From 833f68f88b11b4cbd4149c14c2a18b22f2a60343 Mon Sep 17 00:00:00 2001
From: Will Kurt <wckurt@gmail.com>
Date: Mon, 1 Jul 2024 09:00:00 -0700
Subject: [PATCH 001/505] Add abridged version of the .txt article on Coding
 For Structured Generation (#1012)

I've added an abridged version this [post on the .txt
blog](https://blog.dottxt.co/coding-for-structured-generation.html) to
the cookbook that should provide a good overview of a basic workflow for
developing code when working with structured generation.
---
 .../images/coding_structure_diagram.png       | Bin 0 -> 73586 bytes
 .../structured_generation_workflow.md         | 215 ++++++++++++++++++
 2 files changed, 215 insertions(+)
 create mode 100644 docs/cookbook/images/coding_structure_diagram.png
 create mode 100644 docs/cookbook/structured_generation_workflow.md

diff --git a/docs/cookbook/images/coding_structure_diagram.png b/docs/cookbook/images/coding_structure_diagram.png
new file mode 100644
index 0000000000000000000000000000000000000000..c30e04a5b07a9bd962f4e91d1cfccf22d22dc702
GIT binary patch
literal 73586
zcmZsD1za3W@-~*B!8J&57Pka~69}?+Ah-otB(S&!cbDL90TSF{afjf+Z3zw`I0UzE
zc<=sq@7;a*&BA17db+#1rmO0yrwCI~l6i*l90LIX;Tc%=tttWnaue{OM0*N+Q%8v*
zgn)pkWhp7C0+y7dR&lU3v$TdFAb>tQ$F|BNzrpQYPPOpY0a=*aE27ZI!So;`1bDr)
z8GRK8)mN`q9XOIPX+fP29fK2cd(g<tdVjuf4C68lK8fegcCD2Z^Imfe$Mi26og?ZS
z#XAb2<!rc>bMqFImzEdhlKlBVYNYUC$`q0%`*nN3|8QRWObF!gGjd2@=z0=yYVo(a
zDZlfvp!waI>w8-cL$jA+-%KK#sdC<iO{P;Nw<rp4OuXyyh>H+?Avb#Mx$uYWIe1Ne
z4bd!BU`=MtqF#}+-hQCGvVwf$?9QF3hWbJ!=EHO5o?JDi&whwR@(-71g<@gP?5sO`
z1;Z;>#v@ReDfmu$YV0~dBcQ5a(mD>u`%FXmlBd34#rwtrvJF8Fgn%)2b_Yc!wLWtu
zxYJ1X`@~O`dZAXL7t_86UNk?G{C<%BfRU`VGQM3$LlZ{WK2x@eFdmf&9`Xc4tpU*j
zn<*+HumI<12#6t;2q?fABJg<*e1H{+3r0W%e&Yk5w>e1vcNMuQ2l-#;l#e&QQIiCN
zf!}H-4iJcqqlK+gJUlN3c+`yLdo3p|MFl|<TPTN-sjV@D!wqWp*aSh?O%OPRLY$1K
z-JsSsj)HC?w13?p2%JBD%}GoB*DX#~BD7kHD%6s;4iIWS4lWKZT2TyYYHDEzQ!_!;
zx6=P?4*XAq*22ljPLPw+)zy{5m6yZT!JLy@KtO<#i-(hkhaI?s-O=5~$;ge}#*ywH
z5Bb06yoETLI9S>_S=!oAKR(yU*w)!egqHTPqksMT$2cKwmjCX_#_^xi0w&1$c!iUj
zgNyTD&jy+bKYlByV(A94)_!XV1;hjNA<8enFZ|d2|8wQvJ^rnw*1uZ{yn6ZfmVdkQ
z-!0W0Ar6wZP@qdE(SNVZKO6u3;y)V-b3V@eZ<6?jnE(0~kh3U;Fz3J4Ocdi;PuUd$
zf;a;B?VI;*h<ll+F!J8Bh7jXP^j<;}K|}(iLf1&krwSD`xsu<&)b!%dMG|lqvat~*
zl)o{~iU0cWssGP!U+&&Q@=D9j%k@l7x~*t4O;$^WvixQDERQ;fi_6y1^vAIb9yNt~
zrO3A^Z^i%n5De@{pq#*Qmi+Gvz^!bE%uP*r|DQJDzJyEBO&Px<(L#Oy>mbx$LZ+og
z`mcjhN<7>~MqkNU`hU9#FF!!U!~L(rFPfHGs^JUAS3$b}eikj4vj)n;^=$ia-K>F?
zfzO@u>3+kqnTWzBFj5TJ4^2CTysTxlaaa4cK2}?4R^>?bDvJ1jf4<)Wx&rr=P?6K+
zw5B1ne$l2`uD8|@<!#Og8e8oTiLHWU1D;lo3!9pxlE(j-NUQ>N{O*a^{c+63^I938
z0Tg{D_xOh-P95}S+U`Fl5sLRc|5r4n5T8qFeV8wr69Qs_e%OD=$_d|3A^ad894}Zj
z_B#T&B(t(x(SANt3;K^4?<-Ts=PlkZy~p|Ag=h*A-zpvArbb4^eksq{>SUe#*cV|W
z0(kSM$6>2~o!kiCe@y%qeX8IWHkI()Uh`qwRfF8Ca9^|yr4>K~@p<V&>gs>*TIDU0
zgFeTa`n&Z1?f4@jusUW=f-3*BoH?N=`M-#di@z!Uhvfd!h*t!$dAhuUlz;E#AF2ZM
zrRo&nrar~^A3}W83coaT+3_;6Ol-ox&*ZOj5G|mfQY|bW<Nrgjk2)zn0t{m&QVA89
z;D6~T;b%a5X$^?#!2dgVGhi52(qECK{^vajmjyH!uOfVb{y$ddagr*$z%X8Yne_YL
z8B5UvathUyg8qm8{WbV+slYJO#&^X3cgD2%DNUhKs{bkX937;1Mn#_P;Agzc-Z#*l
zAL^R=(|>knD{9N108%=j26_@IuldCM*Z6)&p-_43aXa4ZH>8soxOKMNo(yeey8h%A
zyX?dn)^i+zY^9iBPd3;Z@>=|$uIYSAo<e4pk{j{;TS=zU%H72>zpibobBs-v^NjWu
zN&BT#3^Yn-^_G9xvxbb@IM5w<?z7#)wwv-)mjz43hx@ycwFpidv1DBvfts~&R=#4g
z;s#{&NHX3&J^NlFvvB4@E}{L}c&}*J5B!e%afC~H_QgE^oVf+1ACBl{2Zhs*H(PP?
z9L!cIg9j0uZCVT7*E?TzH!mLU4<CkYn)fvp{+j7;bU2{fTd|VZ+!xC}K5Ly7t><a#
z{`!7V{6FV190r-qs(OhX)Bi9dy+cwKIZM!XnTN>Vct6}MrzabH8o*9V;QN>x>1*sX
z6l)RiPvxk6;fMOd^T+K`h*%^`Be?ZyBVKvfc~(EUw%Ph-Jx0uiQ7J>KUZ0$W0JnKC
za-pgx-L}n=Z{Cz(m*#WfM&Gr|q$0nz)RO|K2}+cYUf3CqqXH1$QPG>dYUsj`cwnKk
zglif9sW?!X72t_{i?;2pZ|uu7>gUbUrzGu~g;xU!hVkXujP%*7^uX14xXsoX3w2Yx
zW5OmH+7P0EE}a-X=V^7blZ}D1_qisR3e+*%2xFHj#*1A%OKlvpn>OvpQ$Bx*)UTeY
zay!c-jsK_Hm(mwKEB6~rVpU;o(lJ!fEU$Cz(_dd=o*TPmS~qSc*;MWEIwiBR<~$2S
z>EXarl-YB;Tn!lsCbvlugasB{$cSBkPW3vU9|6<>0%N+w4n|gOzdx&h!XYxg@0dQ}
ze%}@7;JpWS*`0S6fX0oR=(BDDuU-=GIPw22FF8`S_jM7Opp%8R$HRF;dur?XLW#8y
zxq<s8k>g%fe_BP`_297Z>F_v@^(~-A_!pKkf{0b}+moXnVz)<<=|?Q@wRR=}p;|_Z
zUR%_xM+>Fxe9dkvP50bZ654<12)FOo4oW_`IcT=k7J&Z3LUI>Dozb-w3E0fuzMfK3
zhMxcQu&n`1fy>kUr{@ykHpr{RAOCN3Fa1!d+;a8^O*0i$9henP(zUvZtHW#G0K0o+
zM|tzXN8Z#Bp&3ITuad}No*}l?@`>x0lL{u@7tPD5ORmdH6wXuc%z}tCD#=pipVU$#
z;USso^a+KCfn9*Tq^3p{{LhX+6o_#uL3zo!v7s|k)Gv@8?l0O$0F~9!ATvT5UXGwm
zm%9G5V>0@BJ(5jI9k7!*(vO=F%Gd6HZofHbD1buMRGZ~%Ck$f9@WpCh4RN-=gjoc4
z)z7);<#?R^CL-ov?~maJ_6Zi~SE!oet3-V!L>yvEy`oe{LIytp?=Y*Li6B`0T;kz9
z_aUcaoJL$l`~AYDwS{h4x{KDb&%=$OWbpgKKqHR#dOak1c5k|Gzt)bO(Fl*$@%Mdu
z?gE1;rEulG7x5b9+2&RdyZ-fM!k!J6?zx<n&*9J35e43n_bhxi@YmhcgTD@qn|CIy
zYevk7wN3GAfc<B&F-86f9j2b`maX?%7}MVI>6lpi2#111&ByXL`5)73!!ET7HH-$?
z%30_8j#WQoEl3X;0*3w<^HJ>g;tgrmZ$>6m8ASnype_+gQK8Vg5dP({F|h}@gTcq5
z0=Cn+3lS^+6APhjhh=`(pGQ|62_Kb3ukvYx&Ks7!tq%1M2)4MgY?^l9Mg_`kV(?S?
z=j&AK0$l5nyw9&&CgfPpE~68$!`Wai)29?@*+%!4c7|p)e@Z1hWa-gDE`EB9Q3-wM
zK&bgMA)l;&S}W*YIKdIP{bcMJQ{@_kFfX`6>gIJnuz!(xJ#Acne#I=+@YzYd;h!7K
z%ajg4!PQ)Xzh-+ctVnB@#DdQox3eOs+9s8Rq#9oM+#WM#Fb27MZKs$-Q*DlkT<DP-
zo{#z5Pg1yVeBJ{5l>Mq~B7<0!nZgVQIU(14L$mEn*U=3Z#U-DsA;V8onG7}ToV`8D
zKkfS|A>e@9K@DB#D@k3uH)n>Ir^bK)Zc-n`?KJp=SgWJWH<^!~?Z?5<LcHhjGYVLD
zitW`f@0e2wUwuNdTEaCl_UjHiFJlK_l@=P)Om0-HeQp+)7m}CeyzA}TyY>K&r%aA}
zgCR{IJ5`k-$t|kM=+BVheOzxa8uE*((siHP^%o<6k(rKl>|+^;KMf)=n4=r^%h$fU
zKC20<fuT+oDy)67t9Vc6UY~gyyxg!~zc{MrzM1F}%dl^-p~GXbz5%>?)sD|6m$mll
zSVJbHTNAgaa27YCw|RGY6K=xYli~v}m*oPyIa`kWGncQfcPo^O`@EUJrG00mFLCQG
z*CLYP=NI$lFS=`GJo#e;wn)2cC>aD9g*hLNV|?a1yXNktp@3rrsxH<<X$nN3Yt=G!
z6NuZ8@g?z;mb(-4{!vMq^ZEnLD#Ua~+kE6Ozqp7U{m9`31}EXoVY`odc%UKbOwz}o
z<V=;yoWPS2{$)~w6_v<%6sgcd1pNA-^%_|>y#kX;x7gl<<PA3ex_GfdgRZoOzUNkR
z-NhoDsJ(MRcae!<b-cnHMfX}>T9t0AimLQQ5tmac-y$~5dNMTdODkIl<F!JPHvmi2
zFhMgTQ9G@nOZL=mBG854VF5%Y>glFEjWrmEEDQ;9Sonc9x_b8s7;GveNxLtHOCxUi
zc8lLd{8u<@`8SIl>$i)RhI8HrK3h+jl;6a$HSbk3Fp${BQM$}^ezK)cylCJ>_<p)`
zGQ`Obn8~6$%JYln%}RY6j|KC~Osv^%6O%E`_TlV@TXV-AJekY%ydh5}vLo2(sL(zF
ziJWEJ<roQ}?$c7V#IT?rm(G>hH{qv8`Zh_j<4|JDOc@lxSOg9>OU@F?xy>$T^jTIm
zdYH$W#()JcU3uIA4X)1bUji72Go(CDaXtC6{;I!}Z3(RwE4<H2=C*>N<U%f`WIU#_
zQc^542=G5yKKEB~dE~5<*o$_MttT%P2YUS#t&2Fd?D&4Vtq+I$3lX!5q?!W8|9$r#
zL9j$)-Hc9&sgxYcY~=n0+);Vn@*Xhs?>8lMG~V<*i`v>t!~66Gwhzu8rtzErGZkPv
zye!$T#Y?_=$&kz=OrCkP)Ze6A_5ty}m($y5y}CrCo^10<DHCu{)HA~m?|+w6*swQi
zv!4O7u-$U}z-%7t+gbJzUe23Mt*R1kgA&dMc;%*az7g5^btrcF*T8FS@{kchrm-SL
zqCbU;6S5|c6`k*hj>{ib@*CQ)H1j1M)|y8jdlY<=^Stw82$D?cLRrJlh4mkRH`f2t
zP=wFBA<pc{(a^OV2d9u#MQeSljKDG^S>G(N@bOg>%U>NDen(bAJRxd)8al+_^4U#{
zP0&-&C>$r+bwbwVfFkO0gJH^bej{sj0XI8zk<P16zJr8KX}Ft0jWE=N5gQ+eFu?*A
zMatbJ2In}HXdRa(mpWwrEfdzaMu;){bch03WNPfxcH{eJqbm4Jn0#a-`<Fm|o#!J0
zWp{sUf8@NGaTpZAO&^ThiNh8?>iD*3RkKJHtoOqk9va5!^7A?1{CE`&{ds1Kb#b%f
z!nIEj(AQHK^ZXUd^FhC$Y!$?())E8j0v(61FJ5<@jz4$@Z=cRq4fiGJUB7fY`ao`_
zfc><s;npVb%(!S7Ffj^sV@AK1&C;R6o~I+`sKXzg8{muvMZC;q7{O(38p$bj3>6Q-
zU)hvU6-H`_H8NU@LEj2{K~Eg^!YR&@o?)EXKzS;!6y^TUSFuFqCW;7Szr6Fw?E9&v
zN=V(g(8X8?)u(R?;|%o#yXLK3@)R@FqM0ib3)Y<@pIrhai1pzFl9-H}cE`Pn7hCDJ
z^@EL$HD-pqbM{%%7zfe!5un*1kIop&jawLdFh>L^$L*SH%>0*PN5m@)BcB}KPj+P6
z2G53gW!L-7)rdQ%6Z|r`QgO$cr6Ckd0-6f4fzuY2FSoUBjtM0`V^&<LDT_eZ3qb}-
zDh75Y&KGTqqvos&qwbA3=&K)cdeEybU@RfvZrE&Qb$H#yQ8y0f%}>X%gX$p;$WM?k
zjuC#i$-6P$g!<RJyvxiIqAchK!`xPE0ZPK#T^Z<DW`KDtgkUgdVQ3n9ITuVOo`0cO
zf3{N7>y^ID*d^rI8L<xd-8-KbJC?Jzs|lBcEsHtue(@l!%A+rB_uB*h`XLYgm22_~
z)~EQFCT}S=>t@2X1+uh0-n((XorfeWMiFi=f0~}NM#3`2y$RISQm8aN3Xy-oNu>qS
zWfo?l0*qZXsiW~&I&;rv6AAdCr6rKk>(4kOJejrI7tuZ}(DpJleCgihp1!2g1MGSy
z{<*(koP@Cb=t<ID%bdwb6#9&&upjKf6G9@t!`=X&FhK}<Yt};=>|em*G6H%uG&&I`
z471afqinlXdyOEqD!*8^GYI>n#b}kw%9B9Vw^X4jp*0&+es!Ft%-97)P*>=RcURml
zCRwUUy9aFIB(vXHY-j$vNK`?B-!rjQ{4DBQzK{Tpd-wE4&(0?#3Cv}79n*p{(h^yE
zx20lkojG{OYE?f{D#gIy0n+!|21arm%hf|2X!F@6b)+HNBh?&I{Fn@hK%gFq9}3}u
zylmCsAF$Y5AY~AzMXNSG>372EAvu-_UYI*WV_ii}ea(As$Gc!0!M-QUcQ6`v&L~c-
zLWpl%Bcb_nmKTnFwFqnorPF2V1H+Z2wZGxfPqJ(2oFX<E#1se4Vh&$2kn9R3rLy^}
zGbXe@+#QC=m9xsLij?|bjb8dMRvj|?GH|uT94pAGI_T?b>;KZfD$VAWYDgn-&T52h
zeMJu^wt=_3OsnzXXoxKNq%h!o1H^=mH#e)So*<6g+PwW*KXapR8^KV_*&4U_Bd>mc
za%>bMBE{!$^s&ibXIc`kuZh(h^mTucxZ{0i+{eUV(>Q)f6W}!LzxSg1P?7W;q2^<o
zn77w7oBViLaMIvo{4O4W&gDu)S9yEoKb`cNR@yXdep4pUU2HNiIRV&n$IQf*Oyd<x
zBa%yxEqlV-ab^Ye&b}i>1o4Am2Q9+}p2xi+2@&}x?l#+-PcNovWl57|((Ujf-Af2_
zTcd1{2Mfg1RTr)*Q?C3IMF+lgaQ5@qkASAzba`@#mF(Q<KkI;|8RalyD$JL%<L9U2
z3$dDb4hZD=kr6GsJo|I=7G+1l(Bma#69`Nu<wfG<d!{OSvoZ^z{o|zr{hO2P$FOav
zc#XdSEu5o$XSIWpc#kc0#OMArd(jTd#}1wUP)^nI6J0dU((*hpIjNDyPg*H{+q`0B
zl%`PeJ;;w3^<7#*_Cpp<s@oseV)a_??(RM813tM}<U=^*U+T54Wj_Y0`}GLLozwfW
zGH_05QK>AXAki_l5FV>~Eqh(OdS$}JDN<!FQyLZZQ@XT_)9njat5O4oy{C4d2b03j
zUd+QvqwZ=7n<3oj)rg8;*BlAuSm^Cth@uS@WZ6|$QuY-#`7hHPMWFq3-;-6wq3Vm4
znyoaru)Vx|#}_Ay156n7KdFo&S+R^SO2!JchfFiDpKyk<b=^g&DwMXw*Iqo;WE?WL
zj<#%>NTX48N`Irhjq8Wafa{o8_Oe1ZBoAcT@SRw|ZbkeiglEXJzej5Ago2=&E6y|U
zg`3`wcW$rj@aU$g7D@knV0i(z0~?<J29$TfuJo$qM~}fu9_myhx0OVxpt9QWQH<C<
zOs13)m9PLxM82GZq)otT)QJ(4>?J|-kw}6oqvfaz)28YVSq(;_3!jnSwk0S^!wxW8
zd&;kN9mhmmKOSq;wPqc1+2@B)x}==BggJ9i!0jQ73MnhZy4Pei9erEc${$VL6y0q}
z#k@V2ke7{GGnTVs_=`l`t&{W8QKhD7ko4uK%@RKwzaLKb%V*X)yzGXAT35vYp=)iG
zmd<xRkBcSOMHuC>9EFHDIi@1Q?o(n9W6rQdjISS7Ax|;6wE5C$s>bikGWdmM6jDD*
z4{Oau=arhE5B;u>tmCjX;;>`#&SPm985`ZNe|5@R_^ZqB5NB*w&c6YRhEWNP6O753
zqxVIixqj+Id!V&Z9a;hDh(`Rh)f3e?G}~3E-rqyG^s#iYO28gHY%ngCrEIKh7!gpp
z{XvUt;ALY$)OJ!kCUjwPgy$Y{EiXOehkvNa?~~T7x{b?CZhV<~FL~S?)S93Ayl)7@
zW)RCI@YIh-UA~5JNg)a<L1VBYQM6USC0w6AgC6l+Hw9+_tT3)`ksAa*Nb<Z6hb<$t
z&{TtvOi#~vuXz|08KfG<Gw!@MTH#92W!GgdXj?cbau?Z=(~2v>E+cY~_tVum?y?2q
z^pmQz<$W!ORsi#9S6>>~Y*BS`?EEJ87#!!^p?%1ub{|jDwXtMx0g0{RFa(e($af@A
zo2Ka*hcPTveEDb}hA?Rr8^}_twCbN45}~v67Fvm$PRy93-SDhw;tyk=LDs;6&e(^r
zY!Yg|(80TTxlR@->2158A6iVRK}~hFt>YP1mTEI53kCa3nNF(j2Ov~X2MVstIL^S!
zd<fV15T8ftwg)KN=iVL0Y%gLKB-3eY{yA@PR8tvB@`z(npv49nH}BWiB#CZ4jcR;(
z8V8ov7{qHQchnurWiHM`M@g9pDeX6kV~Gp69F!(Vo0w4Rc?vbw_?g64ox>E<7_!!-
zt4O?7Ti&bp-O%Um>=(s1%d?Hvn8rX!sS5zG;?qJgL1sG3Y9G(og&Hm#@BeVA_TUXt
z0u3)Kl-e|d+4@A8nPf_Bs}#Oh9UAY=<=LcHe|x6Z{11E~O&+0vK`*Te&NCtf>fz2s
z`*PG*xPUff)e1xT?M{PHv&q+-oICrbY&(X*GK}FzQQiHV{aIy15J}JIcqCMTflw}`
zQxI-A@`;VAVSJWR{zeu~jDaB<{*EqxJcE4POw(Mb_O6fImMuLxq!TX9+IS|6O0hg=
zzSl^l#7`(R3dM{sv43Wg<gVF9DO&KWzFS?D{1J!E2}2>)wdVc{DH`Eh9OkmM1t}m`
zKSs1#JlU0{GZA2FK8F}*x`;?}eh&F6_+TKXj!9qSz<XDQ-q;>@1(G$)nI`GJW%X{3
zJ${=YL(k)&W%w_LnO#zm*Xw=vcQ@&FpHlm&Jcc(4_Edd9Sk=9HY26M{dgk}bM_H?$
z<{9?Vl7%UN4<=pxF{GX#jsOE(^qhTK7`50^-r6@*m_j=+<@jiR_FzWunq7*2Xj+r(
z5%U4uA`lGJ6lvYBcVY(--H4SO1_0Q3i0MLmbC30s%OVs)83vh+hG>iwTO?)X^gUaJ
z7nu^5<qs=nydF2hu5cPeM1Gn;bahtYqhDW^^z5uUqSe&w9X$XJz;)7er?~?;aJxwd
z0i>VOrr1*Cyh{l`C!e|cqaZj6b-(@Lew^;nG{-j}0lv+6^hRwJ<pR~ILh;Ke^}YFc
zJ_NNx$DFjG&X=OA2vd}<@XwYAzZ2Ayn1$3yv-LD`p78;ZTR=oU<6j221P#~Gb>p*)
zbQ$R_%w*sL3oY0RJ(etfZ*Er4QBkUd1RMZFUQtxf7;7xpN}$8lE}My(X=bCzl=X<3
z;{ccolexJg(n}9}6V0G;JX94)?hT?hjA;E6w3))PwO2F7SB2G?@N99U{^Q)LHXg=t
zj^!ixkd|)K#9St8b7PE*L|+R4$df#o>kC#DufPZ26=rL?v4s+*yH|83xXbMn-0g2p
zF6sT2SxQP(HHc=ojNJyT4ZyGvld9aR_2&ZuE^WdMM6LVYg|@SYNAthxPX}7-&SrFP
z()Sz|5M)}%{Ho*C=j;IRGdXdEUwA(?ifIUls5D&Qov1JL>dTfM<<Y=ec>b3>;?Wg=
z(afk{eHFVaO8r<|G&nkn$r}D<;hW3+n}t|bzX*<|?~y?G#E5YNzzx>hy;6vdGpGg*
zcI=d|?Q(LhK~>58gR#B5kr0Mx*7LQuJSDZsJx3YZvnHJ33LZqKjBdBb{Ud_Ar6vFl
z(xfo`MWN3C3rF}!oEM0H;!Rr_>v2ZgZ#Owd@MI|PL^NCXjEP;gB&9!NwOi3=%zIrd
zK|Q@{xu4C2ZnC;8Q`mJN^2OiGRhfJ--+Z2n3t;K>X4zinVM#{o09>S5A9EA&`fWnb
zT!hfYC4d{LyXa++yFCpKgX`US^4r;({jT@9+coSPa~pc~<<9c~b{O~D&be_++R<g(
zvd4UUk;+l>0m!!LezLh%aBDlhZ2<&(H{x=YYM#ZsxWzYhLt%GjdPhGd{p*>-w7TZ}
zA$=A{?h)Ji4Jk&WprBDx-N{Nx<yW1@ZSrxaJVYyP`1$Fa;S|86YSqRsRVp`LyKbeF
zBpGT9T@k=m{4wPc^Hz8cs|XA}b<xw=yvSCFI=NYX__;h~;vT_X52~4X3>wTXk5+?8
zw9AE#^)rxJ^?nlH18P3bGUKgER9Xj*0o5t^bc)CfUjo^V<uCi%8Y*dF52m8;4ehQZ
z2l^MHCIkRvP=}S9qt;uuGS37OsU7m$NX82U&7Fv*4XZ}+NMHgykkWbyyZB%kHy`dG
zA|KXrSl(uc>NTJ;TDuK~#C@PZ1$$x)&iTIpF~j?lWV_@yWJbS92&JD&h~8hWPl#w~
zz!1sp%C?&5nWrT~uwUKM(EsU_7!e6zXdt$v-_}`B8{w9+5juVD=yiL@8_u@uN+^Bs
zhNo&7@K(m{aQasPB%BY_1-QW#i&N`0S=-_<;XI{(eoqg1o68n&Vvbq0ah8-f71Njo
zZ1auDKMj%SydygdQ;JPuZi?jTmtL3nrdR2L2T#FW!#76PTPS#uIW=l#KtL5e-1%Le
z4k!LoASw>0e#@`YmXpD=_tQU28fPVGIwKz83eK=65sj$bmYAf|(nekQ1dJ0qdbDMS
zWSmC1aa#?8nryfGY^M?dd7t?>>E+l5Gah)pSGB;8x8>>Z6UOTalwNn{90mKQ+qe41
zMdl6FYIAGA@`J%TU5;x!6cJHn$&|}AD&w1Q?sKE`v?`@rQ8~W(P?Nn&_J~&58)(2g
zh4BKc@B?HV;SZ#;m_eP`_aDjNXuHLbOup0R7ZK{`yXzhV-mbTIHe`^Lqkt5^gWbz;
zhq5{>US$eC+@6jAwg~>ESy;Kmei4FT@;#R3+yK(lrt;ehJMc_hXInd1$bnS4!|NAL
zSI_TNQTmtkGcj6MdA3&b<jmX>k$La0O{h4wnz~D)E5tZBr4p3I?rNu!<#1YQ>)Qzy
zr-rRgfarTMnd{))vU-t5`7nvBzLpFr16y(6fnzQ25Q4_Ln=dX9x{jA`LAAN^tTHiF
zA`3dp&tG{-)DydC7U_Il?A%W<*Nh|Tdbjb(<juyDZ*d=%ta^BT&1p1RO1qpmr(3=m
zQ$li{vMSYeT~J$!R~xwWC=v)I2Z`)&r5SjgO;5YzmY-;~f9Utnuk)`HX4fF{#xlz}
zGD|Eit)5BNu(>;%ZOB6LoMEX=UiKihfvpCIO}~tDTmw5UyxXX^tsR$sS`l6906>cK
z%Z-)70Y<_P#;S*o)Eedm(63paiqDF_ZQWD?h-Qj;w$}{He&TQ2<U4jXr;;n09`+En
z2-8KAp1{vbd0`J-hKkGKGcigZm21iek`28r?HD9UrH1I_DXZkCx<RXNP${Rt$x>cB
zxJZ$K1l`@33LxPMvd;7N03))pZ5iKWYq&&=;O}qMk?B=6rS%IRXHLYeo!Jo8_ee)O
z<th3L(QPJPc>Q@-B+FKqo^q%Y_*OydP$fFf^lPR|HRnfDOq?;bouC-4ruMsi+pR~C
z#%A2;eBvm>s;n-luHJY64!~|*O9f{-^Nkplrl%)lI+<{9QTx@t>cLl0BYu?CD=Bum
zB3Txdblj9-DJyY(X<ncW7~f>Yn}Da|8E;pqy5u(i_Y+)}{>lBICoZ7CosB;0^*VNz
zpO(w2a{I*fL2*9TMep*%y{jY}uN*kZ?Yp#HbWob<aUs^>i`j<!G3nY&$FqZ_=6>c9
zC<!WWCKdNXRj2N`NkVsMFK_Hl2q@(1o4t#e6Yio50o-P#s_haV8y>$?%0=&+eMPqT
zM%D<XHST^2lT%?_j5x+sSLT`Wqsx@9hFZ#Q3I@Aso)@${jjE7Hr@pN;;~pm<EG%<!
za}s0zl_`WRdQ$a*gHRf-8tR2!v#0O28e|6KGoxmf!lOX+HhE^(4nV}xkVhu?RVGJ!
zHt-Np><I0^R2cUZmnyHvnL3g>@9iv|n^cZ(U{{1by9U<p#=G-H=;EzlRntlk%?=gP
zc7|iLJ7CWT1r=12K9<@+h%6+hQP5_o{qP4#vS{4uk8}qQ322A{C8uuuj}Mv;bq-nZ
zcXP1TMn{M6hN|p>i-kU~B2;JdWb?{ByOT89);CVlHXE>WI;REE^t|ZCCs9biyA}d`
zk?E<~PxGtHosJYF_D}gOFb^rT6Y}?3N59fv(1arb{=gdtpV;0^dV{l{Oz>qrl0d6B
z862};(YjuBsrfe+H~t*n9`=2`zaFz2<o})%d3ILX!DZWG${cTT)HQRNS7Myl(9_<d
z)i^*V8ePr%*vmB9%DBEp*l^bi8|J8a1Lk27My(X(J*wQR9$HuqlEZMU+h$h1k^ua`
zseK^fQ>167-^u=TmqYa>q*_}}orQjX?@bQbY+%Ss<Tq#+c5L#dG1BUd5?HvzSbCRa
z*2R$giOg!ea1KX%p4MPat%fUaxuy_=@IGMMr;0K$6+hV)dvL$#P*MIDWdY+&4y1f&
z`kbnPB2MJB(M$hv;xM$@ID{lNCd)~B7|*nKlOWj&jY!K|lyF&u2KuM8CN`D)aJvAS
zn&Q7Pgwnt{5bb2E9s0p?POZr}IJzuq^bl6LA`ES1rg}<Mem)RS0WX#j=Dmt9IrHk?
zuDKJHk|Z}#LuX+0bCQdv$c?P&+h7SocRiZW^$x*ck2h?@E1zYkst%Si2xfEgwMgI=
z8LwJwm)rPwE0>i@vtdhX_(AEfIo`ou*(q&3lXsAU7=J1i=z*+8_t?YW#zFKyLpofH
zo6ItuXC`*OPYvD!sO<No)rFP3h-n@|O|fI#d`X|H&Ezt9+!@8++bvomyY?i(LtkP<
zKOPcYM`My+U~pqvo^=f??z|C(#_f`hBlfU2?(lzqt&|;>{NZ;HkcemBbFs|y$nx-Y
z%zb~@ayr`BD3qv__XO{|AHuQ>JR!{aIlqlw%<}>1h(+rO*L)eHvVW6H#IaEp<i?+^
zJDFf0+2i&o#Bm5Sg4|&`dA_}Y^n?nTelgy0c5K>6?>tWl6WB{>!Q0XUG~@naP*m-2
zF~TR`Re3`_q`A+6w2;NJ$cA3aQAq5eai{%3g|L7D1_Ox)l}{`}5Oh}GwJzibALCEa
zqYh+`_p5|&3M+ic2NzaDUi|!m#ro){K})}8S1nXzqu_vohto+g?tr}R-AHZ-l};VL
zET23+jQ%Y1>gM(ebmwU3Fu+itj&yJ0uc~YfW#daa0ag9UyzdPR#0@~jsYjQFJ=hbS
zll4ZA^Wj+#AsWYjcN%gG7r4$i%Ags4axiiRTW-Dvm=^zHz9gd6M(y%wbgQ^OD?=t=
zRH)S6yLABJw+T^(F;pV($h-%EOhDFcc>`6e^ha8BlRRv}XJ*>%Ku#4Z-FpPDT}Fpg
z9*rwAnfjb2x)3V*d?H60y9+b|wI~S`U>BmjdFK>PwcUM!Z>BF<6v3QYPFc(T-eE~p
z6ztOT2Z5U*`7Yyuzx^tTrc+iS<zgr^o6i{ijJ=~pJRY(B`PC+;k0rZG`5^_lfz^V-
zT5ZY)s7p1l>W#e<OnW}(HQ8q>4n^6W#7r6F7p9jCVTKov2x-3Y0tZqMb;m{g>wT^O
z*03jJQ=k8~>bz)#2IEqly-_43q?4EeZ#pq~XSmON<ym9!19>7ad6QHLWeq#Kn^yGP
z^)o^8O0vCCMA0VpH@#kD@)ZONfe|)T5<-;{LNr$798DSIvpYuJ!HEfUrJw#xqGOE$
zNn=$(k8AL9#_cMO4~Y2yRB`X=3?sY2AP%RWU;jl?-U(BhBxceya4&e{E+7AxNG)Xg
zs!S&x31S{D724&P6y=&_B+$G;#!jHg;#9Q2*?V_sC;!z)>05+VqEsl^pY7*wb0N5a
z>sNosOc-!Xu*j_KQooa7p`+?sAm6!O&;_8vIMl?*4sRm|6VA%IOK-Z~KU1>vO9~Yq
zN9W=!d1!IoXUGWLQKtWw643szKnhZEO8u64Idl}WgDXqke!gIp?s3rgwZ|X!j_q?>
z&Pwi)KqEjD`ZO#MGQ`f&JSrH%Frta}!t<z8wNK@@49%Z)xv=rxvN=~oY35-`i;x-?
z9{X42J)`X8Qc0wmTQQC^`qyE3zXF}^yz@BN7GSJ6FGS%v8G-`9Yb)nmIiktq&dk7D
zY#U@$>FX$fNe_Lys*JB!s0{E$$V=qOj9r#)s}D#C)!>m~554Tk?ucHH=?05t6FAca
zqIJeDopmMcO?mGySQBjbrDtRwJvA<j-N|d4V5m(DP24@&aSg|zLyWOUS!lKGbSU@I
zc*$XwaFaib{UB%b(}>fPo<jMzIloXInLCn&hbp9%rnri<|9(xP=vC09lCh1YFds5b
zwvok+sFlWJ)FADL(&4X?+{RNb59)>ye+TAEYQ3~Sl?sbiv{BIZSs+~VJH>6^2TxA^
z5G^fGO0gLrfmbyE3OvDB*mS&IAq-#)a`uhoxQ=@G*-Zc)hDqr#kSAPqE6}|wE)dpY
z$HB4n>dzB-PY}I>LzhS_y_UYo$*$o;Q*COXP9jP~WyD3_^Ig`#`zpy`EM|S%L0;Qz
z0$1VPHl*G2)13shBr-cRBJHQVFjPxkn@p|FxE?W70}OmT3sfX5(jfGuqY0xA(fL03
z|75%Xv77~=UlF;w{j0M)+z*XXDHe}hDh@~1zU!d85C$&BTn5~Rar(~iVEm{OHtt;B
zk(cq>N6kH|RgF~k(^5s9L8O|-h5}nbWd<~KwG}^_6~?MQ-JM>^C9`^Tp)-+|hHBki
zI!Z8_8In0I4p5I6f)o}5d4hVM%QvJaWIE7?%Ef7ESS+(d?>br2LLbr?r^l*Hv!Vy!
zj$_ao$#uqIuZ~xzyI3*DiSG}dEbCIU$un9pQEVr8!`m`1<-RTUF%esA$0;#F7jSIX
zN`akMUx*ao*WC064}mZb1TVL-hb;9`!r%J(RR9jp81}a-<N)FI<EA%wNq^fQ=+Foy
z#u?D@h-D4GGe$aWRZko5yd5|q;Jn)O>cyFTEtWOQNy71^<&sz`K}k>Jowr?uB$-Wq
z;7lXnlk!dONc3g}sLa{FI_E|B^8NG9>qzr+Wq<NNK%&>;1GcT<@%c~zJFgaAjsMNI
z6ipO(4_2v{Ru^9l>f)dxj@Tb$t)S3&r*m;gl+a%~e7AlpkP%O)U;3bxI{@Dc?8Lu6
zqVgppORy(bJ`uO;zuG<;+dtn`LkXAjy~#KvB^&jksYR+lKbQYE2b9_tDac+QTd403
zKfRFy@yi5GLVtj1M#uJXc2dva3+k;c9mf}?6den|7B^_TCR|b(U@);sNt~tB-Tmrw
z|M9o1;Ar-J$n{F_89g}VG!Cl|)U8D+$v`q|LL!s)9(A$%v@o#@VpwqPrFkw#JZ2X?
zR=R(@d&aAKqp@4b0q+kB!q$f|_un}K9i#(D!x*kuCL@PgI?sLMWc0Ww9tZ=mMeZo=
zI4H%n@KgMMu0&T}`c##u*R__Eulw6~9hY?X=KxcTU9Hq^--)B>YlcHO4W1ma{jrJG
zxhKD1f{?LR)!^CT^0E%Rk|i#mg*lpU=&Fn7;MRIuFOiVS)pPhs%B8js3yo$Hl*@4-
z3GPw~dyI^9>onfs%dPsi06Y|3P2JbHH|bkHb&GTSFmsWpvcj=Om0-3LJq^(Zu=Qw0
zVtxvtpPGDkj`WIQ8^5VniG7BP<gL$IFMe7bwA~y~1z_~3bc<vl&rh#rcZ)SFRw(0v
z1Q8t&xe`-f_`e}GyN~O_5C6$jr~H3>)*KiLM%gd_C{pRdGyAYayD|ptDQ)x2=>V1_
z4gNEXst*7<Wf60CkCF6}&DWb^8(@;Ev-q>}9-D9nbv2x9eO3q>o=^%cQMBSG#${`S
zA@sw&z-*g(t*cK_*_KTIND1I<SN2Z)3PeOHoI+1DI=?Be26NYzXGwcqbN8<bFe+!k
z=@{oVO}nd0Z#1rs*ZaTnJp7qdhSoz22MVGEwlm^!Sb=cuvo70>>2ulyJ7EHgS%ArB
z`mhq(HKf;ekktg>83aW7%V~Sr&3l0h)JRVlI&!G0cD%EY(*|ozR!>`U^i^v9rIQPx
z;;TgZT6q*)?JOExfmh0@nk25TncDniJvn9qxRALZ_PY82M{LhXi?t~JQpx9Doz0V*
zfuwh4%w<oPQBSD&E(0JrmRVKSXiX~nV9>RT7Q*Z&w_};-V(#989d2Gf)if>j<+%S7
z$vaxfQLUgJXeUZCIVc!_4)`2J)}ijsYsyvY^{>_mf+)}F9Wa6foHKU?F(3tQ20nJH
z0MJl7A8XQ|BW_zgz{sGawK!lC1>3M>V`}`ouD;TO1sl7KT!N?5m(Yr4?$pKBGHYh=
zC^Y74nV-MXy(P7FwJ9_(rm_Hgp?tbHfXW6-?}=m|j44*A8}640q`6n3fBxV*fS!d7
z=dCIV)g$wIged=Uh}WSkNQ@V#X_4Q(?hn8bookvDBr$+n4G2aeg&4a4WS7K|x(K4E
z^^6?C+zr~cI2l#OuRmbbAU9C5lD$!Lw$xN&$!5Iu7)+NARyi^Dg9VI4XZTqk$Uew_
z?aGriQ)3#wkT2z~2B6rKN07@Ubst83y0&+boC&%fUfspS$Pk0z-Fy#UVzEE&0o*&U
z3;lua-L1i^?BT=a-^C?qk6ye@pYF;d^LGes<b$6W=?d%oeMk6E@LqNRgsuO;o5sp5
zB{GDkElwfb*uU!zNOdH+umhRAf=vz1AN49Kru0#C;wOyBokEcwP`LJKQR>d1Y*PQl
zNU~LVqjY#V&fB4X`6kW;AZwoDN^?-YO)>HLXJ^vh8T9gkd*KpL6Ud!DMZ%te{1{-r
zTG%vh61&RAFuP_Le^+92of^3<G?ik+<Rvy`3u((|a6cJR*AZRuMV7<hPXe<|19aQ+
zI!`;LYUOpEJ(Sts0OM$=Y~gNnEyN+ns)1muet)r{LcjNJmiZb6y_rPB6AHnk8E8|1
z+cCfa0t=hDCP{W1c};sO*%0glLB(B#&)rbp*{Y2aEz@@Sh)CS4pEC`?CeVT`Ah|yy
zeB498XTpDhtzpB2g?|Zdgc<79YY}dxkT;7Ra$iu|3~H6bTpO7h4ev;I>~cCY)0w~O
zMe+<*E@NvJs%LXY*pw_xuy1cRkLs=UQ#eKn{K^T~I~{oTtri)1?MF9?jj8vx5Y4|U
z2mD$9uO4t$4uOJ;t)|^_))A)glcru`-Go9j8dlzeY)~N@XTWW5VDj5S7kh^lKh)=i
z&2|{;VN6*TmzYmxg%*5sYBy-K3f@^iGV;i+!c`sTj6$SJq(8Tw46@2kJ~3kWljutL
zez^{BmuY4#`)<#?zAaZWWWn3yDzFpSprH+4JagTy0-o%Nr*?17G)D1OlneDy&29L!
z{BV2`C?Ukz^oy3<gYqDv%w!tEg(8LcMmw6}#Q4R*5G?9OBJyQ<KWv;13p=)Tf#fNM
ztlidu%wW@rw?NXnV(=$og5|QG4IBM~d>)Ipr(?@|YQH_H5cPb+&bDIwSS5%SR@1o3
zR9-)!(9o~m|GOc?L1M4@bzNRl)gxCmiZ?4eV`AtT6nh7I{Y3$08FpLE_9gO2q7Gn=
zUd#pMIRR*!?q)%M^g|h!4<~2*BcsyVpN;Q}lYmv8?fl6NaS3ZZv>;KV+NKvI%^g9x
zSh;7M%UDj*nWuon!qhfY4frE7a%>YymA{o96GVx9W%J6T8-sSDrQaWk@Hkr;%VEpg
z`BuMbj*|5$J+G^#z!2H;UIgUq+#EOF9K2Df-3CsLLyqP>WBvMg9!e9ME$>Gl)T4Gh
z^92qfQ*PZdv>!q?`_Ww8b#aMl#+_|SubKXM_k1k%eOv0{yV1LpZL=)h?cZPoevcM|
z(EbW%jIyj(d^S;B2x2IHym+&BfcW?i@{6_qWr<4A6pr;H_1$CjeG&K-#iW1iZxyf%
zU~CQ6p2S~Czk6#xFEvh`#hzr!Vohfs|H$Nbo|LMRu#9GwTeWvDZwJi#V>wQ>I!9#(
z(u^G7Ax>_%hpXz-Pp^4uIfn_EeWWRsXv@=Lf%iNyF4kXTC`Yy*L>sWOW9`?clwx%X
zC@l!<`QSKWPar+8Z9pS$9fvd~V5hm&?AI)S%mCB)y2Qk_1<88zbj^A-LnF)ahh~Bi
zN)X(gVD6dvJ>7f^dCD;k<cQIX;bQ2$jyC{<$)k6&Bu?L3qqJTBX1&3p#m*FaTK|3J
z*)H43BV!EP0m8f(&V!{CM1>98y2&Levq&CgMw-T7;ZN{$5PQRL;gDkWW#N;$0g%`{
zL;0>4(EpV)ko4r|tn6ybnCMOEGv<v<yb5+hoF9%%<ahNB`lB-ihw|J(xN`n#gl)_5
zZ9RY3%d)zH*gC2dI(FDApqzsg;ZPx%@PwoR)rq&C_^PQm-2ef#bW)EXJtCbI569uK
z`mkpsSas-Cui_H=!I>0MwaQKEw}^#NB@2>pj^k3;kO?A0oRbkQ<dzw`S3cn@w|6OR
z+g&U_<Jf*j&$Hg2895pE8;wgAoPr{(ZD!#7RDNaPMS%N;SqEJx$xDMY4Fh*bx1)mU
z3&cVnOvz7ZWwe|D0J!wlSLK^sFltw6HL0JT*bx;iKl)QdG4R*L>~nX(b-R%LIB1Pz
zG%;Dbu3Lpe7n<JLlbP)uJ5+$@Kg&Fz$^9!(g7$qE9(vKFBU3(s;h}{T|4#^%$nO@7
z>x}5&52%U$R1UzJMOcYVlDWxeT@>3@**izs#(a@3$b8&HR<4!yM8Xw8plCM=-1FFw
zZAmR3kzLSoVBn=Yp-LYb9t~VVl@f957F*JRW!m=wBMJQ==A)q&>vY*2U@dy-Iy}O~
z0#MoM8{}u+W+lj=uP>og@VAUsV<1<w@Lu`%8;=v|x3@;=nSF@U7Jsbu=$Tw`{$^M8
zR(PC}Jz_kn9^gy@6=kuPyEs0V0nk^+!yJ~HR7)$U=nWn{V+fftOT}M!(QubX<&jd?
zz~$Th&COJ6JU&2{Rj0wTp$wEg{EBiA$mEgOOneFe9)p|Q0`P!mx4s^2E7<;ERxbm%
z8c&e4aQh?4)N<cLo53s3!?RK8;Uwj}2r*395+KD_^Cz|ds9)Kn7ljN-Y?y!g6mcR!
zuc_OX(a1J#52(^GGcSmv#bppj;^yI^>nSrw^N5*Qx9t&>q-6{ItijfVk0(njZp?JT
zzy;M9c97S59X1}f<*8KMFusVMjMn|uZq{Ma^40Zi@q-5vfH+hVp9A29SNl78K!A!g
zzB1Uks4eob)P$^7bo5*7jsU1U&lSS9y)LjSmVFEwWXWJ6+)_^_70Lv04CSum4Ic{f
z=B@Oer7zzyvgMs&03*~u$X_a>6W`-PPqDMb`g&lgU$SQ2=S9XZ&PH4HZs_vefNrWy
z_zME>qb}y`xR{A78z%8c0Zv(kY$5{u7^7A8xSqDfN6zo(w7D55O(7<0$Hx+&j`<s|
zOQbnAake~1BR#yp8Av)l#i#z6m`bfePtNR?%dKbz(#~$A*FR-YnAkWxO(wH$1<Fot
z70)vV$k0FS{pup(pDtKF0}_5+ZOUt}zVA_LP27>azpd6$0!rifQ=`snQxLSX2Y7}<
zjlM|ppZ=yDsFIGRu;hi@obt}<h=CQpH;u$(iE%oVo$bI5XvFbFx4!ok2n=ULFOmiD
zv^Wt;sdPv@V&+|nIu(8h<#whzO!io(ZSmhD_m+l3ynh#cWn^g^2ERzR%nfGzg$v98
z3ngX$i?+l`k&5xmOeNOOkGXk6E232mi9KVzD}&|j_t$d47nE4=sR6~+q5GRd+Z2H}
zs$xLx+8OWrXS5&CS48Ec)x_>R$}IF;-ZMAW;qM)+tJt~1?Pb#Ogz5dJIWl5|*(Qgb
z<E(-vUvfQ5&gVukON9`3Pw~^|82=8}8m2O590Cozdjx_qr4U@_RBYO=w(QjA1&?7p
zo569tUIK-(L(r>-c>2!wpy4f=fe@fF%JX&cUzI{YeGJ@xV2yO}*Z7xb2gg#MxgzK%
z*G;SN2@CG_7qHj9xPu`q87i`&1ZL*XhAi~|qWy4Dx6kNUomspdBwoIHNz?43P~CIZ
zPq-s&UKCnaqF*1>?i|=@+0=Jio>fRP7*p`^u@29#3s1{&<kiTxcV7k>Tg;uPE~Yk<
zbjvrP2g~j4&J}0m{&k;8ox$tqxGZn<I40j+IRIco3P8Hxt?{6l)HjOiKg;MOjEk~|
zCTkg^cYCH<O6<Kpy%VGfc&!#=O0<<gAC^tvAdGvV1txt@114V%0wN%C<Fr(hlriBJ
zN0v-6cnm8Decq1OT*-9p02Pb$gV>H5ajvqr8&cLAH*^{xHLqRNuC>?5H#Jr_$f9Si
zMKkOk#UIZJ>tAQV(d_yJBtOW~`Kc|vttD8#*N-}Wgv0F34jQ-XH<R@oz~ucqv-<ly
zHo5SaVM_TtQJ$H&?wsdRNM;8=^DDf?EQD;ng2ME86A0ZVTnGnW`aJ}q6XZ$3l7U*Q
z?rN{VLIwUMC&*&SlJBd;A0Wd}#9jy=IBv~8)!zTma6+#o&^ajSLm_W>Grb}##As-b
z0Olf8|5$z6YjXFgDC97ZpbfLFf#MYD^ln=iVCj^dajiA>irbf!UL6Fh-Cu2=EfShF
zf7Nr_clYLV7(M-zC8S{V>msoTxkgm^()*-zTi~<QrGZ_Az<>h>b5nGZ=wB=#N%TD+
z7(SQDAi*{)Z#~z!!B0DTvLMDc1~I6YT<v{1P{zkra(0cL;kFhA(Z~W2iEeChF0D95
z5WX9BpuO2W_i*EYm)o!05NRUWIBW+|!%8PMAc(jx?18M^b={;Vg9!stxNv?Q_rI_u
zK<hiNc?L-?T5<s#)qL7UYZ>jX;!(pib?9TMms*f|b>Q9Ywiy2d@IxLdY0otO1F!%l
zrv?cwZkQlEpfIcDkv>p*vb)cs3-p3HV&Jh4X>g<ncn)I65xcDvK@I<;pmP7=U_rS(
z^*XZXPn_}lFg4S`L@mCnIe|H8ec0AAu-%89O&rij2B)Yg3H1KiU*s=hX{OwqWB$k`
z4CK_XpiCr4S=?YK(*weV_Jl7cdZ_VF!i<DZCI<W)2tak8uIgesWE;p!YA1CjTL#Aw
z>N^DI;4-4iaV;OF^uUvVg`U)U*BoX3KodftN{qAjqT|%G`x$?&(d)#g<Fb9*tnrJ`
zhRjC5N#941U2u%({Ln@)jS=vQV*7y1_4b_%B?MdS)~<Q*gY1NB=6cr~<TfUEW!}_T
z$B><!E)Qf#?SwcE22Vh$CkdG;8tO0w|HG*-GBy{9zSBEPf_|P#;w?I6qDvc^#iJab
zz@&equoj{_9w|5H6Jwaq!@_jfYex8V=VR7@>ZRr1X3o=BsL!68ZX06@MODYrE4_Pz
zD6i%qkd@3&7*LR;GYEIz&V;Dn@0p9SE^f8|NZ9|wm$MQq$AO`$FtGU~@-4Kd@`khB
zz_~yQF2Kz1cC^xyJfXV!oWyOuSE3iwbt0moB#p#&jr9VkP?);3ERW!93CuH%gL8vJ
zm$r3G_UVJgplQNvO-ip1^5d1NIgWa~nV1_BLiqQY@D_mbZ5v*{ctt+&yA4u8d_u*9
zUu<0~lozD^1=I4)%-z%o;V+X~=&ykMvs|-^=wlhL5U9t=-*S6pKKxbNZL!Y({Vf2Y
zux3Ns9TXyB&vT;i!UO5fmA2DOAP(9$y83g$aMqa!j+l&MZUO@eyGex=Z_s!OQ_f<-
zf&o=O*UvChJ2UU|MH-UyUFOpg^h*C=(FV#L=vb9an>>xGR4g|HXV<}5!Om48W7|q>
zTS5^+H&@%)w&~wpKNr=ppc@dadVNPQPm7b!RCQwgJnj7#nGFG{<<F@<Q2qpRe+UiU
z{$3q@HcV1#pL>(kZU~Ta<}AKM4YDYvxdf7UvT0~B7vm+!Y@G-LjIB=;eZoI-^?(#s
z5k@lWiwK6kJQ9q9+LpIWz0ZR;mjoyCR-|7vxisI^dFT1&Q3TJ#%AFZK*<hd?dZz(a
z``#s}(p2SA$n0~^4N&x~od%z`N45{^<!JqW2d_bc_Uoh+QvA>;+@=ZBdF8dqKtAgg
zJ;VZ_N48R%gVR!XwV7jiU^cQTpi+|(YR1q@&|nUCmNfdnOp*WL2ZPz>mXL1QG(Kl6
zP++%&ofWvPq{Gx=yj#|wv;b}3O?e-I`y1~(*&`D%MY@Sfb8@C7Tj;TLWqxPJHtom@
zlG6oFN>wN2LT{JRZh>wz?STO36hNf}GY~lWZnZ8k=*1|s=1x+GOQ?Xt6n?{bOr)fE
zxdY|147L7qdR#{W#$XmgwqPxG<8TaO9^=LD`BBcutOlf*vR<&9T&-qAY~!ZNlS|I2
zmL6kB&nph_>;BDT!=EPna?}UmZ7L+EA$<wd@WLuAZqgpt0Jqi@2)<M}OifIBQcR|9
z2=>so5K!_u01cD9Ha3dA*O-W9-j0E{<e@rc^L8r28QrGkXr1wS4ZX=yGk{YPEWMi?
z4cdJ<OH@Lfl=*`}L^vtm0v3<YFR!bTCh`Ax`^u;+w=G;iKtj3{Bt$^ELqJIZ>5wi7
zK^mnyL_k17y1To(QyPJfE~QhYL*TBLz4h#U&b@!{7!Jla@a20~&NbJZ&wL){#>4~m
z#1+kc?F_a8Y<sNxk4I7yZRR&v*s{KYBp2o0L&j4W$`Oax=Yt>*CC3fA*1FR0WsfcN
zWP~A`q5JF2Kb{8q-qAB1kUqsU6h1;aF+%2b`^tNN!4nj7*PzaRl(v5?Va<rDBGb0U
zvjKvIC-UYoU(s`*gtBIW^BSv)qSNKwC;Bs)Qyh$ioQ(o*<W9P&xrVC}27OxU^0jiO
zw|#7zmo(AGemLHlG=36?y?0hB_|<-Bl;yh7wBK3K@rUM;$I<Pb+P%&(n!2ctPEY1H
zjKeMlH4&XyJ*jsr41FrT{#3FPrM!Wm0eHI5=<P!EUlbSbW=#tS=GATP8aOEkQ4K!8
zt|&kCp)4looc7y$*Jy|B*^B;WL#n%fbBO=@+_Kn*`r|1>tOCS_{g!iBi3@Fgr0^%{
z7j|Vsek=iOMzqG8qW3!7F>7`eob<c$mnR4N?*_s%PaUBVVnwDa0IC4RJF6i0Qy?67
zfHZ0K?TI1V9tcyErKh3zv5f5au&HXdlvoJvly>Z(x1`ki?@G4A0fRlK7a>{4Udxrk
zW?1~%Kv$=nF=RQ}P~m}{b5>HBdB3#%sZ0i0@gjskBeSm{m3|z}gKze-_Sy;4Z!GD|
z=$BC>!`4Yew7$JD+`DI1%}d3yb9vS<Q98Ro{ajAmjy^8y%+%DS_8o@%v=p7D7r6Cl
z`#OT4NWMsEW0B*IQr%VUa)<fn;}Oey+hn(1rm^aE|NMGHL33Xqt-bGL85GE27D#I6
zkasFH#J)xQCN(^_e8nYO*g6Yevv|4ZxDb8a^7$a3Q#I{Wm--3EF#q>PC7eiC<XPee
z!7h8)^thVKb=IdIJye9Aawx*#=|TQ#Ce&BG@*bnh7x?^Gv)M8;X}f<#_>v^xcuL;z
zk(c{+<Hs8xC(9J1=}CrhhQbkcDmZ;Q(h*n0ZI5vZaRzRH1tn!p<@r%6+|O^_xh*Vr
zN>2mq6-+wVKYoii;+AZGtgDM#hI!tOu4eK{<YxyT2{>k=i7F+?oI6XpnU}K@CX&l<
zQIeEqUx7kymRTPaw(J$9u>G=5v9g%$QiIqqRz~oP=r)$Z2K{Jr#&TK{4a8*a{vWO%
z4IB@@NJ!isRR~mQcBhW9)xV@fe;oLMg^hv#LXkqWjX2nswJ^b^636$x(G1a^J^lgX
zJKyMMs%avfdX#ANPu$Y1aQE<KOW&Qpv&Rt|j$yqb3}V#oFqC}0D#)OI+_E?>iJQmT
zhUdedHX7&~RT`ksn}vzIa@PdCC=r6|j`zPeG!U6}1I3Mp44baF1wNK{yt)`$@91V?
zj8elUb9aN?_BypcbU0~Fg5%HN76t9~tHsa&LDr??4DHuAboCmvc{kKAcG%wP-O{?}
zT6B#g??F9NfUt$pEI&Z(WqCE9mBq7l`QM#Z5XOkpBlsBn&0jL}vl&SrftogsRr$7h
z`iGPw-(}HB1IEELs6yt;D-{o3Mz*rMD?hU41Zn!&rL;GgKVEPaXPUkvz|DdY>E7*d
zetdb!n8E5N)6VXstk;G??J(HRUT7$ZKF@J!!xoCem?9~n!)C7V_|!MZ-U;iic5Y|)
zkdF%cO57^%^Gi1LOi#?1sF_7S>T(lYO*PY$lJ~5^%BQ5Zs<W)#_xl>ZXL+&ralbs>
zMZMi&u*lX(Gn#$;K$L~>RN=1IXG#E?)zWYhjb3hNfAx#{-is;u!pkyOiYLu++hDxH
zKzW+%%hDf)0T|SWx<5`H?V`^|-(ziCeG>Hi%b<v)5tOV*^+9<r_}nho*<6cGU8J4v
zS>^`!Ng$O5y}M(7tS@qq9^r$vk*<~ceSL>r9mn~qf$$!5O&J5p-1bb>vR9AQ-70D?
zY~&*Q6}D9D46Op#(5;<xBVkkh94}f{9+YC0VELe0VNO@tD#znR5sfK~;@|6I_kTm%
zoWOM9mNgI+tl=~-aE)ig_mODO#<HT;#$Q!>Nb6Y;^UwaY)w$~s!?R=Ga+z(PFlFZZ
zv_FcR(##+Cyl)KX!sNg=R2(!w?Yeki;c61RmdJ*>oMmW4Z`olGd|~m}#5d}3!E2!`
z+m*VOFT+W(D*V4Z!WZGuJ;ImTNsWWo?+tlHv#8s4c)#^*X(_y`lj3*d>7}Bh=+$-X
z_FsE<+kYaz`!dgz8?CUu8QuOAVSMM<ZuU%s`rXw9OL=3Y-(E)Jp+B3O>6}OwowA*z
z?z^ndHB<)oz9CJhxs|f}_xd}Xq3XOg)Ob}lTwrQDmBQGk;PkOYZf#RRgsAULi<U$k
zis5^SVB%GW$)jF9ya3PkWBq``=}G%=dQ2K&4ccaN#tRYihqDU5D(J6Q;D7kO#+;~A
z@ZRr#r4h1v|G7>7fYQs1z8{S}NYnD^&&nU+a(qf;`jyO@6l6QOrq0;z;YPQrBX0jq
z)!o}Pw{^b(Pf9gLIhlaA3xZ*&;4Cy`kuD$%S57sp67*JnP!VjrzB|&UMOA<sz4?rn
zAh4|WpsvII;lv`*H?w<p-n3uh(8SwboW3Gw-e?}ZAJlMsa?#Thowd=OYmxwEcI+O*
zuPGc7=;&TKZ}u-ErK;NL`-x{iJqNYhvV`bTMeE~l7x6yO)i?8-4Twd%(ym>dx)*Iq
zr=<&<X1o3*(5c1=lKw{is8P>vH&~Vq-?gq|Yeg$rJ4r)uo74+6igm78)Nv@Yx|1=Z
zY&+b}Eo)alk$E-lpqe1iiaLnJ@_?hMJY?Kf(B{cxjjkBa{Woa~lrqcW{356XI>!<4
zF~U)--0X?zn7$oqc4u<zcl{jY9#cGXPeXSfKp<(j-ND^(6vp9$U#pele3zBQw&|V9
zs9LFxyOm~RCM&jDF+obO#~!%rOwvSrTX@@^8gNZQR!_q*_##EX-R0E}nuhw4&wgnP
zPD!0lXy$l3<m;IIpH(tL*a{sX3Yit*m;fHLtoJO!mA;6?;cM6^1QJ7i{Xr3xpPbaP
z!DY+Ckukun{gk1w#-s4f#+A6=#2cvxTjr_>I<pM!9zVN?jEduBWFu70kG9oR$H`0S
zXEj#+ri9dz1MQ*)C7Ff}se{9&o@(qm8w{75ZJVf`z2SQ>6x46J!Q6K|g79J6_VKC6
zetzt%O>AS}xj~@S4q2v!Oe{Bjq!bxu8KRm#Y1D+F6mw6i$@Mp)mRvihu@j!{x@Csk
z5dQ>&K;fPd1UbT^H(t0{Wb3@ozR?TrCGd4j{g7hK+EAs;gaVV4Bi@K^(n;I81q3?l
zxa%_=k22aoH{Y-jT}{WfjP^>9yRQ)1UQh){%9L)Mzl*}yhI{AAzh~LpnVqEX`EYNO
z)x2&oM5@2`y<a#izW;4)gkN|LJ9))>0xs2NoJm4ytB=B;8zUyl1pLq#`y@B2Kc*9F
z(!c$JQIQz&p4LvwPt6Ilx(k`z@Z_<Pp@>qx!U8~-y?OCs@*WjRioC3)(8{Ypmp#2S
zj4o%S(`e7WB^B`n<R_Y?#~1HZh~6F$UtAd6OTN7PTCN@Q+t*^ISfVL7X$b6BW1!C7
z=+=F&Io&Pylm}I?od^d++3U-Q6xdVz4jNQDXeIL~OfjM<7iD9el_lTXLSf{`NNkQ@
z`>PqSZ9P}1?Lc{fy;%O$;c|4r)5pgK0?zv@1F$zqB6;1E_sK|#FB<2YM-0}hYtA_w
zEF~-b(^U#O_kL<_o8>H@HlGFf{MbXu*|V9B?(oppKd2z(&?Cf*PaEk(Ruk=tt2dO=
zaFVMbBU$M^xL6vBIP9(Pa`x|7#~K#<7PoEhEzf6NHJflW_tkmG3+06@Z(F^5dIxLH
zJHf{RDw8$G>@B>uhW^vhXC7)K;t#|obx>~-wL(NVAL^{R+C`KwM%5XML1`Y!gl;si
zuwnU(SwR$c&)z-lBZIxlEFUv%iZ6qp*vB&D@wX337__-}y5F-bz0#DEN+&euU@i$Z
zvCT`PL$9sG_1S0$=!xvlq&Gkf-YR4DUw=Tx6UmBaTic?t&DMo#u3J!SS@3w}tuFq&
zy0~@A9;;#bti+U1?A*w%ZHi=1_ul*)G&(hj9ggT;KEh}`{^|uw+A=^P(r}Me&nxLs
z;#ovesuxF({oRPIW1fqOxM5Pp;^y>zYs9hUGauYL*r!%QB$kzm(@zmw-n3Ggr?aP&
zXveq{YZ5}iH|7UIey|ajHY_CuYGbgf4y()ElU%IuTPlVH8VX%&iO;7E4WpY_+axup
z?Jb19V63*@!eVqiL(Re84-KGuS|(F+#TKlKIbUVSvMp)(p=;Uih?+FP$d$3VEH{vO
zxOAj`2u_5Hu#HhhQ!#t`mfM@d+g0rXhDkE}&8Ze9U+#p;OJNg&#KFB&)E2BK8SRTz
zrA*F+87m7QbfDUsC4jslf<%H9$J)#pYCEYZahhl{ZtGtcE-LL{5@#gE=cz|Rboxej
zCD#8F{=8;Ma|qr|uc)x=Il5>Zt!biT(`R>xnvIWZXxI4;M?@`gqe`EC4UzyHS^fAy
z;10`ppv<g)88x9;-a$k%7vtXMvc#}_B@hkb8u(W1RQ;xh?ootVWX*FNt<Mmm8NbQj
zTJG~ds@sKm7`%NOVED^0VH0gDg{R+U=e+yddHJ|XCVSC`Ai47Tc4|V`&4c|S+bf4?
zNoFE??_wDXZ0f{@$tVoL6ql{M@_`}Ab=1%z$IJJ4avPt{h<MJY^d6}0ztNp?8BTI=
z!%5~ZodSi_qw)7-rw9^=rc2%<`j&Z1x5!v%5YFEvu>7m16xiL!=Ngt;UoK0ZRz9dM
zBCy;Zx4k+W_N-<z*1vv^Uk=E`4yR;BGvAOP8AMN#-JU4JrJvPT88C{Yq2fc$VR&%y
z`Z1BECTY}h7qW@L7;&A89!VdBXSSl(>Tw|m61)T@Q`)N4(HOv%*-d-hR&6#5H-gzL
zxP&|~=Eq;#|JcRuzZzIcrrDZbR=o-fr7B=JvIgO<yd2@URTdgkDXg(uzN*d!Toq1%
zo$X8@S{kR0XR}Bc#I!bj7juB^GD?t1GZ}+!-}aH^l8<o2)Y82%bO?$Q@F9P&dWSwd
zef#;w%|#wKZGam_3Qy`C`K@ior1$b{NRJZH-Wl#So#@)mnC6`MKVp9Ax?7bUziFll
z7!wkbI8+~up?*rt4Br78#z;)T4x;uc-~Ec5z1r%_Jc#fy9G9M6Jvc%H{>?1<d}~Sl
zRz7=-MGKR<@0(dn-#nR7#+k?zPE#c(^=ZBry@}3iaYnQEQ0U52)!zr=&b@s-Y#`|h
ziqv)!%Zm04O{P2B1zip|vblq(1n=DO6`{;iFLYJkdn9-!S=(EjqY+4VQ_)5W)$Sd~
z$}uv+rNlaRfbiOVLdz|_;kF+dtsLCk!AeaS4EwgPe=M~bElN)=PFa7LC3015a0kqY
zMOuYhF9NMpYL(UhPU_D1v_hn09BS#gUxxx_ze7<#1+sw2z`+Q~;n<l6y?hYalf?Yp
zvO<HSd;AU-_?S}E>#Ih-om8;rk6~f_+`CzyqrOHgi*|+YQhhlCY17EZz65=bSgzCy
zd*R|~^l234yqtfA)}`Vm5LVDG`;pj^(qzIFsGAI~iV^vyhO_3rh2R5*YIzA)U5pu>
zMo^7hRKE45$2YdOO<2D01jQ`HM{HjO=E0V>Jho&Mlk~wYGbAaq44{stEr=HW_@-EI
zA_d@4%gdcDKhR}K0)ubcUFdX21gdh<|3bE1d*a93<96bd+}UpW`CESl6@EC83de10
zCpWU~L!z2B1>)5srSWylUU<WkK|roy9_z|0i4q#%jPdhV0$^#0z?7aZF6_^9s{Fl<
zW+H?v;zOmL0`!wDJ$*U^7`ex5lZlGgmipnR|2yyWj|@+kGi$$}8NHS5{HD=MI$9O=
zSeucnnf=|jc>DL{UASG3qAhOMh!|`uECKE(T_1i5s~$jqBofM)MjHr5pnz$4X4R7&
z<3+AA+jju0cV`MMC~2Ny1O;<hq7a1KXFIE;azwB^baV*BhKH6WP}c!Sv_wKnoBB4X
zodS(C$*N*u9I`Kw>>P60zKu?<prirIj--)R9qGw2^a2YCuZxuJ_NLEXNW7@AYwtZS
z=I%wd=EuSq(s10x=umQA>W}z1a7=mG$5WYUv+P}(vE!+Y>CU_b#f&G9sj^ECX@9g3
zAP@g2{qdAqA<;1T!PQS%F*mG^_o}b<#6B6)v$B&!yiHn>`+D-RUV#;hcLlp;CYtFx
zD4dv>*?E*s45$Z5O*#`E%7eY@?~Ucb;FgCUD5AmG{FTBYUb^C?r-eqK)yF`H1rtI*
zfwOgK^$Z}3*g!~6w@deAbF4>CML%+0O>f@2@k6(+_`HWe!0o~O4TGwOiUH&i{=QVE
zQV*`5SjEK{Wj~rR41MZCJFp|s_DnIKYJk9HzxR<ukiJHihbJ0_8aAQK{3=8Jt9xQ)
z-rj;R?4G4)4LV+eK8b;-=_HP0NkQkYLzFU6xIxrIHWn5Asn5Tf4gX3i+y0=m%}`x-
zjPr1is}N>}Ur-TmN&Eg4@5y@ksuMf5KXbiBN4d0&;ljK0zKT{O0pk&RLisp5-AI*}
zUsmH^;WU!hu6MK{&#>8TC|txSV<lq{fszzaC#ob4Oa7CGx}0fI$y0CHabLf^C10?(
zsJ5~`L~LI8#gLv0`QhElLMo@2rM~a`_usfzEJ&2quh!qsNVohec0xIVg!p{CGnfA$
z$5_#Z6qix9R{Zt7G{>G=c}5G`U>J&3Vzcu0)j-*-MMZG1Pf>QtULx%S(=sF4>#x6>
z_og!C3k%dq8H_41LVLD``UR(D4+wSX<1X-cMjUX%Z+)eMBRN+H{(&&$n5rUGY|fC3
zq+eJ(OXMK{FeqQMwI>uu6}l6DB;|?!dL^UnZ`N!MAt|f+)NC&mvLW6=j#r<<%;Kki
zls_RCMd!FB8x_8;GFwV1=vW)$#_-f9sti$6KZSP)OZw}y)^LSfpw)Br?cY5!MvQ&I
zDeL4+g^3x25l9e?r;DxlrYz)mZ=Qem$A&;8`kemTBBL$;N+RXsx4PO1yw(<)u^W<C
zH8?fTi5jW^NgFi{pJaiLJ1s;e*HBfPYLboG_(rr-2Or%b>r`+qq)bR9<9!EGjaj6$
z?o7npQL94Y(t>P{bAMmVE&^X>$*CSa1goY_nQX}&qpEQG`4o;Z7->Pz?lOmi%nUjU
z%jZ&kI6oqb^CX(>!io^tR$sBFOn&T7US~Pj$}t|Ny+3+?QP)yC3ChN6-l&DzpZkuE
z)Bb$Qlu!D;?K0)!t_C=TGA3Dz=GCj~dfvLqzBqRd`5F@HFG3w;dukChuV!i)7*%4&
zlovYnll3_E?3MG{IqpPCmeuuXx`3leMhYG8{ldcwZ06-}yUD&9g{@zIUY+AF%FeS+
zFJf!aIIQ+QN@1K8<8H1DnCZtHK?a^&G!BKcDB}S+i+Yr<h314wQf(Sy?Ot;MHOp8h
zWwn$->>Ci_IsQRHWr~FW>ha51wym3vqV1`lc^mABX15CC<Vm~{YyuXjDCqSmf`)lh
zXa2tSvo;E3Y;;sl+6gyuGlFu?453fFF`W{fc+2mS#2)({H-hK6#c_NaiOnBcrPeC)
z3+ulb_Gte8rjrSKHFCRR$+4_MvbRT@bocxW)Wm2O@Lo!MO>23TN3+He)IO8%5%GQ#
z;L=LVzI762AleGz9_W}Vq%Cj*MB}8xlCO86rh!!NtKbHK@Ao_7Cm+7iTYBdjef&d*
z!m4JpRrl$-#k!%^*_`AzKN6ndl0ser?WYitqotn*a>-WRde#PC`Z>Ng<|9DH+QIS$
zV?H8X>lP;|B?8i;u<Q-O#!a(i$S|!)rOwq*PHw5h{RV3OWv_02a>jtH_}c!&^dETZ
zpAW<`fP~fuzQE-_2z#ht8gf9_aKqAPwHNqq0tA1fKmRs@K7c^il2uQI<v;ZYzqf@p
zpqNbP*Ln?kcK`lo!U~Z4+AaBi*JR)PrwK0n+TG*zdmvBl88K8ic4K18P@aEbnmAH0
z|I^B6sm2IB<erUR<!>q;=H-Nn`{zAw=U)H)*1sr!p_{BeKzKMzA*ne2=WqWa-c6lF
zePyGQY!FCSpA>P<f*|{l_3gK32VW$};{e0U)K%FP$b-zZur!~xyB)#QvX0)w+1KoX
z{2MpKeOcuS^5I!VO#dE+M|gmd*?}GINEVS;atRRjaay-rRy$u`o=697pZ;7Pe3as5
zFt-i}8qGi|mxDvU9q|B^B664W=4<`Nv)#0={+$qwm?Pj})qxjH!uKvjEG-WKM7Q*1
zwB?39AyuBV`QMvPlFycWunKK1z<XX5DM8fK|NLh|87BE9YTHJxQbkq^r*-QbP$0wu
zJN^aY=`|kDRYX2{8L5739|+N+CXEst{hOYf5imy=GuGE@@x{TFl{Sl9fCzj1mkbH5
ztm=r1-Sg@``hr@qf53pDqfi1H>i#o%7%I+bNGresn<P80-7`<?k_r-${$)|qmvVLZ
zr~=g1JD8uqtpBkbKq>^RW>FT48poi*E+}{H44?*uRmV|kx@-Wkr#WkI)`sPw=W?`V
zlpSYGJl5c7%>fdB(E`)m_192}`QBA_GT#*eV1eDdp)^3oARe!l8UHt7_H#;XPjpNv
z25}y^bXYr1Er3l9ief!|1t`IAe3Z00ce+3#IN<>qVOG>I1ZKY~OEMAwb;k~4Ir8KU
z1M2{K`10|DS=hgJ6n+wzDmAuxv6heXFaca)-mYVi-sfRyIt<tMm{#otaBdTb2$=)K
z#p8gl-E%bvk`I0@5J@I|0i!(cYGjdF`Vhmvc8-o1#K^7kY<#*M#|fB?K>*Tv4Y4iz
z7)T#48>11bGZvt;G?r@EeG!vw@wnXKn^kzFXhZ#9%fIUZ7}m#=B1~uG@M}3u+r<q7
zP%V3u)>@urCH6_+xB3wg^f+=rRR|Fpb4W9my}#@rX7~>*_pd!k4z}>@x~9E_bU+S-
zQ7SZ${>T()^$-ckp_e*_Q<|wNQUfqxNuyal9L}r%QziP>YK)fv2IW9~Tju)V56D^u
zG3|W@L`AvaHg`3K&0TxnFpd=5`5_ABK{SR!J;Abt8^)b~tY*F#Xoxjp_)m~cREHth
zG+-t;Q7_m9tRGWhV^<pzg6su_<$)Ry<VK32=oM2H2Ja_&f`6MET25$5mopQc6B8D{
zonv_@5yI)X;G82%c?5ES34rDG$)rR0=H1C<i1zvp$UYVV7GZ~(WNrEMzJKm4@OdU2
zw5W#3HCvn~sukibwrV<3HLs%P-v=DeVcTcG^|q>*v+0imY?_02h&eyg*8-|smEv^;
zqyL(sBw;Yq43n!vwNuW=@0`ucrWU|lPmuC{e>@8nk(?|;IB$#6He~K0?&0GC=6`G(
z&_C}X>c6)6`T`v6@D9-m$Q{pXT55<-4@!eFR@DT?KDugkYp&LCA|2SjdG6<L)|k%o
zaV-fwU=Ah>T!EcRc$Dna&WfR+%MO?^5^h*Nryy9@1c7rOUCuy(%YS_d<wzG~V1IUD
zHC%z@<kXpkqD4Rj^&YqU1BQKwTLjQ$8dfa0Y{%&~xvXk5ZT$~Rxisj1!itEL8%<f{
zfZ}-v`IW-+*)X`VtEG@%AS|;YkOCGARVXq0s}IuvS0@859kXF!@r?i4FO(xzi0lBY
zz*bLN|LZ};0DBO$GcZ&e0KJX+>Pe5_RhgpkoWVRUWHte*X>HXU#Er2QAurHd<Xua~
zv5GY@Bg;MA378HYSLS}?Q~uX5f4up6LY?dTpgOB&uei4H{9QSe($+%y5G!+K9IKDq
z@*$e!ncla~faCZlAaBXteGHNOLo7i@7PbRy{u+++_LjygvOM$S1Hk&%O_cois(J<p
z*yHf$A%0DIDVVApfJ}HUTw##%jgy7=S_50k1`sMO8i44ku)m@LaTsv#^i^u1uY3=A
zkQ^=ZdHBh}N62mtfGe39;L5>(L<=U=4S$t_$j8Fa>+Qm702*>@Osxjh8=R4FsNZy*
zz9L1eyp3300d=$+ij4iK#-X=tbOVyEA!^)~2UW>iw*^nVH2`fchV9$`;L_{;q2md_
z4S1krAG5MTqe9=Q>yNhEuHz6SDc<AqJPxvNjOCvxRZ$_#n+D*>8UWnQ;~5q;#jT++
zv!5PHX!;&^5xuOG+J$)0S*h+kD?HM1cx7M-3dKb^fKBJx(ar}EfmC*H0R0-*2E!MH
z?jVG^e|ww(pnE>mfzh}8AlqcW>GXTDypJisZkPvj1sGPPP-JH>&dFUyok-W?%;;j>
z`nB^A?>eWt8(^Jo3n#cs9ZRMDLOv;vL_f;+rT=ZK9(A*_(e|~Y$(KA3#g^#o9wP;9
z71bw$C1K!YDBMAK5@g-@v)hj|n{4|H8HU-)8EV13Cr)q329-(w>}pFPiU_|C{L>H?
z>sq1SN#6;;C<_t9=UzW73J7^18F|{F!-CH)n34Et)e#W0S9o@Y;Ii~R@}%tp)+=TZ
zmy~QV@Cfoi8UcvMVQC4>A@_23Fozqv8q+0!&tdJpteB|=)yxTi7h+m{Hkf-RS0e}J
z46IT<RqH`)TNB{huJ3!uK=i0EGm0J5s&elfrX@IZ$YlopG0kX@)!DAW_x&LFvXnjj
zS9li@(p|@y#&x};VNfV&qLyKiahQHLg3cx0G@I^H7BCSx#K{{d&I@SUPj{zi(M;U!
z-vZ8r;`s&02)Qy=q*%UmKZ=?kVn^En+{M<Km!n;g3g(4neo2#r4F0(9eI27F#8P7s
zowz~yu+>^&#ZX_B_ml5VG?hx9XNBZGOzOHaO&VjPK{5d=nmP1;yhU1WZ@Ok%f`qvb
zH0e&~eJsT_jg1h{CHQgDmJEg0sy~rLwj1T!alpxBfLB2~#0?+I9VC>1eV5xJ_Q5eH
z4HRFJJtdkYnw<bnLjuu*>Joz4(@Q-e)&iM$&R%Vgv(+!~^2wf1kqT+!fM!bH-xqr?
zluAKwKe~E0&f1wz&jVxzYM<cQ2JTBcESWKum(^Dd#scc0>25}jEWEMDMWyX%#$5zz
z%KNnQ_bErx5I?}0fS=Rn(jznhA<d%7wNxb&EQETWg6&~J=2O~BTh~MCn0VBUdIIV@
z|2|z(D^NLAzQ~G74$%K>p>3yaEos~jRCyO**!5I>Gy<;7j%vAGDTst!ifKB}aLo4b
ziIO>3cS+Au6QXqJ>XPgp6#y-s8@m|B;@wVLFFD^{5zniy=oN4&hDyd|7J2WfOC^X`
zIc~*^vTr7e%`6nCv2441jEbJxfTA^pAX3QvMfKB6UN!J*D?@~M=o^%5v>aPD2mb!v
z5llq1l8tO(s$RUfz%$WJ`{|&b5K?mvqKKxuF($Y|1TcCT+*}Y#nT4`zaw(n40>A1q
z8EY40xnRnscPnq|2X`Vx=f8C`QmgSa{`RDVs1YoSHdMC~TT(h{UcGZbW9$0?d}XD+
zL>ci{*P%`Zk%j`(LyrQFAo_UM`}TNv(zrx1NpbRm92HtLx<i+2t2nms{9@HFTRc2C
zs%{PkF1!uo{^H2Fcx$Xbav(Zlspk0efU9V$5csQ;xErsG)TFE?7bdo9|89!1tX^M!
z=m`8@9rmRz?WIYGeR@2VQRwcJ3jsfk2V&WbE@t$7BIg`xrOYz^{yTgKX#7nZz3PVF
z!sfw<X&-pEwiUX{@0x5KG2jTP=~w}2nP;h5D0YG64AeDRd+JBO!*~0@3Fdj3oG%Nl
z2Q<;td1zLB>!QX1^7=1Fj308u_QLbppw@*Y2{l0TQ}NO;>W^><Ep#nR41nej_~SSx
zJW&56<~{!C=v2N3?Tpr&l)OQ_)*;S_O-F^#&RvN7zMy%_LB9BllEQOpXeq&ar!#<;
z_@HO2{IRnCluPkGeyE@fz2stE+N|vNU_innlrKrEZn{D&lxhME2D0`mmESxLI&xqz
zViw`Hnn61qacOS`ga5nWk>=l<fmjwikbzoo0}Z65^;U&if$n0MtMPj&!b6TntC%to
zw0lxj5sH)?vQ{Qsy6h7FtSUk<UKI*Te{tpMAq=X5CmenZW)}Yr-U&lD%g??F95>$)
zg&3c;Bh_oUsxndcx2a5Z1yAfIn-?YDkh0NxOZDvm7c0rdEh|aMh5={vrhPZp<3{l0
zU{K~3<UM<1QiNsiB)@4#%~pj5=l0zFt#^Mns}cv`;VtdT#+9J~H9_{KGI9i7LpCAw
zAV+ch(1YKu%@Po^*&Z7y|LYH_!(zaSQR9jH(>(_8MoT37q?-ja<0aGtjWT!bXS4Lb
zAF-nauOfBPhcQ8~`s^cA1U>O&qX6^ITxx=wdr?&4-zlaC3PsSWl(xEK|9e%UK>-18
zmXP=xaS^cxVJcw3Xhkw)JN$k36>`LjHQ09?oktmWz4=;jUG>8vw_*mE<nnJE1FSw!
z@!!Bmn&=>K+tsQ)zxon!gCtd99AJX(J2w3-kOIDL1)B7QXDWar6oMdlgHn$;P)JN`
z%c5ul@#}M39*(^h5J&oRjy+2y0}4XL{>dSbv+vDJRSFKx1b}PddNG`}V%(l7^7m??
zK;Xp-6XSWEnlA?}uzKHCpo?L`4C2<Ed+94=paQ`d<XU7IGyZduQ;wh@ly7jq%in@_
zmp3zo(7Sv9ccq?r@wzuEoTMuw_|JD^(6HM3Y^xpp^$SuNy?&+|&K1r^a7s%zMN-$i
z7eW8?EE6Tg<SL=}r0LBDj6nH#pLcZ5ZcNtrix?3XZ3_^E+RNO_`)$Z65G=*Pm?Y@1
za;g13CjQ)vD3XB0%-ESp$?p~MS2PCN`UyB7R%1*gGx%SZAg~Oc0~5iM6neul_~#F9
zK2+X=R;EMSxc2YmabtFFR<|A4O8>7{jXe4LuQq59K$JhM?tH6Y5Xjd|%g(F56iYQi
zM8>>tA}4V63exSVF2vybZ6{WQg~rfVg%$0gxtF_fbs${r-ch~?7;!rU{+&{vIuNcH
zLpS~1NBl%D?HF_q@9`%%8>$GvCseC((77b^j>dHYkT)$Ck0kBC$7vtP74cvN=7YEf
zg(JeE+kxz!{mu9GAs)^b3_PcX_Ph{cpu7-UeOHUPSsIERwKrXMY<l4$;HF>;Vl00?
z?O+KypPSSQEndKFXkH>D@pt935Iop8OiN;l5#T#M#yxSix#wqDJ>6{qd^)4qDoESK
z0RaRUO7Q1rer`VhYh%2_4coR6X89O@V_N?#3Fw0Y2^jmTw_}R``t8kWjszLxR3h=g
z|G7Ki<OTynJ5#$!`@a(g-j%L3f}Uag$DjFozj~pJLrcnX&hC#T`Lhg!grQz-_$vOV
z@cpk*0zb_jf@W}A@k#y9I!sLkrkL=s{f}+_%L%#pSp^oN76}jO|E0$YU`e67=#Ml0
zZ{ufxWh)9!2>IV0(?TW<sfpDr`hOR_C1}P(x2(kf_831BjMF>ayIj)$+jwr!4D;pe
zj6WUy^8rl?jMM-3=@NRb7y9&=HAcrlqgLa0cH#Gk!h_z!V;2Pq;h=(u3(*<~vav#(
zjzEA*A15IfQi1q0%^!sH!7wrnWGRR~qagAA+TkxfJ?JNu^{0Q}Lw#iHxXz5Vy%yq3
zU@QI$8U0yEfCdUo`jiLVRKvWu{^#@Hb2G3XC$5B=xZdq}LSJ<ZgSjmDvvvrtZ1Vo4
zP6q6pSdG|{y(UhsU+YS5gyma-q;jC?wb)m#h4C!vpLS`37)(LKno>bV(la<CN1*Lc
zEnU9jFdC)u@}KYi&xY279vjV+oG70K9&80jomAyg^MPu`fAk(WWFXeFu9-_ssetDV
z%F1U4z<t?aD-EKm%;vNEcNEVK7YuBY9TzO7n97TUqt6Ni)dm2LR6ldaVbwBKF6bZM
z>;#sX3oFpj+m!;%09;|8%4ds-6m}u($?t!kqrif@5g;=m^`^^79HnmnhhGh2Ls7#W
zzKVY|R0tlLqgF8myk`_F2(%4=6N_7Jx(VEE5ArpS{@I}qLbKSwsB_0iVGwv*ASRYW
zWOQ*LAia<X#7d?ymFfQTdeDvyY%r+&D<b(M^TJ;0bBy02=bPOxAiCRLZ4evLz<hDz
z7X4}Q1`722H}-tIc@aGjd5&w~gDJm=gUC<n*|12sWuoT?R+Rs##+Nz|jD_V1L+){q
z5E}gJ>7>`m&V@#5Rs(dUmH*slsfM77yRp)FAs>Vi;rjMKdB1t)PWPEf!9T}g9VfQ_
zpgxn#p5GfVv-U$t7~}T$sK?g8H_DQaKD+qtc|EHd+kAPfHM>yAR<FOFSPs#gYxOf<
zfa=TP(%6arxCM|$_z-&QhyJOgBqwlR9E@c%93Zi7)K_1#2BCXUAJa^CYZP1wI~F1N
zE`{?`3!Vo-HelM#thg_AUnkiQ05^i#sKz+(UlN3=q$rh(vmX2kDkLN#OCSKr3N~wO
z9nh&87?T&YY`!|*9(B%6a?s=eIktFTLe0+uK+DUx``&9#0BNYW+=dmX<!RTj=;4S@
zH-Ru=VYzgBJ}4b&nk~D@Uhv}RQ9xJZ{yc7#Tky%VcgE#FMf!1sk;j303ebET1ef2}
zYbl-{7p{*C{yHlbftz^o6rG76tQER@lezs6J~&+tO@J6PtFe)yoyMkLr_2v<&st>5
zrwAPPw0I4&yozu#CFV2<Q*rPIuG%cEQB7EkR4}f0^~2VKb5N;LHN0MJ?j}qkQ|<mF
zibgj66_CK%XyWX+!=YY6?m-Uec5y&TRiNyQs8a1OW6YTh)_ga|#Jzl~AAmtQDsmRe
z7w{Oa`RM5T66}2D3M|ePKu~hnUe3t?Q}qo|p)WhZiVw=UI>B2GMzunDk?v4(Yk{*B
zybCC`M(iOf@$s(yu#e+dsK?>Mij{C_%`T#nM>i^$cIaTs&f8?orl!<VzXTG1#}e^t
z?DS1QgiSMsQQLz03^%Mifkhs6<G2peEQ#2)AEd)A<8+bV>IT8zN}vN)0(d@;YwQ7L
z0|jfW9$Joe_~2Vf?RjS_I3V38D9AL(P||2nR0AB*TvBhEHPumd&<T=$wW=VY8l0s#
zO4FzCIdCi>)RonpdKljL={`oyoK4W6c^PnnNUloHZ^GHBW6*M}WS9}Zd=Rn?;2dBz
z9DHmu=*QX}m1eQTv#3<$nQ6{ba+tls&A&yda2KJLSvpWsFNAlE^%oBEAqcL<LY083
zUz>h*lUS;cK$H~+JSL^-v6J#e;GE2bMt>`A^8+8M8^A2C0cydDvANVG=H}CH#(7!M
z>|lAiCvk2603QGJACLd|p^yo-=K$R`5XGBm0m3gZKP<~Cn5uEp(L}|qk=;p-svA(w
zZ=-L90RQbhq=<I_)>IrIHlJO&tMn!TPN0J>vAq9Kwy!%8J~j|CTS5tVdgo8eIU)+p
zKy=wpAssB<o*Q{Vup4V6yJCl)-`tFMA=OKM-70R2gU$fF-OzQwm29n(Q$1yX&Od2M
z;*gb*@Ufg9C^yVX&^3^{=fQX(Ge?UP3uG7vfkcV%=TB_L19#28Hf^K_4S?XC^)Vm-
zsQ~qaF@0vh4Fkb?UobKbbxM|VL;mC*QxM>34TMO`1*2=}r30mxuFD)B{ZNy+Chw6)
zzohrYHhPIGA{i|bo$31{V#l}qSNZn~Ig#>TKPrM~FNnn`%|8?ZGKv1^a^?5Rp5G5H
zlFm|GTWr6d>^ZJ2OgcYlp0BfgJ(TD?RP#M)I5Aj{r+(5e{DBu7LKptX;x*T#_Slul
z68)gQAYj?Pm@o2*5vB~vlV3ogP%wc<m=_5Oe@T(*d<mFxkvTv<vI6AUrf3v6Utl^>
zHVKuIZ1#(O#TJ6|PDiI*(g7t6gFs}}{MLf^t=sXCbx(fkO6)EX20f9Bj=dp%iPlx=
z7GLun*hen2@<tHkVYe5)Q(~Bf)DT3(!W~4&B;`!KHCaL)l)w`lnNK4Bl+)}>y6<HR
zELGSiay@GMwu=In^ZqM-zJhp;e0g^u4Qe|Wa)~WO;fHYEkgFI!2{aH7z-EsFNf4M!
zENfdL5bHFaHi`-UdQb5cB6cnf=#*Lg3^m^*ByXqfvYEwgva!(Aup6P3E@~T#fUc}m
z>I&R~go+_PhN3X5IeejFkny<D-rSNdZZ=nZo*@y+4bf(O(T?>{nuHsS%A?S|_4yD0
zZA`r4v_}Cki8a_APY!@hhjbqE#R>UBtXxzu?`26tTUZ0E=N$~2`rY(%jk$hu6}3JB
zd2cA(Ro!<E5~w>q)gYxcln641#u>rff$OM>W}w|@-X?QxDoFPBi7T~lmj@s!w0KTa
z1BXR-$9HpRwL8pHIsgu_Ycc>8uQLwQ>sS`^KY$s4&~r`o+PBXSLphW9?Af7I=0&YV
z2|(zTeF-+Q2`fOLTT&L}lf7*N@Jb1wJd-KMX6t${gn!Qj5+)j<P>ajw)@VBfas>Kh
zE*ohh%Wfe3m+g6d#l3U|L^VZf0`~hr9d%!IB-0f_JemSl>Cwp{#BLAtP4#}s2XcU^
z4<twxuQ6b;+?2QI{D5;TEzh%%#P1j)#xXL=&~@;Ec}iuIL7&qYHdk#rM)TExR%pCZ
z4Q)+(v>aLQ4CF2cf#uy(jrny1(gxgU?Vvo)l)3^mb$Xvgqr7v-9_*jjGsfL@gff>K
zVDPWs&vV6O$1@htNWfRevxSoKb5Dt19w{3(0N>&Ec4caIaE5rWlsZuk&s!GH#6mjr
zCrF1pxoRc4;xE2zy1uMx1o8)dmKb=$9O>g=8{{D%)<?M*NjvG*PXh*&f<X}de|3SL
zFVCE}kEjg*FO6@p92y*U&2k#h8l=#iz}UXE1-|fYu3WSI#;<-P@ervUP)aEovm<BD
z`|fI^uqDucbOZ!MQ5|53@zABrbfM9?bcjhNfqM{=DXyN-+8!5cyJW7lbP|K}?jstR
ztJkTwV|$S;!v<yWc>x>J?xGKYs(RYKd@{!~IPWQx5j$S)82WgmOJJ`Cq2g8_?W6gJ
zpM%%z>6P>Fcj%|a+=9^FfIx_DZsEg-(DM!YFgGx8xbX`NKbX{Ps_OWf&wB~^o&vp1
zVW>*_g-KWe2L8^XO+QmxHF2W&rHdpOj3flJdlmi64Y9noYq5}db^|UC47E+;p_dR9
zQtp@F6{7%&H2jtP(2#gX`ct}Y45uel`6ZF@=o+?+`GhSX9!h{(3Py4EEJej9FH;je
zm^<I2LCUN*6+RtkEQwR8L|g(8RG$g<L#~r~)10K8(D_r(Yg#>+p5>Np!8K3C>tXI-
z5}s=K6@Uv;Gt`TLl2Resdy^Z@z{U!{$un)RGt5hniBU#A`lOb!1xe%<$(`k|&3NSR
zUYEtw_jhBcW_i?tL$o}O*ESo9KMw#qeD~D}AT3i*$(Bus2e$$NLk`$|;GWE{jg=Wn
zK!E~e=W}$IyFbC}W@sYvb<m{i<;)E7%yZby@ilZ9jyR6ZH#q28&Z+b7)fGEe7!N`Y
z4xYpF3(sXRBQ+^i=95D%o3hL;95WIvu4ZKUYs&zD6JuFj+#K#wdN-<VROErW1_8qP
zb%{rVi(a<g2>|hx9sE%3>a`6=p};ov=JWInU=FZDrPF?e1QJL0U6->+WR74DRs35)
zve?4<7?OF6h4?v;dAlATfMC_MaPY52E}L>>-hS%5Y^6buC?E^_JyJF%;y_hz{d_my
zIhmwRLpG6XI5Yg+<vkALyGUss^}uDceDM<L5xAaISbqCHIdnd220=TxftTeCfW<~T
z62xCPE#T;DQp}Woc~n?6>B2%!fERtBmZg0>{8RAyER^&){c5+TXR1I`Jf8(P{%bCs
z2nx-^*n``Nqw()DsP~V{j6H$`LAV*JhyneDgnG4baKrf>xg@@-ZV1uBz9sz&9O<xA
z8HU4=Z-7L+rYR&Y^5vu)7FXmUw*IfvpANid<T{QReL0Xm9z?D+y(%-j_*5IJH6MUz
z!Ef0o2#*T7prqFIttQBiF1{g(Wgve__2s)Qvm_Jw;g1Z*?+bN1Ljr7sBiC3@-IPYZ
zw`gA7oK-exBX)I(=WI`_lY=(`ip+?VrA5O?#?a$L%cKcb))au|S9<DH77&bsG0`;A
z-<8I+sqz*VQB{J=g_}UCBfcZe<J}A22OM@IKt09Y5PJ2c<Gmf-Jg3N(XpHH;xhNoZ
zr-jsAX5!bChlN!la&h8SwDKs_=w{2ujiXJcUs&c)C|AjNgcW#ArChwIYDA;A*C}W6
z%wIOHZ>~O28*D*7dSO{9P2aL6=O8KDV^4ahM?o(+Px2f_mwK%Q0aq_uGADFBwnMh?
zzZT*MrOp8>w%Uxq_N3V?fHSEVNM0Bvz;*w?r?tUEfZbgQ?1e3vYP`V8l1AfG&Zq*k
zOtC(ge4kD92)X1{6_&#hH0I&xO;nk3%qY3q%cQ}kPbp%8xw;n<>kO`MhsWPGOY+!j
zJdnG{_Zp+-I)l9iK>+&#Gk2BX405}IJ7}Y@Yq>GTc>pCqg*R09bH6yQ>Uj#ZbbUf~
z&NwHn^YM*^XugpA<^qT|o0XX?sD>n|F}c-t3&@^**2LOSb}dL%l&m^Z)3J~0t@_m_
zY@?0B@A*{kF+kqDyZx}f8DOK*nOe+rk6hDcw5gFcUBbe*(w>EM8So(5;Ewo|0B<DT
z4})YzQ>SHCOzvJe9s&wQn7`2FlVhVvdi)WLl5>o$ZpBMZNf?Mb3l|p`qkpJ;;<%^X
z^q!{5$_a41nYm@_4QGjB@^;?2%i7vhRnqrTR!4<DM@Z8<=sc*}YMxKO!Jt|31&qzG
zI~FL)Qlz04AQ%r86p&LRb<Ee>SIAazv8puFMQJpjne7=)`m1xLZb=uBSX0Bo&CT$6
zoI)m`L2*!MQaW*zwZA%ls>E7AWg9ij_1baUPIf#~tEz5B^&oBwUb+qs-m$Jk=qz+G
zRrnUaP*gYiV~|?i5#<p#WWKgfLqt)|fa9u%Il*k7yq^ECjcfzr0&L)t=^Rp5sx^m*
z?_>&>Y;e*WoFBC4rcRirlz)|H=8=D8c6o6k&r4JiKX`sQO~|Mow#6!@RRV9sd0kXb
zU1_~2aQGOF4Cbi1p6SfNMWG`Ax+36g_Ht5Qp&#TAwXs*WnxymJs*$gQ6jet^xen%m
z@_NZ;k#=|m`AYo}v9h##w~iD{ZPpzCjI!_*28*ve^Iq1+vlT=kp~53#(^i2-XAy>O
z{^a!iu+|6L$RQ?;RwhM&k*#deH^d7wT@1P&Ey#P|z4mN6uke#?$RZZw;!x!<J|Z}}
z&@i)Gw4stM%FOONhx~?JJ!xVMt;=Wbx1(fl`Isi16EJM_QD?FDQ?q!I2P^D^Z$H{y
zXi$rC22RPIrH1Z{q!vN!ad3R=OYCDZq`pImne#H;ds>1LsU?&#ynUTQ)^0)}WIy}a
zc*n~oCS@5MdB2cs8u$aQoo$gdP&Tr)s-8&2v-^VfGJjU#s5A&EzHdJW;dp2b@`coV
zN=*>I*-4(|TIZhEn7gFeRdsCc)4^WH3rIX%um57pl^YS?f9_+niL6@mWRA9NI+6*m
z!ealR=IBD{!iVQ<Hqw<BcDDV!h`ch{ZdCFf0oR%EGjaYh90V~!g~>Zra9=9TiFuZQ
zfDGe4PA?f4%~MbKc;@RCBxyQCPIq7ANMIvLMH1k$r-ET@gcP4IuY=Za*T;KuG3HHM
z*K?YTQ1DxCLurTwaHl60E{;cI_(3NzIxZ0(G&%SsKL5^EPXT-~ey%kT%xp6zuBDGr
zPS9X>HcRnEC!*QOzs;p6lQ;B|YqO5(GUdqJ$+p+W9h=tAiwrxzAz}>%>0bxFMcID_
z2ahNMJOvCR{VY6%3NZ|UERxyWiI27A`<i#*-_2b+Dw8W;lE0B`)2)`Qs)Ya@r3qH&
z<G)_|@R|zMKyp~fb@AQLA8tqU+3(K4kn8wBY03hLOh9D*{doRU5NW_dhV7e^zu1i!
zKJN-5^507~YdYsIAe6jiN;jCmGt6l^O2rxKrkZW8E8uzcE3rb3tWu)IGmaa>@s`!9
z+Z%^xsm~GO@wBnV$6>-g8fJx9xWG^-Sb&>@0ea74PoyWr7^+-YdB`~dKQ|Q*u|cDs
zMM8IBRQDGbhno?#!pPr$DaLT+m&JegMPlBn*jB2i=(Bv0t2XFUG^3pb<(ejh)Mkr7
zdfSCgb6-MGj%x{4s8OLr_7aOqJGFqOslFFeI-Xt@YahU6Fd*^asm;jtYTNB!yaoU{
zE?-mMUC_<>nQL>7i#0C3reQo^g`p<)M?OvnOh(Jwz24ai&*2*ef)Vb>EdkhuF1X}~
zU!OR!zQ{&z)5poB1gk>rOV`k`GMGu43=*rOk_RfyaKcvKxURspdtj-}$0rKh8(vPp
zC(Go>CJ?T?1r$d9$aP!9OYdYY3)ed9vMo+vZd1&rc9vYS#Y|6up>CteLslL|o{sJ^
z+T!7ngPkbDo8_-eHiq8kWrrC%u+k5f6j*X%tJz2~w+M!XtVzgO(G*=)g|0mxdUTeZ
z4=){MM><n&X|7pqVaiF)g6d6+zM=oIUI=8^<c)z)jUqF>ktm!u8V<!sQ5d?$#@Gl-
zX}sKX+2{@{ICS?X`wS&;!h6>j+cZ0X{+W}M&+ha(%9DlEb*BvVJBM?Gw(SMb%buK)
z<%%$?-pR46^1X7fnl8Dgc+hI01cGeuuG-F@j?6(KOa-V3SSsF0YhB9HSTV}dyN?8)
z4^|n!@J^71!YHKhGfk7*P+Y73Yso<j*UK_#NOG*b1Gz?*Yqxm1D*jW$(qo-7E*Owv
zk9w8$Y2%Wr!Z3ul+hnQOZ4_)k!kywTMVD&_d$*8g0Ai)gwO`2F8xdWRP+`oi3VUpB
z7@WNpkwH?^*g~lYh$i*g@I+-JIjNf3TnuJ2m3h0;FE)PlvS>nw_7-x%>*3@!-;wW)
zYl@e^=eK+@vgC8k1a8+)Q4i*5Jd?MY*20ysSIz8MYL^f|oMZ9sQILvIQG5%yWHgBD
zt?(GiMb)${eD}$xsm<q76s!fV>f$1!oyPSrSNMLb_bs2{VZLofsOWT-rp0omlI>^<
zRv~Muvb__JXl^(d0ih;2rsUDDQZ<Q-wAwN!=!Rt?)~f9@Dz==6&or=ZjkjpLV0#LR
z0=bY6_YWpS0|ADyKLoev$SjKTf~k6{Jvk^|K3}6=`s#CcRn$wPG3|D#{3s>X7C<jP
zyPbdLsv??bt%NrW;Cj=}!BH_oFDE4y(6W0#SS}&1Va6)Txz!8)j4b*Rhc>{g9(Ut=
zswRWo)_Cv6cdDBM#ShNgicQ%tIR`lt@DrLj{A#vg6J!b`r&I_QF1~BGKzvzSojJpg
zL6rsfa56Km?iZa4erp}Y&}!Z?EvpJ#MyB4-?sB!{$qcv!@T+B2D=x}i=5mSCco!cl
zV&}m}$^pggrYcEsGgw+msomV$o;awevwIPxg|2dcL{D34MqsSv(Y``!G31g=d5l>V
zjh*EkbHVJDEpx;NR0ltWBg~iDfm~q%M7MTIt8*3usLr&C0i65^$|)2LJzUO#c#Ic1
zFSg5dR4bg3zTiEQt86gqw7Psx&DC^l-qumAPl}+3J2Lm}+%uC9{d`tP1X$U?2SQ`3
z2D#mj<uNSM;Y~Ry4Gz`@4c%z)o4DbF#R~-D&<OMOleJ+|6Vb|sxtfAV0{K;dndU?Z
zxz1xGn%Npg69VMi3V}1pg;^5+nCnu7DRuFV1HhZ0&Yt(|$IOdMklIHo;=K~g3w_Hi
zIp|J+pFeF;-?lzeW&S#W8!9$S^DfB7`QoMJF__!O=6-(oes?U+)K#`OlV+tjuXKWr
z?j<aL>$`G3ol9lXl}9j;gGzkceUGazeb~MT=ElnlC#UAIn?o}M0B03dwJzTu=@=p%
zd+g>fy3MyZR3)_xS%O$zwX;pRH-ENS(UHP4;ajNQk5O}}12ltjOPU0KXeJJMms5HP
z9U^W0n!jTvc^D*#Dovg0C2@(OlJIPCPAM<u>M@!PY(R<Ut0IPtKL&~M`N{VbuZ)~)
zuti+k>uVdo*gfWWJybau$rWCjO*m@Oa`n9VEV&<`+RP5rQgnhp77Aq5(QvI#U~E8e
zdWl0FfZ?(Xi(#%k49yab8@JF{FZz^pxdyv(h+`}WC^oaRmn){Tr@1(r&CSb<4`Y}b
zr|s@zNOGrHRkL-BlD1vkf|msM$6um^b{ahjXDZUgizI`ag~V$jnan!%Qy#<2EJz0&
z#shDxD_m4$&jGG5QYGkpA!chg!Nb%RwtN_@gTX#V>g2l2KrS9n7|uH!y3S<IeG!O_
z@6B@%s=|#TUZKZU2>Z0FC#YC!Q;(`{6s5AyEOr0gT%$buyL+^)N?z$H2GmFPFYq=*
zs9=xtJ<B|7Bwap?*4uAuI~0`x($J#_d8M%hi}({`C6A{H3SL@)nG`Mwmeq7!X^)W+
zsBuOX?^dM(StX(YNR8$K7PHcN9}ylgU))Qm#*Ep1_t80sNlgWFPrr(o;R!h<%)eCF
zjf*$9KAyEM!?22ovwsmLPazrx@WQT$Q25XWwXkk2{usU(ITn-y&3_d1Rjcdkd<$+i
zHi#&84C|U}_B3XgP~zpD`zF;%Q9r{n?p|Tt^P}Z=TBP-2L4@MAO3c83i_5a8Fr=^6
z2JfGO9X7?Q6S00Y;5Ug>w-r&~ZN+Z+E`E@0iV!qZONdGS&{4|MU(y#xVKC>)svs+B
zExo8Ra}TBddZck=%TW&6WN6evsamwXvFjP$I(D|B)t0kUNdapZu+OEo>qAxUZsXT-
zgQzw04q7A|-Fxpv&<dNqr0N{<PJFawNg~YNB!r1ppaECyv%1h_JM8JJovE^W-vP=5
zopvLtEMZK;eGqnAU`TQm$|vkDnIqP}Z4Qql_1-`Q!PM{Zky+1(#gt!i=O2BP9!<=F
zYK_joF}E9Y1=Ph}fys)VOyyA&JkdtpruZCuu9D@3q8!o+-HjbU>31J*8cJdSnMeVD
z$lAihNwV(IRR@+}r68IImxoVt2bZD9{e){XtN9pRmtK<)+d+=MI{W~17^N$AJxx!s
zO~D_N)WDVoic5hzdwB-3Iejc(mCrG_J+jXtRn1k8Sl3L}f8#vCIL*-Lm3|^|UcEP#
z9MSFvx0Mk0jRLxM&YnsH2~WO)Z!t724^8lYf)L)~696nltUpsM=RrVm0ykGO@hWDs
zPSK-lx;2wh-%8kvMQ}<E3|YH?OIdfd{99(4>G6#%n&&T2HfKh)d5J8Kg~8mUGo%VM
z0H|ohr6L`d`Mb592~j=Ru@yi>q#Za){QP)0pWR#!MQD>+o`lTjrf*2aVdpvkH;?`r
zV}z3*fNTA}jE(0$lEujEUFBjO1E>qb10>z&lQQOZ>%%G5-5>86CfMPv5bmRe5djww
zTh<%IFXBQ-`DbWaAviyTmvpC{tFv8K=X|YNVP3^b0u+=Pd(Z>ctdX)rQtOenR8KZ-
zk6jQN-+UFBEqT;OOx$(=)oMMI(*@CVm3u2&=dYA93WV#DUAE{*R?O^-kdp)Vt+|0O
z1lHc8)bAfatZHEn2$8OV{Zu(0&KX~Jek}epFR#kjV`M2DMAQRa-qMjHpmYj(3GpBk
z<;99AR~Q}T1@0?zK%q2Z#sZ`MXvo9*#B9!16187kB(BC)2PykCWrLYFRna22&<qoV
zx~Bq<6#UfXkQGD+BS6Hm&iyoR^CUePe<L3jeNGkuZBsWS6uSe0k3X+m49-EQY5Ef}
zM;;wxPhNgq`8#mZ=z^+>Y}2vqRa$YzkUR><!;LT1^2LRmXpf-7S}4^8X*^Vo$(1vu
zvHmLe+R743raupsE`~l<sxO-1|APuVL-7(!kQ<U=<26X)?-C^^AB(e@N+HQ|rRL!Q
zh?j_#^t~<{@iZA?9SwIIt{?cqR`|szMnceXc#yj32hvXl{KVDGCtOmt$`NGs=6W*0
zWV>sjp=KI%U=E+X1fz65nQKBlPLv!@lNhQwiud}dyDHNvC}I}feHD+MNdbnZKJLM1
z*;`!_C=^~oQD}@+<4VWgAPr`1A@K^p_(LBuq@8)ezRBEr@#$tzfQ35v5_Hx)mMzQC
z&sZTtwVrYbg6AKRQ-Q!=#*Uj8TMo14Po1y=j|s#&F+&<dXe=lZhk)zF@8Dysbqc6!
zA3DpKH5UqW@9Dcyp-ZNLfgl7MqpViOmnG9ohHF?C|Ke?a?>>}&nW?e1uD9D%I*lj^
zyN_zaup(kZ47D@>as0KKj`uji20+AC6lF`<ZG6Pa*pq@ZJmdK6&AKm2AN(Ko-a0DE
zHGCVDkwyd+3_ut~qy!Zh5J4D=6d1Zg1%#oLlny&kX=#+sp*u}#EEpO=rBg}ibH5;N
zxBHwwzO~L-XRYu1{j(Q)oO$2pectD}pF6JWx{>#M7#qi$q((8SaXHOL=>CcWb06NT
z2Z%}ev{~nDljWX7MzXpi6(o{i997qPvYZ`2QAcG1&u+!NixLm=Dh((<$wce>lbUhp
z&7dB0egY7diWY`bErSDPbhVVGw@ZwffOfzv4|8UQ*|qpfqngtPAfHYQX>gN2aRM2F
zzOVCmM>zBbjMU9D4~N9gL3=Pb1_JMAdLwUcK~eYdVdEmx4I1a6EI&E6??~11RKUqy
z^it1EvLMAAz^7e8_buthPoT{Cx!Mm0=})A;_^|`l=TBNT#}B8Iu6=vseDfOwhq5E*
z45Yhx)sxPp88A_jk$(kq&Po{7%>LvB?(!U6mdU=|Xig=r@8DfWLcWrCPGHN?>S5+X
zbPE=bPWyHFT@9px@!|?lBZM-@<VV6@?+DdU^}|VbBQ+gxmN*q&KR*JGBn<2JsbTgR
zW^GOuG0RU0xOV14v0#1b2b$8g34VRr^5Mlv4U1p7Ycmu08%oeH!Tr11z?C153%0d>
zI2DpG5P;!7ry+YgOINQ2UMCamwfO6r3b?}*rAm>a0yP0b(l;q%a8+GU7&tH;oo1>y
z`}P!AQX2d8J~xL5WC`$+8)!d+c%4PeS(f)vWZiM5p?BfJZ|6uvQ};r|W(LAVgU+Y*
zR=A&GNyKWe+-Qz+0}<;?KmI43ya-exAZj~bQQpO)>!&ggHR5ZF?rM2Cz{BP8cnaoQ
zCRHbxf{gsqA(+Vc2bw9*S0REE<)58>sIZ8fO&U~47E0s(IdRw;u<281$%y)dJG~JQ
zBi2!rx>2v6GWHPX5zh+@EG>gWFt-gU35E-YpK>o0XmA_N+!HRO%7_dA&gs)8(Ej!q
zUbp{E{oNPhN6W@?3-^+jsur^Zllfk!g#|cW+OLWCx6NpV5@PujZ{IsUbx6f=w%Qdx
z8I&`ENrB~bPSpgccEWBy_ro*CpW!djvQTx5zW{H|2Hu*5aqRQpcNC?nX&+aB@v`q`
z8jp751z1ine7U1>WDNR6wl-c+K$~F={i)uaT&!imB_(nP30D~*T=jYIbewk?wy<*;
z4@D_$Ww?Y`Cx6}C#+{jDOvVdZMJG;Qzi8hm9fk<H!i6{%uAJKr%PIu6xaRft?mgj^
z*yn*>E3?|D?i7P<^}aA>G|mY%vMaqs6RP?l2}?{a?{|g>sct7{y!#kp9*-=|8ikQc
zk>@*y1ffbMs&r|!%g^wV5aP3d?cgU9tIuJVl;r~W$9ovBD(|j-WPAMDu^9=ri(BYd
znQkQ{zom7CV_=wv>5kC)R3W#u4gSV6drxx@KdL)zNwe?f4xXo2k%Q5G(cAYSc46bm
zO!>2can^459L|USeZIDZ+JWOcu+`wXoTM?@m^`>6AM=OW28EGi){t}lwi;RF{aADi
zJoJxp+h7)Cr0c`o81pT7EZV(VoWJj|KAu<VV6kIb{nh^ZJOQR&iR>lRPXD;Zmdz!B
z)KvfVITPHJ*F?vEbMza}%BTXb|GJTw!e5_1IRiJP*p|Ne7{5mPl@}KM`%#(Hzdl!k
zX-)FEi1{@$8`t*1O2S_I=Gx8k_vf^5Q>J4XGXFd~^4HyDcz#I&YvNy@%Yj1{iMmbN
znDt-R-h%qUg#O?^=oAkn%zsT7FC0ed)!Bbf?*Chpo)UZM6f!NowKR|;N86h{$Nn>b
zkGtYz#N1kM?)+z!tPidac?2r<G6>XukV9Ept-P^r9~rqCt%!BovHmL_8xMdrfT4f0
zPvqf7X&(NNcSWTMb~K-amGU0nDLTi|mdxc_apW5}>*K|8bW4)8?}sRu^FxX%rvr3<
z4F`@_#L=qd+!Is>w8<H%$!APb!AT#0lX!be&gMOkcjDI1BZksMX9FIJWd*4SZLY*x
z8K0zvc8eT=zusgY{EE~29uNAj&qZNX{m<9`|CpKCy!NLy|7^qOT*u7nnewiFc~MS_
zv1&O|=G|TTF?FkMuIar6fGjqG+Z5SaVmW9|j0nfH=x|PY)WS>u$_e{9*6O_c=7oI8
zeW8_e`?mcjMJ^gj*^RV5ewX>IUI=PL4mA{e`WzH$Dgl0{lii3yQEOCU9akeXD=}B@
zZW<LP4HJ;m6sC3NR*td&6@P&|sI?A8c~`j2UVz+;!{@3G63%d-=Hp3xgVuWl^h^=j
zq_Fp@hXbe?4MZ_Wzz+G3)i9xWCpmpV7^<MGV|=*ot1s#K&&Xg(1@!W3hYfZgHL^x9
zY&W_POpa}LX@4+-!~q|y2gUQI9h{Z7gV{FMHcuoLAR-U_x&?~5K2er|ugb5l`~0fO
zo<8vDni%_JzTC%AlGnqO=(D>#y8ru`F8G+tR|S)oP%MQU<Rrgz6sBxRv6kky)p?`g
zo~`(#VGTvj_#L*1Cw=Je+1yK~VFmxj`psDH{v&JvZ0`r;iiMl@0xM3QCeX%JEAZBC
z-`jKOMc7Lpt7L*1HwRp+fEI6C0e3sELt2n^SKpg{1yyBo1sH(7o5Q1jCf%@_q6~Ub
zi6K11)O?dLbkR`N8MZCHI)F%evpA)Cee)<l7x?&fBd{<8G|$XDwzDs(a=^oH3u=uS
z{h?osZ1ct`CcY3`YiYpH^=qm!qF7KAU~C1@hjoA3HC3jRV7Qam{VKy?8ea?WLu95)
zaeSM1ii<-5XoSX|ZgCWAPOVW6mi@KyCFS3b_LG1$xMApL;B{jM7OfA*f%WS)q@uHK
ziSFiL;VAd(+Pjvk>PTMC1J5htvHkZ66?_&gSR;7aR@_VC#9>qj60YxS4aoa##Bz7u
zb%T*dlKM>{&XC&)$0X{@Z-a&tK&jHh$kO}8o=l&MLxS(ud0baE2acXXsnWqebFT59
zCEIZqIg*bE?z)8(b=C)WpPMlRnc?fLPos9Eh{DM6WnZI?Zkk2LgSZJ~TKb-6L)OE7
zH#m6^F8^K+^*edAkxA99l8vHAslo~F|3Wiv+wTQ&j$C0sycL<g7wX%{`7S^z>4hLq
z@{P@<*vF4cL$2P#|9TZFy9#hdhin`DeRaOwJkH3~UsoAKokVfL)dk3e?);vYuROR6
z<kyd?=!5t9UI#zr&8VBI-sY3XC7^C?hbLEbTR5>7rO6M|S`ZeqOLp_gg?8~E3nDUH
zOdqz_D$LsgYv9(+2gm(DEl`$`moTsKzo0qD$O{uAXSkjA_t4L2qg#=im$n869@vB&
ziLX1^$jru3quP<5pD(61y|)Eb0PCqm;T7fP#B4!TVw2#VXA#~Ap{Rq^U}Azh*yuJF
zObVlK2eM#jOFl$VqK?C2_8-0#-?sT{Yzc)L1zfzpQizR%jC|zmAN~S13q|V;>eQ0L
zP##<d)<G&Q9MHw@8o?;AU3;}737p>&bD|eq4f_tBTD<AC%iEJqBW*9FfwKQbAZ*$?
z^oXw7yO$2uzowGMry7tQHaEKO?c(m}V(Ob4YtP#v9z+3i^`rtDU3FvIb7=T-sp2gJ
z_{6(AI4=YXR>V+pKmi`GIifj*iQhLyu2EJEFnTLJaK1}jCXL3Kq998VtO^hR@F&|}
z{!QF>)HWH->+1HuSL()I;)#b9yzhs&$mWas<pz8KOGRg=@Mv?6el0OGWS?_N-TBJ}
z<3Tp?7p1x#n=`(#sNx>Ne3-o&{P5RRZta1ql=F%o{9~fw2LI1P2Y#3sm+a<B+*onG
zLa+!zcf_9k%kO0}f(iJZ)c4!|Z5U#l3@kaJuMQS}UF8KB!QG#O**15{uP6AARrz(5
z>u{A9k%63lP1in{u7<nspZ#?eTErIre>drnGQAS+{>ZL!At@Xh%S`Oc=0;JQ9?Ax=
zR>b!GCr$GOSP_G)y6ZaBFNwgR-&b%+m7SbV+2jVj=rOCz0<G84d$^%oA3*IWy)vWT
zl&a1xY}Jx6f*U-t<E~FNR1&Qbthp8-Td9L@A4KM$fDWmynfF%vny)S|6xcQHhCmEt
zFQO$Pk!l{=HGl?y4R_Bq_3z-6VeAW?N^>|uzPcO9;1FvAICGGdaf$OPfF&6F-VESX
zKPc?DKAafV*qo})-|u+hG2hO+=i-(C@u;;!Jo!G>ksLaUvPHkZ?L@4=ZePG?HRn=r
z0QBHEIjC4<UNTjTp=Q!V@AS1*egCBKulY|?K=1TKJmY<KNUvxp@yDeJQgT1Q9;X0o
zhsk#Fz9%=&!kxLloU()c05#V`YeYOLEu;jH?avKN>u+Pn$c?;LoE$)L=xtUP(_UXw
zUpewp!0IXshP6XA^8z4ntun5_-t9Yl%<gkJiv&cQ32#nnA#<h5EVmj-0`D7P#GEfE
z4yN<fp>2$o5DV>3`ape%t~o<N>9gPbU%P|)l^hsunLQ7l|Fu7w1z|On?a<#?iof^H
z|9qYLe|ctbRdY&hsNdEKy~`JJVUb|k+m%L}Q!kISzMs%)=&1sCF$oH<;cTLgks!L#
za0D_5UWLtSid@#<N}pMhYQ5Z7=6&Yk!I{I+x;O93t3qaBD;eHV7<*~grk#OnBN{OI
z*6p8G5!o0SUv<%8Xe2{BC*2>cp4(pVrqHm=nSLrl+S}{<0`z6_o8^wWdL~<{=@#VE
zj<&6(#6PNcEXpWcf;<utM62>Tk{MYibKbJol&paK#I4#uN!S_}5L4?nnEh)p;apL)
z!3K+vk7|T^?C9r~#Zr?R)HDFfC_86C0Z-?hSq<rkm9chSWO)WlpLI~Iw%}5-M9el2
zYBDW^os}{6sXv@}<+Wsyb$t!Eq))2A#IN4&`@QURd<Td??F9nvhwGqDdZWK%x6EU9
z8zwgXOek@iLn$oPChY#wG9tQyfLIVy1DVT2#P(N+%l<ltd}()|05-B;o;zfHh~=)+
ze)Ge-C#u#~d%ppGtUBef-ElR*7B>Ra*34TOjMODIKS|i-zPVtxLsxd<w)~psltW~*
z{)U!`A%5yC7`#Sy6#fguG>St(Y1ks2RLFOl0|z)?8YD1ea<-qXglWR+QslDIW@nu5
z@7>&iP$m`mGzFOPE1<V>QojvpRNK6}Sn6l*jhcI@!+8<dCCicK)Bz~z3B`I$-3vc{
z`F7^GAyf}9EX|E8g$Wv70BH3xu}Lu9H9vxyKlR=WB*ocES0Kc)HS+jzei})mf{0eT
zHE=p~G0E?e8b=}}Y`acOLz<7uEL_U7^cn*VoaVNHvX5F!RH4u<f)tln$`wv89geiV
zbkcFigesU;sQ&U7z+e<nB2}lY4^l-97-+x`h<=jJU!^s&I6w4}^NYipKP3;Vu=zv7
z0?1BBLea2g?Wq}H%$~rT^hMV}ws*q^xQe1;ziz@fR0&^`!4%S7P)~%c^3k6ukV&qY
z%C~H41o161BM_Z2F`Rk<x?{FW1<n&);tTN~=7YFZL;*5a6gv*fYyi6G%C#wz_4$x(
zZG^71q1i^A1r{G)M0rfr(ToQ5^r;O%zL$?Ds6i!Ouq+^~$8~Uj9!Or7JZ=QWl?_5r
zk5h5@UbD8Fl7ZbClH}QaB;5*xpWX-#Q}oq`K$+~6+z!C{_{+t;`ZejoBucj@*#I%W
z&29&k!<xcV?sJDe&}&}eLM_k%Al&<`{crbnCwU8WbB7}?*~GB)bSkJi8BRl@vD1cG
z4uG5QoEJcgM}*dB1+9I09P+ewNGm0@9mPGCXRhF&)s!0S0Y(CM9%yzQCNDV|YIT1F
zDo+uPP3--8=6jAM@f8g$ISoT9MKt1e9er2g?9^Nn2$EPnlAMzB)@*42J(CMpw;Ar~
zS%#%0gfx~o=BWX1Zwzyl>W9@(_sRK@`PZLadOW-FkQ|JwMy;)B$5-zQ+P~>(;D{(M
z2Q!SZatH$V>gB|^W@!^*JiTFAlaG?l=_=hL<<A3C41;fR2S_(1S&Ar|sI^YEXBHiV
z>?I4pNf?`;9Rocg_D%ym;uFnXsaOfa`hrIMnal{&{qGZ@_Su!y<VIA=NEje8NPhR}
ze8m%|`CWju0(=RPpJM^C>^YrO{PsE@&QiF-4qckNcvJdde%5O^)^8YSKA8Lj+Nq((
z5`**!R5UDx`tl46%R5TDMav;(zX+i%-8VVG-y6*jOE9YN)5Bu+XmND*<%2Q();d?Z
zBN8Ct+TX;kXwhG?&rHM{Y32<FMr<1l(g+0`daCx!+p(5R{bP&MU|x352+)g8?Y?hl
zj$x1jfn7gN1m`ukOybs(C*bGCBlcKD7q)tp9HMxR6!S7qx2~pXY8?ZTNGoMK?{}w#
zbV8;(W1QX`&91}boh7#bM$jw};k`N?Ti94byBka@z>i8P`VR?jTJqj16bl*;`3>q5
z%Mo%#YQDi%eq~S-X(tPrY5)5A0rC7ELsR7wU&tL`l9R@I%7&97OEVe)P`TsCU~Vtf
zSC*^1*6hag3W*Ud*&1U)qZfTCqt(>tuck>AvryO5XD5zhu4@YpLqsoz4~&XxCg{G^
zR>P*(WCx4v5AI<S&)D5#S6B`+-?{RW3d!t!?~yfmT^DJv>Gt@h?H=kZvctVh>5wGB
zG<{nvDC}tAqR{bbx>96RaEG>wsXBEygYtsC<;UmeZI35w@5#0t-0#wL+>AX<z7qH$
zRvt!x*G6koY31SlsiXNYHXR~A<(5)u><f8GwRLIYEoW(3U+FIt*>tv;6MARCMa*wt
zIlhPx=;&EQvM0MSO{^4O1@{(lOfjISnGC=g?@ciM?TjyLW1Bu&%ot+$wZ>y*gq&|^
z5RvkWJ1G=@IK9R{FbJhjQL7@gkU)G%p<UO>{I!B(s8lNF;zH>_$;DVp!<_oD)<*mm
z0!XnX@SS<#=3RA@0q>dzUB#2SUd;cPZe^0izsMkarJ+vnBH~|U^|cNxVybu<UQodi
z7PdtxZ2Ifve}ef6pHY|e%;nW$ucj6_wD_N<z3E3Zu-e4+5rn#gk18;Pdapx)<!a>M
z%1MWTMl^#q-M)<PyR~#KQcN0ty>z@GgEzunzT}b2I3)$0L{d+}0@UwS+RkmgU(%QL
zBaln+p?=GiaLQpp4%pUccW@5!#4a;zE-Cb7G73z@W6fc(L`6hhOtf;Ec|(USuwg{^
zstIH(;sw*mo@tyd|0u)S-ipw0Ed*Q6BtVjmqJ0^bNsaxKAO87EwVZ~r*6filzU*5}
z1DB`kP2G$97ul9TW_{L;MEzbc6ObdeA1l)+TZvWaJ3<HFMO%sB3(7)SEXBKD`rT~2
z4q>=BoTmd9-u@2wu;VDCqhUKZ`~FE$stzp(&1AcILQP1p27f1{e{Mc(o^K)brVTg)
zd|0$rscmRd3{X3?S|uO7Z!qf${v=-O&Uu;p!Jy<~YMa@dY&66mu}bM$nJ?~NY~?Ku
zmd!eyC8TN+!S<?j>mo6vRrkvNqK-;DqY&n7D@NT7NP|SsAu|b^Rnh4g?LCWZNE4KZ
zq2OMwe`+jWP+rq!q^VOt2dcTfx}0vCIB?5J{>H)7!(vA)o6W<f_CKuW@KNAA_dZs-
z$1M8$$or*HZ`_cORVk-*CQN6I{;cj?q#hN=az&i^3tExc?^p(!WnRzaalHN<A)qP~
zWg@mGtK|+wm)9`~5E?RZ50O<Dg_?HUPp}QI#RjN1-K~BU-g%7j8c6OZ`}HcuQ>7P*
zi3?QBe5&E|&teM&_TcTC1Me4<k0zLO_$y~b3_=sySCsy%Ifx9#E>WexLf&)Xnye#%
z#`>8~2+VN5gtH?@r?}&uBRY)bE_Be^GJbz?Lq!h|g1Rng5h9Li-!NK`FL<?TP#J+w
zah)DmJ?yO8BwWH`fzfus^hMa#6PNTab3+({Jxu{1N}RF~f|yu8@bR=Mgyf`{5(pEm
zr_ID%vQV*a{8?Muo&hVN<|O8t6vD2GDo;BVC1M$9vpN>>rj)*EHr=VSdCm<^78VjZ
z-2+iXwyEr_^7fX99RsJe5lRti=BjsvmYPthEi-Tq7Ov<b5;3f}?d+0m(7bGW;Wm)!
z69cWyyn>JAg%MOb3_C<V<tTQjFzm9I7it6Koy{uXR*#KDX566z$7frvgD2qz^JhQG
zM1{Fa69x=-&y(Lj0wIv8N$&5g)1h9Jq47(*Ph*5ug96(Kmdq-sS&Tid@((QoCo1`f
z8ZOCIuWDg{_1k=b(&x&;dD?`@;&0m#!P2r+KN;z(%?c!q5uLhJMzhZl6<eE(u#*>e
zTqgDLZPCum8c00_KKj@f>D8Z?g|GHj_dRc$$5)QER^GM^Hc!O&k}fRx(FvyHg!m1D
ziUD6{0p5KDfbMr~03nW<x&%=nlPnZWUSH#addr3<f<A2muAi`%bQ>VePnh-SleX~p
zYS`d))OLLX_a+jY)=r*k$GXY9^nhr?DOrZO$ajkkww(}Qca}3h<w^z!+*8$f8H)53
z4?0zIC94@=CQ%a0yM32yV%J@x)oT29_E4rcIJKL*AVfag&2;+WQcJ4(+bk_(JrYkk
zjlKP+-isqHW#Hv;dk23vVqjYqT5Q_RWnIe(?%{n%K}n1q<yCm6YnrxwsctK)#kne=
zjhF|}MZ!^jl1ypoiKLJzfQAu9eu%A@6tc1oa%Mw9e;=`FM6I*Xh6o(ZuFm_OB<dHm
zGr+;OLioR(mbYF+NbF(37Trbg5O0luO7$?}@D(>Zt-C^_dZT4lA(~pp0gQ2JXgKwK
zhBZPv*WZ=VvxDNP*B18D%e(nCDV4q@NY##a6wuk=WB4RawOx(3Qi0NcziUEA727E0
zBt#+{F;^|R`G;bBZ(^eWKg^b#eqQ(20U0+!-fX8y7V4!Cu2Ghn)OazTwpHuVnP72J
zv(##iCTC`cjo97LHQi^Sk+Ae}yJ9UX0S{`#*oV*hrcvXTw2%Ytm?<~FO2^9E{>ZTT
z=A+U+N+C2Ho9%%0H3_GHdGtr99M=YTHhrEgDCQ%_+~+_SUa+j~aC25kuCg}fhi$?T
zt(2!SJ=6;!M?MjVLvOB>ExyGxDfjYZEL$EcvNYq{vM){u5Fpud2H@*qjFD%5=h~??
zl6+(yIy=x<Egm3EJOce9i+me^><1h9uylWlbU63V6=|8S1Lv{PkAm)Dk~3{&X(~!3
zT>T6>7pz%8g$UEO&5!_#wnc?2*Dw`LKQE_#r37pETG^%GJ-`2X-ISm+A&}5xm-;KE
zcsCM0?#kNRe}2>8Y%F~_NEt_l-eK5uR(@UV>tS5Z4OIby;txg-AZ>bV#PF3jKY}hx
zx)6Rz3bImpFKGFnLQU!=0)R|{rN85-`0HQB%-4e)De&xuN>Sbip4X*z4Tt;7yhYz$
z(uaC+81{V2kTQr0Zp7Z;bN?tkxZ~(J2=$tP@<NOYZ`IO-#5A;CY_Nt_BIXDV0lpL|
z4iF|H0w6r;GGV>dup$=3gg(EmKXe1IEfEkZp@Wnfa-pT3_HL%_>t!IEz3-WBs$xHe
zeL{zduOPiTx<?)wBYi$h)_A-hU-39&4Gs{+u9RA=OI;ur*^`)=fT=w{A<9sK7BNcD
z<f;(?v1;FbEID=e<5p_VXMiER!aM!x(5S>jaeN67=`S27AT&xl2+ge=zWIwK9Z?jQ
zzU#J7Z}~E3eL@vDakWcZs$OM@YL70Pc}F@ntgf7fkiH@EJtQSQqnJA##XpB}xg}PC
zmxuJho)T<BfRIeZ3#p1>f=Ff=(HIbxUR%y{1$vvm-^DN_2o`o4(QttO@1$w#UOZC{
z^*Vx#nVKF1gLDCAr>9{Ev~Ibkw~_oXqb@QH=jO<H_d@kA3Rb~u!%ssoACRzsv9A#L
zpUX=<Yoe^S#TWh07=v8S04RN6fK6T)3(#+ef&2Vf#}IXAe;eI(d<Uc&NW51;#WZri
zb6ZV}2lx(YyL|Ixw)o<F?^~C$xg}*BN%z4YMw&;)Z3D+X3}Hf_xnW?_T<+>145jlX
zf^n5etW{RYo0d6Bew|;jAb80*t0VsN&tf4xQkf=j!WGIuOeys&-3ak^ZlM`12f`#d
zh~PydNFWdqJz;RU%vuq%o}`|v%qzR5F5Qb1LMe>^hn&5lfaKt3M6Mx79AX4UF?e;U
z8^J<LEq^~-<^{?@#}TTlD=e_FoT{H+>5!l|9PIN`0C*)!W;M*faVm_YarbA)6WtJ{
zA^iy;`W!L)J~T%fovd23?Tf34fzIlzOZ|wM4Ucy2sLOziR80~<vqb<(s*JPs{EqN<
zt2X(&ZKB;T5Sc8vw3V2uF0>jl9cfWoi#i``B_Uo`Sw14`4hS!*I}xHbddE&^fovoZ
zUcg?DfulSrtAO$umwSij%S_Pw3QTA6NXvH2dxRD>-n)CjDDP6(Q65RUb;Fjk`|A99
z`-1Uz!eHE{Aw3YGZRGm;9s9E!0Y~GnR1PCz>F=G6KHKrYi6a*d=Lq1h^CsRhPsHsx
z;8`^++cOPu!?98-Ty_fCftBC}PoFnadm(GF09SfAp)C^Wr)Ou`D<2|ks1~aq{kT5G
zovE93xDup?3v=gB)-B=RP<B)CmQ30f`3n;Vy;NLNCkaz(;ig&BA6Ghnf_!;A#%s~i
zUCyb7)+%wJ{v|&cMlJM;apb4a#6z-89O($1k$n{DQBCVcxoz*0tJY%pSig^uvXT0t
zc%F6SaTn6&8<8izxzfc-VUI-9!`)>?`ew!)Ank&rhCBO_Opz$ji7cw>8HxxRrvm*P
ziWL1L+AM{HDOG`#h}EQsm@D8Y4}fO57=ZO2c<YpTK$sW-G>`WdOs~9`zj2sD3S$70
zd@60qHA!AbZy_Wr#O)C}g2;{U@klsamx}si62B&pvJIqtlElB><s4(7NT%C;<QP4x
zX#ONZmdKouX#k9B_Gr-BM~im~i&HMfcM-ysXYH3lO>`DnQLtk-Y-i)3NesX9f<e3M
zx|P05@bR{LKKUKs+Yz9_EdJ%Dqj4PK#dVPFpFK5J6l~8rx(Z!+Y>xqS%pu)Rn<92`
zRmEv7j-DAeYINS0pe8y`mUd$AiK}t0K>Z)DtXt#h)mRYuV5+l-FcDWY(^i}&gO_~_
zx9X3?oF4{-3QTE{`jnDGMUT2v8u`alcm3x25~dOP-m8;w<1;=nkSqMiC}*6w+h1sR
zCKC%uOuTlMAxV0oE;gNJ0;&nIjD29AN|PXITa!hL#yCCli_)-!<9d6gMSc}N)KG0p
zH{c3yUM_wgJP6-p8G}~wXVZH-xNPDhbcrgFQO)mXp(~agwLNq8qWtMpKP*Y_C?Hu^
zWoa6a-u5q+H0&V78ODf9YquAbXJ9i3%*0=(3;1+ObNkYfR;y4e-Iw*fQwsCpowwKB
zJKvu)LlTlO%=c$%g4ws+p(yE4?g;_`@8=&~2(e6P{=0KG47#O<*k7rn9gqS_&<)M-
z&r}v2SWWhyr_&~smOIS$>oZfN?W0#b$?d(?XESpuL(8l^$zWpMaa%!D(^g06TGB$X
zyI}NVJ}Nn|w@SWg+a0^PK5`fmE7G|G+?-Ul&X)KACt7RzZK`-$ATWX7E+X>DU_mNv
zE+rg3E7%PU=53G9XnONtut9Si(q82@FUBgb&YX7wMWjPx$y1hT`@@)yrMW)AUneaO
z!MkHX*<$)t9WF|zKAcERZb{ELDMqU2<B+xsQS|i!@$bwAXwk6W)CnSk4ktdB!}4Z7
zg4PlT2`U93>vTEZyiPE6pXPx;Nu=cqn!oj-;k&wmAGYz<gz(OzVXX%BVGY;XAejM&
zlp6}Knc%+NlY5wSD!9d6yDVKxXOUff&mm8-vIH9gZMGui8`7l-OKhHNqpPQT`WKBf
zl?aM>p$zjT1(?W`*|5>)36riAW2Iu@LT2|Xq*bRchkc@$T8z;oX6P9YOEw?TAuCZ_
zbuD%ZuQYe4N}e$D5+@y6fDCnRe_(XEVRhv2kM`|k<O=+_S>G%?9qE<qia_v8THRYn
zs&E+4tXy=c^A5ZYP?QKnlHSvFt1+hTn)gqplmxnZq(4|^)l0KD=|0+Hh6>wrk3tbC
z?r<S6)Cvh{qjUzz%KJNAWR0mOe&W4Gsq9axgymNy(O&iR41OHJy5L(HLCt({_`y4_
zsx^XsSY0}0vVEOrK<Rv<2XQVUB-eVQoE;}b4q{XitYelVr?@K8Frl$C$CjaunXud5
z3_b0Fxhak(OPC^156@mPET}?>Vw9YF*7TC4N=J4kwwMZ*96#T+<VZLf8p(CKGnMWs
z{*05|*9n#7JCmi-Q~WzXEMfZ{)U|n>!HTV6(NUY^YiI4{+doeZe}1oNopNL(esr-1
zzio$Wy8AA`t3OAkX##S?n%_p60Y+&J{zJ_oI7yTZ6AT2clzVpe=yabSkh;adwv5@|
zrf`D?2wRgsVr@%pv~<lv;s|$Ih8wE6Rsx)cL3GLK0S!SwzcAlZvAZ70HVBT?B`g+|
z_Id9sb#fy$U8I+JmK;;*TcuRk;AIjqYEXbc$*=0ne&Lm^?~e$Hk7C=kzK%YlSgxF%
zKj<~ZQEjyXt;+hnt-~-jv5KoMgb8{?k6;O_k0GQ|-k2#<wfs<jA&8@oPiJoWnk$c-
zhDnx>aK>1NvZu?eVPupDb(msWKatDKy=qX5&~nxXs-ev%GAOsqa%^vI^I-;WT+26&
z^&y|GK4;%8E@>=@`j+7E<7}(1>6NUDoXOQV&r7p!b+^j>RI={!sv|&f=RO74Vj0PD
z(6||M$1JGE2E*o*N&%Wn8rQ6q!j0%PqYrm$j+>m{)B?CAXnD?pi#{%K;Df@5%2<p-
z2lHp?w2yOJ@ssgUdsdD%LvYHO?9qIrflK?r_~)<@OsADrn*B94#OwF1-$}jnXwJ%n
z^st@2ekisC?_&7L<bt53NowD5*_rS+`&H);cB`c9%VTdY>+4Z0c1krOHMb{fY*iX|
zICViKOpSOWUGHMIeu*BwTQOz#$;jaGmCuluVKki_7^c^{P6xi1gTMLL&u@=FY`Bji
z?c@i|aqBecU@!XT-kvkTYB#VJ!KcDz`r|lG324T!4C6Q@_=l824Ah6S->;zcF)tM$
z7F21=2dNR7N%e|r#<UTlbzIWkMOjY?Zh>Qg{EyvZhtqq=J2`&xH?7d+7#6l>Fupys
zo<Z{++4;fq=9*7ibcNrku;xYnBe%u}?UJcJk4U7G#TszvtW6t)t>|gBhMfp2?>fPC
zacOZdsYWPIzNBliWa;RB?(TW-()YQ%oSiWCFBEB|!}e(=qT%8T3?uT~(WX=f<ZNXU
zR%YI=#<))$cYEOisrExxn9EgMtk-xm2R^D#Z6%?Dw(aina>I`X4pa%4>49YYjhA(A
zNk^#ot#omDR?=8Or?`pxVdEHwr5_uRF>DfQxYA09$Z0L1B#^M0g%3%>wlXLeRQNSI
z`kL#+*hUZ3JMm`RBP?YUYFEjW7TV-3NUE!^dA;X=blnB2)l>SSW=MR)Q;F4V9U6F<
zewunF{&39*Rr+Z<5&K+@SJ4nX6yD3G43Se_b*_uGyz3R#<&o!=nSu{#qWEZQTu~}r
zy}F90G^|e;=uW%kv~p_(UCr^jq%e1AI8Twqg1=EcHW_aog0E;0ZfdnA=-OAN)(7A{
zr&h~1@-FB-WV2$V<}-rBWg6WHj?y1Xovb=hN(=#*A@#K5*sjWp1eT}j-6g?YTj#RJ
zzve67xKc`%=eQP_rXSH$z^6PYtGgqG5|9*ZS-AtrBo5J)9nBmr^c4h-**mJ$>F9T-
zcu8Au7k%3SQ4lay>ouOpN@}0ff1dm>buouWo#Zp}Kt?IjK%HZaA%Ct_U9oCku<=u*
zeDSU-IYNhM88*2~E9A8LffCi&+4E^$Z^O4;NbH3K1Yz6#Wb-eA*IN}_y!4bYS7n}6
zI_6cFCkn<slKq7X*X;$@f<eo&nvaO?WlumdgS``m`R_PlU-cZe*7fu;RA5hN%4|(H
z55kEZr63`NoCGTSPo80Pj-<xUs>f6B>--zX2$uoA`}_coq!i{*Le6qe$9oJQb&)11
zUSW@A<eJgjB2v1!R<3PRgl^H)`Zil}MB+dotLN2Q%v#h*^lIP3B&B0IjR;WIc;tC9
z9jmr-FGOV4XH+bH1w*f@qh^zvbVAK1%Us!?1z7gssA20Sz#@d=yfAA<*(RqutR30>
z1BbkS@KMhP^}Ow{Gw-weCKEH)dP`splCT%kZ#cCZ36Ja4Pf<{q{X+GrRL9O6ak7Kc
zLTkM;+2EMzA2_+S1&n*IMMS_s%}*=0-iUx8mTV&wQ-bk%av81Ic)=2?*@+5w&7xy#
zF~rP4X5^6}#_xCfzC^Hi*OlGZigq4rJ8r!T@nbt(2Lh#$iWDO@-7;Wtv9#~;(-iZM
zc+u#n847!#+f$`u9#$!`pc@h)1$6LPc!B<;%3eZ=vtcHIBfJ08jIN5^N;#r#|DMQ^
z`Z44*frUcN;k%`vR2afR{-zQgH|rH-jTuyqag%EjPSrZ8Ic0Un$;fLhs?XED#60nc
zvAwJ7<8>gD5BaiG60ile=$R<8+|ZemMqz>a@`+^(xbX~15W<<CfSQ(}Xid6#!1uC)
z!|a^(!ynEv(7%%>uAk~R1=VQnG2>6xj!D%Io^5ALk%5%V6;~4CSI483<O`nTlQQlR
znLji_Z3Ch54ZxXAXRkDW$6$%?gSvwOg^t@N*H%LpilsEsCJyZBpx0LR_9TO}s#=&R
zJ7w{HCW+#ks~@BQKt`j!m6vDc&MWDuTF!B-E}kXx8!&@N6NFW9K+Fn5V)*h^TiV=$
z3PrccKtEH@eR#vRa!1>;K5fjT7JbpQY7m;_W93zjG*cP6_Z@;4%R#hPe~9)@XKh_5
zwDjF{a?67NfxMtaXb7Sjo86fmqfioi3K0#DNSw<)D0{y<ThO)m(V~UbI@|Z7#+lV7
zgCb8b!_EB;{eAG#k@OCC-076A@@zMrjLB2o?kQG*mHbX?&&*$YDP&z!a^Z+0OL=$V
z+rwV>MW|X7kr7ZP-EVntGyGIDz2M`vOM8`sk)A8f?>~7o5$eV*h0f{}x0A#^6(32}
z3k)e6LtuHUNcyPX&JTsAAL@-!NPF{<;53t`PM_pg>nLB7=oJ|B=vneTFSYS+(9y+o
zb(Y}WE}cc&s791ej<|y1u*c-q%c@)k46NHyvzBiz%80`D;xO=e6o($?LRTP)$l+q!
z7-OZ?394^3GQ7J*1cEMTNQ?$nn$8CCS=zgMw*MUbT1eW}$NA{Q!M#z6WFGbcKR_-M
z|Gt|(_{*FVf!3nC%a`Sznl!D?OO)Yw5D~*KyU^wO%j)}(&+K1(wM&J(;4va*y(q8D
zqH>maKg7s$mUQ#6RT%?<f@n;T5Ay3UC=i_1e_GNAii;1TD9i1a!lpC@hBbD4DfPY8
zeK|#~=G3|GRuzi6%8zbsL483nw>#>3ni;C~Xh<*A1WG>%fZX?rxzuW|RFk^Azrs_z
zl=jL;ci`9C1nsur2o<Rg#NP=DT3j1R{Q~KMvs)!7jA>snQ7@@z8%pUVbEl3Nr+Zb@
z)r}L5=4B^zXr4;c$h#;#Y?)mj?9te_;JM<dCNL_WRmr5G+}h0Rs=r@$<(g(*x}Mcq
zDVLoDqM>4<X>U}27L(?t7H!+hPq9{aI<v>4VgJ!qa7(62?5Ea`LB^wGYZ8K_ihJHi
zzaCw>GSaj5vvX)kwdWT8JacFv(1j(Q4uD((n`B?hd>rdT`K4xOjyU_#)$^WOFDY$5
zx#Y6H1Hb2+a~Grb_Nu7W+@-FH#leB2OG`gTg^z1~UR_<ymB@;4v@&4ob{VcI34EtA
zuUguwW`zk;IOQ2HAqa_7MnT-*B)4Wev2=foijfs*Mw@fXqP&vM%4*Q+6;-{KAwLe!
znE+K|dQ638UTd~=FkhbEW&aZy3B3xwS(M(_A>gsjyE0Dr7_9DD*S<e!_2Um|B;TVI
zf5__V)P-0cd>oBufG0E%O#%*|Gm~XSlH$nQc+Y&oV)uZsam=H54-vnDta(JcY=-NZ
z`CAs7lOdN&r&9L?#1!M#L~T0ut|=bv_FBB46iZTZfed9H!|{B(X9V464ceXJ3$f|0
zrP3yMrJD8>^%#8ai2-N_7L7--bhmPlRoQeFnnV1h%5R>~5P$6))UGa0N-$+xQr@$!
zv=gpiz>OaD<ai$~(P}|{c|U*Tw_w(f^u?O`;hHpY0SO@*J3du;ui5W8HTiAyncP%w
zY^nB#Q|rnC0IGPOo@k;XU2>Rog1hywd62f$TWxG+^~r9H^L1t)IhH#jE^|)?O;prB
zc`trwFcaPGFBrY7Yco0fN}@#TgmtV(mM(Ao9xd~hjr=tlff04a8eoKAQ1n^mDR-?t
zU7*9OCS{q7Bqb_@_a(G*E~iF{c38ysDP0HiZN`*Ro77-7)At9e^!T2u0&&gvl{f{V
z&{y?lU#eB$8wpi4lb&a1@33@c1JJ9U8|!boFeVQ0;9u}LEIEb9=lgFiee8`M=}4Go
zd7>CEnBE!wMi@!c`2|{46FMTqsaW$M3&{m!JOLVEDuf@nk;}gOz~>EmAGEX?2I)+J
z+(kAkH+iww3Eu2mskY|}U-~(!16j%`1fukvB_;NbjHt#Z6ZwQ8x~s(J-u^rY$rLX8
zn(>2N6QmOhnrI(^^27Ew!;j*RpF)z(R2H8eG6UK7ZVV(*X$bdZbhk&UrQAH<i_@jO
zO3Wp{?65HoJ~=GeI0*P%@pjZpYq#ktK28kgA7XLf3qj!7C9R>tzl(m1JaImh(FXyV
zElIxDNu0g)p4wmVLQ??JM;>5`{HLz4K2xvG0vhpu>FYHqU#~^;aQ!pWlqe+qiEvH2
z?Lqmr-h@FER2dYQKoxT*fo~Auv##GtBNf}4WvJ8;Dlq<jd0{$T!?y{5#wbHCs*PDk
z^0$04CIDTiDJhQVHlctNBn|_$3r&Yqan)b1W+TZPcM=dM>EBK|gl@j?^_xI|#yF%!
zOdL4(uPTJ^2%|}J|48w>sLfHI)4;CCfySoIG_(tPS{KC4>d(@Zq{Nzsw4Xr;BF&kR
zI!8g$!%-N@wY}B|Wb1{@o>c(YA!3RaMd`*>mBXOxm;G4(=}D)w1<0j6EQ|)c7yytS
zLG1f4#8_8S%_yXtFCtmyP<>nRJV2F%8+)&ike*Wmkfjy{s(PVW%OXO}X}dDfRWg8R
zLQ5lwL|y2)Km@&2Yv}Cu&4jh6cgh&*YK@43Y316gJ0cyA2-gbReU{9F)N#9KXMxqA
zbKN3F-(hVFk~%~hSQY{}xWp3Vbu0n=+75-#3m|OJD82fF!P^1oDu1I3eX2q82Sk0K
zynE=jrQ=xJ$1Y&qm_sI%K}svG4Qxm63N(+kUO)3IcIOi-mnN)0JCwi73mOGppmt>D
ziG_~Y1Av5+Z+q_@f7u8ErzGH_t>PV^?M8c?dJUpmX|XsneEIR9C<`>rHq;2$5Tz)F
z(DUjG(GC;Pq#wv{QWi$~1$UAvOPBTYEiOJLOv@=F=~>XbHEcxO0N~OEE??`IFL_$F
z4Jr=jK_lCTW$>#M%<U`@us$43g*N&{_aS}96NLTB`9g@<@8$1pK|1^$q41;CFwX=t
z^FJxbki{rM^UDZOW&uyHn=cPu4mbH7MD8CjdNfk4FF;s+CwU2YFoh+6Nb!3Q??=(y
zK)O~FmyWi9LD3hgKk0(fv%Vs0#>{RgZZ(0-*nCdOSZ;$gE(5La>!MQ%RryA!I55uz
zEPyj;lRY7YbR+S9_nM28#y1J_uHBhwvlwj$vy3}i>rUI|wu6j^wN<31BJhTNrw~kh
zYi#^4OgxsEHf1;iFKoCa#k@F8Ktug2A}%3nre%{ASSSv+_m|dugjXPyw<RK)yTz7H
z0KN2@ZoB*#AbD{7M&U+X5qw!J@8ecu?{0P#3<1B}6TtvbbO}Prxly3yx@LRuTSVey
zt2)JX0BLz<BPqr!VSrt=>lD#fiUFmfSjOz624jhI*J;QWT!vs;Tt7(i(yO~@bXW);
zCoI1Z0qAVWVrbD;5T=@E7Csym3u+H}fa{}teYK!t1C7-O`<#n=(C62F#*L#8j2gXQ
ze_V$K#eG(reAVjbPDlam+6s<zXCAigHYoxi<mgQd=`&DL7)~vlaY#k&w{`)h+Eb*u
za-Fc6kGSmLS0O!|yJB{O+56}ic!UE8I#+99#rH<Ynm%rP8o?$aE=jysH57M*!GDu)
zyRhP<DLT8N$zei~g>(bHVZ(GRNrARF_WD;Eq`E^9nNl|edvt$!6{yGi^*honM5I5?
zRur=BD$emtw|Z9DjKJlW+8(ls3F??b5cW5*O!LTh@Rg~q5GpfRuJelXUX=t`#tjla
zZIqk7-Vrd7l<D~h7L2Ok5z7ToQFufP3lLCfHV$8RDW9l`0`(!$pBLJCz3rMWv?YSG
z^b&blYhU5G<<8KazO=_@bSp`en|n{SRYe3(vI3vC4o4T*9f6e7D@f%Fpon374@^8O
z2fPG|P~(U+a;-jHTAdf%+M1q#Uzfu*BBPAwBS_2s!ejV?iX`=0LEa}CK1`v@CJLUH
zf|`7(*eztYutTS_6*<pyAxy!RjQpfHNVtC^qhjkN<TkJj>zaK!uHilw_7-z5E2pW4
zf4Sq*ngeO9)V!S~pHF;&bJu8F>L<`SU|$S)X9d7w8ktxv(6V~i1i4WRDnCMnzZ+mm
zbFu{uh}g+^SQ$vpcYMrk3h{;7XPhG}oW#5VNSMQ!3;*#MR~)6v^b-E!C}~UH-)Eyy
zmNM+A0YJmW{w!!4H(}$iO%a9{Tme(wVe$iDt?4;RhxtWb<7`q3w}^DFLW=UjLQaHB
zS%PIalOw$JmTAmnXAxn4;vxt%Z{HrT=d9h7<x>fQ%Ffkk9`D3!qW}(2s|ZDO*B?2;
zo5$jl=xXRH__TZbTBjZ+!Hb^IU<y2l!j;S2?iC2(&^t1F8G)inCE|ByMD+HdwF_$W
zt#i)5kV^fG(1J;_HH&?E{k=D~?QZsvfLQm|3R>EF>C96r<%_eHT0wT+Ggn+BDLzF!
zlE;*b7&`NnAH+%dHhZ6{*FOB{;qN0i18c7|3gr$BPMng{)J5OK#h-`g9W;+6B~>6;
z=`&goW#*})77AXn98pkug287~zWS7&v^?CNcjZN8|8)chp9hHeTkIW^VWac0^qs}H
z0jJTH0?$XJNgef*mgJV0>6W|gg1kd5`Lb?talKFByzfUms>c$tzaRp@g2!biENB~g
z>m5h9pBiV?zw+GGMDN-C9WosC7AFsciPQ!SbZL%Hf+pMe-PbJ4ZCn&)S(MtY__Ua8
zufhn*+C)cBi*C`^ZX}*h7d<_kBq;&~5rIJ{b`;<S5TzG5>#{k`iY%Mn%+-;64ALxH
zQ{KVxfxqfOeB70(U(|aFaVPW;bVW;K{BJJ^sfyDvHM5Cos&RloUnk3~Cdow3OU*w?
znlCSD=cnaN+gF#B?l^$gxLDRE?$CW7*7{A#lEtpcI}x_W_`<YvjF|7<NrW`-XXUCs
zDaFETqbsc*jTr5$Ic=MwB;!S>>#t<k5XTSPVY?*4lCL<NWM<#_9m1s7yI_`L1jK(c
zoSvcX?0445UWJ1}3||=)rQtsGo|sk{E0~(3jk#xQdU-ZglRgV?NyWfECYslf_Qgql
z#T9Ds^%v#iF-s2PNs&f8)@x_7SfxC!pCJ}aY4v)74id$kk8G87>TDDoZgdMzmkR3B
zE;6{Ec$OMD&=|gpO;zbFB77}{%**>2&drl0gf_)ELuN>`>2l)d3inQ|CMOmDUCo~N
zkV<GQuFOwM+<(jj8|Q?iOlRmciJ1|De6yesKQSt^b+xr>)hq@&dg_n1fk-}@!CxbT
z<qIO1DjFRF2X?n$$!<)mZ<OEG{+u|hk>@Q)IXTAXrpp(+TLF)O#(uwu1ui?v_5g<i
z{ryBMN8l2*o!z!i=+!b5SDh0Ep#t793TLKRICQZjdSy{<M-(?ozkc5bcr0B2q}EoO
zZsdVr7u=*t8PHU^N_wR;4CV8D>5H9d^0|?CS*1(wLGi{U59#nK!?7MnkR17N%@)-x
zQ3|8YV_KtzaYF+)=!KSg&dg7A+k6<9TLa-`zi-@j`eXBaFClZr?DiCj$e*{zF)v=M
ztv4WsghXNF!Uo1P!?Vh#6qbTK_EMBU!{aO8dk-u)d<IIBB_emoa^(>)Za0k8A?l3s
zsro`0ZDJ5J+pR=IIuOzHSElNdk_OGm?&q2P%_67k<fn^`poL5g9z&*u&D1j{Da=jw
zNdIg!{(#o)m;=LEcsggPaWZdIWnq^nHgxV<L4)m@Bf-n7t~2$GJG`O`Mm7MwF&*53
z0MM;G2Euqt*h(Ts(JZGI;2h1zOk%b#&eM`gT_E!$2?FWQ%(i#lae`GL38LI$k9F0`
z$AlXekP<p_?7w)!Vs_M;dgZ1+0xaD}XQ6NpoIUP~OVEb))|{|H+g&kuJnT7ga7Qz5
z1~_BnngA|#P_1<N!L!3J(lRw|o+VLyS`dV`!3$TS4&FgvH^BRP3b~sOo-NH-MUX#N
z&T$&nN^fZ=*;he_GIIn{$Cr**$8Zy5Qd&ySBp47eI}64D<zxHcnPqqj@K5wB&jhDO
zxu(=DBM>nh`K)})?~qnm`vHjLt<O64HLZwMOXE9F{j^;gJ!(p_39%{MAta86>9~F(
z$)Mmv+DZ!pQvBit6qtq49tB9kY@RI0Ur!K1<)f=(>xh_*OCldLDnYvY|NiTqjt)03
zi~WvXd|%L6thPf<vcsJ1R!t7A_icpAn24}-ZwbBq1z~|Wv{tS<b~kVQ{x)j(AL<=(
z!}ip)>How@>wkTPFp#y!O1AD_(GKX(LU42zr}Nj)84+cvpGFc2kN*8gj0fQ=elLEL
zw>P3-kP-rh@O+1={qLc#N5{U5a1}--t=|T8V|HF?Ll{}v`Y7<XG28h3CIn=+;}ACV
zpJPW?!lC`6!ucO|{>Fo&&%ss5Z)yHH<1ooTm{T}zJ#Wp_7ygX6Cl_u|(m%&Z*85=3
zQy<aEj(g(@9KYm0NqMv2sfI(1Q7Y|qq5<_QH4|X<<Qx&ZC4<%lD&YjUQ9%u6i?gHJ
zkQ@OQA{(U`dW6^db|3OqnN-mWb}b0JXAV5WgX{qr>$Cfn50}S>$+~i;<gdwQ{Hc%7
z6G4frcrR=jO5acc{m?HdY&(&csAQ5DqE&ArI|t5tIXObKET6k_g}nX%<2`H<)WrHv
zxIDUtCDDC2A2;Gnu?36PLV;MqeN9*1Bd8E`HWW1?0SlD6Q~8!diV$qZd6#?VxGNxD
zjRc@vIMU{#UXU#L0d|-o0%K+FHa}t00<q60jwnuKy}MFdeCJ(3^mbhPUSz@Iy@v&m
zRU-)4tDLpEe<8o^Fdh{^;}}nQ*7eV~0}J3QFXO$Wh93$4kylYgWLLgwJ_GHbe^x*I
z^*6r!*An~R_v_lx5?&AxlU%<>{Y&5Gl?*Il(Xg~-iVDE~5P`>MWedozsXe!M{(HuL
z3Dulb#h#u3yxv$nBC4)$Tf%j1>*s%Ot{vb_`PbgMYkzzKu+6M}SrDuut)*lL0Io|i
z7pvdnpZmuzaIcY+iN_CNaVLrr70Dkr>1~nr9YC8i1$nwXOGcP7k<0JS1=#%g&f|g+
zH@f*zdrshuE<~j%=Jd+qERFMJX8~TpKP(^#bg%T4v?U%mC^7ey9-(%QCWl<>KTClT
z1Lh&cFn~HZxe(pT8~~kf5m)D*uc+=!$?KbBN<^uKPSKjO^erYXBLXGR?U4xGTTI+$
zhAtwUs<q3TG8#bBxxH=QCB#`t71o6qvSjjZGxM`!)aPVjrcbs$ALFXys&%?`xXo!N
zaFLLE0Xq^Yo&et6zv~{N0EHB30{Hh(AR#ne^HJ{|Jg1q+r)caoy;>jYV#-+jb?-$O
zHdIS*3{&GF7BJ3AFlh&`Ng=8lz}s^>RDb5cI*4WFqyJ|Nuj{=jdV51(!|Zfd=Ww^$
z)PFx@A}Buy6kLDxNIM1H8VvxZpx$4FlnpajkEm)2E>N=-eK?&Z!R2SjT?4qaf2J`Q
zl5$0CBRfu*=lB`8;K|FB{eC0N5%UZ>^1imIuQ{)lU!p3RV50f^E#MU3EvVsTdvy|N
z?YmQVj{PA?g;?S^4diVHKB~T^u)nWV{g0{Jc$1)`M(wEhjLr9QFn_I6vU2w5q3eIb
zeSI$>w0iv$sD&_zS|{@VjxPzucif5diHnAxf$Zu3{v5#2yA?%I3o-(=PKW;c^V`VJ
zH8`JK(YRqCd-gv+KZD#|PXx7a{#32gx&Qt=0e*e}!}&zp0PcSNzduLrt{{Y3cm;Pi
z{qN6jB0p#1d?ITAcbEOo&wpt^R<B-Tu-|{+Kks%5k#>_kidxcAcrwcKpI7rF9j=~b
zzoU)t2?nR|pQ|GxP!6gOs--~0_N3l_R)#53Cd<?zFC@&MTX*dG`^BPJ;l*3$AI!EX
zATN0o@t@rn4u`rqAG7ax8a#U6e;&OSIe&L0Qd_G-e~5VQe|FTbWnlx$rhw*r?=+sl
zzMBjEpMO6IcXvFw<L5)P*pGvZV~5?o1KjE#GlQpXCY0RAsO~66`uFWpk-Yd|8WK<8
zK;P{*$OrJvQ>2y9;0j&EH)t4!A9xoX=}-s<YwpsceZp@<@XfW*=3H#ByzO5?6nu&>
z=ohVdz~nwX3?$BE6Y;QHh^!(Am9j{`H<;pWmI4sUQ4KAbL75F4z=n%MMh+n6tf0HS
zRshj{=ryJE@(zO-l462XwtS^7v*{YJ#ydUSW=G5bIEs5BSqBQg|2z%jRn*MK-!|vB
z3DiKA&v^qG<$w6+>$m$$8SdAW_|CviKPbq_y$Ol_Wd(X+8T`w%uC;!SuGYuQdb1fP
ziig~%j7Y)BaVl`R?L6>&o9mMgLyzm999P*Vr14xsvi^qJ?VC*Wx9fvTw$Uqaos(18
zH8NQib7lWhjQhKa!O!mRO57)>bqTHG7na5<THIsZC8{b~76p)J^Ti<!+gFiK)PFuH
zcxT+$xY0w!e;NUJ0%}Achew&y{~y0~{jaOAPL%xs#Pa9o|8wd8BPRr#gUk1mfq;;Q
zBCBM81?njx(*Rn%lj|u$OIAe#{PQ;tAqj*;Wp<kJQXA{PPu_>UTk=}YaTIleq6uhq
znFF|*ykr<Pc!GMFD;tthNa5qi#`>a0<Y{_U4$x!KQ79mB9(SJ`qq*|Uf*B>XTPNQl
z;IWk~Y2%x6ps6HwoVyVG01-Ot&O!mAY~NLCG};&Vcw07hAhws{X*}ysb-AsK)XCV7
z&)W$vNd)5_jTGF0DEcH#YNo*I`Vf91q**y-kE$ie^+~z>8bun60{5P6RtDXN0(RQF
zmKQZQs3L*bb;Uh~8(fDq6r$1vk;c6-AKtAyXeZXsdvYDLdzicr>siv@aRq2{3IDb#
z8@>S}9~f9%v@#0>-ha3YXlz`=l}(%WYtNyPBmUXRXp!GK02_PJ2OIYO-#_-jrXca6
z)5*(XZ|tlNzR}~rI1^n#BmCEukPFwA!NYwfqUIm%<Wgum33=(s%2f9sUtbe}u<fbN
z!S_3-feiEu9D(%jOCELx0ztN;6$<%}i#rpRZmB`R7dn(AFJ-ea{CRmHEWDPj%DH|D
z%+eI4iHc#ONM(g`sOcYS6MS8+2*yzQn(oVWuWgQ`h6|5v4|wc7ap;eU*as64Nygny
zI2$(nk?6y~q4WoLDUK1Il8)?D_Z`PmrCtg$nyG4B(-0#@KsGZm^2WD6JZK+mEeQUc
z_3|T?-)H9U(6a5R>=w(+)XhuEY=GF&Q=R*-=Y{B<{;g|c4l_8LMFhL{T&}m{pB65)
z=b91M$ps)V;m96(`Cqr71Vd|HE#Jb{P0P&p)3#fWq8BQo!V=D7wLW>+U4O0yv?099
z;XuK@ZyK3;q@Ss>+H4qnGmy>3j&Cy->|>z1^BcHVbnXk(Swb2@huw914&2|}pKiVQ
zN_;1(W@j(qZ1y|tN~?6N3n=C+czN@EPGRr9^!k1{hzliMnjc$7*TTyVJd_Q63d^{|
zdWGc61U}dXo93Onx09lS$_|R&-@$>nJ>O++sat?2HD(!>J&fI{#>}hssu?y(hEC24
z&{~kt9ckG5YxR78yxzQxU%k}PvPVN9UF;|C_c{7)%us7bJedO=<Y7<h4FhHs^MI0f
ziSR}W=?}Fx&v@kgsb9;l*N$v{rAEa!%gvF-aZO6L3v_2QU_NwDomonOGr!P=Rn+>s
zCM$^c*s_S(T5Ax1(eg9RR%3JFBgmo#ZYaU{A;Qo3!lCaMThcTef{z-1wuX{=th-!y
zC(!zE`eL&qUTZ7!sS=FtzXY&+J!p77kb*g-XNUMshiZk!A*(wF*3sqEn=>UyuPv+#
zWCttp&sM)#w`N60pOu8veJJY`sr{MI#s=Lx4fWj!fNZ{{Ec5z-i);8`Rux4*;_I^p
zmsxR9R?~_H$k7X)YcCPxUGC*Rm$%IPS|SKZc(j!uh9um)TwX41?&>;}Ud+}0`j8~3
zCCh(ysBumvkEE^Sg?nL-fNjv>JK&iGsewW>ugK)lVyo6H8z2DTKYOovB0Y-Bl1nyD
zaCAQk%CWQkw|axcBz?CY0WC$_L`$3(F9be$dnTcb=v}38&kd?6?_rpp&S%tg_Z!~J
z#69yvs+Q`mPf_Fg74Kl@U_huoBV5eR$an5_<i?tnJ7)sWly$ZXQUErce%L9axCL)(
z3&ZwonnUW|E&Y&8{H_$8LXD#S3GQ2~@gQe^fn`fZIPvN-<Zcp|brw!xVVU}`eq#Z$
zkg6Xa?Z!S&-H9|RJ4lu_;GvdFYr)j^DN)f=%o-Td#KPyWDBkaR%!q|0gX*pl5c>L=
zhC9lnv&3x{Z(KCbvG((hs^|ccYCLfE_LI2!)+-O{ku6!nbRAe|M1UlClGKOA>X2SX
zGV(`QMJ#&L(@NM4rRkIBm>EN0$)1giyvX5)Kv8*qGyzH8qh|_gqaAF}hG$M;3OLe@
zQ1cmDI;nmdS^mBz)G6HBqea@VX&r|?)M)9YZ^4097I5r{@Q^miObY_TK|4*y0XOcg
z9qNv}#9**-Omc@Mo7JRfS-P`1q>D;mjVD17U-YybunCF5yy`>Uv(q&3w#e7FkV7v@
zZfNg-)K)WRp#vNWw=|=94Vtyfh$=;h$oHJyB^lT<n+Z4+Km9z&kefb-J5a{Vg7^Kg
z(2bO)>1*K`!l!uwR0I5Tg%C}<B&ci+0FJb%efQRnTlheaj~d;}>U;fkmL1<v(0yc+
zdjJP=0BM;B+WNEdnXCHrW(J>2ozgAavY!z~&YLk}tBqlcuAbzuOjUxGIz*&O$TWoi
zti*Uf_f#QhB_aJ^oremtgp47xZQfTK;H*OeYvT7|+sL|@1nw@1yyV^&{wK{5bFc3+
z3nMw<F(u2c5y<uD$Tlg^Gr<g_W3V~bkAC+kra&;2kt7|uY7YC_1HP8l3se36w~f|7
zKgqeGn}Tll-W8^!MqR5{JfL%8+p`N1ihapujHs;_&cl-ReL~CFci8z7#>0K+!?U?H
z-4$>B=jZOd`uy}wW-v{uX<eX9QzSz3#Y)1mx-N9U+ep_R<bmW7wI3?J4^GMFUxrIS
zeqBUXb`&)#UGZ3Iv9=sGuI(=O^lF{ccEDE2m3)80k7f+WdjD5@Ul|tV_Vz0XiiEfk
zL;(?06hT0dkPuKxQo2JVq*1y{#Xu?P?q=v_0F#!G0cit<76k^7koR7=_xbPl{dlf(
zo$Gu!`-7L8y_tDtp7pGC-@p48GKXI8y%wBCjz`56z)y%5ZToXvxJXv<X!w}JOWbFv
zSH2CUDU)pg!Z9d5eR`39cDS-2?vhC$Qq3)>;48FQ4jt4uV70^K`pLxAVQv=JkOalW
zND9<3pC&ay<mp5c@}B!Fi?Pok@D@+$4Re%VCUNe?!%8csclZq%TU*i?!0TN%ZZC(t
zPCn9l5mLV1cCSOq?;`t!BBc>1%id1JzUSgE)H@L~w!38|%J&$mCVyz14=g-nE3cE6
z=LhL%Iea_K2kLs6;S#9Xz_43r@NUpTX4~ZfOOh9uSOL5EB*9ZQ#d^mSZEhRG8g;EP
zDztOwb|6NuN}ehpkEd4NAFC-k1mA`ECS&Fj2?J>sYlklv*wmG3e$-$?kA2w7eQmjp
zeTRP8=<%ZbH$KFR`2~*rlH+>uk6|M<a{odKkE4wfjEde;Icm<sR@y)6x4Yw7)D7PH
zul=YFsl14Mn+;kWWanrBtAi+CNVBV;8G?r_V~*r&6~*t2-B%!cH+(CbIsoOwFOXKl
zv8=<%mbPr83(iZ}n^S`(IA50xBY1-R{dObFzYeiK6aXSao&`Wtu7`+ra?&!DVO8G5
zHiK^7ygc4B6vBJXfQj?|db1#Wlpkn*tJVz^ymQ%GfxcIS%%z0ZKAB=2rhEg5QF1Yo
z)*T&<?+WDjyx69x{46ohpz^Z1Z0^jB^`_qh9+5{@a5@PBAEco;M;;l+<G~|J5H8kU
zMLfQrk%*)ZQM_bq#hK-If2}W$3|i>)hxKQK2)BpgGL^|5%HRI-`jKv+0oSiA<`MTd
zq;#=V5c6m|-Y_I=IYbSGrO#PQMH<qE>-NA7a-?#6=|0oEAg@3%cX}Q9bFFY&Hrjv%
z$ny!Ws!Fwo@?eRz-&6^npwu6zvg#fel+}9u6tvo(p1;#OZnWYA>pYBsh;Y*Y)WqRQ
zHZR&i{X4)e>E7TtY?T960xtCFNZAe2OH*Tv!---l(NGECbvZg|4@fT}#pSuikMF(%
zQ4(c!vabtn-lkyd?vJ?9zI_W(^&DY*+NH!UX|(h)Cv@lQHYBtUnS1eO4)GlgsG!KD
zp8{Mlgb8JE2py=kf3u%jw8wdAd`vNp_njC&cycB1k72)Kaeb(AsFF;9UD_PH0nKB`
zQT9jgkGDEgfd5y{TUduA2g!1>42D@P%3|5kna7}4B=iopz896jtxoOF^0ONGs&y1%
zty0MQQ5EypLRMgPRJZhwwAqIR#FGlmV;zRWyn0Cpq`YN?FLvO*jWp@zC-_z*4d4oa
zF@|MH6>twW8l?9&r*xCH)9Q}q=4FDY>+iFzMUM0{*BEYKzC`Fb87N?wd{%ocKyMXd
z+j7#6?8qeEo0+egfC6Nb4ySPu1f}Wg9xUr>X>Fo&7Jsa*Hmd>QuiG%6iMbU>mupa4
zCIlnCE-)XpQ)BYU%YuDvnY7<)z8z(5Pk*axsU0(dgeQT|YDaQB9+$R%3l+DY9<4R`
z&^s~;Ttqx5xnB4VD;uk*6!0`5c)54kRBw*GZpPsasHwjV=x?c9!>!O1<aU8~Vn<a~
zCGV|JKyMnQtA`&~i^75zm>F<F1A8FaJoTcq%kVdt58(Kr2~}uFEhI23_^gL=ADhX6
z0!JX%t=mSo46!`BreA3C^&315ai)AG4r`=5<)9&D-@9{$g{s-xkUczH+o~sg^G#$W
zpK;hT?yFJks;?6vMe=&SfD0r{-$1^mQr=A`^-jbh*58)7nl6XaeC1F5ojht{U-}+~
zx4`_;%Z#^Ycbt4`@ZD0$kZbp^;?<>zi4V0ajhJQ)Y#qh`NM%*knl_d#9ONpn-_yoX
zGhtt>;UIs01@5xh7Gv6NXCSWWj;A;4&g;}S>yIFQp9Q%6-1mw|e_*iKL%Ih1xWrlL
z5~@sTx%BuHMED!FAd@h}eLB+emT|ISuaMdncBlw<#xPGU>*nnuihU>-yIo@>Lza47
z%br3`f}Rr(<_T!WvT>bq*K~GLuRK<tO-hny*=4sAa-p;z#^a~O8Nz>{_^0fd6)zBV
z3{&U|DZTt*Q||{BXZ4LWzt<u;G~0`DnMQE3LvDv*N>bSC!Vvu2jKU8(Z3B}grn!#f
zMo~35VTFjtSv!U^f{4!-u;0_Tb3%J`sVw?XZ;T<8hwJqjbX+V8bGth)-LEE>=g(B_
zV796$Wkyv5GK(uom?<3??dQFQv8<`^yl&Y0=n8>(o}!9(c<yw3_D#^^l>kqSY=m>Z
zDzp*DBx<guxn^a^)Kg+d$F}Gw-8}~OCgI6>5T-tCq+=egsp=3NXwCM$?sFi;LTJ?~
z5!XVW9Z;w8qqKVL(r;C)L0Ew>d#uX{JJ~=ZK`aV{+FgzBy{~tuvhT9D`hY5e9LlAy
z+??gyVKe6`rGE)j#0TOh4qUDwB=6thE|^&&hiH!@H#MLgRn4OuQa1jCshi{2BGq(S
z;G=&5J#O1yPngLInTO42Ny|`(xS2M&3d-bYH6^WQ;4PeR7|XZxGWQ6nG|CRba~yZg
z5axTUy7iGsNU%ZYmA_z*X3w;zolAJ<h-u+?nO2cr9MBnuKJ4OLhr7XDjP`^cQ{KYv
zcT4wG<lS(}5Yz)}T0$hF^hR87GSk||gYU;@6pY{Q;&N3Wvr@KCKar11SRBsTZ+*4W
z^OZrbfedM*aL90uYwlS5T_w3D|DaHtn{_fQY-JBs?D_krrJN)lXwhcQU(tnbY%7Dg
zJN01IxKpZAnMD^vf>oSZoY9Hzz*De?(yjK8`R93Zd+0_jizDVKifM*aU~KT2c#hyK
zed2>DX`H9wwbXdY>1T|-GtBc^OSpSVq-o^8M}L@6(S`6I)Lv4JIM4RTaRhAOhI5&L
zypT@eh%LVG!#N-|NX`=puxUMVB}BATPZuIbatm+VJa`9MN8cX8%05jq2%As-STru~
zy>U-uP{l8&;X1}mq_UT^F5V%*@@Lg`?7MN6!Fp=K^LhziV9Soj)TEpvT1p%>BdN`=
zD*L6jeLl5GyG|!t^z>et$O9k9qqjZPjv%f3X1!v3Ixi>Lx;)!LWyOkv_`+X5)?{o^
z%Wf(~?G#d38qt+o!=k;;pRx)Z)t<4~cOxG&*58pke<SdD@pxJprE<34T<~-vBz8Pp
zFj^ny=qTkRuP}{rZ_&P!2fj~dNAx3E_3V1@M^Sb#dV7Y(EWS{^C3U{XbB8dPQMIv^
zmWZ=3Z0M=IsErw&;Bn;1A|7XnadA|aH=Gd}B<0_+uZHsY@GYKIa#dK1WT+uXRo33*
zD0s^iADTlD!`WnBd>qdy3l!^Z3igpaPKNh|w|l#&oV)Aw&51*BFBz|G>yCKsmtpIh
z9Vki|Q3|OZU-u9!RLuxlH<|QpFuxhO`d**rny>Ti*|qSI_+;`+QT_5n*l3>Rwa9v(
z$?1XlG@&m03pt}}Sv+6$sN*6$p*HMi+?pke$E1cc%3p-HCL&oz2}Rnal*2|V$;i&}
zXlL$YsW;5nA-OA00<%wt(GrmskEiq8her0NnE8@TY%)IXJqLr)?BX??0Cen6Dy=Zx
z$ld;Ob+19w=(8a0`gFz0Ul)Fk$EZ7==i*xR+g<k?q8$A>)@MmBwxV~(Q0k`Hsh`8)
ztzJXt?xSk8=ygdJck1lv=!m|lRY6utyDX}`f`fRhzb43{xV<KSQbT?-y)$}lsLLuY
zq=zpth$W1eh{_FjMAV!`t2e%`lz6Oxd3Z#98&W<$uFengq#3fGapZl5|4^6ytCafM
zYM58-H@gP^VJ-sAjs-!h-b~L^DOpclr$ux!OmZPnk03L@q}kx{x_4_;OiRSTU(|83
zHFo78cBFo5e64=EMhOG4vyRtx8U>X)+@W~0OTUcAZ|>Ntij!;uDltFd^erL&!sX}q
zHbwBUIZ!`0^g?j7PcSO`G}(rR#2Zh;R1uG5v--m}*B@i+j1>8Gn+ks7Yvx?cFX6&w
z%)o#kNAPJV{@Ap$_euS0sSKBCc;X`0P83+mozXVSE^Gw@sA4Y;$N7`(3;4^jMx9?m
zn4-WnN3PJ(`f*6wEK_NO>(o6mqxQ#em0~@_zmJjD>TxS}oWv8Bq?!|y3E*h=l=xyq
zyj16<r8#h}5CkV%e!U9A4xzX4BFFJ-9|cGv*m|-Q=)awOJzAj6Qx1l@4_$YGQ}mXp
zY@Ukm?uSIHfnwuRPJRsbg_DkRJN+unP_A<0T|Iot9H-jDAG$DTE%GwA42MSZs|kF|
z*DM%zsW-Jdx`c>R=a(Zqi%gMVF{*1QDltM;diUhYfvWhFd8L}EwH$paItOSs4H@Lt
zo}N9t(wMUKlmhxVd%9>Y{zno|r<}tLo9^yC{!Q@s|BC8pUbniL>FI=H1P|<s<uDwk
z)JSb3qg9kRlvMFMhf}`Q`SqdmvB)Qq-RoXqI^|6KjnusTrR^ZSw0JC4KX~9U)A5yc
zC%`rY66!{jTa>hZT1YlREE+JWy<gf5wm;d1j5?`gaUk-ZSwC9)%`imnnpm%NKmqUj
z`Y$b>*E~H~iv;JMl}URAbK|Y9Ry>R5tA#cZ%QLb??pqA*N3AREdW+b5e|ZFmkV1ep
z*CmtBt|fLO0vWc`4EOu09VVJ_DdcDT{hLUR*DVyUeAyaa{XRCro$(Wlc1{!t<?)RF
z<SpH7ysU8~;08?(Ln&gPWy-2!Q^}K8k@>fJy{nymIpC`dJ5uLU)GlSS`pn6>NN)j5
zl;pDkrX44WxP)&v@5^v;iQzG}_cP{actckdH6JLvZgAb6?g{wweynm+zs5WFqIPqx
zV9~Rg0bd~Wd)&pEfAteiu=ks_ok`b6&0f*lIAa@Mn~f40uPz7(<6Zr<=mZ%4cMO(L
z_3IB`Em=L<!>1Ok!{+U1#Ow<9jFV*OfA#OjFi#=kj@m)HjLy}_Npwpm`j@==pB&v;
z?0a|C@FJ&Xe&j@f;tpU8({N+O#IO4n8UYBm7<Qc?KO0=oXz4k4kYJjWca+%os7iN6
z(2u#V;chbpG76I#k+Be3X5Utoo$2X97ki;b0~P6`%zsfC4;=Ywc8o2W0UR@Cp`EY`
z4O{`FK?u*vM-Pc5dFPD!@?sq}=i9D7I1oo~<jz`_*aN0UrmIH{_rgg)D6+K*zYsSy
z1%Qz7vW`pQ)%)o_g%k(tEC-=>t2Yae=27c?N3=vryfwM|yB1jE3lri?PNP=}JEivR
zV>FSIx~{=f0}tXWss}X(mAX{Ux$hjtSH0CaG9sG<58@6nrJ-_@2YW~^>*LM7ykGao
zU9oM(H+m%)0wfqn^gtegda3szClgH9%n4VmlefX&(Y6}5J)kp{9?V^a4f)9{Wb<SZ
z0tY$)z)IvwOLWt@*U&jPo_VR1rthRnzbY&abD6cZ#JajUA%eg4Qx*q@3ixJ6ye6>P
z)pMGf#f!isTJjNgY=S1p;#DB2#p{QMs?O{4SROAA^wsX?y!c>dv$oP7$Bo6Paxz3S
z70fi`t?SWWU~i*iX;s49RawhcWzjpT&v^~S;9k!YAOJp}LP7h>CbxAWn02$EOrzV6
z7|HTZ2Ike?hf#qbm<RnZX3JAmBX$13pQzsE8R;URDXrTG*ms=ydi=x?um>-8Kbj4|
z&sf_R`0{BN8Pqxo)TlSpah7!EEp8?mU_C39=F(F!5M3$L@%b&qX{TXrUt@K(LTKQI
zs%@^i7U@+!cMGxgT`yR_pzS}LL*l`Z%yq`0BzdH9DuYF*Ex+Pyc%dm>(m~SpITxsB
z>Rg^jk#G)pJy<q?7I<-~qM%sRJmLF|umt=2eoVLGKD)d!XnIgL(-LOxmq>3NAaNWX
zSEitHohNoZIBlwvOk&<Os_7?uN^zqNFioMoQ*@ONFvuPQbyC}(o-`|c9Hc&N!`en6
z^J7{>nZ9bO`5Jk5R$oq>K4qTrp=jn#2>vL&^P$|)ke9Q?M<*@t=OySP`1mF0m+e9E
z95~v!eZ0q-MD1L2&s1cu#fK1_ItqsxDlCS^r|6_!?)b64q14(#&^?}`{V>Q+u!17W
z6+4PxQLJcUPMj92N-jKGkJ?WqKd`y_6)8SmPFUVd#nXV&r@pY7F<D(v0U?~3==n_I
zJ0j8_NmS%kWYY2AAhiIr?v^ZbN^G!db>LNA2^%*S7x7-=T(N<f!NRDhxtL+O#fznR
zCDJ{>xQ|mJEWyY@{G^iHhJfv-+rZ1ayW+KxXWtLS_-hnVl<YK{5N*l6GSi#U1;ktP
za^5D`c3lsD*c(2{{u#!oxD(5Z@j;eQy`mQ27t4vZML(=Rgi~;&vNa)XQbtx^#kk;J
zeIEriRleC3?<RQ}7o-+J%qs-KTh8v|q9}u8Ny+pj%H4ZC-Z1!bC*pF+N>3|j7f^1a
zB%;zDP%|2<bGb5B@7DcX=A%4targigyphrUe7_CZN{Pvv1*Sy#D2_M)bb1h%77q%8
zax6A3O@edp(>4$T<-N?~BsgqaISE7E`EHAMUGm&#M>oqj4A7<Hfsrh3#+4}qt(h#e
zX5z_FM+RnNO+zjtdrhp@x`5t|DWH@Kc8jh&H;i~v@QYRS3m=LXZIw=iB@NmLOsTFL
zKwWlZu1-W0GaX=VBPiUYx3Y;`ncAnWqw1{>3HcA3hnSF7D32qE;+$gaZA{Hq^7;dK
z5U!X+mRLjEY60*t9`vhHdr>Pak|boVZ+z0}Mv;Eu@U2tA&bgo`)2v}soQKU=8*X)@
z@X0vUxYN?lVv$DW!vrou>flM|FzA9Eo|j#MClCjG&E`VmCb27*ERhy3><ckT{g>^1
z-ryc<F;J+R4Nor`^=6yA3HoBoUIcWF33A#MyRKXv8L}CQ5MQlgBfH3?U*PUqHgr5H
zywv^YR}08eFIS+`1eARN@Xc&XRZN%J+k+2e!1Y7sP|fXiP$ZSUG2q{nC8A+-CkAWl
zdR2=lzzpJH$BacAI?N?Wi&^`R6Oj<3c&82Z;~@ufwY9~Ogh#j0yH>a6UfK>Ce7x}0
z;9ISQ%Eh&H1r7FAGg@d#UY|}4r%LhN%o{2-eQec}8jz=7&E0`{I5pW9Vn7Nvfo*x`
zK!SiBFTy-e=8wX0-~?_=8;u_{AQ0&_)PaoSPK=+TPXcr;BMxDNc?fXq>M5*k;MB%c
zbkL$00Nvp#C>0Wxr#cb<m;6-lp8QEL9jC>9lCioi-SM{%yeVo3Vf-XXYWD)D#WtEp
z!Q|coya+8|E;8i$37ndc63z*ysXg`SP1dhPIE~9YTrK&edgOcNPw_c44aD3YEWUeg
z+H-LnJMx?A<1_M2@x7y(e$LUCz8LZ9@*r48sffpmClyHOmCt?oo-0WQG?$ro*zPOz
zXqY2%!Zu>2tNkBOE*>I}<Tez3#Wi#>+JE4sFAQuJmi0@G+`4^$iz*|I{GScu6|P}>
zrFA=_uv*wxdky@|QS6J9QwOoXeXFSMjFGuMa2Z(jTY@_Zpd;t*d$t&ZJMEDdjuy(O
z0FQz!3!Y*2TQ=^DSw<sh2J_opKktj$X1A8p1NJ=no>43LS0pH~z{_#W$Sj|w65u3P
zv)90Xzk9UeLaXLb&HW4z05~m_NEFE}asdkoOXY~|CY|+#@8j%zYJb%2!QH#oFb>0m
z9Xm=*DfO!cBBzCVldsRNb|M2TuJ>qy+nIORO_Ecq&~jDsItv-Gp7wIuW9DZXm0s)L
z@t`|<7FRL_G9Ny0uCEy3*uZT;uf*iR<939s0K&M4_8&lIui>ffvgdXEnxeT4t#giA
z<@b=%8{9^9#T(~ir;^U*1Mr~bPg~`E1Z}x|6qXxUPfN{w?PB|w@h34+t00u@p!xIx
zDy^<csOM>uC1ADLmee~<D>E+G50z#8;CT5(R*M=F#15>{F?I(TFsAQB&^Jop4IHK1
zwS}#2n?oWLW86(i0GPjr-1bumYhM+@OwTC>i~jX3B<LcFSZOQ+`HqN`@u3{PLKG|~
zfO2If4SBGu5lVVlJIuZtU+_@ARVnyb?x)()RxLqn{NZ0qfMS)NNUjX2%-&~th^Omm
zXk}y!O*jZ1Y+O$<+$oaMEqzudQ@|X5rSJkNgPkgPY1R9#K_9{7p>UcoNE!f@Y)hl;
zl4g3Hj<g(hRCH%;aYPnRc5c{rNo*`k5z5T~>-L8i3Gc)Pn5>q(YE!(4Nb5OP7tT!x
zgkPImH<HVaJfa3p5rPi(b*_sH>u#6BeafiN26Q;!v>%()(94kh_bJgxiH8`zoM)6n
zE=veGLu0YAH_hW5P8-k$9Lw89bGtcQ0lf#){0GQ&vfEdXs`@ILxED-cI}a3<R*8et
zrCnJ8jG$v-BcHH$oG+@52fvL^3sd9k-eA)p^CkzUkdf|EWLjjB-&_IeQddR-->t-}
zz2ob@uRynV#jG_f_C#i7(p*-h8ty^T%^kW+HgPmjAj3&OfYx~EZ&^Z$h#SVo6H2gf
zf4<(e!Ccihjc&lkz{60adfr3!lw>FmWJM=PTvTjxbvBXEZ8GeYrtvwjtyQs*7&wMl
z8NruJ#WH*goK)t^Z3emP4`XP=tO@%Kbx<#`h{e^d?X^y1#H%w0zFbs=a3-=S378Yw
zQm1)RYAOrQo@IQSH?5`n-jpuJV{|JX-|~)=O+K=9ErNJ#COui6ep0$Aday-LP3n4K
zb?&q__5(%{wtvjg!rJ#9MNK2Jz79?{yI>t^0f{7$=Ii8}{hSNYzI_6FX-9$%$%Zm)
z9q&bN=Iy&ye_@JW@2kSotcvsi-@AkjuxhFFO8^1%Rmtp~{>^eXK<pY}H&|90Lb*+0
z1$(hsUI#1F0@+EQ+Aqox648KGrp+A@`fIjOjo%yUscYzphqO%I!Kp|6TY0R_44BbU
z)0Ud8?hcB%zg7>_?gc-GdZyB`Tfdk9UiQ21YLYqptCyPy-3r({c8mo(45zQbShp1~
zk8whpS@W8K=JSNQb&x@=1=(zDK&5Tl?J`L<Di^%(mHl!oztk?^{?+QIRz>o|&ym}s
zA-fX{g%}B|)2r*}J@`#-p&_F+3{%v<wWA2x04CCY$@!qEy0;ph4+7bDEMu$WVPLpW
zCdt5?@SG<}ul?@l<^b0TU}~3Fw{6@7gURqQlvupScWxm+2KA}UG%3IqP%J3+1unD(
z(9|3)-eI6w^t99Tqx$J8J!Ga|X>AUGzd1S}>c^lK>O?sQ)6DgYHA<oWfrf_O$3dU1
zq>}a5UsD}1x4K{Lcc&#SMkBbS2gThrG=PXT;m7kW$=m_&Uz~_L6Dr!rO-I{op|<B(
zf!u-xXp#~6sPD|ZepT5ak1TTQyP$;MKRR=!AV)*I-QM#Jm7rNdH51myJ`^(L+eZc3
zR_6V-e?E?b(40x1pV8ud!U^)nv)*cejqMBDln)|5hl_&`V@B_{ocwMdtxs^aa<A2A
zm;*sB*N<PVj{BI;UQDGp%t7wzLUe;Zgk1Qpwqzfm_k$1TKMzv5n8=97w-l!e2&-3e
z$684}Rw&7ykVQrD5fmbJDOqzEB#Z%nUH$C8e<93>Y93IJA}a4FXIbMqL+T8Mj4iPC
zNm_h_4Az%%Vr-@DRz|jsuA74iktOeOnjLKKIur|kwT%1oHnnstZGEjYej{kOQ|VwH
zgS%CjURaLVxNP#WK>ysR=L17F8&1n^Owi83N?J~>&U3SUquy-ueMK~lRv`_)-_N++
zS_cvYWB)ld2edbJLy6^a?i;{dXqi|8xq}<Ks<xGp7N<vivn>=Hm1%@NLkZdmLRI?;
zJELG4&a8Ax`xMSLdE2SAn9dRR#n8j|fVU9cW6}f-{KG2KjQ&IHpEQ@U$}ZDq-qp^a
z(clx*|44!+)cc5E(X%sbpX-5OOH{~18Ck8X1h@cm_1Dei@j~4zS3FlOulQ`PUObeY
z40F($rH<K0pn2fQiE;Ab8n<?ax=eo(`-oq=Z{#CGo??HJ`{Me6IaP<YpC-N&G2fIu
zE_V~hmx}=4P!30x$X!9tT>E6VR>nQ9%MDhE(7&mr0jXq`hg01)QfY2Y9b=jc-bUg(
z-D-ilgS@>}obJk*Zz&kC(0+)=*Uvo*BFf2U&r@zlr!1?Df;Sv6UO6_#G&;K0;>cNh
zC~L{3>~c<b{29dixQ=LQ?Q!YJaOcU1NGs_$G@f)G!X@HXK8WxPR(U)o;cQbZ3wHbd
z9hk&rt1N&_eJXi1)&+nXO(8g>tuXTBHXv=kBNG*NpkWe9A&VjH)%{DpVBti!<2ZT0
zp-x($^@}n{dQKPT|0c{2hKhzXX0T%B+KOO+qqKH(M>v$?mjCVQAAhV%KFbD7`spv5
zasqe(_5T?nbA4BH0GwR86xOc}6cM+;kW9v?o$E-jZR}$ad82R70<k(%kP;Y7sJLL%
z;BF>I!{*Xw(555(DmR*5Ey#k6nghHGs=7uE<kkae=4)t*jpV;-hW#pw;?xcyS<BxI
z=4qTc9JjazR!6*gfmtvBl~JHm-DYdPWZ{=YZH;(_w1~`aGKcaFTW=CAZB~NQ!lOMC
z?R4qjJ9?JvB@t&XT#n-GZHs7-v|Je}&+9D2paS%}{iK#$sbf4Nl)Ry>RZx4mWwr<}
zPa3;{po7LEx6<0s!A_a5ewb4$_2gn^a|P5);XC;C4^ZmSc+4Y*h%0ik$r}?%D9L_6
za6lc*R`0By6I~0~_X<O;Ost7^6>pa29`jR)B>8Yh<JfwD29PkM`6{CwG7nH<)AI;6
z;L5dvq`K$8GsLuz*O@ZhlzHP*srUq9JgFCP1u<4|<{6&363*IdFCM>t!IPIxk;7e-
z(67GO;n46iNa@K`)eDyqFm_l0i>FV?I7BZ%gdPK!&$wch*$%386B31-7WDa@^+)O>
z)ORY+hWR>QmyljmuaC@~2glvZ&_#K75+08=9>rq8JEHUBEpfY-55f1>4G;F7<%Gfe
z&b}yu>7v3Jm`#5YK3@k#RPnwyu^;EVdeH+R4#d07pM@&t;UvcTdtQq&uv(sM3$Nlx
z?JF_bdCP&Kq;is6^{Qg@s%%djI;~x(W3rmzFT@jo8DP=;u?Eo1$PRJ@!pzQJ`vj%0
zDGUnnJ{*q5lFwV7pD>M)QA^EDdkbMtp$*@$E6w|<k1sWhkIohQ4-snGXMpgz{{6#Y
z5uif)U|zXL3rb&8DPD%SGGuRVTT&2A0U)STzt*l#uGt+R8@*V)*)sa%JcS`F?w(sj
zbm&fzYx<{kJZkTL1;BthY2@}SfC=T-B4B&1#76(TNc-06lYJqJ_EN9P%uJ;?zk{aq
z%3Z??S{Hv;D&b^T{0${$$20T^eAG-=d)Os39SWZbdPMnrE%xJE*TFXF1@mbkcN4o7
zXlE0rrV{6hsYJkQ$4edhq0bkWK0ZL*9+CwBbywDCrRqxW+I$!0cbW6z^M%!75zp@^
z>t-dGzc}WrA5qZdtz9NnNwo865=;M<cPu-vDRQ?7ZM?puC^OL$@TaLoBo?6ZEE7GE
zeo)VN9ST0d_`3z2?V9--m&w03NP0P0W1@FeZ{>TQ=bx_9$Ws%tV_xSc@D*sg11ZXt
ztb+IKbfx9f(_Bdvj#KjLKu8@qJciex$=ii{vk+!ZT8_@UxE_ui)5)ny*ghDkPHBf<
zN8Oj8ke1Qf)U=%ot2wlJ;T-wZhSOq-=0NbU0M^Oc_UsLX)uqUYR4eC4+<j8UeB>|p
z+o8svSB~ZP)Q+NBY7u-Ee`*riRBL5R-{9!EUf~i_uDK!?C)1UB9@4q?!%6BvGc!Kt
z)pmgbn{8)b811~wJx^<B%zRMR-=y=Ekaj;Unw3WHlZ<)E*5<pnX#)&UO-yOJwwRZ8
z2YpxasFr!%uDpkQz<+drZr@krOg}ONzL^Bu%AD=cKaL3-5VSma_x!^-bo5+6vAus?
ztp7)E;788>2{+s(zfTZM$TKRFHJ{O5_}Wdb&IxyDUw|k@*mI6<vxa{Xj03rQpqLqX
z2!ZG_18Ekl&r(w~=a3|Pn@b{ftD1EE?~7%5K+^Ea{s4bMHtKIrumWWE{Aw2;Qf{L4
zdIyfilVEx6lLwrUQh_f*dvCUnyeEUP`K3(M-yezvqWD}9(b6xXT6=BJ>_{aR%I0I)
zVlDsVs2Y>8_W;oXuHYdOt%*)YMDP9EfBx78MOa4*yy@SUw)eOH4Gy;zB^(^+jsE!~
zfIH>CzbHsNfW#Fv=iGPvzr7i9ViMc`zkct}E2P1Fpu){&Uj83HC#my3@b4>y%e?`g
zfHD80oaz7gIhpe>%6~5A@5_294_|^t&dc}z<L5_amQSEB{$H;^hOoL^<?}Dl;{Dgp
z|F4td|BrvM?8ov4df*n2z4)Utpn~=qxyA6=WwMfOT)zz|D7yKonK#2PhI;(Dp%O=c
zmHxneAomLK36>H#ENv0kZC*9)4pO5C&Jw`<CclSa|FFzeAX2$rkvxrNPl){F7`ryc
zU{zfdDBOHx2%Vvl{c_~bP0iGTyHVM=a1w^P7kbc`sLZU*yD^~}%*1?2g<Th7C#sjl
z{(Sk~XW+1%NF+y)t5E>2d3!7<uj4PMXUD=wz{B{f80a-5B<Q2TNyBDyeJLdnTdwfu
zH~TnMCW~vYD!!L17LxssZQ0)KzL^G_i29Qb%-@gHKexIM(6|KOe`THjSubHXpiHOz
zT4wX_1p(W@D#!w}C;mvE|FvE-Rp9eIs#*9~pbJ_`Fi`<gz2NkdfBmoM^)CQdNQsrZ
z{-0;}?>D)L0x~zdgiinG(L#TdAcuSM|1bUj^>003_n-{V!{OfX=b!h%KXTH_QkWaJ
G1OE$<ObZkM

literal 0
HcmV?d00001

diff --git a/docs/cookbook/structured_generation_workflow.md b/docs/cookbook/structured_generation_workflow.md
new file mode 100644
index 000000000..c627ab9a8
--- /dev/null
+++ b/docs/cookbook/structured_generation_workflow.md
@@ -0,0 +1,215 @@
+# Structured Generation Workflow: Generating Synthetic Phone Numbers
+
+This is a condensed version of [Coding for Structured Generation with LLMs](https://blog.dottxt.co/coding-for-structured-generation.html).
+
+For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this
+with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
+
+## Unstructured approach
+
+Before diving into how to use structure generation for this task let's start with an unstructured example. We begin by loading our model:
+
+```python
+import outlines
+
+model_name = 'mistralai/Mistral-7B-Instruct-v0.2'
+model = outlines.models.transformers(model_name)
+```
+
+Next we need a prompt for this model. Since we're focusing on structured generation, we won't be engaging in any form of "prompt hacking" and will be leaving this prompt untouched for the rest of this example.
+
+```python
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+messages_phone = [
+            {"role": "user", "content": """
+            Please generate a realistic phone number for Washington State in the following format
+
+            (555) 555-5555
+
+            """}
+]
+
+# This allows us to properly format our prompt for
+# Mistrals 'Instruct' interface.
+prompt_phone = tokenizer.apply_chat_template(messages_phone, tokenize=False)
+```
+
+With our prompt ready we can now generate 10 example phone numbers
+
+```python
+phone_generator_unstruct = outlines.generate.text(model)
+for _ in range(10):
+    print(phone_generator_unstruct(prompt_phone,max_tokens=12))
+```
+
+> I'd be happy to help you generate a realistic phone\
+I cannot generate a real phone number as I'm just\
+I'm an AI and don't have the ability\
+Sure! Here is a randomly generated phone number in the format\
+Here's a phone number that fits the format for a\
+In Washington State, phone numbers typically have a three-dig\
+Here are a few examples of phone numbers that could be considered\
+I'd be happy to help generate a realistic phone number\
+I'd be happy to help you generate a random phone\
+Based on the format you provided, a realistic phone number for\
+
+As we can see, none of these outputs are even phone numbers!
+
+Let's see  if we can improve this using structured generation.
+
+## The Structured Generation Workflow
+
+In order to solve this problem we're going to introduce a *Structured Generation Workflow* outlined in this image:
+
+!["Visual of Structured Generation Workflow"](./images/coding_structure_diagram.png)
+
+Let's step through this:
+
+### Real example
+
+We start with a real example phone number, in this case for the Seattle Public Library, that we can use to verify the structure we are creating.
+
+```python
+phone_number = "(206) 386-4636"
+```
+
+For a simple example like this, we'll just be using a single phone number, for more complex examples it can be helpful to have more examples.
+
+### Draft Structure
+
+The next step in the process is for use to define a simple regex that we feel correctly models our real data.
+
+```python
+phone_regex_1 = r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}'
+```
+
+Next we need to validate this regex against our real data.
+
+### Validate by matching examples
+
+Whenever writing non-trivial code with structured generation it is *essential* that you first validate the code against your real data example(s).
+
+We'll start with a simple method of validation: just checking that our regex matches the data.
+
+```
+import re
+re.match(phone_regex_1, phone_number)
+
+# <re.Match object; span=(0, 14), match='(206) 386-4636'>
+
+```
+
+Now that we have a match, we can move on to generating structured output!
+
+### Generate Structure
+
+We're ready to see if structured generation can make an improvement over our initial unstructured approach:
+
+```python
+phone_generator_v1 = outlines.generate.regex(model, phone_regex_1)
+for _ in range(10):
+    print(phone_generator_v1(prompt_phone))
+```
+> (206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 123-4567\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234
+
+At least we have phone numbers! But I think we can do better!
+
+### Inspect output
+
+In this case the model *did* create phone numbers and, impressively, got the area code correct. So using structured generation did improve things. However these numbers are pretty boring. Let's improve that structure!
+
+## Iteration
+
+We've walked through the loop once, so we can go quickly now through each iteration.
+
+We start by improving our structure:
+
+```python
+phone_regex_2 = r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}'
+```
+
+Before rushing to another round of generation, let's validate this new regex. We'll add just a bit more sophistication over our last check:
+
+```python
+re.match(phone_regex_2, phone_number)[0] == phone_number
+# True
+```
+Now that we've validated, let's generate with this new regex!
+
+```python
+phone_generator_v2 = outlines.generate.regex(model,
+                                             phone_regex_2)
+for _ in range(10):
+    print(phone_generator_v2(prompt_phone))
+```
+
+> (206) 867-5309\
+(206) 666-7777\
+(206) 444-3333\
+(206) 444-3333\
+(206) 943-2222\
+(206) 323-6789\
+(206) 444-3333\
+(206) 867-5309\
+(206) 466-2255\
+(206) 222-3333
+
+Better, but I don't like those repeated sequences. Like good software developers, let's iterate again!
+
+## Reiteration - with debugging
+
+Here's a fancier regex that should give us more interesting results:
+
+```python
+phone_regex_3_error = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4]'
+```
+
+This looks good to me, but there's a subtle bug, that's why we *always* need to validate our structure against real data. This time we'll make our validator do a bit more work to verify the correct string is matched:
+
+```python
+if not re.match(phone_regex_3_error, phone_number):
+    print("Regex fails match")
+else:
+    matched_string = re.match(phone_regex_3_error, phone_number)[0]
+    if matched_string == phone_number:
+    print("Successful match")
+    else:
+    print(f"Error {matched_string} != {phone_number}")
+```
+This prints out:
+>  Error (206) 386-463 != (206) 386-4636
+
+Ah! We were missing the last digit, let's fix that and regenerate:
+
+```python
+phone_regex_3_fixed = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]'
+phone_generator_v3 = outlines.generate.regex(model,
+                                             phone_regex_3_fixed)
+for _ in range(10):
+    print(phone_generator_v3(prompt_phone))
+```
+
+>(206) 494-3216\
+(206) 374-6218\
+(206) 494-3337\
+(206) 476-3216\
+(206) 484-3548\
+(206) 495-3218\
+(206) 494-5517\
+(206) 375-4636\
+(206) 384-6216\
+(206) 385-6218
+
+Much better!
+
+Now you've seen a quick example of the structured generation workflow that can be used at the basis for building and iteration on much larger structured generation tasks!

From 25b6bcd498d0f335683f0026e9baeb591061d001 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Tue, 2 Jul 2024 09:20:14 -0500
Subject: [PATCH 002/505] Cache Legal-Token Mask as torch.tensor

---
 benchmarks/bench_processors.py | 106 +++++++++++++++++++++++++--------
 docs/community/contribute.md   |  10 +++-
 outlines/fsm/guide.py          |  16 +++--
 tests/fsm/test_fsm.py          |  57 ++++++++++--------
 tests/fsm/test_guide.py        |  61 ++++++++++---------
 5 files changed, 166 insertions(+), 84 deletions(-)

diff --git a/benchmarks/bench_processors.py b/benchmarks/bench_processors.py
index b47bb4ea5..5b4901540 100644
--- a/benchmarks/bench_processors.py
+++ b/benchmarks/bench_processors.py
@@ -1,8 +1,13 @@
-import mlx.core as mx
 import numpy as np
 import torch
 
-from outlines.processors import OutlinesLogitsProcessor
+import outlines.models as models
+from outlines.processors import OutlinesLogitsProcessor, RegexLogitsProcessor
+
+try:
+    import mlx.core as mx
+except ImportError:
+    pass
 
 
 def is_mlx_lm_allowed():
@@ -13,6 +18,37 @@ def is_mlx_lm_allowed():
     return mx.metal.is_available()
 
 
+def get_mock_processor_inputs(array_library, num_tokens=30000):
+    """
+    logits: (4, 30,000 ) dtype=float
+    input_ids shape: (4, 2048) dtype=int
+    """
+    if array_library == "torch":
+        logits = torch.rand((4, num_tokens), dtype=torch.float)
+        input_ids = torch.randint(
+            low=0, high=num_tokens, size=(4, 2048), dtype=torch.int
+        )
+    elif array_library == "torch_cuda":
+        logits = torch.rand((4, num_tokens), dtype=torch.float, device="cuda")
+        input_ids = torch.randint(
+            low=0, high=num_tokens, size=(4, 2048), dtype=torch.int, device="cuda"
+        )
+    elif array_library == "numpy":
+        logits = np.random.rand(4, num_tokens).astype(np.float32)
+        input_ids = np.random.randint(low=0, high=num_tokens, size=(4, 2048))
+    elif array_library == "mlx":
+        logits = mx.random.uniform(
+            low=-1e9, high=1e9, shape=(4, num_tokens), dtype=mx.float32
+        )
+        input_ids = mx.random.randint(
+            low=0, high=num_tokens, shape=(4, 2048), dtype=mx.int32
+        )
+    else:
+        raise ValueError
+
+    return logits, input_ids
+
+
 class HalvingLogitsProcessor(OutlinesLogitsProcessor):
     """Simply halve the passed logits"""
 
@@ -20,33 +56,53 @@ def process_logits(self, input_ids, logits):
         return logits / 2
 
 
-class LogitsProcessorBenchmark:
+class LogitsProcessorPassthroughBenchmark:
+    """
+    Benchmark the time it takes to convert between array frameworks
+    This should be on the order of microseconds
+    """
+
     params = ["torch", "numpy"]
-    if mx.metal.is_available():
+    if is_mlx_lm_allowed():
         params += ["mlx"]
+    if torch.cuda.is_available():
+        params += ["torch_cuda"]
 
     def setup(self, array_library):
         self.logits_processor = HalvingLogitsProcessor()
 
-        # logits: (4, 30,000 ) dtype=float
-        # input_ids shape: (4, 2048) dtype=int
-        if array_library == "torch":
-            self.logits = torch.rand((4, 30000), dtype=torch.float)
-            self.input_ids = torch.randint(
-                low=0, high=30000, size=(4, 2048), dtype=torch.int
-            )
-        elif array_library == "numpy":
-            self.logits = np.random.rand(4, 30000).astype(np.float32)
-            self.input_ids = np.random.randint(low=0, high=30000, size=(4, 2048))
-        elif array_library == "mlx":
-            self.logits = mx.random.uniform(
-                low=-1e9, high=1e9, shape=(4, 30000), dtype=mx.float32
-            )
-            self.input_ids = mx.random.randint(
-                low=0, high=30000, shape=(4, 2048), dtype=mx.int32
-            )
-        else:
-            raise ValueError
-
-    def time_logits_processor(self, array_library):
+        self.logits, self.input_ids = get_mock_processor_inputs(array_library)
+
+    def time_passthrough(self, *params):
+        self.logits_processor(self.input_ids, self.logits)
+
+
+class LogitsProcessorStructuredBenchmark:
+    """
+    Benchmark structured generation mask application for single decoder pass
+    """
+
+    array_libraries = ["torch", "numpy"]
+    if is_mlx_lm_allowed():
+        array_libraries += ["mlx"]
+    # PR TODO
+    if torch.cuda.is_available():
+        array_libraries += ["torch_cuda"]
+
+    # accept very many or very few tokens, respectively
+    patterns = [r"[^Z]*", "Z*"]
+
+    params = [array_libraries, patterns]
+    param_names = ["array_library, pattern"]
+
+    def setup(self, array_library, pattern):
+        tokenizer = models.transformers("facebook/opt-125m", device="cpu").tokenizer
+
+        self.logits_processor = RegexLogitsProcessor(pattern, tokenizer)
+
+        self.logits, self.input_ids = get_mock_processor_inputs(
+            array_library, len(tokenizer.vocabulary)
+        )
+
+    def time_structured_generation(self, array_library, pattern):
         self.logits_processor(self.input_ids, self.logits)
diff --git a/docs/community/contribute.md b/docs/community/contribute.md
index b336eacad..d5568f47c 100644
--- a/docs/community/contribute.md
+++ b/docs/community/contribute.md
@@ -66,17 +66,21 @@ You can run the benchmark test suite locally with the following command:
 asv run --config benchmarks/asv.conf.json
 ```
 
-Run a specific test:
+Caveats:
+- If you're on a device with CUDA, you must add the argument `--launch-method spawn`
+- Uncommitted code will not be benchmarked, you must first commit your changes.
+
+#### Run a specific test:
 ```
 asv run --config benchmarks/asv.conf.json -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
 ```
 
-Profile a specific test:
+#### Profile a specific test:
 ```
 asv run --config benchmarks/asv.conf.json --profile -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
 ```
 
-Compare to `origin/main`
+#### Compare to `origin/main`
 ```
 get fetch origin
 asv continuous origin/main HEAD --config benchmarks/asv.conf.json
diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index d247db62b..792eba34f 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -2,6 +2,7 @@
 from typing import TYPE_CHECKING, List, Optional, Protocol, Tuple, Union
 
 import interegular
+import torch
 from lark import Lark
 
 from outlines import grammars
@@ -146,6 +147,13 @@ def __init__(self, regex_string: str, tokenizer):
         self.eos_token_id = tokenizer.eos_token_id
         self.final_states = fsm_finals | {-1}
 
+        # cache returned masks token masks
+        # this increases performance of the mask substantially
+        self.states_to_token_mask = {
+            state: torch.tensor(list(next_tokens_to_end_states.keys()))
+            for state, next_tokens_to_end_states in self.states_to_token_maps.items()
+        }
+
     def get_next_instruction(self, state: int) -> Instruction:
         """Return the next instruction for guided generation.
 
@@ -169,11 +177,11 @@ def get_next_instruction(self, state: int) -> Instruction:
         A `Generate` instance that contains the model and the allowed token ids.
 
         """
-        next_tokens_to_end_states = self.states_to_token_maps.get(state)
-        if next_tokens_to_end_states is None:
-            return Write([self.eos_token_id])
+        next_tokens_mask = self.states_to_token_mask.get(state)
+        if next_tokens_mask is None:
+            return Write(torch.tensor([self.eos_token_id]))
 
-        return Generate(list(next_tokens_to_end_states.keys()))
+        return Generate(next_tokens_mask)
 
     def get_next_state(self, state: int, token_id: int) -> int:
         """Update the state of the guide.
diff --git a/tests/fsm/test_fsm.py b/tests/fsm/test_fsm.py
index 97d120f20..21163b70d 100644
--- a/tests/fsm/test_fsm.py
+++ b/tests/fsm/test_fsm.py
@@ -3,6 +3,13 @@
 from outlines.fsm.fsm import CFGFSM, RegexFSM, StopAtEosFSM
 
 
+def assert_expected_tensor_ids(tensor, ids):
+    assert len(tensor) == len(ids)
+    norm_tensor = sorted(map(int, tensor))
+    norm_ids = sorted(map(int, tensor))
+    assert norm_tensor == norm_ids, (norm_tensor, norm_ids)
+
+
 def test_stop_at_eos():
     class MockTokenizer:
         vocabulary = {"a": 1, "eos": 2}
@@ -50,7 +57,7 @@ def convert_token_to_string(self, token):
         fsm = RegexFSM(regex_str, tokenizer)
 
     assert fsm.states_to_token_maps == {0: {1: 1}}
-    assert fsm.allowed_token_ids(state=0) == [1]
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=0), [1])
     assert fsm.next_state(state=0, token_id=1) == 1
     assert fsm.next_state(state=0, token_id=tokenizer.eos_token_id) == -1
 
@@ -111,27 +118,27 @@ def decode(self, token_ids):
     with pytest.warns(UserWarning):
         fsm = CFGFSM(cfg_str, tokenizer)
 
-    assert set(fsm.allowed_token_ids(state=fsm.start_state)) == {1, 3, 5}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1, 3, 5])
     state = fsm.next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "{"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 2, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
     state = fsm.next_state(state=state, token_id=3)
     assert fsm.generation == "{["
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 3, 4}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 3, 4])
     state = fsm.next_state(state=state, token_id=4)
     assert fsm.generation == "{[]"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {2}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [2])
     state = fsm.next_state(state=state, token_id=2)
     assert fsm.generation == "{[]}"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {5}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [5])
     state = fsm.next_state(state=state, token_id=5)
     assert fsm.generation == "{[]}"
     assert fsm.is_final_state(state)
@@ -164,24 +171,24 @@ def decode(self, token_ids):
     with pytest.warns(UserWarning):
         fsm = CFGFSM(cfg_str, tokenizer)
 
-    assert set(fsm.allowed_token_ids(state=fsm.start_state)) == {1}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1])
     state = fsm.next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "("
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 2}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2])
     state = fsm.next_state(state=state, token_id=2)
     assert fsm.generation == "()"
     assert not fsm.is_final_state(state)
 
     # possible to continue or terminate
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 3])
     state = fsm.next_state(state=state, token_id=3)  # feed eos
     assert fsm.generation == "()"
     assert fsm.is_final_state(state)
 
     # once eos generated, can only terminate
-    assert set(fsm.allowed_token_ids(state=state)) == {3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [3])
 
 
 def test_cfg_ignore_directive():
@@ -214,38 +221,38 @@ def decode(self, token_ids):
 
     state = 0
 
-    assert set(fsm.allowed_token_ids(state=0)) == {1, 2}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=0), [1, 2])
     state = fsm.next_state(state=0, token_id=2)
     assert fsm.generation == " "
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=0)) == {1, 2}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=0), [1, 2])
     state = fsm.next_state(state=0, token_id=1)
     assert fsm.generation == " a"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 2, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
     state = fsm.next_state(state=state, token_id=2)
     assert fsm.generation == " a "
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 2, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
     state = fsm.next_state(state=state, token_id=2)
     assert fsm.generation == " a  "
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 2, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
     state = fsm.next_state(state=state, token_id=1)
     assert fsm.generation == " a  a"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 2, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
     state = fsm.next_state(state=state, token_id=3)
     assert fsm.generation == " a  a"
     assert fsm.is_final_state(state)
 
     # once eos generated, can only terminate
-    assert set(fsm.allowed_token_ids(state=state)) == {3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [3])
 
 
 def test_cfg_multitoken_terminal():
@@ -274,19 +281,19 @@ def decode(self, token_ids):
     with pytest.warns(UserWarning):
         fsm = CFGFSM(cfg_str, tokenizer)
 
-    assert set(fsm.allowed_token_ids(state=fsm.start_state)) == {1, 2}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1, 2])
     assert fsm.reset_state  # starting new regex
     state = fsm.next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "a"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1])
     assert not fsm.reset_state  # continuing current regex
     state = fsm.next_state(state=state, token_id=1)
     assert fsm.generation == "aa"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [3])
     assert not fsm.reset_state  # completing current regex
     state = fsm.next_state(state=state, token_id=3)
     assert fsm.generation == "aa"
@@ -319,27 +326,27 @@ def decode(self, token_ids):
     with pytest.warns(UserWarning):
         fsm = CFGFSM(cfg_str, tokenizer)
 
-    assert set(fsm.allowed_token_ids(state=fsm.start_state)) == {1, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1, 3])
     state = fsm.next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "("
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {1, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 3])
     state = fsm.next_state(state=state, token_id=3)
     assert fsm.generation == "(a"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {2, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [2, 3])
     state = fsm.next_state(state=state, token_id=3)
     assert fsm.generation == "(aa"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {2, 3}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [2, 3])
     state = fsm.next_state(state=state, token_id=2)
     assert fsm.generation == "(aa)"
     assert not fsm.is_final_state(state)
 
-    assert set(fsm.allowed_token_ids(state=state)) == {4}
+    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [4])
     state = fsm.next_state(state=state, token_id=4)
     assert fsm.generation == "(aa)"
     assert fsm.is_final_state(state)
diff --git a/tests/fsm/test_guide.py b/tests/fsm/test_guide.py
index 68f7e33ee..20ba75893 100644
--- a/tests/fsm/test_guide.py
+++ b/tests/fsm/test_guide.py
@@ -3,6 +3,13 @@
 from outlines.fsm.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
 
 
+def assert_expected_tensor_ids(tensor, ids):
+    assert len(tensor) == len(ids)
+    norm_tensor = sorted(map(int, tensor))
+    norm_ids = sorted(map(int, tensor))
+    assert norm_tensor == norm_ids, (norm_tensor, norm_ids)
+
+
 def test_stop_at_eos():
     class MockTokenizer:
         vocabulary = {"a": 1, "eos": 2}
@@ -56,7 +63,7 @@ def convert_token_to_string(self, token):
 
     instruction = fsm.get_next_instruction(0)
     assert isinstance(instruction, Generate)
-    assert instruction.tokens == [1]
+    assert_expected_tensor_ids(instruction.tokens, [1])
 
     assert fsm.get_next_state(state=0, token_id=1) == 1
     assert fsm.get_next_state(state=0, token_id=tokenizer.eos_token_id) == -1
@@ -102,7 +109,7 @@ def convert_token_to_string(self, token):
 
     instruction = fsm.get_next_instruction(0)
     assert isinstance(instruction, Generate)
-    assert instruction.tokens == [5, 4]
+    assert_expected_tensor_ids(instruction.tokens, [5, 4])
 
     assert fsm.get_next_state(state=0, token_id=5) == 1
     assert fsm.get_next_state(state=0, token_id=tokenizer.eos_token_id) == -1
@@ -149,7 +156,7 @@ def convert_token_to_string(self, token):
 
     instruction = fsm.get_next_instruction(0)
     assert isinstance(instruction, Generate)
-    assert instruction.tokens == [5, 10]
+    assert_expected_tensor_ids(instruction.tokens, [5, 10])
 
     assert fsm.get_next_state(state=0, token_id=5) == 1
     assert fsm.get_next_state(state=0, token_id=tokenizer.eos_token_id) == -1
@@ -209,35 +216,35 @@ def decode(self, token_ids):
 
     instruction = fsm.get_next_instruction(fsm.start_state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 3, 5}
+    assert_expected_tensor_ids(instruction.tokens, [1, 3, 5])
     state = fsm.get_next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "{"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
     state = fsm.get_next_state(state=state, token_id=3)
     assert fsm.generation == "{["
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 3, 4}
+    assert_expected_tensor_ids(instruction.tokens, [1, 3, 4])
     state = fsm.get_next_state(state=state, token_id=4)
     assert fsm.generation == "{[]"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {2}
+    assert_expected_tensor_ids(instruction.tokens, [2])
     state = fsm.get_next_state(state=state, token_id=2)
     assert fsm.generation == "{[]}"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Write)
-    assert set(instruction.tokens) == {5}
+    assert_expected_tensor_ids(instruction.tokens, [5])
     state = fsm.get_next_state(state=state, token_id=5)
     assert fsm.generation == "{[]}"
     assert fsm.is_final_state(state)
@@ -270,14 +277,14 @@ def decode(self, token_ids):
 
     instruction = fsm.get_next_instruction(fsm.start_state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1}
+    assert_expected_tensor_ids(instruction.tokens, [1])
     state = fsm.get_next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "("
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2])
     state = fsm.get_next_state(state=state, token_id=2)
     assert fsm.generation == "()"
     assert not fsm.is_final_state(state)
@@ -285,7 +292,7 @@ def decode(self, token_ids):
     # possible to continue or terminate
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 3])
     state = fsm.get_next_state(state=state, token_id=3)  # feed eos
     assert fsm.generation == "()"
     assert fsm.is_final_state(state)
@@ -293,7 +300,7 @@ def decode(self, token_ids):
     # once eos generated, can only terminate
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Write)
-    assert set(instruction.tokens) == {3}
+    assert_expected_tensor_ids(instruction.tokens, [3])
 
 
 def test_cfg_ignore_directive():
@@ -326,42 +333,42 @@ def decode(self, token_ids):
 
     instruction = fsm.get_next_instruction(0)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2])
     state = fsm.get_next_state(state=0, token_id=2)
     assert fsm.generation == " "
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(0)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2])
     state = fsm.get_next_state(state=0, token_id=1)
     assert fsm.generation == " a"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
     state = fsm.get_next_state(state=state, token_id=2)
     assert fsm.generation == " a "
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
     state = fsm.get_next_state(state=state, token_id=2)
     assert fsm.generation == " a  "
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
     state = fsm.get_next_state(state=state, token_id=1)
     assert fsm.generation == " a  a"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
     state = fsm.get_next_state(state=state, token_id=3)
     assert fsm.generation == " a  a"
     assert fsm.is_final_state(state)
@@ -369,7 +376,7 @@ def decode(self, token_ids):
     # once eos generated, can only terminate
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Write)
-    assert set(instruction.tokens) == {3}
+    assert_expected_tensor_ids(instruction.tokens, [3])
 
 
 def test_cfg_multitoken_terminal():
@@ -398,7 +405,7 @@ def decode(self, token_ids):
 
     instruction = fsm.get_next_instruction(fsm.start_state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 2}
+    assert_expected_tensor_ids(instruction.tokens, [1, 2])
     assert fsm.reset_state  # starting new regex
     state = fsm.get_next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "a"
@@ -406,7 +413,7 @@ def decode(self, token_ids):
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1}
+    assert_expected_tensor_ids(instruction.tokens, [1])
     assert not fsm.reset_state  # continuing current regex
     state = fsm.get_next_state(state=state, token_id=1)
     assert fsm.generation == "aa"
@@ -414,7 +421,7 @@ def decode(self, token_ids):
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Write)
-    assert set(instruction.tokens) == {3}
+    assert_expected_tensor_ids(instruction.tokens, [3])
     assert not fsm.reset_state  # completing current regex
     state = fsm.get_next_state(state=state, token_id=3)
     assert fsm.generation == "aa"
@@ -447,35 +454,35 @@ def decode(self, token_ids):
 
     instruction = fsm.get_next_instruction(fsm.start_state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 3])
     state = fsm.get_next_state(state=fsm.start_state, token_id=1)
     assert fsm.generation == "("
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {1, 3}
+    assert_expected_tensor_ids(instruction.tokens, [1, 3])
     state = fsm.get_next_state(state=state, token_id=3)
     assert fsm.generation == "(a"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {2, 3}
+    assert_expected_tensor_ids(instruction.tokens, [2, 3])
     state = fsm.get_next_state(state=state, token_id=3)
     assert fsm.generation == "(aa"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Generate)
-    assert set(instruction.tokens) == {2, 3}
+    assert_expected_tensor_ids(instruction.tokens, [2, 3])
     state = fsm.get_next_state(state=state, token_id=2)
     assert fsm.generation == "(aa)"
     assert not fsm.is_final_state(state)
 
     instruction = fsm.get_next_instruction(state)
     assert isinstance(instruction, Write)
-    assert set(instruction.tokens) == {4}
+    assert_expected_tensor_ids(instruction.tokens, [4])
     state = fsm.get_next_state(state=state, token_id=4)
     assert fsm.generation == "(aa)"
     assert fsm.is_final_state(state)

From ea5fbdbbb035abe2b46afe7d16f134a4ef973cde Mon Sep 17 00:00:00 2001
From: Rohan Shah <57906961+rshah713@users.noreply.github.com>
Date: Wed, 10 Jul 2024 10:25:22 -0400
Subject: [PATCH 003/505] fix json prompt example

---
 docs/reference/prompting.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/docs/reference/prompting.md b/docs/reference/prompting.md
index 34860fce0..8ea9c0243 100644
--- a/docs/reference/prompting.md
+++ b/docs/reference/prompting.md
@@ -260,7 +260,8 @@ pretty print a dictionary from within an Outlines prompt function
     def my_prompt(response_model):
         """{{ response_model | schema }}"""
 
-    my_prompt(MyResponse)
+    prompt = my_prompt(MyResponse)
+    print(prompt)
     # {
     #   "field1": "an int",
     #   "field2": "<field2>"

From 497ed9fdeaa1583a46fdbdd6eedb04d052fc44cf Mon Sep 17 00:00:00 2001
From: Rin <shinohara-rin@users.noreply.github.com>
Date: Thu, 11 Jul 2024 14:36:32 +0100
Subject: [PATCH 004/505] Fix broken link in README.md regarding Serving with
 vLLM

updated the link to https://outlines-dev.github.io/outlines/reference/serve/vllm/
---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index aeb1126b6..4ed1e9237 100644
--- a/README.md
+++ b/README.md
@@ -37,7 +37,7 @@ First time here? Go to our [setup guide](https://outlines-dev.github.io/outlines
 - [x] 💾 Caching of generations
 - [x] 🗂️ Batch inference
 - [x] 🎲 Sample with the greedy, multinomial and beam search algorithms (and more to come!)
-- [x] 🚀 [Serve with vLLM](https://outlines-dev.github.io/outlines/reference/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
+- [x] 🚀 [Serve with vLLM](https://outlines-dev.github.io/outlines/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
 
 
 Outlines 〰 has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!

From 0383ce1c17114cf5031e3a95d8781eb6a42e6365 Mon Sep 17 00:00:00 2001
From: Jules Gagnon-Marchand <jules.gagnonm.alt@gmail.com>
Date: Sun, 7 Jul 2024 20:37:27 -0700
Subject: [PATCH 005/505] Fix bug in batched multi sample generation

---
 outlines/generate/api.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index 51a995664..8b2561075 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -372,7 +372,7 @@ def token_generator() -> Iterator[Union[List[str], str, List[List[str]]]]:
                 previously_generated_sequences = generated_sequences
                 # We reshape the output to (batch_size, sample_size)
                 output: List[List[str]] = list()
-                for i in range(batch_size):
+                for i in range(0, batch_size * num_samples, num_samples):
                     output.append(next_tokens[i : i + num_samples])
 
                 # We remove leading dimensions for the output

From 8c5f1d86889cf9ea110b5736eb676ba2ef4f4353 Mon Sep 17 00:00:00 2001
From: Jules Gagnon-Marchand <jules.gagnonm.alt@gmail.com>
Date: Sun, 7 Jul 2024 21:38:53 -0700
Subject: [PATCH 006/505] Fixed __call__ as well

---
 outlines/generate/api.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index 8b2561075..f739dc98c 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -231,8 +231,8 @@ def __call__(
 
         # We reshape the output to (batch_size, sample_size)
         output: List[List[FormattedOutput]] = list()
-        for i in range(batch_size):
-            output.append(formatted[i : i + num_samples])
+        for i in range(0, batch_size * num_samples, num_samples):
+            output.append(formatted[i : i + num_samples]
 
         # We remove leading dimensions for the output
         if batch_size == 1 and num_samples == 1:

From b54a96404e880383ad50071af5e9a5560680c4a3 Mon Sep 17 00:00:00 2001
From: Jules Gagnon-Marchand <jules.gagnonm.alt@gmail.com>
Date: Mon, 8 Jul 2024 09:35:32 -0700
Subject: [PATCH 007/505] Update outlines/generate/api.py

Co-authored-by: Patrice Bechard <bechardpatrice@gmail.com>
---
 outlines/generate/api.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index f739dc98c..4104e3080 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -232,7 +232,7 @@ def __call__(
         # We reshape the output to (batch_size, sample_size)
         output: List[List[FormattedOutput]] = list()
         for i in range(0, batch_size * num_samples, num_samples):
-            output.append(formatted[i : i + num_samples]
+            output.append(formatted[i : i + num_samples])
 
         # We remove leading dimensions for the output
         if batch_size == 1 and num_samples == 1:

From 59d88dc02df78d7810e8e51cc7860197392a89ca Mon Sep 17 00:00:00 2001
From: George Tsiolis <tsiolis.g@gmail.com>
Date: Fri, 12 Jul 2024 17:21:19 +0300
Subject: [PATCH 008/505] Update file contributors style for docs

---
 docs/stylesheets/extra.css | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index 8611360d2..fdf3acb9e 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -9,9 +9,6 @@
   --md-code-fg-color: #FFFFFF;
   --md-text-font-family: "Inter";
   --md-code-font: "Source Code Pro Custom";
-  --md-default-fg-color--light: #D8DEE9;
-  --md-default-fg-color--lighter: #E5E9F0;
-  --md-default-fg-color--lightest: #ECEFF4;
 }
 
 .index-pre-code {
@@ -23,6 +20,19 @@
   text-align: left;
 }
 
+.md-clipboard::after {
+  color: #FFFFFF;
+  transition: color 0.3s ease-in-out;
+}
+
+.md-clipboard:hover::after {
+  color: #D8DEE9;
+}
+
+.md-source-file {
+  text-align: center;
+  padding: 24px 0;
+}
 
 .md-typeset pre>code {
   border-radius: .2rem;

From a48f86ffb4b149ba2237dae72647e8adc40e778f Mon Sep 17 00:00:00 2001
From: Tal <talboren2@gmail.com>
Date: Sat, 13 Jul 2024 21:47:09 +0300
Subject: [PATCH 009/505] Add missing blank space (#1036)

---
 outlines/generate/format.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/generate/format.py b/outlines/generate/format.py
index d87a3fe70..d7b7e164d 100644
--- a/outlines/generate/format.py
+++ b/outlines/generate/format.py
@@ -41,5 +41,5 @@ def format(model, python_type, sampler: Sampler = multinomial()) -> SequenceGene
 def format_openai(model, python_type, sampler: Sampler = multinomial()):
     raise NotImplementedError(
         "Cannot use Python type-structured generation with an OpenAI model"
-        + "due to the limitations of the OpenAI API."
+        + " due to the limitations of the OpenAI API."
     )

From 62b7601d3a6aa56cd11d51b1404a6afd6e7efb91 Mon Sep 17 00:00:00 2001
From: Franz Louis Cesista <franzlouiscesista@gmail.com>
Date: Mon, 15 Jul 2024 01:02:35 +0800
Subject: [PATCH 010/505] Adds support for custom regex parsers (for multimodal
 structured generation) (#1039)

As [discussed in our Discord
server](https://discord.com/channels/1182316225284554793/1182317446225481788/1261998326077984802)

This PR adds support for custom regex parsers. This doesn't change the
behavior of Outlines by default. But this allows us to write custom
`Guide` classes that uses custom regex parsers for e.g. multimodal
generation.

Also improves documentation
---
 outlines/fsm/guide.py        |  54 ++++++++++++++---
 outlines/fsm/regex.py        | 112 ++++++++++++++++++++++++++++++-----
 outlines/models/llamacpp.py  |   2 +-
 outlines/models/tokenizer.py |   2 +-
 tests/fsm/test_regex.py      |   2 +
 5 files changed, 148 insertions(+), 24 deletions(-)

diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index 792eba34f..aa694232e 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -1,5 +1,15 @@
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, List, Optional, Protocol, Tuple, Union
+from typing import (
+    TYPE_CHECKING,
+    Callable,
+    Dict,
+    List,
+    Optional,
+    Protocol,
+    Set,
+    Tuple,
+    Union,
+)
 
 import interegular
 import torch
@@ -108,16 +118,44 @@ def copy(self):
 
 @cache()
 def create_states_mapping(
-    regex_string: str, tokenizer: "Tokenizer"
-) -> Tuple[dict, set, set]:
+    regex_string: str,
+    tokenizer: "Tokenizer",
+    regex_parser: Callable[[str], interegular.Pattern] = interegular.parse_pattern,
+    frozen_tokens: List[str] = [],
+) -> Tuple[Dict[int, Dict[int, int]], Set[int], set]:
     """Create the variables related to the mapping between states and tokens
-    The parameters of the function are used for caching purpose
+    The parameters of the function are used for caching purpose.
+
+    Parameters
+    ----------
+    regex_string: (`str`):
+        The regular expression string to generate a states mapping for.
+    tokenizer: (`Tokenizer`):
+        The model's tokenizer.
+    regex_parser: (`Callable[[str], interegular.Pattern]`, *optional*):
+        A function that parses a regex string into an `interegular` Pattern object.
+    frozen_tokens: (`List[str]`, *optional*):
+        A list of tokens that should be kept as-is when expanding the token-level FSM
+        into a byte-level FSM. Defaults to an empty list.
+
+    Returns
+    -------
+    states_to_token_maps: (`Dict[int, Dict[int, int]]`):
+        A mapping from states to a mapping from token ids originating from that state
+        to the next state to transition to given that token. The structure is as follows:
+        (origin_state -> (token_id -> next_state))
+    empty_token_ids: (`Set[int]`):
+        A set of token ids that correspond to empty strings.
+    final_states: (`set`):
+        A set of final states in the FSM.
     """
-    regex_pattern = interegular.parse_pattern(regex_string)
-    byte_fsm = make_byte_level_fsm(regex_pattern.to_fsm().reduce(), keep_utf8=True)
+    regex_pattern = regex_parser(regex_string)
+    byte_fsm = make_byte_level_fsm(
+        regex_pattern.to_fsm().reduce(), keep_utf8=True, frozen_tokens=frozen_tokens
+    )
     regex_fsm, _ = make_deterministic_fsm(byte_fsm)
     states_to_token_maps, empty_token_ids = create_fsm_index_tokenizer(
-        regex_fsm, tokenizer
+        regex_fsm, tokenizer, frozen_tokens=frozen_tokens
     )
 
     # We make sure that it is possible to generate strings in the language
@@ -138,7 +176,7 @@ class RegexGuide(Guide):
 
     initial_state = 0
 
-    def __init__(self, regex_string: str, tokenizer):
+    def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
         (
             self.states_to_token_maps,
             self.empty_token_ids,
diff --git a/outlines/fsm/regex.py b/outlines/fsm/regex.py
index ff8dd9131..8cfd81ead 100644
--- a/outlines/fsm/regex.py
+++ b/outlines/fsm/regex.py
@@ -199,11 +199,27 @@ def byte_symbol(byte: int) -> str:
     return f"\x00{byte:02X}" if byte >= 0x80 else chr(byte)
 
 
-def make_byte_level_fsm(fsm: FSM, keep_utf8=False) -> FSM:
+def make_byte_level_fsm(
+    fsm: FSM, keep_utf8: bool = False, frozen_tokens: List[str] = []
+) -> FSM:
     """Convert an FSM to a byte-level FSM, expanding multi-byte characters as
-    sequences of single-byte transitions. If keep_utf8 is set, the original
-    utf-8 characters are kept in the alphabet.
-    NOTE: we're representing bytes as strings to keep it type-compatible.
+    sequences of single-byte transitions.
+
+    Parameters
+    ----------
+    fsm: (`interegular.FSM`):
+        The token-level FSM to convert to a byte-level FSM.
+    keep_utf8: (`bool`, *optional*):
+        If set to True, the original utf-8 characters are kept as-is. Defaults to
+        False. NOTE: we're representing bytes as strings to keep it type-compatible.
+    frozen_tokens: (`List[str]`, *optional*):
+        A list of tokens that should be kept as-is in the byte-level FSM. That is,
+        these tokens will not be expanded into byte-level transitions. Defaults to
+        an empty list.
+
+    Returns
+    -------
+    `interegular.FSM`: A byte-level FSM.
     """
 
     anything_else_key = fsm.alphabet[anything_else]
@@ -218,8 +234,8 @@ def make_byte_level_fsm(fsm: FSM, keep_utf8=False) -> FSM:
     all_bytes: Set[int] = set()
     max_key = max(fsm.alphabet.values())
     for symbol, transition_key in fsm.alphabet.items():
-        assert symbol == anything_else or len(symbol) == 1
-        if symbol == anything_else or ord(symbol) < 0x80:
+        assert symbol == anything_else or symbol in frozen_tokens or len(symbol) == 1
+        if symbol == anything_else or symbol in frozen_tokens or ord(symbol) < 0x80:
             symbol_mapping[symbol] = transition_key
         else:
             if keep_utf8:
@@ -714,15 +730,40 @@ def get_vocabulary_transition_keys(
     alphabet_symbol_mapping: Dict[str, int],
     alphabet_anything_value: int,
     vocabulary: List[Tuple[str, Sequence[int]]],
+    frozen_tokens: List[str] = numba.typed.List.empty_list(numba.types.unicode_type),
 ) -> List[Sequence[int]]:
     """
     Calculate the sequence transition keys for each token str within a vocabulary
+
+    Parameters
+    ----------
+    alphabet_symbol_mapping: (`Dict[str, int]`):
+        A mapping from an alphabet symbol in a FSM to its corresponding transition key.
+    alphabet_anything_value: (`int`):
+        The transition key for the anything_else symbol in the FSM.
+    vocabulary: (`List[Tuple[str, Sequence[int]]]`):
+        A list of tuples, each containing a token and a list of equivalent token ids.
+    frozen_tokens: (`List[str]`, *optional*):
+        A list of tokens that are kept as-is when transforming the FSM.
+        Defaults to an empty list.
+
+    Returns
+    -------
+    `List[Sequence[int]]`:
+        A list of token transition keys for each token in the vocabulary.
     """
     vocab_transition_keys = numba.typed.List.empty_list(numba.int64[:])
     for token_str, _ in vocabulary:
-        token_transition_keys = get_token_transition_keys(
-            alphabet_symbol_mapping, alphabet_anything_value, token_str
-        )
+        # Since these tokens are not expanded into byte-level transitions, we can
+        # simply get their transition keys directly.
+        if token_str in frozen_tokens:
+            token_transition_keys = np.array(
+                [alphabet_symbol_mapping[token_str]], dtype=np.int64
+            )
+        else:
+            token_transition_keys = get_token_transition_keys(
+                alphabet_symbol_mapping, alphabet_anything_value, token_str
+            )
         vocab_transition_keys.append(token_transition_keys)
 
     return vocab_transition_keys
@@ -731,8 +772,26 @@ def get_vocabulary_transition_keys(
 def create_fsm_index_end_to_end(
     fsm_info: FSMInfo,
     vocabulary: List[Tuple[str, Sequence[int]]],
+    frozen_tokens: List[str] = [],
 ) -> Dict[int, Set[Tuple[int, int]]]:
-    """Create an FSM state-to-vocabulary map/index through end-to-end token parsing."""
+    """Create an FSM state-to-vocabulary map/index through end-to-end token parsing.
+
+    Parameters
+    ----------
+    fsm_info: (`interegular.FSMInfo`):
+        The FSM information object containing the FSM's alphabet, transitions, initial
+        and final states, and other relevant information.
+    vocabulary: (`List[Tuple[str, Sequence[int]]]`):
+        A list of tuples, each containing a token and a list of equivalent token ids.
+    frozen_tokens: (`List[str]`, *optional*):
+        A list of tokens that are kept as-is when transforming the FSM.
+
+    Returns
+    -------
+    `Dict[int, Set[Tuple[int, int]]]`:
+        A mapping from FSM states to sets of tuples containing token ids and the end
+        states of the FSM after parsing the token.
+    """
 
     # TODO: Consider using a `List` of `Set`s instead; that way we can JIT this
     # code, too.
@@ -750,6 +809,11 @@ def create_fsm_index_end_to_end(
         fsm_info.alphabet_symbol_mapping,
         fsm_info.alphabet_anything_value,
         vocabulary,
+        frozen_tokens=(
+            numba.typed.List(frozen_tokens)
+            if len(frozen_tokens) > 0
+            else numba.typed.List.empty_list(numba.types.unicode_type)
+        ),
     )
 
     while next_states:
@@ -883,21 +947,41 @@ def reduced_vocabulary(
 
 
 def create_fsm_index_tokenizer(
-    fsm: BetterFSM,
-    tokenizer: "Tokenizer",
+    fsm: BetterFSM, tokenizer: "Tokenizer", frozen_tokens: List[str] = []
 ) -> Tuple[Dict[int, Dict[int, int]], Set[int]]:
     """Construct an FMS index from a tokenizer.
 
     This uses the end-to-end approach of `create_fsm_index_end_to_end`.
 
+    Parameters
+    ----------
+    fsm: (`BetterFSM`):
+        A cache-friendly FSM. Other interegular FSMs can also be used, but caching
+        may not work as expected.
+    tokenizer: (`Tokenizer`):
+        The model's tokenizer.
+    frozen_tokens: (`List[str]`, *optional*):
+        A list of tokens that should be kept as-is when expanding the token-level
+        FSM into a byte-level FSM. Defaults to an empty list.
+
+    Returns
+    -------
+    states_to_token_maps: (`Dict[int, Dict[int, int]]`):
+        A mapping from states to a mapping from token ids originating from that state
+        to the next state to transition to given that token. The structure is as follows:
+        (origin_state -> (token_id -> next_state))
+    empty_token_ids: (`Set[int]`):
+        A set of token ids that correspond to empty strings.
+
     .. warning::
 
         `fsm` needs to be deterministically ordered so that future caching makes sense.
-
     """
     vocabulary, empty_token_ids = reduced_vocabulary(tokenizer)
 
-    states_to_token_subsets = create_fsm_index_end_to_end(fsm.fsm_info, vocabulary)
+    states_to_token_subsets = create_fsm_index_end_to_end(
+        fsm.fsm_info, vocabulary, frozen_tokens
+    )
 
     # Allow transitions to EOS from all terminals FSM states that are
     # reachable
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index aa260a457..7e223a9fa 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -27,7 +27,7 @@ def __init__(self, model: "Llama"):
         self.eos_token_id = model.token_eos()
         self.eos_token = model.tokenizer().decode([self.eos_token_id])
         self.pad_token_id = self.eos_token_id
-        self.special_tokens: Set[int] = set()
+        self.special_tokens: Set[str] = set()
 
         self.vocabulary: Dict[str, int] = dict()
 
diff --git a/outlines/models/tokenizer.py b/outlines/models/tokenizer.py
index 949414a44..1a5708d85 100644
--- a/outlines/models/tokenizer.py
+++ b/outlines/models/tokenizer.py
@@ -9,7 +9,7 @@ class Tokenizer(Hashable, Protocol):
     eos_token_id: int
     pad_token_id: int
     vocabulary: Dict[str, int]
-    special_tokens: Set[int]
+    special_tokens: Set[str]
 
     def encode(
         self, prompt: Union[str, List[str]]
diff --git a/tests/fsm/test_regex.py b/tests/fsm/test_regex.py
index 7ba4f9caa..fa72ad0dc 100644
--- a/tests/fsm/test_regex.py
+++ b/tests/fsm/test_regex.py
@@ -600,6 +600,7 @@ def convert_token_to_string(self, token):
         regex_fsm.fsm_info.alphabet_symbol_mapping,
         regex_fsm.fsm_info.alphabet_anything_value,
         vocabulary,
+        numba.typed.List.empty_list(numba.types.unicode_type),
     )
 
     token_str_to_tranition_keys = {
@@ -637,6 +638,7 @@ def convert_token_to_string(self, token):
         regex_fsm.fsm_info.alphabet_symbol_mapping,
         regex_fsm.fsm_info.alphabet_anything_value,
         vocabulary,
+        numba.typed.List.empty_list(numba.types.unicode_type),
     )
 
     token_str_trans_key_seq = {

From 82248550b3aed82f8d95271686759368437fc9aa Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 1 Jul 2024 08:56:49 -0500
Subject: [PATCH 011/505] Use LogitsProcessors for models.transformers, apply
 in outlines.generate.*

---
 README.md                                     |   5 +-
 docs/reference/models/transformers.md         |  53 ++-
 docs/reference/text.md                        |   5 +-
 examples/llamacpp_example.py                  |   6 +-
 outlines/generate/cfg.py                      |  42 +--
 outlines/generate/regex.py                    |   6 +-
 outlines/generate/text.py                     |   5 +-
 outlines/models/__init__.py                   |   2 +-
 outlines/models/transformers.py               | 184 +++++++++-
 tests/generate/test_integration_llamacpp.py   |   9 +-
 .../generate/test_integration_transformers.py | 320 +++++-------------
 tests/models/test_transformers.py             |   9 +-
 12 files changed, 358 insertions(+), 288 deletions(-)

diff --git a/README.md b/README.md
index 4ed1e9237..96d02dbe1 100644
--- a/README.md
+++ b/README.md
@@ -191,10 +191,9 @@ model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
 generator = outlines.generate.json(model, Character)
 
 # Draw a sample
-rng = torch.Generator(device="cuda")
-rng.manual_seed(789001)
+seed = 789001
 
-character = generator("Give me a character description", rng=rng)
+character = generator("Give me a character description", seed=seed)
 
 print(repr(character))
 # Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 286df4367..7c1febd02 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -15,7 +15,7 @@ Outlines provides an integration with the `torch` implementation of causal model
 ```python
 from outlines import models
 
-model = models.transformers("mistralai/Mistral-7B-v0.1", device="cuda")
+model = models.transformers("mistralai/Mistral-7B-v0.3", device="cuda")
 ```
 
 If you need more fine-grained control you can also initialize the model and tokenizer separately:
@@ -30,4 +30,55 @@ tokenizer = AutoTokenizer.from_pretrained("gpt2")
 model = models.Transformers(llm, tokenizer)
 ```
 
+# Using Logits Processors
+
+There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
+- 1) Use Outlines generation wrapper, `outlines.models.transformers`
+- 2) Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
+
+Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
+
+## Example: `outlines.models.transformers`
+
+```
+import outlines
+
+time_regex_pattern = r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?"
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+generator = outlines.generate.regex(model, time_regex_pattern)
+
+output = generator("The the best time to visit a dentist is at ")
+print(output)
+# 2:30 pm
+```
+
+## Example: Direct `transformers` library use
+
+```
+import outlines
+import transformers
+
+
+model_uri = "microsoft/Phi-3-mini-4k-instruct"
+
+outlines_tokenizer = outlines.models.TransformerTokenizer(
+    transformers.AutoTokenizer.from_pretrained(model_uri)
+)
+phone_number_logits_processor = outlines.processors.RegexLogitsProcessor(
+    "\\+?[1-9][0-9]{7,14}",  # phone number pattern
+    outlines_tokenizer,
+)
+
+generator = transformers.pipeline('text-generation', model=model_uri)
+
+output = generator(
+    "Jenny gave me her number it's ",
+	logits_processor=transformers.LogitsProcessorList([phone_number_logits_processor])
+)
+print(output)
+# [{'generated_text': "Jenny gave me her number it's 2125550182"}]
+# not quite 8675309 what we expected, but it is a valid phone number
+```
+
 [transformers]: https://github.com/huggingface/transformers
diff --git a/docs/reference/text.md b/docs/reference/text.md
index f364c3d2e..423a26ba6 100644
--- a/docs/reference/text.md
+++ b/docs/reference/text.md
@@ -80,8 +80,7 @@ from outlines import models, generate
 
 model = models.transformers("mistralai/Mistral-7B-v0.1")
 
-rng = torch.Generator(device="cuda")
-rng.manual_seed(789001)
+seed = 789001
 
-answer = generator("What is 2+2?", rng=rng)
+answer = generator("What is 2+2?", seed=seed)
 ```
diff --git a/examples/llamacpp_example.py b/examples/llamacpp_example.py
index 0b478217f..22d0da3ba 100644
--- a/examples/llamacpp_example.py
+++ b/examples/llamacpp_example.py
@@ -1,6 +1,5 @@
 from enum import Enum
 
-import torch
 from pydantic import BaseModel, constr
 
 import outlines
@@ -37,10 +36,9 @@ class Character(BaseModel):
     generator = outlines.generate.json(model, Character)
 
     # Draw a sample
-    rng = torch.Generator(device="cpu")
-    rng.manual_seed(789005)
+    seed = 789005
 
     prompt = "Instruct: You are a leading role play gamer. You have seen thousands of different characters and their attributes.\nPlease return a JSON object with common attributes of an RPG character. Give me a character description\nOutput:"
 
-    sequence = generator(prompt, rng=rng, max_tokens=512)
+    sequence = generator(prompt, seed=seed, max_tokens=512)
     print(sequence)
diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
index e473c26a6..0df833067 100644
--- a/outlines/generate/cfg.py
+++ b/outlines/generate/cfg.py
@@ -1,16 +1,14 @@
 from functools import singledispatch
 
-from outlines.fsm.guide import CFGGuide
-from outlines.generate.api import SequenceGenerator, SequenceGeneratorAdapter
+from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
-from outlines.models.llamacpp import LlamaCpp
-from outlines.models.mlxlm import MLXLM
-from outlines.models.vllm import VLLM
 from outlines.samplers import Sampler, multinomial
 
 
 @singledispatch
-def cfg(model, cfg_str: str, sampler: Sampler = multinomial()) -> SequenceGenerator:
+def cfg(
+    model, cfg_str: str, sampler: Sampler = multinomial()
+) -> SequenceGeneratorAdapter:
     """Generate text in the language of a Context-Free Grammar
 
     Arguments
@@ -24,40 +22,16 @@ def cfg(model, cfg_str: str, sampler: Sampler = multinomial()) -> SequenceGenera
 
     Returns
     -------
-    A `SequenceGenerator` instance that generates text.
+    A `SequenceGeneratorAdapter` instance that generates text.
 
     """
-    fsm = CFGGuide(cfg_str, model.tokenizer)
-    device = model.device
-    generator = SequenceGenerator(fsm, model, sampler, device)
-
-    return generator
-
-
-@cfg.register(MLXLM)
-@cfg.register(VLLM)
-def cfg_unimplemented(
-    model,
-    cfg_str: str,
-    sampler: Sampler = multinomial(),
-):
     raise NotImplementedError(
-        f"The CFG Logits processor is not available for {type(model)}."
+        f"The CFG Logits processor is not available for {type(model)}. "
+        + "Please subscribe to https://github.com/outlines-dev/outlines/issues/684"
+        + " for updates on the fix."
     )
 
 
-@cfg.register(LlamaCpp)
-def cfg_llamacpp(
-    model: LlamaCpp,
-    cfg_str: str,
-    sampler: Sampler = multinomial(),
-):
-    from outlines.integrations.llamacpp import CFGLogitsProcessor
-
-    logits_processor = CFGLogitsProcessor(cfg_str, model.model)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
 @cfg.register(OpenAI)
 def cfg_openai(model, cfg_str: str, sampler: Sampler = multinomial()):
     raise NotImplementedError(
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
index 6b6656fe9..cdf64a21f 100644
--- a/outlines/generate/regex.py
+++ b/outlines/generate/regex.py
@@ -5,6 +5,7 @@
 from outlines.models import OpenAI
 from outlines.models.llamacpp import LlamaCpp
 from outlines.models.mlxlm import MLXLM
+from outlines.models.transformers import Transformers
 from outlines.models.vllm import VLLM
 from outlines.samplers import Sampler, multinomial
 
@@ -39,8 +40,9 @@ def regex(model, regex_str: str, sampler: Sampler = multinomial()):
 
 
 @regex.register(MLXLM)
-def regex_mlxlm(
-    model: MLXLM,
+@regex.register(Transformers)
+def regex_unified(
+    model,
     regex_str: str,
     sampler: Sampler = multinomial(),
 ):
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
index 081ba0920..b8feb7659 100644
--- a/outlines/generate/text.py
+++ b/outlines/generate/text.py
@@ -2,7 +2,7 @@
 
 from outlines.fsm.guide import StopAtEOSGuide
 from outlines.generate.api import SequenceGenerator, SequenceGeneratorAdapter
-from outlines.models import MLXLM, VLLM, LlamaCpp, OpenAI
+from outlines.models import MLXLM, VLLM, LlamaCpp, OpenAI, Transformers
 from outlines.samplers import Sampler, multinomial
 
 
@@ -37,7 +37,8 @@ def text(model, sampler: Sampler = multinomial()) -> SequenceGenerator:
 
 
 @text.register(MLXLM)
-def text_mlxlm(model: MLXLM, sampler: Sampler = multinomial()):
+@text.register(Transformers)
+def text_unified(model, sampler: Sampler = multinomial()):
     return SequenceGeneratorAdapter(model, None, sampler)
 
 
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 4d74ebf60..fde913e2c 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -12,7 +12,7 @@
 from .mamba import Mamba, mamba
 from .mlxlm import MLXLM, mlxlm
 from .openai import OpenAI, azure_openai, openai
-from .transformers import Transformers, transformers
+from .transformers import Transformers, TransformerTokenizer, transformers
 from .vllm import VLLM, vllm
 
 LogitsGenerator = Union[Transformers, LlamaCpp, ExLlamaV2Model, Mamba, MLXLM, VLLM]
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index fae9b8e74..4c475617a 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,13 +1,17 @@
-from typing import TYPE_CHECKING, List, Optional, Tuple, Union
+import dataclasses
+from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, Union
 
 from datasets.fingerprint import Hasher
 
+from outlines.generate.api import GenerationParameters, SamplingParameters
 from outlines.models.tokenizer import Tokenizer
 
 if TYPE_CHECKING:
     import torch
     from transformers import PreTrainedModel, PreTrainedTokenizer
 
+    from outlines.processors import OutlinesLogitsProcessor
+
 __all__ = ["transformers"]
 
 
@@ -129,7 +133,6 @@ def __init__(
         model: "PreTrainedModel",
         tokenizer: "PreTrainedTokenizer",
     ):
-        self.device = model.device
         self.model = model
         self.tokenizer = TransformerTokenizer(tokenizer)
 
@@ -190,6 +193,183 @@ def __call__(
 
         return next_token_logits, kv_cache
 
+    def generate(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Union[str, List[str], List[List[str]]]:
+        """Generate text using `transformers`.
+
+        Arguments
+        ---------
+        prompts
+            A prompt or list of prompts.
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
+
+        Returns
+        -------
+        The generated text
+        """
+        if isinstance(prompts, str):
+            # convert to 2d
+            input_ids, attention_mask = self.tokenizer.encode([prompts])
+        else:
+            input_ids, attention_mask = self.tokenizer.encode(prompts)
+        inputs = {
+            "input_ids": input_ids.to(self.model.device),
+            "attention_mask": attention_mask.to(self.model.device),
+        }
+
+        generation_kwargs = self._get_generation_kwargs(
+            prompts,
+            generation_parameters,
+            logits_processor,
+            sampling_parameters,
+        )
+        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
+
+        # if single str input and single sample per input, convert to a 1D output
+        if isinstance(prompts, str):
+            generated_ids = generated_ids.squeeze(0)
+
+        return self._decode_generation(generated_ids)
+
+    def stream(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Iterator[Union[str, List[str]]]:
+        """
+        Temporary stream stand-in which implements stream() signature
+        and equivalent behaviour but isn't yielded until generation completes.
+
+        TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
+        """
+        if isinstance(prompts, str):
+            # convert to 2d
+            input_ids, attention_mask = self.tokenizer.encode([prompts])
+        else:
+            input_ids, attention_mask = self.tokenizer.encode(prompts)
+        inputs = {
+            "input_ids": input_ids.to(self.model.device),
+            "attention_mask": attention_mask.to(self.model.device),
+        }
+
+        generation_kwargs = self._get_generation_kwargs(
+            prompts,
+            generation_parameters,
+            logits_processor,
+            sampling_parameters,
+        )
+        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
+
+        # if single str input and single sample per input, convert to a 1D output
+        if isinstance(prompts, str):
+            generated_ids = generated_ids.squeeze(0)
+
+        for i in range(generated_ids.size(-1)):
+            output_group_ids = generated_ids.select(-1, i).unsqueeze(-1)
+            yield self._decode_generation(output_group_ids)
+
+    def _get_generation_kwargs(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> dict:
+        """
+        Conert outlines generation parameters into model.generate kwargs
+        """
+        from transformers import GenerationConfig, LogitsProcessorList, set_seed
+
+        max_new_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
+        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
+            sampling_parameters
+        )
+        if max_new_tokens is None:
+            max_new_tokens = int(2**30)
+
+        # global seed, not desirable
+        if seed is not None:
+            set_seed(seed)
+
+        if logits_processor is not None:
+            logits_processor_list = LogitsProcessorList([logits_processor])
+        else:
+            logits_processor_list = None
+
+        generation_config = GenerationConfig(
+            max_new_tokens=max_new_tokens,
+            stop_strings=stop_at,
+            num_return_sequences=(num_samples or 1),
+            top_p=top_p,
+            top_k=top_k,
+            temperature=temperature,
+            do_sample=(sampler == "multinomial"),
+            num_beams=(num_samples if sampler == "beam_search" else 1),
+            eos_token_id=self.tokenizer.eos_token_id,
+            pad_token_id=self.tokenizer.pad_token_id,
+        )
+
+        return dict(
+            logits_processor=logits_processor_list,
+            generation_config=generation_config,
+            tokenizer=self.tokenizer.tokenizer,
+        )
+
+    def _generate_output_seq(
+        self, prompts, inputs, generation_config, **generation_kwargs
+    ):
+        input_ids = inputs["input_ids"]
+        output_ids = self.model.generate(
+            generation_config=generation_config, **inputs, **generation_kwargs
+        )
+
+        # encoder-decoder returns output_ids only, decoder-only returns full seq ids
+        if self.model.config.is_encoder_decoder:
+            generated_ids = output_ids
+        else:
+            generated_ids = output_ids[:, input_ids.shape[1] :]
+
+        # if batch list inputs AND multiple samples per input, convert generated_id to 3D view
+        num_samples = generation_config.num_return_sequences or 1
+
+        if num_samples > 1 and isinstance(prompts, list):
+            batch_size = input_ids.size(0)
+            num_return_sequences = generation_config.num_return_sequences or 1
+            generated_ids = generated_ids.view(batch_size, num_return_sequences, -1)
+
+        return generated_ids
+
+    def _decode_generation(self, generated_ids: "torch.Tensor"):
+        if len(generated_ids.shape) == 1:
+            return self.tokenizer.decode([generated_ids])[0]
+        elif len(generated_ids.shape) == 2:
+            return self.tokenizer.decode(generated_ids)
+        elif len(generated_ids.shape) == 3:
+            return [
+                self.tokenizer.decode(generated_ids[i])
+                for i in range(len(generated_ids))
+            ]
+        else:
+            raise TypeError(
+                f"Generated outputs aren't 1D, 2D or 3D, but instead are {generated_ids.shape}"
+            )
+
 
 def transformers(
     model_name: str,
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
index 452d22c36..0d1908ebd 100644
--- a/tests/generate/test_integration_llamacpp.py
+++ b/tests/generate/test_integration_llamacpp.py
@@ -25,7 +25,7 @@ def model(tmp_path_factory):
     (
         (generate.text, []),
         (generate.regex, ("[0-9]",)),
-        (generate.cfg, (grammars.arithmetic,)),
+        # (generate.cfg, (grammars.arithmetic,)),  # Awaiting CFG fix
     ),
 )
 def test_llamacpp_generation_api(model, generator_type, params):
@@ -245,8 +245,11 @@ def test_llamacpp_json_schema(model):
 
 def test_llamacpp_cfg(model):
     prompt = "<|im_start|>user\nOutput a short and valid JSON object with two keys.<|im_end|>\n><|im_start|>assistant\n"
-    result = generate.cfg(model, grammars.arithmetic)(prompt, seed=11)
-    assert isinstance(result, str)
+
+    # remove this statement once cfg is implemented
+    with pytest.raises(NotImplementedError):
+        result = generate.cfg(model, grammars.arithmetic)(prompt, seed=11)
+        assert isinstance(result, str)
 
 
 @pytest.mark.parametrize(
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index da08bed71..f3fb9682e 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -14,44 +14,50 @@
 from outlines.samplers import beam_search, greedy, multinomial
 
 
-def test_transformers_integration_text():
-    rng = torch.Generator()
-    rng.manual_seed(10000)  # Choosen so <EOS> is generated
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    return models.transformers(
+        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
+    )
 
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
-    sequence = generate.text(model)("Write a short sentence ", rng=rng)
+
+def test_transformers_integration_text(model):
+    sequence = generate.text(model)(
+        "Write a short sentence ", seed=10000, max_tokens=10
+    )
     assert isinstance(sequence, str)
     assert model.tokenizer.eos_token not in sequence
 
     sequence = generate.text(model)(
-        "Write a short sentence ", max_tokens=10, stop_at=".", rng=rng
+        "Write a short sentence ",
+        max_tokens=20,
+        stop_at="a",
+        seed=10000,
     )
     assert isinstance(sequence, str)
 
     prompts = ["Write a short sentence ", "And another one "]
-    sequence = generate.text(model)(prompts, max_tokens=10, stop_at=[".", ","], rng=rng)
+    sequence = generate.text(model)(
+        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
+    )
     assert isinstance(sequence, list)
     assert len(sequence) == 2
     assert isinstance(sequence[0], str)
 
 
-def test_transformers_integration_text_multiple_samples():
-    rng = torch.Generator()
-    rng.manual_seed(10000)  # Choosen so <EOS> is generated
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_integration_text_multiple_samples(model):
     sampler = multinomial(2)
 
-    sequence = generate.text(model, sampler=sampler)("Write a short sentence ", rng=rng)
+    sequence = generate.text(model, sampler=sampler)(
+        "Write a short sentence ", seed=10000
+    )
     assert isinstance(sequence, list)
     assert len(sequence) == 2
     assert model.tokenizer.eos_token not in sequence
 
     prompts = ["Write a short sentence ", "And another one "]
     sequence = generate.text(model, sampler=sampler)(
-        prompts, max_tokens=10, stop_at=[".", ","], rng=rng
+        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
     )
     assert isinstance(sequence, list)
     assert len(sequence) == 2
@@ -60,14 +66,9 @@ def test_transformers_integration_text_multiple_samples():
     assert isinstance(sequence[0][0], str)
 
 
-def test_transformers_integration_streaming():
-    rng = torch.Generator()
-    rng.manual_seed(10000)  # Choosen so <EOS> is generated
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_integration_streaming(model):
     sequence = generate.text(model).stream(
-        "Write a short sentence ", max_tokens=10, stop_at=[".", ","], rng=rng
+        "Write a short sentence ", max_tokens=10, stop_at=[".", ","], seed=10000
     )
 
     token = next(sequence)
@@ -77,7 +78,7 @@ def test_transformers_integration_streaming():
     assert isinstance(remaining, str)
 
     sequence = generate.text(model).stream(
-        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], rng=rng
+        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
     )
     tokens = next(sequence)
     assert isinstance(tokens, list)
@@ -85,19 +86,11 @@ def test_transformers_integration_streaming():
     assert isinstance(tokens[1], str)
 
 
-def test_transformers_integration_streaming_batch_samples():
-    rng = torch.Generator()
-    rng.manual_seed(10000)  # Choosen so <EOS> is generated
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_integration_streaming_batch_samples(model):
     sampler = multinomial(samples=2)
 
     sequence = generate.text(model, sampler=sampler).stream(
-        ["Prompt1", "Prompt2"],
-        max_tokens=10,
-        stop_at=[".", ","],
-        rng=rng,
+        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
     )
     tokens = next(sequence)
     assert isinstance(tokens, list)
@@ -108,19 +101,11 @@ def test_transformers_integration_streaming_batch_samples():
     assert len(tokens[1]) == 2
 
 
-def test_transformers_integration_streaming_batch_beam_search():
-    rng = torch.Generator()
-    rng.manual_seed(10000)  # Choosen so <EOS> is generated
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_integration_streaming_batch_beam_search(model):
     sampler = beam_search(beams=2)
 
-    sequence = generate.text(model, sampler=sampler).stream(
-        ["Prompt1", "Prompt2"],
-        max_tokens=10,
-        stop_at=[".", ","],
-        rng=rng,
+    sequence = generate.regex(model, r"ab[cd]e", sampler=sampler).stream(
+        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=["c", "d"], seed=10000
     )
     tokens = next(sequence)
     assert isinstance(tokens, list)
@@ -131,61 +116,41 @@ def test_transformers_integration_streaming_batch_beam_search():
     assert len(tokens[1]) == 2
 
 
-def test_transformers_integration_text_stop():
-    rng = torch.Generator()
-    rng.manual_seed(10000)  # Choosen so <EOS> is generated
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_integration_text_stop(model):
     prompt = "Write a short sentence "
-    sequence = generate.text(model)(prompt, stop_at="a", rng=rng)
+    sequence = generate.text(model)(prompt, stop_at="a", seed=10000)
     assert sequence[len(prompt) :].find("a") == -1
 
 
-def test_transformers_various_regexes():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_various_regexes(model):
     prompt = "Write an email address"
     regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
     generator = generate.regex(model, regex_str)
 
     # One prompt
-    sequence = generator(prompt, rng=rng)
+    sequence = generator(prompt, seed=0)
     assert re.fullmatch(regex_str, sequence) is not None
 
 
-def test_transformers_various_regexes_prompt_list():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_various_regexes_prompt_list(model):
     prompt = "Write an email address"
     regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
     generator = generate.regex(model, regex_str)
 
     # Two prompts
-    sequence = generator([prompt, prompt], rng=rng)
+    sequence = generator([prompt, prompt], seed=0)
     assert re.fullmatch(regex_str, sequence[0]) is not None
     assert re.fullmatch(regex_str, sequence[1]) is not None
 
 
-def test_transformers_various_regexes_prompt_list_multiple_samples():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_various_regexes_prompt_list_multiple_samples(model):
     sampler = multinomial(samples=2)
     prompt = "Write an email address"
     regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
     generator = generate.regex(model, regex_str, sampler=sampler)
 
     # Two prompts
-    sequence = generator([prompt, prompt], rng=rng)
+    sequence = generator([prompt, prompt], seed=0, max_tokens=500)
     assert isinstance(sequence, list)
     assert len(sequence) == 2
     assert re.fullmatch(regex_str, sequence[0][0]) is not None
@@ -194,12 +159,7 @@ def test_transformers_various_regexes_prompt_list_multiple_samples():
     assert re.fullmatch(regex_str, sequence[1][1]) is not None
 
 
-def test_transformers_various_regexes_prompt_list_beam_search():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_various_regexes_prompt_list_beam_search(model):
     sampler = beam_search(5)
     prompt_1 = "Write an email address"
     prompt_2 = "Random"
@@ -207,7 +167,7 @@ def test_transformers_various_regexes_prompt_list_beam_search():
     generator = generate.regex(model, regex_str, sampler=sampler)
 
     # Two prompts
-    sequence = generator([prompt_1, prompt_2], rng=rng)
+    sequence = generator([prompt_1, prompt_2], seed=0)
     assert isinstance(sequence, list)
     assert len(sequence) == 2
     assert len(sequence[0]) == 5
@@ -217,105 +177,65 @@ def test_transformers_various_regexes_prompt_list_beam_search():
     assert re.fullmatch(regex_str, sequence[1][1]) is not None
 
 
-def test_transformers_integration_integer():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_integration_integer(model):
     prompt = "Write a short sentence"
-    sequence = generate.format(model, int)(prompt, max_tokens=10, rng=rng)
+    sequence = generate.format(model, int)(prompt, max_tokens=10, seed=0)
 
     assert isinstance(sequence, int)
 
 
-def test_transformers_integration_integer_array():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_integration_integer_array(model):
     prompts = ["Give me a number", "And another one"]
-    sequence = generate.format(model, int)(prompts, max_tokens=10, rng=rng)
+    sequence = generate.format(model, int)(prompts, max_tokens=10, seed=0)
     assert isinstance(sequence, list)
     assert len(sequence) == 2
     assert isinstance(sequence[0], int)
     assert isinstance(sequence[1], int)
 
 
-def test_transformers_integration_float():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_integration_float(model):
     prompt = "Write a short sentence"
-    sequence = generate.format(model, float)(prompt, max_tokens=10, rng=rng)
+    sequence = generate.format(model, float)(prompt, max_tokens=10, seed=0)
 
     assert sequence != ""
     assert isinstance(sequence, float)
 
 
-def test_transformers_integration_bool():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_integration_bool(model):
     prompt = "Is this True or False?"
-    sequence = generate.format(model, bool)(prompt, max_tokens=10, rng=rng)
+    sequence = generate.format(model, bool)(prompt, max_tokens=10, seed=0)
 
     assert sequence != ""
     assert isinstance(sequence, bool)
 
 
-def test_transformers_integration_date():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_integration_date(model):
     prompt = "What day is it today?"
-    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10, rng=rng)
+    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10, seed=0)
 
     assert sequence != ""
     assert isinstance(sequence, datetime.date)
 
 
-def test_transformers_integration_time():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_integration_time(model):
     prompt = "What time is it?"
-    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10, rng=rng)
+    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10, seed=0)
 
     assert sequence != ""
     assert isinstance(sequence, datetime.time)
 
 
-def test_transformers_integration_datetime():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_integration_datetime(model):
     prompt = "What time is it?"
-    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20, rng=rng)
+    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20, seed=0)
 
     assert sequence != 0
     assert isinstance(sequence, datetime.datetime)
 
 
-def test_transformers_integration_choice():
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_integration_choice(model):
     prompt = "Write a short sentence "
-    sequence = generate.choice(model, ["test", "choice"])(prompt, rng=rng)
+    sequence = generate.choice(model, ["test", "choice"])(prompt, seed=0)
 
     assert sequence == "test" or sequence == "choice"
 
@@ -327,9 +247,7 @@ def test_transformers_integration_with_pad_token():
     assert model.tokenizer.pad_token == "<pad>"
 
 
-def test_transformers_json_basic():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_basic(model):
     prompt = "Output some JSON "
 
     class Spam(BaseModel):
@@ -338,10 +256,7 @@ class Spam(BaseModel):
         spam: constr(max_length=10)
         fuzz: bool
 
-    rng = torch.Generator()
-    rng.manual_seed(0)  # make sure that `bar` is not an int
-
-    result = generate.json(model, Spam)(prompt, max_tokens=500, rng=rng)
+    result = generate.json(model, Spam)(prompt, max_tokens=500, seed=0)
     assert isinstance(result, BaseModel)
     assert isinstance(result.foo, int)
     assert isinstance(result.bar, float)
@@ -350,9 +265,7 @@ class Spam(BaseModel):
     assert len(result.spam) <= 10
 
 
-def test_transformers_json_schema():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_schema(model):
     prompt = "Output some JSON "
 
     schema = """{
@@ -366,18 +279,13 @@ def test_transformers_json_schema():
       }
     """
 
-    rng = torch.Generator()
-    rng.manual_seed(0)  # make sure that `bar` is not an int
-
-    result = generate.json(model, schema)(prompt, max_tokens=500, rng=rng)
+    result = generate.json(model, schema)(prompt, max_tokens=500, seed=0)
     assert isinstance(result, dict)
     assert isinstance(result["foo"], int)
     assert isinstance(result["bar"], str)
 
 
-def test_transformers_json_batch():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_batch(model):
     prompts = ["Output some JSON ", "Output more JSON"]
 
     class Spam(BaseModel):
@@ -386,17 +294,12 @@ class Spam(BaseModel):
         spam: constr(max_length=10)
         fuzz: bool
 
-    rng = torch.Generator()
-    rng.manual_seed(0)  # make sure that `bar` is not an int
-
-    result = generate.json(model, Spam)(prompts, max_tokens=500, rng=rng)
+    result = generate.json(model, Spam)(prompts, max_tokens=500, seed=0)
     assert isinstance(result[0], BaseModel)
     assert isinstance(result[1], BaseModel)
 
 
-def test_transformers_json_batch_multiple_samples():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_batch_multiple_samples(model):
     sampler = multinomial(samples=2)
     prompts = ["Output some JSON ", "Output more JSON"]
 
@@ -406,11 +309,8 @@ class Spam(BaseModel):
         spam: constr(max_length=10)
         fuzz: bool
 
-    rng = torch.Generator()
-    rng.manual_seed(0)  # make sure that `bar` is not an int
-
     result = generate.json(model, Spam, sampler=sampler)(
-        prompts, max_tokens=500, rng=rng
+        prompts, max_tokens=500, seed=0
     )
     assert isinstance(result, list)
     assert len(result) == 2
@@ -420,14 +320,9 @@ class Spam(BaseModel):
     assert isinstance(result[1][1], BaseModel)
 
 
-def test_transformers_json_str_enum():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_str_enum(model):
     prompt = "Output some JSON "
 
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
     class Name(str, Enum):
         john = "John"
         marc = "Marc"
@@ -437,20 +332,15 @@ class User(BaseModel):
         user_id: int
         name: Name
 
-    result = generate.json(model, User)(prompt, rng=rng)
+    result = generate.json(model, User)(prompt, seed=0)
     assert isinstance(result, BaseModel)
     assert isinstance(result.user_id, int)
     assert result.name in ["John", "Marc", "Michel"]
 
 
-def test_transformers_json_int_enum():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_int_enum(model):
     prompt = "Output some JSON "
 
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
     class Id(int, Enum):
         one = 1
         two = 2
@@ -458,25 +348,20 @@ class Id(int, Enum):
     class User(BaseModel):
         user_id: Id
 
-    result = generate.json(model, User)(prompt, rng=rng)
+    result = generate.json(model, User)(prompt, seed=0)
     assert isinstance(result, BaseModel)
     assert isinstance(result.user_id, int)
     assert result.user_id in [1, 2]
 
 
-def test_transformers_json_array():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_array(model):
     prompt = "Output some JSON "
 
     class User(BaseModel):
         user_id: int
         value: List[float]
 
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
-    result = generate.json(model, User)(prompt, rng=rng)
+    result = generate.json(model, User)(prompt, seed=0)
     assert isinstance(result, BaseModel)
     assert isinstance(result.user_id, int)
     assert isinstance(result.value, list)
@@ -485,19 +370,14 @@ class User(BaseModel):
 
 
 @pytest.mark.xfail(reason="The implementation of `anyOf` is incorrect")
-def test_transformers_json_union():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_union(model):
     prompt = "Output some JSON "
 
     class Spam(BaseModel):
         foo: int
         bar: Union[constr(max_length=10), float]
 
-    rng = torch.Generator()
-    rng.manual_seed(4)
-
-    result = generate.json(model, Spam)(prompt, max_tokens=100, rng=rng)
+    result = generate.json(model, Spam)(prompt, max_tokens=100, seed=4)
     assert isinstance(result, BaseModel)
     assert (
         isinstance(result.bar, int)
@@ -506,26 +386,18 @@ class Spam(BaseModel):
     )
 
 
-def test_transformers_json_function():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name)
+def test_transformers_json_function(model):
     prompt = "Output arguments for the function"
 
     def function(foo: int, bar: List[int]):
         return foo + sum(bar)
 
-    rng = torch.Generator()
-    rng.manual_seed(4)
-
-    sequence = generate.json(model, function)(prompt, max_tokens=100, rng=rng)
+    sequence = generate.json(model, function)(prompt, max_tokens=100, seed=4)
     assert isinstance(sequence, dict)
     assert isinstance(function(**sequence), int)
 
 
-def test_transformers_logits_vocab_size():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
-
+def test_transformers_logits_vocab_size(model):
     # Artificially increase the weights/logits size relative
     # to the vocabulary
     model.model.resize_token_embeddings(pad_to_multiple_of=3)
@@ -535,16 +407,11 @@ def test_transformers_logits_vocab_size():
 
     generator = generate.choice(model, ["True", "False"])
 
-    rng = torch.Generator()
-    rng.manual_seed(101)
-
-    sequence = generator("blah", rng=rng)
+    sequence = generator("blah", seed=101)
     assert sequence == "False"
 
 
-def test_transformers_json_custom_ws():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    model = models.transformers(model_name, device="cpu")
+def test_transformers_json_custom_ws(model):
     prompt = "Output some JSON with newlines"  # try to force model to use newlines
 
     schema = """{
@@ -558,12 +425,9 @@ def test_transformers_json_custom_ws():
       }
     """
 
-    rng = torch.Generator()
-    rng.manual_seed(0)
-
     generator = generate.json(model, schema, whitespace_pattern=r"[ ]?")
     generator.format_sequence = lambda x: x  # patch to return raw text
-    assert "\n" not in generator(prompt, max_tokens=500, rng=rng)
+    assert "\n" not in generator(prompt, max_tokens=500, seed=0)
 
 
 def test_transformers_reduced_vocabulary_caching():
@@ -582,11 +446,8 @@ def test_transformers_reduced_vocabulary_caching():
     assert vocab2 is vocab
 
 
-def test_custom_sampler():
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-
-    model = models.transformers(model_name)
-
+@pytest.mark.skip(reason="Custom Sampler Disabled in Transformers Integration")
+def test_custom_sampler(model):
     seen = False
     target_token_ids = model.tokenizer.encode(["c"])[0]
 
@@ -623,14 +484,11 @@ def __call__(
 def test_transformers_use_existing_model_and_tokenizer():
     from transformers import AutoModelForCausalLM, AutoTokenizer
 
-    rng = torch.Generator()
-    rng.manual_seed(10000)
-
     model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
     hf_tokenizer = AutoTokenizer.from_pretrained(model_name)
     hf_model = AutoModelForCausalLM.from_pretrained(model_name)
     model = Transformers(hf_model, hf_tokenizer)
-    sequence = generate.text(model)("Write a short sentence ", rng=rng)
+    sequence = generate.text(model)("Write a short sentence ", seed=10000)
     assert isinstance(sequence, str)
 
 
@@ -652,22 +510,30 @@ def test_RegexGuide_caching(temp_cache_dir):
     assert create_states_mapping.__memory__ is cache
 
     model = models.transformers(
-        "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM"
+        "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM", device="cpu"
     )
     generator = generate.regex(model, regex, sampler=greedy())
     assert cache.stats() == (0, 1)
 
-    model_2 = models.transformers("hf-internal-testing/tiny-random-GPTJForCausalLM")
+    model_2 = models.transformers(
+        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
+    )
     generator_2 = generate.regex(model_2, regex, sampler=greedy())
     assert cache.stats() == (0, 2)
 
     # These two different models and tokenizers should not have the same state
     # mapping results
-    assert generator.fsm.states_to_token_maps != generator_2.fsm.states_to_token_maps
+    assert (
+        generator.logits_processor.fsm.states_to_token_maps
+        != generator_2.logits_processor.fsm.states_to_token_maps
+    )
 
     generator_3 = generate.regex(model_2, regex, sampler=greedy())
     assert cache.stats() == (1, 2)
-    assert generator_2.fsm.states_to_token_maps == generator_3.fsm.states_to_token_maps
+    assert (
+        generator_2.logits_processor.fsm.states_to_token_maps
+        == generator_3.logits_processor.fsm.states_to_token_maps
+    )
 
     # Just for fun...
     structured = generator(prompt, max_tokens=30)
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index f4596a2df..1404d287a 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -63,20 +63,17 @@ def test_llama_tokenizer():
 
 
 def test_model():
-    with pytest.raises(ValueError, match="When passing device_map as a string"):
-        transformers(TEST_MODEL, device="non_existent")
-
     model = transformers(TEST_MODEL, device="cpu")
     assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.device.type == "cpu"
+    assert model.model.device.type == "cpu"
 
     model = transformers(TEST_MODEL, model_kwargs={"device_map": "cpu"})
     assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.device.type == "cpu"
+    assert model.model.device.type == "cpu"
 
     model = transformers(TEST_MODEL, device="cpu", model_kwargs={"device_map": "cuda"})
     assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.device.type == "cpu"
+    assert model.model.device.type == "cpu"
 
     input_ids = torch.tensor([[0, 1, 2]])
     logits, kv_cache = model(input_ids, torch.ones_like(input_ids))

From 517ff5321db5f6d96bba293ef3d0130148be7034 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Fri, 21 Jun 2024 09:59:03 -0500
Subject: [PATCH 012/505] Use LogitsProcessors for models.transformers ->
 outlines.generate.*

---
 outlines/models/transformers.py              | 4 ++++
 outlines/processors/base_logits_processor.py | 2 +-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 4c475617a..70f19a0eb 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -182,6 +182,10 @@ def forward(
 
         return output.logits, output.past_key_values
 
+    @property
+    def device(self):
+        return self.model.device
+
     def __call__(
         self,
         input_ids: "torch.LongTensor",
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index f829844a6..1d141a2f0 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -42,7 +42,7 @@ def process_logits(
         input_ids and logits are always 2D tensors for handling a batch of sequences.
 
         - input_ids -> List[List[tokens]]
-        - logits.shape[0] -> 2D_Tensor[logits]
+        - logits -> 2D_Tensor[logit floats]
 
         Important to keep in mind when designing universal logits processors
         - logits processors are only used once and never re-applied for a new sequence generator

From 5a7f08205f9d9bff997db01401c731afe4b25501 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 15 Jul 2024 03:48:09 -0500
Subject: [PATCH 013/505] enable generate.fsm with llamacpp by using
 outlines.processors

---
 outlines/fsm/guide.py           | 19 ++++++++++++-------
 outlines/generate/fsm.py        | 19 ++++++++++++++++++-
 outlines/generate/regex.py      | 13 +------------
 outlines/generate/text.py       |  6 +-----
 outlines/models/llamacpp.py     |  4 ++++
 outlines/models/transformers.py |  4 ----
 tests/generate/test_generate.py | 11 +++++++++++
 7 files changed, 47 insertions(+), 29 deletions(-)

diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index aa694232e..2e4415148 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -184,13 +184,7 @@ def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
         ) = create_states_mapping(regex_string, tokenizer)
         self.eos_token_id = tokenizer.eos_token_id
         self.final_states = fsm_finals | {-1}
-
-        # cache returned masks token masks
-        # this increases performance of the mask substantially
-        self.states_to_token_mask = {
-            state: torch.tensor(list(next_tokens_to_end_states.keys()))
-            for state, next_tokens_to_end_states in self.states_to_token_maps.items()
-        }
+        self._cache_state_to_token_tensor()
 
     def get_next_instruction(self, state: int) -> Instruction:
         """Return the next instruction for guided generation.
@@ -285,8 +279,19 @@ def create_states_mapping_from_interegular_fsm(
             from_interegular_instance.empty_token_ids,
         ) = create_states_mapping_from_interegular_fsm(interegular_fsm)
         from_interegular_instance.eos_token_id = tokenizer.eos_token_id
+        from_interegular_instance._cache_state_to_token_tensor()
         return from_interegular_instance
 
+    def _cache_state_to_token_tensor(self):
+        """
+        cache state -> token int tensor
+        this increases performance of mask construction substantially
+        """
+        self.states_to_token_mask = {
+            state: torch.tensor(list(next_tokens_to_end_states.keys()))
+            for state, next_tokens_to_end_states in self.states_to_token_maps.items()
+        }
+
     def is_final_state(self, state: int) -> bool:
         """Determine whether the current state of the guide is a final state."""
         return state in self.final_states
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
index 80db350f0..832a154bd 100644
--- a/outlines/generate/fsm.py
+++ b/outlines/generate/fsm.py
@@ -1,10 +1,14 @@
+from functools import singledispatch
+
 import interegular
 
 from outlines.fsm.guide import RegexGuide
-from outlines.generate.api import SequenceGenerator
+from outlines.generate.api import SequenceGenerator, SequenceGeneratorAdapter
+from outlines.models import MLXLM, LlamaCpp, Transformers
 from outlines.samplers import Sampler, multinomial
 
 
+@singledispatch
 def fsm(
     model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
 ) -> SequenceGenerator:
@@ -12,3 +16,16 @@ def fsm(
     device = model.device
     generator = SequenceGenerator(fsm, model, sampler, device)
     return generator
+
+
+@fsm.register(MLXLM)
+@fsm.register(Transformers)
+@fsm.register(LlamaCpp)
+def fsm_unified(
+    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
+) -> SequenceGeneratorAdapter:
+    from outlines.processors import FSMLogitsProcessor
+
+    fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
+    logits_processor = FSMLogitsProcessor(tokenizer=model.tokenizer, fsm=fsm)
+    return SequenceGeneratorAdapter(model, logits_processor, sampler)
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
index cdf64a21f..52b8c7dad 100644
--- a/outlines/generate/regex.py
+++ b/outlines/generate/regex.py
@@ -41,6 +41,7 @@ def regex(model, regex_str: str, sampler: Sampler = multinomial()):
 
 @regex.register(MLXLM)
 @regex.register(Transformers)
+@regex.register(LlamaCpp)
 def regex_unified(
     model,
     regex_str: str,
@@ -52,18 +53,6 @@ def regex_unified(
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
-@regex.register(LlamaCpp)
-def regex_llamacpp(
-    model: LlamaCpp,
-    regex_str: str,
-    sampler: Sampler = multinomial(),
-):
-    from outlines.integrations.llamacpp import RegexLogitsProcessor
-
-    logits_processor = RegexLogitsProcessor(regex_str, llm=model.model)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
 @regex.register(VLLM)
 def regex_vllm(
     model: VLLM,
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
index b8feb7659..6da187e0b 100644
--- a/outlines/generate/text.py
+++ b/outlines/generate/text.py
@@ -38,6 +38,7 @@ def text(model, sampler: Sampler = multinomial()) -> SequenceGenerator:
 
 @text.register(MLXLM)
 @text.register(Transformers)
+@text.register(LlamaCpp)
 def text_unified(model, sampler: Sampler = multinomial()):
     return SequenceGeneratorAdapter(model, None, sampler)
 
@@ -47,11 +48,6 @@ def text_vllm(model: VLLM, sampler: Sampler = multinomial()):
     return SequenceGeneratorAdapter(model, None, sampler)
 
 
-@text.register(LlamaCpp)
-def text_llamacpp(model: LlamaCpp, sampler: Sampler = multinomial()):
-    return SequenceGeneratorAdapter(model, None, sampler)
-
-
 @text.register(OpenAI)
 def text_openai(model: OpenAI, sampler: Sampler = multinomial()) -> OpenAI:
     if not isinstance(sampler, multinomial):
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 7e223a9fa..904b193c4 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -141,6 +141,10 @@ class LlamaCpp:
     def __init__(self, model: "Llama"):
         self.model = model
 
+    @property
+    def tokenizer(self):
+        return LlamaCppTokenizer(self.model)
+
     def prepare_generation_parameters(
         self,
         generation_parameters: GenerationParameters,
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 70f19a0eb..4c475617a 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -182,10 +182,6 @@ def forward(
 
         return output.logits, output.past_key_values
 
-    @property
-    def device(self):
-        return self.model.device
-
     def __call__(
         self,
         input_ids: "torch.LongTensor",
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index c6d023929..9432c0e4e 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -101,6 +101,17 @@ def test_generate_text_stream(request, model_fixture):
             assert isinstance(token, str)
 
 
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_fsm(request, model_fixture, pattern):
+    import interegular
+
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.fsm(model, interegular.parse_pattern(pattern).to_fsm())
+    res = generator("test")
+    assert re.fullmatch(pattern, res) is not None, res
+
+
 @pytest.mark.parametrize("pattern", REGEX_PATTERNS)
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
 def test_generate_regex(request, model_fixture, pattern):

From 08a3e540551f2ae2f36a89d137553e155a00ef09 Mon Sep 17 00:00:00 2001
From: Alonso Silva Allende <alonso.silva@gmail.com>
Date: Tue, 16 Jul 2024 12:04:45 +0200
Subject: [PATCH 014/505] Add QA with Citations example to the Cookbook (#1042)

As requested by @rlouf this PR adds a question answering with citations
example to the Cookbook using llama-cpp-python.
---
 docs/cookbook/qa-with-citations.md | 228 +++++++++++++++++++++++++++++
 mkdocs.yml                         |   1 +
 2 files changed, 229 insertions(+)
 create mode 100644 docs/cookbook/qa-with-citations.md

diff --git a/docs/cookbook/qa-with-citations.md b/docs/cookbook/qa-with-citations.md
new file mode 100644
index 000000000..b8e034a29
--- /dev/null
+++ b/docs/cookbook/qa-with-citations.md
@@ -0,0 +1,228 @@
+# Generate Synthetic Data and Q&A with Citations
+
+This tutorial is adapted from the [instructor-ollama notebook](https://github.com/alonsosilvaallende/Hermes-Function-Calling/blob/main/examples/instructor_ollama.ipynb). We start with a simple example to generate synthetic data and then we approach the problem of question answering by providing citations. We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](library).
+
+We pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+```python
+!wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+```
+
+We initialize the model:
+
+```python
+from llama_cpp import Llama
+from outlines import generate, models
+
+llm = Llama(
+    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+    n_gpu_layers=-1,
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = models.LlamaCpp(llm)
+```
+
+## Generate Synthetic Data
+
+We first need to define our Pydantic class for a user:
+
+```python
+from pydantic import BaseModel, Field
+
+class UserDetail(BaseModel):
+    id: int = Field(..., description="Unique identifier") # so the model keeps track of the number of users
+    first_name: str
+    last_name: str
+    age: int
+```
+
+We then define a Pydantic class for a list of users:
+
+```python
+from typing import List
+
+class Users(BaseModel):
+    users: List[UserDetail]
+```
+
+We can use a `generate.json` by passing this Pydantic class we just defined, and call the generator:
+
+```python
+generator = generate.json(model, Users)
+response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
+print(response.users)
+# [UserDetail(id=1, first_name='John', last_name='Doe', age=25),
+# UserDetail(id=2, first_name='Jane', last_name='Doe', age=30),
+# UserDetail(id=3, first_name='Bob', last_name='Smith', age=40),
+# UserDetail(id=4, first_name='Alice', last_name='Smith', age=35),
+# UserDetail(id=5, first_name='John', last_name='Smith', age=20)]
+```
+
+```python
+for user in response.users:
+    print(user.first_name)
+    print(user.last_name)
+    print(user.age)
+    print(#####)
+# John
+# Doe
+# 25
+# #####
+# Jane
+# Doe
+# 30
+# #####
+# Bob
+# Smith
+# 40
+# #####
+# Alice
+# Smith
+# 35
+# #####
+# John
+# Smith
+# 20
+# #####
+```
+
+## QA with Citations
+
+We first need to define our Pydantic class for QA with citations:
+
+```python
+from typing import List
+from pydantic import BaseModel
+
+class QuestionAnswer(BaseModel):
+    question: str
+    answer: str
+    citations: List[str]
+
+schema = QuestionAnswer.model_json_schema()
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
+
+```python
+def generate_hermes_prompt(question, context, schema=schema):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON with correct and exact citations "
+        "extracted from the `Context`. "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + "`Context`: "
+        + context
+        + "\n`Question`: "
+        + question + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
+```
+
+We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
+
+```python
+question = "What did the author do during college?"
+context = """
+My name is Jason Liu, and I grew up in Toronto Canada but I was born in China.
+I went to an arts high school but in university I studied Computational Mathematics and physics.
+As part of coop I worked at many companies including Stitchfix, Facebook.
+I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.
+"""
+generator = generate.json(model, QuestionAnswer)
+prompt = generate_hermes_prompt(question, context)
+response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+print(response)
+# QuestionAnswer(question='What did Jason Liu do during college?', answer='During college, Jason Liu studied Computational Mathematics and Physics. He also worked at companies such as Stitchfix and Facebook, and started the Data Science club at the University of Waterloo, serving as its president for two years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'As part of coop I worked at many companies including Stitchfix, Facebook.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
+```
+
+We can do the same for a list of question-context pairs:
+
+```python
+question1 = "Where was John born?"
+context1 = """
+John Doe is a software engineer who was born in New York, USA.
+He studied Computer Science at the Massachusetts Institute of Technology.
+During his studies, he interned at Google and Microsoft.
+He also founded the Artificial Intelligence club at his university and served as its president for three years.
+"""
+
+question2 = "What did Emily study in university?"
+context2 = """
+Emily Smith is a data scientist from London, England.
+She attended the University of Cambridge where she studied Statistics and Machine Learning.
+She interned at IBM and Amazon during her summer breaks.
+Emily was also the head of the Women in Tech society at her university.
+"""
+
+question3 = "Which companies did Robert intern at?"
+context3 = """
+Robert Johnson, originally from Sydney, Australia, is a renowned cybersecurity expert.
+He studied Information Systems at the University of Melbourne.
+Robert interned at several cybersecurity firms including NortonLifeLock and McAfee.
+He was also the leader of the Cybersecurity club at his university.
+"""
+
+question4 = "What club did Alice start at her university?"
+context4 = """
+Alice Williams, a native of Dublin, Ireland, is a successful web developer.
+She studied Software Engineering at Trinity College Dublin.
+Alice interned at several tech companies including Shopify and Squarespace.
+She started the Web Development club at her university and was its president for two years.
+"""
+
+question5 = "What did Michael study in high school?"
+context5 = """
+Michael Brown is a game developer from Tokyo, Japan.
+He attended a specialized high school where he studied Game Design.
+He later attended the University of Tokyo where he studied Computer Science.
+Michael interned at Sony and Nintendo during his university years.
+He also started the Game Developers club at his university.
+"""
+
+for question, context in [
+    (question1, context1),
+    (question2, context2),
+    (question3, context3),
+    (question4, context4),
+    (question5, context5),
+]:
+    final_prompt = my_final_prompt(question, context)
+    generator = generate.json(model, QuestionAnswer)
+    response = generator(final_prompt, max_tokens=1024, temperature=0, seed=42)
+    display(question)
+    display(response.answer)
+    display(response.citations)
+    print("\n\n")
+
+# 'Where was John born?'
+# 'John Doe was born in New York, USA.'
+# ['John Doe is a software engineer who was born in New York, USA.']
+#
+#
+# 'What did Emily study in university?'
+# 'Emily studied Statistics and Machine Learning in university.'
+# ['She attended the University of Cambridge where she studied Statistics and Machine Learning.']
+#
+#
+# 'Which companies did Robert intern at?'
+# 'Robert interned at NortonLifeLock and McAfee.'
+# ['Robert Johnson, originally from Sydney, Australia, is a renowned cybersecurity expert. He interned at several cybersecurity firms including NortonLifeLock and McAfee.']
+#
+#
+# 'What club did Alice start at her university?'
+# 'Alice started the Web Development club at her university.'
+# ['Alice Williams, a native of Dublin, Ireland, is a successful web developer. She started the Web Development club at her university and was its president for two years.']
+#
+#
+# 'What did Michael study in high school?'
+# 'Michael studied Game Design in high school.'
+# ['Michael Brown is a game developer from Tokyo, Japan. He attended a specialized high school where he studied Game Design.']
+```
diff --git a/mkdocs.yml b/mkdocs.yml
index 12e79fd75..bc06c4993 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -106,6 +106,7 @@ nav:
       - Summarize a document: cookbook/chain_of_density.md
       - Playing chess: cookbook/models_playing_chess.md
       - Perspective-taking prompting: cookbook/simtom.md
+      - Question-answering with citations: cookbook/qa-with-citations.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From e684af2d5dacf9b0d06c2d41054f40488c30da6d Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 15 Jul 2024 19:08:41 -0500
Subject: [PATCH 015/505] Fix mamba integration by making it a variant of
 outlines.models.transformers

---
 docs/reference/models/mamba.md        |  7 ---
 docs/reference/models/transformers.md | 67 +++++++++++++++++++++++++++
 outlines/models/__init__.py           |  5 +-
 outlines/models/mamba.py              | 61 ------------------------
 outlines/models/transformers.py       | 60 ++++++++++++++++++++----
 tests/generate/test_generate.py       | 38 ++++++++++++++-
 6 files changed, 157 insertions(+), 81 deletions(-)
 delete mode 100644 docs/reference/models/mamba.md
 delete mode 100644 outlines/models/mamba.py

diff --git a/docs/reference/models/mamba.md b/docs/reference/models/mamba.md
deleted file mode 100644
index ac6db3682..000000000
--- a/docs/reference/models/mamba.md
+++ /dev/null
@@ -1,7 +0,0 @@
-# Mamba
-
-```bash
-pip install mamba_ssm transformers torch
-```
-
-*Coming soon*
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 7c1febd02..80b70845f 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -82,3 +82,70 @@ print(output)
 ```
 
 [transformers]: https://github.com/huggingface/transformers
+
+
+# Alternative Model Classes
+
+`outlines.models.transformers` defaults to `transformers.AutoModelForCausalLM`, which is the appropriate class for most standard large language models, including Llama 3, Mistral, Phi-3, etc.
+
+However other variants with unique behavior can be used as well by passing the appropriate class.
+
+### Mamba
+
+[Mamba](https://github.com/state-spaces/mamba) is a transformers alternative which employs memory efficient, linear-time decoding.
+
+To use Mamba with outlines you must first install the necessary requirements:
+```
+pip install causal-conv1d>=1.2.0 mamba-ssm torch transformers
+```
+
+Then you can either create an Mamba-2 Outlines model via
+```python
+import outlines
+
+model = outlines.models.mamba("state-spaces/mamba-2.8b-hf")
+```
+
+or explicitly with
+```python
+import outlines
+from transformers import MambaForCausalLM
+
+model = outlines.models.transformers(
+    "state-spaces/mamba-2.8b-hf",
+    model_class=MambaForCausalLM
+)
+```
+
+Further Reading:
+- https://huggingface.co/docs/transformers/en/model_doc/mamba
+
+### Encoder-Decoder Models
+
+You can use encoder-decoder (seq2seq) models like T5 and BART with Outlines.
+
+Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
+
+T5 Example:
+```python
+import outlines
+from transformers import AutoModelForSeq2SeqLM
+
+model_pile_t5 = models.transformers(
+    model_name="EleutherAI/pile-t5-large",
+    model_class=AutoModelForSeq2SeqLM,
+)
+```
+
+Bart Example:
+```python
+model_bart = models.transformers(
+    model_name="facebook/bart-large",
+    model_class=AutoModelForSeq2SeqLM,
+)
+```
+
+
+### Multi-Modal Models
+
+/Coming soon/
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index fde913e2c..c161215d1 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -9,10 +9,9 @@
 
 from .exllamav2 import ExLlamaV2Model, exl2
 from .llamacpp import LlamaCpp, llamacpp
-from .mamba import Mamba, mamba
 from .mlxlm import MLXLM, mlxlm
 from .openai import OpenAI, azure_openai, openai
-from .transformers import Transformers, TransformerTokenizer, transformers
+from .transformers import Transformers, TransformerTokenizer, mamba, transformers
 from .vllm import VLLM, vllm
 
-LogitsGenerator = Union[Transformers, LlamaCpp, ExLlamaV2Model, Mamba, MLXLM, VLLM]
+LogitsGenerator = Union[Transformers, LlamaCpp, ExLlamaV2Model, MLXLM, VLLM]
diff --git a/outlines/models/mamba.py b/outlines/models/mamba.py
deleted file mode 100644
index d3dabf669..000000000
--- a/outlines/models/mamba.py
+++ /dev/null
@@ -1,61 +0,0 @@
-from typing import TYPE_CHECKING, Optional
-
-from .transformers import TransformerTokenizer
-
-if TYPE_CHECKING:
-    import torch
-    from mamba_ssm.models.mixer_seq_simple import MambaLMHeadModel
-    from transformers import PreTrainedTokenizer
-
-
-TOKENIZER_MODEL = "EleutherAI/gpt-neox-20b"
-
-
-class Mamba:
-    """Represent a `mamba` model."""
-
-    def __init__(
-        self, model: "MambaLMHeadModel", tokenizer: "PreTrainedTokenizer", device
-    ):
-        self.device = device
-        self.model = model
-        self.tokenizer = TransformerTokenizer(tokenizer)
-
-    def forward(self, input_ids: "torch.LongTensor", *_):
-        """Compute a forward pass through the mamba model."""
-
-        output = self.model(input_ids)
-        next_token_logits = output.logits[..., -1, :]
-        return next_token_logits, None
-
-    def __call__(self, input_ids: "torch.LongTensor", *_) -> "torch.FloatTensor":
-        return self.forward(input_ids)
-
-
-def mamba(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-):
-    try:
-        import torch
-        from mamba_ssm import MambaLMHeadModel
-        from transformers import AutoTokenizer
-    except ImportError:
-        raise ImportError(
-            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba people."
-        )
-
-    if not torch.cuda.is_available():
-        raise NotImplementedError("Mamba models can only run on GPU.")
-    else:
-        if device is None:
-            device = "cuda"
-
-    model = MambaLMHeadModel.from_pretrained(model_name, device=device)
-
-    tokenizer_kwargs.setdefault("padding_side", "left")
-    tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_MODEL, **tokenizer_kwargs)
-
-    return Mamba(model, tokenizer, device)
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 4c475617a..1d3be511a 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,4 +1,5 @@
 import dataclasses
+import inspect
 from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, Union
 
 from datasets.fingerprint import Hasher
@@ -226,10 +227,16 @@ def generate(
             input_ids, attention_mask = self.tokenizer.encode([prompts])
         else:
             input_ids, attention_mask = self.tokenizer.encode(prompts)
+
         inputs = {
             "input_ids": input_ids.to(self.model.device),
             "attention_mask": attention_mask.to(self.model.device),
         }
+        if (
+            "attention_mask"
+            not in inspect.signature(self.model.forward).parameters.keys()
+        ):
+            del inputs["attention_mask"]
 
         generation_kwargs = self._get_generation_kwargs(
             prompts,
@@ -267,6 +274,11 @@ def stream(
             "input_ids": input_ids.to(self.model.device),
             "attention_mask": attention_mask.to(self.model.device),
         }
+        if (
+            "attention_mask"
+            not in inspect.signature(self.model.forward).parameters.keys()
+        ):
+            del inputs["attention_mask"]
 
         generation_kwargs = self._get_generation_kwargs(
             prompts,
@@ -336,7 +348,7 @@ def _generate_output_seq(
     ):
         input_ids = inputs["input_ids"]
         output_ids = self.model.generate(
-            generation_config=generation_config, **inputs, **generation_kwargs
+            **inputs, generation_config=generation_config, **generation_kwargs
         )
 
         # encoder-decoder returns output_ids only, decoder-only returns full seq ids
@@ -376,6 +388,8 @@ def transformers(
     device: Optional[str] = None,
     model_kwargs: dict = {},
     tokenizer_kwargs: dict = {},
+    model_class=None,
+    tokenizer_class=None,
 ):
     """Instantiate a model from the `transformers` library and its tokenizer.
 
@@ -398,19 +412,47 @@ def transformers(
     A `TransformersModel` model instance.
 
     """
-    try:
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-    except ImportError:
-        raise ImportError(
-            "The `transformers` library needs to be installed in order to use `transformers` models."
-        )
+    if model_class is None or tokenizer_class is None:
+        try:
+            from transformers import AutoModelForCausalLM, AutoTokenizer
+        except ImportError:
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if model_class is None:
+        model_class = AutoModelForCausalLM
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
 
     if device is not None:
         model_kwargs["device_map"] = device
 
-    model = AutoModelForCausalLM.from_pretrained(model_name, **model_kwargs)
+    model = model_class.from_pretrained(model_name, **model_kwargs)
 
     tokenizer_kwargs.setdefault("padding_side", "left")
-    tokenizer = AutoTokenizer.from_pretrained(model_name, **tokenizer_kwargs)
+    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
 
     return Transformers(model, tokenizer)
+
+
+def mamba(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    tokenizer_kwargs: dict = {},
+):
+    try:
+        from transformers import MambaForCausalLM
+
+    except ImportError:
+        raise ImportError(
+            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
+        )
+
+    return transformers(
+        model_name=model_name,
+        device=device,
+        model_kwargs=model_kwargs,
+        tokenizer_kwargs=tokenizer_kwargs,
+        model_class=MambaForCausalLM,
+    )
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 9432c0e4e..06d311f5b 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -36,17 +36,53 @@ def model_transformers_opt125m(tmp_path_factory):
     return models.transformers("facebook/opt-125m", device="cpu")
 
 
+@pytest.fixture(scope="session")
+def model_mamba(tmp_path_factory):
+    return models.mamba(model_name="state-spaces/mamba-130m-hf", device="cpu")
+
+
+@pytest.fixture(scope="session")
+def model_bart(tmp_path_factory):
+    from transformers import AutoModelForSeq2SeqLM
+
+    return models.transformers(
+        "facebook/bart-base", device="cpu", model_class=AutoModelForSeq2SeqLM
+    )
+
+
+# TODO: exllamav2 failing in main, address in https://github.com/outlines-dev/outlines/issues/808
+# TODO: t5 tokenizer doesn't work with streaming
+"""
+@pytest.fixture(scope="session")
+def model_exllamav2(tmp_path_factory):
+    return models.exllamav2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        device="cpu"
+    )
+
+@pytest.fixture(scope="session")
+def model_t5(tmp_path_factory):
+    from transformers import AutoModelForSeq2SeqLM
+
+    return models.transformers(
+        "EleutherAI/pile-t5-base", device="cpu", model_class=AutoModelForSeq2SeqLM
+    )
+"""
+
+
 ALL_MODEL_FIXTURES = (
     "model_llamacpp",
     "model_mlxlm",
     "model_mlxlm_phi3",
     "model_transformers_random",
     "model_transformers_opt125m",
+    "model_mamba",
+    "model_bart",
 )
 
 
 NOT_IMPLEMENTED = {
-    "stream": ["model_vllm"],
+    "stream": [],
     "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
     "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
     "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],

From cd8c79aa6ee1ee4f731ea08865f43d4a50029310 Mon Sep 17 00:00:00 2001
From: Daniel van Strien <davanstrien@users.noreply.github.com>
Date: Tue, 16 Jul 2024 13:18:35 +0100
Subject: [PATCH 016/505] fix PyPI name for autoawq

---
 docs/reference/models/vllm.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index b5221e582..735ab4cc5 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -80,7 +80,7 @@ model = models.vllm("https://huggingface.co/squeeze-ai-lab/sq-llama-30b-w4-s5",
 
 !!! Warning "Dependencies"
 
-    To use AWQ model you need to install the auto-awq library `pip install auto-awq`.
+    To use AWQ model you need to install the autoawq library `pip install autoawq`.
 
     To use GPTQ models you need to install the autoGTPQ and optimum libraries `pip install auto-gptq optimum`.
 

From d64bfc29a1518cd3a84b9ded6d58adfddcf59e3f Mon Sep 17 00:00:00 2001
From: JerryKwan <Jinzhong.Guan@gmail.com>
Date: Wed, 17 Jul 2024 21:36:33 +0800
Subject: [PATCH 017/505] add fallback tokenizer (#1046)

add fallback tonenizer if tiktoken can not get encoding from model name
support llm services which provide openai compatibility api such as
ollama
---
 outlines/models/openai.py | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 8625f4d3d..692e64ac5 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,5 +1,6 @@
 """Integration with OpenAI's API."""
 import functools
+import warnings
 from dataclasses import asdict, dataclass, field, replace
 from itertools import zip_longest
 from typing import Callable, Dict, List, Optional, Set, Tuple, Union
@@ -439,7 +440,14 @@ def openai_model(
         config = OpenAIConfig(model=model_name)
 
     client = AsyncOpenAI(**openai_client_params)
-    tokenizer = tiktoken.encoding_for_model(model_name)
+
+    try:
+        tokenizer = tiktoken.encoding_for_model(model_name)
+    except KeyError:
+        warnings.warn(
+            f"Could not find a tokenizer for model {model_name}. Using default cl100k_base."
+        )
+        tokenizer = tiktoken.get_encoding("cl100k_base")
 
     return OpenAI(client, config, tokenizer)
 
@@ -464,6 +472,13 @@ def azure_openai(
         config = OpenAIConfig(model=deployment_name)
 
     client = AsyncAzureOpenAI(**azure_openai_client_params)
-    tokenizer = tiktoken.encoding_for_model(model_name or deployment_name)
+
+    try:
+        tokenizer = tiktoken.encoding_for_model(model_name or deployment_name)
+    except KeyError:
+        warnings.warn(
+            f"Could not find a tokenizer for model {model_name or deployment_name}. Using default cl100k_base."
+        )
+        tokenizer = tiktoken.get_encoding("cl100k_base")
 
     return OpenAI(client, config, tokenizer)

From 9ce0df3cfea68c2b06edf4847a08b4153cd7a2ee Mon Sep 17 00:00:00 2001
From: Michael Louis <michaellouis157@gmail.com>
Date: Wed, 17 Jul 2024 10:38:34 -0400
Subject: [PATCH 018/505] Update cerebrium instructions

---
 docs/cookbook/deploy-using-cerebrium.md | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/docs/cookbook/deploy-using-cerebrium.md b/docs/cookbook/deploy-using-cerebrium.md
index 79d703d32..b7609424c 100644
--- a/docs/cookbook/deploy-using-cerebrium.md
+++ b/docs/cookbook/deploy-using-cerebrium.md
@@ -24,6 +24,9 @@ cerebrium init outlines-project
 You set up your environment and hardware in the cerebrium.toml file that was created using the init function above.
 
 ```toml
+[cerebrium.deployment]
+docker_base_image_url = "nvidia/cuda:12.1.1-runtime-ubuntu22.04"
+
 [cerebrium.hardware]
 cpu = 2
 memory = 14.0

From 1bf23be48f3df1f02b2dd879ecd3cdceba8e4f86 Mon Sep 17 00:00:00 2001
From: Alonso Silva <alonso.silva@gmail.com>
Date: Thu, 18 Jul 2024 10:29:36 +0200
Subject: [PATCH 019/505] Add Knowledge Graph Extraction example

---
 .../images/knowledge-graph-extraction.png     | Bin 0 -> 33127 bytes
 docs/cookbook/knowledge_graph_extraction.md   | 137 ++++++++++++++++++
 mkdocs.yml                                    |   1 +
 3 files changed, 138 insertions(+)
 create mode 100644 docs/cookbook/images/knowledge-graph-extraction.png
 create mode 100644 docs/cookbook/knowledge_graph_extraction.md

diff --git a/docs/cookbook/images/knowledge-graph-extraction.png b/docs/cookbook/images/knowledge-graph-extraction.png
new file mode 100644
index 0000000000000000000000000000000000000000..a24b1c61a9c371e4fdad95dd5ec9fc31cd71c9cd
GIT binary patch
literal 33127
zcmbTebzD{3w?4dRC8Z@*x)CI06Dr-^p$G^fC6ZDW9Ri!~1}W*3kOm3q?ndd(cdm2p
z{oU_-|9RsNKj-M)d#)K{j5)^hJYxl@D$C$uQ({99geNB}sSZJ?s^E|FEllt)U&-rj
z!GBO3)n%SSMSav8;4kQ}o+v$mppr10%NH2n@3-w`^&BCHj1m5WBB#!<3qd}Ya*|Ip
zU5)?F;1~{TU5j5=*^)b5F|FdsE0Jgn??;gCj*#5p&W($W{(7bPuGjv;)_Akgs?I#O
z&bM-_J04@48$;n!DJu7^+4B!kX_YMTx1yO`4RZOL54YCDcUs~W9<|^;eE8^>hgk92
z5|2}#UwDs4pIiU?zB7+LtL(dXz_g%Y1IIIO_<wSl8NiAl$oVNMJm2U4zyA7Dn+uvJ
zBIhzp?TO(YielHvt*=k?x;}eV>wd(hrKve^v^BQT9zykQYm{?udbKxU=x^1WvdgR+
zbH$s92>mKsC6^f|ayK_OwL0g&xt<rt!+9FadxnrtwhuZsM4g_Q`B^y`AvfoFLPJVQ
zs^;LpbG6$-@NRor__6nOJ*)6r66MnEnl;+Z>6%AQe<!0nJU#Q?pKQOoxjNA(HvZAs
z(IF#73^nb2l|b>C38fK{_QR)1r<VyG_?;ocDlVSFWmL=1xJmo?-MYlp`BBF8)y0qx
z3Yz8aeDiRXoynON<kS8HY){2vc*fPm$<PNhEEZ;FKYUsQOZNPqwjtZZ;aw8LsY;tR
zB(+L!#p@Ay#gRQEO&A-`bsQqAL*xBekrA<mCt^iMTbn5x64qPVka)jG@?1g!!oa}z
zJ&^WnqojlO(Ul?uy}$nf)rYd->e!(B*RRZ(I+vkj(f5*c8ZiPkld3zjKiUrCQc_Z4
z)$4W|E^-xOxJ<wO4Hen`+W#)t#QRE15f6Mjy5Z*fJ&%p+`Jok0i0fkO?emSCl;uY@
zORbzjLb1}pcb~j?@#1j)6G3+|umk+tB+xBAJ-x%TwG<6^_s2<(o#bgNwrAh(kZPEI
z!o(vQdYR}7&P~Z{L^M&@QJUc+qBk}shB=$s`*Kg_Ol!;i_>bJnj?gTJrFJz-%UlFD
zq)JT00?|y>xo~?P{+2;TM>BDuR<+H|nJBv8QSeGCdhfMOCi~0exl-;2aW~u<iU6Ci
z&(6OmwN6M#c(@#?GPt|YvZ4n0u*s%E?|gHr)z#H44_12xWa3>P+YaZd@x5J^<sn+K
z%}@yanW2f;ue2WHnyhim>9rE4jf;Qm#3m&z9mpyVLBO;h?=AS<8+^a>Lq^2;&sUjW
zdlmA;Op14gbP^F8gv@G)kQ$pQ0SB8X{tTJ0?2VzEAq^E##xTf-_Q40#f$u@&+=J&w
zTdWTrJSpfh3)t~b;5847&BnfcJ6z_2`Kwn!8@-$cmHr|(9D9=*Wfr~Z!j7w|9icQ}
z_5(QJ>6EQod~F>aL$yv@%Y=@IK@w6@sYqs(Xqi+2gDTss<z+J-Ui04R8b?+pCd@5G
zl=WL+>+lvtdasYG8UqC1D<Gr<9aj~3t};ig;}b)Q;})$3GBX+SH9QiL*9!#RcU)Ld
zeJbIpLJdkcy@GR@1Y)NWw0Z3=eS3UEkNOexVLtS4Cd=g%!}e)Bqm#YPt@O(+!*y3t
zpz_<dNqn%_IXE&&*|q%%2^A1Zf_Afr5?KZz0&am?=fC>-L+=rbCmrI~Y8)Kc(BwyS
z>?X9U{T}Yab}G98lbdrLw@gYClIz7F?m$k3CMVaRgsNG01;cl?8!Is;**YekQC=5Q
zbKUgN{04A5*Vw8#uLR@wfB9?e7d|4?%@<n(4hI#uA2_@=O_VJ(tcei2K7HwyDW9X9
z)av#gxbre0v}*=T0D*2|U1rurG+AYrf>8f*^7j2ol+Ht~H4E3*c4zx5rrb{|UbOhC
zy?IkmEr16h$iWs0x*V)3@nDp@AH8<V#5U16SQEdYAor^iZWzBj8Z#ju%g^gg5jRn6
zo^_tG*}nz35`n|A{WGK@p7jNXe02Qn7h>jQeANEtwIE_v{#iOEF0N2o?@I?abO?a~
zkKdZ1T%UZ0+ZXKg=w7*MikRmz;Vwf**gf*0!f2O2dvuBQUgXjl5CRMSc1yxk!|qR<
z#okf>D#Hb*WQ$+155jzyr@p`Vq{}S-LkgF<&3d$eFhLN_=faxnh5)kliD8-#Y98q#
zF1GD8D7)`Ah%Xl7p?o$?xhK5l*lz<t)AyuPp=k*@xd2_x%d&p4NEv?+M>(Y#z6x1U
zg-P%EFOB5utPrN}Ep-HG^?7=DfZeng1iKkO-j^iOR&>4Amjr$z9Y)KkUA6r?!w-R~
z?+ijl$?HoHAeRZ#!M8M?K1B_o<d>&Kfm}YaB7B%y*>xOCGnJE1@$3FhRnfZ*5{iv^
z^qoypIV8+iHrt>>(*$5v@tp_#;woFCg^~Q}V5~t9tAY$BokC-ZVnyEz7}dL~(LRV5
zapAf;NV!4sL-tr;R)&A(Zmz4UU3Eo`TJi%9e4A!!og!r<`uh@v=>_Pti%t77)VjL5
z-kr?2^w2{C$cJCB=AU}3k=m`Kf$1#n{rU5TLFM^Rn-5dyRe1EakEWfg@lYg&o}psQ
z(}A0SlwRrV>}>zmZ<FhN@)F^PKK#=H)ggBsrWOQRZEi#HWZAor;e4cuSejfU)9*rq
zs%1i=A)fH?@KVg@2J-TO6Y*-j34%Gp!<z1k0nG4L@&`+x2vOKEi8w4uYV5b>2Ohj$
zl1)x+s4lcmSg35N=5$C{ti*tn(14c_Gcqv^mRk*(g>?4!tJD+ve*WA^4;{7G_)g(h
zJ)pNWm~Mxfj^HIR)H5{ic%6vo(NjKr4tZYgbyVw8Q-0YQSo)m<?yF9E$({OxP~CF~
zGQ@%h-pz%Ixoi*DGr|ICPDPwH8NwjGX<)&YKR@3-1YX!p&!Ej278)82wpjp}IDY$B
zR&LJUj~_p_(z@t2Vo?W-5if}zynXu?oOlS8;2<0_tf6^Da3s|JJC#ctr1Z3I-JP8h
z`{>Hea+<EL)f2KLD#T0M2xtx7-23$cDj@1_*ori)29S%lmeU9ktE}y@t+ZRGy9S^&
zjC-8!E~Mi_eIG#dKyh`lG@job%u-12bCEZ<uvq-)V1K*eJA#8F7#jyi>c=xbd{e5}
z2G5JY$B+;$5sQ!Xq7})kX-oCbYKNp8a4iQt_F<<Mb=M5`F|K(lHGFZjC3A87x8;z>
zC-ONB{B_dwk*Tsb6Qvkk#MRVQ!nW_Fr<q3bW6*qFf*@)6c`e0z_RY)Zv&k4iW=}O1
zDm!cj{HK_z)F68V*d4sXQ3Jc*AXb|dzJB@qO+Ttq$W*F3nyGEZv^Gjpi9H^P1lphJ
z8ja~QGrm`S)FqB$K6S1jP+5Ycp;VinlK=HEiyoR!J4mIT=R`yhxi5#ytb^bVRyzk@
z7Q8+p3#_6IRAElr({T;wo9z30dxTDqCAH!L%6NcGLU5M`xI0)`S>=!@(|A=%wi&@w
z992e;JuaA>v+GibU~46Zbp4Bmqwc6KZ8nipxr9|lwk^<rYwS<7TxR1bt5$!)Qt3kw
z?Bd`+tA}Q9GOY{}11CND$sg~lbRS_ul;Fab5S^}f(<mr&ygYSKTf^zBvYX@ThLWHE
zisi{>6g)rNpeLKH|64gFNdx`y1Agy*+H&vuxqLF5s9BE{k(xs)%{I|fTuYU|Y$*R7
zkw)z5P$MubEQUXw-+ELvj8@bPuj!F}!b+uet4;ir0K|t2T<mqTh6`fGVRM*OZLN?8
zq%;iRvFmZKl6eZXXUkrx@KAoqLqHcV(6Lj=xEn7^S=|<uKI#<eGlH)LPCcwLZ?{R9
z5`xAFfbHX%c<f;m8voEA^`G@PP}&^Mi!4BKWwLFVjt;2~aWue!E}ml(H2E&H_zeMz
zl5~SzqKi4jpVx;BvDZpm4(@$5T7frr|IXCuAW>4i9)eTD*s*@4m^{9}(qn#Gf@&Dd
z{PFQy^zQ^9h~qR~?gvev#??P(GLgkkWt`5gu>o;!t&#vb`UH#{)y~Ob;U|XL@xejM
z>!nn^>B^drN45v!s~D-*z;b-#U%W`Gu9&i1Y7begw1Wy4&4sg-5}2&DTvjO}*`V)u
z^r{dWyLQ1>#`O95=L{;TZqw~Gg;aSL5H#*f$O5r)bBEEouRqYE2bm$?6ymN#zN_QH
z_wNt&=+9RZ>?{XTpOj)kqtDT?CDe;U9)Yj~#-N%Wt?310phCKSV2n?|!G5<8h;vv=
z*3{IH+_`h#8k%(3oqvLji;Lt>FK#&RE>MAdgh6b(d->PSJCZS-RU`Yp8|uu0K)=o0
z`i}Qa9S4dJ|J@I$Ga#M13(|U((TC^g=jOuLQ7I`ZAajr6hg4f`k*phgc+_)f*-U&~
zzI_-tC3b>B?=%NNj^LL@b*_8q6PV4_2l$75`=JQq=Vc&&1ap^$KQw6c`oTd6@<jA5
z;&s{+lx-a8O)*qx<g3JoB1Q&=PNLr+3pKkfF*oe_WJU%`rOe)ys1m!}edZ3*T&Wt6
z5+8yE`tw3};6k7;S(Lr|p2seV=+Ur7IxeII4!QOSB)45ezkkI&GQSPQ)~e)QU_+YC
zV9tYGk<586JF}(Inudmo@D0Qd2|2Ffo}Hg}Q~nMhX3o_qHU<-g=P55I7oPPk3+F2W
zC3Z<H&3I5^<`pRSNY;lzd1D^H!p0^a4M`}?2o8H<LLjt1E_c~6#>U4dTXFt7`Fo`&
zMv4M@{d%rJ)CdGJ<jJ3jXDxmN<mS*1c=P&X?s+mOXkE;r_}FZUrEiNs?tbXl@0d9`
zm7@KVJ@)1FP$BDKPste<=n(kxy1ceFSriiK&s%&&LPSK@4LzqhA>3vUzQ8|u07;-?
zJU#%~eUJJG_`a&3lZhJmS5XkXYa^1AxfCRUMMg40xgoX1xfgeQk{7rT10i;eSJCGu
zJKWtiljUjq`)}1SeQH5{_FF~#ilJL3j5Z7Fo`)(D<q(@*HI*tZF0NFP4-S5)KIf_@
z6iz0n1MCdogF2U;XvXw#dYNt_XxOi|Klg&j$M^*{!MlagLWALNTI{u6SGBn>cSN9z
z+>`dYgFcas$qMV?dSS6opS;qJK5_k-Yn($O&Mq$K-B7<1pHOUz1YZ!Hu>L!H`}rmm
zJv#S4B^_qBp+9NcBAa3usWP&u66D}MGYhcn{Q9629XeJ$p>aw<9iTTv!>Apov@sYe
z#mG@f$=;r+3*v=_P1Mu~kJf)bL%QjGDlO$z!yKRwmiVwrBjWrEiI@R(gKRQH=?`yY
zyJz49nU68JM|wrxt$zhZN{i5qhEg94d&<sKL5G;PNi1i<TEUZHUqpxS(zc&%HejTJ
ztV`m+b?SBo3L0i75ybbomLLy|Gz&_--k$#c7{;I1Mo>uR|9p8!US573z9oNA-Fk`A
z%cQO+Uy~4{HjE}V{3pcFZ6MmTJNBNuSZ}=A3G3dw({Bh|+Yc9aP8i!ggD&p=Q@JRK
zxb0_h5FTyM7(}OPM@^Ksz9P#qiawHN5f6W|*B_R(&FRU)^7eMta!hyexc2n~8|8(z
zz}};vs=kbM7Xlqmlxg2u?mRGBGMJ^(#dh6}S-?Nmk6yt#m;AIxo|tv7%+xLO0UPU5
zl&XF#@)^b}>|CC?+tXqH@w=rjH^IR;&{La0)}t9owQk)=X}o(ER!<gs%2-~Qa}+8m
zq+;w_db);~NNfcC45%tRN~(`0dPPuhj#`Z;E-tReyYmuq2ZaY;VRg0pA2b}x(ZaVX
zbet1h>@tfiPAXXid{2YK_$*5~#)7!N6f5S?Nu{N8^6-7H09(>B7Be)(wkd|0$_x%*
z$m?_gOOLJ=)XiqM3)pjlLwZ^(WVGb1CTjV#Xi@|JW7Z-r+ld^6P8)+#(M{ieYRXUF
z?~yiGVxh2R{D|RXqJF{yi2^hr-ue2{4Gg%J{2mM_eCdENACS=BH7!T{#MhXPe)G0T
zlGf7ye}?S^We+NdDGJD`DmxXIX$L=+PoyzJEZ12rk}pkJEZi74kE1*A<aIJtebJ)%
zpuh7Ct)77n_m&zk|FNJlE|Vg!v7LdmPaGfk+fo?S>D)AaP0KA$(S2s>)tNgR9cRRo
z2paa3pBaM=QGqRv0vrY$no*r|DLOw*b~!(VBpR9ep)0q8*?cf#W^$+XbOh#cJ|aee
z)lO3FUW!HepHzfy&Re$GWLcb7zUr#|7Q+P;;vTswogDhr4?YF<tPE|Y)j#}1WD+lx
z_F-n_#7bI>;PYTX3=wO_K&V~ji%qetZEMIKq<Vb$_4Vf=MK)8FrRdRYk0M#OZ>gQ)
z5SmfCyuPO#?ycFoB>Y<L(`4sgV+)Fi=xC#lESoB<nyUQ<P|pFem3DvSu&4(GX;oO}
zmmM=TZD%${4s0TYM+Nc1(X|(TgAu<TN2KIqAce|ADrjoIma={|jb>KW_FXuNqj`!<
zQ3|l@OD_?<Vm(xu@O`HQZi@felYa3V4h2boHpOs!s4<oIvL=mk=g!8^$R9gi`10^s
zHb;SSuB+T9!N{bl^wSYseO|`sK^*zq<x)@atkW_#lItc8`8SQRg9?ExPUd`4=zglQ
z_UKZ@idh9G^xu+o<CS`(eTPav2@o~+Oq4Se5cIP7XBK$V<d`j;P!P-E;NI42{-FCz
zzD=I>B1Dz)@TqT6B#Tpu`UyYu1H|cgr=!<LT;+uZFZreOLo?Mbozp)YFssSpxc#z}
zKyM2UH?I^;9!#vre6nR9c=e96jmt<O@223Ub|nf6#D}Flq{0I7J5Tgatxp!x4b73V
z*{YZ|<z}|^Q5&76@*V=n=%j$m;ldIUbtlhqisA=BtY=B~dXFxHyd{Ss`VSMH=3O9b
zi&54=i91jb7>KP$_j=@86vvYa`R-YwSF%iWmJ6EJQ(33#!|83BNs-kSa-YbqKWW}p
z;be2P8kuJFMf!JMgyFD!b9W(hOuDlA`ncq=y55o<%S`=OivY>D52W5jV-@b=1*u=z
zW|1^<X;ocZ{^ujN#)^B0ekZxivMM}(%N4SygQScIyPF{QENEV?+}@|)ZuL+E=E@YY
zRRA?%5H>zt4@&uKG}iqj43fUv(N*G7*N+VO`MdBn_PhDiunEeeWxu^U$a+%B@}@t#
z<+0H>qnVsxPdo{3kV{z1R69J2_P1@ka@Ir5Xh~RN2|=}5Kz`Ma4D%=dmQy(WO74?-
z_qbuj29Fv#-hR~;`NO$A2c(z$gREsC#(o;h>l3n7HDATnXgvPeAFs9uB^ACUvyrNn
zC=>tZd72waM)KQ*X4_`-CaCM&V~Q{Nnyy8!IC`pVEP<n`@w~`*u<2X(60%Awnkuh;
z{-Jr#{$emM8++MWxDwCRviK;)e;G$p9kvgCl$}bNIpS^U3!;jO$AjtampUwk>S3kT
zrgEQZ4Y{1gD`cW{6oAW{8*uOMm+t&t6gUe*k<fvcyFgsZ>Bs6|PWRhKR==J%baC_G
z^kwE&G~agvVJ$wo6n3tH+UtOo4JWZ8l!rfY-2Jo(;=fq)735Dx$~$1WE&S3AG%EKY
z8(u3xj5;Y$j&mm9lk<59e;_N&*Jvqk=_qk{E4o1WY(NBD`+C(1o$+tG$E$<oS>w|?
z^2z%Og1D993umX>_`y$yCxbl9!rQ#xbac;uVJ)|Rd2|{VR<-HmN=(H(M!x+P;|EAh
z;zvr&6+m=#{u|G+VfccL(RSV`l8$w{+cdmlkCIp}NQ*pDa3ccK_?qvbV)TTsq&z2b
zMPEz(Sj4((?ZakfSKZqE5q+Hik}9sM*?@%w?_U|nN2TeH{v$-;urnI)G7!iquzdQ-
z>TF@7Kt4Y3Pb@8t97ZZoo(`t+Y|b%$isn$tJE$$_{*-Six5|L6Hs?0Z4GuBj#dZoE
zi@T%fb<s~r{nBYakGihfzUNuq|8Bu}e$USpTzNe;c|Q7hrV0IuI(#(d_Pc|{uPLn6
zS*ik7_b@{RZ;a#xrPYSnD|7a{2@2EYh-1F?E2ypFSiLnpRUhGU(UNCn<n?ba>dYQY
z*#C+#TQBr_og+Of{H-_Qs@XU))dk#ZWv670^u}c}b8M_GMjU&Ck%TXvXkK-XcX7Aw
z?wH5X=sk5XNU$X7=8iYjNyy=@VSm~sZbsJTov473wW4WktlU;VH@Ab^anD$mVrJ2@
zOY^9KOAE8lSgHi2ln$xh&YrfsBp$tb#Tq_%Y^G7U5MKG>b}!D3vdz6%Db^WNR_A!p
zv2d%PT9v#DN*`Mmh!WHR#rDK&evOTdh;J&SH6?)$M(z~1eKaE7w|s_V>59y62-{eo
z$zvez;WmiMX~k0&^;U~k-?Fn8Ql`i~zvfAa?&8+f$jt29Exews5WP%~8lLp(Uo>MJ
zoa$~YJI+0@8x`BuXdn9o9{5iV#m33Wk8yhZcd94j@j(jFimP-1QIRq(swwjGG+ovo
z@qOVV^HXCtR-SChUw;<J(^}1`b-FnoWl)iJ@14+-+Lf(gu?0N}L<Umr3%|Y*E@DtQ
z&=wury`@`_nNrBP!lqGbj8}3>u}ZKr-zGoGcy>5x*sV^6WpL@y<F{~>&rdjX8=lYS
z|C*1I^=y3&Yo|%0ml5qqKv|dk=gTP@Q95<`<w)z8vf=Z_s$>7&!gE1JRzmGY+hg5$
zjUzKgnShF8r!tw(5gyuNGl79+dh$QhEg3UY6n*NiqP|k#AV2?!n3So1ovSMUHZp3Q
z;(>f>%HFFBwhxgnkrjyy`U;otd@myQ>h15L4P;nc-h0Lj4P*jZW1z-yO|quCI=0qn
zXoFJ&!hg5=i(>z%u=YCNajoy&@gF^q@e&(K;>Wi)&)j-K+r%^MDL2yr9ioE_LLcxJ
z<^k&$JB=}nkR|Sa<USUO7S9o1u%HkfO-%-cRZm<A6%b{zopxr8qW!UPwkpQ3py7qX
zjiFMr+5Z}B^6F%cTS1apB^mj@1{=-S;nt&j?6CNm14ZIx?nOjx7|Qx>a0T(*yLWFB
zBs?rPheIlh$JUUg(Dt+Gn7b#O+aN9c4AP8Zu*;>Ahk#826~_N9vEG_%Y~&!M5q>NA
zzf*L;?3e|xaC58t?+@miKY`T(Hs(Kb_i-G8cWEGx0j|Q;UGIxGT7>fb9(U?Qh75t)
z@&11L|I`dz=;-L3L~6>)p`b(-9rl!&$$<`Uf$Hq3sw$`!Jonnj|Fawcff5Il<57(1
zfKuuqg2tcCxD0z@ON0TiVEre6Aak|ywclNxJE^UOe5rHVV(JF9o3X@DwZksBAmzZ(
zWUTh?QBWV^6yAIH0nO9i{(AwTxw(1QzonMF$<uvwcRyOFmdCX9mL46bt75?RdgfjP
z*K(t*lY+8Y&=m)Jsk;kQcs!ie5ahIEjuZTTo<`IiN$auu1lR>TsDy0on|3D3a##+i
zngJpV7p)zhZLtW4dti7#&b_nvGhPBt0In~=gQaJ5hCOdd^uyJ9FqNQkw11_|6q5p^
zO7&k=XA8e-s>sQ-Bf&rA>YU!c<*KTNv$Mb#U*EveurPuoac^T#S>l7rQUnk}VDR5E
z_rR)jGq&kB8_-f+VDw?4&ZXE&Dfj<HEsfuZa(AbFNzfN;X&^5YyVUSPF}b=QfEu~J
zVE$GBI4YGyCMI$yK&jc=RGVUG5J1zC9=q)`z~$5EcR^Yo#-mx6Kdk}8!?2~IL10_B
z_NbcIe`lD7tp7Nf1=aKOO|fQA06O{zgL5?)%2A171gs?ksME8zi8mWiAuX^awlE4Y
zTw$%FCiecA1MEA%`3^%bP)}~RdwvEWL@ZpB3s{eG=+WuJFbK+eLM?ah#Vts)MQS<K
znw;A>Q#t8z7p0no1yjgfo~NVG;j5GF84@xwCTmHUeFbQHHzjlz;Aihb=I*ZcC9zve
z0>Cbc5%N?APzn{OWeZ?@VYcLDM=tC-(#V7a1r!vNQq0tQSOm10O-&L2U+VAA0f-qG
zDHk3oZ0>$~db(N2dSTUU1Q<!J%5$3-jPwSK<kOt?%x@hqY|;F|Pd{AihKhM!9gXq1
z?mh?GqyP@;;k;mAPKp-*>9mUsSrwr3RNI4TasZkPgY$*djg4tV4+ox=`iksJh;!`J
zU;Hm5Yf&&b?$Ij%zWj^GO%2;5T|g1cKDA(C5?SBBe+BF#tF<JRpl#->WH;$isN7G7
zCPly|zJL%AnSqy=mu7g-8-VTjU^hC{U+(t+bo6`+B$++}c~e4Fo(wAgf_FdjiyL0+
zCnd^c!ni3geM;M@>2``LH`@zHRUFvE_eBt61Psy=$*2Ad<%9vGv0<qrETQnA+;--%
zJHUX!b2edy?EeGa_yz0vGeBB#gu7-;044~GDPY^vz8sJ@^%xq4AxA1-T#Ba$YyALJ
zl?oyYNZLR_P*lCJ?U9C&K{Ovv04u2mB?#IDTBn?j+O}VO+OwGhKxxd1tg<qhxpC>I
z;xD!T!nm6RV7Dvg1*5C0h4kq6+CSwXb~UG+L%N8UDnpiUr@jEVm?{C_STjyfUDa%7
zYWYFr{Oo?X{$LYIo(FN9?|Bmh$oy})9j&bcAeMkY@CQIrskE+a#ojm9rnR3!Xhq{I
z{|87ef7yYrwraHjXjRk0+o^=`TnR3v0Y^#+CLsM8L!ep1-hiJ;C3uP#94O4MtE!}>
zrIo+iP}j?!L|p=09KaG!T5U9c7~#OHh6aS-gBV&06IV|0JdR^=`7<8gaJjFrIa-*I
zi&h=^=vCi9Mwpm9@Dq|Fz<beeVA<b+k#jay)-)Mfx6dGeF>f|EHzSjhIJ=>31_p-D
z$A2d1yOTs*vH|uP=;7f3Aa6=64Ohr{Q84xMXLK-C%rBy?O*8fheDG8`BAB5LFx3Gl
zLpmfKFl;@PzX3nIQS5z{Ux32LX&W@n6jELIHUR_bXeObRphw(X9+;B9r#H@g$jhrf
z3K?=UGD;Q1+VQz;8=eE4o)wTSEP!mGl>5<u-_zYqG?oliQ<?<jG_>GvMWF}*Gw_i<
zKWV&CYVpO(xV-dGvl}Z)l1+DAP=a(8zVHaU)xKQ`%str}3j+y|jLV#7RW2Yq7ARpd
zL@*by!9GCMaDVd>^B558fd-u@h99OVtAt<fx7vhHF@WW)f#u9Y0J+TqOt$L5s+Q3=
z;8#q_iLxM~+;M_lEeN#RgiKK{33S50fZ!=XSQKI&LRs0_-IM`%e$0xo_p?u-#1B(s
zRXn!V($knVoTxsI04TFhWlnbeC}2CyV$Gn<x;By$F8Lt?CdS%qe)&()Ry~32#b8$I
z_}8QUSkCge$jE043k$J*XnqTotu}#E(S$IC1I<%{9ZJ0a_;LKK+Zv5tK4rk7LAS#T
zVLq~|2et~E;mo$y)2mRE6JVKzy8cX*^-8CoK8^y=xI>+=#4~$9@?`>cYU*wkV8J*G
z&@krB5(I40t%kC@^hX}6dBl*AO`*TvrHj+ZR^;K-YC4T)R-8iN!vizp+69ckK&D*e
zIsh#9Z&iAJI4_${{u0ABU-#F(Sx90C9I_A2E)`9&vaR8p$UQ1Lf&0F-_Ec)jD$~Bz
zoDbYkE%iYZv`+&F_itd)Hj~F+dgC7X&3awf=%M8vPdoO*_G4c7`aPNP3+f;<R6bGo
zvF%$YyZ#^D(G~#}0KmX>;E`SQsjob8W&mn!)t@Z7FFVrSg!miuR}Y#-%Le3S@Eqjk
zcL}mRK>5wG-19tCI|oVA3gG~|E!s5Ma2IhEsqG^WwWk3g?ZdR+?|mJqU2Mz=?hW8A
zQ*Y~!C$0Byt>dM<hrb?N9b#{Q0-gLr`c^uwDaZ3$lL0SLrp=2l(yMdkgb`C(6iyM(
zb3VX}_P^1u0@<W#NRC>DbQj?Jl|eWeXg$T-2Vs@yBA#N8Y6I5O02AK7Wl5UyQfzE<
zIbaI4>NbeK+c<#Cq1pvZ*jNLHQLQHJ131o9e~l%VZxwT&Y*|D0bHcv>W>w8;p8!E8
zUsw@<Qn$XW2bB6dKw-IMzCPL)ob!nMFIs&`EeC|lxI(ts9F=?eWfpiKvw&3yePCAe
z7N>Z~6GTuzK)8%-{qKEQ4DRGy<$|WN-I^uG&NcNU#9qy+W9ahjCPx5#+CQtBaZafA
z_W;_SVZcwhAtjT@)moF>ll}TBQbff3q&;kF`qtw6g&|jbaBjAL_B!s7XOO`$TjpE+
z_4(FfZ;T)t=1KdSe?Mu;BYv?Q58E?pG@rF*epQGTC?^_H85N=jz-_x6&Q>3I#$H_i
zlwLact4tHkpV=P?(cC7smG_yb087=@-ah#C!82GBAql(zfxDB?G4v$bWmPFrDBRe6
zgAx1It*<hvtkekCui&bcyXSpbf>6)W$WdOb-@A5(=FeaaMyiR|*_T<@CA<=i#w<YX
z41gK<OLn&jIk;~XR$2B$bMDJl#F%)+8G9aSodX7l9Y|qvV4}Af5VTfpfuyGNrV90p
zw~B2;@<i7b^Byl@+(e(+)`?&5g)OhGF@vvbgN+;Tsjo+rC=Uac;C`6qlC{n1uQ+-#
z8}yEapq8;>6Ye6VvO{2=K~Bk-c$9#&_8EnL^cAM3=B|$g<&~AAWxl}(Y_Pv#idYh$
zWSPP~iUKjfvz;F&$#?fe-X9kr^f}qs0<YJ+ul;DfPG10NMmaVxB}@R_2QMxySplqd
z-(NU+c1iF8@H##8stIrgm^nPjN0T7pQlyI;Ip&56c>>ri^#$w%LyGnJHw5CB_rtW8
zI89H}513OtBAz~dx~%qmVqyYLLIZ++j^z7N!Nih9Z#-i*EGc3ML-jF+Cm#TQs;VfS
z=W?a9#41J`M=5Aa{f~HJBnEiSAYh)kq~i;eewP5vS&-F0PO>|!s}KcTaq1&b6p@f4
z55BBb+0I1B^j266Ni(PczMtfx%<3qfwIMqINOM2}FAD@b8_Z{<j<6WPkJjBEgjU%f
z?ne<aC~(e=vi=Kj_^n6{H@BKxm3Y1Yz=PywW=akh?wy?EbkM$6oc>KtLLv!dWicju
z98ywJi*_p_YsWitD*sBb&Q4GL8~OS9wT6Gx*S8|qLC(Xq38mTrgL~)8?Z3K}99vvm
zEX#H`tabNL>%Y`u)MYDxP|IrmgcUS(h=qGs0$e(&Qz7bm>w`e?qaYJXO_*Wh;kOtj
ze$CSOnr25&(D{!pPe>QNU2tUE;-abAs&%^10y@l7r9MiewvT9y<}_eB{$17@AY);X
zt;-N`^0p4}M3;d!h`(cXE_OU^BcS>Ucbx{e-QwgoAQkf4u3o}FQ|vee+MbQevx7zb
ztewUiV+P^+?<^{6$wCf4SKlR0!noow(Q=z~%g>*Zbd~`pg-cqk6phaRBok;p4dB8W
z2nPgWcvV$Z^>LnFnFXvc44)qMP3ajj!osJ%Gh|Y49^i1^PYcv9Irub>f$(dx(nbLm
zFIL#j#OqhQW+b9Wrd+ZPxqG9S{Cc_dFG!W%9c^lzgZe11pg?h&u*16pkbiBCf18Kw
z>)Y^Ddhuu;Ybq*GdS`PcX2QAvA<cb91qFq+ITYv|6etoh+J*X3IOJR;=E>=cgxS9H
zPT8*V!8skFw4(Ar(?XZ$z?`^z2=0HFuFIxy>n=&#;@h)s>!JJ(@{^z+m?#o+uxZ>f
z!R;F|2;O0FTt4RVAbhNn;t9kwi?wXY>Pr<luE-!6hc`2OW2y+s){Iw($F0`9dq;){
z8&{yA46>4JxHYXd>wV*SE`GCi<Ik(2G3vZ046UPqTp&rv<Mg|gT_tebDpTwN5PH+M
zvEkf7JThnA75i(>vg_-j^z^5&CNhbqYjN2>=g{?av3XCl>9??)OIPgB9LVkQcQ{hj
z6shXzv3jin?w@Y+Qhz^}3|fcfTuJkylcNQoGI?b2>&`{ZldC=JcAW51@yi7QSh5ko
z3$6d7Is9&TVKvv_xrq8YxADjOlR%N<yCx=W<dI`eD32jaa(8hbd^0n%*a+qZ+gn|c
zSH5^u7PT+KxntM(%OHtHc$b=Pt{QK2qWvxUlQ|Wh6Br+V5F`*#Z_I{L;VNK2h}La!
zLnOKVrH3E&TnXm!k_b-7)cCH{c6|ko$v8fA{t2+`AxkUR!Ck~EGY{Ne3wix@6_;o)
zwtD-vlm*S_8jch~N1Bliob;ALN7k10>;%vuoN!XM2U+Yd#s!agEIP%y;;~o?8RG<w
zeO|0cyUYEpE65tvrvaw+TZo8>0Z0L|63P8#=#o-eQnHisH&_UuKo8MDl26%|?|#tU
z+%CGw`{)s@2T8{bDFgaQAWlKDyi{U=JfDVP4dj9#3AVww(O<9?qvIA1&LacjKw>s%
zfE?Unlvxd~9zX?RN<(jt>%|G&Lr<?IUKJoN;Nr#mIjPd;hjWN>2n!Y{51X;>#YW`#
z_&^LQ%1th_qWhin>Azf-U%*cSnPuve=pmPR>di5-Ay?=e<P%Uxvon3sr#yA&>L)16
z+Uo(g2*~4q<O<5I#eYBFCN>LU0TLL%o^OE@vAjjV^76c|V%VV09`(x;Psp<gq+Gt>
zb~GqZYyIAoVXe!h8y^)V_29t+K=4m~>BLTbu-R+lW`^Dsd9Y`Vu$ko3YiU~sv=n?a
zO-xlvSk_Nb)(OOtQ7v2j?s-|tC&RJ|1{JQ>;v<lyNXGRygJcgBIOT9=BAuHPI!rF`
z$@)3GA~@Z{eEv5lyMsLZpl#VhO?W5D1q6v9-xerU6l8h>(2;*fwZqa^1V2i{R=COg
z_7nwTA|gpJ8mza(%q8z{Jv16pQ(4N}@kfXL!UOsNLhv5mJuyS1+ww+t6uWLiYQSzZ
z7ceQiy`T8hLK4q>aS6?#0WwKFh~Smmo=`+icFS0b=)tc?rM^L3DCM}|Bxr-};?^os
zfaOHT#K@pw;ggtKQo=H4utZ&?8z+k~ap0IQDrD0vqXiMPZ{blyG8+fo>e3jZLb!gn
z!s8MXw6d~-cMz^(Kjr28>jDGA7JZ4V3bE~I18hL0<eCdo8x0^GY~wuem?5^K{#7;j
zYzQ`vNmyS(0EC#B;fNBRm+kl6-qL$;NrzIic6)@vkwXm=`U^t*d>U-QT^?5t><Pvv
zF(js7+1rP=j*hG>EV4+qzs$3>U+mJbPF}eAErUcvE)22n4;*I=1{<){i{FYmZE7;8
z*!stC`3D9@@E0iJr^0fS-G_H4cB${Oq7QHZ@rr9KK<wfQuRbq#d17E<Mu1C;<p4yz
zZAgL&*p%ztBh5Pk_|D$HfB!E14DTR1@IYGhEgU!6f-#I#C3wL5i1XGiVjLWtXZy=2
zr_^6QfA;51sRQ>b0Fhi_EgdZbY^4u9*W`xRHfLPs0Lrp!27<)yo$-}9N4;5Lx6|Dh
zK*r|-rvD1Ibw5=TpI_GE58U>tJ&KrAk~yq(iLoVc;H(tIcDnir@JW)YpJm|e20*4K
zNr7kqd7&k&pZoRS)&{vU9U4OfyA!Wc;(+_=fGD=+1_Tg_<W~nX$Q1AE8u(!#r<g|>
z9&!QE@+;#mL;auD0=+UbJm7)J_+X*7_}1c9GC)KR!?j2^1@hU+FIBE6P(K{J1Y;5t
z-{@ahjemb3n~b=t(g?0OP2^c^rBZ?bh_kM=aGI+u>M7`X{!R~7CCiu?+n^3|<pRc0
zT|&&P{1xE-EC+?V^~c1;C>Y5>V;>j!(_C#>v{hB_3w*(Z`ai-CIqnnL0a6|$tHxJ^
z6_nBfRF_@xXPf)byfC~S*{8M44X%mYbgw|>MM6!@(LFx_hqL9*@bqB*GzdZL&f${)
zWQh+MRXi4|9SEg&JwV$0(17?gz(s)(C&RMK1odKKKA@g1`HW(TmeTVo#Z!CqI}?cy
z^H1P(REt_2Zd+sxBT1m6Ai!Y$iO8_^(*?}`-UVbtgsFV;^+n?iRaYeDo-nLrmx|lz
zTsmri{;M?DWTiQ(eS-JD#2I&{pF|elb}CE6uIw19-`yU5Baa=`-2$*%c;kG>yv_ms
zBo&wAOh->IHDSEM%*%3pRx<n}ivUB4D-G#UFIt<Br;!uLX`~B~<k(TiFWrtc$bD&&
zqm9An^UwVmr+<siY9M{sOM7!=Qb3_T?z3h3pE8VFb|KqAC(ZAHQWb<!9v-#m#GLsV
zIX)}5$19}N*}Tic5T0*PC(86Kmkxj@#zk|Fm%PM+b5$f&CT{{+C2p2hsl0-ttQ)_o
zb`wozP7VN_mn7q21m=6wjT}47S+f-P(sBz+9}CCHS4u`y6`)G|g^x|zRrEdp7H~Ky
z$qn9oC-bx%Fw?`V;<Na5*P&?Q>|m|!V_wmMQUgI}dL6rBW=XXd!89(o-38MdzTUMK
z-#{ytC<5k0g=_FF^G_xt@_{w94TKCn3G+%8;y@d&M;|ZdS<CTZv^Sn#0U*n^Gi;L|
z*~=|ORimfJ<M;wiJ_J=XU4a|cp1A%v(!uW<guc5Xhg-K&wmzWcM^M<JB}&$O46CXu
z=GInZ_12fLD#e^v2JxW!_R%n&bjqR34<?dI=ND1P3uzK5o7Q#WYPcK5)@SVI%=>2N
zCV;mXXLz6@rOD9&wQcNK3Bc`xaJrB36H85Qr=^N7{B*E40h#i&gt~eJ2chWo>5{2u
zYIZx015*QE&{MR!prDDfg#!r|C3i2<ENYIdZ(Y*teOUtxXC4Uz`7W9(gS^)LHUa_a
zeO)=js>*GE=3e-vioI#&(Liu?J5CBm&`i~##v>kU-cii!(p7DBg@Xv(EH!z^wVdC)
zEnAZvG7Z^bz(7QgsPTx{6AhBdTQXkV`Jx;rlP|EQzM*txh_m8{d&AY5@>EB+JX{<@
z%lqD*u7xF!!1VISiO%nO*7muSdo!%Lp(%LJ$&InFq4+*fG9lpvd+a4~SijhXqE{QC
zJaV~9Yfc9dz1q1{QH{)KRXQf^KcOMcXwAxG*PGCje?5rGSejC`@!k!KTU9-1rLja)
zEqghumE}WwiuI-7P()NZVA?7T<#oth4-S4rb-L^$d8zt~S!7kyuW;weOfpOFvua(2
zZm>GYzq3Euk)3_ey=Y{<Z&KY@Ud}hPv%R&&9`;y8PG0iJnqPQgKwmd5;^?*iq7B0K
z_+ersl5UT~Ab9Od$~~c%g%QT8>By)kRz{4CY-dJVuix5}Dqqz1ouY+e^d|DU7auP5
z^CK*ZqdG#umVD>RvrCPFqV7DqD0x#yxbMPy{pa&X>@#kN5@@?5h<p0_1Tc<4JY=~6
zae=*Y@s=}_YFplZZ&+l^l5f=`1#|hnw<8Jp51s_{R$NJIB8jYyrK!W!kI0Pb5RA4y
z4|w7eY>)6F_7WJFUw!cYRk}~zvQy1BaQG;|BPtfRUpC$A>PV-pr>EzzZfCC1#H&sC
zHIu58{3fnRStg&8Qn{?GVt>Xh#<PM;%;><#DA^@Pyt!hFZ(5lQDjJN{FNq2e9i6kw
zjZcNI@v8)>Y)*XJ`UT2oD*RT)AW9%jlpub8Jn>5D$yFP<2~vipEmSb=W|^e~MXUe7
z7-`Pv#`{gRlq_$~pDxPGfL@K4f>&&8>g(?CS<Q`oBJWpBHiydDX=O&1h07n=S>1%h
z)`7^+JR{;U^)Hc!3i}GS+W@8fv2hX#)aEitEgT=>t91jC2#Z=Xbtxuz7^Df?*C=Qg
z<VJ^*6SKZKzsN77(KWnEoQ&WRA<hjn!jUYctw!`aQ(dS;Q_mp7e1x9iU|8M-?x(cA
zIg&3QM8>Iz@*_1>!mCYf^MXk;ZJARrbaFVN{;rX{j_I(SWcEf%d|<@S{1lnr@~;a6
zf*%N9WGpC+uXcXk+sV6d;cz#JqP|O8IeR$M3n&Z4#sfR=3xllm^gd)+^DiP6Y)w;w
zamu`<%$oBGhWw0A=JMx;BaUw>D#}*7bl(&qj6c!b*8K9|j6A^I=x@*0<saY9la~uQ
z`kZICa3<kk8b1Iun`YqRZ1O=-syZH8!4345GLG014S0&FNooURPrnbef*#t2WGvH{
zzG-&ZMAd&Y`J6MHNIc*C{mYh}@X~Z-`YRQ>e0CBVxon{lNp&HNkv);~V`?A-<vpfm
zQmKA$o#sc&TY+Y*`UsTh$Ryi+L?Er{*6J6s<qMw&e|c&RZC993xv|{@LVsT2N;Te+
zkA3s9zpbxfnL8*{JvmNMq54V*H|@jU{LqQTt>SYLTu!>Z=N~xG#z8NE1hIPYNro|2
z=SZOVBL;y|#b=&fGGx||9&!YJ)5vWhIf)B-pRWITfI+J?s)VU`_sv%8&ccsax(;1$
zdhYAT&#tG;Nz9yOqN8NyiS`sDy-76!IBM(G*DdwlrW6|K&obfsVN(0T@vY*PuEa(T
ziMn!@`a8}P4E^pK5oyL18vkN8FKVfo!U~L$7Un?@y}rRR#^oo1)*q>0v|Z9QDFq<Z
zAOPPGlaexa7wO3Ro#Z-GZmlXE=}V@aNMMIcX}ix0O;8$83i0^<G>;4A$UF@z{9*o&
z?v2m>ngQGD0YMyV&c~Ma5NC1?cRh-ziGy+C4EfXP1b3xefQZ!1#jD;UpAwJPVT~cX
z6-E~{Tp5nNZlBGeteCnq+UIE#w({V}dIur{7Y1K=YpZ0*ECH2>bAgqN*BE>D$B@#_
z>wHGNUK5UrUB;V-dKPh6m2;X`OlrRjhY+-k4GH@66W?GljZC^#7p{SoGGmlUwA5=k
zn;-S9lX6Gi!&}8~ai+<Bw&bfT+tLHX?~ZxDhFlhwhNYg_y?rYRNgQ5w)^(K<-@~zR
zw!A?hPmhGkDU52kUQ?31${t4!x$s}+H)P;&=9&KxuYIp}x<jTN2HAw=@ReC-ceB1+
z;!PKqR<G{0)(y-XFPAc;C7m)f)BqaykC8V0KXDR^(sQvqx?Y*mkzy5PwUA07Bg<py
z&^%~FzPJ>P4^U<n`U=nmL45#{NWHyV$yx7-v!v2oe_lPsQvBWBQz5DTOnpfo*FQCu
zr}WpDqD@y(!=0LkQAC2cnO`)XzCNSn%>A<QbC0i^{d)~rEt6vG!zbl<?KX_e%reNC
z**LzE2-lNcAvV0Aob)n1<nG0tedkdL6>IJh{`V8_>Q%{55^-T8RW(8MUCi!XyC`)e
z)c0mD;%hv4R`e}NXy81_5nQuY&PQ#(v)C6|m4fNC)SU;3y6+M{BdnO!TYDxZYX1BA
z$SSUq$z_j?Da`t=iUDLGdw-r68!u5Z{}`!0KgcI<Li40yGpIkQIloX3mGq{h(0~M@
zh6}-<y;2OowMFy0#5Npci9O3NR-u#6N^Cg}+m`gh7;)n%8jH0I)C!?%vk|DD7^!&o
zAwckqK=Ue@YLVsoxu<u33(1W~V29EU;^I%)KaBGm{r#VuS{53AS{ad?tKJz_3)4;K
zMhat=4a?$6WO(?+p|EC2o&4F9!Kt+Ekt~Z34V<j#>g&hr&iMgR`XGH6g05SM)xLi7
zoWYoi`B4t79oskSh}U}VO^QJhfga88#FH0#xuO;Y#~R*!i1(J@>nD*1f<V)EKV~_C
z=8T?}h&4DdSZ*`lzb{cv*lx25I-m&6T{PY9e~OOdN<Lsrc|7{^EwvrxUBj8|T)r>q
z*<2aZ`K7iRjON9`*{0$q<c#jA{2=EsR1_sWsGGY|j*cJqVbSL;e9X^kR1z4mRU5r5
zi%${bitEueI*7)978=GhBx$_C8-MZ?2(DtS2@wHgn4Q>5f;~+o<!{0YduDs`=4V*2
zdFy|bXxMQ4Szm>bR)uZR5FIZ)(FUET=9Q~Db^^PUb^52FVQF`d?r%0Tl8^|5sqfod
z5(z2*HS!(onY&UvOCvXTn($3U-$M2Q01H>M8ZC$$m;QL;4XX7s(pM!fMR1R#zx(=~
z@L4A`Q);mY7vmSW+e#Efyc)B;`?ij3X-9HF{wM2MmV0hz2r1QVotukqSUU}ky0$(&
zB4^C?lo_9UXilwX?)COMQ1Mj80o6Y{K#`FqR`uY18~Yyp*ERO6lDj0(Tw(FY`nN_~
z*IqoE-hF=V3!T!^`?{XnasktH-rv%$+9-P>CU#zx{80IQ5pFT{UDwvo=xA5_GKWib
zG&@tLA}H9*{BOE!n@PFj&SnbU00eKhB|ze{=V<})ANDx%%%VYai>aAEGO|fLo1IZ`
znOy6%SD{`Hqm^YtK#-`YZ9!<LKW_}Pg{+`aPv3hu`Ifuyz<}z|+`(AmwHB=nN6^(@
z%;AO-+utG&M*10B+^VK=``>y?wzRmsEr>{bA!|4|%Et53gLFSgn+SU_N_WJO<-We&
z<?C{yMIA?t*9V!s2DjaKp2+np>kt{^s(DzyVX~3kjG;VnEVy{V*^KrbZh|D=KjN<s
z=!6=qhM3%j<!m)sIzj?MHwbe%`dHN23K*oMNioPW(7gz|QI|TE^*@$A=e@!6RXHvr
zMlMwzCl_kOMLho0=q)P1Rhjl?%#^Wp5pQar@bp46D>u4vM5F%Z`Xaj#?%vq9q`x~U
zt=4WD6ppZC+)qMe{gT&iws|W0xGStFOOHR<z`&I1^ulcA)caJAEF=B$A%k{wSLALA
zhj|IL6yJOFRyN_q_j<hLXPK65WBX?o1a^!b6$?Aa8`RV^*a15M)UEWS-f3%;YubFS
zJ)?31(K4l==KDY|o-iFfezpl$PQ(5)Q5mmZm+;i-+z-0<C5tvZoG+^ai5Qs~wFB^R
z3cQ96aWv^xi?jxD$`8o8Y`#nd?@fCijp*#NALr{#eI0Vcvyw55w=LcCIQ>`SH9_`L
zDkmap-77Kkf^8uPH$clKXHz}2)Y;Z`<m6G})OwVjwv1G-;p3vr;s7<ek5@{!lUy3_
zFuLcxoaBP6+i!UIxFGvdKtm)0bc?wgkmZL>BUx#zy|Fwk<zBjqinxVFb=sg2$_lQa
zJ^)Dt+{m0)U%y#&e^&AoG^i7s`<#!1u5Ju0tRUWERj<9pHYQ^qQi-|m!6<+PJr`|X
z-T+L1)IPd1xW8g8MS`aWf%b!~8yet)_sz%sDtUc<^RFVbnGe8qQ_B8M;!0<f0bYO|
z2u)-u#B@$gB><+AaT72T*&|sff0`4p!GRhfLEKf&%E}r}dxoWk1&zUAO5_jtz?ZE+
zlLu(KY_9%`j)4)tTb1`OOi>M*2O(w!S=q0QF5^OX5lOXpKR|a0&|ek#0t}M5yQe1u
z@F%iJgL2CU*#~gv>?VZg0Fe4xfJe3^CH?6Pryp+g_6Dfc7%U$l&f`8z5^|u6V7BD5
zeFn>tq_Zp1pr#7A?q!I!7XLe5@<}>`5_GDClz3h^V-tX8o(BI!Zl<-MZj>gcfh8!B
zK~!`GiP%e+24hQ6q?S~3QtUJ=S-r3Y8UTIZ)rpd#o*jVreJlor?k3D#`X>K_yz<dX
zXuy~!@op!v${YpvAtwt<>g<pApq-FDQOM!r<<*r`T-CdULL=88a>yPJZcIffzA{rx
z#w4H(UcOBoIyKz%0Zl4yO8;LF+8RD!)FUMJs2A!hwJJ`<C6sEo0hSt9gMFS2p*7v$
z`Tl(F=E_8^LzBm63&74z&BN?rVN`<nUBp6oRfRx31SC2Cg3gI};BG|QyEgN<iX_b5
zX_^?e2_|M)Y6ybn4Z)hI{mF%vo!st*;ol_}LAQ?vvWx_2;hMjI7*Rl?G}QpR5c+!l
z)Yv%3uNcgV$@Zn1np&wv!}VPtG~l8G0*;=Fl!miaDoQ@fuPkUnL!hl(DopG=m*pUP
zB!=508)hB;=`RJa64pe$TN&sX6iNGv_>hkPP;_?P4jVgr2@{DA4!r0)M~6o5-j#Ep
zRQsNrz6I+Rx(BaY;>R)Zj)u0ny6NN}(k4<b@M43_!7Ln#XYQam2;8#(@e`s2jm^{&
zqUV{jtoNF<wtt)$Q}UR8E+7Q0y+C>>1c(qIxKOzSZiD16xI$Ozc&o&s_z)}tJb*G_
zY=#2H-vPa+U(nX#@)QMVJS6%uQb0c$=@CdiH9!w7K<-`wSyO5_K<)q;0^CT+cWM?u
zWM+8ek!V81U;>m201_oksM%-sJ_beHfcMRbcMz|H#(1N*I3AUNA`)Kd0XVQ!Lh#N6
z#V+Exz0b?H<0oH+i867qyqXrMdNTpK!{&9t$xuEKmEwKbj;jYG3Y$)y4@0o#sR4Rv
znF=Vm&DkGyEP~df!JvQ6`h1yS06A<Exksr6gnK~x<U<cX;X_13gjql<KeKFdlhN+G
zeE?Ykjo809s67ctOYxe}sy5NFt;IjoHIH&ZeSo(2!R!PzbYXFe%j3OCm3mAPpSt;q
zR97XIB_(_hmsuda1h6|LtX0BGy_OVr+c#rSnRl*MQce_78~ruL4ti^095{(%%<Jdt
z{`n&3?P)n6Fala9m@E1Dv=|D7QT|UbHM+RC*g7NcG*Ro64@?55<Jr@vk`zrdyQ@Fj
zLkflWRFsrRS0?T6%{Mf8qC#(A<(<n?J02Vu0E49lAZ{zTVHD6CoyDV>$qh?5PHwl+
zAeTefF|$=txU4Hwmf-f!eor4*!M9BXn(lz8@}$V9&cZ1_2<RRIfBZdtQO|e@6OwrC
z<2HeTfvucN(Do)!5-#t6-nzs;DIzX?<xC`YqUCM}7JDr0?8F_zuAnh4UmLU;z6SE{
z6PWPb-`|hm-=cM$M+ukRn{l@LaTF~?cIS=+=;2Jqo$mu~t~lp$5W;e>PUI~#pSc0v
z%@8AGq2T5w3^U22kOc72by^c$>w4u@TA&BaijBPs_Xd04T+D3{c%ObM7`f;De4dI4
zN@aq(<i-ZJ5U`{mUi5t>fT){j9Hy29&b!%*JL!*M;X)-W3)}(dS}F*ZDx;Lyz>KrE
zvbGy<yxd5djv6kuMTVS|V5P4N)hbXA=YTzd)r5dxoq;7IfVS+?U)qaTZ8yUvpa00d
z8*-;Tla-xaf!3z}=g+5`;B5rJ$|!*XSoFw9KJ;A&jf+3SCU}d90H?*Rd0D`^-{~Q+
zUp(9@ZtTto)lgNH0AwF%aaV=arSJP{`##NqM!zXv);so;aC3cPID!YP>p2xX;;RO^
zH`CSOaF??+6xwDJG!<J0V#9ZMkCMw~8351t4DN{Jht}qS@~M{sgs5>?DcVnAN5tK5
zN_TaA(ij6iU94=k>HGbEKK*S+RG%^qunKd$CjRPPhbcz}-~{lyIc6M}6IwqJh^-G>
zt3$6_Y{I5eGycyLGq#_<268+2Haj%AIWpP5<zwwI1&ghI6DB&j-18v=U!8Fr&Utfv
zI0r5yJv2+oX-pb^>_A}XQ(q-MYzu=P>0)s=812PNwQ?|c*Xra7{QDo19vc@7a64EP
zUcqPaX`dJ9vMJ;F?+F+4*b%|+ZSlTbx(9Gj<7%6?R5mRCJ!R|l1oWMh4)Q;!)sJj5
z@Mg9ML72CJ4U%%7>36ymjziiu@O92GsG1pWkFV*Sgo~vA{zkK%EP?3Z*8-GkLEC)1
znn!TM_CKHf@r*p(GP#jg5j)ESf=5i-vT5hPAHWR*7?srb$Mtdu5M$uif^<>(yjrO2
zv8itOKND`SBLbR;OptfMA1(^~vZ)b+KO9x$`D*SXT?hK>j2b-RiVSO>uB@yaf|pPX
z0R}uTJ^f>mNh25PKEN_t9pO|0$Pz?m+}4tYL61qvRjZN9jFhpl@o<rm2;7kCv^gw?
z82d*76BrT_#sAM6Zs2Bfr$6Irv?#lVAkEKw@XU7r!1T+6ThSbPPwndWd<%rbnA`Ed
zD`%d^fYyIlJ`D5gnwpxDZEbCEPckgUUk9-u`~DgWHOJ2X!|RI57Ooos=A3^t`E|<n
z|Eli2<Ej4t|Nmn~9D5%lNhqU%gX~q2y`rd;G7mz@9+_F`pvc}bO2$c&j0(x7agsf=
z2@${hQ?J+S{rY_0pUdy}{rA^DUFtl~^E}66-tV{D_5M=lkRCGW$&(IkB8tcR>fgY?
z&s(?$T6w)aE>Yg^ViARbtwDVrqBKK9xv<D^U!5{Q#Jr$sXeq0KWT2e>Yro%3JD12P
zX)|CEYXrISakJSQZ-OW)ao+di<5gka0nz2$r`kIwfuYlCp#$l@d>D;dEPhW3;!w~O
z1C>bTzrRjL(vq^WvhLs=KaR&q@G4FYHK|QQ){RwdR=#Nvti8v_KB_HjPsc5FjAh>8
z&3X+^9**+K%q<YiAIEp7B!#WbHLGoc_Lv0}DIomG1%1kg)srQ6hC|pLOApsMem>y^
zoR@28Wp9u{=>mx-QOfQ0?(YRV2>+LxK1ds-L@^2TZ2kP|EOPSXgIf!OI^I8jFp?CC
zBbH@keRRbIETqRJ{Qw(fkR)!|6t~H9o5-I&&GCBY*AtErCD35|g1W(aV;2Oj)^bBI
zEQ2OVspZC(w<I$ECo0MK5fDpW{r>s2+*AyNh!A;)D|-o>Q;mp-h;q~Dg-_3GT}fTZ
zGWIhGr?#}VW`Wc2ad%J8Y;C|Q0#j1^iolc_iQLMFnCbNlFG4mUmD6J%8y}A-3`Vx7
zetmLv2Oh8QgEa&<NY*TTSE#6|8OvA$tXwX^yqQ>e<Sjj8I7aHvjUy}ZfA~rh6(Q?w
zBUc>ip5nuvZDwifHKj}6kqeaAo|p~^3L@Q1NZznEOcNi{%hTMRez+^<e}b};hWFWU
zT?e}-W+|LeFP0zoaFhV+T7EVm9z`F5{C=-sa;<C`{-Jy6n+FXpq;uRk?+fT9YaNaq
z^>JJ}6Hkbe)3VSMxJ$c0Ry)gGr>Co1y4GL;V1tK@5TkSGN6vU(q^{-eRZ5}98GL6l
zBqyL3<g5>R$KLw!i9hz?r~@6ufewUb%2LfR?o&&P`BdeULeOkT28C1)yprj{{IvMf
zkcM~7#)V@pQ&Zz97qdLUBgj5*Wbw<9#Ep-MAYlt1zFt?guV_h%&j*e7`l1>=5I0=b
z@#y5fd$vJ&y{uRcy&Ee7jMxvZ#G$r>Rj~e<-+q3VkV$_gL!P6onu(9k+0C1Ar|Es;
z(jJ(po~!N{-Z^t(cV(5wzo8;@p&zt@Z@i8Vb<JR~8$qsCOaZ&}RZib(SY6f1WCBS$
zmO*oLG!*qN!a4dhv@*QX?q-LUtLrl#>0Uum(Q}DKiw^I~_+KaXTV3^=+vG-h#)mec
z+v-TF_;FJprKMj6YLhV7>MQy_%3gcVU;}f{@}tYKh<+*2eq;cD>4HO<Y&wmi5Nf{o
zxNCORpKro?lime58a5!TGd(p5stNpu59c<|xv=>~Z&p^_qT3)Y%96SWg=2K+UDaTo
zc=7CsS=kW;s?qlmYtzaih{c*IGG7e2<mNY5lFF=T=JOGi%rp|Zr9$jHd`KS$ladF5
zVzPIy0`S2>R1JyZJ|A(>4*K+yinmcZB(9CG_;I#0Ou|?<u;JT4x>!o?yo<}RP~ta+
zbY?|t;^e&!{&a1WnKE%vG#v9KfV@?Q2%gJFO8$P)bmNngVfPvpQ0W4SW{H!r9U|!$
zP&eR6$8by<ni`KmT0-~p8y#G!;m#2h9SxHomgSdUjJweeoq=-&5{4heDNFJ1++}2D
zHp(7dw8X$x^+5-zUgPA)FR56c(xpNsE*e+06b7t3JyXHqa_@Cqu3oz0b)H;kuv(Cx
z^P%*VLx@U!lMg#&)4c;im5H5-P3Vq5@*-+bq4pNLEe4wviiwFObf8Lx-0Yp!Zs5pJ
zKjA2PI0|ki<YT&0SrnhrV6hrl0Jze5Jpww8r#q~@n5?RIBF&Hv$NUN*3=hJCe-Ew<
zJoJ$ikGs&YkA{SVEOTaf_>^uwe)1JnO^w5#`VZhx@Z?oV$4TKONjq$B*QcKNWiuCp
z!3m)3!!a45gkirpW!cZbl4<AifqQjzI1(;h@1*ignoxo|)`Wga3GYL4beBVyROS;5
zlFqQ!%mO_D0D};{J07nNuk$aSk{WFNa}>OP!e<Ec3>aj{o)?tD%uVP58n_NyAW+e{
z|M+yA3wH7543AIVpaXi1096omH3g2DC>=#;M>N&szk&!PYw#{$5|;f018%d>TM($<
zxjQA{ze%$<R8n4!`AS*MD1bYKsG$QD58wh`F<_H<^A@?M{wHjyvFJarrH6IL?FVaR
zL0;a?tS1kkEH-|e^3?-4NuDTW+4lfVu?2Kk-hodrU$?flsxsX58m($TSW2UNqH#7`
z<`GIswh_%971D7OjLwpX1V~_cWrZ0ncj%BhFG&I4A3zkR>(>!!$}eymF+qeb$zve{
zX8;c%{(JuRYd}{4t)cTBpMTUn#Y}?@UCa*4ow4_faCaah%>4ZPMjMysi6*oQWfOWp
zm9&rQ@wX1hO81BJg7O+Pff!^J!AoLG0OejM^v<a^iP!kk;V^9o%gI^v*E93u0#5Sd
z2y{#Zlp~uqYm46c`V791o;x2##vM68!KLIUKRz}F@8}<IOZ@|Qd`A!92$b-hp_kV^
zANGbTsa(Hy?EqDfk;^e!dV1`ZdP|Q2q3!K+<Vb#;3?Gy5sU<E9*edc~Uf=jaVbw(2
zxG;21UbN#08dK&pbRl6gt82J1?4BF}+l02=ACCDLLf%SO+j>;O_||l{puyxRUeH~*
z4hs>;FCE}wW7CAzn>7ewm}jYKuz1kmAtKLOun9e>LTaKX&)up$jxpOWT{_T>#7_3P
zgg^_6ydQPctTfyUmL9h7orjb+jlwZQbPU+XWV6bB%w?-ZgV(Qs@|Y@U=q7Q3Fvsbd
zVqB(5J5S+CMkE;tbvqDP6b50dLC0Xd!RsXVzN!3++vFkvr_yD>axY-ZINzT*zk~ps
z%5B;J;1t|}h~c@Edpw5%_Z5KJumT1F110w<2G|sZl9G{W@FcU1P!a^1(C-?g^eaCB
zK#&mp#unz~Pdq%tR<{m?(=gX0RTk+bfY~En3VA$3^YB|_)OfU@py29zf!IBJW$Hp!
z%g2vwHmKVTYAZfw=vkmpBC))&(NTe!Aq)7H2m_Xjen9zieL7dP%2QaOL5~FV(R?Vv
zo_dR*lqAtj=tY)rOmrA|Yb18iQ_ZpLaudeix}D9%j41sgn#=X;xHN5P2!SMx+%<d2
zh4BT2RkqFBD9%5)C(cK*gjZItydrj2zKBPKO~6`D3&INQ9xmc=UUfKcutSD1Zm?FQ
zzfHz(n^3S4k^!nS;LYeA19?VI$%^l<RYsbrZ-B$gqWc&KKSSr5edSn4{&JYo_8S&*
zpcVc%`sp!^E15jch-Tw~!1(_U{*%nU5lUMUIP~)Vj)1VCJD-*T@tO5eo9Mjff|Kt8
z@Yq!g55kR(v=Jl0&@mo3bp;G+iX4F_JU7RabX4N7TNf+s6NJe;k06XE&T1#z=+Jo>
z9HqiBKf8QM*N}h+vi*vlLO6(fM!7$`gwNq%_406`Z9;b*Y(j5M`R#WGQB~dr88EIW
zr~52N2$R{qfssZ9RQ}B4fD-J2!AygbYad(!t!uWQy8>zRmJw(0feUp$FGNXfgKKo$
zF1$-~j1D7^ScDM_lxlP`{5Hqg4a@jjX}&K}WTGq@fE-oDh}g$EkHIDG3&*Twe<hDh
zDt-;@L*cE>%|{d(D6!$HcBjX#DerEf(QqoSK!O3*(u&OF{CQzqm*f|MYc|1`S^3Gv
z&tHG6Dm>C}hp5(q;mBZu!70la<&yKyR1?F4f`b{!RaOw@6&7561$=Oyn5O72(>U@Y
z|7J$zFbZhC=L)zr2gZVih6V`Mq&U}QssvNEaP?BUlCAD{ZIu2Ntz0-jcnE>cpaWBs
z)gHJN&6zq>r5X|SN|pFj$4o%~Z%JW)IHK8uyb*Zr7Qi{n!oeZzpnl1BEx|lJAejNH
zyug4h?&+iIMfCjU2a`|O?1xj*A&*28i96U(JF380)re1Zu>we>Iot2|uNE#vjUYdB
zzvK^pTM!HQaVTRIV!>HBXlZGw+_-X$6wbq1?K@8fmqs$dfJGG@!1d<bV=RCzLbB6p
z#3@_l1<{4#;gDrHkmozp!x0}~E!xve)r(kdA`mVRRZxaEaUVf~qi~A~PLDbkM!2HO
zPHn8J8YIdH2A$K1s8yYB?7hC^aB@`dU4ntmijgOu-v+<J@710xVy-s(FK|7AtTP-C
z1zON*S;rBtRhrj~h(Z6UCN%0R1GdQ4k9=m&9-4)i7@1JVJks^oS>c|6_T`!#nRhw^
zq1x`nF0w`->IOcxVv>3|_Wi(sS${Z0y;Kl!zm%^k5mf;Fj8!LAIX5Tgy<wl&d#WJR
z1VOLBS;usp6W+~Vp9|?DDN4btf&ySMp`<*%1C43#Q_M(cX#V2Q`ES~=9>^lX)LG-R
z35sN89Yo5GjL*w}j@=U*7GuLPsI~lcLgr`20@!t)Ni6zVMvD5n$D3M(mXfiv%8Qju
zQHF>-V6<Q!2Vdvq=$hDqt(ZaNX{8U2h8|`vC@vNS;~k@oRU>e_)FId_H!nam<5=gJ
zSAzyja+&YdJG9PWeMGb<YVu8k4CP_8&g#L}O#@GFYg!g(dm>Ov5Ao3gSr#!h?|F;Q
zNWkm-kv~bueDQJ-!Q{?>&s(?}Ms$eWLGW;9Vtiaf>G$dbVZOh8%>8mr=z!232Vy6Y
z$`}j<qKt^t4b0z)y}p{?k}Bxnj)qscu#!p2&^0b3h9ja?uq^Y<C}z)VLkB16bn%MD
zBquk33!4q-zL7{cVdhig{(8k3h0^)j6<ufpTvE=ZE!u)QaCuy(V_k#oW+aHO{a=ev
z{Q`Iu73rslE!uUSx4scuc-R=BBx86-DSM1bJ^mYEm1_czKRUmf@J2F08Ce1_bs?eE
zzX@0W`q%?)3?r6KH6PGNlWT4-0K-q;A5tg1h&O%W>jSV8s+&JSE=bQAF6cV>4Cwsw
z4tRov<{%ZRBm?p$qPAGp#Y&+4xX>yzqXpUzAsswkFbkpsZGSuJ6Or|)OpMh?=>6+V
z1qg<csVhSz7%quR^;=l4ueJmH2KQ+R*Typ^saYiWoDs7qpuj@)*ZDoZcXPq}eAlH-
zJl1`V6yOjN+Bqq{o~Xwf|HoIEsg2K`rM9ebKa0O3pf7qipE-#}c$9C&*0JNdq@7!#
zJXK7+Z}=9|XkD}KIrv4c(rtChW+kP>8f=QX$mE!cSb=CRvAs=rhJ?@lr)H_Vt@d%7
z5A=fR*V_xW@Sbu$CcOQ}6`IgeI*0W4AOu-3%C|A~y0~ykopk>hD(}NfmD_p$f$d=W
zh=`3INrGAH!n8fMt?;t9e7koo{iger+SN<p`BAEyQZB-KEQFHI)5HSFL#ICDmt#XD
z9j*zlEj^p+ImvD~@-D2fw_J(+uwj7>X{MI=mtV2DcV<t}irkfc$kdRi$injCuHo|R
zj`IIP-cp&Y;KlYz-d9G#N9F2!mh&JkQZ{CZtz2S3M%It>`_~iSMQ=BVO!UcEq{Qoj
z<Zk)C7WGl~o<rZ6<kB8#E;)b5E{Id?t+KhDysF8?N?0{ABN5c(BK7m5Hn9?84y1W}
z`dd=3@42HDJXl&jNkuEo2VPZhADJv`swc)P7W2c+;zcZf&W$B>v2?wN@7`h2RSeyc
z|Cy`aI<D*S{Gnl@c;=5HXXA18gF4Af+S6gNJ;QAlGj|g3OIpHW=UOv^PEsVFrAmqK
z9}VTSI@8nh3Y^!q90i_;BW_u^H+L$Pp=b6mueRHmnIi$8f9N$w7RPewjQ=QmNe)|E
z*~{)ia>0vh-s(B{_Ywiw$bM|~*G;}2u9COWkwMpNRGwCEf2~c3V}FoPtG#?r8lS3V
zHgb3?*@}MY;p5`-vnGJnD4Ns|Zn-heE^n!kSg!ngeSnod;h><e0b#_!vvgssQfttE
z>`H2n7g4{biHc<*4R<$Wai(KHNk4I>@V9d0#Z(R1h12nNi$$G_Es`$$HZQ$)zL(Kc
zv>cQPbG_}6H-FV4vzcJ9{aJqhQf!uqw-)+^kxa9%rLy@G0lUlIo8*|<?FPFyiSn%%
zc&xrvJ1zd~o4QFa_Nz0arTn5q9ZTJb)JT?)$f~B!`u^poAMF%y1$GAmkvZjB#1sV3
z9|#uoIEmT3QLO7>5}E4@EWfT_6PcBG!^p^zjpAeGy!MP@9;?MYOmj-M@YIrLQO!+&
zU@7V5QUt$JFY&9j9lBSDuQK*~DyM($OC34KW0hG~H63Q^-WP6t)q(DNXtO&O`a{o;
zvxqwsyK8XSn9YTgaQ%JWNK8UlqJi<X$myE;`o5(KnT(L)&yAhcm!k6tD7CJ|_F_L_
z`>r5kzI4<F+dr3%fY?PDzrm`xoLqMt8?O4HpC>B6QRo_@q{^ShCdrIFLf=#LUE;og
z9_1j7@U2@{4Ic-xJa;F#W5j84zAfX+mj2oybbxHhJG4}+UWWKZf-)q`n8v}_+=E2!
zT8A7J$9jy|az~hGRqUwO1?P>Twm8<MN;(aR!h^MU6S?uM!GSB;!a4T`Mk}c9WrdJM
z7Os}dqxAJGVlavFih(sgFKWVWPj#v!)^BPRYy^mE)hYhG=uDEPjF6$)d+da8+z2SG
z`~#(-@k4>bpC}ERRyb}1g?8B!pN=h2_x=m1rhEHfSb2M5a71R{n&mxx(f8Gn6FLSG
zKL%2RG(TpAU``ZdJo{ZDnPfQlP(wxG2|MlMHBGHPF6?FYJ*1OfegeVZg;2uCho6q<
z`~&-(rYBFGiaB_yP0KfdT|YT<A!J!=Fl!+?_s;Gf{%6)>h(e6{(uwV@`uP*$99J6;
zC9hvs*m@e>*VMLRWP1fv0CT)`aVI{$mylE)myMA!I2a>l6iwPclNY^#wc#!K^7z8S
zKx?X<6t4F`(#gg1Fm;?Pkv$Gd;yWxeg4q7mFA=9xXVW+mM5f)MXFnJ#P;%`!g=`l&
z0VM}dp%q>|{M~cw=>2<4_}(9)f=30+BT_C@rfK!9ub3R{ALI1Xa=M8>-`9aHB{NyI
zi()E}!|hNRTX#n`tz`HS^f)gm%{y3RCJ^pF@2%zgV)3#i^W)*8jCtwiujq_Fmo}UD
zUiiKBK5V_>T5f;ZditnV1MLes6Xl9>=~2MHo95vYxXrrFG#@sHPKou`8HE^{#+4RJ
z-4h56i?TBvRjs!eCu|fMM8wORuX$WpdYYx)ayG&wi|oTddt`gUrLvGEspY>F%Ap%c
z$rZ{Gi$=dMzP1(2%$T>G`B=e)6}_`f)%$Ne+zP&DxH?PND|IIOva)FNYmPIO-&c&z
zzBu~U@XXke*7>T3<vSj=_JU(W4Eqtb+s-3hjpi1Xxpy66c@^xGT^Wz;ZWYsByyA+f
zmU_X$$Ea?QQu#_cLSPSgc1O;sDcV}juipC11j%=~0oR+MBPGvmJD=N)DToEq@<j?C
zbrRFqXSRDd7n5z9+5Y>;J(2NEhUCh?EgNz{XI+rl?TpQhYgi5Lu+#}Blp97lbg&wY
z;P2pVt9*@QZE9IGv<)4L5BptX)lf*g5^lt16u++xz1yv8cG8x*?7fw~s$kMwgg}Dv
zB93J<_S~U`+;`KSZ!Ij2Nj+AjU6QcY<j!SdF<6$|^kO~ST|D%R#aX(1${e1PeOv_o
z9S9~;MS4<C-iVRkzh*5<y7taJxA;S=P&y`5?{=tOkgS~ptD3&#7S@_#@Jm?a-7Nf>
z7eRLj@h)1t6>qIWmXuQjE=&mM^Y<J~fQ^EUieZf5wQUUT>)QsnASP!0-?c@jd0vw}
zbV)x^ha(PDsA60NPMeGFQjT67nvvg>5-R5iw5XN0px*T@zgervEqsEni}S%{UzS#e
zup(<RC2O>Qg4BfnleyG=%Xk7wUq`mtD4sBMrYBtYbKrTwi!t$iQDSS{f$>Lfi52*j
z4jX)>Vo^%keOo_%r|R^~?t@%W{n#OjCe9tMj+{Qim6p-d8ApyB0kRsBm?cM-TZ>jY
zeFdHV6W;3ToGL{?X*Mjj#G1)me?ucITb*N1F<4W|qm-nnJ<%p*e_g%j(DBx5!IvbP
zOG~Zz71s5RSIk`grtN+`eK}V!?71*b^T4K8%d@9nE#jP&B}>}nb*bO#-3}Ty8cU|#
zQu3XNL*QD;`911H8MAW~<+fw9a{}6>@tU8p*&=sz4s8yElhv^t_YM@rI4Yf^FqO01
z%#HW@;V`Z3Yy9MHKI<2*-_HoAHlk%(Q(i`9d7CwpBjCT^juf1V!iFE8ctj@;l!ojc
zIB_W6<H?6y6N!=&+tFPWE<csODn0k0(n%I;`*E2&cYf)^ctj=AO?*f_fJy{xHW1X<
zP;*^7M=O}aiDi^l>_)#i9>ey5ir?RZuKWh8*h3zjm7V)4%^uedhGE|sMn=WmzZz6f
z>f<m}5fZ9@)%{^&t<mjr=B0+(<P-r+<<I;n|HBW4ds5F94zGVyp)?WtGf;*RhJ_g=
zI~FsWP1A8*Myc2E`Y#g1(rbgVg>kfcWt>c<IaK9!Z}8a7{`JGGl`T(f`>x%qzr1mI
z+-5s&oOJwAOU2#%(Ziv=FLGnbUdss{&20)!6e+*a`PY~ZmM)lcA&rzIXd72A94YO7
zGcBpi@2=6<0=xc2LpC0X!i@MIGLdltXL2_=S#}%F&&=~_X`gVqCXnzMU-Ro}F~i!l
z_0cv1J3iInlh+<Ooym)OD{kws;m-bCQr(n;ZKs(2pF25N-Li`@f7w{<aX{?7@Ieic
ztX|HY<qXy}IUTF9Z#rz|%$CfiCGpF2O=hf0F<JA8*;{P%>F+CwCbe>JpwH%udNY;0
zMQih>D|50pUTshD-cBE#u!y_8NV-D9Yi=hkhT+t0E7qwTpES&}^HIE8JG(({+I)|a
zvgko7Xm4$|^{W;dhp%7IA7dR*aLu^A$HPn0Wp>SeE@6(Fb?~-t<H@=I+p;uXi$1Tj
zX!4z5Z~9oKMP)|fuHZdfW&qud){!Q1IV)35o1gNDVZE}}3o*#x$v=wP`Zl3`fA?H=
z^Yz*GvF`YYXLzUkllZh&Bg^kY8e+4uH|KlWeU8_Uw9##F_qV49U1O|Yx$|H_-CXjB
zNKaFxZ{O_E%A5@L=Y=<?m?D2?Z)cq@%+VVC=3ZTyQ{{6b(~^EbM=Izy%;MeJd9$}H
z_I1}LiHc{6T58%=<t>vp3UzwvHdk(xRh5q^TRFA#C4KUeD~VaO7r<@A@9N((d8ef4
z_^z98J|76CrW!x_zl$^CJ*C6@{VY{9UQkzf+8@}g!s{&`nv4|DW3zL3b~+v7Z7YR4
zzr6AKoP=A+{QFZbhS4weM`C;Zx~_iq=aOa}Xjrz`zi-hlTJN?;*G#3VMFfT{IQv6h
zZlPpCxq|EOjFe`)v=V;XJ0_RV?Gh;=o>_d`S~%x<*m_sLRS~_@XLq9G?;~^FZBn7|
zKBvd+{i}Q17pc#)IW5^c*Lhmr@>82X!g(gxOK!-pvvvQG2O2wRb(Zo=Go%CEGvDl!
z;pYwtUyl-{2@{ch{woh1>mrqu7)_gX*4f6$PFXb9m+Lg+49y#bN1qFBOE>%fxH6d+
zxOy|`$vYu^u?wcLvsN!}=_J+*wmMij5BsKU3Td`g8q(dE_^cGAa3E5Z``PP(7uJnm
z?~N5K6uSP^PENTtCS_TfJ4)~O*vLJvW%tAK$&YjU+r0!+9=~KV5VVnq%@Dll`(i4|
zTKH+Du>IX(wFyacNts%?7{i&D`@WCUT5M>RKDgvR`B&#Np6%~@H8*2&^j=iJ%43;x
zh2`g)Q?kfJ-{#1L8R5Oo2F=;wEn~0z=(g@jAY<84y}K!?L6+c`dS!$5e6zuKquD*S
zt9)wWh~_<hZDHJfue;LGY_;*-gjpv~W83R;A=~0P2KRLO3r~cdFg5;uePlMf(%=dR
znz_Z>D04+vq&zss`#+D-MvS-G9{IWAF+aO-$Rc`gvDn?7b_$pO_{88Phq@q_devkW
z$Isth32$$n?;G*ttQbqJQS3`J8YFqa8+n_yIjZ_oW04A=N4Yi?$N-C2Bv6<C)6S#p
ztW$F|8_ROOWF$Jygly1Td?uP0`t80M?o3((8-ti$&Y2r}xt8Km+N)aPpLe1aJX9RM
zFNJdCeRQT5&&+sw2OsY|>io)Fi?i&me!(#J+bHKRCdFJ@Ph1=0;XV31$!9D}At|lL
zTz&h{S=jN+ed}Gus~*Pr%|jFw_I4ioO@2+pf2qD9`#>dPmN2Aq-yu=oQ=vcR$m@$=
zUzFt#vDUW3Dr#M);^VOrd4arsPDM{(vs_blN3Oj&x<9bO(>unTDPb%B0GFg-+D(Zz
zvNip*^fS+SW3omzTPkGgHHBo&&P}u_NUM4~8~WoTF5xzO#ZKR7r4hFA6V@n+^SGsj
z+r0@pM$?=^#rFwD<`JSNZWEL;d}YMWI}vHk<v8>V=r@eg--ocgmJKdnzh?CFCrhNi
zNaQc~rOanXBhk;C`9nOXhN4YJHPJWq<YOU#01@L(C?>=k@=MrDs0mrN@}%6@pR|l0
zV^=dy@n06ukJ#*;k<L7#p|vd8r`a@kU#sDD@ekFDiP;>3<g^`2rVVBT<x6=tFPb}4
zoqbx(w47Cm)@j4n_6?i7F1u7=XJk;ohX5TPZY7)AO5?UC3Gv4o2Ue?16nqlGb6d1b
zRUWPFG^OD9bWKL`ejOd$v!_)q(BbML=9LBKpKb?9<ul<K`WBd03dh&Qjk+h3o%MD7
z7_}U=YT^@8i;7b*r5jv+yHh!%TE7hY+-Q#r{pxGha(`OJ^e8JvAynq1!CEcJPsjHT
z&ls1oiY-QB-wew%F9vq4ZYTSvScJqsmwfumdtmA5tB9JvXE9?tyTWPpO6TuZydf5@
z^&id`-z~Vk!$yBe@R+ED`O_1+=0{1-C;H?ZfjGg@<D+Pj{vVq;8@e$jhO?=7yol^k
z0a}u-jRo}Cjvq3DvcdkEHSx74YG!UP)0kPGbeyXyy_pS5YKzNDw>fq~TaLF{H-2YW
zw~u0)7xItk8{0qA;M=<K<>sZH{Uz_B<9}3!jx`)SexLq(0E6J6bBez>(?_4&l_+`{
zYs`{=^j^CnA#OxLOr_=JH9<;_c-iOlI@?s?Cgq~(7akX@_lIjNGcsf!YTGc3XS%ZF
zp=O<{n|6ejFL5yTxVHTU;|44>r-12M#bG@H_)Z6(V;3JA<QNm<It-rwsunZt?&xQu
zK`RM}Sg`IED-<cpczG3c|C#Dp5HjcVHA{rd8Nzk*s!o|vFt1<W!70rTmZD6>wHCn|
zVp=z8TMVO^7>l*uZ+Lf{&p31qV4=kn#b1cRMlm;pj=H)D7&}u9M*9Dy@V1*;D0Cwx
zMNOt{J?x+2#~HIsiWHs=rB`~;*AoBC-*|2OZiReqPO11!Y4eC<zb{;f1p}X>Z#w45
z!|2;doJ{c=j9M8R^^IrKG8al)47GZnWn`T>n`X&+(&wG1e!PjmC)w-SMpe9QN=8MG
z9-7~Is5$9|Kksk}>-~P<O``U@gjK1DrH+@E!kJzfW1eZjHot^(_uXQcSH&Otr^)&p
z8#}vEHO|6ZojhY2iYJ8n_&zDspN1q=s)DyEz7j+$^-D<1Nq*<x&9_1=XWE2l>J#=b
zpcVgo2JtAs-k2nYBgMraw)a+lUC@KCXFn3#+Y6$oliF~LUz2B{|Az^B?kEx#)cX^7
zFa}PcO+>*{eW909q%xnO(3VY0ZLA%|2xY+NU%`(WkiY@jVyxu>$h4#X3PnIHG$wzI
zkeL@jUIZ$`a&>noX!gJV{`lyq8jqYi!+5&iud_r=Y83k2zvG1{Zf#IPKn#QhvU=&j
z<J_97{s;7#3rQqYOm-6CO3owlA^_UK4?(u~e-C0xA>qw~5N`(Vqb-#OpCLq0A&20?
z{i<m%f*QP!9zyUSX@aQrA13L)BB+OdHk`EoS|yf{lmsGrjm$MEVIZfmH_U?QQP>+Q
z;F+605_Vu6p;R877Q-EDk5YaHa1HpmmX`a1sHb5y3ot;NA|d3o!4V{c%*w)YNLwBB
zX(6J2V@E6YLr8mRl@0qakEsB(V69ADP&)^TUHmDM@IOMxy`YNw4`~#w`VVQ8s)Im+
z<huVNjY_S_K7m|!z}v2?#LLOQXg2SH4WNz7z|=JH4_1_RUC@$T1&+|%J7VkT=_n3<
zZFMN<2hhj@SOFWX3zk6Y!&c_iTeObO@Np}xK?12gkl4;u3y~>*+xU5P)XMaq(h`;b
zgypfM%|fM+VafU+<PNf2aQ9b^)>+dA5yMuQ$T{IbHBLdY;AYm%|0a?CCtJ`$ZV&2{
za&@HY3%u1Y1zV9d^wUW5`@PDImAs9{cPXqB)gY}4^b)GO2<!pX3!Fe5$em~XfR;7c
zk<M=O3SM)8667t~2OT&-E2XUtptbw(+@<!abRcOBeSfc5Ofpa47X(o1l=S}!DE-gr
zVw9I6BnM3zg*>5b7Tn<k!I>lY0OW2C>(Roy^kcwEK30APg2PrO0}%1UFDBPG1IpD3
z8bUAHRjtv;^9OXCUZ5kHH}aaH5_`SRBQyqmsD5}&CmlMXJtQhMdIa=+RN)Z=_6^}^
ziISj<K)HO>OD~NBp5q*nvXTq5{x5oI)@|q=pjR^P@*xF+sh$Bq7Bn^iv<Ee<4g)On
z)I1GQhy$rOq7O;?dby@(`((Y0PY84(Vy`w5d4}^$UB<is@^Qvt<m0+<;?zj<J*3X#
zUO}6`l9M|K_3Te_vLS?~_u19c=47FeE2494d@Qb+u|eLEdI~aQ{iT*ifhC$VPe#PX
z$@ks;&k>N}_`e~W{*zW@7toK=b2m1|7c;(E>%UM=o*^*<l5WzL2TpL$CJb2n?sh5`
zsQ1C3u)Ww=9;){sLn$%+ASehOxPY5d90&tP20Whr>FeZ?OVs{QgvKV?cDywK&O|>W
z-Qi23@)X2te-5!Hj|5?Amm0moAqT>9%GJhYx@70N%}|7a=Ae&(yzS8@^jCK&(&mk(
zj>ob_{o$Z|6xT+5$-~VO_c4&q(*Gw-`wS6E1DLm%(FxSdX?G?#&B;1>trR@JRM?ek
zx`Q+f1TPPy3P7&(cQ!cEl?+LxLxLP6Yd~~%GOTvRZBYgNMAEyEV@(xj7XCT&q%7H#
zb3!k%?A6a?NQ|SAJbTnv`D$NV4mcG}Q*_Pr&HWt+8ch0v-9z^F*?;3u2g?7!p>o08
z0l76upQkE61C7c5MG>Y<SI8T;F<@s-R~vb1K6?gW;jA}~pPs7pqvSEwHdh(0%~a7v
zLnpHcAsG1A=w<v*J}OV%zxk-apcq1wL!VBsL+1c{-Ut_EoyLPDMJ1oD!vXmtUuPdY
zc%bpf`PHNY7w+Ypw86f2%0v_l<WFPgPiJso<MvRgT>-}q(L?R2(&JPoa9^o{VvwAF
z5aEhalyc#R6l^Sj<}_Y_A{m)25Z!tKnP(96AEbOJ;#<$+WoA~}E1Bg&2SN<@a10t`
zl_6ncTL9|{75fYbVLK>b1}c7P&~m6D%66MHD><)KVMPr&7~4FdFNMBgN!yQo2xYb2
zb45^sdaND+r&6lCb>dX%++@JkKj6nn5}j!!c)e=0V49uAzTQO7eSUtr_f82r6m*1X
zGmCC01`%jH51CICx70#9xRi_Z-^vgXc??q9@Yj#$9qKPYoPrdNGla;yI!}EnvR*(W
zO*K%nsUVh5$%<Ul1(IlgOI1SuUFD3IoBJH_THt1M0h@PDfbV~BRWBl95sDGbQDBo|
z2>eJbC$5Ol&{qL_a8=}X*i@cQ!#fjFCc4mGBr~A_^{#zS9(oQcuJyhK3y}TiazWGU
z6xEA}ZBaf7!B?_11m}S!W}eoyG0<~Ay*(DNs%aCF3h@sfjGLj7DXUdHQpTkB?`et9
zBcZ@WE$}idd+?^byu69~M8K0JDJdA>%R@cFi14%tka~Wn212G?koY4696Y2eXTYk&
z{3edQlgx$HYA7an2yy`GuK=DrlDG+#4KMV*c?B1J3ND)NqPmqH$Ot!v$Pj9Ift#Ao
za*I%NNd~I49tf(Bk5k>eFeV*wV5*J*n^6ezERcVmTAllv1gcck_7t;Mj-akWNLWaL
z%mOkQ8ib}woZzTdL9nuKKyRsRq(XENKcs()>mU4Tv>Jk64V@wRANbWE5L2Se)QAXP
z6}{NRtk((EcWt7Z>>Fz%j|TC}a8-*S#)v4woGXUB0|aoTL4;wTta619wChVeO!@<2
z<w)H3FA!_-=aCoC+CDR2q4uD$fcojvJ*O6G1sqJ$g*W^4FCN$*-Ru<pHU1a#p=v3d
z+YlUAZw|-Oz_IOV3SWOP3wB_`LP8?GYLajTeTBk@Yn3#_w}g^FX?T9wpoQb@54;%%
z{=F`wRAPz!EB3mtB=;r)6+EB41f6f(QsULO1LBuAEJ4*-&j>O?N+#ibzkq6M+nsY3
zsVscwuCT1^BjD<Q1EUI$X8{dlycAP4_lo67!lDrfGBS?Cx9C887>*GwJK=ayGS;=x
z)R{<i8C}ij!7zrDf>hDpg5xsbIF4B5z^kBBJqpGBm{X*<_}nHw+`mL|?46U^H&eo#
z3x?DeRs-H3OH?>FEK6P1_$HmU7ZX9bZ=aE&ArlM<I=mzbqo=L|KkF<A)9;@ylHM4b
z#yCFD&<n@tvp1pF14ZP8hC@G`vP32|UjDA3|Dv|4#C_dg%}K8ETa_5_1(biB$LT0^
zE1{_A^Gs8}?adDV#7h*fhTWV$^Ae)LUV{tZLIk{|;|F0N(U-eD@7TWqvwaBw!dq_y
z^dn@`Qoy?&CODza!zdb#0ro1MCKY(=oJxMR=BI7H)l|lVGA*6Nq8W6tmA{B?7lx!s
zkeLatzy&an$@Nf{s$Gs;35ekM?ry_P+Rpt5(^RYjTTOyIjH!Jip*NIC`_w^?UjiEL
zcg1cM4g+-QN2Km{2&NmY$xgU^oaD{mXTVm_yEcNh&t4UaBkEgQTYr$-uLN@N4DM66
z&MDH!qY=Dn;$%XHRJvQhThQGe2HkBic`MrSYn9G4=r5yVf=}V(V=GVDY!H`aDP0aV
zq31wB43#U1<o*p05t*?r5J~gF>|>xvLuCC&YOJUb1o}u%82&m!$y?jGWn7dwr5(>l
z)3I`M7hk)6z4b=m%OSUh$+ycsOR{5b)I*^AOQUdA19I4lLU^wtKo&3RdIR94g0YE-
z)*GNlbgQ3iSl(Wg9d+B+$-;mog4(9)K;X@*F!Y*Ofi#ROHJT6L!Yay+q}$0W4$gCr
z7hjCHQFh|tSo)DrY5u@v6lkptp{>|et}Iz%II?al`bC#~<8P!_#QMabh}0lwRU>wS
zONMt0WLpMECEq8ht{`P1*!EIyNR1Ele{uLY8Pq`?a0<FT4AH)};Q4)Qym$anY`y+H
z=ZrA!V~uAKW255dPIxzkAnIEd{xk|mVZ@?f0L>QJ5%QL>L6zd&6{TG@K$<=U&&dvG
zH9<`srp`sd3+1veffWK5ttaz#A3cxr=W-22%uGXBLdozgzeR`kGT8tjxByXTG7VIC
z%8Rw0_xsB(=}}WtBOaLd*?mC7JPpqx^S|h-_xY*1VQd}ivImWJE*ON7N`D0~A1w<9
zOsWxWWtz~J|1uG<G`@!xj#P8je>L301QGlg#Csb%(*`>0X)ylegA>J^$NmOl4(#F7
zQ&_{xlV(_kV?1D~qYWpSfyFW)zEnXil?R@emh1^wHVFg`gK0)YQquVOfZ}2$HB`lh
zwkys^4y!r@x(b@<99V)IUb}W^ZKh9~mjbll(IC&(2MCAUQz-9uocK5~Q4{tM%eFc@
zJ3FhNuErwT9bsYNa73Q-dc+)tB&2e9-Np#b7w~Mt-tB!^FX%)ON3nIXHOLZ|g;57^
z2XM-Bamv%Q{#heerdC(oQ__z`?^#aeJ&&QeB*Q8HrmXsY8PkjS?zbg|;M&0^*p#XX
zq^#`iB9H9|SJ)3(%LOT^kRdC37}Y@5FZY(@(0|M_nMl2*vQKXt*Fl5Erq-w6waSn#
zz3_5D(KWgW*$yF9%6#&!ZUeH+9jcwVw!c~b?ym=aCS%Yj=5c{Q?TXuZ%Q=9#<p8J}
z46)=egGB_;q0q=qh=Q7h3Q6EM_9N`n#!pCr0VpN7+4su<u~|VHp7^E*3i8#=mcA^2
zj)fEiWQ;>fGJ*X2g{S!L2hcu;qkQJRNFjCPe{ZjJ7()T|dB+#-(@fy>GM69V1>Y7@
z#xjpNl#b^}9($A$Gm0B@o2vi|`GiQ=k5{xe5|K?MDAO4q6RG!o1DT=0?#}OKcM?4w
zn0(FrSWy?z@C_i1eF~8{Jbq;nG~l`y-CNW~YePSPhPWwC)?d#3&<^=~wQX3@usJ;0
z088-jE*Kjxo!htA{1`Pf2<z5+;Kft3B)F*1rmyJB(T}fm%F#cA@+jf_w|Tv=l-hJR
z<t01RRspikqV`kXy+dtU9^cXU3kDOW4hqy9NU-#vq6rFxs@El`((U~XE<X74<M03e
f|NlS5)pwP6V{e-~85G1Iw{})TSG`pAa?t++)dcs<

literal 0
HcmV?d00001

diff --git a/docs/cookbook/knowledge_graph_extraction.md b/docs/cookbook/knowledge_graph_extraction.md
new file mode 100644
index 000000000..c7e347dd4
--- /dev/null
+++ b/docs/cookbook/knowledge_graph_extraction.md
@@ -0,0 +1,137 @@
+# Knowledge Graph Extraction
+
+In this guide, we use [outlines](https://outlines-dev.github.io/outlines/) to extract a knowledge graph from unstructured text.
+
+We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```shell
+pip install llama-cpp-python
+```
+
+We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+```shell
+wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+```
+
+We initialize the model:
+
+```python
+from llama_cpp import Llama
+from outlines import generate, models
+
+llm = Llama(
+    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = models.LlamaCpp(llm)
+```
+
+## Knowledge Graph Extraction
+
+We first need to define our Pydantic class for each node and each edge of the knowledge graph:
+
+```python
+from pydantic import BaseModel, Field
+
+class Node(BaseModel):
+    """Node of the Knowledge Graph"""
+
+    id: int = Field(..., description="Unique identifier of the node")
+    label: str = Field(..., description="Label of the node")
+    property: str = Field(..., description="Property of the node")
+
+
+class Edge(BaseModel):
+    """Edge of the Knowledge Graph"""
+
+    source: int = Field(..., description="Unique source of the edge")
+    target: int = Field(..., description="Unique target of the edge")
+    label: str = Field(..., description="Label of the edge")
+    property: str = Field(..., description="Property of the edge")
+```
+
+We then define the Pydantic class for the knowledge graph and get its JSON schema:
+
+```python
+from typing import List
+
+class KnowledgeGraph(BaseModel):
+    """Generated Knowledge Graph"""
+
+    nodes: List[Node] = Field(..., description="List of nodes of the knowledge graph")
+    edges: List[Edge] = Field(..., description="List of edges of the knowledge graph")
+
+schema = KnowledgeGraph.model_json_schema()
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
+
+```python
+def generate_hermes_prompt(user_prompt):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + user_prompt
+        + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
+```
+
+For a given user prompt, for example:
+
+```python
+user_prompt = "Alice loves Bob and she hates Charlie."
+```
+
+We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+
+```python
+from outlines import generate, models
+
+model = models.LlamaCpp(llm)
+generator = generate.json(model, KnowledgeGraph)
+prompt = generate_hermes_prompt(user_prompt)
+response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+```
+
+We obtain the nodes and edges of the knowledge graph:
+
+```python
+print(response.nodes)
+print(response.edges)
+# [Node(id=1, label='Alice', property='Person'),
+# Node(id=2, label='Bob', property='Person'),
+# Node(id=3, label='Charlie', property='Person')]
+# [Edge(source=1, target=2, label='love', property='Relationship'),
+# Edge(source=1, target=3, label='hate', property='Relationship')]
+```
+
+## (Optional) Visualizing the Knowledge Graph
+
+We can use the [Graphviz library](https://graphviz.readthedocs.io/en/stable/) to visualize the generated knowledge graph. For detailed installation instructions, see [here](https://graphviz.readthedocs.io/en/stable/#installation).
+
+```python
+from graphviz import Digraph
+
+dot = Digraph()
+for node in response.nodes:
+    dot.node(str(node.id), node.label, shape='circle', width='1', height='1')
+for edge in response.edges:
+    dot.edge(str(edge.source), str(edge.target), label=edge.label)
+
+dot.render('knowledge-graph.gv', view=True)
+```
+
+![Image of the Extracted Knowledge Graph](./images/knowledge-graph-extraction.png)
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/mkdocs.yml b/mkdocs.yml
index bc06c4993..7de4a6350 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -107,6 +107,7 @@ nav:
       - Playing chess: cookbook/models_playing_chess.md
       - Perspective-taking prompting: cookbook/simtom.md
       - Question-answering with citations: cookbook/qa-with-citations.md
+      - Knowledge Graph Extraction: cookbook/knowledge_graph_extraction.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From f6a6c297df8219f00bc7233a5a6e990d070ecae5 Mon Sep 17 00:00:00 2001
From: Alonso Silva Allende <alonso.silva@gmail.com>
Date: Thu, 18 Jul 2024 15:22:30 +0200
Subject: [PATCH 020/505] Correct link and add llama-cpp-python installation
 instructions (#1051)

- Correct link for llama-cpp-python
- Add installation instructions for llama-cpp-python
- Correct first question-answer
---
 docs/cookbook/qa-with-citations.md | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/docs/cookbook/qa-with-citations.md b/docs/cookbook/qa-with-citations.md
index b8e034a29..cb39befe9 100644
--- a/docs/cookbook/qa-with-citations.md
+++ b/docs/cookbook/qa-with-citations.md
@@ -1,11 +1,17 @@
 # Generate Synthetic Data and Q&A with Citations
 
-This tutorial is adapted from the [instructor-ollama notebook](https://github.com/alonsosilvaallende/Hermes-Function-Calling/blob/main/examples/instructor_ollama.ipynb). We start with a simple example to generate synthetic data and then we approach the problem of question answering by providing citations. We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](library).
+This tutorial is adapted from the [instructor-ollama notebook](https://github.com/alonsosilvaallende/Hermes-Function-Calling/blob/main/examples/instructor_ollama.ipynb). We start with a simple example to generate synthetic data and then we approach the problem of question answering by providing citations.
+
+We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```shell
+pip install llama-cpp-python
+```
 
 We pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-```python
-!wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+```shell
+wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
 ```
 
 We initialize the model:
@@ -16,15 +22,14 @@ from outlines import generate, models
 
 llm = Llama(
     "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-    n_gpu_layers=-1,
     tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
         "NousResearch/Hermes-2-Pro-Llama-3-8B"
     ),
+    n_gpu_layers=-1,
     flash_attn=True,
     n_ctx=8192,
     verbose=False
 )
-model = models.LlamaCpp(llm)
 ```
 
 ## Generate Synthetic Data
@@ -53,6 +58,7 @@ class Users(BaseModel):
 We can use a `generate.json` by passing this Pydantic class we just defined, and call the generator:
 
 ```python
+model = models.LlamaCpp(llm)
 generator = generate.json(model, Users)
 response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
 print(response.users)
@@ -140,7 +146,7 @@ generator = generate.json(model, QuestionAnswer)
 prompt = generate_hermes_prompt(question, context)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 print(response)
-# QuestionAnswer(question='What did Jason Liu do during college?', answer='During college, Jason Liu studied Computational Mathematics and Physics. He also worked at companies such as Stitchfix and Facebook, and started the Data Science club at the University of Waterloo, serving as its president for two years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'As part of coop I worked at many companies including Stitchfix, Facebook.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
+# QuestionAnswer(question='What did the author do during college?', answer='The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
 ```
 
 We can do the same for a list of question-context pairs:
@@ -226,3 +232,5 @@ for question, context in [
 # 'Michael studied Game Design in high school.'
 # ['Michael Brown is a game developer from Tokyo, Japan. He attended a specialized high school where he studied Game Design.']
 ```
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).

From a7e3381a6aa6d4209f1f3aea8a048c1fb6ee10bc Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Fri, 19 Jul 2024 11:55:30 -0500
Subject: [PATCH 021/505] Introduce `outlines.models.transformers_vision`
 (#1052)

Rendered Docs:
https://github.com/lapp0/outlines/blob/multimodal-models/docs/reference/models/transformers_vision.md

- Fixes https://github.com/outlines-dev/outlines/issues/787
- Fixes https://github.com/outlines-dev/outlines/issues/662

# Changes

- Introduce `models.transformers_vision` which subclasses
`models.transformers` and overrides its behavior so it applies, instead
of `AutoTokenizer`, `AutoProcessor` to handle the text AND `PIL.Images`
media
- Introduce `VisionSequenceGeneratorAdapter`, handling and validating
the `media` argument.
- Update `outlines.generate` to dispatch `TransformersVision` models to
`VisionSequenceGeneratorAdapter`

# Tests

- `tests/generate/test_api.py`: Test `prompt` / `media` validation
- `tests/generate/test_generate.py`:
- Add `model_transformers_vision` fixture. **tests pass locally, but
disabled because a model small enough for CI isn't available**
- Test all `outlines.generate` generators to ensure dispatchers for this
new sequence generator is handled correctly.
---
 docs/reference/models/transformers_vision.md | 112 ++++++++++
 outlines/generate/api.py                     | 109 +++++++++-
 outlines/generate/fsm.py                     |  17 +-
 outlines/generate/regex.py                   |  31 ++-
 outlines/generate/text.py                    |  20 +-
 outlines/models/__init__.py                  |   2 +
 outlines/models/transformers_vision.py       | 139 ++++++++++++
 pyproject.toml                               |   1 +
 tests/generate/test_api.py                   |  33 +++
 tests/generate/test_generate.py              | 211 +++++++++++++------
 10 files changed, 596 insertions(+), 79 deletions(-)
 create mode 100644 docs/reference/models/transformers_vision.md
 create mode 100644 outlines/models/transformers_vision.py
 create mode 100644 tests/generate/test_api.py

diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
new file mode 100644
index 000000000..b0ad6533b
--- /dev/null
+++ b/docs/reference/models/transformers_vision.md
@@ -0,0 +1,112 @@
+# Transformers Vision
+
+Outlines allows seamless use of [vision models](https://huggingface.co/learn/computer-vision-course/en/unit4/multimodal-models/tasks-models-part1).
+
+`outlines.models.transformers_vision` has shares interfaces with, and is based on [`outlines.models.transformers`](./transformers.md).
+
+Tasks supported include
+- image + text -> text
+- video + text -> text
+
+
+
+## Example: Using [Llava-Next](https://huggingface.co/docs/transformers/en/model_doc/llava_next) Vision Models
+
+Install dependencies
+`pip install torchvision pillow flash-attn`
+
+Create the model
+```python
+import outlines
+
+model = outlines.models.transformers_vision(
+    "llava-hf/llava-v1.6-mistral-7b-hf",
+	device="cuda",
+)
+```
+
+Create convenience function to load a `PIL.Image` from URL
+```python
+from PIL import Image
+from io import BytesIO
+from urllib.request import urlopen
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+```
+
+### Describing an image
+
+```python
+description_generator = outlines.generate.text(model)
+description_generator(
+    "<image> detailed description:",
+    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")]
+)
+```
+
+> This is a color photograph featuring a Siamese cat with striking blue eyes. The cat has a creamy coat and a light eye color, which is typical for the Siamese breed. Its features include elongated ears, a long, thin tail, and a striking coat pattern. The cat is sitting in an indoor setting, possibly on a cat tower or a similar raised platform, which is covered with a beige fabric, providing a comfortable and soft surface for the cat to rest or perch. The surface of the wall behind the cat appears to be a light-colored stucco or plaster.
+
+#### Multiple Images
+
+To include multiple images in your prompt you simply add more `<image>` tokens to the prompt
+
+```python
+image_urls = [
+	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-3.png",  # triangle
+	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-11.png",  # hexagon
+]
+description_generator = outlines.generate.text(model)
+description_generator(
+    "<image><image><image>What shapes are present?",
+    list(map(img_from_url, image_urls)),
+)
+```
+
+> There are two shapes present. One shape is a hexagon and the other shape is an triangle. '
+
+
+### Classifying an Image
+
+```python
+pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
+planet_generator = outlines.generate.regex(model, pattern)
+
+planet_generator(
+    "What planet is this: <image>",
+    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg")]
+)
+```
+
+> Saturn
+
+
+### Extracting Structured Image data
+
+```python
+from pydantic import BaseModel
+from typing import List, Optional
+
+class ImageData(BaseModel):
+    caption: str
+    tags_list: List[str]
+    object_list: List[str]
+    is_photo: bool
+
+image_data_generator = outlines.generate.json(model, ImageData)
+
+image_data_generator(
+    "<image> detailed JSON metadata:",
+    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg")]
+)
+```
+
+> `ImageData(caption='An astronaut on the moon', tags_list=['moon', 'space', 'nasa', 'americanflag'], object_list=['moon', 'moon_surface', 'space_suit', 'americanflag'], is_photo=True)`
+
+
+## Resources
+
+### Chosing a model
+- https://mmbench.opencompass.org.cn/leaderboard
+- https://huggingface.co/spaces/WildVision/vision-arena
diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index 4104e3080..ad01377c0 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -1,6 +1,6 @@
 import datetime
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, Iterator, List, Optional, Union
+from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
 
 from outlines.generate.generator import sequence_generator
 from outlines.samplers import BeamSearchSampler, GreedySampler, MultinomialSampler
@@ -479,6 +479,13 @@ def format_sequence(self, sequence: str) -> FormattedOutput:
         """
         return sequence
 
+    def _format(self, sequences):
+        """Apply formatting to every string in a completion."""
+        if isinstance(sequences, list):
+            return [self._format(sequence) for sequence in sequences]
+        else:
+            return self.format_sequence(sequences)
+
     def __call__(
         self,
         prompts: Union[str, List[str]],
@@ -489,13 +496,6 @@ def __call__(
     ):
         """Generate text from a prompt of list of prompts."""
 
-        def format(sequences):
-            """Apply formatting to every string in a completion."""
-            if isinstance(sequences, list):
-                return [format(sequence) for sequence in sequences]
-            else:
-                return self.format_sequence(sequences)
-
         generation_params = self.prepare_generation_parameters(
             max_tokens, stop_at, seed
         )
@@ -508,7 +508,7 @@ def format(sequences):
             **model_specific_params,
         )
 
-        return format(completions)
+        return self._format(completions)
 
     def stream(
         self,
@@ -529,3 +529,94 @@ def stream(
             self.sampling_params,
             **model_specific_params,
         )
+
+
+class VisionSequenceGeneratorAdapter(SequenceGeneratorAdapter):
+    def __call__(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[str, Any],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        """
+        Generate text from a prompt of list of prompts.
+
+        Media: A URI to construct media or media object itself. Used as AutoProcessor argument.
+        """
+        prompts, media = self._validate_prompt_media_types(prompts, media)
+
+        generation_params = self.prepare_generation_parameters(
+            max_tokens, stop_at, seed
+        )
+
+        completions = self.model.generate(
+            prompts,
+            media,
+            generation_params,
+            self.logits_processor,
+            self.sampling_params,
+            **model_specific_params,
+        )
+
+        return self._format(completions)
+
+    def stream(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: List[Union[str, Any, List[Union[str, Any]]]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        """Return a text generator from a prompt or a list of prompts."""
+        prompts, media = self._validate_prompt_media_types(prompts, media)
+        generation_params = self.prepare_generation_parameters(
+            max_tokens, stop_at, seed
+        )
+        return self.model.stream(
+            prompts,
+            media,
+            generation_params,
+            self.logits_processor,
+            self.sampling_params,
+            **model_specific_params,
+        )
+
+    @classmethod
+    def _validate_prompt_media_types(
+        cls,
+        prompts: Union[str, List[str]],
+        media: Union[str, Any, List[Union[str, Any]]],
+    ) -> Union[Any, List[Any]]:
+        """
+        Prepare media as PIL.Image and ensure for every prompt str there is one List[PIL.Image]
+        """
+
+        def valid_types(prompts, media):
+            from PIL import Image  # type: ignore
+
+            if isinstance(prompts, list):
+                if not isinstance(media, list) or len(prompts) != len(media):
+                    return False
+                for subprompt, submedia in zip(prompts, media):
+                    if not isinstance(subprompt, str) or not all(
+                        isinstance(m, Image.Image) for m in submedia
+                    ):
+                        return False
+            elif isinstance(prompts, str):
+                if not all(isinstance(m, Image.Image) for m in media):
+                    return False
+            return True
+
+        if not valid_types(prompts, media):
+            raise TypeError(
+                "Expected (prompts, media) to be of type "
+                "(str, List[Image])), or (List[str], List[List[Image]]) "
+                f"instead got prompts={prompts}, media={media}"
+            )
+
+        return prompts, media
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
index 832a154bd..47661b47f 100644
--- a/outlines/generate/fsm.py
+++ b/outlines/generate/fsm.py
@@ -3,8 +3,12 @@
 import interegular
 
 from outlines.fsm.guide import RegexGuide
-from outlines.generate.api import SequenceGenerator, SequenceGeneratorAdapter
-from outlines.models import MLXLM, LlamaCpp, Transformers
+from outlines.generate.api import (
+    SequenceGenerator,
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import MLXLM, LlamaCpp, Transformers, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
@@ -29,3 +33,12 @@ def fsm_unified(
     fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
     logits_processor = FSMLogitsProcessor(tokenizer=model.tokenizer, fsm=fsm)
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@fsm.register(TransformersVision)
+def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()):
+    from outlines.processors import FSMLogitsProcessor
+
+    fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
+    logits_processor = FSMLogitsProcessor(tokenizer=model.tokenizer, fsm=fsm)
+    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
index 52b8c7dad..3aebcd429 100644
--- a/outlines/generate/regex.py
+++ b/outlines/generate/regex.py
@@ -1,12 +1,19 @@
 from functools import singledispatch
 
 from outlines.fsm.guide import RegexGuide
-from outlines.generate.api import SequenceGenerator, SequenceGeneratorAdapter
-from outlines.models import OpenAI
-from outlines.models.llamacpp import LlamaCpp
-from outlines.models.mlxlm import MLXLM
-from outlines.models.transformers import Transformers
-from outlines.models.vllm import VLLM
+from outlines.generate.api import (
+    SequenceGenerator,
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import (
+    MLXLM,
+    VLLM,
+    LlamaCpp,
+    OpenAI,
+    Transformers,
+    TransformersVision,
+)
 from outlines.samplers import Sampler, multinomial
 
 
@@ -53,6 +60,18 @@ def regex_unified(
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
+@regex.register(TransformersVision)
+def regex_vision(
+    model,
+    regex_str: str,
+    sampler: Sampler = multinomial(),
+):
+    from outlines.processors import RegexLogitsProcessor
+
+    logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
+    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
 @regex.register(VLLM)
 def regex_vllm(
     model: VLLM,
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
index 6da187e0b..b0b4e10c7 100644
--- a/outlines/generate/text.py
+++ b/outlines/generate/text.py
@@ -1,8 +1,19 @@
 from functools import singledispatch
 
 from outlines.fsm.guide import StopAtEOSGuide
-from outlines.generate.api import SequenceGenerator, SequenceGeneratorAdapter
-from outlines.models import MLXLM, VLLM, LlamaCpp, OpenAI, Transformers
+from outlines.generate.api import (
+    SequenceGenerator,
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import (
+    MLXLM,
+    VLLM,
+    LlamaCpp,
+    OpenAI,
+    Transformers,
+    TransformersVision,
+)
 from outlines.samplers import Sampler, multinomial
 
 
@@ -43,6 +54,11 @@ def text_unified(model, sampler: Sampler = multinomial()):
     return SequenceGeneratorAdapter(model, None, sampler)
 
 
+@text.register(TransformersVision)
+def text_vision(model, sampler: Sampler = multinomial()):
+    return VisionSequenceGeneratorAdapter(model, None, sampler)
+
+
 @text.register(VLLM)
 def text_vllm(model: VLLM, sampler: Sampler = multinomial()):
     return SequenceGeneratorAdapter(model, None, sampler)
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index c161215d1..d28fcb2d7 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -5,6 +5,7 @@
 codebase.
 
 """
+
 from typing import Union
 
 from .exllamav2 import ExLlamaV2Model, exl2
@@ -12,6 +13,7 @@
 from .mlxlm import MLXLM, mlxlm
 from .openai import OpenAI, azure_openai, openai
 from .transformers import Transformers, TransformerTokenizer, mamba, transformers
+from .transformers_vision import TransformersVision, transformers_vision
 from .vllm import VLLM, vllm
 
 LogitsGenerator = Union[Transformers, LlamaCpp, ExLlamaV2Model, MLXLM, VLLM]
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
new file mode 100644
index 000000000..876f9bff5
--- /dev/null
+++ b/outlines/models/transformers_vision.py
@@ -0,0 +1,139 @@
+from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
+
+from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.models import Transformers
+
+if TYPE_CHECKING:
+    from outlines.processors import OutlinesLogitsProcessor
+
+
+class TransformersVision(Transformers):
+    def __init__(self, model, tokenizer, processor):
+        super().__init__(model, tokenizer)
+        self.processor = processor
+
+    def generate(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[List[Any], List[List[Any]]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Union[str, List[str], List[List[str]]]:
+        """Generate text using `transformers`.
+
+        Arguments
+        ---------
+        prompts
+            A prompt or list of prompts.
+        media
+            A List[PIL.Image] or List[List[PIL.Image]]
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
+
+        Returns
+        -------
+        The generated text
+        """
+        inputs = self.processor(prompts, media, padding=True, return_tensors="pt").to(
+            self.model.device
+        )
+
+        generation_kwargs = self._get_generation_kwargs(
+            prompts,
+            generation_parameters,
+            logits_processor,
+            sampling_parameters,
+        )
+        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
+
+        # if single str input and single sample per input, convert to a 1D output
+        if isinstance(prompts, str):
+            # Should always be true until NotImplementedError above is fixed
+            generated_ids = generated_ids.squeeze(0)
+
+        return self._decode_generation(generated_ids)
+
+    def stream(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[Any, List[Any]],  # TODO: docstring
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Iterator[Union[str, List[str]]]:
+        raise NotImplementedError
+
+
+def transformers_vision(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    processor_kwargs: dict = {},
+    model_class=None,
+    tokenizer_class=None,
+    processor_class=None,
+):
+    """Instantiate a model from the `transformers` library and its tokenizer.
+
+    Parameters
+    ----------
+    model_name
+        The name of the model as listed on Hugging Face's model page.
+    device
+        The device(s) on which the model should be loaded. This overrides
+        the `device_map` entry in `model_kwargs` when provided.
+    model_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the model.
+    processor_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the processor.
+
+    Returns
+    -------
+    A `TransformersModel` model instance.
+
+    """
+    if model_class is None or tokenizer_class is None:
+        try:
+            from transformers import (
+                AutoTokenizer,
+                LlavaNextForConditionalGeneration,
+                LlavaNextProcessor,
+            )
+        except ImportError:
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if model_class is None:
+        model_class = LlavaNextForConditionalGeneration
+    if processor_class is None:
+        processor_class = LlavaNextProcessor
+
+    if device is not None:
+        model_kwargs["device_map"] = device
+
+    model = model_class.from_pretrained(model_name, **model_kwargs)
+
+    processor_kwargs.setdefault("padding_side", "left")
+    processor_kwargs.setdefault("pad_token", "[PAD]")
+    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+    if tokenizer_class is None:
+        if getattr(processor, "tokenizer", None):
+            tokenizer = processor.tokenizer
+        else:
+            tokenizer = AutoTokenizer.from_pretrained(model_name, **processor_kwargs)
+    else:
+        tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
+
+    return TransformersVision(model, tokenizer, processor)
diff --git a/pyproject.toml b/pyproject.toml
index aa88fcbbc..f94b3c84d 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -62,6 +62,7 @@ test = [
     "vllm; sys_platform != 'darwin'",
     "torch",
     "transformers",
+    "pillow",
 ]
 serve = [
     "vllm>=0.3.0",
diff --git a/tests/generate/test_api.py b/tests/generate/test_api.py
new file mode 100644
index 000000000..7188022f5
--- /dev/null
+++ b/tests/generate/test_api.py
@@ -0,0 +1,33 @@
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image  # type: ignore
+
+from outlines.generate.api import VisionSequenceGeneratorAdapter
+
+IMG_URI = "https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg"
+PIL_IMG = Image.open(BytesIO(urlopen(IMG_URI).read())).convert("RGB")
+
+
+@pytest.mark.parametrize(
+    "prompts,media,type_error",
+    [
+        ("single prompt", [PIL_IMG], False),
+        (["prompt0", "prompt1"], [[PIL_IMG], [PIL_IMG]], False),
+        ("single prompt", [PIL_IMG, PIL_IMG], False),
+        (["prompt0", "prompt1"], [[PIL_IMG, PIL_IMG], [PIL_IMG]], False),
+        ("single prompt", "this isn't an image, it's a string", True),
+        ("single prompt", PIL_IMG, True),
+        (["prompt0", "prompt1"], [PIL_IMG], True),
+        (["prompt0", "prompt1"], [[PIL_IMG]], True),
+        (["prompt0", "prompt1"], [[[PIL_IMG]], [[PIL_IMG]]], True),
+    ],
+)
+def test_vision_sequence_generator_validate_types(prompts, media, type_error):
+    """Ensure inputs are validated correctly"""
+    if type_error:
+        with pytest.raises(TypeError):
+            VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
+    else:
+        VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 06d311f5b..b1202f396 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -7,6 +7,10 @@
 import outlines.models as models
 import outlines.samplers as samplers
 
+##########################################
+# Model Fixtures
+##########################################
+
 
 @pytest.fixture(scope="session")
 def model_llamacpp(tmp_path_factory):
@@ -50,6 +54,17 @@ def model_bart(tmp_path_factory):
     )
 
 
+@pytest.fixture(scope="session")
+def model_transformers_vision(tmp_path_factory):
+    import torch
+
+    return models.transformers_vision(
+        "llava-hf/llava-v1.6-mistral-7b-hf",
+        device="cuda",
+        model_kwargs=dict(torch_dtype=torch.bfloat16),
+    )
+
+
 # TODO: exllamav2 failing in main, address in https://github.com/outlines-dev/outlines/issues/808
 # TODO: t5 tokenizer doesn't work with streaming
 """
@@ -78,15 +93,42 @@ def model_t5(tmp_path_factory):
     "model_transformers_opt125m",
     "model_mamba",
     "model_bart",
+    # "model_transformers_vision",  # tests pass, but awaiting a tiny model for CI
 )
 
 
-NOT_IMPLEMENTED = {
-    "stream": [],
-    "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-    "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-    "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-}
+##########################################
+# Stuctured Generation Inputs
+##########################################
+
+
+@pytest.fixture()
+def sample_schema():
+    from pydantic import BaseModel, conint, conlist, constr
+
+    class SampleSchema(BaseModel):
+        title: constr(max_length=10)
+        numbers: conlist(conint(strict=True), min_length=3, max_length=3)
+        labels: conlist(constr(min_length=1, max_length=5), min_length=3, max_length=3)
+
+    return SampleSchema
+
+
+@pytest.fixture()
+def sample_choices():
+    return ["foo", "bar", "baz"]
+
+
+REGEX_PATTERNS = [
+    "a b c d e",  # ensure proper tokenizer whitespace prefix handling
+    "(123456789)|(abcdefghijklmnop)",  # ensure consistent correct sequence handling during batch
+    r"([a-z]{10})@([a-z]{5})\.([a-z]{3})",  # email example
+]
+
+
+###########################################
+# Model/Generator Pair Behavior Definitions
+###########################################
 
 
 def enforce_not_implemented(model_fixture, *task_names):
@@ -94,6 +136,12 @@ def enforce_not_implemented(model_fixture, *task_names):
     Per `NOT_IMPLEMENTED`, mapping, if a model hasn't implemented a task,
     assert an NotImplementedError is raised. Otherwise, run normally
     """
+    NOT_IMPLEMENTED = {
+        "stream": ["model_transformers_vision"],
+        "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
+        "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
+        "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
+    }
     for task_name in task_names:
         if model_fixture in NOT_IMPLEMENTED.get(task_name, []):
             return pytest.raises(NotImplementedError)
@@ -101,11 +149,35 @@ def enforce_not_implemented(model_fixture, *task_names):
         return contextlib.nullcontext()
 
 
-REGEX_PATTERNS = [
-    "a b c d e",  # ensure proper tokenizer whitespace prefix handling
-    "(123456789)|(abcdefghijklmnop)",  # ensure consistent correct sequence handling during batch
-    r"([a-z]{10})@([a-z]{5})\.([a-z]{3})",  # email example
-]
+def get_inputs(fixture_name, batch_size=None):
+    """Get generator kwargs, just the prompt by default, but include images for transformers_visian"""
+    from io import BytesIO
+    from urllib.request import urlopen
+
+    from PIL import Image  # type: ignore
+
+    prompts = ["abcd", "efgh", "1234", "5678", "foo", "bar", "baz", "bif"]
+    prompts = prompts[0] if batch_size is None else prompts[:batch_size]
+
+    if fixture_name.endswith("_vision"):
+        img_url = "https://python-pillow.org/pillow-perf/static/space_pil_lanczos.png"
+        img = Image.open(BytesIO(urlopen(img_url).read())).convert("RGB")
+
+        if batch_size is None:
+            return {"prompts": f"<image> {prompts}", "media": [img]}
+        else:
+            return {
+                "prompts": [f"<image> {p}" for p in prompts],
+                "media": [[img] for _ in range(batch_size)],
+            }
+
+    else:
+        return {"prompts": prompts}
+
+
+###########################################
+# Tests
+###########################################
 
 
 @pytest.mark.parametrize("sampler_name", ("greedy", "multinomial", "beam_search"))
@@ -114,27 +186,17 @@ def test_generate_text(request, model_fixture, sampler_name):
     model = request.getfixturevalue(model_fixture)
     generator = generate.text(model, getattr(samplers, sampler_name)())
     with enforce_not_implemented(model_fixture, sampler_name):
-        res = generator("test", max_tokens=10)
+        res = generator(**get_inputs(model_fixture), max_tokens=10)
         assert isinstance(res, str)
 
 
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_batch_text(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model)
-    with enforce_not_implemented(model_fixture, "batch"):
-        res = generator(["test", "test2"], max_tokens=10)
-        assert isinstance(res, list)
-        assert isinstance(res[0], str)
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_text_stream(request, model_fixture):
+def test_generate_regex(request, model_fixture, pattern):
     model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model)
-    with enforce_not_implemented(model_fixture, "stream"):
-        for token in generator.stream("a b c ", max_tokens=10):
-            assert isinstance(token, str)
+    generator = generate.regex(model, pattern)
+    res = generator(**get_inputs(model_fixture), max_tokens=20)
+    assert re.fullmatch(pattern, res) is not None, res
 
 
 @pytest.mark.parametrize("pattern", REGEX_PATTERNS)
@@ -144,17 +206,44 @@ def test_generate_fsm(request, model_fixture, pattern):
 
     model = request.getfixturevalue(model_fixture)
     generator = generate.fsm(model, interegular.parse_pattern(pattern).to_fsm())
-    res = generator("test")
+    res = generator(**get_inputs(model_fixture))
     assert re.fullmatch(pattern, res) is not None, res
 
 
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.skip(
+    "Fix issues with JSON, some models fail this test https://github.com/outlines-dev/outlines/issues/985"
+)
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex(request, model_fixture, pattern):
+def test_generate_json(request, model_fixture, sample_schema):
     model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    res = generator("foobarbaz", max_tokens=20)
-    assert re.fullmatch(pattern, res) is not None, res
+    generator = generate.json(model, sample_schema)
+    # asserts valid within call
+    generator(**get_inputs(model_fixture), max_tokens=100)
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_choice(request, model_fixture, sample_choices):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.choice(model, sample_choices)
+    res = generator(**get_inputs(model_fixture))
+    assert res in sample_choices
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_format_bool(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.format(model, bool)
+    res = generator(**get_inputs(model_fixture))
+    assert isinstance(res, bool)
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_text_stream(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.text(model)
+    with enforce_not_implemented(model_fixture, "stream"):
+        for token in generator.stream(**get_inputs(model_fixture), max_tokens=10):
+            assert isinstance(token, str)
 
 
 @pytest.mark.parametrize("pattern", REGEX_PATTERNS)
@@ -164,23 +253,19 @@ def test_generate_regex_stream(request, model_fixture, pattern):
     generator = generate.regex(model, pattern)
     with enforce_not_implemented(model_fixture, "stream"):
         output = ""
-        for token in generator.stream("output:", max_tokens=20):
+        for token in generator.stream(**get_inputs(model_fixture), max_tokens=20):
             output += token
         assert re.fullmatch(pattern, output) is not None, output
 
 
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_batch_stream(request, model_fixture, pattern):
+def test_generate_batch_text(request, model_fixture):
     model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    with enforce_not_implemented(model_fixture, "batch", "stream"):
-        outputs = ["", ""]
-        for tokens in generator.stream(["input 0", "input 1"], max_tokens=20):
-            outputs[0] += tokens[0]
-            outputs[1] += tokens[1]
-        for output in outputs:
-            assert re.fullmatch(pattern, output) is not None, output
+    generator = generate.text(model)
+    with enforce_not_implemented(model_fixture, "batch"):
+        res = generator(**get_inputs(model_fixture, 2), max_tokens=10)
+        assert isinstance(res, list)
+        assert isinstance(res[0], str)
 
 
 @pytest.mark.parametrize("pattern", REGEX_PATTERNS)
@@ -190,44 +275,50 @@ def test_generate_regex_batch(request, model_fixture, pattern):
     model = request.getfixturevalue(model_fixture)
     generator = generate.regex(model, pattern)
     with enforce_not_implemented(model_fixture, "batch"):
-        outputs = generator(["abc", "123", "123bce", "33aa"], max_tokens=20)
+        outputs = generator(**get_inputs(model_fixture, 4), max_tokens=20)
         for output in outputs:
             assert re.fullmatch(pattern, output) is not None, output
 
 
 @pytest.mark.parametrize("pattern", REGEX_PATTERNS)
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_single_multinomial(request, model_fixture, pattern):
-    """Ensure batch requests work and fsm order is maintained"""
+def test_generate_regex_batch_stream(request, model_fixture, pattern):
     model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern, sampler=samplers.multinomial(4))
-    with enforce_not_implemented(model_fixture, "multiple_samples"):
-        output_sample_groups = generator("single input", max_tokens=40)
-        for output in output_sample_groups:
+    generator = generate.regex(model, pattern)
+    with enforce_not_implemented(model_fixture, "batch", "stream"):
+        outputs = ["", ""]
+        for tokens in generator.stream(**get_inputs(model_fixture, 2), max_tokens=20):
+            outputs[0] += tokens[0]
+            outputs[1] += tokens[1]
+        for output in outputs:
             assert re.fullmatch(pattern, output) is not None, output
 
 
 @pytest.mark.parametrize("pattern", REGEX_PATTERNS)
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_batch_multinomial(request, model_fixture, pattern):
+def test_generate_regex_single_multinomial(request, model_fixture, pattern):
     """Ensure batch requests work and fsm order is maintained"""
     model = request.getfixturevalue(model_fixture)
     generator = generate.regex(model, pattern, sampler=samplers.multinomial(4))
-    with enforce_not_implemented(model_fixture, "batch", "multiple_samples"):
-        output_batch_groups = generator(["abc", "123", "123bce", "33aa"], max_tokens=40)
-        for output_sample_groups in output_batch_groups:
-            for output in output_sample_groups:
-                assert re.fullmatch(pattern, output) is not None, output
+    with enforce_not_implemented(model_fixture, "multiple_samples"):
+        output_sample_groups = generator(**get_inputs(model_fixture), max_tokens=40)
+        for output in output_sample_groups:
+            assert re.fullmatch(pattern, output) is not None, output
 
 
 @pytest.mark.parametrize("pattern", REGEX_PATTERNS)
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_batch_beam_search(request, model_fixture, pattern):
+@pytest.mark.parametrize("sampler_name", ("multinomial", "beam_search"))
+def test_generate_regex_batch_multi_sample(
+    request, model_fixture, pattern, sampler_name
+):
     """Ensure batch requests work and fsm order is maintained"""
     model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern, sampler=samplers.beam_search(4))
+    generator = generate.regex(
+        model, pattern, sampler=getattr(samplers, sampler_name)(4)
+    )
     with enforce_not_implemented(model_fixture, "batch", "multiple_samples"):
-        output_batch_groups = generator(["abc", "123", "123bce", "33aa"], max_tokens=40)
+        output_batch_groups = generator(**get_inputs(model_fixture, 4), max_tokens=40)
         for output_sample_groups in output_batch_groups:
             for output in output_sample_groups:
                 assert re.fullmatch(pattern, output) is not None, output

From 47dfa4bd530a333bce0978f930cc9eb006a1a589 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 20 Jul 2024 03:44:38 -0500
Subject: [PATCH 022/505] Use outlines.processors for vLLM

---
 outlines/generate/choice.py                  |  4 +-
 outlines/generate/format.py                  |  6 ++-
 outlines/generate/fsm.py                     | 24 +++++------
 outlines/generate/json.py                    |  4 +-
 outlines/generate/regex.py                   | 43 +++++---------------
 outlines/generate/text.py                    | 34 ++++------------
 outlines/models/vllm.py                      | 22 ++++++++++
 outlines/processors/base_logits_processor.py | 20 ++++-----
 outlines/processors/structured.py            | 27 +++++-------
 tests/generate/conftest.py                   | 26 +++++++++++-
 tests/generate/test_generate.py              | 16 ++++++--
 11 files changed, 119 insertions(+), 107 deletions(-)

diff --git a/outlines/generate/choice.py b/outlines/generate/choice.py
index 6718f26b2..931409a59 100644
--- a/outlines/generate/choice.py
+++ b/outlines/generate/choice.py
@@ -1,7 +1,7 @@
 from functools import singledispatch
 from typing import Callable, List
 
-from outlines.generate.api import SequenceGenerator
+from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
 
@@ -11,7 +11,7 @@
 @singledispatch
 def choice(
     model, choices: List[str], sampler: Sampler = multinomial()
-) -> SequenceGenerator:
+) -> SequenceGeneratorAdapter:
     regex_str = r"(" + r"|".join(choices) + r")"
 
     generator = regex(model, regex_str, sampler)
diff --git a/outlines/generate/format.py b/outlines/generate/format.py
index d7b7e164d..88acec75f 100644
--- a/outlines/generate/format.py
+++ b/outlines/generate/format.py
@@ -1,7 +1,7 @@
 from functools import singledispatch
 
 from outlines.fsm.types import python_types_to_regex
-from outlines.generate.api import SequenceGenerator
+from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
 
@@ -9,7 +9,9 @@
 
 
 @singledispatch
-def format(model, python_type, sampler: Sampler = multinomial()) -> SequenceGenerator:
+def format(
+    model, python_type, sampler: Sampler = multinomial()
+) -> SequenceGeneratorAdapter:
     """Generate structured data that can be parsed as a Python type.
 
     Parameters
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
index 47661b47f..03fe512b9 100644
--- a/outlines/generate/fsm.py
+++ b/outlines/generate/fsm.py
@@ -8,25 +8,13 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import MLXLM, LlamaCpp, Transformers, TransformersVision
+from outlines.models import ExLlamaV2Model, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
 @singledispatch
 def fsm(
     model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
-) -> SequenceGenerator:
-    fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    device = model.device
-    generator = SequenceGenerator(fsm, model, sampler, device)
-    return generator
-
-
-@fsm.register(MLXLM)
-@fsm.register(Transformers)
-@fsm.register(LlamaCpp)
-def fsm_unified(
-    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
 ) -> SequenceGeneratorAdapter:
     from outlines.processors import FSMLogitsProcessor
 
@@ -42,3 +30,13 @@ def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
     fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
     logits_processor = FSMLogitsProcessor(tokenizer=model.tokenizer, fsm=fsm)
     return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@fsm.register(ExLlamaV2Model)
+def fsm_exllamav2(
+    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
+) -> SequenceGenerator:
+    fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
+    device = model.device
+    generator = SequenceGenerator(fsm, model, sampler, device)
+    return generator
diff --git a/outlines/generate/json.py b/outlines/generate/json.py
index 3837f72b6..6209840e2 100644
--- a/outlines/generate/json.py
+++ b/outlines/generate/json.py
@@ -5,7 +5,7 @@
 from pydantic import BaseModel
 
 from outlines.fsm.json_schema import build_regex_from_schema, get_schema_from_signature
-from outlines.generate.api import SequenceGenerator
+from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
 
@@ -18,7 +18,7 @@ def json(
     schema_object: Union[str, object, Callable],
     sampler: Sampler = multinomial(),
     whitespace_pattern: Optional[str] = None,
-) -> SequenceGenerator:
+) -> SequenceGeneratorAdapter:
     """
     Generate structured JSON data with a `Transformer` model based on a specified JSON Schema.
 
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
index 3aebcd429..815a8b1b9 100644
--- a/outlines/generate/regex.py
+++ b/outlines/generate/regex.py
@@ -6,14 +6,7 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import (
-    MLXLM,
-    VLLM,
-    LlamaCpp,
-    OpenAI,
-    Transformers,
-    TransformersVision,
-)
+from outlines.models import ExLlamaV2Model, OpenAI, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
@@ -34,26 +27,10 @@ def regex(model, regex_str: str, sampler: Sampler = multinomial()):
 
     Returns
     -------
-    A `SequenceGenerator` instance that generates text constrained by the
+    A `SequenceGeneratorAdapter` instance that generates text constrained by the
     regular expression.
 
     """
-    fsm = RegexGuide(regex_str, model.tokenizer)
-
-    device = model.device
-    generator = SequenceGenerator(fsm, model, sampler, device)
-
-    return generator
-
-
-@regex.register(MLXLM)
-@regex.register(Transformers)
-@regex.register(LlamaCpp)
-def regex_unified(
-    model,
-    regex_str: str,
-    sampler: Sampler = multinomial(),
-):
     from outlines.processors import RegexLogitsProcessor
 
     logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
@@ -72,16 +49,18 @@ def regex_vision(
     return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
-@regex.register(VLLM)
-def regex_vllm(
-    model: VLLM,
+@regex.register(ExLlamaV2Model)
+def regex_exllamav2(
+    model,
     regex_str: str,
     sampler: Sampler = multinomial(),
-):
-    from outlines.integrations.vllm import RegexLogitsProcessor
+) -> SequenceGenerator:
+    fsm = RegexGuide(regex_str, model.tokenizer)
 
-    logits_processor = RegexLogitsProcessor(regex_str, model.model)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
+    device = model.device
+    generator = SequenceGenerator(fsm, model, sampler, device)
+
+    return generator
 
 
 @regex.register(OpenAI)
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
index b0b4e10c7..3fe3dc553 100644
--- a/outlines/generate/text.py
+++ b/outlines/generate/text.py
@@ -6,19 +6,12 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import (
-    MLXLM,
-    VLLM,
-    LlamaCpp,
-    OpenAI,
-    Transformers,
-    TransformersVision,
-)
+from outlines.models import ExLlamaV2Model, OpenAI, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
 @singledispatch
-def text(model, sampler: Sampler = multinomial()) -> SequenceGenerator:
+def text(model, sampler: Sampler = multinomial()) -> SequenceGeneratorAdapter:
     """Generate text with a `Transformer` model.
 
     Note
@@ -37,21 +30,17 @@ def text(model, sampler: Sampler = multinomial()) -> SequenceGenerator:
 
     Returns
     -------
-    A `SequenceGenerator` instance that generates text.
+    A `SequenceGeneratorAdapter` instance that generates text.
 
     """
-    fsm = StopAtEOSGuide(model.tokenizer)
-    device = model.device
-    generator = SequenceGenerator(fsm, model, sampler, device)
-
-    return generator
+    return SequenceGeneratorAdapter(model, None, sampler)
 
 
-@text.register(MLXLM)
-@text.register(Transformers)
-@text.register(LlamaCpp)
-def text_unified(model, sampler: Sampler = multinomial()):
-    return SequenceGeneratorAdapter(model, None, sampler)
+@text.register(ExLlamaV2Model)
+def text_exllamav2(model, sampler: Sampler = multinomial()) -> SequenceGenerator:
+    fsm = StopAtEOSGuide(model.tokenizer)
+    device = model.device
+    return SequenceGenerator(fsm, model, sampler, device)
 
 
 @text.register(TransformersVision)
@@ -59,11 +48,6 @@ def text_vision(model, sampler: Sampler = multinomial()):
     return VisionSequenceGeneratorAdapter(model, None, sampler)
 
 
-@text.register(VLLM)
-def text_vllm(model: VLLM, sampler: Sampler = multinomial()):
-    return SequenceGeneratorAdapter(model, None, sampler)
-
-
 @text.register(OpenAI)
 def text_openai(model: OpenAI, sampler: Sampler = multinomial()) -> OpenAI:
     if not isinstance(sampler, multinomial):
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 7c1f5ad7c..2ae3d99d4 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -2,6 +2,7 @@
 from typing import TYPE_CHECKING, List, Optional, Union
 
 from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.integrations.utils import adapt_tokenizer
 
 if TYPE_CHECKING:
     from vllm import LLM
@@ -22,6 +23,23 @@ def __init__(self, model: "LLM"):
         self.model = model
         self.lora_request = None
 
+        self.tokenizer = self._get_tokenizer()
+
+    def _get_tokenizer(self):
+        if hasattr(self.model, "get_tokenizer"):
+            tokenizer = self.model.get_tokenizer()
+        elif hasattr(self.model, "tokenizer"):
+            if hasattr(self.model.tokenizer, "tokenizer"):
+                tokenizer = self.model.tokenizer.tokenizer
+            else:
+                tokenizer = self.model.tokenizer
+        else:
+            raise ValueError(
+                "The provided LLM instance neither has a "
+                "`tokenizer` attribute or a `get_tokenizer` method."
+            )
+        return adapt_tokenizer(tokenizer=tokenizer)
+
     def generate(
         self,
         prompts: Union[str, List[str]],
@@ -100,6 +118,10 @@ def generate(
             sampling_params.top_p = top_p
         if top_k is not None and sampling_params.top_k == -1:
             sampling_params.top_k = top_k
+            # TODO: remove this if statement once fixed
+            # https://github.com/vllm-project/vllm/issues/5404#issuecomment-2175972897
+            if top_k == 1:
+                sampling_params.repetition_penalty = 0
         if temperature is not None and sampling_params.temperature == 1.0:
             sampling_params.temperature = temperature
         if sampler == "beam_search":
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index 1d141a2f0..feedf5253 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -67,22 +67,18 @@ def __call__(
         3) Call self.process_logits() to perform business logic
         4) Cast logits back to original array library type
         """
-
         # ensure logits are torch Tensors
         torch_logits = self._to_torch(logits)
+        input_ids = self._to_torch(input_ids)
 
-        assert torch_logits.shape[:-1] == self._to_torch(input_ids).shape[:-1]
-
-        # ensure input_ids are List
-        if not isinstance(input_ids, list):
-            input_ids = input_ids.tolist()  # compatible with numpy, torch, and mlx
+        assert torch_logits.shape[:-1] == input_ids.shape[:-1]
 
         # Guarantee passed as 2D Tensors, then covert back to original (1D or 2D) shape
         if len(torch_logits.shape) == 2:
-            processed_logits = self.process_logits(input_ids, torch_logits)
+            processed_logits = self.process_logits(input_ids.tolist(), torch_logits)
         elif len(torch_logits.shape) == 1:
             processed_logits = self.process_logits(
-                [input_ids], torch_logits.unsqueeze(0)
+                [input_ids.tolist()], torch_logits.unsqueeze(0)
             ).squeeze(0)
 
         # return logits as passed array type
@@ -97,7 +93,7 @@ def _to_torch(tensor_like: Array) -> torch.Tensor:
         elif isinstance(tensor_like, np.ndarray):
             return torch.from_numpy(tensor_like)
 
-        elif isinstance(tensor_like, list):
+        elif isinstance(tensor_like, (list, tuple)):
             return torch.tensor(tensor_like)
 
         elif is_mlx_array_type(type(tensor_like)):
@@ -108,7 +104,8 @@ def _to_torch(tensor_like: Array) -> torch.Tensor:
         else:
             raise TypeError(
                 "LogitsProcessor must be called with either np.NDArray, "
-                "torch.Tensor, list, or mlx.core.array typed logits"
+                "torch.Tensor, list, or mlx.core.array typed logits. "
+                f"Logits type: `{type(tensor_like)}`"
             )
 
     @staticmethod
@@ -123,6 +120,9 @@ def _from_torch(tensor: torch.Tensor, target_type: Type) -> Array:
         elif target_type == list:
             return tensor.detach().tolist()
 
+        elif target_type == tuple:
+            return tuple(tensor.detach().tolist())
+
         elif is_mlx_array_type(target_type):
             import mlx.core as mx
 
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index d037c679f..0966a90db 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -61,9 +61,8 @@ def __init__(self, tokenizer: "Tokenizer", fsm: Guide):
             The finite state machine which is used to bias the logits.
         """
         self.tokenizer = tokenizer
-        self._fsm_states: Dict[int, int] = {}
+        self._fsm_states: Dict[int, int] = {hash(tuple([])): 0}
         self.fsm: Guide = fsm
-        self._is_first_token = True
         self._seq_start_idx: Optional[int] = None
 
     def process_logits(
@@ -83,25 +82,21 @@ def process_logits(
         torch.Tensor
             The biased logits.
         """
-        sequence_states: List[int] = []  # vector of states corresponding to `input_ids`
-
-        if self._is_first_token:
-            self._is_first_token = False
+        if self._seq_start_idx is None:
             self._seq_start_idx = len(input_ids[0])
 
-            self._fsm_states = {hash(tuple([])): 0}
-            sequence_states = [0] * len(input_ids)
-
-        else:
-            for seq_ids in input_ids:
-                prev_state_key = hash(tuple(seq_ids[self._seq_start_idx : -1]))
-                prev_state = self._fsm_states[prev_state_key]
+        sequence_states: List[int] = []  # vector of states corresponding to `input_ids`
 
-                curr_state_key = hash(tuple(seq_ids[self._seq_start_idx :]))
-                curr_state = self.fsm.get_next_state(prev_state, seq_ids[-1])
+        for seq_ids in input_ids:
+            gen_ids = seq_ids[self._seq_start_idx :]
+            curr_state_key = hash(tuple(gen_ids))
 
+            if curr_state_key not in self._fsm_states:
+                prev_state = self._fsm_states[hash(tuple(gen_ids[:-1]))]
+                curr_state = self.fsm.get_next_state(prev_state, gen_ids[-1])
                 self._fsm_states[curr_state_key] = curr_state
-                sequence_states.append(curr_state)
+
+            sequence_states.append(self._fsm_states[curr_state_key])
 
         mask = torch.full_like(logits, -math.inf)
         for i, fsm_state in enumerate(sequence_states):
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
index fe6d1e1a7..ed8830119 100644
--- a/tests/generate/conftest.py
+++ b/tests/generate/conftest.py
@@ -1,16 +1,38 @@
 from importlib import reload
 
 import pytest
+import torch
 
 
-def pytest_collection_modifyitems(config, items):
-    """If mlxlm and Metal aren't available, skip mlxlm tests"""
+def is_metal_available():
     try:
         import mlx.core as mx
         import mlx_lm  # noqa: F401
 
         assert mx.metal.is_available()
     except (ImportError, AssertionError):
+        return False
+    return True
+
+
+def pytest_collection_modifyitems(config, items):
+    """
+    If mlxlm and Metal aren't available, skip mlxlm tests
+    If CUDA isn't available, skip vllm and transformers_vision
+    """
+    if not torch.cuda.is_available():
+        skip_marker = pytest.mark.skip(
+            reason="Skipping test because CUDA is not available"
+        )
+        for item in items:
+            if "model_fixture" in item.fixturenames:
+                model_param = item.callspec.params.get("model_fixture", None)
+                if model_param.startswith(
+                    "model_transformers_vision"
+                ) or model_param.startswith("model_vllm"):
+                    item.add_marker(skip_marker)
+
+    if not is_metal_available():
         skip_marker = pytest.mark.skip(
             reason="Skipping test because mlx-lm or Metal are not available"
         )
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index b1202f396..a86d3c253 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -61,10 +61,19 @@ def model_transformers_vision(tmp_path_factory):
     return models.transformers_vision(
         "llava-hf/llava-v1.6-mistral-7b-hf",
         device="cuda",
-        model_kwargs=dict(torch_dtype=torch.bfloat16),
+        model_kwargs=dict(
+            torch_dtype=torch.bfloat16,
+            load_in_4bit=True,
+            low_cpu_mem_usage=True,
+        ),
     )
 
 
+@pytest.fixture(scope="session")
+def model_vllm(tmp_path_factory):
+    return models.vllm("facebook/opt-125m", gpu_memory_utilization=0.1)
+
+
 # TODO: exllamav2 failing in main, address in https://github.com/outlines-dev/outlines/issues/808
 # TODO: t5 tokenizer doesn't work with streaming
 """
@@ -93,7 +102,8 @@ def model_t5(tmp_path_factory):
     "model_transformers_opt125m",
     "model_mamba",
     "model_bart",
-    # "model_transformers_vision",  # tests pass, but awaiting a tiny model for CI
+    "model_transformers_vision",
+    "model_vllm",
 )
 
 
@@ -137,7 +147,7 @@ def enforce_not_implemented(model_fixture, *task_names):
     assert an NotImplementedError is raised. Otherwise, run normally
     """
     NOT_IMPLEMENTED = {
-        "stream": ["model_transformers_vision"],
+        "stream": ["model_transformers_vision", "model_vllm"],
         "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
         "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
         "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],

From 42c55bc0db821254cfac76a1e540d80e2642746e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?C=C3=A9lestin=20Matte?= <tohwiq@gmail.com>
Date: Sat, 20 Jul 2024 18:14:48 +0200
Subject: [PATCH 023/505] Fix deprecated memory= parameter (#1057)

memory= parameter is deprecated in favor of size=
See https://modal.com/docs/reference/changelog#062174-2024-05-17

Current doc example produces the following error:
```
/path/test_modal.py:56: DeprecationError: 2024-05-16: The `memory` parameter is deprecated. Use the `size='80GB'` parameter instead.
  @app.function(image=outlines_image, gpu=gpu.A100(memory=80))
```
---
 docs/cookbook/deploy-using-modal.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/cookbook/deploy-using-modal.md b/docs/cookbook/deploy-using-modal.md
index 8da39a3f5..d4058bbee 100644
--- a/docs/cookbook/deploy-using-modal.md
+++ b/docs/cookbook/deploy-using-modal.md
@@ -73,7 +73,7 @@ schema = """{
 To make the inference work on Modal we need to wrap the corresponding function in a `@app.function` decorator. We pass to this decorator the image and GPU on which we want this function to run (here an A100 with 80GB memory):
 
 ```python
-@app.function(image=outlines_image, gpu=gpu.A100(memory=80))
+@app.function(image=outlines_image, gpu=gpu.A100(size='80GB'))
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):

From abe689c1a2723ee797974e32e545dadabb872734 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?C=C3=A9lestin=20Matte?= <dev@cmatte.me>
Date: Sat, 20 Jul 2024 18:49:39 +0200
Subject: [PATCH 024/505] Fix failing Modal example (#1058)

It seems modal deletes environment variables, which makes
outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
fail even after login. This workaround instructs user to manually add
a key before importing the model.

Fixes #1024
---
 docs/cookbook/deploy-using-modal.md | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/docs/cookbook/deploy-using-modal.md b/docs/cookbook/deploy-using-modal.md
index d4058bbee..835924d2b 100644
--- a/docs/cookbook/deploy-using-modal.md
+++ b/docs/cookbook/deploy-using-modal.md
@@ -6,7 +6,7 @@ In this guide we will show you how you can use Modal to run programs written wit
 
 ## Build the image
 
-First we need to define our container image. We download the Mistral-7B-v0.1 model from HuggingFace as part of the definition of the image so it only needs to be done once.
+First we need to define our container image. We download the Mistral-7B-v0.1 model from HuggingFace as part of the definition of the image so it only needs to be done once (you need to provide an [access token](https://huggingface.co/settings/tokens))
 
 ```python
 from modal import Image, App, gpu
@@ -21,6 +21,8 @@ outlines_image = Image.debian_slim(python_version="3.11").pip_install(
 )
 
 def import_model():
+    import os
+    os.environ["HF_TOKEN"] = "YOUR_HUGGINGFACE_TOKEN"
     import outlines
     outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
 

From fcfef33c84f519af5aa24b1ca52c9908fd97d101 Mon Sep 17 00:00:00 2001
From: Alonso Silva Allende <alonso.silva@gmail.com>
Date: Mon, 22 Jul 2024 17:39:21 +0200
Subject: [PATCH 025/505] Add links to the two examples (#1062)

Add links to the two examples:
- Q&A with Citations
- Knowledge Graph Generation
---
 docs/cookbook/index.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index 5e84cabab..24849d5ee 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -6,3 +6,5 @@
 - [Chain Of Density](chain_of_density.md): Summarize documents using chain of density prompting and JSON-structured generation.
 - [Playing Chess](models_playing_chess.md): Make Mistral-7B play chess against itself using regex-structured generation.
 - [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
+- [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
+- [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.

From d78041e81f355ad9f83174cc8ac879443df95e76 Mon Sep 17 00:00:00 2001
From: Franz Louis Cesista <franzlouiscesista@gmail.com>
Date: Tue, 23 Jul 2024 15:48:05 +0800
Subject: [PATCH 026/505] add link to MMSG

---
 docs/community/examples.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/docs/community/examples.md b/docs/community/examples.md
index e606ca019..2ebaf2765 100644
--- a/docs/community/examples.md
+++ b/docs/community/examples.md
@@ -2,6 +2,8 @@
 
 Publishing examples and articles about Outlines are a meaningful way to contrinute to the community. Here is a list of projects we are aware of. Drop us a line if we forgot yours!
 
+[MMSG](https://github.com/leloykun/mmsg) is a Python library for generating interleaved text and image content in a structured format you can directly pass to downstream APIs.
+
 [Multimodal Structured Generation: CVPR's 2nd MMFM Challenge Technical Report](https://arxiv.org/abs/2406.11403) shows that Structured Generation can outperform finetuning, and maybe even multimodality, in document-image understanding tasks as part of CVPR's 2nd MMFM Challenge.
 
 [Chess LLM Arena](https://huggingface.co/spaces/mlabonne/chessllm) is a HuggingFace Space where you can make LLMs compete in a chess match.

From 6e026a7b35bf717fbe67565d934b1d2f74a5d438 Mon Sep 17 00:00:00 2001
From: Alonso Silva Allende <alonso.silva@gmail.com>
Date: Sat, 27 Jul 2024 11:04:27 +0200
Subject: [PATCH 027/505] Correct documentation to disable caching

---
 outlines/caching.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/caching.py b/outlines/caching.py
index 95392c7e8..6fdda6214 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -153,8 +153,8 @@ def disable_cache():
 
     `outlines.cache.disable` should be called right after importing outlines:
 
-    >>> import outlines.cache as cache
-    >>> cache.disable()
+    >>> import outlines.caching as cache
+    >>> cache.disable_cache()
 
     """
     global _caching_enabled

From bb927455112c976e3a777338057534d2dddd5171 Mon Sep 17 00:00:00 2001
From: ispobock <ispobaoke@163.com>
Date: Sat, 27 Jul 2024 11:14:10 +0800
Subject: [PATCH 028/505] fix pad token id reset

---
 outlines/models/transformers.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 1d3be511a..7ecc9013f 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -69,7 +69,7 @@ def __init__(self, tokenizer: "PreTrainedTokenizer", **kwargs):
         self.eos_token_id = self.tokenizer.eos_token_id
         self.eos_token = self.tokenizer.eos_token
 
-        if not self.tokenizer.pad_token_id:
+        if self.tokenizer.pad_token_id is None:
             self.tokenizer.pad_token_id = self.tokenizer.eos_token_id
             self.pad_token_id = self.eos_token_id
         else:

From 26e293445cbfe00b782cc83e47cb50124acaddb6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Mon, 29 Jul 2024 16:56:12 +0200
Subject: [PATCH 029/505] Fix link to `mamba` model reference

---
 mkdocs.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mkdocs.yml b/mkdocs.yml
index 7de4a6350..d69d1e718 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -138,7 +138,7 @@ nav:
           - TGI: reference/models/tgi.md
           - ExllamaV2: reference/models/exllamav2.md
           - MLX: reference/models/mlxlm.md
-          - Mamba: reference/models/mamba.md
+          - Mamba: reference/models/transformers.md
         - API:
             - OpenAI: reference/models/openai.md
   - API Reference:

From 951b0204bcf555a7734b4083568849f9fba52888 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 29 Jul 2024 17:08:59 -0500
Subject: [PATCH 030/505] More detailed mlxlm documentation

---
 docs/reference/models/mlxlm.md | 69 +++++++++++++++++++++++++++++++---
 1 file changed, 64 insertions(+), 5 deletions(-)

diff --git a/docs/reference/models/mlxlm.md b/docs/reference/models/mlxlm.md
index e738abdce..cf7bb7443 100644
--- a/docs/reference/models/mlxlm.md
+++ b/docs/reference/models/mlxlm.md
@@ -2,24 +2,83 @@
 
 Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
 
-## Installation
+!!! Note "Installation"
 
-In addition to `outlines`, you must install `mlx-lm` and `mlx` libraries. You must use a device which [supports Metal](https://support.apple.com/en-us/102894).
+    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration.
 
-## Using `models.mlxlm`
+
+## Load the model
+
+You can initialize the model by passing the name of the repository on the HuggingFace Hub. The official repository for mlx-lm supported models is [mlx-community](https://huggingface.co/mlx-community).
 
 ```python
 from outlines import models
 
-model = models.mlxlm("mlx-community/Meta-Llama-3-8B-Instruct-8bit")
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+```
+
+This will download the model files to the hub cache folder and load the weights in memory.
+
+The arguments `model_config` and `tokenizer_config` are available to modify loading behavior. For example, per the `mlx-lm` [documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms#supported-models), you must set an eos_token for `qwen/Qwen-7B`. In outlines you may do so via
+
 ```
+model = models.mlxlm(
+    "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit",
+    tokenizer_config={"eos_token": "<|endoftext|>", "trust_remote_code": True},
+)
+```
+
+**Main parameters:**
+
+(Subject to change. Table based on [mlx-lm.load docstring](https://github.com/ml-explore/mlx-examples/blob/main/llms/mlx_lm/utils.py#L429))
+
+| Parameters         | Type   | Description                                                                                      | Default |
+|--------------------|--------|--------------------------------------------------------------------------------------------------|---------|
+| `tokenizer_config` | `dict` | Configuration parameters specifically for the tokenizer. Defaults to an empty dictionary.        | `{}`    |
+| `model_config`     | `dict` | Configuration parameters specifically for the model. Defaults to an empty dictionary.            | `{}`    |
+| `adapter_path`     | `str`  | Path to the LoRA adapters. If provided, applies LoRA layers to the model.                        | `None`  |
+| `lazy`             | `bool` | If False, evaluate the model parameters to make sure they are loaded in memory before returning. | `False` |
+
+
+## Generate text
+
+You may generate text using the parameters described in the [text generation documentation](../text.md).
 
 With the loaded model, you can generate text or perform structured generation, e.g.
 
 ```python
 from outlines import models, generate
 
-model = models.mlxlm("mlx-community/Meta-Llama-3-8B-Instruct-8bit")
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+generator = generate.text(model)
+
+answer = generator("A prompt", temperature=2.0)
+```
+
+## Streaming
+
+You may creating a streaming iterable with minimal changes
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+generator = generate.text(model)
+
+for token_str in generator.text("A prompt", temperature=2.0):
+    print(token_str)
+```
+
+## Structured
+
+You may perform structured generation with mlxlm to guarantee your output will match a regex pattern, json schema, or lark grammar.
+
+Example: Phone number generation with pattern `"\\+?[1-9][0-9]{7,14}"`:
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
 
 phone_number_pattern = "\\+?[1-9][0-9]{7,14}"
 generator = generate.regex(model, phone_number_pattern)

From e3247fe633ca5c958c6d65be6803ace772547f72 Mon Sep 17 00:00:00 2001
From: Alonso Silva <alonso.silva@gmail.com>
Date: Thu, 8 Aug 2024 17:28:20 +0200
Subject: [PATCH 031/505] Add chain of thought example

---
 docs/cookbook/chain_of_thought.md | 123 ++++++++++++++++++++++++++++++
 docs/cookbook/index.md            |   1 +
 mkdocs.yml                        |   1 +
 3 files changed, 125 insertions(+)
 create mode 100644 docs/cookbook/chain_of_thought.md

diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
new file mode 100644
index 000000000..2ab68ff7a
--- /dev/null
+++ b/docs/cookbook/chain_of_thought.md
@@ -0,0 +1,123 @@
+# Chain of thought
+
+
+Chain of thought is a prompting technique introduced in the paper [``Chain-of-Thought Prompting Elicits Reasoning in Large Language Models''](https://arxiv.org/abs/2201.11903) where throught prompting the authors generate a series of intermediate reasoning steps which improves the ability of LLMs to perform complex reasoning.
+
+In this guide, we use [outlines](https://outlines-dev.github.io/outlines/) to apply chain of thought through structured output.
+
+We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```shell
+pip install llama-cpp-python
+```
+
+We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+```shell
+wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+```
+
+We initialize the model:
+
+```python
+from llama_cpp import Llama
+from outlines import generate, models
+
+llm = Llama(
+    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = models.LlamaCpp(llm)
+```
+
+## Chain of thought
+
+We first define our Pydantic class for a reasoning step:
+
+```python
+from pydantic import BaseModel, Field
+
+class Reasoning_Step(BaseModel):
+    reasoning_step: str = Field(..., description="Reasoning step")
+```
+
+We then define the Pydantic class for reasoning which will consist on a list of reasoning steps and a conclusion, and we get its JSON schema:
+
+```python
+from typing import List
+
+from typing import List
+
+class Reasoning(BaseModel):
+    reasoning: List[Reasoning_Step] = Field(..., description="List of reasoning steps")
+    conclusion: str = Field(..., description="Conclusion")
+
+json_schema = Reasoning.model_json_schema()
+```
+
+We could generate using the json schema but for a change we will use the regex:
+
+```python
+from outlines.integrations.utils import convert_json_schema_to_str
+from outlines.fsm.json_schema import build_regex_from_schema
+
+json_schema = Reasoning.model_json_schema()
+schema_str = convert_json_schema_to_str(json_schema=json_schema)
+regex_str = build_regex_from_schema(schema_str)
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
+
+```python
+def generate_hermes_prompt(user_prompt):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + user_prompt
+        + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
+```
+
+For a given user prompt, for example:
+
+```python
+user_prompt = "9.11 and 9.9 -- which is bigger?"
+```
+
+We can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+
+```python
+generator = generate.regex(model, regex_str)
+prompt = generate_hermes_prompt(user_prompt)
+response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+```
+
+We obtain the reasoning steps as well as the conclusion
+
+```python
+import json
+
+json_response = json.loads(response)
+
+print(json_response["reasoning"])
+print(json_response["conclusion"])
+# [{'reasoning_step': 'Both 9.11 and 9.9 are decimal numbers.'},
+#  {'reasoning_step': 'When comparing decimal numbers, we look at the numbers after the decimal point.'},
+#  {'reasoning_step': 'In this case, 9.11 has the number 1 after the decimal point, while 9.9 has the number 9.'},
+#  {'reasoning_step': 'Since 1 is greater than 9, 9.11 is greater than 9.9.'}]
+# '9.11 is bigger.'
+```
+
+We notice that the 4th reasoning step is wrong ``Since 1 is greater than 9, 9.11 is greater than 9.9.'', so we could probably give the model some examples for this particular task.
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index 24849d5ee..e8816125a 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -8,3 +8,4 @@
 - [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
 - [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
+- [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
diff --git a/mkdocs.yml b/mkdocs.yml
index d69d1e718..bb47ab45b 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -108,6 +108,7 @@ nav:
       - Perspective-taking prompting: cookbook/simtom.md
       - Question-answering with citations: cookbook/qa-with-citations.md
       - Knowledge Graph Extraction: cookbook/knowledge_graph_extraction.md
+      - Chain of Thought (CoT): cookbook/chain_of_thought.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From 2a2a5d3b77a5810459c7d5c35357578fcb3078ac Mon Sep 17 00:00:00 2001
From: Alonso Silva <alonso.silva@gmail.com>
Date: Thu, 8 Aug 2024 17:36:55 +0200
Subject: [PATCH 032/505] Fix typos

---
 docs/cookbook/chain_of_thought.md | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)

diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
index 2ab68ff7a..d320feb8d 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -1,7 +1,7 @@
 # Chain of thought
 
 
-Chain of thought is a prompting technique introduced in the paper [``Chain-of-Thought Prompting Elicits Reasoning in Large Language Models''](https://arxiv.org/abs/2201.11903) where throught prompting the authors generate a series of intermediate reasoning steps which improves the ability of LLMs to perform complex reasoning.
+Chain of thought is a prompting technique introduced in the paper ["Chain-of-Thought Prompting Elicits Reasoning in Large Language Models"](https://arxiv.org/abs/2201.11903) where throught prompting the authors generate a series of intermediate reasoning steps which improves the ability of LLMs to perform complex reasoning.
 
 In this guide, we use [outlines](https://outlines-dev.github.io/outlines/) to apply chain of thought through structured output.
 
@@ -52,8 +52,6 @@ We then define the Pydantic class for reasoning which will consist on a list of
 ```python
 from typing import List
 
-from typing import List
-
 class Reasoning(BaseModel):
     reasoning: List[Reasoning_Step] = Field(..., description="List of reasoning steps")
     conclusion: str = Field(..., description="Conclusion")
@@ -61,13 +59,12 @@ class Reasoning(BaseModel):
 json_schema = Reasoning.model_json_schema()
 ```
 
-We could generate using the json schema but for a change we will use the regex:
+We could generate a response using the json schema but for a change we will use the regex:
 
 ```python
 from outlines.integrations.utils import convert_json_schema_to_str
 from outlines.fsm.json_schema import build_regex_from_schema
 
-json_schema = Reasoning.model_json_schema()
 schema_str = convert_json_schema_to_str(json_schema=json_schema)
 regex_str = build_regex_from_schema(schema_str)
 ```
@@ -88,13 +85,13 @@ def generate_hermes_prompt(user_prompt):
     )
 ```
 
-For a given user prompt, for example:
+For a given user prompt:
 
 ```python
 user_prompt = "9.11 and 9.9 -- which is bigger?"
 ```
 
-We can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+we can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
 generator = generate.regex(model, regex_str)
@@ -102,7 +99,7 @@ prompt = generate_hermes_prompt(user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
-We obtain the reasoning steps as well as the conclusion
+We obtain a series of intermediate reasoning steps as well as the conclusion:
 
 ```python
 import json
@@ -118,6 +115,6 @@ print(json_response["conclusion"])
 # '9.11 is bigger.'
 ```
 
-We notice that the 4th reasoning step is wrong ``Since 1 is greater than 9, 9.11 is greater than 9.9.'', so we could probably give the model some examples for this particular task.
+We notice that the 4th reasoning step is wrong ``Since 1 is greater than 9, 9.11 is greater than 9.9.'', so we should probably give the model some examples for this particular task.
 
 This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).

From 2d1a06dc13f321598197acd9191d466620ce2727 Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Thu, 8 Aug 2024 17:08:13 -0500
Subject: [PATCH 033/505] Fix coverage exclude_lines setting for ellipsis

Closes #1085
---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index f94b3c84d..a41e29772 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -154,7 +154,7 @@ omit = [
 exclude_lines = [
     "pragma: no cover",
     "if TYPE_CHECKING:",
-    "...",
+    "\\.\\.\\.",
 ]
 show_missing = true
 

From 5e8f7709e3cecd02943120ed01420f00159cedbc Mon Sep 17 00:00:00 2001
From: Alonso Silva <alonso.silva@gmail.com>
Date: Sat, 10 Aug 2024 19:01:51 +0200
Subject: [PATCH 034/505] Add ReAct agent example

---
 docs/cookbook/index.md       |   1 +
 docs/cookbook/react_agent.md | 257 +++++++++++++++++++++++++++++++++++
 mkdocs.yml                   |   1 +
 3 files changed, 259 insertions(+)
 create mode 100644 docs/cookbook/react_agent.md

diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index e8816125a..56334302a 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -9,3 +9,4 @@
 - [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
+- [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
diff --git a/docs/cookbook/react_agent.md b/docs/cookbook/react_agent.md
new file mode 100644
index 000000000..74930b70b
--- /dev/null
+++ b/docs/cookbook/react_agent.md
@@ -0,0 +1,257 @@
+# ReAct Agent
+
+This example shows how to use [outlines](https://outlines-dev.github.io/outlines/) to build your own agent with open weights local models and structured outputs. It is inspired by the blog post [A simple Python implementation of the ReAct pattern for LLMs](https://til.simonwillison.net/llms/python-react-pattern) by [Simon Willison](https://simonwillison.net/).
+
+The ReAct pattern (for Reason+Act) is described in the paper [ReAct: Synergizing Reasoning and Acting in Language Models](https://arxiv.org/abs/2210.03629). It's a pattern where you implement additional actions that an LLM can take - searching Wikipedia or running calculations for example - and then teach it how to request the execution of those actions, and then feed their results back into the LLM.
+
+Additionally, we give the LLM the possibility of using a scratchpad described in the paper [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://arxiv.org/abs/2112.00114) which improves the ability of LLMs to perform multi-step computations.
+
+We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```shell
+pip install llama-cpp-python
+```
+
+We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+```shell
+wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+```
+
+We initialize the model:
+
+```python
+import llama_cpp
+from llama_cpp import Llama
+from outlines import generate, models
+
+llm = Llama(
+    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = models.LlamaCpp(llm)
+```
+
+## Build a ReAct agent
+
+In this example, we use two tools:
+
+- wikipedia: \<search term\> - search Wikipedia and returns the snippet of the first result
+- calculate: \<expression\> - evaluate an expression using Python's eval() function
+
+```python
+import httpx
+
+def wikipedia(q):
+    return httpx.get("https://en.wikipedia.org/w/api.php", params={
+        "action": "query",
+        "list": "search",
+        "srsearch": q,
+        "format": "json"
+    }).json()["query"]["search"][0]["snippet"]
+```
+
+```python
+def calculate(numexp):
+    return eval(numexp)
+```
+
+We define the logic of the agent through a Pydantic class. First, we want the LLM to decide only between the two previously defined tools:
+
+```python
+from enum import Enum
+
+class Action(str, Enum):
+    wikipedia = "wikipedia"
+    calculate = "calculate"
+```
+
+Our agent will loop through Thought and Action. We explicitly give the Action Input field so it doesn't forget to add the arguments of the Action. We also add a scratchpad (optional).
+
+```python
+from pydantic import BaseModel, Field
+
+class Reason_and_Act(BaseModel):
+    Scratchpad: str = Field(..., description="Information from the Observation useful to answer the question")
+    Thought: str = Field(..., description="It describes your thoughts about the question you have been asked")
+    Action: Action
+    Action_Input: str = Field(..., description="The arguments of the Action.")
+```
+
+Our agent will reach a Final Answer. We also add a scratchpad (optional).
+
+```python
+class Final_Answer(BaseModel):
+    Scratchpad: str = Field(..., description="Information from the Observation useful to answer the question")
+    Final_Answer: str = Field(..., description="Answer to the question grounded on the Observation")
+```
+
+Our agent will decide when it has reached a Final Answer and therefore to stop the loop of Thought and Action.
+
+```python
+from typing import Union
+
+class Decision(BaseModel):
+    Decision: Union[Reason_and_Act, Final_Answer]
+```
+
+We could generate a response using the json schema but we will use the regex and check that everything is working as expected:
+
+```python
+from outlines.integrations.utils import convert_json_schema_to_str
+from outlines.fsm.json_schema import build_regex_from_schema
+
+json_schema = Decision.model_json_schema()
+schema_str = convert_json_schema_to_str(json_schema=json_schema)
+regex_str = build_regex_from_schema(schema_str)
+print(regex_str)
+# '\\{[ ]?"Decision"[ ]?:[ ]?(\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Thought"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Action"[ ]?:[ ]?("wikipedia"|"calculate")[ ]?,[ ]?"Action_Input"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\}|\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Final_Answer"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\})[ ]?\\}'
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic:
+
+```python
+import datetime
+
+def generate_hermes_prompt(question, schema=""):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON with correct Pydantic schema. "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema>\n"
+        "Today is " + datetime.datetime.today().strftime('%Y-%m-%d') + ".\n" +
+        "You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation. "
+        "At the end of the loop you output a Final Answer. "
+        "Use Scratchpad to store the information from the Observation useful to answer the question "
+        "Use Thought to describe your thoughts about the question you have been asked "
+        "and reflect carefully about the Observation if it exists. "
+        "Use Action to run one of the actions available to you. "
+        "Use Action Input to input the arguments of the selected action - then return PAUSE. "
+        "Observation will be the result of running those actions. "
+        "Your available actions are:\n"
+        "calculate:\n"
+        "e.g. calulate: 4**2 / 3\n"
+        "Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary\n"
+        "wikipedia:\n"
+        "e.g. wikipedia: Django\n"
+        "Returns a summary from searching Wikipedia\n"
+        "DO NOT TRY TO GUESS THE ANSWER. Begin! <|im_end|>"
+        "\n<|im_start|>user\n" + question + "<|im_end|>"
+        "\n<|im_start|>assistant\n"
+    )
+```
+
+We define a ChatBot class
+
+```python
+class ChatBot:
+    def __init__(self, prompt=""):
+        self.prompt = prompt
+
+    def __call__(self, user_prompt):
+        self.prompt += user_prompt
+        result = self.execute()
+        return result
+
+    def execute(self):
+        generator = generate.regex(model, regex_str)
+        result = generator(self.prompt, max_tokens=1024, temperature=0, seed=42)
+        return result
+```
+
+We define a query function:
+
+```python
+import json
+
+def query(question, max_turns=5):
+    i = 0
+    next_prompt = (
+        "\n<|im_start|>user\n" + question + "<|im_end|>"
+        "\n<|im_start|>assistant\n"
+    )
+    previous_actions = []
+    while i < max_turns:
+        i += 1
+        prompt = generate_hermes_prompt(question=question, schema=Decision.model_json_schema())
+        bot = ChatBot(prompt=prompt)
+        result = bot(next_prompt)
+        json_result = json.loads(result)['Decision']
+        if "Final_Answer" not in list(json_result.keys()):
+            scratchpad = json_result['Scratchpad'] if i == 0 else ""
+            thought = json_result['Thought']
+            action = json_result['Action']
+            action_input = json_result['Action_Input']
+            print(f"\x1b[34m Scratchpad: {scratchpad} \x1b[0m")
+            print(f"\x1b[34m Thought: {thought} \x1b[0m")
+            print(f"\x1b[36m  -- running {action}: {str(action_input)}\x1b[0m")
+            if action + ": " + str(action_input) in previous_actions:
+                observation = "You already run that action. **TRY A DIFFERENT ACTION INPUT.**"
+            else:
+                if action=="calculate":
+                    try:
+                        observation = eval(str(action_input))
+                    except Exception as e:
+                        observation = f"{e}"
+                elif action=="wikipedia":
+                    try:
+                        observation = wikipedia(str(action_input))
+                    except Exception as e:
+                        observation = f"{e}"
+            print()
+            print(f"\x1b[33m Observation: {observation} \x1b[0m")
+            print()
+            previous_actions.append(action + ": " + str(action_input))
+            next_prompt += (
+                "\nScratchpad: " + scratchpad +
+                "\nThought: " + thought +
+                "\nAction: " + action  +
+                "\nAction Input: " + action_input +
+                "\nObservation: " + str(observation)
+            )
+        else:
+            scratchpad = json_result["Scratchpad"]
+            final_answer = json_result["Final_Answer"]
+            print(f"\x1b[34m Scratchpad: {scratchpad} \x1b[0m")
+            print(f"\x1b[34m Final Answer: {final_answer} \x1b[0m")
+            return final_answer
+    print(f"\nFinal Answer: I am sorry, but I am unable to answer your question. Please provide more information or a different question.")
+    return "No answer found"
+```
+
+We can now test our ReAct agent:
+
+```python
+print(query("What's 2 to the power of 10?"))
+# Scratchpad:
+# Thought: I need to perform a mathematical calculation to find the result of 2 to the power of 10.
+#  -- running calculate: 2**10
+#
+# Observation: 1024
+#
+# Scratchpad: 2 to the power of 10 is 1024.
+# Final Answer: 2 to the power of 10 is 1024.
+# 2 to the power of 10 is 1024.
+```
+
+```python
+print(query("What does England share borders with?"))
+# Scratchpad:
+# Thought: To answer this question, I will use the 'wikipedia' action to gather information about England's geographical location and its borders.
+#  -- running wikipedia: England borders
+#
+# Observation: Anglo-Scottish <span class="searchmatch">border</span> (Scottish Gaelic: Crìochan Anglo-Albannach) is an internal <span class="searchmatch">border</span> of the United Kingdom separating Scotland and <span class="searchmatch">England</span> which runs for
+#
+# Scratchpad: Anglo-Scottish border (Scottish Gaelic: Crìochan Anglo-Albannach) is an internal border of the United Kingdom separating Scotland and England which runs for
+# Final Answer: England shares a border with Scotland.
+# England shares a border with Scotland.
+```
+
+As mentioned in Simon's blog post, this is not a very robust implementation at all and there's a ton of room for improvement. But it is lovely how simple it is with a few lines of Python to make these extra capabilities available to the LLM. And now you can run it locally with an open weights LLM.
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/mkdocs.yml b/mkdocs.yml
index bb47ab45b..e5f3417ac 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -109,6 +109,7 @@ nav:
       - Question-answering with citations: cookbook/qa-with-citations.md
       - Knowledge Graph Extraction: cookbook/knowledge_graph_extraction.md
       - Chain of Thought (CoT): cookbook/chain_of_thought.md
+      - ReAct Agent: cookbook/react_agent.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From 3072b8f535cdfc2587979aa0821a1ded5796827a Mon Sep 17 00:00:00 2001
From: Christian Heimes <christian@python.org>
Date: Tue, 13 Aug 2024 10:47:58 +0200
Subject: [PATCH 035/505] Include SPDX license info in project metadata

A license entry in the project metadata makes it easier to detect the
license of a project.

Before:
```shell
$ pip install pip-licenses outlines
$ pip-licenses  | grep outlines
 outlines                   0.0.46       UNKNOWN
```

After:
```shell
$ pip install .
$ pip-licenses  | grep outlines
 outlines                   0.0.47.dev43+g8b78f5d  Apache-2.0
```

Signed-off-by: Christian Heimes <christian@python.org>
---
 pyproject.toml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/pyproject.toml b/pyproject.toml
index a41e29772..ea0ac11c8 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -7,6 +7,7 @@ name = "outlines"
 authors= [{name = "Outlines Developers"}]
 description = "Probabilistic Generative Model Programming"
 requires-python = ">=3.8"
+license = {text = "Apache-2.0"}
 keywords=[
     "machine learning",
     "deep learning",

From d0be920f46b0132d87732b94e6861ef872ca558c Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 22 Jul 2024 17:41:25 -0500
Subject: [PATCH 036/505] Restructure documentation, use phi-3-mini, include
 structured generation example, add feature matrix to models

---
 .../images/logits_processing_diagram.svg      | 157 +++++++++++++
 docs/quickstart.md                            | 214 +++---------------
 docs/reference/{ => generation}/cfg.md        |   0
 docs/reference/{ => generation}/choices.md    |   0
 .../{ => generation}/custom_fsm_ops.md        |   0
 docs/reference/{ => generation}/format.md     |   0
 docs/reference/generation/generation.md       | 210 +++++++++++++++++
 docs/reference/{ => generation}/json.md       |   3 +-
 docs/reference/{ => generation}/regex.md      |   0
 .../structured_generation_explanation.md      |  58 +++++
 docs/reference/{ => generation}/types.md      |   0
 docs/reference/json_mode.md                   |  17 --
 docs/reference/models/models.md               |  93 ++++++++
 docs/reference/models/transformers_vision.md  |   3 +-
 docs/stylesheets/extra.css                    |   4 +
 mkdocs.yml                                    |  25 +-
 requirements-doc.txt                          |   1 +
 17 files changed, 573 insertions(+), 212 deletions(-)
 create mode 100644 docs/assets/images/logits_processing_diagram.svg
 rename docs/reference/{ => generation}/cfg.md (100%)
 rename docs/reference/{ => generation}/choices.md (100%)
 rename docs/reference/{ => generation}/custom_fsm_ops.md (100%)
 rename docs/reference/{ => generation}/format.md (100%)
 create mode 100644 docs/reference/generation/generation.md
 rename docs/reference/{ => generation}/json.md (98%)
 rename docs/reference/{ => generation}/regex.md (100%)
 create mode 100644 docs/reference/generation/structured_generation_explanation.md
 rename docs/reference/{ => generation}/types.md (100%)
 delete mode 100644 docs/reference/json_mode.md
 create mode 100644 docs/reference/models/models.md

diff --git a/docs/assets/images/logits_processing_diagram.svg b/docs/assets/images/logits_processing_diagram.svg
new file mode 100644
index 000000000..92668e6c6
--- /dev/null
+++ b/docs/assets/images/logits_processing_diagram.svg
@@ -0,0 +1,157 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
+ "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<!-- Generated by graphviz version 2.43.0 (0)
+ -->
+<!-- Title: %3 Pages: 1 -->
+<svg width="214pt" height="784pt"
+ viewBox="0.00 0.00 214.00 784.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 780)">
+<title>%3</title>
+<polygon fill="white" stroke="transparent" points="-4,4 -4,-780 210,-780 210,4 -4,4"/>
+<!-- inputTokens -->
+<g id="node1" class="node">
+<title>inputTokens</title>
+<polygon fill="none" stroke="black" points="48,-751 48,-772 159,-772 159,-751 48,-751"/>
+<text text-anchor="start" x="51" y="-758.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Input Tokens</text>
+<polygon fill="none" stroke="black" points="48,-730 48,-751 95,-751 95,-730 48,-730"/>
+<text text-anchor="start" x="66.5" y="-736.8" font-family="Times,serif" font-size="14.00">7</text>
+<polygon fill="none" stroke="black" points="95,-730 95,-751 127,-751 127,-730 95,-730"/>
+<text text-anchor="start" x="106" y="-736.8" font-family="Times,serif" font-size="14.00">4</text>
+<polygon fill="none" stroke="black" points="127,-730 127,-751 159,-751 159,-730 127,-730"/>
+<text text-anchor="start" x="138" y="-736.8" font-family="Times,serif" font-size="14.00">8</text>
+</g>
+<!-- TransformerDecoder -->
+<g id="node2" class="node">
+<title>TransformerDecoder</title>
+<polygon fill="lightblue" stroke="lightblue" points="190,-690 16,-690 16,-654 190,-654 190,-690"/>
+<text text-anchor="middle" x="103" y="-668.9" font-family="Helvetica,sans-Serif" font-size="12.00">Transformer Decoder Pass</text>
+</g>
+<!-- inputTokens&#45;&gt;TransformerDecoder -->
+<g id="edge1" class="edge">
+<title>inputTokens&#45;&gt;TransformerDecoder</title>
+<path fill="none" stroke="black" d="M103,-725.64C103,-717.67 103,-708.77 103,-700.58"/>
+<polygon fill="black" stroke="black" points="106.5,-700.38 103,-690.38 99.5,-700.38 106.5,-700.38"/>
+</g>
+<!-- logitsTable -->
+<g id="node3" class="node">
+<title>logitsTable</title>
+<polygon fill="none" stroke="black" points="19,-593 19,-614 188,-614 188,-593 19,-593"/>
+<text text-anchor="start" x="22" y="-600.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Model Output Logits</text>
+<polygon fill="none" stroke="black" points="19,-572 19,-593 85,-593 85,-572 19,-572"/>
+<text text-anchor="start" x="30.5" y="-578.8" font-family="Times,serif" font-size="14.00">Token</text>
+<polygon fill="none" stroke="black" points="85,-572 85,-593 188,-593 188,-572 85,-572"/>
+<text text-anchor="start" x="97" y="-578.8" font-family="Times,serif" font-size="14.00">Probability</text>
+<polygon fill="none" stroke="black" points="19,-551 19,-572 85,-572 85,-551 19,-551"/>
+<text text-anchor="start" x="45.5" y="-557.8" font-family="Times,serif" font-size="14.00">+</text>
+<polygon fill="none" stroke="black" points="85,-551 85,-572 188,-572 188,-551 85,-551"/>
+<text text-anchor="start" x="125" y="-557.8" font-family="Times,serif" font-size="14.00">3%</text>
+<polygon fill="none" stroke="black" points="19,-530 19,-551 85,-551 85,-530 19,-530"/>
+<text text-anchor="start" x="41" y="-536.8" font-family="Times,serif" font-size="14.00">foo</text>
+<polygon fill="none" stroke="black" points="85,-530 85,-551 188,-551 188,-530 85,-530"/>
+<text text-anchor="start" x="125" y="-536.8" font-family="Times,serif" font-size="14.00">4%</text>
+<polygon fill="none" stroke="black" points="19,-509 19,-530 85,-530 85,-509 19,-509"/>
+<text text-anchor="start" x="49.5" y="-515.8" font-family="Times,serif" font-size="14.00">.</text>
+<polygon fill="none" stroke="black" points="85,-509 85,-530 188,-530 188,-509 85,-509"/>
+<text text-anchor="start" x="125" y="-515.8" font-family="Times,serif" font-size="14.00">7%</text>
+<polygon fill="none" stroke="black" points="19,-488 19,-509 85,-509 85,-488 19,-488"/>
+<text text-anchor="start" x="47" y="-494.8" font-family="Times,serif" font-size="14.00">1</text>
+<polygon fill="none" stroke="black" points="85,-488 85,-509 188,-509 188,-488 85,-488"/>
+<text text-anchor="start" x="120.5" y="-494.8" font-family="Times,serif" font-size="14.00">11%</text>
+<polygon fill="none" stroke="black" points="19,-467 19,-488 85,-488 85,-467 19,-467"/>
+<text text-anchor="start" x="47" y="-473.8" font-family="Times,serif" font-size="14.00">2</text>
+<polygon fill="none" stroke="black" points="85,-467 85,-488 188,-488 188,-467 85,-467"/>
+<text text-anchor="start" x="120.5" y="-473.8" font-family="Times,serif" font-size="14.00">13%</text>
+<polygon fill="none" stroke="black" points="19,-446 19,-467 85,-467 85,-446 19,-446"/>
+<text text-anchor="start" x="47" y="-452.8" font-family="Times,serif" font-size="14.00">3</text>
+<polygon fill="none" stroke="black" points="85,-446 85,-467 188,-467 188,-446 85,-446"/>
+<text text-anchor="start" x="120.5" y="-452.8" font-family="Times,serif" font-size="14.00">17%</text>
+</g>
+<!-- TransformerDecoder&#45;&gt;logitsTable -->
+<g id="edge2" class="edge">
+<title>TransformerDecoder&#45;&gt;logitsTable</title>
+<path fill="none" stroke="black" d="M103,-653.83C103,-646.79 103,-638.04 103,-628.47"/>
+<polygon fill="black" stroke="black" points="106.5,-628.25 103,-618.25 99.5,-628.25 106.5,-628.25"/>
+</g>
+<!-- OutlinesLogitsProcessor -->
+<g id="node4" class="node">
+<title>OutlinesLogitsProcessor</title>
+<polygon fill="lightblue" stroke="lightblue" points="206,-406 0,-406 0,-370 206,-370 206,-406"/>
+<text text-anchor="middle" x="103" y="-384.9" font-family="Helvetica,sans-Serif" font-size="12.00">Outlines Regex Logits Processor</text>
+</g>
+<!-- logitsTable&#45;&gt;OutlinesLogitsProcessor -->
+<g id="edge3" class="edge">
+<title>logitsTable&#45;&gt;OutlinesLogitsProcessor</title>
+<path fill="none" stroke="black" d="M103,-441.93C103,-432.71 103,-423.91 103,-416.22"/>
+<polygon fill="black" stroke="black" points="106.5,-416.17 103,-406.17 99.5,-416.17 106.5,-416.17"/>
+</g>
+<!-- logitsProcessorTable -->
+<g id="node5" class="node">
+<title>logitsProcessorTable</title>
+<polygon fill="none" stroke="black" points="30,-309 30,-330 177,-330 177,-309 30,-309"/>
+<text text-anchor="start" x="37" y="-316.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Processed Logits</text>
+<polygon fill="none" stroke="black" points="30,-288 30,-309 92,-309 92,-288 30,-288"/>
+<text text-anchor="start" x="39.5" y="-294.8" font-family="Times,serif" font-size="14.00">Token</text>
+<polygon fill="none" stroke="black" points="92,-288 92,-309 177,-309 177,-288 92,-288"/>
+<text text-anchor="start" x="95" y="-294.8" font-family="Times,serif" font-size="14.00">Probability</text>
+<polygon fill="none" stroke="black" points="30,-267 30,-288 92,-288 92,-267 30,-267"/>
+<text text-anchor="start" x="54.5" y="-273.8" font-family="Times,serif" font-size="14.00">+</text>
+<polygon fill="none" stroke="black" points="92,-267 92,-288 177,-288 177,-267 92,-267"/>
+<text text-anchor="start" x="123" y="-273.8" font-family="Times,serif" font-size="14.00">0%</text>
+<polygon fill="none" stroke="black" points="30,-246 30,-267 92,-267 92,-246 30,-246"/>
+<text text-anchor="start" x="41" y="-536.8" font-family="Times,serif" font-size="14.00">foo</text>
+<polygon fill="none" stroke="black" points="92,-246 92,-267 177,-267 177,-246 92,-246"/>
+<text text-anchor="start" x="123" y="-252.8" font-family="Times,serif" font-size="14.00">0%</text>
+<polygon fill="none" stroke="black" points="30,-225 30,-246 92,-246 92,-225 30,-225"/>
+<text text-anchor="start" x="58.5" y="-231.8" font-family="Times,serif" font-size="14.00">.</text>
+<polygon fill="none" stroke="black" points="92,-225 92,-246 177,-246 177,-225 92,-225"/>
+<text text-anchor="start" x="123" y="-231.8" font-family="Times,serif" font-size="14.00">7%</text>
+<polygon fill="none" stroke="black" points="30,-204 30,-225 92,-225 92,-204 30,-204"/>
+<text text-anchor="start" x="56" y="-210.8" font-family="Times,serif" font-size="14.00">1</text>
+<polygon fill="none" stroke="black" points="92,-204 92,-225 177,-225 177,-204 92,-204"/>
+<text text-anchor="start" x="118.5" y="-210.8" font-family="Times,serif" font-size="14.00">11%</text>
+<polygon fill="none" stroke="black" points="30,-183 30,-204 92,-204 92,-183 30,-183"/>
+<text text-anchor="start" x="56" y="-189.8" font-family="Times,serif" font-size="14.00">2</text>
+<polygon fill="none" stroke="black" points="92,-183 92,-204 177,-204 177,-183 92,-183"/>
+<text text-anchor="start" x="118.5" y="-189.8" font-family="Times,serif" font-size="14.00">13%</text>
+<polygon fill="none" stroke="black" points="30,-162 30,-183 92,-183 92,-162 30,-162"/>
+<text text-anchor="start" x="56" y="-168.8" font-family="Times,serif" font-size="14.00">3</text>
+<polygon fill="none" stroke="black" points="92,-162 92,-183 177,-183 177,-162 92,-162"/>
+<text text-anchor="start" x="118.5" y="-168.8" font-family="Times,serif" font-size="14.00">17%</text>
+</g>
+<!-- OutlinesLogitsProcessor&#45;&gt;logitsProcessorTable -->
+<g id="edge4" class="edge">
+<title>OutlinesLogitsProcessor&#45;&gt;logitsProcessorTable</title>
+<path fill="none" stroke="black" d="M103,-369.83C103,-362.79 103,-354.04 103,-344.47"/>
+<polygon fill="black" stroke="black" points="106.5,-344.25 103,-334.25 99.5,-344.25 106.5,-344.25"/>
+</g>
+<!-- sampler -->
+<g id="node6" class="node">
+<title>sampler</title>
+<polygon fill="lightblue" stroke="lightblue" points="136.5,-122 69.5,-122 69.5,-86 136.5,-86 136.5,-122"/>
+<text text-anchor="middle" x="103" y="-100.9" font-family="Helvetica,sans-Serif" font-size="12.00">Sampler</text>
+</g>
+<!-- logitsProcessorTable&#45;&gt;sampler -->
+<g id="edge5" class="edge">
+<title>logitsProcessorTable&#45;&gt;sampler</title>
+<path fill="none" stroke="black" d="M103,-157.93C103,-148.71 103,-139.91 103,-132.22"/>
+<polygon fill="black" stroke="black" points="106.5,-132.17 103,-122.17 99.5,-132.17 106.5,-132.17"/>
+</g>
+<!-- sampledTokenTable -->
+<g id="node7" class="node">
+<title>sampledTokenTable</title>
+<polygon fill="none" stroke="black" points="40,-25 40,-46 166,-46 166,-25 40,-25"/>
+<text text-anchor="start" x="43" y="-32.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Sampled Token</text>
+<polygon fill="none" stroke="black" points="40,-4 40,-25 119,-25 119,-4 40,-4"/>
+<text text-anchor="start" x="58" y="-10.8" font-family="Times,serif" font-size="14.00">Token</text>
+<polygon fill="none" stroke="black" points="119,-4 119,-25 166,-25 166,-4 119,-4"/>
+<text text-anchor="start" x="137.5" y="-10.8" font-family="Times,serif" font-size="14.00">3</text>
+</g>
+<!-- sampler&#45;&gt;sampledTokenTable -->
+<g id="edge6" class="edge">
+<title>sampler&#45;&gt;sampledTokenTable</title>
+<path fill="none" stroke="black" d="M103,-85.91C103,-78.33 103,-69.13 103,-60.23"/>
+<polygon fill="black" stroke="black" points="106.5,-60.05 103,-50.05 99.5,-60.05 106.5,-60.05"/>
+</g>
+</g>
+</svg>
diff --git a/docs/quickstart.md b/docs/quickstart.md
index 7d4e86fbc..fcb524d8a 100644
--- a/docs/quickstart.md
+++ b/docs/quickstart.md
@@ -6,6 +6,7 @@ title: Quickstart
 
 After [installing Outlines](installation.md), the fastest way to get to up to speed with the library is to get acquainted with its few core elements. We advise you to take a quick look at this page to see everything Outlines has to offer before diving in the [documentation](reference/index.md).
 
+
 ## Core elements
 
 ### Models
@@ -15,88 +16,57 @@ The first step when writing a program with Outlines is to initialize a model. We
 ```python
 import outlines
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = outlines.models.transformers(
+    "microsoft/Phi-3-mini-4k-instruct",
+    device="cuda"  # optional device argument, default is cpu
+)
 ```
 
-### Text generator
-
-Once the model is initialized you can build a text generator. This generator can be called with a prompt directly, or you can use the `stream` method to generate text token by token:
-
-=== "Generate"
-
-    ```python
-    import outlines
+Outlines supports a wide variety of inference engines and model weight types. More details on different models can be found in the [Outlines Models](./reference/models/models.md) documentation page.
 
-    model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
-    generator = outlines.generate.text(model)
+### Generation
 
-    result = generator("What's 2+2?", max_tokens=100)
+Once the model is initialized you can build an `outlines.generate` generator. This generator can be called with a prompt directly.
 
-    print(result)
-    # That's right, it's 4! But remember, a delicious and nutrient dense 4,
-    # according to YEARS BUILT ON SOLID SCIENCE. This column presents additional
-    # findings from the fifteen-year study that produced the 2+2=4 conclusion.
-    ```
+([Outlines Structured Generation Full Documentation](./reference/generation/generation.md))
 
-=== "Stream"
+=== "Text"
 
     ```python
-    import outlines
-
-    model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
     generator = outlines.generate.text(model)
 
+    result = generator("Question: What's 2+2? Answer:", max_tokens=100)
+    print(result)
+    # The answer is 4
+
+	# Outlines also supports streaming output
     stream = generator.stream("What's 2+2?", max_tokens=4)
     for i in range(5):
         token = next(stream)
-        print(token)
-    # ['Is']
-    # [' this']
-    # [' even']
-    # [' a']
-    # [' question']
+        print(repr(token))
+	# '2'
+	# '+'
+	# '2'
+	# ' equals'
+	# '4'
     ```
 
-### Multi-label classification
-
-Outlines allows you to do multi-label classification by guiding the model so it can only output either of the specified choices:
-
-```python
-import outlines
-
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
-generator = outlines.generate.choice(model, ["Blue", "Red", "Yellow"])
-
-color = generator("What is the closest color to Indigo? ")
-print(color)
-# Blue
-```
+=== "Structured"
 
-### JSON-structured generation
+    Along with typical language model generation behavior via, `outlines.generate.text`, Outlines supports structured generation, which guarantees the tokens generated by the model will follow a predefined structure. Structures can be defined by a regex pattern, JSON schema, python object type, or a Lark grammar defining a parsable language such as SQL or Python.
 
-Outlines can guide models so that they output valid JSON **100%** of the time. You can either specify the structure using [Pydantic][pydantic]{:target="_blank"} or a string that contains a [JSON Schema][jsonschema]{:target="_blank"}:
-
-=== "Pydantic"
+	Example: using pydantic to enforce a JSON schema
 
     ```python
     from enum import Enum
     from pydantic import BaseModel, constr, conint
 
-    import outlines
-
-    class Armor(str, Enum):
-        leather = "leather"
-        chainmail = "chainmail"
-        plate = "plate"
-
-
     class Character(BaseModel):
         name: constr(max_length=10)
         age: conint(gt=18, lt=99)
-        armor: Armor
+        armor: (Enum('Armor', {'leather': 'leather', 'chainmail': 'chainmail', 'plate': 'plate'}))
         strength: conint(gt=1, lt=100)
 
-    model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
     generator = outlines.generate.json(model, Character)
 
     character = generator(
@@ -104,135 +74,10 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
         + "name, age (between 1 and 99), armor and strength. "
         )
     print(character)
-    # name='Orla' age=21 armor=<Armor.plate: 'plate'> strength=8
-    ```
-
-=== "JSON Schema"
-
-    ```python
-    import outlines
-
-    schema = """{
-        "$defs": {
-            "Armor": {
-                "enum": ["leather", "chainmail", "plate"],
-                "title": "Armor",
-                "type": "string"
-            }
-        },
-        "properties": {
-            "name": {"maxLength": 10, "title": "Name", "type": "string"},
-            "age": {"title": "Age", "type": "integer"},
-            "armor": {"$ref": "#/$defs/Armor"},
-            "strength": {"title": "Strength", "type": "integer"}\
-        },
-        "required": ["name", "age", "armor", "strength"],
-        "title": "Character",
-        "type": "object"
-    }"""
-
-    model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
-    generator = outlines.generate.json(model, schema)
-    character = generator(
-        "Generate a new character for my awesome game: "
-        + "name, age (between 1 and 99), armor and strength. "
-        )
-    print(character)
-    # {'name': 'Yuki', 'age': 24, 'armor': 'plate', 'strength': 3}
-    ```
-
-!!! Note
-
-    We advise you to constrain the length of the strings fields when first testing your schema, especially with small models.
+    # Character(name='Zara', age=25, armor=<Armor.leather: 'leather'>, strength=85)
+	```
 
-### Grammar-structured generation
-
-Outlines also allows to generate text that is valid to any [context-free grammar][cfg]{:target="_blank"} (CFG) in the [EBNF format][ebnf]{:target="_blank"}. Grammars can be intimidating, but they are a very powerful tool! Indeed, they determine the syntax of every programming language, valid chess moves, molecule structure, can help with procedural graphics generation, etc.
-
-Here we show a simple example of a grammar that defines arithmetic operations:
-
-```python
-from outlines import models, generate
-
-arithmetic_grammar = """
-    ?start: sum
-
-    ?sum: product
-        | sum "+" product   -> add
-        | sum "-" product   -> sub
-
-    ?product: atom
-        | product "*" atom  -> mul
-        | product "/" atom  -> div
-
-    ?atom: NUMBER           -> number
-         | "-" atom         -> neg
-         | "(" sum ")"
-
-    %import common.NUMBER
-    %import common.WS_INLINE
-
-    %ignore WS_INLINE
-"""
-
-model = models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
-generator = generate.cfg(model, arithmetic_grammar, max_tokens=100)
-
-result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
-print(result)
-# 5+5+5+5+5
-```
-
-
-EBNF grammars can be cumbersome to write. This is why Outlines provides grammar definitions in the `outlines.grammars.` module
-
-```python
-from outlines import models, generate, grammars
-
-model = models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
-generator = generate.cfg(model, grammars.arithmetic, max_tokens=100)
-
-result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
-print(result)
-# 5+5+5+5+5
-```
-
-The available grammars are listed [here](https://github.com/outlines-dev/outlines/tree/main/outlines/grammars).
-
-
-### Regex-structured generation
-
-Slightly simpler, but no less useful, Outlines can generate text that is in the language of a [regular expression](https://www.regular-expressions.info/tutorial.html). For instance to force the model to generate IP addresses:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
-
-regex_str = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-generator = generate.regex(model, regex_str)
-
-result = generator("What is the IP address of localhost?\nIP: ")
-print(result)
-# 127.0.0.100
-```
-
-### Generate a given Python type
-
-We provide a shortcut to regex-structured generation for simple use cases. Pass a Python type to the `outlines.generate.format` function and the LLM will output text that matches this type:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
-generator = generate.format(model, int)
-
-result = generator("What is 2+2?")
-print(result)
-# 4
-```
-
-## Deploy using vLLM and FastAPI
+## [Deploy using vLLM and FastAPI](./reference/serve/vllm.md)
 
 Outlines can be deployed as a LLM service using [vLLM][vllm]{:target="_blank"} and [FastAPI][fastapi]{:target="_blank"}. The server supports asynchronous processing of incoming requests, and benefits from the performance of vLLM.
 
@@ -265,7 +110,7 @@ Or use the [requests][requests]{:target="_blank"} library from another python pr
 
 ## Utilities
 
-### Prompt templates
+### [Prompt templates](./reference/prompting.md)
 
 Prompting can lead to messy code. Outlines' prompt functions are python functions that contain a template for the prompt in their docstring. We use a powerful templating language to allow you to loop over lists, dictionaries, add conditionals, etc. directly from the prompt. When called, a prompt function returns the rendered template:
 
@@ -368,6 +213,7 @@ Once you are done experimenting with a prompt and an output structure, it is use
     ```
     It make it easier for the community to collaborate on the infinite number of use cases enabled by these models!
 
+
 ## Going further
 
 If you need more inspiration you can take a look at the [cookbook](cookbook/index.md). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/outlines-dev/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
diff --git a/docs/reference/cfg.md b/docs/reference/generation/cfg.md
similarity index 100%
rename from docs/reference/cfg.md
rename to docs/reference/generation/cfg.md
diff --git a/docs/reference/choices.md b/docs/reference/generation/choices.md
similarity index 100%
rename from docs/reference/choices.md
rename to docs/reference/generation/choices.md
diff --git a/docs/reference/custom_fsm_ops.md b/docs/reference/generation/custom_fsm_ops.md
similarity index 100%
rename from docs/reference/custom_fsm_ops.md
rename to docs/reference/generation/custom_fsm_ops.md
diff --git a/docs/reference/format.md b/docs/reference/generation/format.md
similarity index 100%
rename from docs/reference/format.md
rename to docs/reference/generation/format.md
diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
new file mode 100644
index 000000000..88b963c72
--- /dev/null
+++ b/docs/reference/generation/generation.md
@@ -0,0 +1,210 @@
+---
+title: Generation
+---
+
+# Generation
+
+Once an [Outlines model](../models) is constructed you can use `outlines.generate` to generate text. Standard LLM generation is possible via `outlines.generate.text`, along with a variety of structured generation methods described below. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
+
+Before generating text, you must construct an `outlines.model`. Example:
+
+```python
+import outlines
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+```
+### Text generator
+
+```python
+generator = outlines.generate.text(model)
+
+result = generator("Question: What's 2+2? Answer:", max_tokens=100)
+print(result)
+# The answer is 4
+
+# Outlines also supports streaming output
+stream = generator.stream("What's 2+2?", max_tokens=4)
+for i in range(5):
+	token = next(stream)
+	print(repr(token))
+# '2'
+# '+'
+# '2'
+# ' equals'
+# '4'
+```
+
+### [Multi-label classification](./choices.md)
+
+Outlines allows you to do multi-label classification by guiding the model so it can only output either of the specified choices:
+
+```python
+import outlines
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = outlines.generate.choice(model, ["Blue", "Red", "Yellow"])
+
+color = generator("What is the closest color to Indigo? ")
+print(color)
+# Blue
+```
+
+### [JSON-structured generation](./json.md)
+
+Outlines can guide models so that they output valid JSON **100%** of the time. You can either specify the structure using [Pydantic][pydantic]{:target="_blank"} or a string that contains a [JSON Schema][jsonschema]{:target="_blank"}:
+
+=== "Pydantic"
+
+    ```python
+    from enum import Enum
+    from pydantic import BaseModel, constr, conint
+
+    import outlines
+
+    class Armor(str, Enum):
+        leather = "leather"
+        chainmail = "chainmail"
+        plate = "plate"
+
+
+    class Character(BaseModel):
+        name: constr(max_length=10)
+        age: conint(gt=18, lt=99)
+        armor: Armor
+        strength: conint(gt=1, lt=100)
+
+    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+    generator = outlines.generate.json(model, Character)
+
+    character = generator(
+        "Generate a new character for my awesome game: "
+        + "name, age (between 1 and 99), armor and strength. "
+        )
+    print(character)
+    # name='Orla' age=21 armor=<Armor.plate: 'plate'> strength=8
+    ```
+
+=== "JSON Schema"
+
+    ```python
+    import outlines
+
+    schema = """{
+        "$defs": {
+            "Armor": {
+                "enum": ["leather", "chainmail", "plate"],
+                "title": "Armor",
+                "type": "string"
+            }
+        },
+        "properties": {
+            "name": {"maxLength": 10, "title": "Name", "type": "string"},
+            "age": {"title": "Age", "type": "integer"},
+            "armor": {"$ref": "#/$defs/Armor"},
+            "strength": {"title": "Strength", "type": "integer"}\
+        },
+        "required": ["name", "age", "armor", "strength"],
+        "title": "Character",
+        "type": "object"
+    }"""
+
+    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+    generator = outlines.generate.json(model, schema)
+    character = generator(
+        "Generate a new character for my awesome game: "
+        + "name, age (between 1 and 99), armor and strength. "
+        )
+    print(character)
+    # {'name': 'Yuki', 'age': 24, 'armor': 'plate', 'strength': 3}
+    ```
+
+!!! Note
+
+    We advise you to constrain the length of the strings fields when first testing your schema, especially with small models.
+
+### [Grammar-structured generation](./cfg.md)
+
+Outlines also allows to generate text that is valid to any [context-free grammar][cfg]{:target="_blank"} (CFG) in the [EBNF format][ebnf]{:target="_blank"}. Grammars can be intimidating, but they are a very powerful tool! Indeed, they determine the syntax of every programming language, valid chess moves, molecule structure, can help with procedural graphics generation, etc.
+
+Here we show a simple example of a grammar that defines arithmetic operations:
+
+```python
+from outlines import models, generate
+
+arithmetic_grammar = """
+    ?start: sum
+
+    ?sum: product
+        | sum "+" product   -> add
+        | sum "-" product   -> sub
+
+    ?product: atom
+        | product "*" atom  -> mul
+        | product "/" atom  -> div
+
+    ?atom: NUMBER           -> number
+         | "-" atom         -> neg
+         | "(" sum ")"
+
+    %import common.NUMBER
+    %import common.WS_INLINE
+
+    %ignore WS_INLINE
+"""
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = generate.cfg(model, arithmetic_grammar, max_tokens=100)
+
+result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
+print(result)
+# 5+5+5+5+5
+```
+
+
+EBNF grammars can be cumbersome to write. This is why Outlines provides grammar definitions in the `outlines.grammars.` module
+
+```python
+from outlines import models, generate, grammars
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = generate.cfg(model, grammars.arithmetic, max_tokens=100)
+
+result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
+print(result)
+# 5+5+5+5+5
+```
+
+The available grammars are listed [here](https://github.com/outlines-dev/outlines/tree/main/outlines/grammars).
+
+
+### [Regex-structured generation](./regex.md)
+
+Slightly simpler, but no less useful, Outlines can generate text that is in the language of a [regular expression](https://www.regular-expressions.info/tutorial.html). For instance to force the model to generate IP addresses:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+
+regex_str = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
+generator = generate.regex(model, regex_str)
+
+result = generator("What is the IP address of localhost?\nIP: ")
+print(result)
+# 127.0.0.100
+```
+
+### [Generate a given Python type](./types.md)
+
+We provide a shortcut to regex-structured generation for simple use cases. Pass a Python type to the `outlines.generate.format` function and the LLM will output text that matches this type:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = generate.format(model, int)
+
+result = generator("What is 2+2?")
+print(result)
+# 4
+```
diff --git a/docs/reference/json.md b/docs/reference/generation/json.md
similarity index 98%
rename from docs/reference/json.md
rename to docs/reference/generation/json.md
index 85e1a846a..a342395bd 100644
--- a/docs/reference/json.md
+++ b/docs/reference/generation/json.md
@@ -70,7 +70,8 @@ schema = """
     "name": {"type": "string"},
     "last_name": {"type": "string"},
     "id": {"type": "integer"}
-  }
+  },
+  "required": ["name", "last_name", "id"]
 }
 """
 
diff --git a/docs/reference/regex.md b/docs/reference/generation/regex.md
similarity index 100%
rename from docs/reference/regex.md
rename to docs/reference/generation/regex.md
diff --git a/docs/reference/generation/structured_generation_explanation.md b/docs/reference/generation/structured_generation_explanation.md
new file mode 100644
index 000000000..0dedf060b
--- /dev/null
+++ b/docs/reference/generation/structured_generation_explanation.md
@@ -0,0 +1,58 @@
+---
+title: Structured Generation Explanation
+---
+
+# Structured Generation Explanation
+
+
+Language models generate text token by token, using the previous token sequence as input and sampled logits as output. This document explains the structured generation process, where only legal tokens are considered for the next step based on a predefined automata, e.g. a regex-defined [finite-state machine](https://en.wikipedia.org/wiki/Finite-state_machine) (FSM) or [Lark](https://lark-parser.readthedocs.io/en/stable/) grammar.`
+
+
+## Worked Example
+
+Let's consider a worked example with a pattern for whole and decimal numbers:
+
+`^\d*(\.\d+)?$`.
+
+### Creating Automata
+
+The pattern is first converted into an automata. Below is a brief explanation of the automata conversion and its representation.
+
+**Automata Diagram:**
+
+```mermaid
+graph LR
+    node0("1-9") --> node1("1-9")
+    node1 --> node1
+    node1 --> nodeEND{{END}}
+    node1 --> nodePeriod(".")
+    nodePeriod --> node2("1-9")
+    node2 --> node2
+    node2 --> nodeEND{{END}}
+```
+
+### Generating a Token
+
+Let's assume that we're in the middle of generation, and so far "748" has been generated. Here is the automata with the current state highlighted in green, with the legal next characters being another number (1-9), a dot (.), or end of sequence.
+
+```mermaid
+graph LR
+    node0("1-9") --> node1("1-9")
+    node1 --> node1
+    node1 --> nodeEND{{END}}
+    node1 --> nodePeriod(".")
+    nodePeriod --> node2("1-9")
+    node2 --> node2
+    node2 --> nodeEND{{END}}
+
+    style node1 fill:#090
+```
+
+Generating a token requires the following steps:
+
+- Feed the previous input sequence ("748") into the language model.
+- Language model runs a forward pass and produces token logits.
+- Outlines logits processor sets the probability of illegal tokens to 0%.
+- A token is sampled from the set of legal tokens.
+
+![Generation and Logits Processing Flow Chart](../../assets/images/logits_processing_diagram.svg)
diff --git a/docs/reference/types.md b/docs/reference/generation/types.md
similarity index 100%
rename from docs/reference/types.md
rename to docs/reference/generation/types.md
diff --git a/docs/reference/json_mode.md b/docs/reference/json_mode.md
deleted file mode 100644
index bb1c1c7a8..000000000
--- a/docs/reference/json_mode.md
+++ /dev/null
@@ -1,17 +0,0 @@
-# JSON mode
-
-Outlines can guarantee that the LLM will generate valid JSON, using [Grammar-structured generation](cfg.md):
-
-```python
-from outlines import models, generate
-
-json_grammar = outlines.grammars.json
-
-model = models.transformers("mistralai/Mistral-7b-v0.1")
-generator = generate.cfg(model, json_grammar)
-sequence = generator("Generate valid JSON")
-```
-
-!!! Note "JSON that follows a schema"
-
-    If you want to guarantee that the generated JSON follows a given schema, consult [this section](json.md) instead.
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
new file mode 100644
index 000000000..dadfd34ad
--- /dev/null
+++ b/docs/reference/models/models.md
@@ -0,0 +1,93 @@
+---
+title: Models
+---
+
+# Models
+
+Outlines supports generation using a number of inference engines (`outlines.models`)
+
+Loading a model using outlines follows a similar interface between inference engines.
+
+```python
+import outlines
+```
+
+## [Transformers](./transformers.md)
+
+```python
+model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct", model_kwargs={})
+```
+
+For additional arguments and use of other Huggingface Transformers model types see [Outlines' Transformers documentation](./transformers.md).
+
+
+## [Transformers Vision](./transformers_vision.md)
+
+```python
+model = outlines.models.transformers_vision("llava-hf/llava-v1.6-mistral-7b-hf")
+```
+
+For examples of generation and other details, see [Outlines' Transformers Vision documentation](./transformers_vision.md).
+
+## [vLLM](./vllm.md)
+
+```python
+model = outlines.models.vllm("microsoft/Phi-3-mini-128k-instruct")
+```
+
+## [llama.cpp](./llamacpp.md)
+
+```python
+model = outlines.models.llamacpp("microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf")
+```
+
+Additional llama.cpp parameters can be found in the [Outlines' llama.cpp documentation](./llamacpp.md).
+
+## [ExLlamaV2](./exllamav2.md)
+
+```python
+model = outlines.models.exllamav2("bartowski/Phi-3-mini-128k-instruct-exl2")
+```
+
+## [MLXLM](./mlxlmx.md)
+
+```python
+model = outlines.models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
+```
+
+## [OpenAI](./openai.md)
+
+```python
+model = outlines.models.openai(
+    "gpt-4o-mini",
+    api_key=os.environ["OPENAI_API_KEY"]
+)
+```
+
+
+# Feature Matrix
+|                   | Transformers | Transformers Vision | vLLM | llama.cpp | ExLlamaV2 | MLXLM | OpenAI* |
+|-------------------|--------------|---------------------|------|-----------|-----------|-------|---------|
+| **Device**        |              |                     |      |           |           |       |         |
+| Cuda              | ✅           | ✅                  | ✅   | ✅        | ✅        | ❌    | N/A     |
+| Apple Silicon     | ✅           | ✅                  | ❌   | ✅        | ✅        | ✅    | N/A     |
+| x86 / AMD64       | ✅           | ✅                  | ❌   | ✅        | ✅        | ❌    | N/A     |
+| **Sampling**      |              |                     |      |           |           |       |         |
+| Greedy            | ✅           | ✅                  | ✅   | ✅*       | ✅        | ✅    | ❌      |
+| Multinomial       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Multiple Samples  | ✅           | ✅                  |      | ❌        |           | ❌    | ✅      |
+| Beam Search       | ✅           | ✅                  | ✅   | ❌        | ✅        | ❌    | ❌      |
+| **Generation**    |              |                     |      |           |           |       |         |
+| Batch             | ✅           | ✅                  | ✅   | ❌        | ?         | ❌    | ❌      |
+| Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
+| **`outlines.generate`** |        |                     |      |           |           |       |         |
+| Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Structured*       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
+
+
+## Caveats
+
+- OpenAI doesn't support structured generation due to limitations in their API and server implementation.
+- `outlines.generate` ["Structured"](../generation/generation.md) includes methods such as `outlines.generate.regex`, `outlines.generate.json`, `outlines.generate.cfg`, etc.
+- MLXLM only supports Apple Silicon.
+- llama.cpp greedy sampling available via multinomial with `temperature = 0.0`.
diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
index b0ad6533b..e878f5866 100644
--- a/docs/reference/models/transformers_vision.md
+++ b/docs/reference/models/transformers_vision.md
@@ -2,9 +2,10 @@
 
 Outlines allows seamless use of [vision models](https://huggingface.co/learn/computer-vision-course/en/unit4/multimodal-models/tasks-models-part1).
 
-`outlines.models.transformers_vision` has shares interfaces with, and is based on [`outlines.models.transformers`](./transformers.md).
+`outlines.models.transformers_vision` has shares interfaces with, and is based on [outlines.models.transformers](./transformers.md).
 
 Tasks supported include
+
 - image + text -> text
 - video + text -> text
 
diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index fdf3acb9e..4078215af 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -9,6 +9,10 @@
   --md-code-fg-color: #FFFFFF;
   --md-text-font-family: "Inter";
   --md-code-font: "Source Code Pro Custom";
+
+  /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
+  --md-mermaid-label-fg-color: #000000;
+  --md-mermaid-edge-color: #000000;
 }
 
 .index-pre-code {
diff --git a/mkdocs.yml b/mkdocs.yml
index e5f3417ac..d24ca9a63 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -58,7 +58,11 @@ markdown_extensions:
       pygments_lang_class: true
       noclasses: True
       pygments_style: nord
-  - pymdownx.superfences
+  - pymdownx.superfences:
+      custom_fences:
+        - name: mermaid
+          class: mermaid
+          format: !!python/name:pymdownx.superfences.fence_code_format
   - pymdownx.tabbed:
       alternate_style: true
   - pymdownx.inlinehilite
@@ -117,24 +121,27 @@ nav:
   - Docs:
     - reference/index.md
     - Generation:
+        - Generation Overview: reference/generation/generation.md
         - Text: reference/text.md
         - Samplers: reference/samplers.md
-    - Structured generation:
-        - Classification: reference/choices.md
-        - Regex: reference/regex.md
-        - Type constraints: reference/format.md
-        - JSON (function calling): reference/json.md
-        - JSON mode: reference/json_mode.md
-        - Grammar: reference/cfg.md
-        - Custom FSM operations: reference/custom_fsm_ops.md
+        - Structured generation:
+            - Classification: reference/generation/choices.md
+            - Regex: reference/generation/regex.md
+            - Type constraints: reference/generation/format.md
+            - JSON (function calling): reference/generation/json.md
+            - Grammar: reference/generation/cfg.md
+            - Custom FSM operations: reference/generation/custom_fsm_ops.md
+            - Structured Generation Technical Explanation: reference/generation/structured_generation_explanation.md
     - Utilities:
         - Serve with vLLM: reference/serve/vllm.md
         - Custom types: reference/types.md
         - Prompt templating: reference/prompting.md
         - Outlines functions: reference/functions.md
     - Models:
+        - Models Overview: reference/models/models.md
         - Open source:
           - Transformers: reference/models/transformers.md
+          - Transformers Vision: reference/models/transformers_vision.md
           - Llama.cpp: reference/models/llamacpp.md
           - vLLM: reference/models/vllm.md
           - TGI: reference/models/tgi.md
diff --git a/requirements-doc.txt b/requirements-doc.txt
index b7bd2efbb..cf8c674a7 100644
--- a/requirements-doc.txt
+++ b/requirements-doc.txt
@@ -1,6 +1,7 @@
 mkdocs
 mkdocs-material
 mkdocs-material[imaging]
+mkdocs-mermaid2-plugin
 mkdocs-section-index
 mkdocstrings[python]
 mkdocs-git-committers-plugin-2

From b10c78fe08142dd6930f1d3b030f7dd5d1a651c6 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 27 Jul 2024 12:08:57 -0500
Subject: [PATCH 037/505] don't use gated models (mistral) in
 outlines-supported documentation

---
 README.md                                       | 16 +++++-----------
 docs/cookbook/deploy-using-cerebrium.md         |  2 +-
 docs/cookbook/index.md                          |  2 +-
 docs/cookbook/models_playing_chess.md           |  2 +-
 docs/cookbook/structured_generation_workflow.md |  4 ++--
 docs/quickstart.md                              |  6 +++---
 docs/reference/generation/choices.md            |  2 +-
 docs/reference/generation/custom_fsm_ops.md     |  2 +-
 docs/reference/generation/format.md             |  2 +-
 docs/reference/generation/json.md               |  6 +++---
 docs/reference/generation/regex.md              |  2 +-
 docs/reference/generation/types.md              |  4 ++--
 docs/reference/models/transformers.md           |  2 +-
 docs/reference/models/vllm.md                   | 10 +++++-----
 docs/reference/samplers.md                      | 10 +++++-----
 docs/reference/serve/vllm.md                    |  6 +++---
 docs/reference/text.md                          | 10 +++++-----
 docs/welcome.md                                 |  1 -
 18 files changed, 41 insertions(+), 48 deletions(-)

diff --git a/README.md b/README.md
index 96d02dbe1..364e77244 100644
--- a/README.md
+++ b/README.md
@@ -66,12 +66,6 @@ is to ensure that there is a well-defined interface between their output and
 user-defined code. **Outlines** provides ways to control the generation of
 language models to make their output more predictable.
 
-Before using mistral models, request access on huggingface [here](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2).
-``` python
-# login to access mistral model
-from huggingface_hub import login
-login()
-```
 ### Multiple choices
 
 You can reduce the completion to a choice between multiple possibilities:
@@ -79,7 +73,7 @@ You can reduce the completion to a choice between multiple possibilities:
 ``` python
 import outlines
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 prompt = """You are a sentiment-labelling assistant.
 Is the following review positive or negative?
@@ -122,7 +116,7 @@ hood:
 ``` python
 import outlines
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 prompt = "What is the IP address of the Google DNS servers? "
 
@@ -185,7 +179,7 @@ class Character(BaseModel):
     strength: int
 
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 # Construct structured sequence generator
 generator = outlines.generate.json(model, Character)
@@ -250,7 +244,7 @@ schema = '''{
     }
 }'''
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = outlines.generate.json(model, schema)
 character = generator("Give me a character description")
 ```
@@ -338,7 +332,7 @@ def labelling(to_label, examples):
     {{ to_label }} //
     """
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 prompt = labelling("Just awesome", examples)
 answer = outlines.generate.text(model)(prompt, max_tokens=100)
 ```
diff --git a/docs/cookbook/deploy-using-cerebrium.md b/docs/cookbook/deploy-using-cerebrium.md
index b7609424c..6fbaab390 100644
--- a/docs/cookbook/deploy-using-cerebrium.md
+++ b/docs/cookbook/deploy-using-cerebrium.md
@@ -50,7 +50,7 @@ Running code in Cerebrium is like writing normal python with no special syntax.
 import outlines
 
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 schema = """{
     "title": "Character",
diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index 56334302a..58e84ae96 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -4,7 +4,7 @@
 - [Named Entity Extraction](extraction.md): Extract information from pizza orders.
 - [Dating Profile](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
 - [Chain Of Density](chain_of_density.md): Summarize documents using chain of density prompting and JSON-structured generation.
-- [Playing Chess](models_playing_chess.md): Make Mistral-7B play chess against itself using regex-structured generation.
+- [Playing Chess](models_playing_chess.md): Make Phi-3 Mini play chess against itself using regex-structured generation.
 - [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
 - [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
diff --git a/docs/cookbook/models_playing_chess.md b/docs/cookbook/models_playing_chess.md
index 3b4f10865..a0df2eb2d 100644
--- a/docs/cookbook/models_playing_chess.md
+++ b/docs/cookbook/models_playing_chess.md
@@ -23,7 +23,7 @@ Phi-2 will be playing against itself:
 ```python
 from outlines import models
 
-model = models.transformers("microsoft/phi-2")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 ```
 
diff --git a/docs/cookbook/structured_generation_workflow.md b/docs/cookbook/structured_generation_workflow.md
index c627ab9a8..b9d7c28a4 100644
--- a/docs/cookbook/structured_generation_workflow.md
+++ b/docs/cookbook/structured_generation_workflow.md
@@ -12,7 +12,7 @@ Before diving into how to use structure generation for this task let's start wit
 ```python
 import outlines
 
-model_name = 'mistralai/Mistral-7B-Instruct-v0.2'
+model_name = 'microsoft/Phi-3-mini-4k-instruct'
 model = outlines.models.transformers(model_name)
 ```
 
@@ -31,7 +31,7 @@ messages_phone = [
 ]
 
 # This allows us to properly format our prompt for
-# Mistrals 'Instruct' interface.
+# Phi-3 Mini's 'Instruct' interface.
 prompt_phone = tokenizer.apply_chat_template(messages_phone, tokenize=False)
 ```
 
diff --git a/docs/quickstart.md b/docs/quickstart.md
index fcb524d8a..8ffffd2ed 100644
--- a/docs/quickstart.md
+++ b/docs/quickstart.md
@@ -84,13 +84,13 @@ Outlines can be deployed as a LLM service using [vLLM][vllm]{:target="_blank"} a
 First start the server:
 
 ```python
-python -m outlines.serve.serve --model="mistralai/Mistral-7B-Instruct-v0.2"
+python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
 ```
 
 Or you can start the server with Outlines' official Docker image:
 
 ```bash
-docker run -p 8000:8000 outlinesdev/outlines --model="mistralai/Mistral-7B-Instruct-v0.2"
+docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
 ```
 
 This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used.
@@ -186,7 +186,7 @@ Once you are done experimenting with a prompt and an output structure, it is use
     generate_joke = outlines.Function(
         tell_a_joke,
         Joke,
-        "mistralai/Mistral-7B-Instruct-v0.2"
+        "microsoft/Phi-3-mini-4k-instruct"
     )
     ```
 
diff --git a/docs/reference/generation/choices.md b/docs/reference/generation/choices.md
index 275b4a9a0..aed5af5ab 100644
--- a/docs/reference/generation/choices.md
+++ b/docs/reference/generation/choices.md
@@ -5,7 +5,7 @@ Oultines allows you to make sure the generated text is chosen between different
 ```python
 from outlines import models, generate
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.choice(model, ["skirt", "dress", "pen", "jacket"])
 answer = generator("Pick the odd word out: skirt, dress, pen, jacket")
 
diff --git a/docs/reference/generation/custom_fsm_ops.md b/docs/reference/generation/custom_fsm_ops.md
index 22f5ca6bb..5c4be96fa 100644
--- a/docs/reference/generation/custom_fsm_ops.md
+++ b/docs/reference/generation/custom_fsm_ops.md
@@ -29,7 +29,7 @@ difference_fsm_fsm.accepts('["a","blue","donkey"]')
 # True
 
 
-model = models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.fsm(model, difference_fsm)
 response = generator("Don't talk about pink elephants")
 ```
diff --git a/docs/reference/generation/format.md b/docs/reference/generation/format.md
index 52b0dc5ec..749baa8b6 100644
--- a/docs/reference/generation/format.md
+++ b/docs/reference/generation/format.md
@@ -5,7 +5,7 @@ We can ask completions to be restricted to valid python types:
 ```python
 from outlines import models, generate
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.format(model, int)
 answer = generator("When I was 6 my sister was half my age. Now I’m 70 how old is my sister?")
 print(answer)
diff --git a/docs/reference/generation/json.md b/docs/reference/generation/json.md
index a342395bd..da9f14729 100644
--- a/docs/reference/generation/json.md
+++ b/docs/reference/generation/json.md
@@ -25,7 +25,7 @@ class User(BaseModel):
     id: int
 
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.json(model, User)
 result = generator(
     "Create a user profile with the fields name, last_name and id"
@@ -60,7 +60,7 @@ from pydantic import BaseModel
 from outlines import models
 from outlines import generate
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 schema = """
 {
@@ -94,7 +94,7 @@ from outlines import generate
 def add(a: int, b: int):
     return a + b
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.json(model, add)
 result = generator("Return two integers named a and b respectively. a is odd and b even.")
 
diff --git a/docs/reference/generation/regex.md b/docs/reference/generation/regex.md
index 093dd0e18..23df35567 100644
--- a/docs/reference/generation/regex.md
+++ b/docs/reference/generation/regex.md
@@ -5,7 +5,7 @@ Outlines can guarantee that the text generated by the LLM will be valid to a reg
 ```python
 from outlines import models, generate
 
-model = models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 generator = generate.regex(
     model,
diff --git a/docs/reference/generation/types.md b/docs/reference/generation/types.md
index 2e02f45f1..5b83a5916 100644
--- a/docs/reference/generation/types.md
+++ b/docs/reference/generation/types.md
@@ -46,7 +46,7 @@ class Client(BaseModel):
     zip_code: locale.ZipCode
 
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.json(model, Client)
 result = generator(
     "Create a client profile with the fields name, phone_number and zip_code"
@@ -63,7 +63,7 @@ from pydantic import BaseModel
 from outlines import models, generate, types
 
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.format(model, types.locale("us").PhoneNumber)
 result = generator(
     "Return a US Phone number: "
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 80b70845f..15eabb682 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -15,7 +15,7 @@ Outlines provides an integration with the `torch` implementation of causal model
 ```python
 from outlines import models
 
-model = models.transformers("mistralai/Mistral-7B-v0.3", device="cuda")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
 ```
 
 If you need more fine-grained control you can also initialize the model and tokenizer separately:
diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index 735ab4cc5..1380d3d2c 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -13,7 +13,7 @@ Outlines supports models available via vLLM's offline batched inference interfac
 ```python
 from outlines import models
 
-model = models.vllm("mistralai/Mistral-7B-v0.1")
+model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
 ```
 
 Or alternatively:
@@ -22,7 +22,7 @@ Or alternatively:
 import vllm
 from outlines import models
 
-llm = vllm.LLM("mistralai/Mistral-7B-v0.1")
+llm = vllm.LLM("microsoft/Phi-3-mini-4k-instruct")
 model = models.VLLM(llm)
 ```
 
@@ -41,7 +41,7 @@ You can pass any parameter that you would normally pass to `vllm.LLM`, as keywor
 from outlines import models
 
 model = models.vllm(
-    "mistralai/Mistral-7B-v0.1",
+    "microsoft/Phi-3-mini-4k-instruct",
     trust_remote_code=True,
     gpu_memory_utilization=0.7
 )
@@ -94,7 +94,7 @@ To run multi-GPU inference with vLLM you need to set the `tensor_parallel_size`
 from outlines import models
 
 model = models.vllm(
-    "mistralai/Mistral-7B-v0.1"
+    "microsoft/Phi-3-mini-4k-instruct"
     tensor_parallel_size=2
 )
 ```
@@ -120,7 +120,7 @@ from vllm.sampling_params import SamplingParams
 from outlines import models, generate
 
 
-model = models.vllm("mistralai/Mistral-7b-v0.1")
+model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.text(model)
 
 params = SamplingParams(n=2, frequency_penalty=1., min_tokens=2)
diff --git a/docs/reference/samplers.md b/docs/reference/samplers.md
index a0f652bbf..56e4e1f7e 100644
--- a/docs/reference/samplers.md
+++ b/docs/reference/samplers.md
@@ -10,7 +10,7 @@ Outlines defaults to the multinomial sampler without top-p or top-k sampling, an
 from outlines import models, generate, samplers
 
 
-model = models.transformers("mistralai/Mistral-7B-0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 sampler = samplers.multinomial()
 
 generator = generate.text(model, sampler)
@@ -26,7 +26,7 @@ You can ask the generator to take multiple samples by passing the number of samp
 from outlines import models, generate, samplers
 
 
-model = models.transformers("mistralai/Mistral-7B-0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 sampler = samplers.multinomial(3)
 
 generator = generate.text(model, sampler)
@@ -42,7 +42,7 @@ If you ask multiple samples for a batch of prompt the returned array will be of
 from outlines import models, generate, samplers
 
 
-model = models.transformers("mistralai/Mistral-7B-0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 sampler = samplers.multinomial(3)
 
 generator = generate.text(model, sampler)
@@ -78,7 +78,7 @@ You can also use the greedy sampler. For this you need to initialize the generat
 from outlines import models, generate, samplers
 
 
-model = models.transformers("mistralai/Mistral-7B-0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 sampler = samplers.greedy()
 
 generator = generate.text(model, sampler)
@@ -99,7 +99,7 @@ Outlines also comes with the Beam Search sampling algorithm:
 from outlines import models, generate, samplers
 
 
-model = models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 sampler = samplers.beam_search(beams=5)
 
 generator = generate.text(model, sampler)
diff --git a/docs/reference/serve/vllm.md b/docs/reference/serve/vllm.md
index 1b4d4bf14..0b584568b 100644
--- a/docs/reference/serve/vllm.md
+++ b/docs/reference/serve/vllm.md
@@ -13,7 +13,7 @@ pip install outlines[serve]
 You can then start the server with:
 
 ```bash
-python -m outlines.serve.serve --model="mistralai/Mistral-7B-Instruct-v0.2"
+python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
 ```
 
 This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used. The `--model` argument allows you to specify any model of your choosing.
@@ -22,7 +22,7 @@ To run inference on multiple GPUs you must pass the `--tensor-parallel-size` arg
 
 
 ```bash
-python -m outlines.serve.serve --model="mistralai/Mistral-7B-Instruct-v0.2" --tensor-parallel-size 2
+python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct" --tensor-parallel-size 2
 ```
 
 
@@ -31,7 +31,7 @@ python -m outlines.serve.serve --model="mistralai/Mistral-7B-Instruct-v0.2" --te
 You can install and run the server with Outlines' official Docker image using the command
 
 ```bash
-docker run -p 8000:8000 outlinesdev/outlines --model="mistralai/Mistral-7B-Instruct-v0.2"
+docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
 ```
 
 ## Querying Endpoint
diff --git a/docs/reference/text.md b/docs/reference/text.md
index 423a26ba6..3d0d130f1 100644
--- a/docs/reference/text.md
+++ b/docs/reference/text.md
@@ -13,7 +13,7 @@ model = models.openai("gpt-4")
 generator = generate.text(model)
 answer = generator("What is 2+2?")
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.text(model)
 answer = generator("What is 2+2?")
 ```
@@ -28,7 +28,7 @@ Outlines allows you to stream the model's response by calling the `.stream` meth
 ```python
 from outlines import models, generate
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.text(model)
 
 tokens = generator.stream("What is 2+2?")
@@ -45,7 +45,7 @@ To limit the number of tokens generated you can pass the `max_tokens` positional
 ```python
 from outlines import models, generate
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.text(model)
 
 answer = generator("What is 2+2?", 5)
@@ -60,7 +60,7 @@ You can also ask the model to stop generating text after a given string has been
 ```python
 from outlines import models, generate
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 generator = generate.text(model)
 
 answer = generator("What is 2+2?", stop_at=".")
@@ -78,7 +78,7 @@ It can be useful to seed the generation in order to get reproducible results:
 import torch
 from outlines import models, generate
 
-model = models.transformers("mistralai/Mistral-7B-v0.1")
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 seed = 789001
 
diff --git a/docs/welcome.md b/docs/welcome.md
index 80632aff6..4c327c020 100644
--- a/docs/welcome.md
+++ b/docs/welcome.md
@@ -170,7 +170,6 @@ Outlines would not be what it is today without a community of dedicated develope
 Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
 
 
-
 [discord]: https://discord.gg/R9DSu34mGd
 [aesara]: https://github.com/aesara-devs
 [blackjax]: https://github.com/blackjax-devs/blackjax

From 7218cdd0a5e69ac4267b9a17a9d900b41d95c83b Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 27 Jul 2024 12:10:37 -0500
Subject: [PATCH 038/505] =?UTF-8?q?Add=20Remi's=20AI=20Engineer=20World?=
 =?UTF-8?q?=E2=80=99s=20Fair=20Presentation?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 docs/quickstart.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/quickstart.md b/docs/quickstart.md
index 8ffffd2ed..2e1f9a6bb 100644
--- a/docs/quickstart.md
+++ b/docs/quickstart.md
@@ -216,7 +216,7 @@ Once you are done experimenting with a prompt and an output structure, it is use
 
 ## Going further
 
-If you need more inspiration you can take a look at the [cookbook](cookbook/index.md). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/outlines-dev/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
+If you need more inspiration you can take a look at the [cookbook](cookbook/index.md) or watch [Remi Louf's AI Engineer World’s Fair Presentation on Outlines](https://www.youtube.com/live/R0X7mPagRiE?t=775s). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/outlines-dev/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
 
 
 [pydantic]: https://docs.pydantic.dev/latest

From 0c9f5bb3ce60825735c3d2d4b6952a0dbc7d11f8 Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Tue, 30 Jul 2024 20:58:18 -0400
Subject: [PATCH 039/505] Make `model_class` required arg, default
 `processor_class` to `AutoProcessor`

---
 docs/reference/models/transformers_vision.md |  2 ++
 outlines/models/transformers_vision.py       | 16 +++++++++-------
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
index e878f5866..b488bca1a 100644
--- a/docs/reference/models/transformers_vision.md
+++ b/docs/reference/models/transformers_vision.md
@@ -19,9 +19,11 @@ Install dependencies
 Create the model
 ```python
 import outlines
+from transformers import LlavaNextForConditionalGeneration
 
 model = outlines.models.transformers_vision(
     "llava-hf/llava-v1.6-mistral-7b-hf",
+    model_class=LlavaNextForConditionalGeneration,
 	device="cuda",
 )
 ```
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
index 876f9bff5..db4ba3e43 100644
--- a/outlines/models/transformers_vision.py
+++ b/outlines/models/transformers_vision.py
@@ -75,10 +75,10 @@ def stream(  # type: ignore
 
 def transformers_vision(
     model_name: str,
+    model_class,
     device: Optional[str] = None,
     model_kwargs: dict = {},
     processor_kwargs: dict = {},
-    model_class=None,
     tokenizer_class=None,
     processor_class=None,
 ):
@@ -88,6 +88,9 @@ def transformers_vision(
     ----------
     model_name
         The name of the model as listed on Hugging Face's model page.
+    model_class
+        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
+        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
     device
         The device(s) on which the model should be loaded. This overrides
         the `device_map` entry in `model_kwargs` when provided.
@@ -103,21 +106,20 @@ def transformers_vision(
     A `TransformersModel` model instance.
 
     """
-    if model_class is None or tokenizer_class is None:
+    if processor_class is None or tokenizer_class is None:
         try:
             from transformers import (
                 AutoTokenizer,
-                LlavaNextForConditionalGeneration,
-                LlavaNextProcessor,
+                AutoProcessor,
             )
         except ImportError:
             raise ImportError(
                 "The `transformers` library needs to be installed in order to use `transformers` models."
             )
-    if model_class is None:
-        model_class = LlavaNextForConditionalGeneration
     if processor_class is None:
-        processor_class = LlavaNextProcessor
+        processor_class = AutoProcessor
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
 
     if device is not None:
         model_kwargs["device_map"] = device

From a2dddeac41314ff50375aa48ec8830b744f8db67 Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Wed, 31 Jul 2024 09:26:19 -0400
Subject: [PATCH 040/505] Running pre-commit hooks

---
 outlines/models/transformers_vision.py | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
index db4ba3e43..9709c26ae 100644
--- a/outlines/models/transformers_vision.py
+++ b/outlines/models/transformers_vision.py
@@ -108,10 +108,7 @@ def transformers_vision(
     """
     if processor_class is None or tokenizer_class is None:
         try:
-            from transformers import (
-                AutoTokenizer,
-                AutoProcessor,
-            )
+            from transformers import AutoProcessor, AutoTokenizer
         except ImportError:
             raise ImportError(
                 "The `transformers` library needs to be installed in order to use `transformers` models."

From c6dd805abe6b4918fb74bbcd849c2d2dde75e044 Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Wed, 31 Jul 2024 15:53:56 -0400
Subject: [PATCH 041/505] Pass `model_class` to transformers_vision in
 test_generate.py

---
 tests/generate/test_generate.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index a86d3c253..8d5daa37e 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -57,9 +57,11 @@ def model_bart(tmp_path_factory):
 @pytest.fixture(scope="session")
 def model_transformers_vision(tmp_path_factory):
     import torch
+    from transformers import LlavaNextForConditionalGeneration
 
     return models.transformers_vision(
         "llava-hf/llava-v1.6-mistral-7b-hf",
+        model_class=LlavaNextForConditionalGeneration,
         device="cuda",
         model_kwargs=dict(
             torch_dtype=torch.bfloat16,

From c5a1788e1f42e1e952369b496e547604f74771af Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Wed, 31 Jul 2024 15:56:49 -0400
Subject: [PATCH 042/505] Add types-Pillow to mypy hook

---
 .pre-commit-config.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index b528f0e8e..a9039b605 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -30,4 +30,4 @@ repos:
     - id: mypy
       args: [--allow-redefinition]
       exclude: ^examples/
-      additional_dependencies: [types-tqdm]
+      additional_dependencies: [types-tqdm, types-Pillow]

From eca98a8cf4875712bba51bfe63fcd7c07c97005b Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Wed, 31 Jul 2024 15:56:57 -0400
Subject: [PATCH 043/505] Add CPU-based test for transformers_vision models

---
 .../test_integration_transformers_vision.py   | 41 +++++++++++++++++++
 1 file changed, 41 insertions(+)
 create mode 100644 tests/generate/test_integration_transformers_vision.py

diff --git a/tests/generate/test_integration_transformers_vision.py b/tests/generate/test_integration_transformers_vision.py
new file mode 100644
index 000000000..62bc6da53
--- /dev/null
+++ b/tests/generate/test_integration_transformers_vision.py
@@ -0,0 +1,41 @@
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image
+from transformers import LlavaForConditionalGeneration
+
+import outlines
+from outlines.models.transformers_vision import transformers_vision
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    return transformers_vision(
+        "trl-internal-testing/tiny-random-LlavaForConditionalGeneration",
+        model_class=LlavaForConditionalGeneration,
+        device="cpu",
+    )
+
+
+@pytest.fixture(scope="session")
+def image(tmp_path_factory):
+    def img_from_url(url):
+        img_byte_stream = BytesIO(urlopen(url).read())
+        return Image.open(img_byte_stream).convert("RGB")
+
+    return img_from_url(
+        "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+    )
+
+
+def test_text_gen(model, image):
+    assert model.model.device.type == "cpu"
+    description_generator = outlines.generate.text(model)
+    sequence = description_generator(
+        "<|im_start|>user\n<image>\nWhat is this?<|im_end|>\n<|im_start|>assistant\n",
+        [image],
+        seed=10000,
+        max_tokens=10,
+    )
+    assert isinstance(sequence, str)

From 894f240b02e9bd113153fde548570800b84d49ac Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Thu, 1 Aug 2024 09:19:10 -0400
Subject: [PATCH 044/505] Adding `apply_chat_template` argument to
 VisionSequenceGeneratorAdapter

---
 outlines/generate/api.py | 30 +++++++++++++++++++++++++++---
 1 file changed, 27 insertions(+), 3 deletions(-)

diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index ad01377c0..dea3108d5 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -1,6 +1,6 @@
 import datetime
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
+from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union, cast
 
 from outlines.generate.generator import sequence_generator
 from outlines.samplers import BeamSearchSampler, GreedySampler, MultinomialSampler
@@ -534,8 +534,9 @@ def stream(
 class VisionSequenceGeneratorAdapter(SequenceGeneratorAdapter):
     def __call__(  # type: ignore
         self,
-        prompts: Union[str, List[str]],
+        prompts: Union[str, List[str], List[dict], List[List[dict]]],
         media: Union[str, Any],
+        apply_chat_template: bool = False,
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
@@ -546,6 +547,17 @@ def __call__(  # type: ignore
 
         Media: A URI to construct media or media object itself. Used as AutoProcessor argument.
         """
+        if apply_chat_template:
+            # Transform the huggingface conversation object into the string that this
+            #   model expects.
+            # https://huggingface.co/docs/transformers/main/en/chat_templating
+            prompts = (
+                [self.model.processor.apply_chat_template(p) for p in prompts]
+                if isinstance(prompts[0], list)
+                else self.model.processor.apply_chat_template(prompts)
+            )
+        prompts = cast(Union[str, List[str]], prompts)
+
         prompts, media = self._validate_prompt_media_types(prompts, media)
 
         generation_params = self.prepare_generation_parameters(
@@ -565,14 +577,26 @@ def __call__(  # type: ignore
 
     def stream(  # type: ignore
         self,
-        prompts: Union[str, List[str]],
+        prompts: Union[str, List[str], List[dict], List[List[dict]]],
         media: List[Union[str, Any, List[Union[str, Any]]]],
+        apply_chat_template: bool = False,
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
         **model_specific_params,
     ):
         """Return a text generator from a prompt or a list of prompts."""
+        if apply_chat_template:
+            # Transform the huggingface conversation object into the string that this
+            #   model expects.
+            # https://huggingface.co/docs/transformers/main/en/chat_templating
+            prompts = (
+                [self.model.processor.apply_chat_template(p) for p in prompts]
+                if isinstance(prompts[0], list)
+                else self.model.processor.apply_chat_template(prompts)
+            )
+        prompts = cast(Union[str, List[str]], prompts)
+
         prompts, media = self._validate_prompt_media_types(prompts, media)
         generation_params = self.prepare_generation_parameters(
             max_tokens, stop_at, seed

From 3a98e44ebbaa501d6bceedacedc10432b39b92ef Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Thu, 1 Aug 2024 09:19:53 -0400
Subject: [PATCH 045/505] More comprehensive TransformersVision test cases

Multi-image context, using `apply_chat_template`
---
 .../test_integration_transformers_vision.py   | 132 ++++++++++++++++--
 1 file changed, 119 insertions(+), 13 deletions(-)

diff --git a/tests/generate/test_integration_transformers_vision.py b/tests/generate/test_integration_transformers_vision.py
index 62bc6da53..c976f04a9 100644
--- a/tests/generate/test_integration_transformers_vision.py
+++ b/tests/generate/test_integration_transformers_vision.py
@@ -3,11 +3,22 @@
 
 import pytest
 from PIL import Image
-from transformers import LlavaForConditionalGeneration
+from transformers import AutoProcessor, LlavaForConditionalGeneration
 
 import outlines
 from outlines.models.transformers_vision import transformers_vision
 
+IMAGE_URLS = [
+    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
+    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
+    "https://upload.wikimedia.org/wikipedia/commons/b/be/Tamias-rufus-001.jpg",
+]
+
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
 
 @pytest.fixture(scope="session")
 def model(tmp_path_factory):
@@ -19,23 +30,118 @@ def model(tmp_path_factory):
 
 
 @pytest.fixture(scope="session")
-def image(tmp_path_factory):
-    def img_from_url(url):
-        img_byte_stream = BytesIO(urlopen(url).read())
-        return Image.open(img_byte_stream).convert("RGB")
+def larger_model_with_processor(tmp_path_factory):
+    return transformers_vision(
+        "llava-hf/llava-interleave-qwen-0.5b-hf",
+        model_class=LlavaForConditionalGeneration,
+        device="cpu",
+    )
+
+
+@pytest.fixture(scope="session")
+def processor(tmp_path_factory):
+    return AutoProcessor.from_pretrained("llava-hf/llava-interleave-qwen-0.5b-hf")
+
+
+def test_single_image_text_gen(model, processor):
+    conversation = [
+        {
+            "role": "user",
+            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
+        },
+    ]
+    generator = outlines.generate.text(model)
+    sequence = generator(
+        processor.apply_chat_template(conversation),
+        [img_from_url(IMAGE_URLS[0])],
+        seed=10000,
+        max_tokens=10,
+    )
+    assert isinstance(sequence, str)
+
+
+def test_multi_image_text_gen(model, processor):
+    """If the length of image tags and number of images we pass are > 1 and equal,
+    we should yield a successful generation.
+    """
+    conversation = [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "What do all these have in common?"},
+            ]
+            + [{"type": "image"} for _ in range(len(IMAGE_URLS))],
+        },
+    ]
+    generator = outlines.generate.text(model)
+    sequence = generator(
+        processor.apply_chat_template(conversation),
+        [img_from_url(i) for i in IMAGE_URLS],
+        seed=10000,
+        max_tokens=10,
+    )
+    assert isinstance(sequence, str)
+
+
+def test_mismatched_image_text_gen(model, processor):
+    """If the length of image tags and number of images we pass are unequal,
+    we should raise an error.
+    """
+    conversation = [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "I'm passing 3 images, but only 1 image tag"},
+                {"type": "image"},
+            ],
+        },
+    ]
+    generator = outlines.generate.text(model)
+    with pytest.raises(ValueError):
+        _ = generator(
+            processor.apply_chat_template(conversation),
+            [img_from_url(i) for i in IMAGE_URLS],
+            seed=10000,
+            max_tokens=10,
+        )
+
 
-    return img_from_url(
-        "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+def test_default_apply_chat_template(larger_model_with_processor):
+    """If we pass in a list of dicts in the format that huggingface chat_template expects,
+    and specify `apply_chat_template=True` when calling the generator,
+    we should yield a successful generation.
+    """
+    conversation = [
+        {
+            "role": "user",
+            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
+        },
+    ]
+    generator = outlines.generate.text(larger_model_with_processor)
+    sequence = generator(
+        conversation,
+        [img_from_url(IMAGE_URLS[0])],
+        seed=10000,
+        max_tokens=10,
+        apply_chat_template=True,
     )
+    assert isinstance(sequence, str)
 
 
-def test_text_gen(model, image):
-    assert model.model.device.type == "cpu"
-    description_generator = outlines.generate.text(model)
-    sequence = description_generator(
-        "<|im_start|>user\n<image>\nWhat is this?<|im_end|>\n<|im_start|>assistant\n",
-        [image],
+def test_single_image_choice(model, processor):
+    conversation = [
+        {
+            "role": "user",
+            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
+        },
+    ]
+    choices = ["cat", "dog"]
+    generator = outlines.generate.choice(model, choices)
+    sequence = generator(
+        processor.apply_chat_template(conversation),
+        [img_from_url(IMAGE_URLS[0])],
         seed=10000,
         max_tokens=10,
     )
     assert isinstance(sequence, str)
+    assert sequence in choices

From a363568e9bc0e08cbc4e23161024a9e4c9772d28 Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Thu, 1 Aug 2024 17:43:30 -0400
Subject: [PATCH 046/505] Revert "Adding `apply_chat_template` argument to
 VisionSequenceGeneratorAdapter"

This reverts commit 6200d402601a744ce483ffaec286f01b468fb507.
---
 outlines/generate/api.py | 30 +++---------------------------
 1 file changed, 3 insertions(+), 27 deletions(-)

diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index dea3108d5..ad01377c0 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -1,6 +1,6 @@
 import datetime
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union, cast
+from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
 
 from outlines.generate.generator import sequence_generator
 from outlines.samplers import BeamSearchSampler, GreedySampler, MultinomialSampler
@@ -534,9 +534,8 @@ def stream(
 class VisionSequenceGeneratorAdapter(SequenceGeneratorAdapter):
     def __call__(  # type: ignore
         self,
-        prompts: Union[str, List[str], List[dict], List[List[dict]]],
+        prompts: Union[str, List[str]],
         media: Union[str, Any],
-        apply_chat_template: bool = False,
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
@@ -547,17 +546,6 @@ def __call__(  # type: ignore
 
         Media: A URI to construct media or media object itself. Used as AutoProcessor argument.
         """
-        if apply_chat_template:
-            # Transform the huggingface conversation object into the string that this
-            #   model expects.
-            # https://huggingface.co/docs/transformers/main/en/chat_templating
-            prompts = (
-                [self.model.processor.apply_chat_template(p) for p in prompts]
-                if isinstance(prompts[0], list)
-                else self.model.processor.apply_chat_template(prompts)
-            )
-        prompts = cast(Union[str, List[str]], prompts)
-
         prompts, media = self._validate_prompt_media_types(prompts, media)
 
         generation_params = self.prepare_generation_parameters(
@@ -577,26 +565,14 @@ def __call__(  # type: ignore
 
     def stream(  # type: ignore
         self,
-        prompts: Union[str, List[str], List[dict], List[List[dict]]],
+        prompts: Union[str, List[str]],
         media: List[Union[str, Any, List[Union[str, Any]]]],
-        apply_chat_template: bool = False,
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
         **model_specific_params,
     ):
         """Return a text generator from a prompt or a list of prompts."""
-        if apply_chat_template:
-            # Transform the huggingface conversation object into the string that this
-            #   model expects.
-            # https://huggingface.co/docs/transformers/main/en/chat_templating
-            prompts = (
-                [self.model.processor.apply_chat_template(p) for p in prompts]
-                if isinstance(prompts[0], list)
-                else self.model.processor.apply_chat_template(prompts)
-            )
-        prompts = cast(Union[str, List[str]], prompts)
-
         prompts, media = self._validate_prompt_media_types(prompts, media)
         generation_params = self.prepare_generation_parameters(
             max_tokens, stop_at, seed

From c2d92b78fc0d5e1afbeadcb59e624f9e2b210a7b Mon Sep 17 00:00:00 2001
From: parkervg <parkervg5@gmail.com>
Date: Thu, 1 Aug 2024 17:45:37 -0400
Subject: [PATCH 047/505] Remove `apply_chat_template` test

---
 .../test_integration_transformers_vision.py   | 31 -------------------
 1 file changed, 31 deletions(-)

diff --git a/tests/generate/test_integration_transformers_vision.py b/tests/generate/test_integration_transformers_vision.py
index c976f04a9..28b516c57 100644
--- a/tests/generate/test_integration_transformers_vision.py
+++ b/tests/generate/test_integration_transformers_vision.py
@@ -29,15 +29,6 @@ def model(tmp_path_factory):
     )
 
 
-@pytest.fixture(scope="session")
-def larger_model_with_processor(tmp_path_factory):
-    return transformers_vision(
-        "llava-hf/llava-interleave-qwen-0.5b-hf",
-        model_class=LlavaForConditionalGeneration,
-        device="cpu",
-    )
-
-
 @pytest.fixture(scope="session")
 def processor(tmp_path_factory):
     return AutoProcessor.from_pretrained("llava-hf/llava-interleave-qwen-0.5b-hf")
@@ -106,28 +97,6 @@ def test_mismatched_image_text_gen(model, processor):
         )
 
 
-def test_default_apply_chat_template(larger_model_with_processor):
-    """If we pass in a list of dicts in the format that huggingface chat_template expects,
-    and specify `apply_chat_template=True` when calling the generator,
-    we should yield a successful generation.
-    """
-    conversation = [
-        {
-            "role": "user",
-            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
-        },
-    ]
-    generator = outlines.generate.text(larger_model_with_processor)
-    sequence = generator(
-        conversation,
-        [img_from_url(IMAGE_URLS[0])],
-        seed=10000,
-        max_tokens=10,
-        apply_chat_template=True,
-    )
-    assert isinstance(sequence, str)
-
-
 def test_single_image_choice(model, processor):
     conversation = [
         {

From 90e588c7f30fbcaa2bff8aa50e2c11e480d3fbdb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 13 Aug 2024 17:44:23 +0200
Subject: [PATCH 048/505] Fix details in the documentation

---
 docs/cookbook/chain_of_thought.md             |  4 +-
 docs/cookbook/dating_profiles.md              |  4 +-
 docs/cookbook/index.md                        |  2 +
 docs/cookbook/knowledge_graph_extraction.md   |  4 +-
 docs/cookbook/qa-with-citations.md            |  4 +-
 docs/cookbook/react_agent.md                  |  7 ++-
 docs/reference/generation/generation.md       |  6 +++
 .../structured_generation_explanation.md      |  6 +--
 docs/reference/models/models.md               | 52 +++----------------
 docs/reference/models/transformers.md         | 23 ++++----
 docs/stylesheets/extra.css                    |  8 +++
 docs/welcome.md                               |  6 +--
 mkdocs.yml                                    |  8 +--
 13 files changed, 51 insertions(+), 83 deletions(-)

diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
index d320feb8d..cc079a7ff 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -7,13 +7,13 @@ In this guide, we use [outlines](https://outlines-dev.github.io/outlines/) to ap
 
 We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```shell
+```bash
 pip install llama-cpp-python
 ```
 
 We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-```shell
+```bash
 wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
 ```
 
diff --git a/docs/cookbook/dating_profiles.md b/docs/cookbook/dating_profiles.md
index f839b65fe..d0fb9b576 100644
--- a/docs/cookbook/dating_profiles.md
+++ b/docs/cookbook/dating_profiles.md
@@ -170,7 +170,7 @@ parsed_profile = DatingProfile.model_validate_json(profile)
 
 Here are a couple of results:
 
-```
+```json
 {
     "bio": """I'm an ambitious lawyer with a casual and fashionable style. I love
     games and sports, but my true passion is preparing refreshing cocktails at
@@ -199,7 +199,7 @@ Here are a couple of results:
 }
 ```
 
-```
+```json
 {
     "bio": """I’m a sexy lawyer with time on my hands. I love to game and
     play ping pong, but the real reason you should swipe to the right
diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index 58e84ae96..a844ce240 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -1,5 +1,7 @@
 # Examples
 
+This part of the documentation provides a few cookbooks that you can browse to get acquainted with the library and get some inspiration about what you could do with structured generation. Remember that you can easily change the model that is being used!
+
 - [Classification](classification.md): Classify customer requests.
 - [Named Entity Extraction](extraction.md): Extract information from pizza orders.
 - [Dating Profile](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
diff --git a/docs/cookbook/knowledge_graph_extraction.md b/docs/cookbook/knowledge_graph_extraction.md
index c7e347dd4..c4c1dc75c 100644
--- a/docs/cookbook/knowledge_graph_extraction.md
+++ b/docs/cookbook/knowledge_graph_extraction.md
@@ -4,13 +4,13 @@ In this guide, we use [outlines](https://outlines-dev.github.io/outlines/) to ex
 
 We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```shell
+```bash
 pip install llama-cpp-python
 ```
 
 We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-```shell
+```bash
 wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
 ```
 
diff --git a/docs/cookbook/qa-with-citations.md b/docs/cookbook/qa-with-citations.md
index cb39befe9..c2111617f 100644
--- a/docs/cookbook/qa-with-citations.md
+++ b/docs/cookbook/qa-with-citations.md
@@ -4,13 +4,13 @@ This tutorial is adapted from the [instructor-ollama notebook](https://github.co
 
 We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```shell
+```bash
 pip install llama-cpp-python
 ```
 
 We pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-```shell
+```bash
 wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
 ```
 
diff --git a/docs/cookbook/react_agent.md b/docs/cookbook/react_agent.md
index 74930b70b..15fb964a0 100644
--- a/docs/cookbook/react_agent.md
+++ b/docs/cookbook/react_agent.md
@@ -8,13 +8,13 @@ Additionally, we give the LLM the possibility of using a scratchpad described in
 
 We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```shell
+```bash
 pip install llama-cpp-python
 ```
 
 We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-```shell
+```bash
 wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
 ```
 
@@ -55,9 +55,8 @@ def wikipedia(q):
         "srsearch": q,
         "format": "json"
     }).json()["query"]["search"][0]["snippet"]
-```
 
-```python
+
 def calculate(numexp):
     return eval(numexp)
 ```
diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
index 88b963c72..0c090f8a7 100644
--- a/docs/reference/generation/generation.md
+++ b/docs/reference/generation/generation.md
@@ -208,3 +208,9 @@ result = generator("What is 2+2?")
 print(result)
 # 4
 ```
+
+
+[jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
+[pydantic]: https://docs.pydantic.dev/latest
+[cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
+[ebnf]: https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form
diff --git a/docs/reference/generation/structured_generation_explanation.md b/docs/reference/generation/structured_generation_explanation.md
index 0dedf060b..aa27a7a85 100644
--- a/docs/reference/generation/structured_generation_explanation.md
+++ b/docs/reference/generation/structured_generation_explanation.md
@@ -1,8 +1,4 @@
----
-title: Structured Generation Explanation
----
-
-# Structured Generation Explanation
+# How does Outlines work?
 
 
 Language models generate text token by token, using the previous token sequence as input and sampled logits as output. This document explains the structured generation process, where only legal tokens are considered for the next step based on a predefined automata, e.g. a regex-defined [finite-state machine](https://en.wikipedia.org/wiki/Finite-state_machine) (FSM) or [Lark](https://lark-parser.readthedocs.io/en/stable/) grammar.`
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
index dadfd34ad..34b5be4cf 100644
--- a/docs/reference/models/models.md
+++ b/docs/reference/models/models.md
@@ -4,60 +4,20 @@ title: Models
 
 # Models
 
-Outlines supports generation using a number of inference engines (`outlines.models`)
-
-Loading a model using outlines follows a similar interface between inference engines.
+Outlines supports generation using a number of inference engines (`outlines.models`). Loading a model using outlines follows a similar interface between inference engines:
 
 ```python
 import outlines
-```
-
-## [Transformers](./transformers.md)
-
-```python
-model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct", model_kwargs={})
-```
-
-For additional arguments and use of other Huggingface Transformers model types see [Outlines' Transformers documentation](./transformers.md).
-
 
-## [Transformers Vision](./transformers_vision.md)
-
-```python
+model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
 model = outlines.models.transformers_vision("llava-hf/llava-v1.6-mistral-7b-hf")
-```
-
-For examples of generation and other details, see [Outlines' Transformers Vision documentation](./transformers_vision.md).
-
-## [vLLM](./vllm.md)
-
-```python
 model = outlines.models.vllm("microsoft/Phi-3-mini-128k-instruct")
-```
-
-## [llama.cpp](./llamacpp.md)
-
-```python
-model = outlines.models.llamacpp("microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf")
-```
-
-Additional llama.cpp parameters can be found in the [Outlines' llama.cpp documentation](./llamacpp.md).
-
-## [ExLlamaV2](./exllamav2.md)
-
-```python
+model = outlines.models.llamacpp(
+    "microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf"
+)
 model = outlines.models.exllamav2("bartowski/Phi-3-mini-128k-instruct-exl2")
-```
-
-## [MLXLM](./mlxlmx.md)
-
-```python
 model = outlines.models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
-```
 
-## [OpenAI](./openai.md)
-
-```python
 model = outlines.models.openai(
     "gpt-4o-mini",
     api_key=os.environ["OPENAI_API_KEY"]
@@ -66,7 +26,7 @@ model = outlines.models.openai(
 
 
 # Feature Matrix
-|                   | Transformers | Transformers Vision | vLLM | llama.cpp | ExLlamaV2 | MLXLM | OpenAI* |
+|                   | [Transformers](transformers.md) | [Transformers Vision](transformers_vision.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md)* |
 |-------------------|--------------|---------------------|------|-----------|-----------|-------|---------|
 | **Device**        |              |                     |      |           |           |       |         |
 | Cuda              | ✅           | ✅                  | ✅   | ✅        | ✅        | ❌    | N/A     |
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 15eabb682..2a13e28ec 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -33,14 +33,15 @@ model = models.Transformers(llm, tokenizer)
 # Using Logits Processors
 
 There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
-- 1) Use Outlines generation wrapper, `outlines.models.transformers`
-- 2) Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
+
+1. Use Outlines generation wrapper, `outlines.models.transformers`
+2. Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
 
 Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
 
-## Example: `outlines.models.transformers`
+## Using `outlines.models.transformers`
 
-```
+```python
 import outlines
 
 time_regex_pattern = r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?"
@@ -53,9 +54,9 @@ print(output)
 # 2:30 pm
 ```
 
-## Example: Direct `transformers` library use
+## Using models initialized via the `transformers`  library
 
-```
+```python
 import outlines
 import transformers
 
@@ -117,8 +118,9 @@ model = outlines.models.transformers(
 )
 ```
 
-Further Reading:
-- https://huggingface.co/docs/transformers/en/model_doc/mamba
+
+
+Read [`transformers`'s documentation](https://huggingface.co/docs/transformers/en/model_doc/mamba) for more information.
 
 ### Encoder-Decoder Models
 
@@ -144,8 +146,3 @@ model_bart = models.transformers(
     model_class=AutoModelForSeq2SeqLM,
 )
 ```
-
-
-### Multi-Modal Models
-
-/Coming soon/
diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index 4078215af..c4539ab80 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -96,6 +96,14 @@
   background: #FFFFFF ! important
 }
 
+.language-text {
+  background: #FFFFFF ! important
+}
+
+.language-json {
+  background: #FFFFFF ! important
+}
+
 h1.title {
   color: #FFFFFF;
   margin: 0px 0px 5px;
diff --git a/docs/welcome.md b/docs/welcome.md
index 4c327c020..a7800f7ad 100644
--- a/docs/welcome.md
+++ b/docs/welcome.md
@@ -6,7 +6,7 @@ Outlines〰 is a Python library that allows you to use Large Language Model in a
 
 ## What models do you support?
 
-We support [Openai](reference/models/openai.md), but the true power of Outlines〰 is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [exllama2](reference/models/exllamav2.md) and [mamba_ssm](reference/models/mamba.md) libraries. If you want to build and maintain an integration with another library, [get in touch][discord].
+We support [Openai](reference/models/openai.md), but the true power of Outlines〰 is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [exllama2](reference/models/exllamav2.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
 
 ## What are the main features?
 
@@ -17,7 +17,7 @@ We support [Openai](reference/models/openai.md), but the true power of Outlines
 
     No more invalid JSON outputs, 100% guaranteed
 
-    [:octicons-arrow-right-24: Generate JSON](reference/json.md)
+    [:octicons-arrow-right-24: Generate JSON](reference/generation/json.md)
 
 -   :material-keyboard-outline:{ .lg .middle } __JSON mode for vLLM__
 
@@ -34,7 +34,7 @@ We support [Openai](reference/models/openai.md), but the true power of Outlines
 
     Generate text that parses correctly 100% of the time
 
-    [:octicons-arrow-right-24: Guide LLMs](reference/regex.md)
+    [:octicons-arrow-right-24: Guide LLMs](reference/generation/regex.md)
 
 -    :material-chat-processing-outline:{ .lg .middle } __Powerful Prompt Templating__
 
diff --git a/mkdocs.yml b/mkdocs.yml
index d24ca9a63..afc56528b 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -121,24 +121,24 @@ nav:
   - Docs:
     - reference/index.md
     - Generation:
-        - Generation Overview: reference/generation/generation.md
+        - Overview: reference/generation/generation.md
         - Text: reference/text.md
         - Samplers: reference/samplers.md
         - Structured generation:
+            - How does it work?: reference/generation/structured_generation_explanation.md
             - Classification: reference/generation/choices.md
             - Regex: reference/generation/regex.md
             - Type constraints: reference/generation/format.md
             - JSON (function calling): reference/generation/json.md
             - Grammar: reference/generation/cfg.md
             - Custom FSM operations: reference/generation/custom_fsm_ops.md
-            - Structured Generation Technical Explanation: reference/generation/structured_generation_explanation.md
     - Utilities:
         - Serve with vLLM: reference/serve/vllm.md
-        - Custom types: reference/types.md
+        - Custom types: reference/generation/types.md
         - Prompt templating: reference/prompting.md
         - Outlines functions: reference/functions.md
     - Models:
-        - Models Overview: reference/models/models.md
+        - Overview: reference/models/models.md
         - Open source:
           - Transformers: reference/models/transformers.md
           - Transformers Vision: reference/models/transformers_vision.md

From ce49ccd495e379518fb59f33cea252d903781733 Mon Sep 17 00:00:00 2001
From: Alonso Silva Allende <alonso.silva@gmail.com>
Date: Tue, 13 Aug 2024 23:03:01 +0200
Subject: [PATCH 049/505] Change cookbook examples: Download model weights in
 the hub cache folder (#1097)

Change cookbook examples: Download model weights in the hub cache folder
---
 docs/cookbook/chain_of_thought.md           | 56 +++++++++++++-------
 docs/cookbook/knowledge_graph_extraction.md | 56 +++++++++++++-------
 docs/cookbook/qa-with-citations.md          | 55 +++++++++++++-------
 docs/cookbook/react_agent.md                | 57 +++++++++++++--------
 4 files changed, 148 insertions(+), 76 deletions(-)

diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
index cc079a7ff..17c362696 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -11,30 +11,48 @@ We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-
 pip install llama-cpp-python
 ```
 
-We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+```python
+import llama_cpp
+from outlines import generate, models
 
-```bash
-wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
-We initialize the model:
+??? note "(Optional) Store the model weights in a custom folder"
 
-```python
-from llama_cpp import Llama
-from outlines import generate, models
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-llm = Llama(
-    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = models.LlamaCpp(llm)
-```
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
 
 ## Chain of thought
 
diff --git a/docs/cookbook/knowledge_graph_extraction.md b/docs/cookbook/knowledge_graph_extraction.md
index c4c1dc75c..e25166bca 100644
--- a/docs/cookbook/knowledge_graph_extraction.md
+++ b/docs/cookbook/knowledge_graph_extraction.md
@@ -8,30 +8,48 @@ We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama
 pip install llama-cpp-python
 ```
 
-We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+```python
+import llama_cpp
+from outlines import generate, models
 
-```bash
-wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
-We initialize the model:
+??? note "(Optional) Store the model weights in a custom folder"
 
-```python
-from llama_cpp import Llama
-from outlines import generate, models
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-llm = Llama(
-    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = models.LlamaCpp(llm)
-```
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
 
 ## Knowledge Graph Extraction
 
diff --git a/docs/cookbook/qa-with-citations.md b/docs/cookbook/qa-with-citations.md
index c2111617f..79a2214c3 100644
--- a/docs/cookbook/qa-with-citations.md
+++ b/docs/cookbook/qa-with-citations.md
@@ -8,29 +8,48 @@ We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama
 pip install llama-cpp-python
 ```
 
-We pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+```python
+import llama_cpp
+from outlines import generate, models
 
-```bash
-wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
-We initialize the model:
+??? note "(Optional) Store the model weights in a custom folder"
 
-```python
-from llama_cpp import Llama
-from outlines import generate, models
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-llm = Llama(
-    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-```
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
 
 ## Generate Synthetic Data
 
diff --git a/docs/cookbook/react_agent.md b/docs/cookbook/react_agent.md
index 15fb964a0..ca4829d5f 100644
--- a/docs/cookbook/react_agent.md
+++ b/docs/cookbook/react_agent.md
@@ -12,32 +12,49 @@ We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-
 pip install llama-cpp-python
 ```
 
-We pull a quantized GGUF model, in this guide we pull [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
-
-```bash
-wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
-```
-
-We initialize the model:
-
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-from llama_cpp import Llama
 from outlines import generate, models
 
-llm = Llama(
-    "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = models.LlamaCpp(llm)
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
+??? note "(Optional) Store the model weights in a custom folder"
+
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
+
 ## Build a ReAct agent
 
 In this example, we use two tools:

From 8e94488d4ee3c5a29a919d0b9e19f7ea4170b1f4 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Wed, 14 Aug 2024 16:42:32 -0700
Subject: [PATCH 050/505] Correct variable name in chain-of-thought example

---
 docs/cookbook/chain_of_thought.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
index 17c362696..bd76f40b7 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -94,7 +94,7 @@ def generate_hermes_prompt(user_prompt):
     return (
         "<|im_start|>system\n"
         "You are a world class AI model who answers questions in JSON "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        f"Here's the json schema you must adhere to:\n<schema>\n{json_schema}\n</schema><|im_end|>\n"
         "<|im_start|>user\n"
         + user_prompt
         + "<|im_end|>"

From e3ecbccfcfc4e7ed9d2243828c60fdeb10ea3df0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Mon, 22 Jul 2024 15:37:09 +0200
Subject: [PATCH 051/505] Remove unused `llama.cpp` and `transformers`
 integrations

---
 outlines/integrations/llamacpp.py     | 191 --------------------------
 outlines/integrations/transformers.py | 159 ---------------------
 2 files changed, 350 deletions(-)
 delete mode 100644 outlines/integrations/llamacpp.py
 delete mode 100644 outlines/integrations/transformers.py

diff --git a/outlines/integrations/llamacpp.py b/outlines/integrations/llamacpp.py
deleted file mode 100644
index 8e18c33e7..000000000
--- a/outlines/integrations/llamacpp.py
+++ /dev/null
@@ -1,191 +0,0 @@
-"""Make LlamaCpp compatible with Outlines' structured generation.
-
- _______________________________
-/ Don't want to self-host?       \
-\\ Try .json at http://dottxt.co /
- -------------------------------
-       \\   ^__^
-        \\  (oo)\\_______
-            (__)\\       )\\/\
-                ||----w |
-                ||     ||
-
-Copyright 2024- the Outlines developers
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-"""
-
-import math
-from typing import TYPE_CHECKING, Optional, Type, Union
-
-import numpy as np
-import torch
-from numpy.typing import NDArray
-from pydantic import BaseModel
-
-from outlines.fsm.guide import CFGGuide, Guide, RegexGuide
-from outlines.fsm.json_schema import build_regex_from_schema
-from outlines.integrations.utils import convert_json_schema_to_str
-from outlines.models.llamacpp import LlamaCppTokenizer
-
-if TYPE_CHECKING:
-    from llama_cpp import Llama
-
-
-class LogitsProcessor:
-    """Bias LlamaCpp generation using a finite state machine.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(self, tokenizer: LlamaCppTokenizer, fsm: Guide):
-        """A FSM-based logits processor.
-
-        Parameters
-        ----------
-        tokenizer
-            The tokenizer used to convert tokens to ids.
-        fsm
-            The finite state machine which is used to bias the logits.
-        """
-        self.tokenizer = tokenizer
-        self._fsm_state = 0
-        self.fsm: Guide = fsm
-        self._is_first_token = True
-
-    def __call__(
-        self, input_ids: NDArray[np.int64], scores: NDArray[np.float32]
-    ) -> NDArray[np.float32]:
-        """Use the FSM to bias the logits before sampling the next token.
-
-        Parameters
-        ----------
-        input_ids
-            The input token ids.
-        scores
-            The logits.
-
-        Returns
-        -------
-        NDArray[np.float32]
-            The biased logits.
-        """
-        if self._is_first_token:
-            self._is_first_token = False
-        else:
-            last_token = input_ids[-1]
-            self._fsm_state = self.fsm.get_next_state(self._fsm_state, last_token)
-
-        allowed_tokens = self.fsm.get_next_instruction(self._fsm_state).tokens
-
-        mask = torch.full((scores.shape[-1],), -math.inf, device="cpu").numpy()
-        mask[allowed_tokens] = 0
-        biased_scores = scores + mask
-
-        return biased_scores
-
-    def copy(self) -> "LogitsProcessor":
-        """Return a copy of the logits processor."""
-        return LogitsProcessor(tokenizer=self.tokenizer, fsm=self.fsm.copy())
-
-
-class RegexLogitsProcessor(LogitsProcessor):
-    """Bias LlamaCpp generation based on a regular expression.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(self, regex_string: str, llm: "Llama"):
-        """Compile the FSM that drives the regex-guided generation.
-
-        Parameters
-        ----------
-        regex_string
-            A string that represents a regular expression
-        llm
-            The Llama model.
-        """
-        tokenizer = LlamaCppTokenizer(model=llm)
-        fsm = RegexGuide(regex_string, tokenizer)
-        super().__init__(tokenizer=tokenizer, fsm=fsm)
-
-
-class JSONLogitsProcessor(RegexLogitsProcessor):
-    """Bias LlamaCpp generation based on a JSON schema.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(
-        self,
-        schema: Union[dict, Type[BaseModel], str],
-        llm: "Llama",
-        whitespace_pattern: Optional[str] = None,
-    ):
-        """Compile the FSM that drives the JSON-guided generation.
-
-        Parameters
-        ----------
-        schema
-            A JSON schema that encodes the structure we want the model to generate.
-        llm
-            The Llama model.
-        whitespace_pattern
-            Pattern to use for JSON syntactic whitespace (doesn't impact string
-            literals). For example, to allow only a single space or newline with
-            `whitespace_pattern=r"[\n ]?"`
-        """
-        schema_str = convert_json_schema_to_str(json_schema=schema)
-        regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(regex_string=regex_string, llm=llm)
-
-
-class CFGLogitsProcessor(LogitsProcessor):
-    """Bias LlamaCpp generation based on a context-free grammar.
-
-    Attributes
-    ----------
-    llm
-        The Llama model.
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(self, cfg_str: str, llm: "Llama"):
-        """Compile the FSM that drives the CFG-guided generation.
-
-        Parameters
-        ----------
-        cfg_str
-            A string that represents a grammar
-        llm
-            The Llama model.
-        """
-        tokenizer = LlamaCppTokenizer(model=llm)
-        fsm = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
-        super().__init__(tokenizer=tokenizer, fsm=fsm)
diff --git a/outlines/integrations/transformers.py b/outlines/integrations/transformers.py
deleted file mode 100644
index 7c1bafd22..000000000
--- a/outlines/integrations/transformers.py
+++ /dev/null
@@ -1,159 +0,0 @@
-"""Make Hugging Face transformers compatible with Outlines' structured generation.
-
- _______________________________
-/ Don't want to self-host?       \
-\\ Try .json at http://dottxt.co /
- -------------------------------
-       \\   ^__^
-        \\  (oo)\\_______
-            (__)\\       )\\/\
-                ||----w |
-                ||     ||
-
-Copyright 2024- the Outlines developers
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-"""
-
-from collections import defaultdict
-from typing import DefaultDict, Iterable, Optional, Type, Union
-
-import torch
-from pydantic import BaseModel
-from transformers import Pipeline, PreTrainedTokenizerBase
-
-from outlines.fsm.guide import RegexGuide
-from outlines.fsm.json_schema import build_regex_from_schema
-from outlines.integrations.utils import adapt_tokenizer, convert_json_schema_to_str
-
-
-class RegexPrefixAllowedTokens:
-    """Bias transformers generation based on a regular expression.
-
-    Attributes
-    ----------
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(
-        self,
-        regex_string: str,
-        tokenizer_or_pipe: Union[PreTrainedTokenizerBase, Pipeline],
-    ):
-        """Compile the FSM that drives the regex-structured generation.
-
-        Parameters
-        ----------
-        regex_string
-            A string that represents a regular expression.
-        tokenizer_or_pipe
-            The tokenizer of the model, or the pipeline object.
-
-        Raises
-        ------
-        ValueError
-            If the `tokenizer_or_pipe` parameter is not a tokenizer or a pipeline.
-        """
-        if isinstance(tokenizer_or_pipe, Pipeline):
-            tokenizer = tokenizer_or_pipe.tokenizer
-        elif isinstance(tokenizer_or_pipe, PreTrainedTokenizerBase):
-            tokenizer = tokenizer_or_pipe
-        else:
-            raise ValueError(
-                "The tokenizer_or_pipe parameter must be a tokenizer or a pipeline."
-            )
-        assert isinstance(tokenizer, PreTrainedTokenizerBase)
-        tokenizer = adapt_tokenizer(tokenizer=tokenizer)
-        self.fsm = RegexGuide(regex_string=regex_string, tokenizer=tokenizer)
-        self._fsm_state: DefaultDict[int, int] = defaultdict(int)
-
-        # The generated text with `transformers` include the input token IDs as well,
-        # so we use this attribute to keep track of the input token IDs. This allows us
-        # to reset the FSM state when the input token IDs change, as well as to only
-        # apply the FSM to the generated tokens.
-        self._prefix = [-1]
-
-    def __call__(self, batch_id: int, sent: torch.Tensor) -> Optional[Iterable[int]]:
-        """Use the FSM to bias the logits before sampling the next token.
-
-        Parameters
-        ----------
-        batch_id
-            The index of the current batch.
-        sent
-            The tokens of the current sentence.
-
-        Returns
-        -------
-        List[int]
-            The indices of the tokens that are allowed to be sampled next.
-        """
-        input_ids = sent.tolist()
-
-        # If the prefix token IDs have changed we assume that we are dealing with a new
-        # sample and reset the FSM state
-        if input_ids[: len(self._prefix)] != self._prefix:
-            self._fsm_state = defaultdict(int)
-            self._prefix = input_ids
-            seq_id = hash(tuple([]))
-
-        else:
-            # Remove the prefix token IDs from the input token IDs, as the FSM should
-            # only be applied to the generated tokens
-            input_ids = input_ids[len(self._prefix) :]
-
-            last_token = input_ids[-1]
-            last_seq_id = hash(tuple(input_ids[:-1]))
-            seq_id = hash(tuple(input_ids))
-            self._fsm_state[seq_id] = self.fsm.get_next_state(
-                state=self._fsm_state[last_seq_id], token_id=last_token
-            )
-
-        allowed_tokens = self.fsm.get_next_instruction(
-            state=self._fsm_state[seq_id]
-        ).tokens
-        return allowed_tokens
-
-
-class JSONPrefixAllowedTokens(RegexPrefixAllowedTokens):
-    """Bias transformers generation based on a JSON schema.
-
-    Attributes
-    ----------
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(
-        self,
-        schema: Union[dict, Type[BaseModel], str],
-        tokenizer_or_pipe: Union[PreTrainedTokenizerBase, Pipeline],
-        whitespace_pattern: Optional[str] = None,
-    ):
-        """Compile the FSM that drives the JSON-guided generation.
-
-        Parameters
-        ----------
-        schema
-            A schema that encodes the structure we want the model to generate.
-        tokenizer_or_pipe
-            The tokenizer of the model, or the pipeline object.
-        whitespace_pattern
-            Pattern to use for JSON syntactic whitespace (doesn't impact string
-            literals). For example, to allow only a single space or newline with
-            `whitespace_pattern=r"[\n ]?"`
-        """
-        schema_str = convert_json_schema_to_str(json_schema=schema)
-        regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(regex_string=regex_string, tokenizer_or_pipe=tokenizer_or_pipe)

From 03b61c5085a6f52c37513aaf47fc5a85f22170ba Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Mon, 22 Jul 2024 15:50:22 +0200
Subject: [PATCH 052/505] Move BaseModel conversion to `outlines.fsm`

---
 outlines/fsm/json_schema.py       | 37 +++++++++++++++++++++++++++++--
 outlines/integrations/utils.py    | 33 ---------------------------
 outlines/processors/structured.py |  3 +--
 3 files changed, 36 insertions(+), 37 deletions(-)

diff --git a/outlines/fsm/json_schema.py b/outlines/fsm/json_schema.py
index b29243001..98d2de59c 100644
--- a/outlines/fsm/json_schema.py
+++ b/outlines/fsm/json_schema.py
@@ -2,10 +2,10 @@
 import json
 import re
 import warnings
-from typing import Callable, Optional, Tuple
+from typing import Callable, Optional, Tuple, Type, Union
 
 from jsonschema.protocols import Validator
-from pydantic import create_model
+from pydantic import BaseModel, create_model
 from referencing import Registry, Resource
 from referencing._core import Resolver
 from referencing.jsonschema import DRAFT202012
@@ -86,6 +86,39 @@ def build_regex_from_schema(schema: str, whitespace_pattern: Optional[str] = Non
     return to_regex(resolver, content, whitespace_pattern)
 
 
+def convert_json_schema_to_str(json_schema: Union[dict, str, Type[BaseModel]]) -> str:
+    """Convert a JSON schema to a string.
+
+    Parameters
+    ----------
+    json_schema
+        The JSON schema.
+
+    Returns
+    -------
+    str
+        The JSON schema converted to a string.
+
+    Raises
+    ------
+    ValueError
+        If the schema is not a dictionary, a string or a Pydantic class.
+    """
+    if isinstance(json_schema, dict):
+        schema_str = json.dumps(json_schema)
+    elif isinstance(json_schema, str):
+        schema_str = json_schema
+    elif issubclass(json_schema, BaseModel):
+        schema_str = json.dumps(json_schema.model_json_schema())
+    else:
+        raise ValueError(
+            f"Cannot parse schema {json_schema}. The schema must be either "
+            + "a Pydantic class, a dictionary or a string that contains the JSON "
+            + "schema specification"
+        )
+    return schema_str
+
+
 def _get_num_items_pattern(min_items, max_items, whitespace_pattern):
     # Helper function for arrays and objects
     min_items = int(min_items or 0)
diff --git a/outlines/integrations/utils.py b/outlines/integrations/utils.py
index 9ac4e2a4f..edf77e5c7 100644
--- a/outlines/integrations/utils.py
+++ b/outlines/integrations/utils.py
@@ -68,36 +68,3 @@ def convert_token_to_string(token: Union[str, bytes]) -> str:
     tokenizer.convert_token_to_string = convert_token_to_string
 
     return tokenizer
-
-
-def convert_json_schema_to_str(json_schema: Union[dict, str, Type[BaseModel]]) -> str:
-    """Convert a JSON schema to a string.
-
-    Parameters
-    ----------
-    json_schema
-        The JSON schema.
-
-    Returns
-    -------
-    str
-        The JSON schema converted to a string.
-
-    Raises
-    ------
-    ValueError
-        If the schema is not a dictionary, a string or a Pydantic class.
-    """
-    if isinstance(json_schema, dict):
-        schema_str = json.dumps(json_schema)
-    elif isinstance(json_schema, str):
-        schema_str = json_schema
-    elif issubclass(json_schema, BaseModel):
-        schema_str = json.dumps(json_schema.model_json_schema())
-    else:
-        raise ValueError(
-            f"Cannot parse schema {json_schema}. The schema must be either "
-            + "a Pydantic class, a dictionary or a string that contains the JSON "
-            + "schema specification"
-        )
-    return schema_str
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 0966a90db..f8d63d7b5 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -30,8 +30,7 @@
 from pydantic import BaseModel
 
 from outlines.fsm.guide import CFGGuide, Guide, RegexGuide
-from outlines.fsm.json_schema import build_regex_from_schema
-from outlines.integrations.utils import convert_json_schema_to_str
+from outlines.fsm.json_schema import build_regex_from_schema, convert_json_schema_to_str
 
 from .base_logits_processor import OutlinesLogitsProcessor
 

From 0a70c42fea61fd8c8d1679f687315081f9ee8377 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Mon, 22 Jul 2024 15:53:02 +0200
Subject: [PATCH 053/505] Move function that adapts the tokenizer

---
 outlines/integrations/__init__.py |   1 -
 outlines/integrations/utils.py    |  70 -------------
 outlines/integrations/vllm.py     | 158 ------------------------------
 outlines/models/vllm.py           |  41 +++++++-
 outlines/serve/serve.py           |   9 +-
 outlines/serve/vllm.py            |   4 -
 tests/fsm/test_regex.py           |   3 +-
 7 files changed, 47 insertions(+), 239 deletions(-)
 delete mode 100644 outlines/integrations/__init__.py
 delete mode 100644 outlines/integrations/utils.py
 delete mode 100644 outlines/integrations/vllm.py
 delete mode 100644 outlines/serve/vllm.py

diff --git a/outlines/integrations/__init__.py b/outlines/integrations/__init__.py
deleted file mode 100644
index b0a90d5ea..000000000
--- a/outlines/integrations/__init__.py
+++ /dev/null
@@ -1 +0,0 @@
-"""Utility functions and classes used to integrate `outlines` into other packages."""
diff --git a/outlines/integrations/utils.py b/outlines/integrations/utils.py
deleted file mode 100644
index edf77e5c7..000000000
--- a/outlines/integrations/utils.py
+++ /dev/null
@@ -1,70 +0,0 @@
-"""Utility functions used in integrations with other packages.
-
- _______________________________
-/ Don't want to self-host?       \
-\\ Try .json at http://dottxt.co /
- -------------------------------
-       \\   ^__^
-        \\  (oo)\\_______
-            (__)\\       )\\/\
-                ||----w |
-                ||     ||
-
-Copyright 2024- the Outlines developers
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-"""
-
-import json
-from typing import Type, Union
-
-from pydantic import BaseModel
-from transformers import SPIECE_UNDERLINE, PreTrainedTokenizerBase
-
-
-def adapt_tokenizer(tokenizer: PreTrainedTokenizerBase) -> PreTrainedTokenizerBase:
-    """Adapt a tokenizer to use to compile the FSM.
-
-    The API of Outlines tokenizers is slightly different to that of `transformers`. In
-    addition we need to handle the missing spaces to Llama's tokenizer to be able to
-    compile FSMs for this model.
-
-    Parameters
-    ----------
-    tokenizer
-        The tokenizer of the model.
-
-    Returns
-    -------
-    PreTrainedTokenizerBase
-        The adapted tokenizer.
-    """
-    tokenizer.vocabulary = tokenizer.get_vocab()
-    tokenizer.special_tokens = set(tokenizer.all_special_tokens)
-
-    def convert_token_to_string(token: Union[str, bytes]) -> str:
-        string = tokenizer.convert_tokens_to_string([token])
-
-        # A hack to handle missing spaces to HF's Llama tokenizers
-        if (
-            type(token) is str
-            and token.startswith(SPIECE_UNDERLINE)
-            or token == "<0x20>"
-        ):
-            return " " + string
-
-        return string
-
-    tokenizer.convert_token_to_string = convert_token_to_string
-
-    return tokenizer
diff --git a/outlines/integrations/vllm.py b/outlines/integrations/vllm.py
deleted file mode 100644
index 2a5f26e35..000000000
--- a/outlines/integrations/vllm.py
+++ /dev/null
@@ -1,158 +0,0 @@
-"""Make vLLM compatible with Outlines' structured generation.
-
- _______________________________
-/ Don't want to self-host?       \
-\\ Try .json at http://dottxt.co /
- -------------------------------
-       \\   ^__^
-        \\  (oo)\\_______
-            (__)\\       )\\/\
-                ||----w |
-                ||     ||
-
-Copyright 2024- the Outlines developers
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-"""
-
-import math
-from collections import defaultdict
-from typing import TYPE_CHECKING, DefaultDict, Dict, List, Optional, Type, Union
-
-import torch
-from pydantic import BaseModel
-
-from outlines.fsm.guide import RegexGuide
-from outlines.fsm.json_schema import build_regex_from_schema
-from outlines.integrations.utils import adapt_tokenizer, convert_json_schema_to_str
-
-if TYPE_CHECKING:
-    from vllm import LLM
-
-
-class RegexLogitsProcessor:
-    """Bias vLLM generation based on a regular expression.
-
-    Attributes
-    ----------
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(self, regex_string: str, llm: "LLM"):
-        """Compile the FSM that drives the regex-structured generation.
-
-        Parameters
-        ----------
-        regex_string
-            A string that represents a regular expression.
-        llm
-            The vLLM model.
-
-        Raises
-        ------
-        ValueError
-            If the provided LLM instance in `RegexLogitsProcessor` neither has a
-            `tokenizer` attribute or a `get_tokenizer` method.
-        """
-        if hasattr(llm, "get_tokenizer"):
-            tokenizer = llm.get_tokenizer()
-        elif hasattr(llm, "tokenizer"):
-            if hasattr(llm.tokenizer, "tokenizer"):
-                tokenizer = llm.tokenizer.tokenizer
-            else:
-                tokenizer = llm.tokenizer
-        else:
-            raise ValueError(
-                "The provided LLM instance in `RegexLogitsProcessor` neither has a "
-                "`tokenizer` attribute or a `get_tokenizer` method."
-            )
-        tokenizer = adapt_tokenizer(tokenizer=tokenizer)
-        self.mask_cache: Dict[int, torch.Tensor] = {}
-        self.fsm = RegexGuide(regex_string, tokenizer)
-        self._fsm_state: DefaultDict[int, int] = defaultdict(int)
-
-    def __call__(self, input_ids: List[int], scores: torch.Tensor) -> torch.Tensor:
-        """Use the FSM to bias the logits before sampling the next token.
-
-        Parameters
-        ----------
-        input_ids
-            The tokens of the current sentence.
-        scores
-            The logits of the current sentence.
-
-        Returns
-        -------
-        torch.Tensor
-            The biased logits.
-        """
-        seq_id = hash(tuple(input_ids))
-
-        # Initialize the FSM state dictionary if the input_ids are empty, as this means
-        # that the input_ids are the first tokens of the sequence.
-        if len(input_ids) > 0:
-            last_token = input_ids[-1]
-            last_seq_id = hash(tuple(input_ids[:-1]))
-            self._fsm_state[seq_id] = self.fsm.get_next_state(
-                state=self._fsm_state[last_seq_id], token_id=last_token
-            )
-
-        state_id = self._fsm_state[seq_id]
-        if state_id not in self.mask_cache:
-            allowed_tokens = self.fsm.get_next_instruction(
-                state=self._fsm_state[seq_id]
-            ).tokens
-            mask = torch.full((scores.shape[-1],), -math.inf)
-            mask[allowed_tokens] = 0
-            mask = mask.pin_memory()
-            self.mask_cache[state_id] = mask
-        else:
-            mask = self.mask_cache[state_id]
-        mask = mask.to(device=scores.device, non_blocking=True)
-        biased_scores = scores + mask
-
-        return biased_scores
-
-
-class JSONLogitsProcessor(RegexLogitsProcessor):
-    """Bias vLLM generation based on a JSON schema.
-
-    Attributes
-    ----------
-    fsm
-        The finite state machine which is used to bias the logits.
-    """
-
-    def __init__(
-        self,
-        schema: Union[dict, Type[BaseModel], str],
-        llm: "LLM",
-        whitespace_pattern: Optional[str] = None,
-    ):
-        """Compile the FSM that drives the JSON-guided generation.
-
-        Parameters
-        ----------
-        schema
-            A JSON schema that encodes the structure we want the model to generate.
-        llm
-            The vLLM model.
-        whitespace_pattern
-            Pattern to use for JSON syntactic whitespace (doesn't impact string
-            literals). For example, to allow only a single space or newline with
-            `whitespace_pattern=r"[\n ]?"`
-        """
-        schema_str = convert_json_schema_to_str(json_schema=schema)
-        regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(regex_string=regex_string, llm=llm)
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 2ae3d99d4..d1f97bde2 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,8 +1,9 @@
 import dataclasses
 from typing import TYPE_CHECKING, List, Optional, Union
 
+from transformers import SPIECE_UNDERLINE, PreTrainedTokenizerBase
+
 from outlines.generate.api import GenerationParameters, SamplingParameters
-from outlines.integrations.utils import adapt_tokenizer
 
 if TYPE_CHECKING:
     from vllm import LLM
@@ -185,3 +186,41 @@ def vllm(model_name: str, **vllm_model_params):
     model = LLM(model_name, **vllm_model_params)
 
     return VLLM(model)
+
+
+def adapt_tokenizer(tokenizer: PreTrainedTokenizerBase) -> PreTrainedTokenizerBase:
+    """Adapt a tokenizer to use to compile the FSM.
+
+    The API of Outlines tokenizers is slightly different to that of `transformers`. In
+    addition we need to handle the missing spaces to Llama's tokenizer to be able to
+    compile FSMs for this model.
+
+    Parameters
+    ----------
+    tokenizer
+        The tokenizer of the model.
+
+    Returns
+    -------
+    PreTrainedTokenizerBase
+        The adapted tokenizer.
+    """
+    tokenizer.vocabulary = tokenizer.get_vocab()
+    tokenizer.special_tokens = set(tokenizer.all_special_tokens)
+
+    def convert_token_to_string(token: Union[str, bytes]) -> str:
+        string = tokenizer.convert_tokens_to_string([token])
+
+        # A hack to handle missing spaces to HF's Llama tokenizers
+        if (
+            type(token) is str
+            and token.startswith(SPIECE_UNDERLINE)
+            or token == "<0x20>"
+        ):
+            return " " + string
+
+        return string
+
+    tokenizer.convert_token_to_string = convert_token_to_string
+
+    return tokenizer
diff --git a/outlines/serve/serve.py b/outlines/serve/serve.py
index fb8c80139..998fbc459 100644
--- a/outlines/serve/serve.py
+++ b/outlines/serve/serve.py
@@ -35,12 +35,14 @@
 from vllm.sampling_params import SamplingParams
 from vllm.utils import random_uuid
 
-from outlines.integrations.vllm import JSONLogitsProcessor, RegexLogitsProcessor
+from outlines.models.vllm import adapt_tokenizer
+from outlines.processors import JSONLogitsProcessor, RegexLogitsProcessor
 
 TIMEOUT_KEEP_ALIVE = 5  # seconds.
 TIMEOUT_TO_PREVENT_DEADLOCK = 1  # seconds.
 app = FastAPI()
 engine = None
+tokenizer = None
 
 
 @app.get("/health")
@@ -69,9 +71,9 @@ async def generate(request: Request) -> Response:
     json_schema = request_dict.pop("schema", None)
     regex_string = request_dict.pop("regex", None)
     if json_schema is not None:
-        logits_processors = [JSONLogitsProcessor(json_schema, engine.engine)]
+        logits_processors = [JSONLogitsProcessor(json_schema, tokenizer)]
     elif regex_string is not None:
-        logits_processors = [RegexLogitsProcessor(regex_string, engine.engine)]
+        logits_processors = [RegexLogitsProcessor(regex_string, tokenizer)]
     else:
         logits_processors = []
 
@@ -124,6 +126,7 @@ async def stream_results() -> AsyncGenerator[bytes, None]:
 
     # Sets default for the model (`facebook/opt-125m`)
     engine = AsyncLLMEngine.from_engine_args(engine_args)
+    tokenizer = adapt_tokenizer(tokenizer=engine.engine.tokenizer.tokenizer)
 
     uvicorn.run(
         app,
diff --git a/outlines/serve/vllm.py b/outlines/serve/vllm.py
deleted file mode 100644
index ddc50b47d..000000000
--- a/outlines/serve/vllm.py
+++ /dev/null
@@ -1,4 +0,0 @@
-from outlines.integrations.vllm import (  # noqa[F401]
-    JSONLogitsProcessor,
-    RegexLogitsProcessor,
-)
diff --git a/tests/fsm/test_regex.py b/tests/fsm/test_regex.py
index fa72ad0dc..f7aa4ae67 100644
--- a/tests/fsm/test_regex.py
+++ b/tests/fsm/test_regex.py
@@ -18,7 +18,6 @@
     reduced_vocabulary,
     walk_fsm,
 )
-from outlines.integrations.utils import adapt_tokenizer
 from outlines.models.transformers import TransformerTokenizer
 
 
@@ -714,6 +713,6 @@ def test_reduced_vocabulary_with_rare_tokens(rare_token):
     [2]: https://github.com/outlines-dev/outlines/pull/948
     """
     tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2")
-    tokenizer = adapt_tokenizer(tokenizer=tokenizer)
+    tokenizer = TransformerTokenizer(tokenizer=tokenizer)
     tokenizer.vocabulary[rare_token] = max(tokenizer.vocabulary.values()) + 1
     reduced_vocabulary(tokenizer)

From 900762b0f240c6220549d7d8594a221e6c9845e8 Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Fri, 23 Aug 2024 13:17:51 -0500
Subject: [PATCH 054/505] Use relative coverage source paths

---
 .github/workflows/tests.yml | 13 +++++--------
 pyproject.toml              |  1 +
 2 files changed, 6 insertions(+), 8 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 10879c78f..b76d723f2 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -36,11 +36,13 @@ jobs:
     - name: Run tests
       run: |
         pytest --cov=outlines
+      env:
+        COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
     - name: Upload coverage data
       uses: actions/upload-artifact@v3
       with:
         name: coverage-data
-        path: .coverage*
+        path: .coverage.*
         if-no-files-found: ignore
 
   coverage:
@@ -66,19 +68,14 @@ jobs:
         with:
           name: coverage-data
 
-      - name: Fetch master for coverage diff
+      - name: Fetch main for coverage diff
         run: |
           git fetch --no-tags --prune origin main
 
       - name: Combine coverage & fail if it's <100%.
         run: |
-          # Combine coverage files (not needed now, but maybe later)
-          # python -m coverage combine
-
-          # Produce an html report with absolute coverage information
+          python -m coverage combine
           python -m coverage html --skip-covered --skip-empty
-
-          # Report relative coverage and write to the workflow's summary
           python -m coverage xml
           diff-cover coverage.xml --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
 
diff --git a/pyproject.toml b/pyproject.toml
index ea0ac11c8..99d4f94e1 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -147,6 +147,7 @@ omit = [
     "tests/*",
 ]
 branch = true
+relative_files = true
 
 [tool.coverage.report]
 omit = [

From ce029a8052c4d631a7e4786d6228411c616d9e8f Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Wed, 28 Aug 2024 18:54:42 -0500
Subject: [PATCH 055/505] Add missing CI matrix step and use dev installs

---
 .github/workflows/tests.yml | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index b76d723f2..344dfb55e 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -32,7 +32,16 @@ jobs:
     - name: Set up test environment
       run: |
         python -m pip install --upgrade pip
-        pip install .[test]
+        pip install -e .[test]
+    - name: Create matrix id
+      id: matrix-id
+      env:
+        MATRIX_CONTEXT: ${{ toJson(matrix) }}
+      run: |
+        echo $MATRIX_CONTEXT
+        export MATRIX_ID=`echo $MATRIX_CONTEXT | md5sum | cut -c 1-32`
+        echo $MATRIX_ID
+        echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
         pytest --cov=outlines

From 067aedb49dc6eb530af5cbef8c1d0da553382e61 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Wed, 28 Aug 2024 23:27:57 -0700
Subject: [PATCH 056/505] Update modal example (#1111)

I updated some of the text of the modal example to add more information
on what's happening where, since Modal's general structure may be
unfamiliar to new users.

List of changes:

- Removed version constraints on the image. I think this is a
questionable choice so I'd love input -- in general, this should
probably better be addressed by versioning the docs #999. There's an
open PR #1059 here that can help with this too -- I'm happy to revert to
pinned versions, or at least update to the current version of outlines
(0.0.46).
- Added a brief note that the Mistral model is gated and requires you to
request access on HuggingFace.
- Moved the environment variable setting from `import_model` to a `.env`
call when the image is created. This is more idiomatic Modal code. It
was previously addressed in #1058, but this should be more current.
- More comments and prep.
---
 docs/cookbook/deploy-using-modal.md | 104 +++++++++++++++++++++++-----
 1 file changed, 87 insertions(+), 17 deletions(-)

diff --git a/docs/cookbook/deploy-using-modal.md b/docs/cookbook/deploy-using-modal.md
index 835924d2b..998e5d835 100644
--- a/docs/cookbook/deploy-using-modal.md
+++ b/docs/cookbook/deploy-using-modal.md
@@ -4,36 +4,83 @@
 
 In this guide we will show you how you can use Modal to run programs written with Outlines on GPU in the cloud.
 
+## Requirements
+
+We recommend installing `modal` and `outlines` in a virtual environment. You can create one with:
+
+```bash
+python -m venv venv
+source venv/bin/activate
+```
+
+Then install the required packages:
+
+```bash
+pip install modal outlines
+```
+
 ## Build the image
 
-First we need to define our container image. We download the Mistral-7B-v0.1 model from HuggingFace as part of the definition of the image so it only needs to be done once (you need to provide an [access token](https://huggingface.co/settings/tokens))
+First we need to define our container image. If you need to access a gated model, you will need to provide an [access token](https://huggingface.co/settings/tokens). See the `.env` call below for how to provide a HuggingFace token.
+
+Setting a token is best done by setting an environment variable `HF_TOKEN` with your token. If you do not wish to do this, we provide a commented-out line in the code to set the token directly in the code.
 
 ```python
 from modal import Image, App, gpu
+import os
 
+# This creates a modal App object. Here we set the name to "outlines-app".
+# There are other optional parameters like modal secrets, schedules, etc.
+# See the documentation here: https://modal.com/docs/reference/modal.App
 app = App(name="outlines-app")
 
+# Specify a language model to use.
+# Another good model to use is "NousResearch/Hermes-2-Pro-Mistral-7B"
+language_model = "mistral-community/Mistral-7B-v0.2"
+
+# Please set an environment variable HF_TOKEN with your Hugging Face API token.
+# The code below (the .env({...}) part) will copy the token from your local
+# environment to the container.
+# More info on Image here: https://modal.com/docs/reference/modal.Image
 outlines_image = Image.debian_slim(python_version="3.11").pip_install(
-    "outlines==0.0.37",
-    "transformers==4.38.2",
-    "datasets==2.18.0",
-    "accelerate==0.27.2",
-)
+    "outlines",
+    "transformers",
+    "datasets",
+    "accelerate",
+    "sentencepiece",
+).env({
+    # This will pull in your HF_TOKEN environment variable if you have one.
+    'HF_TOKEN':os.environ['HF_TOKEN']
+
+    # To set the token directly in the code, uncomment the line below and replace
+    # 'YOUR_TOKEN' with the HuggingFace access token.
+    # 'HF_TOKEN':'YOUR_TOKEN'
+})
+```
+
+## Setting the container up
 
+When running longer Modal apps, it's recommended to download your language model when the container starts, rather than when the function is called. This will cache the model for future runs.
+
+```python
+# This function imports the model from Hugging Face. The modal container
+# will call this function when it starts up. This is useful for
+# downloading models, setting up environment variables, etc.
 def import_model():
-    import os
-    os.environ["HF_TOKEN"] = "YOUR_HUGGINGFACE_TOKEN"
     import outlines
-    outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+    outlines.models.transformers(language_model)
 
+# This line tells the container to run the import_model function when it starts.
 outlines_image = outlines_image.run_function(import_model)
 ```
 
-We will run the JSON-structured generation example [in the README](https://github.com/outlines-dev/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
+## Define a schema
 
-## Run inference
+We will run the JSON-structured generation example [in the README](https://github.com/outlines-dev/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
 
 ```python
+# Specify a schema for the character description. In this case,
+# we want to generate a character with a name, age, armor, weapon, and strength.
 schema = """{
     "title": "Character",
     "type": "object",
@@ -72,48 +119,71 @@ schema = """{
 }"""
 ```
 
-To make the inference work on Modal we need to wrap the corresponding function in a `@app.function` decorator. We pass to this decorator the image and GPU on which we want this function to run (here an A100 with 80GB memory):
+To make the inference work on Modal we need to wrap the corresponding function in a `@app.function` decorator. We pass to this decorator the image and GPU on which we want this function to run.
+
+Let's choose an A100 with 80GB memory. Valid GPUs can be found [here](https://modal.com/docs/reference/modal.gpu).
 
 ```python
+# Define a function that uses the image we chose, and specify the GPU
+# and memory we want to use.
 @app.function(image=outlines_image, gpu=gpu.A100(size='80GB'))
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
+    # Remember, this function is being executed in the container,
+    # so we need to import the necessary libraries here. You should
+    # do this with any other libraries you might need.
     import outlines
 
+    # Load the model into memory. The import_model function above
+    # should have already downloaded the model, so this call
+    # only loads the model into GPU memory.
     model = outlines.models.transformers(
-        "mistralai/Mistral-7B-v0.1", device="cuda"
+        language_model, device="cuda"
     )
 
+    # Generate a character description based on the prompt.
+    # We use the .json generation method -- we provide the
+    # - model: the model we loaded above
+    # - schema: the JSON schema we defined above
     generator = outlines.generate.json(model, schema)
+
+    # Make sure you wrap your prompt in instruction tags ([INST] and [/INST])
+    # to indicate that the prompt is an instruction. Instruction tags can vary
+    # by models, so make sure to check the model's documentation.
     character = generator(
         f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
     )
 
+    # Print out the generated character.
     print(character)
 ```
 
-We then need to define a `local_entrypoint` to call our function `generate` remotely:
+We then need to define a `local_entrypoint` to call our function `generate` remotely.
 
 ```python
 @app.local_entrypoint()
 def main(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
+    # We use the "generate" function defined above -- note too that we are calling
+    # .remote() on the function. This tells modal to run the function in our cloud
+    # machine. If you want to run the function locally, you can call .local() instead,
+    # though this will require additional setup.
     generate.remote(prompt)
 ```
 
-Here `@app.local_entrypoint()` decorator defines `main` as the function to start from locally when running the Modal CLI. You can save above code to `example.py` (or use [this implementation](https://github.com/outlines-dev/outlines/blob/main/examples/modal_example.py)). Let's now see how to run the code on the cloud using the Modal CLI.
+Here `@app.local_entrypoint()` decorator defines `main` as the function to start from locally when using the Modal CLI. You can save above code to `example.py` (or use [this implementation](https://github.com/outlines-dev/outlines/blob/main/examples/modal_example.py)). Let's now see how to run the code on the cloud using the Modal CLI.
 
 ## Run on the cloud
 
-First install the Modal client from PyPi:
+First install the Modal client from PyPi, if you have not already:
 
 ```bash
 pip install modal
 ```
 
-You then need to obtain a token from Modal. To do so easily, run the following command:
+You then need to obtain a token from Modal. Run the following command:
 
 ```bash
 modal setup

From 7cf63d6b7ed5d27863234fafbfe812fbd706c6bc Mon Sep 17 00:00:00 2001
From: Bill Metangmo <25366207+billmetangmo@users.noreply.github.com>
Date: Thu, 29 Aug 2024 17:45:57 +0200
Subject: [PATCH 057/505] Fix typo (#1120)

---
 docs/reference/serve/vllm.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/serve/vllm.md b/docs/reference/serve/vllm.md
index 0b584568b..8e2886c96 100644
--- a/docs/reference/serve/vllm.md
+++ b/docs/reference/serve/vllm.md
@@ -2,7 +2,7 @@
 
 !!! tip "Would rather not self-host?"
 
-    If you want to get started quickly with JSON-structured generaton you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
+    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
 
 Outlines can be deployed as an LLM service using the vLLM inference engine and a FastAPI server. vLLM is not installed by default so will need to install Outlines with:
 

From 18eafb0944a7de6c8e801ec5454c84e88c4ddd0d Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Fri, 30 Aug 2024 13:59:35 -0400
Subject: [PATCH 058/505] pass text and images as kwargs to VLM processor

---
 outlines/models/transformers_vision.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
index 9709c26ae..772645b80 100644
--- a/outlines/models/transformers_vision.py
+++ b/outlines/models/transformers_vision.py
@@ -43,9 +43,9 @@ def generate(  # type: ignore
         -------
         The generated text
         """
-        inputs = self.processor(prompts, media, padding=True, return_tensors="pt").to(
-            self.model.device
-        )
+        inputs = self.processor(
+            text=prompts, images=media, padding=True, return_tensors="pt"
+        ).to(self.model.device)
 
         generation_kwargs = self._get_generation_kwargs(
             prompts,

From f941fc79a5b7e2068705b4375027526af5f73509 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 27 Jul 2024 10:55:30 -0500
Subject: [PATCH 059/505] Update CFGGuide to use outlines.fsm.parsing. Enable
 generate.cfg

---
 benchmarks/bench_cfg_guide.py                 |  61 +++
 .../reference/generation/creating_grammars.md |  99 ++++
 mkdocs.yml                                    |   1 +
 outlines/fsm/fsm.py                           |  24 +-
 outlines/fsm/guide.py                         | 282 +++++------
 outlines/fsm/parsing.py                       |  67 ++-
 outlines/generate/cfg.py                      |  35 +-
 outlines/generate/fsm.py                      |  12 +-
 outlines/grammars/common.lark                 |   9 +-
 outlines/grammars/json.lark                   |   6 +-
 outlines/processors/__init__.py               |   2 +-
 outlines/processors/structured.py             | 112 +++--
 .../arithmetic/lots_of_ops.arithmetic.test    |   1 +
 .../arithmetic/simple_math.arithmetic.test    |   1 +
 .../outlines.generate.samplers.mypy.json.test | 372 ++++++++++++++
 tests/cfg_samples/json/simple_fruit.json.test |  20 +
 .../json/simple_fruit_no_indent.json.test     |   1 +
 tests/fsm/test_cfg_guide.py                   | 457 ++++++++++++++++++
 tests/fsm/test_fsm.py                         | 262 +---------
 tests/fsm/test_guide.py                       | 292 ++++++-----
 tests/generate/test_generate.py               |  34 ++
 tests/generate/test_integration_llamacpp.py   |  18 +-
 .../generate/test_integration_transformers.py |   8 +-
 23 files changed, 1520 insertions(+), 656 deletions(-)
 create mode 100644 benchmarks/bench_cfg_guide.py
 create mode 100644 docs/reference/generation/creating_grammars.md
 create mode 100644 tests/cfg_samples/arithmetic/lots_of_ops.arithmetic.test
 create mode 100644 tests/cfg_samples/arithmetic/simple_math.arithmetic.test
 create mode 100644 tests/cfg_samples/json/outlines.generate.samplers.mypy.json.test
 create mode 100644 tests/cfg_samples/json/simple_fruit.json.test
 create mode 100644 tests/cfg_samples/json/simple_fruit_no_indent.json.test
 create mode 100644 tests/fsm/test_cfg_guide.py

diff --git a/benchmarks/bench_cfg_guide.py b/benchmarks/bench_cfg_guide.py
new file mode 100644
index 000000000..8f6de914a
--- /dev/null
+++ b/benchmarks/bench_cfg_guide.py
@@ -0,0 +1,61 @@
+import random
+
+from transformers import AutoTokenizer
+
+import outlines.grammars
+from outlines.caching import cache_disabled
+from outlines.fsm.guide import CFGGuide
+from outlines.models.transformers import TransformerTokenizer
+
+from .common import ensure_numba_compiled
+
+random.seed(42)
+
+
+def get_tiny_tokenizer():
+    """1000 tokens in vocabulary"""
+    return TransformerTokenizer(
+        AutoTokenizer.from_pretrained("hf-internal-testing/tiny-random-gpt2")
+    )
+
+
+benched_grammars = {
+    "json": outlines.grammars.json,
+    "arithmetic": outlines.grammars.arithmetic,
+}
+
+
+class CFGGuideBenchmark:
+    params = benched_grammars.keys()
+
+    def setup(self, grammar_name):
+        self.tokenizer = get_tiny_tokenizer()
+        ensure_numba_compiled(
+            self.tokenizer
+        )  # numba not currently used, but will be in the future
+        self.prebuilt_cfg_guide = CFGGuide(
+            benched_grammars[grammar_name], self.tokenizer
+        )
+
+    @staticmethod
+    def _run_random_cfg(guide):
+        state = guide.initial_state
+        token_ids = list(guide.tokenizer.vocabulary.values())
+        for i in range(40):
+            # simulate ordering of logits top prob to lowest prob
+            random.shuffle(token_ids)
+            # simulate sampling and state update
+            next_token_id = next(guide.iter_valid_token_ids(state, token_ids))
+            state = guide.get_next_state(state, next_token_id)
+
+    @cache_disabled()
+    def time_cfg_guide_setup(self, grammar_name):
+        CFGGuide(benched_grammars[grammar_name], self.tokenizer)
+
+    @cache_disabled()
+    def time_cfg_guide_run(self, grammar):
+        self._run_random_cfg(self.prebuilt_cfg_guide)
+
+    @cache_disabled()
+    def peakmem_cfg_guide_run(self, grammar):
+        self._run_random_cfg(self.prebuilt_cfg_guide)
diff --git a/docs/reference/generation/creating_grammars.md b/docs/reference/generation/creating_grammars.md
new file mode 100644
index 000000000..78e41282a
--- /dev/null
+++ b/docs/reference/generation/creating_grammars.md
@@ -0,0 +1,99 @@
+# Overview
+
+Outlines allows the use of [Lark](https://github.com/lark-parser/lark) grammars to guide generation. These grammars are used to construct parsers that filter out incompatible tokens during the generation process The result is a generation that adheres to the grammar's production rules.
+
+# Primer on Creating Grammars
+
+To create grammars for Outlines, a solid understanding of Lark grammars is necessary. Here's how you can get started:
+
+- Read Lark's grammars documentations [here](https://lark-parser.readthedocs.io/en/latest/grammar.html).
+- Review Outlines' existing grammars [here](/outlines/grammars).
+
+
+# Compatibility With Outlines
+
+It's important to note that not all Lark grammars work with Outlines. Changes may be necessary to ensure compatability.
+
+### LALR(1) Parser
+
+Outlines utilizes Larks LALR(1) parser, meaning the grammar must be unambiguous at least up to the next token (one token lookahead). Read Lark's official LALR(1) parser documentation [here](https://lark-parser.readthedocs.io/en/stable/parsers.html#lalr-1).
+
+If your grammar is ambiguous, you will recieve the following error at runtime:
+
+```
+GrammarError: Reduce/Reduce collision in Terminal('B') between the following rules:
+```
+
+### Regex Terminal Restrictions
+
+Outlines converts terminals to finite state machines using the [Interegular](https://github.com/MegaIng/interegular/) library. Not all regular expressions work with Interegular, mitigation is described in the subsections which follow.
+
+
+#### Avoid Lookarounds
+
+Examples of removing lookaround while maintaining the same functionality
+
+##### Example: Escaped String
+
+From Outlines' modified `ESCAPED_STRING` in [common.lark](/outlines/grammars/common.lark).
+
+Before:
+```
+_STRING_INNER: /.*?/
+_STRING_ESC_INNER: _STRING_INNER /(?<!\\)(\\\\)*?/
+
+ESCAPED_STRING : "\"" _STRING_ESC_INNER "\""
+```
+
+After:
+```
+_NON_CONTROL_CHAR: /([^"\\\x00-\x1F\x7F-\x9F])/
+_ESCAPED_CHAR: /\\/ (_NON_CONTROL_CHAR | /\\/ | /"/)
+ESCAPED_STRING_INNER: _NON_CONTROL_CHAR | _ESCAPED_CHAR
+ESCAPED_STRING: /"/ ESCAPED_STRING_INNER* /"/
+```
+
+#### Avoid Backreferences
+
+Backreferences, for example `([ab]^*)\1`, cannot be simulated by a finite state machine, and will result in an error if used.
+
+# Creating a Valid Grammar
+
+You can use Outlines' test suite to verify your grammar.
+
+### 1) Create Your Grammar
+
+Create your grammar file named `your_new_grammar.lark`, adhering to the guidelines provided above. Add it to `outlines/grammars/` (ensure attribution is included and license is compatible).
+
+Update `outlines/grammars.py` with a line including your grammar.
+
+### 2) Test Your Grammar
+
+Test grammar for false negatives, ensure sample grammars can be generated:
+- Add valid example outputs which are compliant with the grammar to `tests/benchmark/cfg_samples/your_new_grammar/`
+- Run the tests for your grammar via `pytest -s tests/fsm/test_cfg_guide.py::test_cfg_grammar_sample -k "your_new_grammar"`
+
+Test grammar for false positives, ensure invalid outputs aren't generated.
+
+Currently there isn't a builtin false positive testing utility. It is recommended you smoke test via
+```
+from outlines import models, generate, grammars
+model = models.transformers("mistralai/Mistral-7B-v0.1")
+generator = generate.cfg(model, grammars.your_new_grammar)
+result = generator(<your prompt to generate output for your grammar>)
+print(result)
+```
+
+# Converting
+There are a few tools available for converting from other grammars to lark. These tools serve as a starting point. However, you will typically need to make additional adjustments to ensure full compatibility and proper functioning within Outlines.
+
+Tools:
+- Larks built in "Nearley-to-Lark" converter https://lark-parser.readthedocs.io/en/latest/tools.html
+- Convert ANTLR4 to Lark (Note, most antlr4 grammars are not LALR(1) compatible, so will require additional tweaking) https://github.com/kaby76/Domemtech.Trash/blob/main/src/trconvert/readme.md
+- Extract EBNF from Yacc files https://www.bottlecaps.de/rr/ui
+
+Reference Grammars:
+- Github Lark Grammars https://github.com/search?q=path%3A*.lark&type=code
+- Github Nearley Grammars https://github.com/search?q=path%3A*.ne+%22-%3E%22&type=code
+- Antlr4 grammars https://github.com/antlr/grammars-v4/
+- Grammar zoo https://slebok.github.io/zoo/index.html#html
diff --git a/mkdocs.yml b/mkdocs.yml
index afc56528b..bf863085a 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -131,6 +131,7 @@ nav:
             - Type constraints: reference/generation/format.md
             - JSON (function calling): reference/generation/json.md
             - Grammar: reference/generation/cfg.md
+            - Creating Grammars: reference/generation/creating_grammars.md
             - Custom FSM operations: reference/generation/custom_fsm_ops.md
     - Utilities:
         - Serve with vLLM: reference/serve/vllm.md
diff --git a/outlines/fsm/fsm.py b/outlines/fsm/fsm.py
index 4a7fce8c9..bfcf55c03 100644
--- a/outlines/fsm/fsm.py
+++ b/outlines/fsm/fsm.py
@@ -1,7 +1,7 @@
 import warnings
 from typing import TYPE_CHECKING, Iterable, NewType, Optional
 
-from outlines.fsm.guide import CFGGuide, RegexGuide, StopAtEOSGuide
+from outlines.fsm.guide import RegexGuide, StopAtEOSGuide
 
 if TYPE_CHECKING:
     from outlines.models.tokenizer import Tokenizer
@@ -45,25 +45,3 @@ def allowed_token_ids(self, state: FSMState) -> Optional[Iterable[int]]:
 
     def next_state(self, state: FSMState, token_id: int) -> FSMState:
         return FSMState(self.get_next_state(state, token_id))
-
-
-class CFGFSM(CFGGuide):
-    """FSM to generate text that is in the language of a context-free grammar."""
-
-    def __init__(self, cfg_string: str, tokenizer):
-        warnings.warn(
-            UserWarning(
-                "The `CFGFSM` interface is deprecated and will be removed on 2024-06-01. Please use `CFGGuide` instead."
-            )
-        )
-        super().__init__(cfg_string, tokenizer)
-
-    def allowed_token_ids(self, state: FSMState) -> Optional[Iterable[int]]:
-        return self.get_next_instruction(state).tokens
-
-    def next_state(self, state: FSMState, token_id: int) -> FSMState:
-        return FSMState(self.get_next_state(state, token_id))
-
-    def copy(self) -> "CFGFSM":
-        """Create a copy of the FSM."""
-        return CFGFSM(self.cfg_string, self.tokenizer)
diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index 2e4415148..aa073d107 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -1,8 +1,12 @@
+import collections
+import copy
 from dataclasses import dataclass
 from typing import (
     TYPE_CHECKING,
+    Any,
     Callable,
     Dict,
+    Generator,
     List,
     Optional,
     Protocol,
@@ -13,10 +17,12 @@
 
 import interegular
 import torch
-from lark import Lark
+from lark.indenter import DedentError
+from lark.lexer import UnexpectedCharacters, UnexpectedToken
 
 from outlines import grammars
 from outlines.caching import cache
+from outlines.fsm.parsing import PartialLark, PartialParserState
 from outlines.fsm.regex import (
     create_fsm_index_tokenizer,
     make_byte_level_fsm,
@@ -69,13 +75,15 @@ class Guide(Protocol):
 
     """
 
-    def get_next_instruction(self, state: int) -> Instruction:
+    initial_state: Any
+
+    def get_next_instruction(self, state: Any) -> Instruction:
         ...
 
-    def get_next_state(self, state: int, token_id: int) -> int:
+    def get_next_state(self, state: Any, token_id: int) -> Any:
         ...
 
-    def is_final_state(self, state: int) -> bool:
+    def is_final_state(self, state: Any) -> bool:
         ...
 
     def copy(self) -> "Guide":
@@ -86,7 +94,8 @@ class StopAtEOSGuide(Guide):
     """Guide to generate tokens until the EOS token has been generated."""
 
     final_state = 1
-    start_state = 0
+    start_state = 0  # TODO: remove start_state, use only initial_state
+    initial_state = 0
 
     def __init__(self, tokenizer: "Tokenizer"):
         """Initialize the generation guide.
@@ -107,7 +116,7 @@ def get_next_state(self, state: int, token_id: int) -> int:
         if token_id == self.eos_token_id or state == self.final_state:
             return self.final_state
 
-        return self.start_state
+        return self.initial_state
 
     def is_final_state(self, state: int):
         return state == self.final_state
@@ -300,178 +309,175 @@ def copy(self):
         return self
 
 
+CFGState = collections.namedtuple("CFGState", ["parser_state", "prev_token"])
+
+
 class CFGGuide(Guide):
-    """Guide to generate text that is in the language of a context-free grammar."""
+    """Guide to generate text that is in the language of a context-free Lark grammar."""
 
     def __init__(self, cfg_string: str, tokenizer):
+        """
+        Construct the PartialLark parser and set the empty initial_state (PartialParserState)
+        """
         self.cfg_string = cfg_string
         self.tokenizer = tokenizer
-
-        self.parser = Lark(
+        self.eos_token_id = self.tokenizer.eos_token_id
+        self.parser = PartialLark(
             cfg_string,
             parser="lalr",
-            lexer="contextual",
-            propagate_positions=False,
-            maybe_placeholders=False,
-            regex=True,
             import_paths=[grammars.GRAMMAR_PATH],
         )
-        self.terminal_regexps = dict()
-        for terminal in self.parser.terminals:
-            if terminal.pattern is not None:
-                self.terminal_regexps[terminal.name] = terminal.pattern.to_regexp()
-        self.terminal_regexps["$END"] = tokenizer.eos_token
-
-        self.generation = ""
-        self.reset_state = False
-        self.allow_eos = False
-        self.regex_fsm: RegexGuide
-
-        self.check_last = False
-        self.proposal_last: List[int] = []
-        self.regex_fsm_last: RegexGuide
-
-        self.start_state = 0
-        self.final_state = -1
-
-    def get_next_instruction(self, state: int) -> Instruction:
-        """Generate an instruction for the next step.
-
-        Upon initialization, the CFG incremental parser is used to determine the
-        first regex and construct the first FSM to generate the first terminal.
+        self.initial_state = CFGState(
+            parser_state=self.parser.parse(""), prev_token=None
+        )
 
-        This FSM is used for proposals until either:
+    def get_next_instruction(self, state: CFGState) -> Instruction:
+        """Return the next instruction for guided generation.
 
-        - The FSM is exhausted, and its only remaining option is the EOS token,
-          in which case we feed the generated terminal to the
-          CFG incremental parser and allow it to propose the next regex
-          corresponding to the next set of valid terminals.
-        - The current FSM can be exhausted, but the EOS token is not the only
-          remaining option. In this case we allow proposal of current terminal
-          extensions, store the current FSM and its state, then also use the CFG
-          parser to propose a new regex corresponding to terminating the current
-          terminal and starting the next one. The model can then sample from
-          either of these sets to determine whether to extend the current
-          terminal or terminate it and start the next one.
+        Current lazy approach:
+        - For each token in the vocabulary
+          - create a copy of the parsers state
+          - add the tokens to the parsers input text
+          - if valid, add token to returned tokens
 
-        The CFG incremental parser is allowed to propose the EOS token from any accepting state,
-        and once it is generated, the FSM will continue to always generate the EOS token.
+        Further refinements are necessary for performant text processing.
 
         Parameters
         ----------
         state
-            The current state of the FSM.
+            The guides current PartialParserState, or None if complete
 
         Returns
         -------
-        A list that contains the tokens to mask.
+        A `Generate` instance that contains the model and the allowed token ids.
 
         """
-        if self.is_final_state(state):
-            return Write([self.tokenizer.eos_token_id])
 
-        proposal: List[int] = []
-        if self.generation != "":
-            if self.check_last:
-                proposer = self.regex_fsm_last
-            else:
-                proposer = self.regex_fsm
+        if state.parser_state is None:
+            return Write(torch.tensor([self.eos_token_id]))
 
-            instruction = proposer.get_next_instruction(state)
+        valid_tokens = list(
+            self.iter_valid_token_ids(state, self.tokenizer.vocabulary.values())
+        )
+        if len(valid_tokens) == 1:
+            return Write(torch.tensor(valid_tokens))
+        return Generate(torch.tensor(valid_tokens))
 
-            assert instruction.tokens is not None
+    def iter_valid_token_ids(
+        self, state: CFGState, candidate_token_ids: list
+    ) -> Generator[int, None, None]:
+        """
+        Iterate over the given token_ids and yield those that are valid for the current parser state.
 
-            if isinstance(instruction, Write):
-                proposal += instruction.tokens
+        Parameters
+        ----------
+        parser_state
+            The current state of the parser, or None if complete.
+        token_ids
+            The list of token ids to check for validity.
+
+        Yields
+        ------
+        int
+            Valid token ids.
+        """
+        if state.parser_state is None:
+            yield self.eos_token_id
+            return
+
+        for token_id in candidate_token_ids:
+            if token_id == self.eos_token_id:
+                if self.can_terminate_state(state):
+                    yield token_id
             else:
-                proposal += instruction.tokens
-
-            if self.tokenizer.eos_token_id not in proposal:
-                return Generate(proposal)
-
-            self.check_last = False
-            proposal = [x for x in proposal if x != self.tokenizer.eos_token_id]
-            if len(proposal) > 0:
-                self.check_last = True
-                self.proposal_last = proposal.copy()
-                self.regex_fsm_last = proposer
-
-        interactive = self.parser.parse_interactive(self.generation)
-        interactive.exhaust_lexer()
-
-        options = {self.terminal_regexps[x] for x in interactive.accepts()}
-        # add %ignore terminals
-        options |= {self.terminal_regexps[x] for x in self.parser.lexer_conf.ignore}
-
-        if self.terminal_regexps["$END"] in options:
-            options.remove(self.terminal_regexps["$END"])
-            if len(options) == 0:
-                return Write([self.tokenizer.eos_token_id])
-            self.allow_eos = True
-            options.add("")
-            assert len(options) > 1
-
-        regex_string = r"(" + r"|".join([r"(" + x + r")" for x in options]) + r")"
-        self.regex_fsm = RegexGuide(regex_string, self.tokenizer)
-        self.reset_state = True
-
-        instruction = self.regex_fsm.get_next_instruction(self.start_state)
-
-        assert instruction.tokens is not None
-
-        if isinstance(instruction, Write):
-            proposal += instruction.tokens
-        else:
-            proposal += instruction.tokens
-
-        if self.allow_eos:
-            self.allow_eos = False
-        else:
-            proposal = [x for x in proposal if x != self.tokenizer.eos_token_id]
-            assert len(proposal) > 0
-
-        return Generate(proposal)
-
-    def get_next_state(self, state: int, token_id: int) -> int:
-        """Update the state of the guide.
-
-        Transitions the underlying regex FSM to its next state.
-        If at max tokens or EOS token, transition permanently to the final state.
-        Update stored partial generations for subsequent incremental parsing.
+                try:
+                    self._get_parser_state_token_applied(state, int(token_id))
+                    yield token_id
+                except (
+                    ValueError,
+                    EOFError,
+                    UnexpectedToken,
+                    UnexpectedCharacters,
+                    DedentError,
+                ):
+                    pass
+
+    def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
+        """
+        Update the state of the guide.
+        Decode the token_id, and calculate the new parser_state with the token applied.
 
         Parameters
         ----------
         state
-            The current state of the FSM.
+            The guides current PartialParserState, or None if complete
         token_id
             The id of the token that was just generated.
 
         Returns
         -------
-        The new state of the FSM.
-        """
+        The guides new PartialParserState
 
-        # We need to return the final state when in the final state because we
-        # then generate EOS tokens instead of stopping the generation.
-        if token_id == self.tokenizer.eos_token_id or state == self.final_state:
-            return self.final_state
-
-        self.generation += self.tokenizer.decode([token_id])[0]
+        """
+        if state.parser_state is None or token_id == self.eos_token_id:
+            parser_state = None
+        else:
+            parser_state = self._get_parser_state_token_applied(state, int(token_id))
+        return CFGState(parser_state=parser_state, prev_token=token_id)
 
-        if self.check_last:
-            if token_id in self.proposal_last:
-                return self.regex_fsm_last.get_next_state(state, token_id)
-            self.check_last = False
+    def _get_parser_state_token_applied(
+        self, state: CFGState, token_id: int
+    ) -> PartialParserState:
+        """
+        Don't mutate `parser_state`, copy to protect
 
-        if self.reset_state:
-            self.reset_state = False
-            state = self.start_state
+        Get the token string
+          - if first token in generation: tokenizer.decode (no leading whitespace)
+          - else: normalized (with possibly leading whitespace)
 
-        return self.regex_fsm.get_next_state(state, token_id)
+        Don't allow empty ("") tokens, raise ValueError
+        """
+        parser_state = copy.copy(state.parser_state)  # prevent side effects
 
-    def is_final_state(self, state: int) -> bool:
-        return state == self.final_state
+        # normalize
+        if state.prev_token is None:
+            new_token_str = self.tokenizer.decode([token_id])[0]
+        else:
+            prev_token_str = self.tokenizer.decode([[state.prev_token]])[0]
+            combined_token_str = self.tokenizer.decode([[state.prev_token, token_id]])[
+                0
+            ]
+            new_token_str = combined_token_str[len(prev_token_str) :]
+
+        if new_token_str == "":
+            raise ValueError("empty next token")
+
+        # update parser with new token
+        parser_state.lexer.state.text += new_token_str
+        self.parser.parse_from_state(parser_state, is_end=False)
+
+        return parser_state
+
+    def is_final_state(self, state: CFGState) -> bool:
+        # TODO: remove this method, use can_terminate_state and must_terminate_state
+        # here and in RegexGuide per https://github.com/outlines-dev/outlines/issues/885
+        return self.can_terminate_state(state)
+
+    def can_terminate_state(self, state: CFGState) -> bool:
+        """Generation is allowed to terminate"""
+        if state.parser_state is not None:
+            try:
+                copy.copy(state.parser_state).feed_eof()
+            except UnexpectedToken:
+                return False
+        return True
+
+    def must_terminate_state(self, state: CFGState) -> bool:
+        """Generation must terminate, no legal continuations"""
+        return state.parser_state is None or set(state.parser_state.accepts()).issubset(
+            {"$END"}
+        )
 
     def copy(self) -> "CFGGuide":
-        """Create a copy of the FSM."""
+        """Create a copy of the Guide."""
         return CFGGuide(self.cfg_string, self.tokenizer)
diff --git a/outlines/fsm/parsing.py b/outlines/fsm/parsing.py
index e4fa7b764..f780fb46e 100644
--- a/outlines/fsm/parsing.py
+++ b/outlines/fsm/parsing.py
@@ -447,6 +447,49 @@ def feed_token_no_stack(self, token, is_end=False):
                 if is_end and state_stack[-1] == end_state:
                     return
 
+    def feed_eof(self):
+        last_token = self.lexer.state.last_token
+
+        if last_token is None:
+            eof_token = self.lexer._Token("$END", "", 0, 1, 1)
+        else:
+            eof_token = Token.new_borrow_pos("$END", "", last_token)
+
+        new_token_is_legal = (
+            last_token is None
+            or last_token.type != "partial"
+            or any(ti.is_final for ti in last_token.value.terminals_and_info)
+        )
+        if new_token_is_legal:
+            self.feed_token(eof_token, is_end=True)
+        else:
+            raise UnexpectedToken(eof_token, [], state=self, interactive_parser=None)
+
+    def choices(self):
+        return self.parse_conf.parse_table.states[self.position]
+
+    def accepts(self):
+        """
+        Adapted from https://github.com/lark-parser/lark/blob/be542c2ff6d968817df019b8bf03f37b3111c08c/lark/parsers/lalr_interactive_parser.py#L95
+        Returns the set of possible tokens that will advance the parser into a new valid state.
+        """
+        accepts = set()
+        conf_no_callbacks = copy(self.parse_conf)
+        # We don't want to call callbacks here since those might have arbitrary side effects
+        # and are unnecessarily slow.
+        conf_no_callbacks.callbacks = {}
+        for t in self.choices():
+            if t.isupper():  # is terminal?
+                new_state = copy(self)
+                new_state.parse_conf = conf_no_callbacks
+                try:
+                    new_state.feed_token(new_state.lexer._Token(t, ""))
+                except UnexpectedToken:
+                    pass
+                else:
+                    accepts.add(t)
+        return accepts
+
     def __copy__(self):
         return type(self)(
             self.parse_conf,
@@ -483,12 +526,7 @@ def parse_from_state(self, state, last_token=None, is_end=False):
                 state.feed_token(token)
 
             if is_end and (not token or token.type != "partial"):
-                end_token = (
-                    Token.new_borrow_pos("$END", "", token)
-                    if token
-                    else Token("$END", "", 0, 1, 1)
-                )
-                state.feed_token(end_token, True)
+                state.feed_eof()
 
             return state
         except UnexpectedInput as e:
@@ -614,6 +652,8 @@ def __init__(self, conf: "LexerConf", states, always_accept=()):
                 lexer_conf.terminals = [
                     terminals_by_name[n] for n in accepts if n in terminals_by_name
                 ]
+                if not lexer_conf.terminals:
+                    continue
                 lexer = PartialBasicLexer(lexer_conf)
                 lexer_by_symbols[key] = lexer
 
@@ -626,9 +666,22 @@ def lex(self, lexer_state: LexerState, parser_state: Any) -> Iterator[Token]:
         try:
             while True:
                 lexer = self.lexers[parser_state.position]
-                yield lexer.next_token(lexer_state, parser_state)
+                next_tok = lexer.next_token(lexer_state, parser_state)
+                yield next_tok
         except EOFError:
             pass
+        except KeyError:
+            if len(lexer_state.text) > lexer_state.line_ctr.char_pos:
+                raise UnexpectedCharacters(
+                    lexer_state.text,
+                    lexer_state.line_ctr.char_pos,
+                    lexer_state.line_ctr.line,
+                    lexer_state.line_ctr.column,
+                    allowed=False,
+                    token_history=lexer_state.last_token and [lexer_state.last_token],
+                    state=parser_state,
+                    terminals_by_name=self.root_lexer.terminals,
+                )
 
 
 class PartialBasicLexer(BasicLexer):
diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
index 0df833067..4f372f209 100644
--- a/outlines/generate/cfg.py
+++ b/outlines/generate/cfg.py
@@ -1,7 +1,10 @@
 from functools import singledispatch
 
-from outlines.generate.api import SequenceGeneratorAdapter
-from outlines.models import OpenAI
+from outlines.generate.api import (
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import ExLlamaV2Model, LlamaCpp, OpenAI, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
@@ -14,8 +17,7 @@ def cfg(
     Arguments
     ---------
     model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
+        An `outlines.model` instance.
     sampler:
         The sampling algorithm to use to generate token ids from the logits
         distribution.
@@ -25,13 +27,32 @@ def cfg(
     A `SequenceGeneratorAdapter` instance that generates text.
 
     """
+    from outlines.processors import CFGLogitsProcessor
+
+    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
+    return SequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@cfg.register(TransformersVision)
+def cfg_vision(model, cfg_str: str, sampler: Sampler = multinomial()):
+    from outlines.processors import CFGLogitsProcessor
+
+    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
+    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@cfg.register(ExLlamaV2Model)
+def cfg_exllamav2(model, cfg_str: str, sampler: Sampler = multinomial()):
     raise NotImplementedError(
-        f"The CFG Logits processor is not available for {type(model)}. "
-        + "Please subscribe to https://github.com/outlines-dev/outlines/issues/684"
-        + " for updates on the fix."
+        "Not yet available, track progress in https://github.com/outlines-dev/outlines/pull/1010"
     )
 
 
+@cfg.register(LlamaCpp)
+def cfg_llamacpp(model, cfg_str: str, sampler: Sampler = multinomial()):
+    raise NotImplementedError("Not yet available due to bug in llama_cpp tokenizer")
+
+
 @cfg.register(OpenAI)
 def cfg_openai(model, cfg_str: str, sampler: Sampler = multinomial()):
     raise NotImplementedError(
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
index 03fe512b9..a9338836a 100644
--- a/outlines/generate/fsm.py
+++ b/outlines/generate/fsm.py
@@ -16,19 +16,19 @@
 def fsm(
     model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
 ) -> SequenceGeneratorAdapter:
-    from outlines.processors import FSMLogitsProcessor
+    from outlines.processors import GuideLogitsProcessor
 
-    fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    logits_processor = FSMLogitsProcessor(tokenizer=model.tokenizer, fsm=fsm)
+    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
+    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
 @fsm.register(TransformersVision)
 def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()):
-    from outlines.processors import FSMLogitsProcessor
+    from outlines.processors import GuideLogitsProcessor
 
-    fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    logits_processor = FSMLogitsProcessor(tokenizer=model.tokenizer, fsm=fsm)
+    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
+    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
     return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
diff --git a/outlines/grammars/common.lark b/outlines/grammars/common.lark
index 801c27e97..ee5e00c50 100644
--- a/outlines/grammars/common.lark
+++ b/outlines/grammars/common.lark
@@ -43,11 +43,14 @@ SIGNED_FLOAT: ["+"|"-"] FLOAT
 NUMBER: FLOAT | INT
 SIGNED_NUMBER: ["+"|"-"] NUMBER
 
-//
-// TODO: Working escaped_string
-//
 UNESCAPED_STRING: /\"[^"]*\"/
 
+// based on `outlines/fsm/json_schema.py`
+_NON_CONTROL_CHAR: /([^"\\\x00-\x1F\x7F-\x9F])/
+_ESCAPED_CHAR: /\\/ (_NON_CONTROL_CHAR | /\\/ | /"/)
+ESCAPED_STRING_INNER: _NON_CONTROL_CHAR | _ESCAPED_CHAR
+ESCAPED_STRING: /"/ ESCAPED_STRING_INNER* /"/
+
 
 
 //
diff --git a/outlines/grammars/json.lark b/outlines/grammars/json.lark
index 72af448ce..7429fa558 100644
--- a/outlines/grammars/json.lark
+++ b/outlines/grammars/json.lark
@@ -2,7 +2,7 @@
 
 ?value: object
 | array
-| UNESCAPED_STRING
+| ESCAPED_STRING
 | SIGNED_NUMBER      -> number
 | "true"             -> true
 | "false"            -> false
@@ -10,9 +10,9 @@
 
 array  : "[" [value ("," value)*] "]"
 object : "{" [pair ("," pair)*] "}"
-pair   : UNESCAPED_STRING ":" value
+pair   : ESCAPED_STRING ":" value
 
-%import common.UNESCAPED_STRING
+%import common.ESCAPED_STRING
 %import common.SIGNED_NUMBER
 %import common.WS
 
diff --git a/outlines/processors/__init__.py b/outlines/processors/__init__.py
index 22c10d905..f0f0f829b 100644
--- a/outlines/processors/__init__.py
+++ b/outlines/processors/__init__.py
@@ -1,6 +1,6 @@
 from .structured import (
     CFGLogitsProcessor,
-    FSMLogitsProcessor,
+    GuideLogitsProcessor,
     JSONLogitsProcessor,
     OutlinesLogitsProcessor,
     RegexLogitsProcessor,
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index f8d63d7b5..93d68727f 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -24,7 +24,7 @@
 limitations under the License.
 """
 import math
-from typing import TYPE_CHECKING, Dict, List, Optional, Type, Union
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
 
 import torch
 from pydantic import BaseModel
@@ -38,36 +38,41 @@
     from outlines.models.tokenizer import Tokenizer
 
 
-class FSMLogitsProcessor(OutlinesLogitsProcessor):
-    """Bias generation using a finite state machine.
+class GuideLogitsProcessor(OutlinesLogitsProcessor):
+    """Bias generation using a finite
 
     Attributes
     ----------
     tokenizer
         The tokenizer used to convert tokens to ids.
-    fsm
-        The finite state machine which is used to bias the logits.
+    guide
+        The `outlines.fsm.Guide` which is used to bias the logits.
     """
 
-    def __init__(self, tokenizer: "Tokenizer", fsm: Guide):
-        """A FSM-based logits processor.
+    tokenizer: "Tokenizer"
+    guide: Guide
+    _guide_states: Dict[int, Any]
+    _seq_start_idx: Optional[int]
+
+    def __init__(self, tokenizer: "Tokenizer", guide: Guide):
+        """A Guide-based logits processor.
 
         Parameters
         ----------
         tokenizer
             The tokenizer used to convert tokens to ids.
-        fsm
-            The finite state machine which is used to bias the logits.
+        guide
+            The `outlines.fsm.Guide. which is used to bias the logits.
         """
         self.tokenizer = tokenizer
-        self._fsm_states: Dict[int, int] = {hash(tuple([])): 0}
-        self.fsm: Guide = fsm
-        self._seq_start_idx: Optional[int] = None
+        self.guide = guide
+        self._guide_states = {hash(tuple([])): self.guide.initial_state}
+        self._seq_start_idx = None
 
     def process_logits(
         self, input_ids: List[List[int]], logits: torch.Tensor
     ) -> torch.Tensor:
-        """Use the FSM to bias the logits before sampling the next token.
+        """Use the Guide to bias the logits before sampling the next token.
 
         Parameters
         ----------
@@ -90,38 +95,38 @@ def process_logits(
             gen_ids = seq_ids[self._seq_start_idx :]
             curr_state_key = hash(tuple(gen_ids))
 
-            if curr_state_key not in self._fsm_states:
-                prev_state = self._fsm_states[hash(tuple(gen_ids[:-1]))]
-                curr_state = self.fsm.get_next_state(prev_state, gen_ids[-1])
-                self._fsm_states[curr_state_key] = curr_state
+            if curr_state_key not in self._guide_states:
+                prev_state = self._guide_states[hash(tuple(gen_ids[:-1]))]
+                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1])
+                self._guide_states[curr_state_key] = curr_state
 
-            sequence_states.append(self._fsm_states[curr_state_key])
+            sequence_states.append(self._guide_states[curr_state_key])
 
         mask = torch.full_like(logits, -math.inf)
-        for i, fsm_state in enumerate(sequence_states):
-            allowed_tokens = self.fsm.get_next_instruction(fsm_state).tokens
+        for i, guide_state in enumerate(sequence_states):
+            allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
             mask[i, allowed_tokens] = logits[i, allowed_tokens]
 
         return mask
 
-    def copy(self) -> "FSMLogitsProcessor":
+    def copy(self) -> "GuideLogitsProcessor":
         """Return a copy of the logits processor."""
-        return FSMLogitsProcessor(tokenizer=self.tokenizer, fsm=self.fsm.copy())
+        return GuideLogitsProcessor(tokenizer=self.tokenizer, guide=self.guide.copy())
 
 
-class RegexLogitsProcessor(FSMLogitsProcessor):
+class RegexLogitsProcessor(GuideLogitsProcessor):
     """Bias generation based on a regular expression.
 
     Attributes
     ----------
     tokenizer
         The tokenizer used to convert tokens to ids.
-    fsm
-        The finite state machine which is used to bias the logits.
+    guide
+        The `outlines.fsm.RegexGuide. which is used to bias the logits.
     """
 
     def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
-        """Compile the FSM that drives the regex-guided generation.
+        """Compile the RegexGuide that drives the regex-guided generation.
 
         Parameters
         ----------
@@ -130,8 +135,8 @@ def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
         tokenizer
             An Outlines tokenizer
         """
-        fsm = RegexGuide(regex_string, tokenizer)
-        super().__init__(tokenizer=tokenizer, fsm=fsm)
+        guide = RegexGuide(regex_string, tokenizer)
+        super().__init__(tokenizer=tokenizer, guide=guide)
 
 
 class JSONLogitsProcessor(RegexLogitsProcessor):
@@ -141,8 +146,8 @@ class JSONLogitsProcessor(RegexLogitsProcessor):
     ----------
     tokenizer
         The tokenizer used to convert tokens to ids.
-    fsm
-        The finite state machine which is used to bias the logits.
+    guide
+        The `outlines.fsm.RegexGuide. which is used to bias the logits.
     """
 
     def __init__(
@@ -151,7 +156,7 @@ def __init__(
         tokenizer: "Tokenizer",
         whitespace_pattern: Optional[str] = None,
     ):
-        """Compile the FSM that drives the JSON-guided generation.
+        """Compile the Guide that drives the JSON-guided generation.
 
         Parameters
         ----------
@@ -169,19 +174,21 @@ def __init__(
         super().__init__(regex_string=regex_string, tokenizer=tokenizer)
 
 
-class CFGLogitsProcessor(FSMLogitsProcessor):
+class CFGLogitsProcessor(GuideLogitsProcessor):
     """Bias generation based on a context-free grammar.
 
     Attributes
     ----------
     tokenizer
         The tokenizer used to convert tokens to ids.
-    fsm
-        The finite state machine which is used to bias the logits.
+    guide
+        The `outlines.fsm.CFGGuide. which is used to bias the logits.
     """
 
+    guide: CFGGuide
+
     def __init__(self, cfg_str: str, tokenizer: "Tokenizer"):
-        """Compile the FSM that drives the CFG-guided generation.
+        """Compile the CFGGuide that drives the CFG-guided generation.
 
         Parameters
         ----------
@@ -190,5 +197,36 @@ def __init__(self, cfg_str: str, tokenizer: "Tokenizer"):
         tokenizer
             The tokenizer used to convert tokens to ids.
         """
-        cfg_automata = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
-        super().__init__(tokenizer=tokenizer, fsm=cfg_automata)
+        cfg_guide = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
+        super().__init__(tokenizer=tokenizer, guide=cfg_guide)
+
+    def process_logits(
+        self, input_ids: List[List[int]], logits: torch.Tensor
+    ) -> torch.Tensor:
+        """Same behavior as GuideLogitsProcessor, but uses rejection sampling"""
+        if self._seq_start_idx is None:
+            self._seq_start_idx = len(input_ids[0])
+
+        sequence_states: List = []  # vector of states corresponding to `input_ids`
+
+        for seq_ids in input_ids:
+            gen_ids = seq_ids[self._seq_start_idx :]
+            curr_state_key = hash(tuple(gen_ids))
+
+            if curr_state_key not in self._guide_states:
+                prev_state = self._guide_states[hash(tuple(gen_ids[:-1]))]
+                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1])
+                self._guide_states[curr_state_key] = curr_state
+
+            sequence_states.append(self._guide_states[curr_state_key])
+
+        mask = torch.full_like(logits, -math.inf)
+        for i, guide_state in enumerate(sequence_states):
+            first_legal_token = next(
+                self.guide.iter_valid_token_ids(
+                    guide_state, torch.argsort(logits[i], descending=True)
+                )
+            )
+            mask[i, [first_legal_token]] = logits[i, [first_legal_token]]
+
+        return mask
diff --git a/tests/cfg_samples/arithmetic/lots_of_ops.arithmetic.test b/tests/cfg_samples/arithmetic/lots_of_ops.arithmetic.test
new file mode 100644
index 000000000..1489aebc7
--- /dev/null
+++ b/tests/cfg_samples/arithmetic/lots_of_ops.arithmetic.test
@@ -0,0 +1 @@
+5+1+1+1+1+1+1+1+1+1+1+1+1+1+1+1+1+1
diff --git a/tests/cfg_samples/arithmetic/simple_math.arithmetic.test b/tests/cfg_samples/arithmetic/simple_math.arithmetic.test
new file mode 100644
index 000000000..882f05c8d
--- /dev/null
+++ b/tests/cfg_samples/arithmetic/simple_math.arithmetic.test
@@ -0,0 +1 @@
+(1 * 2) - (0.1 * 2 * 9.42)
diff --git a/tests/cfg_samples/json/outlines.generate.samplers.mypy.json.test b/tests/cfg_samples/json/outlines.generate.samplers.mypy.json.test
new file mode 100644
index 000000000..1a328a9b6
--- /dev/null
+++ b/tests/cfg_samples/json/outlines.generate.samplers.mypy.json.test
@@ -0,0 +1,372 @@
+{
+    ".class": "MypyFile",
+    "_fullname": "outlines.generate.samplers",
+    "future_import_flags": [],
+    "is_partial_stub_package": false,
+    "is_stub": false,
+    "names": {
+        ".class": "SymbolTable",
+        "Protocol": {
+            ".class": "SymbolTableNode",
+            "cross_ref": "typing.Protocol",
+            "kind": "Gdef"
+        },
+        "Sampler": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "TypeInfo",
+                "_promote": [],
+                "abstract_attributes": [
+                    [
+                        "__call__",
+                        2
+                    ]
+                ],
+                "alt_promote": null,
+                "bases": [
+                    "builtins.object"
+                ],
+                "dataclass_transform_spec": null,
+                "declared_metaclass": null,
+                "defn": {
+                    ".class": "ClassDef",
+                    "fullname": "outlines.generate.samplers.Sampler",
+                    "name": "Sampler",
+                    "type_vars": []
+                },
+                "deletable_attributes": [],
+                "flags": [
+                    "is_abstract",
+                    "is_protocol"
+                ],
+                "fullname": "outlines.generate.samplers.Sampler",
+                "has_param_spec_type": false,
+                "metaclass_type": "abc.ABCMeta",
+                "metadata": {},
+                "module_name": "outlines.generate.samplers",
+                "mro": [
+                    "outlines.generate.samplers.Sampler",
+                    "builtins.object"
+                ],
+                "names": {
+                    ".class": "SymbolTable",
+                    "__call__": {
+                        ".class": "SymbolTableNode",
+                        "kind": "Mdef",
+                        "node": {
+                            ".class": "FuncDef",
+                            "abstract_status": 2,
+                            "arg_kinds": [
+                                0,
+                                0,
+                                0,
+                                0
+                            ],
+                            "arg_names": [
+                                "self",
+                                "logits",
+                                "samples",
+                                "rng"
+                            ],
+                            "dataclass_transform_spec": null,
+                            "flags": [
+                                "is_trivial_body"
+                            ],
+                            "fullname": "outlines.generate.samplers.Sampler.__call__",
+                            "name": "__call__",
+                            "type": {
+                                ".class": "CallableType",
+                                "arg_kinds": [
+                                    0,
+                                    0,
+                                    0,
+                                    0
+                                ],
+                                "arg_names": [
+                                    "self",
+                                    "logits",
+                                    "samples",
+                                    "rng"
+                                ],
+                                "arg_types": [
+                                    "outlines.generate.samplers.Sampler",
+                                    {
+                                        ".class": "AnyType",
+                                        "missing_import_name": "outlines.generate.samplers.torch",
+                                        "source_any": null,
+                                        "type_of_any": 3
+                                    },
+                                    "builtins.int",
+                                    {
+                                        ".class": "AnyType",
+                                        "missing_import_name": "outlines.generate.samplers.torch",
+                                        "source_any": null,
+                                        "type_of_any": 3
+                                    }
+                                ],
+                                "bound_args": [],
+                                "def_extras": {
+                                    "first_arg": "self"
+                                },
+                                "fallback": "builtins.function",
+                                "from_concatenate": false,
+                                "implicit": false,
+                                "is_ellipsis_args": false,
+                                "name": "__call__ of Sampler",
+                                "ret_type": {
+                                    ".class": "AnyType",
+                                    "missing_import_name": "outlines.generate.samplers.torch",
+                                    "source_any": null,
+                                    "type_of_any": 3
+                                },
+                                "type_guard": null,
+                                "unpack_kwargs": false,
+                                "variables": []
+                            }
+                        }
+                    }
+                },
+                "self_type": null,
+                "slots": null,
+                "tuple_type": null,
+                "type_vars": [],
+                "typeddict_type": null
+            }
+        },
+        "__annotations__": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "Var",
+                "flags": [
+                    "is_ready"
+                ],
+                "fullname": "outlines.generate.samplers.__annotations__",
+                "name": "__annotations__",
+                "type": {
+                    ".class": "Instance",
+                    "args": [
+                        "builtins.str",
+                        {
+                            ".class": "AnyType",
+                            "missing_import_name": null,
+                            "source_any": null,
+                            "type_of_any": 6
+                        }
+                    ],
+                    "type_ref": "builtins.dict"
+                }
+            }
+        },
+        "__doc__": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "Var",
+                "flags": [
+                    "is_ready"
+                ],
+                "fullname": "outlines.generate.samplers.__doc__",
+                "name": "__doc__",
+                "type": "builtins.str"
+            }
+        },
+        "__file__": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "Var",
+                "flags": [
+                    "is_ready"
+                ],
+                "fullname": "outlines.generate.samplers.__file__",
+                "name": "__file__",
+                "type": "builtins.str"
+            }
+        },
+        "__name__": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "Var",
+                "flags": [
+                    "is_ready"
+                ],
+                "fullname": "outlines.generate.samplers.__name__",
+                "name": "__name__",
+                "type": "builtins.str"
+            }
+        },
+        "__package__": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "Var",
+                "flags": [
+                    "is_ready"
+                ],
+                "fullname": "outlines.generate.samplers.__package__",
+                "name": "__package__",
+                "type": "builtins.str"
+            }
+        },
+        "greedy": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "FuncDef",
+                "abstract_status": 0,
+                "arg_kinds": [
+                    0,
+                    0,
+                    2
+                ],
+                "arg_names": [
+                    "logits",
+                    "samples",
+                    "_"
+                ],
+                "dataclass_transform_spec": null,
+                "flags": [],
+                "fullname": "outlines.generate.samplers.greedy",
+                "name": "greedy",
+                "type": {
+                    ".class": "CallableType",
+                    "arg_kinds": [
+                        0,
+                        0,
+                        2
+                    ],
+                    "arg_names": [
+                        "logits",
+                        "samples",
+                        "_"
+                    ],
+                    "arg_types": [
+                        {
+                            ".class": "AnyType",
+                            "missing_import_name": "outlines.generate.samplers.torch",
+                            "source_any": null,
+                            "type_of_any": 3
+                        },
+                        "builtins.int",
+                        {
+                            ".class": "AnyType",
+                            "missing_import_name": null,
+                            "source_any": null,
+                            "type_of_any": 1
+                        }
+                    ],
+                    "bound_args": [],
+                    "def_extras": {
+                        "first_arg": null
+                    },
+                    "fallback": "builtins.function",
+                    "from_concatenate": false,
+                    "implicit": false,
+                    "is_ellipsis_args": false,
+                    "name": "greedy",
+                    "ret_type": {
+                        ".class": "AnyType",
+                        "missing_import_name": "outlines.generate.samplers.torch",
+                        "source_any": null,
+                        "type_of_any": 3
+                    },
+                    "type_guard": null,
+                    "unpack_kwargs": false,
+                    "variables": []
+                }
+            }
+        },
+        "multinomial": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "FuncDef",
+                "abstract_status": 0,
+                "arg_kinds": [
+                    0,
+                    0,
+                    0
+                ],
+                "arg_names": [
+                    "logits",
+                    "samples",
+                    "rng"
+                ],
+                "dataclass_transform_spec": null,
+                "flags": [],
+                "fullname": "outlines.generate.samplers.multinomial",
+                "name": "multinomial",
+                "type": {
+                    ".class": "CallableType",
+                    "arg_kinds": [
+                        0,
+                        0,
+                        0
+                    ],
+                    "arg_names": [
+                        "logits",
+                        "samples",
+                        "rng"
+                    ],
+                    "arg_types": [
+                        {
+                            ".class": "AnyType",
+                            "missing_import_name": "outlines.generate.samplers.torch",
+                            "source_any": null,
+                            "type_of_any": 3
+                        },
+                        "builtins.int",
+                        {
+                            ".class": "AnyType",
+                            "missing_import_name": "outlines.generate.samplers.torch",
+                            "source_any": null,
+                            "type_of_any": 3
+                        }
+                    ],
+                    "bound_args": [],
+                    "def_extras": {
+                        "first_arg": null
+                    },
+                    "fallback": "builtins.function",
+                    "from_concatenate": false,
+                    "implicit": false,
+                    "is_ellipsis_args": false,
+                    "name": "multinomial",
+                    "ret_type": {
+                        ".class": "AnyType",
+                        "missing_import_name": "outlines.generate.samplers.torch",
+                        "source_any": null,
+                        "type_of_any": 3
+                    },
+                    "type_guard": null,
+                    "unpack_kwargs": false,
+                    "variables": []
+                }
+            }
+        },
+        "torch": {
+            ".class": "SymbolTableNode",
+            "kind": "Gdef",
+            "node": {
+                ".class": "Var",
+                "flags": [
+                    "is_suppressed_import",
+                    "is_ready",
+                    "is_inferred"
+                ],
+                "fullname": "outlines.generate.samplers.torch",
+                "name": "torch",
+                "type": {
+                    ".class": "AnyType",
+                    "missing_import_name": "outlines.generate.samplers.torch",
+                    "source_any": null,
+                    "type_of_any": 3
+                }
+            }
+        }
+    },
+    "path": "/home/andrew/p/outlines/outlines/generate/samplers.py"
+}
diff --git a/tests/cfg_samples/json/simple_fruit.json.test b/tests/cfg_samples/json/simple_fruit.json.test
new file mode 100644
index 000000000..bffa952c7
--- /dev/null
+++ b/tests/cfg_samples/json/simple_fruit.json.test
@@ -0,0 +1,20 @@
+[
+    {
+        "ID": "1",
+        "Name": "Andrew \"The Escaper\" Lapp",
+        "Age": "30",
+        "FavFruit": "Banana"
+    },
+    {
+        "ID": "2",
+        "Name": "Mohammad",
+        "Age": "40",
+        "FavFruit": "\"Any Fruit As Long as It's In Quotes!\""
+    },
+    {
+        "ID": "3",
+        "Name": "Alice",
+        "Age": "61",
+        "FavFruit": "Peaches, but only \n newline separated peaches"
+    }
+]
diff --git a/tests/cfg_samples/json/simple_fruit_no_indent.json.test b/tests/cfg_samples/json/simple_fruit_no_indent.json.test
new file mode 100644
index 000000000..9b7d319da
--- /dev/null
+++ b/tests/cfg_samples/json/simple_fruit_no_indent.json.test
@@ -0,0 +1 @@
+[{"ID": "1", "Name": "Andrew", "Age": "30", "FavFruit": "Banana"}, {"ID": "2", "Name": "Mohammad", "Age": "40", "FavFruit": "Apple"}, {"ID": "3", "Name": "Alice", "Age": "61", "FavFruit": "Peach"}]
diff --git a/tests/fsm/test_cfg_guide.py b/tests/fsm/test_cfg_guide.py
new file mode 100644
index 000000000..d92afa625
--- /dev/null
+++ b/tests/fsm/test_cfg_guide.py
@@ -0,0 +1,457 @@
+from collections import namedtuple
+from pathlib import Path
+
+import pytest
+from transformers import AutoTokenizer
+
+from outlines import grammars, models
+from outlines.fsm.guide import CFGGuide
+
+
+@pytest.fixture
+def cleanup_lark_import():
+    import importlib
+
+    import lark.lark
+
+    yield
+    # Clean up lark.lark.LarkOptions._defaults
+    importlib.reload(lark.lark)
+
+
+TestInputs = namedtuple(
+    "TestInputs",
+    [
+        "grammar",  # the lark grammar to validate against
+        "vocabulary",  # the token strings which can be concatenated for a generation
+        "generated",  # the tokens which have been generated so far
+        "legal_next_tokens",  # the subset of the vocabulary which can legally be next in `generated`
+    ],
+)
+
+
+cfg_test_inputs = {
+    "Next Token Doesn't Complete Terminal": TestInputs(
+        grammar=r'?start: "a" "bc"',
+        vocabulary=["a", "ab", "b", "c"],
+        generated=["a"],
+        legal_next_tokens=["b"],
+    ),
+    "Ambiguous Terminal Completion": TestInputs(
+        grammar=r'?start: "ab" | "abc"',
+        vocabulary=["a", "ab", "abc", "abcd", "b", "c"],
+        generated=["a"],
+        legal_next_tokens=["b"],
+    ),
+    "Token is Substring of Another Token": TestInputs(
+        grammar=r'?start: "abc" | "abcd"',
+        vocabulary=["a", "b", "bc", "bcd", "bcde"],
+        generated=["a"],
+        legal_next_tokens=["b", "bc", "bcd"],
+    ),
+    "Multiple Valid Continuations": TestInputs(
+        grammar=r'?start: ("a" "b") | ("a" "c")',
+        vocabulary=["a", "b", "bc", "c"],
+        generated=["a"],
+        legal_next_tokens=["b", "c"],
+    ),
+    "Prefix Matches Multiple Terminals": TestInputs(
+        grammar=r'?start: "abcd" | "abef"',
+        vocabulary=["a", "b", "be", "bcd", "bef", "bed"],
+        generated=["a"],
+        legal_next_tokens=["b", "be", "bcd", "bef"],
+    ),
+    "Token Matches Multiple Paths in Grammar": TestInputs(
+        grammar=r'?start: ("a" "b" "c") | ("a" "b" "d")',
+        vocabulary=["a", "b", "c", "d"],
+        generated=["a", "b"],
+        legal_next_tokens=["c", "d"],
+    ),
+    "Incomplete Terminal at End of Prefix": TestInputs(
+        grammar=r'?start: "abc"',
+        vocabulary=["a", "ab", "c", "abc", "abcd"],
+        generated=["ab"],
+        legal_next_tokens=["c"],
+    ),
+    "Complex Grammar Rules": TestInputs(
+        grammar=r'?start: "a" "b" ["c"]',
+        vocabulary=["a", "b", "c"],
+        generated=["a", "b"],
+        legal_next_tokens=["c", None],  # Allowing the document to end after "a" "b"
+    ),
+    "Empty Prefix String": TestInputs(
+        grammar=r'?start: "a" | "b"',
+        vocabulary=["a", "b", "c", "d"],
+        generated=[],
+        legal_next_tokens=["a", "b"],
+    ),
+    "Ambiguous Pattern Completion": TestInputs(
+        grammar=r'?start: /a+/ "b" /c?d/',
+        vocabulary=["a", "aa", "b", "cd", "d"],
+        generated=["a", "a", "b"],
+        legal_next_tokens=["cd", "d"],
+    ),
+    "Optional Patterns with Overlapping Tokens": TestInputs(
+        grammar=r'?start: "a" "b"? "c"',
+        vocabulary=["a", "b", "bc", "c"],
+        generated=["a"],
+        legal_next_tokens=["b", "bc", "c"],
+    ),
+    "Greedy vs. Non-Greedy Matching": TestInputs(
+        grammar=r'?start: /a+?/ "b" /c/',
+        vocabulary=["a", "aa", "aaa", "b", "c"],
+        generated=["a", "a", "b"],
+        legal_next_tokens=["c"],
+    ),
+    "Nested Optional Elements": TestInputs(
+        grammar=r'?start: "a" ["b" ["c"]]',
+        vocabulary=["a", "b", "bc", "c"],
+        generated=["a"],
+        legal_next_tokens=[
+            "b",
+            "bc",
+            None,
+        ],  # Allowing the document to end after "a" "b"
+    ),
+    "Recursive Patterns": TestInputs(
+        grammar=r'?start: /a(bc)*/ "d"',
+        vocabulary=["a", "bc", "bcbcbc", "d"],
+        generated=["a", "bc", "d"],
+        legal_next_tokens=[None],  # Allowing the document to end after "a" "bc" "d"
+    ),
+    "Overlapping Character Classes": TestInputs(
+        grammar=r'?start: /[ab]+/ "d"',
+        vocabulary=["a", "b", "c", "aa", "bb", "cc", "d"],
+        generated=["a", "b"],
+        legal_next_tokens=["d", "a", "b", "aa", "bb"],
+    ),
+    "Conditional Patterns": TestInputs(
+        grammar=r'?start: "a" /b/ "c" (/d/)?',
+        vocabulary=["a", "b", "c", "d"],
+        generated=["a", "b", "c"],
+        legal_next_tokens=["d", None],  # Allowing the document to end after "a" "b" "c"
+    ),
+    "Unicode and Special Characters": TestInputs(
+        grammar=r'?start: /[a-zA-Z]/ "é" /[0-9]+/',
+        vocabulary=["a", "b", "é", "1", "2", "12"],
+        generated=["a", "é"],
+        legal_next_tokens=["1", "2", "12"],
+    ),
+    "Unicode and Special Characters Are Choices": TestInputs(
+        grammar=r'?start: /[a-zA-Z]/ "é" /[0-9]+/',
+        vocabulary=["a", "b", "é", "é9", "2", "12"],
+        generated=["a"],
+        legal_next_tokens=["é", "é9"],
+    ),
+    "Whitespace and Ignored Characters": TestInputs(
+        grammar=r'?start: "a" / *\s*b/ "c"',
+        vocabulary=["a", " b", " c", "c"],
+        generated=["a", " b"],
+        legal_next_tokens=["c"],
+    ),
+    "Token Overlaps Multiple Terminals": TestInputs(
+        grammar=r'?start: "a" "b" "c" "ab"',
+        vocabulary=["a", "b", "bc", "cab", "abc"],
+        generated=["a"],
+        legal_next_tokens=["b", "bc"],
+    ),
+    "Interleaved Sequences": TestInputs(
+        grammar=r'?start: ("a" "b") | ("a" "c")',
+        vocabulary=["a", "b", "c", "ab", "ac"],
+        generated=["a"],
+        legal_next_tokens=["b", "c"],
+    ),
+    "Repeated and Nested Patterns": TestInputs(
+        grammar=r'?start: "a" ("b" "c")* "d"',
+        vocabulary=["a", "b", "c", "bc", "bcc", "cbc", "bcbc", "cbccccd", "d", "bcbcd"],
+        generated=["a", "b", "c"],
+        legal_next_tokens=["b", "bc", "bcbc", "d", "bcbcd"],
+    ),
+    "Ambiguous Ending Patterns": TestInputs(
+        grammar=r'?start: "a" (/b/)? (/c/)*',
+        vocabulary=["a", "b", "c"],
+        generated=["a", "b"],
+        legal_next_tokens=["c", None],  # Allowing the document to end after "a" "b"
+    ),
+    "Whitespace Handling in Patterns": TestInputs(
+        grammar=r'?start: "a" / *\s*b/ /c /',
+        vocabulary=["a", " b", "c"],
+        generated=["a", " b"],
+        legal_next_tokens=["c"],
+    ),
+    "Token with Escape Characters": TestInputs(
+        grammar=r'?start: "a\n" ("\t")? "b"',
+        vocabulary=["a\nb", "a", "b", "\n", "\tb", "\t"],
+        generated=["a", "\n"],
+        legal_next_tokens=["b", "\t", "\tb"],
+    ),
+    "Complex Nesting": TestInputs(
+        grammar=r'?start: "a" ("b" ("c" "d"))',
+        vocabulary=["a", "b", "c", "d"],
+        generated=["a", "b", "c"],
+        legal_next_tokens=["d"],
+    ),
+    "Repeated Optional Patterns": TestInputs(
+        grammar=r'?start: ("a" ["b"])*',
+        vocabulary=["a", "b"],
+        generated=["a", "b", "a"],
+        legal_next_tokens=["a", "b", None],
+    ),
+    "Multiple Non-Terminal Symbols": TestInputs(
+        grammar=r"""
+        ?start: A B
+        A: "a"
+        B: "b"
+        """,
+        vocabulary=["a", "b"],
+        generated=["a"],
+        legal_next_tokens=["b"],
+    ),
+    "Recursive Definitions": TestInputs(
+        grammar=r"""
+        ?start: term_a
+        term_a: "a" term_a | "b"
+        """,
+        vocabulary=["a", "b"],
+        generated=["a", "a"],
+        legal_next_tokens=["a", "b"],
+    ),
+    "Ignored Patterns": TestInputs(
+        grammar=r"""
+        ?start: "a" "b" "c"
+        %ignore /\s+/
+        """,
+        vocabulary=["a", "b", "c", " "],
+        generated=["a", " ", "b"],
+        legal_next_tokens=["c", " "],
+    ),
+    "Cross-References": TestInputs(
+        grammar=r"""
+        ?start: term_a
+        term_a : /a/ term_b
+        term_b : /b/ term_a | /c/
+        """,
+        vocabulary=["a", "b", "c", "bac"],
+        generated=["a"],
+        legal_next_tokens=["b", "bac", "c"],
+    ),
+    "Multiple Complex Non-Terminal Rules": TestInputs(
+        grammar=r"""
+            ?start: S1 S2 S3
+            S1: "a" | "b"
+            S2: "c" | "d"
+            S3: "e" "f" | "g"
+        """,
+        vocabulary=["a", "b", "c", "d", "e", "f", "g"],
+        generated=["a", "c"],
+        legal_next_tokens=["e", "g"],
+    ),
+    #
+    #
+    # TODO: fix
+    #       parser incorrectly requires consumption of a or b in this case for unknown reasons
+    # "Nested Patterns with Repetition": TestInputs(
+    #     grammar=r'?start: "a" ("b" | "c")* "d"',
+    #     vocabulary=["a", "b", "bc", "bcc", "cbc", "bcbc", "cbccccd", "c", "d" "bcdcb"],
+    #     generated=["a"],
+    #     legal_next_tokens=["b", "bc", "bcc", "cbc", "bcbc", "cbccccd", "c", "d"],
+    # ),
+    #
+    #
+    # TODO: fix
+    #       adjacent terminals with ambiguous starts and ends not handled properly
+    #       ensure parser isn't greedy incorrectly
+    # "Ambiguous Overlapping Patterns": TestInputs(
+    #     grammar=r"?start: /ab*/ /bc?/",
+    #     vocabulary=["a", "ab", "abc", "b", "bc", "c"],
+    #     generated=["a", "b", "b"],
+    #     legal_next_tokens=["b", "c", "bc"],
+    # ),
+    # "Ambiguous Overlapping Patterns In Generation": TestInputs(
+    #     grammar=r"?start: /ab*/ /bc?/",
+    #     vocabulary=["a", "ab", "abc", "b", "bc", "c", "abbbc"],
+    #     generated=["a", "b", "b"],
+    #     legal_next_tokens=["b", "c", "bc"],
+    # ),
+    #
+    #
+    # SKIP:
+    # Awaiting negative lookarounds in interegular
+    # "Lookahead and Lookbehind with Nested Conditions": TestInputs(
+    #    grammar=r'?start: /(?<=a)b(?=c)/ "d"',
+    #    vocabulary=["a", "b", "c", "d"],
+    #    generated=["a", "b", "c"],
+    #    legal_next_tokens=["d"]
+    # ),
+    # "Lookbehind Patterns": TestInputs(
+    #    grammar=r'?start: /(?<=a)b/ "c"',
+    #    vocabulary=["a", "b", "c"],
+    #    generated=["a", "b"],
+    #    legal_next_tokens=["c"]
+    # ),
+}
+
+
+@pytest.mark.parametrize("name", cfg_test_inputs.keys())
+def test_cfg_next_token(name, cleanup_lark_import):
+    inputs = cfg_test_inputs[name]
+
+    class MockTokenizer:
+        vocabulary = {token: i + 1 for i, token in enumerate(inputs.vocabulary)}
+        vocabulary["<eos>"] = 0
+        reverse_vocab = {i: tok for tok, i in vocabulary.items()}
+        special_tokens = {"<eos>"}
+        eos_token_id = 0
+
+        def convert_token_to_string(self, token):
+            return token
+
+        def decode(self, token_ids):
+            if isinstance(token_ids[0], list):
+                return [
+                    "".join(map(self.reverse_vocab.get, token_ids_sublist))
+                    for token_ids_sublist in token_ids
+                ]
+            return [self.reverse_vocab[token_id] for token_id in token_ids]
+
+    # create a guide and the appropriate state advanced
+    # per the inputs generated tokens
+    tokenizer = MockTokenizer()
+    guide = CFGGuide(inputs.grammar, tokenizer)
+    state = guide.initial_state
+    for token in inputs.generated:
+        state = guide.get_next_state(state, tokenizer.vocabulary[token])
+    instruction = guide.get_next_instruction(state)
+
+    # normalize expectations and returned tokens for simple comparison
+    returned_next_tokens = sorted(
+        {tokenizer.reverse_vocab[int(t)] for t in instruction.tokens}
+    )
+    expected_next_tokens = sorted(
+        {
+            t
+            if t is not None
+            else tokenizer.reverse_vocab[tokenizer.eos_token_id]  # None -> "<eos>"
+            for t in inputs.legal_next_tokens
+        }
+    )
+
+    assert returned_next_tokens == expected_next_tokens
+
+
+@pytest.fixture(scope="session")
+def tokenizer_sentencepiece_gpt2():
+    return models.TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
+
+
+@pytest.fixture(scope="session")
+def tokenizer_sentencepiece_llama1():
+    return models.TransformerTokenizer(
+        AutoTokenizer.from_pretrained(
+            "trl-internal-testing/tiny-random-LlamaForCausalLM"
+        )
+    )
+
+
+@pytest.fixture(scope="session")
+def tokenizer_tiktoken_llama3():
+    return models.TransformerTokenizer(
+        AutoTokenizer.from_pretrained("yujiepan/llama-3-tiny-random")
+    )
+
+
+@pytest.fixture(scope="session")
+def tokenizer_character_level_byt5():
+    return models.TransformerTokenizer(
+        AutoTokenizer.from_pretrained("google/byt5-small")
+    )
+
+
+# Collects all samples within cfg_samples/ and makes adding
+# a test case as easy as adding a valid sample to cfg_samples/
+all_samples = {}
+examples_path = Path(__file__).parent.parent / "cfg_samples"
+for sample_collection_path in examples_path.iterdir():
+    grammar_name = sample_collection_path.name
+    grammar = getattr(grammars, grammar_name)
+    for sample_path in sample_collection_path.iterdir():
+        test_name = f"{grammar_name}_{sample_path.name}"
+        with open(sample_path) as f:
+            all_samples[test_name] = (grammar_name, grammar, f.read().rstrip("\n"))
+
+
+@pytest.mark.parametrize("sample_name", all_samples.keys())
+def test_cfg_test_sample_valid_with_lark(sample_name):
+    """assert the provided sample is valid (testing the test itself)"""
+    from lark import Lark, UnexpectedToken
+
+    grammar_name, grammar_str, sample = all_samples[sample_name]
+    try:
+        parser = Lark(grammar_str, parser="lalr", import_paths=[grammars.GRAMMAR_PATH])
+        parser = parser.parse_interactive(sample)
+        token = parser.exhaust_lexer()[-1]
+        parser.feed_eof(token)
+    except UnexpectedToken as e:
+        raise Exception(
+            f"Invalid test, sample '{sample_name}' isn't a legal generation of '{grammar_name}':\n{e}"
+        )
+
+
+@pytest.mark.parametrize("sample_name", all_samples.keys())
+@pytest.mark.parametrize(
+    "tokenizer_name",
+    [
+        "tokenizer_sentencepiece_gpt2",
+        "tokenizer_sentencepiece_llama1",
+        "tokenizer_tiktoken_llama3",
+        "tokenizer_character_level_byt5",
+    ],
+)
+def test_cfg_grammar_sample(request, sample_name, tokenizer_name, cleanup_lark_import):
+    """Test whether CFG can generate the exact token sequence as tokenizer.encode(sample) produces"""
+
+    # TODO: enable these tests once improvements are made
+    if (
+        tokenizer_name != "tokenizer_character_level_byt5"
+        or sample_name == "json_outlines.generate.samplers.mypy.json.test"
+    ):
+        pytest.skip("CFG is too slow, skipping tests for this tokenizer")
+    elif sample_name == "arithmetic_lots_of_ops.arithmetic.test":
+        pytest.skip("CFG incorrectly handles this valid sample, skipping until bugfix")
+
+    tokenizer = request.getfixturevalue(tokenizer_name)
+
+    grammar_name, grammar_str, sample = all_samples[sample_name]
+    cfg_guide = CFGGuide(grammar_str, tokenizer)
+
+    sample_token_ids = tokenizer.tokenizer.encode(
+        sample, add_special_tokens=False, return_tensors="pt"
+    )[0]
+    assert (
+        len(sample_token_ids.shape) == 1
+    )  # ensure we're encoding in the desired shape for this test
+
+    state = cfg_guide.initial_state
+    for i, token_id in enumerate(sample_token_ids):
+        if tokenizer.decode([token_id])[0] == "":
+            continue
+        next_instruction = cfg_guide.get_next_instruction(state)
+        if token_id not in next_instruction.tokens:
+            processed_str = tokenizer.decode([sample_token_ids[:i]])[0]
+            remaining_str = tokenizer.decode([sample_token_ids[i:]])[0]
+            if next_instruction.tokens == [tokenizer.eos_token_id]:
+                error_label = "CFGGuide required EOS early"
+            else:
+                expected = tokenizer.decode(next_instruction.tokens)
+                error_label = (
+                    f"Mismatched expectations, Guide expected {sorted(expected)}"
+                )
+            raise Exception(
+                f"{error_label}\n"
+                f"processed:\n```{processed_str}```\n"
+                f"remaining:\n```{remaining_str}```"
+            )
+            next_instruction.tokens
+        state = cfg_guide.get_next_state(state, token_id)
+    final_instruction = cfg_guide.get_next_instruction(state)
+    assert tokenizer.eos_token_id in final_instruction.tokens
diff --git a/tests/fsm/test_fsm.py b/tests/fsm/test_fsm.py
index 21163b70d..94166fd95 100644
--- a/tests/fsm/test_fsm.py
+++ b/tests/fsm/test_fsm.py
@@ -1,6 +1,6 @@
 import pytest
 
-from outlines.fsm.fsm import CFGFSM, RegexFSM, StopAtEosFSM
+from outlines.fsm.fsm import RegexFSM, StopAtEosFSM
 
 
 def assert_expected_tensor_ids(tensor, ids):
@@ -90,263 +90,3 @@ def convert_token_to_string(self, token):
 
     state = fsm.next_state(state=5, token_id=103)
     assert fsm.is_final_state(state)
-
-
-def test_cfg():
-    class MockTokenizer:
-        vocabulary = {"{": 1, "}": 2, "[": 3, "]": 4, "eos": 5}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 5
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
-
-    cfg_str = """
-        start: expr
-        expr: "{" expr "}" | "[" expr "]" |
-    """
-    tokenizer = MockTokenizer()
-
-    with pytest.warns(UserWarning):
-        fsm = CFGFSM(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1, 3, 5])
-    state = fsm.next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "{"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
-    state = fsm.next_state(state=state, token_id=3)
-    assert fsm.generation == "{["
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 3, 4])
-    state = fsm.next_state(state=state, token_id=4)
-    assert fsm.generation == "{[]"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [2])
-    state = fsm.next_state(state=state, token_id=2)
-    assert fsm.generation == "{[]}"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [5])
-    state = fsm.next_state(state=state, token_id=5)
-    assert fsm.generation == "{[]}"
-    assert fsm.is_final_state(state)
-
-
-def test_cfg_early_termination():
-    class MockTokenizer:
-        vocabulary = {"(": 1, ")": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
-
-    cfg_str = """
-        start: expr+
-        expr: "(" subexpr ")"
-        subexpr: expr |
-    """
-    tokenizer = MockTokenizer()
-
-    with pytest.warns(UserWarning):
-        fsm = CFGFSM(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1])
-    state = fsm.next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "("
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2])
-    state = fsm.next_state(state=state, token_id=2)
-    assert fsm.generation == "()"
-    assert not fsm.is_final_state(state)
-
-    # possible to continue or terminate
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 3])
-    state = fsm.next_state(state=state, token_id=3)  # feed eos
-    assert fsm.generation == "()"
-    assert fsm.is_final_state(state)
-
-    # once eos generated, can only terminate
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [3])
-
-
-def test_cfg_ignore_directive():
-    class MockTokenizer:
-        vocabulary = {"a": 1, " ": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
-
-    cfg_str = """
-        start: LETTER+
-        LETTER: "a"
-        WS: " "
-        %ignore WS
-    """
-    tokenizer = MockTokenizer()
-
-    with pytest.warns(UserWarning):
-        fsm = CFGFSM(cfg_str, tokenizer)
-
-    state = 0
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=0), [1, 2])
-    state = fsm.next_state(state=0, token_id=2)
-    assert fsm.generation == " "
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=0), [1, 2])
-    state = fsm.next_state(state=0, token_id=1)
-    assert fsm.generation == " a"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
-    state = fsm.next_state(state=state, token_id=2)
-    assert fsm.generation == " a "
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
-    state = fsm.next_state(state=state, token_id=2)
-    assert fsm.generation == " a  "
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
-    state = fsm.next_state(state=state, token_id=1)
-    assert fsm.generation == " a  a"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 2, 3])
-    state = fsm.next_state(state=state, token_id=3)
-    assert fsm.generation == " a  a"
-    assert fsm.is_final_state(state)
-
-    # once eos generated, can only terminate
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [3])
-
-
-def test_cfg_multitoken_terminal():
-    class MockTokenizer:
-        vocabulary = {"a": 1, "b": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
-
-    cfg_str = """
-        start: S
-        S: "aa" | "bb"
-    """
-    tokenizer = MockTokenizer()
-
-    with pytest.warns(UserWarning):
-        fsm = CFGFSM(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1, 2])
-    assert fsm.reset_state  # starting new regex
-    state = fsm.next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "a"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1])
-    assert not fsm.reset_state  # continuing current regex
-    state = fsm.next_state(state=state, token_id=1)
-    assert fsm.generation == "aa"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [3])
-    assert not fsm.reset_state  # completing current regex
-    state = fsm.next_state(state=state, token_id=3)
-    assert fsm.generation == "aa"
-    assert fsm.is_final_state(state)
-
-
-def test_cfg_allow_both_extend_and_shift_terminal():
-    class MockTokenizer:
-        vocabulary = {"(": 1, ")": 2, "a": 3, "eos": 4}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 4
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
-
-    cfg_str = """
-        start: s
-        s: "(" s ")" | /a+/
-    """
-    tokenizer = MockTokenizer()
-
-    with pytest.warns(UserWarning):
-        fsm = CFGFSM(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=fsm.start_state), [1, 3])
-    state = fsm.next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "("
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [1, 3])
-    state = fsm.next_state(state=state, token_id=3)
-    assert fsm.generation == "(a"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [2, 3])
-    state = fsm.next_state(state=state, token_id=3)
-    assert fsm.generation == "(aa"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [2, 3])
-    state = fsm.next_state(state=state, token_id=2)
-    assert fsm.generation == "(aa)"
-    assert not fsm.is_final_state(state)
-
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=state), [4])
-    state = fsm.next_state(state=state, token_id=4)
-    assert fsm.generation == "(aa)"
-    assert fsm.is_final_state(state)
diff --git a/tests/fsm/test_guide.py b/tests/fsm/test_guide.py
index 20ba75893..67b4e0dd8 100644
--- a/tests/fsm/test_guide.py
+++ b/tests/fsm/test_guide.py
@@ -205,49 +205,47 @@ def inverse_vocabulary(self):
             return {v: k for k, v in self.vocabulary.items()}
 
         def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
+            if isinstance(token_ids[0], list):
+                return [
+                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
+                    for token_ids_sublist in token_ids
+                ]
+            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
 
     cfg_str = """
         start: expr
         expr: "{" expr "}" | "[" expr "]" |
     """
     tokenizer = MockTokenizer()
-    fsm = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(fsm.start_state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 3, 5])
-    state = fsm.get_next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "{"
-    assert not fsm.is_final_state(state)
+    guide = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
-    state = fsm.get_next_state(state=state, token_id=3)
-    assert fsm.generation == "{["
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(
+        guide.get_next_instruction(guide.initial_state).tokens, [1, 3, 5]
+    )
+    state = guide.get_next_state(guide.initial_state, token_id=1)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 3, 4])
-    state = fsm.get_next_state(state=state, token_id=4)
-    assert fsm.generation == "{[]"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
+    state = guide.get_next_state(state, token_id=3)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [2])
-    state = fsm.get_next_state(state=state, token_id=2)
-    assert fsm.generation == "{[]}"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 3, 4])
+    state = guide.get_next_state(state, token_id=4)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Write)
-    assert_expected_tensor_ids(instruction.tokens, [5])
-    state = fsm.get_next_state(state=state, token_id=5)
-    assert fsm.generation == "{[]}"
-    assert fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [2])
+    state = guide.get_next_state(state, token_id=2)
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
+
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [5])
+    state = guide.get_next_state(state, token_id=5)
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
 
 def test_cfg_early_termination():
@@ -265,7 +263,12 @@ def inverse_vocabulary(self):
             return {v: k for k, v in self.vocabulary.items()}
 
         def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
+            if isinstance(token_ids[0], list):
+                return [
+                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
+                    for token_ids_sublist in token_ids
+                ]
+            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
 
     cfg_str = """
         start: expr+
@@ -273,34 +276,29 @@ def decode(self, token_ids):
         subexpr: expr |
     """
     tokenizer = MockTokenizer()
-    fsm = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(fsm.start_state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1])
-    state = fsm.get_next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "("
-    assert not fsm.is_final_state(state)
+    guide = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2])
-    state = fsm.get_next_state(state=state, token_id=2)
-    assert fsm.generation == "()"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(
+        guide.get_next_instruction(guide.initial_state).tokens, [1]
+    )
+    state = guide.get_next_state(guide.initial_state, token_id=1)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
+
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2])
+    state = guide.get_next_state(state, token_id=2)
+    assert not guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
     # possible to continue or terminate
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 3])
-    state = fsm.get_next_state(state=state, token_id=3)  # feed eos
-    assert fsm.generation == "()"
-    assert fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 3])
+    state = guide.get_next_state(state, token_id=3)  # feed eos
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
     # once eos generated, can only terminate
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Write)
-    assert_expected_tensor_ids(instruction.tokens, [3])
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [3])
 
 
 def test_cfg_ignore_directive():
@@ -318,7 +316,12 @@ def inverse_vocabulary(self):
             return {v: k for k, v in self.vocabulary.items()}
 
         def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
+            if isinstance(token_ids[0], list):
+                return [
+                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
+                    for token_ids_sublist in token_ids
+                ]
+            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
 
     cfg_str = """
         start: LETTER+
@@ -327,56 +330,43 @@ def decode(self, token_ids):
         %ignore WS
     """
     tokenizer = MockTokenizer()
-    fsm = CFGGuide(cfg_str, tokenizer)
 
-    state = 0
+    guide = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(0)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2])
-    state = fsm.get_next_state(state=0, token_id=2)
-    assert fsm.generation == " "
-    assert not fsm.is_final_state(state)
+    state = guide.initial_state
 
-    instruction = fsm.get_next_instruction(0)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2])
-    state = fsm.get_next_state(state=0, token_id=1)
-    assert fsm.generation == " a"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2])
+    state = guide.get_next_state(state, token_id=2)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
-    state = fsm.get_next_state(state=state, token_id=2)
-    assert fsm.generation == " a "
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2])
+    state = guide.get_next_state(state, token_id=1)
+    assert not guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
-    state = fsm.get_next_state(state=state, token_id=2)
-    assert fsm.generation == " a  "
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
+    state = guide.get_next_state(state, token_id=2)
+    assert not guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
-    state = fsm.get_next_state(state=state, token_id=1)
-    assert fsm.generation == " a  a"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
+    state = guide.get_next_state(state, token_id=2)
+    assert not guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2, 3])
-    state = fsm.get_next_state(state=state, token_id=3)
-    assert fsm.generation == " a  a"
-    assert fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
+    state = guide.get_next_state(state, token_id=1)
+    assert not guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
+
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
+    state = guide.get_next_state(state, token_id=3)
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
     # once eos generated, can only terminate
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Write)
-    assert_expected_tensor_ids(instruction.tokens, [3])
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [3])
 
 
 def test_cfg_multitoken_terminal():
@@ -394,38 +384,37 @@ def inverse_vocabulary(self):
             return {v: k for k, v in self.vocabulary.items()}
 
         def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
+            if isinstance(token_ids[0], list):
+                return [
+                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
+                    for token_ids_sublist in token_ids
+                ]
+            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
 
     cfg_str = """
         start: S
         S: "aa" | "bb"
     """
     tokenizer = MockTokenizer()
-    fsm = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(fsm.start_state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 2])
-    assert fsm.reset_state  # starting new regex
-    state = fsm.get_next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "a"
-    assert not fsm.is_final_state(state)
+    guide = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1])
-    assert not fsm.reset_state  # continuing current regex
-    state = fsm.get_next_state(state=state, token_id=1)
-    assert fsm.generation == "aa"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(
+        guide.get_next_instruction(guide.initial_state).tokens, [1, 2]
+    )
+    state = guide.get_next_state(guide.initial_state, token_id=1)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Write)
-    assert_expected_tensor_ids(instruction.tokens, [3])
-    assert not fsm.reset_state  # completing current regex
-    state = fsm.get_next_state(state=state, token_id=3)
-    assert fsm.generation == "aa"
-    assert fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1])
+    state = guide.get_next_state(state, token_id=1)
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
+
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [3])
+    state = guide.get_next_state(state, token_id=3)
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
 
 
 def test_cfg_allow_both_extend_and_shift_terminal():
@@ -443,46 +432,45 @@ def inverse_vocabulary(self):
             return {v: k for k, v in self.vocabulary.items()}
 
         def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
+            if isinstance(token_ids[0], list):
+                return [
+                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
+                    for token_ids_sublist in token_ids
+                ]
+            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
 
     cfg_str = """
         start: s
         s: "(" s ")" | /a+/
     """
     tokenizer = MockTokenizer()
-    fsm = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(fsm.start_state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 3])
-    state = fsm.get_next_state(state=fsm.start_state, token_id=1)
-    assert fsm.generation == "("
-    assert not fsm.is_final_state(state)
+    guide = CFGGuide(cfg_str, tokenizer)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1, 3])
-    state = fsm.get_next_state(state=state, token_id=3)
-    assert fsm.generation == "(a"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(
+        guide.get_next_instruction(guide.initial_state).tokens, [1, 3]
+    )
+    state = guide.get_next_state(guide.initial_state, token_id=1)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [2, 3])
-    state = fsm.get_next_state(state=state, token_id=3)
-    assert fsm.generation == "(aa"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 3])
+    state = guide.get_next_state(state, token_id=3)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [2, 3])
-    state = fsm.get_next_state(state=state, token_id=2)
-    assert fsm.generation == "(aa)"
-    assert not fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [2, 3])
+    state = guide.get_next_state(state, token_id=3)
+    assert not guide.must_terminate_state(state)
+    assert not guide.can_terminate_state(state)
 
-    instruction = fsm.get_next_instruction(state)
-    assert isinstance(instruction, Write)
-    assert_expected_tensor_ids(instruction.tokens, [4])
-    state = fsm.get_next_state(state=state, token_id=4)
-    assert fsm.generation == "(aa)"
-    assert fsm.is_final_state(state)
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [2, 3])
+    state = guide.get_next_state(state, token_id=2)
+
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
+
+    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [4])
+    state = guide.get_next_state(state, token_id=4)
+    assert guide.must_terminate_state(state)
+    assert guide.can_terminate_state(state)
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 8d5daa37e..fc4166535 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -131,6 +131,17 @@ def sample_choices():
     return ["foo", "bar", "baz"]
 
 
+@pytest.fixture()
+def sample_lark_grammar():
+    # from https://github.com/lark-parser/lark/blob/master/docs/grammar.md
+    return """
+    ?start: hello_world "!" number
+    hello_world: ("hello" | "world") ~ 3
+    number: ("0".."9") ~ 5
+    thanks: "Thank"i " for testing!"
+    """
+
+
 REGEX_PATTERNS = [
     "a b c d e",  # ensure proper tokenizer whitespace prefix handling
     "(123456789)|(abcdefghijklmnop)",  # ensure consistent correct sequence handling during batch
@@ -153,6 +164,7 @@ def enforce_not_implemented(model_fixture, *task_names):
         "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
         "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
         "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
+        "cfg": ["model_llamacpp"],  # TODO: fix llama_cpp tokenizer
     }
     for task_name in task_names:
         if model_fixture in NOT_IMPLEMENTED.get(task_name, []):
@@ -249,6 +261,28 @@ def test_generate_format_bool(request, model_fixture):
     assert isinstance(res, bool)
 
 
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_cfg(request, model_fixture, sample_lark_grammar):
+    from lark import Lark
+
+    from outlines import grammars
+
+    model = request.getfixturevalue(model_fixture)
+    with enforce_not_implemented(model_fixture, "cfg"):
+        generator = generate.cfg(model, sample_lark_grammar)
+        res = generator(**get_inputs(model_fixture))
+        # validate legal with the grammar via lark
+        # TODO: cleanup PartialLark so doesn't modify Lark globally
+        import importlib
+
+        import lark.lark
+
+        importlib.reload(lark.lark)
+        Lark(
+            sample_lark_grammar, parser="lalr", import_paths=[grammars.GRAMMAR_PATH]
+        ).parse(res)
+
+
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
 def test_generate_text_stream(request, model_fixture):
     model = request.getfixturevalue(model_fixture)
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
index 0d1908ebd..3469dcbc0 100644
--- a/tests/generate/test_integration_llamacpp.py
+++ b/tests/generate/test_integration_llamacpp.py
@@ -5,7 +5,6 @@
 from pydantic import BaseModel, constr
 
 import outlines.generate as generate
-import outlines.grammars as grammars
 import outlines.models as models
 import outlines.samplers as samplers
 
@@ -243,15 +242,6 @@ def test_llamacpp_json_schema(model):
     assert isinstance(result["bar"], str)
 
 
-def test_llamacpp_cfg(model):
-    prompt = "<|im_start|>user\nOutput a short and valid JSON object with two keys.<|im_end|>\n><|im_start|>assistant\n"
-
-    # remove this statement once cfg is implemented
-    with pytest.raises(NotImplementedError):
-        result = generate.cfg(model, grammars.arithmetic)(prompt, seed=11)
-        assert isinstance(result, str)
-
-
 @pytest.mark.parametrize(
     "repo,model_path,hf_tokenizer_uri",
     [
@@ -319,15 +309,15 @@ def test_RegexGuide_caching(model, temp_cache_dir):
     # These two different models and tokenizers should not have the same state
     # mapping results
     assert (
-        generator.logits_processor.fsm.states_to_token_maps
-        != generator_2.logits_processor.fsm.states_to_token_maps
+        generator.logits_processor.guide.states_to_token_maps
+        != generator_2.logits_processor.guide.states_to_token_maps
     )
 
     generator_3 = generate.regex(model_2, regex, sampler=samplers.greedy())
     assert cache.stats() == (1, 2)
     assert (
-        generator_2.logits_processor.fsm.states_to_token_maps
-        == generator_3.logits_processor.fsm.states_to_token_maps
+        generator_2.logits_processor.guide.states_to_token_maps
+        == generator_3.logits_processor.guide.states_to_token_maps
     )
 
     # Just for fun...
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index f3fb9682e..cdd57e0c6 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -524,15 +524,15 @@ def test_RegexGuide_caching(temp_cache_dir):
     # These two different models and tokenizers should not have the same state
     # mapping results
     assert (
-        generator.logits_processor.fsm.states_to_token_maps
-        != generator_2.logits_processor.fsm.states_to_token_maps
+        generator.logits_processor.guide.states_to_token_maps
+        != generator_2.logits_processor.guide.states_to_token_maps
     )
 
     generator_3 = generate.regex(model_2, regex, sampler=greedy())
     assert cache.stats() == (1, 2)
     assert (
-        generator_2.logits_processor.fsm.states_to_token_maps
-        == generator_3.logits_processor.fsm.states_to_token_maps
+        generator_2.logits_processor.guide.states_to_token_maps
+        == generator_3.logits_processor.guide.states_to_token_maps
     )
 
     # Just for fun...

From 6fb355901746dbd56634a0e1ebf34f94a9b5f360 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 27 Jul 2024 11:16:22 -0500
Subject: [PATCH 060/505] Add experimental-warning to CFGGuide, update
 warning-linked docs

---
 docs/reference/generation/cfg.md | 9 +++++++--
 outlines/fsm/guide.py            | 6 ++++++
 2 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/docs/reference/generation/cfg.md b/docs/reference/generation/cfg.md
index 4f0285c11..e3b177800 100644
--- a/docs/reference/generation/cfg.md
+++ b/docs/reference/generation/cfg.md
@@ -30,10 +30,15 @@ print(sequence)
 # (8-2)
 ```
 
-!!! Note "Performance"
+###### Disclaimer
 
-    The implementation of grammar-structured generation in Outlines is very naive. This does not reflect the performance of [.txt](https://dottxt.co)'s product, where we made grammar-structured generation as fast as regex-structured generation.
+!!! Note "Experimental"
 
+    Outlines current **community-contributed** implementation of CFG-structured generation is experimental. This does not reflect the performance of [.txt](https://dottxt.co)'s product, where we have optimized grammar-structured generation to be as fast as regex-structured generation. Additionally, it does not fully align with the approach described in our [technical report](https://arxiv.org/pdf/2307.09702), aside from its use of incremental/partial parsing. This feature is still a work in progress, requiring performance enhancements and bug fixes for an ideal implementation. For more details, please see our [grammar-related open issues on GitHub](https://github.com/outlines-dev/outlines/issues?q=is%3Aissue+is%3Aopen+label%3Agrammar).
+
+!!! Note "Greedy"
+
+    To mitigate performance issues, CFG-structured generation will use rejection sampling and iterate over the candidate tokens highest logit first,, completing once a single valid token ID is selected. This is effectively greedy generation.
 
 ## Ready-to-use grammars
 
diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index aa073d107..44a918494 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -1,5 +1,6 @@
 import collections
 import copy
+import warnings
 from dataclasses import dataclass
 from typing import (
     TYPE_CHECKING,
@@ -319,6 +320,11 @@ def __init__(self, cfg_string: str, tokenizer):
         """
         Construct the PartialLark parser and set the empty initial_state (PartialParserState)
         """
+        warnings.warn(
+            "Outlines' public *community-contributed* CFG structured generation is experimental. "
+            "Please review https://outlines-dev.github.io/outlines/reference/cfg#disclaimer"
+        )
+
         self.cfg_string = cfg_string
         self.tokenizer = tokenizer
         self.eos_token_id = self.tokenizer.eos_token_id

From 72377db17049fea7a8ec8e2a9bdac9b546e1d17c Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 27 Jul 2024 11:16:54 -0500
Subject: [PATCH 061/505] Add benchmark: CFG rejection sampling + CFG no
 rejection sampling

---
 benchmarks/bench_cfg_guide.py | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/benchmarks/bench_cfg_guide.py b/benchmarks/bench_cfg_guide.py
index 8f6de914a..14dc31c73 100644
--- a/benchmarks/bench_cfg_guide.py
+++ b/benchmarks/bench_cfg_guide.py
@@ -38,23 +38,31 @@ def setup(self, grammar_name):
         )
 
     @staticmethod
-    def _run_random_cfg(guide):
+    def _run_random_cfg(guide, rejection_sampling=True):
         state = guide.initial_state
         token_ids = list(guide.tokenizer.vocabulary.values())
         for i in range(40):
             # simulate ordering of logits top prob to lowest prob
             random.shuffle(token_ids)
             # simulate sampling and state update
-            next_token_id = next(guide.iter_valid_token_ids(state, token_ids))
-            state = guide.get_next_state(state, next_token_id)
+            if rejection_sampling:
+                next_token_id = next(guide.iter_valid_token_ids(state, token_ids))
+                state = guide.get_next_state(state, next_token_id)
+            else:
+                next_token_id = random.choice(guide.get_next_instruction(state).tokens)
+                state = guide.get_next_state(state, next_token_id)
 
     @cache_disabled()
     def time_cfg_guide_setup(self, grammar_name):
         CFGGuide(benched_grammars[grammar_name], self.tokenizer)
 
+    @cache_disabled()
+    def time_cfg_guide_run_rejection_sampling(self, grammar):
+        self._run_random_cfg(self.prebuilt_cfg_guide, rejection_sampling=True)
+
     @cache_disabled()
     def time_cfg_guide_run(self, grammar):
-        self._run_random_cfg(self.prebuilt_cfg_guide)
+        self._run_random_cfg(self.prebuilt_cfg_guide, rejection_sampling=False)
 
     @cache_disabled()
     def peakmem_cfg_guide_run(self, grammar):

From 8730325fed857a2b55e153c7113b52de7ad811d2 Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Fri, 6 Sep 2024 17:33:02 -0500
Subject: [PATCH 062/505] Include hidden files in coverage CI upload

---
 .github/workflows/tests.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 344dfb55e..dd9d2836e 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -53,6 +53,7 @@ jobs:
         name: coverage-data
         path: .coverage.*
         if-no-files-found: ignore
+        include-hidden-files: true
 
   coverage:
     name: Combine & check coverage.

From c3a1adf770b08aec2345ea005b51b2e30cfeb190 Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Fri, 6 Sep 2024 17:35:50 -0500
Subject: [PATCH 063/505] Update artifact actions

---
 .github/workflows/tests.yml | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index dd9d2836e..53d20b70b 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -48,12 +48,14 @@ jobs:
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
     - name: Upload coverage data
-      uses: actions/upload-artifact@v3
+      uses: actions/upload-artifact@v4
       with:
         name: coverage-data
         path: .coverage.*
         if-no-files-found: ignore
         include-hidden-files: true
+        # TODO FIXME: This is only using the last run
+        overwrite: true
 
   coverage:
     name: Combine & check coverage.
@@ -74,7 +76,7 @@ jobs:
         run: |
           pip install --upgrade "coverage[toml]>=5.1" diff-cover
 
-      - uses: actions/download-artifact@v3
+      - uses: actions/download-artifact@v4
         with:
           name: coverage-data
 
@@ -90,10 +92,12 @@ jobs:
           diff-cover coverage.xml --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
 
       - name: Upload HTML report if check failed.
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
           name: html-report
           path: htmlcov
+          # TODO FIXME: This is only using the last run
+          overwrite: true
         if: ${{ failure() }}
 
   build-wheel:

From e28d42320df05ca4cc9ca9376cdbf8fff655a78f Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Tue, 10 Sep 2024 10:36:01 -0700
Subject: [PATCH 064/505] Add documentation request issue template (#1138)

Didn't see a good issue template to add documentation requests, here's a
simple one.
---
 .github/ISSUE_TEMPLATE/doc_request.md | 21 +++++++++++++++++++++
 README.md                             |  3 +++
 2 files changed, 24 insertions(+)
 create mode 100644 .github/ISSUE_TEMPLATE/doc_request.md

diff --git a/.github/ISSUE_TEMPLATE/doc_request.md b/.github/ISSUE_TEMPLATE/doc_request.md
new file mode 100644
index 000000000..f9dd65f92
--- /dev/null
+++ b/.github/ISSUE_TEMPLATE/doc_request.md
@@ -0,0 +1,21 @@
+---
+name: 📚 Documentation
+about: Help with documentation improvements.
+title: ''
+labels: ''
+assignees: ''
+---
+
+### Documentation request
+
+Documentation is a high priority for outlines. If you would like to see something added to the documentation, please help us out by:
+
+- Checking to see if the documentation does not already exist.
+- Providing a detailed explanation of what you'd like to understand better.
+- Highlighting any specific difficulties you ran into.
+
+### Are you willing to open a PR?
+
+Tell us whether you are willing to submit a PR for this yourself, we'd greatly appreciate it!
+
+Thanks for contributing!
diff --git a/README.md b/README.md
index 364e77244..59a3722c0 100644
--- a/README.md
+++ b/README.md
@@ -7,6 +7,7 @@
 [![.txt Twitter][dottxt-twitter-badge]][dottxt-twitter]
 [![Outlines Twitter][outlines-twitter-badge]][outlines-twitter]
 
+[![Documentation][documentation-badge]][documentation]
 [![Contributors][contributors-badge]][contributors]
 [![Downloads][downloads-badge]][pypistats]
 [![Discord][discord-badge]][discord]
@@ -355,6 +356,8 @@ answer = outlines.generate.text(model)(prompt, max_tokens=100)
 }
 ```
 
+[documentation]: https://outlines-dev.github.io/outlines/welcome/
+[documentation-badge]: https://img.shields.io/readthedocs/outlines
 [contributors]: https://github.com/outlines-dev/outlines/graphs/contributors
 [contributors-badge]: https://img.shields.io/github/contributors/outlines-dev/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
 [dottxt-twitter]: https://twitter.com/dottxtai

From d11048bea7d55dea7bac92be6b74d84cdf4c49e4 Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Tue, 10 Sep 2024 12:30:18 -0500
Subject: [PATCH 065/505] Set tokenizer versions in tests

---
 tests/fsm/test_regex.py                     | 17 ++++++++++-------
 tests/generate/test_integration_llamacpp.py |  5 ++++-
 2 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/tests/fsm/test_regex.py b/tests/fsm/test_regex.py
index f7aa4ae67..824588b22 100644
--- a/tests/fsm/test_regex.py
+++ b/tests/fsm/test_regex.py
@@ -402,15 +402,18 @@ def test_create_fsm_index_end_to_end_multi_byte():
 
 
 @pytest.mark.parametrize(
-    "hf_tokenizer_uri",
+    "hf_tokenizer_uri, revision",
     [
-        "gpt2",
-        "microsoft/phi-2",
-        "Qwen/Qwen1.5-0.5B-Chat",
-        "NousResearch/Hermes-2-Pro-Llama-3-8B",
+        ("openai-community/gpt2", "607a30d783dfa663caf39e06633721c8d4cfcd7e"),
+        ("microsoft/phi-2", "ef382358ec9e382308935a992d908de099b64c23"),
+        ("Qwen/Qwen1.5-0.5B-Chat", "4d14e384a4b037942bb3f3016665157c8bcb70ea"),
+        (
+            "NousResearch/Hermes-2-Pro-Llama-3-8B",
+            "783fd50eb82d7f57758de033861f54d62dde234f",
+        ),
     ],
 )
-def test_create_fsm_index_tokenizer(hf_tokenizer_uri):
+def test_create_fsm_index_tokenizer(hf_tokenizer_uri, revision):
     # The combined regular expressions of a lexer state in a Python grammar
     regex_str = "(?:(?:[0-9](?:(?:_)?[0-9])*(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*|(?:[0-9](?:(?:_)?[0-9])*\\.(?:[0-9](?:(?:_)?[0-9])*)?|\\.[0-9](?:(?:_)?[0-9])*)(?:(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*)?)|[0-9](?:(?:_)?[0-9])*)(?:J|j)|(?:[0-9](?:(?:_)?[0-9])*(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*|(?:[0-9](?:(?:_)?[0-9])*\\.(?:[0-9](?:(?:_)?[0-9])*)?|\\.[0-9](?:(?:_)?[0-9])*)(?:(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*)?)|0(?:x|X)(?:(?:_)?(?:[0-9]|[a-f]|[A-F]))+|0(?:b|B)(?:(?:_)?[0-1])+|0(?:o|O)(?:(?:_)?[0-7])+|(?:(?i:([ubf]?r?|r[ubf])('([^\\\\']|.)*?'))|(?i:([ubf]?r?|r[ubf])(\"([^\\\"]|.)*?\")))|(?:(?:\r?\n[\t ]*|#[^\n]*))+|[1-9](?:(?:_)?[0-9])*|\\\\[\t \x0c]*\r?\n|continue|nonlocal|assert|global|import|lambda|return|async|await|break|class|False|match|raise|while|yield|case|from|None|pass|True|with|def|del|for|not|try|if|[^\\W\\d]\\w*|#[^\n]*|[\t \x0c]+|\\.\\.\\.|@|\\{|\\(|\\[|\\-|\\+|\\*|\\~"
 
@@ -425,7 +428,7 @@ def test_create_fsm_index_tokenizer(hf_tokenizer_uri):
     num_bytes_fsm_states = len(bytes_fsm.states)
     assert num_bytes_fsm_states == 235
 
-    tokenizer = AutoTokenizer.from_pretrained(hf_tokenizer_uri)
+    tokenizer = AutoTokenizer.from_pretrained(hf_tokenizer_uri, revision=revision)
     tokenizer = TransformerTokenizer(tokenizer)
 
     states_to_token_subsets, empty_token_ids = create_fsm_index_tokenizer(
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
index 3469dcbc0..0a98f0226 100644
--- a/tests/generate/test_integration_llamacpp.py
+++ b/tests/generate/test_integration_llamacpp.py
@@ -329,6 +329,9 @@ def test_RegexGuide_caching(model, temp_cache_dir):
     assert structured != structured_2
 
 
+@pytest.mark.xfail(
+    reason="Some versions of the Hermes-2-Pro-Llama-3 model have a broken config"
+)
 def test_tokenizer_vocabulary_decode_sanity():
     """Assert the decoded newline token (198) is the same as the normalized vocab token"""
     import llama_cpp
@@ -337,7 +340,7 @@ def test_tokenizer_vocabulary_decode_sanity():
         "bartowski/Meta-Llama-3-8B-Instruct-GGUF",
         "Meta-Llama-3-8B-Instruct-IQ1_M.gguf",
         tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            "NousResearch/Hermes-2-Pro-Llama-3-8B",
         ),
     )
     tokenizer = generate.regex(model, "a").logits_processor.tokenizer

From ae1524cf2a02b5a2e498eae45636e22eca92a194 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Wed, 11 Sep 2024 12:52:18 +0200
Subject: [PATCH 066/505] Change Outlines' logo

---
 README.md                          |   2 -
 docs/assets/images/logo-square.svg | 123 ++++++++++++++++++++++++++
 docs/assets/images/logo.png        | Bin 372647 -> 6354 bytes
 docs/assets/images/logo.svg        | 133 +++++++++++++++++++++++++++++
 4 files changed, 256 insertions(+), 2 deletions(-)
 create mode 100644 docs/assets/images/logo-square.svg
 create mode 100644 docs/assets/images/logo.svg

diff --git a/README.md b/README.md
index 59a3722c0..f02c126ae 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,5 @@
 <div align="center" style="margin-bottom: 1em;">
 
-# Outlines 〰️
-
 <img src="./docs/assets/images/logo.png" alt="Outlines Logo" width=300></img>
 
 [![.txt Twitter][dottxt-twitter-badge]][dottxt-twitter]
diff --git a/docs/assets/images/logo-square.svg b/docs/assets/images/logo-square.svg
new file mode 100644
index 000000000..1d1eb3a38
--- /dev/null
+++ b/docs/assets/images/logo-square.svg
@@ -0,0 +1,123 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="95.38842"
+   height="113.55765"
+   viewBox="0 0 95.388419 113.55765"
+   fill="none"
+   version="1.1"
+   id="svg314"
+   sodipodi:docname="logo-square.svg"
+   xml:space="preserve"
+   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
+   inkscape:export-filename="logo.png"
+   inkscape:export-xdpi="270.25565"
+   inkscape:export-ydpi="270.25565"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg"><defs
+     id="defs314" /><sodipodi:namedview
+     id="namedview314"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     showgrid="false"
+     showguides="true"
+     inkscape:zoom="1.0083333"
+     inkscape:cx="280.66116"
+     inkscape:cy="25.785124"
+     inkscape:window-width="3440"
+     inkscape:window-height="1416"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1"
+     inkscape:current-layer="svg314"><sodipodi:guide
+       position="-146.75933,64.2455"
+       orientation="0,-1"
+       id="guide314"
+       inkscape:locked="false" /><sodipodi:guide
+       position="275.80377,3.1942902"
+       orientation="1,0"
+       id="guide315"
+       inkscape:locked="false" /></sodipodi:namedview><g
+     id="g13"
+     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
+     style="fill:#d53135;fill-opacity:1"><g
+       id="g314"
+       transform="translate(1367.0913,152.50405)"><rect
+         x="28"
+         y="644"
+         width="28"
+         height="28"
+         id="rect1"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
+         id="path1-7"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="28"
+         y="112"
+         width="28"
+         height="28"
+         id="rect2"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,560 v 28 h 420 v -28 z"
+         id="path2-0"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,476 v 28 h 420 v -28 z"
+         id="path3-4"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,392 v 28 h 420 v -28 z"
+         id="path4-0"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,308 v 28 h 420 v -28 z"
+         id="path5-8"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 364,168 v 28 h 140 v -28 z"
+         id="path6-5"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 336,196 v -28 h -56 v 28 z"
+         id="path7-1"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="M 336,112 V 84 h -56 v 28 z"
+         id="path8-6"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 364,84 v 28 H 504 V 84 Z"
+         id="path9-6"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 56,672 v 28 h 476 v -28 z"
+         id="path10-2"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="56"
+         y="84"
+         width="28"
+         height="28"
+         id="rect10"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="112"
+         y="28"
+         width="28"
+         height="28"
+         id="rect11"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="84"
+         y="56"
+         width="28"
+         height="28"
+         id="rect12"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
+         id="path12-1"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 560,28 v 616 h 28 V 28 Z"
+         id="path13-9"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="532"
+         y="644"
+         width="28"
+         height="28"
+         id="rect13"
+         style="fill:#d53135;fill-opacity:1" /></g></g></svg>
diff --git a/docs/assets/images/logo.png b/docs/assets/images/logo.png
index 9a9f234a746510cb4bb42acd52e2f06cccf1dc74..14d1c727e132c9b07a9d922a46eece1af4e9a394 100644
GIT binary patch
literal 6354
zcmcgwd05i<)~9JE=T3_k?dG`UymH+#r82c#>qc6cIcj66q2+=B;f9KsXv=K^Q*2yw
zFB!MoCKXLZwA{+wLJ<r}O)(V-aRU|ZkM8`_`}W@DxxMcnJm-17{NSAL`JHn<pU)xv
zc+P(3cWU2(K%kwcPr+P3ps&3^pe@I~-3FZ8dHA#%Fn<U><&Fh`3=-v^EsXy%-3c73
zhMl|^<{E?u!(9pW2jOryV`Ly2>w6{G-#92VAZx)&4Fvk$`!wwM`N(WOhY(*7fY*Av
zj_37Mq$g-jFrGd7<BpLhxQO8nY(!sX+zoi@nfv$kYkrQ4yQS_Er?&5oKJhz`<0ZEk
zNs4!VO}Y4;vHpGbY1rJWhlS-gKRn6JQS}wf&$I2m{`UJFWZ!GPg3j#K_gd|!As<n0
zPCd_huF3eNjMKj<jp#31T{9PHq<f&V%NJ-9ah%&3If+X-TcfIM>Kbb51b+nX;^gnh
z-2zxp?&IB1xjp?Knlt58!46LZVf=A@VOvFo$nka0wRrO}+i~9*eeIag=O(Ks+n9P*
zTT9N6py^j*Fi9rVTv}}Z?CgcALqH%_Lpvn9extIpM%A!6MV)*Kw$!0f5C5E<dhddH
z!5a(bG8db}b)lo4dKOz>4tpP9duByo@=y2&`@$}zD}4(By)}6869l-&jX%e!d<k=#
zzSZg%-|dw28Xe=DcV|hV4k;2=g@~G_oUFGMa4>^F-Ah*sfiogGumHA&?dqKZh<gU`
z1xR=NF&Hq%l-jp{0h9DB!pe`|)m|EqxtK}XHFPmf*T($iFCmShF;Z?wTH{=mO3vtl
zjd+gLkAwuW8mq#({~R!^vWr%q(f~DlU1K=8IMz)M9_0{tKs1-JwFQv|j`QWFF(J==
znZ4!vIS+ohE91BIHJ}fO=MFt_>1qNWbQUT(d%K}&`*j9XFJEFO!u~q(4Upj-yf1sP
zUloQ`Dl+d~wY7Sgq1)BzfLw7}ddQMYk)|_7w(<m#Z=c`oNNzGNdX$!MUC-uxhpddm
z(mF@PB94e(zzO#9>xguXy-*ljy3+RKY@~z6o5s*!|5pLO%*}cRafrl{X!x|78+`;P
z@9dHN7ZLyiO40|q{XaK5hE?psleRjNrmRaHbPVM^)#7?5FH+uAr}DpuIrL!ZSVAQG
z$@ovL@dpBKX*!A^>ePyJVa3ChI_<0!CogO-kfObf+UrND*WNS+;Kl06+3s4qpn*Os
zcdHLMHm<ZmaJ}W_)~|58fT+v@CMI;Wr`c@dAbgSuOkG$~KU7cz6NX8*>h%W&)J{^f
z@fTn;Y(6A>x^DEUcbXD!$3WjggRZs>>+gTi`DHuo^$`)@Wy04p_uUR>^qV2v&*fDj
z7`}|7Y3^rRl+bWUT(VeHz^KF!GEG_^Vix$8^1jP54-vX()2jmf-%H{TT&r|QButMJ
zA9OK81D9)8mviL#H|zd1&?1p4FyjVT66vPx;PA8&Qt|3!p?EaiCcRb*WOiKm73f$+
zLg<yl%f0S%Xk~lIUxZ(^oDUrWrV8i>lDzc)hf!pH1G=FX68Us3%NaXno=z+f51r8J
z8`;?s_9G3bgk*iwvEAV6qurD5!H*)hM1F_;Xa9Dp#!O?O<e;dcY}2vr?rOz;P7+lM
zG@=0vkz@HD0mk~yGOsad)ow5_-X6GAsLAj7XJ^3I|BF#w(gYk8vgRq4&<A*l-OCZ{
z5$c>@hZm^ut~^1xZ3Ue+qF}Bn=m@qR^)>M9eHf6||6AiXfV-g=8g^y;Xu<rNI)Qh?
zrbcgXJvtD>|AmenXr>Y_SW{Rbvv26wYj=+uC`#WgE^ZQdM>M@@?~-!#`Yi=4_jO!P
zU(DRZZ3V3^Hur9$@%6#_Nlv0Q1TE|6=rA{Ko+JwbBl^b&XYeMI!a}<95KPq5!1}={
zKV@xjdHlmf*(SEM)H}{&v}`lr7IkwykIV2l6j7%app|PI{ed3O0fTWy`>n8-x?UfO
zNcV8!yTdonFvXUYPkmM|A6OH{@ArKFLX;;{VUC5pek#y-tuN|-UV~@jc?5pGXOQ(6
zj%-t&VAmhmF_}Z)jqEbva>BVEQ#~v#G4j&hCP1Or8|bdbwAd^-6V|4;-uG<x$ibK(
z&n)JWrmtl-ag9%0o1$a0$&<X?W+>{qCtI4#hn{R(uU=UlXDlS!s!+v>NW}*%;Vf~D
z(pzCLx%L;&ERN@Wea8gJut!!Mf^Tc4wmIStj!mV?;s??om=snD$Ff6<L}V7ulZnax
zAI4qghK`PQU(lYw=M)9><B_AavDwM~bMu>v=`sW?x>}&&T4ILAlv_&3Pc-T4Zy``J
zv416p;^YqrL{eBx9GQP$jIg*9EPPhB`DJQbUa@blg)ViK8mpS~=@p`aj%pO*p@B>O
zC7PK=Q`KCwD|@PbYN`@}`0DLxYfMsXkbQ!)Pnyx|Kt_7yPJ6|OjPin-^|YY%ixGwU
zJNMW@S^7dsjv^y!IC_$9hZnMgm*Oi#_Ia^`u?;JQSsfV}zzxS+r25Nxx<^92JRX#|
z-bJ?#;u_B7>GNgZ8$)rqdF+XT;Wa<xgdg|lL&e**^>u<B??&|1EA^=xqZOh6hxbZg
zFqLkl?KMWAT~YySb#4E;CAm}umXOtWF1<(JbMEJ2j&z$bl??H*56rew2Mb6Zic2}Z
z`%)#9dU(hR4N7?bh8TI+sDDqfb8HDi!lq--Qd}k-!OM<t9xkC<N_^Thx53y0#nwSN
z--FFpG~_e&ZB34;?LIk3{>J%~1{SWE0B9+3LpME23GLHmZbUL9SWl&yb$xJNWx<<Y
zsIqs$$<FGCtu3r$K)-+ZdZm1?B(-<p2<<>r2<eoe7s^O0%NrgQG#mqPk4vWVk??tR
zoM;W-Vm97qxXTKMD5TTpq79}V7n5#k$JE)0BW9H?_Br#kCw_8YN0ahws}4lP@WigY
z%~|p$E3*Pw)hkV1RTpa2ZPph^TZ=<ZcPx0E!TMcU0*EMvR4MRumo&xi2ZQZb1+w?G
z7&|!6p``yYQsB^4hi(%fFnQ-)XStQ8)72}2vMIj(9y@h70$sD(UuoPNRNYU*9-*NM
z`)Flz(aIt1ZnQT#UCl_yUe2wbkeM!(C8l0U)L1KJ*}E6dXR_eDKS4oRzhNVy2TN*c
z6w%E$uXHkNqc1OPcd!r5d({h6O;@pLc_c~A){`Q=#uIB6Nsnj)Nw+T1Ss1IUT3X=8
zER5f2S4n$e;B}_Fz%DFCYk+Z)Ostvke`98=>C7DwADdWDQVNl6%&s(#Ir(9(gapC7
zgi5AkZkE!Y(_9wFL`M~<W5&@R(R`^**uCud1n$$&@bG=~!BrS}fBZHu;NHM=Y+!kr
z9Q=1wO86|s{mX|h9|`~h|3ppbX*%Hw=LIAR)=-Gt(>*CGq9T-73}}Ir3ou4(cE%rW
z7@hB`V|#E%!lMsVaC_oKFcebuQ+aQ*z=8`CUxpE)5QQdg2FzM&V3;b0RihZtnQ>M$
zX2wno%gH4z=Bh1M^}yaCvsNfnBSQi&GJE2i%bBTFQR1#wu?v<R??<X>8wtdjYyw6$
zQ$MJ??W@b^p~(5gp%bn$9AnDRP(d%lTirO9n3+1=6VZ1fc2jsyP|t&LkDlw?n8kD_
z_h_jaOCML>Heu4IrhLI6Wde=F-;w(zL)z9Y`OsoX8NL8$Dak;}w4~c@*SX%r4giXS
zqY7&y4X9q{A$uf!TL>dpJ9#T|DsrRu9|Hg9Qv5NSAkf)nWuEj!o|gN?GWp=ArgUo;
zZlp)$i(3y}wPl9W%!E+NaC*?nlWPBIW&1sQ%*`70Tocw>HfA4j1(uzYANzRYnCa|q
zf*qrU=kUO25#3cBQvt7r=h)H?NJ^{zq*x)mg_LnhoTUz2RcX+tb2o!sYIof&-AfBx
zDL5)04J}BK^1rJTw-FS8UxGq48`tu^M;hrdgpP$Gc<^!23*26DV8_C1=X~is>YesF
z9zST;JiLidt`L2cZ;^40@|g-g*f&BDm3o<0^DGnn%52~!+;zQ^$Wc<D;%8Y#TY#Jp
z4IIb+CivQE=dvY1@A7p6Io)5o*@s`N%)<+<SF>j<6|#~4Y!m^n?sWve*PIEtV7zzE
zvQyCoVjN%W#GTKLD6GGVIbt;y?)8!2WYaxHo50uC$->(jdyQirG9p_GRfbj)b?cA4
z`v-C<sbn*l^Siv{TPiF)N?A#>?(XIS!2#iBnNZsu5}q2PXcYpDfx97|3*>L*YO+PO
z?^9;`sn&H}A8!D`Qi`#2ty^Oo=QcVzgrgrh3<5DbyK?8w_J}kV0nS4+9L^vfnz)l3
z^u!h;2R+lEHjy75^_-Gb{+*Wu_ucGMK=tT!Z293?2cGBMWMdWl`1Q4GRzRP1J)1uZ
z=iu8XHZBirT;v|ru3cC~U=Bw`m>8#DJ6YJ9X)qb}esQs!z<b~pr*Z@7Hi^O11kG94
z=vzV)i4YRWBFpG@9twRO%fuISCl@dGVi@K)&DeQAQ9f}k+9zM&+ld`zOIa&6$~X8m
z&Z9DRc!Vy3Y}Xd;lTA8oA|%(=GFNgS+CGJcarSMwj!@KODh3nlD<YW%X*jBbtL2_Q
zE#=R$<P5f~)S0w)*+^JV#~#x{WJX;8*ef6tX#lS>_Jm8$!=fJxG=LLO(SMT@zbD;w
z9Z(ekw?fKe-6<PxOtG%dBRkp=BPZobYpY8%C|c&%HKh}BJz<fnJsZUoA2;Ckb-f~Y
zn|GfLlUnHOOjA0zFS}gN5SGWs<-pb~Fi{A(!r(`1CzcjS8NlHYw5*=DIj|0GSRS5q
zU%zQ|exjMFmKd%Xq%A$3@bm!WjM4H&YiqkV&?FXTNDiU0+<ncil)$izGHMLkF!$o<
zQ){36RjpWaR1c=Tx|u6MMm6P~a_x@hWJ3RHTMW+OdhYWAnOzC=pSlMY8p3&s(D_&g
zQj!}jrMTI`S|-d=|CJ;eY%0BLlOnlK21@BI)8|&@ZODbaktd8W9zrE5{b==)PeTeZ
z4*;}es%(vt2&{W{LxZ<H@OGCZ6W5X}MCk1l6+1@CgbvDz0;8GxukN-OtdT`OrESba
z*kHM{uZkcOujRhu1NS?-#-+QUNhWWX$MON-G>gG5Y`+^Quxi$ElFP8SGRl9ZtPjmt
zf?8O9y4&)K57J|AL6}UUO5rz|x&<5VCCd<aC)(9EOI6QRdCC>j77pi)>dT5@O}Zwp
z5vfYviE4rw3vY<j4YuLq^J|IoIa6FJD}={vOp8P^A{`OBH;6>t*+z8RH$aQE?F>oV
zF!xJXL_0FBH}u86$M;(v?AcNCO{PZ>QAt)Ux~U|nFwIT<9vPRyg*wY<qP33s3%aZr
zw~a?z9_+!EDlj2F099iL3k#h4$rA;3))viCGbiL&a6QmA(g_w8V^U|Pp-EJ8h0_>B
z1wH)c@Xr)8l(k`8y0}$AaF4jqY`&dRWPiSW`~ASc<$=6&hnmea2tS~D<4k_a^ss%h
zx^g0?`DL}aNzWefSY?r@@;yLcjKIrf`d%nfe{bcxCdKf7q+0(s@N#-L^L(a5P1qps
zE4D{vM)T$Xpi2NWY?)poeX^W$y*~8q4Pqgq&VF<wihH!SwIt#LpKfxtG=hTfe~1j4
zvFuD{eb4OF&8RFQ$3`Qjv#+@2Dx2`V{rG*g6r^p{!9W`hj@+pk(}&$lyOwOZSDitH
zPwMiQ0}<Wc2tJued_~TT(B-M8Ij$e;8WR{4G|`b+SZtcPp6kj=nQD5v9f=jsDMH~7
z`wCW6wXULG<=ru;9?X4OB-lYvPn;`NaAXbP$e57O4?{i6Q!i5Qh39VtcLT_{r}``F
zz5rUV*(kX`(mnRs)u6*q%myDRnfI0Nho|rUgM1~)E4yka7wVFS|8m&^722X)QV@L<
z1()N(m4P0sHy`&`W<kKhu|9QlOcgAc{~eM4ZF2GXZN}%`<QIMN<qv0QAxT>c5~9M6
zNSaRf)lkP8aLWVsdiS=S^&yP6te+{dZ0t)fwEEE@XQn^ab+Y1RGLX0Irfpb52jx_K
z`w9)CG&SWMlFPZ1z%J5d$h+7y!tp?}O<h%uIaCqI@V_weKMMn&Jybm#^>*#-bw8Zu
zeBQgWo^unD=U0KX;ucP?nF%u<&6%;_4oB`ck^2P~3};I3Lt5KoJy7DDC5311^G^yn
n)|b3uYCeL;hSnF2OzaJM`&DVMN{0>L8+7{QIatMCF5mha?z5d>

literal 372647
zcmeFYbx_>Rwl<2pyUQTK-CYI^?gVFW*Psau?gR-QTnD${!QBGEf@`n<14*y|xs&AG
z`|MNSS9R-D-TU96n5vmyuYP*1XFY3mPgPH>mWC2G1_cHj92~ZaGEfH&4iW433k?PK
z<SB$M2o8>&AyCiAN5{&K&cn;y&e6q|&L_abmd@7S(GCvIf3+;n@g-G<SL81cu{lC$
zUprO8p+Bnmhu(O$V-?BUjx&S!(zszYMm*1Gnb7lx<q-25+K1&J4X%S#Rqo+$7AIAs
z{iM?&Pm~uR=<%zEhqmrl7LRAXH{U2zGQTe#T6}mrdkEe8K}&l78YjH|iksBw^k?GT
zPvOA(;G{EL(zpW6a{S}d#~!U;=3~2ulsiG-%K#3G!0KM90f~ayt%6@_@}EjB>07s=
zA1gu(?cD9pLA*4WU+x@8q!D=@1?|JR31~}*ds3tW%b>hQcjgJ`chv#+cNX?QLwWU(
zyYel=(^12ndJDf0(^{fl_eN<oue-ZDZVNy9Upti7-=r>=#Vf8K*`A&PHF-l6zej&w
z-U{;Ty2uWQOTo}J`+)mnPkYm_+uaLuqf<1>Ao#%j3&D@a;|E9J4*-!6y!c(S9B#@%
zCw(NfV*5__EIUV3)~;vcl|lmhYEYwbq9oT6XG6vVy0F3N``X6*Q_i5KTW^k?jrTtf
zHa95zHo<KpeZvEoVIjApkDqAU4Xg0N+@%lWJcldC?Z{q;zS?)KQEuN-nRX8gdAQpP
zd0_}bZniUKozGsm>lm8au~4>jFG>`)lM3<TS$OP$(;cei4@V!tXmINHV(DKdb4F9M
z0fABElQVX0kmY7P`Ny&nI3SG*8#%O9J3WG5#wL@N;rR^!X%=r=kA;kC()liRbs3@*
z-{*PFFfFfp&6XIcFlFjE;j<(;33lW-SGB%hJ#|>pnXGMDeR1mgeFgjDRL}QSlZ#ha
z=c4Ub+ZCwpWS|j8{^_rF*}m^T9~wo>s<PKNoj6t3H}1X1e*LxS)UE5nbAUcwL+mbH
zS5y2k;d!N*x9~>rWNq8V>(N>LpIoN}PY>bxm|Obs?q%Oc?!6cNTbF1v(ZbSd1VR19
z&%wm?FAxyKoTh^@SruyYmZto+XT5CR7(2fYH%K5>do~n2nB3}p*<Adh<JWR$N6u@L
zw%ZUh(TcL4(d;*ezCWw=*Y+mMcXA_h^xSP|FUT7&rA>bv&YmA)#<K6juYN~*8THgA
zJ!I0f;(nDe6Mx^hV)pf*-u<%oXWkOd#fXOCGU?XrXFJVr83Fce&EN6o)fmq>6ZK;|
zoH3k^?d#uV)*Vi6+8@gep%d7szX3a*l|*`ux;@F@-oWKuE>w%lZ@fAe>Hu9emp5Qs
zu9$QQx(Z()e4K5SxTr#!K2sBIY1m`55yQ_ytUn8Kee2V<F)T9)|2<=P@yoO;wZSw_
z^JEkE2%oAc|I~+Fstp|JeyvoR+~-Y&4~6f@E{vnQ;9QRH*X+qn#{y%$KOeP4&4A8{
zd>oHHybR-V)R1>{7Ycs(IeY$1;n$je)!Y4;_+uMugJkqgDy-fwa#S^D=PcuHo`{dj
zkBJH*CTx4$hj@cINh>Kzm9jcc_8*jR@GKjt+1alajvHc{8Oy(Q%28Ap8`AtB`Xrz^
zUqrHtnPpkV?##1md)_|D*?O#Q)t?oR->-uMa;#=BIyYyS^>1NZl!%`mOrmhYnqoL|
zuQG0L4agE%cggjup*G1Wsq?1y{#G|LRua9|sko=aWmTXqW$5KW$dQ<x6&qy6A{E9^
z;flHb-mUqA7fOgOCthiByzS+__qjvzqz|ym6tLbe_T_`P2q=iFfGcYbAWZHX)qYdU
zgq?55F^{=tM;CijQ|x)EUWeL7{1GQh)Kx8{jlj{x!{r-Zr`zsc2G<mF{|AdcM89fs
zD7O0M=dWfwIA3sqZE`j6P8dy1$;&s_Oc(x&eerAOm8E7!8tK@=>d3b;coqEvCb;iJ
zj@GxY-#HZ!PML!AOM#!iDS~n=9r&7n%j0DwQ?52xp_vwFc;G2~|F)C%=;~*=FLS0_
zojPrnP0MKBq>M8TdRL!kp)QQ&6QQuq?ba1~I(0we!|v(RBTi?ME&91;q&!*EOcPAM
zQxjGqX_kpG+cbWaE4eRDHev)_&*RA&U`iOabVX96?%57%jpvKt+Fb23tseR+^di6S
zb1QfLLLI9n8lSTvnTO(M^6PDqr?kPb?uVPKcTahXh=FiK1o&xmv|}8-RcuAJbS2Rp
zuk^D92bVG?<53w}&_=Nz4nF@BMXoePo-h+0^^v`Q0b&y7<gw2+m3Vt<dSG)_Bs0O~
zu$e9E1D6u)|MZq?A@lHjdu6T$`<83gSIl(A<z*RY(sUhXtb>fcDe5>r$f5JzW=zYY
zReTib7uh<3YfQR=#G=PCd3GKPDLGvo4e8n3>De2h_1P<Y)MFvMvL9Gn>*ylb2SeGJ
z?0J}H;fk+EG57bI#tXNX0gGAT3=f`%Sb%1F5hw_GehomRC#m^aIpg!hf`hV%78V|R
z!*?2d5n7wO6%e;fKo@2{{YEyQWn8AwlV>AwDaS{ztg67$UX+#;KLf2(T2qwDn_7aY
z)P(8F+g2VlItC8)RDxhQFUE<=gecUd<;R7AEW;kH(zo(YDdVjK=J8nVI5n1!2-RbN
zg!hNaxVtX!u{<2l8PSJ^B0iGF=Athme&rJsl}1s^4(f0teM>MrCkKLqXAE}0>=tlQ
zi%!RPVz>;c&J{343Dx79Vf#2c+%fP{fmhgA5wDrZD__J3O1P}FI_+l3afh0_)bS<&
zHy-WYt8pOBpQ@6bW^H(}zHsT;5?;Bb90{K4cf_LgRIBB|MY}2w?5+IwbOBe53L^<W
zc8liF`_OI;cp!fAhVQ!gKBBnE@w)D!i3PjeCGDp;O1aUz@A32us6kr~+a>#qZH<G~
zi66+A^M*|35v!yOt-ogEyKVkb#BHBL^eK)&B>d-;ejnhMaP#GUX_Qz@M0cegl0sc;
zF2&;Hb{P+;!B2jQxYPX{^tlOSdlOw@{@rnEFQsl;5*??3Af;bnH<tbMr&HVTTsA-a
zX=tya`Z!2F0`|>}-x;W0yd`H{UOcxa=np$azkJKD2ab~^g=-S0kPH1ubk&ZOv8tS<
z>_)&D?bEJiErMz(Pu+0dP;biLFyq6CH%}1S-Z#=3DrARj)y03grGiY8{5iy-@9SrT
z^sd}g5<E1P;hID|yiW~=GG%Xq{awG(f>|7pC=JlNEm~Spzl(#3(xO;KnffL{V;4Xq
zD9&zpDEtSN=b^e`Ey}B$no{n130l-0OA*Ww8kk5gX)(*0?B>{$XN%yKFYtKj_z6<j
zlbyez%u7v0M0R-Oj=^EL!ll1;EahrA(_<$vL@6iLlgIHhn)O2(fj2fsKtyd$(<4ji
z$!?FB-iELicszSRoUm6~cu$$)k>HU^Mu^^=c*GWi)Z&WF<OXIo0teu<nD*p{UV1-@
z6u)sj+@DdLeJHG%@ek35vb7vQJ+bc9^61_uuSzTp5d;y4L-UX#5p@*Qls#j~W3X5W
zaZ>#hM4OUY00=z-AZqfrKrIB<p@k@kU7&Rsx4{MyhwO8Ik^|a{&A^=ID_bf`55hIZ
zB8oQTckVkTXg5AZ?x_UT!xFsC=i{34qF#bT%QW&}mJlI4+?2-43rUr5Nn+Be*QS}o
z&-iGorx^k}xG1GBZeqVlct*w@?_UF}EhE4+=;*S;-DXinag>dkc)26a#d;%J(YfA4
z(80r~dzAO2$=9mrnF?<Yve2aE33xcwQqnI%vz}>1evOjS!9J?nhRPovaf4YlLKoBB
zc$Pd<=!`0h<VZeG1-ikd1~Z_MsS#B1hC9Ctt_T-IGq%GM3WtWaV1aKXt-Q`@QPZR-
z@<i>m9qFH&zhhxdG03a}PZwnhe;yLpeUD!;DHUci&r#=rhO}GWrk2r~`Yn%oan>PV
z)Ugv|PG#3}xjC**8ly$9pBC?Q*C`i{H2-LqpowP!5L)^<=4I4Uq>PPp^(O+cAYUyu
zfO7o`HiJOfaWx~&MI^7`_Ue(;%C!^8j#V*%K6XUOPq1hokd=?Y;bzXMtU;i?BMi%=
z+C`3!JQ?AcYC40Qin|ZEr9kt6Mb#xGQz*?D>14B=!{yW0c?VK?^I78tD#Zk{iK3D@
zynudw;|4cwtPR3>2WpgzhuXz5evdM6n>!1$%)y+nfS7YMfz6^7VxLk@j@Ij7#V=M_
zJB-0TKsw7?ysvzPaIC?R&%vd3^*~<<%i>FJbe#@I8{MtcZNbe$cIPfjKkC#CHJ~>O
z&1PafR;v0sn`5@8J_kutsDlJHcVx!)c>l~~r=|z^bG}E25+~fov4|to0lpwfX`s_|
zWFF^xSGg*0SjT?0QkC=I;JQry5BPOwbgnri7rg>1KUx+WWrANO-gGK*h{Ys>&pJbO
zzAWISq+l^ogFOwTwyB;)Tu;FHuLn}?N4NKWP@;L$)yP;%>EIat6Iwd-DhMIb=0gd>
z>b?Mr4)X(<M7~3rNQwM_6J~@G<c*MupE7FpJZCB$cGyWB_Jj;dN{oo_5Y829xq8j|
zKnvO$^~bC^O$aMfQy!x=mm4Pha?Dg%>Zr?c$%N9I7wLGMjQrk224Zl8IS*o9ws>fw
zc0U8tw~1}XNDk<bP~3i<Sd3*Q+RUs!W3sy{{w6~|>X70N$7eYi8biQ?s(vJE((fvr
z>Ns`}t<k||>!6*-rj7ucvlT(jacOqW#*XKSmc|Sv-uayJH!_OHh}9~xaBw!H{s5TY
zfIHz20vfT!3ovgA7`!vYO2^i;{B+r3(NS*tbK+V=4_L&C>=?A?uz<TZFXps|@t*4x
z^wUE-pD2#<zsx6))5!8MT*P|y83C+Pjw1D$5P>fKze;vVHIqqI<7s_Ov-J5!&}W)5
z6eyg)3s5gyBsSd;=jA3Z$1-lf>-@HtzMKOHXVlO4=J5K`w2PaBHF$k`l4DUB>VXaF
zbRl@!Y2*pi=+Oc2SQtAVG1f~_3y`F2=~cW~*Xk^Utg%=1Nxok;EMtZl%;coD?7RKF
z$vx(qiCH}t{gr)2b<N3%({li2(VNXj<jRaq3>^u$bk}sWc`Tkq<fw3Su5v<~-RUC^
z%9ZZ)*9}`D9}~yj_f*?sD)e$k#tF>2_(Nn{jdkAKr1&`37oK405Si@;tMJu4-Bv`3
zr?T-Xcm-FS(H56Ke&5=c_~wj=8B9*`zTIK7!i4R_nkt75U2DaQFcbyH3Ky9_9=tPR
zmXvr;aDWwy^>$=-6@D}3)xi;u6k1UJ+twMKg+#IeF~u1CQgTsYpO$AhlrNDq4jjEl
zEBn{r&r&|lR-<~u%?!wT7L{VK+?u98qKFWgn)Ekd^_OKz$q5wtzOj@khuDhTzJ{N~
zu%m2Iv-7CN@BtSL3XLQaTQyp#jEUzZ5+m}8dCyn8mXcuN1)PA}8QO{&zA|4?{EGS}
zv)C&}bnjS|PK>~tPP_+ApxOzSK(to9Lw>oRZu&Yt<tr)n1a0XA2aA~xF`Ge<k3SW(
zc#IyWRHC^}S-vt<>3oXFUk|TboF1zI4h)QXKtrw(L8;hDv=>P!NJW=R<hptj8&@ja
z!fXB+XFB;}>*x$l^lUXoi__tCB{F-HT&{;%V(dUIucj)^3SX%a&RD1s4)yk~nZ#;<
zZr7IzYa3&!q}YN+E(5A;zkuF2v>}h|s;VU&sKqA+zt*h&O_J{-Hkx`ph=re$&=u$4
zP%P-O83UrK6tv+xKuZtsPw}&PSxTU{07rr5uFoDrC4+lvD&K<eF;O4m?ASR#Z?5B6
zX26>o=7KgV!2%>Zjbv<g-h^Lqc<68LC>xuHWB}c8!g2+eloKx4S@y5;@UtY!ZbH=-
zS~7pmg-SAhLE9SMD^ADtitTSa+E0DpOy_u3iy_t58E7Z4ZdH9fFcpprw#P{vq8dft
z5=TUSBl#`fj^%;?H3RMiUNfQX$6OwUo!%5vIGHu2E=0WKmkE|p=c#e_iFbDBx{H)v
z&ys^=xR#L#Jig^lhC@;Cy&qXrk}4h1(}x}klOB9ySgFi-@^=XBV>qo<+avtHrU+)b
zo%&Y45A_gx!+3=mF|qa7S$on->39Hli=RA|rYRCsvz}gKO6cl}CI+lJDW_v!k?dKZ
zW&gy`PE%$!kP|5?H<umvia4xq&NqgZJZ`BU@2*Hj(Ogf(*TRzNd+bC!1UR@oQHzlk
z82pP7#72iT3mKCcNHyHiqrno<^(hR@f!5O}EN#yK(ekaU=0kJesBb0F5;O2+zoC0W
z!cbJqBbUYA){8)OHwwYhisY?fBU*FHe-m5l-x(!Qr)Iz`qNfq8@so*^rtTuy`+yh9
z3fG^fsUMJxGX9(rIHk4I8heT9g^LF-T6m7ti*ZZ)R*yprx3^`Dez}-$a{J}5%$y!Z
zau?k3tnDWsW2Ue!+lq1`KI;yj2%ep4P9E}wd2ExUxgKqD2mJ}Q$~Ryw*Hs@ENwID5
zqK6%S;Jm^GZ<~m7I-Q>ZF}`h-kewL6PVi!<k(%E1s{r|SM`=0%uW6FwZXvL&7=!8+
z`-%+I!<FUJBt7aP;-Cn&SNJ)*H&-%nrVQttUhCs$nN@ZZ2kBWn8)UTl4WXmHh~ma4
z{r#u1bf{Ey*0Q^Ld)SWNOQz%2K^~z(h>win;Jp_)G7nEFXm`4;5Gh;^O|Kg=`91al
zZq?}gz`uiL>Xah_R|AVU8v6FtwrY^oJk1%h>BdKMqU1S@WC<M^bErq^U0czECsH|t
zVtnzSTSiA27j5=r-7<nd37kKaS>IM=w54bHnrvNhoZE}pB9lTlngSn!O3bJCoP@RN
zOIAM==Y7Lha|K;u!44jBJcjBpZtct1u@94A=*Kagk2@c-uEKGEMDi-iyX-2u&#cg?
z@e^t|;KUjfQAs<r`sTfnExojVq9Y8zed%X`)WSokkMcXL-jXA{RL)H|kCJ)us{88t
zEtaCZ!@z*&HZ@lTqFdw3#ZpEFz4u90<2ecJvn<b(RHfWWNr6KM&-dOu@=Lvl*T9LP
z>kn*2Zery!sV(Uf*cVUP372}gt@W0E*Ul-#juqtTEh{a5U$nCQ<t*T%>~~I4+cPZH
zpTbt)mmmhJ*^-h3Q0(L4T1p-ghH89c&V%2}2n65<a(L}&`9iyA8m_k=c6XaXCkzhG
zFvwfUV-}oGvf&n?WQ0j_-iOC(Ym;Q6OLjd@j3kBpOeSj|n%_+8IG59fMfP9ae~?cA
zkfbTN_V&wIzD(=hAoa~r#&G@4<BTpEA&{_M?XcC+qr)#7_VQ;&wcFK7>BobgU*6$B
z<%hvp5eBBnSz}Q(QQ<~EAhCviiV^ad<l*h%rr-<CfVi!1l{uoVxxV-IwLBNP1lJB5
zyu64(8=K+u@;s$eJMi*KYb!;BF@rsq?+kOBv1?Cx9zBRZku#-IoX`d`u^DeI5|%D5
zS`=xXO!bL^Fr8UnTvxK$2boad7aeq65e3&pIgTh7zOuwS2!|*{(4NERSEBaPC@nu(
zAfpj!X`<u}H=Ftbt3Jd-ccQPxRzZlpn3^Z3AH>Vr9J9ji$NI8UnY~q}8pR@b^G9NK
zM5*H7{q#bOjI5n1gwL4?0Jj+eBz>>77LDMo!FPcKl_|v$;Yb*S%_`=Gye8z1Yso+(
z%vD0IoT3gib#*c#xRKE7V$N29!hO0~w1G1^Le8{PvuYL90B=rHF-+8e(IG@D^Ky&I
zzIjqCc&fkwkF0rw?#Z!$nu1^!5<#Eny!elj?nv<4%(};VEOEkesv|mSa5!QcN0?_H
zd|U;0k8yZ#;k~z#W_^lq(PF>K<m{3lPI|QdOlV?uEL}_EdxumD5l~o=4SGjm)DTNh
zy{^|sUrxL=RSx)CaBY!tA5(`>QQQj}Ame$s%t5i49LrlNlZA9VR-q7FR_!HPs_PxI
z&S$9$8cj_uAezlP+`z*jp!j2Okf!+dzinHKh#7MPpO>CTj%&XuLvJ5gWL~TCGB`6P
zY$dAHi#BF|=4nCIJ}Ul<$V6bOk*bwSJS}r>86F_?C52WBC~~d#ZTQO1b%YRzlpg2u
zeEtU?njCeg1r8prN&JiY9Yz0nHASLsiEBJRju$H10OOSUoJy#Sj*36WTO;^5INgY2
znZ0nq%2jjud%#r%TkR^s2h@hp8mO9FM$u#af)<>*leB5!c9YwaeqSDLMptx+UcVva
zTwvjOL#Q0R?e^1z8Tt0mr*9uezAD&!a^{8Nr%Z5&kRaOR)`nW4csYg8HYd`&n$Y)9
zFXrd}sM^Yb#hl`Sy9_eMl*%37R>!fpnd977xi7N3MFrO$htjBE(;;dPowYocw!3NW
zsrSdcO9RuBHcxQEr3jgY#|&7~Mbc5wL(ULM3tDo&XnW2t@{d+h$I&$d`=&9`g;A9p
zR);iY!M^Z%8Q^^>^9}vxYWdyD`^itmGTj*39As&+FLwj{71yiTBeI0zjY>)&hd`ZB
zD$);tv&?%QBQ!D){FPB{Mk^>YOES1Tp-h$n&B@OrGXCQ?r|ngA-gy2U^6a?y4*}UF
zAsWhf0eTAL7(oex+R|M0PHEC&yzrJexjDSI%C9SGu#FZq9yxO|S(6bUcZ>Oo3Od-y
zc-bPzOK&}8mEaam-#YMc;f>KDZJUHd2U<d+L9?(EFufgy(_|03ON0r-t0<BNp))EF
zjx`$=yCW$=V<mQSB8Cz$6_ip?(~qjsQ7g0=^}~R2+c_gejGaM{v{g3B9)2^TtQel)
zhR@1_Ezs1KQt#KBUwrwvQcMYagC_bvn|GWJ7djD%4(^&>QmRasnPFRBt3m*20UnVb
z<s8t+WDOCtnw|hpUpTHXMz^Ue2-N&@Emn9IB@tKMWk0_8tYaKLC1f0C)Ptd%t9Q))
z<Ah%>ONRN(8mB*-7jXzVEgC`Aly??$83TKNV;n8`mvI=-qd9}9b+Thou?btLV?WY!
zdxKMuBAbBD`HSACIg;6NmZKrelMuNbuf>)URl%DV#vCMezpj08sD!KpDqiC?5h7u{
z4)Ry%Qhb)elSM+zIxf4|((<Y!G%ZVFfS!X0US#;|&ymz;#3T6pxhp!yN@Cu<FNih;
ze!Zu^f28{G_3E0Oqb*iO=YR$9n4!A98XJLjNlEQQGepl#O1_`hLOvY_utIxYtHq#|
z=@n^%-mRD?RIx!jfcqkNqY|REq1@CsHdbtqvK%2O1wZd_GW8iEP+d2~NhyoI)yOwU
zzp5>RAZE?Wn@W}MsDj^3tVWLD<Nk(p%UrYW;72!OaivKg|4iNPl^2Agr1bquoJre2
z{Q!di_<DR#Pb9%iD<tsZ(zV1bMBQ<b_Aa|B?l^SOjNSN{t>FSWY4p_6;((lQm1gHH
zV+dMt2`5>%kv4H3UWH86z4YP?VtYGBfNyILFukgIp=I3W;_^kx^fM%@zT$bkG_w(P
zh^Vfl5PQ#Vu$?(I`ovi)eh*mOz^$xwV}(&(G4F!zD@TW+7Ki>4*b--#(FSjsjy})~
zRbnDhL0%)082jsdLi+v7mT(~#g^MnF)BG@srOv>dK}MauiXfK{k485+!EfIvmn^r$
zdo>-cRRm})!@pXX^vEgAqS)CP`-oxgK<C})Fv^3aVWMl9hRn)EjBGu%Vfm!7*@qCt
zu#ezpGn@H(I?N}^CkVO<9kn<*8U!|uXQT;xV0{;C5gbpes~PC%6j6~Da7#m%xMmv3
z7oTCR1XJ8=$Pk$$eb>EQz%#hxLUS9MZ&_T4fo(O%>^DWvKWoQ!IyF-}ITpWgb^lR_
z&C2q#FMWmF?Iv#F<B`*?t{zS1sA!Z+>D7B=W{0hO{llGNAq0OPti0=y_K&CZDCWFl
ztTIX4>5R|vl^3^B7Fr(f1@-xq1ms&=nQJ(d!#@alX-29VMSvmYa9h?m9(%`dc2eL=
zr$;7cZL?doYD8WAY}6R_Pz^2$;RSPX%Zm|%Q{?ZHEZ%)`w+XNMYMJO_hwZ+5U&TyC
z$u>M$3TZPBQB@|7qRV&bP?A|CW;rM~SGIrow)?D2$gB0aBX&=`J+eyQZtfh?8xcBW
zvmAN^e8T!mCy|!Gq|=mO&NSMk5ikE&@u4xO`$X>B^rvMSibx#pcBViGBuV$3i9c~$
z(B3ONoE)cy+t8RKO<w2y6Iq6^x_L*)iyPK#$XWR7AR{)_{OO<c2^GzkTcW>2yh_cK
zd{8|3v{EV!99Rd2Y^2?apJSz>!wH0>jlqv&qA-sFA$B=>cwYf3I{8-g@a3x@BNxGF
z2VHLWBKRBxek1|eLrnHKX=oTK>6Tkr4NfZRj(OKD4;im+OKsTXdx-hz^S(XwN4-yg
z^J<7+oo#p90qYdXik)8BFhhrSBCui=ui!QS$3h{V<yDE<!Q73`Tay5chtmjo<<n2R
zbp+Z=qTsvZ^ZpY2AD^mG{phlq-^bLwwCMXP%SV~_^%)0+S|Z(i50l84>X|>UD%>QL
z8j!XumdVhj=cjT-g<4>`xDqeEJAm&SYyHk2p<Zm-%h(kOUX6TyAAi34_~k4#Dra&5
zk3!F#apn!T(^SX<p|-ixkFyubZ>RfQ2Ild4Kj*8-u%rQRzt%m9hnBm@Ag!b+2l8n>
z8rIi}RU#;Mn`P@0<yJ=RkS0<h(Y5N0fgP-(!^s9`Y@P)BFRGY7N3?#zL8wHyL-02V
z>R9pRd3cL_Nvk91S|lfo8QXkfGW?@l*QX~-PeP^0oajq-z^7PZ4h0zoAS!n;weVaq
zmQ1xDALW`~OVAtqZ)_L=->Fopzo=t{zn_ePY*9{;v|g=GGmS?fJMS9P(Gh!tWokzB
zDh|h|{G&4Jd@?b%&6q|KK)d+9YBeA1Vn0wH$7ah{@>Kt5)h~Pm5NZ`g7{=)aLQXw#
znAG8az9-`OT1=wBTmBN$>{CRl0$NlN^~;X|)VHd>Si3BA<(XAx`QzY@upy72ATmiy
z%+x;kGMt8@=NW;P8ACFLi+jY$FBGIhh5YB~U)LRf+KfF()9bfCRtrrMin+pj5h?<Y
z4d&rup*;e%9JZo3yjs|w2Tzt;>v41tHwfRS*Bw`m#+S1M?UQxg|7c+{$i$Jr|F!-2
z5sbUe%SNm{VC!>*%NYd~2zg7pkDR92S$Wkd_Oizz7e!vWEPtxuNlS7ku*lB{L9#Y~
z_#H7>z|sUqwLJm=K4Z>mJM!!Z#X3(vX4|A~oI2>+#R&Yv+aLX*g#A3U4zVv5&V{%#
z9OY)H76T&f`_}``T$I0%wVaX2Z*43LOX%lmv5y)^W3$JuXgJxRXIXDnB4%S7@Xh%+
zxIx}VSLu$!&4t1l*JI=8(5!w{P*gm=sr+y_&e>8g!<IF}*W!m$vcHm{tAv|U5m>;+
zC`KMp0CXrZPkCxzNZQn8m-b1FBa~;5zEo>||Ejf(8+ZO4Tit$7>vFdo1qg3m#<8%s
z>E1(`DOEFQG%|9D-V}}2nM$u#wYa|Q%(H?yk53Z5Z{clNK%}jeh@c@Mqf=Lpv9s%Z
zWosY13Lr}cx7|LG3lLrW2?dwJ9;<s}k&7~3s0?41jyroAF=|4;SX$rJ9{*x!em&HG
zC<IaM<d_vl3n=vMhYV!=vA#{d`^*?{B^{MW)%e5FT+tgkR(_5i6pg{gh@Op8F-W<M
z^JNV*nerYX<n_y9X8#unQ9~G7G#SbyZPBj7u^2%MWv$6Tk(6m`i}Ze?qCjIl*MJoK
zf^hlrLu(TLrxDaW>)6KI>rS?dN?o|Rm!n}H<Kp+@KRM?eH_N`Svtegf<SOBN0TwBJ
z(YsZ9k}BJFbW8iBE#5?quRVVVeX$T_?GHp(hcKzy)rXi6!gcCn5c%<_i&rtBHkL}6
z$fNQBEXf562VIRSuzM;8gcnP{msFzjAwr|r!PGKDOMt2=^wO(Bnh`oy3-TAtUTIwF
z;&TcgUomsF?wC2zxfY*%l6@z#(VU~jlmO1t!EEyPW_(CJWkHwV{2bbBP0<aRD<Ucl
z>l|w>$?npin$UT(HSZ(Fhi?^6@}_J|<jqW%s*CapjU94NcMEO}wwUW0KGgpKaV6Z9
z4IN^Ul8%C+zx-*Ay;L-&pTFaptW##ef4w6!j|&&$M5*COp|HxrQ3~~0csM5F>Gol4
z69|BW<etPlNwo-JO9VSBB<|Mt&)FX@wCH;lW$B+W`{aW*%0C6Ia>9WY^JmK50i(**
zltu*-qv!`RmfLbwtQkhLCM)ZxeUB?^ka$o0P=SK0GMlApZlBFg`NE9!1&*~f;ohTY
zQ9E7tuKrlJWs%T-_AW)G57*)cLujE`cgZ}>L|?%74^_XqGQZB!E#*i|R)F|jfVQ86
z;HP9lD}v9?aOo$~<HdQZp$s=(t69VIeDqHDaNC_Yt6@Ixs*c5;g=a<<F4S@jrN6gd
zbgAk^2g_)nTGoqs@V=r@?g?R{4DMV7DyvH16cZlzPk{}duPPG15NyMD6fz1a&A2Yi
z5egQrjD^mjD}W(R#HDtWz}Rc|2K9%_`o#8$tjuaR^FF3)gc0^?6?UYboraO|)!vA^
z@&@Ld0}ttWB+0|Mb7t+&_ay+o2INhO)mL>Mzn!MKp8r5D+%V>H?HEii$sobV!p)y9
zccVQk81Y+Gw14P>&Z76#o*6Or^?CSyIZDWA-7>gZQl!C9LFjXT_fF58G0_F8=j-_V
ziFkHl!<2uf#c55|wE!pkU|KG=e&bhnnvx3rdFwYUk6MvpHuAMYJ;ps;T!bnj7i|<C
z4+6OQj_N_?DapCsdpqzV=jzA;{MX`n7cEDL-XRX?n$=<_1(Gk@b60SI#c{*Nkr)U5
zRXz3YTrZC2V-k+9DqzDivT&QZk6ztff%nWfu88l7y<D{8-2C-gA$st{1rL8Jp?D!}
z*hKN_)%B9+cpbyO)59l)U=br`(%HrL#{r4_bPM8@r6hTHz6iyT{%uI0pKq9mYEv(E
zVM*`vTBe9rQ=$8Y?W#r>ore6IpaB_FQuR?qpU$Y(Be74+2aK;`x%9_Vg3pvu;9^5i
z{m&r%^{f_jz$MBqBl7eloGL<$&y~%u<Z@?tGM#P|?5(4ZC2(Jq-Y0&<I5HgoQesVn
z50s|44{>888R^oyI*^7BG52b9$c|#`0aRA_-p+Qu=V$|+zbVL?Oi*>=shx^sgF9ga
z#FWgOvPj}Ndsg2#kGoji@EBe%0T1@hlmpVkCc;S~zM7!IE!ZYqT7(sKp`R}<J{mSG
zrM^Bv*{UD?(Y@jF11yTwuhaPb#D%Ogp->J0?^RC7B!*D_tsztrv(cwChp|)4x^#Ud
z^&`vq8rhs}PC&Uio$*-@D13pcHjjCZ&YZxXg2owQl(i6Rd*5Xe{0^+1<S-GpyZJ_X
z?6L4E6NAH%J<-!l)Tg_Xg4%z(kVn6YCWD%_ssf$ULN8~mH0GM%8M!r6uYq4i?44|5
z4>oTZCVGY>+K&2Tzs+<?a5S1H#r+0VvwzqC{iSclu3)|6UNlzp?1GS>kM+st^4fRi
z;BRr=b&`wK=b-s_lna}HUp}tSS<h;`LXRuk<YW|4oRNnis#p{iIRX5bzA6#Lo$oX_
z9CFnZKwVE_j8svM)j{a0JqOfm&e;JU@kxSZ^+6`#F@5iw*NohQaMD41t4o`tjk$6B
zKv{9|N=F>6PenLEr;$$b6XQl<lJ{lwXn{IU%2)xe*cz=l-0z&h*;e?HWEuQIf_uEQ
z-x>fwYO7>!a+=l1=)3G{MK{TVsuSf8hLSRU!p_QJ9VM!bs#Zu+6Va`sZ7P#$?pkJu
zj$6FMA2W{9I<zm$uX(FFkwvQADNz<WzCI%a7pQ6k{&?%yTOi}^lsl}_HXdhQ%0_6=
zff}5p-ueRgYXLs2^ohYA7asnt+ZWCA>$=T|@o$x~`j3GMwhGX?SDjy>vj?AS1wT1S
zTP{9&GCYvxQOn!wt1hup3fLa|2HKIDz3{O!HKWzV5puj(=0y9na8F-8B3n6S?t4N{
zIWFR~7tq1qG@EoShoc*)XPoB=!8dD)B$CuPcwZgPZYiVF%mhYwtqPg)_necuF)VnV
zGk7zJ)RJ7bA_s>W2EVKg|DGizuEwBi6rYEx_2T5~sewY_QMYJli!X;Zst#K2H#H%A
zf>5Y37NhH@i;{9`QIYTG3Nhn|$A0>j5wd$_&7akKnH0M2b%{&vUoXRpVNvYYs+Fy0
z$~!0c4G_FcBS(FNa=r6Q2d#F)%aZrRxf(7CQ5W~~!OIaq8#Hg6tUqXQx}P|@7^89t
zlosnAO=TXlJXhtAk&WKv44^r6%2T4XDhRxIlSe_Wyt>npoYm4AwP|p&yEdlrKte#N
z@ttroUUyT>1CN`v!*gzM>~o>q(ma&id{jkF`Ll-0#5V`A&iCJ|1$AVNs%H-?Pg**j
z6x<+!9#w_f*(#h9{xeOYHiJ3}B<Vd@4XZXicC%yw)KFrveoJW3D{&&8%9qwY0`9#g
z%Wsqk7Gi!Zcdeuaz+FZy8#FCumRNjE*}Tw37kQhL@me1Rkwpx<K1%q3qu8|9S^lso
zKz_hJ#oi>KZt&#I)Pl{k)?26KpGw>yj9!|wyxy-UoxLkG4`<#Xy!x-5pZ90m=_k2~
zHM*NEhYkAg+-YezJ()U9vYm(o*bffgkc#&4QRF*uInhP*z;zT762Si|@51Vnz6}v(
zkxCEwDQ>ymd&k66!0%4$!dul`u^}TZrP>w7@Zm_<oW4b`mX!vlP95_4i0*tV8Tyl*
z;Oc&6`{DKM<7_hVxx!HAFO*-x64wTc&BbtV@DYyk@>(kL^8dOn3%ewn_bNqFc~FLG
z*y2+hfH?Z)$K5wYV#avH$Cdgla*<l>Ehm;<j}gq+Dsk~8+b=wzpvIH!-J0s~YC5#n
zq3xX_tewiSzn)`oT1$wWk9Gwmclm{Wtl?Y4QiFrokssU0O?@#K-ZVtQlST8ynw(lV
zWMoC2z4XPUwYadBnEbh30HU+tA>Ss9wa<1l8loKH&&`+#C=i*N@XEi<^ItE-B#`zp
z^Sn#+9=8J9M<uES=8LFtWON!GTN?=J?K<u{m@l$ejCJN6b`KlZE*byKe~9c4d*3wW
zo5P@NjsN@&ot!K+v7QyZVzpvitiahbi3j{q?v&^BhR6w2n3b}TUvK+=ZTq#^c{<Ij
zV=#y2lde6HAqV<_4F=!$CcAd|BHR{$*=zUe1;Mw^juVqeBoR>B!11R^<zL}5#rr;{
zaZxrg+&N+l1E8-Jx6%>i+GKv^teFJSR^ZA+b*gT_p+@%^#WpQVZf-1Zto7QjK0d;|
zFZUXl362hgU2FDtgk4`YQdbiLxw~>%*|=NVa{0S@z^*sL!AVH_dsu;-ZGGshZS5W1
zBmoy)y#P8#8%cnnpgNDbhrF$WqjI2^t!|)(9w^WmBx(bYmco$m7lQ$~+WJ`0`MbKf
zd5if=0{-BN!QTIV%nhLXW8&j12{2OEqLX*`vZWK?65!(DRPc9v$p?_app)>ju@ln)
zD*gij`y>f)@bU2w<L37B^W*a4=W_S5=jIg^73JpP<L2YzgjsNU2e|oI`E$B?GyI15
zivhIt26;Jp_&B<|(f!7>vUd0Nkpuu><8=S>&(%X+{on9z-v6)w<AdAZ%7dGii-+6Q
zmHR(uc>5^4gn|6y(EoLYw;t>kIJb_ix4W+w$X4N{t(y<Se?r)R{ypEr*URNkI5r?|
zTNhhbn5j2xRo?%$q>_rd*1u=`roi6O)#J}B7}@{L(#O&6zsUM;zWpBg6V87+0-OGC
z-2Z0%uh{<>!>rWR#enW0-{0=303`vx*B7&K2RYh^{dp^3EhJ#YZ!5ydFJdLYDFCt+
z<+K*Gv*r{L2Ju^q^4eMn@bLZzl!}|TkChw9_BRv^oXZi0Bg!MhZz~`u!YRrx!pA9K
z1+w9^23hlSiV6$yTJiDP3fX~d{sThO%MqqZE0_QD>Nk`P49ZrBM-XJqBLMTp#)?yb
z$6A<EM8HmvQ$R$BA0%R9BVZ-U_Xo-bB&O)@<!S{>r=zQty)CzgoBf}G--L_FYN<#9
z__%of>xq_&m5&{4f+RrA(aqQYzXtRiU2S!JtbVh}D<r}v$S*1=EF{D$#3v~DUq%MD
zUfwVz{>J3x;o|@6!tb<*!GeJiYxP^FFn~WZV6lkFd)Zp~xO?fjySqpNe!E2Xd*;7~
z)nVmiW94H7wDPfqf%5S2i}47E@d@ei@rm&WiSY`t^T6KxhrPRvqg}xNoAvMFp_BNV
za%D$v*!lr~9{sH-U0ct;pZ<Pyar{$Cbaa1;Ld**EHw161m$rYC1;hF~1ah!)v$utH
zkALL)ujh{c7b#d-*$CSS+46B(iSpZW3JCC7b6VS2fjIeu`9VUWLV|X7pue>K8{OO8
z&d1Nn%U0GN#v_a?m^}Y*MaS|dQLKMB`?s~9gYEAuzzE}niRAwzjQgK}xqtVJe?=_8
z{eN&G@yFmlnhb2--)FGq1?z>}|89o=a0c_>|IfdFa`FG?5_EL`yU72D-~XlSf9d)k
zG4MY!{$F+dFJ1p52L4CJ|EsS5&*;MVudh?KZm?63AMDGL|IsP}?28t<hqAFZ92_R$
z?-x8=P97QTA*zpxx&rDU5(N<oo|!kt1soh5oC;7@&wur6k6)gRy>G>@yqitPiAa&X
zVu?18Kv|9eK#`JQo0|EZVTEIS#B;e}<g)qw@@Y$ypwOn}FA~aTb8bvBHKT;f94uze
z{(TmD^oRkL1m+(;bsiT%*U2|MMUd*wpyG1-8HTu=I@80=&YR$^G^9Zuti2XR{n1cz
zOYsKml^0zu+KY~sZ05}1Sby^5Ox{9d0q0Bs^HhJ&&>_swk$BWA|MA^2lP;f7p#}o(
z$;&1r4up6>6WkP{ck3j+C5Z?Jz;$gR%8IOl+nSjNYp~q{*hgcD>%rAPxl#)`ggb=m
z3Vqj?<ex1zINgXw6_**_2XEOM^A+_xKJ!&y55TMbOcFMmnG6edZLo_>1h<{69j{$k
zpBpZ2GT<^bt1Ky&s{NVyIl?(XN>`{jxTlXA;FX*`4qFN-+=^`Ekrj<O6cr&8J`-|M
z*2mO`xNEy!<I~%Ouuj9>7CUwWzCdh4*dk1^<xjC9i$X^k<Vhf;$KryogR9e8oCHsT
zQSpmWN4^~3Y<Kl>0=zIi_&u6q*6qr9VgB@6l11^Vkx9U{Q?$R(uScMv;{Ig!t~m~Q
zTfEN?xUaa4vW<e@v7rX~2e?_DtR*xWVHtAL+9x0CjD|cC#Tysi*9LZv9TNt-BW&TN
zBxR5AU=d=$BF;6WN<d78qbY*mBGS+dhw-11J%hsyM!e0538ubfgX-fO^L`?WvI^^m
zW$Zf<8qBH;l8it@JRC;7m!R88skj5R#lQ3Fe(KB~{s*(doa!qQO$lTfJ{je8D9^$a
zkA|=vm>b@LD1{sD7Y6`!nj^|8%M!+&NF%}y-*zM9USyj$;yYk)r6!mYBYGCTfHcKq
zn(<#oVqgzZD3bmw1gO89xbC2cZU^#hO-wM#t(waZr+<7aVaDS^$*+WHBIP?E^z*>?
zS19a%y1d8XkNulfku{_N0q7tkP|`y(!`g<-`BHE@4REEhO$cbRWQNiFJB{|m%KnT{
z8~s-VP{u+l1Kxy^a@1f`pcKwST1@b;n)P3Tg{oy%5G3^V3os;*+ox)*$)4qzfRkg#
zl)%YxW8K+9e_6<z0qkIGJ3{USwjE_Y!K^A0xS;s>cgo!#KK~0z12qLorvFL|%FyR)
zhd(IddW!(H!oMqwd1%gtRmkrI<*BAXiS%EgKz+7Vn`5l<uGf)*wSZD|09t<Uzo93j
z(G%p?L1vNikxgAlcoWnosr!Bc_a6aQvi#Bi1W%I!<<|T0xvv?FISq*N9BCxIRl>i6
z{lEFwIxav)(|X7(d_JP7H4LOVYArMbFC`>~pBAP-H6F4k?Ks#^GfuvR9wHd#Dq!u^
z9HMOZ0IqI?PGQ1;IqG>G7l8^lYq)rbm@FI*6KvRQ+*nZPP0ZMt9nK#Ok7RQa<n$$1
zog^d7{<iQ;?^Xgn;rO@3_)Js%5m*Ah5{rbaz6D%2+zN<<7i&+XgY<R_aMi1@&;Kuw
zFGT2v2;6#$A?Ok_9M)tbD!bl*E3WN#fTr20e}a9C97wJ_!32{c5W-01(ZHY77U<Rm
zO1XUbhf;HZ6)x0T{}nHk1x6{^4Zj6RN*kg(X-d$)9Qmoq0%L<4-W)+KM*$|&X(NOu
z1K`t=o%mk~XuMM<Lws$jZ-HUz4s#Tg235fipoALWJLZc1mC`eF7Sk$-G|q#FN2Bo&
zPlEacHOvcrpc@}7ZNKY3$$~{fLEi%9A;K8e4EPgz@l)Dj#zv=z{!&pwiXmMDkw%yJ
zTmx$_<TKK-W4Q|Z?*AHYRT$9Ney0DUBGh>b08@!=unBMzsoNNclk<-QZ?)Pb;vx!A
zW(0q8ozaYYYXP`o#_t&ZOVgeU35IkbL>l`6>mwh`!s`1R{3Z+F*K^<po4-7H3ZX~R
z8Y~vj`$+J?G_Z773%F9mzl+Eo_^pSLY!0G)n1h%STQCQsJk@?Dvmz$APwh_@_S6V4
zX7v{f5c3_Ee%CpDw*~O&1#r{3{7()Vvtc=cob)2R9{Mv|=e8l{0T=&HoImn+V@Q~0
zCVCQt3N*tS)ClVi1K`S#nd%R(dFOIF9gzEPeZ4%+FsIkKgO!0#U1e<Kzd04zh#274
z|8daQJVo()D?q*=Oc9P5`#JrW({s%tD5L%>WGKik@#i62xqKb=AdgeG3Ghh=DCPFo
zF;BM00bgyFhf-Klut2)dFDz6aEdLqeiyU|hxyS75h0inI|1J`G9McN}K!Nt1D(rIu
z_nBeU_S<(tR-&0ixgFBbGrJUBZnPxsjj`0XSc69(lG}I|s_X2(`JqbIQU)7BPT0Yd
zQ-GU`e%0s)$&;5bC1870$2^Ea4*{wFvaB`okA*w|pk&P>kI(R;6v)+xV1$R=+2aSi
zi+QO1Cm)It18K!t{&mKC-wE;L-)B)zz-ZU*7REq5t+e^Qv4JYJ+U4{^VzoZH<T0Z%
znFW==hs6P^tqFce556MFhJ_p6Z4P|Ojahg38!#A>gxE_pJV4VtR{q)Rz)B4t=$lxw
zs#qE}jy}k9-);zeqJ;j6AB%(a5ZE4NUvK1}0(rvlcn5PqQaKS~f=8^k3s!@V#YbI)
zR(!f;w72f#NdO&S0<P=<S2(a#!#0%}t|I6W3z50Y(h3k#l;Opw&B~&>G>vyShLp-!
zME@p;DfY&YRb{6`>>eAYLdDB#SVkPm|4NXP$gWuHR^M`h!|UELEM+4_58w0h#U)F?
zCd2WyhgoDKa+ZUfcb?mc#~z`GK9HC=IwlGZi9jI&9T`@DISSjFB_1NMw(v*K&q!be
z9YX9M0drr94EtJA2R=B4MC6$Rj_Wb+c@EV>bnhYmug7{%C7hHw(6W5p0cyzQGBndX
zt{{s}w$J%4j2RzBIef4_@JSE2>GY3}^h7k|kps`CDRRGC5+NfNvxEg|&a&TpID8j;
zF#D8_dL;7tWs;AC?FZI%{?FUPA@`X+^&fH=U3b+0uq=D|n-Jkx@amG;^UbJP=Jm@J
zl|g#TAhdjq4WHX=YZUNMF-{_vyhcd3pwjJn)W^F3T~Js{X14jbNQh%L_rniL2fGgj
zgLH@kKopy}+5#BlLbsm2G|{XeCrq!f1PEms`5Q>Y^rEadl8C+{KPbnIN3X=4=_M>`
z{)Kl8mT!Mc6FWK9#5IFnw@mLCbpByS@u<Z!dn5omU14^RFpUEPR-?R)y80G`v!b)G
z?go?E-gi<AQvZuzdiuZc+c!D9?9gJZuX7YfrTs*YZQ^a3B9$DgpOvUQdwvXPS>%21
zEvy`-%!HgCSkL(TTr}|ma-V!j<pch!tBMLE|7^%>xLe}p5$4S#EU{Y!{Fd!ht=>?U
z@tL!)o#%~H<#K#d!#Wj8I6Q*=Qoa)3jc3Cfc?Ix+rR6dU*66VIsP<YxAJ*--`E~!g
zhmT<xaIsYhaYv!CCixtUQ1o=sKM^)RZm0kv5C5iWD3ScL_E+CLBG+C<B!kTwvylta
zdV8wp(>IH;JLzjcXDqu#Z6GaqyAUklI}$+sEvi{{!)(!n>7;kOiPKyp=5>%>98;&I
zl?PM?YR^!eI8iM6US+Lr78`Omx7dT(;4t@B4<G?b&brsv$#cQqeSLJw=3ESP3x6`k
z4#Uo%BDTS>ewh~2-LLkir|7a1(Ue1a!Rk&@47iG7K;?)T2nDl5iL8h+4u&~$rpOq*
zZn87l4C1aFr2)&YUR95#eHfP-tguZ@8b75zm-WvceKkD54$|8JG>>F^;;kNahP)~?
z$$Fx)YjPw&3UweUmksjRg+gLN6^LOjsJHVgQa-K!bs&v_hy&u0fgbf^{Xtbm`RtNL
z0qrcM<s!dT8vMIQ4;Ewp>A>jLI|q>a+&;u93j63#qf>auBLRJjD)#blt<HO^T0^-k
zt=}jAotO9lwTnW3l5q=btFVrnZag?N_)#f6kd8>9`h<cQ<A<RDtgDPTdUE*d65$Ac
ze*dRT=i(zX6&fu$kG~3h01sx-fLNQ+Q3rX>{oE3YcR1}ZexKu9)L8N=Air>K6`G?w
z^`iOaj(*Z=+D5<*yQgg--Djzq)u4e>;C`~gKF!+u-)@s}lH?N)5A01-jLhZDpL(J^
z7w5ww#L3N?Rh)xLt=0XuIU>ct-rlM@afjct!v~ef!Hzwkv-{H4Cz?f~o!9mXz$dg@
z5dQJ!Ye85yVJ6X|C5{~U1_NxQ9x|xsBH8=QCfsKOJC7kA7JS+w4@{1ENQnvl{5}I#
zgCe=Z1GAMUy{KIL&SazBbd>oCbn@$Atp+sMgChLNIE%ds$Zy*l@Gdfn+~yp;UE1}5
z<_rM$mNQ4#elsIf<e`zk_7G=0D3X!Zi#3K+ZN`((tdHjMeMndGQwg1{NDRqn+77~=
z!0xPdi&d{{2#h07p_oY8B<;wGEWrdqPq<H@_GQ?l?O^_|+{6oa{t;So3;pnJ{8YkH
zBqi7d4Ykg{SF&k@Vrp~b*)Q=3z0R7~g;EXj%pgS;4ofIedpwrakxOWxf%anYQPkz)
z=FvJlDLgfautTgU*I~Ouo}-*qGUQ1JifM-}{*`Jg*5AJ@tnPji7;cp(I7+15Qlk(9
z=|`5`CI&Ln;OqkZgI%XUMLec8_A3ITNxgU}(nKjiuC%`;U>|1GE)|FTT$WFR|I#W$
zJmp!8Y_^@rsEDt4OCdyGE@hboK!lGPX{nVgf+~fEs?{x`7p(Ria&gm1o+L^eU%Zp{
zo)D(IglP_vr&h=ljz<w}vj=<Ze%Og^7_#IjfBkAA37-Kt@}kGd2HVgLdU<MFn%gJJ
z4>k9U#ml9gV;X!e7nT~N`AJYl@?>z6PQw$!T<$EaxIN_?xyDaPP<_k;>}1WpMhL69
z5j#SRHEX?{tsFtDy`wz8C65fv=<r@^rUPbgO@>K~%4B2ZVw->&FQvO)iwA<6%~#^w
z!B3dVdG<|<ay=t_hj~c!qhRmCILiCy>v|^V(xVQ8_-n#{knu2+6SN}_b6^>~&^V=N
zdw0jRdLsC7k-aD|75Ccd=PvIB=vm;b0I$wj8pa#@_LhEBsXGPp>rcHh-c&S!KUyCi
zyPlq5Ppjk=vdPj_j*64M;Cr4mODE6tvz!3;B5^LNe#P7>PdP8UPo&>^2N`m&NQgNS
z>xYV>xzahJ671cf<wcZvb*XBk!zodAwZ-=k@WX(Yf_u1!O^Jy;qS>?6(E~H$bn%yV
zSDjhQiruJt*S&LC?ZaZ5!p>q#=Bk?e#(4hox?OhR^{HP^KXK>Ay@S<F{ac7O-Wq)r
zOk&L)n1>x9g&`-HE(JUq-|;{4Xrvq6CXyLv2|1IETvBKGCY}~o$JdKW>TD3!E&k{b
zV^g~MP~Bpv?i-U|$}Ks?z2+c0pkm>4aWmulPPek|MgcWqaJ9bWKK``N)&A$WDR7KW
zGZoD|BIuCB|10i?(4{F+j(RG)0oon@f_om59CeCVd2dZN98aA3Fy-}ng=m))b@G1G
zn3O2#Mh1u{paCu16A&>#bMbhp|D%<e1Z!_4_&~e*dT{9pIq`8%5F_$rYjL94Ku#9L
zu&ny<#v<-e{}=J|y%*1Y=fOua9~oIb_mLj|G;xv&Fg>Hz6BSz*B#E7Uv$`dBCwy>L
zQm?Rdn;@7Xbd-M@bJ~g=TDP}dPc5jctry%qMD>)yzGpwEg4WHp%jlz|V|zNpsQHsr
z)2?xi`Q7@Ab$$QZ=2M76DYish8j`HmMGD{d=8O$nc`vAd^$UFq%Gv4vho-X%YlCar
zb!iI}3x(pv-J!Ts+}*vnI}`{6S}eG`ySux)I}~?^;4VAwx3B*!hsk8t$UV=j<#|eN
zomRw7(95M>eUXHt=3W3j1`)5sZTdaE$IqoNSm6`(b7^*G&VOjll@V_huRV^7aks*a
zoWy_3(E=Ar#1?_Y+*`_H9kl88^KUDojk6c0Q^zq^?Nd3N9e~}alPO4Ly@SlH_xvq$
zt6FBU`_vyXiDGsN?~3KbP$FR~pOZg#G#QKUSDyljg4Bm-X-Ghobgo*&^Q98w4haMJ
z!Yir}%!AKF;oEA>kL#cKE-eMS`w|Gk*aJkK&-als9|h9{S>;2bD689PUHh97elFEv
zl0hP;1AQ9EDjdr+{kypct1mvPz0!R{3aqY<$I+6YFVntGY$-M13SI6@DN~~j$a&pE
z5c`w0A6N=(3w*mNK|}-yIc)A2kH)_&OLroY0;noNRNYY~mBduLpB+-4+=qpn+xCn0
zi3wdh+U_QH^dj6cbe)`rhsnue!_vai+S=9@-_M-dc6S-6=qiSXDagoH6Z!|Amz|F%
zZ_(FvLXqZSd(0Cb2y|cUlywnPu@Nq};nF6kQMb>8kWnJZ=F^|M9sN5R#&_OJJTZ0K
zZszOeHr~skP<-!&{s4){>=1<~`aLJMU?<<uhMa+Q4&P?m_B-1EF>w0$0wG;n)5-N#
zc^m|js`H&SEpR&~X^Y6l{M8qwB~PtgzoFd}vS8CO-u9-2eqN=1vkP%Pj)RUT$e=|a
zd-vyvR0C-}LW2mNe0+=%F^VL6P}ru9)Y&->LsU*hwgX6l{LDhuk2YRuoTT)<yuMR|
z-;4lWK7@%i%35_@{<cxlNAhQ&*ZoX3@`<%}isaMA%wnfKgRk;rKnYkU=u`g_DLzk?
z-_44W*mm9@oM@Q5ho3p7IVzxQWbgcG{yL8PN?D`R{3=VdChg&tQ{I`a9|}N$a$D?~
zGnd1$wShu`8+}5TH+ei1hia*x8}`4(D{N9zw9q{2gI<aWvQeO+b{Gp$Cr7dB5i!sa
z!`E;ZrtxAc_u8OS|Mr=#GS^Qtuiu>kwE4s<mE5QNm{%$)>ta`0M+l4ks)M~T=t<k0
z-DLC-zfb`RV>aR5#a%wx^5pkwlvzK?YG>?N6!UC%Fv!@dbsu(So4jb2XqT8Dybo&#
z_7yv6mhc1SIV8gBVthgm)$W&Sl1ERI>2XcjjKN5xl#C$_hm%_Fs@HB3|7OYNVEM_C
zjYO`OLW(4Ms!=e6iUB<#FGf0KyHAZB8+3H}aoa%*s7gW|a>H_44=LJntJoFUBiG<h
zwo;=|&Zkaim}MDR%fZmi7PBhg#zm-nDPO$s5x8bgOYN&iz3!9H@85X2U(88~#X_rh
zh3&=d_lf8+T#uN87BD*|K>>8R&{p!lQ!dYClP4e~K^;yiopo%?FH{|IOdBbM=~&^y
zBI4bydat5FqJ)D%qY%V|8Bnw>jA`i=0Ui^nMDl=J<7jD<wfDf?u#Bx=**)WE{|zo%
zA%WZuAsq;Ypj6_8D9MQJ<T%sCvodv~C2lvQ)J9@3u<pDnx0Yi_vmZ1`Hi!6lR=P9v
zNj~VP`B5uHMsACO<Wmp@?zyn_E29g7xMtY;xg~1<k=uqxHw;*WDMYd9G!@fWgo~6S
zSbjst?h8QLCU&m_FsbEz_`noofvJl#hgwkr>AL5AGp&hH^^BU!p_^}USAwth=DsW7
z=M*}+v8K17anV>9eK>CiyKi56rwkA4<$Z*|UmKtjykKPPw7}j8&qF?x#I1=<X%R3M
z>)jBac8Dd&|Ie#}F_0D|sxQc)rDOvUy@IO5U34XDRgius!>-kH(sxLS+A*--UQ6_}
z8$ln0fnHzqF@P%wpu`mP*Q;5sp+>Z$NwT3%(%v%od>dV>HT)?|S0r7S5EkZ(lyZD;
z@E>wOB9|_n`bDg`U*2>VyH4IXWyLhHC0n$3Y~J+G8H~g{Np$shM=5ubgOPbJsGEG;
z6ZUNX)0!{?T^T3w@=1xk--<aXnw&0^$heK77r7?%DD<0Ty|Tj|cLt^dJQAXBffnb(
z&I;##U#^nD<OnQ7+%x3)(^k_-IE6n1<c*zT_IJsTc0Y$jiqs-VeMA34?!uQ-qD7-J
z=ta*li6A_sqGpfXQ-vXe7Z{CVP0Rk(8i&Gyvlf5xsWFa}eYk%uON7(%az^8Y(4hSi
zz=F99-7FrfN8Kf`e`8J2cVwK~=CV4n79m46Pji28$3V%fBWwGw=0!8w3MhH<>{2sB
z*|FhQ1WbtF?A@H;eKa<%7&z7_Gd0tQh+bS`kQ`AZg@k;@6>HlTHQxI1lV0-t_b-oR
zFUO3D7yL6aiNo>9Sf<9yHGqJI>b0WkiSPA!==?dkSb8s&Z;QW)4gMltF7#*e2nacQ
z#hBQn4QPhev16sZvz2qdP4~L8yZ%M~LRy@^cQctm&FcDjAypMEQ%RuX^hs3#hVHan
zZCv8fpOj+6bMBJ>)ji>G>+Tw1$E-LnL0zqI<(haa!4;0-a1G)5GqG^m=f6}!O?4@9
zDdRu1cXsk-bu8rMx^J*>GJ~dr$tPxHj6{3-Cm|zh7o@(bNyQGSqZM)>f}#=^0)nzw
z#eZ334qgmm3i+vebCJ~D0uvs~>(}Et=X=~tp4{U{&WGhLTHIdnL|&6C?hkv|(Ht0U
zy;AiXX=1ZG_88F`!9%=flJ?LdJ>p>VIaG=Ml+7CF_aEJrdLq~NC+C3@iQ_7TEbS8)
zj9E)^{ja*ZLK>L?72M5Wc~m{^EM7PNcoFd77-0dgU?ikrnoq>JpIuoL_CVL60$rja
zgffH;J?8QU-fS%S&bK<sLJL{h*^~@uPL1;2F|zYX2CC=`F@Y{g$9t7@&?|MHvTYUN
z^Z$L70cgFPwOU%hNR4Mla2|Js76DReviGccZ}+nJXp|d9io^g~s6aihR=);~q2*DR
zM69EF=m)ypbVjX4iHyA<8DSJK*Vn;^rO(E*vthO?;?8D)E<n720TQ_$-jFOtU3^Aw
zOqxPZl`rI+UI+A6jK^ngoHbCXeU^BB(xg!QfV1|Oz+*k(5aq9m>mvKCb<>+#Oi!Gv
zF26F$4yFRMDmu@#*KYy^$uB*W<n@DVpUrtTU;DeM@$}6lK>u|~b;2Z2l;<{%7f67?
znGiS!xx06Bz}<JByB-_Wsqfs8kIp&+;mpBC1qmm>4o;a*+4ZjSJ!XF@I~D92$@bfo
zvGq#GZmwi;nfZ+GelRp2&55pCDpuo~;^Oo@mUYEa&fe7PgpEym$|jC(J%BSG6s~Vg
zcMZz$v<>ILx(5tM<^Z(!)*_u|pY#i@0Blk5s*rDN%yHQqWAcHLbvaF3aS}sOpqzyP
ztnG%Pf4S<s#Px%)D@eN}ucL0fyXhS5`C?c3rY~0?*Yfj;cX@34%w1N_mO>U-IRu7x
z6BuM|*5p}g2`CGG8GW3A;{v-$1qpL${L|>hbS15i^AZ*2EQRTGtGZTu7ho`jXDn6|
z#@92_zx~ejC@=ATpiIQ$u_tiw&EN%#>2Ytvw5=ZVsQcxV-|K?kes@blUz9c434#$n
zRe5)s>UV!kjnBgSzrHIxC}6GR>A8z;3~%7*cbB&Hntgk<4Ao7S1`&q(_Zc;AHNR?K
zm@P#aT$(kO@`5{{aCC*p*Gw`o?3{SOzvy224Ns=f=X2A&Av8@bQtNw%ZHF(GW5tPm
z)7Q8uJTpg9DLRcT_1087th^RB8!#%dQ#P<ln=udhJqI_mp#DC;3Gd_)GCvL!_HUpS
zQl4kAWm$s>?pK0{JVtUjKSUNQPiX_n=PA)@Z7Te7RGm`SK^j40+fVl>^09yKqemD`
z4WGISXJMwv#QyRK{8k<jF)ya~yF;XB*v7^_q@S2g&MaKCo3(4StgI7}#t^o)3b|y^
zNnWoHttMb?Ix1&gK+NgpR@J4ZPPB>$oa5oJET@~Lg-@Li)|)GkoLjk54;VN>2$&~3
zlJBsg&JhME4b_XRuzOrxduQ|_;hycK<?r?KZNXP;6ZpJp<29pdIcwA9Y2YTkc@2o;
z9KYlTz5ESYH8;nHI&gJ@W)rPAmcGRT^Tge!9Kw7*#e#s7D2x=~Q{V#MFO<^xY-0sN
zUJMxbAs%MWp!`;d(bhye{aI_ImnH1cE}K{sc|aaaa8&S=D}q+c%yM=7u=9lJX!F63
zYGZgiAn6ysnm?MO>#u4XKp{1&(vumt&WNCd!#^-4L9CNQ=0nG>$L!b@u4kzxeU|0x
z;^5o(U0Q4<_;jmsn(tcmjgtH=suLzfqq!;NAq~X=*X1NOq|K;)J3xM5I>*1Bk891`
zz-wn0to`9-x|?M>E`(ya@=K|P)7AdD2|eaoGI0`gCylu%@_6Rk+oQF3yLcrIy1<VM
zfs?@i4%_*+Ax$*T=n+?@%<{snTWS$kVWxXFmF!P#k9tHqZtuIUVYdJJw_gz3^F7@+
zR^CHSDq6zown|z68tl?BwIbAW9_C?es}|<4P(%KZrRNh6$I7sRvX5&aTNm8ZP2=YR
zjGF_q(X3z8ArAFE^?JgMX-1gsvU1Z?p$7F9Ia)fIm=yOR_uTUS<fDGMHEmrp>+NG}
z-n>X|XZf6-IKx0TU1?I9@@gy(rZ%4h^M`7<@#M3x@v*8`UnGk&KKS2bU7BQb*pfw3
zU$Md7!N!NoKaT0%T1KLHHXTN(3qdBs-9SeYael{Ms<PY@%UbQeI0V+2cT*El+xmqV
z*Akya<OPFl{vJi-Q=fW!bdC4WAb+w2>0NG@)~co;M*M8&UFNe#Zmj)-%ZWUNwg1xs
zv@@p10|d}!su9fY(gx=;*4W%m{{%~^3QxCS+1j{fCADw_63Z$qNDnFI`OL%iUU2EJ
zb*oR;bgLLDa1U1u9&_|>HSVh=Lx6<WaoANeOX3yAyGZcvFCt8PdiGD1<S8nm|ENiS
znlllFM+Z(WYxP6<&|H5{t&zzVJTxREQxiV{-2I#ZIgy(*h&siAG4ta4B^dnc^&E%A
zv<dXQ*u{Zch5^?kcp<tJsN3D1#Na*HtM*1#&r!KCZy%=iepqE!a(#yiq1@@L6X#`W
z0)H1X-r0jw!xLC+X4eTjyjHVcUdVW2tx7Zjkxk?-(~gYCAGI)1vN}o_YRpc?!wZd(
zNBzw6zHdM_NAW6a8g7;z4sdVR`|kui3YGO_%q=?!dM+BdJaaKp>a#(AibUFSk|kQ{
z9{R<A5)oB=yfSW<857Z3HTvIgGpdJa8qU_o6V^O`4NgqO(?}Vi`2Pu2=2~bakLjyR
zlm0Aa%P4f%w0nuZ9Ibx&DTG;PK0P=`QulDwkVlG8(g*lUa_35*0Vl!V;JN+1Pe<@U
zi*mqB@`9eQDHXNz-p6U&^k7!jxYY=PhB;w3-j!&Sw*~r5`2z~dijHNBO}2#0P>fl*
z-oM)#w=^7JrR{RnATiBf&`lfh*js>$s|hSfFpe|xf*2J=(PK7u_XO&wTvMVjUYW45
zHJh*H%{o25!P~w37k1fV<AN^{QYr!v>pb<Gh^_P48f}fA#jt-kU-svDPK2_<k1L**
zD0^32=ozm+l}LVhz%iv3D2ya9$rF?+a8X^fBJ*@%y!=|N$k?@*!`pJneh@xDy8t4<
zeK6r^B06$R;{od^F0SxCkCKgOBhe@)6uq3O{ZpOpcEwN%0?^ZM9zc!j`6r)D!h(sh
zs;x?L#%3mEedO+X!u+X?f9URP$|{gimb3^7*kG}Ctcy2ShcwNWD5=&+7N13bmqt6K
zp3l#HspK+xNOX?y4f#h5$;a)>y{VrP=d2x|XZ?(~84VT=N5XBDebkNK+sO8W#x^LY
zVnao}Raz%k)0@;fW(IMJ<E@*S6Z5{~nKx>>i(w$OLJ_YSHo6}?2f-1&Fncv|SsbPQ
zGk)fX)a=Z#<6Oa|6{g`&-*4yagIkUnn{~J9Qy@2INTdB`S2*XifJ;;`dHxZeY#k-e
z_wVM-iL8aV?czOs5qhUhhv*X?oB37F*kSqE-y<H9UI>=$vDUyNQU-XW2-&gdrtJ1f
z7`jYjfDH*zxS4>3+9gMz{db;!VMQ+MRUdGwi7eDwOKO1Zo$pf0`PCRq+BT=(*vlan
z0nrO)^j|5>d5cNE#`W*s<R@X0DuKd~U3eb$#IFAJ(nAD)nj2>9J#tSWTe2l*{2PWE
zc<^i)C3Yu@?FeCBu|<=MU$bvbV`2;OlHk5pdy!<?YTae6OhiU^w~qj~M;7&~qe*)j
zn-Y+5PmAf1!mi_*xq^Lo#3?oYvGwuc`7Hsg%jGBYd7Q+N@$C?dX`89$igGVDeUbRP
z;RloyryFf?=4d8C=eze{Kc>M$4!b-7Lk2Y^jLQXkI^}>qsnf5j806N;U3z9zeF-Vt
z9z{wQxQuSu{YQUa$JQOlp6y=%p>Ags<E^swl(^R6KRCF%XJ!!Tx>$%~RnmNJ=TCtb
zdk*K^BtT+>l%Ey8&ggFiR*70JxK1BE^z%U&BUv`iJe!BAWM55CWGa8$Z;IMLTn*$S
zL4K9kW;p4|bN6ev)OWeDPH+V3spZQ=uhVY@Re)|!3Mv0pYe^A0e-T+CC;u-=qqR%X
z#V8VHM+5?0+grV3MBKiR-<Q&5!Oc;@*GzwkrP^e^CJqpVAoX85g0BeEVx6cnazM8~
zScogsT$&zT{8lb4Y>&Ql9tdB=PS3{OgHA)+wigZh`zrWd+KnCW1ia71CO1*v&nBlG
z2*>qsiu9hx)Aj(fPEhD>WX=q2UH_3`#86|{4EQ=OhX#k&DaE6`4HSJ&?(t9y1~ev;
z(2a-C>Cn@XD2NQRyRb?Fxg`PZrcFB>0MIks-G;+Ap3c4`;S>(b@W|;2Px?AudNV|w
zf`6}Bbj=T2sQu|~Fg66Jev^IRp}&Fgijk#aFeSqs9g>8~Vtw%{-lGsU#UVc34>o;0
zsjtLteCCZxF>3Omm1-}gU%%eo)O0)>6UaGsS`p41XPrjn`^WoanhuSytogxPKvo84
zvi)2VurH4P)qvnznSdYc?jM;20u@xQ#HB$%NqX}yvEA$vVm<Z}Cwd7z2Q$kq_Wv&K
zcnT;!Vsy(S`D7oXp-ylEkF|#4PP`vH%JNo*+j|eVjsLM2JRIsQai`}T3;V--WwNM2
z>-JQ?fL5kXtN~C$GKPhulcV{-4sG>)LjM|HUPi3X&!c7L0IO$?W{3&%79@vcA{Oyk
zD8^Za{^nEgaBhsPS9%5igKUWBw#&*WQ&JWT0r^|B5-F5`-;8ZKZ_pWA?nq?$4iT3+
z2xiQ5Z^`qJ`;AQJOWuez6t+uTa&<+n;<z;OmlRD$;<zR1F%M#OHLO;de1)Fh&T=K`
za{vPC5F>&Js|XCTRnd_iZ<3%%7Sca4)Il0)_asluo=JEAxqM?}f3p^k*2EqEf#U>4
z;xv~zb{PAmyl|_!lrL-f57-fkYzr>O?10iz9$U2(UGZ>VoU49QmqB0G2rg4F<@okz
za|ysr8x&3NF?sBG3dmbq1@g&gad{C{zJ+J*9vHkJqa&aRPyxo--_*m0e8O`K*Jb}}
z;aU58nw99=R?7_2B7}`Dy6W_WuC?X3H7do9El`w*GAIF<L@juh;@i$c{9}_d&l4!C
z(03f19}*y}N2~)W#!Yp>*qTB~Duyp<{JIWjYNstvBvs~|Vc~yANgSs~nUq;|9gdBn
zEs_@*N~uVAHBiv}s4dzOA~ZmXtJAJ*EF8hgUnYJtP?W!v&pEXp@6BtUXkzd%%pX`|
zeCws)x>Np+r-`A4eWu$=@bZE_`V28XKj%#rY|9kS?Y>L*T=F=gV6H{drtw%7)r%jY
zs_qbPPoR-eo_L(<&<<1sNEu8A6V9ULOLD8K#XNXk;K4PMgzR&3LfR9l-h2c{=kBhR
z{*A0|<73-FE|2cpgBhURYB}Q7-fpAcnn@c;NFc}z5qb=hDX<uWjkc(unW7l$<6ufc
z*(c>Jh-=>DBGJC#kEO%{O+Qo`L_-P=F6nOW5q1$ZDNiMGe4e4dm0roMofi}dck8{H
z>&B!6RJQd)P~vb&c~<=+<shy1Q)O$!r8Ob4PNQ(GPz0IOOjVE8m{3va2dmx;q;BN~
zljBTg-SW>G>X;>Q`MHiI;%fA1mQ<{3eVvI)KB(S8#b=%Ix;OK4M>S2FBR#t#hLwxB
zlJA?B56Xn$dWjf`2UA*qp`LZWHu~0jSHmS3$(EYtd}M0Vc}Qw<qHa-l!PwpJLU@ZF
zlBH&3m8lB5{A0Xg6|8nBqj?-EeM2O)a2!&=&Ys_YkBwp**xyg(awr{JC~$ze>gW*g
zd!1B987OhSX<O;Tt^gVnv8)S;uq4UYA+PoW3ff2HY;pYmGx-$>HNvR7fb)B*xaJ(Z
zadic&rpqr}fz)Ptobu&=iKru>h*|t2Gh%+On#{*Z??B$pi+N&nzA!2ja)R80U&FmK
znxVGX1|t-a?GfdRy0E+jLdufCVk6a~F+|jP@$;FbHU?}ZtI*|HtZU*C4Xy@@9<vCc
zv=^2$IY*Pi(__`|feS>AMX!W;f`I%z&8vBTyI=jA-FyO*V7p&m1V$x2^{zx}$JSi2
z!RPCvSF45%XDM60@VfFlFG#<<Qrfnr#PMc#@pHN@siJExu|*ER*xxyWP#6wq`X)78
z^kVAQlA?f^aO(k^`b9<)?J?xsdYJ1E4Ut#8K34~_=WT!b^9kh|$H<6W2z~Ag-i2v9
zPIvRDgL&rK3WU~i#d_VwF#elmGGGy6_4hdcSB-Ej#YJRV%>(MfDE;GyNaG8gT5>&A
z3vA}p<#P>4>T&!Qku&L}5&|&$n#oSz=L#Rr3kAnlZ5w~gS};d9SjdG{*VM>zx|W5>
zRqb5CJ3XQbMDP1Cff-h^rinEW+Ls+cGIjZf#pddgwulZ#Gbpq0-<^GKxjy$&cXJ$n
z3p@61j-E)eer;a}X|dUKwlOilyS=AI1M*=JsQvLNi}`i-9I8>-Ck&d5sgchH)+;mL
z$Q;o=siCs=5h?uW*9~B<6pjDfMORT+RU_H)E`W93n)>Q_ALB8(;5>QI25%5>nnYqx
zxjC~0$B_pcpba0)4ls2Tn}4Pe=kjv7Dz);zuuR7q{JX6GgKee&<Xcjt<uoxp>I@ij
zb>Bth2fM5gI#~)W7(2e}y=<VqqiE=IiPUaw;|FS@kmlZv)u!A-hyJdt0qMQr<1|==
z1pPg%|A7kNCa#`amU}UZFP^(TNYBY)R-`=tEGTA=a2k45dj@#VqlMEu_Pf9n!uivT
z#?qQ>68?s`u}#{Z9oQ8A@#gb^o-0Zs-+pw{Xyq4`_}I^QpZ_Xyl+v#<Y5@NDNn@*`
zq3G2bDm8QQ&3z)mcXw?g&Gyl|x0o?pa7yIh(Y0sTatx6m#E<>v!}bx8L*291@lw(G
z4W1019gT#xbeBRbPj2Dx=fb^dPn-#w<Oj(05{gu>BHI_e4kwt*9ju5duY3N}&g<V;
zW~i*~rgYf4-V{fV*dnz;mB~31YG~kYw!AA|YOeBS=qnHQj(Dx5qyZ;ftl5Mj4(Plx
z#R=sws!pot`d7}K6M~>e%7(Gif<&q@qL#=^&~W5M$%YsAH>n^+|Avuywp(LNWb>-N
zho2ZI26e1y2yn*E!2yF%gsuwBtE2!7;76jybtH?Ms*~iGoL7iqeKfreY(F=`?Q#Gz
z1T)b*h_Z@p@-(rdCq46l{2sKU6XvSE)M5Iawe>fPpj(oMYniNE?R5!V^FEEjUl~UK
z1uixMZs>>>4Vxf}mVjI%(wPx?-bsoZttHJfE8a?TlE&aZYxO`c!tS{_fpvqaM!gYj
zZHUV+oWqJgqh1qw`3#6nUI5=yZ~T?UW+5e0PeGv+3G8mJlwQ@1uz*b9qG9%K!QPrz
z2cGK-QJLZM(??Zbb~yY1Z`T=5_6*5QzioOrw*KdIl+VonS0QD~S&HmMeCIhJC38w#
zmK?(z>@$c#4{w2fkgJ-DWvHoC(-$O}q~jY|kF4w>q5>4ysWoVh#(p_5Fn2l$aAB2b
zY?W16@j6oR8RRy`wcMVP-O-IRs{W~PT<sEjHBF5tU~Mi3%_{p}kpwaNdr(u^!=7a%
z%iN6c_3}#Qjl8Bs(rM*_wXw5lU4=s)$FjvrXkA()Kh_-#cRsRr+y^T-ZE$vU_el>V
zh=*Plm*33Q`ZV|I##9}ofE~tbGlN^NoRVxlgGph?#X^Hee$e#{ZmHrG9`Te@LP{3j
z#&Ur$eupK(=K^3*d9kaqeSY7iS}f5X2LU5p1f!`%uCH?vnKzFFK3wO9I4_Xk%sL`2
z@VlZhk5zBkHH<{?3ecp|igeR)`vQ;j6CuStVWNkNUB;g0ozup>ouR2=!rFmBR|Xvs
zBCpxdceyKZbW%WdEE*{nkrj2wJSZ4;Ex=bvgKj;79ar9z_#^gbKX~2g-7LP2ribKs
zQ`AV$<<O)vwy0`^8sv!6R<Pi6@5Zj_B#6{9CVM>~j+QEjXx*X<J?_JUD0C3Pq5|dR
z^7hpidDK`LP(mgjckRx$7!R2?@NdhG;M>F1Fh2g!jj&kNa?q~+%~;JC>=7V(nLKE2
ztBkIx{}%$v{4V%HatFFSVT-^w(W&j;ox`fJ1C!Cw;mfDU3SO`f-lZKdd#sTe8X;n0
zgPXZUW2(@T!IEIfuPErxYuI9Oc<?5e8}k~=e}J61-i2}+L_zU2m}d2iW#91#)F=70
z=;{wGG-f%E5A8+0hLl3U;|{(&`i9PB!bMnabOi3Wl*GFajfS^g<#+7m$A5!*H%f-G
zM7zsK@P0lh<*ngEq1kyy>c3XIy$;WFGvEOuW7`u6`N6q(s$qp67uX7Y1rBV`#08$@
zV{}y0DAG0&d(MfvO{e-27<F8@mv_fPn4L94Ryb_^ek%9Uud^XjV+vRI*FbwaW=+NG
z0%9;J8Z#(#?3iUVeR+yg=9v%_q98GS;m8}B0=26`^+g=iuD=$T&+Y#@-T<U-HR^eY
zD0{JvzTT3W={oeWW^UCWCwf53%xA$@8oSpaiRe~T<~3<tYqqbBV~s1PC%^)xmN<C9
zGoV?9JHO2aoD`yE8e;%btVQm1E}lK38Ii>ceuXqE;@NSjw0U@G`mMaISR!UIH#?|*
zggvs;!T)JA=h;4*y|&TS5B}v=It_j=BV!J<8O8-yt52I)=9mcpG4PjnZ2VlLU@&7;
z6&4=hMir9W!F0|3$(_ho$7Snw0Y<HP_vEa8ui#C}NZIcht;}{bl!Q6WM~ETMUm}**
zV`_<%vh{U&btAh3Yk&wZ9OZ<JtYR<*u^e}`<YN%mbRXZ=MP8<^4WwB-1X7Tswvz-M
zh#=mhnHPSM?q7?)p*yPK#d#>4TqZ+Lwpqb{QYn3-t>Tg=qCbPl>|wDCAH*?f?TZHA
zC7Ft}RThn|SbQ~zs1{o``MDG`V{cjT^s^VcH@uKGyzT{=pU3TIbhQl4j1W6xwgwFw
z7LAM9Cnyx$KWcPrdQ?5*Td)tOkvRfL=E#eNth$A5Zw&(RnZY<kIM78GE{2v4)wghU
zCI7!P4ZUfD;Wp`8{R8#)xTbgC9I>Wmtqlm`SzKuq+<TEaC2s3xP^6k>d*_xr-HH>|
zT8<f_N1fmq;2OmYRs!%W6V0Psw=U>I-9VYV9iFTe(?*jNs75P$sJ_EAPE?XJ<iA}|
z7}Skl6-idrKyp#70?j(w4!_J~zMkW?3L-QEfnO$p)056S{GsIKG(#%ToOLO(s5jC2
zHt0is%PkMB>EZc~f1o(=j?N?qP?5e1i3(Vmt3bC9f?;>)`F^R=NZiI=U&L+?EEJs_
z!A_g1=NEan{OJEXmYe>2$X1_wtcU6?7}ECcN5AGocc5S^7K0_@F#)<Bt<%{#ZMu^t
zynH!ey}Ypg<%E|>%_x^E0qI(#ToIP!rq#|CW`}VbCo|_)t_qaV{D!<N7#;5Pv#+XA
z3D@dXfL*0A2S1&RR~a@#+-Or2VJ-Gd82S<!@I0yF_{N!Dcz4dSamzOTGr_~_lT%8~
zVhIIZ3Hd|-jKw7`H#axU^4W}bhWYt|PV@aj>d-YIHs7X@P3tyo(vo!LRu3eBYfvae
zQmD-K;zr`T)o~l+L+D}VHL<%W{`%k$`~|1ay^!}2>RVd!HDf_7vI%Z`KgjEx46^x`
zcQBd_5zaB&FBbbcN0tp5_Bg^d-9t*={z45F$E8n385JOQRnL2Rs3_Q?t$@vz(Iqm7
z@!3H-W6vTZdjDfOqt`n}$6bY+)8N5d^3fgQf!&5Mom>(>%b+)8c#<Cs=6l7}y_zfR
z_GrCKAvKmN6LfgB)d^)Q)HK=DEt0}HcfhjzYVBcn+ZSt8QHLCj#S&GMN9Ogr*c$B^
zbQoF}^)xu$bgVUsmA$SIvh1>fa;vnKH=13ua3sgsGK*^cS4XG1qLg)-{TTQ7J*Fu<
zB$4|^jafTSbT}LFl0(nGE+iQ}Me8Dl3$eeQ>ed&AQQ48MQcAj9L)&j@Pxy=1%$=g%
zznDDjTmA+Fh)uNKIQ`ZIp~%WbQ=luyv*qz9jag8KfOTzGP(sde(y6`4ikMEKoKR`?
z&UG~Rl@vfUN4w1?Yl^kIi3jfeqZxztomc(jQrK+a<A5(p3jZD)UB12N$@^P8fMgU{
zx__<u7;Q-@(QA(TI?lj~Y^6MP<cjkhRj|If)alF_Bx}hRV^TmKrvR9L|MlWOz8pL}
zZFHvQ>)4qWO#d3ZJG)_7@-g|_JG0XU=1&{!IHlT5DJ6EClFc=b={p*7-9qO>Mmutb
z3$?7_S#2lE&I$A(BcT08D@F1t8kV6@U1|f#a-4HMd%F+SV-N<LrNH0@AyCa32Tr*N
z&VQoXGvzz(Swn_I=elve{ko=O;t2CSD2Yz+sXfs9s@_+gN17QspvABEsi*DpQ7r1<
zxpD<3ypbx}tGEPgNgUahQNHlQDS*L8P}$?WdZ@5SDuzW)YCcvnh|7N?JBpX+_vC6U
zBg1U7In>;5aBCBY*a8={1x&5@l$1vWG*UBPJRRicdo)8|e?H%~tBZ{^tA3-?wPD*-
zLN9cYdYI09L+S87Sz}9y#ph<5?9=Q?!05QBZe*TUZ411iJutB~1g>tMhV8H_v1yXj
zL7@#sXaJG&yoTXgR9Gx2AdnlwH<CdNDN)EsMy$9ztEcPY$Ju1Q!TgSxwMrBP8N`%a
z3#$b4etmdx<3|1b@g>oMe$45u0+$>+p?*-$S>_=bm++~~-1#VypP&Ka&(>eKsY_NT
z91f8Ar_phj(W@<q7)D50g=XLu@67Y^Q|AjaL#Njs6uXlaC>-JeOx}kow)sKLCA5yl
zMNBiC$3z7j<T4^I=0kre1yln!2qAPiEQ6j_G&1G+okeXSC3D&f$}$1pj_3eP%g$Jl
zOmhhp%MZQxzqJPc5x!L9kky%hex_hB-W}T$sJAW3uk)^R)P5%baJCHO_{d8nB@^#W
zXDy|MBoFlqDD#aT9iPk3`!`u}WJ@Q(Bw9qA-2U>8UVr0lJ{z##be?=n6ihmT*GJg-
zwi`?q4Y#iG?31)JRe9Q{v9f~Q7%WV9_slHOKdUn6x$zF`{pe=U`*26#(juI~m0*5<
zy}$Slqu_B`PY&>hYL?Vrz1Y08MZG7rEB;IO>6-3sB^`LFwZGDw(3Hw72DQr2&A;{~
zBY-NBun2teoOtS`{{^2d@0?|nt&qjz18r~U4E9eT?x^q`cPJCK;j9aIg@xSr_&j}H
zGbRVU75gpN?f%sge8`lPmnbN+=Ap%->$%laH3CRNR<=4C@$fQAT~fNqaRT51`<Qtk
z$9H;G9G9MvAt9fBzqp9TieuLeh{cc1p~VpXQ48VhfgFot-s3@yTUFTNud{r4z!ah>
zL{KG&6krDZ68)zrx7kDN_Z?%QZWAPcD?U__`fYPngm9*QwFqn5B#B$NAC2De#4$qA
zbv#BY{^)R>zS4U<_mfN&Gj8e%H|I>V{OT$ZB?NdKgZ6m}3x}Mg${;Tvmomlu+zi@z
zlP0J4#3~4;UOYJ>FH?rBe_P@-G*4eHK%5rGH$L3zK31JbGB$nQu!`aQ{zd9*h`=ef
zztG%X2Rd9s0KxL$WMUB=l@@QIiGH%)%}W=mAQETwCy2H?=G6p?;6o9e`|OE<*z7^}
z5>KQrNpuaCbi~#Y5bFv7?NRUhxfztyJsNFILraQkgoy#iFMpJLe=b}cNiV8{j(#49
z0jd+8%`vUN(kQIs5ztdb@Ts7~pm#r~=~v3oO&aF+z8nZnCaynmG`#}u!oq*Bvyb=8
zkh(F_JLJvp9vAY*TN^x~HywU-$`l@j&CcrEvJ&!tV6p6pw`&IM=SA}9-R~7zXHNZF
zLaP-J$}Zu)*t2p|Q_E6f4qjNUNRW4OVYU|8taB;oxmPY<omQsC2q^HkQx=kZRF=dH
zSXCa22wIWSMZ5e`nXja0X;2Zl@XlR>-A-$LE_DI`7X>F=(P(2c2z~Ki;gi=pd)CM)
z1hzNr>V8qxKvJ1g({EQvSgdJ{r-S<_&CjhMeVP9toiq<CGQB^mk(2#ZNLfZuMawn>
zcsqKmuPH%ADxa5-+d#~$Tvz7!6PKe&F`l+W6<=_B&4b>;N#C&@c|Y-R^_IWm`K<Cs
zhbdj+YnQR18P%GZ@b=6FIZ+vdJTYpj!F(<wVnb33xqvQqy{sWe+YW``wYAsYLhF%K
zXxl;8eOBw3xxfp6F-1yJS?K!0{UA@)gAHue>!m8{Eg|~fZWFWy#IL?!R(loM#D%J(
zR0+PgOo1aJG=gYVi;1Pw1azPDZ(;SuK@J~Pz>{%&Co?X91k~z_ONGmDw-<u77070f
zt<JD#k8t7O8B*w}`;;l)jlml=Z^nyBQj)LP$6|D=ZzYCG-ObK>w01{6VGhu?SAgnM
zZ(=0-s|dJB+Am<vbg<az{PXlyafbxOl2v6Tx`smWUa*E(^Kv3UCI2ptm8>YZmHKIf
z2i&n7k%CCAzM$6q%%w}mv)$3OF?bPNNf>zt!T7?3jqeuxj)e`01n}2SD&q)0HO4aC
zH>P}`V+SMil~c!+Bj~^ee}W0^)+I?TNElJANJ5<x!dL@7^^ydIeq5eEQ5hNjyqDne
ziJ^SANbl~OgS?>&`L^oB-zngbSSa4-Zf-@$=wIvR-6HHEmYF>Y!1@b=UFV*TuGjG1
zDey2in;i-`Hgyr3g;yKoOZxq<WOx`({L=p1D~~+i``kFc+}L-7*tRnL^7zpkoZHAY
z3VB(cybi(UhdsauM?-P66BP>^2kC=RgNE!whZe{GV{PB)un5aaDq%9Lp*`bVq)+?D
zVKuwZn?#!h0yd4|04LM=t&3*VQiqRP>l6@I%SSkTfj(QvqeRCwDujaDCF;7-HhbE_
z@nMMXnPsvsUgLtS^T~emO2*JAgx-<70!?3l$5em&C)I7dA5LMpB?AOBm$(T0YB-g>
z;h@xOWiAA-orzVtaV#<}9MmQXzksIxATziWjUk+66OGZzHQaT8m~MEvyeLgmmpf0q
zP22yK-yG6?5M{bGV<+O7>}473<q>S-7`@)}u|1yxI7+um2US<tUdbnx&E_wj{Ze?f
z{68%~jH~Ca78aw2e8TNTGu1tJ%E1Gl1c*>f^eX^Qg<F*ToUXgrEx4BgAy~%2QhM29
z%7R@=M=oV&hizcMGAuGWB0N@Jekv&~kDFVHg<Xt?Js>ruPfoV$fvavHYucZ!xDR|k
z?Juyf`CtTtm!|k~Fp{l{$W3l4ty8DR&{|ws&=@q)XIpBjI)t5URKKoOf#I!;L?;RQ
zv++}U5?HGkAx2zw4^cV0qBT5Bk&ahyJYEY8iXz{-&v5Ob2a`!G`Rpt%c`l~wUS15c
zxE}3U<RXN_kqYYvImAD-mdKSQNWXBC%iBPzr!M(!qyz)QAmpP2?fQYcxTveO>5*;E
zu55W=E9G;eTY{dnlH6_R#xq20m-l~IKNt-uvFMlxh6yUQD?H9(rcG?Nchv{gj}W9k
z)DX`P*7;%_dR+ZZWrK}a?LD&x@;3S$?$qt?bDKYNs5=?HZ<F5d<6;ZGIr6<7ykk1`
zWA6TBN%+G<9BfJQdPZ|{CKdWSSPkuT^QL|oOeD*nqtAsu8)D)v`ipE`6ZAoEUM>Gg
zm$Y6?t<o@_KXS5IuC_wr3t#>EtG>gwv5XW^qEsz0TmqJ%CBnYh@P5P;w|i8-2sla5
zJ*oenR>D5Erj~ZW%F&FLW$bfWTdGh>Oj8_GfSQjIBOc0T8^w{5_n&T=SPqZeQaUAY
zkRm*eT+UGnoe4mdt<fITi(nYgSd~HhK%CMxDbgN%zw05NaXSRWZMOr{j~?E(+yM1j
z(gE%I;t8%70=;gnCIS+*jEo3tb}{0<-weIH!puth`)=62&FU-(u=e&%q|1(tO)b|W
zHP|2|UP>h;eMp2`*%?`eRhY-CTVi@rU%O7SI@mD_EnPB={UrhF(CS(&QMZD@(A#mL
zm!MEt#6q-w9gtXSSl>}dNW!s+P;^m~)(cr|Tg?P^fF}dQD<yt!=&jz(Gfi1HHt$a>
z{JCbNu%=~8H7tFY%F0&sHuaDJdHV&zStq-_%rU(RWWL|sPvs4-zOWdVV?EB$6S)tb
zG~Lx)3t9PgBV_}$;Y-xJa*$81<k21((3;2pHx3i1cBm+M`*9HqjoG3vo382cqiXU4
zo5BXK`zfD@&>8Us7OT+Di6>crYmCN(P4|QyZwZZ#D-r@@5%S<<=)$60v$<UasCdVB
zP%wh(eVGY5#kmNl^gt}mQ=<&Uioo5YHPj*^QEMd9X!$;E9XnqkXHc8gowBs-r|zbP
z7gRPEqMm=(9O8myFo)ddgPqYCN;i|!J$+LWMfqcLd23!nER~$(6I~<`xyB)V&nen6
z{C#x|gmC|Ey+8UH!0cv`(EsSsA`@EI+I|5H&Yibl=J+xE&__P2)4<HmKb}y&lDyOq
z4^_0J=rlyhEA%c{CgJ#1W7XkB<r&rsIQ|GIaw-|^{+(YH(}6R7tY&?dsjsn`-9qc8
zrz%2XwPjHX)sQEMG-B^Ny6{xh`7<&^F^tV>X=kwtDwtbF=LdKsXe1<O@(R0IM)&7a
zsPAK4wj(c&O2$x5uqP#@m;q1J8zBG2=PLAMGaFNbuYa>PJmV)N|1CZurjAnlbHcXU
z&5PaH+Pn2NKI(is_gDCO5mpa#v^IGUn5~FUS3^+rh^`dJwI!jZC3~cGu|w;-b#rlU
z`ck+!i+sg1FTJuvGEag2<*~6SwTta2IpV-?)I~kRtw4Ln$$NV<BQtDlfo&ybR*}YO
z;^;W5RTl<OS;Lns{ph~fJp`hsxkdLq`xgP@6}Q=mcsow;Iw_O)9qLu4&vT+kpUtmM
zH<qQWu9+B)|JL|7f`<hDH{Dw5>%me3F<dx7(Kw5LFHpWG#bfN#*`L0_SjUowj6pY^
z?&<cE-*#TJD@)G`nMR~KhWB^n^o`B{`ugNl3x{#m%;9(@PlQu}L}U`%06Pg1GKnWD
z98rl?Om8!nDNwRKAUIH~RxG=eE(cI1kE6M^WfrufY>lcFgi2skH{%idQQWFDS1fg|
zC;Ec1EiiEtnO7|hTl8wA!2>Q7>7f^Vxay3J<Y{kySHa8e4iVUoz&G=-LyM;@-zSgN
zF~`YV9cWlPU}?AIWj#10O<$L`p0rC!^FNZZU?;JS3Kh{ibX8+$!s<`$NI*yLbPM}J
ztSxsIZe+MbeEMWZ@~NCF|MWM;Op&icPNkdokIhp3Cf$nhP$aQ_(ubuPWh<0Oax5HP
z*l~pZ+jhIJIHNMmjUQ^k7DRkQBhlOmW~j9eCJF!k`HeA+Rr9&wT=KWbjMh;OScEG$
zoI^E=&g+jhyzgF=jBQZ^6$H$+taGz<Da17E0Y+LMq^!aNL%aVZZ=-7j2a>ufz!8AB
z_ctyjyE~)j{c_zSZ(EqtCGRz@k-iBjK3<$)57mU3#lSh984-j6XY{cXu$x~;GxvtN
zLAoAiy|}Hr(sk@oOtx`rC2SlfMU<`JjXasGgMa`=z9$5QLAmqngREH;-gy_6$wa{?
zJ4$q+87^dH{um5nwQcW~Zieavy-AkhOE|PgIke<P*8lq2!@iz(V@P{1MQaHwWMlbR
zwJF>FAWvSolp?PD5BX0RSQay6apExkyCp*SDswohG@On>1kVpUA8@{h%_0-x55Nth
z<B~^x&HCf7o$bdXnSMR4<~p%{@@x!oK2}|uYRp$BQ%paIZh0Ok0rluZwC$@(p7b{p
zJxix~<#)}zwNu0E^hA&vCXmM^85~_HBA!AcS{jD%P)d>mY;~F;Oo9^~tq?<@V@<9l
zco4!0fY8Xqfl%<GS8xng@Y#5Z(|>_3JGQ)GR!$H*lLv7ihzT}5SyR+sksgK}&1x4U
z5E-1U>4C>2mie19h>gH(N;yI(gO^$4A~_NuXGJOpqv~}LqS0eLN1eis=kj?cx9Jl+
z?pnQH?n<Mh!=QZNJ&~EwE_FGZ81mdRd~&Bdjf_UQ+4ONeJ=Yl<EuA{C%s%{2#o{au
znh?+}`XX(|z2=|Pv`~@)>kqlncxv960yCk99~8tc4V)6ueozRx>XfEo56@Q6#jQr=
z<(8&Z7S7&1#&LxsESnxNR~;8Mw5&N?(+zwnz?d*s?DLeI&Dz7k4fq6adAc(G(EW<P
z)(V~$f#+`X5$t_N-AbjAV(ndcDndc7F<Ilm%B;`W!*2W2ackZ+ldpL4Nk>d64Q+0B
zu8f$Arf<PigLrnrin>*ge@*zk$v|20X4jt(i}hbc#wux}=ZK<r`SOGc+BX*#v15-U
z7u?HDWmz(^wc7CVjSjH(2Js0~%g2O@95oq98_c@1h`7<OY%O*kA1xk&GlB4^w!Ff<
zn%c$PT7E?Tda2%C?wj65kcw61qKC7%dS(aXVwSQw4Y{&+K9OF@)9_HW=fdLv2OUu%
zuaJ-LwT<ZQJF0x23!GvLlLdFvb+39jzhFvX*mAF$#FAfD?Kl#?Oc}7gltZefJdW$}
zz7hT=_-^WS^s(aQ@_lTX^X=={cTyRzr8b#`ti34<rGiX^pn+c;(tcp{wLiZ#znfy`
z&m0u+eRg(Bph?JF>jXFO7ebzJAHko~u7iMO(!AJ>>@d0D%asO4ljq-n)76Q;Yo53Q
zUU9>hVR=4Lk4RXoK9AB^j9SAZC?~kO%^%enQL?hMCBhjBK8i6Xw<y=xo&!u;^c@zZ
z=u3d<GX)t&QKAUUV==u`w(tMm*eUM6oA4wo-3SU))E?p7#ht2{*S65=Y-6n*Dfa}#
zb?2fF*_>%)jta{nvayUk;NbI(_1&EzL%c&SMc}#GJgLaQsLa3z;$pKoS~Yo{+Q7XN
zPKRZriF~KKzz?MRlYV(2+9GE3R-eBS#L8m4B$#rf=p66&xF-*5y4ukO7W1Ut)yc^G
zx;VmmnUv>jlQoQW(v)huUr7&NHKcb_QO<Q|kK&G*RssmLL{l{i^tG^ZR_$(&I!a+0
z5cP#bHAiu!n+q06n<Q}Cy^f{7oZ7$qxK=MWYTJSaD*?_4)Q(*ifvb*<KFN?p;??`(
zs&%;1Bm7ty*$#E?e(QUF&`ZXT{k3+}LzuwzABlr`$Gl{i#cnqd+_rP9uL!kNTy7r%
zVo)r*V<wt=o6U&b_&-+pwuGtY3XRk8J!0Iy-VSnm-7jjM#{=QCD{DNxVsfwHn_tT9
z_il(hzRewMSz%R6VL;G%!86gTGW2;Ngt-<7BQ9q6b){v$vz3<0rFqqg=_t$3L@^%6
zJi4@|93)!D(c8b=Ub1`iY0mL_(PLduIux_Jcg7Yl1LLX*R=97L!~r(TJ%0&!afpce
zO(LGUK4Mj94qw%uXg2&NxK<GSq|N{i@Ut+$te1MW5lauE7Q@lvP1agTf1ki>Mve}#
z32ZaTYYk#z{CL(v?SgeYs%zYB)*575uo<ZQCB0!XURKU$h@fD^rT9qSjqxY8rGb!=
zmhUXfxE!Ly;)^qSCl?9&ewyC^d01Fi#w(lK+4jXZ9j>Qc0N{M3w_ERdXom9jlOH{|
zF~caQ&9A)xY!-(*IL+pdC9XAn+x>s4gQ_)s{ZxOZUrKABE?_LfpCL}B1~X#7$*^F;
z?Ekf-FfJ%o@Vlzbl}(Moy&@_2`4p>(te}uF=Tsq%AnSsR5H6Hvl!6Pz{eK4`I~Un{
zX`Z*EzaD|wR#H|{T`PEBb1=I-6gN!jlAX5d?FS|dglZ|=Z&mx1l5w{Fwh9dU0vQt)
zKYUC_IAC@Atj@q+8i9(FG3b(4A>sSd64ch%*n9s$1d>ER-zKNFfqHYVp6kY^^C9=)
zxvsk`-&bm7O%ftfO&`3?DyhR@Wn>5e5Wc7S{945B=uxcg_U(%?<Uamo;w_licEnKD
zbacyX|3w3>OwLGd0{}qUvOg>S^P3x|*yXoc_`t6+NmdrqidGA7Yr8aK-Wh%IRae2k
z5d$R}#E~JlE}27#QkAlYi`MX-zR#z?kgf4l?j}bv&qlkgrLEpP+#6JeUt2M1Wx(#o
zF@yF95AzIwOyV~DlSBMxV?HZbGLOVMg6cAYm-LD^r_Vgv16ovy<*xD{>ejZflRKVL
zi&@g>ekK7tM57j>iS<$vECxkQ3J%PUfQ&HVEFbLP2ae((qk?ATXR0mMsXG|%H+wd)
zLVep_Up`zbo!HbOrzicqhZuXE03vMH`ZC5N9Ze?f1^`3_%8jZ|U`@C4$FlZaqd<KJ
zjxw!JudnW@o9Lk-LbAb|+cW_xgTo>;fvsemIdu!#Gm-wN$wtCS=1PghVCQ+Z@y}=n
z%E*~F{Iy6ln2S{!$b{h)8Yw{Z#h$a}o`UC&i5OOns?(TKi{#HL<=6mwLw*AFxH&re
z$6<q=72dmTTfuL+a&S-2$y_a&kv<oV?tAlzM&&g0&J3TsKK!8zN>yltx#1I_6y4Pj
z+a+t)YyB_S-~8H=P`i!%yVpK{NZ{P;=o+!P`OvDuo!<0n*}d6Lclb?T+hm+lEwgK@
zm+$FP;i9kbJsq~35{VA8<dbWJEZdTw!`=7wE?E0#gv0MW_lv$e@{C~|dmjaU^NR5v
zZ}X+m&mBWAcsR>j`v6$}heZ2`A#7x>z@r?zl5WAU8X8h(>;z7k>Sp(lc0oJeb=30h
zH*WJkzv|;EUt9V2-X#L%I3$5?5Q^^vpMH$#w1r_7R+{g59Q?TPWi=dE#=#?Im7<0z
z|Ewea;Tz?*<$joBzVEi4>McAyF>YF!|K=~C%b(84JY)4^apnj4zsu5HESN_&N#VoJ
zj^t8v&!tm>-7O{H=M_(NAijS<?id+fF=UYfJson9&JKgW0q09XJRCwSQ-e9|)=h^4
zCtASEfT}Io!|y-oF!8a%7|_R2(CGS%HcG{@DGj6%B87ElxxZiZ>oEvP%);c!zx;D<
z`*^SVA#l4s4i4~@+7w;zkm3)$AF=U{q_KbbMpJC{w6U(&PHDCejaGYz94!|Sy>7VD
z%$W1ZnZX{m_l{d~;|;QQauy|2@!I?pQv2|X`Yxa*CsGj<I*Dz)Q`>j1jGOVV+CbD>
zOjKZ6Cxr0VAwl}DEqJxtPhDF7{iT!QdTeW*qc^^7c(jNg6tcuM``s<rbyUB7$^|do
zHok+)GdlvFve6Qd=PZA5C3pZdHQ%oioiA55;6@W;DW&_>fA-#KP0?Ic(Vi6`^e<wz
zKI9Zo|K^$Ue~9XjB8yyoZ_84v|Mq+3ucfm4ri_NCvX|7*_d-;y{c7rMd`ZM`$EE53
z(03xUNs-7Z&u5joH1jMUl28Sm(V5wW+N&I-qV#60PIp2r1BM%dav0q#TIfc;sj7dR
zaxy3tjy#?3jD<4E8n?W#NP#B*JrVbc)z42kTAx<1l`mdiKAIbw9#g3FH%!1s#-M~%
z1DR$5w7q}6#~4VzDGCnu1X-oajl=CQO==Lp5zLoExu-s-XaK>-&(B0YVKzoMObN3S
zJc+YRoxXdC9bYH-2%i2Q0M<Y$zx&ZjLY5eOUi{`RU@oqCY4V`n#jrnHB#4WIiJQkq
zx|N|T3}VKH#M3oZm@OFc8K2wk?31n`l1Q{Rj#`kY^CxShRMV8!_Q(0w9k+7ni?5~s
zyvu0W1gQ^VGzR=-FxqHv-V3i|(UP-hYK4Y_THP`}yq5>Q_Gwo33&w`9KO+{kplZVT
z*B)eehQ}5W`DGFLkKR)B07LHxVBsoN{S)Bw-fHxY6K;?)vqh>-SF)9N)Gn^OhzT2P
zUSM0LHQA!-9i)!^|8YCt{HwoY{^6r6PK2ad<<R6LdjwWqa}C%0`meys)nqn75*s#5
zPiWQy&M6{_=}2OZWu|DPNtAn%Q>ig3?aH9TL8V>$6~M%MhcUKeU0lqGP_I}cAO<HM
z=L1Trc#E_Y8WUvuo?`!l53u7~-(dfar)V^r3=S<|_3Cw;e!(RyIR9MC>8r>V_J?%Y
z_(pTyvWZbqC>>b?zEsW&?J0+JSk-oN)u%7MPks+qp7xbETWR_1ObSlAt>=u+VAFr0
zwmo6nIv<fDSPa@Cc-4;EE+%{jO@sxQXu-aNWRGv+;A4-l>#;}Kvvm_=hmO*}a30G}
zKa=xcaU%;ZyBOB4f?Ac-7_c_DFsTVevC^I}*f1fqi4D3M$pzlNJuQg<R%UH4bY?)v
z46OSTfq8SXZhi*WMXn(4%wG<#(Hn!w9Coq+NA~jgows8L`#I~Xm&1bPWTuYT5d9$u
zyzD>yIJ*yxkugN01=aq9;Y0iR@cVv~x4h~~65~-9-$i0jH77_$daksSo|DJDBJxWj
za(8dn?*WGAJ6JeRRsR%td2cm($H};nb>_OKgVEgWG@}}*JVc5^nr4_(v6|xU`6gfc
z&_`%)ew4KXeW*_u^~!!y)?9rJm%ROF;q>*SHVKKl1m3U|F=}m>B131$vR4f`3KGJv
zOrBdHQeuk-OrR<uI7@TgLP-dnV~Y|oyWq1&bWsqRYz#~D4q86=_4_6aA7p&XCiXmV
zKieOAn9*Y+4D>Bv)fs28>YR&NbnzvyatYdBBda7Kz1a#v7Jo;ZH=pyD6@piM>HQdE
zC}X!0E2fl#z?6>rRIo6$q@AsNpG%1_d&xNU8e^7^cNS}$67cY}MYqSPUOUe~C!h-y
zH&f_kocH9j79$2C8ORc>XT_Ru?GgC$8goOONkD^9r%a3^V`H@Le~>-*-pBqeo7uN*
zJ4PIR%THs&3tz<P*IW%7*27RgS+z=LK&&N7aZ4^<QnCgvX+q423)SZ|usH#s$4LZy
z)46+z2;MnVA&EhcUWnm(U?v(E%{cVHSJ|<3BbQuz9d_k8G;BXfJs|`B_U~`uUGM)B
zhL){ke8PasFr-{{@wt5BFWybfX1L5_twBu$4C1G_BRzM^^XEKp_=1SMLqs0!?e;yu
z@O%Xe%T@KyfSY@((K}{)Y(MQz+zj97qexN+oybsJNZ?9jj<hvNg(_^_#+Ux!j~KiE
z+nh14N?%`%qoZT&N*(iF_(ER(+rI**twZ~3Xr+Q;i#Dn$=m0>bVd4;L7*$t%O{bta
zfCkb0<JuNyb{88ZJJp<5o`2pTg)L@54^e}P!+V2u9x?$p0sHo_=e}>T;~U>(_vXzg
z!je^MIpe}hS#$Z7B<t6piv~#v-CUB$0jE^-`?fi!AUO3gg_?q1l)DhM)1SM~Q`rG%
zDy>JJ(VjjVj679W@=PC_d$IbgJcm!#_jb68wWnwbc<<Ud6;ar4L@?GC@7-z0%5bVU
zAc@m}#3@z@qWWz+ICk&7?0Wbiwtweg#*U6q9T?=imtMvC>#w19@r5vd5SLVgW++~#
zA=@l~WX%Kkybr;7rSmWq%02yyPUd*k$uqt}r%Q*HQu8319!LlYL>0BtCSwXn>aded
zIJlK>efghRb@~QYz2x<zsYMpdWABj(&b#szEIMriEpKpX5H#z=Q6Bip|7GQZ3O@5x
zDhXa9QB1TW)Vy~*<Kv%1<bR0B{@zmb0K@YQEF`M>=fK-~tI<22wNkC)AJLjJ(*AwS
z2gtJyR2{CFk*J5myV(4(Pq6LQTUli)^qGW_)+9S8CYgWf4{-f2zXQ4O2S{zOB@D@?
zd5~9bRT&FqjoJulQz1gfQ%bo;xdlvaQRqva_}Y>9+c_9%+ok5eyKNp5=s^y!-Xl#1
z(tyS=Y}&-m2k&R=cOK@z-h&LxU&7k8=W@=~FQtC&*|2CHu3in4k%~dc02{{~(oxhD
zGY)Ax>Xd`Gih#RPAel}(`<VemOlyb>bDbHSXvsH|{qEB#9cTOg>0Nhb*Fl^tSaGuI
zxG9qyDNVGb9W+>;HieG`(M*=RGf9A-ro$6ggI%wJT1yt+xdD<?kz<D#x&LAIed|7U
z-hV$w_UxikuX4_%mvi>@*OOd$AuJif_0>o#1}jNSn$XZ3P-2(ecjR}>?Yzp2`k1+-
z$e$Vz7^r_nq5wL-7YhoR4`T5k?jTE{(oB)&9(H~0cAnU|otM7(C-L(Z!QgT(d&QgC
zcg&OY&tqb&L0{hhlSlUQ=kIw3H~*&_P?wU}1djo45TjGx`}dan(R0$X=_4ZY-$f+r
z?esms@Vq&C!EC)_uAA*_OmLm8LYr<2?GgjpSk<Jh29;D{a-41d_n&$6lb>M4WRn#`
zgNzK1aWG3cx?qTx{mMHSeAz3}dLJ&aSTTh^zR8)*>Jek`>O&VkTA27~Nm)z`@&sn2
zBl1^sf5T``s2UxTN^d^ThY*!QlMsG}LyRR8ho**#&maRvhG}lz#Fo3i#uE>Io5|rL
zEIDm0=f3DFR$X>EoW2qr8X^^osRZRgZud&~v4!%$h{35Z#M=~t>duo`%8Z+#$e*qR
zo?2qf1{bClV^hV$r((OGT0l;{Qm3jdW}c=@2NLE2k4|($n`!k=q)cj`=Liop&jIl<
z%{#X<Err&_HS=@|@SSrpjXGT9VFD3;hx1_q(asB0jDw`*;Mg#1dV;6!zKbn)-Nm8Z
zPa)Ml&cF5=&U)E(R4zLo9U34@Dwrfeu*Kv74H*LDcCB6KW5)cAsea6JE;Qx}D4xl+
z+XnKbu*v+1<0;wBN|8%pgfPkR<D`3^;Ff>-M_zsNo9TbSmAvDBzMoHi?i-lud>j^|
zLN;-TYcJivCqMoHtZ&6Ums)9p`#s!jI?2t0x&EFx9KB$-9$<L>9IE=?;&h>R^p2+u
zBxc)yr{*i<VNJ1^@rtP^uIbS<Bk>uHJMQGJk9?TEJx5qpGf1W!NSln-`#JAN-pHCC
z|Ihfo`Do2xtqppD=uID#3EJig#R)<Y6lL<@v3ByQ#+2Dy?3{PgJEk5TZ@yb|5os0@
z+K663e1>f(7ze{g$TmI9mapH#&IdQr7;mv+{R=qj@~c^V<)!$QOK_%&sTgdMi>^%p
zWW1Jr>Z27br4DlYhVIA==ok!~q;o!Vp{i5LyXpRYGJwPE(&iI+v$MJRA0IT#wV;<1
z{5emrpIQnI5iMy?Oljqr0mVxJggX+fLqjQ?U6I_Q6>B88hQ*11Y7{Pu-~jIgtAZUl
zMDx*2Z2$6|Y`NpBG>;ylf7x=*f9We&_p<BYjI}V(N2^vr5)&;}+xJ$;_9zwEvvA?&
zE^qFnIn8yQpXoL0qjp4#yo&Io%717c&L}3XeFXnLbx77iT$6wNqu=KhH{Z<01D-ek
z>ThHERv^_XK2@|iiA@jkjW6HInuT=)M`*)H?03lb;4U-u-*X8DeYRf3v|kgZ3?J-O
z20g&={SFpxR@DyyOM0u(0{~CQLuSfmI|F)XJF-OUMFjDle8+4~G^kO5&5!b>KmK#<
z#z$CLP4E*AS5Me-<Or)?dm~r;(mP<~3R;y4iLJyDeh?iOwyd$bC_2qTbRo~JL4Bu_
zQ4DTtqnff43kC1r7YWRH161&tVl1I`YaGrACUZzjU~+`U#z%Pki(lZ$2Ogr;nq<Wp
zXLHWg*Rte_%V5=F(pnv=31UqUp~hb?N{0r)8WVnhN4T(b6+vCu6<wxIGKX_Hdoexp
z7CQB#-w$z5GUt0{>gP+`^^>@{%{I}S?i^?-Fipy@XiD(j%rGy2hv;+^4Fn1&CTSPF
z5<|&6f<`@pMbS)2S{aNElYRYLZ2rnuc=)S#lTD1X?94N`?1$gLqU&A^tC!)bbud+|
z*uu>#PrFv5ML1xGdsh*NIQJ)3XCeeloim+$z@XEtb@@J5Rg4K|$#rQu)MZE`#Z(OU
zeeO06dHA(I`6!1b=i{VKQb}-Wi{|h?-upYh#!uaJBUUn0D|zZ;SReV`DY@i?pF-cY
z<B*8_qKN!cubSuqhVNIfaGk1t7&x!D8olG0-mud>$cx;8juv;YHkC*TYBp19-ox+_
z9{$i@u=mSfWJS}_*UD&AEW4W-t+lIo?YrNFz4)@gf3EcSN16ZtAOJ~3K~x$<ZJ;nn
zmpE73ri#U%LJ<-bLez2;W8zf7Ps7yF!uBy2Jr^DQ@=j$fIK?*{R#IplgDsD+<LlpK
z(|r#zdaS|nbsISI;uo{x+83j%mf$K0wq^-#__iY*PKiA8$ZZzs1TSp~NPBJDYusLo
z&V=FVrK58n>z{PP(C=cm^67-iod9r<azWnys@doivw@1a@a{X$XE|m#OuX`-L8EgG
zMVkfUl)k_GzRN|orBu*|1TmZuQdNu^Ow*Gb+RNcD-pLbR{vx{{ew0*OoN>veT>67=
zz+Qd{ELudfDp0E;@n;$@_((C#*I1;TEUo3>M=8;*d>*1`@~1@C)bP|$<!17k)~*{e
zfCpc?cf&JCFnH%6Qx1+E=6}5FJ$&*0?NA-SJ4@9H*~BPUT=W9|?Qj2(s>$#^!CHxC
z62eqWCEwHgk3M&P&Z8po3nFqyuQKQXhVM79a83jZH}qDccg*!>oV|GIOlj3=K4vs(
z1uVE$3z{vgSB5|HIqv?;zhv=vgFZJ&MLb6*8tkoCdCAYeos~cO7Sd`TR1%V?QwTz7
z?Qmk00t#s1Xw<Im1sAIdn+KHTl#2*1q&y9gQrAxe=QUWciUmBGf+xdf2HS9Oa2I>N
zb_bj8yn`c8KFOjLYdH7Xm$BkSFUFs-f~<cKtOP1YX@!gTkz2p!bxkQmt|%O|4Z>zR
zAv1O2=EB>a9Wb2aqDha_Qsg*qfH|}u<(b|A({zM66^RbdycW$1<ANKn&-69U{H_b}
zbbh<%%NFTNMkKiPNB}S{etoqCGCBd9x3Kf`pJ(S?UuDmun;BTSm@}@rnhW0g7C2`;
zt-dNgsRS{$N^WfiCfpa+TGW-3q|1re@yU?eAa`g8@-%FnQ9Cj%D14SqSTXxhEuH~d
zq{MrShBY)sNP$~Ee+NJJuHPf6ETK6eRBH*o>G2asxcAHdo6}Dl0&fdVQ!!VW1~kll
zYW3^`06Br*tS9;RKEw0;<p>%c?^O^z!0`P77KT*yBfwjGtI<17^yZv7j4k%UavM`)
z;<!Eb`CS;ddmLFKql$ygk8s!fK1BANM>x$^X?jPqmauE`7;A65flGh+*I?NSvbtc@
zhU8K$CGwUg>?Jdo@H&LT3nmvB`dC4TmZ%3Z^TVgZ<w$~PAWc2Sd#rOXagd2e9%j=Q
zzr>RdJ;u<2C2Y9lMXb5urT8_g$!dd`y1`mk<RY}YiqgRfnO;EL1}#%xq)<w@n(^e6
zgYf2B$4o62rvBh5^d~xtweS8EVzzwXXJIqy7MGu5urRIUT)NIh?gTc+Q#1XSm9I|i
zI`6SC_gY-zZJmt<(*f_mn&R3`W16{XIFiv}n)lqzmRmo=jt3rKY+{1-mt4+8Kl}#j
zFL@Cx8f3DXASMYD2i0IXRL-Fg+SDY8HnN2!a!6f~SdV)Nsse+?92Ge(Y&<2dN?p?i
z2v)dxQ^8n^s^`eqBv-uVO-yF}c&#!vGS1N8d|D%i`M~eJjUT<~2E-dgl9<jeacU}4
zX_>P^&r^Trc0Xad6%2BcK*O_ePXAFv-X<a=y{e%H7{2EZ@BRCLU+t~L^XtaaIh0e+
z^dXz`i@gtBML}@RVNKYmTiF;@XW-~DHh$!z?E38II8C9_a!fWF97znVGuHB&-~TN*
z^IWq2IyR}{qVB<KNTJm{)IdY$jo3Xjk{3!PUUDA3CX`%Yx;XzT9ejUNAPz0uW?V}`
zUP3E{UE6u|)?3(e*PS#cCOG4QOS$l6ucCg*g}8+aaQz9!+IBXAmQr3N&?<~bteMIq
z674J&wA`+pTX$@3FrfUs+Ews(7Y%dp50N~pLZ8+Jb)qGTo-iDEdYKC)r`$q=Gvl{+
z3?p<F%dnHPK~6pAB@Kdfb=Xg+b=sXp{4_eR1r;uzJovDF-p6itpvYKj$bb|HIUy)j
zr<g_)w(n&BXFkirpSp#yJ-cB35SP8_Ev)&W*TDK!Wc4~ouxIsNWB0x7Se7IGV4qx0
zem0mWo$_U*K4%TYv;U-4+V@l2+BbK(3SX0Dj)u3~^wYn@#w~kkWroZ_#a1!dBrm>f
z9sm3{@1sH!G(mmHVhCWtKtZ+CX=LB>UEQaimUixho*d6QD0;s!=GS_aM-MQ3PamrK
zORD;T-b(b2XL^%Pbz3R*gK{TQPjEkKWcXG}rRJEp_4C~Q!4I<R*f@hWVIoaAlx2*~
zTfk-i^(Q&)M{cHRElFQ3Bpz$DZ7l5$o$*e&MXLn9f5{utA+g|hVX9E*nuE)b`dqYW
za8BbSA&R5(@Y`mF6$iry8NKID9{l3x*}ZuOD^{%Mf>*qTB`<jqZrKuiUlmCp!G>Z*
zP0@w*w1eEFjT{O=b$JhK7dLIc{31`Gi={qW!PaT7ISp_)o}=oNPUngA<mO^Fs_klg
z&XvGiVD2oxyPM};jCb0314Xj+G(e%OQ|S1dlr3vZy-=NQ>N2-N>FJa-It_xHIR|C7
zML$VJRqX5sFfVw5wQ>HOFqCp0VmN_UAv59LwJB5&?`QaHU+0O>f1aJ+ybo*P^h+<{
zqBq_QFZ==AyaAf^ghVPuGeP3d6EVg8mec!GG~`Z*VsoovO*(|(&lPn}=cU><i%7GK
zO2sna;7>pH?|k6He@AU_0ppDni^cd1>qoi&o=>y9-(if!yKoLn)O={!6e2b@Os?rd
zzpKUAY;>@~Deh?@hS{G<Cy|<ji2SmM{7J83=mCcBRj}|TRsAbqVQ)2_KR1+~z`y5O
z@=d?7MGHWuk$xNW1<i?ZYFOC0ov*&<z1U4#SXDJl9z0BIpr0L+4NiOY%enj)-vO&m
zBdb_!l8|e7qbX6~8n>tDQH&frgVEMBr$FR|no6W)c(BHT_oz6WgCr(ByENEA8k^vK
zhBpSyJV{HD$p$>NosFOV6kG4Uhoq8l!By9A`i(cj`qN404WjibNl9oN&`pV?Q7*c3
zak7-HMZ0}%x<4vY&&6z_gfa*zpUFI3w)>l=(M1$aG)OVc5siY=^4FDVC{XA<IyYUR
zY`SW4p>}S<D&174^431R320f=KI^toaO%6|z}-Udsksg1e_JC{3JI^J_Y!kN^7WA^
z$fBqs+P`6r^|(xjDyH}p&rNZpGv3i%jzcA?(bFlO%I>F%hD7J+v?Kz~d91{26fyC>
zDP0Q!-bip;0>mqo;RbBl%;x|5FC6~%ee8MiDHff34ljJ`TbTFqYsnT3;u1^6R03Fc
zal#PuBh<M_<q5U|h?XvpB)s>4?5FA=roB^7-2=REan7k;k>ZuA@aVRKyz&kIg=BCk
zEgznB#yir95kCHrKjd}Sode=AT90SeVdHc&XGsbsCQzF~1pZtr{AQ<xO?~2YUh^Ho
zx~`k5^IkcL6ttrv@^d2cFTF~n2N=HBz`~1E^`pRrz18R)vvoL6JXfLI;Ro+L#u~CL
z#k2%FF~W|I{XIKwxsBzm21%=lR^hPsq-U(*+IPMa&b^RUU!7{DlD8%C(k>u6nK>m5
zLq}=Z7I=zkw0dlViP%~TqCtc#b0Nvr7%(w#&9$&u24hEPJ@^2ReeMf9_3(C<uRDWt
zuYWm%FS&|z(GXIrU}}+Y<Vzb<UTk!^8+B?Cq$3-yZU5R?q|Dv5=y)w@?dXKJryHg4
zJn1lz?XY<iF|liFI@|3!&AwZ%h6{(<7*^%ABO9Pwe<7uROSwBP?KIrZ4!+E}k*L!Q
zR&tc1`9vvC7cTv~tmaiZ3t%yU<0e*6=P`+G&xZ2+gGl|D=8Vjm6*2ez{_!21wBuxT
zEkvH#(D8jIugBhTO&8e%o$g*JxcvHx&yPe(#sKQk#NeXUXsxBN^GSAp_A@+o+ieW*
z-pSzVbzE`NkFfAnH^AaWG%6NbsRT<`uk9R@{O9^OO*B%z+K-#&txv0+a&p!&2Mqbz
z88Ku&W5O9;c>PU`GC=BKa;!nW6=+QI>T9p$qksC#)cX@MO`=YtZTTB)aKm*k8nZS`
z9MbBkF34umEX=gx9m<*G{4|~7>^0RaLT2<&;OU?3hehOPMdTa3Dx(J&zE{A)N>%+B
z@Y>#LJb!@z%xpe0g^(ajEiz~5d=<s82V(-ywmCUY%_r!$H}cs(d=G<9?q`WO5-W@v
z;VGYS<=cOT6+d<}YN}{|Jy-_jPD*)mw!1|?gRHjP&<%BGJC8v`1D)A~V$fwC8}$zu
z2bqPYgJudxcJjpSxAW*7Um`m;$+?%lh;v@^1~~T&(gj2KKHGL`iG^hFr4OCbc2JfS
zoM<g7-x8%rx{F4gF@&sR$-B;lbal<kKdB=wO*^?Mns?o&u$8n(t0=6f;%hIM;5kh!
zhTfqR;H+@3@v#uiKZXaVfvV-BP}RBcUzS1^C~H1Taq-Xa?~JU~#J5TxcP;=JExlJ`
z5MwY&f{E$DCaGX56>KHJ)~cAKf~|&+nMxHZ2~;XjO~59Q*dWF=hITA&=tk$@H4n`T
zQv_2?3^WQ#LncVnrj({R9eEgKb$-|VR{rz2bn+Q07>y@MazroF%hu8E>q<)3=gh83
zb92?|d_7YB9_@N#1!y)D669R-faWsg5C|71JqlD?85}*rvD-e&mRoOS+XD|WxOxre
zz4o=7@ndg-l}kwllB8Cc4fyaNh!m)xt@2&dmAS#Q>Hc;~_VTuifeOaJxKrNx^S{nl
zH*QCyhRYmPK*v%R)ExJH<!>1rOpr={&~5mT4zDo-C)agIOsdf$v&I)@5#Q|sP-mGo
zwC|zL9Fd|?kw15}D+i;<IP6q!ohW3V!*QF4{EUd~?NuZ_!0?<uy!U?rysftqz2kTw
zV#W<{8V8u-@45ZI@#sW`WDXjK*!rQ5vF+B|St|){B4yGl`$SoE%~ibQw|)a%x{R!v
zP%$=gmrKPGMb=Q{+IC12m<~d8W)yS+jj&P1$%G9Fw7%gnlEU~fJn}H#`ON3o{J{MT
zE?mL|uXr`9uf2|R$#VQa9b3_um)aIo_@Xdu7usm21uCU(R-WfD-P}eCKiBD>f;!!s
zo|f=lrjxXbe^o*Vlq<caPDV~4yYM-s%7uj8K<84(B8@}yzk6s*LaPO1<M{Dmn#031
zM~6vAMwu8nMsswG#^@N$(J`9i<7AUf(uoONqZ$5dwb0asB;m}34yupT5cMIiBZ3T5
zT3_<cu_$>Aa1l@@?J<Q)=}&8r#3I&!wU|l(38q#hsZ_CjHEds<zM&!NL-XicFrU5!
z^BG*SgxZ1y^etEf3+6-LAXIBm?F#_I#$wb8PD5as5rZ+NnA{Yn%6Hy?<=R<G?`_sj
zeTRUzM78QjAaCCj)85mi31Lwxmu~3kfax5t%8B&L=TvBSxZ-GZIr1Ki;*6z<l6WCG
zaE!*iU+1Au{0sXYdJvi4&qc3#9qVs?BdlJ<WKBq_eL>@(K14&9*jb<QMrX4X>&%Bf
zRqjdUXRR8dU9Ca!G@a+;|8zU=eE&xn7+6eeqD9pRnpsRX!Ux~=)4cx1mr`G{2=y7{
zw5+0-j+0W+!c{5{!wk=>Qv->PWA;{RhP0<o#beVeS(f1htQbTLHf$@wFeAR#wkwph
zDVinGcblYk(nZl9F~+>TSAFyV!*hMQ@J?0zU%i#+frIYiQ_8|mnWhk!P5@i(W{V<@
zFxGdDbfQ58L*uSHx#Lehz|x~fSx~Psa%6-NQ4SCF^Mmhx7u9QTAgk3dNu`}=EbT<)
zl10$e@}zJUQl^bs%In9-!?(tSBxdobv3NXLs??eu#>W}`_C4JHum8q@$2YU~>~py6
z^>3hZ@x`>3FTmAnRN`iuJ0~fM@!%qeDboWw!=u`?n}SXg-&0YX6>d(sn6b1mcEndm
zXFwGtx1~ts=(M+NbLd62lL}(N;{$ER1?o(0O=(;x)<fm(VXO%g4JHm9Vf@fuM)vNd
zb>t9}hYvA+@F0`JM`#=y!8Mxr$p-3E$XqBsoC|pqJ_t(<U`0qoumXu#Y`l&P!Gc%4
zZ*SzZ{pd~E#j2Y=RcMx_D0YO-$w_fupy($1c%fubpP@b!$xan7Lgp10(~2{N3`2^6
zz8ci(B=ZKT&tE|QvL(!0v5ZBlS2KUr3WiQQ4Tc7x-XDO31#c7|Ir(Y~#w1u{V<$aY
zU3wF)Ti;o)7MD%wb5rI{C9t6-9e{LPE9L7-C__sq-)^y75~k_89>1WSO94GspwJNj
znQLBrm}(glM7mCdEV`d1sZtpor*+3Y?D+icJo1&Vkjxw6+*iDu4L|W#Shs?v5t3>(
z_^dibV<t**{pGrynTRf>shmtv*_@(hx`4|(Vk}0W<sFaj+|Ok{_!br~Sx;-C37Vm<
zLN;-bAHV)ee(6Wwh;gH=zxo<7aTsgj^E_(4@`;k<V9}f+C^Y?EH8b^9N=&-2d1S2y
z)ii$)*w`JZ(jvtq3ASS6gfxV9=GT1YGMw{xg_>9-NrIH6aPm6Y=S_ZHMBdx0N_v3d
zxe6BEq^chW=Ji&ichJ?PnbNK<cNn$HT&Eo=6gQJR6wT0THmUjw?Ay(~@BaX!-@KO<
z>aZ^3i1F;rl(Sy{DlYt`caYWRLp>o88#`Hf#zE|wmBJA^Qgb^V4_ex=yNp50e2$z=
z6YUz)Nk`3iIChZHZ+@KzZoQSGyAN=|71yxghhL9hy_&3V08>v$q!i|m=Szr85iysw
z2|0l9U1d{Ic$FPUV)Z9QlJs=l<hEN9n0EKMG)<7Q=(WsriJ-}wzzG*T)+s`Ucps@j
z9-0jpK0>y4Kf_Nw$+2BearDVujP5@`HatQ)F^P2{=uT2cMS)gEy;31-HmO!ZV3ze@
zjN+YQ6HAt6n8e~;u+DVOhY5!&)(AfH7?Wpjg*Ed<F&k_ji?A_m14Yg0QAOTHOcj>S
zt3*m7bR(T&NK@8nX!v)?+wd4;aNZX-!1-%WYC(jrR!B34Goe^*`i#_j8d-`nLL+mS
zzB+yL=Tl$2n5FC2vGlC<EI55F$-1?$U@`Rd2Lxae(g;cri`WDsIlZfkwdC8tA$P!P
zJ2^>VJ52H3eaTL=6BMXO2S43Y$}8zAg`ZA#$B7n*vpuJz)X8X1;esVA-WNs&SQEj6
zk~qN}J4$-rH+kff|H`57Jb)YM=iJx+5bJ*A&9HO{6V_6#)`MS@^KFm~T_XH0tf1X`
zs#EW(q#6V2)Zj9QwJ_-{=Unqcq-F^391~;XRBHoh;}|bqzl7UA@fY0vZ=Yo4nw6}7
z*=tGDG#E=14J6l2&qgrbw6B6wRF%zX!ccND%LH7e)F&Ej|Kz{2@ylOg;@DB<^;H=>
z^DM6Zxt~YYuB275RE@!V#R8dn5;dTWKtIe{fsQIQS_1<lrhVg`<a~-z5&7w!2)YLt
zo{M1NGFAN>;G*7Ye1A_4W)q1`-OOWdK`ekHYc+`BI%w4;k#21Dm{^<yQFGuBH73WX
zn+k{i-|gK0v5&A~tVPYGRO&Uh4<BW8&1t;heeZ>{*V9TW*lM+SQKtMLE~X6m?~LU_
zsWsZ^Y@?+Ds9g+(*QXd0z>LSEO^<DP7&*+|+y8@y{^NEghDW&g`s-N#=9^%}X*B26
zv6Wg;UP<iK`a+-?-ETC!J~d$-!|MeRA77&>rUoKiBOOeUvmiy+TsrI=JM8=8#^_ZG
zeXtaEo-qU~uUsTf4HVQ1#w+3?FmN6w$8q}(GP&<bj%?Y=@Dook{M0Tc4<E)gCb6lf
zB8IB1P&F3o72|`OiitcX&9cymP16v-=K~+yJ0JbrgWHU^Cb-*}AXHTmQtxmGnR9sc
zq*)ryjrw@5L(vy>2(}QSi!m4z^a5gR5Efe#^(gUkIk3{4cA$9Y@!p}%;WLNJGSvAX
zypI2g3PvS>Aq9(I#bAMIl3;;~7>p{`f=`16M^L;vnynO(;Gh)#Vo`)jy-wylDUdo(
z(`Ae|T1=ueq{86xWh`8`n#CK=WZ8M=vS9rxWW{nA=!YZ;C)gS?F`)x)5=?|Oq4^nu
z&f@ju^QoHGa2?W4FVBILDIh`nFeT@xFrd-tg)(z2>fppIDQA&Jb^g7i3o4{T`6R{r
z0$6xeG%<JyPCtoHu_I$l-2OSXefHCA{q}<-i&t{t>t4sYANvtlFrQXpsM=cK^Q&~|
zzok5F$)7WNTwktdyW3NsNgo~L@EHxCaP!-KgKs>z4aJaV8Hw$~Wh2xl_wo2Qzr_5+
z^YG_xqcXIVb8dV&EuUg4whg7@b5{^d(G-$X?hPf(PD#pNYeSU2Mw3T=?|-rDi+8c8
z){nCp{fVb(>wJCqDF5Sc{*G+JYNTQSi&drNJk^n5KK=GzAUQM&%?5`ihdK9+Z{~`3
zzBBa0o;{bS?}*4xiOBuEDy9b*o-1HsfvWy3@cQ0rJpVw%T)!^wnj181nE8+)=dVi6
z`ws<&YV1rVo@^o|$sA1V=ic9cFXLamm$fD#&6Lr4mCd80T>f8wlCys9=Shu_3|88f
zedUf_0S2b>^otOtT)~icmtsg;+4dsPG2(t6-D|YrNK6U`pJdnVxAX1Kf0;!3xaj&<
zvg)<3CR?@ySF==Xx7$&iDHv(H9VsOm7ajDM{IzHx-Xm(`Yj#W}b|`H_P$_6eGxc)C
zmmGrfY=~GO`ZVYWjPn?;Ko+`X!^dFfPKLK_;lSgMb7cEY8i$UMv>YbOsB1!1g+xPU
zgy2x;5u>45n+QI0ftT-HZ~;Q_VsS#G>nIsQs=+a6GRyGB;;h9~t5C0EDm83>Kh=Q&
zDgy&l2L`DQ4pJYQM}25MwR!WX4-HZ68z33z!_;d?wHidoNdl$~9Ebpcm{dzCd3%}%
zU|V<{wGm`41nxCkXroCsF-dD;oOFDG*60|mv2oIgaT+5dOpXpSd2E>0*f`nvIO*gB
z+G?R`iuFp>fH;p87)UGwCIB>bDaJ!Jf@$X*i3yzhN|Jy{0%-C|BWuz^naEnC#?wj+
z_2tVLTDOYT7oN|ev(JVNXTgF6P^*z4_@qK=OmzPVl&+#McVVgWRPx|*6LUujcP=fQ
zyVR7g<2LvW83|H^0LwF&Bk0L~o^`0fBPfypoWm)1Scww{QPfzP5T;@W4>NYpT|D%^
z|CzC^kHXLp*ZlO)G4D06BJHc=2CGy^5GBz3jFg^DrCE$9DkgIKrv}`e62Op0N2uVk
zj7mkAY$+f9hkxhyKJeGn2NnfAP194WD2)SK`P4ssjH@m<lLR>M;6ofcdX%$Y`5Ic@
zVQSUbd<oB4!OSTz@>GBYowgAcz+uy6^ljY8?LYZ;&g@&r$fU>3_sqAB{^}4LkBzYS
z##i#9-~BbRIGIeukU5X+eu~@Pa1)m<U5BKOy^}|AmtVn^fB5^jzNFB%JWKPVPl?D+
zipa5E)$+`aq_+eo1u*<}4=g<YK*CI2nd!c+lP;qric1i>`p74@ag?hrCM%258kwMq
z(){XI`O+W1pS2T{EV4?YImzLQ<;bFiy!HQl6xOUKttM0|34WRvSs3t?sW)X>fR+ku
z^)Y<NgKgIiMuHuxkYQ{>3y-!OwT8mMJv{br|IDT@eVKWKLtOQ$H?rvEH{zErr`ey7
zh)|&&jMQEuE%Y751R-y{MnPyn>4-TGL6B?kI8;2w=iKthht2IZOC4e&Ad@^dsR)%8
zkhpOhj~b6xhtEJVk8uQ$kSdIgaBTYy4nO`VhaTU^ktep24IiW0Xi-yTz*HE(u+S!0
z=Sh+ZF3U))Ei?=YH~~k3t5)#hX_3+L;H)9_DJ|E+*-*$QgLNtc1N1Lk#G>WPS-5&N
zgUgmvU%VK*d>IVQgL)0B)lm3b8~)2JY~%E#K&2(Ip=wvjPzbqaZZA!@rFRJO%ij{a
zsEI<QIPp0kj6fr&YoU;4&}_l@IE;<MfkR9jKFr9$1B@Iv!0_%p>_2dj@uSCZ6D|6Z
zgnltp62YorpklEthe-t2OsN32)nd>Hsn3`vNi^)SxI>eS?|+oXzqOHZg)ueA!uc%O
za5|@7eksc@IER7t>%$}hi?f1D1htmLR0tX9Q5WK+5I=iWj1|-iDq+eZYEkctZfaf}
z7uLWs-8A9Eq|Y<iVjl0$8J}NL&dg{Rk4ih^!Nv1t6v!M^8zz2k*<$9s=5<_q<ux4o
z=TEW!t~>d{@Ba})|MDrW|HZdcdGVz*`oaV>u?BU)QOVdyWr^Abr@knpJ6y<4plS1r
zO3>1Tv206g5U#uG3V!$fe}Q6g9&9C{l}?gW2D$UDd%61Z3&<viS$4_gWMBO%n?HLS
zXW#g0S}qGl04g1R#`0a-%VgHiwV%foP1t?^L)4uoYc&uHE>)6x1;o>D6ZUV}20l23
zNy&mNQ>7x1H5yn=saBv-&9z`5k7C*f$jLcguc}*tUJ`U~!th-O3qKpf*?LFsCYq**
z>ZJ8(-Z_in)8gw?@zI*myAXVox)#;c!l45^@V@smde7HcBN{rieN~=JVfE`?%f-L>
zn`9<|z8XpFh<7C6>TJEp;98+EN1^yr+SHSrbP>4u0!iS}Ya=6(3D~oZowwY^BX{1-
zk`-&X?A33;UUnJjqQyu(A+dRJ*tU>Vq|?f>;7@C5@*#2Rq3x{Nq7$dFD=Wpulur+Q
zskrGhqK(TDK}0Ur4F)I<-^wuRFfI#UG&uqL_OO4`W9)xuBL_D>!Q`<cB+Zn12Gv$d
zEvX=xi!>QWqDs{!sEfJ)^?~cHo=hwmLMv0o(-sp7lSx8#-XMKT7BX+yX)Ic^nq_O&
zuxRZXlI1I4!2r~1VG58>0E!|?3QN*LWg?}*s*Sp)T&KzmRsRed_OxJlN;30Ip>Q&_
z80c;$mW9U=IC>vaii4e3JaHKC8ML<qMg;%>AOJ~3K~x-!kHY>#9Nn>%qdRtSVB2<v
zckN-~@L{CUpr%UQDTCDt2}++e&`e3x6*xu2lV+`OO`8OaA;U4!YBG*vLOt_Wu3+5-
z=d<R8ms7j+VpzTmti@HTq$a^635iKc*H)NRl^rsD%qN-BMKx_;-0nKh)zj$YPZ^F^
zXy-bYB4{w`GxCDkN<3dCXqj5-sinYEdwA-WTX^i&TNppOpS71?#bs~*@9=_EOeP7H
zN-eC9Q=CSvM6`I#uj^?GT7C98|Ee0aHZD9_tPxtyF`<U@u6PZt$}&=ojRl|4r=zS}
zKEPd{`zO-Lqg1qtc*~<-zJ+?d!ulItL*@lrtKv|^3r4gnS6;qr`_|Oac(Ac{svP`>
zPw>c}{Ta)X0a~U)tCF#FK|cvKb{rYz=<>Dv@W1{&UL7i6jUhuZM-KDZ|MVKpA6!W_
zg9C1q#*1IVmH+cy_<98qJG)ud$#}mZ)a|3ailsMU_-=uP>s0k!y_NX>pD4@}wAS6o
zFS|{YJ=yb(r!W@O(AX~Q7{$aes<0t7Cq}7}(7fYrKKGtKV8wWY6)r^+81L(6dtaTK
zKKMSk>=IIAFtrL0+&670yGneiQ$O9qxWoTliY~syPPiCTaLKJr$kZd_4m;U`{X5zC
z$-iU!U3W8o$#P!(6K_LabS2r)0$PK$AdXc<wQYx*3tx+gLJ0tLh%Uu<<q8)O!fg<c
zQw2KgFWY|jZF&_s@gqzUO!yw>6q6|?1AK}ZpP;d88@ukikA3&w&*AMmk;zG_2z52|
zYj86#SxPmrh|2;6MnY-qsG>Gu0%hDgS}L^EG3gzx#E}dRF@M=|PFuf$3pSj^g7s&@
z+Evjyv=%5oHl#SIHDnPGXgTDJuFL?YG8xKNw;k82bQKE6wXy6HHg|oZJr!b_BnsO0
zS;aS)_Ur1(s3^a7%?k%>$OuL(@dNRkTBrd-b!6F67M_0=3l-Mqm3k|K;St!kkCAO#
z*}r8odpB)n*RH4V%?bL{Q&B^LVP2(zt<<O_hSp>gn<^D`EbuVjTD)4!kx`C(;ZAnm
z{$&oQDGbeL^`#fG_9d6I`s$a!%4Lw$QNm=$iN#pkrt)d7fhY$E=5!P~3<R{C9DbZa
zZ@TBvY#~@P=^=EE;A0B-D?htp;GoNe;c?=SBnDR%^vu<)dD~C3=H)MC)8G9a`@eP%
z|L@0tlpnb1CN})UkHLyXw339XO%Sz#ni)gRLW8Bu@h`y_Ilh1Y@dL?@zX_-Y3ZxZ-
zdc~TAx`7Sr*Rb))v2bnTg8gh(XY<yl7#^FXR;hp}RiAO*jj!a1FMWn5K6@KyzU+0l
zR)$oQI5my$M+HtXaL^eh4n>|qK235NOV2r%V=64KR!|SMY8^=`G@LS#H5pvD0&H}Z
zip_)Yj0~Ef#0p+pWRnf7HBga~c5W(B^kfp0RP|k|x>-c-=v6ko3B#!d3oBLiAApzj
zR-<<uw>y`IZ{=NBAIjr`+Uy0a;u|T_beQxQkNxqVv*XjZab~TLbh3f(OW3W#!W*vV
z%6Gq;w6Bh-CFK4ZUV;^=l&VFE69nj%)af(@MtegJ$;_d+#u~&4I_a>L7VO%}mVf<s
z9{k)LEM0Xv7rpNFEWGwwnv3RRtAdRMI==ne93hv0Qpj6gWVCyiI`an6VUrqDjE$L6
z6c-kuvGNoX#EK%|S{YV6(sJPyHAY~^cJ_YfL3VuWUWT9C0}~VUXNJDiQB?yjLvY}-
z=;i^&hQdh2;;kV?8BJ3ronb^XDhn4-UAdA4tJkyk!t+^i)(fbvJq?Bips&)V1qFXe
zL<~OI7X>(3O0Q<R`cXPTLEGJ@?c&o8QDRC4!1SZ1XUBU+=a>Odrm1y)TxLu}SJA@h
zD=$vGmdHGj)ko$+f!@quqCsohRt`R~g9BT)aAeC?4nO%MIyOd4m3a#DDni8@jHV>2
z;4`d+R^~}+6^yMh(VV1VJV%;MYK!Kx>iqLL^Cj0%yZjPZwGy9LvTBV~ES0E%Fr^_*
zm_)=5y5tiGne)zkyaoo$9v*oznkVfhNIS*2?5Bo?0vi`{95Q1`J@k!D;_thk2S52q
z_TB#g$>1QDz3FBa-*hvrd4lh+QIQ1i9mWe{+t;&1u$5=UbcWGAv+0S<HeVCLd54MV
zFb>Eh<#*rvQU3HFZ>L%tLR%RXYp6<{iDP^C)Ia<wS6+4&Rx_#=T&s!A43B={7W#)4
zar(8dB2APevEjNEgHeCVO0(lbt<qE%B@>}?WQ;HV_y2}{Vhal^eI(vd@2ipa*SUM!
zE^hj>4>JGS7vsg_@l=cnE`MVq-1f%TG4IF-i6k7D7-iLu{TD9(^>^Zp6gJK1v->;(
zz9=F;-V1f>0ftk0c<(<1{6cRfo<D{E@lsf4>dbYdH0S*JP~DXJX3!dMgd+7j4{+P>
z{tk-{9b`$Z&iKd}jjFJ>QscFM@H<prdIK$ENs{2WVLYW!pqY1q=^np1pI?30_;cNY
zs0%p?qGT;cy`gaU0Gs~plRWnMFEDRtK3Bf}CI(-6E$Q-=xGE7S#QX&9taimnsQ}at
zfB)R=yS!KAg9Oc6>e^YV_;P1cP1~|Fw!*#mLJMNOV44bzCVt;j?0(=rcHHwd4s6|o
z4v)}p3`0qcBz4rSr8PcDf2~5+awL@`2>7fe16t~7rYU31NhXY;Z{Y$?JNq0~op&C~
z&b<J8&IVXC6nO8(fFw9CVDm|VOnZ%$eftYymWH(KB4N4Y>91YW1(oTi4O2_^&W}t@
ztDfOn>F~ELjR>XzAG12DcZh;z#uo5Yx0uc^GO_!dQ&LP<F<a=cU2cgPlgGUR4y!(t
zwOI<OgCj@Lolmj<kw@9R@nQCD*+Jv*K?c>)=Y_#)LgIuzBlu=YqTtEMGDzw*Mq5qB
z5Dujcl9kIj`(-z9#`Q0ucG3Azui~*Z32sw~O;Be-ht(6VV`*ot6d4a%POP3j6}fE7
zD$}I0PuD@7%hZBtrWhSxFH*Y4HR3iY?sM;REmWXeS`$*`oei(jW{N#LivRlE-2E4S
zjUG8fWBx*J{MC10UwS#siosN?7!BDX#dJzasSP?C@xQFr>U{Ud({jv&8Pr5g6b27C
zY4OFcKgjEU=C>FcT8?X^h<HS5kdE=6f9QJt=>K~M6`cqZP5~Mjc*}Qg{WMG0p3ch4
zuOO=l#u&Vh&Ug|g4RVV1wwc$Xw9W+axwbOw)~)>8Z~rFEolmm3+J{)j)?>q*{hBv%
z<9pssYE6+fVM=a$vTrvNyZ6u<9|t_i8D}%Nb~S2DX8=TpAnI9Jg%2BJez8~G^Z>)D
znlAi;s(z@q61|&erqTcOK{)OB0~52l#DdE*q~S4>qip)fU-8suKF6w-$4@qB4OZEc
zW(-_*9@o9+4{<}waTTFzr4Tj?T^{nWpbL~D3Lp9Du|&JCUlf}r%m7-Gj=EJixSK<t
z{Vez2`X5v(eO&SSn;3lQb+nc&Lh8W|(`ZrMM|;UOeJ1C1)6tb|cZD_bheV{KSe~}j
zjdqkdBizJn1htSk4>%-qB+dl~qa%kn^4P=d`uexn^Y9}y4j-c0bPQrhyr*gmc!&Co
zsu5i3k)(oCOKK}L#WR*Vj`|j6Xg*8MJcBhaxRBFdcnSSyp9Kqtpprnv;xqsY)6sb~
zv<W@Z&cG-|acF5xjFG8DUOA&92M<!F4wOXec|BiR&zTB8=}MqDHM>pQl2y7AnA?=6
zaxvO=wklQ6(b86fm%WcNTpxj{w*Q{k2!JKmMEDRGCT)i=6^jt8Y9&q-$UGd}N9&0#
zY`gb9_HEq4v28o(8=GLhsW512hzl-G7LOXG<vmF?#N!>Fm}J<eOx7&xF1d_zZn%NL
zYp;RD3(*7`wu0CMYl3f<$pLbAoI7&-oqF1FCNSH?ZjK=0c+-hIq_1#MDm4m3+C(;u
z4qAaODY(Rt0!b?)*?WNf|L@;<?Dqd)a%7aXSHGC^fBx;TdKKe!sGBO{OZLd7obWz-
zQ>PPD=&V3t_O*!i*fDk9p@_{mFxupdD_+OY{8eb0lBEr5Ngbn2RxN;UeC5;l<~WII
zheu{DN1_RL{maL>=<*jcboNE0HH*Q9wDk_+m7KCYY&!5*))nRA95c~`@fO1m-N%y;
zeFqsB;04$J|LmQ2ykup4|KH!|IcH{W+q-?2U3yu%R4G!$j-Y}Du_m_IW7HT8F~$;O
z5{>>eiNx5^phyrAMJX1LCRLE7t+Gp9mhF4X%$ak}^ZottoHnz!Z2^_N^LqJO+?{*p
z%qh?F{d~Wl@*Kilb|LApnSHRSc+)vjQ345Z<XJ9W<nHOIWfFGhg7oCAOd|3w5&5s_
zDrOomJlSC3*{b?&VA1s1*zv%@gd6Y#q~N>*C>ZYsl69Ng?M>8T1NS|^RsZxc23D?P
zAZ{a084|d+9dX7-|B-<gokA3Z*m^zlMTfv-|B-!wGi96Fh#8~5oKuv%up#eGI`u(e
zh@eJsZN)ZY7#?EHPcP-SKl(8v>o)PMGtOZ4nWy7s&n4-vXI}G))jT&LPcwBHO)3+w
zN-lKHyy&H~gw1{=^SLi45K?jTk*tz3c~xb|Mk6t#zVn9@h(>6wSkB5{U(drg-oVfU
z%aK-_UbQqZgkqrGqE-ui5rXllz}Di7&`MKAonzDpGIKTym+ZrmXC2GTgAag(3!$$E
z#Nq-=ith~Og^S6kMN%P4<om4C3O~0}{N^3m4kkj!)Sx^U<(1URXrbz;Q|YQ!2LKgX
z&&2a4rZy!Qv;5lP9eeYFWsGW0y69IO&te6IYF20!%9k9BI)@ljhJ=l=^`*tYuaNg=
zP){*Ql7ZnAS`jS!J!^0M4G-ROJCEOcKeg5<^~6yR4OXGw8YE7TSP^ST9Ym_EYe%eV
zw&~w{FZO%>N$mBa=fa*lLl}DPi?J9RcJ#(OHS0<zsZct{gen~{LB_epQvnc+GhvvR
z#?uM*RlGh)(M~Yy6mia{aEtMD6jFjHBD(nwmS6B)R^D(O%$UOwuX{Z+UUDi?e@IyG
z^J_}#ys4RVf`*Rckuvk5e3o=dcB~UoB1J$`4^o{fDiLi{<B*eI#kyg?MtRy&NMPC|
z>mT9XJFa0y=m@R%-gN>R7);#8jE?fNZ+?}NPk%XPm)%HfwE{R60h3Q+i8k4uRyCxi
zqaoqE9kfgFbG`tFl7@n$>IKYQme^hkpO1eNudTEB9%VWPl_HEY@7k`f$3*0{BJzvr
z>Sh`+JXv62zN-E&)4GM}*JPVaMbO+)&^ZO`Aa<n9h@cK@zI`D#p8G#6k&v|2Che{B
zU~3fi-kTSH`rmMK=c0XeU-)z_=~V%&lUyF?3wX_XRZ_fH$!j8mkmg}bnFn^7K#;^R
zyq@;8zvSkNFJ)-;CiXq?6!v)4*|g`+LwgJXnW3aB3tN}j!%9U%;k&#PUG|i$qg0UU
zbMdlJrdnr3U)72ypI(SB{F2mRoI|1nV`6BohkNhi_t##_gSXyBvSuT-Rzkl*O@YK=
zaac_;C^it{R7h=|b}eAoSk{qJpR<5PhaSRiM;yuEflFZi9B)tRmFWI|DwU{gI*RsE
zD^e+n^DK)h1*_ZMzH47BDJh*5vsMa2-OQ8Tb(Nak&|qwy+P3)qZ^?$T{rja=;shNN
zt12ebVv<u*x^y$^*UZ7V8?aTT9O~m8{GIGl4`4@z;gOZBxcMd?y7?}ixaS^btY60r
zO&K%|0!^^$NS#9ihY4#mla%2&VqFvw&YR6XCmzQhFFY0Y-y8b+7_k9iy^fWzNT4nf
z+9|Cn$I6Z`zGubKzgq)RlD)=`2-~jA@i98ZD&<&BG|e6*L!1Spg->i?-DQ_>-*+!!
z)5;a>^vq*9^aCHnEuKTG(MLTDJP6EqXn88RWD~t2ow|4yOxCQ>iaKlRw|hrkjVY6_
z)6@|JDQ!Yt_0|t_^-Xt!trOu8XXp(r(b`8j|0|#6ImaAK7^Hs6B7$14TS|tT)P~k`
z<@dhE^IrWXl7R(CPgqPM#+&jyRaw`C4Mj1v%uFL(wx&BpoVT^c_=XG4A;wZ{G{r-e
z*<CvU65lv0E|bZe`6|ih9h0jkZQ62`h`dHbo|vv;rUAote>vwq2mJN)nb@(4|B1wd
zc?YfV=<c?F$u@hFQzDGPMG>Z*!1}dZ_pwjG(xvQDuQB}iYRsG&+_ioa&v?UYSn{T~
z5%)F-d+VfH7Mr%}dX#aVyy#ggej*Mb@awd}J1|-Qe&<{U9t;wA3b$>?*k(-f+grKm
zo8M;r@)aELtP|Pete3;?dl3!Pv6)r2p|sP_o{XIPRO#|6)wtw*vO*?wmhzdxSVNMg
znC$Oe7K-IX%l`MLn&1*4NF0&?8#b_J>2JC3s-Lj_-aBz?HZa2m3<M#JM$`j~MiF5U
zpvETxJ1eB5G~<X-lyt^ib~^YVmYjGpwL=bqd2^xOz==U>UWn-J9i_x9R48UbrSux(
zJ{p6j-*snf*43Xh<yAVT3WdFC0lvDD_&T%+<@?deST6e1QI(#wf~TA1Amd)IDwWG?
zc}&smtPJr=F*%;rM3P|C`IPTw4ENo~x?lgA2X4BJwab=KYmU;3W#)_q=}3#7%w;D*
zNW}2O=rHS?lFpySk{6u9L1(-ayT{I8YD8fTu_0jqn?Pj05JHj-)Lh=NHJ<8cw1bl7
z!wmEAL#a%jpc3z1DZz9?N2M+1`cLJW<UlU7$`vI`*fyj<5GC-y3hqAlOFZ%Gn`v7(
z?k#Vj?~Io++8YqmLIU-<Cb=vqckI&Qne^+C(R{+00WOs-{34wcx?<f@uZ3wl;oS3n
z$S3~&O9Z{MiL3!9KDE6y!r3o5na_RveOQSgHQvQ5us9WrQB1Q153S((pZ<hro&6T#
znSDq-z-Z+&bGNN*YO}on!&I(o*@p0+E!n!PY1Z7)6eA85KdJM{?50>Ry9<85Z+MEJ
zN;y^ToH%#SMW57_=>HjG-aB3SOaq4PJzaRCs(y9)OzhY|Vm!KVhl6Ux*3$^!3=Utg
zN5d&~V@ZB>9l!eA7nu3P8tSbP!d_tm0T0&$&it28!ZQvhs)y9;-oCCV47==Pw3=R=
zhYD$$6%9TI0AsRoxsvNX#rm)zl8CS!!ScIU`hVYN`LA#1z(Y=8k5`?A+jS4pfgXY|
zTfAMVIZ%;~s<^t1RV@M~3sV!-qPM4_K0r-OCh*JXJgIu?Nd>GBCrLq900-;WvFY|(
z`R$cgv0>>l%t%C^v4ri2dJsa|#;T(p2BdL{1Qr*BwA8RJZPRS@GHdVM+2@$!*y-4#
zV3(bJ*qImzGhmSAI;cu#cR#<b)&Bh?`?hXIA<;1?F*>QmrF3=ErD9mFQ|RGuPEXP4
zrN24_8lxMal`gMpISsR!DS$CqlYr9ZH_e=aj5<$~+qjz6-S=?sl|SQ&n}35@wVHVZ
z)U5TxIdvp>aej<rqqbSu9A#k9d=7uni`nheQ(=$YNUf!9Y6P`<hu*<e4&UXO1$jj-
z`Ey#ycaY4DO?1o#$+jjakKGL)`*)TTvaFGkIpSgzHHI`*f{mk$-gpzYe))^ERy@Su
z(MNOCyWRzR?!jnHs0E>4*HZNkW<n;Au1qY&H&XH^QKk^uAY(xJ0$~?g6Nj-vGmg1!
z=?Y%<s<#siETm<GxE)is28r4%n%UsCtItQ$cDWuRnS-+N(~{AM(Yx;8zTYq9u+z^b
z?&-mVwxsiD*%;~oe3cNCZJ{nr2pBS|U-_)~GnuxPI86{E7?Tnh3pT{4%LdO{Zdgun
zrX11bE_pdE*SQV>(-`SFPwsjpB5x9ruT3w8(|}=nfrTSg^}E1c(`RGHnk;lX_h^M@
znG5`i&a+51b}l<Xl8>^*5p5i$ZbR<*udlG|I~TL3rbrZH2YXnyaV>)<9?OZJ{1|DY
z4+eSx@m$fYvnkSTlS+kKs?{tgP^~x#IgALJ3SzwDNE$h8GlBIFu<U{hS$gG9*lnNv
zIqY?BfZg|{y<ip+`Y<Y+ZIB|SqyLllpGrrKJnNw9gk-WVVir!ModrZL!!X$-LN$gI
zBu*hQuxSnJZ@r0oe|jma?z)#?q{$3Zr$;RswW&2iL=%D_K*d1lvm%C*Hk%TM4ED48
zkw>xL^G-qb-ydcUpe95cwan_(VzQ1{b!&1JE^V2srDM7rZC&4cYr1FSzqfOP9seEE
zugS|?VRL;()kajLq-hk>#f;5evgj0*6iFRHsxUeV_dmeHS6t4r8*XCbqmR&+#?0>P
zr=}3bsizdCf=bGUR>aCE#_cqp!_PdOeP8-wShN5qmekY;!df=9%km2H=0wgD6ys@X
zl7AE5G}zL@ax$l-EfH1Am|!}4?wyvvK?G;~@dp6+ui(*de}mur^m2M<&FA2ky^Ohk
z_Hxo0J!mb!Pzu{UE5>2OqMCYJWl>Zu;vwDE7uOYWzW)p%R>!8M<FIF)O4>7zrWjIn
zh)XeTA#JYa{#$;;>>e*}uUY_$5wD@rl%Un*p=+)rO@%#9dOmS)4KabgX6hR$n$e+(
z^}k3AmoZgJ+o|~Jp!JD>jzrj0A;?ZhO(9iMLR@GuLEtqC*+e>ZzT*`-lO4M$jyJX0
z?{r#OZ}T*I*D)1*5?{X;k+VhQmg)6y8Zc}tSO_z)aOU*cn0}4jC#!TTvjx`VJWL=?
zBf=<wwX3=6oPQv=XE_UFLpy5H9ISJHE9JPqek*&v=C!mCUj%290M!gINQHx42`=j>
zW-CV{K;jf@GEuWL5Qq<U8*O8TR<Y^IE4kqZ|A(m$a{Qa$O#QH<h<f`78wN|(@ypYb
zrKCNW+}c#D$-Y{#b}TKMT;A<d?c8-UVzv=)R~k5nYc~Dw9$AMjUCM8-xSWS?z6G;!
znEn)cO@K{2$OxPg*1WAOwSbl?8`ZE{5(f9&gT0@9GP^$Ocx1Pop;1STMXb#>Js$u@
zmaCAv6lpddH<c=9ETQsd7NgtL!<$S}$^_s-Ct1oT<Ni*?aPv<O=x^W6Q8s(tbKsy=
z;G2ci8Ihu9FVL}M`XUE`D70Ge@FT3e{#t%_)itcY_dW(k6K3?(X<#9ZNhCpSKw^YV
zQIm~vO#g1XaO9aUVeo}7guy<dMh#c%Auyr0AJvkY=EMgWO0HAN?8=;IrD49gACFFO
zS8mM&VoLC}pxJqgWbfFPQNIktsSs`)X5?4baO;=8ie0yg-X#Ze_y_(D-E#rWdO$s>
zp{Y-V$9e!_#cR174v}CIKs^<>M7i!}$S`&Z&IX+P{MYcnx;AmGP87ANhnBzvw1yw&
zy#M$IUVPFq7!~h=CcahT#AW6bhBR&yj<&et(hFF$_ukAq^cdo>=1mp4XOOv|T=16C
zIr9^(q=s*MX;)0vOh#qJ+u<g6-Of+{>s*E(TS?H4uz}LIXdz3Ub1J)^@k&}f0rlKQ
zTOFyUu<}tJ`^B&5Y1FWRC3O+C`Mc12^l=bcPr>S5Rz;sGzyR?53^c^kb<H$j*p`Rp
zA5`_@(`RDGoHA_oZ{7Bm#bm){8ZXaLHuxet9&J(=f$MJM(ocSZons*w8pa50P{*V5
z1~}_;{{{Q)LmUQJlO_F1E+DSD8|h}p(7C5&3#@vrgaT<hCaBd&QinEE!l*@Z_iwoF
z%U@#Cibpx_)n_yN1ur35G!Lx@ghfYPD`8(1E5l4g*qI#0W&vusvqqXajIpS5sH*R>
zVbCBYjlsCoCoK;}=!3uKuAlsX71#fg_|a9&umLmcb&{bbJ-s0=N(jRmDghz^Z4!pk
zh)oK!_S%O7UvxUNo_8|LnTa;)w2e>PwIUUCMR|TJ4lXX^co<W)?yTCGc25^3+mxr2
z&@sjN)<nQzve+&r02+U~K;g;W5GU=L=b8a^)yL0kV~wfMgXoy5DUQ@QByog^^PzI9
zo?ztq8@cPMpR;P|QhG+3%%3?3QB0UB0g7`l;vDPah}9@dPJ1DTpM5%V(0;VT8mg8c
z?D0ZW#p`hJ+9_36W>A=XN}3g6Diy~no$$_*Dcr?Vc{+5AuDz<m8jCt5&FG8b6t#*o
zmJ|r%1hf26mj3If8M*gf)?+#1O@GeZSDs0`W|1(!WZi6IENPnftc#pNS!g<R66h{t
zrj``PO9{A~I%0&s`rA+Py`SGm*f)n(t4&}9QbD5)y!It0^68JfpMV5w1Fz{a*55mh
zEQbJG)Fd3*z>OFGFGrsAJc8Z#C$0y6{VG0FD}V)$s5-ucZkPRzs+F<xjjy&#sE=&o
zd#`y7b2mjutXLAP7!qr_>+#1p^OK)v{>xv4Gk!8{fK-(=Ke>#{|K@Mmd2o<8Nmw`h
zIEzksE+>8dzlp=Jh$bqitxr8*Ktw(wBA=LE1*ZYS)`Er8RP}p6&-B^YaUYD1@~Tp=
z75>!Z*#McDs8KW(TFp^<ZG(qC_a*N5_IWJnsS!0tume3jI5Nz<Q%>fYANeTl-d=2P
z&C{5wjcnDtPVbl~bUXa#^=uJfm3KcgSEY6<rJ)Htw2XVtKaYE_xrqZ$dN%vK{x67T
zEx-&myr4M~GW(94DI8ZiZ9Vf&Yc;33;>X@mB;^F_jK-1izSGnZq%lM>JpLe$Uw<{r
zuJ{G(mMy1|IQm5h)MS>HDfPOi4-wRffK9Qm#wN_(_W%w#<7F&3?id*CCpLjE8bph+
zb}qm}a<feFf@=JzJcnaJ9ypaz8*&O(X>_!$4DYSk#$}4&WHP`q4GNxwy}@6h>}IM4
z6{>1!=bDSBtM<4q&87eze5axW3MP(V!)hM8{3k5?$u(@e`+jz^z03$hqIR3uDM&(K
z1J;g=ur6uRyVq_!^OdhmmMwDt03ZNKL_t(z-t(RVvu84Dl(135iZ?vSg~0iQK{C)^
zXrKx!rp_mPykc@PO44LAvF^LEn3h&(>|z9U{`m}yMVvzeCB>4aDfRW67`fsKZu$1N
z(2*tshaAQc?|C0?@j}`SM~LwnlER&=LfMh2K7pJ25miI<{5oMh0957ti!SFKANwo=
za~9K19BI_{snBV}9`h{MU-3O`+V%}0yuK{W0_{3zknI-c(G^^G*$+7GjMt-ccf!?7
zW>ZT>d^4pkNh_)Gh4n7mxL#DgXJ*)86xyj{<kp+``8(gu&MqW178OTbEH(&PF}#+8
zUiSv}`M_Tj3z)=VjUf?8Z@h(T|N5_481#~+F&pEJ%s%C04*2*#lhkX7bof^lo7z)h
zFI#yvMdU0I`O)+WH})%-J_X}$2#ZwpL(^bk2fhB4-6Pj7X!ReoESt$)ADki<NE~jY
zMQ=Oe=kItg_U`-Gw=qD{93hxJ$Zywg;IvQtJ$=uAF|B%yAPmVVp}x4zMZuo8>hex_
zW%n|bPK-(>Ag;7IjWwiAUxYL^ZRC+3f1f)p`T+woXYsO+ehLmcfYDil)JzF{M5NGV
z<YG=yDLSM6TvQ|9T=z`DZ!_xMNxaC<Wzo+mE>cYD2!=;!-Sr!m{@`L({rU!Mrb}3q
zxe}>M34#FA=tXKZVo}z&TdZxjnX}j49R0GFGjPhWFfd45uhR~|T0@A<>I5lt5s7+Q
zR+^^Scpo%TM4Uf|HUktHcrZ5GsPgntx@8L!<$Z4tPZ^zJ(LY`!+m=j0p1xpzdm1rz
zN!>fM1qw7@o@;}kdA>%PfrAVzTe2xuHc2#@1|k*3)$6$V3t9B)H?ZjCuZ4#nVfjUu
zaL+HUq51emX4L8|=;_0ajIto8vujTe!w;|K_D_A5^`HAsmb~b>9QOKG!=4M#AS59q
z2x^EJ;?!ZSFD}!<RmqLvPa|6dF?Gxi{pPE}co)e#re!OMW*ITguRo#n_M8rd)T=~{
zwIp+A(0BGrICjt7x${e3V&v8vx$Zsh<><G)oyN&W5%mb6ts_pdfIkXXo~fvYbb<m}
zx~2_AeDLEz`|k~DggA#K6To9pEGr&<oDD+>v+CYOD8FS)KG{)Fmm=08wHnOZnM0m=
z0(buW$2{XjXTe}U8k#~gli6dHXS!mFCc3rjj*K~Jks{F3yIHMX_6W5&KoSRGiWo;&
zt9y&wRMEC8Xsafhpg3!Aw&62L;)He@(^kQmTDEU>nr4(<6Hoct)uXB(Qq|i;<gw{0
zXc{nV31IjTaNP8nm_`np*M*f7+Pq`uz!*m|8WT7rSauIT{m92zxVniokOT=&)I(M`
zX7I}If0<<GU5V;7YC$=+RRm*wOWPR1QF{J+I#O2T7OARKgm@WwDF3<PHnttZo!4^B
zx4y}S$JcT6pPj{m(_g{J-~ja=p#cST!fBa&n+LLy<^gGfh{=*UD=EX7r+txJtrcV?
zhEx>e9NHYgio>j3$*Rk*<lf7#WaxJfGMkV^3Z!-F*)KIF#Wd<Pv1~|$HF2And+*Ku
zr=QABC!7Ft=lFQ+nu1v0P0V=Xc_HUAFz2M?XxqS6YC2B|z<C#te3FCO!rjVMz+$;|
z=>$ILI?kU@)r~n(3AbZEvfJl(ZvLA(96Cihrhi}&aRtCuF_{<%(jslzyD9}mpq2kQ
zHQ@T{r2Fj49_Rcmd%WlEM8Cb4J1)7D6~Fiuz3qg>{c~_>lQ2{k)Rg&wFm%;Vx!{Kv
zvEbl?IQHzbsh{#};;=>12&e@;Uih3fBs`cG#C4U5<52KAzuV+JYMs(W(y~cjOf#h*
zvMUZ1!Pu+|?;O_oVmUTF=ut;=<X-!+{Oe!g;Y%;)nt%H^OMd?fc7EMoFw$sI>+Qwj
z8#&da@X2WYRjIq?-TNx%Ou$;NpGx9{Ab`Dg-G#y49$Xru&U&7H><Gn@CXV0y=C?fW
z*k_<F_3M`t#CovkP>ju2c8hED(Ep6Xn7MW(D{i`y-Jkn1zxEn~s=-Lr<JlF*obMN+
zTl>^(-h?nl{5}8#CZFD<7zcrdI2WU-Lv4apLpU(#kH=A&^7`k)C2d?_5e*?8CLN7@
zwoblxbx~XO57X;7F01|iYP$HD1`JaL3$ItzcTb;*r+4>pvVZHa3Eiw|?7LS5gH)7g
zXq0+bWA%j>a^n}zW#1qqZ6~CIz1*{I4U1p+0{-lyAEq7F2zqK*-`LLls6EfN$TQ1r
z5%VA>zuo&n!%7hXWU_9$h{G8}+)l8=ZGyE=aL+d{VA&17<j_-|&93LX99_68QBO#Z
zk%H3XDcD+e9E-GKFCKI*vt=#ZMY8v}(i%^lBANyB7*EGYl9(V7XpNHmZYj51bRm!3
zax3AwP0Y2HS=MV4YCSci#{vcyS~e!5tRC4w&(6DZ&<kF|31_?*=FEjgo!I)~!R9oh
z^3;=7uUs}&$fq=gu$gjlllm4P4#~Q_-uWg)%fxK5%nKc_cl}fp<N62s*{A2TKVE6X
zIR77}#@Vml4Y@|3$SM(v3=$*8TL@=GUz!D6<#a$#$5ABpG&#lL`g^fQ9>P(F9tQ7P
z!<s9v;`U1}Wn|eC?9|uKfVI?IN-as5+dr4}_dUdA=X`?Zy#M5+x4)gmFFqN<lvWr}
z3maG!lGJ%|a8AI<T^2e<1KN?cPNi9)<D8lCCJtl2XY*|8GOVi#(<@T~ypY?9pVYMH
z4YJ2yy_-FbJer?<>QgMg;QKsr=RF*L&N(ED4edrqXe^|aAy9`|!+6)W{GNHTx^%gu
zsq>Rf=VEF-b#~c#G0Ps=NLcIjY0#;XxR@YpaOKag<^?AlhI2{gsO2(BXT>#1y;g-V
zAh9XC9(^3Q|M+6&|9&aGdo3XeYZM|1DQ&B_&l4|nrncmII4x|P6X%$}WKTA^m|c1X
z8IB_AHoy^(q=u#j4DPolu`A+t$|QT*IGbSWhGr|mq?yfX=<T{iD!v}wh7(UcaCo<>
z-XbFBO|P2cuX)oi4;J=Q)r)`wrq9NX10cG+^~bbR%@#9Pbwd%wi4qM(^rSJ@{ryKt
zZ@H75Q%Acwg3cf0&b7mw_`bKX)1RG16oiEJAp4!#u{9O$sFKIjN+HfAb%ph)s{~33
zQn-S+)(GL~8iudAhU>m{A^ih8bKE=Lf$p_8$?P5<S`0-RkCV<tK(sSQAs=t6fXgx*
zB+p$CpA?;CbRt$r;uLi;Vbq2-YuRwq^(?*Q2Mj&1oW8^mCMi}SOdUoYwZ1xu@i_@k
zv`5jo3)$o36FK0lm&49G;cN}H25a)syOdxm<7P`GiM0E|s?*LncgL1mcC^MT{f)Q+
z2o)iDg-28sHcqXhzikj>BFfVr*Cb&wx1B%au}soj-loE<6EIM4D+-EEZY$i;?arr#
zTBhRW>q}9rdH;HcAtlACU^GSVyOTS<{eP^u;^!<*L;AuR4o8%xBvwhR@bKCV=<I$@
zIQwjNe&wss{yy3@sMY(5k%5Db^0<R9-gTlRv-PZ+&u$G!8|x++_g>Ko4U9_awF9gp
z5rd0jY7alk?VtGq&AV<ToiUFSKJszWL-ru5*XT7NX{v>RgXkC*$<?B@axHUNyP-<C
zGNg$j){{mf4S3s!KE)5NxQ<5OY$E4~V<jNPwl=ZP&U3l;vhy(7MhShhDVyfzfq+hZ
zh-HePqy()9S@keCU40oxJ?Et`Z#UAKDY7Oq3yR`wxhyqVI`Uq&WL4?RJt#~)RDhJW
zf!b)qweS4^Yp(efb9x4;E7-ar3G3V+CA{v(KcqQxfO_C*R#pTxhUPu@a@B|ah6Q~y
zXf_=qLqqI&`U~0r-S5iiTjd5M6QRwg?sd0_yj(=?ovxCm0mHa2=iJwUznDG~Pahz#
zMFyDC!8qy=@s`v2EQCxIr)^|7LRPQgiog9J^Vf{h7#hK#jMNSHxE8PeuXEv`B}6@S
ztOOk$lw5aE;1B8ad^cH+f@Fn;7(<$M*qj<a-EJk=;SqT3cij4o|6$cV%Q@=wmoWF#
zGl}NS!S>aC0ZdncoaaSIMYx<5(pqs1Qb1FD;jS(-TT#G*B#<a15ySzkdWfYLoX<m7
zTnXzoFvs>_+X<4zL<jW%#F7SvO>xY|Fl5dF2Xfr&UJd&n2!k_d8-uA^Y(BrKY~n?d
zb#|<k8f!DhsHwObl^~6d^y8v{%2G@_(~_p$g|?<(|6`mmOjP{K6ahq$DbT?MF&#hH
ztqd3&#Eg|htp$aq5Z@am(G*;p&{<f$o=3iO0k>WHBhoc%SukUO8DX7PbC@tuQUe=C
zBOXi~hrZ@j9QOK`!-9FFriQ82y#P0Lm2Pk)AktJSXPH`3@?;=QH>*B$V*W|4)2fEx
zXubwG)L0S`oHl6;MLh7i|Ky=de?+TZ<LI}(m02%;3GG^qAgm)!{RGui@@WbKhYUJN
zwSib}3R0ma35L|ASZhergjfQ;e&J=j@5BGftOa}0a!R|M5QsyXO-y?O_uTvw=JbcY
z>5xrGO#aoK77LIQOX?FE2%=He+;S6Z9(#oSo^uB2U~i_os-_`p#YHdE1gLIdb*EkT
zUzK(%sHUVMewwmo9qTT+gk?A0Otg9(SVu5x7RS8)4Y*^DLVJ6%=pnyPChADrDRqf~
zw2V%7Lfq3sXu|?9%a|U&e=y|m4P(q3rdRQ4z%br);ccpV?(~`1k&BzMXqH^qtVK>}
z!L2k-X-5$?0=V%ee(-l6;sCXzBQ0ECjmJb8*==`T_W6IOJ)<Af*OSqAC|r8V)}D5B
z-Lig>AJe+>_)?dR<xNIKO2I@e%+PAqUUn(J`p(7dcF^G*_LjHOUbGk;s8csx6NO^G
z^<L+ab?r5GO39`hT8OU8#A<ae6Y}|uDvn}mE`jDS{O%r>UUV@JT>C2qMq*~V0NP#%
z7uE#=a6yfxQHJYv+VkhJ?<vn`_tQ>;Me|XyS=w@*P+b^+bkzUd{M@PqUwHy2azmMn
zY27tH$S$~y`Llm4<?u9sc``S?NzZI<cU%o?6x~ioR^D^kk@Jvy<QrNprH&xY5t_fg
zi5tFiKI?C}o0*!huzx1e@Cd3c!ht?kjSjIsX|d##XLINq-vN8?LMyZc^*#*72X0l?
zBcmC4F{fFHb^r%)QchpHVjZ<gZJ|5cy!v%Zm!|O?m&QVIMdKy~H4dyHN*#^ih}JdN
za_tvBjjkPG@6%3akGH*rXvPd|f6W)^PJOUgeom|BI|n3^<w8`OMdGsWZ*h3~r8UB6
znsWQy%X#4$uV-MV{TM|_q6Fh&LRFHX)m-?E&+zPH4#kKqfS5_e_g!3?eRh`1kh!{9
z*T!rdV(I0VF>lvBnSaPpxZa?!iS+{L64zA{8MZiaVskelEh$QQ!nILIQcNNc$Jz8t
zAZQRLZA@Q?QUCplIYp9^st{UlkLyG+nihMHF}7U8s}qLGbr$)<l&T`~b`kmS=@oMt
zFm!{3Lsj)+VE5^>@$~O@PG%42G6#`P>jo|O?(xtFHEVh7n-_4~SO15DdixlCd=>S1
zgWR)e6EjbHKF6Q)QJRf_dcE#D<LWy}rqCedc6vojYq>aP9W710Gl-_C&%Lxh|5Te1
zVavgad-(Z(eUS|-ALq<BzYC5!nrQwkOz1V1xh~F0nHf<Q)n(R#R=i-Opdfh~P#P;%
zT%l=*6Qv$I*fhlOZ*Jno^Uq`5t+z8Ts4+9F6R#VlUJFp81dU#tIMzp_tZ~Bb&pMVP
zUi&&Y{26Gim$)8c3>mda^0`IsirBG-Z^x&7k|MjT5Q-EGWuj8%57-l*D0|@#13CU!
zC`mE_#M_by!{o<b5$tZ}^sn1+_9~@ZB~_YYx$uQtDp|e)2{v*_>|pr=+<VdYx%=lo
zqi0i--Dl5W_=)xO1O{sZHY5qZ8*a1sg{N@JU%dquFQOS(!mv&tf!8>QV2mllyPPi=
zo%oy?Zv$;QavVD6DOIvw1ul&7_(&Jz7?*4(m1#@X4AIQ}K@?mdqyk|aA-CSkUFUv@
zl{ekS>_eZy5$Aju-FpF1LkOjYrY>_sQ!gaWIQu@KID<K<>0EoH49IkO0HuL;qO2Jj
z<(Vg)M%up{ZEJ8zjB7`P2yS#8Z+q=a`PhfvL70b{VK7eo=Na`)8e?!t>W|C#&OT8a
zvu*>wzU&g7`J7WRi+3Wa38L1Q_@Y?NCMQ}tbjdcYkLC3!sdd@;$#h<l=}A<)9>JO{
zaod;*h1WaMfh`^VDmKH#COcN7IQ%gfy{!<DGeqP!(-qV-U?>Yj=gtRSHGL*_4A3x9
z$>;TdT98j%W`BrL)IoE2nBH2Qdq43nEdSBv>{+W5k8Z>a4Dj2H>pA&v-^YTpU&Dw%
zPa~ud9guhGvo3x<=B^w^=IZVz)EymkP$Q(InBgYD=tfpt@O^H$<VPI-?B}!ZoBo2)
zfkA40A>#GQIF~zXlxuQFc7&7yI+t;JbAGI3cJU%o&=Zn&M4$@mSFz@@tGMgOKW6B6
z%a}n(uLfw;#%M}kf1N}PO#_d|EgHM*%90nqkljyz3Cy17gL)bu&J-+M5%GadS^Bcc
zKub48K&D*gjW<o$TIuUlK5JV_WZ4d%yH)-1Ck;GIRB%6)0Af4<GgU}AIdFLL0c1hJ
z%L**#Tw!|P9hXejZB`NyG>r+Jz{qOu|L%p{_x;NVR&8Y08M8=+hG~(|=<8=)G|HN2
zm_1KAfg|4jSFrmov}{Vy)1P(kz2#{he{P+xkn{WoC!Qo|R5mjq?LJvmT|l?EwUd%H
z`8}f*%}1Whp2iMq6*Y!b3@%Ejt=q)1&wq}`uezFe&TNi)|9j|v_VJ8bN8QvR^-~j*
zuZOwLBx|gQGuaZP6&k1i+|;=&M?q+30d6Ne?<|&Y3Xw()<6_#Q5jA57;!zIXdk#PO
z{?{<60|^S#h}3uiv^REk7|k*X)OgXmQ6!FtmfgqQx82GSr=LaA+Y6yB$W)R#51=$F
z;Jb)zTlUO$8pwK}klzAjqGewEyRx5#!>3mlU{U?`*(?OJ<BFP2E!fs9$4~C-JI0t-
zPghUVfFWOe-leMlF?}YUUQ>i^-k^(amikVcgZ9uSY9##poqtEqefP4!)){GylJtfA
zW@r;{_|{ip$^Nu@0zwH$of1f8bgWsff~k(_yW(KULV+^3B8Om%5RXKJNek}&Ef@dm
zXQ_ofy!eCfgI)KcIeP{{U{F=U(%HPsSTLsO>S#9-ArGC)0_gJZE(Tme8b#Q|z^YZ;
zfANLf`{N%Hu3X2gphjH{Njss|Q^STnIBjEdgf%Xvckd-U_g!y?0}g=MgS5i{6NFd{
zr5;%`dR3+9(aft}I_+b(e8HNCi7pfLoVTW=+}1w&Q=#&EYJ2^EjD)R;G8%Lea54GW
z+aBL|JOx1chh)uw;yqGQ=EN8@Rh-eR5zr<~EUa74!$0^jH-GO!`ZsQ3L0>=ZbsOjr
zrB)9ZO@#+qjs+(_i{sw&X4q{JkyvWA9t2C<`5Y5dboafYNG0?8an{orGk{-sD`~0u
z*$gmLK*Lnz{2hRSQh(*c@U%=PR!EHCj6sQMtQ%tEg%@$d*UzIqFu*fjbvAQeaVG8F
z02|Z__Xi<!d8>54MGaIyOnKaj%S;!X^Ua)CQ{M8<bGYRCWrV%`;1ZH{Odtk{M`@%R
zx#y0n=n+Lsm}w0h5{P%p$}C1rHZ9St5>JhyM1-xB-(7VjJp+9#KKewGF!14=nbu4Y
z#5q#S7K@+9pD>gszM5X13&At<)e>S<Xs)?7k1k5ul%Vz>S%i#;yjw*6bGp8o1`CI(
z>LtLg)8+P#jx@Aztngix%<j!NwB?ZIC=9LTs`vdJGagt?f4hbB2v696$7c@k%CCQg
z(Vm&qW;8H4?Kh)%=G-5Zyi6`>2F2%zW*p+O=i-czrb4aRg5iza^PT_Yy6^vplU{ii
z`<(r1MrsWjGrUEsRtkzl9k$}cqGTF`BK5Vhd222t&KxleNbG2jj#9G@p7=e>zWaYX
z@XPDaM^`gfY6NXhv9f_8y>*(YvbG)5oHc_3p8Fj3Is28c+b$$x2zqK|o<s%|(ws4y
z^KZ4{7?o=~T$ux*GFG8C5#eO2<g@Kj+)V@+rqX@hn)&-w-#hkkO~qb#OWre;$=x3t
z{FuyVYNGG-R8y@=M^G(`+k(#F=`7Ve2&uTZDMErHNDMMM#L7#5$Zh9eh<W%?_UM~|
zL`|CUDE%{Mv%cNr(UDE;`P@@@#@qiIc9}=BW~hgKzMGvoZ((dLnm7W<dfb^uv&(Z^
zH2<?i`ihSENJy90*p|F}(4=2KO`f`)=5!?o7TQ3tvBmJ!*Kplu{)2(E$u2K=0lU5P
z-L(3Ju<o;2opZ&cp%4vA8O-OzWL^L4>*Ci~=OA{9NSiNy=>k6Z@8>gXr$t1~rhj9s
zgz9K-c${DS^n4E9XCY$5YZ<a*L~WVtp{_91P-6-U+cb^|H*VniOTNXko_jhh+!I&N
zvSTu85}IVcC)j!!P1jx58R(Zac*c5qx{N`j;{ocH_mADi`6GTkBqA@}(Wyy0JYbln
zB0aq(3X^U8I>`cABsU{C&3KukElixkeRpxu2S3bysS<3AaG~Q7*P?#xah&++f1xRc
zT5pdJV)K;$id#hn0CGp9%mFEH%4wQSAdDeR9L6}D8j?7s9s~C*<;S1<BB{bl-t#xO
zeRd_9+ef|9@yZKwjH|!|$uyFsqk7JTR`rt$mneo5tR+buNvlPpDX?+{H-6(Qti0iu
z42-lGAfSQ4jkc)OL!5*pb<3lhhN$hb3x~bpEashY8Vt-LX#}W&zyzDQ9Oe6mb3QMD
ze1p#lfIKzZZ%h@DcIApz!<EMG&~LBq|JIhUQ#GHQxD@aFu3NffPhQ4P^bBq7r|A})
zqlu48x+I)c7ng45_{rheSR3Egz|JQNUQTo;6=7o9na?7Hi<?x3hK1(JWe<_D;9MIM
z1EWng{`3lNI`10<t5&h+oY|~@bPYX$#nyXRp9ss_ZT3C;3=VnM8zF4aY=ra#bzGW~
zVh|Bb=2YZ4gr5JOhlhDfSScD4g<xueZ@9(hIQ#P6cGVR57i2Cyp5J3>3xrV%bIU#a
z?EN2LAW>!<e?0qr@I6FxYnZS<3meNt(5@(ay$!4XxD<mU&u4L76fQQKg0}eiFK^|v
zH-D7D`MZ$DO`Ixetc1pq4sGC{|KUBn_4O~pisIDbk_1~bfEQ8cnFr;*;Q^U*fK4&&
zn4!Dw;Gt!AbKvvOAnor(!q6A+X^N2)j78LL^&LCeed$kmKKm|hVMQjQXpa3<{)yQ&
z?C{j29U3r9yCUs4fTAefq-)=*SmhPGTFpvr3(-(a!&ruX@?(DWAD?F*H8fYPB^<2t
zz(~R#XP?Od?|VOy2tiMnb)K{<akEm|W|It)jqcS8B(i*mtdnI8k_cwBiNqUO{@w3!
z^@ab(iKm^#p09pA&E8%bGlOz6KsRS69}i1maFM1Co4JB0CKFV9<)B#5BqeS(sW}G^
z+{Nu*J)f1g+)VFCOke6S&RYOt+tg|yEns8Vz%JQ~<Ny4P*dq=j?diwW4AvMwAt~HO
zD#h3sZQESvt(4giY1cIr6q|i7Q5k1*lPMI*+k-YV)?dju<@5HrbJ+&iQ{^SgINd{?
zfSJE7{mHS9XZ%kSWb90h%5R+bn(LH`^@;A5?MXD-UMpCM0XnTewUR)bCr1~BZhmEz
zBG-W|%{|tHAQ2=Q<?$b0%$;99kDfKdEb1F%ba;f8IC}g0S=}6AbxnBoU;ZU?U-4%|
zfa$Lh5|G3Wk&FhF9j(cZ!&S3f3bBHA#ba-;b6Zh9sQ-TIa&frVWvO+3LePw{_y3+7
zKJrPDHILGB(0&~G$$w&apodz$j{r#0gur+mfXVb^BBuD9w3vz}7$;eyMF~us<qxgq
zz~{b!{@IJsv<X&-BZnnLTTM=T?(uy63m?Rgk~o76{J9JKCuN~-6|!m$ek99-aSj=6
za`zQKWX__+EO^Gzq&+piS*dqh@>bZTK45$1o#(=E$Tc5QfY?G4H>pX(ACIDW%kk-s
z?V7a10tQt*KvgdV_M9%OpI(!Nae|r0ZE#f!Zlm7L&KMyYZc?`)51jiI?mX|i9Nb?c
zTDJ*1cYxd0tmd>&{1XE&dI9awBB3pkldaK;(781D@8<0~br`XJ6fWS@TkN?=35J`n
z;&=Sw^Z(B1P>UCT;9aoK!9;@tgtdTFy*9#I3sQ6(OOTM~C}d)~3I{b0K0<s-X&Oa@
zvBHB7aO0Q1#ADZA!@QQI;T*;V2r0H^5nCs+2^-Ud-hKDsm^Z%-bIegBwg!EHcRF%i
zKIW9U2^8H>zqz;$&-wD|IVk~yWIr$2tfVL_b*g3IQ*=L8KujKdR~AW6#!@@ga`&l!
z4$Eh1EI`l&o;yyJl<-EnzC|uJ-As5e0bu9|;u_n5XM01+ZJ9Vs`fhQRwW=eqr6XXn
zC>6DIxzQ|e(&o-W;-`8EF=5-m+I6h>*7vyU(o30@#LR8<G5q){8dhMSm(|Ub^$TY4
z!Vi9so@0+7b%O2f169&OXzTmr{?IA_aDvIg_CMEUQEQ4xYS|ihMv!<I`}hWab<T%L
zmM<sRZ80Z&^q)v~T}WE5VF~=S1izO1si!4(tMb8hS>s0ynuS<~(q_|0!jh9-M~fb6
zHYOE_q7*BNX)C+!G{~>7{2HMoB#FfY-hL{;R&sE(;<%l|e3yYifp!dQS99ZK-{+VY
zycFgwB&~-<VzO%D=}v`n*ON18n9BR7O7;5_nlwBhBBzVU-P3?!2YNZ@z6MN(a_mr3
zhRUDxkd{CI03ZNKL_t(VLrSS~ZvEe+BpaL54BYVHkI=mK7WS|;TCE|HS$*8=EPwWe
z&!7h$KwPU4W-0fDA9j%rtlcG2;ezB{HWX`&w{R7oZ;(bYwc&_0Ke?1EzVdYrI^iUq
z@#eP>%^t)K)CscVc7@%YcBEwH8>3b<0eRS&KS+aGf!Gm`j#4)QEAHlwFMo|kZn%lQ
zk%&P8q~&Q!Vja{&Mo8HdHW)neaE^ZSn_&L~N$L%xCfH1%trdDiuH!b{=b282ll*--
zuqY^6IekifpRXAINQZp#4^ZZH-BVSdU|dJtQcQWeZJv}406BJ7e!PD8lhN~^N(ol!
z?{{^GJAZd&DxxwjA4tde<@tT=0i#v`hA;Q6nsLsBZow7X((k){3#hGmU%51mA>_%P
ze=Pvj*pB)gYXgeO#Y^J#K}nhrMgnVBa^HF9@xYI-WOm$UmIgFOHehN(FmsUmHmqm%
z(MNFN-@FfY-i_hV(pT$4U6va_=k@f5>`Lp^2?E--=?U(}C1bG@RT1&Juy)+0@x(B<
z{PSmtZoQ8Btoa=I@lTTMw-_DhEi6REdBY1S%%@!@?zUM|#^N#&dd()JF5t))zmsK;
ztfRN#Xsdb~>Oi0cb<^VZUw)hUGaEP;f`GNg(=u&2B*_OeQdujCm<ANuZPwg&BdZ>H
zhyzbO3s(<t*5XtMD<JYIX})fLw{2C+<4=g1^mSv*8>a!ojx$|&y{dk5x|Dvp+`yaL
z@r_wZWR|9-eO`qZtv;*3s*tQ7CWxC{`JVUE|KKX-yOg-qqSX`f=<I%8{KYR5?=&B2
z)C!Sq#<nlI{UVjb<gR*3@}12X(j-BG02e!CBq3P8nw!u43QKRek+a_Y9(s>IhUWZP
z1VNUo;5w?&g343s1gb6-zJO0t#j1nAr<1l@2|cNShknPk=YEk#uKfiI6GOiWp%{`@
zL>O99r^F$wPGS}xa}0;R@h{<!Ly2o4(o@4$1AU~-Zzy=xV|A0Xbj~RD@hWSf<BD}M
z*T?BokFk_(of+nqNSM1j+&*PhE4GWiIMN&<%I1oTsa8!gmDPJ2cHPZGvd0L%=(7K%
z%SN-)SG~MNUBPGXEPlJ#dKCv9E$J*3z<`P`a>qO=6J52=zwcCqY#wl>Px0C{(X?YT
zwp&h>)XFnc0EH<=b>)_rg~owG;~*B&BqE3u9(ss7zx-7mx%L|7<CukI<C^vK&z{Lf
zV|mae9QvA9viF<*g2-A-qnE%0xHRz@0vOPg%v>%zUfHa$97~z9z0PU5-jv_hf08A!
zkh&I)VaNUFe1g$m-AH>@A4h%cAK>UiXf{Hu+6*uVLRucJxQwji0fQP(?+i>rs)n=P
z@vmHV^-av^4~Ru*MUGk^nAjl6dM^0#Cpr1(Lr^ssYci2F88tOeY#swJ<j;c$P7}=V
z2*1AQ0uFrUG4$`f4{5zlqLvz&#gf=Rz+GKhtWKqQ_a|F-i2S*ToHq>^p8jBAXH~rn
zIC#3;e!9#Krs&pEndYPF!^!eR8uclb>m6qG6I}N0ce7~SdU{(e;vivD%@QrypOZiP
z|A>0(*!~7(L%AG(7sX@FpUn#`SIJz+F@yn!A!)T~qzT-1J3spT7iic4p8NNIL%hqb
z$gEz-a}_GuhaBXT)}vLxLSC);bCX#hwWBsQ2RycdyZ+}qmS6KT`bS#yIg5!QK*2_+
z4HzM1q+Vy{k%w{gTiyzL?L*w?Md|^8m~6^XInmWFab>gaKn`PG7ag~FD!9pc=v;TV
zzU%i+yb*2LHZ>lY7z;0GnT=bNKK^(OPP#*^%8t3HbaC^miP98lx=aRyCx-wX!t*g<
z*0_}8vBn8XnHyHUHvao+1wc5F3{3lXrrys6i%${G=|*M$)_;sey=F!--+xTyd~H6J
z==S?B&*+xIK^+r~={BuSlNV3pUDM8#WJSg0)(Kr^rIoDTaXcz3B%y2~nCT5N^8p_m
zrZ^FjxJ@kqR@}|CpZ*-_UCWq9jdW8Jk(Ao389cUWg!OZ0aK<M-4u>2<I}~a?y`)ZY
zF2xv2ca)t^S7u6C3u9do-IDWN0N=bhl)?8b$kJI)$*ip#Wz|>zmxnI9kaX58j{3ld
za7P?W+-MMDa4yXp=e!e<Xy*226r8~saA`tll_VAZ?lTu~?l-^3;6RNC7;VMWto1hA
zLu>i)`(MjD-*`5Hp%&I~E-m#Z8DJ<)DsuZ;itx0eOVx^r9)6HJZ~8gUe8Ed`GZv79
z)<0XF%C_wa!lr)jKCScTe=8!VipV27QpII=v=Z|l|M4F`2E25-*#6Ul1~SgzbIPtJ
zIsK&&4ocyOA(|pX5pw_Uxah5KWl77>7~X_wz`C$OaO@Er_nFVqY}5$*8l)*$>$M8f
z5$r?d_g4CvS(hvCb{Qj(7~JS6_2E&LedC*4{>88GjAM`IneTf)&BY5agFS_HUqL;}
zXg=9C($Ur_sPhIS4_z}UE{c$5f?d0jd;a@euKvV7;qJPRxlu|_Y``gDXi#ey)tFUD
zO#k7Ba`J~i!p^UGE$xMi&>6i1VOVTfj?&pE*JYG|NQ?6zC8p`PPBO+d6&SMvDO5c*
zq{^<?ir&chpC((f!Ai*xsTQZwb+_g=x}_FFI=)XHj3?3|`-KWHtX^NO6tLsxxk^XP
zRDMq7Cxpt8NJVcW)y(g%(<Wm=lM+}+*-fs(pJKWqo%c)8an_=%ns-r)`z9M*X?`w?
zcn&QOUB^Q`m4lbcY(dImZ2MTj%9Q_JCSBv`x+lA?vS{%HRG*wq53yX3+VzZS$I0>y
z4ah{Y`DD~P;7DN6(BNhcu>12*W8vb(+<wPzAa(Q@aO;PdSqqpa!gUw@kmjl<*!jpq
zNz;@h6v80P0xC^Mb0FvFKNY>o(zQp5rcOsALJh`Pl0L)Wkq41%8fL?tOL_RZtJr<<
z?u3gL6Ng|VC?0qzu0d%ikw3e}=iISk<wmaf`8Cw*A+ajzOtwB6LJR#phSOj0JgmuV
zr9}`iS$Jjz7-S!yrgXtVDvgFv2>SXNe)wVB=rH{Yc0+2WV`Zzz+MbkdvW)ZR4)9t4
zn2&wzW0y})7@oe<h0`*j^Ym+nm^+O1mr9a4tQbE!Pg7jd#<m^YdK(v<a}GxYJ-Btl
zXv6Rb!lGB4$^P$uFC$4pqc^~28iK+L-8od1)Oc$@FF4FVgi|o4*aTD+u?B5}8IEA+
z39kA4=XvnAcXP%&-bMehCo?*;hkC=0sWV%1(hQiGN>QF`23%%As?|D5gd~cvaT{rF
zVAUl*<d%zm01vHXp{Zk9U_>#2BMvMZ;}#nvV5j4c;h4AmB`n#Wb{G&eyqG!1SlwdP
z&AZ;((I@ZhL`@!$)=3^IFPu!ed`(t_OwC1Si`J9k5b`d&%YvRttHDAbnN6KKDOFuL
z8C~I49miHmRw`Y_E~>JQcNV(7-DZrfPN+TB`7dY^bZTu)$7Ct@g71biRRhLSysSc3
z(z*A!r&0!YWN?UfV5qV0fpMlMQ<l33jF~q)S+>Vi=26>eAM5%Qt!RGyS<F&`g$f;P
z{2sC7_hxRFnNgK;$Du?OiDK+%#ADz7Hg|sKd=@53BUPf|^$gCN!<u%&gT`^%$39H`
zq>~v|rLT8}4^hh}80xb1YvMcfNfAO9E&8&tqD~R(r(kUyjSVr6o_`6CedkMv>EY17
z{d<Cw4=3&mkzACV*<l;agW?<*gG*Da6`~~Nmp4DiS%3ac`e)9dDMAwYX^TY&HD<v~
z<(i*fOs{o>fgy<#Y!GIm_Npaqwr*6xbOmI4C)EhMZZ+3_|2sVE^w*$s7C_xre#h2p
z8MgL1$PP4d$OO<ong$F{Kd?|&)t>?<PnXnxnm6d?<z}bl;AFHEtvF~1jyQG%?G&3N
z3}5yWuKvPjIe1_uBP*Z4^fmbXXq)3c^j-#Eem1Q@d>7t<&0{#T90jLY=mL_B;tiTQ
zB%?ba;F1&xy|B4G9MjV_u=H-e_iz8gf}MBeq`!I-EZ&9oK*Q%Zpjn~rX##nOTy8^}
zx222yAW2YTFjh!f2|=?B?S%E${er8%^q<&A9${X+mxdE`WRzMhAW=gr6doNNW#Q3B
z^Q^bO0}eWfq#ojGA)&EUnqE^}G`1xxU@9bq&d%#tz4pm%GCTR}Tf2!>gQz?hOc6B&
zX!M@&S&>vpp)I=kI+l>QGO7_@WR1=HmP*H2x+qU$SnZBMD%J^m<FeW3>1MdAwD8=_
zgI%gZ1z%L*T%|bmR2Y>UR0--bvQqA(Rg9yuP(w<mt<DQ76+|`ml`4V}RW`e4v_kOd
z465psb85#dsXHrUEH%XTErUCS=UQkQs#eNn;;!kszGJ&cRV}4Gt+Fz~%bFA_z7gV*
z1d}Q}@));#<}<9k?ph9AxDy*!JVH+`Kxfb9&Nb`V`9;s+sB=C@To2Jk4`C3Jrdpap
zWFlyjTXc$y@jaiM^0N4zPH|a~rC>;nkcgwvR388LFR=2GpW<c@aQM65iah^#+O;|c
zQ)tMH5z4>^pHy!xB&qVi%3+Rr&dV5>GoR5E(zbVwl2oW$#YL;R<(kV`IKvWH5D7|c
zmt?8QS_;I~xPq!z<}~Dh+re;?wRhak+VxMc<djoMdTJOkrBh#rTiEsi3==DbcBt1c
zMdXDd5={e!rx#%OJ5~MX=@R_uc(YbE`%DBl*5c)~?8LjoktQ+qw8rXhew$yPdoBm}
z)oBf_gPF6qr`6=8pZ-_m=o5%SLs<7<Aum&<Dw;NA_NG}EH3Ji=Q><iiIT4&FNi<5M
z8S~KhF6PQ_UBGkBJd54WID_`Q8Q8wSa|E^Qtmds;sgBe0jog25K|e^GBW*_1TP?Wt
zR<8caml;}mx1T5^xmSCeTFnrNvdPwPOZMTEcf11*K9so8Kx&4ZdF-;yT#S@m?nxoR
zc>g-FC(srbRnpalpJbXaQT*Aq0*0J(><bX7xL0-2(z>z+q>_ByagSHrc-7<$(k#ul
zm{d4lyks4HlRJG?d1>9>Om*xBX{@eqM**H$Qx?>q?ERVlz8KRn<tPgJt~7D!{FUWr
z=K!H&+Tv4p{b%J~pmYI~iPr+UfWVfPG`bmJnkd0xynB6n<zP>+`i-B9g{w?uI)G)_
z!1iZ5r&fv=;krtqG<i5%W)=~}33V0P*I&yOpZp{{4Y!!*YOGnmhTeJeSlx_xq;7fD
zzyBK?b_k=Q^fr1)6R%|vYl?eOy2MT6U!^TCMaTLHt>o=U&d_p#lN7PW>p9kqu=;|F
zx&OkiBeQ4m%ny7JIqYCu!{;>QHrS;e#7`Qm72+gi)L9NX=7nhQEJoT2iNlDHw3UWc
z+RZ2U&ey-lbB;WKFjSlxf1?}C=R+M+Asu6>D<%k%qDg{`G`a243po7vXQPYuL~Ax%
z8gvTJ%=Q4WlmGkE6-<c8IU@3@X~6LG0SiZD(}lUyW%r-Xjk#H2SGDV7Y*vMi!M>>a
zSQ?|a6yl*#>a~ESANwqiU3w||^!CvnT2HWG2KT0hQ~&$3Bzx^c+SBVd3C$=xg-uDO
zDXzG{l$i+LMI-+iZ$AQV(+FmC4ZlA3D=fX~Cf<0?Kf=EI5-pgCwLa{Oia0vwtL8<i
z$05e2fEEF4-c;SS+n5M=d^y*A<}<9n{zm4Dp^=7Y)b`Y(TAd_RHbyPth4VP}O>d*`
z1<xa{)uAWEicb`^MZq9>qr0R0+%^XCw)UaM7p9K)Yq!TnHK9y4pB`klhPTPhIL{SZ
z*@D7WEnSL&IScVqe1fdY>IbLZ#>VA#RL&EnlQu+cXh*nKg!}*Oop+oiRhjp{&pB0f
zL#N3z#35%?GAgKmh>D5<U9+pMzH35-RZ%fvTvu1rMOWQbUn9GQ)fFX)2~ZRe5E#M`
z2PVf(H&vbUynj@<b$hyddWOM;eY-v%K0t@t_g0;%^E}TlwOY7li=^2k8E=p@8@P50
z-|66!1n-ifDDQ$Cg0U7c7O^%JsalKJ02_svFv3)-*jk04TF2IE*jf#05mc%Wgb)TP
zx@Jtbc@{-YzHTL5|6)?aEbnT3+K3jtNoi4OalSK8f9=vpx~bo?Z3nBJ@XVQrOYQi2
z+8G%clQ$~T>!H(wD)>wwD{3a>(v<!$PmWe8X<M@9Of^d}PANP#<4Gncrq#kU#<}@Z
z|ICU@E@jWYJ{WgcjS2e)xoOKN`@Q%Kj(Xoe&}ljBP(PNG&Z%P3buK&W>dpMx<mg`>
zW~YVkbgGl_5)-sz?*HPKSaHeaxSDX%x#z)AM-bPlSZvlLWNcH-hg7*ut#H)IFJw*A
z(CUzey9xM22#qIcZRWji|3AF*@6IF)74^2%`I*8+PfD9x?!eYF)z4p7946_|xOEv1
z-g74ho%VcOy_O-|JJ*<b_L%h8kf97W%r*@F3l8u7cYs&SuIvBZ&BatBDz9|ig%VFZ
z-XOwr(}&Jy!{xtbuS$eVTD1E5Sv6-a&${qm=<K=}Y83)9F`MrSF;#OaMYg&6%<;wK
zl&3T_$s1Cnp`}F47#_HfpZ()MVyZPx|KNLQ&6!U$x1Kcxnif|+wO!S`IYA0)idRXQ
zPZ`wgMbVCj@dkEu0}uZ2C){w+cUaW)40yq}6O0jpiV%mwrjDn*+b$gYs+TkO^fOY|
zqpBq^DL>bjMX3|%ubO^dQ&_fc54r$T8%WM%i=W18RM|>1G4(ss8PAq=(!V=NNtt`u
zyOi6EQ^a-);S2=Ai>8G>igAJw56-2p=QW+6lR&EhjWO7~nPk%jTI<)-Siga>b?X`5
zxRKWI2%WJ8uH8mEaY}Vd9HNSNpVCNN%C)w18`+c|rCqeZYd%Z>%|J50DWW0@ZA2ET
zhq0&?Bn;BG9$35)OeF*p5M~WTeQ1d4oFV!bETFz%0ey=W5iXbybLPREA&4qq1CZ2r
z-UV3%UtqI|wkbS*v}+8L7Gu8KP<5N=-hi@gqh8xzicf8KIu$lG6I<VhGjE*=rI@0P
zo!}g$liV-mhQntFPFcPxg>gZKpyfrj(@dmgNa7AbEO7hnT=szv60P3EUVR}Wt5#E;
zzkszq=E44uS6ui7Sh^pL$WpJ?@J<WIBxL3STYVVc@3}rbiJo+mp$%zHgO90hZgAhf
zU&N+g{0ALEp7`<4;EvcIt*7)P=QHZ4$)Cwt=8Xt1dhOf!?O*PLpoVWJ*dV|sf+4~9
z7N<S+SibqW4`5A#N|-%HG}S<f&t5ANYpW@cAkA%_@_hz_7f_9{BMol)#ecE?k;hQo
z`v5dFq<C?LN}7iOaLpduI=*X+dF5=w@YqKeUa6|zo?Xw6z2aY(rVE(Cv#)Sn7eTzo
zIggGs3AD-8?>HC!ax=SD`WTO!j736c?}K>i=RZqla1g0i(rK7aE$Y&w&D>7ZXPccA
z4jJmQB409ru<;~Vd^4fiO4xk)Z}`PmzRs~HJ&pZe`Dz*qBPs)xbi?soCmBjpXwAAE
zF(z|%0&xm~C2<aIbqL!r8-INnzx%=k)Yffbms%gXWt=b+oKZ}*!j^W6;aY`bU-oJi
zzT#!*;1F?NM2JOw+7S7oxy~GYN|vzByNs4(dd}_4!+K_}P1xQ=(iOp$Rl?e}C!F}h
z+Zga_$q~M6Y*;2xs7`RR$;6MTO|IhAAx<%ll=JO;y5T1tIvsS=FwOOA7+tlJ(Y31>
zU$ctwb!!;kw1MX4VSKBFbYf6P)dYxhRD%%X9hERd5=Ue$K8XozstvF?EhrOa7OqPo
zTN#9!%s0npYig^KR!dE(dc;^<Y70i<J!(?whDV6K!y7|8ai|GsdrzW<#26B>$Y4L=
zzyJe_7BaACG4u9X%EJBjU|_dhVbMIO)WM`KG>KJwD5x=52{9(foN2laHOeHK=d!!Z
z1O#Gcg{Vxp?h_^86Q4=ii+Gk$5Zy)tGJUhLvr-7Ql}NKyqFDizDdj}=oB}W1ba>4Z
z+Vf|V;Em!M4Z@~n`M-UYo4@@{j@xB7*53Uf^?HOZT)@p6H}S;3e;s?j=GC;EqcYf+
zP3gOoO39n=NsE(hKQFdBnoo1>hRG74HKTYMoCsdpMB~amANmN5tFLEl!5~lk*!jc<
zE`@p>@g_@yPH9TkNS2N1_{eAfmCt?qzp2#b;adq<@QJ~i1e|C0`8BS)?AwIKp%N9S
zpHV@Jh9SFO)fULMmIMqL7@3-Ph{rcN_y-?g*-h8*glC;eQm<nII{{^o9t(n5)L|#q
zE+VfIk?+hl43FJrVV<g90USBIoF9veb89K#bldL4y|-|25zxdFx0*yFWBmS{_YmB&
zoVniPs-BU)K{oHYh|?~-khng8)Pf>_Ef?nH+-vXC39<mjYl-cYX&OXRWI-)D+9vEY
zxb4gT%FVyInls+?7W$ri3az0EVLg45)~858ukByvd}K^F1{n%oP_xmb<^^uMk!wEx
zMMiJGoyCMm%cagrfkOhrm@%xeaNx5~=ioQKkz`;VzCR+g7Re@tB9l-%vI%eEJtmu$
z`z#Pn$TqcU8-{JWA;~tKXr{ifTH<AEkF8MGQD-Y%0Np^fiRK|%?5JfqYpGqGn6Rth
zZ7Q1d;=y@B?~&MnCMi-dIu7gC(OUT+n;y8I^>^RPmX#|R-?SOuY(bI`Bxz8f^-3iO
zki-$15K)iGoN7d~Lc}HMwNNFrsi-~`Z+45i+=a`~<)BkS8MHKfPCxhLAzG>JsrD3<
zd1APhigV5Q=!wfb_r>B7)Cpb%u^}Bii7E*mZyX(sNi<=+)xt+1^||v{wD0~b+JAqR
z9Cj#n-+f`>0;pEOhA0+i4N69p${M4xYt&>*rMK`e%eH&hSeMQ?jVh<BFV7GTEq$%m
z@oKVWfoZXU$#X~^zJPRU+@<TTniikFo3l<K5HG~`T{@?fkSNtu6K3;iy@hrsp(4Po
zH*?8*KS=-D5tc}e@v$w`7tUivqfKL<U3lK-FCgw8KnJTt#*!qNK}2?8C8Ms%6tLmL
z=f#)sMk#qC=M+$%73P7(yM$o<2JZjT*V%aOFBzLZho^q-ON<Zo6W01Mno>}0F637b
zlklHEy^`0x|I-W%E~Y)+N)Z_!5Qa(;D>fPBj%$9z+<^$Frsv&g)-dF0=XrL75csmz
z!xwK^GtoAhq)&imi(7vFpDaD#Fa{4g0$;HcPFAVQMUgHaYlqvf*`qlAEF!0h$fntb
z;W2o4@6QK5FuRT)E6Bp+X`fQs<jPj!I%NaV<0KEx6OW7%w#NAR8{f&2m8+QNEG^n>
zhy-)g@jUq-Ka!?<R_j=w7aUq}_YJ9whGZLzp(||7x#12lR?!69=5fL;o4MlSpGG9)
z>F<0!?75WAoB<4iQzeuXF%WCgFeaZN6liyvpF3R`#kXTjyA7Mx^QX^!k#$%9j$IN%
z*ik|;IGZNCC86-(_!jyf_XJM*@P}d1uEaxi0vl$!3W|TL=EZb(B43~xnW$Ci+BTZ9
z?cWZtW4D$bZ!gI}yPV!>Cc%>qp?iwuuFY-gP1~(NC8K?bWMW%MDH}eM5EuvI5{!4?
zI<RG!&dLYaxa>|g-E}(~9$Z0d{bsb&#(Kr75~>h5PhbS=eM%IFW5ftXA+QGV+MTFp
ze?RwTcg|yM=F(-;uN#x5af_8AgiBRRL2@P&Nq2+M%9f~I@o~3kHQ6`w%!quQTa&N*
zT&tsam!2DSu5dld=|x7;RPj{B?^6WECs`<)#d{&BRB6Q?9|nxITQm~SXyVa&mHLte
z%-w%qmK=UKyFB4w!abM5;9#no5Q~NZ4pV3^@<vISF1T-s23EG-5@k{VKubD{Op~JW
zF!T&N16A1ZZN89~OFDDbVhu>olejRE$R55TUBp3)(~_bMisD;i*wF@m{`3W`xa8OD
zU#nt9o46=M<_&P`=rGU!=*Ov?d;$#@Q}62|acSpU1XPM_H&V1CSAmjqo{@YwkTy{{
z7U!i&Fy7#%4}F-<9e2@LJf9Oj_D?kD4iVM+GE~nLsn&wy`oG-831_~ZIrH`)X~$r}
z#exaJJBzkAbNNLVaP*=3rVWFnMj9sD<9ue80V%c6`jQmdXrYtJn=0=eSivNUyZ0WJ
z-SB%J_l)P`<_v;07&YndkRlj42~ac2Q}?;L&5_*fG5zsTW6TF<8-~Xox^S$jUJ3Ni
zuHpY3D|J%$qe;e1UKJ}DM$mEiu@-jIdM<tEJ6XJT6H%juRE2dCGWe7eIr4+&(Wym9
zCB)_qkoZhbkkWxXDR-xs!#TQkka@dfEZT6`MhwgT!bP9{EC(KW3`e~Fb+qT#({$);
zqxP9rp(~2*iVRgHE#^Il6=Ij5jTTX}!KO=o!PQ^;8uNx5%ppP&#e@ct4oz!V+i4+t
z?#9#4c{lc$<7nH6N?(-T6FDDz8#xMj`Rz*sIHMDWQ`<D$+u9MSSE#yUBG4X3PtB=Z
z8TYp&m+kpX8D%6DsCR8W^SJ7ivPQu(%IbDp;!>e|>@bpmk6~m4Rz1YXt+%mm*)lfX
z`vA#s1M6Z!r&#9@mx-1Ox`*PD7%OT092!9rpZ$4BgR^`pj`!LXg*p{tb;PQ)U4rwe
zr#%9(DiVY^F_^%jK{|n#AQLZ#78`^}VABO{3|O0r=8fs4A8DG*t4Zp(;$4i!2|n>?
zoZylek^~ngh;!f^R@13!C_-q2$XX&}sTf0`5EM6w5{L)&_#{SrT9}!P)`Un#=P=l?
zf=U1;AW=tCl;KW;mPu%dAzHAI1qbZMe#bwFzQYfLJ@<r4L>vZq6Q&4+4Z1E=(p~cf
zB9e8>MYx=5sevrt2Wc~8virLzj`OV3jK*I^Ac~2`jHlpJD<%GF`y_&sJikG^fSptQ
z001BWNkl<ZLl!lYG&oyZ$x_xdd7sf^z;$A52e{#SetO;q*~=+WoX{NE$l~31<(9P@
z+5N?5aM=6b&1kbjZD1fBM7YeNx*P<#73Mc{Y-O?nSSLX{yKWg>DyGlYsv77dj>?)%
zT>t3{2$ubc_QG8`?sH$GH3x!fr9k!4bLUt;+~nXBPNzPwh*m2h5rssk)B?N<h#Q;v
z=ks32YhL;sL~6xw!Z=VXJ-H{6J;c(ll`zn($x)Yv@8ynIGBVC>m;IC_`|Qp9!;i&9
zLAMF^irAE5o8B5>VcBYQ;y`Ei*nwkAL{1TrKg~7_k1cfJd%(+P*Y9Iz0Z#2uvh^-7
zk;kk(Hhrm()FqB!yaTINaPe8^uwQeWsF~1=ga<lJ_C51u9QgLP5LY533Np@-5}2GT
zn%dpD)EG6L%qIypu&B#7OHE0vKs*{#O)RT^@Dnci?)P}%YhK6PQ=dd<mqAQm3LkHg
zH0VuCH`~>@v}i&--U~(?-dN&Jo2b)9?!AvIFZco@*Z-M)D|Nzlj89^0)uJ|J*fm(u
zi8=M{Z)5HYUxM}x;3{GEniK5mnT*acxwyB?*Uw+W12Q9r?rlj)w|8SId$wy)1d^j3
zvUO?W%TxhMR$3+Qv>K0BpI%evF<!6<5IfksjxBfH#k!mS#QJ5oGQM&p+H6zv0exbq
z1tG*qsu_vnlva>AXW>&mekxzk?t2iysV7EAP-6AOE}`ueCyI?if@+0o{{YpYLF#jc
zsL!28ea>8JbLSBa^brjXBK11-_d%@&)e3|mgb`Q^fl2WKBk4Le(i^!ghDqH4yzl<w
z9K;ED2b~VI+t6sh_&D*%2%V8pIwK>rhqo}caU-J}HqzX<h4%0Wt<h0@oDgEETESuv
z=cyV)s6xdWe5XTzqH%{xWQgM=rQtb;5y7EYn`W1I8<E(Mk)*?@Q%1zm2tpPgekl8$
zcpQ5ka||rq6`~5xSg;}91Q-*fu3xT{UE+O;>gAL?^%;-7yo`P7K$hcbMNpS5PMMh~
zDxTzV#WtIU>6(X~O$(X$21}uM8JgNP8<VXO!5&IWw^J&NAPQ%nOm~GtS}oYz<`3t8
zkS*8#fg=|0!P-0TWWkcftgx0<^9Fd)x4%WpTIjD)$#oQxhoP0?=yoF1Fe#<7$TrBD
zO6{hsI|S$3R99`{5C7-&3^l+#@dOTk-`O<kN+qZjlr?oR2@waMcpA-CNbCb7vcwIK
z4Fqh2_y#X~)=_-+{BsCQEz9)D1^uZlot9~oC7bQ8{y)8DDY92`0q8iA75B32&sXvI
z)1FT<G$(aJ^QqQKjF#wdAujt>t35sMY}2sg$M=mfFQ07~9y{p5OI7vzv+MSM-(s9d
zO)5{U)5*Nx@wko>_yk#YC%=5(``LBFFi~S11_K_@3VXif`RxDZHxt_mCX{0TODPw=
z@F`FG0V(V*MFi*a$&<mWLq(8ggP_^sx(mO|1Ao4QGtYh}e*Xi}#e-?xM%R8%M0>Rs
zA_W}<kyJeElw^FIN^6W2-~Appe(xd{CPE#Hv|_^0LuhD3!iJ7A_sPd|+`HdHGPoFD
zH$;Iga)rvpj7(XqY`dTrrgDp!VX<HO&=bjLQ<g?$_s0nhqA8%KZo}BM1L@K!(*Nvc
zR;YkCiccJ>34u#6>e7OE?HbnKehX`_znS%SET^?;6SkdDRYQ;@L}8e|j#siWf+!3y
zPEiL|4IUwJ9t|vsNz-4)lMc<q(-Dv$!uHju&s)Tt-FIcql3iK2&tCNJx;w$5`7qcI
z^$J8$>MmpqdCa#;mds08h+Ve+l%D*u-hjL5uG|#-e!8@^iCJ)|y{~i|jQn?#1ezTf
z8^vwh!r1DE*z(|Nwyau7W9@oI)~=;JJc1h?r>dU5z)-hJ1wrD3&?Ok3(zn#7PEt;k
z#49EUz*u}>v6VW*V+|T440oEiz8Z@UJ%|IJdNT8mJqGsN9Ss5!W5Gli8x|8=$=-Hv
z*4u<^fb51zi)x>ht#&JPZZgzNdeS$x7EbFFBa*3`2;E6f*ol^`Q$1&O3g%S#!8<vH
zv{;EMU>xmci;5UlfA@Rb^zYx~34=pyd0-9nRgt+v{JGQS<)6O*4mf~zXo$iPm!!D}
zX?l9P?oCOEJ=xiN_-CkuK4|x&rvII}L$o@`4L9(+_kNUtiqikgliB~BXVW$zfr9ZK
z#Nh}y`6aL9p8GcuYn|A5j7!7RB&ZMw?7i6X>z{ptKn-9sgvX;|a*TKSyS2zK>C(3(
zx36flxc#ypvvmIhsULU*J`76NKWjA9VI&~Vs7L`mvw)RZl;PptX(BHZk&9*<hDU$1
zP*c@k1INuS+sDQNECB8zcR+iz3gsWv33Y8wsgTU+!kARx%vck-cR4?K>w7pT?huT(
zVZgGyIm*#zox{SHy@+<Lf|Z~ffF)W8ddc{xWvznuE^|RjwFX`Ue2f_#gAEUH<>xNM
zCx%nsaW2WCA#}kIiV~8AdgZTI^3#@VvMCwW1f#@_4q?N=UAJ=i`5(bNxROOt9c{FU
zs$n`ku3A=&jnUeDS5AM=`{9Vk(F!A?Y9+l7d}$ajQ-nLsplOhDwJuXlO=s?w)WeP5
z)<$2RsFkYMUh!@(7d^`{D;A5%Q3>(5%yV60k4+RhZP>JqO}E_4s_Xv1n%kDqTDKmP
zc&ZvuQ3$O~wFZG?lp;f5Y+54&1OeVc>{53i=agnAp`{5)V5rU+WYN++S#rRBEZlE@
zs(bGVix#F$K^SG|lz=rjHK;Xc=H;G>KTB4(<x^806zld|i%yB`$%OOGG+Lo)zF!E0
z9T*>Hp7Ghm@eW!o7#@ZPA7cIT<*d2$HrC&LFJmj$P)Qs$39yNyYJ|XhA`3~ojTLZ7
zjEJSviK$j1L<~uyn0i0M?KWdx86zPc8ep#{KZ(OmJ%PT%4~Kyv2tvG#P-8LHW_yP8
zAaLCf?G#4vd8#{QKi5($pY%?Gx25fw+BIogE=M~%Pj~BZL!G$5*aqOSwS#MS9r;x2
zf)Uy-rV@L&?KXaL)|)xJHkanwEd)5iU3THRRU3HbdH=wiXFrRkg=$}|7<lB!;!M>H
z5Buz7v_O+z=S<kFnJ=knjKEdb^M?<Aka(cNG4DH{@QKIZYlg@ND4s+tuY2cvxbinQ
z;KR8j#^c+bFtjAjVvWUgHn9Ar%cy9AF(Kl8mfMrv`{jmViVxM}?v*#OM$sf8zULm6
z|K)lPJ@pLyKtI?_lacL3xq#he0_y-@XeCr;8-|B@{82<s5|QR?!|>>C7|vDIPtLB_
z$M}Mjt%jOC-1V*8kg|ruw;Z<Ffn~RF@wx9|?`DD@A4B?u6<Xm*XP?cS7o0(-8exnS
zgCj+Y3AE3IfT@>#MmtiWsCkT}4MWQlbXxf9|HO~~`LpbQ*pWQ`O>d_;XMoC3P>7{f
zeThwzKi*SOoyj=d#-yI+rqRJ{-psNKzskM8yqtxxr|JxWY8vuWfe{P$$DU(e^J?~b
z%^T<h7TX_S)OH<lM9MSg8BG1Bn`ln81#i#eWBS}&-Nwu>Z!uJh3s#6Pi>a{BU9D0;
z!Ta<Te3B4ojC9)28iN)0vGRA<u;RKK8D6oHu$53zCGeh5vc^z_fD}ov0jVpINelQy
z3~hzB8iqR^;xHhtSEwyo!s0^@WVb^OXVF0i67I2E+86{O7=zQ4+%Kh0E~SS}xJo5K
zvQBZS%3+1wd!jti?p37h$iy&TOGQ3|PJPFg-^ygc(LI(onVqI)*9+&e7zzii1Ri>j
z^>^LH>SedG>7IL8f8V{B(NU`2Gh{3Sm4MJICW#^L;FB0aA=cC|Ac$&+7i?5vbF)d?
z#55wq{Cy8(zo(wS{1Z-u-IqYUP9qDcvxJ4<8w8i9X=hw<&1r8LwO>t+YUX14iDb&D
zvq84MS%96F7`?6XdWPlpL^07M-%AHLCbbg|(o@?RZ&7KsxaRD4!SZ|9!&e#Jyq>{D
z^SN_uoc^aD%acCwNyg(2)&9Z4a#~8X%BfE5&Px^6!eUkm1Tzb+GX^If+fdg2%NMxk
zr$47PRN>@Lf1dVUyJ7n)h!|Q9KKrlV;NSlJKXH}$jJC!x&SL|ES4)4T%ILcL_{*Pu
z&MrfJ*v$Skg|sQ?VxlsG=c*KIG0z)Pamd&>H(vZ-9PorAsqcLdu9E6|#HSM_V^a#O
zk`BL8R7fX`CuW<5hk-J@S42KF+b}$On}q{2bYYj-<@)GG6eeGQTa$pdY7%skW<ms<
z$2mvPOyK4}^OJMl$ARK;qb=M}$bE5xQ_lM+1J62*P86mzBTwcfo&nJ&wW-uRI1Lb&
z3Xgr_(xOj1Dj|NfMYZEu_2Zv%)enBcGhXz3=AL#Moh5_VYJjRIC{KU$XRw!|c*=DQ
zilH;spxWs$e$_Qx@lT&&X#FsAD*>Sz5*K4aFqMFHod(fihw<c(e*&Gim`-1X$k@z=
zRq(ExKAL_c>6w~N-7IXYX((T%?(c5f<47h-kf$k}CN_&&GzMKnsp}q;d%@>^v&INs
zU0RqsAxucqkcT%je)nzMchzrLaqYFV)@`7k2>sSlF_uoVMOZNy&CZuk-ETZ#f&h=F
z+3C>iBy<8t5{1;}EnxRUAIILu97p}ILtu}^&{u<Mlu7|ikaB~?rj8^kTa|ldNkPG=
z%ItM>r>Z<jyj#mk*+zaF+j>dZftLQLPJ7`B%17x}v?pL|vcD=liNi99y{&35f}!R6
zL%S3U)Fa+u@hQEj6~pRPj4iu^l{ekQL$}<@$f`92?G8iMQo&G*4BCk?&Y_(+%NTKl
z)?!3xdFY5`n1oRyL`!#Nzo$QqeV=(6?7bV<fQ|$hW3e{CJ5vaHjmZ*qed-pa>Y4Zf
zWzy|*>IPt22EsGL6LwOr$rOLNSKd6s(73YuSY|4qk~^!V$!N{yFf}5UU+}NoanXNp
z_|RP9^_$SXJ~l+KVb?`G?@M2zRSPkFbwXqDK0|pV4>>eb7B4#z4cU9UbQ{D1P%62n
zVJ%JPsc#wP_D_BWd($7-ykH^6edepQ=GTx~MB?DmE3W46-+C_L;yq|39-Ab1?{Qcv
zK|s7^E#Lmyr#bC}Bd~!<HBain=8jZMYxg_VT96F(aqXDK^1E2E{ALb4<+-^2e$)uo
zr&QWp-&Wwr0uNtjHhYwg)fvig=WN69C_KFP7XojdU8j$&_F-BUZ%MN-aR(^egv8@x
zk9CgDRaf)tkA0FoM%wuCCU!xcyPe^b_x%I)Q=d*!tER%^>{)4rOu(Eq3VHrP!GTwt
zf)yb#io=qOkJ6XK-1@mMbNe+napqZPBS$=r&f-2yNFiR!^sR{I;Buatn)!^2;1Y$g
z7Iw=zZv6ZgSaI3q>}~rA+o{c-58L=^jbZE96jnI)4R2uai(X9A8p3MSML=>J!eVo#
z$XVzTe`m7kPe-ZPy6~B5dyyIaM7i<MDPBWM*FHagnd=d0G3PX$s1e5m>cO?)!PRWK
z=>}F@b2Y1PzmsIk7*Xu0+Yt3G3(RwvB%u;jQkNnlsWTClHcG>LhFynfaFF@?EM<?w
zj%3d#9!KSX{a|iCY6F}JQhv2j)ceAYG#y1x47)32Dk(E3ccm#cb@@5SuU%eD_lC4;
zkE>64^g5AxzT<V|)>pCII26c3-pEZ1y30)nM2hCPxPq$1Po@o89!{7|IMo%h69A7%
zB^=pgl%x$?Mi{yE7S`N&6L(*GJ?%AXskS-{NkG3?!o)$lg;ht~X(JVrTCu7UnhNd2
zGfIbbaf_k-_ThjNPh!t!KL_^M4Xs!@VTcVX*qqPpyCGwmQLA#}=CrQUZL|b31B7q-
z4SL&6!y`TBnWBZ6WR2yvpSevmKK+^{edu=L4v`Pwra$tdx4o0&=PqIGJ*%kn*GT#X
zxLuT&e(#&K=JpW`^kWsAruSF+@YV@E+vE#<dq`Zx&y<7`+B=ZUoP-3d(C)OU3^%y`
zwJ%|?WoR9DB9A-oy^PuzYb&f;v682p@?zYg-5GbDFmXgtM62zo*Ft<_n7{wqzv2CF
zeI-Evnnhn10d6x~9~1R3l4&kfLB_}V^TikOxMNPh?r{J~Bp7UFt?i5TYcegJW-lPy
z5|qszGd;d!jCsp!!|>=s7f#M>RA-OJ5W-M=ZDP(snNK`XvycnCa^FPfq1nd$@ml`-
zgCFAn=doikof_QLYH-G<K1=wt<LUI*5N{}_3m9@w`;>}g#bgDgDK<UtbDLJ7HMWK7
z=w^O%{-<$m!zpLI72SI;;(4_~P-(L1T1k7`eVF^K`V&Jk)**;Hw14*-F8kCc7#bdD
zeo#fnT2yNhE)0k(!pc^gIZr&A<KOi@bnzmRT2%0_De@HZ21g2d6;h_ZOf~tO?)q%|
z@i3_|<Ww8ZB)y7mv*yVbgv$!d{D-aLVxNYbIUtVFRjXKe?QePDs^2nl=iNx7MO_U6
zq_(T-Fk-Pru-4*Luu*^)LnDqEj}w|WT)mI^`|ZO)Pku5>jy(o;-310~D2CVuneIR_
zVzZ_-?LLi>qPX`H3YubCZ@M(9BIP#Iuk@B=y)9I4;W)Jzvgi6~dFuNxo~_%78H#F2
zr;>|*wHrfKf)w&*H-CS*zko>B`qgQgr0dC>c?SxXjF#jTHf*AO%dOmX)zv)o=Uea_
zH&Tg}!7w0j34xF4w3`GL0-rj<7-NYOp+!K$z^YE0zWw*%h!;GI-Jkh%SU4YVQq6-6
z!j$hlq4_U=mfmOC_Nq#Fm{(}ZC+iNB1G@cB@-h!i)7%m%HT6C6k8}b<Oia)k!>k$R
zr*C)zOSZHbN(|a=;p%<djOP_!{~Ef_t|YaHz=mj+*HD~yO4J=GuR7i4IAr_IY<GYp
zno@XT114?}-gq51zUza8m54>pdjX4H|60ZXO=6Bb=GnC7@5*rO30y)LMYLjv#SuBr
zaYr1;e}46o1lA!w#EVZ|kPDPXw(C8sWbR#A^1O5$n{NCQTQ;p{pC>;H9js)}CX-TO
zvrWsWL3=R5*~Va})jX-q>aS)ShDR5=@L$05XP4(=cO}j=XF)US2za#ZkycFW4_9;f
z$3M<~&d?qm!`BU~!Wz#y?*jx+I*xd#p7D{vN+x8?cCfAhISo=Hqx3ir##&Sa8-r{a
zhjsUJ`KSJgp}_?l_tvwBciROW3^Rv|)a}R^x|Er0f@-qH!&pNn?jWOW!saM9f9|ua
zyy8lBSHZRw8yai?J`gs<o=t;uIQi_i(|78#X&R{1qs&S#<x%&{fOFbQUaV?2%&W*9
z*h!P3Nm<s?Yc&b<C=j}X@UD=$Jekfb?4E5>%Rb~}cDV_Gi;<)aYaV3F4S!_$Z?9tW
z^82x)4QjC?!WQm5)&^Mhgn>!VhqZJtG_h<*TKJ)U`j#$buOp9V-xE)SeV0OHQ(kXi
z5o7V1ZLZi!@sk`q=$`A|2W)R4KOtdTCs<Zag7Rn?KIkbUCNC5x=4kJX#(w+OM!C$&
z-f?*%Tvk?>8r}jyA$K!MBc^)XbiJ0s1+#d?tg|p_6gv9AO76M(8t%E~TDIJM7jxSQ
z{lFZ7AZ}y4BTm{#6cA(Z-chObG16!<;yMh6u*(yU;;1uUK;^^}@j*abjWAKIutzRG
zH@Yr0leJl;sA!dvV_B&wEW4G?_*!i5{rm_{h$kKFl-7fG6Umhltx+6}@iFQR&+pIu
z5RL0^V&5pDv3U!@!ukAu(?h)W``<thIhb}(qZ$M_@6+4^z)05wLG!^tamM{lDsj5u
zZYhGB3Wbp@#Bs}J+>bBiwy%GdjS_OwKYxYzkOOEAT26V&OSpf`;VTuaQyfAo_ViUO
zp@)GW;nr(@jQ9>nh^oh0L3}qzb4TvHJiJ>qLAGq>x}W@zBTjxUW|v*j$P{}aKJA{<
zM~h)ryYPq{KQYF<V76g+^q>nbSJm&%uFJ;`!XVRpLnqlW%hMJq6^^#;@V~!?pML10
zJT44qZy6&Qs&PkijOTyyvqZ-qgBz@OC%n#MZEC@)`9`0I3y@8+b3<}taVWl{R2mL$
zzm5O;=g)EEkteY9%x4oX-32NZn+HJUO@irKHs*Gost&OMt!9H-s{=RP%+EgYN&42T
zXNfUHi9^K^Tg8RKim`F_IQ29hckX-9xeIW$2rCBXQ<Ef}ka|7+TzKo^a;kE3`$~$k
z_21T6%Xtx<Pv(035KB#jkv#M$bLj9T*_}!4|D1Q2j$k_;Hg90lEjO_2@=Mut(~U%}
z7JU*>NnDCJs1n!!^-2(hI1JWWIxeB*I*jA7bLX??5y!FLX-|iP4}|%HAR#UY3maS!
zJMr!2`!K$gj4GuQS<bpBPbVk3L(TNuZ=35rk#9aZ^7F_Pj5`+dG*ubY&HCszjy`V;
zM2d%lG17zDcr7J)=K%?c$G8OI7`6;EeDyWl@vGmm>9*U68ZG)Q)PPW9G+Q@GCm{+0
zP@$zlLjqQ}8*~;9a{P;5!d@>r9TqO6X(Ot(iey21NVo7$EpPLrcAaFi?Lw&&bl?y3
z3T;o0!z0i*l!IC(TCHk+pOb31NKF~Yg3-M3#O)T9R>+-S`XbA}`G0xByxrKaZY_Na
z=W^ASjlBB83kaTcEDa;nDmB#k%tF|wu0WVnccfBkF#UwE$en}ch|df#vcN~D4m*-?
z@5kRy>()D15em=z(${G3xsaE??5+G^`5MMO1nRL-L@V(GR;Z{Y9$m-1H(x>BIv_-B
zI$WZxSK2{5NR*Vv*J`u!cUNJnAxn-v4p*-tSa3c)PsLK56+1tID8nm6<a@IX!~cA<
z5UA=^!11%o^0Bknrj*=H)GSPNLlQypI0r$J(D>DF`Q1N$n!TDaGCqdPsd9(2Jm(W1
zL5@B;wUhI?xHuQe`t<dS<YqygFRb)Df>DFFATGwV#u@$9ulV^lzQfZ_e=$p*aXQK3
zAvA&jQd(45oLji3nBp^`&>S13?poaO#V>Q`_kY0Bu#V|uc6MP09|mj+JR`M1PX32;
zs66Q@wEO!ALW|LqG9{*zf~r%MV&zz(sT^2z2OET~n*s_u+^scwTFzd`g^zinb!pxx
zq?2PI3z`ekbXC!WD2|~qO6SgHEdRx&+<)B-=*BT>D%4FHrWaU4;5|X+{^1mSRHb7K
zEi7A1!r;<<+3%E-+4F=GV2{OUSiz}a15=uNnyt>u#I-0BfPwN<RZ1vBm*@XTgm66u
zp3O`z1D<Y}(4F2({+uuAJfxs}<sCz=1F@N4ylg4#^V|%V!00GzfB!q~`}Ngqxc+(u
z8y)5b6{_m+?KbLT67T7VVk=e90B<}S8!fg}V6T&&%F(ZW1MImJU$5b!0Ama`x7jV%
zUb#j^^J#JE>snfKUGjX7r~cNSR}c4$KavPTZ>Ox&h%btebybX{1vnNuoi>pgR)6Cg
zEc?cH*h{Jmuir@jlHIs&-5Q?tPak34>Cd34j#^Yrb2&-^ev`S4b@e0@bTvD?Z*>!v
zwXoC`#API&z|e`6=>Gfo^G7~OZS`GLAAcfyf9gEm{`L>>&EMWkZD<~GvyC&L))E+@
zVnW8&ujV%w{~L$yyE{fgf*>6}<$;sidf&3mhCxx6;2&DUoqzldhdk|6bioo_mO9@P
zZ1G5=Q2&1)*NVuAB9hED4F9tY!#h;<bF=I5v4t#5m9x-;Fig-a6o9r@7;B<`zJZ^;
z?|csM4o2HBXt|>`!t*cqGVX{&(V+qGrVCFOVlodV72{czOiY&*Ax80P2}Gcg!1xw6
zTzN59eE)mA=#B3nI_7wqi|1j0s3=w_6)84tAhOA)v1x*BIMRW2E4cjJcO!S*&n{sV
znjL~#gtLOL3HJ|=vG}B?aO~OVk}O()_Sd?xT$u|JnP7W*w0$a97M)TcIDK2OjeAh0
z_np`B21d7O$ZY_9zVVCTNz%0CL?HoIuHc?af5yF6UPfc(8V0<jDi-4ms*Xx%GcLK{
z#2}$1R%O)3j0Y9w?RNl=f7Y`YeCiWn&Rkp=;*4Od$$VB!k2y%!6@@PMf2Z%Cvm5e`
zpcuuBzJxgHM3#BhNJySip4lYld>`*Ufr7LX7~aC@uYb*7etsDn?^s5E;+a>k5sKoP
zZ945Hl_<a>ByA7XI^!s-J8c#o{REEwn-^1i@^ScTA8IT%49S2-Ts~1R?3%MNu4vl$
zXAMJ|Ka)FQO=qjRt^0hsJ9s9Rtvj!h&}{%9JsBb;3sYkZ9oHmsFnaONx$Hw9;pl~X
zvGIXb%pIK19~+xE{+(}T@7KMOmW!#>`ckApx<m3j&!@Cnwrh3iHqN%o-ScuwU<I`X
z2l&Pq(e=x?;caiEZ(vsr`t+yx%4L`E_KPl|e`qlpcOW#xssv&PO+-Apk<WefEMEHD
zQweOCHc&m!ADI?)oJm9BaM)Ih<(FN`lBIhxaM+Q!&|<unLfV*c&Ssxcj{wT>P7%3a
zwqf|6YZjKM>eax$v&-?ZjWYBguoLz)?^6G36-C>MjJ1%z+{}N!|6KOzcwBQFt;5Q&
z&gmcfIC}I^B!hK~no|COFaC&T8@KqBDx;+bvll_cAj1yRY;gAvzs2>xzLJ-o^A0%R
z@gxg|@Z?N(#c1XLBe{e+6R}!wC`qG5rPXHS=a=xSfB6CnM_LSMi1msI99k8I17UR#
z@$_@fVeskCp%YnxT1ejKq+LRH7B2+{ekL>A$w?|ZFUvC%8)}J*-qQ?rDNDJyRdWuB
zrZQ%4(hWH2w28#Q=q8fuZ{RP#xR{mK-%NG9&47U@yPif3p$D<4pfWawCZ2H<vZZcW
ze8^!Oe8zK`d(1H~XAVe^+F=#%SBm9OBzn_r5Vmp`+^&<P*|o+8001BWNkl<ZQ*7z8
zoA4vHWlcx}@8;QR&;62LX!RbkhK~1yV$$Ye)k;=edI`5*eg(}39$=BJF&GK9-N7eq
zOkn7E#d}A+*2ieG$%dqb+hZ}u|IKTe|MXK}{#=@&5LBvZZf_RIs6s};NdvReXBmik
zd=?01RJK|#y~6jIG!av^`!a*??fl75nXWg{+(0H!=~DKLrU8aY2iuO(Yp>@=@A)7{
z_0MPIt`*c4&t+L>69>HJB^+}0+i7%^TD?{vd1?LYOJGM%Ae_Mfi*0E|BzvD}nsQf;
zk~ub+v=KBL+VHKf^7Bu8g(tl6P24*)$Wt!-3Uin2K@fLv1`<i#!75>mWNet%yyz)>
z?7i<G3WC%jDQ_4I)7q*|M;<ED6cp4b-p4flavN(OSk8W@JddPa#TW%IMLy8%^XU;k
z?h%m_L}blu!|*?Fc<(<7oHx4;AN>n#tL<)*jYWGl{)y^Y#V4L%Jcc`N<EQUDm;HTA
zXQY9j6Y+p+ar(zULv+FuN$S;Z;EnT{k9syu@ji2^(XRceHTc-!1A~Yq-qNDla@_j0
zuW;8bH}k@Gorl|d5xQUigQN}?KD9k9?}{pdx>VSlI1kMh!N^8_fBwf=bH#7itABvV
zTavhy^5m<+292pb{z#s9{zq~17UJqvLSyhg=lo{dl%&lJJqWUFW7@d&c(2z7Lucd%
zKkarRT`NK@`EC|=uK2?7MpZFhQb=D?7mwI`)G3v=z}l7EbH&BnammkU-@76W=?W|O
zW}8aYLhRFA1Qo0b9d(SkHsdyA(V>TP*mKUHcJc`@IEbr6DUu+9l{BY9%1$;V!R^!(
zuZN2w%zBysbrA?Apbg#c80l&TlyWA7qDE2Y2|Ehy2Hdrr+kW!j-2J=XQyCd%P8d-$
z4v8JcJ8V|?i-e4$Y<4g<Smog7Jd*=n^>SFefS3SNsbjrIRY@eZvdq!WlwO$yI2v2n
zb?OeRv6(I1ogMhL<E6$#2d%OmC8du-qMo4DhP&_Qhp&Aj$M-E{&As<9xOgtNcN*;X
z;umu0Kb%byPo=MpI?Y7T9x)!(G=HPuiOI}Y&}8RMk<<~bKqAxUSIYV^MW~;P+t}3`
zx#lhBVw;}Ee*5seUtNm~3{vq96~$%w2Vqbl)DBPBcOF0b)>jBjO3|{$;$6B=O}8dy
zbTu_YwZvmLjdJ_tKj*+FJ{8?<w=_hupvL|6JiY!uKR#xR`M_+$@YhEd4pP<M19NBB
z;G=sppJ~p5OyItz$U>Ij>^!#9hFh0$#rr<Ul69ldXkz9?+^Z3%ocA8;r<{~F3tozx
zg>KQ^rR0c^K9H?Jy+=)o7kB~R8mF>(jB7r70mEyz@a%WLht_VpV&_E&5agD1kc9|!
z1!`#!n_;}&qR(0Q(;vC`yz^MFWt<@m(nQ)Q#svY52sR)b{;F58^wqDY6^7VKlv1Ix
zEmbqi)Z8T~$!c(0vYwr~I4H}Iw+6MhF0&?PDo-xiEWf^%5C)$)mq<QM&Hqlu(a82K
z#2S+9YeBmWaSMO%9W49NPq^=zYhcqbLqS9!29r3fQz}7(;_$}MMj3Ok(Y2Vf-~Jr-
zf)}vc(@%yuL&PS;L_wxRNCl80rU=M$<iglqt5cY1!=MlIPMDqJ{Z;n}x;>$HKAQxa
zEb}3~kBy`b1-VO?icg(toWSOFtpDwAx$dGLqW3?<0z&!`Bx$2~60anhP_Y#fU_@fp
z8P6f7J)1*b@fz4?Z<@vtRclzX%$3Xs*=K9iltZ(6?1;BTBVy7!RN0=Ouc@*kb|(6;
z_2-yq0zg^Ak>#?9&8VZT2J+BqE`7_}*nOl;yl$BKz##WG+wA_TGdc8K@1SKoQ52<`
zh!R!J7)n8zQxCCb+H8q-Maa&0oM?6hQ*E(i&p$slj9I;bZ$JCF?03RRJpXrhFg9n9
zYBu<Bl2XzFTO|^u?l*DcA1)`<gfNser$=*J?wM1nCca-qGyM#BNeG&b)z|(30>>`L
zK9#r<q)54LcN;Q$?D(-+M4lufcg!{ne_eFpi@;lF*WaUi-dB8Wnw*86Mqxq-t&tQZ
z(9v;Z<pccWtaq_=c$8Yh(WyA@i#weB{_`1l&M9;TY8Y<|@EE#&){<3<DYrSxS;+k1
z6-%I?BW+~UgZ%1(&(mqdJoDWjpt)ooW~hduL<lDFnOdy$!3vbY&>7uA<YJcp%fE5w
z4}QR2rizZYsa8Wep+`f{Cg&MlvY699{7E=qf8xFxVpID_&C+DOb6Lu7Y9T4c5F5cG
z1c3Erg8fuH=E=*A-i>m{ivykZD^l3ZrWZod#N#w3a0a%FGjiP@xa9}mW!-JJ(bw?|
zgaIJ|E>5UatB5Ef7BvB5ohBO^EzF|%9Q3qP*#9NZgWYx`uGP^<(u7`Y=8}@?QmmNb
z-}%ID(iB_iO|O19C_+ymG<)o{=ZQ?x4l|`VxMbU#1t+R{(gc?Dr?XfOZ06~o`qJm7
z2Z|R^5<;iYOyH)Qx&He<VAXZkGvpkD)*^8S<2qPX@D9b|Ou&dL8v@JXr#+D;z4`B9
zkKKvI5LWxrV$=InZ=(=o3X8n6nLNi~3XPETdY6`y)pf>ecKb@_9enyv`ri{>!;ENQ
zZjn1pdGYwx7~$Hr{P1sI&w)u5yKWO95qF24eO~c59C-GdX^AJQ)DfR)M6wBhs6oZ|
z7~XC_!jL~|yU%)`Yht8is+V%l(jV@Ov-a!X<(HRS$+>rJWK*R^B{11QC8Nq}L}UcU
z*Kp&tS1`8@fmMuzB}auFqa=0F1u|`zM-xRJTE(5${hC9beg^U2U}oXUgugZW{Cb3s
ze=)|qZMI?f>!AzBW$1#LU3-s>9lva%(X&NO)-Yt8R51dv$Bm6+*RAE}XTP0A8%GJp
zS~wfCxjw|>U-x$`e8~%m2db&|fXQ<fic&GB_?XnkTM?`pWVi_rJ;X0A`~m~D0Zx3!
z+i1?2OE^?53e0W+?29yC&BL3#r!zWAG&aJo-g6$E>u+YSYK^3wU?K}a!l)5eC60s7
zd;tf&`5c<o5DnI{Mj=a!&Dm3=x&X;zW;5<<ZWL&;4ONtFObTT}F*{k4u>EmHJ*9DR
zA+w9D-3w}r5POFtF+nGW&8t~|<(1s}vrFl$SjC|8ST6(!fkGroP%(5Iw16$HO{=fY
zoWl<1=oh_&=*Xj>uMZ6Yn^MT~jofF7aLwR@-aLe|8&hU~nL$w6wKtj|l$?sjF_X}{
zo%6lf<6)o+J&kkMH_X$e&1s=QO0U1o=px-;SCVu>MbP%J@*$S}$3@(C<!^}AZDiMa
z9f@1uW4wTKF~(H!sx$)6ea%r0d(Lxt+#B8iyX?VO)liSp#vyS^kh{eeEjCjW-CGy?
zWr`J(8r@7aDwwL}e<Tow$^Lr&<mer6$ixX#h75wzX^auAAK~IRoyF44!vyQb@D<B)
zH_Y+xKZhkRdkIZKszH>Bo--<vh?W!~+d^iyb$4g(dSpd5-TuIfAQ{T48JZca-N>JQ
zb`8(_%vTu?E7%}OKTl@Y>=R30H6$5c#rMDduRP`G{jic?f(Y+Zd*D1*V#m&R7K*6p
zkPHK+(c<<?e$3v79YwhJK}aM;-#PiInQbB-iFtR1GW>D2VfgEy3*Q7@GrR5{eW=2=
zau#$NE#AaAnok^Nw2h2x;1_3|!|v<XQE4SKL(giLaLB9Qz|z;gnxwCqZ7k`=ZcRzZ
znr%Aja>OBh4<1i~p`r$c8@T1m`Oz1?$leDW%rUQfBh9%XmHrxvO92I+6_+lp=mjNF
z2&|>mYEx~;aQE%};2rN`*Rh28s%X1`R1I-x8IhO`l|Ih+%s;`wkE2tq5``A$Q;Lx<
zQ|eWVYywKD)Gk9w43Otli%$PQmq|jrc1}5<-657qsa+xhiBF4o>xmPGaSrQ%l`C0(
z$uC&`i=V;z^~^C5VJl7{cnqO6IPVDq%Xn*?waq4G(Gre&;R{%F`m<rLJ%|a=YM3TB
z`@CUG&xb2JKxNn5do8F&MRAzt^vYx@$~)prGkZJ|&ykt@|2AJubb{D6r@&;5vrpH8
z$sE&ADRLij{<|?!+?z(gyBL!QY~IL*i!bHoAN~lxdJPLJ0TRa;pCE~5697YE14e+=
zuFYY8^L!3{%bRe0{kR|^imG_0+50p_fh#?>$dj$j2?1SnGf!li(r?GgdwoQ%a`6dr
zRDo`oVU7yt!ez;e@bOWCkp`E%?hWj@sYSAJ1GRdM8#<$$@yU--f5s^^LrXQPrs001
znt*iAu}q|D?$Bo|Kn9uc9PK7SlMTBxZW6~Kd!6(G#>NC&$>^G5Q)*Ids74kY-^d5f
zehY7V?ep+S8ynSe-e>M_=HWe|Gs|nu1KJYL@D10qW#fAGI_Y$L)$|NWBtzHo4=OuA
z;F>*Vbo^gq%-_y7439jzaI&giIlJB-n+t5(oCUI^U?X_vah*2dNQ-OU^KQc1?qW{6
zO=8<@sLx^EGfv^Ke>jIEtYB=Kv{lYo$aa15K0>`_zTe=DLDgW?W10=xf4+%}zI-9a
z9Dfpryy#Vo4^^o2r{1N3*9kjCIui7mHEXNcr0zW%|Ko@J{u^IqY0D5a9D(t~(#8*j
ztZpmyBah&O51mhYa4xp5f|U&IFl9<pZx@wM4pSt=V=)+Cg4d^XO4_=}k_nU7t!Ihb
zPo?Qe?ah!O6Bs(qA#p<3iQ$2}x#MR)<$>Q^g=`+BUo6CqKn)e|5phT*!W*R_!W!3P
z_s1W_5r6l$1cx0?Z2FMMVr}Y7f|_E=*QFjzI0FUcO-ppNY`Uyk(2$DBvrZLoHnpgH
zI>LxfGZ&uScK;{P1}&P)f~7$TWulv%k*_E3d_iF(Yv6NUxr!knAxH#iHra5+B`o{?
zf6%$>es-z#LA!%@F;11(Cs+vxDghf^gO!nR%A3w&mov`9SNlk6hEOcxz!j9FY)y(#
zHjgc+N?8$@-+SI<Hm<nLUeuHZ2|GRvcqD`VdJ1su-5<J{6xnmhXL<w?l4gTo!)C5L
z>+LL9xgIu*QJpiyZOu`hd*K%_$37nVYZz}borupy3f<<USHvXK;wsy72Gad2cgYhV
zCXtZGpLQmz9vnq%h=c|;28@uzLSzM-jPt@%Pv-L<eFtISz}9eCv$JE@G203Az0WCD
zB-73$m{qH|<;p8K{8=x+%^gClnb1^cXTCR<efHSdwUNjvBJ!)*hT)M#7k&VoIlJ5*
zozp?t#t%$}Fmywjvd6bqlCc(ni}~$2=Tg1pFU;!*2~9SI!qBPD<*@g@gHFAQu~GI+
zODA=C&VrF_!^i@Ha#1oCRBcKxXt&sS)i3$)Z+(kXpMM6spL#m&p&?9U@#+cCbUJSH
zolJ^G1faEPl&VR%>BApm?Js}L?v;>A$KWL<5l_+|argK*C%o}27QO7Hv};v@YDmUl
z!W1V>x1wbg?nC+1Ef)=EQ={HfC^Nqf+fW{C^&&di=2=SJ)~t}SijNiNT7)rh-+kQp
zz3;H*x4$E5wCNLzj}wAwgk+Yc6=QK0Hn!Td1|s%)^3yo{O|OIbi-`Mbs1*W&VsFn#
z0o~$lipgtX?YfmSi%in+^g6a|-^_dV_-mpL(|LkT{d;LQ>$8SI=*<BrH_`csz8m^w
z3WY(s_b7n@pZ}adlIt9X#~Ho+H(dMO@8DLhVL=pOlMb%aB=8=i2`Uz?)mhbO(cWb-
zr@rm2^q=w+nkJ-Lucn5bKEuF7pcFu-bj4pn*R2NyXR`YOm#yQ6+c5kUTHA7N!z8zu
zW+9sSBtOe?Mg?4Jgkb%8{_Ai5j{RF9vSBmTN}WGxhnIchYxslrL2K2Fu9%Kn5R>%`
zQqV{z1}p9q6gFeGc}WuxI#{0ek~eVstt+rW0FgnlhzN;;Fn}O&Jbs@&`Nj9Yh|olA
z9WQyP9VLCi%=0;V8V2t(t5!jc$2Qw6zw{R@-gkfc4?Yx)Y!6bM(sc76PTe}&ZT8q%
z#}AD$FP?1}9yxU3bXC20cAY);Hu@gXbLo4kIh1R%Kr$t1GzntQ|6}jH<0Px9_3!WA
z`-Do}lgT-m5Ea23P!S2DAcps<Toe=$OsIenGbnn!h>8lPs~|xPL_y?=>1R%$WSGPx
zCU&TF!d~wmXP>H5)zv+Nz|455?PqAas=K<*KD$<U)>_YV%UPdd%}xKyHi-q=G%6#g
zCqI<~&-pY_DM0yNjvZ0xOzyHFIoi}b$g~QqGN=elD`LaX|Cb+p_uCx(hU1v~%x95|
zR8WEE-j8$!WW-T&TC(<=I86vPM6hZl*Sz!X%v-sZc`6%%6h9>iea1Xk9hP|6C(nlc
zo<uxSL808dbgmrQMYucFmpnL-Uf5d|19+7&N9hhEHz@R1Qv#>KDEHvnLXT#&6IsIv
zOy#0kIlp~mQ*_+MYqnwWo!olKWi0*0uc<T>Dq^sSB~%JuD^is};8V9SDHhpzN1lB2
zF>HI-OJL4CVy*B4ulojU-a^J;odo_>1gzM&JM@PCt`2LH7;cp9O#O6_`Dug;2f%>j
z%I)2=l8bhATbvR}(GloIoAp1xiN9R@J(_pj!w!`ys@1?q8)XdbIHD9(iIrvX<OaMa
zK8BZm^c>h}7uumAtdz5gVk(Hv69ZTM5a|YRfN1N6dh>ufYlG^tGCX9DyZL;(D(tUx
zRVj0N=6wHiitKiMoN(C#TzS$v*t-!SE5`7)ox@*bjQ_awGNSF~qbnn5$%0<vd`~9o
z)`bd`=pI2yQG<b(nTaEMK^~>@s)!}V@bb5u!LNUP3&uByuQR<KrAZUP^8}qJ=7%1C
z{rM$$wDH1f=FP-9_3BL%p&N9X#hqQFH8ugf6dfmQxb+{by#GEP`=XbUR!VMmE<e*b
zyOPBgsbh!c=aHx~yi_T5?J!|@WU32S11}vOv#lQi92k76Zt1@`Xqfk*TPx^hjIt58
zpK~rNuKF=M1~n#=26bNupYlwe{+Y9Am&<r1KMN+sAu&a9B<;$=h4vE>OR7C|1ZE;)
z)lEO+`pYll6(_!#kprJfdv=Jg4cgia!M7O&V_q}2VzI^}ZZ|2nW0D(x%>b=HQomK7
zKAYXFN2ygO)QUtUq*cx8I3eD9FAhHIvn1OtKvgQ4c7Os&$!08a!MFnAHGRFNNDJF-
zRqD3VOl2HVn{Y>D#`XXy7&YBZ?DELnSd!R6n&PLxqI>zvMHjR1rk^s}N*GlN+ic^7
zL57HGFv`%<F{>I)Dtqq3vtEB3l^4H&v^+vu_3+S{gCB(SP0gWZ2y8ZWn0qF~Zdgs&
zTAb6yaMg`Yl{lC-PZ<97V7SZRkk9sK1Pc!L*O1x-6Q$5-(f-Zv`1#kqMzVMr+gA<o
zc$0DfI?Z~@wHMM1JnpMEc+BC4^5j!bfzjDCLrpoTkfzcJK$Vh?s=QTFNP4n_L5m_M
zYkhd#`etgPQ{Aj#N^_<I4jjVd=x>4g1bW#Dest{d?5Aqf*Q`d(t8&-q2*+IfZK9DH
zDy-la>@8$Tmm;@mx?@N4cCeYw;l}-7R?V>1Vn~U!;>0t~;s;mVj8~~3Mk7|a88iiw
zfHjno27kWkVzwDA;RO{^0i`o3(2aZ8>2F<r%_ukbm2rf@18cbT=hw6E^A1L~-yx$L
zwzwHDH%EeC-F%JNWWA5!dXVcHt@V-rE^qeVhGj@Zjv6KlTj71#1ZUp)>iTj!1~|?7
z@l8fJji>~rG;6;1eHLALEj#Nn?Z!AVYc7qQ_u%Pgf0DRdK?i;Zao0Ijreq98y1<uh
zT~#5~iq!a-gSTxS_}}Zf=Hl;g%$wfC$N^8MHCjPw;|3qWMF0-1%X&Yx@n}spD2s64
z=g;GY4}6Hnr<!uTg#u!e5LbNeYPQ+t@RxG%SH4bb!8WK$6+~r(j*|7_N(UmON6U3;
zpxWywv<n9E!N_dB!J+@o<w>_go8riH?>#As0d+%<0t=t}I}UxbPUJ4lhKVCAag^e>
zBm8^r;FhyL!qq1p&+2QhWxE7Qse-nq;#V-nVEvGBPqV~>`o8<}vTuHqgD<%V@8w65
z%$tR&dIUDFUi6XF6e<e6S2W$#Z44n*edbSX!pWZQ>u^2Nwq;WSl#Px3BL*c|GMXh1
z@-M%m+<qz7GSlg_cR{UD8cG4loEqMtFW}J2zRLk0dKW83Yt%g-<5jZ}E6*bpLDyUC
zTN|Zu^Ut~bl}EGe`<GI-n#R~<){<uwsR$`5vts~au*zVSCe6j)x^>lusX`!|xzf-T
z*UW8DrzJBFeDla&ver{wt-($^aP*hX<reCMb84hxO?H`Vanq@15VR6(8fBGbmvbnQ
z3@x|H`HE}?#Oh4kY-X+JnLqg*k~r&Zy2_}Yw;ibwk`^p&mcSaNQC2X<kczN$*$R{~
zoi^0of#(O$WY_JW&0F21t3@(v4wY(+iPg)|ws?VVjz>vGN1OYaAzAbzwO;CC+hM}+
zFeeOe8J?}J2ld<7<I<hW?V-z)e5N4x5h_D9RGQ*7Vjj5S2mIlyUtv$LOtV?Xj+9wl
z9pTU~o=2<XA*HMh$s!q5Tn_bh{+P~~PV(Vo7x<fILNMOqo-2OH)t6q*>)(4SwWsY*
zG~2`XTrlcbQ!=NeozYAIMY?8!+GK-Yp71snUHU`z3M;639nViu6_55vg@4vt9Qx5u
z@VHY?qg^ZE*UDM$p$hifm3+I17Xv1Cr-n^xN~o{8uc-#P3+eg&{J&BRwEG&mRrj;j
zLG$vNlFWx#;;2s8iU{sq#6Ld$X|6l|%{=hK>zF?lF%nt)L?{J@pc+D@#JEx{N!xf&
z+K(g8`vNaK|I0*A*`H{%ieCxPc!+4605qbDVR>(2Ae(#|Wh2jRYDg*@`8r%%B6)bQ
zX=BglVUv6Nr+zq9J|839YMyceVM4blGBZCK<@>}{pY4u0hL>J+5xc(jwcIT!ljRbz
z2}rCZ3>CIHPOTlYkCeFk!tZeP>;8kv%KPyqCuldD1V(i}i#it~q-amN8Gk+;{m^q-
zwF#fIhts1-0ac<9o=eZI$oUJbvYFi|mto(>a>yq?#NX==P@X+PWpWbzx4&`s`4<w#
z5lI?36?&I|bjlU^U&-ejHokE(({=O#VtLtko4Io!jWAM}MRUlA(gu-?kmKGZi%H3J
zI<g?&J}hB+4us|84;FzyvtY+vS-WC6DvELC`a%U;J;=|?a6KAgTm5gcp~nD+4v*I6
zO8_2>!IT-oug+ttV7??9<}%u3PbM*px8b+H=jJc{H@gQV;?@|<s<Emzhr>SqS(4ea
z(bXEBWGy|pB7uWYI=!i4)6M{N7V~RNWYJNJrPp1>%~xN?f1Y+a!4scEG^>JV6h;JN
zvpI+Euz+&nO`6u4wU{+4xZ$LeNN)KjJNRXElAr?x6~#oTSQG@j=6}9P@ce^`N2?ew
z1XO3RS!;S|>a?2&$eB=(neR$(rA7_VV|Ld(CA6B3+noYeyPKnM?<L<pNwX9sH4sM;
zG#mIUmT=z}&*9n=UeB5<uVR~4O3nI|Bw$uprDQZg=&?3#uvo>69(D+?yyAO2=S%04
z9B?4ERwXEx3Gh0xj#XWFuGml9pE^7k_x;EQ^+PARrR<bE=#aMidHC0431v#sA%*Xy
zNY9(MMQSi8v;`j@ynuNAJob9W+d2A*%P~(skQEASwM=R?V!#*$6Ak7y8|=Ae4VNEt
z49mX#ZNfC6S+8d^yN)eEX-C6sH9iK<s%%a}6>NZPYEt~*_TZ681*iUr4zM9^R*J6T
zea3$0_@J~0r3%&O9K>@^eJ^)4>v+az2g=-a*%gdm^%H`2f+*`Ka`T~Zal()(nzRF_
z08<EUH*FURtyCJUW9myqF*-U2FomQNu!6Q=M4@d)*L~l8OB|SHoqv!X+J=A4_GL2+
zTB}YKN@K7_QC+YN^|5hiOy=Ef4v5Pt*V!q~n~{b#?-5*wx)^tuFg#2N!->N)^@y75
z2kp40IBV65cB_HR7lm)Mkcx#=Q0tp;+wENakq_~t%536hi+07ZJS_2&bIu~&aXX}1
z#sh4cW=hSD_yTOFzgKc3+-VrtRAJhn#v9!6qpP{$$Jg?P(@(>H{9d$X`}m$lr#53F
z5$Pe#v<lko6g@sk>F(RO?AYTNU9ysGy%0&8h>5Tj#fFkcbJyK?`S&g+-fK_noC=;X
zDC_hRtV=(v>Yz80v#+Q^^<bJRgX-PF?)N=toOG2)2ZUHN-eV`YmLaE^@DQmh1sO=v
z7}aRuKd_!97hlGe$DPQX7hTRajhIT5;%P;>>Y+;(>s#tN=HB`kJ0Er^FZ=Fy+5f{I
zCf;o~;<Aqke2{FQIhF1}W0wI+O&hRmT`p@_Vb~gW{NZkzQUCHAD^&5U%>pjRc9az^
zEk+rr7}6c*^Ne#o%i*6thX>}(WtFnnaw*dhQ5It@$*M8-A6dZN-})XuIqnUVm#@N%
z)rrz1qmXu55mqEap`FUSNN>{Jm4Z_bVMqG^P3Hw3u8go-IqeA9sX~3#wSna%3}OY8
zAqo`RANx9Xf7LOpv~9FD?Cw{%@?#%`g-em~2@<iL{3)wY<jT_BYf}SMS=2`SpqgRd
zw1Qc)N3k|_m1fnQJ#&bDnx=TZ&pnHlp?n{$wHtWWOjCyD!Dz+|rnOLbbLSEGA+7bR
zk<|7~k6008dZ=l*9;Gp^{x{gjV}L`3M{4VG+*5iw`=|=1w$e%`(A8u;#Y#bYLmSqN
z^Mena$v&n+x^5j_t;%xG<B*Si2=;y~QY~RzE0Z?b`AbJ2n3sT6hK^W-RBKQk#tNhj
z?!EC^Zoc6<UjMFBF?&9a#z+}Yd5Ci$oncz*nu2P4(EtD-07*naR1CBti7e$-MB_)-
zaLwD^!xP10PCLQaCiq}y2dp$2{~6EWd0+erab*-WI)cf0j<mSm+d>nN%Kn7akcH-Y
zI)H5kTuDvAwL}#kFDly(_H<Ly3hRWb(k#$Xezw<XqlrH;M&qZ~@zayu%&lMj3Jb;~
zW>Z0>0i{wk6I3!D8?Y?$H0p&f;uY85z&;=O7|lKRz?MUN<KtWHf~lM~G9Q5NO?)02
zB05~dwYk$kDa_rt{_s2jM;v0Dp~a*Y6NK1`MxJ>9M_hanPdMpKEU^|{t73wXhK&hF
ztF-HrEU+m%Kd^=$9s3%_Z}<u2wxw0CqqV^}n<`q9N|r3Be8y>-gnQ#C^~UPQGd=cj
zY<>Y2kzRcSQHnG(QKZh!3mF<CRtQW;qJ@3m^L9#4-=DG2kT&WZFmFe`|E^QBd4^Vk
z=x)+sJ86mYN1e*8Z1gGz?M82#n0pNivu2GrmaA-TqLUYCH^A;?-`}<HewvZZl0X?$
zEq?G>m?~6|^S^Dad)**l*8J^Q|G+YgwTKw!Q}b~W$@GSX)rHNm4*7SeGHmrH!CHGg
z@RH$q+VTfQn4S%KI@;78y17mpaW;6CjJNO_4X%6ho7rWoMX+HtQdZn=LY{fXJE<J}
zV%oJ4e8*3Y7PRfs8d}$uWSv9IXit@!Sx6yjapzUv=T|r0#PRPt4ZX|mMAZtOHdxy&
zx@JwsS(#rOO>1J3QZr`RH@?NK|8p_Bm&%wpMZA>62xjgmcTd)N&dKj&)=@7f4n34(
z0}`j%)1`}59mP#4IE%cGbvj1RLHjM!4Zih=IIu}1LH$fh$_YcH5|g$fg0>+4yp@~3
z@C7DryMy^Az$6KNSR!?Ff1d9#iDg-<%}&pLCeM2NJ7M=-XnPvZ&-D{BD%mvekCn88
zdS%q;PSLuc$o$9&M{SNz$V2&=ArHyF)txP9f%Si`GHw}1=0J{$TTC{wh0=;t+IUe4
zcQ55<XMPx2xR3?q3h88>iq1$J5}RN~M_3hE%FlcnPe0@Rq_rxHl+nhJrZz9NqDWm^
zB~$D{Xa3@0bNdf7d(t2tZn`W#()Tn%#*XKHC{vtQ#H}&>@g_ff)v-K2vcwZDqHSu(
zllJ3)kA8%x;^X@v)-eKk9l*tO*y@zbY>PMLGSVc)^F3k<f4ub$j(pV{F|~QbstXcn
z?Pbc!0;=7nY})+och^$$AT&W%mC)MF)IAs-hN<_*ie!rBZh~AXi?Wv1!o@7U?N2=B
zpu>nqYFSkrlhFic3?1D{&u~4`>jtg$R^58PRaY3?^!PAY*ox4%q8XYF3zotUi}R>?
z|7l(zjmH{zX_ISCeJeYRttVK&4hcNgg(Y@-<twQiazsX-Dmfpw^w`Ue%CF?~sm^bT
zMzw(?X36!}ar5;z@um-a7<S#8s8Yt)3T;!gWXwOgnHL4tC|cv=l*T9c-Min<9pAl#
zUHuSg#b7d(on-bX3r)xo=Y5%3M;}F8E2G1(gXQR)#ls~FIS+jIYwCvVY@fRG%(&k4
zUD!EQWTC>a5m2fpS;#XvrD!)>m}W$<=pO!l<_Ea;o$nyH<6gFxfKsfessNL;2?9gi
zSQaNOlBYhISO4Hrp7r_vCfaikC<lbTp9%lC*}{~3;4T~T%ed5COGQ44EyWp(hlcrX
ze!OMcM!mlOzp0#J!?iWqzClk&k2pwIU0j_0OLxYh%bcT7kP03l1Z7%#?#96vew%&Y
z^j4N>Ls}cbnvj?j<7-T#%}!}T=?{P4`>#F{bJy*d@d?`P2+!yOC8&1AT-|Fggc*_r
zd5~BC2qp6aKK9b*WgIbX7djDzi0<42vEYZZh)V&l_}aPLnvUUD18Qr=8UNYOSatP{
zcx~5u<L0JveUAK!l(11-{2*f(DwPUKW$j6wN>3;25p0@bL`cT#tXsYY&m}KifBDc)
ztj>KUs+&M6MQP4_qIQI>PoS)ua}bNs`92RTzMFBqWNT+k+S(I_Hx19wmJf-h>tY?)
zc3-3nbx2V^bB8n%d<B2~_{Z30<r>P3HgN!xl@k7e2eRi|PNEt31jf&tugw@7jLJ~A
zEcWa=lN>n_tI?4~PqtWm^|k!$>YwnIGe1hY<IY62Fq<t%q$^INGju!E8Y>3bFiC02
z{rvQVlaasvgB^^Irxi(PiOQPsvSIy#`Ml)(3rQdUc+!yyS~<E#E9fjr=$Z1Co}q*8
zd(VsE8*@c!z<rn+qV|=pXKHuNnLs4drchdl+c7jE!m&y2I{!j`c*04nyZJY48$$)q
z+G4_##JAX5$fC(H>O1Ys%f56z&;PG;Y45%VY1tzPO*T|0d9I+GFkPV~!ZEpYDRDDr
zrd823{=48=gLr#ph=7Jk!N0v7?EUOHm>{3O$<MJ;owF@6RnH_75Yt#Ih=s5m63-6V
z{td@+#Dy19-)kQhB`s{Zj3`4Ar;x^&@j6e^Ret=&6It=yiz&5YS`&2#l;kERt7Hrs
z)*jq#uQuVzWt!x7Gwiu^-=!*i?LJ!E-^}oMBcrg}4!rO^r*e1Hz&DzGf)Q^1?5EN9
zErQ90n_J7<lB7Vgpf(vE7;IvvHh+~$VK=es%wyPs`h3>PR`COFzxAJ3D<mSF1aQ#!
z{ew;%I{G%Y2PC6frdAta;(?VJ<*e%NZ4P>=!W|f{M|aGtx8jXE95`Zlgtn}c+*nun
z6r8C;Usq5DknZqr9HnRrcU^cP<G=p{bLtVXO-agz)*icYz?mPXWx)50qa@9T$6U`4
zIkS)=wKy*FtdBkur)jh5*EezfkFVejr@kAu--WnZ!k{~)4-vW*0HaAH#ndDGzu(E#
zZ+-{ctz6FntMOAIl?J9(VuezuryjubzH%YaHglm?fi&l0cdbw^F^~d!u25NjtTkx&
zPg%VYN3CWEHG5NtehjWVbv|I^F>ix%dJ3r$+RZjWD<Qi1CVqO-3EX$lci1ITR8t=>
z^)YC45E7S5Y^apDw^HHYPk)BPF20!7lb?dE_=Lt_tm@F^+CmWBB}bIPxlo<^@E|eW
z;To<-+~#dcdWh=ICv}T@f((w;3Ln5csc<7xT0;W9F-XND-D787@Wpd^_8F(KD2bp_
z!2})=DSlWYSvStUr7Cw__-$@F<*k&)>qO&|C<DgXUQyarICKZ6&WIeusoi(DJvZe_
z`PZ}^)Wdc|dXt6BaC8_@3j@)adq`1uSij7?qmE|pS0BX#s!hGo=1H?>bM^a9N7o~4
zoOGz1!4&2WWYg~d3>GWX+v!#<D&-Qs=XFkq6vY2Btx=0jQj#d)*EifupcTd_teb_=
zQ*fY7tx~EfsZ7!KDy1RtnLU3S#@4RJq*4%6%?@ZUsrAq<Z-#Y*i+zU)!-Gs1P8gn@
zE$b*{rsvf(iGy@${ZzrZqR3Q%lNQaJuIKhkzt7I5M6=OE&z;4pks8nb%x7qX2EQC+
z=$p!zd-8!~DO3eIFY2loegxiFlVv}<ksn_CUEX}=`_X&vN>o#XMxj$77e(lzJV4SW
z99vK8>Z`c^wD+)EJ0)l~F(x4i1T|-b`%_`}V_wbvXMT!yrGy!&WW&KG=Uk_U4?Q2|
z&HQ&S;x7Z+Qq;6~-_*vIetiYk_RxXjx>Cs<)rI1(SV4J;IBMZH6T)RH_|563^ON^~
zkXcKYv#knA8%;_jgDQF0O30Ye+>^v?d)Q&T@`qPYdcliGsy?NPpS1-jNL636TH!LM
z5`D;&ipg*d*H+<oHUSjrna>{}yVw=LDg?T!@JeoA*|>QfF*z2ZFky%t^_hM2OL_UX
zFJk%Jxjdl2lxno%h#*kN<QUt>3G?q+#N|gFP3fK`=na#^X^b{*z`CQ?khlB|xIUTA
z_HSx}@JP1T^}k{1s!NiKwn$+(*zK`)%xk(tTWcPB%A0BKxeMvM8u7Y1bJuR*)~|gH
zKZ%hv&R*HtzQV~vc<-v;k_%zz<N01!GbkVEb`@!3G|E6wDs$&8e<NPL1llb`DU{CU
zR77>#+hp3*rN2tjPbu8#o9pbSh-KEc+fg4MhgRF|sqH2X4|D!ET>sv&uii>G>;=Fp
zhDT@1I?7G?ncmT=LeH)U)+)3(71Y#X$0y*n+qn6^zrfz6#N>ua%$$%#3J&@F7id?@
z_~j7g28WfkZcZUX^OSZ>JE=vhteplFo&_c+X#DL@Tz&B+yy>*}!R~v}s#Wlf&R(a(
zfbAlAt%>6{y3u6p%4@m)th3lTiBZucx+KJvl(^z?R|}qc$~)NOl(*Ba1o)*8ErQ7l
z7Jd2(`7lMVv{T-OKqt7Dfo@u6WQ;Gegy*gxRf;s%DUh5wqz@T%Xpz!_wL)`z62Cdg
zn#(TX+T)LdKm38+O+dM=39L}7mJv@Fvkey7Ce`O1!2kEdD|p<y-%VVpVn)h%7_8*I
z#B|t>I)bdKM-y%ehtqHk*T0)>RMQ321HtLO7QsRw+7BA!>hyUWr}2<-m1xghIrtmj
zVY`<f!xBJ`&Y}?|c$E^quhHw)v7fPA^dCntcKwz3jW+d0JwvyJj1kDSDH#PuGD2rX
zMon$8Jq%BS5%!sq2ibf#pl}mZCk>kVE(MJ!g%wNamxy(l7o2?-x7ie33K?xF?z!qm
z@aMndwb}?N#$>Y$6yz}<68NTiH9LIY!}CnRF6Wi!%yxQSru1A-64I#6(*O7WFyjqU
zAjzh?v$og5CdekIwrUF4r1I(p8jMBHo{f!T?8FAAoN3%X=UNpqQ#p9J9{%+T7XuFy
zhD}Tujvt<x&AVn|d)`=Qxu`<wAO(Y_GUgo;AznX$9^b$ZK5!;`S_Ab4Ud?Ao9COHN
zXTWYdV@9f3>yi?%DOzbKDCW8^ofwYNiqu-HD2!Fenn@;ay^V{`{~E`h{4V@`_NG-0
z35|iox<ERcbx5(KXyUphn2cHY#jo?5&!5j@$|IO|i!c;y1?r(@snHzy`Oh-<WiKNM
zH9^UD6$RB11}#?gyP|CU`=_RvK~Oz?xH4d``ycDHK|t?(v$Y*WQ;Miu>b^0Sb~{41
zW6Dbxa?{E0;`isC$M%hwN;}OGATPlNLeoavyI~z>-+ejyTi@WoPkfSO`)#mQA8i7p
zsH5QAtCqr5DmC!HsLdT{4A*cy0*Ql6!&IjF1Kb&>W&1nQyX}0Lo`b@#R7hrr?ET(%
za`2}=!@bH7&zVbId&CApKOk8%#sRbD^UDvP#qFOvhiYu8PfTPCWNs*1DM)RaB?peO
zQz_k{*_`SDW`c&)G~R(tc>G~-P@;OElTM3Gp*<`Qim~9A%E)%xamf2m=e~*6gq~q9
zuguTR`UHC2dZZmWy$t6kT&vc@rTn4*qcvgR_sp|&RJkPsade<)B@rruKVEVLEMI|&
z5)#{`H5Ds`yD-!HlWB5^!T`Q&cajt+Rj5>FQD48Nqg&eBHae^Y4`bl1G+_{t1A*5M
zkIa^FhSOn7Gg-u3CAeb<twp2JDys-I8x#1x=H@d$!XAkw-Y|ib+N=>_zt<gy_v~lV
zE|n1L*othypdn7>Gz)|??rMb?AEPyPauOa`!uP)M6^=gc1gekUpL(r=uMA>SN81U8
zEcwvJ&~DW6TTK?7`vva3=u-Ca%6QFohV%rQNw386$}A2)|0~#ipGZ;*(aL0wR#Tt>
zl|HxcSUhsprk=>5_n#VwkEy{vviMywFB8ENi%07Emz^k(&AyTd%F{$PCTT{5W8*Bj
z@B*$n>10Z`-^;cVVp~8M5-B0Bge=!74{W;)FTLP=o^$?}Y3{urNx6#AK}PJPGH&_2
z4NFZ4dWOBO!}aI@SNaX8z@}-Msj0u*q4exxF4$D?d><(r(r4_?QI}o9#4bCr!PgkC
zhSr8srA)GJg2$IfS^cAHxax%0QEfG8Z5YpR0<A~{L~9ZOtu)4(%#e1_h?%k*VJbIX
z|8q?9I}f*m>P;4=42l$H$1KiHYYo*ZqlY}7J&!zs^=X@Mvc>5AYq;loml4DfDFtd;
zDX`li8@-I_&Vo|igJ6uN6!?w`wlmq6{hZF*)GWsHXoyF;YMf=)TunJqXk)SgdAdTE
zGvBKj4qwNGs1E(P(i)7<$oy?tw`L{UrdY{nu)C<#Fj?3F>-8=c9wrPMNf_QZJS$t?
z2@cX(82m(`+U{CzIu1{Sq}8JA2mJDVXHvay36;q<5(?viNA=myV8;{QNW0=^s+O7_
z#bqlQYHme3LArGk&t3|8eG`@~;#;3To5NrEauz)209vy&Wly2fypKNP`_HkNWMYg;
zeFJ|z;|v!6;0NsLmq_a4U=pMjFdmlKu+7dKalto<cijy;tAaARGa%PdlP<7#Y<x!t
zaT}?@3`b0*B=5mB3)PJdPm8O(WIv$u*>IHBM6F4}W)uIff8yqMyo0;Hb1^$MEMWv<
zQ?w^|C7)KPSlDW@$19HE@NfMO=IPHO8L8m;IqOlz(4BQyxOtGmNb@jB7_Kc8s&Adh
z%S;H&G|y!f00%ZGHzOqg230B3S`c#31z%?N!H06|`UjvkLX5&_jcT@7kic&DZ{X4+
zUPfifD!lO~aT;N?L*gB;ey(yX`QWzNnC<G<0FW8%+EhFD@ZEYT(8U!>yg9>KZui{t
zsv{xq(>5uOKm8O|%^pFAhPhJWpO;<%x8I3swVYbE>Iz9dgtsq!+m}}uI`Dly)|OXA
zL^9z$-)G6{wd}snUflWfpTWwNs3av-wsU>C^0iFKWrXReoclMjJI1UeqgO4@ol9eU
z45C)16q2{I^(%c2*P}WX)?3*I?E?(kk+!g(8x#Vn=^UaeU}WN2Q?$;$PulpYX3a%E
zp!w&&G13qk@dQz|%GeIO@zjrfoR$tSj(J*(?et3upp7Hfwpk*Pb(uOjqz!c#U&|$5
zIiKeqdN@11=un!KGNG@~X^PGjDs%pSV`z_!;g3)7+cQ4O*sp%aw#p-JPeP!uHO-24
zM02;jIryvR6V024j8>f<bJqKdI4V#(D11=^c2G^5icMuVVqp3(MKGvH)0pqPt1O6h
z)frC_r<S;0Czzb%t}lIwtKW4h)w>olJGMw8BGeXREUF@`Ovh>Ly$7$l{Cn(i@@d5L
z=A+AHjA%T_G%H+Sks)HLAh0$}6ozYaTd9B3bVWAu>Rra;to6}aa5@3n3Z>F0aS*ca
zd)~`YU-%M#n;fU*Rq@KB#73d5##=Ye0YS(`uQ>|-`g{CVn>0-_4FHw9eB+dd2S}gH
zXodQtXqn;b!)zZ0F&Ilhuu@5fwlo(flfs6EW|Xut!plDSVeU>_#741~U*hK<`y^f?
z#-=goE-=#HwlAA-9jY)dAcZM&t)N^kVbipe^jNaF39B+{R}qB`eAd#YwAXg5Sh$$h
zU;d14#*X1i^$b$0P1c|8eMt4FPUH`#lxNQ+X-6<Qf$r#bNGDGljxcPIb-as-hY7<p
z3BwzJaClBOzZ2Z3&H__^P<rCxT%=5#h)#-WM%X|6kw1L%0(SRG#8Dk@R)wX?aOipG
z($)&E>=c2m&5{k9_mJ9-;nxbRO`UJhB<otJW`ip~cMgwz+>_Y*@K-V!h6KJ3Nlsgr
zA#^boD~4o!lW=T;8&5iw<PW#7oe8jU3*}pqs$xmI&79{P#Ph#+9#O4?u9fiU4kG3W
zO9#H6v3Iv0joTO}zYOdSp4Rl|RqVVMSs7AmNsS;Lw3<!)R)V?p7Os8cNi4kLN_G$p
zttLSUbVU(|mbyt<EGf@>&*{AAOBWLDv<EWk<9Rw4zjltpbqk9iwqNIA*xx!_k3LZ4
zk+8$651A3=022h1K~ll@ebU(;^1K6h<t3N1WLAweMv;Ue4Vw@IhW3heJac3Lm%sV#
zZ1~}I_^pUEjk*K{Wb8O>w6f6uY*9dMx}V#8It}!IDBUnv=R_Hf1R_t0bh!dg*_T}p
zJ(PO2L@5TnXeq0%`6+%oA`v$PUYKQI0H`zQOehBHUE~CI<x&YXU|XyeQk$YYgHj$g
z^m!nxu=^9A#4o=4b^K=BnO%^gYPeF<JLc)`sp?bb$P-AV(A6@futaO!>cJLWJ#1<7
zaBcbuT}(Vo7zQK@^MPSI(iX(*WRrfdz_(C{i+fviQqo2OV-sBQiI1^!5YTR{Czw@c
zX|u`8&;BHKWE52{Wd~5%Ool#Z4Oa@I47fT$qCk5gq6|@^O+d=^U;aFEw%v}WzUDaU
zo~Gn!H@KE5fNIIO$c;9{lM{HWSMcl8&!Dz&3EPMe*a&Sjw&Jrq)ogp{OL^kQKSC7v
z=yJw{l($4>{lx~9a_X%z0Cn1wep2akzWtgCYNMLK{Z$Qd!+OT^5GB}Fo65un?)=<2
z+;GO}ELgFcxsigF5QG7dBt)KKaoomx_JJJr{cG?JJBFlI!3%?oamocL)(KYR8WT!s
zWZ<yo5Pll2M=#j&u-akitte*Hni*=I@$y>?21Z9{@3w%KU;Z81`|iOyKc!TwGSO&K
zEmg4VCU{bHl$$^GDIU1=QoKe=5~XOPNL}D)9XUloJ9&m{lRl8Y-%@RXM4=Br>3P5b
zS5{OuNUl<p_DOx8$G`I=me)e85w@?)<5%DOCM;hKtvIWq=5q#)nmX4K+Sq+pg@NlL
z`Gf(g<ua+Yo%SgqtEOuOT4y4DS{uZ|KUS^fNrxZ7irfAQcie_*TCB1i|EgqAYA};_
zM)AEDXgHiAIieu+sm-3v#QL>p+mnO!z1q;$Z|3zz7aI=~hQ4(iu-)){Y+VkpgDG`Y
zbGpH70qGXHLs`;D@S}tqPJI`9E1!6*j;$J&Tj8m1dJ{bUvBV=Kj54GyI2G#x)Q)3v
zrDZKBQD_TZT`0FBe)HAyu%tZi<Wp#tpjtNB%!17VT<vBSv{IyF<CK;z;>J_nhWF3g
zm?M@ntrPnR@kqeJC}xkBy_)^b_z-cqgbD*R3ahdKL996AZ%M7qhQxY|uk?AXB7;XS
zsa+4NM_(c#$P|ofko&fDnJaXzd~Gx~YT}Qz@c#H`u07#ImR)%@+f247wNsQz(V>BA
ziALbDG7LES^PlCJA3dA)wsX*<L1*40bu%KleuV1WUnfMTidQg1n}+Mr8~V$``VLe_
zBPO?{s>sZ$vep)5w3~GRuU4Zy;&ISd&S&mnM{v*N7{RRBOhgTgvdDOy$C(<xIp;j?
z|JFr>t%O#qfhh#EnVoS`i)s*UL2uo8FfH;{5JcuP4ShB^+tw}_h4CveI-3`~`(*C7
z<A@e^vT*x%zDv-GK=Mj<UR70YU!*597_41UQ1{NeVO<tct5sdaypst+hakfV>!lEy
z8n-N4LG@`*p;|7p^tvl4x#~v`PnT>=Vs7RHLW<fI?i7s7--hvZ>s%|G?bXr|={W(z
zwb|BoE;b$}3_S;UD=WP`8rM`MQx~{*?vd&U%lA@_E5{{nk=SC|@SD$kj`HfY=!rT~
z4j4C@(dRsiZD0LHnxzUxOrg!kIYr~>SVgdD;(~RT0kjgB7~_`9zst(y%Q^hC4^a<G
zgw+6Su0)V=NM}^iig^8c{KfZi!+TF>){524PA!Tyx*X7~8t!hl+3$pt+3TIB5S2sl
ze0=A7Om58cA#TU|WF_w)?eFa^B5kUtgxodBQusY#I>ztbWT97lCqtO>U?Ri6TH}-G
z<|K>0eIZw#@g5eeT+N&W!dT;Zf*<;{OR%urWaPy!;iw<~h}HqmBbhgs&<{~ooScMo
zDk@s_)UUHr|05bQ4cBmOY0LM}5{QTT9>fWebqMEFv4*D#HaMFTGgc9nMu~iQ!Wr-9
zfVaJeyW0tRWDb*J5#wXqEgoC0@R#$>WzhxaQA(lRYJ<+&TEr?4?ND^-4C7Aai0uV7
zwyJyAbx>38V2?gSE{0R8F!H?TQ+euBsQVG~b(>{By$XBx-Pl&#t-2Jn4?2FJ!`?SA
zhA9N$U40q@R4WzIB$<+OS;<-V6vi`z<uXfFt%rRc!!~>D&RxIzUl<>Q6jGhte>EeW
z>c*&6r6)n{08kpjky$hw4Txh`VNm^z1;h0ykB#+KvLVL;`woxBmciDhbUmsr++<0y
z%4V%N#p$P9wK5MfB@<Lrt(b`$uV>vaZe~FYaoR#^B_`(2=c(^~AFa@%6!_$+M=sT%
zl(_bqY_M4=K|4x7)sA7Z&WfMkz(4-*XO8*cr<e=^O4R_BcCmSF6e$)-4YIC<zi25}
zz2%)OSUJXQ6c8c4(5@)%X}5XF=^tX3Hyle^3o)L?Q($dj93y9@kS<e%&P8_yG6YHA
zNV6)W4{91mHsJFx1<|78#5(>_3sD@SnlYsZR`AnP-pxP1doep(jo)lg3M}G9*pe`5
zQXVK(IO@N?z>`n=08wp(V6==zky>&ntb?d^%GL!j?GDm8^d7_Yf7S|aj4N()ybDzz
z11eT2rBmU}2YamoQXvS+#3jY{CmhR*Klur6sgEPI8V!lifsjnpd2Be!?ccnZzkTK$
zO0m#xw|aG4To7!rSliVI=yLxL+M|B$hQVqxn`K}7BijPt(P8e(qa9<2{gkJF;9V?9
zV*E5_H|_KL3%*9sY(SD^zh~1<)w?GPn4uP5FxVX%Zh^2Ab^)ZqaJvGuHdv*I(*!Y!
z2i8r3SLO+a97=t|TI#?14Su~%YLm`ZD^<u$rnJUxgpRB5vsvXTd(5wmWY11cbX2rO
z`ni~fAkh|B`??r;m@qhFb=>ewY&mRg%2=*5c11b*PWw$^a!Xt=od=dSis0URxcU4q
zvI~mF#5h6CV`VBFe%5~xg#lj4kfvGtQI4Ql2VW~R+9(p60s*28292q=S@V})^ONhY
z<Fy|=i^+0GP%Gi!{?4p|h#I1DmH+@C07*naRA97*`Z((LfAX`_-^=c;h-wt0h`{$~
zh8_zO!;zmjhw9-kA*qDmn3rrvXQ5yo3;s_bXg2`eI|`=;7}Rd-j&zcVsY%0(Jl?qw
zo0UvtDRuj-EzL$m(6ZF8{|Q%~_*SZSEo8gYpjs)OUm`Y&q*h@On$mNh%gZjmg4X^|
zB^@c_dqLKxns#Yx2kS;m=l6$BW4QjWS;0Z)dNZE${?At5kZBk3F2;$1)e1=j-}A5)
zp?cVhc;$b8j)lnt<ywViqeV3=p`(PI#jyJ7E4k(KpQIdXlD0*ejDbgh)CxwxbCpz-
zblXgO?7~!_WO|*3&9tw5qhj3wa`*lf*c@6C!3)Z;?XDc~+W%xtD<(*)H2--At>69$
z-Oj2E#(1s;Q)6w~vEgc9MM@R47xMem{k~KxfzrJ{Q)r{oN};tzw8vzl$;9|N=0D?E
z$gFwX^{eafo0ABdv~W$xbo92<Rm}P-MY+gop4^l|{Gd!Uxz^1V7SqHbNHmO*w}OO0
zL=FaCG&~Yp4p(CUZj|-V>WsNay51_cQ+dfUH_GXF6Ei->74JKpT~kXsQOB!!ERP}%
zI{9s|<F@Eo)yznl0t1ix!&Y)7OJ_U*6M<bnNxJwpuD{|Ujz9hVwB}R@X9wuCYcQ44
zhzME~vVIJG+im>hy{9p6)mp+RCbbPBFQr`#Sr}Vh{^3s%KJR(NbE;sxju=}`O47}V
z`pl%`tnMmd_9}1o{%<<c)J*`{6diO&@@80@vvx$-i3!RRWBlWsbGZ4FXS2&hM5PfE
zh~OK+tA#Yn0e2@cFL?h4dFqEgMl@0(7%69Ec{jZ6N<-Ou9p(VT^{B9T59awT7$g-%
z)l^^TU1)t#2pHvtW<8R!!hhkjc=;E;z~3jw@aN5CB59zt#cL<*(DqpQgKN0=yf0IZ
z65^ztD|ltIO9H8+-GNw#WGkf}<eT48``0Ne^nITVE=wl~@<_svonG}a*3X(n;`?kT
z0e|_QZ=>oBY??T*NMrNaHKlsmO#A=lOcWTUFzC~PE4KS+Y{pqlVj;DP6)RT44m+{q
zflp)AJ-5S>#n@)-7zCw{Qfp)UT{8W>QTgDr2IEnkJ%`4Ibv;6ad9`V{w&;41i<KYs
z3B!L5&%&0+)}}lry?yqo(3>JQn_18<c#a!&Jb~YR;uCD&Xb^9xqbinlR<p+uucY?k
z7t*d~z4A(BdJEbyMBq^STnNlsz&I7sdL4gkITv4WJ}*1&L}d3JQFB6!wHeAU3S|tU
zGNGAd{RZ?McXI9NA7I<@CP9=ECoQZWGHw+2B*M|>oQrz?(`nDDfq0p=gShu%>Gbf^
zX$z8GM96vaVkMzFuOLPIhJMVxf0KKrAk=_+BD#}JiBUvp8`X#j?!A|vop>T^uDynx
zF)-1l;%khu=!#}(Yl8K=?!v2oa2ftVFD4zW5cok)fvB77sbQ`#T#sUFSA4EtdFJ7r
zZBYM`A?_LpHSk?^TX1YDco@=BfIR&f{Kr?n$v+z{^sM<zL@CA(@uHMnt;gNp{yNLQ
z^G(7i0h{K0q*?opXgpm^68c+H<e^;YBYSJ5uL;!A3hDe&r8~?Ko+)8R=kTl(k7sdv
z9A6S-$wJos_79jeA(7Ng#wnDYYNw~#>_R%w$1oIzp=)pI+Z)GxrN}@lF&<W6>GEaB
zNXVYgeh#+XWbH3*q8e*gnWxKXJ|otnriBQLg^FBZ90ldk*^F;ki^|1P3u1A@HH?+@
z7T$oz0>=)I!j{9?Hr8?4RhBGh1vr!=5rHJdk1h9p{hKuY_-AIv8u1j9o{xIm6WHhN
zZ>3!h@U@>MA8C?>+pf&5^HQ#xVGymbqKU>PA)erp|Nb1$KkPM(KILh&YbAU!<fMSw
zQ^aBsF{JC7c#H4n2k$tI?It2BsYN6v4K<AtEK!<cKKBLGGY%x4RYRlE3dFUAD%($$
zDpTatWGei}pfY@S#zCg-j-CQ0_Vt8i&N^eO5Upu88iaAe>K|Xtk4||PBg@t>D^1Zl
zMVBot9n%cp&iW)zdE=Wn^vf3zE!ZA2T0*HT5mSJ6w5dvZ!;fKa>TnIM;&jQwrUuIY
zdb+aFcPDDba9p`IMoBH&YHT@#r$3QbfBi!KJ~4)jR+*Ft@hxdRVsCGhzkT5=th?k#
zc#|zel1@lqv?5N^KASMze8N_m7^>;ywEKKrps5^kFwObfqm5y{F7f+|zK5D<K$1H7
zPmL9uS#wcNqk+zxTW1Ve`aw53Y({?I(KEo1gK$#tGs=)iiq)`W=?c;yWX=<xKxOs<
z?)&{8p*{g(dy^AYNC9M1v(WC}?^G+?oL*&g7WJ|9Xz9Lwark?<w&*(6#mtY|gyHqW
zGq5FbHZ`N~>>vwLKqDj@l2Ry)64J>C{&5F?`Tk{W7gT9C#tCPQvY}k%ppSl-cDaP|
zvMNBP5o5qAv@&S2R-z=$bPkkfNYL0gBB&<Z_{Goin7y9BuFpS+R;`SWL8XA`jI-L!
zETr{G%+kf&c*Ys*+Kwr;S{N0hw5II|%S7{<&wL*C-;bzPaca?7(xSb5UIVP|h&T^)
zjJ*>G{X>4eQvgN7Jd@d0N)_kxdRUOOR#<CEChJt%5&t;*v;6+t^VlO*1kD6r3^r{+
zpjhh*%NNY&72o<E3yysg@kkXN1Q-FVbu@}j&@OKSQ*Lk)HSAFxuAx<YP)Wt+&Sj+E
zd3I5M2N8>s4)E+Ljjfg7$&canU;hesCKH%yKr?CMhd#tHdjugr|F5%&e*0VWM62_+
ztOIeB3lCHeE1Ya}C)8AZ`pvvws*_-4Tc@$AOD2);lL#m;AT5Qw=+skKl|<nA$jbW}
zyZM*+?WAL<wK3g2kX|;3{56Xxyz0Amf$!(c)l(`&xou9ZAQ=2onfn$mB@qL2=CkeI
zk7MoqE8*UIP?629C5NP>ug!Hv603f;D>{nSB3VH@92uq6s6(1$?W;NYg$(_}R$-;K
zaLlZ?;DwifVZyLQakOa$Tcv=r^-$HhlC>iRm&ph-+2op!e}p~#D$NZW2xkSXv=K+1
z`F>>HEGPvX?Si}~Ur?FGng~i~3c|`Nv{iU<LRfF|hYP<By2Ad)9#5+h;B|P(^HRB>
zl_nkAK(J&H*S`Jjj4oe6If+Tq29YPUt3Hbp;m8ku3JyGgc(jTmfW_we52?jki<Q*T
ziWb_9dXN=0#mIM4gns<3zcp!aNYY^<nIbym612o(T5ZB*OZe&WC$Rb_H?Xtvh&GJl
z2bR=_v{vT+cAMGHc`ip@@NJ@f_QsBsGKZoxiBp3{(yjoJ2qFpA^(t@Kp5x&fuK!C(
z#Up94i>N{Y?U%yrm+Li5Ndb+mgs}hPc+EF1;MVamypd5_s*P0%s?p{N-fVvO?$gkB
zEJntfS$iiI<9hdtY(Im6ra=;K{c87{*{+E6=2@M9*8w_tysum`5r%|Me=^POw}Wz-
z?Zo3BmtBIIoJ33F7y^rCvg;o@e@SMF)`h_HTnnz6T8Th8W<ZTH8f60RU3@>O0LEj_
zXFUf=g$I7|OZ-;q^l%CnGUaUKZ$(WNwCkLiE{Tc4`3Mv$C}ETb^~sKp>|Wb&Z3XM~
zE_Qx&B@8ZE*mZdSw+@#l)A^_lom}P!bXK6nw%h0=;(tH;ab``{@#=NFu*A5Oc<jrM
zf&HIGT&duR!b3so`Yn}pD0gOvthFegtXD?bcu|X0zxy?}-*P)Ie%on8l`@`ZP&QNT
zL~GF5%tEwb1OBR&{PgtqF=y=tYSyB3L{KSH58>W4<*<)@8t<TI5slVRV(<Vhg0BqP
zDzwq0WJ7HlT@ccd0;n=osHi)$pr$-BU2;ItrKqWCn{ozTDIsY^s79OkSHI-Sx4oT_
zr7M_|2-0X$uJ}kGjGL5&%IA4!eu(`)ayIQbv+>GhG~fYP$wcf#Kx>!mL8`zSh0<Ow
zfia-{V7P{BxV8$5ErTACDOSe{9WY000|HtZq*6tm_!M4u?pL|96+>l&rV=D5k<?r4
zR~zN~?>wGhd;_gUBNMC?GFz*90TQM_A%i9=iW6L$JJIU3$yxiXMU=&2UA&%YS)`6-
z&sPDhz~@D8JCUW6YZ>uU?Ba#A{`yx`9Fr>B1<XYCqV1};@*=(ZZis7j^#VUj0DAL0
zIs&BvMigEuXcMq-$x3`Ytb*W)`x4Auz&(HZ4XQPX6&zbqFS(ZVZ$XcqMiI!O!;Y7g
zPqIqC1t^s%m8!JI#?Y!epQna13|q_El`ROgZ^;V7>xTD!%ivWrF=|r)Cwi(4if->_
zV$rQO<3G8Y<{$sa0;`Fl2D(-z*>iVxKJIv;Y7O7`D4RNT+o_f2Hq1(faBHV3j73eh
znfTi+{N(cQbKECBMO60Cfx#ePEk-+@a52Oi+63z-_{m#OW%ly5%+VULZA<{I(6S^+
zIP!y^pnAv)NM_eC`OJcKy`wfWf>s3Ic&^>b_R+{GW>z)L=M-ed{hW4_J8lT4d0{~d
z&50IXY+3QWOS$?3A7`h@gi0c)q>b@XtY=x#Y!L6ZC$IR<cNlr`;iR>Ipd2E$>&H~~
zqCI!sg4^?a&d~(}N^fB#b-0FWxHjF6=atriq0U=wI---ViYo?N4&a5)=8%tnlDis{
zm`a6u(k2Ww>G}!wi8MF9>m5`ZmZ;I{*lKYAhV;n(7B!#M)L?XrTM_Bwuj{qv9pVR<
z7iC%&0X*d?$TmA*5aubLKV5n`y4C8!DGqU0N_7A<#msIZPqci`qw^a*^tkyzHYlBW
zl0-DCHcWtaEu#yzW4rzKW5cTDaPLASN~XFJ)wkz8sMJ8m(`EWA4r3cdwOXMywzgY6
z9~Kz4E(l=@#?E>RUGPD`LBm_WMewSPKB1%-VrFa!8l9;hTOn<>VDVCZdBFwjOh99N
zl2XlMP08oEXMUKd;$b}P1c0)ETdh=<EaWFnX_WRz0gEPSPQtpC{ODU>;|=e4H?}s4
zt&}naVhcuP=)cj~&-&x*`P~QK&-~SE8HrPrZR7bK4Q09Cdc5R}GbtT>5OJ-97T1mI
zbQXln5R1;+f)omq6^08I*ELVkTWVH4iC1q&?|?}FH30nTr{cEC=_?5Ji56um{N>C~
z^7n6ilRdo<nP{P~q@Jc#_E;3BZ2yW^^1`otjrO)XVMnSM3}S7T6p6Tbk={Pr&JeZ*
zV>QN-(bHx-GV}_=HC${g)1MX&*dp(!r)yripLJU>IASeEE7FQ)^vFXw@J%POv@uR4
z2(WQP#rTxQQ!4i@<~JYzFhK%|B%oCXRab?coV?RT285qm3*68}-klhwD0Dtx?AkEJ
zF+-uV83kLbvj1yd!@71%$*Z#dj=SK_yCKfmlC%q$Bsbi>)rO+)P^u?&F~&Ikp1yaO
zR~eMsa|71+tdA_KH;j=g4SvAhFE|LJG;4qVJHkXrZQ7}$^oWH@zuM*?kSnhm^q(OO
zo=2%RLThp&OBiQVj~=cqv<`AH^rI$Wc<u0>ZzYMt01%?r2CA<85lCB2OuNZdANvH`
zG*e`}i9ZtZfYu!J?sp?QZi}vzI?9@^I}@>z^@56{kIVZfmBKV5{Cb0{zjQ7yJ?2%g
z#~!px0lv<rkQIXHtRX5MpCsHc&hI|(LDIk9%4lkdqe-l>q_vQh+T(y@Ph#YVmy*n?
zWJA_@`Bf>f#lF*wGOvg_;btY&v<|wLJ*i)ut@E>LKxmOO;Yn|rAx2Qj!g!tP`Ukk_
z<Wm^C`Il@TTWGXVDj`yqcBRaH?SzBhcP39f^}WQiNARmYkZHQP{-f+g?bO?wS9yx%
zWu@~0$DvaguHkxciNn_XR=RCL%C#kG<<a&cb~*Vi%zn`eStBhd8KR^`3B{ZytpC}~
zthwS!Oe-dKze}gJM+Qu1Nl&t%rgJJ=($nP%*`11T)^e(H77ynOY@nI<qUSI<Zw_f#
zVm={v{`e~VDD4nNbn~*}1Y4hxbq-SsRR&{Bht9gI`0OAl)vfBMN^sL=EMKt#QHsdZ
z1drd7%7SfK_V>SF+HJ(zPTOp^%~eggJ5z_1r6<Yk5MdOaFOQDWY)qo9xOq1_{OE6O
z@!J-Pq4gHH(CvX?!m!mKFaz9-^tuA=46RLCn6_oncP=GdbU#5oA<+@_z-Q-|9FBU{
zGl;`58)D9poZQi9G9^p*oK}KTDXN*ECu9EK|9poXcYX|WpM5a(Y8Bt}GTnv_vyY<P
ztmCIKe>v-OO#boj%*Lagw$MRHQqe4ICOqafZ)E#7yq0#g1bCeTt_y5W6bF{Y{S7IU
z^pz^Q+bND=$~*!CK_)t2R)0d~5HCa_S`j4?j5qP_yoW1~KZ(j6_c1q4v8@PbBY|R~
zY*;;aHm|w(Qp$%NLNaRv<9V46lME>9q2VZ2G1N2z*Q!sQXSjxIxMqM?wnR_<#t@$Y
z`T$el3l~VH5SkJ#6Y-3*Kh5~gJ27SqUa3Yxj7lwgdS!0>^qHvJ?m$m8ve9X!9qp#>
zsXmKiS<6H>DJVi0(2=d=_Vw0)yL8?nWR-9cJRQL3Z1z0laMq+TbA7||U;GNzZa`9-
z&3@(WNLIQpEWJKLa=)Uw;E?D0F7TE<-;lTcD5u3>u{t9IS+;Bip7yf+UofBfd+o{U
zrHf$0`i!W<iLPa{caG7jpH5cwXESPw^HS)p-`C)kt3>TKDj#}RYFK30`qpb*Ed75w
zVF3OEs1EP<76zpXAW%**Ck1kgt}V+3#jH(GjV6uX{g&G=yNsR0LnUogrOf2KZP@#5
zC(|wkc*<mkGEKUnwszotYCGf}sTOP`m?-7G8?R#Bsx|EYh7+ilG$l_#YC%#&3sxK2
zt&}hk{`&FHu=eM_W=B83CQWnzX-#u~Y<b*kUe6w<oJdsh@wCZ;=+qr@+asJx4<UDd
zo6wY$K&IW#fnB15$SI`TE2>2PfA-!xOp>#@_y3&tt?Dpg^RBeYD4+~R5<wEW280EY
zMYI9qjQz9k<>5MDY?3b+V~h~NV2qIwIat_mNrr2Jzz79`NRkjB0g@2XO1s*fo!trD
zU0wCQ=l4g&?&%3a+TB@PXP<p~LQjXPt`olJobMNhqi;M>zeh8QiN5?5KK17>;_$gX
zVW$fclSmIYF6>OYv>)*>p8C%B(m(nrbj^6#f!2jVaWP_*0hCFqiH6nzp|;u%a^IfM
z$LAf=bet&read?P4w{W|`c2QTUUnHjQK1_eXxnu3wA<(5Yc})g7ruyaYA-(OmPA7Z
zBCiAP(Mvw#F9UiJt1tKs6){-=(m*Z0z!M`N7F6*@Ir3L8qSI{9l9+WaVf)qB(&)!{
zd^XLLQrHSIq@P2v-{C7=WJRzxSjaLZBjSh{um*30n}2j0{Uih5f`B8=Ity_L@l7{E
z-=#ZmXeRtNtOQ!E9c6TtSve*Oh4xR-9w&)oNRn(LrVb9P(8jY|`WL%kP}S$w=Xw<s
z4MRH%Qf0zZD$QF#@MzZ|duI60e|Q6ji;&Du)0}Lv(+W>~^(*O*jU!>BfKIG6h0cJ<
zM}ag&?42=qLV})S`ueN6{+h3F@rzzeH-u)hfjCkIBP>`$tT9bIcfa%9?6~p^92r<_
z+)de+nueW;u=xoWaNM6gpD0K*8uHe!*g`~_0jc8!OM8_dJAg(l3ha%>8!Ht=F#?qT
zARDhT2Gkg$ewP-8dp`K@eC}^A;poJpvvXjSL}F++xV_isA;11C9&_ni>8xFc8EX%W
zObsri>Cb`c)xn^)+75bC8Dz7`7}3;7+655F#K#<>Lf9OK&FlHicU{I0JG)7Yr5ReR
z7u=pXw(Ond`b%F+I2V)P@TMG{5z<bBC1sowQ4*_otE-o;e@J{RUpOupr?Plq!VtD@
zVcjpCLEl4DT73JnUx28a#xTjc<D)!nPKS+V-So1<V682>?!!&zyuodaMZMrW7}MmA
zyY9m@ns^U{%H-(}L&hh$_lB?2OdL+lfEHX@%M1?94u3`lo>|M-q&YjFDyG#=I~Gwy
zE$O+6FZlzIOPl@t2A>3+RiEnBR5>s>ky0H;$`OdZ$Mzi8{@rUhI*M_7_aJT04oNuc
z`Okx6k3h#-S!*+E#pW}X+0t~{(N2SoRRhyAaPN=#>=p0fH(vBgqOm67cnjmw0>v0b
zjezgdn(eaVo$ur3_q>OrgOI4R7aus{@qk^XL2%k5dE_fzMK1_2L6{1jsV~@&@^_>o
zQva&^uJU}yz;d%J*acW@I3Wk%M61D|$~NIa49VOalZoTTSG<}V-}zo1VjK9GSsJ0k
zwk^qcgIhcEJn@yU<fIq=CB2rVIi9f~WgUyM_W5}V1r~V+wbfSJL6|-aYRKfiXNXEu
zG8PFDLp(VKr=7qP{^BM4V9yju;8E)c5ys{{+phc)Ghe(4nT>F`5>d}bYbckY6;1c3
zh8Cn(UCgV9kX9;(qC<jOlE8BMv!2G3@z`dQ-p{t7xBWC@C(?!7*hTR%NP#O-B09zx
z1~)vUc#Hz<_o+|!&J&oBpZ#nb*5(4I7B+8X>j@`w=Xbw}nUC-;+cC-LVblG)atwF@
zx+RN!pT#Ge5|z+sVy!`=ZdOdxSJx`Fv$CA}7q(z{PJNCKtgjqS6Nk0Qw3377PK-{>
z?H~In^WVRjRyV@5TJ#!C+GjkHjTb(fs5Oqw1BEE+eaX<HWpXPf^-^vwBBaZIy!|aa
z?YX~CvUx4Kb^=jTB=~AT#bG)T{V!g{*WPhCM}#3R>CtGlNX9Hv7<xyYz~kQdI(lJ<
zX*3D5Sx_s2y1~|o6p=zk9-YCJxWS_}YHwj!>n>kh%g0)CKu3Mv#Yxeh-kv?QdL6#;
z7cXPim%hwl65{83gkgw;LdSZ3k|@9Zu6Hp0D;E=uj}Zoqq7$U5<$iPBAgHaj+Q@+k
z&A1&>A<A6hcz_8f8<`0uz=h*%e$KCR$OTX4K21oP0u4*p3^{)7T0ZxtR}k#HkHqyW
zl*|TS>BJgM#okiA1D4RQIp8%_aXT!yS>*5D6NF6~r=3c?ZUcnE#-PvbSALd8lHfA&
zQDY41tL0ecjPn8{k7Jc=dypC<=O0RggE4}44ujG#sdmxM9Xm0hMUy1mf0%5s<@ATq
znVEvA9pEG_(nN~RlCMa?EjlyH5<f>FY+!9b((7bJPhByr;%k1t=G10CzOh4rx?nf}
zg~A9C=Ykw!K*+I@3Fo3SelCJ-ck-S0yqn`p3++Y7nif-Q*YT*ocro$#7)={sP>hs`
z@qmJG=AhMs_9H?~xcahx;jm*)XZ(yulB^xaYQ|G-6g37X2Az%J>o;=cC2!=|poxuR
ztW}aWOew@i9mkX3_D@9Zkf7Zp5O7Hv%G8x}`gsYEFE+EFX<O|r3}AG{0g(kvAPeJA
z{*4Y5N{HY@iM_+mb!hFK<_mxPr}V#dBU`Y<of%9c#)pm>3%l2@=eIuc5#ooP0c+Yc
zO<MdYit$<JL}lrvh1zPX?Z2I`2ZQ1>J$qB8v!aD?V=lmnf)z{LfOG!l<?LR+mUz5P
z(nyhO>Ge3iy@sz|_BPrxJzU%`Ar+~>^lA!P>Lonmh-~#h9;k&hc2b!k^#)G5@B(IB
z%qFqieeE?cH=F8<W?gd%JqBf%XKs|xnN|vj{K)u>Si@^xjHRWF7h&&QkLj7Yl=MRd
z4J{i^Jq2qN_k){|s8<vvs=h>!a;nz*@u?}3gH*^ZY&3A$<Ja+JRoip2-1--*V91Jv
zvHJY(e^6z3V5K0&(W;}+K~^Lcyo)h7u6yGfc!&tmo*5c#&wUP_{Nk7557~%mTAa&R
zzcb;p!eMIGhP4(a0V)uP<E|?|!;YVA<E&@@9?_b0gle%a)dom?w&J^e?9YD2r(gLB
zjs$`v!Fx|4${Znk)~x4(H@}JgM4Mo|jm;dl2@K9FVg)gR^S&$)D)gZ%QoJ5m7Slz6
znNc83T9-&65adLWeuVw$Px;S3`zyxp-p*v-<N7fY2)?Q82@NwxAIDQZ^lx-G9s+CD
z5M;;I8Y#qmkYQx2Mu2Lot+s5~MFsH|(Y#aZ@tV?FCz|TLIKXA&D`9(#c*9zre#s@=
zGP4WQY+!ss+Xn=*g1d7&AA9wsw6u|qGZg~R`NEsQWN|}QR2Up!e>I}jE<+PMKJaXQ
z@&)Wh7z6CoEXl36A#qZ$6IB>g2Q)ASo+4+t_I~8i$iuIfG)^W8vG+I(9k1MV-!vu&
zGSExlsACylzlD3gdjqXLSr^@;Iuk_~{r(&aWx@8f5yD1;ey>x~hSwoxwb~f_5s2TP
z9foJu=k`E_kcFA94xkDdZX^|#=M~#ie)8V;BR{%@@oo$a&mIp)U2rk-s7K-_$I=qR
zn!%a+?C~_GWFuXW3+5fO-}*NH^_fretpD>el66gNU@)5PETpqUB=cPw)4Tbvm;WWl
zL_MTCPfU+ERJx6jdu@}a{qtoc8`mM@<M~>4HbRvZD?XhCwbl%{W*w3ES2aKfNt1L%
zA+D-vp_ONmOeIadC!C!pxbX%)`KN!*rtP~K>p8sZ<6V#D<QVtF2@~g>$3^daH{J0z
z!J07);IaoQ5G1b!+mzJkPi?i;R=9zoVM!^8p-Wonav(-3`U=u&!Kn}DtUq`@cXnrp
zO@R84vGzK)?TEPS@_*&p&wd5<nIb(Y9ch*h7d}r%4!$lLl@82d8u9chC8Hk-feGM{
zEliwnA}$anf+jz@>RPNzN4><P+Cl>~b=flvPKu^Kcit}!?TDtFx7ny^s`w1fD|g*<
zZw3;jpcCUw4m;r#?)&jA$jnT-D=7K^_t+}6L;7hU2pc5*eo3QShnZDt`!lyT`{@lm
z7C5IqwFeMP8L=5ZY9s&vAOJ~3K~$xys%fjb;x*I7Hjkg_z>j{&jsN~Z)+d5<bI@qe
zJLD)H`usnpKQ@*kS&|~)gBf_GeUZf}H#ZN{JNS?Hy_4s@;7?%lCfryPt0^L;83%OS
zjS1&su6y~b2=2HW8zscf;{zo!0YCM^GcUaqcjS?{HIvxPMW+F=A~GiwbEXTBlm$+G
zG&;1L3)T_gDDNyt##-VMhnel5*M6N({Kd;Sx*HSDcChO30(Sj4Kk3bJ;_v?fXa3!5
z=ro!%vyKTF6A79Ie34YQK{7oHT_IXm`}*2y`|siCMN>99DxOX%zRhGD+J=0m${=0{
z)WBGiW1shj?Af@9snEa?TbWt2mQQ@|cIG!6#fx5bDP1+FGdUzg&Y3M*)wWsA^Bfqs
zS}OVZsGL3BC@2^Cd@JO*^B>P_<Y)&ie)7%lLa&$7ujV7T3wXIlQENy^exVDq^b{vE
zseX|dyi=@6f41%J?P-fzGuW*eviYP_n7MBX?%RP+QhkVOuX!cO{V>IPPMe!9XG73v
z((m<hevN?%<Jwl8os;F*zc2;Ev+8qt;2h+D^Qtg96wDAq$+7SDF-go9U-LISG;GtG
znMKD!?zJJ0`}@Bk8Eav~FcVks6=^fAv_!QK4AX8O+l%?)WpCl^b00_J<WuPcV>HC3
zzc+%jO6)wLbKLsR|4j6)8`vC-(Nl+OKqm;e2g3y~eJMQRVYrELLQJ|)oXuAWIus1@
zrRTxEu`I$ZEhBD6>MSstN>xi~Pn_{YNgwMRohz^8lP`NE$0i2pMfB;>Gd*-7;Fh^P
zJn6-M&2fMDNAz0_8m%A;`q{;qZ&Fw3UqveV)Us?zQ2Y7XYWwd~Ipl;BrJpOug+wUm
zGmBnB?mij;ZY<<C{{A(5v%iPyJM(<_$G7sMR)eYWb?oSRKKP;kNNHMqs<Y$KREV;Y
zbeB$8S$c?5_VbY$4FU~no8`Z?cw;#9oJZ0P4UM2dxVHnh-h$C`%vA@#A{jB1B_pH8
z(25^mQrd9^8`w04c;T*lwwEFrn(kJNopL&2EXfbQk4Ze5kN#Fzv1Fw&p*nG(Wjnz(
z8pQp6%DPmePOJZ_-%mNV*$;1|0n`P<fdidJ6cFiFv&lOFN~V07&RQDh`N0R@Pq1S<
zcCJHXe2l#bocgS1!SN^J+l_LBM+6ygmYTJ#ZHhc4LeTZx{P8O=;TXqW{A@anF`CvE
zNSERXFf{rR+duMe{N&T0U~3rC>&z2aj~dIaL^<=fev@@izKCeNna%d6D37l!l6e`V
zd>S4tD)+m!NVb>TXHZ5oXk?-0n`q8Xp+qjBC5C(d_22mX8~%Y~n{8aDgAyZwB^e9Z
z78suPhPTqb@IvCT7Dg;Kn>;dx3SHIED3xkOCA5ZqYOAfb#hsXBU0jn7F;=KVz2+m7
zKK;7gLLPZ0M_v3>K0mdKZ;7yba)SA=Nw{tUm%Qm6?3#)3n(8}Xq<o-}+2|+&>V;!G
z`wKu@lsHJLuTH@`0tw;JtuzlmoF0x%EyMP&e*@EZI6p$XXi<NxC2Lak>9QANau6%)
zHh8Z%=Mful&%HadWZtLXSO6S;B+WIO*m=v1go!5sJo)2P-or}3L>*P$C^l@8^rMVI
zx3+`8o|Wa=2Y0~${5r6;KA-!~Mk3`hNTI4)ij?K^o8mp%nS(pN&y64cS2hvQizAY;
z2J+B{vh{a<mu|aFU~DD=<ja|TtvVgWieMa=xq0Sq_$uH3?l*bD3tmX4-Jl&BDEyGa
zh=J)j_{KN+#@pY{(Lsx-*F$YWB*IM4WYgK_anv8ZfL^OfXo6DjA-k;xsq#x1&{>cH
zEU1FWt)S>ya7?RWr&-Y@nPVqP80$OkdiQ&{{_U4@v^4PDKF)X&1G094yN%`PZ+siU
z*=LiiNq0evNsFJHp(=fX(XUSGq0-uFt8G!Qr6YbOS@%-2F(L*OAO)}pi7HQj#f$kq
z0lS+UNyazfTN~)g1RZJf&+qy$CI~XuqyxGOI^eRF<9(J@bF`}3IY`PB<W<qcbL`_D
z&%9A4yyLE~U5g}XdY<<glm+dAVlF7xGV<OJJxAWqkTM{vBedDLm!#CKJMP<+l8Oiz
zvFUiy*aYhz@(_0Y<VV=O^BLXk{S{anWJi*rWk(YRxFjZ3jjQb-u&r6HeQ*~Hs`_j7
zxjX<4a0K>I%|BJsWZ#GssI`l_*iOXNuX_{6J4=7}96`&lJ8W{!OI}XWZV;GU6>`w^
z<gIB{MXV(OTxTApe#&P*bp_9P;a?F=wh6~Va4uyvQUMzf&F+Q!cJS#}{Vj(%OETBP
zT8GCl*9hnxa~S8o_V1Z*KqCk+Sw|$_xkxLXirgh5cw#FZg~bqgWpP0^UGRG?H5ZB%
z4o?#GF};Y}|K**0=bi847#red<_S$q5L)^Z<Lp>7$+O>k1@XxbMJLjI1xs2a7@HkK
zkwOeZM>_kut592QwcVe1FxaK6f+d9tjrRnN7G`1tPyfx|WXJ3*(rBU*;JhVhPx9U?
zKF;>p7@wp&a3We+CYBY#SO@l4Eav0W7^!%lk<vC{!#R&)+Q&?WN^;*0nBI#zUxb~E
z+b^X^m4y(L07)hB9@csDV(`YGBKQilWlX@%ol|K+AO@es8`lulpLh!0sVUgK8+9fA
zHc07le>U)BFmDJ#e4M16jXKmE9M}1N$hFNregnq<&#X`7{<Q^@?Unp8M>vcTym#eT
zk4vzz<F+e40zdjG<9%qgEHerxUHo)7;dmHpVO6qauiRNuj3wnAhXf^wA}k4?{lMj%
z{iJ8$4?hZRH?ZC!B6x|xK!3hRvlH`$m%ftq(;b>laIQ;?CmC<By%lot>)uG*G>nb4
z!B<cb1{c_bTzg@3D9~lbcVxsd9C>W>U-(?XS@3wGUXQ>f+;;ip-28zLbBu&&)I-7q
z*A%+VfW4bGa`D^ViQBq`VC@7Ze?LB3-1fuW10B&mu042dwbi!rcu?L}$C9cjcxB&9
z1$B<+|Jn11rAJIcY!$VjVUw-~y!-thLxZ3~c#$%=iE4(LFGV7{x{k|&pJY_n83qK0
zA3<}&CcKZC3`6E_{yw2kwGUKNz(|K4V;HJ(iOayBp`~b1yeM7-F9zqbdl<nf?49k9
zWXw$717dN3u;H{*a8V3Dy$xK#AkW|E*Rp&8mvx@RAb|y+WRG3jL1NF$a_)n(V5p0Q
z15+rB{K4=hucq{AM$-NOQ3AK$!Ob842%AhmoXp}|4dj?(+4`H$CT=!BP3E|K(N4`p
zYzmD41F8wFAm#_3_*W*z)^Yf`Paqy^)4*gcPQjRfB<a!6J~zGTjmYh{vqlWwCzv3l
zV-mKh@|;WF2wT=;$6D$0nn8UUKcp<WG!M8en{F=}(kjbgn#}~~#fleCFOF#n+<57m
zx%K~klA~;h?{u-^2^s-=#1Nl+5*NMWy+j+<B9mhoJ*sA1nY;=b;Ja8Z!qrw=ZMEGW
z7;gdak&mh8jCtt}m9k;|1i$}#&tW#2#WWy^(_OQ6dyMyf;A718(nQxusw0s5;nCBQ
zx&kO>-#R&E@vVyDCpjaK$pVWJd>C-p8E4@<G<?84H-8Tk$9VN+29@oa$<W?h?#cUc
zT-gxN0We53Ur^8Jva^v?70Y}#Vy@GvXeOn{FFg7<!qz06TYpU8%RN3uU`oPL=VM{b
zpn%vg9iv)`4z{*cSu`B91%s;A1;c@?9)_><e8*X{C0dg${q{U&KH<98zmcOHM0;m2
zYnt3+gmYi{QsT)r5(Y&;Fk(t|$?QQ>g1wZIHS8ozfBhT$;G5s!yyyKddJRipY&vpN
zN$?vz$Mz5YJ9mBVi)>cG#~qZ2qzyY%c;X9Rih0Dtag*byPv=UtK;$dFPKTTB3se8m
z%Wk@8P+9C3>3(0{rV+s5dp*YbJ#KpK>)7%6FL0O>B#KevNdlqMY++70m0!N(O?20!
z8Vk-TSW}KOc^yGrSG(!jYO8G@7Z7=4A&Uzl22n#2cX`2|{2s<fSnshWUC5PSg85ka
z;x*Tkcz}v7OvWIj59Iai<8nz>a4#)FjF4Evp^thL^Im93$Xz$wh-9kQFgV(^K#w7-
zp5lH|9+k?T!=MfjWx<d=f7Fke-aCV$SW}5$YsMJgxRLq0?jY#(vc+{BWIbB)h%N!G
zWjh=x8s<mbyN*k%_G;e8Ik(w|FZ@j4-1-#mZ|rEuS&V>GG@Jc5nSYmV!cRZ>F*-lK
zg`n3%#@kFKp3|QF9C*ly_-3<YV8{^hT<5@hpDKMy+F5YjKJ3}U=RfvQF8<yBg_~H9
zX*E%o2BK6Z)WP04^lR61%{$)7Ry&S$3W*X~$F?})lxP1sYoB%j{Z<oEQ}9DeHojBx
zfHS0SIgBG!X=8QdaT=wBtwVoPAal8HgrovSaVH||#C-QvmvYZ%zQCbQah(pv#L%$J
zjE$iWKaKzK=6|F+(IRLyv8q@r=@^ocqTmr>g(zFyYN)NY+8!)qY0zD2rZW(T<;YE&
zIqwl?ph<*nLe%dALaRN&2S4=h7!wp@qEcaXTXLOYg=zb~#L<>aicEShuO4x!nA*h2
zClCb<0vi(T+66Ov@frGFE|h1~-Xp|8d;OTvBSUE<E|#B>6p+eT44w1r+%<)W!34Rk
zkl<Sl)^0tFnY-?!(dpq`s_RqKe*-k)_x}iq$y)i|4dxW}Lc?mebF<w0peq=jQJ=g0
z=LD<kyv0)0aDF*->%HPTUD$ay-+KT1I6P?6@AvWJP4t-KIqW&VLELK5up#Ol2E{1p
zlAETB&XQ5Hsw&<I0fy^7`hFgM#-nJRd<w1+5(<d;Oi5EoX8W*nhR^)nC7jTlfT)8b
zB5uG=6(%3{XiomqKP754uqI3$r1OZ^5u!U;YA?DVAmy7Cvh-nCLdl>Ee7X9y2&jQX
z(w&7+m76bpBRjwN6%MxzlDLO9m2MNJ8;0=Vr*rP#UrN-nG{#z(eBRY%;z^n=ix0kl
z<>&@$r(IiZwLNI%LP~L+i9?n<4F(rS{MGY+pZ?r#oR6?AL!u5f0oQ-?hwPY&P*-|O
z)hct{O8DpYFBr07&#9X3F&K-%!rBR%haXDpl~z9{zWp{#Q7~j02$fl^e4#my$^Cx6
z(g7%`3YC&!FaH}AQ_|fEY~QgfW3|fIlMvJ>lSdy#ch7Fvxw{;tq8uxO6&`<Gcn(A|
z5?<#Qb{}e6)wVBl?|nIv^z{1F?Z*!ZL##+rxwi8$q|7myoKlYjYn1Q)^IJJ2>C>O>
z&>A!B#qjt)`4jx)1ST*zmr|dWOU0>kOk9eDr~z**rmNh2%@^6bYl@REd>T<05(F0S
z!D10nT(3tM_4(3kU&H!d7aPYQp4i0n0+?Ad!TE2vgt*nhm_|DFgJP@!NsAOO+4JgB
z0ySEZX&W`5v`i3@K`EY`l_~w56333Vv0VSkzhm2%zsw=l5_daT<G~uHMOgQkvpMJ0
zuOn)P*k&_r?@B%+EXDgtuR<9*X?0;xTWz)NpQ0hBhgK!k7%&zu0f8uwIrBKStZkw=
zu+U5T_#nWC4Zd>Sw{dDKBQdGBR9(z}=?V*k1DBYpW=(7{sQQ%DA`~{A{xG@*)&z!K
zH{Fbh%Wk{R*pYlLmsdffJkaY!DSdb)CTWI7X?CBSO{CCA%8IU?Q`4y`w<?;&ZnetT
zvBxtvHxGOFWXtCj={Q*~e#p|1u$YY+P0w4~s<wTWdz*c8!Rh0GQ|t4#|B3~M#%*%7
zV6CV(ic|)9M6mM-d%yT)c75aPtaTP?z$|do<IabNpNelc5e&vi$p}&DTv(erRHK5Q
z?;*Rk@zu|Kl8b)(kLb5s1kDij`Tkfsa%6ikx4rv4#5di{IxqBk^EeApqsiS4e(P;-
zBc2#Tnr*D4fk!g<C4VpZkWubLN4Y|&te7nfK_f(ThQ1api!krgH=YIok47<L{h05)
z`Zdg4b1hrc;*td4FvN|3`F5M}GakvA|L4{8#+rofW;)tya+D}tYRjm(OZFzI*UM_F
zt+oSGG~^1<Qc(08tOSHOp8b>yaIOPZk<ifh4%=MAr~mUyNTXPe%O1mLep!|&7DoL{
zR|i;<`(k6V0z%nv`Y+M3mNqbT%Pj;s`kn{IA=F+9;0d~&Zc3CPqheGpfSQ&PMNHEa
zDBFJDE?nij6f1OF4@WV$ySAa;Q(aG%pR4GS!GdAr09Q@veGu5GSq^^C6bw(R&)I?V
z{JEf6-U8Kv^f{$OHTZNCX?}+Py8P{IvMqdn9@}owUw0_S{r>OZCMGdrD>%JLcMvjY
zL^29J42bI!#&dl6Bk$q7Cq4-d--?bm@g9s)pft>Ph_3q@Klu2kI3%$|^BrtMh#SIo
z*X4p2zX*;!37MS4LdiIxIWM&-9Brw#Gx{J5?<*{YZB=&`s#B>{bQV%1NJ*lEpws7v
zZ+t6vfA$M(@sM=7grUPU0;U^n8fTu#V_*MfdaVXwGek67SX0T097;}FOiE*@y1?Dn
zR$Fbg?OUK}@YWcUG2LbU$XKixe*NMnV_k$%k2MA#I~uJqzWR-un4U?}u}UNx%gKg8
zhgHf(4u8thiR%G4Drpx!1;c#S#Wq%G9DfYmhNW#Soo(A7N$@U3{ZsnByacEShN|L$
zy|Z(ez*52QQ17!5uv7%jktO}v=#dBnVaT?7wqvtu!DXXkMikkw5lo2Ry*(9OUBIfR
z58j8QOz!Fn=U}+D_fZbMuSSyUg5dyE3OU-5JNqiVcFFnxB#LpKyRP^ko95<Wwudwg
zyAsD)&-)$Nya~caS|+Gw!ZbO9Mm{=2x)i1k<~scB+G|i3vi02mK{VFF5)hbFy;8)`
zpX<S{U3}q9Z)U3+Y|<yNG0rM`l9*%8Kc9^kJ%#?*I6*cGTXk4MrA@nROeu92gcVP6
zR*qtpx@XJP#tI3Ze!ou;JMMnp`?>w&pW;Z<z{h<|Q%M?@Jw}*#<XN2k>Pv_NXpOhB
zg}`*dxny(_%C&`T%u6IipjHFAR$Fbg{dd5Y%D5qwE`dscm>UGdal(m5aNH4xVpSkZ
zbUgu)Dqs1=cSyWsw5%Q@29GZnc}0g@+EV1#1DkjU;Ghi5@umyhB8VE;xDLB+Eg~M%
z?ZTd!G*Qnx46-;xwRi7qx=38jyG4o)bI}o~j-?71W7)oAR}q)gdyEZI5NmP`t+f+$
zx8H+_-4Kh1F0(_jlEYAe(gqeHYe4HDu!EyuP}L^`52?@Bfxt#|@Pf}S`m_;{G8y5O
zq}PX^-N{Y=@hR5FX&X5f!`jnNXY2`2z_pv@SbfIYAVRTE;P9x}lqIPb!LA*A^O~#q
z<!3yXWMUoGgqQ@xrTPNS_h|S&-~PupvwnJpMlV8$a454$2e)-AXT9Wa=(WaZ1VT2B
zNp){)xi#S7B4E@ez>243ManLQ3LYf^nkaX^?}Oa<;eX?B8>F2DqXa?7lvCE7doE}H
z?Q4mw5VV^(@3S~u1`{pR=1QijqRCtP;o54eZ9hkiA`Q&v<TLg*BSPXFVl1IGT==;2
zNcvr@wRly$fDPOHzt4RUwI(eV0PoVGp={wU-9^w9K{@;HaO53%Aw32@^W!uOYmPoP
z?Zm{6<enV^V@KsgsTSj1IZ!t@JBJlpi~}zWP%>~>^06cpVfXYNJVLr!V#*?HqDi=J
z1G78t#l|k99M{Uw(EU()9x8Sgbh-!BLu)|mAh3sIx%jFt7@ks}s{;iWRq$!0NJrC4
z+^5}#BqmVL=l|g!*${b>`7Xhl7JFMw9`XF=<JXKMVVF|16?j6%g5=YL*O=^)yD>o&
zan;8^#CcDA9QKGq@j-y`9%B?Wg5n6~W43+hqwK!=YpnO+db1?N^g>~$4SCAnycR#v
zrZLt`L47j{u+b%x|51!bD``f^2uw)xeTlTf!YH)nJ52xkNBI5+-p>(gp`V~uiH$Jd
zXfgInXL82NUrDbagyYR@anO{MsUtNOv=S%eewIbG)mB?=4{88OlV6ju9jU`w!Fxfy
z^0X&E9*O4>uUMO^Jex*?tG{{!PV>DvUyfZC^EXR%40LsQyDWfstt9p-T@v0JHa_%3
zBK3sUFu#2V>iwWkuLy67lyu*-^YiHvwPd5hRevv1nZKndR`P!H-kG^#RMv>e3K+o$
z0b_@5WqxW38zlqNLLw{KA6eqpIqQ?Gwvn;g4rDIAFGiB;g5d!9`uw^tcNmJT+Co%~
zZcO*OuhPBgCfa@QB+Mk96P|Pt9D6j{XqK#R`Lc5UxR?ShSEoqS=k{x^q%}FgmdBh+
zG}a(6CL=S={(iO(H{8N^KX3(`up}-*8X;ZlxU1LaS6=vXIQe+|SbKoxX*o%PrMMeL
z>QSuFPo_HyrL*=ZGCfP@b6@6~x4fMrvAFpTfpz%MGTUzBPCAZrUUvz-RzT2dAZ4wQ
zEo>J8odbnKHKx|6O>MQ+b^z0h3aXqXtdbTA)>y!E!m)?3Wy53&<^+a*5@UjZshK%$
zxnl}6(+MaN?Uf{$vZ_4az?f*}pM4%(M3_9{2ogzm^QQ0GhEZQQ@nScv@}APahq?I<
zwz9*p01K-IgY0p9)@q-d?-0ABWMuLlF@g_Z?9fB$?b(CrMgxQ-_vgHiIz`3!@$ieS
zX;lv<TipeNs$K{jS)ZqU3VsayAB4zt;GQ`rw#kt0q}xMgXSnWd|HMP2fpq3Et$^Oz
zH5~Q3ze(I~VZ>%okuOBba_{NYr$qpQ?{wgv+qm(o*YKF9{1#DTg3vZH&QQS{CFv!Y
zsadXm^IJHi(;<xecuDAn%JzQ3u}^v`8=iVG@#HwhOR0760I`fworO^&O1i*BX4sDb
zCrYdyGuMTiZs7Bmyq;rB3m5hA*3mJZPP@VU;fHeGYcHig77&iL5tludS40NqNONlP
zYCIKLNf}dHZMD_5f8xt9`t3Xj0aoDbM?M1Y`dRGpB$^Pk+FbROuY=fu6@W!KDVEe=
zIKZ@>nvaX7ufpeq9T-G}i9-*e>l{tnqPt@Xo9igZP)9u1L{P;!=yZA*V=LW-QIz1-
z-Mv)k)T<KpV`k>&v)`BWscMzhmMwH=_8^`4G(JCw$3Y8<!a61$q;_PMldsx>;aBT(
zwBLcD0VF4<AkBSW%Ke-xcN%W}#78kZ?uG6=wjHu7>T%BVeit@wKpTy;73;O&jFv*V
zOc0Z1yFTt?nd7QYT)}z2{3QJPjYzYB_Z~w!%L>rwC){+|Kf#^1)Ak8TzmIB6Uj#Yi
zFdq4m7ZH!OQjLdvG*c>wlPpcKw^-A4MXpWCguPriOOYd!OD@EHxczoM`KniQlnwDw
z7c1#}ZqEqOA&2nfH@t;-vW2u-DeHwHL*=BHhb@DNETJ1wTWz(~wu;J!ifXWmVo)A`
z&RImgS%NTKjE?(#TCFj@`t|Q*qckG~Ll>afvbu|gp-57&42r<!wfJ_MMypL{_f*Po
zM>eioWx6XkP(5*?^rC)&*o#QfaUNB%<&#Wlqah89)6;u1CMpWz5icN{4<+^vX66QW
zMRd9Mw)EYKVxd^TuH)80nI1mKIr)P<l5}x>iuSWxAKss;`r3?@K@fDV5BJ>7cdxjD
zjbezkN2dwR(@v%JE9cW68$(P`2*;$enLb6`HC<*i0^%HDlyK{pzDUqs&!$J6gKLGD
z41qxtj2P@pkIv`*lc}q(VxzJ2lNg$UHk(X2c*g5rho5Le7^X#t&lDoGx~(9~%KR+o
zxUeYG>PWv?sB2bRuqwEj4vk$q_`iSt7aS2Orq@Tslc-16jI(Qef+t?`W}?mOF=OLd
zb(T}*rHg;QN^4sbKs8cRTWz(iejKXW0<{Q%fwLZQ5>4YF@u>>1wWw+H-J5SE$x)w-
zVQKLZnZbfUety_zAH8?U{#O>ab9TU11K*w?P8{)`z0i+}G2qb|mXIWls27!Bkt|uJ
z6)`*SQg{zjQ+pP~dPE2|ZBEC)W@b|TiyZ!v<({GXomNh`4ki+&dQS86Qpd1^$R5<3
z{6Q-iRP~9#(e?S+_qfMmU*}Tt1$T7jczoQ&;`sVo|B)@RLwlaEY1mD`c`x`&e0vhX
zW-JS2vy&;hk+ZX@D5^2^dvNd1xas=qIrkaAfg7L1m;mE5HRK>5nV*Auw{gwCyq(R7
zA&Fwb1~?4&ImaV^?|0$kQ;6Gbg6u<RDs-)dEAzSxJ~HUOmgI!c!BADMrlshR$<coU
z(cE5|-7cT~-!J6IPK?a=XxOyt5>JeAcQfFbZ+{2zVH+`HEsQEw4B`cAOghSyw~Z;q
z@^nQry|vX=TWtpzh|2v|rbeHS<{W;=X11(f4^9b#fTY{Qh+*%{96NUH#^)nFrYaq_
zwDy9I^zH}7M;1ifTB<Uag2xK2kg+X?rc8eG9dND~Wi7i3>GF3TclwE=(~Yo=AOprE
zjg`f6vsV5tA8`_#S0vjB+`fH#7SXj>#1{30Yd2t2@q4DJ>{zcjr`HjOTTs*`nP|6R
zumY)L*Fk1SXSw;RD;O@S&&|Gb-oqVC8FA%Tx%>wxR_7ApZ+?e+uez44!5UPQ8E-h^
zvFF1nC*qqSRt0M`QzDj9ak9v+6(1|c#a#QDf8(*|KM~n_C>k_V0XG~9BrYL{9M`@6
zO>CHnuuf6e!->!fE#X5?;>h3o1NyBo8bN?})rGO@+MO;cZ(qfxR;shGaEC$5N645f
zoGS5rM6>U?_GPbR?VURq?|Y2)NhHCKH@Pj2c-m{<0>>PIot(h>OsvRjL0hX?;gOGK
z$pY_OBQ*83Jd7yTR@-Vssj3UWiWT9sla9kD5hi1GP=rL4?|kovh&5U4s515F3d`gH
z17ZLGAOJ~3K~#Y1m7xdd`qYYGD5+|vRx>4GY~xm<D1lxKoo+?1Ij2)C{k->{PB$Wo
zT{==E6;ZeBYpuNX^lL3*p`u_cJ9bWy>!#*<4yqvIV`*`*cQ4}0&dun-qY>Q!UE~~i
zp8;rg5OJ%$)j@a(KX|$MgE^9PVSQ>I{N!Sh%Y5M-if&@Qw3RyUIO-!#`P{$!3rA0`
zBks--PEIgAKF(>+|3i{im?7fnqI2ocB-brV85zW2`>_4%*MWp=ee~HxjWFFkP{F80
z#1nM7?EcgznY-~uCXGQg##qaokC<6M!4qG53H??`7}^wVF;a9tOYUl|IN>Ewk!7e>
zNB*obmZ_Q+NLCCgkoX=UaLe1?LHCC5G3f!Hj#xPhclNtH?KQ81laIqqv`M)Hv!H4+
zwN3-hDqajmOrf)|>gWS&J21|h(Ot2+gsg4##UC9|E*1$eBAkBm@%W@q7+6FU=Myj?
zx7>PLK^>t3A8J|DcP}6P4#2w>j~WNYm-;zbGOh;357|tD$EbsDRKRmu@|jmMEMczG
z8Mx>N_Vma)_&Q7tD7b8iS&U_R&ul?8p3}9e2!4DGw%ww?cV<9eO;&g;tNVM!qBd#V
z6JwyzeW<<EL2M6NZvLPY466DV;H3JzJow4QFzzyNrR!j^kUJuuxp+JTNzB}}Uxqt=
zO4y5WLBh7~43B@o|AbBJkuXH8DcKt`B*Ev-$C&iC7s1~9xbf;MdF)f4LmajVj7{lP
z1+buV5&Yz5T>tL(aTpDf&OFXX^jn4<&U3+^y#%&yCKzvLdJAO_RWeov5m`2@46=Ew
z43nz(!%P&XT;}qC_=ILZ=BMv^H$VH2PqW1dz8~RSAI2JNi(}6H<NwXrV;@7jrkzp<
z8Yvh3iUX!aM<y?RV3-l3#+C1H(fo_{8tYF$Z3jomkdD{bVwT-=>PaV{Ne`dIG@2o(
z61Lj>_>OJa_@xekIm*KE{nh%(*Am`QQ}}W**^sYS+MBoF#34Sx&CgZ#1+-9YR?UkK
zVRo*At1u2}p7d9km1LAgTCtB{n3|d%WYx<Bn8Sde(In~2L#7F#qtUe`98smSpMz;h
zj96QC7-~D1?W8P6Uo{0o-CfwnE^NWhLPaYfXXEj?I%RhGCs7aS&2!ad|H9Fxfp)rh
zb&MZ-0+Z+e3URBIO;GAkagwIDHrr>AM6up;<L5rX!_GV#4m}Lt3~^4dV!(J@(#Le>
zxaO^IW$j!K+lx{4%xlEG-8oLa_+sp1A5GjG8>FGDC^?IibDg?q3sgpy6N~X`EZlj>
z`WKn{HGtDTG9S_X+*RE0fq&&t4KV!}5PUmihsGT9l&7%yx1L3>*~F@p)Toug(OiGW
zsHC*C!T{=1+s|J#ulCotoH*d9<@_46tL<PYhEqVq7n$vmhi}D-V6sBM<FK~DPkwwi
zUMyq^$yFaNl^wFxTqT#~LcV=QWy~mDgl?=`j}yTf(5PRqBMkyWK2MHP;k6u(8*$wa
zE;8%F`-oxp^qztgB>xMeN@y*CZP1^Y!RF&r_sg$VD-|iul_q^m5EKQ2R0DNw2VWe0
zkaF|~Yb5DO^(onR4!e3=-3wmW_D!8!%5vlgobuC;ev--Ed$65|)_BOY3g<rW|AV#b
zF<~QB<}9(K$^)kX!l0O5%=FDSv3>hawx0Jm`i%f94FZ$xE~sFmKD$5j8M@#7E}LwK
zx`<d~`ay_p+R7;}{8QqIajXrBM7k8ltjx&{Vk;xun~o&qTge_ndcRbBPDOE|DN;1k
zgPU&Q%GbYvV}llMZWbgVX~2|GnkS#e=`Z<9qA(=1Dc89ADmWv9^sJH#<w>>=GW;s*
zL6rNGtSpvvuT;r<v3&u#pZ}MB9kS%#jXLWqeGj$Ob|B3ZOmBiZ*tC8<jfP1_cNARg
z5M#J+Y6geFt53lzS!CT=SK{|{RrsI5iYF~38k1{MK~L}S-Od1LRLUr50ZvWt*_-iq
zRERuAv;oTX=X~}keFS3zX68ECUR<R#$;O9Gqd^??F{5ayR_Is?!Pk7kAWpD01XEr5
zmO5q~y!N2v=nuGHP}LKGN7v_L-#DxVNk%2xkRgYlTCy)BQ4Di?x$&bPVWR|i?b5-q
z{t;&&XFrO#5vGd9*(Q~w#HfW&^I}thm?(z1IllJA&vW*Zo{o;MM{E;usb+y#gYWm?
zXLs?<4}E|`O-LMdz(n}Ka8JL-Q(ykquy!L7HVfxdEr5zvt)|6JC9;fkY~?Xo-W$q@
z?eH|2Hy}zf7ZdE7<5RDC4Tq`VJF}QT@j*;%4SUyZ<d@(0CVD2sgpD+KW(&@F;h~i}
zV7PF&fBoUwmO5I)^Z!c)_{B&99!P>N%b=3?xA?EEwu8aEQut+Sg>`FNjJKP3mtX^%
z66ttl&z`-^b$Th6kN1@zHOeXLGFSb9xk`i*P0!PKKR_X(G$zOKnu2K>$0cM?>8e!s
z6sC925QL2sJQ~>9OCL|k6ZHzMc&4%qU~IsAr<W=`59I_B1W2Pn;@t8xxhu=(Kou89
zgw0kZMbyOtYHbIHJvz(P4{X5zTu`5jeVZ`JBH6~kXcC%H*`*nO=!M(f{{c32`y}%n
z%$Q|&XgTvw{wGO0Oes@EOXSFFCM;$$R7sEoBu=>P>d$f1F-NiHl+)<j0E=QJ%Z^<_
zFdy;txBN5XQ?uA4p&!o^iD$|uocNTd!lNHe(rl;8#hEZ#k@D-HGg5R(2RC>ablI#<
z7h8fbpuL!8H{z;S{56}V_Yib?2oBmI-A2egjgViv{2fHCfMC3h0lZU;kwVx;w5p1_
zTy|Kej1Qz}SZbVSxC1b1sd&ME8F4F%+?S!>Rrds!`@3cSSCf<4YFmw0p%{bD=4!$a
z)~%hy<8g6<cj<C&qQcBvH`DIXQuAwpQ(o+|z8U}(DKF?sM>wbb78Pi<@#-;N@o}8R
zEy|8LRRGDbd-rt8kW>N0s{P0XI}7>WRz{(EQ6KMpRUjh;`HW!05Vu^xunhPy{7T|p
zM5Ea(k`Ji?uU`avz`6Q^(P5|yhJEb5M!8h26bd=3P~PrMx?R|@jqiQpW2{B6%_e)J
zh+}^F@$irn@j(+&Dbf7g(~01`$J#77^dpkHZ{d#Hzt3qGUPQlPvDT#QKpC3^^F8LT
zx|Zp0eTQ{6z)Os2w&>u9Hf`Y%fAV6YRtsxQrR`Y>BvN%$i?>gf8|GPfTt>vDOlv_!
zN#dBG7xC>kyaD%<pD-TB7*$Lopl=QLpgi*pug7iJgqa-2W?d0uq*(Npp=xe<cUq4k
z{hZ-@B=$J41Sm@tmW!m~i&YgXsa_V)oGkv@S3dGoe=cfU#f5v;RSg6-Z&-&*VuHY6
z3<M_Kce{7T?i5LvDhGs)08>WMWv>!IMP&g=g;q=auz`p{oX00gQS7G_sAad@XFCPE
zr=~%`d6zN+%J7aJ3*ZA0QW{mSh_%ekcTwjGeFrLxOkxe{-Tm?(i?J@|KaYA5VXIy8
z6-qX^T)U`)_Zof>a&_|nU%(pb4#PfAloljDGVElfOqRVP@bJ~QU&f&X_+A8Kmd={>
z9Q)kgfmVya+RT${?guhvhjdXl#^Jm})bX_|{~u@m@{@66YcPRLI|PBjn*`VG!>-+Y
z^WB$ogcE`&LU2S4VY^c<`16;*`n813kiyUT>YG|N{a^4CS)_m%5ab;3P$?>$edC0n
z6Epp>PjcUXUB%X*1wO$BO5X_g_Iq6PS1*RsPQ_1-5hAJc&yHEiLZ?HT4z=B%;&-Ki
zZ6K5mBl$XXJL(f83lsQdUz6P5SGn+ymZDExyfzuNEm42EYCDh@u9Sjo*-qA`4Qo;F
zuwpX?C68j*y?YKZW>5!cnGVAs>FR2klP|B(!5MxaFo-c2uZZ)7vUPdY$^iB0&Ye?)
zK_fke0~D%*-YxG|SC5?Yp2T_jNg9(!xf)|^!OkQP^w?#NG>Rg^R<p!Hw2pDV0Cqu^
zvmcm(0r(Z5QJ;r>>8=*}<xtUJ5DcEA(}z3m<eqD;W<mwkh&{cChd%j2IQj^jwRq<X
z6r)5I(x4zFeXOR&Ox^Hxz_9+bGjU-P^-AE$&<sn^bNuN2?<d&41FH$%Cq!|V>8Q`K
zk3XN*d5^+14U!@9Dud67%1X8nhLE16^SZ(pONnG@)<7{?;M4gCz3EoI_TKk%glQ7Z
z&tt74v9L4ga{4oWjn<QXmEPDm0pOi4M2J+tn{o+oIpMh4mO-^vSTNt82U(Du>+l#*
z^{WA$^x{@kx&MjF;v~T3Ac8FV{uXu}N~V@t$D_6bRpx8vJGGKDZdktwk3o&Wsp7mx
z#js=7o^%{(scTi1x>j`6c&cK&QS))%%qJL0!86KzgR0JGUJ&@~cunozP1p#Ff@={N
zaA=;X^h#Jm;-J@$D%~2vBX|L|nRsXgQCX1Q%Sz8SppH21(`Yt_I=os3yMx;rS<Zf7
z3I<jEmHHg)3m3i2HC{>xl;f7F5IX`%_|Cg8<0uo7%*|ukA#?3<j{5EAkhB_zwTR91
z8uCmkL;bBWX_ukzVb3(*z3NMxchNI(jSypO!JZ_>;%B;W<F~l;v;WD4ph+)@=!c4o
zk2BYt<ct^n8E&kF*Z{SqoZN6WH+m;n76g!L(j<(avaIfQ<hm3CqC)~6T&IJb-o>Y1
z^S5m7#%M3XB|Ty*%ryd<r<}q=p8o=(c7sMEOsB6hy@+BjEfcs~z2ZX$*XMlY1+xx^
z(yZ9kO5vL;hnI<eN;&Ma@BL7RVo)oi+H%j!hx~8L_2z*A1w5K9MryeSQH(*Ci}5<}
z_0auBlsf~@CiTElZ3ifImMSoneml8l4PF%Ie73|Zh_TGh&!<Eh0~ffi@aiwUx~vut
zB{Fy*M2f&`tWCT0!*e=P7Fvq1dulh<nzTazS(iozz<>hmDn%TMV~z^_eoU_y4QRXN
z^@UG4!bcQhdOv}dtRQd+{eDc)Zo>$|*xC-_>-hoZ><_fV@WlEQ>}#B3;b=p3PQOIy
z6(2<;xBh@_*L{`oo+AuGre?Z4;@6%58#kkkutZn~2EZ$brV1u<-2Sy|IrNwl;qZsx
z!VtX2Vi6nQ;x0kI!_{wlE1UXC(urvX2Imd;bR*9Bqd$bLhvOSfa#U3+@jw^r6Ug$%
zV)(dJkI4Y@n^bfuj6uDU^ydlVnCsv4Cf4lSMbPi3O4N-e^9^P1#<e{D(zg<|+XUGd
zmCYW{Ko%IXqGeH?2w0hvTwN%vj1NTz)ME#fWYKI9lS)^BqMDW>#;5x_>1?$rQU1#J
z#h;3lpV+MEG?n{k_Cb?M1!>6lfl_@BNvrWxyw6u>rgZeSvyND#E?%V}<5R4MlWN=l
z*Q%xz(W)LX;Ii()<k%SM9D>IuF@Xge7-nZ?7wrSfG9|-eD?$g#LydqZvaw&XaU{n;
zzp}cUm)$>cj+xoHjP+>v{VhCFgDy-8myvi+FN!L^PyLzkAl|1P5Lu!qlcmo^iMR;F
z{nW39%`(4`+VlK^*aOPhf9`@oRnG*DtB?P_b6Eozc11{dfe?c#nj{$UT>H-Va)>mD
zdp&4EfAb+6`t+v~x5p5%pbneyNlV_^)DkImNd$8{xb>zRIPLrk=^IOE1W}KqnGm+?
z+4k{I6aV<9v<dJ@hbWGStk5{|L=JuCuaPvz2y6ge2NRtf`PWfgF$>-4a$NYjC`gq)
zhvHGMINv9T6Ylun73{h0dN##ilO9fD;xWrT2p7HTQry}J!ihHc)B&s4LMtLKvIgGA
z60g6u#f!g{ikFq^{4(&bI!`?ytW@m;Rq2Z=5WOI&xSp@@jYyvCYsmpXmiT9%0ntkN
zs|Hk1pC$t8)8{G+S5qL81x;cxWR!IoDq>DLlq~2#Pow<Q%TSWBa1wkqA+)vqVpR89
zTwkK;=dayr6a}!c1{<W}G@VYjEOr-FG+anTTonvm9X@mvqMkVg?=Tn==P^NJkS%R!
zY^T%hG2iK8gP=m6$}%w{``V<GuCa5(Ny6UQ`K%jRN<?dt;F1IhZ02$n@vOT1aV<v4
zx}8*mq1nO@%TU*Lu-kE2?*4OG@Bp6p{QB7Mr=&`Tepu-+nAC&e`wqGFN9_Lgx7h3&
zs7;veIZpbGr@)piXiAQq&7xA=LN+R445FSe5WaooXF2tZN5h(RNDz?3N@xNQN7CsM
z?ApP%Kl)*gY&Pk4V*+bQ0>>VKi(dLN++@m%l+XTqEe3-MM<*+oSJgs?dIbs4@SKuX
zR4c_1#ESp+_xR2gS8$vS&}1G3TmaiTJuZC7%i#E9VWNq;ob5u(k*~bA)=KNRE({*f
z=X+^=ZW&o#41`smr5GM;HgY!5t@1_PSS;=-nr+_ez_Q{Xk~p+CBOty+{G{ksl;w+7
zx3W$4(GeM_>!k?3Qt+MotSga!-hA;FgX{D~nWVDl@mMSwtF5*Nbfv1t8jJdJbE}bc
zt1;>4Jx&rd+hvgIH8}DTmt{eO((@mnB1tRy2$C-`XZ|DKX|o|3q(Yu}%`_Iu5hVqt
z_skF{jv$YHB12$P<^775W4oG-(Pp571R=BYol@vWQr#s5oO)~{#LK86M;7%G)zecv
zze3WRCu}w$472^eYzMLAOsehRw)25+@7IDsRi98F_x%L$Py*a#l1~XwCJvLpb(dYv
zk+y}OpTmqncYG~JJ@@JKTj>shSC3{bM)jr2n9Y-xo@4foTiJ2%J)HTJUn2@25QAij
z&Uhv0JHB`MyI40hi;WZ1D2c#yl<@H1crKj$P_)%b5mz&`vmg~WmD3r?@@eHt<C-AF
zqM6_kW57G;IJkQ!pZ)vSaJ&iVP4A}Fu=E1MeM!PGPq>gZ7hFIxF$QWewt}{h3$$hD
zL#imV48Et@maOF#9wr$o4>NWZF(s3TrYKv{MNlnq_Ou0n8u0iOwa-SB#O0j<1@B2e
z!Y;jlyobc4|D6Npvk{`o%zy|fU!+a{+x#|2Z>3yRbee`zHTh>=GUk|4d`U=|Rf@3%
zQ%U?K+qKAq01+yEl=3=XoQr8iN58h(evT{GSA2Cjcb|{>sRx&MG>}};r@Vef`S<&w
zJbWOFq^fI|3OgPM;uR5tGX~pg4GSWb%Fkk$-o1x7c8ulXL#l<=(SoC`Rfgu&cL;(F
zLNrO(xpOyWAf_tbIegM5XoOXPqa}`{41HGtCp7JW_U314wOf@gDK(wyFA@Zh=I*cG
zZv}&@PSyp(zD8fm!a%YOC-1fRzS_QZy%_()?{V+9zQIEq6VPt6qdUXHp7(56vjxE-
zKApuTLqqZr2~<<RkT|5@<9pYAiAO*FeEh_CilDf(D}fbUr-%9e&D?qAmpQTz;v9YJ
zh{sy=){Jxb@Bc2YIf>0t)~bujgG~<4SLL3UNNN_tnnx`?E9DZmWY6bg#mx8l%0Ik`
zwR>kE>eFmCNUUcshUBQ@Ipf7Iq(9ce7>li-!$tczA6G$8@~%l$n5eeu85};F!`<+~
z-&ej8pM5N}6iIT5F(ym&K`l7%)9WE|Gz>`W(s2N=Ku^Cpb&$jmbzt`%*fULcYB#;9
zDQ54zkN)0Sx_kEGyB&Hn^Jo;|;{<VvI*)frkqjYeXTgLi6O4obrqLj1HwnhaX{=ep
z_?FF#9kQ7<Tes3U;xJgh9wx@2(Ew*a43SsVrh5nh7ED&`IIzZo%ZrB$$-^MN(v^_P
zLbqs<%upmVB}0o2osk2G(bo10aW$7-)@vqMnZ=I8D=L!e>nMbQC9!!5s%2a`SdYLE
z0f%=UBQ^yvdVRz>)F{#lix}bax}vhSbl=o&9Auz^jt+H|-M)NWS0rcWgCGdd820R$
zsdTrqqAf|#enb%3EH@Z^E*Heh;yJ4$jEZ=OX66`gx3k4PDPn0I>3(7C3933NBD4FY
zU;rKug!OTM@LjZw$e2~q<;lSjhpaj59M*fTy8K-n*_gn`F=jlZKe>fNF1{Eah6G{?
z5gC(>SmZO(Mg)(;_G5Z?{0QxJ*?8I+L=s>Pm~5;gN@ALEpKC6ECz}#5i9?M;ngKgH
z^E~zCuYe8f5Su!#d>o{FQ%li8>XOH0v9Gm)2CMg_V>20jsuS#d!u0>o-g}2za#d&E
zzqNPO2{(1lYPDhuWeEk45Fo$?35<{kHdrv=u`wPeoW|kv%-A!7XZ+doIft>$V1rE(
z!GtlGAi%~T2_=+6OD(AtI)@w2IaRgy`u^Cv>eRWR11Pp`Jx{CEeQ%$0Pu1RQul25X
zz3=BgPv<)~vdIYDEJs=mx{2jL(&C5z$NMNohDch0i|V~%O{AXmiB$c7Dxsn%hg=Ws
z_Vqj~a%B*3i+Ei`&jN|A)FQ96%C)az)ql1g#Nv%oh(n#jc(6rAQp7I7_%z+eALGE*
zZHzzoAP2TR!t@h6$R;LmGaWQ90)4_75*5UUBt5lK6_%+n80i;?s){X9wfMsrW8%Q4
z5B0VcI(QB^Pbcfeu0ulR6rVI1Ubl*|jmNO;<V~zQ>kL}QZ-iybAxTi9<X+H3aB8t)
zNig8e!2Q64b*9Fk-ebgs28dzSutN>ktw|M^<M}eM^YozohQ(gQ;<aEro<)bxqQkLR
zLz0-7FniJ@830oj0)IFNJoJp0x;og|CsK_lRnK}|Y*brIhleOsLkANOW2-pg6?X30
z8#|yH*<G>ln)mkB=)zv(*$lnBz<6bRatfrso4r@G)5CW&(xKr6C`kvq?=^~1&BA_q
zVw_cL*W(O*xz=)?Z*iXCXt>gt!+-8+DHv4s(#3oJ)P(4PL9kjrDkXQ(dqhl_T@a<1
z$w;<r;lOuqXHD9`Y0jSMoKvs98dk2riz2Q-q6AzYj+AQnQoJC>bNx52<)yFwA-rvn
zh=tr?#G=O0=qVFly_TulZ({>ZFgacvGsO(6&O8tE(pOS6hp-Zv9Wt0e%Ipty7;)!d
zxW?dYb6@w`61p2s4B9K;!7W_(cYnhMW5{NvXr_j)81`j7F8}%W!lq*|!z~O^o{l)g
zDt*(Q!HT8MzSqT+YrX=vG9oGws(5c=o8OlM${tL)z<r;d>xv?LQ{KZ`u<8P}Z0AlU
z?z)3VZu~AgZo89W*FKUgBUOVbz^bP$mX-rG2~HAvDBgKy)T73LF+n0Yv6wW)CMm|6
zfcz&(9C+5MTct$-&QoMP#CxoBq+XERW7M#`wUl-vh1^FE6EHjXk?q*U-s`U8ks@Oz
z?_rk=Gqipqn=U+$jTc=6r=A2W#?Ydm=M7>Th_Th)Hr3EJASRkVX^sC}f(f-EEh;u?
z<TlKgcFoaDIIOaL@p`@jSVWqqwWb1P6quQr!P*4x`zW*^jEs&%zndd}zdkH;uvPl>
zeE0!BpV(ai@CdFmgNVah3qvC)HjpV*W1Sy61PG5mzMaG-5fm{|>_7)T-t5kRNKjX@
zX^JljoQCeku6+maBAD1_@M0j(De@k+*+fhqbeJudP@5uL0D>z!|DMiFhxW)&*(41L
zSt^Ue&qvcS9dh`mr(ggU3x=mKNY_Mu22xXl;HAcWE*yzCZvMOXb8Nd!F@69UwM^P!
zHeLM|ibjI5NlY~SECfYJjNqInDTGIFzM1CGQtUCC$ZdEBYYeEz^^|1yIJdt4Z`mkK
zx)b9xM^juQWpB@M<$Hbx9T`O$3F^xr6A=uRzQghn9smbJq^eq~v_vS;)Kd9rrXZc3
z;-CNLe`j^>@VyR6;>nF+GBIqt<Wfd2e=XTio7CDMBPUcycwa(X`jR=dha>a4lIl^p
z9uNO{X^nXwX>T>YZ8<dT)dy5RP7{k0&<w?(Ho@8^WIcG`Va9K~j;-ImfyeK?k78ny
zmh-e4BMi4k5Z^|cZMs>Hys31=VTKx{Bcqs+5k^+8V%^%cEL*>Z*79XEmn|h389|0e
zpxK19(VtT=eSP)nS(Y;oc@B9OW@cb|n(V*<y8HLj*|(qGp8ZU2-^Ik9z4XS%Nis*;
zD<F5^ENezrhuNP*nA*07?GJyR2mayH>@OU$W+^AX^5vX**`?T1PJwnC2%QIKk}zmm
z8vqdmALkv)j>w=afa39k2<be7ii?H8;&r$HRz2F)m(hU(6L=SjeeWFJIl$9M6I4S-
zY<3WrLk0EO@AnL>)Y)lvIACgDG#=NT!V37rAY((74~2RklIybI5W{1S?g#|~@oU&h
z<fyu*GJD3WX7d<}F=Gq|#-|uCWD$ew%mi86;gMOA>HW;18b4WJ*u0F*MisJ*BFkW8
zm_={rqx$-ua`+=I7*utWs-C-ezn|hHt;$jR@}2>>H&p^wbTg8N?&FCYZs0_1Lz1v#
ze3}d1^DY=$hPTO}XMprMsNp5zFzD1c_uhO1FT4B=xK;~{iGy8&S|OS4@$f%>9DQsX
zN<yREB=bFXPGp?+mbbuJXHcY#@NbL^6gso29}YJL&6kSLhdEWni4R<G0ScFs76n`1
z|JS6CK7!2+jl`1Yp6Q0Avwj0F|K(q!+isC2Nrechml7iHo4<+9Q@!Y{pX%sL!SqPK
zcdpb)_1qQWj6j%UF(`TAAx&tc36L<+%lY=TH}S1&zCd=v^_-q}IC0p}G=|m9Vce1t
z@Hx{?88^^axq{J68(4GV$*etXGou^UldN75=LjsAq=t$uWPm1flNG5+Vb#<I%3`Pi
zG(3!X4f3kqMe%_grn|6bH<OR;VB+yd*>l$x_CEL!etbWTMuVZAuzqNa4NggV%7JVC
zFW>n4kJD)<tT^E$Hox*^ta#bW;rR8?XwVUhZKNcU00!e_P=YnmAv6wRO0aCe;i6Qm
zJZS~t;`O{cg+dH8g7==p1bqMm_8&M9c=Kg3D}oVa$><P1(r#ov*hik~;P~|b03ZNK
zL_t&opS*HQj^(z1dS3xb#pE<TFR(+y&}@aXCy8@y<?-r0Y7I|p-+@gNoO26*_)-vL
zHm@^^HmFy;cT7%BV~9)vBN#k><^U80$xu5ef!7tw1r~G*DglRTmNDcBzBh?UQfLj~
zRj_pm>0<G4G+pPa>Lw9+<S8u}fR`=a>!%LIRbgO=u|x<(`YHw{q8SOwjUW05YfOV|
zVv5$7B_CSB`ZxR#MYA1-D!mI?wM1TN86W_>V2hmXciqUkb!%vDI+o1X7#KVbOVP<`
z?%cs$*IdJ<q(PQB#425z(pkEc%|H1Q6vHEBQ#J%Sk^1ODNV>nzxv7C-p`X`R|Ir%q
zZ>9^l<qq!t#5J6l4AJQxz_n8Hp_Io}`Qcys4OqF1bhrtQ8b?(tWsE*5bEt#v`I~Z;
zEp8v%?bR@NR0{It$>qayu_nQ}40Q$4v^4Axdvo~umu}`GANvf~UVjh9DJP|d*R4I4
zV)riG(gxeRlk~<~46Rwm#*<Iv+zZdAb<%OLav8K+RcF8<6y;!lwe}c5=nqx(^mWdq
zj~vdo1PsB|PXVzpiMKG^LROD4{DKo1RmTQbfGc3%KHS5Pu;=!>c=)=Tnc23TbYhy7
z=_pO#rje#h-u);KJ@8>}`qK|ktXjrNuXzQhyyYr5Zar!eI$}wUB{6Ai7zU2BF%lA%
zB2ouRXLFr5Wm{n73UkF47t4v~>&~9NETnc8d0_v6@gPP(nDKGW<6X|`Rm)0Xajw})
z%~bQ*T0P9p*(<DHB&wM~F*Ol7DD42!iKzn$RA+VJVEc~Uq-n~Ga~Nar72iqb7%|Qf
zkc$V9a|NluOt&|y^CGyZNz_9!x`b?jGxP8c*bw}($tjxcVMx;cr`N#(WAQp_u9pFS
z|CAOCs=8P(JRK3MLb#<b8IJ@mrB_^5knG;c&g-t{colr2?49Xy{1unO>eUb>9Idf^
z<%TK&3JMn{SD4((oj2XU55D!^QJC=h>I7?2ym2(M8SeeN_cJ;<0i+~0A=85WQ!~8c
zr+)!fu0~Q@<9kXqx~9wuF!Cf|oYI${D-G2GT`Lqty_p`(=^6g%PyT>4NkZP8rrAiy
zG-Ge>xZqvygmX^DwJqwxJc#oSYs0Xbmuj}{2+SXj-U9Z}MVwTHmBwP$`$(IK2dOnd
z*{ECOw9*zPNx6N?qx|IuzRVY|{Vr2o4=u;o#!aNIi|=-LBz2tj>Wet`!gE=7>d7>Z
zTMZ*4KoN$cZ5YT4^T85bxn<A~kU@U%;MUT4wq1Wlpg)HZ{g&ssF_rYLpN&ukS^`EO
zX@I0Kyb*iM2G(5iV%EIp9Wc>Dw`^hibvLo)n>XNh@1t#&(00m+Ln*zP8Fqi<bKL#+
zpI~g$I!=G><*a?fRj_^qIR!-{#o9C!>Kas@>KTamNYXQhTdV5=zb_UL&-)`uJ=PeU
z&k-z6Vb9*ZG#d?a?@7~yB(dbcG3!>xq<Mk$au|-mx?mCeG=1LMtdK@3Z`mwW=g9W%
zM%6L8b}d*Nkooc)vIbQbhKG9vd-qLZ$HsyfiM4|rfZFHM{_z>!C#tmd@1>!e<v4tp
z(~FX;ZfXKA2D@}AQPKoT@Q%t7X5koK`b$Lrh|a_TMwcxIlVHULnpFf)7H1fawtcNn
zeZjzDhv7-D4+bPD>qb+hu6*U+5|w0BNea&c*L<9nYVe&Iw4qGd7B6|%+bEhgXxYa>
ztUlV>M?gfuI>#OV`ZZ2G{ajeGGN`?%kl2s~$z%q%-^I2se}#>9h^#Y(X*Z$Sz%CnO
z`BhhuH<~2I^hYF02g>?~nf+l8ax~PB6PYKI6|GB7hQTtg8WwWlX%>zxfBSyYM;@W+
z9f~72mTB8y_yuQj{5!5DA8KMGB14`~EI1bi%8ZE#pP427${{h@qYPU-?2@5Gp*($`
z3S)xGsgnY*SetSH<uCsBV|@BEU*>^F_K~($vvk=?T0Oz19THP;>5E^;o33~z7r*d0
z+KEsoICZG8xJai9N+iD4E~gqM3|44;k*>-ljsE&Cy*dUY3I|f<Xg($V!lF{mD*tUw
z?zG&S{f?T6{t@BtoiAU-BF2KT=!!OG*?Fu#?`+ophaZFOJDL3UjXdzp>v-t)duVGx
zA`MO(U5@VTu>JphjBo$Z`&oL<sa*8$-$DE3FNWb3nHY>|VaxZ7iAyjOaoCa%uCjpa
z;^CH$dej!li`Stdg7(u|2_VqUoqN$jNg6^gD}n^EFIcr=Nldbn>OiY^o9k!9LaTAU
zg8pd*O|-5oCVgN)ArYasXBVnMbIlsO1PY@mXNbHa(GzCh{z+zLa>ka-bCwLyHFa*O
zE}>9Ez>2{D-EJ1=_JYTdDCqcptXLRZLSM)B@a|nXQ&vq$ieX~kepanrk6JsY3$nQD
zkGkuKbNM4G7*ut$s%~Dq&rdF*J11cq7_yQWgi3ma;(9&o<X-Okr!R0k1{zKFc4j#C
zvRA>H4R}c~D$zGBDmKK(5E2^=eCz!?+4bndyz1@mrf4-$HCPXr0`I|eyWIQ_A7aca
zh0AHQ5+>>K*tBxhZ~Yd&X|aikXtD(EXNfvfnrKPC1$`JDhWWbK{e+DH4SLn!k6=we
zespGr?AF`3>vNxBlM{R|M^eZdP4=}KT=AQ~LD61<wP}_78KmzuAqMY#Oulq+-s>=u
ziSpd@>Jd@208I2#NGu6W?%lqJKmOCd=byiM3*$Y@(9jav%a=knNmCx@2QNL3H^1qJ
zxaiCiLZ<IPk`meTSO|UzB2fthG#13Z)+Sf!4^|sf_t7Y46UwD1l_FDpq1DQt1rEyo
zH})CYwf7+^Kt<|c*&Q?z_o4@JplbR<h}J?P=%!7K9=C~e-}V;Rw~v|c-pl=8`xmx;
z>sCg_9mA%<*_&1{^XQ}8`0M|fsj)Vvz2>zX_l`HgrcHFcCuz4a1Vt7qMVo%nE=LrX
z28XE+ty;5~Qara$A)Wuv@!;`M=F>UIh2T1z3`KP_gZhG@M#}Ofqo8@s6@6~AU@#${
z4>;3j{C?Ek3jzS(La|ij^mgyS7X``MwP2FI8eC$7h=>j6?Bh@DL^07frcV)^-F%Tb
z=YKgv;p5qA@Xj&Q$tYB5M1@09A=|s1wB3d!ODY%oy6c=SoXk}Kf<yy8(Sn(&2^vEu
zp;7e;k%PU;;`Q9VHmmAp5xM^<DHwoD7Vq&>Q!LC$P-;C&iQGw55Tn>!+4|YfGuF)!
zO(?7~ZkxRHhu;cmgEYE2M@xc%z^{0G)Stiirt3KQobzF91xOm%sSd?ZbUWCauV?>v
zZsfSeFx~DnIi9WvtIoavx%fida69-kn7$g$fI5pDD)`T}@2Z1PvmJux==-a$+(wIT
zm$W<0wSV$QZ15?zH$!5TuC?s;IhXv{yWyA(NV6G}+<COZ)djtKbg}!o;BhXsIh#KE
zW(3q3JUy&!0m5~+-_P&;**|dI&3DimUP5bR1q>Ca%Q)|hQ+WHEFXsndatcGy=g;{J
zqXaLHINUEjjQOh#-5*~%BA1xGF_oKgX|NZ5WZy>zrLP83p?6d;-u$IO-54#d=SumD
z!TA|eTP#u;Erzdq^^~QLh~V&1qEY&D2uhG<j9fUsayjiwFXEg_&gY!z4tmqA-0{UP
zbMLoqU_+zLa^GNeW_aQwpXZiOe44c{eGwP`?0aDIv2<+%l9ZNdhha(W&o23Lzf_<8
zP|hf`eH;!}7%V0g&+Yxaz@KBm8)QDi)1R7hbUHnTmoK4H6p<3>S+jZttyIv$RA3CV
z0H3;J`QVg5h3CUFZaLU@KCaG#5^$L6;r8qY5n9J?z{T0EG8PquBVtcHvh7hMu@r@m
zpGT`Mz?|NlRw!5~z{(ItxVPRb)13@gn`jmX`M&)mtzmq#6%&80z5ZOyR;5-XRJ_eN
zyVvXD3r{jMgsZ%hq%U25#IVES^&Ge^0iK$IK~)zEhNmMn8`$SQ&LS8@m7<%IbUNJo
zg)eYyn&PH9%!sn$tg~SAi8z}C*Jf*~#7+m(9E=0Ee<$~Ed4Mb4{WIjoU}GRMV)4G9
z(d%&i-+h2pq-4Du6^Bm@2Z|0?y!)N#&}eYJ>`#i;K7$_ifb)F(>N&~|HPxd2H%W!m
zs1~Fc9{KRcNVh#kBNrORldCdA&a%_bV)+$sAa68COgWGz{bsw7V5}2-b{CVMd7e^r
zTxfZWL=6$Ii1m2yFlmaR&6mD?6aU|zeuP`LY$q9A&Db%glJDEY@|NeS*Zm+r`qnqF
zYAB%X29Nh1k)UeiRf4LQuXP7Z<mp3|@EQq!SE1)~VY*1Wm4Uxouxzh?eU`dvFiVoT
zVlOiPtJT;Bv2ztvW=bBNk2EVIrcz+B!5J>lxio%nvxznn<kc_ZtRJ|9v-a)cz_s7x
z`VW5+^Y}O`maSrQqD<d-H=lgdJ6U=01-$e>`~>#AbLr(XnBgWVHW<76n#`@tKX<B?
z)NQ{y*ZMaPwD}gV=hLYa2Ao3?zGL@33h#qqTM;^cUK|@XY$VG)t;7rnj2#-up9?lU
zbH!-C$6-V&Ir0FtDaC;Un9dX?H8j?(#$z!WzSdU>44dr12e;8|wF9tK#Tl*kTKRsV
z7?CQrYRLvL24e*8Tu?$&N7n5jwh<Tr4to1`vvlPOh(MUkF8F2PyB7+gK2_08PS9#K
zq1i%=p-<=3MWg4Vb!Ug1{sT`*!2rBu@fJTB>DN4)xUNPawM0$&XnL4C$e6tTn>2Rr
z!g?W3Jlm&dxbi37j&G!)P%r^$&{||Nk!z_QEW&rb{xwcH^8$RMg)OO%aYV{YdG>zk
zUzoZ3epc8P6Ws}HJK=!l9RG@!!`bKHlU5vRoo{sOdz+712*VhZneD6dltwdj9Ncj)
zcYW*{juT6F<^aB((rcvbNfNI7*<VD5hOn&|Xlv51_2KLwou`W>!+b^TtV~)|5e!ae
zuo#SO@y*+A=Y7Bb0dBf;8%tNLX2rVo6iygv!aINN?Y!;Em$9q?cuyi8NiEKYfj(Et
zgQ^dm1g|mS5A?8*@allf@a&>e=2V}vs1@JqX8j)ju~da_IGC#+PXf(Bh^W7*nJ8-H
zC&$hMlHhTHQ@m;wORu_;%U<_d@*8gDJ0JNJkKKGbtJHG#iuH`&b`RJ7_<L!bvWZK7
z;TPciGbxe;O&X*&MI8888f(nxWL5xDnSHxv@9e{hW{cNR83hzI@bI=B6gl|9Qxq9V
zvxUoYPCj8HX=;dWZnM_VpkkNK`O33vJ(V;gB*;Nq!py+LwkMDxM_LV%70c*(g(QfJ
z<V7%V1W+T~f8RscB%$ZCIh#ioh_V&FQO!+gq+$jc>Q1+Z9}2UE;$Ujue%7p7i)%Ip
z@12$s((~_Rt)5T-Gc3yF-aQNrk3y0Ly07$yqJc$O)uZz|qMZJ)3kFp^adCd(sY=9Z
z;9cztfu|T#c@=z-VO)=!KmJixOM-Pdldj8>lQzSdXHhi6@T1mjbA27@jER@-+RpCB
z9_PXzxPr_im|!Op^6$(c(^K5`scTqb4PDox(Q46^E|cj9m;Tf*P_)`uLzUF2#5WRQ
zUfS~!77O)%>Id4QB{ihqU2wU>cDj7!FaL~XMM070G+J$DB;)Z6E_lz+!zm}?TWR!a
z2-Ff&x#ViU2odV#$n&gx9sGSZHLQ9X>Z6;fr{_G)bcj2*?c&#d>%Z}>yPiNt*COlB
zVDG*sIDMt^?svU`w_g6Tpii!j6pI?ct3kcP5L)^{jmSr2i;A{W9i*m~)E>&w?-@ZR
z=dUwl-XBk^yIe|BmIaWO9f|n82Eg0+aSWu8rwz<2Udl^eehEDA2sizo4|4C<zr)fz
zVN+uWd2AQ|{Nq1G^UTw^^yhvGx%eV_(>=^clhmYPc)Ug>luB`|k2Z!Fq8K%SQdKn_
z%m95v7mJ0X6~c;<Fsx7SJvJgsk39N#kQVTsqL<Se8lso?IPTc>sB?(5bw|y9B0Q(#
zpmR#o`T6ose-Y7FO_Sh#fI?Bw+4e9Lj&#{77;Z-{CRGQ~T8kD1ng|a*ye-fxbH$XN
z6>^BrQsEm>y(peb&MDn)K`(c4e#>K=V|sc5v-t)1MmpCndZ9=dw6JTBCv<AZckgE9
z@|BR7>T|I`99z7OnrpMFo+u&@KB)x*aM9wueUcN3h5lUogOA2Js7HDZf5$y!TOOiW
zB&aXg+sS$Hl~<v|V^QVEV<W{TNV7>b9PEL_IBvV{TF$xPLKq#wSQBML9g0P=oQbbr
zi$3@;Lz>X_Jqjc2?K)m?#g(w}SWK%KX248e(n${2#`6&OLKjH&)R)7prHy6avfED1
z_*cF}e&=0`V#uAtH7y6!GIG}WY`F4D@}U-q+K~A8{?A%!iid!Bbd(84eza;WJ)lCF
zrRu<#Xh;;1E-A>foU}2<q|15lfBIuS^W|^TShAe4wX4|M?Qr7875vg~{0eV==?iEA
zLXU)?4&>EUUI7U<6oi#FyY%NGq+h!AIrn|nQaxH?KyCQPl%Fu}sbIhsU07@Ot%AWD
zkGBR+K8Evu=Rd)oU2OT_CwTOqzeZ!nF3w!Hi9K7N;Hy9NEBKR+<u(7$ufut#(UBg~
zNNJfSuJEXleuCkBWn5Q!zIgRj5hE(I_Uhnp=)y`$2m9LM_3VJI7LLoc-WXn$yYAgW
zduRkPLShZ-J<d7KJmUpb9MwT1gL!Zyor77+JS*wxFZ}yHCSH9YsYX!^<Av!hTQN=<
z-nbFhuq7H7=eYWTsnhAQYu9cX!%Oj6PGrrl7qqsMX8#!d8|S^l;3>SPSGcer5tyFD
zcYBPkT2o;PRg9=}c}1z)=necJB~n=wOddGErV~y=jj6;EOcYLdzQJhEv+E+@unGoq
z*dayr;>G*=G*uAV-^<1XjK@1gyyvD*Uc>U#phZSzm27MU%U=B|@<yXdw7l1#lH`Lr
zi&&304)LC1#}2kX`ZycTJcmL|M1BRZs4HN4AGdwtBP;{5yr6IjNt3KK#>sE}VSK9@
z22)K9hnAZ3y&O4<qpsmH8yBh6TB6dE01=eM6*=tR&2@kGH>?rJIy0meurQS*yy};K
z5jWfne0R|brB=X;;3^dvLDa<i@Z1iI`h6}zsT$P*By_&O7>8IWv<oJ|G{*RYzy1^#
zzUC+S$8X%r*t(N&os8p03qJVU|Bi3})qmtiFM0tj%}4?930|UFh!4ZRR%}R;>-yfh
zfW+V-ipitrdBvg(2^6NN3WW#-o*a+Ilc*=fk`@{S4w?zw4J$bLH-3&+e(Ha-_9w4m
zOKLcfDaVYhV%@HBKKsu1aP7bUS=#q(rB%S>%p{r$#tP2+prU3ZiX})ymJQy7E}vFK
z-QX{s=ab01#ey7#Pw(hIl=@FCaN3O(?!4!I)EWx)q>Z4}-%JxWpLl$w3|0^3^ZQ@Q
zy!%?7B|EZoQ!mxboGRoQ+2ap`2*W3xfVYNnjL+9*sXW4CJ9d$04%NZkP|jk_^>i7$
zNu`%dUkW`qX&1dL)Bpl>rg2%$(7N?#l7xMyK1xj2Z@%=;XYF?ty{7^witY?}f#KmG
z6+2L3=;F3Ln&E;Y$nDJ$yzYw^Z|&2N*lF#zGWe>Z-5j2HlwCL6NW1XlNyg5+!^u~?
z23D>>O=u%lO=RtlIvc=vunM>Q+rP5;+!vwkHewC+aAIZ(_Fnrh^d5ec)I;VnOuNN+
zUhu*zUk~fn<Bi2jozp1aclIjtd~+q&>3x;ThC$bOuy`fy<=pf)?`LFU29sw<qPT|T
zfI2RE+goANMl?<Pj=Z%gJgsT$OMOV<IbLfX`M;8Z|1Y_->J%#mYk}NRWIYl)#6SJ(
zO}ymtALDoa{KNR>2<_n}CZFa%|KgAF#Si}>Z@B1ehM1v|<XB0_<E#TNsHRAbgQv3*
z=bFp%lk*v${ZCB>3T3Gc8;7n=IcH!9%4T84N${>zUoxmu5}G&>vQ;ZN=@);AOaI}|
zSn-b6vn45*=w_U?dL1k7dx$T;<K5i!bHBpqfpL<V39@X8G_m1%_$c96bE2(&=|cNi
z53!Sv#xn2ubWUmI+N^k{I*x}QeS)OXqUbpqX^M9l$F5t$aH~nhBwRo)Tjzy}6qpAE
ze@3Heoh}$f2zKv5$0u-Ui_z0J<86>s6?_2ZViN9E*!s|;c<&Gs<&9MidSZ&qt*kUk
zt4iGWsH4}*0+msL?tWZiVEx*#U!~uD7nyaoRVqz!7Od)QhS%yGm<aO=?dE{HegA<h
z?%$*MI_%v3unGoMT?w4Kcwe8|ByT>UmVopYSn+K6%%>R5b1W981>+=~_U0eK+XjgW
zMr8o)*GRchf}#TZ_AtJ4J13lZCV7%lXw;Y%k9Rpt@8^zBev+kX(L5tf3|-0TEE(aL
ztFI<+CM4EG+Q{s*Q|3Y?a_Dq#Aw7%|=Mf{7|AZ1cE~3G6f$!eJ6aRWG!wQKQjA_!L
zi8=mM*1Yv<dd&tlv9+KXhIOk#N<$H4VwXkrxw(dp6q%dLZq*u&sHiJ~dXe_XivrRn
zyL!r-f9}8YqrdP+JltDKW7T?kll!>(jpuRm7ygK!c;h9czDFv8p@H)OxlRalSYu3u
zmiN63=9`w7Z}eS`Vv9VT_gU0uAWEW!M+1e(3ThRs;1G;h5^u21((p~xHptd);EZ4Y
zRj&BNhZud+W!#xNrV7VNts$D%-N2_`c{#iO`s1`ZJ$h3IaMlHvS|0{m6>u7qtP&rL
zqm1?wjw44176}c{wS!5kj#p4Ea|l}=eT>QJ3?pg8q?Dq_IrG%b7}Z*bK?a*q3$B7X
z+2xs^h^ac9UgMevaNrAOwr)Z43`UnC#~%lhR8r4j#v{z~<%M$Z{ST6)hP=oSExY2n
z@Lti{*;&fDs*6!En!)*kEGxnsk$6ZmW`>4f^(u^*{{64dKF{hfXg{Y5=9=me@0r-M
zlVxjG1zK~MEE_24N~zby>!`fWjXD04Q!udDU3hAPl+2|yYAqY|nS+UO9{l20S!qXb
zy_^Hxj14ck7>+##O$|naA$(~}r4>)Vuc7?V9k+4H>8Hc!62v4}@rZY*g6m|=eEWKO
z_ibfN8u-GI3yk-2PJPv@VeKktBzUK_F#;XH6)WYUL*Pn-g~q}DK88!xs_;QUdmiaz
zT>m$J#U>gs*+E1p8kU_17r*N#VQdLWGsPD!yi&#0u5133s7o0XpLu@ueD`C%U$ehQ
z=lTAFr>Z)z{>pn3^wTY>IUdC%E&k*KpX8huzn-t%crRlcS2Hoan@cX(%r`&(cl_3S
z-a(_7B9V+#QoL921`&fHsnNXph<fFgT>Z^_27byRCVie?Ib0<LRZryl9uZzO(uou;
zlvEPb2yXo<&i>8c;PwCb0ft_EDfj0yh_Rf%bPf0X*<bVN*S?w7ZTHZeoTSs8Kx}Z9
z)zbJ)Y@`I%s1z#m>{l6hnli7}*<#V~oT+2WXBav3l7RDY!*^~&tik(&yy$^Ik!M`=
z;um5QiSEP^;jRUu`N5b$kl@*{LaOy_LN^;1&a?z!?{{x!$fgXh-2h95!=8x7b(A2Q
zD8{7RefRw&4GZGJ{w!VU=jhGN^D{`ekLAF*F##ycvYbwqV{8IgShbGRuKsthZXM$L
zfP)dKS<LhSvf0nIpK~Y^d-pQ7VnxUZtM@MrD4C_Gws;+d*J0)OhudLz(c=Al3Sx2z
zw(4ArQEECtrVt+g>esO|Go(HYD()$i3*Pmk_+}$0AJqawpAxQ>BP6{Uw%)geV^4n}
zPAtYNwld!;NZkxKfAk})NSk!K9lSWaHTba+PI}7~&}>CYNpStFC{r>xgK&s2d`K*E
zU^q?$)lkq_F&GIHCDT*3fBCD7-*-Q@(<QY+;d91|oMSGznDnA^@eLa%#si9?rMqZ_
zz{GMvwED%YP%clZkenRZ+L*N-X8)S~y>f{3S{8UWiA830wiqFIT_kOD&l8he_=cb6
zm;d1JX>Z(2VuiJhj4%BC@9^RO`STnzT+q;*)U?3c@NCMWEb?en`pIS`Wjnx;t>cP1
z@RJ?fq2qZ9{Axfy)rBG<%sPbo><u_m&N(XL6e|{O2;Frnc;WB=S1!BeZ+PT{b!_Q&
zSiNEiCl$&k-uY8}`&a%8L%TbesTs0jCJeJW4W?027VYJ{&%sviLg2vSu=8_o(P_mE
zH44U9zVR>r#?VlTh$l%BRC2s`yztD^qsg1?n~=>n#4raGc;?2N5Bkw#(hN!=3hFp;
z-~DLeS$g&vXd@knxdy;FwD3qO+`r`^n(ZboiglQ1FgY-zrSnI1t^F5O3Ks#h=#79Y
zKnjY{B^>{{H;^|QBqAhXurN>xwFZ*WmsTfI%&H>CD|BaYd5#?$!;2l1O4mb`H;dO%
zk&_-~j(@lvh6@*O>63_F%*qC<`6Q_))FD{-!XrN8wom^ft6M{4d5>NSW2c>pJ>zt|
zwP6;auFtB`0LlcHO6!GfciqLZRqJ466sZjWBOaOY?Em_==-vMSLqL{Ip$*HpD>&oI
zAA)0#$Js{kcZh*UR2NA9=fjB(pT|7lwIKCYs0Xk3ZU+1JaQ%lq$a0_H^BlE8(QGoa
zbQ$Np_dOKt6l|kXlaOeTHSn`Ztr0HsCqGZFXrw+R>u8{dQ~Sc<SRd6HwT9rT+|6c4
z8$<lz-+qjjUGXj++B3~D$8KU~?=JrR4`0sLKJiyvc*0szdbAo%ay1khoSCIE2J`*9
zS9vDPXU|mXsT}M;(#1rx_Pi?ezN~xIS92#eiWnHv_iWeLacI~kUX<*(<9OXiKf=lH
z{c-Nn8D>N{edTIe-@b)wF1vy!KJ+nKos42~0x^&glz5W>03ZNKL_t(o6ZW0hN`F-Y
zoH_5#Rt7s*!t9RLqMG}2XT1eKloGY>7I52l@1fOf;e3V@hbs!!EFER@i5m#A%R1O)
z{yuF`o^}wA{@g~O>juEhJ-VHI_byzc$;g?fQJ5xu{3({iQ6QjGz+;c^!Z{EdsDwVw
zfz$_y>wZgh3EF{OQqCp>UMEviGvRxUfVJpQ8;nKO<8gy$fDWEx)xdQLMfyNT%EtH8
z9vX%uoS9<!w9)YIeZRxFcpa_RVdeOTTQFR(ct4+vpd^DysthKpzU~4U9d9Ks;NCmn
zk%wqz4yPGAr)RkEtv`yU&6uPO9@G7Krr2HR_IPZ|eVlUMxwy1}H<6dBfv0R{y4>}t
zPq9>#tmu-qTXZ#}Shj-W-f}g0qZPTSQF%QoIFxTF98Bj@4l7vC%jnjq&Z6PHy%r?S
z@!-cl%E<0LB)LaSLT(!D&K(!N;~lVa1KMbW)^JJ7in5?qYA%|yR2~P1!=9Q0NuMM+
z&;`pctv{1PLJIwEmMFd`=xL5`j_|gh{%!u_``*vk+7n2uVTqmOOV|7{?|bjnq<N2a
zk|3ny&LId%JQNj+N&-H`c#A0_A{|Z224@y#<5mmV_AR1YbJ5Q_xF9tplp2RQ9{g%7
zm`GW435~`mJ)3aM)o<s`pZPS!S?BV=#1!qM%^B?$cm2hm@x`~li)72a*r{ozdQ%u{
zBA2^-A9Kh#$c4LLi;2gxsQ{fF6BtM+9bzike(v7-7<<M$*rb8WAWc&8!g1#5o56dG
zwS#AYelI`MGl{a6PRpTd7ab@_@|*)V-%678uwzSL<1u)#k!~66QM~VW_qRRq1e4P}
zJVN2(&~n|lalTKg{llnYAC?BABGKFtrh8c_Q3#kQbu88STn8TOj`LFG7o~tmkWx1S
z2X^gb+42>TSP&EB$|HSB>t~{{7>keI>#%ZsbLdT^=K#wV@8^?~Hdaa6EJ&_Q`oq@*
zqcJfrGVZ$O6Rc?rVR{*ofM2$frLTSkMYAzDH&97bnwVgN@deX69wtpK!y7l?OyHNR
zcY)(PlT+OKU3z!i%ZMfvUg;QRch_;+>)!~g*Fqx=Xhfv1^wE0lquF!Zbq`L68Pw*P
zPq|$N2A>s=pUL2XE!_XPe`K9lOs_+s1)WrBoNxlGU-vrlw1p+9R9!^0DsogEqNPA!
z^s6b!o)zbFp_E3ykEee4iTuH46%MwZW*1-cl|BPWo|K1^l688RG-b=qeVlpG>-oPo
z+|3KlJ)516Y~|`JFX20%|65Mouo8=-nWT6J&U;dm1VcgZ!$A|*g4bvmXljL$_^RX3
zPgb;2tD2wrepv2o&&uTEdr6dzY;au$l~rU=1ha%EB#N^8sG(1xG{oQ=4f2&sx#Z9O
zUoQKT-{rBjqddNUKgW+OW&LBj`P{pHmV5s2FBr)@-RTK19)sYWtFn-CvPkFMB7?hI
zw3g63ho93?c^VDGgh&;N$z|Cc6l?g(H@=0m+t^k@?s|A2>*T!hr7tF}I$pE+Eb~Zm
z$`ODdPxB^}erUvdj}wn^o?SQIN}`4pXPgRS?FfW~Sq;%3N31B0Fz}2?@G3ax2aG}I
zcHHXsL@S`II+g^y^A)(5Wm$})LHan%oJmFZG$OMK!m^MF=bEXh1EK=CXYbBkEM2t{
zXKgTr^E&@|wOHvN&DXM+>wkX*191N0y?pYM17Fewv?g%|l{_0h28x*jJbeB2G!;mM
zy_1uibJgo%WF&AJBibPJxy2rEWxLdP?zr(=oOa%Mcq!3j#Yh9i(a5H_{hAN4OdYOp
zXyH(6@gt)gcje`1Ga>Q846fgnlmY2AnfFA{BTD4vBQkj()FO;1-b3mf-~QnH8O=Jl
ztb<s=Nx{CX;N?I5ZghAUY`^#tmt$Ok@h+GPsp5^|MDe9Vb*U6p|9wHm|6nSv{bnyU
zB31P#U%e025V}Szeo#O6@Dl!c5wS=?E8ER-h{{DG6{}1*v%E*6-R93f_*q_f+1r^O
zUd56X%h>baeSGPof6V{-r60$2r)W1CI3J%0!AC(=BLW%Lo>8#n8a6qY9x7S<%<;*W
zvm=$Axh#aa9c0z$VWpbSrSt8O&%2TX*7*~T`iiWK-v$+z)Cd|Sh7<K9NduP%_R@>^
z!H@nG>)-iC9?S|7l+9_Io!5MZYu@lC+S_(wCT7Ts9yUqvQG%5yb1cji#wqiW>Ss0@
zscH*bx-DMM)Vk}(BCXM9!oms4W)9)nZ(L70(xzK<gX)iiRE3vacplb7`Q(8Ix^Nl%
zY%CgOHn3v^?|_}VnSS&Eibj(q=bnSNNkr)5=QBhFxKI?_e*1kSX$t_GBqdAuAnSba
zi~ae%Anod-0D%#~6`^BM?k{6w(O}~FRlS~8uT!4TM*6Pu-g}HF+MB_7#f&bA^JZbl
zTAiO&|1ZiTKM$@$=K6=zVOT5}o&sd2u5dzahQVtXD#i%5<Jk63Ut*;27@t!#VX`sA
zF|T_qdD28|(m$}Z#u1h{44K39gAXH;FuZmHUP5=l8bAunbWV2jElk{TJ0m6`&%31U
zl&Rct&TB7+)$4-mqv=bI+S2%}a$a62?92CXUuFeO6N@1g%ydqE{dMfW;YLPC$jIrX
z!nhY!pZg+`3(v<j8YE(|D53?OJ-gU)?bq0K<BiCUog|YJ*zOEA?;&}H_#C7N+L7WT
zV&vnjU(MMzW<ye$@XQh>mBEr>aEQ8$w{z7hwBka2k3qPE!)B?G)G2B;9X-szA?}&#
z%+P4G`LUn<&%F0Hf0s?izW`el9KU*$o4@scx!?uIAr!P*L*y<DG*=27zF#y2bsir&
z4UrQkYJy8@=}}zTN0;|T`gExI4kDcwdz?K5q|c|I5d{)zIx8Zz;;%laFLB(Uf1f0%
zu=W!nf-#@LAW1@RY>d-??LYFGzx+M6EgfPX87o>TXLWNv@y56C_y<2sqg#+oPGGIU
zmhzdA!Z2?b(7{|o=YbX$(~oC<U6pG`>(0Nmhoj!GYwrYi+;KO<Bcn`COw(vsL<&wk
zb}h#rvnF<*XP*g!#-0ZYW6!2_P=!zLNnB2L$9GAmrfIHN38$Y#k=Pj3D(ybRgl>aL
zEw|o!H#TjCGt!g;$EN0uBy&8jXO~zNg;T*=8)r#`JkR@X(ec?yjUQZ5Ze=drtUfCy
zB8XQe_v~SG$#Q5l`T$P(y41Uti+#YOJ`@~AuD{q}c&2lJn1BUY#Hjs@pm`3xDIWO3
zKeNQ7NKr5`(_!8D7s2|CK`+_W=!~^j6(flEh|k%2*R5<m<1A=3Q8B@3mw+%x*5&Tc
zewq>QaXzQf3cd>y!y}yVj<@3*DPnC69WnUQak=j3SsPXUUW<d-NNZ_UB?x)m!4xxm
z>;Jr;RZj5Tf`(1$sbya}%!}Uh6BM!I;w6T0V`(p2!Mfv4q?noE!Eb$&dp`3iwteoi
zOka0Bvh_jOwFfgffpn&jyo)Oe3Ot4J6yk8k<Iq5@DM7a?K~lsA(A2on%Rw<39JP%q
zq8=efv<Uz274gw81B0ju1lU<r5)OhO*HL7tsScI2wzBmNHJY17j7b7lNL?4iFixAZ
zUj24Hb=^Ij{gR8>ao-l+@!FU0&42h`ER~FAYU1FdCn=Ax>im$ZsIn4JNMj`oLW<%P
zoVSQiYEX0k6Z%+K`B-py2}!J{udKwJ1MBETx#Wm8jm)!g=D9aAD1cwm!=!9unviUm
zLTE}0*Bque=X@@|<`XP?!<9U!j;!Z7d&wwU{`!COwV!+sjfwsEiE%Q?W1P_tS8DrK
ze81OL^^;U_;_yN(C0e|meG9Ko9W(Ly<zo2CzkG{cuZOBZNoWnH^g5Hg=2aJw7>_k}
zp2|Q55&MHwiSjHtkE>+4G)*Jt@$cLKdBKV^&VjMzKvMHLsvFxWnfKg(-y@`HlRPh|
zxhTv6LFOs4XZaRsjNQiCc<y+5-Cjj6s&f)`R(O~r)2@BhC2MWfncll|2TRwih6pB>
zoehc9<`UvkeYQQV#p{{qFeHblINGZE!o~ae{l#Nl@c(r{LbcR!7QkgWcIy^;k8GoD
zT4Xr(<OQ#L`wv4S!4jO@DfOMD-vw|4p0Sxa!1#fEoO$AAa*a0j&Y=i-meYLjUiRN~
z6UQbkdYx$s15=rD{Hrd7jT`XI6yvGAc6cu{@25JHjo2WLv=Bx-yYvv0w4?0!;#cY3
zzm?UtNiWMN8VM74HedDwaOx?LHV_**w<4$!e5=Xui6<~}>~YZT!TvquySKAv*DfA;
z_#yH<XJl-Iv1Q9ywr)Mil4US53atieB@`f5@Rb6PAYKBq!UhE@F%mWFO{Gv}q&BA_
zeO$lq+#P)~Dm_z~<vXAZy5`Qkr`3TGJu0e^`VuXy5T!W$9i6y*-o><s`O<Z_@FV}>
z*BDy1j#aBx^Z0%D@{zy!JuZ9M*<@LV;dT>^v^X2*6G}PS>Z8<_K-HD0YkaFx{&aP-
zHbDkSMOtSj#nM!z`a<<ya`Z~A>eCpI`jqv!UyL9QtSB@a6yiDcSAT&MF20y={QmDV
zwtWZ34KLxq)@@w#h9BXL@B1AZ7hS+iGo#%a4$HXYB#9qPdep~f8#vdZ@~lRepQp%T
z>`Ix?c9PIRwz2zZ63fRw`9+3Dm(k056zV9v;<72;bme7{$FBnqG)_R&qWIvsH($hi
zwj9_-1PTqj-Y52PV9Nv8)(9)lI~&sAW^LkUG{)dv=t?`M>^*RRoqP5$wqjLOK=q_a
z1Lqyqg#A_t6AVI!3;aBabDks#U{$Y|)&7N4#fsEs@XC(jfCY~BThQf9A4SI0_&CR%
zc1AFSjKzX2Jt!6|5EggrQF^^lRof!c`JM^};Dw9j<WrFpNNvOWev{IKgcJ?I<*qOO
z6U!PYl6iWX)7o?#a>nVnBnbh!qIHBg?@_F=2#_Rj&n-7{{HbR_(!gRu@DdAB(8P29
zr$5P9=5P^3_KBfu;M})dfi_z);ni6NTj34W2i{~|-GM{OV=l~H)mfQ?iNnXetVeTd
znj1g*QP!pnie5&ekuqaFI=Ymz-~KkVX|YDbplyAS1ETC~g0vEJYzVt)BkNqwx-6&I
zzmJLS+nE{P$JRUVz-3*A+HICBUB=MT6*N|^#4KM5X%kH>-lhSeO~Nw@<~|-V!PS`P
zXdDK_V}0ceH4V&&riUa-@D;~&jxOrF<(14)6)MptMQT!d5?kP9z2X5&Hyfulw1mI-
z(3kj`U->VrKldz>Ud9sX@|7=roQ*4oFup)+BOYvlvNAiflk$DyDWxg{v^px;UBRo1
z`s2~ys8p^RP~7lU?KIe~4{O}V16FjQ6*ONFvFPJ)umrJOi^jxZXcIZp-jg&MIPYm(
zdJ(TV=l$ID-`>Zg*Wbpn!gFS8h=2U)U*?3XUdMUA`RjD^36jxK5-~v$rYuU-R|Tvn
zcnm7O-|130LZhz5(&DK&kt!t|69Fu*eTthXniV`TU2w~tchg$B3J;`7h<nFvSi`0b
ztFfh!hUx;a5C@I@&z2`#&J2Y7TCv`-|Mpwa?j((s8)5Sa<kliisYHo<&<yj&kU8bM
zcW%KZQG#3~c2Tv8%hma>dPwUcAyAndVQxWz!WGqhK2#1b16>pPodXH3?fL%ww3-d1
z*~A-D9l{zB05wspU_n@KaZetF`1T8dZ+$-n!{YUnJem?ct+f{;e^4_;#;1As8~?^}
zYVn#eQ8-S!{4y9B0UM+Z!4Dw1RwC+M2z1AJ?7^-4;9IZ8N0kc`-5|9m$m2WM@o(3$
zfs~o)DNM^Stp&$ibP=-o1$dJZbxCVML#kmaQymZsp$~_T9Stxm%5FwPLYfYq`21&S
z?c7DuEx<x1FyVz$FMl0uT#s)g6-T~20EIx6aXu!bG1K*;BuN8JQ_Rp1%hw;n@*;<+
zN!-3&^!Dy%eAjOF?|cGxm#5Kc(^|5W(e-O-tzM0cj6&K#O%mu@aWGIKGNqL<S^Tq5
z$@KMm19d)Wy}cY^q3U+)pyfA#wj*fJ%dRRdiOpk!n)uH=p`(NcFU(9$F*LS>pZV4I
z@fRQdGAF(0#q4?XQGVd!)A`V!|97<8CAA*)rvF|sdBA9>IcB+dhaWEvj#E=esnMcB
zd%xfPs<gp<Ig~H8BGJUV<7?(5v;7(tLgnUT$A7>0<y`Hq=llU#&OAw&Gf~mXRI1u3
zF8m#dkS!hJ{NMYZOnmMOeB*!p0mq=6v2rP$ul_5a{`U2}@h|@nHXldNrZk&veC%K%
z7Uvy_#7<}lla!shgFG3TSE+0<F?nk8gOJo#vApt939G@RmVf&4SD48QmNZhvr#mFp
zkaaWO^yb&ouwfWCB%`wnfw>~0dXVS&Og~|XL;UJPypufRp|5`fD+z1QI~&GEQ5CF>
zaK1*;ooWyhAmz4O?+MgQ&0|5HfL9W0QSYl0VBun+MmH?G#XjZ=m`HQ#R{`Y=SdcoG
z*{r78nG;S4BZ9+Yyl3Ce?JQrt8XAp|PgG||`C5Idp?Ow5vBm4D=rVlI1;gRUEG=G#
zpA64VH%gA;pdqG*Ue50C{2O{;KQ>*0S0PV_SpSNbqm5RTR4UTXt4!d~dlc&pPuz1a
zYc?H^4vpZA4LaE|`IDmH;cLGP6B9U1X*5#0zQee;yyWfgL|cL0<9&z}Qd4oL<S@&O
zZ?;j-yh+v|>O0V$o_kX&blw#OW_%BKeDV{lDIC^&a+l+VniNaMIN>dC#5Y=@JzPVq
zr6$*6VsR7hjVF;L1Vp0bst7ci=!zyRTS@B#7%Fo7<RraaJDA$BgWkk`cHDUf;Apm*
ztXR34=E_yDWCb!d3P}@561+`<5>+(JBWim0r#?N#NO0JWI@%h$DZVaMtF^xVEKd3$
z65l^qO3iNH1)C_62wAot>l$>jjFBbFc-=dHm9O4-I~&hBo!t*T!h7Cv8UN$geu}I!
z!%({!jE2z4@Xz#RZU&9+W_wtaU8upja-b*{DV2Mqnzc|3j>S~PRYdM;-U6XC_HnIi
zjnXn}%~ktI4}o9G5fl?oZ_@gd;&yXhT;gDU>1Z1}oVB$mh0;id$;lag?GJIq+2?WX
zZ~ZqqTOVSLT2AWp_}~?9<hB3xKe6gZ-pKSs!SL8<P*GBkScCH(!C(o6b1-x`l)BX-
zdinHr8Tv(<5vln3c*R-EHJ|=x+Cxjh3*D&>t<hm5cf9_xA0${dmOFG%_y2GelAaCc
zb6N06q>K6@!#s2!orfP_w7r6*7hZr%8(14?TOPrSR70R9>OFkthFeJ6DMc*6tr2qP
zK~3;d(D@2tnWtXRNd2Xv9C?H+%Llp!)iELc7&!3td1k-Dd0w!0_ij!*=Y^HpqN#yb
zQhKe(EYG56RZnQ~deZm9A#(mhbr{ZFyoKN28O;JoKp8m7#Gj}(W4%Xw!JVJ|9BbNR
zxZIN`!m?9NhhvVxl{5lB3#ITLU)tFl$Afp>%K5Ln99NpY6b>Ta(o9XV^$TBNm1)t-
zGK@H;oU-KfbKuO=(Pk?QM~n7pAfoe!xdo@$be{X6bH8KGtY%;1JuMttKm9q{2gXUV
zBB&ZQ3=^uH^X4mI?P}E8!8w6}*C-RBa!@<UKt}gyuK~TTzGgx@R+1EHET^?>8SRr#
zfnFE(?WVVD2c3O;*!|cOxNY0;Vrh*mVPwS$hS#jcEMJCcw;@STYrxv7nkzdXzFY}L
zs`(Qi*Mdg?gg|@0YXuo~v(4%xFy-0_EeyORNdIL(ufVniigEtsQ(xrF7rltlrK=bj
zTgHVie>1o5?Xq^$Ms{x7%Afr9uk({{c?F%$H0|~fg>%^Ih>Asv89b0?lg{c=nsXL?
z%^E~!7d_PG5~7Tw*J@MrReq#IJbnK_Jp5kOp;ClrEx7Px^+V>qxKtInP?NXSni-t$
z%L97>TX+&<@gpg0TF0yY>d(3RPyd_;KKVJ;i{+&oj^jJO`{%U3`VC(BCx1jXISC_8
z8j?~tMWPF>im&vAb>4DL7AzDFMv7Rxo{WM)YBIcG?|Fs15VmZ4f?Mvqo3Z5^n4X%Z
z)wCF+oOIGgHlMU9#N-&x1}E}nFFrZw-#p{bGj>*t2082m$M&yY%a|I*Hf(|uH{p^b
zh+9O#hZwj+DE9@LG`aKMt$0a;fQhNmwA7;qrR&zBs+P(}z6!u)XEx7@S;mdAU?FtR
zb&koJj~_1rHR_SVlTS{C_`7sTpnPgA7Sx`H3bTgR7kA!Kex3V06%4A{RMoQ=@89>A
z032+%ckutik?DFlcJFp3?!JrFq@X$byE*4w{sZ`CBP5tE_!IOg`&zabNK`0xK7kQK
zd-*DI5q#lE#DMYmZbs*p8{v^hXm~^JE#3$ddB#O=x)RzW5fKhEUs^*ds(A^Wj~;S3
zux6bI&m8S(sCO3}$^L!Z`?=4tRt3bN*5gy~tJbpSs;h8miYXIrsaxVidwcJzz+3X?
zeKh1L<>;ybHjRVgQOY$GUf>ghHrnXOCejVZ(sYicok@zlyP16a3A+3CG4;fwOg{P$
zw$WmE*-}PUtU^|;fuT{PnW9M=P#kMx-WCW%1LF&QJ&G#PhQ7I6C1>V?gdbMO92Dvq
zG6tp(z+?CD^-q18Z{Kk*zw*0(OuA_ec_-oJuYM<wjAyK0y_EfrKFAk7{MWqVoRjEv
zJG9$vT;Z^ZjnpCEcQBnV|CWVPfce*c^}E(uO49FDA)(_^d6M|D8tSWcRL{2bvZxt=
zRr58%obG|vj%Yz=^|lVD%zy61n}b&q^OFipOqR;IlyW92%#YaMlCBD^CV5hD+OPgR
zr@it;eB!;o$;mF|`1U9>cRkECuYVg?{{DZZaqb!P8lFbd!WTXW2vh~|+=17AaldeK
zc|@Oe@jCJ{v*hF3@Vj`AZ#MYQ$3BHglS(@J|7Y*a<1DGlyZ?R8sk%$Av(GRyFf%Z$
z0*Z(mYLMWri5L}EaETf<@iobtyvb`~Vq!FlNi>N`B#YY{HEw{o00IWw00oh4SZ9XW
z_nztX-dlCf^Zrq%>ejt|d!`45MeX9#pHDO0eQ(`URnK{r-%_tz4vdZSrq{g_buRPe
z@VP0Obg0~wV|@8p>=iroaj*)6#NMIU#snLGbQ84{Ry_0R85B6|AzH8wbKfMzr?7qR
zC?n%d;^6^s&FuApjI;?#a>au%3l2ju2@G*k5Rx=4#nX~AbZIJ~y;_x;%eH&0SN87M
z#-bG~Ac_zyCFjy?>YiDpB5I4hUOF6S=Z>G#<B6(T6_MuQP8goZ>||Hxj>GNzl~oIJ
zOs{5yK<k?9{GY2CXr&lqNGpcPzIhBi_c^p|gbhh<PAZhpK{p6aGUDz1x8KTfr#=xE
z#b693w4_xP<65k{>{8}?i*-t3489%_4-PQzS!d&{4V@8s76k=dL}!sTcJ)lPt3t@T
z>K%Q?5VWF}XZvNBGCVd0jRqPiiM8x$dCoZhl`uRXsZ{WJ!qc4!YAFFPxsD1#6d*SO
z9Ry{9%Vnm15#`O+;uH7?Tohp|5z)X9^N(JgO&@mep|O1{<9l~AxqCOG+qUAPh<I?A
z;iFa%FI<Qn9)fxuwGnD9A~sAH0?bCF)0?9(;tT)k!+0T5(x#%mLB&}r2M#du&C9v_
z``2^tz3YjNUd{Xd>kA-N9@w;xSG@eq?9w_57cOROWDDQ9?DL$mY6;%8sMczD%?Ne5
zTd4^=nDWTGe2ov4y!8@4oyO<x!H>`3FxmPC^C%4)t?4ueD@*!L$4i&Xi0MC*?)#AG
zg~rc&_oB0yYpY&m-T&a`9-b>yv(9C(@Q$Mx)f9R1>AdR8pXED$@Ta(2*0H)ez{!c@
z%HRCoob~n#S@zo3FcnWy>#w7^Dp)>8=pa2GeC<1*iMivEEHjN5oYTy&Lxj-=eEIUL
z=&KKsq%A}fT;dt7M!f9#=Mb4}YLr23eEWL#RM{@`ko0l(w1s#Pni&=6*m?DJBqJkK
z7q22Z^GUe4j+l%roTu?3nd7~UvgX=tx35Db!nyWtWDuPCq6sqdysYadrTAt_@y<Cn
zt;|;G4yOSbAYJ^C!X(zQZ|62nKI5!RA<gu(-zc~#?MBZWujm)xFd_H9>2M2%)92Ri
z5$OC$X9}G|e^4^%HA%Sr>)&Fbtq=h_#>YAKRWE_zMR*%!$*%XsChNp!3MXDsETofc
zeQ-S|J?(6oCdzb|^AMLPvh_iB-*PLfebz=yC1KZ?<McQDDhv*RiAp-mog23XbDpv~
z)5AG@knrn3ysD5UEu!%Q+<wVdSem985gJd!8)W4wmcIBrv{nn}B0ZA{DT=Y~4s^Mj
zFBq|H5}JSy1Z7u-$EXsEDMrw$2DK(j{wo!<e~9>)H7rOH{DA{ZZre;_&u)^jee7O$
z8<kpx`oIwN;YGxY79zvLP^+WWDn7Cpu^|c0IS-~t+KPH764JzzG!yD#VblHW{o2($
z_>C*6li;cqhMsf==YQlQq?L$u>vr&pi++!!Z!z-*43qn|al`*y&VtxrCB@pfc=5bI
z%!%xUI%^pddyoQHW?LxF@CRjy*1T}Yi#+cY?;Hln3LD9!J3EsTJ@^}xrVBc4Yo&k%
z=H^qgkp=!)E*df)jP|E$FQIl^l>E6c68KC(qhB<J>O7juYCQMif9CFw{uApjxspXm
zlVer6@8kc$gFm^Mr~TDOn9wn514CIsN?ZC-dNdVO+DTix)}P&)&jBSK&dwx-TfDUv
zw1rBAORoGb+eW4sUN{IXkF|!0sRkFm@}&&b1X~CiWJ+=6-i*x1UDL<co?LQ776Iox
zc5ENpuDqNm6;?j|nXqU9YD}gWT==hPGDg<i001BWNkl<ZR$Rpvj|XG;!S`>XT8&7<
zL|gKwOth^c6&i#Grx0ra!Bpr_SaNAvB>yI0$8xAGm_nIJb3@SZ3>gj6bnjtwBooOP
z7(yj#i&oINsfX0!adqA1jz{G<{jduLRh=ssj_{NLI!R$=qZI+P<zeF%8V_!zs#T=o
znNZ>67e1d>6k)^!UHoA7R;bxzyauDf?)x8L@v>!5t7Q|^AQR^_#kQ19S6xeOyooh6
zTCFD3Ev-tGW#_+$v|0&HzS1+qmUHfhQ!I4qaJMo0nbH$PiLmkVZxZj_Po#!a6B2}d
z-tmmrzZ&K*M6=m>k@y!Q?=BAZ!OpY8BcP+JE}cD}ywg(be{C$0gbhI>MI(z=V}=$l
zWGGExbQImSo&DQ3GrDht-P?9zqljvMKf{Zc(6@96@uJ00?*|j%qipgZh2yFLtEh7{
zCmPgzOtfzocm4AxXx(x%^^q}->g(sOO$T^z%}U<#cYh7F3U}SMi|7C9TQNtkq#`M?
zALqL7ev7(qQmuiJD4Q%uRxow+*vnV~nYkd4Llwq{vBmUMOU?Zlj9?^2UFIaMVd`LP
zsaViJG}j)Pv2^N{w-c9k$5UrK;h!(FxO}<thNa1&6<RLZr+;280hzr9Gd6FiRO_^&
zDNg$R_p;_0Pvz3z`vcCJe-w*%Y$yHk9em?;Z{j7N_(xg-rhl-VToQae%FTy@3?GUH
zEkxQnuGif02<%Ft;ksu23q~E@!hd}Ba||q83RXyTidw8BqYZxj4Hsb5Wy!DjzzNfi
z;Hw?Z!rIk#npUnnCMAO)h1#6tkjO?8{n0I?8#YoKTFR1NJ{zsaS+O8m6hk7NO*Juw
zAN=6Qm?*}1hgibCn@w#cG`)JWJWA?#Wk{t4_w3RR=bWd5+?ERzsAL|Xlubr+{)7|~
z*T$vn-?5d&%T|DmgU+IMT&c3slJb~dq`Bj<crp%>`yXt=aN69`9qHY=Gab^95aq6t
zw&l3@iYpnm5vJv6;i<1!2FI<&L~$W^=5wrKFbT_mV~X+0w!80Q%}GyysFFDk19+Fc
z_2eiIT=6aDM}qSS#KL%Mf~99Y16Hie%o}pKxJO74p-2wMI@`{t+2A$=?xc|tkB)KY
z<zHi|wP?!`8KEUg-^$f2Jm;72l{nMBE{M1~gPh!<E{XIU{V1LCtOfb2do6MTZ@xK(
zBwz7pB|;W1L>J6w{)wkBztP0)-OcEh&5Z5X!j^mPB)#KS2Iene(TbJy9kU#>Xc<&0
zI2BB#LgEuhJ(Y<`YWuga^)p}O!E67AqqR=5V-FtT{>hZlr=HGRKJ)K1YQn~!ZsBDw
zdnfabS&o-U77hs4f8&d^#wV!v*8*v}&4JO<w56lldyuO>6Np55Bm(+y9vI1?)rwfb
zVl%%HpOU6oa$>D5u(@T39^wmmTMxzed4g0VtJ;PmOh5mhuLM=Q?tICiiqh^1YY)Cr
zw+y6LFjgT|Dp8ChC3^OmTyV*Marrxbn>F+5RL2rl?%BhC|JwOn@accYuR4aL-bdBi
zl3#;_7$xm|iMGqPc8oFgkm=uCLGcjphyZC7qj;A&cYo&xH*(jyd#Nv4PO1rFQW}kv
zXPj{=$F5pJoUwx`6!T2(ULP7Lr4G4tw2B@t`>~V~YfvYo))SkAd#?Bv+DcgRq^H2j
z6?n0Qid)`QQ6C7zk|{)uHk)kNuoLMUgrpf-^9FodO|W31iOv-6+TzirDZkRDKBPgK
zwY;|4@%Q1ab4acTswQ|8WvGHZyLWT!>1W`A)4PNkMCXcvx#JMI|G^4|Q|H$0k**=y
z>I6PjY-5V`S6|EWI0i}B=M&bx;CawDK;Cl8AroG;ZE__pd*~aJ95}F-p`+K3s$wF8
zc$YyqrW{i@{|LWnEB#R)6KR7a6829Sp8diXL2UqR)FwT4Bz;FtqeMD-ylwe56Rhha
z@7QzAl~lIxz&0G-fJPA`t%NgQ{tB2s4~_ExNE>%Ob9M3%`)+Mw_jL7tEeq7O6@-dS
zGEZU^Y&N210Bmgq^Ovt+e$s?pyEw3QGrKo#;@+EoKyuwETIpx`DJQe?NoPV_rDCeE
zX*26C`x@J>zlvp3j@5CUoeyrNf8GFl(wGfPhPm*wpFpdYb+>QgwdcQs#cNMw`_8SL
z{=|iR{d51y)Z_sM2I?6CE~ikdbZj!+YUy%_X~LoOJ|1?_P&zT`ngH1-LVczx=F${Z
zh@!v=31)bCH@oO~c8J7Fsc7h7O!sJY4Ia%Tw`*7;LWhK-l%ajvU0!rPN|GD)8Nh*+
zn6w(xTDp*zec?a2=H0(VeEWJ9JK<@4b-wh<H}Zl%`a_0Z{!*Iasn+{(-WSsjuNl@O
zdwmmpI7%Jd4qNd<mqT;MVMgOXc(=ji(U~eA`}n6YaRv1$N!p@Ti)oA;;9YOKh{*VM
zj0}ciC2M=Sag2Hh!Qe;x%5|EV5ll!8X38^l%k7NZe>XZfkCiVt53SS@YYL?z5kn4f
z%pH*x!@c)E$mm3af&QV62$_3Zc%N~_4z}a;+_r>5_(4T*&Q0HsOZwz(afd9<`;ytF
zI8U;F1Thh2c$l<Uhv_@O-0=&2oO)OVgQ~7p)kSmbcI3KxI(?C|0sv5__*R0w|1NaL
z4x-8+4$poE&pPh~xIB?BYi$<ee8G@uDDQk=Jxh;S1JzpAS#_R@Sezupw7Bc*mouml
znu2kj)GL*x%Nc&=({Pnqrce_}4>cJnZ3quP#OK|6?|mT;iU;o<wlT^bmwl0WX$vcm
zBn}@}F-sS-@OkIr;#zxOLZ&6Rv&w3ZK;G3oaL6H5ng{prOW7pI01;|D>Orf0$cohr
zFJHy*(^44S&(yAM?A`Dn$*w(Y`pP8?G#c!@<9<f&yO+f_;)E3Z6pZg1VPM`6+v1q@
zN3Y_|pZ_#dVz~R3O}zSo_poBkN$eci%`?wh%a=a&VJ62$=^yMbI#Rg<G-b)$8Mf|+
zQg0nm^^*=G@v_FOA|^~6#Nu2^nx=pdSzEZ)60U^=#gB3#lPijN?~7?muVkmw+u_Ld
z_#f`-$U%y;o@rT`o`9An3cZCeyWeTrb=5GB$c?F-LVX?$KglnD;y=0jZ~uV}U%iy$
z$10p24RPZKKEmpaTRHwcZ)JRJl>WgXyw8#~F&Qx2AgIyl@2Jy@tC>ozxjDn(+>LD#
zvKRy2;k|Op{ad*HrrW45T%7T@B8ztkM=hGqxz9Q)#I(>jQ6C_IQmB_{*TpE$svgUI
zxT9uPEug0&LYg?D)&viH>ub2F7E7Ld7MyS#KFZ)oT58y7Xl_}Xxtk}d+;HQ~L^Vt5
zT`;4}<xvgO^%7@z-U`@cZoJTzRL#9o(j1ah-kr6_gI)}P6cY-2T*|)f+n9gUGO#v-
zN_7-BbI05Pz#>&$Eg~Dce=t+Y`-HhQdxR>7JQ2#tPeF<&#&O@buVQ{>fXEq|>gZo{
z92~O(wH8ytG8PZEZ@cE=J@?*o8^@h;8csp1%^WH5_?Ez~?d-heCk)vdzS#ggW3479
zzUVwyvMeh+Z1y_c;1Vgp1dd3?T}o24q!(=(Df_;E1IfmX^f|@18km}8ze_po<u8E+
z3&Gen2)n1#^us!>S*0E_<NxGtr)`N4?|m>S6I2Xp;*4k(Mfi%vM>U!Y7Q^w!vhv(#
zv1-8rHh$}BZu|6?NN&HE1rsT5_g?&DlksMQW(D?%vT@!JZ~m{3qeBrJ@7ux4UveQ!
zj$X;0J-c|)>8ttDr#{U1_!NDE1K>@@_7IbGa7&4J*_*BH3G(nOt|Okfl{LnL!cJcJ
zL{UT(#Ta8ql7yt0pe_`n>P!3<?U-&sP$0=S-EL(U(MQNZ?NI+FvkoUZo94ukEOMlT
zwU$h}*c1Fk5~Wt}r&)!Qe*Zn3{;qehPMe6WaAKv--Y;Ly)o=R^22^N{kD+A~H!-A|
zaa4RRW-zUw?freJm)lE#Ywnogx{8#pse<~1s1ozBf4LYt&`%UuT-wCgl>G-r`R(8M
zbz*VY0QAfG3vGAkK(_7R3OznFkvij1)^I3_2=DQYCak-Yky~z|I<S!AfAz)qdX+NI
zqlgiKH<&?^JT{K`&b2oXS1TZ{eFv6dK8GUxDP2l(Z6OY0r08Qf=Zc!5Gh}rZ(*=4V
zOqIQ+&}uL`vX2Ezm*ay(EXZSA%FP{*)1zk%Fjp`<nh}dO9<kjll?<2IYOvu4H`0e-
zEsRV|amtHdfL3dv8&ZV2;)~oE#w)G;duUEJiRLfLHgB&5%qMPnHe7oRc47?YQm{&s
zgyz5yt6uy<v=V3I=wNVGPN)tZ=O4l*P(nSppwMF^*wH6b+;-_#7!dI4kO)%aaPt?i
z^rbJw#ZiD^YB@5_YU$+=XRHT{;Bts-s_oIU<lQTRJfUg37^@1_cx>p(87#h`RL7yd
zVJ|=V-9P8*Klu<tyLYjqYLR4&)HkTj8^BqQRD@O~;=WoR7k%PC(E0Ot;Li12c>aZ)
zcG_wN++NN+c|M>2)CXxydin>#)F?Q2X8X7+?jQ{WR?#4(IZaycq4x!6O59`?-yiZ`
z%hhzCkbxBg)>`691rr%km(psraL%L7XFeOHA%zdgY2j*{C&KwcJnSuW_IKK&?I`{5
z-lap!1vQcpJPmjQXiBwGqg9Dm@zyu;vQPd!cbW;9KS*t2k|kUA@bx$SI<@`dq+=6^
zm5_jyCk>{<^F^m!kLfm2?dQAKcXP)~u`MI;3f>E;!fp56&lT5PN3}ADIt#9aw3HR|
z1~~u4=Mr0k`YfIpF?e-dcYRwzPm~^NSdaM@T<Vl4>IG+@5;^X=^ea@87K@(nBsk?n
zd?ha8dq?(fXd6nVFx7yc+_ny}ZKJlbymx!&p80N+Y11=pLtMQU*U7o|>vzXjEw}jc
zUZgQ$FH)NO_E4?Wp<2%#*nE01_l(RP52s*I)l=ry>j(#_u5MYI&!^aCO5@f$a69+V
z7sYtvn6wp^p8agRi8GyRDXH>=@x6p3$6(lS*Ile$dm@O(nk*5|oX3F42J63ZEyJcp
zVmwpgnUIKuPkb`0J(e_z0;I-gxXdD%JX9FbhqW+;9Gn6Thrj&}Choq6su~(Tp;d|5
z+wh!l?(<>MVyMI!(bbgU3Wq+2!%P3V$H&_HJvy|;wVjM=`<%*hc%qp-4yi}1AZbdv
zcY<hiln4LwKe^<BH_^J`Cmbzx>{NoNQi&t_YBkbSNGmmx!6EKWE4<-9FNWie;=X%#
z@Rm3HHp9!8F}{C<<qHg-{`B87H9m%oJ(_^a`|H8ruGCiSB-M3v1PgY|jGKubWS#XO
z>#no-e&$=I%Sa%oS7l_vTn03ghK!;Zv64+3S}pL3szY&r$A)pd<eKH7cyIfobU(7Q
zEotO1Qq_mb6nTsSS=ni@pp}%kJ_ghnG=<nyNFv3bvX=9|`dMxtR7R_YczlXgJ9cr&
zt6xWL`(9*p43EpYr~$&^)YArXNKZ&oXD_vk+@_<iJ$KC9Z!a{9)FYxqah<>X$Uk7K
zeWYGVTtXB@G$+S-@7v$RP=6H@a!|#0__h?uZBO5p$NoN^4z~1J3q_Qu;hFg1&FsGI
zr}+9XtIvNK3=CvasuG6U<)lC73)sv!^MQ@q86R(9Bq|96lz43i*{6guAgHU{&6{_*
zc!T#ib%i`}VdyqZ7$AonnzolhnzD1tW|pj632~)U+EwSCl(}PeHNZ?0hLh)(?2!so
zv>j|ia%s|pb>I3r3;X+Usi%c9yyiGqwhU}sD7dHyzT~D^NDn3rc5Qx;RmY!*mk6Ir
z#firy4mR9}Tfd$f;EcglYK$as{L5Ydl>tO0>+~C&nap%ciO!hF9R5a7<YYlu!vNkT
z7!vOI`eh8Z5@>2Br|N}dV3=cH`a*oYo_%)7ryviFp*kD?fI>N@cXcrBg(P&FG#K>Q
zKyJ&-$9!R|CjgP9H8FuSMv1oG##QgWi1nZSH=Ym)gJTVhb5#23h>3}99cd}*lu5O0
z>08K4|Mu_T)DzgaeIM_6_aEZwHImj8$<zcFzTyIQ-~40T_U*(IQ<+*v>d8PO8alQC
z3YT}6Jf-p=FMfhqmrNHH)k73wj|v>5BhQwC<B&~Jf@`S}Y!p+e)-aJJX(gnM6ypS=
z*#t|CbdVVHYu{O*%pT~|))eZAZ#u-?_&E17&I{r4cvNx$h9JEfCRrGZ2*g^!)P`v+
zS<Fj5`6=SFpU&3QU`CsqIOVwXjlYiEvksXY$4QzaoU$a;=Z1#9V^uqg0WJRPMcY3z
z;JTl){Ycs|D<hVJNJ`5m-2A{cuK4aPn5Zu(*Lfgf@%%yF@Ty-4`;f`FXJN{hci**i
zU&!O^w(a;{ix8J_n_LPLBi#43FA^n=)#p4f<1WM%j2bYRctu8e5L~$5bHh(#;Km={
zjE$qVphQRf2*pLN(k2Ia!c}|zOmfe(z_sXAKer=JUIZ$}WK9y|m2~_7sZXgbSd2H>
zbjXyvA?A*`V>UIwT*2@t<ux7uo)b?slZa$;g6%il%sgAeM9OH=<oI)+gVw6qINr1s
z5F|HY(O^689IcUE^bHIU&znygOv=<dR21tIHhlX^=BH5e7U!U01+!>Ai=XyXyvdw8
zo%5*oSR-wTNjX#^^AL81l0$S6z?%-X?_}!_Z(_d2NUMpB3=?U}s^>ftj$MN{nevdy
z(_1MqjgLq%*IhGo4}E*3!aWqlwVg0pV^OCi$wG%88>cch%Ke}CH@^JVx3O^Z4i+~Z
zZhV5MZfL2?pi2^wI>AXqGm2R6EvqlMfZ_9=&i?U~H(u~t>>Y0q8%x!AuDR|C&N%fX
ze)`R;xcyU~W%H$%ao~I3BiglxsM!FQptuZo98_qgjSsXFH`j^xpDsJvagVDM`uUXL
zO`WiWNAh7Sy<td*GFM?w6vb34bz&RQnrhOVY+}913J2%N;k+tZ+H`y0C_Om?p}5Mv
zw0Ir5QrN?2<v)(oTxqYbnD)TRbPH%^Qnz?ZrQSztUY%$B^<S|1&97$t!~_GT&WQ+L
zxab|&TkayBn4;mDZHj~mF7Q&enWKw=BGaaz+EGl*T-Y9&NdB|evdSg@wY)gaQHd*j
z@FO3iI)5qNSX#*xHu8)d*vB9I=G*CumDm_mU18Ejse4{_U)MJ0_c1SgyNgCCxgvX2
zOtXpJbR*+;-o@~W<5>3c7vbwQf?P#rj8>Y)W>rGT^Cz-=``gzMMYT-%tCtk9NXK0}
z{rq(1Jt>!Q4aD?TD+!Yw>EJOGB}(q79?TZiXL9Siw`^wq(xaeS#fQ(X-nR+bbI08A
zPznZBT>}ixt=EwW)iT3R@<8ERDSH1sL}O#b>hW%h(Inxh=ROM`nF4FxR(ugW0go+;
zi6YkDy^b|&PXH5sKlK<glQ$+uS%3ZY^w~H&AKzedYKr5Yb`}f|5`_6P{IFQW=TdRA
zvDJAPq}Uwq?^CRUbyr=DotQ+^7Pew(kTO-R@`P8M548$5$Y46BC1n>=+O!^lHe8Q#
zq4WPc)0P4sAJi?p_dx>GkxsNQ6O-_h8@b}`Z)3~nzR0P}qkcfK5@Y%+GyrNAZv-^O
ztHW0#wpz>bSNtkZc-t$%c;5eef6jP1#ZWcEHTH4Ub(f&lG4GUBocD*n#nWGY9@3m-
z`;Tv8{g=MX{%>7_xqmZeB%3%m1ZOg%pIlBehx6neL!$=c4H~8mPQXd`N@UXtCxSEK
z_l)8tQ|QS3D;}NgsB~qd4P&uE0m)mCD56^LCypW-Q<F5OCbN;fPjRFr;&)a66l7kJ
zg8wXCn{y|zvou?fZfD!sK1YWO)OwUNuMQ%(c9`C&25Tx4)<QtVVH{K{RgyZaedn8a
z#{Yc}_eqnq+2oYkAYXa!AF%Jr@6hL<<y(PF9TFuEqMgp?^e74**|qu^PN<|4Vy7XF
z<{JGt#jEf?KfZ<k`N55Nsgk0I^@vkWT)B`}yyP4z*5bV&*pfd+_SM~~7JYn88uEL^
z=Pu(Wo76jTNaFyvUG`N-l{M$R0FGLYMp1zLmU<{*x@G|J4r?r^GMOgaeCs-Fwb~Z&
zC^SN}=OaB+PEh7pl_%)NNWo)Fohx8q-HAqS_^HAD*`OH=wlT%{$O!Y6EI}h1rVG-e
zsWErV9mO#iYJrZUm#pfEb8GeR_tt|Ei3){M^+d*V$CX#Gus%TAOh}BUvT_xyIR<Ae
zHt!m!76GZ~c6bA>#qM3(IqA%&;=CeJl<n6zOj8-V^)__ZZmOn6b7GvR9?@udPJQVM
zp<08Ss8zTex25Ti{6^Z={Vp60Qbi`lx$nwvvs5CyZ{h_SCSuWPr(;iAOB&|^JdcbT
zNUWAFh1D4sy<2i}W(mr+7NDri;7Y1J=L@Qtp0}vau7MF~CCK;`(fBBLUHl0)e*0>c
zr;bVkaO2ebs*HOF4onnb)qs~w1+P+Ni$L|+Pv`0H`%RKo!ux*nFZtGWx3F;ON=8RF
za?Q10WdGzSgTq63z*$3d>Y1!O?sV9^g`GG3h&{Ldn2|evLS@-17C-q^DyvtcLwyk2
zOk6aV6%809E$)TvzVM<!lvE42Tjzqb2jV{LdRFr(jV~!-J^Bw}+bA$}{>5YBh?=z|
zX~OvEC~-BSTCZh9Xo9Jowxxu#^(b_`y1Y0jIfQqbEzSNDKUC85W8WPVZFiKrPv45P
zN##z7qZ%!lV$nqxaK^krzWsr}<Vg#bv9{6R`+xKoJZ0a0j(Y8@Xr_}?Ykft6=erbr
zv@|!6vYE&ndvp3-kvD&|D+=cDGIe-U<1at_F|4iO)RQ_#t*@VCazDTSp5LOcW-#Th
zxpuyu(w35*9f<Fdzi+iPMaU-f?dR8rgg$Px*mUVP7}>aiC8s=r{^y>9uT(Ln<X_O?
zfSygwy!TiWktT5Oy&KuHZ-n7xOYu#gDGhc|t+Z6I>A^p`mgA}EC^+)v_UcKTBQEh7
zDZ$LjB<55dk@HOL-A(_{ATlsiLLg{Rh|^g}E6*LjpvQ^8hPi^_Q7jgEYlUYW%M1(J
z;=v!@$Z-gfcy^3UaKh`KhpP<~n*Pw1I#tQ_owH=_z>e)KSh5uAeYm0lm`M{Fm+-)q
z-(*M>aS3s381)`|%o=#&8MLC9$k<F<+GT-NO@?_lVh_Xm@z5K9=0#8V-sGK_G^Kg-
z4OF&lqUJp61&u3=JIB*s`r^z`E)+4Uf&@s7ghV1&uaFs9H#6ii(>FDt%WI;S`Iwew
zLp|brPU4nK9;r1sP9=44>rGto*MGy{j@_)*7(dyhQi+hbLJP&(2rm$0i5m&1#Z@e$
z^*Wq(D(C#|p8*kH``Yzf|I>9WS$QJcx31^wmt0I-16!*i<f>CKT8r@&(Ej->J?=yn
zZoZ%0H{HPgP4_YW^?Mjxu$1}7pG5t{Q(*CYe630vNkIg5c)a9PVz6GY#bUU8+eJn5
z!F|S(5JAgBOKs0z^s#6+cE}XAr$O1oK{2*UUsR>lm}GR%DE<BY*h(ckapSXzTiED*
zVeYL3XQf*Ov5dpW%<;O5s>gTddsaRUGx7hWv!U94txz<W%s8rI>o{$&;3Y5S74w$x
z)!+LQPK|4<bye1V?2~M{e;sFi;7^&F9HZJdh^Q&L;@Lz|+v{z#(9=g1nX4Fj#HP|T
z0nOGkXD0g=NW?d<y@8u<zmM9&B{=V?)FM)aH7l3$t1mn^^D9wPqL#NQ)8+er=4MQ{
zhsL8pNY9!aSvXR}=TO!#1#2{s4ePo0n_nlc4spusUJvsIGKRBu1rXJ?@nfE7d&P_8
zs_%S{YJZKyIgGI-=jHAcT)GDb2YCQXya&}xeaU4$GtPUQS7Ott45m4RRNF*rD*!1w
zwr=9+wI>8vukGMw>D(hTcO1N6xT5?Iy&y>^%&pTSS1go(E8skA*o2H6pyIN$YupJd
zo_h}7L|KQfY}Q7bT$ptV3|rUV$1%s9h^i7tCNp<ak8=+8?P1rAH!y4~q|T9wGU+YH
zKKmRP8idGZvYRTIykTgo7$X6@c?83HM-UDW4|R~Xxc3{Ev&ci7Ixs>S8=_@PS@6`S
z;;h9Oli@9M@^DF+D2vTA1Lb>92o9ZlbxMqC?@O46>)fzNGnZjfL$ZI0%J>Ag{r%tb
z)!+F&mh2v3e#;@PCZ;MhQBr|44Me##81tzi5n;4mW5@ENIPY)%jFx(?y7DIe@ckd5
zHnf5r8#i+C#ed1M$F0CtBPvt_;Z6&`9}$oWv<523$tQBuYhKIA7rvE6C!WIC&RuN%
z?sr-DpPyp)*S?L}u$ju#1mc@0tx%MO@pgck<XyVy`uF)7nUcCv7kKN@#U^#!XT46g
zXpV~Ft)p7&V`yL=&8Y?xBl{sq5g!=1E;CIkQYC4pOQP)^S2S+~NEcaphVz;6hUz7l
zC^J++k4&Pevnj&5<goJ3EtDrLQsh6PLFp~^SW}@@s}Mi?sa){sf99Un1by@8vp7vj
zzjFiEz5Bh?6QMadiIcn&Dzxd(J~YXtxIc5aW^aY6>FX)K0DDpSX4e3d%o5(9&+n9k
zh!6e4$EnR<h%w;Y6gF1&9~j|}fA>AaqD0ti_03Fjo;qDjOz4;dJ#uAk@B29axMicp
zjJzy}2Q$^+fiHXx-M63BXFUf_ISE&<1?9eQk97)_22#EcltM*f%Qvq2KBgMud{UU$
z(c^Q^N(`c7!XTZm6&#<lCZiPR+5O@gh#Fo6aYAGF9>f@;dGqkaB-!_{g_z?p%pHfW
z1$u!`<_d-*5vb`!k=2s#ks{5ME#Lbg^Xhf*o|ZZ)OBTV(<zS<LMWmgOsA6)Rc;mq}
zXpHS=aKYl>c<V7JSBOYD001BWNkl<Z#(0c}(VK3dc3=#v2CqU2w8XRam!6inCwtoD
z*3cqyxe>6;CdAFgaB6y=#bmQ$e5$Z>2itDDok~F#Z8194<oIVj2ZrZmQxBhw7}Q7x
zk1>!2XHnyG=!?$k3=gT2q~$(ZTTD-TCK@^JB@6J@7+RA}%=i@XT|edOcf6gE%f8OZ
zQ9sp5hm9&oCB{b)R4cR;sM$<e$$M;?B~{I8opn)-U-{?<pf6(k)^XnVfxn<{-hB3K
z+symke-Xd*%oD)0s7RgE`*x}%050pqdQT-3TCQ57b<AppU-??rzUw!b_v~k&gEe-o
zyPbQ!@OgH8{;QbV?#CWzVp4}g@y4JgyC*z8<0c>(GFBth$9Q6#MyOhhT4FCmK06U-
z!{3^j6{a5926`8eOv0o<B*Vo!h{WKH$5%XkLqpX2`q)3Zm#G8$LjmVdqai8MKp4z8
z8mS0gEa9okej_BZ8NBOYno>e<_t%*yx;>;5AIr2?W?1jMpptw7STY3^MFY~JA`!k;
z!JTk4ulVYxxM$GOstMJJl-h0g^X+&17WJ0WoSF!QLhfafO&YQqHn>eoi8Il2U9?vu
z{sovagsE}pe)xPp3D+X6*7?^j{5R`1Z9!}e*Ys4a;wD<W;HgjKytB_FvKHs1NENil
zE+?g#fs{rI@)&-4V<dy~sv;h2FeY`XNTb2jowsn$4L_i^a2cy!^=s&0KOWQNLsT+g
zBQF+0v&+Mt(UjZoT2EA|X8Ti!G2LdKoh}P{=#FD~rA6>Aa18TblG12Eq$?tnCK+kU
zp6#1iyz*$Y5@(17vVtksy?ErNpL560!7;0X;rO|AIs%RK9zQ8G#Z`-d`oJ~UF%VT~
zX_FD(V#V1{g}z!A3ykT|^*5LxMv|Gb9oVsx1xuDdtq&E07!zWNV3QW>ue^eVwuUw{
zv$iCHfmJJs*B%FAFh+tsd@!NO<Vo5_aeAl>L1)pM?6g(NyFVC+nhEz@aTWE{6RDE;
z1S%D#YJIHvl^5YdAZ~X<e6b#;!&FUYgiq-~`nEblOD^B^%9<CHNv+xr&Ef;TT|I8^
zB$e?=?*8~caoKPE4hy&JU`c97C&n>WXgZHqaA`IHQUrD2Qzf!7sY23M<DSIx@;`e&
zEL+CdsOO#U{9{^`1$cFwea=a|<)W9-(g|u&jg+k8RhY4P-*yHKL}bm^OcWOxSbh|X
z&iiFfdD9y?>X}c4YL&6|8@TNYUu4@?E=TUXhsuHdL~atHl@ZPfT-Ll(mroo55zwfH
z1iRf|_hkHs9z&Ujk(3^OR-0c%6B5lF9wo*EBP5eK4NHotSmrHQf-#oeJ2umrn#lMV
zE``!sXZv<%KOGHbNFiw`5{OxquZQ9>=qx7`k5YisnG2b+hJ1n4A|&{smh!6LEL!g;
zS-yzpfASyMxU7%XK#ig11PkxFhb!Oo4*DlO$>bPnJW-Hs)uKzTJ*4F25QWyyUq^MO
zHjU=)U7@LBgVzNPoO*UoPVwQ7{WJA}d8FPUR*5A>8dH4meZNE1dMqY`zlNf0W@O>x
ztN<&Vl1(AQEl!YD6WP0q?_Yc|sNt-OUI!}|liHYIT3xumPn!~i4u5WM^wZn!pp~R}
zH8`KCM@fLk8Ii+jQ*}&(x}+7_{IZ~I<eV=#uD5S~DM13gcQ7$V+G^3iXc;b_O89JY
zk}Jc>T&*y7%zPYw(1Jl#`+!w*>vUxH(r&lhB6$sRj*RW0x#>Zw9!w<co1Eg<v!8?7
zxKkTB$5E><u#{FJHgC9(6|0XAI{QBBL}1WVVc#zH+;Jzd8X_y8E%r~fIN^EEhk=2N
z?40`=<oNTPq^PDd89yXP;m`(exqgfZiKkB;OpUVnJJ-;UV$@NIEtAf(@X2SuF{|)#
zgfX`0q9`UiH?;u*;S0}zLq)!_GQRY#6WT?MWD}XZ6DQK1F=T!Vp{M2nYiKnlVPX=y
zeFIm$_uXu}^b*z@OEi(7#z8eAtq2lX5KF~IL?TqgV|_}+SiC}7t+2k4@T6aVJp)gF
zGA&|0_`!eV?ky9vn*FRjVI}|aPk+utYl3=QC-sU^lWC^s*wtJMUnS$AU@>B`7@`!a
zUa-!hQB1OAF~cu>9w)r(H(2s3FQ<CUDt6wzj=Mki|Jd=_&%uvw$LtwF8x2wtoYkz`
zZXBKzl{A}rd5`sZchLpvbBZBzwLT2*vq$7f45;P8nb?cqL~+KWkwG)T6Re4{>6Iw)
z-~bC3uV8Y#!OqQFF|8EiL46XS85x%*H%7|ch^3fE`5rKYnP5uNZo~KdVuoi>A4gHI
z{0K>D#oPV-EWgh@GK^RZ5UC-q4>PrJ5x@5TKF`>($1x&?{=~BQ!L5Atb+4k<YS3zo
zk>=1N8=x5~L7+hiw|ox|s%|ji<Dlqz6q;0;Z!|)u1hoii8ek*-?_YeB311`iCM=;a
z*-UxM>wk??)+|SSga)}YMSDn1Q<}Z@f*$E(w+9C}C1I+f2wqy)q{*(YUP^1zCf1z!
zOloI8i=<Wy{DZuqm3Q~suazh3#@a0Y$CfLu{vYc7wM?}zD6-{dtXkp{Je<2s$%Hg=
zGP6?+-Z8xjq`Z;giA=(dE$dml{Ahd}XC`hc7%auZ5}HcW_s%&|!`uPDs!$WmIAK^j
zw?+@W!y{dN&0fiN9$0;v(75e3s)=F~q2Ur5^?oW(IRhsK@mj*s2i(8prYr`MCKKbM
z#DnufvSC8PsF)V)yzxgwW8*|3BuRsoLZddw(&zs&u3AOC7Q+k4$<@KMEr&t1or-mq
zVftVTBt@rN0N+aC-gWpbTNqHmr75YvgcnYH?zvE}1vruqB=5XHXs@s-bzx=-a=p@o
zL)&GQk|1B2Dg+fI1KD|r(i)qf(o(j5<r2Q~&UZ7oelv^G1f3isu9l3T)u1knR|M3#
zY%^0s>Xn8u?5{?c(@*C3x4j-J5tn`GTE2GG4;h}nl<|oJ{L|n44bA2x{gryQk>rV3
zj_wYfEp6-i73c3YbK&*sGjy&u7&Ukil7Tt{PdS|v-|%|Qc+G1VUU@8I+jeo=7r)HL
z&wdGa>m5|~AHXCNpe>AKDo~<fihzI($|7ROd+%i=k{+%dLC@dS!+s*OJ2)8Wi3S9G
zPSEZoB5F8)BPJ`%tzq7x#Vnk^l#!i#8QZ-BqhMTHpWO#7^P<=mnFcGEdAe{o*sNYG
z9O2;n2YK}4V9qHh&0i!mCfcVa1(94eTgzH~fM$Qp^ZxZCxF<b<eHBZecO18Gg0H>i
z&D2LGaZ}?Ynub_tajGTjHld5|1s&&bsC!5slQnxZGH5MI<{;Oc7_PhF7QXWJZ&M#!
zjEy4_-@w?E!K&x~{I7RlohLE|#Fu<Z%I&YtdreCv!=r8-+(TJgOvXVH$zrzC6eb%a
zKfaYazVS^4hL>^j>o0<VL82%M0n(Se8glqs-i;T$&!A4?AVvB1b>Anh_u(?P!K~1b
z(*Cax5tk*Mb%5ragW#>uIRMf29j?J1G*y@!XMF4c{Y#f+GTkK}ixhrAzRM8tp#<r<
z<LCTXd$5AxxViN?Qj>*Vlvyn>@{u*%`@<hHSgj&HVIpm?@QJ6v;BXNU6rDb$2n52B
z?b{fhzZBwXVQZ9g10+dUf9*94M3&U0h=GP{vFNl@VCgc9wb)>LPf0#bRESit;T%N{
zB@jEy_0cXx2b9PP4}AM7<{Oi_QHvn~%z}9=I_nHHj@x+HLOWdA7~ENI?m9#9aIloh
zgp@|SCkoILL#7GuT^LD=B^{rlGP<81zVCh9`A?tVq?R%;-Xu~_wHgQWs!XB7xXk!S
ztl+%jBT(y6lj1~4YQp}b7V*Nr`4GM;+_CNf{`5oth}0L*n%c+5Kl;ZE_Zw=pN~VXd
zCYbJo4M5rup}d>u)njaS4IM7Sk6RPye2RD__Ch5kgDVY0xF?>-^0!>b+6ymY_|zw`
zfBQD>zU0en{`7z0Z}=&8&p2k%WouzAY6UMD#6g2xA4sN3;teKDKW3ch9QNQ!AL0|5
zcmF*GhY4&9=_dDw%+`dY!&}AFViv48nmDSmYr_W8iBXIi)H^Uf6lBG!hAE)*u98%y
zZO*ymYv&wHdlU-2W@C@P&Ak^XmiBEK=ArElct%9^RBQbtRpHtH^tV*cJ&Wyr3eymd
z+27zBZ+R!R(I&~%B!!QSk)jwEAN09YyDZygTLk?4B};kon>7*)7$tQn6Rn6p_>;dP
zs@8Dc;ap0!VrWf{@zK9{KUHzWHiPTsjUFu#`a6rTvd8dr*4LxC0e2LNDghj4aX{(?
zvv)T)eCl6O49|G$TVc&ge8r;97t>ivQoy;u2>J7^S%b%`@WAFRJh){CY9o>)C5mGd
zOX^&qa&%Y|7U_KpN{9KIV5}`9oaxvLGdz_lSnt@iWg|<DIR@fNHa*eIJk&|X$?_(}
zNofvRgc@OP!Z3HtR1<U;49Cu`&ygt`w5KgfE>EG5q-^`~%?#QIBc6s=j(^TMXcQMC
zRPAb1h!{K;Bf{qU?_tF;t8w08Oehur?+onO!~UP%NuP;H)YFPAlf<lj?m1Aep{C@W
zT8v$N*jzfh_(ybgf+3SSU}B0bKfH-SEWXu3E5d|2Rz346uxKGFHaLoQtLyZpAI~NY
z=yracadx8ZB~VNT9}ArvV-?>>u;b&zcihV5Z+RQs^b=M(MaD*{Ry<A}sSkr{lTChd
z4}kpjBP@}LP+K7#9OOY0@q$16eONri$N|r9{^kd0_6_4~m0vpNG@k!UCsL`V2vKI{
zLzm07YGEcPg{@tX<s&2aN+>3*2@@NW;d{LZ>N9AQx0bZOh919)Rj+w9PkQr3EIRHu
zCU<V-_DjCVhR=MC#t(1A?%6}sn#8yUCa5YI@`(d@)1GvAY1^7<*#HdfNxGA=_{YP{
zJ39QSRWK@8HQ5GQ_MMpK8XP7NVjvQ{aa4!rv1G+D9N4>uy_+7uq)m)dR8y?v`K#+n
zm$b9^?<xH|y+F{Ouq3+f+}U8GAEnJ-dM7~A<A=G*nMhvjX7?aoiL3pzD)8hF{sHzm
zXR;$w>PgC)?R&ZOwHMJpGKHHM$BPToG%4UW8VoUX+O)8<aOf>e@v~DX%=(Ik1Ty<p
z(3T^LYy8V6|BHJzZ=(`dNt=^s<Y*<zv(9`X=l#-C>5DDFF>jg~Tld~JeafJZ$!cmx
z-WyW>a4(35+JRBlfAZf+Hb2O!XPry@oU=&=Vm!8FK%||YQ95I+jcia_GJN;?ACWkZ
zL{jV%<fP&tsVav%=u_$8Of5yzEq|a==iBIjT%J6{Qtv#ZO~ywLFtl_f&PYf85d|JY
zJF?6<Fw7kXJ&v8RU{KY$g5d~;a}AJ<PC53>iCMR9B-y<S>lEiblZmqYsZRl0Eo3@F
zTG-Ja6@a8kb9{_}dGiA>CFgf3qzO}Z{Dk`0D6wXN%ptf+m4#0~3m?_mNo8p>5HSJz
zV@k$qdf1<NdSADlI23ZIt%T8ax6|0W2k{OQ8B#psSk|8Re5h7KXI!RrUZn62n%%kS
z>>4?=x0?1i#A`OjRkfKKj)BH_3p>?d$AAA9m;dhnX4S|T!&6PHYY|tn4MnWUNM@M`
zED{l5qe?0XHdaikn3kX}W{<0~=KKqYo_rcDgunjV&#+@k8C<jsZ#^IR@O$yTft4tG
zeJxstJ(9|q=3Kp6a!OWhhA)NKTdd3OgQTxU>x8u|eZyN>`{sAC=!s8ZY{xb>eDe}E
zefpEMuDK4ga}Q>+iSsEwqfBIcnp{<>kQ>#un$@)YSb8_%kC+fp)@jU&1&i?ptC@g+
z(QFFRPK2agEEMWwnLDgTmNc%gaK*6<4=rZnJ@+!UV+WC1NF1I(b__*#xz?4O6rM+f
zvUFqM;ABN*roeP?f~n4;6DZ7@Y<VR6B}}i-W!jAxRQy#68A#FWc|(;-Wq>Ig^PIo=
z2=ktM4%=ES?1XUQM8cO||61aa10)k;*|U%*YMLwgXxq?ghQdL+`r|*Vt*Jx*Vs<}T
zBczEVRmYw8Z{Y9#_1~!1=Vg!`9D*TsO+NI2Kco`y_dal=+HqF8ig2BFp&tt$jIN$Z
zvCoK?z@-5@KF<DYuVmBp*D-YTTGn3lM*MJp78BaCg<YRZ=r+1DD0q)X!ndxvmP)Ng
zk~kt;2@odV!`%7sFqqv<g&Y=~Ne5>QJ!9G&L+^7(_3V!#Nf_O^nI%UZgO942xv93V
zkyjtGzD(85c!PM$LF7Q@j$h<9cvnr(Jz-cgw?q%8P-w#o+IYrxKU;%^t|ZFl8*XM$
zYNUyyiKDXQC^%|4u57tnQe#n7ykr5sv3EO5mMlTysANZ<Arg%yJb3-JERdMUWr`xs
z3iH;ihNa6;V>-HUIh-Zix7$0s$l(N#Lra>LJ3rQW)_v<L22F)DY2kdzRO+cLUxhs3
z6kHtReWwg(J~1v<^C9j_IV6Ho-&eyXS<*fNkzhTbNF+FQ_{kQv@%?=7{eR4lKmKp5
z!C)sRF}8)*1dqq1u2|E&_{cwtFD3zs*q8=}NgJ_aV2CHb{S7ooxc-`3`O;UvLsVVN
z?yZ~n^r!!ZwAG>>2btY`<M8c8X!rNZj;1}jRb9C9cJh{A_l)D9!4JZ?49DN<v$zwE
zVdX1d&FOD_D+|}2f*%{_{;RIyzJL1+qt{+fyk`X4nnGO@F&>i>1;eJN(rNqDnZ9+H
zUKBs`LYc1okX(k!1#2<QC_AQ?TvCghs;jxLQhHV*leuXR^t1Ye(=ZM;-FgeAF@<s9
zlQgTn!aAx1SXw8Ru5%Ju6v8DnE-C5AYp;@5d(6-dRWv+?(NSjrRpsHX>}c;i_1X|k
zqn!2O_c3(tv)B=Nn4IDXP34QPegm~VW9Z}rDfy(Kw4zd4zplC4!GxdXXJCh!^?Byd
zL^5o?Hx6H|@jHL?$M{MWiG<WQA%=<ZQGWk_y^Z5mE-fZyB0l3U$n+_Uc68!9Y0r=7
z{ifx%PFqMrivm$gk9W2o`iV){@Bp`b=5y4CmT>0Ve;vPc9^OP4OxBSPF*Wm@(NdqP
z9bdikf&ZoKJJ8@qH{VGdxMYeJlB*dh+8_SCq@_YE_s%F)+&TQzrJgiRJ77t9;qARb
zbCSK=H#4wgS!TW&R1n*I2k9u}McUpCIw!F@cT7Ll9IRkiJ-0p&ry2{vEDu1T>!1ah
zSipHCN!akc@3L^9jumCHm9XTrlhIlgH5pdk<QtqyhO-XDWGj|k4{l=V%2jxi1t)b`
zlB!-{YK+ad+)BmdorRRqi3Z0%=WM7{GuKF+*3|6?_0lsTd8nJ57B=)kmQXP<#;zN0
zppFtD#Kth`9LJpfG#D5}#AFkXBAIKShsfc>RAweNmy!u7iDZVo$9pn`iljx{N{R2i
zoh#n?4&=rkalGw^(I!zP>uhRdabi)Eb*{zxOc4q#HeiDzHknFMD@|!tt2`j^hKv6c
zttT`ND1ZJZA0dkRnApFEw_o@YPG8keU+lq4k#uT_+1VlM+2fPS3|-bdQ)`dBGYEnj
z!abshcT`fSr5>wFt3M(+VHHc>{CZZu`9cOyKb6M#0k(hd8Xoxfi<!FSdZIle*r{el
z1hg5?qS(+w#i!DysppAZF2SnO_De{`hcAaOKy<UR7wf#AOWC(d>8q;?-6lMi)9`iy
z>BBX}D83Sr3iK~MiervHfi3sn!=6p|VMS5bLd69kf$Xgv`Sd^tUWpm$H+*JF(zW?q
z(VKiddOxMJnexqc8u-X(yHYH0k*H82<V*J9bLgTvPqj8kB91fv@=sWB{<&<?3EX(X
zsj1<zH@=m4&jjhz6h0JuT4a=_cuUn%YFVshZ|$nbd#W^R2?{9-SqR=Y0H8o$zwu)D
z*Dqee5ANImaSio`id2Zy@T3z~@s0~Gz&piYaH-E?aEE$Vshy&oQAHkwL^HpB(<=HB
z?niS-kAYMSiX)ypz@7j8Z@B#<9RH%1A!nRHTCHRpz9M0kl5mJ=H&WCGRFWb_Fg7b{
zum9nXm}sP+2KAEhw2B13q+;}t_9^Lz&$<7DhK7#B5n^pNjmy`-1n60ka$x&rmMlL8
zqDrT^TS+)Y#FXG;`q=PJ=8m70WA$9Y@Cdhnq}-82f=p?Ew|Eb&3HGmBM_mnGJp0lX
zCqD07d=v;xhM>Yy!YFF{ao)Xe7qtZoaRDlkX+C>Q;+VMmZrsRzDkh>a)gY=?nTq=u
z{^fIU)oRE2&CCSQ9^2HzT$BLGP?w?)+>PwoO&^)WWm<`8NR{KyIVTHrMy4wnJq+oM
zVhGcgacN5h7D=miszQ0iBDjfhqR|O<U;0(P_S?V9k{u(=Yb0oM5?fKy)P>|TljAeS
zX1#vNb?1xBFjF;>2uDO(uk)Z+p8gweg*8XvqX9nl_n$!q7f`Rq^u-D9f6wb79Ve1F
zn>?uxypA@Hqf@-{V76afF1W=ID<rB?axM<<V{B9igo-y<FQ|z~Yc<@lt2z3$ui><d
zUQd0+O49u!+<WEMS^wG3)4btE;*o6PfNx=PeRFNY^12O&pu}C!oM72`FX~~Wc)i`Z
zyWm?Q({ktz6C`by4C*C=0{EbKrGlwM_(<?^g_S3tLS!S>|MX^bY#&jYf=@B(vALmG
zUbqN_iBfTTKG$$BIfwTI&-8|a%uaGu9)q5G*XSr8DQNjRbZXrfGO1pPs{=HRa@wE#
zF4gm%!4B(*n^PP=GRoI3d<%UC8YGirc=6c0pbi~g)z0Qe)^wMS0!;YX@qIp&1VwQ-
zXo(2x!KV@TKe&xQ`_RXz4KJnPKqH7NmULp2i~s7w*kp>DjWbE)NKkd9xq+61IHnh#
zs)wgi_t0;n&e=jolde?oqk^QK_`n!<U;Ll!x@#RrKjmpGc-hNIYc;H~neYU@;2WrR
zeL5}GBic^pO%!qYH?PLV^~`r6m$J^+$NM8UMGE{gYi--hBtS={m`Vmxwui{K*td5t
z{mWM4jK!GJbYW)W`?+K8IQ)V^RhI+xx%GLtajl)~XfA$fOo6Xd*TQeOgZ@T?*eea+
zWIU=-JM%1@i3)t8hLNf>9v@}F4Ha4w6VwL=p_1wF`#{W8z%*NI`@!`Li9*t%Qi~XO
zDQ4AbIC>4vRzmkfrllU;1%e(j`8Y&<kavwWNKS@;Y;oLs^_48Jf>A>=aZEW+WziC3
z?eTb1$&xbZT+_BF_Tfc4DJ3hSWf|F=3*k!zgtW)DKr%j#O;XnV^T)a2qyNb2DJ2?j
zW|?ms5;`Mfz<t4z2X>D)-~?Qd990!6Q3Y)}oWg{{f@eR2)vr6BgqZLD=q^6})$7=_
zTS!Ft@)!Su{o_;A<0{U9#kM;}#iXJ49HG?vn+Cq3W9+RxN(yP~2^qy(12%ha<DudV
zkuwOEq*kYO$|)T48}H)Sx4ny@)1Sh=&0Bfkvahi9lm7uX{unc|4{c77;xZTUa@RNK
z8hS8ZF<uc(5E7t;RY%2JVjp3>4Q731aEToJIV4xSdWajgl#)qlcW(*a9I7c|6<@Wa
zam3KFRjfJgG&bC`p56D~k8vK71TQZ8{Q#fIi)Z7>%%ps&yC^w#LP}7A^3&R@&)ka$
zJriu{U`56wkw$3Gx6-!kLR&sMWb$I9gREYy4lqT;bN=chEPC;a*qpYAoMqM61Ydc>
ztEsh`v>M~MoG&9;b0&)$VR9xOCz%jI+wZLmYI*zvlMYoThapV{TSS3`)U{|(<?ZkN
zBNB;FHJGS^#FnxBBmDIr{U6q>m`5$HkfyndwiXkHV!~ET&C74t?NEOB=X^LHzRc{C
zpmLU*c}nhLt>Ez3W`nUSuHeBNt|K~XB`3Z6UAXyug}G@~oSKrsVsL4v-ObP6#W@!~
zr{_RxlIwqPGbXBJCT^vqy{xFL55eksdO06pNZyAAlv0Td6(g8jVN9~KtVo0Z&)$2-
zTUM3#|L<q5z4tk{O#^0zVP+^pH_}1G1dS$YG{*dr_>GBvtzTm_MgpQHYSet=H^xN8
z5?inZv138S28vyzh#*KA>d<Cj`YosIz1H*lW9@d%J?Bp04l&8yzDAhg&b{ZXv-euh
z^LajH+r|wXaL}P-Q4&I5eU-F6YqwMu<jxD%?pz~zZ!l}Zun^cb56wckUF4^d$<f=Q
zW00)NE#LhCOPdqW$?&QyIeZ-~UWzw14_~9qR47m}3|N6}8}DP~L2FQ-5=B~;1(^cf
zl)HcMLl#8|&Szwr(Q&Z$)Kl=yM%ZAy4nbjc7LK`Bhi9LKW|^Gn0yT-W&4wTSl%<j&
z?GBB^(pF{lDJQ`AI6f>`-<oPQVUWFhIprp73Pi@2h-QIm8Ywa|6hGA_-nNzRpZyvh
z_}Vu)ga|v+CXPI+E)=~!rz5G)$GpL_9FyB}<|0lpp-OSxG+5k?j&i>WXTI{q&=jVo
z;B~Kk2aWv>L`KK?!)HE;gAN*Hd@M$^m>?BC;WEq4aM-6NJuL3Io4>-zsl0=Wgy^tZ
z`2lDuOH3a$4q`0b#3Dx?%AwDG4yXLZOBr8tI6Lp(#4j)U9CuytY4paMiKix!Yz8F_
z;(YJHYb7o=h1fon_yr}ecoAL-YsWg`w|fWge3ffjn@UJ7xb?YVfS3riK(oo=k9rJ^
zc#K=Fz5?maU{i;t8OA7Hvv3dQO>`(=O!aB0^!R{@3_<IMx+o3$nqEA8*Le-HFE+=j
z<5K!szN#rx>AhoB37Uq+7+sVT-|#AyKJ|Bbz(U+jS+~8#H(&H}MrOLC?P;<CH8v*D
z;s}&bt#h~azmr2upF5vuiLz_AOVnuazK?#A8}7P`iLr6Sr!*HOOigz=^Q1@dtY<!%
zQELMrVLKQPaR2}y07*naRKgx;sKVMR=s^Yyc^{e8tKSm^08nNm<mW?Z;KM#;X1eh6
zUvTZGK7o2T<M}U!0}sTEG!V75$*R_-h?V|9X0iSP+T=3js%x%e)8?(1ILhr<={Z-6
zy;?-}EJsHArbN`@PIGOAdWB3&x4Qz>@XkYfhV2_SGQR(ckh@SgzzzG9WM133h3mif
zb?~5uVa>w2Jp9eOjt&$7Fh&FS*?Wh?5qJIU=QI%1J9bR&;E3Nkoh-H((b{-a@?=IS
zsVAba<Dmywyx#!`5ypd2#gK&pY~x+{`|l_5!TrXo(&@H2<h0|EI0@Wn=`(JWL520h
z-$06u+(_j+<GUH%^|zC4*@jggrYOd-z0=|F$Nwfo7F#6e_HVL#iPj7&{fw!oQtCA%
zU<zbTv8^`t=Ii<PIcJgl>Q}7D1l#HmMS|i(q9-BXOUmFJUy`IH51K3e{=Ls91Qwqp
z+}kxg>bcK>!`Fg|`P^qO<-r{-CSt>4nc@v+KbwYW=PGHNYfh?3QXLVM0n-KTyBl}Q
z?<&9Ne2H2Wd}S>Klp*E0n)Vn~Vh2eEkq6f_q(>dX+LynSqhI_NL~9Oa^PP8a+m}Ae
zrq6r^bIYCBsTP`M;ZVzO4yTG!4OXn)fR7O$Bi^DGyzyb$VYH9S-2Vo&kEFF{USGQf
z258L@Rt2LG_>co&jfDiVNLjRcEk_>vXl}XgT6Wxj7ZHMHE>LYyoOfmOrBXw?rE2m_
zVW=ei`V(O2C#&o|jc{KV5ET`yEY@q96_;n`x&r4dlNrNanKVbqBH_e;_<I&T^*7n1
zEu^I^yYC?`KkL<uq>y%7RXu2OUb|ZGU=gh5<_@BBVB)fi9p#a4GltI7q&Ep%^6){Y
z(1|QR`o(qp)4zR?(b4_z&J)GpyN;zz&->r;7MeyuZJ5sFfPmKA*F-9PNRQmU2T;&`
z`_5}aDo+5=OyTOM<a@b<6!$im4G;0te}5mH9XmMnS$~E-?G*e-60&2|<GsTeDKk@g
z^LcWhu9nmuSrg7B8*|Z@FCiXjk_B&U#D;?2YY;_OR<<(W1Cu>?2CLs2YjZHc(`Y25
zX_|vK`RjAaL-*Xp$~9}yD9+EP1;f;ZYu>IkgBpgl3+pmpFq)J}vMM3yMF5<K&71HK
zJw!vnTV=)zs~>Y3K98m^xtgjyhiaIJdNQA4M;4Ri>5@hy(3R4ZEm!@F#oiF9&{a=6
zGU$@UG><w8HMT_82ceO@2hzh0F{;?ML9`W6uPDv9?Xt@mGZyuZ)H^aYWMdP=CqJ6Z
z+L~be;T_I{$rw88ASx#qqUcOYJUzwKrQhUh=f0Nx?|+cR?F>m<0UbtBm-C<tluHaM
zx$~6K9^A6{olz5-1rsr4EK>&^#IY}Z4js+d^gx^Q-u)?-?6(@<+Qf%H^m@e25E~Qp
z5ULn!O5v^yT;;*Vh3p0nJ-9HD*>_3F?$FZnTa77`o_zriM-kabf*f-=$Gr5V9RI98
zAvK0AH~fO@FZd{%zH$lnt_?&pGgy}*WLQfsY7Uwluem6hfZX|_P@m(47gWIBz_j@v
zyO*DAf1;}?4s}D+i1Cq-HAXn<)W@)M+YWBK<|ml$45pK!KE(<-=RWs=GS!r!<e^KA
ztUU1fs{i{i*CgG|J!{|1W@f+i3bLrX-3)4aOHe~Xa&OcaAx+@KH=c_-^;mX{#f*0{
z#((}Retgc~GU^Q7_6(THHB9lj1DxhjFm()f_FgAL_p?W$KxUUeeQ<rL6$#Hl>N46I
z^P-oZ%lOg*@Dh=#L#$zD$9DeZ&97zovN2+7LmF~^pNZBT^0YErszXV_UPiY+%;8jd
zw$|>EVp69D+XcJnL2mtzf2Vclog94HZ?fMrexGb20b|Po+nOF>LaxE++e`9)I_JW=
z0PT)&`FF0uL@^#0tiSWK&6vE2DfKdYe!t9y$$F>Zh_Gm4j5vu)I+S`4*J0<j9gHkL
z2$yGP6jO<XmSN$VhimPuhCx+VFRaUa%^9kqAFU-OXDQon_yuDE)_GEw!T1Db-8!&Q
zD98%bNK5iZj#e8wGt-PMS_VbBuJb{FDMBos4L`k-1~Cw$Ug@Nc1CBWsmMue#4I|Bc
zZO5L^bhW-|s^FbMI>UWG`w3&l6B)(EhUv_)|MADe#4@~zz?fM$`NR1nv=596pGh%t
zH^F(rDBbB9OlO7-AO9#ndfPiVwC!n5Wr#S8F?b1jg+M>ak+_0BCE+tDEu2L25FW|B
z{=?gFjRp^@;gA0LTzuldkMh3vUw|1|OsnHL>4d{M?&#IT%?OB44^|A$xjs}~1|p)e
z8wH=tnuXO?xtNZ4sr=0Qn}+&fCSl^BHZ;Znu{T7{qQ=r4A4MMfIDX@A{+2aQeHK1y
z@W4-h$lag%7?WT73f!|1)0)P+ZqD7)kQi-za7;3x5Qw}9$grR7^wi^MB$fW6`uyjw
zppm|Yp`c2Lk)St1^Cl{|szl8MCkd;MIEh0JeiYYy?>o%gw}B`#xK6jEj|3OB{&wA$
zn5t;SMV?*M>m03ezqUaK6u!C-CJZuUQ><G?(r2?<E0MICv{o~t;4XzKaWckCobWsE
zdN(uc4r9A|#@i8%Yj5K6vtCJ4g>HAII^raEF4NHU8ex{b?d)CS*T<KB@TVd@&w|vF
z@|?_IlbF9e_swkUCUl&HvqlYRTlwRsKY>5|-7`@i_9*XblBQB0rG1Icv&{(}>31*n
zji@$iRDsL`J9l#X7cSuLtA4`LBaY_CSNsib@iIaXPCW=s%X8M3nKrfOA^(wu#dI(|
zfNO5JnFk))L6jt+$;eGddK;+Pn&}>ZUIR1*<<n8pXJKr790&#skpU}?`#0Rin!}Dj
zC4fv6P2WPduyD=C6IkmF>Jx^AhGCw{in?%=5_J%43in=dEh9v9yIrQz4oi<d8b(LS
zC&{I_OpKKLN}m@GTeogz>3++@R&H#thEpHZwXI3E+;R&eCZ^l&qGFg<So?&>p`)YW
zl^N+1yXvlBtG(afdvfa)^%*>H5AA#JVFXL3o00-kDXe?k<M45#bO4$^RJeB&542BH
z=|l2$?e1)mq%-{DU;de!KKXf8t3jq(L`iXQ1ji8X!c<4`C2c2c>%Q-hka|8BUOjDu
zt=6#aDNlgI*V4)?_uapPuYc>uESeZ$+vfXt`@g=4ENv4>93~67^`|iw(OyyAV02V>
zS@;bO6sp{6W(5cJXA;yt+FoIC5K6i{4+IsgH~7fXT{_Oh?>?RrUU?3OKjBGCZ`r|v
zKm94c`p^g2dFc;`wr(d*Td1ap@p%Z6;ykFBSyLyiHA-5Q)xo}0z?=Cp1s-~`mmR1U
z`VOsWl?sZn3KCoVh{Y~j#L2()6eb_q!fikNE^#+SIvL&t-AizIuel=$HK(XSAE)kO
zW~6h>oOXA*AP;vF*sTU^_AluznsNuPLd+a;8Wz)S#60!=Z)e-eWu)T`##%GP*Idta
zZ~A{UyPmY$#sQ`juu^jKXRCXLt<m5V@JG^LYWNx#U2DBD&QNFh*r&h9#h3n&XmlxJ
zA<d>4NgRut9p3roH(;`ihK(zYRT+5K_YP&qmi`DfQq|Ab+_*wZfpS1%JkiunCNI5&
zJ1)DF(f!wQ;;UW*%a@{YFiY@xy0PgAj?~iLx5w;5Y7U15Coqcm!dJeIG@G<tH_$Wt
zKo#ni>azt%yZo?qSZkQ;>@w#<+H^R~Q3rl%lGgMz$+G1*WBGN;URk*Q+gu9`!@k?x
zv}c2rB6!UyWEq=(c|BvX!5C$x)8?R)PQoQgHAbxArgegg5yOL<?q|``WjIf-a4S<L
z*WZTj+=;m03X)37OdPZH<dg8$R>jPKBTojPR}XJ{uUX#}R?*J0<*F-LtcvSoAeI!N
zZ7r)#JsFMS@Y?ggS<$g}FKTHJW3$J$yA-wgyb%kwbJ@-=jcwcc!MT6O&dYzu8VuTQ
zlZ;xN^QeH82pmMlmZtHxOaTrw9ixGB9*|MNSw$ir0+7^WM-nENt>CD?coA7*5!2+2
zZ+r(kyIrjA;2&TA63DtV8xhWh&s)UM6OOF?cZTj^kuk@AFJMFCQ3=0_F@x5}3(|Wy
zqIpPR@g>E?B~N!wT+lhBk)d<IQbwQo432%xIgFif0_~Ye?!Ejw+<E@_xGSzE+A;&(
zH0bq=$=!uL-sHtzp6Dx@`-n>=hAuZW%}YbkfmtK-QN}4D$vYFp<;7ONKedS9yykiq
zMH`kZj#+v9$*ex)5U#oG>&WzWqIQN?my6_!m^zaO*Gc}JWq{ZnhSOwZ4-0r5I)~b4
zm2l6!6w<bh76S0z(!3}Tt9udfN-{D*I+F18kAIL2E1I;Luqcb!@~z8R|LzacOch5r
ze|Ab+G(=4uKAZLF!d&?!J&{`btZ^7x_SMbeYPO$Yujtavy4-xn{rv4eyn|)?uRxuH
z%rTxsNNXow{K$uB_%4koB6D@R-|S{dd-`K7wLAYAl&sP&m4fCec^-kmwz{}$ui(lH
zKZ-FCk9)<-;Lt-zlNh4`p%-JSkh74EQ+t|&(A#L=Q>IE}V-TUOo^M|EBP5PdljkxR
z8^Y3R>E$AO3B2jB$E2`pRqzl+F*XW_h%s>Yo$Fb9*im>4V)OL+A|!TU#ISJf!-iqi
z!n(}YgrRucP0j(;u(9meO85Rv#HoUXDPZ*(r{iqY>l#ve<D0w)638-q)`cjd+${*x
z3!5tU{^UwV)qruRainS(S#vO~IS_A5t|RCR+^OwUGG7yhnwy4*RJ`{r<^C(LW{e06
z_{cCV&{(|+R;~)k%BF&}W|2t-{$?(oXx*#3)`H})ka&-5nI^jXcD{Y~*_i8ZX8+V9
z-451zvdjhN9$(8>P({S%|4z^>2(>iSz+=^-k?@L&#U%~y?WUam!WW`Tm*RDl?|tuD
zesSAf9I$2$qTBi3&;DITMg=7*h2n*dr5DkjO()v>!lfU!@3kLNsT#N-O6^TOE|4!%
zx6{|)SW-CE_&i9<ID-{K7ANRI2eI<c{)FFn{&R^A*pHd}?%|G$zR3Nb{1o<<+lXc+
zA?sj_VoiZ6R)KZlbyc;VN`>?p%@2OvH!(zI09cI*Qa<rovn$p-^p8BZ!C9n{;A2a?
zaxEwS_LI2sdzZ28j+=>EU9=PODy-yL<Rbm>V-1DalTx!kvtd`c4TJO$`v@{F+>+||
zoNCW5TA!mK6$M1aqYk3x1lf`ae*Z)7<i<vdA8)c6;r5UG2irdXRhnH#*L5n|h?wBN
zKjgvbnexn<LOe3K`tE2Jr2f3+7Mf|hgBQ#5U;b*EOAeshX)`^w1rw#rOi%Ns*Pg?=
zwaXcABmp=PtpzvE{vm^PZM%Gh`{tJEk@AG8lW`zBX5f}vx%9mEf{S^=D_#!A9fKQ<
zu-4}KmAnyD?VFZp4M6H`Ahi$(09h(rd)+PEbMFJhBO@geSbXm)yly_GNOj_4Zp>Z-
z`;sJ_pMo;IWg}x_qr~G&a8WJmN&O=drnhkYkGWPYGz|M<m#^>WT4PhIqP|P_wi_5x
z#fp&ONk=E(&~>4pk_s|Xg6o3j$+Sk4nW;&ZF5N%(DWwb*@Csxd9{9!2X~Yqk7$iz)
zdtv2ir=ZOV1i{*xb>mPhZ9a;roX=iF6y&yuw5Qm3^DQJMPu3KeZl|m{`6OsIg0Qzp
z$Q&f9lew2&sSBGc|HP295Rm8K0%XX%gPCc#^*Szj^*JoLZzIbxk9IR83g~u`oni7K
z#o?+XVg(Tk1ktZULQ@tgsis(CXn2cF1@)FGVr1Q6EPdMJNMo3uhIhRCW27?5hKIKD
z5AS>nneSr6hDnrEpUs}-)yZyul?-XTO1sHgSYTelhT=_Sz-*%mE-FX)#SgItHG&l(
za!TUCn}kkd98NlwBhNaAHP8A}((xrcc;i*vcHu|qT=I47J@+D0ZBn1&b0J7bO|Pa-
zh9bYggE){3Ocshssz523mpeN@!llF#12S0(3#D%zYg}nnjLF&f7W`;JXLOv?e(w+2
zy7@tV`TcKT+dE*Ujq_Q_Xe(6=>T3m$%QX)S(P!u`r*V5ed64<q9INlEmGzD0_j#QT
z(k}p&`a_;;i9!!yh(;!8A9xV|`=b|dvj}Y?tQ%R%kN@?Z?EK1CX=WLjafNO=Ofhtr
zyOa*Fl-8l8^S)|Ihpl0c0go?$ql|9W!8Rv&!7JXtEf4I(yBN(<nxh6W@Ow{qEPwW_
zKO(VlPLnLwP=`WZ!&}Tpt`Mzzo)wPZVkA%>(-vmiMt=Oh_uyta9RI)m6#KZ}CLJFO
zTz;=rYD-~(ErJs%Nu(iRp3<dDjg-y0$&E(5HGKM_FB6XhhazKah~B9IB*T|v*TgdF
zAPkiqw-G_T$GMEi3dSh0$#)jlX5;;LvwZa-WF~i0bj74eW``p#TnpEHH4LhHAYd2P
zW&U<~DTs`@!GXy#9=PT@Mo4h2l#X*OJ@QBxX;LH$7Xs<Jm1^O<GP!LV%l2P^^R7&L
zM0_wJ+`g6W?RPRpL^o5~>S!Cw+S5+O#|h;0hyn>I#{BbTp*n1fFHvjfXx(`?ZtG4O
zDrDUhFNSIFS@-xe@VV=Q=I=kC*Sbe0bkZ|TsKje!0Ff#giuayqXNQ>|{(!H%?scr#
zF~evlMKr@k2A9=RVe<mfmq?uCE-Svq;V(Wfg-l9eWkBO2Qj>7&l;d~KemNbJ5;w=W
z_>%AQ;PecOmW;CQkOO$k3C9qN#SjG8^@gX=Q4g8|&6cm~o_<0D3I!<<wCdd%LR6cn
z`G*YeL}n#8OA*u-9FTBeHQ3U2$D2%?c_zo6{Ynmf;?qzk+<xh$-1*T@(D}*FNOn$Q
z(rHkaH=sTbB$A%yq4>;v4?-FmpZPEeE>|cQu^X+0s})p)BGW@khuv}lQYJY2u$ne$
zl3ByL(;vs0Rflr*C0~Ip_hVZr-lgT!9|Y7U7j$c7Bp|gtX1DbR!!!(Yarp;%aqVd#
zqH}GO<)o@|y3*c=zFO^5_vF_?Fzz94jG~9F<=G#27wcUIGdjxQjV51z-5cTBn=sQ;
zq-5pyRgu!mzke-sARwVVZK2Gug8S-#QK_*j3m<+DntHrVc<(1J;yXXTp8Z#?pp#C4
z&5)?cqQvsPcfOTayTrzVXg(G6Js?59L)D(QGxG<$4ax{YLgVb6LVE`D&?dhB*0(Zs
z&jTF)l&7)(GyXr^Sd++xJfoi1RWl5zXq|nHpj0#^m*r*y3NszJ<QtdM92o(lcom#e
z#OBXnp*xfTk@35j>XnRw+SIZX-D~yV5CP|jqrgVkxqUkatXKolNFMfDG%hZ0G;03O
z3)jN+@GhJ82<3#~z=buLza6i7&clY_)ng3Yf8F(rTT3i3ou;fl<1~B{=gF15&O(~A
zD~qCg=k|vf9h(5LxXfWR*vq-p)4Jtm#->^{0BeKIYSxTd{HP;9a)+5>dr~P2^LgIt
zt1)?<{-hpBGw%EOwTwg&nx-@oOP7pvv`OQLBhZ58BlUuR&t8|-8nuEV<Xyh$3wUes
z-VseZrY^gj@4V$-S>4J=TAgrk<^tntx}jxGK&`gPKi8U9kQOP@0d2M>=rx@4w8gU3
z3#)$nOzbg76U8yplheHK1D~LqI-;4)yzd`h1??1T4Z#e<^pw1z;2eIC*Kq8;`Yij)
z!m=3%7El;9SRZ~F-_yqK*7uZZ;1&#P@&+#QN|bsM2i^#sWfLrZ<}*0rZ_Z}?<dfO)
z&_>o@@_Fw0_{TA~+>M=T<I^mp7v{*H=BQ&3MhkcwUbkBMrL*&Mmuub5U=FvKdRvtz
z4^tD}D;V|IBtesyC5Ii$DNp(XuKw}mZ2aXF#Aybdw4{AG?{l3<&~FH)Fr;VKng$(+
zq7VO?VN(k2yHn=JNvZA{+k@eY^e9T}!C}70X()6mszFoKXu^?)^2hIeC%;T5@njsg
zWI3Py>sJ!pc^_`7h1y#F%FuAF*=r<2W^%fVCVbxJgSEVXlB#G7PCYZ3=et*4$Ls&~
zUBrz=Oipe?qKwfb!L%p2=;P-zW*nkP07886X_`5j!aXBm>3qFstpjdVgCrW9fiiEv
zd+g+PuKTz5uyg$_9D4d0tbWe(aATtwYpOuC+Pt~QjxZ=T*PKZYNRq3Cd~P%$0ya#d
zzVrPncyQac{Co=QZo~%0QXZxmCfPb@AjMn(i9O9h{rs02Sda%q0*s6_@h&53#EdUl
zR$1^N!DUW-{%kBv7Z$F0y?Ppkl?&@JpS!(Q6()pn=+q3YdmbPGOk(Lo5eJ=m8kty3
z&DEt0am;=FRRo*^5p0~4HgH8zY<<S2Yp!6iwYW}(^N>aniw|B4i<g$jTxsoE--YLK
z!q79Wt`$gOx}w~B#np_6!KtI`9W5f39CZvVSrTeI-&fS_1{bH_ouZ`icr7{P1v2M5
zZ5lgUJn-oY`O(|o&f1QrG1DQoS|&C7n(m~sB+60b+_F=8Q1pDe7YT<)Q0ZbYsENqN
zM%myUk9o<T<Hp9YGR{{ozMKaqr?HJOPCV{ljy>W4Mw=2S0;ad%9CoZ$<R9rnH+P>x
zB16ifIZFW93zO1Plvztvl|W1L`I_Pd5leR@B0Fd~YoGso9{s#O$BmA#{gzv}@nauh
z+t<E{eegld%oO-8VhV9z?y8@Q8l_ey6mC`(#cN(#V(I~8yL>1rf6$8WT<{PyghBbS
z7+o^KDZl%Nbf#Ne|Mkxk?bwX%hG~N_;k_7Rf`zO1y^|H$lP6RdoZO*9ey>B`)coM;
z+V?C~9mMRCT<TgGDo{d9VV7(}qWu{sappg~nj5CJ0EXk@5x)51momC_ithA|{5~<2
z(NK}hApI?0Z|d%_`OEIiUgpZu-YtoFVFyjkdOLNvG-K;zhnKwkRV-PwA6Yje6U8(W
zW_E1jLvMczM;x?_#F|k5N^nJzzNF(pQ$w8v(d*tdJ9ADFhQHlecdZko)5Sitjhp`M
zL)`o0D>(SL6FBlkFT(FPhK$CQMQp!}yY}+6-!Hl@5H6g!@~OXxBQE^(7f71pJ)xPU
zXj-bju6q$OT6d%B*;8s_bDw7}A%>CW2u5IRbOMvaxZKW}oW>k9bbg^#Sh(i<T3K%x
zRxGT?{9W}V<-n83=-#y++vy-qNL5MW2Fcok5fkP7N@aGCTHu*>W~NxQXi4B#7GIgu
zAtINt`MT>FMaWWzH^P+j9DM3YXd^+asSdoFThKl-oUDYY1?9;p%<Q0j$89u4Fg78I
z8nm6T>XcK^q?ucC&N{Jod!37YR`ShXU6I&n@W#+?&k$?I{TF<UYtR38R(nsQ+bz*@
zYXj0_tqJRHh8<-(n7c2Pw@pD;(a5s7tE_q4X^ft945`bQ={U~+_@{}IF|3XGyT5-e
zX_g^*=E9y9{R42>-8#bOGLr0fhbeS4g^l3e+!QjX@sN;`N<MK&)RJTtOH8{tf<NUk
z9RI4<v+~TRF+JVku1mkcE$@FH-5>oJvuzsRnITgb5Jeo`xiaicY7SP4|IeMu*FMg&
z_!P~>xA|1dSX?6^9bd-5r#+Dss}AL7U;83@-(AEl7ZB4T&@K=1t7kV1V;-msXmSQ7
zLF<slLi;<P0Zql;Ev`@>J?Y>zt97$lFR*e;V=Xfsa*C1n7_m4TvFxc&;Wy7co11q`
z!S<A+W(?o{>(?+cmC~A=%%@$UQO{e?UcvtyInaYzsa?%p=9=N?ESO+eN^ic`8544L
zvNldEfBO8Fp=Jb^dYtcofz-iYzVx{~?eUKd#uGkd;ZzhDsWl)nxZ&yH^Ud>wLFaBD
zC7&LKM@4W7&ULZv7WaJmbKLTs@3HdeM{(R)=i*i@N1JgN6qjZUeLqSE-F=mqSPF_;
zIn4=ceEa>I_~Flfj)_O`YA_mMh`@)trlcb(bBr8yj%CoAkwk4DE*J`h7Bcs*P8xA#
zW-Y;FL$vT2i|}6NaptvfEnMZbVxeK!2Zh0`$5ntd<Du(+#RxIpE1AH=+CyM`EJs1B
z&>eDD1j$DYMq%fcO)Ocwv@9+}45$kfmdPn*?tOq*A+d(UM0CAn&0|j|i;^;xxZntv
zAcpEbdH=<(FL{mOHG_NaB5Aj2VCZI!%m^K0LsJt+!A(Z{oH=&;<w=?L>Aepc5NmKK
zDN5RIGn#g}>AZJw{ii<3+9*a_GpNlli6P6f5{q))QL@1L(c=<EQtFvFB@em~5DH}K
zaB<8-#`1($ypYb+4kU{C=C{7bh6gt@wkYA06W6hF=>!RR(2kkC!OPI1MQ0WDv!*<K
ze|H!$We@>F=cOW`4hL6_m1ys?Q6{5G_~ju|gO6j<{TFk<pFD%p&Up<hPdJgC8#ZwF
zm%hsF|Mo%bFV_=yTKKGmN*317Yw1c>RMC6e7SBsNEkkZ-Q>P;2JbfofBDk>;Rvdde
zk9pD`@Y8R9lZSqBCDBY5I$7TI7t}fFO_P-Y-ZnbBP#TVJ55l|W%5K<I4#XbTi~TNL
zI^b26`UaV07F86AIFFTtjulqD;JKXe^8dr#J0}@SBUV575MTexmoci)ZB6$zRD<7w
zFu16Oxm6;2l6*1yxz4fdzAq?rGU}ByOG!q?c+tzx;*L!ZVVW^ST}<4?`iwJ9Jchsj
ztC!)@7EK$6Cd+7*bEEwzf4_OGvN!g+57&^pn0N&GY{)2a-lJKHotom-^FPY<U%iAC
zhaJVqXT2I-wE{+Ca9YxWM6}$mwa?l^g<1PTefaKO&;ckOCo!MD_*+bMQzXho)M6?f
zplBH^*lX0WNAI{VTCw`<i?Bj1^CCQP90lMY-0Hb8M@5Y_5@@9GNTr1=T-G!#Q2+oS
z07*naRR7}|231|YupaYOH`nM2InUUpDGy$E9ixdwG^5>Vv;WbDlSMY5MTOWmr)c;*
z{7Uli@y;FFXpD~s+KqQPFJEz4%Jy4t!e(7G%fP32H-m3BVcij^L?wT^_?{(QXukK7
z!lpkb$r=d>qBzg?U)@O4Im9_6cYWv@Wzo8W!+JUGolp!dD25zP+FNvp3Bh)*o6^XX
z_3!>58!q}fYlzU98KQ<oO%SfdQG%$&d!I{tr4ofptzf8WKC};73{9EG<Y;^*hApiQ
zhd$xAVAUF;qyeg&cisnRj5k;^;duQyFT%CjkZTvrfHI^i*>u2x+&_39d|VGBlnjrl
zny)Ik_A`yoMZmd3Np2nLG&IX%1xJRr8WKz$G`wI{$YMiw*cuLe(epX!FaMHsVuI}6
z+gbmKkFn*VFA+VkiKvyKJ|OS84m4+Qn7wiek$KIBt~H<{I8%ANBf*8syYQTcflG!P
zv-su9IpdF>&DQ(x;l^)&0Xsc~>1Mdnsmqs+YV}Eu_Gh5X`ebAdtIWME{r0iVW|;$0
z@xAL-t6B#C=0&#mH0%gn5e|Fhi#X_iJ%zhy;kNGNfK8jZ;>~ZN;UG-|8n1a1ru|T~
z^i5{zo#*MEeEz%m_d(}~X#dxRy~k+>X(YViUFY+yt8c)JjneL>SZ5it4URcv1sA^O
z?Q}aWn(+u(=5s(pYTWj^xV(Q#qI(Sz)K_WH#H9DndVjdo0VgPRU4kcaE$;iw1^nug
z&$D#xK|J==uR#x9iyw&*pPNvaq8ZQ6Q3?}{ded2bc6B!_HHCH|$fS%y%fY8UcQGR)
zO)~FLBbbQt>q_yt%5G0$20b^ri<=@?qFY3jX41$rC2UP6SUvZ3J5^nnFf3g2b%y0V
zLNQ@jXc*?TJg5+y;Np?7^@f{iN`zCG%raJ;dJ^8)Aha@hIFW{8x10#537Pjcf+))W
z+>{$lk~%hCc_q!f#BriHA|?(#3>GcQJ@Uh^+35AYpd<y&h2)|PK`}mK^RI4VB5I(S
zhbSV842xGC2;++&=VjLqFx3I{a2b9OYc01z@@~jYwHTS{aQ$1}%3WXlGOMvjt4*93
zGM|+WGcLF4G-7*?W-+zkIq3<ZD^nthz@(ZGI`x7Z9bs$3aLRN4lq_ywqEWv5<xAPP
zZ70p~MLhbLLpW&VI7uTy3Ebj7ne(jTq_-@qmwmG(&%+|TbdCwbz8N=#V#^VJWfY0V
zq2aYfY3>zk3^vz|s0n%!&GXlYgmioq``A-B_BDUQ;>VmuyR(xGKe&wbAG?tB>Yo!$
zP9fb6Vm!7e^mK4Bt1_g`(Q2hb@i~9WhkBf00T7tnfFKkfm2lgb2$~JjMPnTGyH8<a
zY%xFn>_?GJ_Y%)^QFRzov%c*E`wQe;`e_nFKXBQ_EQ;Ouvwhkv)TR%;9Jk8ttyT}4
zz~_$Q5n8Uz@o)Qk<jfP<q%E40DKnRUk9+_9V>Hu@%y%l2+&ag-9&4_>&s?t~+3x9u
z_Iwtjy>C|o043fd(&Rrr{}tZ-k^i8%WPe&|ib;ggW<oZ-gU|i@he_NF6O9o(im_6P
z_0ekok^xXeHC^4k>B2z$hUlz`NjU|Tn&tHXsZ2@dEn?fd8Ft$Y8$R+0uKCPGEMIdd
zXPkW&tXqv=)I`11vO03}5T%y2A+x45y-iYKUQrd^h#~WE?G3ka$K4NLBMbOETosJ?
zO3ziF(#YP-SO$)f`uf(&NvzRGdM1mebhFFd#8elw&I{MRxOy6f0~Xd}zRIL3n%794
z&^Wo3Y}11z#$cO<%tTBaeJpCC+y$Z9H+;_nue7G7ShRG1JSHSO7ePB7B<-^8`s)}m
zLGu|!hAHPc;M9}wNi#oC3qSXo)#$uqsFXje{u|Aw4V%~BPV6&WH^qq1$~>!%Jr<G?
zB#O$^-YO7F_uyx0m4GaQ8bfQUMNG=|Z~G@6xa?9^Mlp2Tn8@PPV<L;@_LfEr7>^fM
zA(i$2*Q&rPUYgHE>0%@lyJCpcB2G{fvALrh`ozZ*9km7*!HgF^@ZnD~x_EzD)5^<U
z_80k-$oA0#`m-?x`;-X74El6vQ}9SX1;s1%2o6g|@Hr2gPTmX`R-*{2!3o3}2_{D-
zG@ceqltJtSXPb2PTfvIwy_nxP`)^6s9nI$Tw{z#`Kh2iUeFpB`2s6`Un&wC&H8Fhc
zRr%f`rfPOIG%5vENR@{#m)0KbLLuxG)!uneiDm~Ox2O^|NATlIS@YN@bJAm<%w?au
zfb_PTiKZP{ngvahk)oC?$T}f$ymar%_Z3kx<8emurVxYs0Ro6Up`qv_b3YA{In?tR
zD$0^jq_`V}kz^5_EaUNSe=D8ik728GET8dQ_x=yk`p!>ibhFUJN{&v~$FI5dXjtDM
zJnw%xbvM!pbdM$r1M0^zLAV2&an-dq@#;6dl|{>zk#<5ZKpb0It*w0HqR%nWP{tag
z!6I7gL3C!;oN4xZsSBZX_wE3x??L%SK8xig1N%|!Nsccx3*mmw)ML6CJHBxlSAFjD
zEI)W1XPo;w^r)lo<4JI)18iYQt+s3(Oz_PLrU*5gW7S?Edu3G{&Z;8pEowx(|HGdo
zYK{a@N1`0%uY(?xl;620PBO=z>X6}{)><tsLBD2_^nhgb44l1Wku6*c*X~>g)EkEV
z7uI9Gu&}TTn*1SlSxWbo^(=}Kbf$w#Goq!-VCBj{V#&|!(#0qboAD?{43pcpuxQaz
z)H|%L2q<0^rl*<Suz?su*LgbL)3p%?o_IVyvB4)g-$IMmQwc-&(czvK)}r~rGc`?n
z<3l9QVMVdl(!sLw#1qJ(2z3=wQXV_CWXr**XYZd$&I!*H(y5G*)N|FF{~r%rb~y(c
zi?&<X#N-PXLVQBNoUdpMJqL6@ir@Pom;1Bx`DX`CJ(*XWiP+>~PWYo|;vxkbamAH4
za@$=G(rk`!+%fAo?4bRKayt(#EnOLqI;#Vn(R-VMUZ?n9H*iT*uz*FdSPUWXPQ(Pk
zvlJn3d1`halB{Sh_7_a1L<p`C;f`F#qh9vcocQO@W~S9;%avDi(+595dc`%wGc7dh
zpvDCUF70hYRr;~d-j(a0;>g}U3$<BZJ=LN*<ZG^5eOAYsXShgg@rxRa9CZ}G^M}vl
zn(uy>tylbjcsexTMJ7m3!!Gsp>u(2iutogPq-W`yan1E>&oK188#D-&qiR)^RyhrM
z;Z{_i9vaPYI!$=``_E(hngeO8=ZJWmZ@%G;aOHJKYZ@=UZ!LUTU24u5uR8Vmy2viy
z_3o15YPdly**Ij<;60tT<F0$}<9V-mC1d;VPnLD@v=|w+Y`T9VA9%;xIrN}qENV2u
z!EPz@Z2Fn<{hxGwVxn^_|G@kS9U#@wdIpO2I0zN?*#)v+J%1s6TD{=2jA(M2O`rWd
zKlt!RIOyObIQ_3*1xKz!N8^yRR#*!5PtWU9VCko@QJY2-xj6;65VC-LS9oCCB$s~g
z2SlS|WO%%(_FE%{+<&iX=!Oc_cjLw$^c+gv<w?rCl13w;pxyR=a+US2EKC^|uKB+9
zuQv?)Ev(19R!Lz)@rtP9{$Jj}h>eM@VY<DOrH3C1)*`0LUnr1&W26Up=(HzkjEv?@
z2;hBaClt8vPLg&j(3UhWf+Jzc;Rgpa#MclN4G2uJG0!_kyExoxg}(ZX?mhP+X@+Eu
zI5Ifzm~@UsN3O$L8{A(|yw~8NFQ(e)2Oh=MX6SvugN+CnOWN&{WGOek{T*!m?hjar
zK{{Qc$l|;UlYt`jv!_6-9sENaq-uFt`Bb2&T0H79R3I~!8HreS;tAM855rm^P7LpS
z&wns+z$$E(a`s<77fD0&R|)0v?>dW$B|X!FK}Cr4NnLffu&qb#GANNktPp#MykOOo
z0#@xS&u4l16dd(@Vqm=|a=|_@T{24iJ5S)mH@}gQN1w>d)(5%$OJCsLkAH&bt_O&w
zQ!=tJt*A|lsJSQ=MS~~Uyp|1xcyy2FT?|F)rEt+AxM)>V1?W-kltYKSvT`0Pif>rl
z$^$w5`TvJI@B9_N`ub;xvlQQH<0$|KUu%xE4!vVkusMpVI6SUI%;$cx>ja!#VdIYk
zyx9jH^f8t^iWkM@po&V^8ym?e*^*J7`N4N`|Dpz|S`JB-FTdto?8dEhXJ&HX!1Uzd
zNJ*2-Q-MuEtJxh}YT2Xap)BdOqRs^US;|b-<-ShFlV5lin@xjmX3?%=Vl1LFJ;`}*
zdJRu_%t<sG4ZI7ouqri0%GC40cFcWy;2b8)b0v5viZ_ZEkBSd9y%6k^OD$c2EA@VG
zlG|WAHDv0E(iU64`bB<r;rT3Iy^7P%ItPwA3LR@8zD^yL+7x~ER72{HsXgtG^wjGm
zXq71vM||c>7t?YbGC-`Y(pNMGo$~*MhTRXD=I+7sFn|?UOuW51j2NsKMw*S%R#*Cl
za0}PMwGXfT78-_qbc_|ecDyGt!h<*7#8@2DZntn4mLGRKSrmn4K??4npS{KST&!l}
zvg|MgXFki=`l}llGZCUfCeRgSbmalCWN~m^GsPxXP<C{{;V{1lpi*(FCCk`y%X%Uo
zM4di_BOr|?cFl^O=||9nR_O9T`EH4%7L>1CD6T>_)n=sC;g|pNHXiuKB^>Aln|85L
z&Sy3GxVnbymH|lG?`f|kbk>{iS{nuD!aVVXDWk<VnrwECM?LeYxbbn+N8EDr2ClmL
zW^7{#%a<(Tq?3=J5f?fYN>RG>4QZ$qk}~TU9Nbh4mq!13JQw7;fU*#utw!uEKg;mL
z{BE@pLfV2-6{7}?8)PdFV9iUP&yml17A}t1_VcT_<pbx_{qav}?3hB*8B{zr820CG
zW2Q0-l@P9|buwDPgpr54ioN23LU({luUV<&e=ZejtA)p}4>{Y6C%DljCqL_jjE;`-
zoe#X7WZPDvPMeGjQ);2Y-zyxvv}_~_R~A!cLF`rqGnfQ7{0`fP`Nu=Q^#B*6Vowpk
zN0?*`f8Y}S;JyFKjcyVva9G!H@k?ID=v0U9%ruzX{6H+OM6Z31pd9Uw1>9AcEpxb#
z%{DnGos1A9b7V7)j>4b4<geJF5h^jxImRZM%<Sm!j3=DQi=XqyG~+OVQ==H^r5E)F
zF3QkPQs<nK47XwKlY?td1W9kxDK&Sw`gqp~+D%EePqOKP3%TOM=d*mxS{`%OS@@&Z
z;TAV?%|h>#u1Zmy4XtRI|1zYAFaF^R;}IqG@QDk)M6)?jx)>I~MQAL@nE>=o3Wqk)
zvgh}9&(tRT2Wu>iBp&EQyud_SxE_&)K~<Z;_`-V3Lvhr{;rFVb89VN{i#XrlcC^|Y
zaPo<$NkX?aC=5$N{zA9YW^8PNGJH%cJ2$JIhi<%?*yo#<db*kCfTNFwW>OOpQ(7=;
zUpnf19p?QOqs8w{=GbxDdPb56mpV|$R2W;a3X-wjalRDeU609~qGoyD6@opN<2@#1
zWVc&0B-V5DyWYt?-?)V3#$tUIM6^_NuX2nlIUhN9nUNl0sLt+DX6^Y}6U7R~2kXDK
zF^sNR%|VYp1!qA`%%?v46-F12Gu>+QqL)6Gf`cW&b2bl+>pSGM4{_X8AwKH>|MfaH
zDt$_w+pHCpVVDC}pADAe69;cX@Rso!rVTcs-5g=$iBIM?&Up<JC!Ng9mIt}{v!CRy
z51o(Muo*MeqMM~)O`Z@Pn8=wAW4IV0@*X285iWPUs!f1&Z`ZY#-Xm&q-A3>Hs1%Fp
z>@_ILexrhE#B@eRIOK^>=g}vg!X^LqF64oYBs;rwQ-@fKCr5Wp02hL0#FdmNqb6ub
z=6p-Lq6E$r{-*3R=cGX*9jcp1>gEXmYJ)yTJVp&BSq#S?$M5~iKXP+-C*7Sh9J+Iw
z@Bh`y8SO$R?N*8IMS^&cM3>n&{d9a`FriC_B%=(buhf#Si%r0F=(cB&#t6@O;oq=+
z!xl^y<5Nd-ETS_t%~OBtF`V~LZ$h#*;^X{`sJqlk4=HZ;a%zvI!m!B5UE44h6r(Db
zu>PIN^+2ZfHJWRQ#DGIcog->@*?iH({Pg1&a=_|CIsH|y#~phNI+6736Vm%8d)?5q
z-bcvnYFd@L$wqm*8gO3u$u&1~>z(&tZ4{_aMZ>3H#pGE#Iml2GZ_Ch`(H;#WYk#HM
zcZxGCiR_$v(E<}`;o3*nxT-b{uxw#1=IbQW8k&^zE4OXO@0f%%mr6wq%)x8X+<LV4
z(GUOT^Nnm~dXkZ`al9&GY?v;%Fk%;%vHiC7G!-<>5TN5>`H3gslaL6lwPM5=dT<`;
zb7FY7@48+Th1SG@W^7x(p4jJhuvSRbvTWVq5I4i9TD2Tgm(5MV4b@t(jn|%&z@wzj
z<2#;_Ou6yAcXP)jUt?9Apxrhm@`weWxso@o<py8rBnC!O2m_c8y(8>$)SovG#0i_7
z=eXbhZD__g6?RN#eC4a(Wn{F$L{oVDnWqy`INSK#{#9y8q4jgG_b8w1V9+7(tIW|-
z{HKi>9wt~HC}`mud2%X(1B=1OmUQ(B4tc>#IPQhdCrcW1*Wb>yZ+#oB@BDydrj2%6
zcu7N1=tI+4SaxbbT`!&xt&!tpZ`ZE)EVP#bKjdpmeax)-C%hi@Lez+HqoXW8@i%$;
zvtPhh-uVvP4cF0VJGxoB%z5yxj_OuZjcSZlLuvbPV;LClH>^<@gr*LQH`~i~)VUu3
z>HDYVMOL{7YVh?(g{+w{@wDIO*cUv9`|#Ku(~RAC3+vB&FC(d=>pE3Iny1BwDMQ)$
z_FA3JTB+I3pY+s#y>zi^UoIxj26alO+a;Np;Osa46Ib4P8_8IM)U_C06fra1=E$`x
zdGFi*i6$u{Q8N_drQJ_q3#~mrLxwKtU@K1D#a%F))M-lkPLD~MmPv{=!LHBsUxMp&
zi6?im{{0{1#}|BxRYyIVQ(yIW_@marSVSI3kY^zbK-guru43SzY}UHx@7imi|0NM0
zIR65g<BL&SB*_<3yUGBgXuef@gAk?v;hmF!Hsm(1O-!_A4Qm8On++-})`fY(!u5z=
z%M7q|VIAfPXU`w>!U09SBfa}B8abNWPP;Uh?+;^Rh&2Ov@fBKTjA3SaCnF=Hc<1vm
zuL}v70!;3pz3Cy69IbTXnK6b%M;(cZ#YmNeT7W9rADg-lwj8O0IGq;VhaN;SS31wM
zT$d$>9YSVI9s(leZm$J`Tr18*@`M6Y^bEnt0*O!+Z05NClb_=DFMX9&Hby!f;v~w0
z+#s@XKvR8=(5m!gD-3j@(*bhVGMF|G7%L9`oLk=^qq}%12R!LaTx_s5;*xLukZtW0
zqbYy%^vBYSv*5^-+qinI9pQDfT?dJ9*+uiv&o$RMq|*OM{q#J1kIMPTW36C}Mbuy&
zBu<IFkd7sA#%Y}N_kTz8#1o06%e`N}m|NfbZ`k#BlC(Wpn&x(i!Hp$D@FvK!@liz|
znkSzrEx7UGOOE%TaI}GSjs$9j^MM*;)MFwcofyX-d<cK=s#o#DOTNehKl}#GR)){g
z3WY`R1P7{uN3OLomJVTG@3|I61~y6$r%9DP`ogN&s`LNJDLP2evSmc3LDyJ*<J_}Z
z^^D))A>U%G9dX~6F5$ilKTnc^6BpL5h$Q#@9QfJK)2#5VS?gJUrfasXrPZg&Ta%2A
zILz1t=f34UF1X}Unj=eb?Tpb8L8sdsv1%z7fA*8aS(}I$=e3+v7UgWOFuQ+xpe5aJ
zcm1$-^(0h!XX%+Dnp|gQ)Raw^Ac(a%HTX_S(%r%B?|%n3eC5-uee_YB_S(P0A8`aa
znglD^0-`b0624)e`_is%9%a_7r`V70dT=wBegCJ}c!X}6g-`}Sa-9NFouxfLUuwdG
zhjC}ujZ>=5C^QX`wJaJRo8>#cFjZK%9-(Wg0hTPR!+aJ-c@ZiarT{nHO4EaLDJ}0<
zdd!h%BPwnDstqBh4p0Q*Dp}fr*yejf7>|o6IAzBjcVgR};5lOrULs^-5whmM@-UK`
zdrIBWVE~0ddcPmEO1=AVfg)|+hEC6rL=oNynF!t*mL0kpY><qJ^s?}^58o`DFw{$>
z1ywLzC7DXu_{j^o`XlGFMhvFip^+Gz%SyXiuf9yc>~%mg72;PXJ?EQ$-B-JC&p;*T
zH|I`E5S*D1+G1FF%4x7{1tOLtvV7tbUuN0L18G`#-k(2>kS-9F!c+Br2%SNC<~8;5
zOLvK6?+udsU*O1*T2o&7+lI-LF`>cEe~ZE3;+X8f{aO2x|IJ~4@~3p04QB4XjhjAx
zA=|(D4U!$xm~JPSz7|R2T2096R`h$H$Ize1UH4NN%xxY#a}57tY@XoiL+GH11jff`
zFJ8hEUj9mMy5?%ufB7>swogH;gE3N~x?)Y3sPqSC?PiCkyWI)u0Bm~ix<=it|JW;@
zEz#iIWvcT351<3afb%5L7##zrzWKFuj$X&kSlF+dan*V6!T<Q@*qJseF6<9pt90%%
zJYaBx*Z;YetwT>XNZ-5j#RMr>;?h;(DCWHvUc|>P{u)d6-yiQhNhH`zSQ2|ad;a@r
zN}G{J6H&#Q5FRPBEU=`OCp^1>*;SgvusPH0wkdy)WwLucKv2qodruvl2|C#(*}9!;
z-}H~%@Re_H^yz1C;yHhVJM=*OqL@6qxo3nRb2PfUzL9FGFe!`>f~B@M37`4w*JyWr
zxKE1D1r0Q$vTdm{TFPfx`x=M6xxY&WQcV+))1Zpmck$xU09N@b@J(m6qYKx<wNI`k
z23WkX4)X({YE7`K3MTXHylp*W%^1jNb=xdE=2(0bRl!RoT}FG-8q-dP=I8_vQi;8~
zY7^;ZY`^tp5^G5_4+yEk#36@5V~oPKQfkwM-Y}l|MSshqsSft2!p4oz?hr>72WVtz
zd11+6hk}WEXg(z$r<$neg}cjBls#fho=4%pb=mptukn+Q{yVGV1k>pd8B3OVq9~#e
zh$99yipS+i*VM+$bszk?-lOi~;cGgE@b@i(@dlIT`2i+k(g{cW-s90`0_wT@{!QGm
zVIy7J;E+SsvU=5WM1t_z=F_D@z&>pKt9YeqU3XpN?M;EE{}mpus`!Q6O$%Jm1mg-%
zbz?DJNHQU!NftL5dD0Vk)Ze~=u@g>X$EJta`n_*+&-ovOyEkE4GdRtQYkX)hbE+J9
zgD!qpKFF@tO$OrayI%h~t&y4tmeK+;G6tg|<sw<Kgx2C^Jn^NkV#lVPTzSEViMLPV
zr@Cbk=T)jy`JTqIx2f54!Lsb){!`e4kh8rXJJnvbNG@Xc<^ER%bD1EVSEUh;lEtGu
z<pb|y%bL|plX0+Z@YTP69kSs*bf%4p#~6cZL0$6$Tld*>hQ4NB2dr2RWk^9BS=K_5
zm<zt}O<wme?_p$O88g#uOp;;Kl*I{r<BOl+z@;M?ZIn|AkItfz&7PC?Qj>Z~IlI0y
zbvF+JDQ9XRT3Pk^LIG_c6CrJNh-P+h?LYrBxBuj7j(zgeIp*v)kgYiwMq`WzqaoC>
zZWN$%0Ua`=+o`{UB3Mp+2~^=Trc(IW1s5@~crmH>7#qSoMHS;i(-5>jw$dRD)co&3
z0Y89RR~d3pf=VI~Swka=!q*wu$IsQmwQ$Y47B4gmdkAt4$8Kx$aLsx0PS~;j4kB-G
z>cfQLHy(u-8&Io3z8bk(N_S?O(XmA#yv7)O%>glT!q#8iLIc733{{60!^)$Mge1;U
zG2erv^aoYV8^ElC*9Y%u-+K=Y6}0O}GmkS7Iy#P7vAhx$QuCmfYW!MoK?PI(4DTHx
zf}Kp6y!^Xd_MUfeu!)hG4smSq1Gshsi?4BpYXLWf2wM6bGqjc*t36)tWx?UAKssZ>
zWT%x_%&O%qJMmGZQH0urPh9u~qVZ9trn)@ug@0Pn$#Rjc5rdey7j@5_z2#vUf&N!{
z2sPN>;H7tJW=!$y1uI9NX>d}CgmmS89Pxq|@|c&rnC-0;z3U#Xd-uEO{P<@ioo=4n
zpjB=|oxU@d=3roLqI<Ek4EmEIr$K5{jy#t_363#&8dWkrMt7pgDKC5p4Liz@KKO2;
z?Ni9kcK#gXCJZGSJbQ{U*js;4uJN!avO#!s{g>|B&AI3wRC5P4Q?*5+bqC`T*^e5d
zxC0jRwDaD=-PVyLW31_9{NOczN4$Ls-)Z6c9d+i2z0?7&WkZA83hUN90Md23M2!Yt
z`_2zJ>rHQF>GJ((x4R@07BVGqZ9e<453_RVC{5cSvAMp&=)gKz2gt6ggl8Fezsl}!
z_p`sYhP(8HQ3mCGpc9EFa|)A_BoA%m(sR#c!_R-o8P9q)N4)Hn_?7!%Ml3i`?|RP3
z${b*h7G_S)R3&sS8W8Vu9R;56|MV(0Y}|^shHlm=?;}KOlwqxfL7H9*<K0@rz7-!7
zNLT^{75hNF(oA9+@c=7u`0u{!FI)=^!+*sY!mgR%<yqtfvXu6No3J=Cab%I9v1SdL
z%W&&KOTIpi1lp}h8c7oogZD0MT)9Blq+O=(xQ&Jt;>2Q&(();bk30-#B5Zjc6(X%V
z>O_r$Gf%|w;(e>H-j`P<@4Jsg6z?;<Do&K<vL(=HmKKnOwW!Ond?|%#)gdh=Le+w?
z`0W{T(@lKypWntB1=DI{jo^KLJep#ie5rqsI=`8IN$?Pp!@7UB<VUR6JOjh$uS;=8
z*^+wJ{mz*%F@aG-JA*HM@lp(i6=N|^{H^1OP(-5Ac_rxg2HElH0OWVq2T`x=Zn!jY
z{yWD1j0dphex=pzQezc`!atwl*-VL4Nk=X2v=ce$AKy&2ZXJz(4PXEOAOJ~3K~%1@
zle;gugnKXeEYZfTn06MB+QOcciZ3laVEJs@-~NCq72P9x!mjIK=_^Pz_hOeom+&d^
zNCP*PaQt&$$l=Ex$HnjcznHC?h}$U+pW7@K*2g{e2R(b6mwaj?N-B^^J$s-(Z(&cS
zSdWbPW4{YgJ-FCw$Z1d$!WKcXjZyT7wfxcB{x3JTc9KkYX|BJAU!M0~lIafW($L(O
zAz(5Gg-7QGl=_;oLNwbYqtk5g>2G|G=bZC;79FsPbSA?jjzlf7&T#SPFW`_BONdAc
z{co?6SKky_WcIYDpOv~dGqYJ4cJ*OT5!xA!1B#YS@|+;27CY4@y60BD|F>r`ecyeY
z{@g!j)pK8fTeb{oSj6X_i>;=y{f1bxWaYa(7^{o1Jyt@pvyB_P`~Clc9T}zT9ZB58
zNQ_T?&@|<V>Y@4ewFW+SJ{uOm_AsFXMKO`Zt7m+CG@s=EI(p$PT>n!VhD8hOFpr12
z)&iooZpKYcg}~9+V3#b0Wy{LOpy0}v6c}U5T0V0bHf{twNCJ9Vh=Q5gNq6H$VhkPc
z=~zoUZm{HtBk<OSqFMUv|75s@<$N^_Wjbxn`JCSPAh8j;oh~Mh=y=DX)hp20<ONkB
zQK%eE3I~mXsv`wAS3tLPv3IWLk~h4TL%NP=x<ecXnn00fP<#@mpq1=4_B$&zt)fTy
ze2y-8xQ|K@VP_8Sg`JXc_><2J;z|>7^)GH^N7o_7@${#hN<(~Q(2-|H6jKx(5Z2az
z81AZ381AFKyFzFEb$_si76WzEQEIdl6QUL>4|UB0;~ELs%KbU)6)$G(?>&W`trk0f
zeigTU;C$q|8%d^S(5#Iy)WY5hl%MLEt;+VH6ofr=Vc+eg=#g_2p<HUgE4G>77DcRl
z>Qgx3aZlv4Z+|PSznR947T$I9saCk33kqc=2hcu2bUg`M`}p~TG40{^-ke;KeR8i1
zTl>=X*$7xbBV40F{FKLY>RB)4-u5KpZNuhoeV<LA`~q>Sjq_>wocfx%z`({)b_J}|
z030!B=25YH?`PNXS8sR=i&m_p+X5T^|LnbYyk%8=|NZ{1wf8Bv&7BTI?}`OfL}TM;
zi?PL;C($&GNgiYJi-o2Lq6U)~><O{ND6ybe5Q!Q@)L6g@ibkyTKK<5H_Fn7z{IPaB
z=bm%#9hm`qa+a^*&Nz3@-DQ<;`4nL+ZqnMhmH#;JZ5(~*CZd`pHk94ioS0u49#s4s
z2BiyCp$A>@jSS)}=zyQ06&SxRVPb`0Q^D+<!~FE8Tz1N9SiF4)PkiZLu=#n<!EG2v
z#!OK58sKv=boZ9kY-bgh>52n-KD%9F`N6f<@$K(kLvwP9#JP-Xk(c(eeZ*v|p%5<U
z%@3+F$b+wT&pBo413ychXL4eUJa4@RywyizH2TMsQPs)ObJ)Yfdc}1v;2r+98;R6o
zjL=OT&Hc7OTrXe-dHtmtC<iNolqj-Uu$;jovev134_j{|YPB$`0goG5s5LPA2f}TC
z-#D{3sK|o~x|g4dDgh&%k+H-p3){9+k0YG-G{%~=x+#<U?}s-z9H+O`WypEpxzbdA
zxEVUT+e9<lxahQ#*)X?Az0)N&nG<dn-4!m`A{F$%E;&7iDp>bPCKZWa7ATtBV9z-$
z0oUBdFmcpTaKIr5LKGQ3__5E?Sigw~&x>F1EX1clX(ca%S1Z^2`c)|%m<sJ_ci}gr
zvrsvdr5B1^-lg;|LGq%o34xh-A4~+JI-SNO>!0>qPI%?%==x2xuD_oDdH;LZ@tKRT
zyBA4Y^EmODuR%T!G6Jbq4ALq*gAhm=Aov^@-xrYY?czk9bQ_9mQ}viQCLODD$m5>O
zQ=a?BeBy2A(!TPW#Je0xyBoTeQZZS~60GoXF8uN-LK-mS-e5C#^rc+|Ss^JH7Rl{p
zh|)5p@+JAER*d=7pdKP>cvIux7ydq*p8jNRPZOqFDObMZo%ruuL)7lzoy%ZOQYe|}
z&}&&qPbfU_q}!qG99Mkr$NcHvp3eC61{P*pxTr&2!L;Z2_<y{Y6CZLcX=;cF-BkJ#
zk3sOb{_2n{Q$o;{ALFv=VYxm_jp(8qW*~9E+#Kcy-{O+9&!m$Y9)J3)nS9E#NT%0d
zVq2Qb<))*#@p-=P2DziHR<&0v9Gy&JH+Vrfd>r%7@BRQ$Jtp-^6vbgZr3x-QM;F{p
zvUJhWqm!h4#h%QZQSN48Mk)z|_nwLIv9d_8LZ@gnMq{rZ!>W^pk!jl=pG(6NhfrZh
z0GBd<!wtk<@ZAJ2u=a=}P-}|HWv-QNC=X)+T&F{$K1QznEoGepTuSTuUoa^a>xCq7
zq(&Iuv<W6A3f5F1*(xQ3VaXA_H#VGVS6C`RMYl_9$4(-Z%xEp5>tO96hoGkHL=@E$
z?*q9kd%A(L7>FCKnHF`^;>)i+nJx2kG!lmq4P6&s>SCkS%AX7^`wdmV?iD~O;8OYj
zuK_m|ey#`!!a|a8%o9$+$76U^W?OB(_O<Vjn27z?ZD9ZX*I}&1d+pH*ry~6*GWg2+
z#Tj)-RZs`^9PVWq)qR{i4DOT-S(Sy_QIN?LCm6A$jR<%2K|J8(Q#s^mPhoNQEW5t`
zEq?M(ZzI0$R^m>Zq^Qd0ohB*ZIfcqf<wt9to0l4sLlu$kestVtQbzD;zza5tNhf2b
zA8`WD`jeM&`2`<k`+t3w`b>wUyI43_7w{y}GMLUDS}=JNN+y6Anv}|FHi#u$hJBdK
zO8?$e?om?c=a-cf^@_1|Tmv3@_G{@JdpI2uj&7{sqElZ3*WHY7Et2A~5(E{hYqLL?
z9h^v1xKs+hn}9L=^5#4E<G+3dw%MSy*ugfSX`3u;y`7J|_g{I`gN`LaXhw0t2guKk
z{%x+;Uo!7X_a`^~dThg2&>+7B?ddz^#${Rw$P0%Wyiw9lLcFlZ{H2%jsn?&wnl=0L
z%s0G|#z~JSUDF_n!x^F87fOMZ1ysUnb=6CLVAn@PiV~em!)C6k+;r;<U%2#ZG@4Ds
zgzkY;50R1?=k%Z<t0*ui;fJB{Cw-NR70QjWI$KWl%GBgoQ4m|*dK-<=*sEh=lra37
z$F&am8_8-AMm@7P-#{}mSamEW9oA)#A;~sWM9@5NCWomQk~AT1G&ACy2Okm#pPg^J
zesw){HE72Xi6KdqsROrwje?VH^)OK?xLa9Tf!zz+OxYpPENAktus|}mh;<<e@FFZY
zrMYDbYHW|Yu}DyG5ke;=C<sM}F(hqeOoeZp{RZM2Ze*;}MtwqL0}<RtF*wN-FRC_K
ztGlUzA&(BKPI)rf4g{r>rgV+aGIb7n>~G;?3u3wA%I`qk&`uJ5_c>1_c$fy&s;a7D
z6+W}vVO`}~40DuS-I4#lWn~VpE{e&}dJ2&$);lmlckMV+zxTU5;?=KWrd}i7dJ8{(
z=R0YC^;^{E7lOR8k#c?(NiBsMnZw|U@?w}PrNi~!rX&u514$s3=CDX>Eq*Mf{*Z_8
zjI&<Px39RAYcD#VdduS$TcL`Zk;t7Y4nd8<WAOxsWTSw1ylByt7*v<;ar+(mPIQIt
zg6zF1FqCQB8{m6_bGVuy(`#!J_{kbiKkrRkXBJQqj&z<czT$6*XWMj=4k>w}SkW5N
zr7aC9iVkJg5*+50RFqb1xcXN&@r*xuDN>&%Z7G<932*Upck+$PFX8wj4n|#YsV;OZ
zJlTtuft{44XOav}5LU}!&^`}|LTjrMyoyS8wn^xG8-r^*>T@k_{kQk<#kaqeL+^Kg
zPI}F0=!1_Z9cvI713pyk#VBI5=+bGX!CLtB3_7@Ub<;+rL~W`1;Non+)@r=(10P{A
z(J&X(V<H14hy++87$5G{I*7br)Mk$Dy}|&<Qs0)nvioL*xHVIgtG(UP7>#|FFpQ6$
z!>?VqTC%Y7xY>G`yY&uYv4}OK#<1?FBk*E+s!_e>VPQvWcNej7jUvpK(uT)+&+Z#;
zplK{P#YQ#S7&aVrG(L)o?=K0PgP=HjGkDSB;0VCa%tK;k7T-;R@23eBPiHL?`|sC#
z&{cFry$_Do`QtGisI^+$@ZR&7zvA0W`@rY0QBbu&2%ZZPD#ks8*nux>aMit+PAV0M
zPsL<N1yLJ*g_DMM6fu455!izcAQg)t;v*mVERFF75uRs0^&}#nDQ)ynOqY*#lvM$!
zRv9#pn$;AQ!R!qi3L`;$Fn{Yd8o1*hz{Acu9Xa$+bYU0Qe*AoH`^W{<x9`MsT7bi5
ziZ;c3&u2<T^fW4^3d<7p`hA-i^}r1^>*A@!Pc=vn+{Dvfc?P%NcoW}$*FVzOu^U}%
zq2hydtOVWpvYn<yHM2^VU3m}D6}&=pkMC>mJ~|!9gCQrhDwyUZ9JG#SoqIMvn%_n}
z?XvlfZCriMKQNxAINuIMj*4%`fZVM>CTVW^<^#NvR7t$!7q@QZd4KjYXfz=;SmT+B
z$4O?k@|pj5KSvy}j#}N&u%-u#mv^>$wz#2a9Vs{g!kz=<v>LPa=tF7Yc~MC9mZkjg
zYqmpt=Ps^&^IQ4G2QT2blb*=KPdyoTzeDkp784m<&RNKU?Oa~m=N>(}f+NVPkZn?Q
zWV~vCbQwdMz@1xXxbPDfGdVSl&)F%N=ZWI+K1it=4ZJu}sWv})4jLJd53eY&s)JcJ
z)mt(WAW73x6RX%sM*IC}?6Wa$fU(hYxLfthCB=YD^N{&%+kz>Sg^q{jK?j3GVbg)$
zmTo11@LsVt!ugE9V9G?qqzMbR-9`*J?@?ozNj&TCcMNJ{jJ1_IUauB+Z}(I0_V%JZ
zSd7X6<~^NVJA%=fH35<%N*dQ-?b;%+&#f17RU;L_J1x{mn44wp@=N*gCobdwiP3gK
zl)(Xl7PcWWQ6?GbDiW%Fm!wPk2}|#Tg1}nHyb_4BJ?(((Jvd-aB94C2<DeeQ*mk$u
zTzSPcxNgKD>-T5Vy2<R2TFM<+t`RLmDuGpX_vPQ-s0P&^yh@d!Xi9*M6-$&t%?sEu
zk`4QD)GJQquowI(Z6oab&XxS?-RHrzKgKM!>AEf<H`SuErW&}j)+x|MG<b*Xn*@FE
z`>mQZijg4IsTzozMqq55&Vh&Vq?et}%&y&h_uX$J+SP`Ic9>I{!@By=5wy9|jkrGY
z^RTdVK*wX*^8cXkk$Y|<&v;U$474dt+0^3WNhY5BG#>FcFXOs}879&Ott-FJ_5b<-
z>RrJ}T9`=Yf(_Lfr$qrtOXElDaoskpH09bG?&Qh;&r2}PwX|CfXS>u=Pt={`E0<l&
zamOBkcM--{I1A+mm(tTirkF121?k;l)8t7(4^u~i&qA(>Cs`NDAu!)2y8TwZ{>C%7
z=9}N(F)w`yN51@3xWf*IW&=xRjFxfF3<lI7YQwpawaRieuu3_9d74(w)XF&qVNT~C
zVFc7Uysh)mPhZT=g*m*ns93~W#8|rB6ioL2nFw51;~s|#ZMRSi)AC8@LLN(}vaT|=
z&|W_g0u0uOpe|)io-l|El@u7I3!||Q$Ji)gxLe2ckndBV+o834H$HW!5zN>)tep;8
z=t8NIlUD|_GMD!bV@y!~30rTk0o(jM-Cet|-V;X#V+>s}L`NQu6N7iYNIa^WmsW}y
z)q=b1r3aGx7vzChXU9&A_qZelfkc&PY8=L!IdZxO6-`E(_9oaZb!O+`rl0f0x4wl#
z;&JFE)M6VdK2oU|^&F0wI?~YmKnM3)q&M*?jB_Mcpvgr^B*UmHrWP}AYi#+ghvFpx
zlq;^d8mZUONO;OqACs9!l|q<^RxW9nZIbpe5C)qsv8vQ#v|{r#Er;lNBRynHe8BIk
z8ca$^)ThUo{GBK9s8^hh-LyZgUtZ5o|Ls4R|LWJMEw=GVyXaW>Jm@qPjr+>_?pb*I
zw0u!<n#}^)-1!<W7zgoqgx|7(C;rWAh)jd8zU>_1ZM)Id0v?xjUa};pzeOj@quYm9
zfvYslVG-XwJa6-7RJ|wBk_7G*qjg*qbHq#kgtaF<h8@<h0p;osz8|joB{J706<1X7
zk>Dy{FvirQnQx3_Vhm}kjjHmiTekA#=e>|<e4NF0f|;-wXE5D)K6Al`Ie61FH5<{4
zqAF9n7r&mpBVj4>YB?9@yE_w2-oefiA<6dK0J8$;9cE?@x#k+aa_VX9zTq~W{qmDJ
z=nr0q+cFI`izO=zsjuw+l@2vkr6uXB=&UeJ`lpumwtF=5!Eu_htJCH^=YN>Snkl-Z
z6(N$~lSAGSlF->sQ`ghZ5$zZ7U0wHhc!NP$qRyKA*?VPba$@N1?qv|vXpF|tgrPZl
z4tx77SJ=G^I$a_W67Q(5oq~ETNT>SJk+dK{XUpEDDRHd{YOuzFcZdeqfU2W?*Ih{0
zA*l-r559{unlQblgd?d|$cy`~epggMR5Rzvyu+ZRyLMxo!=g0e8lA*ZpI!$h&SU^9
z8);U7S5>SSI&&@RbG!Kb*=KPmf?t>?4n_P>9`Yr$#PkS)YnfgRB!*R4%@tkp>hsAX
zOAUvMXpk<{>@W;BZJ0|E#tu6I_B#kr8nqf9|HKy<Uq4AxQl9pNM^IBR#+H>mtx6B}
zd+k>HLR!&tACy>(U`MN&m{O^&dV&v#BM_epiOIa=D1oP9yu&KEIKtikejIzonXG&4
zV@c-bx#6>);rb7LkZ9X1vd~2dJOQUl>OJ0A)EmS{;4_Fq*4mSOa+5=n$gCY{jLBS&
zL-jXO#YBSJw2p`V?a7R9T*GJI`g)?7MVRkoJ|>wfJK(Y)PKu;CS1;35%*B3p2jRX!
z#P9b~kP?z)9L@ya2`4EwZs4OjkAC}Gxa;6e%)131ux1?>oc1@EUGpS!?a-~t{RgUK
z>P+3uXN2-jnvyo_{P$Hq<}uHDA&u$HwC7V?lrm;4HJRt~FMfu@x2(ewhc2eC<bhe>
zTEz&7R=Z1;5`*;bcq{k1hH}{kR`D}lnk&~>YAmUD#GN+XE56C+&O8H~IG%piSu}t9
z$*^H6@Bk3JbD64LIbFR3o|{(019J03DjblW+<2ueP$|4hR07di5R5U{sKx~sUCb@p
zwj+_jWiTZp1`;76{N9TX-6U(WqCnUuO8$?QiKiBlWlIDrmXv%}de9O$QsEg0Vk~Q>
z))c3L8Ir({(uL93hod=481{O#FH}xiTeqQ|6!o4InA(3od>jWg5<@QmR!Y;I)H$MB
zoxC<JVvFF?c*pFGHxYZXAPBl@m^$D9sMmt}kW@C4+!#m;-qfJ8bFbWa;X#pAC_B4$
z1_K@wl8MB7>T4$<3a3Qg8SoW1>PUq4?nTC0ZNB{K*Ro;j3~?(Vj!lrYvlQ|ad16#3
z87+5m=&;je=_V^$2AP5hsx$dPEy46D0otM*@wneYqdG<{4$oCr{fJt<!TQNbjy!Zd
zB92_tqzFz{loH=l^?n&84#VbVK~u=O2-PQ>;njPdpfMqE$&`k)W=PhKbHIyU#Qp#H
z4_WXXlAr#BAHVgVFjrqg)N13BHpW&u3dRI5QpM~n|JnCUN2My?LgD?Pg+~n7i8^j=
zgA@Psr5t$J(R}*Vub_7GE!epQJZa(iQqp*oWsMD%?gRr3sD>=YdpOML=Py+a`toj+
z7SK`?H%P`BJo7DY;CdF&B;mNa=bNXW!dL=6=^_$b)N+Vs)knY>gLW0E)%fbSujbi*
z@=~Vv-$L4gS|egZV}gadZsDTy-_OCDCWvc7ElXfUw1Cl+elUZ#Gc9^JQpyjrdd|Yo
zRBpg)%afuUHY8q<xkVZ~X1VD-@8z;{-_G>r13Bq+XTb@_<JQ$kC0iTL_Y}$sBf+d&
zT&_te)=t$hap;?rAs4RN!^1Gf;!|Ov>v`Y5eT3%3T2eq_1Ew^aGa3xpj1inqffTfh
zwJDKxc%Nnt;5O?*2Su^!=F=;5y9y79A*LZgR$ZB1Gg)d=>Czs96i(iwF&g_WVQ7q=
z!`=#-v*DG6JMW@yECwO%rmWqx8E*oFs9NtSD^g*hr)h#Q5r{204B3Wbl7z)uZVU&C
zLKH`IQpegu4*?sM71fF{O;r|D1|R@rFWh_x-hH`kc@3~BBd8jpT8vYrF*%u4TQfB!
z5!3}oW{jbmwrMutns=NFKl?e&t|u~?=8-W0=UTLFwNOLqfnN>UnuQ_S=#{#C)|Dxa
zu#D`frtDi7F{EB`)-abkj(pq+_<E?|{^);xPTLD@EDwL=@tK%OS;f}Auxa>7v8Vd<
zqa<PABPo7`^^+&2Mjb}eBD@N~zpUevb&n8(L5Na_F?5@C<gt(Ak*B|o&YF$b9d~iv
zM?cK2i$6y+I}4po7TAP?J@*00*KO|Gu<zd8QhTk6A9NA$yGnD-#C1Ar*YU7F`b&;D
z>S!)G`_-6R?m%`ehGIpQJO!CZlXVyR9mogYlMW2+WWb;At&T$pjncyL&sP!#4{8E;
zz#43=fj|5xo_5w7`T6_|@$TKk*Zz<jKl(|cc0%g9S)v&3#fS+VSL0~UFCw*wk6!#`
zp7tj%XT#ygvU8?|j0p|r(9SNtcG>4R=J5UT-V$k6?JffIUO$53ZR^rPl~fbueehIE
z$K|XCYQcGfUsxoX-_B3p@+PkO)I}Wp@Q3l3bN+$!!N<YWSdix>Bs@7htH&!McTSd!
zA)W`tg}I6>v-MTFYSqrVF1>J*xjUy=K6cS1+<5C<*mw-*1QSJ>NI=;^u@aP!Vq*}i
zG#bipKlYJqnuu7_$W*p8s3~C#YQh@x7*tlsX!>4I318ISwBIW>BQva>nhc7GIV872
zxG$51QMxc1`)o7}P#-;qm20=M!gp$4HD>YlJBfU718#S_OdoVG-b!HrRR-}=(bE>Z
zI;^$f;4xZ|l@-N!W#*<^uo%30(lo(~W#bWt<0F&9FUty3#gAcG%hS8{kowd*)MHdh
zW@m|v#d(MGj#QQAy0wrgerUcqT7yGz>Zm0tH-GrU+<w{RY_@TLheQ%OGR|jx|I&4~
zfTyaiN?vfKpj#^kj8wuO#bk~cp{vspPh;~s;zJK0LC~nir#|xq>{uN<JpOSf2KnwK
zgz#$3ZzT>x|9?m0?$+Hyl`+qf!=WxF$A)Kq9$6=;@Kf**Ve!od?#P3A@M}(I>Om*c
zo|)mUFMo;a-tz(MZCi;t363-%5;7ua6>2zmQp&#Dv9hkENGYo+rjkTCh#L{<x-pJ_
z(aSjQ5s&6muYNi4jW-g_FOrf{2q*QebFG$0wAU7`8j7>Tv;01o=t`wPP{E|<hMCSe
zL@cgeXZ&f;;K48bOYUkfa)6!W+Vej^>#J8Fb1l%oR|vtPqCRC|c9zD(7#DovGXCsk
zr?BPl!`U&nKx6$lV-gc5GyK=bKF)y~rfAgaG^051AhP#l#Psmpmd=>o>T>U^x*I8j
z42gABvyzQ~QPRX=TW!ou*Kye^U&Sx3`~fGu_{ALm%2P-V+Kicq3Y94JC2tKWlc6#N
z@xgsZC?J`+Ibo0A#S)Bd2`MWRi?Vf?pRLaO3})wOdwBah-pAO)G^qiJY#~{zLCPxo
zsO-#1Q+&6>1^<2?Z++vd`O>E^Ko{pRJ}ux$Nt#l6d}`5=u4V_?e@G1~a`hAFO^_*%
zBFlz#YfFPy)0-fyEVn%xqj9%K-2k=GbGZAp+p<F7K^?8zZ>L63I7mFyH*ZFb4IPF)
z1D}Gu1{f3k2a2|~R$7owS-j)+%rq!e<y+kjYYsmIAJs|>ZIzgn=O((ZLX`{}lwGqD
z*p_~=O=N8bs}kCsglK#X-bdlk$p~de1uH^2H;-QPeSUEM2ian3XuC^f4XJZ@7jUOQ
zq4S*wKvj(v`8D|(d6-ii%p!;vtcCzn4I!!V4sT%Ic@BE$gP~pzIB{e6@@3z^iZB&L
zJo>~32d7-I17!93Y0tywlkNk{um5f7zCnUhv4*Ze@Jo=AS5S@Rb3uYUs4)U!5vRoJ
zP#e>koZ!H}{Bw?d*<Y~3b@10;&rjd=FC^c+inx`M`V@>QVI&I6iRsEdy!Yx%Dt;#!
zX<vexSri%kx-}f~r!V1QPkS;K{KIRJn{OtbZD*BJ!R7Lqed$0ihgtgyC^BRpz3<dO
zReYcM8h8~@8~)w~nP)uIB15-sIpY6)3Elf2%QkTw)f7H;+Dowu9n$#)R2*UihhwqT
zW@2g`|NJli&P!kQ297x9cy?}^BVH2`brZzx;5%RcGMlGj8gUKDItyY<FPvpHg%FWG
zR>__l<5hc8+TV??LiaL^Ww-07brQN?`Y%5Fs#9<a2~R!iOx8c|4@owz$JTA&-g%#$
zYbFy3fK&ywHY8VS5pS{1V7<kt>G`?&@u+=psA>+Y)&Eorad~&fXI&o~uKiP=`7ds|
zbt{prfwzd*0OQTf8pMPLkX<?x<Ks+^H96+6Lm3lg-S{|%9&|vM%O=8kk3=@3FPNeu
ztwVP8m4b*oZ4F$<u)xN~noLbig+#%Rgf2$oUKzDf!mtkx>BvLtjys94MD-e7RmKiG
z7@zqV8!266a_9>hOk*+~&q@$1Wgy<^(%H71nvL+0#b^Ew;|J`I&$tU^64JYsi0qr~
zWw7y?r@Od_F@_YtMmSaC@dj!^oyQu%X5X9aUZ6fZ!=-1xfy33}TODe3L5*TCp_MFU
z*Q849>z6d_mpNQbXr`(?m8?PwTzyai$#k$q6l}zTsdLb8oq)FnF)=%4T5R7rPc4o(
z=7<9rkMj1nu8d<Xper)qneN-L@_$3Ruf5rPA6iK(Fb`Q0TwBQ+{UU+P?7&(}I$1{^
z^$;F?#_23fPmtbuE7$+WhuQhr&r_e7hh)+!;#dFxAOJ~3K~xb;Rw333wmrWl*>{4g
zgFjVzs<Be3|Em+EQ72h9#v#vn0grm>v-t39UXK0wwbbSoa3mpV?upEL*GUytnd4!!
z!r|zLAL)H$uU@J$U6JoB0#XM!)GM*A(ZTV=bI;}aI<yzI^TbX2^Tj`V5p}nScG?J@
zq}5`4Vv=*;^Fhvj`+GU;fyc6Q*CLxXtfR5$Ibdy#|GwnY)cpbz6JtbLDR@*|l6$UL
zmTw2_DQ97LSD{qdDm%G~_zYSn9-UvLzI&cuyz^ar`i*bq;3JRYsjq(%wMRc1x2}$j
z!FykHr}Fz!^{6R*BnlZ)(VlKk|2_4{-{${TbK}!Y`KYX|Wyqf1L(mw5PaU1q@z3Y|
z8{?B}v%ZGU?$Y#ZwJIKLxaPUmJae=2+;q!r#LXrP9nY<IZN=I^da%}J&&gAT1Bwpn
z2g#6yuf2sNQwn>ap2_hsB4dJ-YZW;#u-h~mqcKVt?vq`IejS9cXDF?0TLZxg#fyOr
z8$-P_t3c#r!m8#!s#q&MCOKL?E#`LNXJ--Bp!A|(Yy|73F=ET$v|lx2A6GW>%2W6n
zw)v#p!I+4|DYhPy0`b@cY9qYKs!K+2X@}Ug`0sPx$okn9(&`e&7N;Kd5}b6iB*;@i
z8Y>0aET}_O-huXGIhBEI<7%>_)$_{Wjlo+*jR*CGTXTwG#@1MK+;QQ+GM>x7c@;7~
zfr(?D^rVw;>PrbHT`4qNnn;w*`}=~b%y0gR_2P?o#1z3nKU_=)RLL{x$}IRd3XxYL
zZ%FJIog?qZ@n^lBwg2aF%+Ky-+vT6*miNC8yLAS?xEN$Nt<4foEp+)yPQze9%)Q%T
z=-UG{4~nx&vIzm1wKclyHgWJXf1k%Z{df7~8(#}I+(bOn3D7KKa8f2eM)QO!tV*vb
z&N!Tald74W?t>j?Z|>q+36ec|$J9vROljtRjTl6QdOQhR4(8FPoX#CyNq2WRc=v9;
zd+I6Fe2pF37il&p`Rg;@%xVAl9@ZcAAa2|_Lt}l+%uP4)@S_jnD;HhBbi>kUj$`sJ
zlvD#sSqeiM;ImL%y$thNSeY@Um1H5P*SP{#s+udqNueGk?G$_a?R@>Ur}F)aFXqHQ
z{6ijc`sug_9E%^TV=)-@VXbNCz7{{#n1Ul9s=-A(lb-e_Q!z8(B=~}eg7G;4Ia|wx
z&_(`{g`jrELCax<p+fHqtdZi4<>D_~$}exe4XHIrH0$V#1s8sf$)&T6qBeYP9L0DW
z^WvAC!aL5tkf%QP#msdgoDVO}`xNzQ*j3bJut#04gt026C!b9W?;VrlO=_|2OCSdq
z#zv>bXza^`AsRi0U;ii`xZpb}-JQFMO&kc6^*T&U6{N^Q-KU3|7;u((LXo#r^MFRx
z!`7{sZi?}x>|t$k4UCTk`MxTMMef-H+E?vNeTt+B-Z@Nc@dj)yrZzDaDzC<d%*uo9
zw7LG%7t;CBPnmQ<qg%WWvXnV_%uw(bOpnxF)n(NbB_>${4qc@jl$O|Q$+AE3;Jpj7
zoT_*u)HiO1skNv&B5V2l7ruh2)kqc-jz8{LtO(wP?gX-Ya<HU3GhjN6YDRZ=&E|FW
z5cHeuJeOrLVY(zhQxhN+sszZIwItJH9QdLa^N_!KDcj~2;1@S=?OWeQ^t0=TyIm6P
z2Hp-iA=-p>TI4G^EGC!p&1FpQ?mO=<WmL=Nl`cU3ZK+a^1zkJNVb6FW|K~X`<btzb
zgZb63h!(1yhA<nw_dW8>CL@Xu+8b9X26Hd%<)vG52s-lsJQ&&e!p|ZP$PGy?X1}LB
zgH2EP9kxaWw`+zQKJj1p?_JBr4HLZZPhZ8qfB2Ie_TUGwZL!7Vh6!fwyo<kj(R29d
zyWbAo4wH>0F<J4*^g7d5RXPW5<Dx?iIm>?85HRC_t<dM|A#*Y=*jl1l%|r3sF40_z
z<Xhk2KTmlzJ8!s+=f2`pj`-t0B0YE=S`WdycZzE1&CstvC)pleTw(rul`NU>PmrW1
zoi<sf;JxCVL%qisa4rRtb?I}7Tc;QsV8Dq>SWHvi{?31+J~4%}g4aw1FNY!(1+VZW
z^&X6nCLS?0w(f3m_POuk`W-u=*1$n1Xc3Z0@nlJ6{{H&8bvig<TiSmpes67%p`D&y
zgZE2iz=f3xQll{%du>Fcgkdiy5+0zvh%U^BjZ7V5>(@hlEF&RG2{Y1?(3Wo(Hj1*i
zHw%U}m-vIu9d}Y!#U%-8nv!}?eZw>w$C$`cRqdBO$$Iw2{Z~&m=Mc?2CF(Vt_p}xl
zX~uOTFQK|*4Su!-Kl?d9{Ln`^Kx4WKZHzUA*-`I9r94o?lwr2!&<Ec~e5=Ojx<Xu4
zR((8RG!V@5`|{3Z1W?05;@R@x2SVJ$8jAt0{mCzh$HrJYHNi25Z35@8Hd>*hpaW`!
z(s#Iw{5$t%L3t_oYxRqy2<$S+Q42(=Css)75&lsR=P_@76Whj`)MvKx%lEyP#jjjJ
zeX&KFc2N_8BA<~XtLsKGC8q+XohirMJ--cEHU%wUeYsDAQN(+!QT)0IHa_dQJob6d
z<^$)Pfw}DtqS<cN6%4);*pNJhMBSolO9^371eeLl%8GX(_tvD)O0b&g=Lc>_#=;1<
ziOIS{-h{%4H4%$39Dl~Exp8`o9ZkpN=8b&m;!pA0$3B8@UU@YK9dRTJi)|WfQaW32
z;hfX{f;YYT<<zCaM6F(s>T?uSStRN!LJyRL9?n_N>ib2J{@`)|m(65<5nz|#J*c;V
z^sN?cb%__|x$Wa0;^K48X5+ezJp1qdfyVDV6~A#B14Jspx2DgVZwb-2lB^D@pH#G~
z3fN!{;j0!_Oit?0WsD;s%|F)&ntF`Qi*62MBQE;S%lYMvcVZeb9d!^JB4aU$5Mi?-
zf-l57YO#>IE{RJ}Jl-4DjBh{_L)vj!VJHdbk~OG-L>xIVDI{ORreA;ZpoQ2%UO4dH
zW39nC$J%w%#XT2A#QVT18jb&b34<LyhhM*TH}G#~c4ATonFm9C;|4OZ_KG}SrdlH9
z!Rs^0v;u>l{oeb~!foGqCvkR=naI)s#`oWhi)_)gSbCB7%tlf*Q&y*Vbx0JUVnf#?
z)!;879z;lX&k-#y^4YV_;c%&;3yU=Bh9pfYrMB|ODYzchW2nMysG6ho*5!5iZj1ES
zVoNZrXh>1SE0MK0V@M;zp$|JA7saHBaMzCA+<DhFoQGo{a5OG+qITZ(m_sqNHZ-ie
zaJLZ2qcLRFt{V8WL?u8=I7N1FdqQv}7<|1!_u&0_=o`+&9e4<xZFg|X#TRnx2R=k&
z_Y5ZK;KgB!#NLo|7rkg{FSo?ALHA-%Aiax*oa9+r1$qV3h)LHpIN(Xo;0e$EJ<dP*
zZ?Hf68MT@D&}lG1ZZ%grtK$AO_YTl%=vTcD8R+k<d;bns{Jh@nvZ7=oRmMyCX~u_E
zfbq(hsY87Y&wtl>w2wc5>zY&i`z4=er%Pen&~{y#b!esu?|=In_^TH_8>LGeH?oFl
zkPoU_TvWx>q_4ivQ&bqrS?DFV4+OjoQuoXpoAY<(d@%T<D@427L^s^Vch7ts-~Y&k
zJp81`^T?A=!9VC&()D$mgaVG&zFN61dF?&FUy=;TkQdp1Z+TlROLkQipDJcHrPgx9
ziwQ}qm8l|y?{m)6!E@ffegvscp^?R9ig)UR8ex%e848Ky#5p83MD;pef|62_2%4Sc
z1``rZsRa8X1-5n2I~caK1$q;VELJea;F6T}YuA(=3K`su7$ps(v5$uxB@A~nSy+Ng
zRB+q26FZ1)M7NVLwP6!Pwm?qj*u>twQj7Mi&nj9y;imJnx9`AsAD}+Q&`liUn>V6H
z3fD}LduQX&inFvBA$1;!Bf8$>NHGy;vq6$|PGX~c<L}O9!_02%e2bb18>WeD;dGXB
z5u~DPua#pw7tYq|HAug3=?cmeT5=eQ&DZoEP~t)m<iyfa;ot`zgExvqmaDJ%DJH7X
zNmCwj!h^zl$Q_<lS57X6gfFA5|GqMM{D1=iI>Yh@`;Z9hgxD!jG){Ybila|Hje~yg
z_u08?8}^4k;l^{%L;mL%L@iIENrs;Dg<+UvoB-oPd0R>`)4s?__uek1WiqV=kuv-0
z!P<y)-C7QP>T`JPlYfT`Uwazda0_NvEAwd(R7~MDVlwS|QD4~=2dINdZ)LSVxWBuJ
z!%9BXOWv*ydM>?bejcEy1R3aDD>R)l@W2P~#DDx3zGK&OUAx9y$5M-9q}5?;w!^3X
z^}Rg#w@$=H1PA9F*d(QccD_%@auq2l3m@8(&<uVCRRy8t9g#KTf`w0WC?+<zP6xX<
z2jBfNmz{Mocm3iQJoe9C#G!xra+0HufQedAxKk?7%RWM{E@_OcK0DL_8HJL0wugjO
zGPh}MIpmg``Rd=kf?vG%U#X=*smfay&SMjC!A1Yc&u_XD+nAu;^+ZujY-*_Yc$1Qt
zl*D>c3*H9!qWIt=<b#zPHja>35Gxp)uNy78QN}~$4UscJVQ;81p$<6$vd;l5ig6nF
z8_uzAdQIsKG<2Gb%1fiM?-K?ydJcE9t5Crkt9m-ywqaFpoestr#x`xn=L$yjdL*c3
z_*kXK)J>|@B~^^~w6<-dCg9TqoRXxDwFex47@JA0mDQ)^PoKRXQ)%DfW<+pl5;o*+
zN~DgOF{ttQg$^^9Ue3(-zRx-@kan?FaNc9QU`<pI0ZWs%A;qYI57O`WyeIdsm1w0j
zO2UJBkIm#;yJCo^*T8=3Q4ta^eCaFSran=p5ettz;XznRBRi3*1Hhnq->_sxM(yVN
z+RD%2ZW;He2$FozSy44eBE)e_cdEgf=RAXl|Mjc5qun98`Btuf-}xlp{0_BNN|Lm~
zuIn}HGDuMkE#MjzPD8JT>E2qiRqTG&Eih$rqz>YGgj+Yp5zqZ29{aT4<r8PTg7~JJ
zuygY`(%_6+Kuna1p-k^yG$1A)RDfm=x&texc&mK7eOms#Rrsuvk{2m5f;h(GyYqNw
zxOt|{F;9Lv-`lx^v=K8o-ehd&9M3rR1ipUp$2elsG!s!oBsF}lIF#*MYP1KYGw8+(
z@utcqD+O0;70*hB7+FHW+sM1BY6UNfUu<D_?Ba&^yq}BSbS_hq8+qm%-@v-(KA-NU
zDNH@%<M>Lkzdz}d0WVfrJ%L3A|059sOF$Ht>7K?W=JMCQndY~z;;M7s&Yhq69CZ_%
z9a^0>t&Z}xx4e(IF@Y0<7#sKtcqB4~XGyQCbjEqgl$bKCUB&ySvL{qwRuyM=9rP}T
z-Mdt~FWJ8~`CIaA*sx(>VX2b#jkfC1*w@3DQ8cpWhmZ{TOQGC6+uODgnFyo8e5=ja
ze*5E%E#&12Xi|a6&(k+s@yMwvC<XT2;`VJ=%~VNzO5#1WbsJF0)t7Ye#oYrjl`*Dg
zN3>ClQK8;wfEdPX4da!x+a=zyolF1uoouN!aP2mYS{;YSSc`g(x-`R*hK<<DFy)7A
z`udC+t+Mo_)MW_I{je=5rCF6shaVx1E$yUBi<I?89SWIqwTXo9eCulJwTQ`D%%KNt
z0!=CVN4kX6w(@B?2=_WV4DK6wL$6FD2;K-}=7Nf1y`!dvq}D_p`A8mm?m6t*vId>o
z!S$c~7`I(`A@#N=ZMTsqsK^BPgp>;G-t<(EOL@|}Hy8S4dZjW?3lN+JqbM;TUa^tI
zub*Pe3;u``p86y%eA5}2JGUV-txQcODCmU!s)QkFwcycr-O~5p6_@9)UEGmXKa)_E
zmqK(Rcm5U$`V6L$>?Dixn8p;}zVcd*d%$C0EQa+pYEy#l*vVtpt>@$?p2UWkMW(IA
zL!W4`mU~35y+7zdXn);D`-=5d8m;u*PcI*0srUjJha>!KwHD_+akoqS^Ph6XnP>97
zi$2S7PkknjJnId(2S1#2a*Uc4j1NjmB8rXjnxpA;q}Cx1TleHhD!*AP=BAp#ctg72
zfRwZ~gLdcGI5EX-*IfrD#*ko*@bOQ7o?Gsi!A4Q$%nY?S#-%P3Au$y$i-b;JrKppe
z`e|R{k|&f@I?1`5u&yxImQ2=c!FYnom@~mIMvXzmuwniB%JHT{psu4a8v8a39wiLF
zMu%abntl$+@k)EwE+QMCH4?|vHf<=Jf-}w35Ezr^0wpG2N`ry2a-j;Tr!zZ4Yy}%f
zU?MsmYFqZpTBb|ON%#1REvjWfK?di*Llk5zNM>40Abjf`Z{wg;upLJnM<hPQWe0$<
zCg|5^N=0HaQfBo;m0@33-(l$Q!jvXnxipE%6m5J!oJTQ<u^jl|2cRaH{W##Z+wLH?
z5%)XlFq(B+>RuJFvOTSXj%r2s&8;#EueIXYnrqeP(=Q6#1kEHGUER{zJk5h%_gd<Y
zd?>qTwvk?S71zD<UDUVk!nRu|Y1UcP0>hrm!1ly;dx4YG@BW~J7K8*&Mjni+6HK#C
zvSETlpZ$kC__t2tlc)V(YIoj+nVAP%VUbsbC048EGC($X5T{`ff@M!bBH7cFL;J2G
z_s1!sztAns8Op8J0`b^1um8t)@ccjj3l2K^Aa*TwkecAql*2b{V86R=<;D+vfGghc
zPq3KKUT75`o#Xv`joE~O?U;THd_UZ>;Et4@20b<BA)GeNm4sA^`Kt?UYTM?x^Ftry
zbFX<l+i$poXPtZsN4)4yNDtiv6A{J)enS;rrkQe&S67^0+7G!Ij=Ws`+tkWB%gE`G
zQLHgwBOdt|f5fiC_M`sjhw`w${d0UbK~m^+Jm>xEhp11iCC!s^iBL3y@s^$<IftY4
zEHML0I8wl1b7|zBcTprYEQ`<@daqR*r3~L{aco#Sm35YeYweASOQUg5Bn+dmhX<|<
ziY2n(!JxEu?j*Jr6<Ab9W6OS!xk=`AXOm-L36g};pLp~fmI~cA-Ni-lO6oi*N*cwm
zb~@WY&Aqo>WW?;^Vb~aF47OGy?Iw8dsHtac`&Qi7u3+0&zQ&XXqYfj2F>%ov&-7J#
zB<V!@gQuL}t-ZhYz|OiK@>3G4gYdpeUNy_$Mb^-<F`MpxKfE!h2-n|qC!NHRq@ITy
zcO2?6r4y4WJt@l&SVtTCZ`|V7KD8Dp=T60{9mHaRS}N2`gZAVk2fp|v9Q&d_<Cd+v
zY2J7p*T3tXaN}J>trn>ykU7BTBx>=Rg)1o{zAzTt<A8XXs_oLHl6`(|Xs5YxloCZT
zolTp#-|xMMqaO4?zHsKLL_6Cs-w9Z96Dq%X;8mb)!fFl<)1vfqLWV6?VJRPBPYXM%
zeyx<80MYW=3I$r@5G%~g?`CXlg1<TSEu8!AkF#d;ChnS<r8%{h+i#uWlb3#)Pk#H8
z)aT}qU;UE!nk%^a?6VjfZ_@QXAijK-faE-tOcz!AiUmsmj019tqR25QRxG+Z+M95w
zFQj|bcw7Q3q}XkD@RPT`iSK;$V{AP9NS^hMw_<<$iTDH7VIzwzyhF-D4q0X1qgZ3h
z4sHJdwwmT@j?c|A@$An!Na0R130!aRVn~~oiKjo6r(OJ69{axY(W$jyVvsr?`OGEU
zw0#E>)lms5gjOtR8u$&t?6iCj12k%v1oXXgI4>l~lH;iVV`Rw9%M$*Yg7ZYyBHmNS
zv1!dXx%IT(JLSO97!Az59l6LVet(}xt(Lu}vXme|d-qKCU=^&ztY2Gfe1(!i1ux%U
z94M^eDGTV@YU5fhY*wA}8XQ_#H(gXagnQh6VL)A6g6njxhIc78vZxA^wFV2{zlNW@
z?_(SsO=41r$&7Wf07ZH{7KT6P9CIuEo2|&6y}62Wtqh1FN^wF+owDdHwSx}`E}-K0
z;SYa8+-Q&_DfhqM(HNtsD;>O5mJ8HYcYKV-eY0FpH3|%>G-b6HQ;`$(P%{zT@dlGm
zd;*U-^9*jCYf<02gCD>1?Qr#vsCN^*OTlJAS!O&H;1fP*%NL|i?&dDpyV+=85Tund
zXW`SJf;yp5k8t}<bNrwFUp60jIA4C#>!@v?BWcZ7RE9!=t>OnNw`k^`Ue!A*|K;%2
zD-2dZ8WfZE%0;aFMyr+k9uvy!EgFRRg;}O1r+Lcr|B{b={tFy<_#w>Cx7oCQ9XoHh
znQwgkQl{5T!o&t%aM~-`*`49wbxpQ^?+Q9seh*n}kpu}f)c5kOq&HmvpFm*0>q#1N
zr{jK(gGgT&qOT>B3qVL_)Tf#ypgz18Zyc`cunQgBcfQNzuRN8zuKp>H`-2zoh*M7^
zJ^C=r+Ga=!w8ZWksRYQykCh&tndqLZr^?<f74@!u$v??eysC;^QW&*Ktwvg_1*O5r
z(n&kaIpOW^`T+HbX`J{{napI;zos+-&HX3JU|-8W%kn2)e#1Q`cuUl%WEH7xF9t10
z#zHMJ?6?1hK=|&FV_l97G#aC^_Yc8)Jz-d>=K7m~($m6W7=g~*ERi)KurvbAX7CQE
zbS<6w9?pT}gi5KJxs^dkcXkd*U4}VTyr!6D1EQz^B>832v3+^_iZxl=UO=oNj$<NG
zB0TCm&3c_Xe|!ze^*2(-V8s<W@MLCoUVTPtGbO%Pb?B_oz61gt;C3p*pM^*nUg(kr
zg_bH9nal0?{`<q^`b=ikaLo^Y#@O^2aTIaP(T5{G+vIaeKdo-;{TeVjME^zy1cjqR
z;69f)Xnzvh`+7q~Jp?``QEFasjfmu!!#MG+|H$sm8_~s`{Osc&W8v~EsLgfpi9@2`
z44tYE+$1bkqh7?rUfHrMuoC(L=88{)67b?89`T+iHn>e|dDtsY!!$i#e&^rOXesI9
zVj0xsa+lTlTXbl&^i5$YCsvmW?3el4t2?JXYi!sKP{!9)_)h3`dexw-kFaxb5g(89
zl;`~gKl;V>M9n7cc8kecot@WT#~nBPh=bOTv2J`4m%u@P_2*2Ucmj9K%yZ<1{W<?t
zuOd!Tw3QaYq<6k=&ypoYjIz5RJprgJGX<(UOm37$CfDhm2Wv=CXs5*6wsYJ0@8+_9
zJcqT@8+iO{&tUEIUx+(kKWwuWx?HLm!9SPF9YE-<LZx*0mepS(>G?Jt(DJEvQS!vD
zARc>+gpR%itOx52Cbl36Q7z*AANmAaXS=A$%;}6^Oy(D%rO8iFUeYD!g{rPnmsfgC
z_EwTp9b~yy?ObFqT+R3z6BCov;sAr~Ehr6o(4)@vXzbgB!QWe51zAN`LGNQrU@$kW
z5;S;6x3v&9>PU#14QSLtJ6>2?RZ@Ds%F<5+RXVeCnIW3bHcug*7>CG~!KZNVZ5~yO
ztXQbkP!l$G5rf1Al36C}9`#)ik5gCrO&IBwFU*bGhH{p*Ae(7__u<#S^xmpxrW}*1
z#~avi)Dd8UbEq*9zx?HO7-N|lA7}IW8YZjci7}O`;jkl9RynutYb${J*lN7Df4mNv
zP+CYGdm}gvWM*kJNH?$L0k1oS`hy<G%=T^E_?ZjYdEqCh?VLqhEwCAEC3A~&>d3U1
z20+~QY<~3AZF4A$>XIO&hhQm{>bDon<T!5AIFJ9or!cpBjw|2tCYp0?Tx$_b4r|Ie
z4Zd8T0vI#Ug;)o!kl|kkogKXfW-Hfpi>H@CsB&Pa_gMkNIoj<uEpK?-Q~rRT-*h`R
zZZbaJWPax^HcyRl-LJ03r*o`Zvo-|QfQvm(dCTAN?K|&e_wE)C7#rsYZ~P}}-4y4$
zA&HbKvAyz3Dz0W)-RHEIch#@OUs0{ip+n)T2|hkqw?Vrp@vcS84}QqEUiSvB`qbxm
z$P=H+!(MeVdg2MB>&K8<fS77oXU~f0eQ+bCfXekKJ?Y?gsY7=*ExZd#Sdh}2Z-%6<
zq{o!WVd!WPY^AIr>3U}7+Pv$X=M&YZNqm5ms%T~|r&tLd3MLpYRDgw2h?5ja*OED>
zJ%S~|a5r;Z@hp}lz7@s#5J8=1{n|A|QI!j{M?{d(7>&I(ydlU8+$&uLU8ckE8@mkB
z6xZ%xL`c1-HZ_LE5t11;mBqH4ztEdJXeF4e%IGDtGa24895l`=jWyF?<30l2o~eZF
zoxn+oR<WpHu*CJT!07OT_kxY;p#tPm;>d#YSgCXp`VSXfQ9->Dv+2@ZyEJDQI-|7u
zJ<@+LgziF+y%oAnS$F86cpHTSMugjL--<DYgSKqW^qR^xw(4@V(&gUHYIO%kV{h*g
zdo1U<jHg=3a|jAbMglJ(@`{O@befYK^OrB@e$W477G`!c_qDHa{Xd^abnC4|iyfQ^
zj?m5r?SH8R!X;Md*MA-=WoRw^*PIi%HSc@|$e=F-{sn1_(b=?_XP<H=H(&cBe*W%%
zp}tFT-2_B%fQl)6-Hd9%YpDJlT_!yqXgpR)wEMreOwyM#>Fa&V;o35Ofr7_Dl+syT
zpyT0xes(jDI_dY>Ig>KK=rFNo$89(BvOj+T-?;2TI&<@^n_OFbqOlG;)<idK<VF8{
zE?3=f9UEp-Zu-P$nf>|?&`vkwXqB?LT2*q>-Xub*CVM^GnpSBwOP-o!iIDdS77_=(
zonq$~+5YKI@}Fm($=qFAIqB7>v+*T=hP(ffXfq~~U{sk0uF@09YLOr)iZ^8nD=RR2
zb)`k5vSb4EHd~f*)O^AHFr@*ZcO(e!c<;ZnI}sA1P??uf5lkvAm>_Mdm6X$0Ityh&
z-CG`%Rj;+B%49NRlaZdY)C5siP-WAGbz$wuAbaf5*)SUS*l_oF!Y~Yv`v02{B~5Ti
zimKwgV{CE?wE?Em(@8KgJm!%88ikU~&Ozz|7F7(+35_*tP#cwsN^<W#j}m&VAfEbo
zGg#YaiIR6|Xa}dk*hzB|XSG6B<tB9i03ZNKL_t&|+KWP3t;mVUU^R}c|G&KMP_Gko
z1vyb-$q`au;}M6GWJl=i+#)-7?!<Y|5l0=;`=h1TnR=<{;ixurUtZh#YyacHJ<=-7
z&?|^DSX-kz6|?49Pv*qGeg$(e=ncQ*XXm{ee)3Cdiw>^=PN5kQTC1cc>EBTYg(rhA
zvl>pzu<Py#PAc$>u)M(;qUmw;u>JXiQ(wmqF8><aFa8wunJ%eo2S}D`5D$=iz4!Al
zs%!&UF=eiHU!{j~x{Qy5tgc)Y-04${QQ8ZOP;YX<C71HV=e-DDYtrgE)=bvfdh?BZ
z{KNmoS*N~?#qJ_&CMSY4XHlKbe&AS>=2L!~`#<xU?3mrfLpC13C9i!owcWEM3-e*G
z^P!3?OS;cXAB6N1qWiEsni;4SVhi2{NUZfFX+qpius7Ypx6gb7SN_K*IP#$<^4ouR
z4ik@m9DdU{-g=D5jL>{<$GZRLiTJ{-tbEworxdi>(=$K2v?7ZweW_jfzbfk`_ui2_
zyY))D+vfIdGrZ^h7h;>^5Cu|p5Yf;QtCd5+v^T-bI}D{Fldh1;u3X-c{uH?<=GPuR
zPxb#*6z?2c_S=M!(D~{;RYqep?)iivy{8Z5Az}jexv}SR(F#ciTv}j88_gz2gw0e?
z%8v5jhIxU<uSk8Okj~9u1=M-)4(A+=si{oV@LqnB89ubA#uz19w+5|8U@a0EoC}+@
zvB9lVgdo$WzJi}Ny?Ku;7ZNS~yMZrIm!M?_*ZK4Fq(XQkJAg5kPU2X5&;ej9h_LOh
zZ2@cUQw}-gz+gI^S3gyUR!#2<mGsfL2g(UqqUv*FMgl}JPmql|EQTbqaN-F(@a!|$
zxpo4xYla`5_b;@+`gLk8MKwhYxJtmT$TG2u_WgIr?90fkQV>{|rnDmTR#USd8@NU;
zFpb8VaLm#C!QY?5WtV(_nXmm9^+m;L3ML33crh6iBs0evc)dCx%v#l^J0KAs<d#)d
za8*h%gAg2A)C33LnT2`c@in~kmFMuv*T0Lkn-6AgzD+%Lj2X{W-})+#I^q6!-=$F-
zD>|#1nQw+D+aa;VS<7SIcow&aFrOwIT{m3z%F}7sI$iI<25h}nC20o*gt|<qohk2`
z0_2|s!8XDfi)(utvt2ruex5JB<`j0{d?P3Q^-H+_t4<}k|1mI8$9fGO)IJ3qCdvdT
zG;{M7gHr=)5w(bq!dDGSMS&D;dPv*LcXx&a;6vIJxq4Ok6GHLLWXY>~4V`zB5;w<q
z-9NmImeexYSJzvd>+9ACS?5c85r(>t{ET{u)D?5a%HN5U4)D%7_TO)FY4(~MEsoBF
z(YWVFdXFaz!-BryN3Y(ENPh2i6ILz!&`t2p1=x@Z&FQtMwG|V#(&=+x;!t%St+dUP
z*|{Le7a7DFQib}|1bQD()lsPg$s%ZE(TOSA7>Ht`D9&`|bGgRM94-5^6zE}Psj1ZR
zq4UsSTzjiH{mDUpLZYg@uvwKNICEk!<73F0X)qSAaQhwGNn9XwAAI2c#HP0%H7uf&
zJ(_i+@f$b#cuG|wq|qwK!yt~3G&F0ZhwR70&pd<lutPArcW~WDKf>JSzes(yjqffZ
zHZvB>=c@G9K7+B;pn!T1jG;fdS$=)>1UuCw<N+6q-FWD+JoDwJaPj-zi@Wj)n!8(|
zNobFFAH)D$UX?CACkAn36$bD%WN<a0l5S|POF<trvx2kF+K84+CdVj7U}3R?8JpxW
zPya(Mx#Bz2rY2DD*nj;v>l%hHUGgao+%(B#%~CV9-~e2pSt?xu4Nwx(sG}3YAN<ST
z^OM=_tZ@w%ulh0DFZmi#r%Qszm=N%m?^F8ThgRS;)rV9<l6)qWYO;hutl*P`+U_p#
z4Y%^+zkdrC|NYz8a@djl&fmYDHP8GV()}lqrYQ<4xhi&WHg>3{q@*Rm(F7N0?QiB*
zIt;Rk4p*N2Xzr$5Rm4_V!qMn||C0Gfi0@O5Qs^!!SA6@2eCj{H%GmUJlmug4(RtBo
z+eC5#{?cb%T-V?fZRJHne;-SSc%TeQn#ws4x;V;~{WdOR?>RazM&q7M7`pd~x}5Yb
zFbc<ASqj&^Pp*-mt*xC_CS4~asbVK4aK;AGRb_K1ww`6M$YmGUlg`Z+j>oxCPJL=J
zBoOqv#K=ARkSeBU#hz!VPfp{kL82J3HXJA>!_=x`jFc^W(H`iK7W(!T<qcMl^HyEe
zwf0n*v~T`pDjYtmWeEoLp}N~#Hx1);ynz(U4YzE?)W#80XY<DOB^;oqZ9b^_xu@y$
zX#6HN8MHr@&4MB?fDQO$lcmbB7sNE_Y}~@JfBQ-{JmK+lS}lI{nNM)n$38->oscAp
zIP1$QZZ1cYsn@9ZqVn1!0-$*ZAtQZCFR`@8QM#fx)PD~>dx{ALWU7i9hY?G6Vv^>G
zC-IaQ|1}>y_Z;}2pApUYtepa0LZKr!3=<jD2hnBnejVy*vQ)5mPzoZ&OAu8|rqray
zlWL0uxOwXW4|&8hxaqE05+`UU#dl|U*aNoky>I*{^|-@ivkqQRmw9s=DFXgHhf<MX
zzNZ4UF?jSNIpX)9&rS2YI4EAvW&iMI?42{{LObLuMk=I|(x6PWf9J}szk2WL(d^Z$
zRZ=l@Iw?`7L-LKw`O;ZuaKp7f;mNN#jR&3aIy(1%5b5Lukr*N^wq6hu49!G9B19sL
zm`s7sV|=g&4T?$mFNffQumu_fIWoRV5BH$5wW&%8Mb%PHS2Rz8q`a@5H|TLFP6VI1
zAUo$ncfmrMc1dlWQ_gxb<I|f6f>46BK}|`0;05GKclM^11n#Q3%$eL9;4guW4fS3a
z;5stQh%wJ=efdleI$*zW8fxgI46J&M#%S#O(Y;p^23@h7U#V1RZs}DqM_DGkzW0_-
zxQcYKRkTS{uQb+9qb4&ns%#^wOJ&@8x28(XUBS)IV=@w?;&BS`<OJzZd4IV_?l^xZ
zVzQt`fy7|DT_Q1<C?aY$FmV(#vc(kA658XrrbT<bUqxL8gOSVHsp_%@6id1g%aW2S
zPd_*mkvPxzh7DjcNQsTO{>Gaz@&Cu(dB<B?Rd@fp_I^&e=a#usnPKR40w|&;8jVI{
zFTunXi3*rEuZ;wa1se*+h!_lNumz&8z4w3x#6pl*C{h$D(iIqnskfYSo@ej%{;~Jd
z&OLYLj=-Si?$2kq%$<ABdCpVzTI;vg@Aqq9O@lS7mxK3V#xy2VGEh)0oXWh282!OC
z8pM?I%SMRI_#y#}Ky)y2^g`1b+|p5YKlVi&a_li|oSx>+3xCATAN~aG4O2+JOBUsb
zwdm4O(Set_>C8ZzRED86wM<1Aiv8E@PNc5x=Aq#&e(4CSpY$Xi`NSjn;#=N|x#>>K
zmYI<5sRrep8nz+xpr-Tp5^8LQic*l)p15JoS*0Vd6r3lIb3K$6UqAN(p71x%qu*-M
z^`4b0#>uv>=jF#fmyf^y-%vlp^2v!%925yB;I-%^6`llfFH`jh34IYB@~W3`*JuM{
z8tmsBKYa6>X!RX^-;Z%l=jMPixYyOaY6LBXFs!GadCY8&)*W|o<-eZBH{X2*yY0U}
zN1k#DOP~CB`n!#hG=(I5^iXKzKHXa3tHf`ccxFumfyzR!dd2D{cXO{;W#rT6a=j7>
z<{qyZy|j<>F7RHwM^zAG3nNb*c&`GMG6qQunmN>1KK$t~@$(yggK0KNvlQ<&a0g9(
zTj$&bU%CIv4$h$e(L!LI8i%G<1nUr~Y#m5(s*HsCs8V5^2<*OQb*+P?^HnK}YjOP{
z77W`>Knuhn*uQ&qr0PgT@wxSUG?K|9rl~xZdLJL(j(#GWn!&0^@PYK(ZXqI})k`JH
z%3wgRB1i+}gZH?ZX{;&A4LpwKlCgl=u_INX`=)p(fciq!0n%OuTQ_mZ*S^eHBf&Nr
zWTKe1B_qQK*yu@Nq@Kl;=5}?wQRg}PIwyq~#vhoUBBsMo$;y3|lj-^%Cf4p=6b!Zj
zH{Ey>tyY7H@o^@Xj6jZvRb&mTAwhR~b6*Sue-Q14fo@h|Vw>mY(R>P{NanG4{8*dz
z6CTT7oOl9vtFradi@D+bXOi5y1)1%W5k`D8I^imUQN{YY{AK<g-e~MVXeo9x-yR{`
zZ|9cXJPn<O0{qBm8jFrL+5f1cIq;7k%Gcic2JBsTqcgqONt6;SQ5_(Z?)XU%q;^h4
z%R)&CsmmcEtd*Dd8aWI`nCi}8k`C|x$l1K`#cv{&HmM2|<4va5-^$lNdlqke#oseC
zJHtq`Q;z<K!P+3N><J)2zU_GRU_wgjeMpF#Bja=@n>_!_Q~Bk#+gaZ2GxOg+V&i!i
zVpE69ljv$8R)mB>1-<ebstfy!H_&rRtJgz*`~xmH@dWO?<Yzqkm}7Y0D^4Ii_yA<Q
zO(WhT>Vpia*a5u=t(Kj2(z_QQ@1Z$ba>bwGSY=pCxT4+^3GrFKhjl_rm1H)f-%r6t
zlDr!rH|GXzD5%3|?8Ib{8PCmk-NoDA^8s3&Nz$xOquC6Edo0dL-Sa3u7$HeF1JQPv
zYFPd6vfxr;c>JxJ+ZAJmmG^d;#~6dEr`>9@V)?Rg{JyRQw%B1<T=z%8(A(9>+tA7M
z%1ome1(V@=(^z#hG^5eaFujb-^$UkaA7?g|IH$NPP{SQ8H#-|zw*ncSk;#cTv+rv!
zB(g(Uj9ML$PS1u-&X^FOP?F9_iOh@KFDYhih{1aruFLvwd<Aa0iDWiKjKOwVXfwg(
zqYM$H^8*S+g=45Lbn9KnHLFW4&UZakb6pCF<m-^ws24I1lY8xnGa;ZS!0or+Ns=_#
zZPf~zCX5w`3_&;NIhuRtZd+WxyX}R6Z<XfmrP@y{PV+~=OW4yy8f2X|`k+I3?AuOe
z3&Q40F5&0zIgQ2@SJRxC4vIcf`UZqCm_ZC;K+}Gw4qmo<AJ@eR;;!yNnhmlQBOLz9
z;}~1El<&Xmoix^OL8CW~76p^koR;T%Mr)Y0XuS}tD)hv?L+9B!L<D>Qp+tr0nHfep
z6TI;FH*&%$|HjJw_GEl?DHCnW>Q2J{oPRbCKJ*}zjIohU5NGfqHTLT9&PT<%YPThq
zRSxfKt-&Ng{32<N!QqE-=nJ01X19&~o6Gpl+fSu=*H&D21}!EtrDEEtH3Lue9&)uT
z4ST6~9<!xO``7Ea?)_(S_CLRi70cH0_><nu<X=6V{_0W8h!ARv;K@)FTSeD9x<Siu
z2(PY5Ro(Tqb}ee-o^^^zdA~@!Pn^=U4X*g`NBQ8BpUNfgeh=+l2APkIE*)?KuM{zP
zVQd7%k)|!)bmG6#OB1Bg3?y<P#7YT3mWkXssl*wrR?aeb|H(k{FyFqdwL(Kw2EF$N
z%HI00Y5^Qj^7}wBHprt*jEynY89~$!>?@1$VR7y9)iXf%{zx+GFiKU4!_f7DbirPK
z13!H88~Nt3&*y?yyqL{re}!b*6#f1z-bi3z7!&6&S%8qb+@s`77w`jaW+t?71)PHE
zjD}4scX29qC*A%FS5A_s;(J}JHDoSy98je_I#y9t+O=Cqv4v~cj77oqGuXP3%Rl#7
zR;k6$rZgH&un9@KjY%w-%VPHAYq>`5bX}N6RSTf`%$s!{MGhT4`K-fUQ0wWu6bVTA
z52<&IuUQkXo5uln-@OH=hTV5tJ&@JZ!-5V3oP}Mf#qPrIdt+hn#?%2EQwCIf@{G!e
z0jF3m&>kmUyM{mems4mT_CR{K{EDkS^nUos6(n1>l4ZSu8|5^r5dp@7@fkw?$5-zC
z`D~_if%~GA#ZGbj8Xchncw<L8^d~wz?w?-Gh7Fs!>?3E=*fdKv+lxtFbu^O&N#+fX
z!xs(Np#;_HXkmN~y<)85vW)5O49+%q!r#1vbI$)MOINSPnuNP=yOl>gU>!gF<`-Bt
z+M?ZT(jXtD45L3{ZH3G(h4G|T3cvbfyhX&2NQ1s<^T1dC1Gg`qq-&IQ#&GV-kEi8@
zRMU_PYgCIX&abb?X$szHFak7&Uf*G6Q_K}V<-cF|Dz5&oZ}89;9?QY6dp+F;9YMNe
zl%}y*U;1ho8@n-*cZTM(#q{+HQ=z^{uCP`7DZ5r$s}Qu}p&%s)L+p5Mnc^p3`U(%4
zT*Ad){5;*8Zo=S8sox>4bS9rDFo34?yD8^icq!-p;HR`(ZM@G&8cBi32XM+!jF81*
zq)=C-+NG+G=t-s5GT#$X6*~10xGU5-LW2fLc~7PUmwX8(sR1uoG59oP)rzITw}`km
z4C7ENuEn)$R@VTty9#iVAqiBeeHk903=jvKF8DqlJN8)8AN+`k>uzV{%A5H1Kfjfq
zo_-pwshQxuOF&N4$97$@$8GNsmjxc}&Swb%Xm=JH6X8r1CME+VELRAU3i!9Z_pScb
zIs9xlY-!ej30qEMv{RsGx?No<y$`ObR7h(<!`vF-nooR|u}#x-5L`Bm_m0L$3)5~F
znj57YpcEv?!ShI!^C1f{avP5O?C4Zz#So}KC{-f)S-}{bD(w|ZgWIkMvmT~qGtxe+
zT(Kgw_;cXXFjU_~hqa9si-Z3)T?UEw)!<M=mDr!>13`|}fHT;U5wf*QIpoAwv*e)<
zqq|`vm!9!nrq4N-=2V}o-$N}$QXDdi(u_(6ByEO|BUM40`D`@@K2(KQR6eJX)a-?3
zXOwjL1kXO{E&TE)KVtnCzd&nqINo4R6^~6AYTh_}$EjBjTHz*?BPSKbp~|SE2=r%W
zQERw!+ZGP_lRxJt*WN&D#U!JnEjHYC1Mhm<8~EVqZ>QIrW=Us+1c?%|5;(q6=AC8X
zB!%EXO&+rvilVe`v(+Z;2uGfJGPjwGWvbYp|C~F&d_Ku+pMJ2rtB8JROwK`Fln1qt
zs^YiI(z<&yH+=X%`RZHVMx!;#k*A!(>Sz8H>F!H0O~FR>qhJA~5u^#cET_IK_Tsov
zRTNcfF*+wXYfW892Y{8r>!3PpFBQpDEu0RpD$r=Ka>WvEy5%NTuHKFIl1Y5(i&HE*
zXQz&8PT{loROw5?8%{nI(;g-DeT)I8VH{ZFxR^wNjgVvbTH!3r71PssK%e2yLIniM
zi5yanJehc(2U*zq<2lDKB>y=R08p3pS-WOcl*}FMhRE-{EdAp8gIlw^q+rnb@T{2V
zh6gmX51oVN<}G~p^nc^9W{0ualup-S?%v9wqm$h5m9H{;(ZyjFQ*vIai6lZ@K=MB*
zHz@a7a;*y5?FLyoA0tfC3>$~kjon4&L&eI!78|yAXxa;!-MPNI2&U7?caI&6?c9^0
zp5ZT5sNluK-}fEddL!4J`%Na3CW<5V88o8nYcs*NlVYqOAa?mEkS~xYD)&YeMUnYP
z%|n4v+0MM_kb5MbuP(GAqwcquaL!EwcH1qWl*X`io6_q6&am6$iXib<y@pb@@CQkB
zvf!s=F%<mXcNykI@dv!vt2ve^BpT#4)q`e8(xN{;!v4qq0|y`dLcE*dhR=V2+duU=
znj5Fkej4~S1W|+x6BBY1+?E49c1Te$Y+sxoXqkWLQ7fuPQY4s-$)&L8T8?_toA~~h
zzQ~sEol9$~hjVElqEaHx)zPf^-lH7fs7P!U7VXt!J;<c051@@0==ORfoerP->Uli;
z@qdGFEn{?a8Ka4%)!oVu&iOn?J?rs^?=jjOEsW5LzN4uUz@?(ls*w;}?uFGMOZ8D1
zD+)q18YA$Khp_gU&tScq=D-zeIQx{7G3z&xZSB@fyaOoXF$iiCQiNWbF*4JoaqZ<?
zc;c(M<f~uhVb6RP4}aU+=|A*A_{k1URhmBhuI5hRVi62pHM(%ya8IZ~yg}4bNLs}|
z;hwRTuR#yVn>aovf-{0Mb<Z^&j?9$TQ9a&T=(Kspzn;opy!Bsr=Epxqx^yWt>>P4)
z0VDZ%er*!GaT+a7dEc2_|C>8V;@DU&?^?SGS}}gpx%aWu0&e9#6scwObDtcoKexF&
zi%Kl?!uYL;_lp;;6)GMHs!#a{mgJ+>VZ>0qW3N5d6venyWPyKQjc*s%?|;ECbHByX
zs*zb_by@Xny5b6wU)@4us*7zv*Y#<%8}zr%uy$mEOTTe;AWC_y7;Q0U9E9xzxcO1E
zd-&P<eT=bS3|<7r#~Eg8Uqj7?nZ(pO42otBLc=!BD+{#S^Bc@=2NAhPwZh;Cj|Pds
zBvXF&!4I)I75r?Eq@B=pX+V>bghsnTdu$w!1pRZRg6!uZp|aKMTsDBt6M$siW311U
z4ho!>#ChiWKuJzuG$dnV(CMHi1l6f+)2KJ7^DJLF9ymAU@B>o9kP9pp?u9~baqS%B
ze?L_aS`@U*4f#HHB6FnL5R!f`7}KP`WQ6f&Je7lwI)<#NtpDCQ+;ZlB(AuyC>330a
z7!zPvV-hqf%#~w1TAK}LL7j#<y^x_Ggyv&9xw%^uOk<)02d(2-uYC>Yf8xXFkAFmS
z>ny4+DvC*k(=h)vlFIg_dH1A3V$F+(oJXKOq;E!)Ubj!X-R6z&d=D>u?VFLw<t$mT
zhNRcypjAt`;2U3JzqLykv4)myRZw+Gkz_5M=_{VVQRY%>+!P&9mlBC`n5sa7ezVD;
z#~;m29fKOnpRZiOcmDAejCu!I8U(c@2=4-54K4$=rHpLa!mVe0gwMVBl{9>l=bUyb
zYo7N!`fFDrBT2!laB+V#QB6tJ;fP&|q7&nb&&hy<tIipH(c1nhk#wxJpO*Q21WRCs
zASsLpE)jI?8dg2w$>@ru=y+=mP*fe~k&41nFU|PbHMj8ovp&t}*b;jEJ~n9~CSDIy
zjqv&P%e(P4m!o;h;ykwy9Z)mMfle8a<Su?*vFfm-G!n)8KB^fWi_du^@esqk;Vb-y
z;G?15gU{G!?>!&_YBGdxu?Q<Iu3flhc4@&d6hjrk<SxP9lO%?j^&1&$Bsh_v2r}Me
zw&6)pytUlDX$xcyt3fu+MW@G7fORlQ7)+MsKKN|mHC?w?L^u_JW`p_XhUeQ<2jmZ9
zAPN5*+>nhi&}>$a_Z?}D-1E=wDI4ada-Q96kL;=|x#NF+#Auq)Y*<7+iMOP)ecFvS
zRt%l7QBv<hTi6G4yWEAdu)r<k4`pc37(9bAVDUCzkEv)N4AH<SMjBMmh^XbGRt~VE
z9cZ<~25$^^+_@1=8l>K_eC22{$4te%&gY#921XPX4oNSr`xf%QU*0|+P4a^UW)QFs
zXk(liHzHUpi8olMG-=`*BaA=&NFMas6PO(}Y`N^G+;G;J*k9j<^!j)ck`SjpI7Iv4
zWE*C~m7%)tgWpiu$#YuvFLK>fF%fn)f?GL>efT5zv*$jKPrm!TaP#fR)-Ilaq>7P>
zYSIv`yQ&0pwQDyxIuetrTwtP?&VU$tKH-UfeJmgNe_vp7uYE9$G1lMl8=n2NM{v$(
zK1jpNGSMCdY~U*prIA&rl735v-D*))CT@OU2d>kEgJS_AG@B#LbS60RttWBSrn?!>
zJdK}U&H68Ym!y}wKj)KImi}~?MsF*vpZ|~xUUxi~e)Y>d>E$owVQ+a0-9sOMTRBP7
z7#avhO-zhUF#=ou-3Djs=rof{!iIzO#SUNU2KhPyA1^{)EJ##cG8Jj;D)=||zAB+{
z0jvr_0}_iD!D>nT9=tbW*aL;krS!z`suNF!)(GALjQ~t}7e|9#Ww#fpTp`c!k<e1=
zTm~Xc`6Xu_<RfSX;}gU;Fj)f|t^lGLl4Q94CVJcMq|+3XK0d}1Y!bdIBn}b})<=Uw
zz?u+0#b@lfW@VuOR>TlpsAALNT3q*j!7zP);e7>??1M(P0eqISYQKZnAO=a=m{e&w
zL3={hZg6KWWA_Ih0XDb?8)HMk;eCN3)B~EbeScuN$H_!ZtRw9OK31+9+H5T_fswfo
ze*SYV3xBjw{0wc`B;lTupmxX<uc0U9RuyJMecC5+J-&O!8LZJ3Hg%}$qX<@nwBN%d
zhNRV^HM#^EjmW{!g0x(O<2dUJgCj^X2W6atd+OZIL+#6ep`FZWNmO+}f^={Gx7OH5
zIKL*KcbhhCL0XVzT_%@B1q-dzJk@eU7gDq>@)>qbXW^cS9^6}nL+#!Ztrxo@sF>iY
zTim#T;Gi)X%Em^Jhdzjhzve{z$`wpqb}84t??WWF-G%M-apDS;T_c)XW7h2GHWOXI
zxvD$+)?A9KScnfrc7OQ@tDf{!{^W5_<;!n9iR7+L=*%pRXhtVtT&ALeF!!8EiGSxq
z;5m1pivEiVnVY5A>Tu1izu_@Y|7)(e@n)8;-i?W|5&E0%;Jt5uGw(h51ZHQpGC4Y4
z09JW|sa0j4IUSfXf|ZoHx#Ncg8=&A+M0rvJ$w-^zPanam$3L2ly-hrL<sMvq#(UxC
z*P=5plR`S9uw{zoudm_ali$F3Z+#ofN0#x7Q{KhoUp|9u|2=4Qnpk3yp(W05hF-bw
zeUp=wL)pNf=Y|fs)^kF|phZV1wlp-X)S#1mqckO2FUDdbcv!r2RYu%J;<9*B*W^PV
z{tTC0^($JV6QnM%<5K6MxJVd7t9K{rYEN_a=iED^Rw7lEs;Eh`R$ZZU;&Uz+n;6h^
znq!~$*PMFN>sZmTbgT)-AJv__R{VadhT%SPO;l+n!ajTN5jk>|4wuxJE{j|I;@VYf
z+5l7cOQBF8zY<l1BI;&^<^%U<^`AeI4Yo(S)uGY#GzD~|!-jT~hraYhr0PLz7|+R3
z36V0;?r_5IzA!!!G|fWb5CTCtvGYi>)trzR3PI;T;yvPA4c8K5z}lf^M?07P=#Gn~
zLhg?WN*Ox<?jH>R03ZNKL_t&?Y0AcLd;`7VMp`~s<hFpJVJzFWY{4{|B&`;vJqpbc
zGM@&IgIsUh`x>$sI9`5;5MA|(X^kx=Qa_agRl=MrphI%_96c(+n<&m_G&`fvXrdmH
zM3|nP0jr3ljEy$q3xs)KRYQQhOqA{Oa~yORmfS9`d$+S7{4Nw^gG`1D!HMF6jnzPU
z2i^$YhA|wkios#=INL-IK7_+hJQ>+%Exqe+<nnjF8-MxLw0aIl7V|B|RqwwY)TX8D
z?)LW!L52$mP#IInm3b7JV#t<_aq!<B#o9G1xai+bqq!vuozwU%8)GBttu>{y+FK1J
zlb4&P2hwboMx(<QzWGD`>?zOZ=BbR%$~7d?An9-AlJ9+u=RWb_O!sCO>rCKX>92s)
zO*M6BQD2VXsFWzQRyLr#-(5rIg~%y^rb+O&!5_c!W&B!%%x3J*R(^Q$>qvUSsD<Xt
z-Q4=A4{`pRPQcv$YaaTS&*acozk%N24@6fl4a@=!t~|M^TTMu3X9Dp~MMxQ5W6pW(
zs<O5NsD43u4>yQ0Wfc-;__Yx8@CwExAWFa6=NG?P&%b@}<8;P4^s^o!mLzH6)r4sS
z4GfGy8&ETp?<chfvkodrMLnrBQ!^@|v*JB!9Nwn*B*PmIHjL&Se%LzRapJ3a=2ITe
z$If~`u0I<(eb&-XQ^Z=lKx!e2BgWpu-i`|1Ioho@D_5?J&%M+*4AcjB7uVw26>Dmj
z7Yth2a)}jr*dPMnU^Ss{Jb(VS6Pek6Uw$#Ql?@Hyw%#napuFs}pQhImq*-o`<-@ci
zq*m_p&*rnV$XvykfT-{ANOb!fxer9KsyVbV5+r4JItkJRey@^r$Dc9PYp3969o&8=
zKl{YTSz{8k-$O*{YmiPARb0P^X;`v`V8$nDOf19M1Yemk9u{D#$0qa0JC^=;GDmiC
z_{-@IKW0^*JAZB48H@70Au;IA&W1@`BXmYvLzG^2_L?ki`n#@hxUWHj?O$RYbTQ9I
z9BRF`x`s=xsA6mYRYFZ`aP1b^+7&$DwQpeLh(qb!dM#J~=ZDDt=chDh`gos)J<=;+
z%fxoj&!BROiX=D}C(!x*$W`hnZvOlfDVm9c+A63|kaiPRjPS6R9LL7n@8G&meUj$Z
z6m|Xb88jw*y!foiP=&N=;wL47><s<hERAM|H@xF?jye9-Bugh*vD+TFexH3;FX0E@
z{3^>^o~F&{v^r!y7Ap*SA01TltM(unz@mLYTFqzMwc<UXRg+2ypU<>MVa;m(_#cns
zXVaT$S?FE$GxQth;?Djy-~9VyxbAD`vfsK#^5ECLm8H)=hW>tQk<o;V(tDziVk&qA
z3m0%+IG|Ly?y3WOCdGYP?YL}LU?g_d^6?t=<#SLd@cA;}2?Bjjs+QND^mbC;2%=9K
ze#83#m~g-uSw+jeeNF*5yy~g?FyuUll4C~%#!96rOfI~VvF~1cllDEX-($_{RkV^;
zcz!cS(n!c$(OZiJaz2g)iM6P6tXR5)@$r$UzE?M&TZk3tVpLgN_jbXMh{(2`m8`rs
zBdAoiwt%l?G@4ENV@o*lKR=59<@H>5$z>dI&^neJaRjs4r`;JT62=lqR>V}g4XX3t
zdczWkdk=lo%#~=oHDK(}_t2er%)QI?<jU5sNZqU8x579V@Oh$|Q7-zF_e8xaw%g~L
z&wqy2=FK!yXe5TrWh9c|#bX<mUav>mcZ@8VB<m%Nt=f&d@4ST(6Xgh_f)uf<U?dt7
z$)Fgo3=L$e!zQ`3EFeXOqs%o@^#JRoVzTE@l93kJP#~zn%uEk4328T@-RXq!r5gW&
z8liu_;JZ_W+v55??mi5F8BzeN(QM6>BO6-kziVw^B<z$%4C=v(4-*ciMR(OI_WOrd
z()!egx$WCO;OA$a$w8Z*&B$Y)Kz~e08cm#|f<8-4@v`=a70$D{nODVUSqG&wWlpkH
zGo_2lF+tren2ynI;rG~`r~cEc`P{osVfh}rv+Su)Bb)G%7z|bf9}tDgXRo}C<Xwok
zhZx0WFq@@Fqr-ELc{N}E?gdQlw-=+MW88k@^}Ob|qdDniFJjyDHbzIsqSUD`jNIk|
z3|iqV)Spe2%5iA@?r~A_GAHTQiW@0iz765?W_FAvPkT0#Kf8tL3%|volgqi_#D8FH
z@3pMmeNXm&#*<m{_$Q((mqE+WjPj&DH!jQ1k7zCANzKUao`HKpU7R9(4k^n+EzgNm
zPEhf?bM!;&piXr>NTY*!UU?Xa_pS4EJ)HTO&vDVuu4Kvha{Adc))*v7aG4{)#8{=p
z9$cw9%d;U+l~m0q@aCipLrC0tl5e78xS*Ka0oxml3YZ3$U2!!tY8YL<f-jtXJ~Q6a
z&-$3w2$^??F@-yOI5HQz7$MGNnz45GJxEL`%~~}Hyl<lY;#yqWxtBz?CBWAECs5a2
zIE!#&VnA%RajgtJ{6G$V@B{EpNhM>%j+Vq9(Nf1hH_w>^)9V6?1ihC&6R68VGH(Pi
z7Hl$57|ng6i-0Ke<dv@KWSygTHX7E+g3j^wB^B9zhoK^YYT|44x;@&rUB^x5eVaX`
ziD<7hbF)Ftuz`|zVP?Z-I!o4K#zz@hxg6IpXsQ^MNJ9(KghV7LCp_zr<Cyb#@(}{5
zf9krT$y`)sz79{Wq~X1f)MsdpjMUz8c6J&oie@QI+bEP<q;lM4IlPd<U~x;opSD1T
z&jIRvu-^@I8l)l%8m7=IoiwQ>-+3vTyVVxSjSq#YZ8Yev7-Q`Vk7dckDt_^?PjdB#
zKhA#B%H-3ZM0d1HquD`SB(X-#QO%PoEse-(_iRogE)r#3_+C;Q7mQsKABrQEjba{h
z7=Q7czvY`B{shn8V?UAy97MXLiHT0e5&?w>&`0z-x5Um7?kp7Uvnh<USU<Imr#|7Y
zxpn<CYY*CwH0_aX`3*lf`_mk{&uY5qG~=V=LFyDOgd4OjggT!mZ~3(V^-DgN5KOz2
z&>=BKvAHy>kpf9K0^Y@QXMqg1+JqyHJDN*>_Fe4yY4((aZJFbNuRZ}Du#U8qARdyS
zn~R#5tX7IkDEDZobr?#=V*M>*bS;W%scKbl5mYK=Mclh%vM)vVSfW|l_D6&Z!g{XE
zn|}%i-F}~6t-p(ZdCv#wOst}pb}<qWe)YvY?N!l8{4Y2fqNRe-9H4SAZj}5vtw)ne
zu^-2iiZLe(acl^0G1BJR>u={-#~e@J&2rflS23|<6*g0RrbWSPqROBzUF5<ff%jx-
z#=iUPQF=ERi~3Mx);JYA=p!sfmc?~1uB`^xa({If>i=Jw4T%WW3)VGAY&$%R2CUdf
zk}h+|Jox&k8!Z(H!Ch5%*PMr=D<}wJ?SR_P0-xx?UvR!;J8P*TRCE{?^ilM;{&1>V
z1ZlzG@w7GL%1?frCEXNpk+80cv2g>BE!&n!*z%j(8SRXaHkIaz36il6sRks-N^rFl
z!vNo)HoR6!*cHu(Dtb51Y_{$=HJ4!qJQZ&Sr@?qGhq}Z-+V7KCL#tscN-D8yp*m3a
z4a)W{)KPbF?Ya)bE)21C^V~4ktb$b=44vkbniWJgfA&rIzK?sR^`0gP>C&Z)J^Ps)
ze%#+dQ@G}fpX8R0|0nIfqu-xJY)o&pq6Z&Ge;fuU0U-~%`o2JkAvgCbr6h;VwGYrq
z3!Gk!7K%+u7P=%O4R(9N6L{1!p3av}do$en8)RDqY73cGQ*n<FB<DzJU@M3<bZ2L1
zjI{Xn`9I|0k2{j}vl+`)@5Z+E8#!XXJ-G7w-)8@n9TH7xC8PK_-dS`tk(gXopV%D>
zK~y^nior=9f2h-IO=)tO%aI!OU@d6|iO-mrY{S;Oxx=OW+9Y@}Y}>FIy>%1aH2Byg
z3O>sy4C^YAuZ5YK&fQ^tdq`|x(my82!B{K+LaA!Jwi3A~c@Z<f9g|^WO=S!Z4#^ts
zy?E`73X&;C4Nm&>8ixP-x)VulE2v)?gGn0UesuZ%?U3TWmAYhh&hZ#_rw(?9D|d9^
zzMS71-cX8S1df`xtQ^->k5d>KU&3Wq-^eeo{|)04D@pqr3q~Id^?Vq)3ZqgM^%jw&
zXmw<1pMCb;Gq7^B?Ck7}P`55dmc@1Nt}XXd!7vZPK*fV6a7ZKhRSJ_ZPlT!}1N}a>
z&V^yFy7z6}Jt)Bz<}TyRs^mO_g{6q|^3bAcLQW8$BSA6L%wcC8uIE3-9KjKSmbgqi
z+*Q|b*M<MXq)Bl7E=Z7s^wF|b#Ne|4qD^hM6RV1}8l<BQnk$x*S;bllC@yo+SSW9~
z>bz%ohgWZq<gD{`P3Au{r8;UDaci{NAz6$6;IoXx8mu)KW9K9oy1+Y6?}Phsaov-r
zc-M#5Dw?geilTm4Qbqscf2=JLZ|^`Pp+DZC^^_-b#4BD(zv<a<;RXEaL;u0ZwrR9G
z1J*}CB_=&e?KR6ZzZ_YrA@Dl*oxx`_FDfd7Ecr-<l>$9R65R3;_Imb_9CpM5`N<il
zlialpKijJqI!dWDl#iNv0c#MGuz6~Rc4w3mPkJZMdfu_v)(FccC)s+}Z+OK^U&z-!
z`aVc!STZ^e*ywQ)3AQ<sU6UlJBg^xdKGuD@b1B@gtVWBoIBuq~uq!$&-ebg~Q#~5f
zv+(O{IPb+T<KJKLdY<~ucd+jN{2g}-tY~&P@83_O-J7B}GfQSeS2;fyMr8=XT<v7t
zQ+`ebkyk~e^rCRxE)5l$y*_5T8^*_^yjE47L*ag{^-WnSZrlZhJU2OuoiG{}TuMK+
z{Oh|vz~xunh%ueeIS?(RtC294mK*0;jn3Q=K+G?-YU!Yzd+?OL5;6cMmY)|XJK%`N
zWf@{U{r+Y&ohC7T)H}QyQUfWTUY3$H1S<_(Di|9k0z!VEBsPeKsL2Z^DR?`14BBh2
zJu9JI<`T(waD8R5Xjoi-s0G8|wx>A)BoUz!sZgj)Z9kL&lwY=!n%#F8Wo2$sN+;v_
z>`Qe=d3=>n7LgJe#2xeZsQtpdVCj-~{so*WK0;b1nwgpYG?#wvQ;ek!-%n}TB((U&
z1Vb~lFbFaREPZCy-+@gMbR=QVL;r|bHDo@ZmZo8FF2kuGUJwmp^E%I|)A@=39Vn`#
zPI4a3sVw~<k_H$PO@W}>?UOVtE=`F|ugU=L!UMG|77D*(od&(1-hH0;)v^^)h$w1K
z-ju>+*dQcENHR~;3)$E>{*NEdqfU4u(@C3cKlvdy{rh`KHqMYucTpRQdt&3zLxLyq
zo<#asQdAX>0Uw-zDT)i7N18wjjVGN`RA?n(v&9&TX}0N4OmOhAf5)b+8@b|hAEGhc
z#bs$xtgGa1xv|4)px^7#2N)aY&;Ig-y!(@1V&B6L!_CZMdt3S5e|?769d|U_W@eb^
zjNw%fANESIxmd#7J!a6bY`6hvMGs#HPQB#kB||jDsw>1CL^UM9qKE~_47zCwJF}V1
zU;I3uJ@)17yJioLKku7NAO5GTI_?$t{SP9W?XlbK*K_kbPhm;Bfs2M-U?SQdP0V5!
zNULNn+1Yzphjxq<aOw+ph9K$zcH`ZA|Ag1@ndd%>ZRdWIB<)8)FJ5ya<$X{Qgs5>A
zDxMBSm(3W+9O8Ybyy!8V^||QEYxuxNzQ9P*0ibD{Sd$=55EnbtK35g<1;0O+FSXW&
zBFp4OciE9C+?xjShyuLKPp2z+ajit?^)~XHBcH%A&pHy@o26-#j109(<~+t)j0P~t
z#JFKJ81jj?7;mr&#)Hw|(&2p;Vr-VtYBtzsuf4+W<)EWTZA@-Q!04h__u{%Q3Wm+U
zi=mr}iHd0D!5u_;WMJEu`~SF$U}Ve!52%hc&)Enn$Vk0n(1o=wcBWvHdO~i*gn44`
zajIm!9$foN)?aiHOKdwx%lRy#h~Yxz@^v9E@HCyU;l^8;7#YD?rM+eqW?}+qCV1z9
ztf2{pH6;OG7|uV?dBU9z(M{h`Z=xK@C^Xd>R7+88Vr%c~B2UGu&LeBPP^sHR?B{pC
z>#)7=%YFZL1z;?-uZ5&aLD<Z7;KBF=9UX;3*YT*go`fG6W6Mu2<=XeYo7SCoBi(Mn
zNAQ3(2Jc-!kDV{keCjGSJCbpnZx5F0XDAZsSg3Rw^p;KXq}RWeU;X%d%v^9Tjcr|U
zA$f8>?xWG`B%R(yv(@H;pZtu24}T=Ty8RAz+j9>#+<pg7c+{Wpqw~JOA$zaJ;27_8
z0vAauNtezC;^=~Q>7uBp5ocrKTjf=uVV^)rVkgya&0uQ}x#@b&`Nx0ahhO~$&pZ8e
z4tV81F>3@n(jc8^@u&av3O2V#+0Bk{!`Ht8S6vI;+0d!TlU^N2YzKTRb}mbt=c>je
z+X#sFsPovg&v#CJ2h;!k6AtV-F8tX4gKe|;EEbG$WVo0Y7~(@yp!Qm7VXF^*Ok%(!
z2ApH2pE0e%OJDg$q%n$%qp>Q2^Da^-f>W|7l6w;r_(X8dN20KgHH?o+Os1kxBo(rD
z<&UqP8>tkKLU(p6kAC=rIpghb;O!^=1FwI@@od_-9xHvk??MvB-YOMR*~OB)*o#M#
z3${eQkhOxkjPXv3J=W|7-eHZ!=X{=BsqD767FT_3Ho&IehcLYmU7=l_ehnuvOJxJk
zB_8g<Igq(YLlt}+c2K{K2c!;pUsc7ZVlWs&(%j5?T=I#JvBD{urr4NunE1J?C6ZVn
zbDn0Bu>OWyu``}_YXsMBv2x#im`=N>2^*z&1~%qlof0jRrAnbNAAMcslR1@YcVRBe
zp(MRk%t@;WRQzJCEpb$>zUNM9U@evnztdfZolh+9mA9LBPOk1u>kW7l@NFM?H=;f`
z$96{Xd#>fdC;c;{4|pJ3uf2jR-}f%;ZR=rMFK$nUo^xOmyjqMjP)!0SEH_6pJIX-o
z!;~`Q8I*YJUgPl^qEVVBnHWd+-<PMo=or3o<_F=|zd@#Y(fl%uo~T-;rnk`;o8ZKg
z-^){;{sNXPUrlFpl#O@X&fDMedOmf=yJ%~lmQ65ff@`zRMP}=;Lv;*G76_`u5h?Xp
zbu>^cg8Hxn`fyQ5Yy-rQc6*qa9%l1xT>bIW`RJ>TXWfGz%G3Vi18h6+LG+h)Xf-Tt
zK)i#8J(v|weG2QnXVQ#w-aFqztLt!?EA!Y2t*xSEO5Hq9h86@1!=&>LKzzHH;L<R(
z5XL(RO)GRpJ76VnRLE8JWN1*9D&xOE`b=qR=X1FwM`k?I9_9Ggzm+>SO=Il{iXmx)
zIe#P32u(aHF8nECQ4-BfO_alS7~>J|@S5R85tGDI<jYtf;dv^!I`I&LYq~_wC#5~o
zX658kTnba0XF2HL1A;n|4<o9%mye1sI0y#fSXc0SR&jU(P84r~yGlXRc0G1qv6S)7
zNJ(U`5wREJ#^SnP)+PgN{NES!?}OLR!T+dWk-J##OjR)NykS*7Riq;Kr#tc4b-Tg6
zhUZJAKrw=g5{G6gW%j3+vGL+dSz;RqE}&>e=H@mM+`A<)h*y%HqkH?EG_1vSMp$+5
zLG%)f!y;IWF%=Z0dKgk6f)2VTONqSB_vxh^&(;OuYlE_X6*toOPe~H_7Nr$wZT9qc
z?JiwhyAIa4PYZ}V36#2ulH?;kQVv{rjfv|>2e;>L9Ppas8GGPi%-nDTm!0;{G=A|*
z8dI})QY892<YR=s=-4T0{E)7m48`7e_&HQ+JhMFKgM@T!gyth3%cGz1T)y<ruSf2<
z1*WHHpr~`fWNl&zPkh?5c+XiM=dg!7gl!u)G10L6^n&m3f@eL0p6@d<GJ<LtXDX*O
zwM2U>q6GD?X8zHzfm}Q<+%$eB^{&$JHTcY9444^@Y?`65Wrn-IdoEvk)d}2m#kCy!
z?)S6*i~f$;)sr-r+F)}T=6^^u!FAfKJML(HJ<?`mbS2w<dMPvKejhvA!%INc!=4*~
zlU)9`rj=Vkk3?%Zy2|$M>$&QB`3{^IAp;(9^wGF=`>}q-IDhh@qi|yujE2!5WMCg0
z(rp#3IV|V5kC9vg)zWvuM?Zfy=bZOHv|3~IoF_?Ik={@iB{JZ=EMh_-DnjNnOcTVU
ztXeUSrO!mCjrSSe`{<5cvqLRP$2tg<cM0O_is9Se`ERbe{$@IpBYfg>U#7EUY3P1r
zu5{VfiW@k&0-#nLtR;>X#!<7V`sA}dd+xp((a`;qvT?O&)V8?p&$ZErNS9q~-u8!_
zNJ#Wukis!^=#bI8)xn&Q6(wUuOwsZUkdF@=)b@&tok()#;|8l&tPxz_Vfr&%{Mk=4
zX$)i;;(c)7L}IcZkJUs%rImoDsM0W&o36T?cDqg5ZZo?7-q?vTvM3XmWf|&x;qF<U
zHm!&N$v`&0+6{IJv@V1(&j;F(CRQD_Dd)uNyckm$2_?1>h>OnGIutuMe70CB{C*Y*
z0}`9Nw1CLH8Ymb8QbWhJC~(CXL<6cyqfLM5a@M`{)$H*XPb2&FZCv@DcfsX9r@3_+
z!9^3zFal)6V7$k84X+Uss|EMX=FPk7{D+(rtO^M&vazM?`JAIzvUHqpy!UO`W|Ntz
zS=2U|F-;!u*k^L(ozv|5ki)t4mfLvJV;;eezVjt^8?&?%Ps23v>Z8_#l!<lG_SB(f
z*9+l%)kv_F9}m<HQ~X<I6m43vsafbv!Od52{;4PM*?&Ei!yolnp7Qp8XJ-F(_*Ij%
zjHN*w_aUlN8INtZ(cM>b*s(9*x{Vt+V(B<Pc-zUC4b!Cke)R6B_H#{H&8Td@X;3}(
zS|^ve_jNw%7B^WV2GeX|j(8}KdG7~!+y_3w$fKT)+6fYG$?XWV>en`u2_gd{wQ)Si
zhU>L;w#yB_Ue8<J_HLTZCG`8AMzclE-%!<{s$@ORD&A;81UIfEFgF_sz1b;V^1|ov
znX}H|(;q&A{?ry4(!fStBPSkbf;w1!kR~cUNkxGQJeJULS1dC$_^TJZls|sT-|~&`
z{TNIeRZk)yzGBc9Kd+aZ*lfx>D#n%ka|Q7Nk5kXS`|N|UrZ9^uL%*^ho9@N6xbEFG
zB_iGY?!9~QeB1*Ru##CyoOiE_RLE}uEf2V{N+kdm5m$E-ahR*0K7XQJl?nyd>!Md(
z!qin)Fe(YorA1K`{gzM(YDzA)U4v>xNVFRXY`FPGnCW5Kqv-e&)~tH~T{SpDf}?0`
zOV@*YZ3en}I=3SdG7ROC9n?=nsMK5&JN6K~i%zaubQpY`x%JLh3Y#6LQ;Q=>zwZzG
zQ0V-=yP`u{fVrl9jZq-ECkci4f+r;U;su*D$yP6A?XgF5z>}VV-?W8G-urL#|N9~u
zGktvK%1)eCx@%REn9henFJ%5d9FMOQ3OU+rV)B$U8|aAP@ozYZn{M044?prTI%A_;
z_`RQU&_RE~T{Ft~?z?l_Z*JlAQ%~lj@A((9sVOEp?SLx#QvO+eM4OAPWP6%2I#eV<
zg?Y@+$`KKhD>sP;E4a*Iw)IFh-^~>tI*T)3@j6DEOF8bNALW3fUqE-|2<^!RE_OPC
z$(RY;uUwMULu!;gUvLziLk}gJ>an(;apn6y$cW2GNn<l1AIbC;pE9k@&d>F(seY+C
z6DUL7nWelJJmCU*V@ca>{OBmyka}Y%q-v!$Dq1fJs>&5QIKJr})4gf3#t6@S!AtRW
z1Sb}ZA@f-vnQQ3YgwdY4U635pHv%r}GtudA^ij`Y)kKpA|Is=Q*ms}6xkC^eVvdvG
zNurgpzg$?rYstA$uYn7r4UiU@ZE^cu+we)7)G5X$sLLfDCHhZDC}?DG#8!&hu23E`
z@t#m+{{s$ys*87R3n<w%i)(T1qIGwC%ohuWdn=a^iSlVdVZBs(T3{L?Lox-4(yU@a
z;_VC5qB1!Cw#c;%bSCB`;{;R*@idL&vQK}OWj-(w8`cI_Z6i^|#6;OuDULa`h6CDX
z2{L@!uDg+j86j)8*yF)RFk_oIg8Gtem}pK_A2q6zk!Aj~wH+?wTqC%-?cH1oH%Q1*
z8pvfX1m_q$thIOt#x%$rb6ZD4_=h_)Jhr%gp9+Wj_=?UUjSe%KtBl80C{tpBLQ9y0
z$Ph_L$2u%M>IJO(zyF8SZ{w=ZeuA0vFQB<?HtZ*f#c4o89VQeUI$zPS{Q{@@6M+X6
z4$l-QZqSvK!eYxHoiY5XRlNLTU*Z=(zlnc*!gF}p3trCt`|V5Ox{T82r{DQ6p8KRn
zlJ;j99UH-CX&@X&DQ@!r$Ja$PblzRIVon#8`E*bUH16+3B@qd@UNMQ_vG}Q3TGN}E
z`TjY4;bkx8hD$Ey|Gx9RJp8pMF}wTzq?4nxTa7?uRgYIk&eqkEECqNPVo66v`HR>5
zBkN6{r6WtZ;lI8NS6qkdX1ECF=N4eT;=rsmpBk+dnhT9S)x@^SNf10%Jxw3ak1Z2z
zACuielvcx)6_wIRCw}p{S5dq(#0Z)6vCSs0dDFje*HkZ*K}Kjc@{U04Lg28t;MIU2
znrj`!*B2%8DM>3K?RVL_rH5+ZHKeJ>xp0l0Y9La_*T~0x!%bxB9zU_m?^JQABe8}o
z^`IvHv%#qgoQg1$92A&TqY{!}iiSm=?E6^kWr*>Jc*J_FHT=;*2gI0KRrFhgnikjZ
zM!~RQ@qFBKAW#GRT#n0n-QiICAWk<&X+!~I%Ta?O{wZPS8^}>rsS9=v_M&C@`xOB^
zH-D|*H`9*(<-cUp&wkEi(!jeOsm~%AlEM(DmNTms?QCxZpE^dd-2CH<X*xsJYQljB
zGqTVA^pmL4l9iyQR<yC_AgKdcNg+3@x&trK9ppGw3H%7z84<Exucp$K#DW1y$h_a~
z=u|(U001BWNkl<Zt5>kPoj$I`^}G6{-w#(%=f0mx?p2T@!7>EV%`-a_bncD8tr(;A
zw8ydTsOOWVDOZ2uEV|!3hsNe<eA**SRZ8Iit(DQg;ZjJIyIkjX^E7(p_|hylCVII>
zf9bf;O?&v(7|tYIebbFxaoOb@vin;69h*7s1xInw*FVXMhSF{-jYc~Vn~jMCU`51+
z4n{sASvkb2Ai<zrW0mAu^`sC=$71!BJ;&!Mp`tT=?50_g>wm@huX#OR`PcXGr;mRk
ze|6eB=^t_+>52}?NP>tV^Dd@}6|akem|Ye;qJp*(<e}?W`ezSg6P^QFqg;5(JLt@0
zIM)w~LJ-O0Qfnw^)g3MG4phw2N`yq`i$JyWQ-D_@rKGRP4fe`wXpjomrh08FaSILO
zOF8Bh|9N&w`P65=&X@n|J9upc{~sW8X(Y_M(%@4h-j@~JzQSuTlAuQEJI}k{{~ug*
z#npW5b6@A$U)@N<B-lvKHk6}sd2&979$%}w1Zfyjb-{I6gXfTl1Y=pES!>bR@r7?x
zE(u?hsitxsHF`KXM^hXJAFwy-(;Aky7%LXn{kS#&B=IkIEuN2imO~U>;6|g6Pg8Z2
z`fLN6^Fqz01+X?04Wt464t+~-k&q}dWb+?<BOQva=|FJsQSL71XW<=2`&|6hFEbkF
z_7h`J6O=$49vg#Ps8xJDz(g{kSfy#;H#gh}8#iMo*Wy}D);;zyT>6o#SlO_6r;*2C
zOA;HE=4&htd?mX#(M)b=*Fm;ZugPElt(Y}w#q&dibf#BB=#r#?D7Y*m&H6yI9%S`?
z!=o41{nBC3UH0A*g(0iyNB9?C6unZ}l)}uk2BR+@Bq+UDvep=rPka^+TDlKco^?9c
zede?5({+p=`Bc&+2GeMuK3WJEDmH9dP$jne`L#OY7rok~DqfKLD!5|y-erB-%{D)~
z_!?gF{A1CntxUGsOx<_`pZwSdS@)<%;`=Z%+97orN&W_=0)1&dP9$-rz8s8E(Rh9a
z93rIR^R9E98Hh->*P!yGR)aHk&(m}sv*|9b{rqS7;a9)LArE^bM}OqabeFB9JJF)q
zXcVZjmry2zOCnV5?I<`RCeYUKG{w@lhC^QYGS0c|HSE7>3)_Bt85_?2AC^Ar(WIk}
z1RJ}uxi?3AkDSZqwO~2K7cPpF-0W0q3cb~lqzX4?j?-oN{Q~6iS`c*e5gae2!d$o)
zOprL$(&}8bM|Z|?!>{k+gp*IF*;!5x7x;iu^4g39U@3op6TLO6oD7wqtYw^%&3Y2k
z=3C#nkZ=9q2PB%%99>G9DeC$`BtU~9Nj?e`ijcfR6-SY@LJ!djxR#s7Ni2T7#|q`}
zU9PsICKA2#sR0wzq144QZ4oWilvYemuw<-(_^fzL<_50C(qVD!(sft0!?1qweB84f
zB7b~MBPon)1PC)LLxSr9L`8~a48{aKY4tuLFCMb~97JrmGLx`f{zBzfOx5pHAWKuY
z?G`rt^kOEYiDoHy4QPt`Fb5iuJDCHgSIqZp2x=WHcV2u6O_R_YZL#D5htOEMJf!{_
z&CyJbxJhkNQ>1LClto#Eo-8|i*>pgHa4uocB4o3(F))V5rPFB#$vn;Ib-U$p)mt_@
zyuQ1TGP}6`fXkNr&drB=s^8231)s}C=IFQ@vWXVWKmQXR^72>FTDFWE&;BACzx;U`
zTYC7^#iU7U9uGD1!(m;PIppHpn?(zCH>WN<KQ>BDrjA~B2HR|N!U^x-xz9a@CCgT_
zttUL>v5(=+Z+s;iF1`$FAf3tZ3f3k#h}<V3_$$Xl$VXL*V+n9}-qFat6mr#?3T3KD
z3`@WwCusRd%vD_G@zY(J(>*qS`&_>EvR88L4}ZdoUVjn~`=?hiyZ4^h<sF)hMsZ5K
z*T{*g6iq4xI-QeSOA2#1;SJk{-S^^%mmJIOvzs_%`R;u0U8iC<Y$BVPB`cEBIN~&K
z2RRs^X5z_kcySStH$?|UYJ1@P<DYfht3+K20=`J?-?cK%DQHOg)gv}so9P+H#@Rm4
z`}^Y=Te*hmeh)FWxL?G?1iL6;q%uk-l~@tRu_Q`xS=+>AO41l1kulUX={rRlrci29
zt#CC=*?7^4tX{oa7Vn$F0a{|Y?n7rwN`kU4x<*>*fRw+Jdo^f5)HgT}d+oIc!D~(t
z8Jt2`v}0Rb_sd#e6bvG=STOwd&4rrescAL~bUpA4+(4Qx^(eoi7e_;f!6wD%PvMlA
zx(Y(Jqv@yZpc2*0)yHjFgARLw=en<dl_fJXh))Z1oGeX)5{bq@9G#fzCVReO(xZqM
z#u^Q7{P9K5>!QsOShb8pANv@#^s@jIx}eMw20nd}0GQa>HG#=ljKtO<rk$gKNoCMl
zDz=_Eew)XVwA%&s`KMu|(+;7`B+PVOM4s{vi-~BaDG<x;Uf#v6{||h3xR+l~Lt$;>
zOmw(seW9{4ScaAk%Yco@vjRA*1`T)~d;|x*_5^gzB-elETio`APt({mjqkhA*wiw)
zt<1AjlFYevazJIZPBvCwAlTt$GF;|Ky&<E`^Z)iZzV?-K7$0qN*B$Hmf1m#Z@A%}u
z^XR{O0l)a~i}=M^pQPgwG|d46TcE#0&xO!z2ntVm605brz&KJPQcFICoL7-g*5|$e
zakSD{ycK-6M{|0b#<f>+;YqLKybpbVM?CrIJpJ_dG5e>FB3rYHW;3CIfOinx>`9_|
zTm&a_32D{byek}pQy;bYr2s`-r_JuqI}-ax>zLM*eR_9u&1vsq#7WU@${oK=^iC)|
zW<>Df!=HYJ?K=_|5cISd>8bT`heWpdyJ|Heuhf4p@8+2NPSMC;!GKsi8PnY!Ze*0_
zzx37IHIuPrs*7zmit!?YL0m})*I?KarOPYiU&)i}IAkPBmU?Vr@an?IYqT}@-r>EE
zqjHf061xPYQLM~~(ZwiT@lmB>Y8XeFd{tqpR5i<eQ)=UA@y)z<1(DnNjQ#i9Hx3rq
zImBkU-(05_*Z=BB7HfpO!*J)~DY>UeUrd^tO)wE<i6cF8apiozcn3759MQ8DlBCc_
z7krRaP5VQ%61F!ovHe!2M1-a|tggcCcW~=>zt2)*!Do2&!7I}yHN}QGM;|v_lWRvu
zNjU+sO&jR`;%aQ>nH_Di^ihvydUPDu4l=AlPGXaqoMayCWniADM6@c~sVvYFp=#1r
zLvTvcc{<yL=oqW&>5R7UF2y7bre@Pn97r7ls?L)yYG*I5`=S()h3`z|C<+?q!1GZI
zAscI<hwRPaFFT&4`|QOn=X{RaKk^?m@7{{<^&-(!irE!kF;%Rh^;&7QN`1B`0t@+p
z1f`p--$zV?Yp%PQhd$(qTzUBow8e1vp@(td#oyy$k2!)ApryR*pWn=7U;Qt*@|Wng
zS<+l3$5@>Dicmpb-<0NCl8?y5@72y-1sN}Z7a!fOec&iqW6^F(vL(ZAxQk!B??Zgz
z<*#CCyTfx%Kb4iwK9Y3pGMbZN?%n{-Yvd6`QHV;HBA;8&0ij}lq~uvhKDuNSOu|&s
z=h3IUn%j)ByKQmn*S`!m-ihl^li{L5R^AONc^RPiOp%#clC2xK<4d1s>$lFvZrqBG
zN_Lf_m?s*^4qz9Vul-QGb`f6+<zGd6jwI>u_y72MetzSvV4A447-^!CR0p<-qF16E
zs_4MYO|HZhxiTVHYeGjY`Uz<?Dij180XP!@r`qi*)cGZ~kETQD;RA|>mC#b_+Q@)f
zR}>8?#^^lw)U)ou1ER#L!<Zy?p9bG+aV@T0y6#*o818x1Rg+=^kXAdQ%YyeF^1wE*
z8|CP@ZYQ-bR+6AeBL=r9!>5YQ_9~=jDK)Scm`Dt6dZBeJMKNP~eQx{S`N-x?G&I8y
z2!_T4Sy8FFSSv%&kPx*bCkq-dqKr2zzxejIY4<aXB(QpS4t>gzZ0Tn>9L{H`Pf_nG
zHCFt561)!zL^Vp^4fSbQQ2VytQkR;RYal)MW@ZYN4^@~L9mhHl5~jEIYJ@}eWw}r!
zFpHb~|9Uv#-YpddoOo3h^jp>6mlEs8dLff0=>dmw;EP|y%5{fu^S8dmuRd@FjrAKz
z`)P2%&1nb56v(s=BCAzQLWwFcp5P+m5R)O|>GpdxnxlN=KR?aWp87YmO@o;Yck{Z}
zzmku9_5+wEc<*VT&}zbgd-IfI{+3Ujb_z+?pxxY$0IacbUaBl+iqle{yVPghH6IDl
zVn1<t_tJ;idlJPvf$1K0dJ9|5{Sse3{y1*F?i!x=p7(Iz>)y!BfrsE$EhDiO<HNZu
zRCTODtj5}`bg(b?>++9OX5;h3+Q)HAi=mN>!XG_=1CKeH+xs)@E@OQ6t#7B(4HqU8
zaD}p!L44V5XegMeEqwi`zvGuDpU&4`ay;op7twS+a7zjZMt-Y%T<3Ke%s;D^oh}0^
zj<nau*%3}V<1D`Twet{dGc`RO$JO#{6nu?BVjsqRjF=LH(JIMaiq4Se5c0j&I{{@}
zDOCq^BR)}?gXv2V<0?exnrfNWQ+l0SJJpN)9O%kHFa;c5J?jqIKa3w2RfSYFt7^dG
z;#yqyN5OE%;`z820=a-`x4{JD!iqpYox>|oT}Vz*fl!RWG@7Mmxeq9o&Q4dP(zU)B
z?j#;rARjUCyUfGvR<8NhxlEX5ln`{W@lC~Tc7FaqVRRIkWBI&kGgY=-cOBetJvc{i
zyu%)ke>~mxDA;zCHjM>YF+=SN<d6#atV*sFnP;wgJE4wY$6rd9AZqB(c7y(S9Osx^
zK8fP6R@ky-D$F5Q-r0zhZS7s0PA;y+B4o!qRUtzby&4qb1=DWe_uHKVU-nuKc--T;
z?FZ*`^%?J`edi`*HVvw1!A3eL#8ma*^Lg!<f|l|(WdR;=j*Ki-`d#Ju&weQ<z44ta
zADLwB@?~7~<8wLc`OhTFQbxu{zy#S#uYwy-SoX9hu=l#dxa2=Sz(@)%OK~~SVWig2
zt9rH+s<2WSE6F94W7!kA;h>65EE$e$wvU-f(O+E6H(zrC-#+VO{Ka4YEl+vtJIM|^
z6kRbvqiG7_u=lyyr7yW)g?bakIWtn}xK!v*b2cJla{2@<-(WUrv+v(OkBuvqFfuwx
zdg*0sJNrCruTPfG&_`OliZ64@Ou6~O%gAoMoqd-sW8d+W+;sWnMJ&?aKH_cX9@S!8
zmdgH&hS=hh7GF5$JDm2xGudbFJ(!x_NHcMCIt@Iorg~I>aavbknlG{q>H;eD_Ceu_
zUGsvd$PJ1P=^qTrWozYqw2Vu&o<hmp(3;zYP@#2TaxR_^_s+ow?LWZYW1bGfV&{5s
z?Xq=8wP3h?@qFBaE<?E$VLKfp2_!85vTS}VKy?mBS?0Wc2prOCqt1hO1v;3`Ob_@-
zOJy6|ImhSRG|k#DUZZKB-sKnLZ@Z1eg#yT{g7K*D#}>QH+0=x3qFn692C$09Vbw9_
z;TPw8gO)03dj$5}or53uIPOjzPJ;Djp$MfOF|lZf8?MD*w4xX>1j&~gp?RLfX<!Dd
z{0tOiJ(q@&q@dgFS-O0jX2Z~IHrR0Yws@}NOl%FQ+y2Pe;&#8db}^(;+~pcOenDg)
zX#X1@cmaOx8dm)Mi`oC_PiFJwS8~bQ{y%$f9&cAw-v7U!we~*e+$p(9NPsW}0a0*3
zMG;gSsLEGC>WrwUEr>JVh;^)1sn$vrr;0<xs#T$-PF1ujPE3l-gNQOFgoKdX@tm{w
zTF>u~wf1n%JvReb?CaOFUa!1D$juq{Ir~}b`8=QDeb}3B!ORP4E5H3}!Wcu$C<Mnf
z^i+MMCmFi)8Ae9A>Nhv=#6yncvP-UFm!->j@v+D7oeTbr$rWQr6Otss#S9nG!1uSU
z2HkPX<KOUVuD$TTnfl(37@0}p0nkLMi<CO1;y(1zeMtL4#fS~Ph@f8Zo?zf$4ed^c
zMy52@-Og`6_EEn4&Udr#!ym;9zVJnw&wMVO-B!{XYtnFz)W@V|1xbKJP$M|6<UTrD
ziN&i$)qq;iz?pY3^Q{QtGnhgOR-<}_Yp&uMX|%}3nmpy*Z{@1?X7(Kwe))lSW2ZbW
z?lXm=Aq1O6&VSReYX5z?r`6>5^P5?xmR%qH7;-T^2i+Yvjo3;LE7EK68iAM`C9@6A
z|M`Wy@$K(n`S?;ExaUq>aQ=Vr@)tjs?%Xs9f+b9{ywjSClP@P*zOOkbX=wf5_Wt@W
z+T->y2w*GQVCmZ#qQ0%1KG$q(MTLIZc=;UKlk`ze7mA;C&K2KQc3QraWy>bQ_o48R
z)vZ}*3}L6?x?k41vSARBb$}aQllzF;A()Xd3bQm(fh=3#bWks<s*4{W%fGS)(`+F|
zNMbsrLCI%kd+LSuIY{@$#to+DXuXIQ#DDKe@-8=i`#VfxFy05$#}x5Pq0un|g3s!O
zwg{dn^#B?Yok18;VckW)gj;TbEN6ZqW!Iyh#r(2WXd?;DR34l%Y*IO7)Q1Z_@1hp6
zqdSO-iPVW6y>3$mk9OvvJ&QPxc+Zk0qgYXrM#9~9Z>fo|>zNSSHsBnV`r*2NXA@Ee
zTWMkQ7=_5jnjpGVk8h=9yN<K_aWCdUM?Vv8{vB6*@V(e8uOXSwP;tSv%|{2cz9OJL
zOc_G3V@SVkPv<m7M)|;Lr*r)CPsFOS^YU@dJ@3oB`Mqx-?{;ZSq~ZC!FQ=Fq?J1pl
zWGp3Lp7PQ+yov9B_AI#RPMGObG$gJ3UQ4}(XwR`weF4;z@$^BAYYcfON2lAgX6Crz
z3t!}4UwtyO8#i(MN6uis<6c5%*PTe0H?T<z{4z1f%g3h3`!HD%!1zil5?@f6&$9rk
z>H9xXlwH;I_<4CSrb+UYr?Tr`K8vYrhFv@J+;-OI8S8qy&!f47$C?DI;WNs-W8}bn
zdEUo9%zh`ml&609t&Bh6F{m*;2WGiXqlR8BSTVZLNL3XTG<P_JpIrD`UU}-@kc^I!
zd*!pAKZ9LIn|$E3*D{t^thz`m^5y5Gp7H=KeaX&cHipW=DSJJ@dcnolftY>IW7b6d
zg|^PQ7|s?aM&WB0j6D`2e!a!MpgyVkFo{O6CR*d-*kiYyYiZR|?zO$=j3MMST=(m8
zu}7%qg0gmaUGD3&K}p8OAW5(`h?NFU7<x#)TgMBEq~~TUZg{Rt$IPdvOPv6)fP1-6
z<SdIGBvds}s1#f$hdXZN&daZ4G=?`>v80lc<qi=W3%d4<oT;}dCE6bB>eLJB!1yqI
zNkkaW=DF=V-(kGt@Ycd!yYbLxKZh+Y4;GL<AWAuoTs)b2M2nWiSMM?kpX*wY1=n&B
zrq*KN&_(LAs-oQ-I(gtW153tJTB#*zSnj&(uJEo)dX$L>p=@Q!GhD;<XJ{A-3Q~bZ
zi{d2OoW^NqVl38q5})FB-h*YwzKp$}|2+K0O<eks_mf}vOVT!E>hLD!64cz=M11`I
zgBr`soI`BN^h}qdjy#rg{{4JFc;>T?<foVafL-@ljVdIKMwnp6NkMT?6bYIsykON}
zM_cgFN3#1v59J4+_(#&aySU8N=s0zby2gg6&?@*U4cU3TD9#9(HTd~1jZPPT*(IFw
z`qy&(g%|Rix4ebJ-uWJ8_T3X*wS=ZIB!Knd@U9|=6>O43fS}3+2oq3{`20pS>Nvu$
zBG9Z(UnxYpJ$kpYs997)Bhbe3=r^6pZF$0YV;Prz`SWno9i*L5H;qWlvk?%7HQ>av
z^ofsW%?Yn&^5~<WHA*AC#y`n-ORL7GpxAh)1>S{f&b-UG>V}(n<tgvNo2ATmEi;{r
z4R_7a7#(G%18P!o?~oXaY>bc71rtFcDW^xJR6(>7yZ6<7>CtU3glqbD$Q}z1sX#C<
ziqjm$(X<KH38{ysTC9eqI2fF`GAX6*%~4LSrBvaEv}xypE>J=a$piOUgOwVJ1=aLy
z2r~`WpJ$c5!LVWYlOlLvMn{818(@ur_WZ(V1gX%t?ZOMJAsrhFg2Q;YHwfM7>7J%T
z7e%ud6aQ974GJzIgJ~vQ^X>m)H1j0tF+QvKmL`dZ0gsx<SC(EUh}tev`dMqVy;JqU
zm)bifO@o_%_#^n^tyq^cH{N2`ryW7Id<D5_VQf;4W&;VyMjOK5E)jx|YN(>nlXhAk
z56eIliJGmT^jbqkv{DFe70k|}CQvpkFu80Tn<T7VdndNAxQO&AW2EkdHC)5>zphSI
zV5?wEg7-0QDmnyt6^s?!igA`d=LH<_^5aQoH*@{x&%*uc5}I=!jqW@^N&Nvutik)7
z*_k=UM-#sL??2|y$2@^M*WO9mQoeTX7x>%1e<zvCX^tj{2^6Q|j?o^lBfQWe>`M%|
z6(bz>>XVqcYdxEO_(K|7=Hn!SQZo>2Hlu(zPAt@GnQtIAP%yjQHjVilv+fRl^?{G@
z^?&#zM;!G+jye7BNgsD8Zskbe)kn)@9}OH#c+n+rB5|79>%df@)S~p6yGG~eNZ~3&
zzUr=R1wAPOGhz)mXf&I2myGe0lTYB*tiv9SCO4e*DbjA2oIE^+7&7NW;HLyCREE%P
zq&S<DW)wYq`+NCa>Y)K?L01z@bb?Z^%y-+|zTs|O{L0sp+ZI`FpxMASCOGZg|Hwao
z{XAav=6A7qW~LOe`#k0jEto6Bi?_(O+G8oa*g&M;4wyj(B=u|^DQRP2x=?i}fy!&K
z2qtELRNJA%Nqj9uI__Um+;4HO^m&&@J>sFE>4;XVW#gi{-C3{0HC*>c!*KiXy6o__
zyWr1uY!pcoOp=5<J<sVKJMZ`{?~Ra*jR(${*O&-tXirU7hw_0jVB0D&aU@$7c!Tda
z*tCH=e*SaDjKSwQiLo&ls6fV|^<aHx>obd}hxn-1$bF;S!~=QL)9N^G_{KMBwH0Sa
z@w=?#kYiuWrsyywF2}nJYm)eN4dZ`Rzwk(|wT)J5G=;7B6iH1m-eU;~fnreD*lwAM
z37z4`R_?SMjN$gR8=^2Viq`tX(=x2^!}Y&nNtxhov>x;(rN&51z`VVYU=6K`RVANX
zLi4Dj*yklLM6xcIee&;V|KM9RW<Bb%=r<p4Kx665&tXMaI@03!=f8|Moq95>c3Q>(
z2kyf!FZwb2AGjYXo-|HjLlC8u&Osu5V24p$sZsLS(Ui`rCeMBEX?)`gAH&>sJ4|;&
zeJl=lJqL7C{@IC=Czfv4leQfr8#i&|7r)3?UVkbR<72$!tk04j`E;_?yI@C}q+&49
zDz(f@h}tTNO-zl-zJLK!TD8_DAmwa9dkHU9K@HDmBnYHUX@94Xgafp1IL4oM3{xw1
zrj^6mADs`^+=O;Ld3>|RMC)81GCQnS5)GfLN=3r-s`uerRUzjRYo`)wVR0y)ZO<{C
zd5$~jjcmzat~-ZIGkn(rhRd$Jo)4VyX@2taOKCNi;9|O<NeXN3D*TVA71J^^p;EQ$
zfa$@YjzVSY@Y=$E5S^LC1OYw@V#9=hnBXmnx2AHu>YE?cryTXh(v*&hVaoTA10Ecx
zXJV<_9twRXBg1LKaNU3F_F=>DC)j`ol}>54AfoVz5ty6n0pR<&o)!INkrEa?r0v86
z4#g&6e3=MzHcydgP8r`L+iGfLA<9@9nnW+^yvy9hzarhRk%k8mkBdbb@faH?9M<CT
zwV7`}DX9`67Mr+gMIS2CoX5nn?w1$9<yT{8+q6d#MxS^DjRPLeeAAMN519|~P!TxN
zrW*gpjY?G)Oj3F<q(wGG)rFf&>HjSzObqa}*RPA|pdo2{mtA*85$>6qrc2?iUcBe3
znbP5qVYs%thWyWi9MUKBtOpd9O>X>Sp$!mg=`3k6anw=lbL?|jB5khw#7CGr_Zu{3
zW>7aDauxENnfYm&qZ3^7o6C6o!w%+_>#k$jk}+QWhF5d;*FH}sU7BMlUcyAv$Eha$
zGr4kbDYeEvAhI4InQY-!?ZmTBcrEAr-MeXQ+>Ex{HABxb-7C^L6k`*dDmt6dnCr0N
z+u!G#C!E6it8e7RfBQZjdg2MpuNbAZycK|n5}z@p#idk;L|LE1Yx6aFiw>M_4rE1Y
zB_S&lTvU((7&21ow=|bvzTh01waCXOdFDIb!W}a+?3Il1>yLe$k&cp)#r%m-lZyO_
z${3<bH<qo`#(O(uP#<$9ywaJUCzlakdcvEy<<1$pS<nGkYe||7)P$U&k=7Dy(g-xK
zB&q6V+!t-p@K%}5EcCRH?YiZ*1>!86K8Jtc3uiIsQ8JvL!}~dG(j{~4%EGspA}rkL
zOEZ9KiXDwpoM)odVE_Fd5G_@G9ntDZJ{hjz`m;9-A~I|k?)COwp#aisVp}aRHc*nX
ztj9u4%HOr*GWG)=BGTVAkS<*o(nh_<tCASY>^++i7h&yM;_>3vqJ84JIPw+=nh57T
zLYr&8`z@BJfb&=qL``%QF+qG5Cl1n+J1GmN8an7;ZlZN-a6xJsSmE3J_Uz9xIx~k+
zLudH}hraIhY!E}9rg#jR=K;+myoPXgmr~2nzcowWL4$#@MB3|Xt$jZidBzFK{KmWD
zJM(CsvB&PK@!pZSjLn;8O0+R(G^MXe9IoM7yl&k8&n2~F=)x5&c=uPS?N!q(PD#|@
z##hjK@=@&n!k07J80U(!{)x_ie2s=nn48bZeahI#Qr`ZS5Ayuyoxn(Igq16I;d}r6
zeO~&C7t+mhMn_uVwL~_Pu_ii<Ro?c!$F$lJNhBoR6X>31^&~qV`3zP}uHu?cf0pJ{
z3a&Ittq}e2L}asV8ne*+-Hm+b)Kj_O8(-s@r=G&0?|%=o2Ofai&)(xI001BWNkl<Z
zWho<#CRQOSn}o_4r{7hr7uR0UhOy8rRR{X=D7~MB5h*9h5T?1^u0z@#=B~Jm<j%FE
z#xt{Ig5>bW(c1q2rn{h5Uqb$0->2!o;iHxyHU=sKDU#a1-|X;iC_Vd$c#Tbtr;|;S
zj*fEjX&>a0t8YU{5rNbs@%!hnCc%isAXu?jkwERiV4|ZGL<?VuSswNFN(Oz$7b(8m
zFcp$|16}@;I3;&QGX!bFko#$#{J2N(7mq)PG@GVr963skkc%acI*uZpTBXphP}1Tk
zvp2%??6v!Dv{D;2JcWaoFV6(ZFa&hC{tV~dT5nLFFx)!4G=GBa-C9hui3-@nlFiPP
z<JyADPtDC^;R8z_;3KGkbaIkB^5aDei4odU(-^{-bn&c(#m!aghuR?SRrKbY*>uCT
zj2l7nZXAu*)MO%fd`zh9WqvN|hb>Y=NDB1bqX^D>S~<*IeGQ#Ue}(DH;>Jf|-~Cu~
z#M7Bc$H*iE{uq<QAfIAVVM_W@9YL+0yUvf&0qsKFim!K45+>b&0+kBPtY05rc0d_-
z-+dJ>?|^|@Z@V*crE8%+s(&UFdANpav7?ame7}3V7ew&l%4Pvy<R%mjK!Q<=8y|=N
z_m@22b*Iotm-5rkd<ky8i?NZEbvNJ1V-9`>=Y9KU?7Vs(p8bLsan4V^#mJHoygC|5
zg7*&Zd|#lJsoce6A(678D^+wCuZpG#ok`1+-|{wYx$;`N7hi^HyD|%*z)|MgZQSO0
z(y19PJLA)Q>8<bK;KQE4^UnAu%wryo+i8sENI10mdT4LdiuK&Z+l|K8by3=rB(Dp?
zzG{b51yu(&Qg2-N`9w12&i@a-@tkM#Z_j!<$%fl8lG09uC;k1qxx<dIQpfo9=guOz
zr;W=yA#AU3`;j1m?teeB!vLA~Ro7a&H$@(Xr<2W-Hb;2Jdr#+^-~Sou=s3A*;N_dR
z-$Yw(EiHPZ^H8Bf!IzmKUc#?(eza9j*S^j`ckouJe>cm4fI~&G#?YSG#A&CzlFy%c
zI^X!jhk3&ZFK1@UJqW%=@2U~d`u53?D-^Foynf#0JmkUqk(lZwc<&HnYYt#+XU@=Y
z&l#>i>AH1L!*JX1y4=@pLLf;<#zuos*&2Mi9Z-fk7g1^fG-z!?EQ4VXBQ%yTM;Za0
zcg|zP(3zfruEreIz@^<@3QTd%Hh_(jk0i^v_Wbi10Yn`(F~NPr2L!UnT_{sF2Np2x
zJ)%TfQt`gaq)ru#wc&l|j!8-Q)mdlJxMvFOD($AQ?~9LPdc_K|)P^y<^C63Z$j_}$
z2#TamsoBZa{I&=1wH7*0l)<Z2GO01p6lkyCh{=nTX=U~5oe`Ilrj}c7UyF;ev!#`+
z_&(3?;To<#XK~WLV+>_pfsY4qDTw)nXt=mX#01eXV6Bjkj=>Y2&XYd=Njm8=etyP3
za`A^g&P$%~G-mEt&#sf>eDL(Y<uz}69a(#tR?EhOm>jRZ%w!3{an*eyD$%xiA!Z-`
z9DRmNp%t*CqfOX(8IOD3G5purAAv2K=}dJIYsfRmXXY5qJ+r^QkaOSmCfb|U^Rmx<
zf@Q}%oA&N2X)YZjMM)NVHCH(T6f&(ogNN;gmZUG~x1t0MKH!F`tXkNsT5NZo-+b<i
zJZ<G(?0(A(x1W73O~AGic+kGAdBL&VHs7Ul#|Ccw_77;al@2~oehT|;U6|3M+nk#Z
zZiFg|la$u9q6V+va3qZ}KKj|O@bNF7$Jnw-a_=$5l+@QscpofdHQMPGK3;`Lq|CRG
zu+J$L@oC7j=xN1vXc)5i`_;N7Z}{ZA_oepEYfj4=9{G?1Xt<ni?$~$lHK2m`2_hB@
zcnz=BYk~%EEmTZ!TM~~q1ouP2O_*gFhd$=u@X09AImrgWehqMx?Se3eo6d0kN!M+I
z8irej*JZ~+h~EDhM&HTE7{sLK&fIKIGIGgD)Wrdj-scb$Lpr%MOxvx+*aR!!W@n*0
zU!|+ovKMq4NLVppPzb)#IWV=ETYmIYmZU8*mxV&^gF{InzLN#pVE?O?zG;arIJ1{i
z2Oqr3y@6)#7`byDcm4Z+(43lrR*GA_iYL7CHEd8p8zW$CKwwG~wD7(ThZ?EHx(Wk?
zaEn+{^>AUeX3rR1of;_j+;JO;53U{Jm0fmSiDq4r#Bkf~cZSJ_sSvYa%^t4*eG`Vo
zUBgPp9_>35YjqbDIc?U${AkL|vZXxhGpBRsP1p0EAN&YU+kG$I|GHQ5;|tE=&?h~d
zJkLp5BVif<xJZ92CUAvok!pRipaX#3{^Wu<rS!r4!7XjF^m#|K$AJg&qchH6q}9OP
zJx6o%6v=hJ=I8HxD;J#oC7yHqD>&@!r!cp20=sGgYfV%h1T)O)D6M@{0V&N9w(mf;
zXrfMyG#bzyt)~NPjn0TKCouw$8|F5wXKb`VVhuCX_s|5$JXtg0fL9#D-J_O0$Cq*0
z$3BYOxe=Y~gq=Z4B!q<|{%yUNzMs|pL4#JN7an}pCC{6D@f+Xby&wGy%Xi&_PCE<Q
zI~yFEoYUAmNRNGJ(SWE_pL>=0CX0L@WZN%&J3lrV@E=;_Aw+GO5y6Xc`)zB<)k0$V
z$uBNuWNb-nsw=I6mV2PX<Em!Og!kxbO;m#L=z|`HQHQlQfMcqNF|m;vd@l^waNQ5<
zmIV!ih^z(Xhu3BY8-?OVE`3KW>G%X{gfvOWvu@>N6fNco-eQIJXrD7kKVUD6?6ec4
zDQXOP9ySu!Z9`|id~B&b$YQkAo_g=K1ao}W!C(7Zl8twfI!_}_!t<B{Ess1*kpcBr
zw|T*e*oNRvSWGG+%3=&wVYw<-e)V5T{<sc5J5OhPjIqZ)p1q&=G^R`=AT%y8MZL)*
zGDSqKv=Srs(}A?tZM#Hls^e042VP<%o(Xg$Q%ysrHf#b%m~e=&eEBjOz-YtryX$@r
zQP)v^7i92(ZMcT(&pl<3K36p*qz=1q3jn5^D3<A4;($E-cLw7rjPJIZpPc^#&cEnF
z+|(AHa^M4a-m6}W%W`rBlO%YJsrPY^ASO%|L~NMGX_ciCY?AA*Rw)Qe#lh5SFjy9q
z%KP6m=uS-V$k(01+PgNg`MW=)aq|`Y=0oq~n;-alcH3uvj{fM!$PRfd`Kn2pqa#>6
zR(+XRFH+YX)wEzzS|Du0esq9kh3bN{wrFo7R*I=TA*-Qg37y!G<6w=#(g_ZE<w;yP
zH-{X07!P{+U*UkXm11oJxARUs>aSnQwRf%Kz@~EJM?S=;NkS%op%zlGa16aQl)l`l
zv@+=MQfbsmg@JgSnM0Zl{{4GD=k=$3h^5O`Gdt78rWPb86wpxbi`<W36wz0gwW`}a
zZdbL$=z*!u{WW$_6w!9xDy8N`0Uw$M@u7*c%?aN35C6<LKfZ*2_`<jN;x~Q($#@7I
zlq|T5c_r~mg2&*(4x@s_VtmvKd5;y)EN7=>lkBy}>aZ94U@B6Ygc!4(r;y<qu0Q#j
zkG;Wy2?M~*!|StsSj!?621IC#kD*b}*PffD6uT{m0PdyL^azM+aJ(YeiBZ%Tya+b2
z7#y1Cu=$=UMO1e%D!v%Ada&f&c>ed9G%4D3xXh8J%^IfXN@3z)k=){Z%0Z2cL_wZ<
zk2iv|2ApH3T)FZypJa5)X7JE%!h=pYk=dPgrjrEMB&>oh+?E9Oal<!#iVD#>bX6YQ
zD_|^MFR5oSlthh~9U)lt<nwbdKOccQgE6pr)pD#?ZoJ`-Sc%qn_!<q%>;RHCT*GyL
zRIcLtUIc|z0gECM@)*QNDu_fZUr}|aID9sb&UaXM+j?Gl^zj`3#`p7p7ahlkZn_9o
zjdR7>U%*Pv%=8Q)(!t=o35m~z2shpr9$)T2A<xRxaOL8t>gk~jQVCNPCGU1nV5U2V
zOLLBY%ZdE*pHAmjuY4;j?an;^6Q5%5m!C*?*#vf?g^6=g-%}H`hG0uAwS7l!TNzki
z>wt;$oo%&7e-x=&IbVjr30^~e?s&uGD^B3ZAN`0!zxqY;1NMy|2N)GR${sI#9=c{9
z{5>5u{rY0`${W#6hdd_f6-mSk69^X9ztxRk`=#h2)nwbd4yp~l{vY4vb#H$+D_8H%
z?0k-Grcnz6&Q*teQ=>|iY6vZ5a^)V;LwjA2HzeCXLeL$*V+#F5v44p%h!~tpnCU1d
zz4<iWcgC5FEMGw$PgCN<zp)bNsYUKkXbQE58fJ>XHkYyQJ`bpJgede2#Xae_$ru=}
z;rbt1Jw3s&Vfd3w9_mS9LTl*~NDR$JLN+^#@h&<8_13xihE$+g{fJLM&<ti|l;-$&
zMA@SokMnTP6b1E1wsRiRgTEI**xVGi|N3%TVqzqH2&nNsuNt0GVxfyjh!<97Sulx^
zGF(XumnakXovMbG2%9edInzJ>Z_K6{(vcLmVv?u7{he$|1ez_tVDbuuG8OQIsUv~f
zXCxc58+xKVBg!W=l5vS<#n)(!N3t${{aqx4gjfOgT)ivCJMP-BiK(0z-c+B~4~zV8
z{qL+;T8osTFlJW3m@1{*hm>L`9!!%?u9#Mf8?U>HR~&sbzx~z4eC@{<^7hYulG)w%
z;qW)Sk{^HNY%YJ_-!VDe!M8g&s>x??z7dk>kW<tQU&|^G?Tt5+x?D6u^y=f3!Wxpa
zg-mx?J~E2E@%Mc9{ioB)GFUpno_}=$bIaF|PqZ*bFy3LDkLk<7a<qi7>me?B?~z>X
zwT7#|?6NS3_WrrP2WVB6LS3?g5oqGUq~v2u$eT^9hoo#SJgF5}xf72*@dP$xT}EaM
zmwxUGv}UuA6<|x1ZTUJE3_`co2ub)PjDqoC)ga#Dfv$8R9pURgxPaHZ<8(%rE@yV8
zgCw@hN)Qc+-NoMKi%ENQ1ER*7T6`!_Z*>MaBFTVXTOUGf{atJMf=cUdQM_~HxnsP!
zl(EJbu9JreQQ<y@B{Ps4Pi_JrrUXh>?sGILAt}TL1>(aFcp!;Q!X6?uqmiE1Je)54
zxgg<T!!WpE_~Y>U>=1^;py-2id;({bG&OXmr!diLyzq6dBlJDk--5lSLN~OL(wta=
zBLu8jQ4;TQ>(_-~F<ChA&~3eu;yF4wcVBu1GBr!n2<r2AfR_qo>3gxVD4Ny_1ND)C
zN~e=52{Z)McmbD$gxFO{!lnQGNm_T@L3?h7?$S|s=z}@ng~zkjC-^i4^;oHexs{WU
z{*8qiv?o2+hXqJmC3q`7gd$g=DZ=b+w_(*|jo@9*1NPYq?{b{8tX)4FMcoyAD8u?b
zT>neA#bU-#21u3Ny~cFiBK#{@s&2XGZW?KePo4RBj(qZSS+mzZ{Px;kaPVUu&RiEp
zn+^P`ojCQ(Gnu*kLN<Q>b2M(c9Y5d2MH)+iCcAJF4g%3qxe#BGEc6qqCbaRF5v&;U
zEg7=8gPoq?rzgLSv)}hVp82LzIpWKoXT#)j?!4l+j5Z-BpuNu5pcs{2)q>QY4eo8?
zVv%c58(vhpvYrsPg@;(F^KJ@sU8D0*q;XmsJ`<G?xTYqAn4y*|we0-V$CEtp0a%($
z{qkbC@Hd!FR+>*sjZ%uyMQEj3iwP@09$oMrQP8|YmZzNigCBA7JKoRuPP;HS*Tpsx
ztg(0>DWXLdhE(Bv18JK&IN@m9sTPatMpzipvt7r=@>^JbJ~96%vB3`5d(?xqww%@n
zx2M81qnM&p4Y-xis2BF+<PO)J<?zD}M#Ptz39W^mmgWQOAcJ7I27EoFbg5e4Ri<?W
zb+`ug1g7`QCw~}TpL^GTYHe=Xm|Tua64F+ayweFW_oX<u_<3D8ZIJ$w7$^cF;BCUl
zvK9Ek@d70gn7{k3klabnkRu*;wbuZz4k<-}9p01VZGQK|?=zO9<eiMvCKwww3tw2?
zscseHSb(ON)~EDrfb|zpzfmP7lQt0ST*it{mtTGSbjI>|oO8^NSXRI21*~}Nq09nm
zELj$8W=$gGz6)Amrs}KoY3X+*kzQ1)Jem}#tu^Xc;bU&aa`!DaW8-)@_nrqnXkYSf
zh8<~f-HkVwMO}$r<Ql0R!y-Ri|7&*5ngCiAtByo-Qgv9(NqmNllLc$S&(5{mEL*mM
z7d`g`KJcD@;GfR^6rVZgYjh_jXpO>X3O@H_BNOo8hx7Q8Uc>Ky@e?+jdk!PFtVKIn
zOll5_gnC$_Q%8^GYaQ?}fr<#4#PMO)&5)@%8q(qS=YE~DPCAJ-2S1YIzx5w<4|*Kk
zT~~9=``*EM|8geWvxSy}Of_gKq!wMLF`llby4L#$HkR1{MJ<t9?o9D1D2wDV)By`E
z3uBq3P&mWYni&)8nTjWslzhnuk38WO{2`xbO)C8ABOf816}&psRM5%-tX^c1c}d>}
zRXP@d(>Bs*^1Yv4!mCbwA0wkXF*}uE8c7KIG};sFSx+mL1(N~Y&P`Xh^Yf8yqkW4!
zo%FR#y=NrTGZ`spLn8FN&tlq5oaBXMXvG>E{=F}fyp^$r<-z-`iC4u`jcf^62|GL@
z8Lr;%aoHqjnZ{jBk>U<*dd~#IwXi4XpD+v?h8>(R4EoD-*)lQ#NrWuxK$iCuMctNe
zu9ogtCY%b4?6e{T^4Nf~r6NqNy)$rA7lBAhx1eIWp>-5Y@wvjLO>F$_6^xn$8@-{$
zMSiWA5^Y2WxHN3#y%bqy)NJ8utfcsO;XqxCbT#@=kLKX6`Ymg}_H{<4r-L43tjWQr
zoXXVloygKLq><p&<229{ycRy#@x>OdMFXx&+2B<Eeiao=dK(#4#TvNdh8vMAI3-zQ
zc;NnP$g?g<v&9uxUXNmsD10tnhYT=>9IoN|vuv70jzKi2ECFq)4(Fm^62TZ{etwp2
zTe$JYTiI{T!%0lawKrbIkw-m^Ogv+wEqsv_FOVp>iBWjcQ+e#GPUf1QoX;I!`B$1Z
z-hp-;&PvRZ33id*`^sh+6YV)`aC?*!8N$KLXP(*3Gc*w9uepk^oOCjCGhI&n^4Hk)
z=%bhdTmxi$6!zMeV_){y{PZ8sK)PVmR|md$YA@+85=mMr1N@|R=n7ZkgW`F-?-A{a
z3}8I<IJ-%$Zb&6iFS+m1S68&cVSLD{X&mtsmL75_?fEvk_73hj{{kAbouEUj6+@}>
zpL(~Ro;qJT?M51t5fk|MwD|V<zu@)ncpnoxtzvGrL()hwQPf=g>b*WY*pCeeW?Pgo
zrdxRo-O;Cu4qko6vt<Hvk8ZJkt*QyT@0;TV8kmnkfF4!IbI0C$t!C$4R`el{F(_62
za8iBWs$O}&lxSd~e~R~J!AbVcmtgF$Ul`mI^fwGQ4X@9=n=q8s#z3-k89qr#tVOae
zWO+}FPrWeru!E$(E3TRbZ(w}YDzJi$;@*Z8<~D4=)J0KR-+F45QY89A`zjZ~0C~pL
zuP<VBeh#uMk}Cr(!8sQ>m9|8=23J^FwDL;52uke+E|M#Yja5DB!*Rxn#pKHB+;a8j
z{snWzALvZYLaRZxas^L)_XpU}Y(OI+*I-y@lOzTcMV@VRbnm57)k59kq^2zJkytF{
z?_s-$22rGvGIiS>ARz}s5!URrlJRB(Nm^WS)lD(sv}Oz`(nEE)PtfIX4cC3DT|y14
zubp=+cpH&TAZg&evUzHjG#%&t?>&>JKJggNIO9Wn@$56PsUc}tnu!TIf^u*XU0(n<
zu>^m_Q#t&Nr*Qo*e!@*(`XbF6?u0o|YXf(iKr2E+^P=i1^jw3Gw=gxxM46kNBXN!~
z?eg6Z{2f30_IbScLmy`U<Nk`x4M-+O7{k$Y9_t}XJ(Ev9oJPxX*DroTnn^IWET{-h
zOY}FS4$A`b)*T0wiU~k)e=;ICRXQkL(NJ@I?eWXOhW6I#t?*|AUFg;%ii*Xn$5?|K
z9pMRYdKH_R!iv!)T=<zYk@a`uJ9+T`uV*O~dx7?!SEU4cie|w_>ZsQYHHM3>xPjNa
z>7Ar@l)0%462X%rUa?`E5lEt@XjWw_Tg4!s>p|H64kCK}PD!#o_eyc4tKYS6{?Id>
z(Z0z?AyN@~G$Uc^<Poh}h|Fa?^dScZH!N#<MqG+k`{TeDKkOtH{R(KIX_P9Vy>LW}
z&6rS2@-8fHi&{49(gyYf{SCwKhu7y`P8jM4`q%_6u~;Ld7?_!lW9e$ts5NAJa93P>
zLQ*0@jh%MJl^jr^VT7r*>jF(iq%WRH`ZQDZ5x0p>42sd5KmPavCZvIhsjgxLu~IrB
zIq!oN-eRwD(W)zbF>ZZ`RWZFNY+(spTyMiK*Ui}j%Madj8cQ~8q&+<c<0)?6H9X~S
z-ote>vzUoVTq5M&#U>#-Qv`^or;{t2sM@|&_E}PIa)NF_%3vWuGtcb$yI^)E<}8G%
z!>+rm#F22z+V#viDqr$K`>}wNJ1j)Qb-z@j!T+Y{-J=k-q$Uc8B(!H|FfN5$IQ+3k
zanVJ;<hHffbNJ&A!#0)iC8Jnl$nll!BfjL{dlXGzeyL!d^mrcgikGwL@(bDUh0oEv
z?hcsGYU)-~Qj~&qYB1^aKI$h#=yto9>73C<$~9j*m#@9+be?(4aUA*fcQCtR6kTN)
zA5DT~r2wr!QokB-VN#GJ33(&su$P_0FTe3s*f>Q#lT~i`*5aK;Hz>U?I_CvVLvc9t
z?KzSxqct-{W3G#zn=b(tPt8447Gd6O7w^7$?IsphZ8VUFJb=l^AHuX)Chy+J-RGW1
zW8THYo?4SoxU5Q@h9P~=AAiUx;ykWv`SoSj@rpOR5j#H4?A#1k5KTaBxwnZ@COG$q
zh_`z5tqDCCobV%yiM9JF{`KZZw|~;GKssGc<z-+4*So#-sn}~HB^&h(UOm<Z%4*)t
zIOGurVJRjLy`9b0jWdSe(_(jo*6DnuF2Tfel9JENl3;P|P5=+IY8-}FcJZE|U&AoR
zFu?6zZPQvus>ep5IReHaDr7S=VdSMnda$X9m}*5zd-<ioQmvre3(cKZ(=~!L64V+}
z3*F6IAopRDt(CbRr=@}m?5o2;Fbz_e+Q|B=ucjq|{~yN55q&o%Gz$`bUx=n{<%&q(
z&q=)qQqx9D`AzgjQ6sq2kT(QuLXu_dGItL@d*@r3Y|ql|WOSBIU=DdKPksHHxnp*o
z)Q+HePUf=cxFUFbBs)d&MOnjrH5ymxX?{w}M=!xAJMQGpx!{qWwc+;LgL@Q;808@k
zI{@c%+MPDHuien+DIQ1j3uj5gHC*?3TilWQBqE3^Mm&OxmYDG$SU?O@n`dZ^jq{-o
zew2r<Ifzr<@EX2%{x{h?GtH8f<5-i%4b_JUKv`=Fgg%(+CmzPIWaG=2IObR$al#2)
zf7yjx`-L-U+`5i@zJrS3VtTLB(qoiNG#utL=-l0<=~HH}yn=7O@r_K3w0O~n|A8%g
z?M-`Pgw|M-CJyU!j0dMVl17_P>#$3gl8=n>?7upZ@4f$RjQAAS&C5fBF$PoI=W(*p
zE49`g`ZTmQ1Mm*a+$=wP`&;?k(MR!v*Pn#F`4%L15wI{l!1Mwbd;o2IaUZ%m=wV32
zFfX2mz4jz-_m0&K!_{B?7g#?--tP7uN^5&Tp=BtwSU%dZny|m+<aC_yn`?g0@u!?h
z$6MO%kkylGhT7oh=P~4>kc0!WAqE2?i}HyUW)SI)ZNl1DVwJvbOzy>JIY^{l(;|xW
zYClQ=q4f6Wrb<;8V{wk44|0w)O?dRd55roiH6+#9V2E_@cvUQp8csZ3T(ECe0Tb+j
zO*bQLHu?5vKf{*w8!*-&qBxg_(7Yk6SFBrNPq65O0pP~rb-Fh_eeGb_YLSkOhRxFo
z?M<5_+M>M<NxE1el#<F5L5VP`-+3jPNO>qufbR4(%*-wHEbl{Dg7Au(^9D%f*>uTo
zN!uN)Iz(M?Myd%8j4`F~S+;r`$#!128lSTU07(&EClU2ihbDk}M%s$p@Oy6h=g%<K
znMXa$H!Vw^^DOpz?y;=vbTD=VNeyblx4H<2BW_tL3!PA^`G=HRiwacJNNNg29`%gi
z*?9f+*rNGxIR`!RL3r23dB+u3UK>BtUX((H<!8A5T=hxoToWxv^5wC}Kxe8=)^>P=
z!yk7P*Ij=Tx83~*jyn2@<jk}4s->YRD9|^R1@%g6v>K11xsnQGlPT$u$8gx2PvO>U
zF6MWi`#7z28`1gM5L%^*!Gh*xQts6BEGCEM><r&~@9Ern^$k4#10QD3r$3!p+ht-!
zibb##6dzk9^2Sl2_#iShD@I|@{W<U<2XoEWzQBkJg51Il%9mloRVI!sB6!_8kQ5Ha
z62VOCJ-3|uEv9~Q5syum(Yp2zT<{NPAYB)(V=+~!xdUxwL+Ck1%0TF**6=7aWE6H=
z!y{h&LN?8AVdq@A?9-oSELWU&WsNJt=?bcnDn5SB!No2-W1i)xY4Gjyf69wadObA9
zc-Q;h%D;T+Q|$kMy}>)YcV%8dX`5OI*0o>kEDO@cd%nTCm2bChW&0W6sJf%&R~EaO
z`u2$GM+T2J%0N{QRt(N%h=inc?;YwLE0<5QX3yO)qJ93mGN5T4f=%1~3C8=$L>izM
zAB6bM!(7hQ=YA8(a(3EvHJwhkyx!jLkmTk6?`uIXFsNbp-S9fyo3Ch%Bb-?5#1fnp
zY?6@8Ph+H3M(VbEQX>r2+(RG;XUi}$QCPMN-xvwG4Avszu+G7zEu|Ajou3*~IVA=#
zX_;AIqx+I|UH)*vPZ=>5av!^EEvFTQP^k2T)*U#i^w;(UNQo&x#lSTz1dnQvJovR^
zfz+UR&dNOD_g_DUP5=HK8dG!F6gnd#?ER`&vG22=$4mVA!2kdt07*naR0dJRHt4py
zsDQNwBVg2F)djj$bdD+rN}`3VFxVElXv?b6qg}-pQKl%K5n$c5*OU5CS&Vuf`mp`+
z^Ru+lgbOda0!$-@FKIZ;NUzv$xQ6R~+%^LxBsv4B3eLMo%S*^I$Lvg-=2(j_efg_A
z_`t_-${SDRtbh3=oA237Ye|DvqZOug8gx{p(+FA%ii~wDQhU6Bw*j?I)#H{lF-IK1
zliqv^S6zA`*ZtdPN!G6?pKarfkQor0;^rM`yUpZy!d3tFCC>ksPw<S_pTt9sdl7Rc
zC7Wz9+DsAeNqn%J46Z%CHo?>q!8f^LtdcEF+3UDt`Sq{91An{$JEwRFiNi>|SMRfx
zqE!9~;vr4gvi2@kHd^?*HZh@~+6CLe8DE=7*XyHfS2L&sfvTv>itova;#w`%yyQ5#
zlPhT@W8C$#AHX%g!_9W^u?AKjrAAS+R+QPeKh4g~<C2t5eEwWs{g(GJHo2Tb4tf-S
zdHCaa^xk{$){{>nZ?{QJBf1DVk|aK~jX4`eD!(|<f%BVfgXuEpx88w-4tUIxj#LT8
zw<o7aWvfqL6GY3x2EAq~V7+2|P8!Aj)`a<zcU=xT@IjcU+pC^@P`5MOKz0PCG)BTy
zO9gKvCfI|U&EWQpT>SHkIqGG9O-B{mNW%BRO4L#f0lvk0fdvhNhzuKsZG1s>8xsaE
zhIHv9&Lo&5WqRW#)a3zcD8>3U_Xt@86|9}Q4ZzqK>9XZvQ@~@?Ve*Xmn{UIYlu69=
zxKUJ(5nLCd@YxA$+Qi*gUQN>&R9qCn$%09%0&H_X>krE5E2VwSP`}Qi+%UzUVu1v_
z2^(8tkSt?ObCgT|>7&?RT|~Zl3$%p!R>EGddo8PuK88Qew=u0Tya|GBys6S-O|e~t
z|EF4ad+*U@SFTVL2OtC9({RGt>u&(fP?uo@c3!oNk%q-N&lOkw9xp)*TG{9J;!wji
zT=(C!LCOU9!s1x0pjc+_nGQiwd52?;cm`kl@)x-L$_sePlMlfs9VS;T1DleOlY1AD
zYbhg(tJz%dDXPcW&47>fq&t?<dg2j0=e>W=PtQ4z>p%T*(%U!Trh_2_vkq%=HeK>d
zzWSy&vd`Xo^0c?UjqZ{r?Fl6rPh#@1R(yL22)2pOQRo6x;{6`8AMALOnZ)z5cm5sc
zp8j_vGhH;#L!z~6%@?94wXuE_-^TI1n&K3CjBb{(`k;gOLwlNJw+U{WondnS{lKQB
zgHmC?xxGlZEK-Im<UWdvqjMb16qb$hke9rW4dkpzQZD`Y88mZH@{o;DAm-Y8A{J4+
zhq<{nBrRU|=J)d9fBr13rOQ~i{vPh$csI>Nn4g)VnWm9C8v!O`aJdg+>_Svs6Q2$O
z?sd?lU^}L1qKh;QJuZE<ClFmEXG6C8FuFj?*VnA}?&-l?B==dM^eH&+u*M)h<H?6V
z4r7bl1ziv?E`yqq;ac=8R?k$?n3#?Pkkrl3(?IyfS)b(PC!T~C#WVwRqL@w^u~qKW
z|MlxTpcfc4VYp#<sqRfzE7c7yN^|KlFbUQ+=}b?Ph%a`tFnaGhk=KW`#b244&{$At
zCNx*>g42L*<CWATOy6;P6#5plN+}UDLt%&KOQ=OfP?zDazk<fh6yoEi?IUHXK2hk`
zBJ|&+y2!N{I9^|T++Tz#;VlLmGz4BS7{tVIoVlDmP0D}0{~e6patqn!Ei_v~yOr|b
zli$dir$2+c0c#WR-7qgmOyu82OF?fy3vcvbfh6^VM?qTkQS0Y}(}`HYsAp>32G}x-
zXt0HBwo>-qV^<QNao1gUap$HsR&34cR)&<M;rcTcDXk7LBDgGQ6-Jkh@{b?+2>Y!$
zfWJ8MNdEh0Kf!8~$>k$7Z8J_IJ;n>xB+>rVm-I#>HLX;+b8B_zMuN>TAz-};=*T$o
zutRwDnP1?7^M1^AU-=v(D9x!^He7chUq1D9Y`J@i<39RHnvZ-uZIfUpMgslFD^_bp
z2BoKZWs|6)@er)#gf;w)B^{rD<$JRKfsf{Q=lvULMoghLH3I~#B|Y~xUJHv|zrdh|
zEUO|7!%k0s3Qv5`J6X5y{ygeUr*hCMPsZs8iCRo;BD1}F&DQC#mHn!U!HC5InSt@Y
zd=^v7b|!U>&6izDe)aFkW;<xK_LRELO9X4eRBEO(i!%u?c=2oaub*Cmb19oQ&Czkn
zPcQln|M|mTkTu8n@W(%oG+N}|;S6+LPLiaE7pyawsGXDk)BNIJ*M*aUtzD!3K+zh@
ztLLMs6H;|+Z9w1usZC*u2D8VNFj|!NxF8m$A8`UE){?7|<sOqH2#V`=c-*5O8TB1z
zV9K}-;ij!NjDv0%DZgUD0L4V=Vyl~Q@j2gQ_XqBSeZU&Jt&~(OMxzaSe87+mGhBmu
zfnme&r~0B6#VFbVHYb;p6S#}B8#iIPS*!=1g)qxtL%$-9UgHPltT49M8Zz%lqBzYe
ztpD9jm~Qxz`zT<mz%q?u$G|607Z5s=XWV|##f+K+%^k+DP{7qQ<j}#zTNdg2r5=LP
zzfnkU5sVwVFY+1+yGD&kmqwrkSS86FYk(iT^;E_;ZeaSJEi{@b^NDcaJKxFJp@*|R
zcR&Mg!U4jGBGGnGOW$~nK}J0aY<gsmQ7P^tH`uB}=DW<@d`C#^j7gjaKk}g%?c&sP
z^>sIetwa4*EOUozxc-N`g}%v_Ah<loxem!l!V8W)j&sjBkBhFnn3G<65_2<i?6lKF
zu+DUWU+tYEktEt<_5uO*$(h8OCzvR%E|BI*59OSMAj!!F*#QS|(&s+S<=;MsJ3jDt
z-15Djv1QW?Fa6Y+?0wX;X=jQqNiYo)4)t2_i$hRj5MEaySH)B#%8g^DLQQ1r6)T3$
zh~*)F^$ITg`310P6G_)cA*!hXyHYkF!a}E>MR4wlS{ltGQY?NXW!L9ClOsNH2D=}3
zEZLG4wowO)WSfNPtvQ0q0yw5Rbb8T13A0W7(n%h1;_<A{w(#I(JMqI0olYzB<n!&y
zG&K4>gdu^alXb}@<>=?1z~z6q17}h^hOzN+mX1xZWO50wee*jx;Bimn!Yi&|Z0Q6p
zUPNPUc~Gkk(KWM)(#2-G(YS&6KZCX&UF0!!vA^q?-7GT6DTd!7y@Kc3RH`1fTA+Sq
zPDZ?@(J%^YExB{-vSJx~@3lI-JFi<2Pi*`2F$6~XCQcd-{%VvQN49B-?DyAm)kPO_
z*vn6(oeHf+igUjA^CgDs4X^X|H4Gy1M__t*t+rmAw{`|pqa@2F$qm>>gU-xUnLAJw
znn91=*EsmF>6U+$T3T!NrW=cqbC6=$eEY35yhL$w<*ZQV4a9B1Yus>E5i)N3%@w2w
zKF@0rLAqe4k{zmB{WD-4aP73TE-gLlih^zw(Ht=z5{dremMr(E<cMY@Gi`R>(&oEw
zdNWHi&z3E-G)EiEr-nzr>s_pT!c*9kB$!5%PL?6ol87b62Bb0=z2n72Zf#8F)M_vA
z#-LGJX~klV#R`n1DR*3ZJ&6ySfLF(3A9D~!Go;br!rxqr7pXSLMJF!9HC+EQb*k4;
zIUcsvgBm(nn}<H&!5py1eq8wLUy#ZOX4En|K8n+vJXf3%G7&Hdh{qVnd>-`>br8Pq
z*(AE{#6<bhG(dtLDVLnwcsk2k@Tf;{(9@p4<rn^hy$(BseV=wDv$XK3p_vLc_arW-
z;S|9U!&Sl=+Xy`-6Bxu>j88E>MSOy)!5P8Z!t5a@9ciH>4PJ1{8~D+=U%@mIH20;*
zSENdNUyzHSgOJ|JIzek7Nnw0oyZ|k{Gmxg()b?p5JloQ=Zcnq&GYg67!-y$|F*=g6
z{K&t+tlXVMER9>PW7E0kk#<7;6*K@K@!iaI=Ez8R!SSbX-K}@iRv{zQW|MgGEGO%F
zM#ok#xpWmHBNJrp92++!6LWk#KG43(!+M-DM?2kt*Hjn#UiWoYCG}_Fzpbprwa0yM
zLF-qi5)B0Hwnj=CGz!tOL8uakv}{-tV=$`VJ-N?$_`?naby#EjTGOEs_x7B_`q9<$
z3=_2~Hc0~5+HTVl`06J=#R;!}J?(6U#z+Gfvlj}h1Mhups`P*Us!zvWV4D*Lfa``|
z&Bf2hi!xiKC~ZQrbQxVSAPH`+4efT=L<+i4zffRNRHd+2v#2$U?70^m4GF2%q$CKN
zZ(9r9yyPi+=j(uZiI__aO2R4Jxt{s8>uE+Or`|WH+qMdjZT-deZE>XlO-y7$`EnB|
zIIliVNW@rZSkyyfzRNCm-^GvLa3ag{He04=Nk$vAM-vYGo4;YnUp$RJ&UY{)OK^Bx
zo+Dbsx%o2LwPXMn7ORoYP>h-Jium|CyrXI0hD$HOW)ATl@yY=Y+8c3-PbT=q#aCjb
z;u9CwbJ#8n*Zoze`tw&xYP2{s^IUW7<^1gj-^0xG49iwef=%eE!@CGj#CuI8c*Z+C
ziAk#Dcp2;<R0JW4NN)~|gh?lvkZoE|cJ*bP|K`)!c=07XcygRyp7CLroyDNc0qT5g
z3Jiz|{`Ca{>$UfgC)JT%YSU1aTJ2Ot@?fy6V+Ql!2Xe$KUQV~u4fIfr`V5i2>RV;o
zXisZ22>F(3uxf(YM6`Vh88$WQQxD&BE7~yG&ONDq4AvU_E)wjeNmAMk!(&f8o?AO}
z>@_;hZ$9_W*!kd;CsLSx!hE|!CM}MB{z+W%yW3FH#H%5-DUCEmMRBRcr3smLxI6@V
ziZ#(;PY@eyLrcwql;Ix(Mkjq&RtBT~efWL(Q2n@DmC;Z%CovBqx;0hQsNztl1D~s&
zqAh*-C8c;;7cm1V62nX9qhL7TeK6y&#xOrO&(n{1GNSQJAPbw2q1<@~=PS}1fT)0U
zGSWQZ{7-&{0}nbF_S_Y-bR>qcOZizTfdVdDMM?ipt${tjuwl3l6NbUx=ujjYlS^=7
z1D`(YLT4TiA&$J<?{w>^Qq(YLMDD~`nyYuirwLvm%RF>F-RaHHnZpsL3}MrkUKWu0
z>ch@*He7Q#O=Ix63lycq_MlWfb+oR<(jBQ{(q}_gJD?Ls=sv_)iw|7aAW$ZFw0mEM
z0EG=b!K+1*2G&_dHs>t;!);u2+S^#>y3EbY&>BsdpJ?*1x4nx!k9Y<fyPlREMZF`p
zo;(#Y6h|J!O)lsq0zEn+ViEZ(c&OfMw5wNIMp$?Kwcs7OhL^v5*%*87u_xXn{Nd)?
z*xbp`B7M8(a5XIb!*&1FD;2HGcLWlXFgdv#+ZrVk7#SIhlsu2NbF8MJF^4iGy`&nU
zf%9JHfA<uH`T;)P$3@^*{NW@wjNW(~-+S?^SbD)ltk}?@#W<ZiZ{<rbIhK(-*VCAt
z!&`$I6Q_SJ1YJp>`5}4K9mPlWA*or#<E+D(kaJ*Fu%3{qqJlLkVdi0RjV6pvVACWv
z_VIoA;<Qf;m7}6J$-+Zy<t3*JG_-|Pvzkg%QBjxr+AZC2D7oKWReQoO>p{cOdbX9o
zlMbh|^R9Gf=i$~n*zoQD#`q+L*%{`$8S~!og1>q*SKoYRaIPXC;Lbao&v63IM8cK{
zPE++JQKER%_lS|hXP_}nTT6TI>P^>VFx$8g(kCvU*X*NaW#L832}iY+^=@Llv@@YM
z=Cl@uS_Y%~>N?9J;~}Z**9^fQ+>m#=BnHGeES|#-c@)-c57;(b+rA-c;6QBFlpLGG
z-536bTmN_~2fySvx+4t|u|XSI`wXT3BN#FQw!dKzk!y$Fz#Zspi@l&GDrRgHO%t5=
zv;=hHUBTwl6r0IHQE=hjpfzrA@pB;+*^n%oBpn|KTv?lv8l{oX<JR38ebVFOdacO_
zVz811<SWSyw_kV}EwMo0Dr-pqls-Y%eW+gp1k=^#CMEYc*uGLbY)J3V7zv_h4KyIc
zW14~y&yudeU3nFkyyG;MW*zRCnW5QgF`o(#dG|Zm<Ctf2yX#_`V>lr9IVO=3wG960
z74NhhWQ&vqMMJ-ZF&5(#J2MS;uA@xkaT$j{;So5O;f-+Vudl@GfKj)~uxA*q`>$yz
z&+u{Nu0gbH8!3_`ph`O5A-Q{s=Ik71%M9MRID@OgyrdTljt8_pl)OCkRr;TT->nMb
z6*?K$p8a(Wwj=DS$~~KBX*}{!p7OCzvux!9IQvB}p|Nf=*>sx@K8UMHqth9&aj9xa
zpb%EIvOt7Yv_kg_y;7{JJ|c@+d;Ov@7e+kbP<7FMS?M#5t!)OnK=&}H3#brosAw{a
zD0+L6+;rOwVOdscO-vA6AsR`MQOBJ~3mPLF`ic|SEE)UlyeGdn^J8$&CNx)O@|=5S
zyFB@6$8!1gf56x@m}?{kte|2-O;ZVGe^D4+y3N)04+yE#1AL@<M5#T8{{DBiH7!H!
zS`W0=C?^Rf(&)4pVn;&|v*2^oz74O$owc%0E!=d9j3k3pk_HwE)PsJIa<Y*mHaKW?
z+U&i@?(DXD=O8!N-l^7bZ5y;5__sbPnoKZ(%%(fAb|e4!uV-_@X>Y?>A!(-JXM)&s
z@j1@_@xZ*rdVs|y4A%_5p4<Mi4h)JfNbt>+#^h2+6H)`6&G#T0-HPhgX*D}S`VbaX
zA5oitwu@M_HOBa=UCD}rxaN#V&h#IDhjG;@A-Ez@2DM3)45$K~S?<2(Dy+{CRcz@F
zRH2Fsx+UGAyLa(ouQfM`Nc&OosHp_eq7o1Baay9_H6}K?unCrero6+Fjw8SHa(?yp
zzhRk4nVO!ZHI~w8SoV9vYdP%XQ@As+NTY>!0eK>Z=~(rs%Y*2<*cc>uIERKJY!XHc
zJ`e6x%bS+FuJ{ecg^QA^=NV5u96R5pWi3Da;RTSSl`p&YBTvIMT>tX`NBy81&T<CD
zms$gx3W<d;pLil?KJ_p7*+>3~#B~8B0WbliD6>hZM)>uvqz8fRscqHa*sFuFl=-B?
zhS_z@wdT32X-E!v80}R{dDf>t#-5LSFz3DPWJYE^I+NoCoC&tZ&Uw5SR1;8x_ZBa~
z0Zm1sVS=q4^;~%VA}q5+9^-LPLcmJt=8S@Ov^!l$pquBRx$v@Zm{LchWm|{)g$=->
zkv6hpD6`ZL$VFz34WLvm9#;S{1qd{{gyd0&k?y$<)9q<?Yj5Jl&zwb)By5_=dGh~z
zA)7M=YjG5*%^{b^$L~)$CDBOPDs;d`$}FK^E+KvIr>{AX9otz4U)erm4K3QSs<CuF
zc=FpcAkI!R)W(=Stp*mo)*K~?H)wJGKlaW$PO_uA_uo@>Z+B1LB&`xi!U!P{5+;Me
z0z?ow7-WI~0mgvH59|ln7>vme42WPXzz;A%<Y2(qWJJ!A3<4ogUTKqdcT$IYtIqqQ
z>Zb0V+1ZtB^z7-+XIZp6+tYLJt*Uc==NH5VMw}RM3C3Hj5uDHPxJow|?az&uvT^lB
zaYo4%Oe?|KpkesKTi+^VB1M7tfky@Y&joFVX=o)gtkiBA27`=_T0Z{vH*@M~r@`S{
zaFZ#54H-`(ZTI?>UrP_L-Y{I>3BwpvNrM#vtqq&VjUh3X#p`y(;4rB)0y2uN$Z)KD
zfnb<8AxILYZgMP{F(GR}l#Vg%{Po2c=bGr17K7DrV5lhG3+&pC-?fK?fW9PYQd+T&
z4&BH6{QC`YTjd&wG|(YwjDfZ0Rv=}tQc;wM0JcK5FoKx~sQ<>UAV0l;Z=UsZHe`MF
z?Oni5resqsCLZzt?)|#gb1?=#yMbI`fD0(3dIA;~%Oa@>IitHp`O%HRC`@_f;%}da
z9k`H%p`IfT-^}Le4ubH#@0=ev;0$RQ)?e$d-^0!ovP6n$9%LE<2{m&p{^DZxU3DeL
z-SlWK|LrB9c@RHGq6$TVaS4X#h+~Zmm2b7Jr(Ed9B=k{k|BAn1)1w~F6-ORH`<{1a
z!~ISn0~W30o-cbjTMs*w@4oy+Ow9KfEM(-w085)RD`aYu11Wg;^$xwFKUS2W%F(GR
zGh=}}7+`u`Ci=_l`oZ^T?c9aTFXC{BgtU8=)zR<kb&hBgLprdG>1MLLa~{p1l|joZ
zez}V=O5RBr96HA>AM+6Q*#R2|*KyG&KFprq?%~wCKb-Esk`2P3+8B$mA<)Vb@;jt4
zO_Jt>Nr!1ZIw}v4p|fq=0U@j^gOgP*UNJ?I_<Qa>=|l_L_Ox0`A~_T7Aa->=xE7VI
zMFE)AY{`|lPShbr$-Dz&5HrB#i`XP5O(iC0R}FRJbBT}CI&itiXO3)nk%vCuRBY5Z
zh^!Z@2lDcTzObkRV(6fFVn!8A^p#(I<U>rR!qNA+AH7aOlD2TU3tzu}$#1xZK}0SC
zcCBBtHEjMm?syK}?!@LT_{7pmQ~LWB5RJU{T8eThB-Ei&z1X~p1K%q~!l-Tf=$kV@
zXtg?cYiL=+&P#qvlIInjsTdc>shdP^R`CNCFZwyNMuYK)N-2^q!drDzfp&mSN&El0
zVtS&rL`;b`sud+l8#0ed+aXnot%t-4sZ%yC2{XUCly5xexoq$`OY`%jDde+L<R{;T
zKY7=CSe)BTp0u!On}H6{Bu7&4nuqW2Vg^GojVU%IA4A`qV*<|=Kl=}oz7HA@1e+$D
zeDW<3a(2xx@Y`(*;k{~2;5vS^{#pa-aDze|T0>vs#)OF?WU&d;C)|RY-s7I=El=bL
zZ#|pbcLIu!sd5RR1QUiO3nRP}tqv9wdiq$*JJPl#-*70mdes}b->1LC1J8LM_=IVN
zX)N=}6nA~aYuLW`8h-hnH!`{G==HksP2&1-4)KZ+7lhP>pfyqqihZTbR4FKh#I*Zb
zP@~sqeBi)l!t$5D%!luLAHMbUvv~i5?uY#3rzHJ6PG!t~oO2H5j_cS$Sr2Gyt<I2O
z^9*_BXt`zPzx*jKdgq&HFAebhE*YScFmw02qDLRgcIngEGRMOn{#deBn@of(hS_N}
zc_6wj=@{p&)r$XyCPA;q*)g<4bhS4rjt%b-JZWNa{VtDsz`gn02j9)hU;I4s{u0Jo
zOxi+K$(@grH)&o|0V4)=1${+=9x)X*Z|v~AXPn6$ZgV_VGg>x5eGr&C6lV;oCQKrH
z2n`EGsv}91>5k<te|U0r_FvB?^nV)sQP`iV`e0w~L$g2;mVI51{FlGrN8kDe4|?_s
z=uai2t>~C!lJ$AQ{$IOd53uG51HgZ*U$?byWc3MaA&~2A+(d35O)c}gcLtgZF?qPK
zz@s*zT|qd922O6cS)S>e--5nZa+l$aCB?Aol1njzJPOOSasjA)J_U?-Z2QqqnD7Q2
zWC)O?X<3MmX&(1q=g3+wRk6@Vf8gRMuYZPW;!*Rqj@Q>9Xy#GpaGIkXkCmL+u4CqR
zm+|#yJ%hQWW%lja7x1^)DLD2h{`k#rg<GG@P7KN93<Dq0R{$><W5Z`DY#bv`H8fR%
zNrc3CSX^fDvfp8x;-m2Ifv25<4tn^c#lL;^JCH=NXk<^XdnK*E4g`0W>$@?Hk<SQ0
z5Nxn09@lB{(Esxq9`LGHl5L(w+exg?gU-i9FH?yr@S%_zCT>e}@2eNSzCfBoTDm}(
z@EL<QRiWfPOifQPnC<Y0SDnq(zy1xse8+2<TX6ITgD~$f0ye2Q`*B*JadImX!X71}
z?!h#{SME*49-@VQ%2dWnyZ7?-v){@|w#~79II%Oqm(M;M)ZtxNyP~X6)yo0jYF)z@
z?oZ9oklae34xvK3ld|WV-{9kC{TaW0>)-RmKmT*mB*C~o?aYx)9>!f?@@o3``V&5K
z%_6&ICYbkq`p$tQAeJn5k%m;e(IR8^bH?VN>A0V>Y%14SARlm3QJoWrpw2PbN_fsw
zp1@Ia6a2|(_hY)<!ibPNSDqyr4MC(apVyK?n<Pp2TAwr3hOd9&qdfI759TB9J)0>J
z$oqlDDyA%EMLnWn_T#;Rg(S7)S<bDFyBV7{PE)3Xt^}&BmsVp}Mb#t|Gg(47UXGG2
z^hvUekH705c)*`N3BP3{W-=sfd+$P%Q6$B$zYe(eU#(%daQ(WC#Z;OP^w4@Qk~6{5
z*?b7JQd%ube_<YSS6Zr8;9adimLTXI*_3Oe3}ck^=wrx}l(gMKZ9<v|y0{1ni<PCS
zue1Uh6Py(=F6Y|+_%)Vb?-mk{14nq=cvMyQzoF}aTtsP#u}7}wP(WM}j3c#b#D|l|
zrU_OO65A#>kS7_=FEQ8cv*~x&@Gnm~lewj3miNvFKJ?rq+14%G>6NeI7LRx|*X9m8
zwUJycP7Efs;5^0?JjKalx@SNpR1E4NQP_~Sx#U~t(e_HNfvbAAJDtpwwTNl)xzByA
z%)F>P+AE&#`fL5Qzhd+H3LuN2f|nTNg%5cnz95<q={ABhAZ-u{MZNf-^YO)G!4r$1
zVmpd@BrU|9Q4AaZTO(a4ju=V=Jfxc_X#_tokqjsSC2X}?WSh2d-&ehZYcIcq-@N<H
zOfR(<bUitc3An-uD0m%Xj0pn<4WrbM+8QV|VnV7iK1?RcR9=apZ>oq<=veF^C&e?B
zSX$YDyt@ebQYgmDJxrU!V3h+)W?V6UXLw{It2YSscZvD|LvV{ig)1+;jJeb?<&}L`
z?}VLwY&XR(Wb||+pTE4%UwrElzL_4z<;#6ct4)$5QTQE-WMd7^<<;jJJ5AS7;7BUh
z@UgOtx(nZGmrf3J9aSIDyYO7}mix@l_h?N{v+bH4WVtfv4v@s4sU_1qXxc`=zwpc&
zV{y4hoyS_(GCjc*2J0PdV0vN(kBPcEkCR|SjFH-8D_%aQEH5u`$|-+@pd>b}(Q?;c
z`+MF?okINOJ%~+kR>+J&$!HtTk0_i<cK`q&07*naRNnu8Ir8Y6v*nNOMmE{P1myy@
zk&`GR_1F4q|FyqaxmlMB*00+bFlj$^vY6TiwqX-lD<Me@#`U2;2nVie<+f1Ye20W!
zrDsD?F#8yWG`9gWHyaQkYe|ejmj?7NyE=^N<Mcv@iNrbhojb^P?V%+hk)U)S87|6^
z{V!%|{=OEShn5qD+7zX6J*)4pHF0$P9ef;_OGGA}59pc`g$coXN3zuA=KTR5f8;~h
zaOLIrg*_~JM>>;|ZJOjJ&wK_aKkI*Si3t6PX)=@2cRtXxB+wj+#zC~0ZUn?tykpYA
z<=^-k^t%`z(4?73!*NF+g`>squDX^hw(Sb`szYIR>zvi~*V>Gl^?L13!>h_5QAZW)
zeG}Yo0&=g>g~*gbf8PL@CDz7T<4c<I&#JWL+F@K!BB^nXV{_o}Ww@BcHKuq}1+%%s
z;IIkq{qonb=eL(~$s1nI)O?Qb`#>uSqOOW%6>}6cTHyrC&+z?S9yAC>OjW4H3t}SQ
z-5LwCvmAQoJF~0Pfx~Cmo3v=(<QQmAA=VCY-G`EjW$pKBbu?u(ad_O*F=DCe3qiNX
zV(7b!<L`S{ws#WdHyy^z?N5fOSwxjxJN9zs)Bb|Tz4+x^xpgz!Cp$=M0!Z-Q;Is+=
zRf@FkVmi|lS(N&-dc7?)L&#*z2QTmwmHmuRMup-lZ=G5hc54tStAWfbuYSupy!Id8
z&!0X2FX-itG)XZC&ga-f5b=c%@hH9(KSE*>oOrI-b`4+r_P5#H&H2O^zsV(6UJ)A$
zDVqe5%5)^GB@<l!Tu#>SbIRTBf+gq;O%#@|zt-45XmEQoaV;0>5sT2;)91=B{X4(>
zj|=(ZXFQ$$WC+XCNL^GP^jJQg$#T8d?SIpgX_eZ4wF$#|!?0G6NXDeP8xM0c3|a|l
z1#$=b_6;FOHCsBV=^uuK(Dgh6L2;dw>6;#dw+5f*Xr41^EsIxOffQr6i0tJ~%Vbo^
zGxl6`0UaD}5RfBllS)f5(qycB0qP);>gp{<)yWU1T+0a1RPR|tC^aATQ=>rCg7_)L
zlcJhp>;&GY<Vgs%)18alA)V#(PdXDh?|aO47g)%au@fD7Q&Vhtz-c_<E$^Uz$f5Mo
zDM%;iYaefO)MT+Za)@zw9Jx9~gv3}TFl5(U1>3I-PBJ1Sn)86uPQhh4D$3Ws@g1;9
z`E!j{b?ckeI({~~-egQGP!#GqYVtXPRa5_p+5|uWXC+X9icl+R&Ov4H8Wr`ZFia>d
z2u++e)U)9QF@h@u{?f!gRGKBO)TkR}Rc1oL?<IKL*og_c8@F<&zj_PT?cB*FXTOGt
zy$(0%mp`A67CowhF&e>=5E_=}A)nGFmVUn%0_ar9B}`l-3^$63n|@-s-Cw_qqn`Ry
zt~vfT9CgN--20WU$1z!gDg_lPSfv)R(|V9-lh%~sI%Yr66{R90_debyuZF-XN@*W`
z0*`;od%627Ud4mo_;<KTxZvkM=Y9`<1mC{s0yf-q1Ks5=Ej6^AhuoF+)z&8XnAS|u
zWJ0Bt(5g|R)+m-uVwD&=_U9jIgjT$UHHuLk`IpgWzdoH736vn$?~~e;uYc?NyzQKS
z=BjJ<VVDRyxxs+cTdY$O4eoc+u*!|1XHl$Ge4atBNHW7yp7{#ydjH4sycfTYnGHvg
ztC0H~b&#lFWA+C{V4;V^8nS+$t(!M;;t9uLtdulYIY62Fjd~RDpZE<VdbU}B+{b1%
z<jBb4BJ8-9FMaxxJn?_M5av4Ac7j1j!~|D3Y;;xA+Js^Kbs)9>Y7N8rfM36M`)M3j
z6Ld|qNN1+O2%R*czh@7obdG55q*6z)#%v%(sbDbViDCBmTe6(@X<<mkGMQMeyzrOk
zz=cH-n*d{~g<+j?)enD08;g-(D=IaLP$6=Zp^bXP#iI5z7O2naZN^wP^G4$_?ot-q
z<7P)uSvM+z8jl#oV$j^<HK35%c9@o-yL&Ikw`TatOa7X@ANx2nnPbn=5@`x<^DO=k
zZptZdJDc`h?#PZ-O1fbanTDx_fsop)9<kw<H!h~O3N5c(^@AVB1ZhY$=f3y4BU3Gh
zm6T6?>`SO5@fdRPP#q$cufNt`V}RJ^KA=thy$%OZ)&I@#ysnrTgmS1zs+C5tf>ur=
z)N9y?%oRorLI6IBV&7}IZ~3TC(9(LfQXd6t(%Rml+Cc236opnE_u`fI#unMuN$&Kr
zH?sGdU0nPAw=uo9OE&0+J<cdzO>~e81q2>?3j>k_tj}56yO&nG6;1d}Ns+4dH-j+-
zV-x(`3`alybnbojTR8S<Ps44Q#<XR`oT(^s*W`XCD-4r_A$<TGjLl<#r>KifUC?xr
zC6jEp&x4t{^C{%B(|q((U*eHZK8yZ_Ew~xWQh$jg6^!#FVq;E+3)&BL7%Qcote`1X
zCKVOMtnQg5YdAJ!>|kxBf?1pAY(q1mjsI4vM1{i9D@j{;^|afQOtt37^H9*3R_qvk
zm~siWF+J0g7AzDx1F;4nA<J?mr#I7TZ$fN`UY-Z7ng~_`z1R3^(or-|KF?7daEIHS
zM5mL6wPu6^OhUbJ_6loADwpf^J*qWmqQQZhf@TSH2iT=PpMTrg+~>g$MvlEHS(;#t
z#pfRNz6`T1%u)(_>Gjvh%g5UPz#0Y->96(=*I(mTLg_~>3W;TM;~`j+&@wId?AU=6
z;UY!Gnh7a8Wqf^U#X0(YlqhjXgy~zINM@BpmAvaoHD$+dFC!(v7hdVkV~t=m_-IS+
zxbD{%lSE*wkY9V%U|||}bX^fSkd=RKS#5`qRS&%S!YzQe3YFK!?xY36M<^Ju4&(b6
z=^;48g~!_A7??gXVd&4#b8|b*kKg!Ce)h^&v)LH-?p?qcFdL@vb5q>uZ(hpXp7$KC
zO%w9=6xL3G4;c+sES7`>gHgrhIf7?0HT>#3-zUi}CjKB>58c4cj@ZalTe$eQmvix@
zyUFrmHY5!pdNs{o-?`T34Z};ViJ_MM@I$tys<k{?(J_T>@!q~7|H+t4#v#kgx(|Kj
zz@w-Y2IOQCc-pwoI23xQVLw%+$vmlMBvljQ)z`f6tCV1)kWN~<GjNC3y`G=`{9Ek&
z<a?QLlk~DI<^Xsy^%#@TU0TLiq0=&a?E@d-7e6?UsSUHZ$a@MjB`x>Ya<Z$QL|pLO
z_Z@H+Yyyc1U;%ZKs4zSkJ-A~-MzNHk>PuGWJ?g(q<YgKh7RBTPS~?(ASf1}OxpjsY
z{l%Mk?q9tQ-L#qOTo%ra)RH6?62;-b2EKmL)TxgSQc=@Vgyfaip{`-6p+~YxtU@^j
z7y_MSEgERiRgR<D8)3>|!Q6Yq3Z5MAGQ@(4!}|=3;w<C{&RbNIQ0!}P*fM32YRmK7
ztH)Zw`vLfV(EEs@-jyMTxq3W;H<X!7HVpx(gI<sO-}{~zkwAkR%O&{#>nxBJfx(q!
z#cTCS87(1-DP+ipQ8C~Q>0-i9KX@*k=}jDa{|C^WYLO<kEG|kdi72jYKg{}}v%;Vt
z*8T@PVF38q`t{q-;a*!KMfAKUTCy2oa`T}$l$Nn9?%au}D^n*|t`-<-8l+j7TOmE}
zSQg2V)MC>%sdwzX{0itU<0EyZ=ogj9{au(R-@c2quq7k%`lM<6DZ+kHk9r>F$}3*R
zHy1;Ph>rOYOZ050NfD98_b-~Rz_$g&A|&7~RxBi7T9&N4#8K^Ovaf%KFFodTHZSfX
zo4=0CIn2xy+4M9sr`?D9zV{rs-5t0TMdvor*90%Yi#s7ANPtn#lv=i(e?BZOAnNb}
zxhju+^h3y&_Tk%;eCi9|gjNz$c4O9q>WIV1`fL5QDiTqngBU4nLt_A$4|9YPbQNu^
z%^KF<0I?fIaPds)Xi*yKMh{sZoAv03qt$oN?Z;wY$ko^Ot7by9tW+pSZNgACE51Yv
zoDa#=W>br$Ipu-xd@n!!_K&&tqaR_)r3~^M6;Fz=xcyotrY8B-=fB8@UjG{I{m=(-
z>p!{&d6pw4@Se5dzaE<g3Hd5gXfh0_N$@6c&$U({*GWQaDRyMlGo-Eu(CXLF)TnO1
zs$<^6M@X<l*R6*^pXI&R;ib(ZAA1%b`O;UIJoHfJ`#IWfk(e|H)P2x!cu~AKoOlYR
zl@{vw;nt}3j1|w>IK4vS+0&sjVT>E9uHn#Ki_#}F)FI4IM9Oq(jrUR|tQ9L3QNkG^
z^@gMt2B-1;N<ELMY(1q2tqbpGOo3WUtg5Q}sXpW{w2X4veeVT0d_=v6Z9nvYO6nU?
zK@=`l<Clyydn;2Z$8CK&0(#Uu24bMA-PNW0;~#M0cfZYjp8b4!lPRqviCUbBuBU`a
zLoqRxbvp1t)%<4ldQX17e*MNHJu7`-xxM;;26i@Y!eeNsZRU6Gf|#7C!)T(8XnSkv
z%SBQvrJyT0;t1^Q3=WUWm9_}2!2;~K2K90DYD~F~@s9lR-_XvpKqWAyY!({Y0+FE{
zhPqW<eN(G%^fD5FTDQ+z_287zQ&vZhsm28^1xs34BqAW*AZmjI%0;fN3nIR>osf5X
z%q(UcdCgA#^^7x^y6BhK#l0+KUDB2zo1G>*WGi=i#ozLfSG|@iZHvLw25iy+mm}hX
zOI4C!#AEtPOm>%8yx=D!UdW6kGs=DLdkQU4re-&C?uWiW&r$Ka8yID9oe@878<iW`
zQ@{TDFFJY07nZextDszHO!QSNu^}xI(xrjrGZN-F4jGhsd}8&l;xx}?iW&5{`ePsF
z(~o~V|MrZh;=lg`Y(GbBa2@g`ZKWCIwjvK-tN0nIAqNt9#ttP@rA0!!HAz1c?*H0<
z;JlBYi~G!HnA(-lTk^<qpXsezdFM0FqMceE|2Ka_*S6@Y$EGGC*d_oYanBfyr!+X#
z*f99=WDMbvwHydwEnJ4tE3b^UQ{!BbWK1zEHC}zyC=_j>x~QqJip!xdEw1T#?soc<
z_|fH8A(NJ+UYDdjfvCk`qX?upPpV#0BDx-EVc}YNDek2-Ep1h5!!T`VqN_o|&|Po*
zXFcp~Kl{+g<a<b<td&ET)W9s$=x0osJrqnW1mlC=L2;<@vEdkMc9c4)ujXVZjj2<-
zvRL(4@35+2@*HVS$N#~xY?yA7+N8{0^6FRFZ*L4m?NFNb*w(i;H`JJ}PF<qYOjeft
zHHx)IMeDUhaeqyGbQhP9J=^%$2hZV&FZ>I1ZUWOzpvd}`raC0p-o|HQUBGZqHUIu5
z3_o4Jga=YSNtGffA-ZT{(`LF^ptbnLWmxWxqU+`EB%>(540RB31?25Evqv38rXi;o
zpEKn>doH*TiJ?Tr%s>@zIlKPj7bLl&xkK=XHPNQ5LXK2dlh=*EBD&g!ujBviO0$Xj
z-;DAj3wNX<jLMe>LD_snYg!{g-RQkyQo$!4X#<jzE_!a>wS3~K&tTVQKh2DD>{(bu
zfn;hDHclgVxdRV)-}{j}-jQudf}Pt)rj|Sh<AoHVjW8_<-}~gJK`h=2*0k6-+v2Ww
zxh-B2<_81L`_}m(#4YmBitAk0kE}kDmfr*Fw7#MK=k6!6!W~_~>%LYnc|2J@z~mm|
zJcDci1vsGfGIM3eGaZwJC6&)iSt3X*0uq$VKYA|TeeD}Ke&;?8zwAoReeSd1+U>}o
zpcI*!exbR5>qqynxlp<5m4HE|zcPSxE(oDF&agaVc=S8}k#pbh8uIV|kQvjW`<?Ib
z!KXcyvtIbu9DSF&vz!e`XIivO3Ti8NodQ{G=m)BTH%xA7U=4*}yf)30HDlt{dk=v@
zGLl9;4m}yO1|vFOHdzYZ#XFgJ2K^qIw)x8Uf5si||0i6#*r%t4Og)kW8b&cy4woY)
z8JGN~NWH<AyQ-A*tw_Tv3-UOH5I^=>G%d1q2n8G6*>r#VkTQf=4{5=q>Uu`<1c$Rh
zM9Xlx#;@Tux1VOPaPt?b1stN47@T0)YZGCq*XPtzP9ZVTo;a>#BesUS4_~hv18B2a
z9iX+99jBb4400F10~3+`S`h1S<1voKTv7#V>3>0Eqk-y$kS#2d4l+LQx>xZ>_j@>U
z>sygcx3GlCQK6|SKczGcSJtg;53c54OT+L#zH%6MJd961t~r``4Nb$`95OM1NmJTZ
zV9y>3YtZ5MsW;;?X2~|hg;v;j+^xy2Ax$k}VA2ZL{OkhkAQsXwkVX{}gq;`vhKVF8
z4=`hFHKvo5O#SxPKGiwBLlcQc;V<KlLLG-#tp1&=sYbL%!+=`6i6UstL0kyC6NBd9
zTo6P{NAQx9Ei7}=<OY8E=6~S3FL)6f2R(ZG_AwC9)WXa(`5_y)^UMF5d;H~Ju)Q<Q
z5((}0ED%K2lcvS2Y4h9fp9j5t;B$-#;dsw^_EX3fmYMBL@}UoY5>3*`&DJtO+DnD9
z`r^Mw4&c>H%GWo&|9|(CVYA(}^lpMv=y&_HykXZ@zskaS-=O1QKt^r?<zg7wANw2?
z3CLw5G8```WwwDwAWgaa{EN8drp@T~T})(MHnc77x=WEH8V$ta(u<aQi#nnxL&?Ss
zZG+Zg<|F803R8w)CQXv0=*BsgHcay9SH6IsoOdplz5Va`$ZKB38P9z#x_J}XM2Ak&
zB2j~NKGIHxkmoq1Ed2M&g1A<kG*QqY8`T3T=7#!=kY+zrvS_`Su3xXp&a{w+gl5B(
zKZ8vawHbrHqnmel!OPyrqn`ekbY>5w-wmZd#v;}R>QV8G6rfgt-r^DOLr#V$i73JV
z5mzQ=8={*~BY3mY&NPu89a1{W3PLv-X>CVnOry9_v*0U-9;2Fo2AitUoeVYRbqdYM
z^r`-SH62hvO)49e@ORD#PHedDKt9OHea>kQxL3%!C|P$VP8h~D_HyIet4g6sjbRWc
zBoZlDJ>SO|OKza2kmU|B7BTTjHlu-Qx<Z{(1Wbw)j&6Vrdy)D96KTS?-u5=O9)449
z{?OCtPg^X}`OF79;Igifdc$s-ld|>K@a;JGhCxKS>r;k<Kn|s{`y;U=vvUkI(1w<_
zT^ZcYYa;|v8CyupxaoxnhlTSndyA7;^x?3!trk;8*!J_ElRCxwq4&uh^Os*mssTAN
zQP{Rxr{6pU_CJBqam!z1Wdya<Rj%MeKJI|79Trlzq^pTSif}v;2}p;8ya{6sd9L`D
zC2tGeg(Z&A7OkIN$Y&n@5N0p=CGDl%?CZ^=7Ib=oeB%`Dd)}Q>-~DbT?{zAdSy*bf
zafv0%GfbATG0$24>esMd@jl16jH7S5m7@-s126pS=fC0V-AiQ7M>HyKx+T(AD@$~E
zny{M7{(&M<azIb=jq&jM@3=!Z3X!_iXrt_p9DJ`#Bq`^=@-<xiroZQhFMTDK{KI>g
zC>>pjB-5r;S6BSTI`otkqF^m83xjvLa^#7(;>!Lq?V~ob3xm7q&9JvVnaou5j}nou
zt!o|k33SYl+E~+-)H^1~0g3}UbF=Jd&2sW%p2DWP+><B1<t=c;kz}(eHnEX%;LGA*
zB#~K~-(~##<A%GzLM>?;kBVzt0F@>wM;r3+F_*2_+!dT3(Cflg*Us~>$NV`T{p{D-
zeCSauEjiLu5N|;f)W->f#_ttH;6+A7(G+OZG&t|#)i5<UsG^%xTvHtt^~tdALB~!V
zWMnd}Rnafayc(@Uf{fVajXXP}Q>~Ihr3V%<Ew!^3DsFyVb|Na(vn-%$P^}sUmkl^#
z%SLW}+%XtDiLqh2qZ;x-R-9P<*JHKgH3d@sjK%q@E*Q5Mfqs90H--W5iZ{HKyWQs@
zJma}94uN()>{~?uq^x2mKKvLJkV;Xf2tuk#x;NwUPkxSzF1V2UzwiZgCxR6)QlnYd
zoQfNP<?A*hs}2Za&417n27n*0U&ob8wUt(!kLk)JmdQi5;w((HEd5<OFwVu~P#J>S
z>!Z6-h(#rg<t$*WXYSUwX3+(k&&)fFW-M*r1`FNhR;)mGneOiEkUVhlwNaxti$guY
zdUQNWS~N|ypGik*|Jv{4gncshJ<FpaLI)aAGKuIy$diajeMHlY4QQIdAl8P1veP1S
zITQ1alLsCC?TLTR74JBQ4a>^>p6hS|Novr!S={C=-1aYD%md%>f7m`Z$F^MYX@|DJ
zRBwSFe(YRYE&=E98k)sF`?JThuy+UANjUdoU%{lEc;5MP19R$veX9qnI$pOiazM(8
zlO9+tcu&_fEjoawef@O<Tpg9j$t@93ZFJN{K`AKs4EF72+eH^}*oHYKrsn8r9uGz-
z_lWTlSzY0i*NeB{yKj87FcdI&obmj@8IR+H=RA*VPPiR6d-e->=zGp3V<QPE$q=Vq
zywsEoW`&YSN9h<b|3SQlLcv9Pp(x}zQ-dD%(qHq-@BWa^9Z!UlZw)7(0Gl_{vVs^#
z=KCz?19I!hO;#md4>d*28A5Y4=ne2ploMkkF>F%ltH&Ep-wnvclm)Daj$L*d_cfzM
z&V<8Shmhm4WvuD&g|GaGQy%&_F4?vhS{?Q-F4LN@;U1U>PD#Z(5(5qwXgP6WZM-eD
zF~MlAg`ov17#AF{LZY!nqrkqrZZf1r7*iCA)c1dliQEn~!7Rf2Sd9CS7+wdns8hmL
zSamBzJ~Xo%WaazRy!$e$KdBpdj2nj(QyU{-45B7FO*yix&%N$^M_NfRDe&G0;13hm
z>Y7!O|A}ovsj90S#EMk(x*anfbpwp`c$V>al2(V0fA*`~={}F;9Uu4?dd%^w-(JD)
zJ-cbBJ!&H4(J5T5d6q!CREf_)oLYjS`}%O{wS4LTlR#|0Lm%PkFM9=kvP05IgCR+D
zL(^J6D=YQ2>#u{V_4nH_h{$@wuxeqv!f|0LRAOrDVK^hStzmxqHbi~H>s-g6CHteC
z7!(fKj7)d%2CTKT1lof>y-R-w&R32j&Xeufh4eD4_oc6Tk-jQJ-)|%VQTLBXtQ0;Y
zA|tsA<>9jC5x>-7hcW2u80`l{v_2vt-j}>9^$zhKQHN=Rq_Dhqfn(d7x#pu^;2TeV
z5_8+GptWx|OUpgf3NqP(P1A7O6M67^|A`Zy@<c8gblK%{CejX9{_H|nm=DF32v*^K
z_r5b7li_^A$3F3I^m1@6ukl<*rzmSbI>wba2g8HaF~{dR{&G;|;Q@#7tyD%26^sQj
zM+BR+pgqM24}B2kE_dS2FL@3p{`r&WJ5Lf3btxs9!_Sh8{Sz72h&GUKArXsCPH^lQ
zkKrC~dIMV@{b&X=mUJRo+K*YX6&G2@Iwh5Ou+$J?A14Y%%Kbx(A@>8^z+taj;&ab?
z5%)XwUbsGVX4?#6h0yQk<V(w>I3`;iw7Z1QdKD#7n_P9X%D#2Oof!dLb$N~~&C|Z}
zYW95Zdvun2<co{s#b+p{j#8s$#n0w_kwokeqqx4uIauy_Uhtw<@vN8pExtWPACFIz
zR+_{ULdczq4pGJAwa|eD`%xn%BC<J&coIBf9L75Gn7I*2Y0c&!qRvLgSX7RHQ1Mw7
zZD3?Z4Bkbjry2#VXfP^|CfP3<E-Sz(b%WnwbX!Ln#NiB+{k0@b+LJH#^=RiR(bOXE
zE%W%t{7K9Ssd4*7S1{?-|I(d{8lA8{X)GL|LSqNs=b#>A69)YylBC7<UGqHgsn6lr
zFMc_DyB<3?3l<JLY%?=68}ZJS4gc6?VQ6nErtJnItviUEohbdKF0#<$+&908$DjE$
zIQCdv+maHVqbjdv3_K$1r`ios>#t83{u7YILR4mAP*D4j!^w=LomjH@eURrB^4=J|
z>d;WN>24Ber*Zr#9_A*PKIUd*swEv@LQ;19>SD;Ey9XYl%C5^Vr(+UqoKTdGNZya4
zZaSvC(EVSN)%6du;(N&oUprneT^(3j1~%x(ep?8NrJ=u!$B&ng7%72I>?MOKVYxfc
zbe40-HP`UvC;u5c{^>(Z^)eRr?!zVEQ_#st*f_@#Pdt;;KYlKAfBeVnbc#5~<zN0c
zTH?u#pjO#11NXoGACYH0=9icF;+KDjGiiBP)YXflej&P_XKpo@b~t#Uw6KoFtZ|)d
zJrAMQ(5|7u)`;Z!hWpFt48TzE??d6~1^nz3$DH{Tj(NcgY2D*q_{k|+CP8YbtY7J_
zudrl93-28og+nzZaEMw|CsAhsur}rtNRgVX4X3jq7Qk5PbyJ`nC@-OA@Fi_C&pnw>
zu(Y}AOJ8DY<Kb}I+tZUAQ6=#fa;4pB(e8Em+*!}#t7ko*EB^1tXkEG;<{f#cP($NY
zSGqPniHvQz5B!X%(@kSrVk}&71^@iWNAvAxzL0m`_daxX>;PPq!$V#CEu{veY#gl6
zcd)E&zW*PWaq45A&IkVe+az-v=x2f^!O}AkgLf`y9IPqrK&yn!XwQp9QF6_Kz}!RP
zprr=ma*U1U8KxkE1-g<-UQNHo?H4)<Clw8bFWhcobC_aisgP)-)RALwrQkXgC=SHe
zc=U~=kQL9Q;c2zXD42C%=e?}Z>=(~faqcK4<}=vDuyw;6w?FZCtfYX&dxwY}`V9Je
z*@NpRgi*yvOuUXZv=V8MsyRubWRkNmSVlW@oc)2XaJ$pa<O@H!6f=7`%L7Oe=+5)X
zm;NQG4N2>{b0hoZxC>ocMnh>6jR+1Qa~TsTUwq4JIp!9}ap=A8$?}9Hu}Peg4FM!7
z>%nZlrwPN)fj#Tja`g?Vp#v+zLJGRf#!a|(iW<e_Ib1g%#BXIpxt9N*@r>W<%_W4e
zS!2l)%f{Q>mIa^3DnUuT@Vg)X6b5-EI?OzKF24eCF4%)=NCqv^RmWBo!y<|`q&beM
zsK%mBimuTkX0^>zX;v__<DLKjAOJ~3K~y)j!Ez%gh%&%Mq}a?sNCN8Sz}O^kA6piD
zhAi~BNgsZ5&inZHCp?z9E3YKoJI})MB9eli>X1)Q!Of24POp0{4|w@YnV*>Av+q3z
zdds9fIAdh4%QK$)c+h2xTHgEa&*CL1h0P=Tr^>2F-VJj+tn}+Z8;k!z$jgC#h3v=k
zrKaj^7?2hT=2Ei`uG2}P<*`>GkL8t1Oep07AwBA5o7A?(K`nI~#2OMOP03O-Cm@Oh
zjVXIvAE`rH(<o@80T^ATqgBdq0al1es{psq#t7L`23K6hMgRIm?)BuSah-3Iwh|=w
zSoL5$ZmG|u-}n)i{ri_W^@J0c`_0w7?GX>d{NiF<FHmjD>F$U-*|-w(`aCpEuXpP`
zKmO>a`2*YGIFqtvex9$s`+ZpNOT9p`WoqpnDqyU|;}{H-Zf<z}+y9A&J>eN#b=?xu
znWUTLNJ{{Vs3CKC^!S(ZJ*BRhE1SB2fSVjT&M?>Qb6Bs(^xo^(D2jE8iS)|yenl6o
z(m6_Gm0i3#LyB~AgQ@ZUC2A`aYl@~RCNKx3Uzx=tI;zN#gW2T@i(DC=F^qyPqbJYm
zNPrT;?@GOe6}tUCr`+)lbP~lBA+5%g;6SxD<v(f9KOl&?S_zR>7<392#Ie-t0c|eZ
zwwwDr{7hc@`ZtrcCXsX!NZ7n-lK%2CPkP+x-0BubW2_JEW=2Z-aDxWjz!0^3zPzt$
z6UAWEGr64etM|Q+#ogC(m#00Q<>`cWyM=SnreWB9YyBuY(AS=|#3-ya?(RqH*K-Xe
zL*tB%5555=TG-hcFwnLs-Q7E&2vcjga_9<ZoRphgX>aEgwVp%&@OCT=`rsViS=yxR
z_{~Lhd>(azf&036%{JOC8^_N+#?s5kky+(13}q|qcPg;L2BggfVg!5|YD`v_VM}AW
zAW}6^!^m&bvbN?i-{@fsMoo0BFjy6m#9|VQ?>n|i!ZF+T@P*S)XWKu%gSl>(g`GR-
zYetfS?@Z!nHZgbFBRK1mpJ#gOX10Ixo3u>}yeGBFt&Ta0<Bysn$-4ad(#yE;m%D?-
zDmhJ-ZBb<SF%yCtbpEx<zqv8l`Q5l1htf(|6^%mUB152{`!x&mT1l-M+<!F$00!q7
z9|glw7E~H7lYQMAULX-6fyzNd0vCFu5vbWKs}48t($J4qxeN6<SDo7*zlRC=4`m}$
z2#Adg(R;M14$vC?L5BbZt>|1OT`~qT8e_0|#?-Y}aN(Q(o}1^w((Fc(xfV(u+<YQI
z+D^7P_71n>@FNa~ZM)d4o(G+DJpcaAzo(Oc^Le9C@pbBg4y`!dpI_VfwN*BNT5I0C
zBU4Lzb~Aexm)W#=3sZK2!Jvn+1(&_L_Ug54%#0Cw{VusAT=3f~IqgrM!rTA(W2Cbi
z=;fiEN)jopBUQl~6NS-Qik7Q_xyskjI00#kWoD3ZM{jt}QOEPVTl^u1?A*tsH%+_H
zVgjX=edt(Q)LH;t@wI`H5?q0Tt*Kyb8a{&sXEJhfNDV$s7^p|pAew+rK$A-Ev7ho<
zSD0hzuuQP_e9D;ns0Xw!gy+zEAMJUAKBSjrJnGR8Bo#%xY0wr&>2UgAxQj_e87vKy
zqgKHKnj<DdtYhGo@mQ9<;jdnIHm5$~DO_~vbtLH=zH2ZZtmJfe?dJBkI+_<g>&aw;
zK9)3srcwpoHTjlRhEiET`4Uy}CVtxO1LnW^J-+?LFLV0eyqex@N~@EgQ8X^KhEmbr
zd*Ai{>1$0j{#qM`AFf}|wIPq?K`R>RLk4WZ%%Ml1VrV5Pd#}DaSpF3@WyK_-=I&5O
z{z@th!8^Nfx4;N<w>pk21@(fGA}*)D^D5ZA19d(W4BoTnvdc*eWJ$G=dfFhr=&-h6
zJfA=h^ddEAPO{=4VlBn4>KJ6G?toCBl<{Y&P+Tcd+rkJk@Uvn_kwAtc6G0~va^o26
zUE=uGX0G|rC;94Q{)Cwyf18Qxu48fcJQ?6xz{Y9(;ahmh2R_dBeI7DM04Iil&w2i{
zpGLO0fanD8|IojZB<%>M_!54tEGG9GYjcn|xUABMjcZW$C!qfCJ2j9I<Z$$v7l<jx
zq0sl<<J9BOV3nvfZn{<`s@kyc(592QnkqO;Od}+#_$m{h|J-|m(Mq9jh(&J{UusSN
zP^GTMROoVCU!%n}lnmpFe=TVj6bPW!#L`;%&zDWQRHKFBvuF{hRQCQ&Kj&h*D-wGV
zhwU%(qqo0{V|VOkVquxIokpj(@V|<Mm^Z-Y&2)}A0-x<g=YL1OeFwdL*Fo0DND{=g
zwYpYn&AN^}1aiHfDvd##IT^Qk(&==M+{netySdsLPJPlD^m`dDy2vPsiy%|r_j~$-
z0X}K-=5s#4!_Ig*mu|a`bYhZTKgXm&*Am3a@wtopd(psY;o&)4qQ(h`5sy)5b(KH8
z?VUJrq0e2-Bqwj!0)s((4r8;ZL%v2r4f5Lms|XhqF^G7)=Co49%7DZ=Fd5D%s0|Ya
zscC6M8*5>;CP9W~v{DDAqWD_&!?)G7-s|!?HsNdr5^Kp_hFC+Vm2mgF-5zV|q|-*a
z>6rJ`8*n!pIX#EYdn&3|NevQ=F*ui#y9{fm`21JD$0-l{Q{MBj&tNB~8HgwIJ(M0v
z$KeJ8CZxv)-+eZDw#;O!jd~X*T|#MyP+w(Qr8PFDY9f#uk3;D1SjNuJ^O1L*!{eX(
zQrwo!*r_Bmuc{^e8_{MHxQ=-p=xa?i{+b#F5&6OT^;`>9t3$P3K=ajjrVcxjzEiA8
z*|+Oj48C+TF;XNT>ZreTZ4nec+)4~HGAV36jH#_hk^w@B6=6f-*>~ajhzF-0G-rAH
zcGTsyyoHzrA)^z9Mhz}2`}W8I?`e92T=TK$Qaxp~L^JAlMJkk6wQRUTL#1FHDlve{
z8k{Ht>qsUoF6%NqFdVt7&v#z@YA$%m%b4DA6?)wsx;`UIEZUx+mkTF7;?d0S+=c69
zB+?>to>T984>rxVQBC;5SAW2@i;(A)^TVk1?0?l$yVkG&?>BkSwf>!s|GB*eez@^r
zLS?Ke8+j3_1qEqD!lc=>7R53pWFK=lYR)sm_LNc~wqgV|K@d1Xvxxgda2p~bxQa81
zv6PNt6;E7OuoNAZg8B*(G!>nq$ZwDar9q<-I)<=C^t-TUjINR~e_?p9(3)970%^+B
z%oLaHx{7Unp3dQOVZwrdH6iI*6l<WLDi8U)zvVMO{1z8nb}_%$w~q%s`?<KJUEWGB
z;~QyNrL|sr*;aL!zUDojwo=%*k%#~N>p9}g$MNL%zlYWdCt#A4#MtN+uUI2kW9a7@
z12J5^caaC4@eE$`&i6xS8s{y}=U5X`i+vf4H8ia(!ox&rjggjKIF(heht$)TF5PT_
zb}M0!D@%DsmU^7k;P_|E&{@%>a_I`EL=%*f((Oo9X|;q-Dja*{Rwl*Mj=2!71lwHA
zf;O&=Mrhk;mh?)A;*7~{sQ*E84tM}NNNqA?P0%Iexg#+`zt`iGyWEk<b}Q<As7>&d
zF)^hF2@>g=&&{$dDRb^B(UB^6@u>AIYlbEhT)CsiX@B|x&UohEvU~3YVmH(6I;;^o
ztroVWn0`*)-OKym{Whjjk0r%>#afGZz9H?BhRtmRKI7&UaR{6dJQjLACM4%0fA>o6
z{oqG1amN#Jv#pS=f<f>lEwa?f)P(cvuLF3J)>z}Ob;9ud^-H>XUq6!CTzFPXI9=OY
zwvbs%yVYiJ-5%(6qp+{Onra*1I1*kIpcSD;s)|nxho5vix)`#7$0P}p9<Ke_&mdEr
z64IrY`Y|UU(2=}qod{La7z~f}u4j?2I(CCo9hwfTu?rEJiH$3LO_9ZbRxE@jZ7hyo
zgW9V^Sq#CB5O1)O;*B9opr7OndKnutOa7A!_{?J-4gdZx%=Q;pUYMr~=oDlr%-&=R
zA{NO!9g~nc<@wKkA}agHT%S*V_8YX5i9kazD+#>Uy6Ncw9yqdU%R7F$aU<Z@-iX=@
zUljWT2a7#yH^VRYgC(VNfnwrZSu_`;_~N-NxY?>w3otXy1)wzYSt|ENDFuP054_ig
za8QelT^&M(c6+3`kY+-^+XrjHG$iuV6~BVvt_<ZO)P5h*0Sw1pncCNyxDObki1Fcb
zsc)zs>Sawqi?FJA2U%x|+q~xGocQ@qbGy%cobIOSGE2roc`em@5^M2?9m^xn{RFpo
z<!gAvzx)e&kGqnmkP1x?{T0z=JzzSL)T-4m39ci^f)zt%4fdoHIQ}VTlAd@1c`8`3
zaa#`smWZd@8=y&xkN(S-xZ8am#;<<6oz}z*gMmY=M*n`|zb%SgQ|*Zjt2ke?t*k(&
zA}h#?&{s!qD&f1kcXH|6BwyZs6~9<sB%ie9graDKeVOA>PwkblK_4R-8->%8C@q_i
zO3r6L{$4)ufpd8G+un@L2AG)Ytv-79*9HS7PT<yR1u^1sRB>=t+0NwP3IQeZ^@D9H
z3JiKY<&4uwOcCVjV@)uw)jF6qRS)8$Htfra%sx=C26CU#_X^VH_3wB$cfQwyIRDqb
zqcbzha@UcE#8<4O<XMIlkM<XM`WcVlu6H~M5uu%?sOBghW2G{7X~^ZOsfJR8Dyj({
zNq(V6tLOOq%m0Qghu?(b9`#6iGfB|E6m;*nzPwkQ_=BJjuD=en#$W4%VHfbT^=rDn
z2}4t?=mGT+L7baLJCits31e})_QtsLxVf5Q>|5J3R2#8JTDx1$ho-@aaM($=XDRoz
z+ieEwX<OmSpZtV`=uohCHxoF-=OK@vJTPkReItnC4Yd3Y6g^+n)w8mK9hWG*;&D+E
zqBneMN`F6(H!-Nl8;nMdIEqt`PXtL*v|~VWCYN$fVwNAi^l$mmW6xmj2NyE2YZu+c
zMFu9vDYViyi5C*DB;N6a$3Kh*-2Ilw;0oUIrdPAn%V?2u@YB<A2hBKpK3%23kQJr}
z!_4IMjkvsevD#D@oLfHeuD5dDCq9gwXyLP92dbhNYfCiINGO0*Jzk=A$yDKefzIQg
z%vsi<qP6<7Qqv&yB;cw<^Kn;K6qg6Ol)dt|od3pG^V2uIf%bei1}%kjcq1jzsYw@L
zDCM{z1g+4~;fS%~-?@^9ABq6ym8_T1A2|Bng~Z_)Y*hjmrM5sTTUFz+C<D`CFtZsq
zxsj9v<Es!TBM?R>A)Jx+B-z%RFnQ-w@tco9C)y-lD<`KlJlQf5q_)530=x<L#@B+4
zH9EzrS5hoKZ<G5Lxup03fr0~wwG6T@nNPW9Z_cCs>_t5HuiuEBK7?Kt%o$o1@B>=L
zAP#&UoZgHMdBhY}ytRF@^3g9u%Lc&)_a;r~wc31P$2MMf(fNF;yTG*@H?Zt1#<T)B
zl`D9OoN*m$DeIt2t;MWq985fHtXM3DJnQlYM;yVSb5pdkKDRsJ7EDidFa})av1x)B
zU&H<_F{XO7*w+0+AkEHhWor#RkaMk=B*tFEdyjV>n^?pewr-f=kM495RaEL)GIIWN
z_S)fSSVNKP;jzYH$jN$(v`m`|e|-g~JnRX)@thBnZrnl!<YXkR3^b>mwj(g&@R{T0
z$86!RUi3V&!GM+xLS$>x2$0nBn4+etd7XpET|9^he`j1qr;~90KfQ<Di;LX%e?Oo8
zw4s$IBq~@Hj2bK{AUNakV(>;vW$60rfL=d~HU55I$pLmCf4F{a*RU|_f;kh3MoGfN
zp@-rwzlx5r%x}A#xucH^cR(w|y(-h1hd`N{TJZ8kW5Q0(rrX|zuJ3_KNLvZ&GqzuG
zIdpsY3CsNDS1=`(M5DKg37UuDreACCuTmNvUhl4Ibc>eJU4D()FQGotUF|)!vJ<Qw
zqS|POYU!0xZc;dhNJCRnQJF{#B9`7#mm@k`$S>K&m!EYO8}4~`?)coZ;3h}1G_cso
zDUz0e&q<8nG~*xM{&IFK^|<Pa%gF{ybdo8=My~o;>dTs$_AC6^!G6t3H|6(dxq7h7
zUcDjTxu(N082tQR4ms>F?(vAz$(MTAsTMvSiYD^PjS-x4V8ibYg~Wo(NlX&CJwA-)
zbsXVtbkd{s&#e^|Ny}y-G$PjL_zN%Ky-zrU6K4*gn>#-Ium8&9KKHNWM)66oO<aj~
zID{2f149{I*i|qSwLJvzj()exgi+YNlRZ~m%f`dE!iGaxo@mjwDJDuO3a8<60#nQ>
zL`jR~8hnr|ZoA5es1{Wc6GP`j!D(ezy27qjXfz6|<{@>-ejkj=NJtyIwkoIzO<|rZ
zh{0)s2w1CF3%;M>1p3bM?*IEC-tg`Zkhdo2%xz{c5Rz7gx&ep}#XA~&{=JC}Vsot)
zW)O9Up~}Wtt2`PFfR&Ymffx2o7?xD%1KtFn%R4FUyhW=xW36Y8NFA=wk@8eZbkd+M
z-k+q+j$PN`)YD4aT(ff*OWj3`Ovg!GPU?$^nUo3BS_ekz&O2)^nPr9DZ5TXR-RVq*
z+>MINHjgJw+bk|Da{7Z$AtA>~Yv{Xb*`$wtueIz)YumHxMhWV*k%<(7Wx*KFV35&j
zb?D~sl9#-i_kZq7Fgc5~rdS$yOlt6jqn20(-5w@Utbp&%^Uk-vi_~Pa(>AI(VoVvF
zT7))A>GD|Y6$KU7gqlQ@93KK8cV}GsZ(ra$U;74U{o}jnwUt&U3AVgGSk{`-@FYy{
zd@WsG4*JAcpDx^B4TFe$S5=?6er;Fp6udScTCbsKP!VR1xCwi&zLt)9mag5#M(1%h
z6rgq2A00FL_d+Ag#HJRXdBiB(<S;s0x8Qcq11T2Jee-b5Zp^Vqvh%VlvAM(efMBWz
zW2iX(wIV=dTmrLPuj?X1&`rgq92#+q`&MHtzH9P?kuNM=%SDEe!O?%D5nUCZfCi2?
zHb=b2XqYsp2b|)_1Cl8Tna{8jDcwB7csR;Vv3%YS_}rJj#$k8AEw}sAGf3`mXZB9C
znVgxxtKc+P+HSH5n~uF{JaBXk!E7r{4{kWjY5#}P@1K&SH^9GX&P-(F4`X0!^nz&~
zI8u9-BV!FSJHsiDJCnShVW*PVC{Q-`#Dh_tN4iR@unYB$d={L5@mimqH3a%AB_m5r
zV-_7sP$@*laSdDB;ev0T$DI#39J6bIsmYYxJFg_a>JsFZC*p9W7r(6W`4XwVdPw&f
zi@NZ4rRA!J37hci5BwwFfBU;=_l0Yg_wnH8zlh_W^(-<Q_9D?hk7+a_qIeetyk#RI
zK`()c*Vs1-{bJoqoCb-dG!-Zd+Nx$Av7fE}y{tI$Y_5#BSNp0U(h7cw#cuIV9<hp8
zhouE%Wc@BGEq?y1-|@V^eht6A_-bah9D?Rb?iH~fJ8)Q^W3q^(+W0KSwBWjFLer*K
zqu{=F@3g*~M9|0^jD+XW#h<~ta1Xt@kX%e9rWcLr(fAr-uWKmNqhW^&V%2Rs_wu|K
zzluBG;f}oTgCC~nCvX_l1a~M;2nB595{WPGU8QoBRjrK2CAp2aOqB!JIfJ0w*w_@R
z50pEvXs+~nJs$PQ2hwRJVG?AAD2K&as|T~w?Z45e&jdq&YmtEw?wPfg%=b}R^mKxM
z{^VDA*_+<P;$F|h^bz!ybEM@lHb<Retiibfi7`NqF`j&>&)>fI`5brb5u_sIdDOZP
z+>4NcIvLl3G##bN&uT)}f?)A`de}>@=2P$cXP*D6zr)YXV5iyvRESMRvp(@ruN5_*
zaV^Mtv#|C`K;*k?{aSMX@B6*=OS_+$Gz|Hehs4J)juTp2kD#BXSeww_w+FJk3QUqQ
zAuhG9xg7l0zRr0?j78fWw*288SWE;p7AuC$oe8e_(N8gXv{W^5Fe<?Tp&F|)MDCaU
z2HG4{;U=qWiLy$wAp09c$OzIie%o#4OK8KHr3l_J(W$0(K*oLEl3<C5n>B$pVGYtU
z2z@$P#?9MvOrQT7zWcvl$QPdQFg9HAJIr7Zic6bdt)bINNUaTH>za#BeTy3>-W!c^
z)9Z6L*7!c-g5cx;rWZG|e!gA?MggTqaZyNFv#O7@pUSdV;(??k724ocz<OlL26F)l
z;(?Z~JE}_J1-rD5uRrrSeC3HxX7G(~(aPgtCmv%=t#HthtcD_;SBLhOVM<1*-9oWw
zjo|Z~Ke)}UxqRPlT8GWC&nO&m1j#Ws!ATwt^=P^v4VU|sT9xsRICV=_tu?{k1#3z6
zb@|qNKFo>j4!7B`i3go{0$(}j?fBih@m@=N>H4EpTAjv8L1i0T3weVw+^w)yEdJaN
ztr)4VFrk?UtChQ22z)Y)E=*T-<1aq@kPm$1A_)>wXjLQqC+sDzyM*h*p5>h9zT^#@
zcKXw~bjJdd8#j{mbMOwMkoJ|yJY(a&d2ZHIj>;Xg*?^SLh?aG;%1XsR)AGKaS68Ee
zc?2ZEtHt}^j3l*O7!_CM1$ptM^Vj$*qvesPwOkP;(za3w)Y*wCzJA`1_^a2xfgKCW
z<YLj(#(PpYKs6g6p{eaz)3!6r!juDWL0auir6$@}2{c6S@p+FUx6E?Vt#5$=thJOv
z0c}po<@%kfWB<Ft|Fp!(#D|HLwc*_9`vK7A{9jzc|6}jX!zHWAd;ib7*4kBdhMt(3
zW)KtvH436p6i4EO6HcfhiHUkOi6KT44HABuXfS$%8gspIo`^%7O^hfGs3;D(C?>`U
zR8T}lq3P+IQ&oGf_5S`?Ywx{wojRvaH=y@EPnQqcHr;iq_TFo)_kF+9W1jjyc>Y<h
zpwu4HpJ8cSfsJT_`uI7kaiDMoq7|knDHq4w@195V2d6)Yo{8Fwh=66mgCVq;AJVSC
zrrimNvX}Nv-L{0i@n+6@&EN8j=R6yG%pED`iWo$wE$boAe2Jpz1hO@QfPJ|_d-z&&
zZNKJ*K}4<qezfn}PTyRbV@qk^QyNRdbeT7-qa4HlHz~*H_FHJdaA^*FZHko16FD>>
zc~HeT3srPLIsCy7XX|)`pg7~07vY!x{cQ$S#g1#QL)0e+0h8GpE{BhA(@FBO`gVe5
zp_813-S3NeF)bq5EU!M?W%XjV&Twe9uRu)02ImFm9Y!6hCQ-D~eH8<0BZoJ=hc!g|
zIR~aEpd)&tF-KH}L$A4s@4xwNxJ8c_A&7~Btt#M(9#S}rvB~9xj^Fw6?M_$eq;9PR
zpRDnuU5mI{|8HHMiCZ41B)PDl*q~SqSOptQf(}BO-229XMiX)vlH+6DXh^SzuS3l+
z6TJ7@F4;o}T=Y+G=Wq)TKJt!y@YSzDZ@L*OB_x<#Ov{OGAp1N$I*I&px|}A_YCB*}
zR;S(*)T2FN?&L>umuLSGR~>p7>mT)Gp8D?d8HIIN6LkQM%`#+Ws**J%8TCXb+KH@F
zXzFFH3;fzoKtL?LP%-Zue#cJ6+jlT;z=Xw^wk{CFB=K)85T8a&7qTE^vYIGMb#Jd>
zdv>i-^VPPr^|wz`cHgm@tUc=|5lU_Jqh1tm;*?&SeaS>RTJZ4_8qe4Vb}o$=xdG?>
z>*qM>)ZgJF7ySpb2OUCL8OjjwHV`CEOPzuF@lH<JauDY{>||bX%7eJ;j3U({)))v=
z#M?S;m?Et+O^v3h=C`atW+esmY3;P75(`eE4#d;}Ym@MnE=OV-tz6dlR>31y7*-yY
z8TMPh1wVF(7BE%{wb)=HY9j$P6~UCK@gxkSJ6m6?&x5SB9avq>&;*4i4xowwELG(Z
z2@Hq3c<jj!rtj<o{idsn$@M^cj}T{#{v?8qprImEV?;gJl93-X#<6Wl`16;(hQ~hb
z5Bc#`H!;8dAjZ3rDS;V5F($}}N);7{m;zsU&@sUa>jsv8e9fyEn1BXH2qq3bQq>+b
zBwc1`4)@y1T%dVwYs2Pp#}dV;<X_HxHOJlWJ{<I@hp{x*rzb9U7M=L`%=Wnssno7^
z=Kt#R=>M+Qk5X-an<ESWU)y(mci-)cb`MjnIg5p5-R3Q*5&Fim<C^P8TH6%*qD|0J
zn^N4m0@agmM5%lgC*Q^)&a&~I$1xsQoG4U=dDr9GZ~qr&X^8At!1#csMA1-p{?j%`
z-P>=y!k*S)wXB?-u-Y!JE4}rElT$-iT1Cnji;+A)M<-?$6Wmmh)`>(SFI_`OaY%{}
z0X2$u0t3sqXIKnEPv-d9#b1Y?{1hF=$fLMX28O;wu?#C8?_d38GG`oVeEL`7w6A`s
ztthXPy*-rv-<{gjIVZ_eW8z4y3#>LUD28qcwj<zd3&EzSuJZ`;<Xc)fun~zm1y^vt
zN1n`9^?d)vYk1=Go`r1O)Y$)OeaLBa{LqO}L|t?>gNIDyoENQ)z|^D$##%}XcYfY;
zc*MKj!SQFGNjPAhVlD-HN$1;i9NN2C4AMEM^ANDcEhj(Eo5TfBYhW<oE+;>f>lot>
zTgUbgIOw>0!F~rs8CFe;N|jhsyl0}>#wIkUk4;JaT^;vYA<`>d)(|RuSz@XR87t#)
zm1t6v1@T0Db$VV*`n5K~^u!Y{ir6ev6%_^Fy8KEWd-`)Y^VP3oY-R~P$3i*6^&)bm
z#-ao~C4murzr-Is;$$}b>@x1S?FJrn*cK#|ltHn%c0rd1WiBh8r2El3|3A;*QDU*4
z1}aKOh+9BR`gM(@i7rbQ&(W%;sUiyhF@o2m<uQiQC?Li~j=ZrA^3~a$B@^FIoF#gj
z8Vl*#-Xl?8LJDn+VRSKvVLaT)=}&(G&Pn3^ODpino=Tm5Nbcbm$yyFGx#N3$<td5*
z6=gI~mOk(O*aaMS(xZ6aCq9dt+sIgziW1W!1?o9`>7!waL`svjf>^~Plsj+X+*h8>
z9S^uVX_No}AOJ~3K~&wJ!di^fbhpra4U=*y>c1ao5!e_^s8kX)U4C&$AIleBb1v)V
z=egH&p2y;#$IPHls6zLgSEitZtY(k3?{yp2^lp#J2a&I;>a+G;V^*C$1a0BqIzIu4
z^rZPMhp^+SA2TQhY`gwiHt0PG1PnIHf2I}knQwd0fJwnBb)ji&fgE%&`yY7};i{_}
zXuUVu0ayGK*|iwqYvb^tN}7q~o9(hRWZARbc>1g)x7X9%qpH&uzD}btNlOulTEfOl
zdb(J5c^^|-28PB9F`Jr6CXp760?az}N28cB=2(OXy*i;1k619qfwk;s#(eK1|IFQv
zIR`(OCHNB4EBMHVKgofI9K@p@cp|&XlD<=%b&VBe2wj0+SFYW?HLcB{w4R#61S>tC
z0N01rG->^wOuPQy5S4W(DT(G#PrpO(QFLzNy0Vhm36+Qjg^7?MiGo{coX~2qYZ|-O
z;Y4*29U@Mlg)n%)iQM<T_lE7;VRjvrD=;R@R7jZxbps(xIV2~nD(i`NgD_~OK21q(
z7En8jSI}N`9BP^c$x_j!ML0nwl~n1y0~ryyEtVypn9@ww1Bx+$(6^lU7cb(gjQH=%
zf5_&ekLOWmzk;Ee$0f%fY@)m9atE{=O$^IccBcDkC|R=n8PM)W*Vc?Qx=RHx!wPfl
zRa}4B_nBY6Kl3M?z>-!JgF*79kj|Ac^@JkP`BAWSV4jJ8AgHjgFrpar`QDYk;N1W9
zb}qR1%gprWn3>y;rA343MOS(8k=juYA;6%85PE%na^2M&J~K;M8g}j+QQ01qbqN8`
zCK<c_lClbzsFPcJ5=4huJRc$ztU*|_?Zq@Yf>4vf0GB8s6V;s7bfnPyeK9s68>yvK
z1sC10q7z{J?p0ON!yzeJyk1kyQloBY_QqOtwJmHd`&vvOS05d~f_jV?42m&uheI}V
z_q!eyMbJLd4%1@a&1t!J`>7-Wdy0l{27<<c?HQKFi_G+9`SPXT<Lq-@!zDlX2^%&Y
zfQ|!8Wk8Ck3o_Oa#vV~aukWaq#@JreCHV0u>LtgX)1LS+o_NY{;j0Q~1z!b>lSa2x
zhg!BPxi&zKt1Tor&>&DjvEcajJKoCmKfjjWecfwW=y?XSz4Se@Jdf@IZf=#x|J!HO
zuN8b3`ReV~FaTfOcZpXoO`54;mU=1@`SB_+bKpVTjG-@%rJHYt#YJS}Tz&(yR9JHB
zkO=9C#wirFe|n@Rl}<S9-uGeaHCJIPPho|yZ5!eHmoXC}b*2h|o-L?+mBdLa?mm0m
z9IRMs=1vwW-6@gYn(24@F&i4k>c>NNePeH6CPmjZT`8j!G!ocUoG)u2lC7gp5P}E1
zEvU+h8Ed)n!i%{3b8cp_uMGMFsFWu?`SCpD;ivJ=<L|=Lp7KN<^}rLD7sJR`^pf2r
zjkkVYJavm9(TOXs(jYH7H7GAL56pPt`7_pB57Ffl)oJR%-EPjm-c7qsyX{DeMF?fU
z8b|OxRv;$o5LC1?xz2#8c0FcIn5<BJNbT&R7LQUHkb#5#CPIkG%(2F6gRCS=xvIIi
z&E&TZSIYPo_4C+iF&b{UNsc;^|7#rhU}6SK8QNwg`+}Y4p)&<d&TL8tN$5x`wM#7&
zQEc*pR;%~}58#2XcvU=c!hnF_`Zle@h@Lijf{CJkciH6mXIya;5thAY!-!V!)fjWt
z5Bc{OyqHbnz|P?kiwCdgw0D1y$~mxZd1GB`08%htNKQ17(&HJ9N}L;T?M(~3;mvR4
z<DdUL!^$wXVSjw7jH`fH3*I)NopDdIsPRaEaZqluJ>GH2H+jZ!_hIuPhj7tFUxUqu
zMjGU}Mu0-=z0<7e$>~DgT}noYq0t9q-DhW5RExSB8i}+6hb}K)Tce9C*<4c5sI{)6
zL3iTc_o2;+2tje;kRY9XtkLkPS-!^Cy2s?tukIznkOEmDV8u~|B<4m~SX|`yp7d0D
zCK({q5$w9s38}Ywhg0ni60Egg0;6HcU@&0&&Lv)W&R_G<&wdU&*vQ6B2eLHsxWW)B
z#fm@`2*G1DTELd&I65vORO1TkJWk<&gJyZ{EB_WPhs^c{v01Wlud5r)KuGtS&U)4y
z!Ai}7y=lzNpZ^?Rzu?pS={w%d()vFAb^R!|*O)cZyic}f*9Lg+YZh+nntqQcNh0zi
z_URaQGq;dV@LJAy6b#JIqrDka1AQN0>sBOaT4eRu$0XP;I~EiFwBE!`N=OU=EP@uw
z9q#{NwwGi2gC5=)=KGF??|++O*ADO{#-_z=OlPHaCUO&n%WkJMyDJM-m!l1n@4Xw!
z;i@5RG8sO%wmT)#ePhv<0+Dk;bxJBl&1Vi6;>^}WF}Ws+fGAcJHyU%W7{2+gw=owC
zizvYc<_!GBAO9iW`QbI3b?#ev*y+#Xb?^B&H|z|QB9y)&gc4(7*cJGUT~Kqrbt0WK
zbG&LQztC_*L)|DflRe4AgA>#v(FtU!>+Z2<)3IBtvAC7%F_{DyvB`2Y6&I$ZoRR1b
zCOMHsuqtF}2+5I1bD#O72db~5))b+ZVpM+<i{Ox$3aus;Nt!i$am^V_vOCWB&bi;R
z=mZN|F{+uTx<(|#)Rk;{Zf$%LOe)BXmB#sIk{TGlQ~gg(wKD0MmyUOUPKiF#)YNFI
zUd0N=M56!(>!_H=yB^YWn4q}2=}OnvG<pJU3#|@)aJtO4rZY{Vvy-2&{Wqrh6G6fD
zdR+dcuX2R+IYicT*O>$0hd*KGe|{4S5&2k_X_?CB5Rzj~023Jd5|3ruu;R6Eeh;Ud
z{!HF~{%7%S20Jsua2z@O)>(pyL7FNOpN(K*gr6EDfK!VO*7I$zyyD_-aptGK$PJqg
zV@wY%E3lFN)C>zUEypq@p?3FsoyKnLOo3v^9ukXV5t1i=4-I{6D6o~pOb7y%4apO~
zrDWBJK)rryk-9M^-sjqnRH93cV4{9O1uucXgsm?rjkwM;*^_2r;tX1TuB@$Ds0nwf
z5WJ7md<}>O%F&P~Kj{>TWL>^`%R9-h+0A|?Q`?0&BnUo+;@QBM5gtP@1K#qUPjK%?
zJe3c9{_FJC9n3g5hT{tBlHF+m-gqjfs05r1Y3<LTra*n5Z~+Gxzk_$a;eRnGghA1#
zs#1(R^}xe4!IVw@yJTmoCfZg#RM@t_&hKB!`~LZz{OLcumg>MciVcH^)TqIw<m`IL
z(=lV&*9+VhuOFqF{`P4Y((5bxuJi5-qgqVYbk()3nGx1)*@8zI2<*7#Dr}Rw9AZjj
zjx4Cqk@ZSl*oNreo+cPZ6glcBbbdAkjwyv0w*B-<T&Wb+rF6t@@jES`NOyxU=n6Hz
ztf2tZX`6NCUbGtQI_;BmKI<BicXY)jAe+_ZjRalClrEm#PQB{Mpaj7LQjr9Mp#0*Z
zFCtf8g;e8MhWatT^V^SP21C)?pOM?hd;awzPJPCo^G9dDo^M}%HDU&gRq%d{H4)a9
zCIqS$(o17Aw>+eKGNx{skyq!^cQ4Xf@*i|+vplr{S@F=^UBvgT5d*fbpqc?)(fZJ+
z!YCkBns|mPa;)oSL0dL{Zd<9-kNag0Xvca~HRZE~^hBlqNbdA6sCclRp&v626;+7)
zfd)^|`0J3({aZeNHk_cQOG|@-L?JiH1v5>-C)Z7FgrEV8B8k5qX?UU8ck&)na>it%
zq^--CNJ6N^;-W4#D=yk*)&*|v&bFMg*e!ZxN}_b<$&vheNH-5>85^>tA6fw&!;E+6
zQpHRFmd03hEG#WVzGD)dH~WohV_O<0x*9p(!1krUdH;F=kNNFq@w&I4$Ic)yKg&>*
z$|$ZVP%%;bTO&d#6G2lTjPVwWMZ{rbhN0WYjT<-eGiSM34942%9urz!i#9#h3YxS=
zbD|l#)OC2usfx8vz7^Y=WQ0gvse1`4CX&yDB>t6-2)TK_llh0LP7Gp}4?U3f*PV=!
zYygm-U9yM!!(K~bwTrw-66%XFgy8X^<W6_k!rhL&6Sf}B2oqTz+U1K0T{LIKt#g&;
zB3ZW!Fs?>m`uyiluI359|A(CU%GWS7ef(gCU6qHzQWQl*dO}16eEr;|oD)d{$x!-8
zDU^zZZQFU_^Pa=;cRi9B>mqt;X@(t5=8?2(E6xnsdlV!@UkMe)TX$k^yq35A-OG8-
zSuZEt@kn&j029)>4cb8hWs;Y?Pp5EOy!N=3-{XcsMD{fdYib{~69%Lx##q!D)*W^v
zBV*_rW!LrBfR8+Ql{AfO96JINyKlRp25kh;sA5fk1J<+YsH0IzjDf23Y`^g)dd5MO
zZCcHtBAty|_Q;2NjXEQlR_#wKX;)S}t9F})oDLzm5Ooa*L$d*Os8^CvezJq1nc7ak
z-pvPT#%JlDG!OGp6fM?*C<n@jAARtBtXq)Cekm*!${A<;KIQhEnDGpy%rl<d#Mgdw
z9nU=bA2{{T&fy~$UB)5<0tHowv<GXF5HuuITx-`XX@XFj02?)}GfHyVsztK7&E#^_
zM|ab5$to?RuyTsDtb$4D3cq*j8uyxS9D<Kp69V`O8ING+R)*JKLl_RRrWH3wer-07
zOv@}LjFMWlc7hg*rB;}!qLG81g$vbm?|Q<jCJPP}&=9CbC1zMLqe8zl6eC5wf-Nv&
zF~((bVv_rlL>De2_>|WYB!}>8Yf?pwHhBg*;U^|aLwzboL#nDGXkb{5Vu)AJEdPOy
zw4%24tkJ8Qx>bLDlO<A1LPeSdYg&hwTHHMu_PA<zTBnh3H;bm3$en>mXiWXx8_h$Y
zG9wN<{#bsoa07m}WU;@<()<hu-{YQ)8`MUU(_AJ*H^Nv;N-W#P!pAQ977zWM=kX_h
z{aSvub%f0A&j^dyg2ELziOzPat@XC4V$!Zj#g-ZmCN-@JRat^{_(CW}5hsLk+0gB3
zizpFU)dWpDLgK9Bo6x0bO&Mw&tbDYF^^qbn8jo<!VhY2k8bxF-ZGT25FtBEQ&_zMB
z!gpn?Ql@BTIxy;Pt4CQ4Vc45lghV-d?2X#>WUa8Uu*e^r{<Jua7t?&yF5cB~2bgG<
zW$Kow<ZOeaSAaB3h}j-xD5*r*wiI~&U;iT~Klu;1^zv)zZ8(Tw856p#O+|ebk#G|t
zI#g7KUWlP-iUA2YTVW;OihwOWk9))e_@ih39!1O|#Rt<lK{{DZ^}d^a%d1b1nZaYx
z9VLUE+j!fHUdZo1=ef+@^JuF5df4_siPrD=byC^aB;3}Vn7ydw_aO|o+0~!eghaXv
zNZLFGd(c6Q3x^Sl-?0F@ma?!;=_W6CekG0YHs5eHIl5GZ<&X!Q%#P&9FzENFh6{|A
zhFF)u(5(mI3S#2DE6vvahh^>ZT3J5eS%H+miYL&NvbLkqn>3V_<&&UBbeHe@Q&=N3
z(-x=_Wo&lr_}OQ_3RmAiC`WjrNB%RP@DL7LUqD$Bf?>GmDNBQQ>-o{O3!L-D5Af*U
z`zy|U=f86O&MG}gV_=Aw5yq6Lrb4(bWNQ>cLnG>LQ(2|ak#zpkbQH>TxJLJ&kYDS+
zyX{1OD>v-j8?8y4YO##A-pGqje;OBm^5ZzyL&wo)UOFjgt@}^C3MJ8Ug3?fj@{lCS
zERq{>XQGgyMD$7NaKRNty{D*@TR!zEKKGPo@YUaW7FWFPH59k(K&x^0&t_->k~_#`
zdqqZx(tOe$@?go#a!3Q!Xas&lZ`%#*xa8aX@~am!7<-nMhIo@V`r4!`H-BGr;_A~R
zvNx#SBzGw}BS{l@H!(!5(F*N8ugmW=%bIyzAtAdf(P*7_OyMy1JdR_3=QOU^WVzw6
z{dv->UkP_OG&Wp`P9ub-FfYrJfU>Xz7hdvTJmp!>=UHd`CD+`rmAUnsK>Db0SZh(Q
zO;T+L2_-af4qaR7#zwQA)SK8$iQuiJGJ-mXH${}PI~zd{BONQ|)sWhnKGy(2no-hp
z877eQB?c1=Mgk^!<S=l8`Z5NRxdOx{L3novBxDq}X`wpWMG5lewd2G!l|Czbc}Cre
zNU!`e`}oXS7eiM=fce7mlqWtWB35Mf>XHY~P5bCNk-O3&Sj*Q;)ObX)f?*s+Q~}iX
z`PjdImIpojNxc7KpJp(>5s#tNlnRcE!uBd8Qx<7xkcMa(uIaluD^PiGo=^%0tTUW@
z&i}_;U+7r}i7MOPqm0I!Fon$PFIg+cpH*&Mpcsz$z*%Q;(nB7`L67(?miFt%2CJbo
z(a`*MLtohU`V}J#dvzN^2;T-y*mu2|dU$0i=-ta46cG=V=jzXXmUTDX%wp-;?;#JR
zchr$+TG-L5Ef)PorQ_usXa$gU>M8*Y0Z+hu_e#F>+^4gp?6Ei&2DYNtn?Zw52UDK>
zDa*L|w@aq4wwquR$dpY%*-cQK>U1r3GhJJ4u&vWui`peg@(GVl@?o_pwnR*(cTI`2
z+Ro#e+{w(%MGa^LRt-Cff{mv>p1Z&J1?-&dF>ea0vFE!#xQ3^n@z>05K8%GbF!mvR
z(SYd{^m-7qq8Ehw9D7H8`*9EDq4&KDedkepY(P^EK=5gcu7#JX?eoHP`We~uw|k-G
zRuuKJ3gWPmRDPvDt9$jhEnAiK+3VvZ?S?CV#JA4BfXDvfA2Ae1Z`L&l!X(O!m1O^z
z(UIa{QPD&NY>>KoLY_pU^|Mk>`~!khjZ-vgE_N$WHRBQHmpl2;lTPP>>kndd%K|^x
z`g2bBn=?7|_n%5BCO(7GEgo8QNKgss6NW?qA?+D;a~smn?@lTcmjP3j{Osc&;0v#P
zEe9?XRDO)V(;ax~CoW(#Yv`HKAPJ=OSgC1B+CEF#i0HfJ<FuzlSGe6Ob!w|Ndtb9}
zG%iKjID*AD3=JJAey+B8Xv$g5RzWf^K$9q418y{iGEzrNEs%MK1hCC`O)0AqT#sE#
zB^Q75KY7g?-@~`R`#lCT>nUaijJ#rOa;s^Y%qVmZb*6_zA1Q;4z-0Y>>-DSJgi&E6
zA<*f1nbf3)+@UHf*E?O&<esTr@w0ZV6m(+A89;rBD<BLPnJH|VLK_yn#S`gRgo&we
zGZ^lgC+G^8(Xt`XBzLtVdYv=?o}}5Xj(6rQr0&7?BC1e_EUMy6bU{*QC`V&XIq|NX
z_m<btONgA2i9K(sfKNJSTz5R9^`(t_ybpLultB;Ua>QUT&)2?t1uy;E|H=0+`w7Kh
zJ&Tn?6OA=K8@72yLV5<&M9oEP7EE%;@;EgSNtKdus9?Onzr5pIe(S{hK(G|n5JJG%
zWV+C-v5>9vS}7Loo1^2FVYsl!obB_;Gyj^xk2&QPFJriFK+*4E(z-C38OMZ3wQBa*
zeNDq{`T9>|%)NdU2m`=H`>yv&rIJi|m^bA?ZI%qqvgxopQHjuV1zUe{H6}!MvQG=I
zLG5!9b#3<uU8~aCddvuN#3A(d-!BfRO!Q7xtrMq1*NtUTcX4|(3)7~HtR`RFU0bUw
zPpGHyKUqt*ara1(m5#K`^SN>C$bMK|l(i|e7O}Con!|GW=ROVBUWXfbd<Ya)x##hB
z;KbvPA&f^cUDFwiD=>w_SDuBXF{N6@T5!n^uI2@Q_ck8$?3Z%xTmFqJuU`TwqWIcZ
zc+r@oipN)>>00WdMRl^+)NC3YvhHm1I!{A#ru{Q{vvxNPl*xHycHr#&EO7F1shw*q
z5}lDOY(Dl*JoW`Iq%17``68PB8Pi-RlRVa2>&D0d3#y5Sn`nVz+JGupM(l5?Fx}h$
zZ7gt8N{d)125i2=;f%*i6g^8ZJIA4Sy)&WAa*M;{pt2RJcXA*k4>r=oo-?$TPnlbm
zCeie$;%o2yFb_O%3ny+loQE872h7i|Wca<y(JDHC<&QxtgqCJPs#7M~t+-+t5Ye)>
zGSs0hdTFSaEeC!=urJwjlG9;CGFoUCO=+?#>U$z}FQAlzK7OvxuxBvy4pV_m7O7=9
z0)@qq<zt`wI*<LG=kmMHc>zDX;wskfe-K*qs8X}<OCM*q8Uq1U74M@Kh2*GXYEFNx
zE08SLfVR;K+KqG6kTnI;#aY_Sg6W%{<U5;AJhL5$M;lJNF@b?|Y~F7jAN#<&`P;Kz
zfT@O9En~W`aZP@ZPVT!?!rWGB@YkYZnwBoD)3djy?p~8uYt67ijq25oYbLrB`cPr4
zVPSEJ=RE6oF<Lc9Z<eaJ;!f4MZ^>#QWZ$Z8HmkBqR4C7|Dp6Z73~<I-=kVC4J(KTW
zc`eMWW6>*^n7?3>{zO&dd5{9t)I>}0ND+$Gzj2RB!!D_)@Qk<b<RyRphy2$4?~76B
zxipM0tv)DTk7<J?gAdUHtwLGaxs$nKz`vdKHw<^dDKGy!hJ!x6elJ;EYon_&CW@w;
zj2hY3EZh#Sy{h5&s$me3eGS8!N1QdaK^m`cHNyPihcXHtYb`r({w0)SG}-gj?h(s2
zHL150CURKg1|T?eevU(qKY<d@pl2B^jZ&cz62j5-Ilrx>rfaR@mIsOL9%7aiy8?F0
zfLWWsO{O3nGPRu&p`-X{_Oiy=#w3MTOx$%wqmPm=IUrPg?>+BjW@rf>u#Qnx@zS%O
zPZ;h@0>W5a_N<{NP#8xw8Z)Z4Gt?#goZ%OvUA*mrFY}n^y_{3dcopyb<R#p=<Pc-2
z)S?2$#net~;)GU%F1zO1aZ-`xT4$NwP_2ALWsUc4`f|}#8_Lz55M5S4)S7~tZlfVd
znvDlhd}Z*en7+k(k5NTJoVwLn1t$G*SA}Fd7&8+L;8Nl@L9^ssI%J=Ex@v+@1dYkc
z`S%rVWH9P1tnc&CvtGpZJ8$9V2Re>;@?+p0_rMn}xysO3`O4&#(wdr*(S~aKOciV+
zw~;)+xXtAdz!mg+hPi<u*de}GVR`}V4xEWjZocMJcbs*iP`5+<X>=WQQj&EdTcvyN
zrCm^xkkeW-kPy`v9G0=CLq=iAoRTJOKbz*I;|$dRg+O}0K37%+r;Z{REXu$-td23J
zVpNXt0d`dt=Y8Y?PI}@q`Mp2;8-DQfZLB-saMTuz$5oo@f;9%0Qht3XQ;`*OKAb9r
z7m5Hy75~?2+fde&p+-QObOd$sa?X2iY-m+iM7ujd%$;X18*weu)28(vVxC0L_Sm_w
zjW_@A*K*MMIez=er*QHE@6UJ~wLeu^;hbAxkJXMlv`%%qnn)1e4Q427-_cjxhh)v?
zoGklWt$t7_ER@4BM;yE#zj?xM;H)LodO<O}(S>QpCbZ6rk1jz`Ur>!pjCx#Q38rL>
zpl+6Te)vnA@W|77-}zr-u<;PazQ@o9u!^yQb8+}Ms$yMXogrw2sHYbUeYKGuNz`Ro
z75V~JD>Rhgm4`p<1kQNo?-QzkvjyI_XsV{~7tgF(*Oj%S8a109Lnuq;*Uj>O{^1|F
z;mXT->MLHtsIR!Wf?9W!8;yhzx-`mtuiL&H+shi>?d8w=647C;*mvpIT52a3k_3S>
zKZDNA5r!T!8soQag)LhUW685t^VqHq;y_m;AtFdn#-bc{(#c%<g)g$c8Y9-mkQt2!
zEsa;UthnduN)!CsvZK<Kru?w_zsVC@cJnc1f@w1`+0+&O&T=2~t~X)LV{fL!lW=(L
z6RrWR(qS0J%#SQTzwis({j}eQqYlQq8G6=o<RR-g^^p(YLtnWBidoddxC}J|0%r<v
zhVeMyM;=j+@dn%P^W*Eb^SXC_lsCTpz1-{WNAvW@J%R_`@9xYcu_9VVv3;ZCZc395
z6-m-YEyk`O7MVnLroipGyQ6-OI_;I80-ci9qdA{mCd<&~o>mFw#u?OyXiwa5@Z$kr
z5iqJ$rN?<ium3sLIqd!iQ4ab@(Ia>t*A>eP;d*Rd^O#LzQPttxBmr(Lj8(w*4DOzH
z=ALhO6$}?(u%2qhMp{v54`tA0u1%6CEve34E)vc<1eLt8N%Sda2rlq|r$3b|&wC>W
z^m=UDwMe;nGlOI9$e0Qd7mIG097^lznnYRIwP%v1-3UEvH>cEm<2_mdVkxPjM|=pK
zRA7;8Z^&{0B;!7bCc(MpQ}%=))!`Uo3Bgm2JVoL7<+h!?`=39^2R{BWZoFv`vwkz{
z_d5Xfo-(w97bh0=zL~b0<Yv@fcWulEx{W(+5`3pP&B7I@c-c#qMnAa-WqC{4g><yL
zLL}RkrMdp;gPA}N%a+XtGvBvV!@%K(Y$5oHLQD*?3_6+Tmtlo;G!m<wo|CgST>%f?
z9c^*7afa^kz9iXwsP`BH!6QLgTDXN5ocUMui$vkFZF0QSiK5x(RmjeCol3{5_a^BT
zgwit(OE@>n<yT$B^Z)Yix%9s;XLiFJQp~V0EOAA&DK<!?Nom{<oONlK5C<hjg({3e
z49@jp{sjRMgReXyV`f}I;l{l44gUz$keR^%uQjDP4^q@Asx(Q1&$yRLIc6|2eD_W7
z;L<OAnLl~Qn;6Xttaow04ccK9cc15d4a4p9DpL*rt7#ZSWM>HBi@>AyUHd&OhcvmC
z^c?FBy8{bXT)~WiUDy7C4O_NQhJX{3C=qdiH&C@MFOpj3L2@uLP~80<l-4j%Q0pjz
zr{^NaJ%k7$tNLZ84Bd_U_QU{hO={c3;kcIEg(+lN)$Rc~@qFjAv1J8n8>lzsvumV-
zgj$0_q!|@HZk}@nzWKp-bDzIDo5cV#MsVXXXFThveC)!DFm@Jigeui9K2)*r@Pf5H
zR1G1Fkf8X55tXS}as{Ow@RiGc!B_t4tsJ~@9S^?m2|VtRCv%TGA3)!d_|nNKrw;t9
zE!Jao6Zx!YlPF&MX|Kiv?$DvLG}gUSAIK_*hO`{+Jmf0Nql)Xmw81B|9^lsXJ-@sF
z03ZNKL_t&->M-00!KB8aM%~0jul<y;D<(Rh{PHK4@v(FMmcoxX>Oqg-p)Y(s-Up1e
zQ5PhVWdYF8awZF<IBl0E`bso8F{R<Wh~T6k7(whjqhf$5ipGRBXy}Gm>11KnW<=Em
zX18F^XU8q~I}MWK_ZSpxdCKo%dj;2C_B}RkIFu8gd^%-sLrjwoH8o2*&3K;j+zMTD
z5%!ez-PEni#{n(skp$+W5w;u=uKESLhP&8s%-vY59By_dCM(mDJ47cZ&|QAMWNroZ
zXE!82dGU-wVC)UXIj*>AD{p!0oB81RpP?!(bMqUp8#hn_5=K}{qfLn1q0nY^WE!nn
zhhf!ZIUS_Dg8XThY@1Q0iG6e3S>uG&3CGM41nt^+YBw6UZ6fNkMO*4(_Ffe*t^%L=
z{8#v+-+nX|@Xhc25S*jZib{#}7i02AZR+4@O=pX)I)m8lfN7<6Vy#h=Q-7Movl9`l
zEeKTrJkAOm*9|!B36G8$JQ`_c#$=FAM`x$4MQwgdvNK(MWFB}%gbIqH&&~yS<*WXI
z_kH4gyv#GRZUf`t7~>+^8NA0DgZCQgjmbt&s9mZ;W8H2|f%k^8j0PP=VZnQXDY4Gt
z$Gdp{+up{&!CWz*s(ci8r);SD8I(*tQyV^$@)!b&51t-9e*US?^Kb9}FfV)Sn<>|C
zgiSNJ46*O*4w)uLk$p_0+voK~5!rcLkHlABK&oe|>fi6X_G{g6v<Y~U#D(M+*K*aD
zzr>ss%DDlDJo;3;bI}7i2{4<97)+!^O5>!Eo0U{Xa686a`t09j!_Tgw3W7)IJCk`E
z+bED+A#hpuVnf}{v(oywXVuPR=E3gYXBp7FM?i>9K{a#&4%JSBkmx+xrb&~dJ=$f%
z)|}ls7)(=MhLI>*^xiY<&vBI%p7xfvF}(9p^ydeND3w}X`Gybk;m>@7YS4>3)(|V)
zIzcwN{3QSW<jg^#j7Ak91d5);8jJCYRwJBO4n1%^r#$$+obp@u<1R;UrXU)q8x_>p
zRCEMFh~aa!=-HU)=+cPJCLx)Qur<fPvR-PM(olC-DNLU(X`V8bQgvei7=7Stk(`Q&
z-z`zQ>V)(<d`xmgK)Xm@eQ#A1@$o=30$lxb7OuXA{$Yn>4mlX_iYO&C(iBLOg;2fT
zbDQ-bZPIq_4wR>rn<Qop$zoX(RiOFYDaO}pZ2aoxEaqOtX<Y|-(ALMkz5yY9Thmdu
znr0I3QCuj&SSU?=_6CCY`5KtGKUs>qtTO3dOW~K_qky5R0{tD=^NHs?mxb^Cn10cx
zIBYYge()0v_nTqROmL&1STYAHNe!CRS_p~X9}gymrBDa21@#!yV+r{3B|qd1Z+aKs
z`1UvG**Wa|Iz~YfqCs?s@mXqo_MP)Y@VcoD=|MEAEp;k;PDsDk8aqfPf~HA8UbcL4
zTj)-TE?KiClOAxLf4dB=(^jl(l8vxx#f&Rh*tv~k?{Y`BZ{Nyw*KehmUB|FW_R&ga
zMLcD@=<eEdy}Gosx8{ady6T`=Z{;bat86(F2TR)OyavR?a<^|i3p;o4drx{aul}1C
zGgqW}!-OwX+$LLCuJg|IK(KZS3kud!`4J+HN`-&<&?ou3|Mf<0T<{e0bBsoDFz6&&
zXN!rwY7$pu%ma-fcX3LYM9vC6CYM*0B4Q~DMg0&pLq<Cm_}jDogg<`9>CCv4&mAN+
z9AvJrh8SI%QRC8p%o7Ner#E)|;`|Hw(EolrFM91O;fO7aH_ze35rS{%hWn6&Uoj6x
z<Rv0<?ytIGxVNf)Yv1MH^F2xs)Rf5Lm@CgepAFTJB~=bP<?;AU8{<Ki><=+f1Z8N$
zmNc6z)@#kAc=yUGRR!Dr>22Kf-gnUx!?053`US!JxYUh}0;+uQv_g@8yEP2C?UC$`
zTQsfU-#dn5UZB+?@LbE$J%S?lziw=%M1o}(FuPW&t>{u)Pye8vaba2P3H^KDm;1f^
zwfK#FtTT*#$&PWsBcJ%k4Cgj6tfIgcE7%}$nq9;K4gs~<Tv8RImhs{cY=uE_J)!3Y
zl)j8KiPCe(fd_E%eeT6$AAWy+<LE663X;IrV^Yp)$T0g(Gnb~2vKkYhPMnqQKy!I=
z^y;I<m4uPC!%L9blOtVQ4Op=dk}HUsHgCYh)L9cuQ>b?Z?A=nPo{$?89g>uQEu%0H
zmz=f4W~4zONFYD@Oif+TE&?!lJ)|KI)l=utmhdU)(Il>LW6`RfAc1jm__AUsjB6tZ
z>cS>VZYKA`PuJH3nYb-A1tp+f2x1WNndgZ#6rrXx*6u`bI!-ywwrmwEv+R8;6$%44
z{rji*+L?dO9qmT!5N@n?aL-r2igl+y34|v2Qd1+{{+J}<)Y>~L#K~{^d1d7Z;xKC1
zS(beIb6??~{`p_{&gEA!7|by{Q!pMYRcgq@Md7&C4o0ygL)2s<_G}-i@0qEei_r0t
zmuBBsUbIh}a8IA?Xx>NfzBAA5{yVoK&WUBBdX7Q!p?PnwW1O|9H;hAtxPqXj@!VJG
zI-yR%kFtFv6J6(2zKK9UT@i8Eds77Mw0OIJJ4tAyDqv0I{Zq1Q`^{YRiT88dU5<=8
zq57HVXazMJ@U^H~O^FMMmLn#bVT^-fT%Rvr`a@p&s@L<KA6~)y{07t~I<rlPRic6s
zBz-;_l&Jo>LHq=yC!j_AdBMjJJ_6R2RKqc+J^G=%?(fc~PleT@u}_Xd<mCxUk{We`
z;E7bMK%nPg=NG@moBqF-@|Wkl9&_B?812`?^=-23)g~`!-|JW2F#M*7T)H=Z*WHRA
zdT9vZ3gD=Hw_s0qsaVudIOaDWOt|)^IE959Z)9fUhWN+Q2A~<up|go*)1XN6jSebS
zU|hha2c5)^-}6om6v0~8*zd*vw(UW_0?L5M?RXiaGl^PD87a3B<gv`2Ri-;}EaPh}
zAMZ`Ldt{|itWIKJm`W2`6tLi7F!Ee`@x^fYWrX98A<WNE*n+u2`O7n&$;;k!9>x6r
zgdmhY254Dl<Kd7_Wc6{%R8J&>56sT4LuG^pW$ec+EiK{tJzUYpSh#-2i1&T+B0lu*
z7jWqQGd%EqC-CrxJdop$JBsx-1r1SV0Z@`CK5ZqitN@TM6(y%PbU+}PrdwHFWXd#@
zOG10rrDE5Wky_$SAoL7l?{U^dSSC>iqCriX0vpXjr81ZBJY==8Fowzp#2N&LFCuTt
z2ah9q&KnbHTeWM5)CIe#gV7|T3)Y2vUG(Rlpe<W$N{$U$qaw5@jwn^|FdAdkGvfwr
zc1ATK^a@A;by1fgO&&uxdw@3OIq&`&YL_M0m{Oib<FxL@wXG_R_xiH%MbYJ!ui1)O
zE?CuNNUTXW+8PKY%y@$=3@LgG%z4L-P;%{+Kjb&$$xS*pA!ZoF`xiOcRaN0^)H>iS
z21^;nvDh;`g0Xz}r#JBS^FG4ozx)Mm*|rlmGtb<{1E_+JB2oco;<_~<Qp36%mvqvv
zb;8oL)|50aGm+}6bX@V4=d))u=j887%Ng%J*K0^a$pE{CzLe1l>xp<!FwWsam4Xi~
z-pB83t&7?k)3rKngD&&tNt)+XCY{t&JG8P+leF9RU~PxO%I3~3MNn5%N7T+#8GX@@
z5gl>--R{KQjy)1ftEZ8CFp%l4a-EOlgP9bpI39VVu-vp`fmi(fxqR$`3&GB_Zv7_4
zV{q1>-Xoz+UDqIhO^vcoXkq)nn54^UfGaW0H7{IBxHl!nTaMhko>#u?B@9f7#U=W(
zX}Z<?*)&LphD7@Ah5*}!jK2F_-gC}h^A|7sE6j1nFg&PF&$_6upiPs`1K;*pA^l2T
zSL|)GaH|@I^ty0g!?1Q_p)K?i9l;sqwrpYNH9w^=mTlKv!`v}<rfO(1EgQNaCa5$d
zg8JZ6Nj<(3#0Ls16n8oryI}*ma~Fc(LlkK?f~a(w>Fumm*e#6J=?2y`fMyv;A-j9;
z=_rM+$dZ(9cb0|kwIN`cyGR!9mWOGHnnp*N#>Tjygfg(n89x8^*Yk*fd=rb6;+#@w
z#S<QVKi>4d53!Acg=!q_7i~=00E02<y^~JDYru&m2FQ+6)9C0Ki<^ngKIKwLC`UMP
z7;6zb&-FV;eDw4GhYx<{KREcnjojyM$MEoz?#q3D^KNXIvzRo&4k1uj+R0!oyv#VJ
zlIBXPxg0v<L+b>&dYZwNOb^@2Z=xpqOS3eFTW(?dPp)9Y(Z`?%tfT1nqgYvlU^6~{
zn)=miuKC&N`blOtDT(5w;<0L?^H^;gs*0-*W19*`fw6*);-s$UqG!Jnz0y_dVp|$<
zU-v=O%vQA4y^xB1i%d*|RiUT?Tz56!c*EP-aqW$a=KGxbl9!-I9f=GItZ|Wg*c2$;
zqO(R=8nn{1*H}Ys>zS-&({#QL7Hb-;djF_Hz|3-tsNp&5?r0-2Q5a_v_|zcJ9eZcC
z7KTMz(K8j-FOIq6osIyXl;Ff9<{DKB9*k975s?>TquWfS(C_#7<<4Dv@#3%Wjt_r|
zOTPOfycg!?3I^-;W8@>EYmLJiB~(5(jy1)w7Fl;bcAhVj?J9MudNVYr%@U*y2$V?y
z^0e}J`aYc_%(?DanNGhXW1Tk|3(09onHHjHVp=_un3n!kCxe@GCe;}wDrw>^I)R>O
z*L6a>bR?P55v&eOuFd~$wZD}oyK8<QgbrfYHr0{_2Z5e5EbUt2nWsMuSJTe|7%MHU
zOhX5G&cnzRAA<mmq4YzH?K3Rlynp#PuYU8p*tJ+PKer#Fky4FI3X??EqF8GweOxET
zgc#mwLfXsX6%D4oPo_Z}Qhs+V){P-(Nia(ky#XvP@~*eMh4p<!gNvwU?zmQ)5M*dM
ziE3Lv7fL{Pc-+so@tVJU8Gn4n85Aczh{a9w^lW?H8e0!Tbxkgb`(C%{>%!ZumRkeW
z<SDB9(S7$|53X5jaZQj4vulZ~&;K_zNuL`<xYJ`EO*J<cH<Lh9M3{OqnGV#D!^ib-
zH-HIB<%F;pxc+5lbMu$JNMHIC;#0E`(q=$24V*?jy0z@k>2IdpO?yH=u-m^r*)*&u
zoYqwflUkDyYSC);IVLGnniao7mo_XNHx^pwr1bmk^e>gwMhNPoAA2#w6@CYgJNFIr
z?tg#m`W{vdWu;vFpFiWdXa56cemxb@rqL%Od}AWNz47XnI=NMw1H2VphX(f$0>P(3
zxvVe-tdZDoSc8;f`YO!y;J9P%$b;{59QQx*c#b&yAPN+NXPk;$D>02MUJsk#^hw)U
zO^1nX?1DAa!D6ilSaQ9{lr2f>@Vrz#SRW9rxagJV@X7annAd&!0yyFj#yy9cXki+R
zwvIyDS|}QAr=P*Z37KeX1vrhO7ZNvI1DXA-prOK#N(Kt_i{zE=VYY`u384*1Y=Rfb
z1cQ=7jdGZ4B2GZ^{~wgH^7KbTKKgslX1^<Mz?90)b(U?1-+^a*;N1k5XgOv|%3)4D
zX&R+g9Ib$D!sbjvk&69Ggt9`$tB<!V7Armog-vcwby~J`NlAOQlVrmAAjoimZJ+rB
zm%jJixZ#*h_d1E=&U_)G4f|oGNY^UrI!r3;R0Cz@Q>cBxu!PHheiiRH?*n|{%U@yZ
z&YjTTgq>Z-!f**^14a!&ZM4o+0j+S>MnCzqaW`UeHJx0IHps4<SX?&>O>CC54TI!+
z!>apOcPCyJf7o2}tQlABIP6W88*8gRYRBR%3sY!~1BBqCKd<_z?@=Klr*%Sa?V^wj
z2<n-O(d7z;x-pgM1feyEo^Kn_e}#^1*-)w8tAe!<DrR+w?|kjktSgkBDLRF9q+Qi}
zSABjbt&I;@2fi$0+^QMyt?ysPOU^on?_B<K`uiV1sgc&?)HMNSB+Zi#u?rS_%*n8}
z+2;Zwp{k(?8#Ng4tz(#xDAcdQF056SZrR2^zx^M1{6kN~s$#|A)7dCVsbFLhfk|g$
z;`^yWSSacJ>_-0S&;Ekbe)}}GJoeEn9kh;qVIq|>ByqfrNzQZ}r1rgjB}5mI(?sNB
zzb=H~^S~JB?Yk9g3qET$sSq%>fLV)Mzlq_FUGxl8zq}EdI}#tm+{u$5o4q{4r?n1U
z(b~T_itC0w&mj*yiEF=pG3&Izk3A4@g~3-Exr>Q9p||&%m1$F*HNXL@o)J0)W1bNC
zO^S4<OtqUTSMq6{G*jp_4(bGPHCB|JBUm%L$!w}5HLXJ<7uYBcOI23fX?B(`ocs4Y
z;r*XrJmZnVA*SSk_c@At+~GievRxUM6;%kBp2dj9U#r1s<UJ|GX_7>}gRhdvS~Itv
z<eo(;>P;`g)T7D+6=fADoW;5swg=^FmtVuhKe!V2{!g%J{W|V@_q%ZN{qDuRPPhjf
z=Akcu4HzE~7bzOCu(olco{9m%HXis<dSN;f{+7NF8n+qMDIAPkN|Elve#aie>(2jo
zIQS6CqDYfHNdn+hplcp*WnwaS`ne%jA#_6pI&@!Cqa>!8%{GGfB6t=*|2Zyt^V=z%
z;R&yQ6Xj8dp)LuSH8Vk;OnD6%bph?rl)9y2+qfEQU}nHU>-T5C4JaNTD%das<q}?c
zP`JqH7s+TYt)Qf7z&iAyZje{NH(_Q{hl^#MRAW;*DzUbQ?E#A=r4?MiXavS}5xM*H
z?#|+)H62oLfiTx+-D97^!%jI3f6G?v`u(V89L^YQqDqMx6pscNjVoNgk5X{+;*fv4
z;0wIx!=K>4e*9ysn`6+QMc2*7W@l-M*`B5HV64UAqI+1Hs)yj?kjAF_UpHNb^fRLy
zjUnpJY-ZJ@jqWF-;xt1O$&3T0ZG)2*v8(Tv)83{QvrigjGy>~P2iK7rP4afqfW>RA
z5t0^d$SG(VF!(gk$l2|+K4K#EZyB7hCoRFAjk8w&(=z2VWUf?cSRkey+$|1AoPOFV
z^b3PEZE|`$5FiP`Yg8hIc^W|+j5Hv&$F;ZY;7|YR@A=HdUuOO624?m@jNy2Uvk^@u
z>27?e5(+8Ul(uZKaYcs)KE!H>=xCel0Y2_pCavSb3O<amYLH>c3;*O9ochp{sLGO_
zE2t`ub2dFYwRfI&&|{O{+nY307!E0}yO!7g*`M(EM?Z-zr#y<qjWhHM)10BkL=k=P
z+N4d}+{Sy~>-K()Q_cSCK^S7)8$Jm<Zr{CF<0jQGjbe{(6qbH?8C!qwT?z%W`2Y@h
z@NZFxU{%o+MK3jxxUsU-jGLG^)X7>6fu&F~bJg{H<+;yh(~c26jo$JxJxU^fkSx$k
zmRr!>*8T-Gl|d)G*7pWsP*y@mRufdxK^U5J<E)}nod^NbT=bk_^(!4}^~oDRQvO8)
zZ#RXi!6&itjg@l3U!KkWPkA)ex><TyhT|bWyKa$F|M0Ic^P8y*jJ1m1+%_UlHpI<W
z5WMjO@eR6RLK_$q>LM|Nex(tg8^riH9SVVR6nXna&telyAE+>Xj0~4B1dcxB5FY-3
zd-CuH-;cW<dj$P7)zaY8bv2FX+oYs{4sBbs382b@#so8E(5Dl$Fh#u}QX;0HS}Ku&
zi=yV5TbuQF;uAH;#$pGtk>{SKVl}^48Zk62`a#RXmgYLvE-nUBjgaeZ;G1Wj#ZlI=
zLzJH%d1oGc&P!1vNRW;V>M~qRWx0MQO)%^B#ngW;{d^Tg$3u2}{8N1M_5Z*|S8(lU
ziTj@UA`X4hQz%U|H0Ut=YE)&SW^e6XX_A!d4#P(apm+^K<~^?lNkYqNaY$b(E_>Je
z_{xQ!=H4egh<pFRv+?_H!WKo+cSw_@tlG6sT$!Pig$jl!+6)qDc80`45|A<=MbxbL
zP^G=VqT<*w_I&-Dm-5~Ze3Xm-<6A6_1I5gGLch;gV?35tB}E~aN^s6L6tfVLI~8C>
z(!^G?>1rCZr01wsp_s~J#Zx4PwgO&EbjDH>@!+(_#^os;gaIa<#g^|elL(k}_QI}7
zLCCpEAyM^+2($}n9X5F~P&N0e7VJxxe$?z;nMS#xsKxK#L=p8=K(O_cJq}Dv(DXCg
z<nH9z6Wf|8z0t~#iSG4TJb5}y_GsFHRir~bvo@l$&Kh=XznM!e{0K)Mb|5pxM&~jF
zk;v?isp^S38bb4(t7-{~85R`Y@P@bWzu)p67K#lN>-$toinB#*K%-d_A^65XBQ3$y
zr<qKbBm}9FxkI3i{}i$Kpcq^lD#r68gb}vqVOa672j83X-ta#ekCvG0&osIOX?8qm
zjU19TNE5rEZVLUj1&SMP<n1qfKKFaj$=u@^|AWN?=jji6P11aKZkNtpNjiporQ)yZ
zb-pp?@wes;ZY{!qh<r{}_caV_-PSVCxoBfqF>}}!hClc()>+H0UtSO63LW$^XPD$(
zBdu^#(QbH(2tIL%ZHT_}9JoI-2ONm*`UTzxd{xowJG@uKW^|irq=8JxbXzqJavNSy
zX{OYx{Glmcgwt@0hLgSQfLQ@E$_Jv^yG{!Bn!-HYAFW+|tVMhUWyJwAJ-+mg*YmVT
zo=nvn;QGR#aNOz0{W$gH`*Y!c{D2+7Mai^DaJEK^ycjtUVo{auz4#V|63yWBIt{o<
zz}4i{Ef{04Vv)WFA1D`>*fk!a0?rxwMS&{@6#W4|xoHPK`S9m>>jyr?AqVZp-H$n%
zhdty0oOGYNbHMr{iX=@+O-7q3S$)W9&@49~1&wJe3{{D?u-aN=HutA)3@vz#VoPRR
zV-acskb=}i1u07y?t-NuvT1+l$4m=L50MayJqbykkTru7nFSpRMSUNWhvD$X>)3kp
z4IH-rW`26z4VYu^N+>J5F&JZ-VyUTvLreY0{l{yOc<!Q<e;%KuNUQ6y-{T(5Z}>%a
z-FOT4I`Kq`d*6q#lQ?a!2N9V<y*n_DRfjH1UQZ5G2r90jp@?KiW~z352=x22eB;G0
z;`<-@R}R_lK)&|XH!xqU=h!p;I5i8V5jS?4g}h;1aj=leXe$^T>eFp;U@D~wV{A~$
zN^vtYRJO-0yB7KCCEwy-KYbAwe*GJ4-L{>X!5sZ!J+^0J-a#noi;GFo&c$ZRSt>N@
z#xRJD^a(|pWbecYOIDK6y0;d%OC^BEqRfuBGZ+mi9RxREr_4bwStQq}1A+!qM^iV0
zgpP1vwZwFpHkr>Tm6Gr4DG8g{zzTv14jYCPVF|EQMyX6Mt{s^|Nn%P?hO{V2?Y<=n
zm4=9Jsl#~(_$2z4%$+V81*q7hcc^J8?a$c#d9(K?(n=QM+U<lgs#F>lDR}R(##1fr
z;!!94CdVAI87mcFTNl8P&h=z@9#UF%bRR3Rc7}id>VNX0m;W6%{&Fj`>kpt98)O*h
zSp(uRR`69^_pz}|E*j1KOzm6dM^U4QYHBfZbkwu9h+T#TOfR6-2$d31!yOOb#2f$q
ze=;sd%=Tv@I_=`RRh7<})o2#fC!JTiIy*{=Z6$Ac@mZX3ulsS2XFivOjdRQlTx^JQ
zg8<D3S7yZ=sC}>7jgIuWUw;j!daSBmw(oZA4LlQz7OKE?U-}}mJ8oj^JsTeIFzlg+
zX2nZe=rs+4<dG?<;L(5v#fhPmfL$8#-+%fi$ajBCPXkn*qVMoN#D=1#%1jyk>2zdz
z?+g80!*ClQ3=J9}D<38@>ANu@u&xn?)xKj>faS{0Wf=j>KdI`3gGq*9Na6xZDA8V@
zo2;_ssi$+7GoDA7EpXx(jZ1cv1*bm!Pubp|XCd)FRTK%D1csmztEo`u4g{g=%#r!L
zPuLT7qO~+o`U+9SIU56fQg(!|JVv4hqHqPMr?3`ZdHiTZVU(U3bNJ>1IN@%0<H0AL
zz`gEq7dCC2!%0BYH>3oUc(IW_l!a(zPD831g>kj(Qi!~KU-M-flS;*EO?~kV$HEBc
zO*ithFJH)LJmxNsJq<tGLyLmoH8myLrJ~iD2{l)DVk(}eO_-W*uyc{0ocC6S*IdKq
z6Yj-<k2!^MZUZhQYj@mXYV@E^L6wB=iR@-`mTvDW#Pn|QAUz1+O>B}vvVAo5xU^1N
zP2aO^yt3j<b3552&dq%PO=g{1U!zi#-qs->IQgL*x#U^gwv%FApKV7S$*CXtIDVZ=
zLTFu{UMyXduf8rJcp3v@Nqqlk9Tj_=s=`-QELL0rh2h#;Zs9Xu`VT(vp$quVfBzRl
z4M=~E-s}v<*+{+AXgO(W?M6uYp2mtbK1bMy26~OoFrsy+p*D!FG@^e}bLFv>0aLNx
zsN$G~TR8Qu$IzRZ<r`Q1g0C!&*-`YVOr&m*9BOi%aVNVZSGng<?1Zb7(i&gr#FV0@
zCu)=$U_#6bsY(x~pf?r{kp+(1FvG|Pez<D~zcd4ujEmE2k!)OITJ25U9CUtQ(o>~S
zIzyzyC=OK%TH>sw3W=N~0Ws0$(<9NXPZJk^LgAdFVyfKQ6KHKAaJk^SK^)SY-Nc!@
zDaV9eH*x-l-py|we`gA75p1J%l3FJSX&PTKu7tu_uDkXI&VKo8_}Eu2#m&zrC&Y;M
z*9fOillRG~RJ@;Um)|}w-5e0Xr|V|%0fJhbEij^39Rf?ZqF}tZol7qKGzZQV%+Jn5
zI$w<-q8Mz0Bs567C<cR14N0}VWN_0?-uvP+Ir^BRxc?bvu(baSgSmc0BaoF*{(Z4*
zzmA6CE+TUIuNz^wJcRInfD`xKlC|$%n(=_dS=Qg-2o}F{BUT*SuD^zZwrruYHTMww
zGLC+`vx#Wts3Mt&)Z%(GjH?Q928<>6ki=8diuEZrM7Q-9zb(XStGXlUB>97Umu-@Q
zWu-Tr{z<29aLZ&A(rH!NKA~+EG(_J_v2l{?tBQ3fKlt#lu##5*03ZNKL_t&sx!dEP
zjz8i+s#z$kV_x7nzxyQ4eecJyZa&gdoFv*vNIC(B4d6oEAOwQZ#y`70eMO}?a2hv_
zbUnO8hl`?!lQQ)oLg5P58l1DKXp7UYvQ)-e5dgPt77#|>^W&{0mtFKtK6>#bNbnrC
zc{3;c#xb09@8kLZ*?aGBOOEo+|MOOL_c=FDkVZL@Bytd#<iu~iYwU$x*4PGH;DGtr
zM4M<9FkUcT8;rqVgH137v)EotFv-FsBSaE8NMw|y(MY3-H=J|2tKR*is=KRCxN}F0
z06)*Po_XeK?!A4wyQ-_+@D2C8>v^28bB^4EMaQ6_S|&a~KWZhG2!#obrHMKZwR~eV
z;WClr-Q*mBKcO1P{e$5e>P9T|=NKK@Pd0xNAXsa$$tY_=&DF-7wcB-TRa|^y0)_#W
z=DGb-o?gdJRO-oyo(WM=N8+>fCUhn;BS8|^);MxW9FMgTHPEqQv(^DWfy3f_-A(b}
zV&zUO)<mMX4`Eph`=bFD3C|{}5<~1xdW@%b<AA!7o-8*~Lk!e0SUteTP+2N6)H*f}
z+zfIU{0eHighok->-D+3)}@X}+>BS`wF1KPtXzgcDDoW1bG~==jr{WmKE}KL^#gqC
z%I{EmVcu?|KYt>H^>`bei!UL!QE5`!v2=A6xshNE9whjnGLs?UdD3d;*U7fjR8JGS
zcUB#_2-bU6hO1n7{$03lKF|74Ip@yjaoHEXz^W86l{f8rpk2r24sO0%@}9berp1#n
z4IWJ73t^(@fC#0!ppHgG;YZweN5QY$?)I>>!W+NvMJ`@lV~~ZuR68pI-L<<ej{j{C
zuVf}iP+V1okU}g{7^-1_G?GR6)h6MRW^Rel?P8B%h7#IQ%PG4R`^XVH7O9zNVq>l%
z&H7m<&G)|Row>)|&I|WhHxBEfOar4djLHp6-e>=y;`MKN2d{el>o~k2ENnXoKXf2f
zFc8$VyV#1In1%-=iL#+@C7V?f`P5jXFg7Y%DHek@fN_}U+tAMn)(;=x-T(MzPTV%f
zLY`BV!O1yEkCLz2Zm>3-7>jHS=pWqATb}+@?r^)ibN?qiiQ!U67HXc%ROzeEEIxkx
z0FO(*-(=yKBn<KRx8n)Jt*E%iGO-t7;k1)EaCy$0w;a0RMt0q01an1$8IIO%jT3FV
zgSUUHw(U=ABIz+JP!nJT;uV`^c;`AV?2$wdzyI4khSh{5sGGJ39m&p3Qpvordu$Yw
zjw1Py%{C?J7Ozx;0Z#26UBf)G(sQaceC{=W$$ek+Jcb2<OsYd3^$Y(W|9sKK+;CV~
zSEXu9fh-LMT@4l?e7A)vq=+mX>$@}ovuLXgA0$tryJw=zRjY0o3r@+iUK2=puVHRU
zsI-})2ns}1$yNJSx%T4E@-H9080#x`p0LE7Z+9*axc6PS_g&8CoU=}4ey+gP0fTz*
zE;uOHBxM)pPSQE4tGqU<fHZE#MnNr=2n(m5&AE@*8J?5mlnyF4!WauSkYzEJ#yMKM
zYFPC%luqQ*oog-z6y7^*ZK$Uq`7sgVAu8a48Cq6kR2w5?u0UN3R86<XRqwNn-^v7D
zUGNeJ(1G|Ed<Im~>Zl^dQ<hPe-9TQe^CTgmNl%<OU`v<n+SCtBtTHMBOMT|fza4wO
zbtPsYWAA}gHkQr+TSS-hx`R*`ttVBeM1q=oSyqj@mKcMvmZ}m)9&Xq-<ny1sjE{Z%
z1AOxHmvYnITPS6Ythd0z!cHVJR2$*5tQzn!ku@foCQ4hCuVGbTkxV)`V@_t;wpQAF
zzwNMuYp+H012-&FSW-i^{}6@d^ydrg2-H^iI^j0S^Qb$J>0qNBG*By(nJy<uz${%;
z%oOJqo@Fg+QFC!T*bz&;MOLz$p)=@kh*07sYvhAl8P0@HLMpP=6?JgVky)Zzj>q`{
zy6VX+6!UYeZH&lm7T#gCiP<PkHGUihCCT)g^y|^1SfRJLAx~C_8aSUs&lM3*<wrdJ
zw;qEPkBJ73+BmPrHB~)|qbxH%@!8Mwj2FCuZ(n@_%O|doi=}dkwUI~YTglee89Fsz
zGp&rZw6Y66F=$I2!RrYn@%d^j1SlnHD>l|Pc<~=RgS(%17V~-7HN=_#&(ua77lVoS
z9N|zJ_&5f^9bBb%a6j*Q#?v|fj(6gMCq0$*opbb;dN?2H*0Lp~ljFw^?)bN3agt+_
zFo?)Ss``TCV{#OQb~LtfgR%<?3*_5Rq}sm^TMel8-h?^rRGcx`1P-IpdNZ^xVKp%`
zZZg(D#2iY|3mcVJT;(WwdD4>awz#J*rN^Mra4eDq-NJsMN46_!*RZ5;mLdRVX}C<?
ziZ<Pml7z15YDf!%iKJ+Q1|u?~eD}le;euayIQBmGp`5SCvmWKZ^GDBn8o&1B7m_V3
z1=vZ%Sxc<Kn^-mV;)9V{qk*0z2huVvJc;yS@^zENsy4x7_<bL%M#%<F(7LCt7FHhB
zA*xEH4pgy4hPQ&9TSTe2c>w?ZweRq;FJA^51GX<LaQdkybC)~biTmI8ZrttsJF<Oc
zfgWIR<R*|&MSXZjj3gA%Za~{aUNHu*G(+-j!O#nX5v7PqBI+ti7r7SJM5)1UHknX7
zJ9D7rF$B3=uO!m7k$Vw|eCl8{@)QR2vaj;#k6p|I9(^H0mSL_J>dtlL)a%sAR(l};
zEuLNyee`G~)%b)s&uCcEM=3@Qw1oM2N(Z{oC!-~s*Rrx!YXW`y`98^3M6{8ftsQ2E
zTJflt{3-8#<*WG4mDh0RpSwQ~dh#<NTcj4xZJ8myJ-39;hlo%*hcOu@FK|BeL-y=F
z$d@kvCZGAt7y0Dpzrq#Q-o*N#q}MYTlaZ~Q7Guni957UhEQ>C@iXjgQM=CMLZ2bV5
znrSK)QI99uc-(|(U>bf&pmHV9B#5~3DBo&?y_xWteS6rp?If~ljw=qVa>#i!4{Na+
z+l8|waHNJ@3!Msj2OOGoMt##AtEGwCq$ily33(NQJ8wc~&Q~6cWy2JFb^khFIoY#!
z?|!~F@C@=S<_aD&rP?8FGrYUbMm~>GMIG2os3<vUsmHJX;-mQLSH8$6KKljcWS-JF
zyb($TOabc2k_vgv+tIYhBy!Y6E8*5SG>pMmYx&>?wHP)FtDf!K7WkQmJP=Wb`Wz>a
zHAZ!wq4RihzI)|v{@{=Qf)8EtDf0R4EN?%7a-$^66z4~%wWLzgn(XPR46{lLBW%g&
zsXM)CiVi~TUSlF3$Hy)sWO;>A%On5C!?^I5AAwXua2B;e#mhGyBjFwQ>SCf4j38_7
z!RSCq?~vy$Pkt8X-QkX0@YH9rx^0pEQWg`9M9iU+BOFf_eyAa*$VJC2bUhY__x>Be
zdB=xlt5tHXrMrX3yFz)@)!g`{FH(pnKXDg3ANb=`Hlr2LjM9_1&ZbqL)A`Ujuu+}I
z4@<uHq8D@ULmy?X9AP{Z{Q~Ez2$z_h_-iW7C`Y>h&qz8F3CXQ;7<7|=&|7;&jmo5s
zt!vUiLpm^`rt;iPu_mFlV;lxaP0nh3c~rsi_2w8hY4yOo<(4x}=cix!I)*3Akr(qg
zudI9Fg@5)|-uLM*G7^hq7L9NV(EyLg#ZcmK#wX^$H9SVf?$Wn9+LAs+)0%rpw=Yq5
zGg9JoJ%=Hvpfvn}fHz@OIVvp~xe-N?V@!tF42xpQAXiywiB=UxM_6$zEi7{WIj3{i
z+nvQ-&btHWo_z*qo_Y#Pi$yrGcOjUMNs>V@SzSq%)CQ`7^>`8^H>p*T#><w$AEPE>
zx4M&Q1-j+}wA`1vGhx!iSVJP_6!H=dz5nfe<!|1|Hc|EzeSYRG|G;qD64nGLSvwcU
zGO{$`l(GCs>j1gF*T7}+r6(VbSpC2|x%f4Ii&<Nz+_}KRU+^-@3+{%^^5)%xpvE$u
z5Y<FglVbgA`PEp(6f01TAj^Oec*Cd)T^q5s;ZX!$eF$bM=bJI|-ZHElH}2iXm%sWA
zF8#t6`O?=f=Zb5tXMI#66y(JMniV*$$V_lEca;X$bLX)Ut{D=g3U3tg6}_xNRVZT!
zqnc3}8nWg@(V$VEo3v9FT2Gb+pdqdCo$&!)XQ~ZOT|LB4>^K89GQPZa1K(Wfv2HC$
zkr>+11m!j1zEwEV)(2)v^}89P##pbO11+4Z>E1ZOB2-cFswQqLo`v;_ldWT2GY%Nf
zpa&{><Lupm2#xIzNvaxpRK5)oRZA8ZbH4KL@1$B=WyiuIKlQN3@QrJ3rm}s+C{6@5
zId~Ve*rjXWHTyHC?E%MvGCS74(Xmdm1YZ*4J=OYtp814d;ki$FJg#)uJa2#kLtj!k
zWmJ0p>J4w_rLTHD&dxEv+-F#V&4|Ig4c$Q(MJ%G?mP$=%?n0w9No0}<kU&nQfk~R$
zTup6|UzG~gl1fJ`_7>UJ7e4u+chZ*uh3N&U*XA7K{6)b<q=JikLv?safA1Rq^t5Ml
z_Bm&8|EK;gYr7WcE#v{(7dzB8i-G}iJW2QwIlgI(IsZqd!w`>)jwcLTEoLaGa-CrC
zwxl@aMEp1UWP?@8{rg~jh%Ontm~LgP>;2rq2x>gE(o`huTtDc^8^Jq=_dba}OnWVp
z8R?JC<GOWy=1ou9U5x2SaZeg4F#+LXYF9$K&#$A0pHzl5b<PYVSXDvShg9GDF4zC-
z2RZeTkD$zgaz^1Qe(!f4$-iCl?<_2x#A>BD3@RR{%ACzmUk10&SdB8V^=}ebLz^6E
z^B=Tiy2c5#)9M~`2M}p=`hyxy>_$XbQ&52k73DlL6j_f2GskFsK(#(VeF$uAZV+ps
z2mz)@fvVv^37`A+4SeRS-v#+Fl`C0U=(B6v3U@l^T<&__9XbC_x8uw+PvfLrOZ04$
zM+EMJS43RUOgF)4yAykRZKP&H!W{!JEXD>LaDA@2#NWsn#sJec>R~l6CixKGecmFD
z>n^#3bITEVIi%1b{^~2y9d}DaWyi|<X|;LGgDun+67`~HWK67r0ypmG<A452&RTQK
zsb$~b7A}76tNF=$zdGpoH1p9k!I>wDsc6&!@`_6xfK6MCZ+t7V0`H)#AjdbZ+TLrU
z3?yQ48f1h=YPoLrKEC?(?{MiCzrv+o_%heuw43$yK`dZcvTTv8XYhC`QG7YXWI|a*
zo%t*nS=mTT_To_uUD(1`^veO;4~@ujIG9^jZJ(iOWpf*`2uTggCi_U5RJ^5;Rn-Qo
z-n{%K$Q`Q{_AD*%wp;dKZI5bkjt!HA`4gXW&EaSir&6rGq?P7`2UFXdCs0WrllG=6
zExubm(a>3t6pwB+aF9e878%!wji#U$qY?@j<AnI}^)@l3Z1;|ulb~d|;m&8D$vog~
zfa{!k^2vPpirwUe4N5zNgxgB!lGbabrezozFL2P|R<sH4QSKdEsF_Oi{}5Am_Zor%
z%ZnM0`^^j47>!sg`pxf@g8`VFi$DKa{^xW5h;Lqhkp8w4!5JzAn+M}HG4-nJapHn>
zsxe_awOhoHX8lba+chD21ob43j|y0AF-Hk1h(-lpRcu>Y##S4A;63jqa|5!h7Xx}z
zTl2&<(X^~Ch0NFhfyKaJK>z05{N0m(kGtOSUflO7Pi1}kJiUcHDLh0dW&%e%-dOmN
zI4(L?Cpu;cgNXdAsy^-b*laCg^-Y_;c5t?6-ecvI(<!g`I<_iVyKy%Qi|62y%_}8}
zHA#cEElp!gly(cSh^#*sFy`87h^NlB+9}v*Qbmmc?GO2dpMWAAgX__z(=k>B(Fp;;
zZm`_>s+Gr7o!UZb+PewzIU!w3t92q04A4f7UBw)qdBbb?g$F+v>`rvKPrnd$E$2Mr
zw=U#muYN1}j-64DKF8+5Xk(2m58B^OyROS5`ueR3jYb#W$(DbO;*jyD(9z*+oy|gp
zO`=aoz~mZiQ2Rij3cuI01q+LF0rTub(ot67XPi1J^>GukAwe#dv6;oHV-3$0H?48a
z-pjb?lFw0<BfKlwwmip4C!EM#Zg&oMy~8=&>Go%G+L@=ZW5){f3%xMMOAqSE;)ZQv
zcHtGU7HdQB?=`v#BMrH_o-0}h>WL{_>tH)3K^gNx17u`il;u-TX3v*D$AWR}+fYmq
zl+oK9OlEZNlW~+pVXJ9^1?Yf@vWvD(Kt_(<>IS`GtK!qo446oMt`mihnOJbAu9MnO
z2@i6lO`trs7{r(sV|r9o8S@W8#j-J~xN+|-eD8{D`T94m;tOB9oNrxuB{%Nb%gB2q
z?;&Ebd7r!&WIeqqm6u4|)j+tHD92m-IyB~Tf$;7%R6wmY^hS<zHXM&Q`A+1S@{Q33
zA3m^`{r#ZI<KtQ>9oK7_y5<PuZpW#FbxhGjP&H@<l@~@fgO!Db-&4Qunvh7-4wx4|
zP9bUFC?>+xq%I`aMve8mE2cS<q{XA8lHrqLk}6;`yn?r#r9Sa1(+ScqHSCG5P)8#R
zPMhp_;LxE%VSFXHxqkQRtLXLnF_BMrc6IQZcnQqvDWoh8MW<#1(^T&CsCF1hq?2px
zeo&;dfE$hYh5z<5>{wU;S2e#q+8A)?@Q_D8<~R7#RacVFEi<>YgHjDyZeuRzz=v+6
z$AdAAQk&9k0C!eoNjf>rzxzOZ*3hjG1Y<4EyZAx|CuLF3$ZfD|+YS!wxsuOZ@?p01
z9dlVPQEc;x=S$03(C9Ml0gJ&MSSP#w8s7S>r*rqaUBG>Q=XY5@VV+{Chzp~3=8})y
zW|7T7EXUgC>G&}_Hvdsf7(VR1zZy8}_z)d=k!s*A5*$JWkJyZ*(@$g16<?>1;oyxo
zuyD><k+@_VB_zSnm0asqUYY>JKsG<$(EFTsWMoKGY?)|<Cv8s=IhHZ{mX_@!eon`*
zRKL&pI`-K*dr8xYkC!RU>>1aZkZC)#hgy?80CYecMdqNYIMpa${HvF9?-%_sWm#d1
z0+o_qe&kQ_-itoMw+{vgNL4wEHS~%er61yA71;O|KI9{*xo-QnMZyr6C5)J6^EcGw
zXPO|TMk{F3u2-`9w?*(+)L5J$YOS;|7HtwIkC9+TXsGel!Fxf~1ayl={|a^Z%D32E
zo8!cgE%nF=B*cMteD9VuuDs=IyzdiVzzU39$-?3s+gDcDvAmtLPdkOPPd}C0opBnw
zb}W)xW%t$BbEiAqj`PnyyDr&cqoXRmqOo>0mX<zL<LDATk)~WCX~tQNpqk+YDhkfH
z@bTDlZjU>-!FlK34bHi1-MpYilWfZA)V3=It$S=E>r|UeDfrA{Pu#}(juY6k=X$oy
zXWX=5xbr>k9#maylwVFdPh!G+sS_pRBX6aiBN2Ayqq-MSjm3#_XnlhlZn}kQuD+2y
z`wnv1<yY|KuV2BH*IduOwf$k9Nk*3CWHy6h1>4L?hm)w_<~<UfUyU&ky+_ovrget+
zY<);hy^o~x%uxCf)*a@ab2;~Mhw*C2SLV6oU`1sEkIP_i>PdK}Q-u;_hQpU6r2Doz
zWAQ$yVKsc|$SVj&b0$az#|}{AkI-~>eH(*oEF>8lrlB!S+sddXvj*oYjEPdPb%Gro
zs5K#$)A4~BTd$0147Ef<r?S1n?woF)hjtceiLhEV!G}U!C?H(+?W=jotN)%KzyAfi
z<6Zy8*RR=4)?W-`ET*~d#vPh*hw9Cgp>(Dtjvj45Iws51sE~<4q=$13aNw1~SiBdW
z`sBysMjI5x93H5uia}QJlfUv9cHeY}eC1>|O2AlpCMa*!El%&l957-rQO&J!y4Qv&
z*6rkU#?CkUM=WbZ4vwW%FgC;3;G+|Y0wpW+3mn*c1Mm93Z)Vq0hNTw?)9L(+Bx>(N
zF~Ym>{?&$Ut0mcu`+4otp2>ahb3g9=#NTCg`x5huSy(3{7cUhwXk^Hv)U0l$7<l}+
zt&Xcj<ikI52}3;I2R!Nc7;Qa5R`qcK51T-?eF?j?f?40d?cWCn_8}*1??|}S>Md!Y
zTd(C#HK@$QKsB5111=g>wWPlks;}d<wWew=9Zcmh?-<;gq~RD{%+a1g#wuPD75F3k
zW*a<rZQfLTG&WM4HAv-I7%A6$@?!3B+2wH0^Qp3oewMQ~9Pvlb|6P9J3D2Roa0(Tc
z3P<G~DmLnX`?gDzIMk9lJ=a`?#_cyG7EPeg(LWytZ|TH%)#OV@8{9FW`EQzM7b;sV
ztgeBG#ZH=s;1=v#IEWbVRq%pvRf)->V{g_6Bbcm@v4-4xDkI!<XvpsUH}Z{ZcSHFp
zd^y6G1ByQEzivNgoOCjGyxrNHbm9)qJ^NJ7J?9KgJMCmnIdKO&Ru-7+6+u2Txm#be
zKfI5MPI2Rnj=<IgSWPgivmxoSf;;VGPW`1v1?+xQF}Jj8Ujd@#+kmKs6#CeG@R~>=
zw{R#iCO{p;gszec)k}6Q@vzsug3JHuAE?$=x!_(GaMHsb#xRGTkDnFihpE-56e<rQ
z7AhB=c!g2rSv|bL{)7A3d-Fc7+r5YHefJ8!eZ>`g=Zb6Cckm#GheOuZOH9_o+CIhH
zGGgbLTRH(-WQ@vT;2l&I-g}G*9Z-{K7}YHsRmHoig;zF;KSAprf*W~leD5T{pFD<<
z81}3k=FpClIj}n5(*3(xFD%Y_a26kQd|L&oP8V^!%&VOWX$n%*C}Um2OX_zmT2hC5
z$P&3+X|?Rr1-qfjYyGa)62sA-M+4j~FBBR27EY?;-u7Lm<Syr|HTG<ZEa@82m}w=D
z7)^gOF-@F;bFaHER)Qg2Z*GZKyzU?PlUKhU$#WLA@1QIlMV{eD0kUE|jZ&4zC&ioD
zq&l4<S|LeHJznt9BdyDtuH|iK$7#SLn#{&E%(1?HfD3=|;heQ=C&R%S8BjTg*o?pW
zo4@C(JqIaz3#^Yq(a$787pjuTBgbcF>g14AA2$`ug$W3(#KbO|z8b-~kX)*7P4_Zu
z@LEx11-q}lg7?1d@445VZ-*7gjo`GxMs=sQ%n{66ym#0dhU5fU_wcQ&dE;}R&jTO$
z5bpNtzrn`NMds(T#AhK95sTeFQ^z$T$CH8|{_(zJi;X#!F|g`GRrOzwkI`mNTw7D8
zbOCgV$qe|C(YL<CO<(;ga{zxpfWLD&y;IL%@m}|)G+6^V5TP|y=?GQV;aO1M@ovO{
ze|bOO{Ii#_V=!Xml%gnth*D!9RWED}KBBu7sqb_fZY>1pNTlb@l7(rcx9>lV2?`LB
z;>L9|<2ISN!X}K?*{<F?mWYMIpl`I`!^RQ9N0T{>tB{drt(SAt8E5gZ*S(eDQbyhn
z6`c*A@!EfQKY#c3f5#Qe47H5xXV;8><EmV%aW&>`iqU{m_39giB#EW|Nupz_)RO3d
zM~l(=JR-G+fsF6Q9XNg`gh@MN)A}`uH$jtFYwS$aPzmotwb@scWf}0|Hh5SVl>j-)
zEV%(PLpchLzE&N^!Rq=trFt@xp;hd|dB@URkCo+lcI{Z<<P)}W^3ENcv}-3jcC2vn
zi96Z0vc$@^B^KxAnd|q+^Ng%86!BIpF5ZuA?rVsn0`mA18K$Wynkn;t%<HgyCXrBG
zlVcMG1LC2|ph78Bu|r`{4PZXUW^>Kwk34J)N;U=qRu2t0yf$FZp8f2;=@zcvy_Xwz
z-@r|K_i*#U{Tw*F%7KH2SRV{1M-?{95pS657l6g~a!SDVEWRp3wbX~or?<ff4Wqcy
zQ{)-qZA3~(^0;beqBKd9NmL^1^&-mO8d?c@HQ?0XeZ`V>?6~<5XO<Z*%h)|{xUrW}
z+0f;1s%-*07Np7)P)i4Z*qC~yBxvn)C{<ffPeW*?V?EJ)tqB}VqbY=_0q0N)ne*hn
zq8M(F4>!nrIs0r6-|K}nLrhR8vZmn<H5F@G{|HS#7@d$W?XJDk^o3fmlT54GYxL1T
z@y?K`#oG+c9qWTt1R={Zv<ilap}AxbD>xGb3`}hrIG(gkTj!HvhM|=*t`)4(``fI;
zFtPu}ggc17nP9AGg-j|69r2O(zKzpXdh|1ksN%~Jt7^FCgMW_0IuBWabDpf<rz)$U
zEGB^OUfY<yQD?2}c2wGAG`dwbb59agB}MRRF-9mdaKjBUV;Bt&^0F5{hhKg8&*I7v
z{k#b8j?gF{wR?P;7=~TJ4okTF+r0J}&*Ue6`sX?CvA@P}`vUn&7CXRWNshL~q2pZV
zAE9tl<mZl+LeeqsFo?)S-ur8TGma0{kva>h9(Pz1ybp_VampF^Z+#Ct8Zp?jm&J_{
zW+98}9Eq|=;=(FSN}#HM%r8^M%9N@mE~C<J&Qqk%>YwFDjl;l1B*nC)_Q(m0&itUQ
zVk;Aq;OUu72P@sya~f3&X+`ydF`D8CGNS!T$gjAPYyZ!iIpfzaq$(iG3alUTgvb0m
z7hilSR~{I$;XGL`IPH{8TMvNxv(zUBWaE=$RvDqI+uzv?5_J<FtB;a8wT^G~<4RjP
zF1MRjt6GDmHg_>ClnR4-hc$T!II^5P?*-_cDB?WoJmtVMV8o~#phl?FV0(RRBwd<f
zk-~;7Lwv+!yNV58v36*k-G^5B>J?W9vZxQ{N8%mo9a(0`YzT(t=R(r3Jl|(wagiO{
z=2)8RF+bO1ar+{ROUv}<7qLYSB^YDrWf^L6de)L#OOY3i!*-VC<hdowGV(mfSVLwF
z)&{tan1H8|*m4v?Wwaa(STBbhKD^4|L#y1f|1f*^9^~-hRrcL-kXsI{ap2GaZr*!<
zvK+Bd4H*nZRFzVW940GpDgp8$AX$dB8EWV0Z(F9P9?A09B`}o35#E~U<ZGzZA<j|Q
zjB+%f$cvCTco$$Cu8PET7u8&dM9m13GBm+<nZeAq001BWNkl<ZQdMY`jcVubXdorg
zo5mtolQDA2E!($ov-4=4G0J?DN7dLpkqP*rbQXoN*mO#uFKSu`??m9VO*&-E&{4*e
zZ%8^CX~IH)z!(o=$wXMF*14d_`PsWah_79D1s}fU05|83(uC@_F(#M*nl|_o9jmx^
z4B$FOA~Rr^<K~sKrA-@RzfpXpK?O}M>H=XtD{`tbbWkNr_AB4Qx7uE#Oe6*=m!wuE
zGsaLkAKa0nG<j2XPu5r6;u_vf+#;N;5h)L5bU*#T3pi)jGHx_NtVPQqw&?LUfB7a>
z*A?b$Ob#us@~sGort?eE)ofg}wzF$>_#(}mCj}xmyp%xB_GC7rEC<XNIimx6`NQWv
ziN`$rXBm}4=JFn1D|}rTitm$8T;XdeXz_|Uw1Hm!HQw^VKjc6C+(WtDFaH`FCoIsL
z&+xvM+?JM@olYw`vhR`{JD|YvWAo$MW0@=*^MoND?**QCe5^KkT^bx`TDdY!WsQX%
z^Cz8xyXGqV`iRlKgXHsbI0>zMBB@LxpP^ZZqD))s%pep?D`0GJz)dAGYo!qH>6Vu#
z1Mu4{0XW)Lmvn<oonbkD<ivJLz9BP;Jx9*{&{@BvElIYm<Rpor23bNYL6t^Cz-CAm
zg3^;R&u9Ps)tvE=hoI-2&Tu)WpBHQlDqi}+XYk)1_jHz)Ph#NfOfkAYNAhZG)YVo8
zYgmSA8WE)+sVOzMn#rpj!Ns=v3mvFJ&CeJgAxU8vG98axCp&G_pe<)dx++2*kD4U-
zkO0q74<<9@xgoQj9`pDJBN7Z&ZNR^)2A+gg*a(vm6TPK0i?hEEg+%$<`bq=outXWD
zvOZFD!*j!dHJq-73Y_<32&xs;sKPr(6@3S^-n5N01a*zB)eVWfmim-HbWV-j)CV(1
zjpS{w4s{+|xnSBTHt;zl=>I!~Jj)Pr^4>ho<Y4AdEwGuv=7L!W%2Ey&`WLkxtysis
zgRvIpBH7tfWWifNGQr{KWf?Z>AzsOQIb~T=86SKwyf%IjM!-vOp$=kR8hsxUsc6w?
z=x3sXu3+l`QCh(PFA+lG0!e!46hBvlZdzr61FmT7E_4ejO;tU84`lYr6cmbm6S~z{
ze5u!O(GHo|)4K3A$+V7D>T43VZ^Kt(z`GIi)jE&4?}Iq`rUTsbu4nPhORndj3PaZq
zUOUmxqJ|X-6Aiw}cX>-hF`e=I$~2#4QG=S0!1z%Jvh%*Ba-OVbsLBw~iqqzKNv&3;
zfy84pB&nLN8BO;)LhvMD?J?oq@ud$&jI~;sfa-W_YWDuq@VA7dD+IoB81WPs2E$dJ
z@#Npcl|!luiXuS94wWNb^|x<Bip8+!8iTD3QYDdZPUZJ>qSQ3O9ZYwyR?Up4WabkE
zB58bd@_vSnN(w6&_5A#QelSmX%rCRHvB6Tmk8@R`7#E>nH3>a(7l2>{x#1}<zlv8r
z_1Qe?u@`d2Fa0X3+vb^@lcpO=XWP)UV7zI}kn&BAHz9r?$9s?U36FWgAR_No)#C}n
zrc1F*xTb5Z%%BEuVaK^=vFExg=v&JzS6|C1C!YX$)>y_nbr`W}_}|tkk~cG7S_q5S
z7!uukjU>lz)%do~{DrAq@tKRrEJE@3PkJ6BA?wjsKep1%NUK3(rKLueR0w0C8nosS
zgJeSHJZJRg_~8G3A^-U=|B_)p!}od=J>je~R(RZ_AHiGS`4QA+F@Uh45+}Y1B!G^j
zWWABLo&kw;W;YDz-h)n&pX|8fN~c@WRy=h|7olxo%UI$ioq(rJ2I@fB)Y|2##Z$xy
zH}J8@gU~d96Vs^rIMXV;*r;b@>Sq(d@y6KNuP*BMhwe`eBQsKKjmNjyww@S9BvyXG
zD)bi&Xt2vOHj)Xg33}|c^EZmGBCN~@Nfqb)R84-P0W;n_BNP9v-q5{7b3VlhSW$AK
zmt&UUD?v;a$>=7e2okzBwX<wyqwJ>BnB*v#3Ecq`WuC?RHi4og#CxqmAZ!d#gsOoc
z`3BwxYJz)2fU^a&sTzhBCEcU!aERo*w%=0%r?hRY8Y)%P<{^M{tw%+e0IswCbUF)l
z+#_8Bt?92s*YPp!!m744c})ZCF`t%M71;Xs)G0<RuB^Z)3-b#cS}!TYV7x=K2xjqh
zpetiZ|Ey|ASL#xyhSIueiATmO+zGUFRF!06kj!DMP@oK}5!U30Q`AW`SNC`?__|Z3
zlNA_k=1q8y#3`bVes0+)E4;U4CJ*HKs;)-J><wl{<G0R@>w6LevEe-)tRLbhA8;@3
zamRC6Ut1%~3!L|qzThqY^iK918n9#6622M+AA-2%nCJ{`lkU{CHhKF7cUM5Ey>28O
ztmqM8S|@V#N@j)8a6l#=amwi@ZR2lW`M)`|zRL1^Fw6{c(Q!}kF7TY<{$q^b5Zth$
z@Rq$F`UG$J)0gqvPkT1YKk*Z6?3`n+s6|pj(q>|Tt0_E5%{7C~c_>FN93DT8Z7lvD
zAz}D{_x?)YoZ~|_>xJoRP1hB+C>|2rXLqcSEpNlEZBVW4$L~LgIbjFRMAu1UTa%w=
zP4P`2L42W$9V_H>b5O2<len%%RgW=E3hfl4j$sWzwqE_;Qb*$ZTx~RI^N#N#M^pXi
zEcBb;m>N`)w*8$(*3u-p_9AFyFj-WfsDxZR<+s1i&F^_XCqDAgl({F%EN(dB3BU1Q
z`QV2>j?q5*RtHEH5}}$-7*t6l8Q7NB(5=A95qdv5qdk+z&^8^2t&XTp2|}0z^2c)r
zIm{ZzAF;emfSO=uBS~0H4Y|`s54C}ANbqZDD}Nfyi9|!yw*Gt*WjOIs$}*|aN`OZ+
z)#QjLH+7~S#LuXJQUa&JyGW3(g8B+0Vq(&Em?UbliZw`P$3lwM05A}tTWFyJ%6d+l
zx<Md#2|dWtc@T&3qp18;1`2oVUIa~k6U_d6R7Q$Hw6DpjfH&G;y^Xp$TEJukj|~?f
zV^%C47mK|<1}EU^XiyE}gQrD|<zr1gBy7HsoRo-jH_ba6x=B8&xtI=?y(W(vYeF?T
zYTT#ua~f+Z1>Q|v2c#t@OVq$O_>dS{AN(Mq6SqeioqK5_(i-zU2G&a^e~{c1ybGpt
z`y}JzHyvQObSB@rX)jl}MGh4iBXL-p1wRlUUB8SZo;s3rHfNM$^e7ka8Sk;8STR&C
z2rqc!5h>ZZeTB+d_S|$cHZLek7uGdGiz&Jh5O_X5jx$a2SiiRhgL58}S%w=2IlO<J
zrENPY(dZE%sMQElOmc=z*082%SMLM03IW8qAN&oBb^QLbpT?oVfVrYTte~Tk!=><-
zuYC(k+gBL6Ds)e5;5`JlY|}z3n`Wv^B7apT_K2ofq9($GTAu}pie+_<TogGv8ev6Q
zC^EJ$T0Zi=w{qanLAEWgQ29WLHqdYu#Fz$Bh*&(}Tt$&9*Z;#m^433l1;6+F7gPM$
z0~nmRNKu#|q!HzCqpEM%OA-!ZU6Py%aXo(gFpeul<O9e0q~<od*moZvudM~cO4k<w
z31mob4cpJXJp&c8QOVlw>qFbUc|n_6ZBtj0iPguzTrWCF#&&L}O3!q|sC1J#mS&o?
z*v{{3yOPb?o^r&k?nvJoAEDW|B%3!_Wpc~XM6^f}<sh#fQN@}7h4Fw|gHgxHS;41Y
z{~EaNI#^$)@{akuzz>JK>ZQ+TZSM^%%`XHrw~+|%ur!=GOoYFg7H2I?ZG%s}(?;-p
z8eD<SJdv?KTd7udzNe`d#2CG%hVo3~Nedk_F5i$RO;i@D6~d%Zh>}>%i@H%5ZN7@c
zZWF8Uw9ah}E+__SDls<?YC{^0xx$~IniOz1S*WyPv009_c_0vr#oCN4&oNnsu{lVF
zb5NC@vh<WAPc`zCBNx7|6z8J{ffu|6^CusJo!}m95w#dE7$2X5$<ToKZ>1SylqPlA
zUbU-C-;9H4&M9?bferU3CXzdYY@}eCfZxXsQ1E~etlB1rt0g6C*pbCxRib^iChv(t
zA}}Nny(cuGGlieAYRFuuq)OCJO8H7^VsKxtqMcVkJMZU|MKzh{fhS{)2N{uBK&rsc
zAVA*G5XeqlyQJ1wYPI<5d4mxuZ}3YydE58C#|ysnDc*S9Rot-Hr|cPsrf80kh@}^s
zcIjjWcVY7dSSFHqnc0ufMqMbVgIA1(10L{W_vByR{&&3n&40`N?tTw+ut63TdX%~o
z=##Fbj1{k9G0Q}sksL<V20#5159Wn`_#Ezb-Z}KGM{rT8EUtA&!?tbkk<yHbi9~W$
zhjGdS?s`}5ar?8e(J)ZdA-UyU@BIkZ-f)mSpTk;%_r!$1i<+#9a&@BES<srIRXWcj
zp;>%Pro#LYBW+?H3pTk8MI&Q8D}^v`*ZIUn@8Iyk18iGf!Ba-L(-!*V$5gcNplc<?
zs^=T8cnu$T?dy2<pS^_Q{`Y0$^hL5c(?~Fj8Gp9Exj#+eRdT#{@FQ~ECNcPJk}!zK
z@q}Ruv8ap%{TeqigT)&qJ7pJDKO<wr@TT1`9N?V~SnUp}s$JfFQn*kI^yld>uiy}j
zjqaND0%?+VCu4?PB1aKqZ*8GImR$?owCcb4&ura<QecX`*{ml0W_iQZ&l62H#~L3U
z;3#SYE%Jc59}L(wSmols_+#>+;74JRo-cAvKXI97J@Ij@9k>NdaK!Zz%%qH|m-7&~
zhKjpy+S@dyQb%|OI+;67tJ#gGSGrjfT8_}kNKcsj(V5|mjE$BgrdF-PcAHPj)OnGl
z51}$Qh9hnuD78DAXiM}<GNs1T)%zG0`sj4qhU4Cc%DqFWG$;yH>Ko2W=!ygqa!qy)
zGOG1m_@wMG?f@2{Zb|m#1)Mch*5brcNrsagPmXAT@qx731hnGuCj4oXxFisBz4265
zDXqtu7D7~6PnmnFOsTBK0=TpWVo=+UqXLzLbClLIGL8~KO%Iqu=t1O?)b}bigfA+7
z8E=B>j@rom2$G~?G!of;^cw&Xk8y<GZQvj2>nM}%#@cq|q)93Tr=DWY0_)uU4rg-4
z_BndqlV_oOW%CTLij%Nj`-Z$*D?~{{+O8i)6YzSE6{u{+q24k#&o6MWpW~d64Kw<#
zhZ!l>1rV3PcdWf8`hN|CD(Se*EQ)jq+MAV3!t*dt3D#Q*YY|=N&S&ky96rdlzU8NW
z?7pn6uaQ}UF&4G;UQ=K+YZ7sZNo)Ol#S7vCr^t^+JmAOf$=|)|4|&YP{u6J0%}ba!
zjx2T!)rSsS?PTvK--hlIRzun}I(i!m4xDp1t;p2lhim-))1QPJZeTGHM&~(Hjd;l`
zU&qS!T_It_Vy$iFQtB~Lk85N8SpOG4w_Y<s;2pY{icb;`4ezW7&Im5PV|l*-=dt2h
z5JO)FeDeJlF*taToh#eI9<N#8C&UHGJKv0dJ<o>gBWz~);$Od>5B>AIdE5*Ci0Ym{
z#>T1JDU4uM8!?EwxRSPC)4gL_f#`Un;fHtJRx$W(l`wn^xcvCAF)Os}l7%{kK;c_f
zGqfn!cKYc!Um?Q{hI?)zOI(7rsV>!|+Oe=aG7Ni*i<DIrWcVUUOubJ6-Bg>r8(fXQ
z^*=g`fZpn)M5l2Zb=sG0<wR#eXl7!6Cwet>&11Excy~HEHUw0naYHj9P}-i&vHGPi
zu<xRa=&cQdQjakdw&Kwbe*h1<$L%pX405SP@fu(#zQ*<&iF<Ry$ZMR7kNC`l339Z~
z*dV%02kDH9OoaE7fq(aV?cAK^G(6tj=r~fRgG~oKHD#2Ogs0ZjkNW;v2hLK5{dI7x
z-P&4Ii)sRbw9!Ym0<2@?-Nx9DG0LgT>T^jlj%}8ronnDBSpV@LDku;6aF6l+!XHiU
zYCNE9$HvrGE!`Pjb1}8kf$N$A)>Yj`nz!~HAaHPyXk@5@`YLjbOhkE^j_H|b4davq
zwh=fyq3u&AR4(#*>i4d${%a?>oV~=f=2@x^l_1@F`0&lV>GiMT^{;v<@BXJZagRIR
zp3&L{ViZ>eoRxYXp=R9$K*o&IYodSM<x&Z0K#66f1{YHUumRp>1+l@r%ZqQgOYIyQ
zgG)~3Cg{v{Q^pDjGF8(osfx#S#*%UEwYxdvq;1U4_qpWKPqVnZ800^F=*nm+(@YR6
zsFUQTt5tUk17?OhoOLd%2RB&XzYjZdST8Zg4y0X~qGX}d3KwIBoj%H3iw(RbF-kRB
z<)<EWAMSVWI|t*oUV+n+QDykQZ~p+-?^%T`kLv_2M#$89z+;64X+1rJ7B(e~KTQo6
zjX$TM+u<8^!_Xx)#$YpxD@&Y?=!t{!5TE+c2N8dmohvISRa`TJ9~05zMU2Eov!hZN
z9@wDgEgyR3^SSKfpXaH6{W`eoJ#o9{>01dx2eE52F@1#Q-J0D^KF(A4As&~<82mp*
z!Vr&tIX-T)ULa|!StRbC(yA4RDq=IXoqiT28Afvs-M9xvWgrh~02fFjKj|Aow_2k<
zBp{Z>9Xo=GMp6M0*<3Y4<j5)$KeV(!kG}ITt-~-a6Y4I0G9`!|PZ8wE*Dom?NxMm+
zbU#IXEs5bVE(}<fWiaxbYBE0h7cYg~2l0n1DhIth!?+=T^1LTuH}-Sl;yhLZG$hM{
zB9Ao&(W=oqAMY5bOwP~g_t7OxPufRKRw=hSE?ujrBm9<5ZeXoMa6*2J*DO=;!(Ala
zG#J}>dXw}lbUc7c0|DurKmC~m_0uLB+6Mcn(NP24s5Q?!iHLS*VX7{pvp{1B1WS`(
zyQve#R6)kG=@76>{A0X9yuqkoG?2Qj3N}KYZ2V(lAa9b>bq#FmgvB8?c%q;Lh!u)v
zN>&J^+W@Pm7@RR3y&i8Ws;s24B~?~Y39;6mqsHK^#Yb3}!Qh<>Tnmwg-y@rJ$hr}7
zAQy`<h_RHuWarWnr|#TN52c@HoVe=*5W$rmlV#Y<b|iwQ<gtlw$~5uQ06_`l@UUA~
z#>cLOMkjIKCW?)64%0moWV825$9i*<t6RrXOhcze)MKqjeFff;_2>BLC12)o&;ApB
z^T~g}C7-zrl_IEASrfZWp%9`bCUQ!vgocKo*T#;VD%NIPa>*xIa~X3dp2jz?-OYv?
zMm}(9yt+myN;!fSxVG<H^G-K=sKJ_yULT(G?58j&EAlK4B<)ejhJ%;C>J2QfEK~Zj
z;Ys;utE0Y6h}%C&O0CIy(#$=dC?;BnkU<cvwXr)E3P^dLA<lzV%;ivIBR+S@N6~VX
zZA&XSRUm5%5ZV@{TARcSzrI1w9p*#NdJ_Auxq`>O<bP3}b_UgsMGB-QDG-sIGu%03
z=30OJ_@N)SS?qnAB@80+uH)l&<eq~zXxc6jHdGu0mggz9?V$2d9yo}*Wnb9j8<Q37
zYWt^HPzj1Ky`3jenxHn~HEN(oHI%M=&*n|Vt>^)LKi@t!zoJL{&TN4tnHrhSS`uYe
z#xbj_M7P7x?k7o|CPr08Y$QZSlP06+#*ow1^3fN+fP6T_4F-6i-^<y#2!H;9XK?Vk
ztLPPhC<r;m6qLSf;+94u97(EPM>gD$H2WDN%a3b|Ba_<gTeJR~5!jCxw`9E9nRG8y
zQl!^p_d>_J8MCfJXE$0zCNbU}9}1aq?KKD`0X{{kQRy<7+WV5)%~=|!_U6Cl-<a8j
zk*OV^bWCL|G2X0mw8=Etf@-HqcP2D}ZPev=-eGdVnE@uN7;GG7p<hJ#SRcC*Ho{nJ
zI}43RaY&R{#c9Jc5e0RDOr2+fuPVg1CU{a4Y?BV}IOe|=3aA~GovSc5W8cA5zV+QJ
zuyYF>8hXBb**D0tB4FWNfTM}bdiFAr9L#z>BoQcRm^8ImPRh@zi$OEaAv(Jot78>;
z9jnbxKs{Sww9qZfLK0RvaHinBAO1KW`~2rw*twI^gWA-C#>UxSJJYvwBXsM6ID=RL
zW5~0NYp%YI-+IzBdCK!%$p8P0XEE~8@f#)3XcGC-t?UfR|711;@A2N@)brr`-H$up
z;dDejc^(o5Tl9F#Kfaqi`$mvkur79mYOw{KSeLYsww7dWweLd5?SAYX3tgKQrj}(E
zH!4v|77LK_0GD3;A+$Wqwv`1!>PYLr90#e9VU3AQwSSG`z#8v);!|P0;+J0br)-?H
z3%{#}$!h3QLdH-@xC#0riZ(cY{Gej*+jPvm_n!bBbbJUmtEo;O67j)lQhh~r^>y6%
zm9H{yGy1!Bv2ekCD2>#$gw_nlB!OK8k{jG`#QxX6k?+3gO)L+CzI3lQ7sy@s0CZ2H
zy(5~Q&74NFjSW(hy1H^ITXcHX%_>E=#%--r<@0UpebP6G)B#(>biOzFaby2F%NK4!
zW;9gkS`{4AG1L<UUrTnD4sSEAaU<^igeP;_uU^P#sYjk!DqkWd=eaL=6Yse6Ym8(8
ztsK@Ga`EB&1a&Nl$}vG|uipLJPS=zN)gcHrT!4m{+9tc>)y?K-NWY&1Av$X|)%xqx
zE=MP%DLdCf>jvGqnsF_Dol-eXM}HIPUKtzp@p+Zb?P-&D{M@Nm8QN;krziLI^AKsz
zUrY2xh@r@=#5Xm;N+BVe1X$YDSl1-5uGNSpQlrN9cxtNNj)T^8d_|paB4MW`54z7C
zdF-QpnNv?YnaBV3vpG0WMlN_ycn{VTc<;%?fs|mPjH?&KXIL^SJZ0>-8e4)NFkIhY
zVR4CJrFb=%yl8MKBCSL7IOnLDMG-RMtMK!gDC$Ps>z;S#`Wtt1;|=@pCTsi`5(L)F
z%~mIRmD2S#(JAYUG<sQ0(lc8QdJ}`nu?}O}=aZ>t?5=OMK&%?2F0&hK_)?;(mU@h}
zL74}Wh!)iK_25_Hqf!!yH-M@S3R*67M+fByC(3+(4$<J;Ez$z8O!_26r`YACH4lm2
zCow3m9C>CL4EFJv5BxJHZR=wsXRcS^%5?@}xXT6qozkq}j3QZ+aZW-sYsqNwFtb11
z$qqWeR}l&7S45wMUJo5r2qnvf#Z(9R$ot>RTwbw#X#wkf@Y)vQOsL-}8@nI`_||~|
z_U4;;(=&dbGfz8#hdl4Otnb)KwwMRm<=Fm|X^Y#Yq`zCS`5r%RtK*Z#m<Rud>oCMk
z=AFmKZ%fZ<va)Gv2{lW8;>no)5{1BM|6VwJ7*W^3FSm=ZA@$@^f%#KTVNjK!l9C#T
z8Pkag6}=79$VpOg3c=v}U2=5mIk4)KHtQyt$+SL$Zngm+9a3%AOk1W`pe2!8RgFC-
zZSeHYd%Ovn`e*j%_|zNz2EKPSx;6;LTqb8Y9Pqz>|JOOO5atPLHXaT7^pZ$&Cg^)Q
z*EQuCju9|s5t*k!iS$UxpB`Z-HyP6j6(zb*ZJAP1)RZ}`lbxi^D4sGNcF5-XSx3|P
zn)Zrx^+%Y1Dz(?46G}GIo{5Y<gQI<~bgUOO6+)w%pP*CQ4v75P^*4~fE36yxKcD(L
z-23*YbLI&raQ+={&tS9>Avp$PZBP@7FLEs%>!fys4scD4#D1F@HdgoX8^8KU-uI3-
zvu(cL@W8y+hO?qGW@{(aQNg<)*$M_nj@o%X{iSbk{hq@(lOs{(Ea=sn1`;>^2wK&8
zo!|#(r^cTKx@p427@2kpK)~_;6&<sLZbRScz15WK6ChD)qT-aaGM(X`ZKTlE#%i@g
zdHVu4=ND}|3>u3wwGx>TY~I6T>CG=N*I&Z<45t$1ffI>kNXC<n;#X&lY65ziE<GX~
zJa7}g{n$rx)+yV`RIt|I)PbGj&tCC54sLj|UQR_-+Z#5^V`l_KJ1ThHu1CT-lCi|F
zPC&c_p9O*$;b=6Z(joIbIB?T7eDb1;SnhdtE-hhDTuhp*NK<UkXbkCrL0MuBJM4F_
z<gb49H+kR#AId|X_hL3q*hwCfh1$qcQov~!+1YuDa>V!k`0+!EC2!N1+uX+2e>`E>
zv__e#9Zf@!#>J+ev+bO-82S=lRvg^Dn+!vx0m~{AS`bN4VU&f_PiLeGVxv-0tWwSj
z@APQzV=Ldgb$}*0%J56%*j&4eg^4m#1xcpVoTmjG(~>3ad``XDi4b+?WA0Eze0&~o
z*sKpYU2@*>{NJZH$Z&^Cyejj(0){1j_wpAZYd5iD+cLfi)m&q3l;myo&FikoM1^du
zb>1No>lDQQM4mYtk|VPoe@C2a{DiYZ^GPQO-P+WgvW0JrXR*oGGA-3S`l+Z#_47%G
zY-TBWS{LPL^O~`cbmsMLF`*O!j3^80FO~O@6_l0X($8JS_FX4%c-8au?|d&P;aE@U
ztDq_+B_bYgmC71Qaw^HGjNpu6WGd8pvcgc6hxwTYU%=BIe<61|?GzsIU;Yal>jSK0
zje<=b&^A00oh1F%*KLfUsvJgya#RIHBuq#S6soF31Y8w5K0a37r;&<}II|KiP6Ho|
z|JEvcjjup*6w95V1vz!!24yT!RGBbutcmC88zc!S!_z{Af~P=`kM>E2urYv33GAKM
z-z2J2^<0kxcCBRyqIIC8)u1V|;L%tSuMN4{#?Lkuwk;R@-jjcajrAe@-aLi1Fs#_U
z=OBOc`hQ~kj_nLb8|1ld#X}M~(OE#u2C67s_^(kDv1O+Df8zxky-<oQ55i<xu{3X4
zJG_@qe)xkFcF5vFzwuhJ5>%X2+o&S&AArg;KZLc5F5(SO{4E~!xL@H87yefUr!JGt
z*PVs%JZnX@CY4p4;hH1pt#Va5e*6HA|F{W*h+K6%Y1qOJEmOAT#N0)#uyE!sMnz7M
z^*M0$wJ;pgc%do{zPYJXMn_14fs;;vUeR<P>h3}f?U;t!oz>i)rLDj9y5$(mWa^@H
zg!n}_2_m;bh(0QBnOjThAS=Zo001BWNkl<ZlFieKF}TNM`z1<2Hp_RA4<rk1>Urnb
zw)+OY`m#TvHyYtbL%>jE!Wlc}c){=fIyYbQ9p>^p?3*s221Lz7{Zblv$ry=fM;~;e
z0<JS%YG>bny5Z<}6g%;N$3J^E(?B+jxMfNxG+Sv&bo1A<lRoOC--IYrsL-3oY_lr%
zGr1CUY1qJwpPBx<Q(x5a%xeo7glysKN%??rq2dr^TVt@+qBx5F9It%+TY2IOUdE%I
z^dt@qMuE7DA<GO9V3mji4dm-gi9~A~o|%ol6wV=<F&sKp3QJ}US6q1&MbU?-I#Vn1
zOq=gAC9L=AvAGSVVR;Pr#Ny+cltpv0JkJ6QCv_4NnaZu{J{g%Iz%Z>5AhSS(TVLC1
z!nI1*k7G)*Iq?>xY6*>-JRJo#jSlA|egnwt#HJ;1C##A`B=puvQ{*m)ZMb+wlFUjp
zsr=JI{G{#vszwzdJVrgkwL?7TDZj&FZYVMvv7nw}uFs#o{MAUY$jDWAt#HoAWW_a$
zMsrJ|9i!4Vp-986<U1(@C_1R4p3FuyBu1<(WE|Xk9Ur~uJ)C$#pXH^wV9pu$h=%u}
zQQ0jfY3}d{d#K`?H@uZMzvxeS+_V3H6Mp`o49;AEEI3g+=eu165>W-&Y)?$a`}pxA
zlPtuT`!+kW+xiYf<Q=N|KaP*(CV}OYqG&@<^|AHZ?=g4Mscc-ooB0ez*Iq+$?m3KX
zR8*>U_G2E{v}HwMc^SLB4Eu-pvI3LE^+_5%{&qE=U5h-jAMp2wW!w^6a@K`x;iF|s
zeFgo#(u9fdX_^DFkW?LYYGvF<4eZ*q$JNfU2C~AUrDvz{eEVPD#vOkA$LT%jfsCxD
zmu2LB$fF*5U*7-0i@E&TTR3ELO79xeKcfNuff1|(P5IivS#&a5)&%dK$`{I}LvSSg
zzaHJAJW?{*`EcE2ejep#$=2P!Rc2+Zv!9A?Qp-<!P|762T*uy3ospzWx)vYBbC#oO
zLv&xV*#k7DA{9u<S}7_Ss{{Dpzh6pj3-B2>%4=%Cz+g-bi|}Oe-&7K}^ytd1UXiTe
zQ(yQtPyOS+Cexe`edbGen>QRDYr41O8?!Z?@`P&zGsS4^O1N5`PMV+%t^GH&lU%A!
zH$NM1?oJdH8gYWLs;+8c)uznu(N5v3)XhfYT0sPY@)2r7wOgxZUaRP27gER1R~wJT
zYpVl$5}s4TXPXqN)}$*S$$Zx4+}aZ2l5;fAw$q$RtN+q{ma)U?ox`i65aF~Fck#H#
z{vy@-Vdi@CpfD;&eDxdO<t_j6ezxyCjiE0w{UH16oTsSm-0Mdnnut!*IHw1b9Z%|!
z(r^#c`;0yxd6D6(5~Gf#e$L@r_V91-csr->SYSR6J}KUVN%j`+DvY(wT&WHX$eiPI
zFMl~-{`41l;#*(O=;Tu|rz{gnhw&UHVXLTggbC`gn!1`woor5~8^@0yQY0zzjvqj@
zdK=fzRSQ+U3OMEXkREmYr?s!gV-6hVx=&xqG7jG_*!F-2Q}zn1kyzmhCM&hRbdX&w
zk+lJzdE%2<yzIO58KIIPHdIlx)q5i)?oRV6*enFpO}vlCH^GJkT`%+^6JdoMp~Em0
zexh4{uHmzNrz=yJwmJ&^TAOm2qvv*M#0l%zcCVA&7HzyVV#!Z65(r~3A2;86hq5ZT
zs>Jo@`IZqL{l@>p@T^nGmglen>l;I~*W<rG=2_gR3*2;IoqleyHeP^llLT$r*)FNA
zYK!$v1~HX}CE0|-G@2qGnFK`A>_GLs`iZ$B$!9dg-|?V31_vqOvD;pnZUUFOs!r!y
z>qNV1Dyf;x;d6e1^sEypai_yX`)nl!t@TSH$-iszOd6f~+RGpYs7B1Rx2}8rURx!W
zgkV4U;%f7L%9K`Kjk%UiYBSyh+$2Scd>tbJs)Dg-1wz(Tr^Q-4j;g962df@G-#NV(
zk5SX4Q@;M&RamhM+&ZeB`T1qmH-;2N4}6g0G{yunv+CsaNlKhgN#6-$xE>=-xh>;K
z-DZf!DeO|6o@CZS01`P*DG_~kC3w0ZgNnA-7&#qMyOVpxstaR7P}^&}k%AY8sW}v4
z(vv3ZT;vJFfW132WkR?>NQ)EG{PWWNNgAU1m=Q~tKUVuY7@;gntj(J7R_Dl!XKnv2
zyz{Mp%LDFxF1hzulcByu^En>+i@(K}zPpFKC@9Mc%}wyX@KE!fyJP4RMWL*1w?t#s
z8q%pIytBqcwJ#su-OPk_U8Q2ddIkq>;*%fx04JR|$3id1#<`8t&`ERcMlB{34Tt+y
znJ*n5e(oQ#|IlH6@kK9Xu(MCTYca^wO5oy*@z2N9e>W?W#P?n$8O!6z!Vme_Eh1-&
z$mq8GklUKWK)iu}K0cneHk+#@6oG{WiXA)gTH@B%8SUR6Dg^|LsHwdc>cu%$KPoG%
zoP9PvR`R_11a&kfpr=PRB0EyWapp3k6ScLD1m}@E07-#EzK<MOH~+?##bjkG12}~&
zoM4bOjg(^~pz#Ns_~2zFL4HusP=!^YssovF#988a{~tVu-oQ|;S9pcFJi~7cc*CE+
znDUkzIA!}Xs8Tt_xjNzqfvr)Cbp{(<bAy>u9Z-yot)y<ITjQn3tOq7Xp8ZD}M%{9`
zTqnB$w`z88@(FV5w#;!-^JagK|NQKiW*Q}N28`f{FWwCJpG@*}whT>7Za|~Kq}S+p
z9s||zpF`D?<vD7g3}B$(UY=fhZ3Gv)1458#wKk)2N|Dde&*mtHC3&6)W3|K#K@)$o
z25u(fkEgAsrn*G6^~YvVqNxoR8B4ws{QYcq<!O^(OXZR7St3Lx1l}1^MeMu=!!R`g
z{^%m?Mk3!qwC)H=TqDM`&r*byX(3Z7>?=(OHosE3&o~{4G|Qg0-qwbd^&C?r&x)o4
zq|TuzrFY!_es|+1?spzkkmr3+hdSXsAN)9<{qi^I6@4nF*gPjQ8NSrkFuPF%Yi>G9
z|Jr!v!B8*UZ{&@M4NyEW(K#-B@Bgv)-r<rQ^}YA!SJgc;Cv4u;YJ~z4N?@dWea$r(
z8%!`b05%xg7>n>?lM!GLCW8qg39L&LISFB~F@_5^*nlzSVi4hCFqkBQK%lhJYIjeV
z&|UR=|ETJo?wOuBbIz`iUcINEXZJbJITO0Os;hqC8^XNI!8lrp!@Cu}ap~vTyS2qa
z%fw_MCfUZI(S<oU!CM@P?G2beaG3Wz^buGU{_vF-vV6=o+Q%$ne25FYAF7hoykQ#E
zye5o_N~KG(P9cTO<Hk5XaAPM6KbM3dE*6^!!-j8OT=Kj@tzplpC(#2^)8>XBUV+Uk
zhEAn%Rk7*Bwy;gucFL)A)yFoZskqDzw?b_!D%=<gfiz_i>q}84O2E!83tMZY>ZEte
z1`?0aJxnG9i9@JB74KRmijhcJjAk(N3oUMsL0cA4&s;C3{evIz{kOb@cE3lS_fWBP
z<_vpwro8#(&t>__%UN9LAY!m-8Wl|hW0G>?89Gm!5?&_N22?E+(t0ShCU-;|Bh*oC
z`kATm6EQZ;*jsG~iq~Qo%uE!9af0@G^H!ZTc%KgG(phSKdUPIPX>=A?>k_1bzmM@%
zoN0y9jYQR>pcGTYg#kBitJUn<xr8zHQpj@><UBQ~8mSMu_C|u4nz2LG9+mOaqj81~
z4_|$KV?%2V-TE@SlFF2+iwNkN^M4kNIceNewCn&#>FDcyP*_M>iYdIoMG#}jy{nLs
zO@#Ff|4HSVN}myxA<9*Uds3lLBom(HY0u@@6Dre;Z={4nc^8Pi)|hy{%3RCPTe_YL
zU-FmqyM5+5Eu0T7-I8|r%NJkB!j3)k^9(UjNyzKS>NRwuFJZuywNjdrJ%)UW(CrYC
z#M19~q3ubM0SoN`U;V<TNu<lxtsSfwa<9=-q%vuXbvuW2DaT*>6)t}4<M|(Vx+8ag
z?z36h-J!E94N7@tIMz}~sFs+hVfm#BdYoK-sFVjpH@OTqb(r{bim`uQ3B%`r&u<Rv
z#&%{OL8c-W*}fgNY{wgeJA44HzaHx~b`J`BAsLdNoFLY+_2e_ilN6Jt0lrjf@@a!~
z+LWsB8u3SCtzmS3Uk^5Y)>o}VR&`pTTPD>7*CXMy*7cW(QZ?X~=!~nw>b(xz^K~_s
z!r0GXt-<Cwd%W@=?|C!Xzh6qa8Y+Xvv`DSv*Z%v7Jo!%_MECG@fwSO(B%-xKo@W^2
zaP$J6+4$(bDnr94X+Zy1RG!ulGpbW_n;=ayjl9fo<7M*eboS?M78AqCf0`r<aAc;2
z&0irC!}WEki|LU(hvO-vR+F^G=cb~w6BAUa1e+MGLlrf2P^}uQMt6Jhc=34SQ1K<a
z#;6$*cXfyvHw2*vQv|X)#L7#>BD;#o9%~t?mR-5=p;)`i+N>F+W6551jVL`&7#g^Z
z<_a4a@!>P9QL-69j7LqM{>pyDFXJ^M7p3cSG*+=gLEH=ttgAV41?yKbKy)-27(%#8
z!D?-uN0W-%(wrSkDb^^)TCyy1GBoi11}n=v`kV)In_C@2+j_hU_j`cl&F}piF2Ck5
zN*Yok6J&C|i$wN{13Ogtdx{s0$%7B(fhb}W6^BY*!rYY5RZCkb&S%WeC8Xr++1laK
zPk#cxvdp$E3*f=!u_+MYqn?lehzQ-44C!Ya{>aC<=&8@<Ugw;{FFf(dbdTN1{9*^I
zKETJKS%skW<+y6#Ss1=GbQDT*e6;|hX86?0W|!e6IzAs`>{~{{5Eqlb-5l5TABrF1
zK%>#xu9Ht=HCMI(hkkfj2r|n`T!Ra*G0Yu*96bZM#>yH)qpKzd<XP+fM@@lsy`iCu
zhziPV)<u)T!RFXfkLWs@tLs%LmQF-?kBYO32|+$l0#7su#dz<*3aN$LZeQddU;2FH
z2R}g%t&n>qH7R-C=kX7{7kB+1XELWAr;Z%RbAw6J06VaSS3|XIO1ZJ=h}@}8O#F0>
zR8rTUEy_A)nl=Oz4H35r>uev$h8_ahz%wJWUZ}2Bot-iD85yd{w?-${9GkCghMQc0
zbd15eN`tyk@eaXQG@Tqv?PXXwuO=-JJsPM5bi+GV#beZ+%~VDd#k4AcJoF@(<}W-{
z!P15ltNAWzwR&9f(U2oukm5}l0#SqEr?o+0?aY&p<L$3~CHJ}eoiV-_4C;~;RmfZx
zc>}{GS5+ERSu0S+%A-@FE*s@wC*7IJMIM=4qN*9@`M9yAX&QG_AG%JxKF@g4V{zFk
zMl3EL&>!^Z<(3z{<aKP_xtrYOU<DV+?KZOHgTF+x30o;OTa`a6x)V}y<c>v~)&2^$
zCG=JfaOyEz`P#?-9<zLiy;~Qs;_*@;A4kZlGfIxoTk0`4$ocxqU(Nr%=v|!qR~NGH
zx9>x5*BqVsByz1Qr9_n)A7s>Dp%}Y*e`GWnF4oV;%Q!-<uFd2BJF&I-IbPS_0IQqB
zyOE?JHVGr~Hr=xe7UxOCksrJPcW}8RpcVG?hENT|hf1A2W*g~_MRF5b&ju-_=tQTx
z+Rj?{o%KCUU!^nr+*qJvZO+B$MqD*K8vP<YDw$Dpf;0Kd45j3)46)%<7Vg3bNyK1O
zG2UZ*sCrqOpt<6gy6o%a{O@N!oz`j#f5_v2cG_Yv=<&K2p3B}jW%u?iWIYX)LM-YH
z9>K*5Z{iGzm$<nMvC74!qty!RdIMmYiU~UMjYp;@7_#mUJCYNbaq7qn(@kc1Z?ihO
zj>n7Rd5|2zWy^X8XPB^QgL-dmp2%pbS?JaWthjiK@fInZam5ESu}V@<T{J@Q74`T>
z1ADCm_K^!V%0MbU*OvkWy2kKS6V}P(k*1!5OyTfPC__w5*`$_QN!jdcwNkP{k5f+E
z%ZD$13%`2%GkN6;o<my%aUPd@tTiM_{ch+;f;A?cEK+M1=rkWW8FTum!(7%;yV1&3
z4|&jnNo)zjSzTG>S<iej3-b|PC59yNv^pK0_mWrAXD)KR0%XV-gIC3=i`|A{RlZoG
zOCP!7HyU8X9)lNyN)nX5e1MfEsU^z?%*`8oeuz_VeGLEh_m{AA!vS_KE(RV>gsqhY
zGATM`mh@nc&Vd7b^4#<I!B@V@Ltpb!2DiUGgJTwGmB_sS7m`v9$(PTUR_D3K^Ho!#
z6f|;$!nqHVWFM0vql$^%7NTq>4L8lP8e{8QR>B}6SN}XO>=u+i*189%fuzOGTc1cz
z9c?jO_k$mj_<->Z3v=*iux8j=)K0>-<4+*778f_V;3T~^W>+?T)>Dzzvyhx+YEXC-
zgS;x{b@Yp5>Rvzfd<qP$P6{MOepf&u8t-4xm58#Dg*=~E&=k~r7My42{;T-n`Ojgl
z*TwgHD9|=33<kXG^)F)S^2@l@?p-8lD@gg-1VdVfKa?Akl#EUpNq}Sw8!;jBl+mzC
zCTp>8U|eNlGC3Y4>l)Sgk?GJd=VaXl^jR7OP1CA~$0p@JljhL0`!h86m9ZFe&7gOt
z>_%9V+%BW%G9i)Fkw>iZD2k3hfRXl4CQfB=A2HU$Za`h6wtUA(f}xdj6feP`O@mzP
zFfgv*z^Mk2P0r}!J(1ca*G$qXkBW{Y5L)k8td*KJndX(n2m^V4Ksz-Yzjqgj&sazj
zta=vHHffrcilFL?wa1kK=8y<of6jy|#y)m6qVJE6sgKE+H${^_$t<(8_)S4>F5R}p
zkPk9YB~i!8`*!oVM?4VYa^^d87$fw1j&FYFN4)brA7<;$z2x4N@VKHXZ$!ZRiWj!l
zV3vikUSWoY_vLjBIUkT1==Zy{7cDD?uI5f>--ge9@V(q{_#oSN>_An?oFf35eA!tr
z7<`^#4-7Crx}5ht_)+XU=45{V!oOm0`mJG4o5T=dOSPuYNTp7ZqJt)t<gX#gmN;&{
z(p9U%38}(nWpYt)^SE&e@iDf(#dR1Ek<El*g9X!2&x%i@6*{*%hHj^YbXs%|Tm?5A
zz&MAC533PV2rz&ca!J{9+8K0JBhgjMfOF`1Gab+{y_$3MYqaaH_^+*dDwDRFMs<>c
zOR&Z+yaonbC#|4DCGhtZ|D$T-zN#NyA(<H;R7lj4lQ1v_Iw|6Ew)eW+@PGc9D?apo
z(mo7wk0h44wqeISy!WD4aLtdtLo2lm`Z;PWPC|{WaAEc8>Mfb_q6#PR3Z$xPvQ%;~
z>TR1cT(9EaRFB3?m>j@2<}R}*DN;*rrpzYYSl%;AMUWZrMD<H+!p+ah#Gl;}TaOk?
zx;jl5cZ{x^%}vmC)`FZlgiaKbungj25F{!XA9xL@R03sRA!=4k<Ra841V&L@KqUNZ
z!D$X&X_L^(6H=FuszrPycnrO|vXa~E=&WP)_+buLNY$EQ;t&fC<IpFWd3?swD;b)v
zwTvNi8UOLM|Kh3xOSp8NA6<Pt*Df#9b2(zc8IL!Lu?mt0<sUVOw>9_o;vCDgtV|9R
z8F8(ytqZj~q*K=9@q0NE21hcziR+`r;)al<!Te9%DzCrr73BSl)C!3;c%MP5%`;y3
zQnu{cOK*?`x8!2lcpr->g{yLu^^JTEFCHI>;wk|;oEMVSu|tdob3tnjSvFv?lW=(d
z)jaSw@5u+>bP<P^me{d%8+eb4#k}0Pa2+BTBlzBcWGQ3mQy=H8kADnzdFX@rZ|6OZ
z-YNUATidjXdmWV$r7DRmMb~B6JwW3>X;y<Pdf&xo7TzC<$H!zmFMNE^KrUxoVQllb
zWyI7!?}Xu>fv;>1^ZG)h`hpt*OrKb`-TKxnX9KpTmV@8F4C5o7(8$nZAhsgCHSD?F
zFVHI%bTb@&uj7laR4m{phitOu`5$ShF$uG)Q@6ymKWVL3$%JaDtQDA!#72y6&{Emt
zwUTg&jFHB4gm5WiDn(%&ooXPmTN6ladCm#a;!AIQEz6(zKeSd>@tTv6Fqaz6KJ^%0
zaNeJ>a@|kaGT#Z6w3t{5HdM51=$vaTL+VJ`U1};M3{`W4iUD9#?Q_)O`beLUjT^wt
zX2_<^_@IqIcqZku>#{muD=m;2Pi1yOxNK<EYWTg>E5=fB^c9^}e$a*-t@0em5W-R#
zj8%ORVuG!;)kB9OMKMtKqIGWv)smQ6Ktd%Z%j%=DoYzt(NHC4o-E}z7_`0GR*X*(j
zq=<SF<B<q|3g2tueT!XWsjdr?%Q#J*IH4OO7zIDb>6jKPeYn@}KA3y|(Zl(jKYAE;
zei3U^RD?XwV-RYF#BvHn9UYVW(xNh6PWupW8w#Z!ZC=H7bbXhRRb(8e)r7Y%*H?w4
zmEe6&?s8VUOWfz4cjuRW;RF)#bXpy<JfoLqeCBgs<nv$pFWOtS;KY^g-IypdE0ISN
z_hJ(@uS;%0rE_Kq)h`jmBzP}aV?jLKewX=$7C-yZWjycP$8hn>U&Nq$nB7|zz~{JF
z8Wa;vTd{axu-wD+l>d6eyZD<|zlldZ?-d+>|ND|3zev&!zDw#9sldR-{4A@qKOKR~
zsZ304i_fgO&=*|{V-mb7UL(gtWybQ|=5d6_S7J<k%S;$zknq0Eah^547;iaivC9c8
z9Crd)(xxqj{Dwp12bRH`a+&rnv`~%rsL4oAJ&ArN!PwBV6gylL0<SRg9?C?as{OFO
z5MQST5i>g-%+zF`HIS)RHJf=7<F2|BuSA*5<JAdgC?i~viWZ?%#F2*5gq2xP*^r85
z+lxbSSgTkC?-XZ?>Z)>Tn)0z1KNtD&e%y6Ea)l&GNNk^n|K6`~pF93H?BFnStrVC0
zs3$L2lSK2e0;@W1?5Y!l_Nba#trc|YQR+#_j4zN4ev?f5c@{mf>Aa0eiDxrvOV^y|
z>kjPY$X2Q5_RiSCKlCJ%;S;GcloOu(NR0$GG2&HGobfp0@U~D(0`UpPCuLw;R7a~y
zO16r@#v$T|Rn$tQU&Ujs#XH4g$!X)%khBcW_i;%c6YC0+-4|N?Yo<^!X|d$FN5urp
zxOW(%WP?>))(r(BQM?fp3z`f`+#*w~;bbhCs6MsEg!>qbh#mcb$J<3Nzxog+-HPvl
z_LaoLytJ@t$h?al7$$Z)9nLzu@f8S+#Hw^mEIe4X!3V(kU?}Q__gbo*h9c0_ai5QA
z3S?vJdAvVX1fM$uPfG-w<-F=;&!fM(ioxO{ncF2T&VBj|*t%mcy>2(`k%gL(vmrT%
z;gX5jsZhCp8oWTn1d$O4%0x!N=N=Nl8Ande+<d~%e)wJ9`NmiC?8iKem6c_-%m8;l
zh`-H6Vt?p#6<v|kq$o-<aI_EKz{j5UBrf~Pr9AS@7t#C0v&r{uCsjyPg85-_|74^D
zSPA%z4!Z`F0s<8yw@P@Z#Xl+3#R`aC^ng%seozi#p1Bf^rAK3V**tDmv9!4fuJe7s
z%I0vdw^kRIR8u$yThzAMeex-+<~eh=#i5`4j5N<tb+KqYR6R&h*tHAUz6+8BRgEsf
z)T$&@0@dkN*tMb@so_|s2$pUz7L;|qTgDVRs_xXIZqu~@NG4qNc)@>q=+va%)6k9(
zsnAztBz3G}>BW?ir-DJ9BX{6hEqdy3y#c4v=0oS6OKblP<Oi2fJT2QMTV3VN=Rb~L
zJ^Mub>H!jCN(yJ@eQ7)t=0@DawBR$CDfr{jNLLy-3{7cAQ(I-mk)Ad2(AnO<zFyRN
zLhET0i4obR5xA2uj^UXS4$k^)G@+AZy_Y-78zw9y<ufU>nm1}KLmlF?!&73sV0^&W
zi?2wmnnJ?U2c@yvvn}K6K%qD%f>(!F$TgUmi5=isMHw*HYN6gofeAspDf<o88#-!b
zqC+iG6pyhII~<-Q5jri;6Cd>m?t8DhlDK{d+l(Q18WIfeBY{^Z$f}NVG{zG+rxCxN
z7;<aC8v27AjA7t`BteWBx{On(0}WMjsss)RQGRJM2_W8MRk3+a%V)H3;Ia^~szgx^
zjmCL0?3fg4Lo%v}H2y^CC<{+Ql44@et<KXD%kus!x%f>Nkm@R(L};}VoXauk0xy2a
zMclA#FzH-K&ItU4!f!&#HPFOjjmL@BoVE*Bc8$-mk-#|!wiN?188Ejk<>0lK^QC|N
z2!C?lyVL7;Sy-6Idm(d?n})$y8{XT`Ne=YjOJC;io^US5?A*isU-1fh#~+K?H(z!;
zf`^9G*JRa^){WN114J9kWK|WUaqMm6qQvLMcqKInWDe_0=zdXBH`Z0c*gV#Cti)LQ
zmYpz&$W_2!Z;tnR0}L&P%o9j>Vl18GkE53+kR<rkLkzC^8EMqGHdb(6WAG=iWeYn`
zJ&R0IR3bi-K$bO9EO@>oj8$DZ*O%o~QLM!tzYb%ewPL&tDKb>(jXHDeIR48<Ig{op
zbR*KF!C0tTge~L3st!HloB+^R@iC!N=N;DA&>l7xpIQ=YX!Tdww|YIFdGRypc<3(o
z$$^Erc@8fx^QQBk%-JXIrEMJ~CTv>PLShgJZn?hD9G3|#l(hy!bz%}X_7PL!e4SiK
zWrlYh&DeiKHF%v+a+a}<(@|)u98ve7sWmp`_UVkBl_}l<IzmdH3c0HCQFJK#>oGW^
zcpJM91zA)hwmljjs<jO8coUVDqQi03>b7)y_KLWm+n-uN=yT^^JcWzj@G}13zW2oS
zR!EFRBtZGRQF0TF=ctgTt5>U)KursLxDRk20qrgEq(6HQPkZDa^0H?>kp;<Vxs1G@
zk)%mbsj*2g<*cOyI<cTLnpjp74PV@UV=-x3n&=rZrRIT%hK`N5h_8y+x{#=;V!dEh
zuvo+yycduJ>tHThWwC#l`N4kLvO=!C;AAg?x?zG=jMwOt)d>>KHSkR7yd0|GVQ&k>
z;k@7HH}3Mw+yEUAE&u=^07*naRQoQ!jPWo(H;3Tpt@ilV53b}b@B9Eew(TT$j>M)I
zYcLK_2eFfjXvMLbP-Yhg(dVF43sS*o0z~Om<4ETW?UrDAD}3{d|H_%i?;-E^Xt#p9
zb8tBiW||s{H~pmy297J<{{b#~;(7eaZ~P{AIPY03ow%2D#~dk;_^^MgVay0crJ?wt
zjHL%*o=S<&D}9_IpRG(bBD~L+n{fET?D{U&r(}ng!gD0D$zizhkH3zw^ew)_fQW1+
z3>(_(1*o+9jDj&7bMk2{XO4LcOIQ3D1{ruCT9hWpVj2<#sc_sKeu=|62rj=iLDgf7
zghi(aQY&QKjfF<zfsjmRmaaRr(V2d$Qv$PQV3e%qOr#-+sKnDvy6VkIOeKJp(FjXw
z@6$?lQFv}B9x;9>l@~?GeI)bE_Xn8oevj|H`HjqrrMJ3*_nxf_3-nf2dG8yZ%c;BO
znG=u86emLNl-vcUaEZ{2A*4#G()nZ1i(2TX)4UjT{R7|FmAEl$S-OGQxyGwyrdirZ
z3~sC+%LcDYnj`-41Xr7-Pnp~pY0ivvLj9Ye`|vF1J(gu^h4?&FsWB3&-AM|fE@J;f
z3~0nu;73gt6xS>dA{HYCbvY@X7d`8#oOaw!p8n)>Nv#1K-q(%GL?(E>XuVJ%M1cw)
zJylf5UB;QGokqXgqup+irUsL=u&Kql43Ee8XjUg?Le)Ym&x(uzD^;YPY6v3*p6c8S
zpW~wpl{Z?cK1+nHP$m)Lwe-(03UjW@S(@{RlTPMgXPnAytkN!}mkLOaAM(6LI#NpW
z{-F)?qAXRJc-=myWj#LY^1ojBG6uaKb8{Usm(lC>SlGJA6QA;2w(dHHex73_a#oy2
z#f6w(3iT=?gF$Px5K~I37H}eik1p3PA0SrA`aQN;_|m67!rn#0e3AqxsMnHUFD7sz
zau0(WdbGPe{?D^sz~?^v2_FB3H*l-p|6{tl=b4*}_UKX7#1si&*{F<aRt%waqn(Ck
z$7YW+imMt)>+nhS-j_eCcwCl|?O!78b$R=F&*ZxA{t#oO=8>XD&;#8(ZhkTJO_eZw
z2Kdb8u&*x`A1?k{Dl}k)_P%2obW%J6Z0=aOW<SQqs-TJ4zVnzwIri+^(n~<C!Kud@
z8|3mvg1aFpcOxtUsc%vnhzn)<1lqv6Hfq%@vwci0XqqN%Iis}F4A);&d?c{!z<~3D
z=0T@hOoB}l8075i3E#Tpy<Gmz_p#{a$uo!Zu(&XX?{>NP!WWV(UC#--cc7Vq7;J<i
zh*PYZsJ<oRBu<%LS)PWm-gv!!eM37<dUD+moQ<BbPK%>uvLpJW22Bx0=|)amCbn$W
zEv&3lvyG#0Q*~<NqLd<5i*X^y4;=@G(~PXY3V8;39xl7^6VM^7MoZPCy3~UtLG32b
zeQ*mNWOTdGTG)<{&sd)OKu$HoQx%o5q_1)2Nx+^PW5e53k#v#Qzw7U~?79`+{{D|~
z?Mg<^E4g==#GpQ>ouqi@$E0V+7Ntp(TSrS8(hRq($Yj<*EfES?bR8muQA_u2hZu**
zfQ4*@Kf3Md+<W_O?!NOlZe!<Zc_sHYz$zmvqVDf7d<JB~Vp~nSLgG}}ccN#JwF%2h
zhj{W69>>087D;SK7_Ax{hBv(P629^6?_o@vL7w5&g^oMHSzJ_0F2Ya=O7J3qnk*@S
zAt8V*=c2ccf%a*2grys<<rhyqnM*(QA-3p%Iorb60s<P40usj`H=wh+MCYg9<-?D8
z1liIu4}SFp3{Jl-gS}hnSi!0yJ|-?jLQ_@|bx6RE9XP3R;!LTKq@@Ou#f!nkYtrH`
z#dGB)668&IUQFmjC##P1Cs*)>Cq0G#ar@h{@2uO>&+~FjC!}?o$B`eOi81ug{YY-Y
z4C%e9dbiE7pLMewUEDQRm<`yJV^6yc*L~}oY-?}f&=prPw|xh)I2V?6CCb6IF#C36
zcW$Tuv;DM10>Ms(8&?5^k+L-`6RVq(J#4GyK%=$9+1Lxy6>M}`6t$r{EIQ-zruE`~
z^UHL~;#jdh6Pf%Kvi@t(7<i};s9T>qG6LbwNY3e<1-|w6H*w6a-OS(X*BES@!%B;V
zP6{hsKKh<F@VgIrJUe%v$o@mC7-O(j@VL0?RMSTTrJSBjO@PJ~=V#NVuM+iihGO|q
z6#ty5)3F}%eq8hF6+f*Nw8y_;O~{w7`^jm3-MFHatmiW~eDkHYW=s|d9nrwiA<w5?
zlJ+G`LQs;#FxRo%{Vw-l|FzfhAOHDHy156jXd#nX^`)qmw$+EVRl!KW?tA0ORe0I!
z-@;w)dPgq(!vE!Pw~w*&<@kElvXdb-C*8_;Rc(m1kaUe)SqDjLjxYbmH~5VQJ_utH
zlKDlvj~e{mhu7hQ$AC7CkTiZ}O{S_Mi(J4?rOG!NX%N<=IaSB=LappevC3V*d2C|X
zzI_|nva-|@u3K8cw>y~BfE)S#5#(c&Lwmh56_W%V-UurOyu(Fq$cg)Q@x({{0dCM^
zvE9MBoPNK@fx`pNf7wNB+qI8w)+I?(979DQ9SWI8Tv<%CDt?=ahHDTBL||)iKF67i
z_MGMLwO8`6`~DWMeZ})3TVXC~hi;8moL8((@d7E3FAYetRStgqpZL&gF5+H)bPmVg
z|M%$b+k;)O*!Vt)9e^qbKr8$M9V^k)4fLuzQSoPAUXqqvpz=4dTj5bGg6sBa=Z5S4
z{fk`k>eq43qo2T@yWNRy>Pb=))z8dkdT=Ag&NpG*`X&m4RXeKs9pI$RDX@0JptUV^
z7*UwwH{S8nFI>tN2i`(!`*s%2IGrp_Fe;&<Vih|W@SW$sfbM5M&-}n4J_`wgF@fAC
zCSYTuTGI3|MP^8FbSyT}QMtDb#38Cp0n=J`arF?ys~c`sBb#Ws3>P}$#!Qd1MiL3t
z(Qv3TlA{@cR@JC<%}7xz8$<AG29J+8Y=cG*4r2rrhpLh~L9-mvHs3cHzyHQ}(L4PV
z?9O0XB+jwi^IZ1R!`$ytPh;B&XRxxo67adkpkx^5;-(o%&*G_jF*MwJZID9L>W`D?
z&<v%4%dr^VwyF|Sxk;*y1b&t5OSR%rUGuAEY*;Fra6x<>yC#`1-<0_5&=~frXvO$T
zYh9WmP$^pfWqeYczv`H8L5+ymbgoV3ht}`w1ki_;_pCW<E7u5N3_rbA8l|e^5ur{`
zyJW2Fprg|%!i)@}j#P7A^8BZA@4NmAOWhv#{lkZIxToa4g(SuEUteXV@;_R1@C3~j
z#9>J2_xp776}HW7#hNt85?iS#=QNghD$n=$?<E#&5)qsW60p_^&U>6!j172rslbJ-
z)**}4bxX#2YC4us=vdK3hmUMjTGNJ5UB(7F_Pfhpd3@wETlLKMmbm+tIqq=m>0Ea3
z5TCf_CtNe%AvcCpl8{$m>K(NP`d(urt0Ji2H7YP+5TDaY6Rtma6`%X)`#A0R-6STV
z(@M$RDw@o5|A#%1FMs`Vuq{lYh#EYK^MW@%0Iej5J$S1XA&No;OR-`w7{r6MhCI)3
zWV9D8*ZkxsT=4v-ao)opM7Ot0tJSU~H5N5eGC`819<p+nufOqi{O6@#=Z~NM47lwt
zpnJDtZD=HT9m3tl_MCbSIxN}M22g8)AA#;6F=5aswHuRejs6locuIutG|1r`ams(b
z{}Mj&5C6bZF1Ubv@3Gh&9mFVBg2$7JY{HLjoa4tLa;Av%Zi+8AH`P_g3%+D?{MX9n
zCV2r2eHm0hjp5kS&tNI*(-Fnpa6RteQc1WAl7hlOl#_n>4lJt+6$&%d7cX?h!)LGe
zao5RRfRSWj7!XbeHcj+P)2V-*@oaVG@0=;!ZVITT{;VUL;phfGJz3L=sJ4z&#S|A>
z-m2t2sCi_ml39<%b9$%4hoANYI@euC|Axb{j+QXjHk@(NZr=3L=dkp%%h}RN$@8rI
zu2Ra;1fwD~6<Naxgxy+5BIBF=<i3thAXUpzDuKvMqc0nr4pTGW5i*TTkOmre=e@D&
zLq}<IKPsjxvc_7t)&fy@BUp0RXV>=a<lP=*E+i8sL|<AMTa6?t8Ux<pcxm9a8o}BW
z=REDsJiE5-MI^ziU=qvFlcb`;rj&`CRz)BxIGY+zzzB{&%r;To$EYEZ1iWA)G(^>4
zj15?Md@Ui6kz{b~l|*W3qmFO~#$GMc`o2;^WkoGkEU0CWEb_UfRo?LJZ}5?8f69Tm
z6x&Kk#86;r6kqbAhRSd{q83-voh}PH#cQpJ9S#ZIp+g6*;m;rS5Y9U7M2vT|(gd#=
z-M;WQANf~4ed#wz7q^m;;o>^yy&^_Sr+d+$UPUC%2kwi(h`~sLcNUcdwL&)^;1W;L
za$I-$_xaGf-o)b{@<4L!G1u-4{Vz;{nrOt@SJE4L$YtN-U!L}CZn*j|4}Hb!;Ow*E
zxa}Cju(QAjM62>cwcAj23)iZ>!YJuZDmzK+q>U~W5{(R<&(Ont<lr)&KK~Vb^WQJ!
zY5(iBWPA5uch7+VBMEutLQJt~^?YL;m)umz!cCem#KmVbVVH6MokaQy%Vof8COdXw
zw=9zToJ3)H|FuZwO2VypK@wr^c4yK}q}*IpeKc<iz76#XuC7%poINn9!DMA<L^mY{
zZ~9~rHpF9?1UHe|mN(_3XO72Y%C$%%{$JMU+l*bWG=WGgkRTG;;wk~mEn;y1ZG%dN
zUs~qGe$EHaeH4q`7QOvFu!1!y?KJ0i?tD7Wd+a$JzV>HqU)+Ls9wWAND;2NUC|y{;
zng;MQ<G!|Q1%lJ%Xm#qm8ZnEQJ-U^n^t{Rpb3~gP7L(mdj#iR%<nPm?_Ia|-51;gW
zGU5&$Dy3c>UQ^!r*B{~#DVO~HKXF4hBX^4P(WREEv+}rNS-{qpX$5)if=ZEhc$ded
z)<^%E*d_43)~%94W~ixTJ~pKzj8m%;uSHThbWs-#=4X_aan-=T{=YW3%4j8Pm5Sn2
zHQ}ix2Qo5QCp>@}9kMlYLhuo0C6?v(HV!V%)9tkJl0Xj<Z_&(!3O%-j?5WnzsmlMT
zQn`!Hct{W<k3plKkWr35ejm>{?{TawA7*i35o<jC)jroB%y`mYoX^gkd+2pnNvs8%
z;xtIj;=%d?g{0WXXGqnM<d8T>y~jGRiQrSus`jz-2FOVIJ-+jW&vMT@o=wXrmIT$%
z`LHG}rK){5Bk5kxfscHEzx|7+a?)*Y$K9U$B34d5o@CD!Y*6tneGMi*N;=%*3l9u&
zQC8iDeW3d6XpnBzN)afa+!QeQ=-%xCH5Sbs=>Y5xzt1Jld<t{(DSz_H3s~K~gVydX
z7_TI1Q0K9UjohHk{=zLWX1>Ws)q8(2@X*bvu-=V(2(Ku0vYp4QE^)<|zQCUL77jVs
ze#U9E_Ur)L3LB{zVD=y6Q;&WECtQ0S(p|-SAIPyvNx;))&6_DxM`t<d8U)3wZp`gh
zr?Ujs40x296WL@s!a5*Hwb;F?VAOOHI#O<ru!o`&S=^Tqd~y>DL{UUMK@h8js^Cj3
zw?;x+oGY4R#i4x<ShB5iT(|24e&?dM(%YgWJGYUjkh?A<bG-DT5Alw_`)9Tue>~S6
zI7Fg`v}JL59-YSvqG&V%R5xsTR|1^kJWX^s@)aeS0uL=CC3r=Lkfh>ulORbo$W9Xv
zVg^W3HOedRyKg3xOn@W}tIEDHI~WE;<FJ(K$uvTeh)^aOO|+sErJn_oRJ{+C`${u|
z;m9CK6ERAaxN`*dI8h~`dG%49qG07csj;XZ(337`N-LcscTqxAqT*a3AFWbKNa`rA
zuyAxXVQ#1cv2hl~&kDwuzMLyEvQ|eo#fBl#7)*|?NkfQTF`I0JjW{uQiNU1TntQxV
zZf7+uRA|HMS9P2w!sy1ETn)_qN}y;;2k#)=?Novbz0(qcRm7-bY$@qoe73=ih^9*A
z`L5(GCK{Al6BU{q?UrHrz%_jNGaqBmmKNLRJE6Ot^{`2Y`#<=peC7LB(s%u6Tow}#
z1Q8$b`^8w9m<0Kt=41*YF$R%@%;t18V}3DZ>AEYq^O>jf5AS^kes!6J#SVhSITvA_
zhC~AQo`Zdo{sF%F@|SYekFMh0kA4K{9qvqb&lb`&iEyXFr(!siMc1II{aSf)OX{+(
zxKG#Ar=%+19Tj5Z9CZrTLaqkCx=O1rT>Y6(@zJ-vllwmSLF~K7Z!y?DN3xLO)RPn>
z2+^2`Y;a%SJk~niV~ly&O?d+5roICra`EOo*hr=~grjKGfHq=bev!`3y>#=OdGU0A
z{3FDBym)d^z~bf?xz(M1l~oO}5}PLAwd632CI!vgjyCo|X30?Hs8mKb=Ib!}k|XlX
z9hFsW2zVxkZ#wY>qoz-d9P1hdB04`4##tnqQz;mABnl=E8=+#c?UckOq=#47wf_n}
z|KjuL^c=lIUA!ns(kAPz@chR=fM-7DPq_A{Kcv;};1Z#qJG?PLTG0o6az~W#WrPE*
z)J@xTgzuo$kg9>_piLJ%-ocotTdLDLRZ~jU8=Vmwo4PW?M^A*i{d_m(^*O$;Gp%nJ
z+L0$>>J^t*_L!Uwa)%?OWp~g@7IA&WxaeH&72_+)z~YBSRw||2df^srV$g_!u`+_{
zh)|Y-v^%C;IYFYj;7!Gjsjo~Bqr>wkVI?|rqZZZB6&9V)t<$m6!f5iSW73*qoAb~(
zP1cwNWr8r$<B0CnMYpAps4bl9#YipTE_@8iUWnEco1yBIl(1G~Vy8ttMik#)=7PU`
z9>?w4%0gN|v^;~{^RB=BD4)FaKS0_bU`%jl)*^XS#Y=`-kGDbM6d&dRUhoPk31V9~
z9FCmD#e@St{yu;9`}gBh?|TPyS6P^Eg9qn4)&$QCBan9;$#R$EKR?5VAM*&*B;4l}
zuRwqOUS#{WGnXbLQJG17W$G00A!*u4WxOM)kw~vWRjHUyLE)+3EcoDd?zI4gh0nJk
z!E=XJX^HUFcf5_idD9zt_%oiuzTdtdy**n<7g9t+_oe74G}-8F9=EJm`DT#@dhb64
z+;ekEY_!u59Wo_WoK}1M<nv!(?_9#bWz3&`Cg!+f$R#0Bp_lb=U;a8@c*>JF&IuO5
zN}<3Iohh|ixmruERA)WeVf75AR77X^D4jhNnb<+7b{OL3J!%l7WfkAdevOq5L*v%1
zT8Cp*$gEA|)oGVnCy)oyp5P1qcG!T8fjm|-jh3xz%`%L%IFLHF-uc(~r5C=K)%k?h
z+ycgVGB?264sZI~PxGp`eUQaHdpUS`nN~YRG{>q7Mt{bl>aYmvG<;vBqF>yLv3pPn
zR<s5RsdQQzI}B3ezevrCc2t6hTK7h#M(888c5Uh~R7w2xV22qx3}q)|v_{y}VW<Xf
zHzw)1N~EmJRxR&&%~7z@ff(s9G~UZ~bQmT~co}&fr80-cJS>WeanU)*HNbbQDLY_+
zY^r6oOr>rDR#TQ4dBU_NmJ{cKF~e}S%4643M`3DKSgp91{E3*Hcob(tw=2~U0HOwu
z!Al_Knn+M(0*RT%u7i&D#-=T_4IepK_e`Tpvs9FehPwVHbnYrjKUzW?{TTgzRQ)Fm
z>(H=Hy~idd6vZ?n!Exu?oyp(5`_*LqF7xd<yvtbWuCO$)-0pw=CR=ZHDu<U=5j@6f
zX;Nk-D8d;NJ3EqutMdjMFVZ8JG$rc~upLX%&e?z2H+cVh-pT{+c2``#%UrvSQy)4<
zfhb7k!K`#?Egj_B@3@%%`s641?T0^-g}dE_?g_WTra>u5eW=m<_^cZd8di%ncHwHz
zc6eXQM4=E}jfr!Ze~JrVNbWu6@PKr+%g0`PKHZf84}9@6VA~G*+ZF=%B%0OL*N))~
zHk%!Wqka5~G3GaJ>XR@xhdUu6@7|mb8>{sdDnBZr(kuq(wApduiFETmEn``^<{G4%
zVY1NDPON3|l#?)XbGS;O!dhFZcr+XStQC&0ql$YZSH6y9p)qV~au1#GnWONLCcm2!
zJ<zb3PVU2uR>3FtaAb{6lQML>En+)U1eZoaU}vMttrOG<&P6pD@7SFi4*uKc`1*yf
zVZPU4u#|xq(zK2E9)JGed+?+`{vDRDzLM>8TgfuTdxOt|<G9Tf8~HGWcf-i@T`7p^
z3^PdToOzw4{NJGY-$+xKRI47D3wrdP(KUZNt-~Npz>Ob;XR7g*$`M}O^q6xd&xk49
z=IP~$YH%MVP7R0^T1nK%PXs4E;_Q7?i;0fu720sQrb<I_T%E8*_7KXf!zM{(?heVS
z+IZ`O&c9;3A12GuUnN0EypY6BM^OZ@-be2OsqqxX6N71Ya-1hKdvk5Lap;7RC=C5H
z$|r^Dw&I0o90%&Fr0a&x;m}nstac_A#}JH%mQAoCsOBtmTDa~C?|b*FSy^78-Rc04
z>wqnb+j-ERK9>1i$8c!rAf!r?CL!s}gA}V#un6P=4%7vrycdr%K$;?{(DhyH7I3<X
z9J-$Gef8h@?K|B8=LfXfZ9IZVLSjNUCF|vM23=a;`ZAw7@3CC>op1367yd8IZ{CN&
zDYqg?p+#_|mSXHpL(r|0O@Or0o>P(FsSvwXUW7dE>lPsaj4$%_=<F;Zk#;Vp-OovX
zcp2|{!nthSwTB13@@4e4A49%<fmSp_EDH~&x_^Z+H6m<UwB0ge<eN*v@E+iso6}<B
z&7rc|s0KB{mfM`jVej$AGB|XQ!L`?5i@J<SVPOl~PCkV^ZIQ<^ffm(qhFH|8`YAFC
zmV%=$$1EHCt4#i3dh8+-f~?9vslh3vBxbLfG)V)0TgDZW3hty#2&P0v16`@ejSdxC
zf+AK?V=|~!@~qDu2S5A!5Av-GUdnuylXX||0-dx&maX!XKf4d--1}E>2d`!O)<tp;
zVp5W%9THWw5d&+=?=B33%8jgI4_&Q<%UXeK4Yxl5+R>!NcqD=J4Dh726XF?9Kxa51
zIr`6+=sMo78+pGT^@4#iSxlh%TOY^YM||}l%TS}_c@`_Tk<2Vw2gHriUvxq(w=o%w
zibXmpd6n@cM2F^cc(7_Q7>t@=mKBN3wsaaUIy+j)zoj&a8!J=jq>0_kQPajuPd2pP
zRxRGNVW}0Z_YkVo^HKA*vYu116}L9Xf4e+qq21x|wO8=TkGzL$^$?45t(Zs*KsvnS
zuinhJf3%;RG^!!xIG2TFO@@UH5rat*#3YCc*4RMa7GaS07`Q&IHY{ELGk)>JJ$&a2
z|IFU`goRFr)TA*{4S`jjgFo1#v$VpM7rljle%|@q?!W&kcX{>;>7IHDa;q&Up|jw<
zk7jd2-j`RcC)=V^?{akmkdZMiiZ<d4CB&G3#<4MC(bbIBvg4|cev((8dmg{_fIs5b
zp73~j+uEet+QA1Tngyzh&fF@od*Oq$H}&rN=5bVxZ^jsTGhwJC4e#2V7aNanO7(~c
zLGg)U?-{3aFk7LO=5(*V0<tbnJx&aRw8a^B`E_pagP`&4bG*wN)EKm0;Hd36I@>lP
zGZZ$DS`BwS(wJJ0H)r{hNwD8qm!aI*Cy<`&oAR)UArouTwe3LaL9~?IS7oSq5i4;O
zl3S?E6*whD6G0P$Bq^yeoM<gqe(b|s{`R-h?i;dRAS|}hHm<+K1<!ssfBJyCbKvS9
zGrx5Uxq*J}AWg~RIp)rl@R-_-Dm5Fm;U-FR3Av^`tLutOk4}Ip>kr0n6acTX_C#9$
zIZh<x&p<k!Snxbbw{hCl>V%NDsMZJL^AKPgD`=wl1iUerBn_^_h1fvhW?Rg%cuiia
zOI#K5=7&l`TG1gAiT?_S6BXqkEn!d9S?>X7z#GVo;#-znL7l;B3-L)Hp-PUoiW5Pt
zjpW?oybFaQosbyROG?%8)oQ@5I`Xf&9_h?uTpH^Y(w+?YAVV%hh3moa{!3G9PXgQv
zBe63Py9+uL0!VC&)#Waa{nJ0=tW$1HJ86;F6veT!n)A7@UdGE_cQKv0ZR9Q|u_-};
z)uEap#^J=#$B_wSYVc}kB^|Q9r`>Af^I!y+wv>ZcUcu8J{c!&2{ck7reHP}oP?#<o
zghYhFz>^FH*sFiQXaC|nzVX@5a_^`91;_sBxh$Q2Yg*eDXsOa>NR}^CrfQ+djP4C<
zsS*xTu7Z;xUa=ZT^Qw7Vn*(gs84aAYl^$uY%U9m=2LAEA7xSdoT)>{+z6aScTWNPv
zY*Yjb=cXiyMV&*{QMyY@A-J*0WVofq#>vg-zPuZF8gT696xrZ*DnrLi6+?U9t#FtB
z1S5TXe}(*-tFgx&PcA8HVawTPF-TKfmIb1f*s{evj?S**#7p0XDX@#l9>Br`i_-OF
zNR97bkL|3?a0)V8rL^W3HdL>XvU{TwI}AgoVl68)GhdU878I+X45?tQ<6jCV+^F;J
zBy3g#Q8drlXXp9Ko8Lrd%MOlx*n{W|`m|eZ+N}k8gC0+P+ym$eyzN8(!nWNzSvtHD
z6iSRiU6kb&hDsSD4x=&#%+Q&N(w><>6IA_c*EhMQPO1&&2IDw>T}M7sQ;!mD#4<Ii
zJspV*YsJ8Dt!tCYx-(LmK6+J<+Eq<Hudzo;Lsi&0H&li3K2(qis$ybd<-5kG83wVT
z`%szB1p~DrMe1er3`gDW>)hx1*9@!cE@Hw9BCg9EE#uI{k%%W(AIP@Gf-wwSM$#ml
zKdSRqatE}8ES3zLxGsn>CANL|5juRaOm*ff-($vz)KW<%#$~~V2HRRw&@0v_?*nwq
z5D4ex-XlX*IZ%x6a+?$PaQ-tMN59u)OJ@#G&T6+uZsvL9xzA$PzLQv5T?y{eMw>x%
z8vp<x07*naRB)L`jKs?mj946skC0tIa5xDa*?ykWYA0m<gV^pepM2ljx#L-<A+F1O
zXPy{}m!sc1yh;bpbszm0pMLAxxbwY#kK-S7AKadOxUFsGDWGUlQMs!7>Xbf9Ivf{K
znb3~ahU9!mwh;sBg4As2d>g!XIAd^mPO_?)>-Y1|7hH%WmPcOjTJqhCuxpW`U#vd(
zWSHpL6h1e?@k(*uQ9^^T>GpiY$JH@5zF8bMrw+q)z&khR$C2!?aUs^Q$a=LL|BJuG
z;nj?mX|sIQ)yQ%e$ucAnIN=0ZC)}F6NCm7^iooOU1forK@96=LPAU%`N%?oGFq5uX
znLL^~+UN&YtKm9JhoK2zF>6J(Dz_)qkt#6LQdz(2Frc6x@4*oQ5P@72^<8eeb)HKv
zx{&KW@-aHSHvNHuF(gR~%~yHR6CS{$e)sMy?Z29BTegxrA7B*5hUCZ%6=TYcZY<EP
zY$noV<~Cj*ZK|-zbg{<`h39KR)sDa%PS!CeYtpG>Z;;8E>k(%<li*ylYKqB>cVYH>
zSM8#XHvCGifz-Qj0b^}KRn!x8`eTT0Dj0kdarKooOB9S3#5e>Og85MDMtxB|kNkr|
zvp;t9q{@ei23kWLem$BuQ4c@7POvI?=dh{AYmZZo+sDrNIZ_`g#?FOqgQh0J3w5N`
zGg56Xs{9flh(5Z(8={|zSHye72lc3+sH0_2Xrxm87vI>#kXoQyffmWoQFx5P6a!&V
zEeCH1zzX}2!IaQ3V}^f|#NeWjMw%ofS(lIh?YrpqR@u^-1HcV(lFlN3_>f1jGEi1l
zGqB+tUFNVh2|zEChWp_m(SYyAD``8$8xJjmwVp%!ui}56c?#eE!oP8++nq|MWk_uc
zg9QRNz&Via3Z4Bw=JU^bDxZ7ryLsR<p3d<Pc?kIlw}vfkS`hEK217VQ4fi@UHZ`c2
z`idcwSH)_1uDqLGBit(N1Gx*yf)mBt6hClua)W;T8(i|#r*q;dr*ZG+{w1r&?#Axg
zf<dvWAz7dzg&X0lq6tEZs3DJrf)xL5oE`W7kH<S>Y`mE;OlZCIH|NL^qzvMRfrhov
znPbQCr?Aq`k*v?*pI%O?8KC6z^PKwYzs`zLyh>Td)zR`46G%+5Hn>Qp#y1<pnai{o
zW<BAF4lhGy3}q(q8Aig_+8xvNhp+2-m5#-eQdaEhPRp^~9XAnUBsj(v8$hc?W;}!b
z3MWg#C!YIkuK(X3rPIyH2LmvMxmJgKu)_16^gtf@fP1rY-IeUzu}E)_k>?t0gJc*$
z_!4L!_4BKCcyyMSRHszbwARs@luWFvq&;nZ&GNGKjOSFRXPvDpa3nQ`rVL!y`9-px
zyD{V48g)djC3ezO!I(JyQt67y$Qz2_zj$4l1fUp|Qt`(|WhFx(Ba>HZ!qrAgRfklR
zv1&`jeYfBwm{{fa;z=wF23?-~_($`BzrKjSefQhw7>jW(LUDvFa{>M_l0xZdS}ns8
zMkpkO>j^TUvAS+eF!=(!cTv7I1nBCu&N&brs|M*>%d%98P8Ghuv~F0zw+j9ThTM6q
zmDqjph<D7jg&X$&n7@A8Ye>z2t)2P!-W=QN@Tv>n!MA?6A2li7<XB^(00cpTx;#r0
zREbOx>?&E-hs0po%JTl7aKUrV<saYk2HNykoNr?#MPssn1HYPMZ#YEn6CdXzfBqM=
z=y3m+z5@TtcVKYhHnbJE63!2+@TGJvGzgMLUDYZ3uZ*q@l`dq_Q3~jzu(6w99)eMe
zJG4T(=lI@R-oXc6`5J!jArIlU=bS@-dx!S6lspCl-jA0EJ&FTzT)Fld9VYK`#JcI5
zM?CsBH>m#2En)Z`@Q%%SvhlsYaBmFy)FF7AJNYE~X$z-0`O*!@p~Kj`k8i<=cl=ed
zq>YNf=V5~kQf@vZA6liZO$cVDzy!v`hjfzR(mJa8M_RbhwK@cqE0`uUqSGs^qnq~V
zYsV_3>-4lJPq9h5QHCK?sVY^u?}fg3A<1ietcE5DNP@{S&f3!9GZ+3PE1&%ot<{V{
zKM(3jtq$4h5-&LS4|vrx&*iEgf0r#=7s*A*oeNd0m~fU|mZGBASP_@H_r%PVbD0u~
z>MR2`IRjgF^?k;V7F}l?U1ip9G)ZZm{L)BOn30z>@!WN$Z$I*tu6d3$`YP&_L`)e=
zDl{0FL>WpUS4BuM-ax*J(H>d8g0YJ8;dL0TnKf#~OQflLHVnQa_oBj~@RHF^9C!Th
zx2Jpk5_`5Sa+{M*2%Wsd;PKd&1&J`G$_i37?iwZ%K1hg)$D2M_7tF-clw57}7631b
zZ5cE%bk*UaZ$i=i(Aqp44`Q`xomMH;qsms0juxCMiA|AA86hJFYw%97CIJtcd#n*s
z4B65_Uh&+&;Lg8vdy-_hKOOGp{OcG0lk;Ey4s3fnS?-Y}BxY8?_#j8C;({}{A#@wm
zVo2~tNVm_?^<5Ti#+Ux(gFNQJ_ahth=yW<1Yy|`O99ZdMmagPGuX`o`_{O*MD-S$}
zJ3RM%R?oOCcGq@NqGW2tgmE}AD+yh}cPYoO`H7KgSEpjyH)&t>1s}x<xf9${mv+9&
zC;#$!eB=Lqg+F=u3z)yhJsIrTO1mBS3f2qahg=y`_JF{Mf;g~{TZfw9{bR&qa6>uW
z4dIP9kIM0m7!z+M3{#SZO<2+q1Og=%zLcL!K-wI4yW7)K!K!EFnyax{Sd8V^W9aNT
zmc9mM2=DV?Vk4$OBtshDB{KvEvR2z+_76O&O->n)lVnots8f=bF;SAbo2lx^hfS!)
zt_NH*`bw$AlvNG+;op=(c_ZK^f>DnzlL_M?cOIV@TmnAJIL)^C@YA2h4Ilp}=3P!d
z&+u5<?FF)|&u`!LEZ*~)m(sodN_K8tq(5+>l^*Ahhz&4+=nAb<7`JPw<!&s6SWhXw
z35UD!;JEo4WU}=@qd0)BYgXv=Z(if%bV?Ge(~~h-`+Sb3*_=+fN^Q8WDx)cVJ>ZqG
z&+s8AQpJiT&0*fHa577rk@VPVGrVRP!>|%xae1za4am4PNDYP}mEcjL0bHIbpZn66
z*td5J%PToQxcqX`)Z(2(6})q?ks%Y4p?ViXBVIqivWyX2mXRjVYI!U<;BmR4-eOEz
z0~$4snym4#jv4V)k{TU(;8n_CIk-;vAj3+8?g+8#5$fChe83(4+u1zv5r2fsR%jUu
zc$Qa|>A5x!ee^Tgef%jb^}CR0Am@q(gUJea#QP9X8)Nb6gZyX;*Ij!hcfZ5gT>4KR
z;M9HFXs4FBR;RXv_j@^JwM+iBFY=G)K9%Kb5A(p6znoq7y+8fETWQT%3U58{wZfGt
zJUWK3Ekq+YZD__fETecv9gzy@-Lj)p;nrwe!@Q5;7O`=X_Xeb0Px6y1c<Un{$F}Xe
zdEiT5#Neb8a63ArX@G(S5w77_jjli@_Etm<1{)D&#p}nEQ8q)fTTm=~vrPZI_wNB7
zvN>hevw`Z^W*LZRg&v1+_}kxtL)Vjg$HJ{n!0bB?Gw|$x>)ZI&CGTgSPq4;eQcGTu
z?o%l=z~feGDvN`TbQ7RXx~-ZS=}3LO!2oOK>zx=~uBFsYt;#NA70Qw2u`v8HX8e0p
z$If3XMq!l)f28^|QhBMk-)P0%v{W`4d;gTGx=u+ds^bv@3PU8`nV?(m!|oZN3fUk)
zF&5^yEX%m-b6&{q-@FgK7PQ*YJjrL|+T!zH`w8bf`MK=9^_g6C;2>#YNQ#EFSHyTk
zee8yqQmLb61jf*)O;uwMR=K?u*Q}CMT^+~D->P<QC6RpQkX9;Pm<GAl33ILJ=G2Yb
zD#?81e6<pJ$k_U)_4gEx+3IWNbyNzq>en`I8=jw*9GB7{Ph_m>t|O*y62rBjNSd!#
zil^{mb#z_Ob+igq8y{`OkHoJ{PP27!1)WOHu2_>xgz$KSSfSJJanj&={^Zn?*>mg(
zyynuc@B`anVAJ>@N!<`n>MNqoBEu#DcOIPD9W61Oeb$+L`&-}Ra4$n8!5bT}`__UR
zL|Bn6J6ARPxnjQgVlG+47x%-HCPKfzOqQ*%xG<0NmfR$GX@f$deq?pgk`%uIAMcT|
zm;mjoxL;TOF2n?BS)cbwtOe)!?w3D|^j2ADcd%mUWy?r9&u{(l<N3~y57W;BWF`^C
zyDU5-HkzYZ32IbHFufZLa@w61Vw9DG*YT?JpT+<4fP3S!RXS;#z%~ij?J}^rgBz~m
zTkm`a-~Zy5x$}MQ$L`;}AFF$gAziT1gsqym1vpskIWOl+kwlrXyA?V=5x!KpN7Yry
zjI3?Gm~mba8#o8PJ0Q&+KmE{$`G@y@fZu)SIUIM-d(q#qNYb&`Xcku^R_o|$CzWJJ
z@o~;;rJEIU2U~Ne-aIyRTw;uQ=*{h<%+2vmiO41_=}2<F3d5|B{BxpgJ?*x1y&*N8
z{O|$X(hANvj=j^JSZXCmJ0Zc44+do`vM?knrSY1)1!PhsaR&VLddrGhPZrjLZdyBJ
z)P^*y`5bCrTu7r%G)vVf->TCx8fFoptrH5H`xG)ZKMGxU4fQ<lB0f7sG{t#?w+WIc
zzI&KcRrr@@K9j31em9*#M%L@2A|xiIB|YwR)~$Ho>(6KPnjf)u=Qgs;<2C5Xht8CZ
z`GHj6Q1wz|O{JxpaOtG*ckTUX{XB-Xa6fXyGOl_wp79NTw$66_()@JDIzL{Sh2Xt0
zRVQg2w-NuCk(^Ja5{rh(!VvrKFzT@6>}j_+d*?Ci$%W(Q<}p6U7^#q{rD0x9@w3rt
z)jgDUBeejh(0A~Kul^TH4oKSgsMGJ&;nhczY143vDnV#bi54PHA;?~~(gc^U@Q3%m
zH(&Vl$9UQEpUu2gTB?W--9b;CuOOv8^BT|0*jh<5JAL%lR}~wIghvG<2BV5|0~R_R
zR`y@TCqMX3B<nNZN`phQ&%rM6oEN{IuYUK6p!y~bG$(g?aN~{~E+RqKBnIy?@@&B3
zmKK`bz%lKduYKYJJnX)Aqs4%Qw1Z%yp&R&>0rv1R`M>=OpM2_h9RBgO{O+?}%<kX&
zL;5G}!!87<Qo&UyiW#Qj#Zj!eLvh{W_Gk%bE2MZ;d`<GV+Jz}ML#f|m@sVgAmGC&&
zB|Wsv=U@6NK6S|lc<eKu&vAFVC)wVu*iM92`Kl(HRthh}Aui6z%TU6dal5_E;}#qP
z-&_-he+B-2bK0!8<Q^-BTB+iQ2{=>&Ol0rrr_s#^^p+1H{S`zrIQb;zPdbIZ8!*s5
zCQV{>K~Q7LbzUk+`w7OzV;i}QMh!<ALP=d-Q#R;QJDWH46yvo~_=wavnhoa$(*^J+
z|EUIHs5rz;vV<5qyh+`$dlU*L0iq-kgDqu8Ds&vw07-bS?f3bGg$`eP%j>xElK0T=
z34?AHs-iZfZ8Lu9jAQwSi(kv3pIyc=yLU!k29L2V^0>)3EjArjn^Vo|N28>^Nki6j
zUuUg<%Jfla8beuE2tD=lWkc`V*!DctU7-S-o7B9MSwvx_`B9l3RIKm$pZ=_9t)f5P
zF&r`-gsKa3)b5INQVL%a8;UKx#BgQi`9W^^_LAdgy#WIgl*N3}>ku8zkDBt$1Vy=0
zUQ|l?R=k3jfTh=<Jm&J)?J$aZM`BE=K2&sz>i4_o{`%kw9jnmZIgAk!D>y&k*2nMV
zRnPx(cD02+{++vU&L8|9E2~RUI<pWn*0Rc7Kxc;jGRjP(vI=Bn#J;_DAF7L{GBgqu
zV-WSsC6*hmzLGax_!3S(aR=6W5|g52<bxic{_K}|?K|Gf!qy#lpGVjE!UW9+H)9Oe
zrdVTf-r<ap&LteW?n)kXzkBeNPk)GgJLi~BEtVA3fZJCG4EmU*pYgSeUe4dY`c<5D
z*L!fM=e&@mv(6?zb_WSc8$l{eVKGwcGiuSJlS(0_R2Hi4oPlRyy<ojWj#)_J!ff-1
zy0Biy!jIDc<?<GCW5_ba^b_p=%ifvC$x@Z)|NEX(-90_a-C1N)Q55$W7c?40#ULog
zJ>njhxWp}LViI?wF)mRN!40Ab8q^rnM8&9>VARM@5;qi3WD$_%0(Y6MyQ|K5e}9}>
zPE}XW%)OuyGVMpb_s&dDb=9d;@AE#-`#hKaf=|5c<y?Qw)jaFO6Ii~-o$v>3!U|z{
zzYO$@GJT5wNF%vCzjh+QO)K$2;tkeY5^u47on5#IUT5Sr@Ne8J`At1zB66y#K5l*O
ztoel1l1U}npbGmwXFPV>KFsWU5cAuwVz~PTrn&~E=D5?t9>isrTtL@L@F-8T)G`M9
zaV=?YDK_#6&xYpU;Y5|to{un$cQAFeuHi~0%4!AQ+QY)?F$<H}4_Wnnq^+DBJE2u)
zCzcwO^9qwnLoL=w)s+&>Ga1z7p7J4_$Ak)qR24}qVlA4ExJ_@CZ@uepxz8ZuHZOb*
z3yVw4%uEA@>0ZYE8x7z5{Kt6wb6*AhF1r^Y6&D_AV!|+~Iw-|0Av1(hu0&a+R3hw2
zjw+SLB$g;O6UFv)47awX+J5bgkkT>Xt=d=Oud#0@jiXspM25D15GMcgWT38VIKVX&
zrWy_9#?rppG0C`uGUzNS7@_%WLcU5fNw|_JPX7M)$+XLL-3~e>;DQ{eU>RcWtdJ;G
zpk3Q{phil&7XJA_Fzn2Gd(H*x;zA&!K5_|bV<W3#P&HVIs`Gg7NvvVprcGoSOd+Hr
zN5@L}`$bG(8fvj&@&B&jl1P(j(MlIro)4+zKB&*HPo77i-r?1e0P{Py^Sq}$ny3E$
zLs?i_WM+CAk7H?&vasZM*{k2k!H3>~>vrzOT8nCMo=TDqX_f_-AO)uxI#RmR2^j<I
zXp#T;{At|%u!FH~K)*jjrk*5slp8K)^wcr>@%Q-hd*8`{`yI%W-uXAM-=Pe)Y^0xr
z^K21Zsu(M|OwgAMP>nH_@k|<tzeU=ma+14rm39>(_{`@lm*mOj86JyUSfuZ#_{ABY
z<!h&Zng<^95bki?V;JqTfzE~=&O16L2^QF<0X!v^u2#(@Y2HpXSCdE%DC_wPH@5cu
zjXk#C<P+z;{{ryH^(C|xf<YVF<|@*u>_>|M?_jXR_MiP2O-Ia3Z=}E9ewfRD$yqOY
z5&JA>bkxvE5=4Yd9S)l#OP5okff}!3;9DY<3QBa+41?;#2Bog)^TLgvVesXqEfN=>
zb;wYJ8?9OaMeZLfIzh#=5304!nvZq!vcm6+hnkTv%tu`(U+dOOQyL|;!8SB73aX~;
zT&gI}C?qC?UX4cd6zo)=%Z83a{^0T4?G>+KS2kk9#x5>XjCCw8rwr-y`_Fg<*ZC<1
zro+<WBK;&h?$iZGjnev0Oz^O;^)2HW1{rgBitBa7NB!ce`zeuYj_O*YysMC|y1u{d
z`n9h-^()UX_(;DdGR_@TGYnG8%&KM>WQ^S^W<J`1{J4Bhq7zvAjH<QAe{)}{1(B(*
zcZOA01Vr(>?9mLv1msf9FvvvOL>uQhHYTV)1Y=2&k7REi>x%u%Bbp&%aB9kqwTj8l
zQ-w5TMS#^n2&IHbOJ-N{>uE%AIkPC553p)jL!CymsLruSRP0BEsn;N|F-l*CeErN%
zGuP>oit_t^@Q3{L%Ii=w3r0w6iuioLuYIPK=6WW9>-gTS_BNFwkW9)M3GWREAx${T
zKpj3^=I~qZ&$mDKQFbpbFgrbi5m+7$S(GV`I`+x@a()P!;7wpwyvt=YPB2Lr7#Cwf
z6l-B&=XRd<xZmT0Z~aS_mgkx7&tNNl2Is+!QrNwn^FH=>{N%sB!%>fU0$Ywemcf>-
z*jY;=AkOD0)Y5A0thHyN)}Ir3;<Drbl?FtMgZB)6y!j(0u(pIe5kc|0m+7WMzVoh=
zIsbotz>{A35@vq;-V8TQ)9V|&y7HXYj#G7ATlByG`l2!BQR{$V?~35ZsOo3e*U(LJ
zHbWQ8Yv5NGcX8<tf5`q@H{s_tF>N}0@6~@v_LHA6=b&pW4u><KVzKxzEtUfYozO_A
zbxXBs4#)LGec*Wy%6(l^?|*G9Mby%1ZQz#$2lb(6UA?54P&=&&FqA{qNb9+qq@eNm
zlKH7uMp7GyZw|Ps`e38$L5!`wC;zDogSSE6V92nVq9cbfeXjH)wjFgW_kZ(SSkxiC
z&2w1i5#w1{8ewOq`Qz7pkS~7o`)s|{0qnS87hPk5EWkL#XMxoan|D+l@c?w}JX6!s
zj}19TfFS}#(g4(JDM;hNRV@npiGZQW2jw86W?n8ez)<+^*Je{2YjAbIQ2%ry;b)3#
zyLP}3Hyfn3=Eng;R9ff%763})Lp07~8}T430{%4u25A{C9|P4|D}rjr(O3_XGYm!Y
za{&EPMGO|7+2up1rGHKAeAO;BYc?R_2#IQFR4^KkkZPskDfio|zprTC#F4QnliEyi
zWSUT}4Hr7n{SM6Z4R^lFojCuZA9MBgd9ZylmmtPqHJBb0<GQ*=hHf?Dwpwr1v3g&O
zEClDGBuRo(Qkv4Y4x24-;kjp#E-bKdb{2q9Iw0vx@tjw_m9L-kKa9L#lm~m`eUNi`
zhkAqQz{qDzO?A<9h+W#vr%pYY```O;ge5xkaR`>&C1^QA21AD5`8sEvaw=Q*-H#)l
z|3dt&4`H}v3s!>E3-5F9rd-;lg_Mh^5k<}UsEWAO2p;4tQh9ZAd-SOFXYjfCh$Ek7
z1`$#`9b<6m2(yqPS6|C#-tZ=7dNUmJ>X)MXA3(NsE*y&wA%v=g{EZsPvj&H__1CYO
zU3jdBd~O{u>^;lyW#H)brF0Wq>5>r+`p1PUuVCkum$Ui61DWZ~u>AQi@n0YQI}UJ)
zbs0hAC^$7(42Z{vXyDZ=V{Lp(Y{C+EGWJ9Ut}6{F?>WP;7Qo=kWY`ISK^S+FSTkUt
z;Yl6=6p`BgJqa-ADlSUpd$e>`i%g7=xC)2~4p@?fLc?3|9n=~e4sdijAw|_XgBoRb
zuS;^D-{JvpejD@MDP}g#K$d}l#laBMo#C|4ev{Xqa0;6ayfr(o+f8R~8u2MM9fp-8
z24@^Lms*4ls;`Qi;`8O_tp*s%fFq%qmRi#ek2zvZ0t_+Jz84IG>Nt~gV*x{St=o>L
zDBgXZ%*$$kfojZLCol~0Tu3SRDhu`c7;{t#t?9noW0$0scPN13S^-0Yn5*%ywJR_2
zu)S2GqWs)?N2$%wLXj?}0T$|W6pbCF#_o~lAB3c3t%aObfS+19S$xrG>?$<5qoAeR
zT3{xO8mx!Rjj)Mferb`JxjuM<_ZCe8fKRlT&M+(TajRdqx{F-ZEQyd!ac_7f>2^^U
zWQ$!REbX|0@BaIj*>}#eVWuCXVgP_Zf4|PL>^=YRsW0*;Z#sc}_C1*8Yzei1w_(lt
zP!wlA!zYGL*RpW!)jaOl-{sVIzM0X|GIP_t^7APmsYiwmt~#G{Px>I&UUUiff5P!>
ze)w@LZrMPuZ-Y66g3I@vLIx_Olr5iC$c;lvydWASn*|0};*1I^N2U5hl)6e#t8o{#
zIh<G?kYviPZ~Ye^f8PhV&jWsk`#$>*S)Q4rvms<FgsoCB)-?9Rb#~!5<n?dHm`D7k
zmyy5OeR0*ttLkUg*V39X!HQp~9+V+qW%~sekQmr7+s9sc9bbFR8`yv6PI@DUBuVbt
zVe`U{956`q+}4SuLalpIt3h<!a&@f?!=&PO4Ueo90Ym=a)$x8{yCkVmpw#x;y^<v5
z#bQi>tRzXQfhcYNvW_2{#`drI--{x>pvW<4c&_;bQH$4Ly^8T6q(>Da2A?{NB<$)c
zddI_f=--~mLcdS4c^Z=%tOf5zc+=xQ&b@>``m?vP<$y!EY+)Wz$86@XR&e4;{fLg3
zybEgZ-k06cF;=O;Ff^qdt!l9~GYk<sHc65+VMUGGs;&+gq-9KHuOvx1U?`KkTPB(-
zKr>mARB4yBr6XJ|sIn7Z-s7tShIl#q6#?AJMJJ_w>I#6NknBZ>S*V-Jl*}tdpk8dm
zTX*sijS8z6;0LcF);j!KHXvdrn*3HbTHTR83f3!}?o58ZYXGw`TwR@xh~CXJdBQ#J
zF+R*i7z3l_5uJVy?-a%1H3YZe17HxTr?ijXVQZHoHKDvlz$_eN{<UD+!&CxSU2+ls
z_UV7%{`a~w{XSSMF3lLEBQE*n4Lsne$FcR$!`ZpGj7<_U4YLZ~Ln4C9Qlx8e;&Agj
z`P7Hr%h5;Noh)0Xm-NXLl3;*@E@pIgUc*n%_#!|1!sj^b&UfJs&wMKGkUKEiJWVIR
zPF9ON8}j{18_%$&wX;NPPooj5@{SKbCH16*kOb|6Wv_+?l94>E8jmv)*v!RXnfkp`
zKh8JK{5PKcioam)zISD?X_};C0u!QPCQ)+aDXqY79WnebvkOlUk$+wX3^$%-_&4D9
z)|b?tfd-k-5HB|er@)A-e*6PA^t;U1SuQyBAGqPnFR>Miu^sRPJ3mtls`@hbakY&`
zYF+N?VOx3QdzcVpwK~H<*^ysM&Q@KHP!(X(k`CL-Fo;?4`M7q5K_i4x6(i+*Z<H|*
z%>ve?RBiW&j(64DF;xs0^6r%6!V8sLryY2GXY!yv@4^70LY8S5C=^iVNb=7z5X%+Y
zHuJ<&Kf%(LIqb$B)(eS%_d^EGaPcn|dB)4%zzx&;v3qfuzF#C!VWdK@XOL_Zq*cX0
zpK84mUIPrV48}KBcgwl>H39~&zG3+p0fr_6A>;S#uL>|!UH3X*sKn$=zz}Qg>&R0h
z>suAs<A=%G)85+#pe6u@v8IBKT?L6GSMgp{XH29H+RK`*N-md(wyTp}D?<PPAOJ~3
zK~&$Cy5h#{i>|V2kOLO1h6bT$zyv0QmDOf*a#qL}QkF=9x8m8;+CY0~1c33m^hXR%
zJ+@;=eTuU|f=a6N%01FVfNFvqR2v<lhLe~!_Bn~1+ln=9iotPFBy(iTJ9zEOUc`(3
z@QJ7nnC?vBfrZgBJC{?AJob3JnPd0z5_CHl-$6{cAkI5_U5nRcmUi#rn1?)&kDc&V
zbTFXTvm}y0E`sy}=rCgV&2R9Xk9`QMeeU<9r(l2k{w(jak?vG*hiT?2l$izEY|0gX
zh&uVjH7n17)L5tL99J)^bw;hoR|r9NmTZ{Pa~3YTgfrg#UR>sR;%onm;kJE9woN1X
z`qIdryO=91m~10bW2j`+%#iihujus^W6W>7lE&N|&ZLNZczr#s{qef`*TZG^7Dzg5
zxz(*0xFKm*x$T1=z_8!LVex6M+s%jhG%j|9Qir-01lk<J)xGR!h^+kgJwwN`%7IH&
zoXN^>T(gc`l_H73gYMmhTr2XI3I8yu&?uZ@)Rcq26nb?9YmzDrR1BI3PBZ#0<qpFU
zpLoo%On?7-^mZ>Y%m$%hXgl;0;noLq`O+s(U|(HkR!7WEcTwlDwuAS<%2D$FS1)-e
zRc=3+r;Nw53Tr)AH@WfYN@?e^*H>x{MfN1{zqYj_Ykc1(xYgLTHt%#+x}INY@WkXU
z_DagDmQa{N0;Sp{BU)MoEuTxHZD8`fHd!WF5);01U@|#yu;liZMF&`Yo{18^E}92-
z6($8LV-l;C&U0<A&5K7bfDMD&f+h1gkW$HAlx&3)Wuk#|{gi3j^o*27+!~uVl|Yp;
zv04O-_%Ph-M?CtdBYD~Lo`|{yraL_pWpQbVq4YWKNiSrO%rTz<ogS7BVg(&J#KY|D
z6n?b8EF(U3(!2Qh32()^WoD+j80nx0!$pVPF`{$nFZjt{zL9Ue<z3wAuzT^qzjy<B
z<dF;x-bgY<l~W+mTv>d7lxJBRsp1;h$+oQ#&FxdkO++e_T~qZ&s%NK$1UEs*b}!Hy
zD3_l0DL(e6uji24+>R%{;f-Yb2e9CAxjPo6gHqLyB0twMc1-JFVZFw^*)t3|%Wx*}
zi1nql=a1U4p~GSbPQi=Ah_d5~3or^BmplCMHE+OOcnNcofK_}FWFvURY93%Onls(?
zBDkir6rtwY5eKeCbV2E+v<J<`nzUw(t5_#gN;k#ObL|k<G{aEYgan#e#u)~!%YFBh
zVUQIH!g7eSv|Wtk#O3p;&}fJ6Givg|(0cl{dL>DfEFH0N%VvJKYda5m&Fk3b@qfS~
zBW9+j15?F_`Gp0N{uW;Qj?*~fD_>>5gKp2I*Ui)ISvoEaZckowr;`Y<)ofSeW@elg
z-8S?bi6|PK{TdC^6284w1`F!lIq&NXL)i~5JJfn3ho0LRhC+TL6T0&f7YvLif5*d*
zm2A-0ezsbdQhLf)vI|}Fji`G&@LXHND@&4M$zu^xSe}iVE?gV7vTen$)ZMnqf&1oi
z(s7^>b=Grt0mN0S!{p9kNk>S%N5q67U{$ObT=Bl*sRS-z3^v@a5*6gN4;=P*Q+58O
zn8_7`*Nj9GoVxrA0YG@~!%(kx<w~iNq*rD$;@p>tQHYs`8rKtmf(q7JP>1>qRnP4Y
zJcxh)>`4q3cd=<^BLGXId6MoN$35p&ocoiXW9<~FLdSLhN2U%dmTWX)c*Ay{{^VnM
z&zt`YMnigCOR(+`GKEg2$nr9SuY8$rfAphl*}4zE^^|Ag?|B4E+otH*BEKTP@5Gd!
z(x}!FsV8w^n$Lzwxnbq3MD(NO&xp^5?0wm=QPF&j=R@ZPTn5P?Lq_wQ{nzj2vY%bX
zvCn-r+kWQ(jQR=P={(9h9BKYLquZk<MWmkazIV`l{dMDCUpB@Zvkn+;%Fg+^W718F
zBFcxU2@({Oan*(AvA^xJ@GocbqtiagHY~<?GEp!|kRrLPbkjj}#T<cJ>7qK$I0$w7
z)inTxu^EQ3B6?+YW^<@@weqAY;LVzLOkHVX)yVQF60g@H2O_{g<5}AV7&I!tR&2U!
z8{-p6dt$kPHXrJR%*IgqrE67KdLZc}kb3N3M1Ru;F4(n$yT9Oh-1dbp;rihsn>TI*
z7bI;yOUb%@PW#l?c>l*v$8SEEtFGJ4RH9g^Xd`tVu|;C3LrfSBtyOz9mZn{j6a#=s
z!3sLThP6$S)N~bUX>Qa2OL=XIB2#t69XNJbiieGTKKoEVo@(gb`=%smWx!Cc;H#pr
zd4H)C*1NHtt%)*Pf^(3pRPV>;UGcur8uBEofQ2;-JvSYav@sjpQs7qx4N?|&&6Yx(
zc|z-O`?x@<(x!WCv1NH*7IR(eku+%1GeAhn83rkARZZwdSFT6J@t6{s6#hyD*f+UF
zauvE(>_Mj7W4!leqab|Alt4o?0*ez1V;$f9v;ywokInnC8sAnANhjp6c%L%c?~yGm
zaNc*%M3(2-I2|%x(sYSVXNG6L?Dc%+?EhwJ^H#EKh?rpc+>s6@$=JE$I_`YH{`~XD
z-p_&iY(QK}VtZsh>|f$MvMk6IKjZB8p2X6{7jxv3o=pGfBN^^{0KPAD^7Ue=%o4mb
zjD<@xfXRT>NA6ig_?Z-IGq)x82qIc#Hh9e`jmoi0^EB3Mn9<D=MnC#7r@#Ge-1?A1
zIr8~`2nQWZI-6jPMRUeVqgi!{@98}Zot)_Cz5co>*@fR|XQZ1YV8|cD{|Y>6eTnT2
zMrb_9@{K%r;tLp0dc$=r-SA5`?p)+sul_4GUw;E#t6&V?CqZc`9wAt+N<->9F%_Yl
zVTc~nwIG|d0S47(@y*`NFvNvlqZ}9;dM+OfHwG|BRoJ!yhRF;AG@O#89fvgiS0`k>
zjRA(5?5gaXwN=}B@^hgiP+t|Z*5JhBony)f*}{;SnGO76xJ-J)5j^Uok8s2CZZ>Y)
zfc1cQhAzeV9#>wo$nQV*<)qWwSRRc?M=Aa38Au(PWk@2ZQ;cZPCwrf3cqeJd<LjVx
zC4M|;8@tPBv(T$teAf0@Z46y|-xA<8d4^%E<cxA0R$>^W#%Kk=V5IG4tOJI)XC6MG
zuX(={frFA`DAHejWCxkoH#TotPxM-DZvaK-M)rVRh?@|OI1^kNm|QE<(^|obwg4ui
z(J61swU%0=qwW|r`E&cc<1bziA3FYtSGs=0R5oBH8v<~Hog(e_LfEslsLLmEH80}D
zp`q+;9X2nw0kt5$>|93?iBz_%@%)2CAfb>C$@tod&Q&?QTg($_{2QYCPx3B(lz{Dv
zIdLx`wOSqOQf8*7*md=_T==82Nk)ron(bpSqzvdV#XCQA8gDt_R5t8)5W`F`>PU^J
zCxRa>GNnUa_sT!!IZu2vOUsM&`#ppttZ)T81a@A<&p-Wn{`)hZ;O_T2g2SKnM~n`*
zHM*^b0XjlB@3H&oq|IFzTQdc<Qo^?!tQG8R&4nw!^Guz3D<&GkTDF2yMTSGnaL6T}
z`7CGu^A|btxJPrFM?Q*l%O>oM37{jFidrM3d{9JNo<-vgroB-stTPTb$?FTonB#6v
zwdBp_rK*oq)i12Cv76vr3JrTfj}aK|xE8y6fuEo92^P=#8vUgutR&c^L*_$I1S7co
z`zM6Lwp4FY?{d~{h1jDIwpcehMysyAjg2|A)Y7vWkYO(whQ_0H6^5aDK>J4LUsoA=
z)_N+?WWb=~J@wk~aBCN;&TvTcy(9*1f*1mjR6zzqOfp3_vw;gcmZzU~D$AR<(AhXc
zqJ~rjZ!!i;Dcb4tqSw5e|2+46Hte%M+jlL{C8291A0l>vdB6wiJ_ig%apTLP(T_>r
z+5qBM+f;n%_EnGQm;@Lm_1yBK$R$MupsFzp4F+LFhM|$Epjrna@lvV<w@pBbm9bHM
zm>gU++f-5pnim_I566P~pm#UofQ%7@<Y7myg^vMlo0VqF4n)#Kkrw&Cu2QxXfFXhe
z4XJ)9yK8&T3EPj>+6NO|l-oKOvG-g3I4O$8yz?xUzEd`gc5+W69DUf~T(R?dzI5ev
zT+uO%l7vxiL|}~s0H6Yr;!S>R+8DDIEm;TSO<9<#c)a*(21tt{wizm>B@=__#Lq2b
zpR8bTOf{__5sRc+8YZZE#2S1S(vYnQ>D8GZvSnt5t1tUGKl<;p*)k2&y*^ef;7GB2
z@^k;eOJ4JKW^Q#b3xg%h$TQpTkh09++N*fNaS!ADfAe}00VIXYAW0{j@7*b#(U9za
z&f%O7y`SaXBaV8>bKxHMVfldlFx?KuLt?-?ADoC9_lUFs$<`rn3izCFF%kiW#{OD(
zqncm=n-2!!yvG`YdckKS?2?C_SMwhq_z1uJ>BT(drLTg+Z%?}42CxBpNzpI}o_Cdz
zMR}a@^{Y0BJp~Qx48u*%E<8#^zPJt;eswIvKLJlzUut^{5o()Vv4t2h<nydpQ8Zm*
ze&-HmFS>$n{`srfv}>Lo9jpa4hSU!+Vlmi2X;19NMIF~>Cc~aG42|b~Ooi{YOA=9N
z*$QP!-4$hRhd7yGXtY5iz)<>-@1Z2A*0EhH!w`jTNo;@<>#|#dy)-)RQkJ7pN7@K3
zmwt2rKOBZM+PNuyCMl19?}y-ycV=liVR~wsk#i(?76(ImGqZf;@6Y62ANv%&eGg^(
zjzy*tM@K_|QR*|Sc)STx4y#JyC6^0Wyt>fYi;k(;y)HHHWBRm}BuQ(ZVTksqS|Pjl
zlBB(yVOWu2sE}c}e?5g8k9eC)cXI!I3AT-1prcBnj@fLH%|2z>!}jh5h9(L51hvFx
zR<Ht&<Q!28cTa4ROW%I)@*kHv@hDN~Ufd9f=cWo?C6Yx}vlX&}tZ-6aLL2|vpz0z8
zolk_R;eZ2{w)3((-hsP1gV}GE_g!!?|GAj5OS@#94v1oc*?u@C<8e}1z=}3^-ar#b
zS+xi0sL%EDCOS{CgiLkyhD*T~lxi|5=C6n{1RTz4E`cHkC?>3rOfZ4Mz<43C7VlDw
zW=!`xTyyy^IO88b!h`N}XFB3Z>=d39rNh}j_yv!A!gJWV|3NGc4TFwjy6;%J`U>uT
z_^mnRJ#XXigKkAO8qn`el}A4sK`(W1_2qp3!yn|QXP?b|j(G&PecE$b+`2ExrcRI_
z`J&9vVRK{`9<x)0jA$j+QJS+a4Ly*TT>6*0Vi6eyx`K(%nOwh=(h<v!ubsu`Px%lB
z9eNl?z5Ep{Zr)69V>h>WErKHR`@R4nBEgZX=4;cI(=BUl)2*`#H~s4~#+WDGoDSIB
z+|ICwd|-W@?F}N7C?l*In=5tEMG?`NrZ>A8bDKli_in$<pfi=vFgQ{-3YyW_kDR)t
zqO|@?{9tPMh0979R$73<N?2Ez?1d!bUzJI=vuo*gt`&#Z$Kdv)^ju4{m9C=X(iNMu
zQ8k?ZU4;hY5P_GtP>mfn1jI<5HXJ~hQ^9u=yjA?}CGH}9PJ8Av`NgL{#k3~u-mwt$
z(mr8sW)|x+o^$-~@s-m*KxgS%=DJFEs>@R9@Ydk1#Tkp*99$X=57yjIrj%AST6o$x
z54BN-*wW;t>YeCn?<H%W(w9}%ysWzP$A49HiZuU*rvH0&m*zTqF&H)U(jj-VmS^Ab
zt~_AtHfB;sQl9ss!`<#YHeR4e>9(e8y&>YCttC*ml5F)85MT2mzggP$mT~VXSrIJg
z%IinRdj>&YZgRH4W2|M@?nQPjEd&6U<k81vAR%GBQ%QZAfnW8TUK)@`YO@zQc1qe?
zCQ`14s%sI0h2r)p7zK?kh>7HWKFoO|Di{@PV#v}FBA$M)%jFkdz$gFiB#ylQJuohS
z1$2aS$|YCM^9Rp-G5a2PAj45cnk`}7GH(8Q{_0gP<{O{+7`NYl3yC=Ty(yeRHUhUa
zqPMWfwV(MkpM3W7*zt>Nc+%V7%ONj%8S{r8NV2IDlCm|b=andYZJfc82$VJV^v2Xm
zex5aQhzfS7A;(Cg?N`eAAWYD2WzHcBBRca-eEV(h;<G27!m&?!3P=6vpEKNdD^nZ0
zxg~F4{nXW;xhk%-r)F#iWaT9ImMujZ+l#QC*zkX~mQ8MkC;98%`@aXCxxVD~yo_zS
zpcF-45nP5U-Cmz`gX8eWJesq<_XGNFIk<Qv37JTsl#MB!c~qq3`PdwNmfA3?O6BXW
zE0)Z}PN0qFb<k*z_xIbo{!OqVElw!9l5Tp<?G>E}7i2H4*(zwJ0UFwgW67OP>Vck<
zxRAE7_4?Tyi*m}JclgjMD|?##Dg3~(yla_zZQYNbe(+Sz|JJuT`W+`S*fGRxo(c^y
zEZweW{|&;oKmC4Q`lb^(^V>gQ^VS2|xwwSw8oI`krHXiiwV|NOjDmNy#`eUs(G0{@
zR$V)5zp0{M&<Yx4ukYElzD<n{X-Ob&cDiM(@D1s-Dq0n$d70@X!!Y7Do42t|684+h
zgu8YgnHF5Cn8f6B48Fydbo>sKRZErCb2D_}xzTEmX#gdOPnJfq9G$!lw0T%M;aKB*
z3vC3QDTsvbyTdZDhV7=yxeE%@bNuAW?OZULVds>=8ILhZkhXb2LJ9!hWH~b<dCIfY
zy5B_)(5OLkN2SW$D4x?1R!3^lzESMwv4*P{l%^a~KoG36bd-{*^2X<%KQIk~;4=r}
z>DexqU2+kpoO}Ym_u%`Ijh30|^}&w-Wyid5>~T*=U6<kR6d`331CBfDLA>)VucDKU
zusWodbnqBb2QpAPs&L8A`06|VmK{I4kmLU71<XABkt}T5L~qtG?L)?a%Tug<{`@oX
z+M$xmRW?^9ay_cyvvPJpqw`W_O$7GOY6z1nn=v`4QU$f45$Dqpy$m?-JpTE8A0RP3
zp8bxup$8wtXk(9VCxFQS791iLli2c7hWU_g8BHTP;kJ@h2)q{1tgOqDZo=2!i^$h+
zz76P`(Mwg2P}OhW92t`THIT4si5m^C2_PAA)eZdc#1FIktbb$kXgO^9Qv%*%d}J3_
z()qs8ON@qTTgrrHYHItgruBj-0zEY;%^GGHnx#NJZMHp>sGKQ~wJo)=xrLSksJZMk
ztW&Gs*Am>;^66?4jGCQE4W5l9Yd5V&wbo;9J~m{9hlJKri$e%$kP(RpP9183zk0Tu
zGTWPBso&%0lJJxhPk_U2$9z|r?az{Wk2Oj<SjP5edCw_d<RhQ>EWORQV&`Cl%QCw8
zgH<%eYAC?U&ns1GZe3myDdvk=1<?pF`W7~<^6hVG@>gpkd*58`^;wa+T&CsSCeAR_
zfqp#0;KQ)T_zZ)L(@?5fvpp^~*IETjY4GMjHAs>SN_Vut{buGkY-$t#ef2N7aH_+S
zwcrxONDaJ7!_7>nTwlgo3wg)7br4fm+RS{*WNLdToK9-@ukiL!9ax+B($?LRx+igT
zbC#vNf;W(fK{Lmc8?iAPFyk$YCSga<lG#G4Wka&F6AT>?k2869S_uwF1%oC&2Mton
zyd-k>(TG=3ZLltD%+V=7r7+eKDgb6q{9$v;ZWYaEF2ZvqQG?f%Of%-D`)t4bV&3<T
zH}KTQABVb>=}sq4BzDYu!|~601wZ`hWhC7>c3giAhaIplA35o5+~ZCMk-BBNwvX2g
zLxN8;birYluIEP|JeBW%{)^oG!4Kjd&wnQAeg~26vw=he@t#CXVAgZDCR0c*YgP$5
zCU-07*VZRR9FEsKZ)MP12+RwC8z$U#reJ0izc9iM2V8La8T|Vv|B2r{<{0k!lqZsH
z-GbTJ%g=ELV=LUP1XBhAIt>zIX&kq#2oiM7#o?X}J-5j6)?ekuB_h8gBIm3FhF{ki
z_x?oS#p`Qt&jG`vd<Komd=ZadUdCK}8UOO)=X1!ykZu+RAtmXe-iL>hF{^JP%?Gnk
z7FBh?AS+i0WAXuAsh>QtzgZN?^+ewB{(7xQTjS6(6Z-o!igj6S=vg8!b0+V-P0f8B
z_|&+VvDpu(tWw9jM{5BLQ9-R$dbbgum!V1i-!Qyr#4(jQTn3%FO`NyudLHzuS8>2I
zp23c#c{Xg>fOU#W#vo0}vJUej!|~7mQ!d%*=*-PBziWY{-wWO+!xUq4IgI$0O-;uO
z9&LMit_3ixL`JLz7@Ga|)c`{@G`#{~h?>3leCSzL8+tAPe2t9=QnG3&wzR?m3*Nbq
z8Y`aOQpzk{vTlc+0(3k0%+)g>B+ht<@+4~OG`@?aeKMcbT$Fs73DK0%wcYsb?BU8u
z^~zc*B&}z|luTB|z$$?i@DdWaZIMv!eNZ3?I3K>p*aYWY-qjC-tde)ai-h(fhoNE^
ze080*9;o6=A~`>sA`|JE)+TgmHyb5kH|I{YsD!k1)3PX<qa;hnS@14ps@GxrHJ9_}
zFMS@bed%)<EG;lS-6wNOA{J+bXaDKz_|zF+WB-lYaJtN&z2Zf@`00;hFj%J7O-M{9
z>@k_bLWWIuv->~)ldpaBgKX%|a?~IH38OpRjlq@;bbHVRjQ4p|8nwnkaTa}K*Btpd
z1eYq&%26$D;AjKX$Js+#CA=5c!>h~L7K7&3J<Fh%WpL@moOSZaTzT2GJnDtdXY;-9
zOFGx5(=#R5tNCU$_hyM2Ohw+u+DL^Ji$h%tI8atc->tXY|4A_Meq+o_Zm#v@&GMzH
zhvzKA=5<(b;~9r|sJ75Ps$`hOMJ{;viCp#B&$GWIB;rVoiIR8Q-?SR?Y_}fNs9;((
zj$SA9LZZ~l)w}+U&slVvMaFGaBaOe$zqjxwUkgB3RZyW?lP<(V&uxHVE$V^#AZ<NV
ziWb&Gqo+@E#Ivm%TkqAk0KCF_6^EE$cj<kXOcPWLiBVKX!KQZrWI|`t9G5Q+nK|OO
zc+lVcHA~$toeeWsFQm@V?ZRl3qIQatPCJwLeB^Yt9(XX<j0R-GjA;|38tNRzNdAxv
zgMM0q!T8GiN_!ozch#}SYH2$gEgl(LALFYmwoR2r$Y5=-b`5}GY+<(nhWvetStO{>
z_)3FKD%rW3i)_$V8*fS%9iO}Ul+IJmV+zdd1Zxc=^{Cj~(lK^|syfWEs!%7~)GD!t
zyd8#3E>s!`cq=drT4pO~LsGX#=zj&k5QBy&FiS!_o5?MOiGtR=tEO;*Rk6w0np_&A
zLB1nuLgC*W6fgG&N9=E1%FYnHp~}$EYF{3&>7>>FOQJ4$dFJwtyr>C{2Mrp1jDR;7
zg~Wl{kT2kzBXcR!y&l*9@={*@!e{cPSHFbe!aN&hXHm^<x=Hw}6Hei6ANmMeXSeZ?
z-?<NO{mWOd&!#?|ZidM^h)qb<W0n=WxP-sr3eG<1WG?&u_jts!pTYh|KZ@c0Tk*3!
ztQJ<OzP{JXILlP!Dm1Tqc-ED6!s4l9T~Sz_MDnD9d`D3u;1n@IMx<DbRYp$HG^IB(
zT>F(TbN0tS$-#FxjQc(1S@_%B3b$z*gU4d>J)yR}k9Yrmt#w1w>_jKDNY<H##%o8;
zE?lq<7=HaM!(RcfU0;K1yV2WCAX~~F&1q93xNM2!!b|wV^Iy#EhAAdnCKH40bi-g?
z-aRZfv^dCSEJ=2(GcEP>RVIK3o%9UXJ>G}vjm%g=H7SZnTb*fEYo}-<{dXKNw7Vp&
zCP`WmFx14gF+Hx0<B+l9b6k=%{vc4ay&I`{A!dr>@MK1CxyheZMRkxLhXn;ohisJ5
z>GoObD8HPZ=1G5hGU@Ga!)UI{w6(YtOvmH1A;V0#a{GuE{>2;l$@U>r8#gi7y+C3F
zn{@G+!>9wJc&Aa(6$#I^>Hv=~2yJB;;-Xx2Es~_jB`E@K^-D8eVAsKlZw-vA>`0|P
zf1a~trL)TfNm4oZs#=9e6>A*LWh|i^?!*VF26-D)eHd&n!n}g58qSg1%ETeETFdrm
zzqhJ$ppk4xw9(au3BEL1FZg0@cq=q1t9SNzs%Vv|Y%RO@@mmU_KKJ4moCy*y<82O5
zJsykq0f<=)PF<d?9<mn7VOa6Cv}ltL`TNNJRD2{eEAQ>xeJ39-ms*t>1G|b@ZfzBe
z&8QVpD5)|`3h*u9#bCYCEsJzzq$%A_!r~3r@QlYD!+Zbwwd~xzn@t;L0aFfBe)yDs
z<WJxDE{;9+_jtqWU&$Q~*oSPiKzFJ~7`YTi3!ZM8!Bv;?lTV$-xnKJ_cfIcs{Py!-
zz~YuekS$Ym6EHqx9C+#}X0l3mM;f2A9CFUnmkR(_W8)e>Rqi_)oRLIb8K$Pj;8IUF
zb+F?a&i%+~{Ph3*KOXv|r?CIg$1<Fop*I_3QwAyJQxlbuvM0RRp4Sd52a&G8MYPx{
zH=7`<{*GJT>n+BZ*WH}!$(!v<RrgobZvwYlhX#9g!6$~96%xz5Q|l$e>>Bd351+#J
zGfroAG@@g>h?tPRn``^qM?9rfD<4Zes!%swSu{D8^sH;%8EY=46RZ%M$IDpq@}7;$
zti;ex1`H9Su{VGr?zFZ7hUm{RV32x7fv)iQ(MiQy6waq8&I*};Q6<46E|*7Of)~3F
zomS(*&zk8Oez0RFNB`MNIq+#uV{xcVZJNPn;H{%0Fc=KTk{M3;`!Df#pZRC{Q*#V_
z!eZ*_hza?)-dD~}#UyrF<mMR;!O*Y<?^<>KwTEKFysh$ht`eEsYucK%s<bsi8j_?I
z``36u6%0cR80xoEona{0rNVEX24L`cF=~oIPAxJo^3<&;^iDMg42>dw5^Gq<=cI~+
z3$WdUE~><}MgAV9t)tpt7~=b-N-@qqqx#0+vPQ`r1Z!ojgNXJFlkv4L8mwt!U3szJ
zh!z9m;F3bsA|bpmq;ln<gi_`u6bD&-=5U^jf|_7nH1z-gAOJ~3K~%yW$rSZ#&Ok`$
z?;ycmH*)b21T|6y2*(abXuBo$*h-hEcrHMvkut4MBqUc`DcUd=qn;#R$EovlW;-li
ze-(fHj3@AxSN{o%3yVz8Ok?xw=78^f_X6Jdrg!k>zj;0PzRPVOb956xEFL9YbR@$O
zc7Bmd|Mhcx`=h6E=<ROJ{r>m`aQN+6+Oiqb=>{w3MxH`7#}Thf9g}L-KYvUz#Gu2K
zL(uV>s!mmImrA%+VSVHS$yU+j3_J2Ho^uXgKItU3?SBXle$G?i&^wTAnh6F1>chEl
z+A#f)$y&!>U*$En^=TX!)F2e3CTZ11%jy5r>*pf!KoPlm9Weawfrgi>>U-9qz}_+r
zGO+|IfJ5doI#*uDna_SY`z<eF(=7Pl+YZh-te9M~qh(h`3d=;Vt&N;|4#NsL1Cz*x
zmd@=;j6x%ut+vUxN|s7VQc;lX$<Xu4U6Zj4gF?<C+86}Cv4Ek;+(g@Utgod37*=Ta
zOzbq3zt>=VmF1wu11>m%<nkmnJ`9|z(!~bXlu<@!rqA_m8M(`yIP&f9AltTu(Nve7
z?ID?>q@?N?jymjIa=h{{-^jTaZ711!Fgte5(@jS7jK%rTRg}EAwI=Tjs>_R6Q)3%M
zTMTHUaIty`F$qvbomcN0;Cx)iUoi~hcW3$aJ(MJA6vQX1Fbrr_=a()!S`30qnUkOq
z$SH&0Ot5n_@p_5Ar`dh4or{``Fj#IHW23b4d=|r)Q1BZm3-6+ZEoDZAZ!kd7Z<SCy
z8<8lWv7U3{oe#ltEmyVGIkn|_dhZb{kOs*~VXbe(AX$dkz!hZPqhi2-@h+UZJpDI7
zf;_*#XpLzpS%or%UBdThP~&?^gNu?ks47M*>OD4}Rq*BYjXNhYp~Kt0`8Hkf%G_un
z%qNIq)zR}QUE|198I26u?c$u$?@uvI=lT7I-;WQy<uxn~mYM2xgToJydgaQiw)2DY
ze##?{ejwQ>BT1CRBzT0uFe4ddbalw!o8ROMC!9=IJx3k?45lCQ2zKtfnW<^8Ug?<J
zg~^)y<K}C6;&5F|)inmjYVt0L{=Ir5ZzvfiqxA=xFANDxXuf3`bTbb(T+equdMf|(
z@8964CqJ1(ANgp~jT=d(3?`S4InA3?>KlMrer@%^_^NMI57D-dNOV6+nakt7uR>c}
z!Ne`56kYw?iu{R)y!V#SaK1SMhD25W6S&_x9JuLriZXZyaK^)O%C)C_lpmjd8vDA8
zuJ@?g^5=}nA5h*$cCA$QZV-8&Yn8&PYY1*@VLI7chM}d9qNThm|6Ft|jhS3Ak13L5
ztYB{i4685<WfN7_HWyI`8d?K`F)$$R#5NfQnLM<<`dnM+WR}*0<-^tG1BFQlF2d)E
zH!TM-6BTsi=~}~lVz|7QaNL_-L;oQUV@KxLG&hUOf@GtU7*g*^GvVy-{*)KI_C3hd
zMml}N{O%=euUCc|A&{m6u+Xugs4Ry<wOEx}XC<O*&sg%WuDD2xMB65t7d2{EcS+Jo
z!`AJ9p?FxYE=g)K3>9nS^J^h-=QMY?D%gr*=CgDR$_p$N6Crb*3^t<xz+$!{978DF
zvou4E$)z7YfY0C`nID9y0E24HCRid_lP`Wx&RCd^!Dl`IMhOxj^|=hFOb4$*9BVs)
z)-GJFUmey=-nvh7BViRJF}T!W^P*aOaISHHw}OxbxtBA!MQU!hYIDvdMrx76P63z{
zAHKz8A(7jdLLG{6ip3ya5MxO5&*!DV#1Iw2GN~b;WmX4KPd=9Ghzbe9RF<-Bv`BBq
zPBw0y<C+m%+wUM7Hj&Nm;Mk+?%ZK0jCPu>nGrg(WZ%9+ea+)zWJq6BzF{CcT4IN||
zeQ)s>|CE1u$KR6d+R5)6eGFTVIf``O0~u_drXyhSbkyLSj|?g_pKYq9OJt1pi?Kq)
zKh$tUH2YMbLW-36qOi~3J7<r?K;{&iKnmzErI&VC{@(Za*ONYgvC0!(@-p-ehceo-
z0fQ&O<gUcNlzJLt%6nGif3IBAIzbmTB=TouzAgj!_a%#6>nDk6v9(nGkIXlGM?`*C
zM6z|j@V^flo}#LsSg-nTf{sGe0q{<-2;Ra_NAxe>&KF+zVs5?jdhGH5NxH#vB1(^~
zie0IsM!Hf*yiv?gD)?Kw_Ijg}o@J~^`1Lak1z_;pgbYIj3dXK<OA>NXFi(_*vGO%4
z*Gz1YDZchF)?(PwQ+ksNp$Ig_k`6OE_>ef<o8f{T*RkKD9?nDG@owgaOH6H@qeE~N
z!8$~Q`CW@>GRL32<wKnD)&F4ILAPaBn&GpQo;CP94cJ(abd)FICS_n*SrD{PSyjQf
zl3DJ!^=}Cn8X#@5B&qox?F>Uv_%tO+YXuC=wGvA-iXt@pdjS9(DrNZ+4%LTi7A%;3
z^~jeoOQPmDO%0?5%TGm}(CvEMXo!eqgdmcXMHmunFH0925sX%v{sQP$41*FfbUofH
zouo_ZQj8d^3BV(Dxx1SfoXbdzt<S8)!G?@)kOkji#G=ll>ak+*E<<n(mX?^B>Ei&B
z;O3Jzj&Y{Qm<|KO)(eUEsHrSaK@l&QdLlq6rz_=ux%5cA2>?VDn+nDl#JJqjIJXcr
zfq}r77V*t0`&C3eDzQsa_za2hY+PF8uGxS)N<zn{Tsz(2=lkr-WlKvOao0om_(|_&
zVR@eZ)HI1ytd4gX#wMt0-b@g(5wJUBiWGi%DPKML6fQdVzj*j@kKy*mAJ5{}{a{m<
z#6V9#vj8%D{wYk$L#^2BYf&i^_UdXa7ac9FBzWchpd`ME1J!(9#pJ9LB4ER?vl(PC
zT;h9w_Ya)&<uCD|qaVeckADiI4I8i<I-o($<$M8>B?sq4!^O1DF8%tTLLE_t#$fJ@
zW+az>Rjgz&FH_FO#P-S_u$9D7Wvr!gOJx_HDkA@QOQ;ub5ijrkXMxA8Lx4R42(1Ih
zdEpq_G&)gqaf$gapUn^7{nu<6rHHpA)`xCZZsF$DV~mt;9HGl6)G1A^%`TLUU0F>E
z&|>|$7KVX_+fc(LX{<J`6+3H~VOXX8ud7fVm91(8P+|}MxbB+-9>#!1qwsBW51-Vv
znv@~2@^D(E2`^CsKJxZA6_cUk=UX+L1Jf0Bq{z_I>Cdv<G5lb($a7A97u~zxn`I3B
zxfxQYjKnizU^rMJvvYjw+@JHhcb>>EcM8*6=h(e#7wSDpw};}f=`atEGUd=_XdHU#
zMzpc>(*&wT+BpP8D@jsQ?iH~=C4Dba^Pwt$Y?FnmvNQIcVW_PYErXUL2cbfgQb=WT
zdtC9ARj3v^ddX*T^G<)M*qCe{P)I|H?!U*=o0=lgWsW}b!TjjFA9Kl-+sUvwvnLo6
z>@Y)@JX|y5i%_y=7+aM-?41v?l+@Foonn4z7yED8f=`DGGRN{z@oKO(NLWOgc`ofp
zxi;w_`AmmGR|Fwt-)$Q?`k@cw|DAg-7hZZfOPNP(2k$blN#0P&@^x==yH#JwyC}}F
zEk@`mAVw@F3LiD>(dvbsD%L3-pOW}6_mLT8m~?Pna3&wVErGJs_QbJIE_I?K-WPqD
z>JUIv*JW&3+|2{}!aZlZ$Y6n;b6vhQG;F=oVf@2~PGEU3Pj6}lgTa&KcFq#5W>2OD
zmkyX(0Oqgb+z*_>kH35tcYDx7`0YP_7CN_)rF}ObNe8P+R|M}etXNzrl&bEr5Xfd~
z7PB%&p;j|18iU%%3blmGC8I`Z$67Q$z=T;JJWc{L;1)+DBglUEBhGlwd+2mJJoc4;
z1cx3@x_JZAN$~3EhzYY-xeTeeu3`njl6jAq0%-rbO15IwEr;Ne@9nB&qXAtNNJsfO
zve4-<>?nzJf;6wVwmC8A)Z?=Z@4+EVB~^Bd<a_@um}B^yG3K$il=|@&@lw@?s_I$m
zkl-d-5mE3J86jZCDE9hgzV^JQv-R3*up>vJinX~8UBOyg6(tg-JU2MaraYmgm(es>
zX)kJTluMGX^1YEHX-|fpYtoiBhNE8j$5NnJV?Bl~)&dw>GCAwn%w_di`#m&)*oyaB
z3klI|Fo$+@B?Zol9UpaXCDIFr9=E}Z!COOT<j~=Oj(ECT<~aX`?cCw9PvW;;^=h0o
z3^z>CvpsC)AW<^sSzgM}?lf<B_i23cOJ8SdW)t213_BK=Fb1ZqN1Y2rmX-WrU&)f1
zA80&oaFRVqJ+QEJNkSqfMd~{G+6+VCCRUw`F)&wl&+C9e%kH|b0sI&w$BSNwoQ}r8
zx7G(-)kk2)G}K1LSE|srHgn$k0wHoyf=-fPWXPGPe~67!J+8iCo<|=4T-0vBi{Ld2
zpQi*Hd68NGJT=fzR#VI-AgDDMbr`?IpS<L`Jo&g|$VNjR|Fjoy)sAJ-WIB}MuCU{s
znDHF_H46C33;j+a=x~vP_uI(7e)3~57_xI=nTI^~DfsRjBi}(1z-L%5HS)sQP}mzS
z&abPA*oIuF?r0WR6yw4<={nCG%WPY|mcut}#dQ)cT^e#J9foEqD5gv}b5hD`WMUPd
zHTqYh`(8?!Qn42M&aq`_nIk%IXwP%2ZL@syyz@Bh$;b2OZ+`;|%ZtqPrvbrvhoKm_
z2PgT}94X1-5Iee-OaAHe{M$!Q<M6xx7DvAHC8P)3mf^-ey(vS2AYMzBrx;ofw^~$y
zs?S3VC9()^T-{oDBPBwrnMX_9LWt%RN|wdrJlMP;67n<%4oJJ_Iq$U7`L8d3mHQq2
zaPD^elUbbGNM|k~$-C{E%chD<ASnXfiU)!P7q}`ATN)GmIy{9+3uY7c%8_yOoP*0Q
z;@s0e$<^oogmkpb?8Z3`x$phC<71vkx?vxr*Dd=LiC2cs(My2AJQxE>218-Q6n1(#
zN&>oNattE!u$yD5bK5Jq1znV{d+$#MUbJ5I?>*yCG6PXB&6qrRNQBXx<sr}gBfj$b
zzu=J06yhCDB>)AlIm2KpAm9s%x9(smV_lt=IfIpISFJzHCUVTuD73Y;rk_9n?B&L+
zH|_P=8?YDqf~-2wDJ%WfxD7ekb-T@0rc9!ZejqY%s%YxsVei5KsSyq7(EScZ9HSkJ
z++}V*uKv<leEh55;Bl{e4bu;M2=jiP*$s2$%*5PWLh4iA{PN>@)>9wNYu|Jt=Usd$
z8#nC3lFL{g4(M824n&4^pd~oavgpuKn>&7g#J-GLt2L;NK*U3nQa_K{a$fV904<z|
zT5r_$^WS#a+J=v%)ZN$J!0L=oWrZ;2Ws-ldk3y_!PF@kCF_!xQ7-K>RT%Kt%8V$M2
zZ4YGg%rt(m7#y=C*z=-bjg7!faBC{TW(1##^Ia6w)|dbSG}w)HG~=<4d>9=!WXq;m
z_T9LNYjzIFvJ7jZao<J_t*zuzyf0cb&B$Cv*BE-%U^PQghJ)q2*h&Hu;$7}KCFL+~
zSWa=^VC7OJwSTEIr4rxS-XW5ZkT5-1<`G->=k`vp8>adB>>QuEatGHXin9i%9w~>s
z>x7U9W#V?-r7g75gSI{&zRpbp#DwW2bt7~zVrs)?>?ZiuMHlg=&z#Oa_qiv_>4@3>
zOljRq-c9zZs94A#-8E!J7ubH**?i^Hk1#vEi5I;4{c!6;S==^5XSUCj5As;e*`^}w
z&qV7t_q+<XsKl;tIg>ZmW6Dd_0yxC?utdyTxyGyf3^bS<7-NFG#(JEBAEk5$g8T80
z_~OZ@;6!-do8AI<ybZ&RGjwfa7Lc1ml=q^^5r?k;O#saRc7?#wo4i+80<UmB!g+V8
zgNuK{=U@2>X6I8jVNmPPYi{8C=U>DRzWaTi^v)CUoo>jwBY@<dSH5@B$y{^RS4c8p
zxO_d!{b?Tgnzzxp_q`!8$Srq_;pAHiShz(2h8#4!PgNfa+-e;P+!zE=>GqxXrNf5t
z2I+ON_rE_I?|B67{PQ49!}oUtW5dvM8JblN-byaL%_S8YiXvT0BAG<sAoe<MMG528
zl1@?>Xsa!B9jB<?6en|S9icS_v`zZCMq99P6ysG;^IKQWyQSWFE-gOWi`HhLU3I~s
zRZ^2nlB@v{M>g<?4blgfbjW%c>EZ$#I$gGO6TbB4uVmvr@5RI4{0@fRGMwwMA?e^U
zgE5BLzT<ZLbUE#PujPL(yp%V-^JIQ@#SW(D<`_A}X-3a<@`B5kMUfP1z3lEry8gIB
zsS?eKYBirB4KN!^kdS(!n#vfUD+}&=^&QEEVz2k+vLBZMNeg3E4EUA|pthtElud((
zi70@N<}(GQhP}Sd#*>_5M<IvFEp;_dG(P{L%lYBOS8>>F58)rq_<Kg)!YG4|DQ!E8
zVPuu2)g^U1%N%@q@j)r19%F@UXu0P41!nf&kE?bom+joeFt>yCTG+N~jlCO}eNhXR
zD@=khiil<C9T!}373W@X1&7`Cww!+E=U6mdj1MwHV@#g%OwO7FDO|u_CNc^zg<!7C
z9u_s_*P>~TaBeVQETe45l&0LqI&LSzzztyY20E9IaHfxwAR8^@L`CPl2yQF~YYSa@
zB$*3fET5l{Q1E)uytegBi?U_b^7G3s=CJ!ej8o41C$67g0F-{m2Bu03s3Wln9>`LV
z!H~}KwfG<Zkk6g?0Y-}%k9pb4n7Qu*nU^WLn>tJ<ghql$(D)M_w^VnC#&on)gZNql
zuTn9eEl-Y`#cV_2DH4I`C}r7u#U$a_>@X!++Rgc&ID>Qk<xAZ6;m2_2XFQXonLhmu
zmTobut{(M*B$Tor)CB8WQLsUN=SA@3cF{Nr2V*G^^{<Lmh<hW^O6!h#{LiO8%RbAF
zUK;kxFt_75xHrv(KR6#Qy$tSl=kRkb-&>O|^FKL{eXqNYzMrFSE$45)n)%DFq;t=E
zVQlpvzJ+0it3>2|x3u!`7WGoqKUdYatylgxf^Ud+(M*G%waNEcwZbS}rgz>2eD!57
zXTQuN>4>foG;>&!RJuoF0_#xdX0_0@=Uo{Y7corg?6rT-Dl-gP3q~CuZp+;qngv?Z
z{xxjWvW6|XEi|I#SjT1<SXGj=qU~(c2fGDB)jyq#574fGAZeS$7#jsH(t)Mj-(6`L
zj*YwVVH!5ruOc{cVX)39ZkS;+Pp3b{NDb#NFZ1A6zKlcv@M$a!GR#bup6N2m0xM?3
zu{d-rr(Mqc#&>!5Nq^7o!3de_Ga3Zn!%ot{6>=Y!23MZaG(C9Rdwdy2=JGD4h~&X$
z`FxKQ#dMG)jXAQ2Rs)jwvcQpA;akl-RFb4dIya4<E6!^!!-%h^)Mg(Vvb5L&RjRp$
zCQ}qkprnyzTwZytQxHX(a${eve;yUYhG4BM8zEf}POx^0QJVYwtAh9WxvuP9kt|;>
zruv-ZvM8x$D!doO2#NE|&ULugz3<H>mtMj}mtT(f4w(yIH_^cN3Oi9vzg}9Z3V}7j
z(>0dSaDkaEvn(&BNHRs1Rsm-cA8S$iO6b@ZlCD@Muf=e55!$5&XTlz8iqEUyeXug#
zH@$`z>^I9@yB$Vz)10%kn@{|5f$g1HaJ_sjO9e^^1-}t9)<LMGq7|aF#(-12cO*76
zh(?-{Z2tw1IN~0>{dI3-+wBixU`EWCX&hjwJf$cGng@^!cd_`xb2#U;k8tHrF6H4*
zeHQy4`zVI{Y`|@rqZ^E2!Ta1AH$HDMIEWaVMhbQbbhSufj{Ti{G)K|gTk9qUV&Yr{
z%1jGI5Tp3z6q7mJh3E5y_nnAwU5<O<bKs75Vz_07B(ddSd`O(GOuwpY9T-hZ7*6*#
zTr0)X$t6iXHybd&BEV4ZKWM~`3O@Dt>qmV3`G3IH?bl%jSpY5qUE>+pgiF#94|~mP
z+4R`oM@^9W84vh`b6);p`0jU@kvXi}#WmJ)+n2tYgP!~ZOh=hWo!*R@2NC%T5&4^2
zQiXVnVi@w*354P2Bh~@HUiKHXddMe|Z(w3Dq8R7tcKXQScjC~8AH(Hmex7|2gHMNa
zdtKZx12rUtEFkxwrg$J~)h(1pdsCW{S{=^>Xz$faLnGa?>5M-LYRpB|gKgiWbg$3n
zWPU-`wCU78b=z2`w1E|EI{{LAU)_Vx*;+!4r5{IwfwjqT#)ngC;6`d;S|Sb+kHdhs
zAO@c~tm)9}LApF3ad4069_PRN-Te4-pXInWzYgwkcWzi*V8h&I5`vpjzmqW4$vEzJ
z59gQ%p1?;x{y9GMPoG2m3{zV+vb1{<pCow!6!IF>ITVM%=HB?BC<zvaq4O;+$TLEL
zYVc-U_9aabK_NTJJGKRAF0Ecm;HVR>gpOZnrC!)c8&$T#HEq-8Y5UR{9|3RC@2Irq
zc!)#{2cc-nQWvC&NiWPEWGR}tpy_tL9P%s*V-?L^*|c=I(qjIkC<@DYkUV!I6H>42
zUVty2`E{meXE13u*s59$SsnS|8tNGf+9v1I<{2>M@b7Z!Fx^?^=OHmuq=O7=lb}zJ
z{QQFyNXr7Y?22o150=XIIA2fJSMxxiWa7a(RFoUcET8zrBKO$$RxIw!_|fh~u1_}O
z#9*xnWoHrKmb+Y81StaNB<$BA|01k)@03ob1DS$zVNktmARRJXSmLM$-;1}u`UGx$
z&_OH=Jtj$*Dbn1_AUE)tU`Hv*!cKPl_kZ!9AN?>pueg@`JnAt#^si55G1<V<)`Y$_
z<vDiVftbn=Ln|I2O6JJ>PBOOLNG&L!xaW$`Zfjj_ZCI@q*$QQ=J?Gm5#T!6ET61UM
zF$=r+!6!b>cfNcU4|?q5xYHA!#BzU{?u@0Evo)nzpplA1%&&V<l$QhLCCyhPt#>fK
zL0Za9KgXbL$T-?ptfX`YBkW*6M*?#IAkL9^Pj9%y!p<FR0=x#-DzS=mLAGpWsgqy_
z%8X1g94(=^LVpmks}mlQH$UKTPF{=NqF%Q+z#t;as`_44J$)Sj+^7>-$xft6+`_P{
z-2NF)<%i$=2Ac=F>3U_joYCn94{)D3tWENctdw#F(Yz>YN>lb=Gic0lGUodvI$m^-
z4S-7BIx&84nN%*XsW-Fc?^wZ#f8wv!mA-M6u(hTg?VfzXiN8B8IOE{C#<d-Pe;s=m
z6s4OFRS_PGSiz}7oCgC#9MVm|3+eKR{iVm~^6U7_(_g^tj(a%weD%xm%Uu?{Ff%hl
zngNMn^M+YmI^sot_!yq>gyT5j1E1itU-@^oY~F{(Bw>DjnT{wnF_^>zAwcR-Bc%VI
zz4s22>?-g4KkqsBR#k@yni&O@6AB0<fdociqR0k>3C2VV2MjFH;pesevzNzycGv5*
z{lmsyu#L&dV3GxxB(nm6KmjC@azuGX(rBi;tLol+&ini0oExjErzOCaG~GNrl6rco
ztIj?5-1qyw-%w->=K5rL9@%$EgM%)$rC?R($pmGqDX#1~wa6OxyEKy0_xQIO+MU(6
zrtV?;O6SgR-bgy-sXp}vYC&g_km_u$4Z)6%ka%orZ&8sT3T@Bm%Fb|o|5SPkv=PWx
z{)7O0K!d+Q5(5tM+%rGFfXhpC(3Zfq%>;A`IBHW{%GVDLGB*6vTC`9kh6h?}D>t~d
zmP)(v2aVTn+wUTEaGErLga+<aC2~g<Czk65b6mZB1gS;mHsi6_@-=+(8u@mi9RtrK
z(x;pYi{I-H7><VYu=G<Hxgk66zLOU}`&s<q8{foX`|pDk%V0f=^SwA#2y{NLCRm=1
zm2~NDZaeSmeC>VjW_as%PJhmGIq3J!AV2bOmNw4Qv*ln<u_`!?le`GhMCy>m>JDhS
z>Q!4)JFZz;J9JeS)rTg8Xe-mX(kw{r@>lArswrK-LKMRxy}~m3?j?NYZ~m6KO^f{I
zpZz5qc?hG8bM%a<taM8XQPm)5s4JS!o7$BseBU+?0MiBTKTFZGG|s6qbsnVGr+@H4
zjPKk|>cd_UV=%UdFC0Y<Y}i;)+O(vi0s9?xFh3fu@Syq4Xi-p%OIl(n_?|TRxX0bK
z`&J$LHW3-F1BUy@#rmlC{<*+Q_OdGfr+1K4HFaHBSM(eYJ<~_G?ZeYv|9ZalXK&>I
zO)vz_gIGaLsecHko$+9N=*MZaX`EU1=o(x5-R|ft0MN9MoInK9;r37C^vI0w>CgV!
z%zmql3B)wSQ-dODsFJI~O?1`9@;w0(?S~AF&T02gS3^j(U8Ev73r<`Nq5`o&uvvJ)
zSc^#v&MFj2S|~@!2JZUQSNYf%zriDZ<<~j-S6|F%Wr&~ekz&CKiS09Ilm}VQTi^UL
z&itk4@TY(NcYN#e>)E(rUwq;jj>d==Qfs0~iO}mUu(W#@X#xY2K#_+oxDi|_vW#??
z$*R3Q89+NlXXWqSEW5jfX>(`FWNyl8K)Jj3S|$TC_GGgPldgNoYJX?)&n7>GRP-(%
zgWH^Xti(=yv#W{0M^SL;jW23r=0GfIO3`V9LQ%1+OyHQSf=w-jb2Uf5G4_#?V4|I=
zNdwtS;|%C3glZaAps1)s3ZrVv8LiauR7EhR)4ZVNj3f;wqAG_$BCW{O??ydX6ROR^
zT3Ad`6$+nYO_-3CMqUaHEm3?u9~wEv4)E$wHP{%K=Q2+}>65z=yLW8odCz)0fA!Yi
z;h>G%SQ+ME3T%I_5~N$eJ5X=1nZg}6aMf9#=hA=sGCOa*l}DZNY>s{LOBrw6pQWt}
z^i!DEV4dsnKxi_F5oD@9OD7>}>v>`7iA_!X3ecmRc!<>E{>JMXr4BPJb;xnrg%uD(
z;l`wy!mhjd_CLIz@1J`<&w1&~+5cHjXE^B7-(cyf59091JJOkNHeSKT_l-`;{-^~m
z{{h}e-D4%*Pon62!m-bO4&S`uhx8?yfD~c?03ZNKL_t)=<^oy-fZU&B8QA9`562O;
z7@|=NW7%@_QQR_GVllZB)N@;wal%4B$YLT5>jA`%|A>fu^nqGoJ|MqT^-<Agb#Wai
z{M7J6*_;cOcqS*=IpV@U{BL9zoyR5*sW>t(h_PXOZUnCu<H4${c)Pyrq<B^j3Y+oK
zd?zV-f?_l+5UcyzRR?6s&q)1yfEg*9tC5W+(=K#Iph6=^wi>oq?`y)fYUX~Qc}q{`
zFvw)M*cz_;o;QeALc{jk=n8VX7FJCV7pNLV5|VMP7ZMi=DX-w-*xIXNY~Y}4A?Wg$
zL2tm^+&s68b9QVtJo^v+9~Pecc!u7R&d-w&5~7_b)(gXxA<oQk&P7-A_J8~kS6+WJ
z^ZRUNsE*O{3f3giAuI&HSyw>es)wOsgTH;v^bN#?s#t(l%0ihFpnd{&mYvP;IejAq
zxj7Z;0)j?CS>>io40k5P=9O)0>)kc?Wk)w5?a4?*`KhKvjlDWGUa`9a>g%q}fk^}m
zzWfZn22&M%Lk-r~XHf0f)C4W<Nlb9qly+dD&<Gn?<qVQ;*x}@euxXH>P0>_yYS5XK
zlxZNU(@=l=4OPt*Epst!EG`=@fLNG$R|6U^4s16ZtE@kLm?o%Hg|$f%oC_AkJu3{X
z!I3dqzKc^HaRPt-N59AMhaOC#D}kyS0fsde@MJ@w=LJSN+<86U{`>cF>$lIvkKx28
zKb50ia0dSH0~u}Egtew}U9w*AIE;j#MG>m0P0g;H9$Z%0O6b(VR?|nTnh5&FUY3du
z7UKR&{f%H6_fts9b}JdF%enL1^ZD1meLII9^<bX-#@EB9ZRA_$<0Qh7hzTGpip%fq
zvsu>IH~ZN<>D#nYAsCw2@d{VG^Br9E#m|whWZ1Zx%cLCs)MxOp-}|rReG|q0o`hgw
zZL%@hch6_%rPs1^$8Fra^Da((`I#&pb6hn+`;YvTKf|x3Xrp@mI$*e;eR=Qy1o$J?
zzwRyS3WuczfmjvD;sJuCT$UH~esno!z2UVSo#*Ii8A%7I5v&V*Z!byYv^rFZ%GIJQ
zkXU8Ppv~fFMj^9`!gul*YSOM%8-a1t-S)Lb%bEoWCbJJF2smB4rY>NhPOjC4t)Z;Z
zJ*gGkrjrN%YJfr42o%?bD77=CS@rU^Xm9oSp2!m}scRmuN}@^CX?4-BUJ(i}^n8M<
zqvtZL_OSCQH|}0$;h_)X32*!zIOS-DbC#snM@R#h^#u-LY3wNIan3it%RhYhlU#V|
z^(<`KN`G;V;qsW3ETi8~G2R7%toLZ8=4^NviB@81_1fwhYgY{1DB{b+-sbgE-w9!Z
zb{cW{5U*@f>l_BU6o)2Ys07YU8<?i}d=<c;P0=*eXkS_d^tAhE=K0c?JV;C{cF|#>
z{(bdar4k26U|DHLDi3~{@oS{3Est11SHb$z!OwKSK(&swHl#V~@vUxl-8iTV-qr-D
z8LAh-O>Gov8o2{y`g6DzPE9SQ)nFJ<pC=7)6(NfWKN7>wJcSCz_7FdUY=wuPa47%d
z58lK{#~sOddBoykA5p=%V1PlPj8{f1EG$62%-!EOn{R&j<J@ue^(>l2j(yUTIR2Mk
z%y8QQ=%$2z2^8WzqE#52kD-cTACz{w#+d5bdP)jlIbiALD@<NXX<VDq%}fQ1`UtXO
zL&irZKXs@N>BS)f0rHVDhr-Q2=Dc^mhfBZv1AgJxUd8@TdJ-$W0fU7kg1?~ukh003
za=)*zn)U|4`KZ7Q$8g6TaO>^3(Ga^h5Bu*2n>SO;EmVPQ)qNWWTqzE`MI{KP!?ibv
zF;xSQ`@g;FpBiKS_<`J-egN0Ix<ysL3_N1J`v1xL4S_xo^auFzAPB-HyS%~;ANU~G
zz4!0gMtGRmG=WTma8yI5RwRqTL?-6JNH_Cp`hhi#I@2g_s)ymkgkef;(lqFa;XX9_
zDBS|WR~%0B6WwCECzqr~r@0lVb`K^DZ4v8a3URZHtGdp6m@url!_`X1_Oxt|I&=bU
z*v)FA(<~CwQd>VOrIDn=sI5B9RiWV(tO*5=7T83Q5tzc$w;OOK;ks;vZBPC=9{2j+
zfTNCJMN{Vba|n<-MUtq?Szck6OF8%At9jo?KFKAQUqd>-kt|IajdFZ3rk~m%rWHY4
zx!x{_MI%+F?8eEoT&xcDO<KM8nkQcs|8zx}9j!#^-LD~GTBg~%PO%5BtdE<ifK0@p
z&zLZjd$<D_ng#RB9jjCR%eM+*&-5UtR!3ZXdtxzjy2cErwFXd{36Uq}B2z)Sc0PL>
z<a7y4I<?0kEgei({}W!v#6sU!c_t0YVcFdAb#ukXH4<y8y`dU37G<*}k%T-SGne!c
ztK>z_Xk~?;f7D6*-fLdPV;=EfQooEff|wNNgZThOfgc9SV7B8{F8su2xcCcSV2c?u
zkd%W@K9y5{^;Nh-4rRPyjx^Gh&@dVC@dozlF;W@|G{x&-+GJF!GsH_xSu~||g-%Lv
zNl)vFc*=9Midm@vesJUx#DFh~5Rhh-Z0xblkvQR&&wrkCKlE`9J^na;?pI!oKX5bo
z<~eNi>JnAdg?q(VL-}aY{~!jy&+6IVH4oF6zak-Z5Ag-D`+}wk>Z8qED{a(QyV_&0
z6c`6;(#SkC)pe8mH^};qh&)w9?py~9_g~QPa#j7ndg*_!)~+f*t75b|e?yjnAyAX-
z&fT2*+Sjpg!wvL{&?&UmP!!P?P(0vbx79SdW2+yiGguBCtI|EZR@do8wb3<ccLXQw
zAmi^4Bb~Zy)`X#JU^?w{tc8*^=`f^goiM<JO{{dL4ZgJ2>`WMBMya><Nke18Ad_!=
zvJ7iKRGYzJ%{>yW$^&pMfrbVj`ap3jC=^ISQDD_Vp%|z1k{-LAV>@uzlTYVSZ~9%_
zz8lEq4FlV!fIxc^M~-EASdc4RdBb+z^`TGm#cy1QmwC)wAFmFl1wCw_WjGi2m5V+q
zii=J^()P=3O{sj<-JCpzYsyfw<=kAzdVMtld9S`I3ajZ<cMmuBS!wer4Zx|LeOJ<;
zq`~)Yx%o^(;oUo6`dzKvy`~bozR?o30uhbcFLWyK*yxy~Eq%@mX+1MUAF~k!djdU;
z&!Ll_Yg!)4RFmOW9QUmIt#g*BRF<<Xx|7!8=n71$ViH5)eZ7~w2InRpotsRGB#zP2
zP8J3|p7)HO=S{!%GLC-G0l2YaZr&pfuqo;teykMBWBUCb?wZT_%wPO9cl_uQ_DK{^
z!u*j(^Q<?$33JRtS(!^2%-JAbSFMBgVnCrZtcJt#gv4zh^CC*hQaPck){F*?v7v=4
z&x=VBantpvov@BZ02U{7B+;~Xl#>(=uDXo>`|b~N`;Twn>92Y<n@)QI%kzEa7UGn>
z%vDf9EK!Iar=;A^dHGsvhqM%ws~|p7z*JFjxR~5s;4$hklGLEK>RE)Ez+jNrG;~s*
zkyW_f{kfPHd4-64@PS+tKHvcZfcO4`z{}Plz*@J&YF<2z2UXckHJ-vP)4S+OKJlBc
z<+%A2Ga6xRii*cMNUZ@^i=e}=aP5faS%+5prfi4_bz1Df^y*foOd52udpKRg5T`5m
zdJXT5hM_%SDF439peL1ow{4R-9WXQjhO8l9LT7-9Sv#ekFfi!>I>C=^gNAm$SQ}0!
zv1rilZ!SAm!9T}iBDgCl4LMSHBsZ9Jfsq$(FrG)e{DnN^6)%EqTUe5WIXed)3LpF+
zjX;s*tT@BXJ2U?I(_i9WKKt)1jYrTQV2!2798w6<ODKFsnhGu}LSR^BcOSdDl{KTb
zY$01u)rOs6E2p6<L^}$Q%2w1wN>Vvtm~umODKFLJOxpXQ3qo}=RH@o)GXTSM6M-q^
zaC5p8cb9Q7;WlcWniC5XYeXNZC8h<cR?*?;?0a&Le#bg(w@&`qw2wA*vM>P{XbH@x
zu9lS9f~8WXX}9dF!nZY@h>6N7;4}v3sY84y+6IY5ixGL@*t&I**Z;=LdBM}4%+|Rc
z3CkE`vB?1SLDb49rx+F_>44(;EBWU?{VQ(x*4H>}<01>G;f~>u$NlE3c<`^D$v9oW
z40=HTUD?Uj(=*ddS+72iRt0vcGcW{NkNCE=Y?aQ8G$xXh*RFj^G!vWSBtv6>oqCK3
zM5m(2Nrno;-Td%<|IBwj`&mwX^64D^^5?_011L7lRevWD2~!0cPC{xZaEKU;vD0a9
z_nU?xb$*hVXv1r&A&x2(xIiD(pgl^cJQ3?#qy(=~7h+1=D=}?*hx>PU+lP!XFW(!g
z!@Z$Xeu}Do0ob+<2<}mXq1C07meWqzr3{6%H=blE<C=H9haZ3JpV*cwnh9xN5R;H+
zLo5l_^g;@&_=&FT^xw<WzyQ@wwY?h1(N5!?HdlMrY(lHzmwQD?(oT^@Cvd0+40S)#
zmql=A!Z0%*LhBCBL|9k{Fz7@G*xED<YXy@NfMGQ;`{b;MCQF-EAz1VD>(a7mO&FeR
zUM(s~m}WT_I%3uZd^&XCJpJGm$z6HD(n61uU-&`}`;{{pZJMW;Gt8L*N)Xl-E(a?(
zuM9^8W9jpyuU^2rKKN;Vc=b&T7PpYh4Hznniy=MZG2YYbr??`=7a?#-y{kl?K89`?
znojGcMYWO!U9LSP=n6Q)KTSzV>IS!UCJaDhVyGP&nRX3(I;qW#g0z}ODBT_0z<4uq
zrza<P+UN=!@4|a@P)KmgDT__xo9;Ie!dCakjOoHE6NW0Oy6Xx%C5?P0!Ae)tPBlfN
zlXDkY?Vf9_VYN2Ze@nf{p{d=-+*B|oaPkpw5^PcnN7A>5brji{;_lr%_E8Vx4X=JB
zPde>nbTlSSpqH4~(YIk;$I8miyIJhbv2^_p_|*UY8*aboBK8wyzGoSl9(N5EdCgz^
zFVcrUf^jM&X&T;|uLa97p01Toq78t6hI|8wqV;BOMRUIg(+uhfag~DITvTlt@ye7*
zt<kz5RueTx!<@tjE8jSePru`x*kq39yydr$2Oopk)Tfukm+-!Ba7c^~)Fw0x=CTg}
z2}7KOnyO(`?ztj&s21VBjU_P_6@w+DnU_i7rVQXTKAIRr3YwY<g!tCLs1~rxjA!0`
z@N0*NJVQh-S_cdd9MJIls`@|HVZgl{Hgs$g%Lk>24=)wS7}MXr#M%GlwQRoWX8Kt{
zDn%SUdW^FKCTq!E4JU)h9@2%UHU=9Hbz#Cmru+9hOpDfpp`><92Mlq-U<|7FcHuez
zCsz4ggO;x7xjSK4y^GouJx|vTFa<PpH4IbpB0AHZHQZ}c1nG^A@+u#s%bD*IyUOB1
zr73WY4=EB#0jOxCuL$H43IV($L19Hn>;PvhyM{S;Zkp$$SG|;DU-3NHxRGUV=$i#l
za9)u_NsZ!s!8l7<QN#Bxzma$R!^io~rI#_30cLK1C&y)ZC>dNqf6zn3;PV`t81gK~
zizhKe+eMi{TA0PHZxn?zQ-^0HJx>AM(zt)>G+${^$*OhlTqv1HqHacHX&rPmp@DAh
z_sI(%rU`-<hxLLHMGJ=}H679b6rhP<im=b6>53#eNney%sqxZ5+;mY+2LnteD0T0J
zX*nz9-BkmV8Z?;7mw(!oZU*_469yX8BZ)Bq9!X5R_Az=+4Bm(Cyzw5J7`)F|886XK
zlxLp)ME>~AujSCKo3Y-JCS!czN&EAm!*2VyJjV|+QWEl;uH`dte+QR;{cH~D&#_s@
zBsi9>;jYcwc=fwKgx|UqKi{wSoQo5oxW@wU>$*iXb17<3J29-U9mZDfU*DK3mY}f)
zPGwqlwf5RX=XC~%;PEEV8S^Y>kSB2aZG7c#-@(-vewSyx<V76v%x5v)wi&a~L+~U-
zZ32yO#j24MrykmHMyqUe@2?sLt%6oX>hnQYR)QbDiJQ`-pq8<CY%;)z;$1LB0VQ{i
zz8G+VGXXj#DtJp7zR`N-)sIN!{(!nf<PSvTf9@?M;obrm`f<we#C0U_Q<^`-12J@x
zJjPhE@tFQaKj1TOd;`bW6d5n0i9u4p1vd)OK>6@p;pLRhd~3ka9QRKxEN20RP8y8X
zR(jGQR&;${>Q21*Nmr}=L8iJSm4$CZThM&#+KyOg*Ood}V%hGP)~loqG4~!P44uMs
zO`lU{7iY5pgLZ!B6o;DleeLG(meq%k!dOw92}Bl;!&r~12|+=HL!F@B({~ACVCTrQ
zvM}J3Uw$!%oOves=02l2OJ8~zz!rg%tg$^g&T_o=xp~(T@A>d&`Q+!n#*Vvpvti4=
zm^5KDTEV%{EjK2-A8X?Jc^@bko+=co5#C$<U)rLB=}@k*9+QAUrc4-USfjQ_U~TeR
z!<8wnZS(sz*kBESVai;ou|+(=smBZ86mVD!xB{^rti=~Oh>gA5=-uvV#lveCZ^^99
z+(s6y`=%L|PB$1#_kinkQsmkas!Qj$PoFik<z0=bZs``(rE!+A3~KOBFvjB4#cYOU
zM&o7nnIG`N=RBP^{OT*%Z_^^;GLls2+ayqqMDPWWL%uYkpBTnh{eZLn`u}kAB^R=f
z8uk&zXF18dWv7BY{CHmc&c7p@pT}+-)U>E_A8MQsHGTFQpiEo(od_tjQS?WFcl3qo
zSm(yqDAUneql6O1HqKgKg9tA{_?_jRexb0miyxo$MZWycAK{4OPvA+fe=UB~Hr%EG
zX-~k1$)<0bUv&1mpGYc+dLmu>WcI!Z7zm)q2N8Xuu;cTe=9}+%KjR%cFk+b7x{3Wy
zemJMR^ko!>9!1eh>BS|oI5c-$^)DYI*>NYEw`?WZG)J;wk+~y|r8wkZL~ISJ8$GXT
z%68wz4F5My7sl&=;eiJlo~^29trz=06}V6{Xe&Q!!HZybj=15&AL8ow{v-P;Xi<>#
zgT<;kNK6_ZAhpo4*;$@=jB2|%sB3Taqto=a4MC@fbS4*_?&rOlFwA-%ZNRYh2}66)
zbt*MU*L1sAqUVZg#j4W0dkPq)PZ+xWT3vN&q*Y-d6T=$S?zO62pG>4J8KQ7fl2DAQ
zqUvxyth-GNo`Sq^7=@%zl03!q2MmjXTL$p(UwskBzU+mtbzeq4p~oPEIH{-b<>NF5
zV_6wG@^rwLzWP1h_mNL<$q%olx6mh<pGUoAoEPMIMz5DpaeiE+7#Ru0g_Vi$qDFXa
z{J<LBW28aDq0V8@s9|U<nsnE@1sJ*-2Au$SHEpN!{VHc7?yH8NR2!C`I<*a;v>EBa
zQd0>IY~DNiPFToutj}1<M)WssV7CLGCJ2J_p+Qk257Szc+Ffkgh?17U!PIQ9ncti-
z>6>N9(17jXo{yd9lFJ`88PYOlm@*=fvO{kvoDa@NQcrqHJx4+hU*zO&i2962oq8NE
zef}9d`$>;uUJR)%NK<gRha?F?^Blw*SScX)Y`^$heC-1t;K$#)kbOy6EGIwYr1O1l
zA1l4%kL5*w{r8MF_A&E)P{kss^Hi5b%KVf%vAUI#QM%&9iGp;(%Dj}0PyviqXG1gG
zu92fqks4bzw5Af=sy9gL@lMfEj&%veg%|RLcfOk)w{GW!Z+s1lKmSO^n-@q2z3SeJ
zP4F4EpA9SI7X7Tg`|jsOQ@Cy(vb-Qa{~LVfx8K4MGJs)$OP~<AQyq6~?DOi+e1Y+%
zO;LXmvX>|<|Mu&@j$U#BX`wiyT(|3X9`$Rl;SqoI$2hg2;TZRM(+;{h{JSq!t0M9o
z5&7)iQuih7MPDNFS?~SZfH$v~{68I}S&i#8h!Cv^`KFXZU-ANOyx{BXzU(qKBq=T*
zfwd$y!R1BhzL>HzI$ihNYO+nSAe?f6CutPYwK{Be>1BFlJfjMj-6xv%Bx%DQ5v^<x
z@9{IFORtz}8QbWl&CoDR2Su``YpvV^peC#Tfiwlo(!Q>BhiP(<uN{tfvskU1VvJIV
zQkZDv=PgD(N^tE0P#5$wePct{+Iq&K;Fg&e;n*VK*1vr}SG@DX9RKuZaP-U1ga;qU
zc&>*xkR}5Xpiqkzj?MFi%sZZQ+HpMnac||?+wS0FpZGWa^-Evn?%g@Izk$L00$!AC
zn1c~gV@QGx4C)mZ0h}s9ylkR$Pz0wMMASx54Tb^g6)`op*sG%wzcxUicDA=nO)y{W
z!o>`UZnwPzGF4G7EuQ|=Ks41Udx7YXWN^3;%qETnGPd4%Cyzh+1o|5{@}<izpjg<z
zvU8|SP-7}v*Gf*?1d~%H0&RJHs|QWnu9j(m4ST$C+pME_9n@DYYh`+MdB185D+|)k
z+TytmbBh4q7`#)gv8WS_Bn_c^j!6WgIe5n~yNja^Ka>|d`+2<ZnNMZ^Et}{WgCQd|
z7HbpKdy-ztcr-$tAj1(!w!+O9eVudO`vI1(x|WTaaIjO1&oE-JJ;5aduE}8kM?Z$=
z{ncME@*X?i3)+%6Jrf%!fmPCVNl8NRRh~lI0jVAdXq6RFUL{)siPAz=%GdCf&2Oy+
z@~v~kt5z1sY6P|5Gf%3(UAJ-GJKxKX&bx?Hp7a!+{nxKzxM7}r(U2k-e6VbeN$$-f
zz8w<R0G*<piNgIGER@#GWkaGb319obhd8>wNP6cGZxS**i5m7z7ubIDZE*8ln8l4)
z8yuu?3KkAJ-~h%~EMkTZqaHuEjb0Aq9-oxZPTEuS`+iINZF>_~*y{j81Py<ws;2`F
zTQB>6;^sqTwW?tPji{0s%V^Q^nE(2_eBw8LlLs5YI7M@TUZ7!!20D{hC9xqHv+h`G
ztDu;61~>WC9cA&V^{Y0zkLVskAR14JJsZHzDkvJZbe#%rqG_Vjs*ld31er)up0L{A
zLm1u+znGazpgQXptyVeBu(zG<fY}B#lV(p-Ss~pIqFM)`srp#!`ACznp|F;qbQA?2
zA85(IP!x~}6vKiIX22!}Z2#Q9^S!e^$3CZ>$fIBJ%gjCYQTW6&#**~=f&SrxpmW2#
z;;iSS!!~p5>t4#6U;Prk_1*9D{*Qc?ORl&YXDsuBK6%o^We!(lNTHYna_{K(dvOY9
zf+cb8C^Qt6l7wuBrmIjn-ScS8)1I}BWO8fhgvxpH;+KMQX<UQm&sNi*PDj1T5RrB{
zK7rniwr7^97t-B1FM0IiIHE{le!z`e_hZNMZj2q^BE`g)MaHzPN4qVl6P3uME<vY{
z)moobRsr)9xE+s)lMj3yoM$AYmWxTMRBs~hJ4q6pcU54SnrJG4qOL%kg!9uI@EJZ|
zCSMtI)Deep`V$_@OMmIP9I@ZN3=H&afvCe6@XlaNq}QxCief~+*TZkWm5V=r7T^Bl
zr*OC5!~y9X+cGHJ3U<)Lh{c)|XW*7`#-UGtI?w#m|4ELgw=jn<IsGo=js!Yjo!4O+
zx{W3+tLul;-KTXDGYxYJsrOnr0ghIzR)hDycQGR+jw=j`8cXI2><H4V;QG&ffiM5l
z-*eaz59ZhY_kV{Yj%Ku>N1E7bqE=FE#Z*|PwrHJkyo#>&9Jqfr1?!_<VU(k{+)h7p
zh<I`wh4&<B+*`RPyL(r#T{T)6csU2j`~pirKM{%|gTx@cB*I7AvKW?j7UMMM>HEUh
z_eJC{_qLjEZv+fCs_I{;>igD9{(EzX)DPLo@MEb0d6IM<J?KE5`v?CoXaC6`bBGME
z2!323N#K&JQ>>VZ&QNvMs)uysjZUNBG!Bxr4SsdvbI|CUwqZeCP)&1eojguw<S~>T
z*e302I-Hn(BI=$$hjxU7(|HxEzpU=@o@wYtRzn#Rp?MvGjp*!VwzR{@rv3HFVzaRc
zs(RI6yv4Zkk6c8u1~t*s-Is#8z=yXUye~pmG%@7!it`TdmPmSVsH6-p{2u3=|6Sxk
z`|_|eUc!Mdd=_lo%5n<Qo2MUkJ<*hjjRR$)2+w@v2|WFgC$Vdo@uh$NI-mUFSNPUt
z*MM0dP5Rh$5ON@_k`D`#NaD`(oHR-3+XQuvY?y&~thIRWu}1Jn47`dS@(3{=t18Wc
z_j&2MR5r^S*(&kLthAfxWuaEnYlLYEnaa6mW{0x5ZKhmX2{bDe@ZvCDz$zkR7Odiy
zSCGDCd9*}f9L{P0P9_ldeHmxbAv%ZOWSVo5iZd_Tu?p5%BSdM9RAkx|u>nphS05{Q
z7aAvuLrAJZH)wf7HR^r%cbg;_5?q{^^m{4ZIlK$BC7W2BC4N{(QH)txUgChQ`|!LM
zKAY!1>nR+6)RFY8AvFc5&9P#!rWYq_;X$9RIFh2kxE#Iq3cmU;AK|L6o<q8OnFD)^
zXr7{3f$60rX+n`{5T0AZ&Ep}*KkpZL;vfGB!)!!<aS*gPaXm~7-%`=ayB{@eAT>qj
z;+uv~YFaeJvWWDiUVV*~yg?8jbV0;v3%;Z!5~dk8JgbV)h`#Y~`H%R*-@l7HZ{ErC
z-~1M~JmxX1%q`HH69(d9%5(V)3W0a+iH|<Op5?2hA4wgg7NHr5VPWgO6gS^Wnn03R
zoPaS(=7yy6Lgq#cjMlDiSWFMR8)Hp@$6=+17>C-l>NKh`f%>FtdW78<>d3zkksH?m
z!`=ZJ-tWDCBJiqp2yhRl3Qc0+wC<cJMjeVFU!0@&_{VVQQ=iAJ|N3bT><zHqkvSJT
zI3Yx+42LR5$}`X9P9|AJ?Jf)cW|)t*L)p5HE#<?z)j85haO78YCz?h@>a_lw?hi7B
zzt1Y|lZo-l6!<1HyyTlMA`{VUHIP{g&Z?dn6BCAR*LWHf>9oFORm3qJc&)bVk*n+_
z1_V|;Oy8NRNGF`k)CHe_I1@U*8CYvUJYt|Q9+wp)7`EDZ^tNR#{l|ae8}EMy$2{XH
zob<|<z|lvO(I*EO@A|d?03ZNKL_t(+Iu|pj1YanFeu9#-Fi7~NXZ<`cdcl+V@m(u?
z;q3GH?7yDP)i>UTqlZle)3YRM$z9MQIOjvh)PhM<dWj_)<roATBaqaAmoiqiOv%-D
zhnBXZRC`88>|oa|(l&T*W-<8a{1PcjG4WNUP34hRy{(Nz>Pco#ZFCf=H>!-ZU}*qn
zUwavkIpSEZyZu_O9Pc9AxDcZ8jMNmLDeC5mbkxdhkxCrib`5K`hINL8aVPgfr<{rL
z%NPy%iWIWZQoKqjCP{116(=?(eqQgO@I`n(-eHqa5cdq=JlV=JIU~m7F^3$qKaV@@
zas1L5zre$fJBEP~2C2gsPv07XH>nAq$rWe@e%WJ5=`AmD?fGBltMB<o?!MwOHl>Db
zl48al=aw)j7;72jV-iV7dMU~L96#Q*#L2(>Vov+r-(#dX{lx*^hin@0iZKDqsK}J0
z>w5p`l*Wckv*3+Dq1;!#A-ES$pyFwyL3st^YZ1H+4JTs+H!LujBi*@!^WOI!e)QF^
z^O&bSjc5GMD{%X6BHK71iFqPEhFt^(nZp_jQHxRLM4%5W(dVjQp^1lmxE3RU{ZD%|
z*Iaff`z4m~xFDVDQP@DwEYdl8+cpIuI;DlGf&+X;GS>&m0uUbCpg~H2NXwJ^&-%Vc
zMBcx5HjMYy!dLYnQ2>49It=*9cac_<qCgi2>F|bek<-8JZq9zoTiEBv*O88fm?U%&
zQ~G41F^RRUaYfb7tBlBX!XmMD4bhfMJd>=0nFpo%5U^7f&ogok>Y`tm5%o5s4c?dn
zv<mdrJMKt9b&3!C6wStjcl>HCnVNc1Iz!c>=RJg8-7_TKa@(oDH)Tzx{+mn%sL~K?
z>m&y$^uJ=_ah(C7tpc%Lf0-CkREgKsP)O^LK@+!T7C6m`_ZYRH!ir|>w!-Fzp2Vpy
zdI57!c^uhXLO$Qe8jGbLtQY~8gV7vk6fc%M_t;*-EqCnVi|1U(=e~3<7hirgE4jlA
z7U?DPU@R_kG003Y;t&cfJ``#(c*;qEH(EJF5u3GP`AsOY2*FAv%iEMxEliv^glen_
zBFkKbmi1^gJ@eRDJbMvbWhvR+Jkt_4Xk7=5l&a8pcX(fr`hx!MCDN>b`2piWkD)b)
z8N|@7Km#*Xs#%~Yc@xu2#Dw4BeJ!Lc-JT*ns9Y1(a=mn-u$`Xl&~_itay>&+n<N2a
zGEO+`V2(KC5H9@QW!!$(62n4pr9c}Kxc@PoKu}B?Xi45Vifj~hUpWsx>M)-4*hh2v
zV;{rAj(so-{T}^PFj|l%F<ICJM=&dpkMf`o7!QL1z>U{)(I-F2_s%|>V*7Ts*#UDR
z6nP#dA=cnhgZCl)&q@z(9Cna!-SUv9z4<L1cjj*}^q&5F5)15Tb|ExuO>2l?T0wx_
zJecXgb=q{wS4|d=Q4Rk6hDc!z7QD{I`!`M$oHO+E0(RZWO<(#7pZ~~*Iq=9M`GwcM
zp78;P<2Lk|OKUq;^-&QKU$6Y-sxF!ln7yq!qNPbvfPze+cjFE|{km7vzwvtdet;im
z_(8&wnP=C&+c@*nA7?bM^lboM#(8kU*M9GJ*?r-;Y@T0aWhv*j-McvPm1lC|Z@-1Y
zhjX^{>y!Joe&L!ZbpFvgVA%UW!)sLa9qSeUPc*KfavUm)313d?P4K}V?WTX#jhyxR
z*Kq`fV&x7}YbY|yAWhK1SN5L4AxN7h-8DU;>i$jhlF>bE^N!0&6)q!{U~k4AiCkRi
zSSGFoFf<aLXLWJzAz<h_O05krOcvlxz_8}vwmnVdCjdid!Y~zt><KV*f9*s%R*Uwc
z6Jdbb$mweazO?~EWf2^Om(sB8RPiBg7R6Zb!vIba!bD=|VM)gn58A|u&p(4BpZyFt
z^Z-WbJW3Cf^f2Cp<Xcf3o-oA7P}5_jP}C;eu~hJ_@BWZae&*k~@OziCYh^@#ZV@vW
zP^hPH8Ad%u9f^iQ%~*p|7e@NICowj31Wj0r+_^yii(tV^01-BdQHyx*)o@rUTTHLM
z3fqc`^2VCFYMSGbIN{kd($tQQp=qsYUFmY8xWZvg5Kb2Az}O&M)lgo0uOyLj6s8z1
z(wbt$p$0FCb<NPfSfFYh5alDvaV5xTPV_o{tKHbOO5hq9QX)bESmW4dvByW>^L7q@
z&=&suZ6D-aAO8d+--9GUF}TcO9Q1k#0u=d>%q`)vj7^(2aPmV=;K@&TEKfQ8(HyYP
zM&@inA|An!#QQ9<He9QSda_|bQV7O*itV>@)%oXe-X}lKjw`QZvomZ?6H;|Zk>d-`
z+}u2Q0T@H3F1Rci3pPP}J+8ZZJFj}%U$XIuPbcg5=p~lILC`fv+t6~qn5MO8JCC4!
zhL!h}w)8NyTWVW3rqv`^!x`)}3{b@2oW}wVK`dm>k$6usRF*FIHeY$y``Ep-!V6ya
zUzq>7lNl{6kn~MOR5L!d6THV*)4t}a(Xus8V`E6%{Uu54r@|zNxfP{Fu9J|&F4tVm
z?O**8JFdKrVr9hKmMv^O>Ug$1?lFuHJqXiF!XEG*s|M$VWak}_?S_1eI)|vkEo?^i
zI{-0h%UNt9i&gGR<&L~oMBcd<_lEcGFYo<_ftRdT{yza=P@SZeBk`dUC<Jb}LUPV|
zeBn>t$`K;u!)0tT2;*cII*T9;=a*Rg#V%@P&DT)HDpHA;r8QlUmM3@pUdo3RRcA>`
z>Oz@P+bMc#TUEP<j{7Q_0oCSGO@%Sj07E+`wj&#rc86V=o=`h;x}r*xE~T>&!7QEg
zDjJ5_E=<$Bt=0}0WGdKbg0gOs0276u*V%4XW$s2Rg<0c1w6whbRKYuk0zK2iW}zV8
zJsNShcn&`8R37@g7qa;HN5Ym(48>r2JuEhG|6O<uN9gPu^@s{t4j9Ww;o=`%!Dl{u
zHedecdEC6?Zj!+~>HG#v(nkv)CV4(4giE*wN=F)}!KP5+eUQ5Xj9`)oC<=%7N}5`l
z;a6pcFLaT$Djb>|Zi(FcnB__->X*jaOkAlJAF1Y{l<RB+5*OCRt0D>H;=u*Vh*3e~
z6hUL5D!vSDD=2J;6&4jj9#jl4i`OWS)oc_JYp`)FX*9vD0kjk(H}npb>zG!dY2r0{
z>}zoAiPJ}e#iCdP)+qnqU;PRDZ{Eru{{Elyy=$)}>CNGsfF$5^aAS&W8Jl>HICy_f
zJM|=<_Jqgt@P{4CVt+2!r6xIvg!9MP1S<ybz!-3af>Yemh=B-<ce4GW3%TH|PjcgT
zE<u)8*pkkX6afqr-eE)$qonD8<>46Hvxr*M7ud1GC5oAw=j!1SuYddB&_C_*jOXV_
zZJ9@4s(sd^ga%O7OmfzShD4_*zM6{GTG!B;?u5ZnyuU+ZC#*Tu^)QJE4#inc;yhe=
zDd&FZ6I^=!`8?r_7jVqWUct&>K!2l&v$XImXgW|yeR;!FTO_<@%(t-G)vxFS^kLpL
z31~F!A_dHeVhe{zTAd5t#K(6MCX^N%K$tP8b9iF{t*pT!3~O5IPE3R9+_6%<Z&j%u
zF~+=fuP*oY_Ig)UPgd2>0Eezu|MzU;Ow%$3>O?Fc%ER7-4qm!@#7!Ui6qmpI|FEAx
zo)z@_pk6}f!V4r)1>V%@r6#zEn39TA`oNbDB;O(UG!8!XRUunZTBH=vxd=$=?vhTK
zE<kyx-RnKp0SvPz48AT^*XYdDEyZQkM}bz|^afRD58$H4VbTsP+9P17lYnOf252*$
z+hDIW&)AcprN;eI7kQ0|1NF4PSO47@QZ#&h%Y2Zs5cV<Kpxj5^$Lxf1%3}prc*J?E
zw=9r=ri|Q(+ec%Dn+(T1<8)5><rl$m$G~8L5h+Fu)|SCk7F#Ili%8P;V1;o3Ch3zI
zxZ#!^eC6zK@#S;A&Bfol5@#Vv`y_S&O?s#)LLuapfF()*A(46%fwv2^lf>An81@>3
z;l2p&JW?e>M-O(>#0zENTqfB@NUJiX@SD8i({_N1ASWpl>w`?#8y_YLKAa~eT0*&a
zhZr#dEE$J4K1@zjurAoLsuBtn^@u8|Dyjx0#Tkb)Ipl&Z`SCGXR-FfH$b1O*l4yxq
zuce?)uo9{y#f9mH31WIvD6S}2$%iZ~4oLb5ew2WB^lXpB2t{77@4lOP=<&yJ+RvTL
zqfa@JgZA6XK$O%f#uOnRB(*W6E!=jA2%;W0EU;bz+8mGAe%WPQ@Y&CB^|!u(?7V|5
zi7=-LiOa)&HeveVfJ6+1Y6#G@30{<<aP(~wz|_JV*Deit!@J&rdC2kPn>UjXu!0dn
zf_is4B=Njz0R}p1k$Cl`PIIR5y$T_8>TxLxOHB3p-Unwf^@7hcY;NK1JNe-UKgLC$
z`y5A|d=ig+?W^E`gBWidkR}PLK7g$V%&XLdc=l^&np;k@O{AN3gW>ju2}5ZJP(>9<
zaB5S)nFt)KeV|p-BLI36)5$$fJwg)cf5CE>%5FEVay7-7(M|i$|GGg$&JdA{*8#&i
zXn4J<zI}Zq?lojs<<F(Er_0ro?slC2*59YN;2UhrJyMJz=>-8@VesBn4>+o%Wn0mk
zs;6f;RftYEQ*&?|u87S}eRIr>uI{o_f`)SZ-t1#fcSo91$Sa+Itm!wQN<Gn#89l9<
zlGKiR)-}#=Tlu!lLV3#XQdV(NouygUDfFRk@a>)eLsdM~4-wtNj`NX5GdX3c-M55x
z!f~Cx-T1v_9zx`rH-T0Kt|a_iNojEep04wr9)dU_K}pR3;~hiI+35<*e*1CUQ=h`&
z&wdUZbr{U`7>c63lvEO|sfEmi5)!Hb)Zsn&T=1#ItKs7BUB<Z=T*!InU&6(gU&C@X
zW-z~zG#%i33yAIEiV#xeT!Ar$qR4|Nw<xL`#F!YIW*|+I09u`oy4fK5_ujYWKcNV1
z*r;mT>r1D2=jw!|s#YMwN4i*2tFQ6VIZ0!<T{TG)U){4Pc<TdYOij>qctO;mB8*K4
zI!la_W*PH-1mht-g_T~9v7f_e0$RinwYVNhf;VY!q4J7)Pm#L_js(qxH0hy5LD~aP
zf%ZJcCM21lR>|`WDGEeKZ0Pqn`p|<p@r2_!`NU&+^uteP%ce~jh5lTMh{t<}F;Y`A
z<6)N=iz{-h3Vvls-zyZ$WLI3yMPK<Wmwx@5jCb6@Ch4<i1f0iEpdM^uD1v`$;Nm;M
zU~olLPPhU|varT;DLR<r>ap_1cm5OpxZ@}m2c$NV3d>1b>1|wQeQ0^bn|EG$t)v>~
zwsH!zX*?kl<$85<#))!B43<i3q9SOnBsnnJ$!%Xco3DQKBlLP3c<O6j!{Vu@1|EGc
z(vP%m!?aGwP5@9#RZ)G`**2;V+zCTRZ&G#s8-asJZ5k5p;u>hgkkhVdZZr8r+kr?-
zw_LvORH!2I1`+w&^{Q|ED-;~-)TEybI83Xb%LbGWHh0ECmJR9Mxx!hmeiet^eHSe4
z2AftC14}3#j16H_k=-0(c`6!?k41{F9zf-Tw%t8$&?ri~z0w@I0T`M|p|hp~Re%l5
zILX!m7<7-6r0Ele*`nu3`p2}OgEsAEtr`Zsmv~Y4aN_HXIYw*3&=9&elRu~bUTsZI
zri-N8g=;Ivqf9LpgZfZ_s?@fS0u`^n{QKV5`^qGN3gaY%iMb+7uE=06QA}Yl^cbsU
z<TCCkMkEJp;poSo#>1X_20Zk5==I46qkEh5!fQ(D3hPl(Ts}s6NnELvf{@m`YnXA}
z4L5P_dEe#SZ(qa(mt4i&D?@COFu$;oq(2X0A+doX;KDRR)yIsA0_Vf~uI`O;>Sn_G
zu-4X7x;SOQmw;R2Yf%M`HBs4g(5Y3ruaEj<qjmS$N2}O6^_%FN)9FIS>5ur(rNk+s
z4oP5)!6Wp%uz9r1{!2G;uy>e#%J;ox``o_x!6GIBpXbptzF?R+at)TXMm_yrLcc$t
zD3o5>LkntJmu6(+6~?UKT+V!&a_FIlal)|==G2Fs$m1XVa1Pve3mfKoSR9E}@C71)
zuIarZVsXxs*dUQFuelo+^hkq|{mPJ?*ImWMU;H;NJNIi8+izpRdN$dF)H#Sb2r5$1
zmBc1dw?nYubpeR<k{-?}mI7mQd}7Gw2i#bAUiZNdQS5gJvd<zGC>$ix4YaGL0jg4o
zRkam#SLZR50ej7fK{=i9jS|&#I4R|vFHo=)HHOR=n9P%oVRXTF`0_vf1EXC_Jo&}X
zXX_JBr`T^F+#to`u>?1zrW2mDG!nYU!btbtC;E)qy=y2@1{7-j&z7-7Nk!7BkEE$K
zA@@C*$X-uHS_cf1h2hDn`YGV>_0{-UtZ%eov_ihJoAlNloOR|aINGG-%R`JRNt%R*
zqmO>x>M<mt*oj432uqa;D0uIyhnN`Cu`aFMZ(2fF)o!gi(fQYkx?C5AvmY|E!NY2R
zq1#pSRY5y@!q8}WRG&rJ(@8t&4m1leOiP!Tb#lz23QYip85wR<0YiHvRBJbM&%WX1
zlwnd3-uNA=-R0|C32nV?U)r<Mw&yK*+oHj3%V{2Xy4E(XULD${9;{eWRZ^c&<c>@n
zyYd{rX(I=naw^9@{h2I2_EgxofiVu>>wySKnv%qF$`z3+VM+~!$EXQKJQjL^c)k73
zA(vluCFh?1UA}q2C0up&O>E!2M2?}?pC{=J==BDew2w_J9;GOJOkofE-xWpZxO*R{
z2paA+YfPCKhfOS^zOo-K^$b-vS~|gL60{K|h|n_K*J~>iLgM}0aulnkZ%S<j8|gwC
zY&DHY70xg=ifY2V7VN)r7r(GD;)J~5pyLkXQ#aqi7k15abJoYEkXWHmFk(pt18kCn
zX`dIyqcN_?$;M-{Y#BEiky^(-i}M_Q_=7n7po2L1#1nYr!ym>G2Oh|V#Rd9_rSKVp
zz9osSJ|P55YFpp<o}&*hE*sM;lq7S(R%_P|etiD<T>h<dx%Q%q(OtV(v<dU7(ULWb
zPDQa{Qi3%fY(bq5phq-VtctbdD2Xv4Xf1cRo?+Bq<o3A%uYdph$hK_5MzG)-Zc<bZ
zTW!N;(Wde$W`IcL#T9h{rM`rU)}usyf^_r&Q6~Z#_cS5b0>zIDQXE`%8Q*%}N4e&b
z%XsuJK93`xcLw?5CUh|&jr=Jon=fKoX3cAEi0tJB;hLI79d4)S>L6^^6LUt2(S7CX
zIuZHBy`74*4j86`hIQAZpYqV5m8BdDXD<%Fyqn~LAMml?`fZL~T*Quc2LY4~LMZiN
z93PT<qgFv(H3|<8t1yLV7K$RKvdL_UVWm}wO2utfjiEzy=qjZqBNo;s+BE8#Xjrrg
z^vbpqof#`YTq|AcgkcIC&{Y6~HeTnaVyn6;C27wy<!Ms}Eeq^vKe-B}sri{?m5D}K
zL`Ww>TN@Lpjs&`1_hvw#51_TOdyG^vFDC3&4OEuE8xI;DBZ|))Rs!G1AQ*2*wMXWK
z!onTdh@r{Z_t+yi;b~9hpeH;Q4nGX$7Vwf%7>g5wNh}FOOHpvSi(URA6lW;L_9(nZ
zZ35P!!1m=a*WY*>-~ZuN{P6NCxcrLix&Fr6*|~Fxk;{<8lGq-D`31xpL=0(?M#$#K
z$9a^zDn*es0tsDf;?;8U(i#ZX^+gRLRfvIlO^SG(nbG99)qsI8eT;v}0LDwq;Sj`!
z_cInyXX*PP2aT8cjiWa5h<?i5#S$OA{uVA+JcL_{E%eg_qXk*!8D$ySD8m;S&Sw~7
z*u2jM_T9Rb!wx%`ha7VxM<018#~g7mhaI>-8x|6bS5g(kh3hK@ymusN63(&2l4m*A
z+OT@L!z#FOL8?h;<BP!EcXIO&e#oU?J%?*9xPZ~l?aZrV(ZZZbutgR_?h3#lHl%`=
zAlp|cUgLF28hMF9i&Mm@l`(=%Qg-)J?%KQ`uX*o3GG1ID-Lx2W0?{(Hw9D4gfvR!4
zYoli`+Q@RQ38E@dflwBcW#&%xRs)a<&Uz+Jc|2Am(oh{b_SiyU$1Qy4{U6}UbH2%8
z#~#n4&wLg7;KLbjTnM7<2s|wk($`}ifXHOGeEsX54UQ*I@@8&#-PmcK=CdH0bS97u
zta#sBkG<R#X&o@^;mdpf9^lONmH25NK4l7U=!Q8QcGsBgpZx;o{q_Ii7(1X?8ItzG
z(c{%4&LGwVA8Zd|eBG^z|MTi&k>-QDN$h|ciJ@yPN>N?Z)B>w|3PGpqtc|4G(q3+D
zISlQ?QYVVuNuGTf5~q#-UQ3r}wqdEv1PrxxCY1pE9)Sf|10`wp5wj{KNn|D%nhKm6
zYudC_onq889WXSfKIJ}WB}m6_G+Y;(6Ntc&^I!~43{EWKJ&6ygwkn`D01Ne@jKhP=
z93F=#^wJdd1wEUR6oE6poO@O*uyr&0J?x<z^@PW;`BA69LHptQJqj;ql7;{<+b6}8
z@d&|vNfl3+78I^P)MKK7fS3f6TFUIRG0Jr}+`{!Y-^R7q-^exB-N-f9{Fs|=-ND^U
zE8MYbnanwy7Y2g`Y!U>rz5X1YB;2D-nR6MxtbR&Zmqs_^lFnrC5-2JXUDe7vpvfcm
z%}FJS*CI6O@qxw@BZJFg)u~XVJqqU;6a|AFyE%RHA}8;=g<F>H<m}sS;p%iNyR!is
z=L#0)b2e_=%n?T&%;5(e$dLyf$PtG=h{Fy!fCKm6#(b}bsFEZGm*=6eq2PQ$nkL}F
zbjga18U#hvVMKxk!5Z?F5q%YsaRJW5XgAB3U(QWGyqF(-?d#le!wvNEf(5lK^!m7>
zK#byC9>T4pbgrV+BEXnH)p8LuY3&FXE^?rrB}te>6%Ne&fLlf*dJj37SHANf$Wnt@
zoD1}$NGow(F^z6dEm+oQKMbO#3I(imVGZK~FLiE5^+P7CPZbOwC&oU46OFowe5CY^
zfxB+whO^G%+<*N%8}~nm$G!BWY&_`^j5qY@r4n@tAsa%KU?W^kM+9_j$JYAudg{5m
zwr9Mo@)X)r`{+Il<lbwHSr7MJ2MnuD8IDucPXG^DUyYyQx;8dk`GCpENJlxBzwPh1
z=0hLmh=omfw-mkQ;{(uGkYLj$(XOJ@8<^OM6cZlOUR~gO`^YUf4a?NVgrn->``{qc
z6%uRX1Z#?(JFSN)RjI7<`DeGts|kZJ>0Y#l)aw@QrZzL5c`R$QaGoN1o)to^Q*_=<
zZg5i}-7fC;G+}7`o5(5?5}7hZ(Kf*ATF*|~d{gm;#`N7dc~NPJq+6e}>YQpTt=0nt
zf<_8R01ORLZz(<{7I$7k*qZU=c@}_guFRz=eN8Zu;GIx-W$cxaEm*e7;vole$SEgr
z@F}OT&k4uDe%qk80jCCMEt-Ha7Hcfhw3b(b3z7|mtm5%S5!=XuN`l8=Y|v38V#o<b
zJa?{S+_rruH{QB~n{K(C>#o0%n{U3ATW`OE+qds#ILa7~MhwTJppR&9mZf1$Dq=&I
zy-Xr5ZIMOW|Jt~<O1CucH5QZx=L?(`!P3+yP>Pnz-{Y}KiVPN5?9a1hq0c`3K8N*F
z4%xPiEeG#UvTX~S4n3CrwjIDBTNc=~XqX!WzP?c~1sE-Y9s(c!X+yznZ4ifBYw=#g
z;L`|N+Lz`<z?b8Up2P`)Dl9Fr^V(~<;ll55%>@?#=0F+0aL4smVMZ&=iLz)?5?_Ei
z)Ms%q0%b{~C_Hq_gE^>b-N9Fa7+91#FC-T2&KIP;l3wL;3U=cLuH1e*CqDO=IQ>um
z8)GA+^K;=`aKRE*y;k<M;X?Qshft&?#gy@Ox467%Txd{;jr%{^X+|--b77=x(jJZi
zS1A2lVVHB<Ip=Wh2S11}GERTVE7<bn$CK@|51MKS2}Xj=FFs~8NEOsq4W`#0zyDzX
zV;TsU^>eFRX8)&u{YXS!AR?EoU+HzguqV*)N>zQ|`bzw4fJ0dX7;DMhm~{7;AO6{&
zvEwUWVVm?&mt)h=1rPOe(`lp8dGuxR8`3|^WLNK944n!EkczhR-J}LC8mAswx<HiA
z$I>%8(w5c)7&_mZoUSywq*II6CIzN`@=h=0J0Wyae^6GPFu*+&J<l?R(zQ((CiM~R
zpPN2mSY5>2oDftNy7ve$)MIijU!$_#m5Ha^ENYl&!>=Ez<l(T#s0F&g!L$+xVl_~5
z1bhre^Uf2878q-Ton`?>QgARX3RWoaJwtEnCJs9J5gh!;liBCQhr(e8!Jv=J6=&0s
z^N{vPO+r!zlW9FTQJ^R^bY=0uTGTtdcNIz$u^2HJ6SNLuOrU!R<W9+Q&u~<*JREZ8
zu3g-{YneNC?qp?o$gbT>EG-XNUS47M@(R0`mRVUDGaQe}@`A!S3RmD<_?_O@*0|OP
z)>=}V(Cel2(v)65Wp2=Cety7UFksWB1vV@$uwi}!^9yrq-mrmf`)p*(#swDU=b1}-
z^h6n?LA31>D11&Lmdp`9i(!<FF-hp+TU*+hnm~)BP60<tTTek<9?qL_LC*?ioMBv`
z#jtxB?!Jp1Kf09dS6{(3-@Sx8Z~8GZ9?>t9J_f5nSD@-bdNHw}H-HaQn*<v`tc_VT
zv6F6%sZNZb6$l}BYC@W{SEVPR@sdb_dc)W`lFf^J|MojL<9C0XhrRL@WW5BNB(?BZ
zYa*$O_ma_=YcVbNT4Jzaq?t9{>WmUowc~U`LeQK>dsrVWb5#O($egF=gTOn#=wiO|
zk&kiLt=oCR8PDgS=RA|~h7Fj34cAC~)PI=jKrhAADh8Z})aKIqSJq*}S_?w`53C0E
zI7$D{S_hF|5s~+=?}&B4u%<8X{oep@SYL^s>P;9sT~yxP7-5|6W-!b-=ePcV{NnGi
z&46`d$g0fJ001BWNkl<Zk|f2ci-Ngen~14<%o4l*K0d6<0GN^v;3B78VmGi&k7yoM
zO`=g<;03O9O?BBbO{g8DMx#idNFdgh??e<8wlx>JDqv^;k=BG^dNsdR4Z}oHKT8GI
z)iO7@>uUiFlb~TbV3<B(=)7Y5sLuOFBKPF`oAoqoGLXFYjpD+zJEd)}z;^0y{Bwx_
z)SHIulWHvh)snbRl^|Qbb{Rw%T$*A>J-QKjA5z{@1nVUh?3O}3Lu*;m5j(~s%svZj
zIpQFWJM~l!d-zFgI{IkXx<8l!xCD%l8-t6fZW2%l1{(kh6?3r4DhWD)kk_D<*toGQ
z4m476FeXe=#55$yWtTioW8x7|8dOaf0W=I%q(S|PP4|x8p$dShEI$PweANs=09#i3
zHe?WpC~550d+#x6Lhc;;|Fiez@v>x9o$q(;9g%tRO!wBkRkv<U%_Ja-3YH3r10aeB
z4uAqG>T7xG2f=xu+jeO6^?T25yTSRRHaxXO0YL^4MFf;aW<dlIM24Df-KsmB`JBv*
z*n7P{_Kt`hk&)+AK^0UL8J|zpJyT{z?ih0Y*7^+uDkCwkf=DaFE8-MCEy={7<rHx)
z*piCkH+R^(^%gFC?BksIz=znl{szupcRj_<CRtfB@Fhz&BgaJ#`Z6SAi$}5XzDON@
z7NCkrofqnbsDtk6IiaF9hRG>lh0+*QmCXB)wNP3n2(q@qKcBmWAOGzau=2&9k6yAu
zm2^YQCy^jHK)vE)Sf#1%wP_P0Qj$V1$2Zf4<giqs+W~}D6SkNjUOhNheBm+9!`M@P
z=%0AQ3;u-b-}inV@eN<kgP#6WbnOJ?`VeD+W-XIwzl{kNz<@PbY#RCC9;Tj5EU0aq
zwieqbTnOG?r_<Z+BX~%I>eILVx-sU57H7o*Fx;Lg!<DM~65tVw)wpv;4ORA<h;T63
zW3*H7`k(j-hX4A0E-939QZiVw_?XUVT%Z~Gs*`UTYrin6_t6P5d|gc%L`~p&V=A^u
zZKo+!QnVxmU2<v6+NvfU=}yp4<D%1J<=Oh1`CdQu^QW{7Nvr@{!D$_<qIsDKb4u_o
z=l>{5Qp+AQ@|m^s`t3}JPXI8aE=P0ryWjrSr+zv!f1|30gi0ihf+Pon<QS)OjxM1}
zwUZkX(@&e4Hgs;K3Vutieyiv3t_idfjf$4|JEMW>6W7P9#;gP(%d=1{sYY56ICbpq
zPnh7@njA1Vc8n8OUBPAdy&w0v_Q9O~?9awtc@-Qx4tWk)m|Unel-}W^;8|=InL}Cd
zze&P*y+&cINhWgA8jQ74#WhWhR>4b36Kv;V)hjJYYPU2<LXsB~X?9FfP}2ByRy6)*
z|06K0_Q$)DEVGo;5_JxoGO!k=4yy`931tD3J@lp<IXHJS=Rfi>&VKA;T>qheW3si0
zJ2)UKJwq`_Su$E0Q5FR@&nTmyoVr;O?;sMI9Tnx#U{I&=ob!#%C>`ox)fjO;PJK)W
zqY;P0V{ML~DoY}izQm7)oCnTG&QJfrpD;Onn&HXi7_3E|@?DGAr8;^h2{Ht40$6Qg
zcq^fQyk?>;uC!9TD+}H@w26tqVg%=utnf(b;KTpSTVC>KeDK}x=CiN;QoihazlX`C
zm!ZoULmRa}T7l>w0m-6Sg+u_Z>QUoSV^9o=!xEZJ3u4lpUqtqOhTPV)rn}Jg_agEX
z5jnFsvJ1d)dqKk!B4`*bR^tx5qr}SASE4!%A9&O84#SN-Ui<84vwr=Jtg2wOK#j!%
zxv#ofHhvIAZg{7vt!PF@gM&yJ6Hkp1OtffiIX0-rh^a-=tr(%CLXHMZhur9TZkN&&
zOnsK0Y3;9?xFkuN5Zt0F)t&C<>n_l_4d*s(^=|T>J2b0VfFn)tR^rKBPZ$zcs6#{X
zdcQvcFlf&{M`Y!@Qykg@oyJOa{+OWxjdcK-%+8=%$E`YFQdcf%Qg9nCKlSe$ZAARN
zk-)!}M5L)5s#}E*jl}7L$GKp48oI0k<O~NH)>jsmmJ*m^80(b%qGal(Oo|drMz*%h
zvCA*#iqF13ryp=XPJPzBSv!3dTzWYyFN4j(e`4d*L~)e>Tm(!679s(p3$e&;MLl;y
z+hV6$dX8mRQQj5KX!Emr4NbC4osbh3#n)QdY?dxMqj`+?;5-@C$Ri@4vczp|v3>Ry
zw$I$m#gAXd#`QOH@unNuy?85rcZaN;k`<nT5prwEybD&PWr12j2|;^7>tLwLe`2xU
zOcFHaU4*Z_^Fe!3YpEP6E>1}eo|woT0frTWmZ50(CX9OQz)%*BQ9i;=3+&1gAKlp@
zyWjozzTf?Q4n`wJCsw0{Yz!8UjTYj)*b5kmf_3%rTDMUd;gd@y{B5`mxNz*X4a|bB
zz~G%jYzE>eJ!n}nm}YRx&Aj6eU&Q<W?sYuy3%-Cy{2xz;fA?8T)^oBvhcJVQLi%_w
zGSoCIL0aDbhz3el+(V0C)bzy%F1AC8!9IP#yr?CP_}$%TJBVQ6)r+%X0T@0}(D1XW
z`iqN|xWfkw39X>gG#I7_dkk*5$g7|E{an4XhuPV~WCQF_@MUm&Q8h^#r)lUozA3Om
za;l9dxxslyQI?1SD+ZI9Sg`nL$=DREDK*AxGH#cI=bPGZRMwnH%Tzn+u5NHmeG}IH
z2}5&%&T(d%NqQy;Jd<8=Z+heJQj%Kf$8(GN<fEXK)JY4>Xp6cdJqLYW4uc$dSL&TG
zs5afqp4XfCAk%rPQcYe`EriTYLp0TXwEThUbI7#L?evgzj&M>Ww|6VPn^;(TA0`Tw
zKRtMCW`jltm`n@2&jN6AC8aAd#*z<mY?j4L6C2K556*>((cvl7aZpb2z<4@ku(HbV
z*b1XdPIB43{~af;xrVjVrx;y!l9kI&A**Y!x&ot7m`=nA1(}UNq5`s!|8E)!mZv>2
zYc7Hg;jCvI)m$HeDpY5csK;t>SqiOePc1so00M9trUcUhcD5<DcNlMMa`Cz|Otv?;
zaQ%&J-g=e`XU|fM_wds(1Me9q3}QE1ONYrbTrnlfb4m@OacdeIQuRWX+1gN{Dws7+
zWh9E}y(V)jC3M{@fSEWDd1;y&fi0!@if$9jcb^Ha@GKjHb0y*p6C8ulh=1AK<lc|@
zI-dNi&trEoX7!TQTF~6O32T@jNS$jLftmU^K`_l^p*lA;y>dI(al9rC!LduM#i2+s
zB`X}<ax))(_3L=&YyO&RKI?P%^6&XJ^gj1xI$UC~Bp8W}hsw@XrNvfEr@iTYyAsX2
z$|OpCJq=xWB;A&BP<!G&Zy#W|8%-11`zP{qBJxX%^I!oOJ{iD}tLlq@CoNXvjt2}{
zbq1v=I*WrX2G^hEPrv7Tc|g9zbZ?s>N^UIP7jeXFF&e{pY@9Syoo(OET`+aRbLmRF
zIE*zx$cY3yM^EEG(kWnSE6%!!Wn$NjR>!!qD3zqs+=yut3e;@<To;zwiom?BY|&0L
zcHW&inHJPk0}(oEiuMQbJP~n<c(el;=Fo=vL*bG_L*zET+rA%ak7a7RCZEAkJ@heO
zSwPkl5Se4^Ds2jukmNu}D|<wHj#XP9AGvwj1{CSE!K-3S5Xc5Tc+#DY<^U!eV6CAn
ziyF@v5y~=jrkx8K4-@_F)rE7|#DZQ-a4_;Blz0k+Qan@VabCe#R4f?7V0px7ZIyg^
zg?w#=;o2(8Cyp^(Sz&Zyou#!^hU;tOt7~K{%j8Q-$Z!OM9IORn!@pu0af0cYE=3JN
z{jYZ+y{;&5(<yE;r5qpN4#pJw2TTrjne6N^+1X{hwZnApfa&foTjw_z@9r?!*+<F(
zH=cs`=(J#v8BFOInwTlCF3toTN*RiE6Z8%~7S0kw%G8Id*g3^yHU!AkF1$h1>ow4F
zd;~Enfs)hoIHY}rNHlxEP>(g^&889%TM@hl4NB?4xnP|OkIjcn@|<^l<X?H(&;Ms0
z@{I3iY)XbJ%LJU5P@~exn#6363W~(Wh!YpjX^C!BRc}8WpXT|(x$ji58mXc-+GLj{
zb|SEUk?UUh8vgzzFXq^>6MX$Mo`HPsLn%+K!ctzh4YU$Orv?!+zf?`LR3vSnRD$fg
zd++-$G-<epuH>JI$kRlmSeyY1!0^d}hA)bs;eLzNxZ^tr+O*d(CfxkTJ3EX%b|ZiM
z{ol{$ERV38J4opnEP?Z(ps~Ih?^kWHM$nv!o2$=WjkJlvO>o|kVaZL7^&T}9XE$*V
z(x&+tI=nV<2GQ7!lf-*GEy(M^L|s%?P9l93u(mqrhg_mF6n#yjCgxJ~yWwsrU}*k8
zT0J}MebH`u^SlXz%)Ae2b>nZ}grO<2`zH+H81+$-8mdssVz+nF&}cjQ>}K0N@Y$S;
zt_ZsIA*)T+MtfLP9X!(#uvY*zr6+}SU*DJpL?<WjT+sGFO;55Wboxai58VjfHDY>S
zYYHkpxeIG2$%&*A0EUE+#4xdo6OJnH0Z$<wCj_x6fnlhs!LbLA<XI@DjSXL0i)0y+
z<%qSxohYH=)Jx+P=Y4Djczoe-Wr;6ba3?BVpaFSTQ{e(=3qZ(fpzQ=QjR~$xDs@hT
za}JD9Y6zsm`}V{wIXni51#IdT<XhLBdc~TIvRccSNfE4}N$wHVm(4U&RTG*`1H?^)
zEouQ@l98cmP+#ID*s$g%N62xmq+A+u!6|P)dyb#^!xwPnmpzi{$S@k%8n6iM-1nA6
zaPo6sQKw?Cnyt>x*!{2BU@=Fb^jf<q87nvhnJU(K*gMC@oBp2HzvLxM3*k}U_1#?Y
zm0!VlxI#7(vSySJ?WLs=ynjNUM83)ZKdjZT*lyEmw(pyAL?h>J2rxbp!NR*1M|%Mn
z?kLdkJ*xWr#Y)@>=%OlgedWFptoC$wlk6im@TWiUEbi}fvgrYtDewr+2-d~hSp^zQ
z6}+V_rg>E`sl)<?7+~b5;9y|0MtE5%B<oIFRX_xVr^fDDl}<}-MHo7zb!b)R64RJI
zRo8`eb>A*SZxdkXma5$Vy51XlLg9!7V+$};KuhQW4XM|xw%gyG2}3nWIdn471q^i}
zu=KI1`sOL^g{UY=b2JR8c)UOOO*<2Y_WmFJq1Ofs={on^M>={68h{E7VC@ug>8nUA
z%c=*9m_U1}3+8spczq2>10aI%IEejS)bKRLZ={{L#&S4daF37YyD?RCVknB6Q(5aX
zT9qhnuIPQ?y7n}Az^;1^*7}&{lnhgms{E7i`Vk}wzUHB*RL|=>=Y7xzc;`R~I)N$`
zy2`~!gBi739;Me+ET-#jfAvlRPu_VS6P+!pWph1=w1n5K=TG!Lw63~H$*}<pXVwHq
zA<2LeYcdXsG0V#<>`h9{Xv7WMJA80^kLSMeb-1gpBwJs`RyqxuDQ8pEb#oSKgH?Rg
zC_wO+P+!J;8yl}Dk2>Ep3dCY?IAbZxf?;7`Typk}Z{$sX`~oK1dpzoC-@(<7c^uQr
z)^VeZ9Bo{hG^}gobY||t3BaGhAJvC~guCVY6M2S+ykK#p7l7do-MshD1%70)5_bY1
ziyx3-+#d@L<I$Zx?1!)Cl|T4wK1&U5cLzJjF?r~wR*9Zzr#a|IB}sFRlLIk$bs=zx
zP+WQMnNEXCiZwQ}0Kx~X#Bd>vR1S>=xkeWpt+~cvyhKaNz79Bv=(*d~le%b5DM@K2
z#G!ye^)}tLmByh-NJP)wVhvh()3a?NB^5o-#6HZWBz3YV>K`YQpih*NluQQNlq6}b
zQw<pA*m<O)^R{;3sBu~|;UMXfCpC2P+C6}wt960azBVG*_P(_}EKPX*_*m%(;p_LW
zoe4*k`fW_y7j4!-DrcwYZ|^k_zkDSy55yZ|Bb~}hJI^B8w$>mo&HP9L5q;g+*94+h
zrM|Y@g0=o4>J{n<U3{F6;NC~wNBp@Fi8Msp9_wl6*lXw5CwUbjO-4aAcOU|WW>92m
zM<|J4PLoT$FZZN5Ya{3pU)#f0#j~g&E(HB~57y-5`4T$J*(!F~!}BjUY_PX>oL~RD
zzh-xBh3wcWd6Y}z!u92&*QxO`dpn8W_&SAK>*Q?pLDU=s);73D#RiY}3YiIwl(HZ%
z6{ZL5zV)5F>5pE*_Qf3@@sw}kn#X?~lfiLhS;#X*Vq>NPkP@x7>r983x3d5YcVx+b
zNBrGAYI~kB=0_Lj$O16j8B>N+s`^6UF^ko>V{c-u8pdmr8Y+t0+9tp0W?u4ten0oK
zIgEG6)sve6&bv^si;etXZTH2!Cf(acaOvxgnxRxE+!Vx<+dS4jCeRl&=o=b;{|YQr
z|AkSy@!IjvNeX&xmjxAHSUPPR)efnpMKS6W;c^6E=-m5fMbAAGh8clxn$~=jZwAfZ
z&k~P2fT2T9k_IsJe`M(#DPFg8-%m+ucmDN|>pSh@%ZvqVYu~!-E7}5!eF~1M;O}O1
z%;RTk_po=4wWL3iwxd@U(5QBw-^|Qww3XYD2(f(-?$$+IHHoPV2MEP^;~1436Cd;s
zmF;aeY*1RrDVex+&SPIs6eO7@QENsTs!gC+S!2m^8^M#b&XIap=I2bq>eT22pw$0p
zJD{ZH0H?iAqun#2cD>SCj4Kl8W=0HNJgP1_$Vm;5j5lBvSAfY!?2ijJiYYr8{M+_<
zPCfWqe&Nqv#GWXFW9#*+8Vy#lYJ){JLMZsFU9JosL|f_R)noyqQM_%#6br?w3CiLu
zBn+0$VJ3<x57>X_yLsIU{wEi&JHun1_^n*?jZZ{RA7{FjW24iO`ru&XwNVKLEltPI
zNERi^rw%aOm0#gb+5SdEzC%QAU7RBez;I`QhDWLD9|2b`R^!f^HfSwW6A9p9va`q1
z4QKh&AN&FCQ;r$!9pI)g99o=$cMgjM6M})-I=7}<DjG#`v0(NB&SZ!c4ko)euMD$1
z&=wT%p3K;$TbUZWAVEwmjW>B(GHPcEAe}aOT5EJENzy7<=3lwJfT1_hxbm-$p6V5C
ztEa`-QZPt6$ZjSFV$OtN)=pI#QF_Y_rzyaW0t|AyCk$OUGY2qqELGdfA&mxw`RUgh
zZC%ZbsJ)X?tqFXYYlqu_wRAF&PKeTj*P0%*Y)W+&FeQ4MG~1&)eMtq`AulD+?bKHV
ze-r}8<K0m4qy|q?PZ+9e$0XKhiQdO+#{;Z#=xJXx2HP&m#Uy53$@>;XJ-D)rT4|3l
z!Q?=sSuZ0pdkxKi&Gz|I379nr<Eu;D*5D^Ga!{ECgpy96OnvQy#)Pk3;HqoIXqc6h
zfVCsWB5W38&P}Gg?S`{_%X5B&r~mBFuss>GvNpncDx?*rA|~9QeAKK+6eXwoq6rmD
z(rbc{Wm6ecc#T1r0$M6&?2vLFegFG-<Da~c>pt``zU1q_p3i;SQyDLxfYpp)R(U#w
ziIP*T)2dCShR`%uM`{4G*zSyS{<i+?9t#-Gh{)4L<c*8-V*wcMY|!v5RsD^{O5E}H
zG_AUZUh6Jm_{=r<y-l(kZsirvekND%?lba=@g-uj;8Id`h)pcEsxG2v;As0unTWpZ
zq2>^r6-w{1#^6+$Ob<{MhJyiCY#7H&G|P+~N*j@b$2YVC<D*bo>Keg!=L1S&N7#H?
z`#K=3v2pE*QtqvF8&skFozET|k*;l8+@_$VrcqSIT}#8jtgY&y69zigwF;f0s!d6n
z4ZoXlDN5d4udQkiXz2T;CY^5In@=IF@Vyz!SoO{8S7P}d6juwXX8_@hZPJ{5juWr$
zflCHB{l{Kg_PkX=)&Q+`cv=_eYC=D1?Tv~4_EAG5oqP<RSnQ_nTU#Jn+QDtTW3mo4
zNjv+O9LM(lNz4h0LT?|!9ILDYb-rT(JpUuI-3Ul$MD387@LqiFv=$9^Th~-I&C!@_
zppju^GYMG{#>V8i7>wrwFf-P|(h|GQbKSX3w!PzH+gtqS-+v+ZecYocMwUzh7_bl=
zrBdBciUvZc$L#c@p5!!_w}|4VAxFWA#~VxG9Qg#yq@euZ2YJ&Ap3j+o`7mGnxX1Cp
zCp`td<}waea|Tc;Zm_;dW3Q}ttF@@XaAm(M?Eu9EF=;UmxI<S!<k=$fn~QT{0T@0d
zoA>@VfM+dkFn925p(=)a_v8#d^yTgz*$p@I>K}R*SMMK?9ZXSU$uq$h4l&WirFJRm
zbWo#xD#;YXlTb=*2%(xf$HC+PF^17#sqV05G5yq3lMB_5By9vIRWO&NeMTBqe{@vx
zv-PyEi@rG;hCVh>T29UISNbOms(MHnJqsAxIdN%+I@K_A`@=muIx|t}j#(neivP~r
zKb<4B3zqvn9@~DvaMXmMA24)M2Q}>^ON+Rb3XYRcrk0q&CnDz=>P@Rqm)4Z0nM|ab
zofdnhl}erre2xT9I{&n(AK&OHXr>aaetHj-OSKs>;kWalog_2+Ud!Q@wm_1QE|ElE
zppAo83ZOd2yrL_qPbMHOp?(SilRTB~3dVw4T2nS@Ics%`^UCJA)-EMk2bzFEV}o64
zO}UZ`BbF&Jc8(j)pJQqa8x{_hkMX>hz7%)$Rb<CUu^FPVlW*#C!AtEh8i;7(oBN+M
z1iaMyDRoWu>JS?erKOl+3kL_gxc9%0cf9l!-1Ola`20sdk_SBXsZ7={Lsu<>JWLd%
zN?@+U<utawP2eK+eP0M>Ne(3=Qlb{yV!LCv-!#TNd$9})!0@S@GF+jmf5<&j&v2BX
z|DBIR+H8b&PJ`RsBD?V{FMrmvxW<7T958f_fwg!o@MQ>hv(W}sD=Wb0EnW*DHKyj)
zE-+e)7M%pHZc#c4UosdBFven>Vr>v0Rvz~10{_<cP#jueaMgESXTpSh6#e-GFtn*5
z{a0=S80NHz)6RQpL0eIWIvR%F{&or!j?gebJL$Q9!jP2t9l+2QcN_*7dNd4kCJgBW
z<*2Mz-xU`3X(yz|rAeBQzt%ezTE%X|-8T(vn&TAPW*5y#x2O@r?DV3QR4$$GTRIy*
zmn+_yNc9|tO31HAiu0i_=v<O@m!642GnqN+e3)rHw8On-vkqq_8r{K|wzMS;)NOXW
z;q6oww<%bu$yJ?+PSx!fiB`8(szBvnVHzFR@P@(*G8AsUu*uCETi9c(oY~srq2Krz
ze&Bh($~ey$9y>-6rxnJz;Kb%bV4jPwYa-|})hbt_bYiidfNifD3w5E=i2KrErUH{4
zcHaIr-ux#oWaFbZ@Wqe*TJHa(r^3CjV7xXWx6N7^Z5k-yUZaSQnK*_zoK7Kx`_+IY
ztdS+`$6;yli|v#7z&*-x_8vKj_!|-VE)ltTaXu^n!>1ZFJTiiYYd-UCH2O(D%S<;?
z`w>H5PQ!G2m*Lrqyz)nXm{YskTsBe0d)ws0U~#H3?$e2HKIe_pLPLokL_V2ph=eX&
zVEsBYf&dTfP4+=yb+i(vA0BIL=;M_o#u!}Pl9QhHZKNJ3^{bM)qeR%y=ArjQ==A6_
z8)u?s641_f=+hK*+<N96BcbIIlTH}=Y@E8pi^C=!DPU-<X4`<_Hfa!AfWaL45qRkJ
zOvfoF)d{pT41Lt8$~j2d%IH=B+J!s)@23xnr69Dcv(?UoK{^wLW(v{RhPI~`%nk9V
zfmTI@N{Z>`T=N~#^-QC_C#;TyJtqxl(YElp*1Yrz_@!I@x7VwlFm_f|54-Wqnu3Xh
zB3HWE8&YKvv}rRcT|m|}A|TL=f=FE2xhN)5r<{Pz@K(5We}kKDzJ)1<!OA7vbTH+q
zKlPJ5?#F(ZJvU`}bt$+fiHQ?70!7Y5T93x`ZFQ|kLEmb^1Qmf-twS`8F(It4&=3f>
zcPqQ^cn5#?vX^uI%sC$V*vIqvPkb_yOHSig3`<#}_^1NzByB?^DZ;6+S){q{q#fcJ
zSV<;9af@xSeNx*E5iI=eVks7Y;nM{gzE4$uXR#7@()y-_swVZ92<7eu`Q`y{c+P)7
z{`mu3IUM1NJ+j>5P~0@|ehmiaL-J5X8)@73#U6k3dv_JJ#Dvkfk0C}=KV>|bFwBMw
zhr=j}b_^tHE*fY0rU98&<s!j%KCyolk(!UB_)fts^P55`9$_}{E3JI`d(+xu5Yw43
zq+I<@3UM_d(Z1le!*;kHD0HAj%B`8x<(9)JNvRfL4h=`TMeSkT^gef+R*$&b4^IF?
zRTN1dsA)sR`5p33U%MlIPPjs~Slq`9huVs>66v;pPiN&i8jcxSPt*X^lNW8TWBqmQ
zAS8*FBeCw)SpZ53WYl-gqggSxw$53Z1qU;LA?f&Q4`}WT<z_j6-Q!`}&{8VhRGQ1u
z28$Y<-h8P@X7El!F<zE&lHdUwXpw3lGmu-(TsY57=Wb<de89@d%UHVPI5%I|;TL}Y
z`CRpwuVI`i!z>H*A?LtHnk0fZ3H>Ybkt)mT*1?b8Iu`=hLY{}>)Zoj4+*#P)=gjM0
z&s$&ozu3R9$-}?lN!<VO-$Z%MDW=N<2GXQjTTNV#WL8kq?gb=r@d_x^C?UNun8k<S
zV!LCCU6Jn>k>6dc!~!sUS~u_ge*=DEu@Xm56h0~7kTj)LJz|A&Z=Y;)%m;t%x!myD
zzvSL)$C>W!g7IXM;hn=(9X@G)TGs4gZ8~l0T^(#x`P!2Sg({L|6yCEx-lK4imC-7h
z%@ALZiwOXv(lwY$xDBQfMN1PZm3q^M#}U2d6Z_XbN>X>~Qnx|-gUed9cI~(^^-LJ%
z9mKu~gJ?Hgtl_hFi@#aGK&0}-LN4{76y|q)dv>r9*7~HV<MvG&Vm+lZ!EN(J*J+9~
z6`d@eJ8cp_y*V340fyEjq9+9!)tM<k9b}klE|RH3Yr>EM&N+!pt&G9c6mj6F001BW
zNkl<Z;YT~JUkNR$t#@I@<*D;24lNB1GXaD?0ZDy%v)ab14Mdg(h;}JeeZZg-t1)8&
z7Ao7e2b#tSBt|hlikM?Qgi2#QTRAGBiuVv)pe5LxdP4zBL^!*%!}-k(_V)G|sNwXL
zSD@?boGGXLzc2oMmLBjR#@PUq+jzc%@h0GdsKZhdv{I&WgcDzfB31*?IEcVV6e4Gq
z!Yf)93>DZu&&_XqGw*ompHm)qzU;~0z!i`E8m22J(UCCBsRJ(S)CqxhY+*Vnlo{CF
zm|D$`I#gLq78ctbvi*uN<|h}+umB8qg(<^vRsH{fZ(fxCciwBx7Ei%0u^jJ`?N9jl
zAO11#`J?~Ey$4HV(|xQLi~;9BOdf4iwN74@D1s(@Y*(HoO)(G)6&HhvN))FcVky+K
zGucH2R##TYVp!@R@|D$9&lU=5lend5w2H06vAl`mK$oPZcA|jOuI0>!t4~-oZ3mrJ
zftwUgRW~{mlCzxgR$(Zup-6ke&`ElpnJ~=j_{Ya-ZJlV2On5%(Ec7*C=nGy;5|U?n
z_}vMEp8*W5jDS|b-w{2}Jg&QsI+r$fB+$|H+)WW-I4{!aS!(y%4SMT}()Bi6+lK`N
zN^3Heti5XgQX~MQ`P+o1*E-))YEpkNVz7OsGf_*g7iL8i$sz4tPkl~Y?F<!LCry(y
z0*ponAkmor5*h`bkYV7;%1KSYf>Xh%;H_}W?l#xma*my;Bh!rI>ucQm%B$F&I`Vs6
z!w>!Lb15!6fh;c}gvo+eS9AO;5N|3GvxcsHn3O8sQQN8dsAWJyj)Hdr(<z1l?49F-
zfBJIX{pyz^rQxfe_-&ke%;V6@*O;yiu$4gCH^-;ZC`c4|&jP!7+9NrHws<=s|6*Hg
z^R`!r$iEkn&BanI0K;7cG+e8yF91G&F+2Dayn0=Wgy5i}oRaU9TzuVM^OoQE74G9K
zy0?emu!AL>bI}edb~>X!Je6Ze7|J7RYP-Lx2+^3}>tmo?;0J>j!=99Ej&~VkImee)
zv8BhLWGZ#%OcMu~@LYz}2$YycO$8V#w<zszhR)VZeS+5%G$ap?)mW==o%x9F6#Qv%
zrHy?gEg6IM0)}>BEp5xzj)p;!%!heLUgwGSs>19D(eq(l{f71ut#Z4?%wcXqH6^M2
z&bxG~cGrE*n)lqI=chb3-!wv2>Ev|Ck>D#$y3m>HJQ2&!Tirf<+jFe6>qb;_>V3k1
zlVp<EOh|P01$qln?fGQWxd$&gmq`}@RxU!#^BW)b%~b+t_1Ma$)<oc5Pv7EcSy?6v
zcnl6<3OJiF7C5)J%Z<0*LOCf}&sSI-j5vMjB+Hi^=Obq?@USO;BaeN~PcR+ju)M?|
zu0fzQg**gbi={&$-%zpQ<i>=354NR23@rxA0<7~G=ito0@xed;E8g`tf6eOJr9Aw}
zPvZ2W9>rv|4$GFIX;#O37fl*Erl-2dSzBy(x$U1s<a<Qq9~Mio01S6E(D1FQ`g~yR
zo(&W}4S=E5kPN*4%1#llU<^*DWV;9O5AWsGKljsIRS0G}#)!uXh}xPmqy{hs?_Hf3
zt7_^(K%$>{9dza^8^$owR|#{8!Wvv=*eed$8}G0-SY>s%f|M?VzKI|{Cb&AS`QSD|
zJ-#(LNDFb90Sv00OSg7DioWF?zWA)WQPV}2j;qUI1!LR2Mp_ho?WjasWn!Pslw`_K
z?~u$=K#l|qv=+wqCOx+SNpC;Da$<>{QPn<ybHC`kK4g*rhIYj7kp=Et&`{B7db9)G
zJ*$dwnF9c2X5y04(%Pcl*6XFuT|cg?CTZ5DrGYr|E|ECn)S_xlre5*d?$YKuUnsK<
zVf~<?Ghoo5S*i>D^ePZdasr|qXq^0&5lq^6?-6ToWyo>y(9jya!B{ArQhJZcG74+h
z6353c-paY%9j312)cOfd<m;@e;l%oJicGoH!IS^PbGX-&z7A&$WO-bVFv*ZA=Ot#d
ziM6rwAFQFh^B8NuE6xd-5Sk>$Q<jibu17at$J_q!MSS>8Z{XAwSM!LcJeAciemLdj
z>l8~FgXF3+9$Yn@h`I<#bO?)WvE9YC9T9nki2UhdDHed?t_m7{N>zV(u@ZM4W@v|Y
zHQiM)!3~7X9pw5O`0JnjY1TG(SSuVlnP6=OR`JdSVQ8rslht9vUVQVsVxkRZC3mc-
zMWN8}@%2=|i%@1lX_U>$KD(0xj*pJBHeA6^r{n_GM~gsJP~YepRPoqoo9?@6(5AfA
zHU$6@5BvB<UG(*u7^U6mwA0<mVxwK{cdbghHl30PUDF9e6UfvBIh}%ClH=I2V(kl{
znA1_#`81<0$f!%*nHGCR>uG?O9#U+!T=Xk@-kRDby}@niB&aPO>7!8f;*xqXn+b(L
zlJ+4zLiw3If&_$AiOqdt@K(ZiwfaepM0efmJ(-yzbl=BJ>D~gGBJG@sYDN9b!4N#N
z(Q56}tv!`#O(`U`0Bw*)HSoh%z*r;Y2-FHL3G6dsP%qeeI_5){-vr80p+IgaP0oeA
zea>9G$hrL;hNB@X*@%;?YpkoVYK2pmA7jTWdnYgB+yC4D4cFWYH(CnPrc}Hflbfq!
zYQ*52#{~8fKIWo%9|h|skRyvi8Hj_;O{VXEAMg5$*YeSSd@m39+%Mvbp86!@fe&Cj
zTqR$IOyd2>YxVA<6|BKFp(R1PQ)SzTa<o<FVq0vVqIyB(IU@337b~#<40q|xd;eR&
zGw=DO`=^hZ)Xr(3CH{pHx3fuh>lSbPzy1s5zkG<(c0f7aV~~ZvH!2uk@#cedpVN})
zw^|z<MRHQ$Cr6QB02f-1_>D0XcnT{xYnh6&b+F4pG3L_sOBslTqF{(6_rZ~bh)uK(
zRuaU!1}F%vZhj*s4cY-GK2AVl;UEUZ-G)|QBM`0&XsMoO4jpNx3ogu9wI;`4?sI6T
z=x+tI$t)P`n~d}noRzCl<Iut^@KSBlfJA5fW1D{UtfS8S)Cp=WMoiO~3m8&bksOk-
zp@*FBEx=JvG6JDo;}q;rz>p+&w*bU#aE~GZ1Co4vbX|&i!{~bI_wFlH$4|Oz*2Pwp
z9cL2qm*URu5t;=svpZoT_uePxc&L9(3_6Qo&?^{Aia;u31B0RT1=d)m>hXET)MnhU
zagOURoTG4_<15EmAFeQxoa0N&3{+Si8ZNnXjk7zOT=Inv;cI{Kzu=B7!*H3xd#o75
zYZI;(C!8T?(nQS!R6b75$LWU+lMtOMW-^7{J<h-L&AjKuFXQ}2uIIkjUdzM2>pS7<
zdow+`f-L3KvmLLE)oLPI7d2696>pm=H4TfI-0Q`**gkFBZyRHNV6hAfz;Kt}VyN5i
z0gwMo>-~SKpQUcHMvksAf_TLnhub|M-`L}yU+{Ze_ZNS~HP(<#_Axf-3h)Ckgk;q^
zI4ydai;n`w_&;CK7s7u|93qNUgSyBKwnB-=WtP$!CgM1^yG@C(zOu?%Ho{D%<f`Pc
zt66CrkXZ{%N!q-smE%e%5S(#|6G~Nx2SKGs<4RJOs1?m4Qa`-+jfP={gsgSPeTLGM
zqO=~$)QpB9Dbi;(476+%wIvwtPXe1aVGv10&xaJ;+{QeIG>8j&Om)&JSCSOo1xTsn
zpw&e`B$V%v>mW@6@4~p6lGKwipuM}@1q}TZwBFaBaa~G4S4Zru%&8)^p|;fkMmstj
z)t*UJ*P~V9_1fpM69Gm2_f1-N5C%&vB=@4IxR`HYY3B$k1|OUI7S%F5Yn(EwSe!zc
zXB=SJoK88jy~+8D7dXDW&gu1IESnr1PdGkW!cLWuIF?2^vOeVF`(qybEl=WMKlvXS
zdyicTlqLk!`4G4!@%$3OS6^1PxQS~j#v)Gf2PGK_8)vxX^{?fvfBx5OZ}0Qaul#C0
z?`cn=xbjMh%rG39VC)+6V>}_5SxjsUsJ5U$&}7;AxcCUl%tFAh*zOwJUx>)}F2dOs
zfZ^^08Xm5yzYBccVm0o>b*?(5b$W6VR%JapJ-}{G*m%urdDrj#4yRSI#Q{<j<k=FX
zmN;*Mc+ff2#MXbpX9p9-&{csHC@3Zdy;a4!_mJ6~QhZ2?7DJI`9N^eI*kkM_oLoN6
zTDF8Krx>RUj38y%bhK+4P@HWjG}-}+kH2@u2UA7`qYVdLRhuyg&HhqpGFsx>rU0q!
zU?sIc)PzG{0g*Yq!=ok)oy68=%Al>{@lavyAxY1nLmdjMo;<l$2-s~4fNPG&ZHSdf
z)>L}zq?pOEI0_se2^d=YrHO=IW+8aT?J9j#ohW`NaMAYbi*$%5{mIvKW>#`kQg@$l
zU)%4hZ^AQ6NlF6LX5#=NFb$JF69q|LptLSd6AZ=)Vr<kdMXeJaD}rDtai|)+@u;;p
zn^Ril;~VF>e&Yg8mD9&harx>=@&kvsDVacKGL}>rS;O*@!DXIX#qhLW{(q2%J)Fr>
zPG+*|gbrZA#xTlil425@+R|Z67N$&@rO*<cc!q_*tvB=jm%f5`{rPKH8(O~P2~Xsj
zZ+blL)G4MTLuLhAo$Edn{LZINUxDsat!D+vOntq^a5ySQCg%9jF1E$?DcjyBBHt$>
zZ(A(I0x;a&K*P7jP`69&`KOV7iXJt$PEDG-p~zR~Fqy|sOJr*g`M|&OhX4M{jBY*0
zWs)=9-vjHBY=FAZu@5FARhQbwjD*mc^y-6HTD6stpe8sgnMhHR%1sGyqF{51+^|c*
zxxH-;ih|45PjW0@Lnj9e#gY4@kkuvuRS-}^y&H88Fj_Q<i2-s|>TI;uti~g8pSnd|
zl?EY55==cu*n1la*?E9LT2!OH5;Oq=B7INTx>e~RWC!nk9a`2s%e2##*O|YPPSP_?
zN^oy@-fVC|CEV3^d%M9*w`0OE?^oRkLu&S_hqy$wp0DL*CG$d_d-g*UO45AIMW2(E
z&Y2i>-<QS!G0?dIeBJM*wnvp#2v{{8tmu83B(4iSI0VH&F%5AD;jg^GcoXg+G5TFh
zFiP+Wr4L32emG=mGtTdBbL;s{wsyBUzPiTgOD<>K=IH(|>OGka#ds#dk{w{nlGS4)
zc8qXv`6-_AtIuP4;!<>VL?#CDK4=n*iH2$wKrm5wO*DE4PDNgnvh>KfAZH5KeT?_K
z_!a!q-@civ?)v~9`EB1x_648I<oFuo!oWuLhmZF)X#(-2_nf1?A5}-r1v{pwZV;rl
z^flZSD65O@GjiLA;cN?$^8zs3Z9&72s_Oq(ti+uUQu+(-Ac*$rup67m%@=t0Z~YGE
z-}zQf*#Vg!<Gm+C$uy7SbwCVK7kDNL{~}ks{x=Aq6~QRcn6;vnhta(e3nEN#xGba0
za&~>m`JHXXZo-N66C4{3$z6$Y!XS1&)dvBiN+KSvEg(a=?**K#DjmgWS%aP0;i<Ax
zL^0T!zT+xarPwKKr>#OKmX+B6vo@2d`jf8xXWLn)<C4_LL}=M(&X$FT0*0zXE%Hg)
zs@9HErtQAdd*_r^A+23s$%+;vX}7#@3ULGmYpS-M;I|YK9nKtY1-RV~z(5*^)|xQL
z+||6j(~}x7Xy|Hp-&uD}s;`qXr=Gq^eO{AH?IQTcggRM0);lbf?Wu$-uX?W)U@!qN
zc=33ncs0myg<UPVY4ZYSwl+{vE?qg!<tvvm(i}b5AqSSSAdsF;_gP(8#uf$QVg1A<
zd~D+a4|?(wdDwsaDfY9BY&67RV?Kg!61d}eXb`*)69EMzGiV8=E67V=a=_#R|IFY2
z!3+5KyWYor9{do#^6B4&eaM3tudQL1Gi-GwrYYK;^1e$OY{UEnaErDC>C#J;TWpK%
zQ&<s*{FsP5cd-%+z;L(TK68?w??RoW<O4Sp*}>+t*x4sPzs*Nq`BMJnWiRF`F$~KR
zEyiSK2~eC<93WU~Cl(`kaj2A07ab$swBg+w3ATf3f)kQeAyWYte(!(+hvq|Ep0VR6
zoY~)Dr<}6Byw2&hlPpaNGFLD_u|D?jyaO9-9Z`=V+QVW4-&m{8Y0P-2=~SL(ilB9<
znb9z#2=BvG&>?W7Q(II?t3_!SkKLMAyDmvGqZ?>LgdSI(UP@9IFw6z!&?8o!0R$Xg
z8EeO>N|{rw&S(g<RfN@VsO{r<yC)3oG->IM(%J!MsY_Ba0_!?oHEu>8j*zK#%l0lM
zX&#8vJ}wY)DB!E!P(9)6vFRm&q74nJ0St8td{hRjIBEmXxO6VAO@@opiLkCtf)h~4
zSP9ox0YQ_L=j<VDP7BU#oM&%;i<84Ou3kOCdNyL<LCa~p_NL&Jfl~%zSso0SPA9Cc
z<m{+$ZkX{+KmVUN{?%X2L6(!*JY*qM*Rla?0yIz;Qsc#kNv1Kv)G2aMkgI{MTRHpY
zzvCbN?B$%lX@k#u^kewKr+f?UUjL5Ca?T+0WHb|-3S>nir*@9QZEn!C^*fEb*`lCW
zY@fDmk>tDp4EKypRo7MZw}Ee8tj3*-Rq7he69begT1gP^sZqvzyNtHBDc|;1-u(RM
zGrF+BaTSW`6bocoPT_n=Lox>M1+fM%1&$(25<-weAOLC7ot8#`Y^ty)4c_ctg)GY{
zRdEu+n~LFpEgT!;eKz(kaD1@F>GhMWpy<?LapZB|C0ZwH$f`q2f*Xn`-UVuvPddk%
z6!%V=sf*DtbulVQ;RQyt6D*woKux{&=YX?HJX;se-QsXwN4n_~w_o?Lf?s+8!~9#?
z5z~dL@Tn4qrJ-WX!7k8AIGzzb_kd{aCJrAVdTvh`Xj_lgpqN<=LyuEZ^Yf0jK>7pT
z<^zWG^$*+sK8LIR3s)0BB%QVCJ6~Gq($y&%DP$6-2_aEkt6M?HLh*p)F(tXgiV4AI
zfX7lAhh~<kI&8i~VGP^FnCmysvo)RI6fVEyB$q9%vRY;gCXUQGa$`_&6y+3S!CJ$T
zGZ?2>u?*MN`S95rx$k4Yibwv;&r@E09F~{x-qz*^A(h@C*3|hJ;uNtK6{VPZD5m5k
z$c3AE?;pRE_rC7$IKH~ZSAPBDIsV8;Q(k?F=}JzP5!)TaNvR@UT0R9Lv>mAy^r6Le
z_uT$iM4l-k+l!@G0ET-CX!vYZ{Y~J}_vniK)4T^Wd9<1opiU7JN-MXwi@o(6oV}Iz
z{=pw`@tyDC^wJ2Kj48{4Jj*~0E|y^)k1-Os!&G?;)o4Ca2gJuHg?3?GQ3_P6e}hN5
z7{RGWvjNjwm@L69`+MBFyU9S7xbnnl)+8e<3NkGidIDEC3NY({T^|MuB{p_eX~e8?
zVZ18d2H~Pnjd@rN<wjEH$$24aX|}@*g}-mYpq;#ccKJ8MNspGXdZ>m%lVh43_d}D2
zX~0fT@L9btDPVwJ*PZ^`=bQ;cs~>)tF#2%7&|0e``MAOnEfR@H?<4C;MCE3bPGsaL
z;G#2yeHzHep*FPhfk*{>QvE{ZlGNJ!gleYRPgtuqpIG%)4nCg1;xk$~3{_J&hcSXv
zixWecS^UzFsaW=$=j??GY;12*Ob@u?lFK+TSmV;ACFJ0MEVc_&gee8y1Q#W%u&f!m
z15=b_%S&vzDIXX3)?fO0PCxbu><<lt+#;n5iQ@6Rufy0-453+}IHu!*p?cWcVDc~T
z<1H_HHUIVx@8N+Dx|WB3_xF-N;C@VpOZa8OFbfR<(CSuI-v>+tAzh$#N8HI5YK6u2
zY1-Z(BF_?$4=$Es0T}LSpy81bG<?oy259*7e~fhV8Y-tB@xZ|zZhI4ZVW0DFcpdNj
zlb3LrdIrA46a|?CTSMn7YE#v+tDZ5GBR(m@DvMJuO_5qvY&Bg%Ob}3e4XkCc6vi>J
z;6_9Cv22g0+`4rhoO0>06P#RKVNg13QIZ3hDn>)Fl8Hr}Q-fMf0D}hiC}@JOY)nO#
z#wAI71!i=P)}VR@W`oB%3q7M@NQV>JW2@RKOQmyH>aqFMq{wYwTO95FKA&^nexBL)
zKL4&MEx<5uU1~ee+o55YSt0FrpQ4X-o?5Fms^A11=-dS8-fKwcWH*>(>d{XHJ4~S7
zEK36oylP|IMF${A`xrIswRlcdiO`aos7f28`RezU(gwqTi58gA3Wc#86eTxZyvT)}
zZKinEm)AJCyv~W`RqVK6$r&<VFqClJIW6(lP>Ms1k|S8Pm}$w-<P4W`KC*d%lMnl1
zzUjaI64OgBh1FFo8D1e%!er2fG<b=6r*PAOa_Y!*4BHp@=$qcayI%ZC$_pDj<PneI
zbDsXKaQPMNt&OllOBNso>OB_0;cNYaBq`@o>oojaz0YD>Y<H9G0}(9z?P4hwfZ?7B
z8lI}EzX4o&&jkv1&a;2QISt9PR2@7)F-3{@1>=n!%*GZ=H=f~5zxivp8*bvtr4ep#
z7sN5l2TZ30c!Oq!(gZz%MG>F$Qll8zsHtf_DR0{nYzv8?#fXHz^?|<Rff9$$bEcyB
ze8{FNIJ2|K*7ShY(J`(%c8X<`p)1H-Nsf~FGA2k%%$3j>ss!FUtYon}99YYuVGTf0
zW9wo~l2S{hUBpYPpqAD_seUF&AY5bV*#wMLp7DJYlNJM7=Ia===ACImBWma}Vq~tn
zj@$;o;59LoO~2?&R<sW?%mWMwa%${k<&c1ac4u8%w6OHfEp_p%bDpRDoRXxc9bVAS
zUO({>DJ7}ir#=RIhi1l{{0o`m`P$qt6WkLuCMoEu0{gU5;{b$fN=>Tn2tdIZOKB{{
zz%sTOTSmBk^8$N2JCp}|T)MQ(>6J@awK*d(n4-XlBQpjx6!k?>z(6Rq#EWNyW#}@N
zO3$=(43;y_yMkMc^5`FWHV^*Jf6o|YxI7G0Mi(qu#SkYR9%C|`bEs3Yse|blz5b*8
z<EviDKfdXIbNu9`eCgLcnbTkUSSBZrG0k9@!=QHc3ipr7KG?@x1XG(pBykxPXh>#g
z^O8>%Q-j5J7v0W_$g@S{#fzm_0ET-yX!t=@{k41UDRtNRn9(k$Lz95F_&ZmY6nmSP
zn{Q@tFy+jv|C0B<>Q!7}bCy(blQ9`!?SQFzrZ_Y=NU2z<JQ-b0F*0geY0>q%RFott
z%|KHuxw6D4WHw{!9VR!F3guwP#N=#?=iJ^78{1n9^NdTD*SK_jjU^m1osySICW7&S
z?<BD>G*)n~3RN=oPO0%>-%KUm4!IJI8&4Oi(cg*I*)DYbNNKmIyWwTCAz~?D=z6pq
z&V^3{*A52^)QQJ^6;cZ@%(+d@NpOfjRF1m#+n`AM;74CEpK$f*Si~kpyU2X{PuCJh
zZ(}G#|AaxBY>s3KpnZ|fHMpT&?58_YC02LAMWWNF9;0=lcXch9Xe+Go7={Uy#{~ON
zYeJx+&6tYi064q9#ku`Gc8h}5)e$F$tDG7vv+ON)Trd>HsKb;2;5&6-p!6QG852_=
z0@fIo3K&c>as^q=`Ot+kT=MWQ;;BFX3z*NkAH|X-Gs9q_;9Q)}5QInurc)p2OJ3k$
zn=@~I3vYkXi#Yeuo4EEX9?i8+ei}UBD#oiT6eAc)@bEH~klQIPTB=o7uQAC)RDDQ+
z7h)J(<p9<emoGrVV!PW^86rO<BEP;^i3MQzjM%*QzX<&7VkPe6(_6GH&aCP*7wrH|
z<t6UW%`Nt~H(A*z;f9a%`rr9&{EcV0dUXZ%#-UqpgH@tehp0o1pjZTpN*Sp~jqo{y
zB1tT;s!m`mY@#H>OYmbdsu=GPuUKP)fLXw)!R3O_hfFf%{CJmJ_V?ME>@ymUICcCo
zPM9G}&Xc(ktBO@eU?=%j5^rU*>SeYtm)aqy5tM3O=y!ZlDy>78<(`CZX}a(-$F--g
z2$xxB9Z4xQa|`jKgu931TCIxTnWvFC0Hm!`(V2P7Z2=7P60Ms(Z;Os;<sj6r-b(6D
zUvK{IRF7#VJ@+(y19xRz(`|Qb=ft+zv2B|j+wQ1i+wR!biEZ1qZQZ=jH|{Ukd+gd(
zYpywGRq2bZ7BUU&F))|Ylst6}El}|iF0_@SPIAAZRNl#9=jkM^N5-sXq(%`+A_UAu
z_F_umutGzEXm{?fPtvcodCCm7RnQ+~YH01v&tgj#GT#PrG}U!$363c^qtb;G`(Q^{
z4?Gi+;)f2z8bp9_<1({FpXv6nzu$jT{_Ol=wRnt}l7z4xs37P@o1y9^%nooJeXS@}
zZGMOyJ9h3vR$t)@JJkBG6I<gaL;TU+ZT?lJ##p%)x#6$d@B8_`*RU;^rb_x~yt8nx
zL@d2i;Gvl?8$+5r0ow+(8=}qn%!BXihFIS`6N{<x0<9cYc8f+7_uai*cx1ZsUcj6}
z!KQ=>{b>N~$YrJ3%mv*d%)$4rEHXRa0UDrVY<-0QlPIgiPiADPD|_lx+6d_8)k`Qh
zayTMYwm<>ZH&g@g!M0YXlGA_$WQh4~S$i`&*;FoE<77M+>q`QA2e%MME(33m{v--|
zlV4*aqbPeIKSm-4Euh>nBD6>pX)BpPsRJ2Q-^ymy$F1U_1!JajA>kS*WuVCl<pb;`
z+}dkr&tr5EaxE?%F5jei!5CFggV!K2EsMOe0@r-CWp(3^b4e#a;Vx`F!L1AK$W>yp
z8J)KrJ@pT#{SQN<)(C}h2=xdCubrwdyk`LdvKt<(axDe9MvZ}YTy0yHf;<QTsMl(#
zB|~uPm=QJXbewH%11~rDkQ}#ygCjd!DN&0WkBpA7van)lLj*Fz6$6J@kOyF@1FMmF
zw0(cp{$6%G!7=WWix-10e*0ys2(BC-Qurf>zIvznPR4F%#+a0I^K65j6fEueaX1id
zg>K}p_9uLAQEXqPA9NhHAa7Nt+#%M_^de(oC{y73%)?>$mWhU398PP=0q-*cxd=!y
z(Jrzs73+CIRwV0YfUxn1i_7K@*G&4v?>c<<#V+>$y#POgr&y=$TgpDEujdCW*Rjl+
z2P6Fv61WWbL*rT#!VvR)28|G!CU=|JuxvT&vhX&aGr9&YL-_j_&c=*Q!ZHA~2PFLg
z%9w8sXpA^;<Y?I_J7KObYbYfG*sWiJCAd6`(%%_{<_r%Ac)s4b<i#|i@<yWbG)lux
zO^>82%FMWQFg+tSguvD39Yi0ZH_~OkHS86@(^QPlg);d;*@#z>S5y~LDKxh0v#3@+
zYZ}vSO<cz%bI?26XdcB`!?&X{htgC;H~Y!_S~C(zu0s5Jx~T>sz(PN>KSFjpVs-~e
zBnj9v6!=#Fy>X8zg~x2b!K_oYkY*RHD~yp24p*7Z=U{KyBB~1Rf%y9v6o-m1w6+f~
z9KAkTvJ}|T$<x=FGV7)P5j6wx8+?hh@*)n8-1VgXpX4yhpvPb~Hrw@jyLVi`4Lui@
zlQrw^><m|Vd~NcPae}`N*q(GGwHlnQ)^C;w^{IKz{$LmCaDQveDhqapie<QbEG(%o
z;xr>8u0r6hWA=59PkMs>EqY?#H*Eoh8pJV>!4X5pn}&Lg%Havh{`sPRec9y=n~zHI
z+*|DdLP7ALdTerb&Phgq9qckkv|ho3as@k~1vQ&Key`7fytsI2v|8bEq@yW=23XVe
z@~jPguL)hR(+K~r)iU7lIr(K%pq-|**lw4ao0hH=7R46qzkpSL6hhWGhCVMLzAfL>
zW0}4mXYOw~ouLm7J%#zm1CM+q%o^B?E+)$79t7MzQ3W+9M6Zbvdc~mZ#(q*GrAlN>
zZvr3oNo!#}o&#DDOGs*}FleA-q+?i!bQp*T%j;`KPG$iz@Dxii-?S@JG(5zxb0mj&
z5q4%WafeW(pl+CkGKOhMn1QW`w&u6^U_-1sI`kz1?Mxk(9;#NluHkknlfiR=@5T(n
zF@D5fc&36Heyp@Wfrzxa`N6_K=+L`;B!k%+2IjqE)S2*O!?S~3^p6!mgH;JDtm_{~
z!qbJ+BfVes(I-YIZ@O6oC5wX~H|>%1lKz4u3f9nNdoM)j`h1czi_a`HJypwL@J+>}
z29BlM+Wwk<iQPecey4?RkN@?qw#JR*WM47yRN95Z)z&xRC2*AFGuHSMEQ&=bZYxFa
zp!Bb2XIylNSsZMl6S)Qvf-~U+j3zi2tM@X{Zz68ozH{A-B7nj7wfXMf)u-Hh*jVg>
zNk4Sz555N0PzJ{Mo=b9VyFWBv*Jjm*dxIIz?^XnZA35)=z%SKEaguilEX>lr`RPH!
z#?i#seI8z(^b9N{YPyI8@Duaejf<wGu;G}{VaWgF_neb|JIH0PI@=K#<qi~C2TApL
zY+lE!2|CI{S%gxo+1sW{*0pU8EVgFjHx>}uZSCLV&~k7#Z`m%rt%H)RyL%o*>w137
z{M>BW{?fxRAV*WCcE@Eh16a5i72X(D;}Hsjt6rpwUT;!B&+YW6InD?4?aXi2$DS#e
z23wHpa4he=JLfoYul;tU*X18UACjv(C1Q|rJo?_KB00^=%0q3V80{}NCz7ESHnJ3+
zkSd}*h}lNje*f*gTj)tHD6w5>wp49s<vLJrOMr;Os;w##__t^!(U#yN0#aE7-U3!5
zUe3r}MkvjFgn!b}Smc3LkLPt_>6`lzP|(fqr5_Ddm66Y0GUNPfKldV>Y__Sfa#Q|J
zy%QoRJ1|)tFTxUPW>s9!zJ!uqZgyTYt`pJUlmdbQ;ujX*R4V2U;TVvSYK3;l`odQE
zn`Z=!kx7r?PKo2Pcue1L#N>9DiHnTP=kbBDQ~UP?N&X8&8eq=88l*`x4RteSI^z|m
zDpGdLKldEsgnN2vtv*oX|Ln6rp(>&Qg?K^1?~z9@`raSZV@$yJD!6%9rha2*=S@*8
zuh*Bt^I`$lb#9O-J$4=wXaE6Pb|RIaw{cuu5Ta9@b8S=O2|>>a8*yP<)1fTlqayO@
zQ;P=sEe`cU-l3{I`2KkM=B<kDH|7wf&>1@~Ga;h#L4E&jZ5g~JYY!J^ql*E3Kg>RN
z-99M}dpR!~+47b-|AHjf@u0<QE=l$z?c|q5pr;YJ1N-gCpwY+^RQfDen~Hvkf+D8=
zb13QC&f}Q#i)si`;z9i>CVg(S+}LZ%dC<Ga#@)8IZwNHDiK2<l`n{s@ecR?zhg8BU
z7R1$TnKEY#en1JFme?=OxYlD~YQ#wh0~zbrGb<{A57Pc6$~1GhAdWeM;iINl=tc6p
z;OvO3uV6B%TS@XSdMX3^$^9ludpz?;%b#Z@oFBg=clmzuY>ZZ%CKsayzsldUZ4eyK
zmSQ6nUhMjq{$-XfmAYfH?2b)LltmEQTiQZtL=}P!I0nKEunVgFWBTg*jckXq7%!<)
zM=h{_b5l!<HR(x8z^EEB#!PO=NoiozK6U6mUX-B}TE~NJZ_R4BaadtBSmR8~8AAsP
zkwz@~14aRF$$q?yP)r1>=Uh-YF_;>=7>@C{-{8i0u+9#KZPdF8_?XLco5nx9y=vt^
zGgjhP8nVvqXxEc$|7gnhs9wFp%izm9R>t#8(-r(}i<+=sVR}@9d{&$$Up`%Z)Pg)+
zP_E0rQp+))GX4nzSlh68tN9<{-2UhPBE2fw<5qI&3rYVSX$n{7Gm0C^1+h6obV&VQ
z3a;Ty2~l^63A*pV?nnbN?VY}AFZ3&)71$PL0((im?yGY{8{@O`(ujsxZfgnI?u;@8
zI^>&Lfl#4iTSiS|bId60A6MFmI7!oFst+2qMEw)0@20u-WT}y}FrwLthsnH(lk!h1
zln6i<GuZLiZ@Z&<72Z*xreR5RoSTGZe0g6PY?eCmxBiKxZN?CY#MAf<V*KO6FI8_U
z6f;5C`rfWnsrlRfkkpFnhIlNr#z*Tb<Z!eWrC^n^V4qrt?z5>Y^0XSj$RfEc*+?7}
z>lGi#dqv9k)$DdP%9gW&yTdV?=Tw%t2<j@zt<I%6%f32MYvt`k54UF}(nkH`h^{@&
zJn!EqL-;E1(I_|*L5k!+A|MyWZs1Px5%nUon?B5p9W@)-B~`e_4W3>8?P1AWlxC7o
zxiU+I@#MNfM}amFxInwN|2u6j;O9_XW={j5`0%_@M_)fKvJqS%4ddb>TglG^*i({@
z;>BO!zARC}Kg4)T%zg(ND#fn-FrJbtuJG<Tx&eN##=4#Iri2`$F~4LgG#N;Z?RK4$
z&GcQ<_PE;m#LF<V+x5wL0_WMc%7J?FCER-;g8_Z{ksQBQ>9A2=!{!%#dvJ{M75(E!
z1XZ75TZC<M=x|KLFf7LZY*i7J1gt#Ds=nYFdf&NzfhnTsxTj>K5|H!q$i|If*8_6!
z`EzKTOUCVGtLv6V$7>4`IQ7cal&XY(Zh&+v7+lRUCpXPAG(f5rAl^<9!HWM&a!0L=
z_E%{mDUobCz)hk-p`lSDeO8>EA|i$g2(Ex#I8JNZ(BH2$IB0LnY+%J@>U74#>{5Xs
zo~fg^Q)e`UiNz=y<`T|9rVft^MR0-j>zLL*yJxWh<GY#B7Q#P?I6i+9OjqpuFoU$Z
zSAq;+2U$^=JCF|LJNJi^Zc$a~my-~R)VgaKy)BHmiqe)ZXM>EDa#rt|y9?bnc_tuj
z3ge!oeBVs6JiE6!MJn~BXEiPOR*Z~KN-ryU+BA>PGrvx`b-NO#LL(_CZhutkP-e*3
z<5B^_1OFPGfFpc@Ye`$^?i{N~&5^+J3+qj8B@@?vXZ!DkVefK^u{7t|7)=F2NChG!
zZPKXJ&Txdk9}TA1Ypke{gf`eZpIn|l)el8nyv>-New90coq7~z4aW?kpm@#!kqZTb
zL?=%iV+y$&c>WjMX#UIE^VOdTUSaEp6%pUlwI@41=-Tg0s&k*eRy^<A$Lh)^@7f;u
zjyB!4B8W(sM@S`21C1flz>Q@Deo70_^@q)#R*B8n7TSzqSB#X+)>$`7+f=@KWr7#d
znwP<d{QQX=>w%BLh{ZAj$D9xv8Tz!@@s!!2{#E<LX_oV*M;aH4bwh0E&7U(K>W+Mc
zaJK2W`Fio|y{aQFf}xEj#=FN&k$+dPAw_1nsu&3E@t5<__J*FCsCjCqPGQ;Qalfhf
zCixoY2wI!?c8>T~8vyeyq*qc@@aLC(^$<IVcfGi6Nsxdr0Vvu%u4}fWP?}fXZyN(E
zjh;37gjSY;P<%JgdT&-;;oOd>Pdr|j&E~07?4`wg7jo+Igf<++ShOTPdycn7^kjSp
zf7JJhQr1t5rW=_g02qq1r>gyXPSq`QDoFOk{i!IU|EA3)GV(IJowH2fFVc^e4Kwm5
zGOMj^9SuEUuV@f(fuVi%pt$K3ESeBSaiqAbY!m!Q_Si~|L~ne~+-JYkl~n``qxvr!
zT5!^%O9x7~4tK)~yao!a=L4z-b~+E$3Jv*$>T@vyE*&>YUq^dvUS7*CFWt(U9l<XO
zglQLN!x=kyS5{_~zm5Kphae4@(FEl|wKm#M!-mbs71m*SIufcXm@py6uX&NtMw#Ha
zT;mOJ;5nK5gQ?o_tm=oYYc3)@l`100{_chsE3QH;l^mO3(d^REHa3oqouc(RHENy+
z6ORt&t&=??iIQ`5VkqS(I6P8?L=52YlGhId<l;2B7}FL=+Ha4)iZy|yc`sYwVG<I9
zR<kg*QS$=MeyDOBO`SdUdbdiKk2X_><vs*H08ZAlLQI`I3%<uPb%zJfujH{y4-EXT
ztgQ&&W%fQYjQRyATZkmMY)EXw&Z!=}HST!nb|;Sr>}KgQ-&X&gz@XY<U$%&MBjgvd
zh(KtFJyu4e=K7bg=^e%`Nx2GTp=2=OZmYmbZ4QS%WI(Q9ik~pCiqBwFwJ(Y9M{)JG
zAvL|8elf2PED8>}s_gu9&UQFg-t+osoEJv_!OnKOgSpGY(eYHrx40yxbW7x4)@Swf
zFr8DuEvbg8+T6tpg^;`8oYxu~s;zHyX$eS3w}E8O;TZ`toxDM%Fi)0#lwVwKX4{*#
z4gh^fbYYO>Fb~AE$94cgA<nM-CrMxb;kdxsxa|}X0fjwlgf3f7YXkrk?)@xgh92vy
z6!xL}U~Gw|h8H!=gme~d^Sev({;G@DxuAqgT*%J+8}m5=T*<$)qTDjw=(=>>x9Xa)
zDv^mC$VaGJ2`YiAYZa1wy;3%4j~kWTLJ7dp3dww)7fQSu+JVJ5ajX&yZ>&Ogp|=?u
zl{-462BbG((q+|bP!6TjMbkI86s{D7`K>sQ4t~G++%JTA<)Si~)#Nhh`CJ1O5>#S7
z@~H&OIU{%vA^~Qvv5d;_pF2LR#1T!H#FKCLSc?+QMBWn~Uc=fmc_`jvi;g{u#sp8m
zQ%#<OVxP>HL4$XhS<}Lds-yI!t|0l&6T4VC_kQJ^!2{o`H2VQj>m_u9!yADl@^Hs@
z!v1+`Oo_GBS?dtvqlYS4Sj>q<4Qe}`>C|rvB*b3RzcM{lUugF<(Y<kRY{L~#ee6We
za60yQ)bSq%leN|B0?S-IT5AH=*3Ta`>xaS&U&?$Bht1!O!x6<SyCd32&?-s!@O0Rg
zCAd*yBE(6pUg8!X#s#TZ$+ZrDrB?D-tw~Q;^)e4Nf1>9_2(O+h&E{d3l2$rI5nbeC
z1Wq@JNLG?Waiq@EX60Anu3@)$bdJ=TX5?#UoH(dwd*eWIFs=;ktoS_cr3DJ$^)%D*
zHA$FXW0#QHDG}QkIeO1(svvp&8RxpSiEVdsGSO&p-(Wo~Y>?_&RCVwkFu)AMlv`+{
z4k;n_V2J^!hTy47>hh-1nGzM?NTk$u8!dN}1sOTk?plzNH6XDFo4wheGMHe7ucUO6
zmFkahH{czRaE~q)HRP#7*7xWHW7#yEYWK~a#SU%-g;Io5gi>(iB8zK#0CrbYc8<L!
zO*OttRe~sY$aI~ZlEaL3K)@Uo<Oi$GvCteFiA{n<ZiKrmgg=5L`p+DKdLxe9EC26e
zZltBKJr^N`HquJ5O>KS$dr5k@V;aL2?iOAil)=v4Qi^<e*I#^B`E@3};`;^JV~rqg
zw>zBi6&~~Xpv~!VPl@Q>j|#kf(+5*u4=VhV9;V=t>u7=NYwtQ$si=tt{QTwc-1!?-
z87J#004whlLAZ16ETrAA9%hiUr#f_aDpaCIA@w(-ZH75VtagH);S4aJA8NKb@Z9u|
zNkfp>t=d0l;?fn(g`=GQ3%VhWmji}f5ol_eJhObQ5EA`E(CTN@&g;!5$EC(!Y%0(h
z+EB<8_%73yj`0iRjqaiQjWN4_0$Q?UVRhp<Gd0}k81m@JYLpJkmA+AfyfHfX)ik*h
zmCCM#u-n<mJ_pB;5^-hu^hsa~M*3<!mUcAVx2tE})XH<j%L)hMx0*Z$$HsPElnX+I
ztY$T}WmLb3G`e;mYr(0ZE$obQ>jmkQP~=kt^JvT+v>OT%8tAP5;NKyPhX}_&6>&-5
zE=>tm*ap(x);Xt@!uWiq>J;4viRm6|`9=<^$xfVu&vzf`SUCE@6WFY;?omrssj1k!
zfp9}@<<s0V7|49L#^~&)(Hw4J)+Mr*m41e&k8-C{N5z-(<{c<^e<#-6Wi4uo46>o-
zqfF~2&K7fX<kLoI94jka9euM6Pqy!)$mT{?8V~z)ni+ZG!kQi$l<Z;hTYV(613jZA
zk$a_<p@@8?mff@@-)G|<L1(oHql8CRoYK<GwfPqxZpvxaspDwu)U$#A#F70`m-Dwx
zPqO_0O#SO(>-MJk{g<uPh!Me^`Q_`lS+Dn%-=FU$XYA4-NlrNeqpsYtJ6Xz!+dqO^
z)s;y)$YZ@?1#TTptee2UFBj|K&g()3{yopHB7Iq^4;X$J+N_fT?Is%XRH9u?fU5e!
zpE{>GZa_>T+U537X={SEw2wI~l(|@=i^ZMc#`e7@pOB32NrIQmr7KsK&Yp7r1W&1m
zz`Bc-?V|eBQr4H}rWp2f$1j0eZwcEC$W2z8Sgt!M#p`Ic{hDB$MB&2zyO}7$DPDzA
z5vcM|k}82vGgL%Vd|+1!S7DnoT+7a*H?AXLt|K#*Hqlc@;~BvttC3+cEzjEnI;@&z
zeV(RJ379v6rA8tRqO0J1)fp3wfWXwZGne_<1h>t{+Ud;;HoH^F466{Aq{KlRrOfW)
ztMEdO>A3#oxWOdtga)^txcfR+IEvU+&>u+m3X~^JWTK?~N#VWEk8gAei3@?5w6Fs?
zWApW>RLokbs|)mot2814@f`N(GsXNvqqWY{?gscI88&@k7n~Fs$rCSZW3FLMe!s3b
zflZTsBiN7>RuvIuc~T1yqzx)4Gflbi^HU5S0>1AfqgH|3%1l_h6x1J<i}ZPVcPM_u
zgG^g;i3T@SKpg|ee{{emvPm4vw7M1gmiHdSZD{`IKKp@<=ULCK!ME?mRZGMt;B79?
zzUUVCQzt|111-g`-Q22hg4D7gtCmPMH%Z^+e+Mo+-eTY&lA$gU{@xZ9rkEJyzJ$d5
zG{o#0lroH%TdC&2l(J8d>t5Kai&C?9`o`xT{&pz1;?=ax4o(Ui6sK7>{>d-~pSYUL
z`dYhy%7b{SUselfe7^8k?|mswAL9JI8%vK&s6^INitzI;S$`VG-iAA-<Zzip^5Dd@
zbXN(Sx)SsG9VfevYjztji!Sht;~J9DQ~tEvCI3m9M<5m$Zaeif;so=5jGm-&aH1uh
zRza`hGa-szRJVq<4o=RG0M8gBYqqwjUDw%GEgh&;^FXMwsZo^I$Z<!QB(Lz0q|wPQ
zQg%W!lwd*bNF56p>qs50y?O_6^i(_1%k*Kdpt}eX2K<fRLl@4*qgqM<@FAWg>O=Oi
zQ60)s(m?HvS<N@xI;ngdA3_@jLTyTR&6M5OCVTBN0COEmk5dw=UKQ1}N8KdYDAA5W
z(+30ruqRK^9x7}pehD^xBM0#lHNUyp#*Ku_Ef9Wlr7>%3n{8PSYN^)lSdkMcHO5hc
zl@*x?iS+twC20ln7-bsG+kni)$%~69DKXT3jdR`v+~mIcToMoy6Lpc>?(3|dG?fr>
z<__@?pr9WGnzkL}frg~LLL++JgRWGuyMJwYhtcux7^w3`y>)DJ?{tHCi7E>#_cUZY
zoiHGtbflaT48W}RdV89mU^m%FK4H!83)aQnLsuwJyX>9D80cuUD~WuDg?df?FX@5a
zB7LAL4_ly9_2%n-x564LJHDqeGWre@$*NEudy<hHy0iG)y>v_7X)6VW%xSfeBQ35u
z*3({}*q^W8iH}O~h4z#>p#kexvkZc`$-P<>L<t;4R#=^8c0gs#D9%k!Fb>L8*b%BV
zRefzj?|C3sOCN{=$W2l`UA`kL;h^NLRQ@SxEy!ICy(+`kEf>X<yE3w#OZ2y_5>Pod
zmeS@HI!Fsi-Wv-}xQK6|9%^D=(V}00la*V<v#S5K_z?(-gHy4jyjYC7=q<Y-Mh-<U
zW;xLd5{pUL09-E0H;(%F*LcSMgKD!SQv-uI)&Mbnf2mfwuMj7dUM@UDGj0dB>*;Ek
zV)CP{SOaS<r?9$W>YQ4w)~*>lUhX35xL~~`ql5j2mrteAECU~W2c(h^j#Gvz8Wc6H
zp)pNUrggG48~pd*seKVY;P?~!8fvc9lr4?9lH6Km?yN>{L514m@SmtqL=2N}3^B^J
z6oUp?wV~*6rPc;*cweC(?0j*$fXcv86RG6kVUX6<Vq4Klm~lOm*XAvC*?l#!3c_?O
zzIeg}m^P8b#4L1WMoo~J>oKES>sLq1yTMj(N6r(T{RspVt>c)cZcH`07L?%0ro1S4
z-Mz!hD#Xso!*O=nl<@SlYLyHtCI(tHa-1(UaZba3=e_>SwY}<xTX#Jjw%MIBNnfWA
zU67cquL4rMH0>RD#96h8&lnTAl@)M>%Lzm|s|H{Nd`)P-aP!UCTYATHTUq{%LSn!t
zZYX`ABL`I4ZPbC7hgR}1gj0b+QJM;Z@kmIxWKQL-&^X=rJ99JWi-()+`38x_O@s)!
zw;Do}XR%0J;I{yiVhRUSd<HNV)fD+x1*IyQDW|~|vY1T=;&inf?sw5-;@R@MMTdDy
zZ<<Zoa7{=;HLSEcO$}XHzh3P~zdSq)qo`5AvnOJIXO<^tTGfZR)L5t)i;Df|{(^WF
zXR^}k)|QM&#o24SOBjxD<6W4D7&CZfZXIIV>yaKaqiY5>C6kc`j3TgvstFmOsL4tf
zfuz9-A2D5Ht(87q6X~Z0YO|TDvhXY61bc2JQc{<tlzOqf=CYi%0w0|qu}n64Yqm)<
zwGfaGqXetnWkLOjPDd9!YIL)laO`Z+DQ1FH$qWTvyx}Cnhu1NDF|k|lwYlFGcPZc%
zBt}LHVtZ;n&y*#jX>C`(!n)pW&gs0=c2i>Ki=fW;abo{Ov%L)=Q$#B|_cqYR7pDIp
zg!T9*N!Es2JWO4CS?mfGi>~<M$k;gLDa}BK@VzDUodx+NHmDaYz+osfc&OyG3q0Fk
z2I$TALw_meq;t(g0&kY@m44ua?M|e<$$l3Q?YmQdY&g=>2N`BNiW1Y%J73vn#y6t>
zG3jf1cf~iBRhHiOfR;6t*6dW^+K+kN2~wBiPjca=9+Lp`DmY_1%lR0^GHNdsbs8?2
zh`<M(6*@f&?Qxz6Q0(_mG2&_gkeNRHD&D?b7vsB=qcN(FJze|_5riq+6|B;ck321z
z70>SwxPCh?<rubUn=)=RcIY_5EC#UojRg`hT{ka*V}ySUIF*dT7%E8LrkS$lDttso
z)n`hAVGz2&l_RAnOOM~je4E>hTc<9A$zQf0YUT1P<Q7g53EF>TJ_*ScmH-`MLj2R5
zj_;cxznOt#VVLm=OBYd;8|K#p&zf5ZPbWyiiY(Mmqq6xOS8~;fhAPc+tJ6`IN^+TY
zvxyfK_)rS~Vb(e1bZq93k*(ug!=wrlY0)CLO8u5Fdxb8P{Fcp8-t$`nK(%4*6kVOq
zX{B7aH3*bmRbEF%JLcL#wr0*J$&$TTFm#@ts&cutf#Y9xT?L+aMt5UG!urYP(+<_v
z8}awR35q_^H;b&HDV}(K@zVGG!D>y5gA{co=r-8l>NiN+Qd&{9{*LU$#*xAND5p&J
z@L_(8xTpV)Z?_6_cmJ|ao@<qeK`bQ5=AGcTm!-`1GE!ltt)d<^+fSIG>Mx!J=P1i9
zL4U3Wz$xJL4@^#Ly4%OjSj*-2+KswDgFW77ft9kW1DT~yBie!?n&AD&BpfLy4h}6i
zT5SpUti{t5?Hi!<Y8K?6jr)_dT$}Y<Sz%&$T3N3m;t3hHM`RV)Z=_1Bj}(s^f15(b
z*1U9FLnk}dS}1r9_cf7|nf718*Z1^1(lH(8B(+9Qd5NwL^f^pQ87b<Zn4rU4CFh!0
ziKf};Wap4p0XB8=B5d_eQ24pu9xmYlIUZtvC;9<p2=+}MIsdGH>2Q?qCa_ufLCZ~$
zdHBhYR5i=sRNR_;AvmIDWEZ;M9<Bq>licl2&t<*bIU<6(%5UMT>M~i<s>wQ<Hy&iA
zbNk<=2UsEzIbHD3gaYEiuCwW(=nT5wZ!o-R@dYHRWLxmrQ`#I%GLl?UN(|aj=#gZB
zd08pkdMY=E3~{<>B7R6;Le^z|auU+=&KLu1a;KYgzl8ro1?1sskFTW5y6-!62H1Ju
z!0yCz&&VzPro$RusLJzfE}AG>nc<y^f>RB$btSd4AKfafV@XS?K7zEh7^W}`>vG7w
z0U*C{B721Vm%*kYeF>@$TR-IM%+G@`z@&T`qdx9U<$wb7J3<i+Yu0+cco{>shtr)A
zv1R^c#1cvI>XqB847b@|uf<-ESyu<UrXz)CZ)t&?6J9Ji#~{o-VgCfpcX7oZtDy=1
zzZYPEf>T#NYWr2$ARJYTZjHBfN1rM<wDI`6;o+7Fo29iWJJTOpL@39;EV0N!Kti}{
z6`tbgAuF`NwTMS%-G+X)cU}f*6%*+Q3cEh^7^3snmip#nqWa(6+vy5v?i54YA?UIO
zPEEuzT47-vk}<Yf5%GxGL5Wa5%sV!GvW@B@5iVw^zXm|f;0rzrSvjd?vNUH-$wt^K
zF9wxNQX<A11ZY1=UB$5P6)O>ad)F(us}Bf0lGYY7!rT3%Q0dd#I)Ao2BzZsRaq3(y
z7&ggzl2IP4<&Kp+<<lus6anjYOw@;ZtgxNS+%n=`_bVv{Yc`@?6Vi879j&wybqt~R
zYnh~B3#kdtDSG)nP79<HLcn0w=DgC@1+FP2-l1RiDoGEXn*qh|A6I*P^4<^*O$d;<
z?maK9UtS0SmqrO4KA=f}fV~gI;T~~etY*9wM1?-!M7ylWXHe*S<Nw7`v~N!3K_67U
z!8|hxamu~g+wPK_lZaZI%~3j*gAh!|+@NKk)~Y%`i(y$qe@LF)9PGWKm)%c`26zIc
z|9aQ84O;CBTAz@yZs8rp)f8wF!1gmoz9&fluAoN=k~){aCB@dDc;A&AD<k3|M&7W+
zuhDvv@{^U4WQGHCB-g{7OP$yeC8tfnwc*vd<FXbfGuk0UxX$v@>ME=Gu3f~3G_NoF
zb8>^mw03szdr%=!QDcowD6mY*_ed87H6BInh@xRPp-ZfVj7l9Z{@Tc%=Mp?FIygP#
zLPq$j&AG^`!|X3yJxW<i2gIsr+U;Yi7xl1>U@g1~edCU^;h4)NY}SkaR7PUmwUl-m
zL_P-{U>HHWAh@w!YqC7?YRO<qUM~C2cM{&ee(Oi#ZkQ{MfAB*?mP_zHR?xj8r_T5S
zT@vmdDVLFo*THh8o#Kn6Wmg0~zpfEJvAzu(Cm^Yf{}!3&jo_2v-1j*k>>(1!R9LnY
z{PPC)C=N(N4clj0c_TkiYfd=@-tS|MJ-9TqCc3X5KAb^^MvfU-Pl8=FnzwTUh^l|)
zx({RamX6&PYx^w3<in32?7r=nwf262Sg$+i{)|viM_nQpaUk^LQmfIqU1)9(d+Iv;
zMcaApL0@Txdyim)ZCFq0E?O?hSF3Jl?Iax`F6`R8U_C2(u1zu+Jy^QbDP5t~JgT%T
zeZamooW*hBWQb`^7FO(oq}_&|(>F?R!nP$cv6d6OV-?)=B-rHqn(HL7c(bb#{+K35
zdQwsun1?6!XRfIv{q^7u@9~zhS*bC1z(XCXDo-eX;B;knnr18Zo=%e)vxCVT8Q#K%
zYT#6RiQ?QtSiURI>D!%Y7@(?zJ-}APcS4}n?dttP&yD0-9TD+8IW^p+W5bH5g)UM^
z8cc`{N@k(#k{7+uvl<b?G|NM`IhkFGDU~5zYaA^PYIm3g;Lqk~xvNYj!IJ^9#E_zr
zqBlwMx#{G-ey9H<<Vr-Fv+N;M0aW#C*$6;hWk)uLPSRJKGjeSjZwRVoTX@zk@H!P@
zfZ6QlW?Fkq4jc4ML7s;Pn|*^7<{A4YaACRDhUEe!Qq0<|vQlqYr!p}x<4!`{59&4c
zjZjqa#w$XZB`CR>(hwv~#De7bsdv09?-v4x4NtydvWLU~7g0EAq0p8G(uaQT6i6--
zj#4P1L$#v!ofzS>VU7i7(b@J_5p^{I>JJ$8lmS6IU{bcr))C*m)yVHwG4(J54U%*-
z+6E-Dd}FMWiBqnOs#`c+F?xt|9T%-hN3^b`ymAQ_)xgLB<(9NYX#w#l?R*=_ND(-1
zM=i`0MK0;H;W7hK2`Xxl55iZf_~qG?wkP}HdhnHB*hxOxF^`}4wOv0iw)&KKf;E<9
z+DJ`x*x6auUC!GvZrdaK%pfhUJ(e|w1MZmeR)CC6$X8Z|+9dVi`4KkJZQZ)~w@6ig
zZV+p)kYoHma9Gi|soTp#(^pr8LQ7?<?O3q8FaqnO*xDC~n;%y`U(En;ecM^Yx78y1
zQU8;4g2Nypd|eIHW*6~Jye)=G2d4)MI!LTZa{`-?Et<@uH-<-yhOQ|gAN55d`YPou
z3iEF8EqKnL*4+c;KVz0|O4&cbFN1M_V?6;CjoZ8aCGa+DyJzX4b=D+}^A>F=Qu#kq
z#C`Azi$%mYfhwf09pw&RnNGxu5qY-OZeA8x*>+SS4H8nYF`<U|0*%s-jS`P;Q{ivV
zPalV~h^AKj(`T;+yLE<2n~BF#x}rn9`EVn)ycgVqxe>{gGrgtnR_w6bq!Vr&Nm(bS
zqtSAz-ENydO5=kNuIM1SWx8R?TXv^<m+e*_glfvoBYi~fH&OdMNmSts8GJ~&ZOW-9
zU-U9V1yFHH>fK=!xCPf59-Df1vdo~Z%k7o6OBPk!I#^<SF7Mp0>#KToU7>5UyDLuL
z+uyDWWOZ6}tyGucoyIMq&MXVZ8R1Z2gQ0_AWmt6z%f3)DEI7*Zh#oA*I7Dff6DJj9
zBN7|AnZ&hqa10#c51$pAV(|KiR35<i*9a%s%qj2<{N3Q^f<j(!y}#N#hRXFO;$n0s
zi2Q`-s89);V1?HHN`nlP67HHqInT|}Q}5>5{IUbbHiesoZn)1-Q^(#Ix}e_xWIC8~
z!3em`Zen0bIn^&niR2(9%Po{&c?{ACadF_B(OcDu(TPY93AD0h=N)A^m>4bh^5aTY
z7aed%ey9aSD^%3twY`OpO4vOpiHau-cWB#xMH;FzQI^xwOlgJZ3AhbzHnK&vcj+=<
z4j1_v`6>33+tHtTVdgp8n9?JDz%oHxTOe(z_8vY0TzCie%-OVy&QKGfJIgp*8Yhxr
zX`iZ{s>S((?Ow943Vnt_LldwnmbQbR^5V5poK@Wc8yflDYw}!u9CPiuBA<<oL}|Yn
zAFIsWwE*x<yeE-9v8@o9H#KH2f;4Av!a{nqt3)<gCuL|=VKv_I12fe~O>Cv1p*c!Z
zri^RiB|NrugUk?_FpV<XV@o$JUGD^5ukdy|{bp4=@ekg(dO15r0rX8XvFaXbAgb86
zE<X)X?W?4q4RIgO#i{VUnYJoXJca3!*8K|%+<i7)+AbHznF{t!<aG%Rw{}@4`yowo
z>*5nJhOSCik$)~V(?8V1c)!h00+Q<JB<k!$ijN!QH>pgCTu~#?Y)s3in%b95e<Ba<
z)z79ihgEIvN|?j<=0@81;Ei(Yoz$;TwboJi12;7)%P|uHMtjtwjNR?Gb#P}+PDq9l
zTaEK<BXMG-nMO*5CRK{yIuo;^(R4`(_2FX#&kFb-??+C}LwpdAQ|2@wInczaU7sl&
zaF#JP)2;~#!3LRRo5=(I@EjT69}{Tq?g)TCoU<=7nqtIW&&kWV&1birJV~AsxnLXm
zJNGvNnN5Bgzb)+G@Wz?>e5vH=qOY<NwdhPlAK6(@bbyqSws}W-$#LRLe<)<@S#3Q`
zn`56`hgF-z+AaW0$T}=N)6*M`nqATfDK`GnhmG)RLB7Skj0BiEzJn{`^%@-;HT4P;
zgm9(_J6F%z&)qNht9cf9uKSOBPesk-)#tTW%q<q&D4%mfwdulZMkTn%c1D*Pt5`zJ
zuCll0HVrqeGimi-jvwXX&gJ`Ouk4=#hhomStP8t6ZXAf}H)?qRvIUT@8JTFrpoKN-
zOuK#-SA@_sq49hGxqc?2eEJ-9_eW;&0_!xL(p-c8opO`5i8MBw@K~;=ZqHulp2zy#
ziF_cpu9&v`?{uyFb`@_jV|NatIgU3fQ%uk3X5-r5<;t>*m`9=J?m__gDGvUvCeJrk
z#ja>Xjqx*P?jiFF)Rp<N-8Dd(9dp+~IjgqzKls^RXYYWKPVbr~2-S&M&W5tse?X)X
zp}t2*{!#JGYpyEqCCC!H*0hMz^p6DWEjASc$@$?ti<<;$n+oWW?(vs`ZzO$99qO(U
zv<*URA>PTA;{|B-QrAgQi<aMbGgb|8$@aeG$&@=A)o}38*qov?$A$kwun}u3c;doG
zTU2i6;ws}T_=3$h;o2+EeMv`$q{O%D`j8Z}L!k@aktgGGolIrq5sLa4EzIf_3F&#`
zk)K9XvO@Wuwp;Dk{^zx|cR1y@`-MG%i_kG-dJ-2DdNwA=YSByTiM15$I5?Zp{xOQ&
zJbK8LlRp0A#nir_ytiYz^z|>>4$oFqrtz$x;)G80D(xv6F6?S1I3v63L0Z6g;>0%_
z5gishWIS_y4tqvdc@$ZL*`cS~t8<A&x2V*b+Kwm!YS1a_G^W_=1#3vi>i4HEIAgMe
zp%p6YQks5TP7weUD@p&6K{12q{e2WoXupha`TTIZ2OxM3S#u2x{HPTa&WP#->EHMM
zTksMxeF$7cG9t8;L72c+noKt%CG=o8mfG@6gL^V8Auv1^35@2la%Dqg%1nQfWD*_w
z(=2PwL#lh}haK|+qaL~tn^JCXNZVD84#^D_{?BN$#ZzctoR^m7Mrm7TR>TVO#N|la
z`q1f<6kA6=@aQ<+{hAo4!iRmiRyf{KMx6(-x{Q*y3r5XPc8Ffn8X8!rG_`2HfJ|&{
zJEb`<Rqs2RRoza2*m<hCexOZ&n4%jlXNi4fugY8!A^Ol~=G1Q6mgH20;Yr+FG!ECt
z5%KN;fW4~yA6qvmJ9B+ai=FMZ0|ovV`1^VFNw2CC%La{Xzely0`IJb$_>RdYAPm_C
zV2#%i25ga~3vWW!m;K!uom08T&ejb*;|YQkrt<UM@uv+a!5dhP#~BXraDyQ4W!D9_
zf|{)?J&c-<mJp1Ux^jv}wgdDC`X!Jbp1SS4gRx>Gk}0di4MKm<rRZQ=%xn~pARgG7
z$Dr%B5-vi7J4j6Yuq1tD&4Yr`M*gT|Ej^o5duo{KYU1;|F~s-SA3Nu;*bXa6Y_NE@
z#Z30Dx@rrm@%FhcrFXiiVUI<x!ILRHcRFr3e#gO;X>=jXw$T#BZzEtl8+sC1m=pzB
zU9pr=a7I)B*-N)pB{wb{we}!z@p+7XHyhCx4$*6{9(a$87!<&QP4R5uXfNHj-SS0m
zfg+@bb#`ZkY1Ge|vV?p#97*CDy!t3S&iDvYJz}^c{GZ9Cg$^4(R<sY6A+ot4JHHxr
z1Vqq?f1SqRdi4Unqg92G8VRYGQGXhjTWXyTdJJ5V2MFU8m0`oU;`<{@&YU}dNs*Q9
zS*;`)o)uol37);<>_X>*{p#?!afz9;W#V%r5hg@M%38ZUpvgu_^yf8SBjrL>a7jxP
z>At_;U$+szWlqDR&I+QdVS=U~rOxEW7W%#`^_@naEfwmKGV>6cjjyt0Xxh=?8S#eD
zik%l52(@p?{}k=IGL^>RaAJ*9y0$)e*SM*^?T^Q(r+6%Ia0}>hN$_F=PP*uD9|CSN
zyvB-2&<GUh)Jz@0@z{Tp#~_ZpeLnH^<)OwJ(%EPe#S%TNHD!Vxxb+8<V!UY)-B(qC
zkn6CIc>gMwpc(@<Jq00i0y`%t8}}YSXk1y4sOXuR>s@HP=3-s5Z)(Td3^Hx>(gu1d
z{<}%t3A3)#L<D~;hmS?pSZvSz=Txx}vTcYK<5i7W%8txfsD^Cgj%Cwq$6?y$(K)1p
zoSQA@eLohg2Uq<-hm9drqqxj7kt8n1`6#QRo}2tg*l!nNd{fW(uMIk%I}vN2N0Ch@
z{6RPCFTpXiysTqWQCi9TOxBpwst!o`HcaO^GN+|i-S=8yq+DSx!GuY%yu(JNTm)d&
z>T0{=dl89RD34ZxO9ou?Hx^T~4P~v^{PUL$3j^3r?5yr5Xew1&%LQV!JvO*=iTg`d
z?B(xZ6eoncX16qFQCo=(TNg_D042-{9Wj9k1}0^(!JAf)ccEfrfqyw-Fd*GeTZzMK
z%t51i{kwI7jD8w#5mmV_3%7is<_vG(-${_Wbwp|`F_z{=zwn;E`YWxIkrNN_F-@SC
zrw%9sngzLzAbU>iJx{hY<l_x8ML!86f~GY8r_3Ftm;;>v@C>!)=dylrYGnxM!9L^4
zSQkIB4+SC;Rwh%a{f|+y?3nmn;WTacoETbn&Ao$Uie%UxdsZWli#tlO@QLX}df5HS
zG>Ic1s=A^R2wo_cystsM9zO^SJ?;UEsw?*2OnPydA6UD<9Q_*rg~ES~qMwOc&XRS%
z<HJTo2fyw?D0Bpe7~U%e7v@-YUT5Y{-{>N#fYG?Nv)WR4v~PZ!E`)T-0bV^qn`E+(
zb0|UtpC`K|Z>6*i^W~Tx%xN8CSj&N_z2tC}=Znv5w}%SZ1ijmZ!O7HSl9(SH=#<BV
zwcG4gFhJkaoERJTVKC8zWSg4ph>jUqj#XNVU`43qT+i)@NZtqK$^G7_^wLuwc>Du7
zX3-@*4fzB-|0>%<f!G46m30S)x3m31(rH0JX$wU$ar)}ik;BEC28A3GZqNF5O_O&0
zcP?Yk@64RM9f9vpT5q0d<DuXq`l{@zZab=ef+|jzP@$t5YuKIQ1AjWbxJ%mm2)KEc
zT&+xySli_(kq#(&P9GWcF=h@LLG#Mmr0}SZk-DT)zn&qQK(PQ^7cw~Q8xd@`B6`O%
zbbj?UuoIZ%!{e4vFP6c`iwKnoNMy^0#xB+kI$NHXQNQtUuoj_OFd#(T2TW7<1!t;g
zP`y52q0A>D&1F;^JiW%gJ>mb%>WufXvF>shI;phfI}I8QC(x&2kTZd<IdhgB_W`_c
z2I+KO^QjuVM#(WXbG9Vf7cW+khGx2TkTl_lfz)X@d7s7I>!Y%o>-WUAla-<m`G>cJ
zR08CXt|1W}UO=n&FA+T`x9*?C>m9B<g;Y^9`zENh!Z?UHgP#`i?)QZvHTR!cN#Y39
zwuDrkb<ByzxhgU4KwT;2@Z|P}odivg9KJE9gAZ9m(B6OL^o0)F3joh=HnYs@$A1??
z2=6EcA~cTBR(c~ud_eRj)i!e7$ianpCDSy!h=pOan$+4M+U5~y+9_K2^J-2kLn26q
zLV`Cxy^>@B%*#?7`o);*@TX=J9X(8yO{vk}37h9pcx&JHe4VErM8}T>-=Q0lC1pCT
z#p`Fxw2Fm)$ki70O3ggKT%moyVzzZ0u^&t4<5p9{`#H;Yxo&32iUpu#x3Uj1$LjHe
z#%}39L3K0Z8E4{9+4C1mZvM>0&FK-?;b{fc5gIJW=iyu~8eDcCj7p<j@WSFM?8H^A
zBE{!sx#xrwBq4z+z3dbT|M0q<IUbX@IOY+vugZg$Qk6Gfl{ia$Y41^wo@g@BgudJs
zr0@Dl;yKptN^0YI_Ol<zGGHTHba`divOS>j7yFhJ{}8Fe1(aB#uxxl@ie<w{P#hYg
zP3c{seLBm*$~&y+{>2J*7<_)A1^E9*CGJ3l<c#zhY~(pS@`qEWtLq_T9m|p7d0~Xj
z+7hhdU2wnb<ED7wDfuTSRAD;E7HsfHI^B{s!1DN~G(3=pQ|y*3n$6>|loW0Aav(li
zw<pK<E<a3{6BBI<_y77Cz1e#+<r|%!^9i@)eNW=`P@iMuu!qc+>{LXF`fX+?L9>Ln
zU{R4s3jN0f-|UJG5vOuM_#)7c3Ek4`&tWfg!P|NJ?adR~It&2vOcCZM6X9u|AE+@x
zIMvC`SHS$*$Yl_Bb?_#WczeUtV?j=l`#@$D%M)wAvoIA*3>HAerbaH1b;2wQPq7Xs
zmof+0cP0CA6;bD3uJo&i+n<^)J~$q9Ty;f@TKdRh$+SY(0s8IAnbXa?`=e#|C*poM
z<VMVCd>><J*yJ9$g+j~PWMXEr2`eN@n`))ppJFqrd=w{NMO-ldd(WtFmqr@tdRHiS
zs?36mkYs~Qx~4jXXR`mus9rR9{&dkk6F?L%F|Gu~!pEot9!24L*>8=HJOF6dO9Yd=
z_<q9WG_6H*)AYoiyN!}ix_pRH;4!aHYp}j{pQ@FFs@cz-P+~anSX(S>QZZ2wU{kKG
zwg#*HKGcp=&d#8`cX+kV9hPnT;T^3lpHzpA5Kd`Uu`52}M>TQ5(D$7mv<&$OZ?d>j
znrLu-5-o|Xp(TZ1mAd`u@4jZ3b36laKO<(Ai_;G|9WJ7nkz#>@3^YuU0C%E>pC@Vq
zt|LxMC<`>OR9d!<Pu}>(2t-KvySkzWyrb8+nNO^>1Dv_8=>F4z$;)D_1;2r<c{T~t
zP;|gi7CR8d#0wKb5NR95V%AMXg5~YqwymT-w>IvS?+vfGt#7=X_SK%<g&_t_5D%m_
z^JQLLQNv+ZG@KvZj<(RJkXLP~!|tZS;XfagI~nZy5ObW|j?xS=)p5V^Rx|DY$W~Zw
zRxe)w{8OWOW=BjO9PCHPl9-${_BrN2l<1mlj=P0SK=B8}KNYIb7oe69?Als?C7oe?
zAm%l_O<kCxa_<Gj^9dYvNQPY89}uO3Q&98GlblthR|7C1Y!rzFcvT%7!UlV<jK@_5
z$NPHy?PW9mK05tpV)3qfvgD!YnauS9gt$>?gW7+@-wnK`b6mX5m^YhDrNLm$7TT-i
z0tX84-amd(e0^XwCalxs1X~yZ<ZEHoYz<B$w{>>Vc+(Az*_jG8;ZIPHc0`enUQYv#
zV*^n`Dy5i;=tQf^OZp{LLie#igT#`8dfS5_Lg$ZH0~KU7y51S<``mWx>{<~a8M-oQ
zs@};THx3-j0w4`+VCJZUddYq!<7+H6sW#aaE^+sx)n4ADOzEpdArCbzo?A9NxqQ@6
z;hf8GU=C{TvS`w_|L4?VR9HAFJMH>`Z!`0()!Auw%jA_+eMC=iX`D$eni2K=0H&N+
zZizXdtgw#t%y@Iyn5a~+Q9)H%r^~SFub(j)at;rq6TC8(I_#_IczO9n1uT{#h0fTM
zrN7M`kjpR~uz%P7t)v6qkOOaMto40C*?mu(GI<;W_t@9F7iWwzIsQbNT1BfH_~+Yz
zq8^2f4X?f?{!t>rF72PI#6gEdG2Hr`wbPijRakYQ`ST}tx>|<L5aTWPbQVYOZRftT
z_22B25~UPeDP5@rh#W3a!CE*-6hyM@CUaq&21V{RIl8U4SR<d&)=r-_F5h_WSoQCZ
zcxm8scpFp492rph3M}NYF!y0Je-br|1jrMbnbqvzKc<?@`eUOJ^kAtg2j>{##I4Hl
zLK^l%yP-Lj9xZW|qs}0cpj}<$Gb*%wQOIEV*9Il(D8;^19}6wETc!rb9dp5XApzpG
zlwR>E=>(TEjcf>*1`%uEi$CL*P>Pcd9A|m7`&B}iI$A*$^d!ZO0vQp}CdqC{g15uR
zY#u#S=?2+`+4<{1l8=OD9<ibW5yizAS!>J0?J`rccuc*^9<puA4%7r4d;LRF#xl?2
zcfebM4)@7vF*NUK6OJ`mIfw%V(uaq!iKvstY!~8}JOSmFE*d{XB1ajK<NHO7&6l7R
z6Wj!B<x!5zMWAPPic#%<J-l++GUu;q3{~4Pb{zjm@P_2HzcFA#&>#Qg)*n5OkI?M8
zH+=8;D3xMNGcLZ^Pp*)grCv-tKuo#Q#oaV~{C#|gr_zukN$CqEp^uIg++XP2c4;OU
z{?Sl3#=VLbV32v%&ZG|5G{|NA2Bd9a!H|bieX!q#5rzSN&}tSFGO~lz+2&m55u+49
zu}4U|MX3eUMEo@Ew5y-hToc*R?K<ec$$UO|`8F=Hk*Zr@nV7D6z%ch)p}iU46<(s&
zx5DI1UheN*ei9~8lgjBV%nH)%=3-M&lb<UzMe}w)bUG@91}?vxlVgIZRf4NkgsW8s
zhyL_rGxS*SG;ebgQ6TH@3T*LY^Um8+*&z=v-Mw;bb=7U^vJ)`%1>gx7ziz%>2Gh#&
zzfBGKL60{?LY%Xj+BmRoYX}#3)uclquVlt}xCfPBYQdb0vFBY~1OzwJoqN^T&@mjK
z6vjD68KJWgW>CTT7^7(8cGG*qH-yAIVy+%f3X<2Lg!^Bo1X)7LfmC0N7VBjFk}FvP
zj!wZ?T1c-taP67pbaqs34tu6abd}7YMI;4QN5Xc3WA@YghN{tx{!PZcAbVR>;ikWA
zZ~;~4D4#GZaXKrd#1_c56(TiW&ZGR*#UXhi?|ivryF5Gh!Y(mGp4iM7#$1Bx;Nk9Z
z?G$Sbh&?2oT_j8LIXh!~K66yMgdu6EDyNzf2KwYrk1pou!6#C@iVplG@=M`y&g7w(
zwHe&sWW($pSawqpmCZ<)R#BwFKtB1Nnk_}Lsi9J#LmK>0`wi;k-_@#UacdPYzl-)9
zGQ4UPYpQzRi`xI|>B{4w>fX40UuG<0ETN&ig-{bCTQMU`ku;dBLqd`*naY}sVMK}$
zB}y1d$d=tO5%H=-k|iWc24O7mJ2QHJKJ%aRJoj0?=X;)W?{m+^!=<EXz9yC)NcC&|
zUFXU_^L@Y^UlJIAKN5a9cLV;-XC?X^#B?^VW#Dc-c3<yNl8Qyzi8eVOk7Vl~C+kj{
ze)*;}6Xdne(cc{p->X#AH#MA_<9OR}VLrKAFOulo{pHn7$IDX>UgUY&O^4g(TS*<7
z2DL*WbuL|reQGYXmXLUVzUF#3O&<D1z(M6|VZ&h1hqn?EJRcKzb-&-*ChBwy5VntM
z4I40=e}1=;={+h-EUR8ybHaJ6OAwiz?gDW+9j)VHWZIe%%(OEyB@_11vCc`JSh}nY
z4iP(Pun<e)^YQcbc6{l_*R`h%htN7YazZ)JXSpaCRRk5SN%$`J_0D`u>Kv*k-_LTY
zaqCiN@;%NsWk$;b?#l(=_M7@EecTS8{WR~dHv7zVcc|}xhL>D|wA)OT;H|csNePGb
zvL9EPjivQtO#jS8gl;q_A!FEP#Ma6WA~U)qMaE+_S|2@5C~<v##63XL>t&p!J(rac
zWCZI|B8G=w&&&cfN{C=Avr*R+O%<&0COpDSwDw6Btj-hBhheqntF==P9P7Ql>Enp3
zx><G1B&Ny|S>^4Ob<Z&)K;&(NfH{hRocy8RV;|ms%--q^RJ!-6;)!84HnDAU^szIW
z35jAKd#sNp2)WiBei3MPX~V8W4Rr+Czg-)U#ABKkztxy0a^r+@X{PhZhj9YYrwmtY
zo+iJ_&hs*<aXHo2T3Bi;c?2d~*XVvnL9qQ)TLNlu!PY$_FhipM-mj{VX}y1>KX!Vz
zQ-CW(>vLsv3zjQ225747ndRxI0vz&yP>W&#1X3(XD&KD-Z)F+z+r>^H=xbq78b5dZ
z?#!Y};UR54$Pv+~=iK{5XJ4r)zf-;<20<L~a}VN}g-ow+tzQb6z1>|m5?kx4_1UPU
z^;u1pB;1fqNoC?)wL@Lg=-ll6yQ>Q`bt3&7i>8GsYdn{;e2(WSk0$+u&v#_yzw+-r
zVw~V?b-<|UgZtDv>}fqaOns7iNz5c+K;{-Td>JphUU5#!m1t8UOdV9M$Dt!!&?6UE
z)L32MT?2}nQ7^C{D->7EO$40hYA<@2pZe8p;CpJZ%Ux>ukJAGgRa&uwM-$uN@iKR0
z%f(*~FONQxe9I$%C~7x1bLTMM>u2Sp!Dgzmsru|-(UfcG({9y%&i=2dk;sF?@loUR
zSZU9`@?zz?3)}f6BSJhu2k+zl9j}$zg|=B7`tbAH-1^N%Ti)l4Evt)(dmoW2%{<FB
zVhH==)XFu7Yt4&%L*B#;+Tk@58`@OM9cqS*@!-bdFTTD2MKUp`$~!yByC!Nxw%ykH
zYQawk|7qp-RO#^U)5kV!xMpDdA*0e2(r1MrX>boQm{3w1MQhEBls9|S)BAmAPfw5K
zHHr3t{Eg5rk!yiPB|-};ao<}n=)6<ZzdX=Y{QQ-h#FCIiu;iD{rVA@(gPyoE+YRrm
z6{a0L4HPX>Gd$q?c%&95uKbx3Eyc;7OUH}bs@yh>q5rIEu^?M?SV{XV2@=bu8B=Po
zBu5tQ?oyYF*e@7o0u%NK!_3G<*If=m6@7Fps{iLf;ER<!yxFOIqYtoM>nray1Y}Mj
zD1L_OQphB6_b+9V774)-$6cW_VvKink6bgAm86j{K~Fh5*=_qRXV<cSd^LyS)D4bH
zZ#%C_hSY*nC**XUQXBitJAJ+CTq{f}#+LNW7sBca4qqcoOuB|l5195^+fN)6dS5DH
zB1;Uk)4d<tT6&gcWu}Di=v@reL`{a)W7)+}K1;q&{Z%RTWi8!!(cyOBe54WFN$8NC
z`BOowPEOjFxv5$7fQ4<<ScxIWC){qhfq06|y5T`xkKy^IVzWVvS%){fZ_jJJg@4!7
zFZVhuz4>$?GxVy149<KwwlA^W?8)V?==cY2eb394luo|z@D<k_qSkGMw48?tdCV{L
zslm3oSt;)!dK0oKTft+xFyRU9rbxJK>@xq$B-5fL^ROB1d!MhItEWhO_u5Uoz-1B{
zQ9E=zj!P$8SNF(cJsq-$)CpXz&L^A(4!3b+|G^T5%J1rbD}^PiwJ3e05)%B=?f|qW
zGUAb*+|ZN4L5Y~}s;lAGj3lgrEAD`a9}UhbXp&3`#j(QkNkTY%+9F2RB>9l(#19Ch
z&K&gOT#@IYUU~X!{!7g@#n_{_on5jJpXPeMEq|2n-51gHc=dd?L9(_|8vptwo!ZYE
zMFf#o<ePoq8LFK>h9;TXT8ghG3|jM2Q5xWTh`5>&mcuIv0gOy1zd`>#&BnmK7hSGq
zeT%j$UB9s$Jn0oF&5#5cj1ji%nt#IrZd^8C|I*hl8if-Z(yr(16=HrOEUZFAgN{Il
z4DLNx%Q|@0-&ZP<iYnDjDHdot$C~@sZIEHsFk#V3Z+PgF!rP8(uO7CV8(;anU@P=p
zWTi__vh|VWlfny{SI!<#l_^M-IruQ7=E3vU>b-}4Bexsy@M&&^PWP5?lFWMIAxt>A
zJ8%3d$jn9lgwj=x>ZLOy2U5YRe5igeId*ji{<VrS2Tu1#64wsHiFi0Ib1WS*IU3)g
z#iCTyz)h;rn|@*)4k<*{6q>j)ZCOj9yQ;AkK_{57b!R`ZSWN~95(DcNo;qO(#j&l-
zdt;(`<BMuCE*JVVfn1`0#%Q;Di``QhPz5Ymq3X|IxAz4|k(pxn;q#4Zh;0{i#QPmU
zdG*mhsL7OII+>DIIXa$Ybl48PL(tTG%DQz_&P#8WsRNS(=t(K`q$*5+3Z4w*^2aFv
zb2c1Urh4!WOj26scP3A)nv8r}v9QJqmY`{GaWHl~oF#?NJ66dPp)DAfiD?5Dtcm&r
zOL1k3utCcR#(~vQlh!(3OIfdI7H2t0<L=M{m&8m~?g(d}MGo%)gtwzt3exnjy)~Ux
zAy82U<8%|sdU7=HS*AUSOBvRrjz~y~(Mx76Bs5V)CT;*%MR)XG?U?r}3jkeS%%X5K
zBY<EI6ItDHbDBM;*Lb{oM{CU=G?j{7_KK2XB_uI8!w8z<3Jx@6%>rWJYrZ%I`~Gdg
znmP<f7Znx9$RlkYhpq{;BB=R`(C60A!%U=8D@6blGE|gh>-e79p#XnYGRk<@E<WEs
zSaO5udQTpx3MgAhIu#}iL_9<nt|Jz^5R3dfx>kq*j9fK2d3RQddk_k5ja$N)xFeSF
zVyMp8<!!ErcX`YVXLCn@E;fH?E!Zl7RU)xLUG8C4EY)_$b?LEZxPO?tDCWO{k0Vb+
zrx(q3MOmsLm~h<Hd=jy!+6>+;Yi->bVT;l~jg1swM6fIl)LRUKu3@29n|5M19T&ie
z23Fx&RyDF4f*xiC7SM@-Tk%9HW&LimHT%Yna=$TUdNE?r7qN)l$##$bI6Xs6u6y4Q
zGkCkUlm%l!XDrjFKs<lk{M**?vYl8+jXa2L&0&Fy7&x9ll%_`(&Tfl1U4rTw9k_S_
z4<al(aSgg=-Mp=Y_zV*P_{75tXVY-V%?!+hQ{By(H}hSi{O*e_nu(uj+o}jqHW_;6
zf{qXnzN`SkeaaRWU7t)FX$Y9x%l?w_iWBKN-uC+j9$xU;7d<(EnS2Wn$OKD)l3b&P
zVBBjQa%Irhc2fb1ewTlOTcQPX40+?9{XHD#;k1FT&LG0A#-LPY0xy9mzyM!23LG<p
zr8Mk@DK}`7_$UH-d`U+Y6UJi3>WZ@*OaQ?aOv(V<jOn1CR0e?a<#?zJ)0@m!y7HSX
zqfKBb6)G@0Rl%|h*HaScfc>Nkm(0cL{*@+{y|Ee7!Ayc54+tqUrJrq4!$yK}3QCoo
zvR+dm9a0)6Id;d#;>%m&qxSCW`vXXe2V&0owrL;2-Vr6ljMQ)n2sCB=aWfh04mE#a
zf>UhSB~wWx&Rc7gz3AM*#B=GoR82MpCv;66dbJv49&9LSnGzz_U9IDFKnAV#XmYm~
zH&V=58Oa_f4~1x;m(55dsch$OLKDW*O`f@KS;?$RWn;(ba@cTwx)YhUE420<yjhoP
zciU|WUse_a(!!rh>091s9P8cv#VklQ-AFc08%*SZgbsgl#}7b)%xSW6CxGyfvgO}g
zjYD4ACZ|#IdX6Q&e0#^-o57ghGPG~nQb%4~dSJ97-YvhOt|qF`MjlI$=A~*LfNvS2
zw<<{Ww4Ka&fHcknCg#$ApBE^LX({qZKpp4eV{dW_daLT%q8B-(Q<L}LAa1FCdcAk4
zT&FRU(6j+jjeDKxn}v5m)gMD|kxBJ$fU;_hs;PF(U!*qsNo-H8SA0h(XQU5Q_hfF!
zHFu#>=#2?la%7Tao>$dHeiiGWPg~((%d^%9t;^d4DRXx?IGPGHHHJ?6K9A>SL2UX-
z07Dyk)zB^cWi_nWc}w<BLvZI4VLp+UK~6za;kt%4oB4+*CL5o!=RE$ngzi}v;5uXH
zZ+;R9r>#+cpeNVXWPFWVSUJ+wkaL;HzW$t~&r7JpoKtz#6*)DUe<AuSf-mVKrg7$a
zPVdiL(ed|&^qC{{NrjEL5(9=sp0$fmU9yG)89dUOs6UH-c(=4jyTA?Zg-9-gfO}bH
zN!T0;#fW1w3N!J0@}^>8l<A(U8u_lnVV#>a*@d*nhL@$gicSq+U0Q~mCod`iHN2bP
zNUP_h`u4w*!dOfs*{)xKCxR>EEZS<1)E&2fzTS&_6k7;|DDQVMILq&G>rr*k+94<?
zJhHjzb%;g4g)V;)E{H=7TVuD`>D1;2>Ihdyly~CHDn&7=P<=$tSUI!CCV^X=ml&87
z$VmrM8;C`p9YG-vh)6n{M$gyr7OA23@4orJCoE#}Y7d#$n;b?Uyt}Rb1>8vvJAsZ+
z08p<5x9gb`m|7FEK#!rUyN}JlZXMISoPJ>6RtbV6;#zs_20lA@GmW$vP1;P^nH0pp
z&#_c2W!<D%^u8Ng=fPb6lW!suv=sSzYLSY64)2NS6rc`rZl;hnb4m4`JHenbXtC5k
z*kxX`th8T&EiPq9VT08PL1V2hMdy+r$9K=-IhcfakHO(H(WgM}=m+%1WFp#}hTO*}
zg;V|Wybs->GfJEtJ_!IoJskd$Dd}Ein$<frqS=#3AS6&C6zJj2wAlR;T%j`RX3RxG
zI~shrOBFr&>_2^<#6{9MHBR?Su_u?4K!I`ZY7S|WpRR%kYKWyW;RN>NZPbc0%dyRZ
z(II3K9_~d_A4@|lT0qw<{|oWFV4NPO#+m*mQT;8adKB(%3O-xB-WI*(3lr3F*x?)a
z^iu*Mj}jqDH)x)ML8+RJ$DZ?Uoj`-%$fa=>DdV$^TJaoV)szS+`i<Zzgl8HLb?^`X
zqYL+@4itCxKSkAq8WlPRHk_T##Wbj3Y!_m2Rt;w8|5uGu#@F!_@J*|+tIaVhLzp-F
zK_qEwC_3QUZS>@yzpfK)65<L9XG1wg`G>m@?r<rg9L}ve(q`z=39v9R*OCHN`Nosj
zCQgRP*d<^wYMvoRH^NLtCUm$monqxf2O;0+jLqTHm#T>a)lCnsk*(ei?V^GxWu}73
zU5R?e`k?o)m#1-8O%oWhqHoVxDwP6W2+_g)rzKNDG}*S_pZ6iR{zqWSbs`{>eF!ir
z^#4O#Afomhy=VHpIC{DE!X&9E54C6=QitqfX-@g=zgKEdlNVv@J=O2UX3xG_RtV72
zbGSJRbd)yBVkT>1LGcAZh0iYL_?puE9=q(*{61<Z&mCT!$wL4XDa}w_GW+ZLbTMI2
zR3D4Y;n3%1oF^yXv+sy<_`e<D^M|I0!^C!w0jfdLbvA2oz8kwW$1g7%tpjd`!l0Rc
z<X+^T8~<Gm?!g{tG!{=;N00IERZJ8-s#tG=9uWbZgob{*wRQqc^Am@xvNvL|3ZnWV
zzW9SZeujMYmgo`w&1WPhkN=OFQ&l{`E?+=N=9cHwmqDFN(OOYF9=*yj4^0X6$_=<1
z+rHx<+{*>6FGonF7zz+1=zMiFX|HuZzA!dI4yj&adwsVcd1sPqt~`y(@a&_L)b=(?
z6X0WNM`&|LxeSHEG|4wa*%TlteuB6Dj_h%UYQ4=H&!wzySBDm;Vq@_VQ8_w~KfKI4
zW-If7XJ7F~(=9b|*$);kyYt@3AwK^><+wrqkRJIMytw4{fHSLdRd=+JtXFd@*heM8
zjZ-wBiE>_xd>vbvegT1q*^T(R4Hpy7XgM(uQXNp?iJIHA*Vam2qg*5Dc4Pa4DDfmA
zB8kKpxt-C5eEQj@R3~7U{ktC5IcfO`8@<-c?1I?JRG{b}9D#^vzYT9Y?9_f{AjCGK
z4c&G0?E14SFNVv{V95?`w)#bsxib>blokkdL65h+X2h0cSu-5+1tWK^K40VHpVaV@
zsuH(+iXB}}gI+?hv$aK|c}&o4qlWuwnnXc(yZ1nd61?SVgOPiVpY;0cAA_w~_k`<d
z*Ranfuw9lV*#5JP22iN95EQyFSM&X~bVK_42ecv+-A^H30%RpkP|+>-qVGk>y)HA+
z9eOtq^u$?Q<hh`8msCUD`UYcoapu&@i{jNok9J+G4!daLR!q@eV4oMAdN5-7cEqm#
flHnAIWHey!XyoVs4O>fO1ApciEA(??w;TTh=r9vP

diff --git a/docs/assets/images/logo.svg b/docs/assets/images/logo.svg
new file mode 100644
index 000000000..7d3b4f84b
--- /dev/null
+++ b/docs/assets/images/logo.svg
@@ -0,0 +1,133 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="397.00375"
+   height="113.55765"
+   viewBox="0 0 397.00375 113.55765"
+   fill="none"
+   version="1.1"
+   id="svg314"
+   sodipodi:docname="logo.svg"
+   xml:space="preserve"
+   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
+   inkscape:export-filename="../../../../../Downloads/logo-square-side.png"
+   inkscape:export-xdpi="270.25565"
+   inkscape:export-ydpi="270.25565"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg"><defs
+     id="defs314" /><sodipodi:namedview
+     id="namedview314"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     showgrid="false"
+     showguides="true"
+     inkscape:zoom="1.0083333"
+     inkscape:cx="280.66116"
+     inkscape:cy="25.785124"
+     inkscape:window-width="3440"
+     inkscape:window-height="1416"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1"
+     inkscape:current-layer="svg314"><sodipodi:guide
+       position="-146.75933,121.72017"
+       orientation="0,-1"
+       id="guide314"
+       inkscape:locked="false" /><sodipodi:guide
+       position="275.80377,60.668961"
+       orientation="1,0"
+       id="guide315"
+       inkscape:locked="false" /></sodipodi:namedview><text
+     xml:space="preserve"
+     style="font-size:117.641px;font-family:'PP Neue Montreal Mono';-inkscape-font-specification:'PP Neue Montreal Mono';opacity:0.99;fill:#d33134;fill-opacity:1;stroke-width:0.612713"
+     x="115.73952"
+     y="91.977005"
+     id="text314"><tspan
+       sodipodi:role="line"
+       id="tspan314"
+       x="115.73952"
+       y="91.977005"
+       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:'PP NeueBit';-inkscape-font-specification:'PP NeueBit Bold';fill:#d33134;fill-opacity:1;stroke-width:0.612713">outlines</tspan></text><g
+     id="g13"
+     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
+     style="fill:#d53135;fill-opacity:1"><g
+       id="g314"
+       transform="translate(1367.0913,152.50405)"><rect
+         x="28"
+         y="644"
+         width="28"
+         height="28"
+         id="rect1"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
+         id="path1-7"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="28"
+         y="112"
+         width="28"
+         height="28"
+         id="rect2"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,560 v 28 h 420 v -28 z"
+         id="path2-0"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,476 v 28 h 420 v -28 z"
+         id="path3-4"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,392 v 28 h 420 v -28 z"
+         id="path4-0"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 84,308 v 28 h 420 v -28 z"
+         id="path5-8"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 364,168 v 28 h 140 v -28 z"
+         id="path6-5"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 336,196 v -28 h -56 v 28 z"
+         id="path7-1"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="M 336,112 V 84 h -56 v 28 z"
+         id="path8-6"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 364,84 v 28 H 504 V 84 Z"
+         id="path9-6"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 56,672 v 28 h 476 v -28 z"
+         id="path10-2"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="56"
+         y="84"
+         width="28"
+         height="28"
+         id="rect10"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="112"
+         y="28"
+         width="28"
+         height="28"
+         id="rect11"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="84"
+         y="56"
+         width="28"
+         height="28"
+         id="rect12"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
+         id="path12-1"
+         style="fill:#d53135;fill-opacity:1" /><path
+         d="m 560,28 v 616 h 28 V 28 Z"
+         id="path13-9"
+         style="fill:#d53135;fill-opacity:1" /><rect
+         x="532"
+         y="644"
+         width="28"
+         height="28"
+         id="rect13"
+         style="fill:#d53135;fill-opacity:1" /></g></g></svg>

From 4ee420337dfccb9b75283f40f9917d77dd1b29a3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Thu, 12 Sep 2024 12:47:57 +0200
Subject: [PATCH 067/505] Update logo size in documentation

---
 README.md                          |   2 +-
 docs/assets/images/logo-simple.png | Bin 0 -> 2196 bytes
 docs/index.md                      |   2 +-
 mkdocs.yml                         |   4 ++--
 4 files changed, 4 insertions(+), 4 deletions(-)
 create mode 100644 docs/assets/images/logo-simple.png

diff --git a/README.md b/README.md
index f02c126ae..4988c77cc 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 <div align="center" style="margin-bottom: 1em;">
 
-<img src="./docs/assets/images/logo.png" alt="Outlines Logo" width=300></img>
+<img src="./docs/assets/images/logo.png" alt="Outlines Logo" width=500></img>
 
 [![.txt Twitter][dottxt-twitter-badge]][dottxt-twitter]
 [![Outlines Twitter][outlines-twitter-badge]][outlines-twitter]
diff --git a/docs/assets/images/logo-simple.png b/docs/assets/images/logo-simple.png
new file mode 100644
index 0000000000000000000000000000000000000000..2c14b4d9f9d8e38431c44c004594d83ad6a111e8
GIT binary patch
literal 2196
zcmb_edsNa_9Ea7cY-(l8=XAB1bt<(qEfs4EmmU{A2rTe<GGL<Ot28OJJhRlum7B#3
zoKij@P*d=Mg3G2k#fE|S#7f0PiC7UnLiclZoSkFY&gqYP?m74OyXStspU>y}xi|Fy
zY_Ea7xjqO4GVu0-9s+^D2Ef?8R1a9m7jLWvre%>{M`J*sjTtioEMB`E_yc4X>v<#=
zPQb**1xFtT#l^+h{SbkVITjpw+>Q{9&3@``4gxJ*>ka+paQt;;uP~j8HEUFlkTB=W
zS=G1G?0H=;ojO{)hnx<JsDIsDo@;m4<bZFgh0fTo6Op`nlm5EG-byy1@~5BWHyn*m
zxRO!+<=LnetJ6@9&wC=OH^P3~C2WdmBYBKeRTg`W$$QeX3pBD3;b;&)zKS99&{#R7
zmfpwu+No^5s%M-8%dSBeUAUmPVx_^C>&$mUosur#Kr2O{b$MXu>kpmR6V6%;D~2a`
z`B^BrxWbCsP+f?FI*U<l>q=J;5SpP#JtUh$*7`$YoF~-JQqmea6g){|hUXsXR8)6Z
z;`E?&a_D{pwU%C?HtNJEP0D(Ga#}G#8bf({t!yNATYthQZ_aurh@;&Vwv*`eB`LWu
z8%LI;m?0@^YDx7CsLBf0BShdW$hn&HjEb11?+MEOGOCXtKmtn|S+-);yNY^!&_I3-
zq_Yp@Ub0m@lu0Q~=`&y@+PLEA(_1Hqiia(avsns%raN8Rm7}z?_Qn{>yN$T*9vlWW
z0xYN@V#mkCK9)oI&|^3qTh@IwkD>^qDmPoBC9oQX6ibvSH+c-EGSc!6)^ZaNF9>q-
z&1oj1eVaMkc0Qix)_!?R*`!kS@}J}AItR+Igv1x6OkSY~r&L2_!m+@M;bQ|BN$M}O
zgVHH0BPy=X;!;WuJBW(D^Qd90P3-2k-;zv6+A>3FE#(1%D!jIc`eHLuC2oZas0ru$
zTe6x>n2JA&v=0SUw712c3*wCnoQ>q^R{K6v{uvh6i7YoNk-uu2<BoYY0Jgc?2Pr}l
zcnR`R%IBleK2O3=$03yK9a$4O$nw_d@adM$jGl`bHo}u9X+9XkHR)Qzz4ep6oCnoz
zJ*I)K96!E;#;UQ2E*fbqZ7A?9Y2K0tap|s&Z4{$>HtexZbVl~_yWq~WLdFvMWxOu^
zn1!J9*E8*XksbDjCF4cf9hU&@=SMc{^sf_%hcwc5g{{l!O!VE3$-sQsqm)ohGE&1v
zSc!<r#yzZ;D5mNbz$%otL!&2sK<n&k<(D8&9Ov?SP)*Kt`%X9njv^$(Aq|?v@9m%k
zkj&F_A8SV#thP=wYQM(1$K71!2(<o>^<?!;c=O<LxGL&8hujd^A@FFAk5E8@^y`BK
z2M|LAvi=<(;7`sOVjGHg1VNxC7+w4BUw+ECqDr7A;t>!aW9sl9;H(9LXO@I}U8K#!
zL3lMMR6{A!CV9V@5QP!OpNa97*l_wmR$?JTSP-yiy=bC~lqFhNi`!hnehlyGecu_Y
zfC8HB-yahHN1mhI88%A77Aw;f*X=WGN8QrFiOse%q|Rm(Tm^=kq5!V(W(HE@cZBYo
z+W8PO;-uO^JS8U9t!^k~DFr$}IHvup_P>8Otf72#p7}a2gqsV1o=ghn631VO(Uy%1
zQ*^J34=Iw-ijh2cF*x&ya=gg5k_O-ego;E!+&{7-yoeaSZZPCwA;X#Z)Cbd0>R*FZ
zwQ1?Xq<fBNZDNG!#NJFwJ)^+oPK-0wt-0~a+2jY^=rg!J7zU*gGFBNJ$wlgnYo#Vv
zDCE*zuDs||_LL|H_`geUmjAbI!?65CD)&>570tyLMadjGsd6R(ZeYR1fyHVOnLEx3
zk$+)(o;kcT%E{x|K<4x<#s<)m=Sm_F#I#RqNGy(;O)Wr9<&-TPPXoK`aGT5*Q{+2=
P7X`@M69#2}8+7g;FTJ7w

literal 0
HcmV?d00001

diff --git a/docs/index.md b/docs/index.md
index b19ebd57a..e7409c0d3 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -9,7 +9,7 @@ hide:
 #
 
 <figure markdown>
-  ![Image title](assets/images/logo.png){ width="300" }
+  ![Image title](assets/images/logo.png){ width="600" }
 </figure>
 
 <center>
diff --git a/mkdocs.yml b/mkdocs.yml
index bf863085a..c95905914 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -19,8 +19,8 @@ theme:
   palette:
     - scheme: default
       primary: white
-  logo: assets/images/logo.png
-  favicon: assets/images/logo.png
+  logo: assets/images/logo-simple.png
+  favicon: assets/images/logo-simple.png
   icon:
     repo: fontawesome/brands/github
   features:

From ba8a52bff3a07747bb2a1861c7c8e94539c27620 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Thu, 12 Sep 2024 11:14:07 -0700
Subject: [PATCH 068/505] Improve sampler docs (#1141)

The docs did not fully convey how to set temperature, and I figured I'd
update the sampler docs more generally. A user was having issues with
this in the Discord.

I update the documentation to include a few things:

Changes:

- Brief overview in layman's terms of what the samplers actually do
- How multinomial sampling works
- Added "Parameters" sections for beam search and multinomial sampling
- Added more headers to break up the various parameters in multinomial
sampling
- Provided a simple comparison table to help users choose from the
available samplers

Misc doc convenience stuff:

- gitignore for mkdocs-generated folders
- Manually triggered build/publish GitHub Actions for the docs
---
 .github/workflows/build_documentation.yml   |  1 +
 .github/workflows/publish_documentation.yml |  1 +
 .gitignore                                  |  4 ++
 docs/reference/samplers.md                  | 77 +++++++++++++++++++--
 outlines/samplers.py                        |  4 +-
 5 files changed, 78 insertions(+), 9 deletions(-)

diff --git a/.github/workflows/build_documentation.yml b/.github/workflows/build_documentation.yml
index 7902c1716..639e8c7a5 100644
--- a/.github/workflows/build_documentation.yml
+++ b/.github/workflows/build_documentation.yml
@@ -3,6 +3,7 @@ name: Build the documentation
 on:
   pull_request:
     branches: [main]
+  workflow_dispatch:
 
 jobs:
   build:
diff --git a/.github/workflows/publish_documentation.yml b/.github/workflows/publish_documentation.yml
index 4679121b2..d861424f3 100644
--- a/.github/workflows/publish_documentation.yml
+++ b/.github/workflows/publish_documentation.yml
@@ -1,6 +1,7 @@
 name: Publish the documentation
 
 on:
+  workflow_dispatch:
   push:
     branches:
       - main
diff --git a/.gitignore b/.gitignore
index 9add6d8c4..4984b18cb 100644
--- a/.gitignore
+++ b/.gitignore
@@ -7,3 +7,7 @@ docs/build
 *.gguf
 .venv
 benchmarks/results
+
+# Remove doc build folders
+.cache/
+build/
diff --git a/docs/reference/samplers.md b/docs/reference/samplers.md
index 56e4e1f7e..9cba5cefa 100644
--- a/docs/reference/samplers.md
+++ b/docs/reference/samplers.md
@@ -2,9 +2,33 @@
 
 Outlines offers different sequence sampling algorithms, and we will integrate more in the future. You can read [this blog post](https://huggingface.co/blog/how-to-generate) for an overview of the different sampling algorithm.
 
+Samplers provide control over the sampling process, allowing you to influence the output of the model. This can include controlling randomness (temperature), biasing towards certain tokens (top-k, top-p), or sequence generation (beam search).
+
 ## Multinomial sampling
 
-Outlines defaults to the multinomial sampler without top-p or top-k sampling, and temperature equal to 1. Not specifying a sampler is equivalent to:
+[Multinomial sampling](https://en.wikipedia.org/wiki/Multinomial_distribution) is the default sampling algorithm in Outlines.
+
+As an example, suppose we have only two possible tokens: "H" and "T". For a fixed prompt such as "Flip a coin, did you get heads or tails?" The language model calculates probability for each token:
+
+| Token | Probability |
+|-------|-------------|
+| "H"   | 0.5         |
+| "T"   | 0.5         |
+
+You'd expect to receive "H" 50% of the time and "T" 50% of the time.
+
+### Parameters
+
+- `samples`: Number of samples to generate (default: 1)
+- `top_k`: Only consider the top k tokens (optional)
+- `top_p`: Only consider the top tokens with cumulative probability >= p (optional)
+- `temperature`: Controls randomness of sampling (optional)
+
+### Default behavior
+
+Outlines defaults to the multinomial sampler without top-p or top-k sampling, and temperature equal to 1.
+
+Not specifying a sampler is equivalent to:
 
 ```python
 from outlines import models, generate, samplers
@@ -20,6 +44,8 @@ print(answer)
 # 4
 ```
 
+### Batching
+
 You can ask the generator to take multiple samples by passing the number of samples when initializing the sampler:
 
 ```python
@@ -36,7 +62,7 @@ print(answer)
 # [4, 4, 4]
 ```
 
-If you ask multiple samples for a batch of prompt the returned array will be of shape `(num_samples, num_batches)`:
+If you ask multiple samples for a batch of prompts the returned array will be of shape `(num_samples, num_batches)`:
 
 ```python
 from outlines import models, generate, samplers
@@ -52,6 +78,25 @@ print(answer)
 # [[4, 4, 4], [6, 6, 6]]
 ```
 
+### Temperature
+
+You can control the temperature with
+
+```python
+from outlines import models, generate, samplers
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+sampler = samplers.multinomial(3, temperature=0.5)
+
+generator = generate.text(model, sampler)
+answer = generator(["What is 2+2?", "What is 3+3?"])
+
+print(answer)
+```
+
+If you would like to use `temperature=0.0`, please use `sampler=samplers.greedy()` instead.
+
 ### Top-k sampling
 
 You can ask Outlines to only consider the top-k logits at each step by specifying the value of the `top-k` keyword argument when initializing the sampler.
@@ -71,7 +116,9 @@ sampler = samplers.multinomial(3, top_p=0.95)
 
 ## Greedy sampler
 
-You can also use the greedy sampler. For this you need to initialize the generator with the sampler:
+Greedy sampling selects the token with the highest probability at each step. It's deterministic and always produces the same output for a given input.
+
+To use the greedy sampler, initialize the generator with the sampler:
 
 
 ```python
@@ -88,12 +135,14 @@ print(answer)
 # 4
 ```
 
-You cannot ask for multiple samples with the greedy sampler since it does not clear what the result should be.
+You cannot ask for multiple samples with the greedy sampler since it does not clear what the result should be. Only the most likely token can be returned.
 
 
 ## Beam Search
 
-Outlines also comes with the Beam Search sampling algorithm:
+Beam search maintains multiple candidate sequences at each step, potentially finding better overall sequences than greedy or multinomial sampling.
+
+To use Beam Search, initialize the generator with the sampler:
 
 ```python
 from outlines import models, generate, samplers
@@ -108,8 +157,22 @@ answer = generator("What is 2+2?")
 print(answer)
 # 4
 ```
-
-
 !!! Warning "Compatibility"
 
     Only models from the `transformers`  and `exllamav2 ` libraries are compatible with Beam Search.
+
+### Parameters
+
+- `beams`: Number of beams to use (default: 1)
+
+## Sampler Comparison
+
+Here's a table comparing the different samplers:
+
+| Sampler | Pros | Cons | Use Cases |
+|---------|------|------|-----------|
+| Greedy | Deterministic, fast | May produce repetitive text | When you need consistent, predictable output |
+| Multinomial | Balances exploration and exploitation | Results may vary between runs | General-purpose text generation, creative tasks |
+| Beam Search | Can find globally better sequences | More computationally expensive | When sequence quality is critical, e.g., translation |
+
+For most use cases, we recommend using the default [multinomial sampler](#multinomial-sampling).
diff --git a/outlines/samplers.py b/outlines/samplers.py
index 8b64ed768..b1421971f 100644
--- a/outlines/samplers.py
+++ b/outlines/samplers.py
@@ -250,8 +250,8 @@ class BeamSearchSampler:
     Attributes
     ----------
     samples
-        The number of samples taken for each input sequence.
-
+        The number of samples taken for each input sequence. Equivalent to the
+        number of beams.
     """
 
     def __init__(self, beams: int = 1):

From 2a4d72d5f8dc7c2603549c53b8e807ead4928ea1 Mon Sep 17 00:00:00 2001
From: Chris Sanchez <19364189+americanthinker@users.noreply.github.com>
Date: Thu, 12 Sep 2024 12:20:16 -0600
Subject: [PATCH 069/505] Update vllm.md (#1137)

Minor typo that prevents model from being downloaded from Huggingface
hub. Corrected the typo and verified that the download now works.

---------

Co-authored-by: Cameron Pfiffer <cameron@pfiffer.org>
---
 docs/reference/models/vllm.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index 1380d3d2c..fb1c830fa 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -73,7 +73,7 @@ vLLM supports AWQ, GPTQ and SqueezeLLM quantized models:
 ```python
 from outlines import models
 
-model = models.vllm("TheBloke/Llama2-7b-Chat-AWQ", quantization="awq")
+model = models.vllm("TheBloke/Llama-2-7B-Chat-AWQ", quantization="awq")
 model = models.vllm("TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", quantization="gptq")
 model = models.vllm("https://huggingface.co/squeeze-ai-lab/sq-llama-30b-w4-s5", quantization="squeezellm")
 ```

From 0b9a3f1378c62449472cd6430d18eab9ac39c2bd Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Fri, 13 Sep 2024 11:44:14 -0700
Subject: [PATCH 070/505] Correct pathways, update site color, front page fixes
 (#1146)

Follow up to #1143 and moving the repo to dottxt-ai.

- A lot of our links were broken because they pointed to the old site.
- We changed the logo and the color but it wasn't fully integrated
across the site.

I fixed these, and made some improvements to the front page so that it's
a little cleaner.

New mobile/desktop views


![image](https://github.com/user-attachments/assets/9c068733-9cbc-4864-a5cd-763cb7403fa5)

![image](https://github.com/user-attachments/assets/ddc5c638-6129-439d-8645-403af1ec9acf)

Current for reference


![image](https://github.com/user-attachments/assets/2bcebe41-a4db-4fba-9a0e-d8adfaf8b1f9)

GitHub contributors are still available off the welcome page:

![image](https://github.com/user-attachments/assets/ce4b21e6-2969-47f6-b351-19cf91759868)
---
 .github/ISSUE_TEMPLATE/config.yml             |   2 +-
 .../pull_request_template.md                  |   2 +-
 Dockerfile                                    |   2 +-
 README.md                                     |  24 ++--
 benchmarks/asv.conf.json                      |   4 +-
 docs/blog/posts/roadmap-2024.md               |  10 +-
 docs/community/contribute.md                  |   8 +-
 docs/community/feedback.md                    |   4 +-
 docs/community/versioning.md                  |   2 +-
 docs/cookbook/chain_of_density.md             |   2 +-
 docs/cookbook/chain_of_thought.md             |   2 +-
 docs/cookbook/deploy-using-bentoml.md         |   6 +-
 docs/cookbook/deploy-using-modal.md           |   4 +-
 docs/cookbook/knowledge_graph_extraction.md   |   2 +-
 docs/cookbook/react_agent.md                  |   2 +-
 docs/cookbook/simtom.md                       |   4 +-
 docs/index.md                                 |  20 +--
 docs/installation.md                          |   2 +-
 docs/overrides/home.html                      | 120 ++++++++++++++++++
 docs/overrides/index.html                     |  11 --
 docs/overrides/main.html                      |  20 ---
 docs/quickstart.md                            |   2 +-
 docs/reference/generation/cfg.md              |   4 +-
 docs/reference/generation/generation.md       |   2 +-
 docs/reference/generation/types.md            |   2 +-
 docs/reference/serve/vllm.md                  |   2 +-
 docs/stylesheets/extra.css                    |   1 +
 docs/welcome.md                               |  20 +--
 environment.yml                               |   4 +-
 examples/dating_profile.py                    |   2 +-
 mkdocs.yml                                    |   4 +-
 outlines/fsm/guide.py                         |   4 +-
 outlines/generate/cfg.py                      |   2 +-
 pyproject.toml                                |   6 +-
 tests/fsm/test_json_schema.py                 |   2 +-
 tests/fsm/test_regex.py                       |   8 +-
 tests/generate/test_generate.py               |   4 +-
 tests/generate/test_integration_llamacpp.py   |   2 +-
 tests/test_function.py                        |  10 +-
 39 files changed, 203 insertions(+), 131 deletions(-)
 create mode 100644 docs/overrides/home.html
 delete mode 100644 docs/overrides/index.html

diff --git a/.github/ISSUE_TEMPLATE/config.yml b/.github/ISSUE_TEMPLATE/config.yml
index a396c1e76..90a4af686 100644
--- a/.github/ISSUE_TEMPLATE/config.yml
+++ b/.github/ISSUE_TEMPLATE/config.yml
@@ -1,4 +1,4 @@
 contact_links:
   - name: 🤔 Questions & Help
-    url: https://github.com/outlines-dev/outlines/discussions/new
+    url: https://github.com/dottxt-ai/outlines/discussions/new
     about: "If you have a question about how to use Outlines, please start a discussion."
diff --git a/.github/PULL_REQUEST_TEMPLATE/pull_request_template.md b/.github/PULL_REQUEST_TEMPLATE/pull_request_template.md
index bc181466b..ce0e89999 100644
--- a/.github/PULL_REQUEST_TEMPLATE/pull_request_template.md
+++ b/.github/PULL_REQUEST_TEMPLATE/pull_request_template.md
@@ -18,6 +18,6 @@ A few important guidelines and requirements before we can merge your PR:
 Consider opening a **Draft PR** if your work is still in progress but you would
 like some feedback from other contributors.
 
-[issues]: https://github.com/outlines-dev/outlines/issues
+[issues]: https://github.com/dottxt-ai/outlines/issues
 [git-guidelines]: https://tbaggery.com/2008/04/19/a-note-about-git-commit-messages.html
 [docstring-guidelines]: https://numpydoc.readthedocs.io/en/latest/format.html
diff --git a/Dockerfile b/Dockerfile
index c6e5f0672..117e39e88 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -13,5 +13,5 @@ COPY outlines ./outlines
 RUN --mount=source=.git,target=.git,type=bind \
     pip install --no-cache-dir .[serve]
 
-# https://outlines-dev.github.io/outlines/reference/vllm/
+# https://dottxt-ai.github.io/outlines/reference/vllm/
 ENTRYPOINT ["python3", "-m", "outlines.serve.serve"]
diff --git a/README.md b/README.md
index 4988c77cc..87109d182 100644
--- a/README.md
+++ b/README.md
@@ -22,11 +22,11 @@ Made with ❤👷️ by the team at [.txt](https://dottxt.co).
 pip install outlines
 ```
 
-First time here? Go to our [setup guide](https://outlines-dev.github.io/outlines/welcome)
+First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/welcome)
 
 ## Features
 
-- [x] 🤖 [Multiple model integrations](https://outlines-dev.github.io/outlines/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
+- [x] 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
 - [x] 🖍️ Simple and powerful prompting primitives based on the [Jinja templating engine](https://jinja.palletsprojects.com/)
 - [x] 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
 - [x] ⚡ Fast [regex-structured generation](#efficient-regex-structured-generation)
@@ -36,10 +36,10 @@ First time here? Go to our [setup guide](https://outlines-dev.github.io/outlines
 - [x] 💾 Caching of generations
 - [x] 🗂️ Batch inference
 - [x] 🎲 Sample with the greedy, multinomial and beam search algorithms (and more to come!)
-- [x] 🚀 [Serve with vLLM](https://outlines-dev.github.io/outlines/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
+- [x] 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
 
 
-Outlines 〰 has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!
+Outlines  has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!
 
 ## Why should I use structured generation?
 
@@ -145,7 +145,7 @@ as non-structured generation.
 
 ### Efficient JSON generation following a Pydantic model
 
-Outlines 〰 allows to guide the generation process so the output is *guaranteed* to follow a [JSON schema](https://json-schema.org/) or [Pydantic model](https://docs.pydantic.dev/latest/):
+Outlines  allows to guide the generation process so the output is *guaranteed* to follow a [JSON schema](https://json-schema.org/) or [Pydantic model](https://docs.pydantic.dev/latest/):
 
 ```python
 from enum import Enum
@@ -197,7 +197,7 @@ print(repr(character))
 # Character(name='Vivian Thr', age=44, armor=<Armor.plate: 'plate'>, weapon=<Weapon.crossbow: 'crossbow'>, strength=125)
 ```
 
-The method works with union types, optional types, arrays, nested schemas, etc. Some field constraints are [not supported yet](https://github.com/outlines-dev/outlines/issues/215), but everything else should work.
+The method works with union types, optional types, arrays, nested schemas, etc. Some field constraints are [not supported yet](https://github.com/dottxt-ai/outlines/issues/215), but everything else should work.
 
 ### Efficient JSON generation following a JSON Schema
 
@@ -277,7 +277,7 @@ print(sequence)
 # (8-2)
 ```
 
-This was a very simple grammar, and you can use `outlines.generate.cfg` to generate syntactically valid Python, SQL, and much more than this. Any kind of structured text, really. All you have to do is search for "X EBNF grammar" on the web, and take a look at the [Outlines `grammars` module](https://github.com/outlines-dev/outlines/tree/main/outlines/grammars).
+This was a very simple grammar, and you can use `outlines.generate.cfg` to generate syntactically valid Python, SQL, and much more than this. Any kind of structured text, really. All you have to do is search for "X EBNF grammar" on the web, and take a look at the [Outlines `grammars` module](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
 
 ### Open functions
 
@@ -339,8 +339,8 @@ answer = outlines.generate.text(model)(prompt, max_tokens=100)
 ## Join us
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
-- 🔨 **Want to contribute?** Consult our [contribution guide](https://outlines-dev.github.io/outlines/community/contribute/).
-- 🐞 **Found a bug?** Open an [issue](https://github.com/outlines-dev/outlines/issues)
+- 🔨 **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/community/contribute/).
+- 🐞 **Found a bug?** Open an [issue](https://github.com/dottxt-ai/outlines/issues)
 
 
 ## Cite Outlines
@@ -354,10 +354,10 @@ answer = outlines.generate.text(model)(prompt, max_tokens=100)
 }
 ```
 
-[documentation]: https://outlines-dev.github.io/outlines/welcome/
+[documentation]: https://dottxt-ai.github.io/outlines/welcome/
 [documentation-badge]: https://img.shields.io/readthedocs/outlines
-[contributors]: https://github.com/outlines-dev/outlines/graphs/contributors
-[contributors-badge]: https://img.shields.io/github/contributors/outlines-dev/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
+[contributors]: https://github.com/dottxt-ai/outlines/graphs/contributors
+[contributors-badge]: https://img.shields.io/github/contributors/dottxt-ai/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
 [dottxt-twitter]: https://twitter.com/dottxtai
 [outlines-twitter]: https://twitter.com/OutlinesOSS
 [discord]: https://discord.gg/R9DSu34mGd
diff --git a/benchmarks/asv.conf.json b/benchmarks/asv.conf.json
index f57db9a0b..92808ce83 100644
--- a/benchmarks/asv.conf.json
+++ b/benchmarks/asv.conf.json
@@ -1,7 +1,7 @@
 {
     "version": 1,
     "project": "Outlines",
-    "project_url": "https://outlines-dev.github.io/outlines/",
+    "project_url": "https://dottxt-ai.github.io/outlines/",
     "repo": "..",
     "branches": [
 	"HEAD"
@@ -11,7 +11,7 @@
         "PIP_NO_BUILD_ISOLATION=false python -mpip wheel --no-deps --no-index -w {build_cache_dir} {build_dir}",
     ],
     "environment_type": "virtualenv",
-    "show_commit_url": "https://github.com/outlines-dev/outlines/commit/",
+    "show_commit_url": "https://github.com/dottxt-ai/outlines/commit/",
     "benchmark_dir": ".",
     "env_dir": "env",
     "results_dir": "results",
diff --git a/docs/blog/posts/roadmap-2024.md b/docs/blog/posts/roadmap-2024.md
index d1413b05f..2832660fc 100644
--- a/docs/blog/posts/roadmap-2024.md
+++ b/docs/blog/posts/roadmap-2024.md
@@ -23,13 +23,13 @@ Thanks to a refactor of the library, it is now possible to use our constrained g
 
 *We would like expand our work to the whole sampling layer*, and add new sampling methods that should make structured generation more accurate. This means we will keep the `transformers` integration as it is today and will expand our text generation logic around this library.
 
-Making workflows re-usable and easy to share is difficult today. That is why *we are big believers in [outlines functions](https://github.com/outlines-dev/functions)*. We will keep improving the interface and adding examples.
+Making workflows re-usable and easy to share is difficult today. That is why *we are big believers in [outlines functions](https://github.com/dottxt-ai/functions)*. We will keep improving the interface and adding examples.
 
 Finally, *we want to add a CLI tool*, `outlines serve`. This will allows you to either serve an API that does general constrained generation, or to serve Outlines function.
 
 ## Detailed roadmap
 
-Here is a more detailed roadmap for the next 12 months. Outlines is a [community](https://discord.gg/ZxBxyWmW5n) effort, and we invite you to pick either topic and [contribute to the library](https://github.com/outlines-dev/outlines). I will progressively add related [issues](https://github.com/outlines-dev/outlines/issues) in the repository.
+Here is a more detailed roadmap for the next 12 months. Outlines is a [community](https://discord.gg/ZxBxyWmW5n) effort, and we invite you to pick either topic and [contribute to the library](https://github.com/dottxt-ai/outlines). I will progressively add related [issues](https://github.com/dottxt-ai/outlines/issues) in the repository.
 
 ### Many more examples and tutorials
 
@@ -44,7 +44,7 @@ Let's be honest, Outlines is lacking clear and thorough examples. We want to cha
 
 We want to keep the current integrations but lower the maintenance cost so we can focus on what we bring to the table.
 
-* Deprecate every obsolete integration: `transformers` has recently integrated `autoawq` and `autogptq` for instance. ([PR](https://github.com/outlines-dev/outlines/pull/527))
+* Deprecate every obsolete integration: `transformers` has recently integrated `autoawq` and `autogptq` for instance. ([PR](https://github.com/dottxt-ai/outlines/pull/527))
 * See if we can integrate to a library that provides state-space models via a logit processing function;
 * Integrate with llama.cpp via a logits processor;
 * Integrate with exllamav2 via a logits processor;
@@ -55,14 +55,14 @@ We're just getting started!
 
 * Improve the performance of existing structured generation algorithms;
 * Improve the correctness of structured generation algorithms;
-* Add ready-to-use grammars in the [grammars](https://github.com/outlines-dev/grammars) repository or in a submodule in Outlines.
+* Add ready-to-use grammars in the [grammars](https://github.com/dottxt-ai/grammars) repository or in a submodule in Outlines.
 
 ### Keep developing Outlines functions
 
 Functions are awesome, use them!
 
 * Implement a CLI `outlines serve` that allows to serve Outlines functions locally;
-* Add more functions to the [functions](https://github.com/outlines-dev/functions) repository.
+* Add more functions to the [functions](https://github.com/dottxt-ai/functions) repository.
 
 ### Serve structured generation
 
diff --git a/docs/community/contribute.md b/docs/community/contribute.md
index d5568f47c..d29576b75 100644
--- a/docs/community/contribute.md
+++ b/docs/community/contribute.md
@@ -16,7 +16,7 @@ Note that the [issue tracker][issues] is only intended for actionable items. In
 
 ### Setup
 
-First, [fork the repository on GitHub](https://github.com/outlines-dev/outlines/fork) and clone the fork locally:
+First, [fork the repository on GitHub](https://github.com/dottxt-ai/outlines/fork) and clone the fork locally:
 
 ```bash
 git clone git@github.com/YourUserName/outlines.git
@@ -127,6 +127,6 @@ Then you can [open a pull request][pull-requests] on GitHub. It should prompt yo
 Do not hesitate to open a draft PR before your contribution is ready, especially if you have questions and/or need feedback. If you need help, come tell us on [Discord][discord].
 
 [discord]: https://discord.gg/R9DSu34mGd
-[discussions]: https://github.com/outlines-dev/outlines/discussions
-[issues]: https://github.com/outlines-dev/outlines/issues
-[pull-requests]: https://github.com/outlines-dev/outlines/pulls
+[discussions]: https://github.com/dottxt-ai/outlines/discussions
+[issues]: https://github.com/dottxt-ai/outlines/issues
+[pull-requests]: https://github.com/dottxt-ai/outlines/pulls
diff --git a/docs/community/feedback.md b/docs/community/feedback.md
index 942809542..033e91870 100644
--- a/docs/community/feedback.md
+++ b/docs/community/feedback.md
@@ -53,7 +53,7 @@ If Outlines has been helpful to you, let us know on [Discord][discord] or give u
 
 <blockquote class="twitter-tweet"><p lang="en" dir="ltr">This is amazing - glad to see more outp guidance modules! <br><br>Will try this out soon I&#39;m wondering how they translate from regex automatons to token boundaries<br><br>Also why Open Source will succeed. Even today I don&#39;t see any guided output functionality from the big providers. <a href="https://t.co/Ity2H25Klf">https://t.co/Ity2H25Klf</a></p>&mdash; Hrishi (@hrishioa) <a href="https://twitter.com/hrishioa/status/1691181499671080960?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
 
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Outlines 〰️- a library to help LLM developers guide text generation in a fast and reliable way.<br><br>&quot;Provides generation methods that guarantee that the output will match a regular expressions, or follow a JSON schema.&quot;<br><br>Need to check this out. Reliable JSON output is a common use… <a href="https://t.co/Bkbh8vKogN">pic.twitter.com/Bkbh8vKogN</a></p>&mdash; elvis (@omarsar0) <a href="https://twitter.com/omarsar0/status/1691179888214966273?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Outlines - a library to help LLM developers guide text generation in a fast and reliable way.<br><br>&quot;Provides generation methods that guarantee that the output will match a regular expressions, or follow a JSON schema.&quot;<br><br>Need to check this out. Reliable JSON output is a common use… <a href="https://t.co/Bkbh8vKogN">pic.twitter.com/Bkbh8vKogN</a></p>&mdash; elvis (@omarsar0) <a href="https://twitter.com/omarsar0/status/1691179888214966273?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
 
 <blockquote class="twitter-tweet"><p lang="en" dir="ltr">Woah this is cool! Makes open source models more usable.<br><br>Give any LLM Function Call capability (and more) with Outlines: <a href="https://t.co/PtPykR5ZGR">https://t.co/PtPykR5ZGR</a> <a href="https://t.co/RRQjWHnIxv">https://t.co/RRQjWHnIxv</a> <a href="https://t.co/BwNnH8SMwv">pic.twitter.com/BwNnH8SMwv</a></p>&mdash; Yohei (@yoheinakajima) <a href="https://twitter.com/yoheinakajima/status/1691231912466223104?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
 
@@ -75,7 +75,7 @@ We highly value the insights of our users, and we would love to hear from you. I
 - What challenges are you facing?
 - What do you think could be improved?
 
-To schedule an appointment follow [this link](https://cal.com/dottxt/outlines). This is exclusively intended to share your experience, please go on [Discord][discord] or [GitHub](https://github.com/outlines-dev/outlines/discussions) for support.
+To schedule an appointment follow [this link](https://cal.com/dottxt/outlines). This is exclusively intended to share your experience, please go on [Discord][discord] or [GitHub](https://github.com/dottxt-ai/outlines/discussions) for support.
 
 [discord]: https://discord.gg/UppQmhEpe8
 [twitter]: https://twitter.com/dottxtai
diff --git a/docs/community/versioning.md b/docs/community/versioning.md
index d64a56e7f..023b92537 100644
--- a/docs/community/versioning.md
+++ b/docs/community/versioning.md
@@ -15,7 +15,7 @@ Each part of the version number (`major.minor.patch`) conveys information about
 
 ## Releases
 
-Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/outlines-dev/outlines/releases).
+Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/dottxt-ai/outlines/releases).
 
 ## Version Pinning Recommendations
 
diff --git a/docs/cookbook/chain_of_density.md b/docs/cookbook/chain_of_density.md
index 16c2838f2..2a6b4eb39 100644
--- a/docs/cookbook/chain_of_density.md
+++ b/docs/cookbook/chain_of_density.md
@@ -122,4 +122,4 @@ print(result.model_dump())
 
 Not bad, considering we used a smallish model to generate the summary! Chain of Density seems to be a very effective prompting technique to generate dense summaries, even with small quantized models. Its implementation in Outlines is also very short.
 
-Note that this is the first article I tried and it worked out of the box. Try it out on other articles, and please share the results on Twitter, or by opening [a new discussion](https://github.com/outlines-dev/outlines/discussions/categories/show-and-tell) on the Outlines repository!
+Note that this is the first article I tried and it worked out of the box. Try it out on other articles, and please share the results on Twitter, or by opening [a new discussion](https://github.com/dottxt-ai/outlines/discussions/categories/show-and-tell) on the Outlines repository!
diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
index bd76f40b7..b814ae048 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -3,7 +3,7 @@
 
 Chain of thought is a prompting technique introduced in the paper ["Chain-of-Thought Prompting Elicits Reasoning in Large Language Models"](https://arxiv.org/abs/2201.11903) where throught prompting the authors generate a series of intermediate reasoning steps which improves the ability of LLMs to perform complex reasoning.
 
-In this guide, we use [outlines](https://outlines-dev.github.io/outlines/) to apply chain of thought through structured output.
+In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to apply chain of thought through structured output.
 
 We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
diff --git a/docs/cookbook/deploy-using-bentoml.md b/docs/cookbook/deploy-using-bentoml.md
index 6bee77441..2926df0f6 100644
--- a/docs/cookbook/deploy-using-bentoml.md
+++ b/docs/cookbook/deploy-using-bentoml.md
@@ -2,7 +2,7 @@
 
 [BentoML](https://github.com/bentoml/BentoML) is an open-source model serving library for building performant and scalable AI applications with Python. It comes with tools that you need for serving optimization, model packaging, and production deployment.
 
-In this guide, we will show you how to use BentoML to run programs written with Outlines on GPU locally and in [BentoCloud](https://www.bentoml.com/), an AI Inference Platform for enterprise AI teams. The example source code in this guide is also available in the [examples/bentoml/](https://github.com/outlines-dev/outlines/blob/main/examples/bentoml/) directory.
+In this guide, we will show you how to use BentoML to run programs written with Outlines on GPU locally and in [BentoCloud](https://www.bentoml.com/), an AI Inference Platform for enterprise AI teams. The example source code in this guide is also available in the [examples/bentoml/](https://github.com/dottxt-ai/outlines/blob/main/examples/bentoml/) directory.
 
 ## Import a model
 
@@ -56,7 +56,7 @@ mistralai--mistral-7b-v0.1:m7lmf5ac2cmubnnz          13.49 GiB   2024-04-25 06:5
 
 As the model is ready, we can define a [BentoML Service](https://docs.bentoml.com/en/latest/guides/services.html) to wrap the capabilities of the model.
 
-We will run the JSON-structured generation example [in the README](https://github.com/outlines-dev/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
+We will run the JSON-structured generation example [in the README](https://github.com/dottxt-ai/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
 
 ```python
 DEFAULT_SCHEMA = """{
@@ -153,7 +153,7 @@ We then need to define an HTTP endpoint using `@bentoml.api` to decorate the met
 
 Here `@bentoml.api` decorator defines `generate` as an HTTP endpoint that accepts a JSON request body with two fields: `prompt` and `json_schema` (optional, which allows HTTP clients to provide their own JSON schema). The type hints in the function signature will be used to validate incoming JSON requests. You can define as many HTTP endpoints as you want by using `@bentoml.api` to decorate other methods of `Outlines` class.
 
-Now you can save the above code to `service.py` (or use [this implementation](https://github.com/outlines-dev/outlines/blob/main/examples/bentoml/)), and run the code using the BentoML CLI.
+Now you can save the above code to `service.py` (or use [this implementation](https://github.com/dottxt-ai/outlines/blob/main/examples/bentoml/)), and run the code using the BentoML CLI.
 
 ## Run locally for testing and debugging
 
diff --git a/docs/cookbook/deploy-using-modal.md b/docs/cookbook/deploy-using-modal.md
index 998e5d835..15e200cb5 100644
--- a/docs/cookbook/deploy-using-modal.md
+++ b/docs/cookbook/deploy-using-modal.md
@@ -76,7 +76,7 @@ outlines_image = outlines_image.run_function(import_model)
 
 ## Define a schema
 
-We will run the JSON-structured generation example [in the README](https://github.com/outlines-dev/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
+We will run the JSON-structured generation example [in the README](https://github.com/dottxt-ai/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
 
 ```python
 # Specify a schema for the character description. In this case,
@@ -173,7 +173,7 @@ def main(
     generate.remote(prompt)
 ```
 
-Here `@app.local_entrypoint()` decorator defines `main` as the function to start from locally when using the Modal CLI. You can save above code to `example.py` (or use [this implementation](https://github.com/outlines-dev/outlines/blob/main/examples/modal_example.py)). Let's now see how to run the code on the cloud using the Modal CLI.
+Here `@app.local_entrypoint()` decorator defines `main` as the function to start from locally when using the Modal CLI. You can save above code to `example.py` (or use [this implementation](https://github.com/dottxt-ai/outlines/blob/main/examples/modal_example.py)). Let's now see how to run the code on the cloud using the Modal CLI.
 
 ## Run on the cloud
 
diff --git a/docs/cookbook/knowledge_graph_extraction.md b/docs/cookbook/knowledge_graph_extraction.md
index e25166bca..6a6877756 100644
--- a/docs/cookbook/knowledge_graph_extraction.md
+++ b/docs/cookbook/knowledge_graph_extraction.md
@@ -1,6 +1,6 @@
 # Knowledge Graph Extraction
 
-In this guide, we use [outlines](https://outlines-dev.github.io/outlines/) to extract a knowledge graph from unstructured text.
+In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to extract a knowledge graph from unstructured text.
 
 We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
diff --git a/docs/cookbook/react_agent.md b/docs/cookbook/react_agent.md
index ca4829d5f..0597eab07 100644
--- a/docs/cookbook/react_agent.md
+++ b/docs/cookbook/react_agent.md
@@ -1,6 +1,6 @@
 # ReAct Agent
 
-This example shows how to use [outlines](https://outlines-dev.github.io/outlines/) to build your own agent with open weights local models and structured outputs. It is inspired by the blog post [A simple Python implementation of the ReAct pattern for LLMs](https://til.simonwillison.net/llms/python-react-pattern) by [Simon Willison](https://simonwillison.net/).
+This example shows how to use [outlines](https://dottxt-ai.github.io/outlines/) to build your own agent with open weights local models and structured outputs. It is inspired by the blog post [A simple Python implementation of the ReAct pattern for LLMs](https://til.simonwillison.net/llms/python-react-pattern) by [Simon Willison](https://simonwillison.net/).
 
 The ReAct pattern (for Reason+Act) is described in the paper [ReAct: Synergizing Reasoning and Acting in Language Models](https://arxiv.org/abs/2210.03629). It's a pattern where you implement additional actions that an LLM can take - searching Wikipedia or running calculations for example - and then teach it how to request the execution of those actions, and then feed their results back into the LLM.
 
diff --git a/docs/cookbook/simtom.md b/docs/cookbook/simtom.md
index aa96005b4..f730d029a 100644
--- a/docs/cookbook/simtom.md
+++ b/docs/cookbook/simtom.md
@@ -17,9 +17,9 @@ SimToM calls an LLM with two consecutive prompts:
 
 To implement SimToM with Outlines, we will need to:
 
-1. Write the prompts with [prompt functions](https://outlines-dev.github.io/outlines/reference/prompting/).
+1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/reference/prompting/).
 2. Define the JSON object each prompt will return using Pydantic.
-3. Generate responses with a Mistral model using the [transformers integration](https://outlines-dev.github.io/outlines/reference/models/transformers/).
+3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/reference/models/transformers/).
 
 Let's dive into it!
 
diff --git a/docs/index.md b/docs/index.md
index e7409c0d3..3692d32c9 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,26 +1,8 @@
 ---
 title: Outlines
+template: home.html # Note that this is managed in overrides/home.html
 hide:
   - navigation
   - toc
   - feedback
 ---
-
-#
-
-<figure markdown>
-  ![Image title](assets/images/logo.png){ width="600" }
-</figure>
-
-<center>
-    <h1 class="title">Generate text with LLMs</h1>
-    <h2 class="subtitle">Robust prompting & (structured) text generation</h2>
-    [:fontawesome-solid-bolt: Get started](welcome.md){ .md-button .md-button--primary }
-    [:fontawesome-brands-discord: Join the Community](https://discord.gg/ZxBxyWmW5n){ .md-button }
-
-<div class="index-pre-code">
-```bash
-pip install outlines
-```
-</div>
-</center>
diff --git a/docs/installation.md b/docs/installation.md
index 1017b627e..bf2da86f9 100644
--- a/docs/installation.md
+++ b/docs/installation.md
@@ -28,7 +28,7 @@ If you encounter any problem using Outlines with these libraries, take a look at
 You can install the latest version of Outlines on the repository's `main` branch:
 
 ```python
-pip install git+https://github.com/outlines-dev/outlines.git@main
+pip install git+https://github.com/dottxt-ai/outlines.git@main
 ```
 
 This can be useful, for instance, when a fix has been merged but not yet released.
diff --git a/docs/overrides/home.html b/docs/overrides/home.html
new file mode 100644
index 000000000..1114895e6
--- /dev/null
+++ b/docs/overrides/home.html
@@ -0,0 +1,120 @@
+{#-
+This file overrides the home page to use HTML tooling
+better.
+-#}
+{% extends "main.html" %}
+{% block tabs %}
+{{ super() }}
+
+<style>
+    .md-header {
+        position: initial
+    }
+
+    .md-main__inner {
+        margin: 0
+    }
+
+    .md-content {
+        display: none
+    }
+
+    @media screen and (min-width:60em) {
+        .md-sidebar--secondary {
+            display: none
+        }
+    }
+
+    @media screen and (min-width:76.25em) {
+        .md-sidebar--primary {
+            display: none
+        }
+    }
+
+    .mdx-container {
+        display: flex;
+        justify-content: center;
+        align-items: center;
+        min-height: calc(100vh - 300px);
+        padding-left: 1rem;
+        padding-right: 1rem;
+    }
+
+    .mdx-hero {
+        text-align: center;
+    }
+
+    .mdx-hero__image {
+        margin-bottom: 1rem;
+        margin-top: 1rem;
+    }
+
+    .mdx-hero__content {
+        max-width: 600px;
+        margin: 0 auto;
+    }
+
+
+    .md-buttons {
+        display: flex;
+        justify-content: center;
+        gap: 1rem;
+        margin-top: 2rem;
+        flex-direction: row; /* Default to row layout */
+    }
+
+    .md-team {
+        margin-top: 2rem;
+        margin-bottom: 1rem;
+    }
+
+    /* New media query for smaller screens */
+    @media screen and (max-width: 40em) {
+        .md-buttons {
+            flex-direction: column; /* Stack buttons vertically */
+            align-items: center; /* Center-align the stacked buttons */
+        }
+
+        .md-buttons .md-button {
+            width: 100%; /* Make buttons full-width */
+            max-width: 200px; /* Limit maximum width for better appearance */
+        }
+    }
+</style>
+
+<section class="mdx-container">
+    <div class="md-grid md-typeset">
+        <div class="mdx-hero">
+            <div class="mdx-hero__image">
+                <img src="assets/images/logo.png" alt="Outlines Logo" width="600" draggable="false">
+            </div>
+            <div class="mdx-hero__content">
+                <h2 class="subtitle" style="font-weight: 400; padding-top: 1rem;">
+                    Structured text generation and robust prompting for language models
+                </h2>
+
+                    <!-- <div class="index-pre-code">
+                        <pre><code>pip install outlines</code></pre>
+                    </div> -->
+
+                    <div class="md-buttons">
+                        <a href="{{ 'welcome/' | url }}" title="Get started" class="md-button md-button--primary">
+                            Get started
+                        </a>
+                        <a href="https://discord.gg/ZxBxyWmW5n" title="Join the Community" class="md-button">
+                            Join the Community
+                        </a>
+                        <a href="https://twitter.com/dottxtai" title="Follow us on X" class="md-button">
+                            Follow us on X
+                        </a>
+                    </div>
+
+
+                <p class="md-team">Made with ❤️ by the team at <a href="https://dottxt.co">.txt</a></p>
+            </div>
+        </div>
+    </div>
+</section>
+{% endblock %}
+{% block content %}{% endblock %}
+{% block footer %}{% endblock %}
\ No newline at end of file
diff --git a/docs/overrides/index.html b/docs/overrides/index.html
deleted file mode 100644
index 74a4987f4..000000000
--- a/docs/overrides/index.html
+++ /dev/null
@@ -1,11 +0,0 @@
-{% extends "base.html" %}
-
-{% block announce %}
-    For updates follow <strong>@remilouf</strong> on
-    <a href="https://twitter.com/squidfunk">
-        <span class="twemoji twitter">
-        {% include ".icons/fontawesome/brands/twitter.svg" %}
-        </span>
-        <strong>Twitter</strong>
-    </a>
-{% endblock %}
diff --git a/docs/overrides/main.html b/docs/overrides/main.html
index b4183d71a..5cb6467e6 100644
--- a/docs/overrides/main.html
+++ b/docs/overrides/main.html
@@ -1,22 +1,2 @@
 {% extends "base.html" %}
 
-{% block announce %}
-    For updates follow <strong>@dottxtai</strong> on
-    <a href="https://twitter.com/dottxtai">
-        <span class="twemoji twitter">
-        {% include ".icons/fontawesome/brands/twitter.svg" %}
-        </span>
-        <strong>Twitter</strong>
-    </a>
-    and
-    <span class="twemoji stat">
-    {% include ".icons/fontawesome/solid/star.svg" %}
-    </span>
-    the repo on
-    <a href="https://github.com/outlines-dev/outlines">
-        <span class="twemoji github">
-        {% include ".icons/fontawesome/brands/github.svg" %}
-        </span>
-        <strong>Github</strong>
-    </a>
-{% endblock %}
diff --git a/docs/quickstart.md b/docs/quickstart.md
index 2e1f9a6bb..81a067ad6 100644
--- a/docs/quickstart.md
+++ b/docs/quickstart.md
@@ -216,7 +216,7 @@ Once you are done experimenting with a prompt and an output structure, it is use
 
 ## Going further
 
-If you need more inspiration you can take a look at the [cookbook](cookbook/index.md) or watch [Remi Louf's AI Engineer World’s Fair Presentation on Outlines](https://www.youtube.com/live/R0X7mPagRiE?t=775s). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/outlines-dev/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
+If you need more inspiration you can take a look at the [cookbook](cookbook/index.md) or watch [Remi Louf's AI Engineer World’s Fair Presentation on Outlines](https://www.youtube.com/live/R0X7mPagRiE?t=775s). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/dottxt-ai/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
 
 
 [pydantic]: https://docs.pydantic.dev/latest
diff --git a/docs/reference/generation/cfg.md b/docs/reference/generation/cfg.md
index e3b177800..f0104ec63 100644
--- a/docs/reference/generation/cfg.md
+++ b/docs/reference/generation/cfg.md
@@ -34,7 +34,7 @@ print(sequence)
 
 !!! Note "Experimental"
 
-    Outlines current **community-contributed** implementation of CFG-structured generation is experimental. This does not reflect the performance of [.txt](https://dottxt.co)'s product, where we have optimized grammar-structured generation to be as fast as regex-structured generation. Additionally, it does not fully align with the approach described in our [technical report](https://arxiv.org/pdf/2307.09702), aside from its use of incremental/partial parsing. This feature is still a work in progress, requiring performance enhancements and bug fixes for an ideal implementation. For more details, please see our [grammar-related open issues on GitHub](https://github.com/outlines-dev/outlines/issues?q=is%3Aissue+is%3Aopen+label%3Agrammar).
+    Outlines current **community-contributed** implementation of CFG-structured generation is experimental. This does not reflect the performance of [.txt](https://dottxt.co)'s product, where we have optimized grammar-structured generation to be as fast as regex-structured generation. Additionally, it does not fully align with the approach described in our [technical report](https://arxiv.org/pdf/2307.09702), aside from its use of incremental/partial parsing. This feature is still a work in progress, requiring performance enhancements and bug fixes for an ideal implementation. For more details, please see our [grammar-related open issues on GitHub](https://github.com/dottxt-ai/outlines/issues?q=is%3Aissue+is%3Aopen+label%3Agrammar).
 
 !!! Note "Greedy"
 
@@ -65,7 +65,7 @@ The following grammars are currently available:
 - Arithmetic grammar via `outlines.grammars.arithmetic`
 - JSON grammar via `outlines.grammars.json`
 
-If you would like more grammars to be added to the repository, please open an [issue](https://github.com/outlines-dev/outlines/issues) or a [pull request](https://github.com/outlines-dev/outlines/pulls).
+If you would like more grammars to be added to the repository, please open an [issue](https://github.com/dottxt-ai/outlines/issues) or a [pull request](https://github.com/dottxt-ai/outlines/pulls).
 
 
 ## Grammar guide
diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
index 0c090f8a7..a14818514 100644
--- a/docs/reference/generation/generation.md
+++ b/docs/reference/generation/generation.md
@@ -174,7 +174,7 @@ print(result)
 # 5+5+5+5+5
 ```
 
-The available grammars are listed [here](https://github.com/outlines-dev/outlines/tree/main/outlines/grammars).
+The available grammars are listed [here](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
 
 
 ### [Regex-structured generation](./regex.md)
diff --git a/docs/reference/generation/types.md b/docs/reference/generation/types.md
index 5b83a5916..eb6d7382b 100644
--- a/docs/reference/generation/types.md
+++ b/docs/reference/generation/types.md
@@ -73,7 +73,7 @@ print(result)
 ```
 
 
-We plan on adding many more custom types. If you have found yourself writing regular expressions to generate fields of a given type, or if you could benefit from more specific types don't hesite to [submit a PR](https://github.com/outlines-dev/outlines/pulls) or [open an issue](https://github.com/outlines-dev/outlines/issues/new/choose).
+We plan on adding many more custom types. If you have found yourself writing regular expressions to generate fields of a given type, or if you could benefit from more specific types don't hesite to [submit a PR](https://github.com/dottxt-ai/outlines/pulls) or [open an issue](https://github.com/dottxt-ai/outlines/issues/new/choose).
 
 
 [wiki-isbn]: https://en.wikipedia.org/wiki/ISBN#Check_digits
diff --git a/docs/reference/serve/vllm.md b/docs/reference/serve/vllm.md
index 8e2886c96..14277a526 100644
--- a/docs/reference/serve/vllm.md
+++ b/docs/reference/serve/vllm.md
@@ -65,7 +65,7 @@ curl http://127.0.0.1:8000/generate \
 
 Instead of `curl`, you can also use the [requests][requests]{:target="_blank"} library from another python program.
 
-Please consult the [vLLM documentation][vllm]{:target="_blank"} for details on additional request parameters. You can also [read the code](https://github.com/outlines-dev/outlines/blob/main/outlines/serve/serve.py) in case you need to customize the solution to your needs.
+Please consult the [vLLM documentation][vllm]{:target="_blank"} for details on additional request parameters. You can also [read the code](https://github.com/dottxt-ai/outlines/blob/main/outlines/serve/serve.py) in case you need to customize the solution to your needs.
 
 [requests]: https://requests.readthedocs.io/en/latest/
 [vllm]: https://docs.vllm.ai/en/latest/index.html
diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index c4539ab80..7c5b5e808 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -9,6 +9,7 @@
   --md-code-fg-color: #FFFFFF;
   --md-text-font-family: "Inter";
   --md-code-font: "Source Code Pro Custom";
+  --md-typeset-a-color: #d53135; /*this is the brand color*/
 
   /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
   --md-mermaid-label-fg-color: #000000;
diff --git a/docs/welcome.md b/docs/welcome.md
index a7800f7ad..728a30a97 100644
--- a/docs/welcome.md
+++ b/docs/welcome.md
@@ -2,11 +2,11 @@
 title: Welcome to Outlines!
 ---
 
-Outlines〰 is a Python library that allows you to use Large Language Model in a simple and robust way (with structured generation). It is built by [.txt][.txt]{:target="_blank"}, and is already used in production by many companies.
+Outlines is a Python library that allows you to use Large Language Model in a simple and robust way (with structured generation). It is built by [.txt][.txt]{:target="_blank"}, and is already used in production by many companies.
 
 ## What models do you support?
 
-We support [Openai](reference/models/openai.md), but the true power of Outlines〰 is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [exllama2](reference/models/exllamav2.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
+We support [Openai](reference/models/openai.md), but the true power of Outlines is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [exllama2](reference/models/exllamav2.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
 
 ## What are the main features?
 
@@ -48,7 +48,7 @@ We support [Openai](reference/models/openai.md), but the true power of Outlines
 ## Why use Outlines?
 
 
-Outlines〰 is built at [.txt][.txt] by engineers with decades of experience in software engineering, machine learning (Bayesian Statistics and NLP), and compilers. [.txt][.txt] is a VC-backed company fully focused on the topic of structured generation and is committed to make the community benefit from its experience.
+Outlines is built at [.txt][.txt] by engineers with decades of experience in software engineering, machine learning (Bayesian Statistics and NLP), and compilers. [.txt][.txt] is a VC-backed company fully focused on the topic of structured generation and is committed to make the community benefit from its experience.
 
 We are also open source veterans and have authored/maintained many libraries over the years: the [Aesara][aesara]{:target="_blank"} and [Pythological][pythological]{:target="_blank"} ecosystems, [Blackjax][blackjax]{:target="_blank"} and [Hy][hy]{:target="_blank"} among many others.
 .
@@ -119,33 +119,33 @@ Still not convinced, read [what people say about us](community/feedback.md). And
 
 ## Philosophy
 
-**Outlines** 〰 is a library for neural text generation. You can think of it as a
+**Outlines**  is a library for neural text generation. You can think of it as a
 more flexible replacement for the `generate` method in the
 [transformers](https://github.com/huggingface/transformers) library.
 
-**Outlines** 〰 helps developers *structure text generation* to build robust
+**Outlines**  helps developers *structure text generation* to build robust
 interfaces with external systems. It provides generation methods that
 guarantee that the output will match a regular expressions, or follow
 a JSON schema.
 
-**Outlines** 〰 provides *robust prompting primitives* that separate the prompting
+**Outlines**  provides *robust prompting primitives* that separate the prompting
 from the execution logic and lead to simple implementations of few-shot
 generations, ReAct, meta-prompting, agents, etc.
 
-**Outlines** 〰 is designed as a *library* that is meant to be compatible the
+**Outlines**  is designed as a *library* that is meant to be compatible the
 broader ecosystem, not to replace it. We use as few abstractions as possible,
 and generation can be interleaved with control flow, conditionals, custom Python
 functions and calls to other libraries.
 
-**Outlines** 〰 is *compatible with every auto-regressive model*. It only interfaces with models
+**Outlines**  is *compatible with every auto-regressive model*. It only interfaces with models
 via the next-token logits distribution.
 
 ## Outlines people
 
 Outlines would not be what it is today without a community of dedicated developers:
 
-<a href="https://github.com/outlines-dev/outlines/graphs/contributors">
-  <img src="https://contrib.rocks/image?repo=outlines-dev/outlines" />
+<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
 </a>
 
 ## Acknowledgements
diff --git a/environment.yml b/environment.yml
index c267f86a0..6fc980fed 100644
--- a/environment.yml
+++ b/environment.yml
@@ -1,9 +1,9 @@
 # To use:
 #
 #   $ conda env create -f environment.yml  # `mamba` works too for this command
-#   $ conda activate outlines-dev
+#   $ conda activate dottxt-ai
 #
-name: outlines-dev
+name: dottxt-ai
 channels:
   - conda-forge
   - huggingface
diff --git a/examples/dating_profile.py b/examples/dating_profile.py
index acc00126e..504ec943d 100644
--- a/examples/dating_profile.py
+++ b/examples/dating_profile.py
@@ -26,7 +26,7 @@ class QuestionAnswer:
 
 class DatingProfile(BaseModel):
     # It is possible put length constraints on these strings using constr- however, this appears to dramatically increase the generation time
-    # This may be resolved in the future with this PR: https://github.com/outlines-dev/outlines/pull/272
+    # This may be resolved in the future with this PR: https://github.com/dottxt-ai/outlines/pull/272
     bio: str
     job: str
     # Ignore mypy checks here because it still doesn't support conlist or constr: https://github.com/pydantic/pydantic/issues/975
diff --git a/mkdocs.yml b/mkdocs.yml
index c95905914..4189df1c0 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -1,5 +1,5 @@
 # Site information
-site_name: Outlines 〰️
+site_name: Outlines
 site_author: The Outlines developers
 site_description: >-
   Structured text generation with LLMs
@@ -19,7 +19,7 @@ theme:
   palette:
     - scheme: default
       primary: white
-  logo: assets/images/logo-simple.png
+  logo: assets/images/logo-square.svg
   favicon: assets/images/logo-simple.png
   icon:
     repo: fontawesome/brands/github
diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index 44a918494..b7b121fe6 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -322,7 +322,7 @@ def __init__(self, cfg_string: str, tokenizer):
         """
         warnings.warn(
             "Outlines' public *community-contributed* CFG structured generation is experimental. "
-            "Please review https://outlines-dev.github.io/outlines/reference/cfg#disclaimer"
+            "Please review https://dottxt-ai.github.io/outlines/reference/cfg#disclaimer"
         )
 
         self.cfg_string = cfg_string
@@ -466,7 +466,7 @@ def _get_parser_state_token_applied(
 
     def is_final_state(self, state: CFGState) -> bool:
         # TODO: remove this method, use can_terminate_state and must_terminate_state
-        # here and in RegexGuide per https://github.com/outlines-dev/outlines/issues/885
+        # here and in RegexGuide per https://github.com/dottxt-ai/outlines/issues/885
         return self.can_terminate_state(state)
 
     def can_terminate_state(self, state: CFGState) -> bool:
diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
index 4f372f209..034a65ae5 100644
--- a/outlines/generate/cfg.py
+++ b/outlines/generate/cfg.py
@@ -44,7 +44,7 @@ def cfg_vision(model, cfg_str: str, sampler: Sampler = multinomial()):
 @cfg.register(ExLlamaV2Model)
 def cfg_exllamav2(model, cfg_str: str, sampler: Sampler = multinomial()):
     raise NotImplementedError(
-        "Not yet available, track progress in https://github.com/outlines-dev/outlines/pull/1010"
+        "Not yet available, track progress in https://github.com/dottxt-ai/outlines/pull/1010"
     )
 
 
diff --git a/pyproject.toml b/pyproject.toml
index 99d4f94e1..dfda81acd 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -73,9 +73,9 @@ serve = [
 ]
 
 [project.urls]
-homepage = "https://github.com/outlines-dev/outlines"
-documentation = "https://outlines-dev.github.io/outlines/"
-repository = "https://github.com/outlines-dev/outlines"
+homepage = "https://github.com/dottxt-ai/outlines"
+documentation = "https://dottxt-ai.github.io/outlines/"
+repository = "https://github.com/dottxt-ai/outlines"
 
 [project.readme]
 file="README.md"
diff --git a/tests/fsm/test_json_schema.py b/tests/fsm/test_json_schema.py
index 21571da8d..7565ff642 100644
--- a/tests/fsm/test_json_schema.py
+++ b/tests/fsm/test_json_schema.py
@@ -1018,7 +1018,7 @@ class MockModel(BaseModel):
 
 
 def test_one_of_doesnt_produce_illegal_lookaround():
-    """Reproduces failure in https://github.com/outlines-dev/outlines/issues/823"""
+    """Reproduces failure in https://github.com/dottxt-ai/outlines/issues/823"""
 
     class Cat(BaseModel):
         pet_type: Literal["cat"]
diff --git a/tests/fsm/test_regex.py b/tests/fsm/test_regex.py
index 824588b22..7418deca2 100644
--- a/tests/fsm/test_regex.py
+++ b/tests/fsm/test_regex.py
@@ -667,7 +667,7 @@ def convert_token_to_string(self, token):
 def test_numba_leading_null_byte_UnicodeCharSeq_remains_broken():
     """Assert numba UnicodeCharSeq w/ leading \x00 is still broken"""
     # EXPLANATION:
-    # https://github.com/outlines-dev/outlines/pull/930#issuecomment-2143535968
+    # https://github.com/dottxt-ai/outlines/pull/930#issuecomment-2143535968
 
     # from https://github.com/numba/numba/issues/9542
     d = numba.typed.typeddict.Dict.empty(numba.types.UnicodeCharSeq(1), numba.int64)
@@ -685,7 +685,7 @@ def test_numba_leading_null_byte_UnicodeCharSeq_remains_broken():
 def test_numba_leading_null_byte_unicode_type_sane(input_key):
     """Assert numba unicode_type w/ leading \x00 is working"""
     # EXPLANATION:
-    # https://github.com/outlines-dev/outlines/pull/930#issuecomment-2143535968
+    # https://github.com/dottxt-ai/outlines/pull/930#issuecomment-2143535968
 
     # from https://github.com/numba/numba/issues/9542
     d = numba.typed.typeddict.Dict.empty(numba.types.unicode_type, numba.int64)
@@ -712,8 +712,8 @@ def test_reduced_vocabulary_with_rare_tokens(rare_token):
 
     See [1] and [2] for context.
 
-    [1]: https://github.com/outlines-dev/outlines/pull/763
-    [2]: https://github.com/outlines-dev/outlines/pull/948
+    [1]: https://github.com/dottxt-ai/outlines/pull/763
+    [2]: https://github.com/dottxt-ai/outlines/pull/948
     """
     tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2")
     tokenizer = TransformerTokenizer(tokenizer=tokenizer)
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index fc4166535..ff247b0f4 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -76,7 +76,7 @@ def model_vllm(tmp_path_factory):
     return models.vllm("facebook/opt-125m", gpu_memory_utilization=0.1)
 
 
-# TODO: exllamav2 failing in main, address in https://github.com/outlines-dev/outlines/issues/808
+# TODO: exllamav2 failing in main, address in https://github.com/dottxt-ai/outlines/issues/808
 # TODO: t5 tokenizer doesn't work with streaming
 """
 @pytest.fixture(scope="session")
@@ -235,7 +235,7 @@ def test_generate_fsm(request, model_fixture, pattern):
 
 
 @pytest.mark.skip(
-    "Fix issues with JSON, some models fail this test https://github.com/outlines-dev/outlines/issues/985"
+    "Fix issues with JSON, some models fail this test https://github.com/dottxt-ai/outlines/issues/985"
 )
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
 def test_generate_json(request, model_fixture, sample_schema):
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
index 0a98f0226..08521c672 100644
--- a/tests/generate/test_integration_llamacpp.py
+++ b/tests/generate/test_integration_llamacpp.py
@@ -250,7 +250,7 @@ def test_llamacpp_json_schema(model):
     ],
 )
 def test_byte_tokenizer_regression(repo, model_path, hf_tokenizer_uri):
-    """Reproduce https://github.com/outlines-dev/outlines/issues/820"""
+    """Reproduce https://github.com/dottxt-ai/outlines/issues/820"""
     import llama_cpp
 
     model = models.llamacpp(
diff --git a/tests/test_function.py b/tests/test_function.py
index 24e132d42..62f7ea29f 100644
--- a/tests/test_function.py
+++ b/tests/test_function.py
@@ -28,29 +28,29 @@ def test_download_from_github_invalid():
         download_from_github("outlines/program")
 
     with pytest.raises(ValueError, match="Do not append"):
-        download_from_github("outlines-dev/outlines/program.py")
+        download_from_github("dottxt-ai/outlines/program.py")
 
 
 @responses.activate
 def test_download_from_github_success():
     responses.add(
         responses.GET,
-        "https://raw.githubusercontent.com/outlines-dev/outlines/main/program.py",
+        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
         body="import outlines\n",
         status=200,
     )
 
-    file = download_from_github("outlines-dev/outlines/program")
+    file = download_from_github("dottxt-ai/outlines/program")
     assert file == "import outlines\n"
 
     responses.add(
         responses.GET,
-        "https://raw.githubusercontent.com/outlines-dev/outlines/main/foo/bar/program.py",
+        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/foo/bar/program.py",
         body="import outlines\n",
         status=200,
     )
 
-    file = download_from_github("outlines-dev/outlines/foo/bar/program")
+    file = download_from_github("dottxt-ai/outlines/foo/bar/program")
     assert file == "import outlines\n"
 
 

From cfde8b7faa02df89fbbe30c0f0ef8cee9824799f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Mon, 16 Sep 2024 18:31:52 +0200
Subject: [PATCH 071/505] Change the color of the logo (#1155)

---
 docs/assets/images/logo-square.png | Bin 0 -> 2233 bytes
 docs/assets/images/logo-square.svg |  45 +++++++++++++-------------
 docs/assets/images/logo.png        | Bin 6354 -> 6386 bytes
 docs/assets/images/logo.svg        |  49 +++++++++++++++--------------
 docs/overrides/home.html           |   2 +-
 docs/overrides/main.html           |   1 -
 6 files changed, 49 insertions(+), 48 deletions(-)
 create mode 100644 docs/assets/images/logo-square.png

diff --git a/docs/assets/images/logo-square.png b/docs/assets/images/logo-square.png
new file mode 100644
index 0000000000000000000000000000000000000000..0592b8fa9a2d43e08f7367e187514beabc685e2f
GIT binary patch
literal 2233
zcmcgue>9VO7=PXM8_DUO${Zw>>J%v?8>;KK?o<p-k(Mas$8e|ld9J&pj4P!(W4j2M
zm<WY6+G)EYn%`%b7Ny24v5n2RukJbBQqmvy+&{kOJ?DMi_kBLk=lML(=S@6<uv1+H
zT?GJus)N1tF#u3f1<!#M%fLJN?c1xt%gSK;<DmesJ!S4Fwd<OIPXHQWHXdOvL1)9T
zJ{VsBi^Uq83k(QF`vm(M1Y!J0qi`qytgv;kKIj^eDeP-WAtN=|d^udCvcoYr-k+#z
zaJy~CQPlBlPS#4N_0Xp`CiRZG7i%iko@ltrT4R*4!Q9p30n0nYN3d$!`L*VG&^8U|
zMx)f<wwYxOJ$SAfk}+{DT3(AEMzv&x8fG~6mz!NeiRip%m0ew4G^*5@zWlJy1~yZU
zphs-YAa?DkgQfWc>KtHGj*_)C<jR$0Yt&UYZh{`LHjcaUPr%8=FnAYzqOM+1Q(nD=
zNJM676!k?_#P(Ae1yoD*5{cAHXvkM->TBK|HTY6i|Ax(k?SPtS#=>2rO*I~#^P5WV
zA)Cu(9Wx4vj4h&aEp&t(a+*7(d-|9$q^n*c;}WT3<TEVslCP!z6jxbkkh6i-{s?a!
zR!tf)WQ1|CnErcHtwctGRrwax*Ki-(qB99<t~4~~`7wF#(*Y8L)aH2LAsNv9R)t`S
zlE$7iC%e_WeU)QH4NXUw6r3jGX48q7(5WZ>M&~0Hb<!fK=Mc_**w5Xx+3I@I1)Y;M
z43_L=ts5#V`cD5{a&HtyU?IVr%(({}UXG##ZQhF(#zguKOryuSiL?G4I)dUribz-Z
zmU=aZgVPf1lJ6Twv{HF@sQ9GI!z;gF{nN6+c5IG`#JtUh1k|7{Qk(8my$L40q^92c
zp;M<THG=F!LS=M7<2W4c52NMng~1xFG}5uLMk&#OOPp$8Gz6V_b5O0*K*a04%n#nv
zN>%Wi3G2y(meZqdWHTwd5Zm2oM=_UUkaW0Md)lyT?dJ-ZhXs_zU#$`uDrM_s{?tvv
zGt+|TxTQn4zEsY<_cqUtI&{blN_3_tO!^0x8EMXn>q9(Z42b87XW+Ex8Sfr7?a9Gj
zt90X$+RFEnc;$vBZ$^|fA>maY>~%$3jwclB)rS>4?re`7c_p+)U%WF$40zO_nxp2l
z-fdrVpNUwz%P~N4>nCv!*eQy|Nu*R?udxy4EkgE_+H94BGM^z_JH`s7-PaZ;d{9_x
zG36<iP0c~&pdx(*S|xd2!ZCrOb3X-RO7;)WXURD5%7^L`ZUqHAz9Y9YUS4i^7iqyL
zyLZXopa<yc_@L}+ePW1((!%34kEDO+?*9nP&GwdC$x#=??@On%3#D<}cM@Q<QCPT7
z0I24&QqBgya?$E&xT#osH$KD^b`;4j;Gqpm^gr1TU&Wb<kB)3DLR5K)vT$mD9104@
zf&88E$-Rfej5knm;H<<|Vg5T>Qt=9p)=t@s$W_%Yp9je|BBT1uX5Y1Yj`;fIa!@tp
zI4e-uyAwOtNy?MH4}BR1GaiA|qTtkC6i>H)V?Tbi{slMc2c&s=%SFS?2x)s{-H%+e
zOi<MLerGz{Yn>@{;6l`&I4#l~X%vD!^}KE_ANRuooB3nr<K0#I%+Yt{mPvjEVD-;`
zNx&<r7^R#^kun1xRSFTEUA*Q65zh~lmFhTgEvqH=V)Hp!B)aLlAp*52I0Ah#Zy^u)
z3@KfD4ap335h<m;8F*6Y{@^;Ez!I>g5{d-;LMgy|y0{N~=1(8LJ>R&%09O((XAoD5
zy`;7YD&ex^G?LZNElMRkC%djkt!&xhJ#8q9G;(<uoAnyqP?x8N)e<n@G2h(arBfY^
zPmyu*sEV12Gq(u?xgQMkdnS2j9%NutbBM)<aN;X0lM*UB)!-!Z(i3!3m6T-k>_OxE
z@|4iG`$LS~$XuBp(nhJ<9QVSdSjL~&`9dJJOvVOglH{??i#IGa7NYgUCOf3JwIBTL
O032)()+L8d{rWfXa;-}M

literal 0
HcmV?d00001

diff --git a/docs/assets/images/logo-square.svg b/docs/assets/images/logo-square.svg
index 1d1eb3a38..29b1c20f2 100644
--- a/docs/assets/images/logo-square.svg
+++ b/docs/assets/images/logo-square.svg
@@ -28,8 +28,8 @@
      showgrid="false"
      showguides="true"
      inkscape:zoom="1.0083333"
-     inkscape:cx="280.66116"
-     inkscape:cy="25.785124"
+     inkscape:cx="280.66117"
+     inkscape:cy="25.785125"
      inkscape:window-width="3440"
      inkscape:window-height="1416"
      inkscape:window-x="0"
@@ -46,78 +46,79 @@
        inkscape:locked="false" /></sodipodi:namedview><g
      id="g13"
      transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
-     style="fill:#d53135;fill-opacity:1"><g
+     style="fill:#e35a26;fill-opacity:1"><g
        id="g314"
-       transform="translate(1367.0913,152.50405)"><rect
+       transform="translate(1367.0913,152.50405)"
+       style="fill:#e35a26;fill-opacity:1"><rect
          x="28"
          y="644"
          width="28"
          height="28"
          id="rect1"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
          id="path1-7"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="28"
          y="112"
          width="28"
          height="28"
          id="rect2"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,560 v 28 h 420 v -28 z"
          id="path2-0"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,476 v 28 h 420 v -28 z"
          id="path3-4"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,392 v 28 h 420 v -28 z"
          id="path4-0"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,308 v 28 h 420 v -28 z"
          id="path5-8"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 364,168 v 28 h 140 v -28 z"
          id="path6-5"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 336,196 v -28 h -56 v 28 z"
          id="path7-1"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="M 336,112 V 84 h -56 v 28 z"
          id="path8-6"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 364,84 v 28 H 504 V 84 Z"
          id="path9-6"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 56,672 v 28 h 476 v -28 z"
          id="path10-2"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="56"
          y="84"
          width="28"
          height="28"
          id="rect10"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="112"
          y="28"
          width="28"
          height="28"
          id="rect11"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="84"
          y="56"
          width="28"
          height="28"
          id="rect12"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
          id="path12-1"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 560,28 v 616 h 28 V 28 Z"
          id="path13-9"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="532"
          y="644"
          width="28"
          height="28"
          id="rect13"
-         style="fill:#d53135;fill-opacity:1" /></g></g></svg>
+         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
diff --git a/docs/assets/images/logo.png b/docs/assets/images/logo.png
index 14d1c727e132c9b07a9d922a46eece1af4e9a394..005b7c6d2195cf0af295c15fd2e167f3f30b1fad 100644
GIT binary patch
literal 6386
zcmcgw2~^T~yQb+Iota#QoE9~8(oD-yGjheWnMlngQ!8_W%F+}G6&1~enx+YJD?@QZ
zO0(4^H`G+riL4MY_bnwcB{v8a6<PfL^v!qfnR`3k@1E)0!-4nkUw+H`d*1hbo;UI6
z5qrh&)V>3OK#GSQVD2E$Kj0wHx&z<H0avcxuB`;jx1kPR7!YVzn(Vu-XS=B)a8Whl
z(8&l7bU+05boe<C7K=3sx`4#^pAJ1|gbokP<yxzOKwEK#VLu;>&YK&Fzd;I&(RwS2
z>F^r-#o|D-N&1Oaxx)r+J0n*1MyNGscd6#xaCUfh?Zn>N@AbQ1XW#htp#0E*so%0Q
z<D4BFHI!S*tHRHuE0k<FokNK`yYh|yZN^qFgUc;L8K}z#^9BaeqdG*3X}|AUe^a74
zH!8}{D4pSn>R9B66Z4(uWlN*#J_`h?mg|+9X{V<(4V7Pf_j>V-I3xZVuoDZSX}mmO
zK^C67fW3o2>H2@q5WJBr)f4UIjnF82jSI6@P1}cB+zHCb3&|S=d~i9Y0TNb_5+0Wa
zxltwwNNFic!<^Yom*jGkMo1aofIx43A+ylt;NakeF?<xZUpS$!jLwBN1my+(#I0xR
zc2dY&InIU?lPGzmCKNEp+(5pDr&&>>e;x!ya{-6NAJGPZE@9Fh+sQU__y^-JVSbO7
zx489uKNRlk2!V_~4EO0-&E}2^da}x8qg&+;u#4<4@vw#{U=bRAky9;&2$$>s0B-Hs
z?3p5a<nKE5&ofczXKzmul|~;<9@;-n3T`@DAe=pEPj8I{KEgR^Kkw=LcRB*|3UW6r
zb~y8jJzAZ^VCR$2iTWK3`o1a0;Rf`K)=11<-Zo%O#f+REjH@Z0nH!rdfIUh2>V&!O
zUxC5k!lH>3H#vW*(KxcOPEWgaH@tvAJVa}A|Egqt<ZdOq-`HP+K&MEUUf{SuKPJll
z{cZCq^v(JgOWXm!=~HVKfjRnG8i!)I7J+ilRwhrjY#r9!FigdSw8i^38U+@bP2`s=
zn-=0mC?2j3_)4P?Ytt+qbUm;N1M?`pd%@tb0-IRIL2}b#^O$j}=%@5}wn`h(&iUQU
z-o`Z~V&>zc#e^@5kfcgTWZrsR_R>_HFnyEEGjDyz^!{Eb{WTI5I@pZy)OI*O@3?El
zk$tEbcNCr2YUvSNpOq1KsI)F18gNhtJ;518-|sq%dp{b+{E0hEZA9JAzmXBth4NKn
zAmU~;lm#TfRhKd>bJ^#xDsyv>^UpS(r{AZvYuM;rzppa9IiX5baURiTHmU*Usqq+Z
zu{q}vndyNV7Lh;us({I}P+7V#7=Hu?%z=&1H+%t;KIWAtjOtl=`AE-8FW7@4Vm|Sf
z^LS>TnK_d(*b!79VGmpjmMq{avc3=W({~DUWvK$Qo~*0f1?T45Sgf@kcl{$LmM*Vs
zyA`Y=^gg0xJujs~fJd4g>w;x*cmu}&N@m`Ll&Y-*9SB^eE~yezGw+Y)R5t~+mH1q!
zTd4FBtvjVF%NKXm43)`hZmjg`CZ%Dh&M*A0nqlFe1G{qn*Pu)G<xj92bcVP4s5udD
z974)vk;~ODy}BK}4!GwwG(V|tl2Klt8GQx@94qs*Y#kub4z~v2_rDwd*i#TFk((f3
z8pvIl=N3!mZV%$7+OeaV)mosFs({N66a;*~#h%s?IA>hgxkVdTB1naPTn8BH(vIs}
z{-4Y<DgdPB9rdC9bt{)W3e(GDEvn<J{Eb1k4$8cp#@?l6Eq?2<s<f{8>BN_W%jZCs
zewww8`EGP@#MS5eDuIeQpk-#HIB853q+VNP{=7yl=S&fu6?2kSndO`13R~9aDl=Iu
zY-%OH8j1A{9lJF{DVFFVg?&rw?bnQuI4v$tQkrxmA5PtYTst-x!P7OzP1GBVU_zoM
zr(7W3W9E<Kd`{$ycM{z58*0&_rD93<a8rV(*c(;3R3^WnY?Pn8_qbClx5!QOc0!(K
zb+a6*eP+zGo<bO#AKx3LJf!F(t9>JpQd(rkP}Q-fdP|+Q;2ax)08J~2?7^lUCI|yh
zcm1SnpG1s*udXLzcHWPLSH}B5lQ+orVDx5>HJ_lYnXm`BM%!%M7P-j1M?;F3>*D__
zUM0D~_eQs%W~ZPpyiZ=3=D}#j%cSuT-w(gFKnhE685CQ(@gbUg#nSTFyN4@-53blU
zhMIhjH15`xo-K;a!LO;zo$7Xsy1TZasCBqN34oz=#jidg1W7{#&LNZ?yW>oUvgiZ*
zdli)IXw>8pFu15k(y0Uo5MgPDwWoF0Gs&fT2=Ad5a==+H7gn-C4SaJLf@QhWO?4j@
zrmimAh}lnyh#$b@DX2X;SoO_jlyxO<8*=PsH~)A1%F(`rFH=VqL`tAg8?U^PMx8Ul
zY(I_@>K^i8_RXDk56_9%vrNwjl9qyf4O;rFOskFYkLJRQZ!B<VqJBq*LZ!EydS;nl
z-iYu3ad3zbH7pgToeiXivm!kkqGR@54DT1J746Etywf3Nwcz<6i5Nj5!>v{k<9$|i
zMwR=9vO&R{t}0>_sjNu+QaFE+=uNJ)s9!(>u7Rsb%I>pP1<Zw8J`IG`R}Us4SSGiX
z&+Xz)LDyRbLoUVywc)0%arp!z9B&98j_MzyXu!3==6aLs-ntR+NjmgvB4+uJV~M$4
zgqRcHV+StZZhEw!^8wYoz!yMPoX|cR1}oufwF)<}M0X9hpi-kPOuFcb)5eDkMNd=7
zu4UMsXbda{)6d@^-;U8?xA}4ODA9vv*mw8vwGTO4Wl1WqJ@8Y>Dtn*<NOA9dQfkY0
zg^hiO4M|nAt@qAOIGQ+Lr!O$@5?zaY_NsUUaQji<^}H5Joe<$Qm3noMWW3w0>pIbF
zC@lmU`Ys+wzn3z?{+Bd+@^XFIaHQ~dN!g4QA~@z}So?-!?B!rX7Blt$)}?2}l%3%v
zLaN)=)M6)9{W!#-qO~|-OZ}0v6YJiZjCd7_`QElwFj5L$<8XcM^F|1dYmFFR0hkY~
z<K;$hSR9K<epGGUv~XFi+bz7J2amQVu-XP~sQ3DCa3|l4YMk)Sizfdt9wWSF>f+ks
zWgQ-33X;Yf;s2`E-u9GfZkk`coA-;8_=R`dv&ZGTom9eHQdZf;5>&Ui4VZ2@lH}O{
zDsKL|ur|b!O&W3Ck!b|(q|7N2HLX{Bd#*)P0g%`P)-8!n#FG=%Y`Y_)cbkcWWEEr8
zo&A!s{u-;YFU>EH6mxpVCS$N$LF$6dj~CMLRqP_v_<cuOu9cR<Za-;)K+=stu3aR*
z`eQWPz^UB;?noQ1Z)!KN*FN+7xs&0Q$Au-NCh3V8zxN*^M7gewg7>hvLZ0*HWu`)*
zoi~3Y7`>Ha-t%SZiL6S52AHyI3U`sGt=4K4ulQ%jw3-C}PDU0;>Pi=iN{rL+xSLzQ
zy>$E-Qg}jv?E-|Q53h}ZY2{=tv1-nJKv;F8u@AaUt3V(n2*H-%?w_#PdBNi?v-|v2
zGnRW#Z=1D(53P>k2Qsqouq$R4yfg{MOQi8wpq8r?!-QRn(r0TWylsKwK_cZ5bug*Y
zm^&O9Q)Oxb20eb%V^5hQlULU4Q2Rp{7seUJOs-W%#<@nQ&|FYBY#SBrCzXcOPkfjg
zXK0a0!s9&0zOr_Cj<P*%jEaf-x6IhNqzB&|7&v;ou%Uss?e#Uai4-Dp#L~o84IoR+
zgV}&U=&LOztj@{`lB0*w^uOpT=l@z2Wz`J`#jH4Wa*ud|KbSp)O|5=yH@iFjTC{>w
zVK;7H$UU3kNNUdt792L`NQUO9#fevD8S{?IXM>&UP7hyxhI@@o<$a_bAF_gJm{sPJ
z6FVCt67jX-w|;usY*@eWY9j>acjAuOW4E<Vdd-f<e3mHngT&WV#-JPufM5sA-g3_r
zibfvJ57ton@ATmMsr2wkg7fD^)g=vJEYQp9xy76PTGtI&S>>XzT<WlbD|+QZE4OkH
zy=9rvpo5H2rS<wOci@U<8a|2*W@JrYCl~KV=Xz$guzQ$<-oQeb1}yRCA?ZX^@$kV<
zax1>7ObS|4yfxoL|ERkSlMGWKPbeStq4m96%E^?FwSi8Lc+Mq3>*s)P<13?A7O|do
zDLP@bysc>uGw<6yboHIeiF)lDwfq>HKWkO?(MQbDvfd~wHl6Bko%wt)TV0#2XaGMp
z?kabsE9TATrce=$Jxz*@YN#ay+>?Q6A|Q0Qdr9kQ&7ZOjDD$Q{Gdk>QV6ip|1)7Q9
z$Vhmz)#uQZo+t_R18E2MjOdjP3vP|L`hM^v0zClUr0z7A>dCFG^|go&;H<CaIwV(?
zX5h2Ks7jh4Kq<5hUF7YBNN%h2??-yj9zL#T_CKh13iuIpV&rRAHR?NBWPRpShEYx7
zwHi%}RpG3&pP_G!=#gTl^d!qt`vh-!GQU5BWtOFOgeJOgb$zv`?iu+#^u=BG29WcU
zjr^q|xA@q8A;m^K4(9jL97u+)8DzYhB$Q>zR1_FG6-GPtPo<bj55wY&OkOKnYB&*z
zZjz+Z1;;Iy1ovU0@*=nXsYxcjcDD&LdZ-gh3>N>wGLe!e9bmrYU50a^E7zlr;V_%N
zD)BW$!JGve{_cNO1b+&HznZgW*2?U3hm<qU+YDos-aiW`2tUe+C*dkS4bEEDs}J!j
z$5@s!@G>($sUvqTLExaiD-!`RcPLj_Tq{=3{H*tMQpNnfqQROyP>eQmEt;N2WC%>9
z_J&~(W}>0zsiaLss>xg=vRgQF8&<cwK-yhwU9AAG-UD6uwOF!t!=3KYKOYlE?h@1f
zm}TfACFh?_Df=$nLFX}sl#BB>@@<hc9e+lXm3F#RYGnHqc(2W}{zX4*(5-NC|7`-R
zq}m9pmNH+`8+{fvY-~>m8@(p2^q%JSMlUJvs&?+0iQnmB=-FHG9YS7rGB}0FGu&hh
zGT0*%%}IE~lvUFc^khj<N)=mIJHfPiHtWZ8-`6<v$W3*;ZD6pE@;1L@6ROhY<@Sc{
zLk<RT?P;Gr@yzX=)a(hj3~!v3c7hc^p<~C|!FQrBN1nRKeMvq<BS(^KZR%&Png+Y`
z-HmsFd2NIHsXbk~+H$#?V)I<Ig*0^fU_vvcPKTU|cg2zr15k{|!<!xrFv8)BjU8u?
zI}-8d?dH&^x{3EhXh#{{LzbRaw*O5LBr95ZuE8kV(S7X_l5*8Vis|$Q6&;sUZ)t(T
zbW7>@oVzHSVexRN<-Q~2DRm|qHU1P|yVInT-fIDu<1<UgQHE=^J*!Cow|pM%m(SXh
zH7&i^A05;)a4Y-UAGHfISM%N3?9v64x_RcpJJyXN0wT>lEsDevn=j|l8I@{sZWoGO
zBrHD;(q~k~?LGwJa>G{NMSQM*ENbf4`WIdaaPggWSw9dpEiUS7D@BqJZ=efZ55nL;
zODBkzH;Q{VNyQXjovJ%%uT!v)TQ?pXSXvrckwkQ-ulHHj;)+)dxg))Hv>3Mr--HjP
zD&S7ZazIqSiq0eU*AC|Kr=%S(vJrWty6pyScXxrcSQWSgNy06T){}|mu?`rFgXre(
zhoDPznMv#-vx&-X|AsVG;wK~Fa~-&keYLCmA?6ty%jST-WI<4A_9%{{CrSlQv6;S-
z_!lXw65IRDC49>%p~l?&TXR4&?39}rj<?V9or`rMsEdy9&pUc(ss>U^)NQQrw>n<Y
zG1H6l<`q$a>OB{q7z(=MIV<@kY1<M`6+z|FaPL!>?zjY%VfS6RS_LQ;w|qo_i}3ez
zkgf50{=DTLO1wWrf6KYq3r_n<<uz!OMq|WL`Z4-^=i4ztBeE#-vC}JI<(?q9_N(?H
z=Iffh`eoNCBLmn})3D&_las3zPXYrq78rBgn1v3f!wd0ots6MirYp%r|JW$sLBeEe
z7?Xg{r7+tO<cx1+y&?>}Dnph};$<xPe{=*sSw(^J`PSJn5#6VckTrGhQ23|*?29BC
z!tZe=S)0JTX-iovyV>iDHtW7j%kPt;ym|rydcEtbVSaBQ3nG6~;%XvFKJCq3_`Xxy
zfEF7sO>IjA3GKK*Dqj=#M<s~Q0=wgsjR6e4+>rnfmktBnR;zA8#Sh@?TV4uQ^~=MO
zg$I2&lV#3MvaZ*^5bE_gO#bKhe)rcJSbXhC)LgyENyP+wFNoPtSx>d#R$k(oRS9xD
z+E3LjT&@s1oY@8}QoEqXyIZT`33d{Zq1Kcu{m~*#QFHC}y>|0$xZ00}U5xuAZ~5fW
e{|6U^O+Qpov+E8&+y~4KI(+B|tn`;tzy1^S?}<qO

literal 6354
zcmcgwd05i<)~9JE=T3_k?dG`UymH+#r82c#>qc6cIcj66q2+=B;f9KsXv=K^Q*2yw
zFB!MoCKXLZwA{+wLJ<r}O)(V-aRU|ZkM8`_`}W@DxxMcnJm-17{NSAL`JHn<pU)xv
zc+P(3cWU2(K%kwcPr+P3ps&3^pe@I~-3FZ8dHA#%Fn<U><&Fh`3=-v^EsXy%-3c73
zhMl|^<{E?u!(9pW2jOryV`Ly2>w6{G-#92VAZx)&4Fvk$`!wwM`N(WOhY(*7fY*Av
zj_37Mq$g-jFrGd7<BpLhxQO8nY(!sX+zoi@nfv$kYkrQ4yQS_Er?&5oKJhz`<0ZEk
zNs4!VO}Y4;vHpGbY1rJWhlS-gKRn6JQS}wf&$I2m{`UJFWZ!GPg3j#K_gd|!As<n0
zPCd_huF3eNjMKj<jp#31T{9PHq<f&V%NJ-9ah%&3If+X-TcfIM>Kbb51b+nX;^gnh
z-2zxp?&IB1xjp?Knlt58!46LZVf=A@VOvFo$nka0wRrO}+i~9*eeIag=O(Ks+n9P*
zTT9N6py^j*Fi9rVTv}}Z?CgcALqH%_Lpvn9extIpM%A!6MV)*Kw$!0f5C5E<dhddH
z!5a(bG8db}b)lo4dKOz>4tpP9duByo@=y2&`@$}zD}4(By)}6869l-&jX%e!d<k=#
zzSZg%-|dw28Xe=DcV|hV4k;2=g@~G_oUFGMa4>^F-Ah*sfiogGumHA&?dqKZh<gU`
z1xR=NF&Hq%l-jp{0h9DB!pe`|)m|EqxtK}XHFPmf*T($iFCmShF;Z?wTH{=mO3vtl
zjd+gLkAwuW8mq#({~R!^vWr%q(f~DlU1K=8IMz)M9_0{tKs1-JwFQv|j`QWFF(J==
znZ4!vIS+ohE91BIHJ}fO=MFt_>1qNWbQUT(d%K}&`*j9XFJEFO!u~q(4Upj-yf1sP
zUloQ`Dl+d~wY7Sgq1)BzfLw7}ddQMYk)|_7w(<m#Z=c`oNNzGNdX$!MUC-uxhpddm
z(mF@PB94e(zzO#9>xguXy-*ljy3+RKY@~z6o5s*!|5pLO%*}cRafrl{X!x|78+`;P
z@9dHN7ZLyiO40|q{XaK5hE?psleRjNrmRaHbPVM^)#7?5FH+uAr}DpuIrL!ZSVAQG
z$@ovL@dpBKX*!A^>ePyJVa3ChI_<0!CogO-kfObf+UrND*WNS+;Kl06+3s4qpn*Os
zcdHLMHm<ZmaJ}W_)~|58fT+v@CMI;Wr`c@dAbgSuOkG$~KU7cz6NX8*>h%W&)J{^f
z@fTn;Y(6A>x^DEUcbXD!$3WjggRZs>>+gTi`DHuo^$`)@Wy04p_uUR>^qV2v&*fDj
z7`}|7Y3^rRl+bWUT(VeHz^KF!GEG_^Vix$8^1jP54-vX()2jmf-%H{TT&r|QButMJ
zA9OK81D9)8mviL#H|zd1&?1p4FyjVT66vPx;PA8&Qt|3!p?EaiCcRb*WOiKm73f$+
zLg<yl%f0S%Xk~lIUxZ(^oDUrWrV8i>lDzc)hf!pH1G=FX68Us3%NaXno=z+f51r8J
z8`;?s_9G3bgk*iwvEAV6qurD5!H*)hM1F_;Xa9Dp#!O?O<e;dcY}2vr?rOz;P7+lM
zG@=0vkz@HD0mk~yGOsad)ow5_-X6GAsLAj7XJ^3I|BF#w(gYk8vgRq4&<A*l-OCZ{
z5$c>@hZm^ut~^1xZ3Ue+qF}Bn=m@qR^)>M9eHf6||6AiXfV-g=8g^y;Xu<rNI)Qh?
zrbcgXJvtD>|AmenXr>Y_SW{Rbvv26wYj=+uC`#WgE^ZQdM>M@@?~-!#`Yi=4_jO!P
zU(DRZZ3V3^Hur9$@%6#_Nlv0Q1TE|6=rA{Ko+JwbBl^b&XYeMI!a}<95KPq5!1}={
zKV@xjdHlmf*(SEM)H}{&v}`lr7IkwykIV2l6j7%app|PI{ed3O0fTWy`>n8-x?UfO
zNcV8!yTdonFvXUYPkmM|A6OH{@ArKFLX;;{VUC5pek#y-tuN|-UV~@jc?5pGXOQ(6
zj%-t&VAmhmF_}Z)jqEbva>BVEQ#~v#G4j&hCP1Or8|bdbwAd^-6V|4;-uG<x$ibK(
z&n)JWrmtl-ag9%0o1$a0$&<X?W+>{qCtI4#hn{R(uU=UlXDlS!s!+v>NW}*%;Vf~D
z(pzCLx%L;&ERN@Wea8gJut!!Mf^Tc4wmIStj!mV?;s??om=snD$Ff6<L}V7ulZnax
zAI4qghK`PQU(lYw=M)9><B_AavDwM~bMu>v=`sW?x>}&&T4ILAlv_&3Pc-T4Zy``J
zv416p;^YqrL{eBx9GQP$jIg*9EPPhB`DJQbUa@blg)ViK8mpS~=@p`aj%pO*p@B>O
zC7PK=Q`KCwD|@PbYN`@}`0DLxYfMsXkbQ!)Pnyx|Kt_7yPJ6|OjPin-^|YY%ixGwU
zJNMW@S^7dsjv^y!IC_$9hZnMgm*Oi#_Ia^`u?;JQSsfV}zzxS+r25Nxx<^92JRX#|
z-bJ?#;u_B7>GNgZ8$)rqdF+XT;Wa<xgdg|lL&e**^>u<B??&|1EA^=xqZOh6hxbZg
zFqLkl?KMWAT~YySb#4E;CAm}umXOtWF1<(JbMEJ2j&z$bl??H*56rew2Mb6Zic2}Z
z`%)#9dU(hR4N7?bh8TI+sDDqfb8HDi!lq--Qd}k-!OM<t9xkC<N_^Thx53y0#nwSN
z--FFpG~_e&ZB34;?LIk3{>J%~1{SWE0B9+3LpME23GLHmZbUL9SWl&yb$xJNWx<<Y
zsIqs$$<FGCtu3r$K)-+ZdZm1?B(-<p2<<>r2<eoe7s^O0%NrgQG#mqPk4vWVk??tR
zoM;W-Vm97qxXTKMD5TTpq79}V7n5#k$JE)0BW9H?_Br#kCw_8YN0ahws}4lP@WigY
z%~|p$E3*Pw)hkV1RTpa2ZPph^TZ=<ZcPx0E!TMcU0*EMvR4MRumo&xi2ZQZb1+w?G
z7&|!6p``yYQsB^4hi(%fFnQ-)XStQ8)72}2vMIj(9y@h70$sD(UuoPNRNYU*9-*NM
z`)Flz(aIt1ZnQT#UCl_yUe2wbkeM!(C8l0U)L1KJ*}E6dXR_eDKS4oRzhNVy2TN*c
z6w%E$uXHkNqc1OPcd!r5d({h6O;@pLc_c~A){`Q=#uIB6Nsnj)Nw+T1Ss1IUT3X=8
zER5f2S4n$e;B}_Fz%DFCYk+Z)Ostvke`98=>C7DwADdWDQVNl6%&s(#Ir(9(gapC7
zgi5AkZkE!Y(_9wFL`M~<W5&@R(R`^**uCud1n$$&@bG=~!BrS}fBZHu;NHM=Y+!kr
z9Q=1wO86|s{mX|h9|`~h|3ppbX*%Hw=LIAR)=-Gt(>*CGq9T-73}}Ir3ou4(cE%rW
z7@hB`V|#E%!lMsVaC_oKFcebuQ+aQ*z=8`CUxpE)5QQdg2FzM&V3;b0RihZtnQ>M$
zX2wno%gH4z=Bh1M^}yaCvsNfnBSQi&GJE2i%bBTFQR1#wu?v<R??<X>8wtdjYyw6$
zQ$MJ??W@b^p~(5gp%bn$9AnDRP(d%lTirO9n3+1=6VZ1fc2jsyP|t&LkDlw?n8kD_
z_h_jaOCML>Heu4IrhLI6Wde=F-;w(zL)z9Y`OsoX8NL8$Dak;}w4~c@*SX%r4giXS
zqY7&y4X9q{A$uf!TL>dpJ9#T|DsrRu9|Hg9Qv5NSAkf)nWuEj!o|gN?GWp=ArgUo;
zZlp)$i(3y}wPl9W%!E+NaC*?nlWPBIW&1sQ%*`70Tocw>HfA4j1(uzYANzRYnCa|q
zf*qrU=kUO25#3cBQvt7r=h)H?NJ^{zq*x)mg_LnhoTUz2RcX+tb2o!sYIof&-AfBx
zDL5)04J}BK^1rJTw-FS8UxGq48`tu^M;hrdgpP$Gc<^!23*26DV8_C1=X~is>YesF
z9zST;JiLidt`L2cZ;^40@|g-g*f&BDm3o<0^DGnn%52~!+;zQ^$Wc<D;%8Y#TY#Jp
z4IIb+CivQE=dvY1@A7p6Io)5o*@s`N%)<+<SF>j<6|#~4Y!m^n?sWve*PIEtV7zzE
zvQyCoVjN%W#GTKLD6GGVIbt;y?)8!2WYaxHo50uC$->(jdyQirG9p_GRfbj)b?cA4
z`v-C<sbn*l^Siv{TPiF)N?A#>?(XIS!2#iBnNZsu5}q2PXcYpDfx97|3*>L*YO+PO
z?^9;`sn&H}A8!D`Qi`#2ty^Oo=QcVzgrgrh3<5DbyK?8w_J}kV0nS4+9L^vfnz)l3
z^u!h;2R+lEHjy75^_-Gb{+*Wu_ucGMK=tT!Z293?2cGBMWMdWl`1Q4GRzRP1J)1uZ
z=iu8XHZBirT;v|ru3cC~U=Bw`m>8#DJ6YJ9X)qb}esQs!z<b~pr*Z@7Hi^O11kG94
z=vzV)i4YRWBFpG@9twRO%fuISCl@dGVi@K)&DeQAQ9f}k+9zM&+ld`zOIa&6$~X8m
z&Z9DRc!Vy3Y}Xd;lTA8oA|%(=GFNgS+CGJcarSMwj!@KODh3nlD<YW%X*jBbtL2_Q
zE#=R$<P5f~)S0w)*+^JV#~#x{WJX;8*ef6tX#lS>_Jm8$!=fJxG=LLO(SMT@zbD;w
z9Z(ekw?fKe-6<PxOtG%dBRkp=BPZobYpY8%C|c&%HKh}BJz<fnJsZUoA2;Ckb-f~Y
zn|GfLlUnHOOjA0zFS}gN5SGWs<-pb~Fi{A(!r(`1CzcjS8NlHYw5*=DIj|0GSRS5q
zU%zQ|exjMFmKd%Xq%A$3@bm!WjM4H&YiqkV&?FXTNDiU0+<ncil)$izGHMLkF!$o<
zQ){36RjpWaR1c=Tx|u6MMm6P~a_x@hWJ3RHTMW+OdhYWAnOzC=pSlMY8p3&s(D_&g
zQj!}jrMTI`S|-d=|CJ;eY%0BLlOnlK21@BI)8|&@ZODbaktd8W9zrE5{b==)PeTeZ
z4*;}es%(vt2&{W{LxZ<H@OGCZ6W5X}MCk1l6+1@CgbvDz0;8GxukN-OtdT`OrESba
z*kHM{uZkcOujRhu1NS?-#-+QUNhWWX$MON-G>gG5Y`+^Quxi$ElFP8SGRl9ZtPjmt
zf?8O9y4&)K57J|AL6}UUO5rz|x&<5VCCd<aC)(9EOI6QRdCC>j77pi)>dT5@O}Zwp
z5vfYviE4rw3vY<j4YuLq^J|IoIa6FJD}={vOp8P^A{`OBH;6>t*+z8RH$aQE?F>oV
zF!xJXL_0FBH}u86$M;(v?AcNCO{PZ>QAt)Ux~U|nFwIT<9vPRyg*wY<qP33s3%aZr
zw~a?z9_+!EDlj2F099iL3k#h4$rA;3))viCGbiL&a6QmA(g_w8V^U|Pp-EJ8h0_>B
z1wH)c@Xr)8l(k`8y0}$AaF4jqY`&dRWPiSW`~ASc<$=6&hnmea2tS~D<4k_a^ss%h
zx^g0?`DL}aNzWefSY?r@@;yLcjKIrf`d%nfe{bcxCdKf7q+0(s@N#-L^L(a5P1qps
zE4D{vM)T$Xpi2NWY?)poeX^W$y*~8q4Pqgq&VF<wihH!SwIt#LpKfxtG=hTfe~1j4
zvFuD{eb4OF&8RFQ$3`Qjv#+@2Dx2`V{rG*g6r^p{!9W`hj@+pk(}&$lyOwOZSDitH
zPwMiQ0}<Wc2tJued_~TT(B-M8Ij$e;8WR{4G|`b+SZtcPp6kj=nQD5v9f=jsDMH~7
z`wCW6wXULG<=ru;9?X4OB-lYvPn;`NaAXbP$e57O4?{i6Q!i5Qh39VtcLT_{r}``F
zz5rUV*(kX`(mnRs)u6*q%myDRnfI0Nho|rUgM1~)E4yka7wVFS|8m&^722X)QV@L<
z1()N(m4P0sHy`&`W<kKhu|9QlOcgAc{~eM4ZF2GXZN}%`<QIMN<qv0QAxT>c5~9M6
zNSaRf)lkP8aLWVsdiS=S^&yP6te+{dZ0t)fwEEE@XQn^ab+Y1RGLX0Irfpb52jx_K
z`w9)CG&SWMlFPZ1z%J5d$h+7y!tp?}O<h%uIaCqI@V_weKMMn&Jybm#^>*#-bw8Zu
zeBQgWo^unD=U0KX;ucP?nF%u<&6%;_4oB`ck^2P~3};I3Lt5KoJy7DDC5311^G^yn
n)|b3uYCeL;hSnF2OzaJM`&DVMN{0>L8+7{QIatMCF5mha?z5d>

diff --git a/docs/assets/images/logo.svg b/docs/assets/images/logo.svg
index 7d3b4f84b..440b3f0dc 100644
--- a/docs/assets/images/logo.svg
+++ b/docs/assets/images/logo.svg
@@ -28,8 +28,8 @@
      showgrid="false"
      showguides="true"
      inkscape:zoom="1.0083333"
-     inkscape:cx="280.66116"
-     inkscape:cy="25.785124"
+     inkscape:cx="280.66117"
+     inkscape:cy="25.785125"
      inkscape:window-width="3440"
      inkscape:window-height="1416"
      inkscape:window-x="0"
@@ -45,7 +45,7 @@
        id="guide315"
        inkscape:locked="false" /></sodipodi:namedview><text
      xml:space="preserve"
-     style="font-size:117.641px;font-family:'PP Neue Montreal Mono';-inkscape-font-specification:'PP Neue Montreal Mono';opacity:0.99;fill:#d33134;fill-opacity:1;stroke-width:0.612713"
+     style="font-size:117.641px;font-family:'PP Neue Montreal Mono';-inkscape-font-specification:'PP Neue Montreal Mono';opacity:0.99;fill:#e35a26;fill-opacity:1;stroke-width:0.612713"
      x="115.73952"
      y="91.977005"
      id="text314"><tspan
@@ -53,81 +53,82 @@
        id="tspan314"
        x="115.73952"
        y="91.977005"
-       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:'PP NeueBit';-inkscape-font-specification:'PP NeueBit Bold';fill:#d33134;fill-opacity:1;stroke-width:0.612713">outlines</tspan></text><g
+       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:'PP NeueBit';-inkscape-font-specification:'PP NeueBit Bold';fill:#e35a26;fill-opacity:1;stroke-width:0.612713">outlines</tspan></text><g
      id="g13"
      transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
-     style="fill:#d53135;fill-opacity:1"><g
+     style="fill:#e35a26;fill-opacity:1"><g
        id="g314"
-       transform="translate(1367.0913,152.50405)"><rect
+       transform="translate(1367.0913,152.50405)"
+       style="fill:#e35a26;fill-opacity:1"><rect
          x="28"
          y="644"
          width="28"
          height="28"
          id="rect1"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
          id="path1-7"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="28"
          y="112"
          width="28"
          height="28"
          id="rect2"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,560 v 28 h 420 v -28 z"
          id="path2-0"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,476 v 28 h 420 v -28 z"
          id="path3-4"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,392 v 28 h 420 v -28 z"
          id="path4-0"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 84,308 v 28 h 420 v -28 z"
          id="path5-8"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 364,168 v 28 h 140 v -28 z"
          id="path6-5"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 336,196 v -28 h -56 v 28 z"
          id="path7-1"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="M 336,112 V 84 h -56 v 28 z"
          id="path8-6"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 364,84 v 28 H 504 V 84 Z"
          id="path9-6"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 56,672 v 28 h 476 v -28 z"
          id="path10-2"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="56"
          y="84"
          width="28"
          height="28"
          id="rect10"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="112"
          y="28"
          width="28"
          height="28"
          id="rect11"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="84"
          y="56"
          width="28"
          height="28"
          id="rect12"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
          id="path12-1"
-         style="fill:#d53135;fill-opacity:1" /><path
+         style="fill:#e35a26;fill-opacity:1" /><path
          d="m 560,28 v 616 h 28 V 28 Z"
          id="path13-9"
-         style="fill:#d53135;fill-opacity:1" /><rect
+         style="fill:#e35a26;fill-opacity:1" /><rect
          x="532"
          y="644"
          width="28"
          height="28"
          id="rect13"
-         style="fill:#d53135;fill-opacity:1" /></g></g></svg>
+         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
diff --git a/docs/overrides/home.html b/docs/overrides/home.html
index 1114895e6..0c97d5ac6 100644
--- a/docs/overrides/home.html
+++ b/docs/overrides/home.html
@@ -117,4 +117,4 @@ <h2 class="subtitle" style="font-weight: 400; padding-top: 1rem;">
 </section>
 {% endblock %}
 {% block content %}{% endblock %}
-{% block footer %}{% endblock %}
\ No newline at end of file
+{% block footer %}{% endblock %}
diff --git a/docs/overrides/main.html b/docs/overrides/main.html
index 5cb6467e6..94d9808cc 100644
--- a/docs/overrides/main.html
+++ b/docs/overrides/main.html
@@ -1,2 +1 @@
 {% extends "base.html" %}
-

From 1894fa3ae297a75f992849f53cd85dbfeec2e9be Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 16 Sep 2024 15:11:27 -0400
Subject: [PATCH 072/505] Remove Broken pyairports Package, Replace with
 airportsdata

---
 outlines/types/airports.py | 9 +++++----
 pyproject.toml             | 4 ++--
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/outlines/types/airports.py b/outlines/types/airports.py
index b4ed6784d..934ae1844 100644
--- a/outlines/types/airports.py
+++ b/outlines/types/airports.py
@@ -1,10 +1,11 @@
 """Generate valid airport codes."""
 from enum import Enum
 
-from pyairports.airports import AIRPORT_LIST
+import airportsdata
+
+AIRPORT_IATA_LIST = [
+    (v["iata"], v["iata"]) for v in airportsdata.load().values() if v["iata"]
+]
 
-AIRPORT_IATA_LIST = list(
-    {(airport[3], airport[3]) for airport in AIRPORT_LIST if airport[3] != ""}
-)
 
 IATA = Enum("Airport", AIRPORT_IATA_LIST)  # type:ignore
diff --git a/pyproject.toml b/pyproject.toml
index dfda81acd..4d87e1fca 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,7 @@ dependencies = [
    "datasets",
    "typing_extensions",
    "pycountry",
-   "pyairports",
+   "airportsdata",
 ]
 dynamic = ["version"]
 
@@ -137,7 +137,7 @@ module = [
     "uvicorn.*",
     "fastapi.*",
     "pycountry.*",
-    "pyairports.*",
+    "airportsdata.*",
 ]
 ignore_missing_imports = true
 

From 2b1aed0bc4464abf2f185b1d376430554a52514e Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 14 Sep 2024 16:40:33 -0400
Subject: [PATCH 073/505] enable actual-byte tokens in reduced_vocabulary

---
 outlines/fsm/regex.py   | 12 +++++++++---
 tests/fsm/test_regex.py | 21 +++++++++++++++++++++
 2 files changed, 30 insertions(+), 3 deletions(-)

diff --git a/outlines/fsm/regex.py b/outlines/fsm/regex.py
index 8cfd81ead..6b105a7b9 100644
--- a/outlines/fsm/regex.py
+++ b/outlines/fsm/regex.py
@@ -905,9 +905,15 @@ def reduced_vocabulary(
         )
 
         if token_str:
-            # invalid utf-8 sequences are replaced with � (\ufffd), but there
-            # might also be tokens specifically for �, ��, ���, etc.
-            if "\ufffd" in token_str and not re_replacement_seq.match(token):
+            if isinstance(token, bytes):
+                # Handle BPE tokenizers where the tokens are directly stored as bytes
+                # https://github.com/QwenLM/Qwen/blob/main/tokenization_note.md#regular-tokens
+                token_str = "".join(byte_symbol(b) for b in token)
+
+            elif "\ufffd" in token_str and not re_replacement_seq.match(token):
+                # invalid utf-8 sequences are replaced with � (\ufffd), but there
+                # might also be tokens specifically for �, ��, ���, etc.
+
                 if re_llama_byte_token.match(token):
                     # llama-like tokenizers have <0xXX> tokens for all
                     # bytes >= 0x80 and represent all incomplete utf-8
diff --git a/tests/fsm/test_regex.py b/tests/fsm/test_regex.py
index 7418deca2..1789c4a7c 100644
--- a/tests/fsm/test_regex.py
+++ b/tests/fsm/test_regex.py
@@ -714,8 +714,29 @@ def test_reduced_vocabulary_with_rare_tokens(rare_token):
 
     [1]: https://github.com/dottxt-ai/outlines/pull/763
     [2]: https://github.com/dottxt-ai/outlines/pull/948
+    [3]: https://github.com/dottxt-ai/outlines/pull/1153
     """
     tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2")
     tokenizer = TransformerTokenizer(tokenizer=tokenizer)
     tokenizer.vocabulary[rare_token] = max(tokenizer.vocabulary.values()) + 1
     reduced_vocabulary(tokenizer)
+
+
+def test_reduced_vocabulary_with_byte_tokens():
+    class MockTokenizer:
+        vocabulary = {
+            "string": 1,
+            b"\xa1": 2,  # Qwen-Style
+            "eos": 3,
+        }
+        special_tokens = {"eos"}
+        eos_token_id = 3
+
+        def convert_token_to_string(self, token):
+            return b"\xef\xbf\xbd".decode()
+
+    reduced_vocab = reduced_vocabulary(MockTokenizer())
+
+    # See fsm.regex.get_token_transition_keys()
+    # FSM transition keys represents bytes as <null_prefix><hex_byte>
+    assert reduced_vocab[0][1][0] == "\x00A1"

From 55919506da45cd9e0289a23d9689f3e783f67317 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sun, 15 Sep 2024 16:38:48 -0400
Subject: [PATCH 074/505] Use OpenAI API For Structured Generation (json,
 choice)

---
 outlines/generate/choice.py |  27 +++--
 outlines/generate/json.py   |  38 ++++++-
 outlines/models/openai.py   | 213 +++---------------------------------
 pyproject.toml              |   1 -
 tests/models/test_openai.py | 116 ++++++++++++--------
 5 files changed, 134 insertions(+), 261 deletions(-)

diff --git a/outlines/generate/choice.py b/outlines/generate/choice.py
index 931409a59..595513d52 100644
--- a/outlines/generate/choice.py
+++ b/outlines/generate/choice.py
@@ -1,3 +1,4 @@
+import json as pyjson
 from functools import singledispatch
 from typing import Callable, List
 
@@ -5,6 +6,7 @@
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
 
+from .json import json
 from .regex import regex
 
 
@@ -24,13 +26,22 @@ def choice(
 def choice_openai(
     model: OpenAI, choices: List[str], sampler: Sampler = multinomial()
 ) -> Callable:
-    if not isinstance(sampler, multinomial):
-        raise NotImplementedError(
-            r"The OpenAI API does not support any other sampling algorithm "
-            + "that the multinomial sampler."
-        )
-
-    def generate_choice(prompt: str, max_tokens: int = 1):
-        return model.generate_choice(prompt, choices, max_tokens)
+    """
+    Call OpenAI API with response_format of a dict:
+    {"result": <one of choices>}
+    """
+
+    choices_schema = pyjson.dumps(
+        {
+            "type": "object",
+            "properties": {"result": {"type": "string", "enum": choices}},
+            "additionalProperties": False,
+            "required": ["result"],
+        }
+    )
+    generator = json(model, choices_schema, sampler)
+
+    def generate_choice(*args, **kwargs):
+        return generator(*args, **kwargs)["result"]
 
     return generate_choice
diff --git a/outlines/generate/json.py b/outlines/generate/json.py
index 6209840e2..f75878d29 100644
--- a/outlines/generate/json.py
+++ b/outlines/generate/json.py
@@ -70,9 +70,39 @@ def json(
 
 @json.register(OpenAI)
 def json_openai(
-    model, schema_object: Union[str, object, Callable], sampler: Sampler = multinomial()
+    model, schema_object: Union[str, object], sampler: Sampler = multinomial()
 ):
-    raise NotImplementedError(
-        "Cannot use JSON Schema-structure generation with an OpenAI model "
-        + "due to the limitations of the OpenAI API"
+    if not isinstance(sampler, multinomial):
+        raise NotImplementedError(
+            r"The OpenAI API does not support any other sampling algorithm "
+            + "than the multinomial sampler."
+        )
+
+    if isinstance(schema_object, type(BaseModel)):
+        schema = pyjson.dumps(schema_object.model_json_schema())
+        format_sequence = lambda x: schema_object.parse_raw(x)
+    elif isinstance(schema_object, str):
+        schema = schema_object
+        format_sequence = lambda x: pyjson.loads(x)
+    else:
+        raise ValueError(
+            f"Cannot parse schema {schema_object}. The schema must be either "
+            + "a Pydantic object, a function or a string that contains the JSON "
+            + "Schema specification"
+        )
+
+    # create copied, patched model with normalized json schema set
+    generator = model.new_with_replacements(
+        response_format={
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": pyjson.loads(schema),
+            },
+        }
     )
+
+    generator.format_sequence = format_sequence
+
+    return generator
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 692e64ac5..89c26f217 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,9 +1,8 @@
 """Integration with OpenAI's API."""
+import copy
 import functools
-import warnings
 from dataclasses import asdict, dataclass, field, replace
-from itertools import zip_longest
-from typing import Callable, Dict, List, Optional, Set, Tuple, Union
+from typing import Callable, Dict, List, Optional, Tuple, Union
 
 import numpy as np
 
@@ -74,7 +73,6 @@ def __init__(
         self,
         client,
         config,
-        tokenizer=None,
         system_prompt: Optional[str] = None,
     ):
         """Create an `OpenAI` instance.
@@ -89,13 +87,9 @@ def __init__(
         config
             An instance of `OpenAIConfig`. Can be useful to specify some
             parameters that cannot be set by calling this class' methods.
-        tokenizer
-            The tokenizer associated with the model the client connects to.
-
         """
 
         self.client = client
-        self.tokenizer = tokenizer
         self.config = config
 
         # We count the total number of prompt and generated tokens as returned
@@ -104,6 +98,8 @@ def __init__(
         self.prompt_tokens = 0
         self.completion_tokens = 0
 
+        self.format_sequence = lambda seq: seq
+
     def __call__(
         self,
         prompt: Union[str, List[str]],
@@ -152,107 +148,17 @@ def __call__(
         self.prompt_tokens += prompt_tokens
         self.completion_tokens += completion_tokens
 
-        return response
+        return self.format_sequence(response)
 
     def stream(self, *args, **kwargs):
         raise NotImplementedError(
             "Streaming is currently not supported for the OpenAI API"
         )
 
-    def generate_choice(
-        self,
-        prompt: str,
-        choices: List[str],
-        max_tokens: Optional[int] = None,
-        system_prompt: Optional[str] = None,
-    ) -> str:
-        """Call the OpenAI API to generate one of several choices.
-
-        Parameters
-        ----------
-        prompt
-            A string or list of strings that will be used to prompt the model
-        choices
-            The list of strings between which we ask the model to choose
-        max_tokens
-            The maximum number of tokens to generate
-        system_prompt
-            The content of the system message that precedes the user's prompt.
-
-        """
-        if self.tokenizer is None:
-            raise ValueError(
-                "You must initialize the `OpenAI` class with a tokenizer to use `outlines.generate.choice`"
-            )
-
-        config = replace(self.config, max_tokens=max_tokens)
-
-        greedy = False
-        decoded: List[str] = []
-        encoded_choices_left: List[List[int]] = [
-            self.tokenizer.encode(word) for word in choices
-        ]
-
-        while len(encoded_choices_left) > 0:
-            max_tokens_left = max([len(tokens) for tokens in encoded_choices_left])
-            transposed_choices_left: List[Set] = [
-                {item for item in subset if item is not None}
-                for subset in zip_longest(*encoded_choices_left)
-            ]
-
-            if not greedy:
-                mask = build_optimistic_mask(transposed_choices_left)
-            else:
-                mask = {}
-                for token in transposed_choices_left[0]:  # build greedy mask
-                    mask[token] = 100
-
-            if len(mask) == 0:
-                break
-
-            config = replace(config, logit_bias=mask, max_tokens=max_tokens_left)
-
-            response, prompt_tokens, completion_tokens = generate_chat(
-                prompt, system_prompt, self.client, config
-            )
-            self.prompt_tokens += prompt_tokens
-            self.completion_tokens += completion_tokens
-
-            encoded_response = self.tokenizer.encode(response)
-
-            if encoded_response in encoded_choices_left:
-                decoded.append(response)
-                break
-            else:
-                (
-                    encoded_response,
-                    encoded_choices_left,
-                ) = find_response_choices_intersection(
-                    encoded_response, encoded_choices_left
-                )
-
-                if len(encoded_response) == 0:
-                    greedy = True  # next iteration will be "greedy"
-                    continue
-                else:
-                    decoded.append("".join(self.tokenizer.decode(encoded_response)))
-
-                    if len(encoded_choices_left) == 1:  # only one choice left
-                        choice_left = self.tokenizer.decode(encoded_choices_left[0])
-                        decoded.append(choice_left)
-                        break
-
-                    greedy = False  # after each success, stay with (or switch to) "optimistic" approach
-
-                prompt = prompt + "".join(decoded)
-
-        choice = "".join(decoded)
-
-        return choice
-
-    def generate_json(self):
-        """Call the OpenAI API to generate a JSON object."""
-        raise NotImplementedError
+    def new_with_replacements(self, **kwargs):
+        new_instance = copy.copy(self)
+        new_instance.config = replace(new_instance.config, **kwargs)
+        return new_instance
 
     def __str__(self):
         return self.__class__.__name__ + " API"
@@ -313,81 +219,6 @@ async def call_api(prompt, system_prompt, config):
     return results, usage["prompt_tokens"], usage["completion_tokens"]
 
 
-def find_longest_intersection(response: List[int], choice: List[int]) -> List[int]:
-    """Find the longest intersection between the response and the choice."""
-    for i, (token_r, token_c) in enumerate(zip_longest(response, choice)):
-        if token_r != token_c:
-            return response[:i]
-
-    return response
-
-
-def find_response_choices_intersection(
-    response: List[int], choices: List[List[int]]
-) -> Tuple[List[int], List[List[int]]]:
-    """Find the longest intersection between the response and the different
-    choices.
-
-    Say the response is of the form `[1, 2, 3, 4, 5]` and we have the choices
-    `[[1, 2], [1, 2, 3], [6, 7, 8]` then the function will return `[1, 2, 3]` as the
-    intersection, and `[[]]` as the list of choices left.
-
-    Parameters
-    ----------
-    response
-        The model's response
-    choices
-        The remaining possible choices
-
-    Returns
-    -------
-    A tuple that contains the longest intersection between the response and the
-    different choices, and the choices which start with this intersection, with the
-    intersection removed.
-
-    """
-    max_len_prefix = 0
-    choices_left = []
-    longest_prefix = []
-    for i, choice in enumerate(choices):
-        # Find the longest intersection between the response and the choice.
-        prefix = find_longest_intersection(response, choice)
-
-        if len(prefix) > max_len_prefix:
-            max_len_prefix = len(prefix)
-            choices_left = [choice[len(prefix) :]]
-            longest_prefix = prefix
-
-        elif len(prefix) == max_len_prefix:
-            choices_left.append(choice[len(prefix) :])
-
-    return longest_prefix, choices_left
-
-
-def build_optimistic_mask(
-    transposed: List[Set[int]], max_mask_size: int = 300
-) -> Dict[int, int]:
-    """We build the largest mask possible.
-
-    Tokens are added from left to right, so if the encoded choices are e.g.
-    `[[1,2], [3,4]]`, `1` and `3` will be added before `2` and `4`.
-
-    Parameters
-    ----------
-    transposed
-        A list of lists that contain the nth token of each choice.
-
-    """
-    mask: Dict[int, int] = {}
-    for tokens in transposed:
-        for token in tokens:
-            if len(mask) == max_mask_size:
-                return mask
-            mask[token] = 100
-
-    return mask
-
-
 def error_handler(api_call_fn: Callable) -> Callable:
     """Handle OpenAI API errors and missing API key."""
 
@@ -427,11 +258,10 @@ def openai_model(
     **openai_client_params,
 ):
     try:
-        import tiktoken
         from openai import AsyncOpenAI
     except ImportError:
         raise ImportError(
-            "The `openai` and `tiktoken` libraries needs to be installed in order to use Outlines' OpenAI integration."
+            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
         )
 
     if config is not None:
@@ -441,15 +271,7 @@ def openai_model(
 
     client = AsyncOpenAI(**openai_client_params)
 
-    try:
-        tokenizer = tiktoken.encoding_for_model(model_name)
-    except KeyError:
-        warnings.warn(
-            f"Could not find a tokenizer for model {model_name}. Using default cl100k_base."
-        )
-        tokenizer = tiktoken.get_encoding("cl100k_base")
-
-    return OpenAI(client, config, tokenizer)
+    return OpenAI(client, config)
 
 
 def azure_openai(
@@ -459,11 +281,10 @@ def azure_openai(
     **azure_openai_client_params,
 ):
     try:
-        import tiktoken
         from openai import AsyncAzureOpenAI
     except ImportError:
         raise ImportError(
-            "The `openai` and `tiktoken` libraries needs to be installed in order to use Outlines' Azure OpenAI integration."
+            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
         )
 
     if config is not None:
@@ -473,12 +294,4 @@ def azure_openai(
 
     client = AsyncAzureOpenAI(**azure_openai_client_params)
 
-    try:
-        tokenizer = tiktoken.encoding_for_model(model_name or deployment_name)
-    except KeyError:
-        warnings.warn(
-            f"Could not find a tokenizer for model {model_name or deployment_name}. Using default cl100k_base."
-        )
-        tokenizer = tiktoken.get_encoding("cl100k_base")
-
-    return OpenAI(client, config, tokenizer)
+    return OpenAI(client, config)
diff --git a/pyproject.toml b/pyproject.toml
index 4d87e1fca..937b0c9c7 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -122,7 +122,6 @@ module = [
     "pydantic.*",
     "pytest",
     "referencing.*",
-    "tiktoken.*",
     "torch.*",
     "transformers.*",
     "llama_cpp",
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 8a7870efd..f4b97f36a 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,17 +1,14 @@
 import importlib
+import json
+from contextlib import contextmanager
 from unittest import mock
 from unittest.mock import MagicMock
 
 import pytest
 from openai import AsyncOpenAI
 
-from outlines.models.openai import (
-    OpenAI,
-    OpenAIConfig,
-    build_optimistic_mask,
-    find_longest_intersection,
-    find_response_choices_intersection,
-)
+from outlines import generate
+from outlines.models.openai import OpenAI, OpenAIConfig
 
 
 def module_patch(path):
@@ -62,44 +59,67 @@ def test_openai_call():
         assert mocked_generate_chat_arg_config.n == 3
 
 
-@pytest.mark.parametrize(
-    "response,choice,expected_intersection,expected_choices_left",
-    (
-        ([1, 2, 3, 4], [[5, 6]], [], [[5, 6]]),
-        ([1, 2, 3, 4], [[5, 6], [7, 8]], [], [[5, 6], [7, 8]]),
-        ([1, 2, 3, 4], [[1, 2], [7, 8]], [1, 2], [[]]),
-        ([1, 2], [[1, 2, 3, 4], [1, 2]], [1, 2], [[3, 4], []]),
-        ([1, 2, 3], [[1, 2, 3, 4], [1, 2]], [1, 2, 3], [[4]]),
-    ),
-)
-def test_find_response_choices_intersection(
-    response, choice, expected_intersection, expected_choices_left
-):
-    intersection, choices_left = find_response_choices_intersection(response, choice)
-    assert intersection == expected_intersection
-    assert choices_left == expected_choices_left
-
-
-@pytest.mark.parametrize(
-    "response,choice,expected_prefix",
-    (
-        ([1, 2, 3], [1, 2, 3, 4], [1, 2, 3]),
-        ([1, 2, 3], [1, 2, 3], [1, 2, 3]),
-        ([4, 5], [1, 2, 3], []),
-    ),
-)
-def test_find_longest_common_prefix(response, choice, expected_prefix):
-    prefix = find_longest_intersection(response, choice)
-    assert prefix == expected_prefix
-
-
-@pytest.mark.parametrize(
-    "transposed,mask_size,expected_mask",
-    (
-        ([{1, 2}, {3, 4}], 3, {1: 100, 2: 100, 3: 100}),
-        ([{1, 2}, {3, 4}], 4, {1: 100, 2: 100, 3: 100, 4: 100}),
-    ),
-)
-def test_build_optimistic_mask(transposed, mask_size, expected_mask):
-    mask = build_optimistic_mask(transposed, mask_size)
-    assert mask == expected_mask
+@contextmanager
+def patched_openai(completion, **oai_config):
+    """Create a patched openai whose chat completions always returns `completion`"""
+    with module_patch("outlines.models.openai.generate_chat") as mocked_generate_chat:
+        mocked_generate_chat.return_value = completion, 1, 2
+        async_client = MagicMock(spec=AsyncOpenAI, api_key="key")
+        model = OpenAI(
+            async_client,
+            OpenAIConfig(max_tokens=10, temperature=0.5, n=2, stop=["."]),
+        )
+        yield model
+
+
+def test_openai_choice_call():
+    with patched_openai(completion='{"result": "foo"}') as model:
+        generator = generate.choice(model, ["foo", "bar"])
+        assert generator("hi") == "foo"
+
+
+def test_openai_choice_call_invalid_server_response():
+    with patched_openai(completion="not actual json") as model:
+        generator = generate.choice(model, ["foo", "bar"])
+        with pytest.raises(json.decoder.JSONDecodeError):
+            generator("hi")
+
+
+def test_openai_json_call_pydantic():
+    from pydantic import BaseModel, ConfigDict, ValidationError
+
+    class Person(BaseModel):
+        model_config = ConfigDict(extra="forbid")  # required for openai
+        first_name: str
+        last_name: str
+        age: int
+
+    completion = '{"first_name": "Usain", "last_name": "Bolt", "age": 38}'
+
+    # assert success for valid response
+    with patched_openai(completion=completion) as model:
+        generator = generate.json(model, Person)
+        assert generator("fastest person") == Person.parse_raw(completion)
+
+    # assert fail for non-json response
+    with patched_openai(completion="usain bolt") as model:
+        generator = generate.json(model, Person)
+        with pytest.raises(ValidationError):
+            assert generator("fastest person")
+
+
+def test_openai_json_call_str():
+    person_schema = '{"additionalProperties": false, "properties": {"first_name": {"title": "First Name", "type": "string"}, "last_name": {"title": "Last Name", "type": "string"}, "age": {"title": "Age", "type": "integer"}}, "required": ["first_name", "last_name", "age"], "title": "Person", "type": "object"}'
+
+    output = {"first_name": "Usain", "last_name": "Bolt", "age": 38}
+
+    # assert success for valid response
+    with patched_openai(completion=json.dumps(output)) as model:
+        generator = generate.json(model, person_schema)
+        assert generator("fastest person") == output
+
+    # assert fail for non-json response
+    with patched_openai(completion="usain bolt") as model:
+        generator = generate.json(model, person_schema)
+        with pytest.raises(json.decoder.JSONDecodeError):
+            assert generator("fastest person")

From 289ef5d7db674fc159f56fa14f5bcf45b4e4bec2 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sun, 15 Sep 2024 16:39:10 -0400
Subject: [PATCH 075/505] reflect in docs: models.openai with new interface,
 also remove references to gpt-3 and gpt-4

---
 docs/reference/index.md         |  2 +-
 docs/reference/models/models.md |  6 +++-
 docs/reference/models/openai.md | 50 ++++++++++++++++++++++++++-------
 docs/reference/text.md          |  2 +-
 examples/babyagi.py             |  2 +-
 examples/math_generate_code.py  |  2 +-
 examples/meta_prompting.py      |  2 +-
 examples/pick_odd_one_out.py    |  2 +-
 examples/react.py               | 21 +++++++-------
 examples/self_consistency.py    |  2 +-
 10 files changed, 63 insertions(+), 28 deletions(-)

diff --git a/docs/reference/index.md b/docs/reference/index.md
index a5357fd86..0cd42ba88 100644
--- a/docs/reference/index.md
+++ b/docs/reference/index.md
@@ -10,6 +10,6 @@ By default, language models stop generating tokens after and <EOS> token was gen
 ```python
 import outlines.models as models
 
-complete = models.openai("gpt-3.5-turbo")
+complete = models.openai("gpt-4o-mini")
 expert = complete("Name an expert in quantum gravity.", stop_at=["\n", "."])
 ```
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
index 34b5be4cf..4d2dda8c9 100644
--- a/docs/reference/models/models.md
+++ b/docs/reference/models/models.md
@@ -42,7 +42,11 @@ model = outlines.models.openai(
 | Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
 | **`outlines.generate`** |        |                     |      |           |           |       |         |
 | Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Structured*       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
+| __Structured__    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| JSON Schema       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Choice            | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Regex             | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
+| Grammar           | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
 
 
 ## Caveats
diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
index 7f610c179..5c737c916 100644
--- a/docs/reference/models/openai.md
+++ b/docs/reference/models/openai.md
@@ -2,22 +2,21 @@
 
 !!! Installation
 
-    You need to install the `openai` and `tiktoken` libraries to be able to use the OpenAI API in Outlines.
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines.
 
 ## OpenAI models
 
-Outlines supports models available via the OpenAI Chat API, e.g. ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.openai`:
+Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.openai`:
 
 ```python
 from outlines import models
 
 
-model = models.openai("gpt-3.5-turbo")
-model = models.openai("gpt-4-turbo")
+model = models.openai("gpt-4o-mini")
 model = models.openai("gpt-4o")
 ```
 
-Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.AsyncOpenAI` as keyword arguments:
+Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.AsyncOpenAI` as keyword arguments:
 
 ```python
 import os
@@ -25,7 +24,7 @@ from outlines import models
 
 
 model = models.openai(
-    "gpt-3.5-turbo",
+    "gpt-4o-mini",
     api_key=os.environ["OPENAI_API_KEY"]
 )
 ```
@@ -56,8 +55,8 @@ from outlines import models
 
 model = models.azure_openai(
     "azure-deployment-name",
-    "gpt-3.5-turbo",
-    api_version="2023-07-01-preview",
+    "gpt-4o-mini",
+    api_version="2024-07-18",
     azure_endpoint="https://example-endpoint.openai.azure.com",
 )
 ```
@@ -111,6 +110,37 @@ model = models.openai(client, config)
 
     You need to pass the async client to be able to do batch inference.
 
+## Structured Generation Support
+
+Outlines provides support for [OpenAI Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode) via `outlines.generate.json`, `outlines.generate.choice`
+
+```python
+from pydantic import BaseModel, ConfigDict
+import outlines.models as models
+from outlines import generate
+
+model = models.openai("gpt-4o-mini")
+
+class Person(BaseModel):
+    model_config = ConfigDict(extra='forbid')  # required for openai
+    first_name: str
+    last_name: str
+    age: int
+
+generate.json(model, Person)
+generator("current indian prime minister on january 1st 2023")
+# Person(first_name='Narendra', last_name='Modi', age=72)
+
+generator = generate.choice(model, ["Chicken", "Egg"])
+print(generator("Which came first?"))
+# Chicken
+```
+
+!!! Warning
+
+    Structured generation support only provided to OpenAI-compatible endpoints which conform to OpenAI's standard. Additionally, `generate.regex` and `generate.cfg` are not supported.
+
+
 ## Advanced configuration
 
 For more advanced configuration option, such as support proxy, please consult the [OpenAI SDK's documentation](https://github.com/openai/openai-python):
@@ -146,7 +176,7 @@ config = OpenAIConfig(
     top_p=.95,
     seed=0,
 )
-model = models.openai("gpt-3.5-turbo", config)
+model = models.openai("gpt-4o-mini", config)
 ```
 
 ## Monitoring API use
@@ -158,7 +188,7 @@ from openai import AsyncOpenAI
 import outlines.models
 
 
-model = models.openai("gpt-4")
+model = models.openai("gpt-4o")
 
 print(model.prompt_tokens)
 # 0
diff --git a/docs/reference/text.md b/docs/reference/text.md
index 3d0d130f1..f36a67da0 100644
--- a/docs/reference/text.md
+++ b/docs/reference/text.md
@@ -9,7 +9,7 @@ Outlines provides a unified interface to generate text with many language models
 ```python
 from outlines import models, generate
 
-model = models.openai("gpt-4")
+model = models.openai("gpt-4o-mini")
 generator = generate.text(model)
 answer = generator("What is 2+2?")
 
diff --git a/examples/babyagi.py b/examples/babyagi.py
index 6af2ead88..0a7a0b13b 100644
--- a/examples/babyagi.py
+++ b/examples/babyagi.py
@@ -10,7 +10,7 @@
 import outlines
 import outlines.models as models
 
-model = models.openai("gpt-3.5-turbo")
+model = models.openai("gpt-4o-mini")
 complete = outlines.generate.text(model)
 
 
diff --git a/examples/math_generate_code.py b/examples/math_generate_code.py
index 5bf16e86a..7eb1651a7 100644
--- a/examples/math_generate_code.py
+++ b/examples/math_generate_code.py
@@ -35,7 +35,7 @@ def execute_code(code):
 
 
 prompt = answer_with_code_prompt(question, examples)
-model = models.openai("gpt-3.5-turbo")
+model = models.openai("gpt-4o-mini")
 answer = outlines.generate.text(model)(prompt)
 result = execute_code(answer)
 print(f"It takes Carla {result:.0f} minutes to download the file.")
diff --git a/examples/meta_prompting.py b/examples/meta_prompting.py
index 8f9ba5cee..cba18b5fe 100644
--- a/examples/meta_prompting.py
+++ b/examples/meta_prompting.py
@@ -140,7 +140,7 @@ def run_example(model_fn, question, model_name):
     parser.add_argument(
         "--model",
         type=str,
-        default="gpt-3.5-turbo-1106",
+        default="gpt-4o-mini",
         help="The Large Language Model to use to run the examples.",
     )
     args = parser.parse_args()
diff --git a/examples/pick_odd_one_out.py b/examples/pick_odd_one_out.py
index 6c4d45927..6cd9f1daf 100644
--- a/examples/pick_odd_one_out.py
+++ b/examples/pick_odd_one_out.py
@@ -31,7 +31,7 @@ def build_ooo_prompt(options):
 
 options = ["sea", "mountains", "plains", "sock"]
 
-model = models.openai("gpt-3.5-turbo")
+model = models.openai("gpt-4o-mini")
 gen_text = outlines.generate.text(model)
 gen_choice = outlines.generate.choice(model, options)
 
diff --git a/examples/react.py b/examples/react.py
index ccd74dbbc..34b3c6eb2 100644
--- a/examples/react.py
+++ b/examples/react.py
@@ -13,6 +13,7 @@
 import requests  # type: ignore
 
 import outlines
+import outlines.generate as generate
 import outlines.models as models
 
 
@@ -45,25 +46,25 @@ def search_wikipedia(query: str):
 
 
 prompt = build_reAct_prompt("Where is Apple Computers headquarted? ")
-model = models.openai("gpt-3.5-turbo")
-complete = outlines.generate.text(model)
+model = models.openai("gpt-4o-mini")
+
+mode_generator = generate.choice(model, choices=["Tho", "Act"])
+action_generator = generate.choice(model, choices=["Search", "Finish"])
+text_generator = generate.text(model)
 
 for i in range(1, 10):
-    mode = complete.generate_choice(prompt, choices=["Tho", "Act"], max_tokens=128)
+    mode = mode_generator(prompt, max_tokens=128)
     prompt = add_mode(i, mode, "", prompt)
 
     if mode == "Tho":
-        thought = complete(prompt, stop_at="\n", max_tokens=128)
+        thought = text_generator(prompt, stop_at="\n", max_tokens=128)
         prompt += f"{thought}"
     elif mode == "Act":
-        action = complete.generate_choice(
-            prompt, choices=["Search", "Finish"], max_tokens=128
-        )
+        action = action_generator(prompt, max_tokens=128)
         prompt += f"{action} '"
 
-        subject = complete(
-            prompt, stop_at=["'"], max_tokens=128
-        )  # Apple Computers headquartered
+        subject = text_generator(prompt, stop_at=["'"], max_tokens=128)
+        # Apple Computers headquartered
         subject = " ".join(subject.split()[:2])
         prompt += f"{subject}'"
 
diff --git a/examples/self_consistency.py b/examples/self_consistency.py
index e7468d5f9..f1bbe2a18 100644
--- a/examples/self_consistency.py
+++ b/examples/self_consistency.py
@@ -55,7 +55,7 @@ def few_shots(question, examples):
     """
 
 
-model = models.openai("gpt-3.5-turbo")
+model = models.openai("gpt-4o-mini")
 generator = outlines.generate.text(model)
 prompt = few_shots(question, examples)
 answers = generator(prompt, samples=10)

From e07f5500fb92acd3647f2aa13aebaf292b3fcbbe Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Sat, 21 Sep 2024 16:08:04 +0200
Subject: [PATCH 076/505] Remove link to Outlines twitter account

The account is going to be deactivated soon.
---
 README.md | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/README.md b/README.md
index 87109d182..9b01c5662 100644
--- a/README.md
+++ b/README.md
@@ -3,7 +3,6 @@
 <img src="./docs/assets/images/logo.png" alt="Outlines Logo" width=500></img>
 
 [![.txt Twitter][dottxt-twitter-badge]][dottxt-twitter]
-[![Outlines Twitter][outlines-twitter-badge]][outlines-twitter]
 
 [![Documentation][documentation-badge]][documentation]
 [![Contributors][contributors-badge]][contributors]
@@ -359,10 +358,8 @@ answer = outlines.generate.text(model)(prompt, max_tokens=100)
 [contributors]: https://github.com/dottxt-ai/outlines/graphs/contributors
 [contributors-badge]: https://img.shields.io/github/contributors/dottxt-ai/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
 [dottxt-twitter]: https://twitter.com/dottxtai
-[outlines-twitter]: https://twitter.com/OutlinesOSS
 [discord]: https://discord.gg/R9DSu34mGd
 [discord-badge]: https://img.shields.io/discord/1182316225284554793?color=81A1C1&logo=discord&logoColor=white&style=flat-square
 [downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=89AC6B&logo=python&logoColor=white&style=flat-square
 [pypistats]: https://pypistats.org/packages/outlines
 [dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
-[outlines-twitter-badge]: https://img.shields.io/twitter/follow/OutlinesOSS?style=social

From 77c6d67913b24cef75fc7f40ab079c51715209e6 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Tue, 17 Sep 2024 16:50:08 -0400
Subject: [PATCH 077/505] Don't re-use logits processors in
 SequenceGeneratorAdapter, copy them

---
 outlines/generate/api.py        |  9 +++++----
 tests/generate/test_generate.py | 10 ++++++++++
 2 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index ad01377c0..4919f2090 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -1,4 +1,5 @@
 import datetime
+from copy import copy
 from dataclasses import dataclass
 from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
 
@@ -503,7 +504,7 @@ def __call__(
         completions = self.model.generate(
             prompts,
             generation_params,
-            self.logits_processor,
+            copy(self.logits_processor),
             self.sampling_params,
             **model_specific_params,
         )
@@ -525,7 +526,7 @@ def stream(
         return self.model.stream(
             prompts,
             generation_params,
-            self.logits_processor,
+            copy(self.logits_processor),
             self.sampling_params,
             **model_specific_params,
         )
@@ -556,7 +557,7 @@ def __call__(  # type: ignore
             prompts,
             media,
             generation_params,
-            self.logits_processor,
+            copy(self.logits_processor),
             self.sampling_params,
             **model_specific_params,
         )
@@ -581,7 +582,7 @@ def stream(  # type: ignore
             prompts,
             media,
             generation_params,
-            self.logits_processor,
+            copy(self.logits_processor),
             self.sampling_params,
             **model_specific_params,
         )
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index ff247b0f4..a96ce8673 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -253,6 +253,16 @@ def test_generate_choice(request, model_fixture, sample_choices):
     assert res in sample_choices
 
 
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_choice_twice(request, model_fixture, sample_choices):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.choice(model, sample_choices)
+    res = generator(**get_inputs(model_fixture))
+    assert res in sample_choices
+    res = generator(**get_inputs(model_fixture))
+    assert res in sample_choices
+
+
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
 def test_generate_format_bool(request, model_fixture):
     model = request.getfixturevalue(model_fixture)

From 538e7147bf5085a4dc94bcfe854e3243ebca14f5 Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Mon, 23 Sep 2024 16:09:12 -0500
Subject: [PATCH 078/505] Fix benchmark workflow triggers

Closes #1164
---
 .github/workflows/asv_benchmark_pr.yml | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/asv_benchmark_pr.yml b/.github/workflows/asv_benchmark_pr.yml
index 90fb47423..d7546f854 100644
--- a/.github/workflows/asv_benchmark_pr.yml
+++ b/.github/workflows/asv_benchmark_pr.yml
@@ -1,18 +1,29 @@
 name: Benchmark PR
 
 on:
+  push:
   pull_request:
-    branches: [main]
+    types: [synchronize, labeled]
   workflow_dispatch:
 env:
   PYTHON_VERSION: "3.10"
   WORKING_DIR: ${{ github.workspace }}/benchmarks
   BENCHMARKS_OUTPUT: ${{ github.workspace }}/benchmarks_output
 
+permissions:
+  contents: read
+
+# Cancels all previous workflow runs for pull requests that have not completed.
+concurrency:
+  # The concurrency group contains the workflow name and the branch name for pull requests
+  # or the commit hash for any other events.
+  group: ${{ github.workflow }}-${{ github.event_name == 'pull_request' && github.head_ref || github.sha }}
+  cancel-in-progress: true
+
 jobs:
   benchmark-pr:
     runs-on: ubuntu-latest
-    if: contains(github.event.pull_request.labels.*.name, 'run_benchmarks') || github.event_name == 'workflow_dispatch' || github.event_name == 'workflow_run'
+    if: ${{ contains(github.event.pull_request.labels.*.name, 'run-benchmarks') || github.ref == 'refs/heads/main' }}
 
     defaults:
       run:

From 6035e86ac8089d4f8aeab07ea116093a2ed0e03e Mon Sep 17 00:00:00 2001
From: Jan Trienes <jan.trienes@uni-marburg.de>
Date: Thu, 19 Sep 2024 13:55:08 +0200
Subject: [PATCH 079/505] Reuse jinja environment for a prompt

---
 outlines/prompts.py | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/outlines/prompts.py b/outlines/prompts.py
index 01e900c96..a7824451a 100644
--- a/outlines/prompts.py
+++ b/outlines/prompts.py
@@ -24,6 +24,7 @@ class Prompt:
 
     def __post_init__(self):
         self.parameters: List[str] = list(self.signature.parameters.keys())
+        self.jinja_environment = create_jinja_template(self.template)
 
     def __call__(self, *args, **kwargs) -> str:
         """Render and return the template.
@@ -35,7 +36,7 @@ def __call__(self, *args, **kwargs) -> str:
         """
         bound_arguments = self.signature.bind(*args, **kwargs)
         bound_arguments.apply_defaults()
-        return render(self.template, **bound_arguments.arguments)
+        return self.jinja_environment.render(**bound_arguments.arguments)
 
     def __str__(self):
         return self.template
@@ -182,6 +183,11 @@ def render(template: str, **values: Optional[Dict[str, Any]]) -> str:
     A string that contains the rendered template.
 
     """
+    jinja_template = create_jinja_template(template)
+    return jinja_template.render(**values)
+
+
+def create_jinja_template(template: str):
     # Dedent, and remove extra linebreak
     cleaned_template = inspect.cleandoc(template)
 
@@ -210,8 +216,7 @@ def render(template: str, **values: Optional[Dict[str, Any]]) -> str:
     env.filters["args"] = get_fn_args
 
     jinja_template = env.from_string(cleaned_template)
-
-    return jinja_template.render(**values)
+    return jinja_template
 
 
 def get_fn_name(fn: Callable):

From 30531e58f8e15fd4f47bed5d5718209cbad4b3e0 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Fri, 27 Sep 2024 19:34:00 -0400
Subject: [PATCH 080/505] Use more performant fsm backend

---
 .github/workflows/tests.yml                   |    2 +-
 benchmarks/bench_cfg_guide.py                 |    5 -
 benchmarks/bench_json_schema.py               |    3 +-
 benchmarks/bench_numba_compile.py             |   34 -
 benchmarks/bench_regex_guide.py               |    4 +-
 benchmarks/common.py                          |    6 -
 outlines/fsm/fsm.py                           |   47 -
 outlines/fsm/guide.py                         |  247 +---
 outlines/fsm/parsing.py                       |    3 +-
 outlines/fsm/regex.py                         | 1003 -----------------
 pyproject.toml                                |    7 +-
 tests/fsm/test_fsm.py                         |   92 --
 tests/fsm/test_regex.py                       |  742 ------------
 .../generate/test_integration_transformers.py |    2 +-
 14 files changed, 24 insertions(+), 2173 deletions(-)
 delete mode 100644 benchmarks/bench_numba_compile.py
 delete mode 100644 outlines/fsm/fsm.py
 delete mode 100644 outlines/fsm/regex.py
 delete mode 100644 tests/fsm/test_fsm.py
 delete mode 100644 tests/fsm/test_regex.py

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 53d20b70b..36e6f8526 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -22,7 +22,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: ["3.8", "3.10"]
+        python-version: ["3.10"]
     steps:
     - uses: actions/checkout@v3
     - name: Set up Python ${{ matrix.python-version }}
diff --git a/benchmarks/bench_cfg_guide.py b/benchmarks/bench_cfg_guide.py
index 14dc31c73..477104435 100644
--- a/benchmarks/bench_cfg_guide.py
+++ b/benchmarks/bench_cfg_guide.py
@@ -7,8 +7,6 @@
 from outlines.fsm.guide import CFGGuide
 from outlines.models.transformers import TransformerTokenizer
 
-from .common import ensure_numba_compiled
-
 random.seed(42)
 
 
@@ -30,9 +28,6 @@ class CFGGuideBenchmark:
 
     def setup(self, grammar_name):
         self.tokenizer = get_tiny_tokenizer()
-        ensure_numba_compiled(
-            self.tokenizer
-        )  # numba not currently used, but will be in the future
         self.prebuilt_cfg_guide = CFGGuide(
             benched_grammars[grammar_name], self.tokenizer
         )
diff --git a/benchmarks/bench_json_schema.py b/benchmarks/bench_json_schema.py
index 8d1ceeb24..8990b015c 100644
--- a/benchmarks/bench_json_schema.py
+++ b/benchmarks/bench_json_schema.py
@@ -2,7 +2,7 @@
 from outlines.fsm.guide import RegexGuide
 from outlines.fsm.json_schema import build_regex_from_schema
 
-from .common import ensure_numba_compiled, setup_tokenizer  # noqa: E402
+from .common import setup_tokenizer  # noqa: E402
 
 simple_schema = """{
         "$defs": {
@@ -69,7 +69,6 @@ class JsonSchemaBenchmark:
     def setup(self, schema_name):
         self.tokenizer = setup_tokenizer()
         self.schema = schemas[schema_name]
-        ensure_numba_compiled(self.tokenizer)
 
     @cache_disabled()
     def time_json_schema_to_regex(self, schema_name):
diff --git a/benchmarks/bench_numba_compile.py b/benchmarks/bench_numba_compile.py
deleted file mode 100644
index 2713707e5..000000000
--- a/benchmarks/bench_numba_compile.py
+++ /dev/null
@@ -1,34 +0,0 @@
-import importlib
-
-import interegular
-import numba
-
-from outlines.caching import cache_disabled
-from outlines.fsm import regex
-
-from .common import setup_tokenizer
-
-
-class NumbaCompileBenchmark:
-    def setup(self):
-        self.tokenizer = setup_tokenizer()
-        self.regex = regex
-        original_njit = numba.njit
-
-        def mock_njit(*args, **kwargs):
-            kwargs["cache"] = False
-            return original_njit(*args, **kwargs)
-
-        self.original_njit = original_njit
-        numba.njit = mock_njit
-        importlib.reload(self.regex)
-        self.regex_pattern, _ = self.regex.make_deterministic_fsm(
-            interegular.parse_pattern("a").to_fsm().reduce()
-        )
-
-    def teardown(self):
-        numba.njit = self.original_njit
-
-    @cache_disabled()
-    def time_compile_numba(self):
-        self.regex.create_fsm_index_tokenizer(self.regex_pattern, self.tokenizer)
diff --git a/benchmarks/bench_regex_guide.py b/benchmarks/bench_regex_guide.py
index 099f94df2..7aaef6bac 100644
--- a/benchmarks/bench_regex_guide.py
+++ b/benchmarks/bench_regex_guide.py
@@ -1,7 +1,7 @@
 from outlines.caching import cache_disabled
 from outlines.fsm.guide import RegexGuide
 
-from .common import ensure_numba_compiled, setup_tokenizer
+from .common import setup_tokenizer
 
 regex_samples = {
     "email": r"[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?",
@@ -21,7 +21,6 @@ class RegexGuideBenchmark:
 
     def setup(self, pattern_name):
         self.tokenizer = setup_tokenizer()
-        ensure_numba_compiled(self.tokenizer)
         self.pattern = regex_samples[pattern_name]
 
     @cache_disabled()
@@ -34,7 +33,6 @@ class MemoryRegexGuideBenchmark:
 
     def setup(self, pattern_name):
         self.tokenizer = setup_tokenizer()
-        ensure_numba_compiled(self.tokenizer)
         self.pattern = regex_samples[pattern_name]
 
     @cache_disabled()
diff --git a/benchmarks/common.py b/benchmarks/common.py
index 7d999ea9b..e920888f5 100644
--- a/benchmarks/common.py
+++ b/benchmarks/common.py
@@ -1,14 +1,8 @@
 from transformers import AutoTokenizer
 
-from outlines.fsm.guide import RegexGuide
 from outlines.models.transformers import TransformerTokenizer
 
 
 def setup_tokenizer():
     tokenizer = AutoTokenizer.from_pretrained("gpt2")
     return TransformerTokenizer(tokenizer)
-
-
-def ensure_numba_compiled(tokenizer):
-    RegexGuide("a", tokenizer)
-    return True
diff --git a/outlines/fsm/fsm.py b/outlines/fsm/fsm.py
deleted file mode 100644
index bfcf55c03..000000000
--- a/outlines/fsm/fsm.py
+++ /dev/null
@@ -1,47 +0,0 @@
-import warnings
-from typing import TYPE_CHECKING, Iterable, NewType, Optional
-
-from outlines.fsm.guide import RegexGuide, StopAtEOSGuide
-
-if TYPE_CHECKING:
-    from outlines.models.tokenizer import Tokenizer
-
-FSMState = NewType("FSMState", int)
-
-
-class StopAtEosFSM(StopAtEOSGuide):
-    """FSM to generate text until EOS has been generated."""
-
-    def __init__(self, tokenizer: "Tokenizer"):
-        warnings.warn(
-            UserWarning(
-                "The `StopAtTokenFSM` interface is deprecated and will be removed on 2024-06-01. Please use `StopAtEOSGuide` instead."
-            )
-        )
-        super().__init__(tokenizer)
-
-    def allowed_token_ids(self, state: FSMState) -> Optional[Iterable[int]]:
-        next_instruction = self.get_next_instruction(state)
-        return next_instruction.tokens
-
-    def next_state(self, state: FSMState, token_id: int) -> FSMState:
-        return FSMState(self.get_next_state(state, token_id))
-
-
-class RegexFSM(RegexGuide):
-    """FSM to generate text that is in the language of a regular expression."""
-
-    def __init__(self, regex_string: str, tokenizer):
-        warnings.warn(
-            UserWarning(
-                "The `RegexFSM` interface is deprecated and will be removed on 2024-06-01. Please use `RegexGuide` instead."
-            )
-        )
-        super().__init__(regex_string, tokenizer)
-
-    def allowed_token_ids(self, state: FSMState) -> Optional[Iterable[int]]:
-        next_instruction = self.get_next_instruction(state)
-        return next_instruction.tokens
-
-    def next_state(self, state: FSMState, token_id: int) -> FSMState:
-        return FSMState(self.get_next_state(state, token_id))
diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index b7b121fe6..a3a2c7369 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -1,72 +1,31 @@
 import collections
 import copy
 import warnings
-from dataclasses import dataclass
-from typing import (
-    TYPE_CHECKING,
-    Any,
-    Callable,
-    Dict,
-    Generator,
-    List,
-    Optional,
-    Protocol,
-    Set,
-    Tuple,
-    Union,
-)
+from typing import TYPE_CHECKING, Any, Generator, Union
 
-import interegular
 import torch
 from lark.indenter import DedentError
 from lark.lexer import UnexpectedCharacters, UnexpectedToken
+from outlines_core.fsm.guide import Generate
+from outlines_core.fsm.guide import Guide as CoreGuide
+from outlines_core.fsm.guide import RegexGuide as CoreRegexGuide
+from outlines_core.fsm.guide import Write
+from outlines_core.fsm.guide import (
+    create_states_mapping as uncached_create_states_mapping,
+)
 
 from outlines import grammars
 from outlines.caching import cache
 from outlines.fsm.parsing import PartialLark, PartialParserState
-from outlines.fsm.regex import (
-    create_fsm_index_tokenizer,
-    make_byte_level_fsm,
-    make_deterministic_fsm,
-)
 
 if TYPE_CHECKING:
     from outlines.models.tokenizer import Tokenizer
 
 
-@dataclass(frozen=True)
-class Write:
-    """Write instruction.
-
-    Attributes
-    ----------
-    tokens
-        The sequence of tokens to be added to the current sequence by the
-        generation process.
-
-    """
-
-    tokens: List[int]
-
-
-@dataclass(frozen=True)
-class Generate:
-    """Generate instruction
-
-    Attributes
-    ----------
-    tokens
-        The tokens that lead to a valid completion if generated.  A value
-        of ``None`` indicates that all tokens are allowed.
-    """
-
-    tokens: Optional[List[int]]
-
-
 Instruction = Union[Write, Generate]
 
 
-class Guide(Protocol):
+class Guide(CoreGuide):
     """Base definition of a generation guide.
 
     A generation guide defines the behavior of a finite-state machine that guides
@@ -78,18 +37,6 @@ class Guide(Protocol):
 
     initial_state: Any
 
-    def get_next_instruction(self, state: Any) -> Instruction:
-        ...
-
-    def get_next_state(self, state: Any, token_id: int) -> Any:
-        ...
-
-    def is_final_state(self, state: Any) -> bool:
-        ...
-
-    def copy(self) -> "Guide":
-        ...
-
 
 class StopAtEOSGuide(Guide):
     """Guide to generate tokens until the EOS token has been generated."""
@@ -127,64 +74,15 @@ def copy(self):
 
 
 @cache()
-def create_states_mapping(
-    regex_string: str,
-    tokenizer: "Tokenizer",
-    regex_parser: Callable[[str], interegular.Pattern] = interegular.parse_pattern,
-    frozen_tokens: List[str] = [],
-) -> Tuple[Dict[int, Dict[int, int]], Set[int], set]:
-    """Create the variables related to the mapping between states and tokens
-    The parameters of the function are used for caching purpose.
-
-    Parameters
-    ----------
-    regex_string: (`str`):
-        The regular expression string to generate a states mapping for.
-    tokenizer: (`Tokenizer`):
-        The model's tokenizer.
-    regex_parser: (`Callable[[str], interegular.Pattern]`, *optional*):
-        A function that parses a regex string into an `interegular` Pattern object.
-    frozen_tokens: (`List[str]`, *optional*):
-        A list of tokens that should be kept as-is when expanding the token-level FSM
-        into a byte-level FSM. Defaults to an empty list.
-
-    Returns
-    -------
-    states_to_token_maps: (`Dict[int, Dict[int, int]]`):
-        A mapping from states to a mapping from token ids originating from that state
-        to the next state to transition to given that token. The structure is as follows:
-        (origin_state -> (token_id -> next_state))
-    empty_token_ids: (`Set[int]`):
-        A set of token ids that correspond to empty strings.
-    final_states: (`set`):
-        A set of final states in the FSM.
-    """
-    regex_pattern = regex_parser(regex_string)
-    byte_fsm = make_byte_level_fsm(
-        regex_pattern.to_fsm().reduce(), keep_utf8=True, frozen_tokens=frozen_tokens
-    )
-    regex_fsm, _ = make_deterministic_fsm(byte_fsm)
-    states_to_token_maps, empty_token_ids = create_fsm_index_tokenizer(
-        regex_fsm, tokenizer, frozen_tokens=frozen_tokens
-    )
-
-    # We make sure that it is possible to generate strings in the language
-    # of the regular expression with the tokens present in the model's
-    # vocabulary.
-    if not any(
-        regex_fsm.finals.intersection(v.values()) for v in states_to_token_maps.values()
-    ):
-        raise ValueError(
-            "The vocabulary does not allow us to build a sequence that matches the input regex"
-        )
+def create_states_mapping(regex_string, tokenizer):
+    return uncached_create_states_mapping(regex_string, tokenizer)
 
-    return states_to_token_maps, empty_token_ids, regex_fsm.finals
 
-
-class RegexGuide(Guide):
-    """Guide to generate text in the language of a regular expression."""
-
-    initial_state = 0
+class RegexGuide(CoreRegexGuide):
+    """
+    Guide to generate text in the language of a regular expression.
+    CoreRegexGuide with outlines cache
+    """
 
     def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
         (
@@ -196,119 +94,6 @@ def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
         self.final_states = fsm_finals | {-1}
         self._cache_state_to_token_tensor()
 
-    def get_next_instruction(self, state: int) -> Instruction:
-        """Return the next instruction for guided generation.
-
-        The initialization of the guide builds an index which maps FSM states to a
-        map from authorized tokens to the state in which the guide needs to move
-        if said token is generated. Therefore the authorized tokens at the
-        current state are the keys of the map returned by the value of the index
-        for current state.
-
-        If the current state is not contained in the end this means that we are
-        in a final state of the guide. We only authorize EOS tokens in the final
-        state.
-
-        Parameters
-        ----------
-        state
-            The current state of the guide.
-
-        Returns
-        -------
-        A `Generate` instance that contains the model and the allowed token ids.
-
-        """
-        next_tokens_mask = self.states_to_token_mask.get(state)
-        if next_tokens_mask is None:
-            return Write(torch.tensor([self.eos_token_id]))
-
-        return Generate(next_tokens_mask)
-
-    def get_next_state(self, state: int, token_id: int) -> int:
-        """Update the state of the guide.
-
-        We use the index to determine to which state the guide should transition
-        given the token that was just generated.
-
-        Parameters
-        ----------
-        state
-            The current state of the guide.
-        token_id
-            The id of the token that was just generated.
-
-        Returns
-        -------
-        The new state of the guide.
-
-        """
-        if token_id == self.eos_token_id or state not in self.states_to_token_maps:
-            return -1
-
-        last_token_to_end_state = self.states_to_token_maps[state]
-        next_state = last_token_to_end_state.get(token_id)
-        if next_state is None:
-            next_state = -1
-
-        return next_state
-
-    @classmethod
-    def from_interegular_fsm(
-        cls, interegular_fsm: interegular.fsm.FSM, tokenizer: "Tokenizer"
-    ):
-        from_interegular_instance = cls.__new__(cls)
-
-        def create_states_mapping_from_interegular_fsm(
-            fsm: interegular.fsm.FSM,
-        ) -> Tuple[dict, set]:
-            """Create the variables related to the mapping between states and tokens
-            The parameters of the function are used for caching purpose
-            """
-            byte_fsm = make_byte_level_fsm(fsm.reduce(), keep_utf8=True)
-            regex_fsm, _ = make_deterministic_fsm(byte_fsm)
-            states_to_token_maps, empty_token_ids = create_fsm_index_tokenizer(
-                regex_fsm, tokenizer
-            )
-
-            # We make sure that it is possible to generate strings in the language
-            # of the regular expression with the tokens present in the model's
-            # vocabulary.
-            if not any(
-                regex_fsm.finals.intersection(v.values())
-                for v in states_to_token_maps.values()
-            ):
-                raise ValueError(
-                    "The vocabulary does not allow us to build a sequence that matches the input regex"
-                )
-
-            return states_to_token_maps, empty_token_ids
-
-        (
-            from_interegular_instance.states_to_token_maps,
-            from_interegular_instance.empty_token_ids,
-        ) = create_states_mapping_from_interegular_fsm(interegular_fsm)
-        from_interegular_instance.eos_token_id = tokenizer.eos_token_id
-        from_interegular_instance._cache_state_to_token_tensor()
-        return from_interegular_instance
-
-    def _cache_state_to_token_tensor(self):
-        """
-        cache state -> token int tensor
-        this increases performance of mask construction substantially
-        """
-        self.states_to_token_mask = {
-            state: torch.tensor(list(next_tokens_to_end_states.keys()))
-            for state, next_tokens_to_end_states in self.states_to_token_maps.items()
-        }
-
-    def is_final_state(self, state: int) -> bool:
-        """Determine whether the current state of the guide is a final state."""
-        return state in self.final_states
-
-    def copy(self):
-        return self
-
 
 CFGState = collections.namedtuple("CFGState", ["parser_state", "prev_token"])
 
diff --git a/outlines/fsm/parsing.py b/outlines/fsm/parsing.py
index f780fb46e..92d3cc166 100644
--- a/outlines/fsm/parsing.py
+++ b/outlines/fsm/parsing.py
@@ -34,8 +34,7 @@
 )
 from lark.parsers.lalr_interactive_parser import InteractiveParser
 from lark.parsers.lalr_parser import LALR_Parser, ParseConf, ParserState, _Parser
-
-from outlines.fsm.regex import (
+from outlines_core.fsm.regex import (
     fsm_union,
     get_sub_fsms_from_seq,
     get_token_transition_keys,
diff --git a/outlines/fsm/regex.py b/outlines/fsm/regex.py
deleted file mode 100644
index 6b105a7b9..000000000
--- a/outlines/fsm/regex.py
+++ /dev/null
@@ -1,1003 +0,0 @@
-import re
-from collections import namedtuple
-from functools import lru_cache
-from typing import (
-    TYPE_CHECKING,
-    Dict,
-    FrozenSet,
-    Generator,
-    List,
-    Sequence,
-    Set,
-    Tuple,
-    Union,
-    cast,
-)
-
-import numba
-import numpy as np
-from interegular.fsm import (
-    FSM,
-    Alphabet,
-    OblivionError,
-    State,
-    TransitionKey,
-    _AnythingElseCls,
-    anything_else,
-)
-from numba.typed.typedobjectutils import _nonoptional
-from tqdm import tqdm
-
-if TYPE_CHECKING:
-    from outlines.models.tokenizer import Tokenizer
-
-
-class BetterAlphabet(Alphabet):
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-        assert anything_else in self._symbol_mapping
-        self.anything_value = self._symbol_mapping[anything_else]
-
-    def __getitem__(self, item):
-        return self._symbol_mapping.get(item, self.anything_value)
-
-    def copy(self):
-        return BetterAlphabet(self._symbol_mapping.copy())
-
-
-class BetterFSM(FSM):
-    flat_transition_map: Dict[Tuple[int, int], int]
-    trans_key_to_states: Dict[int, List[int]]
-
-    def __init__(self, *args, **kwargs):
-        super().__init__(*args, **kwargs)
-
-        if not isinstance(self.alphabet, BetterAlphabet):
-            self.__dict__["alphabet"] = BetterAlphabet(self.alphabet._symbol_mapping)
-
-        flat_transition_map = {}
-        trans_key_to_states = {}
-        for from_state, trans_map in self.map.items():
-            for trans_key, to_state in trans_map.items():
-                flat_transition_map[(from_state, trans_key)] = to_state
-                trans_key_to_states.setdefault(trans_key, set()).add(from_state)
-
-        self.__dict__["trans_key_to_states"] = trans_key_to_states
-        self.__dict__["flat_transition_map"] = flat_transition_map
-        self.__dict__["_fsm_info"] = None
-
-    def copy(self):
-        return BetterFSM(
-            alphabet=self.alphabet.copy(),
-            states=self.states.copy(),
-            initial=self.initial,
-            finals=self.finals.copy(),
-            map=self.map.copy(),
-            __no_validation__=True,
-        )
-
-    @property
-    def fsm_info(self):
-        if self._fsm_info is None:
-            flat_transition_map_items = np.fromiter(
-                ((a[0], a[1], b) for a, b in self.flat_transition_map.items()),
-                dtype=np.dtype("int64, int64, int64"),
-            )
-            trans_key_to_states_items = np.fromiter(
-                ((k, z) for k, v in self.trans_key_to_states.items() for z in v),
-                dtype=np.dtype("int64, int64"),
-            )
-            alphabet_symbol_mapping_items = [
-                (k, v)
-                for k, v in self.alphabet._symbol_mapping.items()
-                if k != anything_else
-            ]
-            nb_finals = np.fromiter(self.finals, dtype=np.dtype("int64"))
-            self.__dict__["_fsm_info"] = create_fsm_info(
-                self.initial,
-                nb_finals,
-                flat_transition_map_items,
-                trans_key_to_states_items,
-                self.alphabet.anything_value,
-                alphabet_symbol_mapping_items,
-            )
-
-        return self._fsm_info
-
-
-nb_int_list_type = numba.types.ListType(numba.int64)
-nb_int_pair_type = numba.types.UniTuple(numba.int64, 2)
-nb_unicode_type = numba.types.unicode_type
-
-
-@numba.njit(cache=True)
-def create_fsm_info(
-    py_initial,
-    py_finals,
-    flat_transition_map_items,
-    trans_key_to_states_items,
-    py_anything_value,
-    alphabet_symbol_mapping_items,
-):
-    trans_key_to_states = numba.typed.Dict.empty(numba.int64, nb_int_list_type)
-    for trans_key_and_state in trans_key_to_states_items:
-        trans_key_to_states.setdefault(
-            trans_key_and_state[0], numba.typed.List.empty_list(numba.int64)
-        ).append(trans_key_and_state[1])
-
-    flat_transition_map = numba.typed.Dict.empty(nb_int_pair_type, numba.int64)
-    for trans_key_and_state in flat_transition_map_items:
-        flat_transition_map[
-            (trans_key_and_state[0], trans_key_and_state[1])
-        ] = trans_key_and_state[2]
-
-    # use 2-char strings so that we can represent incomplete utf-8 sequences
-    # as 2-hex-digit pairs
-    alphabet_symbol_map = numba.typed.Dict.empty(nb_unicode_type, numba.int64)
-    for symbol_and_trans_key in alphabet_symbol_mapping_items:
-        alphabet_symbol_map[symbol_and_trans_key[0]] = symbol_and_trans_key[1]
-
-    initial = numba.int64(py_initial)
-
-    finals = set()
-    for final in py_finals:
-        finals.add(final)
-
-    anything_value = numba.int64(py_anything_value)
-
-    return FSMInfo(
-        initial,
-        finals,
-        flat_transition_map,
-        trans_key_to_states,
-        anything_value,
-        alphabet_symbol_map,
-    )
-
-
-FSMInfo = namedtuple(
-    "FSMInfo",
-    [
-        "initial",
-        "finals",
-        "transitions",
-        "trans_key_to_states",
-        "alphabet_anything_value",
-        "alphabet_symbol_mapping",
-    ],
-)
-
-
-TransitionTrie = Dict[TransitionKey, "Union[TransitionTrie, State, None]"]
-
-
-def add_to_transition_trie(
-    trie: TransitionTrie,
-    key_seq: Sequence[TransitionKey],
-    value: Union[State, None],
-):
-    for key in key_seq[:-1]:
-        trie = cast(TransitionTrie, trie.setdefault(key, {}))
-        assert isinstance(trie, dict), "key sequence of incompatible length"
-    trie[key_seq[-1]] = value
-
-
-# merge default_trie into the trie, only updating entries not present in the trie
-def transition_trie_setdefault(
-    trie: TransitionTrie,
-    default_trie: TransitionTrie,
-):
-    for key, default_value in default_trie.items():
-        dest_value = trie.get(key)
-        if isinstance(dest_value, dict) and isinstance(default_value, dict):
-            transition_trie_setdefault(dest_value, default_value)
-        elif key not in trie:
-            trie[key] = default_value
-
-
-def byte_symbol(byte: int) -> str:
-    return f"\x00{byte:02X}" if byte >= 0x80 else chr(byte)
-
-
-def make_byte_level_fsm(
-    fsm: FSM, keep_utf8: bool = False, frozen_tokens: List[str] = []
-) -> FSM:
-    """Convert an FSM to a byte-level FSM, expanding multi-byte characters as
-    sequences of single-byte transitions.
-
-    Parameters
-    ----------
-    fsm: (`interegular.FSM`):
-        The token-level FSM to convert to a byte-level FSM.
-    keep_utf8: (`bool`, *optional*):
-        If set to True, the original utf-8 characters are kept as-is. Defaults to
-        False. NOTE: we're representing bytes as strings to keep it type-compatible.
-    frozen_tokens: (`List[str]`, *optional*):
-        A list of tokens that should be kept as-is in the byte-level FSM. That is,
-        these tokens will not be expanded into byte-level transitions. Defaults to
-        an empty list.
-
-    Returns
-    -------
-    `interegular.FSM`: A byte-level FSM.
-    """
-
-    anything_else_key = fsm.alphabet[anything_else]
-    symbol_mapping: Dict[Union[str, _AnythingElseCls], TransitionKey] = {}
-    map: Dict[State, Dict[TransitionKey, State]] = {}
-    states: List[State] = list(fsm.states)
-
-    # identify all multi-byte characters in the alphabet and build a mapping
-    # from the original transition keys to sequences of new keys for each byte
-    key_to_key_seqs: Dict[TransitionKey, Set[Tuple[TransitionKey, ...]]] = {}
-    all_key_seqs: Set[Tuple[TransitionKey, ...]] = set()
-    all_bytes: Set[int] = set()
-    max_key = max(fsm.alphabet.values())
-    for symbol, transition_key in fsm.alphabet.items():
-        assert symbol == anything_else or symbol in frozen_tokens or len(symbol) == 1
-        if symbol == anything_else or symbol in frozen_tokens or ord(symbol) < 0x80:
-            symbol_mapping[symbol] = transition_key
-        else:
-            if keep_utf8:
-                symbol_mapping[symbol] = transition_key
-            key_list: List[TransitionKey] = []
-            for byte in symbol.encode("utf-8"):
-                symbol = byte_symbol(byte)
-                if symbol not in symbol_mapping:
-                    symbol_mapping[symbol] = max_key = TransitionKey(max_key + 1)
-                    all_bytes.add(byte)
-                key_list.append(symbol_mapping[symbol])
-            key_seq = tuple(key_list)
-            key_to_key_seqs.setdefault(transition_key, set()).add(key_seq)
-            all_key_seqs.add(key_seq)
-
-    # add all remaining multi-byte utf-8 bytes to the alphabet
-    # (this is required to represent `anything_else`)
-    utf8_ranges = {
-        1: (0x80, 0xC0),  # continuation bytes
-        2: (0xC0, 0xE0),  # 2-byte sequences
-        3: (0xE0, 0xF0),  # 3-byte sequences
-        4: (0xF0, 0xF8),  # 4-byte sequences
-    }
-    utf8_all_keys: Dict[int, Set[TransitionKey]] = {
-        n: set() for n in utf8_ranges.keys()
-    }
-    for n, (start, end) in utf8_ranges.items():
-        range_key = max_key = TransitionKey(max_key + 1)
-        for byte in range(start, end):
-            byte_key = symbol_mapping.setdefault(byte_symbol(byte), range_key)
-            utf8_all_keys[n].add(byte_key)
-
-    # cache of intermediate transition states by transitions from that state
-    state_cache: Dict[FrozenSet[Tuple[TransitionKey, State]], State] = {}
-
-    # helper function to create multi-step transitions between states
-    max_state = max(fsm.states)
-
-    def create_seq_transitions(
-        seq_transitions_trie: TransitionTrie,
-    ) -> Dict[TransitionKey, State]:
-        nonlocal max_state
-        result: Dict[TransitionKey, State] = {}
-
-        for next_key, next_trie in seq_transitions_trie.items():
-            if isinstance(next_trie, dict):
-                next_transitions = create_seq_transitions(next_trie)
-                if not next_transitions:
-                    continue
-                cache_key = frozenset(next_transitions.items())
-                next_state = state_cache.get(cache_key)
-                if next_state is None:
-                    next_state = max_state = State(max_state + 1)
-                    map[next_state] = next_transitions
-                    state_cache[cache_key] = next_state
-                    states.append(next_state)
-                result[next_key] = next_state
-            elif next_trie is not None:
-                result[next_key] = next_trie
-
-        return result
-
-    # create new states and transitions
-    for state, transitions in fsm.map.items():
-        seq_transitions_trie: TransitionTrie = {}
-        state_map: Dict[TransitionKey, State] = {}
-
-        for transition_key, to_state in transitions.items():
-            if transition_key in key_to_key_seqs:
-                if keep_utf8:
-                    state_map[transition_key] = to_state
-                for key_seq in key_to_key_seqs[transition_key]:
-                    add_to_transition_trie(seq_transitions_trie, key_seq, to_state)
-            else:  # keep single-byte transitions as is
-                state_map[transition_key] = to_state
-
-        # handle multi-byte anything_else sequences
-        if anything_else_key in transitions:
-            for key_seq in all_key_seqs:
-                add_to_transition_trie(seq_transitions_trie, key_seq, None)
-
-            anything_else_trie: TransitionTrie = {}
-            cont_trie: Union[TransitionTrie, State] = transitions[anything_else_key]
-            for n in range(2, 5):
-                cont_trie = {key: cont_trie for key in utf8_all_keys[1]}
-                for key in utf8_all_keys[n]:
-                    anything_else_trie[key] = cont_trie
-
-            transition_trie_setdefault(seq_transitions_trie, anything_else_trie)
-
-        # create new states and transitions
-        next_transitions = create_seq_transitions(seq_transitions_trie)
-        state_map.update(next_transitions)
-        map[state] = state_map
-
-    return FSM(
-        alphabet=Alphabet(symbol_mapping),
-        states=states,
-        initial=fsm.initial,
-        finals=fsm.finals,
-        map=map,
-    )
-
-
-def make_byte_level_better_fsm(fsm: BetterFSM, keep_utf8=False) -> BetterFSM:
-    new_fsm = make_byte_level_fsm(fsm, keep_utf8)
-    return BetterFSM(
-        alphabet=BetterAlphabet(new_fsm.alphabet._symbol_mapping),
-        states=new_fsm.states,
-        initial=new_fsm.initial,
-        finals=new_fsm.finals,
-        map=new_fsm.map,
-    )
-
-
-def make_deterministic_fsm(fsm: FSM) -> Tuple[BetterFSM, Dict[int, int]]:
-    """Construct an equivalent FSM with deterministic state labels."""
-    old_to_new_trans_keys = {
-        trans_key: i
-        for i, (trans_key, _) in enumerate(
-            sorted(fsm.alphabet.by_transition.items(), key=lambda x: sorted(x[1]))
-        )
-    }
-
-    new_symbol_mapping = {
-        symbol: old_to_new_trans_keys[trans_key]
-        for symbol, trans_key in fsm.alphabet._symbol_mapping.items()
-    }
-
-    new_alphabet = BetterAlphabet(new_symbol_mapping)
-
-    new_map = {
-        from_state: {
-            old_to_new_trans_keys[trans_key]: to_state
-            for trans_key, to_state in trans_map.items()
-        }
-        for from_state, trans_map in fsm.map.items()
-    }
-
-    old_to_new_states = {}
-    old_to_new_states[fsm.initial] = 0
-
-    i = 0
-    seen = {fsm.initial}
-    old_state_queue = [fsm.initial]
-    while old_state_queue:
-        old_state = old_state_queue.pop(-1)
-        transitions = new_map[old_state]
-        sorted_transitions = sorted(transitions.items(), key=lambda v: v[0])
-        for _, old_state in sorted_transitions:
-            if old_state not in seen:
-                old_state_queue.append(old_state)
-                seen.add(old_state)
-            if old_state not in old_to_new_states:
-                i += 1
-                old_to_new_states[old_state] = i
-
-    new_map = dict(
-        sorted(
-            (
-                (
-                    old_to_new_states[from_state],
-                    dict(
-                        sorted(
-                            (
-                                (trans_key, old_to_new_states[to_state])
-                                for trans_key, to_state in trans_map.items()
-                            ),
-                            key=lambda v: v[0],
-                        )
-                    ),
-                )
-                for from_state, trans_map in new_map.items()
-            ),
-            key=lambda v: v[0],
-        )
-    )
-
-    new_initial = 0
-    new_finals = frozenset(
-        sorted(old_to_new_states[old_state] for old_state in fsm.finals)
-    )
-    new_states = frozenset(sorted(new_map.keys()))
-
-    new_fsm = BetterFSM(new_alphabet, new_states, new_initial, new_finals, new_map)
-
-    return new_fsm, old_to_new_states
-
-
-@numba.njit(nogil=True, cache=True)
-def _walk_fsm(
-    fsm_transitions: Dict[Tuple[int, int], int],
-    fsm_initial: int,
-    fsm_finals: Set[int],
-    token_transition_keys: Sequence[int],
-    start_state: int,
-    full_match: bool = True,
-) -> List[int]:
-    state = start_state
-    accepted_states: List[int] = numba.typed.List.empty_list(numba.int64)
-    last_final_idx: int = numba.uint64(0)
-
-    # Iterate over token transition key sequence. The transition key
-    # sequence represents the FSM traversal rules of the tokens symbols.
-    for i, trans_key in enumerate(token_transition_keys):
-        new_state = fsm_transitions.get((state, trans_key))
-
-        if new_state is None:
-            if not full_match and last_final_idx > 0:
-                return accepted_states[:last_final_idx]
-
-            return numba.typed.List.empty_list(numba.int64)
-
-        state = new_state
-
-        if state in fsm_finals:
-            last_final_idx = numba.uint64(i + 1)
-
-        accepted_states.append(_nonoptional(state))
-
-    if full_match and last_final_idx - 1 != i:
-        return numba.typed.List.empty_list(numba.int64)
-
-    return accepted_states
-
-
-def walk_fsm(
-    fsm: BetterFSM,
-    token_transition_keys: Sequence[int],
-    start_state: int,
-    full_match: bool = True,
-) -> List[int]:
-    fsm_finals = fsm.finals
-
-    state = start_state
-    accepted_states: List[int] = []
-    last_final_idx: int = 0
-
-    fsm_transitions = fsm.flat_transition_map
-
-    # Iterate over token transition key sequence. The transition key
-    # sequence represents the FSM traversal rules of the tokens symbols.
-    for i, trans_key in enumerate(token_transition_keys):
-        new_state = fsm_transitions.get((state, trans_key))
-
-        if new_state is None:
-            if not full_match and last_final_idx > 0:
-                return accepted_states[:last_final_idx]
-
-            return []
-
-        state = new_state
-
-        if state in fsm_finals:
-            last_final_idx = i + 1
-
-        accepted_states.append(state)
-
-    if full_match and last_final_idx - 1 != i:
-        return []
-
-    return accepted_states
-
-
-def fsm_union(
-    fsms: Sequence[FSM],
-) -> Tuple[FSM, Dict[int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]]]:
-    """Construct an FSM representing the union of the FSMs in `fsms`.
-
-    This is an updated version of `interegular.fsm.FSM.union` made to return an
-    extra map of component FSMs to the sets of state transitions that
-    correspond to them in the new FSM.
-
-    """
-
-    alphabet, new_to_old = Alphabet.union(*[fsm.alphabet for fsm in fsms])
-
-    indexed_fsms = tuple(enumerate(fsms))
-
-    initial = {i: fsm.initial for (i, fsm) in indexed_fsms}
-
-    # Dedicated function accepting a "superset" and returning the next
-    # "superset" obtained by following this transition in the new FSM
-    def follow(current_state, new_transition: int):
-        next = {}
-        for i, f in indexed_fsms:
-            old_transition = new_to_old[i][new_transition]
-            if (
-                i in current_state
-                and current_state[i] in f.map
-                and old_transition in f.map[current_state[i]]
-            ):
-                next[i] = f.map[current_state[i]][old_transition]
-        if not next:
-            raise OblivionError
-        return next
-
-    states = [initial]
-    finals: Set[int] = set()
-    map: Dict[int, Dict[int, int]] = {}
-
-    # Map component FSMs to their new state-to-state transitions, finals, and a
-    # map translating component FSM states to aggregate FSM states
-    fsms_to_trans_finals: Dict[
-        int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]
-    ] = {}
-
-    i = 0
-    while i < len(states):
-        state = states[i]
-
-        # Add to the finals of the aggregate FSM whenever we hit a final in a
-        # component FSM
-        if any(state.get(j, -1) in fsm.finals for (j, fsm) in indexed_fsms):
-            finals.add(i)
-
-        # Compute the map for this state
-        map[i] = {}
-        for transition in alphabet.by_transition:
-            try:
-                next = follow(state, transition)
-            except OblivionError:
-                # Reached an oblivion state; don't list it
-                continue
-            else:
-                try:
-                    # TODO: Seems like this could--and should--be avoided
-                    j = states.index(next)
-                except ValueError:
-                    j = len(states)
-                    states.append(next)
-
-                map[i][transition] = j
-
-                for fsm_id, fsm_state in next.items():
-                    (
-                        fsm_transitions,
-                        fsm_finals,
-                        fsm_old_to_new,
-                    ) = fsms_to_trans_finals.setdefault(fsm_id, (set(), set(), {}))
-                    old_from = state[fsm_id]
-                    old_to = fsm_state
-                    fsm_old_to_new.setdefault(old_from, set()).add(i)
-                    fsm_old_to_new.setdefault(old_to, set()).add(j)
-                    fsm_transitions.add((i, j))
-                    if fsm_state in fsms[fsm_id].finals:
-                        fsm_finals.add(j)
-
-        i += 1
-
-    fsm = FSM(
-        alphabet=alphabet,
-        states=range(len(states)),
-        initial=0,
-        finals=finals,
-        map=map,
-        __no_validation__=True,
-    )
-
-    fsm, old_to_new_states = make_deterministic_fsm(fsm)
-    _fsms_to_trans_finals = {
-        fsm_id: (
-            {(old_to_new_states[s1], old_to_new_states[s2]) for s1, s2 in transitions},
-            {old_to_new_states[s] for s in finals},
-            {
-                old_state: {old_to_new_states[new_state] for new_state in new_states}
-                for old_state, new_states in old_to_new.items()
-            },
-        )
-        for fsm_id, (transitions, finals, old_to_new) in sorted(
-            fsms_to_trans_finals.items(), key=lambda x: x[0]
-        )
-    }
-
-    return (
-        fsm,
-        _fsms_to_trans_finals,
-    )
-
-
-def get_sub_fsms_from_seq(
-    state_seq: Sequence[int],
-    fsms_to_trans_finals: Dict[
-        int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]
-    ],
-) -> Generator[Tuple[int, bool, bool], None, None]:
-    """Get the indices of the sub-FSMs in `fsm` that could have matched the state sequence `state_seq`.
-
-    Parameters
-    ----------
-    state_seq
-        A state sequence.
-    fsms_to_trans_finals
-        A map from FSM indices to tuples containing sets of their state transitions
-        and sets of the final/accept states.
-
-    Returns
-    -------
-    A generator returning tuples containing each sub-FSM index (in the order
-    they were union-ed to construct `fsm`) and booleans indicating whether or
-    not there is another valid transition from the last state in the sequence
-    for the associated sub-FSM (i.e. if the FSM can continue
-    accepting/matching) and whether or not the sequence ends in a final state
-    of the sub-FSM.
-    """
-    state_seq_transitions = set(zip(state_seq[:-1], state_seq[1:]))
-    last_fsm_state = state_seq[-1]
-    yield from (
-        (
-            # The sub-FMS index
-            fsm_idx,
-            # Is there another possible transition in this sub-FSM?
-            any(last_fsm_state == from_s for (from_s, to_s) in transitions),
-            # Is this sub-FSM in a final state?
-            state_seq[-1] in finals,
-        )
-        for fsm_idx, (transitions, finals, _) in fsms_to_trans_finals.items()
-        if state_seq_transitions.issubset(transitions)
-    )
-
-
-@numba.njit(cache=True, nogil=True)
-def state_scan_tokens(
-    fsm_transitions: Dict[Tuple[int, int], int],
-    alphabet_symbol_mapping: Dict[str, int],
-    alphabet_anything_value: int,
-    fsm_initial: int,
-    fsm_finals: Set[int],
-    vocabulary: List[Tuple[str, Sequence[int]]],
-    vocabulary_transition_keys: List[Sequence[int]],
-    start_state: int,
-) -> Set[Tuple[int, int]]:
-    res = set()
-
-    for (token, token_ids), token_transition_keys in zip(
-        vocabulary, vocabulary_transition_keys
-    ):
-        state_seq = _walk_fsm(
-            fsm_transitions,
-            fsm_initial,
-            fsm_finals,
-            token_transition_keys,
-            start_state,
-            False,
-        )
-
-        if state_seq is not None and len(state_seq) < len(token_transition_keys):
-            continue
-
-        for token_id in token_ids:
-            res.add((token_id, state_seq[-1]))
-
-    return res
-
-
-@numba.njit(cache=True, nogil=True)
-def get_token_transition_keys(
-    alphabet_symbol_mapping: Dict[str, int],
-    alphabet_anything_value: int,
-    token_str: str,
-) -> Sequence[int]:
-    """
-    Get the sequence of transition keys for an individual string
-    with respect to an FSMs alphabet symbol mapping
-
-    This requires parsing the null-byte prefix rules of a byte-fsm:
-    - If two characters are prefixed by \x00, they are the grouped as a hex-byte
-    - Otherwise they are a standalone utf-8 character
-    """
-    token_transition_keys = []
-    i = 0
-    while i < len(token_str):
-        if token_str[i] == "\x00" and i != len(token_str) - 1:
-            symbol = token_str[i : i + 3]
-            i += 3
-        else:
-            symbol = token_str[i]
-            i += 1
-
-        token_transition_keys.append(
-            alphabet_symbol_mapping.get(symbol, alphabet_anything_value)
-        )
-
-    token_transition_keys_array = np.empty(len(token_transition_keys), dtype=np.int64)
-    for j in range(len(token_transition_keys)):
-        token_transition_keys_array[j] = token_transition_keys[j]
-    return token_transition_keys_array
-
-
-@numba.njit(cache=True, nogil=True)
-def get_vocabulary_transition_keys(
-    alphabet_symbol_mapping: Dict[str, int],
-    alphabet_anything_value: int,
-    vocabulary: List[Tuple[str, Sequence[int]]],
-    frozen_tokens: List[str] = numba.typed.List.empty_list(numba.types.unicode_type),
-) -> List[Sequence[int]]:
-    """
-    Calculate the sequence transition keys for each token str within a vocabulary
-
-    Parameters
-    ----------
-    alphabet_symbol_mapping: (`Dict[str, int]`):
-        A mapping from an alphabet symbol in a FSM to its corresponding transition key.
-    alphabet_anything_value: (`int`):
-        The transition key for the anything_else symbol in the FSM.
-    vocabulary: (`List[Tuple[str, Sequence[int]]]`):
-        A list of tuples, each containing a token and a list of equivalent token ids.
-    frozen_tokens: (`List[str]`, *optional*):
-        A list of tokens that are kept as-is when transforming the FSM.
-        Defaults to an empty list.
-
-    Returns
-    -------
-    `List[Sequence[int]]`:
-        A list of token transition keys for each token in the vocabulary.
-    """
-    vocab_transition_keys = numba.typed.List.empty_list(numba.int64[:])
-    for token_str, _ in vocabulary:
-        # Since these tokens are not expanded into byte-level transitions, we can
-        # simply get their transition keys directly.
-        if token_str in frozen_tokens:
-            token_transition_keys = np.array(
-                [alphabet_symbol_mapping[token_str]], dtype=np.int64
-            )
-        else:
-            token_transition_keys = get_token_transition_keys(
-                alphabet_symbol_mapping, alphabet_anything_value, token_str
-            )
-        vocab_transition_keys.append(token_transition_keys)
-
-    return vocab_transition_keys
-
-
-def create_fsm_index_end_to_end(
-    fsm_info: FSMInfo,
-    vocabulary: List[Tuple[str, Sequence[int]]],
-    frozen_tokens: List[str] = [],
-) -> Dict[int, Set[Tuple[int, int]]]:
-    """Create an FSM state-to-vocabulary map/index through end-to-end token parsing.
-
-    Parameters
-    ----------
-    fsm_info: (`interegular.FSMInfo`):
-        The FSM information object containing the FSM's alphabet, transitions, initial
-        and final states, and other relevant information.
-    vocabulary: (`List[Tuple[str, Sequence[int]]]`):
-        A list of tuples, each containing a token and a list of equivalent token ids.
-    frozen_tokens: (`List[str]`, *optional*):
-        A list of tokens that are kept as-is when transforming the FSM.
-
-    Returns
-    -------
-    `Dict[int, Set[Tuple[int, int]]]`:
-        A mapping from FSM states to sets of tuples containing token ids and the end
-        states of the FSM after parsing the token.
-    """
-
-    # TODO: Consider using a `List` of `Set`s instead; that way we can JIT this
-    # code, too.
-    states_to_token_subsets: Dict[int, Set[Tuple[int, int]]] = {}
-    seen: Set[int] = set()
-    next_states = {fsm_info.initial}
-
-    pbar = tqdm(
-        total=len(set(fsm_info.transitions.values()))
-        + 1,  # all transitions plus initial
-        desc="Compiling FSM index for all state transitions",
-    )
-
-    vocabulary_transition_keys = get_vocabulary_transition_keys(
-        fsm_info.alphabet_symbol_mapping,
-        fsm_info.alphabet_anything_value,
-        vocabulary,
-        frozen_tokens=(
-            numba.typed.List(frozen_tokens)
-            if len(frozen_tokens) > 0
-            else numba.typed.List.empty_list(numba.types.unicode_type)
-        ),
-    )
-
-    while next_states:
-        start_state = next_states.pop()
-
-        token_ids_end_states = state_scan_tokens(
-            fsm_info.transitions,
-            fsm_info.alphabet_symbol_mapping,
-            fsm_info.alphabet_anything_value,
-            fsm_info.initial,
-            fsm_info.finals,
-            vocabulary,
-            vocabulary_transition_keys,
-            start_state,
-        )
-
-        for token_id_and_end_state in token_ids_end_states:
-            states_to_token_subsets.setdefault(start_state, set()).add(
-                token_id_and_end_state
-            )
-            end_state = token_id_and_end_state[1]
-            if end_state not in seen:
-                next_states.add(end_state)
-
-        if start_state not in seen:
-            pbar.update(1)
-            seen.add(start_state)
-
-    pbar.close()
-
-    return states_to_token_subsets
-
-
-re_llama_byte_token = re.compile(r"^<0x[0-9A-F]{2}>$")
-
-# The "▁*" prefix is required to handle Gemma and GPT-SW3 tokenizers, and the "\.*"
-# suffix is required to handle the NorwAI tokenizer.
-re_replacement_seq = re.compile(r"^▁*�+\.*$")
-
-
-# Copied from transformers.models.gpt2.tokenization_gpt2.bytes_to_unicode
-@lru_cache()
-def gpt2_bytes_to_unicode():
-    """
-    Returns list of utf-8 byte and a mapping to unicode strings. We specifically avoids mapping to whitespace/control
-    characters the bpe code barfs on.
-
-    The reversible bpe codes work on unicode strings. This means you need a large # of unicode characters in your vocab
-    if you want to avoid UNKs. When you're at something like a 10B token dataset you end up needing around 5K for
-    decent coverage. This is a significant percentage of your normal, say, 32K bpe vocab. To avoid that, we want lookup
-    tables between utf-8 bytes and unicode strings.
-    """
-    bs = (
-        list(range(ord("!"), ord("~") + 1))
-        + list(range(ord("¡"), ord("¬") + 1))
-        + list(range(ord("®"), ord("ÿ") + 1))
-    )
-    cs = bs[:]
-    n = 0
-    for b in range(2**8):
-        if b not in bs:
-            bs.append(b)
-            cs.append(2**8 + n)
-            n += 1
-    cs = [chr(n) for n in cs]
-    return dict(zip(bs, cs))
-
-
-@lru_cache()
-def gpt2_unicode_to_bytes():
-    return {v: k for k, v in gpt2_bytes_to_unicode().items()}
-
-
-# TODO: Cannot cache typed collections to disk, yet.  See
-# https://github.com/numba/numba/issues/4698
-@lru_cache
-def reduced_vocabulary(
-    tokenizer: "Tokenizer",
-) -> Tuple[List[Tuple[str, Sequence[int]]], Set[int]]:
-    """Create a map from decoded vocabulary tokens to lists of equivalent token ids."""
-    empty_token_ids = set()
-    vocabulary: Dict[Union[str, Tuple[str, ...]], List[int]] = {}
-    for token, token_idx in tokenizer.vocabulary.items():
-        if token in tokenizer.special_tokens:
-            continue
-
-        token_str: Union[str, Tuple[str, ...]] = tokenizer.convert_token_to_string(
-            token
-        )
-
-        if token_str:
-            if isinstance(token, bytes):
-                # Handle BPE tokenizers where the tokens are directly stored as bytes
-                # https://github.com/QwenLM/Qwen/blob/main/tokenization_note.md#regular-tokens
-                token_str = "".join(byte_symbol(b) for b in token)
-
-            elif "\ufffd" in token_str and not re_replacement_seq.match(token):
-                # invalid utf-8 sequences are replaced with � (\ufffd), but there
-                # might also be tokens specifically for �, ��, ���, etc.
-
-                if re_llama_byte_token.match(token):
-                    # llama-like tokenizers have <0xXX> tokens for all
-                    # bytes >= 0x80 and represent all incomplete utf-8
-                    # sequences using such tokens
-                    token_bytes = [int(token[3:5], 16)]
-                else:
-                    # gpt2-like tokenizers have multi-byte tokens that can
-                    # have a mix of full and incomplete utf-8 characters,
-                    # for example, b` \xf0` can be one token; these tokenizers
-                    # map each byte to a valid utf-8 character
-                    token_bytes = cast(
-                        List[int], [gpt2_unicode_to_bytes().get(c) for c in token]
-                    )
-                    if None in token_bytes:
-                        raise RuntimeError(
-                            f"Cannot convert token `{token}` ({token_idx}) to bytes: {token_str}"
-                        )
-                token_str = "".join(byte_symbol(b) for b in token_bytes)
-
-            vocabulary.setdefault(token_str, []).append(token_idx)
-        else:
-            empty_token_ids.add(numba.int64(token_idx))
-
-    vocabulary_nb = numba.typed.List.empty_list(
-        numba.types.Tuple(
-            (
-                nb_unicode_type,
-                numba.int64[:],
-            )
-        )
-    )
-    for token_str, token_ids in vocabulary.items():
-        token_ids_np = np.fromiter(token_ids, dtype=np.dtype("int64"))
-        vocabulary_nb.append((token_str, token_ids_np))
-
-    return vocabulary_nb, empty_token_ids
-
-
-def create_fsm_index_tokenizer(
-    fsm: BetterFSM, tokenizer: "Tokenizer", frozen_tokens: List[str] = []
-) -> Tuple[Dict[int, Dict[int, int]], Set[int]]:
-    """Construct an FMS index from a tokenizer.
-
-    This uses the end-to-end approach of `create_fsm_index_end_to_end`.
-
-    Parameters
-    ----------
-    fsm: (`BetterFSM`):
-        A cache-friendly FSM. Other interegular FSMs can also be used, but caching
-        may not work as expected.
-    tokenizer: (`Tokenizer`):
-        The model's tokenizer.
-    frozen_tokens: (`List[str]`, *optional*):
-        A list of tokens that should be kept as-is when expanding the token-level
-        FSM into a byte-level FSM. Defaults to an empty list.
-
-    Returns
-    -------
-    states_to_token_maps: (`Dict[int, Dict[int, int]]`):
-        A mapping from states to a mapping from token ids originating from that state
-        to the next state to transition to given that token. The structure is as follows:
-        (origin_state -> (token_id -> next_state))
-    empty_token_ids: (`Set[int]`):
-        A set of token ids that correspond to empty strings.
-
-    .. warning::
-
-        `fsm` needs to be deterministically ordered so that future caching makes sense.
-    """
-    vocabulary, empty_token_ids = reduced_vocabulary(tokenizer)
-
-    states_to_token_subsets = create_fsm_index_end_to_end(
-        fsm.fsm_info, vocabulary, frozen_tokens
-    )
-
-    # Allow transitions to EOS from all terminals FSM states that are
-    # reachable
-    # TODO: Do we really need this anymore?
-    for state in fsm.fsm_info.finals:
-        subset = states_to_token_subsets.get(state)
-        if subset is not None:
-            subset.add((tokenizer.eos_token_id, state))
-
-    # Convert to token-to-end-state maps
-    states_to_token_subsets = {k: dict(v) for k, v in states_to_token_subsets.items()}
-
-    return states_to_token_subsets, empty_token_ids
diff --git a/pyproject.toml b/pyproject.toml
index 937b0c9c7..7229afa83 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -6,7 +6,7 @@ build-backend = "setuptools.build_meta"
 name = "outlines"
 authors= [{name = "Outlines Developers"}]
 description = "Probabilistic Generative Model Programming"
-requires-python = ">=3.8"
+requires-python = ">=3.9"
 license = {text = "Apache-2.0"}
 keywords=[
     "machine learning",
@@ -32,7 +32,6 @@ dependencies = [
    "cloudpickle",
    "diskcache",
    "pydantic>=2.0",
-   "numba",
    "referencing",
    "jsonschema",
    "requests",
@@ -41,6 +40,7 @@ dependencies = [
    "typing_extensions",
    "pycountry",
    "airportsdata",
+   "outlines_core",
 ]
 dynamic = ["version"]
 
@@ -94,7 +94,6 @@ write_to = "outlines/_version.py"
 testpaths = ["tests"]
 filterwarnings = [
     "error",
-    "ignore::numba.core.errors.NumbaPendingDeprecationWarning",
     "ignore::pydantic.warnings.PydanticDeprecatedSince20",
     "ignore::FutureWarning:transformers.*",
     "ignore::FutureWarning:huggingface_hub.*",
@@ -129,7 +128,6 @@ module = [
     "lark.*",
     "interegular.*",
     "datasets.*",
-    "numba.*",
     "requests.*",
     "responses.*",
     "vllm.*",
@@ -137,6 +135,7 @@ module = [
     "fastapi.*",
     "pycountry.*",
     "airportsdata.*",
+    "outlines_core.*",
 ]
 ignore_missing_imports = true
 
diff --git a/tests/fsm/test_fsm.py b/tests/fsm/test_fsm.py
deleted file mode 100644
index 94166fd95..000000000
--- a/tests/fsm/test_fsm.py
+++ /dev/null
@@ -1,92 +0,0 @@
-import pytest
-
-from outlines.fsm.fsm import RegexFSM, StopAtEosFSM
-
-
-def assert_expected_tensor_ids(tensor, ids):
-    assert len(tensor) == len(ids)
-    norm_tensor = sorted(map(int, tensor))
-    norm_ids = sorted(map(int, tensor))
-    assert norm_tensor == norm_ids, (norm_tensor, norm_ids)
-
-
-def test_stop_at_eos():
-    class MockTokenizer:
-        vocabulary = {"a": 1, "eos": 2}
-        eos_token_id = 2
-
-    with pytest.warns(UserWarning):
-        fsm = StopAtEosFSM(MockTokenizer())
-
-    assert fsm.allowed_token_ids(fsm.start_state) is None
-    assert fsm.allowed_token_ids(fsm.final_state) == [2]
-    assert fsm.next_state(fsm.start_state, 2) == fsm.final_state
-    assert fsm.next_state(fsm.start_state, 1) == fsm.start_state
-    assert fsm.is_final_state(fsm.start_state) is False
-    assert fsm.is_final_state(fsm.final_state) is True
-
-
-def test_regex_vocabulary_error():
-    class MockTokenizer:
-        vocabulary = {"a": 1}
-        special_tokens = {"eos"}
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-    regex_str = "[1-9]"
-
-    with pytest.raises(ValueError, match="The vocabulary"):
-        RegexFSM(regex_str, MockTokenizer())
-
-
-def test_regex():
-    class MockTokenizer:
-        vocabulary = {"1": 1, "a": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-    regex_str = "[1-9]"
-    tokenizer = MockTokenizer()
-
-    with pytest.warns(UserWarning):
-        fsm = RegexFSM(regex_str, tokenizer)
-
-    assert fsm.states_to_token_maps == {0: {1: 1}}
-    assert_expected_tensor_ids(fsm.allowed_token_ids(state=0), [1])
-    assert fsm.next_state(state=0, token_id=1) == 1
-    assert fsm.next_state(state=0, token_id=tokenizer.eos_token_id) == -1
-
-    assert fsm.is_final_state(0) is False
-
-    for state in fsm.final_states:
-        assert fsm.is_final_state(state) is True
-
-
-def test_regex_final_state():
-    """Make sure that the FSM stays in the final state as we keep generating"""
-
-    class MockTokenizer:
-        vocabulary = {"`": 101, ".": 102, "\n": 103, "eos": 104}
-        special_tokens = {"eos"}
-        eos_token_id = 104
-
-        def convert_token_to_string(self, token):
-            return token
-
-    regex_str = r"`\n(\.\n)?`\n"
-    tokenizer = MockTokenizer()
-
-    with pytest.warns(UserWarning):
-        fsm = RegexFSM(regex_str, tokenizer)
-
-    state = fsm.next_state(state=4, token_id=103)
-    assert state == 5
-    assert fsm.is_final_state(state)
-
-    state = fsm.next_state(state=5, token_id=103)
-    assert fsm.is_final_state(state)
diff --git a/tests/fsm/test_regex.py b/tests/fsm/test_regex.py
deleted file mode 100644
index 1789c4a7c..000000000
--- a/tests/fsm/test_regex.py
+++ /dev/null
@@ -1,742 +0,0 @@
-import interegular
-import numba
-import numpy as np
-import pytest
-from transformers import AutoTokenizer
-
-from outlines.fsm.regex import (
-    _walk_fsm,
-    create_fsm_index_end_to_end,
-    create_fsm_index_tokenizer,
-    fsm_union,
-    get_sub_fsms_from_seq,
-    get_token_transition_keys,
-    get_vocabulary_transition_keys,
-    make_byte_level_better_fsm,
-    make_byte_level_fsm,
-    make_deterministic_fsm,
-    reduced_vocabulary,
-    walk_fsm,
-)
-from outlines.models.transformers import TransformerTokenizer
-
-
-def identity(s):
-    return s
-
-
-def to_bytes(s):
-    return [chr(b) if b < 0x80 else f"\x00{b:02X}" for b in s.encode("utf-8")]
-
-
-def merge_symbols(byte_hexs):
-    return "".join(["\x00" + b if len(b) == 2 else b for b in byte_hexs])
-
-
-def token_str_to_trans_key(fsm, input_string):
-    return get_token_transition_keys(
-        fsm.fsm_info.alphabet_symbol_mapping,
-        fsm.fsm_info.alphabet_anything_value,
-        input_string,
-    )
-
-
-def walk_fsm_from_token_str(
-    fsm,
-    input_string: str,
-    start_state: int,
-    full_match: bool = True,
-):
-    return walk_fsm(
-        fsm,
-        token_str_to_trans_key(fsm, input_string),
-        start_state,
-        full_match,
-    )
-
-
-def walk_fsm_from_token_str_numba(
-    fsm,
-    input_string: str,
-    start_state: int,
-    full_match: bool = True,
-):
-    return _walk_fsm(
-        fsm.fsm_info.transitions,
-        fsm.fsm_info.initial,
-        fsm.fsm_info.finals,
-        token_str_to_trans_key(fsm, input_string),
-        start_state,
-        full_match=full_match,
-    )
-
-
-@pytest.mark.parametrize(
-    "function",
-    [
-        walk_fsm_from_token_str,
-        walk_fsm_from_token_str_numba,
-    ],
-)
-def test_walk_fsm(function):
-    regex_pattern = interegular.parse_pattern("0|[1-9][2-9]*")
-    regex_fsm, _ = make_deterministic_fsm(regex_pattern.to_fsm().reduce())
-
-    res = tuple(function(regex_fsm, "0", regex_fsm.initial, full_match=True))
-    assert res == (1,)
-
-    res = tuple(function(regex_fsm, "00", regex_fsm.initial, full_match=False))
-    assert res == (1,)
-
-    res = tuple(function(regex_fsm, "!", regex_fsm.initial, full_match=True))
-    assert res == tuple()
-
-    res = tuple(function(regex_fsm, "00", regex_fsm.initial, full_match=True))
-    assert res == tuple()
-
-    # This should fail, because state `1` reads nothing
-    res = tuple(function(regex_fsm, "0", 1, full_match=True))
-    assert res == tuple()
-
-    regex_pattern = interegular.parse_pattern("0|[1-9][2-9]+")
-    regex_fsm, _ = make_deterministic_fsm(regex_pattern.to_fsm().reduce())
-
-    res = tuple(function(regex_fsm, "1", regex_fsm.initial, full_match=True))
-    assert res == tuple()
-
-    res = tuple(function(regex_fsm, "1", regex_fsm.initial, full_match=False))
-    assert res == (2,)
-
-    res = tuple(function(regex_fsm, "12", regex_fsm.initial, full_match=True))
-    assert res == (2, 3)
-
-    pattern = interegular.parse_pattern(r"(?:[^\W\d]\w*|[\t \x0c]+)")
-    fsm, _ = make_deterministic_fsm(pattern.to_fsm().reduce())
-
-    res = tuple(function(fsm, "x ", fsm.initial, full_match=False))
-    assert res == (2,)
-
-    start_state = list(fsm.finals)[0]
-    res = tuple(function(fsm, "!", start_state, full_match=False))
-    assert res == tuple()
-
-
-@pytest.mark.parametrize(
-    "function",
-    [
-        walk_fsm_from_token_str,
-        walk_fsm_from_token_str_numba,
-    ],
-)
-@pytest.mark.parametrize(
-    "transform",
-    [
-        identity,
-        to_bytes,
-    ],
-)
-def test_walk_fsm_multi_bytes(function, transform):
-    regex_pattern = interegular.parse_pattern("😂|[😇-😍][😈-😍]*")
-    str_regex_fsm, _ = make_deterministic_fsm(regex_pattern.to_fsm().reduce())
-    regex_fsm = make_byte_level_better_fsm(str_regex_fsm, keep_utf8=True)
-
-    res = tuple(
-        function(
-            regex_fsm, merge_symbols(transform("😂")), regex_fsm.initial, full_match=True
-        )
-    )
-    assert res[-1:] == (1,)
-
-    res = tuple(
-        function(
-            regex_fsm,
-            merge_symbols(transform("😂😂")),
-            regex_fsm.initial,
-            full_match=False,
-        )
-    )
-    assert res[-1:] == (1,)
-
-    res = tuple(
-        function(
-            regex_fsm, merge_symbols(transform("!")), regex_fsm.initial, full_match=True
-        )
-    )
-    assert res == tuple()
-
-    res = tuple(
-        function(
-            regex_fsm,
-            merge_symbols(transform("😂😂")),
-            regex_fsm.initial,
-            full_match=True,
-        )
-    )
-    assert res == tuple()
-
-
-def test_get_sub_fsms_from_seq():
-    name_pattern = interegular.parse_pattern(r"[^\W\d]\w*")
-    name_fsm, _ = make_deterministic_fsm(name_pattern.to_fsm().reduce())
-
-    def_pattern = interegular.parse_pattern("def")
-    def_fsm, _ = make_deterministic_fsm(def_pattern.to_fsm().reduce())
-
-    match_pattern = interegular.parse_pattern("match")
-    match_fsm, _ = make_deterministic_fsm(match_pattern.to_fsm().reduce())
-
-    peq_pattern = interegular.parse_pattern(r"\+=")
-    peq_fsm, _ = make_deterministic_fsm(peq_pattern.to_fsm().reduce())
-
-    plus_pattern = interegular.parse_pattern(r"\+")
-    plus_fsm, _ = make_deterministic_fsm(plus_pattern.to_fsm().reduce())
-
-    fsms = [def_fsm, match_fsm, name_fsm, peq_fsm, plus_fsm]
-
-    fsm, fsms_to_trans_finals = fsm_union(fsms)
-
-    assert fsms_to_trans_finals == {
-        0: ({(0, 3), (3, 9), (9, 10)}, {10}, {0: {0}, 1: {3}, 2: {9}, 3: {10}}),
-        1: (
-            {(0, 4), (4, 5), (5, 6), (6, 7), (7, 8)},
-            {8},
-            {0: {0}, 1: {4}, 2: {5}, 3: {6}, 4: {7}, 5: {8}},
-        ),
-        2: (
-            {
-                (0, 2),
-                (0, 3),
-                (0, 4),
-                (2, 2),
-                (3, 2),
-                (3, 9),
-                (4, 2),
-                (4, 5),
-                (5, 2),
-                (5, 6),
-                (6, 2),
-                (6, 7),
-                (7, 2),
-                (7, 8),
-                (8, 2),
-                (9, 2),
-                (9, 10),
-                (10, 2),
-            },
-            {2, 3, 4, 5, 6, 7, 8, 9, 10},
-            {0: {0}, 1: {2, 3, 4, 5, 6, 7, 8, 9, 10}},
-        ),
-        3: ({(0, 1), (1, 11)}, {11}, {0: {0}, 1: {1}, 2: {11}}),
-        4: ({(0, 1)}, {1}, {0: {0}, 1: {1}}),
-    }
-
-    assert not fsm.accepts("1a")
-    assert fsm.accepts("a1")
-    assert fsm.accepts("def")
-    assert fsm.accepts("match")
-    assert fsm.accepts("+=")
-    assert fsm.accepts("+")
-
-    state_seq = walk_fsm_from_token_str(fsm, "def", fsm.initial)
-    state_seq.insert(0, fsm.fsm_info.initial)
-
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(0, False, True), (2, True, True)]
-
-    # Make sure the old-to-new state map is correct
-    def_state_seq = walk_fsm_from_token_str(def_fsm, "def", fsm.initial)
-    def_state_seq.insert(0, fsm.fsm_info.initial)
-
-    def_old_to_new_states = fsms_to_trans_finals[0][2]
-    assert all(
-        new_state in def_old_to_new_states[old_state]
-        for old_state, new_state in zip(def_state_seq, state_seq)
-    )
-
-    state_seq = walk_fsm_from_token_str(fsm, "ef", fsm.initial)
-    state_seq.insert(0, fsm.initial)
-
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(2, True, True)]
-
-    name_state_seq = walk_fsm_from_token_str(name_fsm, "ef", fsm.initial)
-    name_state_seq.insert(0, fsm.initial)
-
-    name_old_to_new_states = fsms_to_trans_finals[2][2]
-    assert all(
-        new_state in name_old_to_new_states[old_state]
-        for old_state, new_state in zip(name_state_seq, state_seq)
-    )
-
-    state_seq = walk_fsm_from_token_str(fsm, "match", fsm.initial)
-    state_seq.insert(0, fsm.initial)
-
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(1, False, True), (2, True, True)]
-
-    match_state_seq = walk_fsm_from_token_str(match_fsm, "match", fsm.initial)
-    match_state_seq.insert(0, fsm.initial)
-
-    match_old_to_new_states = fsms_to_trans_finals[1][2]
-    assert all(
-        new_state in match_old_to_new_states[old_state]
-        for old_state, new_state in zip(match_state_seq, state_seq)
-    )
-
-    state_seq = walk_fsm_from_token_str(fsm, "defa", fsm.initial)
-    state_seq.insert(0, fsm.initial)
-
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(2, True, True)]
-
-    state_seq = walk_fsm_from_token_str(fsm, "de", fsm.initial)
-    state_seq.insert(0, fsm.initial)
-
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(0, True, False), (2, True, True)]
-
-    state_seq = walk_fsm_from_token_str(fsm, "+", fsm.initial, False)
-    state_seq.insert(0, fsm.initial)
-
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(3, True, False), (4, False, True)]
-
-    state_seq = walk_fsm_from_token_str(fsm, "+=", fsm.initial)
-    state_seq.insert(0, fsm.initial)
-
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(3, False, True)]
-
-    # Test some overlapping patterns
-    join_fsms = [
-        interegular.parse_pattern(r"JOIN").to_fsm().reduce(),
-        interegular.parse_pattern(r"JOIN LEFT").to_fsm().reduce(),
-    ]
-    fsm, fsms_to_trans_finals = fsm_union(join_fsms)
-
-    # Matching "OI"
-    state_seq = [1, 2, 3]
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(0, True, False), (1, True, False)]
-
-    # Matching "N"
-    state_seq = [3, 4]
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(0, False, True), (1, True, False)]
-
-    # Matching " "
-    state_seq = [4, 5]
-    res = list(get_sub_fsms_from_seq(state_seq, fsms_to_trans_finals))
-    assert res == [(1, True, False)]
-
-
-def test_create_fsm_index_end_to_end():
-    regex_str = "0|[1-9][0-9]*"
-
-    regex_pattern = interegular.parse_pattern(regex_str)
-    regex_fsm, _ = make_deterministic_fsm(regex_pattern.to_fsm().reduce())
-
-    vocabulary = {
-        "blah": numba.typed.List([0]),
-        "1a": numba.typed.List([1]),
-        "2": numba.typed.List([2]),
-        "0": numba.typed.List([3]),
-        "<EOS>": numba.typed.List([4]),
-    }
-
-    vocabulary_nb = numba.typed.List.empty_list(
-        numba.types.Tuple(
-            (
-                numba.types.unicode_type,
-                numba.int64[:],
-            )
-        )
-    )
-    for token_tuple, token_ids in vocabulary.items():
-        token = merge_symbols(token_tuple)
-        token_ids_np = np.fromiter(token_ids, dtype=np.dtype("int64"))
-        vocabulary_nb.append((token, token_ids_np))
-
-    res = create_fsm_index_end_to_end(regex_fsm.fsm_info, vocabulary_nb)
-
-    assert res == {0: {(2, 2), (3, 1)}, 2: {(2, 2), (3, 2)}}
-
-
-def test_create_fsm_index_end_to_end_multi_byte():
-    regex_str = "😇| [😈-😍][😇-😎]*"
-
-    regex_pattern = interegular.parse_pattern(regex_str)
-    regex_fsm, _ = make_deterministic_fsm(regex_pattern.to_fsm().reduce())
-    byte_fsm = make_byte_level_better_fsm(regex_fsm, keep_utf8=True)
-
-    vocabulary = {
-        "blah": numba.typed.List([0]),
-        "😈a": numba.typed.List([1]),
-        "😇": numba.typed.List([2]),
-        "😍": numba.typed.List([3]),
-        merge_symbols(("F0", "9F", "98", "8D")): numba.typed.List([4]),  # '😍'
-        " 😍": numba.typed.List([5]),
-        merge_symbols((" ", "F0", "9F", "98", "8D")): numba.typed.List([6]),  # ' 😍'
-        merge_symbols((" ", "F0", "9F", "98")): numba.typed.List(
-            [7]
-        ),  # ' 😍' incomplete
-        "<EOS>": numba.typed.List([8]),
-    }
-
-    vocabulary_nb = numba.typed.List.empty_list(
-        numba.types.Tuple(
-            (
-                numba.types.unicode_type,
-                numba.int64[:],
-            )
-        )
-    )
-    for token_tuple, token_ids in vocabulary.items():
-        token_tuple_np = merge_symbols(token_tuple)
-        token_ids_np = np.fromiter(token_ids, dtype=np.dtype("int64"))
-        vocabulary_nb.append((token_tuple_np, token_ids_np))
-
-    res = create_fsm_index_end_to_end(byte_fsm.fsm_info, vocabulary_nb)
-
-    assert res == {0: {(5, 3), (6, 3), (7, 7), (2, 2)}, 3: {(2, 3), (3, 3), (4, 3)}}
-
-
-@pytest.mark.parametrize(
-    "hf_tokenizer_uri, revision",
-    [
-        ("openai-community/gpt2", "607a30d783dfa663caf39e06633721c8d4cfcd7e"),
-        ("microsoft/phi-2", "ef382358ec9e382308935a992d908de099b64c23"),
-        ("Qwen/Qwen1.5-0.5B-Chat", "4d14e384a4b037942bb3f3016665157c8bcb70ea"),
-        (
-            "NousResearch/Hermes-2-Pro-Llama-3-8B",
-            "783fd50eb82d7f57758de033861f54d62dde234f",
-        ),
-    ],
-)
-def test_create_fsm_index_tokenizer(hf_tokenizer_uri, revision):
-    # The combined regular expressions of a lexer state in a Python grammar
-    regex_str = "(?:(?:[0-9](?:(?:_)?[0-9])*(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*|(?:[0-9](?:(?:_)?[0-9])*\\.(?:[0-9](?:(?:_)?[0-9])*)?|\\.[0-9](?:(?:_)?[0-9])*)(?:(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*)?)|[0-9](?:(?:_)?[0-9])*)(?:J|j)|(?:[0-9](?:(?:_)?[0-9])*(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*|(?:[0-9](?:(?:_)?[0-9])*\\.(?:[0-9](?:(?:_)?[0-9])*)?|\\.[0-9](?:(?:_)?[0-9])*)(?:(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*)?)|0(?:x|X)(?:(?:_)?(?:[0-9]|[a-f]|[A-F]))+|0(?:b|B)(?:(?:_)?[0-1])+|0(?:o|O)(?:(?:_)?[0-7])+|(?:(?i:([ubf]?r?|r[ubf])('([^\\\\']|.)*?'))|(?i:([ubf]?r?|r[ubf])(\"([^\\\"]|.)*?\")))|(?:(?:\r?\n[\t ]*|#[^\n]*))+|[1-9](?:(?:_)?[0-9])*|\\\\[\t \x0c]*\r?\n|continue|nonlocal|assert|global|import|lambda|return|async|await|break|class|False|match|raise|while|yield|case|from|None|pass|True|with|def|del|for|not|try|if|[^\\W\\d]\\w*|#[^\n]*|[\t \x0c]+|\\.\\.\\.|@|\\{|\\(|\\[|\\-|\\+|\\*|\\~"
-
-    regex_pattern = interegular.parse_pattern(regex_str)
-    # Not reduced, so that there are many states
-    regex_fsm, _ = make_deterministic_fsm(regex_pattern.to_fsm())
-    bytes_fsm = make_byte_level_better_fsm(regex_fsm, keep_utf8=True)
-
-    num_fsm_states = len(regex_fsm.states)
-    assert num_fsm_states == 220
-
-    num_bytes_fsm_states = len(bytes_fsm.states)
-    assert num_bytes_fsm_states == 235
-
-    tokenizer = AutoTokenizer.from_pretrained(hf_tokenizer_uri, revision=revision)
-    tokenizer = TransformerTokenizer(tokenizer)
-
-    states_to_token_subsets, empty_token_ids = create_fsm_index_tokenizer(
-        bytes_fsm, tokenizer
-    )
-
-    assert not empty_token_ids
-    assert len(states_to_token_subsets) / num_fsm_states > 0.94
-
-
-@pytest.mark.parametrize(
-    "regex,string,should_accept",
-    [
-        ("[a-c]+", "😀", False),
-        ("[^a-c]+", "😀", True),
-        ("😀+", "😀😀😀", True),
-        ("😀+", "a", False),
-        ("[😀-😍]{2}", "😈😈", True),
-        ("[😀-😍]{2}", "aa", False),
-        ("[^😀-😍]{2}", "aa", True),
-        ("[^😀-😍]{2}", "😈😈", False),
-        ("[^😀-😍]{2}", "😎😎", True),
-        ("[^😀-😍]{2}", "😎😓", True),
-        ("[^😀-😍]{2}", "😎😈", False),
-        ("[😀-🙌]{2}", "😎😈", True),
-        ("[^😀-🙌]{2}", "😎😈", False),
-        ("[^😀-🙌]{2}", "🙏🙏", True),
-        ("[^😀-🙌]{2}", "🙏😎", False),
-    ],
-)
-def test_make_byte_level_fsm(regex, string, should_accept):
-    str_fsm = interegular.parse_pattern(regex).to_fsm()
-    str_accepts = str_fsm.accepts(string)
-    assert str_accepts == should_accept
-
-    byte_fsm = make_byte_level_fsm(str_fsm)
-    byte_accepts = byte_fsm.accepts(to_bytes(string))  # type: ignore
-    assert byte_accepts == str_accepts
-
-    mix_fsm = make_byte_level_fsm(str_fsm, keep_utf8=True)
-    mix_accepts = mix_fsm.accepts(to_bytes(string))  # type: ignore
-    assert mix_accepts == str_accepts
-
-    mix_accepts_utf8 = mix_fsm.accepts(string)  # type: ignore
-    assert mix_accepts_utf8 == str_accepts
-
-    def advance(fsm, state, seq):
-        for symbol in seq:
-            if state is None:
-                return None
-            key = fsm.alphabet[symbol]
-            state = fsm.map[state].get(key)
-        return state
-
-    # verify each state along the pattern
-    str_state = str_fsm.initial
-    byte_state = byte_fsm.initial
-    mix_state = byte_fsm.initial
-    for symbol in string:
-        str_state = advance(str_fsm, str_state, symbol)
-        byte_state = advance(byte_fsm, byte_state, to_bytes(symbol))
-        mix_state_utf8 = advance(mix_fsm, mix_state, symbol)
-        mix_state = advance(mix_fsm, mix_state, to_bytes(symbol))
-        assert byte_state == str_state
-        assert mix_state == str_state
-        assert mix_state_utf8 == str_state
-
-
-@pytest.mark.skip(reason="Only for local profiling")
-def test_regex_index_performance():
-    from line_profiler import LineProfiler  # type: ignore [import]
-
-    regex_str = "(?:(?:[0-9](?:(?:_)?[0-9])*(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*|(?:[0-9](?:(?:_)?[0-9])*\\.(?:[0-9](?:(?:_)?[0-9])*)?|\\.[0-9](?:(?:_)?[0-9])*)(?:(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*)?)|[0-9](?:(?:_)?[0-9])*)(?:J|j)|(?:[0-9](?:(?:_)?[0-9])*(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*|(?:[0-9](?:(?:_)?[0-9])*\\.(?:[0-9](?:(?:_)?[0-9])*)?|\\.[0-9](?:(?:_)?[0-9])*)(?:(?:e|E)(?:(?:\\+|\\-))?[0-9](?:(?:_)?[0-9])*)?)|0(?:x|X)(?:(?:_)?(?:[0-9]|[a-f]|[A-F]))+|0(?:b|B)(?:(?:_)?[0-1])+|0(?:o|O)(?:(?:_)?[0-7])+|(?:(?i:([ubf]?r?|r[ubf])('([^\\\\']|.)*?'))|(?i:([ubf]?r?|r[ubf])(\"([^\\\"]|.)*?\")))|(?:(?:\r?\n[\t ]*|#[^\n]*))+|[1-9](?:(?:_)?[0-9])*|\\\\[\t \x0c]*\r?\n|continue|nonlocal|assert|global|import|lambda|return|async|await|break|class|False|match|raise|while|yield|case|from|None|pass|True|with|def|del|for|not|try|if|[^\\W\\d]\\w*|#[^\n]*|[\t \x0c]+|\\.\\.\\.|@|\\{|\\(|\\[|\\-|\\+|\\*|\\~"
-
-    regex_pattern = interegular.parse_pattern(regex_str)
-    # Not reduced, so that there are many states
-    regex_fsm, _ = make_deterministic_fsm(regex_pattern.to_fsm())
-
-    num_fsm_states = len(regex_fsm.states)
-    assert num_fsm_states == 220
-
-    tokenizer = AutoTokenizer.from_pretrained("gpt2")
-    tokenizer = TransformerTokenizer(tokenizer)
-
-    # Pre-compile Numba functions
-    res, _ = create_fsm_index_tokenizer(regex_fsm, tokenizer)
-    assert len(res) > 1
-
-    profiler = LineProfiler(create_fsm_index_end_to_end)
-
-    profiler.runctx(
-        "create_fsm_index_tokenizer(regex_fsm, tokenizer)",
-        globals(),
-        locals(),
-    )
-    profiler.dump_stats("line-profiler-create_fsm_index.pkl")
-    profiler.print_stats(output_unit=1e-3, summarize=True, stripzeros=True)
-
-
-@pytest.mark.skip(reason="Only for local profiling")
-def test_json_index_performance():
-    import json
-    from enum import Enum
-
-    from line_profiler import LineProfiler  # type: ignore [import]
-    from pydantic import BaseModel, constr
-
-    import outlines
-
-    class Weapon(str, Enum):
-        sword = "sword"
-        axe = "axe"
-        mace = "mace"
-        spear = "spear"
-        bow = "bow"
-        crossbow = "crossbow"
-
-    class Armor(str, Enum):
-        leather = "leather"
-        chainmail = "chainmail"
-        plate = "plate"
-
-    class Character(BaseModel):
-        name: constr(max_length=10)
-        # TODO: Add support for conint
-        age: int  # conint(int, ge=18, le=100)
-        armor: Armor
-        weapon: Weapon
-        # TODO: Add support for conint
-        strength: int  # conint(int, ge=0, le=100)
-
-    model = outlines.models.transformers("gpt2", device="cuda")
-    json_schema = json.dumps(Character.model_json_schema())
-
-    def build_regex():
-        regex_str = outlines.index.json_schema.build_regex_from_object(json_schema)
-        outlines.generate.regex(model, regex_str)
-
-    profiler = LineProfiler(create_fsm_index_end_to_end)
-    profiler.add_function(create_fsm_index_tokenizer)
-    profiler.add_function(outlines.index.index.RegexFSM.__init__)
-
-    profiler.runctx(
-        "build_regex()",
-        globals(),
-        locals(),
-    )
-    profiler.dump_stats("line-profiler-build-json-regex.pkl")
-    profiler.print_stats(output_unit=1e-3, summarize=True, stripzeros=True)
-
-
-def test_token_trans_keys_identical():
-    """assert two tokens w/ identical behavior wrt FSM have same trans key seq"""
-
-    class MockTokenizer:
-        vocabulary = {"a": 1, "b": 2, "z": 3, "eos": 4}
-        special_tokens = {"eos"}
-        eos_token_id = 4
-
-        def convert_token_to_string(self, token):
-            return token
-
-    tokenizer = MockTokenizer()
-
-    pattern = r"z[ab]z"
-    regex_pattern = interegular.parse_pattern(pattern)
-    interegular_fsm = regex_pattern.to_fsm().reduce()
-    regex_fsm, _ = make_deterministic_fsm(interegular_fsm)
-    vocabulary, _ = reduced_vocabulary(tokenizer)
-    token_trans_keys = get_vocabulary_transition_keys(
-        regex_fsm.fsm_info.alphabet_symbol_mapping,
-        regex_fsm.fsm_info.alphabet_anything_value,
-        vocabulary,
-        numba.typed.List.empty_list(numba.types.unicode_type),
-    )
-
-    token_str_to_tranition_keys = {
-        token_str: trans_key_seq
-        for (token_str, _), trans_key_seq in zip(vocabulary, token_trans_keys)
-    }
-    # `a` and `b` both are workable, but `z` has distinct transition rules
-    assert interegular_fsm.accepts("zaz")
-    assert interegular_fsm.accepts("zbz")
-    assert (token_str_to_tranition_keys["a"] == token_str_to_tranition_keys["b"]).all()
-    assert not (
-        token_str_to_tranition_keys["a"] == token_str_to_tranition_keys["z"]
-    ).all()
-
-
-def test_token_trans_keys_walk_fsm():
-    """assert _walk_fsm works using transition keys"""
-
-    class MockTokenizer:
-        vocabulary = {"ab": 1, "ac": 2, "az": 3, "eos": 4}
-        special_tokens = {"eos"}
-        eos_token_id = 4
-
-        def convert_token_to_string(self, token):
-            return token
-
-    tokenizer = MockTokenizer()
-
-    pattern = r"a[bc]z"
-    regex_pattern = interegular.parse_pattern(pattern)
-    interegular_fsm = regex_pattern.to_fsm().reduce()
-    regex_fsm, _ = make_deterministic_fsm(interegular_fsm)
-    vocabulary, _ = reduced_vocabulary(tokenizer)
-    token_trans_keys = get_vocabulary_transition_keys(
-        regex_fsm.fsm_info.alphabet_symbol_mapping,
-        regex_fsm.fsm_info.alphabet_anything_value,
-        vocabulary,
-        numba.typed.List.empty_list(numba.types.unicode_type),
-    )
-
-    token_str_trans_key_seq = {
-        token_str: trans_key_seq
-        for (token_str, _), trans_key_seq in zip(vocabulary, token_trans_keys)
-    }
-
-    # verify initial state valid only for "ab" and "ac" using transition key seq
-    token_acceptance = {"ab": True, "ac": True, "az": False}
-    for token, should_accept in token_acceptance.items():
-        token_trans_key_seq = token_str_trans_key_seq[token]
-        state_seq = _walk_fsm(
-            regex_fsm.fsm_info.transitions,
-            regex_fsm.fsm_info.initial,
-            regex_fsm.fsm_info.finals,
-            token_trans_key_seq,
-            regex_fsm.fsm_info.initial,
-            False,
-        )
-        is_accepted = len(state_seq) >= len(token_trans_key_seq)
-        assert should_accept == is_accepted
-
-
-def test_numba_leading_null_byte_UnicodeCharSeq_remains_broken():
-    """Assert numba UnicodeCharSeq w/ leading \x00 is still broken"""
-    # EXPLANATION:
-    # https://github.com/dottxt-ai/outlines/pull/930#issuecomment-2143535968
-
-    # from https://github.com/numba/numba/issues/9542
-    d = numba.typed.typeddict.Dict.empty(numba.types.UnicodeCharSeq(1), numba.int64)
-    d["一"] = 10  # \xe4\xb8\x80
-    with pytest.raises(KeyError):
-        str(d)
-
-    # most characters are fine, but "\x00" is converted to ""
-    l = np.fromiter(["\x99", "\x00"], dtype=np.dtype("U2"))
-    assert str(l[0]) == "\x99"  # fine
-    assert str(l[1]) == ""  # 1-byte null converted to 0-bytes
-
-
-@pytest.mark.parametrize("input_key", ["一", "\x00"])
-def test_numba_leading_null_byte_unicode_type_sane(input_key):
-    """Assert numba unicode_type w/ leading \x00 is working"""
-    # EXPLANATION:
-    # https://github.com/dottxt-ai/outlines/pull/930#issuecomment-2143535968
-
-    # from https://github.com/numba/numba/issues/9542
-    d = numba.typed.typeddict.Dict.empty(numba.types.unicode_type, numba.int64)
-    d["一"] = 10  # \xe4\xb8\x80
-    str(d)  # assert successfully interprets
-
-
-@pytest.mark.parametrize(
-    "rare_token",
-    [
-        "�",
-        "��",
-        "�.",
-        "�..",
-        "▁�",
-        "▁▁�",
-        "▁�.",
-        "▁�.",
-        "▁▁�..",
-    ],
-)
-def test_reduced_vocabulary_with_rare_tokens(rare_token):
-    """Assert reduced_vocabulary works with rare tokens.
-
-    See [1] and [2] for context.
-
-    [1]: https://github.com/dottxt-ai/outlines/pull/763
-    [2]: https://github.com/dottxt-ai/outlines/pull/948
-    [3]: https://github.com/dottxt-ai/outlines/pull/1153
-    """
-    tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2")
-    tokenizer = TransformerTokenizer(tokenizer=tokenizer)
-    tokenizer.vocabulary[rare_token] = max(tokenizer.vocabulary.values()) + 1
-    reduced_vocabulary(tokenizer)
-
-
-def test_reduced_vocabulary_with_byte_tokens():
-    class MockTokenizer:
-        vocabulary = {
-            "string": 1,
-            b"\xa1": 2,  # Qwen-Style
-            "eos": 3,
-        }
-        special_tokens = {"eos"}
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return b"\xef\xbf\xbd".decode()
-
-    reduced_vocab = reduced_vocabulary(MockTokenizer())
-
-    # See fsm.regex.get_token_transition_keys()
-    # FSM transition keys represents bytes as <null_prefix><hex_byte>
-    assert reduced_vocab[0][1][0] == "\x00A1"
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index cdd57e0c6..1d26a9ee4 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -5,11 +5,11 @@
 
 import pytest
 import torch
+from outlines_core.fsm.regex import reduced_vocabulary
 from pydantic import BaseModel, constr
 
 import outlines.generate as generate
 import outlines.models as models
-from outlines.fsm.regex import reduced_vocabulary
 from outlines.models.transformers import Transformers, TransformerTokenizer
 from outlines.samplers import beam_search, greedy, multinomial
 

From c5be3b84ad83b1e76e8c7510a02dc24323a5f4b3 Mon Sep 17 00:00:00 2001
From: "Brandon T. Willard" <brandonwillard@users.noreply.github.com>
Date: Tue, 1 Oct 2024 17:13:13 -0500
Subject: [PATCH 081/505] Pin outlines-core version

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 7229afa83..ab3ecd775 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,7 @@ dependencies = [
    "typing_extensions",
    "pycountry",
    "airportsdata",
-   "outlines_core",
+   "outlines_core==0.1.0",
 ]
 dynamic = ["version"]
 

From 4bf14705d62b87dd655159b26ba1dd9e124f8478 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Thu, 3 Oct 2024 11:04:55 -0700
Subject: [PATCH 082/505] add missing comma in llamacpp docs

---
 docs/reference/models/llamacpp.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
index a84f0c187..0ab3dd83b 100644
--- a/docs/reference/models/llamacpp.md
+++ b/docs/reference/models/llamacpp.md
@@ -67,7 +67,7 @@ from outlines import models
 
 model = models.llamacpp(
     "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf"
+    "phi-2.Q4_K_M.gguf",
     n_gpu_layers=-1,  # to use GPU acceleration
 )
 ```

From 80b82f16ffd3a6eb15468c8891eef73e807bb098 Mon Sep 17 00:00:00 2001
From: isamu-isozaki <isamu.website@gmail.com>
Date: Fri, 28 Jun 2024 22:59:05 -0400
Subject: [PATCH 083/505] Exllamav2_filter

Fix comment

Fixed precommit issues

Removed text

Basic draft done

Passed local test

Fixed tests+precommit

Revert change for pyairports

Fixed precommit

Wrap up

Remove | for union

Attempt changing to List

Fixed for 3.8

Adding exllamav2 to optional dependency

Fixed model

Changed to fork

Fix format

Changed order

Skip exllamav2 tests

Attempt fixing coverage

Attempt fix coverage

Remove flash-attn requirement

Fixed fixture tests

Removed lora

Passed coverage

Added back transformers install

Fixed per review

Made coverage 100%
---
 outlines/generate/fsm.py                     |  13 +-
 outlines/generate/regex.py                   |  18 +-
 outlines/generate/text.py                    |  11 +-
 outlines/models/exllamav2.py                 | 410 ++++++++++++-------
 pyproject.toml                               |   1 +
 tests/generate/conftest.py                   |   8 +-
 tests/generate/test_generate.py              |  10 +
 tests/generate/test_integration_exllamav2.py | 363 ++++++++++++++++
 8 files changed, 641 insertions(+), 193 deletions(-)
 create mode 100644 tests/generate/test_integration_exllamav2.py

diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
index a9338836a..1950812d2 100644
--- a/outlines/generate/fsm.py
+++ b/outlines/generate/fsm.py
@@ -4,11 +4,10 @@
 
 from outlines.fsm.guide import RegexGuide
 from outlines.generate.api import (
-    SequenceGenerator,
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import ExLlamaV2Model, TransformersVision
+from outlines.models import TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
@@ -30,13 +29,3 @@ def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
     guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
     logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
     return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@fsm.register(ExLlamaV2Model)
-def fsm_exllamav2(
-    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
-) -> SequenceGenerator:
-    fsm = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    device = model.device
-    generator = SequenceGenerator(fsm, model, sampler, device)
-    return generator
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
index 815a8b1b9..673880e49 100644
--- a/outlines/generate/regex.py
+++ b/outlines/generate/regex.py
@@ -1,12 +1,10 @@
 from functools import singledispatch
 
-from outlines.fsm.guide import RegexGuide
 from outlines.generate.api import (
-    SequenceGenerator,
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import ExLlamaV2Model, OpenAI, TransformersVision
+from outlines.models import OpenAI, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
@@ -49,20 +47,6 @@ def regex_vision(
     return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
-@regex.register(ExLlamaV2Model)
-def regex_exllamav2(
-    model,
-    regex_str: str,
-    sampler: Sampler = multinomial(),
-) -> SequenceGenerator:
-    fsm = RegexGuide(regex_str, model.tokenizer)
-
-    device = model.device
-    generator = SequenceGenerator(fsm, model, sampler, device)
-
-    return generator
-
-
 @regex.register(OpenAI)
 def regex_openai(
     model: OpenAI,
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
index 3fe3dc553..32530d0c4 100644
--- a/outlines/generate/text.py
+++ b/outlines/generate/text.py
@@ -1,12 +1,10 @@
 from functools import singledispatch
 
-from outlines.fsm.guide import StopAtEOSGuide
 from outlines.generate.api import (
-    SequenceGenerator,
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import ExLlamaV2Model, OpenAI, TransformersVision
+from outlines.models import OpenAI, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
@@ -36,13 +34,6 @@ def text(model, sampler: Sampler = multinomial()) -> SequenceGeneratorAdapter:
     return SequenceGeneratorAdapter(model, None, sampler)
 
 
-@text.register(ExLlamaV2Model)
-def text_exllamav2(model, sampler: Sampler = multinomial()) -> SequenceGenerator:
-    fsm = StopAtEOSGuide(model.tokenizer)
-    device = model.device
-    return SequenceGenerator(fsm, model, sampler, device)
-
-
 @text.register(TransformersVision)
 def text_vision(model, sampler: Sampler = multinomial()):
     return VisionSequenceGeneratorAdapter(model, None, sampler)
diff --git a/outlines/models/exllamav2.py b/outlines/models/exllamav2.py
index 0ec6ef033..f06b7e46e 100644
--- a/outlines/models/exllamav2.py
+++ b/outlines/models/exllamav2.py
@@ -1,12 +1,21 @@
-import os
-from typing import TYPE_CHECKING, Optional
+import dataclasses
+from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, TypedDict, Union
+
+from typing_extensions import Unpack
+
+from outlines.generate.api import GenerationParameters, SamplingParameters
 
 if TYPE_CHECKING:
-    from exllamav2 import ExLlamaV2, ExLlamaV2Cache, ExLlamaV2Lora
-    from transformers import PreTrainedTokenizer
-    import torch
+    from exllamav2 import ExLlamaV2Tokenizer
+    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler
+
 
-from .transformers import TransformerTokenizer
+class ExllamaV2Params(TypedDict, total=False):
+    max_tokens: int
+    stop_conditions: Optional[List[Union[int, str]]]
+    seed: Optional[int]
+    gen_settings: "ExLlamaV2Sampler.Settings"
+    max_new_tokens: List[int]
 
 
 class ExLlamaV2Model:
@@ -14,108 +23,218 @@ class ExLlamaV2Model:
 
     def __init__(
         self,
-        model: "ExLlamaV2",
-        tokenizer: "PreTrainedTokenizer",
-        device,
-        cache: "ExLlamaV2Cache",
-        lora: Optional["ExLlamaV2Lora"] = None,
+        generator: "ExLlamaV2DynamicGenerator",
+        tokenizer: "ExLlamaV2Tokenizer",
+        max_seq_len: int,
     ):
-        self.device = device
-        self.model = model
-        self.tokenizer = TransformerTokenizer(tokenizer)
-        self.cache = cache
-        self.past_seq = None
-        self.lora = lora
-
-    def forward(self, input_ids: "torch.LongTensor", *_):
-        """Compute a forward pass through the exl2 model."""
-        import torch
-
-        # Caching with past_seq
-        reset = True
-        seq_tensor = input_ids[0]
-
-        if self.past_seq is not None:
-            min_length = min(self.past_seq.shape[0], seq_tensor.shape[0])
-            indices = torch.nonzero(
-                ~torch.eq(self.past_seq[:min_length], seq_tensor[:min_length])
-            )
-            if len(indices) > 0:
-                longest_prefix = indices[0].item()
-            else:
-                longest_prefix = min_length
-
-            if longest_prefix > 0:
-                reset = False
-                self.cache.current_seq_len = longest_prefix
-                if seq_tensor.shape[0] - longest_prefix > 1:
-                    self.model.forward(
-                        seq_tensor[longest_prefix:-1].view(1, -1),
-                        self.cache,
-                        preprocess_only=True,
-                        loras=[self.lora],
-                    )
-                elif seq_tensor.shape[0] == longest_prefix:
-                    self.cache.current_seq_len -= 1
-
-        if reset:
-            self.cache.current_seq_len = 0
-            if seq_tensor.shape[0] > 1:
-                self.model.forward(
-                    seq_tensor[:-1].view(1, -1),
-                    self.cache,
-                    preprocess_only=True,
-                    loras=[self.lora],
+        self.generator = generator
+        self.tokenizer = tokenizer
+        self.max_seq_len = max_seq_len
+
+    def prepare_generation_parameters(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        sampling_parameters: SamplingParameters,
+        structure_logits_processor,
+        **exllamav2_params: Unpack[ExllamaV2Params],
+    ) -> Tuple[ExllamaV2Params, Union[str, List[str]]]:
+        """Prepare the generation parameters.
+
+        `exllamav2` uses different default values
+
+        """
+        from exllamav2.generator import ExLlamaV2Sampler
+
+        if isinstance(prompts, str):
+            prompts = [prompts]
+        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
+
+        if max_tokens is None:
+            max_tokens = []
+            for prompt in prompts:
+                ids = self.generator.tokenizer.encode(
+                    prompt, encode_special_tokens=True
                 )
+                prompt_tokens = ids.shape[-1]
+                max_tokens.append(self.max_seq_len - prompt_tokens)
+            exllamav2_params["max_new_tokens"] = max_tokens
+        else:
+            exllamav2_params["max_new_tokens"] = [
+                max_tokens for _ in range(len(prompts))
+            ]
 
-        self.past_seq = seq_tensor
+        stop_conditions = [self.generator.tokenizer.eos_token_id]
+        if isinstance(generation_parameters.stop_at, str):
+            stop_conditions.append(generation_parameters.stop_at)
+        elif isinstance(generation_parameters.stop_at, list):
+            for stop_at in generation_parameters.stop_at:
+                stop_conditions.append(stop_at)
+        exllamav2_params["stop_conditions"] = stop_conditions
+        exllamav2_params["seed"] = seed
 
-        return self.model.forward(
-            seq_tensor[-1:].view(1, -1), self.cache, loras=[self.lora]
-        )
+        gen_settings = ExLlamaV2Sampler.Settings()
+        if sampling_parameters.temperature is not None:
+            gen_settings.temperature = sampling_parameters.temperature
+        if sampling_parameters.top_p is not None:
+            gen_settings.top_p = sampling_parameters.top_p
+        if sampling_parameters.top_k is not None:
+            gen_settings.top_k = sampling_parameters.top_k
+        gen_settings.logits_processor = structure_logits_processor
+        exllamav2_params["gen_settings"] = gen_settings
+        if sampling_parameters.num_samples > 1:
+            prompts = prompts * sampling_parameters.num_samples
+            exllamav2_params["max_new_tokens"] = (
+                exllamav2_params["max_new_tokens"] * sampling_parameters.num_samples
+            )
 
-    def __call__(self, input_ids: "torch.LongTensor", *_) -> "torch.FloatTensor":
-        logits = self.forward(input_ids)
-        next_token_logits = logits[..., -1, :]
+        if len(prompts) == 1:
+            prompts = prompts[0]
 
-        return next_token_logits, None
+        return exllamav2_params, prompts
 
-    def update_lora(self, lora_path: Optional[str] = None):
+    def reformat_output(
+        self, output: Union[str, List[str]], sampling_parameters: SamplingParameters
+    ):
         """
-        Update and apply the LoRA to the model.
+        The purpose of this function is to reformat the output from exllamav2's output format to outline's output format
+        For exllamav2, it mainly accepts only a list or a string(they also do cfg sampling with tuples but we will ignore this for now)
+        The exllamav2's logic is
+        1. If the prompt is a string, return a string. This is the same as outlines
+        2. If a prompt is a list, return a list. This is not the same as outlines output in that if the list is only one element, the string is expected to be outputted.
+        3. There is no such thing as num_samples, so the prompts had to be duplicated by num_samples times. Then, we had the function output a list of lists
+        """
+        if isinstance(output, str):
+            return output
+        if len(output) == 1:
+            return output[0]
+        if sampling_parameters.num_samples > 1:
+            if len(output) == sampling_parameters.num_samples:
+                return output
+            assert len(output) % sampling_parameters.num_samples == 0
+            num_items_per_sample = len(output) // sampling_parameters.num_samples
+            new_output = []
+            for i in range(sampling_parameters.num_samples):
+                curr_sample = []
+                for j in range(num_items_per_sample):
+                    curr_sample.append(output[i * num_items_per_sample + j])
+                new_output.append(curr_sample)
+            return new_output
+        return output
 
-        Args:
-            lora_path (Optional[str]): The path to the LoRA directory. If None, the LoRA will be unloaded.
+    def generate(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        structure_logits_processor,
+        sampling_parameters: SamplingParameters,
+        **exllamav2_params: Unpack[ExllamaV2Params],
+    ) -> Union[str, List[str]]:
+        exllamav2_params, prompts = self.prepare_generation_parameters(
+            prompts,
+            generation_parameters,
+            sampling_parameters,
+            structure_logits_processor,
+        )
         """
-        try:
-            from exllamav2 import ExLlamaV2Lora
-        except ImportError:
-            raise ImportError(
-                "The `exllamav2` library needs to be installed in order to use `exllamav2` models."
+        In exllamav2, it needs the max amount of new tokens generated.
+        The reason exllamav2_params["max_new_tokens"] is a list is because in prepare_generation_parameters
+        the max amount of tokens that can be generated by the model for each prompt(by encoding with tokenizer) is calculated.
+        The minimum is picked because otherwise it might be possible for one of the
+        prompts to exceed the max sequence length.
+        """
+        output = self.generator.generate(
+            prompt=prompts,
+            gen_settings=exllamav2_params["gen_settings"],
+            max_new_tokens=min(exllamav2_params["max_new_tokens"]),
+            completion_only=True,
+            encode_special_tokens=True,
+            stop_conditions=exllamav2_params["stop_conditions"],
+            add_bos=False,
+            seed=exllamav2_params["seed"],
+        )
+
+        return self.reformat_output(output, sampling_parameters)
+
+    def stream(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        structure_logits_processor,
+        sampling_parameters: SamplingParameters,
+        **exllamav2_params: Unpack[ExllamaV2Params],
+    ) -> Iterator[Union[str, List[str]]]:
+        from exllamav2.generator import ExLlamaV2DynamicJob
+
+        exllamav2_params, prompts = self.prepare_generation_parameters(
+            prompts,
+            generation_parameters,
+            sampling_parameters,
+            structure_logits_processor,
+        )
+
+        order = {}
+        if isinstance(prompts, str):
+            prompts = [prompts]
+        batch_size = len(prompts)
+        seed = exllamav2_params["seed"]
+        for idx, p in enumerate(prompts):
+            input_ids = self.generator.tokenizer.encode(
+                p, encode_special_tokens=True, add_bos=False
             )
-        if lora_path is None:
-            if self.lora is not None:
-                print(" -- Unloading LoRA...")
-            self.lora = None
-        else:
-            self.lora = ExLlamaV2Lora.from_directory(self.model, lora_path)
-            print(" -- Loading LoRA...")
+
+            job = ExLlamaV2DynamicJob(
+                input_ids=input_ids,
+                max_new_tokens=exllamav2_params["max_new_tokens"][idx],
+                min_new_tokens=0,
+                seed=seed,
+                stop_conditions=exllamav2_params["stop_conditions"],
+                gen_settings=exllamav2_params["gen_settings"],
+                token_healing=False,
+                decode_special_tokens=False,
+            )
+
+            if seed is not None:
+                seed += 1
+
+            serial = self.generator.enqueue(job)
+            order[serial] = idx
+
+        # Collect outputs until all jobs finish
+
+        next_text = [""] * batch_size
+
+        def token_generator() -> Iterator[str]:
+            while self.generator.num_remaining_jobs():
+                results = self.generator.iterate()
+                for r in results:
+                    idx = order[r["serial"]]
+                    if r["stage"] == "streaming":
+                        text = r.get("text", "")
+                        next_text[idx] = text
+                    if r["eos"]:
+                        next_text[idx] = ""
+                yield self.reformat_output(next_text, sampling_parameters)
+            return
+
+        return token_generator()
+
+
+# Taken from https://github.com/lapp0/exllamav2/pull/1/files#diff-26f303de07c10aad998e33d3df52581643673a598162cc4b35ef051f52d7c60b
+def patch_tokenizer(tokenizer):
+    tokenizer.vocabulary = tokenizer.piece_to_id
+    tokenizer.special_tokens = set(tokenizer.extended_piece_to_id)
+    tokenizer.convert_token_to_string = lambda t: t
+    return tokenizer
 
 
 def exl2(
     model_path: str,
-    device: str,
+    draft_model_path: Optional[str] = None,
     max_seq_len: Optional[int] = None,
-    scale_pos_emb: Optional[float] = None,
-    scale_alpha_value: Optional[float] = None,
-    no_flash_attn: Optional[bool] = None,
-    num_experts_per_token: Optional[int] = None,
-    cache_8bit: bool = False,
     cache_q4: bool = False,
-    tokenizer_kwargs: dict = {},
-    gpu_split: Optional[str] = None,
-    low_mem: Optional[bool] = None,
-    verbose: Optional[bool] = None,
+    paged: bool = True,
+    max_chunk_size: Optional[int] = None,
 ) -> ExLlamaV2Model:
     """
     Load an ExLlamaV2 model.
@@ -136,8 +255,6 @@ def exl2(
         Disable flash attention. Defaults to None.
     num_experts_per_token (Optional[int], optional)
         Number of experts per token. Defaults to None.
-    cache_8bit (bool, optional)
-        Use 8-bit cache. Defaults to False.
     cache_q4 (bool, optional)
         Use Q4 cache. Defaults to False.
     tokenizer_kwargs (dict, optional)
@@ -162,71 +279,62 @@ def exl2(
         from exllamav2 import (
             ExLlamaV2,
             ExLlamaV2Cache,
-            ExLlamaV2Cache_8bit,
             ExLlamaV2Cache_Q4,
             ExLlamaV2Config,
+            ExLlamaV2Tokenizer,
         )
-        from transformers import AutoTokenizer
+        from exllamav2.generator import ExLlamaV2DynamicGenerator
+
     except ImportError:
         raise ImportError(
             "The `exllamav2`, `transformers` and `torch` libraries needs to be installed in order to use `exllamav2` models."
         )
+    config = ExLlamaV2Config(model_path)
+    if max_chunk_size is not None:
+        config.max_input_len = max_chunk_size
+        config.max_attention_size = max_chunk_size**2
 
-    # Load tokenizer
-    if not verbose:
-        print(" -- Loading tokenizer...")
-    tokenizer_kwargs.setdefault("padding_side", "left")
-    tokenizer = AutoTokenizer.from_pretrained(model_path, **tokenizer_kwargs)
-    # tokenizer = TransformerTokenizer(model_path, **tokenizer_kwargs)
-
-    # Check fasttensors for config
-    if os.name != "nt":
-        use_fasttensors = True
-    else:
-        use_fasttensors = False
-
-    # Create config
-    config = ExLlamaV2Config()
-    config.model_dir = model_path
-    config.fasttensors = use_fasttensors
-    config.prepare()
-
-    # Set config options
-    if max_seq_len is not None:
-        config.max_seq_len = max_seq_len
-    if scale_pos_emb is not None:
-        config.scale_pos_emb = scale_pos_emb
-    if scale_alpha_value is not None:
-        config.scale_alpha_value = scale_alpha_value
-    if no_flash_attn is not None:
-        config.no_flash_attn = no_flash_attn
-    if num_experts_per_token is not None:
-        config.num_experts_per_token = num_experts_per_token
-    if low_mem:
-        config.set_low_mem()
-
-    # Prepare the model from the config
+    config.arch_compat_overrides()
     model = ExLlamaV2(config)
-
-    # Create cache
-    if cache_8bit:
-        cache = ExLlamaV2Cache_8bit(model, lazy=not model.loaded)
-    elif cache_q4:
-        cache = ExLlamaV2Cache_Q4(model, lazy=not model.loaded)
+    if max_seq_len is None:
+        max_seq_len = -1
+    if cache_q4:
+        cache = ExLlamaV2Cache_Q4(model, max_seq_len=max_seq_len, lazy=True)
     else:
-        cache = ExLlamaV2Cache(model, lazy=not model.loaded)
-
-    # Load the model
-    split = None
-    if gpu_split and gpu_split != "auto":
-        split = [float(alloc) for alloc in gpu_split.split(",")]
-        if not verbose:
-            print(" -- Loading model...")
-        model.load(split)
-
-    # Autoload if no GPU split was provided
-    if not model.loaded:
-        print(" -- Loading model...")
-        model.load_autosplit(cache)
-
-    return ExLlamaV2Model(model, tokenizer, device, cache)
+        cache = ExLlamaV2Cache(model, max_seq_len=max_seq_len, lazy=True)
+    model.load_autosplit(cache, progress=True)
+
+    print("Loading tokenizer...")
+    tokenizer = ExLlamaV2Tokenizer(config)
+    tokenizer = patch_tokenizer(tokenizer)
+    max_batch_size = 4 if paged else 1
+
+    draft_model = None
+    draft_cache = None
+    if draft_model_path is not None:
+        draft_config = ExLlamaV2Config(draft_model_path)
+        draft_model = ExLlamaV2(draft_config)
+
+        if cache_q4:
+            draft_cache = ExLlamaV2Cache_Q4(
+                draft_model, max_seq_len=max_seq_len, lazy=True
+            )
+        else:
+            draft_cache = ExLlamaV2Cache(
+                draft_model, max_seq_len=max_seq_len, lazy=True
+            )
+
+    # Initialize the generator with all default parameters
+    generator = ExLlamaV2DynamicGenerator(
+        model=model,
+        cache=cache,
+        draft_model=draft_model,
+        draft_cache=draft_cache,
+        tokenizer=tokenizer,
+        max_batch_size=max_batch_size,
+        use_ngram_draft=False,
+        max_chunk_size=max_chunk_size,
+        paged=paged,
+    )
+    max_seq_len = cache.max_seq_len
+    return ExLlamaV2Model(generator, tokenizer, max_seq_len)
diff --git a/pyproject.toml b/pyproject.toml
index ab3ecd775..ac94ecf57 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -64,6 +64,7 @@ test = [
     "torch",
     "transformers",
     "pillow",
+    "exllamav2",
 ]
 serve = [
     "vllm>=0.3.0",
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
index ed8830119..abd9c72a4 100644
--- a/tests/generate/conftest.py
+++ b/tests/generate/conftest.py
@@ -27,9 +27,11 @@ def pytest_collection_modifyitems(config, items):
         for item in items:
             if "model_fixture" in item.fixturenames:
                 model_param = item.callspec.params.get("model_fixture", None)
-                if model_param.startswith(
-                    "model_transformers_vision"
-                ) or model_param.startswith("model_vllm"):
+                if (
+                    model_param.startswith("model_transformers_vision")
+                    or model_param.startswith("model_vllm")
+                    or model_param.startswith("model_exllamav2")
+                ):
                     item.add_marker(skip_marker)
 
     if not is_metal_available():
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index a96ce8673..b36baf9a4 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -20,6 +20,15 @@ def model_llamacpp(tmp_path_factory):
     )
 
 
+@pytest.fixture(scope="session")
+def model_exllamav2(tmp_path_factory):
+    return models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=False,
+    )
+
+
 @pytest.fixture(scope="session")
 def model_mlxlm(tmp_path_factory):
     return models.mlxlm("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
@@ -98,6 +107,7 @@ def model_t5(tmp_path_factory):
 
 ALL_MODEL_FIXTURES = (
     "model_llamacpp",
+    "model_exllamav2",
     "model_mlxlm",
     "model_mlxlm_phi3",
     "model_transformers_random",
diff --git a/tests/generate/test_integration_exllamav2.py b/tests/generate/test_integration_exllamav2.py
new file mode 100644
index 000000000..12c4143b3
--- /dev/null
+++ b/tests/generate/test_integration_exllamav2.py
@@ -0,0 +1,363 @@
+import importlib
+from unittest.mock import patch
+
+import pytest
+
+import outlines.models as models
+from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.models.exllamav2 import ExLlamaV2Model
+
+
+@pytest.fixture(scope="session")
+def model_exllamav2(tmp_path_factory):
+    return models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=False,
+    )
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_import_error(request, model_fixture):
+    with patch.dict("sys.modules", {"exllamav2": None}):
+        with pytest.raises(ImportError):
+            models.exl2(
+                model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+                cache_q4=True,
+                paged=False,
+            )
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_attributes(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    assert hasattr(model, "generator")
+    assert hasattr(model, "tokenizer")
+    assert model.tokenizer.convert_token_to_string(1) == 1
+    assert hasattr(model, "max_seq_len")
+    assert isinstance(model.max_seq_len, int)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_prompt_types(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+    prompt = ["test"]
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_no_max_tokens(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_test_stop_at(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_multisampling(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, list)
+    assert isinstance(output[0], str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_prepare_generation_parameters(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    exllamav2_params, prompts = model.prepare_generation_parameters(
+        prompt, generation_params, sampling_params, structure_logits_processor
+    )
+    assert isinstance(exllamav2_params, dict)
+    assert isinstance(prompts, list)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_prompt_types(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+    prompt = ["test"]
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_no_max_tokens(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_test_stop_at(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_multisampling(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, list)
+        assert isinstance(token[0], str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_seed(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, seed=1, stop_at=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_reformat_output(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+    )
+    output = "test"
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert reformatted_output == output
+    output = ["test"]
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert reformatted_output == output[0]
+    output = ["test", "test"]
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+    )
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert len(reformatted_output) == 2
+    assert reformatted_output[0] == "test"
+    assert reformatted_output[1] == "test"
+    output = ["test", "test"]
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert len(reformatted_output) == 2
+    assert reformatted_output[0] == "test"
+    assert reformatted_output[1] == "test"
+    output = ["test", "test", "test", "test"]
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert len(reformatted_output) == 2
+    assert reformatted_output[0] == ["test", "test"]
+    assert reformatted_output[1] == ["test", "test"]
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_max_chunk_size(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=False,
+        max_chunk_size=128,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_cache_default(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        paged=False,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def is_flash_attn_available():
+    try:
+        importlib.import_module("flash_attn")
+    except (ImportError, AssertionError):
+        return False
+    return True
+
+
+@pytest.mark.skipif(not is_flash_attn_available(), reason="flash-attn is not installed")
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_paged(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=True,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_draft_model(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=False,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_draft_model_cache_default(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        paged=False,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_set_max_seq_len(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        max_seq_len=2048,
+        paged=False,
+        cache_q4=True,
+    )
+    assert isinstance(model, ExLlamaV2Model)

From faa7c5c77fbe7a0d2f1117d9fc8c407766614294 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Fri, 4 Oct 2024 16:02:47 -0400
Subject: [PATCH 084/505] automatically download exl2 model in tests

fix exl bug: sometimes piece_to_id not populated, but get_piece_to_id() still works

fix exl bug: sometimes piece_to_id not populated, but get_piece_to_id() still works

enable exl2 in generate.cfg

cleate OutlinesExLlamaV2Tokenizer rather than monkey patching
---
 outlines/generate/cfg.py        |  9 +-------
 outlines/models/exllamav2.py    | 39 +++++++++++++++++++++++----------
 tests/generate/test_generate.py | 10 ++++++++-
 3 files changed, 37 insertions(+), 21 deletions(-)

diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
index 034a65ae5..b677040d5 100644
--- a/outlines/generate/cfg.py
+++ b/outlines/generate/cfg.py
@@ -4,7 +4,7 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import ExLlamaV2Model, LlamaCpp, OpenAI, TransformersVision
+from outlines.models import LlamaCpp, OpenAI, TransformersVision
 from outlines.samplers import Sampler, multinomial
 
 
@@ -41,13 +41,6 @@ def cfg_vision(model, cfg_str: str, sampler: Sampler = multinomial()):
     return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
-@cfg.register(ExLlamaV2Model)
-def cfg_exllamav2(model, cfg_str: str, sampler: Sampler = multinomial()):
-    raise NotImplementedError(
-        "Not yet available, track progress in https://github.com/dottxt-ai/outlines/pull/1010"
-    )
-
-
 @cfg.register(LlamaCpp)
 def cfg_llamacpp(model, cfg_str: str, sampler: Sampler = multinomial()):
     raise NotImplementedError("Not yet available due to bug in llama_cpp tokenizer")
diff --git a/outlines/models/exllamav2.py b/outlines/models/exllamav2.py
index f06b7e46e..821d4e591 100644
--- a/outlines/models/exllamav2.py
+++ b/outlines/models/exllamav2.py
@@ -1,12 +1,13 @@
 import dataclasses
 from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, TypedDict, Union
 
+import torch
 from typing_extensions import Unpack
 
 from outlines.generate.api import GenerationParameters, SamplingParameters
 
 if TYPE_CHECKING:
-    from exllamav2 import ExLlamaV2Tokenizer
+    import torch.LongTensor
     from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler
 
 
@@ -18,13 +19,33 @@ class ExllamaV2Params(TypedDict, total=False):
     max_new_tokens: List[int]
 
 
+class OutlinesExLlamaV2Tokenizer:
+    def __init__(self, tokenizer):
+        self.exl2_tokenizer = tokenizer
+        self.vocabulary = self.exl2_tokenizer.get_piece_to_id_dict()
+        self.special_tokens = set(self.exl2_tokenizer.extended_piece_to_id)
+        self.eos_token_id = self.exl2_tokenizer.eos_token_id
+
+    def convert_token_to_string(self, token):
+        return token
+
+    def decode(self, token_ids: "torch.LongTensor") -> List[str]:
+        decoded = self.exl2_tokenizer.decode(
+            torch.tensor(token_ids),
+            decode_special_tokens=False,
+        )
+        if isinstance(decoded, str):
+            return [decoded]
+        return decoded
+
+
 class ExLlamaV2Model:
     """Represents a `exl2` model."""
 
     def __init__(
         self,
         generator: "ExLlamaV2DynamicGenerator",
-        tokenizer: "ExLlamaV2Tokenizer",
+        tokenizer: "OutlinesExLlamaV2Tokenizer",
         max_seq_len: int,
     ):
         self.generator = generator
@@ -220,14 +241,6 @@ def token_generator() -> Iterator[str]:
         return token_generator()
 
 
-# Taken from https://github.com/lapp0/exllamav2/pull/1/files#diff-26f303de07c10aad998e33d3df52581643673a598162cc4b35ef051f52d7c60b
-def patch_tokenizer(tokenizer):
-    tokenizer.vocabulary = tokenizer.piece_to_id
-    tokenizer.special_tokens = set(tokenizer.extended_piece_to_id)
-    tokenizer.convert_token_to_string = lambda t: t
-    return tokenizer
-
-
 def exl2(
     model_path: str,
     draft_model_path: Optional[str] = None,
@@ -306,7 +319,6 @@ def exl2(
 
     print("Loading tokenizer...")
     tokenizer = ExLlamaV2Tokenizer(config)
-    tokenizer = patch_tokenizer(tokenizer)
     max_batch_size = 4 if paged else 1
 
     draft_model = None
@@ -337,4 +349,7 @@ def exl2(
         paged=paged,
     )
     max_seq_len = cache.max_seq_len
-    return ExLlamaV2Model(generator, tokenizer, max_seq_len)
+
+    outlines_tokenizer = OutlinesExLlamaV2Tokenizer(tokenizer)
+    outlines_exl2_model = ExLlamaV2Model(generator, outlines_tokenizer, max_seq_len)
+    return outlines_exl2_model
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index b36baf9a4..9c288c21e 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -22,8 +22,16 @@ def model_llamacpp(tmp_path_factory):
 
 @pytest.fixture(scope="session")
 def model_exllamav2(tmp_path_factory):
+    from huggingface_hub import snapshot_download
+
+    tmp_dir = tmp_path_factory.mktemp("model_download")
+    model_path = snapshot_download(
+        repo_id="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4.6-exl2",
+        cache_dir=tmp_dir,
+    )
+
     return models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        model_path=model_path,
         cache_q4=True,
         paged=False,
     )

From 0da70390d63fba8d2b6f4c1a439ff091a6ddf23c Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Fri, 4 Oct 2024 22:24:29 -0400
Subject: [PATCH 085/505] document third party exllamav2 with logits processor

---
 docs/reference/models/exllamav2.md | 10 +++++++++-
 outlines/models/exllamav2.py       |  4 +++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/docs/reference/models/exllamav2.md b/docs/reference/models/exllamav2.md
index afe542112..a4e727840 100644
--- a/docs/reference/models/exllamav2.md
+++ b/docs/reference/models/exllamav2.md
@@ -1,7 +1,15 @@
 # ExllamaV2
 
+The `outlines.models.exllamav2` model requires a Logits Processor component for compatibility with Outlines structured generation. While ExLlamaV2 doesn't natively support this feature, a third-party fork provides the necessary functionality. You can install it with the following command:
+
+```bash
+pip install git+https://github.com/lapp0/exllamav2@sampler-logits-processor
+```
+
+Install other requirements:
+
 ```bash
-pip install exllamav2 transformers torch
+pip install transformers torch
 ```
 
 *Coming soon*
diff --git a/outlines/models/exllamav2.py b/outlines/models/exllamav2.py
index 821d4e591..78da796fb 100644
--- a/outlines/models/exllamav2.py
+++ b/outlines/models/exllamav2.py
@@ -300,7 +300,9 @@ def exl2(
 
     except ImportError:
         raise ImportError(
-            "The `exllamav2`, `transformers` and `torch` libraries needs to be installed in order to use `exllamav2` models."
+            "The `exllamav2`, `transformers` and `torch` libraries needs to be installed in order to use `exllamav2` models. "
+            "Please run `pip install transformers torch git+https://github.com/lapp0/exllamav2@sampler-logits-processor` "
+            "Documentation: https://dottxt-ai.github.io/outlines/reference/models/exllamav2/"
         )
     config = ExLlamaV2Config(model_path)
     if max_chunk_size is not None:

From ef4e819102cb3f94efb85dd8d510c09edabc36a5 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sun, 21 Jul 2024 14:41:40 -0500
Subject: [PATCH 086/505] Version the Documentation

---
 .github/workflows/publish_documentation.yml | 26 +++++++++++++++++++--
 mkdocs.yml                                  |  4 ++++
 requirements-doc.txt                        |  1 +
 3 files changed, 29 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/publish_documentation.yml b/.github/workflows/publish_documentation.yml
index d861424f3..d1235e085 100644
--- a/.github/workflows/publish_documentation.yml
+++ b/.github/workflows/publish_documentation.yml
@@ -5,6 +5,9 @@ on:
   push:
     branches:
       - main
+  release:
+    types:
+      - created
 
 permissions:
   contents: write
@@ -14,6 +17,8 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
       - uses: actions/setup-python@v4
         with:
           python-version: 3.x
@@ -25,7 +30,24 @@ jobs:
           restore-keys: |
             mkdocs-material-
       - run: pip install -r requirements-doc.txt
-      - name: Build documentation
+      - run: mkdocs build
+
+      - name: Set up Git
+        run: |
+          git config user.name ${{ github.actor }}
+          git config user.email ${{ github.actor }}@users.noreply.github.com
+
+      - name: Publish Tag as latest
+        env:
+          GOOGLE_ANALYTICS_KEY: ${{ secrets.GOOGLE_ANALYTICS_KEY }}
+        if: github.event_name == 'release'
+        run: |
+          mike deploy --push --update-aliases ${{ github.ref_name }} latest
+          mike set-default --push latest
+
+      - name: Publish main as unstable
         env:
           GOOGLE_ANALYTICS_KEY: ${{ secrets.GOOGLE_ANALYTICS_KEY }}
-        run: mkdocs gh-deploy --force
+        if: github.event_name == 'push'
+        run: |
+          mike deploy --push --update-aliases ${{ github.ref_name }} unstable
diff --git a/mkdocs.yml b/mkdocs.yml
index 4189df1c0..b200ee3d8 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -45,6 +45,10 @@ extra:
   analytics:
     provider: google
     property: !ENV GOOGLE_ANALYTICS_KEY
+  version:
+    provider: mike
+    default: latest
+    alias: true
 
 # Extensions
 markdown_extensions:
diff --git a/requirements-doc.txt b/requirements-doc.txt
index cf8c674a7..05f9f36ec 100644
--- a/requirements-doc.txt
+++ b/requirements-doc.txt
@@ -6,3 +6,4 @@ mkdocs-section-index
 mkdocstrings[python]
 mkdocs-git-committers-plugin-2
 mkdocs-git-revision-date-localized-plugin
+mike

From 541a8b0f7a1c76eee085d9babc3673b0031a0f2c Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 5 Oct 2024 15:25:35 -0400
Subject: [PATCH 087/505] update logits in place for GuideLogitsProcessor

---
 outlines/processors/structured.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 93d68727f..d9a912c89 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -102,12 +102,13 @@ def process_logits(
 
             sequence_states.append(self._guide_states[curr_state_key])
 
-        mask = torch.full_like(logits, -math.inf)
+        mask = torch.ones_like(logits, dtype=torch.bool)
         for i, guide_state in enumerate(sequence_states):
             allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
-            mask[i, allowed_tokens] = logits[i, allowed_tokens]
+            mask[i, allowed_tokens] = False
+        logits.masked_fill_(mask, float("-inf"))
 
-        return mask
+        return logits
 
     def copy(self) -> "GuideLogitsProcessor":
         """Return a copy of the logits processor."""

From d1b8728c6c65355f53a2c78432f8b07ad7518b7a Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Sat, 5 Oct 2024 16:29:21 -0400
Subject: [PATCH 088/505] construct logits mask in batch operation

---
 outlines/processors/base_logits_processor.py |  4 +--
 outlines/processors/structured.py            | 33 ++++++++++++++------
 2 files changed, 25 insertions(+), 12 deletions(-)

diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index feedf5253..9a52abecd 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -75,10 +75,10 @@ def __call__(
 
         # Guarantee passed as 2D Tensors, then covert back to original (1D or 2D) shape
         if len(torch_logits.shape) == 2:
-            processed_logits = self.process_logits(input_ids.tolist(), torch_logits)
+            processed_logits = self.process_logits(input_ids, torch_logits)
         elif len(torch_logits.shape) == 1:
             processed_logits = self.process_logits(
-                [input_ids.tolist()], torch_logits.unsqueeze(0)
+                input_ids.unsqueeze(0), torch_logits.unsqueeze(0)
             ).squeeze(0)
 
         # return logits as passed array type
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index d9a912c89..e3b9e60d3 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -70,7 +70,7 @@ def __init__(self, tokenizer: "Tokenizer", guide: Guide):
         self._seq_start_idx = None
 
     def process_logits(
-        self, input_ids: List[List[int]], logits: torch.Tensor
+        self, input_ids: torch.LongTensor, logits: torch.FloatTensor
     ) -> torch.Tensor:
         """Use the Guide to bias the logits before sampling the next token.
 
@@ -93,19 +93,32 @@ def process_logits(
 
         for seq_ids in input_ids:
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(gen_ids))
+            curr_state_key = hash(tuple(gen_ids.tolist()))
 
             if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(gen_ids[:-1]))]
-                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1])
+                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
+                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])
 
         mask = torch.ones_like(logits, dtype=torch.bool)
+
+        allowed_tokens_batch = []
+        batch_indices = []
         for i, guide_state in enumerate(sequence_states):
-            allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
-            mask[i, allowed_tokens] = False
+            allowed_tokens = self.guide.get_next_instruction(guide_state).tokens.to(
+                mask.device, non_blocking=True
+            )
+            allowed_tokens_batch.append(allowed_tokens)
+            batch_indices.append(
+                torch.full_like(allowed_tokens, i)
+            )  # Store batch index for each allowed token
+
+        allowed_tokens_concat = torch.cat(allowed_tokens_batch)
+        batch_indices_concat = torch.cat(batch_indices)
+
+        mask[batch_indices_concat, allowed_tokens_concat] = False
         logits.masked_fill_(mask, float("-inf"))
 
         return logits
@@ -202,7 +215,7 @@ def __init__(self, cfg_str: str, tokenizer: "Tokenizer"):
         super().__init__(tokenizer=tokenizer, guide=cfg_guide)
 
     def process_logits(
-        self, input_ids: List[List[int]], logits: torch.Tensor
+        self, input_ids: torch.LongTensor, logits: torch.Tensor
     ) -> torch.Tensor:
         """Same behavior as GuideLogitsProcessor, but uses rejection sampling"""
         if self._seq_start_idx is None:
@@ -212,11 +225,11 @@ def process_logits(
 
         for seq_ids in input_ids:
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(gen_ids))
+            curr_state_key = hash(tuple(gen_ids.tolist()))
 
             if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(gen_ids[:-1]))]
-                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1])
+                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
+                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])

From 9e8bd6cc22a003bc805ff259c2722a3b8155e06c Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 7 Oct 2024 11:41:43 -0400
Subject: [PATCH 089/505] ensure valid link to docs with latest/ in url

---
 README.md                    | 10 +++++-----
 docs/cookbook/simtom.md      |  4 ++--
 outlines/fsm/guide.py        |  2 +-
 outlines/models/exllamav2.py |  2 +-
 4 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/README.md b/README.md
index 9b01c5662..832703ce6 100644
--- a/README.md
+++ b/README.md
@@ -21,11 +21,11 @@ Made with ❤👷️ by the team at [.txt](https://dottxt.co).
 pip install outlines
 ```
 
-First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/welcome)
+First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/latest/welcome/)
 
 ## Features
 
-- [x] 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
+- [x] 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
 - [x] 🖍️ Simple and powerful prompting primitives based on the [Jinja templating engine](https://jinja.palletsprojects.com/)
 - [x] 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
 - [x] ⚡ Fast [regex-structured generation](#efficient-regex-structured-generation)
@@ -35,7 +35,7 @@ First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/we
 - [x] 💾 Caching of generations
 - [x] 🗂️ Batch inference
 - [x] 🎲 Sample with the greedy, multinomial and beam search algorithms (and more to come!)
-- [x] 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
+- [x] 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/latest/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
 
 
 Outlines  has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!
@@ -338,7 +338,7 @@ answer = outlines.generate.text(model)(prompt, max_tokens=100)
 ## Join us
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
-- 🔨 **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/community/contribute/).
+- 🔨 **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/latest/community/contribute/).
 - 🐞 **Found a bug?** Open an [issue](https://github.com/dottxt-ai/outlines/issues)
 
 
@@ -353,7 +353,7 @@ answer = outlines.generate.text(model)(prompt, max_tokens=100)
 }
 ```
 
-[documentation]: https://dottxt-ai.github.io/outlines/welcome/
+[documentation]: https://dottxt-ai.github.io/outlines/latest/welcome/
 [documentation-badge]: https://img.shields.io/readthedocs/outlines
 [contributors]: https://github.com/dottxt-ai/outlines/graphs/contributors
 [contributors-badge]: https://img.shields.io/github/contributors/dottxt-ai/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
diff --git a/docs/cookbook/simtom.md b/docs/cookbook/simtom.md
index f730d029a..4ad78846b 100644
--- a/docs/cookbook/simtom.md
+++ b/docs/cookbook/simtom.md
@@ -17,9 +17,9 @@ SimToM calls an LLM with two consecutive prompts:
 
 To implement SimToM with Outlines, we will need to:
 
-1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/reference/prompting/).
+1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
 2. Define the JSON object each prompt will return using Pydantic.
-3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/reference/models/transformers/).
+3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/latest/reference/models/transformers/).
 
 Let's dive into it!
 
diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index a3a2c7369..697597234 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -107,7 +107,7 @@ def __init__(self, cfg_string: str, tokenizer):
         """
         warnings.warn(
             "Outlines' public *community-contributed* CFG structured generation is experimental. "
-            "Please review https://dottxt-ai.github.io/outlines/reference/cfg#disclaimer"
+            "Please review https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
         )
 
         self.cfg_string = cfg_string
diff --git a/outlines/models/exllamav2.py b/outlines/models/exllamav2.py
index 78da796fb..d2aa84b0d 100644
--- a/outlines/models/exllamav2.py
+++ b/outlines/models/exllamav2.py
@@ -302,7 +302,7 @@ def exl2(
         raise ImportError(
             "The `exllamav2`, `transformers` and `torch` libraries needs to be installed in order to use `exllamav2` models. "
             "Please run `pip install transformers torch git+https://github.com/lapp0/exllamav2@sampler-logits-processor` "
-            "Documentation: https://dottxt-ai.github.io/outlines/reference/models/exllamav2/"
+            "Documentation: https://dottxt-ai.github.io/outlines/latest/reference/models/exllamav2/"
         )
     config = ExLlamaV2Config(model_path)
     if max_chunk_size is not None:

From a2fd35cfa58b9a9fb87e885a851af40019c10b62 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 7 Oct 2024 11:27:43 -0400
Subject: [PATCH 090/505] use  not  in mkdocs

---
 mkdocs.yml | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mkdocs.yml b/mkdocs.yml
index b200ee3d8..c0494c6ea 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -6,8 +6,8 @@ site_description: >-
 
 
 # Repository
-repo_name: outlines-dev/outlines
-repo_url: https://github.com/outlines-dev/outlines
+repo_name: dottxt-ai/outlines
+repo_url: https://github.com/dottxt-ai/outlines
 
 # Copyright
 copyright: Copyright &copy; 2023- The Outlines Developers
@@ -38,7 +38,7 @@ theme:
 extra:
   social:
     - icon: fontawesome/brands/github
-      link: https://github.com/outlines-dev
+      link: https://github.com/dottxt-ai
     - icon: fontawesome/brands/twitter
       link: https://twitter.com/remilouf
   generator: false
@@ -93,7 +93,7 @@ plugins:
       cards_layout_options:
         color: #173a58
   - git-committers:
-      repository: outlines-dev/outlines
+      repository: dottxt-ai/outlines
       branch: main
   - git-revision-date-localized:
       enable_creation_date: true

From c1fae8aca517d1289e3f934750cb6ea52804ad6d Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Thu, 10 Oct 2024 14:34:17 -0700
Subject: [PATCH 091/505] Create lmstudio.md

---
 docs/reference/serve/lmstudio.md | 89 ++++++++++++++++++++++++++++++++
 1 file changed, 89 insertions(+)
 create mode 100644 docs/reference/serve/lmstudio.md

diff --git a/docs/reference/serve/lmstudio.md b/docs/reference/serve/lmstudio.md
new file mode 100644
index 000000000..e8c336137
--- /dev/null
+++ b/docs/reference/serve/lmstudio.md
@@ -0,0 +1,89 @@
+# Serve with LM Studio
+
+!!! tip "Would rather not self-host?"
+
+    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
+
+[LM Studio](https://lmstudio.ai/) is an application that runs local LLMs. It flexibly mixes GPU and CPU compute in hardware-constrained environments.
+
+As of [LM Studio 0.3.4](https://lmstudio.ai/blog/lmstudio-v0.3.4), it natively supports Outlines for structured text generation, using an OpenAI-compatible endpoint.
+
+## Setup
+
+1. Install LM Studio by visiting their [downloads page](https://lmstudio.ai/download).
+2. Enable the LM Studio [server functionality](https://lmstudio.ai/docs/basics/server).
+3. Download [a model](https://lmstudio.ai/docs/basics#1-download-an-llm-to-your-computer).
+4. Install Python dependencies.
+```bash
+pip install pydantic openai
+```
+
+## Calling the server
+
+By default, LM Studio will serve from `http://localhost:1234`. If you are serving on a different port or host, make sure to change the `base_url` argument in `OpenAI` to the relevant location.
+
+```python
+class Testing(BaseModel):
+    """
+    A class representing a testing schema.
+    """
+    name: str
+    age: int
+
+openai_client = openai.OpenAI(
+    base_url="http://0.0.0.0:1234/v1",
+    api_key="dopeness"
+)
+
+# Make a request to the local LM Studio server
+response = openai_client.beta.chat.completions.parse(
+    model="hugging-quants/Llama-3.2-1B-Instruct-Q8_0-GGUF",
+    messages=[
+        {"role": "system", "content": "You are like so good at whatever you do."},
+        {"role": "user", "content": "My name is Cameron and I am 28 years old. What's my name and age?"}
+    ],
+    response_format=Testing
+)
+```
+
+You should receive a `ParsedChatCompletion[Testing]` object back:
+
+```python
+ParsedChatCompletion[Testing](
+    id='chatcmpl-3hykyf0fxus7jc90k6gwlw',
+    choices=[
+        ParsedChoice[Testing](
+            finish_reason='stop',
+            index=0,
+            logprobs=None,
+            message=ParsedChatCompletionMessage[Testing](
+                content='{ "age": 28, "name": "Cameron" }',
+                refusal=None,
+                role='assistant',
+                function_call=None,
+                tool_calls=[],
+                parsed=Testing(name='Cameron', age=28)
+            )
+        )
+    ],
+    created=1728595622,
+    model='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-Q4_K_M.gguf',
+    object='chat.completion',
+    service_tier=None,
+    system_fingerprint='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-
+Q4_K_M.gguf',
+    usage=CompletionUsage(
+        completion_tokens=17,
+        prompt_tokens=47,
+        total_tokens=64,
+        completion_tokens_details=None,
+        prompt_tokens_details=None
+    )
+)
+```
+
+You can retrieve your `Testing` object with 
+
+```python
+response.choices[0].message.parsed
+```

From 64fb30faae5201e4c9957770fbc19863c7d3750c Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Mon, 14 Oct 2024 18:43:42 -0700
Subject: [PATCH 092/505] use pre-commit

---
 docs/reference/serve/lmstudio.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/serve/lmstudio.md b/docs/reference/serve/lmstudio.md
index e8c336137..db383186a 100644
--- a/docs/reference/serve/lmstudio.md
+++ b/docs/reference/serve/lmstudio.md
@@ -82,7 +82,7 @@ Q4_K_M.gguf',
 )
 ```
 
-You can retrieve your `Testing` object with 
+You can retrieve your `Testing` object with
 
 ```python
 response.choices[0].message.parsed

From 866b9a3c25b88a1790228632063583701990e364 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Thu, 10 Oct 2024 15:03:04 -0400
Subject: [PATCH 093/505] recover fsm_union, get_sub_fsms_from_seq, walk_fsm.
 Add to fsm/parser.py

---
 outlines/fsm/parsing.py | 215 +++++++++++++++++++++++++++++++++++++++-
 1 file changed, 210 insertions(+), 5 deletions(-)

diff --git a/outlines/fsm/parsing.py b/outlines/fsm/parsing.py
index 92d3cc166..e48fb69e4 100644
--- a/outlines/fsm/parsing.py
+++ b/outlines/fsm/parsing.py
@@ -1,10 +1,22 @@
 from copy import copy, deepcopy
 from dataclasses import dataclass
 from functools import lru_cache
-from typing import Any, Dict, FrozenSet, Iterator, Optional, Set, Tuple, Union
+from typing import (
+    Any,
+    Dict,
+    FrozenSet,
+    Generator,
+    Iterator,
+    List,
+    Optional,
+    Sequence,
+    Set,
+    Tuple,
+    Union,
+)
 
 import interegular
-from interegular.fsm import FSM
+from interegular.fsm import FSM, Alphabet, OblivionError
 from interegular.patterns import Unsupported
 from lark import Lark, Token
 from lark.common import LexerConf, ParserConf
@@ -35,11 +47,9 @@
 from lark.parsers.lalr_interactive_parser import InteractiveParser
 from lark.parsers.lalr_parser import LALR_Parser, ParseConf, ParserState, _Parser
 from outlines_core.fsm.regex import (
-    fsm_union,
-    get_sub_fsms_from_seq,
+    BetterFSM,
     get_token_transition_keys,
     make_deterministic_fsm,
-    walk_fsm,
 )
 
 PartialParseState = Tuple[str, int]
@@ -920,3 +930,198 @@ def terminals_to_fsms(lp: PartialLark) -> Dict[str, FSM]:
         symbol_names_and_fsms[terminal.name] = fsm
 
     return symbol_names_and_fsms
+
+
+def fsm_union(
+    fsms: Sequence[FSM],
+) -> Tuple[FSM, Dict[int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]]]:
+    """Construct an FSM representing the union of the FSMs in `fsms`.
+
+    This is an updated version of `interegular.fsm.FSM.union` made to return an
+    extra map of component FSMs to the sets of state transitions that
+    correspond to them in the new FSM.
+
+    """
+
+    alphabet, new_to_old = Alphabet.union(*[fsm.alphabet for fsm in fsms])
+
+    indexed_fsms = tuple(enumerate(fsms))
+
+    initial = {i: fsm.initial for (i, fsm) in indexed_fsms}
+
+    # Dedicated function accepting a "superset" and returning the next
+    # "superset" obtained by following this transition in the new FSM
+    def follow(current_state, new_transition: int):
+        next = {}
+        for i, f in indexed_fsms:
+            old_transition = new_to_old[i][new_transition]
+            if (
+                i in current_state
+                and current_state[i] in f.map
+                and old_transition in f.map[current_state[i]]
+            ):
+                next[i] = f.map[current_state[i]][old_transition]
+        if not next:
+            raise OblivionError
+        return next
+
+    states = [initial]
+    finals: Set[int] = set()
+    map: Dict[int, Dict[int, int]] = {}
+
+    # Map component FSMs to their new state-to-state transitions, finals, and a
+    # map translating component FSM states to aggregate FSM states
+    fsms_to_trans_finals: Dict[
+        int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]
+    ] = {}
+
+    i = 0
+    while i < len(states):
+        state = states[i]
+
+        # Add to the finals of the aggregate FSM whenever we hit a final in a
+        # component FSM
+        if any(state.get(j, -1) in fsm.finals for (j, fsm) in indexed_fsms):
+            finals.add(i)
+
+        # Compute the map for this state
+        map[i] = {}
+        for transition in alphabet.by_transition:
+            try:
+                next = follow(state, transition)
+            except OblivionError:
+                # Reached an oblivion state; don't list it
+                continue
+            else:
+                try:
+                    # TODO: Seems like this could--and should--be avoided
+                    j = states.index(next)
+                except ValueError:
+                    j = len(states)
+                    states.append(next)
+
+                map[i][transition] = j
+
+                for fsm_id, fsm_state in next.items():
+                    (
+                        fsm_transitions,
+                        fsm_finals,
+                        fsm_old_to_new,
+                    ) = fsms_to_trans_finals.setdefault(fsm_id, (set(), set(), {}))
+                    old_from = state[fsm_id]
+                    old_to = fsm_state
+                    fsm_old_to_new.setdefault(old_from, set()).add(i)
+                    fsm_old_to_new.setdefault(old_to, set()).add(j)
+                    fsm_transitions.add((i, j))
+                    if fsm_state in fsms[fsm_id].finals:
+                        fsm_finals.add(j)
+
+        i += 1
+
+    fsm = FSM(
+        alphabet=alphabet,
+        states=range(len(states)),
+        initial=0,
+        finals=finals,
+        map=map,
+        __no_validation__=True,
+    )
+
+    fsm, old_to_new_states = make_deterministic_fsm(fsm)
+    _fsms_to_trans_finals = {
+        fsm_id: (
+            {(old_to_new_states[s1], old_to_new_states[s2]) for s1, s2 in transitions},
+            {old_to_new_states[s] for s in finals},
+            {
+                old_state: {old_to_new_states[new_state] for new_state in new_states}
+                for old_state, new_states in old_to_new.items()
+            },
+        )
+        for fsm_id, (transitions, finals, old_to_new) in sorted(
+            fsms_to_trans_finals.items(), key=lambda x: x[0]
+        )
+    }
+
+    return (
+        fsm,
+        _fsms_to_trans_finals,
+    )
+
+
+def get_sub_fsms_from_seq(
+    state_seq: Sequence[int],
+    fsms_to_trans_finals: Dict[
+        int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]
+    ],
+) -> Generator[Tuple[int, bool, bool], None, None]:
+    """Get the indices of the sub-FSMs in `fsm` that could have matched the state sequence `state_seq`.
+
+    Parameters
+    ----------
+    state_seq
+        A state sequence.
+    fsms_to_trans_finals
+        A map from FSM indices to tuples containing sets of their state transitions
+        and sets of the final/accept states.
+
+    Returns
+    -------
+    A generator returning tuples containing each sub-FSM index (in the order
+    they were union-ed to construct `fsm`) and booleans indicating whether or
+    not there is another valid transition from the last state in the sequence
+    for the associated sub-FSM (i.e. if the FSM can continue
+    accepting/matching) and whether or not the sequence ends in a final state
+    of the sub-FSM.
+    """
+    state_seq_transitions = set(zip(state_seq[:-1], state_seq[1:]))
+    last_fsm_state = state_seq[-1]
+    yield from (
+        (
+            # The sub-FMS index
+            fsm_idx,
+            # Is there another possible transition in this sub-FSM?
+            any(last_fsm_state == from_s for (from_s, to_s) in transitions),
+            # Is this sub-FSM in a final state?
+            state_seq[-1] in finals,
+        )
+        for fsm_idx, (transitions, finals, _) in fsms_to_trans_finals.items()
+        if state_seq_transitions.issubset(transitions)
+    )
+
+
+def walk_fsm(
+    fsm: BetterFSM,
+    token_transition_keys: Sequence[int],
+    start_state: int,
+    full_match: bool = True,
+) -> List[int]:
+    fsm_finals = fsm.finals
+
+    state = start_state
+    accepted_states: List[int] = []
+    last_final_idx: int = 0
+
+    fsm_transitions = fsm.flat_transition_map
+
+    # Iterate over token transition key sequence. The transition key
+    # sequence represents the FSM traversal rules of the tokens symbols.
+    for i, trans_key in enumerate(token_transition_keys):
+        new_state = fsm_transitions.get((state, trans_key))
+
+        if new_state is None:
+            if not full_match and last_final_idx > 0:
+                return accepted_states[:last_final_idx]
+
+            return []
+
+        state = new_state
+
+        if state in fsm_finals:
+            last_final_idx = i + 1
+
+        accepted_states.append(state)
+
+    if full_match and last_final_idx - 1 != i:
+        return []
+
+    return accepted_states

From eabca69013298145acd09dcda2eb4fc2d58d9c2d Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Thu, 10 Oct 2024 15:03:16 -0400
Subject: [PATCH 094/505] update dependencies: torch is required, pin
 outlines-core==0.1.14

---
 pyproject.toml | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index ac94ecf57..fa7005afd 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,8 @@ dependencies = [
    "typing_extensions",
    "pycountry",
    "airportsdata",
-   "outlines_core==0.1.0",
+   "torch",
+   "outlines_core==0.1.14",
 ]
 dynamic = ["version"]
 
@@ -61,7 +62,6 @@ test = [
     "huggingface_hub",
     "openai>=1.0.0",
     "vllm; sys_platform != 'darwin'",
-    "torch",
     "transformers",
     "pillow",
     "exllamav2",

From 6f36b71e995cec4ff17d6d4d4fca28dcf5122cef Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Fri, 11 Oct 2024 10:13:40 -0400
Subject: [PATCH 095/505] update RegexGuide to conform with outlines-core

---
 benchmarks/bench_json_schema.py               |  2 +-
 benchmarks/bench_regex_guide.py               |  4 +--
 outlines/fsm/guide.py                         | 26 +++++++++++--------
 outlines/processors/structured.py             |  2 +-
 tests/fsm/test_guide.py                       | 10 +++----
 tests/generate/test_integration_llamacpp.py   |  4 +--
 .../generate/test_integration_transformers.py |  4 +--
 7 files changed, 28 insertions(+), 24 deletions(-)

diff --git a/benchmarks/bench_json_schema.py b/benchmarks/bench_json_schema.py
index 8990b015c..62d9b3c1d 100644
--- a/benchmarks/bench_json_schema.py
+++ b/benchmarks/bench_json_schema.py
@@ -77,4 +77,4 @@ def time_json_schema_to_regex(self, schema_name):
     @cache_disabled()
     def time_json_schema_to_fsm(self, schema_name):
         regex = build_regex_from_schema(self.schema)
-        RegexGuide(regex, self.tokenizer)
+        RegexGuide.from_regex(regex, self.tokenizer)
diff --git a/benchmarks/bench_regex_guide.py b/benchmarks/bench_regex_guide.py
index 7aaef6bac..fa23a724f 100644
--- a/benchmarks/bench_regex_guide.py
+++ b/benchmarks/bench_regex_guide.py
@@ -25,7 +25,7 @@ def setup(self, pattern_name):
 
     @cache_disabled()
     def time_regex_to_guide(self, pattern_name):
-        RegexGuide(self.pattern, self.tokenizer)
+        RegexGuide.from_regex(self.pattern, self.tokenizer)
 
 
 class MemoryRegexGuideBenchmark:
@@ -37,4 +37,4 @@ def setup(self, pattern_name):
 
     @cache_disabled()
     def peakmem_regex_to_guide(self, pattern_name):
-        RegexGuide(self.pattern, self.tokenizer)
+        RegexGuide.from_regex(self.pattern, self.tokenizer)
diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index 697597234..d46228fe9 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -74,8 +74,8 @@ def copy(self):
 
 
 @cache()
-def create_states_mapping(regex_string, tokenizer):
-    return uncached_create_states_mapping(regex_string, tokenizer)
+def cached_create_states_mapping(regex_string, tokenizer, *args, **kwargs):
+    return uncached_create_states_mapping(regex_string, tokenizer, *args, **kwargs)
 
 
 class RegexGuide(CoreRegexGuide):
@@ -84,15 +84,19 @@ class RegexGuide(CoreRegexGuide):
     CoreRegexGuide with outlines cache
     """
 
-    def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
-        (
-            self.states_to_token_maps,
-            self.empty_token_ids,
-            fsm_finals,
-        ) = create_states_mapping(regex_string, tokenizer)
-        self.eos_token_id = tokenizer.eos_token_id
-        self.final_states = fsm_finals | {-1}
-        self._cache_state_to_token_tensor()
+    @classmethod
+    def from_regex(
+        cls,
+        regex_string: str,
+        tokenizer,
+        **kwargs,
+    ):
+        return super().from_regex(
+            regex_string,
+            tokenizer,
+            _create_states_mapping=cached_create_states_mapping,
+            **kwargs,
+        )
 
 
 CFGState = collections.namedtuple("CFGState", ["parser_state", "prev_token"])
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index e3b9e60d3..d2bc15f77 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -149,7 +149,7 @@ def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
         tokenizer
             An Outlines tokenizer
         """
-        guide = RegexGuide(regex_string, tokenizer)
+        guide = RegexGuide.from_regex(regex_string, tokenizer)
         super().__init__(tokenizer=tokenizer, guide=guide)
 
 
diff --git a/tests/fsm/test_guide.py b/tests/fsm/test_guide.py
index 67b4e0dd8..510faf4b0 100644
--- a/tests/fsm/test_guide.py
+++ b/tests/fsm/test_guide.py
@@ -43,7 +43,7 @@ def convert_token_to_string(self, token):
     regex_str = "[1-9]"
 
     with pytest.raises(ValueError, match="The vocabulary"):
-        RegexGuide(regex_str, MockTokenizer())
+        RegexGuide.from_regex(regex_str, MockTokenizer())
 
 
 def test_regex():
@@ -57,7 +57,7 @@ def convert_token_to_string(self, token):
 
     regex_str = "[1-9]"
     tokenizer = MockTokenizer()
-    fsm = RegexGuide(regex_str, tokenizer)
+    fsm = RegexGuide.from_regex(regex_str, tokenizer)
 
     assert fsm.states_to_token_maps == {0: {1: 1}}
 
@@ -98,7 +98,7 @@ def convert_token_to_string(self, token):
 
     regex_str = "[😁-😎]"
     tokenizer = MockTokenizer()
-    fsm = RegexGuide(regex_str, tokenizer)
+    fsm = RegexGuide.from_regex(regex_str, tokenizer)
 
     assert fsm.states_to_token_maps == {
         0: {5: 1, 4: 2},
@@ -145,7 +145,7 @@ def convert_token_to_string(self, token):
 
     regex_str = " [😁-😎]"
     tokenizer = MockTokenizer()
-    fsm = RegexGuide(regex_str, tokenizer)
+    fsm = RegexGuide.from_regex(regex_str, tokenizer)
 
     assert fsm.states_to_token_maps == {
         0: {5: 1, 10: 2},
@@ -180,7 +180,7 @@ def convert_token_to_string(self, token):
 
     regex_str = r"`\n(\.\n)?`\n"
     tokenizer = MockTokenizer()
-    fsm = RegexGuide(regex_str, tokenizer)
+    fsm = RegexGuide.from_regex(regex_str, tokenizer)
 
     state = fsm.get_next_state(state=4, token_id=103)
     assert state == 5
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
index 08521c672..8d4596d60 100644
--- a/tests/generate/test_integration_llamacpp.py
+++ b/tests/generate/test_integration_llamacpp.py
@@ -278,7 +278,7 @@ def test_RegexGuide_caching(model, temp_cache_dir):
     import llama_cpp
 
     import outlines.caching
-    from outlines.fsm.guide import create_states_mapping
+    from outlines.fsm.guide import cached_create_states_mapping
 
     assert outlines.caching._caching_enabled
 
@@ -291,7 +291,7 @@ def test_RegexGuide_caching(model, temp_cache_dir):
     _ = cache.stats(enable=True)
     assert cache.statistics
 
-    assert create_states_mapping.__memory__ is cache
+    assert cached_create_states_mapping.__memory__ is cache
 
     generator = generate.regex(model, regex, sampler=samplers.greedy())
     assert cache.stats() == (0, 1)
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index 1d26a9ee4..2462d9fcf 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -494,7 +494,7 @@ def test_transformers_use_existing_model_and_tokenizer():
 
 def test_RegexGuide_caching(temp_cache_dir):
     import outlines.caching
-    from outlines.fsm.guide import create_states_mapping
+    from outlines.fsm.guide import cached_create_states_mapping
 
     assert outlines.caching._caching_enabled
 
@@ -507,7 +507,7 @@ def test_RegexGuide_caching(temp_cache_dir):
     _ = cache.stats(enable=True)
     assert cache.statistics
 
-    assert create_states_mapping.__memory__ is cache
+    assert cached_create_states_mapping.__memory__ is cache
 
     model = models.transformers(
         "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM", device="cpu"

From 969887ee3b626380921fa0cb8f6360cb11ff3ed9 Mon Sep 17 00:00:00 2001
From: Andrew Lapp <andrew@github.rew.la>
Date: Mon, 14 Oct 2024 13:03:26 -0400
Subject: [PATCH 096/505] test fsm_union and walk_fsm

---
 tests/fsm/test_parsing.py | 101 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 101 insertions(+)

diff --git a/tests/fsm/test_parsing.py b/tests/fsm/test_parsing.py
index 3f4c1ba42..b7446fa0c 100644
--- a/tests/fsm/test_parsing.py
+++ b/tests/fsm/test_parsing.py
@@ -204,3 +204,104 @@ def test_sequential_parse_example(cleanup_lark_import):
 
         if i + 1 == len(input_tokens):
             assert all(tk in next_vocab for tk in ["\n", "\nde", "  ", " + 1"])
+
+
+# TODO: Remove once fsm_union and walk_fsm are implemented in Outlines-Core
+import interegular  # noqa
+
+from outlines.fsm.parsing import fsm_union, walk_fsm  # noqa
+
+
+def test_outlines_interegular_union_consistency():
+    fsm0 = interegular.parse_pattern(r"abc").to_fsm()
+    fsm1 = interegular.parse_pattern(r"WXYZ").to_fsm()
+    fsm2 = interegular.parse_pattern(r"12345").to_fsm()
+
+    interegular_unioned_fsm = fsm0 | fsm1 | fsm2
+    outlines_unioned_fsm, _ = fsm_union([fsm0, fsm1, fsm2])
+
+    assert list(outlines_unioned_fsm.strings()) == list(
+        interegular_unioned_fsm.strings()
+    )
+
+
+def _reconstruct_fsms(fsm, fsms_to_trans_finals):
+    """Reconstruct the original fsms for testing purposes"""
+    reconstructed_fsms = []
+    for transitions, finals, state_map in fsms_to_trans_finals.values():
+        inv_state_map = {new: orig for orig, news in state_map.items() for new in news}
+        states = set(inv_state_map.values())
+        initial = inv_state_map.get(fsm.initial) or next(
+            (orig for orig, news in state_map.items() if fsm.initial in news), None
+        )
+        finals = {inv_state_map[s] for s in finals}
+
+        transition_map = {}
+        alphabet = {}
+        for trans_id, (from_state, to_state) in enumerate(transitions):
+            orig_from, orig_to = inv_state_map[from_state], inv_state_map[to_state]
+            # Collect symbols associated with the transition
+            symbols = {
+                symbol
+                for trans, dest in fsm.map.get(from_state, {}).items()
+                if dest == to_state
+                for symbol in fsm.alphabet.by_transition.get(trans, [])
+            }
+            if symbols:
+                # NOTE: THIS RECONSTRUCTOR DOESNT WORK FOR MORE THAN ONE TRANSITION PER SYMBOL
+                assert len(symbols) == 1
+                symbol = list(symbols)[0]
+                alphabet[symbol] = trans_id
+                transition_map.setdefault(orig_from, {})[trans_id] = orig_to
+
+        reconstructed_fsms.append(
+            interegular.fsm.FSM(
+                alphabet=interegular.fsm.Alphabet(alphabet),
+                states=frozenset(states),
+                initial=initial,
+                finals=frozenset(finals),
+                map=transition_map,
+                __no_validation__=True,
+            )
+        )
+    return reconstructed_fsms
+
+
+def test_fsm_to_trans_finals_reconstruction():
+    """Assert that _fsms_to_trans_finals is correct by reconstructing original fsms"""
+    fsm0 = interegular.parse_pattern(r"abc").to_fsm()
+    fsm1 = interegular.parse_pattern(r"XYZ").to_fsm()
+    fsm2 = interegular.parse_pattern(r"12345").to_fsm()
+
+    fsm, _fsms_to_trans_finals = fsm_union([fsm0, fsm1, fsm2])
+
+    reconstructed = _reconstruct_fsms(fsm, _fsms_to_trans_finals)
+
+    # assert reconstruction equivalent
+    assert list(fsm0.strings()) == list(reconstructed[0].strings())
+    assert list(fsm1.strings()) == list(reconstructed[1].strings())
+    assert list(fsm2.strings()) == list(reconstructed[2].strings())
+
+
+def test_walk_fsm():
+    fsm = interegular.parse_pattern(r"abc*d").to_fsm()
+    # convert to BetterFSM
+    fsm = fsm_union([fsm])[0]
+
+    # if match, produce equivalent number of states, assert state can terminate
+    transitions = [fsm.alphabet[letter] for letter in "abcccd"]
+    accepted_states = walk_fsm(fsm, transitions, fsm.initial, full_match=True)
+    assert len(accepted_states) == len(transitions)
+    assert accepted_states[-1] in fsm.finals
+
+    # if no match, assert empty
+    accepted_states = walk_fsm(
+        fsm, [fsm.alphabet[letter] for letter in "b"], fsm.initial, full_match=True
+    )
+    assert accepted_states == []
+
+    # if full_match, but last state not present, assert empty
+    accepted_states = walk_fsm(
+        fsm, [fsm.alphabet[letter] for letter in "abc"], fsm.initial, full_match=True
+    )
+    assert accepted_states == []

From 6cff654ad7bb9b1fcb29b244c52987c1e29b782a Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Fri, 18 Oct 2024 03:43:54 -0700
Subject: [PATCH 097/505] Fix logo color and doc links (#1213)

- Will's structured generation workflow cookbook example was not in the
mkdocs index, so it was not being displayed.
- Same with the LM Studio serving docs.
- The brand color was also slightly off:


![image](https://github.com/user-attachments/assets/fd10fa4f-d140-4936-befa-4dcca09c0e51)

It has been fixed to this:


![image](https://github.com/user-attachments/assets/b6c2d71b-6a7f-4b86-935a-bf5072f1d945)
---
 docs/stylesheets/extra.css | 2 +-
 mkdocs.yml                 | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index 7c5b5e808..33783c153 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -9,7 +9,7 @@
   --md-code-fg-color: #FFFFFF;
   --md-text-font-family: "Inter";
   --md-code-font: "Source Code Pro Custom";
-  --md-typeset-a-color: #d53135; /*this is the brand color*/
+  --md-typeset-a-color: #d16626; /*this is the brand color*/
 
   /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
   --md-mermaid-label-fg-color: #000000;
diff --git a/mkdocs.yml b/mkdocs.yml
index c0494c6ea..83106c0df 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -116,6 +116,7 @@ nav:
       - Perspective-taking prompting: cookbook/simtom.md
       - Question-answering with citations: cookbook/qa-with-citations.md
       - Knowledge Graph Extraction: cookbook/knowledge_graph_extraction.md
+      - Structured Generation Workflow: cookbook/structured_generation_workflow.md
       - Chain of Thought (CoT): cookbook/chain_of_thought.md
       - ReAct Agent: cookbook/react_agent.md
       - Run on the cloud:
@@ -139,6 +140,7 @@ nav:
             - Custom FSM operations: reference/generation/custom_fsm_ops.md
     - Utilities:
         - Serve with vLLM: reference/serve/vllm.md
+        - Serve with LM Studio: reference/serve/lmstudio.md
         - Custom types: reference/generation/types.md
         - Prompt templating: reference/prompting.md
         - Outlines functions: reference/functions.md

From a2fa1e0f44e80dd287c215fffde342578e67b1d1 Mon Sep 17 00:00:00 2001
From: jphillips <120260158+fearnworks@users.noreply.github.com>
Date: Mon, 21 Oct 2024 05:59:29 -0500
Subject: [PATCH 098/505] Add transformers vision cookbook with atomic caption
 flow (#1216)

Request received in discord to add an example for the new transformers
vision capability.


# Vision-Language Models with Outlines
This guide demonstrates how to use Outlines with vision-language models,
leveraging the new transformers_vision module. Vision-language models
can process both text and images, allowing for tasks like image
captioning, visual question answering, and more.

We will be using the Pixtral-12B model from Mistral to take advantage of
some of its visual reasoning capabilities and a workflow to generate a
multistage atomic caption.

---------

Signed-off-by: jphillips <josh.phillips@fearnworks.com>
---
 docs/cookbook/atomic_caption.md | 189 ++++++++++++++++++++++++++++++++
 docs/cookbook/index.md          |   1 +
 mkdocs.yml                      |   1 +
 3 files changed, 191 insertions(+)
 create mode 100644 docs/cookbook/atomic_caption.md

diff --git a/docs/cookbook/atomic_caption.md b/docs/cookbook/atomic_caption.md
new file mode 100644
index 000000000..a2b43e5e2
--- /dev/null
+++ b/docs/cookbook/atomic_caption.md
@@ -0,0 +1,189 @@
+# Vision-Language Models with Outlines
+This guide demonstrates how to use Outlines with vision-language models, leveraging the new transformers_vision module. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
+
+We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
+
+## Setup
+First, we need to install the necessary dependencies. In addition to Outlines, we'll need to install the transformers library and any specific requirements for the vision-language model we'll be using.
+
+```bash
+pip install outlines transformers torch
+```
+
+### Initializing the Model
+We'll use the transformers_vision function to initialize our vision-language model. This function is specifically designed to handle models that can process both text and image inputs. Today we'll be using the Pixtral model with the llama tokenizer. (Currently the mistral tokenizer is pending support).
+
+```python
+import torch
+from transformers import (
+    LlavaForConditionalGeneration,
+)
+model_name="mistral-community/pixtral-12b" # original magnet model is able to be loaded without issue
+model_class=LlavaForConditionalGeneration
+
+def get_vision_model(model_name: str, model_class: VisionModel):
+    model_kwargs = {
+        "torch_dtype": torch.bfloat16,
+        "attn_implementation": "flash_attention_2",
+        "device_map": "auto",
+    }
+    processor_kwargs = {
+        "device": "cuda",
+    }
+
+    model = outlines.models.transformers_vision(
+        model.model_name,
+        model_class=model.model_class,
+        model_kwargs=model_kwargs,
+        processor_kwargs=processor_kwargs,
+    )
+    return model
+model = get_vision_model(model_name, model_class)
+```
+
+### Defining the Schema
+Next, we'll define a schema for the output we expect from our vision-language model. This schema will help structure the model's responses.
+
+
+```python
+from pydantic import BaseModel, Field, confloat, constr
+from pydantic.types import StringConstraints, PositiveFloat
+from typing import List
+from typing_extensions import Annotated
+
+from enum import StrEnum
+class TagType(StrEnum):
+    ENTITY = "Entity"
+    RELATIONSHIP = "Relationship"
+    STYLE = "Style"
+    ATTRIBUTE = "Attribute"
+    COMPOSITION = "Composition"
+    CONTEXTUAL = "Contextual"
+    TECHNICAL = "Technical"
+    SEMANTIC = "Semantic"
+
+class ImageTag(BaseModel):
+    tag: Annotated[
+        constr(min_length=1, max_length=30),
+        Field(
+            description=(
+                "Descriptive keyword or phrase representing the tag."
+            )
+        )
+    ]
+    category: TagType
+    confidence: Annotated[
+        confloat(le=1.0),
+        Field(
+            description=(
+                "Confidence score for the tag, between 0 (exclusive) and 1 (inclusive)."
+            )
+        )
+    ]
+
+class ImageData(BaseModel):
+    tags_list: List[ImageTag] = Field(..., min_items=8, max_items=20)
+    short_caption: Annotated[str, StringConstraints(min_length=10, max_length=150)]
+    dense_caption: Annotated[str, StringConstraints(min_length=100, max_length=2048)]
+
+image_data_generator = outlines.generate.json(model, ImageData)
+```
+
+This schema defines the structure for image tags, including categories like Entity, Relationship, Style, etc., as well as short and dense captions.
+
+### Preparing the Prompt
+
+We'll create a prompt that instructs the model on how to analyze the image and generate the structured output:
+
+
+```python
+pixtral_instruction = """
+<s>[INST]
+<Task>You are a structured image analysis agent. Generate comprehensive tag list, caption, and dense caption for an image classification system.</Task>
+<TagCategories requirement="You should generate a minimum of 1 tag for each category." confidence="Confidence score for the tag, between 0 (exclusive) and 1 (inclusive).">
+- Entity : The content of the image, including the objects, people, and other elements.
+- Relationship : The relationships between the entities in the image.
+- Style : The style of the image, including the color, lighting, and other stylistic elements.
+- Attribute : The most important attributes of the entities and relationships in the image.
+- Composition : The composition of the image, including the arrangement of elements.
+- Contextual : The contextual elements of the image, including the background, foreground, and other elements.
+- Technical : The technical elements of the image, including the camera angle, lighting, and other technical details.
+- Semantic : The semantic elements of the image, including the meaning of the image, the symbols, and other semantic details.
+<Examples note="These show the expected format as an abstraction.">
+{
+  "tags_list": [
+    {
+      "tag": "subject 1",
+      "category": "Entity",
+      "confidence": 0.98
+    },
+    {
+      "tag": "subject 2",
+      "category": "Entity",
+      "confidence": 0.95
+    },
+    {
+      "tag": "subject 1 runs from subject 2",
+      "category": "Relationship",
+      "confidence": 0.90
+    },
+   }
+</Examples>
+</TagCategories>
+<ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
+<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG][/INST]
+""".strip()
+```
+
+This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required.
+
+### Generating Structured Output
+Now we can use our model to generate structured output based on an input image:
+
+```python
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+image_url="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
+image= img_from_url(image_url)
+result = image_data_generator(
+    pixtral_instruction,
+    [image]
+)
+print(result)
+```
+
+This code loads an image from a URL, passes it to our vision-language model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
+
+```json
+{'tags_list': [{'tag': 'astronaut',
+   'category': <TagType.ENTITY: 'Entity'>,
+   'confidence': 0.99},
+  {'tag': 'moon', 'category': <TagType.ENTITY: 'Entity'>, 'confidence': 0.98},
+  {'tag': 'space suit',
+   'category': <TagType.ATTRIBUTE: 'Attribute'>,
+   'confidence': 0.97},
+  {'tag': 'lunar module',
+   'category': <TagType.ENTITY: 'Entity'>,
+   'confidence': 0.95},
+  {'tag': 'shadow of astronaut',
+   'category': <TagType.COMPOSITION: 'Composition'>,
+   'confidence': 0.95},
+  {'tag': 'footprints in moon dust',
+   'category': <TagType.CONTEXTUAL: 'Contextual'>,
+   'confidence': 0.93},
+  {'tag': 'low angle shot',
+   'category': <TagType.TECHNICAL: 'Technical'>,
+   'confidence': 0.92},
+  {'tag': 'human first steps on the moon',
+   'category': <TagType.SEMANTIC: 'Semantic'>,
+   'confidence': 0.95}],
+ 'short_caption': 'First man on the Moon',
+ 'dense_caption': "The figure clad in a pristine white space suit, emblazoned with the American flag, stands powerfully on the moon's desolate and rocky surface. The lunar module, a workhorse of space engineering, looms in the background, its metallic legs sinking slightly into the dust where footprints and tracks from the mission's journey are clearly visible. The photograph captures the astronaut from a low angle, emphasizing his imposing presence against the desolate lunar backdrop. The stark contrast between the blacks and whiteslicks of lost light and shadow adds dramatic depth to this seminal moment in human achievement."}
+```
+
+## Conclusion
+The transformers_vision module in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
+
+By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index a844ce240..b163feb62 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -12,3 +12,4 @@ This part of the documentation provides a few cookbooks that you can browse to g
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
+- [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
diff --git a/mkdocs.yml b/mkdocs.yml
index 83106c0df..146a766a2 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -119,6 +119,7 @@ nav:
       - Structured Generation Workflow: cookbook/structured_generation_workflow.md
       - Chain of Thought (CoT): cookbook/chain_of_thought.md
       - ReAct Agent: cookbook/react_agent.md
+      - Vision-Language Models: cookbook/atomic_caption.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From dc31b9b4027e71bd69f213bd3ab7d98b2f0a1529 Mon Sep 17 00:00:00 2001
From: Ikko Eltociear Ashimine <eltociear@gmail.com>
Date: Sun, 27 Oct 2024 22:31:21 +0900
Subject: [PATCH 099/505] Update llamacpp.md (#1231)

accross -> across
---
 docs/reference/models/llamacpp.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
index 0ab3dd83b..24b0fdc97 100644
--- a/docs/reference/models/llamacpp.md
+++ b/docs/reference/models/llamacpp.md
@@ -47,7 +47,7 @@ model = models.llamacpp(
 | `n_gpu_layers`| `int` | Number of layers to offload to GPU. If -1, all layers are offloaded | `0` |
 | `split_mode` | `int` | How to split the model across GPUs. `1` for layer-wise split, `2` for row-wise split | `1` |
 | `main_gpu` | `int` | Main GPU | `0` |
-| `tensor_split` | `Optional[List[float]]` | How split tensors should be distributed accross GPUs. If `None` the model is not split. | `None` |
+| `tensor_split` | `Optional[List[float]]` | How split tensors should be distributed across GPUs. If `None` the model is not split. | `None` |
 | `n_ctx` | `int` | Text context. Inference from the model if set to `0` | `0` |
 | `n_threads` | `Optional[int]` | Number of threads to use for generation. All available threads if set to `None`.| `None` |
 | `verbose` | `bool` | Print verbose outputs to `stderr` | `False` |

From 5f39ded4b872d60de563658f1d7a2642de0e8c4e Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Wed, 6 Nov 2024 00:54:59 -0800
Subject: [PATCH 100/505] Earnings report cookbook (#1235)

This is a condensed version of the demo for [extracting earnings
reports](https://github.com/dottxt-ai/demos/tree/main/earnings-reports)
to CSV.

Overview:

- Shows how to use Outlines to structure CSV output
- Provides simple tools for converting a table specification to regular
expressions
- Includes a tuned extraction prompt that performs reasonably well on
income statements
---
 docs/cookbook/earnings-reports.md      | 278 +++++++++++++++++++++++++
 docs/cookbook/images/nvidia-income.png | Bin 0 -> 185563 bytes
 mkdocs.yml                             |   1 +
 3 files changed, 279 insertions(+)
 create mode 100644 docs/cookbook/earnings-reports.md
 create mode 100644 docs/cookbook/images/nvidia-income.png

diff --git a/docs/cookbook/earnings-reports.md b/docs/cookbook/earnings-reports.md
new file mode 100644
index 000000000..26c2a9ccf
--- /dev/null
+++ b/docs/cookbook/earnings-reports.md
@@ -0,0 +1,278 @@
+# Extracting financial data from earnings reports
+
+A common task in finance is to extract financial data from earnings reports. Earnings reports are infamously poorly formatted, as the SEC does not have requirements for producing machine-readable documents.
+
+Earnings reports are often provided as HTML documents, which can be difficult to parse. Investors often use complicated parsing systems or manual review to extract data. Entire companies are built around automating this task.
+
+This cookbook is a proof of concept about how we can use LLMs to extract financial data directly into CSV. Comma-separated values are well-structured and can be defined by a regular expression, which Outlines can use to guide the LLM's output.
+
+The example is a smaller subset of a full demo found [here](https://github.com/dottxt-ai/demos/tree/main/earnings-reports). The demo contains the full set of pre-processing steps needed to convert raw HTML into a structured CSV file, and tests the results across three company's 10k reports.
+
+## Setup
+
+Install outlines and required dependencies:
+
+```bash
+# Later versions of torch can have difficulty with certain CUDA drivers.
+# We recommend using 2.4.0 for now, but you may wish to experiment with
+# other versions.
+pip install outlines pandas transformers torch==2.4.0 accelerate
+```
+
+## Load the model
+
+Choose your language model. We'll use Phi-3 mini, which is small enough to run on reasonably small machines.
+
+```python
+import outlines
+import torch
+
+model_name = 'microsoft/Phi-3-mini-4k-instruct'
+model = outlines.models.transformers(
+    model_name,
+    device='auto',
+    model_kwargs={
+        # To reduce memory usage, we'll use bfloat16
+        "torch_dtype": torch.bfloat16,
+    },
+)
+```
+
+## Set up the data
+
+For brevity, we've attached the markdown version of Nvidia's 10k report. The [full demonstration](https://github.com/dottxt-ai/demos/tree/main/earnings-reports) processes the raw HTML version of the report to these markdown tables. Pages are filtered by whether they seem to contain income statements, and then compacted into the string you see below.
+
+```python
+income_statement = """
+Table of ContentsNVIDIA Corporation and SubsidiariesConsolidated Statements of Income(In millions, except per share data)
+
+|  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |
+| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
+|  | | | Year Ended | | | | | | | | | | | | | | |
+|  | | | Jan 28, 2024 | | |  | | | Jan 29, 2023 | | |  | | | Jan 30, 2022 | | |
+| Revenue | | | $ | 60,922 |  |  | | | $ | 26,974 |  |  | | | $ | 26,914 |  |
+| Cost of revenue | | | 16,621 | |  |  | | | 11,618 | |  |  | | | 9,439 | |  |
+| Gross profit | | | 44,301 | |  |  | | | 15,356 | |  |  | | | 17,475 | |  |
+| Operating expenses | | |  | | |  | | |  | | |  | | |  | | |
+| Research and development | | | 8,675 | |  |  | | | 7,339 | |  |  | | | 5,268 | |  |
+| Sales, general and administrative | | | 2,654 | |  |  | | | 2,440 | |  |  | | | 2,166 | |  |
+| Acquisition termination cost | | | — | |  |  | | | 1,353 | |  |  | | | — | |  |
+| Total operating expenses | | | 11,329 | |  |  | | | 11,132 | |  |  | | | 7,434 | |  |
+| Operating income | | | 32,972 | |  |  | | | 4,224 | |  |  | | | 10,041 | |  |
+| Interest income | | | 866 | |  |  | | | 267 | |  |  | | | 29 | |  |
+| Interest expense | | | (257) | |  |  | | | (262) | |  |  | | | (236) | |  |
+| Other, net | | | 237 | |  |  | | | (48) | |  |  | | | 107 | |  |
+| Other income (expense), net | | | 846 | |  |  | | | (43) | |  |  | | | (100) | |  |
+| Income before income tax | | | 33,818 | |  |  | | | 4,181 | |  |  | | | 9,941 | |  |
+| Income tax expense (benefit) | | | 4,058 | |  |  | | | (187) | |  |  | | | 189 | |  |
+| Net income | | | $ | 29,760 |  |  | | | $ | 4,368 |  |  | | | $ | 9,752 |  |
+|  | | |  | | |  | | |  | | |  | | |  | | |
+| Net income per share: | | |  | | |  | | |  | | |  | | |  | | |
+| Basic | | | $ | 12\.05 |  |  | | | $ | 1\.76 |  |  | | | $ | 3\.91 |  |
+| Diluted | | | $ | 11\.93 |  |  | | | $ | 1\.74 |  |  | | | $ | 3\.85 |  |
+|  | | |  | | |  | | |  | | |  | | |  | | |
+| Weighted average shares used in per share computation: | | |  | | |  | | |  | | |  | | |  | | |
+| Basic | | | 2,469 | |  |  | | | 2,487 | |  |  | | | 2,496 | |  |
+| Diluted | | | 2,494 | |  |  | | | 2,507 | |  |  | | | 2,535 | |  |
+"""
+```
+
+The markdown tables extracted from the earnings reports can vary widely in row names, column counts, data types, etc. The advantage of LLMs here is that we can define the data we want in terms of the data types, and the LLM will output the data in the desired format.
+
+For comparison, here is how the income statement looks in the original HTML:
+
+![Nvidia income statement](./images/nvidia-income.png)
+
+## Define the data we want
+
+Outlines is often used for JSON output, but it can also be used for CSV. We know the columns we want to extract, and we know the data types of the columns. Year for example is always a four-digit number, revenue is a number with commas, and so on.
+
+We can define a regex pattern for each column type:
+
+```python
+# Define the column type regex patterns
+column_types = {
+    # Year is always a four-digit number
+    "year": r"\d{4}",
+
+    # Revenue, operating income, and net income are always numbers with commas.
+    # This regex permits integers that may begin with a minus sign, and may have
+    # commas separating the thousands, millions, etc.
+    "integer_comma": r"((-?\d+),?\d+|(-?\d+))",
+    # Number is currently not used, but it represents a number with up to two decimal places.
+    "number": r"(-?\d+(?:\.\d{1,2})?)",
+}
+```
+
+Next, let's choose the columns we want to extract. We want
+
+- Year, always a four-digit number
+- Revenue, a number with commas
+- Operating income, a number with commas
+- Net income, a number with commas
+
+```python
+# Define the columns to extract, and their data types.
+columns_to_extract = {
+    "year": "year",
+    "revenue": "integer_comma",
+    "operating_income": "integer_comma",
+    "net_income": "integer_comma",
+}
+```
+
+You can modify `column_type_regex` to match the data types of the columns you want to extract.  Adding a new financial metric to extract is as simple as adding a new key/value pair to `columns_to_extract`:
+
+```python
+columns_to_extract["diluted_earnings_per_share"] = "number"
+```
+
+Additional columns are not well tested for accuracy, so use with caution.
+
+## Create the regex describing the data we want
+
+
+```python
+# Create the header line. This is the requested column names
+# separated by commas, i.e. "year,revenue,..."
+header = ",".join(columns_to_extract.keys())
+
+# Create the data capture patterns. These are the regex patterns
+# that will be used to capture the data in each column
+data_patterns = [column_types[dtype] for dtype in columns_to_extract.values()]
+data_line = ",".join(data_patterns)
+
+# Our final regex pattern.
+max_rows = 3 # We expect 3 rows of data, firms usually report 3 years of income statements
+csv_regex = f"{header}(\n{data_line}){{,{max_rows}}}\n\n"
+
+print(csv_regex)
+```
+
+which gives us
+
+```
+year,revenue,operating_income,net_income,basic_earnings_per_share(
+\d{4},((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),(-?\d+(?:\.\d{1,2})?)){,3}
+```
+
+Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.generate.regex` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
+
+## Prompting the model
+
+Outlines does not add system or instruction tokens by default, so we need to use `transformers.AutoTokenizer` to add them for whatever model we're using.
+
+```python
+from transformers import AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+def add_instruction(prompt):
+    return tokenizer.apply_chat_template([{"role": "user", "content": prompt}], tokenize=False, add_generation_prompt=True)
+
+print(add_instruction("Howdy"))
+```
+```
+<|user|>
+Howdy<|end|>
+<|assistant|>
+```
+
+Our prompt roughly describes the task we want the model to perform, and a few pieces of information it may need to know about income statements.
+
+```python
+def extract_financial_data_prompt(columns_to_extract, income_statement):
+    user_prompt = f"""
+    Extract annual financial data from this set of pages. Pages
+    are from a 10k filing and were chosen because they may contain
+    a comprehensive income statement. Note that selected pages may
+    be incorrectly extracted, so you should verify that you are extracting
+    from the comprehensive income statement and not some other financial
+    statement.
+
+    Create a row for each year available in the income statement with the
+    following columns: {', '.join(columns_to_extract.keys())}. Firms typically report the
+    most recent 3 years of data, but this can vary.
+
+    Each column has types: {', '.join(columns_to_extract.values())}.
+
+    # Relevant pages:
+
+    {income_statement}
+
+    # Key instructions:
+
+    1. Look ONLY at the "Consolidated Statements of Income" table
+    2. For operating income, look for "Income from operations" or "Operating income"
+    3. For net income, use the TOTAL net income figure, not amounts allocated to specific share classes
+    4. Use NULL for missing values
+    5. Operating income must be less than revenue
+    6. Net income must be less than operating income
+    7. Ignore segment breakdowns, quarterly data, or per-share amounts
+
+    # Output format:
+
+    - CSV format with headers: {','.join(columns_to_extract.keys())}
+    - Use NULL for missing values
+    - If no data are found, do not create a row.
+    - Enter two newline characters to terminate the CSV when no more data are found.
+
+    # Definitions:
+    - Revenue: Total sales of goods and services. Usually this is at the top of the
+    income statement.
+    - Operating income: Revenue minus operating expenses for the entire company. This is revenue
+    minus costs. Operating income is also called operating profit, EBIT, or income from
+    operations.
+    - Net income: Operating income minus taxes. This is the bottom line of the
+    income statement.
+    """
+
+    return add_instruction(user_prompt)
+```
+
+## Running the model
+
+Now that we have our prompt and regular expression, we can run the model.
+
+Construct our regex extractor function. We'll use a greedy sampler, which samples the most likely next token at each step. It's a simple sampler that is more reproducible than multinomial sampling.
+
+```python
+csv_extractor = outlines.generate.regex(
+    model, csv_regex, sampler=outlines.samplers.greedy()
+)
+```
+
+Provide the prompt to the model and run it:
+
+```python
+csv_data = csv_extractor(
+    extract_financial_data_prompt(columns_to_extract, income_statement),
+    max_tokens=1024,
+)
+
+print(csv_data)
+```
+```
+year,revenue,operating_income,net_income
+2024,60922,32972,29760
+2023,26974,4224,4368
+2022,26914,10041,9752
+```
+
+Voila! We've extracted the financial data from the income statement, and it's correct upon inspection.
+
+You can even load this into a `pandas` DataFrame for further analysis:
+
+```python
+import pandas as pd
+from io import StringIO
+
+df = pd.read_csv(StringIO(csv_data))
+print(df)
+```
+```
+   year  revenue  operating_income  net_income
+0  2024    60922             32972       29760
+1  2023    26974              4224        4368
+2  2022    26914             10041        9752
+```
diff --git a/docs/cookbook/images/nvidia-income.png b/docs/cookbook/images/nvidia-income.png
new file mode 100644
index 0000000000000000000000000000000000000000..9727bc272fd7457b07f162c81dff8e987278bb48
GIT binary patch
literal 185563
zcmeFZc{rBs+ctVhQdCl=NTiuah6b4`5|YAg3W*Rx=6Ogd$<TnJP?@Jp$&fKqk(42o
z%w#HK=4tI$&-1*$-}`>=_pP<9Z>{y$vb8-Ab-V8SI?wYsj{Vs8{kU$QJtMb{j-8Gm
zh;{O(WK;-(2H#T0u3m-z3H2)6#s5~?9+y{JjW4&=mwfT_W;<C;J5?)VJI9MQMudr_
zm4y+%t)Y#Pk)`crE4vZuVoAJ+pM22?8>5SMrdF2RYNi%Ogqpn(x6l!8WkWk|AweM#
z?gIzJ4hV~h9^^iz#C=vqEikT1fgrdEd6{EsPEUTeI_qdqmZV0TZ%vl%;`+*1PQx0k
z#^kueQTUUBZF0sQe&+AI%{p|poLM?s4G*5#e49Qz_I~06Hi5j6#Qy%=ZGOI9bbGi*
zmljKdCKh$R-phQ}JTW$UVA}RWUSFi`@S>ZsQ(azPzl@Bu%s*e0`|q+W3C4fE7CqRA
z9Xx-3CB6CofB1iF4d%Eq46D}n+lk?i|CS@;T8H|SFV6iQFgvZDrni{#IQo*YagxLA
z%*=v7^CKZ6zx<PzE?o-yu-CHS+bfIYdaL%L#n&&VgV;7<b9mXQy|955zW)Ado+*QN
zVkw##&)?dl6^#4wQWJe2B2=T2HPQlq{`|>PS5sT-la-al>pJ}X$oC@m^X>=P)>gcV
z?T88pSZCIh<fEbIOmo3=QFN<>i(0=}Mbm7I;zI$`LCdBjit|my?&=V}M<S*hwX@A=
z<mBWuH8o4h%BU^Ay(~@9%DOu7qn56DQR8lSxL<m@V2Vy2?kZM6L0+Ex*N;W+u7}0M
z-hBVQ#iekoo}Ql5Z12Mi>+VXERvFyGbA<|mo00L{W~Ijm*Ci$<#)eKzPEty<T&IV(
ziHYUEwR!D0QrY_2k%~w?pK~)U%_^_u_h{#d@`{R!HKF`YeG!fw-QCym$+N#t64Fji
zxj!bbz*mRA=h-x8U3#~1W@e_NL0PH-e|h)*eebFuuI|sN=MUTTtd^YX-!5|bBO$dk
zW9aj3RwMml*{b(Xj%1xB($D8uo>EehH8bO7;n)7{uRUG->x~cn=;UOD9GClz>$D6!
zJ9p9?IdbG}b@iJ7jzb%mnSH3JS1l;NPhU$<uUc1@oK`faW-luzmz=MotIL<@qwP|t
zr`z5CDZ02u)YsQH;qqEyBO_zTC*0WBCpoT=#GfCZooE_!Y<JhVw7FaBq*d{vLs-iv
zEZxuVd7F+p4wuI%Jr=b6ak}w^E(fuJmGxGnglpH#Shvv&24Z}CJl(8`mzWi1x3I9F
zVp>lZYK(1j8Y!F}of$JEr1kZ=L-%QUPYyL34%El?*T>R(EtD9w7r9&i{E}hQbb-a$
z+4=3ecZt7Umft#H%i`YJ2@4CetzCsbnp#>CN)t<&ZJx{SlYZvK9*d(xX~kW0lS7AH
zr}q8$5J8WZ5+zd0v&ui)i<h+yW~<>|B^ZAE_@PXBn3|fJl&pWy*2sBb=77vfqEQrw
zk#~|Er&dj96>;0wxB98a;vC`CTl28{eVA};sOPWjW;HKk94TIIb?*K986F?BSaEk*
zZlR;}dd+th6B84=q0j4D^X<#{(h9tTB<EJ+oS#V3D@1bMym|GSHG6H{rI&t2&CECv
zCGAU7hbYa1P09BjKRzEPhO>QR&8A%mb*k}aD%uKN4><hPT)XA)2E$7K(igg~8{cPt
z|E{x7GhJF$&c#LSNA2TVq5S8c<IJbOd?}NvojqE~Vb#&|?m>9w?c28p?R&3L({?S)
zPF!xy7ya#0sLhayQ~hzkPEc6*#L1JZ;?6|SSy@?WYHPn~XxOA0clwyRy85#sonjA8
zyy3CytJdio8PRZXaLmlj8FZE3_W3Nwr+CzBp*a`(?awZ1R#II}w~k$;WVEvk7hXr%
zwQt{Av558aQ@^MQtTh+DpB!wWK0D6t#p5k%{`skS^NYc3b5>T?--Au`j}KW1G*;cc
zyMdx{<;uS2(b0@-YzZrlEzVDC)X8WC$9fX!9*Zvat^12+R%pD?DaDzkBSTwDYa2KB
z)3;hFto!w3u-*ECZ2IQrcS=g6%gf7yk80ZYf80iRorymoBeQbFiWMEby)+jtTyUG|
zy!Gg)qkzlr{%f6Ox6F%jUcGu%QFOBX*$LXF_~(`H%ga|XY!xp_I2YZ~*-2HX$3Xtl
zWRr$ZwXv~r*NEq`>$8y(FDh!f07ZFoBGj=!L7O`L`LshroKnL19D#lO{MI-VD$RsW
zjwPp<{ou;$rG%oQVq>xf!-HK)Z*W`mArdaX_pDmSej{Ae{LPy;E7q@Hf4QyT2)U2H
zfB)v0KP4mc`{6F~=*8F%eO^OQao0TD&swOrpNPSM2)Ef^ypYLj@^iBJg2q70D~pDn
zit_S=f@VEerP$p&b|l!jy3X^ojE~psXZ_mN({m{!D@(QaOq8UqK}6*$hOPUE@(&-*
zY?XA|!*R$`t?x=}e)j3^Y}1Ba_GcF7CU+h>l=CI?g8O91lf%}>5wU981_sYX%o<NM
zQAy+Gnlke8@(_9_LzER2Rh@CZ9~jq$C)S}L^YGMX8YzZ+j#XsM$jN#5%A&<DARs`2
zOWY|v&H{&FldG$1h?TMNuDH0k!K^!X?mUrjeWmjY>H59`f><TTQg`xeY3Z>SdWF0^
zJW=jbVY#ngCyUe5(~pgg-d)cr=Ii@*>Q@V;&6jbz8i$Rotr7O2B9N0<vu<6}Z`*~L
zF<z@5@52m`4Y8-;w&91YUgXJD>@{DOk?~j<lM6Xy{GQF)+L{q7iKypo{q{|}tGm1D
z2>!0Crzd7y>GWxtwB|cV?ggW7=zS_*@4&I;o6h3r84TMyc$8_whNpuC*4;OKeZ^gV
zpNgHGn^VATHBDR+ue{(k=99F0|9)jL$KjOmD;5^9Cj;0Ax2USBYUNl)@Jcz4U!r_I
zT*%Yi6sN=y5_$LT(|SG^s}m<qR8&_h2n$pi)~8=QQOkDo=FKE=qz=>inA>x0qNepS
z_4V~?U!y#iv;q*yw{!FIZkOtx?aR6HRYgchsM^g6nQ4jc5A5N+Y5non>ye@d0}b)X
zb@%SwOVY8XD(UaP9I`akVr4$=zx-u!wl9h@*RYb1l|f#NU!xtWW^xS$EV1LqW2>UQ
zJhP>Dyty@(mlkV8^te1`p5=d-A1S^NXX`&X&>$DVwPnk`y?giGfAC<t)WYa>&GYB!
ziBCh5yn>guoc=g8sgkBA@ioKn)&kzmW`4?o@zW|%H+T1)M~-O8$dG&b;>8Qzk+=7@
z^YGlicW;}erR54~JF$p`Zm#9KazR_=PoBK#wlG$e<c3J1=%>7NK<T`FCVmq^Uf6kN
z^qH^k`zWak5r-_l9~NC-InBq-EiHVqf`*Wq|0x_YJTkH$iGYdmUPHXH`3k*1XPb>S
zXy3klmwtWAzW?yy?uP=p^((jKSheeQx${{q`_Jp+q;yvXJV;5|C$FgZKtMM?q2S$v
z-O08$*Kx$vUvTQAkyla*`T6sTR_@hjyB?{B5*_}Vh1HxR&g_mj>KOVp%cKU`xCaTk
zB63=Us>IuMGZ)wXu&}W4f%vE-<hEo*W8*kq-z|r)O2;ZcIed=hld6h}nqW(FvxbB#
zJ~^L|o-7R!^`mWY(q#=D-Tks#3^n(*%k2;r&iX=hC`&EI)L-}ZZgl$jHB0XF>C-E{
zPKOD}hKziQzD0McvZ5mKHzip;vCj<mw^Nc9`NC85Pq}6;JKX=uvQ_Ne+qWB(m6fTu
zk#p6APy5jmKYDxR@#-e7*{@%}^4ym?pUhp-)!Q2zws-m^RS8*gY%X5Bxbw4?{_J=!
zuV+;tC+}ivPTNLq&s|!_I#PCw6ifsrnT&N+#D*Snm>i_k{*1!bnT`jTB-ch9)sUg1
z)tnk>zmTBU{j~By>9+2UcfSW3WMy`Ui|fQr4!6EO>%O$K#JgN_joQep3TwDy*RGQW
z1_nD09#oILkbCu5L_@Z@Mu^@$Lci2@m2@EI(YPa3PU`AGTP6H{E<ZhXU8L{y<AX}Q
z?n|?MK9!xOb0*(uaf6eE8yLUl*~VA94#YlG%a4wZR*>>sD)=$y;o;FsM;IF!$zQ&l
zV61LZDz$0Trm=~M2hI~e-XRl<_DNU}qu=vv!)<X=DSCExcF!_4i`gX@v$PbtPVbPA
z&{dT~5c&CQoTDm<&&!#bn({I3WM(BqA`A1a+9hj`02r|nT|a)PC!`|j*!=E4r!;%x
z#*K|8T$>mf7b({3qyZg;GGD!_jF0l5BXX^~R{@KhH^3pf)zP6(NaGAT{Q5>i%pjfL
zyNh0K_4@TUZ_D>CHfM~*Z8B=z(S^{YB64$cp91aiF5+dh#3du6i%7k(pA8FMJz7|%
zS*{sG@g2;yXnBQ53|~N+Ozz&a`wWv<`{&Q<s&RXONk6-J4CnPVj>RfIBo^A|t`Z;f
zZ2OMetHp4B`0!!%+O;z3>MRE>zTJ?2u<PQR8*9RK-2j*ruei!@+_cI1*EdnZD<Xn(
zVR5mh$(V4$28c${f8Mgv3m72SX)%7_!fS_LB9Aqg>|0}EcI?=_J9$YW&p}68TKe`c
z*ZV#`G(adfQN|2$c$`p9I|l~VaC=&OKtby)V~~<T&|9PY*fi;G0WJ`+{I2=HctKpR
zz==_M$qBH01)=^zXALba?Mw6K^#F`pr+$8=R#k0KNuz{yPl`+*6cpU>`SWL^{J^of
zNL=hhU8F=)@kC!8HDTFa6o93pBGA21b9>f)R8Q4@hN9KqUMvOJe1n!@tJ<IyFR(r3
z;#3L;dd=U(mqt5EXS(j}Psb+j)5+a%;?{b~>zgdSzjE3ucSpIcU@Qd$Cb4Z&t7ic9
zjgVfz?!hl*Wzmh}eRWY)2x6j#R>yBh(aGsBDgnhncX)VsaAv%>#>n)A%}ASs{m-5r
zwLR@lRcE_1X@I>I3o9QyKBShcuEfcDg?-fZ<@Rmcs3csc%7L)e%{chFOk36x01DxC
zl~iBqz4MeCiZgYZole|BmU*Zxrv)5Mk6c!U1BQ2=%&};>IXltU{Vm%(*JDv^mvW@e
zW!nJ1J*x*d3wvx877`)|oaHY7VQucyx?3b%n6$LC4%_u_K$*R<Q!%vU>(|X32d_xS
zpNa4*R5uE2Zus7k>9(7mEysDJt*|2WwQb*ev&MvL3{ndlknl7MoDS4CG%#{;1p&{x
zEiX>bbX;c`of}HqE-t>1=y-A$4^NW9jAsyFLP$e>tl~uj18<~Vx6xAC^o)$XudV>K
zegtqJN`j=8d<zQ1uYAp<kzPvSv$7gdFmJAXd}yp@e=+g4-Tr05<}1p)j}|q;4XXk>
z5xYA>2R=T#F4mT5bjSE$*Ml~fA%3^<ce~AgL|6;<Am{j0@7}d5sr$J%`%apRcaocD
zY>(ajIeobb_?SW%^?!VJ>dcvwdU}!J+0!p0U)3rPO%F$*9S}RCq{JT_6&$SgVoqk|
z*x1-9dHGK<%_&-&wx9HuwzGSw!1pC!^GxyP7qY+DbsKXzT*tMQ<2KWpJ{C2<xm_->
z1A9G#vbuqtJ;|71*Up{W_UuV0m(=*-BvD()U*nkRWi7bWRdMI;f{`EH;>5%R_1)$9
zgXW+2?Z8XL#MrY;>P|*Ux;I*lVGj@34>k&2i1PLAPxGQ8Gm6dl#6%g2Jw3P7N6NRK
zkDi^lwc_n^*->%vEkHBGxCyO#=q&~=gSU6Lz5VcE6+r@RQM1N9V^7dIr05m0OM4Io
zDD;P&MxXv@$H$<|j*gCQ92)wbmqmBEE!UcQq<GomURIUl;%`;A`CmCRGu=UqOiVYh
zH7`vYPWF<h?xVR6nqKuPR204p&WgRO16c&u4+AT%mR=gFqCsL8>75uK7s+~1{Vfp9
zXi_)YgVNGcDl`P~%2DbA6}6dY_xZCuG&D5+G>mscLvMCh1(}cY8U9ol7g^+^2N+_`
zJ*}vC;>wkjxVGIR=JKHex^zd69&N-<Y}>X?HuS(&Ti|PZ9l_T^!osSBesn!`_g~xe
zD%@@co-6lUn&46j9~=5nnMNz+HnZX3{tG9OImbSqx4QT>)9A3{@RqpKVb90!xQ1{f
zd>@-D9I?1Y0^07~p2|RP^fbD%D-T(<$pE(IJN)u9ZHU{(%S(r=s8fxhnV(E-kp)f<
z5hMWG)u?QLyg%o8DN4*fsXP<eOw}{k<M$^yY{$Uwk@n)_%|!?sho4`raUONx#%1VP
z_DQ2DP<vYS=@TO$q-^3O8r>7AvNQFH`=7bbPhD<#mHML{n5pUKlcr>ihirFok7YPn
zO=3pNGgVR_fhPB9zg*q!KFwJ)(?xGL(#DAkt{@(ThZ~@T3m!Od47)Ri%Tc|JQX)@)
zxYH<s;;&igA_`W6=&5r^?l*712QR(7X*V~|44{ihnga?5D<P;~LI8165rFbH(R6IY
z@10UqEJJVF=wE?GOT?<J;B>RSgM)e%P1@!50_RN+b}Or!hwMZbBJDSpZ^|m5Q}U}N
zNAZc$8Uo=;sX}0sXPY;3^B~vUHQp6$omZ{WwnULZQfV4P?jos=_wP@?e{|qPYLo^G
zdOn`5QXXrGFX<PlT_y*8&<|b2vGu)uyEN0N=3}zPex=zJG{+1G@_eFszVq^>$XYJR
zg46EwYt|f7QQ4T8m36>#$&Gn~oxMHf+Sf$Y_&1+Fvq(xxhD+<7ITQF}(xSWh*BuqD
z?n$}t=ZWwUJ55bCGzC;dMn=Zy)YMJAG!5Z!e7ZZxvt?+HaeTv@Z{cJ#E%ICVgCbe7
zS;}2JNq&<fo<~3W0LN4NGV7-<aAqp1mPjwMp?oz0B|8_BFBBI&5qHi!ZBIg^SUFi)
z=9Z~LZB8dbTJr79GkWRBv}s^sqG52<;>wjBLP9FGdVjZGS=*D*b^Ulk_5mULE#=M`
zRvvwfEz%%-eSCc4_$yxz`1$^9$x%+CB*%^O@7pI6-;h|>*4DO>l~v=11Dn)t^gH#r
z)`l4`Uk0OSDO|mp9@pmGsz216mbRIekB^UWuLo-fQbX<H%5g`lgG=HjAGRpH_G>w&
zE;C=C6e*tf#Z~v-5_2;x<FBl5WDDalP`|TQ;?sbgfu*HDXlSU=<&WGUTeoi2GdM3R
z!?S(6Y+CcfIbE}_nXxf(aY++^II-_?GBY;<Q<%0D<Qe}=(|wJiFfcH{TU%8n7wI~j
z#}icypqgL~;z3tjTwE>k`t|FLNfgDD)Ktnh^Rz<sL5Y;NPR!Z#SB=`v-@9MTS9Wt<
z!^`H13R(A&Eu5Ua<+nE_{RG|=ZLdRP?f3ZcV+9b2d@nf;GBSR;z-dA<S(LLR65Rdu
zK2yOeou`?4%c7#9KDTbs=GzT;GxMpFq={KX%Key_7&1Pkmny0E7mOayFrr<@9uuhO
zx{mE_)R<FmHS_y$QF<yWs;f3O-g$XPP`^K6QQk)!WkqKA+(_Ru6716RmhpY01P3r^
zdS<5g?b~YxKd0Vd+<vm7F;TT2Jwsyi{tu{x2MjCMHYS{-W?^9=`2#et*NtjIO{+m)
zP$5a4ynK02L1E#uh5?W;KFCblK(2rsl&cLFt#NDp6fI}i`Nrq+$EQ@Mm6Xu*T?2^M
z2LbU&(p^mZ<z*^cTic$Vo&?eN2=<T9W$D54kVMgC;Dy`GNDrmxbx2aqZutA=RoAI-
z%&ywo%Yt%a+_Y)pb?R;PdrKi2g)WonuG5xwzrHa8W2I+hoj_-R7hDHIPqLwTY#My&
z<I`il0Ra`*gU*o=25=q-5%RrAoYngJP|@rUjxFL&XS0e7(1vYbWV}`xz)=RO6`!N5
z#i}^qg2$|^u<9@<negS6#Gyln7QkPSP`I#A0<^;?d~~<zR!6fO0QT|mH+UT*JG&n~
zd>59$Hz??qUYZ=h$|&G*a`V)(@7Ub-VT)^^H|aV%I}wG{=*g{@m)x=Vv?v>gU4ADR
zSN;N#k?t~SW;akz12~jxKg6P<s(P;2!yVDPakp~hzP}Go#F?0lEW_Mb!?{gnEwA|3
zb4jk{UZzxdH@l(%N9-#croKylLw+UD84~}qAG+B|BK)r}3frF}`InSIx%j7b`@i$U
zIUf`aYSdKxL4MsQTDSu}utp!s{M@veO{{NVpt^-S2<S+A+kVfFAOlAiChE%C+O|44
zIP|yVjAU6NG8ie=e?EbP57L0l$&=476nkK$y|c4LkyUO2#9jp5^5vBU2S}kvd*Njj
z5<91-r%N^E#>d3ye@@Z5++HMsK)h&axgNaP{{8z4u!&!gcJB3ofnS5pGS{ZJS~aJ$
z?yx{{PGh6m4IiJ2z11Og`Fj<>L+W|V8X<`>fPV8rP=TS+yfWL8W0mZcRjNNjeO>{}
zM!Eg<kNop#IVm^y@b95!2E@YtV8_lkG&;FgsnGS8-`QW}yN=^v$>+~gn_rFxHK(3G
zgcP)bz$se(XujMhwOmTe<wkIOsz1B^E^B#NY`;#f*N2Fs-AJh6zn?tW@*U_0B!je?
z8q-OCwrePSD!qNZy%$lGa-Ao3iP`mg<=YRb=dSatw4U95<x4u1u*nA*X=&o-dd|6&
z$J7KF2!E7-2r>INU<j6krmy2%L~<^rtpj$WbDA05hGG`$?TJH2vftzx#Lel0!n$f_
zx01NhUc`Cg=33{_+Qkz<*QNS`CHPZaJ;+cvh19n6aVVq$%uJfMCqPD7TJe$}5_vaR
z0+QK$|9*UV`3rDQ)R$t<Vlk`f;Z}TX<JgzvvU}c;GAsrZNecwJf7`)@eP7%*Q+ya7
zkC*KENjyY6#bzJ2A7mmoxOnlG6`Rnd)hq%!zPQbjsw!Gyv^z+u3nX~B*!r#FzBo^B
zps^r-39_xd-28%*Wxt+3u)We(aJrN1`u{=du0DVMJcxQAAuhmapKp1fbxDr=%$YNS
zhYsC({FnoD=?#1mo_YJlIeSno;7CQhLPD6YT)EOQG;{|5g(bXH{mK2$M1j*74=5~J
zBpO_G<*HRYyLPP-i+J&6j%;A?k)EH45F_*M!-uSB`>umoNfrWFy?G|S3_-6Jjsphr
z$yg_F%MtIT#rd+@+I6^ttB#IUcekCm3iaW5v-@ZV6|&SVFx=8<&C$TFnniBTw=MWz
zED}!}2FN7l<wZC-JC`;$vw|+$hQ|I5sC)gH(N3p@nGD_*kQo#?tUrBf(9FI{BW%`S
zo5xrM>tW~d9V9gjPCYk2-=N%=(dl<TBcL5g^n>&~fwzOq6z;%`%vJ?8rt{aYbpW|)
zX?mPESXDT!qP=L#o9t=VZ-J<`wqSW_e)RY68?muF*$-av^;{gfkmLp=V<*OmQ%y_Y
zBuL?VV3_#o@A>xhz+$qfo=sV8PmfWP3$2l+$6ppv`!F&x5ZuT!RG<X)RCa{Sxny<z
zg0V^teB_w8l=%4F`HmwmB$p?mmT!Ps2m4&(R=7gB?GB&Di;gCZ3qH}&+f!0fHf-47
zMM_<7ZKIz*{}?UDK>3V=^X%C(K)g!TNd^LIx-#MnEn5Aa-ri1}r<=Gm3O$NPkmvGz
zJK^={5!*|P78cZ-tJc<~fSh+9JWx$20swpl&HzV^)H+a@1WoIAC~=42*i=j7KMCIv
zAiIc_UMbqyNndPGryih_(Nj@T5gE9{x}OCE5am{EY^>NSZg3)@fQuB5|Bg8pfYdbj
z1kI>sua>6fsQ`A-a?r)GJ~wZgeR~N~01UUrEEEB-%}J84b93K<4W!Hh>n0bV#s`OQ
zWo6xsgUwr4UM>U81$exget*H!da$33N6wu)X8>lfG6CWWiJu7IwRfQZWo3FmjdgW)
zKC51_fq{X6<h^g)Sjot!_1Yn|8cIbd%IR&8>cj?c=|}7bgYzu`V)1$Ogo|hln*oPI
z4k97l=VD?^)JTiuEyi-G)ZFoPa&k&|MYUn~?%m`iaaKFS0&8${3C<pCS6z>bAa?B7
zAwy~dpyr=94^c=C?MIlT<M5NiM~~7I2Si0f`XXJ=xPKn7%S3BMImpkiw3E{aK7_x#
z08I_(-4Qf*9zA-*ls^RxP)4RFt4?y0qoZT(#7Jl7Hcm0S4UG42GuF1Y4KHh(nkWcA
z_K`1NzNn0-{ta^*R&|eL{v$6~0XqITs&8nx9jLM%NXQ@!^w`ttmX;P)=lzxbo3{%H
z2;j3e*`B>B)!p4KI;j*PMrqLNc2z5Mu|EO{(r11*PuIvunupY`Gf(#5&cba;HhAcB
zs=>vJ=lYjGDk!UJ01sB%Iyi*xQVLIKkHLR0NA5<Q=iU5AvhMxe|7kmOLBoI@NSi{N
z2u~1v_C*gKJdlx-Q?#^v0RmIu?Ag0>*B$Y~!6#sk&6m9Nuv<9R@4&Q)TnogRxa0d4
zT`i@AS5WXQXnPr%9Y>FzKN+}XFHsKq|HXx30UjQRg1p?%8wPr~mS=8}`T|}ka*dQz
zlei6?sl5=cssjL{-Y8}Y4i2xtN}b)bQgA)Nva$}`a1RJ59v+*~j$@QpT3N>785UpC
zpH;ge4JjxqD5!LI_V(^X9}9jQdYj+FhYyLn=<1TjQ2>%60SFWIKayGHT65+~!sPv<
zXVpV?Vgn!p0AbGqyi>Zegbm79+qaP~a;`Je@gGTrPO?ucQwDH=Y*u0)hlSk^3JM}M
zdwld5<S$;U3P0wg7m%@&T|n32=%bho{@GK<kv8Ug3nNd)?2p>@bIZ%i-+%m=57Of~
zXCQ@hlN_`kAlZp@A8;Jj31MYsj;#ls#|Nyq3EVs2+s^MzZtiox&rPnRrBzf=@B_qs
z3o7bqB_*xd-1{mx$$jJV=bm(}@Iq=r%8^%B4~>b9l?V1zYXin_Is(;$G5}WPTwS2G
zbRKe^a+6f|{I8)QJ_rP%zkl1AlIL%;R=!=LukV%bx#wb=6V9Dznt()~`1|Qo?sxCs
zV=Y0@CA8P=RgDW4^FRR)4*H9n@e2yl@O;F+_NGnLGKhx63G~4U?-b9RQ5SYXEhBed
z27EiB-}FP+L5uL}c>pp5KTs`iDn5(N&}V+i5`gXfCr?rmL80k{j3DnnJIMNekct~Q
zf5W|TT7r|4lQIXAPjqsAc9t1Dw~2}3>C^sDKNHMZ_N-P94+}FjmSAD+*M4;+Od(3t
zJT&eF;v%8I+v^ncQnhd35k!02T}KLSXo#%L%pD>kYQm2*JE9;o&CJfeD{!9ZL1_%1
zfN&sLV|}0~2f}=F@bu{DS?9_u18A{{^}A0$;^vvnTBfT9JCc-Wg5I2W1WX+hBXIh=
zzTMlmpFqs5^dgDP*oRDn$dY8LX2x-dcNFC0`{1$VO`#1X4PHeE{w*c^!Wq+uUEWG1
zO>+C^$9hEdTn2WANM3roA!G^o-u#8I>(rGs>(__ow7Dokj^g_+)lGDuxKZZ(nEBN5
zb8>Qi%%)J1L#q{)m020@!Says5^Rd;i`GI{M=EYePqBd{^y%#uD8aV%UMg7i#5Z_;
z!CSPGcxv1CAzU0pGik^G$a(wrZArx)*3O}!<h>KvCUZ%SG;@c(51a&vOaKILLSjQ-
zzY%Hba$^GHx0j~Z`=UIz9y@l-$Z``w=4J~go38RTAg*7C>z)KBfsaf@0DGHRl&?S&
z38)pks}{Ww2^WEJ(oO2N%E-zlijD&jQ1y(&pp#8LSidK5Oxp-JuQ??k6p1h}h)XJv
z?0*pWys60ZKHu`4#=2DIbxI&9!*Ws<BmW>Bz>kE~paM_2xrrlGuMRfueIjO0hkk1(
zV)8mdKSZ-VYYVF(l$r24*j~&UHg=15FERb}Fn4POfgWvA7de<UHuq5WSNgA?KU17N
z$T~ed97tl$((APFmhjq74<IJsNFaE;afl7!5+Kb6U@mU~5Ix1(?~G)S&WgJT!0@bj
zNR%v;RVGpHtGJhsPPm!@G`_E`ZR&gOv5DyTl3_TQ3t=l<47o3E_DaEeL)rOeD9iR&
z7Ph>+{OS0_(ZSwcAG|yDThEgSCy~0hpd>{&=pv*%w-Q$&J(3WI<hsV^u)DuK|B^>h
z{B}PI3){APH|_LDyPS<p=C`q&eISVA6Nlun8?vGN9UUD{XRj>W%CnY63fl<0GP5*4
zLeffKzG#+LR+3W8HDD}Ksh@85ELQoZ7LHV5uLB&SGQW~Q?Q-f4IFxSLns0L<EoLZ6
z!j%Q9$DgTlEIlS>ilymuW&$Odg#?vIN>qg9_uOlcY%VS>%;q}$Vn-neL<QL{B$Qe1
z64{%}mX(v!?Di@>{cIIdAD>WR$E--vFks77)G(n7ISFI?hmW3p0wNCmngLJ@i54Mi
zW!UzeA?-L(ZsThf78cUctN;Koov#{MjMy8LxwMNdr%YckSkuJlciqh9y~DFkLTL#m
zmq=p&)dYg?FV7E^BNYf5-rdmOYg5wU9%4D~04&RP_$oJKq8Q_Yo0mIv1-DmgqI!!7
z@M@~72L=Xe7JEnln2s$@w;PW4yo0QC1I$NeW+sS!t`10Gb0@xCGB-~SB)Qez-8*+O
zLqcz{qTH=%Z|6crIR-Q&vl0xUK=Wh(^?E+_6tABB*H!w-ETv)QddJLM{(nZsXTRD}
zXQRV=XUv}DWew6^Batc;zQ{l`z#|~A9!Rk4{!T?fWRX9dCi(-?OBP!`v79hsZ6vh*
zo5ZZR<bc*RmL$(<%pjxAH!A8v?1jg`l)MIHL!Yps&AE@$Ufg}ou9h#F2Ak}Iv)&CB
z{sieZYISU<dg~eSLK->v#}fp5CyqZczMSAOo?SDJhTs4MX1GDT;3Fb;{L5>{{V2h4
z)l3EBS}Knbn<BzjtLNwT-(CJ@eC#eaKmR&#ki5z=qhvl6>w~XoaHjT9yA0BVKI9WP
zJ0_D%OJ`-@)z(s~KnJN-6)#Rbj?-~*$v1w*NWd~C&B;-`b)EF4biP_rb1^Zw1#rx>
zV@FJ*1IzSal69EL5rL4{m?2b-?V_UD$J3fwt8(jT0G97UiS+dJ^yfUvf;{Pa=Z<RH
z7p|iYrzgb5A0Wd;2ck(&&?^=oX(4-^dn4~ho?*EMzsww2;Wgdn$^h1DbYcRe{ZHYM
z3-WcR%*@PgPs2Fj`xiHYrF5H6%2sbfRU;w)x#*MIcz9Nre@-b4-mCiYrKze;zTFz>
zrF<7Ow_=1o2r|I~@0Qa#@Fo{=PB$=;O*$#cmHMGfmNhoMJpQYt$o(i%(<ihvq&I34
zaUinv$MEoaf=y80D|`{%#6dS$`k;rL)YfL-b^6iMInrvhh9_7Km;5jEGd0Wv`_$UL
ze%0b+U8xDyl5yk4=L133A8XJC;IH<egdG4so_$jjUcPWs02&FBeMcezDX1bGfvz)p
zFigr*5&?CK6tuvAU@47;CqPgneC})cUohJD=TIUhPFKZdl0s-IgX7dst$n)rOg!G8
zC;Yjjwn7L5^Rh=HeNw*Cz36J^3R2_5zY()T&`QO4fHPAiXbA=`No0mquxgnpYAdD#
z$2mbECwi7w!L~(*EtGa}I0!puYUC!IOJqN~>1ZKvC#N~<fQ@Oo0EGfc!y8CdltF1Y
z!6KcP{fFQkJb`LLa34I#0+Jbuv+N5ElAWoTS)5|UMkaj&jkXhd*T)Ypl$VgK9)ExB
zsBAk7ok<qVv=I2>a9eM0Z;{t7>?7Lwuqr6&;#I27kPlcG@8L!U8WSHV<YZ@upgzgN
zg+mDkw2ae*-R1#IpowlsN=h7sg@ro>1kQx)9}y@a%^i(t4Flz*Gmf9%2z(^31b~x_
zOdyw(02(#kLbBNsWGioIPz9NhP_L`2YlJZ5;Ta!(Jz`dQF0DD4Q>Lt}3}z~s-XmgS
zNv(!TN=ioHr;<`3eAK6C2`lN+7`5ZzhD-s~h0Hy{;`-^O?VmhCg2g{PH&`&$g#FO&
zJmR`lz85kx4t&Xou6_;E(2qRcQw;0Zs}G=+`79oUa{{hEUXp=ROf5vTjVAYE*-f9y
zFHR2)yIajr+tTVu^$OjAj0~A1<`Qsgt*yFdpHmVYI?#rh(RSS@n~&&|0V2(N5~$y{
zzncf_`21wT_2-I(OA?iRq)hkV5Y(pQb>3qm<*3wANuHIj3DP3=8{KutQqS;L-PaBj
zXEgtm0g_B#(1Xn&p?wGJbw{kbWD>jOu&J|usLcfLwX3LMD)A$`^O=_RrJI00Zx9Q$
z1stdd9!s+(3qVIc2T!vdvfKjF;g-e)cV-~1xA60p>PsX2jsd%E=jBz%N;Yv0gFc17
zdL(Mj0(80tT{#)hfHWt;T^~U6wgv28!wZl(9XNce*RAu($vFfD1Kh$#Yy^nIPMAs)
z(CPTJfEAJW*;dGbd+`BV3FGF+3`04<;HRKDl5|Sfrhv%O2O<LLO;Smu#svPo8;%J!
zVdJ%W&b`c}nQ-@Rxl{{Mfc-t($^kD2bm?PYY?@Z_L-YBPWx_!6lW2!Y%!3d=4vstn
z*{@)6swHV1^p!7pw!6u!Mv^ozP+tC?djP!H0i){R;}-nyU|!k4!9h}3Z&2yLw4)y-
z84Tp@Pp=)V+$L%dky@;AYc?)#2tt>X763KAx3+@(Swt=n61DiowEN5x(q>7rGQiHI
zKzeAbOLE#h?jTXlC8i<6T*KX#)YoqWWKccY4ps0D`QhiLM;HL7Vay<n1!$f7o71>F
zW_wxv*$zB!oYU3Ox%z)1Sj?*;Iq+*`uEgn-gO~N9Alu!?kLgL8cjCj*!?r*CjgNSl
zv=;xbIQPpwCfiA04D)~tuk|%%^x1!X&8c>-g{=M8*K=NfvW@!x>WAm;rSQf=U;(E$
z+WYW=<j{}c7E~d%`R&`cQ(S;pDRY3UR0NvWKU!4?zd%I4r~J7(H;OQ@KvNZzS*eCs
zNYHB8My1{~bual5Hk37g2!xQIKmp`2C!PYMvq`!gCcvS}EnajTY)mxyD)Zc?<)!H+
zxInKW(U`O{&22jW@-icSzWeZ@#<K~uMC(w8i4vGjyb+w$SGJO37Rz7PzJhCzWHQk)
zQ73il_Pgaa&(@nhKHEe^nbDk+HXA-Q2qPNyFnGMYfBuwb3<?oR_kqd1MBMe8j9Oj*
z7d}d5{NaffS|30)Nto#fY0?#}^jJMi39}V*HFNFrcfiZ`EAPY60mZ(O01PKPm%3b8
z=IG`joCMN*YiX(1=|VRMc__L3^CvEgzli4_S(Ycm#61?+c*av6k2Nl_&S!wcQb>$P
za9;lQaxbBe%##bm8r!G{u&z=0$7R2B4ybU9G=-l{TMFHA;J~>;7V)bc(wH&m0JgjZ
zwvbeXV47SQy0FaKklmbiA#HFDj-9LHJ@R&gjcX1bJP0ML0$$NkP&gP?Fd&1<(UBC2
zVAZ}lGkQ@;w<48Fb8BoWl@oXq(qN4q4t3QF`3i2Pbf7>4nQ8US8#fa24Gaw#IXMFW
zw!eJ)_C{ausps+%@b-SYfM~fO1C$#iBN(6ep^AE;KQL@7a6YB1oOZU@AAo&5Fq?O&
zz83--ZR#yptUA$IC%p$XLTRA}(;*!F;mHwdR5eUnG^U={utj&g5fH@-9o`$*s}ZoI
z=PCo(k;g<MB$l5beI?(r97psp#;}oJ`=u8t)L@8$v;lAW3Bs!XTXyo?ci7m3Qkk0}
z!r;ne`?GyJHTY99onisu!nzIif+@d_Aj6-e=8>|=C3yY%=*_j8y~){%YKhqbE1(y$
z$Bd0NN-V90o9ZG~)9ea_YTWsMFvwHl07@WU=|KMYV4cC53W8nbnYV@)n_OQIF4xdC
z7ZenHL>WBZo}D=rozksxAIvPNH<7GvcD5|W0N@=#IQmJ%?_2y*=!(oAp*=?sXc7^p
zBJDMA-<~1`OZ0eT6V5Jj*^5Nt=&O8QoFAPJi--t-ZR*u~cO2j62CI=2DdDVoPREGV
z1fc!%U1=z`yd&U&U_}iLjjM<H%cGlb2SsxeAosZe;u0KWv3m1WT%KyZq)As_kSPgu
z^S@H0pp+8x;jKy;$4182Ba)-*H*MSq?h~MtHx=)abj5mrhv#qNImF7;G||)qNJsS7
zW+{(?>c>ZH<G>92!EW^92@Z`SY)E2#DB~<xM=I`JyLQ1@xNFt*V<99PinAZw`-){R
zXqu+ppFj60g^MtvjiTblH=ka5LY_+c`CRr6*hQ2w81-npu;1_C3{sl~JMSAGA1_w*
z=FM?oy^*4MOP>~@i*Hue3Y%QKxD5xC1rD5$;a|V_A;|OakPNhQ;FBj$%FD`5fGobX
zu__*tK={y?mKH@ML{a;7Tg5ePKOnU=No4&+ODB#)%(4NyK$&{i*9zd`US!>UGUJ!R
zsZ$#u=hkM83=Z;&MVy)353^^;A&(!d{Ms?jKj@YMIFG8^1|%fxX<~rHVfyjuwMQr`
z1gr&_Bo#IpE#9s21Y(~xV72|@M~{5<(pYiZwIk}tvU&5?9VJvCwa*$+`QEy9izK?+
zZzH}E#jlrg<HS;;IJa!+LC}RyMM`<<^%kQZ#!dyH6H)j()4Z#xA)V<|+=wxs@jG+`
zJMBAY>9Oc8?gt0&Ku<4ZRLvC9jD)C?R_s3Spror9y6+w`>1~WN4Ukg--QC+s)(}FB
z(pAGhoM*c*0ZHc^^LrZ`oBJS$(4&O_SHk40(&xWbBLD3aMw8xR{dU4Lj=pdQ?18*i
zfVYEBK$s-xpU}~HqHqD+40|-1;ZU&`va>MeG>XCw40o`qj10J&=(&UVU=&IGjJ7Ur
z#bJ04LgDF568ljXm6lkH*bVnGA(>A&FeIt&@}GaDmcU<WXbMgps;S0wGEy+F6)M)?
z6KEC*ze!WGslR^|Ge#cd&zu1g$5fV?cYs{r7DmQ<uytm<dc{J(FZ|yz$#Vfwjxg4P
z_Od~0Jz~@2%YOYiTac@oTCjqY>r@KK03wXxjb-}E0NqC`!@%dsQO7fqm}&|^(h&eD
zqtS;k31ynE0y??S+DTt#FWhwMKRkjRb^L^{d=a#dAj6n!E<mA@D1MDGJ1tSuz`*#+
z>N`oXyI|eu(m&%WkdOE9@PNP4=&cIi&`79-%XNh`GUSUj!+;>?W)J+aY{{e3YU0Y0
zT|B0zO>FRbf-@3Q2j4VOF%>t-x=hYhUb_LQ(R`frq_jZ`SDV9-8D$QVl-k_LX2NYK
zHAig@EF68n#nQFgq3ON*@Ijz4@;|e?^w~{t;s$P&NIC)n`xv@MMb3t<wQPdYy0$=)
z;RPU~=m}CBiHX8zps_&!H>Fg8WH9ySSY90Qq_l#Y3ok%xLArOqd)G8y@YH?wiY=%Y
z17RY)0a4F@mJ@!*ufqWjiM&z<+$<||2_83I3v?yn<D@sundaJ_?pNPhS{^zdj33L^
zNYQ-ey#D_E`#bpflD~M6T>pb+G~XeP;42v!aUY>*6hH^xqo!SdxtjI8xCjp0boR?X
z8`9n1uh}yK4-|NkXRees4__uOqQJH<Fsa7Lq1T6=Rc#jHH;b7q5+22aWB?W+0X)AS
zb`8<Ti=nk7F^8*H_e{@XXTr0`{|8nP@Yg78aC~~Y*==fkTxIC?=00qQB4>M{YtDCz
z7rqG4#BOxg=HpwoY$1p1fNHo=-q5+<NJ~qLdk;{QHmif~!2U{smUO9@5t19(1BgpV
zkanGwD_3&w*|QpFk9oKZyj#;4S|+$`VkNNy|A|N>U@x1|e|LYU6D%m!3p3V)*PT0%
zTX;mw(_!U1%}LH<Jm`Z&4HIR&rQ$V|19M5K>F>6Kx#H$EA#wYf4-8k~5RG9ZEvatJ
znl*znXkRWvqKPX4qQEG{!$f|Bj%K<(mCQ<r)m1Rke1a{EbUR~ONE(^pjzkVfhVVMn
z)(TX&RUONb5^(pBScmKiVC%woU+ia^+V}5Q5)eeV4;^APHvYD}S5HkU(pZO%^pzet
zvYF6_#q6@u2T0f8ffXK1i7YzZ&&m>BK}}8K8(h5;Eax#)B4f@E-eLh{C)^zz%*3|#
z=EgM4Qpr^OaknJgCyQUV-%pqGY9G&{cQ)Kx75Ee-AiNGb`d}{lcG<Xx@k{a6|7Ldn
z>0iU*Wb?$2Bl`b}Z@A$2>n$9#0#n1S=da{J(z)zyNxf=q2^Mawqm-JEZhoOlPs0;1
zuM_GeC)=W;wpdwr(&);4mB=Bs3nMT?VD9hAJ$vTt?J3F0=3y-x+1Px68%X*D_Jdr{
zWl4wO@3)cjx*jM?UW<z2LU&8Yz!0R8W@u;_9Y`Np2r7Y;zBMtJ6xO1qLaw4;QZ#8p
z9KSR7%Q@UbeD7Z-ZMD++Od*dm{+<`$21;1Sr58kk*?C~S&R9@W*Jj%5e&O~IMiT<>
z(5VN90AvFI`yib{Lo`Gl?{A0?qWhMDPysMYJ=<D&?KS|z@$rY0y}#?~>XdEupi$B-
zQOzEmw)~nYt6Dj81S6}nt>0EKZT0M~V_svx8vo~jLHzLhRYfwT2}||HPbEw{O0I#r
zE5k@>IwnllZWfZEF;M=S_tcr({D*n{`2N2#uVdez&y9(6tH42{&dE`yCU6%f=)^^*
zdeSjHYV{wE>8*%v^)BbiqnaE+)WBFFTz_>X<xVV0S}P8DQb27}stz1D5MBT_H1Vl4
zc$_=#u^Ibh{wI#wW2d^C_ck;%+=C0j-``(l$ntYan3#u+x+;bwz#-E`WWgvyE))hQ
z%e{-hRISaiY*pzcC%poL;N-xLP0k_}gL}M=9jy%HJm2|>X$D=LNYwPNmiRypQ8Trk
zBbC4QCTHrz$9OP(e_W09=M7zYW`3S*yOEltQA1n`7obDz3_LFni%UWvN&1k^F(5Ly
z#nsa~sg0tn&u{MTf*#^MU}|a#%TLqG?sxO5n->Lf5Qf-6g5V1n_9}?LjY9{PINigS
zg}re3KWaph2v8kO=nIS>`=5t|2eIPOqn>5uyZ-~hdF}c?1ZQ)u>$Ct0*n0Hw$p7Zv
zJJ95r@e4y^2k#dIj;i#-hZk=(q@tL%Y)J_$d-G-k7Z(HRMk3KZObEvAWhm$dAXAC)
z<^X`tT}5+)eCP{ER^aeqO&V#~GH!q+gfang8~y+`Oa+jU0s-9@o-&dMG~=%te^v1X
z;~eS6wd_4Vesp1^R&(8a&p%!O2Fcl;O(E#Vq>y$6T2a@ideWBF)&G;lc*Ew+iQ{uq
zzv#J_C;Po8>A|~E?v8?$fiv<b7@Urtp5Kk;lvOKNmcSZ)?b<a=G2|<FVhW6IN$l|9
zVCTZx-A9fdP41>=6(HVv79Br+T;$k+Zf<RPUUMD)=KWi+{v?eJO?tDiaR-RUkoutK
zkq^Fq|IRa%4;L4S+CAtuZrn&mLq+c)2arg|Z!V4W)?4;h0}A|U=Fo!lFxtMnG`%^e
zzb+9Fcm^C7Wt0R?f0>IOhBLCk6_58&n_-9FgCvv)(mDz)gmllLjUv|+$neY8uNT3b
zGbc%*=Ah-JzFREt815u^N5{U~){#9seuun1RS&H+*?R&1lr%On0lVLTTZy!t!B9nD
z^gx7c(kaV8A)Q@aE4Zbcz~9s23!|LO(jI5~0}i$AaXcT{7_ZEQm|qDC&FI+JIjV<1
zuw76oD7~QE5(6=(!T{&*0<gvY2`8Y;^-uu8<-)F+^KJAZOi7pPW7c>rcr`#LCg(Mv
zbcru7OpwG>axx#l#!A3&x4D6M*%usITX8&YVax|7+8fp;(n3bwKMD)@*Yy~70x(t2
zT|(EJtl<oxN=lb+T3b1BljYtt^lEuvIYUj?@|o+9<L;B}7Pw6`AZJsxMalUaP|qs8
zbl$Ga1GmO(>$6Oh!Ps>|5pg1?u;*aef<KHNL(G^WI4@BJLQ@vhT?S+%V)oh_@2AFO
z)AG0zmaTtq-)(;rZRUk-iarpw$Vp{#6rK)!3Jx^c++%}hFulUa%uIP!d&D0s9X&$l
zBv@vBE2|*%zvO;F4zGeY-x@;K@z*=4_e|IqBc5*Dx)lv*f=0FiTHLw#Gn~M@9WZeX
zQreUyXIDbLCy3LS2DGh(cwbgiLodCwz6d3pxLa&E3LJ6(W2#QBt|#IfN?}ze`9173
zI)4{Y+az%d=Xy|yq+yVpv%rH20G`N)3#8*-YuD!fcjTdNob?JXcu@>sn!#A}Dv%Us
zQPQAo(}EOj2&P}Zz7ptR6o%U}jQx=I%LflOAvo_qo$1EJlfZ=Noe5AY1~6feLsP&{
z;OPC4G|8zn^zCGV6@-nAuDKmPU2=2><2>GYU;{$%BiK!lxoqU@Iws*);K%`-1UpAm
zRI@!qExibu2vh|h;2$Ya!BI?F`GBg_dg+XhsmE|-8P^_%K6BCBoF3S^G3^2u;)G4Y
zMTE2@!XZe;2+T8&1TU^QXw~MA>DOr4J1lS%%E20?9o;Z9ot3*e1(B`hAT}HQY5^E@
zj{el?(>D-&m=5Ud>$?TnL2_u<xg#Xn!tIyftVY_adK@ARDZ;M+=5AqKmj#+<J%6@3
zx+zUBmy9*n2A<x&CgV^>qWTS<1YJg9hX7s*pPGf)gKs<gbCMk7eG3gY@s(fM(B!Lm
zn>o|g!&BzNevO;oMQ%7raRrHQJ=Etm+*eeW7iD{R#O)+GvyTa&jkg?&^1fwk7)aE~
z6T(@0^X1DU3<;)QEI+<_+l`MAzX>LMlX(||csLfsMvYBMMPIJXq|?yQSpRsx83yym
zt<P4E$>Yfsy%7AQ4fa#ym)2~%v0*u42jR2fxbHr)W3(4q05btI)!kag+d1c$d?elH
zo~EVMA&N4OZrLWOMZ%-kuc`HojVr-}W**_e6?jbUu6(-T`09HVek}aez+hD5V-+GJ
zBfoCB=fn|kvJ5M9J1{U#C(q_9XW)jY6o{~t2(gP$rI`eD4nN+$5ue;<RT%YI+zGBj
z{>OLAX%~X-KYm<&@4Ebo|NPGzz5aUTpO^RO-~V&}fB!lUt+0~@v$;sPBPILq3z9E8
zqOJa-g-?RK-GAQcZRrlGJ^#Gv>TNRXr#{>yOpFf6{_Cx`$^1CQApLJY+X%}4?|ife
ze-`Vn>#-em*vB5Q3Tp%!BCB8tTt@biO00sB^Wt&em;buuayb)FSY|?rz|QdF#q5dR
zLx+ivJZ(#)pRXGRT2e0*>jfOxE`<rNubi11y0&j$^B*6XD*~EJ1EP;W^IU@27?t2z
z5T)?@tR+ChJqH;uB6@P$jhVA-mmG<ImPW>yoE}2EuUF_&0l)*bYC8`9st!aEDG?H2
z#PCFp`xe%fm6Zni`d%laoz1_3wh+dIx?N5G(2xn`@^N3~8-JZla*fB%@v5h=ZQ6C3
z;NG{-A2x@?`N|-!I*_DZ<$1sE*BxQqyqTN|#$ED)I{`%n@pm0O!%qg!b|IrSfKUAS
z>lZACv6yxyN6$#C69T#Py4Yz)Opjh^D+t@TQvtr>)2B|Ix*7PcvXXSC$Id?pKIJws
z0a+1KQ~bbD6-W&5d?2HQc9HFkHR#}P-`LSJk&_sBLQDP{1>Fx}&-c84d?*OU^ozLS
zv*SMCh<C&NaiULp^}Wd$TJq47`{*0=8GW{9mhn0e-%c(r^#C{oKK5F#>({CAXqIU0
zCFlT1PZv1Zb)L(t5n*443fl*1;fbVsI9_mqU&^NE6rev1N!6lLg5q@@Y&srHQGqqz
zgH<E2AcONLp>Lq^w42uR&GzXQdqjZF8e5pP1<eHSCpqRKVp@L#Z6h7azSYuKuL=Tp
zkOVs+AtCW0QbG}j$P^eaDJpvC$UMm!*{%Yy4Sv}cAj(BOYQ;90Rr3Ej>|FfR)YO<-
z@;e#5O;VDiEe>Fo2XFT2vIy?e0L@2e)M4xA;cvr*jOc*3yTDXpYiPVe_%t%%RpLIN
zbr7u<p%B7b0lJx-q7XKIPi)+D9$H?5-_Cb1;06K+z!QL_VCUd?yq1$Z4=^QrhD#y+
zD^5<3o^cq!QDES9+k&A#?gP#_S38eqe}{{yvF8GY+E&xkccWn9$v3JJ!-ocJKH$+g
zWiaQj@Va-85d&QXZnO7cLV^+=@ywR{##<M5*>k}^*Z*_WZ_Mn61{MSs^z$nuymYr&
zn^X(Pe6U)s#wo|7@v~i@y1Kd`%E8WfANG(8=+YqL%5j?nM2%>-*>Pe4F%`_Gv7Y4R
zGd-6*2#1k2B~&N^aIg#TwWm5{C^aMPLh)8SM`JB#AD|OLN!hQCPc2~s8ho=%TUg|v
z>a%%4uw&=q;v(=8Jg10_Ks&1%MEw-OPfjv;!2$IN{_%*qujpqsi(KA7pe9}dsxe!{
z&7$xUXEQI!TOt^q8L}K-a1>gp`FFl6m$O38fhz1XKRrTyTRyp`>?{=ka-%a96^Z<^
zrA*g@rujg1dt$69_54Gilq~qf)KfH%9Y4MTjohH?^XJc10^-pGY*tt@Vul=?8X5Sm
zPuvY!)7>wx1^3Q^HhvbeKuua5HhIb0NtnF5H46>6Z|@-M@k<Q<N}V(PTIbI<z?Tf{
zxvJz)yIb^&)lYhA%XPI+=snnI42PA7#kXK^5fR;ZOi9D61o`wNW55!z;{D2z;@ZU^
zd=t_htXkSBwREUX$;`fBHq!q|)^Ew}*>z0E3%G>-6`xzSIWy$;Omy$Wkek{C*o33^
zC?rWKM@iCbWF_PkmdNfCPkjR9p8><(#;#}<71c`G&YJN?@)=Uj%EdeKBgspf4j<cA
zKEL~4$u;W63K}{*T!y{`Yt^J@E{|g&gD!Dyo}`VMG1`iDLj1A_yMSIng5~m~xTYM)
z2(F!lPGjDHjY;!NvLsmD+lGjQ%hi8w4$Y*t(SsifxChUFRkg8qWcx#aA6mOr{1M0s
zZtHF9G2eh@m`%z<-0O37p}cL2&y5>`$RxjiZ{&)HIFTi?1_K8ta1bMg>(H+V>3WVI
zS%8`QG-l7?H23E?6j)MHqShW=bHx19b04w&9cbRZ6gUgxeRPW4!Z3k$Nz8Ivx#z!?
zlPG`8k)6#pqi`R%W`w~rc+T_btU$k6J37*S)DFu*N3WJ}_8J^jDk+)=;00Jk4tnm@
z%@+Z_eS3fA4AvBJQNQ@Lz}Yr#hzoLuSc)UgeoxIqlII7R+7D}lwt%Fho_lJ|5k5*1
z6FJ37&Rvo-5oTrzu?isxy02e}iHMPf(oy+eU_f#+^mb^P!N`o~G7MyYPfn8912y$@
zq<9!+pupnMl)y?4Vc;oJ73$Q)c(3A32Cfgdb+R5KRabrdco(Jk{rmTsvu}ms)Dl&;
zh}rofuJ7|{2mm7|Xryf|Dk>rwY5r;a_GGp|-&%dJ<)<GBUgMGV)4y~{20x+oTkE+n
zb}Rqmsqs0V4I_~!XCLj__b>3V;dsBy{}#&mcPIgP^XLzf`xmrX+eWEi5BQ-=;r};;
z6BxHgA+%}de?dWQS_~yA?^XT{acvWl`+xJZ{R6wock_BKEnF4}XOiOb1K6$y`-*wI
z10o_j5I68tWFgfs8G&~Ffa*pf1MDvw8ynpJCtz^Vgc$1@Ktm2rBJYt6SVqPHcydiJ
z-HvW6eh&vA1<0vHwZx6ERe)+cn{Q`Yw7lestXCf}gapb4zM;JpwCPk-(4=9<z(AN;
zOT=Mo1BY9y`j;Z~P-M{osl2>=&OaakG0lLk&=~#{t&MfSY>Al5grfu3IG3(}>|2gi
z5E$>I3@~XSpxiLP5eN|y(8UCOWa8)4nrvX}5~TG+-PZ?UbwW+J13*@fF&KdhMNf8C
z%^m2+tt9=L#<SGEa<UT`HhYSem5E=Q36=vqKq3fJG=SFN2jlnPsXM0;Hha%KUkUaJ
z+H-x8yDKcyb5Vw{K*yo`V&YR@i{C>kOVleAL$%q>%R4Al^}HtgnfX4hEwX=|l`(lV
zJE~sT0dKh0_qe#YBv#Vyf$U`V)z^1*tU-mJ_I$hB2ry2Y;y%CUdq{&kNy70D!4xa*
zI(3>K-B&X3jZ|$uWZ&SYPoJuF`(d<@my#4i@xbg@nB<2baO8?A5*Gmf-VT4XTRH<$
z?l~0BI4uu^%JSYwP?2$3IS4B(2ut=A<j{Jwq#`xNUyJPW4gz$9kV>l7XP_H&A+yR=
za~VhzhDm)T&CQ2ipHRiP)^n^UU+qXDcq#>0Ewv@VT<t~4Ofssit!4J~Ts~c815hEd
z9I=8)POfU1>5f42%@-MNzLNf)t$%j~Bdta5^TFUJ_o9|^sjU!%CP^y4glzPuJ$P=4
z8c644Xo-99K&HVd%pme%MPNofYnX<o&f)U9JCV1(V!qA*&n1w2FX24i6H$d}>v%Lw
zC#jggAR7UyR(F=&3LzgWbRdx9(3w<Ecmf8epdj%=%Bcq75sbW<)KWbA<FT3{#}S*D
z6??FZ^F15A9EZQ3BW+(-T3#vfg^h39LD(;zdGL+?DJ+J%N3M2wWnTWoQvtS<iV4`I
z7a~ZQUqyTS1wb~sr#-cgHMC&)C$U*9p2Ik30(7Orzl^Gv)Pk|U1oC-A<V4Z^-O5{O
z&cX%|GCF!y_MBbu!KImQhU{VJoqJ%#JckvIMN539RWC^`@qN@x&|BcIrYRhV(b63R
zW^O{T+eu?<VPV7cQ&6VefJ*i{q8bs?G>r#ND7H#Ejq3LT%2osKH(5Z9t+f+}%4#No
z?)5Brb$A67_Npb^&3A|UCCpE@+5AMarlic*<+Rs-Wl)~;6_u57kI9FrMW|pdH&*)E
z3O@)^DhP1$-l~h2my#(k@D-^JVYt*U6QfzNz;1pp>li@Ff$38nVd7yfjx%Q_foCB!
ze0wPVH4BeBlRd`-+QQ>p4&-zBWVi9iYgUzy8S^V`V1%$g4@YlfCaR}jY;5d`Js=b)
zIT)`}=J%W(H?p4xTh{ms^Wvu&n~^P;-QC^MY^a5S_}cxDIBf|j-gFl+@;xsabzF1r
z`By3AxECs)GE%~;Oow0EJD~w4U&*uSWjgiqS^OS9azA4MF#T`}Rh~*y8R3Bk3TkSN
z=es}k9#twzMRiheb1TsH0O%bo!cz>42fw2Z9ZNODWc1(^sF<C|u!BX=b;+X&Qpl`J
z0EPwrDDm5w1{HLFc6ZfUwx_}Sv{d$LWyaXd{Wi_Mg7^vlxiT12Sf@FMs;b|;+grP`
z19udHW|9v=wkjrzf{{M?NbMNQ8cVSx5hy3TX36A`N(JV)@bVzoHJ+2y+Yvlg@4O)f
zJM-r_?J+fEeqsEeAZ8k6&uuJz9n+SKU2foQkCES$?WSG$s9^$+SIZs-wz`~sr$@pI
z$l+0<Ch~w?K<^17ufJ<(%*CF!_sDsB!ra^(onamxp@fWCQd4sWAod*42Z0*iIE}%f
z+h~_JgOt=Oa$DmCdIx|T60<;S-m0O-#4?B~grJa6{Wn%FuC$^va|OTxLN32A<B@~N
zHP3LGnD(6Yf{JX5hZqu)%L}%2tO6+$V?*pUu<bQkpv@XIM`Xi;`3|OXc$U^9&tQ%^
z5~%Aw4AYoecFwoo4O>7Ad|TyJRl<1e8`62&gta>z`ck0fE)bx=iDat;C<&A0N)Ue?
z5Ua>>ceB};iG`kbn=ph{jyjl;p3W^M7XI<$Sv(YgG~A+{Qk=H6pFM+zuROI(KKpEi
zi_5Z<k|(AQ8k^s>gb1Jwu0Xg^35tqRa&QhR6FyzW_hhb$m`=fT2>ct#y;zHAW<|7G
z_C*Ih@H{CsL;u^ix4|Os1^0-7KDQ~I7cyh#;Glwz&KC6js7t{}ft$8$d5>P21dhPy
zt6q*Ug5x)-kC7MmTzWDy;{<~gc|VAzZGwV#!Rf}$hJi8fMYotZ>^eCZGM9hp-F*TI
zdnAx8fuWC7N)gT!Zl3nkliiv4cw(HzP{-g}UMUSuB3U=)bCF$u?884vK`nE&@83Cr
z0z)SGXaz3-K2lx;c&LD@jabdV(5!HChv%MfF?-%*_0%q;E3(PP0V6sv2q@_ri>H<l
zAOKJ+vp3S?c>xA!H3&e5Tgb#K&c>bIjI1P!PT`2>k~ms)vgCun#Zwu{M+kuSt!+H=
zQ8d(Zc~OJ^f!%xg`%+G{G&EFqff%!wb^uSq+o8QEF|bYQ9f;vIYuA>)dv_ZSoacKK
z;sW{O;X7l8&4E@Qqmr0ew!sw9q$}wW$7nn@e~N>ZwF(&JW^u7pU6d3FC(PcsppIaL
z#aru|dXLs3*7EuHYDMhW>oBuC^_&mKQIrMg$9^8|edCRj=>6Y?@Uh~V2WFX)+buMj
z&9~>_EM_1rE-g$blX?$&yN`GP-3b2*VBM40Vn6|%0;hY(q>lleiQDo?<b&Xh9<;kc
zN&l~1`8BdDk1zmRX$Vf8OhJn|PeJ37GZp#mBX-X|U2dET5kEE!@17_N+2o3;X7t-w
z^t0F8QuqhK`>KUKdUO?@J3MvmYrcK*T*v>x+INRz-S>Z=+IyEMX;BnHcBqt$WXo1Y
zRw%RVb_iu;h3pk7tIWzME7>7rZz6j>uTNdqeLuhFpXWH9<N4=0?)&PB^Zb54<NbcE
zcSn1B_;YdNuCK6`{Os<Ip3KHA;+HuBsEPkG5tI;x%j6DK*zuIi*5l3tTwHrj3f}Uk
z-_`Q5MS3bquOG*nmM>*+lCl3imFgvi(;N13DJf+r-047)S>+`sv$^eJIS8K#f(GeT
z1&i*dLr@267tkO!$cI<%$Hi+@kqNf{WK+98yG|@hQ3}L$GIl#~50MiuxLs^S!0N^g
z8M_<6%zS}e0pi-YX)o^wG<dcAUbc;T53uZh!_vI$%GIkMaYH&UwOF?e)f;-@X!#9~
zH2A#DmtOGRfrJ2?|0PJ@0O8krk%9M|Xzd3EN>L;oRZBlkNi(XIYbpi&ASbm*jiU*d
zY(hef2Q8OGbId?a21-ABY+oX~1{=Z-2nJM(2TJ3oTqdH_&<aEW=0!l(YG<5Q@%7h~
z3cTw6l!VNEYRCRI3u3+i>2WXXp~57%8m=_J!l1$Kt1>b(GuKW*A?&A`=f|ul;SgMt
zbua?ZKfxk!OZE2ku_Gh~FDPbs^O`jvC03zcm49!p2epX^7A>kHVQ4GhwqGSIER4$c
zmC(CubxqphLVE|Lis|d*_bj7Bga@7iq=2hlVlRWgQXJBlXfIq(;zD+uCvnrNJqnQu
zDc7X#-n-X;v@#$Xmz(<WNNPATQ(N&smuIH6W;n#m&dvfD@<v}yIe|cxG-`D@={ht5
z$Hlcj7ZiL0(&V_bV6*?GSe~cnW(El#%A#7+gAA=beK`5sP^H1hdLLz_{`gb#<LIgT
z>ys<+7WwR_Phit4F2z2pxrQYQy{9z0ZewMoY(mbwpPiXG$bu_if3j^ST+u}XL|+PR
zdcsOxZbCu=|Miygaw+3AOHM9bCVxuM*nb<M`(ld7b%EA`0U?$>Q;hVDYg`1!y>u<V
z7y*I-o^TD<It6bP%hbtW;`9pWN^lp@id}rE@Je~B*xqXmh_%4|RYa1}CMT61CG{~!
zyun|HOZM)LkQlN-i=9Zq7E`Y`c%hW1auF#e2SDO1HhmH9E7IK}#Zx5OqeGqp0ZB-@
z2xGit`E2J9&!JYP)414h{anEdAVwTqsV{<q0|5W+WL6O4+Nbg0(W7dpQ*a%uI*pQb
zbZjgf#fwouS$gvm%{|5M-@PNt2&fkB4SmFhAWQZxbm6W5Ll%2F7QPpL{#=bOKx+Np
zb}8XQjPK!L!G6V|$N^{rSVlSIRVPlIxB^>sfGKAtP_dmb{2l-`1e6oJekm@<G_TtM
z?hJ?iKHA8<cYm*k(8tgc5*iYY5hb?V$TnN&I(0FeYzp4wraTB#8s!qOs!*GK3E@Y_
z2OH$2nDJDfK`Ry$s30cvYN&kQo?`07TNWN^Hd`gDI;WrdpO#`po8bVFsNuv!0H;3e
zs@;UG#wNgWv$waGJM7AB`eV)EO(`|i)glp<FDngtLC2^@qpZM%Z?!4?1^l%>fIZD2
zP)-~wZZeZIi-i}P1F|+;oyEZdJkZ(U3yLnVwx&ICj1dQ0oDc92<plCGUtqCpMSsC6
z3W4#{K4qd9NAA}4y+B=hK;<!Q+Y3ucfA#|!ldMejJ+9$^e3Mu?a2$W}W1~mM7l~pC
zgis1uT(io;3Rw6nR<3-5V`7Kf^~gF$V&`RCLZ;st8JR8k!=tX@teN9UFrwZ>yJO{e
z5r;1G<)kw7wY3~Te^zqX<W<+AjRUi}8X^y9m{Pz&GD_*iu-yAs3lQjYu<dz(GKcN|
zJpb(v!o+hP^(B~$N{KBDp`Y1iI{S*Zv-}@On$zX=G6cQ?7D^?9>L2G@NcW9bor3%=
z7cViESFBiS6@du`zOpqCV{z07T=Y}+N`8(aRCZzlYmHD;Pv051Bqp*@;2KXgs&&O#
z10Lf8q{1}ND5~kq))2Z9`p?|?F-f@8<G6mXrA-d>^u#b$;1rNXR$T4XXCs*l{j<Tg
zc6OYwrXVjnK20#LoC+}fIR`o*3+b(qIu|8zl@i5^C$ehW5W;Yv_V&z4T6g|O4R}Q;
z4P>VnScIqySo*_C(_~TNkD;1F9YBiZx?D716INrj*DTiIqLO%i9Ke=T>MnkuT6MGe
z=j7(397l!#4FdGIAT8|)7&CE%A)zT5GI#(UAlUqI#2XDs4hh7AIfMX<vKO+nV<C~;
z5tRn1O6R1Fvc%Vajf59f5Rya6dD+()89gX_NU(L>^IEQ58c7>9bY^6Vw_N$lSqW{0
zVqK2j8ajkeShKH@UkWKOOEnQhrU49WRO^DAO(97=0Gb3O8MuUm?8A4w|L`G7o#&Iz
zBu55F#OiH!hplmI)@q)?M~^|-OANft!jAK|T&dYEz0oe@wV!_7=q3tDi&QwISs~b4
z%TB=ck6-Mi_XhD_>mh6a+);wlIS!E|aK6ohm^Vr;1&ALzGSHfr$M30tvZu1TS~6`)
z6PP1Rrx&@}IuL6s^lAxZHI(83xCUcxU!;#9htR3pW7r&*pfV@idTp&QHhu~W2>Wlg
zq(+W>FDeps7(s;#<&=n)Rw#%hLk5cX5&5^StgUglT>%4JL1lJYTRRGWhUfeQBpXF0
zQfw=v9JNziB&lCmc>hiRn^oEjaVk*s!F|1In_WIXA75KXhj+L-=sYIn*ZaYhTm!Lv
z)b%V*TLL3uiT2N511Hfh5tFI82YJc;;p;#RgIV8-7L8aIz;JIktQk}6eV*?;%2$8X
zjud>L{RrYCF9V9GSeG<x(S)4uUKy>Cx`qfbLZeYH;rQtOeO*J(4=@MC{=NjKaR?qs
zb*${6kW;tACq~Z6ZNqQyhob8(lwu&uXps4XZn*;xm64hObXb6YA>|E>L{LCGP9?)4
zWJaAqoPABl5OVs_3P?XY1*`H))ExW-d%BA5TmCrx@&CY2x6lIMlHa*fY$67=ehKz^
z%@`68Wu|CoX_@g}N$*<)^=T9e*8W@+PR!)C$Kr}J#U&hEr=P5rajBUH{kdfa9{cq`
zcE+W?`sC2?qpsVq2?41cX?_djNoGKLUk|=r%E?_|fKVeG18^4i+``gQ0l{=T@$Y~l
zZm8XO6zqdx(L02nrTxGos}>CLY(Uky(_PfNn-BHobJa;KT8U=|&(}Dh7mxrqxC6BT
zaRl~TL!cjpiV?kb3AL8}v+kK;e3DEvkT`}#{sAN7P#gg|(2Lgtu^6?N&-~l~r`PAc
zk&)10KD#EX(e8;rc=%JnFx7IP{w9GOsu@Z3qRPtta`-)#r2veEP0D{p)IwN((z6NL
z@c}!M$AZJ_<uDsK1Z@+O#2vVeiS|9CsiA=hoEyPZeEW06WOp#YB+xunNUfF4^%Zfh
zk#@m5k1NZ7X7o@cxe^k<>Ev0^%toL;5@7*;8aIp=E5YM|#iAeU;LD=M{ikf#WqiK*
z*#@dze^|8k?N^PAN|0?!)E9VRsILeJjzf%w`}%GiE{cz$ml2x>qH9WgnC|7~3bASB
z>XPFSWE84eIV7`e*tpTf%}vtcyTY}j{F{t9+1Uv-g6KRy67Ub%HNbo<1Y##f3|25Y
z@95c<jKrjPV-&>3_1%9#;HNxhnm{Lyf;E7!<SG=@{gLwladYx@2%-UaaRGHcG4N?=
z(E~Mr1gil5l7?OTzG~f%A1|P5Ayhd|;8K7Bq}Ik3jYgStG4SMC{P#P~AHa2}vP<V$
zZZ*yw3*0A=U!BH_0e)~0KnqF2b8>P@M4EU>|71d?0aDHmL4I6-a8tBoXimV7uLb0!
zD!A1uQ2-z}0yWr88ll;Vy#q%<ofY~`C1RQk?w|Ju%Sy1by>xV_xj@x8qhUo8RD^mz
zEw=|t5>T564h?orPOzgdQ1-R=^n8YdmZ*?%{v#+oPO$%N2_lI<<0+F?mGrK_&%vTR
zShsUGyY>nskh~OocF<@Bbs7Xq;=qDVBRG&i@c2HcvWR**^1bRz*1M_xZEma+JwA46
zqQy)X4*o+c(4+!R4CFQ`fw1E|&c75$MJzd>TfA20{}GDiw1;l%M%fWx)``LyDiDS&
zk(tQ<XetVh=mR)HxfJKNZ_Vf%l91Na6ol$QVociJ64(g*gBLRe)e5r?xH<nrZU`|w
z+C-;~-G-;F7`#REv1cI!+Bn@i;X&21esu8ur5)$PK=6UyiOHJwy2Ycr=Jh|I5&02(
z{{JrAKdblP>i-LqqPum=KbPMXT6e}Qku7)2F4D1?T()1sku`WHRcrLXml8dJ7L{ec
zX56Sp;*I|P$M?64JUi9u;^~{JabVkj?$19D&HRMBm+!(_?`OyVK{Nk+jsP9QtIzbT
z{x=35|N8%z+=M>vtEnjhP)}v$Z#k>)w`|$#x2cCgW@U+NJ*$tk9$iPRrE9bMY;R`{
z17#1tfR3JFK)+|A`$jk#v`;TBs)K5AcyRd7$A=yTHc!$(ywxjtjSF&ds>#fU?Xm}c
z1yCHJto*rHja*BYB#lg&Y-RSbriRhNp3vWJk5*&<GL;C~Pd~+FAd4*#WIuCM@$F$$
zp@&h^=Z6A-0}+2jt1eN}Q9uBJy`C2LR{bfN#&Dx(55M5UE`M})a<VfcB?9OUL@JMQ
zX7A`i{UWK)!^6zW`;^U0v&4s03jgN4`HXpyw8So3<h4ayaBu$|$YlugEyb5vso;$g
zOZV3R5d{T>^a9`Q_kZM+g5a2bCAx9&x(SoqyA$Mg1<w=reDnJC`9zKg6{X4bSmm%~
zknnW@tfN^hgWt6}#>R~Nw%qL7Olr{^gg5LJgDgVnJ@Bolfc0XFzaepmo>2je<HM}7
z?U<w_B3?t=cBfQzN&59Rq#PkncmCZ?JYDx~g>KcI3CX`E$4Z|2n=d;ri~kjF=*{SJ
zTy`0rQ><##x<gL$?B&8+dd-tBm7>WoM3Tz+b<{<ST0%w&$zRK8IID6t2lH7x3=DO6
z=o`fVfoRz;{qh&=%khh8`33Z+1xN0R0^Qb7n4d3PpV*4`Y=Hy;GV;VdG4W5aIggNX
z=)Q4QanDasBWCXw^;IGigMOznkjsJ*phOf7?NoyXFAV`3+%;0@{z$k7E~~*GX(SC-
z-zJtmIO2lGw{#OVz{CcvVavgSv^`B`LycV&CABB<Q4TKf>TeCCMpR8VI$yYMC+6N<
zMX`n*i@`~z%!)}bhI^P?&t@cPG$2_LjrmJhx^Z$5Q4CIL8eBAooMy#YmY{#3<uNJ%
z%SNg|l3s(bQ_v&sz<$)5()G|x7|jTyGy^Ayd+BF?zpKXLy!w0l0IWKkgC!wITt)d&
z7y9KSBp;9YFI@t4h@PGVTml5t@y`;#sn#KPnv$rYJ`L+gtJ+1rLbzBA8p2mX6ak&W
zjz{<cIX0W2GX;p|&#AvDXKq}!v^CEuRCMDW1cV8qWfLJaz3mXKnuGmO&1~LBo_jx5
zRdF;|jQ{=}0R#MYIy$L@<UFF2QwM%D)RZp!p(O;h5JCO-vv18oYjtu!b>=fY6pJ?&
zM)Kgfx|CK%loN~%+1WWz_B>09=@aHNpFV*C5fQ6ZNg4t~3WJZ@+u2>z$Yo3{!6|~L
zreUFNHCE?4(<<aJ4dcmcHsz;EENZ~&?~a6ttWeJubJ@ngYdg3bWUJ9_jp16u8GJw*
zL8BdLAl{{$NFp7fA<v8+hd_kmy;YE;M!;6{+|$)nyRjyY2*-e=f;KI>`(S-vWyEJS
z;enE_Qt;)TNXsD-<xP7}#a7R}Nqzk-&fd1gZrLv|d4`KR(@kSt&=<->BsI^b{@;Zs
z3t}hC;MqYarEiYaFu{Y@H`x8q;p1HpaW}@lY=w439Kz{QjLON%Ux-Ec0;wm$Bf@_P
zT0RDJy^LTEk0yuKu32+sDn!udsO&Q5JFQL?Al!(rM$mRrVSo?Gi{(LF9muBBX;rgj
zGHJ*T0BH29K|l13!{<%-om96b2Wcj9=e+>(Qs-625J>}Ndy%I;$UFP$1Zsfr@&mUQ
z=lRm;=)f#>Lkg+zO2GvU+};wu&w!U(MIP1g@4wmG9vkWyjsxPAN*1@|S1r+7Wg+Eg
zxoRz}nf5z_DcWK=#|>#VYQRA3qSs3ws6Z^tw6vG3eX-}{Pt>pTBU7!0^+~~*&8@9u
zfoU5W7MH7K0OduM*^q0p$GAQ*0%elX3}4>-C;Wu~P8(MAn$lC8#tx$^@gL~`$xfY}
zHMW|=O}m@jo|j078pgWJ+#M1;H>0)H>3*d`Vs-nTlU)V4sKJW@6prhWy#t@k;wdOF
zw1MQ!<2H|(pC`)-UmEON+sKF06RNw+P@4)hrkO}~F)Jd?CfRlv_pz}bE2-OPIQ6T5
zx}Bb~voeAO;DT?7FFQAP_q5A)BRbbVk&yQ3*RQsPpC^o_gEVAi-!p93pbbFg&|MUL
z*N#%{rl!6C-S=Y`4+LupKy7kBZM!pl&0c4R*+~fb0wMO=C+zqtZ3N3I)vVs5z0ma;
zxfOg_Bu6xJ-q}vXt1`x^b!ES|kWWO7Go!4c+Rkp5o;d`TsBMmF`BAOhQ!``TQCM>1
zCE&O*&f$Uj45uhUF2Cac_u=SfaHT$y7fk1oW)$^>V``{+HU1b{I`4xXi>(fR{kczy
z{#Hmrf~fM^peunqjJ0-kS*Dg_agAX$N6n!#22F4E)=!;1dxKs`-C_5NZ`IX09nn%N
zTHb{T$%vjmk7&O0fcyFui{4vXS{5`k1fX-yvpRaC>RVacAyET|P>2tnOpjgwJ+6XA
zC{sgr2c+@4M~hx~LS8_;_|C_YL*0j0t;VrUmrKtpb8KccneI{W_Ht78LV<2r)`0^Q
zg+wKmLLqPk&?=RcA^7;so7>4Z9R@*vL3%|r)2hY7&H!d))7ANqTyguCtpQ1$BP$KX
zXSKv4)do85CqJ`nS_9|TuUpp+=ohr)dGfoU@rDwj2$d-eE@BxW*bNB7Edk>ka;@S$
z;0mCQ4<RZyQP-$W9MM?!-i{h)2jBo-ZOmF=f$jccFgHtW*z`dY+@UoITA}DNjG9+-
z(QXX>@q_X%a8NU<G>4xBC-foUJ=n1ZwScxTcW@mg;c{Y?tu)SdUh8>Y2dlxh8B`!t
z)NCAx@&NvFX~cnAD0ama;HrO7YeTD1GMXz<HT^Izum9EI;p4Ltm-%h_J^@0xVfyW~
zRHT);1rWTST|>b}3lM0Yla}6`QNOVdoj}dpm?U9{u(wpk2a?}GP?Ym@3%_;m2ehj-
zQ?xWRR4z;YnW?5yT@t>^Db_5<mgcuREMLBA;n#AIKPBUD2DZ8VDN(1#A>{xo{17%J
zwEvQ6WsoZw4z-Vk9GaURYR%Y;&jFP5_jsQ?6l-#)QZ?S2KY&Q24chts<r>Xwc`Jw>
zrq8iEr~Y@OFOxiDVJ_Mn3hYLPR!tS&M^Y^6S=a4i_6fIFusrD;cq&0XYF$uv_HG!U
zO5P^-WZ1U4q@zh){xi5J1?Xz+G)Q!4P`}$f>F?0Le!VPnVn8M6xXKH4ONAY)R;^O*
z`MU9Ny487EgT>555$)t^T(TRE9Q7I&j!V<PqKa2a5QSv}jM2r6khLA3e}l*DE)Vq#
z(d>^~3rq|6+_cqb`*o?8J0SJNDmmFhZ)sTl5Y^q)XAg4HgjND@(uPwM^(IGQZg2fm
zntk!GV&;#8>TQ9S=2+w+*8phFhsNPpfWXyH_Z7VJ`cU|Bzz^FUq?r?YIWHG)`ZC`1
zPw>^ZXT}&VS@k0(YuN{ErknKKPuA@`X5QH=3pe{}9KHoOwlS=VRI2kQ%@0$s`U@*7
zpE<ixBMUNGsTqzV^>u4rg5v2H4itH`;Q-i~<*PSyz?wn+HrNhw=|D8YZrFYf5X!Qb
z+|dA7+0hR@e4hnptQY{iKd~P{)zhDkAF_1$i-XwPLKlnyBKy(Dt(q9%w5vOg#q863
zwQCL5R6d5!i7a1^-cS^EvFk3Tvu6NI4%AB<ADjh}zFg+8O2xBI*PNDpJ9hXC>oaZ~
zu(jyi{&MpEd(<mt?hBe?Yt*>SqHnZ)LWaatVi)z&>@f;3YiPD#)qr(^vC5yvhE;>#
z=348(>f58kgUwlfz@Q;}JqyH~QnS?(@_xM^-{0(Dk-LBJ!Fuo3uO66|bCYTR&%}x7
zdTXkSTEeBR2)Zffht%;4ZdRBVHxp7ItxSoYo*MvZUr%W53Uh0k?qLt~|KwK{{AyFp
zsXABZ=I*Z7fU%AP1J%gqvC3kd>Q7b94r%2`oHW+fK8qUCweY$nK<>{r3v=GWPL64p
z+a7)R>EQHp$#IDeBR>*Nq(9{oMLQX>1KJ(`Gj<Lp&SK0JGHeJx@47-+XwM$#kqXc`
z-$tS`GC(ZNEjmCI+_Jkp+A&bD{{y<$m8boZG^mg*UsO@C8B${)G=(U@Qu|F2M*`W^
zULnghe%vNCL*m}&7=Z)uQO;`G7}qOa)7ko`feU>e2Kz*5*%Ms5;X(=^NL-vuTca@=
z#B#7k-1epJSr8xCG6Zf$ot$#3cpCuzhW?Ixyx3n;Q$63>@}Y^^Hfoz>fk|Hq0~}5c
zi(1+2M`nYJ`1s$V(c(7w@dc<-SejXNkvj##W{zh55FrPN&zFL+bFSc4-%U#!t>%bh
z86(l?@f|W5f?E3>0MlL@$#t0L)m{O=tS(MVihb=ncA$chh|aPH=P|++{bFz|N^*R-
zcKzJll>yuinVEu2XUMt+oH7l)!f}@4yeN{Ol-r}_FtKyBg9DRRu6>}DpUn*<kh|iB
z*=jzOq?B9(yeBTFQ!|=##jW=HYjGSH2&NqIv+o^w^!oKf2-vW$u8sS%A6HA?AT~Jk
zJ@%>y&Zl!dtQtq<BE`5b1aQLRaKiG-79<oxk@+gnjH($W=uPCGZ=?36;lJ*$G$Icm
zb+{#G{ja(MgMBlF`S&ELsKCt{xSN)njtEk{YoBVBZ$_+xqp0UTsSzOuP#vN8Iw^pL
zEu)WQ-OTB0by@;(y9((Rj(gdcp>R(Oa;cA&Lij?ghNO3QHZqs?>$Uh(AA0Z9GM#W9
zSf89F3l}%{s{P#F9SfX${ynUbH17$?nECJzK3r@Py*YM7eu`X60_rFwYPutw+WOIq
zEQ%xKpWV%EbtD8;?GTu@Xq?)&Od1PNb%+yHk%Qqz9!XW5QIq^ibYvgB3f*SlvHAkm
z9_+#S)(;eX%DPvY?ocrCZ<IA!T%0rGw;Dvhc{D*OS?STfn4Q%)_yYuucR}TY;!?@@
zLFrVK@*C~>w3~;Jpaq8k*Asdk53pcR^_Rd0WZ0BMhb>a^{JXgq>cihAt+~*M`IW1s
z8(4jCadBU{&5m!VCFO#YW9m&SqC$de@IEv~Z^7%I9N;;5QWwn;O5o3hg%8Aqmo7rK
z1QbRW5Fgwdqu_r|HjMx&<-@aV9q*5t7z}*@-y@*EvdOT#;#%D1L{#&;Tk6_Z|BsNh
z5r=^O*B2l2^Y4Wok-i3guz{}o!VqvHEK4`Q;_X8AyoVg;WsRM#4cD5*L)A$1m;m|S
z;q3i+Km=<14itSCfuFhVVEROq3&RUMg7od*S`TAgOm{PXKpA1L6tN18n`K`kSK-F}
z?Gsu#aEtj27avo3d-&;hk>uNMyKpSr#27m`bqQesxzPB0_a3h8FUITFU{j%z`jt4r
zc)Llfmg7*$n?JDBV`*<t9hg%RDTY%vUeOJ&HK5gT&K-9iK?%Oj&6P^{@ME+pf$CtA
zjQl<Ykb*1#;I4qH>7OS*@{Fp@ADKzfrx?9YCA?YTXOhx1A*I?-|MBCUw1K+HRPpWL
zmA(7zYzDu7%y5{~)-x$jYR-y#Uh;|7=u|~$b%r!Ou^wPS`Ext)uUa*-z%C=1Yd?bu
z#(DI^vxp08-`n{i=$k58Dkv73#mx4YZ|>noFNO*Qt8nVeWm%gns;4m*{!W`22uT=f
z%#dau(X2OZCIMBSJ@z>a=p*69A3jYEvtjig{8r1hdAs6UvcRB%A0b#inN?H0-B~Ed
zE+(^8zjPNl_?z+%H9cC4CPHe31kGR=j&9ButcAZle`@)v6W7KKP_xFO1Gj!OEjx?p
z1!(TbykwR3zBpE8pBK0d#$Q2xIO{Y;2W5#owMIFJ%j#%p{pQBBdHnZ=>8<0)V2u}X
z!eEY9zvo)h26KKdNB^TY=EGz!e1&ok-*%)tC~fjcIolN-<Dj;vTmcLe$nc|XX=&l*
z<$cX=IBoX!Hmg^W!ya1N+$n|v9c}GY$BM7Yf8nmrY&h&U(4Kt!(`iW$xf#fZh}^Bj
zT&48t)=kI{<cAD(b&S`)+_b<lV|#GA$*^qH@8pFmqKr@Q_x2JQ7Y_SVw|)`yRtC2Q
ztQq2{h`Qux9=-><D~g<xz_lTr$R+twpQuV^Y4ub^^=m(-4~#r732>=Aq&bU$%6~05
zQQ~^Sit#ZG=P`WdTQge&h6ET)c}l(b;5gW17&$RBZ~iobiF^A7<e`9vh_kObf%<g(
z5D(19u$2&ctx&>OZV*`f=jP^MoW0+|Z(9sA1YH}zMmw#2B;4-6fzINf%Nhdvv&fwc
zoj!77{&Zq_W}$~EoF6MHJktR&D678JAJL3gN(?mO14__GgaH$QTmaJ~f4c7h0wD*{
z2Vnhx6OO0AL3cXl-0)!So*!F}B77{b)u{(7y`TR6e)co;+xL{BUf$sT<gH%e@6Fm<
zpMr3bi!s%Je&%U-!kheowP&Q~W};?WY8Zt~L7{>RyFM`rUhEEZ4)P%16Zseb{t-2o
z^x%N$w_*qpywUde=}7c5`KOvonJ;Ct2aIHS{8MesLwrVBPx*N*j`xvaj)qdEta6vJ
z_$e?-1!=c}=A#@16^G-X0v#n>zfrSJ<0uA}$$(wk;9w2a9f%GRr5Ze7gr1p#gQ$)<
z3-1&6$z8MpZ7Ej9nZ<datSaB0Uk|<c_e>MDbjGFO2lv0dGL$DY=CuGYjvz`3IyPKv
zXG_I!EYdG!=Vbd2wS9cmkWxL^6`-l+1lR$mcTr1=kVxnr0HnwYv+A#l!tcN%2EkPf
zG?Ga*^=OoQKudp3u{ZFcZztS6*vL~x!WoGfg%b@DqRUR#@-*##BM$Y?&3$a_S{mY8
z;HWRiJgKC*>b0+(S#&TwY22)=JDB9|1L%wjm|u>|*K9jIz4XVkf2Ii`k>CRi5-6Ux
z)6xdEw*<+GPc60N*czragDcxkLnDsG=fkFjIR}0hYnQYfwyZrZHow^&UMyg+RjMXY
z0P_4Virbv}R|^mjKyr!k!lNfbP|V6lnd_k*icV_=;ejtXia_NQOg0~G%?s7`O}P7D
zU0ZMO7Z3|7IYz#9xsVL1t7b&Y6}m!f!z}+<{^x`g=dY2xv~(>%&?vog>i^xw6)*%k
zEzUdwiV*+1i~;fSh21P_=(;1~$4^&^!&eIy=abIl^~6o9pgAdX$T@vArMvu2f$%uw
zUjd+j-Vk$`Ei~YzqaB|K;#PN;0<sVt!8eIbgz*J5y<y);u>=o%k(4VRmqC`Awr?9o
zynd8EbODO?L(f^X7VF|%W+(gdvBs41t}aK5RW<rVH*wh{)VlXyn+~ZW;YXf@IIjRE
z#zN3<=P}&D)8nW0{q1c|T$oM+cWKbW5Gw;jaRuO9(<aMr3HCVtt~a6E=IsV>zvlLU
zx1~XA2tE#zQdS7>)n#Bq^gc|r#>~z*CJRg|T#A~aKOI%#+wiMiD{k?y;1njQ25v+=
za?37vCU9q$S+$DrAW&Nd8U;jlUYVffYLUFicOK56=1l9tl1AFlL^b*2yXP3O@t^W4
z=0`bG5_iUFW#2l3ZY-{&^PH}3WTqhea<ak;_d`Nj)E;e#>FO!=uD+fJ($$AmgZ#Ea
zNIa2O;1ne*H`gk%>(-)ZM(M+^Utb8?aTo4&E{3wjJ%IBX$@Q!=`~K2Y3*Gth*{RRi
zkWlBzMjFTxfu}1D5}+%$ZrjE9_wxR1KTk}Yzz(oyx=U&km=?XrO&-W?_z9&>UrWSl
z+|d_bJhh%M0@?l5Cr*B*?Ge3m!OnMs>vy~a%CT3-{9Hmvq{Gf*iex_h>Gp!?iGkYu
zy1M6};$>oQTuoBXg`~`au!C3<daa8K{Zf*C|17eYBdn|*reMplq)|N>)B%&U9xkY=
z@>2EixIQ^a5UUOJ>~|p!)U$a<PX~PiE`UQT*X{UFShy0fJ91e+;w&+!3}0znm!$Y8
zSkPhChoHzPlL*iW=<S`u95+A%#Q(W2AbzP_gWI}nH%#weI+^5mkMp<r0q%@=xRDQ)
z3o1d#uY-uv({aHT>*+f}CAI1jV(Mh`GzGEHNDOUb!%wN&c%?!>^SG5ofRr+!t4@N=
z;C2}f0YGZ}$1b(10V8~ZWB;ftNqGSB1a^|czF-Zmv*Pr6L0a_sX-d93`!HI99bx-O
zP4hs${>w_q98YO{+L{nWly+f?rO#2`e*-kVZYOIQb|SBB=Ni&vPklQ8+I9W>Qi>@$
z&NVqsT8U-=m@WJMXf&$9;X@T$4#_O9_26z&^({Qy7&m!#5SJT%x;82U)gggHyN)Y=
zc=twUbZpXbL|z|vgD1)_Z#FF<m~)5+ju(Pi!MvG=BDuK;IdVnI{@SXETctpBbw`m0
z$>`+e<CEodzi(=4Z2WKoUHoEFX^t&m7|X$Lcc8xhg|Q;FZA&A?ARjw`Nr;PyeP}9m
z1K<cNv*+BeHlqdz?NgwHotR3|{~9}#eYVh*qU;~UX<zWz(lhP!;ymh&8$YesK0c8M
z+{JdxUS=(s#Qnmkz8lJ;<4Z$4i7s@WbOJ?pXMUrL-GDN+Gv&?4IBOGfVf6o?q5x<`
znZ1sLF|%h}@xpnw+Or{r@5ZnFp8JH53+xI!gIJI;<1OrYH5b(yGBtqI*R9ERRE|n1
zjKCkXM?wTI!lXneU}Lm#;}?VDf7;jfYIsKSGt-P!a%{Pf=i%A79AdJ&C5rLivMwiE
zZWlV}01{BU^3wyDYVH8aCvv1CJUsAI&Fv_Dg5v9hL*vQxk?$Y#S8x6TuM|F`ZW$m0
z+e3GJ$6xcA%v$D3*Kziv2QTv%3dY;m2AL-kOwxj1?4GEW?!Tw!_36R7IE$B2Q8>4c
zkB;jCV}61hk&JyKQKf!qw#>Q4L|TqdKy-uHd|PQ~T*AT*tX#h<a`dtmyZ!VHsOpzm
z+Kbn+u3VvO@NTA^K{8dePy_TDj!BEbURgAS9!N>R%IJxaji~9Q$AU{jJc*5W`RCOx
zf|@`ODtt@*pskBM=d#UElTEzw-8-U4-bBymH7rw6qpQDc%eF_$*e1q_-RK7>9(V;o
z4=^cao;1<Wp!u~hwR_K=ys39W*Mge?nCM){)Pr51tc%C?<+X9mU>^$fw0r|~$(JF-
zzYDSxa<p!1nJu*L5ApMUFMPNIpi6Eb1yY;1nFjs7A_HUNQuiR9YL?&P4F84~s3fEW
z?U`J;9(jz<n!_Gv7__{i(a5rrw9+e;mX&K~?wvGAyzf2zYp<aBeQd%TePdmNo4W43
zL2&7!`g<>V!&i#?h4v!LvcvSo2{*sv#ltcj49B)RH{`DShbTX2(U@-H_8>~n`q6=w
zc4Ro9=p`Yz*KCf`Z%`8uaHQpO1y!{`Z#PA!RXwx%#=J$!Apx<&q6M#14r!mf>%TLh
zK-2fTQN_3KcA_SCWok4HjXht=rHZa(F{)`fb62c$X5nLwFbTfU9-ni}86M~o>ah#?
z{lcOm$2@O}W+aQLohAKx!(Hjx_>n9cdLMX?u1r3^&09DZ<D=bFmuTsvlT^FJ-1naV
zhY^``=ZKm?EcHa#7Wzy`Iq$|u<&V_9@H%UMF-a?{2<1s%bBy%c{3>C~|C*vv;y;#F
zwq>MvYaSH)M_}-W?iBx<Yw<6cANoIYA4b>O>0A8oK@bO8g#Y_lu%8N!T>Mwn{^zIK
zfBKKcm*BkrFMqt41}N#-UlS@B&o^i4@mJMgm=0|gZkulgCkg*v_#JF0WNr^|3qtQf
z?6|lf_gH3Q&RG@Iwau;7-~gi0h(Y9YNET+F4Ja#OV8S5Kv!330VCMdCN<i!>Ugv~~
zYe*BmCmIn3wdJZbDf<`jhFpf=_dap~4j=B+`W#j8ZR)hq$Y<;wGJpkA@ZuMIoT&H-
zq6?-U;}sW9rYYLfqJ)!1vtEq;4b&n6>DwV{hse5GJx0zIS1*wJ-6suX!iRvq!r0@6
zcmA`t_c10tqM7DH$=5@|MA}iyNP(F}KlHJzYy%)1a=4R(h#y7c``3Q3>BO5xVr?jT
z$~J`VVs}Gs4tVGU<jf#7IA5`#h9{}NBay=>AE9N=2U@(F&-^fo+gnFI|3&lKHL4gH
zL1kPxd-juB2ebf^aGVl^0`yn}8Ujos4TgE0!)|g=f$ld>Z?KUk%vw$%!xkV&B;3V_
zk06vV8oO9G2XJJ75TRoR&3YCW+bHlp$ehQAns|zJOTZEU{aTX;t!Y2o{b>Vm(@(H*
z3EPdMYRT$H)9%@QVnYkNQHTTO>#IxLgR31$9InLa48oKEU4&o;c^=rrBBTlVZeew`
zH^3T#;Q|dHnN4tiU@9G|?XM?m1@KW(UHf*X<0OYpnT(`{vM~Z@CAK<L%0wK5jPf?1
z6PeTKeB@w)!Kn4PIduRyry!#WhYnzbXKn1c%pgvE4~(RdcP;_WhWF^(j{#8#Bnx>9
z>!dTY^6QuU0Nx&^>O+B4f@y`pH907xSH>O?d6?}DZ#21nzN~@5Qwdj_93H4FYBU9B
ze#kw&-jGC6QBl#7(o#vdDbHPY1fA|HXlLDpELsp>t+r+0keJ;?BgpWP%R(@f=^;P=
zIh$6}u|1gYuuK6jdt4ji=6o<NXITvjcdIT);=UEah9^%I4<a@$b60x@f-w#ewuB3M
zBQDb)yenhLt-{C3@Gy7vhWGTq>JllnfcMWeVZB69xG?nUUgI&Ljd)&)&iySpS(-C=
zQV8*LDG3(HfO~;6_A7Q66I5g5lTx@4Qk^r@XqSQZhZ!cn%OTMDVPoM(S@kPu+y=t9
z<0Te^pRORZCmgX3ed4y1=oA<)zCpPLr4C6L!Mf=H;(<bn4`U--z(J82dW*9IY%t6~
zWKk?&R*e#=MWwlM5<v$shD+F!zrp5=U-$ZZRII{vvk$iP)v!u98k4EGs9}D>CYfJV
zC3krkofZCo*SvEL{9R0SqZPtS(8a7mNi7M9{CGM9Aw)$_ii>0N5(uC#M>x>9Hpc(N
zYUK%0Py^KXs;R1ufp*};+e0dUi|wWMj?2(6z!M+}Hn)o5-$#n^nv|!Vc9l&{dqA(S
zMZX_v@4yMi2qFYKt?C#X;5gU|I{{RpPdHLL!4$P-&6*F8=^>Q%Fb&~Suq$c!EsxT=
zi(XEuG=LhF=GT(lB#Bl*afj_sa=>$Jk>Y$7R1A55XbLV_?Wy)YOsTE}r`%hcKn<QS
z8Gt{V>=T%^(bmQQIobhDq?zWk>FHa*cOazvfOr@v5ZCQEdOB3d;hoJ$h^8P^&A`}0
zK@vPe6m9dC4(z6zfPpE*Y$F;$CK{1Z1C<8jrr$dw)9FCzTViytGrDzXuQ3wU;q3(G
zsf$c8HeldOsKX=;I9pr#7Hrb#*;yP2yWn2@5+W!7o6&yEQKROB0RqnkLRPUcUQ}^s
zBPve-p4VuAzys{T5v&Tpu4s<Ma6-($ATlwjfsP4<1(w`a2%FJ5knl0+u5c<rh?Gp-
z3fVS(BwU-u;nu`6ByK8cqV@;%I#;BEym=hrajrP<j-eQh#0Bx8ls*kxi6S>Km7cyS
z%7^v#YJ`gxLMZM1N(aN8Af54sUhrHK<R7b$7UrFFeE)uDgt2fanrsbyNTi-GPv5b0
zlXM9ip%NMXjJO2KFR;Okjgx5e2MQlAJ%_3a>Km_4{WKIu6coU5NoJ6-ATHkxCq%!4
zV{Q`q1rp73_Q5)a#^~}MgH~y9WXpd-yCJ#57-2m)2uUCgimy0R9b2yzSNPW<BZg=N
z;^ZC>!`SXDSYJ_HoJE0Yr3YtX7|u(`OW3g$AzM6yAAs>|w5al{9bPVjGTs?}#H%_w
z`A|IvAhRET1NYiSa<gI&K80Nxl`^FBh|#_aQXBuL#dcqu$!A<2eI&R80t!r~!1>^*
zJHW-sxm<JcYBkhR2t@jV!Iw8~+@KKWAFikya1o>OQ+ape00=2AtRc)6m-iDhhV}w4
z5;rh|or(LCCkEBg8-I=Y$pFN|FC)WMk$0&3>*q(J9Rhm70Z+_16O7ywKqwKc+GN-u
znMr`JL0pW$9jnxmubOW#{D;T*Mes>{6&Z*rfVmOx4w0M!bjBgi4rR51`;E%Q4@1Ml
zmvEVbi-Dl^4z4&^B!Lj~Mf@cB#W=MgPZjlLR$Q!ozxEkz-^4yK4>;%$528czF<$UK
z!%}m@vG@jLF1fOxoF$P5{2|Y#m7z1g0onWE9>dznb5bTF-|*j&Njex@8aq(zN6V&}
zO^{TiK5^oZhpa#pqjf1j0x6KV94U6#Xrhy8XLiq1_zZ$^Wt;;~T)q2o4n4jmRX+wL
z)$F;P<Rb_AcAX6Vk;#DosILD0cumCnP-nrP(ls=ch@jW_W0?7PX=1PQ!*ueI*U4a`
z*%~(k9+7kAfO{4-b9Iw7=0)p4W-Z}!Gz<*hL^={9&ceAE3pGuD3EpZf1%fH3zCtjO
zqEzu<OETZ0U^<GszzTYkb3?=${T697=q%+Wn0D^k6;08@&`6RJLrub<R>mUTO+y7m
z&d!%0$hQTXVI^!g;f!SY+o>&x8E*Q3M=41a4^FoPzO#*cd1Vq+QbfxU)yy*`kN>>Y
z_6b1FW2f;pBygr*V!`Mm+$<7S(ACiF*@L*fV>@YR0<`H!zBZ)b3czKI@PUep&YwH?
zWv}xRdt7i68;b>Hu?f}P@Mn|>%~J#Q^tb|^jz}(Dh2~nxSrw3JxCzQ88RxOzzwM04
zsaS$B7J8T)fn!6GDXIp74X+o=e_4?;z1a<*sS%>zY%Q(`IW4L;G&MmrBRPR<rgvx#
z;k{j*oskq=!DTfe=9ymmocnqq&!)gbN|QZ;jZ1-bN-dnV8kg=$O%I!T4j!H!h=`CL
zrOJjuvX9_~%a}ZUG1+?eS+&q-&t${HK7nmygjpI4&kejafE96eq{m0eL^znNqQ({n
zv&m(MyW#s~KFyM??mh$6IW{&Go==V~h=5r6m`$YOOl_A&UfLI2?wTqPNZ;|xdmDXT
zwdo&`&tCX)SCf%6uaw{1MUBzX({llUBDqNsr9sAFnhE9l^kY-~M7BT}2ttz8g}{9e
zayX;!#>wK{yadA*V3jePh;Ly5=vz3#anUwswOo82c0NESZ-R#c;m063ete?>fTl?B
z6#s3*F?w?kj2XDJQM?hG4Ds|Kp_Zr(Ec0)sMa;@^;NKGe9DJ8vbhAM|SgTc-dKh@p
zkSHAq#WmFA2ZiP`A`XWK2H!afzH#FI#PLpitGG~iLcuQITx4@*n{(e6#Hj5r9*#AO
zkm^L4a=bDfb_OJ~<fFm2(!=qHL98CQyydY*qAvz~!r?C?5%p%m(D;HXl6?A2?OBGL
zHleBqeaKMRfmt=nzirGjL7fVjq02UB$X5%1y1|(vk=%mw>xjw-gzT#i58s`;Hnv%l
zQG*JlZIxw<p~*jo*bRdvk8&V$;q=WcEG$GFO94h8r6`CJ`tujDCF212r?JY5>)xQu
zD0$8z37asuaLLrikGG7-;X!+G;sSt8H6JEza{YlZjle-qtY+w-ED(MUFlZt9>lynO
z%nGp|a2m1m@WkMpg5hEf2Hp_SBZ5Ah{igIJH0ftC_d))ihZe+$xpTt;&`T784edkC
zfdvjPybN_UlyK0vzk|sU{mT)E(U1V<j|#evpCc#HeZ;lcj<5d+Gb&<=?RvXNCD29`
z=xrs;vrwy`mSM!8E$F2%5)~!cQ8Y0mAqyw=%3A^N)~#EIi9*6sJy#3_sy<a#vNv>0
zf~<x#gE7ktr_dcgodMjnL<NaS(UAZg_X^qj<5d$8M)P-DlN@You~Ff*CON7j2M>kT
zrT^tQH(vlq2*-h>^&29RaLz~?^T!3azi;Jsz#uno-1ru>T-_K<6bzK!IHgId#oKG;
zM!igHMw27!?e_BtFN|+@ZejkeieIoFQJUz)4oO+Ld1_(Jpkdhrw2}d0E+hqb9lEdy
z+|%Z4hdchr0)rA{&AN3}1LE;?Cdf(Gf&I(X-CfFcy$B98QXWN1hkgZ$0{JC|#z!M#
z#*1+CgWuVa-opgZEy|fpyIBVo?L%unSYi_I$`9xr@FOMIx$z582+ASH=VzDh^Izs~
zzhKiIcJ=h|C_;h<Tc1+;brmcbH#w41guSy1y&t)N@FQW~`0~?2Oj4lgp7q;{cqXI;
zig!>&DgYj?G90QXScnk$!_uRrT+QUTIK%h>DYnL>QTnt06kD@MV~7Htf|;D2xXF9B
z-Bv}dzg*KJL|hzyk&4%{cMHzb>ZV<&wt$==i?IG!D7(%oTy-z;)8Vh{7u~=1*|Vwv
z!KZtNZj|6MJ=cWG?F{Bx1&TJiNQ*${29g#c#7oKc_)!!l$9fM>Pko&Dl+2y4T2G>L
zzg7`Kbn_%W2(E3+F7-m+*YDM&ivA0z?UwD^BQ7t(g$=n-tor-6NrSk@pbFgy8JEVs
zqY=ehep(Vswz|4H0ck;uFcpC-34-owp!>%$tk7RewI4KX`l=YvhFe*D%Mp*Ll*ztz
zbJeE3u>)%%%<MlIoS$d-IWLUP+#bcu;SA%H(c}co&moEiwG0n{I`HgE;bJBXR=BMb
z_|U_Th8Z*N$`~0g^-A;$$B?Hd=u~g=w~40_$y#QJQhQ=CDGO>Dv>8R{AHaF`mLQAo
zD-P-jrH8K&4~1s0^AWuxrc0bnUPOOE>D@Tg45K=_o^Whj`=x~`ulemX2f;W3D>)Ka
zQSoic`2dD!t*1at5opkaQt$R?`vcZHtcz*{3<gUFjY*E?4D65%sKcO{@ofp*fI@<>
z3YmuKSdwekua_JnLJ<nIE{HKCu)>GZOA<h_zB#bwnPL+mEJQx+>`Dz4sNf*X6hFyv
z5+NE0NB)M@I{tpQ;eJ6%^)rNHkcJ$$gXQ4&X8@GZ^4tZ8!V0s#D{zF@prS}_35=_o
zPosoE3J`R-UkeZ-pDZweOXrT?@0O3HE99^P$>R!lxszr=3G|;|isE8tuUwEA>z+So
z#t})^7YE9PC=<X=k=U1dhDqXZ090D@O2r!`L4S!9B>zA-5yb(84^GZMEozP}jyxA6
zuOQ0h3N)>@rG>^vx@%aY0XOO0$V$u4t$4vFf(FI1ixB3>DxqoI^#^);#!bvV_*|Ji
zjeN;yFrNSpq7Xa?r}}EmuT2>sHc%Ci$xp-v1~`da!SHPXJ=%ed8l7N<HXF6-13)E}
zK*oRSRXl?^0K7GHri$dV0(H<vS&J@nHI{Pj!lar89Wj;TW#bSyfgw<bD<V)b|A}XW
zue<l*6(obvkw_R$L}F^#<B{pD89maD79*4I8-)O5KlQZF1wF<q3&Tr|b0xR)uebGb
z@YlV0`?gBTaK*zsf-B%nmZ40eK`c4gH)^C#p@+K(Um5(E_krKx{+7sdJ0e7jZ>x_6
zAJ_Y9RI7lwD}ZRL;A{c5mHFa99S*u!#F4>THb33r2`vD=w6|_^qt$BADhYp@M?x*(
zdW2H!CBC_TU?9oQzy^i>j&zmZ3ZW@^hK^e%*ZwvRFL!9QpcW$@b2M;r8Snb-Mm6j9
zEf%n7<*FmS@z`nD`C|cmC7_Zzid+`F$|#B=IDadQ+Ra%YjTS`_Io)aCLv~^+70H;S
zD1x?Z9;*VjSc1+bqX3Z^uMyK9LVN1gYjZ0rMUr$1#cQT*PEO9wmt@!p4F5HFvvSv+
zjGDrH<cKrW8C44!m{Pm|E33nLc&H@+@<`FS5mM?04<E(@T8hQHi^Y47LsfWg4zer-
zsuGOtq=Baqdxy~#lK;rJ85x&R5yS%$V}>)5BtGqARad4m!lTaoj#h4EEFLUl{DrJU
z$l5aBVoPh_2QXny4R~ATom;nt;qAp>7@w7-57Ku|Nd0^8=m4WxCpKD%5+KI`g-Q+D
z97Xt=eI7h`KwvRQPmHjtUHV6wGLn~(KizKPhPgoqLuJMop%Pp=Xg?(#F#oR~%gXj9
zE*divlC5Gjiqp`{T^K;storHE7C3&o6O%JiQY_*-Qd3*o5TLi6uKMYebJ$b|G*dOs
zRLd36XOUk_Iv7B|N@xoEv2PuDsAcVUkbMmQUObL%pMU_Bx1j+N3rhm_IQd#sM^L@*
zc!~4b-#`s6UraA}0U4!_IV`jA7nUH#0&nILE+kgjd1>K=NCW}Af?d%Uzl&KlH8v3O
zuz<oOo5W+bi&JtB5mt#7X3H^nQ&IOS0tIHa;re0p>R&Cu@Q-*LFh;nan6VWqg|JMF
zfVag1dME)vr2=^#Zj%IHT=6*Na=rr`OaS|)h_EL;GqYsCWIkD0S>*oZw;uG@ypF&U
z0zZ;o3oWfY-l5c+p@*r51h+d&Fr{HgS{KH5bLfA41*}B@8FC6ZMhQgd-5dkq+&US;
z6?1pvQ7NH?XGRVKc`th9fr?nyiYA!;&i3`&ikET{3)pjHJvV&+eiU(dWYteJndQad
z@iCtsYR=;McU^>T0``K<Fq<26;QmL{<B6a(7w~`11(LWl2VgLS?pWb2mt6GZ-#|t&
z8y>TkNJLDMD;ZTn&Pex~t4|?j#EqPfY3waF^7l_XmUG34_btOrt9i$!CB6R!pk|8z
z1k}e<#05r=IZG>0jRhJcSsf5s1=u74H+|-N*`<O1CK{7EFZ8vK-~F9`y3N1;u&?>Q
z@E8BXCE4x3b7oByhgJ5Etag#7b$XjS7kAR%9am592xg$r3p&mx#s9SqBRKp{=Uni#
za&gYF0?Co<JFndK{_8jU=k!nK=I&GSEBg=4gHQAS!kzrVwEq0#&)*J>MLU~mH6*dm
z#zy_E-l5~ipTJ6fRo~dnY^ZT0>-K-Y_`!R+x0ani2Z^pmKF`r%tf%zC^Rj@OrM_+)
z`}PrkENY}{mfk|*l77dpLLJsw7U2CdOynPuLm^Ss7d_BGA*;V|D9H4jnCE}r>S!C{
z`Y4mu*54Z%$8hsx6iM$E-Y1}@7C$>3IlCK|_i_A<5?|)F=f|k|EIMBT?IL|hY;5e|
z4kQDeeg9VPrnUJ&eUH63J06k#$nMW=8+lU)9>%_JxeoD7j6y=F=X4BYws>sXv}1NK
zMVI`Uygj5U#}z+A1h->o`LbzLU@wAS9EIyqik9cvQm%uGwnMyv_^3Z}S2e;Nh)V|-
z9HJCYTvADqOBLISCa4_u(Z;M3?Ga==V9BjpSM-fbfmwPzxmgiBmpEnS_ceH<P`aEe
zYC1esVy+xdPcv<vP5gXZHS#g5<M(eU4!I3~eE&HaMeY6M)ORpDGf+rj_n$|w_))E}
zWr$TfC+t|($EmNpjkd(=n+*7XvUj5$>>OU3Tn()-hus;4n#_BNof<Q<8~2R21B#)+
z5qw7EksPqS4JHwPA#eI{R1Mz}jW7Gi#kmUaq$x&n^ag$mh(B&MP}7FKdk2fs6V!NA
zuX#?r80{!}W7H)vNzdzyTahI0o;0g_0=z*M_m=HADqjEHs5UO>2n**1vEPGewV#em
zcH2P9`|YK*i)S@<d~f65+=J2FTE^$X@VoWbBnIhT#Aj-;&BmRQYudC6p(*GdFR=3P
zB+dWcj4DYNCp!tL-9#tQ-%c-NFNT)6w=q%8YvQ)8ZNFe1Mw39JLBcgiHb-YgXlS+m
z*Vnfy4^9p@h)m}?fV#^~3vSP!8f@S!@}S8_QAZqt?CEHM3vg_7rsIx#W7r}3_Wirv
zKOPpk?b@1aW9HRFTazq(XW2@xACu$7+tC2Hml}V5VmrNsvgAy`Nxpr^TU~?ZK>jvo
zC=HG{y3>y0=ZiksD~_WBO<=VfpY8Zjs>Ma()8ETy0YqU+cR&)90<?q>Q)drE=UoPp
zP*_z}^=m;hhiPzdNdCzjz=T_i^S1*V8UJp{7z3;aLZ|@`zJbDJhYp<WeLtes1B>qh
zn-?ckgvCXgAouxsWq_lI&A-*KJxZlI8vTbaW>{1WDSpxgvPocK!pfN~+H2TD`5oKa
zRrvG6Y5YtPgym6@T#SXeiHP7lVJ$9$)41cR%$W)<?1%B)s4n&l3PI%zN5Oq&Po`~y
zxv7+4ctU$#Wtj~oV0L`i9Vs8Eh$59o+cQn*1gyes(~J=8MvF&<YToLz)y<pY6YO&n
z17z~63V@m#N0W64aq`h8P~&5%h;~eU_m+qZMZNS|XU-a9Jju`&ARyA*=^#eMp92X&
z5{L*cqTZm37tJAr`2<)*67`Mu;6me<3Qy7E!KOSdVf?m$Q+lpR7w7g7j!RSfm&ntr
z8iUJLZ^Vp@he(Qhu9;)Y0ArR9v!bhF!sTj=`BZOxCle*<d(bJ@9<=X0Hsjh)4aF|N
zmHN`bX#`T7oYw6+`f`sFUV{UG(7;|^bf6Jx^+}mYCGVSm#KNW!{$Tw?J6uO=G_w8U
zg@ih9+I(g*R&w3Jda<5WV;aO%tL<0B>wWrK_Z+W3<#shA^X2$+-1_?J8tUMO9K(YW
za0BDOBO#NS)_r#q+|~<fO5SbBvhn<Q=MHf?$GlRxX>~KqD}x7tEg-I_Nnp(e9E<4t
zHlFSRgOQS^2RzxOn{vG}n-5!J<^J9#vp4v+_*Q0uY=5qe144?V(LkR#-rpPqSSp*#
zv81zj9x;RnajjwKlFJ8vo!peB^Yg<U+d0FNw6rrbR5$wC&uh#8vY<?J9~EO0LK%i3
zYAz@=;Jb)DmH~O|P^J~|XYb+?=Gvic^r1aBZxZVpz0<4<uH~rR2~wAMv!f!!B*6KK
z0q_J7TqC(i1!nlEfQ>2SSGT6)kN8;A&nfIf#AU+#?p;Md1LOSW@A<7AId(Z&Vr*o)
z7t~6d=EvMLJ!M6tg7|%fFTL20#g{R8&^}ns73bqsQ`4`|x^Li*o?LP@%e3hM*;ems
z+^Q3o)5<CLaGa72k{VQh(Z`Pm8d&pGp)K3r>3(jnb3gzMtcX|fl_Al0L4_;>6qJI+
zcy7UVf&dUYF_*x6JuL-3&wJ7*S}O2_;g~LeZe{>bbPs7?=G`<jZS-5n2w3Dg;lqmp
zt|yje+-obZ;F1BY`qJE&dBfh5Yk=y04HlGyIRW=-*OhbUT$>}xwHWw#UTIa}jM$CQ
z;C$xYouFtoz=PEYPh{@GTyd!IQmMsBu9a6NU#So|CCOff%#X6~)UDHs^z_Jzy9j`5
zG+y}`PPI2T^H;4}y~-RTWYL04z=o@mX3K@o@bkms!jnD1p!~*U=G`}6b6y0K-rg_q
z<?c%2=j68<z%$O>p+*1iT4@Z+1BVvvw6vIN1e4jVp>M6$i;D{J>J2Y-2p8ONxah&<
z;EiKBvmq~i+P}2r<BsPnULk$n%<NCEH@|!t1&oaQnW3%p$v{bE5sRx(m>ci{gn9>z
z`p|q~2y#d~KPt_#>9yN&*un*eSNY)Hy?dt9IuKw$%#f+W97VE2A+1wpc=n>@z0PV0
zCZoC8{t~RF$H<KFqH`(uJXFp%wE>2?^8`}8ym1e)I^UbrNWF2O&2EaM#ggi-e*qmw
zZ&S=>v<A8u#B19Wj_Eas1?KVkjf?hZl-J%(FDf|>Q{BStO#zhowEI972@C;Y>7$KB
zZkuO20ptd|^$p+gi)O>l+$K$Ob=BTsXU-76J3LqE^YfelGY-NkiD#Ah9X(3*8xbTj
zy$?U82467Zd=cDm#!ToF7CI+#ozQ0ZYA=&%Dk7$cR?mI43;G(s)S!^<t9f&%`rf?&
zKI2`^!}){g{%n6cP>59pcnt?cUA9ww5y#->Jp>`94xCS9s5_+WDDw*;SEC|+T@V;6
zG#RzEE@SqA2c6)$(|7K!jKkqJ#`aHKZ0I{Q4dzgST8}jRt~+<ny4MUB@^0W~c|w!;
zJ@2f3Gw$Ir(S~^i;^%XI1zFfd*e>GS#U0(<t3F_XELsT^#7G5EV2Fyu$v?k1H<Ep8
zbQwbMh_x)XG6E&i>yHzn7cM{sT^#%D(CyiACT~XR#U*z7Q<R^e|ByQTJW(yZcUK*g
zQZ_GmsOjA2g8~k-Cn-2#*UV0;k%f$8CgAw?ZC!=m*>$!}oZ7LZac}u@Bsf2z<1Bi)
z0J9c7E~Qwt<bAjd<#MJsX1~yG1A{_Fdb&hkcXM_9gf~6XvJzviqhQM#?``;<lW6no
zz=a%(Ic^H-SZNEe!mFZo;NG?^&I9CY!OY=>(J6-md28S@27fpPd+Fq0%SPvh1T8KC
zXEeW7F<F3&-9PWjousm|A1|;0M?q8c*A8gJZ4)+YI9q!;U%$26jV)JfTrzlaIqRuY
z>d6~_nLn8sfO{B%O$1_JoZCb<ULqwWWjHrcPnjRDmX0&>{9B}*ZH3Erjmm3hA&0sC
zjXe@yeTu#L5Ox3=Q0dD9`r-j6Z?>2Es;%Ft{b_oX*pM``YzA_b@vN60ZctnIjcy5D
z+fqDTdv%PI==C4(p24j|*jyB8%Uk)azJnpeT~W*MQ?9h~<DJ2zq?%W<%aPS&J=7vq
zT5nrn<arxgVV~%E_UqF(;7=utGM~-lah5<HeRPhQ)A+trCv?AT0UfuS&u9n0%~Z42
z5Uovn$wV`tD&4J`>qgh|A{bJ5X`T~W*@9YU`|1G^C#;xj-`FD}A_6#Wa{%EAH{E8D
z0CzY1QHOBPN&tXetMcwCx>($-yOEdFnysfWnB{XokRqkEVEg=OnbA20p`{mY$>nBY
zL?4l}^LpD-^Ut)@;`axWS+eFc?<PMN7KYfDLVS5C`sE}V2POa+K(XK=AKOGpxD**|
zhH_(HbC!H@QxL!(**J4O=G%7iaIoC#o)R<irAF6EGL)gAmte}7ZQ1(}YE}`@WiCNM
zYT*U@v(-QLO<0gAQ>fS?Lri<ko*jh7YAJDa=~}Ou+^G_uFf6qnxfYL5&Tt|;c@ef`
z0+mg$iVq7swywP_k{j`ODH?z{i8LDPsz{!q%Y?D*B8w)`bDX%|D^r(EEBl-lw$wJf
zOZ99^lAXZ1dd08iCp3^PQT+@Q6advi^a~>Z3g^(9Ke3${ke@4MquIL~!)A%7?3Ab5
za?P(g@1g3|FpEp3<&KmIJB!S5H++`YU`EE@9d19wtv99%E}<7Hr89FZs_&jUw6z`A
z%>Mo|ERTYtfCM-eC(R345TE!)E=`7lG@S59wUYKiK_26P>9YIMLql+S_V)L$`%tKx
z8R=B}`hd;Ik4nL%g<jc9=S3N5NI3q~DWp<TN-qX@J;|_DOwwR(gmNA0HU+0N_bV-F
zF|jq+d!Iq0VTJDIv3|vB@dE=8pccJ|s|u{hJuD*%$q-(JTp7rX<*gQrr814LUOj^|
z0v)2QNW}Kwjs&fF4h9R4#<3&GN&m#yADq}gGmFsRm_4%sNS3nXa|F^EI~LzTvaU(q
z19UE*%W8Z^w=<IIC5Kz{M{qh(GI<Yp5r1v--f^@3$#R}2G@K8?RMIH)_`lJyT0+d&
z-rO|fYIx8#(Ft8dl?vs;OC*4Z532z(1o85jba5!|kIF}Hs7uZjp5rooVGygEP?tyy
zZmTCpcw*h#t1TdMKsEdjp&@5-mIzPwTGo7Ej!}xK5VB_!yXz<V>!mrOaY2ed7vSfw
ziZ)SWyySm+x+P}>MyLZK&Vr{-VHW*QFtw*{aV;)c0Z7v+DhtSbbYMG=QC1cP0rM0O
zRVI6Bb0=GsR`@V&$DGcMS=J-nRb8d8VDu(8N72ST20tQQTE1O@h@$&z(C3-FrEdJh
zC%f72;@%sT$}qj&lFFoLVn53~;~#5MSxI!gR+{GqF6S(I30yGQHr~2P@bBr>o+m0W
z4maMUZn@Wkf%>GreHGd|RFHP8s!!GsVJfU)lArD<So_d(_s57wn<v_R<*-gWX48y+
z+)liA?*Phqnp3)K0DXew*hGOe0j!PJUapnBC3<IsN6j$zTn8#(@OzK#YS&lDpt$ZC
zupWBs;c;y)-6XE@NNK`8<;JjKKzpn9$1T3XO^!hC0qsJk`bI8e83~4<$C8(zy*k0p
ze)Xp_m2GHiW5poKTBu^9=dN~H0W9xEmj$Q5p6bhP?`Y39HT@d3%XH#AOa=9^Bm%m~
zO-A~Z`pmTLJ|W`QLhe?_T%uZ{UpLQ)S^(lRm2`VPK!J>Cq`pEGb@URL9aL_pA`<63
zZxOA&QAeoLf^X~~x@Y784+#B`(>6ACGoBp$kz?oAl-w<E-yC<jXIh1$=bw>mVvQ*O
zWF*+ZTSGh%Ikl!Wi~1Iyo-D9Tj8Na2rB%+d@8?M9(M15^q_Rh#S@;<a!H`tdlV)Fg
z3f+#(Oy$6Jgcd%NezvXat}DqRG;gFg3Cd<xx;UV>Woc^=E_`^DPl<<nm%xrgCYqvd
zVPfO~RDNd0yvu9BG?~sTiSB_7E(b`#yo1$b<}@?~%pCE{oc-JkbNj$I;0s*lDUq4C
zfq#~cLQfy6FapC=MeeT~q;NxmUrUb-@B+%&UB_SIG!kcJ4i~&;K5zyt!vkC#uOUGM
z(`CV9D*mcFWGje*6Cb|c?7$Q(;V?I99Yhp?4%t4TOl<O9n}xSz-eLzI_dSRGh?X3H
zXRV}bc7wZ7{PmZAB$3lZiG+zs%{w%*EU%i~;zc7QmbxKj@?J?n!F_NMm2dSV8`BN*
zKkz)as{5*Ipnd&Cyv0)22Py#8B6zlzdZk+Bmw$SfJUB<ob^Qy@0Maz0XiH2wg)v~~
z*HbTJdnC7~<a=iq%df#c`Fm4;VZwHudKh(3K6!Vh!*oHi@nClHh>K$)DJ6Tkn2Tkl
zC?-v*2cT1XvX4{LjCUAS!8^O2O{_^tz;{<Vzq)iueYWoinF<F;Yob4Z$Bay5M!wKe
z-f(grnbnDIJFqZ0myJq2%XVT(q{_?xspjMQp^cmMuZpc<hXVn!OpV+2_NP!WQ5j4(
zPL%-f#DT1>d;L`q=j$6`XJIaSjwmJa6mkv}Li?hRXG)PlB}zf<`sp<$ue@;(7~r}7
zVZQd$i(ecK#M<o&x(QIk%Af1m*a-CkvTv<HtESnT3gAUlGO$Hh=CHPRY{Of}If`0d
z{N|=SVAB@pD6bC<-mtL9M~EgG@gp>LZJ$=6PY=hjiGY@a#a~syZ!HtE-f09#=;4Na
zPr$m^O%5~QDP@{C!8o@0n9>zPL-9nmypun#S4%%+prq&)e~K;-LIe=q`tAzy`DNhr
zH194GSGPa41a}hFBuP0SwAM|7qbQ9pDmGW&y*HYu5d@s8GEI$#d<kF_Yf*4vkUQLo
zmcKh^>f-WohL0vGKZ`C8e8zcUQ^+ZYk~<4?s4LeKsvcRQsCVyPhwvDq^W<Zs<#FmG
z7GS!kWT(<`Ef0_f;X7}8%eT3)O}|xQX|(Ajts>ZbJ-i&~v>*uJRJ7W#oA)X98VBxR
zjAU~MCs+x~jCq%Y<J=Gr{?5xeXRdRAWeKn9`XGv-K|{lEocZ_PDMke`tWkWGsluty
z<7)7jNJ}_bAfpoVK`%hsG+u(RvtlgR-CVXOQP@*7lWt_{Pnqs|c%MY1W*<{3sHyR3
zbzBNwJ~yQU+bmQbR#wxqvtx?+53eidg*dTlMcca5p0b^OisFI@8>dR2_gTI6KV|4;
z>fcblhlfHU5OCX67m|1)zo$FslX2>i<mTqzz=#mp5jmw=pQuKSI|WnaDw2W))`oDq
zYaKcm`>T-pl&8BM$~`joFTBuoEAOq>rg>1@v?J#S-Q2sIzZoeF7M)jmUtis@ab0-0
zA2k52L;6YTKi)e80dM^cffONXB!AQ<UiJjN_SU3{`&6k<7+NcTHRT)QQ>`fOH$(DY
z79gl*a}l$Dy32wkoB6F`I|kfPACZn5H{2jnWH8ZocJuCENo)yUD9}FYfojg|$)n?2
zcKUmFx%&U2?5)G9+_tFkje&uRji}g&fS?G7fP^9v(xD(FQi6n1B5i>|7${PL(ygR4
zih@YDfP^634N~7&oOACv_xt_vd%p8L*XL4uzx#dHnrqH6#~9Nl+2y`_GWUG5ia&UX
z7cgdE_td2sx8GzK1(!+Xee&M6B|EuxO1~>Jh(%DLI>ztANM|wFuj;HTW}g1><HoI9
zVegR_DVKd(i`!TnrPjycsTM+7#%<Cs_J4$S)#CxjODB#}tv+@suL&nMk#WtZTM_Cm
zv1T(ic7rhxx@@HXUTs!R+47>ACl-OCzdpnJ(U!ySul+g;L4hZlq#)K8YrPk)d$2=u
zp#Wn*&O>Ea^thu4Y<PB9eSkj3akKJJHdwB$9Y1$z3oVFaIHJ(}^g+9+ei*oH2W2BO
zD+^3by9Cu19^kEHBoi`@+!!BD)Ln!%3Rt&S@4h0AtRfE#)j=wc$&CkvVpQ%Rz==NI
zs$JV(4_^<Knz}x#ZCP2Qz==`Ic-3lHMLcr<Gn&h0m7CVCGZ=IG<&Ku4Hg*N3?py(S
zj05onoK<1BBiiCP5ABp^^8&ihz>VJb&kp(Q+3l8=Hd7Q*>P!V?t>!UnU>%QK=mKH&
zhc15in^Y)<HaiD82Ri<FG!Y*bS&8vWhiw|Tnk-V))6o5dwD17~2Ec*hLCW^A&QeCS
zwol(48->BdsPxpJEvxKx`W%iwq{&HSeQ-!3Mc>KH-ug=)Z)g+UVc$s;%MWOk+UFS4
zqtgM+q+@wgA5P2ZRVzr-2^`V4mz+RO`jL3+7@Yickgt*vinzCE#eECP6$bd>zsFl9
z(*QxYE<-k1Q|o>W2Fu7V8wMp>xR*}Qc5Fd98qR<^dU`LfI>18jt=zBA#eqDYFvBBW
z<ET#}!3Alu8y8nBh#h|?-*7X&r(gaAtJ1u;al^-F5CD|NFA=JmZ(01^@EMXAIF=Kl
z+*gZW+vdO{x|*zhPw?i<Pn;NC`q~|_Ec#O1al2CgsU|P34-UykjBUR?6&3#b720yZ
zepe%NA#5%FtQq<611CpRR~%iN$h##WlKEox^S2_-e`nrpIh|-wO1(bCQE^;szn~?u
z4=mXSrS7|OhQd~Bzu#3Ck2)zJHSYU@*8cwv60~t8%86?pg6PTNY#;AwoO0k1dB}CM
z<c-F5x7+Uj+x}X0zVqMnZqJ51{?Gba%0{Q}XC>~>{ilO<_TiN5{|s{*S@XY!r~e;m
z)Bo>}uTXzPKjJUR8u?s*0cp5l4;Nprlcfm4N0C`EthK%U9>#kjttE`hC?#fS7(XLI
zai&Z3umGpVF=YXDk%kX6fQYwWs;b5WQl$QQu{k<JF9nw14coz82Wce@fDC#<C<Q@M
zI0**P(i~g#u#oN&B@6%reWk?88f1+iMgBr?$NtGgtMp%2cj|>3UpZ8#m>+l&1r%W<
z<JZX$COYCfi4FLbOwmqCB7=J9Ych$2(^<skb<IO)6b%J&|LeAfNV}}3f(s3C@zut-
z8<fLB@}so5;GyDt!}K>ol}>Y*O)*7vH1iE*4HkYliUc^80uJ=R*$CDpJbD?Uj=^0H
z1PWMe-vvmETNDmX7kmtRze0Pp#{ImAuI?e2Zs$S34jDyn3fcJ}`}`Ul^Z?HY$^vX!
zh$*8kd4e1RScw7>BAF|Ui1xyN`3<zwK?ngZNc7vN1V3YZ_^MT_5cBFmKzuj$2#gSv
zMBt2}8oYp$6y0ngKx1%8b3sgDR*n}nf)Lhj3W(Cz=4MT^Q=Aa_(}Bo>N$ml+Q=vT*
z@{*f2Z$68Z1C*CfpuxlJq8xMP(L4;NYXr{}l{D7<RS*x!SUSMNFhwVBtgsTPec}X%
zR$@0G?o7@0?LS<ACBsLM?nzP#ApH;@A79D9@es{_9w)8KcM2kwvh(*h0vX^L*j>-e
z$;nw(jZbf>2f*$kRz?(npeUShM!>3j!vh@^aH1$UuU);mJX+^(-pDrKebR4EtFmdN
zE5Y4Fg_n<y1s6!j9Z%*Vi`q(PD9T_h)jZsEuo{d(F9pc}Atz13UdDelxQE*|Z+?em
z@(%7AaTIQJdo%=qdn<z$Tm91Y!GlP^@lj~<9T8SWW0fug?5ULzXpKgqc^`#48I3^&
z$<Xk1bazLP$)os|)d&$jfeDMW{T%Xu3v8O1DpoK&ChrRSqm>b8`Q}VC>@?RvEggnO
zWmFz?N0R|r9g#~JVeJ_@IzSm0Pf-Q|mm9HviHEasV-TPZC7d8N^RrV{AHZ{xcLqlA
zY!uT%zz4kq#f1?DYe@!IM3f26O2L)lNT`>jxVRk5_{cXw$c{##8Abkwh<gOeD*_;g
zS?{<U|2LM07QO2Mk<CQabJ5udTQ{l=a9T8_1Je^&SrtMaE>nVXXW}Y!9!lsvdYuoJ
z$1=Dm$!zt15$oIFobZ-hE>ch{y@nNv<g!tRPP7p{79pFQ6%m>E4mZMZP(KKPWBfbP
zh$xB`>NO2jg*CZ^!gEM+96WTygKa^m;qW?#w8X_9YW9N(y)}rEtW&Vz<z|uiCkm1i
z-4!BKM!4CM%%OPUZ9y288iQ$)eTfdU)+-}0u<wvjJ9_o_0T_#{eT2EA1mTtvCa`H{
zVywuLyzC`b2MFm@QDC?{>U{$WOq6`EOFD+oefqzauY}7ws%%6x3bu`;awg^+Q}DT@
zlVX93*_MGwH#{*R%KQ}(it+=~wJItqU^wE86Owo)HRWN16hPY>=zlbA)Nwz*To^4g
z2z+Xyb5IB3GR1EFM6!jyK98kIe5A;Wn@`^~s)>-l2hBv^OK_b|Rm?yLw*@R>1iw(a
z#xx6b>ogRtj*7(7<brxdWQlO&=>otUf<`WJZa~~-y6sk-DTo0Bl-B6@?8@tTNKeVx
zjQ=2XVT@{A@(^y{8vhcpoyfv|0~xaYfE8zeDh|sluhn{o!oSzcA(8D=6{S<~E_fPc
zWAf9laE~*V-%)v=lXt8^7;V*iPJIQ2@r8$@l16X}q<5jMC4Ryb=_yfRp{sX@x2_Qj
z4t*@?$y3a?IcHWa%APdb@&rK_b1xqw9(AV}g1ZI$U}bHb#%!%lXh%9))MVf@B&#Fi
z<G-=(j-r=PKgRugA`%mv5==fqF`Y1xIaqT2B{<Gx@Cwr5LO-5k$oj*8qbfyTl=fDs
z1vAp6`tQ+6*aIW`3!kpI0-E%lkQaOs73{rFl{{HkS)<SCkqI299sX<;_71^b+lvMT
zE*5Z3ogJiiuaD@Xhe;;<Lczv)v+W$_ct@x)d;~rK#r=NnaIct<?IO1^dvD#-@idXj
z0U625@kIdPWw3sJq?+x;-VaC9vvBY_mMl<K(hAZQXwdYDzfh9AZ1wu}U+e0;Q4YQb
zWfB6(4Z`?4VwFjJ`V{#U)c*H*Y<AYv)I`e<y!{PWL}3?JGa8`#mRqEJr)3--ROhl<
zlOF9h8I|C7t3%g-ET6O{h;*(%2L{Ay%bq<RPo6w!o~)j{HkX=I*N0<G!W1UKlF9;x
zS?p-85_f(a1}G-2$wNSv>7yI7KVBB`9!aaX6~DrT;EHk#1q9pgEa)15sq}G_q4(yB
z0tC)Cz^45^4Tzg${c|eseZJ0Tc{yfx;5W!PPJju4SB-IeVy))l_yjlLaDyYrQ0Me;
za;zjZErlFN&S3EzzU@}m_3^?Z>uwVq*qa#_+0{P=^+is{fnfLmM8ccEEkUV%n<~u=
ziOjjEUWp<rD5*B1)Fi`%$&_FiKUU!I*DVt`s1;a!9O3|chbWfN03#U-h6^p?b46YH
zZ&K1G@*H`fGhzj**a8G8dTyhw5R=GmO>|mx#HlQbPd?UJl2=;FE?<mYEic*gMod)n
z1IipiIuM{nqWuI44hUVy3<TR*P?cL?<3=YWl#`kQe5n|V{04v_&||EKQF<^y^z=zX
z*ObpiqQ{z#&ZgpM*0LX2Nzxd+k!>U30F2C(jOyhWPbX#_z@+iFYp{G^V}d;bph5&u
z_IW>*k8d5xR4BwMIQ>`2WzzgdtpPwTyw|^(3&<UkevF&8d1umKZK!onq(NK!e;-zn
z3LKvkvBJm%0i;+yycUO{RywSfVvJ_wFgSEKp}GLE?Htf@1Q}T+*Ym-@a0U@u6EvzY
zTD1T|0iw%B(Rc`dMgjE6i2?=~jTqW3WN<c)s!qf^q8S7n1arNS+LM3{aH7)?nk_10
zymA*_6b3vbX@$QD`aA1(opi%;8O5-SZrik#D_6?LYYG4$5kpS6Ffjq*1qHwladIa8
z1q3TM%#tEo3uJDE<kb(~*6Cp<^ux9Y?To&QjYu!JNC3nhCwV7$cRg*2vwsLf3~#*!
z*%MB=9?WP!K1bFPILhD##$G6;2UYnpW-`gc#22m%(bsQbJtNc|1|M-rd+@)Zbd{*y
zd*o!1_|z=ulD_0LkRjqS1cN9^Z}wOWyN)B{hJQPvY?XJ0IsTnLNOF`Au3kn)_ALI$
z82T91yLhMln68=JrZX#e)|wV?V?Rn4_w(tg<BZZ$XDq_g#C#ufmM9OhZ73F~+}*LK
zrT8Ze-NU<UM`z~uPc3{@uN5sxm>L^tm3J)dvM=p&vl~ns8lxvtH6u){VPR&LFblxb
zI^yKyM8c{b((j)?e=1>u3E$1O`!Zzc(aH!Adb3WTjje<w`w|;H;r)9#1SBIMj4Zso
zs$jjIe~Fo;K7M}DNOM_3y~xaAq|5S<(Gp@R$>y4}9n`4E-D@xh7_u!0?!^-+k*bq0
zvgi$g06=qQdin^CzL#KUfK<XtRC|#~Sy_dVZL=1&2R5~v&rRPX7brO0x;#r~77$hk
z1_(jvQiHhYr2w1W-@kRaA*NS`6jBd^Q{cF!3|Xl%Cbr;;Y7719XDW^&84Nf%06t3;
z{$CX4Z9oPuS>MnQ1-3vW%HL@CMRk4L<Kp$NLqt|PL@8>)ckv47O`X{I7-%JVsfY_3
zMH0^+xKS7==|dqK%pL!(R1pOid=^qTC|~<j(SgUk(8lq|*Mp&`SHR`Lm|0J}T~=_z
zKLpE%PArB|6UXEM<Uz(M16TyALRN5-e`el(fdxbmO^Djy(Ct`P^LO`;gs#Qon<;bN
zM=*OD1a@7ts)#xRvE~ZgBY<ZR<Q9b4q4dZ9MCP`=-CCT~LH|t=?{EBk<HY}bJ;br*
z-;Nq1d+e_3Py{`m*r|!q{5R=eJ))VInCzgVCqX0p&o}os>ummaP247MnCk!a$3yDq
zs)3NXj{fwt#wA8EwCFsu0vNEV?PJZOE=JzP7ZMT@W)b?B2izd^-#5?PwV7gn4=1Ec
zR)B_!6Z_1VV%BADGE*LU-)1YT2OHyw7#t&wdo?bJkh?cT<|WU>T%h~z&FX5AZ4bA}
zigaaGF!^-kcriK<CcwL)CL2Z7+|cL;p6?9|1P!o%K(p=pcq-n7p-E^_-LEBHCO~RT
z#_!~S^ls{rV_RVT+hL{D(8#Z*hu!t6EZy?0dVNAenXq4NEbRKV(r2LDCK~xSjFL}V
z>gfX?Z-aL#9c*ePjt!x+iGc_wxvy96%Sq(KXK23^z5!>ZP3%0~qA6Bq*@@2Q8pzrY
z9^XeX!?kr5iacuj+EX35#h}bzg2%~vsAjk`dv7z>i(BavQ(j2>-e%eMgfKEOZE=`l
zhl1@_M&S#xEcvbbF<Wbbp+;f#2w#|t&t32=ab!ByDHF-gmzI7cG>3yU_~B==da&{V
zXs<V?N*NHR&!Lc@N9#SSa&C_}F>{)ZFx5bh@GDcGJrQyfV|s#zmkl=cH>J{|FGWET
ze|zWNKvQ1WPwVFWnfQI{W<?l$Ux~?yCzuMi0uax;u9SVv(HPS&!E!#AUbMNbog6~!
zk#WIpD8VZub;MACT!RJh;GR9|Io9q_>LmhrB{Gq8>%OoOA)E4j5T1De>X7z5`L(pv
z1HG{kRM4<VHA*=zN)k?rmjbbnLQI1%=~-+uB+4j1T;LRD-Z832pjiD452r6dw!X3z
ze;TGfGiqB@Qc|3tT>5z>O*a_iNF9lzd)M3yuU+T5zS=lOqrQtN^K+a;rQXT&TXQ}P
zx1V|qC3_|M;2@K8HR)lw-nZ<_2Lp$m$NFg1fddPj4Otq6g@uH-LPPW9md3XI&AzC;
zNR&enI1(34^sEnWUFm+h{}3bN2Y5e7dNT)0FD5zI!9Ty~bwx0|N7P=GeF2WqK`g5@
zQrtj*Fsx#d_5DKT451iJI<_nCb6yqOK@N^hs8kW!SIfR+LN6Safa{#R!2)=NF2sp@
zdh_t>M_<H@N&0(|TS&;E&ws{0;cMiBa$&}*0|}=4_uJ1^h!Jq773O>x6M!&CoJtW$
z6R%I1ngp%50%oVZfaMV~h@pP8BR6{gL(KAugrdvuq*;Dt<!0yMY==nY#KXklMzrV^
zENSK@Hn>dJJ(B2=s87+z*v<2W%r96@y^f@ZO?H#r>S6#@b<ro4+9C9Y@q)Ke)%vtc
zIgvxIsb|tDizS3O0B>swve8=AFt4k~BC03G@@l?5u$Pfhijli>X8Ed7a3=yXai5Q>
zzjsb|bll3U{u)0e`#jig%F<76@2MNkwXv7wKgz-NtWZQkRyKXbDhqU)vN8cnVNQ=f
z%S+bwi$RnbfB)e~m1ihMp208VMP2w+(Z_VG`KbL3e*{sL&W!eO6+uaT8uu&&^9oLN
z6}ifOee4G1BW$U%26p<rzR~+zLeQ|;|E{U&N4lfp_x?=4kM~N7=Ma#jYfVwx7-84r
z$3(kla4^ZZ;;xp{yfRY@O6(7qumea2$;v&n7vw{3Xpi#qE5;frV<(1On#7KsMgUt&
z$8sM5OetP70EwZFuCCX>A1df>?)TP2Ju54Gbl?F^72~PbYOs7(k+Ck-05FUfCLX;#
zuSm%2ALzIAxVOLVoGmDRxj}c9lZF94^9^u7FP8XXUiv+(PNT7=^xIqP`1#2^C@g+9
zwWQpvD`yZNM}R{oAXg1A=P~`^jL2w_v-k$d9H!-75M*M)<$dYezaK=v`zW^%A@93p
z@S9?YR!hVt1DU|u=yv3q)9=J+QObp8?Lp0O9x4|ZL6JCDcuhQStRNO3gsOm7UWm+I
zK@QrH1#H>1XtMAL4O09A0tk}}DDO|0cvPJ*{rT}z*f()vSul2QbC9gcjykiS-)d=p
z4jq>5gg2*@58pO0ZA@&D)I$nS-T!j>7(a9X;AV#!T#Jw?3MW>!m4KKD!mwp-c?4E6
zdd;Bqo^QNNN+|o<WY#<v+PXChR1MeX5WX6c@OhV)r)mkA4AbVq{oGN@;VZ&9G=8cc
zQ_6F8U%z~zWY%U4<WKaGUYBUu8{0F#b904Tm8Yyt9G4H(n6=IRGI_S$c_{bEmvgN(
z>yIWMvzu62*BIG?#>Z8C)6!>o<c@4Weyv^BoYg=x-}hF8z=qZVeB*nMQvjE-FN~~$
z`0@~H?R%~po(Y96&O&|HU6I9)_RReUuI=c0>~)&^G`H9B3d)RgrVU?K?X{<58rA<E
z8Y=5i48PWx!iJ{Dkx;CTRa=;(T7T9j9+kFAgki=(G@g+qp)v3k{VV3c%B>h9-9F7@
zbPg10Sg8n@uB0(X!#O+P)H4UWsm*DIFJ!7`!IkF%M~DQQ_wonH^?_~^emhZrH#&{z
zoO24q?W&CrNBfGfUeObHURLh|l<qjJjeWUHrK(e)GM>KHw?1BQ4zpnnLU@URMHnqo
zj54if+s?<Rx&W3g+lOXK{=x$BdzZbOlnZxUyNU`a|KoQsdq#iu=P?Ya(tP`tQ1&6k
ze9B~|xnW_H23`VZnM1vWGP>B|kTO3z`N6tmCwPP*iuu@m_2`G8unI-}N3FG`hjeqT
z=I5Z0Wg&QMHn$PfvCC`Y)e5j>$S`751CCZPB_&kTV{M|Sd<y6U>_(Lr7k++p48Rkt
zoJcH_ldOQ{2t`V<v;<gAaR7!Wp@KoaQ#`X^hXEbu6%;lDatsFv*nZ?I9cjbuzqe|v
zkSXv+XTNH(V#+OyiUK3xAxQ8VD0cASHZw4kcs`-4N|D}`@4i<St<Ck>dw`XGIL?Fg
zRFKTh1@#ux=Rsqw<s=iFo+2LG1P(%yeTtv|dKDZ)KcTM*@07BHT^+C~k(aFa)}rON
zeb=rhp`oFZ^@msy0Q3Q`+gH9@iS_g0<UGnt<ibWy{6|7z#j^+Hl8Tzzu7v0|p?w@2
z&xYS02X6As+1N6+1SW*-`JS@3RKW-!j#x0tn04!KZx5blcvXu~w^F#`DrU!}nzm4j
zr$#4)`6JoWRh1oCkl00$5*H^PmOyBqvtF8PNPWq6$)g7L-WfQ93@GSm7*?&zx|H(>
z=Yr@mXQxEKv{-(doKch-RpHG_aD%!>>kT04Q1!!HhI611Lh#!TucdZBvuvs>_!^x6
zV&xD(GmkT$fIozt0Pu%Y*fuIVnnN#;(YXOMNqSGV6M-$&{45#YK>-qZuKD!sX-xR0
z7t+3a(#*R|FfU$)0aHMWVf0I+Iuu~v1ZfK08eJDtVfb}>Z8$BN9S<0suVSd#!-L3+
zfIq=+Iz(D4)tXH$_+8Q&77A>RQ1s?Q)tLOr*Ug{T?iCj0aFZ7^`Ywi54)*rs%&^ce
z>zqxqph}vULkE%YL)a9TIS?-vF5_}%tLt?}%biJ+5MUXZ+<bk~*rd7i(<h}1$}Po*
zn3y<W`i%;zK2wk8t3vi~(jN>l%h5w^$Tzc|_ZH$A)9Isb7(dQoTf8G+jCkU1-m+x|
ze9`snJO|;_Ni-qkkQCI`ZWu=jSOG_6r0ZqG4`Xbss<s2w{@ih=?8(;`(YvK(xfEuf
zX6{BWaCsM{`nM3`twHG<M&KP3FLp(j62Frl(78P<<u4A5fr5L|0U#nqBhdwsfgCW)
z%HiLR3#d=L6+daa5(6SS^Wd<gKYar`%@xefGj;b|T~}o@Yd8h?`(Fd!+ND5ZDbEl8
z)nYHwc+zC`y43`{jMwC9&L{wMM)ot>fM35H8VkJat4a6Z^ywXZx~oY|k2DD`u!7?g
zdd1xpwz1KbJGpM`1E5c?o=KxhdJtd&5qU?%swE2<7kRN2fiFEGJE3!NF4Y0h5Ao(i
zjF1Tb8n7Qg#b-qA9lTdp;DnNzn{1RC6N1stJ^0M^>g8k!ELU8)XTaEm3dE;uO&uEP
z`M5A6jw1{FTvWLeeYGR1-s}ypg7H+zxHphAGcWe#!jPA8&58fsuc=<1k@?{iamp|o
zv^>1HS2Yrfk~dMQSPw2)*hXAVX35H)Q^>GamgM;c_tYD};{vSu9LG1|Q=T(vids+c
z_3^Vf#^*~!`ztbxYNJF;E+;1~HiK}2_!0UcC#ShM?>4k);1i<uSmM(EW!^#zflGqv
zt7wBz1mlC5sb`ss?vmp;apEkqsCVm?fS@Wsw9|mXA~mzpHLL2-9jt77A$#O`r@m(N
zMYe#}y}~F|?_6pUPdtHqh#>v*Q+NR4GBU)eo}!**t8B(l;(j0*%A;jBIBQPc3SE)M
zC4HA0pbVUeO%k}DG^8aOph(^TJN>sOr^74Q4&5h-n!kUHH0StFZWz2Ey&(Doc;UsB
zK$gk^w=hv*VPTK&(q%ZE$l$BQI1Q^xP@;Q&L~g-jp#h8sH!|6bVqk69gnhrENH55+
zG?WQAU5Ls9^MQk`iM3AhCn1WMm>48lx6wRftMcb8x%a8bi~cpV8l^z!N5;nTfH4el
z!nAHVbX_IQ0(RFH3Yi+l`!#8oVzF5?yf)NQOc_5~7(mz?BYn%#QWwrPd}Va><TZzT
z>1&IN-8LZsfPxR~`PZoG(HzmK8*FMYs82$-<ulHWX>dYrvdIY>_;iYmg8hX2Rr|;A
zuivEselM70i09(6<;!04UVl~UsXju0<1L3BX4%jg3z4V_<7GLR8XRoBdLQZp>%oT5
zc?R!Vywf$g9Km5Dn;4G(z6zex>RmhIF30X9xyMA5G9Heo!F&xSf5%@=d%U=(B+CSS
zJiueWfUiN3aJA#{wxi(m9(1K8H3BT?*0I~+H>3jiw&7jw@>!OL?Xrz`+|oj8_GAU6
z<u0+UWS6GM1%r=AJaB?GFj&6J_7%k4=Pr8xvzjUf$96d@m)B+P>k3#pMm2HLv6$4X
zm=YtiZ%|z_jVyf{$uKlzxlP`8f%PIaGswSTuV75p55)_T+}VETKX5s8am$a6Q|nt<
zTh|({jee%7S>#1)dQ~#Dr<ja8F@fz&tXctV4I)!z6cnB}^T9R9R%ggC{psPb?@j=o
zg;NdFT?{0hkFfflsH9$ORc<W=%8hL0TGcv%TU2-M#K71UOCFJz$Lwb-fGpw=-*BwP
zUwum?N?xp5hSQ+UO5f)9_C0*X8VZttamejLX2bIGngSG=WQjN;fO6#y7i|Isls%F(
z<M8VZQM(@Z!pkCT<}@&|C!2lNZLnvQmOGVEcaK0h3ynm72nrjhz*l}wkC2h$V3xjO
zS0z<04%RS0kv^~9(r^EUrf>v`!z^1t!-~5AqAFl^4-c>s=*GPU|GdbaC!$E=Qjw(>
zw{q=!bY8Hyt#+;=_GKWXRbWsI_4?i7z?hTD_}X#|q3$BCWsjJ%s?Fblo{vuY&6)4p
zh&y`?yT<G7o9=*Mtf>iqP%^a*Sj;KA!QDc|eg(kOUJE*;Po@eZrubl}C4txBoQ%bM
z)%faX1a;NXDp}+d1{r>tHaYeciA*SW!i<uQ_&6Srk(H@NUr{{0hU#4_%*goVw$QD3
zb$t*?G0gXaZjS<R`Q)PY|0D-ycLD+!>J(R+QznNR#ZUtwOjXh>d|J%;^m=T5%J$QG
z_v$;3AjBempP6YE_3#S;<g!Y0<?-y^7OlqQNeRg`XcgaG9B81gRjs<)i!7MS@aGPs
z-GDe|g3R?m7$zf#F=A$YUtM@cHcq9%1&p7%QeJ#Aes*}o@X9`-TtvEle`??f0B*1{
zPCL%M-cEn69?>UJIi+(hr0!5Ni(kp_!2E*c^m|_<nK5H!VM*l!(#NZk9tJ)c5m`wx
zH-z^ZnOsY=oqc9*%4r+@PQp`m>>_*cobOm%&#QOOD$*nzI7=UfdBDy9_s_s(rgBin
zI(`)a&E{^w$MZf6&d}+u&ob~VRmmku?fP9@SB8VmR7a_98LABBJX#PsuOjdtF2G^E
zquCc@)%x5uY6Oowge8OdmBj_iI@}W$Kx<_BSNz1@Oxu%Yufn<>oJWhF96o?1f1@Mf
z<1u@**-VnTGi+|b*jGZtD~8ZdXBUA9eaVqONVu@S4vK-JSyk8vr%q1CE5GyI$q|bh
z35)3hN({2N!C@wi9GHrro4U*yyu}ONY{6&d|D2$na#F%uR!Wzl$4VxU0^+(d^hTh5
z14R(GI4;luRpnNcHg5F%7mr=k`IhxZ+CeCj4G^kdb%g2*I2&Q<<tIfduA!l=Q;jo}
z=;-N@K-HI^td9-`DFG#!p{6TBo4|*7{^U$ALow`o@_HD9&?jasLBT)W!AFby`SjKT
z@PT6HLj0PK?g(0H;l;wQnvHYJ@<4eZ5*WSjuk^5s`OdsUIh<oX<D08>_Uv8F_hU#Z
z>RAsm4tt+0e<{!F@+JeZup-^+I!Oh=0F6}6IJqS3ER_lG`S7+QOz(w}l*jnByP=T=
zZ<jxm<-BLvFvn{c7O<(&tNtreJBuVwa>i_=pBebUL2Q48g`G7weg?P$`ov$hBaPlY
zw3$HnsIGXGo`8l5cwbUtL|C$I4U3lo?}moo(mfp`hdM$dr_K1_qK4jeDB_G+zo-qc
zzMkqdj6!g4?=oowNXS?<2jq<uON5{}hbCLYBO$nx7ozD*T<}Tik2Q>goH(nE&dpdz
z>X&nC&fTK7)q3LL5oY+}@EoB3;y0XP{JIw}yZnr9sh-Mfe4b^f@MRatK@mJlz~-NB
zSzHKDEQVpJP0jpC0UzltCc3)!;b$~HSmnyioLC>D0s4Ntm$|2IqI$_b-x|fk3gt*B
zI+3E$*Y0PNUYvqxyPJ(j%YZRedqhtiR+s3}R72|^Yx`ALk=YH5Lfb%@?yF?X(hHXf
zEY+lE=JB}D<FyH;R)E(?9~ZVHg^F2ctLcxA&p<)E#~ChL^-k|Xz(rSHW*%;Up6WDP
zTTv?$jy+qQWe%H9*!6KOZ+63i7g7T#{8p6vM6_Go4tINU8$NuN8J<G+({cuxqGV@2
z?3Wb3?K68ipv$+qNb9uI8$({-{b~qxVW~$k=MKb!_TFCK!E~Z<axS$gie$j7D-gY|
zxpqkO*-NHpWY~pSZ5`q%vGTGt|K9%tVGN=xJ!HN;?3W~|7F(7ZAxJ8Xj*vd1Pp7AB
zVf@Hw#|1bh!(`86GWBb|*GR8onJwGGF2L!;1K($MeyWFE$RO8Y08JjQ$*C@!%EoSP
z#mA>txitOtm<~|o=Ex(}C9*73I9*Tn2v{!Lh=@&1WTdt{*~!;y<=)xcL^FJ4=2|1v
zF&vIRPi2`zcrE&dtoQ`1<{=7sWEa0sdGXzbG*l`LbQAry8Y!^=>(6fJ^BAFLLZ9?o
zvILW<&@I35r<xh+xY$HvyurHN$DObDNPT~%$j-d?V^ZRorp%l73>Cw#Vgkne`;Pni
z{az?r<(cNOdmHXtOH_c701h@%wL8s?xRxf+SB3X=@QnZb6s<pk4cdV<trEC-1<29!
z*S-BN7HwyflN@pI^)o|(KISw{Wy5{=tP%40qZ$e4m<9kvsAt+dBaNMk$a0#GY4Y$M
zA+2WS5?<)=g8+AUF>72NN_?f^tN$zBLYg#x+FA+=3RYp$Sr7b-<f%}(diC8&VR^Qi
z5MxYk?Hq?8r@#@EAfec5F`8MvsEWJ2g6U;5k=@!h%1TONzM2zp?aN!!XsVu5`W|zO
zwOX=I%Z`ayH2+K^*OSokUGLQIC(~x-T!MwhDLrVC$prF+Iuz2jV!Q}G_$qvqGiE=B
z0HY?@*Poyatl0D5$<|*$b^*I;4$aJBd>A}?wyLSAnH+qMIW=;*H7|inRTkPI8$w)i
ziH?qL{(ZM?-auRX6R`AduyL$cbGakmFl;{<$z<gQ@hv`z?T9-fa6S%wmF%kzn$Lw8
z(woPZ7Ea~0NDw%ec!vY0I`!U32wO75;QO?0vQb?aayl}|IET-w8;LpT$U=@Bdf`>;
z;9%VP%?Q`8=0JmoJ}R-Rl-zO9;6GZ=^P%X4C%F_qe>8}AkbZ`VO$~Ax{UEr}Pl3}P
zK^8!Vf(h+QgVt7SDk=zB{mMknt{$kN3Ec`TstX&J=#QY@=RiwQj-^QT-C^lBZcf4H
zNfv~isE}LYP6V_#j}>9ZXmUN7=Pf7(#AzLvFz2;aVdCYj_1znY$%`G(lF*d~8sM2!
zM``fqrAE?|{xZa0MAJy7T{l?u4xc*tO7sbxVg}%Y7_E2*i<hvY-129#ub}nJy=wrO
z8ThP`T0OucxUuk)7o<U6yUCc8-6CZH#Q*Ld^#)zo^Wp;3&D6d9U->m$nxEYQ>O3lX
zMn1lspf$ev@|6tE!PVy*Xk_%3{D9YL2|zyHo1=%|^7Av>>IonnfX8HnTY|;Tbth&r
zhDZrOdJNS)RrbT5R{g}F1xBIyFJSe7-EZlU!H}`UqAdNT<wWLEs;H_}v6d`<Omfm@
zSrCMXMpPS9<`d7a=d&7g1KE;jJy7z&A^#0d+yH~O*mzquZTj?0J~*jt9jGu@!Qn<$
z^%__nsN8E)@2(`1>t?40^04iJs)nY{!;^BDEKxgIF2e>dlyIaI#Nsqe1u0AZ&(C*T
zoMzp@mTGTpCEW3rxlfxGS?CmZGG0qR6~?rGKNao<3X+cjl{%5me!H}2cjec?tt79`
z8w*;YyeR;{ArY@)4HK{q1h;j<a3oU6GTMf|ap>8Z(P<u2q7ZRfm|=o==>YI9i5E=Q
z3#j8U3lBIe@%|#MTXcLZ;FAuLPbpMmDsXs7*GYUb)q+KFttM*QKtoFCvm7BL7YE>M
zBmL|d&}&b`r@qWnGw6+9YYbbDTGK#JGA{sBv=3m$0xl)wm1H*Us$O(I2zK2RZ817O
zeCs)-mH#-hBEcBV0C_--u-qQ-OO31wlkd4ltCD#Q^`qN9US4|S;UJcDZ`mFeu=@)V
zfp5}&gnL9JSm@Xhi2-p=tc{Vlb}gb*%^#15z<5ZCfMAqHt11Qf1KJ73Q4Y-Jyk9D=
z3j`4Y)qc+V{H^czxfl8Fs!28+kWR;_X1o$vMZ0;|mKS=Ia?o)DGORg}yn+Yt73Pib
zP7F88aU36PnsqNdr4M~o6#8SQu24X3(m^-J3!A&%qI0-HY%e*T@Q@2&U-Ozn$K?~1
zFv1Z;yM(DypJ70Yw&q*-L=nms(7pui`9H?o?E^3?M|NH8$M4;<X%l=|6v);(e;)GQ
zWw}5Yl5}qDFdyV1_K<k^z)%Bz^`HIw@Fvq!2o#1^=TWuvt%%UrZ$2CrVVcn*C452)
z<6E?78NQeA7kTLE8Gb*U$ssS#5Ig-<K`G%zz#Amc3`{sc@O$BO*xbAane>e9kK@2(
zE!0y$6$VdE7wkK*mr!ajF15(pghFlw3SxZH*t|b)@%(}v1{7dkXntd(Uk|5#$n4P-
ze_r@FU<>v)C%WP++>E`AW*Yp+!muWZng^9)_sbEqyjp7#mCvI;Ha!v^NS-{op<p23
z=PO{Cb~nRvKSbeX=F}rCWB6@Y$tH3aYBroEDKmf)eP~eT{`AEY4J0@j$O@)`_^s(J
zKjX)P0enYQ)4foPB;5G4?4()47cgnyH5vr)4<1Y+2IilUv3@X}x6OBFz}u=#V;K|k
zJ^k3rIG|et+yV|S&>L096apfA4Qp>ytXI{Pw&FRF$6cY5Uye49W!(jbc_WwK>DB`~
z6|-mA^GvLIDp!Ijo%`O-N+xIT!iMuK(fSv7$EVGF=gzDiDVqcePwY>?9jSlysd3v<
z5;dg}3>U?{ThF!b>R);AON%xfm2iy*n?{-*Oo#8{)L+LFr)8l3{i)I7NneT6Ei34k
z%aL4}M6P!r3Y+&<Gn0w$q!;@Mi|ujXoh=VmkI2HAK~2d^J%>4j=I<dbXEQ~KX<$2(
z)Kk;cHZ7bhKB(A7Z-2KLKRf3~;wn0{MIjtZf6*5Eoc$kFz^`Uqx-0)nx1c?!^Kbga
z{sh3<OG*4+|6~?4U#7eA*L5yk?3WBCw*L&IA-`hdDM9(q2O-Ji)aw=hk3SyPiU69o
zgP#7Gm)AHxe&vX)kPyOl!MAU#wX|AttTN2{K9Xs*_>~W!@9mDw!MOrE227`%-W6Nd
zcoFUo__EoGd+7@C&m0W_C{JOO5-_u1U`HtHVd{Jr$1_-uL9qNl=o`|wB>!;AB{h`?
zQI83;-gqy!;KakuSA>`3?pTcAQ^n$Z5p0!6jBR(Hwvl6sMc2DJUTZPdbcxGHpW!&q
zM~Hw>dtfYI8(jGDhV}2Y@15x&XV<ap>19Cjz*0V<6tR&IY7uVvX2|1(Pqm8^WJ*4!
zV$5H~_Bz~$Xcg_~<2FuTXQ3U2A5DaOFf|VBJXn|FtRKPBHV3_Xw>Ll3j`+B!b&XPb
zpVmwe)*ia}cktl=;6N($26nomz{+>m$<NJ_L1G{V6FLG6T(QN&@M7pr)VJmzhB){g
zeun#6f#O<fOn*u<_5+N2q3p#Bcw&}=6NQNM?)6TE`D9kb=D60zsEQB+b}~d5m^tDc
zLYXuQU06wyNyPmP$uzXIp10K1_Akr>r2v1hWaQ7^dM;i~C(st8e7PQh2G{Y?tr+p&
zA6#wX36H-`?m<J;fwgf9*44RMA`Aqlj!Y-PHDWx{bAY4?aPSAf^%ZPTBHVh_@iW`O
z4;3JOV?B7f_h1JKy^%IVtwczVbWYpMj8+)b+2LNE^SG%K-(enJW~l~#q&CP=yg$@N
zve&F#dlg6_^kNklP=;qkI*|y)I*2I5?E{~2r&1^GDH)cYGG<^U-~l}<jC7C{%7Eei
zAt^~0l?_f7KfEEKfb3Kf8gYE+?j8!Vzd!#i5A>NwaExL!WiDW*HMF!{$lnR)SWORO
z9v<R@7_i3gpE7cWT>*SYui)Y5W8m{A($58V28Uh;QbZC>#AX*8ylixyKY#9(n3#Ct
z<jIQ&X_abk06yDJ_D51`0Hf|UD`;#~qhjIU7_o|CO&A7~Yj^Av*;DYHcm6Y+N4r=%
zuu$cBL8c~NOAD5s20q8a>&OI=twD!n{9gx85LMGir*fRSDwWX^mkgPSGSGV#G*7_e
zV=gi$qd>3*i7djhvcQc#7%HtARN6^~aN?13mbq+=0rjHrMU22_XBM6qYH4Xb@nu#U
zw{)!>wqW^fL1`J8i|F}Wa`ru;_xOw+UOoy%))6qE^b8Fn;idE3a3LnH0JK$R9Ei%;
z<LIGBOQQ}^z~WPK1}D<*4C+o-Viiusa_G=GCIR%zZ`;@$5r#AWfg*F@Vo4oX2O?Cx
zRkA`Q$8+{i_2MXmiLp5|QnGD(_r^;@gK~#OF31`0r;tLJG(F4as71xZEYP(8KfKCh
z5;Wp3KmlG0*rhb@`UKr9dh`zwG{6M_#kRGlrx0lY#;JU(ox}}(0<&>whQ9(r2lssA
zjve2S$c&E;{`jyQyho5?zz#+Y^#E}Mr>cklfc;}E>&k2gJAZ$FD3IX%>&kt-_C{+i
z74C0$U3{R-0yrIH$~5LP!Tx~oRd7!^kb7l?HDOfGCkO(-Z$;R0b#+|@zR*^-%U+n2
zxCxT<b{(W2cM(~zDc)jpVJTO}s__hrq9u-LPzEraa6^88Fr*Zx?t?5lyLRNn?@U)0
z7u2z2Z2An=KT7>|U{}Jty%N|E+_$4%EP}~~(YULjcLkisjM#=c;WxfXT~kr+pat4G
z__1F<e_jSt0}Qr+IMb!s#tIy_+;86=0`exD7lam9s5q^yt%0B3gfAS%`k+&qS5mSb
z@*40x^WYj{`CWpa`bw{Pyhg@(Kz=vP%|F9U1H@o5b17h=YS;4PIR4<&`3Aq-_WL(v
z2rHpjvBcm`<Xp>cfJKbk_#?$+7dlHd);TbDGp0vhtr+uOp}v72+A};nJY}x`%@iO8
zMU-?D+{kdnvT7a5U}GvR_@h7yUJvpT!#n()JrscrTq1&Y<mj7r0`Rb8-wwaf#xb#A
zl;gk|MgY^U%Xna@j`?%wW>yd?JJczO#<|W^NM5lfBBbrm3Z!^=cu?0|LgNZ7CyX>e
zdzj;z#>he%{0Q)O9ewdR`RL{Nnd<7E!!D<sa`=;UM$ZG)CvhN#+6Zl>(_fwj^xecB
zMQgkgiX?>iq#h*0U@@<A#`D;vBsStGCn*n87}D!{VQOY33%!VQzeWxoDV$Lr;2}d4
z9bl*_GE{@+BqMGo=8b7%3rd4+#{>6-F4W{O!7xxQAXKQFGq`NREk`}RFx^f|reS!K
z;hZRT1Cr3SR{&pG#{bkA;93a&qRgT=9AJvphA<R^X+c#c5DUD9Z<HdWf!IpoAl(xu
zd=xDDu$;j8abcGio>>UE@m@#d?}87FqpSeH4sig)7>X+h%@D-ow$4L(6OQbjFfc$*
zDn|(d^aKVTO-XyV!sN$f0dxd~;PdP8W@2bd5T6_XDDbDc19Ihi@a%Dz1(%rV>*u33
zHpJin=vwCx9$)+V5-lvD`Vra(Y<|owrMM>k3}y@whjUzIFzy_Mye}5<i`TEk(SRBr
zA3p@N4D^i(Y%FBc2~C!#xJAk`!aSfHCp}D*_{)$Gr<k>rqOu^#2~MCY>IZP`{*2Ix
zuyg`1eGm@-ijFaG;Po;*Tn1;6+jWDMjB6tg6&ngEgc#slC<44<>jPnGOtvMo4SDGP
zl;DT-7~y5okn|SB>#axRmgK>}#5Zo=PIeIU#h4vL_>o%byR@l*rFX+(3&eLYkWRqq
z2w5-~F3HWy?Tp9`gA;XIu%}?9;BF1Hn9=r3fEN*{VWJQ~VSvD)BS#)0CF@ynnA?iD
zAOTM!&evepY&_ndNsDS6QSdlCiy^_9`Tat+GAkY<x9&kKOW6O!;0+L-<*Q>F>08OL
z9V!>6Ns_~jEF!FMGIR}Lb-sT6T9-Zo{#GE)%vY?9*W7$rRki0d=k<{Fl>CB%Ep&9Q
zFJ35yo$<!+CSv|7%7F7&Qzv3y*ers+52l0HU;?g@YKLXOwe$s-`}gIrM2?ObblFzB
zIl$x7Buj6v{V)~K9)cs+#|7IsNxeS4KTpp^JqNZ#g_9@I)PeRw9#aAahHB72*ob+|
z_wMCo1b!I_NaDBml=eR*PxU=N|2*-a#!8LCHA~g;IxS@WL1nItDt-KwVLIIsM4gnR
z!3?9%2()g9IdKglu~n@QOsP-6a^29S39^fo5u`Kz+W|+@=){CFGOOb@IWHB6CGuK1
zh)4N);Fd0j19Pn$^eeStFnJQEGD3D^s|T~y%Tz|thBR;?{9F3X9)^$);j{zBMCeYx
zMto!z-o|!WDcmPbK{8f7O`#@ELyb-Sqa4#bVDs;vKc7uYOX~?SNR5tu+XYZso$5Q1
ziTJ|zfq{WC7F{?)n1u-eKfL~Cul`tl&obk0*+|C~U<!T&$Mi(b;=C-=A|l4NAPaTy
z_>&AH%MukNQRvA~MPYOB^&tExI1}wbk|CNSYvquL*RtM%2!gK<rr5GnQ8+#ul5l=7
z3zKoDen2w>muB{4+eaoYU@t3UK8G?0dnZfk>s8}vhEeTo3xbn#HI9L}#NzU?xcaUn
z4QBxX0p(t`X&lbmf?xs~{9*_kneR0<%Jud2hVd&!gqC9XH^Lia6<)bwg{}_|yuhi9
zkkLh%wPcg=I#d{_pfWQv6N!M^_Z#rXMQ4nZQiOt{LU0I!qasmOMghv~HEBnTA`4(v
z$_j+Eo@GZXConNVIm^}xJ-Qe&T|Cx`OT!5_v>LAktsqIeeyl`4RR?N59KLBM`VaR&
zVnCSy(Wrm$jZK+Gj-v+Aq~eSvuXV4sW?h<>Jlz$M!5Z4Vf=bS0(q2Q3Q%=<7<fs#v
z6y|JhZG91rM<x7y3(;PdRF3U`CI^ZDP@u<k5FJ9KYN$_OOhZR(^a184YF>yS%@H!S
z4W4Nr2IgU#5T{oY)+D5!1|Lg+Z^06M3!K_ZK~KZ(_m~7wZcl0qe7%K%Vg1}9<rBKF
zHO!=dg6c&p-p|yUIO3FhhIzAI^1yEF?a)fMWCouEJ6|Nc6u(!|>51G??U+@aI$<6W
zax_wOoX{)R2_5zow<pdjzCO!|Z6j(XGMR*ejMoMu(NW+)Ib?xsAIzNff||SnLl`+=
zEA5h%nMpyGMol4I3JStsg`qXJSTWM*X9IjY6NYdW)Nb(bO?i14$`Vd{&H;#S>)Mil
zb>~KpJ=3(ks9{9G=?OYwtjy(xcCSqUBhbPs!-ux4oS&Qffxh*t#37@&a9Y7>;rP(D
zKF4OxcRdLoP>}?T<1#J!YT1ZRCu_U;pZcW|CSL|df|#iTuYMF<yhU@q^%rp}7J-!V
z`ndKo@a%(-Sob$&&><Ve8i_86H>AQ@coy09C*X~+Rai?*E?}~Y*%j9UADl^z?f~Z9
zJly<vO2XXy1bL7|iGfy2gF$Bu0j81P;sg`Y<u!b`zfMP8eK(GIdg~X}#(kvO7Zq-k
zgL3sDaL2zTR1i)&1b6H*kIx3DZn$?!AKNoRIeI%FPhIR9D9XHnr_xvd8J2#*o@;gn
zi3r9yE?=B&StQy+*bEUrE+Wo@4ulLR$9DBdO-;29KxjGv_zgk+^<Z_MkF-H-Pdf_s
z8v!?qD<O+ri{M8LxUt7KZr=QvR%nj+@{+Mj7-rCOQoAG$niQISdF+oPmv2Gc@e8C^
z3UW|=6@A+~Ptif_2{pXy8+lLu>AcRnXPQmFly9|>JP=SHS+ZFjSg6Gn?uVJW1o&s2
zpe5ZmKVLXVmww)QN+8jsp}ziCXJ<Z;B@&tQ!@}rYKb4fQHI4VLP{$@6_f-iOkfEoC
zFCPwwugJ;4?^8%M)j&ZC-^t7~B`KWgZ#28c-aJD!O*#S4QRQI~03V~)bQfE_*fKw5
z^;ecC-hzx0HRb_`4viaA+|h0_NGnBl>U`-~B4Cbq2h|1yeIR}_Aok4w3QNv2N1WcY
zW5+{K8DwT~Hc!tkf$O-<$;Za9NlOrTRI=^UEa$h;V-^l0II5W7a2!)9$iV2>3G>Q{
zx~F{{i3X<9_k3?P2PkA(fbSBH#E!tks4wK8%IC>mk67ZHi7ezG3yWm<FL8Wxcydx7
zXecK7VMLD)k7ivAK~hG?$KliP0`IyES;flp9=i)c@-u|01b>`|S_ilsW?#8tJ`n~P
zLff_th$)h|*Ib61k)V`X^7{r!LHEg-HDh#H_9emxGA@F`BPgM&Ci&rKaF~%oo_E{v
zcD>h*vt#)jS!aHoc&zy!E&vyfo&eJT9G7KK{+&3=IiR91oO0L8WzP$@-ZjTD8xEtf
z&Y?-{gRO;K(?Df`i9?DetyOXe67}^?C28VKe`F6%T**=8&~4#t)Yoq;IeOy{pP88^
z)9WTo+k{N61b4U|a}NP=uGCzqg~cuj5<tp#i6L9trZPKyKFjW0)X8MH50q>~IZWgn
zkYmds(zj0h^}x$Ry{=+M5;i`z?R&^UP}(IV$3gKz!3YLh0Y)T4J@tv<*3UU90JtoR
z3c(;CsUHwr2wW;Lt%P_$U(#~Gfyp**dv5hxQXi2hj0KA{_-#s^>_r4XqQi-ci}Ud)
z4AbtOxMpTnh9xK=Df#)!IU(Im;L<*O_Dr*kZKNduiVm>o384W`K`Bb9C)T*&sa^76
zGo)YRw(hEzFK>o*4~!0eBpP^v9(4<PVPIN}n2_d!{`*ZhH>1Nuq87LYM^)ZCBjN6j
z2~Kl95u7%{%g=9MH@X2p7vO_$Umk5q1CmPE5J*^%N!<aP9;NuYW+BA-)|ys7L$iE1
z3HQ|UBBvGG;-SmPRB%WD+g11T{WXvfCt^SYrczkGLay8OvA<4W$fsh(uS&l&e5g~^
z^me2V`IuBa9avCA$^HN|RhaXlwjhq+WUv}YvRBN^0!>pV`)&NXvVffsV1^7-g##)K
ziO5NV#P;RMvc`pe(iL6YhK~dy8EIAlu<|*5+dTc#V>j5rt`N`J_5wyfD0pnk%IPbT
zBH%W$Tr;bf3yyk-f_x%@*IaTQq2Y)W1f}&=yzc<tlPThdX!zs(ew4PmPu(EaaK`bt
z9V*b;m15gK?Nf8R^pYgjDh-qaci{bjSjahT3p2`77y{IaNC$DUltCjd3`OL6;AwTX
zzm6?5;Vu&1-FnB7=UYHmpxJr^?#`+%v5MSTNnBICIeu<b)6|#+=m*@X$352jRhaVA
zFh-BCZ6sx~8igv$Ky`bN`|j)d7bGuO=#)G}e+<Vbkj;(g!eGp=Psd44mQ~`ZqUsH*
zzGzty8vq>39o1;4&cH8R5{tm2mnz;#iVG%p;<%N_I;&`$F~33!)Jg6FEBmIgH$F)3
zm)`)@EK3iG>bHyZE`xUZjhW~trQ{FHG9g(ZLxf?RqSd(hczEC@n~bieii%CZM;vE5
z=+IWoLpFsy;caa%Y%?1ufQQtUgJzNss5uv7RL)@HH)=r6Bh#zmUA|GF?0Dat09=Nc
zjxcQKKp8-eBYJc()L!XN`{AenuuO*LJ$&>CBJ;Im%<F8|&=74deh<vbeW=ecwzqAb
zcq(94k|$skoXgyJ{Y9Z>8|YI@CzO%<1E1h^oI6DZ1fx^So4$v;tPt#~>c&Zca{z$E
zZ@F9<Y1z1G(-|cEm|1%TJTo)yCB_1{R*|44A==CGu)mv$^|y|Q&LK3!_h{c%!D40I
zauFmwgbM=K=xcrv+mFam%J*7aCnvkO`T5u=3Ypl42m<O~HT{ttn(iwvC0&yz>U*|u
zc%7gi7LH%(4WtM1BR<+=m=f2Cjaa{vR%+~1xf^fTO&m~`8Qj#=)XU<V_l4HLy9pD5
zik=-5OZaJOyN>s;52kw20zD#tD$a`QRe^Aj0HBp@CdJ7wM1LF4l^#UO)|&epBk%EI
zWi}KtC0y;!CP^N6*A~OAXNYiK@uD^L=Y?=mCxgp~lJ~5ckACK}>Oj(7euF0LKy8rA
zn_DF*33hB`FmC8p6e~J-oJ+`xiBV=r*&9*-YLSReT6ZWV5GxH{&6q*+2SQWQ@-NG)
zqm*f6Wo40D<|7(gjyhpn4uv>W+tgI)LmCFlImOwt3hf}oHgvs0i-))n>+0&JJGD4W
zF;Jl1IDmxN%v@k(!H&>q)l$e{Q;4Z_c{TNrS7FtbQ~Uf<_I%|Q$AO>)An*}tU2H$G
zx^}q9JjOYPM623x<t+9vaTueZ$=N*qcGKcA&}G8WuEYaz1_6YlXn*j!M+UlIp4xHG
zEhf|N*%QY(IfwdQ_5Bqp2==O@FC8nGxXn<xa(X=sr*`4Qxr(?zfDdFGZ9P4Yz#+u4
zAs7tdMS%L}I5#3TpmG~^pm9rf0NU%7c$QuYgt37{oOC?#m#{_U1VCU#zk<YkbT0@Z
zhSjoZ=T1rN0T^5yAkzr3r?8Z%;?oTs$1B7LFb_b3h(CyNJ<(<XRU>ncOHXaU(hQb(
zaTu9MBGL${sT|lTlF1eL0*_V%fCO=Bf{hF@&qprU{d+HjCOCkAbNi!=LQZ?;>eT~S
zzobzKUbq~3@4EwNL_8$)2|pTnGS2pJ^b|1cY9n}0z&+6V%>&&Pxa0}rAIFhg7r}q*
z^#EV?;O#OgFy+W}bG!rWINZrYhC@C=XEGaiRq6Tp!;`3=Kao5J*q0#2alqfC9%?Ou
zgu=D~ZhU-21KHPsg%F=K<hV;`92y@<NxyLuj;ep$2n-3X0dGsRu+!j)Oom^;FC+x~
z1fN%C-M69eIYJSd8(vPpQ54N39{!h+$b5rO2r-@|fTPU9vxdgGrDJ9&gFGEfZU#?|
z2m#RR5$Dmu;!kFaJ}mpD7(M#eK&|oOD>VhKqauGwQchrL81snUwE<c`gg1iF+9&nx
zxK92$<Q7C5OOi06+=iVhKtmA5q>(2j3!TM{_BSQVgF4gJvG9neO%94V5AaKIKT6hV
zRT76?0EQGqwMe<Q%Yc3~>@oKJ5L%@J3qBasmrm$)Pjt7nEsdf8zI8+`C1x_U7#DR4
z+&XP-ZDe^h?-}TZ9-x8q{KbnxeC&JF>m_-B%Fz-U)(H;-_;y}yTt5ZCH#ho?dE`(R
zl0!&n=<VY-AUtBpkn9PMtO`@u;Wd%0(a9w3>lORxUGrMT=8ptOL9ZtgADI=KkQCDB
zGQ7m=|LKj6uf`^_8bykLNgLrThz7*%fLN#C><a~K7xPBYAqrb68c32dC=YONx@~92
zDvTMEy<Pwt_jC*m#iLRP268}V#IIeqPKoUM=ypW2mq@&cHyasm562lfG^Z3{+E-pk
zhcRN5nk2(2<AGTK`${-EUdG18CZwcXB;#;#94MizlXL!qG~Q}d$Z<|(J^hJue>)bi
zf~SXJV4~c&M|HF2$3{lvosm#fHOA@JgGmgG5k>`={Cn`=#CLe8C?dmDLg`2_ncKH-
zM*@bA0H4aLvzXqb32Q0V>-+Rvh+g;YvKS9vJuIZEFfeS-!N6ZFEj^tDU`z*COVsOj
z1fd7ug$ebeBO`@Lj=c4^V4Cpop&ags0vh2gC`+PH<n#jnCR4chx{k54%TV>ylCgwq
z<!}(zNC;VWKV7B-nED^crZhpl1QN{ehE<5dVH9rc&j5R5ipuCn$b%<BuA0nLOEzoK
zWDs&xkC#DCPKJ3Q6O)>UtzJW!w2jS2O@?&>Lz2uj$=5IGUB4bphSniz4E9nWW3}`w
zX#fC*FZ053gcX%IkLUiSWZ>*v4nP#veMLDqi?7vHoQ{@$b6Pb%6O54v92vx3%P6p(
z^utI>;5`ND4a<=u;^cB>=QR)xCW!7TdXRLroO@exoIIwq<}y$f9WA<EmzO8vr!_Wb
zY+&FWJBmlE4Cl0jmpsvZXfe7+p>0E89#vJ<NHk$0fv_7y8;>|*pj_GLpMf7Af=@b~
zA6aK$&~>zkz<;HDNYile)!-78QO+-oflE$S#(P!IWG%4oV4k2g^ElX?M&b|lgOtI#
z5=<<4ffgYXP8+qRLw|L&Z31!RJ!x%iH7-kH`6Bccp9|wU9b-+s&StWj9e6X6{gi_u
z8s?>dxxKt<4%K9=A{j(NOIs(gC^R|9AJ;kWxqQ0sTC|Lv?1OLJ=T&*=>u;9yiyx`~
z{>?+_;kob42mKC3>xN!1?|XjJeKb+5l6_}j%ii%UhH$IRp|g&Hd%AAM4^55D+qbk>
z@pU-}9uyMVi$%DSvNPr-;V|5v>%wiPhH)>M>}}<G_Te`q)6piyC(dTO9+`1lT()!h
zQ?DLCaOi36t#i#j3jES$W|R@F+IzreKIi9C!={N~$UqH`8&w~{c|Oi`PI`$WAzm_q
z!LKnI8LxId@p-Bxs;#}hy1JT$%aLzCxBJqo_b!u)s)ms9*T8@~%1EMgKfQ+*??K8-
zLfz)z_|UuI+0vzwe|`%+*zEitx&Ypz`YIRozJJjil6Npt{xQs09r@Mq$Dw^a^z=ee
zM&#4}{W?b6*_rk)Yed=p_jiv_iU0FE;%w*F2><(UI3y{X|M^<siTtvs|GljHPcOUu
zU)SYDIdsbQ_v)?F_K>AM%tXRy_BOXGFXd9ad67(@Aj4XF!FO?*u~M%WdAGBvoxM{Q
zKX}|oYWH84_%w=`hN)au5x!GafE4&#^8(fmT<4D#`|`NDr1SZ{_PzQ(fB`1R2b5AQ
z!e!)I$1XS^^2!9w;i<gW8`Fn_0TuP1Kjl?5qDWDRQ4Vgs<X+DGi+XJ}48kTg$rtF*
z)<S1*wbR$HcM95-Z;6LimRH|Lwh9Q~T=PXq$+f85;H40mJIdzA6P)HlPiRnV*I@(*
zj4mi7WW0&P$+N%T7*BXRf=78}WmK)o_e!+myZibUljdJqYdTu*=B@`A_w>!14gCBW
zy&L|r-h3)qSW@yJ)tjz1UW=BOH%+nzDb8Bgis1cFae^tm{0}v?%|bstI#PFlAZUZL
zvt#J@mzS~^E{K4t0f*R-&3)HFj+<IUeRDRjIg3yUxJ+u-f&RJzJsx}dA_w~$!?{%I
z?vL+gX4;2dW%`}-xL}8l(7U#AH@Cg=1IK<%wm1dj;3nV|@Zp>N(>GzNsB}w3s08;g
zC_4DwgD2ayk(bsmcszyaXb|$lkCrnha56>n`|rG0dSS+PqJIr9FYj(X`;X}&e=V;N
z1)O%992ZU_`+Vl(gOETQL$f*;i6Co|gQKJ1P9CHCNl9E#{!x|q@^sZg?ep%sH+$>Y
zjJTBKUDQLA+#laC<#_EhD|_6}KbR%xrpjq>wKeDrX@}OmcHwY&3l0O|=HbbaPaz7=
zAKkA$JjWtssYutlmo(WZ&PcbHQBq9n6#{<ES%M>+kHR3D-%SB;weZx|!ra_ZJWHs^
zlkZrv=$SKAZ7jog{(dOFF=#+POt+Q}dhq55hJYQv(JBlW^U}VVRTR`1$GN!985yzQ
z2P;zKmEfK*FDWT`V+SWSa?(xR5(Vue$8VJHqvTeFgh+ms`Q}8=NQovxAIXGNbTf2*
zy%UAm*H^3tLd#T0C2W7pzkC}Vyqv+rL?zYC6F5kFW4;ey23X@gcolZoGAUUb#`DOE
zazW(Bp25~xPeBEHe_5a>1hzB-L}q`sPjcsSG}$-u-|BZpB6U3K5^b-M(Y32aj~}tK
z+l>BRb7`=V111o=T-deL_}SIp3kNk<=2K4FO&$f>PP5#>AtEXwH$+?H6dzw3dVb3~
zWL)l(z4xlbZ_~DI4>Ser&Wfq+`eSR`T_3y(gzXDC$F+KTdU5hEr<7xo&RZMIi@L&d
zZDTL}&a=v*E8V4}j;GvL@vs*#XiPQ918`d&<mdoPkh4chk$9=7s-Dr1UO%yJJ;$ed
zdLJJjBFe<N(GH5I)nH)wG1>dzKL4#1lOqE^>_FVW?&d&fCD9SvjC?xZojbw%lz!B1
z4e(Lub$8oceQCe5A{q>reSNzzp$GlMOTWexDt9=e4}#TncGdOk*9?u$;!W9#OE^Hp
z3<(KY?G*q$6(STkPkV9egdu(ndK<Ulh0A*Kc3}BrW)|f~b|;Sh%3B47Bwc6DJhc@!
z_qFSWZ|@~@gCQU<C$~@TBscdq@EM8J*wV@}C^mHE1sWO~8=HNstTYr@ORvylWYje<
zAPL&GJyXw@g630Sebh^2%vfmZC3*QzquW@I=hY}*)3CNa*_`b_!^p^p6|<a@kdW}y
z+q-~~La$>pHOMffI;#eFMkT|V4a~*+wTzDQ?3Z)S+><Pj{pPgIol4~^hIG37`|n0M
zE`0g&&V%PYzQ9`g19-ncX;dQSCQTYi5{|xj-tEsTz7-W69RDpL<#U>TYN+X6*=PqP
zd)@m#Vu6C+%F0}yJfYdiZR+L(`ok4|{@3%9yC`_g^zR-z6`vmfC=f1?2{>NR!~kH#
zW_)<R_2AD(NKg;_DHD1N^eL$+ratj)XskO%2*K!K4;+%2$2^OGoVE%YY=AV2143)e
zhio-H3yNg3Ta~u<YJP{EC~8+w#Kk3-QIJp)ELjzeexSBCbJHB1=^4Rb?mx`3iY^}M
zg#z<|a&|W6O1R*{p(ncwWemj`QY|UVn|Cl|TD|FKP*|7?8eLasQ%o?iTBJKt5nL6#
zzN9#1y3%M6m^R3Grz|b4R<_}3pKi`{paC3sWO(LB(rr$_ozDUTJwan}v%ZmwLgxl1
z1w)9w7pSq==cbIO&!4k!SQo}mWZNw4Chjm0Q~v7jcN-tKgv0ICUN7dSFJFEQiqCnB
zr}-8!4!1iGP7v}rXO@SAhwGp;A`2J6Fz%(1YFFr?4Z&guxwy90GVsBccNOvg?|Ahy
zC@@K`al!kT2x{kjU_zMLi_7F^&WzzMKLS(==sSPjZfc{&n0{Kj?!|G{ndy;(#(cN(
z7Z;|4Nb~;UKw~@FUZ-x(o$4$JAhF~M01`^J{d9L->M;~jWU<HfjiJzL&Cy@~E?y&V
zX7~nx!Lyeo*Mb*CNZokO0kgtOeo$w>IcezYA4YcGnr6vtZf@S1VYDrkeWG2AB5Z5R
z2U47SqCx3C9v(VMdCYAK#7{aj(L!rA<m8-F&6-=!iLb_ZJa?4RSI}7#nsi<r`QQkZ
z$^#rhA58i#Qk=nFrZ_{q@#$8FWPg3uE}VZI&-dqgZ`S}_O$6PR8xT&KZmdFCPqyJn
z;~x|z0|P)3v}nP6D=hrv;=YRm9h*3hK>^#KOJ#2;&V0(JD!!qA6%-o!06BQDOn}p_
z$0*Xap<h^lpnE}z;_K&k5OW(q5tNoHM_r>ZH8MG%25s&}UtiycnKtr45;g|Ac8w3!
z(GFU4?mK;88Rg@O3e@;3Hann_CWkdhz@RG-;~<48Lmi!a0LYe6Ncqobub`;-tFssQ
zfggczTKU%RK2AaHdLxi?72Sb@RC9j6`H?~biE{4L*H@o;cM9%TPj5`ID=8^aDAIo1
zB0nM(Ns@W-&!5lr_4E{Cb4dE~r>w`1padPkI79CpcJf)E2z*BDOY0n>>^pYM^NoOQ
z>+}a5^<~3D(>7e!n|B{P@3!%y0lF1;&L2LEJ>0t4IW)lg`F$6c16X;aX#{oQ$B*av
zPvz$3QXHc%L9G~g!G<o@2|GMaRLr8?nGi;el(xpHXS-#rS#zx+Mm2JFYlo`pVEa9|
zA|dXro@0T6=6<%*+?}Uem^Q-s65VeCYJn4(mkYyc_I=K*{P`7|+@+;&V&+7q?aXEX
zeD94~_E4~4FLpZol|6rRdX(^db6rsw7Qn}i^o0{1%W_K^?2s7ZA!~ndUFG%i<;X4j
z(_p4}{K|9Le=U!i0Cy3lEiYL=7(=Dw{O;X>&ayz-t~Fcs9XvR;#l!7s1JY;ZZbV|D
zj&p@0a1a(6g`%OMfz*dMnW4A8EYbB1uLU*on)ka-t6ZTzNZ2<mAC}`F`stSTjE-Fx
z+9nXkw*Kfs=gtEHm-VP0lx>ebS5UB79Ap3hML`*Bnr7v?;r(FwTgl0ro#;a>%+aH_
zxP2$>JA9sqOYwU^wt4FBzfM_M`Hodf$|x&eOT1=wjM7mgU*|7g+-}?0(AHk374bMC
zD)KE@MIdH=1xoieEG(Zf8L-s;r*o{OVAWw{qgH-`oPxq<M7sNN!EijGv<~?3bmX&O
z1yHpZYgx`6d&1z9voK!~uX1T<ZqEL9SC`27^VI3-=_IU<j*jjdN#@oO4LIevn{vFl
zP$HtHdHQC1YpN}`jO&9n$u<mzW@U_vA3sV<-GKjBru{@z@rs}$W#5r#?SnP?&GrJC
zYTURx6&p5uL=Oc+UCKcwic9Xu%DDB+*O%qSUKqcvM{X9~ZMc*r&isPbRHZ}5gq4|j
z&D^x~!w7Uvo<4haCnQ{!7?;kO^{XIjbOvBa5mpmh1+s+;TIspB3IUw9d1gCQL>tJG
z6B*FW=HB&s+J<Y;3COft+=JAKaOK_zIBftWQ*zn!z{5C(kzpub4#&ButEU&8R66>9
zXnXT$F59+k_(VuFt2C-ahGb5u5J{oIlzB?#sFcX8A|z5MijaAhP|8%M$ULQzDMJ(@
zA=A4Zb=|}BtZ#kmTkBol^T)I9`@Syxe&=}%`@ZelaEdl#iFMnn^+koL3%-6AFCT{0
zI10w?cc@njcDgSZIQ<Mx1=Z0M%toQ$z7Ec_PrEG6P2FB5Y1|J=;llKrQ&Q`rWEn3H
zykANFW6ITW+n1K7oA8FI$fmS7Vwi(ff{2YBcc@*7x6;3~j^Dk1Un5a-xILelg5~>4
zHs&8XnCV{+3)?U`@-$C;XkcJ5f3i;QZlE{x{*2Vwp%!}TOK<=EjmbFI>cqJ71Pa<E
z_wHpwrBZ}>o5CjJCy|lTUlw&<cbli7M0vXMEQSW8Pn=i=6n(9>*{Rod_mVp8k_WQx
zhJ<*bj>e9b-l{xh-39=GhGJQ|R0!w?RKjm?N?#soS^7OSwLVdG=fKym!RuF(*Hlrv
z$F^kgVL3T<i*O)kx+fWBc;8fgd^od!-%y;RAprA1Ht{oOB9oIz!SeHaYkCO=SJt<%
zueL%>MdW#2y1Wher9fAiitdl>R)nY@Ol@4ihR80J8jf7dLtJt-^^ZhE3V-UaDaF?O
z=LPu^B?e$pEQ>=p_yZq2s7STKc#3t+nu{>v-+_5c_T|diuseY~X5R|FMcc;CzTP%2
zc2zAHXa_wu;JNmv^g!-_CNPe2PK!F#Y=YUU+oYuLDiu4$eQ9|HSzX}N&{nyX2n5Ns
zo<XBCP?u?(*z2o3xQ55l?)2S;RaPB2*H){`t-gYjkQlJT)y)fB?G2*lMeTgBb}7X?
zm@j)FCB4u&=!z6<GPjF}tVTn&>Zqoob2)kTDy%x~DiJ=>y=|p;qc-lbIgG0}6C6IA
zxZU9RPz9Hje~tY*0vwEQvp9L^N7c0lzw$Kc^g4-Ie0ZiX#Iaxv_8_?wH8Y*})jZzM
zSe|KI)qZ5citg5!wk7~FqtCO`?=y_(K<`#yk{)Y>nCvk3zPWsKu}RMgd|^AcrtRrf
z5jJBSn}=TJSSt$T+VR0unwh#Y6V|CvTfv=W^EKTk<os$QIp34%`dgIqSiqDl0uR<t
z=+jH5b>tzku#|i`b}zE1zX!XIPgmm@azQ8xQyb1S0fj!+;F-WF1D#umiOIyUSIXH<
zN*k>elswxwIXxTAY8Kya#jn)-n%7$pkqkJ{V|R@l);SIh<-xOD!5_@%9|w152FGxJ
z)t!y#{^2knWk|ti&>d)W7jW3U>sMXF%8GI;bbY>ZwQMR4!8lYnSdaV5AGKSue0h1a
zCv$zi{Bm;pq?-BBP)`e7-SUz80PMgE=pYk|fO<?q?n&j7>9{GnfF&#r@OH=fC*3At
zHyDZ5scHQUrs<XBk#QPnG!F(D+z8safB!~|59k8QWs81Kyw7*at+VQhJ5FNkqpmXs
zqM@hEo1P0%Tf1sDM@hg32b{ed;I&9=X{|?DL{K7eo%o)KqJMD#R4%%?HO}QGc)pKc
zb=M`>VxqL?M_r7Cs?uX2I_xNVYSWCwvQLxdvlG<|pXX*c(+WXuaX6sw(^#mH9c@z~
zByJP*97DTFsRLd~YOg*{<-Yxs$OAYpIPK%lAs3yWH}jr@XR!@O9#?5z?`cn8Nj_&u
z_74&g5O$a06unXTdG15=OJ$xP7==aQzK6_|{y@mCL%49TwlqBD|3W+!=}*v3+S_#W
zNy^h16f8gBJhcLiT#r`^56<Chbj8I~(L~JEAG7<`3TWRzgN`W6leE3LHf`7NO`-sm
zw0=}LuD2QQuM|^J&+fKo7OTg`#}n@_+ZHU%RK!c52^7IkKv^JA7ZjX3e}b~iHb3xj
z9d;$j6Qd#_qrVdpQij59(cLv5MUryqQhPfvl4V;0(9v1pQVn$y;a|0OEtlwL)M*=_
z{roXi&k+@(k5iI+(gC2Hp{1qo8qQZ;PtzCH2t3)C{9x+7SL};@`%H$*kt~5`e(&Yc
zttu)ii&|hrx)vw0EC!&ST#Kjm<udzOZ>8gTk~$uu&x;EWZ=A)c4h$;&u1&4ft*1|g
zV`5`}^i^;4ka`LN;sYTUi798v$&u}xD5_YGOFx~qN^b3j0_<n3Qs!IvBpJEKYdWh^
zs{Mq{%}a2Z1>!C~*O9Z7s&m64DME8*Xl{6_TT?=={IuHndbdjoM_&_4){PZoeB=cJ
zHF%+cF*3CkU7JlCuYo}tj2f3=;$2r-=A{$&B>5J^u^jyTILlKyMO_I7BV<>}`b<!~
z9IY8i1SMKgs;)mkstz$1INGCvV>$b3AKrxGI2yebNO#v!HSJkMe{ljVady;Y?{n<<
z@E3@9ahLux5{;!P=q8Enp8@&P<I81>6AvaZUs!{r8D@8Zp7LN59He_)zp>?5xBD`!
z<|4ZxqAJ9w<y|OY?1F>CWWmFOIo1YTNhJV{UmWB5WLiURV2rv}u44~CXV6*y<|YYH
z2dlU8c0D77w?wS5QCJANd=O<}7ox?DzW8(DW5pVmhtGM^z`d+%s&j|DVfgdQn;5Fe
z9MRrT8zEIb?=rI$b(`oL%Zo=Q25Bg?$4lX>KVWWU3`9w$7xi`L_wP#~_q_bQfxajF
zQDk)VH9!v;2)*Fu;W0AFl#*1B{oES&QokxB$}C+U{o`RA)|nq_m#@~-twR~IcmICb
zjv@+NuVUf|r>W_UZ>@H9my|-dgq@+8*-Z>Y41@9yGJv~<z571jY8}LcQRqZ~y%m3~
zVy0tQb{G~-3I_!`2J4?NGA!G0<B7a~B(MQCQS%kRC@%;Jufl}&HKNkJeC~0nfBoS-
zmID6=Ml8+JLG|R(;r9JNRJMt5peo%Y4Yx)Z+|_Q@xDSyJ>R^_ftT&@8IqC+itQwsV
zSk&=3zpkaeD=*K=!r~A>P6+(UZQ0qIf`j=dLwEiGU}Ly&Ja{aGiOE#m3YY;`$3Ha7
zlqqH5%BFD)PNOjGBo_koXBT0`di2<_V#I=?wQ!$HP}bE#qTIK(m+M~Ie<!XwzeYL&
zATl?%uqeLxO%nshLmC?Q`O@(#6RcV@(C}@Na(%>?j$Rf+{nyrg@3iYT-@Ehp_A2*|
zuTUL5M=BOZU6tVnlff!7M~WQKr;DgUHSGsT_T<3e;Pwl1j*HR^mY1>NUUrXMx=|2y
zjg1awZ)EfgWM;2JCxu0MomA>Km$8?@9@6vInrB9IA-lA$jrZ4VLs6xzt<5HJjt#a8
zo^|(j{&?Z1r2#9-6^?_fxWrao8D!n9_sScm*&$h3VtG!)z+qzN^#^~UY7+WTOd}t#
zi99bhiOc3bOwttFew$4{e@<k=VBLXV#|s#-xw+*Tn5{h!bGqUqGVdyY&kolMj$eJp
zcW`kQH#V*YdR+#BePn$6O*^aBGpLpgOo!A}Wib~Oa`}DPW-P>@)Hj%K8wbaARU;#!
z&PNH?C1}dRrS!lZ*G?xUVZ=5C3kKccLzP;^W8e*4!(@p>7d!yc#byiyvt$zr*vFFj
zO-wQc@;Htp9|w$jFHX_VUIH4O4AQM2|Ca`Regzxr4ZIOslZlCml^45cQ83&Z9q-Sy
zuiXMd${~4q#?`vIg2;WaiN<tsJ<K{^<4b@jMzYx#{hdin$BLeNzrw=cW=zsz%%0%V
zxCHzIR_#{E#h>vT&(3rMg+s%^gpJ3m7E#-8>;<o%=Xt01o|ZY<=8%#SXV?+-s`&S@
zW@2)jbMCVrW+azqCFCSXrS#`q%qNgW%C~bi!iuO@RYi>~)FapaXF{@4MN@MzIxMA(
ztmI+nJv)5<YfQ0{%&DK}y->GjyK#qCI*a;y*8cg$D!1y{{_|@vXc1K3$FMb<`e(95
z{(8*XCqn-WP!fw-r)CG2d8~+ujlcD3Y5U6<u7r2_%yn~$mR?EUp3S_;``7O-U=Juy
z9$v9HvFZM)Jo{tEPqhUxwim{5MgH^1{(i9-8FQBZ%i+sE59sgb{2#ykzh*=+4{*gs
z0oU*R;lQ@f_H1IW?WIdXNZ&!LE{j9)%bfG?pG?ST!m)P)w;gJ4BU4ibVEWFpQ_2a>
zs2V_=B?9QtY3GW{O3=2~g8d-t>RNDW8Wq74ur1J;UJD9h1zw2@L*+|fpYJBVX*=CQ
z*UbQ&7y`=QpuQ`7sjGGPDg&exo~YPlV2}w?fG04#GXKrDP*5-hoA`BsuuI%io0^+R
zQ3-acF{G2Y7Sq&HzO=?CB`w(!pwU-VTzu$@<9g$PhFE5Y16$w(;MsEPxBgrzF1Tyu
zJy9^j+g^S;xFp$bB<6J3fvdtmaaSVz1N=KR`@LeM)*#0R&{8%CvYIz9`}0D0it(`o
zkv?Qw1%CTx<1#-Yax7+BeN~E#U$8~NU|0&`1-Q%%6ych}R(KHxjito5Q}Xe0>L>;`
zm_uEF&jIRSLr?|@QUxttXJ4NerXFD2xhK1U?}pSW`p329cecqAunM;}A=VHtRWPlY
z(cK32aEWQy*z5+AaS@13z!cr=UyON=vEXPeuB}~;Du@jX7FJePBKN@9jRU6}6zLw2
zm}!7<u>~dV^|Z9zkmfD1u&{u*-VhBKL|UTr82*p=hR=pU?uUv;y*tlsewq=n4Ia4j
zD<*6{JYxdQkNaYH!bo#Rp3BT0jL^Akm82GN6b7NSW{7+(fSGaNAfloTaq4!g=>cZS
zMj;SwUT<Qy5-+zD(A3V1Zy*z<;Mtr*t<}u>M~1$bM-l2=x}rzn;aaF~=!$?)ZByF~
zBVF`2LMZe?K!w9u@xb5zMC>oI%LAuQYid^T3+&o;79g7`^#G|Dxz6B<>7FZ%jf)ee
zqoZSR2me{5jLqpsH~qb!ptH`BkHMi>(sU4PV+MEY_B{0%hhx>^nDCqiK{^Kqbze`<
z8!7_;v|8Su)d*EqwTn17aBDPkAtq*UC-eVLPqC~;sCQ5|o0ynftimm$Rp>g$M)i%1
z2&lY@xiT|5Eg~W=Zua~4?_CEE>cv=NfmEVedVuFoR|IZ4M}~gW<X|^Ymk02j7ZDRP
z0f3%@^&ZWxyC!;fQ6U>HP`pg0SK!yMP<N0ksog&?z^yWe8sJ)P?qtB?IVlqel2IcU
z7Cj|KY#4@_T;ce8bTrEY*#U%92689*03FL0sN>#Xcmx5^ji@N2*k7?Q8N}^F9Age_
z3DBv**a{Dnk_OEw_mL0-o;E^?3?jFn;D*ECI*R!DQ?SxB8wCa2Z=?7!&`MS(d~9u9
zM`fUdQ7AobnFd&YJF0Tr8%cL<O;+}<C#xGwzOJ4gI{Y3FI{?s(TOHEzcaaOn07O{^
ztL{ZFKyD2D0Z-<iSWJO(wF%$-{qu`Yj`GW{H)h*^Wfxw01{l4os=R;e>pen&hl)M}
zF0`1|&wUjiq@BIB5}_*Kdo2pVi$4Q8@~Km&R%q`e0cOsvzp~$A_O}i5blX<%Yu9K$
z<=HWVFE=aXLG1f)++jZho$VT4GYbnQMa8O-53PXkUwkRg1rdj(Q!zTO0xTcJ)MRrd
zDLXqmst9|Yh61#+ir32Ae6g6A7{q7W!6@kN{u!K+lVC3EpJYC|tx}{8=^QAulS=G3
z+lL^yg<-Hj=4|ey#5Pok9{3UvHTuCc%J<oKyrqjcE&#*acRcBX?--LXln!%#444o5
z4#_9Zv+kJg0#B)HaL^Y_aWd4w${~VU_<!T8QWVTJ7;%b{88h0HrIA`UguH<Cfb7Fy
zX=!;UDCma2|1$WMjbhp{;+t+RsELJpR6&89vfu&9!c$H91$GfmM_7VP?1f?!sUW0<
z1byc+FmiZg%GjBmJ!=TNO>@k?(eiBDxswqqAFR4d<?s+QDC`rS0?$SU?-EEW!GQz)
zJCLe{mxWtFMpKgwXYu{T2ViFFa3G>+iX3zV_(Gd32>uEi8yopz5P&`qvs(K!D~sO0
zAKQ#rc!JZ$hyos3|Kh4DW(@OCo+OO2clk!F!z5(X;Q%DDItj}lP{e-*1yAkwZ)i$i
zgO}^*=t#aiZY`n>ghrGpXlAsF4j%-1nNi4vfRijDPbfH84%{+@P0#jj-%g984l`O$
zTv*SZJ*xxF5eKB2#Inw+_p89CCist^9}Bt$0>H!kCUbW6Nm7TX_+oXD`6PlwFpxQd
zU&tGD#~Ei2tTkD191#GFdY+l^xzNp^dk=Q}6wZnJ+&jLQDWb8%JwUi4+`PQxwC}aK
zcmMuPRM)afN`8*&EB)bsh>wk_34g;C(oti-3mFzsX`BXfYG3jtumTDDprxfxTd#F-
z^9PzxH6LJr!W5_f3oK|xQwWMs|Gt5dR~w*@P#l(4urH@8!h$McGsfG#4F#aEoSYmN
zKYum&FIwj2+#0p$&DK-DFj)`*#W}~MYV972h8<knwiW;U`4dBc4H~C|gs8c>x%E^F
zTw8NUegyoV!456Mrq#suZo>ut{3+pvQkuARw7^8tLPvmWaDxsem;~#*7aVK?eRXH0
znW-u8AQAHGaS62HqR_(R3!T<Rd`uR$qG0ImO({*FXQ2fJlGxmvbTZG-<JqnKI)z)O
zjVLlGQwxi?<l|qyENbkRiAf-XJAw!wd_0fy<^kSYR(7@tK7=s<EpiE+tGKg`g#sU7
z)m{Va1Y=4QI{#28;h9-j&f>G1;+{96Fl_x^d9XkF(S5vPQ<%D=z}7^Ml{<^I=~v#9
ziAQ>=XREiP3%M|Yf`T3-CGqM&^ot-#@$rcX6LMf|XyrI~Fr3B^9NsfHjo`}nP%qyx
z7#*VT9Q=S;sE+_VH0JTh@f(Hkg>pda&q0tMN_gSZg|066p)$+?RTK9j)T<b``-8i8
zS3QKO=SS#1#wRDuHcMPKB*A;*l?N9{?Kdq(=0mamG*hXSel(ho09%TC<>H5s@42rE
z`CRrdvB*t9J_Tl?g^Zru1Ay{r452GD!--A%5fuL0ABhW6tTC6Ej$HY#<Zo>59Z3Ts
z7ht_QJP!e@8H5nu06z8FXq-A#gw?g4>ihb2S}*hN2M!5|Fd9Sbf>jnTL8)uOI3O-M
z`ZY$|RcyxJ(5Z0x(`f3-btK>4rS%$X`<AzNXUN;i%F{D!?m<)veWe10dLfv`5SClC
zRZ?#A@6a=^r!ZsQXY<0#ivjCq4fjJ##z}59&dT>V=IL8-swkST=<?;F&^%Wo5+pPv
zWUb<{V=FbG-rL9Eo^AaxMCS_9wHe$YH+hO<%3}<m80rVORONz%c9n(N?=Tu!SMdGY
zH#TbbO0^)vyLavwS`mYro<a!R>54$upsx%Gq96q@@wvXQA2<lerDXHiXvxUXiaXf+
zD8A6x(4dL4^u;2RtB3aq_z98n1;7v%xG@gZUE<>RLA&%$Oi0K;{e7|baK0p1bjCOh
zzd`OsYtq=*7z$4Oy}&>tXl@Ph^MQfI`SlEpjLboW(4f%)j~puJ-*~`5kZ$Xp3unJ6
z%6x()EOfC?RW3pkB1E0TGN(1c5#2__F|h6Mtl=T`f$MQ85gL#!uI_;m!yb5c#Kp&#
z!_)V{ty_wr$BIyuW#U(hprKp4YSs2<1wa^Bs38QpkQfOkM2~y~%%iHIVdmxK#X{kK
zF#5APjvuG6K%udY6LNBL8ooJ;aykP=NJ!ps5N3v&lh@64p1462Es&jPaQo=$)=%aE
zthi9cdzVV7aGL0FTd2mzc#jOnMLB366IYpVQrXX*L6Q7GnmK<vysM0;_e7%uG6hwL
zys8RxDp~-)?!OO(qkC>aG*c03dwg@0Se$9%d-j-8&_0f^882`>zy-jowg<-?jUhhw
zCt+DxSuO#AT5!lq^$T73M<p(Fx))`x<9t94t|Ldv>^Hlata8UJb=8NH=)>1jA)%pf
z*^JL&eX&sEqoXT$b(#h<wlhg9*C+-?_xv0iI|CB983hNH@<o;2wk{y?R#GTa@|a3T
ze@|HLzUTstM%nwq)|<8c%<o>6ye~`@=ijyKxe1B`?EPmJs-gWdO?v@cl~ICtf+3F|
z8w|hT!V9ZNA2Q>jZdKLR_Pd^q*ki5jV**ZC@t5A-(1?gFB%9U81n8LN0@8cgl}Yn0
zmIIb>3z(@4?lXb+nk2=<a0FnW4WYIWw)^fT?chuSpJ5!B!l?gqFpM1W4Q$$tLK{a9
zljx#EZ|n;$6l!(jw=;b?sat$s^YhI2oWUg))#hy_Lwo0lb80CD+CT8kFm(^u$QZL1
z;}*)GOq8z{&mj|S9ti9`I2pKi?PAr01Y5w?87DX0sL7^RcVSb|4bjFF%uO-vE#nuk
z5Hv=IgaH!hNGyIr7}0eBY^5k`>+dSY+#7n*y>oK*pn|?pIp^IYkQNh~h1WzmuOFhF
zvNw&o%fJTRu8Zfdd(admjzi280D{pjDDyo%Juh6od>Btg@l($B)_C)lG<-T)MaAUT
z7g2F6_B+nB`0>ups=X^PhPZI$N*Nx19c)I(RSm%|iia<iJ{xXA$pFKJ>p=0g^71Z0
zeL!GL3HxCdPR>}B+&_3(A~&8b`IBgOOVL4gq8G@J+`5&*;*rJ8cM$=v9<^^ve@r`X
z+JA)w1$j9Jg^`hw6gp=xgbi`B*EKbvRvD|^yF*4n43|5x@PS3r5(t*KNA=)*w*jAE
z(c@Zq_);KzQtNi$-7qe+c78%EoA#cH`HoJR8O1XZ0KloDGo;XIa#_*+A7jX@puw3F
zrIk!x<OdK-NQsTx2CIgE>%nGzU3e@I6w~9tOg7oRQGseX6B-&yZ56b0ya)mk<qr5)
zcxW2M6Y(m7wFbgmneb_D;G!e&BqUQvD;$Y0(|9(kgI_xZ0t&1uN}5xF>Y>~Y+Hl?1
z*SDxp@%M_KPy7p?cOOe^U=6={*S|2D{WGtVK2-!t0}!l5r+%?qjn;jPePaaeM{G+o
z>Ix4xjV}jiiU{@zn)+5eRJ<)R;zjq#fQt(7E3s_@8E4hLW%#1H=+A+szQ&*!+^ZX;
zio&ss!}vPNp38Q2e5`*OrTLFI=WdeUh$oBxqa7j$A+D;Z+G2oM4<zh(5sK!IIV*0!
z=yxu-aive5T#lmh(^F`*S3@iH8ZLWD!-k7_-oTPG1L{B6Z_%P5TBBXdMz?h;e*^}$
zWZXxR7~$i%1)aGRzy6pN>9ySjI(m6|QMh5iNnsU?ud2eiy^WiDF_z4A<U$`EX~vbt
z=&oY66r}p=*y}(+(a{@pJx^Z3Nm&Wut5UJG^Xfsf+`JlL_?+EP77$nxaLUe%wC$;k
z$r1(U&C2-o^`UpeThYB$juHrNOivnwXr+dy<Tx=Wpem}X*J$4l3)7C0z?&fG4~jf<
zOH0z$)YQ~~bkM8av1VuG1QMO5hKE!)xVX3wtPkBE8I_<}2ZJ>LJ_*|~EyIHg_LT)F
z1jHCPh8)g*gg5$O%+!TClq6T-)zX8~%x<gnw?+Y*X9+?+uA@b1J#!M+DtWn2&oV*v
zxu}wH(IQywv*T*)2|=~A1}y=wpo1AW-_ZWWH_gRjfCLuMDK)jj_>NFYQ50Akcr6U%
zWkPu{>(0TMeH^rFlx9oucU7Ehg3I5(XIq#ZUyj*GJ>;_<40Kr0<ya@h{p0mGi7A5(
zhS*Xouw=cxy%R!0RANkV{Fb9Iib_q@>zl*m&mZ|I+}zxexIr)ulr=Xu_c39Yas~Qh
zY6y^*4Qb>UB~sqM^z(rn;J6CXEW|m&!ore)woq7S?fg+xIi*-sn4RB(`%E{08kx4i
z=-~i4Jr70IrirExAGFQeq15HuY*4?^8H><o2&tPzm>534XzbA|r+4!fs4z>>NjwkH
zCTJN7254(9f}<vS{<Cf*!R9bKOiW#VYeu<TfEVx0!@(D|thVQ1!ndm=KMy_u0RU9N
zO*F>*7;2qz^prP#DF+Uz{_NVeXU|%!F_}Vrx2aJY5M51NGuC*b_C}pY*i~#BHjn`}
zI7Cy!dw~nqL(Vm9H}>TiMun#aA~^rtU8HA2EkPCR_)cw=ltgUTu<m+sad9+No#4*&
z!9Tbkg|S|wE>3A;;Z3<?m!i>G3Tpv4*sP|WcbS0T<xaL7isfr?i~?=u0(9{JzKo&b
z)OEpH!!?7m$TIQZDV&T<kgEWVB;6e}++KiPi1QV8HqncsE%G8Xj8ub)^sku4XhQut
zIRbxiItrR2FOYTVrW!I{zkQo1(@UsBomBxd=xER*_HHnz4Dt+4|6EUuYQYTzVViX1
zWKMJH^l6fVbP={qxR|kvy{&XfNboDyxeXW=01@T;;Sd}YBnTB(<zvAyMMWj05^Qgy
zbLVtTU%;~<5?vh_zol5!<PD>4SyhE@n0D%J#%}#1L4b>-J~-otSg<H#hZo`#T2?MT
z7R+t6Y6Tu9njFmxX~V1f`dh5v21Hi`tg?*1wZ7g6?Kdr5h9f+rDO!<u;a?^Un}CHg
zxF1Jpt)pLmOd8#2E!z1|bT*`w(p<M;1FvY*;rbm|YO8i_+jc1O70?R0B8=op_;sND
zqbmY#$op&rBQ8O_*ef8NqpZ271fM&Vgp`DGSg3F?xZ4eX(u#D3SA3A@kE8W@q!%q?
zBW>S0;~#C_BErI&019{yFnM$Ldu9rV2#-4)Fog=7pQKFp{BZaNq~epXa(VzjNFR|w
zQ!}$Hpq+wP!8Wu(Pt#7;OQ4Xpal-#=!4Nab;|6ZgP@oBoEiGwko6rg?L02{4YT}oP
zTd3}wBUy?9Cs6bmHu&Auhb+-d){BKF_Y+E>=Z?1|@6Ak4Gkp^f6AMPbY6kgB07yt9
zitYLyhsk>sPm3xc#R+;@i^z~my(p_mmq@C4l<Z3S`Vo)|MQny_>Slzj%0XYQ?Yo9z
zqoS0%Y{91}(|D9nD-GiN$6SFc>#G071vt+>M9hsF8qVM)L0fkl0;q=wYF^yb4%&?E
z&>lg-si=M!U?4}GWlFf}s+RBA5;#%Izpx7t!-E@Xj!k5`=MEy%T5diX)KrxK*fp_5
z-h*M<1`0Rnh)L_J@aTIpY_4bw0biWMhWQ39%Qt)tB?NcyaB~Y{qvhc4>ihQX9Rw2Z
ziDfYrm!4irk|_vDt839#<AoE2?XaoJ!Qlb2o#Y-22*9eU0{^ZJd+5beIVLXT#Kong
zyj#0Vv*^mV_r*j}kG4(~Slk0i^gSrfLWQUlv=ishWADNaL45sn8bd(pU6mohhN~u~
zU<??Fj`ZH$yJvvBzeD)9(62eBqVCps+^M@@6+zFy@DzhB+x}O7D(@=L-5#JaB+rmg
z^Tu$Di{Q=0rKVPMzClZcp3eyTMM+op(Q(N6zkhmhZ~FId;@;2Tjy^m|rnUbg`Esrm
zxRu06Tl^8Z^^;z6*1{S=dr6gz;^u*e2hBq?7^u-#>tvZOo!l0SmaR9!jo^o~;_$<1
z+!gobKI{7RZ!wNb-p?Y#YXT#Rcg4j~d2bR{g4oI&6ilH*TUB0NeS7`JjUt$=qPi|8
zyih#M-D<y3CY(k0q;%wn7n}_rTUE2dVFIGpQdGjBD6vR(^9-tGHe)|1<b*Lui%U#=
zi?d!%ff6pmG=VJ(DK=0{kZ=0+>(@+lA%A*XTREmUNX;@JFazo8gAjg_{AIMOWmt4K
z22Y$kX~w*EmmvU(Gax|t)%O3{XP7F4ptBC<<15}V_(P=ZJ=*oqygUhjQf46GnE0t5
zrxnBm1_+6NVsf$w9zO9U#vr8)o`67MtgZE;F6q~gpD{Blr!wGdvZR@~GBdb8l*|`y
z?MLzduD(9->)@a;n3YKWDkBX%xP*%;eZr`-ewl#a1!4XamzGvTdU(_Z1P5YfVoJQc
z$xrAC;L;XcHgQj$ydnP>PbdpkmiSkULBWW9bn(xdxu5Sk7lwNK7o45ap-nk_7EESC
zSZz0NX4tl!4}$~zm9yYerVpR8T(rg5ApX;~7_a@vZU8z+a)5S0IGXn+Q5jPa0i14x
z@L7U+Xc8wwo+PPVwsFGS>ov&4U7ek$+M`x%zl$q|RDeK|t^;ScyD95^6j+T1O0(DP
z?VG#1Md=yBX=GeiVjw^qU2$`-fXOnJgH1aKvNRMb<&=m{f+orwO+?9!cw+T-qZQkq
zf}&&ydjayvgQHO|;GL@3vxzZ4>NPEg>H8sd^)=WGN3h$;><SG<C(y0E5dGs|fp+)7
z;Ieyv$EEn#Vl3pxJmG1#1uw7+cLls4eK5ZP@6QM#M$KpDE3xpB_a7$D@z0Mn3I1=9
z*}vwXF$_4an>W*LOHg{b^6xlv&&kURcf%riKYp9bSY+sUq%0sQ?4Lhgaq@7-NJYBu
zu%1jof@(X>zdsZHjO!!zJq%k-+WuR-w(0*Dm6}xXzolpbzJFqY{(j3@HK~I%55Pf%
znKzJ%jT5(TM<qqm8gp<wJKwPdtTtjK?RMp;Ixi|+z)}u#zr}Ku{t<}*3=%tX`cn7S
zKffy`?&`w)!L|IlFpbncq$*3dbg)&vb+;aA62QG>4tAUFrU%=_sQxLZXj5}kdi+h?
zR9koLN_jnlv#|(60i*VUgQ8-hk+F$jPfL*5r;N!Ev^5^J0^_~HM%Rk=0G4NyJjVtl
z91+|gkjDM`b-H1$9p0|jj;2lk_H7d1;{}xvir6LRcMgh-9-K5bG2T(=k4^fz;WaJ9
zL^u@GEnlGaC_X=7`OKny!-l~@BA7!(PMli;e@!Af!MJ3!i%4!PlakCor>utv0|cYg
zPu6_4QmBm1Knrr}w2Dm6k=uuTzss*&2;EQi1sfwHj5C%2UN>*d;TIJZy%QYjGd1Nv
zcm_xVE`=A4pi7zG<MrCXFV<|l)hj89zJw8wJ4NBh-g{-%K}`NXkCW?|lG5Q^$k^xS
zoWRqLyr0V7(Ac=racl?I@kibWVD}J+Arek&t6QL&pv9qc`}zG=Yr1|H5Siz;)57pS
z61?*1x$xv$wl@RHPgX!{$HBn?v6Bp}rqEM(HctI;2n)Kg5B2FZ`i8B1`(KvifxUYJ
z4ED(Bi5W)f+ow#DMQ{^x4dBsG2^{}){Ake+<MIo?-mQPM-))PjscFa6zN6N}Ht?0(
zf_{wkVHbviN!Ju5zb~zK9zJA75H;C#C>cy}1@0=zh>|6vA9A>;#vhlxa(XQooJ(H*
zA;?pxId<DW)=1YacBi2unVpBOT&Zhmka2cCKpb(<a^H|YEUm9!(Dc3PB|F5>K1As?
zF}Q}!xVLOPYxMncU#^23vYj1F_QR1_yky+4*2w{;0rVj7(V7tYKCs3fbTyipwwzqf
z<IF9!LHF*_p{fha)#>SH#E_rJ2=R`kwhI3}3y!KoF<k!k&I$<N8T^gQZL4sZ67PRP
z;m5ToH=fSqOO`8o(Dykm`^kJEH~>_FbsIOLBtPIfMymJN;qj6mJ)}p1Y4HW#U4Ape
zfnX>vYiS`?X*MM#k+4}dDT}}i1m_}0?;FNX-aZ~Up>RhTet0efy!zsg22KwNJP5(o
z_RW}Gl@-j*@Yxj{{eDq+$0A^A@FkVUoAuwa-wQ}&(93Y{+M>zN0l~KeXE*$Bd18Rd
z>Z_nlxo6#54Hv3w#~+GcFYlqDFmtQ_T)71I>2sI<WtjLVv>Tw2+@+mUf-&Gqs2U*V
zupOE`on8hvy3WBqFEnfSpSxV0G=3evABNROfI?(kY~<2>c8+Pynl+!?Mk7`FZyFrg
zuA(ND$~=lkh|w-xtx1(2JW->*uv1_(xkJ28C|S(wcCDyLm^U@`a`f^da7vs~gr672
zf5+9=R3o+<fCFS<JJk#T27O~#Dsxa)76QJ=KoBPWCE!D)#6T~}4*cQG;dSdESu=tl
z>C<mA(kH|<WKfzXB+DwQqOH180vqi`gY}Kgo;PnYQjx|tmqBs2(r&`i$9N>1g~<b_
z2>(NI_&hDicl>q3KWER+iJx^@^2<SBV%~fbZ3*rTI%|IZ0nqBC7v`ToyhA~Q0l6>H
zyqV}<zuQgdNB?g^JKcl`+{MP&x`KOzuR-9eonuaeD*mkVy`R;!weqMVB{X%&#~s9x
zVr1ig^-}+KPENVVbIK$z$+Kgy*}UZ$Pm`K@eMu8*va<_~^V<eEG%nQ+3nSqiWdR#;
zLrN|#)u<mj@%LVX^Gd^_WyWQOP2vW7@q+3nhSt3VW=%WQFsB400!SaN&vFV&>E^K~
zM+nL7qxkvCPiD&EZwVPr^Z=%>MJa7r&;UYo)@f^O(9?hOquavLg>0)AEg}LDQp8sb
zfhfCzq&GojMs;FP)Z230G8Z2Nq4H}A7v}ea<9w>?J2@!&T~p5CpisuF=P*diKTe$J
z4QXExJNC-3OJ+Cpr%6dkz$lNHnsUjUc*LFMj}cfWDk*S1+@;5sQ}{b9K*PRu<ffN|
zIIVfXF4f;_RNq~?cg<P`7neEPfUjo1dGy>LAGmrAiFf8!R&UzcUc7rsGT?Eyz@L5#
zX^&B>F^rnb<^d=IYw$WemXYz`0ruLt56@Q?@M&~^e<;3KvoIogCy+-hBf~RiXf?#e
z6A}}<pznEKV7(C(O+uQ(Oc7gX)hfz8vPc>m)(FVqr`u#6o2HisWzurorSfwOOCgK)
zM+;#beTx|o6`_++MKyXaGb&rxq462jD%b?~VnOSMn369sC9R9hF{eVENu-0jrX?7<
zb?xnc4t-pgnQL=VTnXE5pz)<RSjJrdixI-tm0mb61C**hQ#D?oTv<jX`+(d>2PdZ<
zSzF=0>c`7L?a=k}yMYdOxZ~x`2M@(x+=O54vYdR!C=8##huU#@;^zhEjZ(%YN`tHJ
zuf!Kaw@gF*e-Y~W$0~+M$JywSBfo9mzERCrY_O654d@NtS|(oVz`#H!&cXL+EC8}*
z6uxKVP)^(bx8$N0<$ZAF6XIc_49}fo>8zT(+eyc$z`wvFz7Cc9DK!<|OOMkF{MUa*
zR~BU_XMCl+q$Eg79^7OitwW4)B#`J&V|}Kkdu@N3`+NlJik3oqSDEs#$ra`t74P01
zHaxWgvr@iS;R!e!Km9oyB^0bICi%<QY|2UicBYq<XggV1f#^NdO0R9}HiWW-qLP1o
z_iQ2LaY6zG^B#910XRJ-u;z50nR}prRvUQWn)CQB1^7dPiiEqyigf~R_c0KD1J1*u
zd2r1R046DZc5Jo7%+!S{fUth3lZ^1D9`p<^gBB|$F79=8^BOEpIpk7FpE-wodz`3V
zGmrK`2*%Xe@kJE}U)m8lrk@k@`z5zPNM4RP*K>oNlp);U%+H>=h9Yh^jro6z@a$+@
zXMUyD39aR9g3fO1odI+WCF5qboxQ!p3kK=N6a{(Rc5(5^g>UWlnyjs@MLz@-?#Erz
zJjLqN%H8r_TDinmlOrthjOedbif4|U%iMPQno#B$wG{4Ov<XvhHaH1n8XZ!L@m?}M
zziqz&->v8~OcAx?Q#(GtxHMVw?cH;yU!O099yD79cV6Y0Q+e=_UcP1aX|;Gn+cVL*
z=MG)8GjHBBTP#7N{;fSXAfxUSLa5|kjP=s6AypUi8E8h5Oei|rzO%>?BHCEYba4!v
z#jH;h<HVxo<mT3|zrW`~=F%hLA_3LVV0DD!Kw6)R$9<o`!mmv}kdcAxmd0Re)s%1u
zXDaO7I&p2mt)OsXX7}GZm`s9*eH+@qUtXSFU%!T$#(3{W&D@Ws0s;&0>3@I73$xHu
zR#vvWpe!r0eY7J;`R3R42<`<ujwtc2+b)d{#O-&Fj=j7wWlL}Uz9=u-Dpc6cGHh%4
zjKf1GO5g3EYl5WjLL33Q;4^1)p&G<AhtS3eyR^931mnjw0|#badp2KDF|gO_a-8K>
zbsy<?$>^?Ed~JP_fne|L=PpC1!>kjpU!Xv<61DlnCZ`O2FCB?$$6-_duKG4kds}$n
zPXuJ&UJAs61f34D8Xq4`ynRzX>~KJ;b}oh|nl9~NMImWGm9P`DvU+}{>PW~>nV=3*
z4I8ja3|ezWh25<p;vncJh?;Op&C_qsOT2aaRX|NdgoSNT5U<uMcb(kxbQd5<BJvvq
zMm)RbMu7@#NM6d(^|zl(nq&rU`TSX1_7aU_%#Q2rDaLjQIp056%;}9AAuLfvn>`P=
z`H`;Io>zZ;3zZ0;vhJ||dkE+^N6Rsfy?=kIx}N03KQ*u4-WRioN58-6?Y+z6W82+4
zXtn9qHmE)^Jx-lwWQc2v<GF<~uKfaLOHVNvgf-_|gUBKTH(_ex4*nJ~nm4PAr2Wx*
z?BMPC{dA4HMFy=$JX-Yv*pmXf3;BTfx^LQNJMby(n8&E4<x|n`G3ROnYp!uv?iUiW
z)H?lQi&W)e>FUd^7{;+Qw>C8fi)ys4n3mxQ{5-TU&s-E5G)n^p*qH{aEe8O`v~IA}
zP0<tzpO~MYIge(j0tcVnJjWwk0*5L^yK`f$29lxzSGsS|bjD-=AYmNRiP&P#HCBZ1
zZ){G|aO*iRDPF&3ZjWOud)Ow&yj3ed!mJQ>JM(j&rpGBn;p{oyK1;vE5Kib|cQ%~k
z8?C&#uDUuLbOBL2u|n?rx}PJP|MfIQQBfR%>+SQ^>k-<-jZ3E=eutMw5Ow1Mz1j`?
zfGs2i*lwuR%4PS7lf*RZp?CmVZLD_%lyVhm>9r&ULo){snYOR?l-B;VXsl~(t*v|Q
z;~On#^m<X;iO_2x>(Nx*FTCQhDlmH~EMy@b@2h!u-!!A_rQ<`VtsJdAHwT_nY-o77
zMRebcuw#CV&Pn<GTD@4}G)1#K98I4^Z|L=uO$-T4HvUiyNi!-b<MBwkf#;gZb0`ny
z6w2wYbC(VJo1=d1v-|S3E`?SVCSN$0<v|+~I`&m|(x$!XY|#iW$JLo`RsV9)@T0kD
zu7&*9S$=Zt=Il;;`{RCK-q+L^M8_|KY#41-(-O<Lbw{g@x=-l+`$^@+zB}!4pW~Rs
zv-7p2IDnZhARwz|byiPITZBx;&SHpBdN)=vGbg8}ZV1wYOxHIIP(P1OOq60U<c^?&
zjM<Swi-zYya7$}=0jPhz3e#1nOPqQJ3PE>uygX8aCIJ?CJ6G>KUU^EJD^ScDg$0=J
z!ge2?DTS6z<xEN+Nh^OhjK_sO!TqQFRB4Rg`^F~4-0&c~#VP@#ir+Ki!h;;OzT)Yf
zoZld3br&mw7~0}W>R^-pfH)LtLgnt>xDas0Oj#dsjo#zn6pVixZ**^GtrP$92rR_(
zbO96?=RQC6B<@SrwFBN|lO{p71*=qPAEi4Yb5j1Ryg$pbM0!ER;KswPt3Hvgn~g2z
zsz8UlY5vIHwuplJrwB$On4XkjiDc8dhm}sZr0CDpcw{460Vi$^b{|nPz%Y}<`Vb3#
zOpp|Ucc)$|Z1(Wr_LaF6!f)_#9<7Xd{i}VT>vq|@xglzP>h#=dEk*E|u&3i--Sk|{
zIw{yzVf6u$d7XQ{eY?i+r}ML)=l}DtBw!A99Y9#vn-TS}v7e?_EX^4f>Ai4N*W0@z
z|H{)36I05x?)V(2|3ZN{5jO~sZ3AuEff6F)!5g-o!`2^S+FZeyK|D{CdR(Z2c1baf
zyD(4ViRE!A!(&OP)XmJ7B$y6bWANIzXo1WdfF*r?f5fx`*$cojF@y~Viy*bD-Z%Qy
zMZ1i5DiXbP6>nAJ8_vB1aTqv%JRSo))s*8979O#;xJ|^2JM^yNsUA#aae~Lb<(MiE
z9&%at=K&k|<DTg5>heH6N`x|5dLpU2B-ME%AtWSVrq{$!sD$+gr`{VFQ$@oi4>6xd
z_?bUs&`4XEbTk2FUH0nT&Ir<BG~#Ngs5p9UkB!p!W(DV~CpJmd5C|KQ@+}h~VfK}3
zlkUA$FsIYXv(Ep1@J{`ZkjM=8(+I$^#6)8^*koK{ZNovOl6I-X54ACxq*=UlX)LNO
z?T@I;_Oh)cc}6%jcATw^*kyOnHAqgHikq1-MC{|24p(EiWD(Pq<KX};K0JSk5Qd6(
zT3MHuwd;<SvkywW9<}|cGa7dsKOCONQ?COGzU`NaV(h+sh%s!LM~@wQ#&ZemB?>ZY
zv9(A>Ey|}m4<00hNFg!FgN`9#&MD0DY=o=Ah27h>b>04M7KGU`*hU0HK(lXUt@G4D
zDA(rmyJxMAn?W@}Y~Q=Lt9LGL%aXVHIO6Jxcq-lMHwjFiYd``XnxlO%GP3aW#86A<
zhr)M`9e<2+O76dDX$iS8<$(Gi^GfxGRBP0OQ}NDMQjuP-J&e+0s5O=Ja_g7O%N;pl
zJ66I3cbKN9!10y0y1D+I9Ew+;1bGhOzqc^zX#{&$)H>hc$2EW+cVVd8I{&==i4*dw
z3*JHS!a22*;x>q)<<m%Qn3f+uXn{z$6y0oJoIx}KbS^tjKI*)zP+)gEH=j}1@nC@>
zd+GfrL%AKag0PfKT_1|C`Xyj`97k0g#0i$uXmR#l?OwDHSHeT#yDNG6u4!XqEx~~C
zN%MTr+j~9%0p*xlRyely$Scmnleg9@a%;i&55Z*Q1zflwO_lNEfvaHM_77wlKa#i#
zAbz!JZRquzC@l7X8u^@kji>y>J@EV<n=NIAJhBA4dZ0P^ewcYvL&M##l9E4%n{NT!
zhTZ2qnP0z%SFxwtoE&f=ue-gbRUyO_@+mkpc@B=<^*Q`xUiG%Sv?*2RJ{(h5u2@l2
z!nFFr`9?1|DJDEwvD1$|z^XA0SeD9Oll5OG2{9CP{nz^Z{rQfQaff|YZ8~Us1KYFV
zse1CH{-PaEupB$lD}yjdx&+pYtRC>g&f)OID`^xm$^L?AUtQUo>te|ocr|<n4je!Q
zwBPB<h$&pCV8ZP4S!(u9qDxiEq~4!S+F8rACQNRly*+LKjDsVcJ(uh#DbE5I6x-z}
z;(tbm+g3Y-cAKIH#MwED5h2(ePqM0U0jMUba8i9Wp>IHGT`%r>U`~bBm85Qg@(n&s
z5Y(Z{2(c%I_6V19`2CYdF=SVL=*bxzv)D)9vR@<+&J@}T5>$p7RYRVoz_PQ*19wt>
z^NUprdZ0ze;nI@GMGgp!JFj}OWvnVp`8r8TO;b})#eRGD+(O3BeU8reQIMu3BygM7
zMx@TqjY6A!j%7(V2;ey0Lxs+QArD=I`S;-(6tgepM+!x4FK+AmVT^W^_+lbnTUFx+
zJt5Y7|Cmjpzfu&kyMF`axd51<mFu{9Yr(<ku|7^j8o+RbMD{pN3@?54@n9B+vvAx`
zoS#D_j){|b`%diBcr5RcD-ECD$XCz*Zq_hZ>DozoB%jW_(k=mO`lC&AGq`k&F<?R8
zGw_Z0t%ldkt&uZ=NZpTJ+#dw0OAGTZ2nwo2Vtq;HHrZ!`n9Btc4|1zMC1~1U=Un=s
zi~&3^z?pZtzQ*5+SHi&S#_{6f8^{`Ye!-|<Zg_ve!n_kexN@**$Y7u?%Pb!1di2n%
zajA95-h6I5!HY__Du^;fH?O&NMG>}8=zg<h80$2pY6Ms#<`sCT^kmZ8&{W)C-#+9&
zL>ai6jqNBXwD3YbQ&AVi!X_xq7vBNj*+Y2uSVD#acede9L`}SYkMmw)Ico25XEIQ0
z<mI1{!IW)Nn%=_&xx<1OMB)r0Esg7yX6Kq_=S06(E;`}gpG<E`Q$#{V6BT)6uyzRO
zym9;QiP=3Cs{nqRf@|;~G*mfP|KJf8_7iQky_a(~Lca4#vVIP&g2(+tduOwEM|1T-
z0Ng=w3-6f7p2?YX*A3Wy5SWBSI-xWrSjmE`E5DSYC9dK)z@DD_s$v?K(sl~z*k1Ef
z9Dl45CDO|j$oU#^X9WhLN9(+nLNGw4Vs-3pXJPX?l70FRjBdl~zo4wpF1Ij9sN2az
z4f3Sd)V|vxv|f|D+qx-1KRnB<c5JJJ_UTkdxtqVf<I6BJ&QBIz?vG{u+bBqe<21Sa
z>t~E=rQiB(^L%(D^@x24f6?hV^$8&_fe*!P1Z!(YRPB4i&DZX)W?|=NjC7D9I6P*O
zMQU$O)7W(6$sIkq7=CvaYd`)omvMFkvepaErDcunws}P&CcKw59zX20;{0)CR>*sK
zThA|9uQ183Tp&YFF4E=sPL~Ec4-rH9?-fUJ@*M{Asxj{50{XYV4r?pU_hUmTMF?}Y
ztI5karvYTe!Q2OXCMi3O8xB^1x#t^Gv_!(0d0l+YE*W*J!1f&tl?>0H`L)t+K@ydP
zT8f$vnd&1n<77Lr*#2iq6?jbe5`KV-;{Nylw5{a6jP3sji4;Ce``;mvXN}m;E%{eE
zB{|Bv8}F_BpZ~%tiuIpAfPcK5`5qPduRn<YuJu1NBP(8GwfyI=A70BvBmLv8^TOOD
zMr$iqtr|;wx&7b&Sl%FT=3iWZK$ByX;63#uH710=0I&X8V~rX}2u{+VErt6WoY@Py
ziB^I0Rj)yi-a=4o-2-Q2`6s@=UtdxR$cZt?%fwI^fUB{E#WCzGHe`2_^*Tx1zd^h~
zmgGH@{gom7Rd}aYi8Fs5n1Y6gEJG``7PlB|OcgPmN7(o1O-a!o^dy)BcUK%rHPCSH
zh?t#t1ZqA~B5QuH&)<T)iMqx{r|*wpbnAA5gSDGePFWMNe|CdSMCO8)SsC3Q_}M1`
zo`?Mg^`#3@VcU?y0_xCRDL2<V`;qy##s?gIqq!eHW7JZDvn!cdq9cz%BRJATj`a(+
zwuEH_(_XQAPE$spxGe|S1O3|1e?80+vmu<DU;(a1QWOa1gm-3O@EElQu1aZ`MiU)5
zNH~L%xG#zUyW)#tpn4bsRroUQ1o9t=p>~RPwnFk8u&WH{9ALkzNZ8vb*a7wL<l*tD
ztE*f1WWV&s^j-{zF+_$n$qh(_mGMr%9Vo?`zKXjE=))6KXqI!iFScl;XcmLbbGg6<
z6B{5gBtB|%bTq&=MTU4<1D+zOE#Mu1=JO5A1OWf!aREf|g({4KLV-X!fTO#S)(8vM
zy>^4da{P(32QA1`chKUVytLJ~LJhvOYE&<b0Y*NRL2kz?y$@P0u3t*Zs78ge^EFUb
z$-E8ZxY^y;_v?^>{GSWkdm#_t9glkhDZD;nNE1U<3KY&phd?pM2YWdZqy>-MHQ5Jv
zc#d7Pvr{MjwiBDrqAFeziTGg6<nO4N`S8gVFw2%KQuYy$!@<oejWzn>^fc}YR&aGV
z!M~_PUP#l}ySyzkaK{D8#0NFu)rIRTw$s7#Ern8miY@yLFI+?vl250yO8-D?R25cW
zGF6FQ>e&a7)N*LKF+iU;cX10=qM2U|x58VfXYpK2=Vzx8=(B~a+y2^<u$`I4u(CN1
z%w1x7M#;;Ui|!jkC3Nx9r8kJe1pQ|TE4MU6&JD9NWB6B}fccKbv<8*OTbMv2IYKR$
z&-XQMNRnWRu@_A8RHF|wks%hkuV)_|fl{DCjRW%^@)y{4pFsOVQp<37Q{Y5Kr>CF9
z&<V)|*zdsc=UOAy>FuMigaY6x3aSIkW(ihQ-0;5Rfd@BmQBFxo{p8zdlrW<|gyj$R
zb}7_x5Hp2er9X+ZGJM*K$5K)tt$(QuzJ;4Nc75s;Vti3%BJqbGd#V@%J?scFNyApb
zcR>=P2qcb##Q-5TD#k?&3~^Vhcb<$O3L~7&w_y;4StYjA8gPFw4AggW{c|iGZn48l
zr@|kHVxchuP1I<93LzOR#NVPo<WY(7){a1jjk5@@@=akLUxtPhk;8#>jccbr;{9)j
zToVO?Sznj8z|k{jq9t5%v8uH{CY?@IzzTLi#|pNq<yBMN0A$0Ecpv~U<EU*+YKj;C
zU;vzmT@34TbFdbgZE+MLVUD&4)dCF>yW^GjC8rmFN}-n=4&Ehb7t&A&;iQ>%V7%iy
zzw>IdfL~P9HVUGhuKBqc63+ak>V5#}abaH$K))1CI8H2MlI03M5KcB>G{2}TkDwi<
zA*@r_?1yh~u0o;!6umuIBW5@Th<!OqZS#CnkX=ze#F3&ISq$IrVQ>KYOeFD@4;7}z
zKpA$JgjWflzD*LBy%E3{&AkK~8Q?`Xa7-k=d5Uay1a%0*ObMA&k8zk2hksCs!Banr
zehC27On8kt77>mlPMnlP!h@2zv%kL_dOqmxi#s|DQahlgA%PP3aO>fng86+fCLQpF
zBS{kzag8i*!~_Kir7&o>UKlu#!f0Da3JX>T>cW7Q7c=;37DNHYxQzT+XdtgcryLAA
zfO5>ymDnalfF#NLg1uHWPEy2r_~DGB6S%amB~kSzkOF|_r2$g&M{&W!VB8QDH}atF
zqv&}xvG1=-@PZ4n(C9mHA6ew+6*y}pFDpX5Ph;Nplyh|K<|r^A%!uBG9Cy+!2j`41
z-7y})r9ra-0WjF&`)t2p1|cxM{HU`KxU>z`<BSBT{W&r5HdJKioqer6V6A_JG6mya
z@NT&-Ej#SZf;2+}{;2lnkU8F+ohP?Yc<%Jl=s_Bq5HPYIcWf8F2uS!!b5Z?IexEg7
zp2e-FO6(JG8lkev_BzDp8x3HB$9)XEtOOJAa!-oTqr1`;f~h)UTVi%_aam>)-wRwb
z28pEOUCYWEL-}G&+rsEI^*3olD8dr6_$`IFPx@0};>B_MMq@o3p$URsHzsjE)HkK2
zrE(-;G@9Gj7Lm>L6{D=3xbb-1k@~kKI|aq^l1P#Wk!%Bn<{3-NB=j_iCOBP|L_V`>
zJ#h><v{lF#jl*rqi)UbY6?Z->b`2-e?~>aAa&Q7h@=Z)k97BA$oHA=KcG8ka3irRx
zvq7uR1~4*9wUv@86m1lRDN7<%i3iiT$q5IkU5>z?1EGM0oXo^49xI0rjq_==0`a=}
zj_O)@mlN-z9*Bhni~>Z8+Q`xy^Tj!63m`leIvnyeXu0tYk*)PL8vd#VsV_~dH_)F#
z9h^Yq#fU>cCDjoEphgQvNBrKiKj`b6<%eKJt{t+z^$FbaYb=6goOw~&sG=2$1tOxi
z4@sirY2hi;pTd@mx%3#-61O{$@#ugyb#g&?qX$K|`tr0EP8{kvd||96mm<Up5tlH+
zfJna@w@DLn3%^7GnZr#=1o_m22CKZhyxrk2FVo=9Sfy6kP{nNCAa3LLS8KTFlv~uk
zt7AJbxNWjDLxc_jv6jNk*?!;V^xp5^Qzz`<{sf7w0~$f7D=5pmJ$RFrbCzcTA7JXM
zm3f995D7ewJ>Y)<3;Q*QOiyga$$FLKcY?Y?;(o1kAOo>zi19hC1vHlif9874Sks$N
zpojaTCO}<v>{0}Hh;)$=W`8QbA13vUjga?0V0WGJFnZ&qjA&AX(p|$?2oh~$sA4Qt
zt~%^nT3u8`!5$C0h-8sM3kiO+5AM5DnA?fz`J1X&;;a|3`Lx$@au^aPE0ULl%bVCk
zK_ot)x+$mPH9~++;5NdL<V|_GJilAV&=0I9rAK>Ll8d&H*K>*$#XQN~h2ygszK*4C
zJOYaeHVnon@`9qWYjJrt1%3)svt)lrUe#rO^)#l8gKqP)*)x`XK{%;Hz+8;e89{SD
zII8N7e5#m_A(&|7eK#Mw4M*EL(Dh;2A_IOUtkHO2ix!D*jcHe5PVEQ>N58_oNkKas
z+1vQ}<OMeb4p4Ty6Un0DLr@}RSE`;2ZNWXpC><6VAE(=xi*OGBmkgsH*0`$l+x^jl
zuQ2cjLzM)fqSSd~-{C?$aB=yP(;S@$<|iefRW#LxV^#Lx>(3&y4(TJutmAPfTt<71
zYHSotO%iQ_1;FrXHU)K|1q`s_Th7-<(?cd<3Z^~aySVj{TT5PQ7P=n9sfWru3*aKD
zvTU3>hCscVhBMQ#&k8@Z9N^CgQtug*mL!<#vD>^>yQ!gJ5xRahxMksTAz>!K+b;mF
zMSb82_!g88)kc4pbE~LX@6Y|mTMm2Q1g|A4>r&o?C`Ubs-wn3W0gh6PL;$n>iSkVd
z^Ii-d`C0+}cX)J6ix;D7zP$k^G2wtc^V^HhN*}2S?%RtQnXsCG-Rqhr(c=Bp6*5cd
z=&oP>=-i&ulpZR6UVrUw-3K=L_HXa~?SbZJ5W2J&LRx_6vp*Y8TE0MP6P6`OMu9{s
ziMbk-U}O%{6-@U%6tzakMHV6np>r4jQ5y9Lg+thaUZ=BnaWFo`%C&3DaD6w-1?qbH
zonrOGLJ`J-C8}nES~-_HKORz6X6>s9zjgceZHZe7-Y0u;XhDSU3$Qs0qcdm#!{Tbt
zO91#V0Y+|Fa~hi##FguKWT;c-&R(RLsCjwtX(>Z1tB0oAgD-Q`kql_39MFn=_5`~D
z=Pprx0D!5$HUX`l;5=}(8NBY#pcK%@eiye|xW|d5)lELYnuk86F}1`-ocU28+CPQW
zGBW4-hu!F%Lf8`|H#WAdue<vij_A|A8yFcG=jSF3!kM|?yi1$Frzo*^FBS9h`Rn8f
zPY3iwuZoalhxJv5wS^~J*Lrf}8HdM1KZ-o)a3Tlep3A?u!fOS)3XHTMiNM#$z{W#9
zcytz)*cPM`@}Y|D10jxp%ZMyDL{_roh^oNoPoZPWiWU}E3v5HzA(3lkyWldAgTvyV
zP{Ud1lL1>`qS(h$c*V5glPFXYC$p?^QlRJs`Y8(_+I``%@GNh2ka@LQ=R4xK1}-8_
z!bgaPss4o_n3(*nztGJXRGt*@>IbGeU@KtMto;!$kxm*%aoBHs%|-K4$B!esLQqJk
z3%eBr-~7yi73cX7`GW_R2{jGyxoW^_ml+m1=hEb>?W@rbo1Gc)bl6XNIlK@Y2$yhp
zUcT<O0f#P1bro=Gh#&~#101y^ac#St?|_o=FCl~kio{&KU@6H;K{T@ojDB#+pwD`C
zdJbhfnXLi+6mBZu8ax0EuxO(`uGC{l>3sk@B8u>La4K+t-FSSzZ?yS@hDM=O42mck
zGE_-*JBk+bI&w>4Jx~g}5P0LL035;@{|4vIvk%Y7AfzSr4TQVcZ6=lrqW$~q{MX*z
zChKKrZC#DpSY$%%FW&BUoy%+jEVG7lI?$`&OIg+}??lpaX<gkNz|!V(h6^Z-TxY(E
zcbZ6w0cs_dxnxy9k7S+=CbZ-HbU!=q>6D}BhlS^!m)tjMh&hJFKp107Y=Q9)c_Wt&
zS~8>p-Q^jV5Gc_h#UIj0i}B%bVOsX)HpPNF0LbwYq&KN!D>;sgqN)ILIJ)iSjJ7cS
zl!zK07KJYYke#12P`K64-~l*Kw%ZetC>Ulcz!ZT9dqX;by5i?htAOlxXmyYyW)Inb
zG|q;v;DBQ5V=ZK0$0FrDg&`dQVz2T0%#jy_9jGq<7RRA4Dn`I)@qK5wPeoD~;E`(D
zz;)Ual{QT7?kvCHMWk$~wy)!6Q$<!Xe1LH0GvU;Ot|<g}U&5PrbF;JNb&=$28X@95
zXqVPqkSRz&J0_zz1W#R}F%1E$DeGd-5mej7U@Tp<x9`VYLQ?naKtK>be*)i9_}_2o
z4HhgCGtc1J61^XwRY(Fz0=wJX#5&{$D1}R?ry8Kbjyn;q01z?}@wq7UOA#)QIMEA0
zWFTKk?0ZQ20NVyaIP{A_&}X54sszNL1#AT^H~OYYawF3cP#i#xkf2~EJ|zpgq^?;$
z!0;rF*v}J=2M{eVy3pwDIrW|_3wWs~>g1-i47Mcpz6-sFTa!lM(o-AcsN)<%T-PxK
zHxlnKRBf1BG+^8o(B@+O-|&KE8z#YPq}w1q3;P{rjfJaPg~}$LB)CERs*HX!2;gb*
zN1WiSxJ*d*Pl`bDnjok>hDOHi*@t#myU=pu9BOH+#qiFg_!_+e)CVN?9JUIFdGudB
z7e@2i3;l|=9nQQ2AkBOb^pG+Z4R;zE8f~PTK3Ii3BXxV+lL}}+S<(N-0*y&Or~y%`
z8fAjp*-{H<2&**S`7!cwQMjp9A;v9D>EWK^JO%}?(08Ta+U3I^R{>?M;TtIzVsQ=l
zB0C3-d^%b&9Wn+C%R`5$OrbgFwfKNQ^JH3X7+N&&N<lJC%MDIbBX59@kFx9hv{@{0
zGGBDKXN-)ZoF|$!d{M-XVZPNG!TzUQ3(P*~Fe*I_6QN)*X$XAc3ui5|D`GL{sKnE=
z8(xi80^8-Jkhr)8j^H3%)}SLR)latxBkS~QM`0lZ&gp0gxdCYrT0#&Zr4Pui)*{+-
z*3R2*xP}w5J6`Grpl<OE4Q*qT+`g@$75$vJKCz<Fo`Ka29~fO4_|>t1I0}F5n-luV
zqfmL^#9uh^FBTBOv5)!(vn2rN{}Z<PU;p}3L8ku(S_3_gQ7~Kk7ue~&OX+{%uD{un
zx*qI2DXG@@A9Pjbti{^CzR;EszyC|j_1}MY@wWf^TqTUpu#r=C@uQ5Wi68b2!Fgt6
zV(~VtIT_e<iP^cS3?sns+qQ4-R<)N4e)LGGeA(s<0X1)gnmZ2;d)GE0#FYEo`IGVf
zo6&pCH?=Fn^FcYv`RZV6vb?kNth|#R6J`1o6_l}sQ87xr&o6F+@s45IVJK6%etqj9
z0Va6kkMlPE{npNF;pwG#^89)2moLFtV9!1P1ZBOXq%(3USt}kteueC5wA)CfLtY4T
z>Yg&cEw^Jm<?`G<WLqb<5SokCkmWp?wSyT_|5Tli)bC$a<0U`rTe+~=ZDCH_)z$T%
zbi8DEvwl7cZK_(T3O(kR+-ezHF=j#~qG`4t4lITkM__F3bzsk)AU>TCRA{!qKt%0C
zY|o#?NK$I;t|Ht8{+lIMtXnr}bi#eA7Rx(cwcH;S{*Ur&^k=Jn3D3rIFR^M(i8|I(
z#%gz=sO&?S!Wur|Z{_SDD;x^DLJlmD?-TS>H(a&0=Fv}KyS-u0Sv1`i5zQMmUcYAk
zty{)pcACfBO#X@2Qmx!8sd?>Ct^q1V=*jU8wT0cz<^(KekN&fE4re{i$HTrf_~n@Y
z+&EIsmjN>VoFRx4?Z{DcZoGh+FI!wSkL&57b2koNMDzD9U3Ws66##G5?V;9vkY0Ux
z7rNRVW!na^b7a_%fFl`VT22hhF*3z5<Ky&j4>a|stmEAl=1Cerx=nipes@?ak8<=P
zpaYDZpq2p=<ZPGZ*MH`XuA(Mf3d+Sbz~0r;)R<mgqi14f<^cE{>b$nZO&tu@U_SL|
zh2|t}k9@cJ>6`FmKzW5bV|8T^zv7qgA4y&m0e*XWWOIkxn}3s5EGkO`vcNceG*$uq
zM<Qmmt6*P=K=*B;yI3Gc?EF4*8+<`>BjNA^)o!qS|Gw@lap(7fkUlJ4wybYR!mep-
zOZ&?#%VyrOb==b9rqwVrGCg^8@lJpCa>Pht#6qn7ityboSTuQ~e(m}8O{r8N7^stx
z@u?M<1;wflwU8K*A-njK56b{Kkt&a0Cwu!o9v^HDTZlp9PCh;{x#eI}-fAu-Ly#Sy
zqVnrQ=I()I%$LTM_U~Lpy*nGal6utD^9Bc&$U*ACojdM6KF8v+CzF7Kep|vCi8AkB
zQx}6*xV{^!t8T$zz7g@l+0O3<>QQu#h2LKXOx^M-sOQ<1EniRpk_y1Kw~DR~#uqK=
zI^{3M``_Ld=-5BG6R`l0dWRk`Spt)heC@R9`L4;C5!^{%!#}Rmy2Um6YUhp{=>@+P
z`0Xw0+!3EN+Hc^7N;oQneg73dubLW^(igstnTS0AmSwib`cuVN;X)ihKTDU#UmHUq
zHkTAJt=nV5I`r}LMrt@!2in=!fluTCm{u_EXYVe=d~~C;`U_$7t98D|(HheQ?6FQb
zmARb&MLit8(RI@g&l4JO;qeFp22wz=cHmP36!y@_wXxg}&3iQNw@(Fjw*XqIM4geN
zPJ|yghC(PoKVM>~C0*kqG}_znBuc=`(0gg0Jb46D2NQ++W!SSE@&SXhG+ZKd?kLkZ
z*d)#_4q3bQ(LtvWIp%dLSx1(h8~+k@)@K!v98c6g33kjM+Q{gM!m)NU){}nb*RXo-
z$j97(oGn{ed34!^KE2>kf4Zan)vLXi`utj~()9B;C^9q8{~kYysWNu;gBDpGWmYtk
zeo$0QkN0u}++J#+p7yR{B?q!vS}cc%_!Ri-U~`g@MkIxQ<4vZeI_xWwmt{x5@#}?c
z-wvFED2$9Oj68IbZaO)lAX_ixdca>j`EGfcm-E$em2?|DkYYf1zN>xy_cIG5g$X(~
zC5T!R;XXbMJ_!w(Md6zQY<dh&*aU<0EjV@FJ{Dky@b3#2pT!4~RUq|q_&8wv%$o3x
zGcuk8oy1+-oV0H9W*oKjVCr_^?>NtV&)v()BcrdQW7(1r$n|^qil=7Ftq&gH)=OgD
z)6x>et-lHH6daoV@R`5YkEEMqW4y7sc|~rAYqEM8_Y!OnQ;fB>dr=6&-H+yiM05WP
zq(?N6`AyWg`1t#q0!95lq<wch)_?!+g`%OfGt(eR%ZMncL}f&XB2>sq60%3B5E@1z
zGn>pZvPEQMWRnqPW@Yd5e1FG%?%(<6JRaxV|J~n;>-v1&<Mn#37bv`S$Tf>M>Of=%
zT(}VAjC{}iqKZ<hBqc&R2Da6P=`Sdi5JLoFqjt>_l@jJ3dqQOfY^%*XCI_j?+O@?V
zDB}!PE>jRDT;%aasJ@dudohQqy0G;PcG<Rtv66ZAj4zvRRr!AgxeR@VV%!26j=DI_
zVAO{C>z=`06((Ntj1nMhr+V<PO4@-A`)M}y%=`5QC8S?j$3lHLuh4L2cJ^gJKn3oN
z4=q(dWpyiqI5`BAc)sYu4=U7QTZrx3jLLtWeTJpaVN_)Uuu5to^51b*g7;5UFqt_@
zbR}Iyt^uN^Pn?|Bhk}%O``OJ;(0o6(pv>Hm5MZ@^ifIQ%`IgU&Xyoj>n!@2k%YfSH
z3GU_5#$bm{tgIEqvTuN(+s^va(&m?QQ?Re1&~!oBqBvidrzu9(B7hw1!@}w>E>0Ua
zBDd~61hQw&h$p9{oVm0`HT8BW1VG}5u&dkXiea!bm9J0FEG?3$AE;&BJ=__O-Q9`t
z+gl5tFf<B_?n=eQ4tTI)*0dMmiwnzCG?xA^P9fWD&l3i&6DI-yeN!D?Ho+>R!)&mE
z99b41$1BjP(_NWq)CA2D$FA+JP&u8MYDk29fLIF?uAfE%L2(nWmfpaCR8xBn4tiEc
zlm)eQ^#p&djkmFQy6s>YPJk<!+N&WlshwMSYFwi|+r&xvpc)ynfYFgqbkgH{k}*k;
zNr^9(+y$Lx_PhGQ&CgG3FiHHK(QP7tMc5`7;dg83!*7&1bsNq7vu2I4cz$9ixvhHC
zGm)H2-Hj&Q-`Ym>Z|<f*7+}yG=FGs$$M=m{_U*nh7BS-}!CMo0J5JiVj8ym;96GQL
zmR)2+2qWVUH`Xp#(wOe3Xv;#LMh4D<%{(9X9i2{|%DeLxc|}ojHI0oW`*kwsE7^_%
z(23+OIPM5wK?}r@eY(_;(a1VSXaTjfwTsZyyvCw=!*EgC3Y!y+!Xw;0s&Uz1&@hYn
zbni)5?nYnp*g?al(I5S}#XfN=RQpsvM5`j(=;E5D=9oRyWGFPK3e4#NHP?T;1fz3N
zRV4v*gA0@YuGz>%YDe}5Js!FLi11w!5eQ~8J+$Y5fOl=w#Y+jcR-nG?LsS5I)H}8U
z$YeaBw?U`eM+;z^Dnq&8@G#M5{V05rY&d;~La&fU2eUd9P?OXvFq)nLG-#yS^e`qN
zM*jjDLqt^YxbXreAfv+rb|)GBs|A2HAs>~zUiXiJcg^qXGjR@F%9LR`Gxht&6ijuk
zt#RW6Mz?Qcgm5WKRXIqZW9EN2&~m>`&B{`3#|Q*-|3aYjWR5#gy)3ai$j9B1wV(Mx
z2PzbyXC&kuHIJ8-z5Rey!@@L*7gk|wZ_F2`lFzX=hwZqpf9R`>re;q?;M)9rENo0>
z+Z7tBDqftEogJ+nw=K_|O&TrG7eD@}mMJ)*XCS50^BPb&j5h4r8-W{bEvW{Sy8s#Y
zJ^j?dqa00j&hbuLRxO6ap^S^CrF2BH|5g5Wji(v*UZeyB2JmH94XLGH$-Bb_;bi#5
z*nrEp@jhX#ToPNZ0PIn>@d3K-C_u9lgB@bB@`qE=^t^{c4hnbfH+&hLmqE#s09^nC
zV_$_jv@@!98@3>4opQ0VlBr!62~1Gjl1YSM7OCp|3?2^+r6zP>Qz_05Ci)$_LH&zI
z$x^Gezm0{&m?<S*azZhcpqZCozr0y95=hWEh+?9iPfj*@G6m^`@c80lDQE_Nxy!yd
zjDhDoL2ijn;U`?(;kQ5D%P;SpAVh<1e{`@z51uj9Oz2cehQ0Wc^|Jh5(<Q(90|ZMl
zEqP`k)<^HjozH$rSFTFH_<oEhCsl^|{FLEz8`21Xp$V9plHp01l&rF-WuzqzvJwi7
zBaj8sEs3p~0)MKHgPvL?o;m~&o$3(&;`ZJr>#)6a4W+uSgO=?4b>zxdx^*64Mc?09
zv_|<idML^%|8m?Be*Umq@v8cBHGG^@w+2=e6G2wXM1O#Yl_z2r!D<3%IdlYWtH3-3
z1b`p^TByl?`vfLMS^BZ;gpv)|s}`84dS^g<0BX?tQ-|)_r8)$lsJlJVL-{J0H%f;2
z;49TgSRoubPR3_q-)Rq~8BL6IEa&`t?#O}Spsk+UIAadW?b-SDY-H9EvW%U)nMD1K
z051^C4);h(XO$RXJOLkp@_TX;Q><IkN}e$}fzlRR&+_U{AdWez!(Q!c$I=={V!XKe
zdwxkQQLlV`Oy5&A;p*i{3_<V0DZC(Zhn4IZ#NEL##{j+^jo5rknx=)ZUUZ|eRNiOb
zL8HVrAQZIg8Bc-sg`q<}uQ4wc;1+tsQ2($!D2qtW!$8PxM`#8q5bDUHUyMfQMxvwH
z+1b%PeFZ#YelpI(-F@Gcq@6-{zIS378{1?UOGr}e&rXThyH>AgYM+oa><N{_4D$;y
zJE9XI^3QB|q4(A4NNO#&QK~^RU)z0x3sZoG$zU+WEb=&Gu7pwp`LByn^MyD;KIwWv
z<J7D<1Roj1aK$O@eLjyprl+>Bd{-9OJS~9T99&%8KzYC-m=CL+IhO@>1Nt@-SlhNp
zT&f~-dAY~@>55Xl$3rhg%ZCbxTh>qW)hl!&o&(|dugSvEF5MIm5E!Etwf?9<Lml(y
z*@VTvrBK0KUwuyxszmY(+yjCI&ww0S9LOQVF1BT=w)R#85;hkfS2cK7PNsR-Ee~eI
zv{vJ!ioa&Wj1ylMN2qP@ug9sW=9zZYw$9b``!;B%VKShvq80xOi+c#fvYhTa4J`3$
zy1gh|nFDFM+#RNX`I|G1-iRoqre+<7{o@ueJd~BE16NMR3bnB?4@|zf*Wr5Dx2Ji2
zK*>ZMzT>{0UnGE<>4f$wag4Zc%LEY`0;SP+S9cBp0dIWCanKRwmpgFKP!T?*AwkXA
zG`8k_VD9V_RIABI^qHF&dUJ~-ZEud6l2V<`!HJ2zgk_;O`+4k3bd&dIL0nnqa^@y$
z<9)4{ghNa-qi_9KzYK}<;)WL@8NbkXz^wEdNY7uOVubrnGZ6ouid>Zvkx>3@rN*Hp
zDsT_w*^L}54L*%UgIOb2vTm)T1gU3+XC>-CV-+dK*CujP0Ov&!6Ubq6WuKe4kRGp6
z6_=nHXvPhl^s<{v{_=DO1})!$<Cn~;LW2lelySCJ4*YtKsn}n*Oi^z?{NEJ)Kg&7$
zK}U@25Ka1!a<jZ4-e65g-M9II<AA!&`dfUsq1K!$lZt0v%3EH4YFdfs%D_!92ld8d
zIyNAuu#)<F4m$a+y-$D!{AESxbzdD&_wSjHxLRhx_dLz0Rz`8;cYi4o%|C3*kvX_$
z`pFaP#auo^SzD_*UE{K&FN473*T!g_10U)ay0tpXd?aCM8FX`Bb*?C|!eT*V4Rp58
zz|)NW`S^RGtj>yPxXgZkA&Ppioch#`F~NCg8V9~r30nP$7Vd8Mv64~)%4<6$t|Tre
z_q<_v_``h*E~KZUzuSP;8I3nrJUZ8|&VuwE>!N2dP^Hu>T_gy7cW!p+Z6hnh;nB=Z
zbD^4Lv-4FwW1C$mh%N+Y{&CcvucIak^Yc4%stm1xdon_|++F;<7mWh06ld)5ZYX2W
zfWg&2zwe)2!Y(ICZGdoK`uVXEGbb*lP^MphIDXOnsv}X65D4s32w(N9%-6x_xWAd$
zE904O2Xg6Au6##ewZtVeS}Mf&k>F@U<-|xVPniIL)pm;?emsWujwXc1UGMsJ%u4gA
zwE2m;X>!i)xlf8j!;6>K@8awN_RX7~biH+IXkd336UNO~;ywbj$pp?a0yhRkG7RG4
zg=S7YRtiD&$#OyA%~Q!id7d<*VrSlO8ZB^H71-KwD>`>?3+vc=Bg6w-!W*`KbHXn<
zZ-K-oU@TDisG%WlGa?!WY9VYlpyWRUrUg__I5)ef!4^bam00x10QN2e`o?(tmV0I6
zF$;)NQiuVNg(YrDTq#Y4i9~Qknlj&q-32GSv8i674#uZ|&4x{p2z}nEr4M^e{5^!y
z_KJW0cPFz*j=A~y@$TwS7!#8aywyf20d50kuC~l0(0{LwdxDwx^6|TlfMo*sM`^}e
zW!`MQ8VorWNEH1)`Gw^HAmGilhWwAp1PXd^D@F~_FHvD=`C7}5r@G@111<xjDc)$~
z{;Iy5m;P9{;v3Yp5Blp~Q4*<Y$Kw2!d6@@f+F69IWESM%>yViefS@8xv)=Bb6h&1e
z2^=|+kGLC1OzsZ(DBT)e{<zvaK6C@7>IMb1A75NGA;>|QVfrR_bsvX0r2_0INDPdh
z=@Gb~05Cx6^O~(FDk|jX(Gaifv=9^P(q!UHhwDOW%i(95WgouhEiHI`nwt7;qqA}C
z%<@`|@ubEDAtY$JPll-<r$k1nx?btVz4!L~k5Bl?XAk}rOWa!(h3Xlz%Dn}5><}<)
zpeJrm_z@yK7U!oTgjJ}xKXMAhUEP}WK8zp_Sr*TAq<szQWzU}vxm^Cr`>@KPuU2I)
zNMkc-Oz^N7Y|wvb0q-ZSnYiBH;#2>;BHQ)cF={AiU))jyDOu>eE)30O#d7!a#z<e3
zC%sWG_w=_$G2mjXZn@tzW3r@?%Zi=tpOG2atTOK8@L0%wiH2yuc5>?$NlR#&F3!mW
zSI(IaG9o*oy}dorC~B%vQ!(4p1g>X7n5CvIzVd9%CD=@J>vxuzn8w6stTSEZe=-e)
z5c6178JLVSPF((la&DPQb4!co2KFi-hDTTWNrH^T?xd)4*W`}>q3*;-OQoIqo?+sI
zR_=zj_M=)Jo<c6f$OGeAv+cm^AA^hPrCU0y<TujMQTG0C3kFJP*ej~DR1W)tM?u52
zrn%J=0}Zz=O>W1UCSqVFgFKDJIRvZwpe^dI3!zfaTYQB#a~EN{ujp3DN6YUzdcG;?
z(w<__tXXMB{S#k>24X*x-uikoAHnY2yg#3Xrob7Ae;b$Q+4E%mAK`8wgj$<DJ+pud
z0FI<vlLj|;h(%htZOgV?3INjk9?cM%9Ta41D4;{aEfO_AH1k=NZAo1h%*0QO3pu&V
zL8CNuAMVjM#HzEvv(VQbMhB4b2eMffK3=B+axpNF<uB+jaHj&N`7x6}9Plbbs<1KX
zHWScdt%XOc2l%Fp3(=v`B1IVLe88&bQFRc>7VuDP;Ag0o?Ae5dou>a?B<mFI?}G0U
za&o4xX(0m#p(81UjVy2Bk6Yy;k4o&>xQ2LOdF;rHBXRw8u~)rtp8td^k;nJ&`x}L{
zBoX_!u&J5-%X(X?Lx9~_=CjayK_vO@XTm1Uy{LvEGTMsG*|6p=J}xf;H<0Z{)H+(<
z{5<NBJBarDEClYO^NSY#LWYw3LP8*ZlX$MYfKLl@`Wql!Qb#{wO)*2qd3IsM!F%7O
zuAFVLTdq0*14WhGgR=63-P{&v3d~#S+1XL9h{GL*<L+7IqN(at-s_#GDF1h#jAC0O
zNXw9roeu?XlNc1Z(n-K%XY6G-Lyp|&DiWv~p!x9wCu?=mr7(uOf*&q`!u~llB#Ry9
z%)r(KDGx?8JR_orWUX^ps$bsu@sCYUbHUte$ifm35P-ype3o(Ra6!|C1F<3^DcWN<
zix=m3usVNM1qVa6!iOQ|=6(90ikm(fZU<YUT{1OvKrPdoG)y6<sB%Z@-SEQV)_UN~
zKVm+h1bnjue+Hte!jQ%h&JQAwf;dj_on}?_E^#t+2mmuCSJm*>dkDav1Um9+^1X-#
zDJ7J13xZaBqB%MJjfuex1D9Yi3PYkIMpFBIH0{AKq;t?9&P+CHrY_ysrx;0p1FB7<
z0j0D%ZDrp$r7@O<v|H{w_-TcN(&>X6RXl=^Jm}(-8lC<FqLbg5APPoCvRL&qHlNqT
zYi6C){-^vYxtw4547Av$jT=ux!Pandr+>_K$m(S*mVrRWoPPGeiI#NUJ-~|WS_NTX
zOG-(33wZ&?Tpn(dn8g^X2SQwh4p>C+nn!|7OJv))CzlnWsYGBg)SAY7vuIW7i9b_D
z=8-M<EVXTUFk=nWNkM8h_2|(N<gDUbKs`r~hX9_F-?nHKnuhq{>9lFuQ!rzo#>*IB
zT*;FwgDb5Vb9j+bfHA3c_r6Z1VuB<FYbBa?K&~81l@A_dJkMnnkw)M=gADc37E&7U
z!?gsCJ_KduVcofdhz3TIKlQ7FTEhXtP-815-;}Uk9(AhhFTxE=nc3Hv81&Y!#IN+G
z8HeDOE`);Givj_?0gH}1^U(oCi@)XLwxPt@0e$m~`S)Y-O#6aafP>He^3{a>=f&{T
z!mje#95o>l2j^9}rZ(6O*$yV1`BVtwd9^n<WQGq;ECT-4AiIKEB@sm01r-!aK&vAi
z1KH54*M4+#xplg!V&{Jt$S#_hNK;)H6$-Jl%mvze_u@3i0)x6`cY#=_PF-IOS&9R_
z!JzWKZQaIPhK^=zYU)*Iu)NQ{OAn3n<}sQr{k_nE--;Go>Im;TVLS5z@Jo1>_1#;Z
z|1ovgk~jf?gPYmetFZe{np>DLBYY3F!a|T|wZ^J-P;cW@u0kt1V_NV~t}RdiMPWY#
zODNrRi@v-J&OF|QXv*)$Aj3O@%G6c*YRt6hJ=|@#=H|Xlwi>5Kayy*=m)<zz!VrGu
z^V5nzq34r>4H&!hTds}mAUq7af&TX_kI`{TdC%0fZ43^@AnVZ~iw?4)%rEy>oxXRE
zPc+k_7X|zQxvQ(yT<Q=GOk$kK!G3Of827Nlz@ZqdS&Ta7hiFI});sngb56Sdwyy;H
zEo+VB7QW5Mq^w`wwlUl(hid9!ThrS8(S5oF3CU+aZ&OQ}-lq1pX=_&WM%mvhZ9LBW
z=5(!j-IP>xWdn<Zd`1U*3X8A0|J%1g(^j^dbSv?@z8AFdr+;DnZxfA+_RAIj<JTrx
z2(GyDzuDncV?Qo1ocb@)_T&xE|7L)@Ht${i|N6&8Y6v5H3#BJH+Ta4WxF3W{5Oh%K
zYu-ITd#!bQH+CuRUY-9uADv;1o&c`TK5-R^zdzkeDQ^Wr#^EsaI%=p~@%1Y=<R&OQ
zB)zzZD-+`*pp2#m?g`26MF<YAAVU)V)w<8PZy7u!y?zEG7--ruwac^`r2z5K;ixPP
z5Po077_kL}fMi$=3G8!PBm^w^X@Gn&eg~(LBz_vI*lv^v6fA87ia*Ca1L*M4<HtF*
zgPVdQfvZ+y2MI8*!^)NuxQ{XxT_hU>HPBUat7i%T*aw&;xqZBZ-GO7!Vgy3>_AwtO
z-MQ1S?|>2{1`1^Uyg(ZO1KmXkD~Z(r0B4xp{FHW?cWGw^z@H=tUEmKNqcRA*UuyTa
zN;23VBIYF67I9ob>uZEhPD4E5kTwx33w|+n1r1>mqg7M)<4Khi7SbTh2?*FH=qE=J
zmydV_6a(uWQO5AY;k=?M3-|(r7|@s69B$-3$)J{Y`v#CMcZfnEmf&JchahM@>#^%*
z9eB1eSCV%_TViQtRg>r7uzUAzh$?BwNrh4a-~#2kmR2EfLnIpOA`S*;Z*Sl=-x&eS
zNALtw&dGBhY%5N*HBjilKTaZoasH7YUyP;nhc$-k1(fV(9RN&Z`GQDzi91l~<C86l
zyXzq$SBsDf)}V@&4?Po$2NesQH*7S@`|<79$OCd!#ETI)cE92^FUA2&8LOQ-<w#wJ
z;-I6EGhDcM_-`592KG3>gZoeUK>v))t--XC;^J5+1@@i7^OsJ?<l3G!^6(Z^^Kc0W
zDE)#Kg=zpd?1dl;Pz7t`@q{ntxNQv10dBS*=#>wr^tPNSR?WM2lTy;sl!3h}Ix>rx
zMYk`(D8cl>F2eh%t7{FLDJskyJA|w5$FmxM+&jkNbeET)#qhP2(E#3}iK*ulmSv^h
zZxA<^Aq>0xa<@bsI&rgBm=R(I8WR&Ksc1hrs;r>%*3r@m^A|MoH)wpRz_;R<&p`y>
zgS$dU+>~o~oM@h4>OoMrFX4s(H4(xN$s4l%0zbepSeFDJGwGtKUl4<A{Di~Og1Cyf
zoNC75zoil3$ZXgE(G~=3qz%p-V)l3fj&_YBpn}0PUud=PZfe2x=G+58M5d`zu%x6X
z0>}x7Pe@2H*aX8HB32bZ(EZVJY`2DnhV{^_xfC}P=?|f&?y#SauVFM|Nx1qxvi&~#
z4!lNMl65pWMY4FI2|xh!@+*~HCy928ENr4W)6wzK`;GpN<mID-`rOc<oK7xDEYxij
zsOdSXn(?WyWT#KJ#&A#+YXDU<fjmN0gxVFg3^QX<xU}yx6j1zl@TelpmmQbkRnZe*
zMgt~pX_;~7_RJuFYGFvY@UBRh53DXw$E~KP7dTdhcTd_MAd#F%F@}aoG2xmpdb9{3
zNwyG7o+HHvmS895MPXf|r{T_=LjCZzMi1#6ztH-_iAJNQriRA`tK3o5672sUh(H42
z0qLD~#p{>Qvtq8-eRY|8|5>eX$c9JF0gU%%lH5GsNvI(+jo`Go0SxrDKXYZ58HU18
zNN3b1401pU@0sY#i_Ye*d0H37_DT)q!U`>DlfHlh!(68zjaGYgq_I%k5B&}`>TO{>
z2s;7}*k*7aEcNQP_IOF2dTMrUAd9IA!v(Va?+`5_G=7{ld|-D8%Y=2g7k4Iig0aEN
zj=2~;*kMqCd*fMLLhX81E6aRP@&WbAnW6NKW5`^<X|e~?18{}&X@5C|Z8_{4-GC=U
zWUk;Hv_V25=raiK3tOHoXSFTlgRpmqKz<bpT_f@N!qDf*t+<zX8nLt|%8v5$@4?I{
zEPC)Jj9{qIEXvlTzVY*;N-X|h0Y)8X{(FRelEMtDhkk&JUctsY2LcXn=@!llnAIMF
zu&j-~v<H7{-HUhgf15;$qG0OVj=2iY!OU}taN7ZhAgC3LI!Ez@Kb4oe;fUMI=m75s
zG#&^T*p70+a{J}&!O7qVZ=&kO9pXL=VazQY(CG2qVVmT*Qt3!V1F|$6ryY@HlNJ~8
zkh};>pr)oK^Hp(r)E+NJQUUhEpGHlG(ocfB>q-1Rwg~vWbyI%0kB}CPYsdBJQ$A~3
zU`$wvp-{iUOVq8dB4omTdvuH|8fv;V@=zd^J{^JUh`J65D`prUQ~s)x)e$$-{(bw-
z8+V}0qVGA+$b0h<bsehMb(MG|(kr9E<n*Wc98^=*7htZ>wldwy!t&-_W=T|h7@|!G
zM&6!_;)kU!8M?f(mX?<5K~DhP#lnTc4YYF-+SIoDTGM~u<1TVwMcpwqO~KLs$c-}(
zO+bCFk$2pGEhDL>YH_SSUQOkyswzn}iMr+USRVB68hJd*Bbj1?g39=HG2lUBKoxLD
zcw=jL*9cR56bA5Ot&LdaAstWzq^JarLb7TMZ3@pRJeR|hhh%KG0nK{V$La_Z%z1(B
zmr!f3<e14qLVoEQd0aQ=edco4^;5lmii3>K5rI=FUGVTzK^1&IrK_#2jZppVX}BL*
zyp!RaK}0eUdq9O#KVORnLdnrEWA>np*n;&ism73q>9L2PcC2ZIqWrB&36vanVUt0D
z{1CXH#v)n*LJ2~hxH^HJWb?1uRTR|&Q95|pRlt65#TP(jyUE?CJzE$mO1P-JNQD8W
z^K7q=2(}L%Km-mv1T*|nSJJdChMJ4vMQ2hwmi0ZzyKL?3HoMnkV&po`tbh|%o|nQT
zlfkiLuM(#joONtyhe<s{TcTfH-yq>q=Y1vUEVs|~3LNQVyeoi)3sWz;ElTfFuO2J@
zP>4%h%WG?*2Ue$gN*79z+exy=3u32yuV1<HI-%9{rnEG!kgXxOh)ZM7-R$C8rr-dX
z9Ec9Autb0@=<hue=6}1lyA;QXD+3ZMVrE1kgb^-UZ^!UG>lT+mvFl^19rny#xKja%
z)$K&~%}3gj)h|`a96(3BmV(?y=OuJ&BBd#CEmr~%#9>@_aE_`Pq}aWlic&}*qLt&B
z@k1JpSGKonBpx~wz74e)UTz8aLyyXWxu2z*1(-eBsduU3#Qg=d3Wg3o*>^fSSLZ8v
z^Qau!IqMpS*G74P=dS(PaI?NEbYnC-al&CP;)#L`tU{*S<;bH?8F<NPv`K)sMFcmS
zFK7!cM&CM);(s)Szex@mj>6>BR8GXKAI_cGEm4BtEIw#@5MY5w`1sS*nB>@t>2)um
zPF2Jjse6~X;Z#>?I^0yy?w0}XnUsvzw;8|#a>v<(7vY4v1RJEJwDiK^*{FIvNq8PV
zp%0?p<Os6%Tb$d5n275I4@t5S=B~j)7?M5$p5uGA)n-_a!alx0ML|;y-Tt&fWB7UM
zBU=$=o^ri_8sxqt>a5M+H9UxLmjdvWr+y&htB4qWx%K2ggzCqb3u+~(7^0d**8J5b
zj3vO-Ne?;~j+t^?q?@m?{lS9<DzjYJ-W~*0mic$%`qOI}7=)U(Z`o3WD-(VI@d0HG
zmt!qI(ot+zw=VtpC9Q6AP%+ZIMhozC6`lkdMxsOk(87`N`V2@c$C||!Q?8>$S6saB
znVWMWL=A>XMh*%lA;AGFka>o{>uCOqVdd+i-p&570^9kY=7Lm#a}48yaN8B*tHVWo
zROG|7EDjnZ+?>RAoL!vmSU2VD?7Vs7#+L@YHk(^6mw+H&;PzksR|{Zdw)6x)ey<H<
z>fclwI}{k~-?00UVcU$#_hG4+<<U$o5V6>q$w=qVvBN=>xhuC=`MM}{=M?Z_7eBYG
z9b?C#2!!G@DlwwLGLrK>#D-xllFJpCvEc~Y`v!RB@ft`<4t6-fd$E+619|}*Q-|O!
z(9V+(w^NcYUF%S1kro5^e$C)7W~h#^Bq{C*!!~d_p`59$8_^)g8XU=2GR?S2KLMpJ
zltN5qlW;!k;KZ~Y|0;wSXEF{rXAy-FK@eg3bS-`JOBaMCi6@vH&KjA4QhIU<syP>Q
z5`V_4#1zT3UJjM!dh`@r;HP3lt|p;w0cNxZlXGy4<v4Tl@p+=A$M}j*IBOv9p;qY4
z{imH%Ch+S52vmrpi1jy&7ot&O%67zGRZP?n&0Y8-O^mhpjIJ&&*RdQIaHIo+tA;w)
zi4WUj#VK@69c`aos)C%I>2SHRidri*vPgy1kY+4w43`kCMrwJZ7N(GiW4H~`17N#*
z@4x&Mj!ZIB0evQ9?#IwN)bsG*%@NxIv7>Bj*|M7#>3LnSQ0K>G)8#|sQigfpX1rK*
z95626Kyw1JU4xmFQm+nNfC3*Pab0vg*Rm~NQqj=LBTC1~H2iZ%jweM%2Ys0irah36
zYpA;w0V<Qb05fQ?`gdZO0)#X#;g7Y)L<uJ$!f@b#Tnfd1-2xjIK4VnIdbAoSm>h%c
z=Um|)$KK^9Kr1$E6VO@+=cV{&>4?(>n$Ks?pEIDgl`;|C^d6fUO%ax|9f9p)lwvlK
z?iSS-`JDIy{hkij7jik$zPSFqsP&t1PEe6x1suP;%ut=alcr=MJqdj)MlPG7^gIir
zH@XC3=LUg?@@)gIL;5l@Q0Tht1KM2*()i_Qm$d*wOFE11Rl%ntQ<hM=-N17qlWq}I
zO?=Hb+L$oD1>$+@5~3IZ8Vw`q#9viDJTt_2VBJbIIJnl#97gJ5yb|^(VI2@a8n6+R
zz%KwG0ENSE3&TH<ta`c4VOolMlnqzP9^!eyLq`jm1hk#7M{p2ugZqbmAOt0i<nn_c
zOlX2|vx_w4SizDiV^U8c7KSVw?hqf?K*&=22Wh~>f+Ln%p58wb6aqG3iT0wCB`hm}
zvM?4BhX6O^a43U|;dwrM^60<iy8dUJItsQjFe;(um#yJ#u_}3N=}Y{qm@JFyNEcNf
zd{KLuHPJh?efJpyMC1+=>8|EpqQFGE13WS+y0UBB9EZXdb`+;!TR8o2_z><;(Kg89
zLf9tjfk&P7ER2T_A0|0<p>h9ao9y;b#c)sJszJj8EP6O60KB?WMphO-nKyVba89F=
zyQoBJljH8wL{yBP<!`r46oWe3zc7&g6}&f;f@`@h23YpGu_Nrs9ZeN(qC#{8Ucb%v
z74=>uKLP)P&Ga!KTf4s4e?g`5-GCDCnQ0L!*@Jl~F`2J_4g!-Pu2K90(&eKV&Gi~z
zHOBpafKx}j7}Tl${S>51%G&dPpihzPnbd#ZwJwkJ1^gM*SMIX(@(T+editSLR#(((
zis#$Ix`5P?Ku$_9GDi}gZu|RhuNvD?I(>XB*)xr&h-c*X#tl|q{xwM8b6fJK2@1w2
zzp9}6uLzkUE}2dL4W-$B;`o35V@pE>EM#PTycDZ_VaG|^naaw|EG*@|MlGGce-=R*
zy;xbf=j0vO%YT~Oj~w#V1T8xs8W#pJ^O@33oY7aqXi*rC3}ycZn{#2n=rsQb*oYx8
z`*ZBL*Kw6Yu}TRQAkv{wxa3(vw{qnTV_oOK_=6aDC)Biz8yCHq1PIM9ua5GpT)U;S
zT+rwkw1Qa~th*Si^ThICs6)d_uqQotg{Q#Ev~&&2QL3$*60;n2<PAHgn0#hsif*}N
z0rkpg#VkSm<!sy8O%!w{kQ_Wk@}6dzK3eFR<2i@M#&Cp(;ETX8Q{@>U&C6RO=Rmw;
ze}0wSz>h^2w)p_p{{2sxYNHfmZc3!+e0{1l{9WL7(+hI-dE}94pS1KLuie&O<<pJC
zxle3Nhc5*Ro8BUHBMykZ)+o7wYj^q1Q_J@~a(2dtm0qXP$e_M8@SdjagTVM9=Y-;O
zd*X9{%rt#rksymI+Ys<zxN8&_In&WY1B7~t!6yoc8L^GL_&L>j@~n+KrT`HF(OY5l
zD=7XAi=Q!xqu3f9ZhTI=js;-ZMkuqnE}VKffMCOCFw@b{t`9$3>bbi3;PiO+Xq6aC
zo5fJe(fum1MSzS-eR#Hu+DyJp#`xDe-$TIqp`Dc2ZU~1V<m_;6uIYHj;#;P&=j8ov
z@Q_6A1qU43S+mw)F^9_6HDP1XYQ8A`IaQxb+5CW-*$p<z4oe#n++p(Tj}Bs0Z@f5$
zVGoCh{@l2k4=#ps;9W7Ac_*_hhTbZF^$%EHv{1k8>dD#Pm?ws~h@ID4)9P|&R_H06
zS-zPa5mgnG<91Wx1ddrU(F-5iS=wQ#6peWD`}ccv6Rr^x46Ypoa8FL%K>J=JV*_t;
zH|~r4xE0qcS%voj?{t37YSrO;vrfov{vL*)2^zI8gvNRH;AzUYwgtMltg;ujRvPgQ
zIR6nmO004We7D@^sA*~^76qnGkR+|$h~2Z|M`;-jn);o_dr9LCXx7OPYLe&Mmvlq&
zIXhIJJj(AvNTmxF!)AKKYnM1}+Y-<iUhUEANT(?;y5MEjDi?6<G$vxq48$dl`UwuT
zT??4`jk;1k;+_sPax>kZR-!y|F9v4o_H^rhLiwVfJeRk$I196_`TTLSzWlU!HPRJn
zBy!G^iv(TyN0dPDY$~KEaku5(l6^x;BGaDVS5XDe4<Ayn6@Uj=qM1)8`f}Vhm;xGG
z8$-qEriR+eF;o@l$XYn`EBuc8qAg|i9<P-Xt3k%1YWkh8&|3rf&X3y#x5edd)Y%L6
zOW6tOhQaMQgf@kMIaDj^V&*;k!{X(EMq=y#-I}MFE(++w_pL;7dU|g~^Lv0=l_BfN
zA!Z^?>8F~xw06y!x3~TH+?j<v@j5R=l=<{G#cW23!4C;Df!Zk9Fa=N_{=&7LP5e=E
z<BJv@Cb(-D<i~%$;kQWOSoY)&6=LoBYAa*BMeJ4YRdTd1L(@bXqDrdKSf%(i`2cU+
zcd($q2N2SSrG?5+keR=u5%6Hj5$g)ZoFUw6KZkQ{JQp^{81z14Bi+dqn+V721w|>6
zkcNAfiT7T3&Fe0xqhm~UbD#RU<Lg9S9BRpTsCuTHKHLTp?nb{Sr?JDXF0UgNm8__6
z``g2u3v#JfuGuhAe%NfVU9|LGLs3ou(^2aWhi2&#pNvjnv=!D^+MQ<V?18ID(UOFC
zcwLJz-XS#m-8dkv_f(*S)(6^xoMgGX=S>VeS9G2mcs|~%kN8($nA)JyuPyv#sQ`{C
zvUdByt{K<OC@fw8e}Y${9PHlYGemw)(8Yet{qp4tgZg8+PXGbPWK4A6l456{3U#Z7
z(E{rrL5eWJJM;~3?2A89MQ)?>xw|FYs2KzAoSpwD&N41CRILdZ8FA#dKl}y<x`CC|
z9d3PSyYk~|PPUZxXfE&OIcw0A3P~-Ap*T{EeZGEqfgjquJt$2jwVteJDaOX-#rWAK
zi-W@-K_%q1PXQ-|YZKZ^oAwWLU7O!y2o%;m8fE{2TmXH{Y@AHN3QrZe<huI$T)e!z
z3YG>;6r(Umjnxqc(FQQHDD6F#o_Vk*UQWNG?l}F;o|9_kB{Sw81IdtieQYD3H&9>n
zH?JAiUB`zZLA=&=`*xhLq&VU*cXu1CQG&(g_idxqYv}9Ph%Etg`ugkNbFKW-%Vm>x
z8xAd3A0B1K0CDcmxDiY&O!w$MW7YoZ`8Y+V?8wcTyHHI}-pt9#Nq(P7Qnn@rn!N&q
z?~TfK<hp9;E(lW)D@FbY$wXbKnsG(A!k)6Ea&C{8#YyW4mmf&wcx!2rqFqTG2;l65
z{HvQ$K7+|Qc3k+Su;Sh|+xI_$NgUSqI=9v|bDsNYX$P)mzs4~nkI=J}Tcu~(Pgu=}
z49n)@?V^pPsl0gcXQhuDEtDS)0b<5?E75OTPY*}{i7gG_-i$v+zC;R6?6V_Z2RGRi
zzWg3sgPO@CDLpDJE307P!5hIlMU@S&@NsGGqUuk2pGtt8Yct&pkUl)|3fpt_y>_TU
zeJqh%c??p?e^lW7^XscL8se~X9&gEgh(A{<^>p8Hv}~sdAU7TI>!DjN#kCHdJRV{p
z)`ycdxGG+lJ<5D}&NYBtdMSGR=rwgRtr<-!?O4p1<k;lYCTw3EDqJd+0mp<z6!J7g
zfnUrOPs$8fk-F(@%g-M_PJdNRz4GzqN-(2c0K|YPc%hN);y2%qOSrhayu335T<%>Q
z`r8g%)0;R>$qzOOR>umeDBH@j4`wA{)1N!fP5oASC;#9e-zUlN6hH((@+ENFcns$O
zvjWT&d8xZtz_@02^DjvWi80`wH87k}KW9an+4?O!8rKdNfBN+Bsn7oYpj~5DS=N*L
z$#sgJ+PZdkcpWMRj9J1M+vYy|p(!+cRLeOTCQ>_3%4J(3_L4^n@xM!U^X(Z4rJNue
ztAPj*UpTbT;m$7~RSiDfdiWv`W%TB^fRa((UHRw<MG9rKZMx$<5<jm!-CD~Q=Kc0k
z;gb_~Q|?qWbX{$U>}abxk*J5uYMu2j8;9|A9TP*+NLKYjd`qovMF?IY>3q^vQ;{N)
ze5m1oopopZ-ACwJehdwH=z8ttMka=HKx_J)nxa}SPfn>M>#xkpOQSUXGP1Lg#+o@I
zU{Jh8QVmH<M`A2r!!*?tLWL$VA}#AUh_Cw3hk%L`et)n>mX?9((|4<JVxelKp}re;
z*$uQ=-_6>RdM&_lr1hp}5p>6yt&pmEl=Pi*RyMWx#dxFSI~KdMl>=YbLy3+;HS->t
zuv6<)R1l{i0Zo8k?_hcUX{QWac^~f7KLuXTF__CS^<>XSu`oQT>O=<zv;d!xv0zq$
zIgGFI);vXR?xPDIAL!jMy>}M&YM3RSh?oziAgi3@4Om`bY=faGHQl>`W+x6(qz+V%
z#U)xICIH^T+b!9CDVnwg2yJ4VYPm+r)Pi{BgDAN!zVG=u<|pV8TKME8E`KtB9xaA>
zGRk9!5RV{THC8#U*xiqxya(`C+AtPN_S&8I5PI4&MWE6-SkAIYj;A?~V&uhNh?zn~
z7}f6V<6Xxe^Y=}s?<E;_10bH+V1q}pR<ZoB+u?nevX8Z8nTH_^<#%(|cF@E*S_rEk
zULaBc))|_He2!8-0kiHw89x!9gtp8dBO_J7W-=^BcHlJ827@#*F;~>P2M7tueP3i4
zzBoC2ykk&<eeQVemkFiS1O_F0`wT+kR??;*j-_XIj@hGrsxNb@1+9zKH~W&qSCTr_
ze2}$B`=WC~wWXt~s;ePymD{}W!!Z-nSy8uijVIvNAPa+X6|NVJ?8|{(H=7cb!>w8l
zg&~IeNXt=?TmUpF@S!60cC{n!8I<w)amjr;W!_ZP2LtWl#7F6W4yGzI9VAn}g1afj
zs&+4Ol#>x|xB>T5xO5{{Bd_@OR#Z47ZWNpbA~_=OD5x0yIRw-*KV;|a_UnAoMk*~>
zgosaznx6Us&)D4LK!3f7|8oWqr?<dwnhiC`ZdU^|4}J7&5uG(|+Ya{t0&s`z0JoXx
zX!R~M+2+Z2hKDgM=+O;v=YhSy>ly$yxriOQRxLYsHT&%9`?otPA^Y|l=nj*vR)(_&
zyhpNrw<Iww*JoI%BksBcPhw^k<m+m~9GGZSBS_JWjA$c-zn&ir((AYjM<7%7kMsRi
zdLXP>Puz7J>#!%F215OAV*c6*v_G<MguHmQBy$GyKxURNESODa0c`M2lz;FoLW<Yw
z8wYB+Maju(J%O{5lAio3hX^U?c335WHS7Mc_IRXhMUW_qqxx}+KgbFpjaYAEyC=wl
zMzfIy4;%vhX~2v;#43YzaV!cLHOQpr7^dGLsYVB(9t8PkQ1bOK^>WUo6ZdA0z$hlr
zvUr@kW!@apo{Pc4VLV~D3BG4qc_D2Xcl&yn(fqos_fGs%TDv`a7Kwx7c59{&E}rh5
zo+prVL(`~(Xlv4QK>+g-pSd$NH1G3M&SMkU`QLrN5E`TF=LN_s5?WQ39hSqaQw`kN
z_@98uVhl9XaPeNXdNtKfQ|dzN#a9Qrzx*it3n~xIE4D<yJGj3LfPJhdZ5dG}py!RA
z&bcV9Htdv<=~CsJCK?$Y&$&00d-I3E{^E|{Z<dyq8_|KjAu^H7=&%XO!Ya`W2X@Er
z&z_Y&wQ!df{EiF@?ayjgu2gfiSAPeQMLk+&sPpbQOHVV%I{?s8(<M3gjX>Qtvk7fq
zvCvrkCS7k#vNW%;xv8o8ot0^h5jYOs=x%5ri1WtIdA)VAhp1?Cw$55KR5Xd6=b@2A
zWRcY2S-Q@i9!bQ5A$2T#*w^({R#p~jOQ=agq#KStR}mY!wspE|n*ppf+klXSU?Bj{
z8EDIcToG!ObGWOW9hQV^Y+g38Jo8?1{}v@z^-ZT-Uh2o<SD^*lz70qPjD-1=EIV)G
z8#hjSS>_xI3<%h#oP7)s!1t%pxe*_BsDh$wn76@HG-A|*e52;;jj$mP+$JVDp~+z6
zIi-D5W>Q155ciIVP5-jXv5a+v?@hEi^Jb+rd{vFNw0-|hzsaTLz71FV`^mkNgpDaG
zDYR%mk#@Vi0tC%ipja^M;nmS5T0}%x#%|w<8W^F^8HY54CM*gbu5ey5(6<T6XowE=
zU7)#ZViJn4N_GQSp@^l&u2q#@881(yCCaviPtH|$%ye{kz8yM|)KkL#aiUQZBF%^3
zn>TIPaD9Uzf=E=hkIh=+e5a(D7<I|&x0){Ouu(-$V!fL*pN87uJyos=i^ZR_dA`w%
z?g?kR$mkH1_DHYWbN>EexEX2H(+^^*mF3u<h&!z)JSciy{eIW47Aqf;u6*?7f(bNP
zUjhudjx{v!!U2l4{t`gHQ;-Q7szzx&eX#`&0kf3zFL#r?24HYrz!BP}u9+BXYw9Af
z7n~OCo|m(2qH`P|6%Uu+-R9-A`O2a!PqGlJ@gM&EKoQAo6}VgoyM~MI&yZoEo+b{m
zlVLgXfq?-ti|HLCu@`5r@ponS$+TMg&ieb$@8Cimx3in=j|x6^|2HFg10oY@WIj6K
z`4JY|%o}Ns*KU={*fFNO)v3-VOmD*a!;s*xb$etGX%uEG^`f@xH^=%EoP-e;#rrX{
z(Ve3m)g91_yx=swm9yQssyot`M6dD+@$WfT8G!DF?tBK9)T?(_A1pt~SUo2AyX&MW
zi?jxrsR<PJ$$F-@jdxPVYn>lIHX-F_S~9OcV@(Q1*h0lajshG*HP}!&i!J?+es`Jo
zYzPjnlcLrA>Q6}E72BQa?lkrY{V7?Tw$2t_+Vs>RFkvxQ=NlilCwA;F_rXp6-&f?v
zXK0_fxTkc#mZx=CbOak;w%*uHCvD$X$qa0+>vUxgX_-nFMJ=B4s^*(WtkrPSp-z5f
zVtgqzr_wX_+&$0uA10G<Glp3NpIlx1S@YNa+@kio+5J_~mj@f%cZm9<xkFNqU+2%5
zmZH11E!w|9VRppsDAjiZu*`bQas#LcLF1MpAeL9NEqSp5qD~h+u@$o76E^+zE!7!u
z5xinAKW<Ge4AAsdU3TH<{jcY}{iuzdy<R-@Cqx_~UJU&Ekpf9PO1b;vOxQ1<XKyx-
z_Qb1!aij&SV=rp)41d+BDzarGN3q7OwJ#M~s-_vba}GGn{^2~go}KoHh>CDNw6dg9
z0HGJY>`3)NuZ9}rB#N0k8{+JC?C#1{%f8siz5BrkbJuw2Q1(WLiS{?qP>O?m{DJz6
zcF*wo;+&8e<s6aGSc_EPY(P;FS^-nl%@gfk-eZpKL&>lUqgnS1UZRhUR`^v0(c0a<
z>Rrg(3Hn(kINzo06|*t@x0ylB`flh@@A?N=$Q<X6N;AB3@8;$?&QRHg&%rLUMomRI
zKaeS$(;i^T?Q`YcfUJY0xHJ7OR4kxKI=j1_P_;^SD4i`*^H7oYmG_X|kUN+LJJJQP
z4Slsy3^=36^mX>ecQ?vGi~tx0Ip1eqda-}eKG=9m$G?l37;s7ADI(`z;<6eTYRxp~
zL06JdY-ev=8%ZWc38_|kkL2~D446sOkVUNkF6k2{1}Mj?y}$Q)4ft;4))||PfgkO~
ztQ$pC0+7Qk4j+=Y6~sP)J(df-J(q}x>h!nc5D2~Vc(Z0-E?NU#gmRpCBM=Hd2lfhC
z3mcg>7L4RA7DsxpOg^E<D7U_C`vuazje_=aT(-X9ipz@TD}iVmbj*L(x#A@f5B;md
znhDYai>C8}(p@6y0?*#O<wh0z!Kn2W*qrwq2_<O_C}oH)7ix;Mg6Jf09YBa6^wCNY
z5y@G1fnxs%XyOADU?{K&VMXs?bmzqkBD%hyo&gnjaV%i%1#abcj1W|j9C%4D5%e7A
zQ(c8R9)Ys}+|mQm=m(lhYf%Ot1!8=ZmX;PV>pOL6`*zBqrP-*O6cYF)85RnVKSnw4
zrb#aerb1PC#IlKW&TuqYnr;9@d#wGZ?v^AY7IH7)Kt^5WQRcJ1Cf-nGYN*5b$oJ`%
z`Z(+5FNXTdR1Jr2|2f(E6jT-ofP_ZgwmEVg=vE{o^APe1buI9VipOG0Ay>w_tV;j@
zhScb}PchnI1z_%Kd&WN}8$GLl3vM_*tvNG{8j(S>0VrW+GQDF|Gj$E-nIo!nRkr_I
z&d{v~QC-lGU=nLK8h|g-^4E8w72dt?$O}H5@_UnITTzdk#WndF4O3R$`LSjw4;@lR
zf4XkY+;Iu_BnVU{4>1zSH8B5akoSFWMX1&LTaYg%LqsX%ec*!j*%b({>%>;u!maRV
zIFVDi^RjWvW=N3$4YGO(%jM3PVNE{Acf0$$iPvepBI3c21g0`Q4SPCOVlIuQ8%w64
zmeJRcjK{B-1@unU-ul{<t-o2w_<P93w#*ZDe;z^EHa;_^r93-7XugSA<TfLD>wPbF
zt=J)a2BQ*)ca&hWfcs?l|JY17)r6srRLHs*XmL+WgUij)HaxQG^@I5MgNm?;#3)8?
zU}H1dU<#z47$zW2!ndk<qgWtthw+p)wP=QC7?eJg1J{!Buj~EB%H~Gk(ArH*pJq-H
zDH2FD+wOzg^AJ(^gnmawmGgS4#2vrsnfQ|K45|Kwxquq_N~E4p=M_-zGTx&y{GEvU
zAVTQQGw^8#=5^lR(pmxFijxfsSwRYO{Lp6XXm4-dvQOUe5?CUHQWigBDFLBM=w$k`
z=kD&>UkB=~ThVxw;&E~#B;d}<FpFko>&f4)n4MNowQh-JZ;0$QgWe{;r533oRU982
z#z8001r}cqP8NC!jJg}Uoh2}9)F|RZP|Vn0<!#p@^2&JtCd0(Q?yx+$omf>!CUQqQ
zVnJF6PPpapr7n;1DsS%1^9-yZ*ieKG?XQhAK|{zaaQh4NAfQSW_G@^|e7CZ+q$FC4
z<b(%ZIQFj=zyoTmdkaOYGtGv|2+a54?k+S!ML3X7JJG#g&Zly^bQKX6{OFzQ5;d3O
zD-WJz!(9(yC1!m+>oa0lw@#;_!Y@+{CwG3_g5&)+X;Co^xS;$5toQ(?RRkS%gXmQr
z0f6_1TO@!fOx_%V${^iA4C6|Ae}6PZc)?wKSX_LiKlYq6W7F1%VoRPq*^-D#`MmJ)
z<L}K|1x3t%e?bcU6<m54;Krra@Qy_6PJI_%I_9tgw-sigor5lL;#t+3!|W8Vy%)om
zmY2?f8|L1T7<Ea(pzhrTc$`Vn-JMhejP2wC5kZJt*dB`^J@K8Hm8V^|qXHeo7<yb(
zQHr@p;9o)4Ju)H(j)YP47X9b|%{QE@OwjF{8#I#C1@K9(q^&`yE<$9F)R%D84+?4Z
z0AaE`g0MMzTv6C~B(=xD!2uk95g`0c0lIRe^5R}(ngryBQhBUfPORVh^(3%e^h5ln
z#q_Yg$pX?fA49*?oi&o%-PmE@?RGe~v*Wl0VGXeHUxrEMn20L8#qCeTb{Wcv7YEVx
zRRq3q!4N8H8fZngn-4bBb6%jMXUs>8KvTvI8oA&Tp#Y<@ZO0zscuGO%1>tSgc1_z*
z75g*BTO5!Wp_*<0*)SbYeuuD3E{=#vP#7#M%!Fcc`S2NL8{?Aaa(4?-f1dwsdLEF_
ztbN&*++e<69_8D23lA)hI3Uov3-G|`#yy^_@d}YLOhFt+E-ONytxzX!KK51unMb4M
zRYBab?`O=+aTvo(n*^f}g0tdUHQ*3m(W-t1jgEP(ET^-e#fyN2(~!cD!`sqzQ=Juq
zC<Qp}Mjo}a1LvNzX)R`&6F3D#^`<|aREQG!i9vgaG^ouLrwcQ+F&k+Q1ORt$u8uoF
zJ+BgMg+c)3k&60<a-YLh3_L1)Hx?{G-VzTPL@gnxUq);<m4`PqHz#A=M_%;%1ZJY?
zqZS=>CZC~O`>{t7`owo`L*G()o?X01B$O96a}l3U+V5t4#PGst#+&(Yu=Spa$&d5b
zch=h45=BFt{R|a~B0zDK1n5FkT1wg2=;(Fes>C-rvYnM<S@#1T0*p7fqU<I-&+Xm2
zccwB}E;aUYe|tO&zH05z`_xn&pFA-w$>+BwqAy|f%SYe6LA#DsFNv=)%YX-0)^#8>
zL`QHmP6cc3A-t%t49QH6g+g|kc?B0J8DR&ZR8ywTS|_@7pDHTc(OL7(W{phr=vPY(
zc-*|F(s<F;yP>ve{J=ll^jPj|e}6t`I7KXmEAVeO1xD^a$|h|48bD_k%wbUGca#fD
zw5Z@Z5iuWj??}-h$Nr>}2E(pYASq5*;K`Pyo-0(0<Ybv6*!K9A1%bOh^mOYP#xd0P
zgm=Yq|2nshrG0i5t@QcaY%3z|uxU?u<DlAzQLcm?YBueqKdD3#iGe3@W~wIslMvvB
zt20?^{yu_p&w~c+=>JIgdB7kIf`SO{((L)N&6@ld<<$L@JXTQ%BI@Z9UgmZ<%&&ti
zIQMqzD<GvDjJho8A-4=(z!n7Vu;VUCoX@u$XLS*ZvJ1l-v~c<fHbFGYqPu5!>45>)
zU+ntXmM)mF@ecJna&u}V)+kD2?htT4l7@~4CF$w#$uTwcs93yuIo4obLn;r1ChVC~
zBDi3Q92nja*abgK--oDaUlCWA9oSq}Ys%h9$+1eYWnzNB2<PibA}N1=)ibNc(!$}g
z`E*LvzHlu0LM6-RfRFchEx&BIu=H2>FCkyX$=@K|)jt9l8G|W`cLLAf|0Vne7y$On
zHD^YD_SbG<-(#c_zS8*H{nIUp=UEiqSG6A>{8s{CX~XQ%Bd&7VMQCo*{=fd3|DP9s
zsSjkDOaF%j$M)0z{Efj!@MZq_?W86D|Cbr)BQ4qv&=cO~a-cUm%+265uBK=g02aV4
zp64~5?E(KKROr!H-}g8tJpIr2FFH4p!%3N+9CQZc3E|bqp!X*i=+;t@wgB494^gN{
zC7G+f%y<3Hoo6`ow(eJ&slI$+92Gb)^gic=C*)`0$3O{*0^$(>f*Zj0v5Sb;6?a50
zu6o)ri2R};!GRHoh~e;)i5`|n)u)8R!omnEfgijRiXogUmr}GZkUCFKPw!y~TqB$4
z=;&ayA>qGA?m%Ni2x&AIH_&)sMioBvS`_CNDTDh?+F{uMf?ek!SqMN#(TsG4L~sO^
z8xFnWHBCbL&*4doLSntA^ac@_Sdduu0$d`#+(jJJICiKYwvxFd+%nbI+35%@6`kKL
zWN70iKT)dXHC_?^v)NKJoUT*12V^`=ClIbEXt`lT`c2dL<po?qc+R&Fcg&u-Utbk<
z8AJ>({x)D0sbzxihqO7rq=XJGM8xOlL+*<<1rzVwz_1T|2F~FJWGh9#b}vEv!Q(@@
zsI$DZ2&Cx8;NY8|!Z3o6<hedfUVuJF>(isvC=8JDua9#13ofbC=g+&e^8zjpy;1mt
zjJko+?`%0chH2yM>qJL_X<rn0_n?!+C?OgItM1;nZ{IH=DO*o8S%vVRDB6C9wi|9z
zdQdn}2<zRJY7G;#QUDz`j9DCAnXYn*ut(x`UB?q6vN$vWHIR+2XFn+e36`g*CnULk
znc+8Z(jqOEkxYny78@e|jhi=n^^IK+<>$5%Z0DH_BJr}P4qi?9X&3k>tPYIu%4i7l
za=iYurW!^OsxwCEAaMd;F5i3sn(f!r-p`*`{j++D&c*-7rqjBEnK_o2H!0?j<41Ta
zpvC#i3<L^w06GinF}!pYTo+{p1*datXZdPO6V%c#QpEx+rJW}hP0A!}3Ij$c5WO%6
zPyq{50p;5odEh&O$Cl9iUB<uOkH@<IzyTTK4!o=_gy8YfTP*6ugN7*xrHG;<B=jmN
z&*mY#Lv(q+cU)W?5zF%-O%_rQKIZwoLl9;6lzscAfL$qPhj3;rLn-bEbSM#&dPIxo
z452o?jODr?z4QP?XRfbIK&<K9x)q7Yxq9$!X01?RT}BnRCFB+6i*)|`IMW$gXzn6=
z3`Q2RH4#w+xIe$4VH-9NjOTy3SCExWa#l!X4(=V0Sr;KrfmqTelh8I?6EL7`L>e&B
zRQqb$m!d%r-QI}Lh3$tO+>0`+)@KOr15}w5GT8|PF@YY@rq|}$**FQ^qd=t=i3Oou
z8Nft-(62HHXiC>r`fVAqLz)uOwdtX7eGF8E(Qe)ZLK#5+@Iw^gnnk7%NpAowlHi6}
z6HseEA*yA0)R_47f3XxoFF-=R#omYUhDbjU{ZIt*0ZPQs_F0@p4<y5WU@l4sNH(nd
zYFtWfOL(BXm}-OYHYxOom}4G^e~;OhJf$)sjKMUcvMpuU<9hY6TS)-}mXSDjF{=m3
zAhFwhJq_h}j3Tg3(ETtuqD=^^(PI?46L4o42Mjqb@M6EEx}swrFF%~uchegI@Rn$J
z5$7CLAj*9?b_;Ah=wJ4h&i!@c*j5nrD5Mts2s3WKkC%hTPnSJcv2GOIA$+2R01=_5
z`JiEox|aCvq3`a(hI9>5R0~SBN>o=@r$V!s***oxb{xkKm#_+P7$LJMf;^b3S2^Qu
zkIlj-9}0?)#7U!d*}{eidtXtP69$K*G7C(}wrD}m3u28G?CHfzIYm(<RT`^jugvR^
z^c=|DN!&glZ;q%NM0o&goP;JoxHphyh&5(wOMwWJ-}?5wAV2RDEF*36Wdh8DRMB6j
zz|3CxRTZpxP&4-$m@gTe;0ihwuP^z%c@qaA#9$Y@nEgAFE{->BfU<y?*alkD6H}6s
z<Z)+R!eO3%=su*IYve&RD#0M2f?i7rnk^;lXR>($kC~j=9%onB0MRd2AP@0lry@=6
z7t9xwqi8C80LwznWnd1u7TH!S_rd>|V~FPy!H2YLJXuA^6%-L*<-`o{OWAAQF!I~c
z{7C;fsc%C;*Qi9kf(N^p2Jlc>{1R!ZScDbr2cWcl8Bd)DZ_Wo8@=+|U$wCIPeaM`;
zjBN3hxL|Mf539rA6pP-Ox(=ez%lJQ;cJpZpQZe8j_MalA7D$$2u)X8(oA#aBx=;QT
zYU}eBl|V4|owBmBih<Z6&Ny#T8HSLn+!0E^r@^y&6L4;zCd~QI__!ii(pUgWYvhS%
z@y`w-tX?CJJ#X2ac5k4LTgS*K9!G{xdV0G5NxMvg=9CC(B^**nzzshn$k`aA+)@U@
z*UVT9J5ULq5GJ@hZw0A%%{&dbete)#J_?U~DwM8A;awmeM{)`^PK|v37>0=#;V>iB
z<K39If&q!99L1{QaM1XxhzjN#>K%o@?_g;Ypjz@l_z<S5eH*`ox(?L7QIaw4nhVb+
z&cCC(j{AxaqkWK2wI6;T;|@eNYzYB`@<N;Kb5uO6o#ch9f4nC#J2Mj=O?Tbab@HBd
z>(o9`%Hvm(VV6*u3S?F_Ha6;2_0M2OL{U;L$8Ltrr*^#YnB=ha(%pM-lClu8`qYq^
z^_Z8C6gRkU%bXRj6r?a8{0dcBW24I9DX4*!90xG)MPcAFb!kO~%rAsDK7R6K;PWu%
z^7^j$#(eB<2(=QH&E!o`Z|^8h4x>m)R6<=p$t<C#7fo!RD4LYf{#*Yd6C29v>gqyl
zwLYxQ$MrMF83d$O33~J`Au!~ax5i;C#Jl+TdLWDwWma<&F=f7ogUJGf^r+5ANN|NE
zR_7XdKt@6{kKvWSyE6JmA#AGpUe^IGst_v3jr~8KtpwROj?x2Eyss!S`f4Oyo$u(;
z%NRHjr*Ru+2d;^OYEzPpL9bfMkXGPse*~KO53#T}i8w4}w@+b*?mLxg+@WEdHML`6
zpkA6gq7Glh8;dknBPSEi)e9**{(i0gxB}w9r4F1tZf#}7$xNmmY(-)^nNyAFBz0)x
zy-u7%w0kTNPw>Kl{(L%5iIb3Af*{U)?qTW>7Jj|WX+TD+Pem-vyDvb;7z2k$EEt;@
zoJ6q@lkJB;(i|lO2P!)CmK*NC#4bt2Sol+v28LR~IPl_;5|2w&Xil^q16HYG(Ym?5
zf(C_tWi(D|rnZrvKf@_oSy|tuFUkh{eekBpY<6=*LS$ozVK{U^6l6I!7~vPP;cNyB
z*#8;h@B#>~3xzZVce^AakRSvp1Q*wNjfnFhf(7e!%2;!u)DFJj=s4+8PfqhzSs9rx
zIH7SgX6|~6Yi<uzg|HECU}i1@YGdw@a_g%U%2Q#$#VK4WLon#07P9VznU$Xcz>%V1
z9s!loK8(eJyHaW-0E7@>V1TWu+n#Ai)MPjX=?27s*7GP3R%Q<v#6@^O!oLoQ71m?W
zuVRD@lF{M@>w9q-kqH`(_@8zgjh1JvGUmi#X3-WbF)~2XIvJ%wH|j}@Zk_i?Q0F<A
z>N7(UZsg@R;25IQFh6a-wD9pxf$vP)Aeix|=pXP!iB$n-%nhi>aElz_y|6j&Z=HGj
z8<aZa28I@Y2E^=DXwFgYNgrvitJ{LthrdmP2&C4=%efGAF03Yru+m6KhZP9f%&Slb
zML^p_1~VdW1%`t584=0nJHDd)ZOSm&1N8DUVw;GUgoQ;K1_b;EyY7_FB}0=p+yS&(
z9q#p>+NU6ng45;cQ!=23cHVxjpNoP<*7E?(b-X@WH*PjjYj4bqgWz&kke@}tgt!Xj
zh{&Oq@84M+A&#IV-dV!UNrqUUNvL*{4L&JK;Go4toW2yWm!JA(@%=KOHX?Rv+)Xt`
z7G`F3Znc$i*Ke2J8aipTh%wy^*-}SxPHtDT#d6WcVO+4Nn#lF?!-ordji}GrRANzk
z0s~x8E@fc<yl5Ce>PPgryqN+vxPK>>AorI<axDc1nQiwR$6jG0iQ~>1=xf&9BtCQW
zQRWT~wT5!iMw|q6#l3q43zyuH!`Z5kGYcHMg};+2yc!{^#SCH#j+iv`wCW&fFUOJ%
z(O;l_>q75CvIf8!_T?@xbq@`F#mW5<`O5_MM(qN{<|7y}IIM&E`t(}TjjM2y_G6Xe
zX?#XyhF6I0!|lx*7L<uXbF+NHxW|g&jv=Ny93ULnp=6FWC<zR@EJmK1XMnW4tv_Bj
z3C4j>hc<9$`&RwuyN|+k2W8L{nI#38Ao1I;CeY}+tGJt2Quz7bpDo`9za5S~T{xW}
zr6YC|2oNf905YMA!8L<O#4Cvb3}Y49YxQ?iy{=<ewN##~dgTho@rG~TzV+D~o!N^S
zY{oD|^?MX^eQ0YErd!ASC@ah6p-BsehC<i$JE!FVHjJ>asKoO6mglf6bW-=;a3dZ!
zmeM!QRtl~wy$=B>YTpC;SyAPSX6Wpo*K39vWp{NORRf-}qe$)&wOOiaS3f6_nqx;<
zULWWb?)brDCGAa(m1nnXJiFmY!sCOqX&NkR&u$#GKJ9S-0h{BM=;@ZXubV=eo=OSL
zFN*hM)AR4#`Ebm(;pRb6n_#;j;onTRf|@t31NT9+A_(+)@$%(qDXBH-F6M}il|ph5
z1zxcgmoHyFcKo=?K+9hcf>;(2-z{I1m6gC@!He*dzP`Q?AY{8I7Z(>lIzYNl0sr*?
zoheoxymL+pi6n&{-;NHJn%Dfg>n*-FH#0dbjM0$_8o-Pxzw-VrY=PH;M$C9ZPp~FJ
z8Q!5EA_FpUhK#{W89JTuvoRgsYAElC`wJ^2{*|GE1IOhEv~a5!7)prtsnKA^b7zDg
zU~VEiHZ8YG?hTZ_jOfS_CXsIx6dM!ciYuPDgwD&z{A`JF?X8&pBhEo77r$^}HRbI6
zyIap4QOf<fMU(0@b&uMUI$w_SP4xh3>o}w(v~S&dWi9h!)s}0Ds;YBu60R=T>+9*c
zct=G=iQ_-S9WH)oi3Jh+bBB_$&U{(=iK(a21Gx&<fIxxTMcV!oNQF)#!<+{^>(!Xv
zwmb)REL9?bL!;}4n}le<E;F+br07f4ld4`VxX%znbOzN}(~9dRunQq{W=}G*F~HT_
zg#9|A!NbLcvNQ7o{phc?_A6Je!~~|d3MS_yC8=jTuW<{jN}M}K2j)=%7Yw5MB<}O=
z%}YSeDSSHXVf#TNv_<s}4=3l!eBYR7Xkr~PMdV9GR@vL7^Rb>V58)bkj1n55vNZVR
zkC5yMvWCi0Z7%kJ?MZf!Gl*p0M6AJEw{C?P7mtA%E+)scG(prO2ET>Ja~1^F+H!Mq
z7v$%YT%DPju`qciEqbsL-<7{UOs!g>cv?CW4nNBiJK$HkrZcuW@`Xmm-P5<NBpOZK
zo<H9%40ZJ$N@q&tLqm<M`?A{pT%*5M=Kc)0Z&H@23AOflIl113W*)bP52;ZSU;#hC
zJ?rYNtF7%Kdt3Qv!R96S;dblb;5rKvKE~Ru4;k2DBcWuj$vq6Bp5lnka24k=#XFPD
zT0>(8?RwT#`uh6h@d9ZeIU@iuuk;`LtCzCh2E61nt{nhQoM{2Pw4KYk444WMiW2-a
z5@S9N_(~p$L(0gx@OL306k%cEo8@I;@6yT@FWSqZOUVaRgQVoGnCEf|GmGcpsidMH
zf@T$Bh#=Ou;-l_6w<kvG@#Dwjd%$;rn(d>`CeR;rlwUiv73JkA7_<wK;5xj!7q4Cw
z<0c&&sOP~#pgHwbR5#Z*crVo|3Ko(0jT^h*S~+NTV*ma%n9Y!nT}DNLROE%1S0SL=
z43F(A$n0T+8KZA?xxL}fK?FpEgoi(ZUb-8X`xzY+eWzhq_|es+Fgyh>^!|y+2fR<f
zMNw1I+e0xZ5jHRv6gKxt<sQ6ycL3pXpPHIx;*3wibxffURT0a!2hb{0G>7f1R^=94
zgM#9G^Z#P+&7-;O+xF2*BtwcsBAH5M)}Yd0j!L48naNOOEDbVdh>T^7h$JC%qQRIk
zp$L^i6bVU#%=`Gb@8^Ag@7`;#wb$PJkNwAPt>;;Hciq0%_qwjnIUL7v9w(qQfM7o8
zucERZG;WoVIe??cwq{M;Z$=;RQDzvI?>^zLGu8DXD|0$Hsy&wGsjpldrHMYBO8q7#
zym4bCjDq*BKbo<B+)5musGp661h{|b4X^n4&|t9L2NkpSu=L_u(pj;hhYus;<Ab1?
zfIOKo8s#HDq6c1gn~GS;i0?h+;c!`i0%tF4W^!U;2+j)Qn`PeUWCbK7G;9M1GDCU0
zhnsc-%F@!*^af_3Gc3*c?qiw9lX`D&41j(D5#9T9%A%8ZQ@I2n2*MtWgqz(VBSW8X
zT#3now`4;MJd+)Ge>UF6(P~8HDj+G@@~oK&2r3_{7FFP>ejNoz5mk+h5?crR+uOBa
z*;S3+V5rbpSC>=xXwq`@*r*<9GP+&AsQzhFr=c$s`-{z83P9vAakmWaXk-IBIy^cS
z78VKGEZk@k>@b&S+f%j=DRvqDJsHu4EW!WgO>TI0Wo<7;p=^%og-iDA=6Q(1@d=gn
z^$e)w6@j`jF*8e>SNY`@3#lmr99fC`Q-jY+$;=E!F-4pL0c1qCwqHR!CX5Dg#eg$0
zmdyoWXx9>L1<|AG<`^{`Su1e_2A92KV_~ROs{s@tz(+(yL1?gvlxh&dWj5`gzm@>(
zB5d7mU;HH@1SXIr^)FsjBO5z|2?wE9QB{>{bjITvps7@R4}>s3ScYOxtTRp)ID$18
zm#o7j(rAxbO6Fz`i~2_1UYO}w;FpwKiFK*JM&GYNC{qSf3Op2>d42x=y)X-%F(&23
zQDp?+jDCXscK}!djLmstq@<#`In15rxUOz3DwkKlc*qPfI8?dciuV>%qA5E!Hxsxd
zATeer3?cSYvbGjR%qRN}<8Iz`*&{dn?VBYkb0`L6frPyRM1b80pKz(fh;hH?>{aO2
zm^`?QFo1)Ot#5EJ5WVse3Kk)uIFXisUdf6!(4+i9LRS!xEm2<MrNdj_|E}cnR6%QV
ztZ(Qr?sE_xCk=I?$d(2J^Vq4F0yyewNC;m3r2sZ>=j4O|Z|2{yfez6LU-?XFN0V^*
z3cMGj6|;W}JeGz+bWs$_K$na<d<BVb##$IL+qk`7L{5$mtGS@P6+WMhl#iv<XZ&aS
zkq7)cI-bJ0(_8Z3`pN?<VTGh<XLG&h3h;Y!O?Z0!IQbcv5kqVVM3ath?{eh=9qPno
z_yyA&SG&0%Q<nidlhlnUa{zW~TKf7|@pt!n&T?Vu#0T}gcjG=Ak%{gMx9pqhiMt|c
z%y26(=!1pJG-kl^+3rhjPCy21{ehRSkda+j3aS%fA)#v6KXyU8(g2K2<n;k?mxaZ}
z)*>Pz>W2;mfBp8Y5Eobochi(hWQ3UN!ql-}*rp8|q^PEkA5GFmanai-F>FU4`U|)q
z4uYuTdU`fnw{BI3+3@f0J;`tc@c6x6x)vJ=s@GC(-#WyLQSuPDh<rLW26mwfd5im`
z(u{(V5yA_uHms&bwV;0d_zeRr1O;;!+|G-Ni?#wQRz$e?L^=PX1vtA`PA>U@L16#@
zt!=<+gg0%fBdSwCWBMrCO2=a5PoEZnwSBohga~m6#{EC1veCcT14a^)l>DAG*^Byi
zIW^NCE>~h1{Z;gl!v_npOH@heN<(8~0qW-h^zYa1rl;#pS>^_nRdHF888|#f@>h<4
z;KNtLfL=U`I`Yl&2iZ96Vi>2yAK>o9r>6_!fLJ5<S^>CL$4j$8rk0C*PudEEnHRh(
z|0!yQk%;UOkh0;dsIH|I+3?~;Av)MRbR&6DQL7`T7v^UqUESQQVY{q;;6MQ0i4Fk0
zoak_wn3(*AnLaAzoOpO>=mO-I2{=r4(6D2Yh}_90vR999f8xH26UrH5s3W<&w|XHN
zl}zKFi9sd?Gls2JR#w7WwlowK73rgTH9*Pm=EJUu!#GElR7nfV66F5n6mFkl9ki@v
z@Ax_&7}mYXRr3l74_`P9mY{RqVbXWD3-#Iy{OGu_uobcF+;W~zyNrVP^;*Fwr1T?w
zi(!ubq3tVQ6h6+$2`ks_(j41$$h}$nSE<&m3n9^QlWPC6foX^4;Cet-yY6K7k4RR5
zJ0u>4a6uERfZP=T0sZ(NRs|hD5_qiNGjj*^FQ+&_d;oZ!*VLpS0`A~t65wzH@aVvM
zNzqd{M4GVB3*r;UAJF{gW8@N$Vl2`7NJ&X$eAQ&QA5m4MXnXo}j9sXUqUUxLM}9YM
zumiQXla1nAa)d^BF{~~A-@uv!$G!hUWunKM^4D4tK^VAy{z`-?*T(-`75O(){vMBx
z$RBHaU8dLe-#^4}_VwS!Mx74MP-??}|6BZMM*08xNvw*V=28E*|M=<-C^T$OohAcA
zxFPrQ@>ti8jY(~nszCU(lU=9i$%G%CMvraYRrRf`bNlvcBDg3mTdC;ZQis;VPPXXK
zZg%!7ZZR<#5AH6_$}y&)U>NSplx+ItA{=RaW736@FS3Nx$puGzx2~D^tl9U|{EU@O
z%Fh)eJu9D#eMr%}w*naU0hA1R@Qzu%dNsqFJ<gx#wExZ?^EzAfP~{y0{Vu?muQJS)
zP!)t?(!pZ3#2EtXA3OKyPRtP48y+u#IJ!hGsAz=&qqv>tCEB%hUX^IwIrhH(<;$7?
z<e-O#*Kn=nR?OAlEpqH!0c~^*>JYzSZl1N#Z(DT!UdBKJH!Lyq%gt$#RhO<je;zC8
zF2~j#^*obnT7E!IPcLJ9imUId!Rcqgzs#u;%s%f8T&a3x`gcl7PFMg7cZJ>Tg{Sb&
zbKjqsdp$FU`DLOsEebUyGfh@Z4H_B8&TV<Hl11ewr?|3-SwpVCwr%X>mcl7KyC;uk
z)hfC3>yxDEfq}!C(h`REB=S?Y(Wv1CWX<H^;|s69%Ag)E%R!M!0q~y%+&*nTX0Huc
z&U<3?2RpgRm$ROnAFF6-$*mfE@atza9ul?%1%l9#*DzTR2rO>5CB1`Dd06Z3Z>5iM
z(kSd5d7Ofc$ZCcKTem7Mm=@Wx&FbYSLq;v-Fnb{*GxJm5m1P9$0e*4i3YHInALa;)
zpWSa#?it!}Lzs-ca+wJ3c7eAUrVi*)tt_$3ZkGZ0-8V3F71VBZbMp%DtKNuI5IQo*
znVOEgD`m%6ga6~lvQ)*37tCwp<1OB!tz}!myO-e%FKWl@*H>X~2fT*2KQm{=qg5>m
z6Ep9yd(*l(?#G);9HZ5s`~omCRo~FyqxAiwZF>eA$&q;eeFOK86E)P?;Ay9GS;chx
zTTr2{BdxmY;LAI7y^^P2Rr!}-5*stpw=**I+xjjSqW0}Y4o4~Y9JEqtzmI}~;5h^!
z@7opf66g(*gbtX2-X$5FblS3)F1=Z`lbmHWo%)uRx*=&xO-x?Gej&SiaHFuW1+-dZ
zX{tuub9n=UwUm@yhgb=?dt~Jj_X`!7w#t2jsanDqn1Q*Rc;P7e0gK<6I;{_mFWYqV
z7QXiqy^b6YclZ#P1Hjw`?m~r6Vk#748|w@V3<L|K#=6Tw$BK)|^_I=eyn3vYb?p7F
zO9Kpmxx&%Mo7Xkp>i!|16o|Hy7Ci;AsI_iNWTeKEwyN0gL10IF?2EQEeRWpbRJa^A
z>4u`ZGi&O~$GW@um_T(;cJv!@*e`{V)qwpP8X&UI-#;h%E^w1ZoHtUXeNP$hikf#E
zMy~TuFDjajY2QnBA9^J@gmZ|~_TZ~j{rT%rQGN?^-@V8>7g@Jq?Y2JcApDn<cW`i{
z-Zzscmvlyc_+R4`5EW%cnR5Si*J5xi+qr-NQw+*sft_z%fAq}LeSLkTluc{6Q98g!
zsU13G@g>`U^2U>PuETQ2<rF5lJM`Xb96F?k_U**Y1ZMQ5w^3$O-Znu&pTnYOJ<)^&
zj=nEWLA8y+ySthO4kqiryi5+i!^S4K&GXP9v0<A}`)7*Q{Hk%;(5#nx!rySGqoCPK
z?#)}aFh+-R?Op|lpi5KfAjibdNOXc9J!aqpMEU0~Z@jO@=Lx_BAOXZP%qh6Q<$U|s
zVb|`i3*-D)50=%drEL@{3@>q6I*$E(-;|VofTd2c316RaCySZZ!{b+8VKq7)<>skh
zU)il+CI(p_Nl-klL<?@2Ylq7Qgd^qQ_3&{*6N<3h`^9)AkIZs_Gwhi=VepK2MBZem
zRb9{Ql-^O(m}>rfA!$z_-i_8*9S7zv{gz6#ZL~ayu|hs0V<R*BlXNWS&TT*4#adls
zbUbl>{rbH8VtT4+?wcfj6q$<Fn@90Yj`7QzGAd`p#fNO|?%2gj09jwA7bs0$1;lS>
z%(kTb`T6tbD-;>F8CgOS5@T6E((c{E*862=PCN=g*hl34p*q<77kpRgj)NkX_T~QQ
zPpKO0ZytTG|8;Pg>$YvkJ5|rSxS$cP0$!=|3DDB>l6=coSe}cYS>0MeI&U=iahlKI
zP(m^eYSAD7!o&ntN?IlWQ;)Vbr-ESQeznL$SeQOXLc+E?gKhhc?X9;xJxYF!UfU)u
z9UXQ@Ou@&8<bZmuJI`YdtEu&(UU>ZeouZ!Vvd6NrM5BQ%MkWs?{iOr&vnt*-GT~ph
zzHV1O?CJ`|Co_o9f#Kh-=ln0eS{((H4Yx|l?$r|Sac=KNjvtWhh!-M)`3=KzrQ6^k
zAzN7r@2Z-d6nXMaPYA72pW}s;cz)>hUBy8r8SSABUzCuo#8YnHCL=n|b%%th!h(YF
zu@8Gzb8-eJI$BsPS-W;EDtP^qm(qJ(kE3l7T`f8F3_~rerlzKF%o5ZL@B^ezy-srx
zw`$MGNDg|Rq13vD$K$S~@P-W)1jtLgRa4sr0JpxeQBYQPI_n4P5(-`Wy{@ti8Y<J{
z&>>P3)F!O=Me&+%b9EgU80ZD+6CFHb_{MgL=40Di!GNDFk!rc5ocLF)m-XW)E4vyV
z9*E&*Oe{3NA#L|g9v)&*OK6y<Q$Jt5__!W<+&A>$4yYun-WUS|?*^{&`sZMrjEAhR
z@0C6{Zr)K1tH`v|!)S!dIji7LM5XKSp)3@rhy90Q7C@*(8T$ph0Kj~#Y+)11b6oI*
zv)MB6F5}#CCPP0)-aTYq@$8ZSuKVNhbp8vYCw}osQ4mJ>>U8FmoM;@{<a4U+UbX9U
z2CCw=!WCA|_025bkE*H3<7(k&L;>UU>1m9P``Kw)-|(`gOWJZA@;zD#a9YRQ&y@h-
zi0^nxlS{?$n-cfI6~uCKoAkGU)v>@tkY<F;exE)&)Wwzs!7WL!>FEIm+iYCE?b7{q
zX=_0N<qe3eAFAQ;-#@D%<mkg3AChLuqbpZh8R|)=9Nv734eBp+uY$OElF@Tb%KTve
z3DV$J{r4X#$5th-<9+qiVB1JnDZ62@gMV7oYT*kP%~5?D&0$0VEh;orG-zC^hP^4E
z^{W6p`2c!^Xxrfau0|Jh{fL@(0G(RAaO`>Pj{<<ansMc3G$^H<$3_ggCP%gu>HnO3
zDp?V0{(}AK)7$tOfX1r5-iTmf91VVc-tfC<cUoCk;KNa4$=`_s{9=38DOjY>bFbwQ
z!at%F1Tv<ie_>ZuRki<cm)rNSr%(4L?mw4lG<ls=yJ^CV>GP-U><}03RSlvwln2rR
z!4TV)8`x+H;K}6K<sf8dY9X20d!d&)vE7lpN?I6j=NkG(WK%}Fsf_#g&7rs;<rdlP
zi`e|>&5KwpT&)CleLFilWVhFxPJ^n3=Q7{P0bLbZZILX@^$$VUiut(^EXLh?=@$dc
zbyW5Bxn?hW5Jz>)SZ6oMV6o#;0N;1-HgR5$ja4-~d^qQfV&*ZGX{mDC?UPW*JTKlT
zxoO$$%Xdu51G$Z9HKmm2(T|sK^1NW-a{$42JxQ((L?|p6e1Ps#e+$&cAdiF%b<4&^
z>NYYIX;CigWy@pJE|9Nl7hMMx4GnqjSz7o7G#DBf-U+|Kj)KOAU4?PI!aBbFmn-hK
z-79^JY(tEG&;UC3RW03q>fP#!*idHZwnEr9Y48I~ICkvV-`S~c=@;DZ9BfM`#ZK&5
z8t<~^W@2QHA1Ki7c@P?m60Pm8vlOZWU>Dur=ulgeV({||VJFBiHZLFk0edH|g4KiU
zi<7$j;6W$`f|sw8qENC0!Z8Ol6dfdJ(08uCA$)Xx_`5Z92_=}M0Sz0nzqz^F8E2Dl
zv6#fA0FU^Eh92?H*;LYrB?-GO4=|*qm7T2&yCo|iF3tkPWAP-6jjf>K_MTb=pHU~6
zvZVZ8hsG^n#prv>KHsI72MNr~6zOyW^e^MkeYH$N!pH-yC&`gMI7Y|L?iCYT9wI&R
z?peb7k|k7t`=zSt54@Ox@FySLzxgG^)8iv=w?SQ_kTVoydJOoz#;yJ0+ndTz73rh0
zUDBD|*YRd1X%bZ$J>0V^osJ&OKJhR(YY*foaq;n$@86T@np#wWuVAq{Hgy;NH|;jp
z(4_}pmgc0DRb*`Jrs5<F_TW`=hD0ZYYq8DqRGt^r`}F65A>>vO^-<3!fUpPR&EBwa
zV@(Y6JN3hdub}ve?R}3#3VUNe;Ufv35M3EoGFQjEOdu9PiD5x~>9ZQi(zLRVAII%K
z!;SO?E5g&KgOPH^3Rwg`G0=CeUP#nlyLbIcP$8ArX;U?O@u!s6$Gth(fL<9)o)k9?
zwtz}z#YSe}gNl0U1xB~=&3B#uBbK!P;>4R%4Sw3nD$`Du$B$!mST91<g{e>et<!(d
z2z2w`3F&cW$|Eb1>TeWjy^{HapfnIR3aAyB4c9K)2+a>4k&&Io`GwUJ8K|qlA<7O8
z5<vGh6&wh_9f5HnK@%ghv+c?~6&G@0wz}r<54i{wF;ZwP@OO9e;wNEW^x7dS$8aEF
zeSq@1!voTV;eir>ro0bDI7gWr)!Tk=CCc-X0aH)U{RM@EKz#`c`SjA%Dip6q9ss`x
z0kH5sk7kX8Q3ai(B+Om-&(2M|u)O)mzJA5VjT?*XdQ67$pL)!2hN-x|y_1?M;_(cv
znRA1jPtu=1l((x*%aXqZZe+scRq+`^s4E$2z#kBCJDzgiNc=X%(al(AQ|f4%V(B~o
zt3cmt)ku^Z_IvpaXQC<6Kf7I+^Y^{RTCHOj=dYePaUwW7|A);h+ZWqZM!mkz6x_2t
znS`w;_1{0ftEI<*-K9Hn-c*OE$a-HqW?XFN<q90*rwxq^?C6=&$C!pgc;wat399)l
zs+ZSQw+x;%T=4bA`%9?C=XG_K9{D?35fr4ItLTdC->$M-nCE#?U6pt8>oJsGckl-M
z{odmx%DZ#wx5wXc8hqyG6RrhfP>8@dm=3Lgv{SD<{7l(UZPlH0bcEcE9%7eO(@mz=
z7aF4@u3xVLF#Y0bSbDlJ)$6fED3{@WHX2=kI~R_|`{MN_SqIM#53UCDfM&1^JF`UO
zxJBom;R&UYC1Mgvl5rv;YNxGEp5#Yr2BIZ+=U&>~yQLrUsb1tkifR)+^4ugBrnguG
zRe?}>J0+!(S915P$@q&`Z{Bd6QW}4G!c*(=<k-ih6qaJE+kPlRjP+`Mwo3<!h$ne4
zIOs0(;K2(d_!&_WWX_lQ@~X~Q!y8ij_VEg&q_lQ+E)1W;OIDK7#!}N{yk>K$laor$
zCI^kD=2~(WbkB}HQJMUOV!~GQz-zJm=i>&TS|fKJFYuj@N^arh0S=G%jtmake|XHH
z+tD%9eV<+I{X_Okin=mk*XbAs3p^J-l?A<NZ}%%G*otdgQQLLCT+;W?t#!PI&&;?l
zH_+A73-t3-zI$U}@gcv#yBnS`ZRm2oJs8mQg~>7Aop&<QKvaxh*z0g50;fFaj4IM1
zV*7Kze9nKG)&itugN&M-a7EDT&b_?~DU;rS9`kJ5=m9axcfU;No>5Q?K78<?PeFl;
zzf3Wk%^4@D73+6-gWR_mE;`li90*McmfQR~=T_TfP)IfuM7lzJKxkqB$FaMoSN>gK
zVLTpu#D7zoh<4FRxN))J#xm$X<6f__jpy^gfD}BLH;af^07ln5empC(Zjti6F}FHf
ze3PW)!kfpw196I5&MkfN6dW%TuVm6*(H7^Fzri`X!F_BkU%EO(Bs)WAeqO)CNdPl3
z*HB#C!4Zs*JfwJO1z+>1`tiinRKe-eQggsU=;Cza4Gq^Ie*iD6?Xk)}qgT+TJ8<cf
z+q+2dLnGE}Mny!qIPI*g{GB&;7>eZ#>=SSK&P?6h;TU!}R#Ipn*$a7?2H?WqscB!3
zz+fTc_Svk!VHV|8JxGua%5&|L&zwCpIlBO~;=L5uv}y23&zj1gWP#mg1(WVaWo)d3
znhg#P?)F$z&`hyCn3$}4kg)Sdc_>fh&FjAH7T=AzEbCvssLqO&6c86W{$))$kzW8W
z`!PG^ih*8+=+A?L#KIZxqGYGd#?aJn3?J*5Wd>cH&v9JrNk^VVmrQ{D3x9iQaQP->
z%QQSiQA=IGEHam6&##ZBdD5y4Z`~?lTecxlUfp69pZ(1JkGFKdN2}%PtgPyhdQ-J*
z*ZiWiU+r(|{~=}5ybknJU&l+hGDInZJgOZ0DgD&;K}GJF%b=2387n4Oo9;t;Ab$SC
zYfh8Z8#F}<byZK%UE6p1^eN|WwK~y29N?NI55g=`ORE(mlgOG!Ki7;~#onygVF$D0
zV6Z(+iPBdLm2$*pQ$HBAy?!n7OJBt0+4Tm!+xnl1H}B3cS{+zabT+!@I1K`|*q<Mj
z7JPh9FZ@`7d0_??78S6_U`F~LxmZQlXeBw`EqLC&ruW4h0SOWlyU7XKa1sJ-P$wpz
zM`P2$17ihoD`Wx1c8ipxcKg@Vm6xgyzb)y-_p}9meN)CR0F_GYI}NNtC5z$;ZJMuc
zpd?L72teTwi0LMKf-|9C;(~pd8fM%+(2g%`%c3*Q800*|f9#l>n+(N&uj`Au-|Bl%
zI6yYnj~{52<2t{?K*d8^RoxW$x4eTxPQAvPCBDjTKi+!Y?XxzF-s<fW(U_^u2M-l0
zxka_k;>lARqzveGvBYWi%flU_-vae%^Y*Uh;2_u=kSdbu6|F*~{rvpk>UV}J6MM=)
zO@8XHDL(i8<30oe=*jXLKe4OtEOVJE=$deh@A&(*ZBtrWS_Qb~4O6jkaaXaDJkk}f
zgH%-;J7Rm+lk<{*{esc4_9ASXK1)LB9TE}}fffY_;MFJ_YipOIzKtf|6<%#0{j^^0
z5*^i@&Uqx*GaQ^c>xmxCUQiQvh2=!C`jObuR-t0JefO?E5%8W`=j-Tw_^>){1qc@9
zFp^F{U?GWTzP@m%EiFURbbf_LsTc}9vk;w2!Fv~>uyguF7o^b$)XW2vOTj^BU|t!R
zoXmH=8FZ8LOow>Nqz_2;E4WSwiG;S(3hvyw=5Vmv9^MmCC!>I?09qr9vLMD#+?<3Y
zL&5@8@nEvM<MpbF*dyN$l}UnmP)2na6&+nV>ERc(A@0a8a7T`(>k2X?IeV@i=mlYf
zRL;8R**?_CnAr#n3)3jw&qv5GO7oGev9n_4A(a^?V7cwD-f$Sa_pF?qBDFL|U?)9j
zC903!py()gaib5nKqsPDhp^~_3DiDd22{nzkLxjafTFEm<|`{Pia-Jg!!YFKpT6le
z7<!~d<N$4`d^za3B|<vdnJ|ToGLmen1-5hHBNmJ@aDKzh>BCroqQM|uQ1ztkK|J}J
z{tB=BTfk|KJ&`fureS~=$AA3r0X~P8kCaHJp-NIu-;{Sr<Sl`O%O^jQ4U>MN9?#HJ
zIv-yR_OKfxkV~lSYh``+@7xPSO6db$*JTpb&4K3M7ybIp9cM(`w)b@+8m0gma~f+?
z9XP@kUPlxUk2nwO;7rP;f<($yTczn_nw|Y1YkSgbsM42E^9GMJ(#XEU6ot8IeZAWD
z@dCAj2L&)IVefl==8kd0sfmB`9KqhvX|T@-p)rN91rtIkfUdiD+g-Ur6496k@3@u7
z82Tl6$9Ur!&Bsi=%rR~{)SfRy2H>Hdzk?Rl$ivIa3pc6q!%G(OTB1k=%(0>9)CX?p
zZl6QgL{_v@rf%sFDbxT2Zw0PT2NzYy!pUuW7|nralxIIR;Ub=^?0K|9Wm~AmwfH3T
z-6l=Ra>+9dEiI~8$K0Baf|G_ySHH!C8Xy~C9zQyN|Mt~}GLE%4%hK4}P71)fI@JMg
zNM?W_eWnkL+6cTdR?=7zm)7|3%RT|ox!IDk$0@cy(@<71|2D5&BLj3A`>RznG*;MW
zKah6s5JN_)DziCC_;NruJQ(V+8<3`zd$rn(ijzIV4M_<T@)l`RF+^8c_*|L5tHZ;?
zafU^_g}}%$LBa%cp0kDdl~*H$Qou&<dE_XB1=A=*$Ho7I8tG_iY56Q<>R@T;@=Je)
zd{ALkR<yDL6R1oN<pCl5_sy+qR7Kj*QU~x65jSF0d5w*Kopbid$`XZ;Hb8C}i|@58
z(U0X`;SUUp$y`S3t5=m!yyTs2ST$x??)@gq9Hzn>d84DESXZwL0Srh!>d)AFy)P$L
zV3-_=ysENi?#Pf!P<KQo#|A<KNzf|j<>leoMp0mcGRJ>IZ~*ehf1Z~9c1$o6(bAx0
z!|>+vXY1CLTmqgFU_3b%CwoqW078&t^dhfz<$LTQVyWbfhf7!tL=;U-lH9kozpSN)
zFS(|UP7O#pvN`QCg4DHO@nItlXJp|Dh%o>aD!aKAw_3~i3RFW}r?3)YgD@@{u-h;_
ztVZL4L|JMaR6giI<1P32Y$fy$`BpDD_0Nq4SAsCn(qcpPRo%?H$IQ&EVk4tyV^<d!
z=#47$8|_)9d=3r{xJ@)vqU<>)s_gw)$3>FsCYaJKzZUwEk_evfW7pB^1rtIjLWo^U
z_S}pkX4y0|9PTyWa)uW?fEA+QySm?PIjUA$2tnb|%LLJMjzoA}b<T$;`7p-x3-mXi
zDzGAF8Tu3WW=KIlm~#J=U)0GnXULM)b>U&PnA7xYMDd!31>z}|$eyU=ZqAga_|U-e
ziUp=Ti8GU?blE1MfR7S)d9FkKjH@Tc7B~WpFVBi-NJXehb=?)ijwS`XyP#vVf>aNA
zssac?)q1NnA4Nb)r4eCr0s;>B&p2VspN(^U7!e?0+>cycP!J=hxJd|D>Yxw5?mwQ@
zW|T)HNZfmlsZ3NCtI(H3vaunl`;s*}FOyjCAacggfe-O;@$cHjg>Aqjdtp%tEQ~6P
zzP8nj?VWu<^cT6g;V48{F$nzY*I8TJ(^ZX)OdzNf;ong9?f`BjnH?Q@|6~nKmR|LT
z54;}6MtS>VA7B2D7NDwg4Kn90ZZ;nnosm}q=d%V`(|KXu!w5@TAU-~TNT`vxiw313
zU9`Se3qFBFADj;8#Dj65qPYW6CPALU#>(mq3e@6V`+Xuf!*`C6Fs!|mfPdCfdGc0!
z;1Vva>e=55_U}sAaF4El%Ys1es>-!B%+I8u`mZVVTBtcBC$|o`{(E41!NE*7x1R~V
z_;V188On^xd7LlDh$+m`v*g(gYS1lPMg%?F+}woYavS~l(&NgBXlIOr^Wwpf!61x}
zl=l1n*0X<dzAod{oqUTa7qZ<RC;{I2N_ZgomA#x!np{eC(9x%!OOPQW)7VYu2QqtT
zc-R8c?zoXr<KZv-riD(g1PKWW_{<+1{kq_d?nl?jtw2}ckS0WOe0|H8-w8V{!e_(`
ze38_z86SRtndr|NMg*in`}H{!sz`t6KrGVYcF;0TaUVHy1R_{V+y&yCL0l1{5#ksj
zT^BiTi0GwXV#~%UaYB2{6>oLKv9R9@H2m8)(BWahr9i#({igCqsw#yojajK4U05PD
z^m9Jkx9*Uk^G}|;>32QCts9OyVEj;@kDKVKQVtVZU*n-W9Pi(q(66YirFZ``_Ta0N
zCl{Jye}K|Zu7?7|-od+Kv0{n7H1kLN#6$ZaG!UqQa?`5?(v0B48c@v1NhZw%1%*I@
z=QrQl#|*>)U740G{ThLCrpkaqoxXb~Xd@z$jq#F5VS(g2*s+3n(QE6QmHP7X7BgNW
z>`H!+u0%eF;l2y$nVC#Y56wp=2dAV6Kp#F1BB;-cN2dMkE<rFwWV=-Q%bVsMBX7y&
z$?ZKvtvdFSQ_^HP1_@E#3kV6_Nw2=|3<`;$g4vAA(|8vZ1`~%dbIyzDNX<5L)??;`
zs>CWv2rD_^*J9Mx=S0M4s2LHJFzvSKjIEN|Zp|eT5vS9kTl`>d{ujj4feV*{Kd?=n
zcg7MWOhPRK-;njI_Fu31i0^Qf+HqlWe&0I)C2c8A*sGdnH|elpzq@eJ8cHDTY!N;G
z;@AA=|KzxEXZ~C3z<NyL*bZLmpFc}GD{|TMNT58!v17+YFR%D*JRy{!-m^4g;X0p^
zj>fw=hyPMd^n}&*zd6g`r#-B+dGW8=H(r^RoMOYJ|Bs*jKR>%%3b&1f;eY<l1oIub
z|Mk-TFaPlbnJ=w^W2K$!azzm_35gXrP5@{ZWgvOLl(7M)D7~6LO>y(Ta*_I@1E@p^
zubPu1L9kkMg$k&wH8eCZnQ;0@;AC54eSHOl7DOJlKba_C35tW{sOafgrZ$H|XAa*o
z5|cwhmY+R$uJ7yDAiTHH-!7r^@BMMXYMu+YUgCKyiNl#f>?GBu?dvpzALC#Y)sc@;
z?-E7i?fdr`!4o435@9H+TUY<@M|Cdz{K7)Cd+6pM-lZs@b+53kf{@A(Ozc?I`F#w@
z7A{CKph1xOb;o`>^pu5SXU1(P1R`Q%1M%JxWiv#f`M*K_ky!(XkmRBNB_SKrp=jKd
zhXFh<Dghy#Yk%`5B{kI-q!N*2L&a=)m+#x75?IMX1!QJn;Y)ZD@F}hB?aHX$*f=?L
zg{0@gh_DY@eu7*^L`M_JCHy6UEJnNB;#;?F_D!b=P<!-6l)RhUAH8oeBh^TOTzlNa
zFdIP!d=Q$2s(0_WK}2N^zt<AjTg#R8is5t@;yoKO2reJy?ssw47hjSzZH!9QLKM~t
z-Vp#-AD$wr3KdW*d6-oMP}K_!zAd`yBZ1S$@j(i31ad~#R^;%38wI!unr3)igyTt7
zzI@4oD)Y;*X^9gJM8HaDFfl?;?l_#6;B-Svwy5D$E;ly*ur;?7M$Z^KQ?)mM>=A3l
zh^7Ox+)j0vS;I=Hm4v_XTA0_=)2k!6(2q|lD98mdGm())py-n)d;9u)uxpDSON(Lh
zu7Ioq1F5CsQ?rgwT|cAS-2*^dVE68|IXOAx8)19}hL-+F|Bygc*3>Koh<v{|u6_-W
zMx!MFAeh|T+)Vck;yQ=m(1Yf)l?X1^uI+<2P*<VNN*oE|d5!GuOrAgS2FFsG6jX~K
zHkk>*fi@%+tvke%DyTC{AH(R4WGM31fK8CckOX0(-A1}-P;!&Ri?r4=7;u$TLg*J1
zR8*Ei-imuJk0x&(&ffMPpRRT6DO17xQ)vBsGo+;yygy*@M2C46WmFxJP#>d15<)8p
zZE0S9{+7bf7qOE4P!QjZfof+x3L&zEP+Ldm$MkPIBs|!Ytp)+e=H3iayR>r_VrcLM
zgroo%s))hmajal~@Ka#V9v;l^?8Nl^@u`Gyas3^qu%e5$0zy;+Y)QQT8*0%?wD{#;
zPap?cVeoyk-iL|``3^{3-5@k`d@TIW8`g6i)zooq?URTIR`K!iA=0&=;ND_@n9DNc
z7Pf5An4w$CZIJRh1em~v>cUPAzc8=)+4!~G+*>IGpRhe2u}g~H8(;Kjh9~X?Z3PO7
zzej$A@X00NmCVO?7}LjFMAVVEr6V|3-9J#Rbl^1t;PetV|2c0K)I<>du}>uyVy6(j
zH-h@N1dD7k5=-xmG0LsATu<PPNLzt6patq30~EhdWuikn9s2D&86u(ge(3N4my7vH
z(|v#d!!hHLK%|gxAZ3U~wGnhsO-%}eFEPmgQ&%ouev;~f1E1e8mLwzrEXYTT=UNA9
zd@c25U|@sG88kKY-f+6hOI`Up0g`$tY6OaG+#s?$0&5hZs2oAB4EE$2M48vz&j|O_
z;i=(2|NMF5*;!QZM2d>~Bs$)4n!EC+IlfbOKF;7)${rT6pHEj~P@KjLr*E~aC-g}W
z!q*@u`6uzbx&Q7S1t&;ajFR;H=EjsRlA}%OFeWDZ@Chiyr_P7|vwJ7P@yCyTn1!CK
zz?*~0MDR0?y*~&t(#P>XPUf*(j`zM6aJpYuRz^m}t5<9lv0uX|GAT|9F(5R4g@uKh
zbKnR8=g{HUDS$7yYq1*(<2rJTuaQ^>IY2jAwbl#ZkZf*1bS~`&aY78UQE6#m?-FHt
zU^C_r8PIkJCc7e(Ye6_h4oEilrYm4PpkID?dgC#)AB>>w&<@b!Q{Hg4Eyi^c9$|YE
z0$2u`8nB_HOvdP-H(p3h17K3-2*)T=7?12&$-v+p6T^vBw}RT`=-}&{nw0U*ohNS<
z#qSnif@4QRK~?`b?2!ogEd@xNh>l=O7IFvX8R4ib=_a!OzAOUp5;QSrZ~Vg&5)v$t
z-JtOhJI6HLNd#_i)PbDuR;yYwK9`2V2`ztj07qr#!w=QfN*UQWE|_Fefn*g*GvoVQ
z<vI_AsfR^HBv8W|h!xWZe<eDB_8?e{oIjPYnDQfT9)mm^!ks=)cmP!IcFCeWML%)>
zOFhzTSs7V!i7r&7k?z}T##P6C-FwiMMLxcM{W_`m5aNhC15+JnzCdssRjsY8zy#Nb
zuxGy%R#W3#y}B1?bhDTkU3eXPrf8gun1qj@sOU8Uxj-yzlzLYd1FUN(zw|9@Iv=02
zQ;(CP0S`i+f&(^7fE295K!M7%rUBH=5D|Gp7#$o1!lW0kIY9O+L=o2uUG)X1ieS{X
z0(qR0zjR3!%^@W}@D6KXX{hdUFVnO+Fs=-5?n~e<so*)^x7&biP!<5a-^f0gKMz^D
zC<e^r3DDhCRkeh2!ZAMda36+mFVFsEL+20?@pnPnpn=W|>6<9|z*R=d03LFNA0>tm
zHAi+3-fiPXot%@OCN&>0R-nQFNJ)eI%C-OUE)<n7JTP``WA5k(*EAOJZ?`ezG1OCD
z-uiegwx)qjQao^g9t~+l|BtfAIXy_fmHj`Q;9E_Z7L=i!LR!M2&)p~mNWp?Dh2<TH
zd3)AVcaTHC!V>di$U(`UVpcWh<tI$WT>+1zq^GwA_XD>}36gBG!7=4hO+T5KKr57z
zoxL2@71m9IG82Nr5|6*Nt?de;AxK7Hx0fMBMcZKtxCSGo0r=8UQBkdUP=LR0-@e`Z
zZD;c0Yt+6AB?i>h{79du^n9?n3nDVTHuJgHSnuV7gIa+%194y}HAXuRg?der5Z5h(
zc(y~3+mS25ID*Q7iyuo}m*JHHPZfZHX7XLq)9Lpoha<=_kEV;NUBCKXnU31&JC4A}
z5td(oh>`++0Mip>G!UQ);BR&<6dYLBvkO+1%^PjX@wB<gI4f4Pf_ACG#(Por%hrxS
z#4(B-4h&Th=Mf?rRE9zIHv0H$2&`rRUa{q>25@c!uw*0>BGtkzs710V+x;b&^b0u4
zudr7O&4#m(@4?^4bafB7xD<35y?*rc4_*;oMEU24jihQ!*^2dupfSjV%)sZ*DOp*}
zgs8_eZhvTWI5;HaIciC0eVA0R$*4|$W2Rk>GlEP!o`PkFG`3;UdoQJ@4{m3FMBj_S
z_c%-u*rAd&gr^-b8BIV0gbS`EqHVV<8*YUa$_YzKAAKX4GU$S0t}(<8o$GSy8di^l
zNXOi~*+@yjvWN^16eGR;!7xez7)Fo|+wxsYu_|;e)d4XS`7`Jj_NL#vSBE~vuVd+M
zSO#H6%nSsKJ)8pUakuSpSky6UMq`GFJ0oL-b$A8Lp~$nxT^vE%W`*>Qt3gdjr2RZt
zG)%4n1-unKJ$*=Be?B6%>nKRsLQe4$Qs7TzU{q2JAg37H2d(YwUQkjtHiIs+e-;~+
z`fx$aZ+{&clA@MV<Vt{<nE{)$LdUj)_oHM#aQde1?nIL45qw<hV7hDY;Kai?oJjZi
z+0xeCsQE1r1C~?FD^}#R{yuj6czi$ry<J;+XqO8wFK>Ewb`vqjBf1aNUX}R%O_*Eb
zqO7s$0|yCZ2+S0bd%;dyyuT@Ji5dV&T6A^)BjN*H_~?-hGLg~38Aw`xPE1S>56T=n
zM~Vb84}fG~0O$ly6?>gsa%uFhgQ%FC0XAy|3X@G_B_pFbr4OxIJci#EpYJ0_KEr~O
zjO6)JU6>v6Yes2cj&Zx*cO0{*!1EzQJ0$nVk*F&=psJMM#5cD-b*cgT2r6(Pr#Gbc
z908)RmMTU~61i0&th%P=Zq+E7kgWi<us(24WC1`aDT)*a@Q;Y|_w&m`bl#??r$_IN
zyGm?yV&8fKZa=Q2WsYHx5&S7fSZ=1a-4ab*-5BXV8@E{5+Sa4GGqA8o)qHjx>nQl-
zwj#;Gil7!J+Hd$H_G^$Q0)WV+4*Jivl(nOyX-)~Gd|<~)HvIx$?KhmX=e)*A>DSn+
zMDLBy2zUQ+G9(|Z%i?qm3rlG~!oCgW@&>}+<DH2|t3zG@JkM1@cLuKI^{U)EIiV3T
zD^|u&Hk+>{Ge*J_T}ye3w@Us4yQh6YRLMgNLtgas`}Z4g^SG#V@akA7W7yU=_Y3`1
z9Ug2Qj_F%q|H_Se+4_}K9(<niftlnJPdm)LWBO{FwS|RX(OZ>^#0h4^hLz^rxpSxz
z6L%j^Q<vgbkg8COpq{#o-s~!-OGv`i+*6d+jfVI}8r259_gXG4r)2l;4<GF8oo9Ov
zVv}_o$X)l@ztZPN3|Dz03OB)Z5a~=BcTKn4(*tWK*5(-H?GA(F$Nl$kbZaSO(AHqK
zNVPZfVG#T?o|mwIKxM!R-bN(Y8ccN7SvMy(fc-Imvywjk?nb`3NA4K$;(?k8cdj=n
z79%b8SmLk)8TS*`m8xTK_<lwP=<g1KU%JiDUi>mV-1zJddbV=KGnkhN{y=w#s1xk*
zN{t|-30nCciH0w~!RZ-{-J2njk`bBr%IWFtYD*%dYk>eFlsqupnR00(yEl+SYpM&5
z9bI^SDxP#tIG%aT#7OI&W0c>YEE&~m8Ilj&z-iA0#^rbsY0Mx}Ty<)ud|}=a==4b>
zQ&R8>4sz~<eMpL9ADuIr*WmVZ&dzKzp&!upt)*lj{~n8rjASE|LO`Eam7Yg|iw{L8
zesM1M-w>jb$LxeM0d1Kjjz~6AakrY(F=vcuo+R@~rl#fXY-cfJV~-q4CWH=~ka`@`
zas@E&x%xq%L=1n})YTPt>h$S??rx*}b97JGuJN1_?4mg#C2xkvzTCUDzUC8MvJa+#
zO~$ygPW&+TMuXLiqO*nqB;w6|?d$nGo0j#FZ*4+64434UAT4e?bLI>QKLFKgxh(Aw
zF_%-{U?SAzN<0`TC&m~QzS2|^NR~7QZEfwlsuy)OR!UTWv>gGJr*3R~3oQl1X!nt_
zaFn9XLJw%9)`!<2dUZiO`FxCZb_cAg*xr79_7vqxJSGfChqFagOzb3y(R<g2pBw8o
z7#XS>)*XF9F=O1;V!XF<7sgA7chKp5mEF_${_?PS8yJLD2IY<YV<Mp~t*x3TrH@tm
z-mN+RhplPSp(`&%jfYqEtT(2M(HA?R{ij&tk4k_VXQBiX?p!szsxM_~gvpH&XNAjX
zIZ@{-BToaxQ5m}hh~0a*N8z7tLR)+h8kFULG_wYsDyDe#+63mJp!z3EEin!*gnk~Q
ze4K{ib^T;>JdupRn8s<WdpBSTdNeHEGgHC#$57I#Acft=5k!H(X{egFJD}txH_$*9
zs=K)w;bz(Rf*noFhFua4s%<xkF(6182UW`n4^n=k9L>wi^9B@;il%Z7Ukd`pZQ}?f
zB`+OCGR><4xM1Xf;{-Js<L!|$=Bs~975GSjr)G?D40U%ehESt}b2H(3kcx|7U}RLl
zWa#|_zlcoiCk;U!fPF6;>eS~bp`HJZn>!ro4u1N_+Q!BbV~QD&hoObm>eDBrBw!-Y
zp(%T{AoxO_3WxuUIR<=o!5d}!jvb=io;o)&8?NE?=LS%81&-JwBO|wQ?Z-Aj_(Kue
z7+zzs?_zqxjdTD6Y|-1NaCzgJui_7qaAUh%6ecwC*ssf<a-_<1b!TrPRagldEPb+#
zZIQC!i_OiI#L77*EH*xN{?MC_4tmXLdXBWSNdIxzPeU#Uf!gc$GdBeV8oZ`{GQ4oX
zeojPC04m(P<xkNCT}PmeJwbd_h%^<lGL`+kq+P36SjkFsnGm(zKaNeet1}allEQGC
zF^0_r<ABd$KMaXb0@AlYhz;kb2Kn;Op-BbctK>1c=P_DxGQWTu3^xFI`&)2rqoLr-
zKw)A&|6FQnsub40qTXjlj%D9s6GHQMp`{q*fiKDf(8QR&CKduI(;H_T0aZ~~KUh6P
zd5t+%j(3{B4B>QO34~+(&!1YT>$eoIc=ks(8O#9c`3g8X4gJOTYKj6d^x5GXq870#
z==@&u7C9VWcNzNk$d`MN3a`K&0oaO**ET|Qm3w)Skz(9E3^rdb?VtL1+MbSo+=}Nj
zG~SozVQo?yI<qRm&q>YKxR|`Jd-gmz=5_isj}<a3g|W=j@xP;@7s8+I3D4R+vp~&1
zPJ33g#l7HrYHB@afp*$S(`)(9mm63{<a#*L8`Mg<-B@k4>Fwm*nldX!1CgSu1GDo|
zlk-N8qK9Q(ew(`R>w@}VV$1IDEzkINOnxvpOP!pjRk)dbqxc2$>rM#>R9<v8H#Um2
z)NVkxk~(W{Zy$oVS@Hh8;UAhAI|v^TX_sLidH{0f7J(|STyU1eMuO}aLl}uVGtq*c
zUshI@>E6&KcXz^nqw<Uq_4+fi7IoCNIkCmA2R4-#(=GAjSV4TC$1m>w+Cs*HVS-E%
z%9SF}7K9+ISaVx~;bbE@kX6bkHhhv&PvjWwMrbmFM5chlsApnw5)`(GT*815ib1GK
zPX7J8D6K{9hW8jt;c0shZxsSoIM$AO`r9YuD}6XSRWTJ~p57h|4AqMHNXk_B37PE1
zSQxfY-#3~;LnE+v?>cnz5gRuq?69M!F!*(+Y9G2Iew;I$ftG8r8&t8mn1m8UbwfA7
zkw<Cy6H_lu5pS>HbskxJgK7El=cqJV;j&mNXOE?qE1=wu&$-mp(DD(HRXst51>Y=6
z6vkALoj;{06uc49E$GPrqE(=>VAh#fsX4J+o+4~HtbJIyxUK^4+uW3PLn1!9OuFY1
zlu|%Qe(q~-!l7CQg$<cbhV|M8y+_z7Oz=T~nk2$dCod@%Vg29PElwr`W(JB(#>?SU
zkqRip<u|fmgxZGq=0l6mzUmIZMg}G(U!Xha*L^T(-H+1X!rb%@Al-BjCMe^)VCZhG
zgOE}@YT+t~hRjc%^h2h$>B8=_B>-nAvJ9FL$uAJD#ek@m0qcaj0UhxxK=6{bc+OC2
zhwY2LfO;$eQxKm*#U6CBOT5)>+?D{5S<5M5bQLSFmlF37*!f@}A&-E?kUkO`8&vwq
z(avmxE9jh3kw3{!hsGFjvPq2SUjTw?sU7>WDM;BDIp7Y8yu61GRp3l;``*1}6zKv1
zSCLx^WrZc!v!$G!UZJQj*75Rg%f7My3cA9W1cm79*L|VQR>V6gbLy_5KNGQ_f$5+(
z{$&-knrMdzt_nRW8|D%bB^s|^MO)Lnj<A~OSn5b{1fq>vWswd5uz{uF0(N^YPMid%
ziM@lRB_1;}B#fitfwP4f5cAeIZ)$+s{OsS@2qt(fh=V|=l2UQ*$dnG@UitV|0<R>u
z6SJEK5b{2+Qiz%)05dX1Wpy<z=GKVa5qWxuS3I~({j$rYMl+CCRx~s)LNfp;J6zH0
z^<dedXp#7^DGY#r0`VUw=d~YIlRr8(B03<hnW*3cpw%Wmuobb4rjXt!U<CkCufTN1
zs_Pz}3|PyupE`eOmOgyj`zX-+@+c;c6;TFo@$fJMf+$szM>XaLBe2aTPfJQrJa>~_
z(d@`SE(t*dQJ`26TXzSG7Anc(XRUsB=o?J+_(k928YzD8rCZKhyg~N#2UxP%usC}}
zitVe<MxgzX-zEXaDWXCi0!@VEZQmcUe(%I+`g4Hq92D$+0<%Y8;un1@<M#vPa@aSq
zID>>4F%d?P6BMBXl+I6NSg`c+gSrKWKv6LAD%3f_x(`l}<=LWpo_pQJ?;B;JT1M=E
z-MMrksOyo<OK@UgKy#plna6x=y^D;ANpD?1kX%aT!!>kyot};k2fpI&;2rRQ0?Bm@
zP!z+C9S_B@`DyefW*#{ZC=`JcYd<)lGYdz<Lwh!S!AYt5@WEI+OZNyIh3hqrq75lP
zzSQ0&5W}QljhpMC_T>@gTect~_z~C=p90J)6;^8&z<mYx?K4i9Lls^lSw1U>Nq-eQ
zEI-r&_`LX^%B%jP1>nF6I;J!IpFTZ@hq^Knr5}FCX024e4DI{Z$H&K6VS>=|gr|iF
z;TS;sYGLsi$`e5fGDf0Ll=#woOB5}3^r}v;mpEYLWGTmQwemw$&(P0LMb(sc0sV~P
z`-liu8hM!e?yfEafmOIy+L(AJVmJ9gh?xy-?d+KG6SdF^NxObgFHOGpx;nIA<^(#Y
zmCxvV4ht&Ne_J-;149jxLq4dV#reSHQB-NU1ZA4uw&EWL9IhRwS0nKG;hb5Z?}0X}
z5)sR4#^|Znf=Tf@NPP{91zfDjm<60jV12h7Us(|%*RyBOnwguYNbNH|(Ro{ewlp*v
zv-<>+M5L<#JA{{lO#PzrM6B36bGVyGt$>uRmNZ$1^oBFJV;l}sf2XGU5dFFetgeli
zg=K%@+U3MAW8dYC0;lne`el53z)ux8m_Rg^j$6f;f?**8Ghi62@F@UC7*?#PMKRwy
zFyM!IU$Kw77hlk9FlSoul=U}Qsr{D_r9=-eW#(O=v}HAJ2$|w}dg(V4MlBmx1QxJj
z)gDaH*RP-AAlNs6=r#^d?F;bR=0)S)JYnRwz;suVCdTFb`8u?b+xEV^=e00TjIt1J
z@?oAyCfrev!T1japKtdmanLAa9@2B}x1JbQwxEA}xs@hv4cdKrY!xI(D6p77ST=FX
z*k46rd@*|DUjokzo`^`c958gzghe(%-_Z9=6@>esO2_*fgpPxspPxe4?T3axQ};nI
z76+IC&APZ>7%wLqM+n-b#y=kERibDGkUMQZJbsApy&Yzap*Z!fds4l8@N(hVitY(f
zta<lNwr*MA^$7-nJOCn?=FxqdvYKhB3~dD1gR57E%7@6*5<p3iMj0cTS6x0MqL6rp
zuz5FcaG#d>8qDy+%7PgbqHpk_l=y}8?dlYE0sQle!@MSnm-^4|*go?KQ`j2*&#z#q
z|M8nKUp4qBC~IFmxM7u5puFGrx3Wns5*+kB?jtnGrXtHHCc45dAEwF<t2gE5{PROH
zs2@Lh=L~<m-8k)g*!zF^!9n#eru<CJW3ZL_pI<{WL&HM@2BwKLiyyN1L4MNz|8M@k
zb`K7MSbwX4L;lY%t@+;N^5jgDU#Ws2jQ<aRGa{Evt;1m5hE<FRBIzTVoB4<_+o~YK
ztZBJzChrCLbx6yMQ66ViZNsYVr`s6-sx=nff*TH?NiSOn>L)D#D(^mR2$L)_;|Azf
zZu{c-BbTjZuHCzhLcd9NihcRU1*)vxd5IGa=`N>U3TSF_uJk`&5C6YK$w!z94GkrP
z99S({09nnk(h&7H!wBH=*94|ol!=iyZ)T22mAyb27@3rmGa~h$*ZL&WcA7Xa1W->1
zf`tYFP9z&)x%-Ii#kzH7Cr=VD!6;F^6}aYF{H>E`U>#06W;9)-&%!rHOm8gcah#+=
zu3fvLEa-#MWWAgxwy<5g4O>x2?i4|8=K+28-;c}62MA7l=Z+s|hqJuzvu7RwGfeP_
zW5EX;z_bT^s)!~Qt{fWX;g56s$-+Eju#`J@XpN1HLDcu5GDD{Ef&C~vaaTg+)ee^{
zRPUNx7`bLc53GVKf^E_Y;6K>8Sk>9NQ?(%z(hEEc;>3ouJk*sh1R@S<))nYsu;A-+
zU~RV69eNVvP{X94o@;Gwt-vQ0hI~cL#&n=+w1^HyhKPU@-M%F#;kCtjP`;vCC(;jG
z1~#E`q6EM2Yba9jJra0DWhD(t{u|F!(LX##?EL(0pV$04+%Zr%q<_~ev?k>SKMbQg
zgg3Q3&_<ROh*aB}E%eX(DmZPj^d>bs-bV}nYbI27iTgdcvE07?)j7~1eh^~d9@C)@
zo`t<C$$c=sy9$c0AIc8O8`5wyd^|;=79`q#6uqUvTButV$zjCrYIvNYO&*_|41@|0
z#^=6JQ8yU3L3jZgM#0pS5BIX>Dw~iKR6khFOOM{Lz>O$05vudB7pmmR6HCJFpcO&Q
z)`uJl<U#cL57_^~6)FXs1hL9Ohfj8`6VV{~-snwWaBofeKnxTsp@4A$Ic^|_W{4<I
zzqBNWqqVTWmFwr3(JW_Sk+yE)!hgK)eW&t~BP<oMX@=-QS|REtuOBg81$|1yBBg%D
z4s{@MRiO0n#0DPQQ>Xff`V~(n>eOEFI4b0_?K&lh_a4BnI(#K!t_8xAbX829Xf%l*
z2v#Rl0aG=@un)9WWExqKq1281TQn!VD0nd})Q7c*wfM5;Ha3Bf&1`<E3Y)=7ZyMUx
zrx#rb!3TRg%4NR$KWXf<sFpk=1%ha%5TC$x5e6a!fF2gLZ0^hftU`L7KOUDaKP3||
zM8<?Zz~u~H1iycTRzcY#cpH~^V;^n3mJ#|Y4oVpwlh`H%M4v{UVg<u>!_NKFb4D8Z
zU3>PtXap4@N?ip|at(z0i$>_jHsj{_qv_8kHZi2&M$cCNdj{v-9Ok(|x>OVueHGW?
z0qI=`r5LbozH|ZuolKMk9$`TN3N2=ev{Skz=XdUWw0vfv13c-;d<?7U<VfHhMsa2Y
z3;d}kb@uHUktpiXs_BgFXbMSAPNw)#*>BXFpZaM-Xg=bmX7F~gYS${ZHz4Ym(mydV
zF-yE*3mO}J1de?mWYwYYX?Pf2)?zw22V#Xr9)7!uV#bfPb37c94NrMN27tmFFNXKO
z;cW9grdgO{AmmhmZ;gOFM?fjy<Xai&djUf^xw##<e3{@<R;Iw-rFQPDL&x+3s<^$O
zTA+0a5+F)+1m2Q|E$)IurJ-=2{gnkHqZ4l4`~-fntIT5$iL8UoylCQKBfJEwl3eDX
zfm>8#;l+daPE79}B*T!o)kot^VWC=K8oT@jR45XngkD7HOJ0)FP6^ak=&UGjG<EBq
z38Ut7a&^5S3ghop2=U=Sf$C%0zPPCeDK8Q5v)}FJmbQ&dKZ-v})EA7bw^s6e(bmR}
zUdF{rS67!B=xHH_Cj_&^xEpitizH7xYNGj~>cziD*Ph7Mmj(-<pY#3uMTc`|&!Qg>
zw5W-iXu1CHy$jyVVZ~za=ok*s9f5lE%RTp@0gW<$0?`)b4H6o<(XFH3aa`cj(p>l#
zT@J@}t=G@tta>wB{jsSbit^RdwjJ5=BjsR(X{Zdy(gOMq-niDYKI~YE;!^)&@Pw2%
z?zfG;M5rXBB48)zD1=R!YG<0MG+kYkqf8teHLB^@=>y@S2u=rsl2Y~e^@XSwcL!HP
z<3^m5z}q>&z8=nkw{eFt_i;lUHuwj=hm+yasmY%}Gs)N=o*;QrrLq)crZ*JjfL8dU
z8#anBNHY94z@|EG5r@SN`Jto>fD9JT?KT!0Fr7hEp(tR5O5lCBjvkhs5zkedJpUqW
z`+gRd&~*=Q2;VB(c;w^8O`Bds$8OdVY};R$IdiFrz#0gN@x?Q4)<)K8(w>Knjd?3#
z2i}H<oH8?uxU&n7pMdGa0u$>hU}9E{`=OPp%!os43Bw(x2CY*QGJ61q)qtEK)4^EQ
zv=*U;Os7NUmL<i$EvqE@_kV>JkHhyF<>4_(w3Ns`Xj+xc%x+grs3KcZ-mqp*XXLWF
z-aHGVW#bq|F;KEp1<vi(>{#KDw9yIQUk?wft2f=~lmPr$?Q&`a&*<WQYirP}xE)Ku
zmJ{3`bweGxIAZ#U+_H#yfu?qSGl&Z#`|*&_iF3G1r_I?DPP$Apu803AgOnF|5B1o?
zq{9I!7zGG<Kmn&n-OzY^@(-lGYfX|n-4{N`oOLV#N$LL>&=8XK8l;6Aqgyr^Z=DtY
zdR^=5vEHXpVFbalQwV|xByT1HoW%?O-s*y^07Z#b$}F%l${U^Nz*O9BGWLet2VZhl
z{6JuHeq!m}#EKEe`ce6j^#T6=pEGl-{wp}Y@(O&-?oH)c)Gz_gFo%aMlY0&zY6Z9@
z5J3fu>5V?%mD0L<hy&%r5nbUAcgFDhDZX9a2(Pq%3lgWk(QZLl>4Q-wDmFzJ_`<K9
zxWN)62&iM}ej@CK+W|g+SXf=;HpQ2up#Z$<WFVEV!|WX7lm6>|6XXlc(BYgBK#>Ha
z0!AkE{->m5T~j^WL^Od1KtQ>rol$_|uYSf<cpTU)6+FX^0ed<mB0puph99oc?*}>m
zZnc|0XGp5K9Lr7rzV3}7$6~3&(Fj-QkOKk(mzk7Z-twq35z}74$9#~6QgBHAcU?Vx
z#5T|QDI0?sY-&+J3I_Ovgs_dw>p_nB*>D8776D5kBW;B|qokySA9WbMmNGhhcqwe&
zqka<U9F61-_|=g`E2v~?iDw7~^l;4b@R{n9Z)Z%9pNWAfw!4I(iv2%!=qPOlGjb6M
zDuIfD;8(!${DB1nuqLNNRfTlc6f3>$6X+j|a#bQagSAt^Mute-R4`rwu+T&rpU!#q
z8?Kz1cMh-+Cs)WR$wa5a$ES-n)A;M)4fev6X9ZRSqeM6mBh{pabVC8!tOVr3#=)^M
zLHf|Y@At22Wg7V41VIn<qTt9Aw^2F@Y5lI|d3d1o_4e|ke(?u31Hba8KoK7yyXo0J
zv3|)7oL>x~fY?^VSR&?VQ>>t`KnSg3XJ1O;A5&U&|A9s(AQ>X|(L{xZdQ}3iiZVWj
zc(AUXMsXZ4^d>7-3O5`Y5-7$rcgf0%6ulG)Ll|eEK0*KH^7+3|IppT^-=F7}&x@Sy
z$lgMV5hNfyjY$21lR;{S55JiB&8B!2VVjYv!o&j+2qP>pqa26Lj4;HfP#%_7?I>N_
z29}0m1xoX_RuDQ`5o~Y}!~xwzDh!-ZDA88|0DYo%<Vf^Go;_~a6hD}O)zc9j&`2~i
z;?!(POic6xTnfcG8(z~jC_2_CGUR*AxM^X(C5n9^MDN1O;b|^EyF2b5SEX2?``qox
z`=7q0K#1*A@{SwI@dO?Rmg*06aUYNd;4%OCD=sOTBmdo>D^FedKkxnA|5v=>e{}fw
z2lVQ`uVVk1W%ut9LKAsJPjyqv^r!z<`^tYC(^RkfA7P;edbDxHmCc>0f<juy^wwZ0
zQsWB$&7HHevt}RCf(okn8>1IaE*S@gzxjR^bD+kpC|)Mg5NhR&ynnP3&Q%#&&t%cI
z1j=>hs64p2<JA*ANtoxWyzQI2mzJJRriY|jJ5})~h4$_kyZANsh8e3k10qudGWD_o
zkP1IdKeXGu{yU0L{elb9*$LZyg7T$Bq9j*>tzIX4RxRa$@wX0qc*56CRD?<2X6Cfu
zMP31Chwa4$ig-UM>i2L58GDJ5F!05jwj?~rh0OnD=%*|>G(7wX$!&L;1E5ob{nuXS
zBEP&aM!`vBkExJbNJ;20o%_~KOm<kYQoQp~j8G@YMtDfH=jyF&OV_BuOQZF+X*m|4
zvX!{rJ7$qOVhokom3@1O{Z<?o&mE1?cSB=e0&Z;57$^zH<QAM(n8a`HR6WJc9<{}o
z+YMvzxbo&-H@jy&H*XeMx<iv^&mN=4XAkD63r5N#_FG4bVkByPQc@Be5bG!P@g-W*
zkMKZZxwEHk@oi`nW&YAqhd_-9EtX$nZOdjRN3S535{e29Q9r2|v7HqSNZN?zwWo-=
z7SG4~E`XWyV>KwAvzxnwCYc+vmPn312fw~9@f_vG%28*t1<!SR&oY1st1k#g^?paM
zB?$m9L=sCq=RdDW+px(J$|onZvgMM70XXRO@1T|Cym|8`3?k#rz0jvL&mCDHme?sO
zKfe&$EVwfP&<`TxazlCMyRiBT)tu9gz3ULB#ac@?yW?Z(B;^)DQ1K%1?Lp5B*rLBS
zyrsyq+SH4<T>+^sD1g|z=u8_u<SI%~Xlr5NTX)Gt_HUgJs`K-t*2y_5=6HDSah}};
znEZBD)-?=okmb_|d#Wua&+^W_;3U>5x&5EU+m6a{=&MIP1C&T?e-OL~Pc4EvyP?01
zxN4GKmJso3GT-A_m(Wg6sv<RD<K1#Vt?-W_Q-6O?)wAvm39H36*nymem)`?@H?U2u
ze);6Na_RTGj1{;sx1;i<y(JCW^(0LNF|vv;83<}eBXO_6?S+U%xNn{X6CVt1!5l<`
zTsYXQprl0m%>B3GTbIvyh`Nb;otNM|VFZI^$raxyUWZ6GaS<l*6GdbtrS1b0mcz`<
zFh>4_b??CS`}?(?=!)a&_!0gInV>s3IJoTl!5h6pLs!KOA6f7If>SB|>s~UMvB)+l
zq(|DK#?k$3fAp{hv6*!($5H?clzen!G+A+IRC;O0%I~u?;wMU2k=~x;jYVKzCMdFX
z6)M>ath_Afe=t$9bkotqipW+?3=?eGV>BIXaQ6@h$LEk<ffQNg7ZC92^o3oxd9&3X
z%f5dMMgM0$@4FNZQE}N%!8&%aukw8WkHR)R(a9JwlMv$W+3@-<`jMHwAFi1wk9To$
z2E5SAAq(wD&05RC<1{vV@#*EMRY%U)LqV$AUFNzBEzDBRHCAH)r!#a4g~?C`5)!6W
zO?A&B$`skX=fq2ds}4n4vU%5%1Is6@8bsgMRd_ykWDe<C{6w*>528W;tUq%>@@VPQ
zTm(A|@p{e~A7Hx7I36LOMC2Cgw`4i6m_omAFz{-<8kwMwmF4!rf}&P|MO=*I5?VuG
zL<)#6+h)DrK2>3EO4_FCz{&MFJp9xtVh8tU<ij=4>eg>%zBI5CCw$idHRACI$r?{Z
z?8eh?-%cO(9q+t<+^Rv2g84ud$jwL?MtpeU*3bwA3G=>v&(`%E_q=v-dNpPCF)u~Y
z7!@%gQ<_ZkC?H!RL#>;62x{=wb<phn(~C)eh=(d<__t9sxojH*w~@DL0~sZWIy_HP
z!6hlVHz;UJD=uMgXpj%EeKbRvnl{h_kW=s{aW|Tn<36=2@$R@wsH6HuN5x}jX$fzy
zovF=8@CPUh3rB2FSyJB4{dF(q#;`YWL+#NxZ-hVl_r*talM;GXE`5Nsm;ouU>ZVOp
zMd4t}=V3qAzm_k#i}DMVthQ}~OKqxkC^y%Bd#<F~=k*WgQ2x!IC^zZoDAC^1e&3i!
zC}3qB%3rI|t?y}x^9=wgdGIQl@1D~|zijI5zRRp}Yj6C~^BSZ?irR)woAl$P(?+CR
z{+{^787JOi-c_)W1Rlp6Xv}6^v6=AWzj?OHS2W*(8bmka#@NMJ<JX57e6FpLxDEby
zt><m~kEu(k-bG{!&?6jF8diBLGLSbA-N4)01XGA&C<+kO+K!#jF`$E0!c$<95@`(L
z@bsK2IOI58bPCfw-rpU|8|DmUUH(e>u?`LN@{5uiiXgB#c2+_S5=*#zRzc^xHT(S7
zI}Y@6S2Wf)cck)JVs8khWGzEtV-=0j=|i|g<KZ!9|0w0gmkdPw98DW@a~i6|sn0iC
zY-c$i(}|RzAvDXBH$-X;(MP%D)}vg#0;G*Rk($FX6|Wi?fm+J%ad+R~`LO*0lJeir
zv8J~l+_@5jy&|tZ)sUVELGRteGYO;CDbSi#BEOOz1>SUAdrvYr^pw&<h~r1MY;bBH
zJ9+azM=1+dVsens;vEl8n8VnN(18WjseHYUES6G4*=<nl4(=TKMF_}lnm;Qtx+GPd
z;73Q1RTolb6;zarw8Fzo9j#*ZwKx}z)<vt+5J}1KNZH}o)T7@&B6~Rb)XDp+LiPX&
ziXT$>9kACf=iu!@QQBK~)U=y=`Jbu@Ud;6gafJHTii}*kKDV<gA6)dV>0e`^Kl*F?
z4y>>qJ=S*vrwWZwCGsk0!}VBF-|&a~>#A-H6pFSVw4vKSy(w9zi=#_`f9*kxFkI`>
zAA!aL5vUb&NoY3ehvht`{Qz>`vC7gVQ>KTaHVb66wYOGaIUZ@cetqxZtBCcG2FP~>
zzb&=5LwzCz9kvlF286n1ssD|=H;=}$f7^I38l*JOb7@YaG^vm@X+n}T4~izugG{BO
zG)J>U2}x2(6OuG&5QU^@QZy-<BJ9uUdEWj0-nG|W>s@Q_KlUH{S?hV$qi*+o-PiT~
zp5OC4&f_@FL!5eZ7a1rwZd?<x`laX|`Cn7(?rt$)D`c-$-X}Z@o2to)@7H%0M2@68
z2jV>XVymkdFp{9nsL?`m$WFZ6Wzng>`0-z=7F|}@5^lk>T%Uw~M<J-GDqxY;mc54^
z8Lm6{J(-IVwbUpSu6>T49=U~q(S|{AWbiJqkFm>&U)pf*M!slWw>oX_ZT<Bj44{>b
zqo&NsywDmeLRX6XYEFIByX{TL&k=pz)*THSf1%%}C^+G71|=vTxVrP2wQE*DBc&PH
zgtp?K7ER9f+1OP-?d3}urExG|d<^)Q;xO$-#?=WeuPjS%5?PwxZOx-$V^4055|Ca+
zSXVXRmXB6jI4%9<90FdAmtj><fb0*h#s4Utx;J$P)*PxrBRSZF+EK8jX8jgD*|Iaf
z!tD(0fhW_iK8R{+C_XRs3{Ms6J9iRX{QRsb6Kdn^b{CGJ30j$H8$Ay7OoT_$<3r7o
zMcr?do*KStY!o9q+bu|FBh6mujH`bA6VtCi>zRl9D)c06!IVEIoD|y|$z^DIx=uJg
z@H9hR;_DpLZogHCJYg$hRB0E?eUuV1ZMi{3_~XtU#w4uuOgY?hY8`>k(c`@Se2T){
zZ}s6`;sero)27oP6_j$NF+<|76T)Mh@7Dn=Z6RiTSAr(fMcGLodp@o172i{8GY)C=
zp4oLv*6&0oMV8a|!JF2!gliVGeV(K!q;<H}om#qT($uC+U!0p#i@x=}!iT?F{TCNN
zE|Cdms=IHpZpBBDvdHbv?@v;uLIb!@b}x0`U%OA^cQxzlXE5LIL`%dk6o~c9ezu1Q
zyJs}r8ppIt8mrA`3kTVv<c~}R;o#qjwZ_iiK!HFo6xmMem%^-8b}y&#n|rIY`S@ec
z*o=%zp)fPVmADvZ5OC~(KHf#P<29Vx^q0(gGwWd2IIk^a{oBhgkIha!zI5U-9d$b9
zHfes<Zd|`1#$4{Z+#e){Jxa2@HzhjlNK=K`H;mbd``A_<SmT<M(ylo3b+-s~!5!D#
zshAYz33Z@q^-UYZ@jAZFer>Ux&Ht8IrB>4(w>e?MI<!+rW7B4JdE@;3>-W7;-;2Jk
z153|-|Lw%Pjy3S2qxT}bb$9bSX2x$H#dIQ%ucDC9?GP^y8X7~Ys4lo19GGntxfrMN
zi*BR07rd%?yyp9-PiHUgwaRoVI|Xa;g<}cQ=bhVkx9}CDaeal})n%D=fX2oj9on30
zH}>$jG+@)yR)3?GHWzAU`JHND_P(Qwez^bnv~-6Zt8#AYk2ir+ZVsRumI$1y6CM_W
z2T9o4^Z^d2r_vp(*78tIu-SO+rq+qust$j&4DQ!yEyHGKT-zXQJ#zBV0E-ki9-Bd3
zz=W`GG9rUmnRi#6Hr|K}d&|p8Pg4q6mfmm68o#)7_73uuc{sBpkez*Y<=B9wc@LIL
zR4R!`B-cwIdGGSi$N*e`%q`Cu#TOtZU9)c09fUU*7Z)rRUDo~77A;$j0HA_)*F0t1
zqkg)+Ivqw;w9?g_X?1>SlB=22f39kE<`n;Y7hh*b-dLD^>vA({!_a>bGPQT?fF##B
z+Q)LO04zC(tT(M%GQ1lR9VPHqrg_#Tc*gFJbxVZPVeK|DO~_Ey>$KR)s($w@#&tzP
z$L2k!(DXvCSe?V-HWiV{+40R^oS)VZW)j6_Ta(Mm;nNlFhO29g!Q9~4*=5r)HrD`g
zs{Ye@_AYVB3h;5-0%Mlu<bun;j1PA^I9Hzp(lXq>|BpA(dr6XI<TSuT9D28^xf$L$
z#Yts4+NHrQWKr(d{!>!ohndi&*x%Y@V5;ptJc%E>5wUl*(Q0Z#9ZMR;v~PdixO^<r
z+Zs{i#3GJu`~|C+n##px8Le&~j;!E;TL+(rTKn54Z_BI8-_o83WWV&Sg1LjYZA*X>
z@vJY<FO4=$T!YW8teOewSC3w`Wn!2jo|iS(l5fYs6r%5u=HT&bIv@y`d2_fOhiqHU
zFaOqa4Lt;nmP?i}B4~kS@0s0>oj7pdbU8uUlz=J1Z;H68%CtQ}EE@ebD^$Abs2eE7
ze-pOvN?IC~@4Q_5RqV%)I~q<a?6iN3Nx<@%yV~@tjtI4?Qnh+D>sPaGY_>5s_cWZz
z&XW-ut15XKo-zP(1J)IS@xOlGxqtu0!!ZqJnr)xY8lHyM58lRz1@2WO$!zk=ONB)o
z_+ZSok8fNyS5Z+&|AdTaoi%Cxyr1`v^(S98q&`zo`c67IhqSAi{_<AB`ASWArC)xV
zdukdzyb7P#sqT+no2=sx#ZfVYE&FPLD!YH|0D6)<zbA=j|C(>0DO`u#VFRY$+Y9&z
zHRVOZg2J|n&gJlt%yksX5U}Ti<R{06G*zA<DHytQ&?cG!aRp0KFWS5_Oj8+>d7!ge
z%W0O4EN6IzzY1Uc4y@R+=(A5v=RrI6?!8~=oN+mH(W6YXvV4vJR$e6SHM6)V%HbZV
zqp@*yVsyMTcNuK-XG52b0ne5$ZVHQt+J>3#;|s@yebc}Ew~f=<XKD)C?ROCR6qgsy
zg#)kC!eGLnt4dkv{P$<ZuKMm@vFXf#xP_NyViAMBzc|Elv-F7uoOC23V-A)I9qWlB
zE74m2oW#>=g<H3_>)NLtM&E|_CwkXw-Fi0+THzI8D$a3Q|K9Gldsk&W(i-wwvuC?@
z2Fu^yW~WF<<ux-ckU3YSIzUh<*dp{WO+IWZxr+76&*EMTPR{HBng0#_z`cC&)c^!#
zq(JhKd)p2(Br(0UskDjkVo{o0wEev>&Gz<0?;)qD)5K!F#q?=0z;}zke7Uyjk2c*f
zc}FjD?ceVg7-;UlZI@+y?Z&lJjNSna5H=FzR3D*kyX|{BM+_p~Z&jCiF_&!te}BKD
zvq&3mQ{(p9RQGFgup8)sigIU9n~{t^NW4C-DLvz{zR|2@MIqVOZhe{cX?cBd;}fk~
zzsr7GIZMJ}Tg^eMKWR)nIY@;L>#p9nKi@?-t=Dg%EW?X7FGRpBcMfJ`pY=axw5nAr
z7SqZ((BYakJ*-ODx47W2ar|+;)?iE{S5Kb~gRVIz*xg{n64(zQ6{KDn@gx72lVj<7
z8nS~n{W$S$P8uC4A&S1dV%v~;MLYTNJ&A&}3nA-?i_-OwkE?!{`koqVG;!jC;ySre
zM|-B;ds>H1wjsHd^`?o+>_TsMr=JQW-4SSM&Q9{F6B;&OML}q6aPZ!4@!k5>u=6Qu
zeffTdTU^7O%Ip!W`7(a;$d3=M#1FsY!CDwR?CSOODOmO9?KgDW=x^MDr&c>GMB1Yo
z#g_t*&Y)m%GGEvuvb6BwyqwHv{dKvos<BQ6t^FagiKXc`8cXE+)N?wFxP5JV3um#u
z=^h89yBG$=@uh$Ld{@bTQxHM8mOxU)2)WmV3&wjcWw+F^si#^o&D!%QiS?Z~u_AH*
zS-x<K;V8cgGe?jSAUSc1$FWo0;O!%|hwRk3zt2u^-?L}Cm`nBS+1Ke#RY+@;ov<DT
zEPpAZ7NoeRecN^NS&dg$mU#$7c&ovIyO5Ue<XJxwdb^wvNUBCdx-_97a^-TweUNbz
zP-)e@9@-H<67(Sv9O;pM?b?lb98ro3RlN04G>6}Ik(IqK-qDa|aKfcbmiatr5`)&}
zb%Y<Q*05p2HD6w+AG{Sg*T$x}Y+!*mMz&%0$=#)JU_h(5*C3@f3AgIcLZ7oWVA1B2
z=OzqZEs3{fnD|YJ#u(gCQe82TFDofJJjCYSjj;{4i6p`6%dP9HCBGjEEd<W0&8zdx
zZ*`eq=zU{t*1bpIStp()Cq9gOvoPd#Lzm3%rhZk)Nvj3l+ue7;&pA4FiG~--i^Jx#
zm8;Xe5R$FVvJ?$)kwB~P?q|oJtaO3lsCV1jYV-d6^*Ol0)+e-q$uAsZItD`TWYKmS
z;?VJO)we{Bi>tDX<?)k!x{)(V7h1hxm-xc=xMxAz?O|W^W!t5!c+mw39z7L!dM*oz
zT*7$lr+Q^IXqYU0eNp9e`Lv|*p0mrn{sr9Qr<0VOV&uf8O`ELU3_Z03DHgv)IziX7
zRvT4R7qc{qpz3v-`+YMBbf2&K4|xzIKd{E5M^)$e>Q+1QU$DM=UXz9{|8`%Dj7B#5
z&mWHLyi}#~TKyh-y-1$2u7=Tu(jPM~guc5xZd**z(3=W0z*dzsz#&z)R=%Or2_sEY
zUYA+W-m{V`cs%dz?{V8y{^MOeJG}mv%vgCn|KsBRAN-I#JD!oM@G#BSH8Z--WW73Y
z=B4Hx(#UVdZ%hIp19`QsZhE0nNmUX0z&I0&1+7{*Bv9nj#5s5C_NBhq<U(tb0fuS5
z!g@2U-=I}14UDAaz@mL=iK<y>QrJo&h%%)E(1osO9nQ`zybH3ke#?q*EvIV)Bjgmh
z$@Ml}^z_=$E*lBOiMCI5k{UYgZ{eyoznX~?3%}#W1ZG1BT#S-;p^M09t>a(Aej9R?
zNGfEpi<(}U!}wI1b#syEaIQ`<p5fUKg}n<>T)qp(z~K*Hk(Kkc4}4E$MfUstf%clL
zOD>ESMBdxwt9oPGCN6%o51DCLwX$G`G!_!lv%I|2&i}4an#azahOp~QXwE{V5|<3%
z+8P)HtZd8ZjIA6v3@OzBQ}ygItpV?UpM3~A(mQn>daG%ApLz90M~}g0%}lB=)9nMQ
z%~jTvZc>&LDP+b7mU;D%>8@HP4FG<A5jtCc(ueC<ijf#Igs97B+vv*m_{W+$R0iS#
zl5XP*XhJ)ob_twzxwF6|9K7#<X6TKXtqZJB7xfLV#pyz|b!+b)>FE4Kxl{fl47q!4
zsDn5yv<2j)iJ(Sxe-X}<C<r+%7~vOkz6|n=V7s0)bFpaZ3epENIgBAzAbqvFOgw>H
zFl+t&gVSY<Dt<!snl@F#*-vO(!hJ${f6=+ZX~U*XAK)MGUZXe184do;O8A@qe9}wk
zIad=lj-pN4$2#6~{MQPmOiC^b!~W=1nXydc_$ZG&p0YR5cZ8bUTyd}=S&|co(&YmS
z)R1#ThMhir&})H63@LxW#&09wfJ=9WFDrZkVR3OqZUF&{4mnagsZ+wn(%cltoSSHf
zEf2cYJ6MWb<Qx@IA2qh(Lgu0<P!g*1Xt$Nq&OBN)@YrNO7aiLqBc&b{PZJp~DF4#n
z^@ED$2dhKXh38t2{4%XDoR3y9;LoydemFW{*)cH6k{@3$f5^${rns{jKc_d7hnk=H
z>s_pxFtyH^2WbMDLEqvG9?8_{n(enT9Vt+0$1v>B$9*a$1IljuSMqUW$odOPJDJV7
zZJkKtl)DtSaiEG-73fbsGj{h!h@)b#t3`j~3x8>x@-0V#;*JH@#L=Jnt&w5`o#rkF
z(#Kc5Eic0FV%I*+dQ*M`j6+=!%>m6~89S&*zjg<>-VQJ%2wK1j)mk*Q78$Aec}r6g
z(jD|1_zQO2#-#NI1};iFr_%2cOVi;-!tElr+Niy5IUtaQLZ+G7gf=V`O;ebuSO{^-
z$;4J_y@yT%4(8F?R_9cdZm@jz)b&tVOhJEqsmSykOFXUhA+<nKTer?45QS0$<t{{l
z1=b{8sxG*S*k-&Q)292+PPb{QiRpt_LIHC~BkmscyKqFsNt4JI%lZ3z*dDt@9V#hw
zjTbU_cg)0zqgWgiOui<WhyjEe&x5uwTvIWDZ0kP%UiKdD=j*$f1a&UaOP=Y4FezUN
zt(cv}dkHd(|56MO*m~S6u^g?BWgSO;&CBN?$kbNgiNvgr`xVIq8C3;|s!5<0<t^fW
z7b$PSvgO`}0$1{kJkt+p^@fcxms**G>*1PDuc_-qVs*tPQI#qu;rXxC9|#aMxrc)p
zHU5Hw@I{9|tK`B#QjFdAXy-8$F#lUxLr!f{ki)@ZL_wcM63w(-k&nkXTP)1~8^PNt
zWdA+27$Br544JJMQ~bvd9N+TRhLsyOY><&VU=nzvTw7Y<wyRKiJcc*{l~kxk+itLp
z;=i0Z{-h0+0ATs0wJ&E6`M=}$HY(A>q()pjsYS%Giuq}*Sc~j~a$aVSCU2MKC@jzz
zine}BKfswHPj9?pLB&i&N)P+A|Jjo<rfa^v{N;8JC3|yF*Msl0AHv<JQk<7&R8j_t
zA?mzsSCb;sS2b8a>ztW?O@(iuA8St*`WK_xKzh}~w~rjmNWL3A4u5tEB^+pSC_!5q
z6*jMVKxeE<q!$Bt)vci4I)VwlB1dw{1sat(9v3bwR~bF(XkTq@K?;zRoH02J7m#&~
zcxb`3@kL7*1A~>#jkV)^e-NuyEpzFuZh$><TH7zUDUhNTy(!v%xaLSzlXFmDrTNEt
z*uq--^r#t(NG1r>$j{Fg@+j>~xU~<yfM(R-&>f2AHX|cLAT=3yAf=Zm+Q#)6g;yD*
z6oL7+*KNB8ACK0s_VBG^s;zm{-ANK_DkjhkBw`K68$u#42($l>9V4raBZq?m01xM~
zp?ALPr_{?PQ3do2v2XQ5NP@vkEQi|fo+-NogeC93$0?S`Er#9_ZE!!}4yUIU!+l!I
zq-73d>z4~D&Y}I@huXD3eS!6_!opENPx327q@6xR75<=%;)}+?-wp#%aGIxuE!5Q1
z1S*y(@UkK3Hhp*(>-O^2#(uJoNlxY>O1)G_>rG&3jyh}gQH&w?>^-{;nvaB^i#81J
zc5#Hks&kx#w6o*bk<Yl=p}DgmUvdtMrw&}g*XliZ3*oVG03#a{OTG9nIKrx6u|mV#
z>QL1a#17fKoaghgw!3syh+b$we-B+(35eMp*RgE?gH1|2=@s5Se6U;Fp&Jy)3rJ26
z_cPhqHD`ER&H?0EUcu><X$nIl#=g73(|)LTT-wV3%}0iA)FyTEsTl4&5;8Qq;GTeN
z$cSYIlp<Uw;qG(pZGpzK`JPeyhwtCX!fU4`gxNVp=125tbzbsg*3G<cFT;=m4tD9S
z)qL5~r5pxo()ncNnl0bw*$JOo3|p!wVl_xGYSsD(*iEILrmcP(nau#($9gUBgDkA9
zI`^?9xXtYk!|*`n=g_pGg1X1X>VsJ|%eoeED513F<IRTstJG7vn+Gb2@t-9NG2w%5
zg*W{!+$flU6(5C;6;;_D9?zqBLr@(86DI~w?S0*e_Q0Wc8BJta8tE^FuO+&2aa_Vb
z5<6c`l^EEku-szUo1U#Tvs^I}dvJ9ktDrHvb{ja{DGuFB){(O(&DlppYbBIy>8h}4
z<5-V<nR&x*2|Q(|!j8HJ3@=zH>LjIj)qy<nOVJ3+fa5|H@0}NrXo|X2tN|&3yymw2
zXHKP-V@__W^hAFCTp<EgLi=XwNb=EI2sO<^XSD|82WGzX^HJ~UIfQn&q9bf<x(X;m
zV`9;!H?*3Ci!9_x@J4amy@-YvnVdSOjJ$gM_+97AWcC}KBksi<!kNfH&yGL31zs!$
zbJc}{j)nLUpNL~YhKH5$W5=qvXbc?i)Go>P7Ga&Tb=RF#*15tUK0PA{$Rb(iSs5Z3
z@C_xKZff&QZgYO7oTT8}u_Upa9`|{OVJqGlVkLDN0sf_mfU|z5(Hjb1pD!t4gu5LB
zd_yXE$$Dw?@7C4}7gk4tCUaIu3gSSzu(jNdl(hS@ZdO5N{?9JV9HQJ7&L*RBEG;eP
z;5T$BZHoCCY~nPOKAgQ8U#qC9)_jtbxl8UIz$)G>v21^l0E=NS>-gHLteux)Pmh;V
zOkx2w@yfE2wIX7p(O0Q>yCIq~kc{HYigGs|xJ<OBnHfX(7K66EkUdEYMx{g{%ac^0
z*mW0o``f+?XfAF=$F%CYM%^GsT`TFROSQ4O)=sdY&=EMFoz0Ox0!KEN#|^z`5>#t4
zP>_kT2#!)=Ov#`b0bf`P)^0|BL`Enc<Sg2iOxhI_e#)Usjg4p98Tm|UO9B;3!T>?o
z`){4WqeVzZ)*y%{!wo_&#risU((NIx__5%ar@7hCk+96cBO?NZae;LIQYiTngM5Y#
z>fCwMx)p)ijp<nNEBD}CU?5Oh$K(Ctbj7K6k15B@P46nEjZDJUhiP^%Tz>xL^r%~f
z4aU8Axw~UwgX!O@HU4IK`lZ>o$f-*M>-QM5F-m_;mkT%Bd33Fwo4IylWN%K}!TnY*
zd@%F`vYU^MnyULIc+i;?PdyGbZ1@Hz-#WJbz`|+;5jb1jNlI$4fB$}4-3tr7e)IQL
zmYnFTcXP0UAE9FK$HQb4E#WE0@@zki<5Yo^YjSR5|C!Hc-`e22{^qAFS&BxU^+j#X
zt?PhAQ=FIC>W>dpDs!WN^DP`T$sxoX4<pm?Bv?STl=y*<DO02z!4z^<?yWDHs)}*I
z_PyUpNf~<C2~=55v&fAiDZ=;Z>hU<4>_l=Y>cWig_yEPx5E4ghG$mi3oK9O$4#))B
z^3lP;_t=!umlkFQURa68-EuP$63~b|V^N5jn>DGM{=B5ki@Oe--fOQyWNxpgH!Ej?
zBg^tUHk@Re5aIVsbb1o_{PD5gd6&Pa48n{~x)7wRqK+P3dhEy%<(czjoyeM*qc4{{
zBW228?!0N1A%~u{oJe2n_dbliA=#+us8#FM-6LxCA%<neW-X({5~_`8@o~IfXjr>j
z2Z|-Ja)gfWX(234)UKP!ASY-ILj+6c9!<2_*JV;0#RVCYOiqPDmgDfcjqaKnl`{-?
z*M{t;_v7KM{aRCdNksQ(($nnh6#LY3Q-bjbeWWuXev(h9bNL@NwplE?to-Gv4ogZ)
zOTWJP@y`y6te|tHDEfbYR^d$4OY`fC9!v_WpNtfLkLR2HZ`r}hODFvffOKM&D#tiS
z!IEV-ikW8kL(83#1Z8S3V1R8wubu>P7w6nx=_t2d0Sc{sJ5dS<tObYV#nDBKpg1jK
z&XAJlS69>Dmf)v0hi0F|eJP<iq#)({##n!h_o}5xa~i!(*Bk=yed?k$dBMr}M=n{A
zuUp58(nB~sm_cynrhGAU{*z~2Tv|#M+rskl!rJED8%v7JJ{K*>#^x%a9i9+i`WLsV
zgmlc3&P~yhxsbH@?vP{1ble$H4$X<aGQc5>DevO`3Z3GiXU>`^9S~NNPZ?EasinB2
zstkVO*9zm<X4(sCIMLp<evrz3R&M?1Biuvl3Thl-GZ5rP?bjR$KI%fxs(bqMD>j>v
z<<p?9BLoDJdhtUL_0@xeUG}ED-YC(Ar^e#hOwJ|l4;LdM?!y(o2#Sl3j`XbTKWkSs
z<GV?2#bG096Tjw)%1dbnca2$N29}o1R>4nyf9|5^;D2qNyXi04zIzg~?(1&6okf#-
z!ViW>{Y(u1r(GTuaEoF^#bD>^(!sjAAQG0j1D-??o*i=CBX@no#1}S_SU*F%5l4fS
zga%1s8T51;fuoGJ5yR;vDc=qzdzc*_vVJ*{yE=PyG~Zu7gATkJy}J=y8e+7qsC;az
z9cbBq@m8vX5Bz0(+cAuvmWL&o3LPBL(Gdf7Mq`k`ViTW=&J&sF_TXs0E@F``By_m&
zHZ;sAsKpdq+(e-o<5hApX@35}gOkqCI#M;D?C#AxfPIit#=v><r<V@(yN5zWHd15;
z;j`DQxuWdj@;fb~`cgGO;pTUytRQvvyafw7LVUgi+&Or|&r@P0g3<pOu3gmti<DCz
zLy^)e$a3m6XHGnc#dIShtQk52JiLd~&Ar*jy^e>Bb~iE66x7<9&o^$)#wBfT*muAH
z1VJ|*z(&qWw~O|oARo74#YJTwlu+k0T9wT4*~f~Dy;{R8Q`#wGjfR_VV4dzEnHWW<
z#K_-lhp~SHY>t931pr9ayc}E{hjr+AWlVvOhlhu|JFJ&m19J(_BaJ@7zEL$GuCCd$
zk%ZkAAOxr8oumXr(s2MVS-e1ZBiCh3ylzmhUOT?IpG9!o!-r$>OI;gvUC)b0MPcdw
z@yYS2Y~HFa6jSa20Rferj@Q=@b}R7n#EI>9<gy6B2V<uv5%z4w{6kHV<|`jbiFK~^
zB1vtekhL9J<CEa_^2XX=aT&mnZDfc=EhUW6?%0<!hX@W|OMCK{kwJt&AM9pKxbHM*
z%rSyE19L_j3lI1*N%3F5e$`SwvV5#^IA@ax^xaHOxu7y3GVO3ENS>NR27Mi$nx}n{
z$ekfD9O$=bA}_5E&S}eOjdINK#cId4UhXG<RcSk%uPz=?Av#{h)p!NyjUNG59AKZu
zPa(P524MpEf;fmNp*N=uuuZBebSom7*Dc!gcW`v}%2*Gd><+P9pg6ojHAL}iuO=Z|
zAS>#;eZ*=G<L0cyYKq`mlDZQV$ZTuzslYMy**U341J{ssP^=FU5EHkK)0mmaxV?2h
zUhm-LQo_VV&k;1YzN0(V6zHCIYXu}ZKNZdBnKAW;ZulwViI2`~t?3H4a~=t$?uddN
zdW(E|DF3L_n=hf&0wysWC9L+%?nc$GGpSj^JJoJ%Qj+%N;1LV<2_OmbGK%fnkJs4V
zQ`*T#%0H$}3a5prj<3uG<Uh?lRQ_8y-DB5({xIG<t?@sYyJwL3xyKh&>nHVFJ2btQ
zW1Xa%-&E3jJJ!)&l~$)v5s#b7f8y`xS=GHA5Ej-p*yf)<DcM@=IJ#oW(wj!1K^;4k
zHjLU<S6CR8|CpA%w4A=ka$2kZz8cSMegFNMlBxXvKm3319t0T~Y2R${f3BYVl^50i
zV`=@5b@spf;ZRl;7ul!A{)X(rq?v+O_?iWQY9C(NxKu${yw%wM4ks3?J%3L^NQlG3
zq>>X0?herm-hnEeU*i{8<@rY+`snMMcb~Fws%vGoppl<g#RWd#7G6S>nEd!TqgMmX
zyE|YOZy(X6QzwcB6#_p6_JHhSk{FQl>c)q9*gkV=`<~>7efzk}kT&)GXD__f!CjBe
z?|`{bquQF^L-yGREKo!=T9I&A`Nex`UK1+siciMqx(c+p`q1o($;T*7t9XB>$@%i?
zvX;smpbYy*;sPW`>b$s}G0Dlvr#_J5w3Zp2SjT9DI_R>scd8ipRk%OiFuI=Ocsxvt
z-zsm(rGCcN)z7Y~5Mw-`**Te{ReofCj$^)Z$|__ONrf2KQ}SgmO|4o~vq=wk61;c7
z!@3)FN|QsT7tNnKa6~J@)8*iwX}DuH1~kyV*`i`2xnRP#?#~u|jDFSTz`=u}oPB&@
z)S-PqacI`Qx;}cB+gEn9-}$VAk5vNIUz4tnU39#2$n0pRdi5iR8%{{L9Z|(alv^NZ
zt3E%iO^+TS`e{HZ@%l=M>h>geZE4{$e12*&y8`^xAZ%Woz`~-`>h4t&MCkdZUYtH#
z0s^|AN0PxMBvUdZKyl$)4EHyiix65Q7Utc(Qf7?R+(Iz+rCz&u_3C!~7r112mYX_I
zuzj{fVk@3paMr59EF-C?hRCS;ljx3`NAw(Z@dBs~yY|SjW13tsXGLqetU(`J()}Py
zV?20h#0k=ALw~b(_uD#~Cned}#UzSFDe_(peKjc0Zb&u&$`_lo&W!3svwIsYSI)Y6
zjT&v`Y!GKllOS{5?>#~a#y?^4O6v^ed<-+GIHhGO<+<OF{8KD+GzJ$@wU6N2k$fLv
zL=|bE3c~-8OG&Sy#B2dBUkuVo`bB)vQJYn(QJb97c`|ztG*?zL9u*T8rPIP&d$swh
zTa(I8WN#WR+)!>QhyQ)<2!n+bZo&fA7!J^P@zhE(FT}&CHkg?tgyi_bDuj0-lON=i
zYTthNA|LoUqSpb)G=DWh@<(*JAr2UYZDu<{$B~8^_0wsR(4dqq=r_TLO<x_CC-L#2
zD~*W{NC9;JBrood*4Le%vD2-_{uvRorCMLmJMUYp`jNmCIME3HM_nLID5)eS&f*4;
z4OgKFF`VNLj>~2e+pg*T-XEO+hnsYYY1nmz*}&*n`aYc3i4Qk%f#T?czxMco4P*>F
z0GVS#h3$&PVb(BnI*BKSZEC_|A5B`v{PgO)MWY^;^?;vKx(Bez_?1E6=Tab$Gc`r6
zzhc!YL)Kk*dA)TctlV&B!sWau1>CVp2OsgZ2!{OLPxvbVGXo0RF!N;Yvs1$rmj<1T
z4gAfzrB29fg;aNk467tgu^_#^^h*bb<DE|EIkph@FK&o@LyA^EN)1t^T9m^C<JWyj
z;dMV(LdZtd_f2|B#{p`aCZXlPzI~^{Lri=P-!rm+FVqwopQsg63K_&|8f5<d$BzqV
zycv0TX;DHiG79ZA*(TF&Z3vTTLwTPQdMh1ubj+7HE2NF5&JbhA)84J>B6*wmP|>??
zL`&b6EK1o&mZ|l}qp?=!?q0?eo@b19G|*o6&GM*@?SV$UW>_j?2uEbAAZh!7f5Mp6
zhw%mq96J3q{7<m<Ybrrgf)5GxQEZQLp82{4CNKV#mv;bgz}EXT;wRElX=_+>CnO`}
zJr0uGtF!~_$c-%I@){{-us<L|c){s0##sWcgArk-XT{&&fKH--BPulNW@cnJn8IJm
z0C4+0#chT<6q_hxq~V4Mm@dOK00$3K*|jhqU4K6hd<S!kxtc#eZssnKPp6S7pAINc
zOG#se-+0A(i;W#g<{l9dA)5^wi{3Oi577m+0%RpI?rEHH<!a**pkEIFSF5^+!3j1~
z{=~*S(8;dp&tF9mRQ<^X{mg>S@TuBBCP%?xMTE+A)6g%BOkDsztZHy?PwU{z%eWM)
zbd6yd0ZtS1BRg;b)m%`gP30lURjxvpJ?)Ckegdoqj%C!m1(B%_F}tM;b%w@gUUWs-
zq{44$9Y}BwP;jcW!({eZd^DV6IwuXl5#`~55m-{d^?*e_536lLrg`4tR}(20FyDGu
zSO0H2M1AAJDU9>$D`x1y{-9x4Kt`GbtQr;`-kA8o)7azg4&Yo<={D`s8TOJ1cpJdz
z`292xJeW_R?Bfn~%bJD!4&r=A_|{2u*^g53p?;qRAH1Wo0OZxG0Lo%RdgHdQuZ6M2
z9+HiJ<Q6yBziZ>dkw(V`+VxV-!06S+GNd`;CSvIhRtJ&J;$c&CyWpAgmMqaCwck&q
zc)*F=k>H_faCV~ixsQB<AY~sr5^nC^$KFGx5{<ebBWb9-|NOZN9i#mKsiIxA<^*Nu
z^kSt~vJg)2XSI|dYDS2xlQB5&fs*1-ZdB<+McxYh*2Db)ZBDPE&o?;?&V_#m|JEk{
z`0pXbUB&o7*W5&qGBG+dwnwj?7scH^w7%lGFU7E`fpnfpCJqs@x~5iO8F&cLL~A4P
zdQasHWsDi3aP?ZX4$^0cr%+)Pj>T-R)8N6=9!4|$sgr(XBJH2qMP1qH&M7xic|gGT
zx<W3YuH<Ay-gI{d?48+VOrUyasrklgVfII$eF$iMWseA~f8iWU@FhZlx3(?9QLLeo
zL<cscY;Wr!ofn5&DYQ#Qc;=I;bW_>Ert36}0~UjbxTjAi2Dd`k`S4KB;});Dqt#qQ
zL;v}eY;@&;i%J#GE96Wym1!5}Y_`atJn}I1H=aBPAJgGtM#vH91(a>O3^3eB*~ekt
z&GuN`C_$;R*zxC;eE@G-w%cD<c{2Jk*Eg}<?Pt1+w>a1Z>~I$nq>??O>4+~!8$5{w
z;5uPp51bFMG7EI`)S2c<78=9WPcVt60M$CuZk}bS?&?pwlzlMVzJBz^iH!zZJY4h1
zk)5|!Rca5OwQy*ck_@?-ss?ta<mj=jV+5c0OYOg%J9Zos<?x5n{UJRm;vBq)dDhmQ
zh%&}MM}MFfM#=5Zy*EDAdEfi2EdRz`C+=q5e|&s&GTGIYB$i<}&RtIuww({6RSX=9
zHh?**8o<E{Jwv8vEdHV}U0=kvY}l{^C(e5a?OwC6FeZ#Afy0E%**oUm(97)Q21*jG
zlaRd7nKwH_M>qh=*eQ}TfT-eWaC8))h(rWB`3-K{m)^M5y0~fmP7gA%v<q1?!Qh8m
zTPKqwnr*&SOESt%Te8oUeXLy%_qi$t`<*joK6Xm{lI&r^xKQ@#mlu!FDtn{M!HP!N
zM_l{z`VnhZU1_dkv+uxJb;Ht<lEsl7MwXZ5ZY831v&Gcv0IT;A>v_#npUMn8>Bm70
zmFoF*tE6ZYC7<tViI!G8REab%9-dCCp#BxtEAbcwA&@jRKAN1MJ>(kUk<sIre105}
zTscKOYn1~qGm8As?6n<B5~C6er>hYWZh)+GaQ*sS`91TPdKx6fQ&T>(Nrkg%5}eFs
znB+lyYR&DkN%Y%hF_^4y1evGVutVF)n8CJWY!n!Zdkqj~^4`jPK=GV=O%X}3a+?hu
z9zX5Inr_OenK#!d8qsRU-p!+Gz-Q`%TJlOTTKXs>03@%(=e!EnTJ`C1kk_I5%c^aN
zH16@!-sV#}oj)N0M#jcHuUwT+*7ktzx3f_<zqLb0ygK~JLA?98aZ9|O6Bvc!L$5n%
zR>Kj^R^58MVWL?>ELB6JN{qFhV2z+44Gz&c{QT;&=^;L(R^oD8KZZM}qL4<pdw7f>
zj~17vkk#wgt=o=+m4Jyz@T(L@4AZg8Xoahd?Kx4(N>T=rY2!bY<d}eyJnc;~-33Br
zqOleisC@rzNAd*4BILv(at>%g>SRkmPag<qR<Yhwx;Hke$}LH5G4Ysb!Z*-s<?oR3
zRvqZfyrlCbt`lO_&CD_=OOou19w0=U^ma4mGjB<u<8w83lTClsc*s`l?UGtFV)l)w
z#^0j<d}_&&9CC%)0kzWiS6*KRw}^kX$<yN4yC+b_o;7d5{G;{CZ8$<}3TvsWC)$rz
zE%E8eLqrr5gSGhE8<lY}G5SSr!hyarBc7052gZ%Xr=#dK`w_mx@<R!&;zDOgc(10I
z*Z}Oe=X3-kjvqP-d-1A2OJeoI8*k36bK=%lmVx2|Avb4fu!--MiOTp7-~XiO=(O#W
zTe>3Y#Sq9xaYK*k>XS}OXNv4Pr0S%MPai(G>R0^jm;Ax@^{tH!$*PQ4&f-Ub&N1XW
zK`p+*4m6Z6x{ap8JeusrCG-z5TBeE6!p7Ufx2PQ*B8X+i>s$9v?KC^eK{qvyL8F|h
zaqLOYE4ys8lxIv#6X!^o%gK?slVlgOrb=fSz`_j2q*oRi*Y#VJyb5XmU7Ure-D1ei
zrNW}k=!^G<s5!{m556zYv3~GN`|IGb4{yfkzs+<x+dk$1l!E}vLZXB6che6+kMRvN
z>+zKgSUP-8DXDGDw9yR;pO##G1g$*N@WoT4<|0Pge;yB?Labmdtc-JS5pU*`kCR|u
z%h=fQ;DG~c;_V}TG%t9(_|v!-MXe5LO`Z6s;+R2T=eOm(CK!C-qpfntF;X5L$4W?f
zrhalsQbG9PD_1;rd7j*quNO3P1k5+-;!ptTrfYvmtAw5WuKZb`VgOgAxTNI0m{ejk
z2jUz?{};cziRv|(10YlNMEpf7NAN>wB7*Z-sP;Uz_YgPmZ?1F2xL+HN&1za=V?W1M
z?{NlmIDFnam2A<l^$#kN+xPL~nSpyhRhcnap93`VPtvP~nw>jW=VT=i>qrBS2XqG(
zJ=CI?k(&mRYMuAWWOyiubu7zC>M3zK<y^nECZ_?vaQ;P?$RdNx@cw7^1icyPn6=1g
zx{3n2Dg%$juZ{IC#9DC{5cVL$@kZ@w74&>S5^*I`zngcLIhLiWvwFV}HAxSvD>W!8
zjC&`DZxPS7>-4MNHhr~oztX06zmo@FoN4s7zh~5dB{cmNF<f?WPKYE<!Y^LD)7n4l
zy75>r@3)^nyETVkFDY)_uLPl*1mck#92VtQqJER9dQcm5K%Xpb_Dl}*`daa1U}jh6
zH5*g!+b`^&b>_u@j3<Z9rYF9RSifmQK$DLBIx+%m-G0-gO?xxC^h_m)H-Y_S^}XBa
zlRYb+obmBVO_Zv$ic}Ho=m46k<5I-`nlEZLg&e+*bGd_XCrNLn6h42Zu@+ar#GBpc
zS=s1aBt8Z8Tl0eoDX05yN1gki%F52By-zNl^<?xC<F+mo3X_iQxlY#|*hY<v`3`=*
ze&OHWwCUM`qxl|BB#f>BPo&!j?f$V$J3d_Zx>`=qhxDQeKI?O5j!JpXXfuI*VQf&A
z)uuODg>)$v%;qOnJdjb6-)cW!?QAx03{OEV5o1453kbUjV`bsviiBMJaUeY!ueZBR
z^5b8+;pndu4mBy@Qq|Aes&+ZnZr`WZGiS`uyB2v?ggafQoNHuMRc0N~6@{?bNKET^
z_(2n0Pffyuhipfo@&l2c8t^&eW~9~RgQf}RKkYS!H8`Qa!D8k(OLKQ~#`j~J&OAJv
z$YA-C@B2<J8F4>7f7p_e1s{jmr>3XEVl!Q|i6%JpgI;!Bn}1NHcY0~9Ty95iz6BC6
zk9LFN!h&9$rCzh9*{3+ZPE)EMv8J@Xx$sCL)nCy1BPRGwZPH&zcR<C!Zm%0`n=3y1
zRCEgTR=joL5;jIE;+#l?l^rP^-MlsokehJ-#fjO3j<r{9KegU8l?VOy{rfEh98FLK
zPS!0{UMYVqo3E^!G?|)XH0;Lr<PNC0lU{Y!v|2rS;hJHYpKxfRt<!4I3I)<^yu<wM
z#tCI(ceMZhb*4ajCcdgv>wTBM9wf9rP`IErlb$~sJG<0j_9VX-dk!~Gco^{W$B9jx
z{QIGmDth1AoOJ^@^flZj9cuqBbL?N`rH=^9tleYg^;Okib5B}5jkpC1SXDu>dExRh
zS_Q&=0(E(ZHN@M@_OZE18e>y5sy#&`+_Y`pd0V;NQx~WH<mPcym)fLKy7CWcQgvnA
zp&sgF)6!dz!CV(+Twn94_PJS+x-S;@r%k?{e5?vlf19hT>!o4jl=uz}<=jVB`gzIs
z()V$-NNK4PJ0y=BT*(TujB>bcQGzvP_@{m2dWT!>h8Pr5H<@Lmsz~G3`vykX4Zxy`
zahO!ORo*9b#&vDfa{d}#FK}J1#hjnxXYHk<>+-{48l_hJydJrQC(VMB$)l?Aw;t44
z;%^yT)OKIrqISd9ng^M8rIE}M+e1$<<S#clmV>?Dt&?PPE~Fw5h1aOo1(<?mpnI@5
zV2b^ZNyZwsV%<!?%Jr#V=i6Tmy1NveJb3m;?$tP?19PxVv$EaZN<!A8xpY$Fud9Y`
z3dX{z4N*0|RUoz7W`wAGapAOul+mx(AY|z9oZqY!N|hLQ0i1qWI_|4u>U|a;2{wRl
zN=>s!`~{#QxWYJhxe}hW%RYkPyNwjtWFh)Z$S|6=Z{PmNpp}});^e3#<bS}~Ot+SY
z1Q60N>EGfGP@Zuk;68s@zh%qdzdbaiPwwLUIc5#UFIFM<csy$}hJ*4XcALDqZe~<Q
z&h^nH@rePn+Y}clKepj3C$#vPTQ+7dyik&3BRjjXm61^UcEGH4gF1FJQrtPj+4jTj
zPI6N3;0I+N;N+L3`Wy4kknrrV3OOX@D<_~e#CLX82W3IhnX$$hAVDhan3-q%isCCS
zCT6%Yh9Y7C6{I%D2n><^G&5@`&9xV}ABct?qiTTD+1f9DX@!iK>=Z)Dr)mJ;vw(pq
ztAK9n*Uxzwc3P<id>}v!1>0C6nE)|xXR>HN{jd6@0rnv*hpyz~Qm+$FJy}O%P)psV
zZArxMrU+64ciwMG1&!qC3OmK4S@Y%(3DSDmZT^88EY81FO9jd>qa}V9!zz-cj?gF$
zz%?@_z3?YKs~Tu_=<s|>f+tA+Xk`o(sRb~Jqz&1eW@V$ER_4kBLbvCw?N=I)NEoK6
zX{d}rZIGYRFXa1ID$n=X*<ImrCgJ9CRL}8-RGbJ~I{#Cjbp~AwnstyBI%ZF+!A2n#
zk4T=fps`(|D(_}-{zxTP{Q>%&q_zKIvS-$X^iMC(ttjuEylSKoiXAED(Q0(;+I0+)
zuwKxl-hcfX93h?{f^5Lgd_bbL0Pi^M1rO@2T(hPT>313orUO%-_o6kLF}-g6{oH)P
zU6g$QR^x1J0;S<Cm=1tKBM|tsXy<76Gv$vLUu_IhH}?6hRQ8#4i?vzhzX}Sx*zjX%
zkyLdNK3+Hd@XsAPoSatQG((M6%mJ7rkBTy#dt;F<fEsoB)V#7^_uESO%NZm|-?Xbs
zj|wq}1fv5*(IqI*-m6pB&t#-GUEaO=<<E@u>BB3}N*q(UlPc{XW;KPtca$2O%`_T4
zg}<lIkt2a0L#KwGRQ7=u#<P+HR7+w;Th_E%8XD$#dd}rDWH8R-$KxTo`u?rcKfet#
zEXHr_)p^p%9w7Ugn(y&U$t@2ULC~q@BK|xAHBJLdcKH2dVElDtSTi0Tx4N0Ul-$Al
zz*QcTN;?jkr1i%uy@I;#*4wD8pmtI<P#*B~3qCc+TlT<b5>Wf!P)w$=5AHsBGJyk9
z)~Oh{6wK}7SMgz%(MJ+YGnCI^SFawTj2*kJX5$4QqC!DXX&1YFkyF3=|HTEUJvpyM
z3xyu<DbyJr;GPa}7FID_xqkhz6DJ;$9mTVWi;p*^S>KVoDv8WK<kWUDts!T?I0dF-
zO`0LE9n6Wk$}uVLT{LP3Spo|l9h$(X4S(7N@%_JJNGE?lvDKaqN8+o{IWb#Q4Pblj
zpO<DEsTSm@6A-Y62YQ%Iz5@sm++ZB21jVsd5^fzT?N}7XQ-@9N2T<P)hr2;ThIkNp
zI1a`!n!_6G%<7NEGOg_=N=sE;wqFT_M`+BKj(h%mGNcKt<E?5o>e@~?9L#9YnR$>(
za}4LBP!s6w_>bAIc?oy?KV?Ok)m*SZ(42X{^hLtM`*oDXKQ@wTbmF1~=*F~~f^kep
zyS+jqJwq)hEDd4@7)fW^T%I5gwpsmjl&*Ot(=aa!6aW&K@h+pr;2)d;V_L-=&kzSt
z9_HQqgD%K#$|fdEvgHn(IMIV3CN+*AsSZVN)@vRpS*>}XZ^BY6b7iOkU9_(93&aB!
z?_v1eGrQzM>NvwjMZKd5Mg~WtBmA2#cB9l3Xwm3uy2Tf;V?9B|9)aCmi5G{AHQ!%N
zAv1~tlj7PIFjF-T1&we@;33*(`M?vA+Z-MqE_9Kgo9&CT*$QUYS6K_of;vYwn|pR$
zrE}-b@o+PGjNI9%KBb5MU!M8c88c2;%&Pv6gTA(#@kj5zLuOx-i7{04F)YzUaGA#!
z>NJ<B>*O4_NtC@d2Dxh=L`OBzM{hS=_ce7J2b>S!9!Zak=!pkdIKNU^OFOFQ<+O6~
z@t_Xzg-ye3n7tPE3^CMp#*oTO+jn8U9+{dU3&<;ZM}Of7!>+o646a(uMi0)ss(cUr
z<LDIipjWqyj0`crL?*ulo~SQDz<u}bIJVtN9u}DR_F)h7CR02j!s*J!TfyvDspPnu
zsBd~ng2Bld^YSHa%(4HrqVv;=HLJLAXJn$dc!&@mq;8?1BSWaTqt9saJiegm-;N?$
zU?(YGAh6B_C%<^*iV@1kzI4<ew+y)){14Rv;_kF}tz7>71S6wqFPjej?{0VE;Vu73
zmiTu1iAu4~fWrBW|LFtsTb}-Dl>~?X@GAU&fz#l)U*$G<O-#zY>fuY0<G-D&WRoO1
zKl?sCLXdz)y)87*39m@-Zb^0&LkW^JbK&^$8c~ILTS9i%7XH-0grI|662Ba`d;2SS
z>VtRhx0rPrH!trD-19Nd9^S;MQEb<LTwfjgFtL8z&L(liD?bP`aZT#}@=~L!b4knt
zN7df2(f$s9U%su@e1nQ|NaSPJU;l7_w=17<&LYTf{mrw(J{1*?s?)S-BQ2{&!#3sL
z7SK5_9Gioy0|o?@|NdPtVg0b2XqjzuB=7koNoZGPJ>Jo(S1-xdx$f=c7Hhn^Y0AK*
zrOTE=VqQ_%Nf&*QSdZ|^6;Jk8nbLi#RBw}e^wS*Ls=sQL{aTAAeq6VD1Y+omy?XSR
zS@5Xqmq}|+ottv*ExYE0@#G+L0F^Cn0bK}Xpt*Gke$M$fm})O8kv2qW_Wgfgg!5UT
z12rzpY)P)f_$I?Q_67AEKi5GI|FbkBvN#8`H{bcAT^~XJ>Wu<t-}^#GT--n&W3Vfk
zaBRRf1oDO7GIo=IzQ6nMAgSU01M6X^HydhGkE}BBA|yMyoI^A_h!c9NL6vHLisFf>
zf1W!}|2)zAB8<k5kG;m(PP$q2Ek=ZLJe+Uo(QP>es(WoLdC`r2Uf;OyAWL^Y;zjrM
zt4Et*n>1;MURzN<R1DV%S@X-b*A|7fa>^;v{sF3oj8(ISI`5&dN3WNBs>C$qCyfB}
zP>b3`@DW1(gMlTZ<Jhmwv^Vsj|KGcZ<;B}?S2ca?vo~^6VfLHOnl|ZYYjcytG@Yso
zxK4V`-b2>j2@jsycg|ab5Xa7wj&0?O-vMy))?Ghpm01Jso6yvGs;A}$Ow1WvR`+a!
z`hS{L)8D;2xnx>H1?;oui4)!`xAyqyU2|@W8-Ni=3bFEMX_O<Br$5c6R6ZBYk~#9s
zXFHc|oBcUqH0PFAJ{>tftK@UMD#riO;o3DXZnJyo0#JFHqej9=%Jyx(U>os;<ZP#?
zVBDSmG~k{N2Rl|2>Jgh)XVfR27_=F$Jdr<0<N2N8Hqp5TR#Ef3rMZO~Y+h_h!4chM
zEs!M~96ohxxfU@p@tSFeM%`yLgY_ni0GHHdr;`U}&f%Jx4Rjhx#(3zHHo&AQQT*1<
zz4vJ$yyGEx-(1%%J@?Jcq8q6`!CG>IscE~XOF;be<e0Bv+)hXnA2#hx4z-*8*hjzZ
z&2_n>NCaySS{+|=d)=1XQ2<`4t3I5B??5@v&a>y_>Zm3W95wn|v&s0?bk{PBf(~@r
z-+Q)Rt6}zyGO-UYe&^QK^E-BqBj{$EA}(9{`pVdi3uP`~RrB-i*e1RGxxd4MN+34!
zik5kx3*EN4Z%PV`%K@iPFFmR)H2<eFZzlaMjk)(q_sQoxJs*b}D>yy-&|B7Mt=mYj
z!Tjd^R{XuIj-9)z=oI1`Yw4}^s;qL|lR5&0P@4zNym{5)XI%vdTs8OaA3>{M-`q6t
ziS<1e223C+B3tMz4IG$2k-zeDA^Q6Csp{D(E-2kCLZ`XbV#Xnl==Rrk354`#0)^kA
zsAL6HS9<*U9-))E_tG9_LR2+Bw-p9y*Ujp0*QVQ4ddxk}d0yRX@W!N$Ck8pSf>z^V
z?4Ok8#0&%`SDv=nuW1#dugJ$MZ%e|L<PLkv`72a~{BOw~G10v}-A|E(Wv51jTHM&Z
zyLRP7%t4Nyb`8&hAYBEN0p~XNjG@PDTbC4m4JdewsMl|HMY*`n5k0-~Ng4#v;B%kF
z`wZK(?!@3tO*R$&cyc8p)hDi%3vW>4!ek)+F1l7rj=RtKabj?wm4D5l!-l@a0^@H+
zMs`ZL*7D37{$2xl7i^VG@^ps`j!C_*_U7i=SV$yZ4&9!p9D%Y!)qk2RC9<nds#{ll
zRk1HE9kZy?x6i0TrVU$0&8v1OvLL4DYVqOTkQnA39=7S&qi>9SBd6)(cuylzrIGTo
zesgste)MvHCQ-=&T6;VC<-S%?B4fKU`uZI@UhewYO;J5cau*|2pbW=K<NW2GJZR-@
z_D9c8c{(GU)7`eVJ7v}>awBEpT<F1gPQH7iBHGq!N~5kO?UFypxKZ5a9(WHKSUsxW
z>jP1(I*i_Z2b+q0kDg3el>TVo_tb|jg+nS*XN%1NNm;kJGM&EN9dn<PzN#pcV18%5
z1IJx|{n{jZ!oCuMUNK`Fly-4Nr2%=<swygPUmbU^xLMb!!l_IKN)WH|zh<^Q=v<;6
zv-Zmz5ZS;s&D>u`Y;RGwrEa5&BIgP<ttCq|nhmOVu=Au*WV?jI&0mtwz5B7Lye7_(
zT{PP0)PRM-*iwDnbzipr#>)_&Jog%P6MnAlEJ4ce{Cr>>0d}R&_xyRcPVY!aW?D9L
zYBsn5X>GSpLl9$<3(xhlI87QNr|{4pKOT7X>vCgVj$il&*Mj5SylBz{M)ZJAAv38G
zbJs+j>OB?W{inw#Ym?5x-R?1=6m^c@l8EKN$ZkoO&D%oT5G|I3@`wvFSH3S^`TDwJ
zOFCksaMhTzY}xMina)06!o*h%=FVK2wN{bm+?BUazo%P9FK-Q}VmD3sGqyfNggG=M
z0w+iGY7zPTc<HiZ?XQ(&1RT2>m=jaNgf2lyGXCVf!xJaaBE0(v?-OrN4jGu|RM!9A
z-j;5(eySw&`rvSBq(MHs-#d{v*M|-0q2bZAr=h5oeunhQr6&LqGs?i<*z)GOo>W$z
z%MLYZB5;&w-K~Gn2(b`Fxj-%Z6QjePkRZx<Sy5V2*!V{DBIw?K)qlyeQ*%dfKpE_K
ze3j&4q(`F|{M75s2$*<w^wP@eF8lZDFCWLQsi8jn>5S`r7JhrF^Zj?>vrHO{(E~&2
z0%8kfLcjI#k--hoC)UWks3JmoQiHHXxw8;cb@k|WzILtH3j3mzk^#E7(k+i{KDIFg
z#iy73&v#@qR+Ro2(n`;v8O6ac*fl<9%jEPs2oH?4lRH!AOofZeiYG@KmX#JfZ_z7<
zwh-ngF$2A-=`t{`5Ag^*qS4wfDHGDiCN>pau}#XY!D0=7?k41#sc}Y|k|9(T=D!d1
znp~Ye*1coN!Q!SJvTTV(Xmv$QQf@~Nzr%@n3Rhs8^b4(}r3B4IJWqsw3_!H~=;ej}
zpUw4OcB!ji&M>RX6Z%Lpz9um7w68_Q^iE)eoyHyRuJ2I;RG?b@%?Zb?qnX5fVd=pB
z{bF6w@Sr%xxvFbU<0`5qecQE5wztLL+%|KT)&o%w7%hEWgGEUf%{KWTT+@bAF#B~-
zL&~Fz%ggMcwx6+$hEDPpPG3-+vSh?cGR^E+(CS6hfQDo3&=ZmFUtIU3h3HpgCM8+9
zX{7yHBj#%I#fqbIqg*{aYB6McuPLbofIKW8m!O~=w-W=bRlvMQy6vmCly<LUN#V%o
z^0MbwW?OA9kNSRiwo{DzyI4~?kXLNd8sKU#R)Y%@j%`Eo`L{6LuB(q>LWm4dqk>4D
zvOVX~8q}#eb}1T|sTD7K*uXr{vhRe_Clv)#|3<CvjgI4IM*CQvVcg^9{lssQ<6CIw
zeJp}Smgf&LN^dFtW<HTfabcgeU;0K@rV@4w$Qqb=${d;-L*VQSzf{W{*f(TS&cV@_
zU%73yDKm7)EAETp(4E|-u_<0{h3BVrwG7;MsjM($dNa)Hsere!>KT_XE%;W%=MWhU
z6!ks1mP%tSO=iETcy(`oYTn33J<PKsqfHWS9qpR;b)Z(ntq^FEIl5CTv#0nQMpY@B
z<1nC<0WCeszYoMB<g9IGZ6vn@sLm6Qd472|d$+XY*#f9Ff}UP2{Qhe0_hbEWk7z5_
zJBsDS742`V{f<3i4Ti)tCfH}Y_nm)#-k(d>0`1fDu03da{9=(m9b)n&iP(|BOPb(Q
zee`S@22Gy6(RA|d?aE19=VTXUo(9>pjBtufiP*@RF}-A@F5SfYd)m^eKAIABtiKA-
zhH1pCVX+hUUAh7h*LQWm#EapVV-JN)H$5|Ezs~m}_Vf1LEg}a-gUY6m`|TU3Rd<O+
z+rnSJVlE-iF+aUMuUB&Z*6@gk)SBfgp>DMX+9j$|uf%f3NAFVRuy8^YpL<|lrv19s
z;3MMp>z9iav=DV^Zi~r+O={FI5Z^i|$KdwQ&&-Wp(({RiLO*pgLsULJ@%CVRtx?p`
zuBoidGBphYLf7Q;htEFh{JG&jMEat!Oz<W)hq1;*YB>Mio(aEgd?CX5vU<B(oNu-`
z_1v~05q4VX2p(&`E!daI53w1QxOH)SY!x!Q7wKU!57U|}^@>vdX46ZtytJkpX#w<z
z-gjs(f<JB9u|u6Su{!;P1q0{!Hi)+p7QiSbtiHNeJ^iYkTXeMZv7$2Dyakd6ML+e~
z8&L7rHNdXPASGkht8rs)d#6Xtowv9q99h{+%Cl=@8+sQW?=ZTK7<Uq7?jj~zJFlAb
zwa91_snv99e?QYZmTrFEQ_jw=6+cyF`mPHAxJ!9ict#{jEaj*DgFo`d*2`)iANV_n
z>ZGk&omRKWie&IP(iTnU^1pIog3au&cw!|ZbPyv>2&;1J_wkJFcG1E5{s{|*E+G*%
z#g}O-4;h^<T(P-;W^$4MBKLXc{UeGL|Hbr)uq`Fcb+7UCgx+qRi<qlXg}Q|7p0Lxo
z|ARw4TPp3qUZQLK0|1Vs2&kOAe17&`OnM{Nmq)Gl=I(Iiw*|sE+sj@mGNvgX?l}Uv
zwXjyC+M~ZedC|MO92iv;R<Xa=NfNLb%a>beEo{R$S>lIH@;R^JYtNqlz!7a@60H|i
z@<q#a`}rjzHNXA!>97)c2*4yRvtzcn@7#IkQPbXv5}I)yhYKbUt19m^HGCY7civFE
z+^&1cswXFhwd5EXfe@_E>XaU?s<E7-#g4hNloE92E*n>Wo|p0Zx(w~WLN4@-(A*&g
ze)Vo0HMfdO>eq=fI7S^G-Ob#)8O&yz?O5Ptk#=nT+NbL?l6~SAHB_p)d>k0;YjbE!
z%6W|$ur9rCl{!@nOTGj(+M(?4+pnM9U@V=2(N=@HJWiPdC3yv==+dPIDYi<#GjQqc
zko1T|SY(#b(^R%sX8%#+^(UO%RRG?CrG{<I@{yxQe|YDrCWj*IRAO8`iwjo%19R`W
z*N}MJZ0LqQHPHo9sVOdYQGcpW2FW_ud7_bt$v$nnWbSs~tN<TH>Axgd_2-SrBj^*&
zU9`kw=FRn5r%&MDym@#PwL3iY=H187J|w$qEz^XE7u$dSEPCpE2$KBsfQc~=URpN4
zG-%a}u5|w`?aHq%d)1D@uUXlLy~ep@DV1Z4#bxPLxLs&?*C%u&f5PmD0o}~q_W+}j
zDlQqOTl2HPVwLK3nz?`;%zgU$y4A9=K4^1i+_Y~z;ObRmf^}(>PIf4oMc>f1dL5%k
z%!g1SblMPsO^?o!SB;nkaQjNmbFaTrG0~OYB1~Yo@5c6}`|2Tgnu|xH<u#lA4Njpj
z>@2oIw0vm}XH^L`DsdQ~avM&oj19n>(d^UfM>$_=D%T2A4BILq?G{3j;u+n!vJQ0k
zUHjElTUV7TYKyZD53C3VYWP)LyTyRJ)F*xCm#kqk#S~A-0A7+2Z=fWGB#nB-L;uO5
zqQAoFq_xq2Tuo-QYP8jxVHZX3rE6^WfJw)<LkgewB0I5|=f{8)Q94GJ&V>J?n4I6R
zQZV_k3EYfPyW8x>Hu1FKpFCv1df@sOZJ&1*_aE-BxWK5~YEzybICq5DJ9w$8nh^FW
z&pIsqZbeN~jf851@11S^@-97dzU%4ZbO$}XDXWz6-xKUkenWKe!?~i$PhA^O`;8pS
znk7GGD$k~=7c8Sc!Nlt1#hK_nYTz>soM?L8tdq|2q^giIWhhk>@D19e<BtuUmb+#~
z)6AiDa|_S-^?v?#xGNLKGzk0N-QMZ!EG|9m<XT;k=p&D>|Kb<7&m{*V4={5m;<RX_
zz@zu|3g7@0=uxbyIkHdlql_vNUKHDU9^uJqSp<zLnV?X-y>=k&SMVs|WS*ZDUX#-x
zZtwhi&0c>UaM~eHc?<UQwm4;WbMmDX1@VnVgn*_*lxDPEs=Qrm!aSmtM|4(dvP*?x
z#gM{wz+vzTjmU1rUn%SGe6Tptcm^b#7|vwjIdhlqX^AafT9UtLzV8owlU)#m1m;U}
z=w5Px4j1j$z6hEwBT6hW%LRYmp0@hat@Z8e|C2_%q7sbp80cM(2|}M6{ign$v}jiU
z59YcZ&x}Mmz8`Yp_~W{b#W9-`Uj}gPKRrrLI;%{r)1&r|>NRknO!s*nFhMjt(q~_s
zZ-eOD*<@ID^tgUi`*!bMlh%&??~hNtr>iTA9mb?j2{M<uRdB5-(4}6#3=`@K$%YJ6
zfZy67uCYs7j%#kU<Il?@rn#OiqmQl~H<VU^xvEQxwn<Jj3?&!c)pAG^^5{_@CNeOn
zZ##vZxi!?s$HK|WcpHh}9AII~l22v97Qv?1R?J<Ko7iD;W_0)Z`->L!d;NTU=l<s@
z&CB7Gcsr0Xib!C`fli4*^GADIA20fm=GxY&SOp}V9#6ISLt1Wl5xs@pdk->wUq~!4
zmO8AuUx&O)W>*{^uv}cT>oJ6!6I*P*pr@GKsGWIpeceBKpGKnY4#T1|%aTZcWbo?F
zhZqe)1nfYH+0{I-UP;|%F6c9zwg&%$RL|aG)qjcKl&>kavuwSxSBK>mWhTOxJ27-^
z_shYC3obIK;BCs2<45|gUbt{-r<qH&SG2eR>*gO7Ur7V@Q(XalePi9;6{c)yWmHkk
zRju}&%byb_fB$8iKe$1C^O`jbLhYhQlB)H)she{EKssSz9fQhc;dee8EVh~cTvb=+
ze4m5t8VDO}mL+wp{OD{Z!~+n7kGX$%*RX$j6>>W0^=qD1rB6ZsgOzbc{v=y}_kV)x
z|3j9z%fsa#0p&eFVeOkWjfy|#Om;1ro9>@5-2PRs+lAwN+;+{l`r&j>Rl~|V(srW1
znf1EH?nwpp<FAhMan0X8^rchI>f8S<imdOi{`o&;jLrV9KloMH<wnK&=@wROj-EP4
zrCaw2!>cyFZrP!`rbVMAcLUuU@2<9_#}32ses@PL4O?|Oa7XR_ox_?o>Xc%mY8u_S
zg=uJ~b#a!8f7k8~PR_;Y4O3?>Xd1Tl*6f7X-!m31{QP+EkVS<jCoTLW-^{fBe@f^6
z7eDOZ5BYp2${jy45NT#eiH%f=ec*q4Z3c7&|8a-6Bio_kU7I)A%&cCUHhX9~2}~_+
z@B~95`|ptHM$c%vibSq3ZBe!59m;<1s3DdBG?!!1#?u93n0W)elI#vpAnu>i)ho-N
z9B6)SXa$6pfZs>OM@yV{7=DEcsJq2^cS@r6J$p_Fp26uf+cJWE&RVYZ!R(p7$-;%5
zX?y>d&k1-ou;2heB_KOBGat<knDsImU90~Ty&*$dl151vfYzS4&5Elu%mJgti_cl4
zZF$`?)FY?mkV~7zDibj-eggBcK2uysO-}<hv7Bzxp3hub<+rpDj7E>1PkbV?h&#Ug
zpXOHM1Hf2PKjE|`YB1O+;(H?MbdJF|Muz^3QES_5X70~-u`9o(^o-m{njj<2Bml?|
zV0a|rC;jVtM5dI1P!=}5TYaBgVK+?nW^X!ox6g2{aFoe_<h0otkwfV3%Hb?}XI>o3
z8Bv{KNKI3G)nC<at}W-0?jIPTDJ4b7wS<T)h!r5C@IEQZm{hh6>MaBnA%iI=t1CVg
zw?<i5G{FHK?v&ZS$pOyS<P?XoEpsEqr2`UysEdJhj5cf-Bn1I6s(q$CeAzh+L=}?E
z6UzOlylB|#XvSP1q{xk?DUN*+=M5R>orp{)4>7yQiN2T7GSur<XP~o&ij{MdCY8NV
zDxv)%(BZ*6{fwpF8IUdEnsi#m9fFaesaf;&NuVK4ALGsXK0Y)q)s-~PeXSwYS@`wq
zM>=8aqiPTZo)VHzaVXpu+B1)^klS@`b!C5^?^4wT^4$n6XYzGb17QSsl_SR&YK$lV
z5Y3&B_`!qUK`Yg)2v_mN=IMdi+~<ah8GN;ej~|;SfvNdZ1*;lx$MQk~c(tklw4#%C
zX(vv!`d1V&7?!F<2E9{y)Xn?dJ5`9!#0<E@S%}B0dj<Gq9h{_BU(w$1b52M;quf@m
zU*A+&#g#dH6}ARGMy`7XgAJ&(V`&_wBi(8I4dW62#RV8PaNtz7O&4VWgcai(eT(jW
zEqrxYsf%|M_d(VyZJ0#p|E=_$T8S*u%}osf%K0f#WPV~rf$gnn;CF$}(70FyuB~*_
zGfh0??Y$dzhfwCZOgD>@RHiZu9t05ZIqj`(?(6HW5-O0IWxom!&d}x<5lLUu*$%#5
zv$#KET1P2$hn2TRuZyN&oZysw@t<-{HyZhzzOKX;SRZX*^G;2I2VUu~H)!R1qo@L#
zL&FYrpVB?^5<r?5kI(_xzCL5)XszD3qrmhrq7!NKaa)0SQA*9mQUpi$&RrRlUB5o8
znW}twtDH{8o(Awwm1Zf<&y$igzT8yiFpihSJQpq!EnVXqgc8K<Dhl2iCP}-m51bW|
z8dY@zJ#|v96yrw^9}ee(BvOgZGv9FdoH=i`mfh!z&$2D6JQd9IF&&6r;S&u^b_WB{
zK(}a?WBU#rs?au$rKHV2bR8%IA(zbX^d1WBNvwZ4gYmubx5&T`Moa}nP0aDHeQkod
zPGm)S=>~o-_%zI+xSpc%r3*_E^FKFCCS3wbC+~F6=(Lv`qqy+)Up^i^e*83PK0QXY
zUNf#K@RcL)Z~9x@cYB@IH5r99{z#wpnT4i_`F=a*ZZFK-wM#?Fl;)>LRwXIl0_%+y
z;@WsOjKpDy+0Cr9Q2+Z`zrr+Wm)wtt_nidIlE;^FVWtPm+JPt*e|SCZ4nlmSYn+T;
ztwy>7*fga}DC}S33~UHap}l6t3|t%Q4)+$7@timxw&EWS5mAa>+I?VrIWN4ik{~`)
z4G9~$MvWS`g-|=Wi)`cijhB%&Vr<bv&?2T(yXsw@8(Up^gwP`s`W34v2oZ=-y;iMW
zy*5v$?lGy^jiV|4_5OSmYv!>>gAC6kP}oSiggo#Rwb7qLlO{}%j0!W6+F0@t7-C`c
z35OA#U3f_jBvf@}W7M_N6sJmgowM2f_~jFLzQcX6C>gR}rkvwEsEdd}28#6V5q#8_
zhSM;Z8I4)Xqa`VZdQYD)t5h?9J0=A(IjG2T;m&LZDq50kmmT@*Mr30ZgWa?Q{ow3K
z<p5+c`(a->@~Hg8GL?V`r(xvMrRNVs!vw+*=PfmsK(RQ8>||>RFPK<orE-sx>=bV!
z6N2QJc^e!WWk|S8q*dyA*8AUWy#kwjC>nx?LniE--7le6S!;~<Ydspk`Ge}zeo>9Z
z10oY1JTOT*IkV&@Jm(Hix0WyC{`MvEixCavK%sX>`HV%q1*cbh>bJVN$;=b+!{e2|
z(iZvHRx^?*F$THcx-0RuM*ql+4jrh^EGUQ{;#-_Z^JD&`y*sbLViAixPE5-vhrv{F
zc|GeYXvMWSPx>r-rE?y3skEgl%8T@|jI0l1&R&1W#=HfnuKg(FaM=_w1itlW!cY7k
z<(>Ik&-wrFUu7*BA7h5nAeqc4B!k8tvSrB{nvvyQXbi>@()wXa2%$ts6xl+nr4^%6
zQQ1O>Q6y5{rKr@oJu=^O&H3T{0p~h%UGsyvG`-)i*K>J1?%U((K9?6xE^nS{KIW>E
zyM|4Y(ed3?douQbJf%({nCJe6W~&)mfec=blopbvWE)|lmLrBK74xCet^V@|?U1xt
zAly56%Ek_~UOTB@I&8@Q(!`cVlkt(jMrGW#g{$ijv^uJsSMvX4#rOOru4lowmMmbI
z;z1Fy65XT7LVNV+0W>ph(d41Eby3(TMW!qHOM&&%f<=YPrVw2*hyO=XE?d_;Ms+WK
zaUmz&&3OT$`{oho4qEdgew5RNMfcMBGaxrW1oO`*%OnElNciZ6ugOIk19K$x2cC|u
z1rI%xXC)<Dxk#^&*C}ewn%c7Do#58gIo&u^1!u#=dVjIx>Pw#?@w{lK_|lu{5|-Ya
z*ejhCn=tbyMML;Y#S;l!VwGwIxUHU^9zLdS+|)7%z9^d_4{Z+d1b(;Ux_H!0b>99d
zTJqC)ya`oRjdQ4S-*ht+xxdT>;r@C-No4CXA~7AVwzX}|X_*fq+nieQ<Fid|reL#?
z0!b2@c&{234^<@(k0rB^Og?H&+qD+aF3O&TxUnUL5S5Nj{qM{Y#;AMZ+h<B})fEU_
zMWGjKqGTimINvPi8Vk0$w0KKP4Ee%8@*Fnhj~!iru6YV$RP7fpenAq{l7ghf5r?T`
zMp1aiZuIDg)NdrnfsFZS6mTU|NPR3heeOI1TLpTMm5P8?WEjVM8}H)#&6Mb<DCODq
zGjH;G=HsL`<R);c7X@dVlsjGS`ul%Fq%SU#Nk16tq=Y!$X}nY_vnv5c*r*0Li9m?b
zXq&;tm*mLowr$)mV&%t-o5D%tKYUUjdfUMl*OmrFnGaF~^5&Eq&AXHMX@U{wgCA!!
z^#WJYP-?2n+Q{4k9LT4aPrAh1w+K~kvNdcVabkV}0bU$_dT70R#9Z$G)8zgWJQyFS
zQJ2`^Sw+9sH5*@XZCzad-jHOa!)3Q8*Hp9KI-|#)!hJ8u5NU>9HjLLcH|5G;YyQc`
zid*pGnalG6u#%pC`Qb*;e$R7z7G?XQ>C{ElogICU9HD6g&RG;iFHLFOS|^IrZ{EZm
z((?Yfc{@4KisC&ADrCUMV?xg4$~ti1z|+1#=CR8k#b`l1&DcS6jZ{8-=2~8j61+TI
zo}KJ0MLt%d6Y0oGtM)xp#&&W06Rail(c}hKxqg{8jKtlO1)BTJ=uJn^lvJ^B;gwCO
ze(EylQ_z<FL4od4Or+eIV;!V7r6cQn9Kbu!_%nR*(`R9IP(s*JV`6u0J<CUZPOgR}
zpM2rMg}z_uG}OE~Ckin{tlQum-xMyjt3UNRS-3KU*MvuTvb&cC^a8Et(A>3uzgEh}
z#`8<llj^#uawsn^*%zQgZ^kVm92JHYI3j9$>E*C`)hePQI}gb@klhf(_-aeb8U0`y
zTJxH(QD(N*u77fIWfLCNNOBTb-7O^tqC?wZOYxZqVMN}o>^YNK%SnLQ=m=7F8%^+k
zW>8-F=`}$WfTRCGVA$2r`Mw+U6@jnX6l-HpH8WiLy5_;5pZ7e0v<3e9kxX&c*rBp|
zKaLKGDVfiVOs1JOj^Cg={0bTwLUXpsAJv>TDv2D5iJ75kz&atjqCKOZh57$&S>T83
z9p)MQ^iw+^^7}4zv<!;pg$vaa4w--38k^@|K;Eze`ZA&Q`qsqtQ{SLOUy0xqeaJ}X
z{7PD!a#U1?OtBCjkjbD9J!gBO6%9dRn2~i)tV<MQ0y#pth0JUK{}kY{-;x}~^SxBe
zHq8|I5iX}I`T6<9Jq`w)Sh03(+~nXzHVfSQH3?ee=lm3C#b4t#^IoOoSXHDQ6&>xg
zSEI||wOP;Nz8q=8uc6>FWBMc-u&xZqIqX)o|LH3D$Eli~xNWa+a*21xweEcyw?@te
zz30*O)yWMS3Nr54?ncs>;b9obeIdE6L{N$8O!Dk_8&bb0*7L&rW<3e&wH+~&y633T
z`H0dbsf8C>_$jUgrOk7urZy{Iv+XQ5^c4iabDF)v0|OJRf}uY_&iv)IrP?&N2Zwuf
zMz@<Lnc6}Q@AoPI)McWwq;GQA$Y^aY&yUEorzX3*`F=NW+KONLRX5pmuc^$QW{2V+
zX@d6g^ab!$r<`LGIw%~NEy|y|lu42X;-DtwYfQYbzzOw(fOa$%SBa+(s*b){97$ZR
zXYY2<)7wH1x&S&x_7)F7)KZ*z(tL{!3({39WPWw(QloR@Y3G}I*45XR4RR*DVK>E>
zbZZpgnN*zq2`Ec648&3Wh|9dGC@z0=<Pz4jXy5zS6Emy%JNU1)6v(p~w^<M-Vt(GU
zCp7HX@#9-*@+3ZD+O*v^L#cr~kRhP;)mJvRB}7iKhZ?eVfMjp#srd0}_11d{yyR27
zOxCqY!@P#k9$b>MhPCV5xp!kL_UZm`UbTh+vvG26P_KlEshy*rxU+2>V1jusFGyN8
zC4c940}YizUR(c<t|ym}YB4Iw4c>2VkHf*iy_LThvK@YwV$hKoOqg#Uh)lvt#=~FK
zQLJv*dC1@Ydv@Z&6)Q|p5Q_(#1N@TxqD3zs+Qr)Tg4Z*3dpe^Z8^y4oK0n;L%Q07q
zNyQJ;dg;9i#<c7o?vcdZY5X~oYF8`*jiG5P>6h#gs7=EIC{Y|R1@xtXv>7*O%-3Y-
zyR1m)ZM>_Y;a2Iatn6$?5q27x?Ecw*&Ke3VN(OVuCYA#m#7M$UW2*|{Asu)CvcTBA
zP1e_>p20U_!2AXlVRJ?38o_1=)!EJ|O)$fA?uZmtxOi--iUg$qifN%xu*l^s^Nw+C
zOY;HTDRm-!{U{(S$yAbTMDFZsf}NZ>(^LBVhu0(^SSGnkF^d^X@@wVqz{=)eZWo3F
zca-W#@5NuFHy)iaO_HA0z9h4~kKJZ0zFj3{(bUfM&ZMq=AL07=gOJ>aIc^O$=iIc)
zqdUw!Hth<f53-z(g?p?;AXIekL+U92DFh{3p^+R^xqRFOlr5kLqmWwPxAUCm{>iAO
zs&pGUYn`bkMsxToB3~n>JPm7uh>1yl7TgRO_0=B^yKcljXY3@>TA91Vr;wp6EZr1*
z-BhVkDo9D5IAQmfO10nxx7yEZ-|F&nRmTsEw2xsqk)KpD$$V4#62pz0S>1*Vsheb8
z6jA<Y(LZy}KkI*{0R7pceh8nfR({GK<iVBb4$y!!eV5ferjB4Aw5AsQ_%=Fw{%<ul
ziJC9A&sY96m&N7%;C~^pUH*;48vYxJHS~=)Xlzdsug@^PcVM|@TkR;PA+@~|TWi-m
zYEm_CgZ?)!mk$2<{}{3I@82_h{DNCQ{O|C%qZhOn|9e=w&i}iANHh{ClFphoKk4=H
z9245ZCj1fd!TgE8p$ZBV5JODC%8Rm4_6G4!&oQfcctyAlOpTMEa0WDQqSHlDx9#o&
z@c@9cYI@Xvst3}pTyX{HI(LE$p#rMJ;`S9b0PZfC*~K8w5wIHQ_0-1?jB7BU5SN09
zo2p>Vbi0!@IY@@vBAZzsTKcoI`g_2mnv9a2KrtgnUNt#=xYDIvb#DIca1CP9wK(_J
zGdfOmu532L{&5)Vbh%I6@7&Cps#{%qy7Qi=-M{Oy#m8>ASP!&+q#u#Q#4o8+o}Al;
z)WIBc0tKLp_M(|HYdNmD373dOBBDV~czrx!8W2os*!Vq9*PplN$Irc8KzFatFLzCG
zYx49jx5~ImM@6a>1)8RH<6c5h(&1#J11uU?*sBA}z4u<dTrjU_U*&-3AIDi%F4~v$
zy7_<Fer|UNRvP_{a40LW5pV=M7G>`<YXVX^g6@wxLsG^PkL>+A(T!fT^7sMe*}b?I
z*q+iTaO~!Mu3YQ<OHDt;-oZgbAsp_L_b=+Qsk5tV^Y8AoGupA!(sD&}S+?JLF7h7D
z<>|E1Ly-|&9G?p@LoC5@V`H~r!yxI#IOJ2VUe)AiFMt1DUec;ntJaj%Pyqc$SGOhX
zA|NkOK<zm&(Tw)Jm?*s<hcgKu&0sRGg0z4b&?x2pQV*aFdS`O*;KBKhj^dJbg@t2~
zUy{uwa0@bSI?%|GfBZAT8Qz@<DQok>Zvnjuh(L|-TWm*=e@So##W+6m9Zf7tQat78
z7`=EbSCEqYgX<tC-9>|O{(K)q8b_hrBq)y;5I!@Tlo#+K0VjwZYDO9}!ET<Rq1_~g
z9e(Y%fDUcJ+plra!18CrzC$Q4IXR6LKfezIFx7ha=cb(3B9;Z&l)x_L!=Lcy1@(Z-
z6m4g<?K%J&B);TW<i%SDnx)<5A;<)9sY|3tB;{6Sf@1;X)A;S*fAT^v)74oBRSbIe
zoOtOLg&$8`GRk9XoRA7go2OWCng^`T>vKGr)=Z!2(o78Sg16wR%jKE4R#(x5N@}mj
zQK<0)td2#86XKVj19^|{azuoUB%0BF_bE<#Ll=$EvyGz5WL<uS+}jh=-)%EZviK5T
zrUAGo0}jg8P1jXfFk@Oa13JqTkPYHKil4-rH<F1_-odrF{^Xge3&)FhO`bKDs^?UL
z+`0lH*%r|Unz`4lhf;41QtF8onINH!@Os5>w|t~NMl*ouqXi3%*OB5;U(rQ|7xIct
zhvX$x#m6(?i1{5_tT@rF0piW3MG3CGp*LuG`$dNb#BI-xj*Cm>s@AA`Qr%ofcghk~
zCXH%al1LyLjbD6>FH73S{Wtpb<LDP3DdKbRj6r5z&B)LeHk__xP04`DYRiJoo~gbC
zsf}~(Mvao7H$032+p!=d_Bx|*U9$e|ojQ7ZbV|nt)vD>Qn4A=}Dmy+tUIj9vNIu6A
z7!`@W1D(UeW6E0``>L+>(Bo!quSe%)?lary<D+fi<CDIG<<UV=-x^#tZp;|(D~B`F
z(<QfNE0q`1GiHMjbcaTg2x$!iO5fYY$qTP*i&;~o2|NTNI&OcCtgVL*9pbrI(F`V~
zVH`{V&zlBIY)_pZMa{i`s7qWz#jU}b7DT0;o>@<ivvAq6UThY1?$|CjXcpM~&<U&u
zQ<}Xg8FX56m2vcN4kdR__T|N`WHtrL<r|<*{;QJf)Z36%{VKDqsd?3KNEu-9T{%74
zDSfFTj=iPq`jY<zkZ+9DZZF+d2FXJ>9Z7ehavv%l-naP|tZBVXSx*E46++caFVVvj
zr})ukr1Q#5z+|I{qG8Gtj!_PDD_W&#?%YP27LwWp(iye>{X&V?;xZY3X;o1n2f5Wl
z4t#CJ5lDr3H0dlrBQvw9$QZhEfGk|Tyf=l_H*I$jScjGP=RZPjty@jldt^+EFd#Q>
z-VFGk;=b9Sy*}yGb-fg^P>OStQhpB`YAgPJ#wbrv(&fl~u-JF#OBj>u*RG910^A*!
zo_HxGMw$6s^BUvk0;1U4{+y{MZl(kQabI>bO#M*J2i;1(GN++fb^-Gy%5Nc+TEd$j
zG6^gAmB77Qwr>w$iEaf-#!hS{7;P6K#FW2qFD}Xs?#0-9KOF34Ozpwfohh8b`p<F@
zRVHJ+ke{YS_3;_UBBCuBj|)Sx*|_Vw(**5i$!Rx+B-zaMGHVV-v?VYw5WxNQz0-kN
zLH#GdS9K+=O#Flh1bXF=?(L<e(6a26*Ti>{$IFc?{$SRC5j>HlC6(`WWL#v$F<it#
z6061T%-bPTYqvzj#16zqtc+s@lz2+j62tLg63)Eu`}d#OWAN_}`ZQWPziylB(#65S
zVGG!e;wP?go`b}%O+zf>2(2OIS<7PyW-k^k_~59@Kaq{|)c!A@wzh6%tT(T|aip(L
zmc%-EHU?!Fvd{yx+I9B{4Grb;iby)5m*(c?c0Dt50(FHk2!KB#SxL@Q{3JE)bg6fJ
z@Oo$G8x=eIa4=+72POcqPm-lChGR~!bUx7c#>Q53w6JcS2HRNLxhCsB+Nsvx<NKN;
zB8uYmHq&e#Cc0cec{)>egi=Upu$Q%xLBmqSv1R2UA%pin=@0I#|0qy>*)DZQTT4s(
zA&oVP!T%6%;xp$t%i}f)3Co6tY32##OY~1W`#Vj01B(8I(u-IVr;?#Cf#B9ww6W@C
zl6-Rv10stmG{C+3-CVI}&YCqVqLSuS;<>*Yymq3A8^N!d7J&W6hK5x4vpUb{-{zB{
zA;j9(w08*QtZD|AC3GWc4s}dS$;8Q%+f(!~arvU!qOpUFIqT)!eVp1gWDOm*rUi>5
zEX=X9TE4Jg;l^D$b~IM(Ss{joYW(`da_p9Pv!-dVBfS)oG>xomrDM!-wrJmWi_9JS
znniACDE&7Cp$~mC{mIgR?L@yRYRCuUyceG~uTkyzzzx!4?&TQHfE$t@kwMWm^U~Su
zc{^X}t)AF+7s-HAlynfSn2cc?D#r!&b2L=2_SlUtyB~2t@dGFL(J{4Z-OW#up>xL=
z8hWlHWe_;V3$Ir{8cyS`iHwT6Z8d?f%%foT4LNB?1F(gkYkIXJ^tG6U?^>N>tb^jm
z>5xH}8S^NDkTt%OmF<G<8bQM-Y9a#aaSyjroZ;JAt$D}YLY&^r)5o298ybR=Yg*`y
zrIImH!ub7zN)MQ+ylI`RC_Z=}pmg6TY9b$h+fM9^D%BSjmy*Efrl0Z*vL81*nE|Jh
zIAeUkkSG6f^TEpKJ~c6s%d#>#5S(hBX71qUkiYEV1kV5yjLG=Ty%inev9O_ocX;rL
zY>V1__p^Z#2E{zcNmuQnc2#m7ywA^3LSEL24bW!8^5thCTBpBpt+-sUy>x6%>i4SJ
zT`!Yd-G6_8y-a-w+V|J-0P>8i>M-^6y^Ix`BGdZCnZAy<<KqGx_W70A>=_DGk)Vf&
zt_1@|y)29(GB+>O3DE8S4p*~l_@Y8<akNNBY3o5fu6WW*Dk&6kL=F6xO>SspH2U<J
zJ$u@*-lb3m0(j%Ktdx2Xz+p3@eP#XSn0Kf|q^5lQ`0*d5kNQhylJW<ssLZHrN*6Ph
ziY&HdKTtV<{U#16erC=<MK0f^CwIrk53YLmWSa=11`Zr34S@{N11K9irz%vG`G^m@
znwYf0d44-X8U#>`(14e(Uu!8SCgPt~iv6g9X#5{0PKv#dP?~e3i87zkO;IR^WG)WW
zfH%QrmQ7*9q-%k((@+4tk4}C<2QJE!*xT$L#F1-DJ1Qw8@p7+@9W$mm5|_}KYoWm;
z$&DCNr;^_3kzxjf&T55^SdxTeGbUp@Do}k>xK9xZk-lBrF}4mjv$Fh)d|;TJpGW$Z
zmeFo0!M&%tK*M*Uj`cyTq3@Oyv{r(4(SnFFCq@m%*Q{#vWM((D>nSq=C=PM)h`+hn
zsd+>@Xad69mR6OQTQD>$O6TJFR8RPVkr6iC2yHbPN<JQFX-2GZ@ukh)-oG7c6_Rh=
zuOLsb>7XELYw8r}tIcHkDnDj^+T2?Ic(l&6J89}S*4ee-f*KZjE&3yx!x3(MNS*3?
zz}c*-_NTb*Ue|MKhtRG`VwtGi6`7_iF>%5_<P0Cf7~7*zN1yI5D=ynItXZdI_G+|L
zFZUzS03>~``URNRxM0vz%i*^f`kY&H;pMt0es=w<L+cs?B#%(@h!H+2At@xSUySK%
zRhI4uOUq{0=w_#C%$Yq~NE2k(z=?l>puA_@{tYS9$iCh`vxV)9n5ebl$BO0N=?W={
zn~)5$`!#o(#`bGtOwOEA-mkpMbq+vjT|vX=E}r2d;_H@F8OAU)nWaG;y};3N3v95a
zr}~f3hk0pSRN*@~4Gjzo6zZl=dzZnvL8!<W3qWNJPqjhSkek=8P<_PQ&I~XKpMIYj
zS6S0p6IP{jSa6#^HKuf)5}deJq`TZaQ@KaEC~&%Ja>q+quN1o-sHkW&D5ao{S$3QQ
z*}%*}mH4%Q7+GGNPNTu#D&KXR<ELsA%a$K*s2dypJ8KRRtzX$n-=T|aZ94*YN?tLe
zBMzyZ-IB9^ic6+$kP%VTf@8Rs`Ve^ocT#=Ke6Q&ekN$g;YLC;YDfLNlR3Y~E_R=|g
z0TsM#f_WCK!D9!Pj>vxE=PoIuGJ*%?6?||D&M_}4*pj#H_RSkiY}x{qD%5wb_wBb%
zC(54Y9L%fG>11B9@NM-hzX9!=k4lT=mX{ZY)n&77gd}nEm;<uk=y$^+=(O=-jdTLv
zn>N!+uDAo*Ea07FI+5}%L+jXH#ZAN26_;ywrB);#asKQ+>_BytYUdHR@m*A@Ee#Eg
zb(5G7Tx_%-pp348c{U$G0(LPkLI^M6?I^s&*G3y+tUo)Dv0VVV+0nybFqA17jIl~v
zLS}7nQU8XS!%Zcyi;Mm_wq@}@^ZK>=$uX?eaj4702@^71vuW>wRt2esSyw!JUzgcN
zi0v~Kqn(}aHT0ZhAR{>-s%DoFjK*=K)3IMm%9`F7oP<B2R}lty!yB8^_nOB_z#LsA
z6EjS>-=^SVau}0|F-_;xH9wW(*&|7-n+M<Od*?!KZRU5;Aq6Y!KIlFwNaXbW`!B7m
z1LkcwQm)YDn?ae2n^s=4y?vPanWOL3L5J&fZ^vELZKA!(THocJ_tO@@d7|Q@1`?^S
z2+b4;Z^LL6ifX3$g>GyXj>YC4X+Z<t#D7(@LzQ)Le+O0aa8>^7**c%=p#nwp!I?du
zZy=p8)(tcQYZfil|K)<&=SpYU@7TLGi8G0R`-`STbh4xg$*~DZU`A4*>*BD)E$(5v
z^2)-xCLeA#^!Sm2b#^+Px-sr~zKAM~?Ij`*C7Fz<1&om-E+r&uhn2N;cEy6%bAr^Y
zya6W?hH*v6^lOAi?}6B_JbXCEWH^P}n*y3zC8Q#5o?CFQrHdB_E*?m<xsty6M%c>{
zqo$udPjOr1JaLi*G4Ra+xg@Kw{NAj2jsb2D_uF;s;##gDkS=RwAt4d76E_Hx7}E1(
zZ=Il&mv9#!UM^h}+Wz>}8uwPW3!XfF+y^co>)^n4<dI0qrjR+*F-_*one&CPKeGPL
z=MQt(VY;iWlS@|50zv_mfveg2_U-TD2xKgB3TK#@AyCa+q2xT>sb*2cB71u-psW+h
zx$ft~v}%UGp_I367_jzn*0#uk!)D7|pX}&7bSqkPf?`D4Z#yBWmCAeXk>nq05;ETB
za|Qu@FbkQfiXGi+UjHD<B&h><;gX?osHm*tic?*_^)_EQF{Qk*L-wBF2|)KObmy><
zbSZYaQ^>24%^DRI<+o?gCJv-fj_35Oth@2UA##q}oVSXu>qS=YiqD@)Ynv%3AjVCY
zFp4sLP3K|Al|m|>9e<elnyz4w(_Dt8(>!Jjn`vqskIo1D;W=8F4EL!)Mt0eEdiLKi
zbHd|(nOcXB&Dl@{2=J`zSTs0gZjX3GM<X4_^2vq06)BaNJ888x!uuEZ-07khRP?ZB
zQ2?yh(~x1RGekIx0*2Xa;{HFjXGL*!ZPe!{>@>PW(km>Wdjf`D_YIimK~B63exUi&
zo{?;(t<nKP@IG)rUzD<N_Yp(dUV%+KPYR84uKMxKn>S_po0u(tuF})fEg096)ac)3
zOqy?J)5c6S2^yX(8mXB2WpDr4KaW0Mc1-oB4lHS!psn)lS6N*+dYMbW3v)$eKtlCX
zcge763c5P43AqI`h8ydw_<LWrOYXlezdB|YF$r%tU`xolIB%c%d<xLKVN9;^Z_)kc
z@y8rocRpo7@`?*V7M3E6ortfvlCQ%P#%2Hj9xkV5V7bhDFqz{I=c>WLfqLwFs~Xn!
zd&ET{gtsg8!7HZ$DdOdP##smqeZgjK0}%mczuo=OA^R|Gp{63xM(pNXnEtjMJ8tFr
z6Kfd+iQs7XJVf~xF=)l(i~kif9)&qbVo|H?V<9$Yf8ByMp{T%cBlcYizy$*p!8ASJ
zjd@RHca<7Pqxf)uN(VTc!}JBcluYD8Rhf(UH1Je~Gq|O&hj5bQF*!vf&|ZF15*f_&
z3Ui;9NdM0T0DiWUe9M#qdc>%}yy(2U`|9hNv%zKw<?{@RULr1t)vM=SocAEQH2v1C
z`=>(DLkbv1hZxZMePf~?HN+7zd<L1`$eC4L@oLE#4ph(ZykX|%9YBolSrwqm8&+37
zs>sgt%<8<wSW%hmmFMi(s=P9|X?@b7pujO?Gt9ejC#N!UaqXV2yeq{*Ywd3sF`6JU
zNwf)yeDN}RnduTi)Xx24rz+%%6doBUKd|Htov`kmxWG=ywIANd{O&_88abuWHAxK{
zt3qF;g<b8~!uS2DohQpSFi%mjUc63{60`NnEG@&fHEv}&c?DP-@cRQ#D$5YQxI`DA
z(vyH$02CSjEukvHp}M|{&K@?3DemqcpSO67!XND7D(bpOr#CilA75%$mRQ{xia=w?
zp=C@X-cAp3g~7uFI#>vBj`3Ge1<2#;gh+q9y0*QEjSeWaXc8po6w`}XJZDGQYw-{y
zypi|?<cl)yc@)}?w)`&v*Z@+#((N-}!62|%cw%Tj8Na9=a1FLubea&`WBKc%F{3p#
zsqKV&W7_^FafsPu>^yyoE+aY!!T*C<W;nhTbRC@GPQkVJ6MLeVB)szoCyJ;WkjnZZ
zemF|eNOFs2dFAufWa91X>!8ij0ODOqguXNlOYff<!5tKdC=4>yuuq@!Da*7-apYWE
zNqoEH?k%=S&+Kb%o(9hM9<@UVDjtz>0llB+taG^UpQN|TbSZCLCcFT9Y?ZnWGJBl0
zbs8P1q!*EnF=J_2_nTRn7=DgWCy+8T3MxbAT^VH}QgpAVA7(%xqHq6$c3DQ;$r*?0
zSiS55B^k=(9c+lJnVGF9^)_OGMkyZ}<BZ-FK|%^`ACFihk%t_-o$s~AquARE%_+s3
z(^I=chZ`66iti|cegY8lrN5R^3fxY7Ut-nb{Ai+3#b_5j{s6Mknj>p7UB7rzGjx(*
z4P=TcI?Id$Ir*_mH**udU?Esg$K{ZpBfSvH0J%*8I5z>;>N(YM{!7yDY^KJ7q=>IT
zmK`aVg3`-uJfrj}ek&^+YH6!opP-3r1#Cy{(S(bw{KJRqAP<oU?L@6lKO;;P5)e=N
zSCbI~N9k3#s1s)OcVCCFP-3S<$pADA_beSGIPh4OI`W;<Z~q&Z_u0<RJ#589W8al2
zx?cUO3P%@w=)1^Q2~N6RM}38C`Om&f_xbp!yV6w~a99d>HjXRlIz<O}w4}~~Hwfa>
z*QM(B@z&OPpB~dW7D3nrvXr->pn<%wg1B<v3kw!=NX<tGWKl<6xX=NJgC_^bV@Qqk
z(}_bUCM7u!4ES4EyOXxDr?_>-0|!ta^eAP4hpG2V_&OV_Cq{KaSP3LmbGT>nTVqSy
z>pn&bj!1jkRb=-BqDc}OhOgq77qruoSl9^Ppmn31lwlO%zi-%ZS@A=%SA3Tt^<6X!
zaPH<>Sx=n!jUw|ljOcn6Xj<^bjX3#c-nv6pBznFdScCUZsnA)IOlX2dLw*;Jnt^Nr
ze-aGOD1q_7@`Y45U+e3eDBlkm@@x3^poE>nYQ~jS=U!U9Si=DDEANJ2J<8u@W#^_J
zK9ZJk<Hj!T5GzW-fcm~%Iw<n}A;(%u5KnG?+{Vt+Eely=<Ye^b%8`+F0C57R0~p^q
zDf#2Knb0f{Gq9VC?(iR?sv!e)+cDPO--U(Wqo_ctY(m?=n`e_yUe^z>PEbi&lVhlD
zR#m;7W3&R9<!zMgJ*FQ0Q@}m&?l6yr`c?1$9=)lgeOtk@A3S*ALt-LIDrWn!w!0~n
zJ6Z=NpnZ*~4D8gz(h@01FHXcGD2lh)MzrS`e@*lrq30vQ&m4KhhwrU?4?rFM)ugAi
zDgH4$-W&7)VU$|aFefu6br)Ye9gl)B-`)1Ghq`b}9nzuz+OsVL`S)GBopeP?5U@6v
zbK8oOAsmG$61Z3(Y}Ovi(ow=~NUHwpveX?1yzYf>cU0N6H?`}j*mF*RnfdlY854*&
z&YHgedTQ#Jj^@jIqC)sj+g*J65iD={kCgPaS6BYJXBKIO5o<<ySaZ@Aej2N*N-cX;
zo4E~QW9*<c#RiBSzGYFfHm5Ov^Ajq{LU18lhYwH6QmaBcY#;Ak6`Jra6P`j_(bCbm
zJ20CU)=Igw;lnAL%SY0Brmyf$dgG9#OT1iv4Fj6?oPpiR77hQ}c~nCWM#vO@+YwtM
z`p!tg;YPaUre+EdL3j)?H7j3c>3w5TwPD@gf^peK3@DkD@{jN`t<g|sw+FcfD%WOc
zxE`9csz=fC70^_k^De2wb_ECT5Keygi$>i-ZvHkGQ5htT<ng3Ynl#=DB&h>rnN$XP
z3^~nfkUK-NoH>7wLUKx)4=Zf}mKgw-q$t%F30eLY@v9x3RZNngE;qF}CYUQq9%pSp
z@6V-JetuuJOl&*Y^ugMi@$|Ttu5!%;Y`^!9-eXhahk*m)pKljD{=e!bRR0!H{QvWZ
bT95Dcu5OxQcl5HtKjTMFwmds>{>J|W@-@wt

literal 0
HcmV?d00001

diff --git a/mkdocs.yml b/mkdocs.yml
index 146a766a2..b97a2e41e 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -120,6 +120,7 @@ nav:
       - Chain of Thought (CoT): cookbook/chain_of_thought.md
       - ReAct Agent: cookbook/react_agent.md
       - Vision-Language Models: cookbook/atomic_caption.md
+      - Earnings reports to CSV: cookbook/earnings-reports.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From 906e84e061e6f4027522f57efa1c825f36a53209 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Fri, 8 Nov 2024 09:15:26 -0800
Subject: [PATCH 101/505] Add PDF cookbook (#1256)

Adds a cookbook on extracting structured output from PDFs.

I included some extra bells and whistles here by showing how to do JSON,
regex, and `choice`, which should help provide inspiration to people
working with PDFs.
---
 docs/cookbook/index.md     |   1 +
 docs/cookbook/read-pdfs.md | 376 +++++++++++++++++++++++++++++++++++++
 mkdocs.yml                 |   1 +
 3 files changed, 378 insertions(+)
 create mode 100644 docs/cookbook/read-pdfs.md

diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index b163feb62..97bf06687 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -13,3 +13,4 @@ This part of the documentation provides a few cookbooks that you can browse to g
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
 - [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
+- [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
diff --git a/docs/cookbook/read-pdfs.md b/docs/cookbook/read-pdfs.md
new file mode 100644
index 000000000..dbe4ccb02
--- /dev/null
+++ b/docs/cookbook/read-pdfs.md
@@ -0,0 +1,376 @@
+# PDF to structured output with vision language models
+
+A common task with language models is to ask language models questions about a PDF file.
+
+Typically, the output is unstructured text, i.e. "talking" to your PDF.
+
+In some cases, you may wish to extract structured information from the PDF, like tables, lists, citations, etc.
+
+PDFs are difficult to machine read. However, you can simply convert the PDF to images, and then use a vision language model to extract structured information from the images.
+
+This cookbook demonstrates how to
+
+1. Convert a PDF to a list of images
+2. Use a vision language model to extract structured information from the images
+
+## Dependencies
+
+You'll need to install these dependencies:
+
+```bash
+pip install outlines pillow transformers torch==2.4.0 pdf2image
+
+# Optional, but makes the output look nicer
+pip install rich
+```
+
+## Import the necessary libraries
+
+```python
+from PIL import Image
+import outlines
+import torch
+from transformers import AutoProcessor
+from pydantic import BaseModel
+from typing import List, Optional
+from pdf2image import convert_from_path
+import os
+from rich import print
+import requests
+```
+
+## Choose a model
+
+We've tested this example with [Pixtral 12b](https://huggingface.co/mistral-community/pixtral-12b) and [Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct).
+
+To use Pixtral:
+
+```python
+from transformers import LlavaForConditionalGeneration
+model_name="mistral-community/pixtral-12b"
+model_class=LlavaForConditionalGeneration
+```
+
+To use Qwen-2-VL:
+
+```python
+from transformers import Qwen2VLForConditionalGeneration
+model_name = "Qwen/Qwen2-VL-7B-Instruct"
+model_class = Qwen2VLForConditionalGeneration
+```
+
+You can load your model into memory with:
+
+```python
+# This loads the model into memory. On your first run,
+# it will have to download the model, which might take a while.
+model = outlines.models.transformers_vision(
+    model_name,
+    model_class=model_class,
+    model_kwargs={
+        "device_map": "auto",
+        "torch_dtype": torch.bfloat16,
+    },
+    processor_kwargs={
+        "device": "auto",
+    },
+)
+```
+
+## Convert the PDF to images
+
+We'll use the `pdf2image` library to convert each page of the PDF to an image.
+
+`convert_pdf_to_images` is a convenience function that converts each page of the PDF to an image, and optionally saves the images to disk when `output_dir` is provided.
+
+Note: the `dpi` argument is important. It controls the resolution of the images. High DPI images are higher quality and may yield better results,
+but they are also larger, slower to process, and require more memory.
+
+```python
+from pdf2image import convert_from_path
+from PIL import Image
+import os
+from typing import List, Optional
+
+def convert_pdf_to_images(
+    pdf_path: str,
+    output_dir: Optional[str] = None,
+    dpi: int = 120,
+    fmt: str = 'PNG'
+) -> List[Image.Image]:
+    """
+    Convert a PDF file to a list of PIL Image objects.
+
+    Args:
+        pdf_path: Path to the PDF file
+        output_dir: Optional directory to save the images
+        dpi: Resolution for the conversion. High DPI is high quality, but also slow and memory intensive.
+        fmt: Output format (PNG recommended for quality)
+
+    Returns:
+        List of PIL Image objects
+    """
+    # Convert PDF to list of images
+    images = convert_from_path(
+        pdf_path,
+        dpi=dpi,
+        fmt=fmt
+    )
+
+    # Optionally save images
+    if output_dir:
+        os.makedirs(output_dir, exist_ok=True)
+        for i, image in enumerate(images):
+            image.save(os.path.join(output_dir, f'page_{i+1}.{fmt.lower()}'))
+
+    return images
+```
+
+We're going to use the [Louf & Willard paper](https://arxiv.org/pdf/2307.09702) that described the method that Outlines uses for structured generation.
+
+To download the PDF, run:
+
+```python
+# Download the PDF file
+pdf_url = "https://arxiv.org/pdf/2307.09702"
+response = requests.get(pdf_url)
+
+# Save the PDF locally
+with open("louf-willard.pdf", "wb") as f:
+    f.write(response.content)
+```
+
+Now, we can convert the PDF to a list of images:
+
+```python
+# Load the pdf
+images = convert_pdf_to_images(
+    "louf-willard.pdf",
+    dpi=120,
+    output_dir="output_images"
+)
+```
+
+## Extract structured information from the images
+
+The structured output you can extract is exactly the same as everywhere else in Outlines -- you can use regular expressions, JSON schemas, selecting from a list of options, etc.
+
+### Extracting data into JSON
+
+Suppose you wished to go through each page of the PDF, and extract the page description, key takeaways, and page number.
+
+You can do this by defining a JSON schema, and then using `outlines.generate.json` to extract the data.
+
+First, define the structure you want to extract:
+
+```python
+class PageSummary(BaseModel):
+    description: str
+    key_takeaways: List[str]
+    page_number: int
+```
+
+Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers `AutoProcessor` to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
+
+Images are denoted with `type: "image"`, and text is denoted with `type: "text"`.
+
+```python
+messages = [
+    {
+        "role": "user",
+        "content": [
+            # The text you're passing to the model --
+            # this is where you do your standard prompting.
+            {"type": "text", "text": f"""
+                Describe the page in a way that is easy for a PhD student to understand.
+
+                Return the information in the following JSON schema:
+                {PageSummary.model_json_schema()}
+
+                Here is the page:
+                """
+            },
+
+            # Don't need to pass in an image, since we do this
+            # when we call the generator function down below.
+            {"type": "image", "image": ""},
+        ],
+    }
+]
+
+# Convert the messages to the final prompt
+processor = AutoProcessor.from_pretrained(model_name)
+instruction = processor.apply_chat_template(
+    messages, tokenize=False, add_generation_prompt=True
+)
+```
+
+Now we iterate through each image, and extract the structured information:
+
+```python
+# Page summarizer function
+page_summary_generator = outlines.generate.json(model, PageSummary)
+
+for image in images:
+    result = page_summary_generator(instruction, [image])
+    print(result)
+```
+
+### Regular expressions to extract the arxiv paper identifier
+
+The [arXiv paper identifier](https://info.arxiv.org/help/arxiv_identifier.html) is a unique identifier for each paper. These identifiers have the format `arXiv:YYMM.NNNNN` (five end digits) or `arXiv:YYMM.NNNN` (four end digits). arXiv identifiers are typically watermarked on papers uploaded to arXiv.
+
+arXiv identifiers are optionally followed by a version number, i.e. `arXiv:YYMM.NNNNNvX`.
+
+We can use a regular expression to define this patter:
+
+```python
+paper_regex = r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?'
+```
+
+We can build an extractor function from the regex:
+
+```python
+id_extractor = outlines.generate.regex(model, paper_regex)
+```
+
+Now, we can extract the arxiv paper identifier from the first image:
+
+```python
+arxiv_instruction = processor.apply_chat_template(
+    [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": f"""
+                Extract the arxiv paper identifier from the page.
+
+                Here is the page:
+                """},
+                {"type": "image", "image": ""},
+            ],
+        }
+    ],
+    tokenize=False,
+    add_generation_prompt=True
+)
+
+# Extract the arxiv paper identifier
+paper_id = id_extractor(arxiv_instruction, [images[0]])
+```
+
+As of the time of this writing, the arxiv paper identifier is
+
+```
+arXiv:2307.09702v4
+```
+
+Your version number may be different, but the part before `vX` should match.
+
+### Categorize the paper into one of several categories
+
+`outlines.generate.choice` allows the model to select one of several options. Suppose we wanted to categorize the paper into being about "language models", "economics", "structured generation", or "other".
+
+Let's define a few categories we might be interested in:
+
+```python
+categories = [
+    "llms",
+    "cell biology",
+    "other"
+]
+```
+
+Now we can construct the prompt:
+
+```python
+categorization_instruction = processor.apply_chat_template(
+    [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": f"""
+                Please choose one of the following categories
+                that best describes the paper.
+
+                {categories}
+
+                Here is the paper:
+                """},
+
+                {"type": "image", "image": ""},
+            ],
+        }
+    ],
+    tokenize=False,
+    add_generation_prompt=True
+)
+```
+
+Now we can show the model the first page and extract the category:
+
+```python
+# Build the choice extractor
+categorizer = outlines.generate.choice(
+    model,
+    categories
+)
+
+# Categorize the paper
+category = categorizer(categorization_instruction, [images[0]])
+print(category)
+```
+
+Which should return:
+
+```
+llms
+```
+
+## Additional notes
+
+You can provide multiple images to the model by
+
+1. Adding additional image messages
+2. Providing a list of images to the `generate` function
+
+For example, to have two images, you can do:
+
+```python
+two_image_prompt = processor.apply_chat_template(
+    [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "are both of these images of hot dogs?"},
+
+                # Tell the model there are two images
+                {"type": "image", "image": ""},
+                {"type": "image", "image": ""},
+            ],
+        }
+    ],
+    tokenize=False,
+    add_generation_prompt=True
+)
+
+# Pass two images to the model
+generator = outlines.generate.choice(
+    model,
+    ["hot dog", "not hot dog"]
+)
+
+result = generator(
+    two_image_prompt,
+
+    # Pass two images to the model
+    [images[0], images[1]]
+)
+print(result)
+```
+
+Using the first to pages of the paper (they are not images of hot dogs), we should get
+
+```
+not hot dog
+```
diff --git a/mkdocs.yml b/mkdocs.yml
index b97a2e41e..7272e4a63 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -120,6 +120,7 @@ nav:
       - Chain of Thought (CoT): cookbook/chain_of_thought.md
       - ReAct Agent: cookbook/react_agent.md
       - Vision-Language Models: cookbook/atomic_caption.md
+      - Structured Generation from PDFs: cookbook/read-pdfs.md
       - Earnings reports to CSV: cookbook/earnings-reports.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md

From b93f5503a11fcef07ae8c62eb0d403358fd6ad8e Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Fri, 8 Nov 2024 09:16:06 -0800
Subject: [PATCH 102/505] Add earnings reports to cookbook index (#1255)

Forgot to add the earnings report cookbook to the cookbook index
(#1235), this fixes it.
---
 docs/cookbook/index.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index 97bf06687..6d279bff5 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -12,5 +12,6 @@ This part of the documentation provides a few cookbooks that you can browse to g
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
+- [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
 - [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
 - [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.

From b5d26c4f7cc532cb725fc96b529b61e1b8811a46 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Fri, 8 Nov 2024 12:37:17 -0800
Subject: [PATCH 103/505] Add a receipt processing cookbook (#1249)

I added a receipt processing cookbook.

- Uses Qwen or Pixtral
- General purpose message templating, no messy model-specific token
adding
- Easy function for compressing images down for lower processing/memory
requirements

Should help illustrate a simple use case for vision models.
---
 docs/cookbook/images/trader-joes-receipt.jpg | Bin 0 -> 430638 bytes
 docs/cookbook/index.md                       |   1 +
 docs/cookbook/receipt-digitization.md        | 296 +++++++++++++++++++
 mkdocs.yml                                   |   1 +
 4 files changed, 298 insertions(+)
 create mode 100644 docs/cookbook/images/trader-joes-receipt.jpg
 create mode 100644 docs/cookbook/receipt-digitization.md

diff --git a/docs/cookbook/images/trader-joes-receipt.jpg b/docs/cookbook/images/trader-joes-receipt.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..6742d39462449d7264cbc2724668505f6ceef7f9
GIT binary patch
literal 430638
zcmeFZcUTlj*Ed`}ff)vL1{@I-p$AX~6eSBtXa<m=!Vpvps}cnaL={k^NsF>52$&FI
zM6x-py1I%HOn_ictAc=;EFj2tdca-p`?{a!`rh~7_dZ-beY(2p)H%O8bt-f>Lyw1E
zA)d=D=UE8DFysV(Xz2Cir4G@-fe1M}>moKnNDgr^8e##L1qTm!Dm51lMU0LxI8(=X
zj43@A0zYAduK->+l7|Mo0hApD=M!)k0?&tYB^+aD|HuhWWk}`RhxAFChxv3mLKoq1
z3=LheDgYsIaCE>b+?-NZn)k=q(Rr>IH=JgmKg~!VYS1?{wlFZT&^N^mOf3v%SeO_g
zU6h+RtY<wav~Pre3cSYVVI3$vwMY29z@tnSV!%n?(8!RUIjr*=$ix`odw}NwPvOG>
z9Ti4+Mc@@j`0>DF+lS>~l&x7DsoolpPRknBM;CbdNV*~Lj1k@fc;-kR5%4l2ybJKG
ztbcTt9pQPD9wT`?AWvo%O{%|<u_@aA6QA`Hk0c{D8tCg|l7HHQCK<*b$%Aeqd>a>`
zafScXiydG|ZISvx(NASZcu)t?f8rTM!}czs>>5c|hCDJq<zfHC%l^c3e&Xdvc&c8b
zqG9}fL8dYA0!Rx1e)&iqV5OvDix45=qB6MhhvR1tF|sf;vd}m9k7;`PW)}M3QU88+
zgSMyS>>QDIeB=x#Bm>8f^pn81f)ZWhhtJe~O7DO3XWZAvZzb;MvpOKc2REH&zIrtt
zINe~HsTm+BEkXGqjq*1{gFl4k!l8hM@%(+nNH=Ltgk;foXm_d>>Oj~?OTq;<!oP=P
z8uU)8pgxGEkMvYZbD5v`u_L@x2imY4Q^-#n;ng*u01#XHH}rE%p(=+Ohn^uj#Fmj^
z$uQY07E4x^&5`Gil9!W{S5X|p<xd>1t~PPJsw%EEO&iyks;R0foS`#S-@w?|SY6xP
z%FNJmnvt;~B?Oa|m6ey1R~|J=*-)S=F#JEJp-QB{M(;5&Rs+!#FuDRZbPcJ4o=k{<
zR1h2?=%x&&42vzxk%J7U0db5@r!nYECId{z;;3%SP+%$w3~Xh_xcjg)!uf^?+xN3I
zMaM6X^|;rjW#k)?D9ceAr>ru5(qwI+j;^tZ=?pV-3p;y<S&mN5E}mZV<}dJGxX8~x
zAaF%ca7bj-+I8!rV`7t%Q+`d|yd^CwJ7>qvUAy<>79J=nE-5W5uQ+k?)al>PoIUr)
zm8;jT->9m-d8@wR{)2~)9zS{d`b}fg+vb+mckkOfI=i}izV`O@Q+82y{-9yQ{>?50
zu#3iE&>1YsE{wJw4!Qz^DKL;xv~_3sgpbiMOkndx+xH*8EURhc(KgmM;vPpy%lP%A
zcFHuVW&i&SOZ>mG>~F*VvFj<4r(@7~bOmICZUm3Exu0=uo=Q=cW7STZWo_kcj`MBy
z>F4Da4UURtognbD8@H!aeRH%qY++;K%2xMGi`3El-1z>1(v`R&B>w;lnsfC~Wt@(~
zm2D?E6`FqcYzK9<XDvOTZkIcC%Fdz-o9W-KxSd{a!+qi3Hc(TP_hQDfuk$ZFb-HZV
z>Gyt0{l$q@C$F;1nqSw%W50NNABmK=+nDN}#ti(RrqOU}#@0~(wJ$9yl;{m+@AAXD
zjK6&PWH#x1%;DcMZ%h?TI(|RQBCYFVO!L->6^AUg?z_J0neSP<RYrR(u%x7i`DfPE
zFODlTZwz8tFJHc(v!n<67$ceH8p00Mn>OdyDZBG#8Gc(b@zNrV`-fJ9tIh~>YBBQ)
zbFNzS+cJ;e_&er=^ha9UICE@V;NuEa|1+fzkBs+~7l#xrI-UN=bFpJsNsvr{>n69O
z#iq{|HiY``&^zk0qwtrSvL|?KgF@t*xy+D#zn?By>T-PXO9!oOvhzG|(vRq*E@{|Q
zJMrpL`|Z^(``0}?o0N3<YsF&OT}1P_jLov=uLjMEIKz(aS8qSE?P;@IQ767<(wxQ!
z(^)IWRAFwv<{ht8++W+c&}-?3X$99ZW+>R4Emk<z!usBS(Dsd7fxh2+9iuUwdEZT%
zo_F1P%Uxpg!O3$lyDxXIzM7%Umx^V(dKYw;uG9FvZZsz|&M<g%&9#;~uM-XeO}g3(
z9i@k_hY-n)yF!lNJK@}bgG-ss1sh$Zdcs*ZV&?LOfin}G`h7{h<M&4$#imcxqsPxJ
z>Sk=XpCFD`?;9waxaI5S`=o4W@q)mYG0e!ntU|u-bd0_(NZ<bbcsxVqn6YrrlQ9Z6
zbZKi|{5nqV)AbN`<p)i5$@R(i2EUd@Y3gKsx~i_<XesYLHs{6S8118D%XhNUziSVn
zqfY)C*YZD4QK^4+cJn|+Z*EsSnp`VaRyFW@{JSO0t4@Ba>z6OsA3f%~L-G(ha&+Ls
zQC9oOg0XF*+Kd0}uG!15h}(WP@T{}e4H;Tq#!-BeO3}lQ&3``VTAQMs;+Xfv(r(aQ
zCn0}k?z_1&*3<qpB42-&{IQJb9rUngTlNx{-%n;G>!uHUj@$cau%f^r{<!J6!e!UT
z+wo#viw#RAt7cU8d_7XNexlA<Mg(&*E4Me|)AXZJ+~ZG;Jr+MX5n{ONDJQv-^+-GS
zLEF}v{NH<&t`$9+u=wUPnTIK}K7L-)H}+{oMVHuyW17;6`aS$C`)i(9KVvo$^@}c^
zG|u-H&6(qs5?_kGn>=1QW5&bS4Q_s`rumEHW^wN<u>3T!W{REhy5jIWf74wScXwWj
zHvW2M+{Hyzs!4vgHp|#{)oK(TOI{XT@v!5z<6gJKdRfJ$3$`<|5C?t8SoC?OhhJ`?
z>+CDFCVyTm2(U_N&t5r3t1e}Bm{Fw}E1wm6xN&x{!;8}|_pDggzawiGd3|*6-tfL1
z=QliB#WFrCk$-_|CvRD)_oCwI^Xk2Cw=7WhyV=-XvD75pCZ&ZzPqj<>-gbnXKQ6uQ
zQp%djZo46rk>=U!{kmPVmU%Ad1|!7aX{Kmg60W}0QGVuryIPk_!tv`Riw>?h%)!c^
z)f)6f`MaR+UelRX84q)A+#Gz5A}m)nWPj9MX7vZdrsm+M+Q=ak7vKF(boA>b)4TZO
z>E|oP3?a-!g&m)qv5I3me$3-RotGXNokQrMgfUn@gjkK|9J`AZOOFjWOc?jBuYR@K
zxdEI0F3f)qE6lod2px9M+1(X#`c&hYSMmM#ed4CjMf($K7dj0DUkdErlhN1z;A%pB
zA>I?R+_uO2<ju(5-0(!ffL?DyT7FZ(e&ZBY`G9M1!qVsi9nZe{{wY^^Yur@@q;FaE
zeQnDb1+Ly7o94A{Vq+T(*b`@5>O68sy{g0f_%iI`i{Mn_p5+(vbLwtAPg=UA5{1ta
zE|{;%i@83{@vwnxecGXp=W)j*8N5%*t5#@(Ic8sr^4|nqIC1<-bswo(?>BY5gzhf;
z%d0^4yrku$&lLr)8M{D%7Mb+uKJ@5;_Z}0bDqL4CTm3Tdd0?d~Pk)x~{49OhX%Z}d
zt~iC3D4eyR<{?Zg_-~j9KNPQ2@<ndh=uPhn2BPOze^~fZ<CpJhml!L)+Sl*8Ql-8{
zZD9y2({OQT`MRT*2D|5++Ula(-#`BMa%C&-+adJmsNi$#O5cg)gZE~4a9=OGRXWpR
zb!aU=!SR=K{mrCc>Yf|-3a;dn`3aXW4}XPi1-8Vz+V<@4?GlFY@wbdVFRg8PE$`-^
z7)O@2n@i3&rf1ZXQy78YB)Q8brOrd&j+Mtp$IYGTU;6nleX_xz4c+-|!h_19jwH2(
zC%a1rGu50!uG~o8;8Q!ztsyPA<GlP3s;roQ#`Z>VKJjrvXN7|L%_qcwbBQVYQ=ive
z#cdMx@}hZmgMVEib(Uan^LMl^l~ZoJru_Ze5Ne%uc6PC?Lbb`uUK{nct&2asJ2!;7
zG=GV9+nV>dfPA`Eo?WK+Ky&r0<1^1h#2p_Dxm;GX!&h9JH)h+EduA@#g=;#(h^M6y
zzjI%f96GKk<n(TT;3&duzE_+$lX*^TbbB@H<-VTXXZ~2xEW*^6=CErGr;amp+%<Vp
z&0=M%-XUat+Nj`^^5s`EZn+xWr&qOj?tK;(QNvm@%c^45PGeLs2UcHNCpP236SVMw
z)`ef4#N`Zn#-?yL!^s`V>5XP(cklj@bv<XfYSzrrrz#zG&pDF9ZYr;BoOf-ditzjv
z6!D_%(qMf3;nWl_{ipn`Pw43$yVCu$%svOpZ>vLr?)mBy<5IjQ2FtK)ALgu;usFqs
zM?Ia`W@Wka`@E=mbC;~S7n(Y#A3w;b-+6grt-;AO8@-FJ-*>F^9sO7Fyi>`$pY8nc
z_|2iXy!C+ti!PL}iCyWl-mx`acS+TI$JfqHLbXd<wX?@<+!Xce%qY9_o4j7#8y$A_
z`Tb3E?2g%_2(}pv@X73&%~3wRb!&W8cd2KVjjk<iObe!~PhDvJXOP?&tB)>5Eqb#|
z*hA=|+iUTOb4?<fFS&MXf7b0K4RqW5`Hc*&PuiNs7Y3i_I{lIn-s=^RarA@tPW!hv
z`@9<;*9(`&491kpc9$6Rdt&MZZ>swGy`~;~k@sDrXsYAk$&IbS2_8+%jQvL)+v3&6
zo&Gv~2<;!=dH(RrWe|;9FLG-KG?mVM=!|cj^~bRSg}YmN=Vs6Gch;Y1hawDT-cOEn
za)@Qgs%NWxdL|RWeQRf@F!g$jm1}(MVe3T0qS!j7C-Y$E>Cx6ccb#reC{f*7<aflZ
z;$V@+7mXEqcz|yl+vc{3Wsh&P#@4*8sbybo=p;;;y6RhvQTht2Kkmo-C&_bK!tkb+
zql-!hw{}jEOP#bhtk#wEQ|VD}y1za6=!D}&JDCs8EuA$Xoc`=ETDT#4-`LUX64L40
zbgqnTQq9~j{=noNQ>u?lTWEUo(%Xr%A~Q<N23ET9*<8Ljs8vUP$(w^A1s$x5FaXEc
zbd8zMDlc1$op-sQcFb%DRn65t_OK=ANzKcJm0v#0-}tC!;Qir3?$m@+c|FU%|G9{l
z|ECS7Bfq8S_c3>DK7Nl<QGeD^WMCbgxk)4c-tTq1<9^REqWbpn&(L5pjSbqZr`qLT
zJR!f@ymswae167YR}fAA@b^b=T^|oV7(y=DHz)QDv??4O=v3c<jBc-rFzEIn(qoSf
zc6&1Dg;=l0C$)m&$;btJa7j0CM$`hElqH-4g?{a2&+snV31htD>|5oy_(8Adlf&$|
zW#!xA;*H!-R^@xsKL>Ai@UD#w-%8YWm&9y&(%E(Mz(&8Us&<*(X=gTUyvUFLYvED*
zwoLj}(>%Xlhfv+tm``^qe9g4Jk8Rz=H*b5ezbAB%{b;_zhC40S1|}bvJN~#wrJ9UQ
z?ZX4(b~aV?B~2yN`f6;pmmR^U;}Ho%DAtBqwd}B`(lwQYGf&6<c4L;St1EX=*XS{x
zcji{C54)<VmLuyI_;f=4=|xk*dOoQBMxH0>+eXcucq+(?+~l5evqne@nd5PfkWE7^
z#_wIws7X3S*Y)XXYr0v%J_U2?Gb6<l=FJ#o5cr%>{&e{+KI!H8-#@FoOK+HdscB-A
z&AaR&q-h_fdoJSIaXhc_h{5b=Bfl_6j@NWr8F@=DAmirSy+`M~DOId`Y%qipyw=s7
zP3DD!(+l31ytS&GyGlFPpi(1e&Mh;Q{`$c^8SUyd3Bex(1Mg6NR04s*pAVt_*3KrY
z>y`_3Y@!>kXRV(aZ-IWh8~-Jvv)3agJ^roaAnl>xS*LM~0{3$rGv3IiqoLsNgiG)C
z4~A54v$5^<OeXSK@0`;s?oN*J-Bgn6UEo%8WN;s=tvR#Oiq*34`H5|AVV7#QR{b82
zE!en!@Lr$Cs;~Ud8u^Rw4Iy2-ODATu%&)opAhlWeLj7y%>p8OmZr$>Im)4abn`mhe
z$Xznyd-BSpKcWW?WW{U#@+GC(dcrBZ$>$Qpb?*gVj0(oIWj){CQ^>eHA)<ZTpWy1&
zr{^1~e{P@VxaVqk?dip5=YLzk>q`oQakb67Ou=8qvZ^yCk91{SQe7YkR~omOtGVWC
zp4v0VVr%!!J}nGk(~515#?_C$+NHiY)v?}hri0bx4ndz$N5@&qu0|LyiL_g$v*hj+
zDWX6v|CV>3S+70IV)!$5kJrEMq+;H8vahEMb<+AWB{xQ^IA7Lgbu|r!ROPua9iBRy
zn_2SWuNupXu05_ew0pVHdF{H+hW6_kj$!Kz^`AA%TweF>f`7bCO(Nf$Hs<&*<oA8C
z^~Lvy&1I&235FHnN!#L$`aRBiUx;<jwR6s$MtdN)Gjkczy0*b<hkS(f8n=eizq}9G
zS~WiH`^W6JiOMZ1FB%$Nl1d&!D7e{e`qwQt-q=nYLRXF~Y5W)omY9B@H{WN^#d)1J
zF>1E&%|ARkv*=s|I$7}j+?kG2AGL*y+f7Hq;@&Si<v94zMD&8aHO!!Q-gQ^=<x{FQ
z<X1e-T*B>Z4&GGHYV*{;^w&Vj<T3ZQBIlmMSUt533SZ+Fty%Cb+2adzbfgw;HKyA*
z(6y2)Deh$LFY4S>uw@8+s55B1+iiR<`%$OyJ<WB``aG|te!JSi*H^8#w(c9$HC<u#
z%OxYXndd*Ly9C#Mz2s7T!m#M8_J_MOf<M~mX6A2dA1IeqEBbCCQaSdO)milHd-XtY
zAFDTbM*N8&g{AM}qhh&N{$yRb)S<U9BTyO(25(xs_pQ9tc2d}OcaJaYrRos6u=1c>
zJLN`&F}v=ZUGUP&__F?mri^8`4qh9hh?>|-jz5plQk<zCaxPr!W|)%U?rAb|h^PwP
zbec2Udd$X(h+gd}iBswv<tC?$_jYUR7AbG=d-Sv<>qYKZhvh|rUmm~9e^cnZ_qMv(
z&503KHdV$aEHAAPW&A31jc)4sehM4cv$2TVwUcG8CBOcz_nCy(OB5&UH1E$%z8hU(
zim$7#`H0Na2jds*SrjVRQ~jdg^^#ZNPf3lH&U@Bv;Z3{a$NIA3i^pKvGJXBM2M#Ja
zYK<<;oEw4P=Du8AQwPhd0+=KpIZQ0MR<dM`b<VF&4mfRRN|QKw`eCp9ht6v(ukQ-^
zD0e3RTutnQigL?_;yU|fFIIb=54QZ=qTQhO*A8*2)k(`|SDAOxTK*cV2R+tK>K<iU
zJHt&iReot@2WTceVqNMR2t3<dAK%~n*R9_Y-j=m2S4%opj&*ycMbD%6##?&4jlXDP
z(O&Bva-xUTQ9RWkZ&P66dzO25iSfOo%W@OsnuRKTn-X`ei=Ez4Hc(w%t38p_%FVGR
zGz_gv`Dk@*TUx{oHwRjdlcSgVKASt$t*l0(df=!>@D+n6!Lzv?!87u&)zNRdUOAu@
zALUtNWn(wTWV_$`SdQUf>N{Is3#_+cw#{h+xzuinOi#(@$d{Q}7RS|Hd{rW3Q_hZk
z5xsKC>)$L<Om|g#?yG4oeWS7qHa?1ftuf)M((ZS|6NbMOYB%@J&OQ}5$k<EHpAepB
z<ag;+$mEnITSmVXcpcX))6A$!+gB*K(dRq8<@0U7nH#uAHF7iW95#KTIWyr&jcIz8
z1vmVc);9_p%d*VR7;N@$YT{vd%cH{<F@l}Ei$0vP2vN_QeJh(bd(6mEVf&72W*znM
zP(Jh^W_B1?HKl6^aqCIl!++e=UA3e4vifuu=GL@n*{#EayG-xO+03tb`>Ok{kXJUh
zP~W1|&GoMoCx7v|spxYg!M~g2zullo;nyYGV!oG@>MLzCezEn?goH<9@#VAZXxQcE
zU9oEPY2o>!&`e{=D^J_zoNYnKMAUfu`@Ng1)P(_}4q{bzS7Y(gvZZG7uV46eh9O^>
z@ce>(v?;Ed-<I8-GVX<%m!DiM9b(Fd`JcJ{T0^M&?xh_j7s$(tZ$D>OnpIm}I&86N
zb+P)#86738owAMEU%1uveeYVI455gF$<=SS3<TS|#Q#}PepJ)$f!>tQt)*S1o0JDa
z%949L-5&S02GKV9o%dSh^zM>c6KA<&_w@<v_t(A~FJV4+Y&-MjzU<;&&(0zR8LJ^w
z``Y7>L1V#NVwB?V=DIeoye<ZAXnAksrr_Ax(pIj-|Ncp^{dDETEdv#r5iJt2_1R*7
z@TR)7&|O^v4z{A0+UIG0&f0-cw-5S!C5hHORp+^b@3rn-){pN_I9?ezP~oc(tzO;W
zv?L<j^^NW?$k*@O$?l_0Z7T-SevckPp+iVy(lSFn-*e+tn4jIdFy3)M<Kd4R4Zrqo
z+vs>D=nwqQQC}DC)>$6&ir&zWmL@BHeP2m2GkxK)=*AMO`23K?iQl8yf0=$+YOu{_
zX~;}-<?I$G<*e|*nLW+tYXrv^C&)Qzt;Iv99x;sHyln?Eczs2NF3-Jt1Nl7Pq)r|b
zI~!IQJUhNSp)+C@cUjd~y(7)G&7`Zrqr^DJzAuD)c+ciM#lEd!S7)woTG3E?=~4ae
ziSnZv1(B>Qb4&G>1-VX{JG8>gl|R`UY(4j^_JxGAuBZKGsG&~4C?kW*t<jcV#}`hW
znlkaCd&2pG&iX?EV-iJQix<o{W2hMs9OkHp`{o))VKe8ie!Bm@;d&)rQhNGwgO-xW
zcZ=dv+|R8aR~aha%etNK{34+@LP(gtw>Q$Ac*lNwmTgr#MB3^7du(3V+^SeI^YE>d
zZ1l%l*9`@s*S>3-M9kS;FxWt-Eu!;Gzk458>376NFh_2Ru-bKB#9U#@5NeUAJwB?{
z&1%fpv!y>*Oh&Mp=HheK`V_1!U9C5S^pxX-bu6saW{yYR*OI=K&y|KybUpdoGxa%j
zw5Is}yFxs!d0<<xoWp<%_fc*vU$vsFpW7y3JkP(hzgX>wjcsoHo^3_q$rDfZiEk{J
zY4JL;*AsT^8a$L6KYw&PJ-NDlU~ad`j$0~qVc{Qjl3QTTlDk(nopIfml(9U0UA#F-
z*d(mIG<Jed?XAi|^QsS?pD*{#;@0QZytprB>b%m>{#50+bnB~o8FC9@Axte_m&QbG
z^Vx{i|0#Ew%-}4dKWd?dvb8>&CP(^iQtZBc<BW~9?dT$&<FR+|w#9QouFFk2KkLBc
zHC2`=<{!GsTNY;+?<O^qZdQK!&e&wE>*_Ie-+BMrk@uHFCp~z!Q6rIE@~ak(jrp=P
zmC-pEa%(ZWXZ7~BKaW`%Y|9}|=<a*fvaR&l`w%~3Y}1Kt&Wq;WIdqjl+LV=#`r{9Y
zR^slbSIeq~P(|@Xok}soaPgtX&lf9Srl;Q<Tj8KvExSVYZQ|p1dG}w8lG&Y@J6ETq
z+xS=XFz%}G{w|%J9VgFQZ_V4}d&*#*cH;U!7R*u;i#R(q^V)($D&ePI$iKcJ&UEUF
zF{^6K{bLA)U;MRVF5|?KEi!cn(>H!^DXS{6S#<8c!h4;Bw`tX=v7^{f`>S(*+Nq09
z?pNrpMzhSS2QZVS{QcsYT?(5VVXw$lEi(SLQE%EY?ewS@LulihylLYGwi7&diRBJ@
zJ4{F#k`?2_ZNHYM5224qz4v;4t>Or=#Fwt;bK-P1RWJVHR0WfdPex6Rx9RPrvVm6q
zHYA^9rj2$~Cv@kXT>3Vn(Y^2Ji`tg7H9c0pl&$i|I?PxdB?U&OV#;TnRiC=SM&Y~h
zgQ=2TKDDbJ`e|CP)8}Vg+57gn`sa)rB}S8D>y4{zQrk-y*7hxjcJ~M7uG&4=$9<c-
zyT+id;7Z{Py_)Ov57({i9j$hKUi>vxxqpEq+RE~h+|y~1U!{(+J~FIX={5>~Ci5r#
z%A|rTJ2!Ujyvb6(F+9UjXxm$LSh*r)O^3H>w5p<2%%|B}kvq~gUu~JK7JJk(UiY)v
z?S&;S&BiYtuWdYSs~Y+F`IqeN+s~)d;`j3}E_vy-+39jd$nwrSzrK3qZyUDj-?*`q
zwKOli`^)^0M8{|qGD_~$rwJLXGoe?C^lRG3uKVS2r6A+0)Ay*l-K|H|oK?p@E&EvX
zs!RR*9~x(~W?=^^pSm@^>wl0Yauq5p{$!{uh*ZDxnV%lT`qSido&kH0*`$cl2FnN8
z^(jS$atoXCZ1$+kQApn=A>;f1N-=DH`}*8AldWlnS>8Hkm(>auoyjeZ3|up<aJLzz
zkdt_>#_4NMiD1RKEOU{r|JZ6XtH|_t-Sf)NuawmW1^nThUa(e$wu2G*SC{&ZAcw1-
zM6I5p^~+HL^Y@Y6MLXtc@31+qIXeH>T<gLMC8w>7yR?{&0qQR6J~m)gt2`2&U&p%;
zHkx$}3CcI_IgR~ubX>~J+u7fW27~;@6vQIyQ|)8E@(XR&7KcqZ^Wb>bf!i=mH6B7!
zhtP%&l~Q)?a<|uM%6B@d&U1XWJGOHxJ;U7E3`%x>I@R)0;+HNNLSxj+u1{{WpwGoF
zwd^ih{$huN0jI_a>(AY~FZ@|sDcgU|re)RcJu#_x<Gbr<-r1a(=~Zja9-Y;;ad+F+
zZ2t)F$j?m+3YI*(vwZcns3V<Mo!WaX?z&&m*Q*J$?p!c`+?y5F+OKC_eXpa!xp&;x
za9v*G(cXC}yPY}J2M+pmU-x=p?xo><`^(kHfti2AyUch#HSO1c3H}p)TfZbqb?uSH
z^xseHOxIU#VW@3*Q#d1iV$QXQgJ)m$G!6upEP0+WZJr=4{QMc4SHT^{w`b=pA7eT2
zypc~Tw(q+&)l6V}TaA0hCFkyBn+Hvw(ek|IjRkkU@0hwgSzM*J&FxR;r{%IcwD#RE
zxv#0I{Z@OT&7uOWZNc81&dy{*eZs?CX{;sn%AOG{Mq13FwH@)NreWKYR2F}YRt|lU
zp6hGkUC|hcb(<J(9Dnyd`*HD}RLu#IwI3GEews_ad*rpl{WUipRPXE0i&rRNEPZW5
zFL>t}q89RDQR0z0%deUl@rM`OdD3j#%jsxoH>%ILvPFa<kc~rWLjI>-Y%m0AS^0-n
zdL<eS*cTl1%5jo;d2YZ-{noB2FLt$uEZ?$$$$ve5TT2@sZPn83-PNWdnUlKTuL}}m
zy^=D9kb4oscg_%!`#yOW>&U&n`xm~YiPPie3U)U?p{D$uM8cT;K3=bPzI9#dar?7@
z8<yAPQuRFB(`V<LkPXNC-{dgdZ?FdB#&WJK{-STQ#ik~2x$O<ix~PFBYGp08KCO@o
z4mLWtnha0MSmm3<_aFSL!I*t?J^pQGm__V@xYrJf(erD=s+$^Ho6pYP-Sg}if=TYA
zXRm)V`1n`9!jZh`*J?JpUHR>Bzyz~Z)%vfG89XvCoxM`EW6Ifxc}d$GS;_}Te^6#V
zQ!$yar1E^E#@4E!uO=_Uw{Cl(86a4OCuKRpe(RI1-mJlF_es}Qg-y^lyD`fAvu29-
zxyp&{p9-3Ugo~`zR%S-AXqunF`J^jbuYL2hexefKo^=PKKeFV$onzlr;Px;xyVb@b
z`K809Q1d3g=U({{rvfUqWTy0~dMF<-hg(>vZQp~)cr^D~rs?Maoez-{^lkTDqn|m;
zvUwFm6T4b)Yjz*gJ%8e=@`{<eTgq=(>6Cm@GVx87ADi0fIqSpJ4Rr&h#kBE03rDMI
zX_~j>9xL4IQnFiBrdY+|#aB(W;Ic$)z_FwF!559+mR%GTZPzxdxb##fJvsEog+G4#
zYeg|PMX4vl&V|2Dy{3GkZH7~&dds`H-!^1?&bSJjZyrWYTpw4t4+|&v``xSTSO0!m
zJtU`l`=v!wWh#+pQd!r)Twmjw*puov$fh3c_rczQ_wr5Nx7v*NxkReQkOj-<`XAE@
zBj>Ffm8E~-tCsunNu~-)FKU8S){Q#YpLVSI;LT?r#O@OgCC7CbHKfJb?~jRadszO<
z@i91qbgI6)G?_a!Oy{v(Px-`1SJkM-Hc6L>oY9K%fmUqVGv185u!S;uqe*ZVwGFms
z+4&glW}AJnj!CW?bpoVm^T$YIgXyHL{zC5+E7zzm@$0{D<D2xhZ__)5f0w4sn~=zd
z+gF}#)J%J6R^z4GezRn$eTsL^OPB1VETgKiEB-Vad>Z6{?l2NG``ncAf?cCBnshZM
zUA=q9?D>IU)N=IQ{fDKYS8^@VP_c2mmcRDhgni|Ck+luwrO&ps|4z^h0{bKMEjAR~
zI{ejG?G%}DbMrPo&yG(`>$-}0H;{%&V=2GgRU#8(z%v@3=U3MC&O^x2mtnGsC+YLk
z!ag}W#-h$qYn|IC?@zRO7*%`P<Hk#8`n&n3KW}s^ORDa)*hD+|=3CRcqlTq7YG<i)
zCReLw*VdD_KW|F*;&$YlwVbvJahdddrhI*9tunpst_gV~+|}(o!S!&9ulCyOJ%m(B
zXJMZ_^wDS?`6G2x)c0*>9g!_$o;GNi`AqxygI*H%=bRu%licFTUG$+R@P3*6`k<(&
zFpKHa*F;Y9@elP4nC2I{dU~`^7`*GEKOI@xM2Eq9Fac4xZ@`M+HC7s5&R^2NgZ-^E
z7MRS{pBrWu5ESeb8xi0c>+0nfyVB3xU&F>4w~V%kULCeNAj$`iUcG8fq(!uqhE%u(
z@Ko}24P2TfYNeHiH@qfh7a9?O8&5NyrmqLDyR9?S02vYfffgS2vqmJqm6gVbs_WOU
zpSIp;T4=<I=?3QJ=F{~JryCmTK@Pphm^D#8(RynlHGfF34~X=O2o8%14qby&5`BC_
z*G5@sK>fpttqz+z_iw@f5rfsMDQ%^>BcmMFLE--mu}H6&uz=|v0g<6=Bm4p!)&;DI
z()>}}-|z3TVQV8+Ne%G#n;x(#U^Qfm1jG&gu8Mkl?(f``!7GAShe=BSvVZ3S{_jlx
zQF*Fru+YLj)Nd_SiL<?x1|`YDKh!VS-$Kfo8^E8Rk-wgaNuZ&gvA@2do^N14fZhy$
zvl#}y20jLU<_16HIIoF}@>$~-K*<5g(}F>WiK)5Y3}Yi7y+9LVb3J1peIq?{gFtgV
zKSN(rb3<Q0fBityALZR5g29b^R{g9GCC4A+_?a2_LKy=+Q!{;2J>x)gGmv9!sOJO1
zj04Tg4E+MkHE@4Fi&>!&t9_syf>-;j2$&uc7O+ACr|`0{6+2sL7*5k4xfHMRi3)_g
zRvL4H*Q|{mN%aa|9pD+|Lm_7{!${xU$jo$xsgbe1k<pBi9P<MrBEerMQPQ^;rOI2_
zMFjXng+_RVhOV;G`1|EX%Fl)&b_$M!+c6`a>lqL}!mSF%rG8`K<41X+l}4n`x&VKT
zkwm{BpEWB2pg%#I4rdGg4-osB1^W31`kLwK>j%uxGxjwy*7GqoG1v3)pW)|c5&+kf
z^26mLLj$AM`$PoTu7LIiO`)}=uVLasDGEA2@~saFpiG4KGW84$^z===M&8Rb9e%M+
zV>;+Ro$4~frcD2LE2Z~->RvyxL!YJCze9jE{$Khp0{=zezX<#nf&U`#Uj+W&L*So9
zM8F!D>8yt(!q6MqF=sowWnvGvS<Z7D;5B%7g<Z1BZ*?RCUhQAKCMv?iQG~-A**L2e
zzIma(=inlyk6&b%tynx~`0eTc^xN}r_zn98JxbPp%JILXjq-=Z0=!R-Lmt8}EFuc{
z<-nV+j|!vs-N5r<!sZYBA>gM*fP%n3rqX?f`HvL8LdsJg1t5NemxmqLMtxm_ulR}g
z{fYMridYSKj3JMf|LQd$pXNQx`&0ZzieJ5IE$AjC>b-dRfHm_x;j9IRDsn{5@cn@p
z;m8-QMZstlib8sDUIQtS$OG<+{)h7L5d@SMLn&V<6^#6#ggpvHtC5eiJfhkF@&NsV
zZDgR4p)?7T7r}=kFNcOYC&33W*$5404h?<VKQuI02%j)KhYyO@{8K(uf{<A^q-Xw<
zrd0q9n~c!!*ZxWK-G$JZRD_hDg!x4HNb8})FD(#0gX$<nXw*XZNU07!tXVva8{DJr
zqXUl+@&c^p+(l>)d=WKy1;}3VKiExuHT5@c|7)HR{G^{v@JDkMxuf`s3X0>D_{!==
z0(DhYbsN3ulZ;$!UFSO6I?eJ}vd-UQL8$jEr@-`}(Ac=-lw>!*?RzuG+;s`b@zT#G
zl*WxypP+6j5Lm{0IeEqZAJfnk_-sNB$qmpk_+sG4XA|S8uf`~%pcp2ML8mcg;M0R~
z|8@gz!<Pj^Bk%6>=sdW7d-X7myR#V{Ph2pr7S61|a5mSWoRiEzg*rhViIp5t3XTYb
zBq$TIGh^`I%Dc{F%%`fNk<{-`1Rt7Ugd9@f2o<x)BE;t*4h8tXQ5ugwn}L|nHE_%=
z+vK)sPPZPSB~o1l%0O*!1HNPt!`TClVIsld{;rTf=&O)+<&aoT?3lG6%bu3Z6p|IV
zu98j&yRaPYGN`PCm;qW7n24a?jZu&*?_QrVAMPO`9RErLAi<$A^f;t0pN7N)65{`w
z0j_vV!hNcOEeA@ok>}gFk$3ss!XgCn5fdL4i4mws2Qe5-$iiWB;1WneRY)T-4qVXy
zNMjcoABV4Yl2Hy)vu&OvW(Y-<#d>0nE#h$CvpN!^@o8C^lFbN_yky)eB<`0tPn-aH
zDF1In&_Hb-sGUZmks?xqLrWaV!THA>NG98zx!qQm1QC?p;dfK4ZWF`#Kv8K%reL@L
zm;u&+whSSh0C5UH>Qg6Tl*%z+FtI(_fn$V#>A_1D5IV~_Vli8n5&_w0DLh;cX<@P+
z0=Gp4o+YxjP=J7Y|3cD_rZ_Z#gBJ&s2GL$4SYu9992+J^5tN*?osht#f|F<S+_0Sv
zR<j8ytRA#XNERZJKMz_Bh+s=k(sl&^DI(wimi)38JV=O=j~Z}I*qECLpJXdeA$45D
z8kt}mpOk+5$l=qJaBx+xEgrbl`?#gMH|t1CQ-XdX8t96|G};yi1{=VH3!xM{@L1|X
z>Z=%&f!XqD3;C2Dw8|7~|A#R?e$P)8F;S6_0u=GU1u*Ka7{-VcCZw?`_1R(q5=aTc
zoIqcSJRcetLUy$vlp+?<_(hCkvU)mf=w3n{TYQlsRg%DLgD{2%T9nD0<g|mqT~x53
zL<5Qr3y?|~`5{0-j9~9@Nn`V9b%;S-5ujlvv;Yp)BC7du36IGT%leC{^4>IlU+2Q&
zOsiX99H=G%Jaxb)Qb1R!P#X2imZl*#)y(4I1|%gU{fI8fLn0o}i-DtJ+?FY)hl-Je
zOwn2-K{!VTD#Hi~f-;ELDHi?tq}#$5ho1C5>OIeafgwG@oeCxr1U2hGUa109{RE;h
z85~&=6^)prz^%Y!ud&_zizH&t!$C!jI;agSCcs1(jyYfiGzXuSNFzZU2{%Cx3J@4I
z7D{9Zf{?{TYS0Xr9B*D}VJ0n;M>^<nkWdSNNr1{oCP<^Hh4APvz{@qj8mi`hCD0ez
z7s&?U2-jnK18QI#4eHDUFmY*K59EaqE@n&3=BJS8vPrfv*P|bRlIjXYq<}~uH;p8u
zWgs?D2tW}kSi+fx<6<q0<3jKY!Cr4KM9V?q7)wAurx<J`VLAaZ0rTJ|FfoL05hn;-
zcXY=B+W*PLaZs9xkgkw{7;uJKd6E=?6L3Ft<^bq4CO9z(fH}F-_?JyGjJfV!Vx$8;
zEkQ^?KyE7Tv%$#)xJMeqG$+LinKXWBAxS|@c{!Lz2!%YvccGCqOwCOL-rHyDuwxOw
zD>lI?rc8wKfyAUDj>KRmj(C*3=F4o)KMOo#HUmIZ2l^LcKrX<9=>U51pbSGOfkyHG
z+#oABKM+$1A$w4Plu($^D(U=(mUULMT?9_^pd$e2BFF^Y6Nt)HEVRsb@kq?H=QDNS
z8(&(^aVI%KM1qxkBo<S~A;km&hxkjQ5%Y9x5v%YzN-pCNlqzD9cZ~XO7f8#U#}x5W
zbX<Rr`=uF&#*j8Db>|21Pznw;a?;QU?tIXg!U_m}`6R*|NJPUq2&c*xF=?pE<TqpP
zY##*g0#GeBOk&IyN>NZ@SQA+!M8!fR;YlRYMx|PZLQO80QNbWEA%j4)a$i0v=9AQ*
zL=}S&#1@g56Pv`EIDCF-mfFgjqm<&}0&zZ#1CINz#D-E(onjio*-{-q>)|n+8hxoa
zhQ>leXdu2glTurP81j+cokaKa{V>EziSl8OvCt|o_L3M^d%If3JX?0OjtdN(TrHi}
zE-DvOK)^3Go{PmuMWVYI<{*fMI3znog20ui$~Z`t!}g%M;Us7kiGYpJz4C68-HD`-
zDhaAHaS$?0JSYWjI7tH03%r9N6a*pO{t=JDz&A)j>PLx8hP#N6Pzi_vT#0|8KTSyj
zjRA8ls8reu#0Umyz*v&Z#93L1i8+uLj3*#GQPxo;mdlkOf(HoTI!FfwRtA7g3D^;&
zD<VM?lH6=bV#I!5FU0#IRXc$Iv3$_)AHt|BCIvSl833lJE(Ni9RQI5WN4YvgVF!d#
z5|FzSl_rG6LWB2l0<=a9J#Y=FI3|K=5!`ixVSpfFkfckLnJmCu*dl(Z1VEsXBQrJ#
zHF8sFW1Xm|4jL0;f}AIvRe@Y0Q<6Yz7Na7z7yv_zqgs<jna0ia&y<K^?gHlH;CPe{
zQX-_DGD#ve8e!B##9jh=NQvZ=5&(>kXlx1C1_Vmbjn5iLtY)_sf;vn>gQq}9AiBXl
z4?gmukwUwhY?vzHI;|vh6ek4ZIfo$FP!K?qhFv{@EhQ=i7fL5#Y!L!0FoLNe2D}CK
zIy?s&N<?fI9xbP#ylZV1k|2-|aZfrXP(?u}_=5>haT)NqR4p(ZJpLzwh{C1;9zbA^
zeEq;cklb_$N>Cz6cUwNfY&lFEoRtp|iAIC?;*p7v)q>q5jilULLgN)733wNS8iA>3
z?}7_|1N%G>ArOg@wHEQfM>zyo1XGL@0UKkJ5=EwD5-N5_vQ0bKE)P~TJoKf8Yv>Q4
z>`$1p$s)wWl}L~aH(+{9*#|#AkvBs~!<JJCv`m1_=yvLOvg=~uW{`tu1X@I+ZUI4g
zYA~np?}B@L2*w=DmH=TMR9X>Juv}D@0!qk2384;?gEc|==0ymDyO5QD9!4<0g=m;R
z&N?u^wC<s=J3;A8Igf;dkT^^Xj04H1q!-XV#3=#-htu%Q;a>#;B~aZF`VCZw%w~7C
zORT1sNhlQ&Xv70QkuDk72xqVob0iRwkVih&T7;5y+z5I#<dA@e!T3p4t96RT2l%lN
zf|gYzWW%5>;Rz)$IRI;6T86{)o|+6Jfvi8nBqVO4Kh-f2+XL?ZLok&GyCqZt*aM-T
zk0~bdnG8?|2qFwc-Y8MGQp$lFB6|8+gEH>y_fY$=qe>9C0$6}k&4g&YWU$VI(vyk6
zcR&=5KvBw<MP#>-$0sqwog`%F5lBlY_8<w2t>Gobn#u=G1B#F|t^J%)u?2PuHAoQM
z@ssBYArMhoV-9dbS|fy0K14$u4mc#0^h0ZgB1v5W0Se}`J5QF=?_W$FPGPdeY%$a`
zd_^W(CXx_@JTi-M0o(IltE2;h3-kd38X5_C70~-K<eaD?Bo3_w1JEKKVt^wEamYb+
zUOq{JGXUCCN#3LKi*J2FG{bt8%V*D=Sc?VO3vFL@aQ&G=Rw2|#%Pc|!6=^C6s(z|H
zz;p^cn2Cr|AOR`{#E=r>Lap}6x*y?Ki0C<&Y%DUTUi=Mnf{uxF!SyK(DWZyjL$k6#
zh#Qe2;A}uVQhFc0&o3)m#7O|W-9;{tNy`KJb0+ergWI@3AjUZ0t$-aYhf-ibAi;82
zYNr^#U;U6;DPwtkvzvN0xUM+heC^>X89$*_el62Nb#t(BCNv~WNRd5+T|P%bB2gm7
zhhPEG&6dEKh*Y(0dK1N%w9<-xv#H7MUXAgi!bH^!rd8hM(Wxi5KbN#^Y}j-5K8DP5
zdcavJ^M@OX*A<%s6J|^dGXI5iIDk1*$i#8ME3gilkd~N;2*|^yiIoW0y`d$OIP?%q
zM6+iUapVZZfYl2rVM-_Q$+%^v1Ue76h%L5qX~|t5C`!Lc-uiQlN*(!rgN3?Nsb=$Z
z!^ug`c^%R253I`(s^o02N;*f7W@R1w#$|0VQ>&?RDH2@aTUHtEuWK3|`2LuYnj0%C
zY0vYf3oC<*6$q%FEY-v^Th|>4e}8eyW3@T{hPiXd);Ric+q}N2B<ISW7{TM3xTMEs
z%5JR4@Q<nV?RW01Gs*0Uw3t{4coV4klF|6>T*vYtz>w<Y)Pw~5iE1M@;&IsSwl1WY
z0t^6vB_u%*2!b(~M#4Nub}!PiMKEJ#A~B=@vzaLtd()7t7mY7T7Pue*<#cSZNU>5d
z_gT$de-JWyGX321K1ap;yDU%hu5H(?7iQ_Lvff@<8M#dM<2QT6@B1|@+Cr_iWCIUf
zJLfhz$s+1di^J#KnpY*twM$$)C6-6CZ}xcJ9}ORGj{fzak^PeHcNRzQTivAKNI`ct
zCW_VPJoQjcX9~EeYpwO3=c!|w)v&ty7{h$7W1VSc$HuGX+*2E@eeyasu>AHog`22*
zfnI#XmFqZu)PUo0i^&vVG^HGVF@l{1=+m%f<3KRC^+m_D#7M$`iHum33}tW}k$@C+
zrpXe>%LPfmrP(@)lBsMqv8c;~lm)>F2%UC<JPaMSJ(5tP8<R1~o=4x*c@*~Q5MB0u
zy}UxlhjK~3_DR9Vdim69<9)Ryb}h<oj-DUE^D_f15_3nj*djagj^D&}8w}}}_ixHt
zFXP#qx9zw67HY05n>ym`zwk=NUS8^16qmiHbI$#x@49QlKsl11ShO+sT9lP~SdV<^
zEz60d;7(hVae){0(Ns^Ir~92Z{+4PP6<Bp&V0LR>ZF016{@lq8+vT;B-j64wF4A^M
zbKI4}{KCUl4u~g)#W)k(js&|ggh;khB1Z7bJj7x#=0J4;*o<Ky7(56JmWDwtV(>BE
zIf)COKx{&^ljPacQYB(UvrL5II<pA@codtRmT4tSMtlg#E?b(CX%}~%lp>>seWc5F
zvg5|IW~yw(Z*0sAwchip?}kzN+2;gyt<RC}WSL)ExSogWw12v_&+W(t!vf3H)1M2(
zyk6$a=?$(`MhQ91!Pd#5-H%!p@VnxAOZ?t0u)EhiI#Y<lEJl31^FuRzPFw}OaSt~z
zJ6t|hbNk1RI8P_*g~^ZO#GDv&)w!Ab=aeMf+ug@9<PU;2EqNIg&}3j#$HqgUcSJiE
z^JzIbFs6VJ;2;Ep@)oM+P^(JV1H&OR1M=Ha1EmlyAcny#<=ipagQ-}MhMJQIhAL5h
z1&BxhIAKH+;>`UR+g_1W(ybFgctkYtk^TuahpE1U&~<g)@-8<CfeZ@plFsln{Rh@t
zPDzlzwfn1nnY6w->+PUl;L&ZKU(^zkE7utotoq1JjIiF)Md&m&M;lkFakqvVtFP19
z+_yvWGK9y3aZQN5>Mb4fG57MF-qh^iE;k3uJ#JYX0G+PT$<=;jrM9h9;-r*}c)cq$
zFS{LEiBgS@fhI8It52WMNy~pLK~y7(!9G3@rV?PWm#(B3)?X6Pz6`QT2MkCDARrS2
z3MUWRYz?JydohL)WdMyw4ouTwTuH_!iD1lff)y`f;0KX+v2Gfavbzf_<G(Rr7&z7(
zWz}1gIMYCVYO81Y>&Q~S4-+>dL4tK`m+9VvO|$&Y^%Uc@J+&!IYcIw!tQYQ1UEBZ8
zuW<qWUUy;A*&>8`d)l1h&c5+6U)(jSj06;jn4=D?>{qI_XM<(z-X?uktH#_CBV`)j
z@!K|{byocTQV&HvF@daYU)>ostKsIME-D~`AE+CRZx^Z+OJFpE0bFV+tjeiyi77xZ
z!zGe=ARY1o7Cato3BaNlJOT*_l$GJYoC<p=POxOd*&L4?*y`gYgC~eCJ2=Jcw~vwF
zOs2y89FPptcnw|T*&1rGHj{}%qj7T_?&NZhh($>{?KL+}9#4oV&HlnKLTww3E=&5W
z{^%=F?ufg&K_XG*ZV`X0BIH+83qB5Fev>wag|M<e=1UOjGi{MQ^i?(e;b2*7BFr)|
z`NH2(xU~xRL<nol+kEa>59IkeZ}WQVb9K!dH7_5!zruR^{bZPV(~``Sn?3bql3K@e
zsF^L8+G%torZB&_82le*7qA!wB}uRn_BA*-M$ky8k3n6-7y@M^(ALsP4@?pEBQaqM
z{>8u<zd4;IQotR+o3i%1#7-64U5{bH#7DwslX^PR?v<Lu;s14^^&wFNvII@Nzn)BZ
z^-cRkCS9T9F6#+2U1tA{6NsC?u`bw@U*jX_8u#udvHyzoDDLmUsIR?q<>m5C^`6EX
zV}EVDcc(uUywiH(0ns{h71x#Be!kF01yd5zU!BhmV%^bKt349T%5GG1JG2hbU*)^W
zT>VJj)8@u{ITLx_v3mfsnbF!O6FToS>CDxBU`06t%qu7)jukJawigKK@X#$FV8XmA
z84)mTORR%X0okO)hkgTC01(p4*#uL89V0-@GLOU@`pss1KVlI3Z8hqW5&ixKGcQ=_
z!=5N+Uzk)UWC$>X%L?6<F8<nCx{!qOG-Y{|>KA^Dq&F$oq`nkd+g@L*T`y!AwWx5l
zBV%^IiniQ0?o;Vmv8r2b&hxnCG1kT__DjAw7&_iMlKRks_Z3_-V%avA=LI73g7IgM
zwDxatLPE<uQv|zv&TrMZ|7!V?n@!(lWpdN)F2r%Vh_H;D$Rj1$@97C`313}br6%P>
zn${svQnR`}3|(rj%hkg^!a5X)v+>S{ok40luOpIL%n(f2sFE)GAw*K02!aO;lKd2C
zXA(dJcN!sFXn7l7Bw<rq8Gs~0WP&Bh7JG=vLLD#ruEIpEG7;@|Yq1VE4EPya1pf@`
z=z$CG9xkT@bZksAWBJ*3RU>Qn<gQI*mtFIoaphf}wpd>8HH!%!f}5=u&MCG&SAiCc
zd6n}rPC+FjC%SBn^|6fHz%vo6eiNx-H%w5#vRzBZv{u#0aQaf`eN(sxnnWI|x@)Pn
z?RMF#J)BtUn~&*l_pn!PHMvtVCgEdiT;y@L+>KZ7_0&{l8|*ZiXWfQ)s&46Vk>_^z
zX=kZr1U9ykly<1}PR-fCD?3(sR2EC%l7Ui~NJ4QG&x2p;LO`7)NJOQ2i;GYSbdnSR
z;wyMo0*pAiwEP5u0b!TiqC{r$P^^_u&3;j_W&Y-vNjWFdFqFoGZCBXq(ZE(xv12zS
z$f!VAnso1DT^x;!wzkcKNx;I<D{fdFt(!}z>sQ9iQ1{k2F+2HHtg-rhD{Kij*=~QA
zZDmcQRe}9qiWdb+rB8H5&zjd9Wt~p?E!!*ZR}7VjeDh$&>@JTS*vsl1Efaa~fpcZo
z8tY8dvdr30&9&5X3s?JOa2LeR<kmGa_oSB^J-PpFIU+V@>pZRxG|G4JeA0DSLa7Gb
z=j0J{N$jqB1UZ}_eFy-EK=Xsm7)%$SKT%gWiTFy}Q5o)%%zzC%J`6a5*;BP*aM(*P
zLP9<&0<0h$QTt&+Ru;clz_k~!Ik9NF3kGX4#K4i#-EXyV{#3P=>XtorH=F9Bj8TAo
zK6|D0cCPj<%e@(aiRx2tMVqJx;N{cDG}8BXjXEaioPss=thd}JUwR<?Y%4cCK&NLx
zeV|}VaM#={_QsPM-G|!`>*V!zw|@Mi(HHz%^&pm`#&ta4+%=j+0S0*-=!o3FlfkWO
z+|71RMmI+QLUpEzh^u|gGQVayk8aytOHfK7og|CbdpWVo#0msMLM6~()Wu<pft?5v
zh6+B?AQzDkQ8APP*kA<a0^I<@D<Tjg&q?5yhj}G$5fNgTu}ii)F)<%61PifN6k`Q@
zI@_CIpC8sOq=0(t0}e9YBohil_vED(=A@3f`6|_p5Oh|1J~`FAoMl+B^qS7JNSI1X
z`meBV9n$ttOK3{<dp92AX8Fq$m%rk*#U-5@t$oKbf6A)!*-u{eIaZ8+TbEyn(3H!Y
z!GChMmP!!AT;)!SVg6m2D*#Hq=e?%ZXu-#>b;dj-|0_)GGkapqGyAoVQEDLC?#M%Z
z$JBP+2NXwQB()Y5!Igx%O^_TWBB@oH4i##F2(TO0l{_?hj|8?B7%{j`5Mj&df+2}P
zV@eS0M7hu$E8qzOiGCx%31}!ML3XLVtzBzW)vzlFoFEwSikJ*m7AcNoWnW)!<>X%7
z_o{7!^`4LI>J}&SFW-N)!D9SeLZxw3gvCTR2J7U_E<cH3UXa}r$tSfrZ>{+6L17Mk
z`>UQPP>UF?ZFH}$6-I{?22$IMP#bsokDUX=@y)%R#<aR^ODRvy5fPt+=>x3GVg1iU
zr<kzV0t=~|pe+NDMN~s0=mBD66-YuUR9M86<s_4OFwj8e%@kHZkdFgvPaGw05pQS4
z&iG*-1IiOPWPnXp__O49c;0i0u*@kIj7~j<y)joyiG>~gYq78c1lwv9l}XXYfde&$
zyCXGtONq9-<oKIPt_Mjy^Mbu1k^^3ds0|0IdML=_gT^!-cx$Ev%%_ZjJ0OY2rr(z^
zafIVSrXK7?foz%*jT$b2z&wf+m<Z1>!6?t1csi3x;lY}p#4uaP!K1E+2?x1SPFh+h
zq8`_Qol=6?B&9YbUrNf&$D1Xd)MGCok|xEKf&?A|-~g{ZHG`E<)=*kVer&;mW6`Lf
zixGNM3&FDtX@$@vlo*KwaYUpo5)vYIa+%0OcICTJt8PkRfh~Ydb=bt8RDyH@4j|YC
zrhw#7NDuFBLS%ynh}16S2*U9HS63>9Ab9dm`2@AK3lWXYfh7h=0o<hRITfC8LQ{)K
zD$DjDxFxmM!iC8o$6^984}&ftcK4!0Fqi^cAHstgd7+&c@UJA<xC@D5Pj2e`i&~A+
zV0j{aAK{;ehm<^8Ce#Rl7Pj@IZHUHFaRw?KCJA_@lXL<^U8_t!)?Cd$QJ+JDKE~mQ
zK_Eb;lnhfj@B^@b+GPNFI77NT;8lncNhk)P9Y_2Vmw6l>xHt!gD~T2pv&T4cqRkR9
z*vHTXB-zO*1s+5QQ3q30Jfo{>B~>M*3bj{>B;Z2?AR#3VfrHaX;3jAeL^UfU0E1JF
z4u=#dDiH<jl8H)#2a7^4wt^GMPvL{WYx`j{k@9S+HE2i!ijdU9HUMvghWAUfP9Z`J
zy5n${kep)D#@Z85Pf-!IJz50MQU!zr2TEbjR;*iWoiS(KF==`5BESSX1OxV&hE;^F
z3#k+d2+Donrj)2og#!n<Fz_~M8vO6mm`$5Z5Ih5LfVBx?ibb0h%cRg!Pjg_IUIZ3{
zvLpSKssSb!l>AI4->Nx9mIrJxU?$Z8ela*;%mhc#NP{LLF*bp%>Yr%vw350(wONsr
z0Wp$c1_2^KFtd~HoV!S<9X=pH0y6!C6r^ESmZ;oMfE`^S30D$0Qs98a5Bv@j6E0!n
zI0c8I1l0+o*7J0?@Rbso0tftu;W|`%iz)31_!SXF0taecC1A7v72OfB<d5DYl?wap
z5aOx!|M6r9nhpd)<^K>204iZCmii{<7Z*~cBw)R_cqAr5KctbAjX(sb$Oqh*WuO5?
z_Ap(S03wMakEU=0Z5Jd9ZxPTSM7VP#5=E$updQ}g%#-h;suoL&5z=31c6%WKP4<IG
zU~oVk#svzGVWRkdQx|9HY}&N88I_g?0W54G_?(j%UIKv*2O@Adp!G#%JRv{|dOAQa
zfbaoUFfcxwy@Pm94NnMK9*BT;mb#l#qO_MFC(1IC>|oPkj~-~4xQfa|<&hE_6aCN&
zu0aT(1rd}A7N~5*5-KuNO8-RIt!4Z6IiU?ZNotlRZD>#ko=bDoxZ3XU+>PorB&5J|
zq0L}98*G86Yzbl|F@bZiAa0C=Qh@;Islv2O*czs)q=;%^P!Rlqv)FpqoTy_|B2N8K
zB-Lz~rLLeXL=Nx<;Kx$G|A{z6^seTXrOP?7yFfF_8VLyn(GCK<h7DUjl+)q~AcjG^
z=Z^RcStk%fJ&G8n?py#|inp|rQNAaE;$l(aNCq~zgRC&tx?osaVkG(BPBge?z}B=>
zF6Dy%k-&F%^qqaQ8~uNTy$e_rSGGU8tEiTyG0<vb0+CKNN^FQG0-69q{+mW35`zu+
z9-~Gh_(~>%sKA(SgS4ViLB%Ktq`+3tn5a=_@;>Mod<-HQ;_I0F86`*z;~0bHn3?l^
z_qVFiJkGiIE~Ot;yLPSjUVH7et94|5J*G<VxeAuh%9!jlPLgHKV1WJWOz`yN5UB0|
zdHO;Yi6EIYpOr9Hu3-L3W0~ZDvzLaMt?%Caqpk{s)c*!VFxm_9zL3yE+xbmfC$oL~
z=If(6nRQnW#=S65CYaHPRfCXL*&={~3O-W$HCaas(o@N20wjssczh>t1JKP7??|ix
z*PBu9lFT3(ZazY)YfP~%xOsNhr=$q~Z`d-9h7ctb0u>S@8~Xk`fCBQQ``a&WH}9H-
zMHwLk8o+S)VG?!%f1rR2TIn{Rrw4a4lG6QzvP%!9<=)#d$ia_8-F3~xAjve6gpJ&S
z_b2~9dO*|?vRKXs!zj+v^drB1krIHd^zEB7M_p6HUPRymswn@2;n;1M`amEGbL;x-
zA)k%$Sb#vc@tGxHrZPkVKhV3Phz1i+FFUZ88O#PKugt|j!F)}cS%`sWcYSH?#w{Mw
z`&;fkI4lSW{1^WVcX5I#v$k!>5r$by_}i@<OG6|)#x&0xVJew89fALcLlpE8fF*tQ
zL#2ioj=l+j^AZ<(6^R8XM&Klz$RcRW?hxEz7S??~$4A_1L`xNah$`sy*PwX5K!k)P
z+zMg#J0S8%5lTeB%A|7g%ph?RP*KD6=!Tk8*_RGLUlTYFJwfb>e0hAP3wV~v#pVDd
zym69jQ4}U5(*dSwmcugQOb2K~tj5fD{6oFpg$pY<fSw2=1g(PJegkG=44RWNFoVi0
zHoX59=qkhgyo>9_@L~DozUfAwgd-YrsbVi@x=CTZQ#CM?rVGcxbfC}&^J^h0>dacX
zc`pj-a3hN}ne75w$YVQ!7AS!h;=`DRG3I9o;iVtY<s5%xPrnDCYAAs-!Euew(Tx1}
zKv&(b_`h+);eHEP3Rw=yGl)d$*ueVDCnKY8!VF>*yW+I#K?moEA4{~5qB3OO6S$6s
z&NV0lU|_MwxXtF`WclYq-}Rh$mktv$=!9N>1E5NAhU<#wVg3Ik2go3in7RC{YMauQ
zkB&f^f*b^6Dl#w}O2mO3rZWn(z$7JsSV?wz##y>oe5ZhIV)fm{FzLVnbb&|ZCon$U
z^F?K4vS#aVuy+6avPXBp!3SUw4?fX5XL6&d3_K^o2{ipxe{uJIe85`GJ}WInWs1gl
zouGCO0j)&@Q_i>sW+PeOhHoYzmvIl@6$8Ry^G64Ug=6R|*A>MhIt-ysGe%%;;!H8i
z$mjRpV|1{Bwx9ptVXx~Kj16ip9{L+-J-~`Yl2-VD_h8R&LBIF&ujih4)?&cpIT!^)
z2yp-^y(HO^;FDBYaAcrVRC?GFpmp4AnN*mNl>P-~p)&KTI!si9Jrr#KROSj$bxgtw
zJ^n>eY|Q(;3A;YcKY;V#!~Y6U2VYFlTugp=`T^b#4tUUg6X>!>+nlPM4I?4^dWJNH
z)e*A8$wLzjSUoctCvhP|=qX^Gut_Wpxl)AigfRg|eG_L?#v<^|jFke#`?Cr49%PwH
zcKW{nm|G{8HIR+MSUmpd_w3bh%In6B8ByG;?a$EKA~sM%?1iH`kUTMv6`^bk>`h>-
z*t;yjBn>QdYGDr8n_`~IhIzEZ!!&3O;^S5@8r3cwWGfuir#54P6YBekM;koM`>i7A
zcf+!rVmIm&0~&+@zvr%w{)nt;KhDOqc$mTDW1%>JS{fMR9H+G;mhMz-+>pRg`+-GD
z)OsC;hm*}_d^IsFM*a%w1l$C|1%d#k4zY90AX#XFTyyvNnjda9(s47r{0iK?n41z{
zXe7V~02F=$9=;r{&N?G3@6K^?oRo#jZ4XB+jH0Ug5T~-~rtsoACNOeX)?wnHX?oo;
zFqgQSsheAImGNP>)Caf<g-RCMK(Yi1bftf0eD_c5K57N$r_%c`kghvwy+2)h=!```
zk-;ng#RX;hC!z)=TGK8pzQlaztmU|e(_|_tOfdH^HiuGfaTor<aiV}w#cl+Y2bjuD
zb!;L-87hVe*??#$fXj{gBoi}zrl3EJW5Ae`5m+$sK?rG<oaQ2}@Shns|LDOPZr$|@
ziMR%(qqrNaOs3$UBTUg*>vam_02P1$bl$4?$~`*bWzF3Ji#vC7Sh&#BpcuJJFo#8L
zC2}^!z{JEvKnemNfZER#ZXSc$o*~d}!MV|r(JJ;bp_xgSrW=LF?kvnz0(Ahl;RJ<K
zF&G{FHH2rqih;3>I|cPe(H6ZUc_SU~%GlJ^<u}&J;>xV;$lRNyzpv2Fudubk2`pq8
zBXeWHbWH_k?4BVE{p;$B*S-}O#-A7QW9yD>LcJe3060}U!ZI^sgqq4siVugG%1m`w
zFe$QVFcX8ySQ?+29{0ETAdkAE5<@aUWFX*Qd^ya5qyS+EL8`scquSWN!c6EI0Po-?
zLNGIZ(ra4By%nR&?w{+~TjDh;KYEAC3{xk%!?qUG4?ftuCh=Ueb=@0b-+WjzxAxY7
z6T8}w=CwFs_WB%*ln6tHM?lQ$+1JP4>^Hk#S3(j*XqXymu)r2E_)*(UoE=umS-1(h
zW9g^kh+qRmBM%0(WJz*SiI7pN)Q?l`G!WHCXq5sE*)nk*23iYvDEX_xdiVLae^`uL
z%ghvFnMn6|2u_|@*xr@>Y*j+rsUT1kyeKkg+@OtX5|8za@7yd`q<#?}Sa|i`+|icn
zFHi8EdU9dZ<KycS4=}lqhagn|o25ZRcSm3^;7s-0GwayF?gP_}j5Q$gL$-u7UOP_(
zjaYF|1ThmB<3HmQl_7SgFL8Nbu~3*~cNs8_N)yL;M1|uJ3-TbKM7ET{F3DZ$gJ;1P
z2dD!RD@pU*nA!p~bN#A_?m;izkgLZZtow|61>{COUaCYjUAHx~EK<E!ZQkJ1+&RBb
z;<WJrt-<5d_I}j#*TSxwxX{F*5)ch_$Ehpa6#oC&+|uWoKiK1IeQUUZX`0HoK|wTN
zM<_tz#L(^#Mh$a61M~-@zg<-u*hG^^r>hzJ4tp7(bx<+bIHFooI&jS*@J0q|)(Tkz
zNLq?`Kqo1|q>b?m#ArY=#iE9}esDx{*CX{I!RvLJf)~XNLfz7kl>=<&*Q~&z#UQ`>
zEl02Jj+z>Ap>FKn6G7h}&y6m>Nx`dLkt$7e;YNs~0J<Y%QCU+Dr5I7hW``8v4w9x~
zC<Ez6djqCUz90|9!lrO=pYeeLn{FE5zla9aE{g*w<E!O3MGO-lZF&*uAQ(r-Xu#!8
zNwJge1&`|0z5ejDTRBbx5nQB;-^~Ukm{x4O+jTbhG>e#g=xE2cLF+TGAKm;HNv>#F
zv+7_+`||jTm(**Cr>-_Hjoz_wY|yO&K|~dy9}J`tvI?+i1h)iU&*zptE3-cjqwO=8
zI>2K2RI{VZY}!TDrJEZJs*9C7G;%12Cpge#>@bL;NR+3b2Niz=b;Tr9WTkXufwDAE
zn3-gBkblr^NmdLE>QkFK<0_LViD;b=F(C;T!eLH?$Q;&<Lw9N$vpx`g?{?e`es@#H
z4?(5gQ7^H=_Vx^CRJGrls9MvKc<af$cxy>sJhJ2#miSe&%e!({VaKij#6A)qVh1yw
z8ll6YI<K{S>SFopLYc{FvX{u|iW?CRw-`_-3-mD2vrY6#@=2m#nomW>5t!hF1T#&S
zxru+!2o?z^{3agP&scw#2xS5Tq=#{635Gb#52b~b(K7ka3tI|XkN7W<{3kMNYFerb
zeXF$zrm0(x-*hTQ6Q9npKv-ry+TP<gG~+8~7y>p5a8qS+)miUD=O%c2O>4iM8#l<b
zJlaw}rxmM5b_ESsZ!&3c@rQ@@+md~6#``RQg$P-Li<P;=X%3EApdACLK@plAC7C2}
z9IH*Hw|>b^He6*xG!7#hjy;W_7ln@|BZ_fAiATfH+6@JDn8~yXjx`Fn{c^MXVesJA
zn3{1f&pm1TM)MxrR^UA%Z}gxEL-TLv$5baY*C(%D*ycYZo)_;Q!|tNDjwQB+p1Gr}
z0t{6>(IV{j)schzZtNVMU8-eAKknx}cawDK<ig>KJq=cdizN6Z!X*0rz6)jcc^$<I
z3EW7iO_$A(0u`~7z0mBGD92ax7`PXG;}?zXtD5H1xML9t+$Q`7TH%kBn)1n>L%)zF
zmnjy(2DS9rS&<rWA%05o=|v4vWnA^f`u@QwIWb@LTXe1^{;Qbk(j~>0lRNYFF5Q^5
z?xx?XBj3iTy1I78*(DnfkDU$AAoYerOumrZ_D1}=8-vbhO^dv@j9Rv_KB{nEy2p({
z#2QdwZ1B4FrdotUZ<7)_%NbDB&5F*EKrE6p5z@l=Z^tUmUX|_joMA{x=*Fi@3gn>9
zl!HhYdU2XKQ&fYE$)TJ}Um|w%m34Bx)?WJi%Y$RwXI~!t)zbC5KNf*=4|dmk)pp;`
zbU#&lY4_Yy2gL3b|Ck(Y`Dw<^Bl$5^kF4nzoWR-1h4z>!+dCJLKgUKd4+de|dzOr@
zdL%Pr+W6I4Vtm2I<B>ssOV-^A8X$9{;mxb~!-M-(_W3oi$?%J%K(P>tgIi`UO(fB1
zR0$2Q<c1MJ#&9kvdICVjIoKvZT7a#U0w1`fuVo@&K~)@x`y-8g=)>~u%(-WlN1n*o
zu)K8aA=?LTd*;Ry2eRX>XT5`NWyV*RzTgNt8(KQ4Cibh+p<6Z-Y&_Bx_(;8dO}bUs
z#;@MIYE;3-KJDNh>!pu{b=+AJ8`!dX?st}x3sJ}D9kr*cWxoBvOa}w0gYfZ-yw;?#
zXBu`15P>YCK?OA)pn+|je5}FD$^{uLa%o|7Mupb|GfY43NcJ>@T9&y`rvirrN(6>G
z+9N1omWiYpmPx+M==vw21xiC5*0T}*0q4DUR<yG0_?p7zV8g4UD_TyZSCnRL4&HgW
z@w=HjW4dp=ymM<oT-BMOQEf}3gHIbXBHHR>J9F1}{+0R#mp$N)3*0b!choE6SMMzB
z>df6gBAQTg(@Cr`07rxeUghvH5y{fcSsbGiF`L8KfTspkmsG<ft_~ES4HIL4iYXCP
z*6Bm-7~Gvbol|2h36yEOa3PghAmUaq0fA226?9tMz`S3cu=CSl_{>D_{Ta(Mgmv_U
zTgG5tni4RfAL<;o`W45N-%LKeFie^daD4TuP2a>=si2`2!MBYrW^#?6cg7y+<W0Y!
zvsc6gpH;jMFh~Tx1Lapqc!#2FeDST=O)P9)kI9HqO{YGb6L(2AR8y*h3XC0CP-NKG
zf^Ly~13N%GOijkF28&?fW#~<{sgyMmbppZ;lN^m$+bhXy;HF%7%;N{CD_rh@HF8-i
z^zkUBl*zscs9Z?#H^iUVQh2q;VSV?M)HRSuFyzI{ma%2&;9|0r;6H|Gx6^Bq^P!Bn
zF;!>BIy{?$oZQ4QT?ak}zw~P@Zf?qdTSzLi*P;txDCo5p1ikCd6g1(BjY~-)DIKFV
zCMTHRVQR$;*g1k=-l9S=^mv{@kO-swRhE0<&9{HVjfnDIcy-jE(kwaXWR7>b=?ClC
zA@SA*CFS(eczMvyv_|v!oao7iHZ_GV->9LzJjLsyRPQ-s2b^o|VKG&wmqnNNERCyr
zWDKj{f?**m-s@m_P;=JY!iJvSA&OQY^!bG)Ci8mBr$rX)u3d<purC~$ID}fN{Xqn#
z^OkM`0OAG}c?|ljvJcNUGSP1w_BcY-NTmpzW1z4)BcKep!d%RPkG!+<)8jaY%X=a4
z?ASw>w+H!UZC<tOjd<(zb^nT+*1&Q`)SMYQcv@3@#rLnydDeV(`ua^*k5rBlm`bpY
zSKn)IS`=5~cQ8d7e<CML>sS`nb$i~kWpd5<%nt^g>64h&clDP?yG|}Zn4)trjMdK8
zXJ737?7ugCD&(Knr61)g_ER(zC2(C0NGpA?X+@8pCPKxOemGJKDh!w?8o3XoW{>9@
z{H#lb=L88wh%Z5iMC=LY5E&v$%MfI&jLiGBtge4X5QTzX2-LWW4QtHH_Wf{WU*<@7
zfyQMqJCY>-s->eUUOISk&AKlGA4!Uwe6l&?-0=%64*8kHh^NMNpI+7{<J0_Dv&9OR
zVQagGYTqjkb5BOt)!T_5t-sJWvoR-P$LU^4VqY{ikdyAP&JUODxAB4TGW<Fs;r<q1
zzIL_>Y)Bp$*pX(1j3?1|J5n^VvslG(lWo!~T-cIV9o8OTuBr-#P+dPTC{A~&W~c>X
z%rea!ih}3rzr3?^-$Q}w+?LAE&l^>hv@EJRC2K>U9B1T|oJ|d4P_ud^^SQD82Yo-|
z)lrr&{;>1Z(K#!gW&Yl)&cUz7O&OeUsby*Wpwi6wmvUk%rS_)R<AVFH{$Trt`lwUw
z=M!Vv?j#;8yObDs_zqla7p;;Iq+2`!_FMGh51`k&^W5yi7?bTqNFJpROv8dd%7l<`
z9^yg?g3BeOiOxb3oIfSl5gl^_s1s)F+o{(aUZ2lLk6<72F>Y9}5ucJ2-{hh_*pF^h
zaZWOYOU%b9%=NF`==$CxUg8Xey)3L|&o$VocS=3WEzR;u7@ynP@tWH2+M$}4R_D00
zM+bgCF>}-^;GY*~eQG@Q*4**wzFYI7ts9HO$}U?Ix;pcts!Nk1AD2q5aFd@m&mLXz
zh&cXQRQ329IhU9wa1D&klil$;*4~8hx?r8tV@sH_3iw=|sQ!h9w-FGazQ$}e$P!E#
z2@S;SlEZDI@Y#Bj``y<tWv3w@`}dK)GHYumt2AU9P01$gVAZYGHlODyY|YAtv{JBP
zP;nVRF(_dZhV0EBVnShW?D6;Ch^<K-d#E9*(toIYF{6!|c$;n(M3;AE#aC@u0z*wK
zba!B4E0wo#7H2(L)sr7<IlVBt>SzP<3t%_^W=G)fyDI38#bvE4=gbaN1Z+&T%@gEl
zG~kH&wPBOw>nPtdxvatdHWi6sTgO{tw>N}hP{BH<#71|P5Pclp3;GK?orJKIuC;MT
zqvt-*;$j;M`@eZV2Q_oFRq*Eln!*AkuF{)D&?`2SW}He{VVmG&Yl?y=u=={HmSR0{
zhx>4Qf)o_<^iEH_j{UoPQS$+db=@>5Rbkp}o5~Pm3A{kPB!khF6h2V65IY(r2vA5E
z42<%BO7m>SG>c_xJ7%L9yXRw)MB7y`f&hslszK#OU#^P=C$L}91<n`~Pb$CA0QvtG
zG^i(yv%O?EEtuJxxPCx?p~5kslmM=Ij{5t;lERI?b6XXK88)T3TsTjYPg**&s7{T`
zaCZja1P8(l#z|AxCaGv5AG3sNL%8v&wuU44t`|as-T*phhnm6XKx61`KnZKYvL*PZ
z>YsJSGl=Oo;MTuU4o+8gWc<LSUxSV_PFuhy9QS_;I74x#%{M6qY^A~>0}DV}UenpZ
zeW5?td2ntZ>Y=0r+x(=d{W$8_P-<(@ENZ{m#`qKxHu9YzG`~d1iExqh4kLgLBqOsB
zF#9(Tc{u!gKI_d4s;4<!COjX1`-%l}jw5mV&oq7dVzSh=tDGwRRHO!-OnunY3^WD4
zI#Cm<8YpZB)9x^vG$Fya!z{i>Y7WbbG_gZsLo%5i!gDD~Q7A_MGE@e44s4fsuvI-8
z;A!P~C124C5{Cjj`10#_=qf?n{<BS=8d!cQ*YxkmGkD(N4<CxIkh63}WBdTaUV7+<
z0l|Uf{{}Xbar7_P5&aU<A_0nItS=^TI)wBy`}6X)|J272cpv@XZ_q{a2cRP)9(?=d
zoyyw?{GVDCi_mG>O6E*HjLe}%w}TSGi;(RBarkuH6N`;5U{2P=lVN12BzCFl=td-a
zh$M5Eff<k#R}7jR*F*75A!K#b1*yT2|2ypgzL?Czr<kzd<9x^!W%%zf{zX^Ns|S$k
z75b+;A4csrqh10L7c3g-ZM(2wP=?51U3Q~NjZ|7WC}u)qcw!FZ=0*WWLb77=wae!%
zG||z}NTy;wvr$k*kQ|`EI||Y%Aw;pN4qtF#`67M1@|ap2_Rw`^Tuu$$zr_E=j8omR
zbpq`eEw*(Rd)Jw<KYooSQ9%NNVlEeBGqY;%CBnvWFAtD8d-!hrkI}I|gc%ImS_KVH
z$l#!bF`YZz!%{R3n}>QCWJFsa%YyVh<3_517pm>WnVXBa`P~6+do4`$#U5~>_iYx?
zk7S^a3sj+Cl?C^etqDfwf^U|2240;I$#3l3!tUG4qJ!6WmgUTR(%rF)t<KqV@;r;F
zNzRQk-=5!RZd%8hoiT2?Ds?O>v}d=nxEkZqc;-K<qX+&x+37AFgQw*nUn5z{@8qMR
zl5i)WYFFGI5c2PTe}CL9zgL<h=&K3T0N{IsRwNQ;gZCC}4=a-7BA%q_*H1*T6r}iL
zK)jO|Pakl#4r6dyIy+44Qt<QL3c`%q1de;K1XB?Tn{gcSgHNh_uDa6W6lEM^i5uo`
zi_M?qbG3N^;({~TI|fu-k^>)ka_$1Zp*azk>b)wC<kXao+nPHha>tY#L(elWo4aT3
zvj`;(QP!3<13ohKJ)(8wMh9=qTD7OWJ8Q$F?dL3Bh5lKG&NX&toI09cf8^S7pOamQ
zr&{TD0P?!MWVg3`n!<f=+TYgn<FHvuTN&t^f)C6Y0z#mh;VjgL`HbU2t*cF*R46Oc
zoi%ScxSJgs1FcS*C<`B4Ql#kyMAw@8!@_9w8OkDIn~dZ;{YdqdUy~iPE4%(kXZ3`k
za|0XhWykE-Dw%EL$76$jHb+*jZz>v;n)%m4<D;i@_oW?N^3FKs6>$|*S3DAGh^jf8
z8MUObCpY54Uu_p`-yF2G<;83-b<chFZT7e6bEZ7jI(KBv=~rJK*SAT_wsqWc0<#C8
zlWl}S_Ai-8`2{+TunMo-0CJHq0PfNPgOE^(VgU;*vP-^%1PNz_MwH<)Q+%UI&vjvA
zxzT`Pc94oql6cC4r4Skd=As)JV%DbZ&v@PAn{{7dKD{Wi+JDqHVeKso!y0>D9XYu~
z4tyyyYJo97qO!2@_M#YTYM)iM105@-zRnGimHsabUQm83p9L=3m-*fg&Gm-@7Domr
zFORg&P2T@x)VSsg`66m^$?n8)jc+yVS-VVWT5O2i_YwyN4?uUN^3#34uhLBU>!#9S
zp9I+n|A+crFhxets6f%{1WF}|2k>$5jG7B&U9yJ7P5H~KaH@);)M5lf97Bu8C9IbG
z6Ru3)oRPCbs1suO^Mba^9{;Xejy)pCZ>%G@uy5w(8vm^Rm8{QzSGHeD9JO%$i4Usw
zzOZFeSM&TCYY*JckJ`am%kucZM_$M|*yZ-FJXm?;L|J~sg+pPN?ktP6zQk_Mi8Z$t
zl=S`ToA}3CM8_v7+qcagpV{d5;((8U#RJgbI6;t3tDOIKiDrg3s^!daju}hrDg?ze
zO#RYL7LbVI4h{owHBbylxj;vekvWr?pU@=p9cE}yeYnXfg~%)gV>#AS;bc{FfQfQs
zFn!G|N}R(hk5pHdskkXaUrf@s6?A#UwC*y*1v76o@x!Re+t+xVYh4p{W_k39oWkxq
zxsjEPtYuj|a!LczQh2sDdaQX_<dhX>r>72`J7Sb&chtC(Hw%XJ{p!-e%FnDDBfpP-
zGW62KlM4+o*3#LyM=$6?@;Iz24b`qas$r^tE*9+XH__<Qu>J!bw!pI}n8OK2g9^_~
zdPySmff}<itZE6rGFA-18iwa1;uD*&xMIhtUHIHzQ30r$77XgH*ciK;`CeA08){9d
zxL*TR8^ST($&V|)Q^2s)n-_jRUc-|H%FTBQLRXZI-MM}L*ezOEdE1isnqogoT=j&0
zSFaU>ZOXNN8Tj?WrFx<XL!t)^v|<6kgA8ui&136IP=zQ@H(MNKwOFfF45}R{j0poR
z7`AX<L2iaXg>Jeb@aaC>YoDWEvW@SNO+8L04}l%Jup&&`P2~Qn0VIiXy&?uf55iNq
z@~7(jX7u$Nvfv%<L>Riqy>KHijNV)&Dehr3Yrm7PdXMVLIF>)X{Ejo-llZ0<4+$6@
zdt;b=SpG0PFa-*OjI1VrCB&fv<RZetk82c%D6AHDG9Gpo`EqGCqCp`yB50iip~&Vk
zD4XanEEJt!1AvYsW;$a)b%Ik^j=VeVt9^tU2>1n%jp2I4(u=+43~1^Fuu(BEfp5?e
zsDmJ$%llheK1xx5@+6kv*bHmHE@ucIE`eIgav=@tG98n~4TMrL4KcL4zdmh&Molkb
zz7yp!Dzogu+K+>!pa%n@KXicDi!{%P%$@^d-{t)olXa)a(eFd}1!+B&dw85_!t<GS
z_;`?Z;U+LS8pGvxo*Vt`Fm|x(gH~v!<N$Uu8NRL>@XJiG^qUMdF(KpQV`9bzX?9X*
zsN0~yI5?i<<-ddtcRI$Pr#tSR^umeYb2xaZU;votD#T7JC%VGDRK<oZ&zBEP`>KJH
z5&VYV$S8Vvi($}%tNuy`sDsl{W)bv>qH)5`4yOL;>Oad5NY=kW!#UPAZ3z^5;yCUH
zK6}cD0){}#o%pT+iS$rV4@JZ}M7mMvDu;<^N&@KwHo4H{R4D*0gY(9mms##+ewLM5
zizB%D7eJm1hNOSzo_qG5^Qg=(4GA&3s&f)-j{SC+6Uw2P9EfQ>{b5ZE_jKa<D%kCb
z<kF@EA+4}W#g`2Wvj}L+M6HZjrW(p+i==RL%(0v)<V`79vJeg6seUgb&=c?~HY4Z$
zlNVpeT{wo0bN!@eR!jR-baRAC7Gc)({|#=Pi^SpY%MWs>Mj`cu+t5>o07<YVJn~SP
z4vz=@oX(sHY~71oc~`c$XiEZGS&X>L?FK9y7^h>&896|;fwUN2lRDDvg>0A9g2hk!
z%d{`|&{>>%2#`M3j>wIGHe?KK$<Bx8{^tkM487NPM#o+cBcuhal&PdD(0dhk=FEpX
zA$KUVuKlQ#qJF4ghnbPfVVK=QbHO<Y*J;d`!#?$Iq)~)*5u0cuSAGUsq2d%1^g@6e
zxFH3exHjR^oljd%ovi0Mv#@zNpB2ZIo1Zz{lQV1F1mAze)s%+Xx`bPaM-Csct_kJ_
z2%{GQ^gxD=ow6-0AHKw(E_H$qBj0LlfUd}5Aa`R48^}mJjN_Dx+Utr3tl6DO7PQbG
z%m|vmeFG@L#0~2S30NiozxAmIE)>OzhY+m<y)faX*9Tx%yjKN&dh`$1BfovVa^CdH
z1Xq6b>2lec>YWqT>NRd{!Nt7iv%90~=e1WqdU`>0)tSt>wa0H|RCL{25?6in<h^{P
zt^8+GZ+iV6-a#SkR@Zm=o&OVic9deu@861r39do&O)3;7UDe2!(j{n&Yq$wlkOq)C
zhLk`s7B8TwyAFXT5>Z}dFJcR9LlB5v&=}X@80`}$mHKA`*z_wI(HCUq@*;vxE`2`l
z?{99-oL*V<@9Sxo?aVsfw`a?Zplz4eo%x}#An40)_dAAW2W<;l{CxgjyXw!(s;jyg
zT)OQE=b+|AJ;TmeOA-q&uM8M1sK?&aK7GO;P>0?8`}??7`HNwP@x>Qa=O`cnV8o!)
zx>G|LQ;K<*A%rR6M@<b9FS747v31jphb2|8yLN_XizyI0L%|HkwIGD{6cE5n$YzG`
zI8q%{2A)u>-+<SqGkK5uZ_UbEI@I;#fj@oKU312OB8JNUm^8xLGD`a{@X@t*w{&j4
zQrQwwyK?A}`?G%z`*GXNklL>Mk^lYbnpORJCR8r&Gho-J*Y{i=sdo$)MquLlo14S|
z3X5Ci6LnjYu`k6l++fd*?w)04luLwS35S<<F>@^smp7N-!E8<rPe=O7;<@|~XB5CZ
zw0ek-02u-r<Nh-3+#cOykh|pinckPY=!+}vH+^X+ShqZ8N5YRUjm<fShe1udyZ+s-
zqq9;|y;p_xG>=`B{p5f3WW?0InZN8WZ57|&9OJ$7`V+6N+0yyM)cbdL$3OA>$1g0r
z_sOemde0sT;jzG646`(5?H=@foM!sGPQbQoXe7xcW*?R?iOa_oOCj4^E|(dMhoNsp
z!dGFY1RYg_ruz?g_`5JVbOm`kbz12QRM8^zcXWrcAoA5q7cx$VNPQv7GWraV{NMS*
z;DE(()ulBH<6hXj-S5WSOU!&f>q12JyK8SnPsxeU{vPyRb9Kv-efM{N+`M$=g65x>
zw6DqOcxURGE#JN~^~B3NznUFi?O(Ifc!g6Uq_<~-P9gZ&OJSY7zoC{7Ns|XA(QV<7
zj3%rhzU4o(EPr_NkqOCD`@1at6_6?5hhbspj4J%00|3%K9h{!cQuypr^#jxnV`*HO
z+3Twvxz#~}sH0$t-ZepA02zgohdOH3pU61cRo{QghuzL}ui)3#Z^dlA>h&1wQ%4*A
z7Tlb<s_@dir4dthUhep3aOvACN963^zH;i{Cl7xk_QW6F%wDvC|IU5tOlK^-ecQ8(
zohhGdkeJZ@p2CDQrhaE@KY#vKyWCn7mds^~b2>6N3BX0eAJey1lZpaYto0#r!BAuj
zp?K(}<9gUkq*HC>-OR7EUeFj?5kST|n~~&3gsG1Sc(F~uUg-p>s~C^%p--laqY3`A
z+n0y6pP4>8dw4;~QmW=6$E164n7*!!NKT@QK4JSVA6zK^T$+^bR5X_?)1w-F5*)%X
zftSr1qJ(uJOgYS_q8xZ4cE$tI;c7V=k++bGBm)H+*c@4606u~tPF8(6x`m}dD-{Di
z651;K2|3Jj4NTP!F^^KBA$r0Qj8x!|OOGHt4JcmI`PADU*>{kS_BIYS&aFIKiCD)_
zj)(-#Z@3#X#87O$KjKu91S%VAnRyp-i2YbtLK4HKF`>fIVT5j+Xi!P3Oo}~(g>NPn
z>YP__7zjTA4o&rN|3(rB^Z<e~{lf#4vHM!j(e$3~RXfwcULoPI+jEM_+X&?|I|57x
zU$j#|A3jNX#i_9plhF?OMgCXA*2Ruhrxx$R!+tFNoEU}6Co-QgSP(<@o|K1rKtVrT
z8McrL`V3}BI!(kC0u=|0+~oqF^Zq&e0X3);(a*xS6OCA=6Plpl<hKxBrj>>L`cQAQ
z6Si^D(~F7G6AjL^00<rP1yd-DT*&-xT3V9HjK?EL9qWYMpw>CYq&X(Rz$puB4=J2z
zme8YF^J>LPr3x;F14oT+qS5_{uaG53LiW;zIWeN18=naxj6fx90VWK;c!N4XYc+j)
zrG6X-Jicb28$q0hNc{w}2x8Z}n`cz)!T@rVPa9*L<TB#9S;;CAXSkcx^?_@1T7r}i
zuGY?1jF^7!a{?bW$;pbaDkHb8vj<RD3Wp%@X4g|s3PT103WIkYG^m`B8#BJz@m$S4
zo&9uJpG1a{^1{nQuU$YIb3<KG+|&S|QAZ|D1bF}$129L0>bxJ5o#n8Mtvfm&)w@o%
z2Zh>36;JwsBFq^+a9ol>Nyc+UX-+zjE^_G@3QSQX8OBKUdFX|7gTRbG;+bcwH67s?
z`rtADQq%jJ6?6<ib}uhA24@9~Jyp)V{(^c$wF=Y&Zk^i033)Kqy;h%T`#wSp-+ci0
zHNp>KsRX!!0S7zw4p*H;W>b<z7Xf3S;!$LL@#TtK<R2-ht`m$g`j`-nJ4&3;2OYRh
zhJwAINC9~5`Yp!j3)9>(xrpk}fi86mb$S4>P{ejPkEs~fW$}>LcC;akp&GTwkRJxW
zTn%r_t>JVnOcT$yL`DN0K;|AE$}Sa(27=|2wF)x&Ttl*=2n(MMaX_Q1Fw5Y?6O7cv
zcG-xD)kLP-;Xh{UBG6kgG{P<5Zv^B-Ox87oTo8_vcTy0#!4Z<)0BDObRjGIcI<N9?
zJ1o8f!wbnPPBPlXRNvVY=-K6Xmb7NG_4M0duG9U%dJJT2gbaEm7k3vQ_98mK0gRF0
z4uclz+hkSr4_3>ZCLTY*1m6>Mwii^gJe|27!3ocCa09xHKk|X7$N`r6A<wYnaaVHZ
zHY;35`e}5*hogieA*l%I6%!s|=3QrBv+7*eRI@=v4X)FNyHo|pVdV#10-rQT9lh^$
zaaF>IiG?f*476cX5i$gl_vi>E7w}A#G8Vt^wF~)|m`d05f=c|vg`f-|^jD}EkR;t;
zsvQl421h~s#c~ZDB$hDt^n&cnFTC2OzRPi)h1ndYflS7>94-`RopnOeaD_Y20a(as
zashu0KTd+0NL|sE9nGwUq97Vi?!$M94iaekD?F&VSjDjZ-#2!Fs9)eg@J&asc8ynX
zhH&(RL}bdO$fOk11Qts`i~z!&Ux>Tbx&|!fQAWpz(7{q(nB>@lA9R7YBV?V9LPt@N
zhCT2uB&%>9n}2(Y31c~iZTvBYhc9Br0}p__7JUc6HK1AIDM_9F6L5a7ywG#}f4pLf
zM7pP6hj)z{ILg3Z6g+V72Ol}c9y-hnprdn}KCY~T4!aGZ#A#s7IB?aNDNJ6cXfivZ
zno5Wh0HYIDA~#@38Q1!!zlE9-qth&kSdoAY*wG_6<Uv}6QlKLnY4bmB>?$vzbN?II
zI2k#eQ9K$;0}=QReOD+TK{Zw~o%L`zFFW&o+KZMCif|WMZ{}k1I($Gk36d@k$P-v<
za;s)x<t`O;DjW?~wH=@f(p5Zafq^_dKdWgB9u71K<Tw!^sc}VSuid-<J8PE^XzUOY
z9_L3j&*;8aH^^||_#e<oJR<<|6lwI1>2Ou_ovc+iz_bZzT`4dwl-V;un?GPRkiUMJ
zJe>_pG6W<z@ZFH6Or#5ekp8y0wI7H_?j5)R_j0Jbx^^Od3Bq7z0~^o;M=7i`W_+{z
z>#SYHq><4zEiV*VQK*PDt`E$3^i$XErE%7E%)7{ZFXQyTlJji;_*1>6XU1XdHqvhb
z<csJH1s23_C<JZ#rz0Nkrw=l|l{2P&KDp?NKfnx<s4hS=?=g+zGS>S9VIRI38!F@w
z3j#vnq-!4nhTbvIB_>efq3Og?1sN4@c(eu$g`l->uWxq$g9rhGTb4%!rM&vGPj}PD
zrd8*}&qGE$!J8LEKU035^~o;3mml$T$LfsQGb^hzuG}8&CM$)VLIC=_PPTXi1K(pK
zdSFm^1GvJBpKoST$1ZwQ0Osu=3FGE>X(^~xL%}-q|1f$A#bO-Mj|f~g%Ed@q;7TX~
z16d}43P#c;{8+?Z8Do%cs)wPEXatAkD8Wc**AtU80j61e?84VC<j)W|9Z~6*HEB=9
z%Af(^_eX}>s+ZOu>U@G<TM#vDl*Te+_aClbRr%$W?3aBiRtChrpEw7XP`;KUE7qH(
z;y8qZpb3TQ5MTm*^&VWI?Vi`QESW`>-^`D$c?lYhCu7~%HgNIOkgaJP>Ir!Tj7gtl
zb710K2KJ)hTNwtV^%{(UFk(;zql5UJ3I<4+(B0ui)9{)(WdYox;vP;g#0HJi-4;5)
zA##g)W)BZJ*6EY_)_;<H>Z3W?vz|PDI&ap3b9qCPIxD9{$O~pPp05#o63=yazY+Gu
z()t0(QO^|LeB(KELO^2Tpb9squeu8Wn(kg7>*u|-V}5E=&bg~%(8Y?c4S0CzYz<r{
z0{^%*DeOBuuRtT7nD&@hKReQGSgS(-4?nCLWW~-ygk7<$GjO2GQ*Z_z47RUm#`1t9
zu7ot2O#KDZsbodi-)nXw2E{rE<d3N9p?#f7hHLT)d~Z%>d{t^zYh>l4H+Fy9T)l1c
zYctk<JLk2CYTxOz@x<cK2AnIo_0fh?)wx5W%DUz~FNh?784618<fQ=|f0WGcjfz_G
zrZ1b26?OC~TQSqSyme`x#Hk2UEbOCiPwe<MX;^-L0WSa@)dNly*8)92=wWb94#Du}
zB8e~!Rooa^>0^pSG1OFfof%VmfW$eRoq|3HU@(Z@%x13(@ONLIO17&Y%{@rc+A~h=
zdNcdA=xP*e+uZkeUq3VT#vfE^d)H^?xv@{@EGfTKWy+tqd+<WN8>pwF$m6acK^dP^
z&v6Ff;Vx^p<^`pGhwWNy(8bF6GxjuKhZA9!Rx=~-^MuJKYCU=V!!}-KaO-g|kko<8
zRDh!5H!V1f52fh{Ls3d$gtGX*fl41U)e0!&k8uTPl9-)nlb}RX4}&^UY{TfY=aXsQ
zWvA2Z<=T??N>mA)I=OmPVQa_onBehEI|HxQ9~!k#^vPM<eLic{ApgusH}g^(Z_XP<
z80dxvsnKTu!fIV;Zl($P&R9zxp4Xh86J43m@l5&c<ueu>AmV_LVNmE7J38!lEgv*G
zjVauWL^O!9y_)8LkHMqBSidAl>Hv)p26#R=N1B~#_$y3?hPA=A6~wQC25+`@qJA@d
z;tLhmbcC=^nfvGOI8w7MG*n?;NB!rqJBgPRTi2Q`=aMr{`Hu<t7q`jBdlJvFkn?VR
z?#rkLw;inX0LcpX#`uN^9`^yv8hSQ$QqRsk9UaTV#1+)n-mW}ac0nH+(%;zTm-{Wh
zYk83~QxX+})G8h)py7E5RB)?FbDZJ58<Kn`>q?f+Xe<ZZHB&LaYF5c%uu$n0?zvV!
zNiVXD8IeASodO5^ve+Uey6Ta!2NGl2FU*VEAz}g_{kS0N>GE!ly;gc~fP^ggMGl{~
zp7eG^yirL$G^!KG_v7lh(TSpVx!gK|6U}(`(T$moFWb`W3|e79ppVPVWDpIAViNF>
zh;<U0g$)$Mo(BML!}_<$Md==m^L26=o5XP*0)$k6lOBDLx6lQ%@y`Q8RVEKQ+eakE
z{C(wa;lF*^`#&_Vt}C*R@7!|w_DjXoGw$QjSq@Y04V~yXQ2}>r@Yp!o>GIJ>HYPRA
zh`G4cayP&kX2RCeFjhkCjTSucsfeXv6C;9Ol@4K1CXLV&a2a@`7{aR!Bx{nhKf_)+
z>Kw8k&h^H4(cpO*b6^<$Y0rO5{cg5^&Tc>NOf0livWS{RufAb(YlF@P)J1whm}qWb
zf>;lxBETf;cmTvKm;pH`xD{r=AEG1RfJa8|`fGJw+MKNpe74g8=#D^NE>y&nc|4PT
z6fbb$`X%&{1SbmtO9!(xyx>t%KmvC_6C4siTZr%W{KpDB@0YRF8B>+0szskJXHD7f
zZK~>8?tSQJ)%m>m$Ijr|WQhaoBJ>CVnh$`yrcMts#B?_!aKj&RKe~R=GqcjhHuOy6
zq%Z)B(C-cmRA3^21u?^7)YpOajjlM(@N;M!u1gS>xSe3)u#4D7^L;oeh6xM?c%$T3
zx#OpeYjI54ex613|Lnx<jKFJu-hORI-xH4Q+tU5UCUTqv?GHfw@)3t*@-V`}9D-}$
z3gJcyfM+id&%XfA=boDf>!%x%-$C$(cU51K1>05ER9sIpQ>~mSuq^C85oqM<mw;wW
z(O6O~)q+gNERB=e$YvY{RUl1O8XU~8@}0yLT>=Mjzm0BIzpB-iY|8WAd9)i`QXb$9
zIZ7VxQ9}x8Fxl4Nv^is3GdU2_2s(p#Jzk}j8PhIq9XM<fZq_>;(uG0yhPu>_LFJzT
zL_<$GXut)e4um+hAjQ^OBHg5@I1BD+*!P7{I|PYt$(U~XF7qGP>o}c3OJWs)$&=Ef
zk1oA8`Xoas2#iRD>BnhfCL5w3(;0z>=fXpFBGjKOdVA9Isd@_omQq=N9OoL!wXwLX
zAJt`mr5_r&WDPSW;6NTtw%|PNqSj@FrdUQW1K0n)UPoG^*!+`h{$n>=+^St#m?XR7
z@%uw=*q54?TyCeDBo($tkCUXkDCh>jF_GdJ%*<4Mg}FS;07nff(-1wzEXkTV&#P^c
z$Nu!_3%qinA#JsVUO>BG3CJ$eHy!lNaG0x#VwKbh5~lecJ*`whr@d?^%sGE**N&e`
z=i?|#91;Bl4}3{ISrGSl-!*l?8-3f`YSypWdHLM9)<J$J>IS7IjVjYt&$~T8u738X
zscQxtJ<zda0yeQWcP)>p#uGhzvSY089$sNf)$m&wJ^8i{Ysu%EEyc7%N(3fAf<i-7
zD~!S-Cs|a+Cz(TkA<=S|YEg+1y(Y{cO_}wAfu0n>K?JE^QIUXHnK<x+-K63t)g;LW
zd@#85r%Qr+xQ1k*!bJ-X`N&?o>s!Bx*C0Bn?~8|9swccKc#n2+akTYf%;BK{x0gjd
zxwU1HPh)UlRrlwcw?EQ}t&d)te|*Weq^EM-hovtZD(q@`ZTyM3k=-M9G-n?7iu|u<
zhpr0Ly622CcRBmkZ?49ZOn3Bs_h3}@l)&!C!C#<Fl2r{@Rh32Aplxd#ZdXjmTy@D1
zv322s8QzB+OVgcbkh5)I?zQRZf8+Q}pvxH9dPZ8u1ZY(iR8@FRXqtYMShBg%2*n&o
z^B8^fkr^jvh#P}5Bus8JVmrildwQNLogaqdiT$RW`Y5jI(bJ1YRBjxb+q7uH*wB`m
zi?wa17fzt<SAiS9|3{y}39S~dqPFh5$e_{}OLn()ewv!uaVM#H@%R^yL_FQ*6>|Ss
zt!Y90JMnK8F?RB^FKb<M{uuXkPQ_=DH5V)PonFXsBim#c3rb5QM<WJ1CGMU0SG0ST
z{(a_MDf3$U^_%(g!+#pF`>Rm2^p3$$KMz0B37e)XS*`x2lAG7eh=x!a{K|Y&2uvj{
zS%a;xV)zM6Rwcg9WI`S<T##pRflVAIsTT;GK>w;M_j}~F2fCqAeKU`Au1L<<wD0)R
z*x-w%+*$LY>vLNcji}!8QG0Mgd24|ZwSUWW7Cn5lSI~+y{}--TG~RyQ=g~kS{MaXD
zi6MWTj`GF{nF068FYvzaUXI;9a7V_m8*z_siTqb|<x$KUG_%p-EP`F5j*SF1X7!}i
zTP|nF-;Ex^td9onv3>Q%?mGhLtstSNO-NsdxJ;KUC~8ogVYD9}|NFrY>CO&Q3MpNg
z>cB)bUH|bO(ip$l%%oQy3m3i-K*jaM5iR3#D2+xLKjTyH!R6m`so-104r6A#b@<;O
zPxU!)a?y;{2bP2^kEm?!)ZR$D99Gs^U^`Xt<h<^h#^_B2^{ZlPS3Z{VdR$=Xlh;r8
zUG!1J(|_&v;m|0#XXMbF@SjrGoPF;=JnQ@JSEHYZU+|Gf_WKWQx^#SHWGE?TZx{|l
zol^gbZDH>7*-?+5XY#+VcLhXlUGaeaKIDqiTR|+Gm|*08k-<qu=^_adnhnFnhBY(P
zTIYlTaz?t4h~wq`?5K8OHW)cJMBrKgCX)1qfe+tp!-@>=ANh|J_hj(Ka_lqb(&);T
zv8zrsoLo3!|EoP25m#?77!kB_&8njB;-8#*x%FPw=x3DHeAfBtBQL!A?xmhN@s$(2
zqhfme#&w<kWPIkA(uYsZt@|?ONaXv)J+q$^3{K<9-06!a_%?RFKK|YA&PC6s-AxuI
z4NqXewpfS&C?k+g&Dwem&om$Tb109i*%t9gvSEhLr#O@By%r8V6yA<drpC6}P+5to
z_@?LSsO-RuFCK@F4*-M84kjBhJHgPHK+K>F99>KUiQXJv(`75S#9sE0&ithXU3Y3q
z9M+OYPss%_HKp^SE%!1u{bQlO_m(~W>=yGazvo;UAGon|ZpDWkIpGT<8Xh0eSzV~r
z?>fD3{NKKge{5Yw8jq@)a(?-&p;5;c&g9o#j|lFYb-`#$KCB2&{9?rI5s3AF>|%^|
zt73{{X>7&1`OnW=_;!8H+Ee*KM22*2M&QMYSGqnpCpiSrR+tdL91^3SEQLvXK<j<S
zn?7REqjzA_So45CP?$!80*s$XhL#|&tb##C2TOc^_kUlf)VAdBJ?dU{GJe8m%I1vA
z-{qO>4>gV+y5;DEG4U+1)f>BJ<wklQky9QRI(4Fq0OT!XF*Ix^NElmIbHQpVh+_qG
zFCq<?K|3**k<5?U_T8zvc&_VSO+weV3w_!;^5aPrVaA9kre?I&5s?5WBQ%%1&qJUo
z0yU+--V~S2SxSH4fFX>+S+wpQkkQ2Gq;NY7o=fukc*os$CNr#jn_HGf?J#|q@yhtj
zt?iW;gO>aH`#f6PdU`>KqdNJG(bkeoTgMd^(t&-5etjC}_1Rf_@@C@NOYKX&$FzLh
z-?C;^^s|8}841T-qs^V4VdHB3>^e*<*WyQ%7?wRS!4d2*CNnm!@LAp;RF#IcY4V`b
zv56HOUHHXMxaqZTF3bHpxhqfs)_c^F8$tw`ErJh{B)Wnhg0#pS5gwA57{+8xzXG2H
zS+<5(G()mWC1UZM?L3Bz?mx1=!U@?ca9?w!+!<e=+q7f+jY-X~*&2gUP`p;#(-?d*
zbELI!{iUqfn#P#@t3E2Zhf3>IS8VwcBfeRRs+#2cO+Uu$ihTOG*gdMPYC`6uD-|2Y
zZm}JBtGOOyTQD|CGa11)H*uhEW)@1br^}nds#jE>&K#|@O5^J@Hg`o<Z(QKUae-%7
zF3Y|CY)xr;E9jxP4`3&~Bzf7m9vVWAHGHQ(UB2|Ly3ig?dI4dM@aU_j87uLbnSgxb
zA1|!L2*K15H7(xfV$2<tXRAK;fAX(4f=U;3G#~C>78%^KwmEp+t0S#PHxvXrO@1%>
zomdrfwI?&y>)?c;gEt-cuM-!ZkGS_`|E{>o{+7q4<mcJeF5f+OPwVkdGKx0d@GAUp
z_1rWt*@r8*4fe5_j56geG@-7ZcPDXL*J`gWw;`sc`Nhj2U6D_IP=*z6wgOMZU2!JZ
zSC7tN)zDqbBx9ON3~)lR;ex3zr8m?ET^bGu89!e&LfMPC3_~ZF!vNgS5ya`gfKhDq
z<XMSsiACa(9;+H-cK;aO6Ca%6b8x}&`niGcMLje3%8Iboz|<Qb7&@z7N@5+Kr*_S#
z4|&|bPtN6rmZg#J@cY^mNm1|TbcM3Vwq^4_ebHJqrAs^hNqXRv%WX{yV^(o!b112t
zv_NJC)nrHX1kq>Cs6QY6sMOG1#&Tw4wT@qO_1HV($-54^cmuKZK-B8#S_!3uH4;G8
z1=2#CK7$bg(iacbpAc~QkI*QnqH&0gK}#n#JCWpjfp!k!G5(l#?Knb4%((01qPU@<
zca}v}X2iUg9rR_t!-JAPTfcT+`^WvON}u|=PsT^xgC6}jE54dX@89Cco;5Tv>_B_Q
zfk;cs$|#@K)$Vsot_W$vvih>wFE2Q{Og;9=X2a@q>*)p1K~gym=eP%aHxX8h4Rtl;
zjv8vX_BskU{(U{WAf{i=4m@B1w!HQJ{fmJ~rP(D2D5xbwrO_M5$OAc3i6^KfSEW-6
zi2_K}H}FUmOlVXC=0~{1U<@g9>nA--q6vu1w;DM}zWWYEYaqEGWd6)?SMmFCTOzdM
zxwh`+&r&M;zIZvly!o>&wa2Y3O}SlFFAdfFpQ^f5k-BEiGf{g^c4tIr*OuZLSPS;;
z9(74bYnSzlu9}iPdzMyq?2|3)zN}rp`kIVIZDukvB*NiV>>=wo1K7&$tayI&pYC2Y
zmEF#bvfnE2x8(8jB;$T>zMk8Xc9ZH#b<|03DGiBJH0z=O8vg{TE2fV$_ojTH!G;Dm
zP|?F#8p>(}gdH%Ba#=_@qa`$i>xRKtv8vDt!pacAScvg(gKe_9rVjJw+J%WH?Y5M>
zxbmJ!oi(ZL?2WMGg82PjwcERv#s#-z?c9tdl!V4x%sVP}%e1Yz6&rj`Hhto^TU{Rg
zIEr4-znw!Q8noF>0|n5pI=*=16LIJ9D700Z+(*Z)+`Y5#(Y?7iF<-g{>&-c1eG92q
z{g-ZVm5LKY4YXHZj-bJm8`FyzWDT>xVZDVq4D!`mdmJ2n0W&b~C%Fquq_CHb#oq*p
zbplSk@An<|<|lu;@zl53^@q+~YH{YqY%Q=3i1=#d(7Kw!_8w;*)!Z^=YJNz%ongSm
z6Vo)XHpw0C|5O)Mr_k?nLVm1}s}EhLAsRmr=fGnKonx>eQ{pu0jJlSFrw40T{VJ0Q
z835(a;e7?JwnNul<dk5mkPh6jM~lUGv`JE@jjNPpaG<s07?#thod?XtDwaOKzH!BX
z?`2J&^j&@UEper0TjZjM%JrQhTYBZI$fw;spZ9;}RZNV&vsLwEzq}whJ#ez+R>qaA
znb;0<sWUrlUpxM;E#o0S#?}vx{CTALc3$d@1ICa1@VJ!}Oo=yU?W{O?GuL30UU6{5
zDmtQsK;IKN=hk;Eja4ljnKM_Hw||~-?y3gWf)wA`u%bM_2&Se-#LNKl$s+N>;gaBt
zR2p+J6Q=o}st-h@s`g=Mp(`Z}5gK3`C>I4Jn7Dq_^CEsj^ZoA|n*W$OcI(i2CwDv=
zrZra{tsZ;mbZ*r9?)rXzT+k=u<9{U-U&UsZb$*LZp3m}5yj&hsI+QJpu{6IH8Pxnj
zn^)xzqtxclTbeRH%4nUMQPJLmJ@uty2Ui@pmJ#vc$wffTrRYo2y(mZ}Ez6#@`hB-G
zSM{<rbQDBa_HEiMJd%Dr_xi~Ea1Av$c%}-CMLJ+!!x4Zv1jOzFA2{dW2lNr@LrTE7
z)9H(rROP8ccheHEH3A=WW5Mj-16UvOU-sHJD>Yo=(|2fY^XpO7sbPEGTyV9kvafgc
zs-pX&g9ZfkyyRW<%+kP(fBQ$D+}q1jiw!Z)M5QD~-c9+)|An@<)<2%S%JfgD#tS`O
zv3H&gjoN=`PsyeJE?2ardv<)~(e;m~D$$Rh9jamK=kX4ndz)T~3wlY4dEe_hO|~3g
z95uP&#2-{P91CD8X6?XWYhBlX<OTh*2HKN?b66ax!A4}J;xp`q072RVi07U*6#W8(
zh9shl-T2Iokr&Ki$W>8>m~@-eCNX63OeMyvum`_*bbEee(9y=8&(>$UPk-E+6LK$M
zd-vz9{!fm&Qtnmx<H#X1Zfve-X*#|-v}bWhhUr#hP~XhktK^?zhbG4UJo>TZWh09$
z?S7qJmm7NKgdOO}iw<h;tP@bTUqM*aG5q@3{m#QTvFKTN_=}m%_2<`_5*&prN7w9v
zVH&&{<<4|uA<S@^`3M{!slmAdR189P`as9zM)Kh}sK0_d_-@YdiyrhHk(oqKEmz-v
zd*vfp9)bfK+LxwgZaujyHFHb951STxo0mmAeZ2YO*6fgbDOvIRC*r|kUYA?Fj-Fcj
z<h-7O)VIX$_U&&!UTUB5!jB>IXRK~7yZ^V;PIt@a8=D5qx7$+LtvO+MKp6~uMf8r+
znw6AwhUe)83?t#0GrbSZS(5fOV|a{NsbODcmt5u?NU_D-ne*y*aY&Z574j8An8ieP
zfN7du2}LybjwcYG4kT;;H|Qe*D=-(TPHix)6bzkE7LsFV!0cBKBQX*))?9!|$!A+C
zx;mCeRrbwFw>H;fT6^qc^U{d_+?`rKcwN_oEVkr1Va%L)sv^JIZ|&9vUG=Lf8t!C-
zAs;YeM`8QzWsy4u6n%M6T*|kz_QXBqUAbY69j`@~H^2UT%smdI+=Pb*<l_3XGshn4
zdhC&gcsxcUCpdrOjYB7+g1%7fipfz$WQXo5H{my-@Ni)eO}-oBO)rp$DHq7$?l>JY
zaXnp>z&(%hRp}TkUZ&TH0>7n&d=iE(s6vs~DO$e4PbS|J;E4H7Mr*Pu$rbJ&s$eh(
zqjB6ariXdE_831e9d{~!P(s_CoUp6c^7?GCCA8KbD!htq<lAo-jI@^c_sKZe7Fe<%
zZq-ruwG~)7(Qik$1lT#V_I+z#&z#trFAsKf%#WN>a&uXnT#Z1GO)C65e)#0tbUaZ>
zZ%bj43K=kiheW*}az1cUa91?1g4hUR$LMeL;`RYGp;8Do?qF?(!_~v;sq7LzENJ$^
zd=|P4rX~8c-B}!0F<`;BORD|HE<Lt1?NW>D!kA;9ZCo>E%?DK%ub6UXdMAF&yhmA2
z^R2IRSdW_e54WY;15_xCi@BJ<z3@TTA-XR-g1`zUlKz;8=V3(`EAwVnKkEK}U94DX
z-9>&?XS_<$0Ys+*B&X{k2@YmR=IGL`&=AXe?b1{hAW255%EqvzkrUEyN%Jk5pi|9`
zhoHa%Cl|*(_Q;8y;YdTdPrl~OKJ)8SeEQ>~XX5t1I_S~bbA6I-&3$gDCSMy}wIOR&
z*x{y(%dK5=Mp>I*_%h;ZQ$c*t_*1)Pvq8-_c2141E-YKamxnd(o_o&SIX@~0N<~Ru
zfZgLRr1sMv`!z)EFTNm+54aT{w0=$0K@DVxurqI5pE>Vtz!;J!{geU7)-NQyLAyn0
z%tJnO*O_i|n;~KW!!t3WSwUW1L0OgPois=ibwdWOgAI@6#99|ADb~RomqtG^zCH1j
z*7$nllYf<O#Sb;KXY9GuyeK>yGzVVXbFPo!YMxKG*XbWdJ~^*#u5I73*ZU;C^5cdl
zS6zQ+_I>ZR$j3`Ed_H83ibpbYYP%a>j~nz-=HPQ~q^YV)=OK^v<f==Gfw@#@3=sFA
zFx&K;cI>l^WW$KZ3;Q-53fu>(h-x^bp8_}(Q_%_TC0f6x*QbGOI$EyU5uK?92;bzN
zW<){)MK)nhrqSa@ioi3o<}3oQRLp?ThQm=6t4KG0LR2%S`S{{UEa%S+JMw&X&&c46
z8$o`>v(%17&o8-pzU<Ctt(h~|jIs9pzPqi=>*(?Gj_r>gY->(0Z_SK*y!7q)Bi@;K
zdQrS}i|Llib`Mx$dc%9mLGd$dT2Pv4>dqf+ZB7c)dT0rXl+jqOV~?L~Gk)B%c2j=+
z;Hie^=M`*h+G6Xv_D8G^$|7gRV{O~o4x^obln6ZIsy+pPO~7qvf(}E?>jV;#$B7`j
zAua`O+Z>9LVZ9m*4K!-S>jbm{w+E=U4xuM7px(@$XxrD3M}uufL~YLF0|E~Z8dyDk
zwY%lxqU@>561L`7oAct|pZh@pyZz})VWvOby_I9Tbb96Bn7%h!N7VQ=ZSlF<wQT%b
z`^@gJlK(V4wqf<)B^}q0hPjnBdXQh|PAOsEJ^h{r5Z;UXpo`L_OWQA|RgV&Q>R(iF
z-`zVSh)}ueVxLi+v$BQ01W(1QM7fzRha(mA4nT0l$p_>29KjomNvQcpV4>(InMFZK
z3^h0J<tiKn0p;SP4&jQwsgJirv($EH!T8R<#{MgM_aUDnZ|1ohVxGx~*>`Ml{LoR?
z7RCqqt(mhs;7V5PP^Z7|_Dgseq5s&yd-gRii?;e@UJq;N%#C;_!8WDek&sr$_D61r
zKmAyH@@FOXv2}~$pNMW<8f(EOqZ>~+HU0JyJdZ`tN9k0y*OJv|UpzKpFJpgu*}tE=
zX)zs#FT2#X?P{Gs<d`m?%Q6nq8m3D(XewdiBu7kYa2C@LT26$RAyS1Q9j2m?H5r%S
zGe@&JG%?^Cc?goWUdyLDGnQiM#4`tv71)dgan_cn%(>C~58uncPYQmP(f{Ys<;{<T
zeVbGK4eMj*$v80jv2_b4Jk@yTwYb2R%pW6Ko94$?ukW+vT(`VE_1x(rkDcj^8drQ_
zUTn~Wth1xdKLs!9*MnU{`aPMs0l0f$x8dx}j-QL;20b!%z|n@rWzkP}G%byJT;QxX
z=srIyMSx(WplNW$u3FDk5H3j2>vR}IFPWg*Fj*)OKyK7OVCxipgF95gU#JR&Ax37#
z^I>(vra~PuY;$|gd$YX?PhKf9+{sV=$nVCh=G$+)9G2=^i#yQ$4!FZ4=K*=u(e682
z^NNBu@_IbdH+AUFgC}q1#12ZuGh6BhUp`)+ajLa*Dcup1A^`#64-hf=l5FnD^RB}K
z1P@hoe*6U;tPeQ4VcKGi^bm*<CG<&G*<Bz5;gf>Yj0O=QMu3;2YcP99mQ_-wbMj22
z=D3d0Vv(SC>IFd0b$V#LeeZVPUd~!pj12B?ozjV4YU%ey9`Xk*D`t6*d8^|JmR4Z#
z(ikMTLhw*hd6K|dxEXwUC>-ih(918NyBWD^+l}kvR;GZXnv9>$v}hg*D2UWFJYNQr
zYv7oWz<V#W;N6XXLN2t2s!&ZDZ~&<OLx5!<RfNN0C@j$!PnT=4ZHVUPIKsAW?=E1?
z*`tEf`@Q+fk<pu$>~5FZo0HDJ`kdRq1a54q!=y^!Sza!c=4W6$-K`glG$;QZxNzU>
z8PVtefDIl{2zwaNG|RXz=^0`K#P5ofg0zSp{x}3xTLMrmYSA><l)gBL%`TYJpb(C%
zvK+e+(LynRC<&h+1v6@kJVxao(AdWN&Vvy8tq44;RbYK`ZSuUAeYzSypWucJ0CEZT
zp_;HRidg~_!Nig1q~jjo4?g_r9TMm>a_caLpX)W5P|Aovi|s<d2uxA#HP9uqVjDf1
zU!`$@!f}fy_{3>hC|K2lJn%f62LS8nbqJV&g9c+BC4i99CH^KpvUFrkNng=gjT~~?
zqOPW>X_up_w^eMvI-sKb8?IlAja9gpMbG|&rKDI7K){avCz!G^d78HPfJN-Wh=*q#
zrQ;g#PPSOY9fzeSPX#mYk@->ig9^HI#$uvGI%W}@g^_A1W&Q37J3@oM&PxpGl8NIw
zF6nhqci)+hFVPpb?CBQmD@N99elKQR>FjeT7SEV|wD#+Wr+3!jJnRQW*C0Xj+zYuX
z)IjT<(@C?Y(%bLAv}DF(SN;H{fP0}^LU*YLOR-JrvY01@B`2@J0u8tfXG{vOL0QOP
zJw~=!pkEne;Dvjha;z|@A<;~<qN%_7P>T$R=+af)Vr^K!0e)yC!{JkVk^y@84aPLX
z_#c4smcXyixFKtSO9Vzfj;X5()XWxX<~Zg|FWVK0LBn$)-Y6jh>PZb0jgz}k;d>wy
zhmi4{F&vSL9KsFyNY`YBUkr}Fsp}0HH%Bv=HOd-+FA|YX3;AsY{gSvisTyN0dI5j{
zv*%Oqt7r02A9{4d&u3^E3V`iMWQ_t_{$%r={CF$;EDFkmeuYNqC`9Mz3iqb_4>_#D
zo8n3?q40<cHc?(@oQUxR4DSWAHQA__HJ0Fj5@mhrqf2YN5;{&|D%Ff%+4W;>n^x@l
zf4KS<sHl!??K=IS+ZLhG#!(b<T6}a9oht$g3ar!JAT3d8LE|GY6$AK)5fLLA?>f8$
z8I)Eq0}6OR1T-c-LUJ{enE`_c4q|H<ANffxqXY=yW(*SZn|1%S{%@Z)$=q9s&FRy1
z>ePPk+O?}HD&TIwQME6RTsLfqD}hP%ifrOb0tK)8Md5%J0^RtvV7v<qK~@)h7fM}{
zi$NrgfNvR9U7}K(Es)-LM7J<)=bmg+$h9C%7#Fjs8-gJltMsQKBMBSP=dVcw0b*(F
zS*KRH+ruYv069S6=5^J%#h(wHv={s>`pD7)ANe%3Z8f(()@}**i}y-#R!*!s`fsJ?
z_MbIU{-hF(4bx7$5r4&@!3|*zO`V@6u*bV2{B>-evpYR9cS&pCzR;=5Q#V#7#H|O%
z;7#cHb3#$PCMlxl_b~@fx!A+hMSh+Gfh&el2LMl8kRecYDrC68YS!u^hgEOL<B_r>
z-$%j`w2_lhYzUed+28^@DuRnd{EQS<@Q0c<%%Nx}J}3he1$g_7#q=$&1ctp8t}jS7
zUufGJJmqTY8#8m~ZGN@Bx$=)k8fRvX?zZ+uXLt6$qg}H;I=gr%EA(XK*Bci&CzUMz
zyzhmTZJEdpde;A#t9+apUb*bL^33qF>#nGik@r&;1uQR6rqK)KLfi*=Gw|%)M+=7*
zRo?VS-ohmu|5N9I+<9AHJ=i4S_l6mtgHvVJ6Yy0Bj|IxeA{gz7`Z}WkEP#6eokJiu
z27secxN%@cEy)(oY7il%T9A%99J$S;U%82ATK(#y>^Ble@6U{w(pg_Rx#?Mf?MmIV
zf~3pyRyPiBj$SKuPdNJRn3lL*zt1?>ILL8%Uwh4(up;m44fCtB3lBVbG-c_5hyF`p
zM3E#BG3i!qM#u?IjVrQz*}#Mj*WmgEkk{wM4KqmL5qS06MY%_gEf86B*15-Nb3eR8
zl^VjX9qMs0H@ng^60;&eu-B|jER#f0As(~yW=K4Uk8PUdKOi46$MD+?^%xv?0hmz~
zhMXLnz~TnhXDl}46mVCt=;XA}(J)zotAg*|zqPV6>B#8P>l^)xQ(UPYx!mQe+ckL)
zKg~&B@qk}Z_Wdq4X<TpLrpOnB(RV+(Qa9C7*POBN`10#Fq7QZ}CU-Rm?|y!!uV*h-
zbsWfO>F(bdRd$!&j{>eHVKRZptbRuO{t1_U_sz*A$ooCA`E+!at8c%aGGXWJ{NwMU
z9;^q`=7>Z=as+sqYYDK}y<@QV5FH+am;nqK7oS*0aV-ZNb`=vPurBw~Z<xY!r1`B*
zyb{>a(&p!=0UFC$cNAGbcF*D)eS1TWcQzzdG`7AKJjK#{Ag}FD6a4;bOlR+>S9h%2
zZS4=6dU&RU-D=a#ougTyle{I*e#bUM4f^i9`()6mn2aD+N8#7&7vv6{j6t4u-jJ>4
zKANUN1_XadD+$JqjDSOD4(&f~Wyo#%cW1(k;NQDM3`fVldiwgU&9Q_VUS>^bIH)`=
zY*Rd7<oX^901)ylhgrbNT4Hc=0I)ZUK>1AWNekRfBy8cnM6Z=Ct6obiz5DI2_wpht
zTy(lJ!>bO*ooc!NFmvhH!Kd#eoZMbMv^$~t&t3(|169kfg#;X$m3ySUXCLUS30%>9
zXZ?bcrk=gwQ?zeJoD?2rgq9y_WmCOp<{I07o@rY$*t*?V_~rUV)m;+$PlZDsF8&bA
zTH}+|EyLKMV)eHM%07SN__pN*rwalXZif@5b*y;x^fds^BvGzSKwIphD6&NTvB2Ay
zXf}ImiZ0@GF>^eeb=q_r49vr>;txbvL`FRq17y&fVDn^^h6G9Z#&pCN&=>L%zdG?#
zbt8{&lD-l7PAb-<@HMW~OIUEgk6Oj7$m$z=qAI)U6RX>IYf2AFBUuY4-^5%?YUzsR
zyL%$4i)WUuEgZF79DGVmQXyFbqGb>@Va@v$Gku??Ey5PgJ)duhsy}@r=~Ft5JYVg5
z<JDJxdQ*8@W<WfKq%0US*efAE&%Du@XC7S1t4rc)MS_Y7FyH_{`8<75b}O3kCkMV5
zmgi-|ri*jf^syL#VwWaSC*U?!rLiR~Ixb}W>6&DT^`(4xpmg%!(C*;s<=Aztr*Gf9
zvhIu!<%Xn%KQ9b8tXfo^fPbQ<v}3p-<aRtfau{1G=-O|E?8O;;l^}dt_IMxJAwVFB
z0ITN4tMC5wCXMI`6*%h}r`uwF5DUii`D~(i0pns&2s|p_lZGq#WY1-GR7p&&Qo~E3
z2LV+fDj+x}hZo&rZUwWo^HC4}lr<=FodfH5HqY<<?BR!7LXLB--`l=9xwL(EWYyB^
zNw;0?+n0~+ypx4<2(m)_y#^#s9I`DD?_|u^x4jind1!IS#q#>kvDZKLN8h9QD7daK
zsiyM#J?7?-eW4Y)v?we`IO^X_J#=Di5KrU;WyIJUo9#PZZ%f`HKsP*5d~vadhsThX
z{zvbq(}p~9Yd}qWs8mMm76;==eh*DABD~MTZX;|;N|q3SD`uKrE1Gcw9w<nQ9m7CA
z^)z{LBakT%eSrvLA5Fddzkyn&$-TB)WUZS&Jks~KR|03yDLb2g_s#JVj~xF_kDRKd
zi~IUMI~1qzLvoSylzCu-#G&-C@|*R^NIOeWep_z2{5Xqt6r9|Jw2&8;wkB=J8{KU#
z@7WMmiF7Ro>rBHemWYliV)+_DAS$7s{O^@V?)P|UO5=r9Gh-jV72$_@bnsbX(bPAg
zn>Le90l6)J3u6a*+91LqCMT6pK@yCn3}kx2#5AR<?5Sle$iOMEf|+hIcyd|=m<$|+
z#_#|h#a4p}tA3+&WTm&qse#oyr`R{mpZ;nr_CHLU>wj32RP$BD`3WnU+cOc)ZNMJk
zlb0<{s%ds@iq6g(&NMzbG{CkSmwJ5eVpBa+N{v4*jZ40g*SlZY(cH5q(lE3sEXUBA
z7VcLf;Fai)K?RlrjU3pK@pj{=r^jPw&AWJWZiUpk4T&%6IarDj`Sd+eu;ySwl%7I-
z9YEm%s4`#y2ijr03ZUzd6ej1Gi8(p4KKRLm9~s~T8hBzVE7%1HvokG5f=#^M;FHD~
zUxz!qdplGz%ZPZnoN1G?y?ielkb2{piZHH!XR7UX<^7e9KRYyQynhkqd|lAR{mHk>
zhu?`R4}2|sQQO{;ERQDxQ;JiQU}|=RRLAdE29NUyaV@M|w%B%MaCpP~9Hnu&<~A_R
zL{?@Z4wYSK#g11!POYCi#idOe|6TBjV!m(xbZE>KE3ug1)@Ff7>n<Vyz!jE3T9{UV
z*GS4Y5$+kyo})C0vPX_-MYK&ti+{8Oi$YQWQ@B~14;_t{&d7?)k#%q(3-DSc3Ce@R
z->mH0rZSdyd#1oaw2$lx@h{pN?$?#7T-*LQO~BN|eQco>eXwwFXK2OE)UJgA%ld5{
zolo~GBl<?RGPc7AYK;L#mW<r0-7&v>nR*d?lKgpoT|9ZlJckX|$UJJvpTKnDxQ39N
ztt|hI6R#x)k0zZhp4mrnh*cVMdd62}cmUl5#egfB*h~}|t6|geFvW%3h3u+a3CsQI
zj#kSY@xYLXhrtN+W42WkXf!1$IK|3H(d)hKG<Td8aW9Q|#rdo|>>+w-CS2^fu_gGp
z=XKke-lw}TAFe6Q>U;dov{jurUr)CoYH~^44tURZ$;zfAUo7`8t}o60(0I2|!)FE`
z_g>#_TPxm4UGT!}pJ#<w5Zcb4JEd6OTu9RDv5ThyIt<Yz=qs^vU-a2GUq5)iV8&pX
zx^?^Vms$_BB0kK$B#?h~8bblAh|mRf7{VYDOr)vZeg(}Ai*>*UG?M@o2N{Qz@p6&1
zB^D8B1TD}bz<?fMZIP~dIB9uC7WLq-Sx@PPRyPKF4`PKyWOc%NNu{2yBkv|4-gdk9
z>7FS4XvV?-@6-urnuqon`x<tNIBT#kYmuRA-@>UkG9m+()rUOjeu@JE=7QpkN3J3M
zFcMoR>}TIyiJJPyvxxz#?--J@zY71iQ#E83-E6!QF=r9omww}B`BCJQ$ONOc`QUBL
zGs0`&0gT9Idx@x`*Tcmic7{KNgTw0xR(Xy`e%y#ihb*+p?B*X|>$;$tcdzupS!0^_
ztia2WN>_MA0yLYT@$K5YP!2dD)ParE`ru*E<A6M${*z@wW%7aTquUBXoq`|vi*waT
zbv9>DY8n2wko#Iz$C}JHUXA=|2Bwh+VmFCkMKS*dS;jugR%VUEif3@BqD&N-Ks&<-
z7`Vgs8sHeIxb@6vNB`YVGj;}HK!KM7P$tT5<B?|O{qozyqdOyCE@U=y<KNaD9Dio{
zimM-27Y~$P89tC~H^%F^L%S*u7YNh^rW|vbJR`Zl(Lywl)>goUwPx!LIc|U`KA=^s
z>U<|C<JJDsS~k}E`2LXaDE}OYhc2k{o6m1;PNN27CmusLbXb%mpg{%vJM;n!1;kLH
zrz#@F;$;9-QO#-|7F3|bTalh8Ya~4vSv3qdWH8bx>dA_J38BB1CaQw{UNnE!wmbZU
zXAH*8vvODTOgePElyePk)_VAUc`V(pxOM*Il3Ax#wBFc^vyM`2mew2ll}DZ|^Z4BD
zz;g)9pnC=pGNLNtxFv2{Vj(ehq2bEAc7-~JQbQ&E`^{b3=V2D3Ny_J0?U^o^4=lIh
zjo>9k&U;Q0F+`_C5q&f<CX8K4Ds`GpJYun6ryNT>@f^k3MN!!!kup(F!v_->l!?Is
zsskhV;!8KD2Z<E_SxZ93(E?oY*^d%(Q>KtlRDRbBN)*G<)R5y#uiv&SdgoUc&&ti}
z>rai&#(Za1v1>#0!OA-}gf2$jeX_P&q9%cQ3BxVKoPg^XlLS7+J@ga<3!{Bw#FP@v
zsf&3%If~yWW6~SREO&AMNKMxpCk~xfGy_Lg3PZ=PQIyG;q81hSN4s4G8H^T~4M6}X
zB#wFbr{)&|V*S}MxOTHAB5pGCfNOPHH93nAm_AYbg_s>Z3hXTwUhRebuwC^TYvA3n
zNz3|^vwQrGvdhZ~PH~-2lg{*gxifO=A#5H{_IPuca^R#e=}_DSv?Lc%(Tt%m@(E@O
z>j<tf2Cyweh7*jI_Gk6ESB8gZb&9a6ZS&mvkNb{ydi3f9tgU|tAUoEJYIE!wgD#d=
zp)q+07vrJrWCf$v+Hn(u7N}hTBt)nT_tkLUrJTltcE~CkLeWX=m!K_pDM*x&q@N*s
zyD~9h`<Qs`70&^ixv6urxv(dxqPg3#enI77kB^K^2X&VZt-rBnfwMmu6dc_#zv9&C
zone#4H9s4;?P0T!;e)b*^x=?59f>BQ{uJ<p#sLr0I=6f#PRj0cjq?cH^1F_~);%Fp
zu5Ni16T}d>*gb#rn12(I1rgzSa>z{>ZK+5#LahQgTtF&Kcm>acpkPY`R=~P6P`xN{
zG1eT+mB^~n@tl%G*c=*JV4n?DwwC$!<*^@AZ{OX!h6Rjy4PO1~+$%j;yX}`<IJ(tX
zIJ8+a;lf<KD~%nj+T%YiHN<CJTS!gs&mMs-oj(`+ZPC<{I!~Wb)0-hDO4dv02JDYE
zQIN1Z8eYMv)KH2zy$1?|dVvv5L)Se2Ta6o|{110jWgT}lUERJkHgPMYk&S(Q#<Tl9
zH^Gfsl1{xoADoV%rp3b4JTWavD?q1N8d_6|cR>ZnAJR`yNjxSlNQ*kuV|9A)J6;SQ
zP(}V=B2Jj8Q-gCybQp2p-H%QGSmM<08nKC7=@}qs4>o*j2)rLr<sC!QKa<C;Cu@Th
zZ4%x~WB58Vn@#6YksY5dZ>3H-P4Z8m-}P6f=|^^rndpn{&0-s&>Q!pr{x_-)eH4Se
z8Rey$iOmLVk!FskUgLlC3%Xc<UBETem^?HFjs*C#Oj%cJjA6!Dfi)GX0id3MRjt$G
zJWd!H^p^Q<!~%D#BJRlRb|LAjlW}XmT@ReiTe--;-LoKftxMl*%gUS5Y!pX#hfMKm
z1}qr<NP3t$Py_l|;2BoCI}K2pU>?4R*oUKB3kF039^lPM9m8u)$S*GMd2hU;225r+
z9a%Qfq;`V<8bO|*ogi!iDqEdIW(+E1eVGS>=n}KB3k11rg(OlA=DP&$nc}DjHq{B(
zo7e8hW8B0>wb6SzSkOf*$OO-@<!z#8Tk_hTyy*1%uJp($-a?mWTkhn`-=h_S2*rzC
zWxGiMgXt+9#}1Wcz3P<U2+QIHjDkTJgPVX?fqE!}>DXs$dTySUXG%@JU=_GyOa)T~
ze@jcuB4A^6pav?!M0|uNBmjYz-q7+)g!ZFqSg@<7#<W3&V%`cCUnie109o|m`fIS=
z-(QIs91FhZNHZ>NICcBJ(%5-_N5qMEFIC8tqKykk1@Sa$F*D+CLMHW<sdSc=*xUzU
zZ8NAPV-zo=&1_`|Ko}s9GN0$)1c6BQKVG7TZUg_D)V$H|5ClAzkJ^9;8i8~KsTzUq
z;L+(4d`Gm%;~V@Eyp@|sza-`{VIBI!6xdQQzz$F9ZJ7?7XV|5XyO#`Yd&2xJo*JBK
zaNvuM!M>g0(B9sXn~{K<9L2eK62a$N)Q#mF68a7<#;eh$2#ir5i6rs`K>J;Paz-dU
z0{WEAhJ8XkvTAB0cFBYyRx?QwaR&g1P<Rjz=D_?IQ!z!0KtBY@?-Q(QFrNXb$e4kp
z-3~xeD#kuh{Dqk?@uggASz4e>cPS_Le!bCnxjl2y3v-9DSZv%T28=+u5gut}brNtS
z(BlZ*DICPnz_=WL;epEMzX{J_fy4IiDGNeksfQRn>u4ATFpL&~AOVPg22a!c629PU
z0@7RHFM$($7pKV~6}?F$P!KwTMgf{~L_9^23UC(k9}#o-uhb84kF#3O;fO$L3f!Ww
z=Bp?^$9NMzThIaX#Qkb8z6jd_g{#Fp3VL9)#!6rtbOJT-ftVOP?0g1pM$hNp1l)KN
zF%kX&Jg8=1nC9XYdOQMrqE~2CpNHQZT>QGzq6(R93KS=#CK36_*uRWt*faw#u)u)I
z#XY`)?~0Ad&|5qz2bN)}c-gGxLw@xabwSjr*#fpEB!L1I$Wl-PCRIE*QIy0wV2%2v
zLn4pdZzz#}-@af*3^UYWzMe;2H9i0=Cg)@(Bh@2g-pKF+Qt&YQuI6zmIL4Y{)bK*<
zRygp4h%_e#+lyR7K_@8y%5$<Gd9efgsE>wKe?mP_tim)SS|LL!5}ug$W3;6MFKWkP
zBT$;|WAfTSx`0kr0RJ59wFPKX04tnDubaq&V)*X+@2@WiNrT=%Gbr%Sb8>RPBhX#=
zs)jb;(NnEXl7I(`fNN9{0S!0DC<qLVq2rW2JO@3)6FmOW9tIX_LItB1uYU?7@L%hR
zR-UUaKiuOnVaesTZAQ-nADJ)cl>^F}FN5zT*GSKr@1})2zu%!ryZM@@#x__HSe2B{
z%!QA)EDG>SO|D4T5&m+H;YNCBRcC+h!EUy5QGhOkGL>vC=aeo6s$i?(v2fMMTmz6?
zbK~UBwo_O1tJ-!(PU)HzSaS}&Rjc3l<Dn1<3{RU3sUZhYu9!0r9E3I%9ww21O#+7q
z(BMp>w!oUq62tPr5rUSI(kKPm0Peu6$O407wE~obD)J6|Y&V^-KJdsJOItUGmM=R!
zcS^f$^6>C3GuIY*x0LG{-Cc6>wdDggpQhF<%zd@*w@tp>{I;eux}x~bUe>oK^kn$Y
zGec-T<y4br*k|Ii>$m5=5c7ZIdgh)e)Fp=oaFt(g2%Cz~%@VI><db2Ga*~yN6*@xp
z=H1V;yMLaUw1eYUB$V%0`XE}>s{k{D)?p!Mp#cTZjJ!&M9<UhkL3)6S2q}^ya<bZW
z0tqlNzzRd$RfG;<<JHDA4kQLSIc7vFveO3@5)dos+>{1NpB<CBwuY9E>qVGb-@JFe
z=twW|t(|)>x#fQEzD3n73Dw?G^1K&fy55>QrRd#9gQDOyJv8R*ric@r-z>6ToKlF1
zujs7Ngrc`20P+TpoXI7%?>-S0m5-a1D}dON!WY>;k^($Qo2{sV%5`ZTR==)6)e6s)
zFUEY<wiRU2ajW~2mkt6j1w>H>R>0t0KL^tzi2Os9Gf<QYS)%8c__o3XN<%S>m+y1>
ziBJMy4LJUi0Zs)CKWJ8$3*wcxM?c2CooK*wRo~Fw&`HaBuWnaGFX{ey25-I*akuc<
zkLMnyMproRY{G*5fsbzYO^x3^cS=dztp%dh`7kST>dpQMmraWdSKmVJ<gM!c%7L5{
z?Yi|i>bamZwBlBs&60O#XBhe6A|jBAX1oeX6n4z;DaX`(zay;tP&>Dw6({-xc+L8|
z88Cpz+e5qZh#%D8a-7DqPJ$s31S?L8rDP!7Mj_KJvkN#ISfn!hf@-298btVi7)gSy
z6bf~X!ACd?oWXCx0UlrmGv-x{{gU0AJR%HtIko5FOH-FER_>WQrNwh*>5+5#hDjr7
z(fiY)^iNZepI*LgUiIMyYx&{&N7+4-mR+A$oxOZk?vf|o(z~T+q#>`~t8ba+o+$Qs
zGUk5s{+(=nXvMPnz`^oyn$jg=Hy~GtF#U`bp@Tk4Rzg&9Sa(Fg;nrm@ieKlm_4BL6
z=&aAMzaw+3dUpAXg|vH-0((K7W_3IhMG@;AbXbB!ybTcCM20RTR~6p?JN_wR%xc*3
zeCtuBSLdl%?a@^;Bp;@#v$_}^@`&l;L|{{ep}E##RerhOKlNh~kLlXCMO^p$cGA`*
z=bp9Uq|Uat=S~q-(OEtHY3A;>G{CJ%-Z1>E&6yQ`yfY=&f*jlAEK9=7z%$~{vyvV>
zITSYmXUE^zH`lLo))&!PP0!-)@jgdCt?;hH4B@0@8cYq}O%E-rQ>&PSrS6hWV0vGH
zSM#jlW=K^+{f4l~-nAJKeipGO=}6y={n#fLBe}nnoH(>}5ZTeBg3)CV?AKX+AOY}#
z`tm2f33LHgV2Wl(jv2d{;CGnU@yx_=90Vd?Uh@uaXVX^*NM5wt+WGS@?;k7kymrjW
zm>TV_h@9$@S(l)C`@LhLaHdT}b+-5V$BAYANk^6sgtQD|ZH6NE!C=FL2hC6S@mPu$
zH=#!AeJ9+%gQb6XJ@87|o%Dqj2{Usa6p!v+gpmNHFifjKB8{6<RgjA~g?~b7X^X2N
zM2ZYp4x8MlQjbljI+WGbXe1qku}eG@8Yv~r7Z6Njlz=w6a0BV)ah_)e&IG$l_Yqf^
zk%J4_p21CJyjl&tkD4Y{P}+WR{GU(!=oW6S>x>I0*62sEl>PSb$XG?d*4u;q-W%18
zYX)8~iF4W>c=M;R5ewX7oDU5ygK@DZf^kJSTTGPjLck=Em0iW-jjW*7Faowr>KGi^
z6BW?R2Vwt;FZxqo4|(5!{FO4QLWTFWuaN<H4g`rNV*(&jkbmJ;+Q_$p)1lAfd8443
zqk#3oMxxw|JgrnTb|omytL3%?`!Vn77rf(c-`$6i4CBLB=e9KVrI|beH!Pi%k=@jq
z7FqS^+WG|nodvd*3;CK5^YAG8Hck`%gKbIq(4J7(n~WvQLs50pBLnO3-cfr@NN;c6
zvao6i#7NGLBenvgHY$ol!Ej9Af#bfbTNh5|mSJVkH!rMd-G`ixI@ZX8cqgm&be)qs
z29t`!AvL5GU)@@ocoHP41G#7`3c!mLOrm2_Ypou^!_$p|um-dCt9aGeX)IrAOr*d_
zK5AjE0d^f@6U+x*@Lv4?Qa>PM>aV?05`5!|-wkh{qNb`x<F<!a7q@*8EKM3Wz|xA3
zonxts0*X^}vpi>AzWo{}e$C|#W-i43A%ocbZBqMN!N*I40~1y(UtBuovJ>vOi8n)o
zpj{@g**<nGL(#57@}Wt$KBD?&K`DY>R%d;NakSIpiMi*gpvz&a3|I(T{C`XeJOQTz
zECx49lBfw_j2KR@fU>s4L&OMpRDuGB!dJG|!Un@*6pS!g=GZbxf>x6d9pdma{0|Xg
zg3K$<ds++8cJF-?nhSpUj-hR<DLm#wk0-`uno~Z@V)Hl6pK5<clsp30TJpNn&HZyQ
z*a+`Vy7;7I@rc6t!=(2&509=lj(RRGy}ev>DnjbpGSBZJ9MB-R19CxcT5y4M2<@7j
ze2)nSKO6FlE6rlYvXQrfXI^{O|KtG}f6U6Hfz(%@fR6#03V~2K8&Kjhl#oY30QZ_`
z3@ac&K|s<TQ?%747&^DJNvfl1K!ahepi=Rug_W5{4YCie6BytX8F)P}@(=tQOW<bR
zB{}w7)r6(|($y_<<GRlW_g`9kV>BaV3cullf${#WCa$>q&4@Br`uy@oH`5kP@)|gG
zab%pwKNQZ7y;4hSn*F;p8TX4nuDiN5{CM$!OM*c^F^@@*SQI_ODG!*LW^6hbyzf~<
zQo0?fN(_qK<5GZU|C8IO<VYGA0JwlWQ6$<@5ePCpI;%qwwDLXhGY`cen4o*iF%VG3
zD$XjLzN`#nlo<fzS8-{Gyx;*i;R<IyKxL}<0DH{D|6KCdwf|ht5P$J~D>A@qacwT=
z>f9gL(tTrhq<?39ZuiCy8pZ>b9bDf1q=7O7z}CLE=pHiMd+U`~db{}-Tf7d0JZNr6
z?vS3Yd6wDp@a=HF;*_fZw@DtZqh)j`HdaO8;<z`qV$Jf@j0+3PZ{i4p!*LVL7rH0-
z4MkB<(vjqmd22pgLZ^hMa0dfSbQr#cRHhQ4UqLfsX#z{F(glG-Iqev|gy>prPHvf1
zhmZ!$sk^9<^6gIGbQ4=e0}e!78uOoG*_9jOIawaLq3ecsh4^*;F!WJXXWSPdhKFy5
zv~BrvZ}f)CI7V8dJhEJ4%W8gX|7Kmu;*bX}*M<d?#`OwM)^z1Xp3o&n1uVY~2fOW^
z5dY;V_j4Lmv*Rz|;7+**(4%?+Iv-Q`2#h&9_tFGnN_0hE>)Sl!UZa>&h*WQ;o(tu?
zO@jUfTu?@ED#Rdih(rL4yK>wUreroeB^^wX727-_2gMC&gd!X0C85Qj2zmk^PGC}}
zC_|V5bVg5jp6PSyykEvThmU2%78Esn)EYm!w?y-)C>lm|drQK4NB5PmAF-L@W1O{9
zB0Qevc#6%h3UVLxM<uG}?9^NeXz3LjP9YaQsbYAP56UqJ`6OUeFu~v^uAstVEv&)o
zQ@SFn#mInmcj%KPd1c-TGl`^`U=K-&N@^wv5^kt?z7erqwHgy$@Oz27Pn#`+aK+pu
zsSLG^0uPBFL+@(>SJ)Lrv2;Urx-gbTM8|u2-HUhLzPs-MDe|DW@%J`P^(wGcB*cd2
z)(Ae!S9K2W%kKK~wezziZC2Fvpn#H%@ZjsvB_b8L)8JIs(4+1G!eOqJO9MKgIBgZK
z0!8i>=CLC!a%w{A4#q^u@OjnFo0qPN0!Rh{gU&(Z#AH<fA~jMl^lYVyF9a~=_{1Cv
zeEn9Gt*|`?qaE>CX*xVZqc8)MGpXPjqajca8|^~>Vnv`88~6WM{S-3Ilo1bWq`ttG
zvlsoAr<@wV$5rjbzogekynO%3+Cw|e;poL#%5BLXHiTUH9w9a$7I@^<Duuc+5$RYs
z3>JbxI+2`I(-DDp>k@TF^oGx{@mD~}CN=nXmhE62Rx430Q0(&s-4c9r!BAv%S@T$)
zios!1Nbd`J$*8sKMfkpDr$I&n`FVTvB#$uG4-AUNJpRa*Y^*N;LpuZ<lm#e7Q;h(8
zfr%5fWw4~Q$pQsU_R-uHwszXu(%JQP1PlxIlxW+AA)1zt|JxMLze|t&w|2ki)1)-^
z_H3Mg6e)qo`Ilr>i1?o>h$RZOH8IbH&PlysB5#6skBMNG98uMUJchwhA<3ud@s5Zo
zU8%KYKnDvRcr|02gvk8(8V=jHfLF_4SR8^?e*~xk<;o<vsoy>@f&no2?yYy2>{?dC
zW6+X7tf*ESj7*#Z9}Sp=a@Nv`y4V2FF@nhGpD03s+t**Q1de+<;(Ac~?(mZ<k-*t9
zx`ZVe2fuqQ(0BLrw7D-o_H4*qTllr-fy?>e2I8zSI2UB1q~?o)6jw#+^z)*{-$c^v
z#!74#v!wX@o%8jEPKYknuvRn<675ZM2iAVHt%)LfW<i1~Q0Qa824^hEw*@BQc?GmA
z0{b#|k|+UR=slDbrY)PZW2n*?yC_Mdba|C}25%w;67Y|y{$R(}o~jz}FcyRn3y$%C
zhz8%iqx8A_&D8I*KWoPt+|HC!-JCt^T>Tf{oan;(<tF{rOikdP;#Hr3!6|Oa@B$8j
zCAd~}LG8)~a<0WU9TWD<LlV^`7#gwQ=u`~Ov!x*bZct9l{CwdB#4D|e@jT4g-_`2b
zxQ&OYQPL?ih5Jh(LnjwyE4Oky@G5^-*DwhzET|dplV)Fup@A{Yq2sgyWH8bxOWr34
zLIA3lA9eKhZi|Rp*#B0zGI-RpA*s$`%TFuz`1DzSY4+%&fz>X}1fy$oXLNR7+x9|-
zwMeEMNV8;OjllkeXy^q;vk_pqiK4#YR;2$p&8ahaJzl9e=B@XRcjbPZoq~m1#-(fT
zZX0SA$vve-UGQ=w1s;Yqh3}8nA$Cv~h~SdSB358(`e5GUMJ*2%%&_-T9_^AN5E_Zn
zB0M9OCmQipntpz|io*5(wVwFdb62Ea*NpHJOFc>}8k;jCPPFP?9Psy^dFsg7;Fkyg
z?~hHx>5Kf6$LTV{{9_W4BnB(WFed^#C(8ncAiow4R0OkeLJimXZ1Lymd8g+qeVRr#
zEv)L&OsHuDaZK!bRchCt(JdHtMNter3JoiuO)_9|ffoP@A~&9LhusC+@{AW{NMSy5
zObBLjX-MT@NT`I(u;|~%kkSNsS19yP97?hS7=<|4j4aJ1eZb*VX)iL{Ri-g|L&2#Q
z*TC!9-j2wk)2eKNY!O4yMYQIrF^nRv`>hBfItG2Lcl2dqj}fG)vp)zbAk?$s?z{h)
zSmds$1!C2#j)9j56nrzGD$C>10faPvZH_l;y>}K_osxpjLpT7!=0N-ee>js>KYp5v
zix*bn36mD23`p?pQ{ExD!YqhF^h!P6kW{(n?5+s?;I0tE=<bMWOE(s_yJF1!SFi#5
z!#(DnCjs9ESO)dAXW9e2et7tqerRV@m5P0sY6(c&i_xN?YnR5JU(?r@88x-5pw#tC
zw^Zn09r|nBJl=V(zWhf=`&JWs{HDi8M_Lc`Vk7%RgYx&8Ct(OM<g{W$NV{49NC$FO
zfgHINtlvTY9kdLGK!yOhD*~0_pfa%S^jgHs9h{>{Z(MH#GEG4`j4W5G;I4?VbF}t3
z7nm_&+8mDYMjfIbxkS0OoB;f_rhPaosdDf6xFqxaUM#h}(20Rk%hH16BNt*cfj7N9
z%CD~zBdeqxQB|BSzCLM5xho}9d#rsh3x~-spP{jt&mpcg&o6$>wbu@5ln>fn>5C@c
zoG$kmyf@L!Zu;81*>9dp{b;ZnvS)L#8)*^FuI(ZIx7LFv*gWXZ_-qRZs$wGaAWfu_
zfoG>dTO{mY!Vor5ZH7MvAhEe)87BDz7!g!*c<n60R42X9jB%F&vwu!zl2nM81a@d^
zPt%J3#OeeAgV@K`GUd$}^{5=)gc4KDJ@H*f`S<%~dxm`)KJlX|U7Azb7XmyFJQ~v^
zG$>XlMVy^?_flL)Ra#L>6cV{isi9N74kWJ+*Sk^<#Te7x#5~p2?eHR&dOS|j<7n=<
zc@;NP0@I&^&hkZGRti{N|L&6q(&)b6<2M*<-5oxqxIWhfK6ZFMxV!B?;arNe*DJ`Z
zQsEP<3<#*G)Cgw)G6@EVOr!w5UkU1(g(5aRHjk~HR%TEuFr;Q?>}{o&ICw?k7^MFr
z%#S3ufDKmBhT)`?h^T@b&%O(t%arPE5drqZwp{gF#UJ;yZJ&qL3F|jUr`PqpJ>9Y6
zeDym~Cj!5@d&&}cEG@h8+t;RvK_~gUSXzz&=<Uw#HxI?AJwm=p3qFC#izJ+La5}j9
z=Hig`ny%1Ex6Yx5R5+D9<KbY6y!;)({)ev7(3|Z+%Hj3giVH)~tHkXmh}$8TL|_q$
z7L34SJdI&B7?Xp>HA4v)XnZYT|7w-QgT4+-lB(ECR_v|I^M#I8JhocWQ}U6P@jONQ
zIXjpH0W8IPHCSV(QW=CdHjidj-=#RYa-eF9-_n7k2Sd?sZT2|TH^tKUtih$8TYdQY
zo_vp;_mT%rTBG;BJ$p|C4i-=TDzk|BKDn71xI*mlipyQPVS$sS)=0EsIB8(Y4VzDU
zPs##4GEtO+Zes|7aWt-lq1Wl_kq_RG>~n6E8K2y|e!+Io^;8CX{`PG{f3s*ABoBZw
zS{a5s3t~oaK89Ds2(qsVb(@8gEi79^9#)L7YRwU_@)BZUFeB2Wu7NZIZUoD$Ii&BI
zAlOC5&T>Xu2PK97R6WQ<iFtbboR5R~29vnYw9kGVHQEMVuoslKZ#$QATLx!KaptF*
zz^ctYl|Q~FG4qq}vc#V?H=~bu%__A#NjS~CRA+YupMVa2)^NvTeNx3Z>=;-6W%7JC
zi&DRN;&P+KlsMTl?$l5~@qzN6FeTFU=6rwqo{;0W1X9%@DSch%YRf=k>ata&{xU!;
z9*IS`gsx#Y6?!N$%wQk}NCZ^_BnEeb3a1NlDzr^TEjCsl!%ok`zko0pc`j|5BZe$3
zau2_k9h&C1&nV3|{)7a|e<;Uo*PpN4YIJpOD_(Wsg2t~iGpg#8Z>HJRl^GS68<lzZ
zdZM9T_toV0El&<3@y+@DzD2F-5Wg)`538|NLaaJ^bNBGhuz*uD_C;I^dpTr-^S&VW
zh-{Duor9^>JWc7CbHms@@$|go%TpCqEqx8x91r%y^TCym{^}qwm;@V+=d_#(+56;b
z7-{EWoeJt*_Hzk~?gC@1o-Q%F7JevBM-WNn;RKAJ@uoT4Mw)~+_K6BX(XDB$&GU&5
z(m#e{zkl4oB@As?eq_M6y2rZ2fIP2EmoV^pOTw(iV=-Kx!O*{JhIX$HSx#wn`6!2c
zk1eqhUT74=8EZI49q=g^Ep--J#6D3J`*-~TN6j#<v*A<+H_VVT#Vx&Z<>IF_i3<_{
zd{`y`W>u^dXFyrgLCA8UAF0UGr5KjU?h<qxPFctM=~;z&>`Fm_3^8I3#xi(5JD#&}
zV;s8CleVm35f6LKDjLizw7Nzz7RFtg=nPq%6nw5fqr~?@U*LnmM3_>9?*y9I7W8$J
z&lZT#|0*Khh=m~%#$@_u0~sqS`^SmOit+=w&L}?&{D6yFzu&%3LM0LZPD5_>5Wz(F
zLnic7Lb}9nM3D_LRl_YwS@|(%uQNf~iT(;mAIqEmNz2AFaPy6V3N<7>Bv&g;Q_MES
zT6H*-D)kb*PwDgJmBmHhPFm*M-IY0gZ<Od=^@?Tq>bA)1Ni#NQ55h;Z<`7O1Jv1R5
zC!gih8~>#!WW3|8V;%J&M}}P~3(A)@D~}A1?C=ox@ieAYhF-Zc%!*9lF=pg~;I-4_
z2~$Ia1W6C|Dz~Oq6WutAyJ_@{dF_zd6t&J^>RDhoHt9XJx`y+fJHwh&cABzDBCEAr
zOw#Q%9V}_qjC@Lf*e8LU0+vzz!+&c}^4V>{Qxbx`W7*a1buU8Q+*-#|QW{W(lp2QV
zc_=G-*RKU|;#V!^wCO_46?Se9oph)zaDxCdIj?LAjE3OgmOi-h!>4cqta=dx(L~WI
zLO}o^cOgeFM}*2$L!E%bkY58R#X15tsZu%;i&=I|Jt0X5bP7_4Mu&t%x6>x$7`+bR
zSFL~*xzJ*f_m1i*MU5cJ=UHp~ftJZzZ(nHMet6aQQ8|X|D@Qk%99!pgr14p%G_)Jp
z%sjvXy#ZE^8>hc?i2mYaL!)3Mpr9KAJT;|*732Ct)(RSB#O25PBJ|%w00CIVm5d7_
znK;TfbGZ3XeuN?61_sfI=mHFv8&)m-f=i5uwbMkT#bEy5q>v=cNkFP|&>I){u8ucp
zYV{KtJfxLKiKX+<E|IDNjYi%KUA$%Ft$8PkX<+3U)-*AC^Jvhuxi1e&iir}UL?Xds
zJu2dC5#rYWQM8FjdUw-_i(Rj=-ERFlIjrpQKFzHKW&PMDczG{D5c1L=r(94GZwVMP
zKQEYV1%)8<_zqOb15z5yfI-3MxaBoZG-Eo!3h>l$%XN87an`&|uZWePFtu5jqwO<{
za^G&-E0`ocJ4i1>br$F{h;3WzHr1^QI6TW0TzzP+DSH2YMJz=#_C{y!XM9Si3(;Q;
zdn`E2L?`8=1OKydI)Rt(K#0e#UHWEpcKJ~H0>jnq;U`1yW@3z=@_8fWaCwe=({@`>
z$yI46h`|xe6z6UXgrfpNK^b9^6@MW{SOpU)!H2QLgS{G!f_9o=Vm?hsjV~Hs-KfRl
zOG%Of;Pc}2pOq>f?!}fd)`r_(Lmy!@kmLjF<+TDql=HL1U17q+3nva5o(-IRx;1)N
zX3d4gy05jEEeDEa^Fsp*`}1D7`+`L)FfH;msizM$NiL&hd8)0Wx1rWqAdx|V@>{xj
z4!TX0hPHsNCiu{IcaiVY5dJr`>9k{Y->_OHV6bdK9wD-#Cg$@h;|!@r-I#^l5LDby
zejyzqqhg||kd3eDZ=8d`+8{{JC4>#l@D3i%JIhTI*NyDOs<OhQj-edYS&RmGoxqsH
zOi<_ltEh0)6SRl>u4BYo=-s!_si1ult2uBPRBGgv;P*|)qLM%p^cGz83kcvh$WV?U
zpNG_dXmTsErV%)AoPmM7CN4-34+s^<Op=kQiy+l2F!3m$U0L^QwL;eUS;q^bUWm(b
zQR~WM`y#43Fw%>tycL&b(Dm-YkX}<-_s-ma!EM<l1Tp9(CT!G<&d7xzdtS!;0tb@0
zd!G#?uWcPnoqtsduUuY$Xm<lDLKMb-({`>2gC0CXDh)-H=##8FiUgS4Y=|!c!q~tL
zTl2$;7$c9zfrDejh-X)Ti2$)&@Phyi%Y1P%bUazt??uo@8T|d_<0UD!+x6kq@3#ys
zbk_bQ>vBNx%=;tzLi~aPBonkK5RU<Sw%~S9+^zl6WImc@YWiTSCpZ^Qy;(5f!LTSk
zPSIrC?xsKkd-VHPx7>gaKr{tQ<5Tw>2wW2B+(py_Hu&z)cbr}<iZmdcj%b|4f)3>1
zLC_NbyEP9O9@m`7*3mDHp7@U{U`fj@E!f+aWGS4oH2Blc#z%=s1NU6h*ER3gip~3x
z+m+{&1|JKUcM$IWXCe6oXHwTt)H~<-k4v#_E+5*(o96l>?Y-mv5D+J@Z+`#k)@}(e
zqx=mwZSbIoX8;>eB;X3nA$zYNk)#zxNG6vMWcno;SC1qY-t7TWU8Wb0sW5B9&alLl
zk==IhRil{z27Or?clIa$Efs}Zf=?uuFJwvmt8FzO&Ru=!L}1@NSJd^hVl8SR0TzeY
z$qoJM2Rg`rI!f20vN|xk7t4hzN>bNz?VWG2k)4aKNj$ulvMcHDn<OavI^dO|9S3S#
zMSPG$0I6Zv43Y=-VOD9NLMj@{4GF{e5ncl-kcFi`YOP)TgI-rh`VcSI7cCL2EA7^M
zx9{%aK^h_W_=~rT?;yEhX@M(5a50Rod=e`6`^A!aPQ>P5)*QH%PC9nwe-%ABiZJs$
z1r^~o2%FBULMsw#Qx;ZXhp>*q`>;9fJ3_4_+{#MQ0Sd4S8GoRF3OGzfe=*SvvnIkF
zCu=HCQ3C`M&m*dNyhcV9AR^o`UbRLDYMz1JMfG~4h1bPn9a|n_wvKORU=Dh~3Svu2
zW9^?O-HI^|KHgruE36qx*~MeIyVB>A8cX}&G?DQpu*Cm{hj1qQ70AkhYXWyjtW%w@
z&Wb+R`FNagZHl+=ljR3W&y@Qm-$xpu=(Y1v+qqg1qkifZazF%`Ea)Z3B3g4q0p1X1
zp0TwH$Oo}fk*tuU9W6o?0tZ+yYU>Q@8ugk)y-J@C?_45$&@>%Z*>?2lM&%JA2=BS^
zN>z!+r8e7F>4yu>-eyT-nhedmaGHxIekjf7c$4!2^#KNbtbhjor4UqjjUA4=9r_Y^
z8(4z=jboTGGHm(41ZdNG`Qmw+KZ6oMU#WKj@F+^;v<ZL%Uogan;Y2nx7Q71pLLs6G
zXF_e2X!rw0N_4(~n8&H1BOHVT9^lFL3EqKl9;EpcfU%^Om+blG-qV^=ml`u-+Qi(B
zL2T$!L^bsoIfFz0wYd2N?}s3uXBb91>wtXB*m64k#WWK9AAUa-f{)(At49?<YilP-
z!SF2}V^5Rauu3GG4?G|<jg$ceZb(rh7n(VEBc#UcFbEv-R|V1+=(q^gsFoR1QZOV#
zzj$Xm*?eegp49<rJ4lYH0lZGcUH828zw}zg)I7eT0ie1LZvtT)K`=Rdk-(7C0Jis@
z_m8%75^iIRskH(e;MX8?3f>K%MCvH6@F{Cyn*NYP!p0K>b`vFa9?w)rypqs3Rm2k#
zXr^L$xR4Csoqp%}C4V9_f?t^#uZ8;|HwK?q;L$;Og)@GY`4^f6-K7F>pgk#80KY|R
zf|F^HBV9V+EAnpzewVU!7KfnEga4?PbLfdX;8x1Mz1#wD3t^^iL)k*-=+NZRY(*>^
z|AED>gYQUhGEr4Mvtn`<7hp~czK&$o9ofJ3*IwfXIsZ;>3X3Mm46G1v7lnk*3iT+^
z7toXlf`YI)=GHj80iA~5MSXtqQQcYrPh&bOA2O%_i-0v?h~4$^$suCFAPH2!+q{hP
zqeY^Xczh*($DiMNEkax2{~UPlwhps#KNfmDNAoYt0aunNmrtQ)k)g{fxpUM6^Z?qT
z0t$%u1MEpFu?B$0HKRmxNfCFeYj|xex`MiSq>|8A3DA<m$e_5Ma|Au$E^f}ah%ewB
zv>@QP85NW86)lJ&RU)XM7(4pgZGXqax0UYk^dWOI7}4sK7zTnspc-hD<ew1HA(T?V
zG8Kd)1lSZp^$eZGT>&-@qV?qB`*X@TxvCjMCa@cI6czCgjG<D4B}94v)#;BL1^8@&
zwig_bM_?S<!rddNUIf+A9Oy!6RIt17PEn{q6yO7WM;wJQfgz)tirg+Sp8So0!)aY0
zqB{bl4&3i9$a)tvU2R8y#@fH<x1iI|B8i&{B0_Dt3-{~jO5g;n=3^?s{R3vm+;2e#
z@iuom7@}IovX7A_BX<;hiOO_-F<=19P^+l-FPa%W1#PI)dI<=8exa-L5oS<>YUICz
z51ENNi{Lonf>4A|3L2y9v(v!ZOrl@F4g%f6H-YhzsAdL1zGGsAc#WtaS)o>W)WE|E
zkyPB1*B!qi7Ho>9;XEJ@A8!S%@fiNr(gzhUp<n~&=+9bId27~3_p`^o+7WO|P!_)L
zZ)Xt!%LnEehLX>xt{F}Kpuse()tRhU#fP>F#SL42GH|c9+b))!4n-r1vXOU(x2L%9
zE?J(SW-!3U4@lCXmr#NICur2vU1OR~?gjEUsM{DwV?+cMK+lL0fuj(qB{4?LB<Wb6
z{zvSLeTiDZ(*y`YEno=nD`#o%`s|Cf-9RXwqY_ea$fLBybJiWbZ}U+4qDt(eBF*+%
z)!AUHV<CPoM%43~dDS;kLZ_CjD*r#qh@(yZ%P+86$)qW*$aC$Ax)$Nb=B%IpUgW*N
zD}`TzGU91Xr_5H=Rki#I(&Fwfz$OZ8KZpg|8j1upPt=0n7THXa#=GI9H4-B3;<R#w
zE0*U?z5)eSk<Ag-XZZL}dxk~&29*c}jw>pt?u86ruNNm+I<aE=J;@~BP8!pDXScc1
zPsLS#I<>1lW397$jWW!8Rz@A?`LWcwt&qc>6d$B}$8~+sX7PW&rgCKabe***AItwW
zfoq$;Fg)DNJ|5|+&%N0C7@JF@v%X`WWSlwirXdfgrST>TG`v8ZB1s}@xHYoEr6nlA
zVsb|mI8B38GFEA7c`!7OAEu!)LMcruf(^(bTH9_~6?eaS1P4LJ5e1+KL_x<vld_1Z
zH)oZ0IPdRVSn>Y)dDpiLq8c3NrS&0M_a_~?s58A8S-Gq~qo(X<wHUnGR?+wq^9-w<
zch6PG6@3y{G~nm?<G8p9=Qe!MpEP;Y>!RHeZ%`=B&O5eEIQ6S}&c^wZ17kPMFTW{)
zKh&{*UEo*om(zFyB{-5a$VHK(AB`7*a40s=mncc}l|kj{?-Pc_>Tpi6H<`c#MXXE@
z%t+Kx0s8OxGq#&q>=!nb+W>gbQ&2x(Yl8I9;>_O4-+kNg(Kj!DZg=mH!%QI+<7VcR
zAF3-dVX3KQxIS?65WBiB@?LV2&NK587CZpX*pRB?%*)zITX${xaONq?G%;R*!^qjg
z-Jx-3+qUJ~Zs&cO8dZ67W?&k$ux|FH|Lz=0`MFFd2<ST?L2?w-jKE+VgR(h0=)~g(
z^cqDckh=su3rczGb*vFQfS>EY_gDrZLnaIGSuIbsnQBHBKj*|d)!)*?xJ8;&Wx@ri
zL`i(SGqn77dy!?cbX7vFx8?%fXNG4Dxt7MsH)r^?{b|y;1{}grKVfqWxY6Pz8lL_A
zlVGeiyP2~0yQ!v(DCDnT<Cvd4l8=mgjMoXa$(8r~zcY5%7yiR>XVb!|<5DJEY3~7-
zDQb_p!6fILC$2={nNA-oF?ubj2N44xXnm1@m1Gb>bG`HfqDR_Q(@kS_4$SERa)(6c
zM{<I@5iFn_1LmcGr1J0Ezh1;M6aip>Gt|@=SX>hOu}OsG>VL!qu07s_QTr(g_c1S(
zjXg0oBWxv29+!$elD|%m@L%4mEl_sUv81SJRYf@XY569sr`5y{W*NO23Xk>u6YZ(Y
z+Kq=!pPyz<m@&r^kdWC2|AiBR%>Q<?>$Co!%jl#50X^44amKQuUl|uv!oy7XWjcVq
z(k?gb2Ze8}$)v?Dk&Gu9gv1#N-Q#aKT{^5W#hp5>Bc2kk-uewh<`CQ<s=+K;+k?Kx
zwg>Unzef_9P&|_DKr0+KM4qTB6@*6-^3|@Rm}N#(Zu!Eib*!e+g?T6M^<bs=puar{
zIUn(t(=_;lh#}7-2Y-QSHIL2;xW;G-ZUr^WF*4I~xJbm8$ZHODP@+?zfK!ZmF!&ss
z=ETy{v4X(yXe1{asW~w^%b3?Ofd7v@-;0d24pUUe(_jfI7}hN{*8-DcZXL;tavE^Z
z1hyn8?Ql(9-i#kn6tO^$Y^g}eB{EFmBJH<wSwT{0{I~wiiYqUK|J)XGu@zZclB)LH
zUqcrDaAIg3T9)L;NpgTHc`F{eL?4O6p!hkrmc%^+R=cYeyd>ru3ds`&9=IhI93hV3
z0fi^<1RPfp4A+fG;{U+hi#Amq%CuR?g_-(?*ETZGg7_c(vzbPD+1NRx-4G2KPE?d1
z{zGL?b#@va21Eyug567(Xq!+V7sP6D{40wmfDRwX_J7~7jp=Ueh%Vmn!R_il1&Gl6
zigSNmuy6nn(U5?qT;Lj!Bm)+cj7$9G$=}Ndnh2#J1Y`-Kz;Vl0#Mc_0iUJ}O7K8N|
zG$J57LkPfMxsYp!CH5VCh!_7ChKQQd8E<8)`8PjoNJR9mQBS98V5fwcS1U{6YzGhg
zM_f|=j<D+&HZD*eOo)>>OoIkyO9PHOdm;jsr6k`Tewu2=sej>9x)v+bDUA*Vz2)T#
zn1A?!MKsNTgDjH1t@b){AB&e(=Vp&fFZ!fy%g`>c+e+P~|L*y&`t37>Hk|bY7-g1|
zm;kW=@AEMCfD`0}Z-M`yf}q?)LaDNwQjm>X#z#8ePgoL9Lw<3h(}X?2h~gLyV0Mn7
zYLJjugPD??KhJW}xY9ObTglA5PrWW4K67sJtvLHr&to4V`#7^W^T*`L_eXZ04?ggh
zyuINS-)%}PUv|AT8>gG{9?HRhr31FpJEAI=9mqISe*euym4};0JyV)DETA4k;F*B`
z+<F(eoh)C37Q^^ktS{}}J@<rt?3#pGf$5e1Ir;sXZn#Je-Mjz&&-dPXF9($%a9G4G
zOCex|hCo5^k5|Je65L=!Y8`qLWOi5uc(S&e(`H+SM>)k+#W64?=ICl0QCZ8yvgul_
zfXdjy2-`_zA=hej0o-JL$GS((59^axlnw7H32W_pXJPZrb?aGldeg0So<CgZ+V|l_
z=Us1&(qVpbt2TA+y^P}hR<?10{zm#7+uF*WO$#f>)rYKY9?4t?Zw1)E@PS4iI-Y^X
ztWM42jo1%guT8{P7T=s@zC7Uno#*z*dxb+g&d*GKHFd$=2`hKiV0I&(%U$%36N8sr
zpfG?3QAIpL?WpN5uAu9njf{4J8x=&th+tvfg0iERaf2UkTwN}n2H1IUzD39@gcXw_
z2tg1gph&sCJSpl|VjtF=9o#g3ve$L{BU{VhKb8L5M0WM<@ah*UE`?NoSC<mze4IL8
zKl$cj+k^Z4nY`I~cV|fTxPf_hmETTYHq*8_&xKWskw3GpG~@s{-2(uVDBEzn4AKXM
zi%ughaS_#KmQKyAsg1Z-_&rzV%8D5OF~V_X-FR{1@mq)X&Xxp17OGPr^Tq1OhC|Fr
zM)4K*1Taw@mgF&=NjJ?kFh^_NGM=q0%rYjVN*EPt;Ru=x7;xrMZU>K*@M0eD1Pvs5
z&nnU*goUsv1UdOo)}qPdF246^Rbbe@h=7~=c`vt%%7NqGMbw8J-|Ex!)71Fg5#_5l
z?l!{L3%<8vidMJL+}E9=`k>%Ow796;t081_-e8hr-tp|>(Jc@tyiA=f!-~#AbkXxf
z0$VqE*$kjV3P^}yO@J*&Og?_|P*S%<#Fa<Sl1LBWTaHNTRs=JHa=1Ye0`i0CTG02^
z8a`RFL12p0_zK`0P<q&JnfY?85YMW?UwI(7UO!^7-NE$!r`S68<;G*hqq`Pe_v?RW
z_RjOuL!04CD~;Wwnc)H0B5rcm^-EK|u4yJ*jwl;t=JrACD}B4@6DQ^be;IsgbryPy
zyQqBupPE<To9cahK&0ki42TmCk8YhAHu$57G_Ey=JQQdvflmF%=Rr7tCPF>abRRk>
zK~QlEMPOcIv1-a-IBgN+8)~g~ngY>RaV%^s*9RxZBG6nbQ+9B)nG&X2>(na7z0Awi
zpPNCv5580N*4w1^=?8ppI=e*u#(<WvhZ$T#tXz{m_&FG0TB+ajL!(}@!%D!p7i47P
zDH-dsl=0vp&`!R_6$+0Z#~`H+%!q$HZzxo|r(1Dfz&o&hTM`YNUI*<0@0tRQDw8yw
z9{sJ<(Ixx>3H>DZNA55h$mfY^P}IZ-^a^*$XIS~|aN{{N>Th-|^deb6N8QMkMVqiI
zp*}3Z^T#MwN`X+r54ZsY_)c>&tXIOT8ONhrH9B%RL<)2Ru?(mA+K>bU`N(H0vn`m&
zKqr1F?l}Skx=ABFe}Vpj{C<OoN{VhZj}FPb$5%L~1~tYV+}ezDd;T8vk^n>s7?QYk
z7gRS1jM^;lGINXIHxW2RZ%DxLX+oYl#DJG=x+sd+&q9E2I2~=GuKYC5Gt4^B79H{w
zD7390>kB+he-$}0Ditmu7x5xDmKi%Ds7M8);1={P5sivNzzs~tA!zaPtKZEGd&vZZ
z!S4p+qP|F-wIDKqLh6=7$B|=((S+Pv`6g1BQBi&hClllqSsgL7W(r88x_Ai1qEmKI
z(z$zXQ0}Dr2YkY5W&}bL-T4*CA`$L+Y2trrhu&|ruS4&~2p*fyt#SQmLTnS&XaGT}
z$+#qk4un#p^XRF28iSc}6N3S26AJ-k&gE#T4+t1z)jHgK4ZMT1&>qho(?SQz3(MzG
zM@=8@LYKgqySKO{^8~)*w!3g!s8zX0NJ=s|Ft5tWSjzE^D*I0+1(BFW@93fce-vCw
zh2ylKx||_SXoj?@;Jt}p7L46dM0X!d(*QX+5ZDX>uNJ{P*(QM(tgD^2ujKhkfoeY|
zG+m;g$8~C%bLGB*Es#f|rC;z4{A~QS!(dW49WFO^v(ByQ#<`u6BvF@%ZSg|H(VXZk
zAXJn|5-L)G=8~JooCGvhr6ktL-Zjw715{K%yK)6|8?~Fo@ilNXW%d<+QDtBWsv?Q_
z!22+IdR}mPaBO5*Eg%)?4)HT%HB0EzDiZ(kj6aJvurftO|HVf(`5YSjNPPo}(O2*T
zcv!$aCgES63ir{BeD)6Nh{{SzL8#-wa8QO6UPY(YF=0&^^+_OAffL}k7dsH#LZ^4}
z`Kq&HWR68e$UDX^Z=Uf-%c$p^C?XqcJ(@@7^zFNiJzcvZj;lZ}8Bllc<*M{S{Go&I
z##p0f%(;s0634pd9T(xa2m+LtyW?OP#zYYY0)JpG@r{@irJuLrQ_837CikGG5dex1
zKdg<O?qGFVOun$JM)0$nL3Y%*p6)4GHKaOg!7~PN@;Kk`!GK<=e*g3*&Z)zfG(-e4
zLXMZjl^(3ii@=f38bR>Z*o^lJlDW?WBrzN9cvgm;UhFOs6IeX)gj%bac`-RLY9clq
z!9ApBQgR1TV1<P#=1iQyGq5*LKoJoLDDa}pv^*b48XLEW8}xMoKIPd5o}X5jQ-%pA
zEdCG;EPf=*cf#$_cd7?fs&n<9cT|qPGkeeZS+;idY->QaUAyUnS@wa~xgIC0PqIe_
z1TD^;e1C9H<h{~!dm{q2#$>_G+fX@TTT=G$t%a0loz&7B;H^pO7#`ZRsETXt%QSam
zx>65hJJEyZTt`<_APwL<$f<#!di>fU`{*Y;<Q3F}NYL*N#n@PVok+LvO$170nPws+
zK?RIJ2n@<rx~DZ0hY`-P_M%QSa`ZaM00&rW&?)p{t$B<!AkF#oZD@Lh?#7nLwxFTS
z`RV52E?dX_Z~eXEKDy6)B%jVKse0vbYEm~di#?u;Ypc6v1&!K=b_JLH^YtQI1ooev
zu(^E@2i_X6twxA{N$Rbr6U$QrEq$Zf^$V~SIr3cvJ$i#8%Eb6ZD5?J{lkAQ+Chpui
z@aXl8)u$z7gi%%e1{NSLl&u3)8Vx@o^i;_6279R03@LzggK%MDMpmfiSq-lni|Dj_
zewUtr=!r3nFR>rrh+B)a)cDRVJ{a~s%HOGutcu_1v$bwFV>^BC2H$P(&BQ+FPi{$`
zVSh2U-`I6Nby81~@9x7Z!zLa6;pks#o|H(|>ZSF~N&P47?P=IMzy1mKhO1ufnpfR5
zz}}gALLy8299HzIj=JU+Is8#Wo#fBTzjo+Z(b_byqpY)NV$H1tX!-|r@`<Mz@;&tM
zCJ;<DzXYER_mxl=A~;<Tm~o8GpjG6!BxY4KgI^l)5RXwT{;_;M@@Tv5tNn9QX_!4{
z+w{~AuQy(Q9cRZ^-I^5=lT@)>V(C$<?blxk4DKGp-n;1uC*QCd-}_T_a@Ed5*XFLa
zt!?j1jXp9`a_-$P1}m0-6LLIz>C8vwWi!W5=}Jl7T<l8f_gmhOWKLLIo)4}08?bIx
z1_w&As4+le(&SDI>t8Xn4s#liNnQ?<@*wLV^b08UyudmkyM#0y87?a@fTTx!aO3j|
z1qZ;6Wt<fo+94@B-^gQcG*}iM8`8963X1E?Y4Og)<WUD_>FszWCgnp!Y}SpYM;@QH
zI<R|R#63F(d`PH1HDg<G!QP)3m)AqYZN2$=LaP0)r|+GJRbdrH$<Y^`4W;SWw^$1I
z^PZWpqPH*2*gCv(p`l;5&)7GLr8P`J|74nB!WY+Uok;W&Y5rHjm8R4yY=-X&`h^PR
zdSXXAKp{(~(ZYD%NGc6O?i`2#EHjuGScnu9IQT8UFp;Ar_yQBl#Iw*%hVi}@ay0@h
z7CS=aOqf_u+A>@G@r{?(4vUGCw^p3{Y3g?^-AAfFJN_uU8(a6h@26v$kO#|szmBNF
z$=F@_5mlI3^9^i_49G^#wgfGMvqlwnFqFYGaST{R*0S#2VrG!~j8o~=b5B@0{RM1W
zT-tK(M$(yX+&2UyoILa_8_^;3nNEci=qcG;Fe@n>GY2~_fTPI*omR_YDq-s=h4T@@
zPDtOCRN!i}0y8#}JpBV5B}#cXMCao{Xs9vck7(obK5CCby2?HtF7I@vS?Ir!ULqI$
zX`cRa0`U>VnY0K^S&K|X_uW*{p-rsfdpYTd)Sc9Sq9kRbUki)G?h6~Mmex!7$zh`d
zJqteSt_Aj)NaPX8&Z*#CB5Ms*FgeCqpjhmbg(v&Ac&rZp_y}QQ9eEc`kTg+UBY=1q
zWI{rLih%4G#Ciby3c;$6JN{&aF+pZX7P)%Dk;Xw4cf1A!m5X9(mqdvt48CC+W<F$S
zh^G*CM%>6*hsD~0yDB&f^;qOR<&krPZ$<?iZpb~@C@PH?u5JmgvYbODN$G(TmIp7P
zmb(Z<9&ED%z!+2q&zrEW7&je+8FZ8i_7BElIk+NrE_N&gNdgX_Rr05~Ko}uMGKpB*
z13yfy<E^WYA9RZri@0~RdYsCX?fLP!I7CfHgnh=|hD%d#>|*-QUtf07)xYxtpTT@h
z;4xd%zVPu)ShtAHMF=h)Z>P>cc*wQ;rI0@iE%C9l?X6H)h}<iEy_=b#b#M5I<(lNl
zqXKm|`rk7LLNT6~pAkj;fnost!Ce9bgFF`!NfBcSpb-UOaa=qy=&;}rM572M{8Ef5
z;pzz194!qR&>^IDi^$y1AuZ{j`YCn|da-42c!#n3TYbkXffb!UtUEPuJbqsU^3H4r
z?_BX4%WQ|z=U*>9(*1C=(c*M@`C<uj(uU5#bZomsZx&_T5`)81xpg1zk~1s7U%k0A
zqzZclTxtJ#dWc`UPQY5qmb&KcAQ3as8u-|#cqWQ^5qY{IXoj(c3=p7XMzRCcxa6Ka
z#?(C^V2b9Zi1G3ufzZShdTiDP?g5DMFar__^r)C2FY;*f!qwPTNUDpN>K*2~vnkx~
zP_XB!zeG&hS~2=)@@M^6Ueo?~Z}gkn=Dx7z0_)58Ah^@tbMZqu2e&xvPUPyqJ%_{E
zzm^d%CX&>_QANL)8lUKjouEXY2+B?2{>uhRE86ecf*YrLHGI^9UHg5c-nR%&b3zqX
zj*FM>Q1-DEH!zYaBk@CvDzK!0cHn%8_(5(Bf9Z=NHxFtnKO0|>bFQVSY6zSKgqA>R
znhT-cB(ki?A|F|EOgDS^4nN(rsM<2l$Me`FZ1CNG%G}hOb>8Fhi1&$C9{O)-Rvvuu
zvmIvajHFOhY`roRy!zBy=Wu4RuM1&hy~IR^0Inq{2}Bgk4&uc?e-W71#Pz|$iBfj7
zfA6K)S($)C?s^gOsrLT=RDB0P%>V!Ym5XzHa&Z}bD%a8=E1{*kqj8n?AgfeTWTc`p
z(ojajHHb>;+DTGGR(2sxE0Gixg(N;1pa0|aK7OC?|K(iodym)q^?L5{8s0!1nTzDT
z(yMevCho^56rcgd01OmdL33#sPz_j6Jb+&s;`5{nEIp7hA!-b}BEd{z6Y>+Hl(;wt
z8H9N%wpv`t0~3|T$zoA&gjMg-7rJCOq;>O0SS{A|J!E?9_Qtqr+54qs%XH+%Cg;DB
zJAfQxkz?>&1#8oKTkjI@qy5`Tydwmn+df7hN*ayBW}$9cI09i%OQcAKaq-QR7{jN8
zlEa)#8b9e{2->R{6KtdyhKo|*$J;B|U$nL91YuwzGS`#CA*doqhD3zxW{xAuUc+Bx
z@*y%cxeKpBR1ydvP7VyhRN>fhsdytcJ_2w^b|5S)QVpw7BD_|}T)ba+@fgG3@R-e3
zY1yoW^6?|H6yiLwfA9VF-cTB#|K0&<fl^y911x>zdD>fNe`yq62U{r|4N)@Zz}EOb
zHpF>$7$%9Jk7@TuHE3EzXA9EaY@m(U+u0j1S22FtHQXU#wfwRClW~@Ad<bTA%L1(;
z_(OH#_!XW;D2M7pChjK&2>Bini7zg)KoZ~)5Du|nBpFGpOmOkx^&ssq5yXwen@b75
zSKg08y$R7-Al@jth7#9M@!BZt{;DB0YD6i`HczN{0QCVvv`jpTNGuIi{4T^xktNWT
z2<uosKjC($ZpPT$YZ6xbviEj(Y?v!oOD<XIJ=$46<4TyOg)M$ExBxjDpdr~Bass`=
z|KIqcjx))L%vXSsjPDZM$knj5<>FQr98wcJ{kuS*?8S<S)Vk{;DZQkjV2zSFB=HhO
z`{(WAHs2VD&aPGM5=WaMFfq{?mW+6sHo8LUU~wU*IKS(n{$!HxyoY)xEVR9i<d<Uo
z{4gxnc0bx<?~~fifI3ZMuJ=kE?%;Af2`LUB0~eXUJ3(+v><n~-M9+vsM*-+*bZHI~
zW45pllU5SC&xJ-M?l0Cv;lSL$O_>x-BQYI=+UA2Ry`&^}vFA2JmmP{5?gxw#x>T|<
zZ~D!3&b}XPl<>fHjAGq7*O!XpPOOTRsmG?^BA^(xNYbP@c)tX&gJ{^F!UnP_jUHCG
z4qHb(^F|yz2V)PU<|cIdV%)9iM`BB`@1=V`kW6u<(K&dKbd@y!x6=SWo=bR<i|&yb
zxCqIih-eJSh%k*p3GQAEKKAMmue^bGBT7dWqyQYy7|48_K^~XvvlBxb*i!O#gw@uC
zkM6!+E%CxozRu~K1|6~Z_JHC%cLgim^12Q2>JLsADOy$3_aPrj6bSkxhBSpTbdvo3
zQ!)WwTK-lY#Ruwj`ttFGA;1|;gdJZ--q6ekT)CJ;^JwQ2`-~hs5e6ZE+6ZD;R4=Kj
z>9Ew;sD~8FabQcvut$U`F--R3M1=*xkAwus7I6vuK$~!kq7lQe#qfX5@2>5$cpsv_
zWbUKNy8(rPF&-smO)T226&`aQl^z>6GXJ6LYqbqB;a4ITYX=x`;YSm|py1_UVw{m*
z{}r62)3kRs?SC??E<8Fe(_u<!bVT*@k2}bMKG<YE6St0-A9!LOKZV2_sDX}RuZIlg
z0CGs7x<_8X!7I4b`?w><*C)Z1jg*4(z>$=2l)P0ah60Bkh1rz$jntD8SpAQP2dvK1
z<0d&J%I5$Qg>l|m(d#<`bRNB;Axo&EFnx6CCxtp$EKblcb<veWi**C^uaO7=hk6rC
zAXLRq;F1de2w!pxwN-!_Ag7OQ)|X7JYkw1<*%^H{wQ~^08+N_>LyTcLDPe?p#09ZS
z9FUHZ-$<Af(TNk{H{pmgwibN^`ATU`%qXV8gP}=+&skEkJZ=~ekHQb5X)-JZxPSKE
z(0NoDiIhpL&*@PyuUzj>``k>M?c5}B1#2x_+gO}?_xi5!lVfk{G|w~3La)Rv01ZKO
zLH)rnMwS2DEf77ICRL(NWiZJ4t5Ikilk0$a%=5(LM;h_F{MT>>&cMGUtVT359)U7e
z(Z9y!+J`6Oq%D_5w^imRWaem^a3pA0eih<PJXVfI2l=UDJEW_@*bb@v&35|2^g1)<
zzQmi}hR#jJGz>@*1?sRB;y8obm9Et<PQ6<n!pS6IvnZ+=1VK@B7A-RjAmvb}#9x{;
zt!preaGC&|GzyIH01_fh`v9zRih=OR@4tjO34M{<2<pXDhaZw74wd5%G`dNcs>q4W
zE<g`~mI95F=HxsnSgsd~FH%fYk#Ua6#TY}l3J^3LC?awLJB);qk^Dk)#5j=TlA%0e
zrA{M_XW~XAR!Y$~>H&0%dRA)gn>TftM%rN@IS^JR04y#XN`_@pKS-<5XcPuogGcpX
z_(IJ(!c(Tu3-jZTo<c7rIJi?nD%dcAP0~Gu<lX;&kvfU`!ibEd7=R|jv^c}mgmb{7
z?j`Fm0oo!B;E-+suM1QA5LSUkAVrGIgAOa|C2n{aAWRBmC@3wAOOC|`8ZObvuP_)h
zTr^yIZUWKqPB?+{Ebx#NyhU4cCn*jmhDK+jwy-P=NmN^WjiQ#`@C5Ss{}e!3L^y`P
z81rnHTzg5ycr~S7#15MgJ3Qb}GtSeH-5qQ=;$&TZyti^jb=}EFN}WRvHSL1|hAt6L
z-*-qy=`ng2Un_}DO#N#V)`RTcs#HJZulYVYqUL!fUWM7t|6JIXZ=d4&Kf|@fBn0aj
z;6nLLI>g#Tf5Kp+GwF+uW|B+5zd6ZivBQ`qvgRE4$FOGX5l@vA*)zF_Szw|}0;>l&
z$s%wT!x{&g<XHe0`?;xUp!uC|LXFb)b`Qbajvk+GzD?=*M}K_<X;01ix8pCHoju$=
zI8MqaA#9M35ka84O!AM(4g=v9GEfs-eWK(ECSDyst`8}d05YKn3ri>i@Pxfk07~=#
z5(q_%M(R*v1cD$ahj|hbl3^A&0=#h@N|FOUBFjV5qi9$ejn;#pdx)y0#Qo^A4ec)Z
ziW;FZcNJo`X$KmbX$=L$Bc#$;8BrXvFk{2>fQ1Uo5WV^#htzJ7Qs?_%!-ywJ7JN1)
zAWLyDP7W_&;s5^>pqWF{Mt5NENdb;1px1|_KZ-e-ndCQliu98jJ&c}Eg}PIGNWSeQ
z5#^Lj0Le*)gR@~UVtNEfFpOr;(Kg3k0bDLo22z;RSknnzViw#R#=t(@%;>Boh4xDd
zLkzEFi749ym65*?vWTW~mx23fnKTaf&!87v{#~dRVDUKEpkk;3@)REO3b>RQqHsee
z2p3-`xF;Nk+#U@hV3=G00~U&5!hl2O3~&q3go|Vr85=l|04_)KB*+{CsgrOLEJ8GB
zW<tdcH4@S(yA<9QqFpq!3?0UQQiz2@wVIirfm)z?2_6YO<IjI3lN^$G9G7Yz-l}rw
zf=)g&<oA#lF78Ly{=k`-A?P7h2RTjSaA=Y&1jsbIMXIDK2pG(7AsKgCqzu+-2E>zC
zOXv~;4dT#o5YZgR2_w8vNCX1aVhTDq2Ed?eFxy2$cmjfwYOuMOPjFpafJO*CemMS$
zi-xMhfv($`qX1?xWGNX;RYh(<%Lzz`uSJ9Sq+v`tenuIwBQ?h%u)}0XO7OYKTn$ZJ
zSdytRgQgrlj*}HRDk}+N9rwB^ZHxrRBn{eQz#(NTk_B1Oateq4!J51-GfXiCttQ~2
z#o&$57wj+`|0z-&v{{0SkQo=J$ZgA5?2Zd_@|nxmlV^ycC7CY*#QzHXcybDjkR%KB
zg=L1ZS;;oRG+B<tM9C;REs_H|2BTu(F6F%}EJi;FL`H_G0o*vrU}D;?lGslKVIoyY
zn)F}0;aB0USQ{V1gaCi2Jkm2lc;Z5A30Mr0w%~yO(=`%qT!6ezB+*L@JQOAu5TzhO
z{iH#qFc@5PlW0Q%2~>*B6ZjE#PY0io4A%~!MXK(=ns#8}C}s$DbqXsalm0mzO{^IR
zn*;Ah;DSt9q9}!rd1Itm|49p5A^}Ldv#0|E_i+*8@K49cy@b6WwZH>^z&VjNLfA=~
zO{f#s!w3zrps7d`VOL7=xltTEhGyXr(teZ?aO1p7`HI%+2~67%qofh{O$L@^=E(Bg
zFy@zowTv8^U9s38cF5EaxngcUhensoRJBjCmmNo^eI=rj17uUw`Uk91q%*M4?L}iq
z+aPDda1R6w1K;AnG6hm(qd~Hn;^ZwZ(QWw=;tn9L3|0oh&|a#LWf3K$IO0vWFn$>|
zVO|g6DJ4u7;j+S@4?901V8RX5P_?ylq?yBzph=xvf|@vCY+5F20pv5WR4kq5l8CVZ
zBybV~=unrWIAPRl035jI0F0_FS%4N{N;qAphlCHnVU$%u-I$38Jq}PaRV2WZofmWp
zXb>x5IE|C|uYnc37+e|^|0ctAxZKfPCX9dBSCDuK2)!bV#bw$qUO1|h5nvRK<)vrP
zC8V_JOk0a6<mOc|GZ+q_k_D<mF{eo>yMWpu5dV6cg{TQ3JQNQ3LahYqT!bsQ5NY59
z*lJ8jRbt`%vDm~=ccq#xNhpgHI+lg~59(ZU$}pk^xXjE!g&-hFN#a^yB!T#K)Iwi?
zF093fn$GUUcd_^HU@+FpbmMU`#DI(Fp3@TG2e)V?0X|ARuojzz1Y<b0<zNowjuMt3
zg^&aa=$L?8L&`9W2xPh><gBfv5cB~_7>*u@@0w`XA}K4;J%B}g34{EBKITJMCh8w%
zE;mu)(nPdKB?jb_IFrqx(@CBPOX$zwg4|$pgOS6+sAds#hOx?3$8pBfC3Yz`w$R2C
ziYf%B#*p@hNl~$V%1l7uU`giw)i%WEJfDFoe~-7La19CpxDYNRc@OYbUc$Ihwpf)*
z+MAD!h0dtl+F_Kg5)~S+?#A7rH?hi0&MZ5B;o}`siUhU#F0>H+I|uUeKdl~T=CJo>
zeL!5`-QGZ*jEAEf_7>Orpf(C!Fv9RoLS(0BkILajVy$?Zh7no|Mqy}Svm|V#vL3q(
zW`fH=z?yVQtYS+WP$9BTN*L1356g*D&u-u;Cy8XaN^hjJZLyXcB@8oTBz#yjzCB0N
z7e<7nEEj5=k=#l4KJ9}dqlHO0f>A(ekfMM*)?>71zZvHiow=3ve#_{d5VF~FUPQkB
z6dNxk8#N>Lo9%iD@7)$m)OqZ1SoF`EZp^%hbtqnUN_|54*uX-g#}5}wvF_AQeZ6$+
z!s9A)GCwZP@zlLu5c6c4Vfua3vu{pd1vc_JiWBjNF|qPqg#IVitgI9aS%JjE=J1g;
zN1B?Q1>N@Wde~8PHj!&llI+MZS|-p64$j1bVUpq4hdM`7ddq3EOvbYD(MJcyd2tGv
zwqzAYG4){cZMw{Q2nR{gXnqo#CBce57WO-?0J^uuFof*=Px2;|E=<o(?e18w<FcbJ
zRPzZE19P5qt&AOIPusCDeQ*8U0E43syd~3?<}SA5VME+?MiFN|*S+;<NEq~0y4$f)
zPd@IB^0Be&&Q>>Lll2%TLt59mA+92AoC%gr-ud~={brB(NrN8xa%0zB#{yLwEie({
z{=oJ**zYXjaCsE^f+9A&*so{KcvnMrDg|KdZO3E(vK$Zw6q%$YcZsI28r>CEOAIF9
z0V0qMnV=IQ?oluSa&giXdwW^z$vzZnw8RiQAd@f)4<WJx)E(3pmm(@U#?Dg?JLJ0*
zbOjyb_SGlwVj2T<j!DZrxwRs@>v6-OH>Kl}*O<SOd#Ic6gf`N_cHyzH>JGff7yf#R
zf9>1$w3)TfV|iq~^z9LqQVWlpkDYd7PCWudXAfid_OA2?qbfzTuEpg$Yd<iXw;0I(
zWjdi|X+!55SERSn@88a3VwqD+)JcmU86lJ39&GscT`)Z<T<Wi$<>$|Y86;uSB(a&?
zOk9b?-ck&7TMctY;Y3i33DHdi1tlzCc6A~gNz=N;5m`hZWwVCJB6d<(k~J(Jv1tuW
zjs+EvVj;Vn39$(s4Zp1_xW^43fK062pqFO-c0^5a*LbWf;hZ>n&?9zxOz0xN4SIij
z)YMkdBoZD;j1|p4F}prkcS>yZh=%47sh=NsCzk4;FzgM{J61O>Y_Uvfm!e?xp}a;z
ztjN{$<^&hxGv^tY%C$~@kZUox72D_bx-H5#x~<>8^K(IFr=L=LxBL8S6^EW6i9L2*
zw&wfx%~I$PEs7=<;u6+#fBblNpBb`(s6+{_8W#yxpc{b5gy;mYi2BDsZaR4c0+x^i
zN2wFbISlY4n)xzoHQ8Zb%r%YJVPv|ZCYEKvvaU*wG<;~pP#62U%Gf=1T#(-j?9;`y
zg?2sU)QmVLBBF`+x0Ez=SFi4HuDrJ@`)U2Pz<dmC_r>s#m?!q~aib!(etz&~ad6mk
zZ!A=3Gt$A$kk#v^kKN8aYMP@*?5}()A1_<q>8tbi*lDk2N+)t?ubd5zZV2$0KY6;R
z=c#=<KA#>IZZ$YscW3d_gdv}Wigmd<%k)y0dY?r;X#>bIW0x{PE*w7+B0+gOb1cBI
zeEmrd389p;=rn|cFy}>lmE1w&00t0cfD`1vIJ#1QG2c<9JCY748JQ(kca|$HVTPWz
zTW_8f21|s)wUy9hCSWkDB%4mO6>LvZu;M=lk>kF%IbpQ@eCXM+>+VKpYj0C3|1$YO
z<eX#HW*&R<M@SS`udz2g`ffcfSH<1xZFmCqBvJ8l$7)K;yV)PN%u9OXJul5>q)@)v
zW|YUA@-8GeYU={$vj2W~G)P~L!<BaH-hxSqW+Mc*FQ%yogcW8Zo*1V4rp0?_A8yW@
zkZ&j-*U*0<BPPT!qt1-t4F)j-2o2EmEFn21Gm?o&gehpx%;bVb#03q8)j;VG$VxUl
zERv4hfoEXs7Aq&-Oo+MD+$c3}2u(zf=jRhc$ehM!4lcgWKajK6gu}!@9a0axOme|R
zDBjUMg(k@XHWn+DC(LoKd^5e=eF66UCiB!@=qqB~+eKcx>`nEbNbr&d{SDG$y(1Rt
zrb!&_4>C%SRFsZVx<BYpU9V^nY&6kO0fQB~qiE~_TU&DxCPBl1hae-YFDYb8(aHD%
z6j;e(fn{w^F@Y0TOz;e3VWc=9U#dvaGJ$i_R}KsW^i`6@%scwgf-O5{NeC@R<47$-
zim!Ng$EfV)aicB{5}Ba%2EDku!r8)=OOsen07b^#aA;f_)>P$i^f4jlof}q<2a7Dc
zvt{xA=O4#;kGLT1?Xb5Q3ypg0e^TdjdQQhh8Z+1s_K{ISq!C7xG?4W&_~hW`Af~!Z
z+e;fKdtXl4Y(zu;^$>%EUWY@0bTpO8#p_6c5zPi&lS7K`a7G&ad^r~*hQtyLZi|F3
z8nu_GJQ6b?PjdkyARBoJ2LD=KW-v$AVw|dyrD849Tq8wU!j_#SyOuSmAt9ksW9b+f
z@1&ezZW3h^5QFJT#CM=Xr344y;)1{_izfx$&!o|KFm&NjAy3GK|4LXxT77ZD{ocjZ
zm=R`ot;5F%Yi+7DNZ+2|p3sZ^qs>DMxFZzSLoMt(D^uL=Gxu1W_u@GvltL1B3v?+-
zNopF3*cCt*+aG+Mf$@Au=McJsu+d<{EZNJY(3uJOL^vIY#5ftSA~Zhw(jA>3eWp7s
zO46{Njf@a;JdHJ86l|oe%-~~@QMQPqQKT5A0j~-FNIi(QnX|-J&QHR?G<;$Rc~J#6
zFF$OZkz)KPm96dVJ9N{eZ_PPwGtFW7p?<FgQ#FQjJ!~-t(_{YL5mKA=GVbTzka_;m
zXU^Y9Cq54(6`*~jkmxC#eX#~67UF*xbmEeMjXW++X-IMK4HJ?EH;t>I1Tw?6c9}Ui
zhapX9ZFtmykWq;bu)N+t2xkVD21(751nI*Z3sF<9kBnC}QtYPVG1NsES_7^w^9&a4
z8xSYyo<JSo6NT$kk)k2)G}417z0RXs9$R<W*?7+RJAR~x`=R$cUvItk!b3m(klD8V
zPx|T8r<7tr>+3-}3U#KR+1pN^q@D+bU@Q+MIvA9kL5HcoQnapFbq_vk<gCDb7-~TY
zT7Z^}Wrur)v^^43A`K3}fGngW`hjjUCrW8DcIYEvhQ;6|D10kn(20FQByB{SLya=r
z2<R?6A4b!{1F+IzoG={PFx~VVo@YShphM)yo@B0g5<F}Wa`Zu-&few$f8C^;euh&X
zteYT6>vRbI92@N*k$-m=i#~mH<p_zvvFaXkN)@cP#+|=LtOSZCMfd}pal|~x52LD<
zj7Yj_yx@LTMouQ7H*o>>I2<J&Zq?AxW^l*Q2p%Od)fP?+{@OVvX!1Jq5}$UnKC^92
zfK*T&2LmxUQX;qtojHYc)PpHAm~$$wkS;-bM&(t1qfE>a5=`zGE@?^L!$o0+sY|01
z_cjk=%IN)RuT#^WV8`J5L!Jv#{|dXd`H)lo5Dm-Tbd2?GToj!D#)q~~XMGs@LZOGr
z1!;f^AXdaLq`o!u?nN)SnU!Ko*C1k+VlGe@TsI3E%@L6T9#fJ_s5Fz{2S;SLN>&&<
z(nyi4VVe+1W_82(qH~PG6?qnjkJ(DulCJ$98_*;;jNttL?HKpB_r^i+98}KS`FV#<
zTI`(()$NU*y4XA>yC$vQO*dod$PvK<54_LrEq~2?4`O`YfrXHZ>#&&!ccg<O;VY_N
zsDfIl6(J!K?E6zbQHH=d4FH~wIWNQs$Hvu+(R_knu070{Fk37@Nbp6VLJD0<O6h)%
z(ZrMHi)%xS-o#KRS&<yAdX{}X)+@dS!a(Ak$)Twca>3gX{!@B$lBSQ@2JhTxM5%@$
z5=9ZDB!(D)zNc7h!&YP76MU{I{^fQb4ndTLKK9jfIQf1cP)VhQGuk}){hJUeY!^X4
z0RE@hl9xie5)}s-0ZJl37ngD4+v_fI%Z3ro0O2F9BteFuq5arjqtF&vS0Fc2l0<i4
z=Almry-G;%q!wT2hKv#>l{I0CG>8*!DmEUI48v-}VO%Chic6=-E*V2yaJ&pDLbYHe
z2^+GjI@v219|XD}`SB==U`uKgHe~8#4N5Q_s=IMo0=ojzSowBj7A+n|Tks);;08R2
zumt$&cNuZ>s<Hkj#;mOHFbN)MJ3EXfg~jI!k&B~4UfQBVbYhrgQq|x9?jFl=znx)*
zbaYnOavvL8S{Py}=rEJZN<v`PFpMvmG*OC;paFRyDM-biR)<CMA{QGJg`r+V{@@s!
zlU7Jbk#AcrzNjJ=TsWjC_77iV;zSsfh4>ljMl6`5b9#>C6A@TAimi(K**P>xM;Q}L
z5XVNO99cIEqO(AFhXkep0lWhAn1HQ+4I4Cj?RY$oqne~54#xjv6>vu-UNa1HXs|$E
zNnjKg01O(J&cMnK*f70-Y`;U=L<h4*L7aq%;fY-sCjtdjV}I)vJO&j>-66$6HC$4a
z2AU5wjq_9)H~z=_cQ={KeSmsQPcY#ya9ToOj#X%+uxA6n%>u)7vFHqlMqC$cujs5K
zFsE|FQR)T6FOZB>>d+p?845vsh<b}8JrnaSvFQto78a?bMUOJ-!RXs_;+9MBNL5IO
zkg74-8SjAFCPl0faUgCvkN}{GcYT7L%7sw;g9)0H9#-=St~HwIYM+3p5=AFrWHstA
z@CfV!7EP0K+A@KDfSn6@#p6oRpm{ko<9~0`W&yk-Mo3~O5Xlh|5(LN~2bmG$l~{8*
z%CqO_&?j(hRor-6<}F^rS7rbE3`oZ+{!)^!go5_)g(9Iq_0OkKFjLJcl~YbiIAeI9
z-><?wyUS&YWvU}aj-9MFdg<>wPaOGu(4(68C3r5+ht4vSPb_Bf==e-bK{@<LxI1TL
z-><^med1fy1;3(e5>*QlrwQzL2jGS8$oa$;Ue%t$!cD)5g}YPftv{Vywl`3fkgE8s
zePbujcDrM`NARbCGg0tdd-xbjOY=tp^>?ga#)XCb9pMZPKCzt0*>2Ubp5*S14$Fi4
z$t!9SrvcVc<T|-!&afxT)s@F^i&R1#sD%?gc5)+HGA$r{(Pe{wX-+L|qCc;jW|gXh
zhbt@{2}B~5NOP)VNv*=&hCTW<szRH`53L1gQ}SHzILBl)E1~G;yHHf}$a+1eI)Z*u
z!(u#c7HC<d&h>nF`**#800c-OAxB(9EJXD9lZRtS1LFZ&o3RrU%v#T1{j$pMr6hS=
zeI;o}i#9&u!e^15M#u^D1;!#9Y2b`>k42#g=nHRTJPHVJTokwaZ_=a3Z6#J1YRd#n
z@g_Vsa-L8y*AtN0K|QBGSD<<X?d(w%ddD$T;snE!RXS=6XY=SX)qqQ)>Q>XkL-r0I
ze!8r=D!&1^wjhnd=Up|O<yJk-DuE|gO`mwU7$+0!!xwszi;k!s7eEnHwTvy&=knwn
z+DHX-q0PkK?bWX*mnoV_;3+KQICk4u00M1FKnvftvNBzB)32h_FHeeJruxH2e4uXi
zu{}vW6TlNqp6y|I@ZkP6sp2cf`KXOQS#1o?IjG9q5a?YkeD{@50M$=MPiIXAp(&3X
zv@LZmKf#qp-%_(xHSwdTFr*=O5k2uMI?XaOMf-P4-BNq|tLW6Hb^_W(2MEQ@c1xPc
z@921K7>xQB(}jX*JF-ua_U~k<bXL<feZUAHt!jdG)H>+6oshyee*!D&V^bZ@=)iHt
z6m=_xkf)%jUhw-;;^1MC7#4u22Dy9MZZ>rSoYyE&KY)iOb0VY^UbY18SFNl}Ruedg
z$&*1yg6iL*aK;>wFzP0#Zt`rE!%|ySKXryTj-3pqnd`kw<&|gran;1e@I^Dwf<f<t
zs+zvw)PzWNi@1Mjy%*cA^b6Dl>Q*}igq`BX?v5f&``B5-bfxGtpHHZtG|eLQl`tUP
zpC^?=T2r;F8SMmFfLw|b?+DrDp!=;`!Rq?FsqU&j06mH}DZb1&M<~(9$Hy0>7(o{b
z4A}%mEv;Wm0eH{43%FTjuv-P)B_P`YDkWrV1|ZC)SSi;k^<6_-i*Qe#@B8ls>Q<>b
zkHL#Fx5(?opPe=82f**(0=$AqEj}s=c<79-rRs<vRmh7^>Vw?PwmWAhd9;{n#qOwZ
z^b@T3)Xyk8CJ5S$2TrsmTPD7BY#pXU#1+fjkP>aNu?M50s9AkcJ#n18Y)g%5Vgs6)
zKG)l-2Pfq1LH{>!Fmm1<1%)<?xJV~EwVYa&Gs@dT)^6oBo#Guh{9}5utSmkCf?cDw
z$@%Awnu{FXi-dx2lbC}`EQF!a7qkO;t91ur`wENII*gg$91CV<$OcRjy$}|~Z}~cY
zSB?L|!)wx4mO49K_F`J6e41Wwru@V8_@sUc$P7NKn!yjqBjhaxvk!d(0GVi{Nx+$L
ziDgD7U=<in0rB(bljjV41XQ&=YtmIB6yyvpZiw`M-hD4I$>&F5s@?L@Chk86Lublz
zd=Ivjy?U^0`*uB(<i_~jT1C=d=T^*3)?4|_W{XuG^8!5}+gDkrJ1{-d^-_VoY4OKd
zo3~s|`Q_YP`PKi~k@2fT^D4{I%k$qkf70?=BY8F1S1aqnrXa<~F<p}{wRxX={V&a;
z^!&Ctc9z1<nx&=3PFee0mhag-{K3v(jmo4Wu35#Sg`GqH(vqLAKD+arsb6D$)${x!
zex8=^TaV|7&ks9_x=n)nm3A97%XS}pTySRhF@d^RXxI=m?d)l<D;0}Bo{x`Ij)-Me
z4VfCDO6Y|@`IT_DXX3}%9yl{?hsO4vtLq$xZhEVkx=tJSw(`+|qS4Xz%Qq#j@IAf5
z$aq^TuV?*nhYIcIPG8T|U;XZ+w#4rE4&QPYZr!h**%b;=tDOf972SXQ{Nm%u(@)B{
z*Y)3e<FeFjmO)iT@bU8&v4`(59`i5Mb_wjCn(5jWb~z@?_4_zqNcDCtREwAKFn^e5
z<=Y$8^|AHrmW!+QHHV+@)5^K=QTLHBP9|%Pd(B@B_Dq}GanCxV_7<efZ?-?R^810v
zL(7&YKYMy#qhy7J&(<G-C(Z;2iA+=#hES+Bl85SvQleB?1qkGhyil-J?PlVDh`QVZ
z(LZUtK;N?_;}C%dMIea-p2FQ@aUE)hQyp%b1d?$uVk!)8O5QbpL+`o#{r*iGi&l(g
zEXr8l*VMe@Uc%t_^5RSPl51nFH_dYBoTS`UTzi3)TUIA`YiX0RyK8C)w_tzbnWgpR
zZQ6}hUNWDu9epjHl)dRbkW$ER3er8j*I~@w3R(L5p_N8Oh06I4E_(?q`l8g?1E>5h
zKH7E4q&muT;_EUkH|yr!=}R7EWv?z$vCXrdx6Ny!kbiEJXU<iZxh9)@ni{@O(#p|D
zcD$IZH!&@*+-cJ!iy7zJjo0S7g^wXR;2VT1x<j~FU=IytxR4SL$q+L_c6eiBBh=5y
z{}h+Bpc_Ih7Xla{$SD8u)B#okV=+;mfQWiT(_ovj`ly1ZFKujII{D^a$-8|l#P52s
ztYTb4zJuo{$G}skjz!IX{#a?L`@`j~ix#P$SDxlmCev5BYlarLbMO8G(k=e+ZSqOC
z=eEf|Rj^5^cC~Lio&Ub8%Pu=8IjiNvroUC@2fmxX`l7wHpKsS|TMt!NpV8J*8?`di
z57%r=O5YZ6y6W3t!*;2BqqO4O{7Yl3SMQf|Ogwu;OFf{yFrnD<^tGUtZg2E_;oFn+
zDMaOi=s^ULN~5dM1M>NCSEYc+2h!_Q^q&BUqJtj-^=xuOV!crCiimKA%7N%>Uks!^
z^nCvmazZiV#N}?st;(uNYvoe%#^rzd+SJrv_|#b#rLMZw+1lnug@4;cN4Jh6o*$M5
z2kd^`bp3qzsB?CQqJ0OG@)O>9u5XJ9Oi`%a7$5$tWxTA7^@(`1i_JlueS$LGbMHp$
z`ow?uA>i2s7FmU)9|%;-=$ItZ_0qR)m4DW_Gq*%w;TIoX<-~Ybqe>W^GOudVJ*<e5
zhy*=DOHcxFQ3*rk-4>6I8gXY0?QR%OAw3)~P#d?mH;9{;_u*FWPEYsQ2g5rnCXPI`
zaqHuNGlg!MY5g+l7Bk;;oO=Ct+l5UBgf5;_|H->>`oR7t`Kgf=MWaK^Dx-Rgn5{Xs
zR*XxhcWPT5>aShpwXA*h!{@f2dRD11eP)?vZFOq<F>RdnrDw)#zuHYo8FL_1-}0&J
z^sPC%OPoS%uAG?gZFS^er;sAVqcu!Ev6~a1oJQ&s%vh%mPeH4HD?lQvMM8moXHB|W
zxNckutQY89n2IpPaAwiiodn7Gfz6@)uKT0eYkdEu$?G;Q(`8#&j<#W>waBn|RXd#A
z{A2cLoAa|m*XAE}-CT9xbZpQ`n>(ihWW7zlZUP?<E=`Q=Z@U!uIQ9A@?Nfg5dbc~=
zcy_i~nBDIC{a{fWgV*!J)lpzJkaQ`{X?xYhvCFp2Pd40BBS<Z<Iy>)NK=0gF_Vms3
zbe-A;Eeh?HdtYA-ix9F#NK8#ABnaEt{=!IL%t8nZiR~|$;WVLjq6jw2gqFC`su>KG
z{ej@EWh&u<9y9q3at-fJd-Z>{bW)0)ayGj8wED+8K*C*<>q%?;w+tswT2i2KAU}EP
z`m0I){A!a&Rr}5#;wLsZtg)|LH92O1&zjrH>_)S5jC(<!S_T$e)Rn*Tq|?>Q<S={i
zyZ-79=_1VqiYFqP4OHT7IvOkv&Jvh-lynqa9rc-KJTuvFLk|0guaG}FE_>D1(X;ZR
zIR%pIFLz&@^vWO(yg~Sdcn3smG%GQb7`-*BKQ|^0euN7I!%|@iu?x)64MKZ|CoKbh
z)K9D3x^)wMe<wdL<jYIB)n!FmFPcx>S_Swoy%-X&ygj>iaQV{JrO7F8ZXXh;=ATzc
zzN<8M!Dgv7lcV!qW(^gVnYu6==4WKzuIo<qaC1EDn$C=?n55NZG(X<&?qH3{$MjJ_
z{zWeT$jz~fzdTd^n;Ac@F*@F5b77CBk&OD@)roe&4Qi_suB01Ji{Em&c9HhV($Lsh
zCfj+*+PjngzWYUcg6js&kE_7)Xmv+ThT-@~V#^6d1q6eXi4_Psgu6s_o}sdmGO*MS
zED8^4#OaUfl%xQU^$PTPbf0kBN|=H4!2F@!g5HU%c_UcsrrvPnYz~6jSdeC=FC5sg
ze9zx0f*a#aXIG42Ip4@mvOkz}Np|NJ9)D(I_AB|br`XAjRsXC%_jbCRx}Q_MVCPlW
z?=!C}9(TQe?LMQY`$9}dHpibgd)4uqr|wA>{Lod+IqPy>OUb%;dLggBYju9joq_F|
zWo^6ooz{a#J|!MYR%$6UkGrvbYfIoMxxrOhRmuG>Gtb7`RJC8dI(G7FN(_RxC?}CG
zGvJw_!PnnH1fsn_y@;Ta`hciy=mD9KiCuSGAs83wDz>?{au)xi+3y&g^-!`{cGcDq
z7UPl{zMFal<yN$5U9KHH{gfyBD97Ve$QEPY?v|+BPp6L<ONYAG>W;H{X;Ia)sKv#y
zY15HydXM%0(d>Hmc%`29;-(F2?&M6`7Ep9P)lO>hg~*vI7Mc&P4W_jC>RQ)kJ1Hrs
zgx)N&n9+W^OXaVk+IT0eMN^Xv4%gm%y;9BfKxbNGU`gOcy$+w}PBL?Nhd&l>`cA}n
zcMa)74^a522A)dPt(%FBLdv!U#S(iP<~ilb$or$W7C#v61gDF}kGY}RLznrO8S7zT
zbn&!c^Eut{rU&7wdi}y9yL=b=uVZMu8aCLFHKn#IWcso`Uy;6)#^%SBZ_0ySwdP+9
zzgRf+AC_5urOP7Mi@KE?v>z3EroD)1{rJ{Zw|-ws)(hst<mGY>mb#~3|8m=`x%6B2
z<+2vHg6a{=RaT6?VRuCPg5!Pjf-yg)JyKd4t~<e{Sl7QW+EZqunO8kG#<ullUrFsV
z5_>p&^q%J3=6yfps;)fNuci~ndJd67Tm6a1A<jNjMAY4tD_0U3^xH_Lx@*?#Swr6`
zdbQLI9{vvW8BWNnx}?yZ45cx)rkVTY6`wI{)~p|j-?(&AMbrTm^SxERV-7n=``9G+
zWH4K2G=w&OIGz7|I=w&Q7RNMc#l2G-D-WICkY;sQbMm3aPu*Xi&2$Rb@bP*1?q=&f
zKlk*>vV0#3Rwy?)%0GXmn|EG%$&&76!l7*@h5hamG@fcU$*Ao8$L~Psj-jNXTWb$m
z4K)wNZ%z^Nw>qYp3i>q4n};6yEy-0|xcyw-ITiQava4PqI%9pIY1)o6w^SiyiK8a&
zV@{zfRU(z>pl6DF5sbJ_4tB>a1VN~DHuMe36mM8-5)-;!ZBBu_?0oK<m?zp3N{ghd
ziW!j;{hzt2={c88RB@?dKGD{W3W0!I=IZb?$GJ9(d&=mH=BuJ?p+oD#^T~PB*V@?K
zH*w01Shl}8bf(eSR=HM(*V8?H`8zLGU4cnS2|!|pXiAA#Jcacf63VeOr#x7db{pb(
z!~4oZqbNbr3;XNKVAZZ&`P1zzSrv5i274R7aT_hxtBiL#I#fS&``kjOve4k6`ZHr#
z6;i&MYnj5s4e^hrZ=$;C04W76L%FYrvg{HOLfA@?{>Rq`rv*}Uy%V>7tOkRVcOx>T
z-rE!OtN2tEF&aQ70ur;b8vHC4iK>Y4z@Dxlj>Bw^2-FYwapEhf*Qus^N5j67YZnUs
zX(X;-x<((VX*uZ=ENI*f0}J7oh1;s>GPhI{sjv*46|EVORIPzG5lfxa7rw}PG#+T0
zX7%yTO^P*DGeodMVcU!sBpR~PAdq%069<?Ch5=Q3oSi?wN>#%{j(=>~G3M_mTldc?
z!ug!|MA8(}0U`OV(uE&74RxiwkqiC35##fLE6C&lGiHcYQNAwKzqn-Wt1ZbfoVZAF
zKZQ0U2^2_68UG~SAh9#lT_hAdLf{}{>||9T(GHyQ@I|nc4JqjfuY-DT23yk)Fd_c>
zR4urq^6%&}LyZJH#?r(?rygT`U&x6?<j*zbWpC}}s}5A)%!a>6mBd%*GG7-`5grn%
zL?2FBZvNu6>gvfF(zSOd9(UqT4N*EWkA*u#LIEr1{BT*x%hGgQe{5(cB2=LI6Je;P
zYJ~vOn4+`qnr(2h_fOyj6T3*yzBSs&<8`|l+2{5+J3hI(obE&4N${eY?oYrZ2l%?K
zjDkil0N99SO_v!&XLrY{#^h+9Xo-Jv)x~f;=^DH%xEBm@G@_I8K0pi!gey_MA<L}W
zpWFue(76#Yl5U^!u=mc!H77JctrSfFD3P=MkCF(8mPOi~IYM?6Vm?ZM`E$}7=qz)4
z27)uh)*v7hM}#I=gc=tqRD_g~x<SJ|bP~V`w-sBQU_cT<W0}224a5vpx9Z7|pWH#Q
zCaz$VL}UOSe6^RAcXs$pt7AGvA&5ocedsIvXeEZ(d#Z9`trFK^(L`e9@#sqQ08~iY
z@|#Gyv8vvewSB7zSkY*I;*yXD^b;tPf)%OmDj;DRxMYZois2&j$(Fo0bk=)^LA+Eb
zFq%PZjgAfhVL`<106Ugh<4M%){qgE~!ogz+!-8eeSyYVPA`IL?_%1;-k+QDMo@G8b
z(Ngl0qTI)FHK<i}&@FBT2nkjk+&>bWJ8_A3+_-4fmjts~gqnf5SeS04KPU1iPe^5u
zKIQCi#Y-&+&KlK3gh<gF2Blg#1Os~<$n$Gn9s~_iVgl!^evF1IiC@F}@lN%Xe;O1S
zm>J)au<l{+<F7a^&>41Q@kmr|;>3w5C&e)`n9JV#Ky!#)!3H6*x%dWk0gHO(JE|sJ
z18|5mVRo8b2MXaRK(i64&!caxOVz24Z`tBudg$rT6%N~5c!)dy)&>Yv#_8dC2V+SJ
zFX8jz(gxka83-mCn+PeVpMo^J5}q7^Tg8H1h5$n<=BiRwAFd))YiX`Rd@w4YZ)|4l
zO^zu~B~kS^&=(VQ$hlohSm6D|Z&h3HQuH59Be#L%Np!;8o&rD2=+_8d?1y?9Id(j`
z2+9xFZ^WHMVyV2UF`7I~wKYB(F+BfC<0%aX?CApJ8=T_n3;PKL5$ny>HHr}E4bqwr
zf2v7@NfCD^l#r5OcBeWb34sleoc<kW0QD@@ZnWu-bnJaCkTHG#{f_>~;>^=>JuR!I
z$~uqPchxM&d($buL#@M7tA9?x?If&tBiupIM9^(^Qy6mL+iV^L$5ds;a4O*eo!^NO
znU4fhg!CEhysF)OTMUloybg)EujQ)lE$!qo?Zv1PyT~U6t-0^7utG*ZeY#<G^D1?%
zQsui@S2cdhcdr;YFmrr%mt0pxk#VNe2N4)zte6jgXCjvZC?PZ3?KVhKm1Xt}Vga5M
zl#%L*viOvUpr5VHAtw3;3Y}VhIi-cEOE1Zum)*C~zk0V^&!n@?0aXKA491x>ecR?P
zcZ6{`t9a4V$q7y?pBX8gT5&Axiytpo(Z$*|)R<fJvPkvE6x9qvqDH9TY)dD+*92a~
zNQ7F|+wp?{qNjVo<~9RHomWxWzB|DOj{9t!KD6ltTxP$W_nhr&rYX(-@apJ}I^Q&D
zfox&aGsmZ9>rV7Ho)&(wd#9aqrZE4!#{0@sWf>3awZ3*&Ph~QOeSk-Mfsr8z;+pm4
zqjtw<Pop1{d&&G-n_Ihr`SbYM@bg>0uOIS1)qP#N(8=6ombU60dT0Iuzokh*J-_<2
z)Vm^OS(|-B?`R2JJm&A-yDVIPfXF^ry%d8t5kbL1KlEjODu(-guPvw%==qj!IOCu)
zUe5Xb@g39q<u8>74HfFG%YUC`c_eD5aG`Zw>r!_aZ<B!@yDUSOv3(O4ecpE9xLy6D
zw}Z?}_K-wk$xN&<L|HvfLCF#!PSF8!J->d0`+b|aa^1}6Kz8L}odK7#TcW}jxQMcX
z+6$k`$h+z)DK@f)Zocm;ugnm-=A<@OpNiFf^XPhWgS{ndNvyjuujr0g{l=ZM`tg*~
zOaQ(q<JVr^{2%NA*$IC=vk~~Kc+HSeX)iii)7oJ(e#rUN_z7<>xK(MDUNop!uu`=D
zx&JkxMLF-`gOc%fyK_%JtEQ9G8;P3NfTp52@e!!=UH7z?;cq@?=ULw<n{H|_^ZoQK
zceY%1|NFez4Ap_`vgrqJ%6eE<R92bD+V^?4Uv`%}kacfKt*pX>!Oizx%kJwMReLe+
z(+ZWH5sfPyTP=MT94%JgoL|&>l9)lMoVYZZ0Zx2(4N_gi`{K!Q(yz#7Cp%c~zomOD
zdHxx%yze`v`)x>hTH~-`Lh=fIzGbGUXS&~7OQ%Yap5>@>BWGlFDYlySt!_PNI@2-A
z(b3U#Mpl8adv%n%!A!gG1Fb11HfQ}y+q%NbD1MoRxx;|ZuRi%wry$K&;TI=vD)^!s
zx}_cZv(b?RSQtO$7{kk7)g|zz+Mz8-gVDFNlOU*SU24a*rf=UgE_GSglpSM67*+%|
zJ=t?U{QSDS89pw4MiHg4N2K!(DD)0(YRq{1&%7nET0@&|&5XKCkD4B|t~xeqji#Mi
z`M?YPwx)B*^P5_2q=&VRuJ~m4y!@Qgy+kRaXM3tl1X>H&k5*(iK2~pT=^Of|r*@Yo
zvNnw2^$;BR{4<x5nk0VD;>PTmD)nSVbeC~tm+X_gi@CQq8n;!i*f8J6VZ;3Rr)FhI
zp^x6T<ru5kXk<Iv7F8T)Ur*uGxh4;tb&L3sJ=EBGXPNag=l3m3HXIq)HC~i*a)sBd
zzt68Mz4hamdXm%KpXEVy7MByxH#_wqKche-*Uy9SssIv5y2anVO&2z+m<VJ_&u!CF
zJ7vz-`=+Iuy`^nx%2}TuuBTj7&mGg!UR_hZc3U%V60>ZM60>Ug8UL8>nd6<ljCZ}`
zzS3SR&*;Wu_Q!_0s8_`c%98GMJ@UVt5_@lVlt8geSLxhI#vJ2mC;GIun<&0_FyGWr
zlDu8#N^aif2-Tm82+HoJvP9~=5QUFb37SYuTfM9N*P>H9++tnypSdcVrih*hkIJfU
zFxeEayR+|FXHtdja=U`jn>)2sD=IhFUr?JNbNE_pw9nK;OUnsnstMt$yPTF@b;{5{
zUx)F<g0)_T;P}3=K9v0mL1)7kJO*2jLGCzyKNUTP`3LQU=%X8@4%*}Qz0CFh%s<p`
ze0JiBkL%_wohcLO>)oGZk}m63X=c;n`2K2cHC$;?M%=uNGy=^?bvCFgai9}I`*+sV
z=&IF2&fvp!mxFz-Ot3t=o&HXFmaJW?=6ffZqXB33L~DAgJ%61-4>BuOR?g4Qp6F0m
zGNJS>H=<PKte&;(GBwNi4(A@bv&#9m-oEm`_SWli>~)dK#DD@d1%>1@qXsU_Jz?i?
z{zId{5UMI?K~M6f0p+@6hm_-9F?wzrOx%}lxqR%C>w>do@0J!_`=QdNl77!}!;h$?
zr&?`hYW4q;Red&aeqG<n9O<8CGi1k)>|5P}uY$4@?hDthcD|PCrJKP>_C6A2Z}DJn
z)tYlD4Tj;l8r}Pu&URJ1vvOO8N549;7Yt=I<^M9Er2kF$s-y}~3VkXw)b#oIV6SkN
z3Q-_h?VB3o1N+RDN59b9G~xWH?}hD~9uw8#mseS=?*5Bk=3a8WM5DiV)1Lfi4{~;u
z4>>-$vv`wSS(;X<)ABW6&bVDLE$)3dOLo_(Ru?br@ZhLEGZ(9>nlHNZPeXJ4HbsDV
zmF!s6Bs7Hb*@$HY6a{pRCOPR!VlV*vjR^Q5nL?8PB>PV^;h)U9n$<^}d+12U!p9@W
zZczPcx%=3$&&i>AzE#kHew(hRybU^8WWUCj*=#+jSY><FthF@`ccK5hjJn*b%z~q8
zL)V^HJaKh|(1cfz!92*gHN8Z0<MfSrXP$0YanRKL&ZnRfflE%wFPT!_x!l;#(#FS~
z15Q^Q_kEvJ`Y$agDfHcnv-5wNEG-)FC#{l#6+{L_fUX7sa2UfM#Ulqi>r&gp7lHJ)
zwqvA%&>Ft6>W^;Xo{q*C3nB;OOE)*^U8*xV?scQlHKw$|-mWaXU@2#HbKv&rC3E{H
zbtT#QTGkdM1!-w6RVu3eXX^bA?R$<0<hwOQfr`RVuXRa6R#92TgFS5`AHC^)>mE$)
zYPhmVt7^r*D(&2hI=b_+E31y4dU|}($?uMkYqN<|S*^dUr-pL8Dc?I)2g+ft89lLD
z$mpr2@4oc%bgZ3<lYkR9O_lLv$AuX($Fv0NV`Epo*c$4(Jo0*=Z`a@Dc;~IYfJtNT
zeVbIHS8#vHoZ05Wig`<ZJh$n~DJ?5Jv~h!AuAWHKv#NJZ#cr6^1XufrZ~h9I`$?0=
ziwf?0M-$hNWm-mlTM$39L%7Eh-2HaJmoK%`SdPMm@e7^CY};hAF(u~Modx3uC%ijx
z235h^LWp7ORw5GEzXy%<!cv<|90}sJ@tcX1P)*jUMsE=ip@vyBXDqaDgpI##@B&PE
zyqL<ht0sP4@pAaW_enX#fn;zZUm;^P+l+WoW6YHpiy@HX_tIq=!u8MA3$1=SLXq|m
zQr$HHLr`1{d<ni)PhcSi0@#LsX;UXseHHG&@EF_b_$O0H<?I?<CsDF`qFszlh<~v+
z4#8`k;Tz&=dg!~)BGZ6T8}N?iRKwK$>6QL3tth=>_jShbLSzszNC4!OiXPyF<eiU5
z)vYdrBn(1gig^ExNaB?u&J+3S;nRnFD#t-OZ*%p2qROh|_m7#Sx}NqOOYFXxYTwoM
z-e<Y5_>(}t#`uDP@2^y`_Z;Afx=D59Wpllk_dJKMXMB<X(*DGrqp4duqk?LeFRm~P
z$}&x!Z)az*Q}g~i+h_fw+}4>`E~(XX+plo9Pd50LOhDC%y-l;0t-KzTsB92r8MSiS
z4o4GpL5twknwX1uj}3f(_LJC9TzL=WOz?NEFz+RAxkYvFyXp3SlD^6s%!)hW{=!7V
z_vhurZ+S2G#zh@=UX#5_%jtbvQBuy+U;L71p=@NUcmFMTe#2$p<Kw@}y%P!(3R+$7
zF>Y21wpK6Dm#%Y)^qyO|_E?RR``w0baw->U>mse6wJck5^zgip>pde<R_;&kTk7r7
z-8H54-KvV+a9NF+3{C_**Tj!t?{v?MD*y#=>Li$zkv}y5z<)nENcy_7c2({5Zv}qI
za>9~6!MM(XbJG$hrNoc5eE;`>V=BQ1!h`G+K4`jlDI|~FnAM}T(Rb~Z=ID=){w~k!
znjYJF$}{Dx`{D~uZn<Sqie2u%^bJOKobtPI&aPJL^T}nK+BWQeTBC96?vKKth74X%
z!=o(AWtI6^LBlQ2u3D1!14SLbLn=mMPj14iXJ(Cr7_EOhtq2uFC@ja#5Wa{9#Oa4#
z@sfQrUHX$<tDHf_U%q{LueWP8v=$DXOxIF3{&Z^N=yH+4ePciE<OQ`RGaL)fq>64P
zZ$GjRnn*L);%Dkvr}x69e`)iOppiSAquu47cOdS@&tKV>pQPDanY752KC4=O-nAo0
zvw!z9J=2*1o0EQNd|V)H?fuigZw<n}HhhN4Y0~xrEpIq7gy&|1qHP}6sFLupPRZ^P
z#$$}S@_a}S(ER#OY-hoiA4N^VxHg4naie8Ci_T@#7923xG3D+*3P-vhf0h-x-b%UP
zt~qTDe6Dpj--MK(yJO%r^KR-Ms>zenx>glGy(Lh(zCYVA%Ke3YsLj`qxbEk!>g$4&
zFMSz0yRPE<0w1%$UFD@MT?JXe@g32-*3>37y?b2tsm4eH$xkY@lmWigFNK4RFqv_v
zybhdyYA6b6@t+`_J(O!p-pe;UpG_`j?I$?qx+-(~A6%d5dNZ;;)XOPrf&TLQdt{ge
z%5IKA#$f}UtxmJA=!crAcfC6Gv})<;%I$M=CxmyH`Z!b=cJ~d;(s`a%Y`pJH&hDPp
zCAYlz=N@f*RM}P(ndyA@_N$)z$M#eLtWHlJTwQtjNIPlek10CUQ<eXW<2!+oMV<&c
zDrgc=gPAE6%V(vN-yPv8AFjksQvZ0gQRjMRlJ#}b;e$J}r{0ben5uFL)I^@=nQ8e6
zM(4Mfb!qjjYun(rp|neQj6W+OKdrb`{ku#+-d;zO#**89!yjIeMB>mlL~q;>_eXRz
zBaH-rDLXC^K@XMFlxQTWx}29odpcLk3YKh?5qYl6av!53r{b~K|HPUfv)*q3#ljSM
zgto#Wz?ioUgcv@NP&Q(JKUG;~e=Yi#*3$pek~n)%6`Xhs585|2fQ&*U0jS<;gxW-u
z6kfffjD$_&_u5?j5_**g+i?ADpF7ErM&uh1!^1d(OG2A!%7^Anox@xOLEAPgk9P|x
zqv8*=;GeY%U5N79fwVV?{DM5IW9J|k3-tTY14S9RI2k}iLRqHeRnQ1kUvN9lAP9mk
zp$0rgOc!)-G^9ygB^A*IGN|U7xe9uykfJ!Ef*1o`bNjE;+q+3yh;}2P>cyipkwT)<
zx!~XM30)Mds4)iVKENj2xkjS6(fOU7ntPTRyceG&0V@P)_>=Xl?MbAGGOAFY${>J~
zY&2P=NI`$iwk<^|`T=$&$%>_OHZ6ted>BIu{fne-?$63UW_@tD>X##-9}<8=8Wy~Y
z04o^)fB^v{)I$Kyo4O*=Sc}oPh%K}+WvCR;cY12>x-r!B!Y|)xwX?y1Y|7teu3f5K
z;S1S&=bn3W@!IWs@=q+L*E+8DwVa_?I;3xLZ-ny-trCBYmb+iV6Z0(MBL1b#Ejy^q
zTQx>GXj`UPjGnt(k2*uIK7Xc+&e!-4)xeV}Ap+7CH~gY9ZB(X#F?<2`!2lv<M1OJO
zMF<*dt=ZC<a>j4DcCywiU27NRT9>Ci-+g+w+kg6G;3rkn8`}7(szRY|*%Hf_mU2#O
zYIyFuY=XWmuZ-K%k*B+~!t9?3XKO?B_RNI3d*Qs&-pbL+-l2Jus7I@Ke%$h}n@Lun
zQLHoJ(Zo|RaA;8P_Ub=bKAG=w>??1Y?VmUs&wQ5l@C(Cs`R4pqaT}f$mZeuJoNxMI
zWEM0_xw%UewX}SJa;t5@o0!0jMH4<T-*?|{Id%D<c0uPa-8D#<2e=&l@^`_v-futW
z|4Vatas05tnu>8#+pc2V*)tFg1qzI6B%w+YRGjKGRn+)B>iwO2(VN#yVC5GU^c>M0
z3fZ;Gx>qkx-m<mnA7*UwTK~)EE*#&$-r_vqusqM@p37gmorS*^@*W6}eQtir(<`d(
zIg(QLInQrZS3|2tu#Me|BXN20O~FPs0anfdd3m39wSNEc4|zEijs|1NFP!ZWjpPrN
zcp@}+3PUxwi4uoH*jHZpAG(qJ<avqB9l@o-h761E>w^xDvrTK>YQ5u#-#De^_2uVW
z_y=m8R#qG>ysI%w+w-k+n|154`}U%;*A=csV{@My9UZ*)V8z|^jM78>dd4&3o?@Ow
z+~7H5@4SJK^r58-GLGB<UWlWE59LaLTBzZ?h|F8=cOcRb<{vT;5s*vBB$W|@f!2rd
zr9OC5bJv5^(hcr)6P9@D^=K|FbyZ$>*2_-SbpMSXb4okPESk(cgewF(c9(jdmDEZt
z?Q5NQId!daMV*t0hu4>c>@!dORDUENzkh3A^BYZ(l0(gpajPwMo;I1Tn)Fy$I8blx
z`ye^@L|pc^o}pcB)<22<^EgQvK@?+8Ze(;nb9JlUdS4Xq8c|<s620XhrWKIgRXw%v
zq~&1q(n<QcQi2Bmm;^0PfAjS0*~;9`Br8sAl)u22*|xG~m3ERvu<7+GjYs}PBj-17
z573%#AsGENaDfjI5=aRWr3KLf_qw+;LV-czmtxZy9Hlk}v0);2etu!DsXYC%xvd$O
zGe%>u0ZzvqN^GFwYlOXk*I@<bnn$t*AK#i|5bSJ^eT<%V82ICD_&e11vEwHH<1bLf
zP|jzzIX~SUwv&jhBoW8dAROY6*N_YxBswf@4CJPQE>HvCec4xwwYCr6?X5f09I2$g
z^?M`xUs|zNKgLLXE5$HFDipsvkG^Y0kFi^n@jGVK6=%|?hs*c4Q-;nXhDx7pUhD9c
z+z&5Ee4AArE0*8wXmj6d5qT#47aLij0Z2D8z!V>`94A_H>NXfqH~A+v_YPkkuIDNE
zsFNFd^Uz#~m$~Ez`Y(1P%;4Tn5!(QZ*fN6*hrXX{T&~u)-?@|SDGXUg2ER4oX0(C~
z`2SXBBOM4>LzI$H?!P0~yThXvy=?gs`jfshSlkk#KS;`42E(QyTwsR^8E|!horKED
z3sjy{X-8&!GVV5F42v{}ew{9Q*-f5>o<mTLLNpnJiQgo{0F6jV?nz2e&d8;`0gj2M
z0B|6|g4+^vt&pLW`JzX_g#kj$T~~{GzkVmpAznMtCGctKAV@OKo{ar1uK+3+1&~*W
zRRO{{RMv?MP9I*bUM%ht$$KFSjt+xvC@l&_q)M^}SlRl?AMFQq20L~o{9|HL0ZSYz
zZ!*_2E=lGQeql((Nt^R_b>*q1j_FPwP66wSB!krTm1`5TiWwyMAlPzH-@45FpmOo<
z?=BZVl1ys=nn!GQ5EjOgpjvSTit%&A@4?d)<zYZd_@em%HRIvbDYL%^YB8cruRSj>
z%%14#tXX`fEw%je=Em+qgJ3&%O#^vd0~N1xf0;LCeLk>h&Bk?85eJN2z2}AR*X2~!
zfy$#$>1+Z;eBR|z*%86;A;fy&@Z*%qYox)OjBTS9wOL%BHU06}XW79mX{V>NlU?7s
z|6Os9QQWomo>`v3MAr|>akW*4pX-J$*mR)OD8ExwcE>OKNyxTGPi=1Rx%%b&MY%ie
z@8fso7+qRm@nl;3Zq4E^g$G4FkQqd3P+1UwKp22wr)Dh_BO*6#qeun6%o+es`vPyR
z3fDPaCCe=Mrz)7~z3wfOl8<X?{<=6lOZ7t0xRB5JsRhe6t*BmccGX{3%}S=E*!K4~
zhx)AWQeV88S90<EJEv8Q&I5j%)pV0qKlD}6>}$_2trf&N?PrIyn6$?~2K789oI(t;
zNTr1Xo^=U)B&9Hddv|1nL7>)G>stFZmkq1_+Sa+WBJ_gVdH<%T4tpvST7DGj+*>ub
z>q6}uKbI=uz8B24V=q=XShPLsjB81Mwy)yc9AWGD$%j{bDXh`>V&bKJ%DgczE~Rc%
z*LjblzIEkQUwjV^6VikoKp8H`w9Xlv4YmO6)!WS;E1~nH>7v}Ux;Ec&YgbtME&i7O
zB)=`zGJUPIU&-HH{wrhK?A+yQOmd@Ux@>WN|KQl>?d6WX6|Jc~gCWU5_D=x+ocBZ5
zZU#66oOHNjwpnUv(z%#R!}cFq3TM5nLBM?n#%=waFl~oNUwvL{fnC|MS8`dZ(wi4v
zE>KllVtgPrY11<wX)UE6y2hM0H+ucnRU2NCO-w%RXgSbw_-LUs@<6h(@h-7%0+%KY
zP3`~bTlVVRy!_Lz2ZxQl6fSCAix_fvqJF|+)o_K$JMfqF1s&1O<pB3@efKMwe_~To
zwf}_caT$@HY)n~^tV0LW-F2;eru_7Ia6$gyUl%pYZcZ(WYt#B*T`N4)<rHjkv5_>M
z53&WYJ`GSxK#5_~%hnv~F)GWbzi;eP&{sG%u%_sQl>XuDF6ZOPB`x2~{ga+Mt}v<0
z{eE?mzopN)M?KG%#5#=+h;)i<UlLcb`W`jABVuY1CWc%Ih|~Ki!Q6FmQ{U2s#zSv4
z&CZ<?6!hJ5@KP>49^ja?B=TRHN9)$vI`@7~aw!|k#MG%1godE8Q*{O#Fr4FyhC^Rd
z{<O9|gBkip6~))1cy0M5TCuG~DdUf=ANDQDiTZW0=qa2MB6Ay!MeXq={EQbTwtw7n
zHZ^C)mic)`1#8!(OMh9?s;{|Vf233FlsDvoJ*SA2IuRK$fAC-0(`u(Noq2}ia_%3R
zFxsv#Uf`c>5xS<L$hTs^+1k-ROZU>thCqW^efC7~d-s6F=tRT`j1l|En5WtIZNB<=
z)c|e&1WSkE`ME9Kh@W_OwL{!z$Rn{P^JOj=0K-y{J^%KRP^8@0)}Lz}d`}Wny10`F
zZ>6b}cmKr=BQ%mI?+sdUaJ|lE=RNs_eruk3BECdCJi%5(t>9?LG$cX2C5qh=nJ@V|
zo9*)gJ(AU$pQrK*ueGG4x;*<N|K^hasrmuMnawU5U7@D>gLWnUX|FdOd8Ju=-s0f;
z=jY8XHhvvjH2p^Z7)Or}7ICG&MlR}d+N-75Z>Hb${BI{&vqep9npgVddRDD0ja`wj
zts(DX({o;tN?PoS`pXNQ+%AlMB=~8X+k!EggJ`%%9|7tActYRG1h8^2@MYis&0c~U
zfYhxW+uB@s>E*3e<wMeb{n@coy0n7U-NAt7X9kYhZ5DFPt!C?z&z)Vj|AB2`g09!a
z%YGJ`F}$mBdHL_#qNb{uyp7}+d6lPc+urj3n0ga<sMq&>yx;kDPEwr{rPY!xjO{2C
zg^zU@S!T10tw@E2va}HXN0i8#p$J)qnPDgy8kCPpLK0@kSSnGLC}qpm|9-~#{{KHO
z=Nz4s_w#<1`@XOHx~?Z0qq9e@b@xvxkt5SQ<*(fp-`}`NFz6Qw@HeT9WZL`^HHlh$
zvWIupTfspu?s?%v7Ws{$q+VP@yo%i`M<=ffn3{r>;_ZC=JJNiCMLqUOM_Fj<u$cE@
z=~+iP(yo!z&yP=x0!NyrTfW@~ksqL}LMv?yX0f{&4jX-FZ~^%PMO-wq&*2RqPRzgD
zZlD6K5MZ$E!C4Jw+g3p`^yk!Fl0y?L{iX@S3e1XId1aC6QY<_nm|2lCph^Wew;5k<
ze&kzsFZP75op8LW%g^TQ1?}H|x3X=US&3WuwJy%k<|3K-a>c-+`z$z*#PiM;e^mbI
zcR4Mh-KHw{?(C6m%aL}K48G%vb@!f|j<N5~JC%f3l-#}XKXdlZ?^l7jiBTh$TqC?a
z*vw;c;L(MV!|qX(h{@siaBwI6ql_pxz^X+Q)k|xEZPf1NS_6k?oE1tt^q;vF0)h3E
zNUA7~9MWKnU!beqo5{?1b2Elnbqh~;zIS1euG>7h<*bBmXD^*v+FN{{*Gj9?slrc>
zjq7sUM9U5IS8*j0*e^3HcUCRp`?@X|{3q+o&tO+YLtk1KyrtQ_&L0gdzMi+>)85bS
zRat0>2pv4JWS}dE2;yDp`!F-`gx##}l~;Mr#5N~yWWV6`PJLcshGAK^rrOOm|AM)4
ziq*;<)cSxh44_j|5^SJbjnH>gSV4he{HpFOCZIFp;_w`nNZM7=ec4gb&@q0bvq*4c
z!(~zd+b!*a51{^MdRpW9oF7tZmA`1TMYIX<Kgiz`p_n_goYK0!HC#96o=xSI8d9%S
zHyztAkUmg<>a6Nli~Jox-|w}4PW;jRJ*juBmiTMkzFBFg;F88_NxqLlUR~s?Ym>~#
zK~gDSA1GIi3<aeZbnIJIJ9jhirez5)q7khipie|>zg4)REyfpeEQK(-uyDZg+TH%3
zjyoJHXf5<R{%hTtmCCB>7Lw(~;0Vl8%*e5dTWss+pTcSh!8>EW@Oo|D@PcD29KSRL
z_GeO=issJ<d)a2`c-L(@s+0bm8JVmgxtAl2>z{si#85p0@gs!t1I50ERMknZy`EjZ
zX@{f>{B)%YA|<G4%ifxbA)llkc1@oCxq<Qa-S1oISU1OU-%Dd(=rxgxN92p>?)_ws
z-~V`eZol8G3<qRGupQtNacmOKm%`V#=23GRkd35{t)~X_^HoITEK;(?_9dPmm^}qX
zqCMxY54%!){U-A~MR?|cI7Mya-M=aPNqw2tSe#z?C$%xz`FZ8fyNRTtJ{(hdpuj~!
zLgB$n^+U=xHj*<E{E}mW;?lZ@vH8Rz20cEcJPm8zNy)<b{H0pRsTKVZBt0}6J;@Hg
z$=n!}L!>gvio2E#3+dX|NWSl;?)?nQ-{x%>M05FTk!E%5fd+}9-O{?bXxKL_R;S0u
ziC+0@oq%MrA~9<4JuS;O1jYwLgIp7_(GdOvEv-T{+(<KE*w_Jy`O6y^BJ)yD0jEai
z@uoL-Gj={5j4IPKmd@;tc|PJ`$Pc94-I3?(=Vj)*KaPjr6RXFlcebhfn00<qdrlya
z*CwTiZ=LH@GgGrsGZx!cx>E8Cx}(DW2_)r6)CpO>){+|J1H$SgJPn@V%KF{ajMce)
zp9GAG7;>uvMMon%Rzq6zR-b8~@q*XhAUC`1+2<ZJ>CHX?IyRMaK8_DBjNz?a<w|_d
zaUCwV*rt|ylR4y3tsb2|yle5_&Hqenq#IyrI{H*fD;*<jH`MUvcZipnM+Bsl(<Dbd
z&dVm%k^}c$9XLbVip?+YZP(4n3q6l%2vv^PSIrBseK>~^ru)zlLnu;C+L%xz1o119
z0LvnJNf|<_FPxKst0wsgf_yoWiZTGr?rpuR1DG!b_vr%zj0r&0GaUp2nM%o}w@*2j
z%ys!^VI04-#iteCAAjVnDb11SAv*h*zZqh!TBZ{(>e*#4OJpb~)11;foxQ~U8u033
z-V%(0E3rKDP(fNuQ5?aE(xIzZB`5N5&NjmT`h!k`Zo2-(O_oJd5*Y@&VlhsPM@6Zu
z=hsR#@k?Y7C<%l9t)kG&LeiL#KlHMr_f|8Bp#?E%q!xn{DwF;gW&ZW5!gvk$mtV-Z
zRC&7Zm$UgKGAVxyx+kNW#wF*`ARGm4nIvi%wp{`;NKqbncCS-is+N&DHLWgG^OU#p
z*S&x%$}HNpHd9#7l#4Zg_fY0q(3y6x9iR$0v>a8R825e4>ET(a#!0M^=Od~@tLH*o
z;#tUTqVH9#ZVGgaaBQ?b^K0E9{L)Sm0ggHdPMsq1b5@I`J_;*=Tl#pB$)Py>q$@5t
zqczUz)!g=ON@NvJPPA6#?6%Z`TH+}+WAP1>r(ye_Y9n8Bl56oJqQ~yDDRU?s<VBV3
zwvjBgj-GVxrmLrxvgqqSabJ?pZmfFfUBQY?zAJI1rj7Hszw~j~K8RUgx>eVQwz{1k
z4;#EX9%z5|(3hRccjJ<-Pmp`m_d5;lN#wApd<n%}qnakCs^Nj-p_OVZ;{~GHR69Yl
z{xHHNmK63@GoBcS5he$^_sz*|Oz-GFl%9X%Q)gKVoYYc+Br1@vs&Pv$56v#>IijVY
zi+`({=SMxY8?VN2nC_S!^N@FaK)XCPuNtYz_>vu`$9zU?%U8Q;RgSN#lww8RD&79=
z*E;-wigZWcmlSWxEpkigdGeby=NBXM(mYQ7gY9!~7%4=?WZ8RQGe_EkO1!-#DIVo-
zS5O#{8)(xliMBMeJwJABlHMN{m`yX1)3Z+Z{o=_|9M`IA(IIboQR7h5;{T7dPA6wh
z>({yjj9hF5#s%*hOLE?mJm8utuwLJMhzC7q3^l(Xm5{z$U&HqWrdV}{vnh9cy#W9H
zT#xPLlrX&DRp?c3>bB*m^2D5mCz1HtFFjNx?;J-q9~;EuJFIu^ESXfP8kCw11+FYL
z8^B+{z+r!_LWLc-x$pataqfp`S&^N}31+e)k{NPDGc_x-1U1t>TgB5laa!p#0;T_N
zKiMfk%OwlUW=r|+Nv*q*1wNt<XLB2DWczjY20341e0XG0{25be+fs6%n&aSD;oAAx
ze7_(xx_x@i&rX*;Gxn5wf{|DwsXHlCm8wC*lP4cpUz-c*F}+(#eeC$fgV{7jK#|}Z
zwh6L{MAsjiBZ3_T`LQD!&R(Cb)5%YbFj~*D{o{Kvu|LIn$^RT=m2QMXUF>vXut&!7
zH<R$815!U^9zbehXt>!Ah{-bhMHYpT2GBo1AuX5Az;7k|abY}?gFW$UT?0|$6kmCC
z)$yXm$Ttc=u3fY*j|#`61aYoTQ3`+cSwp^A?H{*HH}u&3?w@1aRY98)MsLAla!)<T
zOH&Kz41W6KH<XcMoR0$=4mX)3cr-WIsmmxmV#xPv^~k47_LQo2caPLb4>2!ghDDLK
zsa!lfd#zeM?B4tir->&I`#+m^e09kR)`84&RIFW0U@&n=xiP0?PNM>}qT;qtCZP5-
z?tFuc;lI!_hzRrvR{HIue;dmjheam&z_H)BHR#v6>yxJh;k<iX%neS-Pi=O^Ev`iK
z_T-BJc73=km9w4DH|pX9=3Bc|JAO#hjq5G(@HX0&6Iz?UL80mzzs-2;np#!v!MI0}
z({}A8bDbAB4{Xz)-scTRe~4(IUgK7zrAv{!3&i)ZE%Phd7e$}wN9aeUjM!qs$>xt+
z_}tWnE^qDs@;|@7sAFb{Ic`(={Xb*4Z0r~JXsGd(x*7oT=>)cC`{wYY;R;)^gZtVZ
zV-@&w$w6v+Dg-PoS@uZiP4hIa#NOg1y}W8SN3UJ?7jgDqrg+xflnO*xM2SCb_qiUH
z$S~<Wv*Ox-fklg-8uyRWJjeb>>D#~76_{D#q)mAjPQ8dsVb$i;V}7msJwQrRZ2Yf}
z*QyiKeR5|)|2ZQ3&O-K3!O+Y=qnSrZaVx=#XYi1XrL>5b53%Zgt!va6u~UK*XJTJc
zJOptg)jxD{^WaP${i%E~fVe`psjT9+{3xKz&hO;>IlC=Ao?dXJhTLm?LQqm`-(4hd
z(Pnev)%f<IPWy?DU-BN}mO%H`ft4tk+F4z+Uv<LtWoSa_^uy%@XHBM?VyUx9ePTxJ
ztf?l+rR+WXW5JNo*sN5S&)Cwhb)BT?H`9yZ?IFe7w8>cw@7F;mMnB6}MjGI#59)Mw
z?6!GW)HL`YH;75}TsaBZ1&aTW0iaNtE&oy26LymS1RQy=xqw9EB`g-B0r`PApB7#Q
zjdG~zKX|6Z#<(szH7z4Ki*<zar;r#*t0GR1#=j=FIzM!z7F34OWSkY#mc1ZRE#Ucg
zUYoz%_*Ca)-U!qfm-<#M;)rDPpnmqulIWnmyFBkk^|k7_q?_n_H&)(Do{vl>Puou~
z`cErfGPkKGG+F0<F1usfe(nXO&YAJ4a39-zU&+Y0ktaw5)}D8*rdYiWzXVLQ8^Y^h
z*aC8p0|Y6y^pZ|2>)ih^rhFrgnjqbb_hHraSv&YDJSfHE$&^-dn3m)<{O*H9Z`EYl
zHqOcPe6CoDoW!<6%H?E_kK`7jwfkp^(vgWT#UG8{IalcP$E{LS;+TF@mSnC9oo$PY
zWzfC%KCEV>DL@fqYpF7n=h9^TA~dh|lHVo9LjSPwUB2I(A6?>2nmx0Xf@^IyH<Hr8
zvKok(tcW6sq}uO#<cVJPMZG}FskV^Us-)`&*S}81Jk`;>NGNLj2pLV3QSmSunGpZm
ze7<4*`DTT0uoJA560IhR4)^F&?&4puu_DO^8omnJ7C~jiK7$A1Vq@|?4X2V$-mu-#
ze6Dh&`Pc*84Rue`^#LOo#aBhiPiy`6H062a)rK%Kv~HgFen#L0y56v{pX-M6`R3z2
zZ|wrG+77DebUiazMOKpUUM84HroQvb?Ad(s%y3#!UvSugG@hSi$0KtSiDxbW<fWaK
zvf8;FsE;dDL7Z%VV+W`qJoq<27<zKA!9v=pIrqH@F68?evlk=3y=bI^9PmkeSca2G
zwhF|Yym;zy<G5A}Yb4sUKU437n74Ki+n|FZc8T`JJ?gY`_pqYQ?G?LU>pTt=HLm;_
z+?=Dl;M5u1P7skFEku1>goX$2OJYC_$eqIYD>zuT#o14hpB8n0szR9CV@7F7ZD^o2
zvnsI-$XXve`r9oO$lfeIw{GLBU+bvrBZnoc=o?z&#u<u7UDEX@RckjpA8%Z4bTrv7
zNaaw^F~*c@h&Nh)8`w#yiJUNFu!3Z5zLTA{J+L~-<iyUZ2X~os;x$y&HujCX-q$ZA
z`E>}oUG1H1H?Rh~>nv(;iJx3v=!bH2HD5qO*Xt{+TKPO-Hv|MsM|WvKmsF?(0m=-@
zYd7x#>@yhvj#OC3<mIx^dx_je$gc_R-pVySIOSPxn9dH>{!H@QmUJyX_l%Nitv!F~
z%$LrpmIwpMD#w&bZmG2kYp&pJnBs_cDz9j!dE87u|NWE1lM&pNnj;c_+1IDUYyVm|
z*3)%yh`c=F(CW+y3!)5g98?y=3nM}&exn-b=M-!y{@9sBb88>heR!Hz%45v^z^Lo&
z7Z1Q$NveGAxX@|yH?4-ItNCA;X`oL}7UpJMl@f9UQOj5XVE=?bAr@!~%$Pt}1V{MM
z(r<TF>lUq#HftpBYK<{3x@ks|c<3nD)8|msts~i(|F*D-+vmhrJJ*akTksIqr?M~;
zgl~-N;9T;kNTPaV4i-?2WSCzBOfS~Z)e};Je}1jA%%PCQF<Q6?qs0JYiJf*f@8Vap
z{nWI!ed}=5)1V$w*U2_;JjGS#{$*3o`{I?dTZOK9FyqH!LTT^p5lm@V9Pa{+S0x}F
zIj6Rio2tSM&XSpGu+A(d6!BLAVGlw86_FP*(S+^z<1e_VjREa-nZkQ*kL~6Gm=H{b
zQ2ZSPrS4;<wVi~xrwKoN5Qp(+w%|GybG9&fK=lj#V<(<edc$vrOUc}uNw(A!gC9`T
z5}K0M+2SLC8`M`_SixSHJe}Q4r6k0iefw3vGtEoAGwS@WbzSVUJhSMc`r9OPwNHL|
ztHm6BLhMTezyR<csfvMNzOvX?-I$sxr?y-|DOQ#4q2YC{kRQ>qskOn^l{B({SNGto
zr}7{r8<*u+uCR(=rL|rR?AP~@<^-b-kDn0}&K2{<fM7?}Yv3WA%(~0&HN^(<r))Tn
zyK{ylE)YZI{msQa@B<9PP03g}ny&ptg`%7SZ^h-%BdL|NjMR)IN9xQXH0R^}e7fo=
zBKTqExW!~?XI2?CWUsmR3|Z8l3w7_y5beO8UYk>yzt-I<2Z=vS;4G&!q>WL{%Z$Bg
zUNm2`+ongG1dDT4j%r|(!MikeX@)$<B08TOWJWKhmqm`wZhBoy$m$?V-msB6N?y{8
zOYDezIZhH3j=`$)nmIcr1qmJSN<{2>?V)VuZ0F?5?T?Mk)G)KO;Os3(k;?D#EwAiQ
z*~e6w+;`JVOiX)j>)f>i{6N=tC+v=UNzg3@yE{k*#hWd0rH>7&pAv(=J&x<?|DLIu
z4IJ!U%wOv|#{ZZM^QUf>I9gD(7+wjqrhDZT83_{-D?IUhjWh>oj+jK5`3H8(46Czb
zM5U@5@Y8!FZrVGqa?9eMn9!WIxAxx`6Vg};>JIsw9_fNhUmO=%3?|;4QF*)kLK5`O
zaV1*{{#L6huIZSVGSqXLu)9px2+hXTlFf+wL(tAjyt_R1FC+ijW?}nJX@2|e>caW|
z2)JJx@ZANfr~h%>m`47`?>|k0nXHDd`}o}7S+Vs}m*m|WndgSNVzQDixMzJPtIdjH
zS<eA8s1r>MQy0&azxH5HL0?2-QKAcBPan48N5vLTotD%v8-t<*Ghe;fNr~99#oGN<
ztkeW+d+2viYcN)m+J0g+s$zCN`3g#eGJdBBG7|vp4YYpckL~I_wyX!vP-yEQh!A;N
z{a=kxFueG2FX@I%9&9J2gV`@r5V#+5CN50-qW|g5j#_T7UJ$Ka(~T%?>2w8$hpBXL
zl0lk7CmH)|ojG3s39lb`DZPHARk(b27qUzNPAFI^0iFSRuADT;A3ZnsbukLW_m(e?
zAJ}VmDlYX-A=~0{RBQ(CJvq84qn@|f<3+pCjGIl#8zj`xZb+<y1ZY^OHuxy)xti`7
zew{K0UI?@2uzYw-MW?;5R)6oszx>qITfz==DzU@G0SMGUKIuu}Tc`Nwi5hy0)?|^=
zpgk@Egml!yf=G_qS*B?(DvCn8I+Ius-~ro*IwPqE04$oLH)l^QAA3&ca$f8~c*q;-
z2R`(ddT~<=9I+%0rCJf|B+xJkcFf<N3JwzHJV7CB<Xhe7eA;$Zzlng@OPlX>Mw;gw
z1~v)&o{-!rR-`X-Fu;n};WoK1!QvaK_?*B*VSkyGpUW0f+MN9y5!=P3q*oPa2HaOX
z95}Kg_J!cxXie?kUZ2_<gaYKnFB7O2nhL2nv|XFm6d_T-hH7dw&WbD<KMMvv5awqO
zAk{YD)D{~Ra-X1~U6s^>rp5O+r{0Y(W*I)-Hg+<%0Q%VIp?lhQ7-j?Rs827Oe?h{I
zmtPkRYYBUXk*}oOf3iSp(&x=5<Jz(fx^q67g<{IqlhDhIs9JBU4Cn5?BswV2cEO87
z3R2W406}ADZzpPVi`(_WF=ji<Hg0>ch~c_6a#{AW8|aBs32lD%9LQWV6TNX((%CIX
zIQ!+>ee#KBlq_W`#7%btslMKGjByjS;@=(B`pBXIA+vwKU<1j{B%c3t&THJ@RB4wZ
zE6IhzPtW=MNwSI)ALO)@7FKH3axQGreE!X`B-Txj%lQtzjKx?u(`K}BFUw@er5@o0
zl8M-w7gZH1Q<dhq52@C<1ts~KPomVm@fbG^rfC`-vnj>(Y0l%kK3H2OLnzxzjmQ*{
zr-Q2q;tXF=`(FlGAwY8B6_66#E)V?kGqjaUd(EI+qo2LO$q84zG^+WLY;yIu`(Unc
zOx1}=NmzZ=zpSV`N?aPoZYc(1E`=N2yr_P9a;8p#{yHyE#ksr3w>8{4?@8|S`a~u%
zv-<bblosvI9)d4@!>JmZ04g14wLLeC%ouuz$1G9J$<dx+9D^!prJm_duW(M`+x+`q
zAB1BKn!|bH4V3U2QtFN@QvR23I`ITLTV|pEv#Iif?VLXXdE=zMp4=o+V^Cn`-Uq=_
z@I93LL<`RV;fx{#w;<Jb={Jr~c`&Qc)Ma7t@tyM;CRy6f^QZD`NbHZ^I_D~L>@Vzm
zD9vD!E%ZMuouPJ?!w_+=xSBzKJeIs`nDvBqC0H`2KW~^dHF(F$ygP$7A<;6X^H`tF
znR2wBb`<b4gD>YcjO!jeE;y6K{w(vbt60U&NSTY<<eTK(3ia)7yXB<&Zk2LAKi}DU
z(6N+u+tfN@xNnX=;JDDcS=qiz;SmJPgp=7y*B~<xksq=JSx73%KZPb|5lIT)fU<W6
zRZjg-<9Sn8C-Tb3VO5KP`8a~}nRkY?y5K8b%dF7wH&MqU+U=@wwaN>#q_hxa3$^O2
zT^C_US2d^>Z`r=u?mXwvB~gSdQ%@Hkb3CTwUG{6;{u)*3Sf<JK8w7%DXBIW!_sG3R
zLql`rcnn2;3UiBU`EbWfhZ**7Z?p5`SI(BjHkPHo@i%Oo>Hbub@+#!z1*xAb4x{^9
z&lSZTCK#L?H-3thkQ>G%1_0R|pl8G~SkUTDaD)NzKxB?KmrufRV4S0kmUDBpBpYeJ
z!D!9^A`Nr(C&2usDqo{2PCU`yb4bFX{Bvh@V$`Zua|qeMC>-AJB>Wndq2@69`u&W6
z6jHDpk=sPBJ`g{$vzq>u?mg^j5Vyq3XOv$2slyzki7SUj?sjZ@29xvxXU7;kG|lg<
z&#>T175rK!WrzPrx2X!FE-6l0&2Rp}cVdNJ_e?mlEvD7a=TLt{vWd)02!@JGpg;~H
z#48V$ihYBI+LarKrUWwL&F2t96j&+g$siia?Y@5;bac{(m5N}-q-@Bmg4ox1c8qji
ziui7us?oWZ+3BezE7c-Re^J|Twql9bD)@UE&0UhsgSm=qF1>xv2)X2=U+L*bqh4hg
z9TF~CZz@!)>fzN!vTwd>imx6q^;$!2Vnv$tO^=WQU-=B)8T$WuH!N`_BKHO9$G2`P
z@121-IpU`hj}}Mc5_QbwnSz!Nzt*LeGp0MIS5oKv(zCMg;`we41@{H6Q)x<Awe15J
zXb$7CY){ur%f(n8{naYnt#B)r0*g@_#39ftFtO^~WdEW9mr94e8)_zs@pAhh<c<>i
zW|O0p#Pbh3=va0zMO4LH6?f#cnt96cycY}boAxt?)4Ql_$!6yB%JlGJ)4iS!p4?|Z
zS7Y%)mB#~WGdt-zj~T-%#28lGr$d{U4rxj1W~vH!z9jP?LR3y`TCpu(x-az>^-QJr
z;D0N5$2M8n`K-6JrMuOMO`~%fG5m~yTO<1?0ZDbHEK=>JAaGpa?XLdkttMWYmEim#
z@Tq)`&NRJ+m4P$59p2fbN4{0y_FC^a{Qv%Z63KzxjHd?F%o=>&9P;vUmIIl;&#TOQ
zaw=yO3|Z#NYp&O%^0|j4sA8iv1p@lZi~JpzPS5ToM0ndgMk^R~_79f&AT|+dd4sNz
z2KpP+ZTY=WY>6Z`Yk4qWQ!Qcg>B<Ocq^Lp=VN$gIi-|`;&-eTmhXLSfcx44@oec+k
z{aiA?>fVeQb$io^C8y^O1^D%r60sw!ZVua1m6BfkTlu43>%2Dva{E*&hUXsGP>Boo
zYK*)k8myb~Z_>Gxk)ZXHZYS`VE!Zu6CEeW|@BAY_BzANYollUvV=HJo_B5vRnZDd9
zlUtv$WS4Rm=p#!`m|s-&$YtFYTOO@ABB#@HWOLz2&?DFT0z2}#&#L@gLz%O_lZX~p
zCd}%=KN2yGL1YcmeQ5OoB@84{+0I}abTix#QvQkNqvx=r_WRoEG$ySBw%9t#ci11>
zUQMedzP-pU=+p60Q;+yM-SL+E9Y{kb2D!whu{nt=!B;{T6DpUlXkUprf5kA@$K^8l
zO<X&5pe0A2tk!uwJ>Ta1SddTQA>PHIS!vB<`5!8e7BgrDM;pd#L%oMfJ{AwrUr#TW
zhd%S!zt3~kc7s|S3z{&qGEd#`-Rwzjd1pnVzIyZr=Ic+og7RMMncHk$NwpIs^TTKb
zS$$~VEY>Yd$bk2U6wL~`)zA_|=a!M3an>cbs>DGF-@_pX$Wh*Vl|fP3Suqb<(rvG&
zaJ0A_{Uf4J09|v9WsMcJlHl~MEcximjipH*Qim4B>ph$3;q_yM6V4Gq_>p7$visfM
zNz)u735m?$qa4Q{e&g1uhy4~9f#r-|Q;8No9hzP{nUi~y{gM7*MxcssMc9whPf3Z(
zGGSOA*X7y@cG~7_nCGljcSZlIgcC0O^N;!_?D4pG52oW;=BuEsLqO(3Lbd39Nhu0>
z$YTWfb*OMQ7K(v{WN@L{$jfR0QYldRG7eU6s_K|*B2v}UI(?fqEL8^Nlz2-{bc2+l
zVeF^sS$YM_R8Y~eq)f#ujG1Tld>N?I*0L|JICZ_)I%YWNguQ1Q_Gxw3Bn-YHR~;wq
zJLvtTdS7GuV%m-jc|88G>Lqrro%MHvZNJtr@;j>}N`k^2ec8(aA%i(d*Skck^2shG
zmzbBH|5_)1m3MLuThVCFHKhCG$M!ntowoQ?zs1oy8EPSy%+b&iLdyWIHX?OvLlW>6
z5$zFzi;$B3Z`TpR=hW#upr_z_QbUB3m*g+uVx1~I9Otu{KHP2AM&0&0|HD_OVys`b
zuiR2nNmWL|m^$pF)5QXEFSd#8eeL1tEizi;7Y;2)+6eNk4fNxqy)?;1RUIzwlTWe&
z7(Qclw7L;*v(v>{2FA<hdZzK;I=rJUkDv8@f^i+nP2y8;6pq9NA2rOa4}2o!(rEo6
zlK)>Q@7|zhPq>ZNjh`i&!_xO5PRD!+X}-ML613;Yaw6TWUL0259fcXK!aE1hJR(XX
z%;Ta8I%SZ>0k#oFH?DLQ$s~Kw*aVp-=N?yYvZ2oLR(HLPDEDrv&`HTZ^h08|MyU0!
zCh38eZUNcKPkfjB&H;>kKfy1>o>`ID<_Fm02S#j9ScwLiQF6^*aF$JMGCf{Dd?cWF
z5~Aq!TfF@)&g!Kjm&dw&7Ry$LT&A>-6>*LR*Lf=aT4$OK{x!wU9^~Fy18Y0G5A=n>
zGBaE`mmh??^+09&&xvoOVYafvMOE?caO^i;klPi51jk2Z1j|#j_3wYJyBxZ2{B6-)
zn8iycf&C2V2XdQ*BNN%{!1!j+p^8eH_lR-{T8c4jPs92v&v#dUvI-6ElH8P|V@B8-
zz1@#$jF%uQ?n%${bk|`F?0FlOidT`_L9e{yNiM!T85m|MXW~1<P3!Zsj#PB$SMt22
zhB+|vg(m)qNWIBuZ9l`rIWEl6+fHtw*6BD0hWS?u+WG%UFlXKHr|k`tGdVMyvo~^I
zWrvxk5~b`VQn3NQMne;9B%xA{GMs{wW5^2TN$72hj5X5WFhMy06As1WO%s6A|88aG
zsIT~@?-BXU)XCk%qpE_`*JV$2{B5LLuPc^~@<Q12Y1ratTpe-}Z(6_IWVn0So#Q6t
z>O{-R0QVGGn;`d>wKvp)#05Bjqr2dc0(_Fx{7U%duKpnub-=rut=wH51AYh)?$0dv
zEFass4_e%*%4*`72BQgdysTt=$+t-jtF)o=nCR>GJzFFGs(SK?<{3#irccD;()Gj=
zxrZa;+>KjUU(b5cAve9^qqq^tT~S~=D3lB8b7et@=+k0to&1Pvokr73^n?@$e5lt*
ztZ)FKxeUBp2qtoH?tn(tSmq(b1I<2cfanud_vDhP!=bF2m*PIPkH!^sl?A?uu5_H=
zcIL|$lp2RYOv9gy3F(-fm~V?<QQUxS`Ao^c)b<C8R3tL}zLH7IR|az#byS-j_h>tO
zo&*Q6_mX>PQeJ+f$Ij%lp8fB-I>_xYl;(xO*=t?L%;*|QYB#Bz$T&SB)tzp0H&Ce5
zRQUIp1OpPD(4FvK^=8gZU>zf7bmA_c=JM2G?|pzdg;E*{YBW$up^-@5#zs)9B*Oeh
zI>ILOfatJOp__6zMCW*qYuYlfi2xZ^n;@Pp?qgF&BBd~lgn7y_{(AtPlz%=p`N1Jb
zi3<MYmbV460ts2$G9NhKzv)~&)}LXr$;Mv}_o(!ufhqov2UYvG&CHco<E4Jm7@r+I
znyinvANq~WG~W;jXQ527z^I+ae5j=L$~2yY4hrFnJl@pWnwBBir1eU;3{h~#@{A?V
zA=N3!2Vr9=&8jU!k`0UG3oMe$&^oWIqmxI99}GM;qRLcV8}JKIT^Q@c9(rr65_B%~
zpIV$@Bykvrxn41JRidE&CFas7;CN}X()xx<_H7&sQpT7|YccanyUh5XUhd91drEBh
z%fWM-e|$46WMM?>SN<YV+;f2gWJD-xjpuP~y>4@?tvA^VeQ7OF^hFOfR+t*GV{^pl
zRo!p(RH?cHIWu|W{>r{Bx&HK*<Uov;Y)M~NnfFfY<jh1XJG{}NzeUqB&7k?>28PpK
z&TEcxbrvr`;<(q8^n_ot^JsU6rI+ImiRIurlCOI)<Jm2r@uQic**B+Gb85r&M-Gmw
z{#0EU{j&FS=V>gS+id*o_LCuERkl<Cw`3M`_Jh!lCP?!6VNN5mQG-<=8wLNsyZqN^
z9%z+-<m^B8Yn@3mJUya3`6d(lK)U8dwcXjl4gW_)v0!XaQLM7S!|P5O8y}eS!l#Sm
zCf&*j`JcDvZGmTYyIh?W<{v00r)c5Mn-^ozB#G373*n{=>6UiS+U>X`z44mAE4wr4
zr_<TE?8>s-Te!*e0~G~M6J0a&a-AVlHQ73$zt(N>@ChHoQ4>_p-h24p;a}@U588x(
z!F$enJVd$Kzq=6G;DPy2vT>9>cT+Mfk+l^9>jq}R9tMUYD01dPa}ntS{|1Oh$j=8H
z7t)43fPH8)_#ep8eu9w?$L@WWHzQr#l#?Kae_3W`7ZRl>#S77yk65oX?%p%ff!}VS
zqeDs&+uN{FDG(SznLr-}|6OF?6`uy4`L3v&6N3!2f_9_qWBV0&ooQk-Dc`-u&=Wu&
z25f%Npg1E?FF~tN*iS^@+etG*&fqP+KgFI9(#%@L)?{{BsJHR?_gkKS7jRsAi-`*z
z4&qHD3a&JgG2KNE^67fGsI>%6Lx7G!ak(GXUW_{7DcaZ#QqGH3P|Zb<p!bz}T1Y|2
zf#IX=5AqLTm-uuZa(V#f`kGBE)>t$wl&r|vQG>ysJw$TBkIzzNq~is5Cfg2q<`z$m
z|M3ON((8%!w|JNC;3uj28Icx8Jo8SR0ORv(`W042`a19s$cYZ}Wcd$kX559h<|+^-
zxZVf#I8f7xBrGXmLUzNCvunc=&*GhJBBvdjtee=rfq16K<dfWL#x{Q_`HzMyCsJP~
zSmYOfUX6FF-(fW_^i@J2*?Och$2KLZbVgh=1cND|;&x5i0)#9$<$f}VeVrHY3^305
z%TB2j9wf2l@5b?_?VVK14_qLd9id%`$iA%VJ$_%V#i(s%N3G*WB)Nb}Yi;g8l>$f|
z*hisY{S%r>SKjVREQIMBLj|#jspRg1K#lI}3_p@Y?Tk%$&*1xn<q)$%!q=;GG!QZe
zY%>QPeOUx5TgIj>MD(0NwJt{PKu%;n$Xs7TfI$iw6i}d#rH|!-j`UxRE_hXeq(T3)
zWfvIg!Qy!<YZuU9X11EMilm7kslx5tR&;Fn#?RZJbAe135~VBfgy1i=*2hKQ$~7w`
zy21pQ&Rz!qHL2Dl=iT_PbsvtcoP-3+0w!ZER~FhULGl6V3All20WtF#jC5`8%GZZ0
zfHq?e^p(HXZJN;oV+bVc0NqJ@6M{Do$A6P0IxYg?PXn<9N$VSAoE27XqAg~bO6cR&
zsM2=qHT9<WPEQ7rmCKSmn)Byoh7vwUik1w1J(XwP5_49oG%UJsM)?U=P07I68BgEY
zkLnmzHC43(x61sg-UM<n?s|MYUR8rO$TrfhC@K38j&++NX9hgp0N+x6g51)r<$sFi
zUi!X-g1|!+TlylA8Je-`z6toA^OSCOThY=Clh7vRCPw6K3Dg7pu_P$sf$yrDjPtIr
zb`{mTcmE2pE-%&Rn5nS1Gz?4MH;tv5!CcmC6t=2Rt#sftfIEROtTl=uxC#Jfh20JV
zU5@?9?G`E^j|o#4CrY$z3#3pQVvPomaZf1z>7m!<*jr*EAkI6s!tfitp{4PSJpIPj
zq&&SoF)7yLyMe91XZs5s0+(#i(-lxnnl;qKEd<Slx;HVWvg3~OabI>nq&zA%*V{r%
z!fTmJwp?dS<EMSbIjhRWj}|Zyn1UVB(BXC5u%a3IH+v|*KE88o9GA#Rb4u~T`FT?Y
zF{%qs*i#``ZhWXJqa@o;g=jv?TjqG~Hr2p<=Z%-+gv%Kuuh2qme+i>PlZ#HwpIUSF
z0Gw(gdGY!Eww{uZih+ZkB+7e)&Z9z2AEb<8IG0e`-jGvzDlGbhsgK9SZ{+WOV?Ao0
zna>=XOi#F7o>&+o`4Vw#eBG=caN2w>PyGCg__0bwu1&_~?X1ZK1EZ|C>bUDkWRJG-
zb3^rUN%3~X^64?{S-0ojukTG{35Vyt#wK~(P9T5ln3kD!0K&tHOQcl5qI7G^6NS=*
zfG0!Q)?G0Uy!^ZG8aJ!B5kgsTt|ecGk)d5?Eb}MKVTjPxYrP*LSU<nhWyEPlN8rg%
z$ny<C1)SXfny4B}iVIUIVTQMUYm~8d0R9R;tL-d4ud@8z0_H0}fgH&3G1x(jDo#v%
zTAy0=qQl(Wn`TF>%1!tZ&O1H8{5aD=y;PBu1f4jAH$`<h84sYFjQ=t1FdZ5x<!B&D
zlL>sE_j-IxdsNT0We{ViZeX?SGgh0!u;m>s%?zm@hgyQg;Y-hEZa;eL$!IC*5&J{l
zkkNR_$C*{5l5p)nmU55Yj^f0t;`;*B{r9vkj1~SJuERCpVe$)VN~TXj+7Sh%GO}6B
z02SJK3@){%0mObsA?rpl9eMKQL0*r33v@z4XJes_JRnXeB7j$4wwcQ0xaOTU?4P|b
zvGv~^-jx>*kgIj$ANfLlp?X>MOTkA*Ye&4wc-TSDQpH1V<Ek~?x~YlzAG^460Aa?}
zQa!BSvRXTiY_Ds?Qo2j4DTA7JA0_KoBqYSxRO9d-r6PCADjK`vNkOD^mK{mScjHgY
z2NnJCWr@qGgpg;0DK=GIGkxu3LHX6L1=!^p<w;!&t7Ta<-`eF|0_mImbo{E_Ukuk`
zYPiXd5Jx6xa{wQME;`KdNDHV>kBXH+Kw7~e0x7Iylqw(?Q#|HhjoN19=v-GPRQZ7k
zPUqHWP*gxr78hm}u`%k8&A%u0nfkr+aVRYD^YIAtlgk-D))E@tX!c;VB9nz9Mpivp
z+-x2;ahCpZcQc{DpNdJRW7SP=kvS`N7bcZ2mn@84D&e@cTN=zs_*v2j4e#Qr>pBZo
zn&c=0t{J*n9amW&#KHw*YNwv28<<B*>`A~~tn^E1V%l18<%XkoJ?Z=EgCCq|k#?l!
zr?_0@P3H5cM)q$>lVkHaB@a`_4YSV&?rU#$FHekmreV5imb|AlvZh*2{i0b_)|VY6
z_VjiVxEO^WSxk_@?=fWQsBqj@Src#x2t<VABZ~Bw?yiN*?DGxTUNdLpkpUH=RYRNr
z_eH2#q38oV7`;w_MgeL;f#}-pHL8UyU=$G3he$NN*z;=(%4FAQo^>_hU;#l~%o=Yk
zrtMG=?PU-AYsP@wB^o3D0ZZ;bpZa5yqZQJ1@r=#`p9(k@`KqbIX3@J^H|R))spa}G
z3o;X-X<F1ht$)jvU6epBikqi5L)VEh!+r8py3=&mO{jD&<diC-PRQ6uNB(?MY_+Sv
zr_UZDmgeuL26>i#y%A|+Z@-rWm_1dO4)Aoy5{n&9r~ht{nx(c#w~}su6`Rn(6O7DJ
zbgJ$#m0QjXP^~K+^3#>v20&0UtZDLnMDfL?<51lBtdB^6ZXpV5uq{qTZA5MZDCAY5
zV^CNCBFX>XUjMskgtAKH^8-5!^rW&PULR$x104I$B$;<O96n7Bl%3j_Xx>5InnpNn
zB{2!i3+t3kZ@xB$;k|d7J875<cRdI+;chk0)-d^$`{Gv4ZZi$Lt_S-4*CTWS_s89b
z&R*b=(&09Gp;+nG=m?<B*|}SN5T;Z>GENLh6sD8N?gte}kRTUrL|GBCbKmTP%vaGf
zLy`!-pr$f=tdt62d1D!A*aIE$>a6=Wsws(|?{o%~OjeQAAPNpgR!5UOffz47&i2%X
zrf*-8-}av#|6O6z1JzZ*)m*iVki6(rG)BEhd0re8erH!7p`sB>d^C5BUV5)0zV}AZ
z0-&yZ2_KAZJYDp)MMY|>Xw!5_27q1A0g(PK`9f=cWg^?xLx=J+D7{-S9@Rm}(A@d0
z<dt4PXh+7STz+IE4XJnE!}|MQx$?I{1WKbQg-m1CI^wY7P0qhgwfisRLh&~g!7CK1
zbg&>s@>XP)%z$opD*7A#bxsDOvebz6VlQG09%C<?w-BgISDPB}J!tLDD#kf6NLFTG
zT0p~-Rz)K`>l3fv4BFZMo^{B_ibeDLYLsOCj!!+EWJi4GVKZ3!z^3Y9RMl>~<GkgM
z9T-w!&VVUz(xwJefF(Mzb<&Lh0p8f^@>NO)!#hi6^;y<r>k2wnBlg5+j>=kVf&yf)
zU_My5tkzCzp*<6NYy?Eh{12R3`fxl$#(}_+|Nr1d)dWwrSS1P(%u)P;@P%BMQh~%F
zBA<6uYx+1G{=$YKRQ)sdbRcn4{pM_dM2YoQjJwKYam!y7C3%(MSUBHzdM-`arbeJ=
zq}ma=(#hkgX}f(x*=gfdjVxXl|030a3j~dvY>OcNhfvp+i}(GCDej?9ym*tRlW>E&
z3*m=+pBUMElz!JbAFp2e+2)vmHWS}e-0}_cl<>)eAUniVc3_?x9Kg1{DLf2SnYI13
zyTvMj!b27T5>bYdNTLS``1*4&1mQfP+LRKlv{^`6oQBJ|AX<4j!}!p+5;r1KwH&=p
zs|n#@A7y=mKx)p}wV+Ix<Xz073{gAlg9Bag#QqKn*jwyP@jp|F-LwF{fqIU@k?#0o
zGW?J#<L6`BuRaochYcJPX$llSoBN@8X(^1f4(-CeyTfyXcWjjor6<o`vsaz-Yz?=4
zakjHMqPdSpnqCap^~GW7m5oCqwyN=Kp_sS7=n7L6ugX<S!b<E7)W6F<MM;?CF%O~c
zV6?1ZoY0J3V@)(7tV%bBOTuhHwFIMw`2zlh-S({(xBp(-qmXoJeMj<LV;MbBx0>9q
zQi9A@zm3&EGN0pPi4XAJdugg*CFI#cTRL>S9R_9k0<&uS7qPci;)L?4YUUt$B+TK{
zVy=|tjvRVWd6BB_C;N+<|C26u2MBtrc_-3t4@+6${!CxYPUxkR9iQ~P@Oj)@Jhtrh
z!{e)1heOS3fIZXCK}D-)-|Vu$HI8oK8g`U_$G4-{pVMsy1FIcDTMzI+p{^Y)z<>;;
z57{I1b>S8ns7v7Ps6bfCYjITQ*B-6(JBlD{BVLiV;={mb=M+zTO-bN9DX4H5!Nyh`
z&;&I8<3&Z%r4dTvmm4kfByr!Pl;qN9f{E*jFPlns_Ky_5S`}q;)5c9D97@8C((991
zeRqs6hubA)pYdMsZ5}^TRkeeKogSP2)6T#Ax55Gp<=q9st)Owc_j&Jl3%xGVLl~_d
zMI)_oBJGNHPJ$;dH4an%RrT+S7rKJdwO%!-dd8`;{_Hc?<IKumc-+!~F(6m=l&%P`
ztwU>1b~9X(g=_%i-=qJoA@oQ-FEsgkL^}|)<Iw}_O<>+GVp|shwmH$~Wv#36)_W+K
zdW@mxu{QFzU1O8uV(APD7`=&Dz*jd!csA<Bdt&z=W?P#5Q)(7iINmMLV+=po`d}wk
zN7AYD#b`o;gwhQszROm#=pek{LDgT>X3rh_Zi_ME^fW9-sJfLF`I>NH58G{Dw3s}}
z%yxrf^EY4r?$C3_QXA3Xl8d?t!s$XzYw)<>Y`_J~;s){dzc(p?!$4C1R(fB_R3+|)
zT&eVg=WvR9EMu5#o-KvnqiDg1KWZDQO?j!)<C~aqCN93Ach|(5ut|^e1+Iy9ZzV^&
z7Uq-Z^?Z2skb}<$(!4><$v0zmmvDu=C1k6C_HE^wU<(!j0l6}7OGFz(6J0?DhZyPp
z{re9#*+*;50_)yTWNEW_6lQyE2u6Q>zQKJh?{a4n{krZB%Nly@p}%%z@aB>Qby`!r
zxs<>5HmNfjKXqOY_pRA${4z5%vDrpFjH*)eX|?XN33qC@{RP3sK&Rvcvk#@C7S$3i
zan?bB<LUQiIL&F}<ag+PGbx|7`2!$$0-`q78hJ6xUPfv2-RF=igL^e3#L*Ekpfg)D
z#T@&brwp<$VLlYC#e(OQ8Ib~>FLCALT$i)o(Gk~2`FEc!_x}lMMF|(xWwN`+M>|_C
zf0-{Kp4E<rOnPxXbhfVz|Hy7<v8!sw4*VrQ1s>}72iy&Hz0t>Xyof`97P=S@kR2p&
z1B_0r49*Mibr1qwt*t!prVxTYPoF+Vc<zcELVds>T)*iBIz0XW13=^njovqe7ui3F
z>x;PJIJcU_?-Q%jOUje9`4m@gj;Jju?Cyt0CC@wu6DzTDWBC5rTdDE%;^Oaq%A9b!
zQ>E3s`BhPaU+df>Raso}7q<F^mf-k5)XU$q|KTWN>Th+;F`J3+ObSLyk+yh#=S~&W
zse`*6O}38c9rJ;=3I&ENKpt@B|C?GvdW1e?FcF{NekLLI622Yovb(Ml1uO(2{D~S!
zP748cC^{C4N2j{SJkTcv?Rw61wTh(Vva1)qmDqc{m}zmpL78KkZ?m(jHSaTSmdNP~
zH+)BYx1c;!KG74@WA>y3&pzc$Ke>EYG?kJn#Fjr5BvI*CbaP(Q-M+86&m)K*>Hzpa
zZ}j>OCQLb+aM_JBfp(({$X4p$%uf*cA-q)S4}_zsL(e2}m+5c9ncsfu?Q)|{ny%7a
zc12=~A))4(%EGKW&++#wtWqLXHRG+Zkrc5@(~w5Sr;S-w+S%R@WxspM(w<qWcMqef
z@);T3teyiIH_EaxpNcXRCh+}f+cI!dj;e&cJa!FAYKTgxO%f9WHF(j#CRbuVM$lLA
zW@U7(7I(uQ63U=^-wW?_1at?zMSzh+;Sp73r#`%P+-&L9;a~qJ`4TUhOW=&Nev*7m
zeTGU{r7U%0@ooS3d)u9VCca2J8j*W^X8D<f3t^tZCmWdcTE8Iv2u<5drUrS~7q;$Q
z{E?;Tdx4uKrX3`iXckQLd`H*!O_ESlCy`FdU10IgdtK`^b7xcEAkqtz-zdl#N+OpG
zfCV8%hO_9Fkug+RyAh`Z%ZIBR$GkV1pnC)eq)&RW!ZJ*&(ZEdQ!nhe!qJAPKR=#e_
zFVYn|-bC$lZ?iNe%IPSXTiM8Kn0dT?XZ!hzpf*p$?gsjM0j=JV94!l^DR4ozg-j8u
zzwu^N_@SII%03swZO}AE&J-FNa6<rO^XOoT23s1aWJVx&gcpodP0*ARk%d>PkQNpq
z@h0qs^w62ll(1NJGcj+?PxJQJeCE`TFt{af^mBpTxw=(@X9KNnD*r9n7?c#Ge}LXp
zbAaBW10G*gvzGJnbk~Kwgb@k5pp1UW`|-p(Qi(hD`5im3YQYBmAYs~9B9a~BVizRx
zXxBugYk2r+76x;s<agZN=DvSkSAHVX{8GFzH9V|7FDPYzQd>h<>%^^9@F*LOfWw`(
zR<^_LLF+&gz6S~y)PMx%8ChfmDm;TxD@&3fEaqTi36n38(FVY9@BR(fZ|%QS>^wG1
z(;$5RJ~Z?}5|SxK-S~6qr>U)e7rbPv_GRaf=Lk%*&fKw3<l|#wEI{H)y8;trGD&|O
zM%_7G{yRA?n9JnGCd^DIPgWjCm-Ut7^@6^uTcxY@Ftg<mBYlgrxii~e-6a2+iUQ|V
zTG7Y{*4@g1AlIqjNso(C>K*0GE8BB}Y`=0gDAo}B9h~gg@Y2c;nKBccaS2>ZMWb!M
zL?(+Vfaa2OK<LOxU{tvTXNUtlCqykoP!bHh`u8me+CGVVz7ZAx_&Ulh|F-}F3Te<A
zNR=z2^r4i!gajDEkOav1s1>Xt190I<DZSN2;varr83Rtb38Dryw4?cR_w?}F7CMfO
zxTRHf{-A+V_OEriphvF+HIY9DSZV<zi#NwK1MIGiLC5(NZJBh%l{7C=(S0_zogN_Z
z*RIcvwto#}hMgG<{me8<#7NuV1Iq4ln&Uq91zo7PMw<_(npbd6_3LSR5xR2<DlQDo
zisH}e932U<w1tlHee|40JCS$O8ZvdorwjI0;znpc+#sSt5ArZ1O+t4+=vS;&4EnIr
zVO1AywLsY#WZQ7iA)42O@XesxA}XUhtqy*JZQ!n~X#Xe;sA?h<<Mkpj8Uwgi2~FRm
zP#0q@kB2fUQ={+nxm!z@@s{3)%>6q3^eqZ?D;fhXQJozp3zzU5#S^ZpqO1GenfUmC
zVpmOH9d+@-61PAj?=6{*z|A`E&LsPmwy+Q6H-?0zI3X6;%&OMQocS$oGx&Y88>Sm=
z{kM`i%AeTol<fA!;8hi_y(42DQ#;|?_1HJRvZ&1eNnwEhPJ=y{-EU)bysK#`xG;TM
z?d&(4aIWEE*pR~9>xwALT<a(cizAd@7b^<yM?ykBfXsbhhT&}ikdc@M&YjlrH}5QH
z6<HBr{|{ZdhkT<BoO~S%D0W$IBq!;lFSv_W7(B9(|FJ}P(rr;S{^Vs^9&XSDFJc^q
zx1vmfqX)=>lWz98^@U?jX@@>jJ_|R`ja_=gEL@CZb1Q5AvRjSX#~y;g{#rNutx#|5
z<qw`(u=3^L3wz-Wa$MrvH5Y?(JZ$J!j!Xo2MDk9wKU2BX4!QiLBY%bS_atZXt^80N
zc!w@}G+Aq*h?wYqN5Vy+iKB4qCto{!FR?O17++Dz-zl;pYG?~U*GpGBH<W8v`?=})
zEX)MtQVgknmn$#q9eX_#*!{kIDDR`hE%H)0#tWBaVAy7W1qe1S_var1sGMGQR>ws}
zXyvBX|9;w3xmX<^ml(j>7O=C}qq&&Ts7o}X;?97_9j%|!GBx6BZe&-?a9yD~{C6=!
zbe^;;y>iwdtJ?b4I$5_jEw=5CGWCvQU(VJYQzBGDnY!&gYIUGAT6od6*fdOwBJveX
zdQ;eZ%jm!+VO0xPRaI72v(RGfn=R)g1?tXkA>+6Gn6NR@<<wQ^LZv>#fP<1UsDDtD
zMWnJGGXw9`WG20BbfD`++xm`aj}6Kl_0-&Aq({!bz=;1!e_9YXxL;=^j=taNt^|SP
zO5ZPdM@ZEwhvu;ExiB0}x(QQtnx}9Bq;(|k)<5yP&pm#uZ$V3Sn`LE4^!*PdlTU3f
z@iA?x8lEGB=ZSAnh<$~nR7dCu>z*qrsJ)2V2cs~qd01{@9l0sUgn8$-XK>2|ggA(C
zYj+(d*Z&`wm&CK*^cZwz`45(^R7sY?en!d*rt<Cg7B5PA#r3CiZ{N0E{~~_&s|9n}
zN2Mun=l-~1pRiNI6XdVol~7;ra3SjdK<tOLFo%8|uZh<Hor?&#xtj^@W3%@3KE`6Y
z^mmb;{PQf9>=eDtd5B5i*Gez&)4AUi$s|gglHU!I(j3PLH!B)~7m&_RJ}O-G-%#`1
zD?kdo3e+$7(hoG|u-$*bjqfn9kExKCqVpygl-R?~qR_!Y8MHFcHqd2<;~(Hes)HDN
z6vE8&)N6kd)1vn!rlK6=hK+hC<)9Dw8yxr>f(-j7!N^>q(^p<>)1JsppQ|>B+s$Ld
zmfS*A>7_U)dPRng3Cn3!G~tn1G;FF1<SI^^*vrSh12b@M^#49aYkXS9Zsh6b`t5i_
z5*=0#3ik>tJd|q`o(D_Q=S;*QK!pVdeOh*-bp;=mW43po-T~x?Lzd05CKP17oE|(8
zc5usk_AvkL?4IIVb1e%Yz8TvGa|8;CTRt<kRc~=T6=#b1NythOlgL+_^_`C%O0|YM
zs`UoJ6__%RcZkitfx7{R(LISmJItSlc4mmz*#iP)dJNs;3CXYR1NO@`gO^*~%_#~6
z@C0UoRr#t8S!I6dyu5)`;%W@I6&%$`A&<}R>tQj;p0)0?W<g~Vuc>B1eCLBp#P_qC
zV=B6dA(MG^vj(@Sf5!^a1l6Lw1h_OrpO$e*yciKA5Xuer`5DHE$Xmi2nFfh!0lKDR
z?Y11i`xWx8u9x+NG<QcH;$g`q)@-kuRlUJ!Rxu4BUn4NO3s|nDy?wktsdU4BiHz?S
z9HW;Bk?)qg$#H{pZ%Ia4xKU!RI-#QQ*Shtu%dA;v%FGB5)z6MQfC3p>N5aL3xK03H
zwZI+fxirZ5#Q#5V2i-Qk-}1aDrSOWAI=Tp+pcBQ1@RaobpuPU_f`+K|PgS@ASiC^X
z*t@JTu=d`)ib2sqH;>Of)r|M^MrTS}WB%K%nw~}V6AXvBvs}V$w!I`fc^>skD_n|$
z+j=|j-)~w55UM*68nOP$VRWH+_-53|qX+>B$~AN&NGM<~;J6AG0|OnxcuzBweeekz
z=hm~|875m}9$$uINx@usZroAKLo|_ai;Ahf?amVye|QppLWWXtUjx}lgR|{Nkf60-
zAyl)LCdscgE2>`Yz*opwJjkY6NhqqByhzz|+@Y9K{Sqxu!y1vrSK^RooP3?q1ECr&
z2^xhPR8osB9e_*dVi=Sd00$`1S1i{Y^k8a4jQ17vF&OdxW$Nc+%?X#_!jywce(CEc
ziyvwAC~acpYt~#POC|<u)`Z<2?;aPzF9+6u!%1N$3nk&9@FPLvGjTjhL=4iIBquXx
zyPzzrn!E$Wpu51=tAB{GM8;T6J@RdJUvisL=5J+ub_=m4zqS($R#kmgv3ir>l3MVA
zYD^IHOH|=cNm*^bN5`6W#70$x^3k~s_n`+P`HvAHHS}Ods<51b;?;9I$9fZDVgOnh
zY7g6^V0<gAdKZrFGwGMRD@#kZU!WbK#F#xha?Ww^#(E!eq=NAt%qqdikI|Phw#8x<
zvvCWsQt=A@17w9`5<7OAYR+zIpHdji_I+H5(D*1Lmw)Yst(06g39&_C*?}$ugYsvT
zfgkS?w9zusuXVgj8B@EAkKD&p<|WGD>nY5iAt0lkYpL~P<>87|9q>DU(B`^&pE>)7
z>~Nw9Sijc!Y`9-kF-W(4ryKDJrG`BUu$aJ9@7q&!y*=_37W?*jY#10IP^!a+@MwoI
zBUt4+upDD#&CucrO?)Nj4y+b=gtN+g6sShG!(`m+aeFpit<`SkmoECmJ21augpkj2
zF#Wm0Dws6D1rkqxD1aRYRGx(ejDB~Qbp~8NlHm_e@|zXhPFiKv4UYpE3u!I$h28Fh
zR_cuW*Wpy*jckA#Kq2c1*ah0<TaYg?;0f?q_kVpn!fp=>QJp*WG;<XMFf3=cMxAuh
z59Lo!?fV+W4;51}KMNcc7GqZ`>O&Ksq>|#a(<!Ssx<9Y8JL8Aw{TySqnei0dm!o^@
z1kmXwqO2CVO6;Td!V5Q1F9fPTV7~y`9@5T3+Ie&|QHC1T)BrIoRzXdrk;WG5M1*eu
z+zXW~G}oO)4Jb>__vz9jA$yO7C=jBOe$WPz?E0?q{JX0&Jm$Z-Et$|%p7~S^n?)7)
z>*Y6QOY*r#NUeLCHrIJS%L=3nI|bgfCW1bLcRGGO$j_eb0S!3X4Ua$@bO(}t;Bo7J
z9v_CYNqCiFA(Vx1h$H?EDwbe_#-Mb6?m<femWw+UQ$C=ZR_$;$U-h}A^4xpq|8`cW
zFTW;Z_x0<RNJZ-0pLA^(`09QlP)kRfo<$69R2nhf15kDDbu=Gq5Fz@RhzOHg0wo!2
z3Diz6lx@Kwj0sYCIG|*HL*EP95d@7v5L^3vZ<0Z7OBLw|@%>LtzR5X^-EW28Fu*8S
zo|UN{ZiZABt~%{bJSz0sgv9B3w%14#G(rCT5dandRD{3-l<mURP>^q-_z3YDO9m%d
z&ECRWMIlj#*SdI6M}`R1Ylgt3JAD-u&WlIT-^5pk`Pvhh+BppLP$46Eu@}K*T=c}y
zL(-hs2`1P>RtKBFlT`Q$0U5a|?6=$*J^A_ps`vFz7bDZtI}6~I1djmX=A#E-@>@`0
z2EMDa-_V%_q#$^xx2%|#!V9H@kqX8TQkabh`Y5qV$`+c!-!j_vhj#8&nQ&NA@R5;`
zu~P+un}uU#v7ygV2+YD;h0YOp?!pz322P7`ibn{nl$hwJ!0A_Cy3=>|9$2m!!ZkM^
zY`H6|>rIjZCN@w#27n}Hn;5q}ED~?$+_#enOZd*g#TnT2Vv7;MiHGs<LnakBnk$Wp
z7warH6CAItdIV}SCh(H^`j^TcIh&cMy${N38A<1!j&W1f*!D2{H&%aY<d)EqgIcdf
zI2!*iYi|M$_51#hDv@kawj^sZ#uB1X$U2OKF*7tHTPemaD#?=UWNU_!>=R?MW=o{(
zQI@e3l4M_#t*o8rJ$=8||972p{?|Fzbxzm&)1^<-d!G06exCce@7MjhUpLx8wjxHR
z&p+$*=hMbnP?5rn710I5*%n=dArPPU2R4s?0P}$|{S19$tgl@NFd-mld&nd|dSm0O
zMv(!Ekl)h!kz#SKj#oJUj<FdwOIvb>*Iye`Tlk@l=QjVJft*4@;8`^wa}!1IRP^w(
zV0#b5HNOn>|BjBU;J%UZVv!tlw3hSE*o%bDXO({$ic)PXyg#|UG{#LN)5daB&LiCN
zvNyM4K%>Rb$E&L^x!0Bv^MJMl1r=QH>C{-LY3R64%Fw{CdjY97iod0ABWgyj>nHCp
z!~H9?K2S!^^9~%-7tdg_=C}vP`lvi}x}%{pcHYJI=3YTA$LRZUe5iL#9+UTEyqCK^
zk(T{!!tk2HaJu5kT+}V$u9mNPpk2ZYS>HT9w5Ixb_?_<#RO>>Y(n0C?ds&F5Bl-)q
zS0Frj7TQvP!-oJ$!E+w~BMg|t$IXHW8d%^UD+{0i$S(#!j-aiJW$WyYLvGG-ra<#&
z@E6=(!iy&Gf=^l^@6bCHCpKy1PFm%j;9PQL>d;03&}5n{U8LTk*61x3Qx?9~)Ia{q
zaG&3w#QoH^)#72F?CCuWg~z900p%X+3Y!{67q+*knYm0;c4TfmKWkk+@OdloJdfqG
zkfd9@B|$&pz3`j&Vg~OG0t^E5GJL%Dczv}WD!8I6Go~j)8$Lc>L@yGxjZ4=uEh*U6
zbLpnsi~SY-BYcEcYE0!f`j*xfTr%5DFlAt@coK*Y#Wfpo=SNTs4O~C>-+qGohSi*2
zHoKIb7tt5>C1<dMW%L6W=5GzXxK&yI5%ni|`Zu0S>Rrh&pF6K&{%L{ey~uqN>9UC&
zK0Zz+TVv*e*ji5lr1g4N3y{4r{fcM{Q3g>noy%y-&i&gEwYUL+f-TsU^t@LWd`|!#
z5rW6>Dv<v0-{C7HfykiC{I|qyHT&}O&O0irWLGpoBI#99yRh0L_^mpH)B(3oFY>dq
zc{Y|UY~5xiylkHiDbjwta(5l;N5O6vEfFpo2Q*z3ENJ!5zq*80DlTl_kXx@IBt6=3
zIfu?|6CO_O-{`URM!7oU-GrPdn`yu1{RFPKVe+$$Gx}^g8c#?A)nu?s0I%+mR1Vbq
zHb8dw5-_Ka0bUL=8zEY{V6pc7+6p>$RqZJbTSn3UtT?Z7w8EMLpL#D=JtO`*`Z@`}
z8IX%OY;^6o3&0*KJm9@`8r3i6kf)ANF>S<t=H9R|f5~s%QHS1b-BFWns7QWeP^k>c
zfxDa!7@PO?bZ#VUbYGHq=55vU(x70&K}SRW*yV38d7k9&%86cbnn*mNDrdyYZ8%?S
zK`+7~jlBmTC9wBCA(};+1(3nt)n~OO)TJi$cw6;p0R3B>fxeO60^IP4X~>j`M}4}R
z(4W&r<G=X;5Qr;LX`7fUHid50ONZA38v(Tx@}D!Hb7U3(p{x*}5hCs(pZ-z0{Q8GZ
zN=ruEZ=8Rfl!oBjc?XLzHT(Se3d=py?hH$g{2bvp{#R;4bDl%Sa%295#w6kMspQDh
z<3X4a7KtCN?0Xc@y{C+x?*lN|0AR*JX9*}B2g|03+<YtN>&MrYVI3@AtD>1}mf$g6
zA5~Ur(a&#FhmtycqTnm_Iaeg-jdBKEQp-kzXQFS9WETU5@m@$>ezvr}+9DQHP}1&?
z47A}z`_G?yLQWxt0?@xDV*u1Cpg}y%6aN5GL<es;#7GJpH@<C2sBuEz)tk7^i5E*I
zfJ-4114_jjwTOirN1ZCTnQIp1n-A)YijVQ-$w@Ux<l!hP<;cjfr3X}RPwXo1kt?%c
zmxrJc1y0p!;cD)}ucty=BBOtNRdhU9i9E!6Pz7zE?}F9RHX`JZimvrc13}4_Gac_H
zE2Nc+zmjW#R+xI+&-X%S1~Pglk^+U+Zd%Dn<N<rcsx1PYi$In=jGpb}X0>)*i7s)R
zeEr0mg#Ta!dVRo#{K~jfX)B(uN1u%}tfGYqrv2SS+Qr-9Vza%9WWC9?*~m^(vM|_B
z_;!lKRo-fu8zpsh=k3p_7FH&898G<V-+$vd2ORVrG71vPj!Z1PFwpa*c<TsU?b<5b
z-=nw;h?4~#He0AH=Ox&nC|(F4hg81dD>6#fauK--PX8tlj{#>LI9>ld1jzM)%FILm
zWoV!dQU81+sM-OL8g+gcoiu`N{K%VxyXBSJiUDa(){^KQ|B~_%Z>PTKTKHp|7E<?%
z`P42o+g3!2nf1(W`g6yc&nvvOKHsKNYO_E4*YaZF8LclC$|`ya)Z)sbaqU1yYQf`S
zFT@!nZU%sgs{X==nZFDxTH0633v?6zGCW_Z$)VwmdUJTO`BawAS94}_QigS2ag$Fq
z-uFe$C)?K{pVeQZ_hg3LCYlO(Bqrd^0kc0Vau~p*bY&QN*D9#i=x7Ll`~gmW>)2<o
zZOp(wfe!QN3?ij6y=M>gb|_G(Xy6?0@ebV^!~qx2*1rte7LJEcXemNPYUU1AmJ4g!
zzWHgb`M-Kh0crCt{A9oIn!~;8O|2BTtkSb0NA=_yvw{h8&n}ZMg(>~ba8shdJay)?
ziHUIO9L8czibgWn!}f;Rv^OK#3wj+3(1RutAvK00MMk_u8gDC1Mk~RdihBupjh9Os
zg^rju04x0cJ5U?efljjj^70_|T60q*S@`RvRZyue+4O9Vx3bhe%)M{a8xsgB^D|Py
zurk|`=ztGK;?Ks4>%_^%;jV%+#%q-EJad7_&8zTYHFJ4KmGa+w?a~)iuDa~D=5J7+
znaDZbFl<%B&8Mlzxt$s@mrfrtVkN1OV{N@q6C26q#nKiPLlzE|MI>kI^%r@+whmWq
z*0?MeEbZ<YfsQEmDUjg-xxRlrbOHH<-rM+J5-L=AQt@4q`C?N`<65aP&x?K&VAKaL
z_B}x9IT$~UmV0E?akZhGy!Xx3P`D#TiP%HUx$?{c+p8yGDW#d5&E2#y^6qu-rnLBW
zo$2(N$MjVH$ny1!jFY>10nES-0kl&BEK&N*y@s-2z<Gk)PB0iKUkXBe5CB8=3K&6;
z`pbVAs{2QPzV#~b3h&=H{L|Yias>SAyTBs^-ajW)Wsm#^7IQCm=j=M-@8DRr{J<IW
zy|vFs%j<#~*ez^^62N{#k4sagL+ZU6#{f?7m*EO_1jHL%oq$aV(fy%v#Vx-XL_biW
z0)7b?n?dzI`DidthQO3U$wgpHB@F4snUIGK9zg7&V3zS#dsPtVnGNObp}UyoL!?d?
ziP`Ib$qH4SgS1srD9Cw1Yah51>Iw*U$j1heC#Wz0X|`D?i{(kbPhc!JEy)M^Bnyzr
zh8j$QJpv2?y{%swaEdu0>i8cs`W>LZc(0SPPO>Qn;{_zPxd|w;1*4$RG_9do1?dF4
zR=yzUhy_@>dP_cParYWN92#UHghyI{F9H(;B$(;MX>bmKtOS&i)(4;0vOK`uuM<DF
z%Je}U{$7iU*`k1O0>~e+g18YZBQPPrtCI$0W<+`yt*t}ACpQF9^kzYRi0oQYHvC)I
zO2(<wuKnx&QfERPM8B#!d|hAEr)3miHfd0j2O1!td4}Hq@DGipbECkR04CR2f|!j&
zEjG$Xr<@v_=kV|d!@oIlyc<m{Wj<L~Xpq<ZnsZBIX29$=x1+wd^6y&(b=)IelS>uc
zUx(n?z`2pozcX1-OSDM1xhYbBUIYyz=-FB56@W!(`i#!$k%JiX^d-c-21+B`z(WHL
z@(Gy4%P_EMSCY`)R-it*GrGDSD9u<W6zS<*>0z?6;k5!-xdfk?ORnXr%4*A}Etw?D
zxNXujjLH%t(JyHGu`K$tAvN9k9@Pi);FEVn5q1{iwU}Q}-h_X$VFG6<68vJ0i`K&m
zl??gIN~<GRWpBz1A2M-dorQl&#RTNwyD;{|+G{~=yQml6z@+{&hyRIzK@J&A7BE22
zV*VH9&4|WRp*IXEsl}%-k;4F~Zojs174OMu-jgrCNWV%v0yIXGIb`k>WW+slM_&C(
z$?10gCiuwM*;OC~ewO=_LC)=E8qjrW8yGS$PNls{8agfm8+cBOPkPl~gdkUYMlq&6
z;;YnX!U$AZ3J-@ioWJ>EvCvk{_S$Wp3l;YpcYMxrKj&xpEtB@V4gn<mJT|oo&|m&}
zAs|wZe-4gWX#E2O82s>Wi6Ov)U9eaO8~SlGps@l8bc8T?*g$sQ1Yqrxcj37?w@D%{
zs3MKAxKz@czF)LKgfGyLH5Z(<frF-;u%Dm}SEryRz&^?g=B?VjHJ)T9bU};)kAgn%
z2NyIUCz`6FQEo6WK?`>wET%uJtO#lYT73&Jzj?p*bgj#^*{_e{ib5XSlRZxG?l@`^
zMhmn&J^{P|OfOiAMCcR(KgQ>P*#)%|Ax5A%4&dCEfrcDU0R@%*xwzP6=p#`l*So>6
zQywGVTuJ8LR*CYcR4kd>opa48i(XZ;tB}MqsU!o%awEqu&dC6JmH7~Qb$0Ij88=Kt
z`m@%#b3q65Z0^0;CY^8z&;$(Xlk1%N`OhgU7#U6OXD2Ej72c?*&C7Q!dt5K_OPd<t
zlM2Kpo;hD-mMtB|eLiGvFVea6$<)>IgMQ^+!EXJ2sMZNeD${8wbRq(<L-D!bLBJw`
zDsygHiGt$g#~SxNAzT9#OyutNK^qQ+LcjGXtV~@YS+G+&mQI?JeBJtQqXu;YEOjb#
zNpPMUsxGcM<FLu*e@{`PrLvZv6ibB5eOznD{v;nj2lyPnm;AcDsiK0?Rwaf?&3<)l
zw(t`IUY7bbh(nt~eWs${E~z1?@z`un`lNzCcY73zT3PZ*T<TEN*2x_yx6hDAK<9k}
zE&zZ=)Qo^zV*!!X5Iv?yNQ3Pk03^ol^}Q7|a{<Oj6qxv=yJudygg_l~CBg&a?)d?9
z+lcjQH74woZfdHUWce^lNN!%$r$ty_X#OA$5)HH(#w?OHGB-jSJnQ838hD<%#~WS#
z&@eD<>{n4CW-eA(?(g1D$#}T47jH7y|8pnm%dI4V`ReDd@f}9<GJ}#06$bL^aYFzE
zhP<<ce{>jdqOGC+dGq7YIO&b*!I9DXUq>If{2&dZ3Y`n_P>W2+Z-ds^+zLH2{z&RO
zpdw?s3?kPsySwu)$Kfqf)#lHw-#>NQkgA&9_IyF|+<JDb++<gOh;BRsz&--UCYT@Y
zWk{xwo@9nv3UIG^Y6C706jywYD~FgAv!=}S_QCX}ogT{#{yq4WV33a*em5EFp->cd
zt>`kp5;o2Dm*pXo&r*w=MsDZC&-50nycV9tMSU=~+a=z$tl+<1Soh(U^xGbSLlP)r
z7z|v3UJ6}8o(GEJ)K^T~Gh#%JLx&4U;OIi{6X++sxaSQGRO7m|+xNgB=e9zfxW^4$
z^3gMb6IZ3=eY0g}6(n+@PQ|qd)5F&TFwq_lPN=JO+lVjMzah0h@{aXwt2X%Hqrqps
zG!0Cr`|J7w^%OZZOO$yBM;>d%sdhZmuyH+WFZx!ZzM)Vh=zpUY`6FKLa^5m1N;{dZ
zb~5A90nlLw%u`Rsi!LZ4+Y%6c<VkN=U~%g1^o*TO@;BCJ&c!kXxu^wkcyxlr6%*-A
z#L1j15i5r(mTY~G0k}K0{j}1M7R#qs{-nt4Jw>H&tdtM4&mMGeya?q4g16Qt1y3AS
zQ?li5iW6zaFs%ME6Arl1zR_^h6-RdTBCc}CQW6<GLxyKBW|!6aK32Klb35bbT(JRI
zzyKqD?s?5Rtp%L^x`IX!vbPB(VxE3xR!;H(>6wq<_Yy=m_^8pTKtRl_hP&xigF-~>
zBBp%!Yixv@v>&3T&ldrIjlsO(?*#2CUw$%2U2bb_tm{N**c1-G2k^?qVJjD<hdW<|
zGu_{(dzh7sx+6b^q<(G9t?JzfBn0M9ynbox^jqwsYn`{aHuU+JN+2GshC3%G-|?LB
zbphPi*<1Jr`&1C7{a>-c<CEO)fvn?6fT=^#JuvrAS+yar@?Z8|GaO{N13}V}bY=jW
z;5X6>T7wp_0m&i)Ubm9?IE|WuXF@mL{>xAcQ~`zpwXuEr#O&epm-@Vsr@jGgU16oF
zxHv1!swY-3zwn9ljjH^GX8{O;jo4p?*zD1Mk|a!OBC`L(E$6d+KsIO~$NhMy&>;6~
zgO=o5{@3sxS6U!H9AX=Q22cOcF&1EApx$PJfMEgI$Z{~zUVsxDILERHY~XuKN0o`!
z2RaHCU+m$En>mLpzM)qn$nYA>Gl><_ZSybRa0L@O)RtVO42k}kha_uCRmh9>%M&LC
zi%O{&x$LAR*4Wz0h8*j{>|y+sg0VO>p}jJ^p~jx#5`=ZZ#ecsvm)Hsu?|jTOb7x-T
z4Aju7@nqJ2XrEjFji<ZA0PZ!K)fQ4R1#W5pk^2S68$A9DBKezE+^qP!aK}QI@Lr%`
zNB(eKE=fg9+2K~aYh*KDvh-NwH?LW_7YU!sFwRk?5>Mrk4k|~ZykoCRt@t}xCG^D!
zc9y%3xM_b%u8`BefBrOe>s9(Afj<;j&43S&cTc|@?DWwQ(hYx+hJx2^ZO7W`EPsRE
zS@a4NYwO(VDJ5;DTwl=s0@UgP%exB)4(YvakS+oSZ8po)<H`|r@3gClMQ_6IVUBzM
zWe`G!oIE{89n^wFso<hNvxUET3OmK5he}%Yd_mxS6I2jAvmX82vd9tm(-4<jhr1s-
zq(H@G&j)qr90p>;3ZTTT9iUR@U!8sYTt3AYgX!lkh$Qjt9lV`gmUj5HZ$U}RWPMo>
zyRGx0x8zlLHG0P;19$!pKDybrQK@CD%QwHjrWmtbLyt6oa=?GR`I!<RKT-xzZ!3C}
zd#KJnBT?N*>tdKj?comR)3D~6%=J)*d2jv~1rypu2rtwphwO`AE35&(OkY27+o?c>
z8+0#^_sy@!L&!?rY-=riiLo8TRPlW>C@7sPN%VXZpyhUKVyvL25S4=+xcg-O`3@Yn
zS7<UKOZPmXh(R>I@}J8Qq6z*0E~uOp${3MA!#2<q7Vv}UyeRJF34ov4>4Sb!sK}*0
zA1tzW<*<+NH=!?rW$}v+)iwwzcUBvn;v23pbN){7p`u}ZNpsQ5i0CAZ%6jB%H2&y@
z{-V+K$aR$}4ar9c<DzKq?Ay|tcjp&x#>J8R*9q<*|NOXF#GbCwf}vyfAQklDUo}Rq
zFXTc;(J4y6gnkB71(Ez!uR}_Ffb|Nh?@T?|{0Nv~Q+FUgaCz1YArZPjWh{6pQ55v+
zJ)#Z!rRz9Bw{Bq+zL=hY9Qn%-Qn?FBhwhTht$uoWc%1L>ro@|$8jvfL?-1PIAY_HV
zQ~hyil=h$A2k;sFls8xYRXvKa2WFhyc>*%gvVgb>A!ktYHZW%TAxw`?Kt=*e?}IZ5
zxC&qm#ih|ZUO&EI2-U|&V9bvkHSj}R$9cmqHaF`}l@ufZLt~@!wmnc?t|qUt=L+M-
zy})LuBGmQ+Ie=IjRMwRI>AHgy<8k-WPXNcsdFJ^@!cb0M&(*iUT`?CNu)7A3TL?u%
zpq%Fk{Rn>Y0c0IP+VKT2gwT41&SG#5{(F}ol0$+u1{XJ8ze&u0ou<}C%edoE*oN_`
zqso2HtxuYOt4}wb&5$(GL&m?h<e7QNpIO`QQ#cA#W9zsdXAHVcZwbDsHwn>9|Hp=e
zj`9S=f%*SA1W;>{1RcOowO9r@2`XIYT{IF?A>4{Lx<qvYG*d*@wTRx3B;C6Lo!9dY
zR;KXXqB8e(ls$3~Tbw+`*|YH4=N|HGs?NzEJ@iO35Iwslqy#-A=;8Wb8Yu7`K|c?O
z;u22<J*84scTFvQAXNVv2wsB>!73rq*<9QksLP;1Dd@|ZkPXVE*AMK#>%N1`72x4O
zCDb#^V2dY%nW77u00clC?0LH4vifx+>BU_nh%v=q9toY5Y!mN{{(}#_LR%R9DD8LW
zZ1j@L>o6a=FNDs>S>6Um{xVg4bEP<5Uy$gL_t8IUWRB6}`S2X;YqNP=>H0?%2}AR?
zf-|LTYDdd*qoZXsc>yS{e;{xSI#TK3GSrEk-d8pOJXa3nR|4s6b+B+YTpsWlnea0;
z%@sYc-2ak)dQ+$X*>EYXp)+RC-0+I_NFnMgr;)7Olu!~5|54_8i9thhmf-L+-w>4$
zRf#qA=X>`bZ|nX08i@J|d~hh*JixtrG9wtQg>(Fld_I^EJ)UYy$Mf*_eu2>?;x9Y(
zQE2zaL`-PU4Xp1KrE}H!Ws#?%ldkJMlF(~74}0|B3LI`2iGS&dSi+p&aUP2w0KJAG
zG(Ci?<yEtYafn&^v3W4x?A-4Kr4P6q2r@(eqyZ`_sI+}aub<@SKenYxCwDcUOrj%*
zP^KD(!_ywwfui(zc$z1O4Yq*Nga%B)P+|`7f9b0kKqLszxe}5as|GW4=4A^1(H_tn
zavuI%TIX)gx-|ITVEM5q7^NRNS)CDfK><|FKzUR@Z+a@x1&RiNlOQh-dSm{todI0n
zL>m1zX=d8-^%*JPuOy+9JA~!4B-l>*I6N39po{{%gPslU=m2@s0jSe2AQVI1Y}N^g
zbp$SnF~cy8`T>lV0SFO5Rud92JmCtVa|sAx)syKU@kIY(D406~fl#2d8blYsX9wNQ
zlkOb_b}v`@0Dygf>KscytAEhTf`G68E$`w&AX&yC5@XZTaq#$(#!(i}vi=PqLC5nu
z)a?@x`mC*A09Q^T8t?)Cai&2MRS5bn2(|@dP3(cDSZZTy_Frn3t}Km!B&RbT1H#06
zNH%Mx;4=M7!PyIs%wVP;5Fwg9^zqyV?HnK@2hB79u;{OGy&F^z8zksNehY<6G?ZQi
zF^J)S5tMKN`rC)Lc%&Ei$|%E=^ghy1c`aD!|Ll3-6iBegfL}!Yt32gZ5fC1967bWY
z)Hh#Z<H_akL@0r5p#^MLeK)@pt8C<}+$rCmYyIZ|g^Aq-N^}E13pzhQFao9&z+eEV
zI}0&)0Tc@E1bi+QIXRI3OOL((1nhF~L9vHIiYr`sRcmTL(q7vUPn%$Vyb&+bZ;-O@
z4D0E9V58ETU9C@Y9eKa=`fGkc{=>AY8}9vAyU5+YEYqIeHoFE6UUAvXy`EeCswMYF
z1@SU}87eE4FBMereARN~cQ(1T=6fDE`JjG<o)(AP7wBmr)bBq7m4bpsdQc2DWa~TV
z1ZDx^9}+PqixR-6rG!$BZCwEEVMODk>1Uk?6RUsV_1WQ&yJU6#8;h2Qaz>nrg#LKl
zIj59zYgvAvy+G=W-&G5jvvIavwyxqiPooJKABkfu`HGHOU(vs4Knl-smUukn-GXXP
z%CCOVu?8_EGHfVV)?@jc5A_1x57=2pc@Ii9%fi;DMJm&2d4NtqIt$t>)(w02>m>io
zp;n+Fm=8>@It}719sx58P5OOlAes!RSc3P22B8AYptSMi+VyIZnDHgCJD{071;Qie
zft&U&-DCRPrJ3Ib`L&Hri$>nq+sDg-<O(D_RR)D>myCGHDwZ;Mq0_$kErwHV&9oX~
zvKh5De-L>&xvCRw$nAca)jzsg2GD0)gA#3B&ayY-YEBcJ5ACgh7RU?t8U;sN5hbmw
ztb<;<A6g+@>A2`G@S5YMjwX+mX}!DUmMS~(MvmuX%=s)frC;{hivBzs1;oAUf5KhK
zC4U*b7xi8fuSVOL=1@S}LNkc3M1O_Q?gURA9|pX4a14P10x;ENegi{#%BQn3dDaV_
zEAS7_4HJ#81Dh*&u>mN>f7>WR#t>p^Z>(f@43}QmDp;*>6p79z9=)Ivx}f-4{P*JB
zp9;Ig^JhT&AmdhQ$q<`520sc)5L?dgL~omqMiJQC2W$eaPTxL4O>URwnJbn#i)x5$
z;xo6}L}ffq#gB#;$ECyuJzt5{tBV=}9e!db^Uv)>tZ4Olh=w5)y@A{{Q{bnC$(^sf
zzvuT>Kw|j)w>z#`!;U!ch}4Qp^q`%7TpNnUKLi`+Kh{wP*cC|T>zgW2qLKgJ%9G{g
z{Zs;P2>Bdf<0kfjEuJoYGW}(eX*2Y+yNz$m7wRa7ri2Cs#?;a*tUtX(H=b1qlKa3i
z+e6tLs_nlMZSmu!lpE*>z+PxnOP#W;!L+YA;9C6Jhou+#w_$DhImZrD$I0U&dkD&~
ziIx!V*J6s((-TvuY-5<)%O}W4<aUVcthw0a$><$lNbnWp+_O-qWFbs@slC)n#v1mr
zrR?ht4nl{)W`V3?J`k(qP1gDt#J*rTTD%?oR{cEFDrgXGKm@S?cyu$!Z2iYhK}kbU
z7KQ$h0ZPSB0;iJ2uKM{+%E?&~!GF;GAF#x^5G~+IA7K3Y+g}*ilPuXIeI&w!o?f$O
zJ)Z%uI5KeV=JtZ%7h1;CS<d@1%`n+tmPadrb}n|H?l(sVL2Am|Ap=g_nFP&Hgo{DX
zIcE#7IA5;|1L^$}dfeA<kN?TdsT~I`0Iz@B=xuftcO+tY3aT#TO^wt0k$gi>dVV96
zV2usjKy@@4EMhk)PP{qjg|SGm%S3x9{Eju$dArNIQBXYj<|)cN>9Y(mFQsTm<9TxT
zO2CQB9X;TUftL%Z`#_#6eM4$4`G7+*>F)Oie&7%=Lwb->;LO=nKw!a1kOe)ir4KjR
zIyP}+_zQIuc|5ZKJY@xW>YmYGhEh}0(eUbLNSD(^ImcKwc=!;TeFe$X+f`jKEauje
zgip4hJq-M0X&m@{V9(Z~;+aS?P3HD*$$|n!QMBz{&^_k3iI((LN#b&=iuqiec`UvS
zpI8*8Z@JtD=l}@~onEuBPcI#d;b*6+4uLisN;X$I^HoFv>F+#;TTYTsxe5I2BBo46
zEyu7e>fmyepoIM>!je!#&xR<zSHIphvkX<to&+eJI(MrOH&n<6O2R3CN+mCakNsEN
zB0&bL9=dj}gI=M$=YSfHEa#%Jvt=Oh9IyX5m*k;uW4Bo~G5jC$@_se~!M%C&B_B}X
zaH)^Kg8ALlCscv4(_%$H2Zrs|yVnM1zp>rg&wFIzawCdKFu@>Wiyzh(q9kAX^5(kw
zGs`vU87Th?fHyj3E)i!2lz>k@ZVBZ=;DHrYxHAu|zJR=>FO)U<4p~`nskD9qeo63D
zM@3bdbu6Isbd+DSM?TdXdYNJsma14XcICOimmeq=uGpEVL&rXEN~t3~v)iim+K+y$
zt80p(nbV5!=I}#im%p->*x2R_9Qn;N3YY1}oZ2W>(GSnF=U)kYW<l7kO)@>;EJ$}C
zfI~vZg8=*kW$4f#2!9U}MTDCCs9f&%C!A)bP%J=l8Z3|UT79C_3foZXSB>u126^%c
zd3!$&ES|w|6j91jJM`~N-yz&@24Q$U>Z@AG(B;>vLRD}3qU%!2T5i4MRpXuilYdI|
zj+|v~Won$28|Pn!7kxwZUSKO4L42;ypw{x?L5Ktho$SE(2UVCqKmr9wQ{+Adk`r1E
zz!|uC#j~Xm!+JTfV>7}nvl^W`dG%2TVpG4-$r1Sl_g39hs50l)m0_oor_%EPPybvX
z^IwJp4auV%S$X+F_2J>PJmO_sMM`q2ZjcG3;_l6<F3^UY{c$VMF6xB*Mo_9q-!<S>
zY%0F1ll}v^i+~Lc&^!n{fn6^T1=%R5!S>1Y1JWDY>ZR0UPCr!_%Z~`|;Xqs<W$70I
zvckwobgJtmjm9LI;v4FRJ@p^=c_7+;e=sWfcHmQhlZS@oP0)MNAn$Z__N{)IT)5oN
zUG*3JAnOR>GoXA0@QvvzX3%#oPXPS1*Cuo)5&{jXfy<1~ofWa8PlaEFt-@}4hD#FS
z)Csw(C#FvTQBn=Wox44!NV2c@ZZ2l0$b|^!U$dulScatsXvLd+UDwFi)5>#~(=H4H
zrKN!x7KCMB=?tEXb%FdS(6U89Xc|D*ARj~xN@gZNp7*~;w%Sa<X&-9TB0%4b|Bx*k
zx5_!JOOUOmg4RnSQ*hU{VN>&(ytw#Bg?hfG=Qp<8hwi1%4i#K%PWA+y@(NaZe{K0j
zq?7fRw|pl0hc(X4uAikq&8VTa3y=%{PoWlgKJ>8f-!Nz_e-a=cZ`ug-S>B8L_thi3
zcDY)-{v@H~9zzd8S=ddaidgnb48OE7>$HwXAs&c({$)7-#ylXe#v-6Kf9ZJv#FS-y
zG({+gk-t0DY`yLDvA$K8<MW@)#H|1mYA*;>E<j0PC?N|;D?mE}#3rG@^@o5vaCAh2
zfRb)NAda}v)}hU#>%Y3wlXy0EpJUqwPrx4;B0nYdmoAVlNemTK@)?(-cT%-IWbka}
z@@~rE+?$paXK_|4qW3OSdSd6&K!X_2CgHyT91FeP-{^nDyU+<upAKj(LL@t=hdS_%
zJ4@H-S*^qk@Y=Xn6Y{DSx#BWbw;!2MUxMv&Ob2kRA4@3$h66pTqywNGdQc20eL)l;
zE^R250oV#?&J5*hfe`t1D9f!`3c2|i6?eh2LD(t;rW2EsfGZ8Y6f#PnsssXA|BVIO
zpm30&CW~Bw2-RRDzvz~WWxzP<M|_!1GKl7$VL(^`)hB`K7U)^x)ac<Hw5`M0AwK=T
z+#Ivf4B`{yXi?3NE^(2ow%T7bgL{G0GmzS%zXM1`3)&D~pjH7S^`YvZ#sj{YV0*68
z{x&fPaqru=k6}OKK8FAP+86AZS(5Os`}^M^29|w{`}S$=n=aYF3g#sLkL#i`<Nw2@
z=1h6*Oez20d`v_1Fnqkv{O<qtCKLNI1^>q-%ysxZgA;QwxPnCO%$OLU>mUnb!GQ_r
zN1tGxeHZE1EW>|)`uFGm)9-n2T}?DTA{jDS85X*r`yC90hBkQpKeznPjhi26KCA-M
za^l~#>}UM%tT2FSNyseewtM$)UiL%t^5=cBde;BEE*`s9@PEC8g8y%acFcwH^nZU5
z|MOhm<-urukDmU2bsY(Y?f?IFG5)fhW;V@*ru?2(oha^qy^5a<-5me_U5C3L`2FX-
zLT}@LbNMmcFOmCy_|mO<5Aed|xfu2x*tehQ09deKL4)_UkAaI@QFOo95d||l&x_jj
zUgp6M#1oV<kD{{ZD|;UUv)2B_#BByGyMqf1zpel`@nE*)Jg+^n{zfqgNeX;9Y%5NJ
z?`9Mqj{mZ6o8dwVVbk{q^!YKnmCRqjLPkH_yE{~WMcyXJW)dl{bbe`tOwK9&ZNEZB
za*e7j1)I0UN<3z{L%waY!7^oxm<-l7PB%vdTnJNZ@g|34I8yF!F~l%6kdXp&aGt&W
zPe<;Typz~;G2`T3dTZ8z>_gj^I(ha^i~k~CQ`0tIia({1$|G#>Z9dH1oI@zmmX+_5
znDNrX_V6%H0U2GI&{=#DkEYT{6YMv^e2IeH*8%>Ia&`n+;|K}fzT<>l5*9X}Xhgtx
z-`t-P%lggVMiMYS=BdWW6~Ot4RRLDj(Zc;;kj$9S#&`Oy*Nmy|%R*emT@)Eva~1y6
zqGR}Jm6CUw=t`6wnduej?ZG4k7S)CTZm-`D?l$E&m3BB@h>ynxIKt6X8EkQL-shjD
z2VhekgfsF&$*Rd`I%1iX0xskb3rjX{QM;2bJe)Btt_W1ttnKKMn(UfRvJ#8(h)`{X
z3&7ZBtJZai))Q(sPwXnL@le13NO%8i?VCAkpLR_m$=Xgq=t0}2C;5QdFl^)9G}~pz
z4ScERc!s|mm19f)CZV6&PJ;hhV0i0ipDXu;{bAGT%EQM9o(ini*H{I-Sw~#AS7U_4
zT8?MllDRR`5w9x&%dka~;F#xBY;mhp#=gtkAA~3ux5rFgg(|8=d4`Lwe7H7~r4vCC
zf5qT`I$h6Bcee5NaFXYUX{u?iP{Qid36I4#HKDd?9ioljn`<?X+ay5~#AKy7uT9VK
zKqq!Yeg1_4!UV2@uyZ#+bfPhZk>ywfQA;50vSbJHFx#wW*3*uli$Ow(541v!0)DA7
zr=;6<-nWU|ytH*OUN%!>;Lk3_KDkX<NX3CQVpxG3<>?vUFoKv?%6Y~i^!`EGP415u
zN@wD&DhRv0;1LAT)Oypx;RIm;2hUN)PlJ=;jECOYX=(&6*27f1W2M(4N?qQ!<|T8$
z)Y{|{TC9Xxcth_r|M|EV0=?|wK32@5H8yOajWWgZf$5XeiO4t6rEfFuyC7?r90~n<
z(>5yYWMu7kuxH`{e)IG^{TtUZ*9n+M)Cn5-itTLTOvNMmKq!_Wr`0jzujb^yt!ZRr
zX+<D=v1GdbB(hctN%HgDdndBZ5s(UQ_3USZ#0EoPh;86ue&L}C!ZI0|-oJI+a#**y
zwavFnSYE@RzYdNWEB!J4sp77s$xAugi$y<*?0ywe=lO~Ua8=~16q}OU-}WMJS5Pg-
zOFtVq;R{Oy!sA>O+SNQG9BE{$QEKr~&bNw8?79ce#NJFmCy#0Gm7W+SYz5rOoKK3g
zXV+mhZWye_c@K7-I_`>ybhEu1_vQ_=JXe#V{>28}%igVuA4IaR6?clU>k2VGXjkiU
zu|2!_`4L#}Y?fnVvl3Kxsr`A+HBv9+YVatp3Cxc_eRFa4Ar)885w9-+YfyhOlO}h1
zAAdNz=Lwjfp^sRwpOf1_TWu^ObL{VlSfiP-+?+JgaWba~95eZ(Z3Q9vT_9BH81seq
z4`#~77+MSBl`fx`o@9>`i$2CK_AvZH!oyI$5drC@@6i2fypJu6F~fu&>Kg07GW{Nm
z>5w#z@OU%o^2HIg^dpq-9(i-Nj_FRCVg_E*v7VtdTOd|K+G4)yU01d$<KmWn{V74k
zQP@R=HV3Y2of++%f__S^N8315)5H$bbjCOhKPoshKa4e16@7oe@8anxgu+8+S@Q<}
z24GlruTon`@MXnh76*aS$xg|b0|E|y?+>!nbJ|T=+i`vTabA2^|C%3)Y;%Tm;Z-{<
zo7OFe6LRqD4W2s4>9;TEXsFni&s(BgUZIm5hczRAyTKd!so=&_mZ{T~i^YPNnbY(+
z-XbIGaq$cLT7YUw*H1Y(=22j@@BrELu4QP>W6gsa8DRb2xx2pl0WL6gCWZ8?jj<-)
z)iyjRPGC&dE2L4hReCa1Ae^X{pe9;=i<wK{yL!~pJv0SXG$mYKV%y}W5U;GL9mf@&
zeo;H-fTF-b$?yZB%`vfzE&9=_?`EglB?gk&HltcyW7u`r&8|fO9l`3h2`dMoY#xPo
zYWlna#Cur_^S`*(dWaKjU80I!A7vF;B!YDr9tvwC4q5be%?M3a-h9F}Wt=9a+kWhT
z0*m5BZ%&E4j2`Fp(hmacjG^e2gNgI&8>FBv+^#Lz3Rf!)nZA1+yL#Xc6@I9f#eL8g
zEeZ?D<twy7Q?O@?*8}^h6Q2{un=6akb;UY3)I<gZTKGt^*4}KP0*YqMTp^q|E+=Qi
z(a=5I1re*EDza(d>5=!|TYgL}O!d{yW_e<Xw}SO~<<IZ+uToo<n}c}r^%T9jSV~ak
zEhoiCVfn3F(+n}~w20gEEo@^Yug1i_uK^t>EKnLj)RD(;O*XJp?;5FI7G3cvd6yOJ
z<PpvedY&@1FjiB^L-i_%5I5Vu4BVgFUV5y*_{(5gJTUr~0YTaty<)rI2%o^!QY`0B
zP$y8}tf}vh#)N?%!Cp15?@ev!U$2?|?A)q;LB$VMeD!M9m7%wT91>#fhmSoBZV^xv
zDHvlvr>9^ob*M>M>`7bp`v{Ntj)idZ@Na-B$7bB9n_%Y5A_Sfyj(d1`=pacoeT|H)
zll;wcD_A=&t%aPsqQiOoh8n`1iSyQ=$AcyHmVRo>n`O~v14fQ7T+1u1m$k>LpZw+t
zHN2JBwoylIsg+gG3=yhE9ly7@JQ;WGX!m?Jt|BOR!XIWUduzlsE_*udLbIB-6I<wH
zBjxwotzx|kZ=_C?ic;9kbdNbLJN<+c8nzR}9&B~QAjeeM9mm(R75C(MJVVIQEcU`&
z!uEnsI*$nJ#%6ilRk-5|-@RY_*~M~ugST**gX7J`Mn~132@OP{`6GfQ<in-{$YHsm
zm3n^EOE=uxU4ezT#~<XaUN>Ff`@!l-l4}Uzpm3-n#=^;PWG<VK#7Z?TA@n#h_Py;y
zIU6=t(G$&ztu{5@w4JFfTzMqNREhsB2XENf&nx%!?V>>PpfSH{L}ELU%gs<xpMQ{#
zHdMm69IA>7e0KTTqO${EGA#Rj5Vzeubdh-Un2ckYQk<v2Jv_F!J!|Yei+G;+@<uCn
z=q!E(+9+$$ssuNfL<CnZYck7O60(fe?CfS(Xu5-Vp%W5?+L2I0nA^2}(r%B<EV&g0
z-p#nuatOPbL!4v?o5kD~CKzW0$0R4q7=b&T-qCjrPJdce_Ryp0R`9mm&VZ7tm&2`4
zPXS{CNnIAZW)7qkMU~5nA*4Te5G64f!jYFCkOIs8%g|(X-zR&BoR_%IXf<sNxL-Le
zEvpH%_c5+j7d6*t+$gI?W_G}{bC0_ymuk6X#)T5pAf67Z$f9MThDW^kq(&Kb5CXKS
z)sO3Q^TU4MY!pReH*U$neuy~A+^=-<+i1Pt!W~<fyc;3>b_MKOdqV%Fg0g6fB^$p0
zzpV8SMWO=SN;EVdaT)O|7KSkPh@{b)tBF-?8I7KqA&Gt8?TqlSrg93Z5<i+<EXzE~
z@g+?%i}#f-co9F2rI2>Dqg9EzuEUu$o;|PL2VPJ6%3se(8XqLydJw$UKk3l8zj9L8
zu+cS+aS``DS;kfK^Du8id1l^-=TNNAcZ{@I^Q3zv4)3Uyp2JEUYvAwnM#G#pr!o{8
zjrn4Y!{xmWb1JBYHMKIlO<u4P9V$Ee0FR4p4<qAFpb!2~Yhc{>Nz+kIeDnAZw6cPq
z*8`T>(_log%uMB(LuG<#Ic<mY3V6)%-gfuSOtq~a;T~y3$|7DF^T`UOo7?GEwq0Nn
zA=UKuXG)(X8@o^0j`&x8UG0_W$6Ai5s=0Ps#i8$nu0LC7$9x<0<74+@>J<B9+3By$
z$F3WMol$NNAFMX!j!hnP|1REmqKw1=zjt4Au+hDTd>9Wrg{$(lI=C<4EZ5Q5F=@I#
zpEjuqc$UgBJ-8Ns=$!)0ThsE;$IRGm1&LAeauLss<aVoA`?uJcJ*&f(ffli6@mZ9Z
zWFpu(ngdmBIl<T=>mOF4aY8ai0?fzK_H$4=2yR<f(tu{!*%!jDszwey?ebm%i%~5>
z@IJZI6a%J+ArZqdc#r^d@6$#k^zqs5;J#mrPcN3DxH8V_f~Ro=uQ0Y<mhm8ME8$(q
zHqN}Q+6`u)e~{&Du0Z7rt_UP7AhSD9RV8vxnc8~v!MpNsN!QjRLM!`pl{88Y%lIHO
z+6A9Hh+sXI(9BE=*JrBvp-nzg{Ultpe9vswmMt=W!`3>I?N(@!#Fm82o$wzGN_E8K
zJ`B~420ouv^)CD`g9!&gcu40+FpEz$yN*C|a@A<1n{DX#>VQ->xAu{|-qCnDTY|0q
z!Gy^Nx=PJp_Hp{sO?wK&n<Ed68`!O3+HIX`Ovq57^2(VKT{6>h<-75HRqlLoj{=?{
zNsfeXBp>yK=gVTT<*;{^?dJNOVv|`0CriQwu}MQ_4nrwyB2@=ZTxm3JXZg<MaNKy#
zETorKdGUT>26|hf4VJ!DD?=M~Yfm%Y-ztA3daw$?wOv@3{5uN4@SgdLxak{yI0prX
z`?`u0^SCUOwNd^d;1bvDt(9@4mf;V3q-Z&k+;o2*@E2E#t2Ufv@u4e>V8+8U?ju!?
zmEX~Kwk0Sq9k3jVCn{bj%^TDa)yT=stU9k~hXu8x;^kR1vS(J)ZI6%QaY{5<inT$i
zmc~y)e+3*f(pTA5UaG2%eVeGqOX$I;F?PK&kR;fVPv}>hX?saqtMbY!_J1JB?3Abz
zDFLvv<X!E(zAQ2VDZ78L3Xio9(m0_(U8!?qbCn#IK}EXQvhx0rk{A7||1gwCg3Jud
z9=m){vCusehI@t#YjDmjgq_XR>=s@uqcxH{y!iX-i{CP^npL@<`O+UowHzDywP`z_
zcm<Wv(=C~qkHTD-eq($zRxTk$(tTTQIOYZRi%xJ)+_&U!B=1E$t=IxM^X4`j#M>>%
zP2K&DOdEV-oI-c3Wv;W1E&Hx6&RSlE+-#d852ChhRdpm*B3h-Q08Ch?LA^Kg`G--{
zZppjt29c1j67pO<V}yoX4&Z-}r}MJ8az4Mp73?mKsiol4RO_!yRyG6-{2}t#!+DfD
z+{Rcje;IDuW4YESZDB~_G)L7lk9wIysAkjw(P*+^VZ&br`0?9B^9dZY_ZaP~dZL)?
z@b=kHd#Pq}@0uuxl!F{Y3WG*-Ur#go8oLH!v?eWsW!}gc=!XdKOAW>l{Ukl2WLa!m
z2YIQSTyf8BK8fuQ-YGwPThMD870@q{#f0?PK25G+kPbzjq>(oo9SvvZ%*k#*IEhFy
ze)ubX3!UpvvDUA?|JjMROb`w8SgEqskHx#kECcHzkUhPRl{KwGO&>SNOTiNbvJaPO
z1&5GebE@$b<m%*PScwBS&5ayncC8P2#wpWN{na1DyX4}U7$zhVR{Sz1T;BXv@*sa-
z^nUC|OsSZcKT*X$6Q^B%b0o7!XssI0v=;0pZ&(=~?^{tM5nvpH9QxdWj2gVyzR%jA
z4`#N1w3k-&{_X19Se(G?fG4lF^gH)Ot0uGTo@<Z4-S=w~4}<UL5aB#S>n<<t-3&PW
z%5CXgLs71|K0`Op_9X0a_vzA~tji}UW;h#ew<lBwHKvQ+eZw%b9FcI5FqYl#j%%;A
zqxUM)D$W0(;<Aj5_ix!^le#sYnOeyBHihS+=TxGcHq2F@BT_GHUs<9aYtT7d_QQck
zL5;-9mmKQoN_I|8sHgpa?`JG%7`!-fUm>8+XAZqlObR?TnxvX4c>KC!N4w0fg(05F
zRG?Q!?PnOZ(S0EODY6VWufVBOC*3Vjkmb}bua!TtmZ?l}$bGxOX?7e5choC7GNBlN
z?1m|rRpH0vb&ou-qmeH<{dS_|C7w7lR%y#pJf&ASyi@vT?KTJqNYoYXPBZWK>_~Mp
z+ty<|R+D++Lzchzm~uuqOepJO;c4w+{DL~_#s_p{effq*s_WXr8W$A5I|KVidb2y-
z5F(FI_C1a*Zn~+_7tYei@rsF^5!>D)87*sjo5g$op0eB*1>D=@7k%YpWR4XWL38e4
z%}8`ItvsV$j#6cxahyYf{2YFBU!IcH79Lw@s{_Th0Y`Qi%qN8}ZD+mR_a_0I`I_Br
z5qlEwK>nldRr-W2eF7Vb<BhYnjU)?xTdnWG#X=ZiM_QYJFP!B$z8%@S6moFTaQ41f
zP+nT5otREs9<t7tYS~uINc2Mxt#&E40$f(vtzwDx8}%_L*eoh)6C1gcT_Jy#HoRS#
z_|&03zc7JBb~azG4m86J8|kw*@~E5Rk`tPRqW8I1EzF@Q7DE+*0xUP3l*=Q&r7{2U
z`g1LzO(*xP!>CiWKEW(nFnOagbI44MN5~l5XX|2SB-|`K{6}k4br4(~yB3#84&hW2
znKW8kqq1R(JF4dQhjBDNfD3YK8<=H<=Zc`%bwYVAcbKw<EyqsrJnn5h1<)YO;O#=&
zLh{u&uYN|MVy6O5+&$I5aRP*R(>8JaR1kkH+Ipv(1w2^-XLrjzjI?n4&E6sZT25{4
zOzGK*ZaEtO0D)-#6y#CW!CSVQa?(1M!<+&Ib@4~Y#t{jH+O0@-JONGS{lj*z?SZR}
zpc$UBc$mCvhHVdmb00e)@2h;T24E92zh?^Z00*!Q6225JFYp?{(5&rNN_$AaVcYU|
zU;zVrpPG1bG#Ay4<5RdIoD$I6qE?D6mfXJ`VmN()Z=C5T8OOvA`z}CnZUqPZfpZ8S
zM`l@TX^sY4<FLH>@@h1}u2q=7FMjmG%4yaW^9dWiEQaQGRc9w_S@BX@$jAvp{?j6$
zh@slRN{55o&^yG$$1d;#lMjm+PWAJJ5R+V59sr4s5_Z|C_^S4bHw&0LioCEkHgA-h
z_4AgDBP-v}Qhc&t$cwNh{yp@zY7m(#uUrb5;VT&lCuZDHyAQWUq}udT3216J>g6J?
zB2bKM^UB>|-t`V%8b0u~Q-1u%MnJ#jysPAVnqT=wA;3+LW%aB>wbGV4V<=9ATwYZH
z78X?ApobMwE3A?7!hG4@1zR`^b}z2N1At_6m@&=M*Kgb2C1}A-t`%UP*fwE_LM2Z7
zJObG@+)4I<FSc)k*EGWw_hoLoT|Xa|U7k4^5>mPBJZ+l`Zc@*PDoQnP#Af(188Dme
z3*u&C2pNeE6>EKXSigd>{o@<S_Zi;4nXEH<2!->Z$nq_!qe&di)b|l2WNJTk5`;R%
z%IQr29Rf~GmlaqT#Sm!KJBR6f`v5x5SVlRbLog-ps<_(sVRI}l74M3(U80cQ2x%r7
zn0?1r@2mE86y!afENSG-!p|MbX1i&=)LgAN=8TGOJe)l2iv+mIxv=pO)N`vR=&FmP
zMl$l@C(Z0%FN?EL#jHf0Kq1w4`N#O#g+xc$8Q_v_J2`6=XNdwGhUEbK8OgM8W*qTM
zIh@BxwJ`i`|2s{t$IXO{{IOGDE!>vXtEw|*=IkYZ=0kAhj<_+)3=g$Q8La=a{ae>^
z3vr2d<NI?bY>nLuO~(!>z%-xh^kp>k4Vk%JNz<X}M$je+ZnB6m#JC2m@&)aEmx&T6
z;9<k0<xqIE?Ps8UvS@2Wwvi&5+l@1_1`vN(s|tVQBid{03mqub<^v1;W7VdaIc!d5
z);;<#`C#7=9;I&J$^8D9$4X<=!#Ue;Pc`TNGPEv*KQ}CraBGlJ_=>YK#ga`3{l!5n
zw#6enR-%pjyS?NEMBgT8lM$27VkbVcF-eRN>>d)FFMhY(iY;Er%pvoLAP-Z@ORWus
z49NL~_X5%4LtcILbqXOveit3oF~GgRjD8+?$b%y(IH6*^q#L-INz$?c@f`+aK|I)4
z8@MN)6lb2V;t2v=kup0&LY8uQvyuARZVB4PT$rrfv$@zb9TAvF2OlkG>;$~0{|_$F
zlJ+O$iIX0mddxR}+>icw3aH)yAmuU0WULJovts7&pedCTv_Wk%S}&WU%pDS|h_7kA
zIJ!tym8dM2%HKGYEVc60Y9UxB;7JLKG|e8*RR!-~`xJg$+5U7*u?d)j3iz@caDhI1
z&=%XHTE02r1{aN6?hIMBkKl$I*2AdSHDF)sjjw6~z?Pk%!Lor=7Iv8JhSXqOT9sam
z`BF$l8QY>w`>?SqT1!}uRQB+0Y<v29(QI^kh2ZP+{^wd%*9e%fjY)XJ_NB~lAzN6u
z(y&~Qy+BGDh@t}w0}4;)aTteQ`KV}5Ca`n)$vk4Bh*UnfdYdauEJ(<)fiGE>YbxFC
z-tk3h!<%3`F7FoHz>=Zu%ymCOvj7=iLG(6JbGbL1d~!l(yAj3Cm+?|7ve0<z-TFuh
z|JcF546vu8$Mi_#x5L-f$cwnzSu9^Md4PWyO)X(&5ZyAv+p(tn@or5dGl}MFGYLKk
zD6XPCB;``Nu4|VgrAAC^n3(cr-|ghP8E=2#(inKW7|L$1ms#ne&FSoK8Fp8^&TUPi
zOP0%PnMTHkQ5yL!Jc0#?qYLGl$cHP)Yh#qm2{wZHCMs@Di$(6VppFQsnIe)KIBaUe
za^P@dL^!Lq?xg8tOJb*{>4k4%1fl}^FT+=<L4K{UvOm|Bt7CNjpz`E;#_)x5Y!S=d
zRc<n%@G1r-vs~Ry6DppI^nO=7Za`|fHxpb{&RJmA?4{TwT4|QM=hyn$qb6RRbm78S
zz08dwbJh3gDz|=kW1ybsr0&9UR$g)3^?MQ%t23=yt`do|=TnI-D9eV(fP|gZcvL;_
zQtmkxJjFqva{l^(pD8Uz3P!OaE@2^|hD$64=;Po>t!&|y=BQI%jn}JZ2iO~kQK%nR
zb%j`oR=RuwM3C|j;E*wPzV%g!?=M5Bfqk@=(-z~5G{E%)fm_vBPQe14PvyayW9;cW
z@Bory6W@}c#xRy4GlVu>^_~^G5uMcS#l$SnDn#pp4tEfU!Tmw*MVbDGO};{5r?#r~
z>Jq_(PBR%_r-Y#xQeFOUeP{3UL19c{2L+bMJlF-kbYHl?<#%;V&pexD$6I}B)2abN
zeEi5e=|6g1H5DAxL;76hnN_$q!MqHaKZNYA-4{D-3c$$I*%qUrU1!bga+Ng{^j#mo
z*l)X^G`AZTwLY86E4?H)q(f@5@#f0h!eQoJ*yi|(s}^l|XX0~`)1;nXh{FFC(6A|~
zXSM*-VpF}Er`jMUM5bKE5U)vIcFc-Fkyz3qx-!#CUoBO3qDgXwhp!-OnkP*Iuywv`
zgtpaSRQFr!$}dQwQg0h9dz$;HtjYn_Fm@ynE!2|S+a|%u!z^SgS0v!G#J%v)RKoR~
zT|`!1MF4a9?$s^idLKaS!=RaBU6A<1>9zOWlN0-b7A_X0&o8qb++{hIGsz^pWbJ&+
z2sPv<SfYj0NEPJo#>F_#(jKt#DCjEi563-4z6pnXlJTqIep&dib&%f5(qVx+M30zu
zO1hb_WKRkyPX+e7l4I3uzg-INZBz7s%~@NKN^o{^Tb>t@AzNk>ZFXieklDTj&fm}m
zPR-@Lv0eMb!Tmh!NRC~7e%06x-GlN<eXQ{v&J+Af(eaLXhj7n+=6M!MK6KWVD`nM?
zgZt0^9@G6K5Zk%0y{}v5$VznIq)08ZW6ofGk&V|TDqz@=%-*pJ7z<fJd@mUDykanh
z2N<2T-`yJ(5J?Z9F4y^9`y|k?iC^_q4porb$JfvHr&&+#+mlYo+_N2v7xkTYU|O5D
z_n#~DJcR}F5s#yYV(yW+XQrlHFNEynY7(c*{C>?zmH(c?dwixgNzs;a6jtIJUa%By
zSMCoV!&aM2O1gFruj8?O)Y_l<=}x*Vo&pnsZ@NPLGkJCtWm!Jft=&U!6wX5H09nfx
zz_6;rDwNR>e)z3Kg6Zu?h6j$Dxv9PL$8QyY9r1XHVi`;YcX4*s=fpliuQcG%f%w3A
z)*u)GRdF_LqP%={)&jUaSlUA#w1VHUeQg>;d?exJCO#3NSDl9mJB}WNEiUcoF*o7`
zq-rd)$hV~<Z+YY;h3Bm8rBF)M#Aq{+D!IHhHZ`m5nX`Fu+fM9&uS8th{s@K8{1abT
zHYRB=WzsI}3#=sh9__waZ>Jo#+}|lINWN;eCEa)*Byw@s1^bBG0dfOv&brr?k}9p?
zTAIOk<==3vBE;1gIfcZ}SL+-N)>SfFV?6#d6*F`&O(B9YF|RIo<D!Go7(TPNQ}|+$
z-&9!TflMoJ<-*^dqwHp9Msmbb_?SfC12<E-ob(?uOUbEn{$iW_^TgS#o0jb*^RtIm
zRPnQ`XrSW$_PSImmmBePc&H$E>t#Vt)%01C)@deuQZuh`$32J4&jc{j#ruSv@TD-(
zm8NG1?qIR|FcxtgmOIz)aNNFXA8-T!I{j2^VWBN-^Tn>?y79uTaok#E;d~3C%{Pp?
zrV>W|#r-P^_?X8nXMeI{#tNo01Tv)l$A8+$+WY=CJl^LHzrDdjX@Z-12M4z^Wzrs?
z;>Wv(Mq&8z#!s528fNaTzET68vlR_yDd~FO6}N4llC?+zU1Q~G<h|pQrVa8~Qr<KW
zIu;3Q{GclAx55j@gi+hpQT4@TqluakK78M1&aK}c*8&RkivVK2mU~}RBl%?@*QR+S
z`Ylgwyd#@AUlpk#ivz~#jL#u^*?Eb|^(h)~Pi0sjOX}GmXM0y{t|nbTPCjVYfU&!)
zj3}~6^v`tUgzIqIvWNxaG!s~M9bti9w25=9BQdV_Z$Gy`yg%|R?4WAWyHdF-<;@H0
zR}D*qC%yc$n##*fc&@F^8k&)r)tF_rwy+s5l?xA!?dNXAF6Nz($B&<o)nXC*D67l#
ziqKU5i%+Px1Gum8U6{W5!rZeYa%`dQFl{%}L@r@VuQ-j71BtdjTAryesEyyA&}C5E
z@<xkq_cDis=UaRXGj=@Hr+mPAoTDsBEw2a5KmNGiSMxcNr|J-1I!u;p%`9RpPV8;M
zP9~4&{-}>!g$8G2gN}vmyR2M!clu|qvn5E>@0)e+os%k+NUVC~LQXIFKUlgFc&7jV
zuX4<l&3%+G%u()Ca%8q)t{f4$a^&cebU4clbLCDHGN$CF1Ldq_shpKcUlY|xwKQkw
z|60HQ9^c3J`z^M8KJU-#c)p&m=j;96#LgibpO=5r;Sx=K;QTyOdt24k1IFL;)s{YQ
zi->=ISA2EXn~&+PH}CsC3G7Lbly{>Y-eDh}gbMb%q}DOH&ESz*zl}uIi9t^Z9&)tg
z$6K4zwvV2d@$@MW3eVdjiHLM+Htvf&s~OmdnIV2`IWlRlBGaX-aU>b}RBXl4o+ti@
z()%#&+nq}d&l4N(3!O<9<TkeU^89kRy}`WOFLHmwz5+Fdj@}-Ficg9Q+*reo5AK{Y
z*jeP(*iUlHcCt&fHd$6b-&X1|cH79*EZB(~eP79eFvqTW!EM%BaoJsa{VCzDb1u}b
zX!H*9rCrUI4^K?T$w!L>iZpNQ5e{|OJ6|pM3qQqS?pKoYg&TYmc&`L-i~qSl@Gu{F
zx&O0h#pi~^`O8in;%YceAK`n%ch+mScL`p4oHH*zaP(K?c}_Bz7tuU6p)YOk^uV#u
zIG(AJo_TusI;0skW+Aox%=%^1fDSgz^EvxMpQ-!3*zck1hd`@v)*<(#N2he@_77l@
zv9%7deK7&L9b>m1RXI<D%a3ijpByrLt7FNu^HfIl;PZ#4?~kS&jCMJHa+&+9{Ef3B
zck&{UgM)vjnoJt@Z=J8-Tym(4+~E8IvEF#ZX(C^6t@cB}Hg1Ren>JkMxO{KTcVu_*
zs%ht8vpQc_4X2#`wAJ*WnULHc#+^fx2{&5yo=W4<a4;1w%RX~$t!l}C*+;-9!dHD4
zyBn+ND(h)Tquo1Zy2Ih*geA=((k;n8ut$3R=4awWK~*oFJ=vD$TJTzYP%f~iEM(I?
zU)P!~+;UFC?>n+jP6W2M_*eg82=x59(l+^mA+4S~JTdBRsn*i(oGgGksdezG%7c^b
z*ZL9W67ylh6E^3SZ^HXR%y-b#v=b4v*(R?&+YXJM^l^p){^$*QkF47WaqGcg#DMbE
zO+snu0;>lT^5@4@u-PY&o<zrj`0qEO_wM~sR)*c)6e$i1sZ_FBs{zq!V4uXp8>SKs
zXJ2Z|*!&7?+8Q04zGVXa+5A%7nQ6}Y`~9xb?#^cL<9CoP243-HS4yAl_p$F-t;=U!
z2`}%`xyQ4s=l;KD47Uqf^IK8O1!BX!&{d%*0bF>(sn7B%YP=<H-o;P_Mem2-vv-p0
zKfmH<{Isdy#*#Vlojt9YfE>Ohni9EFS40?rh1r#v1Lw<5+BHi&_?-HBWY>%nT!QKv
zTBL2?3G&U)S9olMAWvaO7A|WXrrz&(t#~R0a#iaS17oyiplQ=<_CvQsljXzI&t;Eq
z1fH4_%X((ps-|$+JoEXd?`Pu$wk@91V0FJ5U2727tmSbSE4ca$5$tP{^`xAfXw)be
zB=~u|y_wq9w;yi7tw*|%zg7Gg?9kn3Lx`Ra%B@89itp+APKxAHi#)=cFtf{BV(i?f
zUmrf7s(M~*HE*?33WC^FV)Z=Pbo|rioCi<J>DyoO*z{zQJ?AaUj8}Ws;)PAgcPE?7
ztS0wzm7N}MD-g)AQ!A%uJ<uR-nEuz)(5Y~Bmx{$&$IB*}zg%HE$gbZPe|=Pu6c%nS
zIpFgo`q-b@hK_(B%?Q_*8us{}0yC@oH=dm(=I`27wZYo^ScZLXBEP?`H*a(L(fbF>
zNYVwuJL9rXYUykr_+^~3u#mV^yc&9_zpi$#<W|4D2g5Qw<ilK@g+hF12hw(VLt&FL
z5P{hw7!wEwhPx<@zzFI3g&SwfED}YwWi(6_d%iQp)tqii^3UDtQ1r98`s$@Gfg0;U
z^|~TRpwYaxza(}NEK}9pZr!_7^Ln+^r{UbJ_M&}f%v7#)x?yaq;!S@MA-Mp1AI&(+
ztvYF?!({?;&IFZM9cf-+0m<cp5APOyBNcCL#}9l6Iht}+$TK{xbnxQY_+WGFgS|JV
z8=n|7o>JL68`AHf+}JgFJKO>l?<bx4T#@^XmhQIFS+(0smbBp-?Q)Mt)%UdTr7FfB
zjaT!t?B~;m+C2S}Bfgg>^k{|BB#BEYSxLHouI!%fi-65!3kk)Lz2Y{!(&=e`xg0IO
z$CeMPyIFxUW2aBV!JWj9Ruw4cctki>L~YjTgEp-R@g*)hDbEwOCec%?@(tt3R5n%%
zNA?Kh??O-$gb0?KQ$KA#|0rGZh2Ed=@^>K!KbQ2a9tXoe=&azw8qdVzpG${0J?44}
zji$K6NnD0?R;RyN`E`+X5`}k4leZe3N!&_QYy6f!efkTRQitchql-z7@%==hgCRoF
z;<7H>ygb`zJhwz{5Y2)M1T3um9;+Q3iSIcqS$aT3HA-Qg)z=aZJ(TV4-S`AaKIwU3
zjpMHM-H-X7Y$F-x4x(i~^Y<T{WWCgY#(pTk$EVkg_Z`sH?)g>NrEtGLk(77j^f|Tr
zc`izs3H1WKuO~rfHrQ&J<w|`lGON~6V4MS-@{aq$^1Ps$2wh9}q~6kJr-i6h9Z*C-
z^-DtMSKY1ul)g7>{h$o6yZ5_v@uUr34>=o|zR4=m20GrdL%V+q*Hdu|B2NPR!d>bx
zNv-4RCyN{luFh-u)e8sq?NXb(r=Y8G5Z|NLk*EJj@Eq%dcD{=Enb+fcW{*R6bn49I
zkG#UaZWySAu=U5!%^VkG9!lJxIo*2n?G2C(uLXPZqmI1lJT&7*rTNt5Cob&(A>>XZ
zh#xm?;oc0I-v7P0HlY-ex{eAUfz*jVP<Al8(KoDni1lmo|7DPy%oN{H>yq7bOss3Q
zU$x8+w%wy;^ZXuEAi+>S!KjMpR8X1MNq+E`>-q4Wa~?g&A}q2ft0;>?DN5D4Sxwq&
zBH}!_Fr%TVsqUU@qEmzwb02YZbmWdfOW}iS=<YD`HGoq@*eUXl@Q?i486ANQCbbnB
z1wWN2RBP&2qjqpf!Tf@@*C=gxLrBq-e}w2n%_v=~MTd1C%;J2$eLOx;BEN8YKxKn>
zEeuW(s##NC8Kr46u*l+5Oh2X?m8nKAx#OKOKTM_8QJvPfG&SEE!6xhqg=RL;b0Tb1
z-!ck-eZR3Nu}PYG)M@-O+*TfKxWKbek3b+`FY$0LY(t}fT5xB^X5^eEOrvx5a5o^U
zW|1+BA7;+*DPH{I?cy}`OPlq?x@C57$A~)y7Q8AwJHBK>pDuAwO-I^^SPb6gyl7gf
zPOvrj?-0(HsCyUc51DGf|7qj3s=qCEY%`H4M61<0_#sj$Z)yg*ZQk0f?0?$)cEcrO
z-APeK1~2Yr9gRYjagX}mCh;$~8aU#`9hp&L$p<8TROa0<tv{kc(~m=O24gcq5_UWv
zNK+0x9xX=O&3(LEHXysoK!bVS3-Nr2e=Xv^&&sjeov4lHj5-m5J@@1wjC0IO(B~|b
z!z1rvELRKS9k&d!O+>IQYBlg0cuaJfAv!VJ)T83z+S|T*7S0UVM;*(BRRcS~Yc6YJ
zscbcBv#M=&q2x_39+?0?f{=u?(uSKq3<A+37BM*?{~3w!l39e*Wo^>(s2X+NBz@k&
z2&<%|gx)Sa4C_I>YuU5Oy+i~XhS}0RcnGd08bp`bbYu2qb1B}vx^N7_Z~@T?Oo8)I
zmbtLVQf?)<R}#F}J*VL2E*U*b8bYD>E<6jp<NPf6OzjPKbOc$UL)Iz(EEU0&PmGaZ
zMdfB%lvwn^l0}z1;h&zDKPhu-HcCFxgS4zkYSGjLeSC@Sd@MX)J1zKR8`^Ri|8FPw
zV$L6i5Oxyz`T0ze0CS!|W(uH1&AHRuP>JQSI}Hi8>;V!?wVa!>6{o33IOWmZId2+f
zdrAG}y1E<odN$_gAH=TgjGi%0&MSglIawN`65aUGX!$2G+4GjKGW}pA06|HI2YCB7
z34tkF&F}V3)g(AH&;1#WWTKD>wgWsd@Qz~<oMq4JjNeTjB`-I=CVCOx<>=|PVP*8I
z8ft10DuaVh;7Sz{_UUjXM4!~Ww&A>X@%BA7<w0<i&V1)D*y1@GLpRQ)8<#x6!IvtH
z@Cr%53u$d)HtUEFJZ1~Ie(ukR5nMEiW!ZbHaT~LY=`E{SdE!q#{f9PM{*!a-6|1ke
znC*`FR8%-vq%~s!9;`7B!wZ2?YE@gDft?_!$l`ta;M`*1==LeYZK~DO0xGgp-c*z0
zFU$&t`xji&M)zN?&$F?c!7e(4V8|?7GiSwQpb<8;(h@2E#_=~arf<tvT#mU^?~u|!
zIJgz1+mYxc(QGU;LmuU<Scna0LDoh)Rp37h91rDZeGK8OzV;+m%=-1ehi5f$Qe*nc
zT>Pe+PxX}mTEIdqpm_ySDRBR@JnA;{>v=Q{bPhtfBy-k{C#XzD2G-|$#d1EgO|-8z
zzX}aZ4kK}FL5_8}rmtfg+#j+Sme*?8wkl3;S)8VIx6P-A#_SY+!-nr1RG1G{U~7_!
zH={yCekA|p${)4gtzeVwC2!4n$o8qOuCB2L(K6Ems*$E_b1CI%dyg#tgl*D{)giW9
zw+~N@RD6_txN+i|ur>OY1#?xu+ZcKB+%Q|f2_{Yg6#1vg3KTeaR9crcJnK}z;4DDF
z=eFT*|L*l1P5m5Q1|LD5r7^`FbB$rY0)*YEYZ{g}zs~L0aD#Xy_9m~B8$2n3j5J66
zdm-$LQd2or9=A6j)Y>4BHXD`EqIm&hZ>4EFtH3r9V3fn)QV|3{2u6rLY=nrP0W9?!
zc8{17vS@h5rgpiB;OOBJeSAQ9HiQ*yUy6Osi5@g<&o@Z2YT+=9s7MIsbvyon6K|gX
ziIpN^=7+N^`=5YE!LT76?uzrm9Dt9^kSNr1t8yjt87b~AmIx-nih97bT#keDC|KEZ
zO-=n}@T@%8^BB`3%q?~>!<5Ey$HSYjND75gtV52qEU}u-AZnnUn(}C~LWO8;Z$+`W
zODo=QFbSh<v9ODon;6>Lbu2y`Kb`k6GP$SB84N<ed(-1wMwGF2{KFZc#RHY(&2n$r
zK6?1Et$DB_PAzJS=;_R2B#skciB3%r?h4TgR6RX0qL;QUCoB~)dh^;h=h2A^WT`%g
z_4e0C1)f8wfqxFs9eLD=Zrx00Y^LA7H%hvkaI#~=wcw^t|3{>`BX^iDi4zN*>(D$N
zoN2k|%rKm9Z~8woW`ae(MPQE;z0`@`6!;!H3|{*&q|@yOYHH{?{CCtk9Q-1Ui_ZF7
zl;3>O#D-C4pf-Scw|J;30=DLpv{X`ur`}}ZHqU0$xlyPX9wiMAI=hW9?2UC&C5O2M
z`0!E6$I@$&<du3Hrl0?9hPqAFxlUf&?CkuJc|!SMHL1>sC(qHn#GxIo3Jy2h;SnX~
z=!R)DF0l*ROqp^@eF0uP5*$6-77`V5DV-b`?)bB^6@wn5%j}aF=#^oT-+Y6NHbs2#
zFSXipEKS>PxeTm{pM?$e?ai0^#O1WC-_5k=9(43D$@D{^A_R=NSy3f4)<wbF=_S_F
z&r}8AZD1&VQ@?Mv`K%pZ=ex7vk?!boAf28wlM{k}<0Y!d>va@i4@;DuPNiVQ=myaX
zaWI&QKvzOkFxy<<lvh)EPEa#<Y{~ee;3J1kpG+K_An>QMgK|u*r^%3xAg5GYmNDJB
zBNQ_6aC>J*ABk+@@yH8kU2)I45-`_F-qp0!YrCm-UBvX=*b$#&>cYW`gZa<Sa3k?~
zq?t=o^^A$^5ZcF7SB8B}-1!Bc=hzVb(VFtXI)N7@4!GfqXu0;iGo&mR-|j1I6y=N~
znpha3McAi$aa?y8955;fuT@~x_gvGVRT(9%p)rM*pf($qMlre_w*}V;FzN)BN7G4q
znf@P~Q1Ekwk_>kU|EX$ACMkVsguS{W&GQfC=cu|3TV1CEC-_366|bt1I-AwJGLmBh
zG4uci=1g4rt;W2cv3K3Hvv<cP*nh$Mn>FtHFKWhS<-!ysFH0rMhWt6twPilOt_nfH
zO>}L=F$Xj{`^eIpG=GV&jcq^1rg?V?Ba7YK8pYbv5B2@B6DvbIbYE_j+$0fJ*pkS=
zocpr{Q6+dGgRWJbs`yYE_m@j*J&7LSlW()!m_-f<3|=%DI6GeApg66X(1q`o;`T)L
z=Fx`SV@d-S9k^Ko<;<EW`J$u0<^5VsC<-}367tDao)X;=urDAOcI>rj$dLf70HipO
z^614u2f9gA{%`Zms#dC`fmpR^@a7h!Dci!4pVp9pq(^uWSS>`VF<%+;qF@>Kd^9Q?
zhci$3R=;LyS+4rfJaE7M(ON^ZoA$fY%ed`M40)5$;~(v(qS`dJ2+cVvrV&Uj+u|ti
zmfoykYmQr8RuOf^iR4ykeF}Dz|I5X5Izj2r9usDxKC>q&?M!v$b?mMEQe?ya@Hg_G
zZ{_U#Njp~WdQ^P?t?S?ZIA=*0Pt(ud+<UuUxwHG7@wI2#nWesS#N=iRM+UvP@a$a5
zu0zi@2NXYPpKxXMV+41WAVLiVVPJh#MsChGu;}Q+WzDW#XY$8;1@AvRb<{0V=g;M(
zrxN#@5kJP<azl7EUdtWYMLCjm?TP<p#b*x)pVZxvjkT?k(Xt;{C*-n&AC`{VEh@@l
ze?0yD<N7_H-JHzP&Gt#kmykeJ5nr%^NKLESxnozXG2Sg|^c=l55G_Kgz~Qifl84;g
zz0<SVn=BhWOI8Q1xd8wH(vyHR^j{W)A|6TaM@;hs9+r>n|MClx<mE!sqD48)khFxt
zVncGlj(w@AEr2aM1eW!y;Lpdt*26B;6?SP|jq}e(&iv^uKS3^NoJi{G`Y~9s@asL=
z{>gV|rI)X4{%jdfiR5?rBVzP2^}C3Y@bsfAr%Kd5$k@I!T+L5xP5WSS<%yVL?R2N#
zSSz9ProRIZ{zw2Z;6LdWtz_Z}TE%r?tg&!2;`LIb^!;meDYgA=i@IHpkJgtTo(PxY
zF<VEK-)G=<{mN)o+h@0qsN8>O@t91gTQ>9SSC6Z_HGfDa+nymarrgMOhU}LeUQZh@
z=mvDvylr`iwf)G8wRlkU)%E77_4F6*{oCksiI4PjOxk;QnIyPok0zO3P^<iX$H)A^
z=f+6~v<g}{c*wEEbgQ6W(~rY<gmR8#?7bi-S^H$W>x{QdoV@ENM{PzB-H|)oe7h~}
z7;Mio;#v`rXgV<WEEfW=egzvNx?|th!D#vEvx7KP7`{?SB1ZT`4b)Rc^$_1j7P{SY
zjj)j;bJaQf6mpEVOB6WDzzZ^Lpj5qb7S6Z2rhmjeR|!IW6c4JiKWe)~_e@z#$&=4j
zEV1rcAHK`ws!<{B-==Zg&ZDK{r}R7`!;J|ZCr%pARyD_xCyx5n#4Maw4%F4L-ulsQ
zRZ^ocL#|L*FWua$ggg1QW3Jk<|9<S;N(ti0QGCO<<omGwMYGJghP?Z~>+2QoOdNe=
zenUk0hK%x0bcucMCd$F&@mCqQ=-e?WAy>|fzx+?gb57;7zf#v5x$9;|%1>8&Uxaw|
zSoUG}-HEhZPzv48R!+Nq^H<X0+SGSabnyt1Zr@B}aB^<V4W)0wWQFeyDJF`Ujh_rn
zqjcZT1xb2)9U~+LoZ9>Q=$l3H#O}1)J+b|Vt=!`*v}*ryZDBtPed|TR0&GYAi~VSl
zK;rN+g~8U$z$BEdr!#$hy*=bNM+a9Qc(nTqX_Y$%N~{_3O2{JM;}UlOPFHqoG9C%u
zzL7e;XmS^ZcHwbAF#E-F$`7%rvi*jzkLSjXx<nc%u=TrhV%+e6g)j`rpv(>taH1WU
zRwJ=^rVAVtma|*k_IkUiN#*{5xQtGJ{?f05U*z8#{}D1_AHLVuuemM7drGXrrq;4=
zS%Ilo)f7MOwuSM2_M?1-CuMxx;~}~Iy!bW0a;d(rOwl@8m;aU)``{Z7nKY@K@prbP
z8wFUWroA%lFQ3hX>`b8;zEC=8SG~M-fxE9Mpul`-*&>3Tr+ag*P+v;rbr<>Cpwy4;
z-_>wm5BrPujqep3W{%C18HJ)z?yp{8@zk#dyRvG!d93|i?`S_wN@mRM>fFM9`4o@m
zR8QQ260a(hbo_0od6hAx?~L_=)ehW1asR?7+v9^PBHI*LutYsNEefRx3*!Gu%eAl~
zIKJIT9YC)PB>qTDP9!AZ+qbI|2I~a79vN|?=6x^^9ENB3e-T{M-jL8dtf-j5Vkt<M
z%uX%=Komhl!aemIz0{dV1xoMTzg!|+&*qd2<xJb&tS7dr|F>O5#v?Szdg?&No79;0
zOku<9$j1>s$J`$bQkNFKuL<u=)=xDUj}>Vb(?279(NTEc&s#@Nr=*M<VzQ)Uonw0+
z@e{3{pBX-}hew657+04<X?oGt<{xObNumRpZ?`1liAcV)b1d}h_3ZoRY9r%VwdVTZ
zSMiFy0Vk}k?_s93+!sU2JsFRgYW+yDi@8)|@~ly+va#A#if2gM8QmNwGAn~n5?!tW
zb|l!E8CB5DFy@)8lzDN3@<jg0e6;-2ujQXGuofby2uxQz0viLx>8BnsTrh>w%_#wN
zT$~D~Xm6U$c<@>+y7VhV?ag%trMuN_SfqcoTrN&Y=MXi&P@yhLS#dWFua#FGh{vJI
z&AI~8`kIs&KQDg!%XJh{zeRWSTVPJ%H8mB3H!@}g&Qi%aLPYM4BrE$(I`l8A@1uJC
z2GB!yP`Ab(A8)zi++E~2kZ$7kD#QMGYb%_bcx&PmUh{<A51Z?g)jPPnv9*jSbav^9
zgq^7I?Hw*ZBoDof9$fhH9%tv7(rYz*Qpw_~dP>Cej0yjSWZ*&hKT3YEJTG{9Rm|mj
z*T|`Uzp$9dsPtEbKa*^;d`>fn>XTw!o%|bhwVcwrJ5zP9j)sKb%`*Sqi^Nj}t=?a+
zEt*iKuQ(Wzk*m^EHjLF-8V)tYtS4CYK^<D~OFe2wm8Vov%FQ~P>7|h$=G=p~p%Bm{
zbqK6-2Ly(?5*!isH?o010p7tJl}JrhtG|=izBRpY6i~U?q3j-NnGoc{U`p=;Y91{L
z!Ue_(hXW}5;OmKP<<r=w{O1_~B`@sDzwsJB5^_;y+Y~g*-YF+!zM1t1I3JVIV=srz
zGD$zhxbk=!x9)R?;Kkm^#A_eTyowPnt9}%mdTs4wd0XGqi}TOE#qfV<KC?4|ZeDAW
z-rtz6dvGEEJ2f4guCeHSve3{mw7S&q%N2RMj5FA_;uj{xA!&U}sGKu-OT$wS9bWIb
zYSQoSo9lQ=wMS0N)Jj~Ay#MKyDZl%CF4u+AfBZ@DqwJ41U;Mc2b*){8jKe|lMp`Q$
z_u*k3WSc}h-Y1WJ5{zu`FZ(6-F2p;25^AJZmRTKeJ`#e`vDcxGtQYhNh&}-A^G<69
zB!q@S^-aW-ZFVlB2wXOU_#!(`k*C6N$noG^@5EyffCO+MHFQpN9fE%qDWbRu!u4NI
zBxN!@F4ufWUDd_E7TVv)Fj9`YdX4X>oI<F%@#fi3Yr>dk)?u}3hM~gCkUmGO-{T9C
zAI&QE#FvZKe#JRYTjg9BdYY|d9DHv`^K2sl|K<FYteFiX`sa3)Ir*KBv>xx=Yv|3~
zCk$Ia%KDv75dGMwZ;G5a8IsMgOUkHQS9p6~t0$7bc&d7TxM2rQy59}<&5@WUeA-T?
z+-sHUKFjhb+{OE<Tm6xXHR5Cn&$8Ob5t~8vg<%;P27B}dz5nJW|FJ7pqqh>eEFl}T
z4Hw*WYBuWO&;i<${-4@n1zZE@kZ+CZ@@Q+;Z+42Bqjm?vOWe`j!q3DiUFy@0!#~K9
zH*~yY&~Wa@?KZ3!kpdOA;BcD^dI({TKK<(WZ%N^<)vdMT*M)}{rLRv4f7Biwx|427
za(O6&c&U{d!+m#qG3!(;)v49ug3zFz+xW4>&{*e%b0=rt2@{P>;&l-6s|FuO1??HF
z<#xW`Yng$fv`0T)<X-t=*D_>!^1SS+T^$0gts)n;-|Kr=_Ti*qPyw>ApXxE7r^LG=
ziNW4kR5?fq(pC(m4k|w2mgu`z`DOCgot#e=-ZzQUn0=PHCw8^{L0HdWTyt>`w>I+e
zv*fyF&riFB@0wycwzGHX!VVTju`*1E=@Kkg*zIBkjaFRr&zUzaI!L(AV99TJ>me|^
z2DS9F$3zOJvLAdT*tMXV(C12-PRr49@M9Wtheb(r`*4a-fa@sf;hBg|AxyugM1Fw{
zZMOXY6wt4Yv{Hj^Gcr)M22skpD-?xL<toRYxV&+mbCjnjq}|!r8GNEWyI4Hc7uOUK
zYw-A7)v8A$<B68@oT<ZhL;36m@=Srz)-_R7dpv9UxOG#K-DADQnRUl>dD|}~PiYuQ
zRzoy;S9YkOGci(iQ|bk)v8gh-LcLV(`^+OT(tctOscDK@LMe;XBJIS)AGRfviTW;q
z4lUV%7{-#fP;_<p^%Ftc_3mBOzXN*$-TiVSi6zl`MdTYJknu+_Y)|m>=c0F?D8+th
zy2>P;@JchZlfB)#SD{#C(fZj*o(DeT#V;lcslLx;P=;=2kL`TZM|@bDTKTb><@$NO
z*J_26CO>CSm*6~6VZN~(+Z<wJ;`fPVZ%zz1dDc2-Zse`BS83{3hz(W1%MVr)-@Wh_
zxJ`hKIo8~`<xmBZfKx$T!Dp_0%<}={A)_V&!Ob3MLu`nnV-zr}j!}@#WpKk0yQ0&M
z=Gtb{w6aJlHER;TlSAIixP0<=7n9L3zrPdlZb*mTD}b`3^DKE#BN&INexmcmGFwiv
zALg7^wz=Q(o<H=Md0<cekv&K2EnY7V9r<As;9T{SWSQ;LvJh~7&|&uNtkd;`6FXK7
zDo?p-A}=~ll@#UFSr?qQ^)Cy4@d}^m|Fg^f>tsrymao^&oWL{dTh*HbjgxxpR;8r7
zKa>mY+%JY4D?dwrMNK-sf8gloFp1l^BgsLvx96cxz)+ai`9R(!#_3^|oa$Eh%^#{}
zmz8CoikIt&Htn@sMIG)xj;{D7M?WI&BpK)ERO8TnBxRkfnznpn1_2DTa&|Wgl>iJf
ziq0Lnp310C7<^=4F0(^<9#6kwiyFW?4r+JDqBvygwk$$zm|OP@HCQ_*^JFj_(>q$!
z)YO!!Mr|<$Xruo&>KFwp|8W=hvYW+FNGnCFTt)DRlFv@GCQdcwcdAuQemLSu=tMmI
z>{025rj(Zy{Ri8V=z{uUhdMG4R(HFDG1-BRh2rv?^!EQTlC{o$HI#bWwQNz2>2C5W
zDU9_@ajEhdX6%#v*r6MFOA*I|caho`{^X|w9LU$|vs&i9&^9#1Q_xo&_p%Qk9<k3j
zQTlQ7U#@uK2fyb-`l1r9dv;|nSx`+?_RA<Yk5wBU{&MAkoP1ahHt34*&3k8G2QDW2
zogAMK*(BGvN4QDgNQYpx&IzOG;>6Q&#8uI7Rp#0KQ=9kf`IxG8!)IMe^WHbgMMEXI
zZC$MWhnwWF#al|{vbJe=V<T*q?>Mo);$o;K+~FogWJZ)kn2Wj5YXCigrqk1|sW_3?
z)5+$t0*qhsn`c%;bUjPHwkl^?{zKl1M*rwA^-Eg|^{E1CnM9u?$VI?nnv%w2m9uxr
z(~jFPIK(FcUps_9W}BU*{Fh76;J)PCMQ-^X!{cpoGANU*ndU8s=lEj%OE~NCfY(We
zfj4U`vXMCxmI}Sr&P=OUpJ<k$)CZSW!X@92?>|43uhNw}*n2%^Z`oNr<nijTz=0pT
zx#i4tjehaqV$&&44Ib<>Qrd%fc@nGdJID@Bx%UD0!FT%2X9J3<vQIW``@=J-t$J#@
zNhi{3D_-e-*eTMi{z&kmE}x23c^SVHoXT+8*LG1bTzFZh`RH37QQ;Qj$Zeym%=(76
z!K&xg#yXf?ER#pK&l8icZw_r7cc!j|b@FcTXnei73Y~~oDSeqxiC9nfrrx7@X4AKC
za^S0)jmk6TOR$eJ#jcpJ>3o-ep};5svDk4+eGXUdg&NgRYm>n|HqvwT5~~NkJ5U>k
znX*rzA~SD7h2ygg+Hyok<BiZ*Jsi%GFW8{u%X8rr^@g!hrJ4Z!vXBQox(wYrVhVqX
z*{+kN(jOWp7<a}e95u(rwnzpaEV_s5-&W$GFHClMc<P9Y^VoWFM&M8Gq8BcP_8AWZ
zu1X%G(?8hCWH>(x;JzbzJti+~E=e-xjHsHr@<DtQqp`+j3Tvjizg9&?>eJ~6ts<ua
z52q|k<#}t_5?YuyVB@pKiJATrwplnl?1`GAm&W|>StECrIr!=}B+wLhfJ6h72nhmq
z!Wx1OWK<j*Dz4RGJ$S)7yfnh1vhy7A;f^>k9<U%zQ&5q=g4ua}H_3fN`@&2lftSpW
z-Fm&m%HXO9-(09l%RLkFUIt4zIPE&SOfTpvenKu!C*Vo;JE2ieen*w~lY_$n9*^qR
z9tVD-)iif`^GsDr$0T2?bMV`zK#C|&*mg-Qf;L+rMps6aj{<>bvKynM%gehjP@2(I
zZY(oO-*Fh4T2(eI`x}LtO!?%p2>K3Z#eb1!VKX~Xn;sVmfQzQy3B9CxCJu+Et_eWe
zv{GPelkN(3u}yC^>FLuKWAkXN=fI+t7!a@#PEh<?bNEeTzRQo!t@laXOqf>9EQ`nc
z53+~mfV_o+AD|XvK`RNE)l^`Ce4g@Gz=}wZ@Mw2!JlBMNWrq;Uo(Bp#f&j*$%7^Ee
zL-WtNI36tuhC_cxDF56*NwFpvCK?JP{*hTgkrzemtot%^T8#_0WkKU}n^bp<#IJH?
zXmCY`a`5u)3{DsB7xkjN^NZ-*@ksuDk5bgOki&I?&m3B(wR0zIVFx|PRHlMMB|@+-
z@mn4|Y&U2ku*^iw6R^Q#+%nL+sdvqlgV`hC;PbKH0PE{_%V+v2g^E$1vb@w*6w)3V
zlL>oAfI=T7#YtMRqs6l4ypCe3VxYl_7X8cB-!#QMuI!O_k$;N#>@ut1%C#VLpN~<4
z0~o&4yDK(*kJ4^qGPE}_wDP_jDEpyY%CRd{H_A+84!XD`1L(^^DN!+Pj}MR5ABWBX
zr8Ivi0#@HYmlrrLNW^F~rx>x)1zL^w;%H`0q%M6fw!aHX{)?eII~;;4tuUXU<}XPu
z^9*+W;fCpezL(Y=NB!uD4Uc>E(CR{#CGg<!IOxZTeXty32XeZHQb`mbco)$**08}X
zgF-1WmG*t6Ds=8+sD=U!Si5%lXTT>k&T&Wx)R~b>8s|pgJ7l1zj8IZCP1R)3ITUw^
z!%3-lS%EpNf~*dQaDD~Kvq>km0H<)5k@08%-}e~x&W5skVFb!<2csW*)`8f6Txr#(
z0<C`zJ&xvk%Wx{F;It^vXb$P&(B0VuV8M|aH4Gh?fPZLTvG@$#5x7>jUdnqZ`j2Q)
zYi^%Hj;pOqY37bEDUDL0idvTZ@|UY~?ZHJ;vUaN2X9(n!^fR^U?lvpUR`9F3l|te!
zmNt&=pcSxML4O5brKGUn2dXJmKt2YLv>72ppCTY`biBAxy@`t+@=vCtXL1m#xt|<V
z|9x`M*bqN;O>S-5a0CPR1j~9$Dztcqa`eRLz-QuraoirRib8$^!2;;_PVV729(9}L
zhzeA#rxZ3E{nfIO8(id<Q(lHRO+yZcG~c3e`VW{;LFbo&o(hQw0ZsjX){TuEf>oZH
z;(T7r|9_YSv=jMFBAk%L5%B3gInWx)hwcZbFABW~Q86cI_rTR-umK;Jz^!4csihg8
zTr<Iz-Uc#seBi=LP1GX@|C}%=K{)^bpDsO8>M9w+1JjKs%!?m9&D6nv9JYEZ7Vq4x
zMf>*{zYl^z781QqPyH|h1=B9#M?Cb7Vq735b_yx-SMgz8uo3=8IB5XH2BYBNQNam)
zZbm4C{O93|R)Ysw@}e_;a<gMfrh)D96q|>Yfy=gpk4@X)&AXkk&w>4y>qU2j_jT4l
zNghpei0Ff9+?ES_$n_@rv4%vl+W8u%P5Nn&!1oT;iaowHexrZ748#SZmf+gYekmzw
zPo07_viSx5KL_`V@<rLuiO!aPuZI8dm&@RDi0N_YrdH?F5$`Hr5y392dCLar_uobz
zxUq10J9pZkvC80$0mUys6&c(jSuxT`TA(5M%ca8UpTIgNQK|eU4%ilLVdN{zZ24*a
ze!RFPUuB{IqwO;X{35Mo1)SGA=O>0|>x!JR>B=+gz_5M@q2y@_8>wLLv2-Qj3Dmw?
zH@;SIya{X<GdTgMC9ho_Y*9HNK=BH(Re=QR2+b)%ge3?1NhuF@vtuP#0h~((alzTj
zbJ{3wsIJ9nX!m5OV3>WE9&8+0^=H<TIt;XxOs#b3`$D~WJfNE+SiVn!9s?+cJLDqe
z-`)HkNSYz~Og2GNRMpBs8S`nF38!;JG0L@Zff7~R1Ov`(8Zo-B?A9|u7$=5DmrTz`
zLcK%U;qjH4Wo;sxRt7xA|DGQXV?PBVj9u&47uoZKXasCbX^t|gvRS^sV)<A8<=X8Y
zfmUl8IH?0+L#&?Dc9ehCsvL@vo^mK>uEI$-)73c!$|WL@MIiC+fMo}k6YK)R7|zK3
z_rbgNfI5V;*XmdB+S1Yp+QwG!(Q7&`E-qf;GzbW}X8A<O-&Pz?=B%D^Zgynope0{b
z3?RJ6U$k%1;q-?XHW<Y3ABLCa-##3y*wETg{`SuWiVZ4lv*$T|-R=x+I{zqtr2us7
zcbf=kQNtmWlvpct-L~dL8otYIVJm{ubq?Xy3OR`MOH!tiZ#Ry-TPXN7+P1y}1WJ(M
z!H?T~(%>2b!NaA8935p%B8Ko19I*o#r-(q4v<x7b<RH{KB1}@0!M5OAp~<QUT}!Cc
zv6m!>UNYw{t-Y1{;#A-{BZC0x2B%-x{Xfh5pZS?=&Kuxn*zt%takLy78s9eOb#`$m
zf|(t_82NiR{pEtQj{7HRc)YqT%d#7I0&Yn3Bu+#eW=~P`+R#q<5bp;`0@I6wOA2Wq
z3YhZv_u25THDW4cYR=D%CCVpk#lX~U7#+ZYLV1nAqIZ;nQ+>1fZ5Ol=+JBZ;-2t!#
z!<uC-;|a12d$N2Y<4`vfs%q$!PY;+$07@!k4Ded@4fu~OVTut-jAwj95g^VGIQOck
z6@6Qr&eza%HyTOEgOq|g4h`oh1yamicv7!Hx>P{)c}yM-F=OlCCYp?R{QOd1aBLuk
z6D<Dc6ga|O)67~;^`(`t;(wu5+!5HMi1tkXv69Q}e1jdb^0&%7rQ&lj?_#p)viZNL
zdJ9w~X1jPBCs0DfAt|9{z0_^i+SP5S%u8zz(MxRw-9+a<Z<%l;N9=|dzz$GZFxV#i
zx1p0CJ=-nS4?3m#oOU4*NOP`b&ilUcH7aMB8>vUQ^;X1W`9|xj+Rm6UW{8|MBmo?Z
z#N=+9b<yiwIdA#nTx@BSgYsG*2OUO==*HyJ=Ny#J4%(U)X@J1OX7DtpQ_DX@^iHW#
zhJ8N6EV(^{A+`^>bC_#1yr%1jkP-XfAjNcWQKt=I&-`!#%~-z8t$RxAhAFf%Q>MW-
z#r|>?Wd65SBdx($)lT?UmoIi}v_pXLNWdh!<Xl1a+~9#mhPG&kX0iN}ERMxwIaoO@
zh&DAVEArQ>Uy`I!1yVaVgDpHyrJ&1(@ZzqnUN~`AU|gUVCr*Rq7~%<DR08XnS}>cR
z{#M0?J2m5;PZtts4Qi;)5LnhT$A@B-&^!tynU-bHac9^Y-2$xYKVD*G%ts+<A@0(&
zQWV&Msq2|l`%o0!SE=ZKTs&Jfb%8-%LFeql*oB)QeBmu)X}R0-B{+UgCJ4a|L7Igg
zJAiV%U^2#QCFI<zhEgsNytD<pYAA7N-d2JizKCEfs9n4pvIIgQlRhOSi5+Esxs0Cv
zzCgOL!FSP*|7Mw)3e$u#n4yi@hB_XnUO^B-TH@*ip1&xZp$)+S_t3EK;ON9QizpeV
z9gf!HIJ`p_?X7IaWJNXvl~Vt)0p)pfloVc_7(m~la&{10KgfOMVDULvBc!EuHE9;U
zdo;&N3G7H2Fs1~mw_x^?vEw-^GNVmbIQ!)AI~TD|jV0lc*r8nE|1#8~FX=ytBh>wZ
zX73wUIAhc1utEt3TXz8V_5f7~wk$#-xIP1w7yS+NG?`ZbC-DkUhHSvQYzs}<w`s>X
zi6V<b5fUN+Ho*xWMS%H)&MbpwG6@P*km{&Mw1J0f_uCArwNd&u^fn`b-Z$cOLTE!E
z2RerW3=Ib^7{opJoguNf{Z?RzgYeTxL1*FMyA1(Bvk`QQW-p85ytg;-S){BCExW5m
z0XiXe;ZO2{35u3Kt!*CeN7~3K1{RdzRO0)7u1Q6Cl)k%S#Cy@ip~@g_dU1$5yW1GR
z&iqFQr<TB&OzcH@EcD)|7e_AU0~1%M2RPc3Nz$GSA*QC?8p<`YPlM$37&jf1i>U+E
z0s-8g6NHeZP(98M12BNt50p81SPV8gM~{6HotXQ}<#m3hk(wpX&g5%YUNp1B|3|u7
z+`X?>PWk)HM{eDUCDE~@ix0Ctki8<F@@O&MSIYYLU3L8w{7}z-`$)qVf(S1YclTjx
zC&S^l{59Q5BcQ5e)#E@MG(>b-)&`Iqf9~wngz@MmYXCt{d}PIvlCO1GPZvp<^HqVS
z!`&UC8_~OHdB{M1b>3X6+lIEL&hmXWISjfIRV(`MOExR_`)uSIK5taMeRKUwZ)ro^
z1AFg;_Y!kBmD;qSD?=HFWowd8&5dX#+sovOQwve6&;A_$OqIu^r!|?}>}qXbNhbYK
zJVEX}y}WC~f6*zAdJI7KIS3mztR4ukfL4qeoFLeDs(^sWpaNAsSf8L8Reyl?4I1KV
ziwXpgJHRiQWr7@ngGt-NIBYvs%r<ThyYo8nlcutv&fxsh_3cXQ2@6Jkv9J1UDEIAD
zKZ)i(OvU}K5Y4E4XKFbzHo_x!-3`Cd?h(7$@@QgqK-nShCrBcCJ^c~`R~scwQi_SF
z=74xOo^GxJEEfs64DbrbN#L9xx5p+TP9lnsH0JC*t}!|Uf%Y=OA~-_3SqkM@Kfp;A
z8G}<aJ-(bp-f&x;pmb8K#r<}Kr(O_!^C*K~YkD{J($Ur8y#@{lEO+1IBWjfW!S}FX
zVu2%>qgK!Clvi{t7Rb91@2GLBBVxQ4c8-e1KC|4?Q*LxTZmO*7yZ&FUO&^$1!%U8S
zncDGB-?;lx8KgjoVcte2lohd86iQ01R-vAQGAf+jmpod=_{=O3T`qZoSb-vl+b5eu
zHSm5&`^$B2<JP3Z_U9>QaX(Mzmu*WqOxIme_qCO^>lXfVeDC3ET0L*q{i%xtuJVj0
zyQ)Hhe^ERQ|69Pje#X5N8-~+TZ?tcZ2-&0W$REm<dO)~HuAC4Pn{C^rekrXauhKRR
zDg=o4bg0K#mgj?Ts4RdiP*+r%Mx|1wGo&6vC{`e~g4*bcRr)y_C>7Jsz(qZ1`>0hq
zx~5w9@k;NzmIVT1D46B6urqjh2~$euy$w6#Q9RRZ48{}wn{@K&vcy|?z1Sj|abx^M
z49Esb$gyb;rRVo2-KGj|JU%n#^JGq>Dh>}|B5wakj$Z3<iDg#u=pzKQut^=L7@_%~
zoJdTQQjAzNbeuV8fUSD7e5}NZ15GqIPZs`J)eA_^%>8f^8w1_}%4iu&GYeMZLz3D@
zf(i;TDRS+SMR9Wa(||eMyW^Ed>TWbFf4e66+t9(f<+N=1BaOUlqq{{<GMGcII(2C$
z<;a3W$K!F^8tx2Pe`1sBc-G{0b5}`i#wFmHAJh)l%X<HAO%@$`ggcRgb7z@2MfB?1
z3uk^xn5B)f)qVIl$Q82gJZS8~L6Zd)3MD;^gMLa7eehaQG%Di;C}BWWdMoSG^Qsww
z>c2LYwC?N7Zdm@5<xvQvud+%3bAhsp^8o&}<}H<RrL^UfCw1a3$UViF@|-^?Z<@z8
z&18#qdH$+g@eq(xy!_*#>aj4hL6Yb-{l`scl5EW>0x6HM-{P(G;0Qsi_T}KkYq#sB
zLVgatKXT)_mAlWDjK<)?aNRPYvDinKxUYN{%^VfGcJ^lTk0C)%gNprS29n5mlg_lv
z!?**93<9Z<F*IjMhY}sMts%R~!yL(hX{Tx=#LVNFJ!A~WOehWmG<7D_TL9Tgo;9M(
zjPf`TTE@#43-uzPp%jcA%Z-N-gbSF*qRvuzn%_?DvDwkk5~Y3m=w`;Ah-+!g^)lz_
zj6HT4vNcb`W6Qz~*JrN%<?{FZ9WQ6#(ZRkh(AQv~mzt4YT5~E1^@*8uamR&8In%oq
z*#GpLH-!C)`;ttJX9`zDMHPRwuim!3&bA)bunMMMlGx$2Aq&8%^c|Zc3(kUDQI3Ja
zttry<MYzkMG7b(6N>H`x>%+#htO+>HIu%Ubq)>YEI)&9Q@o_+Rj46_n<{+ru`tO>K
z?=xNy@pI}l>zja^mPU+da5D$aTpJJ3dO@`MuC1hbd8PHboW~_7%^pwK-;Y?g+g|>e
z`EGIN!GT5on?ugKZ~e(Mr&6)w0m7L<%=Cywv!```xni|lf2}@7j_o^!)wz_QuIHO|
zXg%G8XEkHsOzOm$lhm1>-ADd=dZK0I;>pzwHB;J3E5x+!D?(xZB#s7|12gpD*5<(o
z9d?qDoE9IALmh#q18(KCgMXDF@M{Ll?#!Nc8Ax&fBh%h4jVL35-|Bq;O+WRB+LBIw
zQG%~kKGn%wF~Ri3O^PP|I<0xp&-~7iqF=3SO{;RU)neo6hIKpnYk~I6Q*7O1?=$~g
z+w3TwRF%3p7MD$~S3X@gB^9h3`*gCe^FyWb5%0Vz;aCghrv2xyt?^mkl-<@z>PGtP
zS3;JL{gRLiZdg5)?UbJUA)s)_Ss@?M>*{N+m*UT<6Jvp{>m*zRP7oZ3(=-BwK!hV6
z2h5yQ_rLlQk_&4F_#33PQ7mY(IDg4$n!_s(8gBF`Y@tH=#4SZL$vkMfapXvEiC+2@
ziu|K<8xnaQCQd&`xDQRaSZ5!#lifq8Kk($vp3MXvRb1_J%;uh9pC5)%XH>hz2)@a!
zH9LK}{cMnQwtLjl-Ow^6y$~VMVysN(r~ApH9Wu)obwB_e%#nrlrWfh~Tq=2=WiG{y
znf|ENj6;=e6N}#~&3MM+DFyOqDaRaBP76#)X&12tw8@VQOss1(b6PYRHS`jPraEFP
zn$CQ#IfawQuFULJQd*fI0%y3OozPtk)i#V4gxlefmV9NN5-cyFc}qG-nE{)DY33K|
zmp5&oGP+C$zoRWIkJ+%U?cuE^P%MU0PrG>NeI+>EL3d=bcF?0vA32jom~^d`Y{KGn
zI#b@QxI-9C2eMDmEFS_Cr=I>y)5^p`5P}SUt{d+upU86S&Q&=UL0_VUAB4m)$bz;e
zm3ajKb7>OTO_b_2-EL6t0;n7kn)ZR24r{O~e->mFY>Za_GGTXgCHIdL%Xk**W`LSf
z+R<W4GhtH2=<s({i=QkLx&Bt$0i>4dd#@*Bd&HfC>Zq?G7B^Q?sSYn5J!{O-otd`^
zroEF%Zg<Z~-ZHL@b1CZCdO`Z5>FwZ}nH=4@z^_MZiz&a$cV8EJUn}!6M4d0;>vD19
zg7#<d`aIeOFT`ClzHQeuIBM`^J<CBV29N-fQo39FlTXjbHU03zV$Zs;8Hc!Q!Zb+O
zh3Pwwu5J5qcu%Um#r$CHF$2*(w^ctkeDqF=u<-PhKJW@Lc7z!~9l>4kdfGskw0gO4
zZS#=arnTJH9oN=3o#LK(RQAnSx)~Ko$xgO8#=>_uln2HRS+I>q@<(au#M^N<f3fAR
zwH*nr{Iag@uZ1^wc)L;FEM;s73TgE?n_57i@I0B|jSx-)2E0Jl3O!@!fWW<Qeii}K
zE^uItrvUsJFk!rMxTSJ%!r%gD41_^v25=HPsEIm@IXAEUGy>Rf(0eN)9G-k!(*&+r
z>BM=-Zw|uK;n8xHto#SP)#$zlqh1W(cA@-&9Cy*aM!w5zyR#C?7e*w%3|LU_jllsj
z$qiGq7DG(b;k=gQIbp9UhcPo-;$i=XLnX5oc3MP-i0#2Y^iNXFUYDy!VZO<&D*KnM
z64o#3AJsX9UvpSxSK;~xL2o8vhl~_a1cC>G6hs{yekel%MKdbs_LLVs3~h%FhqGfq
z6z>DNmXVh5t(@0L0Mstp2ivWHopLjBrhhRm5{rkB7rVoQ!J%%m%Ljp+j=-S%BAH^>
z-?G&X-2@4ofaY|E&Gh1>t)ROur~@OIHJ4QfTAsuEg6oB9Xbdr(Kn4Goi@*Jnz-z*m
zmNtiBowLIH2QSUYjhtFF44FG+c{$;8ahsk><AvOY_1=j$<!9d|8(z1ZiNI40PX(^d
zN(i4(Ow%e(IAa|+{WMZ`b)gey#q$H__-?=O-2u|F<a<IOoQCp8;Mjrxwj(E+iBDvY
zT71BjtseW??2G-&6?OjDK$%OkncFWITmiVyJoii~N$9w3p<Zkj2jp-pAR1zT@sse#
zHhC5x9!|#DoT?8|0)AIjfMOzI$1_O<AV<=&?zj8I;p8M(-aIlx|FRp1fl#JuSQxtk
zY^*4y{F$281w$Z1?iL0>dWLXlkt+h=)T$sJCkTba?VE;m4<CKR51fPzgC%8Q44sc)
z@SF7ilGikKrTxVe+jn)_TAmJnuNZvRnQ(7iP?C&(*60{1=eIvXdr&vMI$Qa^NfrJC
z)$8RZ&kw?amcDuT-c+F|@sf?hkCEh6VOB?)>5XpL^9Ju0<C3aYcmL%I@_yUs%i6_l
z9%gv%esM|d<4VU7``Mkq(u!?-3N=R|b5=NYSrKGxBY7as5glsq7f>I&BJTP^K$YMC
zEUk)gFed1=7`($zQk$BX<~(AcC2K{4s?Xe$nRZJ#;FuLu&vUy}VH%D@yi=gS*QOJJ
z(Aq*YJqRZsV22fN9VC1@!?<E}!A>SHnW~&ll1n?fzdx3>tM0z5uBLENlDlBS7cbW%
zi$kX*#N$tVY(rUD87SYUc7+)V{!wz&IL`a+IX88`M1-o&pSH(D`rT;4^V6q4bxzcD
zYSdXxllYncLH{V91>-YrgU~picTn92q<E-zQSRgN@pN-!TJD^G2J&DX%3H(_q2pqS
zxt1_gP`0R9d{J@t?F<R=HY}3v!!s=A%1U)wcw&DC?wkWjK(Uj9EKMYu%NBIUgaA#c
zXq;7=+%B?l^3Ij_TALK^T7?dFcG?Nerbf6^Wj7vpxU-Z$EPtPD!=~hQ@}P4f6*7qM
zxb4pxKNa!gyEdP-Dj|@}X<@y|YlHa~eY$jsRqGJbKVgdzA{LWXX7Vh%WcZAU6=T7~
zX0ROk8zw0VB(%H`NU-sN$87Tl4v@4c^x<9@Ks!d17_f46gI@y5W=ArNUIN`DPabrP
zDOnA|20P-yV0;ZUHc)H-3(&!rR?aivfKpkdyf%mx2IihK#fX;C0iHq5_zF}5IS7ut
zk`oGnWOEEG`Pz~429A}<1N}DKI1?qUTMIwSqvx1dDJkwIt5zs+Iy2JD3Yh3)lptDk
zyIoTm<-B`g-V4$#sE+wpK!fPQ5$o_oL3w6UIM)Z$H28VoT$cf<t6sPG%k}b%fs3WG
zSUnRuVtRxipUwp3K#5z>!vzP+(0*(t8G9FSypIXYXjv;S{BJ6wMV}TZUvcJ^om6F<
zZ2&OGAWZ=~CJ)Ete(g+YFk5@%&LjWcxv%xE!i|68a-gqgpbewDa`sDyU}cn|(ab+;
zI#R!XGy@449$CD@a_EYI>fL9m`((u`AB^3InnXAxiB-mNzu^$jup(<-Kt_Gxm#<LB
zMj2@zS}XElRAv=ncf<l$Ff3kA`X}gsp)Cj?#FZgqRQE-RutVJ$c+cSa<tQ*Fz*$1s
z>|t;az9Cr60L;{!$X_n4@o(B!#HKp6Ie#~AXN8SBgy&!ZWe{C^ML<sxe!E0qMO@7r
zXo{#{j)=r4Wjhu)zDVeJb!VnR=j<p{-RyE-GsY<xx<ws&d#=-f<JIKZHm3qsNYt$y
z7#!>nq%t7@#pKTq-r(=I7|T=+!(qq)^$rlFXG$VE;Ia1Ri2|swPeQyXWZV#91p&3(
z3Ed)230O(31ym>l3JMtmP8YqCixcbBys>^}FCy)2?BwDf9XT_I;VIw8leVq*))ozY
z&1W_wMLy^FU2h{KD}EcDJ~Xhd5l3VFdrYNi{L2q2VMQM6cYiC3QvZq#GfjI9JTLbS
zzm~u-SMAL#6C-Y<olrq<+H~7Ay;v3Mdp91I<2mx$b=EMu2T4ie446Xov*Q2>kq*-X
z5Y_`o*tmcNor5PzNjzn7&0#Jxo1M)yM;-juw)D9_)NgoNj@Nl}jE_I<^1J<s`4Lxs
z1<Kio;9|F?@h;o&dQZiOH5){;@5=gqRQ$xgt97l~lIJOMh}oL##~kz$;bF!K=gncx
z1uJG;%Cv&D6~(YPNP)I|w!4JCD?J?u&v$@Hc?IA}zbT+c1SLfwBMsX72GbH|q<|Ra
z5P0fha|cRuB9`8m0^okG4bzNw?ZOETR|{@$+$xKafM1A<erqNpcj?-T>I_-O5(}x3
zu#^5lu6EX$eZ`i;-cNTuH;{b5cXZC*-~=*DGR-;%-PmE`Sk~fkb2LUc{8{6P=4|$y
zpovGoBtQShOs}c3wBjdU``!qITtxPII3?zMF1&i-2v&L#6X6$=m2@yP&1@60)o@is
zu7hZUK2-PSz2QCK14bXr4)^`4`!bKQLPxJ4X<BK^po34~72u>ui$UN4Mr`V#Q>q8;
z9%*eWC-yq9yd5&^7c7_%Nif!bH^_^RF+s6je9LL-NZBXxisF#EfYq`}DT_dL$Ha`%
zm~eC!;4L3IWZ;~0Ls260|2+u0UpS^UiEj=&Tf4OaI~LqiQLjhnWghdCn>YLk(&D?{
z&8)^w$SC+hHZgCM{3I=R-pnQBUQo<*?Ect$KC>f1jjJD#K}S1p&KXAOqH~ql4sS-*
z(--8?5hjT@U*3(nkbS3oE-slc^5CX>i^x%V(-$$Sl#?&b4t)4tKPb5%=gMewn{Vjc
z!-GRw<_>ZC_+6{=y^-k8z}#>~(V(iCPQwq`cUF0dyE+w=&)r@P8o_M7C~q;ggtpkW
zOADSJ_z+DiJIs7;q0MQ7=gpc<5gLcYaz-=R<_2#6+AK+XVHG*06zIg@wECGHPO0pF
zJs4<$0?8(FCTiHMKU!K?Y*CMgo*|w6@Y~G^eNsu!^a^ua?rf}S*W6I><Z8lL><#JQ
zxQ48gFVTn22QkwE{0?WwC6D|`W`=BJ==lUO*(w16${K?&*QHGPCyI5vuO~0d8SNC3
zCj{(aTi+<@dos2d79#@ceZ%$b*Vsi->5sp<5q>@>U$^+`&3u8qbP4$Es<%Fdz*Iw5
z2(&c0#)Z=*J_kw=4p$)C=7<oWt;Uwh_ECu`oc;qe(nZW6({z?Sx7gL>W{d^XTz1r_
zyEC8Y#mRl~GgP6%p<J_g#66leP7x$n|9n^EBq|(Ftpl5?Cou!ki<A^m8P0~6cB`NF
zw-Hhrqk9X#ybvpVny$p&&%5>?bIv*UUG~w`0{<r2+vn{gt}Uxx+}W&>CI;YtdTf8m
zQ&&FgzJ(&jgz8jKwg0=-&B{NukFcZUpX?WkdfjYm{9Z+KgD3YZ;lYY%&3}KSkB%Rv
z*k;rRW$WxHAM9S`3%V7YazQsHD?TJ4I?8NlI9#k%<n+{EE}Lp11vqvPtH6E&e=s@n
zU<)iD-Z_m%Mm=dc7*c16yYVamW(uhnVE|UfvT+ViH_(78GfEF0wI(Y7S`K|U{92xB
z3QgTpV%GKlh`J7Ns`vljkgTq~vk$U4M48v#os3lW$|w;jGRxk3MaVHqA}bjg2if6J
zIf|%MhfYX`?Edes`+J`M^W5h?w|k$<Ip6PRyx*_&e)%r9<X!X|P3Sjmi)Mj=fD1^;
zC?O#sP84gNy?g4z?z`#<DJF5YdfE5OB_e~;Oapnk*rm9UY1f}mJY2YH9S#MdQ0#)4
z*Xk~HbQ`c6#fcJqJatsa`@go|D`7;26wasF(O;)$6CMi<NM<eSoz@M}N{OzR<j#_4
z&}^u|WW1L1fW6SOebg*I2Yr-D=x}@Nlg02*m3IHXhlpV>;N_QH=A_c14zY|l%@<ea
z^~^%#j%XnxBa{)!q9*U)KB>|bGa4EKZznBqeT4x}cVr@(Wd^CV!GXaWeZ!7wU`s94
zO;g#^$i8G+_stdtdXn#T3*%s<O-xC;b-V;2e4U+S_d~T?PtBt|eyz)W*0mOh%475~
zx21NxG<nDhFLhb)f-z=UHlm3|z)AR7fKIzY?!P@M9*ft@8{K)M_q&q;+X-^Jl&GdL
zZn(6sw!B_LM*3(VlS%NFCR>!<(h50b%{QS`+UMH0(-k{V1o1_nThvm0?rKsVzYFX&
zv~**9$M6(OLD@L%aqJS>{Zsh_+B1TV0|7vx@&O55kwILXRsvXQL%~LZO9p)em7>P_
z32(f3(ocLGgP<vo5dH=&%W0ns;dtg&vAB4wBO42RAhH;h7Bko;4_xTETaUH;EnSR2
z29$(FkjH_qX`reRJgBL1%?<G=2FLz>D6QmhDGH|o0pE5F$sEmQFr;9t6qf1Xq{5(d
z;E3=<P8nqXWTJ?04Jm?KeShp-0>m&lY?{D9d;sM}ct~Wxm-GYT`V3IFh`bLvRTI%n
z5-viB!rI-4UJJVk1gs}TI#d<c%*hh~G;msvYXXM`{0I`AfE;LM$Ha^_K~MSJoJvq`
z6@CQplOxh&TmGB-NeDrW0Ff}r5?oDF5ztN>ng0MN3B0xJB6ObX)f415dXESloYYtY
zA=~>xV9$R*-lUo@u_0;scSC!`0VZs~$`?06v;8bBfp)!!%YfeX2PokfLllr72iQYe
z>~!Q&ncZ&E-mTnkP&R@ewi)m@Xl=oE=s%HbF<B;Nv_6JQR!=Q24|1=-Qi+=07#=J+
zQ<OP+K7B!?<({r{D@|@?R;wu^pBC|vKCsF$Wg(e^_ehQ%x_}skfCug@D3LVbw=()e
zd1hy10o;d>fa+odXzu|lN0==A!Jk+KO@`ck$J?P-wk%uW@rwirXmI6LiD2r%>ki72
zh0Hh4(zV~jPCl$=S-W#tVNZegNVBTw8z3`dTPS#sNUN)IsHc{yUOlJ7NX49!00|oQ
zTuy66&QJzuqS6qx9ULprG9rESLc|?Gye<P&&Vjs92;0^O_PrYzfP&x+Ofavyr4>i2
z(*WVWDU&Q;f$4}lmjodVnxZAPyiuPqJePxHZVqS@1O1DBibjFTCa%BvkL%i?Dupd$
z$PIXgKpJt*r$!{4YF7shEisV(Fffzw=t?f)?BftTGY+tImt-+0s}MEdAQbJ*|KYRn
z%{i&Sr3NcQUg__lE_+BYuu5|Hgh}`j7KB7502&ekG9>zRUY+xGN6xB7fjxDw<df?0
z1j#_!etVE+fRYdRaquc3p94{38rdg+{KEb7f3N(GT2Y*61T)OcgtrZ10$5m?A|~mu
z)L>b0s3*>YE(y3>)Nd>!LtNphx!oQS+!Hs7HhhVGn}NpzuIi{pqrP@}E~a={mZ3jF
z>5t#q@t?N#W*8@#)v0kb`T&i=mON&>I+1G)RhhE(IBfH=<0WU9&m++vnS%@C`2q2t
zRP`&D;EH>B`nzCk?H|`I7~>2g5Dh@OejpzcJ~eD=;~pJYNvfIC8Z->~nZSBk+uuW0
zPkI_~raIOR#}J}ZHB*P7(6UoL;P(m6`(L{zT(?ttxfs_x`QOheudvmL`=6gHihg`G
zQ-HzfX^R1Up|A(1MosNzw6qEf6Q}0wG9!B2Vsb`BQP!y|pJt*O_!FGdT9kUk&uH1M
zokkxaT}yjwX6|CPM$f+Wyyc~NR+&EyvN#Q(zyv*=+TeFB51_~9HC1`bfgZSFWk>I<
zZdeIgpOcKttDm4vC_LC|S>TE)B|+n%9E#mi-(E2=Z|99-03qW=l;}KsUF10M2g(|;
zT!3EF<x|um(kT9~xfGq3`DPidEL`iUNpjDwykUGUPl2bVb6oMk;<nr2%<TCJO_F&^
z`j#iXxHNAcTa8V@pALVY4W6RdXpS4&F<t2bZFgh9Y9@w74c6u^2~*XtVfsW?>MSFF
zt{Sa4e=ot#{l4aVfkSTm0AhrTcU3xy5wcO>>5j(n$3@AlL9;?f?8_7X@8uZTvFrG0
zqKZIvrXF)g|6}}u%U}v;inV%rF_%onGvSR;w9<5LUv?HOZz#PoG#OJ8ci~`10Kq5_
z6&0_2<tlLXp6=STjZNCrZfl^~$kc^VV#(W)&y|JVhH5<ZBE~1#OT{+Z)Ge9)qE6S_
zpglD=PZ|l2;IDp4We|O8HRe$RyCj{h4Q1wKCZpHr7oyiSbc`>|O7ZRTRXqyfJr<{W
zWj;hL{A|wW1w@!hEt3F8bR_U_h%f*V%D}tnA4Ha#bQxo%ExLo=)MTyE(Iv-a@E{FD
zCSqG+;n`p@YIR^!h_zddi+ymV5(HQp_bQP8A0)AGJ~3~;@!ogg5{p6UAARpE7t9e!
zq3jt}d6>uMi`Rrh$!$YU1&_Vn+K(P}t7t3q(Zi&(<;<L|mpJvWAK_8&Mn5(?DR*JP
zi%yGI{pp@4aP;iV>tj<&p`85`=l3pM+a)CLmxJGn1=fb<$qYl#`64Rp>f>o3Z>AFD
z_*yiAU1hfB<os0%aSvC>u@FTBItyDy-(MgtwNC<}6o5~V)PqQz5IvZPf4LYWDnQu;
ze<~EsfCjfHt8-8(T=u!P${zKA)ys?vpQ)Y*l(^40%<M#ICI-dKEEGL^!z8aKeIvXi
zS|#rf;pM8S>SD?*y87c!d2!VCJ$XJ{DqraX`z?$^n9ul8eB>=D=G|_5nesmD@!ZEn
zP6dY=>9DAtcQGww@4)carQzlNAgTk+GHe!muGKA?1yI96f(CQxE@Vy+SFfG}p%TV-
zjh=8*cmCqB_&fX`u7H}O-0t6Jw!3GyD$Tg|qeD3H6s}O-zsqV5>xGL>)>0GERuwU3
zY7BNW6vxf(4(q$lGCUVzRoG!n6>y`+#cT?lq}L||Uk;{Lw!h43@8A51i9KV2BYT5`
znswf&TL+&z-aoE-Y{sarK00^C)#0qhzlML9=qV;~L43zQIF)Ei{|OS>IM$$3;g>Y7
zEn<>!ncZ$$$xR=3SLAcQeRzQ$aT*K|04snSC4IDX{C{kTn+Y+WW{Iz586fX-I<{)`
zI>Pu{o(<9q0j}nipA+X*2^qTRP`IUGPXLG2*kQn*3{go8qt*#$MSus&#bwANgcu{o
zyM3i5z=9S#2VlkSR;+wiB09fah!3I0=+e^a+~Z<C{GK|X+|odJ93SVtU=r2Uc!81n
zF#QEb6)7&+fc_gykM7LWht8<=$Xz5DBoAWdei!tZBw<@53h%(}6L&7(LgW^I{=wUz
zLM5gqzaS*fq>vs^<bn+Ba_&sBCxvGXy%wKz=77HPhYAqC|C(HoDU?MLqIL;dXoeyK
zj7Go1JX|(SD?2QYwb?WZ^yc@u%(wU>4Ap4o%Rz-<zH||n6ASlYxfsUoEEr<ZchJ;b
zWFcKx2*B`7BLZ}2{Q>BIT`q=bd6B*#q+xZ@cozX&!oRx_r;S`KBh-zcVqC}vf(Imn
z@`MN}Pa_9mJvxup$Z;}w4*W}9Sak757Rv0i=~8YWSon=xCr0MvXapjHTng)i=VAyL
z=88~8jO@FR?E+9Lk#cnrF2ef@shI=_W)c+bKagy)Q-FfR#1A84H=xSJbgnKU_ft7B
z3cdnGQ4&~^A_vP=+g*NC2GvJ@syxf0mI@+>6Jo}2L_ptSpmoi^T;>}{fZJi?!6$+O
z>wvk@Zu@&kc8+@tXXNk-`DnS9v^)%D5;*kZGdzkZ4HqH_fgo&fqHq`m0pdP%vSD`b
zmR6uN5sxWGB>foF@5iYqTju^TzzNjTWaPd{1~W3x2mTlq$tso<?blFVg!(@ub8NL-
zSBKRDZs79WQ<$Oa;af5D+l=a}0FEN9lf)@3nbKs}WA^VMRz%%K^$t2kJ-uL$2tk=%
zxPBq)N6h$HFnrdJoc`D=@IenAB>XL0y8G`T2{SYJ=(fAZc)k9pMom;{`EMgRAqdmn
z1X7ySN+Ubz7O>rKE<)uBpC<^=%+?PM`%WF|(a@s8MUGtVIt;qF(*S%G_g-PU)xXLD
zzY?;0hRffjzX>|k%Aj}AAQ<%UC_ls<b3Y}V5`I|jr6H)zL8hus`F-EHqtyaA)hS%B
zfCPo=9Ds7>Q22YaUo2S{(D(=Bukec&H&W|s41@;N4L=pNLeX~>G1p>3(IDJAvn$E+
zxO_E~6O!k&JU1Xwiv~(84~WAmG{94c!ZiVkJ(*GO@NW`xC4g1_=*yGU301H(-}M2E
zjia0h0xgIx15%?ZV@)mgE>Lu@=7Wf_1DF1&fl&ndgTz!^xef!Odram;Ybro7j6r2;
zVWBsY#;O|xGr{$A!xZ^N%qGxI<<3X#J~2=F@Ark7KphT+bBJZR{IlthY`+VGyapf%
zgt~ck@+=akhWHLp&?E3kiJ6^DGZZZ$T1h~C2x1P~%)#X2b*Ad_xN?`SK;uVeGw*+e
zPfCrYio^7P2-Z#j0B;Zt3}Tso|MzNxROYr4{tz7^{eck`3p>ODg1fk(Q;2gLN%8@C
zA|$(DbB*!-WkA`j84|}1U<x$_2pD*jIbf<BqEXtV2VrFTQLS$#;B8QVSd%eK7&N5R
z;9?13#o1Q?3J!lMQ><!*AJ0_f6hTBHc(|YaUugVhBn!eFJqT68Bw+6FwHY~fxy-5f
zGr=&vaDH9%-q$ybgQo*@Rn^4DWq9Q8FGvQ<Pp5y^3KWevBNU~(>H-V2buI(>U6{H+
z-d2zrPNW!?{MUy3!cuL4APDeGq>|@?_Il*Ld>IfarauZ5fGWsEE=f0&#J`8~I_H6v
zwo_>3PcabEFM*>%LjKXm+?s39Zrl06sR6>=bJm`%)bqb|>8V37l}BP*HiQO`$ALQ}
zYU0*^n`GZ^!V`be^Wrf)BvkwnElL+kR*z2nqD4fa$52Gxd{I-p@;WZi!(;%1n1C!$
zL21P-qUS+*D|4@94UJo<?mg`<H9q8b8&zt}(bVYNHDaG;W@TsBl@&E#)|d7ms97}K
z&}@eK?fKrnhn71B+!4!eJotT?2hTF+B>7^gfGS4zVwsMP^+YZdEl|#-`vtR~52O}N
zrUl?g?<Ul+@bsB+6e5}M?O_$<eu@(Pf_wwxDgC=C3NjIeiI2_}X*Jqwd4Ng-Bnl|^
zIuh`nxD=@BDdDs%@<##|R{^Dm=Om9+*GYe`e6a!{I*=ciRsO$nev`XcNa#jXTAeC4
ze(*8oS9!%i@z6W_?)6iJhDi?7_y-JQ4U9aq4y2ohi@Xg!rhir%ygS)b`^BlSKz{Gm
zW#iX>4;8l7sZ8lsZ*>k_*9plUjS~VXodTE@3_v(%@dmN@up702k;xvh>9kfNzg)@H
z=lkPw9sg(+AZgSAXfRwr<m_VL2|2i24{ia3Qy4Aif<U9g{zz}uICv6d^qH_9WfB-7
z1aH}aamoYHrG(~ve}n>f%KowLqbWT&;e{Axx2)-xj(s*CA-E=KJH_ie(uKZ`POK23
zZ}>rZR-;}?xwL0QXJ}#8zA!bE`^FGcz^OfL-wI!j<E=bJ`hO289j+!+#H=xV)Vs{-
zKl}KX?wJqL?glr}sw=;}?PH`9C)jmwAZ%L|Le^-og<N1|uwyyEHxV`iYCR;Mu`jFq
zN_ZwCh;GHf4F;mJ$&-9~foSOBFIj>#6cny-TR}fYjuyLYe;<F$53Cqzn$(+@cDqLi
zFCyi~obMC`)8>%Q$~5qyT$xh3c{VD>^CGNNnW~$I*h5eKn}1@4p?Tg#d&>CL8tVLQ
zmU`|Rfj;lvX@?lu1f4#sD*(LBYv&ezbbZ15ORK`JW+*lLTyy}A*1PW&mVR${d7TJV
zC|tC^H1rI#VDn1@Jq)Of@zy$6dyA4fI31*@m?K2vRSghSWpDy=N?fALrE|~u@u&DB
zJo-tNR`A(89po(&9ZbmwEF-Jk@?f#fmrtSuG{o8|$4XJ^M}<;i)Q{KvD(x%iBNL?k
z`7E?uW6CHVDpFmaPQ1Qwb$>fN(Be*h0pIxtnauXMNNxpuxYI3{KNPR)!+-no9kIzl
zhpF{S<d1DEI;^u|_hUF8>j*WF!*jpQ0LlK}Z`{8EXBBo+Vj+mP?3ZelW{`|x?A-@+
zbiNRUnJI$lEM9RaC4#Ph2-eXcDgqpoAd*LP{(}oUz|?`rMI;^svklNcVo*SAiy4jF
ze;39Ip^Pb}O#EMKH)EDxC3&gb#J!ZN7@WBzxjd!Ws`Z^@bIB_G4o5|_twO#`In%Sc
z(WLN8=X*5-UcM|@E$7IOadF4&_k*QGn$K7KZItco$c3Rfv;B6p+5}rcx5-@ord|4$
zphcOSgDnn6Y=0!rS^2ezT%D`114NIY)eRpe=I9QKQl6s(I*jhV{9vIIv<?uMqjxS2
z5U1}DYq+hRitV6*Tvvm^@4y$L4kBb%XMw(giVZ}KVj$H*>u=_;Qbf;$e#^imbK$RH
zqj_g&or-Kqd@;URo;vw#=dZRa^1NWU?vc4NiD`~Lw!sjOeKe;_aN%mBTB6NTBs;-c
zk~;!%uDo05aJ%rGrJ64SUpyJ6c6d_8rp8p%LK5I|M{J2;z!8$jxU(q_lY<$rAqoj)
ziI^Y+ojE%ASVwS8cLB7#^pKo*e*r$9mZv&$9xNjNDANNVPtmhx5wG6Vk~qS)N8=62
zn%I_w+j!$(@~2OQ#8f5$FFgxoL4#gwoTu9OEWuDJHcQZ;F{vW_+uH>#4OjECdOK~Q
z8YU*~KQ7E)>WDfm_xK6(CB-|@Z^N#6cqhTZfF}iKK0f!~=G^qoDxVYyRHJ@~bULEi
z5V}SV<&=L4ABFjs-$B9KSqL8~s#%~u(7CxIMmTxs(-vB7-T5Om=F}G7hERIx+u!My
zgq<0|4OB%X1p(Hl(!^|@X#3m8sZO+P)fU!mCwOIRv2VHAM>$zppi&FNl&UhOgNhij
zUBM|qU+w4qDQeTxzx_P&2$(k$pvFN$H3k*>AWpc#0C|>za)S{yGzc9rnu*O_UD3N$
z+-s#6lCwa(`W<IsXx3hJ!9&+iXZWLgT0!cYw-#=acPxkaBTQc>O>g}@)B|7>8V#Zc
zaMV=8MY86EtRMl6IOKki!I>e&f>76avU<MzamY|C6ack;Mizp1=?Ea~yAgohVqQU%
zpG1Xm?bY1hmIb7LD$$@7OUbSBy;C_UC)5Ump#+Zc@_76hJP|aT(y;(7jK*n&QNTB1
zC?4q+l=-&iT69ZLa1#vI;qghquj5&WEfwfrthl9j$F0VRrjs)KcuHf!p0o7R-LxlE
z#tV%zU;vDP3zB`98V(mwiZc`|?B8A_yGHGD=&sHpRVrxm00>2VDDRHJ%`r51?twLB
z-UzwB_8(6ie@ai((^W^Fm(c?^xolE!PqVr%MfQe&YN65;3}Hd!P6x@h$fMX-M5=!y
zhX|yKqu@<?mVRmMcm=?Lh?lTo#b(Paz34p%pAh}Nd!afR7qNSsh$sVKr#VrXdIAex
z?q)FXYFD7_jjQ6)Yh(Sw-i&FFiQsrt?P*I@%Im4_k$(@JAp9BEMZ9C!18SQNiRO&j
zH_ECAMEsJOgODGS)h<7nax0Xkd5=MjW@res1TX2qB?u5~m=s`v{)NclYX1<BQUy)@
zx|3;|ktf;V;&(-u6y~R}dcZ6A!G8M)_-_-TYS`ueJ#moz3K6_ef$iS;p9?9H@{D@I
zrWzwjPa=JJ@OL?dEytJ7$m$C_0b2K4M8N+4Ou~=7;*pqnB=xW=Z}fP?^98unWuvy>
z_Rd|E7rR^dR*loh8=ZwQ5(Qz_Gt*6(y^noQ;cf<q43;QV7>m#Wp+U);1PVpIp1J%j
zadC(s6c?91fHl|cJVAc=<Fb7m;?~R1R8}8rIGKIb0tfKJCoGK`*^naa`tD5MReBwk
zk+g9at_0Pb9(l&J+i5m@P9$NcL_a#W>Xl>HNs8Z&O?4!kAFlmP;;O#Q?OtW3aQftr
zjmBms|GMFLMY({4Cv%as-0^Qhl78_wKK)AVdsXnQZiVr;OL8RV_uiXAX$1?xRGB6J
zTu;mQaN>B27X8CiT_T|P<vU)oYCO6BE-I*JIV!ayeyS8-pSZbg-)hGh=0hgRL}W8o
z!cIl*5D<}O5;g=yVrUYYWg;MaYxBLOrfNn|_CN^{!#s3{X1r^SS$jM&LhxE7B|W?u
z$VR>53!xY~OR_FTdiMcy;R0{1L{TCNnnmOSCC>rk4Jb4~!LVh5oQ`d|{X5xZZ1?(J
zv$>b+3|&Y&v2xB2Tfp^YNogd@O`LSy&ALuzeQJ+>a!R9iAn8Q6jsK!lar5jnXJvys
ziLa{Ec_Xlg-9swXGb0DZeVf!reVoer*7f)DBsNS~ztf7ox@PoVqr1Jfm`aYR)p(_l
zu-_RWr>&?Arl0r*Qmqmv??2y6muodm`W3vzSi7x$ipezjyGfIR=jo(-V09D!9!d(f
zYA;~)VJ1Z_F+5p*ot&y2dKxG2dlEP4e`Yr2YkUVyd&Od{t0Ma-T3inUL8)n)3NV3Z
z5qw_HB9;!s#LnPSWCaI5=tGm4)1k6EXYIq>4fze$$3CZ4l0YB|?Q7&oxK+R?0-cf$
zq>N-3f_4|g$O5{o;}0VQ!%m3#mx~iwo<*YEg6j2PkAW|9pV#@k#l1SKKE0kB?>R1!
zsb_~iWgOv^eB_TJ?3u~f4YidGgv7j)AQ%o5PDoDe$1X0_T7BWjn=p9Glpvivx##cI
zpZTYMKO=r8^>@;qo#?iLZQjJyPk49t(!IhHGnYzMF1?@n^Bgn2Da2<c?nxS9BJ>dP
z+oB{|c9Q!~&Xk?e#Lr!S55>Do`hC!ED(VjYB-Eu<9w9o;Q`|I5?>{r*b6e2e&piDs
zljF>sZ^OkSTN>l;K!HRsr}IV<^d%1hL6L21q0tYtYDm!tlC*SS=?A{i&k&)zzF^^0
zKaB%{8Wd_cqF$;+7sSMlfdU9}PzU&v;lD;gkZzd7FQ67epsO|`L7~dHz8PQ)3GuG5
zw-+-|wKBZKVAEK!64&<`g|w2ZVw8t^O+%OSGI_r+Ez?hmU*=98Q@H;;w}nvF`%Kxr
z<z|i8N@?Q@LDh(UyX^oB`K?Ut1zM*U=6l?+HxJJQU0f+%f{?TKC-?|^M<yH!ty81a
z{#_PlW-D$DONn1r#gKZ{#vZxD(I~X{eKRI3eMfu$r=nVc&_;g*U>^ojdr^`N9o<%{
zOR+m~o5?G0s^W)_TwW9BnUEbjdcJ9(1G7$)PC%qvw;<oS(207NXIB{(2PJ)7h?;)g
z&ngOA!CSBfKt=o%)KX^=z+w%WBvnqt9ct<cUUCGhLaSr01I+h0nSC=)o01Aq^Gcu{
zhUNO}dEbY8Cs=s*`cST>bN)QaeUy8W{`^)j{)S>h`?$vBg^r)W1btD=x}uB`=<EfK
zK#X?7TP&v5$ZOA3;@t+CYam$fr`qrJw5jE(jYzxJyvxikPAj_mzsZezzUVH)Tf&PV
zTD+V3(M|m>+2_5+RK-+-+Hro{VvSd=ro3`0WSR6;Wpx$$M^4ndQIZC;>6AZ?7~th8
zWHR>>(0Zr=G8o{HOpJ-#U!MN0=DTo1fWmzQDjgvqI5}=fCpZ)_e}~ITjY6dH0A@#|
zs`#@Wtse7aRsBeam;ldk234mBL-7f)%;>tAEK*_m?o(WlWRK8U$vItO=*Om*s_*!q
zSma9`eFGD9bXU%!L4}L%gu%r3!Dx>dz9-UNT5p6LJ7=UX`d&9vR>&TRDAF<_Ive|X
z76-{$jb(x;VO#W%+yM0@f_2c&Kei@eXX})o@R%?63!7Z&tVy1wGN@sUovv!E3~>85
zhw9jqc7?r2GPz5}a%CY&mF>y~@9R!Ks&+LE%l`MM&O*h|yocGONw($USHQvPZbv@0
zao9icq>U|ubSnLNm9Oi$#p0J}5eF@Kk?F1QDP4TX6d+!jJOPk@184P+gnkpS@gRh>
z0vl}&9V@#oP@xCDC1@dhHx)qH43``>*JfAsqi(NofgQP6G5NwVNu4_iT6!W!(XZ5E
z78xc&xX|Y_tR<F(83l`lLKy3E&Df}HUS^jK^oj&9m)doYsoavIWItxwmTF7tjawC1
zxy@Ebq&xM?>3Eiq)}_f49O_$vXHJxwoKQX`5$xwS6%yyiCe#)C>36B%!lm54gz@>`
zf}@hD7uvTK_fit;Xbnjv0d5NQarHCuXDaK2g3~mbFU!3kxw4>O`0~Qb{=lQ8+3%ZW
zYHb<1EM8|QE8@rP`oH+5Buqcrq@5ehSw_n?kxw(E%|!DK3~x1w=}bdCk>7o5;w*@w
zQOUtbgffu^A?Fd$8G?f4fD4Bx$>B}e;$o;Quvs1S)i=w4-BWK}G)j1mMANSV;sU*U
zA*T>T0!n!rPeS%OAdqZu<bEE=a8__J-3Hs*fC*DGCpE5=S%B&WuO`Hz3d^@pKNB=j
zDY$pV&Wl*i+-})wC!JLOZWmMD-|<X@k=+QDOsy7S==B0mQd*8`Np8Gz-d2~Rp6p^>
z8kD0a{xo&7dQ$6Z*)MFi=m+iY%U`j7;vy!msE<x_TI2`5&RxC0A`Yk+1Rj7v|4W5u
z$Sc2K=D`J#wv&ADrthndffphH)P(@bAb9O287^F)UB(8f&J{m;3UViytAH?=?%0sG
z{^0XV24%SmY!Og-202mxo&$>uoSn-6#>(U{IApnaEj#P@2lBVeHPcvVv-HS5H61Uw
z!%au&9>tH^!atrQ`7j->QKez84K%f|id|p7`XFX{n>xOK;!V;W-si2HWd#!q<)=AI
zzQ}3ciMU!2fGd3Dsv<yAx!v06tUY|i?L{Q|$}C4_V<^?*>$s*bOh!qUwJom4))ome
zXL31xJF7&6)to<9@P+gAy_*ZpKIxt^UzsMZPTs7&FFa6e)cwsPtx;%3G|X<rrDZ$?
zBXD+c{ghrr?A>+sR?hJqH`P4Z1D6jX#{v+_u@&MauG}yq%MVh&y#)f=J0P323rhlp
zn@E@u@a{jTF!SxHes@r-jQ)1u%-y;pBmp+L)u`RTK^~TgWm%DgY%0z6;}ah?PSfs?
zDfjxmeI7k}6Ti>Hym#w8WQEq`+uP?|Yvru{LnKjs?MT)sqO)75v1Bx>6Icl1w%_Le
zO*d#p8gU7(OY{uXPVD+6@JER|gHJz3PBzOWo&LhlNP(~ET{#=wERD<FM4j1Ozk8=d
zp5e_pqL1vX=ozVZ7g9N8iOKC|O-Yo#A8rsrZE@F`)1rs-iCX;4s`Y{!MR%{Xc_gvT
zcbMXBiT~yvJ+i5Rh3Fp0R6sg^R@OlNlzX<US+KYna1o0p#E65-!3Dbsa#0-+VI(j>
z3F2>_yNLuB96#to*7?B$x^_7+l#<sw)}vACMRN;%=<!qH$(m_>VzL4S%nht9de=`J
zts)3k6sjGXB@3Lo%9QOPyK}9Mw8wKPd6IcB<|U0V+pMGcvL2yymmyu`4pZj59j`pA
z^29J_&0AY~C!2pi48KJGNZ;<;NiUo)*3#dK5oQ;hPyMsF-={296d%@YE&RdYPM-R&
zY3|6cq`KWid^f{~;I@RBnUlivTHkBqso4lFvyY4tZ9EoUYW-PF{e#D<CN6xOG=2;v
zSPc#DkxiJpQ8a?=f53qr;n&NM{teN@KzH@3e6g4X%oE#)aGuW%wmLM+w<dCR#WFQ=
z4}^k(vSwD$Pa7OuxBIlu*W^)n=Q(lB(+UbU2Ei=CMhQ>j*6FhM?l&==*4xO`d*wB9
z*tB4<*vXiY^|r`&Wm2#PrsRZL<kbgp+IAXy>nlwhCPIvD_3BgUOB~~d(XVIx%$Fti
zaShj;YG0K31ZF(DxlVW35%p$`rE&4^A%^v_l15>jz*iB5)b)6c2A!IewSYP&R7BO?
zK|hXwpE}~NU$SVVGd)OjCe-Vty8T-EV=k?-9!aXz2(ii0Rg9}Lw(9DzO|{fkR-@NO
zxw<2_1U!j@*sO3-U^4!*z}e>EQ|JL{0#Z?Qno0)~!E9g(;_(Q7ynR|RFCEA38F&CN
zoWQuUh%QspwrBy;r6CUE?;*MSO^QB2Y;rHR20zOt(KidIrbai#;eKPsm6qi-9uKAw
zPInb+1kI-otaqI5Ncj3jT_aSW#$a|J&-K`{=!J>!re;<j(`c`>o4zhXat@KV<YK}G
zx2;`^y&tCx$h!(TCNw=QZBLF?m<wJ1h;NS0G7ZyzLLxL)*$l>8ohw6S2(z!LMYr{^
zeSIZHQ^2SunxnW*<@!OWk1O!r^BQ(1`L*aZ=>|csYIa@L?0uaHX3#-*C?H3`V!LcH
z1wb^^tD)(By3iuWoRNX_ADK_x%$3wIO9bJxLXQA7x9_8ICV%|Fsr&#+)#0E^j}5P{
z2j$Q9PhR2(;Ype8t#SOuxYXBo+S~gg2Pf~y9^-Hf&5I-!$G?Z1?dd#eOqdvbuqGMP
zX@$IA3{S71&H8KaTex1QOT9HrNA=Oz3mu!mQJ<t`BPST=!!+a?BXiASUE%#Lx!d}_
z+=-peYktmWA4NFZs98B&D}OWP{k*BCo79=PXC{Ve-r8|Z?;`ZsJ!NC`&z4VB2;e_w
zH(Z@i*P<J}DcH#cTJlAp%of4MiE@Qd0fypuV8+O<qn+;4dzsU4|5d;{rP76gOIgp%
zuVt`8@=w&h6@Vm3srtb40AezfM0nMD+TYV#R<|pjn;Ml7WyhXmP^e|5{<)|3l(REA
z_JNo)2LW|%?*hRsfjU~=!FO{<&T-ZIA*uV9&v9iRy1Pak>G4x+3FBHUX&>2AE-y^8
zQ|XH-y^R-sUqV^%JpS0sh=%gzf|~)7Wdzdze|Z@v2?03bXS#;|xibA@OSe@&%^X)v
zb^pY(m-DYnyidlfd3jZloNJ<d^&VZ}7A7Q)jqqKV0v#)Ww5_3xt9sg`9l6mYU$8(6
z0rxWzvI$V3ypSsA>cQlP#PpQQFx3J8YiKyu2EV35khjZMq&bD*IWUAM!%)O<w^~Ne
zU;x}adZ){G34M>Xug<@wm7_O}Qcy%m21g1DM_zP({P34qs=sATAMsVv*~@I>wVL{g
zPoF0ruevEBp+gF_++edhabLwFkK7t?s*pH)HEUC8Lb8px_WZT;I|(Ni!X++D*xV3y
zwD6r?;#+xr4cg#2T@eS-ruaA8l_3XjAWM^7S;iSE650}k;EZcl?|^d)%!?9eJ*zmE
zPsOPaPL$z4B?wF~3BH?=#AvFH?)-jZ)V>IU02&NIrb7fciQ!Do==-H5p|gIhLGOBh
z$p@R3S|ehVY;KS!`#?fn6_X16$>n3B>3JOkvt&`hGdf<{{hXoCJ&O7|sy!B4>YU!S
z;m@?6VAIX5W1N4%A(v{kLvJ_wc#Qj_cKB0kj#>k1qp?uVS`|O~<>W=HZSI65GNox`
ztK1E{?3AASR{cdWC#Omj`Yw^x_fEVruI<XGds51+&yJ11a6H2+{P*JZ<9A&L#4Rpx
zPSJ0+r|l$}(Ysf0H293Pai(NB-ifv``!md-RfuOD3D%Rzys1aiarDCKm=>^Wx(kzm
z#8b~eWP`>i^CzZuBQ#QwbqDZ~i!iO_j`|>^pt($^FGSojhg_a4L}mt3L+F#PUH&r}
z<fNJhMG+s105weY5%Wjf2M278@z$Gi8B6A`ocHHL%5;iUXsKdT*s2G6V!6=gdfd!r
z_FR+*<hfJXf0SaJ6t?)PSE<80^~tQ0URLPOWHm|?^4(4S?W-Z3+AoG#_#cV*xCTTu
z{?=Ah36?Fy3wXcE;@WHH{x&4^?l#3mCPPPuhdPg3C}+>s*Ity?spsFY&J-cV`^hy4
z?PP@isR>?^=Pf<S6Su4)^Eof;eL-sT>mv;Tv~gp;PwpSP-7ZmEq=VAYwz+@xj}~X~
zjZ9xWM~CT35CmNu)77<Q?#mw<S<@tufz*iJ`u{!&jB9uz<O2aFRhhvMbzq1E0WYZH
zMTh;jk%?%*3{kR3Ak8jI`KG5493NMB--E|c4sW~#*Cr(WBC<g(pi3Rc8aXVix>@k7
zIosz<UMr@^|I_{6?kLZB?puB5zFe2Lx{vy2eusNM_FC7xj$7b_Pw*6K!!g+%<Jfi5
zll=XMMzC)4BuP}Xc8Z{=vr51Z->m0du|B)s!&O-!Nb5rWSn&6d=wdO6*u!(zdHRR$
zvoF>kF5Z?ZYONEE46>CGDxTn7vEvWlkl6fSu=AA5`*Wh@x;Fj20FTz-S+vx-zlU^C
zqB=g!#nOMyF>lf!=9eOjGxZdhyM<2j*;&GXwRh#AXJi4ZzY>a^g^rMzbOXW(94MjN
zl;89(pCp1-;3sg{op+H#0%1~28H&4)S4-!_Tlh`X|4Y|M*ViT_OOlh=nF=IuJk1%I
zVJcx^vEeIosdCyn#izv;C~X({>No!Vs!uT9JG^<1gyN~J#axP-_Mahpb0)p;2>N?S
zk9<XDL(RS$k57GOY&sS+d#|lOg1TsP>4QCaZ2?RY+uLfIP1fQb&X{FXN-y>6;pd8a
z24KZ&%e&86e^!##WT8z_ws5;S-6D{olqUhL$fbAFLHc{UtV02nHV>?CW5_AwPca0r
z@(o!Hk?|k&hR92u3PS6hOF(8JM5H3PtZD-Z+}O4<xF{s$g%JUX6F0dKZSQ^fv&>Mr
z4)Zfx)?bf1!t4qt6Y1>f@QLc*?=l+r<a`@1h<f~LwU57tV9Fs|VI%jRu;<iA{1rJH
zT*#QP!o~m7-(K+gcb$e-jb=9CMbV8=q5AblMD8CmE(=1}Z@ZcPsR_ILdt}`suWxgf
z&N!1~r_}z#`D5>@06R|chi>KA_yoh1%;YP74+UU7BlS`8>l*Rej{}Y`(LEEHDn)C0
zS8o?2B;P(-8(n>J!JtX8D0<B&_+5IP@l-@Ea@9rhD3mb-5zoJ*EdgqF$76j8G8Uo6
zeN{yFLp5_49>y+-LtI?F@j!981#t&16}f9-aEBuA9YFs~(B|7iDiIN(wxRio#Y5y0
z)zJ4<7DCb^U&fE0%^1EFm5QBQFDtIiGd4`%u<9+lX&~Lqe$hi3I~5TnHA1JX6-*r}
z=tWSny+UuE+%ugh@3L~jqxYFhsFIw3K_RVW((t|b;DT@A58tMm=%W%_I4}YC7;cMr
z3K$V;{e0R-TJo<jOHWWU&zs7%KQ5dp(koq-x+thi2=|$Q&`|9cp$V;#Snagx$tZbS
zyIyCOy`K8^5pSG?VH_%#^!)X!s_WUE1ACay=LzaM8nQm;K82s3>DcR?z$j9u75RI*
zv3lhc9h$k7d@8-1QPGU;q{@e{e-D|;UXuRb)WTLVV<1!KpN`fsUJFkF;6JxVJ#G_#
zC+dYl$cF;**T~AseXSBiNoi36jMSU!44SCLn*XugaSM^h@nrDN*I=$AnsKq9V`5=m
zgQz*1at6cday;40)ZWq=m3XnqpjmK6OGn^D?!{{x2KRz>yRr%!BWts?F1Ju?S>I-A
zQQ|8N9{p@4+!Y_%C+O%>T=z)wHdooe7)jY92mj(j*#t-CoZcW+u4CFzR%TntlK>W5
ztu-e(cC$LSl9ea{W;Dm)hMp&{vux;mS_aKNc?;E3B>dX7ykUGn=Y^#1%-Ug7Hj}*k
zw(O}=><iwv<+B&l+X_c|gP2Z5`zmju3eFAgcK*cDYDRz14f?)T(0-$sjSjv(FyBbr
zm<C1265}v`O!09dStcfscHZGAIBs?{I-Wo|HU?zbrqHVX5Y#7vU-Q<GfDAZWh~@7f
z6z<?;4bz2bPes@UP(OempHU$x37J{~8V4YAbJup*Dp@l*%jAz$^tru@=_hbrmlGA#
zZ1vG%{T|8vKAG2qZ+zIUcPc`FRwgCdjatUi`{WF}w{rSR+frS4a!Twj%`?m@)Msyb
zwpqUYU3$|fbHhC8vbA*8FSUy{yMvo2B-_XzChLr=nYY5nwWvKc1hLrN%W#`U?OG=Q
zUuQ-}3o#FaO9(L|a7~b1Q~$GACw5gu;vfZ91XI5Slu+SKfFL1B15(~i{2aI6d8UHM
z%ij=;w}I>_aa;Twf`XI8DAxmlha=EX(2nJY**eoj_0;0M)e4sNAA7ltnTWP*Q)1e!
z74s|jG~0;hKA({dOOFhG`{=_%qXK`C`0AAD_(DmUo4lut>{aNDZJd>EEGt<4kt>`N
za@~u&*Rg$FMs4_XU%{6?J>vj{kAVV@oG!dDxcaV#@Qw9dr{n7gU%|3>)^D3~W4_gJ
zid&v>GxnD6@$>Se-IA-Sjg-;#Tp?*^*WH=^qE}U;SF<lBV>c4(5|OiWIs%t_b3Aq}
zV8J?g)a1Jq*Qeqm^2>F#8&5I0dz7aVH#PJEMeo@XHG!1H8NUq!`G$Z7GjSX3j)S8o
z3Y-H-*6G(`X9G~+cv5MfEPAEj0IcTu4*DZ9_d@7?SCWGGe24>M$FZXKRH^n=CYURi
z5Elt}7XZ-EB)TMnpc~}Q6rkZccy@ze=BW%nj%M$gq{*<Ktm_c=aZf6BVISZr6?!Ik
zem;?Rj6XkEO@U&{*K{W;n2u-u7JBg3+~Ttx|C6O|DBqf|a)th?0o~{YwUaBizsT}@
z;3~b+jK8B4n)-%O0k5;k%ji+WF#Pw>in4r?<y(o^qxWw}*`-eWsK|ZHEu<g9ov=F6
zeoT<jMvni!;IxqM^yIoi)n}dxKPMvI<H5s=gs|LSr@VVJmxV8P;1X5^%6VU@q-BSY
zrADS#r;WlM=6uL?jCsMc??4{U?J=GoeeS0Tu^lX-R9j$Vj^Dbu1|b9pi;Q-!A<qMZ
zktqU^ex$B`Wi^HW{&nVxfmRw=+k@C~^Aa8p8ig`E?2my9g{ZlJM2Pa^tYCL;vb1yl
z7hcxghxw_h8-H8dLor|AX9f@Nx7RBE-uhdd94#uxH=Gg^E_n5Ev9%vx`Ul+NrwL*4
zJ)@l@_s4j-9D<{9*`5Wf=bT>GD&)@!%^B`27+=-8L+?}&r*K<@=a1xYA=)9RS0;;p
zlMq(t7v^sx@yNSAHtSdEP=08<zm4PbHuR5~bbaTGWL>Yp7Lxug0})AiGWl4Jfr@cD
z-%p16)v!<7;k_kps29n-AIhzRGh#zuz5PA@Lge*Qe21R$9|^;sC+UTKC@pTT{XJ9@
z{maf-uD^#0^SvGkcp_|s9$|?6@DLif2rOhnW3-~_frryPNyOIB0{Lq0DH)tu0vRhv
zo@M9jiG2p=dj%CM^<N0W68ssFckWml5nGz&e(X%$comvta7hX4Wrrr`>@Cf`(7qgF
z73WB26HE@vn_7CCluJrt;G1-Z0)?dycj9_>zBJ~ZZ-&ThF*X`AGuCN--SwQ*!)MEV
zT3_2rT&s$jdgiRW^7IeIy!lx0gnOa3i5*ko^YW^FH}2^yqYRA(L6dWqFi+ALzi`Kf
zJ|HNp_?Oj*5&d)@s!LT2eeWsvGm1tZq*%LM;_x|(b7<zdp7C+`Yj{WWDoAw7b$FxQ
zf*KHThxvPP-^6;Vtv|)rk|Qi@f(xM|1vown!h^Ui_XhrC>MoqWmBZ#sL^5l*;85T-
zVmQ!C5@EHTZ3s(v?nuaAmU#Ub$Ghd``ZEGwyLVVY^mvkNJT1nSYQg^ded<RI8gFcr
zi$i6LZ=^qo-r`6WChqyJrka)v#6DL-x%)}=dxw4In-oY9$h=8uuq>0AC!w&U|8D;8
zAx}4wbL!x~Zbk3flXGppk7l3o^s&|JxjEB&OH3@+({-QRc1~f>&5M2#Pv>tLd^ng3
zs|$61Eu90YEAjDQ;|qzYqp5ix$~hWwJ)JA1PgQq2m(lF^9E9FHNY|p^kef_-f*b`w
zpVTF)cf}I`=K$L`wEnJ)*`O}FwE!KCniw*DSph0}1Q>vku?)rkge!tmsa&kMAE1x#
z>rd~;3^aP0kH%%lofIM{7;<D8-jGGl{Gm#0aQ1rKChw@XSfSn{cZvJqS_HeC=I~Q>
zsXn_1?fBGNGaILF==3$1X4t7d7|4fOB+^J$IMj`sY@<&eaw<Go{GD_AbGRqH3qCxQ
zOVy?Cbu&qrmy|Z}PCEO6$&)iCL-*;#o8p`QESRq-Ng2g$dKHX`gr#%H5SorW5kX6N
z=xgb!Xo?h$L)H7QvC2O`ea;OA|JbIyHoh=wl9uF@@4h`=QuPV_4p-tFH@%k3KK+Rk
zxW<K$^Z@rdU2rpxY0-!tBtkzgw-$qQM85*;FPVdMVaIeB7dkkhKaFC2cyHk#Vt`Lq
zX5JD~waMa}Xvi>l1(FhIGvyymZ(uZprE{M!+j^QABOV7RkQq6);P!^|2rxn@CO>_<
z9h{6~(()7!Bu*cF&}}E{2D;PRlCK0p=u@JkT-&0E$v;isIz-iByYW83aYfr@IlVUW
zPmG<J1RJm_hoVJ#4bvr#;>y~5_g>3P8NK80-<8)b4Zxb&cXu1!X5`j?$XItw&`ZmZ
z+N)7>C&57eLgeXN>C}{B%hvo&KQ>?GEfuI;(`<6~*BKWO#9!m6s(syv-!Ktm$(fTL
zYZy1o9*z3GBYQ?eD|$YZZY^}tA4vfeJf%CQm*2>K?{ZPI76-&{LZo-nTH}{YQjnZx
zIx!R?sNr0JBuhvpHLj${d^BFNq<t2#OM?;c_f?3K#E~1zcMbFq4xH~Fb-J`D2FMS@
z^!z8+k}H?_tT8C}$f<#wyZs+G*IFfBZ9BJ=T^hBVF9JUan9zT;So&Ck8u;8KTT3tS
zax*;1#-X{N`UU{P_^h*_j#SnBGOFl^G~HG3G^wWbAN$1K^)fwuBeYQ*w!H}?1m-h4
z)Kho$bD94zvNL?U%A4<w(&N9kAV3xz-v-vBC=P=w<O4zAflm?2c`ZhmaZ3d=h5gHS
zpWYXV!=P*-%J>2<ga!H>VxfPlee(d6{84&zA>9y(xg585&p7zqw~tL*oLt@Bfwo>)
zlEQ_Q*Dp+Cak{@~jpI$)V>-Trxaazg60J=&iVVS;raD@DreF_^+7t@`sWb?ZSA(vC
zL{=Cik-dHhetNO*QCf*=krv&JAOkHw(VWn@_<1M><se-ww+2zN8&5_s;~=P?KVn}Q
zB(?KN$b8g;P(SdUTY+5*va_It+dL1wD+~faaG&r)QrFmBQgS_qYy0hk<ZnuLkm&qY
zc`7e7&mFG|RgmfyUzxG7X^Av#Y_srQ315BP8I;QO!kK0v^XSCn@<rd3@_E7ecDcdh
zRT7azKzg-@O9_rs4>#n;M%uwC92e)iI<b4c@D8MdnKz)Wu)ypT1#zT#-1GoMO_NbQ
z1ld*4Nt>A=F6#rq9wU{D%_Nnn@2b1uFW58tG=R1(TwKE(OR7b+Q+cJLs$pE1*C?g!
zAGT{vJ<5L!v<@#P*)p@`l{~l74o;=g^z>wt+2?#gO3+l!hQZj-jNKZ-WEcSL9gJ20
z_+Gv*4lK%XIv5^?{Any%k*ui;1{h9emMYkk;2J)tFT>!aUxQHJUr01Nl4N3D6oV`s
z)bC0PxpE{S7Sdq;w+Nn&rbUT<nn=@oRN?@tDMTq6*TRpx2oo=l!Pm-Nv()sjeSb6u
z&tON}HBeTetJaqIl92YZFMdQMwU+vhh$7CTg-7ek?16?Xm}^1iC{OHos<nAGOS?jA
z=sLW8tAvYq6VO4oni~3>GH9x%M)J&esv;A&Afr^xK`&4qM*)`M_;D`jsF3uu!XpqP
z&(gXuTI{+80|;mm=K%>sKuioufPD;dtswTa=rlhP)e9nYz;|G|!DfLfDT4u2<|^T&
zv$vVy-`K|uzKPSOgmH4PTx4+YqR8$w%tk(VDzZdm7mRjrIKs%hfHX3&`ZqzNmkK`W
z2IyX*1dyCv^F{bSz~Kc$B#>gv6&4-_!r?*H0|_FaI0PRN6dBJV+9B!&8ScyR_t1ps
z&I$N|NU=ddL?+pJ8du!HMdV^(60FxFf!gUXF7y49Q(UlEOg)1{O?Ba{WkjOlcgw(|
z4yj}Rl3^Sv@Ddw{Qx$mu{|jwWO~3hHJdKS0+%L}p!4#ELj?K42Fh+-UqQ;%Jfg?9L
znN=xAUMjY?uT*5IByHKbQAa;L<6>>j^S3R{g%|uUz+jx8JG=|9dqQr+z%y~|*9A&I
z0d7&WQ0W8%HuX9S3=g4xx>@U3E@N*(cEo`?Dni{ra3P;X|Ja7^Jh(-)KyXnFrXv2D
zFd8tzZNh6QGJsb^k7#rK^<cBF=~hB|4?B$OL{cDickA7+q(=Rldb4(f@5NE`4+WPh
z7ut9QUUzig@{2dC%Is4$OGsx+c206qniirAqd7@W%V<%zExK~8Rn648`ws8tky~lT
zx5<q)W04R39;&114~}3R8LFg^^5*^>XrS1m?UQ;0<Cc2&=If8y^OJq~f_~S_`Y98?
z`Zr0aznU{~jjzsorai+i#gOYe?<(?LhhlA)SL6c4;o*{pT9zwo{zBf{UR8wjlL4n*
z(AqRSqU}7Y{!2#DnM0?uw(9SpH`bQ>cAf6wy~Qj-6WII2)_xp#ZA`CPca*QfOwnh3
zFy}0BDue}GD(wb{AwC1m+shD?4daZFO9URzjygFGh&DbQS8xqMKjYC#1wpis5NWW5
z2e3_am&yIk!1vmo>3;^Uq*V#Nztc~-Qc~|huDZ#0e)^fr!$&(*l6A>5UOquY4IPJN
z0N(yo3B9itcu9A%SbI8aB`5jGcU)Tl-(m&&)!DjjHB@NJW%A+DwurNwB?F<!saFK@
zc=|%i)pI}39hUl_MJ*VTbZ(;s6(XpH%@l4XF*|OZ#<IlYP$uc0O)_%0OMH#5h!gB-
zGfF017)sWH&AxB(WHb3DWk~k~q)({im95uN_M%&j|G|m<<as7~IQ2qivr_H*BG%A;
z?~l)hUyxRisdzCEV3lH&&<G~)2xN^678Q3Cl*{+UX37dZG+)=IgjE;*4`kMHOk}lc
zc@$ng=u@KJ6e?MR`QO<O@6X&<aH=%{Nj};?2YWZ7LF86=^z<D+^&?%(mRH>un7S+i
zECQ`cnyO2lA7xtpl-SPqYF2RQfjk-(NiIRDoo(<vdz241Uj<Ce(}8EE@`~B>wDdx0
zIjMg?nREM_<HnWjxPTqHJuP!Rsd{eWpJmR%*yAnyFH{CS%$GkcI*ko7^0^eV8%>1O
z#-L_Xw<%RSkLJG6>(t^XGNby~kF}aQIy`Q+KiRp)p&cmhcx7g=V*qpAM}^O+I`~UD
zs%iOp&h?LGUrb+AUthbvqq2KR7Ic+1zq%PB|Dn=m?~yqHG!@#Pm57}X1Y1o@5C~Y|
zs*p~E+}w%WCs~D8Kjp?0mlRU92+Cg6hnQUs7*tV?7{0)8>V8QJ2H7lGhy?OSz;sd8
zo4>>N7SqC82yToN|NKD5a-AJ>A=aXQJPdEj^L1QdroZRf_Zbz>haH<H+VHMpIqn<%
znz8=w<$qhl!C`{g;pPnkPYZu<ZY)aJ%bq+tFgu=9aSU}^{&hsy7fIT?lfSu{)BR)>
z?HZfezXg+DV%(RmMP58YGfzH$fu2%|6Q!p3=CZN3o@^1#{^v;1V-I57AczB%DX)V2
z<^8fChPf%ZtU7IO{Fqcq%`(L!Au5*u4O}9d@tXCYkhZYVJ(dV6HBS488~g}VW&m!@
z+=?G?27;`0snA5YoHfhgsPw{GDA|^!Tns!ZNMl!d)<btR6&v`b5E;scgk{1_XKlv<
z6KgU6?*<hVWa#dzg9W(-kzndhARPk4nzoE50woa0dlZ>n4r+XG64I`bOl9iWEJFRx
zVLEi4ZRj5n5zBPDAx-bpoRE~v=U1ac{t@n+q2#%z7Jg+?Z>-(6xvmuQRrq!wqa}E;
zA;8Ut(SA7H(QU+gn@Cf$8rLM>N0EYytz)Vn4yaI$eT?7k?=*5D?U}NzXP{i1uwV6>
zm(ExHL;mQTwnv&d8gwNp<HHtP$|t&@JlCDdBd_vZ+I(5Pmu&Lp@4xq$AS4y=?d`UO
zjkINsMyWq0dm1hce=0hXdZmO<vgk6e0gPxtq6iQrD_GNhVnNsdMzSi8R;{ZaVDsG;
z<>5vG{0!$U6I_!vCQ-W9MNv0LU={#<MXC-UEI*9#RG3emVb<nQbwNnQ6FbDNcSnF-
zM8O_ViT{l{!k~0?IAG)|sGXtA0ej~z<yrw>;7&~{-Vsxl{Cdu4at5W-*EF;&Ok%Pq
z977A-o{)3cL*GvF9>(D6pXTJdrbc?bOy!EIy2KXwuzt?_SeE=!nt2;>p=Ksei>a-R
z{Gu+FL3+e9+x!>Zc_n$~X_A+_Sn65H7fSrKxuqz{u-L7bu=_0~Ihdo1K1r`S&d<nq
z^E_nFc`t>2_Rq3bkg}(X%ISf1&PmASHD3CAC~I2m##yPtp@`6t!wG6OjyiMA!n4%e
z@r!(a59wE_>zL^MIFcl*%6XabWR^{xbxv?dXRXN!mg>NV_cR1F{-p#YPD1p^4K+ma
zi(KmE{BV^gX9zQwx@q4IRJ#;(;m3!cFtno(tw~+|j-me;-oJPimfI^T-9gBm0xdCE
zLD8=HPy&Gn1B!Pc0g!Bh!rBOCO$ohRq|(%gxL6z^^@*MGePC3Raiyy!%9+W6S*JzX
z-#irG!gkSj<5&mI!tS}>+>NhKrBw4$8$63{#7>Gw)_ZbSowx1fbfRlyE7nXc@w+Wl
zn&a=JM5t>WW$2&xsT-n{aBAH*>oCq4$RE#mVT2y<Z(3Gx&Tz3U7Ubz9i9AzJS<JpC
zdonn28(a2sjRbTReA~Ei{q;xFe^$Y{3kJxO2u&PR)E!%SHhbpZ2dUa8JtMLz9Q;!>
z@2RCLdps3z;*zeTM#`(sbSe;00O;A-H<|$LIL*+=4`tbdR&h)1Ie(Oc<OD<U-TM{P
zXVS`zg>pIy8+@OJYdf_R#(X`?Y1US<EdCKM)ungGhs)97`l$<9n=~Klz?B=}5%c1P
zU6(W++D&rGSxB%q$LO5F4W<WIH5~}^RNCHSDwOz&D7=5Q1oe`$XDUqr|C3sgj7LwH
zB}#8fnyNJ#Yy@2xo7eb0!QU|YVxd!Ut}_M;f;q#<OC^Eq+C=vZH459zw4CC~TSr!9
zn2G0f19*F_C$_{QDc1BQ|L}_FOG(UkFxu1NQR$2!<A~vZ*B`>6KzRmR1ygVgszkWl
z477lHKESgTAc6ANurI4(6T9fdNILIlLsYu;yqg)DjIwrs$GY`|(M~(3I*mBdgR^Z_
zI;_|7U-}I;e!^gxyHlp8$26z$kq;f0G6*qAsMZhYaouxYA3k|sw{@XT)=u&lPd}Za
zOLo@l#E1<_*ZHYW#>#4ayT42OR%1It$=5!cXumrVPuYJu6Xn88;xBb-ot}TZy!a-!
zv@p=vu;QCZtNZe2ub2^I?=Qh=gFe}~wpkiwN%nl<j{B6Pk6_A@!oFMeDp(MZmC$K9
znMzopbb6ybN~Jv7&3?h`<qOK@a53x--#uG)?P&(xnNgjuk50lylC_dKx`bgKxI@eP
z$znFM{4+oJ7GSy#%oAI(gt;{eFOT0%O!f>?A71DH4*wuZa8;86g1(tSt_7+->faN)
z425SViQALP2k|X}(4IsQ5t9YvAtA2+(KuNbnB6Yx^1IS_3RF3zaQ<wt*X#)eSAIOR
z(Bj4wfr<;Cg9G_?rPpku>x;}8F4f^Z@*blP;lt!rFS+oty<}e6{XK5=vGFAbabwfu
zg{vd``|;{|15|7B>lcL+2e&C+pZ0v8G)FK{Ki-vNFYC-3TmR#n{}-o8dJna+fKqt^
z4}<DWtoyvpGBfYd`CU0HBRnhX?8Db3;mrb1WzVXS6D|C7GHki8oeJVz+01SAw)ZUO
zytcfoV8dJGw=BOB@Me_#-b=bvUg<sVq;Oftp9EhqB~6R?0tDyKht(j>i@|}doeDh7
zi9p0stg6OUvp4@%bEuq$>_{WXfX-Zb0>X_&BRd%!==ehU9R7jaO!G9o#Q6$i5IG{!
zFW@h65%-a_1bDz}@Lyq3Lf2yITfKW6GWw2fM_SpL^P7f`ORe}mSyig1=9FY0TyqN2
z{b*n~quG9j{m;Aifjf{=ucya;ub^acyPdYLI=14t|19>+9VVH6jB~c<PK5u&gDdsU
z=-w$CX1udK6@3gkHd#pag{WVx*FdQD+{FmG;MR=dLke6KUo^SczE*_4&-i@Pc-4&i
z?;(tS!-mgz?auMR)$zCQ9rsSXX&*6af90jroaWH`vVPU<><q~>oiX^@Ns_R#<;JTI
zv@+}EIQ3yx&irucxw;KTn5@Oi7<B3XSUS(Jrn0sRJBrc`5Q?CL-UXzIKmZA$mq<~R
zDn$XMC?e7ss`TEg1VR}Vr6^75RZ$3_0)nDO1Vozj=Jy=mZ+^@*Gh8zhPR==d@Aa&8
zFSY9dNw>fLHO<c;Az7Q?0p|Xh;UtlUm*@^#(VapH9VMVf7l}CjToyB-LO+2?_c=7;
z=1SsIa(O`qfbB~NKw008;Ys)i9=UXJ|80$crIzptUiHZM942~6xj;gd?pNd3De#(o
zmOnPQaQ#%xIez1TpC2Dys%SgwcKa}gG|Z-L<Ot&QI0sdm@+|rN+m5@V5_dMV+Mj2{
z-?O>cKE)+cdxs)ikj-I#!8^s1_;AL~C`vgn*Xdb{p3A<#zBZ4^OT~!C@2Rh37ANR^
zchgoY^LYC`C09k3@6Vyl(1twT@kzIyA35IHKC?3SrOgyYo2VaftZ`>2yv^*M6uc?z
zSWm@XP`+oj@%!>yrRVmRKEvHwP1i);Wr&Y;vk5arGr8u}6;dWZpLHI;QmFSmIU35;
zXp;UpO#R`jZCVw|NG?Tlk{~e97lfky3i$foh$gcX>p94C--ZMrbMR5SB+y&HN2V4^
zJ7NHKvGkV;Y^*y#%|kNpXsZul#m_4BU;V6EqaOtfNc3y5O*PPf?KHKTg%ZE8sqF~v
z(q5?clfzKswKP);6<vP?Wu%u(IJ%2SANlN2RHowR6GUs4b&jjG*X&xb_tlURqZ9Q4
zHnXGC+P9vg9*w=e!JOuzxWUxc<UYeg!F;n^<K)O^PPS6L8&Ug4@hv4FX#^c{o7o37
zs)#XE)7UQA?{VYOJCTf9rI&Br{c~t~M=}fF+O6}nfA}8%LXlt)BO68F_NTStXvqZP
zR8O8QgxmaF<?!v>+de@D+|W^|aTMZdKplXyKUAjQW%GiE41njl5OGSklceL&qgd(>
zf%g~Wqqw9uNT1f>*mbt@F)$P(I2b4#Wmcnu%XF{b<bTjqIQr3Ol_|hb6R2czAMs2^
z!F^C|C9nO-L#h5hCk@Hnf^TE`{G!i>dOF*~FUZt<)NWy1^EI~!Jzb7IdA4JNA(nC{
zVNGLVmDAGwEOxCg!SWJ@D(CmmGu5J!(YIIbi5*`5H|BMX=Ipga`}mt}X-idBb522>
zn7oWU?+4ofj3fRuH`rIk6SzDkX<y&Y;Lv>I)1BobHqsDTf4fonzQDL!@cZN+M<bhM
zcaQy;(QWI$%%0Yk-w@^4%>MS>P^{I$AU|8^_n4xt+((h8GiaN?)V$sBciYXA2%KLD
z$dh|=d)cN-{$v2%Z&aYxa#cc+<NFNeA%rTHZ}NS}?*zr~Q!}b^)-j9;03dq-BfwMK
zn+M%dl?LffM8M6=;YX#r-e&dW$Z8uwi^Oab?6#1_wcwUHbY37b15P?g7@99#kp2ZY
zKGG#E->ll>-kFdb%d5Hj>*1Jg0Zpo0-F)7qo}02x#+<^D8ZK)A#RQM9!v%G2Gbcjd
zeib<>!y)dx8*}bfz@hv>2m5+C$HS~yT5oUxGt8%aSol{+Z^jq+bjI{)-O@+&%I2iG
z+m*ubbth}ig^%am_V^K8_%7W^$`oz%wxBgOR%7FFoS~%Zt*{2mFIy~e#geY;_dWRu
z-^Vr-gVsfG^wWlE<b+QZl?Z217fC&ZFoOjNjHTYCfUCujp!fybL^q@(lAn9;-y(!n
zy!)Q01o&scxfk-l9h~W-{ADhKJcY<t-34lctsN0RTeD~gk!R3~)-j_GIg5@*c}k55
zWW1#PSbpb&hQgKnv`aN_mH)Z>J@aeWB3H~J#1_?CDz5=)q@Q2B-(P^!?(m(M`NEjb
zd5+@?=Ql75MIzBF%)RC!_+dmk%re*-17YasGJwspO#Ofl@t9?q_lG}023zQbGF=g=
z8E%o6g<xgcN4%v}th_(w2$<jv(4V!D#u`yy`L^OkJt}R0pB+$v&=f)qVnD=)qsT-7
zVI=IuNI$UN$t9>piQK#$Tui*=hnIBcSw1TJ+^1sMIlevrlQy%ejoIhRCX`jFvWHLe
z;{!t92kd!PJ^rp!BDAd%()YM`wXi#%psc@)I<#8j<ewKGpiwE1jTcW;KPJ(%9rE!D
z>1We$dLlsvz~2x>1V(2VO@Eaxq(=S{vozp7Lw52l49Rc=dW+2s<dWvT;OlK{NzeQ?
zedr-cq|%?GSi$*h-3x{8TfW}A`C$R|Br0ZC_;*P2MawzEATmM^7T_F=Ti2RO6eaKX
z&^#R%YAYkFb>a9u<!JQIUcy#&9PzajV$N4#Iy=tzcz(>d&-DFte(k~y>n~<cn=W(A
zk2o$|JdyP*FzTx(odM;%YVy4JsW2_#)5UH|znkisomcp_d2>yAz|^T%;^X;}W4&!v
zhH|t#kDmyBetYjgr#8Hxty?idBAvUhL>3#B=<!W(c(>(|LEDoW##bz9Rh@0+%`h!}
zFs=`^OtXb1u#XpcD4iKMl#3DzT0{49p$$5GQ@AqPr^T9>e}A}(bjsu>@j(SCAC0z!
z5Xwg<ZM^s^mL?8h0WlD6JZ4j;%@4BHjf-DO`pa4%GJ--LSYQE}&J_rk^1)drT12OK
zZVztnRXf-qWR<{%*rRogaWV4(?$oDmk(ZlepPZ1}l{d6w7*M1ps)p#41{aSr(gxhB
z_hOzH+fhV)7@-OfYQwd?Jtuznn!5>07yI}?@kg2G{c`-%7XDl<v!@GoQHNQ&%xbut
z-gk&rAK}vDs<7U2%E{1R<=6`t&wsEeY<0)6cr5Bvj$ML^o$6U%#=D`*25IYq6C*qi
zzkl9Pw4{Bonl(ufNxpw{d1$#1GyoJWZdT(5_I~53l*byUPP|`x19xq(Ap8k>tqPbR
zln3F8^~13PKR2I{d8qi{m&V^P(ee*Eklc1&HY+&>L>pKILA>q<0?+5Z*2|Nd!-EHP
z`(`2I2MK%-DIM(`m37JaAhgA?+r++O+y|WNTa+(E)A!UvkK1?r700z&LhnfS9RCti
zj}{#yi#`#ovINS#5hl3r@#kHxQ6r0hhBfOw^=)4Jg2AJim!WUWe(8}F_yo1M^M?cF
zg+AXn&*eLMcK=C^qD|7V1P|TGLeuS`-!5BgCMp7_B;jVy0HlnUPRBp`&Q%9CcrR#Q
ztrc<}-w67Nb;Yz^S3tz^RXg=la|LILUIEQyZUge>dxOM;4+kl&!+lOOgcI8I@}G9$
z&NLdQp?_57;pu-SDbyWPR3(ftyz`Ly+U%VRCV6)og<K}h=8m<5vXZ@DN$Ne0kn~_J
z9UFK-qH$Ge?9)gO)+tjg{qPcPvhY#Y`SuQ1aPpd#PJdC!f#%w==A8Ox+>W1*>Ar|3
zdDVTnQ}z*(zYN#EBs-32DCWt%bnDJSX(#UOwbmJlLF^b4xoqo5PK&Dk!d4PGGqjAp
ze;#wB&6v74k=dUCjaZw3LW4Z8524ntfBqG#T2vtOY*0Kix}AsWuMo=tNs(KCRtaP!
ztp;VVlIG#g{oQew0KjwNUxeDUk&p{NC2l(RXUd4@e+7*1yYW_!$$AIzN^F-t8VyL%
z+^`xa3X4n{8Q~U8azAalTiH?;Sg+qJP5bwTJ!d<LSDgR$3D)cA*4wN+x6?<=WN~^d
zR{V)woliJI+b>bYTH&@DMV}l_m+y*bATLThUGvIaMK=q~n{6v@=fzjlJK)E5xE|@L
zEc+-i7|&SRe$pZN%o2B17+16JaA=)e+#9dzvTL;IawQuSztCE#m{J!DEbMJORoE3U
z|LI64euX138{3<|f)X4Zv_FL|L-8qreuGR(VBd@69<*So4JvZtqn*yRVbZxEN&;>I
zZIwK1>tK+z%Lny&t!bcjgB7rC6e#!Q(CIc52IYo7a4zr0SuaT+E%Q2gr%ab$fKu^Z
zBxOx%(BYrj;mJqWv`UR}15I|N!<meiQn6<Kb}1P)|IJ1B;`H1<eAWKEDiI`mcKEd0
znM#j}E2-@G0@PIt?bK^#0=I*G=6FP^xwR^$*Ci6DBX$aPow}U)HdXC#q+7mgE}KJ1
znL$a_vjl2#(J?=fObw7k;rK+7EA!q}TPI1MCrLd;!2UA7KSO~sqC&&I%wYPZ0_$G{
z6|6q95XF`@+S50sg~BOYuB5Z+C_Yr>1!4mXodD8FWeJcMJ*fQV{zraree09?ZS?b7
ziqamDOQx3!LUrC%b=rs9Z{M=%Y2n0A6uUhEp;<Ypz(l*R%){rbf*f}Oxl)&;WcBAz
z$mbc)nzxS1!z*M0%PyU51SU9|pn`SF@8bMQIP9yl=A7#1#793%A{zn_qR4q1vN*7%
z*(wHBH;s`vHrJ#R(4OB6_iD?*K&V4xiedn+5_p(75ukyt@dIR9H3yY{rs-g^nnoCl
z98o3EJ>E~@TmL(sOaf9SB?7Mved-C85@dN``tSlTi~oH_jx@!|7e?yw0YMDTPaWH>
zcz!)L7n1LO`?ZnPvg*jbBPBTL{LL(3Ov#<u%f$lj_YY3&n><brK)(eV3#4WsRLAUU
zqN~b@G!Ld>4~_wMMyASF;4kHZJS({F{PdB(z^pMoFzGrW^_)OhFjb(7MHmFv^olLG
z(n&$kSOAY2<S^FZ`2-U?6onZeHHN7)Kun5DMd6T<TDUM^`=D-;M#O5UN<q#jaMn2N
z)*0TUo+AH?Ky7Sy8a!<J>bU!_dL1{BZT6rhzF)^E&esjRCYYsWzU&lG?<bL2orouj
zH1{n29CF|J_PK0F%p@oasQsyO{6MBlfQ+pHEM7U3^UIj<zPguPUVXs$7YTot1tV8K
zIWPu~Cu?=cOp~^e<Xo7nnhD&R(2K&rp+z=$hiqyf&jW*9a4V8&N07gtCqa21)Vsog
z({OFNCq2wx$hjCHlKIgNp^PxSNDKvw?tzY4k#S|$5WyFO;txY=$`VHxz4M+cCr-SK
zH~4z>bA4m(G~aUG0$=G*9a8vRLE-1hlKSY2xw2Nmk&*#4L|^rn)UA7|2hMABvO%!s
zK)Lu<@T*2~=MgORgD2FDfRykpX-mR2ILB$u`!mE)bWRuIr~fj4tk-)xkx(&2M5Ckm
z!4RdveK@^ouq$E?tO0<_L5#lybw*^*&2t`DvC_IGw!}e)fF$93I*D&O)rEQs5zDwh
zdC(Db`-j#^Cn*;T2YQTC;*F!!KVf-e2K{AC+KuFio~na%2Ww%84pU{`I&A~8V`drA
zjz(STd(vwOQ@6|(&aZKF>g%MW4dfHd4V>x#oRB({)Xd|+FNOLEh^&iZ`p`H=HlQHa
z6WCIS8-L`d-Zz8qCn&wb41lnJtSUr3hLPg|04P3y6el2noVj3b0CImFX_obE^q_+j
z!Tki%s^wgbU85mD6gJ7=ojUJ}iZ?07Ss7HsqOHz-oZ8VgJJC&Sk|w5a1rxa~F$&*}
zYRZY}9m}Bg4yFj<41hwxXV@&=`XV1O%tN*ieUT)f+U4U=?Te~h&{oY+(}nU&rd*i2
zo43*w+_K<87>6HieMJ&{f!o6y1{?_3M(051zzt0VUL;%vFRk|g_JI9k@0ZU}fyY9=
zC#^9ob}!5IZy*x89m70n4zO8j1CYc8{<;o%Er@xg*cn>0p#RZdwt5!YYtiR7ol-T_
zED=P7b`mm6yz=@2q`ItSX#~E20cxQS479fQwT2-0=s?381>Y!*CiVLqSxsPE!2ew=
zL`Gfd;N-)LoTi9(RlN3lukyELk^Rs2s!D;C9Ow#(F_BEXaU5lLZjfK*9P^sU(3)!y
zP-c!|dUVEVq8nsZ-o`@p$<HfV%xh25UgKBi*2P(EsfWV6RC5F&E(e9V4H9Ii+`wzC
z!iW@VH^T2qub7jh(gC?sE;zR?5!w*=kju%QJK}lcCB}|pG(sLn9Z#F~kNJC|Y+5?C
zd=;ZOjP#rwJ4z;@=Wpg$Bloj%Zgl71bzRx0kE|aZEOf(svbP+$C+kX$<z9U7zH64l
zA^&qy&e*E-@+G=<EV)L@!)3&$XO)lE>$ASf^l8-cg)c3skB50Cr3&#V$L0_2H9qgT
zp4=)N7Q6oH*07geIeMu)e3PV9+G|?S<kJ0`=y%w*AnP1jDTK3{F{i&JJ&kV=ve2tZ
zHe!6YZ-;{3xOHYWZOtQ^eCT>o4X2_-NX9{eIdsTki=5tPJ(*2Pp!=ck=g(xz1?^0t
zJDilh!Ye={gz7rH7Z_ybZMj}i5JS-h`nBd^IN$<r2a&VhXF)?X;?7VLHR7Wsf~nxl
z#VrK{qv)K#oV<Zp_>XEGuxv}G9xJc3N%1X=z<9yP^lUzEYfgju-@)&Vd|Kl)G^>g)
z=id)n2LHP@spy{_@Frz+F!-UwZ{k5(z{IGqgj-SF`7Lr|e@Ttcj7_*@$kY4(D3}E-
zO!<!Fy*Zs)&+)e4_yn%ZYpFCqLSD5e<QVfxd4c*!$)al3V+?0i->2XBTPo9F0SaFw
zP6`)Z>X>E5Q48!va&(_G`gBBO(qq0FKk9VH;`G3=qVDKl?lKBZpS{iyySU#yp(#8*
zb+hAfujxXYP2c+%TaJ*d+VvHIp_dTw<|}CnsN?9$j{)~Jl0JYN`x?Tv7K9RC*R%y-
z*|#80dkFF3wYGf45nyu7-ZNGbcry^^0h1Qo+5a}B0<o4|%>Rb?qDWT7zWh}v$ZPZU
ze(R5t!Om&S+A&fEJ`}Ce)_a<<E1C3Zbi;pSU8iL1AbyGkRk5@?Lgl$nqgPx_)iSi2
zUr@x&u<U;2{wcEOWX9u<1<Q-Gy>qVj$Co)CF6lnLseC0+HtaNM_S5FjrSWlp@&Viv
ziKJ`NIxl<1CqmZ$9BN-v|8uBF`}YB#`Tp$H-@7L_+Yk0x`uajXZ`J8@e?(!=(pv`5
zhYyvtR)4|F-L(nsv~%d)N$gP-+>N9B7)sKwJN0C<q;m&*xq4csX{ZqI2cSr`bH=gI
zk^SJ3)0%(eTq*+_#vO*I*0F5M;BbQ(s3B`>II?MD6bp7SelaH>9cQ(**2*Y$3a%q2
zazpE@#&@J+-XAqB0<0$#dl_)ep#f5W3f-?9q(8mFLFlv1!G<M#PRWt3la!Vbk*P}9
zBv=|Zk6FpHC{c->4O0Ivq@}1#8`Y>kc!nmklVHDd<jKTdLW+9{N{?FgBL)D)lN77@
zu8J(LrI*PQtv{pnHttTV&)*irTpC!wE<4RomL-!~bc@~d)%${HTK6M&<QMjD-YZDb
znUg(KS!eg>P)n01n2Fv{N!}HBI=;&uv+9@qA2d5m?F<QcLQ|0Z$D$tb!K>-UUqlH4
zrN6pLA8DF$+Vq;Dg;&?p76%e0+}zbJKi|GG0?#7eng=NZnFG0Znf=NlK_okB%LR@E
zWL4*XF9lCv$}4AMenS1W1o8$3WURrK?-t<DO|w@1<AvSGmNcVhGW5uMvSdT3xXVTq
zd20Lk)$S+%W-e>MYJDragZ++j>@Nl>un@AgjIg_44r5$OS-9%NTPaXu!|~+Z$fw?U
z*1&7*gl8i=!ftzWJ74Wu;+Q%OtvNy`SDk-~wlHWmJiq54oIUBcDA0FF>&cStu8GHc
z&lm=_4(7c5IYYnNQm{V9xt=>IL^%AaBwqW%SLVKYl6l|t&s_KXIh56R^Dxnhoo9_^
zd-Fc!XI&Q?ZqBe7^U8kxq12CYQ1|9M94H9~JdZp>*fq_^d|TcW%dSOr8s7R*Noc^~
zXKR?AZ{GuwF5i{ZhSRlRsp@;NzNCFiW?EH<{4G<9vcc5bp$1#zfnAtd&>OCTM*8=3
zCB5v2jEW&(P6QfU3SjNMKZgO2RHrzQ(?>WLOlm*$PZASRW?eS4eZ>H3J|PvWOAf>w
z=`2&4fgfIu3r7}32wULs!-(>w2lHmk;hYDxV+7$JuUabtipqk&geUY!I7u)sSUQIt
zzNvlLaozhyfGa9XOL$R%Hi=jBRlVZ+n<?C6k$dd>o(`eUla~qgch2mqvIm80Z@2|u
zT??rAJsTouCaLBLDS3*=6k2oyKk?13hTe6Eo3Y1i#V>IW=tf1>dTK}X4bV=?l-NXa
z`Z&@rG0b;xJ-Wx)fr3c{R>odXjaht;vD&-JPT(Iu83}Ack<QQglfu8K$`uU*N+qS3
zEcG=SektB+E9`9TDt<S1QYvpa%PFdEL1=k+OuM#;s5VTN5NfD?-1s~?=vUPv(|~lP
zvt2?-87J3X9<Ehp4=AKi0tAM93IB8@UGa@CYQ#o2GU>GKFt6&oCUa0Nw<ZJP2X})Y
z%z%_n#3JJ$P$?Y0RwRbnMgtxw<_8XiJBOu{iqiXm2jeF;m$`gs1;edo+O;|KBI2u~
z=1?IO1ws^DOpfh>$#9yWj?-MQDmeKalR(|9ed~;L=#{i3!xQA@)gCit1;{1cZ!iy_
zM5O1cRmfLde=3+7UCc5&;#KxaDs#QWi^h=mU#8M=*NNH^$FqhLRDQI33OT3aDolr_
z=yX)#EdHS!eXiYS5c;ZO{aLFi6?3TGCeG<iI@9_myBWb%LAK*~1t%6<$?IhnnzwG-
zVxR8WZ;<r^<O@D!7W>F`(^$#5R4;6dba2X8^Y`$?aity%im==E-~4z-;_iVMJr{A~
z^hJCMy+L_vc(V2SmAHWVM_gX(A4d)D%}93j+2SdVC2g?_u}K=f^50T8GPOwg%z)Hp
z(V;224<1<Bk?@~05pc9c8@ifU;7HdHvj`&X*ys?UB9N^gD2UUKXT!Et`zT=oKxn&i
zrNADp`*4^%`KvUG>&>)}dP;>P`nKYq<$VFnmH6ov>g+I38e8ZqEez11z#x=$h9(!h
za1pq^c*%&?z~B-1+~uqYk)`&oKbLTp^3rJ|SCNJuk+Rt?65%kua*o~1`}%y7Vu?fW
zfRS5}aVMWUCMIsuXuP0-T$i!nc$Q%@8TVa#*j2G1ccPgyH=eefF=FJS?&P4^<T>Gm
z+<Mur`G-0QJGbB7eVFm*(4h&ZcmBU7U8ksqFV@?LZ;;u^>DN{t_nw8#;i?{$NM?Yl
z*mPWk;}%UH4zY)V#;z8BKqaCjeV&MrG{8KgEl-kyd2JdN9SX;o1*GVs1L!c`rn9*}
z{v#4~1)+lGBZ@W!<b9X4%-9Pz6=OvbA8d>g*<AIwGAe-kxJXb&hlZ|!o)EPDzZzMY
zeh>{uiCNQ{cfSAHbi3M@ggShYJM3b@;OJ~D%AWsSNvLG<RAV3~&Gf@|D)&eHXaUj8
zJ0*r$MK^j_#U5|SoljA2<l6jTpnb_?lvR`Ss#o^*Wgi=2<}*?%epx{6J=3^cHU6qL
zqgzS1H5n5T%Q@yU7*6H;Qj$D|Wng1WSNPRFcWBe$;$RS$l>8N)Z$t@O`Gr?(3{}6d
zvDzOzf84RE$%t%%-`Mxig*pN6NB0T}UspB^hAvHN67@;FhY70m5)`YVI_XY?y}11P
zOV0G#P}N;N2Cul+`sbV*z*W$;!5bjApm7S_+l^Qv-~qh?v7En!>AdYGRI7T*H^Y`y
z<uxo*ilHV@%-aTXZvVlVm_{&6A&?80%|gxZU+7^3+RN??`_=f363=KTzrPecaj_rX
zpwOvf4QDmLxU9n_JM7#jNrLk7o^?tF5ncImwf5#$221H9qD$pFO~al~H{3bydBnve
zJx`c99aA4=D1R%;<+*F?ui5T{+;f%BG-p3OytBsG8~0J*t#ED0q?bHlbk-)bpfF6V
ziDl$-tAWNgW@qh&!$$)F_b!>>I@cw2JM;AD2(3#U!U3nVca8}*ocpM-_f=G?tK|p1
zWagVBzM;U`jiB<277M%++T^-7FW2lA)HUIHHApoz0;WiyKTvMWSmZ&K$pS$zdBL%r
zA1GgPUA`vk&@|Jm)-|UAR)H__?Re&%-04S6oXAl%9&S&K#Q<+Q=_4O>`T`0k(x)B&
z0!>VjBbo@1p4N`^6E}!j8wd+)25Li0bD0P_5GcG>vrYl;fxZoblDmd<`zW0OBf2dd
zc0m^!qylKRC2#{km;fF#3fDp>&EI19ds$J<?l<b<LFf9quSIJLN7;_jvoQr(v8mtX
zSNlv}cBHAy&$q1eEPV*TX4{3aFX&DhwnykhQTHWmUVF|sbj8B^j6sTWn$3W3yiN^|
zX070dhPYdj_uoa){pu^?&!`h<SiSmv!nOJL5c*zqE`!`mQK^SZcj2$AFrqn^RrUQ`
zC$S-DUF)hQo%@dY#x0}uADUP9qx~Z7t7-S`_$PmHcX#+F(X`BfGZr>V<wGyn??05N
zb(KcX8yjdWiu6N$BL~kZB;IXu3h>HoD?}G$#3qPH^psRRRsC<c+vHVt(D5|8fHf`O
z1WWmWDnf>bLx?u+Zj44)RVDx=IkkSaQ{lDnEZVVvmKgw*%EaVwk`Y_R?^3)qGg?=t
zR5;1;BYvoiK;@fstP~Nzb*N^G*haVaTBk9O_Iof1c1%@-V_!E^Ph7!uzY+Et)>=?`
zG!=f=Z$$dVq9khL=YXHOq3>^M?}0$Y>RH|U9VWDfE_4Ztr?k$pvslw>7JJH-czH%W
zAiX}KlREDz`{=2o8gpSR-)c-@CpKH#!=FWRTCjL5`~B#?<{fRxWz|22@?UeLJ*PX)
z?SqckCYM}zoS~L=r`klwHR=19PRhCEM5^;i!`n<_)-hX!vymq@{4t05)^yOly|iUW
zp!K@pPZ}J*tG6E2O>hV{slmYt?xtmQ=>P7woLRu)`@_H-%)KCl))_~OT*R|zK-a{R
z%SO8VWPGxAZcx%c+6a`&t54*}kvZ0;F5^dRFe;xzGBqx6H}}>G!iwT*uvNR9^5+n#
zq}TV<%-BjmQRPGVvRM(iNSDPgrYCL;mgQjvRFC}ki})_DWZgS^?uPK-OX2@^c(gz2
z>zugeE~lC*ssB{lc1IhNaOI1Nw588=!r}BOH_L_c#QZS|KV5SMuf7HK$(Q5!;N_x;
z(>e(r-8jp~#@}u@dk;$O6yL6Wx_RJE7ZBj)rdpgX>vi~e^N9OrF1JXg-$bEZNqMxx
zfAtNx>&x%+quh74nRjtN2kppfHo<G{;Sf2|MNVeias=zbAB~Hky@!D<a4k~W90YXp
zi^IXlaZH0V*k8RW2H0%OBxtvw^HJ3eEJtbqQ2(&k_-Bl}+A6>6Ujml-o-dr`;Mj%%
zfQy)gAd~49=gRMEy)FZ^*ZNuS;fn+aE_1X2B~Rrn{K*n|M|2u2JUWbJ+A@#U$Qtzs
zq*vUyB!9l`!=jUA%uToc^5n_~^-Xh^dM@>dKNUU0r93HlyjyUx@+6M~(`zv_itgHS
z-MFWtr?7_C>G?#w{maBDdVhzau`zv0R&`MK*saS1ytd~;`q^XjYE1dhniia>Jr2Cf
zbBY}{LN<;_bsgM|*R%Iwcajgpb*j|zzV|&bZqLwpv^{svTfir2^pyQr8(FricQs`<
z49};02FxX<>0t^IVKopbP#yHr#ZpZ>M%ZadUuFYk3#4j3H5igL2^5>m5uqLAF6=dq
z0lms46)6I<ppJermM@QujQ(+9g&9=7>A1_d;G;iVCAY38gP(Q#8whz$+rx=KHQU0+
zRCKqgKQf_mFlOM)Jj;@YrjS{*BQ@OIOWHKb+2wJR71~P^g0I-+Ec;u%CAy>d+r{2J
z4?8O=qEans`(NecFJ3*jp3WO(T9M3W&$0}gB;X3GK)We0GAeG_7UFF$ayN}`j<s@k
zt@+_1_<ZIoDUbSCrXN{>QLVweyY!#y^xLz<wx@MBK*fAdJN}R^{UvRQqy2w<4cp6`
zbRAh*QaOP6uLa#2>`6pgwcZjC6o<e?ib%m^?<P=!$&j-eg%WTrFFWO{sooIqQNs47
z>HDhc2gPov$_Mr)Z_I*m%&vbh>6ntZxb)wXLmw{WAMMWufy12~e-0fpxl_vOmlCU^
zELMi+V^eWGyh&z%c2^<(Up|?~{jc7Y7}RUJCM-mB6Vo9inMJ?a9MZY$?R#DR98Y2-
zqu%hp=aLu7qU{7mOqz)Ks)L>d@;_m@_NJ2{Q&C`d?#G)5lM-jJ+Kxgqa@uGWeW#%s
zuD6jY6$(Y(&*_S!D)FX4(|Uq9Ma`@a*=S<~x{PUhm(5AT=m$9Ta*13!ZIw4<nh+7E
zC%sirxOS%CR=6e@mgkRouGW|?T1lrYZX7M?;JG;#`N=VBB2xDveM7a{jLbK@lAEUP
z<v)jbH|L(KwY%Ss+_*lcre8c?RBPq&i|lTE1SQf~y<2*I>IoIc=Yt?@W<>72%!2A2
z2N#}yrzgxvH-0okJUo(wb3Yk(=jl`53xzL|s0Rnw;PhsM=>h^iN4Ejd4-aF)RDanm
zi4C;~3WU}Z3HH+<IG{=(Rw<HtLR!%8nW(D4H~ms~bs6izt~-x>qo)x{v9$RCl#78L
zD<St-%n)LjZFb>sh<vRzt^JZMXBK`gT)QI9`?nftSF|F4=1W%JBhz(!@J^T##z-4g
zGBVR|G3%f;T^mre@7gk5_wo7FC;om)<LKZV0~J(o+zqStr<?EMXqNT;WEj60cPq1Q
ztSL2{`EK(_l)Q3zQgRTab1{5PyVA=dU#e=Hm4kmmt!S3(9@d3(ODZO=ErnwstJ6aC
zw?IEh<RSJAM_J6kZ+2{No<Wxm6?4qAMFQ&fsWeS;g${LP;N4@clh6RAG)fv<1|9g|
zMtUnF{-&vaR!=05M@4G+nAcZ(+c`ODoblI?Xd7Nz4PN)oV{0N|$j_hG{~L<Ss!Ga!
z$-0TL;fR?QZBaiYQB?d>MTv8#WJaa(4uf}x-^V0-p|QJlMwfs2NZYT~Jw2P%s47yx
zI$S$erc%<*bv3eH=k_j*xczMs>`?WtG=hj!S5CR^M%U&jt@~<hIFPfI_9}5d%ec3R
zX~Zh7w7DV3|Kbq&jM7o_ef3O@`O6g~RGL=M;MoG|FzG*szM5&i<1syovOo1wYxzAu
zJ+NA{6WhbjhV+;@ASDY*8Cn0;pF_vfTl4M~dIJvyza<7B9IVTg#Q-=cUVZq#<?nzI
z2|iJ(2zTE6SKg!BVkyzhW9rH|h{p(wdbWRg?}1IqRt*^$L0xHHnB3rtQaL`q&1@M!
ziF083?1=qP8L`svGX7*AZJf2RF00l31D7~@yHVA5p;43SzLr6Sb@-Uckioj5e@YGO
z@Z((NQ?zsSOT&>dCLp@v$}f)XhlYg<72ztjxprDC^s1kDQU%F3Qa20e`?8%I4nvuZ
zE@RLO0VuJ7LB}@_w^09Za5F{hwbmeeP9AOlr5_;Z2*Z3ZTPjQoS)+4)=t43M-alYf
z$%67CIJ0q1?_hqK=qTGBX@l4F(Lj!<nAaMCwF~}oL+d(2h1wC;(T)D-f8&y}&u%7g
z2|JsQlYcy#-0&^)c8hwnms9&G8n^h_SijHFz(<u^aEd7){0{Hbr1{{HF`|l*<pk?9
z3vbii4PK8m7l}1`2~MTccm5p0#)@6<jPosX$dfb+Uz+>j{}o+t?G<uf_V~7=tH-O<
z$I}`{%}n{y86~ODBod@eUG^eTj#H$gwp}3y@-L1qt43o6cA20Af{>Zvjbn<C>9<fu
zcMdP}nBH$5PK0u!QPR(dz0-3WriJ(aZMu^J4iBfWz)ge@qkPMe0;dkmqZoHusBEX3
zeo3cB8G`SeQOb~wqwSNLOI>we5sS=h@}jVtMS*PBlcO9`dbAAn$8oxW@g5FuC*@G=
z-X-WL%WsaNKO8-$_up`PM=8BvbL<c{>a@Qi{DGfvIWqrZRtv3XoyJXaC~R;CA1+Ns
zWq3#bIaERHb(FsvC;R%^V7Z=|b-Kr(p~=bSlZ{yosSL$b>xTS4XAIU^ZW+_GaG^U@
zb5Z62M3^n8=?R2$n}bOA&V&jR=}1Pmt-2aVuxf}vw=t4VLst=r2W<GXlA*_Axz|u$
zG_4UpiI*TlH>oj9OG(m$v>y@){MZQ=|H31kL@#XjJtph~98v?3VtL5U!bDR-zP(WP
zIH}~7(AY071B_lnTJ=KTv;nt~lgloxWw8Vwi@vL^5!3ZdN*9Y>cCo#Y$W6f${Fk%;
z93pCdnu1NJkc-w={%^~!CRBwal;*r2J^HnPK9>LiJ|nQp2o6$Bv2nx(WTb|`&Hx_*
zGUEn6s0w5EgRnHv|JC}c0t(~TM*LIynZvnRE5dXkF$N9Rf~<Bz8N*Bv4Ahc>UrTNT
zvGr6HZ;E=+_N@;=g@kJSs&S|q*Tu<v)vIHU3Wd@+lHb&8qRbmdjvV<#qnCC@$pz}K
zjy_LdVo}XwVDW7_6K_rtzEm%&bd%b1R?lN!iJ0d@&vNt5o$8dtncnz~)$3%`v3nPN
z{waQ+92B4Zh$FHDts(4rTW;vODBoG1>+@AP2`C2*xq!xHd!`H?M?PZVRQyA&?x}^-
z9&BZ*IVS^M$RSC%>Uir}QZrTeLVU_vJH<oC$KxM~lnATr*qNl`zkba8Im8#Jzzhy0
ziOlqG<*>B4oQGK#Tl(L2Dt>+#MVjO76!6G*rq-KZ_3blGnhIsQOW(E08<4C;hDBQ6
zE-_gTkQvk0{gND%;KsjQX6R)@U1@jp%U^x_H`+MTn!}p;Z3I;O+EIra$`wZbenJJ^
zZy7lPKlGL0ARuW^bxZFUT@sq#`>Qo$Y|~<LzU*)EGOH@BV9JrtnO^1%T-Mar#`G<Y
zzPfkaeo-kR*`Zm$i9CO-?>Q6Ycf)w9MXDX&{HI<A#^{DK-VKUN*N)8?n`JrHYi19<
z$yf?2_lYu<4++PJkEiguo@P9Cf8p)jgRUD*$&Zc2eUc6AomQ_7JaQAfOr!f8ds!^8
z-htSCs$FKp-)i`{S3OH}F~2CTq}P)9Z03cmP_Ytb>A-56c1iBIaW~_uyH93?b1Nbe
zvPWomR-U+BojfOl*=%Icv2Z{!+(;r>o)N_J7xt*cv%fwaBPu3hMgG;WeI*2^tE#_C
z+qL>r><3N=3dS<p15scbfCiH7|Ax}Y834#K{zbcN1oj{(*mX`7BSf!ZxvmDJ>Tz&g
z?8heSfx9libv0uutzXIIVFKioZ{CIY;~IfVfXfD8M}h3+#Z-aGS8XlPJsI@9L@^AG
z2PR#z&X&>h-;qWWr=+r!{-d7N^8DiXbf<_dLCKb5G)}9k)doRlE|%~yQ`yNbHgLXh
zXJ@cdxb#BaQ;qtGlKOb*%&we5AX&X4U~Z>piu3%E@9gKVsW0Mhc-sGb8bA-NApujq
zstNmc3-+VMAG@2m1pOE;;avKQ=ML`d%g=<H2cd2gn&gVy70*5QpV#*Ds4z{cqmPSL
zNjy#<G@<DjKFf_Th4c_3HnF;O5udn%($<@7lTuc3!95K7v&6tu^5d3^c0(rP;C!qC
zxE7%X+C9kru6N<x>xVIt_6b|jX{2Ds1(|P;q@te)LMe~o3B;7~4NrkWGWHB+5)_&~
z*ZQhCqdF*@RCx+h{_mzl0h4yjG6eSdnEM515e%X-A4eNp?!d@Rf56`yzQwodBHfU}
z(NaW<b+wHIBZw)DN2(s#|C@BGO3hK-LV{^Ek02y>gK0kPoM`ceK0f&f9_j}Mlday)
z?O44x82fF^kH%lWbJ&(8!)WUq(^_;2zdVkkaduEa6$JH=UqlDI()pc&F^PF%Mcbjr
zmM<tAu*j)6xqntd5zCm!ko(a)_i$x){8Fu0QSrAZ#k-i-5$_rP%^E(var0o`{)EC*
z=;<6U7X7+x-3KmR%GI>>&JK5e66+U&PG=Gr3Aj-kmR1e7_0dQJEmsMzdo}luIBPL5
z&S}VJxOF}BeRmM5_Cqx3#U#$4@Y{i6n5N2U>{H>~o{STV4fAPU<1c-1IBU+}tG<uF
zpoA9(N47({>`n=coXX0KSSh~TLfv>|Qs;E)EkIsp5z&^=&4iTKRxvZ>j_ScW5x^!n
z`J4G|p=$KPX<w9=*N0SjgojPS|1Jhn0Y<jSLG5mk_YN>axi;IUN0cy>wBl3DC$=uY
zUdYD$ZE}n%eOrFC)+3Q_Isr1D*Zk{80%6MBP;CJ3TMOaMSA8nQzD#<#V-3XfPt6$g
zFY5xA3*+eLrvLc6S+(O>#}9tYXRXSBQqeO)K8issjAs%u7EGCI=Jd_#u5YPelio2+
zY;i3Mu^(02Qr!P}Q1XdHvpi>X8m*eTKbs}Zm|!COhG%Ea2y=fpc7i|0aczWC*2lIf
zxjLA(MEI-*Yt`CV^<?ma{JRuKSMd$mD08_pzEbonMHJPMBXaj@8dOJliiamJ$=n<@
z%&b=H(#qkzwe6!y3n5)>ZZLJllbEFkEmxFH0IH*TjIPd~ZKaY&89Ab4O;;oYV7Vq1
z1q#I4z!H4PDXBo0keCAaSkm(4K^$~HgHe0$=GI+=BOwQ?FMive_R(DXn$(K~j<5cM
z1w}mJHR}&^TQG$#<KONLkxzY!Py2bPBtpfzCsi9;&<umbcuTk99VgE6lJ%}pkAO4t
zCI_`?v&Uy7W>wb~u}r&c`c50=Izutm&jr}E1gkGpw|(AFH+xLC$0oI-I7^$9cJhjD
z33s&BJ%_#w>~kfFhRX*{ZqRgeGALZ=W$0I#vBlDN!Ott}Rb~flx)nOl7YGnd+w)7=
z={{#Usz!898mDV3dpYakPC&sI(kI0Qr)uPPum4&RrtnbKBqS-Dw?LuTPS|f={y-RZ
zO(3Jif;V~(6G__@rV5qo5!)3TBB_CeM|ymt%=#Bi7okxG7mNw69QXg=ITpEpEHb08
zBdVV7{fso8XQ}v<LR+zwHRhzF)zXQ_E;np;Y~6q0E}~zN#yl+gkAAyzhaBFp!G7au
z4L_w*-8$py>NOQ^H%DhdnG2uP@F%^Q5*i^kE;%8Q*LaO2Ufr^El=B+%v&BB%(@#+2
zR)uLd#lyzw37WMYy<beU2z0dWVR%9@T}HQJbTd6wT&9vhUo}QPNJTaNp;JNz|4D*>
z+cV=?hfCS8BTKMluzF23C-C#K@zIFPib1k0i*}vRv*yx>h#0e4SfrWx;d%NSgg}y7
z0^m0}@eLG?m%aeD^XFZ~>@4G9lSTXfQGWrSvY7YVr?9!cVRL7McIE8@oU4AUDfH4j
z8dO)64j1-@ZCg>bHQSuNh7cSWg+|ob-mOa~{&%b;JFxV+zHL4A$mea7bF-Q!8e|O@
z_&sLrZU@l>mb4jiN188{zMT?(svFT)uZa6;$S1*XWY?Bz^ahil%;5fhhl>yr#-USS
zqJb*rmqd#VJoz?eJ1T~DS21%`S>ImQKQ^4zz1=!sjzn0&2j*(9JZcU&Iw;mDnFN<U
zeDjC_pSe9S=^sKhl>O-lGKZk_ZKL~M&56dJ7@31`RVuna(cAoDnfh6Nes42n-k7nd
z3F$$am{R<#bB}NAw_>}{1GKml;ZUgsSDlG_fuecSF+yVb>G-7=Cr{&e_ze2aXy_-s
z;A&3C)+{l*(Pshpg``FIYy+*5I_!(B{+h#S^TF6;^JLCBuA)Y!58jR*C*5C3x__RT
z0KgvDci<Q%`q+;*u8gRJonV5lZwQ{axLj--q}QpYh!*NLo^cGH-2ufVd#|r=XCOjE
zTUT=|PN#U*pY_t;9VjryZbgW6ErN;}y<K?<%6p`>B4#zPg(dUgPdr610|eyGEuA-+
z`3;kj$-h2+nqAv?RAiFC(LvF#ELHuUH+Hwkz+w9JJ#MKlqA~Od<1r^Z9}B-7HSi8u
zcYir^hUHaX<e9Z=FYB~tuKy$|STjXD*K!_Of6AY*DB{CN^xJ-}%GxBBCV0&o(zceS
zY5g<U2xzODNLloALHFg$Q#5!4nQ4M-ORx;EUnt<0*}f35bqH^qjdM-ZkA|bl6W{^B
z`E7Nn)9&bX!#(qLB<5_{<T8>9TjhVU9J)9CZH)_DD4i{h47*vwEJ6w-G3}l;0Hg_U
ze6P>6J6|usbFyUp>OH#75r&V|i=H&w-10*Wv%Y@e?^u3|?BBraJoHesm^NqcG}5Xb
zo!PJIq8erCzc@AK@LYr5?BVz9HQv7@L+I-ClP~EHs`LEkS4BzMB!RkEmB(dQ{ChT!
zJLeY!#O&v^f!I}8Nn7?b#-Q9;3uhy*n*ZPZ^Y4lO55CO-R?zs|WIX;VSlHSPR?TB{
zimj9WDG)Qv#E$<MjvHG;6Crt^P?>2$*NDwECD;%`;~$kz8hoBH9+nWh*B1>|54YyF
z`fe)M{ldKcsF><sr>B!Xn9@{e_Lm4^#Cfv_bG$@X(y>2>9yU`;wvqU`e<}J49Q7S_
z-<i9^GL<EU7sAS0Hw2Dmk!|X3OhsetDP41e+$W@?5DE!+(EzWK1@UMQdZe;eifM-c
zBm^fBJQH9NfzAVKMwoOvNvx0oiq^qwn*z=yP&cQM;Is4_6Ziuvz$E#Ula6Nxb_0cm
zMUP8kMp~+A_ZwC@*fdaEEpzjA!|64;8!nB$5<*f2{26BkUOKp^HFM#}52gawo(axS
z#hkvGYDdK{UUZk3U|kdJ1%8_c3&o*syv-U1_my9s(m5S>J;sXbtGi8if^bgL(4fio
z_Prh2EzE12ANn=_s?M=<2XssYR}mU`y9srPDt#v$!jzLE_PEUczgv)>65S4ZEZ`;z
zF2EUiPCr^Oe8h;=pUDab4+|w#-pGSAec^W@?t{&?thQp%jtoVW3LNAW-jbqm(=E#O
zG+K{EdzM~Puj%O7c9%TAytUvbcl?sjw85W4B`2othp$En%a|?NsF=(*W`$-4tr?wh
zOgr^+qIpbx$4BU5<I`%zt0iU|LXi=!StpM?y?XhAW|y76!tdM7Zl`NGj_6&MVeQ~*
z@GjiG=kGlXogQ>{9iXAUr(%Y3)Fw)Q8a4$;-^m*4$I?k^@qcUTknbTQ<8Kkwpg#mj
zU&w#Z8L$&+#HNU_#0rFuXkGUg^B3nx!DAwC6VppL+KU`|^IHtG(7M=yLmE^+P3ISu
zLCQRzh5tzsrzC3UCqSlI{^sh3YR2L#PpipIyuA~3)-R7=TU5@A*EiP8_hox>&*-x1
zm3_^8{^^_cRG30z-?SPZa&Lph0cq$kn#(@sBHr+4OU?+hXmhln5A9h|g$t=-XJpj>
zH7k>CqJkcub`wZo!m^BBsnkdAzG>jOA!&4^p2TjsP8$raq9tST0Zj0|o{n4%V17%Y
znh%a0!E#|HDDh@fDQ~l)SI{J7bi4BZhT9OthZ;#B`^3kWOhS7wC5)OZT69MDzd1QF
zZ$bn=SB1W4lN%v1>Pebj0hg23pF>t$tI9V%1fA`Ba)yvsUftBE<cPvE2A{#{x<5x(
zs%Pyf11g74Ig|=<S&slFp%OXG6Z+w@=Kzfzx$0Y>jM8Nh=k!ejpRwpcZ!&!`gb$qF
z&lV_Dv_lFU1c?!0#CiZko<>=CGi%?ks4~opQ!~}c6pks#sCl1s*sSS#&xlP_txI=$
zks*h`?JMlNOeY-W#ySmdH>+v?7ei{!mXbl+kjF#bnO~Q%3D3JN89AmUovkRiz<dln
z_slgXsQt)TPx;OM>Au7_jSK;_T52}M{N6OYAa28akV6yh7^Dehx>%hnz4g96jH4mI
z{X|GAc3_SM$L6IY0@emI#7e@36ulKXzV(y074(5;-nm3jI7Tp0dCuUG6PXsN0+$}>
z-O||YLJ63b^k0wFA1AKtlHXrz$VlvVdBNk6)jdsccAR5;dQs&R3yt5%)k)|%cA2;p
zvTjye&6(=UPHEu!+25ott()iU@4z3!>}H#!8f%|Lb1RSidS2kyExDeUzT$;7sH;BU
zhg=)7_h4;O%auwi<pcWT?3(DKL6Pc#K^f}eX)K6*hBJQjlnp475g9Zn2MaF5dT6)e
zUosX+P;EH>lpXH3D1U$x%pBBNVdadL<1k%#M!XX`zp}}*@2L7B`h2V}Q&e*vN?P01
zS;+ZY`BMt-%%&{e)3euJWP2?qY~D@EmF~@*u4sbojNu&ZY6tX|8C3cI2ScB%ez0K!
z9ijgyW&-JsF!4V+BIwUb5UJPIpi0I&E(5y-)~{`mR5bAb?jFi-@bC$QkFKkt*uy)_
zydO^f#{A-8ul+)PbcK+$(REXyoi^ZQcX!b7ie2`DiUGIw1G6JXmapF#exr2D@qU|u
z@<)Hgu^5;Ac;Y=tL1VU0!ckwPck1*rqsvpDO=ZY;F0NT74VoEXb4F^l9}XXBdqrQA
z1@*m6u#!INM<{FF(D`?8P3-QVHr+EZf)nKO!P&f~<>Fj21IC2}v0{tMut0&;D-4X+
ze-8b_LFt8sk5<)SdL=9LoU7j;-y%U(Fs<zc^A{{Pvj){If6V>L`OBbL4XsIs)&;D_
z?SB=T!wdwJy_*ZEQwiJ@sogtlg5jO~JzPH|NWfdW#;Gz_@%<1m^i}zD=-tC6b3#Oa
zth2k^Zi%<5rcV#4&R;<_p}!iROQv}dQ^t3{-Tve1^Jl$66zxcd7P^5|=TxR5DK!_|
zO{5_cxWdo@1c6G@YLWxk_uF~G)bzg>E^P3o%9qEOFJr!LVQF5Z*a<=J@EL^66j<2B
zjQ;TuDu?LcCX3z{ppXuZ!&j(Ex1mad%|*#j{AJHfugrnOM15aB$>Gj0NdGLx5llK%
z00RVf7S6j4$Ipja-VV9?`aRaeNqs?%P3*-M4f~bmE`zm`l>ASZ)5>rQ&ybRnj7fR5
zgac|IdmCq*c)gT*SoJyGM2KO<es7eO`t#__ldQVQ9UaDBmHV#SgqdDvt5%hr>;U|w
zmTq+XT^v&(0EWT&@jtEz&K217OufR$L=67XH@p*UUbW>lVDj%=9hkGKt!)ve26VE*
z91$%`q&+>ZpRMS5TpO`PK}6s)9!%u=<{BGNynq5BqxKk0gIZEm&h|Sd<JK~gPBqu(
z;}N&~3k7;?l<)noaDA_r+&sh6Bq5mPNS&pF9kUc;p1$@;ChhPP)0Rs8VQufnY_;zn
zKNvr5zAUHN!B)C?;<$L><@Dm|1-^{l`;FrvqcZb)HV2pvt+A9~j@^P+U}Z%bXPpL>
zacD|OF!GlOH;bPf_E=v6DRPZJu_Zo7OG))rrQ?Xo>1(+{l?)$R<0v$1${XWX5zNZD
z5oJ)m0mco;G2J!N_B^mo$@3SkeKg3KBPjd&U`1D4TIe^#k|vQ}dL625S6#tWf(`ul
ziYlRiJCcsALbXoyel;|@_SPGCzQTX%&=s_DiCKEYq#s?1(bDj)^7M%C5N59nS)Xb>
zwy+&Cc~kHwBDKHv)}TUH&c24gipY3iI(W1Znp-l=ah1(dG(Vf_2^P|>e0o%*i6zP`
zz()lu83<pHE=ev#J(j-sw}?TcnnQUQC<7o`!w4hS*0IkoCVN>oL81NYh!V%!9NxHZ
zBq<>C9rJenOQk#~66Py{hb7~#EI&W9Yl3ZZ?%4yq5c-b?gSXt-m^+7_gvGByM+9tR
zxGh>Ztz|(L&df(O&RzvqnzqhEL1v`d8<=*5409VcY)FVFZBW~P&KuTQwprlDs>UEq
zsbFLW6FQBN!{54bZ<00qb`W)R4i_m71UX9F&L;qa)23<`hyhcvrT1Gch=M^DJ~RF0
znFx_L@M<Q(G;~cOfnGWsAp9s?mO;s!I{;m*!d*bl)}JoHf};ei5LqcY#fkFH337$8
z&9fSU%qQkcy$FJsN5!*2C`aWJ6a+))oB(Mcns`?#CV!+b{B7f~AD{fg_EX70P=ZP$
z!Y|9zp5O^3o8r7laGFeDwGequtOCM={H05mEXuK%`_U^I{R1uF>p*{n&<LDv;C|Vz
zgf0MWBP1XJA0triakjYliH}fG-yF*EH2honmUk!RbO6ekG>z4bj{8>rQ7af867K-l
zHD2B)3HlJ}j@DY~kfQzebw}2zO1HVj^>59@5Ks#H2&t|#jeu~bF4ljwRwQ^!>$hDe
zN!prEO&Pf^sWbwFVn}nXJXpaTuBoX>C;d6}_Mh`*tv2^S5QT6AQhJ8s3oQImO_i!c
zWgV=bWeEfH5oB*?#LOI+-g*iqS$xXk0#GD-1$Yd$(Kz&FT;K?R2LUk;Pk~-3A1$=N
z0WA}LUE7$lV`P1^dk5FlFHZ_XZaKG>1nSltX;0ktQI1G~uE?|v;1f%r?wbyQ7l3lM
z4O<g1+DE`4@|`j)XzLxW3M8rITr$)%4BVe{vGF})ZVF5$C-A)ir|>7_J62gsnfGKN
z#O49%78jfg+<t+a-z=jB56+8g*%W4-ajADml9*|ZT>h3?<z2ddjgA!oI^xh@0Yw28
zVrY*4b4V=%TsYuDvcN-yRV(#T2=ho~Emy*dS{4lLhfflie-j0M>KLZ1JXkc!TFthK
zi6y&nIq1~xZANq{Gn!Q1ym4>Vo$tS@C8w+AO)^H3W4}FXtGnlIkJ_Kyp=7--gNfS=
ze`vqlJF_6h!N$Inqor2@Sig2~fArkJENN4i(_UJ~yi<%8dL&^hDd=T>bFzNhJ%(lV
z4Y{WDc5!%hGLNCt#T*mS-ay8(vwS`UJ-@EgO$846d<}+>Evz$2FieN_fF}|@@%P-T
zNSiY0Cu#c&?C-V(U^f-mxb<Wj>UZtcxc&zik;esIsG|QyET0KLXGU~}h^kX*#L3<=
zEwjDfyw@R?8ucru)bhvmTbnquO3u@n0SP75Qn@ix?c&)VX~ORvCx6nOY{iL$JH=>=
znEW1!4|A)zx+I5J%d*`Uz4@q&#`LJpP!DOr)XRx1lSi&!Jeic)JKj{c!T7ANo0<##
z;s>R@kZ~%ibm?jAyixB>W*6Ml01u<ulme>S)atR~ZglJOgfmO42M?kz^m3b7dtEH%
zk`7lt&6jBCWo~ZAfH0Q<hY#<x!XCXJrs!pnUT`Mb9sM9UTHuXT<o^9)!O;h1w4^Ae
z0(dhFBI^}ECjm52rn@CfEE!vM&dz^zKV0L>fARsGjK{JB!LV<6HsrKQ(<%Lzu|5)*
z?x^G(wO4UsYKBzx4tHI?Ea-pVqQCvh7+<Izd1oq;MIy5NV@$@i;rI1Fj_^a%Z?S$J
zdx*o?q#(Wpb%PR@n8}%8=$@Su8d@#q_f#98^<Z#v{e&u^Kj?OBm(uP|JmEUIL@LN<
zoO59RAs{q%QhR^gEV(qxu|Q`>B`|h^<8W))8B|cjoZO8sC%I1RaH~#4k+0F+;%6wz
z;KDsl9##GDddJel;F~r?1>{l|G!R*=stO|a09v+_L=|Jf`+h1Juhshn=Nd+&nT7#h
ztvW>KXD=ZK{(^g7VE+$^sS4!L+pLlYhb?JzgM2qW2~Spl-PwO|^Fsg>kk?@N=H%3x
zg$6*h7LeZ`m<19v3QjaSf#Hv8rOY3&4b(&Xtd9oyH{1^{ayD@7yp%JFkid<INIL~r
z>$54+JDt6HR*%MvJGC{^P9}`y+c?qWwCAH|MqAuXrmIiy{re7GaxI_c%Nc>#EndMF
zaVP5?r7g5_&u+<&E+(h{_D*wK(6}jYP-sPJkL7Stt+yw>%oS3<a}vGa?|XCj(u&D~
z*~I4~M;P<4YTDmSj3ys!T7^5k$V+@1AHGx=Az4kTq1tSr@~EP46@00OYsS%A$gm8w
zWxzx!iA2Q$kq6Or)P?ElcGT)}KOgl+3j895iA1`;Z|vwEi26u()4l`mCS?97yNF|3
zi~KjDoEuQM!=wQOk;|;`#<P)YKj&g%4Ut}2BnDjIhY;)1#=z2iPOV9nATkcU9D&}>
za2Cuxsh$yB-L_O3P35t=a6!GoBbQ?8GsLa-(NyuI=ySV&6D72#B34baxbxM`^fznf
z%L>QNn1#yTQ#|}xh2~8C1w)BScJiv2{70|%KE3*4jSjAjiLZ&g?Lpi{#M`UFQ?E)c
z8rAi>qh-cv>nH7(G<)iKdfuc9J!~O7H8)jwricq4Q%nn;&pn&Dp#8RkH__Yn;FfaU
zQ{Pn0hS4VwD3d6w(*2?D>kq+glq0_%dYpE82SV*t`gYe~EmmVlK_X#@#3Dv;M+}U~
z5dbV8?j?iK;qR6Z8U)gt0O*I4b%}s6Rj}BIf^aDlUh5&@Uw}#tq@<a`(}*;RfjbAr
z-|#f=^1(J5U6zpEI&g%b{qXFm5vTg{c^YSF8mpqvMXbWVfwbLr|Hi&q9I|Q?VCh=7
zov;XJ=ID5AyQJ0PA7ag#l3}VWXm{F`qaimycJF+|a;&#phm?ehIJR&)Mo!;0Q^~49
zEqamCK8=s;_&=Ji1Rm<PYY!n4VyvO8V<aIGPuA=PL!^xm4Vf~uXt8HWV+%7OJjmLH
zvNX032}8w<6o#l&DqE60edq6azutLz`(>Ir|9d(2xzD-IbuoQCcXd}v2bP~xy>lk~
zjeD_NLGpR;Lp(%J@`8+p|JK&-GbecO=?dR8R>);*WlG*p^;6fj<2ttQ!4;YH54m3n
zi9}1vGV{L6wxSYYd~I9qPAxvhrsPT;30Pm^z#iNQgvsAmca29wJ=5uGu4$djWR76I
zg}Z~L2Y2|Em%qv^PNY7>5<|b67;hDEiGP~$9HA-!{f}6FQlQg3(%ljR3g1raSP1v9
zP=PL<J<9D(6k`BUAT~-ty#eIdT4PwyD|#Cm&zW$ROGl)s{g(Yva;~DHa?i1x&y%&v
zBTQAUX;W+YEB=$2`Pq2C_=ht~J74(^bclUGZ>}I0aSp%N^>1`PepUEXn*-s|xjT#7
z4q=q^w?C7v$Vla#opJCzcP&H}NUz8Cym(J9(h%uQYR(R9!*oBMY1lH!8T}{T=2WVN
zO0Y2RiQ6N!_swu8(#k*fa~m0*cre~R5+_;9To5j!nqm{I<*|K@`c$9v6@NY~18wBq
z^EM5dnK2iF8jS>OYOU@(g6>7X_ISl)=0w+rjL3-%ktAp*1)NvLL%kjv$ejccE%}j^
zmMO0dV*_%VaANOWLnR5BpE`~x;)L6Hv;jZ^Nvq##Ry3dRTR+@fMtOWDt^*^Wz>wd`
znzm%!J@4d$DLA4jml4onD*R*9sC{c|&W`Ei=w_hDK!@uMhMu``;)-m`8)6j3XMe@5
zHl8N;!18z?E&x%pFu{_Tb3LX_f+caCi+xy;f{c%Fpn$#uq>4N2A##DPY)FYMWg5-L
z>gA(L^a7}`J!rT~rM7o0emJ}=R-^*KL$Fw~sh~XEZeP(-gU+~@{#K3db<71U7ZD~C
zco6n))Po)+Nsga|aJ_|RwB1Uf=XF*{U;w>h_Yt$W-N_$j9tS+_bn($$a<1KS=JM-W
zgNQM@^F_D(wSdk{*Uh<wN*B}>%cMW`9Bq;zz@5TSuJ$q3r{_ycg&~q60=+gLU_HXz
zY=9Z?MLPVE033n{<UVelxQ#X63Wn^1mS&aQ2OiH>f7}?nX}67)am@ZH1ZLm~du*YJ
zJ`LdA4>xY;QA>tfb^_JV!s!NV6ffZ%q#YbLBVi^)qzE`WF;pX8T944CB|E=hxqS11
zM&#2QZY6^k+z8ewgACrgWqaLz|2b9rompurkvcP5d}KR2b(~7*u>M%F<&gdIa@g(u
z@*Go#;8QCiKP+u$)_JD=4;sD*zKwm&U-OSb=O_=_uRiy<r{anxxzwCgk+rgXMN_nx
zwYDcl9-hs4<KYnE3BzxIA?_mfIR5}@RxQH0Qx4EO28z}3Wk%IKmi%XJf5NhQo8bgF
z%X?*9ZA<0V_aIyD7zz$P7}N5tpt89Yn8<?Q%yf8OpQtMN$?6XKjZg5WBP<0+{Q{jU
zqD;4>1!goMd(%WuOK_I9cd<Ty_`O+NXuh5duC>>#XrCl?NS+b9VEW__TF?qpmy%8M
za0qJM7N6esb!+miv?{TYr}Et+Q-#$f0`d1$bynO2IxmM*Rn#TDAdG$xDAv86{<=r|
z^->Fpp5^7}U=k9~|7CP(w?vg#WV)_pN7cMV?P&8|y}cR@%#>6Y5I1R77(NkKgrh*7
zHw`8kVJb%il0ebCc!Ns-R`3JJTNEroCdr}c!6Ch{ys(Q5U}rzMvN3$xF>edXBLU(9
ziT8Ln_i8UDdT<gijAYmQ8IH*3A+j$F1tcc_&>VLO8Z8Wl+|>jzfnT-iq&sVW1cU^e
z*zcBaI<h269=45)$9U*Om29b*@Ht|6zO4I%WXN~G+^hOhE@zAm3@Uhbrqvf<AMWU=
z7tv9BGDZD|`8wpeUd%3+PX=dN3odkSi&dO8ia9qEnOC}RY*=5<`Ed#FrOWznUkkp(
z`!aMc@{YZfJUuuuRo;E^s@+k!)b>N-KS#|C2u}i@0Cj<ILZqtIu_GeyKke)eIIl3u
z3H{wiP50;%N)Os9SD9;;Kaa4_s}9eD>K@dT84IQg>ik!s=?D=u+jDA+kz2aWD4`F{
zywr1o7z`xScBD|z=Iw7G``s+2UO@xmwWwQ-LNVV0Ac1)u2<O*}Lxq+>?U>b(^v@@=
zmcnVIQP`oS<*0JXFJE8|wi#;@<hc6X;KhXyA<K+w&!b=N@eNLEF$?8bb+EkNIAh^>
zVKO$H$@Y!a@sjlokD2z`y3*+`Eo+lYcXCU*QIR556FWb=u+MSw#lEs99dC78ChmTh
ztQ&8^4hi4CmXcICup&E}WhOIYrl+sTl*U&aJEJd|^utR-n<4PzCCmSEnDzpfy?bx8
z0^fuL!TK$ZGLk;xey`zy@zVA%9_K<dWAWrr_CKz#2etAZ|8$tw^U1u}uARS&T<}4=
z#&md^`KL^3$s(bnGchbn<3&v0%qMWcfcFD(2yho4GFn0v@=h<_eFti=V-Om_wo<Pt
z5Q=#N@{ds&EjEB>ZWIkZ++HLuk$yY%$P=Phr<|j~*+&XSfe@1xtLFuuvN>Hi*`f6L
zz7_ZERld90hhN`TbehGS1HoKi0RglCfQ5k@s)G5^>|?^cy?NKd`<NI1S(dDr-+Ox0
zL}=LU`oPHI!`tD(X{S41tXKUJ{7rhg+wrqt$aQAV;SN*ja?8Ee?-UG*UMbRM`X~>+
zq-!*7R&J;m_qz8~A|@xM-nBxn{fg23iEUe{K|(%h?SeFl8hP9=cK>9s;_%%^gF>-0
z>u8DAr&O!S7mlAxytm)tF8!ED{zl6-x>1^d8B5)HK%}l#BTr529gkoM@pKG$SjSX@
z%8&`hV!sU(u5u}+B*0*o+d%ndi#x^s-bV8@+~CT!kdxq7gttbto}yv32^t7PFc^jM
zJBC}Z@==)})s!f9#+?E^*B?y~C|L%yBAHD-d-}uyqu6$iufy(Zso`%y4g<kE%e;ns
zW2d<<@xTAKQ7zwMds6k~Z9VTXhX(8?4HkHl1LdFle^PauF3+|Q4Liynzhavj7B#-W
zpDeYgIlenvHOgFnTWn`qb;(LNrMHy5r)1^6l3T1&+|48DuJgyAj7_QsyQgQ6+q<55
zoefFm&&;4dcNrA_SX#z6@GwZhx5_7HTEM5QW8t6*o8R{GV~RUt_+su9?@4|3!?cl+
zq*ov<C&9PQy)EMOvwGo#*(-tW^D9P8QXWz8{NR_sF(EAsG^pv3B<w)^uILA>B#&5R
zbeOiqeLo)VkkkUF{S$(15uoUvaE2dKn1L-(#RC8gQJfI@2@CJLeP$mtJO_T=`!a+S
zg&&KaaJT*?WJMdf^VH$~{xa^l{Wns)W_Q6HEG?lr-aTQC7Y!J751f4ewL~tY^utyw
zH=#`7)1CB0o`wLI#Fg8>Pg(Xv4)9~CLLXg96sz7DtO(19Cmr6L)S_^keVD!|^W|f3
zaq2@Sc`V&cTY=!DD0<9%!B1Y?>|VpS(REGT(1){Es_h_S;t5@P&Iqg_stP<WaS2JS
z{ke@0#LPNSAq|q6nvx3#FVY?5pr@eulLdd52?~Ie4tu*yxq|NRSaXkPM=oIptIGJ7
zDJB*fub%><K5AZXteo%J3N6)To6t%|?Dg5JTMXkm%Z~#+9wuDD;Qlh7tkiO!^5}Zx
zy0DK4;7@izmMs2dUNAvkFCT6#S*%;YWKyj91LAW+cs=Yl<hmVY2zNn=7Q0Ls@El1k
zs{&~Ch174R)@gj7jR*L=u9rYetN=VmFKk)i*S&r0Qq!0qgh`>)PiIa*zTDjf(|+Sx
z6Bm%r0~8pcmEWaa&B%>)KHNL_7nP~InOm^%$B7H4sD+UH0k)ih1){<?WV$!ffu>g>
zG}=O8dhpLHM}oqv#rVJc=i(uk0Z@<K+aRAwg<RatcsQBT5DR0L;gEv(g#v$6h6wEk
z>?#XHQ`OO`0{BJoW8j1k@&buSpr3=xBPH%0!r2!p3O7GE;cpm-T`*A~<So<Ic-t}!
zQh<<D6pldTBW?TtOC*3^%&uMt?w>>dTM*y0;h-L-W!|HmL4{hmx*ci5-^HJXYZ0<G
zKgE253<T^_V18HJSx~2%TV%#p7evAvXuSYJd0Of?ki~rewkT7a%()RtDO1%f`T*y+
znNokgA)|Bg4d}cGj6sqxQ~LMLOn4ZGJbCRYgmZ@(<Ji@|D2sF5Fyoo0h8+Mz{=+Vl
zh1sULB&ija65}CYeufEh;!eB{ZN(ZLN0p##4xJWnlBxBPy!MG0pdADbr@XMuDY=&r
zaqcK2C;nf-osuYq=)7H(DKc`~obpQ~H%swG{;{VR%OJ(^RblA48%FY~*)r7*1>n~q
zkc5a_qj2D+?PR&+)o{kAJ55#%NUPuPpHMG@d3XUmmZa^_Z8PheSRTm0ny<xMVb2@|
z97X0v27QGK6ADc5kfq*Gl24Rqj|Xl9UI%Irlgknm{z3sld&C{;=7w-_ygu)k3WN+=
z0q(KY?RhSvI}6H>&vz<{wW_w5jg-CyvBzL}z4FYxOu&{2K8jis6Shn?YjelyGA6Nd
z2>_n@ww*RU?+qKpSG7SdOAHigAqqy)Nx$;Z%Y6l*7zS7FT>$9`0OT=8C~vvIe}2c<
z(I_n=2Ty~lHfX&;=*lb|unIud+X`A6?bVHk#f0tQOWcg*Lno;Ce&a36iqyc~rw-*&
z%fva0u?G%<w%6QuE@&BM!jo!46uNh5kJW+FM@xEc7eJgw;@OI>wDIRxy6Da;B7{1&
zEp;T-33xO=H+O~|dwZ){7Kn6Ag%Uo&`N#ZzReRYM5MxyIGE_?SoI(=v|D8`AnFtGn
zs>a3=B4W<>qvv|!3wYnaH-SD)%v#Vp+c~FsR>u7w<^mD_j2w$;Mt^YRxZoYfRC8L(
zec2PVCK5PCd6u=hfT`Ovo4q@iA+N}%a`M5;k><~rH_whhPd0WgEYF93n@aF~JT?g-
z!_x~b2GR4YabuJ`+EMHU2U_nVNU76a{Rj}g3j8OB_KUry$Qx@xj}Cz0eWTPo(f#|t
zegE`m?Xgg??Sqd3CdXizC;X4xuO1|~A52n~(xMEmY5VhNe0W@^<<4K|*Q!{y=CAf5
z=VBL*?%VQ5V6j2Ct%3CPyfsODTF_pQ7hQBOU`0i~bi7`>OV`D`tnGVYx4P}^RPV7$
zvrZF_%l<~;qifZ{8pJtZOMhP~CWO~}(lVwYK)w)?VjpAfk!<6X3P+8ANM>mAsh1Ya
zTx|oSzhRllr96vh*)}IbhWf_6s6b|*qG6|I-|-y9H!(}w;a-m0FDcM(_GHG2H|UO&
z1*XcpBoLospYNAkd($*3#j$pub7V_>qqgj(D*~w|`GV=sH60Wyul%@lw&{}7dbe_I
z^+^M#FYoxx0>Wa*#4d(!`nfu{&vKMUH#ZOZw|&3s_i*PulFlQxzMyNgXxQ}roc!;c
z&#TT`d1m}9{5P3xRp&ic^2I*r%I?XdT+XL`@iOjKB-|zWY&Kud)3>6o*Z-mOReiO*
zIhwIn-aBvcCMpOU&CF;S+sphV(^0XR0v&v7o#WKdi6e&Mlw||T_=0-EoAND&&<QRV
z6~STKzu2e81y5dd0m?~tuN4je3N!9gNocw#w4k>{2^=@5+%NZAq%G2j%UdVKWJ|K?
zko(6sB@BO^E6IYrK5x$Y3EX(D_*;<fZz(soUR@CIR9(aNFrWT0<AGCmFAJ&gZ2LXp
z@x5P6LrSGHX!{RIN$(K>mkTPp4X<M}iTli7eyBKUCRkW!TRkbKbWx(#awj(hWaaY=
zce}r_zD^MI4R|^qK?%v&ej@LI{IEXJXGv6YCH7N8<Bo^@_m)Ka&Q%D`8y_%j@4509
zrMf<ZdC_@r(P%uyfh+WJ-fmX81Mb#rm&CPsc-#;u`N$guKIKN3%wI<FHb*ud08d|#
zx%XYp=^kEFlHLQkXs9VcFfzHi3pN;&dkzrQT;|Xc5hP@VXjNnyF=pN4tQ8a{R$;5f
z@`qZ~FP2*{Zl<4Y@O~%luk+HfYgSYHk7Mmq%FN^89UY$=D#A3${q>6Ykcw<FPD6R(
zxA(#is-d9G)|tqOQU0w1&Fk^wCz29f4R=u5Qq+#hE!(&LG_pzal#S(nIeK1XRe+CZ
zXf%KNl+3i?(1cy?joQAUeF2fO6H@0+26Lx9?+fz1p;$gbNxrzB=^FIfM&YwG`m5|*
z^R%IO#+PC!0Uk-V43>!e!@I0OdzI^@!F{MLSgV85q+BPf2&rV{Sy5mlJRvp#mwy|%
zc`ZvvxBcKr`6Xs&ol@+9`im*K-)Fyu0_BtGEE5ni5QZW#eiTG(jc~$i@j?Y?`F^;9
zh^0y8l5E-DKQ4b!k6+y=dDppeh4=Qe3<I}KR^gahwXdRuFTEaa(%gQ@KT}pwcO~MH
zD+qNrf0oHM#=ba`dg<aoe7J>D%<p26vA}zixcu^uyZC6IQ!?r5@$S*s`Jhj`K8s}E
zADIoWdtbNZ$@&{f^dm*tYd;IF%!?W-E(X>V>`ugK?!TbJ7%RQ98|{(uHb7eN_~Egx
zhQFw;>12z!e#SUlSraR+lI8%V4g#@#4#e(Wl?B6U1{TE3o#v$>8}(buAm($)PsJ#B
zYF@dlZAQsuT28l08C0Ph37gaAv)_B{`nIQCaE!DF_5C+qW_*GGdjU~>0j%dG5T(c%
zgLH&Y)>QG5QG1>qYqML>gRiGv$LMMgET}!W%<6u2WOmB8WLUO5;E(LIr3|mixXrRF
zg#==jS<~!?p?^^(>D{)zSHgTU2Fs->hWg=0p646x7Ox@SV@&U2mkug@vdSJScPdq8
z#u(QTRz0n*`q#v7V~rh;+LjxVvD@*>X&vT+?$jrf6#|Q+#|e|w&Tkfo*`Ess@E*!h
zH)<Dyjg}{3SGm=6EWZ`(`rv3bbYosevR=Y-?e+ep2a2huUFw#9^-Vra#$`OdY4J^L
zRn04R>ifmEdRZ>_=;6eWO1#`nKpy*Vp&-Oqw;;`hS4_ftR0yUlDCY2Lwt#jq)d)ZS
zZ#Cp%U`%qYEg*mN9V)8cGA}N8U<0pvd!Gi9zpv(&e2_K*K1N_afm`BHU{Za#u02uD
z`d&|51#e6HmWhdL5qXx&B{^#~79%A!G1`B-gZ5nd1BP6agXMpx@VRE#RYc;XD<7xU
zPacC*0>6v*D3+&2!ZT_2RPAlOih`Z;lY4c;G~L<?a_aB=%JFG<a^@b^e3wdXQLF3$
zL%DU!`}B^G;i-p5!>jLF8OzT8$u<8Kj7hk)FTeJF$}=Z=gHg#<!GVd&k@lq9ifLEH
zZ@y7!_}u+w-=aabT=M33&bg8^y&fVXLSb#o*qgkcmZa(oZrp~JT|tu|$??@bgwbWJ
zPcY?|s9_w|LHLoEzCuMFSyt!PoV|t7rrt;9Z(6v1FsMf1JR)ORpQr`57$Aw4d?6UU
zU8Y_%L~=Y<PbGXC!PDp%P^TQ4NR9%6g&tF7ZWkE#WW&z5-oZ=n11_CRUH;)MvnE9L
z7;9?!i^^cH?-~m?=$h=?bV#36LEFRBH?jOI`-HV1@w>pSHfQ{gi_!b8ub=;z$ST(^
z2+MH(qddFsf-`WLiM32<sG&dSz69TgRMcXWk63pBf-U|K=pWSX@(B_Q6UIxlAd-Wl
zz~seO1iUOuZ-@*)D&A_ufR-Lcq@+OGsS@r84wdEB8nH+Awroyma=)$Io=^bAsQBgY
zF>kV<*gUystqbT2^{(joB{xn=X^`F$tuL?nJXC8k!v96--7Ptk>&dfcAiob!V(fO+
zoUfeEHe3l&8*6W0(FwHUKQr*mGxE#evPW^RM+PXQ1g<X697g1tkdjvt-`#PzLSxT4
zP}9!9rF+3lHrnQJ0pukqiDwZ)O%Zd*9lJwhAN&hRXKBEbXluS{+%oo4vD4jZ5dH2b
zHpuY$IYaL85J}9O;y85bK@QEROSkvSS39?X_2=7NcMNG?KXQgq_tn|ou)Hq3BcPDb
zy5>+^AYP$3x?Q3vePT&6C%JYx-AeCgQ<_EX;8$;Rqx7_a_M_*e`k!m<lNx?l&PGpO
z7TYuNbKGnDYwWRzx$L;|as9H(f~CYeFNW<iqkkJ3A8fMtUcGv=X_C)7>;{v!ig(Ab
zmB;6h#z2Xlg~JR-uJ4Bofvj~wb?i03=n8IuxMCIy1XBYYr$c<HR~4f`V2Pp<D$$a2
z2JUT5{D8~Mi?(qBI&v!p?-R_RObihi;uj%_^AAk*giwM-uLJZY`$E8-3(8pc9$Fy$
zz1UxX21v<^v@Pl_S+rkD6Xd=*uTE%?d7fA!w3vSM-4!;o*B;ark7p>p8Q<9zdqtfe
z|IsnrDla8<XDnOGl3_OKA7h-pIYT{;kQXW8-&PoIrRA?#*8bRvOW5c}!tg6uCFR<p
z_&aOKyF1gQ)y88(ZXML}bvWf2QlV=6St_z;Rg*K>)~74ij_}6${%wcz)tkSI%GBqc
zGr<V7%8ptVmq+l_!{p}#aF}};S>LD#?Wz-fziGnWcihN6D6qd};s^zy^3Q=J0<7sg
z?h_jf<sgbD2M#J&CN|LOb&>KoTtnf&|6~cbIbH=^ozd@=?=egA`?B6xVZY5ozWY&b
zz0)_iD)NU)xR!)(Pt9EV=AN^je#?kG=p}uf%KsN-hAq8p;GIc3U!7SuH4&f95L#c6
zEmxkd{&Onq+{<^~fhjgS4@*}xp2yrvJ$o}fm$sly?|AsDFk!UNzc#F<K`MssV{Dz^
zZk-r4d-85=--yQ?iy$GWYiVzG?3XzO>xIv2G<ME9dsL;Ir!cM1qVpiNGkH>lr)WJo
z<KkZFuorXk5~kcAk8X~(s+fd+qS@0iA&LsTb-{+bZmB0^)TF|C<Yx-3$6|1`eLoNY
zhgOg_F7~H^-Rv8=qk1QAqL%`Tmr9D1$2>rJ(kJLGboxGn>$sa}v#313s0my^K!c-H
z?%mOgcSG>W{O^2K{aSDjj^U%UGqdOB*C{sH#1j|Kg{P#go?Q&1*~ZShZOgp=WPxZR
z^`c<^RZXj#Y2xe2gqi#t4?XXB!SJ|MF5X#fo55>->aI7@E!8_Ns;5nHmC{ePXRjYu
zw=<VHhwcB&k?bEhvX2q<Zc}fH%OQN9@T+f<b2jYr5i0(>9$7F!NA#nC-upu@L(bq0
zgj%dpl9dZpL7>JONGl6~W(x4(rk9-N55JCCHhjVNf;1=bI!@Uf6iCzH(g}!&V{m@G
zT+{s^2@lJ<Rjff+E{EUEVVwkaw74J?QnN>ngr&6nW`<@!@UJ&&yHJ0H%e3&=>a$rn
zUWcbETX!vJo(SI3U1$6_<h#h}`w{ovmW(Z8XdIF`S!a4Rt?a`q_@13p3w4?c(Dn46
zEsB31V9WmJ#Ps+x)w<I)A<3tR=05t{m)@^$4zNfa?vc`1CK~AWSlFI_+d(hw5xier
zqv@~glcbAoY-64J?3(7q3n0_lw@|6I?N1x5ZYugxEzz0lfXn0f-Zl9@P{3PW>^vd`
z2A8e^@Ofgs^QwE7s}auUInzL-@oFQhQOx=c;)!|z&w1_RKmfOJfRZJ49+csaKze7s
zr_ca|e4*;7$ZcU^0crnSZDR7_I@S|2^h^0E&D-AGW9M}=2UE3Nx@{%&G5XDl#UJyf
zM6+_#Ej3<Wx1Y2Oa5$iD`0DpZ=EcI+Aj^==Nqf}Jvq1N7|8(z*>)JLU+a_-1$4#c>
z`^m0VXAiqt#i|wd->i+s?R%Zs#UJn>rg@-z4XrRdwwHr{J(|84?yM8ubx5IT+)}Y2
zg7~#y$4!p!R|R<>$>neXt?(gK5^7w5ptr!oH@33mL!mUwK*udy_D?Z=k;4A^E~uK`
zDrWr%mX|~`4KTAnMS;PHNmxYi%cFaty9JJ2tw1P&g}F^<2zO>U`0?>M?+u;bcQ?#w
zyFtSJdLy)Sj#jqIQ+e-82WOQq=jvIjye}Z4OfOdJCJnTiaX;>FPLcr#wdZc3*mne`
zWV~Hq{@Y__n{h6pzKimYo1ASiV;Vd5+yjbHxP}?o07aF9Q>}X9xD{j4R6(QX9we4#
z!+4ifX>8DHNrIFof{%jwTnw~6M9rIu+@0v*B6?qB%A>zRJK;M7FlQR>%nH<;uc<F>
z8nU8UJ-+L@e>t6dtTsGl{M1;OY*FxLhm!?X#Ok8<!n~Cu57lj?HR_%nm#WtfxTtaM
z^1G&l`-Li+_IBoBEQzX-@e5J|?;8`wJ+Li~$`5zCOP$_(JE$f<nVX3R5Z=5{UZz3n
zg%DBa4Y^f=Z$@Y86s__-q!5m0sE9g3WesdqW`yswYLNsuMAK~m^EDnQ?QsS>MLyiS
zY4ZeFu;F55%u#u(P&#MVEN>{qU;Y)pz5B;K)S2=l<*o1DA9uP$3`;yzJTm4JyC~C9
zC#N(jAo*o*sFZK<-OJWfc8|Y@_+u&A2^Vp-(8gqY<f^zdwT0~|9)i|C_sTXBBrJp;
zWQFC~=;H{hk9U?!DUMzCe8~Ul>T-?IZV3}Z+U~1c!aBtO=a|f8xh4M4506UF=E&Up
z6I2Xu6(eoJrXm0W1a}XDnQ08jo?7Z5yU3UiXSasB0|~^k;{d1o*vrU`)WgxrC!#K?
zCY{jjv4o5y5F{%j)Z)j3J<-MEa_<-W9<P78*xL}RxDxHD?}k0Ipc7Fce3srp*I-y}
z@_zobls#W`X~s^hqWW-yJac(_*0Vn*Gm}vyrMmYwTNmR^{oW=ys(GY}pFLd>l{cSP
zBwsll{(R55uSG64aRu2|#FLWZ75_AzV+ZM&j^}?=YLMt|EER^a#sFnH7wmYz_=}kL
zA@IXIZbX*d0jqpMxhF>?)og@nBGF=mXF}grpI$4(cy_{;nf<@j2A&2kZ}|#>L<rW!
zEielG2Fg{bz-1Vu43R`FFc?C^@HardGpLuK%o;3~013$4=vOp`RP3Uy^stO^s`o@2
zV>wv$_P~iA#av$JTCM=qg7~)m0&Gvcq`=RMr1j@-#f(yHK5n&X$fPy=u)HbW<Kory
zWM~mYC~CI%osG;7q3wy13ff|v??aW&OD})RGge*ZSmI-)-`+-R6-<7@;J;0XDeI05
zZOC#;2^KZfUfwdk*b5+=-)o2aaF=4%5`onSx+~$P1{24-dqLUMF>;)BQ*z?i?6-e`
zP}Lx7lmq$mT&~x$9Mu|v?vF*7f+awY=0j6Owadx+KY(2gIRjYwqZ>7CQGGxuGI&**
zXNuzv6gpgx6_9KW3f>hP5>Ncnw)8pv#lQpcr=ENyW3!TTKku&i-aB7gb}hDhcrw~e
z?G8P-Z+@To+}HIS?JT2Voit|1uZP7S9vr^sv`a~4;KfA$-N<_mkMCam@_zfyd+ywV
z9HIPkr&G~Ozr~=>6|`BQqyd?;)=Tf}ouG+5D<wwxW2?}O`zJ^@UWN!6)N6>e1&g)k
zU{*6;mkQx|__|FUj)cU^6v2I5Bz5Yc$@hb^z;O&ujip|di@`n40uVaDdn7;t?!R7;
zjgUy>1Mlf%RlN!kpPTLl@>S=i_xT!i+RvnFxpJFyID`{36pKZNd2Po0wtbLg?y5Ht
zuU@~|jX%-*7xl|wQONwC6A^v;mSjK$GgBx=;tRMhetBaY(jl}@9=$A-CkrTEf$ggk
z9+p%fwv>aSiy`{X!3Bdh(Vhy!>w@A*U+gV@-IwZx{Z(!a`zshEXU(BWVC_vYGb!Va
z@#y_env?qTX4CTvoz#V`S3GZM+^W2udotkM`%fd?`Ds3(RU!urlBHb@JNBhdn%8)$
zbY}V<B_t#q>xujpON3^qn560WIu#HG2Ox4Lf{e{To*sHo5WXOaeY~^Nx4aJ|#3w+r
zj|L~;VL{%YPvn`^=JNc>qsmR0ib+B_q6ldRBDW>c1FfceOD_4w$a5LLdw{l&sbq;1
zmlqX=3I0Win*b9fucO5Ygb&^au*4(;7pBp14^pQHWD0n=-0!dLm|%bwK~%w|AC8hU
zkFY?hlE)>FUK2Y}uwU*T1e=hN&kY1vE7CP1_I(Fqy?H6ggGar#QCw(sT{vSrgo;(r
zgquw0o5o9+LV#gc4J7AA?koPW(x*I#o`udJ6$b<|i~a@m87Rlvg$C9`1pzW#<-oTY
zyI(!Z!Kp!zC%FQit^$h*1bAARH8YR^6j)=Wg)bu>tz9=`TV}`J&*?1xTp*%wO1MVb
zH`g+Mn;J&JF0$6<F^lt(V}X?NxOaY*4R^I$^|OQvlAAqe=$J$!lS0j;PNUm?(dLaw
z52dOcwr}ywPpT-qpLEIO`|J2>`OMn>Lnnumu0L)|)Eq1(kJ*Uph}Z?u<0cDrcf@|J
z9b!tw?sM$aPdpfCwPjIqFd^VMa)N;E;kypu?7o{r1r55LqH&k<A{{~SJO=pfmyoRd
z9ITmTsXDnR2V@R)0Ap_wbjGngBf)geFkWUz4rxs#Knd%=yk7Dtq!$UHJU4o@@)sp`
zGv2irnEm40CZ%+}x(Y3XH6M(g_jbxC+UF4_@$|K@s&yGtm8V22gKf!v{5CyPOEMu^
z<exp5T1$SzuIU+zisZ5KDO&yw^Ly6SqQpa4%!UhBm=7p{+SRTk@i>uyoRbNKc$$zk
zor$F@VCsdfv;6Jjm;%kWQlaXZ0Yx^J$k%%h;Y8&Dtx8i&=sduP^i0E{ITjwv#^ckO
zahS4N_2iVMK!c%r<>)s}e1)>gRee|tIPe#N3{XJF)k0zK4Rokv8ue7|5G)XR_poG<
z_tlvHF1tKsnNQBz&chklJ7*G2j~L$Hlbx7D*{mgEm`!?i*lNIMzBoQ!SJA*XIu5Je
zP54rWBTyDenP$bIeA=hfx#%Y(H+AXA-xW$RJt@k#C9N-v2~4mbA`~(oC^>kMjDG-c
zQXqqi;t(771{$1wpbLaTp)>^}LVi$$Ag%2)$IP-)pm~!BZNQALrof&6r5$f?pCBB?
z+~F02Ho3aTjSK2ADnY~xhi5hW07lgLkoISV<+smzcX=FsI3e&11?ymKdg(P#RNWv4
z%RGEgZauy#lL$cV+#YjFep2YAH|}{=ApeC6j8XoT0oP-aSJ%#b9Y|N0BEyLYeCCdv
z?bq7{@+5{^y5@_SlO|RPj!<!X8hAEm55@kjWwojAp@_ezy;JkL50J<vSj6T(iIAYZ
z-SQyj6-okNEVSbBI*faQeoZ}V{`y;?80ZYaaAAZFfX-Z^Xg!cUZ5w|`V-)KhZn=b^
zS<y4);N~f0!qp3a+$K<u=4<=68an<}L3$L3$G^}(djcwpZ=e{Z2#?*(G`<ZNwEyyI
zkmZ?Yp`oBZ`wcouCUYmB{ZvI8jYDfBUM@L4!?sKvd>U1&+JZ6ZCaqyw%Y6B7!dLUu
zmig0Bpf7udUq9opX+R0Enx3LwFkIQ&28nB{#0^4hYz!_DCTHVmhopgxfV1dy1=qG+
zD!)sDmB*X`PzOKFOLIJuFjfju3L90~$qqgBsH+o-c^B&5wW>~1{&Y_>6ltAS$~Dfy
z_=%(K6vO0;ou?jJNo5pHnz5j|)xz7gb54vAnmbfglo55wG!TuYBALUucY*bdDvq17
zN0)P-4cb5%*?21)bwCkiybJ2Kw~8SH7Kpgdd7y<r{3cxELSPGn7zaW;^)^00Yh>Pf
z#E)27dWANbVEK>AB~|HDMT$A1juyRF(Q?04qrRQCeF(+WYfl5ef<9qt@AXfQCao9?
zmQTLjg9ew)`P=7YyxJ1=a|r==ri(1my3M}~DG}XHX%a15sSw9;%XH8w0Ahtov>^qw
zY;J+52gsE;rOjIoOI)iz+3jNzn|8kFMn#~G<xywoMmrU~T}<qPyTzMz#b?1#aZ$ZB
z0V&TiLChv3iu$0XthLO<zMUwxI04;ip}u|7L4sr<JSIs>woze^)w(ZP@CyhLI6;f9
zvzuiYg*kCGAs-59x8>3OdSgj4Sx7twN<2t_CzNGrZ&iU6({AJ&vruJ8orBW}Cic6l
zCnsNlP>ybaP!eeVK{toUM<6?wH|8Nh-5B^@m_K+m)_o@h=b*yQb9FB$sd1Z=qS-D>
z2HYPCJ42{Y0$|w0*9(Y2Ro~&JP?r4zrytN>^iSW95~0o&p+{08&pNJMxkx?(1g>5E
zSswep?9|XQO)aTi0`-wJxFdcmxv2QnwqV`$-3)#`%wwTp)AD@~QAkA)fnq`q)<A}2
zK7Leh&P0M?`&KL%B5U5bUvTR-ss<JZBr~I=Vo97!JYCrrDs$u7=?jzrXu{`x62N&C
zn*W;gd_zQ6b613EE`ax-v=`t%nS$k#^60V?Y2jw9z$?a6-@les*QomxAH7hfellxT
z(I+RewB%E3L0Hyqz<NG5ibpypK$qQ+0v)MXh-PzdB-jB_c^wuSy<FVW5h7o39?(*h
zxwnN2f&hcuA?QkzXc_ChI{?telss-kEw~&y)7=t!ciOq_$wKUo0U<imz?cb0<Jx?+
zn+u{CUAJKQ=EbaJkItqhP9Hv*6(aC5ZlpSI&vnKZDaEgLO73m=0+wpGWuesDn{I**
z6C&u`z{T5d%Rf2IKh>YrRLqnLE4TEst@Rg}QYgRF`ir@RK2wxt6<L+%rF#?Y_I~0)
zh4pKnWMQr@GF8v;&h<38ACGhPwFwQCjg24k#&xB6PfXV6N(YBiI+D~g)^N`{O%*2O
za%^|IsdP34wm_HA9BV%|$9Ak<1Zdj+Xzq(yy1zXQgLDx{$t@Vhbz5Ls02~3uF_1)t
zwzv;^k=i{<XzSi+Zq>IFO4ulCRzgYA!6mVmKO(jQ*vmT@Ht07hZ*{RoRIP@<9tg@8
z;QA+&BlxgxSy!(#PCCJob;SRQiWDv<r@Ah>C5lOWm|OB-bDRCd@-})OefZ%d`oPE`
zv2!{xMY0ZmOyn12vO8bZ1fTp8aiz#{qQx-%hv(KKpASzb+Fn+!_d55bb<btBo12SY
zE(^T~5EPf+Icc&bo3c`A^)4ZPXH84N*jn71Os`VL`AHiu!v)?~f`3f1aq*H}Br2^{
z<@6^z_xrh1Iyn`+Ije@cf3l^NC4<BtKFbP=t%x&@k-|#f13vbi3wWm9qr>JhUO@Qw
z^UTR|hcI9HeR(@}jYrvjNSh57g3w6<`?>U|#m{_8ILDBH%mMJCReG&KOmx9nr~z=j
z2eQ#h>Yia7p9!>J#H?@4c#wPD*WEG*^3#y^&!vBDZn?<4B5k099hkRO!#jWR9Z0w+
zCNwWZt+FbcHDC-!2?gL70`9o!n@$Y{)`x>k8Bd>|TM0{S;rS6H<V(9z=xjT%IAZ*`
zGo|DTR^Mo%XRt7r%Y-=?XOyEOa=vOsr}|Y`!KZ<|{$i_vx+eKbty2-wsh#tyP2F4y
zdAoA53#hnBf?=lo#J5n4x?67Cc7?&9ZA!zng{H3>_XeP9bi+K8@tP|=qx|`&9zQX%
z&bfm5U2x=tn#6Rl!96VVQ`nj7>+h3ghjgk*A<4bc-Nk7O0#_qq*!_+M--4?Kj^=~c
z6<Scw`e)-1`#1jU5k$sF<jcJ%K4cI@p5v#Lsb?$N;m=!^0s7Ij)8Q4cA<QLS2%}kd
zc6F)VIc-+tXCf~wrM5ZqV3I^jU}(KVn260G<8D*FC40VcpRqxC8XqypTM{L5FO=_X
z#=$A8-Ge8O{}3s7(IvS)f5t%Y;whQjociWk>qk<AoSA~e2>))Y7eQvj9W4K4<98*X
zt@h}^?<J+vO#-pJ9=(1c%)h8QflU)>B++qy{qy%}5<`OK#HP)GcOz?AMq_qIqa_>Y
z6c6=uiF4fzq%*GLuEmEQ27gO+4N}z!@UPx&Zc^i;t`>Wuwnjf$!RQcH<Z+tHK&p=A
zju{6b`AppeOV`XoT$Pvgl9cjFzu&h?8HZ?9?i<e{Ce}qx9v;d4?1qI(#F?*(!_ZF-
zlI6V2iC|xZJ>@GDjD`>XM5;nDLJk0lV}*-B%UYqTWmm#8;2@lc-4YL^;}{IbeN#=P
zY)pkbpM`nu-2}6Nf0Ml+1rvqvuED7qSl@Jg={o?mRBTHEV_27rI|OQ4Wj^gP{UWo6
z_VVicb`?{X^44>rr{2l6`RsRI%`3rc)MlDEX}t>a%1~6Y3(4M6cCJ1{dH#v5>*Kt>
z6!%x@N&6qmyVY#<eJS2FqHCBPCfl^*)9Lg5K-f}tXJD})!b@Yb^837jIGdyA3?+#P
z8DG$W&>qe3V{hBCu$Lp@a)xJt=CG?{r!n@2@A|y8pRv-yw9f?j-&Y;`R)~jg<QT7d
z_m{sslUbD|d{k{ufV}7x1&hWVWsHaQ;`*6|!}~@9J9;m_Ykr^iID($C{_)L{&T4iD
z<?<-sk{!+7TG4#@J{KncTY;eb^hGV-CHz>qAjScyaS$|Gw6jdZm0%d<^#CGm8Bh^4
z@$O>hVMRi)<>MIWlHxs?gsI?bfwR}cfQi|MHgqs;^hCddSTgN(CN}CH<uN0Q@&}>$
zt+m^p`eV7$FZjPZFA~=66dHF<-t@dbhT_$4)cG`9&`WZqV}R1N>nVTtj8qCyDg;)g
zWJdWPH4(yxNuS8wV^7pyCT;IC96NlXP`7^ivSjnczK|b9;RzzE^X9}iE>5%B9;YX^
z-Vc$C-q)u2;fD6?sd3-a6X_a?mxV>Ov1SrXi{X}b#fsXMUD-*+3H{A82S-@dvk4Wx
zx=Ee;Q}ar8u76pjIsc^Aw*O1u3&8xGYYQ_k94{k^cSar3Ud@p@qoTShANrJ6{P`a0
zc$`JOI3u}qZ2E<t?FFi=Y%6(Qf^Y(8_@F{~RWjEMG=2C!_BLq)WM%?$3CSUS^mrC~
zDM!}=ws|9;j(38kn*`43dgX5Y!p>*+iZlyP`KLUx@Vcy|{J~V@I06y`W_;T+mEg_!
z9^Bpl2ZOdDp}cD=*?HI6UQ|4eCVEXfipeD%D%CoxuNFKQye-o}!1syVq(CAwJI_bf
zyYqcNWv}N-qsN~cP0>V8R{?@}NcMwn(+i2omU%?u30+>3k@2(NcPNJ+w@BBP7~7i6
z6hEUGw{4-yBC_hn{M|{}BDIs{It!TRZr5Xy?i-EoK7ZncnX%om=$*d`_gwZeZx{7U
z?E8ya#<-P?GI*N4V$>ffh>#k-mX6)J=vM6{abBTz``6L@<I<Tu4a2EpFT@j0>)#IF
z+V-iX%DiKz%dU#rcWuJ{^1Jfo^*5)RR=6mQ^ej1!K_@wvvcmG|I<8sGv3n1rdlBIQ
zMEgDd66AX337Tap5fH9H5a_|J<)Nfe1*z&IlsdzBWx|4%YL1^g$#R;;0B#afNkZp%
zJ1rk0UYcSOm0FU`ml_Ctkf2hhr?SATt~q7@!pnDeIY?`m!`(;>s)SOhNO=pmjKZO%
zM#YNN9OHMR+@`^KPu-gI`g{+wP&ZfGw~D5G)llxXcANmq=f_}0oY8aHW0{ON9Vzp_
zsC7wu8!L+kvzo4-xAW!?FHkIV7o~E#kKO9wv~W|k=qwHo79M}~Dvp1*@UA|^&NQ^e
zSm1P$bFez+wwMRiF8Kpl!YoyW-uu>XR^hQ#g-46OE|)zDRxuyU4FoHpII-jmjV2II
zK{G1?7Z1C&LKISt9FuYBXmtQQxb|yv<iaij`drbS_|)-ZdtONHE#F*`HQ^|9-NiJ}
zXQEx@n~k<u3>CN~%;jEx*6Q)vawFeVOtA&71(4HowD^4)`jnZFESv{5R#<ni@IZY8
zLbx+4kjszu7!TgN$IHx8;&tvn%v;sg+a9)<_ZK(`CziLXj9%8f)e&AK{I<P(S$L=Z
z(ZW5vGZX#d7~O4(TUW|Wv_YW!ay^foUuf2TvuGoMne$`Y_R+ISemvRv=*+&tX3L;D
zBfHQcU7q=yw)c~L{B<`K8C}_%UAfnmcFNM}J#aNN#|}-D#}z4CKy)5|tKI{2ML~0i
zijb4eB0L3I@)F7wfiePHdLf2OjugLt(`}RDrh)nZ)7stz#joaoZj*58%_`waj1xJ^
zTlBq^ghZxj`-X>2jq`qH&Xks`rdMjygpy5{pg<O_1bYs8e>Q?w0Ba&dUo4UQ-@3;8
zeEXO5*B)(27Hhy=^wfLsb~;!=*#9Hjf8g!#`>&pLPp5qziXLjeUjJD6H&@@GkmT*{
z=Su|V!>h#Hb<ExN3di}@iVqlN<Zn%=h>XOBXjw%%Uk=8oOrIgnS(?mPHv2F}!;}Kq
z6-44Y&@f9JVkU{#fP^LR^79DA^l>AQ-AurP(ij-+mLOk%=t}KnY({7#0Iny~=5-;g
zdgRzuLCF)zKwL5j_5LW4l0eo(wY6`GuZcayQ?v*!JO$dj(-ijgy(vwk+w6(&Ap!<t
z<ARa%4>0sN0e1d(Pt>i#U?L;cW3G@@WbSa@X_@icTw)?5?@q+!llKJ|q$KPx^3#Sg
zJXNnV_g<Mu$r8jFAAG4lk(gqqAmSE#Z>l$e5Hg}DucI$rUH(y2t!@#rp`}3w17BMz
zj3@W3U4GYiY@8q_a~-I2Ujz6IaMi&2*R-YUW$Jh69YqK-5c&x!zpPijRe^G(evDg3
z-PK&hUeGU4|MyFf;KgC8wOpR|#JxuI>gfQv>yH9jHn>qwx#T`CKr?qn0ioqh%foaD
zO+{0^ePU^uNOiAO<ixf;9-Hs1ZGlwf@#eKczfj+PQ^|~r+AESPFRYZxLTIYBrG)fs
z+uMdcqp}Spg4beaEk>)NhTfN22)tg6w9meZ*3WJgIsV7T(u|Qsov#jm+n9X+9{D|2
z_Kow$m0KuFOp1MbPpb~lVX+rfc;kM8^M6~gWdqWVvv~tK%OmlSd<V>IrMh+ExiDEv
z(A?@!`dT<<=)lu-0d!ZR@csLs1G&#Yax5`s4yut9rp?P+ATb5`2TLRFI+{#^wHfcM
zuEHAN$r+L~gIfzskg=)oUvW|>GQuGQ1i(Foi_ouID^cw*Yg>_7G4nW?dAD8;%dY>6
zqMYh+T&*|4e7;_{7^E|Gf={B`>PO|1LCimBfl<$$pUcoS4y)}3=#nYF(?!kianEj;
zwc)<`-hUTq+_RREXkDI?@L$FAs-&XgS0_bo7Y&2}ZW%{`9w`6-8@=2-4s5@nzHYp~
ziBalV6f9JWL{cY1j3NVR$fXGis5y|cKV2?T49L6WpU@zb^@`LRyTwgn*-E$h7DHyA
zJJa~!B3wAChORz$_+e=MG<<UtCABHgMX0PLV<6b>q{Ex=Y(>U{SiXt6Ad?=$m2&}p
z#2r_>Pj&}{34ItinlRjwr%^Z`F|P4j5|e)wW*-2{*7awR6$KehaBk2&Sv3h)Rc(*W
zfZ{9<4i+hZCQ=~B$lnjdBPPnxFMJ&*t-1}X%}uy)v=gLDSul^xNk{@jK`D$i(NQ>T
zS3e?AjP3`LCg?^WxToD}+*=Mf!nK<I;q^nmxyE{9N{@z^y=aVKos(0as2DFFzMb8$
z&Fz(+6OY<atYW1|o4RGkQu1%+nMT>tL)Y)#YbWKC!czm+b2H|wZWE)jo^uYIlKIp2
zbMuph9gjHc>n1|tmvsN4WY-(H`$2&8D0lE);C-AZKw1@`v*Ub$P?ob5Hubrod(=<G
zT(f~W%*@?qNSCbR<Wap@bHW+O>v>ClGM54Nv>SC?y;0i1qNR-E@C$k^Pv@mg6*JG;
z>~)3GKkSu0YNKfJw;aBo-*>4%asiNguxNc_`gpd)R|C*5u*V@}*~-?EcQR4yB}-Z@
zpyFEou)Nd3R<k4T-5u^vm6<Wik7CC~+&^PfE#Jk_<o6|}`sNh<ct)2wSYe-$y3_8P
zk*QD!pM;gC@>;~oJVtV(?Z<#i&@}nfAP{TEKq3JWRlD7>My73ZL3R%4F$2sP^J$M)
z-8YIE6y2di{V8wgjS8-->Oe?iBcmpG0WuP=p|s5t>g*R_!7J_GhDjK6coyU4+{owJ
zZv!bisLVkI!ijf-N5!VAt82y6Xfu-=7M-uPx|H2SzdmW}%Bw?1m-*HY{jo(9gejDS
zl5)rY>SM+8%7)4nUlbNqW^)Au#nU9mA4SIl4V{B&leRsGW2O!-D&|A!Okrul{um$w
zc3dnG`Leon0kUub?KA~w3n@jKb)42j>mezmbtF^ILcLDaNo2`J$LX6TWY~-sCvbJ3
zNcufES0S%%QkA-YzZLxkCDQCgp=@0!(#v8|wC6Aa!oT(#mw6&}B=_Qv!e`Y6E=J>_
zx#OvrOQyb{stvo-VirN}XYO4(4jifb<#|n(e&65{0bwg(JS>`D!)%lD8W;?Q%yX9$
z6p?m*`8t<8H~L3;<>vUJ;w`1lr}I~<Wd!tJ{@cBgK8LRy2QBH^{O^f9nqh+mxdiPc
zK{!=mE{uS5$L$~-#Dp`78(=70Ad>O<uY)oQDp_ztpc%+o%$|ZZ1k_MgEG<*u=7A1i
z<YZlXsB+xH(wwJ0V4MYjP=pNzsQ{tx0yT&>08s$R5m&Mw7;opiE4>BH9YkIpRUDpE
zF!23gfQS<oc7*^3Nuaq!Ke`A<v#HxCTtIt-Op!cj2+3L@$w|m)b7C-OfA{Z$4vuAL
zBL~zyqSyC#2*tmN1I3BjghHW&rkbeHXYtD9Y83I08Cp(=D2GwgjyeEYLewU5CEioP
z_%G@&>JNQcXNZK$#_Qm*Y$Baa!j`17Bj{}$MCHqU<RUg1$0pPJ;Qf%q#inAT;Qx=)
zNJsyoR&j&MPz%hoq~S}lEh@;B-B>C^t@6-rDoUVo2rq|1$))<CqWsbsG;2IDo$W^#
zAz$-*7qa%Qh0Gx*ec*8d`c~8dR4O(F#UUr+%|#<$8&wf6`_CpkZV1dlz7_r*n~bHW
zk0L)9M|(>9Q2*bmjh};6DTrz5qrXT;YX*7PphvNqBg!W0IFVbO{wVwLu>C5JdsiN`
zqvKUL=q$G35c!GHlj=-5UI~n%4tiH#t0X7+(QT}`q^z}5eVz$9tRKg+Ic;6Pd21*5
zWzCP%o10;r<_KUD4jBge|J%eyY#PIRX5v@`Z1cuQv6ZNJlrX%@j*3^J|3$5kPSseW
z2G#zeYHN-wB`IR*L;RHtC2A*kB?ATpn?9&iO&=n`rxLSCII0f=FV4VHT}`kN>uhp`
zxHY+w;fv)&DTFl*wVJ~iEU*J=<ZM5n{!CO}sR652$vQAn@O@wv8D2_7^bM;{^6ebN
zvA~2^$o#Qk(V;kzcfl-YMEpfD6zSHSc=F6+9o7NyXmZIQk;;bGPGLM5#**4YruS0G
zC4FQ%3y-D23rr<ruymbPvQD)Y?IdT{Nd>kX?l_DDeArYoJE9Yn@J;{fNxB&n_5Z-&
z4<X<5Kch%s6qpX;GU%<00K_OZk@i2MkfE~HaV4)tOiewcTf_GUEa9Ao2->&hXfO`s
zbUd+=ZLOl?;upnD?^f}%l_Tj;O>is{Jv$tajYIp&IkApwzfg^`#vYfxjQ75Vvc{bY
zuf)UH%v|H@nKb@R5`Mn1reGL`@O1b(!p2D9m5Sj#8&++xw2hcUtnx#w;z~iGQs8X|
zcw7V6VN+A8bZZp)Y(Rz1kv>upjf|xosvd=zIuM0dDiX$*93o(|$kl9;&JejMn_M(B
zM`-?*se&SAtdeWmY>8CDoD&0&isDBJE2SJDz+lB;*`?V*D%vb$q_W`)NtOQ@1rLgs
zFRodIzsK_GxkmgV!FD9V7!D#PA=85A>egWgOfB!!Ie<c|20YDZX9vLa;^xRDR6-FA
zuahGAXq8OQCKR#pICBm<l}L55Y|Ej)XOq%l*c-TnM_XO7=^mG<116|QPgDwjHPw$D
zv5-!~Hjo)ytOb}QWW(=9z6^$mm<|>x)vRue(k2!=VD9V`eGu%Z&g7zQn3RH-V<RRt
ztf5ex){5jO*>727>=0rMy$`RQO@>zqi%{K1)?wl4@Uhv%6p^zVTh>~zSy=mB#2-AC
zQ(hDv4Ta+3<dD0@o6c6lkt<4EIOG!fS6Pc23x|wDzIDSUa0hs$GMKt`hMWk?l?>Yi
zOk$Br;2nrb$ohr75HJOAL2Ob9Xso$FBWqKUwOpO|=fA1araAnflC=M#$R%u;Ogx=N
z#tjkZRDu$CkH0ADfhUu47{ByqCOpm7axfm8sLCO06bg+M@~eDy$Wsf(X=A7u|8Evd
z6<&U-d3BwEx(P!KR@DrG11n)6b0=(zSk+0zM@)fBK*8D~!;*2X`lizd-wZQ0Ix~RB
zu}RI%1T2fNuacnC=T_Xz(1w9tAw+QyaO><D_=n_3xlO)uDJY!~G*>FN(i)A$^Kwzq
zJou?qdRhA1C}NT~vclm7v*!o1hn@mt_zGI?veSPf#Q>X%V4T1vydU`TE-GH*4}7#7
zmTE#$d(th@JcvcSk;hBeD?|7q_%j<%@77-f2M6=XLN>OoS`iA5e9IKeW2#mBxEg2o
zNrh`@u!QTHis&`84t00l)L#_7`VAL#oxRYvVGq2Y3?@<8)_i1SIn8n?&$)oX$n=3x
z`2RNwIbdGX@ZiKNBfzKwc%`SWtiZrObT<6ERrrs+n@x%UXUih2mt_QG<83e430vcF
zSup8%9XN}q2Ub}$G6{>uVzXgEm~ajfu!F=rVm3aC&=6Kbs=-n#-wwi2hHP^5hFQe)
zA(U$X<bC$=8d8T~L$Je|@JitK{x?!s6dwDMif^u|jKHFV51_Gw-$p(nPf<zGrs0dS
zBZyffEWG;gxL3_#;00(z@CDUWcq%*=e992sR9N^pHYEz^=kTb5N|nbscub4){L`tn
zY7J9OBjqW%)VB;+7}rp8$mGFDk?AgA(mF|wV=bMHSQMItS6e0h&mwCq3N2?x#g~Bz
zY{VoEZ0g`BeN=Y^W)1m1*n-+rhcz}{d&t94t)e>X7i<wY#=%>FWw`gRgnfa&4JE{(
zld5*-xQdRoc{={|;A2iMAs>1V?3|f@i-g{Ga>7zYVN*D;*<g_x9R`WKLXr~%x8MqY
z5lF==u}N51-^f10!tW=j_{K(<PS`&jSlX+%e^I04s<Y%NDjQqGMprWY{-Pp=NU#I%
zYT$Jz?_&qRE0E!Fi{Od#rlL^r8Jv80u4^dNHET{dv4qvTNlHx^-D`18tC#;|?*UJ|
zVNMEa!y>T37O_YnTuz*Ai?{(8Jg_JeO!@B@rwt3Q)|d_v57n9j%N}iGy?0$_Ox3K0
zRk5|%Q|oMe2`zv@tfs0GiiAnh#)N7%zPg40!(Yw9a~`-R#ETV@(-wgP0nHVKitQ#=
z+7g@D#?+YwuECXR;pUV7vxfr*6NbEOZ2zoSD-X)?u<>Az4i(weh&^C9@{7}@53g1V
z_FPM`M#)J5cxoLu*P-`e9bW?936n!37qO|*ewD6sB<(B*+biHSNpQ5l=j&vRavs2P
z+R?#-3vob9@HmPCi}%G2V$*H0=>ry(o$1gf+aQN5BMxi|8V4h?VUlauD!z$<5+A}V
z)7cyQWMk4yHVm@nbZ6N{AjfP9GLf!-Q4l)80{DyaqvEv@=Z(w`E;ao)ct%@1w-Sq_
z<VP;*bB$nm!@;+OO1Q==M<qP&_7>u_OYaqUjN(vmB0lCkKsQmUbo8c2DiY+E!6p<K
zi;Tv9HgSSipNqJRN2L$p1Hh;u{N+`W4j8}zYZ-;YbGonsjw3!ZMHohIYcp3pdkV2>
z2E1Dzd;}i$7M%f;L3QQB>-3SH43TSAN%SG`6D*QWIss*kjmDPXJ-MQ-QLv!4zz6ZB
zP_41j#7ZY?9HS4<!GN#>Rv8vv*>I!~>8`?HQ}b#i9@e=nxtZ+)zyC#r>GXo_8%FVR
z4fQ!O)Tj(4It!0NaY<Vt{s@M+5;;I{Yy#MXy<IsNzToPF$7RFbfH@6uUHFS4_J!$W
zpM~X3g$UU%g@eo`%zsVj0M0}$C6%Dly_?>P!g5Fx_z;`+E3MjbGO_7u^g(jEi#XyT
z{|F;Cxx!Z=?h@P-1Q8j><v8SU9CA^3l=UrGPnAgcLPDYVrTeScQwxKL<H;t|hw#`e
zybimPP%~%KZ5_WthAD&5E6HF5`9K|<5oL|qR5qnOw2ne!qj+ODc;&8{RCi<R@k-Tj
zi$Lq(qrfH(^K_P7EG>lIpU&Q$?vlX@O#g!tCg9$)XTTfJtb<wLpppL#T|<+eMq*30
z0FQbs{s1b<j>?9EonIE0e%6nLj1~O=&v9PxiTG+7K?l4(MT-%^KnAW4Uk!%~_=!Y1
zneH$%k0<Yx8zkg%n)#(1ND<~h3893$i7}jPELwN@y)_u+sl&3nW_w)DFn!3SvX?&R
z)%-_zuj@>+2YK!<>Yxl8In^TmqTEuPNZ0{X5sh@$`{jTcA;pR6ikCy7W$TCV^xIhK
z5V=l`j(0`2S#fo4H{58};grhuV{zaHuPkF(WbHme(U5J#FIVK~o?0b5-o!;U{#Kx~
zIt1<tjbq4hxf130{*SKrfp6lv^F>EOP%DJS5m~yFw>N^0L})F@QlotK<|f#L^lA`O
zx^&CdXcLoGa+son?q*Z6ga`?T%?-__6(cHwl>$wL*|eddvN*cIK}z!SwkR$)soWZ*
z6EnHHG%Xtx?E@mk9whGfjO>u!eUA~+Xf$)q_x%3<e&@`1{T3?OV{OVH3kv#w{nkf`
zLQ}h+)_US|j^8b%x_w*32lssznDot8lxP`abv~*!u0rw4J}G)mWeOlAe)~lw6H31}
z@J>>c-Ek!Yi=7*LM2e2r*^#|syi2*3lptz@Md?My(9t@pWG4E?^u}IGT&(Bop?<+0
zyMHLyW>JHOJzwtyAlvNM-aH`2)^~NDN)!%ifo!$&Nl$w2Ln%}5cQ&v^3D(H#S8kHc
zqp~wDUOkKK;_1mX#T_MU+Dq(0+xyyv>%z_fWl;6~JSiF8W5Sh6?`Z0gArRF*<FSgH
z$tk${J2%QHRVJn7><jDqUSH7HCa4{X$hJfmT4I)0u9@bQme{+l?%M!JZNk32;yboY
zCr;6K{*(4m{@oSNcm{plbh!8N&+DSb8ZbU52L9%zQQaS3{lu=e#itYAJ$CQ?)JF~r
zvw!quC>>=HJdk$qAS4^a;MVGc@O5E4lO(B_MU@xzo+MWh)oI$c0VzB*@3qC1Q?>Db
zLG<?UV~uR&!+jRvF)1&m?-LWJn_5xgxzS(!dQo}fx+5>CJrCS|iZ<wJKHSXcj^KJd
zbx0>T!Zz=Yr`L>n9!+}Y>~fEsTe95+NXLbnl0Ri~XK@QjdoL9^d;a`BDV1emI__kW
zctlKh%}&vhYkwt`l<5FjrY@IuUh6m8`vr4B+;`e4-ooS~wakXv7)-FIxT9!=&$YZC
z6W_1}9mmpd|7QLxS<U^rEB&uq$RV&}u4b+ywJ4?7<^;<`<H0A`>gRVYhmM~JK*7>x
zb9vGSZiIsS<@ZMwF|N27QSA0c?V^ZK1YSl6J!;Uc6IKfmX43XV`Vy^CuPDHpn1(ja
zEQXg|+M|mI357iazV*;H!M^;Q#r*l^k>63vs|&wW)9CASIzLdHRNJ~Hav-mhW<wke
z-u|2zQ#aq-wK^9SXYah23eZwe^?qmP<8#TNO76<Z@2l``s3yp+ChIGb6!3ujo=_hw
zL0o|sfF+pQ)OKcx^YwmVq1{Z>8rz<_bkWbqsRP0@;_f{W@A_`waio$KbzateEPdP6
z14DFO=Jfo3`sXc9%C_F8%+FsJLI{LMJ@ZPUjd`4Hrsl6KZQDvtnHEIV_pJ?&Y*b*N
zQ)vWkTdJ28ItIT{FM#8s3TZ&_B)N--oFq78eC_uvh@78Dowgq^S;hy}A@>{<$+n{M
z&O;*9<Uo1HDqPg%a2G{&|IoLd&18P};-WO<rziKe>}cZV-ZiBK98z*MePi={&2oFv
zpE>srm&k5mSxNlIG%jWjvwyWA0g;?#(pd!0q`^@Iyy+Z;{Z<B{drWaB4Ro;RNoi%l
zIbN^Wq3^<?UpDm*2i6^5Vcy0biJ9D`ZXy1d{su?%84<Vg97-Jj+td-^KqoB>o-N(`
z{xWA-|M<a)pX_8Lsi>@L?#VN>yy^XOvz=u3ZmmIga|*Da_IG~`;|EG8sZl9~W3SDD
zeMU+LhH<%$E64?28kLSD*A2pWNzQe2JnGrq-e!}2cPP?)bFRl+{%uSM9JWbx)zWbT
z&zDpC#YB1R&A&bTi_kYxwKKFRpIEabwd?N3XR#IcLUgw;XYZMFAVSx+0Xa)75ZsK#
zz!NC_)8`wArbN($zsuR6D-$Qkjt0wc`(zLq)c(?`xD1==9@w%1RB!CRE^IdKp7D^y
z)@?^X)DG{z#2#OkfL}|CzbJAF&8y^7&kl_JP5kiPEB`i=m=IOPy>WBq^xfBmnXGdu
zBYC+q_nPpyH2BWU20i1rLvXSU8GsNGG5qN8gKQgbhI1#Q-k&M=y%kr)`a}jwHF`#5
zJ~t$2R2J3yQ>ubkf+sKU-DF9{x&@vv31UBrbk=!tK4F%IwtDPE$*y);<}W?`!)D^m
z#A`Ed|LbK*sJZgUz~;`RZ`p6z`|l$(BiWiDkRL5-Gc{`33c<2hvYiG7f5I2|(F;kE
zQDx{jI-D!IUIwF7;>b_fZ<0Eb@?cS>5FaKv2jor$->64U=zkxZY3cV^7VHwAA(>k}
zV!Pl(7H$)d2Je4ry<~yrtCGFF%QA1==~<U}`D8ewD$CP~JI-olUw9JQERiB2wuPN-
zmStDS3xcJ;4K`o(A=L}qb7C&Z3`x!;sU1?zoPhG{h_MPORYtf(<8Ump8&;9Y?sA02
z;SFnuCF<wSS)3asaaT~hNwOp4RRbxtn<qTa&3plomrbw$k{M-#ZnrO>zH%QlmC#zn
zJ;wcOFbzmV?Si#OP=wHJ=3$;zSfYZ8Bzz>Wa)pog+zDMT%b5ZAW(NYL47@Fq+$!lD
z=;agy2SGc8lVpgF<kh#@cyq~rvYUnW&wJ1o59m#b{iofX>ycXsmZ%j+EI`J{ag?yY
z0op8yb+C&eyd_k2zXlqCO?bIFyAbD!d=&h|FI5*1T7(&vK_(|8i9M7mDm%|BeMLFN
z$Y7kLl177Nz+yW{`WytOTj;k6s<enbBt`{GJEYs>ey#fp58d$!@I>1TJdzMYJWoJ7
z7a!7jz|hf9CC5mu)EO0oZUfIoc~Q4ulk5hb1S4Mv*ktD@p#uucAeb$xvY3?JjHJ%$
zS5<N(2`(xZklJ#qL{t-vJtXdZY@~8>wQ!$vy>*e=Z5Mdjy<8X#T{iR#2?EA$mMk%o
zz=9UFYfdK=yUap$cu5_e71=GvgZS7hsvG6wf?%`Sg$^(bJuE)DQ865=5~=GU1da@Z
zk_JJOR?Jy%0O6515GtHpOj&`uidGC&Txpbx6Tf=$v8K%*jdcmDy2NLiAGcdgKUsLo
z3m_7bF%C+M3BsO-R;$t+3>u&!U>aIR&!HOQXGI~)n^d2CSO5nFp(E|XZ%Qb;e);(v
zyA+pJfI($Bl~i~*o3GF&qbMw?lAGZ?LP4%FcW$m+Az!cb*u{aFwBXoxUAWu(QH@pe
z(1Fm4f3rW`LWKOz3PhGCy#^wKOyL@u(L-4sswO@sSYm=&_JB4ZlYkt~K{=Va!rA{c
zgxFO}6~{#|XMphjlQMb+c~WHsSa2NGC1+7u+$=n?UxxwkA#9?Im)>c~fw-Hkha_E{
zGx&~ps~De#WU*QKm8!2xl}yeg&H~$D8g*nMI?1Bof<#JCH%fLd;q(G52Sqwk_IRII
zu19zlEO)?2ayuz!5OtRoH%F*&(FVvk=~U&^*<v-GK8@lz+@3@IHWDSfC1zG0Hwkm=
zll|`qw~9kkfIOhU%{DXK1d@8e!U6e6jvH4#Bp|A!R4__NuaQLS9oXTBOl~QFpzfE!
z4;W5_iI(BKM8q6SP7Ppi5F81X(Os6gR86Fc088-He8je*+5=GxWl1XEc)5PzP{O;c
z<letpyb^r6`{ubF@#UO#7C?y!R&k&QSuFMikd$q~e-KIIc_qmeKt;P?a+U!LpD-JM
zWe1_&^8EMU3%xwXgtz4`jL`DB3LSUH!*|uWO_bc{Sp+T(%#y^!`H`jStcy0OBXJF+
zYE~LKR>?!wQ%13h*xQAUCtXPc74k_T1{qhGvxSNi3*(2%0ak54BRL~e1R4<G6I9)H
zZvayJldUh1YLgo$C*-FnMT*_mRC#b<3(2uyVx*Cl@^$^vWbkx6)W}I3!-sL=E=ng5
zkT<Ri4<k9;Y?g#}aKL02tQ{s-i&kO`sBE0BB9sfPTEJ0dxyAt+tYS_B2gNAPkU=G!
z85IPW{{;kB9)02kLV=<P0pGF}3@j5aN&{LXwg1|is<1{$6=lRqz}uN5h@{ZSvO1)K
zlTTNU7Cq+E7PH3^HK9${7ee1zE^+I`357m+w}51CLCV({scQxzURr`=iZ+EKvbJQ(
zINh_j?_UWVN$UhRRJN#eGG1z-liogvJ0NgjASqm4l(O8oGz%aWSkQz!WFSvYdT|z$
zl1g9GTyX83yLP!3^g@G3qj<ZSC?az^ChcH&jsg?aGAgJHOg;D&K??}PsSJ~5!U0@+
ztMez4dsGpVk~1mO%Mw<*UK|`yoTIXP0LDp%{5v6LXiAkK{}56*-ezGXCsGBrrAAy-
z!{pvq;n#O9{uLzJ>NPF*`ME3^rC(7c6#aQ1!jM*SlsCmvh|97}RP`xTF_73EvreIk
zdcOR`X>bZw7J45M!K1S5RAu8*$qhwhP)Cx1Ur|Yokp|ToCuoUQ?uk1dV@jlCpWzo`
zqP^W?!iB$Z4K4Q?G9oFwyEyV13;GjGt_&kp-e?I{5K`q&LOC!@6$TK&gaBV1g{v_k
zDExBc`GOc%p!l#w^~q8v>!DCjCi8a#Sw`(mp>>QFg{mg$Jg7BGR?%j;XOAUzzo5=M
zj@b&_4=#ykr7yd(>m@O2g%?wRD8&(pQ4VP)>XcQCh?X$EJZ0D7*M&=vZH@?SukP=%
zM?DrpeLvUy=dE*tst;#FL>g3b7G;=Cu*uCS3}CQS2ou0KZc%H$*_}iS9ZSTdC7QpV
z)XtzB)5ss6=cliASujVi0S?Xzr>gkT6A(p3L+~w-B*QAfpcs%UfD|-@0+5_cguIzK
zu|Us1r~N40Cm<l#g%pb5MKV~1)t1C0R1kWG;E6a@l8U3nNneN-z9R|opk=M*OAX&o
z2gpZxEpmAsNk#j+1e?h|<Bl2I0LtV~pulE;qA?E2++|5Hg`U@gXxhL;;Lw5tvKxX5
zOp@5&E!f(?Lx5W7yooL_gqEYapep%Ml1Ul_%IlG$4B~eVNS$aN%8|qlsn#xDJ+DKZ
zE0f&2Ybad-uC%<8vY>!2NcO*cY}*?g7e*wJG#?#q+KWM-I4-I4JPSU<ab;Va2@Xhz
z79prkFAp<gpfPkoc-Y^+oiK8u$@uknoz<UU+r^}uN&=tkWXiVlu(%vwhHZf<&`^V}
zQ1;!+$_jjg4)V7O4PCP&6_DRza(`ug3l(O)R1Sc0Q)EaolW2kk3)W6htHv$59#j=T
zN~wy|7ofmnpfa1j?R?9J;K9LlQ*zeMq@ZQ&foxjJly$`0U@IUGOPxtF2o+Q%cUkHj
ztAfSHKw`LyLY|gQM~-8B+{(;39^J912uTO27&mDEHCQk}c`T||q6fh?X%3>CN>1mf
zz-7wKpFGF{n79-`Q(!ZFJM*@eWEDpZiXhzuU}Y&hEHxnZRY1B{?5!lpUDt&(xK@lS
zVp;CIZ>ue-Aoeh{8;-Gd+O{=jLkzU`)u^yfSRz8OqBqHqZ4vqlR4w862xc$l7xbXy
z5?^z0y=rY_7^f0&1|bmLW#_msL{m@dm#~NM*m;SQOGRyR$A^gc5DMVTl@zry(wWdc
zq<S_UNl`X<sB%h`Gin6Elm-WeOn-<Po%mMu*ad_Imbm}6v4R+#jq4Q{8$g5J#jI#v
z==YKgt_n&PI32Vwy*2XVI60M6+{{7aem=v`!R^TWbLc|S=Lf1XNew6>dM@`iR9~o8
z7*x-`uS$c^;@xS4hX<6EFK5rBZJ-z=9(<@^w6r~h-YQ5P3<kauIKBZSP~p)HamD)-
zFfq2k0Oq#>m@mR%8R+2HdHg_NtlbbJC;%?t$RgrzA@sPju?$d>H~_#%w0xng7H%>!
zIWuFGSE$>EBuk2s6)z>+DsQNXO8doSe2)j+)hp<WkzvK1?7hU4R~CvbaOf^G0NUXd
zG^i%Vz)%+1WD>-IUVb&0#Xq>k<Yy^(=Qf)M@ZhVj^S8dvgRC<ROx%`qmP^j!j-40M
zr`w>OWN;Lb%N2<tZJ?|nrrk{jW>0U%ZbOY!uH4d~R|q{bcu0n~_=6l9<CK%p6&#My
zEjMSaV403aHt~FBx{({76@F%tpAVx1=M**B1tCKB1|fqikn{Cnx&Ssy^<5Z19O`hm
zasb!7><b)<Ov)*AK+BlStkj_MoM?EWZfmQ91RBg+18Gkvi3Z3L*EGT;5OR|67p#ZX
zu1YRGnXpN49m#3aEnIHxoH0&>m_npK0r+=~SH9Xq(gpvat!Kwq49lI|Nz0uB^2)S4
z+>xjjnLD|zk=Lq~oc0pDx9-rD23CdeCM!*?!gi1vlv*d1xP++s%&;{x#;HeAY4oAE
zr?J6enSC;6$R-mxW$@Inh{hWfa0(kY78sKsL7`xNjK1~&tPnyXKL5i+?NXVDqcSae
zw?Y(?gbwufjWtIehN-I|>dL1VtY24LJ){Kx$LyxY<z-^Y%Q>xrOZ$+)B??R0EF4>$
z9O2mD_~k(kZ>`Ynp`VviW%ex-`5B7u*#<ob;x##_8Y6D-hM%{I2Sv8;3acLoU*Ks_
z!LMyqB{xD0Ra$vnDEjgYp5s(b3+J^^P2;xi!-0`EtoVa*5Z}~rOWyKePrk7@UvM(d
zytw3!D?e$mGmgki=CGJdM_~bEM7+*}p<rBZ?pb$t4S=!}$MzrM(VVl4ws0nZC5aoc
zxCfl?U3Bx#f<s-ilw}|#Wt1tEO7_evjzgsE#S9;0s;Sj##_UNL<jJOvFJ-T*J0|)V
zY3h8ORQHc@6E`j%s<@-p#dmaW0JV~2*-(o+KxP<+`-pQa6ih~{ZW5{M;WV&hdHc1(
zGUF3Hfn`93%DUi@MD?$QwGVm&G%_i54n)Y+Jv_>1oMhJpPzIhC=lk3m&+jqH<~o?P
zN;Rif@)skp1^hIS)956_O0YjiYudFkvKj*)la>u}WO$)~kb*;>O*+r+*g8roHJCIf
zzdS7=W#L<`G+kV%aE&vr5>`oA`SZ2=xAotz5)9x3r%{D_+jt0MVZmXTMZJO+LOC*I
z2%A>rjrrpMiXt$&-Z^%rlILIEkE!6ufEbgAKRa>QdM8LIYEp#vq&VS>0dZ9693{?Q
z?g++}@MwdJXXtWI9SV*X_}muXE~erVygx1yY#g5LHya8h&4M39nu&rd-#0+;_aGqv
z>q-=qsW*LK!c~=1kRgAsjbV*hB+_CwKu(;h>62>&fhnRI6=o<vv}nnlWB3Zw+>9Hv
zOOjJuN*VGz<-hztdNAoD#oI$!9km=x+ZovMdL&yA;|l18=`Ao8udh2}=P?)HGf0su
zpQ&rL0cCMr8XVr*(;7+^6NwNvMlSn86lC|IJDv}M(@kJ6xi#W?ITKBwF+uQdN8<~Z
z-1g5*aWzBlBSJYcOtPFX`SBw<fnp5i^{pCuyFA0t`bu4cd2GOzVhO^ys|La;i&+E~
ziQ{vtI}T%EyPtN^YQ2__hqt_6<}#d_+&VJ!K=#TOH^c_!PQ`4JauhZM0#}7Cqo>h2
z+eRySr1wdwQx)3HGK|{7C7qy}D;xPJe<q?6s1?^{r8BKK^4jk}ER6KBaP$*5199Te
zDs>|%tf1n)vPG0hj*b@fbm{v(qQXAiRot<aJ1?Y>8ZC!vqcSY97?<;9C6ko1In6MX
z6}*$ngn%G&merav;O7Pq2B8>0I^)uhyMzt-f@?T(6+OS)M1@$7fKjlocV)>6s)KF_
zN`+ZXRKv6u5*Twx$8{lBLDc|s<C_j2i>BXp9{7I2kuEq{C_f>VcU%`nd5s7Ch{z3+
zJjdbn_z)!sOhv#2F(PRe$zs+BLW35hj(8C3OfXmMhUxBhZ^)O3u2+B}N0ntc{fYe{
zI#%6QQz&F&C`_`i3BAC=wViL*YUAOozLHk?)Zz5j1nCH6Bb|GH?sL$QECUTFgi=L;
zm55lxa8joGjYc9xt`s<(zmRl-n;0PE1sq8tw~G^}D)zhf#Z(j+w`96(V|gywdgG+t
zkyBRU)ZbyW$&syB<hwAS+S`@vIpID%{0W#wRtx3XJ-~-1Z)}hk1vR47xu5Z47Y$1Y
z!~_CF)g3`BAUP><5WFuZD2Zxpl4p;uCTgbq5(5V8A|4!Gu-Fj4c%uDG0)?q(-dPY2
z7b4FbiY#k}^G*Fx$ocROcIwa=R;^VlR9_@SZcq%RbC`q%7}cojktv{sPR`)1;&VT%
zG2%bVFj$gFf{5G=hS>uv=*U)}h4Te*G1GuH>J7u{VIa_4G<MQk&FWny5#?V~2rh~2
zT;J>Hn%6sxubK3%tV23DR&kWL^VOY~<;8VagHRZw6oq9ss0JYn<+YED!Pi-YU((1G
zIM<S%s=fy<kx&iY!zRm_xIz&|{5UiLf*@jtl+giiFp0v<bz!@h3T8KtgV))K)Tn-0
zGNZI>>$W__dNB<P?XF474GJ1`7~>t`DnwV2cm&odGAUz0>m@Zra_rHrTdKxcH6zkN
z@4(0}09L3uhlj<jmsVbR7CuI)P!8AK3K^jd0}{SE%QE7X6dXE};j>XaF{lTl8=;WU
zibX}O!f5`wux&IH(kh|lTON_szpm??`39X0E64Q-)+Jwv3@DB|>l?*tO+DCUH7Hd<
z3y=sF0avP^cmvTI0AiZrAR~G(a{1W|ZwtX^MwRgDoK^`ZBM?b3t}L8Qpr6iiaL?nB
zj)aCNvor1sryn_!V;{0Dm7IoH1wpS6m!1sl??qQ(gAfqkDTuu19Q!@`&N9iB*E_F3
zY}nT4=5Tqd7!>Aa!GIgiX`mr21yX5fz5e+7>YkNovoDT{8kXlk-axyS1>!rjC;{l;
zD2PbJ0cFELV|lU{<JCe<h6d{--Fq#6ExJmw%dvh>n<+5Vi`EEw;2RL)Ez;EC#|E!H
zOGl-9_Oqjx8NQOyC5{|cC9rT!tJZW5vytPc<I+gE1aN5nzSMzZijze?vg#oKNOQI}
z5o_<3I<(Ayyj`SnRu3L5q{=%aCE8iz1_5CQixja~N3R%M%$7LZMvWb1c;3w1b)SOQ
z0@-5SS#+1xX_+(*>dSIK2<0g*S&`@QWFxDiIap~VYxrNuWB(v3`^ND8Z@Xy^6BDfa
z-gD-3$wl$<2GdeULUNrf5Va?e51o`!aC2$9wPU!!6_{UGLl+t(D^%Bmv^043Os%|j
z2J^)i-m3i(LQZF@1=TGm8!EI&!uz#jNA)n^T{>rXHnmVq3uAl$r-d+HPWcw~a{z+A
zn!h}3on3Gyx{|0-CPsoUI!7%}R;-H_NiWRa?%WFjM2M|mN@K7XX~dNbuQTf;+o;fn
zi__<lHgdIm(TSy-RO{39A;*ed1D|e~7Cr#FE`sU)w31~?1matdw8&chrhWH*K3Ha`
z2l&ZZ{0Y1!f<chsy6{W8=S8p;g62*lASWVI2FDY}@^WfGc1Oh#e=HhM3rnodtna#z
zBq69i(;L}~q$br)8l^MmXh~d&VWX#<e5e1<TL7nwU$v>VW-*{7S*0=g05>HSeF|jb
zoTdG&9?I#E8m>jy1xOGVnE7UCzJc=ao@4wInn5z7%HT^p*YmnCq*~<*f!(oPbAI*^
z^*Yl3-gmLHs0_xHP`>0ICCh5CRoXSmS0g&pG688RyTD}pwJVL25geP|npy?ioc_k8
z48-GG2%WUH%Etodbf`8}G=TnT2qg@>s88mM_JI7CKJkl^IK0S8QRMJq8J?IZZSE*s
zIOyw83x3+;!h*B>|NBMhX!DFXKq_!tW#Y=A<^0IjEXt@bol@uA3S<h(r<hHDZeXQ&
zZ*8Cter!oV>j;65Ywck~pfFCDu-kA!1ls+{MG>xuxR5mQ3($hrq2}*}z0YRKo)Rb;
zOna>966Qd3DV)c8vBZQ!{k0=Kxpb<DYVegK&52jheHG$_o(RP+HMT9OxXX^i-J(qr
zurfV<2bWP!X>|DMZ5ks##<3Xov&zL1U&45RA&wXpA3tcHy0(Rx|JG#FKKzZRMvp~3
z%e@QT3xevI8pM`B$xm=1HMK1r^xV4rFJDqwM*@|BMk>!QzIe~c(92Yh0DnX$8P%hj
zEHj9p24ldTbgdXSq~lZl@{9hzx29gLLwjl(nLvEU&#Ci{r(_<?X>B@$0iAIr<vY!P
z9tr+@ge;Ok_!?BMEQv`(Z;;B1_ZSE%E4k7=K|WQ;f=|hT<I9?ju&w|S^Q%2~6=8pa
z?a|NjdjHJyDyu!|->Y%Z=!5tF|GzOXi*Fcu%jXY^YO)vKS?I<^pTM3P1}8Uf8=wDj
zSbEltFR0c6F?ImGC>e=0{~b}KI^x3IM%v5eMjZ*@GFl)*+!X#w722FONODVRFFdD2
z#q`(OELQSjFK-IqYX*3{fr9_%#ugir!Vib2MgTmHhalEk(htbm{|Ot?f-oPw(F@_B
zAEqvQL)?cVP?TZg5c#c8y%F9TsjFUx+C$RqrUY|fy*b?)$LiK#ynacwx?Ih*w%HQk
zSC*Bmf>j-r>{i#V2RLAZ@laNL=D2q0XpWobBUugo0%S#Zp(=6%WF8|!O9uIIt|9Nd
zod$hROdiXpecTRZ5J)4Ga+kdo%KJ4aqX%dS*20m9yN-LsBZK?V(&g$_H7H;$g7xPY
z=7JPS*;YJ5E_X$R&RNOLR@Y&JsyLBTH4EqI>*alNpds{WP?KXvpoH)&DOKSLuM0au
zITkUnarMBtVFLorqi@cOKZ6fnq*g(6;yhGYQS^n*vq-J%4XB6#MqG)KKP`P}mbc@_
zN3&<QCB4_;mjF;O+2=9x0#<Z*EV0?;2UL4hvd`~0-dn@ifUSxDbp}9_<2Ahs5>)D(
zt1M(l0p6-tWM?|WBRmYG41yT|NfV#|5EtrP7PGI-`Ze(JEreA&ob$+>9eGwePbF8b
zWOY^=hqi|jh+k~0%q;lj6ZyTNEW31FcvC0{yI=49VK!K7v#74j=TMvUBjIQhpK-)`
zx$vVcj`A`bU2d|sri06oV<$sLaokXhyjHm)VXZw%4}hD0{DX(NZ0Ug-;~)H>0WtsP
zzS#FkmPPc53~I-yNx<WZMK7Yg=9so6aOlH>>}R~N;W%|+cyW#W2x)%2>oX9hxMCNu
z(hD<8>vpR0petjOEDHkt09WDQIE(WKLwUVGOMx6Z%$*NZSOeBGr;|x<Us*=<JF+sr
zkrNC3@~wvJZft0V<X$tk|BPo!7!$@^sTDG+;#MVGkPzcL_#jd+SzEqM5CY~oK`4?-
zrZ@}Yv8?`K2@B3=o~VV$u;z!7@#ysqum7jL#^7;sYpbz)aLrcA(9x92Ryc(t^)Ti~
zXiefs8=23)-K<wzSMZK*YzEM)v96!-aAMd;i*o0v0wa`BWurO2DjfsC5S3unSX$1M
z@*vIS4QxxeCPN#YPYdm<^lbwc!7Q!NWdxd?v00vu?*S!HO!d@&5b`hy5p&}Tgp*DJ
ziV-B&w}&QSMWoLB2i0o_<@$7#bps#Y!`RT+Rn_|gg27)G1|-KW`^#m9^##si1YW+6
z$qr8c9H~KYC#B9Lq`pF9^L@vfp%0mq;R)@T7rO1{HdhdFOnJ-Ws}m!spZPwGk_XC$
zsjM_{!-p2W(hEjuAZw#8hfh}A<u38inG55&anzUo7~k|a7Uo>P8f8Cd=Jg1D+<QR<
zd5uLZa&3TIO>%=o_*hhFaas}&@e41g3?vQL|Ioi?NhsfNm*qv9#Xo~@!w?3GTYbvB
zlKmDRw1x{M-y*OihRx%`hcO%9v}&De?il{F%4O}*_}WhGVm4oX_V=u%Yy7Nf0YJ>k
zz}(+N<q#@W7s?OR;z~9>fuJbNNG81nU4!C^8$dY>0J)35zh9w|H}v1<*GAXqm0ZC|
zQ)bU+D#K`ZSRPDn7TPRNy3!meKe@FTR~ujr%C=G={!^o-@sI;>`Gzk86QB)<HaY)&
zz1jq5zI_@nTD{tCc~AoMw8YCrNl;ti3a^D4XNj2LepAV)>+H+UP&iT-tx|muH=mwI
zH~G@y$x!2!wb2=CGeueO^){$x-t5!>7ARWV6jm6>7FO(3>slqtLO?dX8NccvI|qyZ
zJlG?jz%`5)1$gin%!uCqxTMU&S>Y^hwWX-Y@syb7hq&m1N{+_E&f*ptTZb|O3#b!$
zmO8|(?S&sS$4@miIv~ws>moI+LiI;kmsk8L?9-=+JB_KvcL0o+W2!ZcM}AX|@sqI`
zcpFn+b%eE;Wt+YnBLH9xP3UO+yn6p5SRPuaVa0ExbqaQATyc->e;GUICR<hxEZXr(
zP*~tkj09pR!G~(4A7w9^o7^~X4|iP{%xcbEc0siu6pP;cvIc?}FsD2H<IgIPS9oPv
ztF(TYLJClPtKe%W&mz2WoH@NU@pDTl7-^nwU%P<?sE<377LjU2>0YMLxP-ebc7&KX
zAYm2Qxh1n0NqB>a!5pjPSW_K;pss`Sdax*~(#=uHF_<j4+1gU@&>^&AkdpT|C5f%B
zn_8s>nCo+h!<do~?0xFwDIBe1H{LmF{;h`QwdJeV_XMaH2#g*o!#%hce503~fLU<e
zx#=cL=GT_5yVTNVtcBe3qElVL2aNJ$lAF=@qT&)>tSTZa3PwkXSPqs*{&on-ClXt&
zf1+d@at{48wJPEQjvBw=Z3^@hsws`X0r$opPLCL%?_9mDfJ#Omu8d1=CT9w14{i^?
zkS=W&U-MYJ@6x~Q7#b|X7^5D`mSy#jO2YKoY?H=;T{x|zg*#Mf`5G=A9<M)nJ6g?8
z_Y;}k;h$XG@Cz(*1oM?Syt6Q3uRIhGG`zu{)-_YY)B2JHI)ek6-LiqXkQa<oq~1wI
z{6}?-$J$v)6xsxvWS^xXLno-Bcg6|Ox{Pb|NS0MF-fNf;e!9&T4+>9PR(Zw2?4`Y~
zAKb!;FpUzNYi)pp3mRFy@=>z-GRJ%7<<Tn7c9AUeaR6{@GmPGi0XL?QHnr&L4TxPz
zeP!k(PDaFM6D1SyAZCE?FJRToAPgP~8|BUYZOjEyMyOWFl?AI{>$2||#b4YKR&|SJ
zkL9>vm&gxhv~U^({RHMGAhP<Uk01^gk}yu!g+E;0@p~S}LJHZZ&IQ!*mM|G((?Aw*
zb*}#rUjVnp3G;DBLz_#4ae!qrvH*Ipgq;i`;CRU?xRJELu9Q|{YG)ms>F>5d{)(5d
z#5Xo4)-K$P)j6tUDUoCxGW#D%`UzHu07j9E;0djfYvpgeRI41V>7zTU4<dRph6e?1
z)EkaZnn0*#C|HdNGar#U)UiE66rA6$r8p@S5I2+D^60PaG`{*BCRn?Lq%{gd-P*h%
z7|~(0OI3u0SOA7{Z@;l{8%q~vi$AkHrriJwg8j!$*S~lWBL#+6bw}mh74IkXfxk1(
zoV~JjwjR!3Y{3O#x#@-UfVAfY%jV@gsgvm_=3YV<uzIUV%xvd+No?!7ExPLdWpzHE
zHbkh7C7=k|AO$e?oNsPb!c;vbyv$u#s@yHwJ}=J}U+xFV2~Su4Bg*L`r-oUT)2mr~
zHh*np?zxZZKr=LP^|96ofLd|j@+$WcMl(=odOo#R9xO^I4v6O*i?e@F?@w6$)=Y60
z7K*o9=VCY?Aj;t3jRemZNEl@@Tuo@fiA9W<ziG2zLLQiHU08|e@%ay_JNeIu3wggy
zgXPl@Tc1mDV?>?r=Er}Y{F}vf&k@e|D;EcigGK;kAm#3?k~fqwR6GBEX>F`<2Xbdr
zjY_U0c{Sm&x9vMU%NL;FD+F0`rQtbvisMKe<qFFON(-3b0$7W{?`d%1YgC9v*Q_a#
zLuv#TiLx+14KPE03{qFX8W6CmUp$gisyS`&p!P2(3y-wUGLRa16#rh+40qn3aA3cA
z<ZGL~exz23f<Nu?N#T!Fb1>n(bbP^P3;LHja8g1gs&oW96bt8&ijllwOD3GJiE6iq
zX!zk&3MMV;#UoUb>N*1^hSlmCs6i5FVXMzts}xj4Gy=5=)i7Fq!q(hKC~siH5Sj(d
zUw-OGcub7;<NtPGbLvOAk&dWYS~4g(7FQA_hdS@{@{PECHkdfKlr@>mBWtHdz03G?
z8RVuFZ!`umc0MQD1(ri%jg`;k&0g`)4YUSM-Ywr?7p8Wmh@G4MB?3}b>spn6uS_3o
zeU2Nbs~8f{;#ULpuimqKXC$Edt_!bIhUF<gc3eoaZmILf!mR=IE$b4>sgP2o4iBE9
z;v>`n)K@WzAYIE>29@nFvt~>Pm<2(VvB0B5PO0m3$N>lpvp9|524EiLafs8u7~=w5
zPC^}Ddif80R4suwem3wnYW$RUe{ugdub3{Sp6(J{YEe15YM%G=vU6Z<!I8MczT@ja
z)jQqe;!AXV!NgZb<xGuap#MpP!YZe)S3YirIRLz^rI!w<((Mql5Eq8*G$O-9LV|%o
zhfg&qZGgUU!l(Bz(Kh`x`^g?I;)BtF_}t~Sn>TvCOq$KIyaa~!$d%>lTqza*mK4m~
zb?KzfpDsCK2R37&DN;#c)>wg3y$^V5(Kak?<B-<`qc<VbmBi8ScrW}&l~zIzd9pGU
z&hp+Hb>0=<0KAbmq?TtsH#TEHTsQgldkBFc9rWz<8OAWH{2Lm2=RM2Djfu$uEqWF=
zxXmqY7<QzKqk^p{eV5Z~QZOrS?QQ>Vpi(1<AygE*O0)o%YcWf%6r|Z3jAHHeu8va2
zkJ2s3pm`Y3U<CD6uLcAoR|B2Vt#Nn?V(W|s)o>h#0JybfjNj2rOO6b|`c3iX2Yt2k
z?BOEA1mOjSx>j*G7~fWu5f>bTEao5%(#jh1CookPgZ<*vjN`K-=CP=Li`v#6#m(p>
zC+xa;WYb^tqnyUA55JQ4+GG*_fkU7?|NIoOr*ZMm3e{ScxMqOF%&YcC`5f2UkH59{
zK_;pt7%tQ_P-hw&+p|=w!}}%))yA<{8&KWoTPrNiBpdXidz1{a?T9Bk%0j%0-#+to
zyU>pN%IT$k>-;F78lZ3iUfg&_Kq~@iHo_|~>(dCQ_jR1hXA&Y!T#`4|?r=ii*lPT`
zkKeQGENya29bdY<O2$ey5chJ=GDe08sKC3J-fD;<u-ap9JQ7sHxyoQ&&$0(VJ=nB%
z?sYA{Si(gopP!uHc2CTT#NI$lrs-tx*GL)-^8S(Os<t`-qxtYCCK&P28l4sfbb|k!
zMWN=kO*6-Q!StFL-Ywr!kO7Fb^BbKcg!;=g&emsom8AMex+Gd&cnv1LA<{}Ji2#nI
ziX`n-I+K!9l~Qf8Fz3<%`|~kSm`9Y5#2Wa|w<^x%JyT%%<E9O;4UU!DIi0ft*f=LJ
zW`kgc9@bd3_V1VF^{51&dywtiLmPD^MK71W*oy|}yQ(^ZntTTkqKEWbt29Vn2Nnw;
zzQM#_bNqx#-n-?x(B~26l9cL6drZQu;!E6DOo;-wXe6&209)_VFe7{?0uYUy0TOy=
zV61VBO3UW*Q}_@8(R5i7^SVFXz=FgJQRmqr+hoHkn$p^*e1UxhSNg3kAr%uAx>UqB
zA*zH<YA~des+Wkc0GK$KgQc)h!4$=_B>K&N&qD;Tnj@+LAa{mb4mgk&3jT8?tB1H;
z7(~2bm2M;opE7&KSrE!^dB#`>%&DYtf%akZjO9qWCy-{C-&iQN$82~8%K}Fl1>}(>
z;ybwWla8lG<+QaOwa%CozRQkK0?SBIi#2~<I^NvbZvf>%0l049BU!D|0&2KTuIqqp
z3<sYq%=bb~oCEg=AQ8iE@PmBQL(SSI>v?{Bc<ZF(NPlx?+Dj6?fP4LVi7r^>4XWNi
zqST?z@7QOv;O5KxwfLr2<H|9tp5A5gfs%qHYPY~0ATPj_>3?>ed~&+&DI^@kC&VP+
zEy5riF+{G$cMXJ~S4(j(a4d!>w%Td3NUiwD%x$_KyhHSv#=20|b92;Wn=euyE)UX@
zYq!n11XY9ROf;Nt<k3<Zf|8a_RA`SwZ0{0m2m*j{=HaUYN`ZVY-fA0WhBWm`omF6S
zcuOpd1^}wXVNk4s@03l^9H&<RyQOPf<uR8Q!P#`|y;<Af-;tPfGDUntnD3$pR()Hg
zFLWG^@~i)_mN!EnatXZe+JT0WJ<g)?(gDG?dCAg&hujJ6Dpp!s1&gY9JHB_wAOP<l
zGi>M*Fi<SNE_|oS5yCI&QxI!97lIt)wK+q>s-I$S3uu8O2F8ECWu6o~+}|w<_k9)O
zx5{I3vinm>VHr&w`(3&RZV?bg6{mXfn1QVjnj999N@gPVdv;ul%bjK1%WtGsqjF)<
zPY+G8N1yoUP&miV0BoeB?YQ&yN*);j@M!=Y%IkdG5X=@nKsR=H27UJHaJTZq2DbM;
z>n@2=^PEIslMuU!43-tTUp{Q5a#LmgbiqLjF1pGr#$BpU5m$8$@#*mjB2$ET>Kc1{
zd|imIA&h&>xJw|uc0~Dp3tWVjmG9=a*R&C;{r1|xi`(x8nlQ9D4~}y!H_#dU)p3|p
zx`8@Ee#o%;ks3mo;=hEJBhx6j5>MV?cgAHb^OVFPpGXU0+8Y#-Th%pwq1|IQxzrKA
zI^)_8B1h6-$>Sb-bQSKhlT%P?Z!iXruC8U`Q!QMicJG#zC$}HVNuXNz{BYm$DImC#
z18W*zIgtP{46y^*H0N1}n;Sxa8Jm_6)`|};UGDR%!1qp3+BS+?K$^@?3wAY@kWXdF
zL=rXddF#9I0FgwGi&~dXo~ivN7SI~U9k@*aE_3s1qf%~E!mv_hkJRy$j;^Zt;YvQY
z{pI9Vs=a;c=Y1^>0#t3vt>_#frxdV(gV@P4C!k)8;4K9XbFrgik9>Ap32A#u?S!*U
z;*+~B%8--^!;-KI;^s<>;2rQqRoce3Vs*EFblTOY<&?U)y==@4<Gr}HB)*^nfe{Q>
z>%*a()=^`<uO>R*%M;)MTk$K`1)gVfr2O0)8^AK>+4CPP(;C&`Y7M6)wtasCvPf1)
zx$*Q3uY_#eT~rWy+4hJ@F<lT7JA$#fAq~Ez(4jW6SQE1SSREtRLwZoy<+4k5AubQ5
z?dAc(TtC;*VI|A>j0uc$v!;b}lf9RL1%%UJUI5KNXAm*MflYZ}K*5y{gYc0<p_&Fc
zdDOe1;9BnUSWv_Gg57Q_+EIHGtQJYcB8miv3sVq)UhGpP7-UQr>AszODh<M^UKF1e
za9`0L+`b1-SA^9I_K-mx>;kwDu;;YuHhuhxUK&HkkTea+OBFl(WnhCL);Y9DlRY+2
zey}-Cg2Q3%7?Yb;JtDc5C*{~&QT2(5lIsO5Xe*JQ<Bq2-SiZV7%?*evgSY^eY-O4p
zPx>nKOtx8hsbCkt<StYX!D_o;M*jE|WQT8AL=8Hrl?iCVYsDmsC=s|A$4~((d@($+
zp^^|RjK9ggJHSSoAoQ3XZlT^b1K22A=w`u0iKLn1P=(4mIk32<H7=;g(K#xdfS=aO
zhiX+WEifQw(CQhd<Vxc{o>h{&3`*P8RUU~P-IC|%n%W13-Q`K(U7WLxOriwJ0H5Jf
zfo<x=$u@;fe{`UPQ+2|0q(V^U%U*X6p1;*a3mATSqsNIGMD-p1BsV|}iW*58D_vku
zz~GUhybR2W3?X8wBR7Ofhuj`ayt`6yzu*yUo>@}xNmx5h98GUF8u^iz4MpUK{84KN
z;PM<CN0v3f@l3wf@*5aHw2ZG%YJNV4m)AFa$tgMDnjliNN2GA}-nu?LFz7V=b++jP
zbP?i4m(>`m!_A_EqPyBH9W4up_{9O?z(T<+%^Gq2xCO4Jebg+4YZ@=S^Zxu%d$m^L
zt=7;;tNAH;gJ$G_3Gi*W_Z)QRbm7qWv42^FB_P}>u?}_BT9~>tj0oYkH`XVaf56+v
ze<Jv!DnUBJB{<LcT4$IH9&_eLQUyV1a}D{IkD!*+W$`+G3^@bsn(X1`_{;Tjm49n(
zepU8Ec7TQ>yz$~(1y#dYzY#SUS73~K2uh8QZ+$5tb*Mgn%-rSXwpcuL4G$~|LFjU6
z$pOxmmWcbW3jtmwD%tH7T!J8vo|oPKa2#t=k=#9%Tm!O8ir~(VGZ7^+z5ziPh6KHV
z1H=3X!~*e8A0UDu;Bt$(1}Q6Jtt0-}s$K&h1_{kDJluOr6Qze1c^!O2g&_A8;Cc86
z?q7zyslE!714q&?_Q`gWxz+bISe2qo3!=?r>4yoAZ(ttN6Ui-&Jn_12McbS>1|5L|
z1A<g>14E??U{sEC7zZJ6dWCn2X?Wvk7E$c&tqzQ9xln?|`WE;F9^3J99}99&SjUU!
zd%_YhOle)<Z4?lLiYY82C#nw*w&@&|Ql;BdiHXeqeX55)z7cfC%SvvAYbRM_p)(|8
zU<3~!9qs+_=_Nv!{T2u=Hx_wLT&;r6sb;8YmW5xjM+Fxo_0;<uAV>%rLGjnY#jAym
z1o8?APlgMsJ)G!VxI#7Na&cr#RM$%=jEerE;vBe`0l2v2E?A!Ivh;UZfGIi?`Ni4;
z<j*0F=s9s+aOE|UIfiZ;p{y2s@z2z@8Pz6M1WfQWte%f(LtCbxfrX7<h%sTwzpt~S
zwDqnT6hR~U>cmKZ7L3jrEN=j-5%f@9#{q>TN&)(6oKp;ty?tYnS;pGqyPC}XKP<c2
z;2ieF1fk!<AJ`{gTE+E?7Os86_teWhzk41jrT#!x-@4+J1NP5VU2*cT4if<<@!*>$
zAaa~MJHX&39?IM`C;b_X=1!jbeoIxi{~0a*1D*nTe71{Irohi;5e^tJOS-2cA?409
zIf24Fn8mtGb2}2HFDRJ<q7DARDlY0qPL5U)<=KT*%KYF7a0e9IAO8zF+57&!b#>e+
zRM&&|>TTOk41M>?!%LxOdHwBbZT#5hAQz!y3jnZ*VwKRleci{-MBY0E?BbswC``mj
z<^Y_$U*%?^qWU6eulkZ@wo5(jS0zWf2T{Ogc@PS>F4yQ(70f^48B@aPJMcUM6?IQV
zv~|s)Og`Yab-Mv9&u{%HZ{MnpXCvf9GN^oUU~}dRxBaUhz%GK7XO4_LK%T6`W`r&+
zXi(PN17x4u;wie%W-r2|7^M8hZnMX(3iC!HlvE%aHAV)cFm4P7Fa7z*m9n>kVo^~E
z53(j=D!wi=N}Rx@Dhv1UyO(jpG|<^3YbGxBh<~CRxU%fqTR)!NR|*{I4INFAH!eLX
z?OuG_3y(VM-{s6bRy;56OT}hzGeWgyfus5hYrN6O?ikY|I77cJ6*Dadp(H%z#$8;O
zg<qut*7yrfZ(jm2lO*yrlJFTLejzhf*>+w|#Zl;#4@dDs>jp$I7F=@Ny67v-Zx|vI
zdX*!qv2{?*Yx(LaMjje_#+q8WdS9MgfdJ||E?oX<&`sYl;R~yPc>D$r;WHPzj?fF%
z4`6MelRF+b9uv$AxQ<cDZ<$q=oTFbUpjzUfK9~*%@7o8uxl%Mqs=0Kb2GgZrHLqC4
zUAls~ONe&OW5Gnxv&gsbPYhh-P&$;yLu^^NHXNz3pgpm1?Y@_TJ_rb;12DG#Zs0yb
z`<`3KqmBxfXmrxG_=ClJmg>m~H_R?v)N{QV%sF+Z-<TbOBaC%BeL)xw3i6gf5W}_N
z-510nBI2T@6>8}HvMd(GwAo_z$o4iX#pDKpyCwF5<A>aDRpPS)gXf-c>>(O%xnO>i
zu*zil$v^SCr61!#N#oMNw=Da_=t6p))d3X5>hGGV_w9BcALiD!+5ES6Q_&P(dq<q#
zQvrv|lb8S^yb2;M0FIa_IMSs6EH7DEUw4Vgol)0elfB)nh`c;T=$Yu#-E-*Z+5?WX
z>X|>t#&2n!B%#a*Rwvi=tpPjMW&CCL35`#{@a;ok4`ewNy`9FTyFiv^o7gYRd|YtR
zv5S}TIv$zyzy4&v$BSEo1&}2`51@vGBnpph3MP<{mRxENLXRNUg+RRIP+mzTfj5X7
zTHSHtj0X!_q*T(QdaS8p?*g$O?Cx*7r{{l`NBp<-<a9WmtwHDK+bs7ZrmW==w=rKJ
zX0t+TdJ&Um!OY8T5YCxjly0SoB0HovavvYKO)WSQJq}J2iN91{^oB|GC)R(30|6XZ
zcpL|nULBCYKv)1#l_~7}u6fCz%L(yPCHH{duDrkz-JFpc%Y7$e-nv#nq856{{yjmw
zm5jZOJJjoEr4Ea&OHA$Ox*(sZ;eJu8&~A0TxXI$)fA-h=0qNQ^f0h{GM8XQ`baWSm
z;DA|P4j|!Kt8$|cTjy6X)8KUC*<4^2%HTXE%CO}KvH6XRNf1`DoAucXpHfYNZ6=kj
zW+ys^d<j;qGP20+TVkJ!8_06WiKC+*-)w;gO{tK6OSajFSuHk6+&shXE)TJT+^$~}
zzCpp49&pMKX)pxwGFLc20WyNsqZzA<e?{<AE28`D&axEF>+W(81$X?AN`gd|m^5d>
z!hTp%+dtkrvT}dq)XMSaS>=6S=-ad6OYFA>E?07gCSJZW(QU`%xTyBln0OavvtwNX
z5S_g5aOU0Q#OFAokK6x>_Mmg#q3UxbKody>(Gor^g$G6`9O{6aVpN&~!5j_ojFdmB
zuaRTU`ZzcUN9hJ%1yMEcuqrPUO?KO_?)?bGlxz97pVz~Ktn$U)8K%y;TPI2W#0P_;
ztGZzP#h6Wslq|TpCDvtcx7wp$6$gL#;?TW@{7>5c3bM)--sew9NHaLPG+_Dy;Grrx
zgFZjdsRdX11CO;j=?kSx%b=>0;ry^-h!xiCh*6Um^|U1&Q6Mrux8U0+Bcy%^31sL8
zECgxsI%3aY7=+f9Wn<?&qS<6%dPU4KISs!m2+7sYzf^qjBGvGa^RHTOOZYmlpjU;O
zVTlm%lw9B(Xr}fAmw`Mz*bjiO38}B*g5~F)5u&N#85bE<PC>3QPq9Y;KZG{>e3E*{
zY8Uq2+V<23q1+F*G{+ln^aP%s{Z^Z$e>KvL>%xnDMRS*kzygGF7F^J774Z)U<+0}u
zs=c_UQT?j-0f<bZ#|3k!`bc^N)h0jf5ff5S9f5%o)q<$@1O^NKT#uMHk~<@QXfgz)
zCxB^;3_|+~f*ESpWl4&ebC`B`{;s#aj~L}vxwG{4+2Km+slPYb-wMnmvsOv>G8KW2
z)e=?hcNIVe4k-q9dn~A;?H`Bnr0N>qe}U;qfN>0*a&aaiUM_X0U}&KO&gI1n!#pyq
z(u*Is4iQ)qzE${;8(0-kF6H8DQQ>JdjsPZU%9JI$@O*a5Q*`jcfd`I{uN&2ZF13V*
z{4D$o`AYFGx7piGu>EjL_PGUn45&O}j6czBfJgQQD9<x{(*;;(m&$E7q{)$Bt_0AH
zU<eP*NtN#uZccNr5or24F!c!^2|Pwr(MUg#N$S|GfCVprKKN&~C#^7pcHtns?^r(a
z(J!zv^zDlF;lp+5za9Vm9g71?FWa`+`|q&#ckz>LV4O!(yDd=M+2U1_)0ct|o$5D#
z_70?&%Qegpw2pC!riCTq9+K|r?qGIx{py5N!B0%@gRW-1T%LfAODzl43LR?T6j};I
z#i)Q{qJ58Ipqe$v+z9tx9eVrKlAKR&U7VJdAFe<20$H>kbvGow!p-0!XI|6;OdQ|r
zJNA>jj`D|=NxbX(Ar$TSE(e!6PLwuF)RTZwA0g?DIOtbPtGRUt;zC+l2@W=<A}~G>
z{xz$Inlca}&WUwflP>r`JlW^V7awDx+(&Q!uq4;@&`;-{US#?9eTfzR%k5AJByXDz
z$E>2wV~KU4ZLCBWUwn8!;qFJH^xc1ZQ#~iCnA4NQG?->vO?$9%i~tsYv0wt2I2B>U
zxLK=#$%wVVHHZNAj++c-UhWn<i!$nS2gqSYFcEP3oUC5SYua9hNP$IulKu`7AL;N6
z|7Dwi708zMwmW%b$c0Ocb-|pQwv%L-FM&jjv;OYnvBbHxs^nLhi%J0%UJt#2jhvZ|
zgwK)w(j4U4^Pma1JvYm1kTk@J1Yo#7m(aH>Io4OQ-w8IXvhVR5n1mUQr|+)T$HQQu
zwo4EF<ifupRt%<D!2D#F_;goB6*8QA9=iZcf+e<rg&jjxtR}DON`CU(cm9{hBqmA@
zP>OO7$ITM$aRC<WXrd4Vu^{$bFDF9j2*_ioV2uTaNoIImWdiqpihCSUsCjHuIzV=h
z_+x@~`Hyz*-?H>X;r4UkJ^A2kmOI)6i}>dMd>v7V-P{d1GKx}?*xtrfPejoq2R!u&
zKtl8HKK^t9qb;~pFNDXDD8Vd131Jx^{T-l|>LqE;Sdwn;SAe-?6u`zvXOTxPZf2pU
zEm2=EIzx7kDoIC*`D5*}?ZCPr^6FmSH$FadCK=phw`~^JTP^bu%|oI*mZ)HlcG+RF
zAWjPs8miIZ#kjeVg@?Jd8Q=a7X~~i1HUY$e=zwQ<?*QErkT6(M){&X(O8bI|f<MU-
zu}BV6ky5It(0X9GNfv7bXHiP2Yw?_~*{;JJ(#^}UnnSvDoAld1$p3JA<g>erLjS1U
zzE`jao({OVo7zO^Uq8&I`A`EXGuBr-M={_~u3<nrmhN%z-eJLkjSChOE&zC79~g(%
z;TViNQ2})&q-Y0mLU4(D6iMqAJU9Xm7Px?bGivE?lav14OiT+m+ehx%U%9Yv<oVOu
zM+?Gd|IjYccIohEtYSMDC#eIRN^k*o;j_TQMU+LBVZ1{;p@-3_jeh{G>qr=CrW(eI
zGfHwHwo7=qLPGlYse;8v623rsuS@mi8kwe6siQ5!Qju&8F3iG(R!|l=o+>%gT{CC(
zJeI8-ix+mE)?U4*I?JQpF&Gt}`mM?Gq+stisq;*S1^lzxt*8v;0E|t9vSIv<Dh#@c
z)fQnF>~(@c;W(uX0s#l11sC>#EDeboam2R-5<M=gTtIoQ1qPw?Xc%qqVTRq!g0UP+
zP?Pe4jzL`2b^qt8&ZhT-TKd)Q*HeG?p)l*YhX!A};|oK7Y7tDpWuMh$;kX20PEM;*
zt&-Js|9@@N3cSq<=1orc6Iw0-;&_9q&!KuRz=DFCX&*`mV!Gs_1%IOFGtB~mDi^d5
zP6qjZ4Z!)c(Hi)ZOTc*%kNHrgI|NnP)A2%qq}N-8BTfD7kI+M>I&xwAhpF8+g`ZM4
z-Dxt3w|0wZlQ^?2WfLq>!Q#qkd}NKR_V?THHycL7e5AlNf#<xATu4g+2p0v@KqIeT
zEu_>D_+}6;nApjBH=9*2B<peF;<HY8el|oNr}9Vh05@hi+(nP772K*kEOK0uV4s~|
z?7OR3xtE3OdgFkly;C)t=lvSN)M?a2Ec*(U{w^^pEUHittCfm20kEb8YOInUKPswL
zKLjgXa6uKR7e*8jFakeNCqUm~=NS|L3R}rmRb}>Li2{fN27;M70aa5fNvX1%DX=+M
zmymQlg?2n9yY=2%ckFCF@UV{BN@bVWe#<%Y8mOHGX29IWm)amehze%Hg}tVMIvOO3
zmu12^pemh!%z#x(frN((kn$dwsQ@Cb&Cj4$PCQz0<VvAu@z7vj-iQ;p&jF$7xQGW)
zPV!Sa@z-LI;4aoDo^cFf5Njm+f|Z}U2ub{4iyi{7cSK>`rS_*qq0NG<Ou#=LN!iWT
zxh-Hu-4DL00ZYnV7=-J)eom>4H`79p`a0%$5ny$QCO8){-Q!Bncco!1uo^myn-#x=
zbak7~?HqSDRXToGbA;dSgfKLYs@qx>!bJv4F4fa*-?MW0jxT;#U53+HB_;~DNOxKx
z<rcNuE&&$Yo=1oZc!U|Z4~?c4YK@gw#j#uZ9YMH|=J8ad+(XjRvdIz235QsK|MBBJ
zp$FB|mP3G1qGxH+&9ANxp|y(PdGHWe5ERgm+c23HCESWs5)<F-5=XN8ytaOR>psV2
zF@FWP>;R;8317oUw_rn5Wi@bto?v4LR;Q+c>jtb~#_r*&SMnnoR!2H$p+oI~Km{mb
zc9u#&^x0Aw(@@)kbeEIuAQWu?3eq$P21R8O55Pnw4|^IXOGp%Yve6?xL$7$3{@c1f
zoCx8v2wwq-L>SQaHlYi`#UoLKdOSem(pNW5!+xGKaD2vG&t6LcHq;y7ss;*4n}r1d
zIs!8af8(G&|L)8HdIx1lDuGtCMuL+G7dJ3Hl!}wVQ6eTe9}!eN<2oz>RGau@v$TS;
zDsmenV*;L+Trvxms36$7TcAV(lo*1DQy69rilal(iZf&WdU;pSCUKjE;uwVoDlJ@I
zfGLNO2&hoZf<tr{Q9L={6HueF>Psa<P(r?iMcm$YX_U}nID@A~sF*<}^IcB^l@F0g
z*BmjSk+F$T{i59@M%%v0g$l1O0so?cU<NO{MPa9oaq>9|P_zaQtAl;x&sbC+4)_y3
z5h&<Lhgx!QD+(^a_rYfL1eb_$<m(*9g+w%`Vc{uhYE?Y>IjH%E$Y4>XhBrv5^4+eN
z5XiCsYLENoo1Sz1<HzexTkR&er2p$P?^tXmyAAt-1pxyf*}6sZP9p~kXAMg6Mp-C7
z{><v9SP+C)VW_-*ZXO_4l1nWF5<XYj=cjLMTt~}zh>Z4Vl4z6K2f}~uc{ovh36h^H
zh@*sSUct&2gB9gr$3OXklzIGrI`gR>5*u0j+}kq`+uQ80`&gkXnC#IWAzD}nA|b@H
znM5J712N&@S?t5-r2tIMPLK;!vU)59m?*@W9SPv^i3zVr6>bF}=jo!AeiP3Dh)+1N
zq(GIEF7?+y;K&+|JwhvJBU_YUK#?}!_hG~^L@g~j{M<~N^sDCz|M<MzEiCewqYLe?
zFF77Fb+*rMQ|I3wZ37dHe><`E)80ReRp;z;Wit90nj7C^f?CciiL?|*K-Eee5GfUW
zC?F>7NX!P=9&(Akhc8^EYm15%-C?D1YZ(pVj(P=B1i=QYU<jzHEfG1So`-Jkzj>a|
z6zln`=g%y(+XT_J#Qr+TykoMcvxS&-Va#gp2cj-K9u63=$3m_X$Ld%CISY-IZt4<L
z$*Duh@Ny)AV4Pna4V(<sK`}bwNO+-qU}c}|Oe!;$(ohX1v&l%VacW56aEj6sxGjN~
zg~CgaIY0=M!}ispwDz{Q{_44}oW3B~XMFJ<=lo!ztIckCM~FVXIcnmfB$xy<5dAxL
zwC#OEmFvQPubzk9JF{jDIo|CCmr4Y-0Ny1%5&)s_N|f&EV~)hmYiVCl-TG_UaViaU
zfcUi(j@xpC$1K#Svu!GF;nA*P{-)Ub7i1P-d6(ItP&du|#;SLht~T8!c<kR(7PW~f
z1oQp3+M|DJZ^w>dt89RE9Pt(0kuz1VJy}_Ygr@TN4BQG%M_{-KD3g}_w1nCLp<;52
zoR@=Tc_lasqve8tQOMsY>fRh3c6BUriExaZS`c=uYwC<|xtqKSv6MgtVMgKaf8m*F
ze+N!j5C+=Jg|@@MXOr6YeNAW=Ed3r7JM89&RsprFcYm<#@TFgQW*>q@yZHDFEIJmv
za5Gp^)saROVZLLdw0a1WrqQ?ziw7_)n3e$8MH@81<r*;%FmWr5sU+NDS;l<YY80!G
z8^G}Tg!WjMBPT5U#Ww^&+Rasej;MN4Q4<W(9P_V*SewNxh=|ksv6#R>wBQ40`hOGk
zHULhQce?nSaKJ>M#=ugQ?e;+51lT5p;uzXmB|&ua5+am#U1F+$O^`S_OxIexE*%mA
zbw#)LhBx#elNBh2aotL#hj>yQnV@z1CuXTM(~BnOLR>qzr4VNqH6w(?`@F&4KY?K;
z$>coe_woEb-mkG=4Tn{vqm6%wLI&t~>Bplgs0a=8?4zBu3B;xw_=zj$Kw1bp5zgoy
z^fsz}utiYjM}>1R_tKmX+xY?#f0!I07MgXz|2kneHCuMs$@Fw6Z?h5`{KzYtO;8yT
zxeJV``-fYlq?q9F3FceQObjP+r|VCF5Q@>~sUC`GMqFKub=9sy8R>x4j+~b8PW8xr
zxq{eLtmNSAE5>H!c|KJOJ4ck^I&AXve8f(~9C(_dF`Bf}B%zbtdBbra?4HZ5)1MCZ
ze7k4=vgJwplx<6v%66YvW0QhXkAu{S*(;@{=9934#(zPFH-2r6_GqVm?TS`pc%q)g
zY)elS;!@C+f|hjTzefC5_6YahIh1$->C~aZ!J+6tINI2^$GtG*E&@=iuh1TA8rL-$
zTIbFh&TpdJgpCPy8Ie@@E35v=^p{Uc<k#Qd{`(|SIVri^ylb6I$aLQ<Y3a8jNvjJ6
zmYkBmI=k<`U~V(R@1_%j`n`2yy*}}=-dZ)Mg&pvC+$=IEz0}dly&WUM!;x2lP?Cxb
z&v;O;-FIour(}XljX8!AV1#a0K`pt9EttE}SE6YlC^8}PQN3}1MRDBHH|<lX@5b$~
z`=pd<mf7^ByDWr|zg%Vg+Hr`>x*V|5Qcl47BchCjguS*O{UmkmGGmkGcr~*;1SAxk
zslxpkh<uI)6*7m`=w7?i&YdwfBgTLylow5c1U`$wt&QDD33Y)=;(yZ~e(Op$QP_IT
zH03Z<@*bmSQkHx&F5R&@`Lqu=JmFsXTg0+3ZJK|~N8j`>P2I7<>?M1^5{cFZSe=Pn
z@|rgOlw0mKny*X*s~9M#@OyAF2Qo}8tmKgF8IEGH*xT&rpa&!_ld0OU_v-p-BrD?4
zL1e}Tn}Pm4y@{rQTdodf*g|&mH22~D_m8pp7n0um@1o=+*}wfY6ru~Cy#C+**^6A*
z!FM|E3DMMK#AC9@v7fAH5dy|1Ra183u_FO{{ry+2DFKaw;7)hCxZx7iU5=>7$U>u_
zu|ukGesG0I<g!TJT9k+AaBTJ}2m=!ufa>Z538N<?Jhdmgd7F|>blzigMw6!fDYs;v
zdN{R}rYAq3dg%@1K*dCCX3rbnBB?tBpIepv<`tikqzRzp$#*<;x&G4eX=@VNHC6Xv
zn+y6PljBGV1Ox;z28WEGS81*SuhV!RyZ2t~LsNuqbz@t)F$@N`>WwD6-)J|3dBEjN
zkd%^!u{D{yvIE1*Xc^r51xUbbl5DUbt8)Rq2I&{p#?t1KkWyAjni&NkVj=GEj&sKs
z9iJKKdu77I7~A}vC*aV-5uqWVF%Vtc2TE=p-l&(l)ek(a1w|!{HF~Y3jI5P-eGFuP
zd7npB0~Oz(0rte5pjg-?)cP)a=z(LFWQwg#+wV{MQBO%od_mvB<p@HaoCG2h2`iGo
z1nMH>r7h1VhRfdy^d(x8J{hsU7SYQQk;AcQI)p?#t{Spzii6Uc-CAc1v%X;2iDK3H
z0xHk(+DY?uNXJ`3pIw`8>_f8Btr$jFzM{;Gz|21-ud|JBRF=zV8=CE_(<TTKa|I<}
zN_xSuB<+>bOa`fj-Nft@q5{7kxz6VPY4Udy0<cd~bs+ym-Kin=y;D<eJ0vow&<4T^
zXX+vY*D*1B4MtSeGW1J#9eS<~%4~_bPom;h3fLDAphWhPYcUMkyzDrv6y9TtBT2tB
z(@)3}|Id6h<)gMcr2SS|u)63`(-f15iHGMQ?IeUqHb337d;$I6>4TM+_+%^?QUlQ3
z*_ZnU9)cnP5xT5WJ&92imt2mkEro}paD!~`S7O#VP*%B%NBlC{OSBS1tpM;IWH0-a
zn;m&YzZj5N(${T7*_UpMirA98mOjjgy$gzaDYu|sn}(2p2>;SVE>GzhJgqOPfMmwk
zC-1}=PI`G5J9CP7)8#9OfD<c~WltFlCdMK4K5*(lqU5Qg7KM6;fz9}l(T^pFw89h!
zF1H)c#J=kO6+ZNGU>5RRDaGhkk9V76t)0ognt&#PY=-PmCoD}1jV?V*fblpIcLMjb
z(r@b92j+<NR^#uP-TJb49z|VkRqOx-JLrgaxa2#KZba`41UK#uoQEW_X))K0kA^NK
zT5Gx{1vw$i<as}|cUyrug$r(e*8R_2P-_u+z~&@5Vd<}_Y_=wo0T?<ldtvA@UkbY-
zB$y{KGvKno0q7IH1+$m(mknic=e?Pf2NuJKi>vNC%R?82t6$04*}^g$Fs_DYv-Jb9
z#jO?X3u*~M>|ncu-3BbbuL#R(K}-yPbC~*sS|r166Egg2<QK*9*VzUT0fp@(y{@PI
z^za>b+E+*i0E7^`SFSn;!S~B^(47F~mAOT_(SP`B-_67fWj@B1XFkGvv4MbFt*#$W
zG%tqZ8W?`2qJ^W{@nylmj&d-QXq<)Y3Zq%JIFAKC00dxTt&nE`pkPsuIA|i?WYHtB
zTcZPW#ye#kU*MO(QE3yZEP;;u1rGe<K53@~DwD|6G@cOO;7P1Qw8$Ioz2%{in4|N-
zjySh@XP;y<8gJKQkVX_nt1&1rf+*-dT2yk`MB^Y2xPll)TXt4&c|IlcTbSAODLJGm
zqeg6O+(;Dbog${1todqOpM!N+$7zbV%#4_m!q<ow)4{2@KBk~xgvf6rCTJ~<7x9vr
z$v^}RCSVRZBj#v!jz#31R;U~yUqzu8Kz8oZzz-~<Rvp_LL-vxQo?nEWzZ7?knYOf)
zOt0`f{QDFembSY6mnJ5d)xU||n0V|<vB6^J%sc-MFZ;$5pvouEGC;tS_&OL4gaS3u
zjL)`2fRcd8Y@#VE6)g@z+v6y&V~)aQD}`_A%a#h!l#nJk{#;rK@&Z1!>L1#|*TfKc
z<|!pHo8WArQ54D=u1xIa3wnEgVb|$JEo?^V<ZC{TYXDVC%g9y$;xI%&a81b`TLASD
z2!-Li`Tf_LBtaNN;wIQ6MLn8kZp?9M0+ik<Q?*L@GD$bwJQUT+E;XbHiG{&Tah=5v
zJhq;nf9PI=W}2DUu`9x^MJ^?J2uicw8^_IL8pAPI)tpZ0s}Y>%S-baF@=c}*K0ij>
zx+;S>l<8*EXNc@ew-W(DfCVd<@qt#Fw7LsM;(f8##AUW<3&P!-%XK(bIl@C5^}whO
z>4e-`$y3!wE{7aEOozsb*y1Y|bCW&FOp&fMFO#KOXi<G^xI}9~1mZkrLoVH2s4VkQ
z3z@Xj|Kkxf2fQ#Cd8e`q3}DBl8MQ6VB_SAKz)N5zCZ11P5e^>9W29BT5<qGWBZA><
zxfnu{;OvO+sY(JdJ^N6|jiClDFhL|I=N$XZk$teCdH_Mow2<8n?`K$yrGvk-BJj<N
zs_tUNKXnGUCJBj5m2F-bR~3LPatham$t;BhQ5&Oa%Qp%5+X|F2zItk_VK4v`a_AUL
z)D6&3L@mYj@*&s{Jbq7pK+*SV0q~?-J&Aq{_K@1)&{~oG+_ybxnVlFmaB9;QVNi`D
zuu0G+quKD!sY=gOm@5snq%gr26hz=mQV18e&ZzihsL1RECJAp(o6rYB^#a2tfan?d
zuKfu&{P>D*jUE!T!^3Vp5al6kGJGv_XX|Ap+n^>~3A{Wu{PtK8BmK(OcrLjfYYRHn
zd;uadF7RNx!t`uxp&8MFor}LhvH#Cp+%@7W*@07yE#zq?GyG_^*O0j|nu6Sf0m<n}
zpAws?YurS}&oe4+ZgLkopRg}>jo3M(>Qa%)O+6@c2ZW}FhhsF;?x_Te!)%ZTkTQWa
zJR@al^ywnmZR^-_FAJw&B6}JUJy%aw8S?!V)h~j*cQCrjN2TvYxL1Bsnw?`(nn^Xk
z0FX9Ke7q#58Rd(B3G(vdL#Wv#&)ACjH}2;;pPZX<dNF3&&ki40@=*0bR8?w;9_hiY
zbB67@J+)vi8|$|H1t_CG$769#!5|83eV`9W&%unG82}4X=(-kr$o(Wm$-dz(5YF#>
zPO@{m-RG_NX%c>Zr@Y!qCnPEHnV{J;Mr9WxCFz=diuGju(xl@(Pcb)besU?12$tqm
zzZ{u{c`D!o3se#tJF3Mn(zOgX->Vcaisv!kN@s?+Ck!z(a(u-Y59TW|3K?aC5GekB
z&3`-XRM^^wueV6A`^`@qsEf%eCN}!=GP24dZ#zcqJKnIvlHmJYF;Q-U<@Y>&C#A=o
zTiLvP^Ga`J#pn*PsLD`&JeX0g)b4>gsu3QDaIj`zBXBfv3J>E3_PCtvtRJeZXln?!
zYu>LjPc7!I#0Jqhz)_s7p!bzcv{_;6mZYuHajSWN%Ql~!dLl`Dj>j!>(qc(Zpyt=g
zC=^nJCOrm_rI?gr)aJF(Cy3#rx#pEv7V~`%`~B<|EL0h(gmP*ig1&$lvO_|L_3W7b
zm{1P!K(hs#32*6(BaG0xF@GF^x0snz3=X2vb4Jvj*wboLb>+-(POseSpJcO2t4x!Y
z{Z5*8llR|_%Vs~B{td4=X-U6DlTW_^(IFX577Wg4X0DXB(w%Ovj5>!%dgsdd(VRUS
zd?ZsU{L5yg*=$Ah66MUL8WG-#>D9vhfeShe7;GIwUfX|Jp+{D<JIQI0Z@j<Tc=W^A
z4>8|R%$W=Kwfbx96VvBmNdfPS*PeVMiLTR5$!4+nj++ixOuyJM(e8JV$LViu-jqoM
z5VFYNKAPD8zoav#f*SUN9R~Gx|1^MxWvpkQx6wEyz|d}!wT~9y>Lq8#7)5cV1(YL7
zLCIy5c(%TUZ0CMy&*Ca<XmI$vYd5;mF#P=9Mt&srbLymOC9ux6%_1e4?Qt(bE=TT=
zBr;L+Iu<e$H+znmNP4Se1Ej^R0TB?yRd;%c%}2~wX3-h_hi7&GQUXbKl$Qicf)w;B
zedZ<17(|Dl0cgsw(SVpBs7rb+NF18iiu`)^vVp*+TF_~%`Ol#kQYTMMoVGah`M(1u
z7Ln~1qFQFWYnk-=EXmGfGm+9JuqKKsEHxrFFY<0FnP{OZG&TrKY2;UaVPnqD8%GlL
zmRmO>Aa5SnFgqL(n1sBXXJAW0T8&PvI|Mtd8T7Cj83V*dRr3JrsVKMh_5CQT?3gF4
z#cZ1V{Jn{;l;TMfF~51&<G-^c1G5+W!P|4=Qa&@9^kWT_M;uNKIE0{GpN&vsE&-yH
zpAwTT=@W=P#JMyS(H=2Z3ji8Q0IC`=*G8e+Fd0L6H!1>h)vYn*a4D*Wh_u6@Pa3(a
zdUoKE?88gQ49o-fKm{YQFHKA2_^DECI+=Ryft51%y+ACcozD{zUYcg>v||&~PkO4T
zpM1bW%v)O1+@$J4mryXJ&AUJ#Iz#@=qlGB$UX<YH9B_NR3U!IFhj9Y;<;{;F^AL2d
zf*DXE?1`u%3`ZaT-Io>T)6^7K{eF-s7MBO}Bg(VWHs^t8mW)o}t_katrd`B3>;3nZ
z&JNRQJ52|kAp28#Ns*EQ5Fv}})1o}QmYCi01YeKc+w(n))7~w8@0&>3)Id}dlA>yf
z7TfUqAC2^e;ByX+sN4oe&Li8Rx)D|PMAO$tv^RNh@mc-&wd*`PgUG#|*;uZk^bEv)
zK67!7H%qe+xo#L?pF3%>5sURZ-}BDC*|VCgm3`aa-R3F9rL>h0lVo1JW}2dT(XCP>
zlb4u-V(b{6_e=dEs@RTTAjsfRAsy7T1vP{&6GaM+r%^-dUcbaAA{SRHzMD5@LOFf{
zQ#JKW`27L2=VBg1@qjLC_kDfS7b!o*W+APOr*5z;@gdQlm;A#-vTTXhH+8@Nc^Qz_
z+Py?h!#x%oKXB;B(5rWAg?Ys8wN3QF#K}QL?=N(OvKFCJ-a#Uej-YA-MGA9AS~)E$
z5fj<^;E_!kr7?hFM9HCiGjQl;=@yQ`u^8fshYPxVe!>^2S}Q%U;Wwm5uaePaYf0b>
z_J56fR<B)FU2a}IH&@A%JMCTwgb^{_YbyX*NXDUwm}doj(KI2u<2+I5yU?i?3Vfwi
zN&`w9_M(Q}oXwjd24S7eB}-dwY|xp(X-C+o8p|paEb>JE1m>TJ@SrCn!ZTENYjCJX
zt%V$9m5U^q*-leQuay3SteaBzep@zQPj{8_W=SZ@eihLKhk6W$A|b#L_@A7%&=etW
z--NMA<+)}h<U&$FsuYsZUiV5M4i?crZq{Mr7ppD?4I&vB>tLj>p;r=u$Z1A_#cl3n
zQBx`i(Yadi=fRmCUJFg!ZwKRe7Y;glh;>2Et)KLANQS^prXIEv(oJj$xf?A}ND0Sn
zVHTE!z_hQ02YTiIz3RTB1(1wdRp8zdN*6$$rffGO)ASe%u4VIp29b?JBsHwWsyhdQ
zm{++OJ&KJL)-!@_Q<R=e4A1sd*erL^1B#sQ{mF6H<Nm|HSo@v*<PmxnhZ*oC$nH$e
z*Qe&M2SxfNyI)GK03=}_95+1e3E571%)m|LzoqI5_uy&?wF^o4C6Bd-vrJ=1QC#TW
z=eYIDG0Fo@qp!&@8M0AmeS9Xkk*{eC*E`@KtWVm+gM*onK2XZ>@Ga3IjuT=JN}1JD
zI6M5OyT5<f`o_~Q6urd0ijb7W#;2yY;-ZAueIj*A!ojWf5FxvENOodN`H784E8<#c
zva~fOVxEu=D57&X05wE}s~PCbAg;`BA($i(QhFn;Mh4x%R=X+uQVSL@_BSpE2O-AV
z3Dj<L#N6aFfpwF1hBx&5L_hOA&;ALsY%gSO7JCS%l6V(zYkDWbKcU=@Y(&EUY!;>}
zk=<U|ZnF8j5Tx`rrfw82AG07NPbU-QJ%X2>u)24%2#3Q}`m`|OuGqA*A-bl9qMXHI
zOykhFflRX1H{L2@ye=EeWgxJ(E>uoCy|<$L3dE?%HqRW?D%+MO%^SY)!oDRIJK`7d
z+K`1Ku-Qxq9TG?0JCz3j%Od@wMN0X5?(hQZ#9xcsz48Qdf7NWYmt4@RYAE|X;o#Au
z_+Edf50eb^01T;(I0w5a<d7p>t!|CJ(e8u{_+5(?GAGubx|BO9{0~PLA3Ab5fCG4M
zFeR7sFTeTM=_MB13-wiAJ7xkC%Yq~UGKp9yVxkSwn#e<OEo_mlu}WsVL}<z@j|_`&
z)~Pylrf?KYU4j&Tj&4)QQK=PGd}FK%Ky-kW>AOrUJ>0jut)a9tnnF~_0YDdZx56Kd
z7GKI<W?zci&1?4u(>>*XokVsF{(Es1SvZ*<PMW4C(Ih}>(gj4|{Spr;dYDSvz?~N8
zj&iT$<9&&<dUc{+^K!(xWmZh@Ut00jU{vA3N*E3q)$m&|t&TX1GL4!s<(+9LlM`H-
z(?KGty!5f||GWBVg&pvIQLkTS@>*hDo0dp5qNVRVjDrqW`R&G9TuMqx$;n=sfR`X9
z{E-kkQAMk19|UAt4}47eH=(*j0~W1^-P+1!3a#L?gN2M95FP~}!&(@4z~zd5ipN6H
zUFG7%C05KA$^ZEB3Yb0fTtVYbh&cvyJ#4WD!N5&c`lZ}4d2r&co(Yk48`YJ*o|Mh*
zeFL6n$yzOTq%l`PO+?`2CD-(mvP`B<7q@GX-i|!6P0>m&b@eJKb*E-5yVnZFjO(5O
z?5I@>pH;(NCyVNDVL!F?omkmVtgegGUQkO8sejx6XE^sZj?RoZj|2w~y?og_a}jE|
zuPi@i59WV8%l>2i!0fg2oa(9@)(%Wtl|DJeY@Kq`_44E2l9R}dyu=LCjc^PQe#`Vt
zVx4XmoiXX|_~lUKpUm02pdB_4n{Ql(B;U_~ZH%#TWYisg%y%R*5Olw)mE-(I_pYRs
z9*b``ZR{;9tT!vM{BRRHHi9P|x#FL`eXH2;v(L2iuYZZwa`Gof4*aX<>ksfHH?Y*V
zD#yM(*KC<<ekY=?l+6}|V*rBaVJnmV3JC*L2^R3vzf+DUN7R~kHn95lURsQuFsn<z
zrorci4P`Uai+XKYknTQX2HQrp-uV9-D+qZSUuh*eC4D3&T(M_>{o*2QyL8k&1U=T+
zmz7&u=RDgsIR9y<^V#*cevf;<?eqAWrCaBwt#i#g*3S4AJYRI<rU`4%7@hP}%k-p0
zTID5nzx4JNWBn*EyEVhDuXgW18Jo-YKZmSiw-KquE=70<r1Z!DvBP#8`cs(VHaO5i
z%lsU$06EJc8O$n!MlhjY>dC0@OuH5DhwmxrAF#nCRDVHTaKQ8RFY{&BN0Sq4ft%)H
zRy27zvP*!HfD5TW06tEoX_lT8OPaf&9$eqT@j79o+^${fNv%vvu%R9<?Dk@n;ucQy
z()L2oMnel0un$fP*F*=U(|74%k-Oywg3gNKtORLMJYQLH(Dh8CSYxi)nq`t@*`)t^
zBW?S`saapT$gcO<Rz4g!8BChyrk-5)`1Jx*%1X9&l^1Zx|7e=TmqUF*Qje}zl*0H&
z#S?UgT17+~h+W9WdK?-k&v2ahXrOgIg3IN^WC6$qP{HHkORF&+6cJ}Y>FM^C29e3}
z-o8pB>Uc&0D(Z<?XI#H5_O<Ql+<VLCv@Vyv;26B=9^<t;J%lD_Z-y;%KbS!3qEZ=^
zcD_baK6G|uyI1=A3iB9>6PHz98bs57WDQD#W&Z7^xE>Jt#`8ec(960WTGT2d)<w{P
zM`DZmBy_aZdx2xAJsbP5_mMNz=OTNQg3;@+J?@*ab;H`fUaFQ|qoznL5IMl&+OJI#
z*g3mF6H!$}8EWlbU|F2-cTu0v6q8}<mwE7cr2n|%+(7W94iC)Jk|rc!S_wgys;4`3
z$febyLy2)MG~XyW@iSsd6H}%{_EvXbzb+hSrn9_?8o1u3H!AQu$L!<&8LzEf`@){)
z9v4QB5IzZuN=r;aIZ3N*N5Tek^r9ywjC{$RWWsTpw2*5}#73mAV>2Ga1k`uP;Yt{8
ze~G3|5)X>l0xM(AN^**`>L`9(G{hEOCk8_fL_9D=e?F_^vVZSh6mr7UM}nS#V1D@2
zcQ076`ky)b@Auxz9*Vs5e8r$@0IuaqF_ZMViFPreJueG6eTE5yxqZT$$n$!VR$?d8
zzZ%uNj|ZZY0I9E2EA{JTzn!`88=S@Y>M0!RbL$9wfCmy6QbKgw?8(HOnZQ_vT^M1s
z!e(b<Z3NvfPi-`LSoy;HmT#2i+iWq^gdBiirYnhL_R=JgILb@t$SS7^vAR)$A%o0i
zo*+p=%y2qFdf`|KPc-;r4j2st+=YVvphK^^F|U$76<14?m!tY`voUr7UGRMhX2R>$
zeC-5~P##91;H)Vg6bj6(G&Mtsy>I(PVAhhtkiRMePoY1@S6W1>OMJjX?7#z}HetU@
z&A^8%`JFomL7#z0DMH?hsDTKc)Jnn|yMH69$Q7a-VQ;5P2N>qoFzD5OQDf(v7a7fV
zb~i)Kk|(3!7+K{A(p_k5MW+_H#r@0)T{V<>@4w?JZE!x?%L7(wauxe?>MD~&e6r1i
z4})N6(o54s&c;=gLNFxLF*#-34?y^PBw9@5)N({z6@aQ3rHBBEh9ASw5}sLG)K3{>
zh953#MAJLmYT2!hA;qL8ck26;{H|BNyRY8P-rVMWAG=@12GZsDeUD_PRA=41`Ag>e
zO>`%Cqj8p>gop@gFCdbVCcvRdlazJi+$6o~)<6F|<vad}zKNYkU-G~#uBaiZm)sge
z162VtbE!gJ6k5zxiMuEJts!MU$hzuM%c75xZRCb8={y$R66>i$Mh$FL9yb!}=iUZW
zEIXC@lR?9DYh<tT`zNrjx!4}%UWS@STo6iw#*)D2!w4@fCJ<GSm|OEGDdB0DY7C{7
z6*Vk!Hyv><;08FnjWdS^`I@KZC*%2fG@T87<W@0z2uoqsD8*Qwt820`s$P0g54H=f
zE0aGZAA3<->4uyuTP9(jqREPLrJrP@Lt?{EFSG^AWe}3gy^xkn{N}K0IE)X+k+NQD
z^NCh81i1*~5iUKvTxZOV8)R|SQ20sM5r(|toGj<dg88g+bg+cU&A|rv8fV><l{8Pp
zdeHUioHzt3fMW2aD83>!jb3}7u*sOEvLtZMH1|1u<0|0l#-LFQ&H@av`)C^2{7Bhk
zLKXs&p{N)D_={t7iI3Xx5|KRgu%PIOTEb00J`kO<OALTVR3ok?f!*%sKC35#yt;RR
zPhV}BQH~q^#r8qHb#)qN0+pu7&HKk&j_7TroN7cjOrO*Hm|vcc{mi^@h)u2bL2kB)
zbdq~<89<^*0?yG8o0<l6qwwh`iJ)%bj8Y*+pNgnq*nz;oL)~A2fis5=N`dR50g%Uu
zTLb8?zFUrLKW6&!7{r3P9+*<vl~STxAV7>lSTIjaNtxsQ!uc04bk2HyJ7;Xxf*l)I
zd6$#$?c%ju8iF9;V%QeIF_^liD#ookO_)2xmVSDv8McmG!9;I!>D<3?nW7<MsJd0W
zPl(M8L))G*BgtliwV*x{d=b-6##`*sR+lxKCtmdyuQm-a&H`bf627IA>Dy~KzxY~g
zR5#}T5B;^hr7!=+^UC+JL4Y`Ir9Z!&5X2WSco!A;)5=R!7;Ediq5+k9(u;T4rx5Ow
zijjQ!jz|*FAtu@tnWBYgk{?40qtyor;-G8>;?^rM1$kVOJX0(oLUtu&^fgPxe0@cs
z-YB;EUv0QN=dgEOX^zs9y1le+@q-U+NFug$>WOBOz#zn222qT2zm%N9AptRmqpzVu
z9Iw5+)NDzTpbXUW&2+Zk4#<WqF&8hndDwx?NYN7GhGTCSshfyroS~nj8?0ib8P>D!
zvwQ^-D%QTebZG1aHH3oi8KI`g0CU_LI+xSgoaOn{1;e$wPm#@V?xrv!NDIiQ9h!u|
z{AdyKkgL!hZ?zj5fSs(yO@E9f-8uN_i>o%^unxRG83!u8<RZpoPspLNh-M5M@7ENy
z$d&Qa7#_=W2Ik{ly8?}1Xdf-`vzJR0h599G#Hfms#>c)IxiOYg*;h7P|ASA;?jcys
z2JceLysLwB0ZIla4mJ5<^^)Pxm=HN8;^c-yF8yAwls3svx`ij6wiE2VjqVS6WMHi_
zdOd5kP((x?k!oDK<-W7UM6-=a9zX?1Y;mjV@KkXNk=B@ALNCl|1anz7IbQwQrqKBp
zRKxLjD2^CA%p$2bBRX?ptTV}lltfBkh7WpY7;!EpXNX5Vq2s7qbu}YqA3Zq*-a!9P
zKr*@bx7`}XPbYrJ=*U9m-ZgVBTtW<ZmLIr3>=aWk6-OoL%qVl-U074LuQ;fMT&)CX
zU-S#%4X3r`*^sRK(siSfQ%fL0%z||dBV@tBf$3AS51b89LO!@MW^SI3XFrp79rH8-
z4s8kGM)guu2r`IS(6dmyVFTZ6eXMv<C@{_giXQGe&Z1Uq2^&(>L;8&NFhl;Gj9{H(
zPGg(U;V~c}h3;j}Qz*=5ltGnxg*leqsEi^k=;0_%eHuo<3Q#Qod+zL(Y+lKZZW4oG
z(o9?x$<I|>V#3?SaFkcP(lp?LnGnf{sv>ibt0g?+;^oa4a~*#ZHO-ciZXEHyvzs?w
zz=<e&bxR$K&(jYkKM(GSphu9wP#%);<tM$?muF8Z`_&SgckAE4;=8a!Q+=}=XO4-`
z6$aII_U4e6*nIRVu4!?OH5Gzd5<sVOBmn{poZsDpZ;s^zLzNwR;47y}`YA{XY!z0p
z*)DEUE8|lULmuSkH^T*$8!@Kd6sIuf@JSg3PDy5(mzgFliDvtYu@|q|gE%ilRow`!
zy`a|*f@iX0!>#6-9IQAE-T4<n!kNfG7@r16?WTZ)NM~b%_^ul3f%jehBWH>hcECN=
zQkSYi@ZreNgE$S_t)Cdh@F6^05Ws1CN<x^X(mAbH!cdrkmZH8@TWQgaNHTKlWj)Id
z*@F-YZc8=rhOrBp0RxO=`-bc9pGqPKnh8Nth1Haq2&y7u46w!^sbRp<(o6AGFGEsZ
z>6;foOnM*=-&3m*H2|5%^-=^m1T2CFJN5n?FKvWD);rlCt^~UacSjhM%gj@uFu)~j
zx+*AE2=2vIV{L$fa0Mt7qa{x*5Xm<5`bTF#3h;Wt+B*pX9S;o~eL&UMG!h@FMns2S
z7;aE~n){_!b37bZ!CzI>rx6`jf&4>M1D}>7NXqM?2GEWEgZlA-b9DDrPsTkNbPpXo
z#>4UyG>>(lGJ;r$P_Uz#fvEKGgW{cqGfb~RRzS6>p0p$3^@<3G1&n;D8s3B`4~aKN
zVpt*PTewV8OYUSv!qhCub%B?{#v~O&P@y~s%JAwX5NKR20UK^v(J|(Frs~H2N!nGP
zfkN{|EIww?Sr51~t5E0?vknJ>`n|zcM=iT_a0%ov4DT{xE(5b^2J0u&KqI(w;xA`S
zOnTL8KFa<Ou-ba{>-I|3WHV2@RGRBDTO=sCywKUdONY=wieVRE3EO9|JIcdR4I5XL
z6u2hEVldW??jmd$j?0ShLTnIQVE~oS1an-$<c#=#`~-HTmK;zG{$lcbKm6=Pq^|k8
zUjIFG!>u&!ocQ6oS+r!&?@}K|VDA6QiE`M3^S;ajiI_YaBIVXe_aiz)&dmW7w+06^
zBCo)9O1S&na22YE0@Gl41BZEB$SRm5gX<Y>A$t7HS3nKOkWY+37;g*l;DvbYKiYZ?
z$P#)BA{2k|Y@_+seVzX@BV3ti+Uj^V9vfWEflKrs&NibTt@4B{#6E3PY!=%ET;4gq
zY3Pm4MXWi8RfWNOijHVgrPtNsI&<k}1#5rAynjsPiN&+%wjlUn$3ijOh|QOs2Be1<
z6GwuHrZ92=z}QlYj_Ea2aFDEcTbD1D4(6D?)o6N~$nKb&yu}f+C_Vp!=YOw$6FWye
zczNANf3>jWms2;{A6xq+6{PE(qwY5&c`qMQLCBC+K;a=PBnAt^z^n#AhPTwkHLHt<
zVoI?wjzx^a^^HeXjA4Y~6p|-g9olQ*(hc1Z*<kz_YBY@g3upq*Rmfr&8m0K#bDG!d
zr*^b6{(jaT|D6j~Y9H)-d;Zav+8@ptzvnJ=(NneW+{B_eLhRnZSo*w#E_%tOGH6cG
zK&D}@Ju4`ThvIty_SASwx%eqL*vR%*nA@EdmxT0IKzGR#-JG;3mfJ#U><}zV7-9Th
z+=)UR9EKn2Xn%-j#RKc_nrHf3Yl%s=d!^KWI=0>r_1s|Zhf0edw?_wJxqDYJu*uu=
z1~4=67_phYaL>QSZiNp557<mePZMi(ks`ne^mFtgI0S?io_${q4eAU;lH0LAyL5t{
zXcSsVD++~{f`rUaZZ}vP7m1%=Ct%cJYxRx@MXj!;1<F2bR5!kKp%*PhLe^2TTH1cx
z94xlnxV1)8w#`DCDA|5=jWO3Qat_lMe)A_e<Ur=9GAvX^5$)oF=%RWR?&#3&InKlD
z^}vYYsg-iI;0I3-w9J%ZD<Q6_mm6@|tO73ucQ@45rsuIPsv}@QR9D|4aJB~VBUW$?
zkD`3$J~3uw!?51c8DcE>t&VZp?F_rM00#T4v+hu2J2F61`<^ki+$D&R8%aDOxCq}s
z3TI%-j~zd=fjgbG$DBMyGs*$~%>H3;6lBv7j-VSSUaDWoVE(5lfXeZ=$FB%z;sHxA
z##;&W1@YRSA2^>v#{dK+cgmtLv><&~#VtT+E%?G9CJp#8ZR2ZS$JDmR(FHKISLSpm
z6TvEgi$SY#y&O>mkwtkp^3&Tc;#8=Tvoa%qF$W$<yin}xloLQEP-6OLt9N}JDr-TK
zpfB+1#@vp+@j7_h=qp23;WhXohEu>V<zbTc)70NgOd~Z$JYRbCV9G|Pj$8!e?Hq#&
z$*l`6xn}(u%;!D5WR#<Dt41{lE9}rhkX^TAt4<Xv4jVC@U&@yjhLFv@Htz$51uU(`
z*$mb)f#4DVWr%-&4h|rX&GgXAedFhU)DDtGh@OzJzIhR_cTFyFBtpI9z;&i1AvTPS
zN_@qiWh(w3C259*T6F7%RhMR2KeSM8IUQ=~B5+33a#R=KTpCQRtY{)!DSU~f1x%EV
zVhV9J&y}`VSHWB1aYJbu|2ba5IaU3@+r(Gn!6JJ}LHn-ohvpeC^2xUV3c}mH`N8p)
zu!QzyWztH=u_d!aV&*3{i9@RRxUNnc7-r((7)`Ez`21RSGgyommqwvP9)jfHsZ=u*
zHAH#H&C78f#+J0eI!iY-wAq<BuLYs5PP^X!k>^CA)hL{X$@g^{1*kV6a`sn(2cvu1
zh5xU4>;bhtmKbmJt56rhS`j1^ef;~8&-ooF|C3CzQhiOaK6X`V8wg>8xK`!q?_Q1v
z&)f%ZF<ReM4HujesD$xX6<`yRD#I__+EY|5Aq<~#I4i6Nmq-+hzE{RX;88BG!KXGv
zEU(3`1}r)dTtsec=zkd9>YToNY^-y7@xFThSUn-8J9kRbjq7h(LZL<*cHp3@;O8+N
zube}wh80r!9nXL5hB&|Pr^qRdv(q6HTv~`I97j0U!c;{(5FQbchT?qMJgffg*dT0v
z(@0}Svk#j`8GS>s!DEhg3)vX+7=4e_VrY9K@i^$a4WI_Mn`b;O^8emP>aESVgIkKC
zZ4jMpGN-(<9jO_iJ0keNtB)>6$3zyAjp!A>CX&K+*pbHg_-{3YrvN%^5e?8#Jb}(c
zgvY#e(>za18O+2ywV*R?T6j-A@FjlIR9s(Cj_M^3XM=AWJrxMo$W?-RR+$s~x}v}5
z`LVnnW9|)M^oX5U9MQF-Lf(e{Ad$dVcB@pPT|h89AZ*P7|BA8U>QeZiZmg5Y=ZydS
zj#`eZfd~i!`Wr66!tf2W=-~ckF%OI72jV17mP`oZWR=&coct@uUOd`*Nu1Nz*TjxB
z7<1yvkFvpBjm?QZ6C1x6jvsg{0dP>3U^L;NS0BntyHLQvFNuYuOp^ivA3i~3+N~I|
zGbe;pjl!^0j1`+dXMFr!t*i*n1=RvrDGF>_SQ#uUd7kY@Rz;R9H!E9VOrE6N)D2uz
zT-LZUT<WQ@97lU4mdk8@WJJmBzC=ecT!20KFum@xn?TvsO}V=c(rUl7D{Ue+3%n3P
zh@iNGW5TcNn`5$_NnljN0gK`#Sx$b=82b*$R)Zj6W<cG4ke4HRSyn43ioJXXO()H9
z{@}}lMsVQ-S%@FG&utiva-!_X#^x957_QQh*r5;~3HF>l8hn+{bf^ZR&zO+N|1+tv
zB7S-)mxN=YEk-16kpfVWK-Mw(*1;Pkuzb}egjSj-le>)<zpICZPIXW&i)%ZgDtwZ;
z^^(k&`3Q>nbQs-nq;D?NGHxL;_#~LZ7(=R~DV_l|>w=yUL^3v52_77MxslVWko?s`
z0I~ui2pSOZ4MdpiVxywtY9ghm3cl+q;k+Vq?+hWfxHiGXOH4Mes29JW3Tg%7C<s<J
zsP+`T0;oHOnc8$D)-!N_$&F_~BuSA);4YO;L&+E~b&y^vJy{Hkz*UfvJT~X-Gw40x
z;{t#q5{)sa6Qm)+Kpke66kRb3NqUKynFF|kE->BbRg$P$jEHK)ahwoG`^0>A{oG&Z
zVK;yYWme&~aW2wrMMouL3*TLuE6-Grfb)=hJ2B@PxN}HJ!Sz6HW3TkWFFh5O2YF<I
zC<>laUS=-5u9myV;3$o&(7O15t&v3bP4EXJ)WpsraDe{@?r??*WQLL^E`FSmB>eo7
zzeQz^XEi8}Awt9Q;y|Y^sRfINxX49Z$l}y;<WLW14dnU&2ua{VdZ3M3T`WBUd0>R`
ztxY&eBD#kOsq5h_;%&$yyXb_Gc3S-xl4fKb(57ew^Q!v@#EJVHM<sSR<Iua%_M>)K
z?ZdR6`nkW&*}Oau;UHb+<}3<kQx{`U>Rf~~JcYjzZL{KjDWlAexzqW&;G%AwG}i3)
zHx-S!Kz?YsWSD*mU3Ax;fHDYh!kZJ#7J_Wwe~MoLy!H$GiC@Ug?2{2IK~$&3^L=`&
z*OA}m#w|e#he?#5w89hWo2Pe=C5ph7^b(LqgKgNMLkmM*G@W+F=s?Vwz28uBSvMR3
z85dfy9TZT)g5VT9_zQ#OFoIQ6oG_kf^k~3AK->b^7$)H06UF86&Q|+^7}m3q(Ueb0
zO|XM0>M%m##=Qa-44ns*JkCt00GV~)C!beC4qlFeZuf`0$}OW>i0V}eC)P{1z6LtE
z<s&Hr$+#g1+V#;sNCH;!qup&i!`9=&59#%TL#4*>=Fv{;hwfLIvf&AbWc1a?#@+%j
zNbECrxFDsjkyFdJBUwDj>;absk+noMa|t&B6M*3;hsi{3r3u!!@4YaDr5f}p+<h3~
z+?eGtjt@em<@NtP$YV%pnq94`VZ~#(2=>Pi_5&Z))in*>c(jEm(z>=*Ph6~zK|E_$
zGiB}L(%R`W>SLW==iW|YJp3tnCH!dLQo16?eEY+GlJwhQy0{}z2G;xVDP2@zA$m|3
z<0k&>!z*9@2qWa9Dqb?;`jxGdKp*TOHos|vyJi-^^PuARuAD_e-$f~I7CCp9-$!wB
z!T6mZtNPgGQc-7~5#vjN?=n~9-JeN4SV%tpR?lu2TE}>QxV6y!AE|4|9+Ot}du)Dt
z7+Y~cP%wrrEVtlQz(PB1i6280H+}!6WEN~e^{XObR4;EVD+4eQVfO~bo3(l;hE{s6
z0x;qsV1e6c1qy7gXn%O$a~Io<4`T=k2vrZRTFcx^U;jGkIkiT-`GYTZZ{i<Xk<M6t
z_qz0T@*c(mjc=s<JP_wAU=9u4fWSb3;2Ohm)!k$>yJ>Wq9#%BqFS%euyBbG3z!Xrl
zK!oGTGF9K|)ZZULUcq$k#r?(H7lV`so2WmxRG8-!Qd{L0Tv=FSw+zI7cF!^9IQDuk
zrF)^J&i{O-31>KR>Gt7swC4NV=j40k!|93NupiR`akcB&Z6H$wERjZUi%`3e4@lVl
z`yo9H>Q}TfwkFg4+aOMN7kX5$I`~`T^~R7gn2pUDKnp64-*MnVjp6DdeBN__Rf{do
zAYyKAJrF)P6kHU3H|%)#*xf<b?(>OH$&=Tq^|2O?{A-RmU%Ru1(P?+)5G*k!#f0+~
z;ir&hyC||%EXw6Rih=DEx_)f--~8NDqC6Di8{8ZeA}WB0kX9+KpIC(7$pjxk*r^~k
zF(mN#tk~diOD`HAT1<578Ca1U+33$a6qv?5Z(myU#ZiMUJvJrK=pJX*dF7~HN9PmA
z6(qbeBGOHFy_+&gW{C%e0SV^SHHm^2f)d?HZYD_!HG5HR%AdFEN?r@ORDi?ZD(oqu
zha5Wal?tuI5OVU~-iSs=Heyvy+Xx~Bgc}+^9O{jrI0mOySYItRJivN(hsVG8hOy3D
zQXXKU-G+P8p~|jx2}Ao6rrde$iSbcXnzkUOGl)+TFQAnLmjN<m>+A_I`6&yLSpwO>
zJylRGQYk8!ts2)u$TE<%xiy9Xiw_?v-CnC=AB2%jm@e`u*(<`wd71@P48(3LDhH7F
z@ZjL7&8V9`$S9GUyy4i=ZPXd+oM{{!ZbMa9Q^s+zvY6PcwgfJM?V;mZ|2Sx+3k>iv
z`7waJFuKibrYXO?a1|j|GVz0Z4B(CGMnr`!4fl>kqi{v89Ljof_`u8F8L*^`ES7G}
zbtCKP4v~Ja+i27fzm?ojsizWtzfw;$>-|G3+7rBMK9rM@?JJt%darg)Urc;FTrYH#
zv>$>%rH8q*qIcJ)pT>tb1ayQE%==*uc;L3RJ_*f@{azrGCq1`G0E$4Eg)#6@C61vo
za7Ms!6f8f8>>sOJrpS)*b9v6jGHd!pJZ{!8XGrHPk9=&rgg#6T>Kut3nYQ%4@re49
z$2-W`3R7^sFc|F0hs+5UzY0s{erac)CJ@#Uq5k}67b^_Hfj+O5%|7{tsRU`X{keO}
zVuQ<(&3&Cxp2waW740+JLVbf9X<$#4*!bNcDjzMh^|wpzp@h&;Vi-EW|5uDyZUnf3
z`T%;O2n(@!VQ+ivkGoDBcz?WJ)Pn8iW^wgar**eS^idWA&M~u(W<fmN7w|EyZngr!
zRRV$$e)jGSqU(|J33R0Ja05Ba)i+;0eL;;_k!U#JE_>^(U?##DB`GM?ppJ{k4%tVQ
z-hLQK1OBBHt27H^Q;?SpiyTb>$VDdaeV3!TsV^q<>ilO<w>z8h*;N<csH4m@Er#DF
zZP;N{oNx5!2jIZ~0FQ^JqI#v6zat76pM-1gJatM`Ieo>}6}_PkEH~c`a9Xv~K}IB;
z+Hsgx6v`=(=>uWhvrvaIHs0F8F$k>mpm^azt+~DJw%)r@Ne4fS<%4676pY13675bH
zk~v)*@E^v^7@q)_BVJSqFKC|$1eXRSU48J?Q}Y=nr~RU(1q(%_+I{NO=MQGZ%cq^Z
zcW@x5M>u@fT)Vp()62>rW>|`Xw-y6Ph^IX)azNZ@J2!B8u+pLfLBx0*fxZZ>(BX5h
zuHAKdoDH5hkJ~WdjF|D1=+qm<4YLWc`xdv4B5|DG&Z%8&g53W6XlU3=mov$1B@2B*
zc4-K1@q6LL^+u-?Mo0>f!xQytTw@5FXx^?P<{CA#g;r`_I0?T<`wMkQzpiwQ4-6$=
zgRwQ<AHiwixRU4;AoWnv_`kVEBx9ad;?;V)h~8Gm*c>BiB#Z%p<cdtJKt7sHh@U<a
zTf5%Apx3Opi1~m?T8Oe9vO|5b=?96AaA96`JntOva3U%LNU9p)7`V~k(zs>j{0PIa
zOr!`6<bnp&6QfAkMwojTwR>>CnWXJ(foXo!onND$@->avSN2^w)Vd8;@NrSAF*Fc}
z`3}H2PKoIo#sz4qD8yfHMFGE|uL1z;#BW$+6j^C2!LYsQ!8bu8;W+ngQRpf47*Pc%
zs+HZ(J71%G3N>fs*`Sr0#7m3&;g5XtN7)!8=E=r9r*}9xE<~}Fu56lj(49_<E5Gfm
zi5HBGpKHx&h7W#9RNOlRMy`QLAD0EFZmri%8ta=K#j&&6k|0exu35|Eg(yf-UHRzN
zhd1JqRj2M8frqJlIS)tm5<Pl==v9+6fk+{Te!Ux;n+EgfzgJ>z5C*0h)Y!r$3{UDc
zO8oN%cLom=>yh}=u=rrS71;RE@NL<!XMD|~Hsuy61xZa5dh-&S$zvp1S8WSgY_;OM
zg|NJ^De7q66oCe3AaV53ZQpnZS1w&C{%ro?t;mLyfvQU{fjZ)7VCk6m4qxV`9!%P?
zb&O(XU!z!ZUM+qzK~Du_f|U?41dsW{#-kLtFcab5JH3y*q3z2CUOn}9N7{1QL;)z(
zwI$eu0Cp7@><fLb#_q8OX7bRT@qgT8qA4#-hM3`%4{ii89g3&|NEE%+PC!vh2&P@u
zHJJIB6&*B*WGTPP*w%!Tq%rw0p6#buNG+KAP$^tQ7!f&5>xW{qH6)8~HhVi0wcB|3
zmAC&c(}vM^3rwWKP^QlVBk8hrj1Kj|zaR_NGtHHs4QKSCdDaz{=eOZy8}Y`f8$b&A
z-Pou>gihv0=S7C0k`Xc+Vof5NVcLUUo0u<MPzDELD5HAR)*v_fnsreQz#Ho0F^H>K
z-q`Hz32M2Emkt%{w+z!X*c>0=C~n@0ddXAnf@;NFnSKl@fd!EHI}Y3Y)DDpfK}p-T
z{D~_xDe*91c^H|Bh>3bvM|I3guGD0joPaukg6M2)zAd_cW&#KbMq{+t54$*8Fd);0
z))&Et7;7FCn7V;xJh*YHcFTCHF<fsSqG^k`CtGCb0WV1g^}wG+<woJE3V8(c0oG&3
zl~Jz4-mw|Q?eCjw$eqm)589vtlUuJc8^Sz79GNWPofurhS7dxhbHmsM>jw}1L-|2N
z5NCT+3<LR)@vaIh4A&o=CBU9)bj~Od4UCmQsOfd@Pg|s<iDaCH*GtG$E#!u}1t%QR
zAH%gE0+tJi3nTX-#6kdX+t6Kjb^)0r8UkHN0SCm23JME+HV_jv6Df&M?uYE%Y0IvZ
z|7r%>#wWK6Z;2|-2qN(HptA%O&WRia6Ls1z@PG@upn(T~eY2t;gJyhWr}Hvyj!fJY
zlHh!WdiAV+$pss`kgpJGVp7sCrI1z<8Njw)e87`-GDYj+Rl~4#2R1J`nz!1B8SfF`
z(^GdP{jgg1HK=r7!Eo~$Vc&iAbC?o2IDjrj!wElZ)}M1f15{8hlBc-Z&FulJ71t5(
z%w9>%Jb>7e7TI$DP`fyEU-56+hgKw-m4%b#Aw{-XWHSa!khC!}@lSW#+#0kEVU`pv
zj027?s8w;I1EWbIaW6XG)&<uJeWxpHnj=9S{cJ2(Xdh{fojG?D_u5}*6^r!;i<|qJ
zt%HtTr@Od<_`G}-0_+k&42?`8b0*Jp<dPt8FP$(0w}l<<i~4yb1I4Sy2)!;R5*@f9
zNRBgd{m<{V`)O6=4{8~IBO213Qr3^E0bpSNxh!q~Ed@lRM4scj&Hdw*)36Q^2{E$S
ztRUX$^$RsdTUZjCzE8Ze!7Z`qyA@;tFJTn#kHF&PK6z#YdXh~M%s3er?}9Kp^_Egp
zih7yReT^irLzl2)?jzH*4YrAbr69F3tc7afm<tp1gylpC<V6t<AvdtYj#)}E(O)cC
z$6HsLEhIl;rANH@7B(0}is(YN9p4p1Q~6G5<<wo(@iXG;eGR$_A<-w-{{>PfB&wf}
z>PE4CsR=|w$|1p|xLZ_hFljpwdC9aF|0D=h(N;%z$)t?#&`S_oWJVbn-4y*ZJwwt)
zJXuqI=EA<=M!yb%Xd0MrLI+r()A~R?(Jox-F#;bMYeu8_V6kJwX`0E_&Y8MMt;P|5
zfykV|O<9TThSsmB)gtPHw_DJ$0(Ni6LKmP=VDduM#5(rS61M+FEHd<B(@g*ZY2`98
z*&gIpuR9Me!a)51Mbq%>am-j1G8ZLv69Uj6rZvjA;=CACI?qsS8_8ptc~1umxJBH$
zJV7`la9vUa#bA{K){-GLAYEK~70MgKOy-TaM^s0p=wX-<Ob(j4itPA-hq#vt=ZmXQ
zAzBorE;2%TvdrzN1$(kV0J&_qG{ntB=k(Fyxev06r*>cdlu;_SkwZ9(+HXL8M&L+n
zu>x2zu$(-()4ifd67VCjS!Y)*@MVINFU7ry&N71kRSLNnio=m8%G@O0V8wl?Z)$hj
zOfz6sVNgVzukw;Gz~PgFx-nOo7guA{I=IgmJ6LE$lJk$8J~S{DF$ZAu?3$gMT0hJx
zwAs)HK&h8*cKk{~F&&fBUU~BR{~uM#LhS7GUJJ1yNV{<~ml&h2kB#ew=ni1sf1H=*
z(ezbN5SL&5;J1HD2Zfba<9wL~tA8FxIAn9-eH?7T99DfKyy=vo90@+A<kw`+(YH5m
z<m~nu!z{!g2Db()r*$sw!D+Aq>@WKZyc(lPlWZWwUUjr{LIQTS$m|Cu*$Pz6x`;i0
zet`ePgNI?uvuGN*FoKW3#^Z1OsU#9V&4_?Xols~k!o!a3Ww&0AXCfaeK)6|fWeqE>
z+4nD2mJOb{Ch<xyZnU@ERMwBQisRMM$_g~ds?X_T&$6#6I<J~&5*g2D=vu%E)AG->
zd!Zz{1K=PpSiE?MTpS<Xit11pUKdFa`~&Egs~aExfPsWeoVhU9$IjwfD5~8HnMI3<
zaj=<QcJUSC*q+0Px&L7>^KgEzksGOxHS6#1Re08LRurr++*k>(2Olh2_qAZ5V|XbT
zD!WCDyy2n(n3N>n6iEw1P|?9x3Gs>;92B7p<b@0wss?8z0So^xug8D$!AfE`2YJYi
z!8w>c2QuLW4t*J$jZwmK1Or2*7mK}BN2R3j!AgSP*>a#T)U)WQpBNl~Wh*((7B|ZU
z(+$DmDKHx3wP(v!ZMN_ek<x4z{42?3l#p)>0}kXgc2#DO9^_@X4?}}jf!Xo@iu%Ui
z{%HwxN<94EH3XuNi<cEG>^e-ml3fwGNXe~**1Gj#u^h#Yn@kV?!?d;N)zlB-E^)Ju
zi2N()90(Tm-Z}GL4Hla<hMS`Zgxw7BGNU#ClX7YB0L%3D(%W>SA6YnF2fBLuk7n|~
zTKF&_67B%+#pZe&|8ZHF>S6qXOU3XTGE?|7da<T>+U~OlT&Om9*}-c~URvkLkDs}l
z!!pK=M=I<^C;?*y`7Cxo!q{8TvpaCJ{A%zyEJXyZ(j8sUXu~i{yvQE5bKe5=Ps|hO
zuSR~}Xk0?y`us;<SDNoi+aFqMPRj7sB!PDT6sB>$^@-?bM$rmGhc0h6*)eu=J&H<`
zY+J6EP}own>QJi*?RRW8R?Tf333PZL-s{j$&+n{hUlzwR$9h^Z%&_~a@@g=L!B+ul
zYN3ueTtDoW?S3geC&%qQNrE%Ue3r_Wl^1?8dg`sgm3^~c)OP!#scp!V0Z4+Vh-J?E
z*ZtRIs+O!;Na6NnvS|)bs4!M?T%M_@lc@$vVh@EIm5rgu)=j!tg^J~AwZ3MB#X(xU
zFP?dfR#rSX#BxUX2g<KX>+?t(36rwO%+S>_Y*zUI`<X_;-`Lc4gMC|Uc4dD1(2hrH
zvF&>{Y=1)fsW-|eBR~rWL;wP^f1JG<WcH>Bj=8W(Q>8g1aP4x!W|t!zs)VJ7qWn7u
zTQ`e7>t9~%^_($~$Ml@4w_!D$^;GaRh@`@k+5AgoS77~iI8fTR^qR|KN-MmOp)YgR
z=eF6HoMOv&Jr&EHoOc_R$VW$%U3=fzwS$s%-n36ndZlz15WdN*f5*(!GM&Mqpi{om
zp$F!YZOCcdS|Fl=CYVE8en>YKD#cN)ai$$<w)^2bed94VNPc9+rVM;M_>sXdm;PCI
zep)0+7<wF1R9o@lk!4LF3La?3W#pu(vTM>?4c%@PV*bg;uPToZ-8Vf=uGtW=nOsSj
zl7*OI=w4~-gP)XLszJB!j0$!wIY0{<d<~XT?nXuo)ibp7VQp?b@xQ_s{u_yRv80cz
z0i~Y4J=-^EAp8UeA#ii=vve1ilISB(KgrHsJb#-WxX4Iy()Z`Prze4F+Z26oDgV^e
z`Ja-X|M2#H&j9m0MJF+D+B-MpC21=l5kT6x{<kaR+;DF%MYL*EMQRtIg5d)Zj63RZ
zHb}{E9lm#|w!)?7K8%gkS46T(a=g_(m2g%9=sM&wu_Ti?x$ap(GC8|nn%u&Mdab_U
z-&W8@^(}b9MUlywQF!e`Gsv<>8ESaihfZ}~_sYb-Anlca$7Yq1K<}qJ=YL{go4m7H
z*=sMOE<;x3LgtQ9Z#4jdil~Mg>_k_YX4e~;=Z?hkub=M8C=`Cwte<{y2n#xDC=pru
zIJO3<|1$C`f>VGX_?s`GzjWKh3rpZ>zHeqCl3*UotlQ}J=_8h8boYHPX|PhWNlN;1
zNeBjigpW$ZPq+5{dl{W(HxGFAHz@}e@RH#`=pGbb6?H{ZE4e2;OxE6`EgRENWw)K#
z7%jBM{^gP|h)bRd`lyU%gKY$_U%K~Vb67wMHIbaO6#f6a&M+Pzcg<^J%0KI1Ja;%X
z*M6spCU@9^1dnpRlp<Y6BY-U>E%Jm-{}b6v3A%FY$4p{E3X7`@J|Eq8PQi_6r42L?
z*Eo@XKT?@L6uiB7IuF*>hw7g!Y$b+EG$YzY`iKZzo1ewWcA`3yO3)dG<A+bJ^3kM2
z-lHT_K03kbm=n!$!;&hiFv=Fp9*zq8gYt-!Bwz@LN;ci{Bn~LKJ}#_z8QVV-HWU|k
z@Rn$?p4GsgWtyrwq_kCfW1!DYt~-=z_3y4+*u4f`e0r>TEU9--c@P3?^}ka8s?Z!k
zZ>*skeL3Xyl~q=Wut_0D607i^@=h7NGiTmvNxg-<2<pBH3e+e}+6;5TFGHq}JdFz*
z6gI4u<5--LNi@~-j@HM;R^oyR<y9568c*+V9QaKp3`L>Mr=Zg}HgxV#0h2Zb7DTIr
z5l{1A-`k&(XX&d~L7mWH_Q(<|$Ro*eITYERPA)j<pQ2BkV&675^F;yb9YRt*IgMYi
z;-B~fP&ThT;gNq1Aw`ifO$mibHyRFK>MLSyKWG4<1hV2~Vd?3oBI;d6xLrVE0a}K~
z;mz3jyHRW?MbQauLx+n6#G75a@bQ>!Aj+^_s`U;3o>;Ip+RUQ+h2@y$PM_#9m*!6R
zu3o}I6UbEnlRpDy>Vku&ANhdPVi0D}zF1_h(TA4op}pale7$=RsVKZda1~Aka%*qw
zJ%<TX0EMw=)TCl!AyNV)PGG5X%?w1ts9GPI`<b*7O@LAn>fW$fae8E`<Q6S#`;Azq
zPwb^B=^Js^yV9_29~G!fa<Yp^m=~5G-tgx3N)V%9y73sr%JQQXV;y1hY!f0nN<WZR
zSkdprx+8~LvoF1>u{Ls9WgHV5jm9gi51HB32Vx5>L6n1k?;6GxxZf@ol)#BT#dc3f
zS6nXzf++H6aw$1Baqfa7<zL{TtH|)1Ff9{ACag>3QDCaOeluf)&AEtCR^{OE#W%sI
zIGYSjA(YS&ow;}zU62SbyLGQ+USHFIWkG3`H5b_}LqenH^16}BD$Wrl|91UDwinq3
z_wqG`%L1l*I}nkn`IA1HvWR5iX~*NKm6W@93C3mt4|EY(fQTlqB0{$>@b^ymFzSFF
zF%$%+VI<QvI@H-n%|j6#mKAcTrPyR;{Nb+cp!%z~Tbmn|pq^3ON0hP4YkClelWvT{
z9Px|8LHSZ1Dqao=FWPTs@>ZV&y5a<z7f3{sS9#@aR1HK93q%YGBm|%ss{F2)Fj?RH
z-{|x&rid~6@pU~ZM$LYjG|_2pv>-Jjs^d7WutU9f!~H+X<}t?vc4G8(5}o7Ld^Yyt
zk-k1X=t1j!M4|QSFYg?LRd7lFzpT9nSW`{cD2fP36+$mklF$-*5u{59q4zEz(n0Cc
z5s?sDC?TPvfCLZ_QHml0BE2gFREVN<jT(6aA|fE~*}UKR@Av)pJm=nXpXcs}nLRUm
zF|%g&+H2ODS+hcZaX9f0LiA>I*})R@W|G`=bli-<ECkT@3C<RhbpSd2PlQ;2QIfdz
z51rovmZW?G+*1xFIXc}H+TCEP1Uh<KWM_6rCs~Rx<p;!_6W)^9p(P~$8aT5w@XZ~b
z=tjV4E9r4rwrggO3GSgCfZ7raAf-`qNRYY*G*`gbJ_Wo{Kt8kpx@{<#mkrRu374c}
zWB>>oP#L&@{`&(Ie!{430p(_N^rjMkTySzROX342-9kzLEBp7Mo$o{L0lr5UjZZLn
zA3|^k1QuX^_5;pqeVlYYM3I2)UjUuX<v)GPd+(|5p;mPO48bCnV5Kx?^}TO-Kz|Lu
z=oTa3NPL+r$zXmF7#-;=W)j@MZ~%=lp1dN$DG541pi;oarzE`@12<3(000O92|9*b
z%o3>vuO|H4E#yl8O11zPlYA%HA0QuY3xyH@E%$&4fT_Ru>-Z1A1SL;CuwgoJ(5e5r
z=uV$hLw72I?$#BM?L`T$g4Vy^0UonXpE_m0_2aC+zuokM`&z}tCaUd?pPZd6|1%@x
z?dYFw^w++Bx?KsM$=`JJ{dAO5`6+^*x1Cfw&K+yhMbnKL<X+$ZCmOEZIy^ph7(FSa
zoSLPtn@Tq;Wi3<`rGr~`IhIE>>Cx%Ysjgyk6a^dG|IcsRcK&O`sS)GpuEu92WsT2D
z`2<wgh@U(h-Qxcv<MHG73BdpL&|?alQyz&;`*+APnBo59;dA$2!S~&tA7TEZaG95V
zmLrDKUH=5+RIa0C*S~x8AE}FHLHfOoYNzj>xp?-LDBVdP8#+KX0Z^cTHBA}L-cLDY
zb6pv>6~L%B-QTE2e|H)v{-4=~XJx#<)aa`|U2#~N=Ct|sTI)aZtV!jm|Nc(^lxw-?
zuhalY)wHo*88%6Uch5(~`nvLehW}RWY2MFj|33u%=$!!As{p3|=Wv7N+qA=-eVz+H
z@BJ5`?$hOnPgi{I{J(|I167Ylv}V%ug0a`jQ~x7mHE5LqS>SeDxVNkgdzDgmwEl+m
zdiB}Fntyj(<-d2+pgT#8_wz|CPTukR{L3cb1Otcv-T_A4e~M0S|0{l<f7--4yh?HW
z<^K=Vjn&~mgN|SO`rB<Y26v{-uUGRY`bhoyb?!Lzl&-*{Z+W&&%EepN_+QD(e)%a^
zeo1Lf4+Dso-FsG6^jVF&3iywz6s>aBYvlzEdUT0&G+?1B;plVqtB}9&nfh8|g6DQm
zeg5$;5FiGU2Fc5D%O9_+@-C=>t*-w~hXUIKNNLR&RF9tPOr!(odECKc+R~vs8+%<v
z#X3M5Ha$FY{@(#0P6xMwk@My3|J?rT0nKPp@LmTPdreiaK@D<e8hfoAZ*_eY=L%GY
zy{>Aoat|ooJw-V^^I7-!`oAMJ-Z3>5%j>GJhXE*9{{JrcD6N#;)5>~*_9^=WIcCB7
zmX}*zR|DYvRi>x!X_Xg^si~HI3UD|#J^8;r0^0jO1^uTMpOTZBfW2PUf9g6s(*raV
z01Ny79`qgoJW&EfM^8@=u)S0NlCe&hA}GT4it@!<ToT}df0!cu2~%7{YpV@OZcL4v
zN}!pcs!oyfq#`A?yZHld6rkXZ3^slR%?zYo6)PWnP6eirs)PpfvSiKulrm^i-o^Vr
zmNp1xZzmsV`<M-msoykzWI&UzJbMjHjxc0^{&vOv)Q=Ni_&ju^ci+IS72byXMgdfS
zeABB~!qs(g_8YhYl&8vbNH~x9g0FFi@4?Z&?*XN8BY(pUW$-8vXpUee;|FiTeOsuj
z?pP$4ZB)Mo?Ws6i5&RK#@=S6Mzz*6XDl3sE>U2+Mv9ZKp7-VFvufy7+as)iCP)y|c
zjaDYT)s7)WzRiz6GGyQ~33eJYvSI&u7tQ>2#>wfPE;TqfwpQSL`J3i#K|VK50i9_s
zlQ%nNGW5Oo0hw*jOa^O4Rl3L)ZNS7Y|4{Op1Nl<quTQhqJ{R6h7z*<iK7db0Q$aYV
zvae>tj{v1r>%5ELK_P15!cP}LirLy727)!C)2Kf4qD+TD*)WR3X;gbH3I0%p%F}hf
zYb>b7=|L=(@6wo&$)ZG_5NnI?$GdrMtPf-T2sW?b)6`)lA*hOh%)NBVsrXf($J*fd
zHob{yfn0kV^}@s4fq1gy7b)Obs^quLb-!TaRdp`ZTc?@@{Dc9zlV=X-qZQrd3nqyw
z#*|Y|pO?}4s6$=g)qs|(2b^Y-D;@8D(%WH{$TPkFN&i9{_Mz`%^Z#40=xUPIK>}c(
zn4y|_7gpfU@<~7_&S@O`{NrOo|ERVwb02%sYtV|3c=dEj!{<Wr95x5vK{LtUeq*&W
z`RuWY57^@c0TO~vN$ck(qxDTH*l(t&yotOjNELdNn>0<K8Q<{z@f$bu$-tnD`qNeL
zBz)Q5vaxfP3A{i2<+tRWs0Y?FPT6m^Z#iYZRyGtIajjGm4H|$CQ_SYx_y7&q1m@K^
z%RIHN7L{_XtV(xPnVmHh%^CwcpPfNG5CsbKP?{U?0W!rYtz)+NOkj;Odu$Zn73%JP
zE8~nspU1!lsI7(!ApoTK>Q}p7urqt&Yk-orm`@7Gra+v?40Ww^V{P^ADEL=Z(SYZp
zs7O<f`?WyPGbhv-%8((5`Cr*)CLOJRf&T9jt#wuE@PYind?XORLOc+p0+`~ywa+#)
z89WLHnkK|VD>{{V@wXnxCly`G0x($cuW<}7#+{d(3p&YVRphamxLb((0Hg}9gI`XP
zKk3|9FX{m>66A5W%w8aN68V(kS(0wo7Z?ciDK^30qIV2Pot-(!oq+#Se3E#MYD}+f
zss9I;uj2#%^?ZBY7G=z4ntx#>&k48Q!S(ojXWLP%4}V6VLc&hl7SAJa&^20TiH;Uq
zdz-vh`s22*pcg<<sPEgrjV%CAu@(kU&p(xY)sD$c5NzOTW>O(HI;9CxU&9A-W3TdE
z88ujP9S0z{89?{JQ6s4TiD1w*TU^e!KYcl5(&LjFlM>AqW_#pzGtoYa;&P&Q(wB|u
z1ivzmfv@-!{KsgswKv5D765QB*lr{X0I9F|ByHwbqcGO$WC2sjJ39jY0qY*gMTJc>
zKGAjX<Xz9I6Jq&S0N#QE<9BLolt6;LWo#6GOjt;AlVrsNn|fc%OY(;N*QI4+@vMgc
z1&pGZ8yc9?UG<TL-r$5}UC<(8{c<!o;VAuM+yOI`zznU?<`&4|27Up9kg$@wCQy~L
zlTM)@9Q`S554_x9NyYoy-0B!br=NdP_#>}!N>;S772kAuW7s>$3~51ZdKLe?$$xxk
zG9*c`+0uOW)#RqxI^yg7a)z)uGc7mCDxv_^-eyZ5RhGV1blp%`Tfjg5yWnF*r*%gz
zLvdflbbOM|cd;}fGQcTGw$EJX#Y3roJx-#6mhYeioZ`gCMi1o75?Ps{EDWG5^m#`m
z-UeqxtKB~K3v0X1c{b@isS@L$^a6Kyca}#p!_i&DrtA5Q9Rc`?^PJ{{WHmsaI~UV8
z9M@Lz{glA97t+x=dfLOF0<;0MM>F2P7%=&IKl{uj8$X9u1r4T8K9cm8F{<KKbovYt
zfH&C)NFLus!;0YJ01b>E5NsH@4Ee6|WdTzT_5*9X!4i;_{59JYxCEe58C|*>AEWhJ
z(FxhNN-6-DSS4c93HBz+nw|kZ{%ilGjok^#{|b7q5?0XaMeq?^3|c){=5BhbS2{}N
zZG?#rgFNdITqXY{V(`8SuJ-&IpU+r8jk7#JfB!vU8S5iZJ?>Q4VcUCzR{>t9Tcy<n
zKlsRzjtXmsjtbk@gv>?6Ha2<#ZuiS;sDrEba-JdE%!U#lFtI?Jgoc$23VM$bb*10u
zP3O6Knn$vxz;Q0<!ZFEBXlwfCos+uDDYM3`K)q%E5wdleoeIUjhWMQ@fA~cHG~z+o
zH`!15vis?7D&#DuLa`Ca!aKP>{0W@beLJ{1#*=P~mZeX&+iYyfC-H(F*EpXbUOK;2
zJt$B$Mf+Dvms7wAF&88iu?I(^gp{Dm@h0C-P&EL)j_F~=NgCQr=7|=IOgdY8H`_qd
zoNbRB@#sE4dg=YWifHC}+ir)oXo7WC`g&2&_(@$y>Fcj4rcO0mwuZvp0=Zl$MP>=y
z>q0W(!6NlB{3R|VpZU#fvxeToD1ckP1`1;HoLD=2U;8Hh7lcu6Q8X}}%>q!`<_6bC
zE@0p<^QVBPzq(#?HEZZ7uqi+7)CS!kL?h(~haRL2_5dYM;xN87>qx{@TX;y)(p6@+
z^zCr@q{jfWa1M#R#1}B;U*jAS9RkdY?R-I$#TjsbK=QSRQ-~Jv1wXJq|1t3VB<>np
zi}916F6?dF^}Upr3LpkBTMd2Jg*t-Sg6&D;K+%U3K#Y>-0yIj%b{V)%G5$-IIt2oA
zsTkUMM@pnB<7?V3+77JJwn<ka{t>AF@9+j(xb_6dW`XW-&8}Q(*CETGcaW*0x^Ufj
zYLHD}XVwkE#W$eE`F3V*wQ-lkP>$RlfyMBI>~B#$n5DJ^6{#t%cw{~&T{nv^gP^i~
z=_f{*1&z0U|3#s%?;O32%Wt+nPAR_`f5)-u4NK%&&>M4LHbJ{~Hk;+#v4~9`^5@Lb
zUAr(Oj}c51M|Yq;%MEV90Je5C$3%g-%UDVu_DcfOhb$$ZJ?KaLo(WkzarP2dR(35D
z$j*-5iy2BZqwMOj0bs!2H*+kna!mqHTh%0V6TdfpC}+%ED|D&Q)XJ<*y5VWR^NVLe
z?YppvkCWfC>Lsf%jad~MS|YOt89Q2DcwQ{cU9MwsPj7T8zydG*)W@@JkBi7aGOjl^
zt+d>=$4+_%-(<xXn%dmLNVir<Oaeoc7gm%Vlp?+pdHg8lDCuuHXna$WrDps-uV)^U
z(`-3SgVG@U@|pNm*kwl-|L~IJj_-IL)2!U+ynPw8K3AcrsvpMhZLs=48l!1rz3hVt
zH#gTuf#j+Qn&3=jJO<Bz^aO!8ARs~ID;0ymiWod7jfq`Q+1C;%7(mxDqS*QP_;~AK
z(HbOUIhl2gMDXl<?QE5LJ*==^{O<T~$z6G6b5-_hpQ<czZD2`Rry5pIdoEK9SbYFZ
zByvhWVZhfrmkV<lp5}Ei$r_k&s|4-2`0uF>XCsl^fC1}BDd5j#DVJxnsE;X`MNQg^
z!H-4dZds@Wbx*vYwB$T5GZMufY+;@&c!N<yP7!OLWnW}j>(jDj;Bzh0q%qH?6FYrD
zBzp2dj`E?6Vb+OXf%loHoSdrPM4?_e6u49aEz9?3Fg}*Kxwv)?v3Fd*^qxQ<K<CmJ
zjZy4^FmwCjqZ($zuVh@wkgwa~`?Rx6=|%@4A_}~H$fIoH!{PKxS@p7?jbmy|+Qj|i
zwiz%%Id#bfmkN!z3_&1)OG;i!i=(PsFKt!~9`yrOA;N#t(QLi4|E5dtGTS;QJMI=#
z>pzfcX^EA6{E)K8ckg)I(^)sKm=9mY<YM`7EdATlT$&i+PUX0g%1;3eP?UrsO8BzJ
zSXZf%*EU|m*qXhMTLj#LaX_D*v%bS7`$CE`MXE}V^XbFd?<=LgU<{c-nNLn0s4Wbt
za$(*U^S<CEVr7Zx{=9y<nQgiSS^r!Er@2h}?TfigU#CLlJLPLsN+EU)r)PKPhP=Cs
z-BZ=fbnu#f3XAx&pNzxK6QJnw71>wWf-s9FH;a13dd0`}u$wMxi;F+WDqfkuJ6qmQ
z!hQZCJJrFnD;uJpHh(zxQ|JOupIvfZHROjhPGvJNz+#Pi<)`w>b31)bR$Wt?5(#$Y
z+S@W}qLN_`Sxh?SUVQm_O6}lONO>v5dVO5?Oyic$r8)Jm3Si88vZzV8Lx&FyYeX9c
zN&+NWAhoNpDH@fzVoU<1iGre&0eUpogWXg!MQ~+hRjj*Oftj<ikO&@A1PByVu~ILm
zsy<p4c++CDJjxn-dK)+Bj(-|p^qKzhMEKo#LvG!?b2WzgOugWz#?d;|(RIjC<#g3B
z(Hiq45qu=Gc1Z~XW>!b=xEkjAE~ulX2LGnpw)hiyEVD1ZZ|c|<W#p$y$WDT`>&QHh
zGfv>s2rTo$yNAD)-kFaVpkC%Ae=@nszXEJ<sb=VWsr>7=q;=ib>?8qT?aV#p^w{0^
z8YP_q8?M`|5jf|<63UK1o0tEN`-HiD5iq|urjIYQM?aR#pYSU%JZ<F;x+%5mf=W~)
zQKokTYZNgQZR~-ZTCKEYIY|h(kjaz^n{*h%4f<qFlBWP^%^!A?_i9EOQa>w%Eb1GV
zXjv^MvNhxVaIC<ggNxc9d#<T6P2?q;?u!lc57{M?d>(9fgB37YlNzJRRXGflN@1DF
z8}?w9x+67L1Q$kFh$+j^!tZ&?e%*@&uLI$wJ~j2y9u9*YzNg%VGHr9!wF6$ruH!R;
zq(9<&(BlGN6><VI?2bufcx6`@L!7#<W4<}D*?#%^5j+#g$`=nO*+cD_W!gOo7Z0;#
z+HH8=R6Xbed5W6WuS_>cdL)=UGMY{>el4#gmU%pOH|vK1dFu~W%aUfdD+-~fiKSDV
zNG^k3x9VwLj5PZrC=`4$2F9iJY&%|4*GZxvq8&rw@+8k`>sN!~R;CpE-67xp%jvt2
zqFwQfv{0G>J;NG#Nk6i}qRly9Lu5-&SQR2IrSD+&04&xqzzlSl#L~o^3oyRc=C-?}
zTs<QvqxZlr31AIPiDTtpdd}AG*u{MbV2co93ztpM7D7iiv!EQ!fa`Q&Pe_lwu29zc
z=uGO(;g9y(v!343f74wuI3OmdyRwy<nSI|&ctJ=iY^(@1PF~)1B&V3T6;M3h#T&Hk
zVk2-Y#zK}=FPRU;-f_o{LzIzM(5z1+QYoIi-OhGgnAd0s2qZWMWn$;!ZA3G}fh~7Q
zv_8`Y$+N^^g~nOy6xs!=|0tbQOx{Z1)Y{O+(1ddb8)SFIB$iuo5)&+Q427XrL-@{S
zMC_X037V+1HRD1{X9rbCrV}ARrGm1JL}utlx03I8(hX053vv{Nxbn`Ep8*F>F&KlN
z>BoLYIg#s$UZmNPEt%MCF{8NzlIQC!vB0+czv(_DVRU2+O&HgY)Z59q>715fYhvjA
z(r&f>)OWh;(|YB4F9OStcMZOoiiv8`Fc|B{jxv*TI|?k^5|ucVi?>?AK?db1qcL);
zMzE5opCJx?UY756j)lMIEK<YR?yb1%Bl?{z88Qn8i=@S)5Mqa<9~3HGfZ0JL$Pg0r
zvJPbm6>VaZ^jMST!OSQGSBBZ_Ae_?NBCos;7hH^+xNNdC8Rg<1E4P8y-pCp69V6vF
zL|G0xpW9h1?r-a_l&He6w|#q9I?N4Gwu9Q0Y!;KL&R+g<b4A@QiFpIsdIEK13!;#@
zW@=l;XqIX2b6^9X`x3BMqx^z?$wjS*q!R@BC`0w~T{4lMiPRPo6{KZFWIZ-a;@T85
z{1t3^Rmh}p*3DmKU$x73c;?gwQp?Mqdbx!idszPcICEoiFYiwkyY*=NnXJhhkRzUH
z(30`@1F5GprL;Fp{&7B3Nu+698rbO5Asbz0*VS_vB`M@yO|!6P`B#l3a5U>vIHif^
zTYFkWOUI2WMfFHMzfpy?5inaxK3(V4-?dx*d~&CtO2?$EZ>)nm*6<v`Oc=~UWR@<6
zN}$!G{JLd{8OJE{zNrMjNWsh#HIOU$K{yoEjH8wpPc|_lG7Ag+s)zk*@hEb<646*c
zS86QetYW?!@`9x{QA1jD<E9Z(Gl}K4ik%WfTcNrCLoG1<HKqI(YX1FbsJ+`Xe8L=w
zj()%4<y(VUM#pG_KV5pcnKNx8pHTf5;v5g@Z#o3zZ@K|4T<z>GECLY6<Z6dAHhEpp
zM*_7=d(dtZ!kMM<?_ZGHL=Om5l(!pUjau+Wa!)ft3f@lPmW*PZxHXW`Cdh|u6b~!d
z1dIOT{@iQ{Y%H{2v33+w&|GJ7Noo44eNjE|KI{FHi|hbCTMx6db576fcT|88R8=#Z
zysvUs__-}Gl2E%LuiH9gTLdPMqm%YUEnhD22-ktd{d~Ni`e*BbfKMfe6y(aGuDmN(
zix~_w)?{(Jm-n4`;dB_+n0RDocxuldErFxx7e}}MrekET?uJ0PY0w(Q#wOBN-SOQU
z=5+*SK#;UCn%$M7JR<gCr_1N1=D+hc*NC?8y!%t-XOu}z9{&NI)`!?QC$<^*+`-bu
zmv?xKm#`a%B}Wc-XBBE%0M)Eb7?j$rrkHWKa`!gxN6E_cL*4Np2%5}!nu%@gN;9|5
zCjk9pAyq>m<JhFDCVnW^EQ$Pa8HD=SrN+$x<2DsHi|uZhIo7_^+`iP~2BOri)6QJO
zzG&IvOuYyX%!0AT+>?d!PVsTLFi-#DKBCQD!_F8~E)*1`ILHZ6!Z%>_(?)Xgna+6B
zbbRPElS{><1@<pha=d%=HyuOvPRjB5qv~~&Zcc7->08e*A9BX4X22W!Br{P#%9t_j
z(n_*Pwiuy(DG9uPzI;;+Fi>UQ#L55QypS{H>j9(6MC!~0H8868vb`-_t>VYi>NHn(
zy%@A#&TEDr;Z*f<wYXWBONBL=n(-3-tCIOWB0Zf*qFEH*`Yys!uAtyz6D<P|if#gt
z8MsVa9C7x`jEU0dsm-c)4rIM5qZbIF*OfMiXvh~1#jI}RW#(S)U>84bLtsI&OWDyL
zP@sVUK`Fp2YW<D+>rL)qHGto=GKIE1A-nhmJ;YKT)m$zY$&jJReXkGf%NEW97$|GD
zP??PhTHMZ2<F6=9=*y1yQlcgMG;_x6Jm%#@t#|7Qw_ue7j0iXWdYQ8g=DWcG60SQE
z7_E<X4qPSNVE94+ezvN^c3TWOjN^VQkzoiz6I&2W819R+1oLvMailjN-&qc6z12;>
zh~dg?_d$`zSX^}MqnaAPb%~Kn2S?RN<=st{4H;IuzKmr*2fRTn{6kJ<!OBn?18!GJ
zOwm>m^Z2l^>vsELV4Eb@O~z;!lQi?kk4n7!_mV(l(Z}b-P$qsXX5Y;7Syt-#0I^ww
z*If_iyM$)Okc1U8V9}TCwt(+ubTm<qRS)~Td|lvT`J`-8Rfr!>{ksfGQs+;e4(x5g
z{sK+0sd$JpPhcJO2Vi~zWFw#U_UAk(CJYA9Q!KkgqMfr20QgCzb{9$OXyuEu-Ar9N
zA}@~4aRWAa;|bN^dLI641NJ^v`%>54c=&e^Sua<}#_M$88%leQY-qjjqFuE9JDFj|
zEqP+ONo()8vg3I{?}?TCI<HiXm0@(c!o)`S%I%s?X@zx!H3*94L4rZ2Y7j{ZycCQ6
zUiCi=IV{v7<sVg6oRNCjR2PmKnwXw;)o2owdZ<6%1EJ=0oM=A|K%|JIDgS^Kmtc-~
zUH+z%HXSD$x3$)d6Cc-J0wkC!dk=gpxYHs+-6MsOCe8!ZA4!@^ZfGqG8~sr2bE66z
zeO3e_z?_8{5c1?BttosnxM04~5A6|iwjP@;wcHk?zhD%G110l&1!(80xJv|=6tht+
zrsj28qBT?YWu<5@j`4HX8`brZG>0Q-c=`bh;F2O4i8C5CzW^O@TCOADFJ!FbW+j3}
zj6>;Ep|@OGe>^8$DMAr1aH7#g8<X;L1>qhilCpiN#RCay3}A+&u;ycmW9ZR^{79~e
ztvp55pczTqhAGj1q=bxoMpaNVuX9H&J~O)$yldh&HXA(<N-#dVpp)70`YxJX6O*f`
zuCn67OjGN3CacEy5w7dc_b1wD>yGSs`BfBC>H=&hburEuf+U0+$>n%)T9JKNq{wk2
zl)T{^x-(pj@84wsftV43aJXVwga(C%q)2lMe=mbW1S5%Ea_v3hk;z5k-lrPCJjYZT
z#EdeN;brW+>H@7Beq~Vj{1G{sA#bK2bybGG=8@612uF|F;%A?_O>z%#jI_=M*gZQo
zuN|MWNfU+stQQejx7e3?Nxac8$OQ=VHbdt2d8V^FoQ=s|19o1~*mH5e=d%iIOMUd2
z$A=}#tzKp?4vBH0!D~4KYO|GYtFuqXd$Xt+ScieFoR!Pzo)B#$=xsGO7EmlFWn0W>
zm=NL!n<dh;$gH<^wjE1$ZCAe2K529bB9Rpla*4P8t)OLN2Ig^He2JS$c1}%pJGroF
z#pmWku>e##ES}l(6|0LHhse`I-HppwnMI5D`&{JAplE@04s42Vplt6eD~~qC(F!GE
zw@7Hgg8dzVC1-9;7ONjt%ipFPU%X!UUG!p8+DqQVbfP9$Na#fB!Tff#FLiI(w=e3p
zFI@$C`^6lX!Q#>iMm}WK4y>lxeU&i3_}X+rAXX|$UDF65WZc&EGmN>}TP+q^$<~?u
zT|aYh#NdOpmu=s7zLB^P;*?(b7T|8Qz_?yE_X`NbE(k{;K1SylAs}EQ`g$ms5UdJj
z0E1GE82KVD7T`0ciS6x}_agpsFgHT9jFtIFtXJS|nU1LjjtLUqg%?3%ZHc;uGDh3R
zyV2m=;WBq>N<^tS?Xk0Iltjp`z&*7=9}ffX=z%!sbS7!u5YSF`$ecHA>=e(u7M=#b
zIL+jgQ7!MasFHHuV6^YrA1%O$QYwaArjebCnY8~+q5}O^g<0HF2No)3rlc)aCJH43
zEJ^E@<!TSaS}|az4DGUkPyVdbl0ly+o=~~*p7hJE;p7*IQV4{Q1fF@<*;#$G4lIex
zg2PD^mll|tC$AiZ2H6Hz;$(IbIKdW#1PIlkgG~8Fq9z5^8!KX_Qh7tVRoR`SYZ1)C
z2-C*K@BxBqj;!fZiAvUcn1YK(nPokBI<rMn?F;M=N47`}(BVY&B)1z+>LD&Nm1U~q
z5P>FLL%G%|xw7TV*~OSf&E?<FsQkBX`5wWVHLp@?hs;*RYPaQxyGjnPAF(GUt@s>;
zdlREJEH})5!zCIUU$EL20V2BAH)flkNyl;`u#{0R-Fb?}O$I<6Mg*%@6TZM?R_aON
zJKdfUp0H@sr;PFPA-kxJodQRmXlF?t5ssj2yn?%FP&Rj;VDoC;>}-GLtLjU7=OHQs
zoJO-h<uLI$(5fm;)4d)0onxfW##_cm2ij>u5G-|lbOwVH7UKZ3MBDY)vz)K>zX0p2
z_D8X^5ZTXz!Zaf$)mDpi5%JF$V*GY=1SmgLbvmW8&kh?g(DXhRc(`C3_bP@I#(HEr
zJ{-;mu9^^)kVdHu%zI+&`rFp5ut~4LM9?@@uClDRy5sn9c0cYE^~`KT+OUZcB_HFW
z*_N~2sm<@g%Dd;ZM9dG9u9fo7VYYg~?)jkZzDSAkjSQ2^tQiuOm(=UDB3E)-hl<Z;
zdz4E}l(TjT_;5evTE@%x%&c_3*u4r&YJgr|Dg`swfd2+Ezdd>D6^p`GMt{CU-<iAc
zp7-7xz3Nv*FV=M5M}@y%v8><!5PVaBHrL+XUR1BNhB^HGk_&0HQpMHQ|6#7a|9#C*
zG5V2M#F5Ysh7V{%RkBLM8@bl*yxGf+7Tqk8H8)t?)y9E?5nG96e0+k+j!r8-#T+l*
zdY#_C54wD2Y50|EFIguW%fmKaWyu!{xXUv*{fWd-)zIXx+97Q^<C8!d<eI(5<aOrO
zHXF+Fm)~^TS!6r3ig7kdV!3`d#tJUd=^N6no3RkmT;gRwws{aF7?N%^+^Pqil#8&=
z`Bl4NQYq%GdCci65MZKSDKo2~=O;Eq0}Wh8%@cY%6eoYx;a(@;zZ|G6TkaHC6Qd!N
zEQeZlE5`jMO2S#S_c3#apGA1{a11n<Yct`3Id@RAJr=nc!}QSCEcQm5n^(bv?Y9Tm
zOk&<`K44ydfNhz_-0sQ0WW!1Zbvh((xpF@<pnsstFL=w?93yvs+~=LjqoRBBL8gYb
z<ePE8Td>d0S(&siMfd{?P8@l7@|mtTpty+&V=lkydpf~|Xp?i47m@?ktMxvzejj&r
zM9PgRqvWPR1RrwM!2iz6*2VcME$I2AGR_Y%SF2p#_JYmKEn@bCi(S_?96~E3T$Cp@
ziug`9eQzF3hQlVaZl{%Am8xSUbnh$gEBm~Uf}asHFV*jVCLDfWeI8jtHEq9Ry#<56
zcdWIl7ldh$q&6(={d)5^3FwNsjiB;QQxvd1NC<PqO4}D%A1D91>tgK7Qj9HE+pe%b
zI$1TdXtKDe{zB$2SCb9gJrmN3Y>SmLv6{ZSSOfp+>5a7MnbTYX?xx)g(Ky6c1C^(~
z)MKWZNF|6CQAbeV&y`1+B$uz&ka)-vvG%1h?-50LYj=GB$Ixl=hai`sR1=RK5u@#h
zcjmH+nbiNK5;Cz0SPOY#4w#Y1D~*lK#?czGj=9fy$f9^{5{EoPo*k5b574hhe?uZI
zX+nE}-kT!3hv2Z>jGvSr8c7+?1<;}G%EKnJdFA4PZOlD?3bITU7%*ocxfyAfZ9nt}
zyMvoCf5dko@c?C?#lVJDKkw_NFC90uZuvg1jH70<Z=)`VU$QT<id7UFGsw=l)^1#{
zH|(U~`qARCcl-0Y@dZuBYrmLe_`hf9qk6C(l&|IBg}U<h`JEe}&hJOSn1q`G3?e)X
zlqcnJq0&;^gNXUKvD1MT6S=BZ2g0_r_u;-eiVbn9PQxnSX@33h%IoGt{n6C~MmIQ3
z!5~iJF~e6P(k8xY&A#{O@?gm?2-h0dc{3T+VaXXkBEbqJ`;_9wg|T6!>gT|x$gbN$
zF^SRV8goi#_c2;;gk|PpTq-dVIjqjI9|~)n>-C~t@3c5RT#qG4J{<IoN#^#;o6LA#
z@`fZ<0CSyN8WxFJwJON#Xd;eRHWsN5F<qKkezWM(2kP2|HBo7Qa-_+vrttx9hyVH%
zfV6Ubi$5>2IXBnKbywK@z|!_Ju<at{ma@+t<AnwrU+uIbM16pVyv6Tw$S-*)9a1cN
zOjRR!2sWu*4=jOYn9xnt!>M!gp^+;2tQ^j_b16vD0W|7~lBw}p9mhH2<eKyEf(eiC
zuB~Masgc~DYwE5s1+yS)eKenmzR%Rp3GQ|y2{z{T;Aedeaa`?Oc$ITJR-0M<Uu}$4
z`q<16ucoDzrtVq!Uxkey<qgk>S<c;$?1F2ry<eH>6KQj~SFSd`9#+>CS|o}Bw;8;8
zI!L+|VuSF9RpZEZcD5D5+<;w!OwHwKXE{Kt9ZQsxQ_6#T6eQqOUeAEV+&KNFR?ijO
zeHS5ct8f0j9z%@5cawrW4gsH*LZ3Chm5}BIo+^awnwcoiDl>vvhL|W|w}#x!lE}(g
zMHl%Th(L#D*mPxTIOh@#0`KmCV~M-w8%YL4os^F8ktN`OffZSdaV~yFSdRg8o^AYj
zbd_ljI8E3QAq*noK$Z~R^;ZcqDxW5(rkkP_k@3naRh&({owDuNL*J>FkRQYxzwm)7
zK%K8uP-`)4+{HZ)GQOM%4Qw-XL`d8hictTJ>xa!-tk^EcLVBvGd#dNunzFthaCQC=
zy!};kHrjgpR*UF2gMIN=U-!88&V_JjbhPE$<d?v{nhUq4WGJpH>y7{DoD^6T$Eq8|
z4AmyT4Czs+n`}02(-I9u+hqNcop^1pIF6)Ui7V;o9}Vz#aN{3!9Hx6{mW%RfOj7a^
zV<){FQ0(4c{C!zmTGjNyEKb;USL4$RwdwiXZ`n8Hr=i#Q1R(C&?s2kqt`$idqOJUo
zt3H75b($Wv!9RI*8yfqOj~NiAA8r>oMCu7!sqgvAFg$I9Gd+*rz0`C6GhTxo4wj=a
zuq%3PtffBS)b)#O>Pq4bjeyzb)L1H1SlQ!Ol<5{AwpC73EDT|`J=1<~C3A>J;ex>q
zPY%T5w10>=TrCd04Dy@u^8Ow*w>*6Z!fL~H?Y@)$WB47Lbm0ikcWVCJAB!FH2o`x_
zxSFdI?w*)A_bf_BYIk@$sdiEJ0nR60xm~^NjX;N_Ze{39jmC!ynG41uwfNZl=Blh9
z=LAMQ;epjlI^Zk!lzgN%gcrM>-5k;yjl`kiLl-+mY-2;u27L$`@r@I{Kw|yPZsD9h
z6XN@V6TYQ|39V1g-xMN>y$An_RL$MsVSnN-a_zQ{;M3K?B7B5`=FXVi);!-OkG8C}
zprs}+PoCQ~*SfWsTI2!$FSd8CZv0RJ)(Jgd3~b3@W_S}zNk)<BTLx-NPSf1qv!V28
zLk2PLyYmCk!11FO#ycvDyDWRE57&~}CaOuhe|R3y@F|S(!AU9j8)aG;%@kmT{E}i(
zd_nhEsvu=o#OO7k;WH((-3W5`*;CuZ<BOz8z?M|BvkTdc@NLhDjD9$pZ?}2LZa@z-
z4;#mZjKup_W|JZMlizt|3fW@7i+|JARC(txuY{94q}}tPaEKS2%{(<L^VT1#5Qn<t
z#h+d*T$v2<l4%itru_NJRsFon3PsTfV3k{(2;vz~AI+<|5BTsKiqg%%v(!m<tgHlO
z3EbK*$<eQxz1?F<wb%4v(>PLj^08p`CQj%N4=^YsDsRLA>J?}!dog%tv^0PnqOF>h
zlGC60`gWGlMA0w-8zLpYVHBtKM}&vRXUbcV(W*1bu{e6DH9)vs*EetZ68Ns<JL1?-
zR0+o2+M+3BUMRxO@O7nV+n`ap`xYv}Xpqy{)dy0OdN1O-{0k{1UhZpn5suEc3kUqB
zVT#k9V<XH@as&U+4Pas8)kd{bWNJpo<421lYb=+R-tj>f<a{8bno0_VZ{!(o_eER!
zvQ#{~+VV!?{rV7s03D^>-}L8TUaGETw6s(G{s$}M9-GpIZhG2n<INv0@w3E!akf4o
zWZv>p!|m%`W3kVSmOj*~u?l;ibrWy1Gq|8YxY``(n`+@><T$}wT{#e=#%U~w==U`#
zol`cpVqE*HuMzWO*ge+`8et9%uest4&#!YmXRQ$RHXJVts#fIK&>Ta=R;OaFbNsb2
zz;7AvAMi17aaUqqR;40SZ6i9*og-G=>TUD$Y$I2Zy4Wm_vWe7MzNsORvge(~qqiIP
zeC|ooL^Zp06Q0fs4N=}bC0d>*_&ElHzbGh|-ZZDXthq5Q_~s5Sikh32+T(wbV3}Ta
zNAPbteLwHqUxCkxOr;RknCILGNi-8)lpx8?Ey*n=2@Amqi~KVEf;b0F2YoVfp5KjW
zpnQ4G*PpT@u(%Jp!Tt<s^@gXhpl^A)d1}uF|J~OG42?RJJN8LAUF7(5VP6-gqcHJq
z7+C8wo0>D7HJ!sC5iCLdSp#{Iw0d9Pu+e2hOvop(0Gp1%;lg7^Q)lfL*|S+w?P|~{
zyl<v74xjNW@b+&#;?B2W24;#4gW7~Dg4omzeT*u$z-8RCLzzxeiKBn=4uu~^3q%wl
zUoq_?Zq&t}F4)J5uzp}b``qwm(oi#Q?=&G(r+3E-%K1mleLI+#tpq1Ma)Ye1vilqC
zv<L0BAxg7UuSngtO&|M3(-mv9h*Zs~@t_R$3U*COsD@8?dkV2V`gm-gP_`HGby{A*
zzN2143~mpe80pzqPr3&jlpX&y&tcMGYxE$QTc(CvB?x`3gIPpq(3cM<;F}cX0<2>h
zb4Z%zb0mFSf4?>gZyT(I2TQzxS8<G}M?{z!+XOhu3wm$si%--u<^I;aZRCeb@iC9#
z4U8HOLs(TC;iO`Pr>)LE%eM(7ZhPzA7Jna9%wy$?Jm2XjU)BN<4vpIfDM#Y`J}+g$
z`|sU*h3$rSX!y+;g?Z?x`H4Mo=*al+r`-8bCp^VSdsnNTzbjk~&+S^#7>leJ-l&cT
zOg~;NFMmd^`Uw9#BV*>z^UTF$boET%dGmNR1EVhy!@3#Oi0A7uN|F~|mL!*}zE?uF
zg`x9W+6(G0!+FcK-7MN_lYFH1E<ejaDU5hdPNJjVY_3Fy)DfhI&pXdE&eq!H&I%Q?
zr`E--+_}jb`pgt8iPb#f-pm_oeQThY!w`~K^~OOsEFGP7s}CYbBx!PF%{dwLU%LAx
zY;Id6M!+zU_mkaU4GyzP9g9`FGHk^90nwoL`HiQG%ibbl{k?;}t89%iXX<vOb3b$+
zV{y{39<}K|RVIh?+nv#{^_*it8u%>!teq~G2~sSx$)AnOt}^087<Qh7r7(~rqm3{&
zC;J5;K@m@Z!N5Q4jp1W^{{4CVebGOrxHbWLqSC~?Edu`V%_;P^49^~6qda~y?&%LF
zKwo96E>6i>N)97}(fN5QX@T-RUF3=c)r3TcRh4~0C(AxwvzI3YZl8-WvH}Uhg>Evk
zd>Z+Xlb+W>wVBAz63){RFL_e7XnA?iTWcay_UcC3f|Xuw`aAu>r7vc25~-LAy3E;w
ze*`I3ii{<rp>>=o!#}oz)*c<3l*8t-lZ%gsmR%mTnGX-*pi(Y5UbFYRoyUsXxo_tI
z2LW?Z-}5E2N9zw8KB?Wezw9Rj_#tVZT>O5w)4EXbdZa_ropH}_p46A|a=WkeWl<2T
zdD(}*AQfSqzkK46nP^E|A-8uRrD_LRZOlu!mms1V&rmTab&kXtM+}Y9JrtW{Xl>lW
z*2)~wCkmUXda^eXRIU~I+I~mXzU}pI#BR0Snw8=7Q*dl~z)7$~jTH?wpWAZ&HU^#6
zzC8Mrft)P`$xgX7e@zDs`Ua2lQ?cNMe$k8*%-|sKXR^~|$BTOb<E0@=C9CS#vsiy}
z3D&G@d;!Lsl_T|*L#j;!uQtJJNw_CMgbs@>$nh~7*A>YK*DVpb-ffgs7QNx$6i>au
zua0orDtqZ?wTwd9a8&#93v*-3rB-sAS}mhMqauPt)TfkzV@ua<?k-R)!lK5wMRxR0
zS(@?0hRL0s@;Fd50dO;D!f_zLQ|UE)fY1wLKPTA9vlc%RC4Bv8!aY}0;l2Ve^(2cK
zzV^+4wO4X^;a0-54|w?Zc@mUXH%!g63o0TTK6Xs7Y%^!?2glU!`#qeZ{(|^ZOcGzy
zq|<4F2SMm1w7M7L{I}KoNo<xxVd~SjXm1T^XV0r16f@CcmHNe_%=+`V8=gMwdIh)Z
zl+t0*-3UH*<)bWWP)4mi!*e#GBGqw`CU$BWh3OF>Q?)iG3dOiKF&QI$46oG5VKd4P
zJwa8N=PXEY`h;GUpBbZ(VPkb3y#+?yd+_LcwKK5YR>iiseRWQCTU89VLeg#0gwIRs
zgL_ysJyuX9vrDb+R$cceyG0jPc#A&ph>s9uL{wGUbr$!6AEw0;Yvj~YA(E3S?%GD&
zEfGzltrjNAPl4U>8j(L`gYEbA0B@<_MYmxr;|O-r5;-^3D^hzWrwDhZsXJpUV0^yk
zu{Ff~^jVMseAX!^pRqXR^GXh^r<9}8;dDFyQ7OA*TK>cE)Tyv?-QOT0XO@Oca+D-N
z3NWhS@#np_7{*?3ThF_R`mW5}@p2}Q2Ct<L|7{-al{K(AYzBpM^pPc@>z=%a)K1D2
z!P^fti{+G}%rP*2EI0Xi<Cl%b;f79?)Q3YH%`|7-lAGjO(v7*Zi$3pQC`oS5J9$-u
zr3`s6W_x5#3~9tETA?`{s}bQvyS~RRWG;i(69lyA!o@c`f72oL-LyKbYsCUvch2IL
zs*ehAc0QD-o{RI9yu&xTG&rpu+7xDq5)D&6RJD6Jh!$+~#T;?{=8IKA@5@mh^x28l
zPA70}b_+vC_(-dAt<8}NQhQ0yxsbUmUtQ{d&lLgIHr>Jmbv-Xi_>chkaq&K>0d{Ss
zbx$H8eUg0E(sb7bts^j0>Gdq-w;1;>*fha-`9L>4)+=S=GAm7SfR~5>543e^i!~{(
zKkImgQ~4wXEG}92sYkeHNYkWQzo^%7ofq<ML$nD#oRN-I15t$8*tsJdOKiUwqfGrK
zXwu3#^A-WrzVS``Qhzf>X5}({c#S3rnWnJ^n-JM7{@D99NsNkb7-X4s3o)5}@!}O}
z+4NSVnypF1!(`@+Hp#vb%w-f9_3}pRAMGnk(XGJ?y38V+kOGh2Q0r87j1OvJreP}k
zjBINqE?-U*V?(qhM5#2(b@TqtFSH?;?kC(FttmHoBWb9FZb(K4Dv(sPY@o*_IJ0-F
z8kEKsCVQcK<_X#AiMsdhKFdSIn|eErYpU^poMKis0yzXmaJ8=EY06fw*fTts)h28t
zHR{TYKOr}+=M6WN{$+n%99wK8@osd}v`Z9oC04R9Rf@m0O~=<{eDSShkc-QE!Au1w
zpEFXgKIDE@u&u&bBNf1RYFI+dxtT=%a10Yek<V3Mm0hq-%@I8tRiqRsF;}a(f_(P6
z3TOPom8<BD#f+`HB}=v8hVl4s)!ga3?-xERj9L#LD&_ARESUxDwJXv5O@hm_4CS-A
z3>Wb$CaE{GNu0Bpk_m|K(A!61&NgSzfp%Xab*?PCP<y$m7`03iS#DR&{x-m`9FK<|
z<)(ZxDwzSztpna#b3WYd9Egr~+f6vXx4Z*wcP`DM)ed9wK?0^@s1aB+M!F6HD8h4J
z(iq_|a|i8>+=n=%b(Sd#71*KIVZcJXYN*FE2PJBgm=QM_SM1yYtGV&Fd3EM<*2-eu
zdz@WC8~9)Ri_ZI^5%CPeb6i=zHt5^JQDX|D1^4ySUWd<qXcuaH<~mh^6uvSdMT|6(
z{sw(KXQityhq29-OUm=WU*M(TPKh6hBS*g%@<jdikv(8Dyc*acin_#+1nkstGvKSo
zjE)|BQezi6530Q8-l5yan|H_Yn1SZSe?GYOR<^>{V(|rYq4i>wUG>T@P3+Ln2Yi-R
zd*F3ICd*?^Ig@%Zr%a8+y6z(E=kE}Z8Cmu91Lyh})fF09m!Y#dSie!Umc7noez^a6
zD>vffJJfrm&Ej^$>9>01<Mht!{ly#hp~4?`;#Jh~VehKj9sPNJ-POAjHIY^JFo}1G
zXxiO-;deVm52TnE77$pO1Nbi5m-Lzt40y-G-TPoKsyx(cVZ}E_Wo<D2h_$)<T7B#}
zm7#own=cj^VX6Kfg;o58!u=dYgUu&KNN2z1aA@m-LL^tb9Y+O)6a~!fE~o{vE-%)|
zg=UaT*a*yp*ntq}`9l%7f0R==-b}TqUkYts|2%75Q$#_yF<knk*HFgsbSQG9UiuxN
zZEe+gHYjWgj*k1Bch(@rE^!rUcbZ7Oy?r272W%DoP3MhU3?U})42Tlcp%Z(mtEHT(
zipzeU`dQpJSkp!<W^>l!AICE#*9zE$Z6##({$SZ5Q+1JuucFKr7{mT$0<oq6@7#v^
zRCa@yB#vOC*t{(U;oCI(kypB_y5&UVe!q5f0#<+Q(nN`SaQFgXn*@27(HJ>U7EW4G
za!@>PW9V6MVgkz6WbE$`nCqHm0Gk6Xg7d=u?NdtaJ*peO_}8P&SH(`na2e)~avEix
z{$}v)Y=`o|CTwd&q@5Ar01i1X71VE>BkPwQVEfh|46ey!pke)e$Gv>)AS8N@eV%9;
z3IbePWP@**&}H@h$x8;T>;`|y#Oeo3elK}8+nbAim3uRc!C8Io4_icY51Z^l4eWjV
zqDz=Rtn2N^r%BynRndl4q3;s}zjiPphWCRq?nybi8M4;apRo~}tho_WedeC~nK#<|
zuzRk4%gk2oc1UT4qPWprErRy7hnTOIZf}^y?Qq4ra|W|CsQf7|Ur;?Zn8JAz!iue#
zX!dpqW<|!N2N0Ho8(B0a=@xl@&&<}Bdb;;AEEMl<OkT<JlQ!^g4ba01ni2+W>$$zq
z0*w+6W`|wh4r}ZU=at`h%N99N>Z(Wh;Sw-*od$*8XHOs=2q^LYO(!MqzLkp{1zg!Y
z2ykh;Jzju=tT@%%1Pmty5^NTN5~v;7-0wNuHa!v-xO~3xMx}DZYLPm^TpQF%G+8sU
z9TYkj<oxsqQ_(O;c9B?%u~%Duri1^%n+G2ro<|q@x@E=PZc)y7Gk#~baPgL0)OxDE
z2LGqA$-L2Bu>AW+M5BUG)4j!E!M&HmtC~wI6J<+=kDH!wz3*tv{pJ7H;`EceKPhQl
zZfoI|f)J3Rx?|?}+tPB*<1C!g!1@6zb*cxmy$aqX#t1Oz2p`M?3x%F@idV*B{^V4Q
z-hJKr7il#R;p;M_&7q+Wa=n~&P`vR-{%KF(GpeDm?OY|Z%A67!u;MWl&Uk3@z5&>!
z4D1kC?3_+{jVV*IA%D?j@SXAp_OXSMtADBL<m!ky%+9FoIt2a^V^5HI=(288ELO8I
zc0CYXSd%`GF(Uh*R^N2GTor{vO_ErC)bGo@Rjg62!cHo{%pDEc2s(hJCv>$jL^Jw8
zc^^W_&nU^uy#OmfImll|vX`OI2xPO*0FR^j^T$_Y`rqUwLTmbGwc{NhN@rJ5&BjZv
zR|sneN0g89ed7O6_*_6`mHPsOg89qb;8drxHo%Cnq&Qv9otkTrENLN}jPs(p(HeR=
zj<`5{QIzArkw|~Dyqo7n`H(W|eE6+ljl)yNH4OUiY?$CYH(sO*FVmAC=8_7~7CctX
z+N|#WryVC}_U={nbC#2rO$Zj=ykk{Y`nm`<Pwoqq_8H4>$=?*IqRqt<9h>Gtgzy`?
zdREU1^4fH3{KAI8{yp5L#@zjTx@w#h<Qc{bKJ}%CLRVHAT7*o3VRjGmtcL+pm|eXH
zrg@6ypnI_2j-)oW<r<Xh3EW|y{r!akFHPPTsr5@)nB)%|5&J#Zzk+LYE^U3vX!5m?
zSxgVF;tKA*8aeU!fZ_gE1<KOXSA!LnV5QTbF86W=AN6lK%f3${Si@@V<;L*s?|rAn
zD{IZ38j8%9eRiN;bC2D*AkUelt!VGGH{#?x@to_eMTB(FQz;#F7c!c+wN!|QPn0J@
z8X4A3cx>Ls-FV1LXj*z@+wU5igL_oDaEY9{_Am}AWH;6~l+qT^C3t9Txy}NaU)*Hg
z?pp9s5AdnEn9%KK-J#WCu=1Ua(E<{SvZ}W&|8bzurN}Sfz}B0v!D+0YHk)g(aXDKE
z2j_B6F}`q&Q`?Pg+dCcjH{EQXN7YzmA(a*c(KVFpG(A8J4ZrO$z!FPMzw<0EwCix$
zpghYc*|e#!$1J*Mi&Y}3iN%0XnjE#-?`B`dO!>y=^5^1gfAFhfQU}ob1L=+lU3~0?
zXK?Bdu_1De_G=eXNv9gA5)S58=LXgP&WpKQHBL>h7yrq<FA^QkHfy<iD9BZz^0X#L
zBy4gsUrVMF@h%&KC81t|6jAxsM27JSVb2`-wx2>c*%HUg*r{4=%B;Oq+LhVa*7R2+
z6WxkKaPB;E2{nJshQ6XG=2fYK84(m5ZCMXPY1Y1xGx?dKvCb#d9MjuIv|&#y4RHMN
zNE+PiAwsKbqh9?wLNmJE29Fr@MzD^mgvdyJPiagvzweARvzF<gF2{>K+xp1T%#ZW_
zg;(ag7naPzK(m#EAbWXNy0tvV?uTt*3O<ec=-6B*A+4}H7WK`OApc9X-eZ`louNvN
z;%SOCdt`clMSR>0y;O{}R~pk(RIcsyA^BhZflOzU?pR-ZH^fiM@ebIlp8W_@3P_h|
zEm@z0c(q+S<#|)|Y*o9iB0;QWy?l;d%qaNvCU?d4vkZFnUFeW^pM6=cZuom>A>N95
zrEhWjLh|VkuRA)-kuR7HL<rrwV{hJMl@~O!K0qJp<2#}MntP3JA_lDJp%dKY+Vabo
zy*aY?^bU5FIAOqWYPGH9Ky3W#7`{Yq>4=peGN{%S#%q^ZFOuaDoq`Cx?J8XV^v_w{
z%51bnVf$7@YxS08y+%Bb%G2-dOrOHej$in(>-XqtAE)`dRt@u>%(}A0YnRg1hD2lx
zM^}s-!d@*y-^EjWADR%RB;YdF8J{_6Y;4vB?kn7SV%z~DF?g~{tyCbmfI}Px!}YKm
z0s~$s%R=D9g1*J~+fT*3$ti(PjaLK=wIr`dH}T;Y-%RTIzDmf3POQB!s(E2j<mjG{
zTXY#9Gk#22%-PKLZZp`0KL?x+>M}<qaBAcVmBs3;7VrK|XF}Rd7E!<j*^sEZ^ykz8
zlRwagxC6l8JT){z#&Tsw*!SVCKX6c<*1{69+<0$8u9+BwnNs!GS0#-pZ8#n$_>M<|
z?T(9HE<OqnC6ag>L(j$<CFQYD!^xgSBXK4kcug5II9NXze_y7b+!|*&(Mp*y8G7@q
zizF0+srHlJQrQ~KDP*qcZIi@e5j-|T9<!B)nX}5ZOTZ^#TEU&1GW$23BA3<k{HM~U
zU-rgCc}CY+<BhR;@ft&JJtnEbU$VcU>n;(4R}aJjh%`wZk+#TcY5!GJQ`kGRh{a3T
z`u7<7Y1yLu$J=>1zN;v7$XJ$xXK-gP$8?*wJU?SWXX@9x#pnCXi$R@d-gIUc^Ij8>
zmb0eOB8QGjnrj!$RP>ah`?Uq1#cG#}EM~|#&cDZ~j$7W5E*Bnu6XN$fDJNcq;aA)6
zbF9ad?8gUTRa2t5;F<X85bkp%$0dtQ{kbnRn08Mlcm}6ac88-fBOE6C5$9uUA*FSu
z)7!k98luSLI0oHtKbIf*9Tbu%6V%vC2={g$%8=quq5mH|eRo)r?fZU13mmz3Vu;(+
z9BC=)D}uPmQI>m;a#dz-({K;OnVFb4dn<F5tK3^t(T0>JrWW>+Imq1oZa?4S_;DPi
z{}enA_j8T&I?wa!68LEzS#^=~`iQ^Z&$qS#kFF}wSgx-kjO+6}&a3F>*sn_I{_EaT
zq=`{D|K=k`py9TlWxvu!WGwTMMUf77m1%z%zr1fag=3m4X6ymoMCwrunqy&9zA_-6
zym<4jx@Gm-4U{lW`hAz3Wq?EQ#i2_bsUi<=+!3fRx{^9e&Q?3`Aj0;QGWpWMP?DU%
z-Rl}!=XlU>OS(_0o&W~bCvS=1q&(MMi3WaV6)M{xi<>=R#b6Xw(1~qQsd>%ByRYt^
zEF^@_yGvPpz&B%)hvFmrSM5*w3y!fyhaoaIX7`d8G0<OV=Y{L8)a-94Bhpx$VU^i`
z@sA8-mhzVNp^s8m<X3n%_P7wbRrq8lp9H8QAOn-_An}lN0RbM0#F=xWxfPP0Xh4$-
z!A#_yvz~Q*C-=Fuz5ItkAWnW!`k_&ix2ST~!)~>gLcP!gqm-+4oIQp%0X&r+@5!8+
zK5ashf<v6V(*d1N#KOIKXsHYZ+N!|3pNr3ff_kgt;sPN!!$9sW?oXPF6CrAOrbF%e
z=OveUZt|Gj%-}2`iwtF+x}|bc?0tFfHt$nE$?6|oMyrWWXl+HmjI>73e$SWT-hxpN
z1l#Z?Xxw#|{E?8eRkmIZibVS>mj|B}dJ%c#8CU&f(21{ohvBCrZQ*0{^j|lRdJW@k
z$LiCm%*LP-VWf$v?i(IAGd^gG7hkmNENH2ksBtW>b^Rkb6FKe7Ta8*CHp7;?+O?g!
zx+jB0C18?iZ6i10hMJ8{tzU;~YbtAN23-Dyj%Rkgzi~WA;d9x9W5!#(S7#rjaAWu!
zPA3y?7GBwOtv~CUH5Dny8FtR7ir10rj<v_?6w-k+sR?pT)YC|+zE$M>x%7<#54TGP
z#<%I$3M1XESKNbdh^xTZmC>Im4XlF;J;F%Y<iS#D|BDo;`=+}OY4IyHzdtW<+U}%$
zA3WELDphEA_2&&gzx>F#IAx8T{*3RmRQt0mU6)5sKToTx8uyj%pO%4p7-{w{h@~5?
ziM$>97A!M%Rk>eMz-CXnxuebYOhGoF0P39w#GX`JK=F|enX|u46)$^oqL@f!AEFJU
z6#EIZVq0kAM^=-8O_PcWBXh1^MH(mj*djP>nN#p5-yG_v50UiEFi}r)pWF3o#7WzI
z0q?PiM}18}n>N9P%^Q5wMgKKAZ*xi{TpY5@n*wzG;H+^Q5peffRXVXA);bgMVN!mS
zX!R#*kJJxxpLYQQn|n}~#`&@s_K8(!E3Z<&>csnCeYk6DLULcLRG}nh6)y>gSjhI%
zzp(WG%U6lw$tXGIK7#BMS+}hFiev$YSH42I7Vn#VqQKDjcInFx1nYtRlO5Q;1HbhM
zEBRh~)g=w>Doue`V$aR=yW(Ur*rY7763Q(M+U~ho-I6xpP?X#JsI9Ble5N#on2|-s
z$GN~xD0&3R-9UM1ph(aRsJDXi(V@2C0NT}KE4)v`TFe6%A~QxZeB!&yg(%PIRfySH
zSWjw535u_Gb|Je6x6Lbe#WO+S(mduFmFFq3ja@cb)hBe+=H7dkVxD%>6h~$CldoP{
ztje^_Pi)EyLP;OlD}H~elrb-|tk=3ozWUQSA$dX0dS0JPreEy2^}U>3vEgV{PPd9C
z;&-P54lKeez$<M)22ZxIzTvNWQ@S7@l|7;**mT_FWt`+C`|z&Tbia#Xyqr>Hs*+Y=
z>iylOMevS-sF&_834wJQ&4$L|vv6rqyv14JSd?+muf^$9i=>iO;(oNe(O{ve^?5`+
zIRheEHb~%(Mhqu05S0;XyBXW{pZ%_dgt2mBZrMG^q|y#B@4|>QQ72(l3e$x(N+9<N
zEpbovH>rE~?B)b%blq?3jt1$(!O8S6p0DUbg&^Q^pTP5M%mFn+{R!8&O~l}N=(Qii
z^fi7OIdynfY{P7^+M|JCMAff{AfjN$AAdvk{i&K;n3bTy3Kg{`xyRAj&1hvVYX`Y7
zCM<1~$X>g_7QTI~cUDmLY}xcCMJgMZx1j-N{5gMxqRQ8f_yQvh4OK`o-?8zhd+)bo
zo}}G6+p<Nz9;vElu6ZFPP99Ab6|1O+q8yYOtsPVeDvM%a(<2b#1!}T`I-yl_?7rLw
zxOs}^j#d?SIu2J_@sPKVyCqsU4pWD|*2kUowL2vD%NT@4eRoHiSKhUY*9uX5@x|>B
zqkc_M3uC!#V?*?{_ihI1Rr)DEN*2D)-{b`B_m9Pw=H?&mks9{+O&m_7<r?Al;2{E3
z{%?0)r7k1q21@{`xi(Y+@r`dOEuX{_0+;oaW!d+OFEF)rVO0PxF$@M{q+!ICnrc0n
zSD^ib3MlgOrcOfE`J4@mC!AM)@-F#4-1KQ)Q`7d2%B<WMqNY)H^~z(cZp3}%Z-Uv&
z9w*;!F2e=UoFN2TX)0em;#PPnqh{P)$8#G<fXICFrb!$EQLK~P54#{(3aK#?dO9~k
z{rv%cKjEr>;;cYBq>nwa(0B7pUbW>zg)`StqB)h#XUk3*Q+`+xZKT#M6y&YU$hjSg
z74moFDgdESGc_?D#gccbq+{<gS@88<v(P|CPv@)cwP*4zcUq^_is)JE9sFaB9rHi8
zO194XQe(M*VW2cLaL(ZO-y^c=t59XU6SUlVOH4}<P+K`d0p4Q07>&P(&E4Qz3|ZXd
zI;&YQhKw)SP+T<(M&5r<F097-XYzT)c7{uXJ=u#mmT+tNb>-2N+Z9e3I@o)-Tr6=n
zhHz6iZGIdq261f8+cYkwXzOLZlaSlc1$t;gEI;AwY=eq#37cZVzOM?ak>kl9SXye)
z3RZdC?S8)Wsx;x)xz?!@9Q;q~A+i|dtP+Jbg~0H}auNZRbngYZBOQT|tJmcPRGx{P
z=K{X*CWa{JX+m#M7q%`2{!Mp~u|Kj*GDM_#ooS=YTwEf4=H1>ecqAPvJYeL&bs-wV
zup4}n_%VCS{S6P1_=n7$7{iU1d7z+xsC#nUPFk}}`IUJfE&7DL2lta}T2BG`!Si2?
zz6<g15wOug@qKK3M`tXp@k^gMy<>v1Z;+>8Z*n@tF|9d@Wn1Dw^!-60SHKfD*pdvj
zKehHfxjkI>GxC`;A`0{O$S7ftB+*NElbMI!wO72Qk}rL$KPiQmeu>HxW{5>Ymk%=E
zIru34JtE0;sc+Me@9fL|;x%;4`Ol_CU{dS8T9qPmXnP0I5R+M@l1Q=hxDScu(Pj&b
ze7nwY#%)Ft*!6$7)1i*~Xec|Tf(i0Zu75j6i%=;O5dhitT-TIjbT7UxT*4M|eXf`>
z@Ag_9Fn+_kD_0CI%^EpS<kN<DWkN%gqnjn^-3#^yD8R=?;{7ykKc2cjJ$wV2F(aV9
z$do+ax-Sr_`8wN>TA07Zt<qUkw<)4F2Btp5Ms41GPBE3m5zmA4t35+b3C(8I>=E6g
z^_)e|oDvCQlHnrrY??-1+sO0{cdKJ)w1@c()5C0OI=Y(DgKE2PiF-JJ)c5fabrzyG
zv0OMfL2#XAAXx(0*_-c-2!?<9dqk`x4=iPw+w*xLsv27bk1cZmF0c<shR%me*Oybn
zQFeRqTd_xjo6F=g^A()9s=X$~wJ<y4iA61pA?n9JWjtT8kCT~Bw<msKyrR4H45K;<
zXQIA3z7YwYqf9a?8iX^#bLS<jA8|zKjjSx(Z9BC&){5xS=Ck1WdH`AG#y_l=h;TyC
z$BiNobfERtqa?I)HLj;Ge84NSQmEg$D~sS!=#sMT$#9<sks4BAgri?kn0iV2K=+>P
z7&m^BJ4RZ)e^}%;mjTB*X`Ei{AZC79YufTeFrog%&Aa67o4+*n2?fU-JRh+d%hD=$
zB@ux{-Ba%j8FPkC9iA55(T0|zRHO?9x0OddQOx<eY?u;=8kBv=<oz(1EvcL?tTN1J
z&PlErP8SF@%ni5tF5JC1Oly~zZsImyOiyUKtprBiQJ!J&`g5xF1t>^$KnWALf?;Bl
z2&<uh=$*EYSivzu-tLnm7FUZ+*dD>dB<a#a->_LGe~<Kh-a&F?7XFa=&Ju!XVn*L*
z)$?Av6vDA3fiE`nUTT}Ud#+97m*^H!mB~_7FW(LK2M!1gXrLUhHK8IxoXM*cJ0p$|
z7uF#omF${v_{u)}E!F4Uigl4(Y>G^RfyfDs@lf9Nfws+17+*g@6e%g)&F3UplN!O<
zo;vI~Bq9PWhwI%oHbmINd4#;#QrnP8MYMu)q;z=FgY^SeEzdvNSIm_M@N5^{+ZX>H
zncnGJ8JI@TrnWU6;M;NZ{O;}`xFzonVm)mo*|DZCvhHLX#f<7%M2$(A=mKz`;+D^X
zl@$w-OgL)6Z6WkvsAJvbY@8qLeqSk%oo9=bugTX3>r_IsgSXdev`BDz**7HGdM5gT
zXleKIer)c6%DM|v=%A)!vPNyWOa=m|Lv?Rb+Jp*q|6?~^g>iBjigSeE_XY9_21y-n
zRG(ptfgC$o>Ktv{(;0e@M#&1(&sqLc`IML>sU!6{7biRY%jdJy5)mYLptTV)Vy~c>
z;&S2K^T0ph;uNcV6d+p}0bvHoxW%U~lKZ>KkGoHP*kN?p0&^L5KCen{g$zmYtPd?F
zi;nwF0S<cj{SP;u$=!M=<Tm}M>)}X>5aNX2G`<}gjBeKJU;%`F?uV!-HwZJ58(6%6
z_u~iwPp5){(x*J;C==TiBx>q@k#9EVq2cd+dPE8lp{!veE<7JEi4b)Z@95F*ixsEr
z3GJ`}b!O1s%iXt}*>6IlF5wTQpefI`T8*$&CnA*vcDyrX0yC+wEF7FFVk6h;IYtmI
z%Zc{s=XRIBGm4NSxTngsQyAE@#mQ;YL2ai~7-7(5o~Q2!!k?>DYnHIVo4P|kv%|QK
zvd4!>F5eJ?i(vAvMckOz0dwcJhM_lHB6{3ps?frr*80PtFbDjEgXz?n4B*00xV@N8
zZV`I${?_2F@cO6>bXYs#T#Wl`y<ukhAIviU+;627L_R=!clXnZ;mj8k#NgysipG6I
zLwj2r{@K`c`;JxljxC|R%YTo2&u^+0H>Cj9V{#WEU!y`n{_u9mheS-K%{L7%yM8hI
zx6&<Ji|k7>Su{~L$7a8f{5kbQI`|(ZCV)-8K0<~2bLVu%(4tEHE^Lv6=nR9j@;XV_
zqj<8?OZ5kUX%GShIcIB>B5O18+;np9HrW?kOv>{~a&E;|0-&)MBH$dtj4rE?3dbyi
z=^^e-^Y+Zr03wO;cq8p+;mW{_^8H?<F{v@gJnNzDKNH0edO_*q-9~d*hHAzA*lppC
zy@h$kT`zfb?=bk>u%@dhSAsE0(?C*8<Ng+|C%%-7jF&s(#6y$s3M->q{zqSJ%%ge?
z<%}gIuR@y9yu4c|zHcnxYyu|_@R{T#_z3qZuUxg!MqQ@_p1&1_-k#W0*Cj^Z00k<T
z(U2i$O71<Y@RHBbXKOnXZ;XZ|Fhe9CwICDLsVlu6hwQ8N@lX?~!=wIt`;(^^;2vNj
zY-q(C$|oU-cupkYar~j0(g~6d_v|>ZH`mo5Rin5cz)wgSJfK8~%65bB>iX~u1$l5a
z6{X3WLCnl5u%XFj^I5mE0V6%*fdQtii?;QyGNDwLtb@(U9NO4XsSIc3z|$kxmrCc3
zg{vGeT(dURc4c>@8E3Wr9-(|lbV{nW*mfyvFl6xeh9XDTxM{{T<7PNd=F2ChqXkzD
z0j_}NHK04pla4W`xE@sM4kC(80rL^oSRo(FI=KgAz@oW82y0!irkd^n&whL`@b}0_
z^ihhM0T~LB0xBA~=%If`7npKciE|r)C)}>mcyO?{4|>4&LmDx0#u)AX0D~fv*^=$g
z{MPI6sYf*8<ibT3J-5}UYHdP<4c&4&{vfk=pQ~hxWPCSP4Unnt?&o~plJ1E`YwEg_
zy0%dI`;ZP+#y{;VHn$z?W~kqM(pPh6C@No<kde0D%Z{@#pyW-K`mJj^FI1YJ&;;yb
zGMQ3g93ieiBnij@IqSHgkcYsNfJ`nhTW$52zt3vuTOGhzTIIPJo}UXpx)$S-0c=UM
z2y3zT+z-xL=Q4pr>l#K;QN0|lf#Ts#PO}4wM_XGgIwV4c>dbnI?I-6hwpAA$0?P}Z
z-WxDfbsimASMXTHuMjQXbLC!p^l_g!x1AFHCVL4sM-flTzH_kv@z}waPZMpoBza*)
z>-*!OCBm8Yx{x<oc(XVmV^F4{H#|d)ULGE`+=>r>j}SsYVPTycgn8%j*?Lz#Mnq9*
zu{;##yAE@!64hm&SMMD0@8dG4oLT4GRto1t5?P&%gQv&(M5{#hf&fkD*Ivh05XolO
zt{6Z-3VUb>D$;ULZq^I~ri{d<>WjO3%ebn?BR<F+*lFDPI$rrMep={1U}m2Fd8}VY
zpU79jc;le`2-v@i%Y62GPqo6a`gAb07wyhY-73kEZE5IbnpN=8q!~&3kjxZXJbL0a
zFc?9*`b909I{BF$AmNZJS3JkNW0TDPGrbI?!JH68eT)n9(0B&F?onSeeo@aT=7iy*
z3x{`~dMnGE*B(7=ggV37q6X+PzTq}+rz%vRCPoeblL+#D<E%`Pl+hCO^?3bmU3607
zLdzQwzw(f?<uX3tYVW;Z_E!2cTv)v^ri^Q|CEhxx8JQHt>_}{~#;wSd<$Q5v37<_n
zTKk~$amuGOdQGg&3C9(Wv3+8=7v1doq!deW{K9@_IVQ{n9fMw}+Ba{X*HRAEWZIc5
zAms=U5|JHm`SLah%24h^w$x<s%>a2*-rlz?0GsevcmXS&lK3s_!!TkF2*hJRC3P4}
z!S`ouz;d_ZUkTK(QYji{!R0O_n=Ogtk}0u{5I;VyVzTK&L1t~AX^7p`sb+Z6B4Sz1
z4y2bsDxhg4zc{8P_nUnsYW;I<4xsT*+?9S@0x5yejd%F}p#&Pg9~KS8MVy_<sbfr%
zof6W19)O_E9N*VE0d1g=0#ocDrqk(#PSCgGa>dll5sdg#dP03n+HGxDCwYNZ#m<`V
zLn`^|i{T<arF$l%3Bqd-diOa%18n`^MFcNW?>g@604P#Y0gs(kaeEvO*JoAfE?Bjr
zM8L=sBo>C97r`3{-Bc-=9W0Z4xcWQ+P9dpHtPc@}axAS$qG`_S$|kp(u7_LpN7f6;
zdgr3bv1n#AE+mp$%qG<1mBwt&bJcGisfsXe%i~N4bQh%fj>T9C7qNuO1y{M~lK_MB
zxk>5ack~`!9#IfgqAp^yHhA@Kc9_37dC#I*o)U<=P=$EgK7KA&IyfvP7-xv-q(IKc
z8R>loAl194>zZT)Gf`#*v&eJT{fCg1G7sRAfZ-AFaRBj(#kXS9jzUTu-P$kwDG`B_
za%upGQJ5sKZ;&^mwX^L^$wW+sf&3dTDJkn7dhRmC%8C`(s{EgElFS{v<~A-*-gLO1
zdhOb^92w)$ERwP7Lyt9=I9~;qoR=iGW7dvQ|CEuH(*=XUzmh;nXd^2^UA#i23wh95
z5m)Qe#b6qipgN3?X6y5gRG<FbFw+Nm84qN6{$i+kl)2AKx&y1Gz-#_5xMM+5Hld2O
z1+T*KHKU(-XoqpU89@${DTG5;gN!CVi5bQoiB$m(*KMb!d)=SI;@z|;bPE*^I^t*E
zXWcK2Jj(hUaD7lC8H$4Idk6aOt~d?-gguapJ=4?4X^WYIK-WDT)Ut#v7DsRI^H6?-
zky{mSK4JL@*_AB@{Uv7AM#}T>1yD6Z39kCC4us_(Bi=h_<xGw;gnqwUsrQ|9c|HS_
zH-#_i#s1wz^lTaut9<r^Lu1PZ_c_1xZCVnjFc$A^l{TG|4{EL!WC!M6Kw|G5UI^a;
zw302=*5v_z#bj=Un3zP5?*M3*4p;!M4RoUi4yKS+&3RwE<c%B|^W{6Oi$fyNN*&cN
z52hsk7vX`=lYGqGg=->g;QQiRy=TBZA77d{Bl&2X^3kX6Wb-OtskRL}K0+YGN&Z?f
zi0V%KR+yC6u-;9~^g_vEG~I8}Ifybner96L&}VR!B(pEFDz(?stJJ)HyS{x;VUDpO
z6rpB8gh5tu$7;^D51gS3yr!*l?xwBll$eC7ZliO$rF<<DIfN$<fVd0r(>mgBKn$L9
za^FB0qR`|;7qD^Q!2a)%l%mGqY76LM5G&uR-K1I!q0vR#_VROi7cb=&*={vpPbE8V
zKbaT$Pfq1<tIvn@bS8FEj(niY;w1b5s3|)&<XPpN^KaSO=&I?ZOsx9dMa{;fG1fiT
zsk(&P5~($XJ7vH2eR0qOf}0aeDvY><-5-2Dc(?YyP-a=h4&pg`8RjdGG6x;8YU8z6
zuq5O_YQcs2G_R|jZelpeX2jszCCE-t9mD0-V<#S}(Sj<B>;<UNr0*>(Mr^5vA??~>
zDF8~+J&m%6bjS2_52#rO13Q4yy~^QoAn$ellc#4_W~z)V5l=NM+n&3>ZCx;An%ZbU
zci|W}FNW;svWBKHY6fca#@Ayf>3yp6{2(Soz*>X4T#kKr-j*vJmu;0T6(@7mIjOb!
zG(TR8m3{on7YAx79}k%@D<N{p(RYmzLKJ^;+S#h176?i?A6Rr0dK(`IPv!|6)^cUy
zQSu>1Z8p;#(#+9zHo-qWbqu50L~R9KHo0fd{9oWYay|c7RipQ8{}-Yw6t`6~IBwVy
zWYwd9j;v}ecD^pQT7SKXy?~M9%yK5BeA4%mBPh8ZZ1Cc6Oa-c{ostlKRfX#Rq=kX?
z1p6p)XE!9pg?%HFmq|&fm=HS}xqW1*#M7}Ph-!zyp!skiSf3&^^vh7g;J#2ZC^ZnB
z1T{=V{*5{z<<O76EVS{hY^@6aUB-?~3tBn-u+GaJDj|hL(4K&~qNK!4so3{rgDTuP
zc%czpe7NphL#JiQy<9O88<?tlZ5qEAV2u9s$@+t&N|9JB6WOc_u5Iy79<~m^r^&kD
z{}NFF_@z)0ywp4&cKhNd-igXoJ3fdm`2o<LiH`Me_bZWowBcQm4PLp}9llm1eKoy^
zqscw;6|iJNs!;9X02r^`agXS)2!8WG%2vP7?}}YjaU(f+t}~MpL2bdQBU`Vdx1=ZH
zK#YXGGn}S4(!U1koZirsUNc@p^0%KP?+bNP><fmXz(J00PX~E&C`Tp)IO?Fd<AqfL
zZkP4LlZi7z!m3JxJX3f1XX|GOIzq06>(Yzi4_;T@1H?ULyzVh6%{JsiJkSOsPlDJX
z8xJlNy7<t#1_UD$pi7@90)}>Sc-d^f`>Ysr6e|UzGLHwPzO^rSCtMF_qUxSj35YG$
zu?MWN&jXzY?BC8&PnUx))&;tsH__S%ofUsBRL2kv3Z=q~+o0O83yND;4nHam#buD=
zu9EW;KDe&lJt$1SRv(kPgMC~xJiQy?r26rD>#UlQ3<a6OhnirjkatwbzhzcT{Qz)x
z09jEs>YM2ULsvlU<fi#gW1#-T)9Oka+9zFK46Lh2X<Z;S)P@%#ueewMST?;`Ih<wY
z!Me1sYqtWT9>Izh2F_fqNKCYz;iya$sdrh*@1M@<_#ajv^_YUa>Hv`i&m_2$vG+mE
z;qUf%Gy^Ki7Rc$au-35#*Y9@!dqnl`kyOA#wk`R(;@tP3h07GKnu_?(KD7-{cSJ)E
zQWG{^SUo1nrU~_`$a-a}&><QV;!*KDR@i5CvS#on3>{9;OS-%-(4Sc3?`C&@(k3@l
zj4Swc5%2SLvv@~IX-fPa{Cveb;tP*;!yn{FdPz$|ZwnbJMeqjIn_6^>?&70cR<??-
zIg4uBYFmOBX0#u?5UjLHF~>*qsAH=pfQ9#O>Jp@_Y2x*-db(wAbk?Cj7tlZUY&-tv
z-tfO?_J4ovY5d=FD<8ueRxqV4JQ{{G-gVtQ@#v8S!+9Vh-IYKC>rq?)q~0;%oc6Ub
z_IschuVY8u=Q=R&CvHl0JAc8yuG@4+5~0#vc3@hc_T1b209Pgz&BHXhR8-mo%y$}w
z(}MRo6mu_aw@iPJgn3JEZw>}#HS?;{4$`_*DDU^U_<^0qWUJgj4n4B8o+_@0dPHJ8
zD&pJhLy|tZR+VA|dSpza3UBql{<s;Y@jAHVWE@Nt2*mi9Qu8Vl7z9h@;O{s>4IT*|
zSI7K_{Xnt}C)l8pf<k4R80+f&<Iwr7xBvf=yLbd^OCX@zvIw@dE}Qzs>gDAPx~#!n
zQBRrLU6__PXVM!w0l`3Wl5Tzty0_EqbF^cC;y`Q*oudW}xVr8p<z}Y;$jpu90uZ*#
zn%(w`y@CVYnQO2f+`MpB*30Ur;Mlyt15F)Kc<F&E`S*nF8JDB_kjgt<kPZ0b8-1~c
zG5wt1_s!Sz-IvW6LrD+IPLn0PJ;T+Is*}>Me5?h4S=+vG<!|VJd7E~j`voKXESUfz
zrKbz>e3EU%-PjB#z;$ybh@^Dv7`MsEC($stIwn*G3>Q2g)fr}0b3<H4FRLr;T?15l
z(Q4c!qPCaG5DM_9a6w0lY=N3U-iQhMN<;BH&YE%MhI#4W2!37HRT!c_jALk~<1B8r
zgE~!RoNcIn3-ug&7xd&bVxpaZK!u1E91mppH+ee#Jp%s&xzFPh9}Ko8)xVGN?gLGd
z<Lk@}ZKLy5(}hlIbrJ{IvQ5LgyD;KKc+?l7__Y3+jpT(*?Ti?<TcISka3WRP=>!YB
z%%G}dVn<*~smq~&zUf=ZcR{ik7IIh6VX<enZ-#eQ=paK92N|ecMTqJ$2RvL5*x0hZ
z5=^K^vhLLMw%syUasQ)_)45B9NuC#c%t%~2+8XJ7JIG{5hE1EhCAtRc2!&Q`Gp@CX
zUf_&6akJb1Rc8^y=7JgZy9;VbXsEL>XfL&Fyh=CrWGXPeK*0Pr_$--|q526(o@gNo
zO%1-@FzVt{_iYH2xh6m;>B72&OTd6BP<?$mG1oitKNA-2uoQY`aw~v!jkHzb>*^eg
zjCU35d+j8{IkdRH6+`a*r?kRqx7+yhk*{V~!>#Q!hIE_H>yl$Hf_{JS8cfwxg7s}%
zr|h}V{K-BiAf5+z=kK6oD5}N}{~nPbFw4VsP{COI@8S&D?Jh`43*Wa;&B%l$B<deC
znjzi;pGGH__{{Y%20YO`uxu5KlL%Q`7Vm;d*rL2i0r4lwC@oauR?@WEss*7o+$+2=
zV7H+gY52Ti(_?(v<^DvA7`<|RG;6(1Y884K#v<(~iR1}Q7v4evSHd8(#ixx}-=rbz
z3fF|C`N4SGF1#*rXh&QBUNg>7Xi0Jd^B<;aX1lgU42g~72}uu0D+TQ!&sCCaMrqCT
z>tx<+l;k-N@HC7i4>JE3Q*k7{SbOXfNSi7(bGS|6CoF&~taw$ygx-^A1EvNacS#zl
zi@(?d=Nev){M>^t2wCw`Jtz@IVc5!)PXZ&N>l$rknK(JH?wKGe0DOy93rlvsx)<uP
z|6AIu?)T<3|5G>4IMK;6+Z#S>rZ<a*RXhZ*0Cz2+fjivgP5dHkr=X|#drpsiw`9|l
zEpw01BC63BPhC)Bt?oM@>I3)x^{k(-5+rWv^}byRXUTu}w5}{Syq|RZm+ojyU%s$8
zn4&XfJ;PkcVxYaM!E&nC)t$5kcB|2e&RptBL!jb0DjQO4a)t?xNEb*a*REyj4J}rF
zPhOP_69`Qzgt;S0&uaR(41ahwa4^Npaz=3?O7*(tElIxZ=*_`(V4{>pW6F_%Q=4dD
ziDBsXnaB>GT2hK`vS%|=Hr<jyas8~)|DUOTws~o7QAd2uChzUqO$rMd6qJo#PeLw}
z_28zXh3e&7*3Wqje%Kk(#V)8V785Sc3^VF40;lIQ@hs(GlD_0kj9s(><4x9=xLi_I
zslolu^wu-YH47{Kt<J|H5EZ4j4_Ze8C$C(!xG0rT-Qqb^xSH}yX2xhR^b4!J=JgEI
zNgEDyCMr{P?XK{;WOld2E$<iR))^1Wy!eD4Z1Row<*Bls$+D>;8EW>t3_Lrlzb}#U
zJek^_qNRB5`J(Mintks#VSMF710;0j+L;PxxsMzT@4EVJgtAh(S^UP=N_Om+@Wy6%
z$OXh3vC(aJ_fUN*cgbc#81#49L~_qy+OM*Y$~dJI$A*C#XH^>7jF51M57hc<FiP~x
zx=JV(#mqYivD<k?T*sLgJ!{6i#y5BY`XX7C9goyUS+c$|w%Cy<JYGBXpzW2iyCYON
zjRwkbvkm+n->QXQ6d3|erhKFfsI%QeWKnuVlzAfe(tP1E?3ns@3^)anp+0R)7jtsr
z?1PObx>H^JR2W=m`k|>xM;08qhWCsn_5RPT7wrV5T6qH8CO;w?dRK+4PG+@_y954i
z_!)p-($mPw$p%Vp5ELi8G><d4Od<C8en7KNi&g{r^WEd`HwE-{T&A~FgN(hYi!@a3
z^U6!11p{EtdeqJ#!4DUlP}I~(QP&hX5^HEguP$Z48D=6H$m{ipw!|My=_je~ws8O4
z&iWO`YqneGr4%p!3-LbIw4-L}C~npHVXfiDrbv#r%h}J&hf=Sr_X`G_kpg#4v0LA%
zf5T(Ikjf&l&Rnqy1HFjYQ2|lgVkkv;nO!M?Wunvy#s&X2asM4tF-im!z>~RUo}no}
zIZr>AlkI%>JGN|gN3n_i8h3E~xlm_uR#EFg;@i(myE7lNy4v<hJSy?0CAB4&V{r9_
zYF9i7YZfq~u)-vLn(99q@JfIl5$h#r(R2$uUy_b5s5(5$gnN#D2g&oP6kTjq&9x{x
zXji-hmQ7_C66p#K+1nnlt2y~h=`LjmktaU4`l&Vju$);2z!4aRsOf?h@X$yp`5~|Z
z#8M6)jV)CU0qzI^N`z1)!DCPkWJ8ZKSUO&4|8{o&?PJpGbLQ3sCb0;a_MYZ#&a~**
zl~^4UryfP-)-x5)(OdzqM{KD@Ms>7KkFyZB3zPAq)=ul1KYh-HFBv2+YF+xR`Id{;
zq0No_f?v4C@SbHqu}OiR)v-wzOGQ7cbZSt`lP(@8%^7kz*7KQ<eCHeUe0<&&1*>5k
zb?}Sl7ZiF6+;7OIQC<={^ziXj?hY8`Q~lO^*O-@Sz=FX{BzSqL*O)9OjqK;ntp3+@
z58Vvhs{K8>A|-W@16D9x2Yr`+2q^l2?d01*1pwQLXKh-fw5QR4{UdP+>d+i6LI0z6
z!Q5h|?hTiuS~xtu+lnO=>yzT;1)Q|pVnJ*tE~{9JX^D^8%#MAd{+v{f<E2$Z7f#8T
zr_E>c86WUhY*_>UYLwskLkKuE8Syp53M0^gmmLLQ!8`;7Y!G=uEE9mQk-X^$_@hc#
z7}yVf<>mgTfVf#^vGR|=RUvaGzZUK*fBUC^7U2r!R<Qal31B=}BqeWNJMhovG>@ow
z%Yoybp_d+N!{Wq7d~X%P+58`N?d{oP)o52M)$<_lFLwo%`ya&fyaL<e>N*poHUksn
zrOW_H6EZD~@{!V*=fHlI^4Bc5R~sg@vm+NDSRu{>g+s_*@}|j*4M`nYj9%})rgJ(|
z))w<cGK^<4qXmPy!dT}b`&ToS+7fi9*FD~r3HZxYtRV*r<f4;*yr93>N?VS(Kb#tj
z-S>6O#Hs8`5A74Rz36Emah+zL&`XRI4;-V&#;+Q;r^-De*SC+9w>7k9QTZNc|A!-Q
zeUx%X!!_*R<rpI*gi$;xu#%$?BP)zwjMf66aa4GzS4?Ri=1Ax&H-5_l3vQj$VN9#|
zuw3#;rWRO=sn+eOdg~QzX$10L*R`AXQrSLrJ`XMC$6n|^C6MRodfNWrl=Q{2!gJL1
zOHJVs*IvZ+e$u<3MvZN;)9UPOPEV6@;G0e(cz5-_SG)Sla^JW1MpTNFR6FFUZawpi
zg=(osh#*`cD$%Qa%D>KJM{vn%E7p+5eRRs$1|OaBC-3`dvEe-Os_MV{7gdQ)S=X05
zx>t>}9XYLy<wn|07PQ3ZJlz7eUoOi!j{@b5bWN92iue~w*f(u*?P9+=Wt5QF&*EYi
zJ%)HARcMZu1K;Ht-%Tn9o&RGR&6&F>5B|(XqLq0PC~k~Mfr5iV9my60P*mO(BJS#l
z#kv=aY$|KU5TUXE{AhTr$70X1lMA;0Mt5(!>%@!=8>f}K9^No`Hg^fQx55(MIPXNB
zDRsag_92T2%d-lo)i=nvWvi)5MxAp72bsHsy)yDcZij*oz+XxrxQs(1JC`*urco|%
zC=(ThWf_~7^u)%G{8M5@j!BIFG&iv!Vz%lc6)5>Eh_jGf`Y%ZJ4B3*-B(;~&&FTe<
z^ER}h0J-MARb>s+M^(6!Drq8`s3G;+Zj*#vyb|<jQK5q6cTxiB?6c3!$$u({KSj(~
zH%}}SUwNQvz;2BY;Ry`Z@b2l+G8tnJyn2x*uc0==t&%iq&^Tji^hH=If5?DGpj7&7
zEIp+^a!OkGg(pfQCCWh_Q^_jQIZ_s(peBZQ>O+C5<nIw6u$WlMqt~QS3ZBr=fz}g$
zik1j&-(u@j2-B@B7kEM%!L0%;pH;o&-`I(D#LI$DD)j|24zEmPQ(#9!B$bKU2nlF8
zFF*qDry1)iRHaXpa0r->CBjVd`yV0RMYv$^t_j`U)e#2rRH-#pik*GGUQp@KHH2WV
zC2Nr!l097Iq0N^JKjt7#6!cl?p}_q91na+OUI2-=9**wyPkRE`=+Gq*G$T3Edp3K<
z;OL9W_ON0MJvcMyWG>u%N7wbT2q@`mH#|Tfap7S=(<{ucVB?}N!AnDGQrDlSw57?#
zvMG97gnHL0Xf$o{Gw*0x)68iV7v#RMKj))lq7*NmL%;Sz4_l`^@-w|zCO9XwXvkYv
z8_DS)dB+yvJ9|!MC8{)N`ao{hcXv;V_M7IuEuXO?$r0#oHTZR1*IqT`0qolRrKIFf
z{R?K!YJIgYWhNs}*!OjQ&1e_rb%{&iY#PcaPHlCQ(egIimkpmUVm_<_PqIqS2>|br
z!F&uX5fC$8hoQneSEY&s3VA)@hZ8A!MTF<0GoOdCiS774MC>%KhzCv8N9)sKCkV$U
zqI|yG{}2u=pWBEdK@HO=n#rB2DfL75b*GGQ*?gP~PkhZdy{FJ6mU1=HgtCi-M(5Jk
z5re6`O)%Wzqq%jDvV{ZaE#6THEu;j~&i$1NfAT);^3G)C;`S@7Ki&kOpZ~d(iN3Qn
zkXmLc0;O$9{{PU(z1w^CPJxXT02)=?LS657?XqG`NLfAT=yvb9?7>2<Ix)5Rpqv5!
zDk%3`bu{V4H?&Q>6)xKg+Hh>^!I=ZWgNmY0<cN#^bylWd&iq={;C{ynXI@9n$3L&Z
zjC29}T%>`{8`{6JagF&Ho)#SefthQlk~qo8C(Tb395_=`|4iHFdiD=8Nd49RWP^Y$
z?r$-P`CdfnP^;04wvoJZXO_#S{*7{Z8|D0(@U3snA^F5OC-F@RRw?`{vb%cq@@FL^
z_Z+p$Z9*NTIr#w%mkrMD<wdC_T;X^erAq%~lI|Q}yD9KANN6GUy9eE4u|odpgRHzE
z7TEa2T&1$0d$o`$CX=!#o1?!YdCR#^AUJ(B1vqwe>-a|Eph?HH4Jm&_+6!H>iaKA*
zhA(1+og?7qN#k`ipE$m5BOU%$$_h>!@dlDEfhw)?HSb`ta95=ORfOr5qM~m4rjgwq
z`w}4(r$qAE)nA!zwbCYsxI8x(rgz;J1@7VGE%{iENliOomwN}tBhGWk9*EfAlaINo
zfWVeU3s78)&|aZ-o{SO&>jU!`g=9v2gedfaN}a`{%e7S}l=VG+8UEm8T78N4w@-vU
z1`efRd=^i;t_A-T2`oD1rFlP%Bf%@v>skHVu!GT2>i7bxd+fc|^^wA1L^#x~jwa|x
z-}s#tdA^I2F+~Hk!iA;V#-@p*`33hixucT=h01(Y>I#OD+x-9KL_!4x`vl|@6Q92n
z?R<qy^HHdKboA1@<!nZVrglIFF`qioq;31hRlR;+{U+VhhBnXc-~RYxFzzCZ6RbNv
zG4NI}VgjF*s%4zd@;p+!*;IPz`6uCogL`J`^Y!VrvpQwxHD_J#H-$xKcm?q8FE=~B
zX$-z4trU~k7fI+QQ%p6VUQ}f@HWgDNp?~ZJ?fqhi4dP@Ri{BR8T0S$OJ}9hh@1Aq1
zqY2EhSbb7zgZYOSt{|mC8ZDq|y8ISAWR2G>PAmoNW)<?gANB$Z7Jwypd?3}SVK(P8
zX!3V>LR?!d;DQD7q|r^B6klj0DG(QZ@?jPh?Zo|X^acbsU=>nb;grW;&J7mQ;caMd
zu1Un*mfsAgOsiZr0K?}g?QTGWpznOhtu+7da4AWiKq%Ie&Wt59y&d^VEs2YmP#6E*
zv`*%hz9!bgq`9T=y97J@$;9iE=Ar}}&9x41u@?S*3D@axHP|l<ACdTkx~H|JM$Y-!
zIZe4myZAI;V60p<qaYr7({1-BJJ05qMB+=|+0~Zd4Uvc4p2lJJvMU(R`8UXas(;+R
zy(LusE>JaO{0~A_&U;K$lXpS0#OOveyoIS|^jX!=u`xDXp#0@Ir?<U@t*djy>u-iu
z0|MCU)4vbhd*mK@{^W6LrsXdvKkcUOhIzDud{@C@)62?OeofT0L&l1djn&*2E_D~i
z!n(@+y`aS=%_~!9@fok&QxPL$`ZM$1<8_y&u`tAEh6BV(tAIf1q_X5eoi69A7Ic@C
z)@OdIoKkEIYyY@o>71`@QZ5@OJwf7rBajD5o~>~yfY<>B5?Hy>Wn~wyb7;;iPNc$O
zF>5;lo4R-vHlbUF%P{zH$J#h{qR;^-!|Il>kq2Bms4yHetYJrKU6Bli-_F&flVFP4
zrjkGI&covexMDi0RKP-!{2vS9^29^=5ReaDvA!e<`0r((fX*GHrA<V`pm|U2YnG4b
zJBO2KQ<r-kY%Uq{rgl?`%cO^(l@B!i362Ow(i#6o5(?Tm;C?J-Pa3FWsIEkE4F<5a
zd5qj_$5s!&*fVzEpOK&RoA_-;Ck!L2{b$nyK2M}ZPME^|V_DZ?iEIsLpPf3j8nDM{
z(473D^-^S4rO6h|+v0>~UF)Y+C7ZCpgGpN-{o{ALsiCEkQ8zk%@)5>TXCgVCTyzkT
zZNAj_%4IC?w2}M_oU<w4P~zi$#^gr>CCk$ng>Lm4Wa!eWGzSOIUs<Sd@DqBFZYsNP
zkox4QL*~bnM4fv2+@?s|SNsU$`mdvpKYB!*800S)eDpH+nc!UZ-y<1XJ#LE08X59%
zW&9vbV@jz7rtX&i;a|roSBKQV8{VHNY(Dw2aSJ=BSjt3AK)q~E^3?zv&CQbKi_!6m
z_%m5q@#rm-%36D3Bsw|=C=B(kt?ZMy7K@GdAOaU8{v^=bFN9H|%x1a&_1|v!I<TKk
zn)^vQxhte_fSE*^F{LEWH3J(9LlRHS$pn%0%y)dZzBijpXATh4T9$#Q4v)|kCBH*{
zM439ZK#A`&A~y_p5u}w%y%t{ecv)@^Mu_LGxM<Jihoq;c<7J0=NP}s06;vTlZr3z&
z81HH{Xb<~yE93x;-GC{uc^^9%j=Q)d<mN;&C(U_=D))?rjSJ^>%c4RgodkRVfUY2B
zXizFetk~2~Jaky^%2L$i9lML}9SjOqJCFavS92cT&o$dB-AbO%ShRnGthTyf-xTJS
z9{wf*J~qT|+QHCJ{%I|KR>R}l48l0_SqfIwy?+0y%)0>H!BB<@qVi7=I)L}e;|!th
z3s=e{Eu`ZNLaXC+7U&Ug9>)K%@Ah=RBP{SD*x>w`n(}d0u{NWw4MW5R?BCmE6poAl
z;!T}<>zjt^w%PIO(8ZjxYly|j8lk<EtxHtcpyjQa0rO|iw~^@jO94vc`<EmMaZ*I~
zTNKA8OS~TzHF*oa(U))XwN^ZQUYIjd<jyBvV~gyPWgWuVxu+C}i&zB)y=efzW+94;
zhpvd4&q9oZms1`HZJQ3C6_-=%E<oQ|zMwAa*dsAatMyxF{s=X#hx07z|2mbo$M=~T
z$u>_wK@=P94%pe>9){YkSY@4awpgF64M|7R?a@fCyYx&x`E`kq8eNba!(ayCB*q^^
zz4B<N+o2*bmpSev@FQc;2>XPG2TY>?ub>QVaI%!{0(ns{|G4;H#Wzr6srVQLmtqAp
zFV2(K5^T>RCD7eIq|mS<z;0c)9<X)J42~ra>-BI#T$8&vUmQDNJy6?Yx65N~q%<2|
zgG&>0?A2bT$(bl+C79i>I_7SoIXNPV<M4k{c2Vs;tTEZNl^0q+xmT}sSDX#qI-U41
zQnf-;yf~rfm<eAIy9+!(`4~niYf~ieE7atgiEm#;4epqQmK)ENz^@dxb?q=pdi+m8
z&dEJHhcbE0dc}u5xx|aYWw-stPX#^!9My(j<eNooYh!7*JY+WNES*hwQM@zSQe+L~
zr%DHDG3xs~DlgND-OhuW8W5?_7+QY;=(xG@XrL{t=Q3cmNN{9#fW?_^A%0lWi$BEg
ziUa~5smd^}WBaf(fW5fP6XNZs#|LWdL=Hc|6_m%hlll-*4Z5FYhh{Nj9XY*BXE^O3
z>8pwhJ#l2`tRDO;Mq4GW0MNpnMMYOaH4XE_jm7BUm13g)OBQl7`tCkZm~E&Afr}~Q
zzl2K=>Sani;T(9h4?6)DcvFNC!{7B)R?P3^&3XP$au-+vaR`N-7u`I6>pTYnKQ}SU
zkmJ)7*piK#R%xkm%)B~^oEzdJ|3<$uUhR1@y+ewNe#{i=BurFE?)o&yw&GAI2r)CY
z(J22-dj03p4)ax^QcPv_-y`?;2DYpl-e0qKY-sP<4f#qE=1urr#wV9lF_uf-sxyC@
zOVs0t7wbYbczH~F=A4_pe2Q%5^+Z|pLjEw)-|FA8%Myy_xjqs?s~+KkpR|M%lg-In
zuQX*VzoyKJ@oKT?!k*sua1v@s8Jn+DkK=d0HpW_;s5zuyBumODS$Z+LwYO)B0!UJ{
z8dJp<OJ|a7+R%-q_a!c%Kqas}x2om?=1aIVG?=B<P@PDNp}o$>JHx=&p)B;@0vJPG
zE~V=qr|u}PC^h)z{2;jNUCX>?ldl|CRZH{_gb^cz%oui5hLLRXm3GIZHuXW}r;nW!
znU;w{qPzy*q<md4cPboExcYF}Yy>MtUKFrfb&<(gA5l!6iMiUW^2vUHmpfOlOMcGm
zSDQ_VO1b^Gb({@}wgk(q_&ZX$Gl!&B`6%L&VG_^PD~)@HH}%2D4{e%cP1Q(y#2<J%
zksi*$8@NF$udbob3d5&410E^O8GaIlPCS2Bs_|Z^rAL{La@(`#ihM90t+FGfFE!~r
z%J$apCGncV-y^k8ip%>KmzqSTyomn+m58dpb<SeM=SR^TrH}GY9?io)$S-=G67fNQ
zun!J><$GXd!y=zI_&fATp1-vqcQMg*Bs%J{#p>lo@!^YX^<Ow7Zv+~_MDHF8dE}cI
zQ`&j6X-L2y`A7`@x?9y-@fG6Wq$O10{-@PUcDD#?OFPN9rxuWcnOaj5oyD1P1HVwy
zj>)^x>S1$+lWC|}c^a7W;-ujJ+d-$GAaC|uxd3!Q#@ojVx~}Eui>!fB8Ca1Q4`7*W
z{bU2sIxR!D4ml}GUh>82RAt62(T?~hL$^N6z4>NiL`rwexQu0P2-@RNY!BFL2+B%i
zwgC+ta>#o}8);N@nLROe8GuATP^DV^FNT3PI9&Z={`O@3?}HRKt2!m_>2u8=Fk1EZ
z(*`Tn9r8lro{!%5j=#F{JU#ofSo-pZ8cd+rz)8IYiyim3t)H|`ZJ&OhM5z1sHu5Za
zC3AI`&)A}*%Lma0HLq=8E$<2Q{rAIIN}#2xoC&{+=fL_IJj^=eTS`|^EaHRiSAN$A
z!<J|dAzArUvJJnPvz1AWp4Xap$ZQ?2gJ`S&UTi7BA+aSyminz%tn>Un;n|fS$uSe3
zo4n}CT(V$>-hxB7zt@+`d#8+kpFB6}hd35aIscPAT0uAP5zCD0ioXdx79|nrar3(7
zFA1`gAEUqQ?Q3?oV;@gROz{b_=Ku7V=JyHB7TyRc(b{&|dN-u{6wm$7YgStC98hEk
z9F|?XDml{?l&B0H^ycBJ=^qjFoWPC>6><qH-8rc;AySVy;H{&84m$(1*y}ut@#k}3
z0C|@J=;;U@aIWYzsq(kUI?}hWzH3~dlgawSHzi9Jp}R&1OzQH`8>wEoc730C&%@hF
zSRm_>5XaPc!nqfc<ll16<d|;2=3vNnOIap&(>m1l)K?Fyh)F*^+v@5#I@UK{150e{
zA{CO7cMyolJmJ*{K_sgPaa%{ksernI^+X$KpR+2MH&P|gGD;b5RG8*!N{=LGPCuZg
zTR#114Z%0$uVk3Ck~4nVXkL8b`EgyxVzsYh&UK)6z`~Chjpy#Z>f4*ty@fRF(^7nx
zJ}Ld{q)xvEs>#rKm*3s*Rr*Bm+%A{HCwo2VQJ2Air?)D2$|s416a%iRebto%h<_F;
z=PQNN@0onri2zBMDRvKXWwu+QGa2(61D^`4v#yYaqV~5gh{+{CT|e89!H{O}xs!hP
zjA@?vG^*fSYjNcMX)!#hw39+KYIN|^vyT<p7m3UQE7$gmvWB?ar_ENx&cDR63jW)J
z5EINDH=Ox%aIfQ3XW}}b8f2hzsM5@_N*&huy}>=H@JZzH7M@Qx3UpF`!a^f)A~fX5
zI1n2p+_zvQ)a5gPHv?HFr3Wrmy-@3534{lnKfZMS-OJvhvKorswfT3DRvzJH;Z;lX
zfV;AkSuRnD)BAj#O@qrp6sHoVok3WK*WKhby)<OOpwBD9Aadc8GLJSjuAOyNAEV1<
z(@Efd_sZ~x@cH%8P4`{4Rpj-8R(s`TW~k68FZ<h95e)|u#?!h-9fgA>;$Y}dwjbw<
zjQ<{)uW{))?%N8XIPqVxdycVJ>D(pthK6&gpRwHXa{HQmz-8iQi<ADN=9u#mFvYa$
z#MKYd+X+W8u>MFmD?P1DoO4SN80xvCsn9&c#3tzv3Z``nUNA33giV>-gtVxchNmxv
zh)w%JS++wksVk|$bokjcqCdUkb<i7*>ltKn0Im3EzM9kHV)yQvg03-?oz%L{tuQMS
zNz8fX<G}QKre{iR?%hga=ganMdsm9IXkJ3OroNIz9_lw~5vA3By=eoUI$FJ3uara$
z9tv*rQ|>G=1C?^nvy0b%9X~sq(87QDV*g1^$8?Qk|4mJ^oQwXC-=#hweo)H7R}jw|
zmz_H+OPGG8B{-po{Ja>Xc`*Uws2Ruaq#A3tqgh;<1l$@tpl_7TshpQXAFM^!?I#CG
zx@RxF83<@{Fg?J1{d;8W%7@r@>cN(S(Z@o`X2hfAxF=zfTbLX1t;$KQ2!ZLTx<${;
zGYMTk^m#fUKaAf{PTShuMNTzhV;2oeK1=%B&WmH8Zl#Uc-5ybznr@{mVt(v%BUUV@
zdkPp&2geVxNW$%>>Fx`91SoqRPE%**=ZgmMkL!c}9llk2=r<j4;(+}7_25_xaaj>;
zC2YK_8clFx_$04#ze1atRVcoHI>flsBAf5EL^>K$)}4H&m?%zf)6e=2R!;4c<Tlx`
z5CZy+l!o33Wy{1+s3ZP$Aw#HSQ=6fe9g;US;asj>=a{@yFRIJdN{QmH1Vbln&-6+)
z@$_)KWvtf^0yjYv8${lv?;}23`Np%&RXFYjDUozeapa*iv23Fv5tJKx8lVakY??oz
zp~fAL-O#5Po7lXIXI2@HWE;ko(gUA7x|%)xRE*uDZ942zPFiP8p3^*E&?Cu{`8@%!
z#LyQ<wKyW#*@z5jo{)B)sENdH4fh3CdS-YddC^x5{8KdRb>}s<wGNzwr*g<9+GS_N
z#Ib1;hShbv2TXra>3r$iwb#|DgZlk@F-hx-9vh^t@TU0E9Fuj5i!lY78}!Fde_~Vg
z6^>XZ^`Uzh{?ZFt7&}$_dl!B?PHo(gaBK$qBlaY)<Z;W;GLX~F2iqQOYJCf5&gKay
zYQ-4E8VW}1Pba9kxf)9aJv?PH|My6Feyy2*ys>|WPqLFS9WG4~8s^4RSa@;h-{go1
zhnqoJd-z*a*xw^Qgl(i;5+o(K1LMU1?Ah}|H7wO&W&G~0=*B(?(Hq9Yvl;yW&>p}>
zDj#`mn|E)}VKD??7sN`7fF-$Wit^IVuOgFjt7PLy>6?OEN%#Ik1qTRziqW)5+ClQk
z_oxi4YXp8|B(&~SvX+O7oVIHFj8z}>oC#M|Ohv!C<R5Y;=h9fBL;kG#7rl<fgxU|1
zp4aOkfr^SBh*Tisc-Hy>lmTiusW(IPHOUBGW-<+LACoSG77!3xiYJ#%{HA1mmcQom
z6)$zpb)Y-y>aeC0d#}eI#N8dmL27;fBLm5J=~soX94I7Y+VN@Yi(|<)+)1lYMMy)8
z(j{uOGFTNg<K-WjA%~;-(6HGH6GWbuYDn)C&9fpNJd~Thx6EQr3O*@S@KJR0Ti#X<
z@nq|FwrsptVBk{$h;f<;ue<miM##6QTcRbWRrjFQ)OyQt#oHmvxXzzl)rOZ-Hpk<h
zYKnF?K9eOt+-Ev3zWA0BRRdKlN7if0KT8fFKsE7CgF@4M=#c@z@Eg>b`eukuLM5+>
zf5(K-{A}@wfV@J!V6q4FaQ~0ldd0Sl^!nvb@Xj@YUhPz+=V}_gUAT7jjZhNqV0E*g
zOxDOkB4=B%(W8|c&qg+Qnrr<3SUU4~DA)i0cZ#xR&z4~phQvfgF~bML*kvinR(6G)
zl4Wd%24fv#myofvN|q$bzH?$|F;NCnjhbpOO_rhG<$QnhUwJ%myRZAYulM`?dOcq`
zO6hGr4SPEj=FGJBg)RJsn0yj1WU;Km2MS(+AqGqCnM3si$FNiqs_n|E{0kf$6(Ijs
zti+duEJ_SKgY-v^NP+{jOk~XdWc)4XUX18Cq1_|oKGe1t8)-f>B_*k4@m<y<tHQo_
z7mXlIFhMXmpqBvfA8j1awoCqnSum8Lv0}w5r31h0;!&{`W;?kYDBIhoLxzJFcczT}
zkomJ`-%O61EMRw7h{(Ok@@+q10#(Is&o3WLKpFfiyL2A1d3_~ufBuaJL)Dga?*r^a
zRjUUCifYsJ{7j3XR!eAsigt!n{5*e>!?~q%7aAj<8VNz|UWiYf?Z0z(*1U~;gnoUs
zsvST2LDEK0*5%{qkF2F_)czQY#D6#vLAR^)fL78C#I?p|3tya64jQ?)sTNU2`b#nI
zgV2vpQIG)l^y{+7Rh5l^!yhxsgd?3Nk&9k-%0Zo$9{G%l%%e|us*96bkCPgZqY4WS
zmwrmoQGYEGIdW05^o*apSAEPM0zYP1#RpjT`zv~*sm}gYpOv_X7Za2h6MDM-&+bBu
z1BvOA_eUY$@eUNNaQ%?fgsJC-1#{kk&k=m7zE$gHcHCILUOS=E4jJ}wGht@c6X3F!
z4@v0{yeu(u_iw<6k4~Emv4d6{3eF?T3>CL4EKFoAguE5}3J5UdO{Q0dfhI0S!6#YD
z+U100{_9Rfj<dAGwote$A*uz4f$7`|yYj3xAX}f(Nb?&BDIqr}kfNq^8Lqyys3wlN
zJjCRg|L<B9f&%h*iJsb7zU$S&pMZE^(tE6uBMa;UJF4sVet92ER5s?{Ay154j;mJd
zbF!K&7m3ha78El{{c!x@|HcRxKgxVa|CIkR{je@(C!?qjw&L8K`2kYz)_J~SsF@o%
z-D^(Y6i1XB-Mt?#eGXZgAb;&c!oXZ6WY^7aylfR7O_n9N(1M%ZOAkU<cg<=?E&unQ
zWI{%ik04dMUPxt{h0;8lVlm0Cd=aR<(Qqaui(_KoWJO2w0duwDZBe`-ifIQAC+!KI
zjQ>kj!vo>ly<V~+-2dXhKk)}l<-*A^O}Y}y<P*St<av2_GHe^#Y|yt~K!?8U-Loib
zA$ze2bUPP%Vi6{HkMGnusC1N}aX@$^E5HVsfhTL(z$6IQq^l4K?2|PreWG!D-DH1P
zM}g`YZ=8ARl<a43{v9~)aKhQ0td)uGjUz|Eu+EGU+d`6Cs~+P?F-WRhja3<exgac)
zR7Bjez_t21(Bfn`8N;)%&8OEm)i1^VrlSM1r6@-{w4`gNSrwM1yLSadgU5rG>L{pk
zCTz2U$>a=)R>F(kSi@yHT42bO6CYf@E3O8^5*hL5IAYotQS>uTv^@RGllp+kbBcJ0
z#dP4{y8m$?YA2I5&~IY>D)Nwf_iQo-2Gd~!$Hl8=i}X4?yiEEvs_E%l!g|(jt6qYd
z4bqMtLIoTJOkpq>DjQ@Rc&EHwj__6PQ6AF|0tPh?-WYBN=zVn(OV9JsNs?wP%g3%9
zobQO?U#Mot&>5TbApx&Elo?DJk^-gxi(c+iWIECm2zmdrCRvZ}#H@pR0lxdL(dS-I
z@?*e8XL4kw%>8AW5%^rfg^Tx+?xWT4g;cj_gtxT+aPYJ3RF9^(^5F{7Ue!207)&L2
ze{LR({lCRI{c)r_{3hSKAdV<S5G^V!5ZZgn*1aefIHFW{J3MM&94n873f5on{Le-g
z1gs6UsNsZKjw2q3aI!!d3w{^ic#z<&AweXnJO26(cnsDmhh$n`4N*=^_?Ul2QZyQ6
z1!n6(y1I6TQ^BK06N^C0QNY0nfj61PF8BcaI04}7D`=?3$RDRE`>h(of_@>B=e?W1
zZNY$?7}$20DC$p&z=Cc&swO%u36ejB7X!gY%GVRw$e<UDD2oejRt0nKuC*&y?+{}4
z$H_z+NTN-ie1Gs(ig&bh>WY!vG{0rC)wHK>n=MzF>-RO1oIm2~590ewF*Y)dOcREz
zkCX)M{~CM=XM%jyp4h+=IAMSG=<=@i7Sp>1zR<g-yf-s=nTyj7;34qgfeFE7tljQC
z*uAVK-UDET=3)fjK(OaKX{=!+#S)bG1Sq;PKtzstAGn9&<=KAOrPE>79Y<Jfz378U
zcre9g;Sb~%tD1w4A{TeG7|xQfiS?#>cA3YgPUpS@^E6wA)enQ)S8?T^52uY@DUw_}
zQ#zdA65ZnR0m7`f*z53|IUWf{su6hUS5F8V^S9NA7el3M{qQQuKtM*RYHDhn{%K{!
z$`gQr(t!EsRbgRaDQiU8pX(af*#Z?@^#{xm6&rKozdLxXPCW*xONSiW%&6g7ney$5
zQbRpW#a}>k;$dP&;vr9W)FQrt?M+|eJp<0&z5>t-v@{h25`A#^5C`{i)f>EvX*NKv
zM&{t?kExy&Upf$Z*cu7#|Gc-tY-vzfpF0=0#C35oVU7WjgaM9*Cc=wGX!*k&Xb9R#
zJq)dnJuu!E(04y2w2Wm(@0<{Sd?vFPIS>nrGd0N4BgPXOni^we8)J`S3N*hL&V-J=
z%^@pBINuyq3u4Io{;P%%A4`r<%e_5wyXC&5&7me9)W{XsT`aMtaT<;XW;mJ)$F=vj
zSDf5Z?ZB>LUCvhhfD_^enq$RG?_NhW&)PBxZvh+!7=?r4GZc;FX2Xr8TGo`Z{z>&p
z9M`oPyOfMf+D=-M74Wf5(=N2VVmo@mu2c%mxYE=3BHN9js`<^f(~9krTiuyk_;e-2
z!Y&Ku5O-TTsL9Vc8jm4tw1n7axy89p-reUfwcb=<wyvk<si>YhM$vwVtKVq)X8wRD
zwhdob0D;(kxVs<oX@Vfe$Fv<AUorF?IFu<s!WIWM;o=4G+`J~JJ-HRwlaDFn`ZiIj
z`uIIy>!@s*!osEvTq7nuu0o?Y$xBV2&o#0$<LBO1V0$>a)1RY34LCIFY93$83cf~5
zM)7GYH+Q4QzuNxV77E==qZ!Ov581qd+#^k4_@i-;g7N5g%*VZ6;i|DJ$<;4(xxVow
z_kY$%L|kCR+_mh2-O`6Kk_Fmz*1_Wdx>W)#)d@n(S6yHyjPR|(So@J0>)n7qj0gg8
z!3tU5AM{epxi==`6M)#!Sd%NcI+ItsX9@a~AqJmd!apq_4C7U<-08)JnJF2fhZ&mT
zay^J};)O`#B{XS4XWxtGI}Av#TMKECtg~S;h>58FvSh4w*muU~-njg4r~&Dvs71z0
z;e-!z#0lZ<f|ehmx16J`F^11DB*?IjQsc#t)Sp#Wr}act6ZfjTxTg7%tfrxQxgPrA
z{g6fa!($)M$uV9mgt8fem&zuYEXjf|Hq|rMdi|Kmn)}&wCr3@MzPD<k>4pkF{7h8E
ziPdMTsl67CLDWnGkGvq4FYrDcyKM1dGf|mI6z5GRsX+LD`4m{(-)yk#N4@{sDuHPc
zC^3o|LXPGE)?<t<jrig|v|RbnKX`{D7Zg5?9Q;gMK*Q8V;`k$7{kfz1%)SB#jV%Rk
zv@7S_EJq?OLy4I2g#&M#w+w$(aKxK;0@VU$_i@D|t|3Dqoz`1{&-%6G!!WC&9eM;b
zU%$#5$(vD5=EP`;r3*%W(fdq10oq9^f0%k8*N!GHu;WszK28SMPC<C}jH}NUWN-Ux
z(7akM)wM&8@tdP1yv;&J@^p_Mv;tSKd;1(oR_sJX;-~Ktr}c1MjYPBX!I81;)T?Q7
zANl%cWx0v+pn+SoY34KIk~6Gke;`|G@T7%(oN=mcR=2HWsk8{okJxT{F4^(>mb`Y~
zp=X5tCuen{PW>&wT-s{|*m0#dgt_0&`<#8&=yvwVf;*jkE6a9h-{+aHE$R+G7t@=6
zyK!U=XWg~Vzp^2TY<x0!=-&@<3W~SH%RLk*W%pqECz-sgf3w1vk8MNE6MgOKk8g^9
z!RG$H(dHfcBoR@1aullrX)UrDf5}Yg+DAi+CacVi(-tmg6fA`-UgbNgpmXF(IZ~=G
zMp;xnQsA2gkriqg-XV~|&=JZ%6`lgSLYCi<p#EN60#7KmZj|`i_7XTLzk2@POVDNa
zv-?Jicm=2kyRStInPm=6`iZJP%rb5W0$4Rm-OZotXzmZ@AHuALIzRgq@Px4fI}FRV
zE9RRHLvdzjYwyiB@rp))#|x8P+r0`|G)S!8sc{9WFLJCy@ReBey^)aK2Kb*n>J?X+
zwhTK=2?0up+uvNvjJB6X_UueV#9Ol`H}uA1sbE%<d+TCyUHla(H@~Qw_&R%Ie=msz
zGp2WSt?>odo{0ZC!$6pb<vIi<8zOD~twjWJ{3FJgof&MEN3WjTqW3uRiJp1aNQosq
z@#l}wWnlzII(RRuTka0dH|8)ZmLyW$J_+edNNF(x%BK923R5GV!A_F>yfELrO?8P+
zzq5e9A=`6AXZA=4^uEu}DiII${^u)h`Do*+>Jz$60}=kq8f~1Tg~1tf@WUMz*X+};
z9~BbBuup5~>c=@f3S5SWKhI}eLnr3W*Ro@+I*Gx{`&WH${Wt~ra%tabLL&P^nciuL
z{`o@lCXZS&%*Z|97v|RR_kho*XB~cPL(*gZPVT4mhsfS5tozp`z$)`O5BKYVal79~
z7)tAwd%5<57zM|q2f3+(4}#llVRc_t!<|h(rr!S+&WS(w-O8qXpp=2vG+ydj0W;}3
zaf0OJQxFXQ#b|e<GgS^SL|ys~A0;I+TgN!6i7##_3v;9S1*x0J=x)FQYQiTrOEKdG
z8CyY=c8v%ey<s!s=Q^eowS)8Bu|4hG6h~2BkGQ+3z%H4kq@{161%638Iy5yIZ|f;~
zR53aU)3i#9=q<v(T_xqcvHs7@mt{wCJeO<2&1X4+kIDQO$knt;-|y8dymIKl>vHqm
zp`+I)Yw~UbXK=MYYN1w<YVBxKgSEUDIKxY4<$NSuize+w<d8uyu?AnR5iyi%N5<>k
z7HeaLw=x~zuSTxz%+T?f{Sk?xzTU|g1Dj9zv#QsyN5s?RYkM35G@AUQb*I)U*IODo
zNKb5vVt0lq@yB46MHW$exu^9Li=H10W93&iF(?rVq>Hz+W;C(|tTY-k<A{&i%|2!#
zSBtp_#mj@Ug7T%;-U!F@=e#+P!*}=2W4q55Bly#0?P;BjWx`ilj3>=vbS)#p+$srO
zjSr+sBn#iZ$e>G;B4$2gS_(Nnag304=#_o*=jn6YKYLn<GVu<{71s!1rm@kGncJbq
zgLQL%i%%C~5oM#qsJ=i*u+7k@56+5Jx~)8#IFow!M94eE_!3vC^A_4a5t!aN?~92(
zZ=LrkT=-~k^4E^?ywWMoPU>$1nf+ARPBS#^<-juvMIU-B;;@)p8?Zq?ed_%Kqi`lx
zr|}gUFaEi}21#F9?cyji_7BhZ0GGL1zFW;S$G3@#%Qrn`SJB&{%muT#B=^E<L!nC$
z%jJA3M$ArXM1Yu1-!uk?K4^FF0;PPOxe`b%JLsITh5753n6P=#4+q`Tq1<R?+F>X&
z{glP&l+}w5>u`cQ8J~YrziaeywF#XA<HWn*c-TRRO`si=_~*}_2iR8IoXl7)A<eZ%
zY9+hp(ub7C85y23g{8;Wb{A~1wVtyoUPmeK6sme$-Qf01Mgqx!RZc})(S-jM@B<sY
zuK2Bsp|Zmx3__$zk82Wanja2-mlpvuH*;K`)s$$t`ci=PQ-bwU9z)<Nh8qm%TVRdM
z=19KkXg=np5WSKdk?o~<-t%yqEhT{{!An3(Zxikm`_$pRlzXqb)ze5htm@UE6lL2N
zMc*A2Y<@0iYE$|AvOX7H&iY-qcc5Ke#N>sTNY0PNqUdPBj|bEjgRb}!h6@?VMeFbr
zH&Wq)%cQ*?Q+k71XU*$r0u8D@Z#1x@?PkMI&H8k&y+3);aoTHnRG5yau*mn@I1^K}
z0a>bP!gRZ*zZ=^+rNQoEX^g#Bxo;T!Pla;NS2fk<Ye%fBPmv+oriw9k>EE_{ghTRs
zDy!K}9%DC-2h;U$cBg9?6im##O?TXoIT_`Y{gPYZKKTqcbKt;K6&#b;HdfcK)+>ol
z+bO^GB72;1%jLO>=TNTIG^)`D_;sR-SsV(uLxGJ&EQh>PkQxq=eS{ITTbX83Q2h3e
zpK+zYk10t1<<He>E~a-5O=?~%u7?b%V+KFWv(n~Sp4f<V4>&Ux(!XlXsF3%}V6q{6
zG32m7s%ILlEAV>4@K0m^MY1vvtz4y)K{sXvrf*?jzrCwb>xIJ(T3yYwEx^wNy4oC2
zcI|eFck4C5x|3V$ySmu_vw{oOiO8wXESu+A&t0#HM(BUzL<q6Hh!y-%J}jD2;}Ny2
zlautvQs|Hsoo87=q$boT+W8r0@-7fd(^x2Iu&I<acm<Ly;__*HHa4ft5-iFKuW`_~
zAfSTB19V>D+Z5lBh@hKTl~UQ?B;MHSF#7{!?a8W-G~>&;{MmpzHI<5T6_+Vz-tjYA
z(%&H~sBYI+tmSm8KB;4lJ__8cs2`;WyS=4Dvz&u2&K77%MLXVTD;Dd?-x9d+#`K06
z%0H{+(ebCTclnl#-621dWv}d15sQ|z?EOj}kBlAgtDP~8&r5!@)R~%bYKeVz<ka(<
z9sCC^f&(;fUI-f!e}-rzRHWpXJ!nu4RHIayg;Y<>6dCsP50FF7ImK+t$2d`ih0Y#F
z#8Q&jLklHwvKypD<Eqfso9o<*(O0xuPr7T3i%E<v=Ta9QXbpIs=t|OAczLWrThYuX
z%4$xbImCGS?bhCtBMi-v&km+uxhcJpP}ET%oUjmjA^9AgR#N;FbN*BF!S`+J=IiEc
zRMHRPp$&;K{A-Q_j)I1=91yEm6Ho#OjIu2p;HxlDkmfVJrFL?Ir!j6{y)AzQ#R$(d
zwCZi)0h`Md@NV*6US-sx#sF$6scQGg1_WfOfV~7gS?!m5QUQ`y1x%aikzGCsXUYJ<
z1)9{@NUd{5=1Djw@BgbY?r25`3F{hF<1lO4Q9{0A!-l+95Vc6FqN92i{eAQUFLr`F
z7Kyl0_qMGxUxVePz?AKBegt6kWde+tpz=R7Rv$AUPU>++Ou+e1jz4!|XO<RTHK?CI
zC0VJob68N!R?H&oL|w1=>o1mh@!6K4Ds8I8&xh}hp9y<$I4Gs<@oL11=`EL&-DUQ2
zdE~+;J7OA-`z6EEYcE=6hKh4lpNBDt8U1ZP5u%Fvow$QVzn^1I>AekbICI89ppIxu
zE=Fz2y|M4VJUfx%HySM~cwbt9l?PSSX?`F>wTq)ZL<?MQdy6P=3!y<r^*qk>PxS>K
zv`m`19Q2e=Hf`N?1;zfer$RwA|2>fZFubt0)x4F5R}Rq8M6N<yWAV|-qZLFZN0NQQ
z05Sb)TNE<^8Y5V>EI~$Qvr_SJC$6rz%2V2v<_X-hv{##0;0%a(^ayy)ZFzOizTCK%
z{GXl=+!O8m3t4fOzsp3DU@9a;j4mUv@vsZfaRTb1YNY1VO=?xPyaY{@Hd7icJHvCR
zy1N6n*D=T*4dDuv9tX(t+~;tsd~CFuJ1y(ao}XGzCW5x4uZszxT)dL;Znge+)1a9F
zdHf;m_g;G~1{DQ7EO-aoNLI1An#J%&7J%mwwjq4@6G3PKFyI8G#z~wq<0xW9tq{q8
z`NiSu8uX!xEp|#O5~J635<&+^d8fqRI<G?;5ITyBha`^1$%H@j8*Vy!;V2^IS~~Mt
z*U=SXfjyb*?04+$DSf-hG`-@yoR)iJgFI?T#!)<q^XuXF6wlOhNLn31Qc@BIgNn$=
zNIv&#gz^d2yB~c#`sqLG)b3Js;)U&I1?<)u?Nk5#N5P|p^PsLocAEk{mQgSF2z<m_
z#9}7;9m`9Z_63*aW;M19iMqRh#zsWKpve>!X0jl%`gO(RyT;Yy<Nw=gxu<rG?p(cJ
z$mzrRpL`<fHX{%XX_p_DnnlXZpH;=#uMK&TG^IWA7?&AEcNcvfk*a(R$s)n|YB#dj
zfDZK>O@dR|O*ag%8dK0XlOY4QCR&6_@({xqKI^I8co8$dX~W|s2v+N#hA8Q;>x8ls
z7wU&7)8(gRd%ku$tTx;P2lqdFz-=lyan&hfT}nU=L?LXU$Xjnt<PZ3x%uzrF_<x3*
zZ@_lr@DcpW0;~}KL}2EF3M6DjX3m=8{USA4{*9Gu5I$!@jN}_bR_wvg6JhtYlMG<H
zB3}~&m6YynK;%gv1yZ&&nIrNxQJa&*SN-3s4m?>+^S$d^k^YN&dR&SI^X3U#JJc0M
zI`}et9YY(Kg*{(mGa!%7A^l}9q{?>+g`6Dk(t<>0#x=zRmVQkTqb;f~BhnEkqQ!CD
z=kzdHr&X`Tv>-<oOTQ+*A796E&?)*523HH-kEIU*E2v}RONsk?@2!&(e+8~bpMor%
zmRr?2lqjXV4m^-~13SQ4oSt2z&vt<MY@iBxZpT+D(N~O)5!6y&P*8lHfPcCR?GYhv
ziOKz9-V5wnH7pBEVoFFx7!?)N8Jb;b;`&j<oHloCmJfORcLKQZW`Ng7uAz|n<6(%2
zVXM7BX7xU1LyhYdj_Z5PR<BuQT2f0pdU#c07Cr);Qn3~9h3};aWL7<IQare&je{8i
z-->&qbDkk`N@=B%tm(Hc9DVO>J2d?78a3k2Gndj3rzexe&Uu?FrWv-lj$ywGJ)cCA
zn7^~50p_mSUSJ|+Q)9$#W~2yJL_;^Kvmr1K2;c5=Z|xkAU6Wt6E<-JH_-manuCbZH
zg+T^b&NR`#yMO=L<7!5Yqa}Fk@Nal8eqlEzKJNM~=Lxy*=g`ZUBdls{Kn3r!Oq7DM
z#ZB=@0*+qc$OXd~B$<J3r!cq4RA!d4r5D)|BIuEa%QKh+om94?Kcu%yEsi;H^Sg3l
zPY#WwW)Og`$O^TqAend*Vt{xuv4=If6*#NIxDP?##dOhwp3R_&Y@M%}D5lvfH{Xeb
zZ5qB~#%2?LfZFBoe26ty(ylAX`Rbxu$+C0zRIar`wgAGaCtljvOkxBs`X)wo6snu}
zR%#l3zV1~{&6Z|3nML-<VAy*P_eG&1hp1bUGwTJsBj`ZF1SKlvneP+tYlojzEy+58
zeo!yHR=f7sLkyXwRG}K-mGx50Nnd?cW(+D&Y#@2SxhEm@#lh<MGRHeVd~W_25&xg(
z-pBoHdg3=5`lyZ$4>F&gc*nTE(}DMH9l5Y`GZuQ%q9;Ml*CQ_czbf3DKLRwk;M9bt
z7*arBGa|5&&l3PFffo~?<}zuW1?>qt(hpkzHUl_Q9*zLF1Shb@VINH*V^>=37m2+T
ze&SmCHIh*}`<xk#@V(2#-rL7+02TC15P~%%#{&R^24kowDq-<z<j!!JV1tih;@(8M
z!M~@rt6CRCo|!Wre=ci0+yPj_%|pBj=-XtgkUWiwai01YKhOMTq{T(2O1^DUkNh=~
zgy;t5=6;Xwfy{?b4U_<y_woGjas;wR=2N^fHe_x#j--#o56nbQ{6#l)<+CsVdn3j0
z7`%DhXleL{UnfbWez@*59kw9}9Xs}k*@^J)^{(#8Nj2b45NICPxHCJ4`2JE%0wS>}
z2;7+#gM*VZllY8LX8i}T?U=*7(>&p|hKWJaS{0ad)K}J>Zq@I|M%aOE+9SV*6EDx4
zz{UeWF&SV$8(Bc@i3LV+QZ@4yW5O?<%(W9ISfFKOWP)M^FhbIr3y@R?UpQtOlRt)O
z0n?8L^;MWh^+?Ua$l`$Bwa$U#$y|5R%qqe)>EqQM^@=aC3H{}3T*r*c*Dixz8Epqs
zwR%L$;ZrS9bFNSW$M1<gXDWuHWto{ZzV#Zm;Flm5IO2C0xh4{Pq1kQH^b4vn=BsZD
zxXSR5pQ=4{33{}n)Xl?C;{(+4aL^8|9yjW#E;tcs1+j@c{z|h6nT#fKgknoluIoj+
ztC3v=IGU7J;#_%_|8S1kq3;zssGHq{9OWuRtBtkeOH26D>p<+Jci@wEmo`M#j)Xhl
zK<8WFajN__()jCmm^3BIf?lpCXP;$eae=U7Rbb|?-V$=b@$j+@+rg%G!~0k5`s%}$
z{=1EhU3>9rYG6+X98jtN_F@m1G2p7=q&epP{{Po7cg+Kyt!jR=E8Wyl!&_(~Ft>qZ
zh~dHJ0C(EN(fUSG_p=Lui9gUjb0C$(*ZC>!<9w*pa@u5<za1crnW{GJWJTG847qjv
z)wMU$^6{6aqn}1%TPG4(>_#C)s0goAz@g-t2_KjwAK5TyakOFFZH$*40qUZ2K-vJ(
zcXpYI4M|NI>Jxqd2X`kYnC|1K1WCoKmlL=A0SDX?ursrD>NWWCaK2ugcX>>Fe!Jl<
zP4Z!;RJ(q`*>1K^%8|erO6PmTis*sK6CM&$wg>M&|F-Q(l2Ir;?z(R_#6MG-`?rXx
zf4^puPWipK*rH9Sx&DELr~}9I>WY>3XJxBaJN*6mQPssMiqzTkC%2voI~k-xt9_1^
z<Mx@K7!eHWXT}Ivy5rH1EK_a10$baws^aYaezbI$AhivNd(!pxcgg2Z*$;5$Y)3Rp
zZ4-49N%ijbR(%apH<+F2-$87F3dxv)CQoBhxEgDkYsg^wf14nct8sWavU`oHS;^=l
z;K1Maf6OjQbRyNmnAvq437Evu&`rTeoKMjGpI^~ke}9}|chS7~P;%BZS*onJNCIf{
znK5HQQMcRHHC91$HMkg|1+gSC3RR6^tb-x4ZaEl$nb(MXo}Q^VfV53qpWz~stvtlB
z>R>G-7qo*>h<cET!qqxgb2>?83(lNJnHu6DYrR=oWV&X(kP(69>iZ!#aVig&tlDzq
zm{YyAE7dJer|!fc5qt;fxR+wvo@2i)pB<gPBy4x&wbY`$Zgp_=cM+?mA@_D%tlVc=
zZTS_lh}nr6_T<B}Hz*!W8gE9KEm<#ezH*-*%6K<~{cF)=RDaC2!Kn8)DK&L+qvxN(
zAu46xC%w|{=Fr^F@98D|0cvS2L&_9*liZe4t!|3!<@SLmDHa)|AuolIwl`5>F7FMC
zsX8TI2e0+XXB+9Fvra{aLEpQJ)~C7GrDo(^z2<N!*8FsZc#o`5hT$8QQ+|M&uAY^7
zW2@;@*~y(}ExKNVCpz+Jm^>9roUNyn7Atca%e1f2d<wZfai@x*`UBn;-~T&tMC{J}
z?iGO*o;Wb!F21*~oAxH2c~OyKzDi@zl$t-(E&`rZG#N;WFZfL)qN=~y<^t^`i2(@5
z#GW%$Akh4_Rl(pIF#wi2eKn9!##|Go7FXqUZHSq~3hmlOnVqSbg&Y$!{Pjfgy6s&N
zj!%`>K`u#r<s^}nx$;57b$tKu`qMZyuijtC^!J>?8Lr>8=#g+Ny_W`CH)?73%c=)2
znj@3Rf*1$qJMH2zLM-lXN=C%RTJ?|FSwRG~jQcM)|Li$`A44U0g_lfKkGWirxnLPx
zLcj7+i?DcJCqza^%d_N-LlQB{-k4SJX3XEkJq|Kt*!JR8Yi7@w@ogP1;J)ckoGh&S
ziO`s}8$d*x*}e%2w>jFkd3YeLWB8ZoE#G4==l$L^mBeVNyIy6S$60C^2_ksMtB@!T
z#^qX=;5_B;N47$xS8ldGeY(_5-s*lFR;yv`qflF-b*FpUL)v{tYmQ;&ak0y%!_rLq
zb-#4w@yL^E{bZ6|vl!RV=)BiZldJc?pyw1KU+i1lqg=grUGT1eXOO3H4_fT+*C<M~
zNE`0XY9YiVKK)Y}yjA-<Up<kT9D((e)3~t(7;hjl#KK#op~@K}LC#HU6d<--tnMbN
zkchDmFz1Q|!>+?%D%E!|`Q!LbPV3gO(5*l!H%6rbZqvmK24`x+W4wuG<DC=_B+Kqs
zeENF4496RnPiix#rOwWrx+Q<Z1P~+w3kCE?Du!}n`zyp0&;p!eF8|R19Du#Y<t%l#
zI(}OyBi(89hz&{%qj?|*o94$xS^;l|t2xtCD6R;=*mAMH|5?1xITe4gcFNAq=9SGi
z{K`7ePd?sqQ4eK|Q)kySobdZ_rKWX{hGxM)PyJ0K?N!%#bYJ>wmDN|rbXvcAXP^z|
zIa&iLxVGe2piAeRUketVk(<_0lcY?~QDzH^XSlJ7u$h#QmaeL%b<CYnpqrlyjeRir
zr222p$r9h=F9P9y^bycg{wZJcRyz5up_6<kH!KmkFUU&OGD^_AUDa;n*6NR%`zF?o
z11vKwB%!OSjOQ>_IWwjh?Dz>wPI;b{((@wVco=xMH2&FBA0oUpv)l;8u*7pTB|b#E
zw-^{oB<!OW_Q|$*nO7O)KiJN7*f&ht(+QHD0~?iceXHz&&kQsyHLP)dkzsSeG(O3b
z0P}#U@Ev!s{T1jI^J(!51^s>t!KnTuIPA4xL(P9uxmx0p7<lJt3tnDhf@!gkG1Ymj
z-RxdW-qEW?x}dv^7FkJ{*ozj(AqzSGT(EZ9`EP`H0#vxtk28B@+Ej9!^R`f}vAVG#
z6##?;+<4{y=LrWbEFsTpexy7@6(Uf($^bZ$hRw?Gj{g7kJ3OHR2$P;=>nxdrMqXWS
zA>|@}HPsk0ibFCI{q3)Ea}vzp52ta~Uyf|&<hzsy(cVg%Equ~}zPg}s<ov13rIGi>
z%nVpxAndW5dGM#VCH>XecMqE6zM@21R90=(Py4-mS~H}3;{^fXa?7?rwRI`pv@z>V
z-5cM{GNH)P%QD{`c&y_N4jPV0(}zwPG}8{*jS7^z40aZrfHYq_{6%wnNCBbL$g?rc
z@37f4{vIQ7qLJj8wP{`XbKhF}i6z*fEt2!Y4c^u5BXQ55i9v5~B%8|VfF>PK^8vU|
zFBJ&X3rkUX#ZHC*$rY3{9NZWdU4|JrU5<L_N`Y5#tBWKSvg>RdjSw+po5N{I&Hu{1
z&tk`0{@R3m<0MigHhdEOqkSSdwq#G(3y2+%IAx3d{W%bS6`&)taLJoG^n2i@Zvt95
zHvhg^RF65B07k$kDH)iFYlg7b_eU5Qf+W7F9V2fbZa-X}PWNq+MMPTMRL2N2JBLbK
zm-r$5EZoPm4wdw3ejuT&^$)Q9c9VZy$Z1CbJaMOZV0Z$q#+&~*6*QcUhEFoC3{L?m
zOZPB%lO_I&tou-9q|LAQF+^>jhP)MB&h(m^S^?y3!PwT(ComhqLTqc{+$hhL)qnBZ
zUW)7AhU~z#wEWwUFM%;ziih8wEi%7Za3$*2;dvcfZ9c8S8ms0+Ut71vD&bKE!umwz
ztu&qR(0MZkCSyx+_?2Sdq1W8-?1pFUhnY8EWdjaX|IYZCT?r`WxPLxr@zwI(OocgS
zF7(}_%!{Kz_J+qE%q069j||(Ea|rtKz{Y0gwDE9_wo`_6iF=N1S%>()51%IJ?^Hb(
zi?5E+us&9D>xg!GO{7tOOpC(biMS_Uuh()STakV-2dRJd#A-ZWor>)ue}azi*-PLh
z@X}QtP{WH2g!D!4^cmN{$*ohVKUBj8?T7?hk!VyY0@m!%UD&rJZ7cAI62zIziKq<@
z7zOyU<JE{&2Q{HxFj%S47z4Nxq>o+Pl+tii%44J<wB_j;3oUQBSl=gWDB9-{R?bWw
zL;$QLEon!$a$P$~lK%I6#%lH4UC28sJtSv<J&^d>Dc<UX1Q(nQD?=(Q&3F4U1wh+*
zNNVX)+;NON=r1VyM}nsGE_M9~_+qINgtXB1kz@QT;uOqgO7Rd)_Xy+efuw5|(%J`F
zb;z2-_db^ua#T8HM28V3V|pQ}DF^pSX*cp@$J9*D$;D=$0SQW%62t5yo|&P72ieh8
zf>s@7@;{2%?oJOc-Ave5l6CwlmCslG<B?Z;q%S*OOUQDLY)ppD1w=iS)zu5>J^S2O
z?Ho}jzR1}By|sbXU|v7^`+KuqIeRNEQt01pF(=p4UiNpa1){$j&lGOhnzc(aM+!ro
zxK`y$-XlWmfkfl?X*<L}dj{vyD%LbSbt}g4f`6NO!2+1)S<^2FkU?jr^Uf0w(-*Rj
zFqX=f7|;=q*4US)4{hLXNw#Ij-Yb5-CLdqQ#YwR2B8k6hrz4VU^(^!Se-v7$w0ctz
z{_5`yQ!up24uN<n|EqjuM0dZ(d6=%`M7r~b=5fzkQg@2Ylwh06WQ-ULot9fsrqx{7
zdSU@KClN|t^?gp#U&VgekYZzy|Kvz%CgCA@*Md4f&ozDuS!qJ<H!x>sXtElX%GDb4
zEgg#UC1wg64WuYA0LL(2p%jFc9$($PfxEcZU9lDsyy>O^1~yBO%G~8nRsadWhEC)1
z$+@d|<;~8%l^Uy=6ZwdnGt3czyq90obqT4`%A{g`NjG|ICbpvYo>um>XQru#l%m@y
z$w3<N{L#@jB8WV?X-tiV)uk}KgAZ3#t!k6Hi`<Vo`4<ioV2G$sa&os~bZ&EzM<)<L
zdH<f48h$;E(z>mL$e#Qh!aZhM@CNZDJ>8>6_S2I)aWdG@Rh_vq$rwa+>UlT6`5QZ}
z&rH#n0?75@GcXhsmEWyh-A|HOgj0Za(Ok2t(e^hy++p9OQSa^LNAW}c@OhRMy=Vkn
zvT76z<{=|%8B0`c8xA?Ycr+wWnVdbVP)`ocQ)65ESx8_iOnyrNGGA}&e51XDeabV_
z4S^rM_CGgtn3clI$E%E~QU}LaOHur+vFMDeMmc?lTl+gF1fd2ZT6rjfD?PMyO6rp%
zARd8^ZY5w+7Mv?AeviR5O(o_Pbe&%L|1eLf3r}kbz|{CvvHi(TAy2?D#R3Su$v&}e
zpk&tdo748{de=tHT-@<E%-nGZ)u-ML;w+ubKHur_;SBBC5!;!pd{5#7pI__6n>h={
z!5!8ESZJ^%E0_kY&VPk#{x12lWRqm}0rLI~AZS$QSiA&jxBjhvYt`gdc9@2lvO@ZN
zHr`O=Z@vx>W*ad2@@K3fKq?QZo79iR^m-^&i{`RXR5UbiepNXSuJCCm{Zksq&lHTd
zF_&3c2f{!9fQSitD!+)_+9t9Iv{3HVhdX)+{WP1&p+OvQ_gt?MAkW`Qfll};0Sm^C
z;L;Q{tR0)ap@TTJDEbHhHYM{kcbk%)h2VH^!iP0e9*g<!1njN)fakD=@m@?!3f47V
zDP96oGo3K>?=0KL84I%amt}pXBjrE88dUA*@?&oylm6_n#0hU|hVn#SaRb|u4oW1~
zSk#c9v9h<di+gww<R&r|m5R1yW2}j?`pJfhq^xCp>-gEH(vjOmGs$aW=M}Ez8(E@F
zze2;G0^ww>(bFwsf6JU<>Q_83AtneMX7U+1Qm8twgZQq%*H@b0pDKO#1^t5f;kI!p
z+lSWiwp>ADK2}B;dZuC))ESvI@i6=}YYWEOO3YqPh-e28qS{@|<vWBQ7`FnN5}2t1
z3}1y+9q4$+9rTq3s%m+80YIi|o7+yT)k(}v+xcf;r7K04hug_{rYOR=+>64-i`^+c
zq&kii_;NOy|20WcGU92icD;C7qv{FFU7{zF=wRtsJ>%=o#=+}vJ~nRBm+o|zJk*<3
z?3KpHO!qi6Xe#HC?iBRi%zpCQGiF0I!XMJ_*$lC%N_`laHfOBT5CGt`JLBsKqa0+|
zLnhc^2p#Zti@kAoIX4n5+)sV=z32J?7G=3&KZxmHon-;-1ELK_X=>U4G0`}`3KGZF
zww{&MhIFwXip-p=_;_1J{486#!8HOrb$Y-QU$0RPhn|E2)B)&g{6{XfsQe26l`jAp
z@JmaTWG$l>&<qK>^7$`gFt+B|P<Ij%CI9R(o8MNFZ0Yltl9G}J(<|Y66r<fK*P`^B
z!+pMJ#oaL`>Sq7%0FN$OZ%}Ibqhrrdf`S!4r0`;Q8o++6-RmSi)Q4azUb|P^0kPOp
zk<{CPUH2DqrGc4P3=~}sx}N4!+i{nSrNM6|i+e|Z^`ue=Ji5EBrCgASL;!+Xooaqh
ziMOC>Q2%xv3`?`JKx#^66HGv*{D6oo=<4)>vy0KegQ8hZ1?R%_HWMFC{MjSj#Sa*=
zrY&2*{ZO8&mC5vLda-EVAIT9;PXTi4^<E&4UDsrm3q{~3+?eQ_KXHW+oyGDbl(tRL
z@G9a3Swe0~fIt=TdDJ^sCWpLT;x1MthC*1$egaD<Ss;BK`BJ<xCjp;<*X<;kMpw8*
zS5meh<J?4mJ#$~}5^~}N<lcIHc4mwG8GSdM+<Ujfx?TOzgx{*^74Qv80BsgayhsGJ
zyjoOG7T#$&aeJ>ZyLgkr3))F)tJsZu5u~3FK?HsRlIp;Em-ln+nDcWtR5l$K3?wnQ
zJ1@zxuvT`b!SPpx5l+R7K#bJhQ~HNIOs!(oX+yn1WaHl2b;d|uGr~_T7YvtcpaNJs
zqo;sm(NsL*s0@M!zy}Dig{f$X+?&8Io}+CkDStzpYZi7WAHW~c_gVo+yxGvWc{h4S
zZ|0e`=hfc0%STj_#mpujy=q>tS*tbk6+6Qd$>N-8Pyz8zi92PaUoIBIv{^7N-PesO
zkPChp_v;Q9*T_O?1p|&&A#lQ2n$!#u?$CTx(JvK=?H{hMRc1&n8OcxSO6;JT`EmKy
z_`vFINET=THX|FALn<s@kR#1$vVm6tP9+{O_k4~~0mLPjs1Wj-{i4FLIt;NporE7b
zJ)wM&6U{U_+R{o}B%(_MIY_?mDZ>H1W_b)*8Ne6ct5{KGu~-~P>~o?B?2cQ{|HM_D
zfcR7*pa0PHm!uOxZwhnBb^SQchWE1TI2fiTW&o3c_rE)CD-4eh|7Crs-MPg|j+i%(
zz>C-qKgPBF_FJ9lVoeAq;)9kUdB^EjOc+>u$9z?d1h=WeD&w35y4E=ae4og}Pl>7;
z5nT#S%+1RVems7e+Rr;e-wU}Mr)*g`4v}9y*?>&1*!!WdLp`d(bR+Qj4`XQ(0&O-v
zhKG)!jhp*Pdtn|Zf*(|Ht;dc-z+0R!-bI#}@w4M~Hf8H|<MS<ns2)gtT`X#){Ya)M
z+!$THng|m#fYbbQiGs=y65DnY1~`EsZ-z1F<PQFW7*BjdBjSt~V2f%}p23Z9+xO2>
zzoGBl>CnM6#pSkQ*`~qPM3p>CR?Y}G`K$#aVjZKK=2OkPmt4C0iOFV~Se2a{!PZ{C
zI;}g9%_CKZrD_3EzEMPiTTE;Hv#0Ih{aK#CT0&~qy4YQgF-PpccJZ;^guxEzFL?07
zcPy4sh{IL0Zonach3in0m%%8ubjpa@0hJJu(Hfgc7RV7|og0jky3ar-C%7q*BeNoo
zfB~0*6vkZxsEU1uTgtOyn_}@2Q_vnI7+M?zPMmqKVeL0!UVL{rX9QX$YawCniFzev
zX{Mk9S_578jQ{L8{LA~>7|(pexHkA=cMB{yAAtvZfG6apV2KV`9G-9`6qN(>V_GoT
zln4khhbloI2AcSuOU3({7JZS>6D~M`*;<s(spC3u^9Un^SBeZ58G)BTyV9IKxPBy!
zPsR{_qeWYH4s5H}k8wcLvTxk=+VcZtWZG9Rsb2IYcv4F55oYawV^pvucfea>$czh8
z)7!Gs8qG3XNP2w!17t+Bcv%m_lC|vs|Dq!TvZHBAHtWh32HCCVVImnnv{v2SeD}Ni
zwaJ>$RhuB092`aJ3#ivJFhqDG!cK^A`5gH8W!=iX<*<vbsh!FS0iSHiK(MA^-aEhm
zxxuIARm{=8UnTVgcotmRJyF@PI0&@k<^Zpv51irXSg@yO$$`Vdi)DOhqk`^p#oNYy
zH&iD$6USC}(o&kok<UFe77FQ|q$(4yE5L8}vb?T><p3Y%xVq^q$$eY$_J6w!md=DO
zdf&?m#|^>n02PN`dZj)Dk=a5Ivs;>;onmhngME2}1T~tcYE^O~P77b1sq5`fOze&q
zl-Mu_SsV4T{l6tcK|BHj(+`0eU?ae;iv{#xz-hAiVr}wyYEAs)g{zd?5M))wdn4y@
zZB_ynge7&!tb6xad!BO((dDImx?hAcHR|fMIlF7@kWX;iku_#?^><$7$?ixVd5L0v
zocd}C4hQdy74zz47p1F1xF8}j*$M-?(^)B*chQ;5p#r^yj4;_qbng$JpMlw5o4!H*
z$|2`uE;;2o6`MoDsPu%PDQbi_C>?zi|K21Ws~cQ#`UNvqUm*$HZCkax${Swe72=V-
z+oB%cRnt7jNsW;Crmf%`!nHE3LBZ?7CGL@$#mHt%8p&o*+l9y?)3KI4H<e^BctTGO
z7ToLF^E#OJS^3guKh6+%8oJid{`7R3RKMHP@!}5QaVsT31CZz-?vM>$B)j#*f`G$-
zx7pRIbsfUjXfbHojhaqe+g1zuf<y%NVg&NXg|5*quvbg`*;9ReLwSAJ76V;%O3y+O
zL3Y6ia8Q0PjC651TRr(WLw$s04|;whtREl|3bHo>MA49H;z7hK_%Q|dY9fKLzfqhK
z-?s|soyX|j?wsJm*(PL!YqhOm9*?Z?$Ut+10Q)9epA!*pjw#PfnUI$r&uI7dKLziy
zwww5hEK*wP27k2^(GVmP_^avBlADrP*v1hR(mx>S%5S86R~Hpe@u=d8+`YT0eC`mH
z=H(Omq}<VyD)Yf&E~k$l@MA5Mx^6yNFKbhRt)!9RFPy7BgW2BrxEsaFx&F!s8~C%1
zcSnnCf{YpXPg9ec-QGS$0F1LK1?{~8Pw!nSQeO!=^E0sZ><o40b%H7<TB!f1PU2wd
zh4WXd{i*5H^kwsM_0)lK*MWlC_WEF8-+l#1Kj6BAmnKF?SYtLd6kc+$gEW$dFohy6
zg5WDb50G`5Z}r9CzLGJu7+#f;z~?pGP?Ukn<J&IosUVJnTu~|Pv8?DL-^J5efs+kY
zt!?H^BClTNv_H4MVaM!C;)v2T_tK~`IEB5CT;CDAWXycl{;E}@pG}-2%&*IQ=uC@k
zR3h+IsCRKQAFGCMOSWMF{v`pt(uqhm!>O?F+iogV=F0BH8XQ6YWcB$z+?3Rl;g2b%
z2|WwHI=>++TEU9anPZzDiQdkih0{$OL}J1MOJXe9Vzn~Qv2t3vV}dauB$yXJ9l?fE
zT={H(JH{j1f*8a~dkXBb!wjKksr}U8mX8sNcC=3h>CC%Ly6Vu>7M?_b8>lE)u@Ymi
z(U$Q7!O&g$Vb=JcJx6^Gw=iK(F7*!Uaz4Jc7Y+G0+3C4rH7j8MxM@B)#ipL?t4ET4
zU@9tI2|pT^K~!{D?DwOJv^-wIPpDU4*>C0dSX4A)ZhRHhlw}v0Vh}2x(o`3setk#x
zzx;*bZm_cH&|40skAht9e={09ZT{Q2?PMut5S>nK`a6*(&QTIbUyL3KU-gO01%XD<
z5;++}kZD0>##_ys<C-@r&T?^;j5J)D)sjJWxWT^z>$u6P2Hh?BRY{OShlY9orAfFn
z!iSZRk&##xt1=IIG|`9fTLE}h&;#^0H}^hTdi?!y8aKWmS<LzMuXUl(aWb&jweN^+
zQKxz4iU*r}3(opTY*?<0{)i=9Q-EN%F!HWfWSyfAKf^-&&Ln*{T%#nHg6$gPXh^v>
zd3qLo6ZzV~%-;0MaQBAl50q0PqjLdz-0aqPC+LbaEby5E+ODLD0t%!Qd*ax#?7=}=
zo~}{<i;Tv6<&h9bng8Ljf@9~EhPSEB>Pg?thgVDGvOW%-`p_NrwEG?F-_KSK0?^yl
z!GVts6c7%_9PZ#)aLyfA8Mx>vQnWv77OE8*g&Un~nP=UX3>nPeYAdzI+?EqvEr!Q_
z`71?R+;1-J$=y9A+C!^8JZ2^TPAspU*T!?8vq8+&2^V@?#s2%}#+$2s>a@Bv+lY<?
zJWO9hz^d%Yi<!cgc-7aY3(7X2xx4{y?nj#q$B=x{BL9re3WuP8_~Uky(f2|?E-kIA
zm}lC|VtT)!{h|duK&?RntBHuc!Pav$t984d+@J_-HKp*~x(;N>{V#t!MiB8!v;H7q
z>`iI@bGpo(oS_cW2#kcQWe&;i|I8j_r%z&y%G|ULPIR5O?=Wg-=yf?&C^8>d%G9WN
zv+Gj5+p+*^(Wn%k{lY}!&z|HhszaVpukumLpH>*DDWjA~;+B*Vt^-jHT#BHKutD>+
z!E1ob4`>I|_N^pRIz+d0@Pft~;va-0w9SPwGU*dopUkOTFpDez8=KxZXyJ+BV~vIZ
zBtV9vx2Z^P>0_V=KblwO`pTaQOAmRmA5!GSZ-N+UduMjP#1co=N#c81+EiZ5aQFHw
zuU>99t=b-~Jdc&GY10yEc_H*AJNr=elIgrwUP-Jw*-0v}aOIIv-=lcJ4>E>|qRj<m
zw$|qs{Hb9j|Gw(iInw7QfAU2QHHsBhrT%GbJA1woWM7N_3hs*(A=VxW<(`~z*2^W7
z1w2%G^6|oYYrfuy<M%JU;g-jSL~I`_(SlN@LgYS*4zxaGJheC~*#40c1~LcQ{W%~m
z=LWA_0cdwDxyS(h)tIjQVJ@&R0HOmM4$iRfnt1pxHYkjWKM7J|pt~k&?6I>n#)bY|
zeA7^g(?l{TqPz`IaQI+G1SGk(7bAgB35ebB&IYXM%AhTyiB|Jj$<^Vf2MNH&1m{-?
znk!ac73)~XR8|iVc*WE>9c_LurFi+jk*1aWU?^3{S5BM43Dk`gPH4n`>%Vv}85Fq<
zh2~h1{{d-^F!M3#FM0qT?SlNLLrlF{DKl)QkiizjQ;knRB@86wbXJ3W*RSfw?>Hgi
z(hYok2aC2MpkE*KiGLpyy;&r6_`~-hI|Pm4hj*DvNPTdAoE*}0R76OpBS5#=^4$2W
zT)OAm^M`&xzKyC(f!*(VOfUMFBm1b|p%3HUnddyJrH}WUKAhjre5Zc(G0ykL`%S5l
zGBJ3^3u<-n(+j6b1qP>0-;kf2%;bb6%I*NY^4{I}hvYdQYKS&V!L@qtRI~on;d}=4
z)p+!9t#VhUVzgGwi9_Pk=N>4u{@PibU9HUxeP@k8+H?bL2(cYEr5m`~d*wu>{JJ)o
zGu+Q~P2MtJ&AbpJ5A<1SRXG`L7YcS8kAQNNT+pQs(B0^gPYp9$vMjc--v-X6ldQ~E
z=ECpj#u-fjJB8T*#fTA^GXyU(beB9u#rYw5d+V(u5HTl6m_xro>mqyB=cmM?*Uibi
z{f{jUa#15@+QKpvOV}}|<pjB7)*wl0zrgFe4pp9lohi1<mMVEx>*f}gBY>+?m@Y)+
zy8wcAe5Z7Eb?pA^NzA`Q`skx-XQxJlH&LkClfMmHn^N0P2fw#@F2IwQtNEyow8Gm*
z<q@rG@+GB2L0rY^4aL;fQz7rIE|ooLx%aD>#8oVTbB6FY)G+xk8g3^2kN3mEawd;<
zaH`$lSVutLq0vmfC)}Ld36lc_M|YAt_kU#07qUXq{)!iiT1ou&+jz0+UZfST*xr$(
zdjE4?tPeB_e+jJkfOkdfFUdzw7F*gIv9kN;O45s<|Av2QzPS0rLIBM32A+NTn7vsQ
z)YZ3(3WA;hg$lohbt+X!g1Sr@Q>Qco!uQbnN$Gp{E<eXcl629-YZ6~L-Xc3Bp6SYt
zUzqYsS)YS%wv`$<G0#}6KyGZ8&(DOBkU&@FZ5c3Hbxb&+3DC4G*O^YYxTd0S0MHCt
zeXhepA(?d-y>(aYPJZM1%phZ#hKiC=cL}K@5)m#=S^t|Pn4p13HG?<unxmBU*z8O!
z#Q9Ezls+q3q+KIj5Njb!$v}&=OEcE|NUF!wUkbRxHebg|0Q~9yO&0vx{Rj$j?-j9^
zkI}Yx*%!~x+5Cd`yiBVKv^g~wCr~Ima@l2){;Ix$ub<cA%&C&HY~DIdSg)?X)T~(V
z=AU%SVEUtSyO3^OPA+aW4CgEL97}$R^SkUh#7=G`-QYK+pDS{&y8YbEyt~?g_Rm6=
z`}{tc@iVU6>ik(qulqZV?Zj&3^-MPAJg7Bg<Y(qIBu)EX1x*<|kEz^*mKv9YYq1B7
zUX?7lTf11yZ631Id^!O)udA1?^z}j7n1pO*UEa|FkI3Rq(8uieRRVTb2Pu@DJ<4`^
z=Q92aAQTt|37Ugjvdu^0bk_s$0eFlc9XCF81*A&cK!OC6HHn%0F)bErVka<Z3-td<
z#&le$=Bo<d*{KPhn9U}NPl9{Fxe*wqC!TfV<ACN$B1VVNlU#ocNViB7BO*v@i$>_;
zs>};;ShgK#&v+|%O4XqYBlOx~EZ_Uy0ugDGWG^C*B3sgV{$@XD!K^dDlwef7bHewa
zn^-FmhD?UVDN@~ME!<VBK#jz;Aeb_h(|vnmvm=!+Df_|H>gN+P3hYUxmtI0+#2+h%
zU`;C?_{CP>&B#_3Fg>4~+#1c-1_#K8&GQ(6e1T=x;mt@*$datiEy>72sSUgB!VBjw
zeH+;)KNIjB>N^*Js~?0On>9OyAy)^y6OWyH^j*aD#+e7HKZd%$><>T6ZS=D!eC61&
z=BfPY3!`gNdHw6zQQr}V6a6k$F7?Hr`M^28ZaL>^ymdG@HK=nivrk9nihD`?{0z<Z
zQQgF*;Q#*YDczP`O3%DHmZ9~&PR(iAI>S@}<VZlk*&uzvsRLpaOZ<Q-&t0o-G`27#
z7Q<78+BNbK65LpU?_@^WeuE>+aeP|ytGJ*mpkCrfJ8w(RJiHG7zEy!_b(sw)t{G#a
z&_kS<q3llDmxSREinBjPB2KJIk<0?g!h7wB)LoLo?unUFuWUwxe3|DsFYIJ&2H@e#
z66=~x9hepq&|9wEt+GI=6;W%hU8~3w2d<GGK!F7b6TC|TCKO$uiC(N!*sZ|LjFaQe
z4;p7Ec7mKm)25)b66KlCYWE|2$IH%t?9($-)R8}#)JC(g@Pb}+pD`M;zcmWIpXZbA
zL%d3EwW0e4r*OjyhD@>j$@a2j3mKPw)FlD`m&9uC$vT*2DzacW8G0$Z&C6_JU`KVQ
z@<7)}Vy1pi5+d}hCBpsj$TfIB=ET|jOJa%JsnCy#dLn7jJ1Nl1iY3Cr)#B9pP5F@!
z#ht{)+s3vjo8GeP;eo^eO=d%texwf@=@<p^>G0<spL2+^f`}qKTZ+MWJJNM#BFciy
z@#ZJUM8jKffin#h&eYidW9iJpp<e(0-zkZZJzI^8osfO!U}o%OiDWDLzJ)B|WZxP4
zQe<q&mL*cQv9B>mCS+*Ts7aINjD61U_WAy<%U>=omznq6_iK4RAJ63CQm@63Z+YQe
z!W9^Zh>9gSwU4zJ7L%oAN1S_97bV)v0w(Uuv1&79vomZ0{!l1SO@l52&2`XOD>D_T
zHS>sc7Qqb{4+D)oNCZY(-XKd7nkHJ+&N|vS&Xe0<e0lHynHeYY>q}zE<o$v`UYDF0
z?pGD^2~WQBP-ZlpIFjrMp{O^?^HZU2U~zk`x6Wzf!l#!corsdBrqZ)}vg&;qgVR@g
z2K&m{&5;BO@8qk>?OH%Z_&bh}gHPjai*sE`oxygc`!oim*+_=z-g!OEU0ULyY)@MQ
z{<v`aJrU=;z~EnpSMm6m)IM@`m3tdb5|u6~{wx6Kyq%fBi9P3XDGBoWdH?vE?D6&L
zEeuG!xa>^Vf05|f9w~4&H*p@!CeEHM)b}&v7ck-@Y~c}-9cwv4fV6n_5{XV_Q5ot?
zb}G?yrV<jGJdQaxdd=s}UR2QhzJ<?#-xEh*?~VfbEG|9Bxj--Yt2$3=ox=c(l*7b!
zzy*$)Z~z%>c?~ja_P{|Qr0X?GvT=)-^UCM|8v9}}7$CLfY>s7Z#AUq_5uZTkCNFwu
zl3msnCG@Ox*P87wG8MX0pssDkZyLmTO&m%eA}_12kNeNADMp3U84<Z^%IS<%ud|An
zh#=BlGvnG`{&(*8wa20dbEKd^s_xa(WdEX6snIjOz;bOMGTbfv{nrQ#Bs1KA)GUUJ
z<@IY%pPB)4XaI2b2B9l^2!znYl;+&S&)#`-f+!8vvo!WfuvyqT`w^Bzf%p~-UcY3Y
zCumM@(ZE!T+*qbJfZ_kmIaU-o&@ppbcob97Krx%VzEN$1Aw+>mFW{kv)w&b)vBdXZ
z2F((k8Ik!W+!Dli1<>9ha8)2jJz6B-!Q_SK3G<-rLqyA&dWXKxO}qhImq0A5eF2h#
z%QT?=caG0v-*mgp`np$gB&Uv8t(V)$Lk9?g@1b#9h=QD=$f5EzgoKT)u;L-VHTJbc
z#La%2;YT!`1`62Ew11jZFtLD1@{kJpcIcs0#9|I`8SovuHlpzwTjqYoD+En1twft0
zFP*ww84(>`0fj`7+%i0<3{X>~XrR|OcMF9t$S#~kPn?1!0sFR*Gg|9*Wf@CEDoC0~
zs5y(*0CdGDT_*5Bgo0`UyjcAl)v-qDSrefBFfas^_`(9U^@1bFO!nsj73dDj*sHQv
zD%OCC3swk%z@qZMqsz!qY-MF-W3-1jklAdLESB%y$fgy*QqI`uw}JeAXk~w2mR}@y
zY;*ZiU3nJAC4N}d#Q1^$({QMUuBW77sbe|R;O{pjNGm&d%#&Z#Ye4Gi42ccg(;gZa
z@v0;+#LP6h4hj);9OV$Y?B*yr5Dpm+j!9ss0J2&~AS^G8lu35aNAvJ#<WFk>IAeD4
z(AfkyH~t>@pnHQy^`7%n0pLCW!AWz@MRIcTORd#65jYbI1#lj4+#s=r9YaAV0VYhW
zB}X^Z%OR=o%oj9)w6LI?gDWoxUUU>_r%kYo=FQbnlY|#oz&U`-lZQ4r`9Avg#OM8k
zPcLG4N7tfH<m&43lZ;xfgt+0$`)==>@eVM?s{4FX&oFw0Z12j_F<o9o3FY0qDPS+x
zL-4^ry|z|vs>Pu5I3?XNVSZseNR!Wo{BSh?RV3IMB>i?K7@yQ16d$ybzUJI?pHF!O
z$QJC5sRBrRKP1N{s(4WxO%S8Nh>`Nxe0-%464ZcUHHSm&2C(~M;3OP70dG!W!F!;K
zth@n(e>$_{u!g&WPcN`G0U}EwaNM+Ydxk({x78%&R>}N_mkOLuXIHg7=gAP(FL`SS
zSFRy(xL?!UoQMIq$j&}Rw3U~h|AF?fey&pE1S!I?!lazb@ZB6CQDn=A<2r{94T|;|
zD*9TlT*YL;W=XhvtO5exKNpA3#$JjSBxF{vP=L6LCjZ5XpE6%o@;%6d-t<Ko?L5iY
zn+6SO_(gFH90=k#M9^<7*kI->qn(ndf9cVaLM)U!vahjhtTl}VDvO3zGSX<F8p~L%
zxcU0SU#Cp}oy(>=@*nAOR7bLfcV<?2MAPl;*vR}<K}dciwvVA<Wd$%|ko2o&`Qv|2
ztLbEuo7=&czSs;LaK(s5LsJ@PcEjVKiIfTGLaC&5$t_T;fuXIrZo#CCPzdi=Po>J0
z>O|`FsJ4E#S|!;E`>|t0t2zAnQU9Ed>D;SqEX~>OZ|E`M<T9a8L5B}d0g+Hz0Qe=i
z{qb7G$bP*P(##oE5Wo<Uy=+Ik-rN=n5l?&WNb5i!QLtV;TY+c!pOa)JOAde>h2m+e
zS@hAT8(;gvF{#e9!PL6{&W#M@#ojsicv$gIH3$(9a(WO3xd0mATW=+ZjqU1!lY}{h
zCm`y8L@I=?^h74u-2+n1|AqUA+qyd<QD855Ch#u=?*Im3AP75C^!zNj>33)0MnA*j
z_V$4+aT7wtg`fJ52zM*)KEv+IJc!cdpPBwA3mwBiRZ~w_)s~E@n8Rvm=egVfmX41?
zV8sjPB(d);S>M6(I_Xpw7<(Q8=*}8MlXtC@hu9Ca%LKEx5ZAUBu(gWK*2A=*QmK@`
z>pLJ~w%DQG;`QQbwu-;P*l>o;(yz*LLFW#Gxtg4X0<+v2o=3HJJUIoSM_FGG6N!xo
zkue2@t9jnH?n}(4OsqxS9!pL}oS(Ys$-8@?T;;iiHF)bl0J$9SXX2*sFFTK|RDdb>
z&JX+mX5pCbtwuMy<Io?s@VQ{XR)Runhn7GNo1I-{7Ep}GtKUe#6!!BI$n}Zj3b!~c
zNFF!CZo2)bBAk)jjHXE-;CeFu|F-;<86a0h+=L%=k|1K{F-i?$%2lj~Z7{s<v!NTq
ztV<S~%sUh}5D!&!sA~1h_2N<FUm?ILCe!h2InPA{G&;4QT<5La+;TegaSA<GXdZbM
zRs8ikmN4h9J|YRxKbD*f49YcHhsX*hg8em1rJ7{{m~DWP1-KLdPzl%~a=p9~mqotq
zQ}4vva<lR~xK-P$W7F(A?(yd7mq=v!dm8)S%Q<gRJjz*^AN=gvvW|Pm4vz`aMqwt4
zj_LaI{5EC8;`w@>JK8VW^Oc`6txY0g$X^=*x^N%g)q&rWT6I(EjiHUWM;`0vyQm}|
zTzi}1|Cchs`p_Z+Z*faFJtvwQ%f_~9@+?C#v@ne=huLLSgkqc*@FqfNJerZ!k4)RL
z$MXO(_vraCvS#%_Aq|<@@*+Xf)lPn_b!=>AqiSa9Y0XQG+}_AmTE#?HA;E?dj=D+F
z%(RAHhR3NA-Wh_NtoakaYqAcMt*zLmVGwts)QiMP-lx<{2V7He84nxv1k`^ANQvFd
zlC9OAT45TMPTE6XuW*Z`6RZa-p;N9``|J&yAQPQnx91WaJEy&;_usittF48Ibi!N6
z#3Y!nfuO7(^etMq$A^9x16+%Rk+rp>I8Bd!sTZ#*m9fi{Y#%sp<)P`WV&4Ar%D>vb
z;Ctf_W8W_aVt=0I*NW;aXwZa$tRz-;lqf6;TH6juDgu}fxrt-VoF1H~G1I`s!<5qN
zGkY%jLV$}eO4nxrhc#oR4d!E59V}A2{E#{=qqzez?VpJhA%n9}If7v*oKy>YJP0}f
z%R+Ejouui-&<P%J3vPz-F|{oUMWx2i4Q#9G+Np5Kp?7V<4OUlHe6DQVT9|vI<e;P>
zi_LO$oojL*B)4!s{tPqI^`^*g8<j+`eDDr;ApERd;_0f|>DZ=in*X5J_4a8VDKU+0
zjOM3xenzOKuF|IiRYH^R`K!%k@z*9B+PpZ0ot^*9;a82*v*{n0AC27D<^D2L4i##l
z|8CNH1@Vd^F65_~SavP5S$ZBdrLq7?w)h4~s)BR>Wcq+xSx<jN2x#prMS?*_TR!Z|
znk3=g1N8#TrYIZ6!z!AZT(8q*9;^(v#hc$Lrz7@oTj&v2>7fT2d#z-RD`l822PNKT
zL&=RQ*gyvB+MFkg%B#A}h{9(DqV}i5gL$77L~MHyI`-E#k8kz+{)0Qn^LPApm_d>$
z6EfQ(K6t1VulLi!&isKuhfz#&>=l8rU{AE0p_B6Ab^i*FnTbY+#O81#`*#Kudnkio
z@Q<+kRba{U(<*&L*IGSq{#Q73G+cAMlsYG~2Ue{n5$0G3%^T-xE%sB0I5pjGyvPBp
zasn(qOXlwib|tEa>RktwCspOF$9n-Dq`-^fv0M%n6I1U%lerf?pr3FRJ4HGz4MGyG
zSB_*mQ^G-zI2E|!Px;B!;9eho=J)=I*#+s_&*G|pSo{28cyx}uiy?%@>>|mzgrioR
z2B3gtzw$<Ylo9_40(2{cZdKuzoS*50C`Gob3*7@V;zXGFy)WQ60)0-yMLk^*d7eEX
zG?#4UA&=XulAiFU7*0!w=Xr_k3*{-m%+VG$h1{SKpyhM*d;iYH{1@*@G^434(hX;c
zQ88hGJ84~d@l{rZX_(8%-m5V*kgZ)Q^>GXLxb&WI(MsK6>}E<)6x(nVAI-WQmgzZc
zXtkciors<H_B%~AdZt9H-5oah5XB7~unJ9P6Gbfa{4Q5h-+thUwx!&3e!wuBg8c8?
z4k2E{fTex=!7=B9?*krR<CJvsd(01bad<iVQPih2g;V}zjQ^{~ekC?TU$;`hXc^B9
zG>3>N3GQW1(JZQ@oX~4B-I^Uf=3B*t-UHfs(ukKF#t5b0SuZt8QR2FDtjs^RL!7^i
z?gkF)s*u$qPT}sjpQ;c;l>OtmlMj2w4Mf`~>m!Axe^glcE&_k%{nyAx@9sY)CGTQc
ztYzc<;VH!(no6ZI7m?q8joo)?X+F|ZyB(~<LM&}Ppbv&dYC{bdlHc+~E0{^Q`<-N6
zm4vT^yo;09I<LOT(31V%xvfZL0_@MFn595QGAJ-Wa3RqCfzs*TeRTFzzY3_tYsD+a
z!;a)J6**N3wD7gh`9m6R1Wlm`U9DM}YAV@kothTip*heNwm7aKgOTP(g(CB`rij(6
zIMIAszV-yC?+Skq&E=*VSGMVQiYQmHz{^y8Bt9Zz(Z4ZSd6KqaJ|Bs601uQpNU!F~
zb%gy}WkJKL3TI?vBU^&|m7|sfaJS%Au~v#^$6oB`34jOCe>vul(s)rn|HrkzJz54&
zVwK7#$8fyQO-+OyE)>JLYrUG@7aQn$162g~s+^8v+6J_43mi*vG~>pX=#1q&dF`rG
zy}K`8Ftg*~Zr-kw_kLH!_|Q&I;M%pnDL0g6g(O1d(o_DVhS<DkGC<2UScmq~_bkdS
z@rhjE2(vp8C|&mPgXohwU-^~s*(({Xd~g@kD0j<;d~km(`O_K8q}h^cX->sq`AmeR
zFVV6QWLrgAg^SM*bUfM<-XEy0oJZsyKiD~`C)+_PN^V8XF%w{Tf8{?f*N9`IYWLUs
z`sqJ4=1^|<Y4RsfF_DjuTv5?V$D%T37q|rDGkq+W4Q7`)=D5p~@;+tqNLH?_#i^e%
zr%*}vK};GUpgp!^XAA8YtHO^Qi9??7<MmFm{$A)|RoY*lIPqB0iu=jWF4>6^`WmL>
zJYN%JxJ-_tu~GMX$6f068`wb5ixS)xYf1CSa+U6gtLQvFZRj>ZULF<6K}1{PpUC3C
zsQL!kvjLc#-D4n$tjPW9iFhb978Ko}V<DaS#*I<MoAfoZzt8%8I#I{uMkER!vw`lQ
z?-Llf-!e^^IzedxU#CMGapG&516|@%;UjH*Y{k^E$0>DA$1BP!UV0*>ifn!<LHC_x
zdj_3S;86wQp<Ttd=fkdQoI4D!OtJWhv(g;-xwx9WizjU>M?yU*mG}gbu@yEv7$IVh
z1AGJennNJ^tCa%?#40*??V^xdOzxgr$T%izF99(uQuT&(W$Z-7-{$5q)iiO4ni!}}
zdv3?1!=LYc?%D!Q#81JnRDPOc#Ckk%q|5W6S6y{eDT98K#j|9>)Dc9zMp|XljFG#K
zuRko{#>XWKsVj1uoffwq*CbKeO`W3tJGV|>>UQ<@!Oe~1!6X}!MrzKRQ)HTl3~^2V
z?dOqLwVYQ2)el*R1^f)_lE|YdXLb)oU~iUe|Mt6@Wl%WZ@{_VV%{t2|nl5}5RF<tM
z0oUOF&Y^g`dxheg`3_`0rrFWroiqJKCgc3+t?<s-aaE}3xE0UeahM3@PB-k-WL{a?
zS9D28SjOPNzo-M?ugZ6wzS?2nF(q2qHT<`lVQf-?dUwYirX9uRvh}R$k94b82^X({
zo%R@sAF1A|+d8dsT*TY{FF!(Eu_9TO`B%x*&U&C(I=`!#iU(esNQG)NAqom0pPGu2
zk^sVaHc1f8=f%BvLB@nycG`HtVhY3WU@a^tn4Rhmv2!pB_wWdw$1`)xgO`ap?UAxr
zTcL1$FJu)>MP`3GpWmDt2wusb)ViH0){=(HF|x%v14~I^=|W*)g)SkIwKULx4Y>f7
z5Y&tB%@~ZT-ciTxv2?qBg;U@Jz&NHRH$TfdkRI)Ra5pmr<@`|N6>^p>0!-Sf`5E>q
zQ-oYhD*vLd5h;<FM~c2Ux|#$w^m}tI1`bTp$-T@b^mHp(195p43v#*SwZ4*18<v=X
zzX=Z3+QNPMKN<IZ#@|%?#V>$vh={mZqN8e+Wic00Bh)=j*W~etW{vfP{>h@eMKas_
z$CpoPIdYnjD_XC=bZU6FCyPpJe^NV6Z7-P~hDm0lYrG}mTwI1t?iEWFc)9j197x@p
zeo*DM%gi3=>UMGQL14;hf^Ec87SF$HABdH++26HourUZ$<VnH8rBr^U3fr+;qxw=a
zW4fJ*UXvQaSRd3MA5o$!(|jetz4X6x%SIxqh=Ob1IsD~5(g-bH@V9)uZ3gQDDZ1xJ
z$n_WImqm|Dg-*C|-Ew7X&>LA#UMj%~ET_%zez!Tnx!y2`@cuc<MPr2P^mEeQ^>f@H
z-4wN-40Y<h8n2QQG2RYo-1_g_kX}^lvt`25tT>mdHT<}~VB4E?JDUqOLkvaB5H`+>
zLK!SQ7#pwNjK}oMupgVJnR4>i(QR^BWn=zQFR3upNe&;i=Nc+g%Hync1zsFDj0^o3
zR1jzF?w;l08An94JsL1*;t>GO5ij<eW3xb$$Bdc&3noXb<(|;cc13u^L{P>Dh+yLp
z7!=B#6k^~-w(c>dKXc;^xn;o+HBl>bXaArzuK;AodV-Q;63kbSg_Ul5aI8AhKDQyR
zbV<WqgXf_FL+OC~!((<JZxB)e^9bJM_JL<`n6#lNkWK<^IDqBW=v-&VeD?(D<zeD>
zaCa>-2q*-CQ_$3<nx`BO2qz*+b^R3?VLa>A0h&pu3aklkQ>B^F`^CLtHw8SKJ5hRY
zaA8e=N8zUoJY{8^ZrD__)M7yuL7j<5@dPCwbjPm?+RI+iw^vIgRB4!5%vk;98(s2S
zXvn3(@vTz7ZzMljb~EoC<(3wcN5J@P>y_<*!EwG9%2%|$??^2HyJ<D=cw)tPAb<2_
z!yy%6nGbkGz}t7sZplPzyO<y$No&siRnLj`u{+StuiPKbGu7M_i{d17?IL-&5XKK7
zU!F)h#9X<ZC8Z0M7&vBW;J<ivySI^hCB)DD-hp7q`U_G&Ted{)a8+x7uOr{VS}a4>
zm{rW>Nmu=C2CH)=b4z7YK?YexvqNj%yn$C`9XZ!j5rJ2~5#FpDGgP(pnS8}UHj=2}
z+9ah!ozCZ_pV1i!`5Qom=h~F!E<N=`GKnW()6LAJ(LUrxj0)zFJHCEprdvc8+n>)T
z((sO!W>PG_=q5gVLdcHrfH#uE!f}ni-E(6by-`>XQXuEq0~B<C>&F~wxEIX=gXvx8
zM;!`wDg<dr$^X<$B+pq*NKTtZ$%?A;iuddzSN_Pq9|^3fyx_^-teg~+_CJ}UCMsI+
zPJgXt=~s8HsZFOwhf;7=0MG@Ud7-#5De=HONxF@!k8x&8bSxppYg7y!a+0TxG*gc?
zu$qJjI>HbR%(-dF@EgE(8ku{H<YOIWr?4yei*Kc%Y$x^tO+e%>vn+AqmmZjba5!*!
zk%3KGE)vjXK`aQEd6mio1l9qj)A9%MUapF2?e99%`h6-wFq&8S4X?tmWA2LDdy#-v
z53vNFcY0$@$@gdKFzdY~KN9SkP0T!|O}zJ30;(e)8amS`mLAYnJ*LguTd`bB^{Q@Y
zY_yryC~%!(vXM}t&-p^X`*-h^TVr3h>K-C)JaNCGNmgE2ld{LvW_Cb-N=b)_h&bd5
zGl_=SqiPImJ`vxrRb*{*1iZ%ghLN&O?UB{q{CeR_June@zuckPu4<3VuSr2rS9e94
zPtm8DJcdODb1dVKDS25D#6VJ}vr-^c-AP9`$umGs_*NIrU%pdT@o9Ft7FTr9#NxVB
zZoNA@LD%VO=(f%AWv``O#3VPak1y*(=-*$J^EQT3UVfyQg1qiA;^>OPF6Nsa<rH#W
z{VK)fWSR(kw&%4kRd8&_zMmEKFd?_fyAiOsfrP;Y<SP=C0(YTs1E8(uXyE+=E8<bE
zG@U~V|4m@X1T+mxRC6#UJ2NzWYA3H_V<=aLhI1&2VNt)0-}7@1ut&JL`<RQU-*TNJ
zR37k~<e=WdBQ%qrrhPii75j1^u~Q&ZJ?}J)h?ZX!Ez-}EPQxXhC`zsIhwzdjAesH6
zk<>Ze-r+We=r+w9dA9=*Hq&H2Xgj#nEP;<W=uG_rKr*4g5l2-6N;Tq(eg<4D4MCJ(
zlOCyYUZ3{r2DeU)f`YD>pC3ccz*L?HQJ^ggv5W}pcAXJ(32OiFOhmsHL21rXc!3WP
z)bnQY8wVn=k{{+p7QVnCArK^B(O2wPkom#V#8?#%Oe>dY@*Bvv)(2>{lHi#|&dw8G
z`Cg8;I8EU`%*8O%g)~1KT5^A#9S`+zPV43LZBVpzomh-KVG7Ig-?Q^qc#!oV+`VNi
zSRW?v=x*$IV{k33>LXAFCA=VHCP=S&joCaR3p;M|5eGaOqF}Wrf7Pwm$LTRd$OXc8
z#tKZHejg<`jd_%m>O>{~aS+HW!{(dZ{~^Ntt9St_@>JJ)ihbpt$9h`GN+v2Ki}R_~
z=lqNLBJPIW^TzQ(VtIjW?vBm}4ktzVCwI>F19#d}gMIVs_IRC*CqbT*n_k)H-5`yZ
zJ;8ngWwzG|53L?4`q>BKL|wUX)gSm$bQ2zD1c0voA&|yJ4jGT0mH^<L5q1D!636!L
z8T6fieG2he?lb-3--6rJ<YM9TVN`=&uICj=K!!=u#%#u+x*7jFSF6seG1J3N#X_=^
zPT0puYV@5}iB3ZgIj21fp$HJb$qDKk0)R1)X2^58p^^)dNQQ7SkT}3eQ{$)JgwuSK
zu0xqTsAGUv31Ki{HuX*;;V$xLl{|on`}JefAauoWxTH+pp%ykr$jJbB9-xnbF~Z|j
znW-$PAMx{>tC!vr9MDyDTBc8(^<qYr-z~EW3+Bd7s^v+E2u4=0RE*L{HD4)ihrK`w
zWpaP`5ejDg+>H;{T!7e95$Ai)qDv`;Y7_J7qtD&)Cf4UmK?93|P20l+`*&8s<qWN5
zV~f11m*n<vg)al$AYXW+refmvp?iuwa?X>JKTjFU-KPXP5{{fdKlyH~z+^2cm!9yL
z&P)DgIj(i+19Eo4$!D<oWq@j*q3HMQrGSA?Zr$QJhD|kWX&qFg)vYa7ONZqALO*+y
zgwoN4g@2tze`j0~dhzD%y;jj_E}gN&s)A2PN_hF7%i0Mss+cPyE52QXik-{;Z@#`1
z?hTR6yI<kL1+CfgAGO<L%tn=j%p?e~g)d8Kn0<Wm9iQ-iI&R`M=bqfLa~>k{#@m@P
zaeQQ#h0S5L)x_xmM-bMvnVZT$+Sm)k8H9ESRQ<C4tY4{n{Znv8HP6d8EOLv10Y&y!
ztD2X69cDM;XeLsJr%u?GVwm%gpWk<};v|uiNPEL@tx3T`Hydvy!|u~M<2?!AnCiUo
zwh9r+HLGDVEREHZ{@>rc5e$PGD>0elIrss}oLTu$&hbzLHOKZyGVy(#44R<1oG{yN
zp(c2RmH%I^VGVW}U8PA#mT-t*an+Wn9<TF7MsntX{3pT36@L1DzvTc++c#fGv(sD%
zy!95{Sc>tqV3+W`S-doI2YRFlGSrF)NIKI8;D$OF@}>jnhvq`dWDC2y|Nj{&7|_Vy
zR4h|UPfV#Im?aV-6g#T61i<bUWRvXJ@$k?u!fxpiDzxMse8$F{Q~nCxKeT#ErU-E4
z2Y>L|FirD%N&gMWswVEV)kV9?V#(=IXnD+28o`$IK+NN>syP%B_ee)Vo*db;No|zp
zPH;dbR`_b=EZtS_(MxHzhyq1^sPl1%q&u_uQ7q#cS!$}y46RC9)VjRjeJc$mivCvf
zxWm$&_wj|!5C5IBNbeiV8uo|3<3GP$>E*VjFe2W!I^V^O94pl0h1~O>omq58_YvNI
ze>RwZ$2Nla;*Z<Lp5I4vLV<8boU1Doe#~<G=&$byZvbv)?6(xr!2MdcAncI3z2@J%
z{Tz8PFp;bl{&7hfp(^Z<-C*BHv_Hkny(*`@T`C?s+eha`%@2xd!x`>HwU!_Lrd~L4
z+UM#hyclUMqh;((B9tXncDuZF=wkXOFa8_AvC%SDwtY|1v?v<6R=4!gRTlk~N@MiK
zPvqs*MydK`fr^~pcWk{Prr5*9Z&phyXzzIDMLKs-i=y<KW3CU3OvkM#gv=c|%CHhd
z)8L$;Q<ALYbdqg+$w@`;9!xC$(pFT(0KM^nH_Wol*C-k?DhL7+D=*){+D5_g*`-?o
zgBgjZ{Lb+#WnlTPiR2QCE}rRkCrS{V&mX`E6DRGIsdqOj(m<|DA^=4=RbJLc(;t9h
z7^C%fg#rIR8Otel5Odv9r_{BAXbzhs(8W0t#^F0WJFH~S4$a&|fh_4Fj<?Q<Zf?B1
zyaIGON+<8`BM<b_3{=XdOf$DWb8zysb(M4{8OeFx8`DQ0a4Z@&b?{}j+>#Q@_m^VZ
ze7+T&ne(Dx;B}l!(^sOEox#kPhVRv!=S>upteE(i7N)+}AYc2KIUm8%zp6@nCGv#>
z`U)Bfn_R=RQ(T|q2VX?8Q|JjM^;h6cT36Vw_SBWDPmn8E?=P5k{UW8cb=mvP9+qw<
zZ|L1#E#UAl{we<u<I`<^soc(J07cDDi75qbH)css5|zQD4PolM{jWZM#Ky#jKIzZ2
zzx|aKInx!}{8HlUR%vuuGki|b6<)ee?Gu@GdW76F>?90_x`}2cR>_V~+@${exA!kH
z8X7ljsq&bUhFD{K(W(%SDSlIP`cl|2eG_Ag=|rHl&vejh(W3PnT>E5B5nHIcmhGkT
z+#P)5#E$GPQ&+nQ-xd9XO4q0=q=CWhl4pjChayk-L!->q;30xVqib`)UtD$s+F^vm
z+$!66I-yv#n|`3>^*dgqRUEiR2Gak74caI-|4O0}jHJ#^mfnhHeq}}^%h(bcb8xo)
z7X!G6{j?4>`V)J8mB50H<TCz3H(I3#-L4qr->){;OP(M(qXYN}^&U=u9u^N&9Bxgm
z?fo+pkUY$l7b*&i5CQ<8Igc>lFkIxtXQI+&%zK1#;g|$kMh-8-i$SH%#e^@ZA`sXk
z#_sDg6;HogV_7jEUKZ6M60J4YLCeVf<M$94`p~(OWOIxxGyRWG%IV)lpb=`m0XBRA
za}Vy6U?W7Te~Wy`Ux{-1nJlsUqJcMB6Qjqj$(S?Gx8~ysy2%}zeLh+pH;J$*FXOM`
z#z>}DuQ>+TQ9LLyoldV0dX=mqAEdX+TdWXH<JGJuf%|OPyF3qVU$59&|Dh9?%8k-A
zb<Q3USMoYd9_T;@tVJ_CcvO6ztnl6t@@4UTOpd{UR5AZ9*UXBU!Lf@$C^b_<_Ho*2
z7Up}(s!6HSEHO}2HyNw&O1P@JnmrZOezk~cWkjQBc7WhcOSvxaVsVO1^|EY(vSc1i
zlmVay@Jy(_D|@c~bdrK@ujBB4>a|4=?K3}5huq7*UH!E#5Deh*FK9I4Os?cTAbX0G
zcl>v*{&Tm1T{OdrctQ`m)|9rnl&|-$>K%Ll>|RYdl;U^d_gZOK!z7I$ajkGs8svz0
z>bm^EH1AI1`$0e4+CK1f0ldTtrUpn`uXy@BbCW7({i=&S@I<7oy96c_%>|4>xQy0J
zZopgD61ZL>AroI%NXeumR|<mMzZon}l#23g&+F2CD`Vb91Sk{J2MVB|`nOs@BUcwF
zDU3P|PG_d(Xh0nyGZ|AcQejF{Jt+3()|)eCYb3Cw0+&F#W8fCp8UR04shp=BSVI{%
z#v)WS<uPiPN!Fi%4c8>q!Atj2fdKzs22QHix_U6b(z?p7trJ%RnEmf%hlW@OXy{IK
ze7DmLb4n=VC0+3kO|*XF-<OqPM3EQFK>GH|N&iGEmBW)rH*|q4WwZ2Ijb22AxCJLl
zRcy$=<szfPvYqt@ftjymZQuQuj-<)he6=;_jp3UJq`b<70cZ(7)V$Q0xY44$qRAd=
zX;~`KwI{nIXnw%@KC|m(l#qmvK%R)zbz_J=S>H;!&8Fr`m9f%FNbF_B*!O{=6m5!J
zobM@LRhjSXX=W7a#;DLnnUIn7_=vZh&pYU|>1fHWII>3Fs;newRH)d`CFr4Coh>2o
z)>r#Rwd=z54ov}3*Za0s6}v`TS&QBcVZ0~)J2#qelCyWGXkQy~rP-{2eY~dkX1b|;
zO=Ynm56)w&HU)Fp*zNfX)TfZmhikfPc6OZILpSaHr}2^Wm=Q%+sGXl{DaZb1h~c4j
za#Gm$c!145!_U+%vv1N(Q2zn}tzhjUyyx0+y=3nUzyd9d0PMXqVnrnZkn(LYcdOK@
z7b1~gI439=I^Ec7r9jtdV{r)O5~9Qmz)91arI19DAW8qzX~leFL-s1+31QCa13w_=
ziPUm6^PcHk0f+$(p|y2^Pd*`gR*Br)gN(ObN-&~>9#{@M6E|*hJDd2WCTnqR=Bo{X
zM5X_6n?X3FXCeYex<1SKxgbYg^9$|UE8a#nO2c>d5BU<j-!5BN`J|f&uqhq06mkl2
z>V@Eu(R>L$1+jE*SNvE;kx&hWD+PW5VYS5HhF_ouq`yZWXMH!?6Q32nd;ClHweKv%
z-J%aOgq9{b@;+So7$Q1BF~%j~cq~e9&yH)2TJVKoRtz4t{Vec>iYU#vB-pLY{JHVu
z>K@*0iQ6@NY{J!aDu?m?g*Z(6@*(#m+>$b0r1Sa(n^Cx~GAKH!W3t|DrcOTYs%|gx
zA%beKz4SWbfUCwZhs(6V*UO=H{BBkb-_RxsBdwGXq1Q~kbio~Z6vHm`Gp>tiG)1P}
zR(hqPuL|^bPQByMm3jZ2OD#0jEX(xvHC-+k%zfEFt_Bgy`P3QVdv_@Um%E>a9*M8&
zul|L#R46oSo?#mEuNhj;b2V$Gj=uRf6Mn?%GPAdmO?ViRaCLd)x=0cttF!T0V#Csw
zb^UfESzw1~<B59Sf<&gV7B;Zm63GNC9w4yh1(;UXSc6~i7+}BZ;Qv>CtQc!tpfKzN
z+dfJA1U{`_ob_8b>(y#0R^b`=X!`An8E}T5X))yFyea8V3Ff9?=Le-gYJcXxWeSZE
zq!vW-pI%ry*07q-c!Epqr=!5;*|9*2?V0*p8%D_FKJ?tQbz>)yZE7V1iw0cb?=%F5
zG=3Pb9|=^6_W;Lr`&s7#m;+~b#hEmR{WmbHxSoysU|zK`JUjVyh5J>+$!!8iaHeal
ztxBJoqMEp3PWZr<K>SZAe5mi$ovo0Te}yGFw@Pk(aVxLl?e)BHJ!r*&>91TVJFFS&
z>tcMzq=1H{V_KBtdC(Z|rT9rLpgxHn`ANxn_a$lZ`c{mpQgmIg-;4aEMyjyQVl{qS
z_Qge4ubmR2PawnA3;y)HLke=$)@o~M*VtB8I$ZCY3gr!91`p+f(la~GyH9MZsj#!m
zjynB!j%hhNHd8PBkwZp^h4F;n9j=oMdcCW&`)tDg6zR~3H6JDA5n%!>O)4*8>4WmC
z04igwK2`683HeAlK*qh=i2im8-@0Gs$Zmm!(nnQHo#{*loxkDWGd;uIbfv&U?RZv=
z@*w^BPdFm|c$;sZ{DL+j8pHlMJ=7^*?m!+F5E>1Qik<ugC(Y?7C-&{IPsdzuq0U5m
z>W+RqC|-%s&X=}H?+H=&Orws@vpE27;Tc_h55QGY7h^a!iyHyX8HK0L!H-4JYJv=2
zmX#jr9(1Xt*MYkz(fZEkH&6jZfMEv?YYU_Fy9z&J$6EM1W)$|cWv_{x0@;-@gc%zJ
zGOZ)&oOJN+@d3AsD?5v6WloqZ>3<ggwmbnoN-#H#J=2{CdFh*`bq{OOFa7&0$G9hZ
zJoF1%jWbjq2*2GuV%fg|Ipn>wLeu=$NfUarQzf%KNvdTm?wmizSO%kF2Qe_pRRuAy
z=B<Rlm!~9-7nHD$<t99Mv~$hF%B#mrdM0Sed6u=vkjWGl@6LD)eRUZB{--0J0YX-M
zGNr^j=cc5d*qD5N;hlE81If@mXANJz>hd!*LWVFHyzNNntL>q)ZFk6d@bg2*FLdBz
zFGE3Z`DeL3`A_)?6AfJ*ms&3Wh)sOZQ6lNGpRL)$S!+B=n|`xpbh<|p7cB6T+xP+|
zJrJ({)x9?gqh{{?M~SbQm#;=us&7&FoQa~Kp<~@0dWTEWaZBo9Zd5wo2f}xsjz#V;
zSG~!5oty9ZYEon>z3WrJjg+Qu*XY+7vpNjcWU6+RV?=U4UL%{RDeZ00StN6(%g$S;
zbRWqbbiG((>6YQFesGP@W+CFAuF#QJ+k|`27*-k2Hu$V8hS}0B@-Cl#xfi^x;Hqup
z>%oThWt{K&9jX;|M8f=L)!n2?&0HhGG#c8X9bX<>EsYBQLy4FmQZJa@^#TTe*X%&-
z6-EN}QU<kCxq;vE5hA)912O+=D+U=<Oh|a24;Flt5;`<qsJ6b~1qtDh<i-O1IFv@w
zfhW}>8WcwjVBRsfSN$4w#IEBW9&$G_QN=AZ=CL?26>u&=#6;nuOzp8uO(YkgKMtjR
zbJr#E;rqGB`4RdBgDHYzGVEu&jCQ+pPE8cFy}ciscgi}Sc1z;V30KD{gySeq#r&TH
z5Zd>@C4qDeI~FYA<i42C*m?qrUt|uC8YHz}0c012fk&MRxtw|<;Vi$2;9n{z*)j8g
z0t(Yi(iE~@D%=S1F1J{o<zHqf#6cZei0V%^Y-RNE?l>Kh&97FxpO?;*<3V<ZeZS=c
zK(D-({&`n*4g4(LCz#*&>&2l~sPydWAAAnX_jx<U@=)mJd&-l6E3=#CRoH3B3wy~M
zh0@pVNtFdNMVrjIuG}BEbiQU*Iy!<w`KpO7t9Y-{d#}f;O2i%JFi|N&y3nhlp#nu~
zz4>Wz^cEjo?_!w|R+u+s3Q``n1Zh?dDa$}+&^?Ic86g^?mWrRu{s3iK9!wS9#2cR_
z*l6{Q25I7SQOHP2mAjeEhA)OLU8HU!C<5K(VLhHPxV^89)<%b@i6sO;bptvluM`Cr
z6V{*c6`qzbto}2b)wz>qBNJqN*Z2JCG)drOIp>6D&nRnG*Bt-t6Bi%1ahF)c2uZqH
z^CwHfwhrsSU@<zt#M5Yzn56JPOz>5K;eY3P!m=9WS<3^q1uikjhu_pSl~iT_+OIvt
z*z&~}sV$%tmew*!(M$Be#wPY((v3~ulsl1S3=Rwz-W?v<SScPMNEA(SmVz6>8+>U+
z3ee+0C=j#mDC2=f14Q+4#n;nkOK|?wis_?Us|TF(d?bxB(U3#WcS5AtNy1?2xCotQ
z=mxLtM0|$kgq|aSDo^aNMx64y19ghKxGh}V7~j}wt)F)gSFM?bW_V79z-)CQD~8SN
z1!CF4Y8<%<WU*`vUF!W8F9!ZHCV`NuRPL5i2eY?MdTYY(I(7jJI;|bx3NY}K+kn#H
z0K)8O@_t_?-i~59yuAHC$B7wZ=zZW;t(C8r>7G3);QD83PCXOh;bC<vMM{8lt>E)v
zQp<;)(9G)xG8dz>9xMoDbuTJG$MVftbQb*AWFL+;rI$IUYM`emf(5J7a0KN+TF&$l
z%W1M)ofhI{Pm#ch+@C0)X2p+3E=p4hop-N|D(RaBtkhqS$0PH^fnP(R=<cZ$&)t#t
zcVw-Ron1jD=Ay_*d7dM2qO{fweusaCr#?nVnW0k#hwfQv+Ek>!EYSG|9YZJwhP*xZ
zZ7a@0=4W?YS{f*leo0`dJ<y5q>%+=3cOSEme>cY^(2(POX746f1>UCkthz+P4F!k4
z)0y6z?m1<q+;}bXs@d(c2u*toeE-s$Mj4;j@yELlLhY65DB^}US2jb7w6)cgsk+}5
zIj+94eY%j&2j822{wwjOOqKuDVGi#uqi`#Tah%7>0SuGZ=n=GW{`0}jDBfU)cJ|PV
z+)ExyGxAT}oM~k$_vU*%whWAoWTnrbFdT2!rSl#B2ch|&p+LdrK$CteG6P}~O_Sl~
zKO`4Z>51iXBa$^l|9M8M=7TlT#RHq9cLWb%=D@w52Pa9G##g}d?A*Dixd1)#FMKDG
zk>c6TA;m4^(<ACafiVL%07!Ov;y9rH)TzFvZ+e9S1FOaSX<4xBTmY+XY{g`=)^sh5
z9YYW~k|ZK<t?(W(eLo6B2>6uFNZ8JM;ubV)Q5ChJa~EXvw3r0~^VgABA!H~3MgvLv
z6EqrBMA!Ks)IhvpZ-kfMzuWrZm5gt0IRqn7|DMv_=L^PsU~i~WmghPytg>sfnEJ`i
zGIPv)-i7+%G<&_`J&#Coh#`ThE?~ok?^KLQd%{MBdLYU>B2*U^a>U+6Dr&F5*$HL}
zQ&}Y!>kL<Cc~23=){<W9Fw?-^D7y0c1^NJ4x?Upk1HSZ*x}N;~fEJ3BAtj%!Dx9@e
zx!oD7z^K2>^hwA)TQ^QkB=88G6m{k9FA{UiEzgCT96>Xbnriux%7-V%_ap<%VB;Qf
zo)iJ;D^9SI0?f@#>5(nw{3q|V7?<8TV^2Qrqeewj2-fra<=LLj1bc1$1?EE!yH#HT
zF$k#z6RaD_xRZXGX=+--?qMX*y1Nj!BN@Eg!*v~%;J0zT<NlZNyS%j+L6V@C4}$A8
z@#AEs=pA#AufF=3=A|sRgnnWVZ-#FQLfro0X$B6gg=`&j@2%v~y~qQ{icHc1OC{ZF
zAVMjqGnfLRg#_S{67U6pK%kN#U0MU(5fGDW?_5dOH_0x=fvU)TR!NH^^U$Dc3wW+N
zi*?kv?F9*S!|sl`)*vO5^^NAm@U(6A$!zSEu?mylA(elm+6|f~p63713k=B^AV$PJ
zQ$Bf;)DD*ej9(4ld>0gJ$oVH=4IqHN-edn<y9IUNUO1D-P-#GgWkHjV20T}P_BKZ#
zsRmCD^v54-%L5s(B@zQ1v>xU+F$qqWFtT|k=_7dD%xxrB<zycF#&%V;iz$bi-b{L8
zU*<EDf&xc6L(Tbmf{T=VCkuX=H;KXUc67xodg{M(EB5iOW_J+&?+4mkg0O#~gS_Je
z$E3%9(`Y8{4j1dE>bG=$k`BddmL<#$953AWop9$s)q2;ZG{+VXd)X@8OqQN&hGZHH
zv@h7#c=;rP?G{ZKpi4X{YZ;|(g`Ga9$i|Pxxd>&O`ViJsQ=3@3Zzx|3kJ%|cEx$I4
z51HH**d^Q8c+d>J&fdH((hTdm9<cAM4_#XjIL#sIK~?P^HeGH+7s(-eUr#F^gs3}X
z&1M6>5$&@i_%tF1$@H7=R5*p2^X4$J#KB;dkBNWprqN2<cg9W3{5-hm)9Ut0%1{w+
zoA;tjAK{wQ%r#v|RXZ`vI~hNI{=xC~U=S&gETWZv1pV)vqSBF=p0`|Y);IgQv5rJ}
zImNEZT>*6gS>fMooeiRI9ZJdl?uT*Qk~MqceG^F}Fi3CpRL`Hxp1MKOb(f#dm(0s?
zUCExrJ0&@S;NEDaX#V@t=+ZzBwYxE}C}S+=KIhsovyBh{*i;1lVJNK_9-HIY0*z5<
zy0b8lLw93zBUldvi4+#dFYFdjq}0Snd5687y=C#Z3b<TM!hwS|`m(<<-m@C`vgy|$
zuy7tMMegtnAWL#IvK;s=abu4ysu+XAFz5Eq#mMP!Ip=}rD6|%+<pn~tJQPDZMJG<|
z<~7=%>aD^9gR#&5CYhGt-l4q@m<J#AT`Ff+zE!E6?%mS~&&-|-s$sLjOFl8;lL{hd
z=Jjy&BxowD^7cK8Pf@uYRKz|kf?>wUXy8MpBMt9Ze2Z{nFexGQ8RA{uRVVP+*aV~%
z^0;5A7O7XrvujfFx9f(X+|XEKx%=Ut_(Zs_H>Z5^-&@%znuEYMZ1AZVm%B$C4{8=I
zWG(19tZ>T)pVBgo<^=oFoknCobZx-x&0w+PUvorhw;yIl3EC#bWST8ooyM2PhiH8)
z@I6$fcl4dlb<i3S6?V_k_4$PJ{w+EE1<F`M5pnxr+BTzXR7nWI@W&xOwH)#~c>CSR
z2&v!+%X&9a$#BG)=)F+cwwYi4`m{8nl_H&-@jltm+VJ+XkhcB~!|6}diMP#<4_V4M
zBY17PwO$!;K01N#ryGiqoN|jV7BX}lqQ5M0j=!xoM0_%B>_XgJYP#+AHFk7ss2v}E
zT?B6a1=yRpj?~@y?|t_5#08DNuIYVnIsU$F*cn6H|EqQ|LxgunR{LY!#qy#9xwOq{
zNVZy@j3r;TS2F6|BHwf6KNCl(pFRi4MZ0-NSSxlf5ZyfT#n%r09PnwI6A%>f0DVEC
z6`KA;R(rgUW|Sg?k4aPkq-Y~zzj4F&X)x&~$hc~@i&o)e7udP8ZZ?3DW``Z4b-PRt
zOlA-8?{mvQ_6fL+EDNR&{O*83&n4m?_EA96i_O>UgeXL1{N%#oHB*a)9Z@Z0QOm;6
zI}4hu<XemGJ@d_zl(WeaaXc@=Bg!p?eYq59VL(&I{>8$Ah77P`Uh(fVCyCGK2=Ox0
zv2oyl#7E^XmCGbLgnGFKkUvajn|r1BJ+*@VIugXyjtVYBTC8X$hLd~6hEdnq2dAad
zWzdcVc6q=&P2Up&->wnqQm-BxS^0c-e$#@9(Qm$aaE-0QR+F0^0E3Amzu#0CGSb=9
zx`sNYmcE+EySC0_xk4IAViR7tu^?178V|o{ZCt^@en&F%u^;c)RnW{tG5Klvb3!UR
z3C)QtO&x_e`>c{C*J+W}SW17Q-xEl=2y$Y#Hb}(Wd_{LQkFF|PEB)LH1E(n};%BPC
zzH{5_)M+WJbmhhCu6K7E-ODout5(SxnJ!bAmXFI%2?HVJ)*}V)lqP~JSJLmH=7vKH
z;w4Naw<zka*)?g6ua}P@m{F&hYt|Lv*Wgcju1<3ADFz}L+<b|T=cGk83_e-Olag2x
z^XUbe>DAIQUrP#4bUKS6af!}2nRx!G@_bxB%NJaVf`Ji*-K8U{T~!ZxOGx^nLybQy
zYIzARvNEPTWtLeHEFzGd-l95pUMcua-Rwy7rh87<TRR>>1>?yq6kDc_v3T^!yU&%2
z8YW47E69mAm!1TkNbcFhEh8#HVp`;x?T5;L4@`8+#&p@+kPElLe8$w^9dsE?eOXzo
zHDKhZoRK~w=WGMWg~5TSn}_C@|F6_N)$0LU@c}m#-7a>5Ir9;wp)1yI>of<-)k=<z
zLqrdtB3UC(k)d#mh9Fr5UolbMzTQYq9o8HGp8k&EWAEaVzaIa0j?GqkIZd59LQtHy
z#_8*2JvOhZgmZF2bV0JP<PR7`_e^60+%~S&{F?ya`IT4o9$OUWq7;cNIaw^+4aA(K
zKQ4|W;q-?zFM*TsQSBE&v|9CfR<%~6PRs8Y#yy~>V2ST=-Pl%V3m14@@#<C4%on-w
zw)dC3@56p7=2Uf3R;cpo?l)7rR%3QUn0l15)kPoJP6{Ury5|!bJH0JgVlt2(S$lG>
zJ|TKtHalgILe#oW*R8@ZI%y$FkW1eJSWoR^Uc$-)-1I{p)<_n+V-eh46O}Y3i-8F%
zCcoSrxc$wR(-6ZekqGo;;qLFM$ZobZr%%V1?l+Jh);-8`Y-{8PF~MRHgQ+bR%ufa;
zyx?io_gAkpvni`$(%=ehB9ji~%_^#kPc%5yvJs`x42Vu&sf-KW3}U@pw6s94&^nPQ
z`URhvA#5)tO;Z>kvG4l1P1^8;HIhM@*RO1^<QW3%=OgDCH{G{G-KOa`XDLb}CBmS~
zzVxJHEimIr4XZu>DB;YJi_phjijp&%$h7*LCb((jX;@F7Ol10!_5Q8ZKdsUbJ(5h(
z-t~MU(;*G}b|lzVxuE@recE+obsyy@gHh8)tU@cs+l)7n|DNR6??uoZ!$MWMA5se%
zCO7L(#heeO`0w^#TG~sFWCgOdstP@hXc7Iz8~V6`i;)I<QtFHq{6xb@Su%=WH<k{H
ziEqikoiVzRfaI2ZIUO`#`ht4v%&)w3OiSK8&zX{Wn*{>t20CV|*|2h#2~HQ@r$?s}
zsX0oCg!CjBThvZ_I<Y_QGE6{0m)+qCV^j>+Smt-7)0{oo;%Sr6QG`)9a@#r4!%2FS
z+e|D>y`JX9Px<@rwbufn8nk7uw&C%uGzt}85=vouZ~X4xmB_y3Q7J=s`Q5-UoR2J&
zUDMDJ_`D@$BA|`V{j<Pv!++<ZG*cQxREmC4(MI+%iLb0Pqh-*WZnmZCQTE)Id|bF^
zYlG4MkX#ra<a%X5eG3;eo*ps!nsRC2?wCM6D;PO0*Be-Er&spzaR}8%OvJN8t-ZDM
z{eRq3Hh}lsH4k~SW5fJCl$3w9yNXFN+RyjP#h2Ck6#$~ud(8Wk>D_ooy6pJdiUhv&
zyxVtuy(F8;{ogvj`m?ebiRV#nvAnwU;vl<Y68up{@>CP{>Z?~glBfT?se5U=H~Quq
zuW6G?UgJUi7Xx2h@d5pK))r*UVTfdBu&Acf@4YdP!KKM0ZyqD9^7WF3n$TenkyhhS
zJLpF~QJI+OH}!=<@R$7l8NaKkIWPrT#}0~iq3Y^4vS+d=^;??O^lsz)afjeLH*NC2
zT<NSINl_fN;^w+-pC^D`7{Oev$*&u(pSu?Yg)Z;J%74od-UFn*YBZFFaLNfnEFy)7
zo>dz3*w~yS%$k#F|C}fBcbw=GhAN2gk!X$H$}#ICi&A+#k(dcG@@RT)<)@rpRf{VP
zq$YgP?ai?(>=~Cj)+jyAPtdLs9@O9?9#JRO_@ku1YQmsUKz9eCtS431v(#Y}tt|mf
zxz989u1>Tu0(%1MNBMJq)K{{9R`3t@YNu7u+CwQ(%nIV~Ab#&b5$s~fo`7M+@H6W%
zPqa)~{-AGKIJ0fua^2*IxrbX2!D=1JeM>1BlF{n9F`Kd3>00UGjEcOTZ0+8SfcjrF
zU(2pp@pWHPaoNrNC$u_TQsgV?vxI$bj(ab|*j7O1v1zYM@8xY7yp`cw9OB+%L>EW)
z0pk&^#8#aetXl&0e%ullzOH?-<zv-}{^?FXtuZ(W<_pu?5iVKi#)-uP@#Pu6)OE&D
z+1ze5IX8P6sCdF8THG<OnihzGADr-$Wz*}bul>oYbbT3`m7ByTXYkC>^N|{~bi0*k
zfUGN9Gg|f*C4SZc#&*jF^QxMSEY&Dpu&=FC{tYXNFp4{G%#CHt?Hl~avu4#a{E%R1
z%U(YAb4A^Mdvu>x8Br7GJ}%XjI(u31#usTer?kaE{$b$!V8-nG9`gDZYY(=zi9Qqi
z@HRS{Wu?qtd<wNG%q}uz5EIv$ev<Rs<b-R>2sg}v+QfTvej~g5HCgrtHbhoc6<Y76
z8B}x-?QZjaWO({msV}uqrNp@H0b}~umm%`~(EMnkLh4h$0)^t6_OarqjU&>?WPY#!
zl%;Gbb*KFuPf+T3Qc>Ej;N(&%_+P4P789cd{yTS1rpWUZi4{W@-Wx04)?(P?p+^D7
zHafnw;D4j=%hr0$kRdX|(jQS5tx3gNjNMtVxeV-kjyaC(z9VZ(LlO*<EEMnsR`7?d
z9s6YI_*(3dWY3q_>8;4lzjIy{7e#AIMf3LXR8M<adIF~vcIn$azKGL;HKpy#0W-xn
zhZCO9@xLl<n|{#KK5w;Ve)Cax?Ul>DP<t_*ow{kUrrKpz00S1FOHaMge;FV-MD+$O
zqGD2S#0vsaQ1Lw~OX7|Bc?_KnYxl2hEqhp#xfj19MuI*lP0Xl=<tM~cYH`Ic{$?**
z#M?+N8H`Ik@2#>xtz@L7w1RvIufL6m<8?m2$)AHe-NVi+Q=jDwFf&>yYqs(B^j6!a
zSA2*ax0JATC%aHd6{oSxJs!o(UC5Do@*De;Ed2rEJ&+EQfb|$V)a4Z|AjF(#7jt%b
zPkj6Y!P|w%@-^Q6<|gbXqx+EETO8t3@%(9SzA^bmr}y*|FU2Po(lv(JUcpR}BFt>V
z<cG%AMV58=mvsa2Y5W)ZGajwI>7duHYU#?G@~?cpALY`y`gSF%-D_O4G)`{~Hipf(
zEKJj*^kGUgTG{oE4_ca4<do^^m9Chqr;+Ho;6Q<EYh~Pa2F2>k&C>O4`9ABX<>UXI
zyX{DW$-9hSo#?r)<2SG-Qtz9%ou!>a*7_xR8kb5n`1U7|_3bAgk=0G5lweQVjk`DR
znWP8xGNX5tMv~L^BI6ZJQ14qsFteC}JmjU#E00|6aJQEc8<1}wzUs!c=4M&2)<~tG
zPSl<M?NH`~w{?qtF1clWzY)%B;z-LUN^mMB_yb-;MJ5;!B>Ct@9>%bYJA<u!0xz`S
zEyzR#-xdI+HcO?3w_U!PEvacvze?lsUd~BM$KS=M2>q&92C2O%Vj#X!yWRY}hL^!6
ztC~iF86k44!fwm&cgWM68@6aJTsY$YadaMzRPX;E?}{=rPO^&3V}$H7zs7NF+1cS(
zWhG>vo7J(Mb8I0ymA!ZNv9d}|WJMiy=+H^WyuY{KU!a`h^Lf8tujljec;2kD#kwv7
zY9kki{@exONJZp!aQCr_7IE{g=d(mD>5H=uEkqAygPsDDWpOs=0(c;TTcrWu7U+UU
zGXN$(NC13r_Ny*Q=5!33Q&DHeWy4bQTmgoi<>exaDOLR?bC`9rt9PwjueIai9?hS9
zQ>?Xgf9-{F?VlabV}|wqIZfQTJ<H%L=H@o9=ugm_5Ge5*dgeC-Ywb%#6xo~(E&e`5
zA=Z8CKVJXHCXdy!D4>?%sna01zgFNK8Jc@spjr@YH)EzGO8P5O{7*~w^VrtxZP}k0
z_YHi^B!BmN&9(mM3NZfMR1?LxRJ1#w>7kG*<V$<>)?PfeuJwsXMjq^zJklTym3Mi7
zZS3>mzK?6AcXLf}@hMk_g8ioDQya3MSKCc|S<#T?;>&`ejTfAoL6zU&dy}b)evVyz
zWsomQ<l6&YVJzEL+<Ka~T1C!=jqS5;3DF2oyK`6Go-=jT+xBzK+|#K4n_;=@BeLtD
z*qztntFgs@vhspq?K%4gGiLoyv^G}hRE{MAS~L!W7hcrH?}bxh#NO3|xmYKsfl!rJ
z4azvViv5@~p_(M<S_)O<y5(CI6)2%0GJWc!ZhUmg#h*6b1<#0)ha);>G)?3^8`Y=r
zMt@3WliWF~meX)u=h*-X$P0;}&dZ(WFol6kzy$YJI6ofjfKQMP;LiK?X1*5kel=#I
zR2;azU_ap{5S{T3Rm5G{FU7Mr-VnA=1zyk?XKd3CPHDhTMb9C|`=0pS(;~seApV50
z=yYncO^L?d#V?0bBK7BAj=Ol!?gHZ~K(nKOF$NzXIC&db08*VN{%-}$N%w@EtTJ<Y
zsmn{$6K76b{98upOSWQ{5hBb#q7oK<HO^9M)?$l&#GX=ix5?<aB*v@Xzl11mDQ7|G
z78C7Ee8hn;u3Vj2l_n$pw9-5WDoiTPca0{xokpvVoo8ye?`F*2RYj%pY@~M53XSPE
z)QqWq+ZXj^o(=d!?*X&wYw%@FQrEAwwLcpcvHZyT(8es>wQ*<o=UR(x@BvxD-^x$v
zvan2bx~tuQ^kQ~=`^=KZ#@##Rgm1-3Gr#B0QqS_PEq5v-9$lXKglY<r5h@!;>Rx2B
znthn0VEmm%QHp5u#8@C!Z{7L8?t`ydPx{3O<ULT(bZyGO-|##Lg?yyVeN}eEYn%8>
zlRo#r`$|P?g{M(vHBaNL)@%`AvM*z-5*5B-GO-Q)w~}BwOPUzdE#P2E6c#Y#m}!b)
za)~2FrRBVGIq35gSx2F)SGibn^W@xT@Q@Z3K!#+Bq<lPaO5aGysIlEay{L%$MY;Cb
zI(q`2pzVWr@U;JpSP-{Jji1<&NVe{Lh-AGm9k0`{r9nfB65Q{yevt)|=$IGSSCY^L
zTyicxM#(^3<zqTBCi&lmR}&rV0n-H2McMn7PhalJ+Iobcx{=Tl4m%y8&hw#`&1<hz
z-ii~u##U<tFb4w1a4vBqW7OR-i*;T)rZH(`<-D#rxs72hc6Th&q9YbVlYDA<w1Oe8
zRS7?5eis(wdZ!<YQG*Q(KpTV#emYbW9zGaiBg6InsQ&pn2EKEEDG;hZljvh<c{_%7
zHk!)&9QEQ^wg`UKc_Z|k{DxJ_Q(wtIB|Bgp&Sfv2MtR^m%O69#Q2s}vlzdLk-PaNX
zL2UD^AG55f-nyKwd9u2K{r1PaZ*2B-W$r5AGE+t9WvvLq86gF!64lsYoX%I3JU2FM
zn?UwrMuhmsM21`UzSKxTXjfof#=B<84~?szRWx(QntWBojtdN4=Spy`vWs`vl)&sq
zdu_yH+=>QGJrql`emz$GlgC>1KF33ZXM-pyR5lT$e+w^0GJ0uWMTbO?(<5nGpHI7e
z-=X6WG@cC_9ci#dC3>_@vXSQU(v%7amX#1?yL`M2(QBEt!$0uodgk9}d7+mYZSB@I
zmXG(cnevp5Bpp3dhg5Q;NRmK|*P=YEmO5F#N);tRQv0K$`KTsLo9h{m1*XVWlEQ*+
z{rj|4e!<QpdX$&@whWYqWh(K>y94U1-a+qYO1a#QGJ{k*iY`A}THb1N83d9?Z)I(r
zFxGj?)DL}By>g&`7wvQXZS+v$-pytdHqKt~>1KFCB;VN2_iy0k6hG-p-?dn@zHCL&
z2ZbqbrK3D{m96iyJ@$Y86W%4iB;Ipn>{Whd=)Sh@>Pb>kzkfa2-iD{RDc)B2h$X$n
zPbd}Ot{a6L?Uin3Qu+!Z_TVod#N0&zayGG}j?X6aPISNe@4{vfTbB>%l*&*4_MhnY
zBj%e@x)IsXrbXC7s*36muUKOjZ%WbSmnns`(b$e`Xd_e}B-JSM8mjD4@MSo5XjSXS
zSKMh{@G&%&_*<}rW7Z$J@l;M_@o!&KNgra}L-{N|RI8gEZwsO-J<{W)wX;`D?;e|m
zts%&8crAv)r>&NC;C$r1Kgrif5p#}dQUgUY(|EBUyE)<QqJelq6z#FDwkaBN>;E;R
zrFj%EtDsS%KW|N=aN|V4SCPWqZoU~qP9pR$l)ys5T1`rnS%X{xKty7$w3Jy(9*orF
zW>Wj9{HC8Q7itB`%zW4L-xzfNwO5&zPa2)i<ZNwn^)ygv#M3==@yii(Sbn$+K@D`U
zmAX67y7eo$2WJd+<#g8tke4GO^}+@c>p#YN*o}>6<lBos%sh90tXh-T?)%Zc77M@b
zU5Pb)79g3XqrYw=$WAwvFf?%+>(|ZHX7ib?JKgtnM2V%jY3cP9qT53Wgq4gUw}WpV
zmeJYOD%}|!&>pw;dpQ!3vSWn$9DOErU<Dz#yj9>}e(Y|asIw%H?$WU7pyEJn|BnFF
zJQ&w1Z9Zq9uDj0l7>B9XfpSa1Y@YEnmUv9H4a&}qPPple5MJ?JSn&>*Q6Q`KjrHUH
zo;*6CzVrEYlRJ#rpITo>|A~V2qO5?q6nX7oko$xqE>E)0YeET$pi{ylAo&XU-$iDj
z#RBvFZu`yuUFdQx7VyO!+}UwLtM-wf&UAAea)^&w)UCDA>?HnnlN^dd<(#FOtrd`y
zi252dJDHkzUbNcmwB%{Vn!{6aDD^G_@lQ?z=n7)TRMB|q5Y4o4P}JdOQCv?Z{&&GA
zc);1s;R6i!w}==jM;MC@rJ1$C|4@Pasj77-vMu`BazjsRomDES1%SW$5FzO@n|!%N
z)ltblBf16Uz0+L0%(IcbDIqFUaKS~X1hUgVEN&Im22tMX7J6h5`&GpbN|e1zmz925
zKM|cS2-SO^bzpvY&df>pMB}IkR#04mf>76EwQ0edu{YS+0e~axL{RV?83btcAWa4E
zAkLS<1yyPGfgKN;9Yzban0Zq+%RURJrfS8=>)r;6GR>BjR*NZe$H?opk1G`n(&&nQ
z^Ha;;wH(x;K!w^Nf#}ZPZyzPNRw%MdUu1jQ&@Ho*{J<I4dv?gSQ#pZen!JXRb5xHX
zPtVTyxt}b;dQDz8;jPa?eqAK<<CBgPj`Wz1zh1kBm!W^v37jmQMmkZS+2{mn6rAMz
zmLAQwqVPwu)ckkhO2aSk4hNv1TA(Po^7*vsY{b<4l>Lq&!K;ItRB&tWn9rO#aOpNt
z>Awp<Ms_e!sa<ctQDEcFUnD-PRr(UgGqU-j^RsZ~YCcSeL^f=ttv@;dw6$&%j+Hi(
zw)#AijuCelhOFk;g3?QDJuiAhu<3njR$njO%04ivd>?^Z7J7Pf*K`()AR!Il(zX@I
zAg_NABE`x<cO=^`I3rRZH<ofRMvM!)iHd`6rW^<mN3-23!9>H-;!Al1b^>7(4jotK
zBZbR$iVaAB6<%XkcoJYoxrQN%eEGBi-ZUuu9L%gfMgq^G#=i8=(wNL%M)KH9OozB#
z1e?l5$8op*4?@ncAg4IUIVe+$7C0=tgT-jXg2swk=%pJD8xFA*5S`*LlqDW%Pi`&C
z{M6n$7uV~dwoZXm(-EAn^3Az<1OFG#z7YQ$#Ef5=-C@&_oJ{l^5%;y%ufcHZ!x|Tg
zgv=dspR`&^rZ!yJEhvB9K1|8!Ck#_OUZf!Nqd65~W7Bgq3JHB|!1oa?Vz!+=)!m4!
zPtz7kYOj!0Ps*WvnEgFt+JpV_h7+DI)512!zK@u2cr4A9|CPe!d%1*TLdygc!f3#^
zFBdZNn;93HNA(%Z!r<A?tbd?EPpdb*>R_AEiGld5-K;m*XMJ;h+YO27bH^$i1aQK9
zm+&12U4rs$N|NGy^PSSBKlI**%ZKJ>7*hqF(SVtv4ZKWZ8$3=mLE1-dw)m1!M(N>#
zKBsidk;Ghtz!v8=N52DqpF2P^E)iGJ!(C^v1GXUWc@iY|7=Xy2dYsQyHEz$A0Q^Qy
zZG2p$&aaKN<0v&QXCpH7Uox*&eBx#pm+Hia=TxD*qI;6`P!Wr*65_PNOd)OO?WE)i
z(zzE2Bo_gyt=MhJNncQB?t~_2oZy>cf=kaf6=JvGYuj*RU~_^KCyL$P*LfiKSan>#
zh-7?Rae8Bo%f3%xkg1v>`K(5%@EAipUPA=@ON`Yf9!!zz-Q!gU8q2A*%{3w3Z~a*N
z1r*M7)$W~1e|=x~^)ks$a<q;QbFw;^UaYKk_hhLcI+*g8vUm7-9yutZ6Q6^wx^(uu
zHO?j^qa#;uRmc&2O4kAYocv$wrS{YPcY(<A7rEF`creYH2yUDtn1qCMF6hLr_GeEX
zi8Y;OQB0f0(>}s6Au0j^i=AEDIkC}+#ALF1X7Bib;zk(IkOEAWP2q!OJo`T8dXR`m
zn5zIWhW5B@K#90EmB&)RYPeYey9q7;rKKl=gwjOt#<iK%jzX;pSYQt4x3dGn`P<@v
z>A>@i#OOYdx}Wm;AF1S>7L&^yA#AOu^u(a_3;vBd-=2I)ci9k(L$mSAcg}saU(X|e
z+b%&B)v${rqW`-fqS~m+%Oq7EYH$Ktc$YJ3rvt<0V&pBA*8TuOn6BQqMAP}1kkD@f
zBb`4UWle^n2&m+*wzkf|C?Hnks*VGoE=#MDDzAsKCkCsG(?*3?{>6e=DdfH)cW#h{
zgUAm1rKKMqbuzKM_A>eoo|t<ZKQ`wjF}Gjr`C_V5?3)-r{FNc^JnEDcZn}H#(Z$RU
z6=5tN=KN^gBmy$Pra`F{ChKqMJIUPeByMDYy<iVsOmdz=A?XUkPF|cz_VM6aPk!q3
zJ_P{ScYurrrM&FBe42hyM9XC6xez%V#d|3<x2~zP$Ov-{SESC0PS0zaTv^1v%kn`x
za+I#@!Zq5XYBJW+1{vH!;drvdLUys>^1dRRN{}Hon$;+l#{b>?6F&3RXA8o)!mLi0
zR%z~@nTnqh8WPfUKLts4NV+q)@FPl_BWfUHfH=NWnmqHSG@Pq(Tr6BYuPBdCr;x9V
zJBsB%Ef26=ODUYav&7`hupM+5ApWM8s&S=_M}Y@HfOWZEKX@Y&EqLX7OwHp7;gEhv
zi*&=cp%9!Zb%~I;4$x^5rnS5zn#MAqN~BU{cCf7Ga?DxD1HXTfHn2ic=%>Zv)CZWU
zZK#&$xdPDNRDw7+51*s6LOHnu9;HG)gxy@uT?AzUK7kUerw5`Bn^s<F5lf3k8rbL`
zCO&?B4Vxg-b-g^^eqT)w!()Rk?r;fIrxCBrjCr-V!}40c+_fMF4{_mF+BDF(_Zu(#
zUKfz~|3ol=Y}=<{KT{H4|5dx+BqYagWa>%r+UmYq*dhNz@Fuc_o-jywK%a1&nL!8C
z<KD;RdE^t*Y&UgSplk25gyi>*)#Lp$ZNU)b7X{ikt126+-iNKPZqPxk{qg*_t~q%t
zr(eepc8P{i#?;Qk<r;$1HZ041q-;u65^_8BxYcct|4l~PpG}J)NQH!~rvn;XL$uC`
z$P|h5C=t+{gIrKJsB`5J;hWwE+B8~j@3hE*-S%HlTTKmpt!@LSsXSVDOv(1{$C8Z+
zk!39)Rlm|4g4@{^IM1#DHX?X5f&nC`rqeB2N?W|sTaB}U$5r83BpP<1V^Hw1N+kj~
zoFfaMDyZBn%8N?BT?;<lu#9Jsd0WkfXzcRf!x_rYq(AUZD8EC)jmF3#Dyef|`VHPD
zr_sK#X!*9Sk@SnxgIAbUk{{<0jaQ4gf9<c@yb<f`M<zYr?^~}+K9L6Lo2lZfkKde#
z=Q|&zz&3BMt>KUP;%de=-Ft0a-KL~lO=EzM^$fMuNUI)CKm&jf-^V--)3R;vY&c$L
z-(=WC%zn|7TE4?w&mVW1E4+wd`K*(4lvZc=Ao{`Tzf1Mj<KAA8>5C(cN#77@)M&sm
z(E<Pe7-cnjpiBn8RWtW(Jss1_Wg-#4LGi!8{KH3r38z>2sN%z5k5x2aYD+>zpnq=|
zlg0Y_lW~~DcJ?8Epvj18X2TC?oHNznz%0<cH0<iDv1oGG0Z9A3lw&~oMPTG21;`Yp
zzh(7dyua0T#UGnIQ@v8MKC8mcj!PT*TS0{C!sY}Bz4RySStqsW)-nS>EzkbD02&~?
zKi9A~>k|^1os%A(*%*)3X$U!!{N9k%{Brhvq_LDIlwu0qBY9`>$2BKQe1DtfgbEnm
zrJ;VV`IiF_Jw|;I3szTQw<8!Im_;UaB7(x{dRdjZguh^0QR_^3*;@1{0e1fCIi(23
zFs5>DyzZuxX({9~q2d!6LNTwMpft_O>*yrK1te`}u1$xV#B+)}WXKr!g$slO9jmXG
zDP7heuRu3*1IK%S+X3n+P$_}I;8?-7k_}b87-f;>3fMyCvG`34ogh#;w8N8i|3U-R
zbs!DTeps%hpXBtv>@g#&N=hbvH;DaEa-CTdfPD|#cgHqiO9C>a3X1u5Fi@Q9DO|dv
zk>1xgRv)6Pbd@4Zr?zs=TmSvft5KOdmPU9DH;c*Mk*lvZtaBM}RyK_7+7dHt)$&<3
zEepa}g})_`>P6m)bZKjTDEg$RO7qpV-ri*!t>huXz+LkQwh4>-<kk6YRUo}k>g^~#
z!C0z`)04`XIt`wlPtjt@11iu4!Wm;RJB7FV1xPOS4Mc-WYe@++Y0Ww3>{o&?UEB&i
zg%*`qYm~RcdSD7|poWwX!hs$w4IPtd?Vcw{p(Q3Vv92YoToPSc^UjJnMcF^%(QY-S
zdlg$<uJl7`=LQpRNmWdT){GWJM}TAsx^%PQpd}ZTT(ywRo*BVKeAEiVKNsfxC-~|Y
z;e)6A;K+HgBOSW$GP`z0>oZLlkRFl<QsFGZU!vV%#cx7mXA2p}eKw={eLus~kHF0>
zejM7U$1TN0cwe$EK{grm(+M0KOY%qCFxtE<G`n2LK6U_E+;4;yd+eYBf`BYny1|6`
zwHk_d@-O0WCK&~ryZr!U&Hxz_Ol3UR)$dXoQa~oIWTtp}ztk0CB}Vpf{&6D_>VG}-
z+NZTe(T{COGO<B0qcg}^!fg#{w{eQd2&3+hi)L!T%#g)w@7J@#+(8hIb~Bv3>hEO&
z7klX3&@CrpQ8j~g;585_vd0#;Z&wGMwRrt^LE>0+)n{$71byEzUHGD3?eH3Wr(8@~
z&Y%U*pc2uo%Vb4?&s?wkb4Em01<c(nbQ(~aZSB>N^`2E>5u*Hokg(DnRdWQr&_nyr
z%#ZgKROgHkp)3paO}5`%O0~tG5X138KkqBDE5Wvy+R+wfCH;^3Z#Oz9y7`XIEX>2H
zE>_lPzc$EP@$pRZovfz*VeBsRL{(DP{};sk116FQ5#U=;d(mysru6N!X^^^|Er(DG
z*G!M%l4(C9c2J}X2|&ieCy(r3U+rK^3j8uZ+dQIU)v8n)ph~CHwKB=^r#hO>%+B5K
z2K^#tUqv?ZQMfy`^PPnO^I^v!*{Bgo=QZ8Ivw9hzT@e=39R~)xaL}?b1EmK+ux%Il
zsoBxRmqCIN&_%XsfmQ>e6{UXtfimJZqqQ|lSTas$Da@T!$we-6W*mBXyHPw>E2O{1
z@ly*nRC!!VL#=gc30~r$35fr5lB3Dz%6F?MT~iA6B1@ob#>c%;9dn2XRB=?*7<=wt
z%7dgT&<|zpfVExwE4N6>N?i0tC8(&Ca}rJYMU>f8QNPklRMcJe`|!`<t~Xni*@QeB
zY6p%r!_ghEr|e762DKjMe{Qf!v4UtKW$qjy2w<Pv(abhwoRgg5(1yO8ui<P;ah`;i
zdvyZZ_IZ)1Izc6&qvmpIOFkjdX(`^%6;A1oSZ1`p=>BllYPK6UNL2Xu^|sgw$AKDC
zIWf%3fN31)cs$ycj&7%O9L8D^+TL&G7hfDnjCym-`Z^@sh*r{~1bK1djZ5`kU&dWi
zDvE#hUZzorA+ba9ib?bI#nwqKX_A1cQYSQ~V@t~N6uLCq#+yJ1yETGZZKpommt&xd
zRdnMi<sjevX6BD?98@MNynZBR5cj#@*v*#d%2nr{5sbcYT<P4lb%*RvDp$4y?Anuc
zA$tb52!8l|uQNU2>jj*&#@+pXk6dUTVv<nru=1_TcI+ZaSIq0PWXBeV#8-CcR!iAA
z*im!VAN9TVWx3pxjqr^#-+q)&S6;3{ZZm9kFb%s(<yZCHD&aL&KSAfJ4Em$B68qnm
z$YvRfwDMts_v2m|2tMt8@$iK)x#0Ts&p}`8DWX|XEG}Xc`Y}3}LpBAdSDp()P`Qz!
z%yz1CIn?GWd>2{oA|R{AJ_~HBA&D=|_qR3GyziH+No$eqpl)i(gt~rbueVB@{b}e2
z!v&41=vEJCQ18&2(`VBBOo{snT0M|3N$zW@(Dn}~elsC2+}rmJ=bJ_V70@czEsy|v
z`mGEX*6M5vY3yQ*6v#AME8>P3)X2wNff2mZMnOW`Y=8%>mc02UJN-gUUeRR2hP1}m
zJTsLtgdAR|xA)%#npT+q3Dliz5#sFg+vi;>Gn=Xo*cCViBJu~K!KMrRG|)pe&zlRQ
zILehT)XswJbfzi|Sv|JQ2#%*kb>5b}+kIw!c%Zk-rfbLkB|!Qm_6g^{`;2leBQ|=>
z>azyn%iFhuijCf_<BiLCC9t|%F+O@~QzL$F$D;e1ZVVj{vj%LQG)n#(_FDQEQd`pI
z!m^0411$SuVN1dP$o53%#hv-)KULIPRU*Fc-it0QX5Z+4_(o!VWIHA2`QJ*j2VY57
zwXAJI@99|;#@i%{2U9#VCXQ~z*eefZO-{<b#TtjQaTgo(Axbc2M@pCKT4#mHwx$;f
z@LHTj+cV?B#;8ht34KP4B0ef9^9R=H4)3c-jX#XSAAC{{HwX7k=TB3LpIyV$W_<JB
zf)dKueF{%<<)&k<55<yQQw!cD@ZZtZb{znxRgv}V_q&nIAwAJzXG(loODb1B5kmUc
z1LJzFZbc$$Fh;M*BiTNl70>SVHERV`bNnWmFxu-Btv9ElZ?)Pv<?FnwsgNq(N2-BE
zMHt_eCFHyO;x+g_-z=50C=fY7wP)UxTg^$wR~w)}{nrV65oUFIG)IcH42Zu<nu_s!
zacN|pK;y_n<5y*|A119*!X9V}6cBnZX=fYaQz=^4OO#@nhENasQ^8K^RfRLl0XTB%
z9Kbzb;X=l@tx6i4-N8YW)9Y=Woz`-uJX5F|1JMdUQV^%GXyOH%Xj1<cYJao7xUq{Z
z@DE*p399H4JsNndw~;hr=XwbsuQRiGXR}exq6w`A^d}haP!#$gLRKXl3{%-z<cURq
zgMH*q7>!JLuYp;XUOcpeJvq_eJh=4X*S=bcLZL><uDJHAN%j7HNfFP6dS8hpD(OUe
zTw;TTw;B%#FHI!LKVG?9+ng<2deye>h-Wl5Sf^$0G+6@+|0O#}#DA6vlHfGK)ZXW3
zb7J}lczrioLYd*YuSXADb{}<yNV0RMSeWV+=z4@GL^C5Rp{HMdhO$pS>C(q=1&l^X
zAM09OW2BoNJd1n#g|WY2AX!S;2w~2L$4Y3#dR?u1R{PX2W`g}yAz59p-B?lggd%k)
zxMEg#=}USrQ~EP0=&k&ZnfnRVY|E{4c@ZVIkRCd2k<WcB5G(v`Ay+~IBp0N+3wq<$
z<`R_iZ$|^#!ON(WORa{wcLNSA1=QIhIgLN^Z1J2FZy_(Yt#|IdKGFM<?^S^E2mUar
zAI&a4Rz8Ty#xb8b8Wtq=t8)D*sT2m4Z5dii40Dq||K3Y+5#o|RS+aowz^2_y$uT*t
zIgK`=ktx3m_0Q6oXKXx9xdP8arhlTRW#P|`)P|&NZ!`v#21PN3a)(&0lB_mIF|~OF
za8E}HlXM->J*$MuojxCf^#hYc9*(i70+R4go$K6V)OnhT$rlAil1RzT#zRMtDlC-+
zCUxy$nGUuBfU2W#znP!WEt(9S_5~_wHU;bQG~B?ashk<9C3!X&e_(l+ub44AuKJnH
zaqt>`s6U5g$2EG$oD)=}={Wrk5+VUGq40S_?o;vg==X7G=gf2w_bwmo5L_?0T$@-S
zAtBXd`&OjsH=tQ7C!t7iyC{S0th}4)z<&bb76PTE^fH8TqIvp&8azUe=v;c&tp{Yk
zm^yG6WsSt^JkRHxS=+W4?Kai&c=ZuRkk!r4?mT-(($91)!!?iJLBhLVeTDK;zlgJ+
z*~824AceYaE7+>e(~zc0N%8jC>(7{6O7amuk2$m6@0Bn6@v=1y%9_rM@psq#rPT{v
z>3ivJT;P?CeO7jg^bFb5dSG6;J;zIq9ZNN>8hkn%4{dvv?rE@oId|u=IY)bf8GY^>
z-IlI_YgH3R()}#>nb<V(x)Rsc{7+QX)m^rF@7(Usf8nb5?}E<VoQe8TL2;u-zDt|g
zYWt_TvJ}nKiTJ)saXs7o`$?f`q;KLU4y;0wC3j}BND7tk^5T8fh9_%?R)3Ocy{B6M
zg8P@Cex#^}t5xe=+{6T{s8Etn-%xnXods|XynOjG-?KFPE^hHo^LK=VrjzWneq;Rz
zMQx5-_XWN=_DJ<In?Q1?ygnnoyXp%FVenqVoS$x+-+e<jm0A*V7T>mECuxF5>a;ri
zPW**LP@EeRRHSEq6qi*ymxA1r#M$1A!~l?{qI6d&H{y%Vch^dI;%Sk_rcQEPT#jB*
z3J2iMH`E~xJZzYb)Z(IHohli<I;?X7B)7)$|B-7meME?*CEaw;0k*&r!NnNe{o5aI
z7>q-w;UmbIys@i)h`rm#O5Z6NoAAhEL^}f_XHIUhuHW~wbOtsXfR|HW&f-3X*niGy
zbt_@`#jUsNK7m*xI%xz3XsJ@$xJrRx+Ep7oMt$0s<^Oj?b@7zx%!TtepBBsiTTz&`
za@K>w*It1F+<Eeq9sjEm(7(mSKhzNOlzEvXh`slUoMN6zqxr9UD{+*{ZAr2#sr!W$
z6`zVJXNBtAxp|zvTnpiOaH3mWHlBWApH6Dn5bL(O`GxZsQPxYn@Uy(UO<HoKL7DSQ
zg`Zl=%E_F#zv00hy+`6AIt;OH_HPUn?SeeS-%8^$-+!7(`1Mb_!e8p9i3yzDwb?42
z>V6xuuZ>*Wxs}}7$wL*KIg*{urSgd_!StI>uFLjV)&8;ZJq_2NMdtrv)JI=1Y4Hdx
zNAWKx&iS^FUiB4ii>A>skZ$YuWJWJ?T8FWs)-70-ZhTx1^lcC`yB($_<V_q()0e&i
zx&22@pj9tj{fJGfpjoLSP)H(<t$i$BL-<>&TI>^^zb(U@+2W^75tMPiGAC-!f!|Td
zV)Ua3lNIL@BT3w8dV*lJ5kF7M^((Qg&MTMnKP&hw&Q!RDVD??RXRUc*PDL=8?~G!$
zM$snG7Noj;*4>2d=oCMR{#&goZ+lci`L2k=*Jb>RHcZD1c`&{sF~+>2i?LDA@=K&b
z5<jo7@Bmx8%!5F87ttD(3D)~&)<&2|^4`^zT-o!tF^W~!CK2dp_R>4{pU8@v{9A3u
zYTr{pK_1Mp$$hYWfJ~j(LT|o_n<S6>jQ;!QY}#~QqNR-QP_S+L+E&KO8JYK=J4mY9
z;!@>XJ;*3&TcUz}6Hq@1lb~xAoCaGgo1oaJbB9G|L+GssC-wZMQ6S#prSYzm0X#ok
zQ8H&A63uEt)Aj_P{L+!uaPDS*xVCFAZY8de{thQGNo;@WFOW3O&Gn8SQ-Ep#9JHWY
zplB2g!kC7Kel1?}Yw0}@JT)uCaIJImeLD1IVd6KZ@=GfG;mrII5J2h;VU;(#NRp^w
znAh=@ApYk4ikp7CF0)o00EZPZT{}h%sE;hSMr@m%7V}V~{9brlk9^gJ80V^aKjd+@
zbEY!Nn-`r|VLMh_tT(Nh$UG8ge*9|g(O5Cf?iDub?4jOd>io7Ft~mXrHer8c+Wklr
z>TAhO*Y!1T`FZ>coCRyd%e#&WGzD@&>sIBRGJ8Dq{XD5}!(&+DBA$~<`a6dkBQNd-
zci!bui>cE!ZrKZC%3Xe4`Uq{MHJ<;;J}Pa=PopVNYgOpSXv6A<&+n{jeO7S;AA4+~
zn65kxikJ=U{I0?-zlX(N>3mKWc|9A&U@Z{o)$)CP^a!fV&tLllJr&HP;W04t?n6OX
z57&kpQ=^IO09#S1kc>8)ajQ&=;{&g$6a6InJb?;X;$tm&iwzd2-n$&7;LXRx=o_el
zJIs4!Pd1e>c_B8gw&)GdMD*M#c9dv(k=&bnCyHY%H?m<`cp(S&;RrJ0J#Ur|?@r@7
z<rt**SuK<7d#n<qO^8k#1rv0%Vp%`OtnLsg!PXzeYfN$d(L3A>9*o*g8;sH;@k~RB
z?V0;CANM`Rb1{(&)pQR04AhY{?nrUxG^0GZy*RKe^eVk`GxaLC%Ym&;6Y%Ijr!T1o
z`Q{xC0QN^BeNvZTBdwYN@5MQ%p(t$6IS7Qs6t>M#LeTV!Pj`(7hm>&W0Tzj2rO|<w
z*Yn=;kCpNxC{6xABy!}>qedQ_0KYOmAOTl<l_E3k9m=n*vlK<W$QVLWEptd5{;*ez
z_E8{?JDLjKQ~7*HoBq))U(V`(m-kg!Jrht-iQyb+!@G%Mh1DTXT%O1u$k^@<uX{&b
zvDXp~kt@*hTWkMopZ0e{(UqpxD=Tw(^T{ogiqT(wwWwHjp`CU;Y^<UB@T_JX^~B*3
z`{E^HfbQz`0D=3%J!~$s*aAB7DO3;LxTY(XuBP5ctzd(*foEXTu2rXvWx+U91imLp
z4uf1{O|4x>lf?HLygl?wS5_m};e*zCw%4XULdR}vCVw3ZjypkpC-v|qf9sXF*m!N}
zm~jpHF8Ju$<<f!KRzGqjWMYgeXZ>BTg!8cz!&-&tl3ayj$$<bjH0$j7H#xkbvxhPb
zMQ|wTL>P$MPFTS~T;=}Zl3G(spQrHo!UO5P?hMzCMt!BQ4@RdofwqMDJrde@AVL;k
zvj$*f`5mTkCO%Oo9in_>i8X{MwU)i=CXv4AsDTGdKqEyz+*TpkQHa9dA454r<a?HC
zP)E~~B3Px&BKUiUSvSE)XB%Ac;q7@w$NZ#$6qlffa|ul9(qS>!oG4LY$8U}a+5;dd
zzrrX&+Lul+mAxO{2zA-Sb0?wf{>)3Pa|LKFPCthF5a&n&!ocU=OK;xTq`|2L*#{FQ
z9+KktXvMxX_E=p~XuRHCxEWFENGIGj_+TnyVQVH^z<?g{U4e$9<BqTub0cS^xP*^v
zYFUS;+hB668+Xlph@MsOKwsJI_s0?1J<pnvv!iy-W};~UrC#tSUZn*|t@Ha&n&}zC
zAA%)UJ!!E?+<TNN%Q&=~o?EN7{rTI+6>xXjhgfx0&RQ$C0iHlsdYs{%o3FOgO?{Q3
z#l;?P%>@cod$jFqx6=6jLT@XD6wmx&tWsR|dG;|FQ*)Vk%+=ggMN^sf!eCc{v$8Oj
z>M_rlTKDx-`*;luONg=nb8s)7uWqVp(-T!AvR1E{r;>;3*I5cNY3jFWN2U99XDfFn
zS?^eC=e?7j`OEu<%@prcX6w5ACGO1)qJ*0Kks@Z^mZTIdcQ)Mol?PEw)84@;7Vq1G
za!vY9<6)ZmB1wCk!msp5^1<X&X`0Y{{ZF_NN!yX+Z^M+16N$AynG?8ZYv~Gi3;K@l
zA}>%wZN)G6Gx~-79g%{mtyrVq@V$rT^0##L_Cx#Gda^e-Vb!DW-Uwy0FO$6b*cNo{
zR%bqtr0c6LcY+*4-(&y8W!?zEvD4hgr#VT{d1gUM+k8j(9KMEc9P4F<s}v5Li$+E?
zvNg01bDf#~lyyBCNVwz=xI@$*e8kPi;T`TezDRBMl5W?ZV8~9cJT2nCqA|7*XNbsH
zEGucyJ>lYcK^v)CC|_iABoHXHE_?9M>Eef{3Khf%FYH5w5stO=n;%Sl^iqPHFw&!H
zTok~z1o8n{H9gqp+a2P>(%QVpq$756^00tn{-GQ$ftZfvrSF86iRg`s#yMt5nXU1<
zl_-%zMS_{jr%OY`+N0f;4*5GEy2{eAYLF6D+Of^##xO`p8f&B^|9L(_dNa-rY&J{f
z&700quU_EF3oH|5{w29X*EVYhEhb$5yP%Q07*w;E!goznSV<5%u-2WbI9pl5B2-87
zY3XcmF|O^u3tKOGek0w8Pj#8&$#O22YAXj{<0m3HdjfJ(&(e>$6bq-_bS1wY1DCO`
zfmX1!tC@m8nMvIx<4n`sk2jD_aQRM9FdtLl`O)(0Y%NiYA)49kb|G~!UbNb<H6Z+>
z#9gvu!5VsuVQP%0H|0XZt;a@x+$nS{U5Fm8<4R#BONg(E&|~hb<Ez_OX)zY@_4zuV
z{WWVnd~b+E^z}<4wBPfM^lqJ^3J3I^;90Cqy5HL1&f;=tLnSLsfpnpEmR5(lgX^C<
zif*3xnR6eGF%X6ZjDu$pzT)-IeFWB`x+^iqkj5JCwuU~RZl>&u$!qc6StmEqbUKM_
zw%6Y{GClNv+)H+j-|m_v^qSh1Q5-oFqvdn6Al@OPup1zHH1nlTa_ab(e0@qX<F^Y{
z^P)8(ch?^vX-=3q7niggduhlCLSOhH^q<f|`65i84wt`foH6)XALl`r|K7}b*MU*h
z9=g5F`_F$Dv{e&)rv+4TX>lQNH<Rf$f=R1U>!g!t_naU(jCK908eXlbzm%k*V~9ZD
zCeai@em>m@NQo0GfiIwnV|MjKclzlO6yx=!XeezJuMDQSPK-skxLtW3b=U!~Y{xDf
zS7h2Cuju@51t0FJYffoq#ZFcT+%x0vxrJ6xErkqUn~o(Os}JP}>2Q9c(Ruhlff)5X
zqX>e8w^#KOlOv+HC>%razKaLZV_Qu#re(aWp<DqIi{aRJg(cdH#X?~M+pJ)&2OS4z
zeRXyg_`7iSS*Ik4#$ymqh!Kzk6<=<!2b7QnOg>uhb%d31_FWisZX1{N)xEuhKCu+w
zHp$cKz78?`A{7`Y;Wc{QzH1Rvx0c`?^Ird|J%2Kdo0!HOC1w}<vr3xHC4ZuY-x0zS
zra`<bM^wub>mEK$8E1#wie*%J0E<YGR4|g1Z9@|D{!Gk1ytapQ^>L_d(0w@TDdh(h
zmkW(ix%>6#`p@8sM}gPg=*zFH3>~<B({HV#$Oejx9P<ot`^sC<T%VuzX4xNqxXzc7
zri)EG%bj_I>T5Z@k`-%9a<6H#;t{YeKsI@~i=P%G3~_8t6O!lz^bg#E_V$7t;3ueM
z?j(gh`2cP$J@2|QH0~RkgJ=@z!FzAP*{$@h0_$Y+^l1e5-CrEXJkwM^j+~c7%j&N3
zKaD$<`B1YSacv_41XQ&@`pK97^dw-vQ8Xt-J~~Hm8u*cy_WRxe+2FEWk5%R4R{c}@
z)f&kl*~SK?d3w7t4r-D8^aFl21I?4GR_59@Jo<`O`Yx+R*VemqssfjgLNYH5rLtT;
ze80>Wi}p$~c2FXVfcm_|G#O5?2|+JE+pi;LjS_bKXN^IB0NeFAD0&oqKwe=5x)ZP|
z)gei$aq-KKdeC=$xLG^kmvkf>!Bn}!%kfL|8?sDliX0l>>e?kQ_Cd=$wNtR`58TDQ
zD_sscIIo?U1_3@%cXtmxZIPe~F<l%ndxsa{y8IV1;9)Of?;(ArEKQ}x*LeK07|S&3
z`CC9EP^~wu<&@!NZ?i;TIs%SwZR-y*b#ofqZlp-x>fvjY!wV|$l>+*c4{5#{0G*EJ
zAlhoY`JjBxPYs;mj^MbDyb{g4Zbl8MGkV)Yt2%rTfvo^p+_|j(&mxx6{)9=$B93AE
zR`|>3;Tu8zODNEyWHi2WdS)!PbaZvKKld(PM^Sg4<5^oI^-#4;5&P8vmkxOSTvXzJ
z7g{f=#rx6ELH-$ikmSOe5>tz;S1V{DB>IXWecTSmeZK#AYw|_6VIC%|n+*m7ZKre-
zExg&K?aqr>DYF9)`6{vyMv^@%cJVBks8F_G`N^$O6|q0tURf+`CW|o|N@N*c%c}$x
zroP>qn1^q4hOccF#7A(!oE$&1=&$$HGH+>=usI(jh35N`WLBcBqDK~Y^P(rBN+V+L
zW>xJA`?^TqS#F9Ib^FvA#XyF!ki%8Xp17Z2Li%>OcjRoYeqIo@$YC>1MAd$;lnD~~
zvUjQRi$PRy=91#z$FAbfGGAWyOF@rcgx8M*_YbUytm3K#<CXY7@J#p<B7QtiS$HK<
zfVao>wtk+*Zv>pEM*43c*?a3_ImT#x{F~*z;?(b4TS<Zt5qIme?YS_5{s$7%{kKvK
z@AIoV>FDbVD?^JOJsFHs;2zc95OcNn^ABp%4&3G-;3*vaFRXX^N<Jairk_fjY2%Y$
zkpz@wW`jmRxx^B@+r*BgW4FDUkNCNYaK}3P^2F))=|lX3WVbk26u;8i^njBse=PTD
zu1p1?;wDJhp@~rtiAmoBNU3w_VWyqMxBG$pRRP74r&$`CxIMv9GW=2(J3i^L*qq!J
zQ&)kXdC0V!>a9KO%Tc|Gxf~-h0{7-TGxD5vl<w-Wv=2B|I_qxA;A3eV3FrTu6PCES
zNO7SUQ}a$~t-G75!yyqY`oP6;gH!wiIC;u&XM!DfWEQZ9s6r%&)={)YRE|kI#esB!
z{qf`C4kb{=ig|VBrX7y<<K}xZQM^>s2&dh%Th=jFGsEt3`@Y7Ci>2nK4fcLOGmYzh
zEy=$>xatrHIZ#$sgr)^ZCMlkzyCx5`oLpnUKe>}2bIs$w3zz;GlwJ!MyF(0+WS}+-
zS~e~b3Ul#`fTHJQ^r7!Q@Rc`g2VlJK$IcO?wLd0T@_s4uJCuHpY{z4)CP1TH@$m-x
zYhO#wKbUtz`h$JPN7sHM%8E<$)h7=W?~tqz>w?~PBK)NR+}qpom`ANr-}soLFX@Aj
z<NVADEB7Fl{jpz=MaGg>znVysrJrNf3A#9j$??0O%H8fdl)QOZK#2ldI}Ufvn#0X?
zi+aUpJnHK6;JUS^t@ca-$rGc;&#_4h>Ui18nyX0-$b{s!Q$(O?_+qm9g?OQ#*Mgrl
z8jhb$#q5Atq{efNXD2peKT^{e-f&D@T}l+Z097I0{C5tCGm2Zd|F_hteJ!N}JjwJH
zzgwrubdvKsrsV1~@qU;!Ke?ea&i7kWCINRo`?|A`5T$&7uV_$Fkiw@--?z*RE5xNT
z+jQrEnq!~0^hFWEq`0-h`AAi8s^o^;OfNmT!Y<R=)9||iz?|PZh)#GRGO>HlDegze
z9oTryt#;!+ow_qv&kF~?-qE?!JRgh-3v{=ytT{|(ak^>bzYCtRhbkG4$l0~%#I^Uy
zhLmli99{CUdh@GM_Z^ZzxSHWQ<F5&c5DRR(@n_XR_^sbPP;h4jUw@MVLT%E}Rk8gD
z?E5&B1HBaEc{DcjK(L1`oTJ|v=I$O<dU!L8ZtahU!}w^a+)vHT(JpA&kWdaa0pxq3
zus`KJW;~f8jUBEkpa>`<fEQFlA7~Oa*PxmebDZQjh>L5)p|C%r=6rB(^u4i>!irB1
zXZFnJ+!<^d?~QL4z~7WxNSWz%C^pK;CGIdeTtTzW3?pvc!Uku*Il8_%!BolsX=t%1
zY=|wy`jm}7eAWJ4_s6)1(zA<`CM8@J&Qi}i{3w<meBa(|@ZT_z-KJd*gCq(M`ZSH*
zS+l%~<kOXyx{6Qqyu|%m_f3DyWJLG%P_E!x{9}@<f7r4~H|=9}HQC%T8#0gX2%XR`
zGOew&w(QDezP&9Yh?kL^Op3}(sxx$X<oy#@nct#)Ra#)V0iIR0!gyU`f{jxQokyFH
z*x|BZxHogC^zLyw-=e%z@mN!^OsZeQ=E#6g_iXj%S7rnBLtm1s=$s8rZc6PD${${`
zFGtETM|S8G?GyFQW4F2TcX!iC`;OMD)F9b1m|lSOo7Y>p=hj9YbU=5=)5<Uw=P;jT
zL$KzTKW5x?6I&nTYS}7%oTQv5(5-PFSF_F$FEZUQqPOJQ(lD8+`uET_f^W9HLa;gP
z`Jw{%p<rbYP3vQ)@tz#&ommDn5}x}s#s&(8ydWkx7m0+v?B*at8vq~|+ka9783;9t
zZ~T80QwqgV1RaFx)Jl#sPdb(^WD~@`1Ht<w@SXU7(;1-aYS_tZaq__Ce;SkDWGi-Y
zcj%Prb_;a6hp-=>#hszQW=@7BT2HZ0==7S3WKs~Rar1Eo>(R}}sT>kOSqx65O0#CB
zTDic=7mIWDY*52!UIdT1o1Hgf<d3!+37SAY(S&Ca1XAWuR$coRK)U46<j`>hp=~oW
zBl09UoFy!23%i)M(gHNKtkjXh5)wQfMya5+euW7HEHiIM<RGuk8nLP~zRs8$5|%EG
z$6?3Uz8e?aoy)n*=pZf7ru~`u;yyH0N_S|jolvf1>iO(;1jRUufxbX8es5L5;(?Kp
zFjY;xKv<KpmvXXH8{=m)YW8QMs0r$-MnPjsicnZ^ilC3wu^XG-M0YY-?OV`77-yNj
zHO`e}u(y~!)f}uTvZCRhl-v_`9c0F|eL5H0uFt&-{dp1ndeW`RPDjdbN;8G`e$n$m
z@sbG6n}51GG~C5LbxGAn_Far{Q=JMN<rf#74>p%W(4<W?--nIgP--KDYqD(lnl+?0
z2yM!>Ta0WN3v@-pN%!y0BIA15Nl*p-)}8$)@o@{sm?WzOhun*)MWac>Z{N6E3LC5{
zmt${!3h&v!#q{(cQE}7HC~3oNjA9Lz&7js>mZ1riEE@PG1kXB#;`Y?hYS>fWD|e-d
zx~&6+0)aE#i$q-KwHPR%rt^S7*6>`8Aw38*$lV7p9+u=LZukN^!^|-YA60a8Gm&%6
z7gXqY<H@F8zXiwBZy)E)<Q_FFY_%e1c~CY%r#W9V+@D%ZQ#b=%x*#$Uf^R6Obq!q=
zt@jB@ezP#ghUTu*M29lqS|?p8UJI6!;&Ba`J`&tJxwd+E?aRGxHkHiUL4uD(Ljulz
z5e|5$5)$g<a@J}X$odnD^ia*wDP+(u(QySp4FLkk@}d98_le?gxML_rMX*so5&(vr
zIu^X?_P}d(9Mqm}&Je;`Vz9?2oJBv{-gVDk`;Mz6yHcbuHorCkKakRI1@qtPPI?%r
zJP&!bFlGMZ`3u&(Cku1;yI&AKs8krrryb4`|L3k``S5!;(}x#p9@?v$-vycy*{Aq~
z+BrAe3z^~dY3fQ)Zj0b55uBy#oWYAnCF_K@p$FbgqVE}p6z;9|qt@)bpQzJ)P!x{G
z%m+cQ!2PIwP1Hla#_XMODD$6PK{8Q%62=m|oi4Y&K$!MVb@VCsI9=h1q^)jUKA!dO
zy__^QN%!ge(`4ue2kxgY4E(m#2AJQkdY-2IPam(LLjod?JYwp;G{COKmET0AjV%^7
zJJG%(OP9UtHEpW|#&=MQzHh8Pn>HCXA3g_mMDR0aEs9vQKo80x)j`8;N1Nc*leaD1
z6rJ?vNcz7ElI)Alp0Otd;MgzN)u`HsP(ix?gbrizryT3#us*l1dR^17Z|OMU9qji*
zZWD`R;c;!}vm4&nc`@QJcucI;&9r057D1<p1t{u64WpvNV{wp(OivLbDi7=UxvbJW
zb(OL`WhFXN0~O~b#z?PK5T{^R>!qKh$kohk8%&}zZHE4T9b{$(R^B=5U@||K**O6D
zc2P_El8Uyg{2L!N{SC!STVhjOky%W}tzDGn{PzR%h6>DOZebLe*M9W0`KkQmxH(ZZ
zPjIw;){o{_ac8~v=-o0a%f6YPP$dHCV;WG?21R2Ohv)6QP**G08*eBc@X7~Q+9T=G
z5dreJn%Nw0d*Yy#r+(?}kA^Ya;@BmZDbFl62L_q3QV5j}+g%Z?k-TlVCdx-4U#a!)
zrLVZJL0R?-G?`Ja1Jy=LpRS!TyF{)Ocz57PXV<5k(jA*0txu1gUET|CD#q<U$Q-G)
z=|YGY$bWG}35H{BNiwZH3qy5x$l<%gKC$MSomr3Y@4TOw@tDT_Z^Q}Z$b2bKFnuIT
z6<P01cTP7pTqE1b0#^YjnZ~HV_ML@7G*#ObNT)>?KH-j$u|v}kIeaql7r`;6Pc}52
zb;C6mADEK2Z&JFghHM(|3x_`FH3}TfYDE07TuggjLNrCBhDs#xL~{_U#BQL;|7$`1
z5$rk54f*c^!djY;u-Z>|aZFsI`IP<0xwN2;xO?X5R+xtKTViX!{)=@eu7i)}^y-1@
zPevOLXI-tOij!+!Dh@>KUmq*KHVr{I;hEkhOdO!E)B9}2`|<NVCz>F4=?cmwr$fQ$
zKZiCk0L7OlbP!q@-UGCofEM?5tO%ziNPdufUyBuFa6ldt{7H|W53V+f6V8`~s1$b2
zPXN+nN~KP|8Ahw7x6S*C3}esZcH>6ybLH9|P#Vgt$w>Z4)S$j=-2#ZA*393HVC&i#
z>ShvN63|AQ*q}o+w>J8}*sJFep<}xDcHx^3C8mXc2`FLy_w}KfzP{*-=48XmJISJ3
zJ_$ID%*s`_;V1E@Om76Xu=g9WK?(y*{dJcN59B3TDxBqmV?CZ<{Hnw@NNmhO|Le6x
z7_m0^gew;ps@Ix-N0OQGXqZ~M2OsM+eu5T{<Zqs7+FejMrL_{a5Hv%{7t4sp3E{Sq
zU8a3Y4cccthORp{T((^sA6I&yUy`6J%(qM9&E{<>GoG4mwTGnH*KT3<<)0o;B}WiI
zZu|4DLLslsVeN*^ZBhKJXd=_StUYua)s%+o8!wQ+oJM%eGF+GQGBO+;X7rTws$#&1
zb<aV6%zbVy$34Gz-=%U%ff{CSnE&l9e>6L&A-#%`*Zd|EXsT^B?(C*Du0A}+cRy(5
z2+*=(j%iw5+sx#=`rzp3S@+^iXF#t3G0uMin;?fiY5!fotM|E^u-?ef4~5TsY(9c7
zn+~{x@si~AHZLM%j19ydK>a!`DMG}?Q~GQ;9FDSh;=648ls3HSLvId|;LW~~y!&%z
zb{FJHx(P)b$84c2p!5i&zRzs*;!NFOxZ^w#dpQYkpMxr*eb;u+v~47FeTG7wlA&KI
zh4=2kN%m*3wlkPg!xNdD5NRm!U&D3QJ1`R8V^!zb8*(3uD%9x-Z+lS3o~?0l0<<>=
zOWtILy-z;kCWH3Xm*Kysc@!aS?ysb2;5)04-ikvY|9>$p0LBFzBS~OKq&HT|dD{|>
zC`vI)<u8}HH(zs2In`oyxTr{%*ZTP$|7VlttR)v)C-Ebh+Y;3@&8BYANfQm5iV1@~
zBfdSqPjfr`ux|hJ6Su%tx#d_N-ak$8q^?>9zd^~<4N=uPqAQa{Eq%uc>_L)s>&Sil
z0-vV_QN`yQvV^ND{uo*?=6Rr)U<E<xFuw2*rosqFS^tjHq+@9-9gC#BGSwXl`y61?
zuMgp6NPKM|nV`1dzxW=5sWtSxTsm&vB)&A6lEqB@{;I*$WU)QE<N?&DVf<P~k}2Bz
zv2@IA!Br$f+3Nab6Z!T9QZM7uiAl5#xEc*rmVfmwpGvK?Du(%TgJF>Nl5gb%dOWg<
za-*ePe3KZIJtlSI#dW$DiT;f~h~H;goru1Y@2B<!=^EQSjd$5+dKorawn>UVBz#Zk
z_cUOK&dnCV9Af0BcW=xe1o9E(e;tVk6Z>LJk%WY)BW<S}$L=8(Bd6E>*BLs_lK#6u
zN#6fNvVVG7SUg#)1uev^7hOIWUI^9FZ-l`};q2gPAyF#Z1@G&Jjah@sf6&-L)K8=H
z5Iiiz#<OEFl$iybfAv;x71>^Xb^|5kgE;~MaIim?;wP*174msF8`a{D3>U|o-S;&-
z=LOf(Aw<))eb0=<F`LJss2~IUHHsTSmG%$&0erEr)eX;~Loya~rU}>RHv~gcsXRdK
zHU=bjK|WCsU;skI%(oTp6vVU+!ks0~v1phGzuaCD&S#$dq{~>XZzbp7Q5cL1jnatG
zbUmgtR5dHTlY4-v_<G9xB&&X%eD)kql;^kcS{~b#JJa<Q&YY&O=;*zP+6)&jk=Qi<
zzC#X_5p;0}v5w*ruC?#S^gEH43ddHSD$X>oV<~scThIO7s>l)_7_$@yF>pm=>4t**
zG%M3<za<70euhEn_<IX>9^Z*}mv4uZfnu5Cz<!`F#X|o0Y^1V*-C+M#OQ$V^aDrNj
zZg8niA6n3LvBpe>0kVDu6`UpdwB+?|HavP?w{JRlHdydpgm(}jsS{7W)$Z@<x2gKq
zkVY{ReWVzVD!6h*-e9eI8{u0vOK<xgr)gbeI5Id(l~gRy*KjI`&X@5$<h@IpSsyy)
z`(N$q?0*+n;@_Vp`5()5=zwtgKN2AxQb<=_@X#jFw4UmGJ&OA_9>qJV#rA(3oqIgf
z{r|@MDsqfDpT(TTB2f&t!<<<p3R%pt916Fbca!rl#}wf<$C8{+i8-Ic=&(8D&}3?a
z)*NQLf1iH)+v9=9KHK~CeqGn~yu?qG+UW{idu>uYxum>UfwjC?nZpFuT^Du_Hb7P{
zoy2$*3gWm=|4a|5#KRiSjEszwhDFnDX)h?Q0>K&{$%XgV8Htu@iHrtOquJ?}r7ap}
zlvzcb&vrZ-AD>44{W(~9_0m}GOvYQ-UeZZQjF=dAnz{D)tfx|yVwy@gj$^OPs^Qaf
zE$<{n;9%m|t3DOD4b%-dX_KjH*j_P8W(<gU&YW9F1m=UvN+$uEQ1Q?aq`Y?N>7DB_
z>>ndu#olBsP3|}zlzOw3m*4e}CZz-HRE)iAN<*a59~hAACk5()i9RJ_Jt&i9vvmlu
z334UW{NDELiE@_W&By#6b@(9#g=Zu07fy7OA7v{Vyj^<Kl=<8X`S)zRt%UNkXQSay
zDK>OuLMh?vgl|*HqZq-C(Mm?T$IYzq)29Od?sO$}l97qKqR6AKp{_S}{-;D9OY~PF
zj(5W^&(o!^2IeF2HvfD}Pm}(H0g@~gGJ`1JP4Y9*An~yoHd_V3Jf0t8?|Hx4?8Qi&
zenRgbgdLD#$Zc<%PlaD^ZO%>|<H#*Pf^29$;z~)rS!sZ%Yk6zyl(!D^D70lH<yj-`
z3E`21cTr%164`&QZb~)esRD<~y^YJx*D9lYU~3Eag*Qt2^BBSEeFA$v-+43~`Q57e
zQi&Pek~|X4|9X<un>t*`{K1e0gQIdkB%(PCOo=61m`DIbGMg{B)=37sn>HC_6*_yC
zfo{@R1+vNwxjK$DtN)he;CWn>v*$ieUf$G}HYqJrgq+pj+?GzQsjbNXiW!bgoxcOU
z!+S9Zvxh({0A15VCXgyn0<Tn@#p{O<MNX3_#5KQkvY|XiOFzdWY}}ygM}-TMQ>gz<
z9^D`p{@<BA;OwlRX+d-GQgVHloxNwhy}X#>QQ~(irE(*jZfybX&HPK5Zzd_30GcGt
zB*5B{$-xu^%No6}hV9+Gm_4f{ehhsbFbenlKzSobm(Vi}QQ<A{83)|1JEe`svh6j)
zKRv>|c}8;{-sw0xn!W&6A5Y5QeJ7Sbo|q16x$$?TRZ~gL)-~az#RnpK+PSr@>IF4=
z4PFcP<TI^UUZiQwuk*$!q`ezU)-$UcNY)=vT;j+Yd8HPdvvPC~hHGZg5lqic&j@_h
zR8^F?c>R`EnC6|Pmj+fV$qPj*goCONuTNy|%Xip%^fb{-GWzo*T}l@o=}_dfJNfe9
zAbD9idHGAtT>PZOPf_Ru9Z_LdB8LDa$1RhlSN^5?mzxfQCvLj$JhucPy5dxH^3fh?
z3y3Oe7@e0<l#~3PZMO(P%ZzT%evLNEg4R?Dw`*2=nt#=8+01D0j~Wi<G;zO&WUYJf
z&&QW9?#+z6qad*z4<0Z$PYd8%=F1{1w;uqLLP9XY{BuPocNo#jI!n&N)u_MiYeXu7
zp12}M*E2JXI59i)jpC)jHp;PXC_xhfq)kd0Co>Q~T=LXm6M!BAvxbW@T-nTw_a8>?
zCXc=vH3~b$7%aJNR~jenN1jc=!GILDI_?0<OI)E#7p}$0hP6`d4(KAh6vc<*+oeB`
zWso$}Tws3j`31#BN?BCHedGARo3jz{a^eQCgO0Y$`tMBD9wVEou#&b@mRW{zp(GEg
zx?`PZkHsURc)<+5E8j6@2$OpyEj<ulpM!C_as!`sk0vojC<>B$n+h?D;~-#~B|HAg
zBkzF(ZoRQ~xWdvoQ8qa_eY!RDW6P#q-bc^@@jnniU~>Xqjzu=Vs#Wq%UrD6x`u!7?
z0vhz`gGC@^z32lL-Q~n@s1^l#P&@+oThCT|5Eepnj;8=~V#Lg#Ul#idSycH3!M{+*
zJ#gcO?e``vLq*ElU=5c6j}bK;>vT7`C-!*LOh;d#QE^i?k@X+ac##~^=3#}7b(dY#
zCkwMt#b0LId<su|=F85R5f#|Rq;+Z9rFM!*n*6`9NUO>G2d#0s4hQn61Hl3L-gwr^
zQ{`x*;r7S{l>)>C*uIJM38F~PDDjjJzLAt4<rG&@w?zrK_wSd>mlO7&EJA^ZULmK>
zi;KkuYt-=I?A=TJ8FmL@_=(}u?g%7Zd2hd_&jN|_)@4qX%n}VVBW3uR7vyK>)x5{b
zbSP@Wec+L-P$xt9Mno&wro{6qNwygvNF=4YmLi6Mhr-@bxo4#h9qh58=XudGXL4g|
zkoRObDt_YyuEpj2zh$f*XMY-y_MV4$PtDCDCaB;Yfo74=W8fW=r(V_-R9$IRuUi^l
z^7b*HI#)%&qe1$bgI?4)0X(!pBMTI63>%1sMXOjlJ7#U?X<d;I#13!ur5gQ@O|r~R
zW&3%kfp%S8-7lWVMlv<9Mol_t-{M~1SgU*<d;42p=*WD2d>4W#SYS1ii&2<K?J5&e
zX_d0KSfgUm39;FD!@&H_XA%4nQa82}d!@&NmT#`4QDWHlqUE3!jRjHKGOR$vx0Q#^
zMsry!^~mG8WL2{HYsne$Tgok~tFZh0-z8gLUOR%ubaN9<ruc(LHmjDNeIgk?aVr^r
zb*t%=l^8-jCp)(v30l~G*PAFvDZEc5WDHNJE=q=*hp#VnsaUk5e8W@TJ4{2Zjv7rh
z3?)u=2&S$<2U~0HxZGGm+bZd|tI;?!gcmsDhhR^ZzuFBWp3M8%253c{rkT%oK46SV
zB-*B)ClXJ`gMpD|OuaOCRyLm?K(qPp%s^oW221ZgfxKfBtkR`T+>#0Ww|$EiP+qr2
zD77uXPb6V*bLD%ot;+?piCDfaGg+Y6*l-owWm^%;TOs402_s|^q3Lr&3Ac#SY?912
z7ksIhgklk%=7Y6NleaWx6PPkZsQs`ap(-27EuKJp&}5f+i02n#+3zG_&h0yZt|X3U
zce;oD0w-g-gU_wVMq^|6`8+3;CZUxZc3DS^Sq{N%&>?aXI<&G)6_9bb4lAP^dgtOn
zfA%qmG~m1e7ITCK%2*X%44U)Dd?EGSF51Ap!tPlX-XnY9TTCYKWn?oEsj`l%9B{$U
z>v<NLHbpfnrR{;PnOaypzn_c$t{d3)E4Iws(07veQJ8MUSEc?U_Bi%O4^yVxWp#6B
z)eF9qmNGTYg*fNjurV(@xTcU8z^hlUZa29#u#lZbELt?+Z@>@8?GQw<&viFNF>$D~
z#n;mAL2)#rC%}yvK}*$azi%G&2Dfy5MrSY)-=F-T_)VSBP=B8aJpahxIzCnyvw;)B
z$3~sn#JS9nqH*Z!^Y~NI|IX-&<DUL^<~mlCib#-fUGCv{QKd1U7OgbINRJzCs2U`4
zu7lhPJm5!al&F<vg(XJ*0)JPPO;xywz6{cvCbaBs9q9-O_XY2?!~m<lD$j=i0X6P4
zB(X0uzrsDM+uv$3e)@V1kOc;?snJ{qImev87y>KZo6in#-ch`%My9}>vHy((6%j%#
zN;GISbfO^Uj^nF*dg6h;hh34=^>8CI^*p>TaM9@#V;_er1JI^<Iym;qjED)s%YzbG
zPfiedx7$b=^M;44MHk$OD>Lbv?z;mV?qvA6Jrne&1_8Iq&mP!HRtZ_qzN+$sbW?GT
zF~6#r0P(B`=Fx|dxO{<gJ48D?U#HrOzlJ>LVsCsm!70-K|JavKsLh6UXcSvj_^hx>
z)0+uq#>!LDLFw2W{9*o-pQ)w>4^45Mmjy&i8x`Iu8wAWhSH6o~nO~sLB`FVsGyq*%
zM!9iy-yo&4LAh~`Y+VT-&xkIR4JSQ9El*zB{>>a@p%^IFG+fG}CCPuBIhPvfz0crG
z4|Q_=iK2MM<37DW$2U@JCr86%mP7(B?~`xelDD=~$k#OpnJN{=u;4=)7E(uJi-yXR
z`_qpUjRuMvM}IZovmCiLsE7?*(RuLE9%SH@yrXj`#ApKDyv?0FOS!sPIdNjPkANA%
zd&Th5PU#bv2jC9o?EeIk6yKWh6kVe8YQUX=%EXqw3{9?xbil!8#{kh7=S`(`DZk+Y
zR2LO$n)1&>&H9K{;zr+e_5}M50=DOIXPs$7^{BeKbfUhCplXJx?c|)p?a7A6XdOp8
z76EWAYXo=W__ztR;Vkn*l;-jh9U8>QjTF+cidL#}h$F{`z?6H%z-k*>#%ut?PXvaC
z0V}4!=WBldv`lX183P66TJ3y!mPd|Qb|?s8?yePXpb!zI72|avWh~$IeBFed4&KZS
z9ZyT0s|#2Z@5ToV#0pLYW_Am$;h04$Ahj^Cq2)UrrJ*B{Y6b_jaz*SBcyqi5w7~Ba
zScj=<XQ%T{<OXD;u^GVnN&<Kth0^1dnoD;Qg(JR~?_Ga+6y2F916Edg9}u->Y#aF7
ze|AK8zOL!AWXZ)hw2ps&-*@O#NzluU5;E6Q^wIY^soLSW@dnj3bz3!5^FvbD^=kXP
zOC8`#OZj3w|DBopa@;Fi7p+hd&lB6YW^#l`Tv|i#ZGEGiqSwdXlTqjQ$_q~sYhRu0
zt|@IX1E=@WC92GktxDr>)8~R(U=w$<`X(puYK@;LRW8O4H_UCgY=b&6AEN4s8}LB)
z9Rs1s`1IhRoe&D9yhUc0f9bo)t^dvps&{^u=oTSDh;9XRrHkcz;3+Cky*ge0GqIFS
zam055ViHfJL7qg6?77UXn#xBb3+}CY11`;fBH`kUG8}OONRSWihEVnpEt1@NN(0cg
zsJ7Zl4PG-h<w`?K;@x~0C#Eu<V=c4zgJZ!PK&?Ux+&6F51LHyS$}O}1i{<s|QOX6N
z4u7&%^%g)NYQaEE!WvDnw!R3|z&TjAd;0~SgqXZD!pkJ+=Jm96JI0)g5vcUJo0=#s
z4)IQ1_elTZf;1G(&@;&fm9qh{d1xyeC;BB!(Anh|5LOk0%iqnC1>IP@NWBthDE;4=
z<D<1ojIw2f?~dWQnG37Y*h|*VnDGo7gJAVn9jmvCF`iFPZU(8RsNvZ7t&}@dYwxcJ
zep!sU9h_^Ig_qQR56x1nlsUBNGVXcao4=gfZOijk`Sxe_XT;;)CiOmj%>JXw`GAhP
z+edpL^-C7DMi(p%?u@+$^aB3n7$)1JW0^_+KHl{fJZ`uTH3{rwvUw{sTPdv(Q5-R<
z&NRQd=Y{m5b;FqtqQSVz=DytJxc7Y%|DB;u_Mig#rE}etsOVQEls6NQz&G1Q<lMh3
zP|g3HxgVKyR|yaBFI|=x&kWIP<M~2-X@9`L4$|i;JtiV+TNxq~h2fKLEo#Xzjrh1!
z(rhP@>^2Bt^Vlj;@-^QqdJIj9czlIm7O1;c(W%12Nfp%E$*&NxsLF75sRADraQ(Y~
zpU=t3V%!n#f1ebWcEA^@s2C~`3|y>VXH}DCz4Yac*@SC=4=4?|MH}R38;HeSI8riw
z7k{>C#&*_xc;Vjjz*=kYyvJY}1s+(=5J^-rJAe^EphWB`fD5WQF^N@d1@w;541g$6
zbzmf0&6YDy8O7}$AqL_Fok&i5bD!$xaOZ4;iW8sqsJGc!dflrQm|sxK8~i#DFR*F&
z_~23X<d0jazqZynnEJv+>!bDSP0Yu)_7UmK@VnA4;<?@&&Ynm_f+z^DO=C&iJaj;6
zfgXKM2aaDWufQe5Pp-seMOXqeKA2(wp~<*V()%4yd$71d+Lgyh^F{BXH3Ml(EoE@i
zKL&FaDj)#B&dSc-%{_Owxi;tzB;Liwn(jx;Qc8^U`!~szuZ)wtR8F8cr~^BKOg(`(
z;bjpba~4E%CR=RuWz$a7%9jD;5AKkE&BDf!&luoeMf@Q_Sc>^zSFrlB2Ug!_DjeLB
zeDItv(6KmQ8};;PkO)6$9^79zDJKdd|Ed)JvCfk*i2&PA$R`{L4MH3md5JQGbZA=N
ztZy8~L8X7s<dKC{&=RAJ$R9c$YFR8K#AjZu2DJB-0S<XHzUY?3#zZYcDxlV>{E1M6
zI~Gisr=$fdmdx%RX@`=TcJbbFYp5rfq*ETQJLP~ijTa%bmv{vAX9{eM0&GygKN+vP
z7yVvoFt&X3R0cSAHq5PG;3}cGoSvMZLqwJNVQ5VJ*hG2YK3D`$HTl+Kh2DVNz_Pp*
z$vv;fY)^bS-|^_nQsMzH(%jOXM<PvZGkv1W|D8jujB%tXp+Tg>iLObo8OJo0H|;CX
zq8!z7rP6*fa-+d>N9o%IG_{>m^e5V^nf|RfsnPDx|3!`OX(X8t<$%@i)g&`>kW}S#
zO13k4I)*Ko2I)vgBu0v?X@{Jp#LPZ#e%NF+#L*?q=e`4Onw2G?T4Ife>oB_twY*5N
z+Lr2TdIDF}dzkiLNzmy-DK0y*!|lXOfK?Z+IViTy)qksx?Pp$6=rBkGhPW7`Rrupi
zvjkt#0-K>s^u5t?Q<nrKv+%Soo7-?j?gzHp-5fRSaeU>FO0ccD_Gq~;@Rg=(@B_u8
zrEn{jpQoxtOEm!Va2|?e8$w1#E&Gm`V0_CPRD%^8&3<mAEQBkn@~d0-iiU&+j6PHu
zvMrW3`jT&!p6LHfu_0FY#1>Q%z%D1?)F;3qoo|H|8RY^mgzPIeTI?-<-N&N$uD<p<
z8jP@$J-?5Vmwi>PIV=~+@7D~XgWdUFhNe-<FTHtwH{%2_qw+(`seeUE(1uGv57SOy
zJ2}er&21S)3h>kuCAK&z&7Cq{%C4pT;lpe<*&ar#G|CH_oq8(qu!L+sReJ~xh|9vU
zgfZCZpe3lQ6&@6h!L|U6+<+VeB7!(`fj6Vg&P{+6$G0*N&q*mKyVqbK2Y!={E!XUU
z(=VF_quA*@rUKO?{6{xO)b`XSO+#n*FO}#9r9(NOa!~MvH11+)<yyHOCG5M(?jl0|
z(_zPT;I{xpd$jSi%-EL`yD=^7nLyYB3^-?=Yr#icyc>v8xeHf_TAc%VT2XS8(VWp7
z0Ezwo%-8=n{0DGs*@_xwxKyi39C4~q-i)IEnHu7qzPer5U)S~V)|F!&ft=KajAHHY
zUcb3#kvfY@TZy0hdH6P7H%EBgZzDEq6!6*QTQRoRn0!mAm-)XD)r5MJsEbI{dE-1A
z_2z*^ExT)}!saJf^*o&?Jr!C>J=ty|ZkV7vx2#~<+t^T=x?i(EfaYuQ5c7NHf<@qK
z9E!2|OYIF&^T&>gpG_!e*-7vA6AiWDiO6|Z+0N$Mg4&rsy9al@GdtQI@y-8Wq0PCZ
zUHv@3rdyp-J>ILHKKgIB)Vn99RP1HhWa|NuTbERtDsmgO*B{?UKazbY;b(DG(l;}-
zaZR}o&7b+YlePR5k+>Z4p_M8o-#`DPxqn&U>Y=W9HG8L8vsSOr-~3az#}^><HAlbp
zxs>m9{e`OG(sj`f5qZZU2j%P63ftarLn7i6({IA(3X!jb1_|Q|F1`df>+_@|^)XFv
zZTl_QtmXx;$|vn=60SO>{M)iN%P$PuqgzfL27fXH%szWZ%nDA9?3tSw&YIoTRKTao
z40v#FAt42E1+N^s{5r{Y#NnADVmy22TmI#|!yraFR7*_`AcE~Pm2P(;T4-YX41xSO
zo+u+E9qQc$QqMUt4=?}oVZ^tu{gSPUvace_9qBA`XY6hfR^Jv*WTVm-Nyo?r_R`+;
znu!yYz!1P4D7cc@8C1qb{qnyJGaLVViIxfhEoe3Wa_Y>`a@n|?D+$1-alE{G$#nS_
zAF00=&IPDU5~o)+97v|7b+g`+JJs8~s*qN|WD_C3XP+pctZ8%9bZ>0%v_jiQRTUtY
zh$|}aUQQ^Rp4=@Y;>_KXezJC&A;KhZ8w$Zg2&3^6?)z(zE@aeii>$N|?yEB8<1|?m
z*O94gi*3`E_6JJl1AUyS@X>+<B$cY05EY5?MKlQ42Wnm{{)hopPb%{rAC=bq%_QHv
zu7j%}EqxMa4&&(06&Qf9H!J{7i=<vDLhT5?@{tgBTG6U#$GHiF$_0Czds~`ya5Z-^
z@~G5`UkVu1L7ykJp!-GyJ^w)b3*9wWetdhFmhSD>*UT0{0`5f~H};XE@gt9uKdq3g
zbW6Xm<rQ!}dFUPFI}v&<?S(s(H}uuRwva_Y1C*EjVq`?o2rkxN$$UUWE7#f8Wf&&O
z$i`*EQ55m_8!M@~oPP%ulYEZDj-=yROFQh^cO`y*7@5LR^t)N!$Dz6owbXpXZxqW{
z$|8Ar6GmT7-|%cxi=h}15QLZ4f{enWA06@I<J}I*F^Yz-9vL<Fm^b=STs$#CySmrX
zk;Lf~L)P4?LDL!~OxB4&P}R6<49_~WnWKde4eBfic`vb434Q7wBl4bWo$-^r+Enu6
zNCJtq&q<8GkskvLSr2DH@i?E*x#Jg4cj&S`Ov{HjCmZ5w72?cFTZ|8@uF0`9rN9hn
z3A8>w%-S=L!_GL6#CiXxR1t(rxRt^Hy^}_eemX`OB1LTaX1VlOvT`t00B<i>3gfxP
z_~YfWs7f0jhY=InJTfNLeqp(GgbgrPKFN{C%=S!Tf#XXSnk%Y-Eo&qC<pdrt{QR1)
zM5G~Gu#$O^HIv+3vIhl+{33rt<uf(30hdo<0ZtfbKGyo9xEt}zXR}kv@Fdys&nZ3+
zA2ju?_D@yVJF0ZMk9TwGchZ$68P7NZi^vbwaTT@a|6M=^C5QHkG~*)a6216eU8k8v
zY?ADOvXhV@)0+2>k9EYH9hvEZNk5UN>3z5VJCi8i1t<F&7BSG$IW!U7R9f)K0%`$V
z!s6a)Y8845|KP)#FQAL=`cLrJ>;7b(%bXD%Rb3Qq_w5`=4_ozGbJ~a#n7?|*(f4P#
z(!}KPlg`izMpjL}o8+v(c`?_xJSn@93&lci2omY;scPCNHaBpja1YlsZTayicsi<s
zi6;lDkCLU$17dzKs`UR?^QxYF|MVRE19Rf{DrMQsouke5_nhfy`0SA`Y9TLFF@%?H
zf|^@<FQDN5LWLC}Z`CAGRu8ViOd~}e6@;BI2l)b)Yms$fm_rH2%`GO|RyYGgLU{(O
z_-;wBHORn^b>habr@U*am8ZO?k{3KDvy$(;d;_mwoNs&Oj_P{EhWY1HxyB6pgDfkp
z*o)PZMc#Iy=^BqUI9zTLwhtQ24oS-<n>;uD-g7yr_dd~(*P!;zjLQbE;(0z(oYu7@
z)VP>(Jn#C5cXMgZE7F;`4IPq{))FO+Y$6%xSE+}z<txs;pyA-+Pw@j<+4|wIaNV!g
zzfsuBkusEB#Q5_bC?U5;#G1hYF-0ivvt`30b16f0@8BI8xkbwG)9i1<ek~5dX&0+i
zY@f+(KO>iEH&(gSSi;kc&jCbGU;WEYjs*@dA73CuIeGscP@2r`KkL^loC+}4W`r{k
zr7F%igsNjEV0HfE;~)<eL_V>6u!%{yqW-*aMXlL(Lxilfi!8kkiTEph5ZuBOgJN-!
z-SaP03(fvXiV*$QRugHfCp#(L#oKx$9x>hFwWRsi%78>}X9P6!7ivNpRmvK<e1~#c
z!a92AucjrmWV%m>x%&4?Qov#Q!s3-D(LWetEx)ne?*x9@5e}Wgeksq;VEf;av%BYV
z-&xRZhFK|)JRYpPO1@~PWj@kURc<DNq6ko&uZfJDQy8^L()YQxu*uzjcZD$I9-#C9
zj4!iXH4T_2UQ2DI1SbA~I!@UH+&;WsMwD)CXSUhncu9U|KLdtLI9)N0Tvpg=S2*F7
zl0N3;8rVn`gLXQ`i2j~NAxmd?Z!l3bqo>T@ZEt~%)%%r7jZow+f6hCjpXLlWBeNF*
zG)5lRQU**Vwo#uR?VH|FOndZ{56q)89lUMCqV-){b!Y}QZ(9=ovap?wW_&8Me{Ru8
z$uyXy>+)3cFTJ{YXhc}uYb=(Tcp(9N>Qcr8Ofti<lS^<EojLeg;X(bA;YtCTqDj>H
zh3bn9UvzAKY#NlPy~}UWX{4B>-|S0i@ID-Ho3K05!O0O%eA3U&==PL9ZU(24tu$mR
zGHBD!=Otv!qqWT8dLHl_*=cyw_(>4Wwie8!{(=y{rADSuOdp@-tsa9G<5R~N(i(un
z4E%{MaSh(w?j!R@<LxghB3si1R?_tty>d?cDV*<$s$5LxqBudlgzP1`0~kIVQ%O!z
z8s$s!o-e%b0GND$-cc&XH3MDc?H$iS1SL{6_a`I(0A|vu*gOC$mS*(<0GQrMvY<qO
z_UE_$gpEBk{lM@SA@YizWZ$R}>vhYHZ`}r&8s;{$gd%>rLJI<`4_SLK#6y2d0Q7hI
z!fe&QJxSPDFagNtbn(O><`y!h0x2CcZi&F%x}?A2%|?{LjQj!d8;9HVHSo3^fRvI5
zomGB%7!CjT#lPc+rpv&0SJQuIWCP;bt{M*?@d@Q)Dv#3=b$_4;xC(}(DEtDdYsKnF
z%QXb%_mh!kuS1g>*m2l_Kl-}CyCoNGvOKwDL37x-_#VHhW0+WtF#X&`NiEgQE^#w4
zDcuXScKQg7<KgUI=%>D0+`iyV{+YWqvy-a$_D|6$0|U_oU6Z@hvMJX7kz`lPqAT6Y
zby)dnx7hN{K92j{6}kW<>~XG8>#>cO@Wpl)-cU(47q9nshU3u);)T0CbUaUN_@X#V
z;X~v49ln@t|Mbt`l6NR`pD9K`Z1Bf0?{2ZTwzJYHzg|VD!>`_qm5Y3}>+Se=<Lc>!
zJ39G6jdC4delv=g6?cv&8nNGT+vj~YO-R;g<5HXq-H><fHzHxzz<E`}j>)_CQ&ZeF
zlJ9_wODN7Sc_wjBlepYYwu~s8FsjIx=JrqO8(D2JO)46&EvyQPKJ~xT_Z-c!%?>}O
zSQwE;$^CcsCG0vlLW!~(c=rKK!yMy47(UkpB4<w>T_&hA5|<!^r5;3Uvti1rIu#SR
ze=Q%g?7B`H8EuI`V0)N#wcQFoY5a7$w~htA{5*s2b0CcPL`nvx5+4a@XSeqFEQavr
zw%dR1WW^?YWtKg@+~K!?r?K%Vx9GtwJV07?!50na@kQ~<8DPAwd)s?pE)<fZLC1*7
zt|(1NDSHu{9!vwhKj7lsiG=^TyoOsi{oT~4wS!_we(ZCY`&cC=ZFe{k%J=n1PYe(S
zzuiQ}bDj>C6pgR8T$c`$M-FJDh2~1%I6A*Ey7Qt@6X&mXq(9vU*jR58hO;I3J04xS
zS3Q@$kv)oo(V$E7`vAtnT~4A6^@IB+CC)a^2>GvWpvu^uOPg-+Dy3HGUk8dX97j9F
zcz)Q(8aiGjTW#}pSbB%4yNtKHndQ}x=}zH3?I<}5j(0rv>D|*L_wi@5ic^fy(=_ax
zo=D*=%ht8tXHCdN-BJJ2hlZ?Y=E|{ak-w)*stuQ<hkohUEC|5-SDQJU$6%>eEqguz
zt<7u0bf8uND?fIkd*h!tlvl>sS)(Rp+WZN>(q3}(ikWY+-Qu;8Ab#{yt0TprbDw&j
zpFcT26>z(o|A|jItHGmiUFR!y>dBjtlMhYpE+f1*ll&-ViDV0X$s1>7LYd=m3Fc@P
zB|Y<FUtq;32Qk+2>uZj>$D$>H#tzD@aehq8S8MN{H75`W-PZF3p&<pU+s|m9#^ySv
z4-8fZLW*mTgi{>!$FV*gH0QVtX6o1i5WbLm+fdKv*(Um{<I4S6;yYFZ)h4owrl@13
zJ8;^N>w~DN-yD61_Aul|2punKwzSxF9LBM03q$H$U9NYcU_qWr-$Z0<1;EBzY}^jP
zV=;1pDL>EN$hQAgQ<8(>uY}A?&za~2wXAy?|I_|YYy+|LtNA{^r?z>6AjRs)=bG9i
z>>%-0*MsE`B)aSp<>^PN#YN&<oxL)`F^FcdV)$%+vn>0^iQer<$dZUD#U?4*{4Nk-
z*;V<5YyO)#{35;r!bSDNjj@}60x-!ywjB@{fn5fO-6UEzU(c~D6^jR`z@yvF<ra9L
zqDy4$4W42mU_zE{Q3%W+?+bolq%F4+zZ;+R(RdD?$G(}BWiby7*4ixk7NgdyvI{Qd
z>Q09`)}H$-62lP;B27#H?Q=jv4r7WibzzY#i_l)zT9qrgZd$S~xFV!K++qKd6osK=
zPftfM<=$P)za)oO+Ov32^jPz{FLTs!T_`r~AoYt`ob7%P*S;Z|lD5ptU|GVE+;LrH
z#$oYk2T9oyM!%m0rv78Zc_I;S5e14EBSl|$!*?h-yE*zT7}^{Hvoldki!3GvSxD5>
z4C56yg_Q%0S<ihMyy*ko<YTd^XqD~sZ%~l?+d)3btixs<_HoJFK|YjkZ4;YbEOGxh
z3$HA_DRf<}>nXV~R6BfT^m^JQZbw^D&`rT<n}n)int#fCNG@OdRWq}d5kd2*cCR!m
zvx1~*8Ltt`L8se)cgONV)}GDDWw&)pk(Uq^pP-;<RpH7yD_I?Db<UE%-4BL4Mp;z6
z#AM$s4hskER;TcoiEv$QAYfM{SY@X-OA>#oOf0hLWpDMr@V|HD3tjNyY^pAcknYj!
z2^HfF($Pv<>lWzx>PcBK<F-+p_kGW?20NABlWC?-<;Ogi2#Fv1)HM%IRP&>Q>V8)F
zM%bY%cEeuFY(!&GSsZzkF_0q7UEt+j(sZK(%P791U`j}*lNPFy)ynsDPYr)<_}}rq
zcI<ZmR{y|LxymZ?^nqOkh9KvCsQI`Ug|we!@W`CPCNVzek#OXuOO4K#&DxdJ%%ZTJ
z0bLH-$Z%t2yGSLlK=!h~vknFh^kn=KY3sK>dyd=eCQ*N?<?8C*Y5*Hql{@*008BQu
z8Ja&K5;&HRFZ4h846NC!7XS$4<aa>_*nMLVt3^Tp+~VKjEJm3+=P<>^Cn6@lf|a)`
zG3KJ*hwrcs+u2lFL>Y?o2C()tlcS#=F{v-S%@=meg2#)3ME5+KsFexImwVH`P|B<*
z*%MP{C*s{Qr3}%gXQyHtDy@Q>Cu$Qxs|Scj+23MnDdot4^11xhIP7f=*5tLitVzhb
z!YTCgW;F~6!@hskLfP~o{q~92F65t;rjF6rmQo<(0I;h|Og!#<QP>Z^s$MA6#}Y8N
z^h;YQgboHcH^bo;>}BjD*PyK7I*HeTmQxP8zU$?!Bg4T)r30Lr-H_|%HeYvy*kA&z
zI_qYZxH#JFtn0-V<^QD_2M>!@Rec%W3Fx_E*7g9aq>|_nE$k_9cGP;U_9>TYd4on5
zsXU#hCq6lM5xRCYJ(6<k4xGcGNYb{$_<#58p24n9sqrN~RBDd+t5|-sQ)pyY!KqAn
z2DD}xBz^Cr4u*)<+7&iyj=?Jv`8U)4E#6`GF3j_=7e>pG>%PL=czG_!F9j$}i67A3
zoN}6dn)UKI4Zc{vd>fDS;JjIS?b4FH{oh~L+7MioAA^3Cm)<LP6y54S#J39xE#^DQ
zcuhSBWZp2DSQ`<qYr~2kojiGhMPvoEQ`*gMzx3|;L9uazw7ALlq1~=;^ANAbFc*eb
z33*DLWCIXj@^5Kr)r)(!2~^tjaZXOsfnblSIwS?bJ!=*Zph0LcTnk*KJhjTd0wSMF
zX|cxfR+LTC;!YLwzj9y!E4bWwXxSpvVGbj@>L^Ayklrf!AotLz%Zv~HWV=Gwb}E-V
zee4f~y;bWv;A(CbNwQg&IZCnzIbO_txm`1E77k~jIv|iCPvf3Z>W%9wCsX<?K8-D_
zr6yin`b<9I+Qs^kt=DK4;R?MnepDM@h7|YT1F$5aO|GBP8Jp>w&)$`{j+~ZQru{yV
zozz=`p7qFB=S{*fg=m4_9D#2#yUFzHS60#^6dub3CjX#{^!MgcVvtg;=kZj(By*_3
zaq7e5KC`XU*-)|)mUG=-m-6?Ns*dC~uX);-l3jTrgZDtT6Na8>zUgY9P7FS}bXM-Q
zr9Se=z=Q87<wC^vu-f^=m=FQ@=%%<tN|Mij#gar%A@!cM^--elwgHg?4AliszSDZs
z@yw5{&I1&oe{p!j#@^w6J;koL+k}kFNU1zh&l0LSP2<V_NvOdKUt#U;@{Q!ntm<p+
zDir@rPU!7DB)J$l@tA1U<830w@dF<P^5o?rmZG${?NPo2$&V8^ujO_YE7<O+K~|vr
zKNNWgL}O7xZ>;lxxMwrjQAvViIi_Q+Gi8})@D0s9S#0x2t2pdtePub&f}QKB?Y{-<
zqRKU(OanIc%7(O;#ck-K3iD=5tNnT4aDjnzwv5IBSl4gYD;NxJPgppMsc_`nh8(c4
zu&~@61{lh=$a`Mp$r`KzK6%8~+3pSM7L}j1DV2rY%f~SpSa;6gFy+<U=g4C&3GaN0
z|J7E+xjx=CLd@k`q`=RL@K(y0kGb}VVf(E2L0q9!wzLm@{^?%Hi44;OcO=(0W4=v`
zW*tv$%DHdxM24Bv#p7N;EC-swYTnakMQARycti#@dz2w~_?T$G9D8DH!2Z^3;paNl
z;!|xiKk2}S(uUjSAje;-9a1(YL?_SbJN+`4J>icEO!%G_9>q8C&97_Uwr^UI@*s_l
z9@e<62uSIr@5{iWkrQ(-1IiPzlga?;V|?NOD(zWWNSE==x90>b^vH77D5m`?DJuKS
z)fJvv=7mV+S)xhFxs|MVehtVwvE2@Z;ITiOPSCngM3T1no?%)XY?86vh+j^mG+JG_
z-NT-{w%2B|DW?xD25kZEb{kTC3k^)%@&{K%gMUL>$Dt*ckrnpnD2bclPTj7`rIe|Z
zNX!{Ru(o@aoSf*zwFt!M^w?W%ShqK%JoJet<PuurZG%c%(b;F~P9kjJcbAM`!RPFs
zYbkaG4jfqB84rQV2j0Pp4upgkYUOt7z7?&QLJ&K56}?}QPjpXJunqU}`&5iPxwN~Y
z#z`JS$2=d3$v^f+?TPYOx)Gf5q=xj=#jBwF(tl^dct7A8q@8F1MW3+D(RWbG^hLwC
zzAI%$1X<F~_zO<Zdz}9_PBWXT+lC^|=S{XwZLzyX_)ldU81PNJ+0s|f(3%hV6$`wY
zK%Tj@6m=>p;RDqwe~0<W80%u*e56uoDohm9L*72>WrlbL?VGw9coc24qtexc-fX^!
zuyk7vLxDpyn@2hAHWAI+g>ONUg^yyLZ6-S%*}t*h>%$RTvoLhw!IEQJ1FBKuL_1|e
z!+b>p^QADtG+`~^WqX{+P<*mmT%sSMwZm{`uCoXIc4m<|;}}=;@55pSpfKata3=Kl
z%GyI^p|K>L1BTG-F1Uy;ZA40C-)o%93v>bsFsozbTJ_8w0N<U0y#opVK4Gn`$W8hz
zgaQ6}ijh6LY^*25>!6YD&W4S}j1(L}T0+QhBF@AXm@pq3+FH`$XL%27{>h^xspl<x
zzk}ebMg~sgY4!b;1o)SMaxsH6z;7n5z`tyE0GO&0q;M(uaaBPl3=m*Vx*7KL_AhqX
z@Vs@QYi464rHUh-X9O3)G9UU!7eZ09QZ(FC#Ih+`pdl7K4smw>CGFog%d?I8!V3(J
z;TEP;6`Z1ew1f#<^L#7Sr!_ul1{N7VS_XuVCM}*H@PcWkC`$dnR}(Tb8?)g`L}MJG
zXWKjmX54^e435>;d^K-7rmOzIAn6Mw#E*qo7;?jRRv=54rNK<~%4XVIn6?X1R}8OU
zGh`G+iQxyrrs#%h#k_CA99mXN33}=74h1RJ`gF+w0UUq)*o{2m@8}AAkM>C%`~3ZH
zEp<sJQjS3(ymdH=59#gr+XZ!vC*gt!vq%I*{C>W;ikfgkQczW0#H=~gE$Fr`$k!V+
zJn_{kFY7Rmx1+9S=<&#v`ui2a!}ZC7ks|V0VT5$K-ZlL51D*IO4<ci9VUqZ)|HX@w
zE5F;4yKReUhB{ZAl8PPgnUhf4k-{7NzI@Ba*w|}FznCl^zt6DlBeMKc*~$)o<nBCX
zrQJmr+hIg+o4E6+=U;uF(<E1SG)|LFbIX+M&9V+GQ+B0WFrL*m(sX5z$yU}G^<`-#
zBzTMUrV>>Xy&liLNV~*Y$^BZ@<4*8R%B};P23lTd@9YQ_p}@Y<6oA>SY!xpVpSXm<
zzGD)IfQhMVhXK=NILI87X7#oh9YFtRCS$NYboKnUm>4Vq<E3waUOr@Ez@~MYego?&
z6LBM<7bAK_v8x3Z1@M39=a<L6X)R_@HUW=m>A77$LN`dNygwU=ruL?V%F~qk(^TCB
zuv5y|jLGR-VDHRsr`cfG4!~`JG_f~N47ercV%~{Qx3rWt&?%&2k;m(-!IOSBhrbKv
z0g|nCHQ-XOIK<l#uQavy7+`c>jC*IS@WfUeX1RHnw5qfgeUG-8sA!)Ut1!H?>z6ZJ
zd4V-@zE-xrC-Bgy+3j8zzbjakWXzp8>+NFF4O(K9&K}GCWNWA;XHTA(1Uu}DXPye=
zFCZ={>@X(|n8R$na*-F?jPqOMWSenkk<q%OkphHWyt&5Ed#<MsWzzze&=>ZFU#AuT
zqv=%@Bt`4^oxZ4L3cSnpG7;Wi^?;-4UH0?P&ypF9ww$s$HZ@jz!`!e))cAM7&8mf5
z@(6<a>cssmGP+<{R*`c7PROW~t<-lcpz1f}8<}V-KOJN6a@#C7{La0!r^&yPCl_#_
zc;4pj)2N_DpGI?ozO&M|>+)}+^KLDN)z7n9y-x~uu2Bf0^O>GkVDFrKG`I{iF%Eln
zM>Ahn(b=d%?!->`tKOKD?t$VOjIVZho1eHA{s!&trb|^Hb?Qn!<#_szOfA2_RrN7e
zX~Xrd&gb$c!vCFl?%yfws$lN>Z3W`3*b#RzsYJniNaD32TgGeod2$SQ47+vJ%1XLC
zW=ZgVq0+hM9Zy}9ha>~C-D*|5Z9Xgo^o=}!BlCGC7Sv_Maf|znj9=Z&zdbSl$FybH
zUD{(8noSYKhG_V{3w(b_%Vl*?z|!U=5EV6@+Y<yD1hNidTB7*L7G^fU(&Viy{<{p{
zT+o1BZak_&Y0;nvAWU={d2PYJWNae8%u3PuBUFIL)lSF81~_oz5GHSxD-CjU5BGtQ
z5paS5Hm<jEKzp%|S>9hC0*K_Pz>$%75K!(^%1TpuN?h~u!0^ihcz9LSi&sN2S_Y}l
zs2{Y<U0`IZDcK2@TIC{Tj2hJ<gdA&;h$Lv*Tckr;SK`M{LGERvQZMs}t3)AB&Zm=*
zDhEVhz=2_f+`5=n8vL9=F3I9q_HzW%0Cc33cP!I%dgpEi9VNTfVvz^I0(E>%Jw>^=
zKqo>f`wIbT5H^Jv?i)UEk0_v{=B1R!qjk-=q9V;sH?29$2uyI{nmCo+9laED<r0WU
zzUf|*;_){#EnH)+1VPANQK$&XW719LC?ENM7p(L@6`fLbbD7IBOw9Ns>V5rs;R%TJ
z#nMWt2e}+!C%H=5b+J(JDYU*Opr^@8(jjJ?7!MOA#DZ<R=w8Ox{yX#NQtBWx>28Dk
zX>-hFaG*P?8Lhl!R*<*MAi=Zm>7h);g>&^gy}aDIRD(7Q82r5Uu$RlzrnmLJP|RAu
zajpoWPza=1D><fiQpF+RhPy-vB@BpnwHbWdzA^XIERy?IiXK-};6R?<hZo!(m5X$9
zBWj@R#ExP-ShWuVtPL;3LsvW+H>g`{ziWUPGYN0{5!xMIyJ$FE<%ToW9TwfIYUP&|
z(z$WU9i-Wtvr%FXYL#sE`qm-Hvd0r3+e)%_RpQNnb(0e={$fqfQ~F(b8ba!`TAMW^
z17!pzV5uu-$u5=vB;F^kFRjP>z(5R09a3FgBjCtgR3*W}wht7CGgAs>aMIvZL=0V}
zr`#`U3bi4>nLIxB<e2{`E?ojlu5U4Q1b5h1`(rbvfo$3l;2`15cneAF%QzKN`)M12
zH0}TEBM&?IjcBJF?)*zKbW_)h?zPWPHCdCt8&KvT;!lGFDfU@~rnEBW0tS6_JOg<+
z_uLXyaKBW#t)@6^fMk`0H_im@RsQ=4d9?z^nD3Dr{NzUo=tN;(V06GwXCJLWy{#x2
zj?Xj?l8&9CwT{6%SuuW^Yf$zvb_Xobwk)LuS0ct13|PAYa-R<H&-+<J+dU1r?F1XK
zZMPN~St3B<-e9OVAJX(gqy(_+2<eK}jR@T}o`Tva9$&7HnR5hvc$4iFDj2_+wWo4h
z!iDAI=xAVlv(b6E-fBjX?sI30Dmz(^vh(#_U)~i)Sn+ds;6aY$mSgF9)hBY|{C1k_
zutCugIaduMw%i??q{>DXNSzEN<KLB!@Ah<r6jF|)Pc(U*|4DvT>>QM7jpul=)#}xe
z_a;U>?XBs8st;%XdjwEQ*wpM_RF;y&X;t;*Tafa!$BmuvQ6PtNz$ubre9M_HcFGrR
zMZC(~w465GQDPjpV%p;+?o36iCHqNLkj0H)1lrq9Y3AtlOL`-zgIVbxI(*Mm!HGP*
zo+o@y^Jbqsv%a#t*m5mIUxt91SPNjXy??EatBC9Ec-#Le;0?kUY%PmJAMaEjLGnL`
z`mG%<lwQ|AYJjY7k($(OY}Z58Ur$;esu^pQki2ZQ%94;;S?RsQ4m0DIJ1okmSMQ@)
z3Ew6o@1K%C-odN#R22t3bnJwa8;~~l-P3lod{E%v(GHL6{AiiPSY3$MYUIEkUzu&h
z6fwuiSW>!JJR+sHasr^u3>d&X;1}~lZ!Rm4<jbVa|AXxO4o)hE9iB>W`p1h0Rt`@t
zUc~l%0@{NcfMy$U$HMynfY$`vt3t^}Z!<Fj(G?xVL+|qmPi>3XpRgqHU-T#WZ^n-;
z8xGXk#tq~xUAPjDrmTAT?m(J#F%$~({NsIo-N15gtBq+E4Yamn(N`#6q6NF<T9QGV
zXZ^guZNN40(b}O|FJ_Ku1>y@948(Srdnb8&2PSgJ%1Y)vKj(Uv1D|Ce%;ru14BUaW
zk1T6dqyx&<3BU?TlDEnb5S1%B&n`rjck>>2n#HBx&al^hk(npx_*~k<v?-rGCn>E6
ztD?qc{9Ez)xtYWdIcrN}{_@Frkh9(!AN{{ny|K&8j~TcZZ}m8qqu+P*iR#>Hf{9Wt
z2Nr%-xKH|~+O7L|H*1TzBxHl&(N*HNFC8j1ALRw~g$I*CR5?d}%jHVKy-Oc?a?R|&
zHebv7HxSkQtlRqKppCcsb>R)Jo&ii<o*5)i{*vU5!P`gcYlP&Ntr^MgLGo~ux9v!i
z`&d1z7$~jADx%(8;t8|)zVdFRDo3THsIt-C&w*H(sVwP}yVB@eeVAYFqzboN0_T>@
zH@+yBDbB%!HQL6&oOV4E5AYQ4U>6oEem8}4H_7?+Dn?o>A~e{`Kpi)Lmxn%Zv2Das
z@n!hjGQg;6C`OK!G5!9hi&3AE^I@Hz41nQL=X&es)qp_Syw<uLONfdLP1Y!!zFn41
zw!#BG=<soJpY~npMOmI0Q+ps22Tl8>`4aA{K8>S;j%+Bk%JAi=+LAyI<G(p1LGueh
zfDPFyl}o^P`+ucF2aeA#%aqd|B@$39x^g|;Gz4m^qVKld>K;VyoO>dI<-44_U}H<)
z)bTFRVh2NNWEGKexlgyt6K#vkKFR}+hdGFn^52;agGTZX+Xwv(+j-w3gaF}<k~UyH
zO1wu5Pm;C1dHz#V(!~mnDDeZwg8GUh$#W^I2^|OpOW2duI#h1WdPT|0+RGDyJ>w2l
z3w5ST-u3)8Y9%~@z7@NAjDuBJ(2wztcXaS*q>Sq`{*IEHEYe@sqE)JdLOb)Wn6D6O
z1{xNmw?z5Yb1Ww;a2`A2_<^<8gp4NA5a;rVxtA@F7O{FGF`mK{RfJ&&nwXN?7cN-5
zS}x+Ymb-1^Ry7RNME;naAb+ag3swK`%%}}mtRu_I=u!;NILutvcm9pHwf({)RZUfZ
z1R-;WUuzZHKyrLupR1RzrSLe=+d~Xwn{iEs<d#<2BblX;P(KLX1|9u>2b%rx<J4N~
zNCO-4ULjGea8qt6!w@+G4K%mzl1{@UvIAAlcLAd(*}pSD+O!{v8~BKV?Tv)_L@*}s
z(odQl{_5!_1s(!sH%ZGqgt?yT#kxev3To=S(!`E<#3lEe*h?tbW1r;Cs~Z|SCn8t8
zzQJko>3jSM<`aU|KznqC<Bz-tM$&)of4Kiyi%jY$<;j81`mcUXJXK1uSb%buv2>?4
zgY=Y44Ds>jOF&DssH>Tg5;u`^3{JX)4`fKuHt)N6(sMJUnVcM@de3{Dam7H3ub(b{
zHw)XSgog?qK`Ih8Zp6(U2y7jxj+~e($MWtkJj*(VjV=w~Hm>>dQ27(E50x(ogREr*
z!`1>@FrO{nS|R3AuFR^W8+Q?E$zN7Zq(_n;1jjuELz<s{D~Qq6wy{XQof{c`B2|<9
zGuKgXZduiMK$AzvGW3Cx!Lvi*>3#{NUzK>?`%>Xo8={Kl(4IzBpS`BTOF=dH9L$Zf
zpdnL)V4E99yzVP$0^jei_N+%m8^}AtXiKAEY77@D)o4A`oAATLd1L;^gxq^M)#Har
zyO0wJ&aVM+k^uWZDDKk0t^r8))8PPLf}%6Y<I?jN6|XyPCTe&aEGdj<nAayhrAeXA
z-7-s`9q%-K$G(?8Of54x_G>N&k55EF&!_XPp|JrSU7AFPb2*s*(IRD<fAd223^}J_
zSZS!%G&IF!ydwoZC6zgQ&w8`c-R6Y<fw$7lDPAo|FZ*WpbyF80LtL9mxxaN}uv$UY
z{O#H48JFA7#9YKdXo^2ilg^bNiGaDr6&}hz6tUcw)5sJ8;lCpiGboK6-bqwD5Fij`
z1=#GDr^+ZBvjcUOSAUqK5c7FP)oPqc^7PBa9T2|<(x_p}RjNYQ4budJLh--ZmgVSj
zo58j;WGkEa$E+3|^Ry8{B3~jOy6s+_oro<EKy+vUu-ZWpQGgM`3C<mUJz92s0AXCK
zLSKlIAaA4a8}QUaJ-*(vIp|MMpY9r(Ly+J6cCw`7fk(N{?$L?t56R?7nTgojqZeuu
z@+;?p7kMgjow<?&krS7kuC$%ij|{>suUWhH1XQm*5cr~M9AVW+3Kbj+W22$XR&G|G
z2tIc}DEf~HMB3d$h^BdbYUwZ5u$><>7q1+OmgzTV{v25RWPRHFVEKLeD9ORY)ORA<
z<O2~~5Udlja_y+%)MbFlW+#0jiw+sZ30@Z&`TpM-5u>l_bLnnEE^8)jg&0b%z~u6H
zc)eMn{F0*Xt&zGLRr~Sm_kYAHAqUZ(^YO#q<_SyDLZlAdKVCnaZy6@5PFg|6+h+W>
z{M2YygE1{Kj1;zPHyLs7-i)R5InEos*>53lxt`j8W@JVAwZs&D^V-$CG&BC@oDFd}
z4?LNCuxb`Y=QxNbN?Z48HUepj$rEuM87ZE#YdlNxhzec5Q=x93$AKk}32z7EU4SLu
z+uOJ;%fKYNoH3Il2mmYq^uWU0T;QsG^oOQ0Dt5d&#eT1@wx(Kz-IOJTJs2;w2%a?B
z(LNRJn?VQuy9BI?bPhxL>`40{-inTl6M;Ap=kyvq@VdeIbZIPIGW6<zW4Pr(!LkQc
zezUFj92wc5<^9CWH~I=*Oa^(k9M)&<9+xICD6%fik+S&)PFK3`1jyZ}e+oPI^h&5V
zeHoyWG9fxF64KhswTUv_GD>KC%*ayGr34Vq+d=??=2@g$Qp)RL54cP&#W7Y{*Pr(A
z{bx8`BD1%gjD&O4IgY-*@x(0*N6}KAsV5QQSw}mA24e%{_XwoNTm~<D%t9X;jFwDV
zPDvFTjCHd#?P(NajxPk;^K(>x(t9V#aVHT)oCwtwx+oed<{=R7_p<6q)eI6)t()9x
zsSknRl!61);p+x~zz@e2m5&)daGqb?oE<=a9V6j55#ukAB%2FGET6;pZ8LL<`0*WX
zU9&>ms^W&-EP3BXj9<Qznt6Rk;(xk1CXDg&<}kIPIQDACsIc5GF1O;(r_@_rZ9WYd
zIE4&wIwDT_)k|m^vi+z9^#yXR4Hc1ocWsyLpE3%teIQi$piI_$H9P^S-=t5LGFzy>
zur2zSiQf9IPu$7*dMx)a=VL41CLc8O0V!m`i0ngSvL2zZMn3v~9O%O$5yJ=LM0hhK
zA&dhG!E1<+Q&}|H>4v`t{-h_c4RDhm&ec*t_ph``-{kZU4iezu;>>8ThPi@akRzb(
z%%PL-^(?Z_)P4!77gz!TLhg_(<*|gjR|pn@>yXxYgdBI&MV&N{x5+dYpU2oX2lKy^
z=K0F?E9c&z2c@_Dx4X05DNOIjk16S&DP<G!X+fvlgb;ijn)wOw&uoE&u1UhHA-19j
zl&|Ex1Z3?86YL>F)p`ZT|Fsm3(KhKF4}7^(KAHReE8j|0|HAF-QYreL5&B-=XaxCQ
z&@TV&rG~M~>8;HZS}*lHh8vzJYT5SvCatyg132~1fiy@Zi>18~`GAVB7wo>VJn%3m
z?t+TY*TI@;QpB^T4^BXVI(O34><4X9z|O0{9pd`+w@fo^v@yu9W0k<<cJqIJ1vTu)
zvg8bz{<wPTY;f(Aul(ws*Bw<G^)mbEA=>o0|0C(l!=Y^7|Nm5E&%TFdu_Y$^&SM?B
zLX>T?Q+ATwgQ2mGG1hDuON%5+)-v|BFqDiXM4?6vO=B>8ez)(>e-6jtV6N-Fuj@R|
z_xtsRIL**MZSLPuh0eZr&3_&zPNZG^H&ha4qQN7`oIzWNPXyX)3--CixAF36ZIb5T
z)Hk!IP+#7#Bty&e5Q1B|4bddHHsMpOd=CW<JJxL?nQc(v$?MGFcw?%LRe~t#6=Rew
z0VqNJ1`_yp4p@J!0%6z2wiEfwwc&teF66iMC22z#kRqyj<*{M>nQsT+p~xw}l*WqK
zGHnOR&jg5nCg@x*i<2o%mhi7MyW4=<uA_&Z_1vUK3eRjEnU07opTk;1C4VG>uRlgf
zyB_YX8boJPsH0niCfG*WSlM2iY>9=`%v3UuO~BT@&#>XCBC{+y_C8TEF@`bXnlL;e
z9!X0AMFV~HAF0;;K%(~2q*(5W_=$=JCh7$K-%o;Aq(1xp$OlRn=@@bvn5!R?KAxEM
z;~iui@XkMMUDDw=+2M{ex4U^aDTgE<#4{FGFpbpO5?v(S-RZINzr`WP?AY+(@`4-*
zDL8rWl7Gkl{_*d9tnX#iFTyRbc-Um4GRD(*bS;_HDtm)B^xk`L$e<8%vqCW|K%6B|
z-F_!abY=)G6+wFZP+Fq;p*%)6^TjGQ3S3jQ?Po51Qra_+BlGN~M5tp@TcVzvP2`&M
zZ`%AKK4q?If4<FmKdgRBLTbYAkN0|rcWp-4rE~mxL#*e+#OM{9x1J(J?@aVf6<KJK
z;=E5ET~K~|##bg|9-79Z@{L}vb4J&Dzr7TC%{5oW1tm@^#+XGJPJi8SaQrSvQSg!;
zPtYgM%C?-c((xwRh<>-_W8MP<(Rz+zxUxFaRq}ZC5H!;Xr#$F<b@|mq@3iHt7zG$R
z$(@F~Oojv&pWp61@BbK}&x~8xOIqe@>!dt1o0xec6pHXNE55elM_60+GpznByZ!3B
zW+-NuCz>}!*7DT@ATL@r(G1usZ}@rO-hy*7FY5qgJa;7!XWVQ5joDCHm$z^FTIaIi
zW`WvzE_$UhUu`!*Y%1na4;5&NP7U^1Y?}+2n6k6H$^=xOKx`_(!aKVjF=)UIe4QNz
z9{I|ifTmcf_dN6E@bsKTk!Uf*eDafC4*fMm>atUd`h`{MWYW%e`B0I@890i*JvS)*
z8qf?aeE3?%xi2H-Sh#I>9CC@KJ{6wNt?X;dxvxGk?AthQhAj<@$bT-t=xGTpu73Bo
z{G*prH?yBUKkPs)0rl_K82jYOI|iyqBsAJg_o0#EIKk^eW})r}3UVmJ-s2t{GszE6
z*iNi4IsG1KS!mRdGRV{)oMq$X75c>Io+9QvNinYeZ{F#Mn_ug~d|@;km0jUBKLXnW
zMl)#veW$Mer}p1ZIloRZJ{7_DTOKZ`wwk9~x{FrsM7WiCT<0<G!_7D3U^cm)r+m`6
zsL!F|!|rLl|0v~YSe>Oz74uz(FJ+<s4zk|$Tatp|R<p(}C;#+==R$I5jYBQ<<kPI5
zO|5+yqWBjQaYv?BI>FTA^eKtJ#sQn;p#S}2_hqx9$Lo!f4g=OZi`Btllo@y;0s2fg
zV0x=7uNh0Pv<=fymTeV8N&lMpYZ=!pzm(+Zbegwatd)2%T60!*7V_XoLc5ZbFL+*|
zlhl7HvL=PTu7do`XyS(54A{Ic5}FyO(LP7A-r@gpY2<_@#@EaO2ondOkb8*1L||e0
z`34^y$swm%)^)Bm_N#nac#Bp8Ry=Lo$(zdn`opfcFLPpI;ys}M0ILAy-oDAg;x*i{
zi2gBqkc-;sqrNiPUHVjN5{*AA93X0ZkJN8{9b-jpaYX`eCbBTO&8yi>9JaY(olx9$
zKP-+Z&db94{^(%L$+=ULo7Q?CUMjw>J!g!&$!79}VLyDe8f@wQb)bS3q#kPdVrnpJ
z<T%6FjUz)yI4Thdy=nYIlm_Auy;a@Q?|J)bjS_R{i+k*{kSH+9-CQxHS>*m_)7E|t
z>u`mNXBg|EVU&4!ba*h1$1u`O$Q=4wZYt}~-X8?hp0nM36+jjbQ9JzoLU7>J%(hIR
zqy{!Lyl~*UqQDs={^~(h%T|Ka@O#aqlz6VJS8wc2)76_u=E@sK*W1k{0K$v%-bm+l
zHI!?`;I5JT`&?0q6lox&?U*O{5%Q)_?(gLomt)D%{YY;60;_-lzf}L{1EvYA!9!7g
zsI%PaMgTQjC0$?bp26MCR^oKcR)x$W*)n^|S`2aAR@oojlqU`AuK?L!9sbtK7RwQM
zj9@(!5TRzoX;e^ToQPHZ0B=UNPFqAsBu1iyEEuI@_G;`~-IG5#!UR?Y?p#cCjmg+y
z>6S6o|Cmz%_+^J2G?jYsh=Ldf^Fv+2FL+?S|BV@2@4RR*MbN-B^7tz0(S-pCwm`D4
zlZQKxNYGmS9u+LvqdZ*d<eNA=d!w72s3NsW=ROo9a$;igP+<6cLATU&>m|!hLdN4~
z_lY7aynWki%z-lmqGo-p{>-P&i3uOheR%f3BO#RpBVUo{A7zDGn^I*Z!?Dk;H?5vG
zdC$*(cgP6-zF{+nB{<zH4>X&+Tx8X$cJtX#&ir#<)fm1UD1pVb!Y(6f!@<API&oB5
z+~g@_PSojou%?N_3#FId15zy|<mrM^R<A434FZ#S#nlZgZlmen0@Q-uK2;5<720bU
zB;0$_VlvaAgn?hX<e)p0ua;jq1a3m+FAvr_`HiT9e@%#g411w8FT+w7sGr4M`@{WF
zPnnSNH9e%xEa)YBoant44T*?C4ks02+uB4yl&Glf#o{1AZ=!*MfQvPApVV?e9iy}L
zpnwG$-g5mBI+eS(X??6Wgl??s&hJ^NXxhaGb$rpLsG4`B==ZHn-wlqYOe{}G3xB|4
zIP%*09cko{)NKp%<%aD|$c^{m)3T)@n}+M(BExDAW4H)?2dAj@Dibw7f{TZ{%u|`)
z%FDlkazvXpU7?mugn3&NvX1}0JZtc157k!wl=7B&K)h3yp)xHbMOQjs-Vn$&MM$vP
z6?8jfgsbjHKRe0{qMLt`>rVitD>LTdUv2)LR2@BHfz5mh1XwDa?}~`rRgBxcq?qf4
z{R-SX>q1vWx{Q9$=IAjN&;SiK&pfXg2@Nqnzr;q8P9@dvJ4wj_IZ$Jz{fX}z^I;OC
zgh@DLpF_hVtSczxRaS6X$BEgusgWelEpQ{20>9hIx@QvFP?SCwmEOyssWm6zcD_})
zZi3GL(>#c{DLBcqKkg66BpQ(FYp>dL3Bl=c%|Q#$u!Q_|D%?9`Gxgn)w%j$|4r7|J
z+gB;IHFpE`YYTOvssFx0HPqTd!iwQ9o=%8md_H#xwkXnWo)h<c0n#GB2x&b9P&^ze
z&rlLZbd0u6?}t?B&Q14dAw!{`uDLt@^8VJsA>ha9i^RNcFt<vn=x)Eq+kH)!5UHUj
z986Mxx831bV~Ir}Yd=TnEy|)_{%y{^LY#S)x1cZFk^=7DWnb5p5+C08@)$hbT#>j}
z{$n`&vF2kP4T&P$N(k({!O}fR#D#*lPw!K-qc?0mjis8j?`Gv^8*9#LzT+aQiMHLA
z(VLn2)Z<<*z-B#g2~<#)bJaV0Vee9Sw5ZjH#hm2S#b^UraC+Xy=jyyvU;ovX%YmP0
zD9<IN|9w+XR;F>%LgS8se2mU%n%Cv#(k^hwp}IL0d#GpO>4=bPGXlFkY0QwERlDw4
zs7(4I-*}Q0tz>E5L}PicD4*sT^yV6Qv-QROWWJl%5zdOfllMy1RuS^OA>TWMr#EnP
zc4z&=)Nf@>ij_o5(Ug$!)&PFo<(43n?&HPK)l@+X?5W<hC2*COWMKAKY2aZZL$%oG
zx%TU|*d;y#@b#iSQ~j=Qhg?YY78dz=4UZyfl>9o?DLH{4xJr@zHLw}3jo);6P7UKZ
za^@Zo4SnA`PMxIa<UQ3VgkUbWpALUA&PtWgFbZa-G&DB>R_2d8)}<u3q_M^pLw&{_
zf#oA#dWTgEh#QU6MfJcu>f2qUoYEck{qFvP5ieEM0s=3N$|*zuRgGSY2(yh6RP~NF
zbR6^GaCtJ<7f{wrmI@KP9r51~5KO*7BG;-Th+;Eu#k9l<$<ir6M~(Mz9@Q@*)b0dZ
z=qlrh5j1mw;*AY)tUVGQ48Y8hzm)%1v(5RC?Ds|7;;C;qh(h?&X~oL@xZ7pPL1+uG
zx-8#v4CBP+fiJvWmPia?4jjnTFd_LB-CC5K!Wx^+v~=qXl=Ic!b<H-BHO<0cpR37Z
zS<wCp?;mNw`_u!o4}8MMOv(hmev*FM%Q%uDv-A44xceJEy?<B6mieOvW`-(FmSfUC
zvm38oa2n8Iel(zP+t4%y@tfu}5c(oJp+%+xV87sY71@JgGGZi~<q6o-(9P9Pn%DQL
z7LdYiHTv-Vq}St80bL6x5lSbhfC#VC;Z1`tx`1-pux(1j@V0|v>cfti4}YWx*S51=
zX)uz*Evb>~1=LHHH&Ji-)9%X>?V~N-0+b|IIH=EqcJw`icbo3isDt`2%FMb-jS_|(
zuCnx8dXjNHN`;!_QT1u2%dLc%d>O(-wm>$Na7|XSrAP<3CDw|Q1~KLO)%uo>Vf8=P
z?!8u7iQfu)xziQLH7`^cIf`R<c&h)mEFoPZM6NjfweI_{5lrTR<NL$0UvmM%N#1?a
zFdye*b^n7|4o_Z*oS*IGr>eA>{E1J$xeS{NHP`mri$>zA$Kf{$vDsa^vxxa7i~GX_
zpX;xWl{cQ1jWAkc3z4s%LdHHheopRCX+Cr_{q#&Z8RSLdYWMuQ3{_jM378I-kfz45
zd>l2^G66)_7KVb&RxN06k8Z9N)7SL(mBJL4kcFq`BvL5$4|P5t)E5u)XbAnZB4<-f
z<**BtRzlZc={{rwx5`*P)#zCVbt2oEeT@9CQ6K4k1$OXdK<D6!{xAAD07}H+_Y^u!
zI_$c*mf^Uko^f(Va;a_@;YiVg2>B(?A3Pd7uUL#tOl0o}3zOin8s8yrQK1t4G$p;~
z#S1PdzX|Pu;5#~bYv!TZkqH47W7~TnGhbehxb}A>KrU~Z#g(#eSMuwai}Pqd30~H|
z{r#prrs3X^R(UUDv{8uJzLB0)lV`B;yUGX0l0ArK*-%@g`YvN%tZ=A<dz&{XA@di?
zbwhx_`CsuoW-HxmQN}(@2cy7kbtd@8Oba&N#R{_X64RaC)>IF9ZuGF!6*UgsDOwaA
zhP;l0DdLu0{`b$5mDfOu{Y>qFr!BzegLtuv{xvT4S(vj7VPTIOG=9gjt4}TYG7zG2
z)3hCrQTy~(?fLxp!SR)QIay+m@1aV+JZOYA>F;@_b;}?4JvLgt)L43^wu+CM-(6U%
zOTpcM1L_xVl@3QpSlD=$Z#9?G<433rOc3tJEj4_z=f~*#$yy)YRsHgc&1aF?QKt*(
z_J(F_T_o9q>dU=~Y`i6rKVx1!>g<3>x43M^rFgB6lo9-Qxv-OB`LoMi!b|E@8GG+1
zo4MEUh6!^nA)6uANk4LPEecLYCiX|B$XAGtpX<j(t6D=+XUPKNoj*O+U$v;GDA)VJ
zA5jpg`U^f`bJ&0wd6}h<YPX1(xJdaIL4Mr+D~+QH`~uKKhDT2^dhb6h;`>%)E83g3
z?}vxNHiOt32B9>FdHqCLFP;hL?3Qlr>p$iV=ATaw@^vaMUu*9GHH_x(z91-ZLnq$9
zg8hUyL3wmj3~elnB+M$iG2o3`Lc9*2lk^zlifp?1Yv0W_jpaP<ETfvt7PiE5OPEB$
z&Pm9Nr?^x>{4ACGr`p^n$my=!2}>0FWkCo-rR-N0g$Z{ylSoHzv@tgr=L9e1L{WTn
zu`oM*$$cJqIVDp#Q3v~Bwi30Cdd}H-UxewSTHGbx%P$kk>-0&6?j*}3QkmHpKxNb_
zquP6$x^jOnI0&qCk?t#Of!CK_$*cRAd;og-A7PAUFG3#l9Vs)g@sJ2VT>b;N{WBOM
zkrkXb&&)eOr&AWVTx_^TUp3YgA(iYAEB~c224j}fbf+Ucsg=-Z_<s~ISdl=n%i$0w
z?N#zcSK%(hazsNj2i~G3n``d55COV}Kd-vaPF8P1#$_<C9j_H%ot&6xuW3a|@*84j
z_(ifvsN_y@l;^`ip|#)~4xyE9j^n$9fO~dYK!r32hY@)z7&83B5k?wxTaO0bqbx}c
zTN0Wh10(w>7~&$7WHf20ycuj6hK%8~oB+QgP}FV~GUrPoJ4I?ku3&^)?}zJXi$%6`
z2i!$8D@hKNv12VBFM^gbU*Y%{Ez3-0ESyD?Et4R#s9Ub8Ck5dqMXybbH-AaK9uRRK
zcN~dy$Y{&BG%_a|%~e~XW>~8g?DBTRVDW(l=q9;ZPaETW+t+>MO}$O~k%T~*GsW|c
zNX4@?7j5PAM;W*q<z*%(Syf%K+5wUXvoEJC%y&P?(<G#w^DNs(g!8-Q^f`Zi1$>5q
z7tkF9tnE{iu4bR68ehKg@>Y&$-o~Y$R-PvKW9vce*WjVt31#<(vJG{3BCASE0Y%f(
zZgdCjvlB19z5zc4-8;36G^meDU||T;wEARPf1H`LdH^H_fUr9HjoyTyMVIp1Yjl>3
z@^^9~nA#Z6+B;sB_KlJNGj3%~2^FKFv2wBE<FQ^gMtSz?dQ*Xm_pvC=@Im29X7KOU
zu8!SOH9!g&n$6eC4nW4`thhBzY-|<9ZO?s_3pL#`D=wnEh5wf^aMc;G<pZHl=NTS6
zeV)V^6*AM1o)w`t5h#``Nj)8Umzh)0%Ty=4AH&oN8P^5ya`R_h_0i>fD;FzmL+Ab@
zzFvo|Q3X~CR<e4ypky@936Lf$&HzNV|6kImAK8v3_kT=_esuAdlGy;el{}Hwe!=Qu
z2Hs|sI@v_7Y%8_&xE4(ZvrZ6q4$!YJHe2m8^hb(_7=L3DtrpwbP9IB<7|HH6i!b}x
z{&u%UotrB)8sDcs+BfP|wQ&0{6OQ|h1mzlB^K|sWZ9Z&8V)~zUI5q{@%DEX(@ARI~
z?O!#v;VtqgN=0%a8bRc_rN%e<GLQ0zQ6UrG?0V?+pmSumXqV*y5l`{qTM97PIv?=V
z{KJs7HGp~|2=BI1ECRo&3woappR@yRsf*Xddr<DCBG%I0<toVHOuaJE!dH6Pi~N>1
zp4Z{J5>U>a!E?JTeD}aLl*;Qlg98HDAGrJ<DKX(^MbdPGGt(!mBUM7sRL=14#@O2r
z3Ed5A?TKl%^+ZQHI`joqs?y%dg?|!yx==00+WwaFL%_Je2C?PGO;^Pv1|&G(*R$_G
z<^@^UPx%kl`)_E~yPGEqIevXuh-LZt{Vzh?a&I|FEo`qF@mT&%;r%9~w+@oS;x%>V
ziuc_)MvIaZi)Ih1e%eSR4_En`W8`gPqYb5_MV72pFCT9gje!CiJzJElCyu2--7vE_
z3rYN@;oIv%T`&_}U_9^zVBeo)#6fM+FRAHe9_brKK}d~U9zAfZ4O6NBapJ^mSrYfO
zN9+Su&_deyZnvyeDMUIy2CTU7iEzmgYb*H<V~vH7hrZ;Fpp({!<6x6!3IFLNT#UD|
z+ZntQ*d|KemFAm(=tHd>BVn+H<~m{W&xREZ|AOvdW_c*tu^^Q?*d*x~mj7xZtBz*!
zrLAthpz@4bk-$y4>CQ$vfVB9#@lm`j`-+i(e)VI0B+w|0NLTChl?wc(Rs2P4m%Eyx
zvrwFmqkG92gSb_uClu?|_zrZZVs8Hq0RUYQg13Ewbj5-_npYTphI8DzG3n8izHvAP
zv}Wa=o_HfhPTx3w{i9Df(ePysD3?Wn`SRKY4O%v1!*9Qv^VR(@WQM7YoFqP17<=lv
z;y%*+Y2jh#o5eDa=m`i4mJ3iQHt7tzRn`1t*}Fp)_T;<lyH8^oE1o$guLpZ>;{!T>
zEcWDFpC|IkNlwpXkC#PT+0D=KWx1-QTX<_(YC|DnY0djV;Z))NA$5?d$BmMgq^9`+
zq3A85!5x*^-;5JYlSr~;fx~o-H`r;)wAB1{g`n=tua6+pmXNi$0xQG7nGDTv51WkK
z08rgEA5S&kO+A6PJTbo37O&hlD&(BL-Z!>;(D><PwFubQ01S;<7uRT!J+8B*NQTtA
zIe5C?tjJgDsNaXSPM4#&&0qAmqDMu6$9--EMp}pu?8cXi_J+Xs@d~E+QLA>dZH!I=
z>S?QGSo@Irq+GJyFtQ$HusCTsw(#;r-bQ9i9uUly?rc59yH=WYmG5b^P4zo|^|5yG
za$pw67Irl@sW<%t#UQ0ggo~8RE5G~1X-9mxLp(61jJSSk%S*}_q}-Y%0Q6VD?g!uu
z3pJtnDhQM8c3X>qL*6Cm9*YYUDpkmr5@!U5kc>l}tU4I?bWz8;ZAOgn>TRbXT};06
zSbnTAm$9*N_Z*}1+Os;qwG}T+5Ed3z1-#S+b;`AYHWSyX^<EZ&RiKq<yH5Cju}ml2
zmE+&?P@~R(iAO)pM{uFBuNcA>WfZC75YJ~=#SYwX>&Y);IYis^Sf%5Xd4tnME7N?0
z=YRHFR^vnzL`6dv-H}-b6upikEqkr02fFYtGyfI$@C8#8?^p~`Mz2x1^ivPFlk(T=
zti6*)3XH5R<%lUXNf9G+M>%dezOF>;SB`TL6(XNae&d#J#i;c%y-}_aDgoapFiI%;
zwEd~Ze{sQEDqHqBmK&O%V*DX^$h^wSJcBgz9M|}1ZNM!f5c=eis48PuNSuNrL(K^D
zJ=GhJGYopzY#0<Z^tq)DMe1)f=f^~BCbEYme^c47I&YQ_%_m;IM_DT=d#|P;z?BBI
z@!{x6QE2ndGFAzZ71C7KL1AU}y#Qm8A<ony%8u=P398KqNrhc_KzI6DrKghg)yM->
zvPf3Hd;4!qfU-uB3-4pgIF?sOAcGEOt6LCvhB;yuM_p)SNVO(M0$a$~qX6)l6B2z@
z8{Jts1_ZfcoPdBaN7J$KZ@lBV6K`atO@MaGdclUpF}$dnW-B7ikeQ_d_fLQ=0MJ^k
z(i5?NSDW!>=<X_-t;(;nC)+uiE8hjScJfnY)a!maLQ>!F|BkV<tJ0Zclq(&OfN(#*
za|hf1@{XrxjCYqYDQTWP5|}tKbS4h{$XIi_mdxiq312X_Ne2=Z+UjVA@%w&T@rB#7
zeyBar)%8SlT}@D^%MgEP3`bdW2$r-aNVqncvXkgg6FtvE(=kiyn~fJg&zub+jY@M$
z#Y(89xjysjEZmp$>7IAg(t?Q2qOrdn6ccq=iZ7&;svb#Yx7Rr9K3_Ic%Yn=W(quj}
zI;=z)|MIyW_*xRPoxxwSidipGW#agQul_}r<nVmv3)k@4O>wX-Qv5~n`ot~S&R?P3
z>XtKMM&p%L1~_~vkF*+dBfAxqlTr=Uc*MhS!m3>ae+pdsI-c#towwqq$~nH%CRM42
zut|N9|NV1*9M9G+X;a&>Z_n$X_F+ACVwNYcwrumhj|<ZXQw_oV*225YGCDV}ePBB|
zE!4{M_OmiU`1`GVHP)AGAz6pjTQAjAk<wVeZDbtJ7T#u)y|;4&_{U;A;eP)C87Kix
zK1PnD1doRQj-S!6vaLl6bIA0r7oz(JWhLoAS5O#FJU_)<7OW2$JIPzYPHanKr9F}n
zhBRT)SmhmU?nqg}-?PO#)e%yBWp?!OS?hv15gN_Ijf<8}(aTu7S#|x7rwe|Q3OmOZ
zvUVeAt;sa*(b#vF_LLtG92hnZ2m=c0fR>%NPdtNE70|GjS>mSp#IOP44S1O%B*i-W
zOf_(~-$HelBc4+)^C6T)LaX`Jy8PA=k2}u+r{)IVrrPl%!Z9F!U$02foe<ep{b(Jh
zbXiF+T_qi($>Rkh9BP1EgZH(Ncg#PHEfPPHx9c1~^kNND;0orqA0#eX)I7Edd%c&|
zT-kPN_#k(@hQRX0@-ODchf!F>U7a*pC?;h*OKh#qGLbcNGV->$e7oqoNBNOMQnwmQ
z(mzL`Z6%oM)IXu{X<~GDm8T4ZFUY^8gq<;p=Yq51H=~>(2qs>9ORH%*k(2JzxmZWy
zD7c$8kz(X`D|2LJS-mskTt|_q^tD|YZ|WY|qR}`=%5bjdhAavo`~Xd>z0`jyv^;r&
z3RCIvSz#S1ZYuU`?wl0(dZn)1w@Om(v~Ma~%Bwe$z?5r_s`u;@Fz)^Q#+!GOHBffI
zsFA=2P?WSQDx<UzB=gYs<=k=bYH^-p2_J;lx7Oy49p1KR&0PLCn6Yis5A~JAG$J~v
zwa+Gt;#G@aly&TaK%jKq4Xh+!l4U-|5}^S=ZW<ZzZ<|B|=mE*`SQW39B%FrV6WVC)
zfwWYN+H@CGfw`BLeicx<bJ@{+$O$ZZs4+^s(Lko3XAD4Vp}U4oUlbrr70KfPbOI{e
zG<(Oi0`*HGLBd{n0TJ@m`uQKmi7G#OpWBSZ$|_FuGc^)2q<3*BEqNw8)d&hN&)MGR
zf5S_S1NN^I*d&dIoBSY0!=Zi(Kzd*j%@L)CH)*%7Va=gjcS!inh*;YLadq*`e2*o*
z<$TL8N~+y!R+kA**`k$0wlh*PPea`t2ZA@uk42^)J~lM}jtiCay2F*C^+j@o2|t}L
zFJ8`OjGHazbx(}<r_SD2aSl`il`{iRC)sx~&^iWS>Ij}P3J$XesukEb-w4K=stJ@^
zmFWy{N#^X4eDym&&-coOwX3s09S@ZTUVX9ngHP!HWjUlH>wN1iE^{SoUw+u<`Xh;P
zt<0)(_tuX^4r~`b4JAxd^!P?YCj;f$FdZgTL_~a-HL-H7(=Ee^N6@yFE$i^=PYRmu
zd8n5ibqrb8o3<(!N#H$Iv0r{#wCb0=SIJ;PNTJOZgt!8MP0ya?z5&vWfmkwcAHK<J
zR)5LR6lN3}r-y>qEM&Db$x&mc(h|`ZVI995*C7gw>cZhC*d?9waOSL9+uWo}7I1Iz
z>_+mm;RCq0&Is$RO#q=NlrxehJ`^Qy)|ZR6<}c*B$q;GRQcAPe<<5BcB$_)*<}pna
zzs%bS?)^<u>Ue0x(L44{LoyEMxO*(QwIndkP!0vP7`e~%^=~Q@t9)J7bAlWmszL7r
zqR%~v;nWnrJ(pGu;Zal>Cm30cTQ<xLu9?M~3x}#_8{eDy#*WAt=<%tH6VJYw<Gtiu
zo<>v`W&PIu@q}<Wjt%hlJdjdZ|0+K!+_&UEP?a1MV0J4~SDkCX>eY?K5^9serR=|@
zvxF1jq2BB#?c`O)j4`r~(n}r<FHMODcM#R5OfSV-H^zIj+-(@huJqFO3ieJB+&*lq
z>O;K5H^jG3%Ku>w)uP0xcB};{#*+?;jP16h%|QDi0I`B#?1FxKMDQ>$O@Dsr#jWRx
z<(bV+uS?OzNnoMPey3%*whGUE`)CPGxn=H^7TZ(vq&YzoI$QrKYE7;K_{j0hHa~ZG
zTPdb1SzA=`4<{9eIx>%#2i7RGG(wW3Y>o#Yevq-bQDPXUi(z1anh>O!2>FUNmu`M@
zPMjtHxnI865e%Y6&oX$m6`*yVTjS{jE7r{yZ&7b8ex&%U4Fz+R5jGM6V9m=EPqO=?
z`b;42N=u%fdAl-eyZ6i}Z{enZ_tiFWt2BBa(^i|T#DzK|c@ek4F5Ebw{ky?|xrKCp
z(Rjn9fpWIsukapk>hLDP&7X`}XO+)4GTjybr>ZNU>*|3|%&EmbP@}y&!LBC`RbD9S
zKhhfpdTAC!jC=)<_ei$BQo&^$14~|6uw4WaHw;4GHQ@VR@;4`)7gkl1#PBl5v3oVw
z%YXPl=bd!PpmYhh9kAq{Pkb_fmU)@g8@(C3pVLIv)!>?C>uLhK$h=<2x2bhD9JdP)
zS`IVlT0Pe{qyE2tQ2TpVnt6=qfg5B*Dv#z|@v#5As*x5ApT$EU#T+WeBi6R|@MZM3
zxpyl>_86b-yz%*mxyEaLXo}^HqU^VdmjC_Jbm4^!$CF3HHoB0Kh_Rx@WT1TbnUnG2
z+SAuXmxI=&yoMSA5hLmM%oSVu8i97`WYm$7uQJd$SY9F-TBZp7NDv$WGF!&;TMl6(
zr4XnjbnYT+TlbdPP$|g(2+=b18@zVp50p!05FEh#yQuvJP}WIYzyk?Y#M8;ZETRjJ
zGeX?@bZoma!xgM$?9Q@ZM{|>P^3SD69V=e`$*DEtbnT&{_{|@d-WC8wwI4td>k)PE
zUozH!wDW)KOlAvm$4P%Uw40BaLxITR?LB^?(gFWE%I!nq`%-JpD^=HiUa1`GA@m(7
zc38cd(qmBeJLFr7DkhmXE+9uv`1zyYKDHGgmPVN+IVDl+@$~6(bVkj40DcqErQ$Yr
zobirf`NgqxPYB8Nz~D-rAu%f+Xd;`M#@)Lf+I_`D{CMWncEie{;uYJG*L}}mv)}r@
zW}h!44!v(EY!O*PcHR->rh46QsEz}D(;a@{Yr4EW_0VO+jq3+v3BL7Fqrk9sx|g)u
z4QrO#Y?o+}Th@a?B?@D`qsiV8X}Yg5^eBO*KsKP>7Sc~SZ29qqNdNV0g}zF&^!(pH
z8B^K3+~8Rc;fP25dNySU!_@BJW~AwM1;5#Ax1bPx9JeSWM)&@0ug5=GSziS4tXw1T
zv(K@T4W*`N?o1EU90Y9{zJQ31-)X-8lr-&Cvz1LA3)r^ca9V!<<)RH~hTYd1OGz)s
zg;|=ubSVLGkP{@Tag@Q2Y12E2Y$U35YoBB&s@qf*e{S2+F_yQF3gM(drm13?$ASh7
zG&Mjm(_|H3VHamiwZ#%jNs=GkmHS}W%V<r(6bb~_tz#CrjZ>c<pFv9H0SXCO3*hmr
zTpj!A`8J!D)Cl(dB{ud@n4}P;$(rHH9GY84_t#|%ZG(mPPHuJ=kH_a6!aWnj3Lrn7
zw$D7Yp^7HiZ(K5r7OdQw*1{F33f?D`ipw}j(Pot{kwc>$&bJ4rZ@R;3^YUXo`C_=5
zJC1n(vn`ZV<97~fxMao6q9vUZ5TAn~id8qHCo2+lHfDPI6KCl!-}n=?MLAo_-|z<Q
z>pHu@QXHR|)5bO|g;K6>fa<Zy|7e7GfFa|50wROJ&%0%0U*w}5_Vqi>^@%;kcc9d9
z0pZ8Ni^efp0eH{OCh$I9QQCyprq1;Cn_N?z{8z3oZLROnXyjo;qd-AYPL{iBci28$
zUrqAG23Z&Ob!&~XMHk$|X<N50;m{z(#$k^xL2vH@f(HwW&Mtu;a;>4+BA{+|rR<#6
z@V0EmOC@Oel}xOtSfX*@U(dF2=C%flJIyAMQ-CmaJMS@+X1FceSMKgtrn4wsExG+x
z3R#&J=}6N<lHf$Q%Ede+P@&Sc_UyEQHM6C}v4+o-do0dl$8$&KDsFDyB5XUQjyWwU
zFG;uX_A?vuZk<tZCjKG;6p7dRCGPo*b-^X`b>OyK@e-I?M^y8siV5~p<bbcwlDZ{J
zUDFQ*HTPM>&B!PhsBKmpi|Cb3c(zlu-p%b(i;>B1&|FUiQ_r6vza4a>|C=1^mLQoE
z8M2TDLtZUZIM!&v)*48#9j3_DMsxq2P5b84<yX4xD_x|M%=XRGrMV^woFGS)BBO@a
zk}y|gdAq<EXt5BV^0$m%ccosmv-^K;Il8S9P@f^o)gkqo#t)WDl)cv??%wu=%R*+B
z$8XKZY7!Qu?tM=mqA9k97dPLFji?3YiV_oOvxpjrjVivdQ%>nvSe2Hp8shPWyZK{K
zdRY)y?Xlq{NH8NLyV?7tjnvOOK+ze;$^0vIa=20Tb;?9Z{R1P=*Y>hRhGk5@rP{qy
z6P$S|gv-GId7${>tyvFugLJFq41y;?3|=GO>faE?ybLg|m`VA}s9uLUI;Jf6)}lyM
zwDgf2DOS9J8p9+1%dd#C{=8p|2X@x$#1hYRKp<vgpbL7_0#ObGbOD?#3lza33s+^d
zaBu}RI?(d&%4e)tP8|?CjA*7-CNK?Ux5H6ZM84lUybZcf-?|P!H?JC3<Cgt5Gmxu<
zgpyZ0+j(7gyB6iz47s_5uG=v~Y@fUFZ&P(ki*`@{o^aOJa%JJ-pIbO0F0615^2T{d
zsFQ*gR$GU}*(ZAv@9>A>98@~UMv-A^vR5~wH8dsF`M&HjZ$0RIGDkAe8yBAQ?G(Q9
zKy@!z{#u#RA|45~v+A*7QZOA3A3V;HE69bm^G50}1wN?*(wf)vT`vyjEf!;*-OH>l
zQx(1A99DC<+8?oxPy1O`L%*E$CxRAzC@GLd_uDgKD&^*F)+YRmwdvCEF>#Q~>LFZ#
zB%L3By==I&Humq_N#?yDUq|QRdKaVXIE@|H{b%dCOV+P8Gus9B-v64`87B0(KapX&
z038PNRsl!G-24V2u8lg8VmYNyj+=ls2-|0gGy$d=Ejx*~=wv$_-9zPgRv}|M1!`CD
z^h*IC+eAvWT1DM$umhs6PDSgkUcLIUiXo~kVfrS%sdWV~3msM&Q%)D?PNXJA%~SR+
z?}M5-#Lx-4Rwe5CzG^5@EGtekAuWpg?d9+PKQ}3g;cY?Adn@gPk|6Br5qITPT^V=p
zgcfC}XO_|?DUeK%_EsF2EvWm5Yf+2g4A+LHKtzCVWi7JySb^IQR(f?{dmad~;t1f*
zS_pK<%jwwakGX`T-EY6HC%+Ud$Wr5RWAgbm^uk_LeS+Y5Aq&x=K&=efZ3`?%psui;
zEZL|kgp(*-YkU4WPMU=Un#OZ^uoINw5#_D4LI}+3H1v*op98AASRn#v_(zr>gR}nV
zI)0@^(C+k)4s5K4y(G|!wG0xO4&9oi#-2xss&vR;?eCm07v3Xzx4b6&V0WAt@KXE!
zvutMy<@wlEk=eoU%cb5y)w`M5Hg~Z&XV15L#-Bq>5BGz6XaiYa(@V5@pC-q?Z#Ne|
zPs#kX3dh*W<6D;%utSC$F8g%~ZFO`z?8F&4lM_dqld_^3G3v6~71E^#&&*kNrOLG+
z=Ir|xnVG$+Q4R*soxid`I?b(x+29fU-<(*4-$}Lg?;}~N>G4Ss=oj3ZCz@p~=Hj#5
z%(dv7?sl@Z-^{WZbTXdOnRn^tn{sXJ4rC{c+j8%()5&jm&&`LHL;so+1PAWrkr8Q>
zIPT3};oFnQSXL5h6c`Xk;?ueRzeidKkcNh)URO)s9^|VrRYf2p4nf7i{j-8{SMM%H
zM<mO|-O@J_lMGy>PLwj>wjh1OgdEbBA?4R;@C%<L04HX6P$X&y^rpZMbM>t#xTPn3
znPuS=M*eVTbi*0(7$FTLTqkM&i6!fg;LiglQ$AOMVp7Ju6vd+MmdnB-+ueHlx!kC#
zJtYp8`qRsz%KX2k#m)MqlP)@dq)sPasYRBNp_t5lgz6rAdEC69Y29MNQG33_rEWG3
z#AztBE7+og#1LFrm%$W*urt`0Zf^&MTx3sVsWYK*nmRfrupiXw|N5?tXdOeX&-JTQ
zoJ?psi$)HTJ?w9dHI#`g<$P;@XLfz4u$DOOaf*?B@?Ty8Wrhy;@8%#Rw1_TKp@1$j
zYV68n0?ZWI1`hQCW;Is)0V%kGdIp|Y&NN|i#@9BdN!A!iZV00@fXqL8?{OievY)-f
zDBL(2<UC>C*C+he8PM>8#ES;eEq3XgoRsJpx(wCtu2bG|>WX=DGDUf0S_99%B7QOa
zCTyY@ir*EL4txrG1|lYzb&eO+^nGe8EJ(}zxzI!^Ks0}r`FW$%+#Z3ai9te;a+vjm
z%!Q1iHc#%(^C6^v14Wi=B{UTyWl;vKAFoCE<Qot6TdFT<svo?ms>WtsqRJJ;hAYw~
z){`)IlalX;?JXcg8)y;aB&NOY4zk~c!)dY<%I3#IZ7jpko%eGIAm4khbcE$D0p_DQ
zoqtmk)A3a~Wsykj(m(}db?+sHgntv{tnKltdoCHplf2n=3ROOC1-;Yt@Ci`vL>n|0
zeQpNUUj>!hj6>}<%cV$P)B=*~w}Bqy9HYtXyWoOuxmpNB8Z$NBf@9WYJ(43xc<*}^
zxesaageAe3tfLWL05vgme%A=DU#qW70|E1?%Rqn+?pXRrcfj{Ibuu_*lWhzdJQkiy
z^J+O|Ap<6Et_56=T)PXY2{|^{m-Vf3lRCBd00(1q0~a`Xg|xe!J$ml}VFLSo3wwPb
zYiHZANQf98@TLN&lE4q=NDa+30Nl4;C8;lfs7}#Kt(5|TO~X$FHxcoItA`a6PSVY3
z%S1f{37}jM6}Yg2Kr-E$?W677@gp7m{_PWhV5>wL1(LXpe)EV>XM6QafMgC%0D$lt
zgE$s7_{z@tuUV4ZzWvja{<Zu@;cS78I2U8dV{uK2q$*4bq*7iHE3f^HbrHr4p4&R_
z(0uy{gfD8`9s;{e9xKIn;%vs6oVB-jlGmQ=MG({H@-~fc2cWCB%p*mX4~*#3-~Zw<
zFth~YQ~h;(HUYhM%C~yw(7nka(=j77?L#3)Bt&hM0+A5E^1=T_;OfaSZ>nYLgR7cr
zE1C$>9X~d=2b9pDw%5?E(T7pP(hI2D2hk<zPm@tLgO}@>oW5D^CK32ogmzh8Z7adk
z$Kn(}I$87Y2xp7|`cY;~N{vVYK$qS(AJ3ikVkA#^k5{FUy~iu~aomFKfcqp(k5v~9
za(%~KUfj$~)?ow|c<*9)4+YIP+PaHvTA-kXV@>pr`3_FLyH*jiuwzsEx(Us-(YD)v
zBnauNPqy=WAVl{IZG?!TB=29-3=41B|H{qI#&CkZg5*)%bDqVw4VAb~IO--OZtROs
z6I}Ij|3VaQB52}CKLM^R1OieW9X-pM2DTub?^#2OX^zw@7T25jOY${^)DS`vqxz|5
z2!fb58O3|Xfip;G*~=hR;aroLa?l#36$*Fm&Us^OJcL|My?jYd3h-;dZQ;vwaB=qD
zJ9T9AYR#||34+rwLuJ*fbc+~XY}<njAYGwST1+qLIxF}?@!<hbb$bT;Pq`pu{@SZo
z!)nWFY-8KVNh8Qh(^*fMu&Xr#;zDW7?{c&BwvOpMvi<g_(>+Y!W#^3U(}DU9AhZ^n
zZ1MKAR$EQ2xe(W{^a6IA7136;p2SbaR|M@j>&BQN#8r~kfJa&+>{MSwe(GvbML8;N
z{kIxmvbXqo#k`@5we6YJX8r=TaWpw6-|1fa_J@5b(_iOm>x#WKZtmMASIEpVHmhLk
zt$(ygmZxY4lm&<XMP%?==u9tkPxF8SF=$2w;3&4LS>d`8CoV8m4@+KqEF5k#95r4u
zgT|#5-2qN4*6M@OuST~PdWCj`qa^|or91C!mJ@HF88sE>7h-f+gt{bfRj^vy6HeZf
zN=YD&1Td7;d^j@sbBx+QX@(#zh)_cz$*#-NN1~xyh<&;KU39&KOh`wpS5QyLk{0WB
zT2!hYk91^K2ptf>i#5x*0iCq8k%GEIQy}YPhA{pM(lpb1`PC$K+FKfwyV?@vz3+M6
zGWq;Fus%Zq<^UH52Hw^2fGQp+`)GzjpfSSO@x^D8-i}1PN%W%<lCOvK8LAr8molOY
zBD1Ye1m15GTuZbkwJOADYm3IZb4w>Ha3g_@<O5K#Q6p`@8^ld-vPML3TqiHfh$geO
zuWhy6Vi~OH^`QAS*T`797WM_8*G66jwR$c=9gc%?Jfm6+f`)EdrxVvWo8LAx-gwkL
z&vTz-R++bD706$uVx4y%%~!z&am_*V=NC&ll#$fWb#TyxGe)%g3T%y{Zd)pa?<?fw
z+c))5w!XH{Rq#yrlN95fzHPbdT|T{}XHS-oR-8}NNy?`73-bnN1l;Q|#yx_VPe@K>
z<c7BeXvDEDnqhhR?>c5@8u75$j-{QLxRh>wTyJ^$-%tJ{-SPLPhlKMvW8G(`+T#RK
z|9iyJMCM-UW}{ZHrl&SWpHf$YUd!c$;mSF?I69<~enR4ZOkaDXRpo4pivgRMJ^KS&
zm}`lXPPe^Jjb?gWw0WAfJ`r@xN8~1D2FrHE>HDE9*4&uWTBK9B>FTt*X+1hmjdbk2
zz9zK{&{)d)UGXH6AJr(<-w)}InojA_S%Ri;A6+FijBihD_hQye_H@T|vVp+Fvj-Y~
zk)GpHpP7&oZPmNQY>ndWvLuZo|A;jnLnf2_{*&B2Vh*A+5dd+qB4do%rueANLgfOs
zf1Rt{><H)_Vf9&+a2MazbmfRD6OVM*sY=I!#H`25P*Z`KU()qT2+z^C2R!0`7PY^t
z{?flfU}t@HoB$g``WL<6N^jVQpz?470zk0p$F{m}H)X@%*X}xCUOc{xm1{D=d41n}
z7CY@<E(!90<W5N1yrc4%o9VR~=N|0gIZn;|M5^f$<3W1}u{S2u8(U=EXo<BDiSGto
zue;w}Nq-X?t{dVNoTLEs_5RD})?-c+D22%E>q@kgSgSW0!BSdi%4Z~iXLYVEuf|0<
zT-x3>Q=1l!M;FH9TYue@TBG^Dn-z`E+=K_BCG`yVd7n`#-v9Bto5?m4(vL9lYJ6-I
z+5|rIh*8q1{QadEX|^Wm3RZXz@N*LR`6zNbg%n5cF*?hduC1ACN=(v1K1`?h5n1qr
z;MIqTkCZEdw2$p8&h)k+BxNmuGmv!5PZ=k$4P*tQ_9Nc=a^DmKjL7CH$WLz;wM*v4
zCIlvmzFe*!QG0J+kogNoVu@?js_Lj;m!zPN2Zxo86%Abat%ucFVnR${DZt;61>B{9
zz$2CZ?TI+Gls&VMK6KwJ*TjhTL%pIr+Pohr{IfjX2^gxAUpv~@6cvZl^5)*m1$?TI
zC#V!qEm*{u2VC=&jF<C!$92k6Rwhb<bj@?FjVPx=G(XWbr9W){U2_g<sCY<wo5)Bm
zJ=E`SFGp`ufz!H_rl-1P`gb!{DG~7<#5C5q^Kqavr2WB*W7%2do$ZdE49yT~G(KN9
z@qNTiFarp;765qU*NZnPu2n#nP8RUH8NEGCbzM2hh+!FC=x!qJaBrg>oF3A}Ju7#y
z4`XjEkhvoR-DDa_<GC|)ZP9~Yb?=a5+%x!Tya^W@qTpxZ(eJ=aIP%}gD6qR#WP&(q
z+W%OOl4gd!08n3!19W71*|wkF{G<7OktLpMT`1B4Zbi0gL>t>)7;_3dAnf2`rLoQ1
zQ|XsMFg9yGwK%XV%kF8${wdnSEtBOn1(NO+HD%B8zkf`$%Ih&3#vKY%^8$bL{hqBQ
zCO1S-^s*fS223>Nx`J00ii?JFDV9%woW^w^U2m3m9GD*a|0Mw7db$o-$sm@OK<V%0
z+zVoevh~)K?3-(L70)~5{(`>VJMa+5C$(L^HIlWC?$4pcR+i;#p${a!J#LQY8j{8l
zawxD=!l}5i<bJoV2Ek{LFPMIVdf2oF>p$fHCTy$wfJ5I8tS0m~R`K2k;5MinOnOo?
zR?`hH<X+Ot2F`4^a`JSm?hG0$h|yMB&V6XR(0A+ti#SpyE<EKtg0U#Ay&B63=6xtE
zhv7t$GcE(e^~_91_mQjuRq{#p7pEg7VwFi$2(gNB7u(I-M1ikTL|j{sMZUvIIwdGl
z>sbI~Bia}SHC@)*&1y#B_(3%l3boT=wn18C)E{8)Vf)an%eFyz#@#2_Evk+ekU(h+
zq#d;bTuBn8rZY-;=@5e7Okt0dDrTy-u*0c7x7-8}ESu!Sw2saiQDo+R@k|$T-%dSg
zTrd+Dz@K+X;s&<0jz0y^DKY~Q97O}2QTCo+B<}Dn&pqUc8gf^yX70}ku*l5XW0UFT
zCsJuhFh~Um!UM_^yPnreE`@Z)ghJwO_4m&@(mA|OmeT5cH$lwlXKl4Bgv-tEF6^^(
z>gQ&hGHLxGce^E2A;gokr$5|#-G&fX-PlNUp#vtkOioLiBx!x%6GxAI5@@o0rztV)
z{}+J~yaRn-O}r>}BG8$bHY{Osl!G^WwF^Feo<Iz!`}Os*Iu)8U_UP=tfT9h!X0O{7
zyzajFLieJpm?g<^Q*UlNARM5ly|OuJjzJKeLcT(S=odo=HjI5{rKK{1br<dDk*0c9
z$SY)q3Vma@04&j>=c+7%*&1zZGgj|%!daR~yp-7kTXPMLvj7IKCM2}dQq?j%glZsA
z0I+4xm`o4?)3=Kiz?hPMOW$OjZssC!G<%f#3!XQfk1~*lfP~q9Dx8Ni5v+hRT<XnP
zlDn^M3ji{fPT@jI4-~Hf9Al_=(m9~BxDg5W>(a##b(#+#V`$&{Pl3+{M-K~@EbHG(
zRUO^#`gp9%gb^f?)JcI_V`Ii9M2<L5P5RE|vkx%2HOFNxAIM@%(fpi>In`5LLp;jy
zo7ynDyS3iY57c0I!6tq22T7%6^3wUI!HIG7`KG=C&#Qob><IGHl^+PwbFhmu`jI#-
zLtn+k23k?WmoyD9-()*UUdvEYlv=dOtZlM<9>A5*i~eolo;jMb!_Kc@tE7bjN+}qF
z&w!HOodK>AxppJLo0nL=vX;&V<N_TB)RkwwY-_-g`5rsYAYTB|10^#%bQzIma?z$v
zw&q{Ye1^bPMm!C;FUJ=WzZ)<}p@V?_o8s98$AbjyA+HS!;J2hiDw^a>b0e%aCrFim
z2MF3b5jq|kjaf6?hG!W@nmB)1yb@dr+BEJGwdRExEG*M3|AZYzb5ZWZn(QSdC(3_l
zTiuAP|7lKZ_tZGlN9V*RUb^Jl6w;prrs<EoNm6Y``H8^DpCkW0lftM0D?0}Tc6ISJ
zs$arM-WnNi_yd6h-LwPnZJq_)el=27^n!Zn3b{kBBOM+QQl_U)2Io1pl;}e(xf&32
zk6_C))$K&~A7cKE<J$a_epubHHl#FA48R63jYeJz6`9{qyBG!L>`1q8;5GmSF()Ki
zINYM0{oBCh;#>khAyTfT=L)Jnw-jDo#o+A?5-Yjx03VVeM!uZ(b*&t7!vIkg7hRMg
z)&0yg6HT+}+r|Z*RcrFXRaj@+0=f?R*8T=LR9@iq#<%J9jjyL{7gjD8?}bO4#?hIp
zlH;9dX8u86u?!%NuIo0_@c(ij5x{`sQz=%Mb(1Rc@X2RSvd^}IUUtFp)ho%F6BOqV
za#acqmJ{whox<VW3akx~1q22O8Y9@qRrv+a@EJI+$9D(5aW6vJ^frD`Y^?yv4^Zb?
zmC`o0^!U2H$1eif0pM_s<tN+$UQ5OO7IFZl(e6<8>_xI8o|kTg$vc*=u=M6xj+l_B
z$ce6Q-=V-KuXjn@ZA8W82I_z4!;g340aVWURXPjJdG>xHugG+jp_TuVyBTn`W`L?d
zO=YP5-h5?^S83CIix$njs$s@oHV@y3h6F0@Le)MIUbKsEsGiF&d7O}FEG;cE;cQoi
z5xOk8iF{~VO-(xA%d5Q5U4o~h&&$0VAB)S_w>7lM>`dd09pHKlVyW@@)KUz13}~Cv
zX9*>yhRBu5(JqFW2AmemqE-L@)VOxVu0K^D5dqv}j7n`~X1tiJ@V--#eUVe3yEheJ
z-FBld<{?d0fONxIUMnG)+%`KpQ+1?Wb2~6!I13477XmmM_$xrJO+KH^81O~_?9V_J
ziE%pjT5jG_#KqnBx=T7z8kcf5Rb3RxnV}f>o8Y-SOZs1@`wk9u6(U`J;4f_in&*f3
zEc9j0#(1Sy%5O*vskbvVeDcR@(~L@NOdG0sT;4e8vXgKw_!q{{xm6Lt9HxZrWOr9N
z80@13Rp8iOD9MLYgWp4{>epC(K_#4g(`P+gv$RGx`@x+Gd&?T}sVXm;rdE@p!fmBD
z4zo@iD3)6jhng?6!(s+QE^r06b?ocoelti~H;4(bzT*6zA>S8LE8oq=39c(qU&#%C
zX74B~k^17!owuI#$L9n<m+mJbEk${FQ>|5I_B)vSoNekh>kESv&7<!9KxS{lJhWf(
zU%rgK3eA2`aJ4c1VA0NfJMKtgfIdH<e(BloCn;axK|tiQb?GeLH-WuW;2@&89gf~~
z5*OS)G7Ee;9lHkj(#iG`teNIanet>cfJ<8e)M1LXx*3`}rI#G53}<^Y$xcKnBGwMi
z$mi(P^rb}Bvfc_);9j!JIfeuZ5+T<@q4}2HwzH*lW;Ta%12ev23-56WV}fe;ZM*F{
zA787|`C?Z9b9BeTO&RFPVtyEMJ?8^7kK0W(VE=3H<ScFi%~+0t=YoXbMJFpOV#!<O
zKD5DiBgge7$p$oOv}*HKX6T6EMpbduz$vf3^h;XxlD4WS$IWYl7r(yrcne*`gevDq
z3@0i?K99eh(|>)HWp+fAWsCIhC#C(hAV^x?<Fu-AaLtQl+I3aCKO&-y#THtPD|Rr?
zw-5a12L$Fb9gn^28LDGZ?ZzEu^fGXDr3uJ=nh#6Us11Txzf$|(KRZs5Zwq(_Dp@*C
zfLtt!b67p(6|@105dxa8qEjZp-0F7J;Mi)Ywo<RDz*N-KjH`W~==~{`3X2KYg`Ym$
z3Hh^g1@)v&6N`_C@v>b6F{}3W57x?0)^T(;M;+5SD>Ae^eCmWhD+Oo)HwH?17fu;t
zeiz9s_CNNLzm0H+1qH4qqkjWepsj*;MI$X$V23x`GMG?8WLwb^hvFgEn~>(auWio7
zDh<I2Vhi<FWKAW4Q}Wg+r{Z^Awm8;h5P~5QNKXtvSehAfrF(Q~>H9sYQs4f8Z<G|B
zNnodD0Jk;nzMw_OOdPI~1kcyvQBe7RES>dVQ;qw^eH4|H?vjm0K)ON32BW0~gei=Y
zR6?4EZpP>wGQvb8q&r8K!2|?!=+J=!CXRf*$It8g2k^@{=RS8_*ZZ=}p|@`C1+W!>
z6Fmgl+Na#u>WYs><G1vF(QPV@0(O(i(!ipqbUFwCHZZZ<qqj#+Ah`2rC7-=);6hFS
za#o=<o%AQm&z{vuVY1K#RwOtCB>!&$2taEwVQYXyf{i|AtfCIR43v&mG*6o~TmtEy
zIPXu1<fP0QVShmO=vJ%4(mPpFfJi09Xkq+@qI2A(8Y+v9;T$q!9}Qb}cMQaH?c8U9
zlQe!NrBXkB{cc44_|56R3(hB8?p0(1_nWwiCb<{(4_n#3BquU^pQ2(MnJ<OIz-4=C
z&9cf%Ij$q_g`({Wt5d@6E9B8Y@mQJowBqELg%H0ny<ds^lWr7i7jK9_=^%>G>gL}5
z=tu6ynMvwRgra4hg7{GRk02G<zGCB?ML!KCL#emOPZ0^D6FItlJeuEqzEq1m8~zPY
z3@UOP!hW53Xp*^}LC5fXqU>OcPZ2z(HWPBWMcp}=D~-Q4X69GSp?9%Eds2f8aOTIE
zm;I16#x+|!Q&F<ZqgR_Mj)Z&pG0S6#v2<8A^{otJb9Tml_5iik&lY<te2c4f^LboT
z-$u!g&?y-sUSgfEc}`m%se{Lk_<r3qCbB($y##r$l|Yp(Bj1?*^qVl0tY%5a7%Z8V
zlC6w;FG5_7QjEC%7J5KNh+(;2;#Socf<~h=RQ&1=IeR$PFLnwUMmc_ERR?VFFi(?p
zK8-1aAK+152<Iox1%8mb(0|MuROPk{Y>+5nYD-$ehY^;1+RXEL&Z5@YM;?dw#-Wg8
z1aq~tD5R7ymbZo>M1cq2*EDUw$n4Rj;17k576=jvnjX4{mVwp?hMhEhQJ~2IJnX<N
zc^WuW7n1kz1S);FIRMr+AVvYi%<)KRFc!W55WJ#Tz|EiEHISz%bN?)7+&GkI7Q?MB
zlZ9<2(1(_y`}dy37j6BGvEnEExt`!1$zhAQ6ia_>RLukE4UDoGFhbBMREcg-C(P<4
zGF+I55v;s;5_f3~ZT@Lndpa;I&Z`VaIWLVn=C;MKQ^Zu^zRd>a`xba^FH$52|6Q)=
z>npW-{}i7xG$);zo=Tmw9rX$iMr*U51c{{wOkojDQg+?`avj{-w)$N=jyg7hrKfQ>
zoC-4<L+Py-#M{`6RZYBar4B!`>B_GE??Qe;v<CHnOn|KJ4z0B!YhM!Woy0$e)(U&S
z!}x$;DsHQ{{fYR8gvdFK!i@)9-zF4Ou*ztUO=jgXr(pU?x%?oM%KG<*<$;C3akH?^
z_l((KG%mjA52e|Kg#JT@!!7x9U6PdI+~hC;dpy4l{&BySKJAhB{oy{SXz7?gQBe5<
zUQFg$Y3!(id0~u@Hl&<mXuD*FGq<vsB0E)}YwFQH#5pXMF5YhN>FuJB@K+mtkty3V
zrK@aZkEv|qEl%qR<n}jz7L(^{o}fsIKE=eK_tWcc#(wA~t&G&xuwV~dy29r#hRtZ&
z_nM~M{V2_b+c{S#y3bn80P=25>2J%6>rN+Z^Z}WRGSmBUc%6l_d|-G$A>gL?expPo
z%N1>?1NKny2zecS#Rjza+DSe{_ln(jirw<uMcoWAm|ymp#TU+2J*B41o%8WsV<E;c
zl(aVKkbTG_D06|m(=rTDYhq~hbCNy+J~UV{-UV11dYYlp`Ov==?%p#BK-Y;BjRMi^
zBvrIOAy|$>LV$il)xN!1Y~K>e$|CNO1`JJ}1_LNt*ECGJ$ODxR&2M$af@Pu=b*HD8
zIkbu6Z~AO8X0PE;r@smcQl9Zo2&9)4UCQo3L|oFf+FFdF+v=77N0YV#LWS*TZm}oI
zRDFZZu;YH8Gd?_}<sOU%^<O4=wdhp&0mvT{t2X6hp`?mX1&5kSjnl`D<nOMrrU(54
z&lj#(!1G`^4{PLn(&BXGnjF4h*hNINK#i9iRN>=?uO!>xp5eyFsw^*QR;)8b*e@co
zql1HS)NR0h$!Gu-XNy~}rt{}-mrL_^hUa&u8TwOW%lI2zmXjZ>=D8dI#$*^M9R%xu
z*6f*PZuZxPcKWaI6#R=#Fm5ECvmbu|*0WoU#DLDS3RUB5dUdkNzH6!ITlGaEU@1Zy
zP7D3HX}w~Kc`QqjC#Gl_fwr`_M5Qfj&r=qIeMD#bA@53~P7Iub{6*Lti|-Yxr=#u6
z$_dUW(|{$YZx@r8=Ze&@kYJhqWvF?-x*7wFsy1v&mDCR7e3emR@E(h(X3J^am|wc@
z7t7(7U?Zd)e;K;FV;Fd|$L?%_%E=rgxw-@zyz*nD_I?5V)QRk=spS^a&=H%$eBnm^
z!>1u<7}jgfl$@e7&f%D1yhHOD!$cf31}{v#v$(sC3^CG|XI{|u;cas8ChK`+8RbQu
zF}B6wBV+MiNTr{@t7BTGj)d$^bo8*H(tc^goZ<mvL0YBF!X-`C-x6+}ei)oKZ6)3O
zILw865GG|x=reHvn&PNa0n@NdypTB^>2@zG1-u2IJ^^u61DCVpXiN)uU~0P(M%76t
z<%}fGk6-fmjcpP2%gV|E@>eDzuBX6plOEo^c2D3t-j!(fh?DDYa$fa-)Xy(@zVRK@
zB05?&qdmAAMRTDr05hfx3>Yl#168-kSSkjGYF4;t+N9G=_Z4tbO1b`3=fMu;6={Ju
zm+B{<Q_h$&#zW#ZPG2Qab>yiI&QFEos~%<McfURJQA#w`K|Xk^m~~*8ATkIoel+yS
z*<`J$zOcEVN(NOLza{ZWq<H%UHox`13k=lY*PHmCLWL-ZXly^hiKFgVpmJlYJw8H_
z$*>8qiMi+Hb8VOLRwjq)&i~<j6DmFtZg?hUk}76<SSl^!Khq}Z#}2hmEPtVyQIyNo
z;_-lhFfR@WQ;L_8YK9L*imq+OZFIa^m=nHS)_uF7(*bon)7G)-?)8AOI20KFtQbav
z>md+!Q4hxhNSsR!hCL$2;!7JbFSAEds_%^)EN3`-E=840oG<K^P4fHSZC{m0+g`73
z;kW<#knpLBc9`+)pzhV*C0biSlCPClddFlnsJ&s#_wte%I|R0Qp7Y4sJf~<L1kblF
z2H3tAq;qmwA1l#N@?yWs>t`iAkAd~GPXWE4`GYO<`BZp7kcQYz{rvz**JL}xswXHV
zJaJ{OMl`~w$#qjWu=?d%l~>%MWjq^R&*<lGa*6a7cbjGW5@Mlf?pV`#CtZzmhv+jJ
zY4rGg;!MBW_qjIH#%6hOfpfGFKM{34N;%PnI~4R~4D0mccL0yJ;SRWtaW$jEBQWs?
zN}e2%tg{k_{PUz3EBCezfV%!&F&u$KxREZo`m8x~B45s!xnekn&x)TMnlA9{1A%%%
zX)kL8xh7>~!1FG!LM*M54t-I%<f~UeNJWGKo|W5BlWMO<POzA>5*jz{H~JSd-qEfj
zpVhr==3KmrkTdenSjAi#OX|%t3QK~*2SfU4M1Z^?kbqRIGw{e+Vs(U)VmPdJp0g^m
z(sxM{`+y6x3>heL`BMrjH*M|aKN;w|#u+_X!FKVrDGm+^9nVevy)Y+`rq(|TpD7L2
zm}~MnfGyFfV!!{HmA(@sy~zvRZ@v*PmG*>s<=PGLH}^}!S&p(rsI6h^to<uX3B}}t
z7rdgiH3421Ut}Udv*D~?i%+!g%cTao4V?(;919l1%wL!n(R3^VTgB*#JbqJ^?*oJY
zx5p6xePyWPQbS+X15x8^!*XnAXtto#G%L1Gx_8E3<#D~_)9kN)`=wmm|7P)^cBUJn
zu|AzZfOhraRkU{77;z}TF62CM!vDzI-@i>uCW_zsTjAHHP}a-%$S{$AbG~--^w6=A
z^~2z|NAhb57Pk1+O`+4cf^R7i0>4v!#&GOumAaqG$Efp3=oid8&<?f;vDm**?nrn}
zWXVlSiYYk}QwN3F<q;(R>CC*c+&6D}2TFV%xD_{h{?-bUW@t`V0XuZw0#ZPk44Hvh
zyB`|%m}1FjJ5HJ+p_(hISB}A-qRz&{&WY>D$bN$jVCDSpfJMo<heQWP7~ZyPc|&BF
z`~2&OXfdCQiLG^E3(3$3I3SVI(t~0%dbk&&3yzgO?#JjMNojMr5v7_yBuO7Pv5zZx
zKEB<fF9aC=>a>KC)WJ3HxyKUkf1W<PV~4l0ie{n$QqJi^8@_w>W2d~tIj9$&)-rNQ
zILw3yyEu=f6A`6?(BcZti>Htl@z5%I7Y=)ShgV1uU6v{V7Q==@0B#D$LF%dj-|BxC
ztbR;x19|b>W2_YAG4C`C-16Z&ZhF)~V}kHZbj16wfDmj(bj1er?qNRKO<&VJ{-<@d
zLc6rtEh+IPDaQKXM++V4`7R1<XlTa*UPXoGzA=dV+jn#P%~zto&@wuFueu=y*s{g@
z6H-{af_g=FHH8nOxc!Iw6PEe2neLA&PD>NsGe*2m|87SuT45aX`SwV~{<4f+ML<#^
zUqzRD=*vh2cd%_oL=5kZiiIc<|1RsXHlc(bthv|3gXUh|9zTEvS~CJnK3Sb-_x1zv
zp8*Ks_2-j-Sb>{eH)HdBic}ZCQn}#2{CddMWzl5Acu}I}GVtaC{J79LC^8xYaQIon
zr9+GoEIIHH3Zhpm!q>7zuy`N`EyzxI{dcGan?=yyaG}%!iEwo?k}E*Lr>m1s*v=;!
zg1a&1Xxa`#dM^`x^c0^<K5l4ZJt+n(QxHi@LP^vBJ4cw%cvGH2dy3k?$pTU5VgPC1
z4chJp&l36RnCC--PAGU)CuO=<cJEn&gCv=5U-{sabqU-Q=%r~G9lH$3?ZIzeyVKs=
z?{C(o4L%j5J5T@mo8vIJfg7(+fZf-k|8=PQi<BND|Bts-Wc~!(TJe(@3ehZwx6&dU
zM&=WLBwXOLg4}0I$|9%kC0#$)j$Y*3-JYM2KGK*<tZd*y8CoYhzQ((is;X4jO1qQ=
zhO_DAFI#4FT{?g~stG;1q8LU`c)%PY3k-}(T-K-`AL4fd7U<oORYl{kUbkw0N9bK^
zEqz+asuXG>+Bk8}G)fIuUt~Y!TH^O8lJ?Jj6(r(V-*zfuAOn`Vv2o0$C!H^uXV<18
zJ-zD6)t%<XQs!THyXxOIMZxD9Y0nBN(e#Bd&3qCfJ`lJmdnyZA(pIG)U1{gA6}l<E
zN(o<=!%-Kt87LNMKoNGtw&<E2i3p{9;wr#CN})kr)^@C9IMgjB_09tgmv%*Zhd!W%
zmfjRfmdl-LxrW<f9dgo9hRA&+4Tog_%Vfzz&**f8D({v&T6yDAA~0ZU=cWxNR6%1{
zsr>t`;@#5w2u4&=78NYhEay&2)KVhZE$OflQ=L{ZIATs3a6iW3jwK)&^Jr}aHSbuX
z&To_@Fx&ecu6&c=2I=bw#fT>#n=Ngs(?>-euoacn67Fx&g!2C`BrZ`h2ida8iQB1Y
z5Ws=F98ow&5_}Yc*$l<dSVBXfQ2qE&wRjAp6`3K(b;<b6QIlZ*@}Ni#@R-xDUq&L`
zA#%8GJ_x)l?w@GaZ-w3_FV>bRuf{FW6iK>b22@MP>;EorC~X?`Y&L=xNMbbp-8{X8
zV>;*Z=!z)6p*0apaOTYpX9g)9p{yt?bz<m{HP0Uhid0@%1o<I$3c8v&MrXJ~h}Pl%
zsp8~b;7D>4NL$fRrDyc@V*snilujpw3cVd#tM;&~6rCN0oaY#^y((@~<9fc_{3290
zB*kd>H7$0gUO~HMYs25!S9q6^Gr~Ky=95p`OUFHfWrLNC-(ILP?osbQ8tPo@U#vY+
zLGRhkB9;f+Z*(~PcY)@&kqOnTojY+E%^<yzF5{JW!QCeuM{)8me)Eq1D#4w?9_Kbj
zu@jL0!!-|$d`^grGJ2|~@?wS(cO98jRxo8^rHRDlgC+aLwoz9Eu-es6(R&}llZ+N`
zkTa~(NH79Lnkb=~=1r)k+W^+9AaCPqJRjXOg_1ojJ>~DWBlokoo1!y1VE3tDA`0+B
ze8cO)0W^HwA@Wrp5JmnHcbe?N%lf-U2ZQGZ+)XdBR9Lu7bKf$>wH1e^6{o6$Zt8<A
z^K)gHXBC~pBpuzCT~nSnt3iV5_+8TJGg1<*Hn?exJK#?xNl3A+^Z?To+j;HVeMQCS
zs+ON-VOhdR5$-YI>T@iYn%lH%DVrBg3nKam8v;iBMRqt0C_YV1Pi2|JzZy?zC)TGN
z1FwpJD3~61z&%P57iFvPsUQDIn|`0aJW<VyCF$A7L{b!@HZZ)C-R#G9;l&ZwJ}{q7
zlK-82=B6K%X%TI;H}BJ)OzV`M!Y3PBlt2G-(I?1Hk5eD$8S<tZM!1#j>d=B0{XFlP
z%A}+e#l2w#`Sd9}=jg{Rl#hVyl;)~mWy15{7}`&-oeNycf8i(h2!t&{Td`W*>MbWQ
zSF3PZ8MPN(mn>{?u8=oaG;fOHD@qxyYT`rwwgs+9MlI%H&Jv8=aa4`kI-58xRHoR<
z>`jr~FD46_;pUJ`2Y?dP49jS0?qRrw8tagcHNM{(iqRu2`N52_rXSUWo5a+L8@@xV
zMAsf!ovKsylI1#UsJ+57nWv|%i=Xe6wBntet9`5-K6Of?YdsV{o}Dl~9DD<^Rx2h&
zN`!PW$Z#|rpUu4FlsQqQCf41!F(#Ya*(D6PjSw>XpA>Xd#o0I?#cuKZ;%>7rK9MG$
zG51q71BrWXt(hB{wO)Ws07j@}hKd){u|Y_430*SX9G#1<$_udGiMv>_i3BVyKd^&W
z8%q=IhRuhHhtqQp>?8l4offRbbkLE!-#`{u&mgZ?*bfmj(I7=#3K&<JK2K4YZ%Gm-
zJFLj;oC36&Jz5YLYEb!urGv*GaLU_TZ<_#SIf~H&yqWTr$1Q1<f4<c_DLX-}S<%*|
z)i*1nT!dP52fuv?_-qa^SHfFlnjGf)je`0fRL<lp5cjdP3#NN@fPOhok$scaGHFS$
z@R6}#?vc|K@vAGR*LWJMDfe4^_&d(8Mh^k?3Op+i@d)8Ck+q)pYv2#p;W^D3-_FMc
zMN*BKZ4uqjv_OR4>`I_p2OE`W=+XI#_HDG1!1pcO?CI1iq^jYG&~eK}5Q}#IrT9zZ
ze)`$sHUq|@MQp|*52U;PT&7#@zS%X!gcsy6Pk7aC*Gzu}pJY+<@b%lu80Jr$q9tY0
zYS`6BzWi%D+3+6vh-Ne6x(SSNN?~-i+#z{LEt)H20dk_oxbzVab0)|e`-G(A8{Vj(
zvQ4DDG3MHM&aUW_4j((12+^+X;PG17)){LM)I2wMYj44HqRRc&?hv=5A~rlb!JuIr
z+D_09|7U2jT_9Xmaop3^$(`%xU|68(Ug8z1PQ?3D?$O5S5pDiPpv4ur2hV%J9+R(3
z)XB8d&2_Au`7Kn<HxD*0Y1L65s6^d`)Y!?%iGR6X#>1VX%qzzq7pdjKsDnZb^zpln
z`zRGf(jA%_>fZ4s6vs6SFyG2g0<z;h!aQcvq$P2aH*Cs`yj?IA^m6oFx|tgdg#{L7
zSqV@PlGSZMx57y^2yBv>;tPiJx$xWt5noMhrZfACoP3}pBlv{ey3gU|jif3vC_+h)
zb-oV`1EQi#x}@?&{fzU+(rII8aR!&6(LmCCmDdX3CCg}y<m{sz;oj(*dlcoy!;TeU
zi~Nz#znvfgSFG!n+VGg1Ut9J?nyi}3ww^FKfc2}3rV(zHBkad?oxyJ`!%#O<)`x`o
z75e{aK^k}&b9S({Kl;2I)p|Wigsz@GBO}9M0XRBGw_LluG_Xe-Bz*7m`<oop6{(rZ
z0B3)&;>(c8r|x-wf~WDDZVtk@!hzt|SJ2CGHUF0L>r{9((mveH2#XjWBK~20p?%5y
zv-6kOnZw=Y@lMzvS$H0%)}}W|r=_|;guf6d^^ndn;Z%OFW>vBU)WcRL(w<KzvDNoT
z>D^QgS=_<F3nrTL(uL{n4r*f}Okbo!&La1OA5#Ccl~Wqt*it0&(q*K^xyYz5%~t9T
z>PS6O!Kq=Bi)6cKB3B%5d&G{+`0~6pU@4xp?czaaM-Tm04q5r9&I!v4F9kQYZw&I`
zRxe;v^m*oO)A-J7t8tz~dM6!%!w=FeHYOT^geb|-WA+uWec5Cc(UnuR%<8Ll&dftF
z9jMrAf>e6b+l-6@dfjxS{R7*_Ku_NJqxE3FOm<=~oO@yid;0TKNSF43b%RoZ%dPfj
z%2}@_GUfOpCb+6Mjs-g9t31GB1#eQ?iWGsrOiRvW<YV`$iMi6YFV+UmB$_K@^GKF?
zpZg~4->qggDIeH!d{9P_eEk>Yfrqhg%x4znu9|fSd<K<76=++NTv)QQ<~;T_RzTb7
z;fk9K^Oz7jyy?cVTq)^fap-Doa-fMvm0OHrqr?k9xYJmI_y>8bBI)4fGPsfS<n<z8
zq(SI41#*+dX@S$Vi3ll*C586w)3y`|b7EwQRETwp4#(6EH-0yY6_+{GI{&d=nR`*P
zH)Y*pUyV?=uTOL^bFx)&Vxq#r+<q&`F1P0}5yZa3-@lpnEEi$b3KI>Og@Ri|T9~Lt
zkVG#~d6Pi1^PnXI#H>RY()S^KE5GGtBzHA1XM?sFW|Z^AhU1KEDlO}evySAu<C<5C
zR^_G-gqD^Kk__TEd4kd@Dvwki>4CKcDG09<+2$AwD}MpE0_pxioI*_;!PWOPTPAQL
z<vY=R?L=l!IH3Ydnnw$94-%)FDU5X$cJ{C!x&)4(HPrwQN&o*YOdu>zh}$;1tP)9<
z35JD^dc<&Eq$bTVb8x&*7spfv-$A?#+x7_q|J(^|^3N-sQA@cQOAS%i-%WDeW@~PA
zF(}vy$z94+2L)B7Ma%b;r_e4{G(zO8n*IA8c_9m|Daz=(5-UQ_Jo&^UaUUFeKEjKu
zz3v<9Zu~%de_ml(jJfhCU7loJ+b-TK2r;cAV3${$WxxvJwhgo%H}R2G#~$4R7yXq^
zJ~M@Qfdtyd@cXJID@{7dhI<G){arEdn6@8Tk+ObYYy;;<Me7eKZsKi`{%`(JZ&VZa
zY8Jf5kE$l({{8O)Uj@p{zcoryHD~;?X#uNB;qu5`W0hm6^m^#xCewpe@gtsWPJe60
zB1`G+W&DgB{M)3);kFIeCVzkx^>&f<-2%%pHqn02Dq^L^-CdhyZSb}CcMrnULgB+>
zpt@c^HJA9xo*yGjY1X3vcA%_1W~3KdJMYmnXNY^vk^#+14bfr@K0!F~ra&p~Pd+h=
zV8)CBTL$ECIUKTt6E|1dZJ`7x5-FZ^OolCJ`yK+D!N!tzz-N<J*eJaS;nl^BNdsf$
zX<)FA+8nc0)hC8p(mQFJiXGg~`D@<(cR`F<JWN$zkTnDn{xJpTd&9~2neY6^nvsAz
zvxa9JW4bi5qnF7Ur$e7fI3K$$cD_}xEZS@eERlc`ap|Q4v=(*?CIIE+h)-v?fT|55
zN*FLCU{YW_4IC3isyjSe^xeQWq#S3Nc-|Ts2UtjcWRm5@!sTmpbNl}kMg-z5!$4cf
zer-5_7>Mj(bH1|lr1*+*UmP@OrV`OH^(BDS{Tl|@tmvbR`iRDrFUn)bmj;p&?EEI0
zoFQ9rSb>i0BbH)!$-GXkte|i{$)74;<6m&zbyD)b<!aQIX*PHx-1v-z=9f5D=39H^
zY*Lsa?Lm+#%Qm}ruN;5<8x5|S2zXdM+1cJVVtvq--REke=<Bc9SI#mWo4h}yJE{Aw
zsR{Uc&~1G)YNih|;~)g<^9A)V$J{=5mbB?dPuTL+Mtj|e8h2v!8a89ut*^P<b6q|J
z>nj^3x(Yt=Tx2uP{$QkOyZH7G42^C0j@EkT7Jx1-hr35xFMSH<sj~B)v)>oUw2+>n
zWObl#&gQr+MB*rlmWPaP+_f*?0|V1l#^(jqd9S#8AHSDR)g(WBN!1&T4LtSOgP%3s
zyZ*eE;5M+R@ZkNHzMJu{8kP^{8Z-mARz!R>OD`Z_aX(~pX_h}@v7|*bxT@9DcgfYs
zEQunKfg6V`_a_M2<LQ_$R3Vy>VFk;_5>3C5E_4-fSLdH?3&TR|VhTSep=nYV^<YvC
z(<atHZ+5%yCIz%D?A-_7L9YOJG)gHNlQuz!T{w(Tb<c?5B~q^U#pSQ&(Hans4NW`>
zjA;z%!7vIk@#X#GFg@_Hu~LI^)s-)IlLZ*O`E^AiIZ=<go4X^eLvH2W<@xDkea<!M
z!WWiC!c2&I8xz32vQ)FS;a4H@8*_Pcpz?lF-<=#>=Bw0;+9KB<EQ<^i=jMn+a|+@b
z_xU1<;Y7kJ_EJxu<*IVPIW7=^R5jQ@ED+99ez`AZl3J!&7bwNklRvM_HDxk|X_2?z
zLd%|VoTxENEq}Wkh5hrOrOiP+kOBwA&NRv5%HWIVlY#0y)B9=<-)m51lR2!G$F`i4
zx)e6ESEtyZ&;rQJ`h8Uub7)qCRl&4q-o}G^5rvocl4(8!qIX4R3f}Sl^zTgQ6kDFE
zIeekqZ-DemDtT?XD*J$k=0;E-?X!;O-5c-{crb^V0jJGCIy_xAw>Z#iyY2?3qU-Cc
zSsCW-tV;IuAoIMbb@)$pkBlJEBI^Zu+0yb;rN?7EryWaG#d>N~jipLJa)0sMWx4vL
zyb+aeka4K~vBv4oV5)Z$@%l>3KUy2F8g`LWt4_=iITL_7uRA2nxaWU0ox53PhZk7j
zeay|D;D)*Ylt)oR`&!l&(7pxwSkDLuNUV-xG3L9`3?SmCUU{%j{FNmwZ41{3mh6=5
z8)mQ*J%+Wf{`-0njMh?hO}+~nRp+eCY7%IJS2D_?f7iotfRfHO?e8*g!8w~9_u7Pn
zi9>^Ev*a>*m{eML95Y`?H9h3aB)LLt#h^m6v|Tsv^-G%HRJXO&9f`Q|+UQf%#1~8*
zC3OTHs|+Y~SSkm<Yo`Xe`eD{dcZYt02R97)MbpU20GdjoK2$g7l(kqg?78`Ja<_a-
zU*ALka1dpB2LE)P<{ZbwqzdT)g`;i&YXYlppJKryDRs?!6g_Vs->7(A=iKD!FWUkg
zZcQ<MPuW(a1Ss9zp^IwHbuD!4*pbi$g7<Mm);5$brHCSgLZVP|DEDk`iHpmLQP^Xl
zw8&qHD9wj8RLjh1?qT$~;Q8HH*ts?P`{#LU8gh*rtq=cF=xIC=H+`VCR+Hd4a>Vj+
zT6e}5=8KZ9Bln-KEO4VGAg);)!A+K6hins_4^@vCL+qZDCPXRk+HS8H+?b$+csGi1
zZyX@QpfEz3*hbQRS~s82h<2&}Db0!}IF1sIc4byg`vo&cZ^&T#`Mks(Znz>@?j6hD
z?rpWn>(;?N%AysTp(pk#Ta4^SvGQz}5PLyA%s8}8UbuKDc-~_|i~DoS$gMm>MjkMX
zjJWJ+_7j-npiFQbUbj_FB?<I*k9H3szryOT^Lq@>;2T35{@jBm4%og7B|K)_-Z`1j
z-_mk~5!uk>#ppH7zg%8fTKF{es>$-Pc1g~1;)q<U7)5;x^i*=6j$yNqV!aWF)lCWb
z7Nv`GZq*%3R&?Q=8vhn0;C+5~8_?8!iQ`A*FMc<C>$moku2P+gp;G<h+4!4`V6xlM
z9ywCvbhi=u-vvQ+`R`5#RNnqCZoHvJdT5r4UEEVo|K!VeC5Z6+Fue<9hdNc&D#N^_
zNIhf#n>McTR7hXf@*i%v%^(mZ%TA6cELDsFd~PMAm-Vh_idVJxrclC(QPZLB`0L^q
zGJ~mL)1rHQG{=0A63H>3s?((ZE^K*3##HtX90Yf^<K<`&=NDnk=M0;i6r{!U(+Qni
zAnrDiI9s~NlG)4o*9NzsV-heY4rIXHs{?=)q6PPKle;BxuPvV?(E&u62A{kL05f>g
zb%e$E2Li@n@71idVhv<^fv?b87bNLlZHg`o)2Crk5<E{oIM#8D)>zbrH^NaJufGDY
z5-{5zM=)BBO@a7wlI=*ioMHe(-5V4P*^%a+oZRlBZQC{SG<|%ELE+Okk!@z%ur^%<
z3-s<y8}=!8tD5TKs^ysb<o=ESF4Xdb!_@qf^vD|2xyf0dw!oQF`|@b<wAC_fy4QE*
z?_b9<<AjAm5JHa^a{apI%8vSrFLO_!%HGFv*DP$u0x3p!9v<D!-soVOL%MXGaY?j=
z^9r#Fu{vOl`SSOCC9$nYaqGS;G||edQ<P$4RP91H7rdH-t3YqDZN&o2*o`hFV%z$&
znvoM^05_@rM=z+n3afE8?i)zjOV_%d=S(l7G$m04<i7F{2c!Ro2T1}pR$<mur3Fp)
zf%f@>7*WoUNDeC(%&^6T1gtk>wMEz1IZljwxvEHClww%>SJRHb>Sj=DsQ+(l_xcJ5
z35`WDsnvJE0|<~t4Cm*^C?(>7Uk+X}C*Ap($~W$`zKI|{4Q;<4_w#{NAsN5ir<aKb
zrU51x=R=iWCyLH*PeA+WJ7<DY>6?@yyiqMhm^$+7v6~UNCnaum@EdOj-y64~Kw_T&
zQTdBaQdTGYS8UF_5B|j6a3^-GxTZSpY8qyaIPJQ3pZVa%>H6k()2hTf+=Q|{GQ>zC
zKhdA1C^U}a065rFjD%D39#Fp8%jSg6DjY=0qxc%f+Nk}&x(akUZz-ug#xP|d4*{ru
zb)HcYBegl{>HtOz*yAIR6&c+562ph4@wyJ%`9pz_mDs$MSW@<UJM+S)^%`6^{tUw<
zO2TG(A1i%G2P7{>e*22z85!2an>6*nYtc+8MZk#Z8;08DV2CJyxs=(-Ia;h+_-0_G
zDb?)pP@@=6vM5L!iV&f-%sRUFJP5&e1KT%wB0o;*Gu03_IE+9O&Y6Rc`5qoKL@6wy
zos$fgBIrqfMB%`{#R%#OsGLplOxn3yXM&F@Dw~;p$B}?dhbX>_b3x&CevwKwvB?_*
zU5|&SnWG|o{!DKw>*8<*TNrWEr10C{3oQS;a1D4V4V@KY#)0P;aI9$KuwN-Q4+Z+0
zb<XGb58da;Mk!|?t;@Y7d&Qs8+lDbZ9FAe8Ywaxqx^EUq#S7d}Hr2~n<bJMo&ZOM-
z$hqvJU;A-P6I7l)KLaXYSO82THmHnK=I`UfVZnQm@)<`Bt#Q}%2ABJ{n#x0oVqstq
zDZS(6m)t1OpLaF&fGzn91Bl`USpn4+!;3g<d7Pum)8(hqhy=h}ZZoAj`(b@KS*Mr+
zl|aVQp0XYSI0R1CMzdG1P+$I%gXjntN3Opa#DJaF-ro9%Pxg~Y7ju06kcUo1J<eg9
zbFF|7p&CU?7Gvqq0sjPqc&Q7t5IT0_+C^+J(Ho#K)Wj3O_2J@i^K%y!uP1bAU-oA?
zqL0qUOZ=Nk<3bPIQvb5!>{WHrmu=S4=CvjFu-c%fG-Un|8zf5l42S?((OJ&TtP%7z
z-P*=vm6DBs&iq)SP0kPx3gG|`m=BWCA6Y(u1HOg<TiRx0wveP=Z{vEvdzUE|q-@3m
zZm#}is2jbFJ7)%P18j@2x^eXHdxm4UhoTX4Eis6xcwlGM+8=)^PgVMM&T#7S;;8z;
zAuSukyE$p8&V?6Y%+&Xh^xuWf{7L2xitS-wbGM$~o?hhDN?p_`ZvQOR9jFO((Q%MU
zRDuj3aEL_nEEdB&=7CFw6QE;9Lc)RHaNLs~1u1AW!%TJ}HTdB9OWJ<*Q^zuc9BYNE
zN9w6J_SiP*<kDJ?!NjbrviwbzI{CcpLJH!<Ff***D=Xw<t;$!DrtnUB1rRZ=M-~ab
z5j}gBx7LVHmHdNo*G7<t0Ge$`bqDB~rA(<(X3^gY>g00EaxF!nhJSV0)HNToQa}T{
zfJPUXgC&7r4Nf4ZV&DKZJF%I$Ih*f~W}5v;(cNqb;|m*0i??>$2^XfKT;rHE{rWGR
zPVN07ufXcZZ99*qQJZt)sATFiyKr*mmsy%tgt%V9LbP98POOPYtep85TgaFUprh-y
zKGuoDb{?0S<9E^_cWEjMEc}u=zE2S3P{BBhBn!q6TYyZAg8-NK9AKcFgU=bW0(*X_
z!pedJZ11^aKIudOPY{j-&lUt_0QN}Wl(Yg;FrRPRl<<S*4;1>L={AzCCmAw^xYq&_
zYJM_OPPk=EWae0eR4&^RlU^=;0)A8lL7iBPpXIOhE&s_1uY&(C$I7Ek7ypI+R1f-X
zIKbsF%s$*00Q^w3$W{L~l)lSK0*Gt<jA6DY5RXM!w2&+#<}(UzIY@Wp6IwUFQX}SW
zJ|$l?r%E{$?SPWgAP;s^YfT?r-|;Ixz9=yjA1#!A%GGE@-A#*JFCS80g**MJsWbia
zW;TE_QsYsxK+h@U0H0Zh2oL6eotWE*)XP`r0ygWZ<J8Q9BwPa<MU0r)B1i;8HbTi9
z_J9^VUjlEEf%F0*(=LZav42i-0Sn3o<{ioBmGHw4z|$#~IOR=6%(R2o8Sg{-fU1_B
z*VA%iBN$pgv>;28(n6((@>u&<TF8TiE~e<KtD*JhS#1nEC%mLQt%5dE81st6<XK68
zNOM2LRiJIXi6}$A3si~^q!Y2_!U1DB<%uFA{0<fblGT0NPp{8PxBoI9r@XHxBKTna
zui-iP${Nben21R$rl|soEfNq;v%?{P28Wa5<zH_5U$Lxmog6k?j$YNs3%J8JE2;rG
zZI(?y(|)}ruo@oBJ>||&NR{~J{coL})MWofUX3<KR%POh6W5ooC=IfXXV2+}S1aFk
zWxrjqYfdJ6=9xYwCvF6PH=GeB8m~&CwB!v8RhM}Ev^YN7#2Q-MHFi=Gy_Naw({S(u
z0r3cU+_lS%nhkf}b~X9ez8Cn>KM?mtk8HOz%r1&cgYfWEZ2vN6qSdO2>3ELo>2xbF
zB4AelIGHZ;d0G-pHRwNC64f839q#+5!1M@+V3Jn3*IAZ}NT>*C18{X_0mbnkNraU(
zTQ`&FyLR=z3nhS*nR8e$Zn|Meb})ux4W)tx-cgRC^?{uA5uzHN)xuh;yI^gkr;Sn<
zgn}9x0bPKl@P_*)pN`>~psC^EU!~Pl*a-G6YY1$gB`H#br>E~CD{K2C7E7OD-9M-T
z9Hde#n_uH3T8+hEi`H5<Z~N)fkXu*qbj=ydBz_X|-ndb^fIg2Ff=hD3<V(7p6i-0|
zHsDB0HQ|R8{=>QNzAn0}{))&SGXkMR2z9p{#Pht%;KSgE@i#)iHm)_%n2@qk0ThR^
zEO@xDJ>>ZiuQ*?%Sn9%Kvk$6V4lgDFaZ|{<Fl4cop1~n5Do1JPN&f4+h9+>Vvj3Q)
z+5Itbm+#ntmu@%L&jw$%E-2d{F$>R^8F{R_r8x$ereTJt2M~T=Ige(PHn(4mL(z9z
z{HcP#r<Is=KN7@jjbY$C>ITD}?C&@f8UU*;a5a+mWlxervW%YO;7Jk&f}A0(4tRtY
z{t)&#E$G(26ED6S<vOEXp#cm5n=z9+Zr}V&H-$tklaJZd6&7;(^qe5<<a^_J*qZqU
zCI`u4(cXNI48N*oJ@^}7>Ux7g92&$C!dRJ#$swK53h(M*{$Cf&*NIbM%UyItK&0^3
zv$Zi+ydU$^%8KSN;!4A|Fqvh!ZsBtlpg7Bt`0fPQI*wfadU#&Vz9}1u4&L*Z#j_b0
zam27tm!z7JToGQmDZA+an&N)~q3ibRmzJM2hCzhftwryMTMxF4YZ-(gnh*3TDU*BG
z1RJyXxY1jh3-&RCwlA&I3DK<W=7ZL2DBcasHRkt9*E53IF_b%R6`}58MYJx`?-RJ1
z3fxHPMT`E#GKk+O*=BX%^>F0T>Xd^N@k>RRcN*n}8QOC+NoSmF3O?MVi|Nvl;`v19
z=o2pg4ZcM3@fxbj4rkwH(rJ58<@Rkyz|G9_XC=DZ_*V|Ct7vBAH2ue$<83d-&LI9o
zPvrNUBq_yI>6#nxhhff7T`w25rZ7mJ7^7gHYr7zf`PrOz+i@knJyp$@+{%M=<E0$j
zLjqqcN_^mmaQ~d`A3*nk?QWy^9^U6^^I6#YEvDyr6BxJ?XELUm<ql-63wfz|>CSA(
zib{_HAv6xMI>aYi+l|Hk%G+);3?Dk4)E>~Ww(77X%uONo0}Q?k^bh<wy@@=nU?B7l
zyPe+k)wS8^=hF;XrvwJ<#$-VV-j?16;4$H#)3R$%Wm+;{aI9fh%7B{$bbtEppTqD3
zSDYL=Cu_S>_jeLdwpa%L#+YJHrFU32Pk>nL`Ep|W)L}Tnuc4Xt<Siw0UweH*a6Yq(
z9bZX@xrXqOA5kLO0DGxjwhnMD1YlFq?fq?IiuV?2CGn1qm}%5s%OBC{wnf_Y^uIWE
zTObqbf1<U?-xF0?;%MKTX&z~Ct8zdpT1bD--r*mm`^8Arl}y?3Cc(Vxz`E5CH<Dxs
zcnQ}0AU^RHtnrDiV-26QkA2=GDHZsV0s+PAGvA+w$^pZWA?}|<qa={cim>>7apyzD
z6%F$|_5m;XH-!APl6Vv88-KoZuo3e7@6Ug|R20*pX}C0~5yf+9c60Nifbq_o$q>7Y
z@J}mm-nm|>^A}~}gXdAJC-!f78vD;eOS{(eHkS2nh<(MUDEaPdzF>M3=;@%x5$x)G
zMx;bcgo{c!m6FDidt>k9E#Y64q9g?98Y1#sn&1(_XoPHj_F(Epd0>b9Jr#6e8{Te*
zhe5QEC0(%meSpixv)3+e+&)(K^s@%cNeVhNrVsvz@cJ#Vd0r+)m3*a-O~ws%NAZZh
ztz=a!e7N{N00aldr<NeY%C(&I7el4hzfMe~IlKD()A<i&Vc$llpXQHUgK11#GZ9CA
zv<@`9usPvtDK1I&cd}%NL-W;dHYYz@k~3I4HwmPMFL<8|pvXFiEwlb_cg;s#S#bmF
ztig<(Xf+0$Wr!Fl!Egd3dT*?hI;5L+EcqG-g~Ij#xN%K@^$8!4*mO+--Ct8ZZV=P*
zOx_wyk=_?FJ8O1+3$*q`qyH`>h1AQW$TnkpYQ2DCM9k`e7k(GW=w_AU9`Xe~kWSXc
z&IkPrggii-6yM8=*ObQ+fglq!Ylu6xk1Gu~xsMnP&1(F&3}85g+2Q44$Cp)tCU)vK
z-vB`a9O7A>_0FnA?Q~pgwYQCw<s!vMsDHjdL&UIJXE|!##o!_Q{Af@lD<`oR_xXz!
z(ooYN$P92LP1Tr>gll`nvj(N95PWuEe}u~cKrCMvPKUnu(u+6)!l?ha0q&3^QS|BA
zNDAzR=SZ#dKX0TnX0jlw_C+=ruU#hBS;c8!kk}Q;Cu({)!K|VLpl#&wi-~0YX_Z!N
z9<TKZVO!^$VPi8*Hj22cWi7=&-YbGuRCD<_O8+V9<#`BH!$-92th}UDH1*%*h`JUb
zBa^v}^JMH}%A0!vPtU6JBZ(%%Z5!>1t;?R^REmEhLj_cINZXwh06(rdy=w*{Y3Iab
zmr_a&=~c@=7<Fvl=)X9r{DUv8dNn$uNhMr9kt?L}CzzmP^3|7MEyiwNZ1f~E%)H9|
zT~ovPG(L%XAJqmu^PYuG-XE*9Ubrcz3W~U98u}D#h!=iR*IGf<3eoMdeMuGi;zX9m
z9P8yn7uk&t5GqGiTU8qkuX8_(-z-V?gafKmDOwmlkq^OF0F-_sZ_NBjnoAznVX!#M
zjHg86oJ@EA<m0ELU$Nt7@SK@riG&}Vc&qzGKYF=WmDbvqyHFP;Y9s&Fj9RLB01&b(
zqv!>vB#AfS1@XEeEqKMD9z}(_z$E_-z?>-ZgE}~=9wFQ+3j|9D#Q1vR>N6u4*#XFd
zG9&h}lnq~&exxsm3c%<;y{uDCaRA0yhw9d=fVD{Bkf!11vBLen$qk!S5^56m=TWEo
z^hnUAFhpI48G=Hp+{KKAH@WqQ+JEC*n^TRcI;;(c5TO>_#Ai7SudV~IC__eoMs>A<
zKSbQ~&M(ifI^VnntkunB--`;W1a*%TwKpH83KXJMl`e-23yrO8GtE~n>4Mvj#qBvk
zime|ICv0WNyzXn&vvIBNnQB+Ufa=@CD?~X%PI7iLntRgmU%kY$;=b=aCvzfVmsOP1
zxJxd^Uw&Cx(Dq?}>HJ)BeCeze5(rFQp99=hL%y$-bhwFrVog!RjI#K*mj+0vop6e`
z8A}cGepEW#{d$%z?Guc4r%QcQfsRd;_@JOWeH+xj;jdcd_Wp%Qtf9F?5U+fm+lj||
z`2*>g-Y~5I!}9u4d*@XI<oedzm@f7TOI2o=PbN!oPx<o9dxXJ4GXPmqRq5fjSh9=q
zgHhvF@renxM~oHr6KDMN&{-8rm6Edib-&|R`1392rvhan!5kCH9|#7n8_L6;LKZad
zJDCnc#@)^f9!oBIzx_%Ui$DFmPAgAoKffrk;XR%WhcA~|ioOS5xpIwghWhWqKXSn5
zNvDz$az3ig+F1ot%j@(sI#1uNb!8?MOJtw7rrL1hugn%>b-#Qj>2GLXBtp=Q7ok84
zSttkSp`(2(Wt8@6*^IB~Fj4!fTLRxX1mg3XS^q$RqQn)}eWIY5yKV5~!w$z5#;fK~
z*FK$I9Z_vKMU^y9T8H=14p*uzSTM{doam$hVMg{D4zLXr3|mLG?#nm^QXJ^87M99<
zMorm!o}$z2c#|{GsspyW>6b(|l9q(FK?3knUcc$5pn~Au28kxFCUiOr^gbu+$APwW
z@eo_D<xjt?i8X%N>AH??GEuDi4l)Kd!OZueul)6om=^WBCUZ*%4Oa-Q+WT|O4ifQ&
z`-(w3KC!$*uU)SQJgTVUh%gX}S@`VG<L@DXj&&P%QWeePA5wWU9rj-5$bF++?a=S@
zj>NtPLmxayI`Y|i@a<b-x1UAG0=K(mkYsvs`(*VB0^h!4IP88?L6s#ThmQQ;g<ks?
z=}$&GdVrIC?&s0Yb~_!Z3jcKiqZ5!)&37kY{n^c(#A2lArvNlN^L&$AV-LEXU=ozp
zD8qTDC+*N@cazRUQT2-=W$G(lhy>p0gMYQRoD;pYeCHUXnm_yj_xOdwg6UFuv3UW2
zZ%1a`5Ah&GHM-+x)9n1Es=HP+A-uWb7v5KqJ#yNx7>Vl>rFott*O5@qFJsR|Nw<{l
z`P{@m8||wF$0n`Y?UeYWR=<cfQ&34wQGZdIM8*P7CvZ%YAA8zDb@IW+c&8>qe~mG*
zwLa&Kr?YvMY&^DjsM4{B8s9RdFc|h^$j%r2JmFoBIuoSpYZa<LEBbmk{^nNx?xNOR
z!bJe=ADtbG+ZSm;K-(<0Stn<!mS0bR<&yv04`WuE-%A(JgNF1wp(zlSvEo6~Mtt<P
z>A|rggx#c~8h{tbZ8o;N-^O-pmk?sze*j;ss5wdFf%@SQj_(w;0H59a9jT(~9Dcr0
z%se^o7U#fk?h{gQ7~aBxUMp8qk{rndBwoZ~nP7aK63KBW#Bn`~KSD@<9&n+g=vZ>4
z8W%%?!<^S}&TTjzati8HBH{B4*8C9ONW1G1jE&gd=AanF3o0Jp*ynE2&MgBCHrLP?
zUfCBxdMDH(C6~vBIIh<zF0Uf3)L@=wu{MXgKScM}VU}K~8bQ5oZWsU$N;61;ZP)~L
zL!#+<lxM>fTZ^<Nm$DtFS@B*DH>zJrS2Upv*Rl>E?snFSCBM6GJ$VsT{rvg!<(C(u
zR!wGE?h*#xIfi>4MZ{)FC66bl4Ht<ic}0!|Yu@*9DMD>53xpzF?>d*ZuHM=SljSQ|
z{Sw2loPJGLIz0WwF+WS*H%I;L`<)zUu|^zaBp*z<nY``6G=7I&Q>>UQq9~dt3Lh~v
zQ5CNgOJsgnONA&v&$A1JOp0(_e&S`usy{2zJ@-NAF25)4lXA)ikpbBSOI1nBzG|)(
z{WwG_ifmm|l-*8*rDA+J9@Jyf;s%5Zq>PhBG&&$t?9E8-QC|Az%g5|qmn83RYv;Lr
z_MTDh9n9I{@h)p8DTb~dM#tLuGz?3W$rK)nY4@*i?VXi8MKe<k3^~u3)7N3W2`3UO
z@<XK$BVib4VrVzBQWUrT7dn0sZd2f40Twhoz~sYVGC8$he>?<y^cdy-1{|28ei{FY
zGs+Z%l#6V{fMr)Jo7kj-xNp60_R?<vurP_<`_P${Lrb{`$JujS4*D2Eh%<6NpBjV&
zN^z|J)~4_V+$R*xhUe7UFSm>Bk5s1h(17Gn4N``9nqRd~yS-XiFaWTW7R2TvjG#Ay
z8fux6dYyRvN?n8!EEhiA5F;6dL0>LNBm?=_3em!sV4sOo4>c5uNQuCN>B3=MSjxE&
zDHzRh8_=&-i)>%9w{GBfd*jpQ7@gaRXc2NC`C|yO`R<Lm-NYCcOXoqoFt)H*{N?5o
zx39cAx@j@_OV^Ku#ybfNo!MM7*>bp&?`~I9Mq{o~{Dc>ThN$MVbTTZt?hdHre-QET
zPwfjm_mL*qqusm{Ge_+&7UaGxFRyQuyYjvDN?@#6h~{dLv$&>=Kh;}xlC8QD5G31*
zNUn(P)S&7Weyl2ITR3Hz@8>v`ztt8Lp-K*JB9);H)(A#E!#c!ICaX|R9}QQBGuY%x
zd@a7yVDw|5l9$A(5!qzO^~hflT!_pXUlAJd6crby{jp7lYdcN17yhPtFTco3MOChP
zla31qWI};J@`aJelbiPdm92bs?4*kbYvN@CZ!+`E<8>ltI98an_Yv=!xGfJ4?PmJ?
z>qKlnCjBlQ3$Or}ueGlWnfz(_Zrn`RY-(c;3Y5a7rkaVXv;SRasrZE#%sXRW4(0Px
zr)LW0T19}g0`<X3rU`OsZ<D|>)?hyABfxDBut9(>$<V8tq4ONKyKv2c=!>_BSGXJd
z9l}&0YQI*xi`P=D0$CVC;bWOA3uwoDkk{pON)tz=<jqvTg@1{5lRIp_x{u~juwuK1
z%auMogPp)diBv%ZZZ6?OrWFBekE^wc>ych-zDuWcQ9=^)i(>jf*;(UV+Cgo^u>bDL
z^~QuFAbcgVAc9TJ-D^rX_YYIIzET)$Gu_W~gk;A}H6pwF%z<6mJ+<2jSuS62wGvJ*
zmd}~p{Pn<<{5|L6MP}I>PBYvS-lH{>Y=d7ymS;!MDy2u-0V<sDeNt!JmC(os80Ao`
z?*O?k2tlKBuBK_W<~^|=r>n9~)^T0GOhpiWI%UEQg8S}d=T@8n)JG^XAf2(v963K}
zRb8Du7yaJ_cQ$#@O{o;!RB~8l6TGYErSw*{Yzhd0OVP30xmMuco^TXnc5OzQpnh5^
z6(;ydRa^nU)HHrB<fuC><P5Xc!DPMWdDWH_li!pBIHaJaNYO4svu!C_gUj&W9x<E+
z&z@y#J?7Kk)-u@P3-_`Sl0%e!$Y4FH^7s>Cvo_M(@=2~qK9!i<j1b`D<dK%LV#vFj
z7hW^spp$kaG+oIUl!3^Sn9YU1pAfLFVy4)vrEg)}o2?l+&1wI)Ep)nWS{Ek93Zy+M
z>)A%IhTJ+0ZN}l$B{MqAdIZR}4h@QtQ^&HqL%>Jf_%rJ^H(KFUF^$eEqIEf#7)M5%
zuc9I)kJ%n&g~>O&(y8mTrUeKWSTbrG;pcPvFBARex_>{p#kM2;Xx{u#=9(JY{3mH?
zR-?XmdMRE$^eLS@@2&LTL0e?bvLGD+bZIDA$Bp0EdhJm8lN`t7j!8ywJ8g45w@rfK
z%Z6sI8F708uAwafmhoUKqdwpGb53J=z({tpL)jLC`3;Hq#)MC#><r@1M>ME%PqGxB
z53p`Bf|<N}wo~H|E!KrUS7|53(LC>139`Ojuu<8`p26`+BJ3{nohFp0NB~$T*_O+d
zr7&L8!$bn}E|o7-Q6|1e*4a^}H`=vvPu6iPAB5oA9G{83dX`VF?k5*-nD9kuUwc~?
z?6a1HuzI$%RM*d6Dlc`<M%^y0BI5;FC+ALm#@;UPY943oxs1BT!$}R6Gb?_kNcC}p
zlta95U2-iQuZePQ-h;=&Tj#eB@v?j(iB}i?o;JZ?vc;#f?7A*ip@)ZUx|fZKv0}aX
z@?x|&+fGFD-RiwoXWwHfvMjEb6(JhdD)c2%+iJdb98WR=#!qkqX93b;RM+vFN5-z$
zR?8SzjLfDEZUE}@6}#p46C@z;Z2omJSth!;ZBArIVOj1;IY1TeRkVa6q^MFM3en;b
zmt&bhMnMP22m_ymYZhN`TYj!mO@kUFabsDjLh3-cL0+Wrzvbn;o>B67MWy_nh9xCt
zj*&;+5O}}>O&oc2`Q4lG^ymkX3tyQI*z<+v4ne*HIp6=y;VOpc))I7*?zXv4<fN?A
zF&)YTjJI6RtgDkNXiq^DgC8=F=Cr!4-*^z!6zBcB?rmTpswKrN_XdX9^HIllcvpG0
zd|E^9uh~zVBGc_HLp-M?kNe*#sTuwe^0Gj$O=FBnmhp4#rd3`_NlnB%CS(PSeWxjU
zX01E*jgRPrSPyi&tDxkKGgN6UXEy-b+-N|vL<hTK@gm#+|HW^SxUqy!8<`vnx@X|#
z$TH-(s&N1y!k<{lzDu*tCRfLYz|JS1)7us-?p;v;dt6O6tKZ$R%9~|qiULV5cg<W`
z3P@G4t?TmKqGUet5hguO)6z4dHPv;^4Ojp8xKS_JzefF3-u3@TI`eoa+xPuH6*3|F
zE(~Me_p*MBv2RgXvSnYgC)qt13}YX=kX<1LS+i!DK{1qOLYAmelP0FId%kzSUw?Y_
z$GFXXo!5Du$MHV&M1_Pl!^m#jM^&$;knBM{$&C!iv)oEH`$ictl}n!D%F(NiSI2U)
zw@m>$TDM`z!#cj;|Nc2y_Je0S_+S)&;w0^Y$Ub(vdA-sd61kWzd%v+)kxN&ixIU>P
zoKc@s_LTK&Cxl>5?svMMh@6545@rSmxy$7>GCt2!BU(uJ-jA?u(6xk7jbl&wT&hpA
zLqI1=D-^>>i`oDFG25Hj*@gTavRUO9C@CfKB&+I!$Z%v>U4)x2K%Y*8>6D&x5zId%
zAkG>niJy_N-QbY}rFtriDw7gb4=lL%1m}{YU=h%`c7RF+h@eGwvYW2Fvu2!$)gu7%
ztF{?HwSfpe^h9@!%hyAkSsypO=8Gc$ryt7+AV>3!#Il)8<vI{vL*?#_r4GVw!HkRZ
zmwfw%uUauH68h}BNQI_6(!vx^&YXV3N5(y6{HzgK5l9?LdpE*%x7)uTq5;ftM2Yvv
zR>{URMBm!Z&cfzE&%id0n6$yNc+Vp-a+Mv=SDj8)^=XRkIHnr^d(`SVjN*L6$9{h{
zPHaVP9r$C!bC-P_(3w*A*MP$YL|_p00tdzwL6S&D>#NaSowicH%aK&*&+H7O>Cgui
zBTS4~#2o^%;5kk4`|nUbX{TZOM+|X#9t~nuF2tUIff023U2CC*^!gxpbHBjugSKyk
z*O7_;I5>W)EjY&&Pf24UR;)y}&EP?HX9ic^>A`xICpNJqhHc&TM`mnN51qPv0dEJl
zUOK9};7mKg_`>7ESrsa>LwPm}AHEuV$1_uZCyB|sze~1^H;}7BC2?`-c%5k1+-gOA
z6|JUSLWo~u+Vxu9Qmm=wXqq9|UtSAGlRN=W^HVkKr^<Z)R~r0!;+sD0`EQ<_2ytri
z7US1zswJ~ubIuBmyX10;Z>PB?359<wIq`isjCrl@s@FeZ-}PM2kgC2V=%JxU2wLq#
z6%*!#NY2oh``TrIMHq!a)`D{Yw5*|_jLJ&}fa&5}8&$ZaJEs<MhNu{KuEs5x9_!AK
z^W6&X<oh~h_$rWGpfvLI8RU_IS~Xxk^QAxo^P||<_*r)iW1PecFxqX9&n|JS_lmKl
z8xm}K&eQ>DjWxlc#xNY4APU6?12j78x3RpcchEI71!BaiG|9lZGG!=>xyj5DVsQ@x
zkPLMZciiJ__JGyX11Pw8maf|`*)IN5I@zBD6-Ni}pRz(7I<Yz#O^(4w2EXp8{y4p%
z&#N2H&zyTGn0kLdFOQs_683P#XVq>m*q%f`_5Jp6hj_L}VolAry7TFF05H6-7nHwp
zOCXY}&QqD({G2bgR?rAhTAx`jZQ6On3UfrBAwmmB&z@ETtz^gHm}FrHtpX_|pxCe&
z2#rbQ&R>HSw2RJu3`W02JVRu_l`jI$<*^fW^S~+jv=mkU9yvTQv`cb+dKSgia^{+-
zDKx%2f7HI!9n+S&UtXcK{en*c{guO6>_s_QeH+X8u-?Sc$>n3jYbDa|z0}$}olc;N
zS$def3m4r~M7eQ%W=~saP`0#VfKq~!0ttx|geM8-dWDub>k%yoGyKcE;WNQO{lkeF
zH_#G+0BVumnTN@_rbTG<M|<5Hir|Ba4$A>puu-ZjO=Lo@y8z7!-m3X!3Sbh<;%LB;
zY>%ZS1`@E3;%`U|r!Qz4+g^2UAims-yd}^>DfwSM$OOpuaPc>DQJ5+`<y_LH{SCYJ
zvuduArAl?>OOE+w&G*?K&@3ucA={(~%U9SyY_A<=-C@k9#tKfb^1%E%GkNN~<+CTg
zPA(SKgH;jZe4%&MoaE$o4ZQ+WmH%8cb&C8Hk;mmi1|{k#qD%zlwN8qcTzUrcD^BW-
z;1vlY!*L;_I3~406TiOvXD($vH|rNS0z-^RCthUTFxt~$I{Ae1i`jiCL(A^z{7+(c
zZ}yqY=(J4Z{9XlZdu41TNP(n(%V!13%lnK)6f2%KyqEE>kuQ=tW!LKD`O!Fgo+{{0
zG|dI))wZ>xGdCWzN%sM|0-VkAHhx#UB@QeZIkk6f#jHe#Gn6r?U*pyWx^9guX7sel
zV`<)-7jZp&fEUcNsy|qZGpVoONBS!h9ky&P#<`ym<AR(5L+@K%?tbwhSa_z@@@Wmj
z!8NU6=FLPPLFsCB#Z6pd{RY^FNp+YBv6%6++)tZ9^m6xVj1Egr4_$poc(dL9n=;{F
zU6`_Nygc89OHIOxiXev!DRnXR{bQPfag(qRx;7vZuj5pRn8QnPxYu)!YSnE2?=Zmc
z+ym|XB2{0dx_={@jfdcFOls)Pkm|{XhxvN43W@{tyMCYc1oaO!C=uemU4kNMai}gC
zLk_pVS%Ae!%r&(23e6O}j4sf{=qE^87k;FH8M@L7=ZgIuuey0t-53FDG7ykQKLCgV
z8hjjG%J%ecfWZ*r5M6ps`m^-3!nwN9p)9YvSY@@8St!ecdt*j3K<p^M_NWi+!sp;v
zb44sk6*A~89b-?EeYUkalHk>I40%CEqBN+VFQiyfqR_M3vr911=n9S^ddaYAE@{i)
zJ4{n)C!V<@!C(UaF~v&s5uB&;-B&jsi_%#8G1QF}H+-3|<NehKqBgKo&&A4MZ%|Fr
zsE6R&>{V<?q*RE-FvUA&CcEeKMqoTjx=(p2u>7uYR><(e8zA|+#Kws(lp{F2#afhQ
zao&Sd8_7@h805Ozy$MG8Pp#Tq$IDM{@=Lx#naN{mrGi&p9vb`2d(+mEGQIrwm4*Wc
zG1L49FIe!afzRzyJS5Jq<yC-pjFi<-5qCbYM>zUJ)Q-NMdOW+IVnXI>-<&=Iq`$X5
zR#SS(X4(4Jxz`C9cf#lC#<>q-N`|#jUtr;&FPP=v#)Ur~9DaM+CO=gZru?;?aq_;&
zwdWv>jTVhpR2yVh`)of;dl1urM^LA9kHL3^+S8GRW0trdr%L-_3dFoS_@$7oJI_$z
zC4|o-U4wtLryqJ)bC|DCcy<$z0P3qt^gb<Bv}+-Tf{Rg+H{>N^shNXH-zf1aItiGw
z>;3S^`OpFSVr+syKFY04%Dly&eeL26N!yYV)z`S*rSQL}<!;6gxn~nQVM8oH_6e{x
zjEIj@0Ft@rVGo*H8r^8)IRRG=4q^=fa{9~$TR`2>ERcNAL)*;RT6801X7Q!TPu9o;
z>}E#ah3~)i$%8uSp1yE$9zn3q{g89mQ78}&{!{PG;acsX<rS?>bJK_wh@0P_@JGb(
z3PMkPd&)#>O^L~(3<i#1ZB_(vr0a`2i4BO1GgwJ#D`iMYR4{6O7Sdnn=g(0opf7b_
z?Fmbv15C`{b?&zoDd!uk==Ia5NYAk@50HMvkD==fUZ;ja!UW+9`WVBY-!Pet`*J)J
zD(&)9X_dK#O`?E&jW|gjaQoeNq*rjHM|Lb;e9oZ%dOg3&qT@B_tfFv0<7R5(L;MQ0
zYHworU59lxTYq0a^b;|ohZ;t<KweUL3pMQ61M^!l$a5%$e5(<~X4|&h(3|a#6)3>t
zf24%A`@zq@{481{6z^TY+X%?Ei;0&sHxu9m1i9dTWhjqRl?3VGMua^jN<Fs*5`XQN
zkny9caeGa{6j~!ah$1jl`ujSXM@BCkwVW)tUHCXbsP+*jhPP0&30Un{R1D=FA(8eq
z=O4x+6@nhE0zlz}38vS$nY3c7FsFdYGE)DB3a;=-Hy!Pg8>Z9w7%hxQHEF7S5VSUP
z!^=xq5y$P95$=h#-9)}&o;}r8bZOA=Bosh)*#>0N^uXHa@!Sp0VMdg`+a;O^!$mGt
zEJ_^UM1r~cq@~R6K)k$gsxZTaMIuGU$Vd`}AHFqTs8#oPIpQIvGtWzS<XYcqOsH&)
zobYnS!wz7ZV2*a?e-W(wzqAew5W8W^!9^<Uh}q5a)wbPm3|l2E$GAT1{jC@IwjYxM
z-qF;IjCIAev3)=~fCbK6TfPoIw6foD+jw*&adcaP&m6}Ewym<N=4EBFXq!l7+c77o
z151o&Xx%Hy!|=?fdL$=5%Br5)V+pE?SQ5>7=e<oSr0rxw^tFrk-V(3wswY^Ei;)Cy
z^e`|$V7?`W1whqNH;32fe%$q*F&sGOSd}QWjyv_DIXMSLJ8zzAG<0scA%OAAc1cN%
z1PrVShaHWJD5Oe7Gyr$!gBrqFmWDNf8CYH9h6Ub%BMH2qojlX&H+uAz0YBL`PlenC
z9Zt|TCG`7ka^Vwsh(Y<3g;sBzmY*pk0$8|y6vLUMs>0b81=Bm+Y7WNYQwfM|^Nng6
zXblY$h_^O17@%BHp>ejk9$F$m`G!51PPvh{R;$gdbhYMtj9+VOb`gPO4Jg$00TNa&
z{K&RutXs6^s90o1XK#c~TRi_GF~)LgHn)b6pgB@ETI~1xw#%=%R<aI8P4Sfs-_*m<
z08WWw#WJxJ*!Yv*zKBGM7$)yV-g^81@qeB{<BBihyF>KDg+h8M!S_w)p;cSbE5B;C
zSDrJqx#Ycg(SbRNuXnoVVn+WSsUR*{Ch45{l%IQYh^y6uZ<!@g>;*8pJJw4F5`y*K
zBv_xR9o->0iunA9aHtP#$At=5gK-BBelN*PZYlF}2B(&WNv`p9#iScj<%apwHHNop
zmY)Th=Og$-@-q*ci^Zuc3zfS{9|PHrfe{`uOwVsdT)}(awvNVtWFDFR-#-ygnClvl
zzu9k^PyoLkej*(IIz1zP7mT}oz`CTv>(E|r|4Evaa6>tjnClhGO9DinB|zYTOMcx#
zgS7A;-xkfhCbv3yQ$pD<w;pS#YhFq6-xcP|YZYVdfW8w>u=bVf<PFU{V0fgH-tlk#
z6tXeKxK_f;-gi(qSw5eudtjX9RD=}70T}9a=!I+HJ6!T}RhWD>n{G{d7bYzOyPr()
zIXxCH=K!tWh9zJuNHF`EK~GE_qsPpMBM{vU(xRPo)-ST73=&vfMv^QGPn4wwFJNnv
zkr1{5h;XuqNWh=+4+NTi1AEBLKx)aG;@w}+Oae;OB2Zns<7PjsSHMm)boZQjbC7tq
z%e{_j+9}=;TD0)>BO3y0Itnn^3b?GbufKolotIi)A<1d<OW}&?5-j5yfA${2FHAgE
z=bmdvG#)ygdy(Gp!^ikEYwx#H*r{+!g<#29mRjQvh8oGny?o@S$q$rUJ^2kc*`Jw*
zwpp%V6-4k8HUTV*2B(4&_4-c>Vyx*f8Cm``YRkFC`yNN_fMl8<_i%bF5ji~;_w{vx
zy+1NyVJdK=pM7i(KVngMy}RhVD@;~7lVlMv<F{TQ<ja09N=_!i$iiW=`(k{>NFu@?
z=JWB2)lZIZ=}F?pj^_xeJ6!+$lSmFa%P1WBVv%L}ELCtCUoG<8ID=KL=DIy;U+@=B
zN4XlT;+6gtaSteMCd8sn(M|R);g^55a}a<~MH)v0+VA|RH>))|o%Hb!L0@cH;7dQg
znAr}q>sSm(jR{5~KVM+UVxfjOt}<oc&gdAh?k$yiQOPO`K(!E;ImnrLa$niF?&GHL
z^c$t`+8BTQX<DUnjJHlY|4~&<iP&w6wOn!h06d4Lp{LNwCbsTfEd1kNpGQLXa~4Ff
za=>;1ZD>x3v^e$pXxSitIKI#s^FI>;do2rmF-o#VX&cA3!_));V0qr1-prQ_0LC&Y
zfhGcugj<Vi8Cnu_9e#Ex6W(<k-HTP;yt;H^5KPhtk^F!;$ltkEp^qy5xIwvX(O8Xs
zF4nIAEkH5@<NCe;6eGSi0;PxMT!l=D+QgND#fhcIdN?+I{hQj%?)>_#?Dns2X-Rr$
zADI9^1j5^kcN0}f{)*$Z_g_EZCTscjT)>}~`itBLFLU^k-4>|IL1ESCO|eb>8`(vG
zA6oZNFOj<i_msgZ!Tk55gQz8-f&TUNl^ei8M-&)UOB#h}bnLZY_Eo~ETfu?61(m5U
zKPnJQ$s?--e)v5mNI*Bv2}AKS>b9CF9q>vW<d8Eu#B#KbdfGPL-(u2l$9u4p+4Y=#
z|7--06bA4!Os6uTu9!82y51vhpds`M2R^hwQ~f}lfRbbt=%s3d!k545C;t>aOQ|>@
zl$Zoy-ih`&<Ar&z_^Js&gV+FM4wwI}xZ|vO>7Cl;HsjuWp+MoZj-;HN0s4RhhTkKX
zNy(8Umr}DeTp|Gp&sZ@2_v6Zm$n~D^L1sgy;f3~7nsdDW>;?AMSr&JqSNiBg==41%
z{B7S2fDVaqPFv~g=jV0=Qf}FCtl6C078F4CN>wj`fLjVx1J%A)UK+^{*7GfVGOLAe
z!sOtKLZet&lW1KyBW8{S;auB|IvD4I9hNPFAlkIGyx-v7QqjFH!On;5D-xKBLdfyu
zS*IYCa*GO%>_9<FjISo~JlEaJaKVljg6h^`0GIXFOg8?AM;aHCFQ%PSJDz<rS!<_g
z2APTHM?7l5-^ELtAJ$Dc-2*6Sana}WLto;HSR+ORKtUiQH<rn0Z#}5ue)*DG?>jwX
ziKvV8!aHv4=V;Q997}*9+%T~HkW66SqHK&vv>faG7F-upb<?_#InsNJPLqvN`W0He
zaIDriAWkFRSq%-{Um<q*2HNk-bwg36E#YEM2?#qoO~uTbT0RH71826%#_b}IyiL$C
z3sU9B316=^lKL32iSJ8?#ux7Bc0bdp5XEvsr|DbUe{cUqN?>#504()u8yD^x3P77i
zIRWv86zoE!s?K~k5WL8HWNBgv+*m1p<@Z41=}a7l8$13bVcQt?rwy=2X1dYJ0L}0S
zV}7As-J|&Ip!`JI!@F)0-PSRu0^k;RofKEd%NbqBh@pCu(unsV*I{54{`ERqAt|D6
z(Qt=tL+#+~h6csh-P$*vY15u2=63&Fxc2~O6q9qMFQ4(hjdO0^GUFT@$X_QJ$DRXS
z)6dK{LH0BeF)UTgmTci^C6cj4TAGq%sZ=5`3m{Xybsb6bKJjp!ga>MT5vo6IubCw#
zay?I)yOpMEIj(7R5-9L6?c1=?q1cgS$-e%$>838mj_uid*t%@^{W_ayrLdJ3vI8$X
zI}uhQ3A&87UTaOHcV<>B&%0|ACdCNCKZaEM16=r7UKvL3@5jPt)$Uu@yborQe3<Sa
zV&V+RNGfeYNo{4rN=!uV^uD(Cl>N|_C<0{D)gQG1m!Yx5tNR;H0vEJ(M~@B?Q!lO_
z0a{?1V_A^0&RX)P0IJV7A@(yBk+bFp%Oq?19Qe`1g9a%_s+*Xh;o^}(@^Dt!@z{oh
z0j+$$=wG)!EbH+x8PE(iNYPY#0Qw6kMZyDEYO+OGAw&Q*A+Vp+#SAKX))5ERlVRd@
z=4h)q!21U9#mG*P1{d}B?tff3=IK?SO6<+V_CO?RB+cK^4{@Uft?Jk7bV`TUhfAX7
zR{F9l^Vg4AcJ$Eg-e&@q;2_j7KlOsP7|Y4+$hnqE$Z(~&T)sEXCeI3?Q}CRU0HpXx
zccnPx;e0ofVH%S0is;X(`P-F%F%TEW@fUoDrMJ#rJi+ntd#Z?UN2XQgZ`)t-dUJG?
zq@ks8X^bm7Yk0d=aN2a#_o`q6W{j>#eO-0`Z2L`@-{Y?O;DUY$mo8Lx(vOaJ>>UTA
zl*XQJ$^NFuMKeQ<%i>4cU|kTKOb(=!=XRSn!q8L)!H+Hxf1E9Hp1q8)mA1-K5<i0S
zVs8o#=57J@1N)=)jp^g{O}F$XzwaNsIAcsK)B88JRQe~w^HVlzuF}=am}|VKj*1?8
zKV%o38UX*Q{p*bb(mZBKhhBL=n>RuSSVe(gq7Q@{c`<+E3IG613r!x(fJVMs=NhrS
z4Y(pjOm#UJ5|Uzc5k|(d;1pWmP1=oU0J;FTe?|PF5Kc;#-I=5US@vFKM_U;+s{5d8
zP6^q-rz(9S#47|a>HVf#Ro`cS1U=xk*#dDh*ox9WXR$U;);5MWK9L*LA?#rRD=V^&
zAvtSc8!%k512RM-O`!Jyaxd!^T8FNkaRcT&zI}-vQ%HuPTsEF-ohvsCl<iQz=b8hr
zpQ7#NgXUqQHwbY&J>srgA0#)@J922Vxvu+Ud5k@Ju^e~n?BSW<;}o3VyT=<&52xRL
zzpzf+U*FtLVW~=l00})bYF$?DO#P4gM?74A>Y>>S!M}l#&YdcWiSw$?5eVtL%JPC@
zNCvB#{<Sc=$>srKY4~EpjM$8JifdW%TvWc5b7!^K83YD1QUyxO)jAm>kY*4oQ2=UK
z`b)oKDF+f=v;PWb0R=UnU@A=hJV!BJlt}tZnWY~TajtELobpB9B^y#=XMWyZ+51-e
zORi`Zq$R5V*j9ky%|?wHaMHcn1`<i#B%z+0x*Wi>v2TpP(ob{!1oHf5Syy$r)^S^2
zkutff<ZEXlPfT9I_UFNAwMWfoDO=19tUV?!dE1Kl>}RyCZ;yg%&jil+d4`!Cfzq@&
z<a@+?CebSKw$@*@^w=7!dfz$)_v5-dO~pfbCf|52vZbzB)w{gsx9$meEh<tYf~BW6
zUbF_uO?+?htZA_C=rOTIodxzC891$)v#ue2(tsrB7$3Ll=w1dc&b;&n0QM!<V4<~j
zeq5jqV+<;JbHiXUYk+IpJ}^lFCL+>K<2QV5c841qUlcLB&|V);KaKJNN-~ZaxB*4}
z9X+(Ev>w?4I-EL~u~$`VJDQlQAJ-Y{#R4+~o`TE6PLucbrYx!KFjgL575n!AzaP2e
z9|7#PMp#~Eo+t;3XXuG*XNwcn!ZX4j1DS^>Vh6OE3TLpOdC#iA`I9@N2Ji>dhP%AP
zO0LwjzsLK#WyI5(rkW=26PSbhpGVDZUise)!c`AeWffA6EV;1>;-LVd4gy#JANK1q
zTNXu`1VJhSbkY{L*F8P#hI)C$gGh22_vGIc#^ycj@-!1i_?eQPPh|hBy(boY2IB8S
zNAvf12b<mJ<a^>c;rx^)3%Wd2J(f?S9$RqgHGFdS_S_!3vSL?+RSQc%ek>cU*AA}z
z@m)*iNS_s8_LNW7RwZu1yvNE!Lt}W2-tCzLG$fA+*hA=xm@2p=Gf`%m&(aZrwm~09
z3_%4`C%V!v%1p{TyiwxVA+pGP+Cp%r?hSt1JF*V5;&qOejOKUDnEw3*k~@@s275dU
zCD@2lA7Fe?dY&fO4I+x!<k2)=>5pQ%7WsvNqw9rJn7W69#oS&+0|pLH$)Yx7WY(oD
zS6$;9UcRA+LSq+@z!QV}2;sbB`P}y6YlFf$Wk5DY5imLO(JocACB8&1P({P%#j)D3
z2-VDr?&yq7mC4Ll<{A>>RH6VFLBgzQmNbowfXl7)LN+JGra*TdXnPd_&G|jT)jIZ2
z9cC~z`o=aDMs%YeW_64M$Yr)8@WnWH;5`yK`bM@CFL6+-SM2q4yBt>;bdh@{(U80q
zJoVaU2GJzhU+^*ix=AQ_o~#-ffsRuoOi*fH)320=YwToz$8_mM;m{~iVij$;gm&50
z;;d-x0UZzu?`uWhWb%0Q>1^blu`H!&OcOj!@4fIR%Ubhl$1wYssgCV!52KT}QR0j&
zCgqm8*hZDgw2z*PO~`!R6;M-Goixd0$^ZS+S|fj*G}YimmgmEc1I^}<tSjO?uf|`}
z6$DP;I<R8J8}F2N=#m3JDEeDY3Y+%Ki4J<g#hM9zo`Km(sGJ-c8K;f^$?LMbRhQti
z0Jh+{^=KHH*6-HsH}r!b#w<2JlZ0sjM&xFoC41yrX+A5LO<nRM%huIw&bG~5Sp(6V
z={f4{`39Oj_a(E&QGNUOuYDBXJr;a#k6vQLh|wzPl<qyra~GB27-#l=S;S&kImu_J
z0X4MLcrRvBaXrAyMjQk(oycNY^oUie^Y=o+YLV{sWo(QFe98~GGD2}R^%J#=UKzVI
z2-$~hE9SM(56q^&4OU|mFugb9z^!@kNCil=Y7=%(rnc7@E#%t;#97b$RZ6k!2zqE&
zF+Cfm`R|`s&kno}8@P>e^cX0+DY-eB*>*q@liMXnpn(|y&g=$-ao_;0%2?PJ#a7N`
zX!1CLl?)Z8*qwP6fs!PZz*7D9KS?EUn`I92^$CWx_Za&;zGcPA?-w)uu*yO#SaeAm
z@HU;^P_+*qOnCFfl(V1z3D7Ql1@F&a+Jb%u%x*tmKp*g+c^D{uwa5+E>wdAyBBI0?
zHA&(wZ`?Jq8;w_==#Fj%K4VsTw`6o>mvbWmFl5T+z?hs~=%x5eOVzl8oTw9O<*>Xv
z6-h11{Ea6^LQjY<mBNBkiifs64f&m}$I=I4x!epm@7*k93wu%~+ug!d3|iK;P`t@g
zZd!a7))M@Me#umJ_-*F3Hy}{2Q=}&OP08qI#;u@_z4ZPJPrU0%llh1;fhNyc{wX#R
zR(FQcr|c<%s8fh1|C37PyW0}n4DDDWy9~pPJ_)#Ih*WRE=!#!lu$R6P2UToyio{`3
zkU@|7MDi)femOHdQ2^^GeskAxqM}zsVqhvGZrjVB<~gCzrf$O@Ry)-dF6Yy}P1*jo
z`Pcvl*!U8dH!frjly)(_AL71kxM}=$%GrwG<yi{158OvfqH5)bl~7n!C@YYoqy(X0
zc1y;k#svCa`W>j?qiDJ5ftAwSVf~WewM^tDAhO(HCGg4AQgkv&RN-U=vO6z5kPkF%
zNVc={bp#InlKZm<aJGGz=VlHjJQw^G&^GQ0jsOPo4G-cZbL*)T&T<;1GN&aaKk1SM
zHh14Lw!5~E+4}sm8ZChUrX{s?#Pah}X+_|cxo*J;{No604J!Jsdd$PrB=JaRy-uC?
zoO7(5o(dxqQK!-hDPfJ)SC@nuRfr*1B+`_ki-bgi<Vq&?A(opA@*1nA5Hl99fMwXr
zG09qd+V$tfcH_0mfi=Az_J&^XzT9AYmapxvVp3YT7@jr+g4B*u#a5GT)@-1bY&mB_
z<8l7eW8dGYn%hzCkCfvA{y62@4?kr%YuZm_xM0=;kv)!dhGq7_%8~4lMOsE4MIbN!
z(3<coo_#!|wokt3)W;cOA8D`Gk+Z&{jmdP)YH0%Iz3p8(dB`!FZN5-9#`f9~+WpK_
z5++V%)O%dQ%Xn5NLpTj+fZvtu&?69k?d9t;|NkYCW>-i_pvP_lym4-)?bOi&KNJ5&
zVa1T`cv8Yxb~ac5t9l}Z(g9X~>{tlT;u(6m@D$ex(lIBJ-zYO&BSC)VM?L^77cQcr
zGF_x;Zx=wD9~o`+LN2GiTjm>#s!FPw?%1o3t-4CgVW+qOM3Cl>`SvAe8K>~qJ@PET
z34>D&%ArcQSLSwUJMO90F0DEOBKQ#UOT_j!<90o|;U_YHk(qNj&JGzQ^!3}9h}bip
z>e>8@VB=$<Bki;P|Nc??g3cSBIgzPR84HFToTjy{or(<@X^EY%^?1xfxknG#FZ8$9
zNtFfC;RqdG=rD2vv0AjE^SPUBWbfNgL(%7U+kmqF1~OHEgrY^@+=$3(w4q0g#eiOR
ztIuxx6GRObUS>+RcO@7-9b(2ZclyO;Q3bEOkIE|=XSK{odR>(oVdvYWCvb7o8X}@-
z!;rGj4)?7Sxv9?1v!Bjiv&rO%J!6Vf2at%9g0;Pnz4W=+U&m7W(V(3k1;zZa8?62;
z3zf`aU+RA6t;w-|ITG#ER4vk$uT=TBVgM{&>;h3^5<rbN3}55r^Rx)G70<3re{pb8
zoKF6>F>LP|vF#}I%`LrpblCY!u;pHp-WFM5Y-I96?SjM1`8o<u@s0!QU5Xa81W?^O
z3G>pL^s(ZzsvD;|i!$>RH@%Kry-pkM6guX1awn8WT3WsgtS6qx=;h_@M+LCafHUxq
zY!IMs3QQsryon?9q;yfot!XEs%q}-i@WsT(bmPhE;yV029KqO7(NWa8e*Xm{Cu_|R
zN8Md3WURP2{RPdzJLFESQ_W>9bW8CCywB71O7LWtH$z$HN-wF01lj4-W@VKt3{uyp
zKwL3ajtY5tH}iuhe9Y}-1A3Hzy*YU3Jtxo-=%-7DptqG*6(;~xY6huHE`?Z1(!^Qa
z!MXCj^N8HkT7$R!V7saFSTK!`!`_p9^&<StnrsWm-A5+ODiZ%quV0KS*-G4cbj*G7
zsNws6|2&dsU$8Zf1#6WN4?IU8eMZ^}IFdx8jho3(*09N`moTFI!B~PK9Bl84GK<kM
zBenv;yz?IP8NcCRVAV3qfWkHqX9a1HJ2l5ule*~(hdNkNCC99&sQ42RV7e5?dBh(-
zPIYiaB+2U~JKb_l1c!m+3)~Fi*#^5<tA%Pq8FK<bWwf$Gsv}2IUGXULVohCPG_(8G
zWaO>T{NVgJ_|q-(#BIx|5wDYTCN0rAib85quW{eO2#j{Zn{2S^LT|EU3OMf6l^Sm!
zW$vETC8{XZ>_sa@Dn)=u$@j)Z7NZmmL}K|^2|BqOmo_EF1g|1;@>v;yXm=)cX`Im%
zKcBqK*M+*VzJrT(?)`6D!`#0Qj$B_~@E(2KO59K&^PS1;kId@fzUC?^ar;Dxj^<VW
zTOw{Omk09%w6V!YhRT4HvAWW@hG@TxSYwVf$x9YVkp5eIEzSZ+IWw8BfcY6sgt18u
zTu4mENg+*})+16B2)P>+QAm<gpiw8?NqJu-xzq7WZMJF&JEZ?uP(%~z%7Wvyc62m}
zc+#3v3WOS2C9i6of`eByCEtiV3M8iIGhtbZRu2ZOlFiuU8RPiV4zI}LTvp`vFgwq_
zmAj9#$TvDA3)qn9htfy0-cmuiz9(#d*tX|8?MQxz-6H$}V_#=-&u@kS&i#giT8pf5
zsIZ!o_zZx*knI@h{9}+HVm-i6s2$9AOB6m6D{c>M%{4=Wn{gX?Q`L^>8L6wozI+;q
zgqr{%`zkO?PihBsexADx#DsoV<m*-bKJpU)E3}`}2I^6`o30!UjV!pmi*q|3S?3p%
ztb$#HR;LI_?KddA-8IY1{Ohac_68srOa^48zpYVV4!d{gWCUw3$D!vs)L8^@MYh*c
zaClru!7a}Y7htF@&uaTO9Cn|5DpT1DesqsO!xFPQoaK;`{mH<J??zzRIB1QgB9jUH
zLM^G|9aBM(PO>rME~amedY<1<D6)w}kD*h1iqML$DaS)J!=O+RGW+eaaqTNyx2t92
z;Ws5Z0`N0<d{2;sF-7jcV-I3PM_WGhW}k(C2vl5*F^s&RC3CHc?`>H*c^K2Yq|dkA
z_-z>BJBH7AB*y3DPbF~gh$^|PQ`KWTO3;aW#<e=^8QfR%V0_M<5Mnwp1kgfH+}Hcf
zJ*I#emT~_vee4_;1zafddym?;M`X_a`)6I_doL@{!p$3*U}F7%U1RCrSp)C`fP@UN
zk6&i4(*P4&3uG4)2*w1Hzm2=PYHiWm#esK2_+pSj`28ScvsoY1HJ=`?42{+)1GK#^
zateVh#nIfPebyzRyqCxb*X;yDDg{epq}FIMLwY$M7R}yp#p)`>-Y0L&I+A^4JJ*B^
z3Nm`C#g9bcR~2tkJ+wMweDuy)zNQUvtTER^Hd#Idouw_aGPd|IRQlAjUgMGG17yHn
zow58NlKW0f0!Nh1mjpy8l#P3ujvhfI3yQ=*t-7HiEMveglR^g|rK&Hfrg!nNHfS>H
z*YbA@JIv2do(ZmRjbNvb+UXN_LKP}lE@U?6k5(D(T>!Q3Vt3?TUq&M%H}E9>_PkD_
zuw6SKmZ69s&_&?*1OCIrK5m@P#Nv<tmG-VtuEuv!s=sNFVNNU*5(?M@L|Do<%HOLR
zmmaXF7+%}gH;T=m2eWdIu>zC~SHz%b7to?-B1G?)RF-Xfxz2~#8yegIzsp+@(}9gA
z2|<`DvtK9Dx+$rSELa^@d)+fD3Nz=J%ck3sHkXZ`AUmYMR3$r4gv=E<j0m8)ebKJ}
zy##21E&19b`L){tP8A)`|5rh5#a0H8(lxmV$|sBNqan?%2vSlwQ$Lt>T#hX5>i{^D
zfh<<rl8p-{PXns@prXLfvSOh6kL<hVo-=pgfyv|jM;G#+^xmqu$-BM=zBAKD4ByVt
zk6v(fHvlgJLxC0mH0Xf^=Yl=)`+ol}s0HbHx-NV~x99({=$X51PrmCG$sI|ML0|pX
z1qO2y@He{$X!7;1xIC2grs_NYcvr{@^dj`Y=P+|a*ju!ts}K@ki5B1f8~KDAK(h}s
z32hD{`}WpJmOXg&?o_CBnMxldv;dg)fKNwerz*i5>`XaHHfhR-%3g&ADt!Q2^0NP%
z1e{!oe!TfrXTd9r_65Hh<FDW0I&_x^G*1%QK#X02vo5h&kVN*8pYm3WoKzIj6wr;B
zC>GQ&n$;mG`awhULe)_1R3L}u*qzaDN_bJURvw<068Pl8`?!7oGd7@C=*R-dp=7D%
zuEWD>E1aMHP=)3ONq1RhWgD7xW#8m#yC)WksIs&PmTm}N<s+z$^%u&bLI&>IVduev
z^_zoyjh`o!L6Zh+eO%cjT;18#qsEqRXXt$wc9JPUr%LJ~C#jm28f`W9b-k-QU9zFQ
zZZs|D_O^Cbg2(4MCwHeYm$!$H-j-%O&Hz|W2qIz%joJYS0&0AIAxm;&n<_fKo-*iY
zWqNQZWS8AURjeK&qSU}ITo8w>_1;L!tfD!eY?}zL?ri)XE%{;cL_3>-6x7i=N*I(h
z-j1fVB)FikR7t}&9!&E$;62X0e5$JZ9s^Ou{|MLe+)tqN`0E*1W!*Gz@5EcIWpCK4
zw0Kw&1B_t>F%RcQpXDLSZMtY+Fc)B)R7)>x3Oqg=R@;p)^c0Dl%)yQ`!s^7$_|CW*
zRYI<^GJaNSpYi}8)3WAOOJ}+Wy$!sfQoZl?4G3AK(a!)G8jnoR;w{IN5dm-Svn)45
zmod=Q;;Ant1=SKXbKRz!cCubcMkuS=4on3yqA>Vts9emOHMB9`E#6*Y6x&BHv!9JP
zrfri6@A{TN#9dtf5bW`MgcWEqZ!u-O+<IZ|$l>VgV2}an?G56crS=YVEXPu52Ff39
z$RQ5YnGGW)k$Z8Bt4NRK&rV|<Iq}mASpF!3uyD|EeE_#tJrb5dQF@1uGB5aFuIMek
z;)CZ*c`tHO$6|Qh3nHU?uRQB1Ydb^x7w17tzJfb?pS211cCXY0yoGz`i!r}O1XlQL
zRm1`Xt$ky1S?yv|jI`uo@Khx&{Yc_CafNf(_^Wjy1~mvZmh>a~;~(GxZ?6$XEqWPq
zvF?`RLo3=@&RY)E^vL$YqxY)L*#gXxA2A>QU;Xq^m?{sJ8x$PZ3N**v5Z$c>%b7q~
zL0)NUsWaM_4s8VzQhy(c3?_Vg!LEULX<(IDCbvY@*us)VlAV-D3f+R}-p|)Ro>zoV
zux(VDF5G?x@F>zOF*C}NwSzj{H#eTZ)2f?lS9;#r3c-sPa?D4`EcUhbmwhjPecvi?
zBkE<~l{Sf1O~qmuH&RDUTt0CZ-y_iewb`irE+1z%?f?v*LLPU_>}prSV@p<-)-ryg
z*|uPPiU`rh*fb%UzaFjBsEf&WTZa7^^ZR#qSjQYoutG5?BjBj;yYM{{;8yw4bMhv~
zE=SBMgNNNoDZ8lc?PS{IgL2sNoTY6)%)TfkGOT37(X_3~sCl^#tQ$0C>VydJF-fyn
zjCKaSgAB#l)}2n2sNSWw`^a;PJ$B4dH{w2I<pV`a$J)a`L%T;jFQ=(vCDb1-Y*`S3
z9sH@Nm!ih>JpqBtag61W?*>%vW(GYUAD;4p+l7HS53efePA)86q+Yo0y5r*icAn&p
z3)|-O$PK*RHm9hXi;zNEC4;rXWzKTDM)h@b$M8ij$>;lGp}pQpRL320QAP<+xH|h+
zW`m8z^9-JkBBol(7jkyWu>o7GfAmPsxysEe6ijpXizvH|)Th~kn7kUffX^)Oy)WY1
zj`(vUnEumlxwe%=!;ZzV*NMEW>fwjydGBNCsgD3;u~!>&1jvEvft#?$VzpYHt3sZ=
znvED)4gWl#{84gZY_y2)M>qE_%HwJ17RT;TBEHxvtDTrGM`gbIRQ*?)oK2VE+vlwb
zKr!fFF}z^lC(Um0M*$cnVi~5Koyr~dHP94@Tu~D4a3$C?0`6|o$wzl8xGBFmYU-(y
zGu_YBC?GYo<ny(DR(#=Id=-$g9QD4_QsNDt@k44<inT=2NZmwB?}4ZsqDjPO&wViA
z9p%*P3A71z^MdYm6O3z<>7x|y-j1&0)qhPSXdbk2mThDgM&}dg5g9;sJ2V=%TQwbv
zGp;x7hsEUu9xIgaJu%tzPvUsCOH{W$-wN}5%<v5eC7?H{GTo|_BybbgFnhC@=L#TX
zC}up2_vJO)lC#%0b*@20Y02}@bg02b>Juj2uT><mpSw}?V6|a7Jl=ghRRDl=eL1@(
z@{)yMLGY_U_tVJwRO|-W%Rv2`drj3H(qGv#6K+Wj`_%ptS7>F7*#CHn%D+swwI%rA
zdi^qgOl?5%Cb4Jbv#b>tNSr3>98~e#G;%jXy^e3WBTGTlq7Cn2^l3$p6aBud8vS$y
zq(W7-u^>C}&A=rx-P1T%GDF2uW~61u&6uQT0vKf;UClVWpzD<CG!`N`eoCZ!p#_b;
z1itsYm*`O<qBQK#D<yOabCh|C0sn$<PFp2fl>Y2t#us0Qr*z+XUBO<s=uHKOId!Yq
zyPai)XEQ5ye!tNa5jQ@uo~F}U^vlpLO{#pxLGMWNdgId*AFY+InGJW-x#y{GcMcx;
zJ$<21vJ0*AmT5oL{=EsZSU)ZO?ptbotWHzppL>xRsDY|NJSjU*75<RX)Th^Ei#-+h
zDG#(U<?-2x&%sW5lFV?%{rJjx#`s<n-^(^VQEcd-I!?JC$2@P}1SV1~y6fF2p7zvr
zYPiF`Ui1BcHEpFk{+TE2?gCA?e~Ll>{iCokq6!44`D$^3yA2`{WbQ}jpqsw9kO#x#
z1LNkTN|T?Dx{j~Mo@cluAOaDA+5&Hq*E(A50UD1`xR%HFM2{bW;2XeA7w9TcbE?23
z()eDY!IR$anr<ZfH$O{ubc84(p8?%^&VT>R75~)hY<H{W7mq-1$&7Ik)yc|hxjy2d
zus#@cs^qdKr$kI1JViI-hj}|L0?W(us#+52R}N)t8cSfo1*a?<E>RzdRNO%`F}SyM
zw$rC+ZD+jUcb!ZZ;h^wi(s^&O@^t?o>!c=cQ%cy0@j`VAyAf_)MxR6~GT~<(ujd#U
zZ#SD>m>x77<IaE_ht#97nwpy00O8fciz_lguVU_6WTG@JnxySi??xyYxsjzt&}`>^
z8SK%wlTS@zz1bsF3rqta;2Q2vCpY0}97u2l#MiyM)u)W6O-{TV8phmc0M4hzhSo!e
z^J%@gNV{wQ$klR~{;^hbLj~j_ZQ~nz;MNaw)2(L6^CHtIaR>60uEKx+u!|oWTMz%B
z%7jl~73$)bm>)$iUVm=UPY!Td>eeWvbyP3G|9h|CzkkGK8qV_y&f$jGP)0N7-fAPl
z*mwGMLANvhO$*=w&&oFHQnc&XmV@eN64DE2$`H5~Ouezw<I~VI1^1H84f#*LGLgvQ
z+jIDdJAJ(ih<?@$r=0XmJ&!wTQLI&jiD4_$ozgV)!FdHzbL<t=q!4-8;nNf|ounb}
z2j4pmtXHKBxACr~Cihp8baGKYN$Liv38ni#y5(PtL7Q}5!EP2!FVv?06jxBxf6nHu
zk#gzRrF=PHTjEAHzvzbV!KqLok!T^pwiClC>)htZFXMn4di6@#*z~yKU$`60ivZZ0
zXKm%be|-7cDYX4Q&a8=h`C5Y03kAQScH@Im@p(-wvn8i;{k{D;(1Wf~NGM_#xbm`z
zh$z*I=hcTV;ILmMGVK{>R-`oRFyebcVV_321Wgh*bzyDDXZBtSGX`;4L8}LWF1gSc
z80`LAFl|k72Cy@%^H#zCDL0^%e&1T#<Av2_|5BAj@q)>rFMYt|?{!CAG2jLDHCwjf
z34+N+s0hSwnmvoDL6In?)oiWoB&IxSfM~%)YW@otS1g<3s;P&FIJxY7*_KrAZ<z^E
zGW$SAzmL>50%o1MSml1`XKX>Ov~u&P7`QK^C`?iKw~}msaJu*sOV)0xK9cM70w!`^
zRWyLAClZ4oBHL4ro=<!RzMnU=Wc6aMDSq&8IMKTU<%w5`+D)B4uV)$^FRI|y0bRHI
zvB_Y!rG86D6n?Ba=uz*zLcJy1TES$II!EGhD6I91_8I#OO~f>63;PX8=n`HBTYeEl
zjSEd)5BVmg&{-k)(<n9N!Lw=RNZhp_cDly6fOf`Uqlr_V4<6Y4TXT>X){lmK;xpU;
zj{0)0k~8DR_rY_yn@bm{8qUHx5ZeV&iO?(K*?OUe{t+7__RSBIUA*F&F@O&`&y{~s
z6C^O8b(I~)V*5f`PSo1q+E)hC;VvBU@<m#kqr6UqDQ8z$T|ob&nD~)tm)z^p;<tI*
zt7J7RtdeqIslP*^n;5`8K6U$?-eRi@7rqPs8ky)p*2=lNbaBe=IrS}HjF$f)Zsti?
zs1xuWOu5`u9=;_wG4#~LQ*NJahSz6D?Su>Kq3?CQK;cFv%X29mu=C8)#k6bp=RQG<
z-#e5_w0s<WW1Jp35Eq<!Qhl<-9Z^jvkM(=*%bR%hk;aX0L`R49RKjiBzSElk%)#Ld
z%|X4kMUxDJvdM{xqfF~{r;vpPh0*iuz^ZOCsNS`CL?WVS7M+M;-QhIKuAo>x7`|ee
z&$L?an)v+IE$tpkO=y<<H5oasFMt*oswKh&2Lguyi&9~qR2)}fok!+O6P*c^XPNa+
zEetGW__!Wsz{$jVgAL5|3#=5(E(q{hWWtyh5`1wIaP?U6bB+z0;QwWL>bLvwv!k^@
z;hydaB4+<<>gc`9^~D>`7xcLdLgxALmD1ciu!9vqmpq#;G`_gbrfxB_Iq?o$pu=X$
zsyY^@qQhdf)6cbS9k>^0rV^O@k4YbJCrFTh@d2P=VH-{@fP7Ni{%_~9$m~tJHSeMP
z9Gx7?hT<us1SeUq`>pfc5U24xySehrk7RiKN3~x~ubQax{ASIO`zW`>Ke^`DcX)yn
zqWJF$grGsQ^x1ImO3FW2F(<0i7q8}i(5odu@$KS-3#%nV;0*9QN+cF0n1qDBrbQm}
zG0jyBp;Xrr%mqQQDfjp*K3=94PmCBmp|u|MhnRcRAKnrCGIe}0J{j$~{OBh@Gj0QL
z48BNxk=JN&6u5U0+VWK|$+uaiCrmctyQMWQFMPBtMO;u(OhE2eaOUxDN*J%Y6@JFa
z><_~=nI}Z-n-yXTr2Ngo&lackZI``A6!nVk$mgZNVTQd)M%9He`jxTAF~!++)%#Kt
zdt^(IxffUa87y<cP?~#O-rJwULnNQUM{7OR4_{|Pr~EI{3LvHX8u3NFR3FTt>%Q`5
z+kEoAGuNJ^2jY328`y?3ml%+(TgIz0z?~zgP$+xn@kDcB2?I+S>UITAR;|<e_$JGi
zbiv8>q`)5XlxFJ_3V3-g%w$t)Avu6}p8dZSsbj%*B^Lp2?Z$iottZs53#K!3a-t4^
zx3%yy%b=H&cCPE!^E%e4uGQjxQz?EbTq0r|_R{&6FBP(%B}h!e6u+Bi#)lACT{BP1
zgTD<f=B3k{MhWQfauTyF9APZRFfueoclMhF_ZY?NFYJ;r^1(1eZReVJ27PsMAB+_R
z2qC0}liIEI^)9c|3WR_hnyYTFzhaLM9HS3ec+(CVQ*U+tKAmLk?E>b1TD<>>ZGQ~Z
z;-wKnZ?|tq42)TURR93o>?2lnes!bmjpAdmK!kdJ$c#U|Ww?=>H_@*u;&s-4|1>O;
zGaegSu_|VBv~+^_SoDcbE3n^!Z2f<9>l3WOF34Fx{VJYUZd+K#^GUh3=A-v8@`7$W
zQ+iCHlTB*0rlX!~B3#-6^C2?P(@Jq=H#BwR@<sR1T=#$f+$9|OB^Im`^k0>1j1k;X
zx$NBd@+`~ll2%4YDEE(uTX(<QwvgtF#a4J4$4$uuh6tTAZ)5{H8wZZ{hv!U<8gMu?
zJ2A<|Xj>xkYZ#0Tma%-!YU^=cXzpaaFXpbRojdCu>^$jm-Ht0{+xpIx?2Kzc3g1wj
zT+EKMuQGxhKCQz-<iA{)lI~bTCbzX2hjtB5#CnS^a2^9T#J%L3bYE&3PqCngCVYX<
zC*M&@#jBp^Pjug1byk_&|NG~*Ia>v6P>^J@MmHTik)F^fqHJ@o!`RFD5xRGwwD6(y
z1*Z9R9&Ym<Riiq`qO9(4YlL7D_cRY*T;m4WpJompoxb8ZcA3D&zQ%7(^borT$FZ-A
z*zM<>=e3l6AIeCws7b&WE<&<PG2698Zqo7sW7~5iANqmC7ZLnyES2RNm}eY@XCNRi
zf+?WL%We|EdXvyBn?~DLDnCtjBQo;>P$>iUH(VSmRMuX9&{5)O<Bm)Tsp@CKyMLP!
zaqKH6i&0AN)Ao>`SPwa8>#10x3k`9^Nb?w~Btnnq9!qW$OpWk0GTnx=%JFOr$kvt^
z{_R-&r9ia-u?mEog>}!-c{&-m!y4y8g7K<vm;WHMct$Xb`GrwQCm%WOZMvsaU(Q6O
zpK`1S#MJ$2aiaeI&1}Uu+>6mq*Y@GP7Z6veV2s6~=n$VV!WJK3&bcc(3fp977XA>J
za}7Kzm;Yj%Ef5M7DQpl-S$Ig0JYXKq=9)qJu!W%9uhrf`)L3zoT$%BMHeVIGR;i+$
zvD%;Jy>uG{??xBrqM)F~iY^rn&8rB@BOGrPVy@bkkW7U|>R(}0dYsJkeh@fXNy;IC
z1tUEnIFibPBSB7<iVB8c=!f<vD84a&;A@8KG_TL!7GL5#&%U1cD5=1k9WOQ5ADgCS
zF8eI9VIQPiO)o}aLVcF%`*hAwMj7c!CF6UY1kCX(Y0HpQ{kmB)tItoEzJq_FXW_w>
zM1%3A3)w<rVXAf)3+b+U+Mn&(IDM{^|Ne2wP<`Ex|B6&Vg(UO6C{Plw;*DpH=W8D2
z(#$CGZuUh>2KYDT(n_k`RX%TS5{JXzaop#%gGY^qi13|iGbBvb!FRIAUm*qZEFbbx
z8^7^ER4OMNv)DI3eY4mYgYFukjIN0U<6g+p`yJnRkOj-)keB7gr!w@v-K)b2cdr3i
zbh7Kn<B@%(7yXF_E$HvJr!`rr^O3Q+EWlzuHMw&s1tKbXL~kirszfkaTlP<(yEZIp
zW?P2ybbzwv7fKv$8wu1p=O(GVN4tph`*R|G0(X{6DumS0k*+qWFU)6fjmTR&o%~8n
zx%R&m7TV_b)nzNT|97A+uR8HGkyQiwgl;KP?dn7Yhhl-NVNn3o(*SIV;@m@sZrcRI
zuqn5$eqQy^GB6zE=t%v*(hEjFyNxo{r|Bn%?;bRHd<IQusDlT&Sh};h$EtGm#iueN
zJ@zQA(4bMgTo=bq9p9nogy%*IndY=do|at$MUjU^v2N(levK{K-dB2}LO9i(KKF$k
zoo8qnmkGk6KN9y*Xzh7}ORADD%bD4h_N%@PYR?T5evw)9X$;K2Cd!=AiPofRDlfSY
zwiTSDeRYX^C^qxC$v^=hm33<@^j@wuk5Kv{e}kJ=gaL2l>z>Yuk|v<#3t3l}BLwc*
z$>tXO!1c!V4NPa2_#WH6t-&EGyhod~8m{&g>6^h!{NC?_Me&BTXa)4Os@)UGneU+%
zMKZ1{+2V_;y=m{1ZC@Jrdrj1`)dkfXmqnP}a}ol>Yr{39Xp3CM0tcSjZB;F`mgaB?
zrVSNZR3~4hd#)isOeK(;JD9nv;P0I`)KHo*6EzN}(LI;Xl*whOzQ!H_D~AJLN4mSV
z*c@}^jmE>*cg?yK>zjzNx!W1}NUH<}V6J{ub4X|c$hM*V@1L}4G*IX^^9(eN)HxU6
zFeCb-RL`UMB2JF6K$$OwG4fr&bRI5&(cO-8AI+8cWDF{L14LKFnPrk<J2kzI&hyt;
z@BI5=*0*EdQO4{1hUtkyXZnKrY4`9<@JLsm(5(zCa_wR^>d8X&ou8cQrre#b>9J|~
z-Xq>LO<1{n_N(<8oRd#HZn5>&pg>m_0<jkGeRw4+UH-p6E<LvIXs7L%U`GomLo#(!
zJF8xnDf47!6yPoXw4ez=3x(q%A|eLf{DN14L9_hLX;_P(#Vb8X4L&=gk5bZ3?ggX9
z&Vp>*W(u$-C`HrOC=S)Hl}U07JVmWQ3N(dt5JZ4$oGoQzB1{3qt}incELl4Tbr3Ke
zrntm+Lf^dJ3J&_yXe0SlgK;1pYO(>#z8Jr!VqQXogPR2Tu@A;bO*?h1AQ~!p&)vVz
z@SHiU?j(fR)uZ~{$ftQ(uFZM!s_mH3x6X})x*%nW7*_R{sIqcLCh|4rg~cc2!o)rq
zme(Eo$*{WU9FcG4XK;V-tx`ZD1^UCNU}Qk{Lr2=HZbZ1{FT*~VBhf}H=A1{oU=vig
zVsNFu#w-isD$b&_{D&$Qlrn_k&**HmabbSGqZKXiAvT`LCrX+43Y4gCrsfe8dt+Hj
zck*V~93!b>V&w8trNmog$awnEXTYD42%w@9)etX@>F8WQwG|8z{})#fsPD*=XW8~u
z%SG&pi0FgRtry`5q!Nbu!VsP{tfC_g+B~Fjlhu|f!SzHc<&SaiA#8}cLTSn%A=xdH
zXMv`|xYs}AM?c7d;qKp~p@`sD2cb&cBha*d1N$Oxe`B57V^0l{*mO}`D5lE3j|m*9
z`yS){Si6Getd|9ZI7IZcLkPhU9Sso@m_!&40XBl_3G`itA+^Ar@HErI!#tRfconGv
zNWl>oUpi9hJeV4}88vKV8X08sYZ|@>SRm-zq4+u`@MAfG+!twNZo<IQ-H7b3(Qjb5
z5WGa0MHE-UMcEf*UJQ0##qFp^N2ZE;9))yPWXY2x(Fj6mgo=iPPb2Ud(A((rC;rKx
zV$09?^YBf+OHM}3(SH*&iS9?~qoNMmWgL!Gcv&ylq;4=~@aaA6R(cfGBt}lyuK0*)
zb`nbA2C|Wmce`6nzXIfDNg0e-RgH<|-4?y{-(+g#w*J{++N3tlU4J9~@ZFK3f7sVX
z;ksD=0I@>6B9CELM#jNW=*$`*Qc&Pi#f0dHq)70Hk9qp10_&W$F;k?{Oh2Kw2A{z)
zpyY*@{V1uN8iS6a$$o`O)nawZRg8IGH;<Ee{EPFi@E_-21SKNJxsJ>Nd$X@3_AjBg
zgnCQO2>WF6^hZRT5Jb`wSVA(i-lRzKB3zvbo-B9{lzWxrr@JGQz$CpFvE)RF=<#9k
z4ERLuQ+XP48s^(CMm>*Zar{KzWFAy3urEySU)a*_&+MMn?3T;=A#8f|CBN{`lm5nc
ze<Haa;de0-Ujphx^oT)YS)|f3G805yj&@bvgblZ_=#wL6L_~zD3m|w2oy3JJrvi||
z+-kNVG{KE2b)_DIZV^RxFU%?K#mGy9KEWU(a}h8|#D<1i#TyK`LvCe_?qLnGCvbzF
zTuL--bs{=F3hUrS>}Rq***uVhAqnH)Aqa?!7}x7a$)nizA|fz73(#s9FnAAEg>FXd
za=ZTk7k|q2Ulc#Xbn@wYrwKlVL7W(*6{Xw%0AZ76KHv0Nns&AtlXCkIq%K3m%{6O6
zP)lSXU5lXIGEe;=zSubxq?tKJQr_|4W%6o8{g_0`Z=_+d#qfHHzo88hrg($Nv~lc8
z_GE12m8<NnF7%xX^Y(j$rm%bJg|OCF`-2*&Sa$juxMnf?#*ICUgU01cQXBT*NkmSb
z2W)3A(99C-Y-5}>=#ogb*N>C<zsUap1OEU=UlmU${UavqgTRE3gd~>*()?p1yA)Y^
zcab9q@+}{6;@S@kJ;tAsoU_UQ03>No0w;*WWKdQ!?1@aYBINl@_K`MWLmvoF6Sbnb
zEpssB&jVyOH^P?LM$QWSwf0|s<aPz!h2&ya0_Y)dh`JIGi1HpvL5B>thoF~;^c4A8
z5z8<rmn%XzE2F^6<W`7hsZLKJ<fIwlq&5>p6D!bEkxiJ@320j`lTQJsxrs=GAql1r
zvl{&$UkWz+(T3hd+!8q(SB56Lg#G9=<%}AwkeRT1B+fuxn;cymUJ#p%K_5*>L_}c0
zg9R^w_$%nX1r$LU89b1PnpR}VlO{~}rT9Uj40N9eANGH$Y)pUjY?APZEFo4c-ewK#
z!QNldC(2BZQ7m+ZEExE3b9oc1MvrQ(hUO`^vVZ8Ba|a|+>6B_TghHDi*mmV2*NDyc
zmISD>{{T_WsT)!vx@11xC3ny4J>r4C(UaMhzQmzTScq(VIQOBFR>rl+XX!8OdzODG
zk!D4S9IBg5e?{w`(CH;mY>FyQ%e!Cobb9hX!v6pY{{TsUnK1YM!eo!5z%)!RUVA0G
zMjWN~KGbzqc(K$x13r%_ZP@dq5a{}7LgAh)S7bKieoKq&_B@1A^(x?2F9bE+eGNWf
zNgZRb$F@sk$>4PxY@pl8IkCT%UoDAF!F3Yh5om_e=zXYZ<nS>+4km1fUTGBH_>}pc
z;R~1hBJS7hZF(ip`oDr3H|TY<={#u-y&h97*pIeHLQaI8A*Iqb=ue`)QMpP|vWodC
z?P+>5(33Ax6NMeYlBBOBq$Y+f9efG5B*6;IneSW})|APTGK<j>g91xKM9A-xE<~M@
z+=m1Yu^P>p1faF1NXSG)L}0;#1u09WesrZNU&?%_#y73d{zv$W{64kk@bD_xs%Pzw
z2`-WmSv{MMZJ1Ow7AF&A+u%~k+<&7AMX+fxXMj6F;pmKtztG%2lo-=yf*_XdHT;XQ
z5?}ZcF>b5+LRnQ26hmV+Z~p+mY(9*FEdJ{LP>`+=?*^WT-5j!Wsu?-JZY7Rv3evS`
zjmv`ej6J8a@2#taOi2&iI}1S!T8}egaGJ>X_$|k+FX38ks{a62*a)%oKhfwjf#OAZ
zNYDNQX70~sLE=PRB2BcXLM!nPa)I_|V<!z23l+F4V4RPC$i@Ev??nCNHbn4_fo+9(
zgH6b<A-64vK_v7fCxPh>Q9e-n&V=95Z)YtS{SNgjmM4(VultocBggP%!!yV{DIRo+
z+{*9!k8}GWcl$l$9>VA(z)A3X5iU`M$}a{~M?!akK7_=62qx^JwAq&?e2~&gKkG^K
zMe9Zf;i87cLS_tX(K?JpkcCW5Ey-IUu=z3!k+MA=#9fgX?5L=MdOc4;w*v|ehu9|U
zMZm*u=qD^~vm_-bcouV{A{M1H_^EX%N;mjt#_RNFNI{_nh(a5Y{K5YKo`>@1@^CA(
z!e8x*a;c)vh@wj4tc`Mvl=>EGGPx9olxywrYxF*_?ibk?EOKDlmGUJ{$$w%TYkVfm
z8!kK`?M@CQ#1Z#Yz}q<OG@q(|zx##F`H5w*i#MUOc@Kh#v0Od!dhs!eai7^v+sE`v
z%Iave>on>)InfKMA|-sfM9+hqir8dt{)mz;3|+Ppnplrt0{N*mkd9lb2?S}rezfOE
zgW6<cEi{C^WP6F~hob!NVSLxY{#X7>`D}&A(FX7@M1Bvnj4n#9<9(itc%<|u6xW1`
z-6P3T_IX#PcDo#MNS_Fou*DKQB|hahWs%5`^%2mQqYgA-65XNni1Yjyf7l)DT})ew
zJOp>iT)*t>{zowe)nYPxe*rH6yV$-Fevun;`vtw*5^_s}sSseOXx*ef#Cv~2;&;&A
zZY%7h-4YXeDZv%z^e5$>nLM}I<|Vu;Z3y$Fmk;z;V&4aVz6X(hk0LzhkgO<s4EE2q
zL}Q~p85(5C1c*i@V>wJtuy47FGEC8@WT6}j6iXYI!aW5Q_D?TJL_!ds0SGa-(BA|!
zaLPiy2qEx&2>K?5HaB0*N>Y@iGpr_%!B=RhZqG61B%Y6PLWib=o1@NOWgg@ac6%cB
zLti5{g!Foq!b9*4*%z^2W_5of{3re@-(G%C;G=k+ODXY}*ojdgi+WtTFukjUJPc`N
z4W#xY{6S<<LIvPawmpn&l+sPd(Bn+^4jY*EF@-lb*qyn}9j7O0!&VJ_x^9bc+`C2k
z7l^5$GHyFewpn-{aNtdwwM0tk2~_#P8-mBhC_)KSI3_pQ3<V2gDPp9QUj`FVFP!{u
z;{N~()&5ub7x+`;oTgsFF?bQ}3Y(eoJ<RC8vKuGad@A%?Y<flPBY*TeAM$(YN=HIo
zguM~ak|INCa)zzZ9&_G4mS?jKy9;cHuPGJZ^)O}s0A%<6WWVNmpV2+Pi4C7iO^~+`
zppCm7jPy$2Nj5}y5~6n^ZbOuLiSj=&(B(K3<s-b?-H7i00Fg15_9NHmp5M{bC2U^+
zuEx70irf)x*q1ZSiF~p|I6|EmK8xc>I<$n<GU+rz5t0yuGFCa@L?HGbN0K`cSdK`K
z11|H883{?`gd!sAL?H-GDRiYpLujaRi8@3$A|mL)f}#lQsqDr~>-IrWb*PTW()B${
z>q3ayPayMbp7gvYjSzYf(dc@Y(dyJG$)3#i$W2H>6IO&F88jgndq2_F;N#<EB@?tt
z=}8n4ZpiXQ8NqZ)K$NXG8nkbLk6}}6f>D_vVQDi*u#`Cv)fUC9T66J#25@2VN?(fn
zZ|3ypSN;R}FTlUWt-2}L@_!;{yJAvB&@Q9Ij{+hfq#H|j3Ge#`ul@+~&nfge*~Hz2
zo&u$^7kNeKw&TnBos0W9$VdMG32k3TLGk^Lr7P0<Jmh`FIbyBbF%on~NJuzjf=2Cr
z$nYO&(3=zR;(p(`AB2l=J6uW2limAdP9L+e{2dAYj~5%Z4g3d@4EGlI2bqhi`ys+F
zK~9EOrc9K`LK9LnMon?>Q?aCDM3*8i8Bw_JMn^{Bo)2Q6#q%O-cd@!Q<TIU#x;%`O
z#GsNsrVMFHT@mtSW_8D9MO~5EE0HH;b~SP|oSh6;;7#40OiGv=NN6>NOHnq;mf~-+
zDZ49{FCs`Mf)n9NUyLW`nqa`%8$$>lh1Eao{{TjMug3-~yTThbCW~X8In{5f=kTp~
z)1QkLi&weF&3qT)L;RCYf5czr?u^P++7SN$=uCV607Tp~8*piag*IQo{{U!Lb~#&q
z%oVNnC2TS$52Qo4^iOAJQyoFSV^0JB0FWDZVw;8LCG1@UiEu<gQ+9@)1dkJDT*_=+
za7DRDT%)o70Ar1B?1z7{D3-~7gs=W5h;G_Sa*<fycN22wu%8I%MmqRprY2?`2?!vI
zp`sW(i1Q!zno!4p@DE0hV?B_BAv_S8-vwP5d<d;@Cf@^1laYrbYr95EbVlGu8<I&c
zfi2*AiIz~X(vXiLF997DnIj`3D^I?pA|fK}$7J$HWWGozv&c@=N{HY{BoHv<XcV?H
zV9r?RqHIaBo9IY|@}+uYjD#UR6s0M2r7yrzl%*+3QL(YHv_S-urq|lP(fRyHMH(+B
z*~X7m`FZ&BqG#dA{tM;D&3qU6i~Mz~{{RtN$H-@)T%-FtKVXjPB<|m$JT`fcXNsLm
z`Y+??uP@l$yMD*G+3ZVpE+HvwB<Nlu7(BF1$UH(jU(r3^MO+28f3Y?R%#mHrY^Q(n
zC%^JB<rCpQq8l+h8tgB`DZq`m?{O>LwtI!pc#x9mX;~RGCPjQhZi>4{koFPNEO-#w
z^j99tkG6TSAtFN^M244WsmeTuxSklXBjF)FsPahkH*9$t#B$%EK80K+O6Fe!z>w4O
zLU!$w$sK7)+3pgdArXT7sY*vjsa+M(d^AqQ71>?r_8^GCf})gdm5s3lX(n#QF|i)R
zIU9Si%z{;<M1;~qDd_}=V_H({N>Y@%Qk0}>R#sA!rT7m*x+|iFCt<f^*Fca&e<^>8
z^<U(qHZAD#{%c3A(fl9HAsp++$!N|rdOwoUh+ma{XYpU?FYu+qv9FN(wk}F8fg^!+
zAt~@hB137z_C;*^A<ytT;=YEVTW{npsxQz@-hPiLM(zICle<37*X>H2uh7x|00J8a
z)yiyJctV^C`?-DHi?d?#61Y4^fe|fX8!JTAnj%}YLv$Z)JWP(oa2xnXl+Te(+t84N
zAhZ(ldlfdxdSbEYPYqdIvF!9fmz1=;20CP2X!=JxX)i>h@+X#Ov&2LnVq*E-Dj^L>
UC!$H{&tzX(PZ4?|^k1X@*;Pv=jsO4v

literal 0
HcmV?d00001

diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index 6d279bff5..c36b98969 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -14,4 +14,5 @@ This part of the documentation provides a few cookbooks that you can browse to g
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
 - [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
 - [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
+- [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
 - [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
diff --git a/docs/cookbook/receipt-digitization.md b/docs/cookbook/receipt-digitization.md
new file mode 100644
index 000000000..67830fa81
--- /dev/null
+++ b/docs/cookbook/receipt-digitization.md
@@ -0,0 +1,296 @@
+# Receipt Data Extraction with VLMs
+
+## Setup
+
+You'll need to install the dependencies:
+
+```bash
+pip install outlines torch==2.4.0 transformers accelerate pillow rich
+```
+
+## Import libraries
+
+Load all the necessary libraries:
+
+```python
+# LLM stuff
+import outlines
+import torch
+from transformers import AutoProcessor
+from pydantic import BaseModel, Field
+from typing import Literal, Optional, List
+
+# Image stuff
+from PIL import Image
+import requests
+
+# Rich for pretty printing
+from rich import print
+```
+
+## Choose a model
+
+This example has been tested with `mistral-community/pixtral-12b` ([HF link](https://huggingface.co/mistral-community/pixtral-12b)) and `Qwen/Qwen2-VL-7B-Instruct` ([HF link](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct)).
+
+We recommend Qwen-2-VL as we have found it to be more accurate than Pixtral.
+
+If you want to use Qwen-2-VL, you can do the following:
+
+```python
+# To use Qwen-2-VL:
+from transformers import Qwen2VLForConditionalGeneration
+model_name = "Qwen/Qwen2-VL-7B-Instruct"
+model_class = Qwen2VLForConditionalGeneration
+```
+
+If you want to use Pixtral, you can do the following:
+
+```python
+# To use Pixtral:
+from transformers import LlavaForConditionalGeneration
+model_name="mistral-community/pixtral-12b"
+model_class=LlavaForConditionalGeneration
+```
+
+## Load the model
+
+Load the model into memory:
+
+```python
+model = outlines.models.transformers_vision(
+    model_name,
+    model_class=model_class,
+    model_kwargs={
+        "device_map": "auto",
+        "torch_dtype": torch.bfloat16,
+    },
+    processor_kwargs={
+        "device": "cuda", # set to "cpu" if you don't have a GPU
+    },
+)
+```
+
+## Image processing
+
+Images can be quite large. In GPU-poor environments, you may need to resize the image to a smaller size.
+
+Here's a helper function to do that:
+
+```python
+def load_and_resize_image(image_path, max_size=1024):
+    """
+    Load and resize an image while maintaining aspect ratio
+
+    Args:
+        image_path: Path to the image file
+        max_size: Maximum dimension (width or height) of the output image
+
+    Returns:
+        PIL Image: Resized image
+    """
+    image = Image.open(image_path)
+
+    # Get current dimensions
+    width, height = image.size
+
+    # Calculate scaling factor
+    scale = min(max_size / width, max_size / height)
+
+    # Only resize if image is larger than max_size
+    if scale < 1:
+        new_width = int(width * scale)
+        new_height = int(height * scale)
+        image = image.resize((new_width, new_height), Image.Resampling.LANCZOS)
+
+    return image
+```
+
+You can change the resolution of the image by changing the `max_size` argument. Small max sizes will make the image more blurry, but processing will be faster and require less memory.
+
+## Load an image
+
+Load an image and resize it. We've provided a sample image of a Trader Joe's receipt, but you can use any image you'd like.
+
+Here's what the image looks like:
+
+![Trader Joe's receipt](./images/trader-joes-receipt.jpg)
+
+```python
+# Path to the image
+image_path = "https://dottxt-ai.github.io/outlines/main/cookbook/images/trader-joes-receipt.png"
+
+# Download the image
+response = requests.get(image_path)
+with open("receipt.png", "wb") as f:
+    f.write(response.content)
+
+# Load + resize the image
+image = load_and_resize_image("receipt.png")
+```
+
+## Define the output structure
+
+We'll define a Pydantic model to describe the data we want to extract from the image.
+
+In our case, we want to extract the following information:
+
+- The store name
+- The store address
+- The store number
+- A list of items, including the name, quantity, price per unit, and total price
+- The tax
+- The total
+- The date
+- The payment method
+
+Most fields are optional, as not all receipts contain all information.
+
+```python
+class Item(BaseModel):
+    name: str
+    quantity: Optional[int]
+    price_per_unit: Optional[float]
+    total_price: Optional[float]
+
+class ReceiptSummary(BaseModel):
+    store_name: str
+    store_address: str
+    store_number: Optional[int]
+    items: List[Item]
+    tax: Optional[float]
+    total: Optional[float]
+    # Date is in the format YYYY-MM-DD. We can apply a regex pattern to ensure it's formatted correctly.
+    date: Optional[str] = Field(pattern=r'\d{4}-\d{2}-\d{2}', description="Date in the format YYYY-MM-DD")
+    payment_method: Literal["cash", "credit", "debit", "check", "other"]
+```
+
+## Prepare the prompt
+
+We'll use the `AutoProcessor` to convert the image and the text prompt into a format that the model can understand. Practically,
+this is the code that adds user, system, assistant, and image tokens to the prompt.
+
+```python
+# Set up the content you want to send to the model
+messages = [
+    {
+        "role": "user",
+        "content": [
+            {
+                # The image is provided as a PIL Image object
+                "type": "image",
+                "image": image,
+            },
+            {
+                "type": "text",
+                "text": f"""You are an expert at extracting information from receipts.
+                Please extract the information from the receipt. Be as detailed as possible --
+                missing or misreporting information is a crime.
+
+                Return the information in the following JSON schema:
+                {ReceiptSummary.model_json_schema()}
+            """},
+        ],
+    }
+]
+
+# Convert the messages to the final prompt
+processor = AutoProcessor.from_pretrained(model_name)
+prompt = processor.apply_chat_template(
+    messages, tokenize=False, add_generation_prompt=True
+)
+```
+
+If you are curious, the final prompt that is sent to the model looks (roughly) like this:
+
+```
+<|im_start|>system
+You are a helpful assistant.<|im_end|>
+<|im_start|>user
+<|vision_start|><|image_pad|><|vision_end|>
+You are an expert at extracting information from receipts.
+Please extract the information from the receipt. Be as detailed as
+possible -- missing or misreporting information is a crime.
+
+Return the information in the following JSON schema:
+
+<JSON SCHEMA OMITTED>
+<|im_end|>
+<|im_start|>assistant
+```
+
+## Run the model
+
+```python
+# Prepare a function to process receipts
+receipt_summary_generator = outlines.generate.json(
+    model,
+    ReceiptSummary,
+
+    # Greedy sampling is a good idea for numeric
+    # data extraction -- no randomness.
+    sampler=outlines.samplers.greedy()
+)
+
+# Generate the receipt summary
+result = receipt_summary_generator(prompt, [image])
+print(result)
+```
+
+## Output
+
+The output should look like this:
+
+```
+ReceiptSummary(
+    store_name="Trader Joe's",
+    store_address='401 Bay Street, San Francisco, CA 94133',
+    store_number=0,
+    items=[
+        Item(name='BANANA EACH', quantity=7, price_per_unit=0.23, total_price=1.61),
+        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CARAMEL CASHEW', quantity=2, price_per_unit=2.29, total_price=4.58),
+        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='SPINDRIFT ORANGE MANGO 8', quantity=1, price_per_unit=7.49, total_price=7.49),
+        Item(name='Bottle Deposit', quantity=8, price_per_unit=0.05, total_price=0.4),
+        Item(name='MILK ORGANIC GALLON WHOL', quantity=1, price_per_unit=6.79, total_price=6.79),
+        Item(name='CLASSIC GREEK SALAD', quantity=1, price_per_unit=3.49, total_price=3.49),
+        Item(name='COBB SALAD', quantity=1, price_per_unit=5.99, total_price=5.99),
+        Item(name='PEPPER BELL RED XL EACH', quantity=1, price_per_unit=1.29, total_price=1.29),
+        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25),
+        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25)
+    ],
+    tax=0.68,
+    total=41.98,
+    date='2023-11-04',
+    payment_method='debit',
+
+)
+```
+
+Voila! You've successfully extracted information from a receipt using an LLM.
+
+## Bonus: roasting the user for their receipt
+
+You can roast the user for their receipt by adding a `roast` field to the end of the  `ReceiptSummary` model.
+
+```python
+class ReceiptSummary(BaseModel):
+    ...
+    roast: str
+```
+
+which gives you a result like
+
+```
+ReceiptSummary(
+    ...
+    roast="You must be a fan of Trader Joe's because you bought enough
+    items to fill a small grocery bag and still had to pay for a bag fee.
+    Maybe you should start using reusable bags to save some money and the
+    environment."
+)
+```
+
+Qwen is not particularly funny, but worth a shot.
diff --git a/mkdocs.yml b/mkdocs.yml
index 7272e4a63..7d48bf392 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -122,6 +122,7 @@ nav:
       - Vision-Language Models: cookbook/atomic_caption.md
       - Structured Generation from PDFs: cookbook/read-pdfs.md
       - Earnings reports to CSV: cookbook/earnings-reports.md
+      - Digitizing receipts with vision models: cookbook/receipt-digitization.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From d842522c704c0c0ee23bf493873116425eb76e5d Mon Sep 17 00:00:00 2001
From: Sebastien Campion <scampion@users.noreply.github.com>
Date: Sun, 10 Nov 2024 11:00:09 +0100
Subject: [PATCH 104/505] Update README.md (#1258)

Fix that error NameError: name 'rng' is not defined
---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 832703ce6..c7f071503 100644
--- a/README.md
+++ b/README.md
@@ -190,7 +190,7 @@ character = generator("Give me a character description", seed=seed)
 print(repr(character))
 # Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
 
-character = generator("Give me an interesting character description", rng=rng)
+character = generator("Give me an interesting character description")
 
 print(repr(character))
 # Character(name='Vivian Thr', age=44, armor=<Armor.plate: 'plate'>, weapon=<Weapon.crossbow: 'crossbow'>, strength=125)

From c406da8f9ecacb96b93982b8acaad2f5d12dab6e Mon Sep 17 00:00:00 2001
From: Michael Goin <michael@neuralmagic.com>
Date: Mon, 18 Nov 2024 17:03:48 -0500
Subject: [PATCH 105/505] Bump to outlines-core=0.1.17 for python 3.12-3.13
 support (#1273)

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index fa7005afd..1fd2897aa 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -41,7 +41,7 @@ dependencies = [
    "pycountry",
    "airportsdata",
    "torch",
-   "outlines_core==0.1.14",
+   "outlines_core==0.1.17",
 ]
 dynamic = ["version"]
 

From 2eab80c1ceabdf5ba340b921511f22c235e1c063 Mon Sep 17 00:00:00 2001
From: "Victoria Terenina (torymur)" <torymur@gmail.com>
Date: Thu, 21 Nov 2024 17:05:28 +0000
Subject: [PATCH 106/505] Turn off guide caching

---
 outlines/fsm/guide.py                           | 2 --
 tests/generate/test_integration_llamacpp.py     | 1 +
 tests/generate/test_integration_transformers.py | 1 +
 3 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/fsm/guide.py b/outlines/fsm/guide.py
index d46228fe9..6b97d7729 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/fsm/guide.py
@@ -15,7 +15,6 @@
 )
 
 from outlines import grammars
-from outlines.caching import cache
 from outlines.fsm.parsing import PartialLark, PartialParserState
 
 if TYPE_CHECKING:
@@ -73,7 +72,6 @@ def copy(self):
         return self
 
 
-@cache()
 def cached_create_states_mapping(regex_string, tokenizer, *args, **kwargs):
     return uncached_create_states_mapping(regex_string, tokenizer, *args, **kwargs)
 
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
index 8d4596d60..fd5be2171 100644
--- a/tests/generate/test_integration_llamacpp.py
+++ b/tests/generate/test_integration_llamacpp.py
@@ -274,6 +274,7 @@ def test_llama_cpp_pre_tokenizer_remains_broken():
         generate.choice(model, ["skirt", "dress", "pen", "jacket"])
 
 
+@pytest.mark.skip("Caching for guide was temporarily turned off")
 def test_RegexGuide_caching(model, temp_cache_dir):
     import llama_cpp
 
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index 2462d9fcf..8acb87500 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -492,6 +492,7 @@ def test_transformers_use_existing_model_and_tokenizer():
     assert isinstance(sequence, str)
 
 
+@pytest.mark.skip("Caching for guide was temporarily turned off")
 def test_RegexGuide_caching(temp_cache_dir):
     import outlines.caching
     from outlines.fsm.guide import cached_create_states_mapping

From f099f9668986340f1167a82c5e75fc1066c79594 Mon Sep 17 00:00:00 2001
From: "Victoria Terenina (torymur)" <torymur@gmail.com>
Date: Thu, 21 Nov 2024 19:34:22 +0000
Subject: [PATCH 107/505] Fix index interface in tests

---
 tests/fsm/test_guide.py | 15 +++------------
 1 file changed, 3 insertions(+), 12 deletions(-)

diff --git a/tests/fsm/test_guide.py b/tests/fsm/test_guide.py
index 510faf4b0..bf25c43c4 100644
--- a/tests/fsm/test_guide.py
+++ b/tests/fsm/test_guide.py
@@ -59,7 +59,7 @@ def convert_token_to_string(self, token):
     tokenizer = MockTokenizer()
     fsm = RegexGuide.from_regex(regex_str, tokenizer)
 
-    assert fsm.states_to_token_maps == {0: {1: 1}}
+    assert fsm.states_to_token_maps.get_transitions() == {0: {1: 1}}
 
     instruction = fsm.get_next_instruction(0)
     assert isinstance(instruction, Generate)
@@ -70,9 +70,6 @@ def convert_token_to_string(self, token):
 
     assert fsm.is_final_state(0) is False
 
-    for state in fsm.final_states:
-        assert fsm.is_final_state(state) is True
-
 
 def test_regex_multi_byte_llama_like():
     class MockTokenizer:
@@ -100,7 +97,7 @@ def convert_token_to_string(self, token):
     tokenizer = MockTokenizer()
     fsm = RegexGuide.from_regex(regex_str, tokenizer)
 
-    assert fsm.states_to_token_maps == {
+    assert fsm.states_to_token_maps.get_transitions() == {
         0: {5: 1, 4: 2},
         1: {6: 3},
         3: {7: 4},
@@ -116,9 +113,6 @@ def convert_token_to_string(self, token):
 
     assert fsm.is_final_state(0) is False
 
-    for state in fsm.final_states:
-        assert fsm.is_final_state(state) is True
-
 
 def test_regex_multi_byte_gpt2_like():
     class MockTokenizer:
@@ -147,7 +141,7 @@ def convert_token_to_string(self, token):
     tokenizer = MockTokenizer()
     fsm = RegexGuide.from_regex(regex_str, tokenizer)
 
-    assert fsm.states_to_token_maps == {
+    assert fsm.states_to_token_maps.get_transitions() == {
         0: {5: 1, 10: 2},
         1: {8: 5, 4: 3},
         2: {11: 3},
@@ -163,9 +157,6 @@ def convert_token_to_string(self, token):
 
     assert fsm.is_final_state(0) is False
 
-    for state in fsm.final_states:
-        assert fsm.is_final_state(state) is True
-
 
 def test_regex_final_state():
     """Make sure that the FSM stays in the final state as we keep generating"""

From b55d31463cb6ed38fc0109e018f53ce0cdafbe19 Mon Sep 17 00:00:00 2001
From: Jeremy Zucker <Jeremy.Zucker@pnnl.gov>
Date: Wed, 6 Nov 2024 16:25:29 -0800
Subject: [PATCH 108/505] Update generation.md

`[Outlines model](../models)` does not return the link correctly.

Tried switching to `[Outlines model](../models/models.md)`
---
 docs/reference/generation/generation.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
index a14818514..930ad9d22 100644
--- a/docs/reference/generation/generation.md
+++ b/docs/reference/generation/generation.md
@@ -4,7 +4,7 @@ title: Generation
 
 # Generation
 
-Once an [Outlines model](../models) is constructed you can use `outlines.generate` to generate text. Standard LLM generation is possible via `outlines.generate.text`, along with a variety of structured generation methods described below. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
+Once an [Outlines model](../models/models.md) is constructed you can use `outlines.generate` to generate text. Standard LLM generation is possible via `outlines.generate.text`, along with a variety of structured generation methods described below. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
 
 Before generating text, you must construct an `outlines.model`. Example:
 

From bf62d119748a33e558d9ef87385117dde040e40f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Wed, 27 Nov 2024 13:25:13 +0100
Subject: [PATCH 109/505] Add link to YT channel and .txt blog

---
 README.md | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index c7f071503..f14c1e925 100644
--- a/README.md
+++ b/README.md
@@ -2,17 +2,18 @@
 
 <img src="./docs/assets/images/logo.png" alt="Outlines Logo" width=500></img>
 
-[![.txt Twitter][dottxt-twitter-badge]][dottxt-twitter]
+
+ 🗒️ *Make LLMs speak the language of every application.* 🗒️
+
+Made with ❤👷️ by the team at [.txt](https://dottxt.co).
 
 [![Documentation][documentation-badge]][documentation]
 [![Contributors][contributors-badge]][contributors]
 [![Downloads][downloads-badge]][pypistats]
 [![Discord][discord-badge]][discord]
 
+[Youtube channel][youtube-dottxt] | [.txt blog][blog-dottxt] | [Twitter][dottxt-twitter]
 
-*Robust (structured) text generation.*
-
-Made with ❤👷️ by the team at [.txt](https://dottxt.co).
 
 </div>
 
@@ -363,3 +364,5 @@ answer = outlines.generate.text(model)(prompt, max_tokens=100)
 [downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=89AC6B&logo=python&logoColor=white&style=flat-square
 [pypistats]: https://pypistats.org/packages/outlines
 [dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
+[youtube-dottxt]: https://www.youtube.com/@dottxt-ai
+[blog-dottxt]: https://blog.dottxt.co/

From 7a9baad809eff63df17f2f0ae79acd5cd4187db4 Mon Sep 17 00:00:00 2001
From: George Tsiolis <tsiolis.g@gmail.com>
Date: Wed, 27 Nov 2024 03:28:33 +0200
Subject: [PATCH 110/505] Update home page spacing

---
 docs/overrides/home.html   | 111 ++++++++++++++++++++++++-------------
 docs/stylesheets/extra.css |   3 +
 2 files changed, 74 insertions(+), 40 deletions(-)

diff --git a/docs/overrides/home.html b/docs/overrides/home.html
index 0c97d5ac6..6525ab062 100644
--- a/docs/overrides/home.html
+++ b/docs/overrides/home.html
@@ -7,27 +7,15 @@
 {{ super() }}
 
 <style>
-    .md-header {
-        position: initial
-    }
-
-    .md-main__inner {
-        margin: 0
-    }
-
-    .md-content {
-        display: none
-    }
-
     @media screen and (min-width:60em) {
         .md-sidebar--secondary {
-            display: none
+            display: none;
         }
     }
 
     @media screen and (min-width:76.25em) {
         .md-sidebar--primary {
-            display: none
+            display: none;
         }
     }
 
@@ -35,51 +23,96 @@
         display: flex;
         justify-content: center;
         align-items: center;
-        min-height: calc(100vh - 300px);
         padding-left: 1rem;
         padding-right: 1rem;
     }
 
     .mdx-hero {
         text-align: center;
+        margin-top: 6rem;
     }
 
     .mdx-hero__image {
         margin-bottom: 1rem;
         margin-top: 1rem;
+        max-width: 80%;
+        margin: 0 auto;
+
+        @media screen and (max-width: 40em) {
+            max-width: 60%;
+        }
     }
 
     .mdx-hero__content {
         max-width: 600px;
         margin: 0 auto;
-    }
 
+        .subtitle {
+            font-size: 1rem;
+            letter-spacing: 0.025rem;
+        }
+
+        h2 {
+            max-width: 460px;
+        }
+
+        @media screen and (max-width: 40em) {
+            .subtitle {
+                font-size: 0.8rem;
+                max-width: 70%;
+                font-weight: 700;
+            }
+        }
+
+        a:hover{
+            color: #A7623A;
+        }
+    }
 
     .md-buttons {
         display: flex;
         justify-content: center;
-        gap: 1rem;
+        gap: 0.5rem;
+        flex-direction: row;
+        margin: 0 auto;
         margin-top: 2rem;
-        flex-direction: row; /* Default to row layout */
+        flex-direction: row;
+        width: max-content;
+        font-weight: 700;
     }
 
     .md-team {
-        margin-top: 2rem;
-        margin-bottom: 1rem;
+        margin-top: 4rem;
     }
 
     /* New media query for smaller screens */
     @media screen and (max-width: 40em) {
         .md-buttons {
-            flex-direction: column; /* Stack buttons vertically */
-            align-items: center; /* Center-align the stacked buttons */
+            flex-direction: column;
+            /* Stack buttons vertically */
+            align-items: center;
+            /* Center-align the stacked buttons */
         }
 
         .md-buttons .md-button {
-            width: 100%; /* Make buttons full-width */
-            max-width: 200px; /* Limit maximum width for better appearance */
+            width: 100%;
+            /* Make buttons full-width */
+            max-width: 200px;
+            /* Limit maximum width for better appearance */
         }
     }
+
+    .md-button {
+        transition: background-color 0.3s ease-in-out; /* Smooth transition */
+        border-radius: 6px !important;
+        padding: 0.4rem 0.8rem !important;
+
+    }
+
+    .md-button:hover {
+        background-color: #A7623A !important; /* Desired hover background color */
+        border-color: #A7623A !important;
+    }
 </style>
 
 <section class="mdx-container">
@@ -89,26 +122,24 @@
                 <img src="assets/images/logo.png" alt="Outlines Logo" width="600" draggable="false">
             </div>
             <div class="mdx-hero__content">
-                <h2 class="subtitle" style="font-weight: 400; padding-top: 1rem;">
+                <h2 class="subtitle" style="font-weight: 500; padding-top: 1rem;">
                     Structured text generation and robust prompting for language models
                 </h2>
-
-                    <!-- <div class="index-pre-code">
-                        <pre><code>pip install outlines</code></pre>
-                    </div> -->
-
-                    <div class="md-buttons">
-                        <a href="{{ 'welcome/' | url }}" title="Get started" class="md-button md-button--primary">
-                            Get started
-                        </a>
-                        <a href="https://discord.gg/ZxBxyWmW5n" title="Join the Community" class="md-button">
-                            Join the Community
-                        </a>
-                        <a href="https://twitter.com/dottxtai" title="Follow us on X" class="md-button">
+                <div class="md-buttons">
+                    <a href="{{ 'welcome/' | url }}" title="Get started" class="md-button md-button--primary">
+                        Get started
+                    </a>
+                    <a href="https://discord.gg/ZxBxyWmW5n" title="Join the Community" class="md-button">
+                        Join the community
+                    </a>
+                </div>
+                <div class="md-buttons">
+                    <p>
+                        <a href="https://twitter.com/dottxtai" title="Follow us on X" target="_blank">
                             Follow us on X
                         </a>
-                    </div>
-
+                    </p>
+                </div>
 
                 <p class="md-team">Made with ❤️ by the team at <a href="https://dottxt.co">.txt</a></p>
             </div>
diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index 33783c153..e71a6124f 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -112,6 +112,9 @@ h1.title {
 
 h2.subtitle {
   margin: 5px 0px 25px;
+  font-size: 1rem;
+  max-width: 540px;
+  margin: 0 auto;
 }
 
 .md-typeset {

From 63b4feb2c4ea3c7767d7440bdf5d01fa887cbbcb Mon Sep 17 00:00:00 2001
From: Neil Mehta <neil@lmstudio.ai>
Date: Tue, 12 Nov 2024 12:28:39 -0500
Subject: [PATCH 111/505] Add compatibility for numpy 2 while preserving numpy
 1 compatibility

---
 outlines/base.py | 23 +++++++++++++++++------
 pyproject.toml   |  2 +-
 2 files changed, 18 insertions(+), 7 deletions(-)

diff --git a/outlines/base.py b/outlines/base.py
index 4de8ccf5a..29d42c54c 100644
--- a/outlines/base.py
+++ b/outlines/base.py
@@ -5,12 +5,23 @@
 from typing import Callable, Optional
 
 import numpy as np
-from numpy.lib.function_base import (
-    _calculate_shapes,
-    _parse_gufunc_signature,
-    _parse_input_dimensions,
-    _update_dim_sizes,
-)
+
+# Import required functions based on NumPy version
+np_major_version = int(np.__version__.split(".")[0])
+if np_major_version >= 2:
+    from numpy.lib._function_base_impl import (
+        _calculate_shapes,
+        _parse_gufunc_signature,
+        _parse_input_dimensions,
+        _update_dim_sizes,
+    )
+else:
+    from numpy.lib.function_base import (
+        _calculate_shapes,
+        _parse_gufunc_signature,
+        _parse_input_dimensions,
+        _update_dim_sizes,
+    )
 
 # Allow nested loops for running in notebook. We don't enable it globally as it
 # may interfere with other libraries that use asyncio.
diff --git a/pyproject.toml b/pyproject.toml
index 1fd2897aa..a8b77011a 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -28,7 +28,7 @@ dependencies = [
    "jinja2",
    "lark",
    "nest_asyncio",
-   "numpy<2.0.0",
+   "numpy",
    "cloudpickle",
    "diskcache",
    "pydantic>=2.0",

From 7cdaeacc89df059be0e302a6708c19b910f1553b Mon Sep 17 00:00:00 2001
From: Chris McMaster <mcmastc1@gmail.com>
Date: Mon, 11 Nov 2024 16:46:06 +1100
Subject: [PATCH 112/505] Update `mlx-lm` kvcache creation

---
 outlines/models/mlxlm.py   |   7 +--
 pyproject.toml             |   2 +-
 tests/models/test_mlxlm.py | 100 +++++++++++++++++++++++++++++++++++++
 3 files changed, 102 insertions(+), 7 deletions(-)
 create mode 100644 tests/models/test_mlxlm.py

diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 6e63ef5b6..d8b7e032c 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -167,12 +167,7 @@ def sample(logits: "mx.array") -> Tuple["mx.array", float]:
             prob = softmax_logits[0, token]
             return token, prob
 
-        kv_heads = (
-            [self.model.n_kv_heads] * len(self.model.layers)
-            if isinstance(self.model.n_kv_heads, int)
-            else self.model.n_kv_heads
-        )
-        cache = [mlx_lm.models.base.KVCache(self.model.head_dim, n) for n in kv_heads]
+        cache = mlx_lm.models.cache.make_prompt_cache(self.model)
 
         # kv cache contains processed input IDs, we pass the unprocessed inputs and cache to model()
         unprocessed_input_ids = prompt
diff --git a/pyproject.toml b/pyproject.toml
index a8b77011a..4972f09ef 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -58,7 +58,7 @@ test = [
     "beartype<0.16.0",
     "responses",
     "llama-cpp-python",
-    "mlx-lm; platform_machine == 'arm64' and sys_platform == 'darwin'",
+    "mlx-lm>=0.19.2; platform_machine == 'arm64' and sys_platform == 'darwin'",
     "huggingface_hub",
     "openai>=1.0.0",
     "vllm; sys_platform != 'darwin'",
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
new file mode 100644
index 000000000..20e59da81
--- /dev/null
+++ b/tests/models/test_mlxlm.py
@@ -0,0 +1,100 @@
+import pytest
+
+from outlines.models.mlxlm import mlxlm
+from outlines.models.transformers import TransformerTokenizer
+
+try:
+    import mlx.core as mx
+
+    HAS_MLX = mx.metal.is_available()
+except ImportError:
+    HAS_MLX = False
+
+
+TEST_MODEL = "mlx-community/SmolLM-135M-Instruct-4bit"
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_model():
+    model = mlxlm(TEST_MODEL)
+    assert hasattr(model, "model")
+    assert hasattr(model, "tokenizer")
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_tokenizer():
+    model = mlxlm(TEST_MODEL)
+
+    # Test single string encoding/decoding
+    test_text = "Hello, world!"
+    token_ids = mx.array(model.mlx_tokenizer.encode(test_text))
+    assert isinstance(token_ids, mx.array)
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_generate():
+    from outlines.generate.api import GenerationParameters, SamplingParameters
+
+    model = mlxlm(TEST_MODEL)
+    prompt = "Write a haiku about programming:"
+
+    # Test with basic generation parameters
+    gen_params = GenerationParameters(max_tokens=50, stop_at=None, seed=None)
+
+    # Test with different sampling parameters
+    sampling_params = SamplingParameters(
+        sampler="multinomial", num_samples=1, top_p=0.9, top_k=None, temperature=0.7
+    )
+
+    # Test generation
+    output = model.generate(prompt, gen_params, None, sampling_params)
+    assert isinstance(output, str)
+    assert len(output) > 0
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_stream():
+    from outlines.generate.api import GenerationParameters, SamplingParameters
+
+    model = mlxlm(TEST_MODEL)
+    prompt = "Count from 1 to 5:"
+
+    gen_params = GenerationParameters(max_tokens=20, stop_at=None, seed=None)
+
+    sampling_params = SamplingParameters(
+        sampler="greedy",  # Use greedy sampling for deterministic output
+        num_samples=1,
+        top_p=None,
+        top_k=None,
+        temperature=0.0,
+    )
+
+    # Test streaming
+    stream = model.stream(prompt, gen_params, None, sampling_params)
+    tokens = list(stream)
+    assert len(tokens) > 0
+    assert all(isinstance(token, str) for token in tokens)
+
+    # Test that concatenated streaming output matches generate output
+    streamed_text = "".join(tokens)
+    generated_text = model.generate(prompt, gen_params, None, sampling_params)
+    assert streamed_text == generated_text
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_errors():
+    model = mlxlm(TEST_MODEL)
+
+    # Test batch inference (should raise NotImplementedError)
+    with pytest.raises(NotImplementedError):
+        from outlines.generate.api import GenerationParameters, SamplingParameters
+
+        gen_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
+        sampling_params = SamplingParameters("multinomial", 1, None, None, 1.0)
+        model.generate(["prompt1", "prompt2"], gen_params, None, sampling_params)
+
+    # Test beam search (should raise NotImplementedError)
+    with pytest.raises(NotImplementedError):
+        sampling_params = SamplingParameters("beam_search", 1, None, None, 1.0)
+        model.generate("test prompt", gen_params, None, sampling_params)

From 2db34c5ce7645ee7b7c67c1734e94e51d301f790 Mon Sep 17 00:00:00 2001
From: beta9 <egottli2@binghamton.edu>
Date: Thu, 24 Oct 2024 17:00:50 -0400
Subject: [PATCH 113/505] Add `beam` deployment example

---
 examples/beam-cloud/README.md |  5 +++++
 examples/beam-cloud/app.py    | 39 +++++++++++++++++++++++++++++++++++
 2 files changed, 44 insertions(+)
 create mode 100644 examples/beam-cloud/README.md
 create mode 100644 examples/beam-cloud/app.py

diff --git a/examples/beam-cloud/README.md b/examples/beam-cloud/README.md
new file mode 100644
index 000000000..5f190d76f
--- /dev/null
+++ b/examples/beam-cloud/README.md
@@ -0,0 +1,5 @@
+## Deploy Outlines on Beam
+
+1. Create an account [here](https://beam.cloud) and install the Beam SDK
+2. Download the `app.py` file to your computer
+3. Deploy it as a serverless API by running: `beam deploy app.py:predict`
diff --git a/examples/beam-cloud/app.py b/examples/beam-cloud/app.py
new file mode 100644
index 000000000..fb6c2cb2b
--- /dev/null
+++ b/examples/beam-cloud/app.py
@@ -0,0 +1,39 @@
+from beam import Image, endpoint, env
+
+if env.is_remote():
+    import outlines
+
+
+# Pre-load models when the container first starts
+def load_models():
+    import outlines
+
+    model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+    return model
+
+
+@endpoint(
+    name="outlines-serverless",
+    gpu="A10G",
+    cpu=1,
+    memory="16Gi",
+    on_start=load_models,
+    image=Image().add_python_packages(
+        ["outlines", "torch", "transformers", "accelerate"]
+    ),
+)
+def predict(context, **inputs):
+    default_prompt = """You are a sentiment-labelling assistant.
+    Is the following review positive or negative?
+
+    Review: This restaurant is just awesome!
+    """
+
+    prompt = inputs.get("prompt", default_prompt)
+
+    # Unpack cached model from context
+    model = context.on_start_value
+    # Inference
+    generator = outlines.generate.choice(model, ["Positive", "Negative"])
+    answer = generator(prompt)
+    return {"answer": answer}

From 66f662781c43154a4411f98c11ea0290ef727c39 Mon Sep 17 00:00:00 2001
From: Ulf Aslak <ulfjensen@gmail.com>
Date: Sat, 28 Sep 2024 22:21:49 +0200
Subject: [PATCH 114/505] Add OpenAI to LogitsGenerator

---
 outlines/models/__init__.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index d28fcb2d7..fe6f861ac 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -16,4 +16,4 @@
 from .transformers_vision import TransformersVision, transformers_vision
 from .vllm import VLLM, vllm
 
-LogitsGenerator = Union[Transformers, LlamaCpp, ExLlamaV2Model, MLXLM, VLLM]
+LogitsGenerator = Union[Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM]

From 568f252da3835329e65b74065166e21052c6a9b5 Mon Sep 17 00:00:00 2001
From: Sebastien Campion <scampion@users.noreply.github.com>
Date: Wed, 27 Nov 2024 21:11:27 +0100
Subject: [PATCH 115/505] Add cookbook recipe to extract event details from
 text (#1269)

---
 docs/cookbook/extract_event_details.md | 34 +++++++++++++++++++
 docs/cookbook/extract_event_details.py | 46 ++++++++++++++++++++++++++
 mkdocs.yml                             |  2 ++
 3 files changed, 82 insertions(+)
 create mode 100644 docs/cookbook/extract_event_details.md
 create mode 100644 docs/cookbook/extract_event_details.py

diff --git a/docs/cookbook/extract_event_details.md b/docs/cookbook/extract_event_details.md
new file mode 100644
index 000000000..0f87d9586
--- /dev/null
+++ b/docs/cookbook/extract_event_details.md
@@ -0,0 +1,34 @@
+This recipe demonstrates how to use the `outlines` library to extract structured event details from a text message.
+We will extract the title, location, and start date and time from messages like the following:
+
+```plaintext
+Hello Kitty, my grandmother will be here, I think it's better to postpone
+our appointment to review math lessons to next Monday at 2pm at the same
+place, 3 avenue des tanneurs, one hour will be enough see you 😘
+```
+
+Let see how to extract the event details from the message with the MLX
+library dedicated to  Apple Silicon processor  (M series).
+
+```python
+--8<-- "docs/cookbook/extract_event_details.py"
+```
+
+The output will be:
+
+```plaintext
+Today: Saturday 16 November 2024 and it's 10:55
+```
+
+and the extracted event information will be:
+
+```json
+{
+  "title":"Math Review",
+  "location":"3 avenue des tanneurs",
+  "start":"2024-11-22T14:00:00Z"
+}
+```
+
+
+To find out more about this use case, we recommend the project developped by [Joseph Rudoler](https://x.com/JRudoler) the [ICS Generator](https://github.com/jrudoler/ics-generator)
diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
new file mode 100644
index 000000000..b51f8d921
--- /dev/null
+++ b/docs/cookbook/extract_event_details.py
@@ -0,0 +1,46 @@
+from datetime import datetime
+
+from pydantic import BaseModel, Field
+
+from outlines import generate, models
+
+# Load the model
+model = models.mlxlm("mlx-community/Hermes-3-Llama-3.1-8B-8bit")
+
+
+# Define the event schema using Pydantic
+class Event(BaseModel):
+    title: str = Field(description="title of the event")
+    location: str
+    start: datetime = Field(
+        default=None, description="date of the event if available in iso format"
+    )
+
+
+# Get the current date and time
+now = datetime.now().strftime("%A %d %B %Y and it's %H:%M")
+
+# Define the prompt
+prompt = f"""
+Today's date and time are {now}
+Given a user message, extract information of the event like date and time in iso format, location and title.
+If the given date is relative, think step by step to find the right date.
+Here is the message:
+"""
+
+# Sample message
+message = """Hello Kitty, my grandmother will be here , I think it's better to postpone our
+appointment to review math lessons to next Friday at 2pm at the same place, 3 avenue des tanneurs, I think that one hour will be enough
+see you 😘 """
+
+# Create the generator
+generator = generate.json(model, Event)
+
+# Extract the event information
+event = generator(prompt + message)
+
+# Print the current date and time
+print(f"Today: {now}")
+
+# Print the extracted event information in JSON format
+print(event.json())
diff --git a/mkdocs.yml b/mkdocs.yml
index 7d48bf392..4888309b9 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -74,6 +74,7 @@ markdown_extensions:
   - pymdownx.emoji:
       emoji_index: !!python/name:material.extensions.emoji.twemoji
       emoji_generator: !!python/name:material.extensions.emoji.to_svg
+  - pymdownx.snippets:
 
 
 extra_css:
@@ -123,6 +124,7 @@ nav:
       - Structured Generation from PDFs: cookbook/read-pdfs.md
       - Earnings reports to CSV: cookbook/earnings-reports.md
       - Digitizing receipts with vision models: cookbook/receipt-digitization.md
+      - Extract events details from text: cookbook/extract_event_details.md
       - Run on the cloud:
           - BentoML: cookbook/deploy-using-bentoml.md
           - Cerebrium: cookbook/deploy-using-cerebrium.md

From 5608dd809e87b36eb395fdbeec8dd48d0f513a73 Mon Sep 17 00:00:00 2001
From: Aakash Thatte <84656834+sky-2002@users.noreply.github.com>
Date: Thu, 28 Nov 2024 01:42:25 +0530
Subject: [PATCH 116/505] Add jax compatible api (#1207)

This PR adds a JAX compatible API, refer issue #1027
---
 benchmarks/bench_processors.py               | 23 ++++++
 outlines/processors/base_logits_processor.py | 21 ++++++
 pyproject.toml                               |  3 +
 tests/processors/test_base_processor.py      | 74 ++++++++++++++++++++
 4 files changed, 121 insertions(+)
 create mode 100644 tests/processors/test_base_processor.py

diff --git a/benchmarks/bench_processors.py b/benchmarks/bench_processors.py
index 5b4901540..db1e4a8f1 100644
--- a/benchmarks/bench_processors.py
+++ b/benchmarks/bench_processors.py
@@ -9,6 +9,12 @@
 except ImportError:
     pass
 
+try:
+    import jax
+    import jax.numpy as jnp
+except ImportError:
+    pass
+
 
 def is_mlx_lm_allowed():
     try:
@@ -18,6 +24,14 @@ def is_mlx_lm_allowed():
     return mx.metal.is_available()
 
 
+def is_jax_allowed():
+    try:
+        import jax  # noqa: F401
+    except ImportError:
+        return False
+    return True
+
+
 def get_mock_processor_inputs(array_library, num_tokens=30000):
     """
     logits: (4, 30,000 ) dtype=float
@@ -43,6 +57,13 @@ def get_mock_processor_inputs(array_library, num_tokens=30000):
         input_ids = mx.random.randint(
             low=0, high=num_tokens, shape=(4, 2048), dtype=mx.int32
         )
+    elif array_library == "jax":
+        logits = jnp.random.uniform(
+            key=jax.random.PRNGKey(0), shape=(4, num_tokens), dtype=jnp.float32
+        )
+        input_ids = jnp.random.randint(
+            key=jax.random.PRNGKey(0), low=0, high=num_tokens, shape=(4, 2048)
+        )
     else:
         raise ValueError
 
@@ -67,6 +88,8 @@ class LogitsProcessorPassthroughBenchmark:
         params += ["mlx"]
     if torch.cuda.is_available():
         params += ["torch_cuda"]
+    if is_jax_allowed():
+        params += ["jax"]
 
     def setup(self, array_library):
         self.logits_processor = HalvingLogitsProcessor()
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index 9a52abecd..eec7de121 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -20,6 +20,16 @@ def is_mlx_array_type(array_type):
     return issubclass(array_type, mx.array)
 
 
+def is_jax_array_type(array_type):
+    try:
+        import jaxlib
+    except ImportError:
+        return False
+    return issubclass(array_type, jaxlib.xla_extension.ArrayImpl) or isinstance(
+        array_type, jaxlib.xla_extension.ArrayImpl
+    )
+
+
 class OutlinesLogitsProcessor(Protocol):
     """
     Base class for logits processors which normalizes types of logits:
@@ -101,6 +111,12 @@ def _to_torch(tensor_like: Array) -> torch.Tensor:
             # https://ml-explore.github.io/mlx/build/html/usage/numpy.html
             return torch.from_dlpack(tensor_like)
 
+        elif is_jax_array_type(type(tensor_like)):
+            import jax
+
+            torch_tensor = torch.from_dlpack(jax.dlpack.to_dlpack(tensor_like))
+            return torch_tensor
+
         else:
             raise TypeError(
                 "LogitsProcessor must be called with either np.NDArray, "
@@ -129,6 +145,11 @@ def _from_torch(tensor: torch.Tensor, target_type: Type) -> Array:
             # numpy doesn't support bfloat16, mlx doesn't support direct conversion from torch
             return mx.array(tensor.float().numpy())
 
+        elif is_jax_array_type(target_type):
+            import jax
+
+            return jax.dlpack.from_dlpack(tensor)
+
         else:
             raise TypeError(
                 f"Failed to convert torch tensors to target_type `{target_type}`"
diff --git a/pyproject.toml b/pyproject.toml
index 4972f09ef..294fbe4b8 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -109,6 +109,9 @@ enable_incomplete_feature = ["Unpack"]
 [[tool.mypy.overrides]]
 module = [
     "exllamav2.*",
+    "jax",
+    "jaxlib",
+    "jax.numpy",
     "jinja2",
     "jsonschema.*",
     "openai.*",
diff --git a/tests/processors/test_base_processor.py b/tests/processors/test_base_processor.py
new file mode 100644
index 000000000..cd9f48278
--- /dev/null
+++ b/tests/processors/test_base_processor.py
@@ -0,0 +1,74 @@
+from typing import List
+
+import jax.numpy as jnp
+import numpy as np
+import pytest
+import torch
+
+from outlines.processors.base_logits_processor import OutlinesLogitsProcessor
+
+arrays = {
+    "list": [[1.0, 2.0], [3.0, 4.0]],
+    "np": np.array([[1, 2], [3, 4]], dtype=np.float32),
+    "jax": jnp.array([[1, 2], [3, 4]], dtype=jnp.float32),
+    "torch": torch.tensor([[1, 2], [3, 4]], dtype=torch.float32),
+}
+
+try:
+    import mlx.core as mx
+
+    arrays["mlx"] = mx.array([[1, 2], [3, 4]], dtype=mx.float32)
+except ImportError:
+    pass
+
+try:
+    import jax.numpy as jnp
+
+    arrays["jax"] = jnp.array([[1, 2], [3, 4]], dtype=jnp.float32)
+except ImportError:
+    pass
+
+
+# Mock implementation of the abstract class for testing
+class MockLogitsProcessor(OutlinesLogitsProcessor):
+    def process_logits(
+        self, input_ids: List[List[int]], logits: torch.Tensor
+    ) -> torch.Tensor:
+        # For testing purposes, let's just return logits multiplied by 2
+        return logits * 2
+
+
+@pytest.fixture
+def processor():
+    """Fixture for creating an instance of the MockLogitsProcessor."""
+    return MockLogitsProcessor()
+
+
+@pytest.mark.parametrize("array_type", arrays.keys())
+def test_to_torch(array_type, processor):
+    data = arrays[array_type]
+    torch_tensor = processor._to_torch(data)
+    assert isinstance(torch_tensor, torch.Tensor)
+    assert torch.allclose(
+        torch_tensor.cpu(), torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
+    )
+
+
+@pytest.mark.parametrize("array_type", arrays.keys())
+def test_from_torch(array_type, processor):
+    torch_tensor = torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
+    data = processor._from_torch(torch_tensor, type(arrays[array_type]))
+    assert isinstance(data, type(arrays[array_type]))
+    assert np.allclose(data, arrays[array_type])
+
+
+@pytest.mark.parametrize("array_type", arrays.keys())
+def test_call(array_type, processor):
+    input_ids = arrays[array_type]
+    logits = arrays[array_type]
+    processed_logits = processor(input_ids, logits)
+
+    assert isinstance(processed_logits, type(arrays[array_type]))
+    assert np.allclose(
+        np.array(processed_logits), np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
+    )

From e9485cf2126d9c14bd749d55f8aa6729d96808d0 Mon Sep 17 00:00:00 2001
From: g-prz <158364710+g-prz@users.noreply.github.com>
Date: Wed, 27 Nov 2024 20:28:25 +0000
Subject: [PATCH 117/505] Add json call with multi-function enums (#1277)

This PR aims at solving #1217
---
 README.md                                     | 27 +++++++++
 outlines/fsm/json_schema.py                   | 18 +++++-
 outlines/generate/json.py                     | 12 +++-
 tests/fsm/test_json_schema.py                 | 59 ++++++++++++++++++-
 .../generate/test_integration_transformers.py | 24 ++++++++
 5 files changed, 136 insertions(+), 4 deletions(-)

diff --git a/README.md b/README.md
index f14c1e925..e9285e863 100644
--- a/README.md
+++ b/README.md
@@ -300,6 +300,33 @@ print(add(**result))
 
 A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
 
+You can also embed various functions into an enum to generate params:
+
+```python
+from enum import Enum
+from functools import partial
+
+import outlines
+
+
+def add(a: int, b: int) -> int:
+    return a + b
+
+def mul(c: float, d: float) -> float:
+    return c * d
+
+class Operation(Enum):
+    add = partial(add)
+    mul = partial(mul)
+
+model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
+generator = outlines.generate.json(model, add)
+result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
+
+print(result)
+# {'c': -3.14, 'd': 1.5}
+```
+
 ## Prompting
 
 Building prompts can get messy. **Outlines** makes it easier to write and manage
diff --git a/outlines/fsm/json_schema.py b/outlines/fsm/json_schema.py
index 98d2de59c..0bab57923 100644
--- a/outlines/fsm/json_schema.py
+++ b/outlines/fsm/json_schema.py
@@ -2,6 +2,7 @@
 import json
 import re
 import warnings
+from enum import Enum
 from typing import Callable, Optional, Tuple, Type, Union
 
 from jsonschema.protocols import Validator
@@ -306,6 +307,8 @@ def to_regex(
         for choice in instance["enum"]:
             if type(choice) in [int, float, bool, type(None), str]:
                 choices.append(re.escape(json.dumps(choice)))
+            elif isinstance(choice, dict):
+                choices.append(to_regex(resolver, choice, whitespace_pattern))
             else:
                 raise TypeError(f"Unsupported data type in enum: {type(choice)}")
         return f"({'|'.join(choices)})"
@@ -524,7 +527,7 @@ def to_regex(
     )
 
 
-def get_schema_from_signature(fn: Callable) -> str:
+def get_schema_from_signature(fn: Callable) -> dict:
     """Turn a function signature into a JSON schema.
 
     Every JSON object valid to the output JSON Schema can be passed
@@ -550,3 +553,16 @@ def get_schema_from_signature(fn: Callable) -> str:
     model = create_model(fn_name, **arguments)
 
     return model.model_json_schema()
+
+
+def get_schema_from_enum(myenum: type[Enum]) -> dict:
+    if len(myenum) == 0:
+        raise ValueError(
+            f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
+        )
+    choices = [
+        get_schema_from_signature(elt.value.func) if callable(elt.value) else elt.value
+        for elt in myenum
+    ]
+    schema = {"title": myenum.__name__, "enum": choices}
+    return schema
diff --git a/outlines/generate/json.py b/outlines/generate/json.py
index f75878d29..703447958 100644
--- a/outlines/generate/json.py
+++ b/outlines/generate/json.py
@@ -1,10 +1,15 @@
 import json as pyjson
+from enum import Enum
 from functools import singledispatch
 from typing import Callable, Optional, Union
 
 from pydantic import BaseModel
 
-from outlines.fsm.json_schema import build_regex_from_schema, get_schema_from_signature
+from outlines.fsm.json_schema import (
+    build_regex_from_schema,
+    get_schema_from_enum,
+    get_schema_from_signature,
+)
 from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
@@ -48,6 +53,11 @@ def json(
         regex_str = build_regex_from_schema(schema, whitespace_pattern)
         generator = regex(model, regex_str, sampler)
         generator.format_sequence = lambda x: schema_object.parse_raw(x)
+    elif isinstance(schema_object, type(Enum)):
+        schema = pyjson.dumps(get_schema_from_enum(schema_object))
+        regex_str = build_regex_from_schema(schema, whitespace_pattern)
+        generator = regex(model, regex_str, sampler)
+        generator.format_sequence = lambda x: pyjson.loads(x)
     elif callable(schema_object):
         schema = pyjson.dumps(get_schema_from_signature(schema_object))
         regex_str = build_regex_from_schema(schema, whitespace_pattern)
diff --git a/tests/fsm/test_json_schema.py b/tests/fsm/test_json_schema.py
index 7565ff642..6f0b59c50 100644
--- a/tests/fsm/test_json_schema.py
+++ b/tests/fsm/test_json_schema.py
@@ -1,5 +1,8 @@
 import json
 import re
+from contextlib import nullcontext
+from enum import Enum
+from functools import partial
 from typing import List, Literal, Union
 
 import interegular
@@ -19,6 +22,7 @@
     UUID,
     WHITESPACE,
     build_regex_from_schema,
+    get_schema_from_enum,
     get_schema_from_signature,
     to_regex,
 )
@@ -237,8 +241,26 @@ def test_match_number(pattern, does_match):
         ),
         # Enum mix of types
         (
-            {"title": "Foo", "enum": [6, 5.3, "potato", True, None]},
-            r'(6|5\.3|"potato"|true|null)',
+            {
+                "title": "Foo",
+                "enum": [
+                    6,
+                    5.3,
+                    "potato",
+                    True,
+                    None,
+                    {
+                        "properties": {
+                            "a": {"title": "A", "type": "number"},
+                            "b": {"title": "B", "type": "number"},
+                        },
+                        "required": ["a", "b"],
+                        "title": "add",
+                        "type": "object",
+                    },
+                ],
+            },
+            r'(6|5\.3|"potato"|true|null|\{[ ]?"a"[ ]?:[ ]?((-)?(0|[1-9][0-9]*))(\.[0-9]+)?([eE][+-][0-9]+)?[ ]?,[ ]?"b"[ ]?:[ ]?((-)?(0|[1-9][0-9]*))(\.[0-9]+)?([eE][+-][0-9]+)?[ ]?\})',
             [
                 ("6", True),
                 ("5.3", True),
@@ -248,6 +270,8 @@ def test_match_number(pattern, does_match):
                 ("523", False),
                 ("True", False),
                 ("None", False),
+                ('{"a": -1.0, "b": 1.1}', True),
+                ('{"a": "a", "b": 1.1}', False),
             ],
         ),
         # integer
@@ -1039,3 +1063,34 @@ class Model(BaseModel):
 
     # check if the pattern uses lookarounds incompatible with interegular.Pattern.to_fsm()
     interegular.parse_pattern(pattern).to_fsm()
+
+
+def add(a: float, b: float) -> float:
+    return a + b
+
+
+class MyEnum(Enum):
+    add = partial(add)
+    a = "a"
+    b = 2
+
+
+# if you don't register your function as callable, you will get an empty enum
+class EmptyEnum(Enum):
+    add = add
+
+
+@pytest.mark.parametrize(
+    "enum,expectation",
+    [
+        (MyEnum, nullcontext()),
+        (EmptyEnum, pytest.raises(ValueError)),
+    ],
+)
+def test_enum_schema(enum, expectation):
+    with expectation:
+        result = get_schema_from_enum(enum)
+        assert result["title"] == enum.__name__
+        assert len(result["enum"]) == len(enum)
+        for elt in result["enum"]:
+            assert type(elt) in [int, float, bool, type(None), str, dict]
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index 8acb87500..92c5d789c 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -1,6 +1,7 @@
 import datetime
 import re
 from enum import Enum
+from functools import partial
 from typing import List, Union
 
 import pytest
@@ -354,6 +355,29 @@ class User(BaseModel):
     assert result.user_id in [1, 2]
 
 
+def add(a: int, b: int) -> int:
+    return a + b
+
+
+def mul(c: float, d: float) -> float:
+    return c * d
+
+
+def test_transformers_json_function_enum(model):
+    prompt = "Output some JSON "
+
+    class Operation(Enum):
+        add = partial(add)
+        mul = partial(mul)
+
+    result = generate.json(model, Operation)(prompt, seed=0)
+    assert isinstance(result, dict)
+    assert len(result) == 2
+    for k, v in result.items():
+        assert k in ["a", "b", "c", "d"]
+        assert isinstance(v, (int, float))
+
+
 def test_transformers_json_array(model):
     prompt = "Output some JSON "
 

From 36f1bf27a80f5877b88a915749341664cef4a1ab Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Thu, 28 Nov 2024 23:15:15 +0100
Subject: [PATCH 118/505] Fix library top-level imports (#1296)

Users are currently running into install issues. After a clean install
of `outlines` they get an error message that asks for `transformers` to
be installed. This should not be the case, as the library is not
required for every integration. In this PR we remove `transformers` and
`datasets` top-level imports, and add per-integration optional
dependencies.

## TODO
- [x] Test `import outlines` from clean install
- [x] Test installing outlines with vLLM optional dependencies
- [x] Test installing outlines with MLX optional dependencies
- [x] Test installing outlines with transformers optional dependencies
- [x] Test installing outlines with llama-cpp optional dependencies
- [x] Test installing outlines with exllamav2 optional dependencies
- [x] Test installing outlines with openai optional dependencies
- [x] Update the documentation

Supersedes #1295.  Fixes #1263.
---
 docs/reference/models/llamacpp.md                      | 6 +++++-
 docs/reference/models/mlxlm.md                         | 6 +++++-
 docs/reference/models/openai.md                        | 6 +++++-
 docs/reference/models/transformers.md                  | 4 ++--
 docs/reference/models/vllm.md                          | 6 +++++-
 outlines/models/transformers.py                        | 4 ++--
 outlines/models/vllm.py                                | 7 ++++---
 pyproject.toml                                         | 9 ++++++++-
 tests/generate/test_integration_transformers_vision.py | 2 +-
 9 files changed, 37 insertions(+), 13 deletions(-)

diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
index 24b0fdc97..51b62eca8 100644
--- a/docs/reference/models/llamacpp.md
+++ b/docs/reference/models/llamacpp.md
@@ -4,7 +4,11 @@ Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/l
 
 !!! Note "Installation"
 
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration. See the [installation section](#installation) for instructions to install `llama-cpp-python` with CUDA, Metal, ROCm and other backends.
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration. See the [installation section](#installation) for instructions to install `llama-cpp-python` with CUDA, Metal, ROCm and other backends. To get started quickly you can also run:
+
+    ```bash
+    pip install "outlines[llamacpp]"
+    ```
 
 ## Load the model
 
diff --git a/docs/reference/models/mlxlm.md b/docs/reference/models/mlxlm.md
index cf7bb7443..d435b9c1f 100644
--- a/docs/reference/models/mlxlm.md
+++ b/docs/reference/models/mlxlm.md
@@ -4,7 +4,11 @@ Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx
 
 !!! Note "Installation"
 
-    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration.
+    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To get started quickly you can also run:
+
+    ```bash
+    pip install "outlines[mlxlm]"
+    ```
 
 
 ## Load the model
diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
index 5c737c916..638107568 100644
--- a/docs/reference/models/openai.md
+++ b/docs/reference/models/openai.md
@@ -2,7 +2,11 @@
 
 !!! Installation
 
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines.
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively:
+
+    ```bash
+    pip install "outlines[openai]"
+    ```
 
 ## OpenAI models
 
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 2a13e28ec..f4c319540 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -3,10 +3,10 @@
 
 !!! Installation
 
-    You need to install the `transformer`, `datasets` and `torch` libraries to be able to use these models in Outlines:
+    You need to install the `transformer`, `datasets` and `torch` libraries to be able to use these models in Outlines, or alternatively:
 
     ```bash
-    pip install torch transformers datasets
+    pip install "outlines[transformers]"
     ```
 
 
diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index fb1c830fa..8789b588e 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -3,7 +3,11 @@
 
 !!! Note "Installation"
 
-    You need to install the `vllm` library to use the vLLM integration. See the [installation section](#installation) for instructions to install vLLM for CPU or ROCm.
+    You need to install the `vllm` library to use the vLLM integration. See the [installation section](#installation) for instructions to install vLLM for CPU or ROCm. To get started you can also run:
+
+    ```bash
+    pip install "outlines[vllm]"
+    ```
 
 ## Load the model
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 7ecc9013f..444492500 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -2,8 +2,6 @@
 import inspect
 from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, Union
 
-from datasets.fingerprint import Hasher
-
 from outlines.generate.api import GenerationParameters, SamplingParameters
 from outlines.models.tokenizer import Tokenizer
 
@@ -116,6 +114,8 @@ def __eq__(self, other):
         return NotImplemented
 
     def __hash__(self):
+        from datasets.fingerprint import Hasher
+
         return hash(Hasher.hash(self.tokenizer))
 
     def __getstate__(self):
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index d1f97bde2..778c27c6f 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,11 +1,10 @@
 import dataclasses
 from typing import TYPE_CHECKING, List, Optional, Union
 
-from transformers import SPIECE_UNDERLINE, PreTrainedTokenizerBase
-
 from outlines.generate.api import GenerationParameters, SamplingParameters
 
 if TYPE_CHECKING:
+    from transformers import PreTrainedTokenizerBase
     from vllm import LLM
     from vllm.sampling_params import SamplingParams
 
@@ -188,7 +187,7 @@ def vllm(model_name: str, **vllm_model_params):
     return VLLM(model)
 
 
-def adapt_tokenizer(tokenizer: PreTrainedTokenizerBase) -> PreTrainedTokenizerBase:
+def adapt_tokenizer(tokenizer: "PreTrainedTokenizerBase") -> "PreTrainedTokenizerBase":
     """Adapt a tokenizer to use to compile the FSM.
 
     The API of Outlines tokenizers is slightly different to that of `transformers`. In
@@ -205,6 +204,8 @@ def adapt_tokenizer(tokenizer: PreTrainedTokenizerBase) -> PreTrainedTokenizerBa
     PreTrainedTokenizerBase
         The adapted tokenizer.
     """
+    from transformers import SPIECE_UNDERLINE
+
     tokenizer.vocabulary = tokenizer.get_vocab()
     tokenizer.special_tokens = set(tokenizer.all_special_tokens)
 
diff --git a/pyproject.toml b/pyproject.toml
index 294fbe4b8..5b005cfbd 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -36,7 +36,6 @@ dependencies = [
    "jsonschema",
    "requests",
    "tqdm",
-   "datasets",
    "typing_extensions",
    "pycountry",
    "airportsdata",
@@ -46,6 +45,12 @@ dependencies = [
 dynamic = ["version"]
 
 [project.optional-dependencies]
+vllm = ["vllm", "transformers", "numpy2"]
+transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
+mlxlm = ["mlx-lm", "datasets"]
+openai = ["openai"]
+llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
+exllamav2 = ["exllamav2"]
 test = [
     "pre-commit",
     "pytest",
@@ -61,10 +66,12 @@ test = [
     "mlx-lm>=0.19.2; platform_machine == 'arm64' and sys_platform == 'darwin'",
     "huggingface_hub",
     "openai>=1.0.0",
+    "datasets",
     "vllm; sys_platform != 'darwin'",
     "transformers",
     "pillow",
     "exllamav2",
+    "jax"
 ]
 serve = [
     "vllm>=0.3.0",
diff --git a/tests/generate/test_integration_transformers_vision.py b/tests/generate/test_integration_transformers_vision.py
index 28b516c57..ee4f84c06 100644
--- a/tests/generate/test_integration_transformers_vision.py
+++ b/tests/generate/test_integration_transformers_vision.py
@@ -23,7 +23,7 @@ def img_from_url(url):
 @pytest.fixture(scope="session")
 def model(tmp_path_factory):
     return transformers_vision(
-        "trl-internal-testing/tiny-random-LlavaForConditionalGeneration",
+        "trl-internal-testing/tiny-LlavaForConditionalGeneration",
         model_class=LlavaForConditionalGeneration,
         device="cpu",
     )

From c8e69c6b9b81fc9f64328d9d3dac9e032f94efbe Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 6 Dec 2024 15:08:24 +0100
Subject: [PATCH 119/505] Bump the outlines core version to 0.1.20

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 5b005cfbd..f738c965f 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,7 @@ dependencies = [
    "pycountry",
    "airportsdata",
    "torch",
-   "outlines_core==0.1.17",
+   "outlines_core==0.1.20",
 ]
 dynamic = ["version"]
 

From dccaaced6c9972f1c75dab0de83afbd68cfde933 Mon Sep 17 00:00:00 2001
From: Hicham Randrianarivo <hicham.randrianarivo@gmail.com>
Date: Tue, 3 Dec 2024 10:49:34 +0100
Subject: [PATCH 120/505] Fix typo in vllm extra requirements

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index f738c965f..2001f4c40 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -45,7 +45,7 @@ dependencies = [
 dynamic = ["version"]
 
 [project.optional-dependencies]
-vllm = ["vllm", "transformers", "numpy2"]
+vllm = ["vllm", "transformers", "numpy<2"]
 transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
 mlxlm = ["mlx-lm", "datasets"]
 openai = ["openai"]

From e4f96fbce45593222e40805ca614ace251728ef2 Mon Sep 17 00:00:00 2001
From: g-prz <158364710+g-prz@users.noreply.github.com>
Date: Sun, 8 Dec 2024 15:40:24 +0000
Subject: [PATCH 121/505] Handle enum in choice (#1279)

This PR aims at solving #1218 (requires #1277 to be merged)

Quick fix to solve #1275
---
 .gitignore                      |  1 +
 README.md                       | 23 +++++++++++++++++++++++
 outlines/generate/choice.py     | 18 ++++++++++++++----
 tests/generate/test_generate.py | 31 ++++++++++++++++++++++++++++---
 4 files changed, 66 insertions(+), 7 deletions(-)

diff --git a/.gitignore b/.gitignore
index 4984b18cb..08390ae3d 100644
--- a/.gitignore
+++ b/.gitignore
@@ -7,6 +7,7 @@ docs/build
 *.gguf
 .venv
 benchmarks/results
+.python-version
 
 # Remove doc build folders
 .cache/
diff --git a/README.md b/README.md
index e9285e863..d34b0984d 100644
--- a/README.md
+++ b/README.md
@@ -84,6 +84,29 @@ generator = outlines.generate.choice(model, ["Positive", "Negative"])
 answer = generator(prompt)
 ```
 
+You can also pass these choices through en enum:
+
+````python
+from enum import Enum
+
+import outlines
+
+class Sentiment(str, Enum):
+    positive = "Positive"
+    negative = "Negative"
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+
+prompt = """You are a sentiment-labelling assistant.
+Is the following review positive or negative?
+
+Review: This restaurant is just awesome!
+"""
+
+generator = outlines.generate.choice(model, Sentiment)
+answer = generator(prompt)
+````
+
 ### Type constraint
 
 You can instruct the model to only return integers or floats:
diff --git a/outlines/generate/choice.py b/outlines/generate/choice.py
index 595513d52..75fc71271 100644
--- a/outlines/generate/choice.py
+++ b/outlines/generate/choice.py
@@ -1,7 +1,10 @@
 import json as pyjson
+import re
+from enum import Enum
 from functools import singledispatch
-from typing import Callable, List
+from typing import Callable, List, Union
 
+from outlines.fsm.json_schema import build_regex_from_schema, get_schema_from_enum
 from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
@@ -12,12 +15,19 @@
 
 @singledispatch
 def choice(
-    model, choices: List[str], sampler: Sampler = multinomial()
+    model, choices: Union[List[str], type[Enum]], sampler: Sampler = multinomial()
 ) -> SequenceGeneratorAdapter:
-    regex_str = r"(" + r"|".join(choices) + r")"
+    if isinstance(choices, type(Enum)):
+        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
+    else:
+        choices = [re.escape(choice) for choice in choices]  # type: ignore
+        regex_str = r"(" + r"|".join(choices) + r")"
 
     generator = regex(model, regex_str, sampler)
-    generator.format_sequence = lambda x: x
+    if isinstance(choices, type(Enum)):
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    else:
+        generator.format_sequence = lambda x: x
 
     return generator
 
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 9c288c21e..f91bc8653 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -1,5 +1,6 @@
 import contextlib
 import re
+from enum import Enum
 
 import pytest
 
@@ -127,6 +128,18 @@ def model_t5(tmp_path_factory):
 )
 
 
+class MyEnum(Enum):
+    foo = "foo"
+    bar = "bar"
+    baz = "baz"
+
+
+ALL_SAMPLE_CHOICES_FIXTURES = (
+    ["foo", "bar", "baz"],
+    MyEnum,
+)
+
+
 ##########################################
 # Stuctured Generation Inputs
 ##########################################
@@ -264,21 +277,33 @@ def test_generate_json(request, model_fixture, sample_schema):
 
 
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
 def test_generate_choice(request, model_fixture, sample_choices):
     model = request.getfixturevalue(model_fixture)
     generator = generate.choice(model, sample_choices)
     res = generator(**get_inputs(model_fixture))
-    assert res in sample_choices
+    if isinstance(sample_choices, type(Enum)):
+        assert res in [elt.value for elt in sample_choices]
+    else:
+        assert res in sample_choices
 
 
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
 def test_generate_choice_twice(request, model_fixture, sample_choices):
     model = request.getfixturevalue(model_fixture)
     generator = generate.choice(model, sample_choices)
     res = generator(**get_inputs(model_fixture))
-    assert res in sample_choices
+    if isinstance(sample_choices, type(Enum)):
+        assert res in [elt.value for elt in sample_choices]
+    else:
+        assert res in sample_choices
+
     res = generator(**get_inputs(model_fixture))
-    assert res in sample_choices
+    if isinstance(sample_choices, type(Enum)):
+        assert res in [elt.value for elt in sample_choices]
+    else:
+        assert res in sample_choices
 
 
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)

From 1b7ec18a91b6d39e0279dd91cfe9bab415dc80c4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 2 Dec 2024 15:50:52 +0100
Subject: [PATCH 122/505] Use `outlines-core` to translate JSON Schemas into
 regexes

---
 benchmarks/bench_json_schema.py               |    7 +-
 outlines/fsm/json_schema.py                   |  489 +-------
 outlines/generate/choice.py                   |    4 +-
 outlines/generate/json.py                     |    7 +-
 outlines/processors/structured.py             |    3 +-
 tests/fsm/test_json_schema.py                 | 1032 +----------------
 .../generate/test_integration_transformers.py |    1 +
 7 files changed, 17 insertions(+), 1526 deletions(-)

diff --git a/benchmarks/bench_json_schema.py b/benchmarks/bench_json_schema.py
index 62d9b3c1d..3a1f72cb6 100644
--- a/benchmarks/bench_json_schema.py
+++ b/benchmarks/bench_json_schema.py
@@ -1,6 +1,7 @@
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
 from outlines.caching import cache_disabled
 from outlines.fsm.guide import RegexGuide
-from outlines.fsm.json_schema import build_regex_from_schema
 
 from .common import setup_tokenizer  # noqa: E402
 
@@ -70,10 +71,6 @@ def setup(self, schema_name):
         self.tokenizer = setup_tokenizer()
         self.schema = schemas[schema_name]
 
-    @cache_disabled()
-    def time_json_schema_to_regex(self, schema_name):
-        build_regex_from_schema(self.schema)
-
     @cache_disabled()
     def time_json_schema_to_fsm(self, schema_name):
         regex = build_regex_from_schema(self.schema)
diff --git a/outlines/fsm/json_schema.py b/outlines/fsm/json_schema.py
index 0bab57923..bae0ad17a 100644
--- a/outlines/fsm/json_schema.py
+++ b/outlines/fsm/json_schema.py
@@ -1,90 +1,10 @@
 import inspect
 import json
-import re
 import warnings
 from enum import Enum
-from typing import Callable, Optional, Tuple, Type, Union
+from typing import Callable, Type, Union
 
-from jsonschema.protocols import Validator
 from pydantic import BaseModel, create_model
-from referencing import Registry, Resource
-from referencing._core import Resolver
-from referencing.jsonschema import DRAFT202012
-
-# allow `\"`, `\\`, or any character which isn't a control sequence
-STRING_INNER = r'([^"\\\x00-\x1F\x7F-\x9F]|\\["\\])'
-STRING = f'"{STRING_INNER}*"'
-
-INTEGER = r"(-)?(0|[1-9][0-9]*)"
-NUMBER = rf"({INTEGER})(\.[0-9]+)?([eE][+-][0-9]+)?"
-BOOLEAN = r"(true|false)"
-NULL = r"null"
-WHITESPACE = r"[ ]?"
-
-type_to_regex = {
-    "string": STRING,
-    "integer": INTEGER,
-    "number": NUMBER,
-    "boolean": BOOLEAN,
-    "null": NULL,
-}
-
-DATE_TIME = r'"(-?(?:[1-9][0-9]*)?[0-9]{4})-(1[0-2]|0[1-9])-(3[01]|0[1-9]|[12][0-9])T(2[0-3]|[01][0-9]):([0-5][0-9]):([0-5][0-9])(\.[0-9]{3})?(Z)?"'
-DATE = r'"(?:\d{4})-(?:0[1-9]|1[0-2])-(?:0[1-9]|[1-2][0-9]|3[0-1])"'
-TIME = r'"(2[0-3]|[01][0-9]):([0-5][0-9]):([0-5][0-9])(\\.[0-9]+)?(Z)?"'
-UUID = r'"[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}"'
-
-format_to_regex = {
-    "uuid": UUID,
-    "date-time": DATE_TIME,
-    "date": DATE,
-    "time": TIME,
-}
-
-
-def build_regex_from_schema(schema: str, whitespace_pattern: Optional[str] = None):
-    """Turn a JSON schema into a regex that matches any JSON object that follows
-    this schema.
-
-    JSON Schema is a declarative language that allows to annotate JSON documents
-    with types and descriptions. These schemas can be generated from any Python
-    datastructure that has type annotation: namedtuples, dataclasses, Pydantic
-    models. And by ensuring that the generation respects the schema we ensure
-    that the output can be parsed into these objects.
-    This function parses the provided schema and builds a generation schedule which
-    mixes deterministic generation (fixed strings), and sampling with constraints.
-
-    Parameters
-    ----------
-    schema
-        A string that represents a JSON Schema.
-    whitespace_pattern
-        Pattern to use for JSON syntactic whitespace (doesn't impact string literals)
-        Example: allow only a single space or newline with `whitespace_pattern=r"[\n ]?"`
-
-    Returns
-    -------
-    A generation schedule. A list of strings that represent the JSON
-    schema's structure and regular expression that define the structure of
-    the fields.
-
-    References
-    ----------
-    .. [0] JSON Schema. https://json-schema.org/
-
-    """
-
-    schema = json.loads(schema)
-    Validator.check_schema(schema)
-
-    # Build reference resolver
-    schema = Resource(contents=schema, specification=DRAFT202012)
-    uri = schema.id() if schema.id() is not None else ""
-    registry = Registry().with_resource(uri=uri, resource=schema)
-    resolver = registry.resolver()
-
-    content = schema.contents
-    return to_regex(resolver, content, whitespace_pattern)
 
 
 def convert_json_schema_to_str(json_schema: Union[dict, str, Type[BaseModel]]) -> str:
@@ -120,413 +40,6 @@ def convert_json_schema_to_str(json_schema: Union[dict, str, Type[BaseModel]]) -
     return schema_str
 
 
-def _get_num_items_pattern(min_items, max_items, whitespace_pattern):
-    # Helper function for arrays and objects
-    min_items = int(min_items or 0)
-    if max_items is None:
-        return rf"{{{max(min_items - 1, 0)},}}"
-    else:
-        max_items = int(max_items)
-        if max_items < 1:
-            return None
-        return rf"{{{max(min_items - 1, 0)},{max_items - 1}}}"
-
-
-def validate_quantifiers(
-    min_bound: Optional[str], max_bound: Optional[str], start_offset: int = 0
-) -> Tuple[str, str]:
-    """
-    Ensures that the bounds of a number are valid. Bounds are used as quantifiers in the regex.
-
-    Parameters
-    ----------
-    min_bound
-        The minimum value that the number can take.
-    max_bound
-        The maximum value that the number can take.
-    start_offset
-        Number of elements that are already present in the regex but still need to be counted.
-        ex: if the regex is already "(-)?(0|[1-9][0-9])", we will always have at least 1 digit, so the start_offset is 1.
-
-    Returns
-    -------
-    min_bound
-        The minimum value that the number can take.
-    max_bound
-        The maximum value that the number can take.
-
-    Raises
-    ------
-    ValueError
-        If the minimum bound is greater than the maximum bound.
-
-    TypeError or ValueError
-        If the minimum bound is not an integer or None.
-        or
-        If the maximum bound is not an integer or None.
-    """
-    min_bound = "" if min_bound is None else str(int(min_bound) - start_offset)
-    max_bound = "" if max_bound is None else str(int(max_bound) - start_offset)
-    if min_bound and max_bound:
-        if int(max_bound) < int(min_bound):
-            raise ValueError("max bound must be greater than or equal to min bound")
-    return min_bound, max_bound
-
-
-def to_regex(
-    resolver: Resolver, instance: dict, whitespace_pattern: Optional[str] = None
-):
-    """Translate a JSON Schema instance into a regex that validates the schema.
-
-    Note
-    ----
-    Many features of JSON schema are missing:
-    - Handle `additionalProperties` keyword
-    - Handle types defined as a list
-    - Handle constraints on numbers
-    - Handle special patterns: `date`, `uri`, etc.
-
-    This does not support recursive definitions.
-
-    Parameters
-    ----------
-    resolver
-        An object that resolves references to other instances within a schema
-    instance
-        The instance to translate
-    whitespace_pattern
-        Pattern to use for JSON syntactic whitespace (doesn't impact string literals)
-        Example: allow only a single space or newline with `whitespace_pattern=r"[\n ]?"`
-    """
-
-    # set whitespace pattern
-    if whitespace_pattern is None:
-        whitespace_pattern = WHITESPACE
-
-    if instance == {}:
-        # JSON Schema Spec: Empty object means unconstrained, any json type is legal
-        types = [
-            {"type": "boolean"},
-            {"type": "null"},
-            {"type": "number"},
-            {"type": "integer"},
-            {"type": "string"},
-            {"type": "array"},
-            {"type": "object"},
-        ]
-        regexes = [to_regex(resolver, t, whitespace_pattern) for t in types]
-        regexes = [rf"({r})" for r in regexes]
-        return rf"{'|'.join(regexes)}"
-
-    elif "properties" in instance:
-        regex = ""
-        regex += r"\{"
-        properties = instance["properties"]
-        required_properties = instance.get("required", [])
-        is_required = [item in required_properties for item in properties]
-        # If at least one property is required, we include the one in the lastest position
-        # without any comma.
-        # For each property before it (optional or required), we add with a comma after the property.
-        # For each property after it (optional), we add with a comma before the property.
-        if any(is_required):
-            last_required_pos = max([i for i, value in enumerate(is_required) if value])
-            for i, (name, value) in enumerate(properties.items()):
-                subregex = f'{whitespace_pattern}"{re.escape(name)}"{whitespace_pattern}:{whitespace_pattern}'
-                subregex += to_regex(resolver, value, whitespace_pattern)
-                if i < last_required_pos:
-                    subregex = f"{subregex}{whitespace_pattern},"
-                elif i > last_required_pos:
-                    subregex = f"{whitespace_pattern},{subregex}"
-                regex += subregex if is_required[i] else f"({subregex})?"
-        # If no property is required, we have to create a possible pattern for each property in which
-        # it's the last one necessarilly present. Then, we add the others as optional before and after
-        # following the same strategy as described above.
-        # The whole block is made optional to allow the case in which no property is returned.
-        else:
-            property_subregexes = []
-            for i, (name, value) in enumerate(properties.items()):
-                subregex = f'{whitespace_pattern}"{name}"{whitespace_pattern}:{whitespace_pattern}'
-                subregex += to_regex(resolver, value, whitespace_pattern)
-                property_subregexes.append(subregex)
-            possible_patterns = []
-            for i in range(len(property_subregexes)):
-                pattern = ""
-                for subregex in property_subregexes[:i]:
-                    pattern += f"({subregex}{whitespace_pattern},)?"
-                pattern += property_subregexes[i]
-                for subregex in property_subregexes[i + 1 :]:
-                    pattern += f"({whitespace_pattern},{subregex})?"
-                possible_patterns.append(pattern)
-            regex += f"({'|'.join(possible_patterns)})?"
-
-        regex += f"{whitespace_pattern}" + r"\}"
-
-        return regex
-
-    # To validate against allOf, the given data must be valid against all of the
-    # given subschemas.
-    elif "allOf" in instance:
-        subregexes = [
-            to_regex(resolver, t, whitespace_pattern) for t in instance["allOf"]
-        ]
-        subregexes_str = [f"{subregex}" for subregex in subregexes]
-        return rf"({''.join(subregexes_str)})"
-
-    # To validate against `anyOf`, the given data must be valid against
-    # any (one or more) of the given subschemas.
-    elif "anyOf" in instance:
-        subregexes = [
-            to_regex(resolver, t, whitespace_pattern) for t in instance["anyOf"]
-        ]
-        return rf"({'|'.join(subregexes)})"
-
-    # To validate against oneOf, the given data must be valid against exactly
-    # one of the given subschemas.
-    elif "oneOf" in instance:
-        subregexes = [
-            to_regex(resolver, t, whitespace_pattern) for t in instance["oneOf"]
-        ]
-
-        xor_patterns = [f"(?:{subregex})" for subregex in subregexes]
-
-        return rf"({'|'.join(xor_patterns)})"
-
-    # Create pattern for Tuples, per JSON Schema spec, `prefixItems` determines types at each idx
-    elif "prefixItems" in instance:
-        element_patterns = [
-            to_regex(resolver, t, whitespace_pattern) for t in instance["prefixItems"]
-        ]
-        comma_split_pattern = rf"{whitespace_pattern},{whitespace_pattern}"
-        tuple_inner = comma_split_pattern.join(element_patterns)
-        return rf"\[{whitespace_pattern}{tuple_inner}{whitespace_pattern}\]"
-
-    # The enum keyword is used to restrict a value to a fixed set of values. It
-    # must be an array with at least one element, where each element is unique.
-    elif "enum" in instance:
-        choices = []
-        for choice in instance["enum"]:
-            if type(choice) in [int, float, bool, type(None), str]:
-                choices.append(re.escape(json.dumps(choice)))
-            elif isinstance(choice, dict):
-                choices.append(to_regex(resolver, choice, whitespace_pattern))
-            else:
-                raise TypeError(f"Unsupported data type in enum: {type(choice)}")
-        return f"({'|'.join(choices)})"
-
-    elif "const" in instance:
-        const = instance["const"]
-        if type(const) in [int, float, bool, type(None), str]:
-            const = re.escape(json.dumps(const))
-        else:
-            raise TypeError(f"Unsupported data type in const: {type(const)}")
-        return const
-
-    elif "$ref" in instance:
-        path = f"{instance['$ref']}"
-        instance = resolver.lookup(path).contents
-        return to_regex(resolver, instance, whitespace_pattern)
-
-    # The type keyword may either be a string or an array:
-    # - If it's a string, it is the name of one of the basic types.
-    # - If it is an array, it must be an array of strings, where each string is
-    # the name of one of the basic types, and each element is unique. In this
-    # case, the JSON snippet is valid if it matches any of the given types.
-    elif "type" in instance:
-        instance_type = instance["type"]
-        if instance_type == "string":
-            if "maxLength" in instance or "minLength" in instance:
-                max_items = instance.get("maxLength", "")
-                min_items = instance.get("minLength", "")
-                try:
-                    if int(max_items) < int(min_items):
-                        raise ValueError(
-                            "maxLength must be greater than or equal to minLength"
-                        )  # FIXME this raises an error but is caught right away by the except (meant for int("") I assume)
-                except ValueError:
-                    pass
-                return f'"{STRING_INNER}{{{min_items},{max_items}}}"'
-            elif "pattern" in instance:
-                pattern = instance["pattern"]
-                if pattern[0] == "^" and pattern[-1] == "$":
-                    return rf'("{pattern[1:-1]}")'
-                else:
-                    return rf'("{pattern}")'
-            elif "format" in instance:
-                format = instance["format"]
-                if format == "date-time":
-                    return format_to_regex["date-time"]
-                elif format == "uuid":
-                    return format_to_regex["uuid"]
-                elif format == "date":
-                    return format_to_regex["date"]
-                elif format == "time":
-                    return format_to_regex["time"]
-                else:
-                    raise NotImplementedError(
-                        f"Format {format} is not supported by Outlines"
-                    )
-            else:
-                return type_to_regex["string"]
-
-        elif instance_type == "number":
-            bounds = {
-                "minDigitsInteger",
-                "maxDigitsInteger",
-                "minDigitsFraction",
-                "maxDigitsFraction",
-                "minDigitsExponent",
-                "maxDigitsExponent",
-            }
-            if bounds.intersection(set(instance.keys())):
-                min_digits_integer, max_digits_integer = validate_quantifiers(
-                    instance.get("minDigitsInteger"),
-                    instance.get("maxDigitsInteger"),
-                    start_offset=1,
-                )
-                min_digits_fraction, max_digits_fraction = validate_quantifiers(
-                    instance.get("minDigitsFraction"), instance.get("maxDigitsFraction")
-                )
-                min_digits_exponent, max_digits_exponent = validate_quantifiers(
-                    instance.get("minDigitsExponent"), instance.get("maxDigitsExponent")
-                )
-                integers_quantifier = (
-                    f"{{{min_digits_integer},{max_digits_integer}}}"
-                    if min_digits_integer or max_digits_integer
-                    else "*"
-                )
-                fraction_quantifier = (
-                    f"{{{min_digits_fraction},{max_digits_fraction}}}"
-                    if min_digits_fraction or max_digits_fraction
-                    else "+"
-                )
-                exponent_quantifier = (
-                    f"{{{min_digits_exponent},{max_digits_exponent}}}"
-                    if min_digits_exponent or max_digits_exponent
-                    else "+"
-                )
-                return rf"((-)?(0|[1-9][0-9]{integers_quantifier}))(\.[0-9]{fraction_quantifier})?([eE][+-][0-9]{exponent_quantifier})?"
-            return type_to_regex["number"]
-
-        elif instance_type == "integer":
-            if "minDigits" in instance or "maxDigits" in instance:
-                min_digits, max_digits = validate_quantifiers(
-                    instance.get("minDigits"), instance.get("maxDigits"), start_offset=1
-                )
-                return rf"(-)?(0|[1-9][0-9]{{{min_digits},{max_digits}}})"
-            return type_to_regex["integer"]
-
-        elif instance_type == "array":
-            num_repeats = _get_num_items_pattern(
-                instance.get("minItems"), instance.get("maxItems"), whitespace_pattern
-            )
-            if num_repeats is None:
-                return rf"\[{whitespace_pattern}\]"
-
-            allow_empty = "?" if int(instance.get("minItems", 0)) == 0 else ""
-
-            if "items" in instance:
-                items_regex = to_regex(resolver, instance["items"], whitespace_pattern)
-                return rf"\[{whitespace_pattern}(({items_regex})(,{whitespace_pattern}({items_regex})){num_repeats}){allow_empty}{whitespace_pattern}\]"
-            else:
-                # Here we need to make the choice to exclude generating list of objects
-                # if the specification of the object is not given, even though a JSON
-                # object that contains an object here would be valid under the specification.
-                legal_types = [
-                    {"type": "boolean"},
-                    {"type": "null"},
-                    {"type": "number"},
-                    {"type": "integer"},
-                    {"type": "string"},
-                ]
-                depth = instance.get("depth", 2)
-                if depth > 0:
-                    legal_types.append({"type": "object", "depth": depth - 1})
-                    legal_types.append({"type": "array", "depth": depth - 1})
-
-                regexes = [
-                    to_regex(resolver, t, whitespace_pattern) for t in legal_types
-                ]
-                return rf"\[{whitespace_pattern}({'|'.join(regexes)})(,{whitespace_pattern}({'|'.join(regexes)})){num_repeats}{allow_empty}{whitespace_pattern}\]"
-
-        elif instance_type == "object":
-            # pattern for json object with values defined by instance["additionalProperties"]
-            # enforces value type constraints recursively, "minProperties", and "maxProperties"
-            # doesn't enforce "required", "dependencies", "propertyNames" "any/all/on Of"
-            num_repeats = _get_num_items_pattern(
-                instance.get("minProperties"),
-                instance.get("maxProperties"),
-                whitespace_pattern,
-            )
-            if num_repeats is None:
-                return rf"\{{{whitespace_pattern}\}}"
-
-            allow_empty = "?" if int(instance.get("minProperties", 0)) == 0 else ""
-
-            additional_properties = instance.get("additionalProperties")
-
-            if additional_properties is None or additional_properties is True:
-                # JSON Schema behavior: If the additionalProperties of an object is
-                # unset or True, it is unconstrained object.
-                # We handle this by setting additionalProperties to anyOf: {all types}
-
-                legal_types = [
-                    {"type": "string"},
-                    {"type": "number"},
-                    {"type": "boolean"},
-                    {"type": "null"},
-                ]
-
-                # We set the object depth to 2 to keep the expression finite, but the "depth"
-                # key is not a true component of the JSON Schema specification.
-                depth = instance.get("depth", 2)
-                if depth > 0:
-                    legal_types.append({"type": "object", "depth": depth - 1})
-                    legal_types.append({"type": "array", "depth": depth - 1})
-                additional_properties = {"anyOf": legal_types}
-
-            value_pattern = to_regex(
-                resolver, additional_properties, whitespace_pattern
-            )
-            key_value_pattern = (
-                f"{STRING}{whitespace_pattern}:{whitespace_pattern}{value_pattern}"
-            )
-            key_value_successor_pattern = (
-                f"{whitespace_pattern},{whitespace_pattern}{key_value_pattern}"
-            )
-            multiple_key_value_pattern = f"({key_value_pattern}({key_value_successor_pattern}){num_repeats}){allow_empty}"
-
-            return (
-                r"\{"
-                + whitespace_pattern
-                + multiple_key_value_pattern
-                + whitespace_pattern
-                + r"\}"
-            )
-
-        elif instance_type == "boolean":
-            return type_to_regex["boolean"]
-
-        elif instance_type == "null":
-            return type_to_regex["null"]
-
-        elif isinstance(instance_type, list):
-            # Here we need to make the choice to exclude generating an object
-            # if the specification of the object is not give, even though a JSON
-            # object that contains an object here would be valid under the specification.
-            regexes = [
-                to_regex(resolver, {"type": t}, whitespace_pattern)
-                for t in instance_type
-                if t != "object"
-            ]
-            return rf"({'|'.join(regexes)})"
-
-    raise NotImplementedError(
-        f"""Could not translate the instance {instance} to a
-    regular expression. Make sure it is valid to the JSON Schema specification. If
-    it is, please open an issue on the Outlines repository"""
-    )
-
-
 def get_schema_from_signature(fn: Callable) -> dict:
     """Turn a function signature into a JSON schema.
 
diff --git a/outlines/generate/choice.py b/outlines/generate/choice.py
index 75fc71271..afb998f52 100644
--- a/outlines/generate/choice.py
+++ b/outlines/generate/choice.py
@@ -4,7 +4,9 @@
 from functools import singledispatch
 from typing import Callable, List, Union
 
-from outlines.fsm.json_schema import build_regex_from_schema, get_schema_from_enum
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+from outlines.fsm.json_schema import get_schema_from_enum
 from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
diff --git a/outlines/generate/json.py b/outlines/generate/json.py
index 703447958..d098d920d 100644
--- a/outlines/generate/json.py
+++ b/outlines/generate/json.py
@@ -3,13 +3,10 @@
 from functools import singledispatch
 from typing import Callable, Optional, Union
 
+from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel
 
-from outlines.fsm.json_schema import (
-    build_regex_from_schema,
-    get_schema_from_enum,
-    get_schema_from_signature,
-)
+from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
 from outlines.generate.api import SequenceGeneratorAdapter
 from outlines.models import OpenAI
 from outlines.samplers import Sampler, multinomial
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index d2bc15f77..64892b73f 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -27,10 +27,11 @@
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
 
 import torch
+from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel
 
 from outlines.fsm.guide import CFGGuide, Guide, RegexGuide
-from outlines.fsm.json_schema import build_regex_from_schema, convert_json_schema_to_str
+from outlines.fsm.json_schema import convert_json_schema_to_str
 
 from .base_logits_processor import OutlinesLogitsProcessor
 
diff --git a/tests/fsm/test_json_schema.py b/tests/fsm/test_json_schema.py
index 6f0b59c50..825433397 100644
--- a/tests/fsm/test_json_schema.py
+++ b/tests/fsm/test_json_schema.py
@@ -1,31 +1,14 @@
 import json
-import re
 from contextlib import nullcontext
 from enum import Enum
 from functools import partial
-from typing import List, Literal, Union
+from typing import List
 
-import interegular
 import pytest
-from pydantic import BaseModel, Field, constr
+from outlines_core.fsm.json_schema import build_regex_from_schema
+from pydantic import BaseModel, constr
 
-from outlines.fsm.json_schema import (
-    BOOLEAN,
-    DATE,
-    DATE_TIME,
-    INTEGER,
-    NULL,
-    NUMBER,
-    STRING,
-    STRING_INNER,
-    TIME,
-    UUID,
-    WHITESPACE,
-    build_regex_from_schema,
-    get_schema_from_enum,
-    get_schema_from_signature,
-    to_regex,
-)
+from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
 
 
 def test_function_basic():
@@ -58,1011 +41,8 @@ class User(BaseModel):
         is_true: bool
 
     schema = json.dumps(User.model_json_schema())
-    schedule = build_regex_from_schema(schema)
-    assert isinstance(schedule, str)
-
-
-@pytest.mark.parametrize(
-    "pattern,does_match",
-    [
-        ({"integer": "0"}, True),
-        ({"integer": "1"}, True),
-        ({"integer": "-1"}, True),
-        ({"integer": "01"}, False),
-        ({"integer": "1.3"}, False),
-        ({"integer": "t"}, False),
-    ],
-)
-def test_match_integer(pattern, does_match):
-    step = {"title": "Foo", "type": "integer"}
-    regex = to_regex(None, step)
-    assert regex == INTEGER
-
-    value = pattern["integer"]
-    match = re.fullmatch(regex, value)
-    if does_match:
-        assert match[0] == value
-        assert match.span() == (0, len(value))
-    else:
-        assert match is None
-
-
-@pytest.mark.parametrize(
-    "pattern,does_match",
-    [
-        ({"number": "1"}, True),
-        ({"number": "0"}, True),
-        ({"number": "01"}, False),
-        ({"number": ".3"}, False),
-        ({"number": "1.3"}, True),
-        ({"number": "-1.3"}, True),
-        ({"number": "1.3e9"}, False),
-        ({"number": "1.3e+9"}, True),
-    ],
-)
-def test_match_number(pattern, does_match):
-    step = {"title": "Foo", "type": "number"}
-    regex = to_regex(None, step)
-    assert regex == NUMBER
-
-    value = pattern["number"]
-    match = re.fullmatch(regex, value)
-    if does_match:
-        assert match[0] == value
-        assert match.span() == (0, len(value))
-    else:
-        assert match is None
-
-
-@pytest.mark.parametrize(
-    "schema,regex,examples",
-    [
-        # String
-        (
-            {"title": "Foo", "type": "string"},
-            STRING,
-            [
-                ("unquotedstring", False),
-                ('"(parenthesized_string)"', True),
-                ('"malformed) parenthesis (((() string"', True),
-                ('"quoted_string"', True),
-                (r'"escape_\character"', False),
-                (r'"double_\\escape"', True),
-                (r'"\n"', False),
-                (r'"\\n"', True),
-                (r'"unescaped " quote"', False),
-                (r'"escaped \" quote"', True),
-            ],
-        ),
-        # String with maximum length
-        (
-            {"title": "Foo", "type": "string", "maxLength": 3},
-            f'"{STRING_INNER}{{,3}}"',
-            [('"ab"', True), ('"a""', False), ('"abcd"', False)],
-        ),
-        # String with minimum length
-        (
-            {"title": "Foo", "type": "string", "minLength": 3},
-            f'"{STRING_INNER}{{3,}}"',
-            [('"ab"', False), ('"abcd"', True), ('"abc""', False)],
-        ),
-        # String with both minimum and maximum length
-        (
-            {"title": "Foo", "type": "string", "minLength": 3, "maxLength": 5},
-            f'"{STRING_INNER}{{3,5}}"',
-            [('"ab"', False), ('"abcd"', True), ('"abcdef""', False)],
-        ),
-        # String defined by a regular expression
-        (
-            {"title": "Foo", "type": "string", "pattern": r"^[a-z]$"},
-            r'("[a-z]")',
-            [('"a"', True), ('"1"', False)],
-        ),
-        # Boolean
-        (
-            {"title": "Foo", "type": "boolean"},
-            BOOLEAN,
-            [
-                ("true", True),
-                ("false", True),
-                ("null", False),
-                ("0", False),
-            ],
-        ),
-        # Null
-        (
-            {"title": "Foo", "type": "null"},
-            NULL,
-            [
-                ("null", True),
-                ("true", False),
-                ("0", False),
-            ],
-        ),
-        # Const string
-        (
-            {"title": "Foo", "const": "Marc", "type": "string"},
-            '"Marc"',
-            [('"Marc"', True), ('"Jean"', False), ('"John"', False)],
-        ),
-        # Make sure strings are escaped with regex escaping
-        (
-            {"title": "Foo", "const": ".*", "type": "string"},
-            r'"\.\*"',
-            [('".*"', True), (r'"\s*"', False), (r'"\.\*"', False)],
-        ),
-        # Make sure strings are escaped with JSON escaping
-        (
-            {"title": "Foo", "const": '"', "type": "string"},
-            r'"\\""',
-            [('"\\""', True), ('"""', False)],
-        ),
-        # Const integer
-        (
-            {"title": "Foo", "const": 0, "type": "integer"},
-            "0",
-            [("0", True), ("1", False), ("a", False)],
-        ),
-        # Const float
-        (
-            {"title": "Foo", "const": 0.2, "type": "float"},
-            r"0\.2",
-            [("0.2", True), ("032", False)],
-        ),
-        # Const boolean
-        (
-            {"title": "Foo", "const": True, "type": "boolean"},
-            "true",
-            [("true", True), ("True", False)],
-        ),
-        # Const null
-        (
-            {"title": "Foo", "const": None, "type": "null"},
-            "null",
-            [("null", True), ("None", False), ("", False)],
-        ),
-        # Enum string
-        (
-            {"title": "Foo", "enum": ["Marc", "Jean"], "type": "string"},
-            '("Marc"|"Jean")',
-            [('"Marc"', True), ('"Jean"', True), ('"John"', False)],
-        ),
-        # Make sure strings are escaped with regex and JSON escaping
-        (
-            {"title": "Foo", "enum": [".*", r"\s*"], "type": "string"},
-            r'("\.\*"|"\\\\s\*")',
-            [('".*"', True), (r'"\\s*"', True), (r'"\.\*"', False)],
-        ),
-        # Enum integer
-        (
-            {"title": "Foo", "enum": [0, 1], "type": "integer"},
-            "(0|1)",
-            [("0", True), ("1", True), ("a", False)],
-        ),
-        # Enum mix of types
-        (
-            {
-                "title": "Foo",
-                "enum": [
-                    6,
-                    5.3,
-                    "potato",
-                    True,
-                    None,
-                    {
-                        "properties": {
-                            "a": {"title": "A", "type": "number"},
-                            "b": {"title": "B", "type": "number"},
-                        },
-                        "required": ["a", "b"],
-                        "title": "add",
-                        "type": "object",
-                    },
-                ],
-            },
-            r'(6|5\.3|"potato"|true|null|\{[ ]?"a"[ ]?:[ ]?((-)?(0|[1-9][0-9]*))(\.[0-9]+)?([eE][+-][0-9]+)?[ ]?,[ ]?"b"[ ]?:[ ]?((-)?(0|[1-9][0-9]*))(\.[0-9]+)?([eE][+-][0-9]+)?[ ]?\})',
-            [
-                ("6", True),
-                ("5.3", True),
-                ('"potato"', True),
-                ("true", True),
-                ("null", True),
-                ("523", False),
-                ("True", False),
-                ("None", False),
-                ('{"a": -1.0, "b": 1.1}', True),
-                ('{"a": "a", "b": 1.1}', False),
-            ],
-        ),
-        # integer
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {"count": {"title": "Count", "type": "integer"}},
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?(-)?(0|[1-9][0-9]*)[ ]?\\}',
-            [('{ "count": 100 }', True)],
-        ),
-        # integer with minimum digits
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {
-                    "count": {"title": "Count", "type": "integer", "minDigits": 3}
-                },
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?(-)?(0|[1-9][0-9]{2,})[ ]?\\}',
-            [('{ "count": 10 }', False), ('{ "count": 100 }', True)],
-        ),
-        # integer with maximum digits
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {
-                    "count": {"title": "Count", "type": "integer", "maxDigits": 3}
-                },
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?(-)?(0|[1-9][0-9]{,2})[ ]?\\}',
-            [('{ "count": 100 }', True), ('{ "count": 1000 }', False)],
-        ),
-        # integer with minimum and maximum digits
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {
-                    "count": {
-                        "title": "Count",
-                        "type": "integer",
-                        "minDigits": 3,
-                        "maxDigits": 5,
-                    }
-                },
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?(-)?(0|[1-9][0-9]{2,4})[ ]?\\}',
-            [
-                ('{ "count": 10 }', False),
-                ('{ "count": 100 }', True),
-                ('{ "count": 10000 }', True),
-                ('{ "count": 100000 }', False),
-            ],
-        ),
-        # number
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {"count": {"title": "Count", "type": "number"}},
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?((-)?(0|[1-9][0-9]*))(\\.[0-9]+)?([eE][+-][0-9]+)?[ ]?\\}',
-            [('{ "count": 100 }', True), ('{ "count": 100.5 }', True)],
-        ),
-        # number with min and max integer digits
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {
-                    "count": {
-                        "title": "Count",
-                        "type": "number",
-                        "minDigitsInteger": 3,
-                        "maxDigitsInteger": 5,
-                    }
-                },
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?((-)?(0|[1-9][0-9]{2,4}))(\\.[0-9]+)?([eE][+-][0-9]+)?[ ]?\\}',
-            [
-                ('{ "count": 10.005 }', False),
-                ('{ "count": 100.005 }', True),
-                ('{ "count": 10000.005 }', True),
-                ('{ "count": 100000.005 }', False),
-            ],
-        ),
-        # number with min and max fraction digits
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {
-                    "count": {
-                        "title": "Count",
-                        "type": "number",
-                        "minDigitsFraction": 3,
-                        "maxDigitsFraction": 5,
-                    }
-                },
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?((-)?(0|[1-9][0-9]*))(\\.[0-9]{3,5})?([eE][+-][0-9]+)?[ ]?\\}',
-            [
-                ('{ "count": 1.05 }', False),
-                ('{ "count": 1.005 }', True),
-                ('{ "count": 1.00005 }', True),
-                ('{ "count": 1.000005 }', False),
-            ],
-        ),
-        # number with min and max exponent digits
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {
-                    "count": {
-                        "title": "Count",
-                        "type": "number",
-                        "minDigitsExponent": 3,
-                        "maxDigitsExponent": 5,
-                    }
-                },
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?((-)?(0|[1-9][0-9]*))(\\.[0-9]+)?([eE][+-][0-9]{3,5})?[ ]?\\}',
-            [
-                ('{ "count": 1.05e1 }', False),
-                ('{ "count": 1.05e+001 }', True),
-                ('{ "count": 1.05e-00001 }', True),
-                ('{ "count": 1.05e0000001 }', False),
-            ],
-        ),
-        # number with min and max integer, fraction and exponent digits
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {
-                    "count": {
-                        "title": "Count",
-                        "type": "number",
-                        "minDigitsInteger": 3,
-                        "maxDigitsInteger": 5,
-                        "minDigitsFraction": 3,
-                        "maxDigitsFraction": 5,
-                        "minDigitsExponent": 3,
-                        "maxDigitsExponent": 5,
-                    }
-                },
-                "required": ["count"],
-            },
-            '\\{[ ]?"count"[ ]?:[ ]?((-)?(0|[1-9][0-9]{2,4}))(\\.[0-9]{3,5})?([eE][+-][0-9]{3,5})?[ ]?\\}',
-            [
-                ('{ "count": 1.05e1 }', False),
-                ('{ "count": 100.005e+001 }', True),
-                ('{ "count": 10000.00005e-00001 }', True),
-                ('{ "count": 100000.000005e0000001 }', False),
-            ],
-        ),
-        # array
-        (
-            {"title": "Foo", "type": "array", "items": {"type": "number"}},
-            rf"\[{WHITESPACE}(({NUMBER})(,{WHITESPACE}({NUMBER})){{0,}})?{WHITESPACE}\]",
-            [("[1e+9,1.3]", True), ("[]", True), ("[1", False)],
-        ),
-        # array with a set length of 1
-        (
-            {
-                "title": "Foo",
-                "type": "array",
-                "items": {"type": "integer"},
-                "minItems": 1,
-                "maxItems": 1,
-            },
-            rf"\[{WHITESPACE}(({INTEGER})(,{WHITESPACE}({INTEGER})){{0,0}}){WHITESPACE}\]",
-            [("[1]", True), ("[1,2]", False), ('["a"]', False), ("[]", False)],
-        ),
-        # array with a set length greather than 1
-        (
-            {
-                "title": "Foo",
-                "type": "array",
-                "items": {"type": "integer"},
-                "minItems": 3,
-                "maxItems": 3,
-            },
-            rf"\[{WHITESPACE}(({INTEGER})(,{WHITESPACE}({INTEGER})){{2,2}}){WHITESPACE}\]",
-            [("[1]", False), ("[]", False), ("[1,2,3]", True), ("[1,2,3,4]", False)],
-        ),
-        # array with length 0
-        (
-            {
-                "title": "Foo",
-                "type": "array",
-                "items": {"type": "integer"},
-                "minItems": 0,
-                "maxItems": 0,
-            },
-            rf"\[{WHITESPACE}\]",
-            [("[1]", False), ("[]", True), ("[1,2,3]", False), ("[1,2,3,4]", False)],
-        ),
-        # object
-        (
-            {
-                "title": "TestSchema",
-                "type": "object",
-                "properties": {
-                    "test_dict": {
-                        "title": "Test Dict",
-                        "additionalProperties": {"type": "string"},
-                        "type": "object",
-                    }
-                },
-                "required": ["test_dict"],
-            },
-            rf"""\{{{WHITESPACE}"test_dict"{WHITESPACE}:{WHITESPACE}\{{{WHITESPACE}({STRING}{WHITESPACE}:{WHITESPACE}{STRING}({WHITESPACE},{WHITESPACE}{STRING}{WHITESPACE}:{WHITESPACE}{STRING}){{0,}})?{WHITESPACE}\}}{WHITESPACE}\}}""",
-            [
-                ("""{ "test_dict":{"foo":"bar","baz": "bif"}}""", True),
-                ("""{ "test_dict":{"foo":"bar" }}""", True),
-                ("""{ "test_dict":{}}""", True),
-                ("""{ "WRONG_KEY":{}}""", False),
-                ("""{ "test_dict":{"wrong_type" 1}}""", False),
-            ],
-        ),
-        # object containing object
-        (
-            {
-                "title": "TestSchema",
-                "type": "object",
-                "properties": {
-                    "test_dict": {
-                        "title": "Test Dict",
-                        "additionalProperties": {
-                            "additionalProperties": {"type": "integer"},
-                            "type": "object",
-                        },
-                        "type": "object",
-                    }
-                },
-                "required": ["test_dict"],
-            },
-            rf"""\{{{WHITESPACE}"test_dict"{WHITESPACE}:{WHITESPACE}\{{{WHITESPACE}({STRING}{WHITESPACE}:{WHITESPACE}\{{{WHITESPACE}({STRING}{WHITESPACE}:{WHITESPACE}{INTEGER}({WHITESPACE},{WHITESPACE}{STRING}{WHITESPACE}:{WHITESPACE}{INTEGER}){{0,}})?{WHITESPACE}\}}({WHITESPACE},{WHITESPACE}{STRING}{WHITESPACE}:{WHITESPACE}\{{{WHITESPACE}({STRING}{WHITESPACE}:{WHITESPACE}{INTEGER}({WHITESPACE},{WHITESPACE}{STRING}{WHITESPACE}:{WHITESPACE}{INTEGER}){{0,}})?{WHITESPACE}\}}){{0,}})?{WHITESPACE}\}}{WHITESPACE}\}}""",
-            [
-                (
-                    """{"test_dict": {"foo": {"bar": 123, "apple": 99}, "baz": {"bif": 456}}}""",
-                    True,
-                ),
-                (
-                    """{"test_dict": {"anykey": {"anykey": 123}, "anykey2": {"bif": 456}}}""",
-                    True,
-                ),
-                ("""{"test_dict": {}}""", True),
-                ("""{"test_dict": {"dict of empty dicts are ok": {} }}""", True),
-                (
-                    """{"test_dict": {"anykey": {"ONLY Dict[Dict]": 123}, "No Dict[int]" 1: }}""",
-                    False,
-                ),
-            ],
-        ),
-        # oneOf
-        (
-            {
-                "title": "Foo",
-                "oneOf": [{"type": "string"}, {"type": "number"}, {"type": "boolean"}],
-            },
-            rf'((?:"{STRING_INNER}*")|(?:{NUMBER})|(?:{BOOLEAN}))',
-            [
-                ("12.3", True),
-                ("true", True),
-                ('"a"', True),
-                ("null", False),
-                ("", False),
-                ("12true", False),
-                ('1.3"a"', False),
-                ('12.3true"a"', False),
-            ],
-        ),
-        # anyOf
-        (
-            {
-                "title": "Foo",
-                "anyOf": [{"type": "string"}, {"type": "integer"}],
-            },
-            rf"({STRING}|{INTEGER})",
-            [("12", True), ('"a"', True), ('1"a"', False)],
-        ),
-        # allOf
-        (
-            {
-                "title": "Foo",
-                "allOf": [{"type": "string"}, {"type": "integer"}],
-            },
-            rf"({STRING}{INTEGER})",
-            [('"a"1', True), ('"a"', False), ('"1"', False)],
-        ),
-        # Tuple / prefixItems
-        (
-            {
-                "title": "Foo",
-                "prefixItems": [{"type": "string"}, {"type": "integer"}],
-            },
-            rf"\[{WHITESPACE}{STRING}{WHITESPACE},{WHITESPACE}{INTEGER}{WHITESPACE}\]",
-            [('["a", 1]', True), ('["a", 1, 1]', False), ("[]", False)],
-        ),
-        # Nested schema
-        (
-            {
-                "title": "Bar",
-                "type": "object",
-                "properties": {
-                    "fuzz": {
-                        "title": "Foo",
-                        "type": "object",
-                        "properties": {"spam": {"title": "Spam", "type": "integer"}},
-                        "required": ["spam"],
-                    }
-                },
-                "required": ["fuzz"],
-            },
-            f'\\{{[ ]?"fuzz"[ ]?:[ ]?\\{{[ ]?"spam"[ ]?:[ ]?{INTEGER}[ ]?\\}}[ ]?\\}}',
-            [('{ "fuzz": { "spam": 100 }}', True)],
-        ),
-        # Schema with a reference
-        (
-            {
-                "title": "User",
-                "type": "object",
-                "properties": {
-                    "user_id": {"title": "User Id", "type": "integer"},
-                    "name": {"title": "Name", "type": "string"},
-                    "a": {"$ref": "#/properties/name"},
-                },
-                "required": ["user_id", "name", "a"],
-            },
-            f'\\{{[ ]?"user_id"[ ]?:[ ]?{INTEGER}[ ]?,[ ]?"name"[ ]?:[ ]?{STRING}[ ]?,[ ]?"a"[ ]?:[ ]?{STRING}[ ]?\\}}',
-            [('{"user_id": 100, "name": "John", "a": "Marc"}', True)],
-        ),
-        (
-            {
-                "title": "User",
-                "type": "object",
-                "$defs": {"name": {"title": "Name2", "type": "string"}},
-                "properties": {
-                    "user_id": {"title": "User Id", "type": "integer"},
-                    "name": {"title": "Name", "type": "string"},
-                    "name2": {"$ref": "#/$defs/name"},
-                },
-                "required": ["user_id", "name", "name2"],
-            },
-            f'\\{{[ ]?"user_id"[ ]?:[ ]?{INTEGER}[ ]?,[ ]?"name"[ ]?:[ ]?{STRING}[ ]?,[ ]?"name2"[ ]?:[ ]?{STRING}[ ]?\\}}',
-            [('{"user_id": 100, "name": "John", "name2": "Marc"}', True)],
-        ),
-        (
-            {
-                "$id": "customer",
-                "$schema": "https://json-schema.org/draft/2020-12/schema",
-                "title": "Customer",
-                "type": "object",
-                "properties": {
-                    "name": {"type": "string"},
-                    "last_name": {"type": "string"},
-                    "address": {"$ref": "customer#/$defs/address"},
-                },
-                "required": [
-                    "name",
-                    "first_name",
-                    "last_name",
-                    "address",
-                    "shipping_address",
-                    "billing_address",
-                ],
-                "$defs": {
-                    "address": {
-                        "title": "Address",
-                        "$schema": "http://json-schema.org/draft-07/schema#",
-                        "type": "object",
-                        "properties": {
-                            "city": {"type": "string"},
-                        },
-                        "required": ["street_address", "city", "state"],
-                        "definitions": {
-                            "state": {
-                                "type": "object",
-                                "title": "State",
-                                "properties": {"name": {"type": "string"}},
-                                "required": ["name"],
-                            }
-                        },
-                    }
-                },
-            },
-            f'\\{{[ ]?"name"[ ]?:[ ]?{STRING}[ ]?,[ ]?"last_name"[ ]?:[ ]?{STRING}[ ]?,[ ]?"address"[ ]?:[ ]?\\{{[ ]?"city"[ ]?:[ ]?{STRING}[ ]?\\}}[ ]?\\}}',
-            [
-                (
-                    '{"name": "John", "last_name": "Doe", "address": {"city": "Paris"}}',
-                    True,
-                )
-            ],
-        ),
-        # Optional properties
-        # Last required property in first position
-        (
-            {
-                "properties": {
-                    "name": {"type": "string"},
-                    "age": {"anyOf": [{"type": "integer"}, {"type": "null"}]},
-                    "weapon": {"anyOf": [{"type": "string"}, {"type": "null"}]},
-                },
-                "required": ["name"],
-                "title": "Character",
-                "type": "object",
-            },
-            f'\\{{[ ]?"name"[ ]?:[ ]?{STRING}([ ]?,[ ]?"age"[ ]?:[ ]?({INTEGER}|null))?([ ]?,[ ]?"weapon"[ ]?:[ ]?({STRING}|null))?[ ]?\\}}',
-            [
-                ('{ "name" : "Player" }', True),
-                ('{ "name" : "Player", "weapon" : "sword" }', True),
-                ('{ "age" : 10, "weapon" : "sword" }', False),
-            ],
-        ),
-        # Last required property in middle position
-        (
-            {
-                "properties": {
-                    "name": {"type": "string"},
-                    "age": {"anyOf": [{"type": "integer"}, {"type": "null"}]},
-                    "weapon": {"type": "string"},
-                    "strength": {"anyOf": [{"type": "integer"}, {"type": "null"}]},
-                },
-                "required": ["name", "weapon"],
-                "title": "Character",
-                "type": "object",
-            },
-            f'\\{{[ ]?"name"[ ]?:[ ]?{STRING}[ ]?,([ ]?"age"[ ]?:[ ]?({INTEGER}|null)[ ]?,)?[ ]?"weapon"[ ]?:[ ]?{STRING}([ ]?,[ ]?"strength"[ ]?:[ ]?({INTEGER}|null))?[ ]?\\}}',
-            [
-                ('{ "name" : "Player" , "weapon" : "sword" }', True),
-                (
-                    '{ "name" : "Player", "age" : 10, "weapon" : "sword" , "strength" : 10 }',
-                    True,
-                ),
-                ('{ "weapon" : "sword" }', False),
-            ],
-        ),
-        # Last required property in last position
-        (
-            {
-                "properties": {
-                    "name": {"anyOf": [{"type": "string"}, {"type": "null"}]},
-                    "age": {"type": "integer"},
-                    "armor": {"type": "string"},
-                    "strength": {"anyOf": [{"type": "integer"}, {"type": "null"}]},
-                    "weapon": {"title": "Weapon", "type": "string"},
-                },
-                "required": ["age", "armor", "weapon"],
-                "title": "Character",
-                "type": "object",
-            },
-            f'\\{{([ ]?"name"[ ]?:[ ]?({STRING}|null)[ ]?,)?[ ]?"age"[ ]?:[ ]?{INTEGER}[ ]?,[ ]?"armor"[ ]?:[ ]?{STRING}[ ]?,([ ]?"strength"[ ]?:[ ]?({INTEGER}|null)[ ]?,)?[ ]?"weapon"[ ]?:[ ]?{STRING}[ ]?\\}}',
-            [
-                (
-                    '{ "name" : "Player", "age" : 10, "armor" : "plate", "strength" : 11, "weapon" : "sword" }',
-                    True,
-                ),
-                ('{ "age" : 10, "armor" : "plate", "weapon" : "sword" }', True),
-                (
-                    '{ "name" : "Kahlhanbeh", "armor" : "plate", "weapon" : "sword" }',
-                    False,
-                ),
-            ],
-        ),
-        # All properties are optional
-        (
-            {
-                "properties": {
-                    "name": {"anyOf": [{"type": "string"}, {"type": "null"}]},
-                    "age": {"anyOf": [{"type": "integer"}, {"type": "null"}]},
-                    "strength": {"anyOf": [{"type": "integer"}, {"type": "null"}]},
-                },
-                "title": "Character",
-                "type": "object",
-            },
-            f'\\{{([ ]?"name"[ ]?:[ ]?({STRING}|null)([ ]?,[ ]?"age"[ ]?:[ ]?({INTEGER}|null))?([ ]?,[ ]?"strength"[ ]?:[ ]?({INTEGER}|null))?|([ ]?"name"[ ]?:[ ]?({STRING}|null)[ ]?,)?[ ]?"age"[ ]?:[ ]?({INTEGER}|null)([ ]?,[ ]?"strength"[ ]?:[ ]?({INTEGER}|null))?|([ ]?"name"[ ]?:[ ]?({STRING}|null)[ ]?,)?([ ]?"age"[ ]?:[ ]?({INTEGER}|null)[ ]?,)?[ ]?"strength"[ ]?:[ ]?({INTEGER}|null))?[ ]?\\}}',
-            [
-                ('{ "name" : "Player" }', True),
-                ('{ "name" : "Player", "age" : 10, "strength" : 10 }', True),
-                ('{ "age" : 10, "strength" : 10 }', True),
-                ("{ }", True),
-            ],
-        ),
-    ],
-)
-def test_match(schema, regex, examples):
-    interegular.parse_pattern(regex)
-    schema = json.dumps(schema)
-    test_regex = build_regex_from_schema(schema)
-    assert test_regex == regex
-
-    for string, does_match in examples:
-        match = re.fullmatch(test_regex, string)
-        if does_match:
-            if match is None:
-                raise ValueError(f"Expected match for '{string}'")
-            assert match[0] == string
-            assert match.span() == (0, len(string))
-        else:
-            assert match is None
-
-
-@pytest.mark.parametrize(
-    "schema,regex,examples",
-    [
-        # UUID
-        (
-            {"title": "Foo", "type": "string", "format": "uuid"},
-            UUID,
-            [
-                ("123e4567-e89b-12d3-a456-426614174000", False),
-                ('"123e4567-e89b-12d3-a456-426614174000"', True),
-                ('"123e4567-e89b-12d3-a456-42661417400"', False),
-                ('"123e4567-e89b-12d3-a456-42661417400g"', False),
-                ('"123e4567-e89b-12d3-a456-42661417400-"', False),
-                ('""', False),
-            ],
-        ),
-        # DATE-TIME
-        (
-            {"title": "Foo", "type": "string", "format": "date-time"},
-            DATE_TIME,
-            [
-                ("2018-11-13T20:20:39Z", False),
-                ('"2018-11-13T20:20:39Z"', True),
-                ('"2016-09-18T17:34:02.666Z"', True),
-                ('"2008-05-11T15:30:00Z"', True),
-                ('"2021-01-01T00:00:00"', True),
-                ('"2022-01-10 07:19:30"', False),  # missing T
-                ('"2022-12-10T10-04-29"', False),  # incorrect separator
-                ('"2023-01-01"', False),
-            ],
-        ),
-        # DATE
-        (
-            {"title": "Foo", "type": "string", "format": "date"},
-            DATE,
-            [
-                ("2018-11-13", False),
-                ('"2018-11-13"', True),
-                ('"2016-09-18"', True),
-                ('"2008-05-11"', True),
-                ('"2015-13-01"', False),  # incorrect month
-                ('"2022-01"', False),  # missing day
-                ('"2022/12/01"', False),  # incorrect separator"
-            ],
-        ),
-        # TIME
-        (
-            {"title": "Foo", "type": "string", "format": "time"},
-            TIME,
-            [
-                ("20:20:39Z", False),
-                ('"20:20:39Z"', True),
-                ('"15:30:00Z"', True),
-                ('"25:30:00"', False),  # incorrect hour
-                ('"15:30"', False),  # missing seconds
-                ('"15:30:00.000"', False),  # missing Z
-                ('"15-30-00"', False),  # incorrect separator
-                ('"15:30:00+01:00"', False),  # incorrect separator
-            ],
-        ),
-    ],
-)
-def test_format(schema, regex, examples):
-    interegular.parse_pattern(regex)
-    schema = json.dumps(schema)
-    test_regex = build_regex_from_schema(schema)
-    assert test_regex == regex
-
-    for string, does_match in examples:
-        match = re.fullmatch(test_regex, string)
-        if does_match:
-            assert match[0] == string
-            assert match.span() == (0, len(string))
-        else:
-            assert match is None
-
-
-@pytest.mark.parametrize(
-    "schema,examples",
-    [
-        # NESTED UUID
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {"uuid": {"type": "string", "format": "uuid"}},
-            },
-            [
-                ('{"uuid": "123e4567-e89b-12d3-a456-426614174000"}', True),
-                ('{"uuid":"123e4567-e89b-12d3-a456-42661417400"}', False),
-                ('{"uuid":"123e4567-e89b-12d3-a456-42661417400g"}', False),
-                ('{"uuid":"123e4567-e89b-12d3-a456-42661417400-"}', False),
-                (
-                    '{"uuid":123e4567-e89b-12d3-a456-426614174000}',
-                    False,
-                ),  # missing quotes for value
-                ('{"uuid":""}', False),
-            ],
-        ),
-        # NESTED DATE-TIME
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {"dateTime": {"type": "string", "format": "date-time"}},
-            },
-            [
-                ('{"dateTime": "2018-11-13T20:20:39Z"}', True),
-                ('{"dateTime":"2016-09-18T17:34:02.666Z"}', True),
-                ('{"dateTime":"2008-05-11T15:30:00Z"}', True),
-                ('{"dateTime":"2021-01-01T00:00:00"}', True),
-                ('{"dateTime":"2022-01-10 07:19:30"}', False),  # missing T
-                ('{"dateTime":"2022-12-10T10-04-29"}', False),  # incorrect separator
-                (
-                    '{"dateTime":2018-11-13T20:20:39Z}',
-                    False,
-                ),  # missing quotes for value
-                ('{"dateTime":"2023-01-01"}', False),
-            ],
-        ),
-        # NESTED DATE
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {"date": {"type": "string", "format": "date"}},
-            },
-            [
-                ('{"date": "2018-11-13"}', True),
-                ('{"date":"2016-09-18"}', True),
-                ('{"date":"2008-05-11"}', True),
-                ('{"date":"2015-13-01"}', False),  # incorrect month
-                ('{"date":"2022-01"}', False),  # missing day
-                ('{"date":"2022/12/01"}', False),  # incorrect separator"
-                ('{"date":2018-11-13}', False),  # missing quotes for value
-            ],
-        ),
-        # NESTED TIME
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-                "properties": {"time": {"type": "string", "format": "time"}},
-            },
-            [
-                ('{"time": "20:20:39Z"}', True),
-                ('{"time":"15:30:00Z"}', True),
-                ('{"time":"25:30:00"}', False),  # incorrect hour
-                ('{"time":"15:30"}', False),  # missing seconds
-                ('{"time":"15:30:00.000"}', False),  # missing Z
-                ('{"time":"15-30-00"}', False),  # incorrect separator
-                ('{"time":"15:30:00+01:00"}', False),  # incorrect separator
-                ('{"time":20:20:39Z}', False),  # missing quotes for value
-            ],
-        ),
-        # Unconstrained Object
-        (
-            {
-                "title": "Foo",
-                "type": "object",
-            },
-            [
-                ("{}", True),
-                ('{"a": 1, "b": null}', True),
-                ('{"a": {"z": {"g": 4}}, "b": null}', True),
-                ("1234", False),  # not an object
-                ('["a", "a"]', False),  # not an array
-            ],
-        ),
-        # Unconstrained Array
-        (
-            {
-                "type": "array",
-            },
-            [
-                ("[1, {}, false]", True),
-                ("[{}]", True),
-                ('[{"a": {"z": "q"}, "b": null}]', True),
-                ('[{"a": [1, 2, true], "b": null}]', True),
-                ('[{"a": [1, 2, true], "b": {"a": "b"}}, 1, true, [1, [2]]]', True),
-                # too deep, default unconstrained depth limit = 2
-                (
-                    '[{"a": [1, 2, true], "b": {"a": "b"}}, 1, true, [1, [2, [3]]]]',
-                    False,
-                ),
-                ('[{"a": {"z": {"g": 4}}, "b": null}]', False),
-                ("[[[[1]]]]", False),
-                # not an array
-                ("{}", False),
-                ('{"a": 1, "b": null}', False),
-                ('{"a": {"z": {"g": 4}}, "b": null}', False),
-                ("1234", False),  # not an array
-                ('{"a": "a"}', False),  # not an array
-            ],
-        ),
-        # No schema / unconstrained value
-        (
-            {},
-            [
-                ('"aaabbuecuh"', True),  # string
-                ("5.554", True),  # number
-                ("true", True),  # boolean
-                ("null", True),  # null
-                ("5999", True),  # integer
-                ('["a", "b"]', True),  # array
-                ('{"key": {"k2": "value"}}', True),  # nested object
-                ("this isnt valid json", False),
-            ],
-        ),
-    ],
-)
-def test_format_without_regex(schema, examples):
-    schema = json.dumps(schema)
-    test_regex = build_regex_from_schema(schema)
-    for string, does_match in examples:
-        match = re.fullmatch(test_regex, string)
-        if does_match:
-            assert match[0] == string
-            assert match.span() == (0, len(string))
-        else:
-            assert match is None
-
-
-@pytest.mark.parametrize("whitespace_pattern", [None, r"[\n ]*", "abc"])
-def test_json_schema_custom_whitespace_pattern(whitespace_pattern):
-    """assert whitespace_pattern setting respected"""
-
-    class MockModel(BaseModel):
-        foo: int
-        bar: str
-
-    schema = json.dumps(MockModel.model_json_schema())
-
-    # assert any ws pattern can be used
-    if whitespace_pattern == "abc":
-        build_regex_from_schema(schema, whitespace_pattern)
-        return
-
-    pattern = build_regex_from_schema(schema, whitespace_pattern)
-
-    mock_result_mult_ws = (
-        """{     "foo"   :   4, \n\n\n   "bar": "baz    baz baz bar"\n\n}"""
-    )
-    mock_result_maybe_ws = """{"foo" : 4 ,"bar":"baz    baz baz bar"}"""
-
-    match_default_ws = re.fullmatch(pattern, mock_result_maybe_ws)
-    if whitespace_pattern is None:
-        assert match_default_ws
-    else:
-        assert re.fullmatch(pattern, mock_result_mult_ws)
-
-
-def test_one_of_doesnt_produce_illegal_lookaround():
-    """Reproduces failure in https://github.com/dottxt-ai/outlines/issues/823"""
-
-    class Cat(BaseModel):
-        pet_type: Literal["cat"]
-        meows: int
-
-    class Dog(BaseModel):
-        pet_type: Literal["dog"]
-        barks: float
-
-    class Model(BaseModel):
-        pet: Union[Cat, Dog] = Field(..., discriminator="pet_type")
-        n: int
-
-    json_schema = Model.schema_json()
-
-    json_schema = Model.schema_json()
-    pattern = build_regex_from_schema(json_schema, whitespace_pattern=None)
-
-    # check if the pattern uses lookarounds incompatible with interegular.Pattern.to_fsm()
-    interegular.parse_pattern(pattern).to_fsm()
+    regex_str = build_regex_from_schema(schema)
+    assert isinstance(regex_str, str)
 
 
 def add(a: float, b: float) -> float:
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index 92c5d789c..7b7973d23 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -363,6 +363,7 @@ def mul(c: float, d: float) -> float:
     return c * d
 
 
+@pytest.mark.xfail(reason="Enum of objects are not supported in outlines-core")
 def test_transformers_json_function_enum(model):
     prompt = "Output some JSON "
 

From 5113f9415f999f8ba64a4332b25078d90c89a1d3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 9 Dec 2024 13:20:31 +0100
Subject: [PATCH 123/505] Convert `enum` of objects to `oneOf` with consts

The `enum` keyword is reserved for arrays of constants. When we have
callables as members of a Python `Enum`, we get an object whose
properties are the arguments of the callable. The `Enum` thus needs
to be converted into a `oneOf` field.
---
 outlines/fsm/json_schema.py                     |  6 ++++--
 tests/fsm/test_json_schema.py                   | 10 ++++++----
 tests/generate/test_integration_transformers.py |  1 -
 3 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/outlines/fsm/json_schema.py b/outlines/fsm/json_schema.py
index bae0ad17a..578ee7626 100644
--- a/outlines/fsm/json_schema.py
+++ b/outlines/fsm/json_schema.py
@@ -74,8 +74,10 @@ def get_schema_from_enum(myenum: type[Enum]) -> dict:
             f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
         )
     choices = [
-        get_schema_from_signature(elt.value.func) if callable(elt.value) else elt.value
+        get_schema_from_signature(elt.value.func)
+        if callable(elt.value)
+        else {"const": elt.value}
         for elt in myenum
     ]
-    schema = {"title": myenum.__name__, "enum": choices}
+    schema = {"title": myenum.__name__, "oneOf": choices}
     return schema
diff --git a/tests/fsm/test_json_schema.py b/tests/fsm/test_json_schema.py
index 825433397..23864e029 100644
--- a/tests/fsm/test_json_schema.py
+++ b/tests/fsm/test_json_schema.py
@@ -69,8 +69,10 @@ class EmptyEnum(Enum):
 )
 def test_enum_schema(enum, expectation):
     with expectation:
-        result = get_schema_from_enum(enum)
-        assert result["title"] == enum.__name__
-        assert len(result["enum"]) == len(enum)
-        for elt in result["enum"]:
+        schema = get_schema_from_enum(enum)
+        regex_str = build_regex_from_schema(json.dumps(schema))
+        assert isinstance(regex_str, str)
+        assert schema["title"] == enum.__name__
+        assert len(schema["oneOf"]) == len(enum)
+        for elt in schema["oneOf"]:
             assert type(elt) in [int, float, bool, type(None), str, dict]
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
index 7b7973d23..92c5d789c 100644
--- a/tests/generate/test_integration_transformers.py
+++ b/tests/generate/test_integration_transformers.py
@@ -363,7 +363,6 @@ def mul(c: float, d: float) -> float:
     return c * d
 
 
-@pytest.mark.xfail(reason="Enum of objects are not supported in outlines-core")
 def test_transformers_json_function_enum(model):
     prompt = "Output some JSON "
 

From 147b03e8d4d7279177fdaac82be4646d10d8c8ff Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 9 Dec 2024 13:43:33 +0100
Subject: [PATCH 124/505] Fail as soon as one test fails

---
 .github/workflows/tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 36e6f8526..f54a888fa 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -44,7 +44,7 @@ jobs:
         echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
-        pytest --cov=outlines
+        pytest -x --cov=outlines
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
     - name: Upload coverage data

From fb43f4fb7f711cc765e78671e991e1eecd51d38b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 9 Dec 2024 14:55:12 +0100
Subject: [PATCH 125/505] Fix conversion from MLX to torch tensor

---
 outlines/processors/base_logits_processor.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index eec7de121..44b55af2e 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -107,9 +107,12 @@ def _to_torch(tensor_like: Array) -> torch.Tensor:
             return torch.tensor(tensor_like)
 
         elif is_mlx_array_type(type(tensor_like)):
-            # mlx -> torch -> mlx conversion docs:
-            # https://ml-explore.github.io/mlx/build/html/usage/numpy.html
-            return torch.from_dlpack(tensor_like)
+            import mlx.core as mx
+
+            # https://ml-explore.github.io/mlx/build/html/usage/numpy.html#pytorch
+            return torch.from_dlpack(
+                np.array(tensor_like.astype(mx.float32), copy=False)
+            )
 
         elif is_jax_array_type(type(tensor_like)):
             import jax

From 9df3b1b2d6c9d4b78551b759758ce5602995ce97 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 9 Dec 2024 22:05:26 +0100
Subject: [PATCH 126/505] Bump outlines-core to 0.1.24

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 2001f4c40..896e8aadd 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,7 @@ dependencies = [
    "pycountry",
    "airportsdata",
    "torch",
-   "outlines_core==0.1.20",
+   "outlines_core==0.1.24",
 ]
 dynamic = ["version"]
 

From 05bcb4b0370e6a7b2136092ac1d1b599a5660cb9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Wed, 11 Dec 2024 14:44:34 +0100
Subject: [PATCH 127/505] Bump `outlines-core` version to 0.1.25

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 896e8aadd..b83275f89 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,7 @@ dependencies = [
    "pycountry",
    "airportsdata",
    "torch",
-   "outlines_core==0.1.24",
+   "outlines_core==0.1.25",
 ]
 dynamic = ["version"]
 

From 5c41fb3599cbd8771d5aae072bda837717d0742d Mon Sep 17 00:00:00 2001
From: Sidharth Rajaram <sidharthrajaram@gmail.com>
Date: Thu, 12 Dec 2024 16:24:26 -0800
Subject: [PATCH 128/505] Bump outlines-core version to 0.1.26

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index b83275f89..1a6d08e54 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,7 @@ dependencies = [
    "pycountry",
    "airportsdata",
    "torch",
-   "outlines_core==0.1.25",
+   "outlines_core==0.1.26",
 ]
 dynamic = ["version"]
 

From 365b566e9f26d7db2aef399c7e987f3b9358b6d5 Mon Sep 17 00:00:00 2001
From: Avishkar Gupta <rootavish@gmail.com>
Date: Mon, 16 Dec 2024 15:14:36 +0530
Subject: [PATCH 129/505] Fix image link in receipt-digitization example

The current link points to a HTML instead of a PNG, so after it's downloaded via requests.get, PIL isn't able to recognize the format and rightly so.
---
 docs/cookbook/receipt-digitization.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/cookbook/receipt-digitization.md b/docs/cookbook/receipt-digitization.md
index 67830fa81..0fa256db7 100644
--- a/docs/cookbook/receipt-digitization.md
+++ b/docs/cookbook/receipt-digitization.md
@@ -117,7 +117,7 @@ Here's what the image looks like:
 
 ```python
 # Path to the image
-image_path = "https://dottxt-ai.github.io/outlines/main/cookbook/images/trader-joes-receipt.png"
+image_path = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/cookbook/images/trader-joes-receipt.jpg"
 
 # Download the image
 response = requests.get(image_path)

From 9b586dbda46bce06363bfb26995e35255b58f55e Mon Sep 17 00:00:00 2001
From: Aakash Thatte <84656834+sky-2002@users.noreply.github.com>
Date: Mon, 16 Dec 2024 17:37:50 +0530
Subject: [PATCH 130/505] Move generation of sampling params (#1340)

This PR addresses #1335
I have kept the `SamplingParameters` in `outlines/generate/api.py` cause
its imported at many places and may break examples ig.
Would you want to change that @dgerlanc ??
Otherwise if this looks good, I can make docs changes if needd, lmk.
---
 outlines/generate/api.py | 28 +++++++---------------------
 outlines/samplers.py     | 30 ++++++++++++++++++++++++++++++
 tests/test_samplers.py   | 23 +++++++++++++++++++++++
 3 files changed, 60 insertions(+), 21 deletions(-)

diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index 4919f2090..5d3c52a8a 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -4,7 +4,6 @@
 from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
 
 from outlines.generate.generator import sequence_generator
-from outlines.samplers import BeamSearchSampler, GreedySampler, MultinomialSampler
 
 if TYPE_CHECKING:
     import torch
@@ -353,11 +352,13 @@ def token_generator() -> Iterator[Union[List[str], str, List[List[str]]]]:
                     ]
 
                     generated_sequences = [
-                        self.format_sequence(
-                            self.strip_stop_sequences(sequence, stop_sequences)
+                        (
+                            self.format_sequence(
+                                self.strip_stop_sequences(sequence, stop_sequences)
+                            )
+                            if stop
+                            else sequence
                         )
-                        if stop
-                        else sequence
                         for sequence, stop in zip(
                             generated_sequences, is_stop_at_reached
                         )
@@ -428,22 +429,7 @@ def __init__(self, model, logits_processor, sampler):
         self.model = model
         self.logits_processor = logits_processor
 
-        if isinstance(sampler, MultinomialSampler):
-            self.sampling_params = SamplingParameters(
-                "multinomial",
-                sampler.samples,
-                sampler.top_p,
-                sampler.top_k,
-                sampler.temperature,
-            )
-        elif isinstance(sampler, GreedySampler):
-            self.sampling_params = SamplingParameters(
-                "greedy", sampler.samples, None, None, 0.0
-            )
-        elif isinstance(sampler, BeamSearchSampler):
-            self.sampling_params = SamplingParameters(
-                "beam_search", sampler.samples, None, None, 1.0
-            )
+        self.sampling_params = sampler.sampling_params
 
     def prepare_generation_parameters(
         self,
diff --git a/outlines/samplers.py b/outlines/samplers.py
index b1421971f..3ab1728fc 100644
--- a/outlines/samplers.py
+++ b/outlines/samplers.py
@@ -1,4 +1,5 @@
 import math
+from dataclasses import dataclass
 from typing import TYPE_CHECKING, Callable, Optional, Protocol, Tuple
 
 if TYPE_CHECKING:
@@ -17,6 +18,17 @@ def __call__(
         ...
 
 
+@dataclass(frozen=True)
+class SamplingParameters:
+    """Sampling parameters available in Outlines."""
+
+    sampler: str
+    num_samples: int = 1
+    top_p: Optional[float] = None
+    top_k: Optional[int] = None
+    temperature: Optional[float] = None
+
+
 class GreedySampler:
     """Greedy Sampling algorithm.
 
@@ -76,6 +88,10 @@ def __call__(
 
         return next_token_ids, ancestors, weights
 
+    @property
+    def sampling_params(self):
+        return SamplingParameters("greedy", self.samples, None, None, 0.0)
+
 
 greedy = GreedySampler
 
@@ -161,6 +177,16 @@ def __call__(
 
         return next_token_ids, ancestors, weights
 
+    @property
+    def sampling_params(self):
+        return SamplingParameters(
+            "multinomial",
+            self.samples,
+            self.top_p,
+            self.top_k,
+            self.temperature,
+        )
+
 
 multinomial = MultinomialSampler
 
@@ -320,5 +346,9 @@ def __call__(
 
         return next_token_ids, ancestors, weights
 
+    @property
+    def sampling_params(self):
+        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
+
 
 beam_search = BeamSearchSampler
diff --git a/tests/test_samplers.py b/tests/test_samplers.py
index 88cdb0fbc..10a7be26f 100644
--- a/tests/test_samplers.py
+++ b/tests/test_samplers.py
@@ -47,6 +47,13 @@ def test_greedy():
     assert ancestors.equal(torch.tensor([0, 1]))
     assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
 
+    params = sampler.sampling_params
+    assert params.sampler == "greedy"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 0.0
+
 
 def test_multinomial():
     rng = torch.Generator()
@@ -72,6 +79,14 @@ def test_multinomial():
     assert ancestors.equal(torch.tensor([0, 1]))
     assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
 
+    sampler = MultinomialSampler(samples=5, top_k=10, top_p=0.9, temperature=0.8)
+    params = sampler.sampling_params
+    assert params.sampler == "multinomial"
+    assert params.num_samples == 5
+    assert params.top_p == 0.9
+    assert params.top_k == 10
+    assert params.temperature == 0.8
+
 
 def test_multinomial_init():
     sampler = MultinomialSampler()
@@ -252,3 +267,11 @@ def test_beam_search():
             ]
         )
     )
+
+    sampler = BeamSearchSampler(beams=3)
+    params = sampler.sampling_params
+    assert params.sampler == "beam_search"
+    assert params.num_samples == 3
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 1.0

From 061f2ba1966222e9b59a19b6fe4b8b9cefd61686 Mon Sep 17 00:00:00 2001
From: Tim Zaman <timbobel@gmail.com>
Date: Thu, 19 Dec 2024 15:46:25 -0800
Subject: [PATCH 131/505] Fix unresolved imports in transformers doc

---
 docs/reference/models/transformers.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index f4c319540..0b0add226 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -133,7 +133,7 @@ T5 Example:
 import outlines
 from transformers import AutoModelForSeq2SeqLM
 
-model_pile_t5 = models.transformers(
+model_pile_t5 = outlines.models.transformers(
     model_name="EleutherAI/pile-t5-large",
     model_class=AutoModelForSeq2SeqLM,
 )
@@ -141,7 +141,7 @@ model_pile_t5 = models.transformers(
 
 Bart Example:
 ```python
-model_bart = models.transformers(
+model_bart = outlines.models.transformers(
     model_name="facebook/bart-large",
     model_class=AutoModelForSeq2SeqLM,
 )

From 9744f00aa7a803c6637175e2f640f01f144aa8c7 Mon Sep 17 00:00:00 2001
From: Edoardo Abati <29585319+EdAbati@users.noreply.github.com>
Date: Sat, 21 Dec 2024 14:19:12 +0100
Subject: [PATCH 132/505] Set docs `docstring_style` (#1343)

Hi, Thank you for this great library!

It seems that the docstrings are not rendered correctly in the docs. I
think we should explicitly set the `docstring_style` because [it
defaults to
`"google"`](https://mkdocstrings.github.io/python/usage/configuration/docstrings/#docstring_style)
but outlines is using numpy.

Before:
![Screenshot 2024-12-16 at 23 00
26](https://github.com/user-attachments/assets/c752ee3d-519e-4098-b943-3aab43c8af25)
After:
![Screenshot 2024-12-16 at 23 00
41](https://github.com/user-attachments/assets/5b5f524b-6921-4dbe-994d-72c079e677bc)


There seem to be other issues in the docstrings:
- for example
[`Properties`](https://github.com/dottxt-ai/outlines/blob/main/outlines/models/openai.py#L23)
should be
[`Attributes`](https://numpydoc.readthedocs.io/en/latest/format.html#parameters)
- only openai and transformers models are present in the [api
reference](https://github.com/dottxt-ai/outlines/blob/main/docs/api/models.md)

I'm happy to make followup PRs for those.

Please let me know if I missed something, I couldn't find related
issues/PRs.
---
 mkdocs.yml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/mkdocs.yml b/mkdocs.yml
index 4888309b9..cca3e3af1 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -87,6 +87,7 @@ plugins:
       handlers:
         python:
           options:
+            docstring_style: numpy
             show_submodules: true
   - search
   - section-index

From 6205c0b367a8c226f219a2b400c9695d63a6122d Mon Sep 17 00:00:00 2001
From: Marco De Nadai <me@marcodena.it>
Date: Sat, 28 Dec 2024 11:54:14 +0100
Subject: [PATCH 133/505] Fix JSON structured import

---
 examples/vllm_integration.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/vllm_integration.py b/examples/vllm_integration.py
index 5f833d19f..c9ce057bc 100644
--- a/examples/vllm_integration.py
+++ b/examples/vllm_integration.py
@@ -3,7 +3,7 @@
 import vllm
 from pydantic import BaseModel
 
-from outlines.integrations.vllm import JSONLogitsProcessor
+from outlines.processors.structured import JSONLogitsProcessor
 
 
 class Person(BaseModel):

From fef70c0febf26521e81f1b072900f7b529a2a49f Mon Sep 17 00:00:00 2001
From: Martin <martin.ku@skysource.com.tw>
Date: Tue, 31 Dec 2024 02:05:11 +0800
Subject: [PATCH 134/505] Update mkdocs.yml: jump to mamba section

---
 mkdocs.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mkdocs.yml b/mkdocs.yml
index cca3e3af1..7a3ca1e98 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -161,7 +161,7 @@ nav:
           - TGI: reference/models/tgi.md
           - ExllamaV2: reference/models/exllamav2.md
           - MLX: reference/models/mlxlm.md
-          - Mamba: reference/models/transformers.md
+          - Mamba: reference/models/transformers/#mamba
         - API:
             - OpenAI: reference/models/openai.md
   - API Reference:

From 59208706d3d502e2a50b82fc13d4975b68ad457f Mon Sep 17 00:00:00 2001
From: Martin <martin.ku@skysource.com.tw>
Date: Tue, 31 Dec 2024 02:19:13 +0800
Subject: [PATCH 135/505] Update transformers.md: fix headings level for mkdocs
 toc

---
 docs/reference/models/transformers.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 0b0add226..3649b9b13 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -30,7 +30,7 @@ tokenizer = AutoTokenizer.from_pretrained("gpt2")
 model = models.Transformers(llm, tokenizer)
 ```
 
-# Using Logits Processors
+## Using Logits Processors
 
 There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
 
@@ -39,7 +39,7 @@ There are two ways to use Outlines Structured Generation with HuggingFace Transf
 
 Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
 
-## Using `outlines.models.transformers`
+### Using `outlines.models.transformers`
 
 ```python
 import outlines
@@ -54,7 +54,7 @@ print(output)
 # 2:30 pm
 ```
 
-## Using models initialized via the `transformers`  library
+### Using models initialized via the `transformers`  library
 
 ```python
 import outlines
@@ -85,7 +85,7 @@ print(output)
 [transformers]: https://github.com/huggingface/transformers
 
 
-# Alternative Model Classes
+## Alternative Model Classes
 
 `outlines.models.transformers` defaults to `transformers.AutoModelForCausalLM`, which is the appropriate class for most standard large language models, including Llama 3, Mistral, Phi-3, etc.
 

From fddfc8f92da4021e58a57af545934b92d79157c3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christoph=20H=C3=B6sler?= <christoph.hoesler@inovex.de>
Date: Thu, 28 Nov 2024 13:16:25 +0100
Subject: [PATCH 136/505] remove non_blocking=True

---
 outlines/processors/structured.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 64892b73f..ede9f5f8d 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -109,7 +109,7 @@ def process_logits(
         batch_indices = []
         for i, guide_state in enumerate(sequence_states):
             allowed_tokens = self.guide.get_next_instruction(guide_state).tokens.to(
-                mask.device, non_blocking=True
+                mask.device
             )
             allowed_tokens_batch.append(allowed_tokens)
             batch_indices.append(

From 2f0740e3ef1df21be19bf01031ac86bd42fc326e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christoph=20H=C3=B6sler?= <christoph.hoesler@inovex.de>
Date: Thu, 5 Dec 2024 15:24:00 +0100
Subject: [PATCH 137/505] optimize tensor creation

---
 outlines/processors/structured.py | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index ede9f5f8d..583fcc98f 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -103,22 +103,19 @@ def process_logits(
 
             sequence_states.append(self._guide_states[curr_state_key])
 
-        mask = torch.ones_like(logits, dtype=torch.bool)
-
         allowed_tokens_batch = []
         batch_indices = []
         for i, guide_state in enumerate(sequence_states):
-            allowed_tokens = self.guide.get_next_instruction(guide_state).tokens.to(
-                mask.device
-            )
+            allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
             allowed_tokens_batch.append(allowed_tokens)
             batch_indices.append(
                 torch.full_like(allowed_tokens, i)
             )  # Store batch index for each allowed token
 
-        allowed_tokens_concat = torch.cat(allowed_tokens_batch)
-        batch_indices_concat = torch.cat(batch_indices)
+        allowed_tokens_concat = torch.cat(allowed_tokens_batch).to(logits.device)
+        batch_indices_concat = torch.cat(batch_indices).to(logits.device)
 
+        mask = torch.ones_like(logits, dtype=torch.bool)
         mask[batch_indices_concat, allowed_tokens_concat] = False
         logits.masked_fill_(mask, float("-inf"))
 

From 6a8612bd10525e95a62d9f9c727d3e54d0723f1a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christoph=20H=C3=B6sler?= <christoph.hoesler@inovex.de>
Date: Thu, 5 Dec 2024 21:18:32 +0100
Subject: [PATCH 138/505] add mps to processor benchmark

---
 benchmarks/bench_processors.py | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/benchmarks/bench_processors.py b/benchmarks/bench_processors.py
index db1e4a8f1..02ea52b79 100644
--- a/benchmarks/bench_processors.py
+++ b/benchmarks/bench_processors.py
@@ -37,15 +37,12 @@ def get_mock_processor_inputs(array_library, num_tokens=30000):
     logits: (4, 30,000 ) dtype=float
     input_ids shape: (4, 2048) dtype=int
     """
-    if array_library == "torch":
-        logits = torch.rand((4, num_tokens), dtype=torch.float)
-        input_ids = torch.randint(
-            low=0, high=num_tokens, size=(4, 2048), dtype=torch.int
-        )
-    elif array_library == "torch_cuda":
-        logits = torch.rand((4, num_tokens), dtype=torch.float, device="cuda")
+    if array_library.startswith("torch"):
+        device = array_library.split("_")[1] if "_" in array_library else "cpu"
+
+        logits = torch.rand((4, num_tokens), dtype=torch.float, device=device)
         input_ids = torch.randint(
-            low=0, high=num_tokens, size=(4, 2048), dtype=torch.int, device="cuda"
+            low=0, high=num_tokens, size=(4, 2048), dtype=torch.int, device=device
         )
     elif array_library == "numpy":
         logits = np.random.rand(4, num_tokens).astype(np.float32)
@@ -88,6 +85,8 @@ class LogitsProcessorPassthroughBenchmark:
         params += ["mlx"]
     if torch.cuda.is_available():
         params += ["torch_cuda"]
+    if torch.mps.is_available():
+        params += ["torch_mps"]
     if is_jax_allowed():
         params += ["jax"]
 
@@ -108,9 +107,10 @@ class LogitsProcessorStructuredBenchmark:
     array_libraries = ["torch", "numpy"]
     if is_mlx_lm_allowed():
         array_libraries += ["mlx"]
-    # PR TODO
     if torch.cuda.is_available():
         array_libraries += ["torch_cuda"]
+    if torch.mps.is_available():
+        array_libraries += ["torch_mps"]
 
     # accept very many or very few tokens, respectively
     patterns = [r"[^Z]*", "Z*"]

From d32dfde3ae7eac71f57af96d5141975ef9322b12 Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Thu, 2 Jan 2025 18:52:10 +0100
Subject: [PATCH 139/505] Skip `vllm`-related tests on non-linux platforms

Fix #1356
---
 pyproject.toml                          |  2 +-
 tests/conftest.py                       | 14 ++++++++++++++
 tests/generate/test_integration_vllm.py |  6 +++++-
 3 files changed, 20 insertions(+), 2 deletions(-)
 create mode 100644 tests/conftest.py

diff --git a/pyproject.toml b/pyproject.toml
index 1a6d08e54..28060f797 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -67,7 +67,7 @@ test = [
     "huggingface_hub",
     "openai>=1.0.0",
     "datasets",
-    "vllm; sys_platform != 'darwin'",
+    "vllm; sys_platform == 'linux'",
     "transformers",
     "pillow",
     "exllamav2",
diff --git a/tests/conftest.py b/tests/conftest.py
new file mode 100644
index 000000000..bbc56afdb
--- /dev/null
+++ b/tests/conftest.py
@@ -0,0 +1,14 @@
+import sys
+
+import pytest
+
+
+def pytest_collection_modifyitems(config, items):
+    if sys.platform != "linux":
+        skip_vllm = pytest.mark.skip(reason="vLLM models can only be run on Linux.")
+        for item in items:
+            if "test_integration_vllm" in item.nodeid:
+                item.add_marker(skip_vllm)
+                print(
+                    f"WARNING: {item.nodeid} is skipped because vLLM only supports Linux platform (including WSL)."
+                )
diff --git a/tests/generate/test_integration_vllm.py b/tests/generate/test_integration_vllm.py
index 4634bc839..d812d3cf2 100644
--- a/tests/generate/test_integration_vllm.py
+++ b/tests/generate/test_integration_vllm.py
@@ -4,7 +4,11 @@
 import pytest
 import torch
 from pydantic import BaseModel, constr
-from vllm.sampling_params import SamplingParams
+
+try:
+    from vllm.sampling_params import SamplingParams
+except ImportError:
+    pass
 
 import outlines.generate as generate
 import outlines.grammars as grammars

From 3cc399d3d6d2a1b0b2242d9434f4d6ab15ccc30e Mon Sep 17 00:00:00 2001
From: Tyler Thomas <36181311+tylerjthomas9@users.noreply.github.com>
Date: Fri, 3 Jan 2025 22:43:54 -0700
Subject: [PATCH 140/505] Fix `json_schema` imports in cot example

---
 docs/cookbook/chain_of_thought.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
index b814ae048..5ac18af25 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -80,8 +80,8 @@ json_schema = Reasoning.model_json_schema()
 We could generate a response using the json schema but for a change we will use the regex:
 
 ```python
-from outlines.integrations.utils import convert_json_schema_to_str
-from outlines.fsm.json_schema import build_regex_from_schema
+from outlines.fsm.json_schema import convert_json_schema_to_str
+from outlines_core.fsm.json_schema import build_regex_from_schema
 
 schema_str = convert_json_schema_to_str(json_schema=json_schema)
 regex_str = build_regex_from_schema(schema_str)

From 4456f3ccba277aae18f013b429a6cc74e6655b6c Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Thu, 9 Jan 2025 17:11:11 +0100
Subject: [PATCH 141/505] Add `from_file` class method to the `Prompt` object

---
 outlines/__init__.py  |   4 +-
 outlines/prompts.py   | 143 ++++++++++++++++++++++++++++--------------
 tests/test_prompts.py |  96 ++++++++++++++++++++++++++--
 3 files changed, 189 insertions(+), 54 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 307d2ba6f..eeba78dc7 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -1,4 +1,5 @@
 """Outlines is a Generative Model Programming Framework."""
+
 import outlines.generate
 import outlines.grammars
 import outlines.models
@@ -7,7 +8,7 @@
 from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function
-from outlines.prompts import prompt
+from outlines.prompts import Prompt, prompt
 
 __all__ = [
     "clear_cache",
@@ -15,6 +16,7 @@
     "get_cache",
     "Function",
     "prompt",
+    "Prompt",
     "vectorize",
     "grammars",
 ]
diff --git a/outlines/prompts.py b/outlines/prompts.py
index a7824451a..2110630c9 100644
--- a/outlines/prompts.py
+++ b/outlines/prompts.py
@@ -1,12 +1,14 @@
 import functools
 import inspect
 import json
+import os
 import re
 import textwrap
 from dataclasses import dataclass
-from typing import Any, Callable, Dict, List, Optional, Type, cast
+from pathlib import Path
+from typing import Any, Callable, Dict, Optional, Type, cast
 
-from jinja2 import Environment, StrictUndefined
+import jinja2
 from pydantic import BaseModel
 
 
@@ -19,12 +21,8 @@ class Prompt:
 
     """
 
-    template: str
-    signature: inspect.Signature
-
-    def __post_init__(self):
-        self.parameters: List[str] = list(self.signature.parameters.keys())
-        self.jinja_environment = create_jinja_template(self.template)
+    template: jinja2.Template
+    signature: Optional[inspect.Signature]
 
     def __call__(self, *args, **kwargs) -> str:
         """Render and return the template.
@@ -34,12 +32,93 @@ def __call__(self, *args, **kwargs) -> str:
         The rendered template as a Python ``str``.
 
         """
-        bound_arguments = self.signature.bind(*args, **kwargs)
-        bound_arguments.apply_defaults()
-        return self.jinja_environment.render(**bound_arguments.arguments)
+        if self.signature is not None:
+            bound_arguments = self.signature.bind(*args, **kwargs)
+            bound_arguments.apply_defaults()
+            return self.template.render(**bound_arguments.arguments)
+        else:
+            return self.template.render(**kwargs)
 
-    def __str__(self):
-        return self.template
+    @classmethod
+    def from_str(cls, content: str):
+        """
+        Create an instance of the class from a string.
+
+        Parameters
+        ----------
+        content : str
+            The string content to be converted into a template.
+
+        Returns
+        -------
+        An instance of the class with the provided content as a template.
+        """
+        return cls(cls._template_from_str(content), None)
+
+    @classmethod
+    def from_file(cls, path: Path):
+        """
+        Create a Prompt instance from a file containing a Jinja template.
+
+        Note: This method does not allow to include and inheritance to reference files
+        that are outside the folder or subfolders of the file given to `from_file`.
+
+        Parameters
+        ----------
+        path : Path
+            The path to the file containing the Jinja template.
+
+        Returns
+        -------
+        Prompt
+            An instance of the Prompt class with the template loaded from the file.
+        """
+        # We don't use a `Signature` here because it seems not feasible to infer one from a Jinja2 environment that is
+        # split across multiple files (since e.g. we support features like Jinja2 includes and template inheritance)
+        return cls(cls._template_from_file(path), None)
+
+    @classmethod
+    def _template_from_str(_, content: str) -> jinja2.Template:
+        # Dedent, and remove extra linebreak
+        cleaned_template = inspect.cleandoc(content)
+
+        # Add linebreak if there were any extra linebreaks that
+        # `cleandoc` would have removed
+        ends_with_linebreak = content.replace(" ", "").endswith("\n\n")
+        if ends_with_linebreak:
+            cleaned_template += "\n"
+
+        # Remove extra whitespaces, except those that immediately follow a newline symbol.
+        # This is necessary to avoid introducing whitespaces after backslash `\` characters
+        # used to continue to the next line without linebreak.
+        cleaned_template = re.sub(r"(?![\r\n])(\b\s+)", " ", cleaned_template)
+
+        env = jinja2.Environment(
+            trim_blocks=True,
+            lstrip_blocks=True,
+            keep_trailing_newline=True,
+            undefined=jinja2.StrictUndefined,
+        )
+        env.filters["name"] = get_fn_name
+        env.filters["description"] = get_fn_description
+        env.filters["source"] = get_fn_source
+        env.filters["signature"] = get_fn_signature
+        env.filters["schema"] = get_schema
+        env.filters["args"] = get_fn_args
+
+        return env.from_string(cleaned_template)
+
+    @classmethod
+    def _template_from_file(_, path: Path) -> jinja2.Template:
+        file_directory = os.path.dirname(os.path.abspath(path))
+        env = jinja2.Environment(
+            loader=jinja2.FileSystemLoader(file_directory),
+            trim_blocks=True,
+            lstrip_blocks=True,
+            keep_trailing_newline=True,
+            undefined=jinja2.StrictUndefined,
+        )
+        return env.get_template(os.path.basename(path))
 
 
 def prompt(fn: Callable) -> Prompt:
@@ -92,7 +171,7 @@ def prompt(fn: Callable) -> Prompt:
     return Prompt(template, signature)
 
 
-def render(template: str, **values: Optional[Dict[str, Any]]) -> str:
+def render(content: str, **values: Optional[Dict[str, Any]]) -> str:
     r"""Parse a Jinaj2 template and translate it into an Outlines graph.
 
     This function removes extra whitespaces and linebreaks from templates to
@@ -183,40 +262,8 @@ def render(template: str, **values: Optional[Dict[str, Any]]) -> str:
     A string that contains the rendered template.
 
     """
-    jinja_template = create_jinja_template(template)
-    return jinja_template.render(**values)
-
-
-def create_jinja_template(template: str):
-    # Dedent, and remove extra linebreak
-    cleaned_template = inspect.cleandoc(template)
-
-    # Add linebreak if there were any extra linebreaks that
-    # `cleandoc` would have removed
-    ends_with_linebreak = template.replace(" ", "").endswith("\n\n")
-    if ends_with_linebreak:
-        cleaned_template += "\n"
-
-    # Remove extra whitespaces, except those that immediately follow a newline symbol.
-    # This is necessary to avoid introducing whitespaces after backslash `\` characters
-    # used to continue to the next line without linebreak.
-    cleaned_template = re.sub(r"(?![\r\n])(\b\s+)", " ", cleaned_template)
-
-    env = Environment(
-        trim_blocks=True,
-        lstrip_blocks=True,
-        keep_trailing_newline=True,
-        undefined=StrictUndefined,
-    )
-    env.filters["name"] = get_fn_name
-    env.filters["description"] = get_fn_description
-    env.filters["source"] = get_fn_source
-    env.filters["signature"] = get_fn_signature
-    env.filters["schema"] = get_schema
-    env.filters["args"] = get_fn_args
-
-    jinja_template = env.from_string(cleaned_template)
-    return jinja_template
+    template = Prompt._template_from_str(content)
+    return template.render(**values)
 
 
 def get_fn_name(fn: Callable):
diff --git a/tests/test_prompts.py b/tests/test_prompts.py
index a0433c0e5..de35a547f 100644
--- a/tests/test_prompts.py
+++ b/tests/test_prompts.py
@@ -1,10 +1,12 @@
+import os
+import tempfile
 from typing import Dict, List
 
 import pytest
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.prompts import render
+from outlines.prompts import Prompt, render
 
 
 def test_render():
@@ -110,8 +112,7 @@ def test_prompt_basic():
     def test_tpl(variable):
         """{{variable}} test"""
 
-    assert test_tpl.template == "{{variable}} test"
-    assert test_tpl.parameters == ["variable"]
+    assert list(test_tpl.signature.parameters) == ["variable"]
 
     with pytest.raises(TypeError):
         test_tpl(v="test")
@@ -126,6 +127,8 @@ def test_tpl(variable):
     def test_single_quote_tpl(variable):
         "${variable} test"
 
+    assert list(test_single_quote_tpl.signature.parameters) == ["variable"]
+
     p = test_tpl("test")
     assert p == "test test"
 
@@ -135,8 +138,7 @@ def test_prompt_kwargs():
     def test_kwarg_tpl(var, other_var="other"):
         """{{var}} and {{other_var}}"""
 
-    assert test_kwarg_tpl.template == "{{var}} and {{other_var}}"
-    assert test_kwarg_tpl.parameters == ["var", "other_var"]
+    assert list(test_kwarg_tpl.signature.parameters) == ["var", "other_var"]
 
     p = test_kwarg_tpl("test")
     assert p == "test and other"
@@ -312,3 +314,87 @@ def args_prompt(fn):
         args_prompt(with_all)
         == "args: x1, y1, z1, x2: bool, y2: str, z2: Dict[int, List[str]], x3=True, y3='Hi', z3={4: ['I', 'love', 'outlines']}, x4: bool = True, y4: str = 'Hi', z4: Dict[int, List[str]] = {4: ['I', 'love', 'outlines']}"
     )
+
+
+@pytest.fixture
+def temp_prompt_file():
+    test_dir = tempfile.mkdtemp()
+
+    base_template_path = os.path.join(test_dir, "base_template.txt")
+    with open(base_template_path, "w") as f:
+        f.write(
+            """{% block content %}{% endblock %}
+"""
+        )
+
+    include_file_path = os.path.join(test_dir, "include.txt")
+    with open(include_file_path, "w") as f:
+        f.write(
+            """{% for example in examples %}
+- Q: {{ example.question }}
+- A: {{ example.answer }}
+{% endfor %}
+"""
+        )
+
+    prompt_file_path = os.path.join(test_dir, "prompt.txt")
+    with open(prompt_file_path, "w") as f:
+        f.write(
+            """{% extends "base_template.txt" %}
+
+{% block content %}
+Here is a prompt with examples:
+
+{% include "include.txt" %}
+
+Now please answer the following question:
+
+Q: {{ question }}
+A:
+{% endblock %}
+"""
+        )
+    yield prompt_file_path
+
+
+def test_prompt_from_file(temp_prompt_file):
+    prompt = Prompt.from_file(temp_prompt_file)
+    assert prompt.signature is None
+    examples = [
+        {"question": "What is the capital of France?", "answer": "Paris"},
+        {"question": "What is 2 + 2?", "answer": "4"},
+    ]
+    question = "What is the Earth's diameter?"
+    rendered = prompt(examples=examples, question=question)
+    expected = """Here is a prompt with examples:
+
+- Q: What is the capital of France?
+- A: Paris
+- Q: What is 2 + 2?
+- A: 4
+
+Now please answer the following question:
+
+Q: What is the Earth's diameter?
+A:
+"""
+    assert rendered.strip() == expected.strip()
+
+
+def test_prompt_from_str():
+    content = """
+    Hello, {{ name }}!
+    """
+    prompt = Prompt.from_str(content)
+    assert prompt.signature is None
+    assert prompt(name="World") == "Hello, World!"
+
+
+def test_template_from_str_with_extra_linebreaks():
+    content = """
+    Hello, {{ name }}!
+
+
+    """
+    template = Prompt._template_from_str(content)
+    assert template.render(name="World") == "Hello, World!\n"

From bf27c77b01319de77e41986dc7cb39396a7a3e15 Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Thu, 9 Jan 2025 17:11:51 +0100
Subject: [PATCH 142/505] Remove `prompt.render` method

---
 outlines/prompts.py   | 97 +------------------------------------------
 tests/test_prompts.py |  7 +++-
 2 files changed, 7 insertions(+), 97 deletions(-)

diff --git a/outlines/prompts.py b/outlines/prompts.py
index 2110630c9..66e723205 100644
--- a/outlines/prompts.py
+++ b/outlines/prompts.py
@@ -166,106 +166,11 @@ def prompt(fn: Callable) -> Prompt:
     if docstring is None:
         raise TypeError("Could not find a template in the function's docstring.")
 
-    template = cast(str, docstring)
+    template = Prompt._template_from_str(cast(str, docstring))
 
     return Prompt(template, signature)
 
 
-def render(content: str, **values: Optional[Dict[str, Any]]) -> str:
-    r"""Parse a Jinaj2 template and translate it into an Outlines graph.
-
-    This function removes extra whitespaces and linebreaks from templates to
-    allow users to enter prompts more naturally than if they used Python's
-    constructs directly. See the examples for a detailed explanation.
-
-    Examples
-    --------
-
-    Outlines follow Jinja2's syntax
-
-    >>> import outlines
-    >>> outline = outlines.render("I like {{food}} and {{sport}}", food="tomatoes", sport="tennis")
-    I like tomatoes and tennis
-
-    If the first line of the template is empty, `render` removes it
-
-    >>> from outlines import render
-    >>>
-    >>> tpl = '''
-    ... A new string'''
-    >>> tpl
-    ... '\nA new string'
-    >>> render(tpl)
-    ... 'a new string'
-
-    Similarly, `render` ignores linebreaks introduced by placing the closing quotes
-    underneath the text:
-
-    >>> tpl = '''
-    ... A new string
-    ... '''
-    >>> tpl
-    ... '\nA new string\n'
-    >>> render(tpl)
-    ... 'A new string'
-
-    If you want to insert a linebreak at the end of the rendered template, you will
-    need to leave an empty line at the end of the template:
-
-    >>> tpl = '''
-    ... A new string
-    ...
-    ... '''
-    >>> tpl
-    ... '\nA new string\n\n'
-    >>> render(tpl)
-    ... 'A new string\n'
-
-    `render` removes the identation in docstrings. This is particularly important
-    when using prompt functions
-
-    >>> tpl = '''
-    ...    a string
-    ...    and another string'''
-    >>> tpl
-    ... '\n   a string\n   and another string'
-    >>> render(tpl)
-    ... 'a string\nand another string'
-
-    The indentation of the first line is assumed to be the same as the second line's
-
-    >>> tpl = '''a string
-    ...     and another'''
-    >>> tpl
-    ... 'a string\n    and another'
-    >>> render(tpl)
-    ... 'a string\nand another'
-
-    To get a different indentation for the first and the second line, we can start the
-    prompt on the string's second line:
-
-    >>> tpl = '''
-    ... First line
-    ...   Second line'''
-    >>> render(tpl)
-    ... 'First Line\n  Second Line'
-
-    Parameters
-    ----------
-    template
-        A string that contains a template written with the Jinja2 syntax.
-    **values
-        Map from the variables in the template to their value.
-
-    Returns
-    -------
-    A string that contains the rendered template.
-
-    """
-    template = Prompt._template_from_str(content)
-    return template.render(**values)
-
-
 def get_fn_name(fn: Callable):
     """Returns the name of a callable."""
     if not callable(fn):
diff --git a/tests/test_prompts.py b/tests/test_prompts.py
index de35a547f..4cc4d8ff1 100644
--- a/tests/test_prompts.py
+++ b/tests/test_prompts.py
@@ -6,7 +6,12 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.prompts import Prompt, render
+from outlines.prompts import Prompt
+
+
+def render(content: str, **kwargs):
+    template = Prompt._template_from_str(content)
+    return template.render(kwargs)
 
 
 def test_render():

From 7149a5e613b9c2bb3c2ed3ee5b7c9228772504b1 Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Thu, 9 Jan 2025 17:12:43 +0100
Subject: [PATCH 143/505] Apply a small fix suggested by linter: Ruff E721

Use `is` and `is not` for type comparisons, or `isinstance()` for isinstance checks
---
 outlines/prompts.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/outlines/prompts.py b/outlines/prompts.py
index 66e723205..86519adaf 100644
--- a/outlines/prompts.py
+++ b/outlines/prompts.py
@@ -9,7 +9,7 @@
 from typing import Any, Callable, Dict, Optional, Type, cast
 
 import jinja2
-from pydantic import BaseModel
+import pydantic
 
 
 @dataclass
@@ -253,10 +253,10 @@ def get_schema_dict(model: Dict):
     return json.dumps(model, indent=2)
 
 
-@get_schema.register(type(BaseModel))
-def get_schema_pydantic(model: Type[BaseModel]):
+@get_schema.register(type(pydantic.BaseModel))
+def get_schema_pydantic(model: Type[pydantic.BaseModel]):
     """Return the schema of a Pydantic model."""
-    if not type(model) == type(BaseModel):
+    if not isinstance(model, type(pydantic.BaseModel)):
         raise TypeError("The `schema` filter only applies to Pydantic models.")
 
     if hasattr(model, "model_json_schema"):

From e7be56284a0ad578ca1b0b20307504be8c9e7fbd Mon Sep 17 00:00:00 2001
From: Per von Rosen <46034369+pervrosen@users.noreply.github.com>
Date: Sat, 11 Jan 2025 16:23:18 +0100
Subject: [PATCH 144/505] Update react_agent.md with current import structure

the old library structure has not been updated to reflect the present one.
---
 docs/cookbook/react_agent.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/cookbook/react_agent.md b/docs/cookbook/react_agent.md
index 0597eab07..15b6ed369 100644
--- a/docs/cookbook/react_agent.md
+++ b/docs/cookbook/react_agent.md
@@ -120,8 +120,8 @@ class Decision(BaseModel):
 We could generate a response using the json schema but we will use the regex and check that everything is working as expected:
 
 ```python
-from outlines.integrations.utils import convert_json_schema_to_str
-from outlines.fsm.json_schema import build_regex_from_schema
+from outlines.fsm.json_schema import convert_json_schema_to_str
+from outlines_core.fsm.json_schema import build_regex_from_schema
 
 json_schema = Decision.model_json_schema()
 schema_str = convert_json_schema_to_str(json_schema=json_schema)

From 03d387893b95f151c74ff6b7d7864e15cb484eee Mon Sep 17 00:00:00 2001
From: Joao Gante <joaofranciscocardosogante@gmail.com>
Date: Thu, 9 Jan 2025 18:45:17 +0000
Subject: [PATCH 145/505] Update transformers_vision.md

---
 docs/reference/models/transformers_vision.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
index b488bca1a..451e54a8f 100644
--- a/docs/reference/models/transformers_vision.md
+++ b/docs/reference/models/transformers_vision.md
@@ -2,7 +2,7 @@
 
 Outlines allows seamless use of [vision models](https://huggingface.co/learn/computer-vision-course/en/unit4/multimodal-models/tasks-models-part1).
 
-`outlines.models.transformers_vision` has shares interfaces with, and is based on [outlines.models.transformers](./transformers.md).
+`outlines.models.transformers_vision` shares interfaces with, and is based on [outlines.models.transformers](./transformers.md).
 
 Tasks supported include
 
@@ -62,7 +62,7 @@ image_urls = [
 ]
 description_generator = outlines.generate.text(model)
 description_generator(
-    "<image><image><image>What shapes are present?",
+    "<image><image>What shapes are present?",
     list(map(img_from_url, image_urls)),
 )
 ```
@@ -110,6 +110,6 @@ image_data_generator(
 
 ## Resources
 
-### Chosing a model
+### Choosing a model
 - https://mmbench.opencompass.org.cn/leaderboard
 - https://huggingface.co/spaces/WildVision/vision-arena

From 3af1e5ac56a912433a948a070e8e7e41a3d157cf Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Mon, 6 Jan 2025 18:33:33 +0100
Subject: [PATCH 146/505] Fix `vllm`-related pytest warning (that was spaming
 user)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Before this commit, when you ran pytest -k specific_test, it spawned dozens of the same skipped warnings message on stdout... IMHO, that was not ideal ^^

Bug introduced in d32dfde3ae7eac71f57af96d5141975ef9322b12
---
 tests/conftest.py | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/tests/conftest.py b/tests/conftest.py
index bbc56afdb..ea99b69ff 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -5,10 +5,13 @@
 
 def pytest_collection_modifyitems(config, items):
     if sys.platform != "linux":
+        if not config.option.keyword or (
+            config.option.keyword and "test_integration_vllm" in config.option.keyword
+        ):
+            print(
+                "WARNING: test_integration_vllm tests are skipped because vLLM only supports Linux platform (including WSL)."
+            )
         skip_vllm = pytest.mark.skip(reason="vLLM models can only be run on Linux.")
         for item in items:
             if "test_integration_vllm" in item.nodeid:
                 item.add_marker(skip_vllm)
-                print(
-                    f"WARNING: {item.nodeid} is skipped because vLLM only supports Linux platform (including WSL)."
-                )

From 79100b293f861ee8bd74390527ec25cb6fe3b670 Mon Sep 17 00:00:00 2001
From: Frederik Fix <ich@derfred.com>
Date: Tue, 14 Jan 2025 11:32:30 +0100
Subject: [PATCH 147/505] Add filters to prompt function (#1371)

Allow giving custom filters to the prompt decorator

```
def reverses: str) -> str:
    return s[::-1]

@prompt(filters={ 'reverse': reverse })
def reverse_prompt(text):
    '''{{ text | reverse }}'''

prompt = reverse_prompt("Hello")

print(prompt)
>>> "olleH"
```
---
 outlines/prompts.py   | 71 ++++++++++++++++++++++++++++++-------------
 tests/test_prompts.py | 17 +++++++++++
 2 files changed, 67 insertions(+), 21 deletions(-)

diff --git a/outlines/prompts.py b/outlines/prompts.py
index 86519adaf..1cc264226 100644
--- a/outlines/prompts.py
+++ b/outlines/prompts.py
@@ -40,7 +40,7 @@ def __call__(self, *args, **kwargs) -> str:
             return self.template.render(**kwargs)
 
     @classmethod
-    def from_str(cls, content: str):
+    def from_str(cls, content: str, filters: Dict[str, Callable] = {}):
         """
         Create an instance of the class from a string.
 
@@ -53,10 +53,10 @@ def from_str(cls, content: str):
         -------
         An instance of the class with the provided content as a template.
         """
-        return cls(cls._template_from_str(content), None)
+        return cls(cls._template_from_str(content, filters), None)
 
     @classmethod
-    def from_file(cls, path: Path):
+    def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
         """
         Create a Prompt instance from a file containing a Jinja template.
 
@@ -75,10 +75,12 @@ def from_file(cls, path: Path):
         """
         # We don't use a `Signature` here because it seems not feasible to infer one from a Jinja2 environment that is
         # split across multiple files (since e.g. we support features like Jinja2 includes and template inheritance)
-        return cls(cls._template_from_file(path), None)
+        return cls(cls._template_from_file(path, filters), None)
 
     @classmethod
-    def _template_from_str(_, content: str) -> jinja2.Template:
+    def _template_from_str(
+        _, content: str, filters: Dict[str, Callable] = {}
+    ) -> jinja2.Template:
         # Dedent, and remove extra linebreak
         cleaned_template = inspect.cleandoc(content)
 
@@ -93,12 +95,7 @@ def _template_from_str(_, content: str) -> jinja2.Template:
         # used to continue to the next line without linebreak.
         cleaned_template = re.sub(r"(?![\r\n])(\b\s+)", " ", cleaned_template)
 
-        env = jinja2.Environment(
-            trim_blocks=True,
-            lstrip_blocks=True,
-            keep_trailing_newline=True,
-            undefined=jinja2.StrictUndefined,
-        )
+        env = create_jinja_env(None, filters)
         env.filters["name"] = get_fn_name
         env.filters["description"] = get_fn_description
         env.filters["source"] = get_fn_source
@@ -109,19 +106,19 @@ def _template_from_str(_, content: str) -> jinja2.Template:
         return env.from_string(cleaned_template)
 
     @classmethod
-    def _template_from_file(_, path: Path) -> jinja2.Template:
+    def _template_from_file(
+        _, path: Path, filters: Dict[str, Callable] = {}
+    ) -> jinja2.Template:
         file_directory = os.path.dirname(os.path.abspath(path))
-        env = jinja2.Environment(
-            loader=jinja2.FileSystemLoader(file_directory),
-            trim_blocks=True,
-            lstrip_blocks=True,
-            keep_trailing_newline=True,
-            undefined=jinja2.StrictUndefined,
-        )
+        env = create_jinja_env(jinja2.FileSystemLoader(file_directory), filters)
+
         return env.get_template(os.path.basename(path))
 
 
-def prompt(fn: Callable) -> Prompt:
+def prompt(
+    fn: Optional[Callable] = None,
+    filters: Dict[str, Callable] = {},
+) -> Callable:
     """Decorate a function that contains a prompt template.
 
     This allows to define prompts in the docstring of a function and simplify their
@@ -152,11 +149,26 @@ def prompt(fn: Callable) -> Prompt:
     ...
     >>> hal = ft.partial(solve_task, "HAL", "Travel to Jupiter")
 
+    Additional Jinja2 filters can be provided as keyword arguments to the decorator.
+
+    >>> def reverse(s: str) -> str:
+    ...     return s[::-1]
+    ...
+    >>> @outlines.prompt(filters={ 'reverse': reverse })
+    ... def reverse_prompt(text):
+    ...     '''{{ text | reverse }}'''
+    ...
+    >>> prompt = reverse_prompt("Hello")
+    >>> print(prompt)
+    ... "olleH"
+
     Returns
     -------
     A `Prompt` callable class which will render the template when called.
 
     """
+    if fn is None:
+        return lambda fn: prompt(fn, cast(Dict[str, Callable], filters))
 
     signature = inspect.signature(fn)
 
@@ -166,11 +178,28 @@ def prompt(fn: Callable) -> Prompt:
     if docstring is None:
         raise TypeError("Could not find a template in the function's docstring.")
 
-    template = Prompt._template_from_str(cast(str, docstring))
+    template = Prompt._template_from_str(cast(str, docstring), filters)
 
     return Prompt(template, signature)
 
 
+def create_jinja_env(
+    loader: Optional[jinja2.BaseLoader], filters: Dict[str, Callable]
+) -> jinja2.Environment:
+    env = jinja2.Environment(
+        loader=loader,
+        trim_blocks=True,
+        lstrip_blocks=True,
+        keep_trailing_newline=True,
+        undefined=jinja2.StrictUndefined,
+    )
+
+    for name, filter_fn in filters.items():
+        env.filters[name] = filter_fn
+
+    return env
+
+
 def get_fn_name(fn: Callable):
     """Returns the name of a callable."""
     if not callable(fn):
diff --git a/tests/test_prompts.py b/tests/test_prompts.py
index 4cc4d8ff1..f59c04ac0 100644
--- a/tests/test_prompts.py
+++ b/tests/test_prompts.py
@@ -321,6 +321,23 @@ def args_prompt(fn):
     )
 
 
+def test_prompt_with_additional_filters():
+    def reverse(s: str) -> str:
+        return s[::-1]
+
+    @outlines.prompt(filters=dict(reverse=reverse))
+    def test_tpl(variable):
+        """{{ variable | reverse }} test"""
+
+    assert list(test_tpl.signature.parameters) == ["variable"]
+
+    p = test_tpl("test")
+    assert p == "tset test"
+
+    p = test_tpl(variable="example")
+    assert p == "elpmaxe test"
+
+
 @pytest.fixture
 def temp_prompt_file():
     test_dir = tempfile.mkdtemp()

From a03c25451f8b5be95f76de95cf759d52b99bd2cf Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Tue, 14 Jan 2025 10:43:25 -0800
Subject: [PATCH 148/505] Remove incorrect text from feature matrix docs

There's an extra `outlines.generate` row in the feature matrix docs. This removes it.

I also modified the markdown syntax for one header to use ** rather than __, consistent with the rest of the table.
---
 docs/reference/models/models.md | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
index 4d2dda8c9..eab70f2a1 100644
--- a/docs/reference/models/models.md
+++ b/docs/reference/models/models.md
@@ -40,9 +40,8 @@ model = outlines.models.openai(
 | **Generation**    |              |                     |      |           |           |       |         |
 | Batch             | ✅           | ✅                  | ✅   | ❌        | ?         | ❌    | ❌      |
 | Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
-| **`outlines.generate`** |        |                     |      |           |           |       |         |
 | Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| __Structured__    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| **Structured**    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
 | JSON Schema       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
 | Choice            | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
 | Regex             | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |

From 7b9012b49de25a6957039a3b5e53452c56b26e8c Mon Sep 17 00:00:00 2001
From: Marco De Nadai <me@marcodena.it>
Date: Wed, 15 Jan 2025 13:25:12 +0100
Subject: [PATCH 149/505] Update vllm_integration.py to the latest version of
 outlines (#1352)

---
 examples/vllm_integration.py | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/examples/vllm_integration.py b/examples/vllm_integration.py
index c9ce057bc..ee1b24a33 100644
--- a/examples/vllm_integration.py
+++ b/examples/vllm_integration.py
@@ -2,8 +2,10 @@
 
 import vllm
 from pydantic import BaseModel
+from transformers import AutoTokenizer
 
-from outlines.processors.structured import JSONLogitsProcessor
+from outlines.models.vllm import adapt_tokenizer
+from outlines.processors import JSONLogitsProcessor
 
 
 class Person(BaseModel):
@@ -11,8 +13,12 @@ class Person(BaseModel):
     surname: str
 
 
-llm = vllm.LLM(model="mistralai/Mistral-7B-v0.1", max_model_len=512)
-logits_processor = JSONLogitsProcessor(schema=Person, llm=llm, whitespace_pattern=r" ?")
+MODEL_ID = "mistralai/Mistral-7B-v0.1"
+llm = vllm.LLM(model=MODEL_ID, max_model_len=512)
+tokenizer = adapt_tokenizer(AutoTokenizer.from_pretrained(MODEL_ID))
+logits_processor = JSONLogitsProcessor(
+    schema=Person, tokenizer=tokenizer, whitespace_pattern=r" ?"
+)
 result = llm.generate(
     ["He is Tom Jones", "She saw Linda Smith"],
     sampling_params=vllm.SamplingParams(

From 088f4399296a038123712a2ac7e263c1b1b95714 Mon Sep 17 00:00:00 2001
From: Neil Mehta <neil@lmstudio.ai>
Date: Wed, 15 Jan 2025 07:26:47 -0500
Subject: [PATCH 150/505] [Preserve dtype of array when converting to torch
 (#1349)

We have noticing the following error with a recent version of outlines
when used with MLX:
```
TypeError: argument 'token_id': 'float' object cannot be interpreted as an integer

At:
  /.../outlines_core/fsm/guide.py(294): get_next_state
  /.../outlines/processors/structured.py(101): process_logits
  /.../outlines/processors/base_logits_processor.py(90): __call__
```

The issue is that the MLX array of tokens, which are integers, are being
force-converted to floats, even though outlines expects an integer
array. This is because all MLX arrays are being converted to `float32`,
even when it's not necessarily appropriate, like in this case. Looking
at the [commented
link](https://ml-explore.github.io/mlx/build/html/usage/numpy.html#pytorch),
the advice was to convert to `float32` only for `bfloat16`, because
numpy does not support `bfloat16`. Now the MLX `_to_torch`
implementation matches the other array libraries, none of the other
libraries are being force-casted to float
---
 outlines/processors/base_logits_processor.py | 6 +++---
 tests/processors/test_base_processor.py      | 8 +++++++-
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index 44b55af2e..800e69f79 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -110,9 +110,9 @@ def _to_torch(tensor_like: Array) -> torch.Tensor:
             import mlx.core as mx
 
             # https://ml-explore.github.io/mlx/build/html/usage/numpy.html#pytorch
-            return torch.from_dlpack(
-                np.array(tensor_like.astype(mx.float32), copy=False)
-            )
+            if tensor_like.dtype == mx.bfloat16:
+                tensor_like = tensor_like.astype(mx.float32)
+            return torch.from_dlpack(np.array(tensor_like, copy=False))
 
         elif is_jax_array_type(type(tensor_like)):
             import jax
diff --git a/tests/processors/test_base_processor.py b/tests/processors/test_base_processor.py
index cd9f48278..d2a1e1af2 100644
--- a/tests/processors/test_base_processor.py
+++ b/tests/processors/test_base_processor.py
@@ -18,6 +18,7 @@
     import mlx.core as mx
 
     arrays["mlx"] = mx.array([[1, 2], [3, 4]], dtype=mx.float32)
+    arrays["mlx_bfloat16"] = mx.array([[1, 2], [3, 4]], dtype=mx.bfloat16)
 except ImportError:
     pass
 
@@ -59,7 +60,12 @@ def test_from_torch(array_type, processor):
     torch_tensor = torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
     data = processor._from_torch(torch_tensor, type(arrays[array_type]))
     assert isinstance(data, type(arrays[array_type]))
-    assert np.allclose(data, arrays[array_type])
+    if array_type == "mlx_bfloat16":
+        # For bfloat16, we expect the output to be float32 due to the conversion
+        assert data.dtype == mx.float32
+        assert np.allclose(np.array(data), np.array([[1, 2], [3, 4]], dtype=np.float32))
+    else:
+        assert np.allclose(data, arrays[array_type])
 
 
 @pytest.mark.parametrize("array_type", arrays.keys())

From 8b173a315d23d872f1347e6539af4ad2e22b18a7 Mon Sep 17 00:00:00 2001
From: Petros Amoiridis <petros@hey.com>
Date: Thu, 16 Jan 2025 10:34:21 +0200
Subject: [PATCH 151/505] Fix typo in docs/community/examples

---
 docs/community/examples.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/community/examples.md b/docs/community/examples.md
index 2ebaf2765..813bd4834 100644
--- a/docs/community/examples.md
+++ b/docs/community/examples.md
@@ -1,6 +1,6 @@
 # Community projects and articles
 
-Publishing examples and articles about Outlines are a meaningful way to contrinute to the community. Here is a list of projects we are aware of. Drop us a line if we forgot yours!
+Publishing examples and articles about Outlines are a meaningful way to contribute to the community. Here is a list of projects we are aware of. Drop us a line if we forgot yours!
 
 [MMSG](https://github.com/leloykun/mmsg) is a Python library for generating interleaved text and image content in a structured format you can directly pass to downstream APIs.
 

From 063291d816015cbdb04b66a8db6c6020b619252a Mon Sep 17 00:00:00 2001
From: Mohammad Nasirifar <m@fireworks.ai>
Date: Fri, 17 Jan 2025 19:50:38 +0000
Subject: [PATCH 152/505] Fix readme to reflect routing through enum

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index d34b0984d..e3e6dc24d 100644
--- a/README.md
+++ b/README.md
@@ -343,7 +343,7 @@ class Operation(Enum):
     mul = partial(mul)
 
 model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = outlines.generate.json(model, add)
+generator = outlines.generate.json(model, Operation)
 result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
 
 print(result)

From 51ef66388799dd285afa7393db15f773193d4e0e Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Wed, 22 Jan 2025 11:19:05 +0100
Subject: [PATCH 153/505] Sort, clean and update .gitignore

---
 .gitignore | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

diff --git a/.gitignore b/.gitignore
index 08390ae3d..eefa0cb13 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,14 +1,19 @@
-*.egg-info
 __pycache__
-*_version.py
-docs/build
+.benchmarks
+.cache
 .coverage
-.idea/
-*.gguf
+.direnv
+.env
+.idea
+.pytest_cache
+.python-version
 .venv
+*_version.py
+*.egg-info
+*.gguf
 benchmarks/results
-.python-version
-
-# Remove doc build folders
-.cache/
-build/
+build
+docs/build
+logs
+logs/
+__pycache__/

From ea4904ac5a2d1064ff4c00a19a66a21dc27c9b43 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Thu, 23 Jan 2025 09:18:13 -0800
Subject: [PATCH 154/505] Update README to include chat templating (#1372)

The existing README has underwhelming or incorrect results (Example is
underwhelming #1347) due to lack of templating for instruct models.

This adds special tokens for each instruct model call, as well as
provide comments on how to obtain/produce special tokens.

---------

Co-authored-by: Victoria Terenina <torymur@gmail.com>
---
 README.md                         | 88 ++++++++++++++++++----------
 docs/reference/chat_templating.md | 38 ++++++++++++
 mkdocs.yml                        | 97 +++++++++++++++----------------
 3 files changed, 143 insertions(+), 80 deletions(-)
 create mode 100644 docs/reference/chat_templating.md

diff --git a/README.md b/README.md
index e3e6dc24d..b3cde1dd1 100644
--- a/README.md
+++ b/README.md
@@ -65,6 +65,21 @@ is to ensure that there is a well-defined interface between their output and
 user-defined code. **Outlines** provides ways to control the generation of
 language models to make their output more predictable.
 
+The following methods of structured generation are supported:
+
+- [Multiple choices](#multiple-choices)
+- [Type constraints](#type-constraint)
+- [Efficient regex-structured generation](#efficient-regex-structured-generation)
+- [Efficient JSON generation following a Pydantic model](#efficient-json-generation-following-a-pydantic-model)
+- [Using context-free grammars to guide generation](#using-context-free-grammars-to-guide-generation)
+- [Open functions](#open-functions)
+
+### Chat template tokens
+
+Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself. Chat template tokens are not needed for base models.
+
+Please see [the documentation](https://dottxt-ai.github.io/outlines/latest/reference/chat_templating) on chat templating for more.
+
 ### Multiple choices
 
 You can reduce the completion to a choice between multiple possibilities:
@@ -72,42 +87,47 @@ You can reduce the completion to a choice between multiple possibilities:
 ``` python
 import outlines
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+model_name = "HuggingFaceTB/SmolLM2-360M-Instruct"
+model = outlines.models.transformers(model_name)
+
+# You must apply the chat template tokens to the prompt!
+# See below for an example.
+prompt = """
+<|im_start|>system
+You extract information from text.
+<|im_end|>
 
-prompt = """You are a sentiment-labelling assistant.
-Is the following review positive or negative?
+<|im_start|>user
+What food does the following text describe?
 
-Review: This restaurant is just awesome!
+Text: I really really really want pizza.
+<|im_end|>
+<|im_start|>assistant
 """
 
-generator = outlines.generate.choice(model, ["Positive", "Negative"])
+generator = outlines.generate.choice(model, ["Pizza", "Pasta", "Salad", "Dessert"])
 answer = generator(prompt)
+
+# Likely answer: Pizza
 ```
 
-You can also pass these choices through en enum:
+You can also pass in choices with an `Enum`:
 
 ````python
 from enum import Enum
 
-import outlines
-
-class Sentiment(str, Enum):
-    positive = "Positive"
-    negative = "Negative"
+class Food(str, Enum):
+    pizza = "Pizza"
+    pasta = "Pasta"
+    salad = "Salad"
+    dessert = "Dessert"
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
-prompt = """You are a sentiment-labelling assistant.
-Is the following review positive or negative?
-
-Review: This restaurant is just awesome!
-"""
-
-generator = outlines.generate.choice(model, Sentiment)
+generator = outlines.generate.choice(model, Food)
 answer = generator(prompt)
+# Likely answer: Pizza
 ````
 
-### Type constraint
+### Type constraints
 
 You can instruct the model to only return integers or floats:
 
@@ -140,7 +160,17 @@ import outlines
 
 model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
-prompt = "What is the IP address of the Google DNS servers? "
+prompt = """
+<|im_start|>system You are a helpful assistant.
+<|im_end|>
+
+<|im_start|>user
+What is an IP address of the Google DNS servers?
+<|im_end|>
+<|im_start|>assistant
+The IP address of a Google DNS server is
+
+"""
 
 generator = outlines.generate.text(model)
 unstructured = generator(prompt, max_tokens=30)
@@ -148,19 +178,17 @@ unstructured = generator(prompt, max_tokens=30)
 generator = outlines.generate.regex(
     model,
     r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)",
+    sampler=outlines.samplers.greedy(),
 )
 structured = generator(prompt, max_tokens=30)
 
 print(unstructured)
-# What is the IP address of the Google DNS servers?
+# 8.8.8.8
 #
-# Passive DNS servers are at DNS servers that are private.
-# In other words, both IP servers are private. The database
-# does not contain Chelsea Manning
+# <|im_end|>
 
 print(structured)
-# What is the IP address of the Google DNS servers?
-# 2.2.6.1
+# 8.8.8.8
 ```
 
 Unlike other libraries, regex-structured generation in Outlines is almost as fast
@@ -168,15 +196,13 @@ as non-structured generation.
 
 ### Efficient JSON generation following a Pydantic model
 
-Outlines  allows to guide the generation process so the output is *guaranteed* to follow a [JSON schema](https://json-schema.org/) or [Pydantic model](https://docs.pydantic.dev/latest/):
+Outlines users can guide the generation process so the output is *guaranteed* to follow a [JSON schema](https://json-schema.org/) or [Pydantic model](https://docs.pydantic.dev/latest/):
 
 ```python
 from enum import Enum
 from pydantic import BaseModel, constr
 
 import outlines
-import torch
-
 
 class Weapon(str, Enum):
     sword = "sword"
diff --git a/docs/reference/chat_templating.md b/docs/reference/chat_templating.md
new file mode 100644
index 000000000..e3f33fbfd
--- /dev/null
+++ b/docs/reference/chat_templating.md
@@ -0,0 +1,38 @@
+# Chat templating
+
+Instruction-tuned language models use "special tokens" to indicate different parts of text, such as the system prompt, the user prompt, any images, and the assistant's response. A [chat template](https://huggingface.co/docs/transformers/main/en/chat_templating) is how different types of input are composited together into a single, machine-readable string.
+
+Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself -- if you do not apply chat templating on instruction-tuned models, you will often get nonsensical output from the model.
+
+Chat template tokens are not needed for base models.
+
+You can find the chat template tokens in the model's HuggingFace repo or documentation. As an example, the `SmolLM2-360M-Instruct` special tokens can be found [here](https://huggingface.co/HuggingFaceTB/SmolLM2-360M-Instruct/blob/main/special_tokens_map.json).
+
+However, it can be slow to manually look up a model's special tokens, and special tokens vary by models. If you change the model, your prompts may break if you have hard-coded special tokens.
+
+If you need a convenient tool to apply chat templating for you, you should use the `tokenizer` from the `transformers` library:
+
+```python
+from transformers import AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/SmolLM2-360M-Instruct")
+prompt = tokenizer.apply_chat_template(
+    [
+        {"role": "system", "content": "You extract information from text."},
+        {"role": "user", "content": "What food does the following text describe?"},
+    ],
+    tokenize=False,
+    add_bos=True,
+    add_generation_prompt=True,
+)
+```
+
+yields
+
+```
+<|im_start|>system
+You extract information from text.<|im_end|>
+<|im_start|>user
+What food does the following text describe?<|im_end|>
+<|im_start|>assistant
+```
diff --git a/mkdocs.yml b/mkdocs.yml
index 7a3ca1e98..8d051d1af 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -4,7 +4,6 @@ site_author: The Outlines developers
 site_description: >-
   Structured text generation with LLMs
 
-
 # Repository
 repo_name: dottxt-ai/outlines
 repo_url: https://github.com/dottxt-ai/outlines
@@ -76,7 +75,6 @@ markdown_extensions:
       emoji_generator: !!python/name:material.extensions.emoji.to_svg
   - pymdownx.snippets:
 
-
 extra_css:
   - stylesheets/extra.css
 
@@ -131,53 +129,54 @@ nav:
           - Cerebrium: cookbook/deploy-using-cerebrium.md
           - Modal: cookbook/deploy-using-modal.md
   - Docs:
-    - reference/index.md
-    - Generation:
-        - Overview: reference/generation/generation.md
-        - Text: reference/text.md
-        - Samplers: reference/samplers.md
-        - Structured generation:
-            - How does it work?: reference/generation/structured_generation_explanation.md
-            - Classification: reference/generation/choices.md
-            - Regex: reference/generation/regex.md
-            - Type constraints: reference/generation/format.md
-            - JSON (function calling): reference/generation/json.md
-            - Grammar: reference/generation/cfg.md
-            - Creating Grammars: reference/generation/creating_grammars.md
-            - Custom FSM operations: reference/generation/custom_fsm_ops.md
-    - Utilities:
-        - Serve with vLLM: reference/serve/vllm.md
-        - Serve with LM Studio: reference/serve/lmstudio.md
-        - Custom types: reference/generation/types.md
-        - Prompt templating: reference/prompting.md
-        - Outlines functions: reference/functions.md
-    - Models:
-        - Overview: reference/models/models.md
-        - Open source:
-          - Transformers: reference/models/transformers.md
-          - Transformers Vision: reference/models/transformers_vision.md
-          - Llama.cpp: reference/models/llamacpp.md
-          - vLLM: reference/models/vllm.md
-          - TGI: reference/models/tgi.md
-          - ExllamaV2: reference/models/exllamav2.md
-          - MLX: reference/models/mlxlm.md
-          - Mamba: reference/models/transformers/#mamba
-        - API:
-            - OpenAI: reference/models/openai.md
+      - reference/index.md
+      - Generation:
+          - Overview: reference/generation/generation.md
+          - Chat templating: reference/chat_templating.md
+          - Text: reference/text.md
+          - Samplers: reference/samplers.md
+          - Structured generation:
+              - How does it work?: reference/generation/structured_generation_explanation.md
+              - Classification: reference/generation/choices.md
+              - Regex: reference/generation/regex.md
+              - Type constraints: reference/generation/format.md
+              - JSON (function calling): reference/generation/json.md
+              - Grammar: reference/generation/cfg.md
+              - Creating Grammars: reference/generation/creating_grammars.md
+              - Custom FSM operations: reference/generation/custom_fsm_ops.md
+      - Utilities:
+          - Serve with vLLM: reference/serve/vllm.md
+          - Serve with LM Studio: reference/serve/lmstudio.md
+          - Custom types: reference/generation/types.md
+          - Prompt templating: reference/prompting.md
+          - Outlines functions: reference/functions.md
+      - Models:
+          - Overview: reference/models/models.md
+          - Open source:
+              - Transformers: reference/models/transformers.md
+              - Transformers Vision: reference/models/transformers_vision.md
+              - Llama.cpp: reference/models/llamacpp.md
+              - vLLM: reference/models/vllm.md
+              - TGI: reference/models/tgi.md
+              - ExllamaV2: reference/models/exllamav2.md
+              - MLX: reference/models/mlxlm.md
+              - Mamba: reference/models/transformers/#mamba
+          - API:
+              - OpenAI: reference/models/openai.md
   - API Reference:
-    - api/index.md
-    - api/models.md
-    - api/prompts.md
-    - api/json_schema.md
-    - api/guide.md
-    - api/parsing.md
-    - api/regex.md
-    - api/samplers.md
+      - api/index.md
+      - api/models.md
+      - api/prompts.md
+      - api/json_schema.md
+      - api/guide.md
+      - api/parsing.md
+      - api/regex.md
+      - api/samplers.md
   - Community:
-    - community/index.md
-    - Feedback 🫶: community/feedback.md
-    - Chat with us ☕: https://discord.com/invite/R9DSu34mGd
-    - How to contribute 🏗️: community/contribute.md
-    - Your projects 👏: community/examples.md
-    - Versioning Guide 📌: community/versioning.md
+      - community/index.md
+      - Feedback 🫶: community/feedback.md
+      - Chat with us ☕: https://discord.com/invite/R9DSu34mGd
+      - How to contribute 🏗️: community/contribute.md
+      - Your projects 👏: community/examples.md
+      - Versioning Guide 📌: community/versioning.md
   - Blog: blog/index.md

From 437ffe4f73f8275ec3ca6e5b19c692913f5fcf98 Mon Sep 17 00:00:00 2001
From: g-prz <158364710+g-prz@users.noreply.github.com>
Date: Mon, 27 Jan 2025 16:22:08 +0100
Subject: [PATCH 155/505] Add `genson` integration for json generation (#1390)

This PR aims at integrating support of the `genson` package (in
`generate.json`) to be able to use dynamic json schema generation as
proposed in #1383.
---
 docs/reference/generation/json.md | 73 +++++++++++++++++++++++++++++++
 outlines/generate/json.py         |  6 +++
 pyproject.toml                    |  4 +-
 tests/generate/test_generate.py   | 16 +++++++
 4 files changed, 98 insertions(+), 1 deletion(-)

diff --git a/docs/reference/generation/json.md b/docs/reference/generation/json.md
index da9f14729..05dc69654 100644
--- a/docs/reference/generation/json.md
+++ b/docs/reference/generation/json.md
@@ -103,3 +103,76 @@ print(add(**result))
 ```
 
 A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
+
+
+## From a dynamic JSON schema builder - GenSON
+
+Outlines integrated [GenSON](https://github.com/wolverdude/GenSON) builders to be able to dynamicly declare JSON schemas. It can be used as follow:
+
+```python
+from genson import SchemaBuilder
+
+from outlines import models
+from outlines import generate
+
+builder = SchemaBuilder()
+builder.add_schema({"type": "object", "properties": {}})
+builder.add_object({"name": "Toto", "age": 5})
+
+model = models.transformers(
+    "HuggingFaceTB/SmolLM2-135M",
+    device="auto",
+)
+generator = generate.json(model, builder)
+
+res = generator("Return a json of a young boy")
+print(res)
+# {"name": "Ben", "age": 10}
+```
+
+Anytime you are updating the schema through the builder, you need to redifine the outline generator to include these changes. From the the previous example:
+
+```python
+from genson import SchemaBuilder
+
+from outlines import models
+from outlines import generate
+
+builder = SchemaBuilder()
+builder.add_schema({"type": "object", "properties": {}})
+builder.add_object({"name": "Toto", "age": 5})
+
+model = models.transformers(
+    "HuggingFaceTB/SmolLM2-135M",
+    device="auto",
+)
+generator = generate.json(model, builder)
+
+res = generator("Return a json of a young boy")
+print(res)
+# {"name": "Ben", "age": 10}
+
+builder.add_object({"hobby": "sports"})
+generator = generate.json(model, builder)
+
+res = generator("Return a json of a youg boy whose hobby is coding")
+print(res)
+# {"name": "Ben", "age": 10, "hobby": "coding"}
+```
+
+!!! Note
+
+    Beware of [GenSON](https://github.com/wolverdude/GenSON)'s behavior regarding dynamic amending of schemas through their builder. Here is an example of how you could lose `required` informations and generate json with missing fields:
+
+    ```python
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"name": "Toto", "age": 5})
+
+    print(builder.to_schema())
+    # {'$schema': 'http://json-schema.org/schema#', 'type': 'object', 'properties': {'name': {'type': 'string'}, 'age': {'type': 'integer'}}, 'required': ['age', 'name']}
+
+    builder.add_object({"hobby": "sport"})
+    print(builder.to_schema())
+    # {'name': {'type': 'string'}, 'age': {'type': 'integer'}, 'hobby': {'type': 'string'}}}
+    ```
diff --git a/outlines/generate/json.py b/outlines/generate/json.py
index d098d920d..008662206 100644
--- a/outlines/generate/json.py
+++ b/outlines/generate/json.py
@@ -3,6 +3,7 @@
 from functools import singledispatch
 from typing import Callable, Optional, Union
 
+from genson import SchemaBuilder
 from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel
 
@@ -55,6 +56,11 @@ def json(
         regex_str = build_regex_from_schema(schema, whitespace_pattern)
         generator = regex(model, regex_str, sampler)
         generator.format_sequence = lambda x: pyjson.loads(x)
+    elif isinstance(schema_object, SchemaBuilder):
+        schema = schema_object.to_json()
+        regex_str = build_regex_from_schema(schema, whitespace_pattern)
+        generator = regex(model, regex_str, sampler)
+        generator.format_sequence = lambda x: pyjson.loads(x)
     elif callable(schema_object):
         schema = pyjson.dumps(get_schema_from_signature(schema_object))
         regex_str = build_regex_from_schema(schema, whitespace_pattern)
diff --git a/pyproject.toml b/pyproject.toml
index 28060f797..0f1683d41 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -41,6 +41,7 @@ dependencies = [
    "airportsdata",
    "torch",
    "outlines_core==0.1.26",
+   "genson",
 ]
 dynamic = ["version"]
 
@@ -71,7 +72,7 @@ test = [
     "transformers",
     "pillow",
     "exllamav2",
-    "jax"
+    "jax",
 ]
 serve = [
     "vllm>=0.3.0",
@@ -147,6 +148,7 @@ module = [
     "pycountry.*",
     "airportsdata.*",
     "outlines_core.*",
+    "genson",
 ]
 ignore_missing_imports = true
 
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index f91bc8653..3d36faa5d 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -276,6 +276,22 @@ def test_generate_json(request, model_fixture, sample_schema):
     generator(**get_inputs(model_fixture), max_tokens=100)
 
 
+def test_integrate_genson_generate_json(request):
+    from genson import SchemaBuilder
+
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"name": "Toto", "age": 5})
+
+    model = request.getfixturevalue("model_transformers_opt125m")
+
+    generator = generate.json(model, builder)
+    res = generator("Return a json of a young boy")
+
+    assert "name" in res
+    assert "age" in res
+
+
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
 @pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
 def test_generate_choice(request, model_fixture, sample_choices):

From 1842fe784085deb1bde0c197685b9702b1920429 Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Wed, 5 Feb 2025 12:51:28 +0100
Subject: [PATCH 156/505] Add a few dotfiles to improve newcomers' experience
 (#1368)

---
 .editorconfig                                | 15 +++++++++++
 .gitignore                                   |  2 --
 .pre-commit-config.yaml                      | 27 +++++---------------
 .pydocstyle                                  |  2 ++
 .vscode/settings.json                        |  7 +++++
 outlines/fsm/types.py                        |  6 ++---
 outlines/processors/base_logits_processor.py |  4 +--
 pyproject.toml                               |  7 +++++
 8 files changed, 43 insertions(+), 27 deletions(-)
 create mode 100644 .editorconfig
 create mode 100644 .pydocstyle
 create mode 100644 .vscode/settings.json

diff --git a/.editorconfig b/.editorconfig
new file mode 100644
index 000000000..e795376f7
--- /dev/null
+++ b/.editorconfig
@@ -0,0 +1,15 @@
+# EditorConfig is awesome: https://EditorConfig.org
+
+# top-most EditorConfig file
+root = true
+
+[*]
+indent_style = space
+indent_size = 4
+end_of_line = lf
+charset = utf-8
+trim_trailing_whitespace = true
+insert_final_newline = true
+
+[*.yaml]
+indent_size = 2
diff --git a/.gitignore b/.gitignore
index eefa0cb13..4b1f5e921 100644
--- a/.gitignore
+++ b/.gitignore
@@ -15,5 +15,3 @@ benchmarks/results
 build
 docs/build
 logs
-logs/
-__pycache__/
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
index a9039b605..3c05ecb80 100644
--- a/.pre-commit-config.yaml
+++ b/.pre-commit-config.yaml
@@ -1,33 +1,20 @@
 repos:
 - repo: https://github.com/pre-commit/pre-commit-hooks
-  rev: v4.4.0
+  rev: v5.0.0
   hooks:
     -   id: check-merge-conflict
     -   id: debug-statements
     -   id: end-of-file-fixer
     -   id: trailing-whitespace
-- repo: https://github.com/pycqa/isort
-  rev: 5.12.0
-  hooks:
-    - id: isort
-      args: [--profile, black]
-- repo: https://github.com/asottile/pyupgrade
-  rev: v3.3.1
-  hooks:
-    - id: pyupgrade
-      args: [--py37-plus]
-- repo: https://github.com/pycqa/flake8
-  rev: 6.0.0
-  hooks:
-   - id: flake8
-- repo: https://github.com/psf/black
-  rev: 23.3.0
-  hooks:
-    - id: black
 - repo: https://github.com/pre-commit/mirrors-mypy
-  rev: v1.2.0
+  rev: v1.14.1
   hooks:
     - id: mypy
       args: [--allow-redefinition]
       exclude: ^examples/
       additional_dependencies: [types-tqdm, types-Pillow]
+- repo: https://github.com/astral-sh/ruff-pre-commit
+  rev: v0.9.1
+  hooks:
+    - id: ruff
+      args: ["--config=pyproject.toml"]
diff --git a/.pydocstyle b/.pydocstyle
new file mode 100644
index 000000000..6663458d8
--- /dev/null
+++ b/.pydocstyle
@@ -0,0 +1,2 @@
+[pydocstyle]
+convention = numpy
diff --git a/.vscode/settings.json b/.vscode/settings.json
new file mode 100644
index 000000000..a3a183836
--- /dev/null
+++ b/.vscode/settings.json
@@ -0,0 +1,7 @@
+{
+    "python.testing.pytestArgs": [
+        "tests"
+    ],
+    "python.testing.unittestEnabled": false,
+    "python.testing.pytestEnabled": true
+}
diff --git a/outlines/fsm/types.py b/outlines/fsm/types.py
index 5695dee07..cdb1d2115 100644
--- a/outlines/fsm/types.py
+++ b/outlines/fsm/types.py
@@ -39,19 +39,19 @@ def enum_format_fn(sequence: str) -> str:
 
         return enum_regex_str, enum_format_fn
 
-    if python_type == float:
+    if python_type is float:
 
         def float_format_fn(sequence: str) -> float:
             return float(sequence)
 
         return FLOAT, float_format_fn
-    elif python_type == int:
+    elif python_type is int:
 
         def int_format_fn(sequence: str) -> int:
             return int(sequence)
 
         return INTEGER, int_format_fn
-    elif python_type == bool:
+    elif python_type is bool:
 
         def bool_format_fn(sequence: str) -> bool:
             return bool(sequence)
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index 800e69f79..d6fe346e0 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -136,10 +136,10 @@ def _from_torch(tensor: torch.Tensor, target_type: Type) -> Array:
         elif target_type == np.ndarray:
             return tensor.detach().numpy()
 
-        elif target_type == list:
+        elif target_type is list:
             return tensor.detach().tolist()
 
-        elif target_type == tuple:
+        elif target_type is tuple:
             return tuple(tensor.detach().tolist())
 
         elif is_mlx_array_type(target_type):
diff --git a/pyproject.toml b/pyproject.toml
index 0f1683d41..5c45b0e9e 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -174,3 +174,10 @@ show_missing = true
 [tool.diff_cover]
 compare_branch = "origin/main"
 diff_range_notation = ".."
+
+[tool.docformatter]
+style = "numpy"
+in-place = true
+
+[tool.ruff.lint]
+ignore = [ "E731", "F401" ]

From 5d64b23e7f75357050f4ef30b9e2fcaa4d9b741d Mon Sep 17 00:00:00 2001
From: multi-stager <133181681+multi-stager@users.noreply.github.com>
Date: Tue, 4 Feb 2025 17:12:40 +0800
Subject: [PATCH 157/505] Update Dockerfile to Multi-Stage

---
 Dockerfile | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/Dockerfile b/Dockerfile
index 117e39e88..7e6f2e68a 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -1,4 +1,5 @@
-FROM python:3.10
+### Build stage
+FROM python:3.10 AS builder
 
 WORKDIR /outlines
 
@@ -13,5 +14,10 @@ COPY outlines ./outlines
 RUN --mount=source=.git,target=.git,type=bind \
     pip install --no-cache-dir .[serve]
 
+### Runtime stage
+FROM python:3.10
+WORKDIR /outlines
+COPY --from=builder /outlines /outlines
+
 # https://dottxt-ai.github.io/outlines/reference/vllm/
 ENTRYPOINT ["python3", "-m", "outlines.serve.serve"]

From 69418daf0c176bfbb2fe4c271c12359b7cb414f3 Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Wed, 22 Jan 2025 11:20:08 +0100
Subject: [PATCH 158/505] add DevContainer configuration file

---
 .devcontainer/devcontainer.json | 19 +++++++++++++++++++
 pyproject.toml                  |  2 +-
 2 files changed, 20 insertions(+), 1 deletion(-)
 create mode 100644 .devcontainer/devcontainer.json

diff --git a/.devcontainer/devcontainer.json b/.devcontainer/devcontainer.json
new file mode 100644
index 000000000..ae3e00d45
--- /dev/null
+++ b/.devcontainer/devcontainer.json
@@ -0,0 +1,19 @@
+{
+  "name": "dottxt-ai",
+  "image": "mcr.microsoft.com/devcontainers/python:3.12",
+  "runArgs": [
+    "--device=nvidia.com/gpu=all"
+  ],
+  "hostRequirements": {
+    "gpu": "optional"
+  },
+  "features": {
+    "ghcr.io/devcontainers/features/conda:1": {},
+    "ghcr.io/devcontainers/features/nvidia-cuda:1": {
+      "installCudnn": true,
+      "installToolkit": true,
+      "cudaVersion": "12.4"
+    },
+    "ghcr.io/devcontainers/features/rust:1": {}
+  }
+}
diff --git a/pyproject.toml b/pyproject.toml
index 5c45b0e9e..7c8b9fec3 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -6,7 +6,7 @@ build-backend = "setuptools.build_meta"
 name = "outlines"
 authors= [{name = "Outlines Developers"}]
 description = "Probabilistic Generative Model Programming"
-requires-python = ">=3.9"
+requires-python = ">=3.9,<3.13"
 license = {text = "Apache-2.0"}
 keywords=[
     "machine learning",

From 6c598c3cb94d6a1419a35700c0be0b275f70b22f Mon Sep 17 00:00:00 2001
From: aman-17 <amanrangapur@gmail.com>
Date: Tue, 11 Feb 2025 11:18:39 -0800
Subject: [PATCH 159/505] replaced pycountry with iso3166

---
 outlines/types/countries.py | 23 +++++++++++++++--------
 pyproject.toml              |  4 ++--
 2 files changed, 17 insertions(+), 10 deletions(-)

diff --git a/outlines/types/countries.py b/outlines/types/countries.py
index bbfc0dde7..56e68d0c2 100644
--- a/outlines/types/countries.py
+++ b/outlines/types/countries.py
@@ -1,19 +1,26 @@
 """Generate valid country codes and names."""
 from enum import Enum
 
-import pycountry
+from iso3166 import countries
 
-ALPHA_2_CODE = [(country.alpha_2, country.alpha_2) for country in pycountry.countries]
+
+def get_country_flags():
+    """Generate Unicode flags for all ISO 3166-1 alpha-2 country codes in Alpha2 Enum."""
+    base = ord('🇦')
+    return {code.name: chr(base + ord(code.name[0]) - ord('A')) + chr(base + ord(code.name[1]) - ord('A')) for code in Alpha2}
+
+ALPHA_2_CODE = [(country.alpha2, country.alpha2) for country in countries]
 Alpha2 = Enum("Alpha_2", ALPHA_2_CODE)  # type:ignore
 
-ALPHA_3_CODE = [(country.alpha_3, country.alpha_3) for country in pycountry.countries]
-Alpha3 = Enum("Alpha_2", ALPHA_3_CODE)  # type:ignore
+ALPHA_3_CODE = [(country.alpha3, country.alpha3) for country in countries]
+Alpha3 = Enum("Alpha_3", ALPHA_3_CODE)  # type:ignore
 
-NUMERIC_CODE = [(country.numeric, country.numeric) for country in pycountry.countries]
+NUMERIC_CODE = [(str(country.numeric), str(country.numeric)) for country in countries]
 Numeric = Enum("Numeric_code", NUMERIC_CODE)  # type:ignore
 
-NAME = [(country.name, country.name) for country in pycountry.countries]
+NAME = [(country.name, country.name) for country in countries]
 Name = Enum("Name", NAME)  # type:ignore
 
-FLAG = [(country.flag, country.flag) for country in pycountry.countries]
-Flag = Enum("Flag", FLAG)  # type:ignore
+flag_mapping = get_country_flags()
+FLAG = [(flag, flag) for code, flag in flag_mapping.items()]
+Flag = Enum("Flag", FLAG)   # type:ignore
\ No newline at end of file
diff --git a/pyproject.toml b/pyproject.toml
index 7c8b9fec3..e76f4f7b9 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -37,7 +37,7 @@ dependencies = [
    "requests",
    "tqdm",
    "typing_extensions",
-   "pycountry",
+   "iso3166",
    "airportsdata",
    "torch",
    "outlines_core==0.1.26",
@@ -145,7 +145,7 @@ module = [
     "vllm.*",
     "uvicorn.*",
     "fastapi.*",
-    "pycountry.*",
+    "iso3166.*",
     "airportsdata.*",
     "outlines_core.*",
     "genson",

From 774fe567e460ce1a1a3b03b7e6e501cb8dca5653 Mon Sep 17 00:00:00 2001
From: aman-17 <amanrangapur@gmail.com>
Date: Tue, 11 Feb 2025 12:26:55 -0800
Subject: [PATCH 160/505] fixed code style

---
 outlines/types/countries.py | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/outlines/types/countries.py b/outlines/types/countries.py
index 56e68d0c2..c612640b4 100644
--- a/outlines/types/countries.py
+++ b/outlines/types/countries.py
@@ -6,8 +6,13 @@
 
 def get_country_flags():
     """Generate Unicode flags for all ISO 3166-1 alpha-2 country codes in Alpha2 Enum."""
-    base = ord('🇦')
-    return {code.name: chr(base + ord(code.name[0]) - ord('A')) + chr(base + ord(code.name[1]) - ord('A')) for code in Alpha2}
+    base = ord("🇦")
+    return {
+        code.name: chr(base + ord(code.name[0]) - ord("A"))
+        + chr(base + ord(code.name[1]) - ord("A"))
+        for code in Alpha2
+    }
+
 
 ALPHA_2_CODE = [(country.alpha2, country.alpha2) for country in countries]
 Alpha2 = Enum("Alpha_2", ALPHA_2_CODE)  # type:ignore
@@ -23,4 +28,4 @@ def get_country_flags():
 
 flag_mapping = get_country_flags()
 FLAG = [(flag, flag) for code, flag in flag_mapping.items()]
-Flag = Enum("Flag", FLAG)   # type:ignore
\ No newline at end of file
+Flag = Enum("Flag", FLAG)  # type:ignore

From ad7a7403c1557f4e05293c665827ac55fd8df34d Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Fri, 14 Feb 2025 14:31:17 +0100
Subject: [PATCH 161/505] Update "Contribute" guide with `uv` and DevContainer
 intructions (#1405)

Also add instructions about different outlines "flavors"!

Co-authored-by: Cameron Pfiffer <cameron@pfiffer.org>
---
 docs/community/contribute.md | 76 ++++++++++++++++++++++++++++--------
 docs/installation.md         | 29 +++++++++-----
 2 files changed, 79 insertions(+), 26 deletions(-)

diff --git a/docs/community/contribute.md b/docs/community/contribute.md
index d29576b75..6867263b4 100644
--- a/docs/community/contribute.md
+++ b/docs/community/contribute.md
@@ -23,65 +23,108 @@ git clone git@github.com/YourUserName/outlines.git
 cd outlines
 ```
 
-Create a new virtual environment. *If you are using conda*:
+Create a new virtual environment:
+
+*If you are using `uv`*:
 
 ```bash
-conda env create -f environment.yml
+uv venv
+source .venv/bin/activate
+alias pip="uv pip" # ... or just remember to prepend any pip command with uv in the rest of this guide
 ```
 
-*If you are using venv*:
+*If you are using `venv`*:
 
-```python
+```bash
 python -m venv .venv
 source .venv/bin/activate
 ```
 
-Then install the dependencies in editable mode, and install the pre-commit hooks:
+*If you are using `conda`*:
+
+```bash
+conda env create -f environment.yml
+```
+
+Then install the dependencies in editable mode, and install the `pre-commit` hooks:
 
-```python
+```bash
 pip install -e ".[test]"
 pre-commit install
 ```
 
+Outlines provides optional dependencies for different supported backends, which you can install with
+
+```bash
+pip install ".[vllm]"
+```
+
+A list of supported optional dependencies can be found in the [installation guide](/installation).
+
+### Using VSCode DevContainer / GitHub Codespaces
+
+If you want a fully pre-configured development environment, you can use VSCode DevContainers or GitHub Codespaces.
+
+#### VSCode DevContainer
+
+1. Ensure that the [Docker](https://www.docker.com/get-started/) daemon is running on your machine.
+2. Install the [Dev Containers](https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.remote-containers) extension in VSCode.
+3. Open the Outlines repository in VSCode. When prompted, **Reopen in Container** (or press `F1` and select "Remote-Containers: Reopen in Container").
+4. Run the normal setup steps. Your environment will not complain about missing system dependencies!
+
+#### GitHub Codespaces
+
+1. Navigate to the Outlines repository on GitHub.
+2. Click on the **Code** button and select the **Codespaces** tab.
+3. Click **Create codespace on main** (or another branch you are working on).
+4. GitHub will launch a pre-configured cloud development environment.
+
+You will not have access to a GPU, but you'll be able to make basic contributions to the project on the go while using a fully featured web-based IDE.
+
 ### Before pushing your code
 
 Run the tests:
 
-```python
+```bash
 pytest
 ```
 
 And run the code style checks:
 
-```python
+```bash
 pre-commit run --all-files
 ```
 
 ### Benchmarking
 
-Outlines uses [asv](https://asv.readthedocs.io) for automated benchmark testing. Benchmarks are run automatically before pull requests are merged to prevent performance degredation.
+Outlines uses [asv](https://asv.readthedocs.io) for automated benchmark testing. Benchmarks are run automatically before pull requests are merged to prevent performance degradation.
 
 You can run the benchmark test suite locally with the following command:
-```
+
+```bash
 asv run --config benchmarks/asv.conf.json
 ```
 
 Caveats:
+
 - If you're on a device with CUDA, you must add the argument `--launch-method spawn`
 - Uncommitted code will not be benchmarked, you must first commit your changes.
 
 #### Run a specific test:
-```
+
+```bash
 asv run --config benchmarks/asv.conf.json -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
 ```
 
 #### Profile a specific test:
-```
+
+```bash
 asv run --config benchmarks/asv.conf.json --profile -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
 ```
 
 #### Compare to `origin/main`
-```
+
+```bash
 get fetch origin
 asv continuous origin/main HEAD --config benchmarks/asv.conf.json
 ```
@@ -91,20 +134,19 @@ asv continuous origin/main HEAD --config benchmarks/asv.conf.json
 - **View ASV Benchmark Results:** Open the workflow, view `BENCHMARK RESULTS` section.
 - Merging is blocked unless benchmarks are run for the latest commit.
 - Benchmarks fail if performance degrades by more than 10% for any individual benchmark.
-- The "Benchmark PR" workflow runs when its manually dispatched, or if the `run_benchmarks` label is added to the PR they run for every commit.
-
+- The "Benchmark PR" workflow runs when it is manually dispatched, or if the `run_benchmarks` label is added to the PR they run for every commit.
 
 ### Contribute to the documentation
 
 To work on the *documentation* you will need to install the related dependencies:
 
-```python
+```bash
 pip install -r requirements-doc.txt
 ```
 
 To build the documentation and serve it locally, run the following command in the repository's root folder:
 
-```python
+```bash
 mkdocs serve
 ```
 
diff --git a/docs/installation.md b/docs/installation.md
index bf2da86f9..5f8bb2d13 100644
--- a/docs/installation.md
+++ b/docs/installation.md
@@ -6,13 +6,13 @@ title: installation
 
 You can install Outlines with `pip`:
 
-```python
+```sh
 pip install outlines
 ```
 
-Outlines supports OpenAI, transformers, Mamba, llama.cpp and exllama2 but **you will need to install them manually**:
+Outlines supports OpenAI, Transformers, Mamba, llama.cpp, and ExLlamaV2, but **you will need to install them manually**:
 
-```python
+```sh
 pip install openai
 pip install transformers datasets accelerate torch
 pip install llama-cpp-python
@@ -21,18 +21,29 @@ pip install mamba_ssm transformers torch
 pip install vllm
 ```
 
-If you encounter any problem using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
+If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
 
-## Bleeding edge
+## Optional Dependencies
 
-You can install the latest version of Outlines on the repository's `main` branch:
+Outlines provides multiple optional dependency sets to support different backends and use cases. You can install them as needed using:
 
-```python
+- `pip install "outlines[vllm]"` for [vLLM](https://github.com/vllm-project/vllm), optimized for high-throughput inference.
+- `pip install "outlines[transformers]"` for [Hugging Face Transformers](https://huggingface.co/docs/transformers/index).
+- `pip install "outlines[mlx]"` for [MLX-LM](https://github.com/ml-explore/mlx-lm), optimized for Apple silicon.
+- `pip install "outlines[openai]"` to use OpenAI’s API.
+- `pip install "outlines[llamacpp]"` for [llama.cpp](https://github.com/ggerganov/llama.cpp), a lightweight LLM inference engine.
+- `pip install "outlines[exllamav2]"` for [ExLlamaV2](https://github.com/turboderp/exllamav2), optimized for NVIDIA GPUs.
+
+## Bleeding Edge
+
+You can install the latest version of Outlines from the repository's `main` branch:
+
+```sh
 pip install git+https://github.com/dottxt-ai/outlines.git@main
 ```
 
 This can be useful, for instance, when a fix has been merged but not yet released.
 
-## Installing for development
+## Installing for Development
 
-See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development.
+See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.

From ac49f578f046a1d6106ddcdf5b2276c45754286c Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Wed, 22 Jan 2025 11:16:39 +0100
Subject: [PATCH 162/505] Add nix configs and uv lock file

---
 flake.lock     |   59 +
 flake.nix      |   11 +
 pyproject.toml |    1 +
 shell.nix      |   73 +
 uv.lock        | 3979 ++++++++++++++++++++++++++++++++++++++++++++++++
 5 files changed, 4123 insertions(+)
 create mode 100644 flake.lock
 create mode 100644 flake.nix
 create mode 100644 shell.nix
 create mode 100644 uv.lock

diff --git a/flake.lock b/flake.lock
new file mode 100644
index 000000000..16d4c44a4
--- /dev/null
+++ b/flake.lock
@@ -0,0 +1,59 @@
+{
+  "nodes": {
+    "flake-utils": {
+      "inputs": {
+        "systems": "systems"
+      },
+      "locked": {
+        "lastModified": 1731533236,
+        "narHash": "sha256-l0KFg5HjrsfsO/JpG+r7fRrqm12kzFHyUHqHCVpMMbI=",
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "rev": "11707dc2f618dd54ca8739b309ec4fc024de578b",
+        "type": "github"
+      },
+      "original": {
+        "owner": "numtide",
+        "repo": "flake-utils",
+        "type": "github"
+      }
+    },
+    "nixpkgs": {
+      "locked": {
+        "lastModified": 1738136902,
+        "narHash": "sha256-pUvLijVGARw4u793APze3j6mU1Zwdtz7hGkGGkD87qw=",
+        "owner": "NixOS",
+        "repo": "nixpkgs",
+        "rev": "9a5db3142ce450045840cc8d832b13b8a2018e0c",
+        "type": "github"
+      },
+      "original": {
+        "id": "nixpkgs",
+        "type": "indirect"
+      }
+    },
+    "root": {
+      "inputs": {
+        "flake-utils": "flake-utils",
+        "nixpkgs": "nixpkgs"
+      }
+    },
+    "systems": {
+      "locked": {
+        "lastModified": 1681028828,
+        "narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
+        "owner": "nix-systems",
+        "repo": "default",
+        "rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
+        "type": "github"
+      },
+      "original": {
+        "owner": "nix-systems",
+        "repo": "default",
+        "type": "github"
+      }
+    }
+  },
+  "root": "root",
+  "version": 7
+}
diff --git a/flake.nix b/flake.nix
new file mode 100644
index 000000000..c82584b49
--- /dev/null
+++ b/flake.nix
@@ -0,0 +1,11 @@
+{
+  inputs.flake-utils.url = "github:numtide/flake-utils";
+  outputs = { self, nixpkgs, flake-utils }:
+    flake-utils.lib.eachDefaultSystem (system:
+      let
+        pkgs = import nixpkgs {
+          inherit system;
+          config.allowUnfree = true;
+        };
+      in { devShells.default = import ./shell.nix { inherit pkgs; }; });
+}
diff --git a/pyproject.toml b/pyproject.toml
index e76f4f7b9..f1ff57323 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -42,6 +42,7 @@ dependencies = [
    "torch",
    "outlines_core==0.1.26",
    "genson",
+   "pre-commit>=4.0.1",
 ]
 dynamic = ["version"]
 
diff --git a/shell.nix b/shell.nix
new file mode 100644
index 000000000..b67ed5b89
--- /dev/null
+++ b/shell.nix
@@ -0,0 +1,73 @@
+{ pkgs ? import <nixpkgs> { config = { allowUnfree = true; }; } }:
+
+(pkgs.buildFHSEnv {
+  name = "dottxt-ai";
+  targetPkgs = pkgs:
+    with pkgs; [
+      autoconf
+      binutils
+      cmake
+      cudatoolkit
+      curl
+      freeglut
+      gcc13
+      git
+      gitRepo
+      gnumake
+      gnupg
+      gperf
+      libGL
+      libGLU
+      linuxPackages.nvidia_x11
+      m4
+      ncurses5
+      procps
+      python311
+      stdenv.cc
+      unzip
+      util-linux
+      uv
+      xorg.libX11
+      xorg.libXext
+      xorg.libXi
+      xorg.libXmu
+      xorg.libXrandr
+      xorg.libXv
+      zlib
+    ];
+
+  multiPkgs = pkgs: with pkgs; [ zlib ];
+
+  runScript = "bash";
+
+  profile = ''
+    # CUDA paths
+    export CUDA_HOME=${pkgs.cudatoolkit}
+    export CUDA_PATH=${pkgs.cudatoolkit}
+
+    # Ensure proper binary paths are included
+    export PATH=${pkgs.gcc13}/bin:${pkgs.cudatoolkit}/bin:$PATH
+
+    # Set library paths, including additional directories for CUPTI
+    export LD_LIBRARY_PATH=${pkgs.cudatoolkit}/lib64:${pkgs.cudatoolkit}/extras/CUPTI/lib64:${pkgs.linuxPackages.nvidia_x11}/lib:$LD_LIBRARY_PATH
+
+    # Add static library paths to EXTRA_LDFLAGS for the linker
+    export EXTRA_LDFLAGS="-L${pkgs.cudatoolkit}/lib64 -L${pkgs.cudatoolkit}/extras/CUPTI/lib64 -L${pkgs.linuxPackages.nvidia_x11}/lib -L${pkgs.cudatoolkit}/libdevice $EXTRA_LDFLAGS"
+    export EXTRA_CCFLAGS="-I${pkgs.cudatoolkit}/include $EXTRA_CCFLAGS"
+
+    # Set CMake paths
+    export CMAKE_PREFIX_PATH=${pkgs.cudatoolkit}:${pkgs.linuxPackages.nvidia_x11}:$CMAKE_PREFIX_PATH
+
+    # C++ and CC flags
+    export CXXFLAGS="--std=c++17 $EXTRA_CCFLAGS"
+    export CC=${pkgs.gcc13}/bin/gcc
+    export CXX=${pkgs.gcc13}/bin/g++
+
+    # NVCC flags to use the right compiler
+    export NVCC_FLAGS="-ccbin ${pkgs.gcc13}/bin/gcc"
+  '';
+
+  structuredAttrs__ = {
+    stdenv = pkgs.stdenv.overrideCC pkgs.stdenv.cc pkgs.gcc13;
+  };
+}).env
diff --git a/uv.lock b/uv.lock
new file mode 100644
index 000000000..f1deefbe6
--- /dev/null
+++ b/uv.lock
@@ -0,0 +1,3979 @@
+version = 1
+requires-python = ">=3.9"
+resolution-markers = [
+    "python_full_version >= '3.13'",
+    "python_full_version == '3.12.*'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
+    "python_full_version < '3.9.17'",
+]
+
+[[package]]
+name = "accelerate"
+version = "1.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "huggingface-hub" },
+    { name = "numpy" },
+    { name = "packaging" },
+    { name = "psutil" },
+    { name = "pyyaml" },
+    { name = "safetensors" },
+    { name = "torch" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/85/15/0fab0260ab4069e5224e637d2e400538bb27b0dfc36f17daf68db9770d78/accelerate-1.3.0.tar.gz", hash = "sha256:518631c0adb80bd3d42fb29e7e2dc2256bcd7c786b0ba9119bbaa08611b36d9c", size = 342758 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/73/de/64508cb91af013aaba214752309c0967568a4219d50a4ea30e822af3c976/accelerate-1.3.0-py3-none-any.whl", hash = "sha256:5788d9e6a7a9f80fed665cf09681c4dddd9dc056bea656db4140ffc285ce423e", size = 336647 },
+]
+
+[[package]]
+name = "aiohappyeyeballs"
+version = "2.4.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7f/55/e4373e888fdacb15563ef6fa9fa8c8252476ea071e96fb46defac9f18bf2/aiohappyeyeballs-2.4.4.tar.gz", hash = "sha256:5fdd7d87889c63183afc18ce9271f9b0a7d32c2303e394468dd45d514a757745", size = 21977 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b9/74/fbb6559de3607b3300b9be3cc64e97548d55678e44623db17820dbd20002/aiohappyeyeballs-2.4.4-py3-none-any.whl", hash = "sha256:a980909d50efcd44795c4afeca523296716d50cd756ddca6af8c65b996e27de8", size = 14756 },
+]
+
+[[package]]
+name = "aiohttp"
+version = "3.11.11"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "aiohappyeyeballs" },
+    { name = "aiosignal" },
+    { name = "async-timeout", marker = "python_full_version < '3.11'" },
+    { name = "attrs" },
+    { name = "frozenlist" },
+    { name = "multidict" },
+    { name = "propcache" },
+    { name = "yarl" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fe/ed/f26db39d29cd3cb2f5a3374304c713fe5ab5a0e4c8ee25a0c45cc6adf844/aiohttp-3.11.11.tar.gz", hash = "sha256:bb49c7f1e6ebf3821a42d81d494f538107610c3a705987f53068546b0e90303e", size = 7669618 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/75/7d/ff2e314b8f9e0b1df833e2d4778eaf23eae6b8cc8f922495d110ddcbf9e1/aiohttp-3.11.11-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:a60804bff28662cbcf340a4d61598891f12eea3a66af48ecfdc975ceec21e3c8", size = 708550 },
+    { url = "https://files.pythonhosted.org/packages/09/b8/aeb4975d5bba233d6f246941f5957a5ad4e3def8b0855a72742e391925f2/aiohttp-3.11.11-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:4b4fa1cb5f270fb3eab079536b764ad740bb749ce69a94d4ec30ceee1b5940d5", size = 468430 },
+    { url = "https://files.pythonhosted.org/packages/9c/5b/5b620279b3df46e597008b09fa1e10027a39467387c2332657288e25811a/aiohttp-3.11.11-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:731468f555656767cda219ab42e033355fe48c85fbe3ba83a349631541715ba2", size = 455593 },
+    { url = "https://files.pythonhosted.org/packages/d8/75/0cdf014b816867d86c0bc26f3d3e3f194198dbf33037890beed629cd4f8f/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cb23d8bb86282b342481cad4370ea0853a39e4a32a0042bb52ca6bdde132df43", size = 1584635 },
+    { url = "https://files.pythonhosted.org/packages/df/2f/95b8f4e4dfeb57c1d9ad9fa911ede35a0249d75aa339edd2c2270dc539da/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f047569d655f81cb70ea5be942ee5d4421b6219c3f05d131f64088c73bb0917f", size = 1632363 },
+    { url = "https://files.pythonhosted.org/packages/39/cb/70cf69ea7c50f5b0021a84f4c59c3622b2b3b81695f48a2f0e42ef7eba6e/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:dd7659baae9ccf94ae5fe8bfaa2c7bc2e94d24611528395ce88d009107e00c6d", size = 1668315 },
+    { url = "https://files.pythonhosted.org/packages/2f/cc/3a3fc7a290eabc59839a7e15289cd48f33dd9337d06e301064e1e7fb26c5/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:af01e42ad87ae24932138f154105e88da13ce7d202a6de93fafdafb2883a00ef", size = 1589546 },
+    { url = "https://files.pythonhosted.org/packages/15/b4/0f7b0ed41ac6000e283e7332f0f608d734b675a8509763ca78e93714cfb0/aiohttp-3.11.11-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5854be2f3e5a729800bac57a8d76af464e160f19676ab6aea74bde18ad19d438", size = 1544581 },
+    { url = "https://files.pythonhosted.org/packages/58/b9/4d06470fd85c687b6b0e31935ef73dde6e31767c9576d617309a2206556f/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:6526e5fb4e14f4bbf30411216780c9967c20c5a55f2f51d3abd6de68320cc2f3", size = 1529256 },
+    { url = "https://files.pythonhosted.org/packages/61/a2/6958b1b880fc017fd35f5dfb2c26a9a50c755b75fd9ae001dc2236a4fb79/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:85992ee30a31835fc482468637b3e5bd085fa8fe9392ba0bdcbdc1ef5e9e3c55", size = 1536592 },
+    { url = "https://files.pythonhosted.org/packages/0f/dd/b974012a9551fd654f5bb95a6dd3f03d6e6472a17e1a8216dd42e9638d6c/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:88a12ad8ccf325a8a5ed80e6d7c3bdc247d66175afedbe104ee2aaca72960d8e", size = 1607446 },
+    { url = "https://files.pythonhosted.org/packages/e0/d3/6c98fd87e638e51f074a3f2061e81fcb92123bcaf1439ac1b4a896446e40/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:0a6d3fbf2232e3a08c41eca81ae4f1dff3d8f1a30bae415ebe0af2d2458b8a33", size = 1628809 },
+    { url = "https://files.pythonhosted.org/packages/a8/2e/86e6f85cbca02be042c268c3d93e7f35977a0e127de56e319bdd1569eaa8/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:84a585799c58b795573c7fa9b84c455adf3e1d72f19a2bf498b54a95ae0d194c", size = 1564291 },
+    { url = "https://files.pythonhosted.org/packages/0b/8d/1f4ef3503b767717f65e1f5178b0173ab03cba1a19997ebf7b052161189f/aiohttp-3.11.11-cp310-cp310-win32.whl", hash = "sha256:bfde76a8f430cf5c5584553adf9926534352251d379dcb266ad2b93c54a29745", size = 416601 },
+    { url = "https://files.pythonhosted.org/packages/ad/86/81cb83691b5ace3d9aa148dc42bacc3450d749fc88c5ec1973573c1c1779/aiohttp-3.11.11-cp310-cp310-win_amd64.whl", hash = "sha256:0fd82b8e9c383af11d2b26f27a478640b6b83d669440c0a71481f7c865a51da9", size = 442007 },
+    { url = "https://files.pythonhosted.org/packages/34/ae/e8806a9f054e15f1d18b04db75c23ec38ec954a10c0a68d3bd275d7e8be3/aiohttp-3.11.11-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:ba74ec819177af1ef7f59063c6d35a214a8fde6f987f7661f4f0eecc468a8f76", size = 708624 },
+    { url = "https://files.pythonhosted.org/packages/c7/e0/313ef1a333fb4d58d0c55a6acb3cd772f5d7756604b455181049e222c020/aiohttp-3.11.11-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4af57160800b7a815f3fe0eba9b46bf28aafc195555f1824555fa2cfab6c1538", size = 468507 },
+    { url = "https://files.pythonhosted.org/packages/a9/60/03455476bf1f467e5b4a32a465c450548b2ce724eec39d69f737191f936a/aiohttp-3.11.11-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ffa336210cf9cd8ed117011085817d00abe4c08f99968deef0013ea283547204", size = 455571 },
+    { url = "https://files.pythonhosted.org/packages/be/f9/469588603bd75bf02c8ffb8c8a0d4b217eed446b49d4a767684685aa33fd/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81b8fe282183e4a3c7a1b72f5ade1094ed1c6345a8f153506d114af5bf8accd9", size = 1685694 },
+    { url = "https://files.pythonhosted.org/packages/88/b9/1b7fa43faf6c8616fa94c568dc1309ffee2b6b68b04ac268e5d64b738688/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3af41686ccec6a0f2bdc66686dc0f403c41ac2089f80e2214a0f82d001052c03", size = 1743660 },
+    { url = "https://files.pythonhosted.org/packages/2a/8b/0248d19dbb16b67222e75f6aecedd014656225733157e5afaf6a6a07e2e8/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:70d1f9dde0e5dd9e292a6d4d00058737052b01f3532f69c0c65818dac26dc287", size = 1785421 },
+    { url = "https://files.pythonhosted.org/packages/c4/11/f478e071815a46ca0a5ae974651ff0c7a35898c55063305a896e58aa1247/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:249cc6912405917344192b9f9ea5cd5b139d49e0d2f5c7f70bdfaf6b4dbf3a2e", size = 1675145 },
+    { url = "https://files.pythonhosted.org/packages/26/5d/284d182fecbb5075ae10153ff7374f57314c93a8681666600e3a9e09c505/aiohttp-3.11.11-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0eb98d90b6690827dcc84c246811feeb4e1eea683c0eac6caed7549be9c84665", size = 1619804 },
+    { url = "https://files.pythonhosted.org/packages/1b/78/980064c2ad685c64ce0e8aeeb7ef1e53f43c5b005edcd7d32e60809c4992/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ec82bf1fda6cecce7f7b915f9196601a1bd1a3079796b76d16ae4cce6d0ef89b", size = 1654007 },
+    { url = "https://files.pythonhosted.org/packages/21/8d/9e658d63b1438ad42b96f94da227f2e2c1d5c6001c9e8ffcc0bfb22e9105/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:9fd46ce0845cfe28f108888b3ab17abff84ff695e01e73657eec3f96d72eef34", size = 1650022 },
+    { url = "https://files.pythonhosted.org/packages/85/fd/a032bf7f2755c2df4f87f9effa34ccc1ef5cea465377dbaeef93bb56bbd6/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:bd176afcf8f5d2aed50c3647d4925d0db0579d96f75a31e77cbaf67d8a87742d", size = 1732899 },
+    { url = "https://files.pythonhosted.org/packages/c5/0c/c2b85fde167dd440c7ba50af2aac20b5a5666392b174df54c00f888c5a75/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:ec2aa89305006fba9ffb98970db6c8221541be7bee4c1d027421d6f6df7d1ce2", size = 1755142 },
+    { url = "https://files.pythonhosted.org/packages/bc/78/91ae1a3b3b3bed8b893c5d69c07023e151b1c95d79544ad04cf68f596c2f/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:92cde43018a2e17d48bb09c79e4d4cb0e236de5063ce897a5e40ac7cb4878773", size = 1692736 },
+    { url = "https://files.pythonhosted.org/packages/77/89/a7ef9c4b4cdb546fcc650ca7f7395aaffbd267f0e1f648a436bec33c9b95/aiohttp-3.11.11-cp311-cp311-win32.whl", hash = "sha256:aba807f9569455cba566882c8938f1a549f205ee43c27b126e5450dc9f83cc62", size = 416418 },
+    { url = "https://files.pythonhosted.org/packages/fc/db/2192489a8a51b52e06627506f8ac8df69ee221de88ab9bdea77aa793aa6a/aiohttp-3.11.11-cp311-cp311-win_amd64.whl", hash = "sha256:ae545f31489548c87b0cced5755cfe5a5308d00407000e72c4fa30b19c3220ac", size = 442509 },
+    { url = "https://files.pythonhosted.org/packages/69/cf/4bda538c502f9738d6b95ada11603c05ec260807246e15e869fc3ec5de97/aiohttp-3.11.11-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:e595c591a48bbc295ebf47cb91aebf9bd32f3ff76749ecf282ea7f9f6bb73886", size = 704666 },
+    { url = "https://files.pythonhosted.org/packages/46/7b/87fcef2cad2fad420ca77bef981e815df6904047d0a1bd6aeded1b0d1d66/aiohttp-3.11.11-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:3ea1b59dc06396b0b424740a10a0a63974c725b1c64736ff788a3689d36c02d2", size = 464057 },
+    { url = "https://files.pythonhosted.org/packages/5a/a6/789e1f17a1b6f4a38939fbc39d29e1d960d5f89f73d0629a939410171bc0/aiohttp-3.11.11-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8811f3f098a78ffa16e0ea36dffd577eb031aea797cbdba81be039a4169e242c", size = 455996 },
+    { url = "https://files.pythonhosted.org/packages/b7/dd/485061fbfef33165ce7320db36e530cd7116ee1098e9c3774d15a732b3fd/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bd7227b87a355ce1f4bf83bfae4399b1f5bb42e0259cb9405824bd03d2f4336a", size = 1682367 },
+    { url = "https://files.pythonhosted.org/packages/e9/d7/9ec5b3ea9ae215c311d88b2093e8da17e67b8856673e4166c994e117ee3e/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d40f9da8cabbf295d3a9dae1295c69975b86d941bc20f0a087f0477fa0a66231", size = 1736989 },
+    { url = "https://files.pythonhosted.org/packages/d6/fb/ea94927f7bfe1d86178c9d3e0a8c54f651a0a655214cce930b3c679b8f64/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ffb3dc385f6bb1568aa974fe65da84723210e5d9707e360e9ecb51f59406cd2e", size = 1793265 },
+    { url = "https://files.pythonhosted.org/packages/40/7f/6de218084f9b653026bd7063cd8045123a7ba90c25176465f266976d8c82/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a8f5f7515f3552d899c61202d99dcb17d6e3b0de777900405611cd747cecd1b8", size = 1691841 },
+    { url = "https://files.pythonhosted.org/packages/77/e2/992f43d87831cbddb6b09c57ab55499332f60ad6fdbf438ff4419c2925fc/aiohttp-3.11.11-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3499c7ffbfd9c6a3d8d6a2b01c26639da7e43d47c7b4f788016226b1e711caa8", size = 1619317 },
+    { url = "https://files.pythonhosted.org/packages/96/74/879b23cdd816db4133325a201287c95bef4ce669acde37f8f1b8669e1755/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:8e2bf8029dbf0810c7bfbc3e594b51c4cc9101fbffb583a3923aea184724203c", size = 1641416 },
+    { url = "https://files.pythonhosted.org/packages/30/98/b123f6b15d87c54e58fd7ae3558ff594f898d7f30a90899718f3215ad328/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:b6212a60e5c482ef90f2d788835387070a88d52cf6241d3916733c9176d39eab", size = 1646514 },
+    { url = "https://files.pythonhosted.org/packages/d7/38/257fda3dc99d6978ab943141d5165ec74fd4b4164baa15e9c66fa21da86b/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:d119fafe7b634dbfa25a8c597718e69a930e4847f0b88e172744be24515140da", size = 1702095 },
+    { url = "https://files.pythonhosted.org/packages/0c/f4/ddab089053f9fb96654df5505c0a69bde093214b3c3454f6bfdb1845f558/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:6fba278063559acc730abf49845d0e9a9e1ba74f85f0ee6efd5803f08b285853", size = 1734611 },
+    { url = "https://files.pythonhosted.org/packages/c3/d6/f30b2bc520c38c8aa4657ed953186e535ae84abe55c08d0f70acd72ff577/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:92fc484e34b733704ad77210c7957679c5c3877bd1e6b6d74b185e9320cc716e", size = 1694576 },
+    { url = "https://files.pythonhosted.org/packages/bc/97/b0a88c3f4c6d0020b34045ee6d954058abc870814f6e310c4c9b74254116/aiohttp-3.11.11-cp312-cp312-win32.whl", hash = "sha256:9f5b3c1ed63c8fa937a920b6c1bec78b74ee09593b3f5b979ab2ae5ef60d7600", size = 411363 },
+    { url = "https://files.pythonhosted.org/packages/7f/23/cc36d9c398980acaeeb443100f0216f50a7cfe20c67a9fd0a2f1a5a846de/aiohttp-3.11.11-cp312-cp312-win_amd64.whl", hash = "sha256:1e69966ea6ef0c14ee53ef7a3d68b564cc408121ea56c0caa2dc918c1b2f553d", size = 437666 },
+    { url = "https://files.pythonhosted.org/packages/49/d1/d8af164f400bad432b63e1ac857d74a09311a8334b0481f2f64b158b50eb/aiohttp-3.11.11-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:541d823548ab69d13d23730a06f97460f4238ad2e5ed966aaf850d7c369782d9", size = 697982 },
+    { url = "https://files.pythonhosted.org/packages/92/d1/faad3bf9fa4bfd26b95c69fc2e98937d52b1ff44f7e28131855a98d23a17/aiohttp-3.11.11-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:929f3ed33743a49ab127c58c3e0a827de0664bfcda566108989a14068f820194", size = 460662 },
+    { url = "https://files.pythonhosted.org/packages/db/61/0d71cc66d63909dabc4590f74eba71f91873a77ea52424401c2498d47536/aiohttp-3.11.11-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:0882c2820fd0132240edbb4a51eb8ceb6eef8181db9ad5291ab3332e0d71df5f", size = 452950 },
+    { url = "https://files.pythonhosted.org/packages/07/db/6d04bc7fd92784900704e16b745484ef45b77bd04e25f58f6febaadf7983/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b63de12e44935d5aca7ed7ed98a255a11e5cb47f83a9fded7a5e41c40277d104", size = 1665178 },
+    { url = "https://files.pythonhosted.org/packages/54/5c/e95ade9ae29f375411884d9fd98e50535bf9fe316c9feb0f30cd2ac8f508/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:aa54f8ef31d23c506910c21163f22b124facb573bff73930735cf9fe38bf7dff", size = 1717939 },
+    { url = "https://files.pythonhosted.org/packages/6f/1c/1e7d5c5daea9e409ed70f7986001b8c9e3a49a50b28404498d30860edab6/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a344d5dc18074e3872777b62f5f7d584ae4344cd6006c17ba12103759d407af3", size = 1775125 },
+    { url = "https://files.pythonhosted.org/packages/5d/66/890987e44f7d2f33a130e37e01a164168e6aff06fce15217b6eaf14df4f6/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0b7fb429ab1aafa1f48578eb315ca45bd46e9c37de11fe45c7f5f4138091e2f1", size = 1677176 },
+    { url = "https://files.pythonhosted.org/packages/8f/dc/e2ba57d7a52df6cdf1072fd5fa9c6301a68e1cd67415f189805d3eeb031d/aiohttp-3.11.11-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c341c7d868750e31961d6d8e60ff040fb9d3d3a46d77fd85e1ab8e76c3e9a5c4", size = 1603192 },
+    { url = "https://files.pythonhosted.org/packages/6c/9e/8d08a57de79ca3a358da449405555e668f2c8871a7777ecd2f0e3912c272/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ed9ee95614a71e87f1a70bc81603f6c6760128b140bc4030abe6abaa988f1c3d", size = 1618296 },
+    { url = "https://files.pythonhosted.org/packages/56/51/89822e3ec72db352c32e7fc1c690370e24e231837d9abd056490f3a49886/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:de8d38f1c2810fa2a4f1d995a2e9c70bb8737b18da04ac2afbf3971f65781d87", size = 1616524 },
+    { url = "https://files.pythonhosted.org/packages/2c/fa/e2e6d9398f462ffaa095e84717c1732916a57f1814502929ed67dd7568ef/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:a9b7371665d4f00deb8f32208c7c5e652059b0fda41cf6dbcac6114a041f1cc2", size = 1685471 },
+    { url = "https://files.pythonhosted.org/packages/ae/5f/6bb976e619ca28a052e2c0ca7b0251ccd893f93d7c24a96abea38e332bf6/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:620598717fce1b3bd14dd09947ea53e1ad510317c85dda2c9c65b622edc96b12", size = 1715312 },
+    { url = "https://files.pythonhosted.org/packages/79/c1/756a7e65aa087c7fac724d6c4c038f2faaa2a42fe56dbc1dd62a33ca7213/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:bf8d9bfee991d8acc72d060d53860f356e07a50f0e0d09a8dfedea1c554dd0d5", size = 1672783 },
+    { url = "https://files.pythonhosted.org/packages/73/ba/a6190ebb02176c7f75e6308da31f5d49f6477b651a3dcfaaaca865a298e2/aiohttp-3.11.11-cp313-cp313-win32.whl", hash = "sha256:9d73ee3725b7a737ad86c2eac5c57a4a97793d9f442599bea5ec67ac9f4bdc3d", size = 410229 },
+    { url = "https://files.pythonhosted.org/packages/b8/62/c9fa5bafe03186a0e4699150a7fed9b1e73240996d0d2f0e5f70f3fdf471/aiohttp-3.11.11-cp313-cp313-win_amd64.whl", hash = "sha256:c7a06301c2fb096bdb0bd25fe2011531c1453b9f2c163c8031600ec73af1cc99", size = 436081 },
+    { url = "https://files.pythonhosted.org/packages/9f/37/326ee86b7640be6ca4493c8121cb9a4386e07cf1e5757ce6b7fa854d0a5f/aiohttp-3.11.11-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:3e23419d832d969f659c208557de4a123e30a10d26e1e14b73431d3c13444c2e", size = 709424 },
+    { url = "https://files.pythonhosted.org/packages/9c/c5/a88ec2160b06c22e57e483a1f78f99f005fcd4e7d6855a2d3d6510881b65/aiohttp-3.11.11-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:21fef42317cf02e05d3b09c028712e1d73a9606f02467fd803f7c1f39cc59add", size = 468907 },
+    { url = "https://files.pythonhosted.org/packages/b2/f0/02f03f818e91996161cce200241b631bb2b4a87e61acddb5b974e254a288/aiohttp-3.11.11-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1f21bb8d0235fc10c09ce1d11ffbd40fc50d3f08a89e4cf3a0c503dc2562247a", size = 455981 },
+    { url = "https://files.pythonhosted.org/packages/0e/17/c8be12436ec19915f67b1ab8240d4105aba0f7e0894a1f0d8939c3e79c70/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1642eceeaa5ab6c9b6dfeaaa626ae314d808188ab23ae196a34c9d97efb68350", size = 1587395 },
+    { url = "https://files.pythonhosted.org/packages/43/c0/f4db1ac30ebe855b2fefd6fa98767862d88ac54ab08a6ad07d619146270c/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:2170816e34e10f2fd120f603e951630f8a112e1be3b60963a1f159f5699059a6", size = 1636243 },
+    { url = "https://files.pythonhosted.org/packages/ea/a7/9acf20e9a09b0d38b5b55691410500d051a9f4194692cac22b0d0fc92ad9/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8be8508d110d93061197fd2d6a74f7401f73b6d12f8822bbcd6d74f2b55d71b1", size = 1672323 },
+    { url = "https://files.pythonhosted.org/packages/f7/5b/a27e8fe1a3b0e245ca80863eefd83fc00136752d27d2cf1afa0130a76f34/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4eed954b161e6b9b65f6be446ed448ed3921763cc432053ceb606f89d793927e", size = 1589521 },
+    { url = "https://files.pythonhosted.org/packages/25/50/8bccd08004e15906791b46f0a908a8e7f5e0c5882b17da96d1933bd34ac0/aiohttp-3.11.11-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d6c9af134da4bc9b3bd3e6a70072509f295d10ee60c697826225b60b9959acdd", size = 1544059 },
+    { url = "https://files.pythonhosted.org/packages/84/5a/42250b37b06ee0cb7a03dd1630243b1d739ca3edb5abd8b18f479a539900/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:44167fc6a763d534a6908bdb2592269b4bf30a03239bcb1654781adf5e49caf1", size = 1530217 },
+    { url = "https://files.pythonhosted.org/packages/18/08/eb334da86cd2cdbd0621bb7039255b19ca74ce8b05e8fb61850e2589938c/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:479b8c6ebd12aedfe64563b85920525d05d394b85f166b7873c8bde6da612f9c", size = 1536081 },
+    { url = "https://files.pythonhosted.org/packages/1a/a9/9d59958084d5bad7e77a44841013bd59768cda94f9f744769461b66038fc/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:10b4ff0ad793d98605958089fabfa350e8e62bd5d40aa65cdc69d6785859f94e", size = 1606918 },
+    { url = "https://files.pythonhosted.org/packages/4f/e7/27feb1cff17dcddb7a5b703199106196718d622a3aa70f80a386d15361d7/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:b540bd67cfb54e6f0865ceccd9979687210d7ed1a1cc8c01f8e67e2f1e883d28", size = 1629101 },
+    { url = "https://files.pythonhosted.org/packages/e8/29/49debcd858b997c655fca274c5247fcfe29bf31a4ddb1ce3f088539b14e4/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1dac54e8ce2ed83b1f6b1a54005c87dfed139cf3f777fdc8afc76e7841101226", size = 1567338 },
+    { url = "https://files.pythonhosted.org/packages/3b/34/33af1e97aba1862e1812e2e2b96a1e050c5a6e9cecd5a5370591122fb07b/aiohttp-3.11.11-cp39-cp39-win32.whl", hash = "sha256:568c1236b2fde93b7720f95a890741854c1200fba4a3471ff48b2934d2d93fd3", size = 416914 },
+    { url = "https://files.pythonhosted.org/packages/2d/47/28b3fbd97026963af2774423c64341e0d4ec180ea3b79a2762a3c18d5d94/aiohttp-3.11.11-cp39-cp39-win_amd64.whl", hash = "sha256:943a8b052e54dfd6439fd7989f67fc6a7f2138d0a2cf0a7de5f18aa4fe7eb3b1", size = 442225 },
+]
+
+[[package]]
+name = "aiosignal"
+version = "1.3.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "frozenlist" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ba/b5/6d55e80f6d8a08ce22b982eafa278d823b541c925f11ee774b0b9c43473d/aiosignal-1.3.2.tar.gz", hash = "sha256:a8c255c66fafb1e499c9351d0bf32ff2d8a0321595ebac3b93713656d2436f54", size = 19424 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ec/6a/bc7e17a3e87a2985d3e8f4da4cd0f481060eb78fb08596c42be62c90a4d9/aiosignal-1.3.2-py2.py3-none-any.whl", hash = "sha256:45cde58e409a301715980c2b01d0c28bdde3770d8290b5eb2173759d9acb31a5", size = 7597 },
+]
+
+[[package]]
+name = "airportsdata"
+version = "20241001"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ff/75/0af9b73babfec14df1395ca60e2a174e887c5d515e823c39a60f27bbbb30/airportsdata-20241001.tar.gz", hash = "sha256:fa0bd143b4f4be3557cb892fa0612ef210fd91a92bd720b4d8221de576a4fa00", size = 903075 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/af/57/57635b0af4a23723b2b43109aa4eae7395d346d1eed8a00691cf3acfc5bd/airportsdata-20241001-py3-none-any.whl", hash = "sha256:67d71cf2c5378cc17ff66b62b1e11aa2444043949c894543ac8fd8dafce192fd", size = 912677 },
+]
+
+[[package]]
+name = "annotated-types"
+version = "0.7.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ee/67/531ea369ba64dcff5ec9c3402f9f51bf748cec26dde048a2f973a4eea7f5/annotated_types-0.7.0.tar.gz", hash = "sha256:aff07c09a53a08bc8cfccb9c85b05f1aa9a2a6f23728d790723543408344ce89", size = 16081 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/78/b6/6307fbef88d9b5ee7421e68d78a9f162e0da4900bc5f5793f6d3d0e34fb8/annotated_types-0.7.0-py3-none-any.whl", hash = "sha256:1f02e8b43a8fbbc3f3e0d4f0f4bfc8131bcb4eebe8849b8e5c773f3a1c582a53", size = 13643 },
+]
+
+[[package]]
+name = "anyio"
+version = "4.8.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "exceptiongroup", marker = "python_full_version < '3.11'" },
+    { name = "idna" },
+    { name = "sniffio" },
+    { name = "typing-extensions", marker = "python_full_version < '3.13'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a3/73/199a98fc2dae33535d6b8e8e6ec01f8c1d76c9adb096c6b7d64823038cde/anyio-4.8.0.tar.gz", hash = "sha256:1d9fe889df5212298c0c0723fa20479d1b94883a2df44bd3897aa91083316f7a", size = 181126 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/46/eb/e7f063ad1fec6b3178a3cd82d1a3c4de82cccf283fc42746168188e1cdd5/anyio-4.8.0-py3-none-any.whl", hash = "sha256:b5011f270ab5eb0abf13385f851315585cc37ef330dd88e27ec3d34d651fd47a", size = 96041 },
+]
+
+[[package]]
+name = "async-timeout"
+version = "5.0.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/a5/ae/136395dfbfe00dfc94da3f3e136d0b13f394cba8f4841120e34226265780/async_timeout-5.0.1.tar.gz", hash = "sha256:d9321a7a3d5a6a5e187e824d2fa0793ce379a202935782d555d6e9d2735677d3", size = 9274 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fe/ba/e2081de779ca30d473f21f5b30e0e737c438205440784c7dfc81efc2b029/async_timeout-5.0.1-py3-none-any.whl", hash = "sha256:39e3809566ff85354557ec2398b55e096c8364bacac9405a7a1fa429e77fe76c", size = 6233 },
+]
+
+[[package]]
+name = "attrs"
+version = "24.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/48/c8/6260f8ccc11f0917360fc0da435c5c9c7504e3db174d5a12a1494887b045/attrs-24.3.0.tar.gz", hash = "sha256:8f5c07333d543103541ba7be0e2ce16eeee8130cb0b3f9238ab904ce1e85baff", size = 805984 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/89/aa/ab0f7891a01eeb2d2e338ae8fecbe57fcebea1a24dbb64d45801bfab481d/attrs-24.3.0-py3-none-any.whl", hash = "sha256:ac96cd038792094f438ad1f6ff80837353805ac950cd2aa0e0625ef19850c308", size = 63397 },
+]
+
+[[package]]
+name = "beartype"
+version = "0.15.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d2/6c/fecfe9c7d7b903a52b732b8f33968a067cad8209848ae8037c4d0ed53055/beartype-0.15.0.tar.gz", hash = "sha256:2af6a8d8a7267ccf7d271e1a3bd908afbc025d2a09aa51123567d7d7b37438df", size = 1002709 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6f/e7/9220de010e015fdfb7ce22e7d9a846bfcd6e5d686c4fc555fa76a22846ba/beartype-0.15.0-py3-none-any.whl", hash = "sha256:52cd2edea72fdd84e4e7f8011a9e3007bf0125c3d6d7219e937b9d8868169177", size = 777565 },
+]
+
+[[package]]
+name = "certifi"
+version = "2024.12.14"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0f/bd/1d41ee578ce09523c81a15426705dd20969f5abf006d1afe8aeff0dd776a/certifi-2024.12.14.tar.gz", hash = "sha256:b650d30f370c2b724812bee08008be0c4163b163ddaec3f2546c1caf65f191db", size = 166010 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a5/32/8f6669fc4798494966bf446c8c4a162e0b5d893dff088afddf76414f70e1/certifi-2024.12.14-py3-none-any.whl", hash = "sha256:1275f7a45be9464efc1173084eaa30f866fe2e47d389406136d332ed4967ec56", size = 164927 },
+]
+
+[[package]]
+name = "cfgv"
+version = "3.4.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/11/74/539e56497d9bd1d484fd863dd69cbbfa653cd2aa27abfe35653494d85e94/cfgv-3.4.0.tar.gz", hash = "sha256:e52591d4c5f5dead8e0f673fb16db7949d2cfb3f7da4582893288f0ded8fe560", size = 7114 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c5/55/51844dd50c4fc7a33b653bfaba4c2456f06955289ca770a5dbd5fd267374/cfgv-3.4.0-py2.py3-none-any.whl", hash = "sha256:b7265b1f29fd3316bfcd2b330d63d024f2bfd8bcb8b0272f8e19a504856c48f9", size = 7249 },
+]
+
+[[package]]
+name = "chardet"
+version = "5.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/0d/f7b6ab21ec75897ed80c17d79b15951a719226b9fababf1e40ea74d69079/chardet-5.2.0.tar.gz", hash = "sha256:1b3b6ff479a8c414bc3fa2c0852995695c4a026dcd6d0633b2dd092ca39c1cf7", size = 2069618 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/38/6f/f5fbc992a329ee4e0f288c1fe0e2ad9485ed064cac731ed2fe47dcc38cbf/chardet-5.2.0-py3-none-any.whl", hash = "sha256:e1cf59446890a00105fe7b7912492ea04b6e6f06d4b742b2c788469e34c82970", size = 199385 },
+]
+
+[[package]]
+name = "charset-normalizer"
+version = "3.4.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/16/b0/572805e227f01586461c80e0fd25d65a2115599cc9dad142fee4b747c357/charset_normalizer-3.4.1.tar.gz", hash = "sha256:44251f18cd68a75b56585dd00dae26183e102cd5e0f9f1466e6df5da2ed64ea3", size = 123188 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0d/58/5580c1716040bc89206c77d8f74418caf82ce519aae06450393ca73475d1/charset_normalizer-3.4.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:91b36a978b5ae0ee86c394f5a54d6ef44db1de0815eb43de826d41d21e4af3de", size = 198013 },
+    { url = "https://files.pythonhosted.org/packages/d0/11/00341177ae71c6f5159a08168bcb98c6e6d196d372c94511f9f6c9afe0c6/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7461baadb4dc00fd9e0acbe254e3d7d2112e7f92ced2adc96e54ef6501c5f176", size = 141285 },
+    { url = "https://files.pythonhosted.org/packages/01/09/11d684ea5819e5a8f5100fb0b38cf8d02b514746607934134d31233e02c8/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e218488cd232553829be0664c2292d3af2eeeb94b32bea483cf79ac6a694e037", size = 151449 },
+    { url = "https://files.pythonhosted.org/packages/08/06/9f5a12939db324d905dc1f70591ae7d7898d030d7662f0d426e2286f68c9/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:80ed5e856eb7f30115aaf94e4a08114ccc8813e6ed1b5efa74f9f82e8509858f", size = 143892 },
+    { url = "https://files.pythonhosted.org/packages/93/62/5e89cdfe04584cb7f4d36003ffa2936681b03ecc0754f8e969c2becb7e24/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b010a7a4fd316c3c484d482922d13044979e78d1861f0e0650423144c616a46a", size = 146123 },
+    { url = "https://files.pythonhosted.org/packages/a9/ac/ab729a15c516da2ab70a05f8722ecfccc3f04ed7a18e45c75bbbaa347d61/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4532bff1b8421fd0a320463030c7520f56a79c9024a4e88f01c537316019005a", size = 147943 },
+    { url = "https://files.pythonhosted.org/packages/03/d2/3f392f23f042615689456e9a274640c1d2e5dd1d52de36ab8f7955f8f050/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d973f03c0cb71c5ed99037b870f2be986c3c05e63622c017ea9816881d2dd247", size = 142063 },
+    { url = "https://files.pythonhosted.org/packages/f2/e3/e20aae5e1039a2cd9b08d9205f52142329f887f8cf70da3650326670bddf/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:3a3bd0dcd373514dcec91c411ddb9632c0d7d92aed7093b8c3bbb6d69ca74408", size = 150578 },
+    { url = "https://files.pythonhosted.org/packages/8d/af/779ad72a4da0aed925e1139d458adc486e61076d7ecdcc09e610ea8678db/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:d9c3cdf5390dcd29aa8056d13e8e99526cda0305acc038b96b30352aff5ff2bb", size = 153629 },
+    { url = "https://files.pythonhosted.org/packages/c2/b6/7aa450b278e7aa92cf7732140bfd8be21f5f29d5bf334ae987c945276639/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:2bdfe3ac2e1bbe5b59a1a63721eb3b95fc9b6817ae4a46debbb4e11f6232428d", size = 150778 },
+    { url = "https://files.pythonhosted.org/packages/39/f4/d9f4f712d0951dcbfd42920d3db81b00dd23b6ab520419626f4023334056/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:eab677309cdb30d047996b36d34caeda1dc91149e4fdca0b1a039b3f79d9a807", size = 146453 },
+    { url = "https://files.pythonhosted.org/packages/49/2b/999d0314e4ee0cff3cb83e6bc9aeddd397eeed693edb4facb901eb8fbb69/charset_normalizer-3.4.1-cp310-cp310-win32.whl", hash = "sha256:c0429126cf75e16c4f0ad00ee0eae4242dc652290f940152ca8c75c3a4b6ee8f", size = 95479 },
+    { url = "https://files.pythonhosted.org/packages/2d/ce/3cbed41cff67e455a386fb5e5dd8906cdda2ed92fbc6297921f2e4419309/charset_normalizer-3.4.1-cp310-cp310-win_amd64.whl", hash = "sha256:9f0b8b1c6d84c8034a44893aba5e767bf9c7a211e313a9605d9c617d7083829f", size = 102790 },
+    { url = "https://files.pythonhosted.org/packages/72/80/41ef5d5a7935d2d3a773e3eaebf0a9350542f2cab4eac59a7a4741fbbbbe/charset_normalizer-3.4.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:8bfa33f4f2672964266e940dd22a195989ba31669bd84629f05fab3ef4e2d125", size = 194995 },
+    { url = "https://files.pythonhosted.org/packages/7a/28/0b9fefa7b8b080ec492110af6d88aa3dea91c464b17d53474b6e9ba5d2c5/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:28bf57629c75e810b6ae989f03c0828d64d6b26a5e205535585f96093e405ed1", size = 139471 },
+    { url = "https://files.pythonhosted.org/packages/71/64/d24ab1a997efb06402e3fc07317e94da358e2585165930d9d59ad45fcae2/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f08ff5e948271dc7e18a35641d2f11a4cd8dfd5634f55228b691e62b37125eb3", size = 149831 },
+    { url = "https://files.pythonhosted.org/packages/37/ed/be39e5258e198655240db5e19e0b11379163ad7070962d6b0c87ed2c4d39/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:234ac59ea147c59ee4da87a0c0f098e9c8d169f4dc2a159ef720f1a61bbe27cd", size = 142335 },
+    { url = "https://files.pythonhosted.org/packages/88/83/489e9504711fa05d8dde1574996408026bdbdbd938f23be67deebb5eca92/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd4ec41f914fa74ad1b8304bbc634b3de73d2a0889bd32076342a573e0779e00", size = 143862 },
+    { url = "https://files.pythonhosted.org/packages/c6/c7/32da20821cf387b759ad24627a9aca289d2822de929b8a41b6241767b461/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:eea6ee1db730b3483adf394ea72f808b6e18cf3cb6454b4d86e04fa8c4327a12", size = 145673 },
+    { url = "https://files.pythonhosted.org/packages/68/85/f4288e96039abdd5aeb5c546fa20a37b50da71b5cf01e75e87f16cd43304/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:c96836c97b1238e9c9e3fe90844c947d5afbf4f4c92762679acfe19927d81d77", size = 140211 },
+    { url = "https://files.pythonhosted.org/packages/28/a3/a42e70d03cbdabc18997baf4f0227c73591a08041c149e710045c281f97b/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:4d86f7aff21ee58f26dcf5ae81a9addbd914115cdebcbb2217e4f0ed8982e146", size = 148039 },
+    { url = "https://files.pythonhosted.org/packages/85/e4/65699e8ab3014ecbe6f5c71d1a55d810fb716bbfd74f6283d5c2aa87febf/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:09b5e6733cbd160dcc09589227187e242a30a49ca5cefa5a7edd3f9d19ed53fd", size = 151939 },
+    { url = "https://files.pythonhosted.org/packages/b1/82/8e9fe624cc5374193de6860aba3ea8070f584c8565ee77c168ec13274bd2/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:5777ee0881f9499ed0f71cc82cf873d9a0ca8af166dfa0af8ec4e675b7df48e6", size = 149075 },
+    { url = "https://files.pythonhosted.org/packages/3d/7b/82865ba54c765560c8433f65e8acb9217cb839a9e32b42af4aa8e945870f/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:237bdbe6159cff53b4f24f397d43c6336c6b0b42affbe857970cefbb620911c8", size = 144340 },
+    { url = "https://files.pythonhosted.org/packages/b5/b6/9674a4b7d4d99a0d2df9b215da766ee682718f88055751e1e5e753c82db0/charset_normalizer-3.4.1-cp311-cp311-win32.whl", hash = "sha256:8417cb1f36cc0bc7eaba8ccb0e04d55f0ee52df06df3ad55259b9a323555fc8b", size = 95205 },
+    { url = "https://files.pythonhosted.org/packages/1e/ab/45b180e175de4402dcf7547e4fb617283bae54ce35c27930a6f35b6bef15/charset_normalizer-3.4.1-cp311-cp311-win_amd64.whl", hash = "sha256:d7f50a1f8c450f3925cb367d011448c39239bb3eb4117c36a6d354794de4ce76", size = 102441 },
+    { url = "https://files.pythonhosted.org/packages/0a/9a/dd1e1cdceb841925b7798369a09279bd1cf183cef0f9ddf15a3a6502ee45/charset_normalizer-3.4.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:73d94b58ec7fecbc7366247d3b0b10a21681004153238750bb67bd9012414545", size = 196105 },
+    { url = "https://files.pythonhosted.org/packages/d3/8c/90bfabf8c4809ecb648f39794cf2a84ff2e7d2a6cf159fe68d9a26160467/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dad3e487649f498dd991eeb901125411559b22e8d7ab25d3aeb1af367df5efd7", size = 140404 },
+    { url = "https://files.pythonhosted.org/packages/ad/8f/e410d57c721945ea3b4f1a04b74f70ce8fa800d393d72899f0a40526401f/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c30197aa96e8eed02200a83fba2657b4c3acd0f0aa4bdc9f6c1af8e8962e0757", size = 150423 },
+    { url = "https://files.pythonhosted.org/packages/f0/b8/e6825e25deb691ff98cf5c9072ee0605dc2acfca98af70c2d1b1bc75190d/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2369eea1ee4a7610a860d88f268eb39b95cb588acd7235e02fd5a5601773d4fa", size = 143184 },
+    { url = "https://files.pythonhosted.org/packages/3e/a2/513f6cbe752421f16d969e32f3583762bfd583848b763913ddab8d9bfd4f/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc2722592d8998c870fa4e290c2eec2c1569b87fe58618e67d38b4665dfa680d", size = 145268 },
+    { url = "https://files.pythonhosted.org/packages/74/94/8a5277664f27c3c438546f3eb53b33f5b19568eb7424736bdc440a88a31f/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ffc9202a29ab3920fa812879e95a9e78b2465fd10be7fcbd042899695d75e616", size = 147601 },
+    { url = "https://files.pythonhosted.org/packages/7c/5f/6d352c51ee763623a98e31194823518e09bfa48be2a7e8383cf691bbb3d0/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:804a4d582ba6e5b747c625bf1255e6b1507465494a40a2130978bda7b932c90b", size = 141098 },
+    { url = "https://files.pythonhosted.org/packages/78/d4/f5704cb629ba5ab16d1d3d741396aec6dc3ca2b67757c45b0599bb010478/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:0f55e69f030f7163dffe9fd0752b32f070566451afe180f99dbeeb81f511ad8d", size = 149520 },
+    { url = "https://files.pythonhosted.org/packages/c5/96/64120b1d02b81785f222b976c0fb79a35875457fa9bb40827678e54d1bc8/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:c4c3e6da02df6fa1410a7680bd3f63d4f710232d3139089536310d027950696a", size = 152852 },
+    { url = "https://files.pythonhosted.org/packages/84/c9/98e3732278a99f47d487fd3468bc60b882920cef29d1fa6ca460a1fdf4e6/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:5df196eb874dae23dcfb968c83d4f8fdccb333330fe1fc278ac5ceeb101003a9", size = 150488 },
+    { url = "https://files.pythonhosted.org/packages/13/0e/9c8d4cb99c98c1007cc11eda969ebfe837bbbd0acdb4736d228ccaabcd22/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e358e64305fe12299a08e08978f51fc21fac060dcfcddd95453eabe5b93ed0e1", size = 146192 },
+    { url = "https://files.pythonhosted.org/packages/b2/21/2b6b5b860781a0b49427309cb8670785aa543fb2178de875b87b9cc97746/charset_normalizer-3.4.1-cp312-cp312-win32.whl", hash = "sha256:9b23ca7ef998bc739bf6ffc077c2116917eabcc901f88da1b9856b210ef63f35", size = 95550 },
+    { url = "https://files.pythonhosted.org/packages/21/5b/1b390b03b1d16c7e382b561c5329f83cc06623916aab983e8ab9239c7d5c/charset_normalizer-3.4.1-cp312-cp312-win_amd64.whl", hash = "sha256:6ff8a4a60c227ad87030d76e99cd1698345d4491638dfa6673027c48b3cd395f", size = 102785 },
+    { url = "https://files.pythonhosted.org/packages/38/94/ce8e6f63d18049672c76d07d119304e1e2d7c6098f0841b51c666e9f44a0/charset_normalizer-3.4.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:aabfa34badd18f1da5ec1bc2715cadc8dca465868a4e73a0173466b688f29dda", size = 195698 },
+    { url = "https://files.pythonhosted.org/packages/24/2e/dfdd9770664aae179a96561cc6952ff08f9a8cd09a908f259a9dfa063568/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:22e14b5d70560b8dd51ec22863f370d1e595ac3d024cb8ad7d308b4cd95f8313", size = 140162 },
+    { url = "https://files.pythonhosted.org/packages/24/4e/f646b9093cff8fc86f2d60af2de4dc17c759de9d554f130b140ea4738ca6/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8436c508b408b82d87dc5f62496973a1805cd46727c34440b0d29d8a2f50a6c9", size = 150263 },
+    { url = "https://files.pythonhosted.org/packages/5e/67/2937f8d548c3ef6e2f9aab0f6e21001056f692d43282b165e7c56023e6dd/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2d074908e1aecee37a7635990b2c6d504cd4766c7bc9fc86d63f9c09af3fa11b", size = 142966 },
+    { url = "https://files.pythonhosted.org/packages/52/ed/b7f4f07de100bdb95c1756d3a4d17b90c1a3c53715c1a476f8738058e0fa/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:955f8851919303c92343d2f66165294848d57e9bba6cf6e3625485a70a038d11", size = 144992 },
+    { url = "https://files.pythonhosted.org/packages/96/2c/d49710a6dbcd3776265f4c923bb73ebe83933dfbaa841c5da850fe0fd20b/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:44ecbf16649486d4aebafeaa7ec4c9fed8b88101f4dd612dcaf65d5e815f837f", size = 147162 },
+    { url = "https://files.pythonhosted.org/packages/b4/41/35ff1f9a6bd380303dea55e44c4933b4cc3c4850988927d4082ada230273/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:0924e81d3d5e70f8126529951dac65c1010cdf117bb75eb02dd12339b57749dd", size = 140972 },
+    { url = "https://files.pythonhosted.org/packages/fb/43/c6a0b685fe6910d08ba971f62cd9c3e862a85770395ba5d9cad4fede33ab/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:2967f74ad52c3b98de4c3b32e1a44e32975e008a9cd2a8cc8966d6a5218c5cb2", size = 149095 },
+    { url = "https://files.pythonhosted.org/packages/4c/ff/a9a504662452e2d2878512115638966e75633519ec11f25fca3d2049a94a/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:c75cb2a3e389853835e84a2d8fb2b81a10645b503eca9bcb98df6b5a43eb8886", size = 152668 },
+    { url = "https://files.pythonhosted.org/packages/6c/71/189996b6d9a4b932564701628af5cee6716733e9165af1d5e1b285c530ed/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:09b26ae6b1abf0d27570633b2b078a2a20419c99d66fb2823173d73f188ce601", size = 150073 },
+    { url = "https://files.pythonhosted.org/packages/e4/93/946a86ce20790e11312c87c75ba68d5f6ad2208cfb52b2d6a2c32840d922/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:fa88b843d6e211393a37219e6a1c1df99d35e8fd90446f1118f4216e307e48cd", size = 145732 },
+    { url = "https://files.pythonhosted.org/packages/cd/e5/131d2fb1b0dddafc37be4f3a2fa79aa4c037368be9423061dccadfd90091/charset_normalizer-3.4.1-cp313-cp313-win32.whl", hash = "sha256:eb8178fe3dba6450a3e024e95ac49ed3400e506fd4e9e5c32d30adda88cbd407", size = 95391 },
+    { url = "https://files.pythonhosted.org/packages/27/f2/4f9a69cc7712b9b5ad8fdb87039fd89abba997ad5cbe690d1835d40405b0/charset_normalizer-3.4.1-cp313-cp313-win_amd64.whl", hash = "sha256:b1ac5992a838106edb89654e0aebfc24f5848ae2547d22c2c3f66454daa11971", size = 102702 },
+    { url = "https://files.pythonhosted.org/packages/7f/c0/b913f8f02836ed9ab32ea643c6fe4d3325c3d8627cf6e78098671cafff86/charset_normalizer-3.4.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:b97e690a2118911e39b4042088092771b4ae3fc3aa86518f84b8cf6888dbdb41", size = 197867 },
+    { url = "https://files.pythonhosted.org/packages/0f/6c/2bee440303d705b6fb1e2ec789543edec83d32d258299b16eed28aad48e0/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:78baa6d91634dfb69ec52a463534bc0df05dbd546209b79a3880a34487f4b84f", size = 141385 },
+    { url = "https://files.pythonhosted.org/packages/3d/04/cb42585f07f6f9fd3219ffb6f37d5a39b4fd2db2355b23683060029c35f7/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1a2bc9f351a75ef49d664206d51f8e5ede9da246602dc2d2726837620ea034b2", size = 151367 },
+    { url = "https://files.pythonhosted.org/packages/54/54/2412a5b093acb17f0222de007cc129ec0e0df198b5ad2ce5699355269dfe/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:75832c08354f595c760a804588b9357d34ec00ba1c940c15e31e96d902093770", size = 143928 },
+    { url = "https://files.pythonhosted.org/packages/5a/6d/e2773862b043dcf8a221342954f375392bb2ce6487bcd9f2c1b34e1d6781/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0af291f4fe114be0280cdd29d533696a77b5b49cfde5467176ecab32353395c4", size = 146203 },
+    { url = "https://files.pythonhosted.org/packages/b9/f8/ca440ef60d8f8916022859885f231abb07ada3c347c03d63f283bec32ef5/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0167ddc8ab6508fe81860a57dd472b2ef4060e8d378f0cc555707126830f2537", size = 148082 },
+    { url = "https://files.pythonhosted.org/packages/04/d2/42fd330901aaa4b805a1097856c2edf5095e260a597f65def493f4b8c833/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:2a75d49014d118e4198bcee5ee0a6f25856b29b12dbf7cd012791f8a6cc5c496", size = 142053 },
+    { url = "https://files.pythonhosted.org/packages/9e/af/3a97a4fa3c53586f1910dadfc916e9c4f35eeada36de4108f5096cb7215f/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:363e2f92b0f0174b2f8238240a1a30142e3db7b957a5dd5689b0e75fb717cc78", size = 150625 },
+    { url = "https://files.pythonhosted.org/packages/26/ae/23d6041322a3556e4da139663d02fb1b3c59a23ab2e2b56432bd2ad63ded/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:ab36c8eb7e454e34e60eb55ca5d241a5d18b2c6244f6827a30e451c42410b5f7", size = 153549 },
+    { url = "https://files.pythonhosted.org/packages/94/22/b8f2081c6a77cb20d97e57e0b385b481887aa08019d2459dc2858ed64871/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:4c0907b1928a36d5a998d72d64d8eaa7244989f7aaaf947500d3a800c83a3fd6", size = 150945 },
+    { url = "https://files.pythonhosted.org/packages/c7/0b/c5ec5092747f801b8b093cdf5610e732b809d6cb11f4c51e35fc28d1d389/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:04432ad9479fa40ec0f387795ddad4437a2b50417c69fa275e212933519ff294", size = 146595 },
+    { url = "https://files.pythonhosted.org/packages/0c/5a/0b59704c38470df6768aa154cc87b1ac7c9bb687990a1559dc8765e8627e/charset_normalizer-3.4.1-cp39-cp39-win32.whl", hash = "sha256:3bed14e9c89dcb10e8f3a29f9ccac4955aebe93c71ae803af79265c9ca5644c5", size = 95453 },
+    { url = "https://files.pythonhosted.org/packages/85/2d/a9790237cb4d01a6d57afadc8573c8b73c609ade20b80f4cda30802009ee/charset_normalizer-3.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:49402233c892a461407c512a19435d1ce275543138294f7ef013f0b63d5d3765", size = 102811 },
+    { url = "https://files.pythonhosted.org/packages/0e/f6/65ecc6878a89bb1c23a086ea335ad4bf21a588990c3f535a227b9eea9108/charset_normalizer-3.4.1-py3-none-any.whl", hash = "sha256:d98b1668f06378c6dbefec3b92299716b931cd4e6061f3c875a71ced1780ab85", size = 49767 },
+]
+
+[[package]]
+name = "click"
+version = "8.1.8"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b9/2e/0090cbf739cee7d23781ad4b89a9894a41538e4fcf4c31dcdd705b78eb8b/click-8.1.8.tar.gz", hash = "sha256:ed53c9d8990d83c2a27deae68e4ee337473f6330c040a31d4225c9574d16096a", size = 226593 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7e/d4/7ebdbd03970677812aac39c869717059dbb71a4cfc033ca6e5221787892c/click-8.1.8-py3-none-any.whl", hash = "sha256:63c132bbbed01578a06712a2d1f497bb62d9c1c0d329b7903a866228027263b2", size = 98188 },
+]
+
+[[package]]
+name = "cloudpickle"
+version = "3.1.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/52/39/069100b84d7418bc358d81669d5748efb14b9cceacd2f9c75f550424132f/cloudpickle-3.1.1.tar.gz", hash = "sha256:b216fa8ae4019d5482a8ac3c95d8f6346115d8835911fd4aefd1a445e4242c64", size = 22113 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7e/e8/64c37fadfc2816a7701fa8a6ed8d87327c7d54eacfbfb6edab14a2f2be75/cloudpickle-3.1.1-py3-none-any.whl", hash = "sha256:c8c5a44295039331ee9dad40ba100a9c7297b6f988e50e87ccdf3765a668350e", size = 20992 },
+]
+
+[[package]]
+name = "cmake"
+version = "3.31.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/50/cb/3a327fa784a5dbaf838b135cb1729f43535c52d83bbf02191fb8a0cb118e/cmake-3.31.4.tar.gz", hash = "sha256:a6ac2242e0b16ad7d94c9f8572d6f232e6169747be50e5cdf497f206c4819ce1", size = 34278 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d9/db/50efa1d3e29cb2a6e8e143e522e52698b3fc08f4b56100fb35f97a70af79/cmake-3.31.4-py3-none-macosx_10_10_universal2.whl", hash = "sha256:fc048b4b70facd16699a43c737f6782b4eff56e8e6093090db5979532d9db0f6", size = 47198138 },
+    { url = "https://files.pythonhosted.org/packages/c7/76/ccb8764761c739ef16bd8957a16ecbda01b03c2d7d241c376bfca6bf2822/cmake-3.31.4-py3-none-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:2a37be93534df04513f0845492d71bc80899c3f87b77e3b01c95aff1a7fc9bde", size = 27556485 },
+    { url = "https://files.pythonhosted.org/packages/ad/8e/888e2944655d7fa1ea5af46b60883a0e7847bbf9fb7ecc321c8e5f0a1394/cmake-3.31.4-py3-none-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:c9f5f8289c5e7bd2ed654cbac164021fa7723064fee0443a2f0068bc08413d81", size = 26808834 },
+    { url = "https://files.pythonhosted.org/packages/59/f4/0b2b1430a441c3c09ee102bf8c5d9ec1dc11d002ff4affef15c656f37ce9/cmake-3.31.4-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:926d91cae2ba7d2f3df857d0fc066bdac4f3904bf5c95e99b60435e85aabedb4", size = 27140820 },
+    { url = "https://files.pythonhosted.org/packages/d1/f9/a274b4e36e457d8e99db1038cc31a6c391bf3bc26230c2dc9caf37499753/cmake-3.31.4-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:929a8d8d289d69e43784661748ddd08933ce1ec5db8f9bcfce6ee817a48f8787", size = 28868269 },
+    { url = "https://files.pythonhosted.org/packages/9b/35/8da1ffa00a3f3853881aa5025cdf11c744303013df70c8716155b83825d3/cmake-3.31.4-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b463efdf5b92f3b290235aa9f8da092b3dac19b7636c563fd156022dab580649", size = 30732267 },
+    { url = "https://files.pythonhosted.org/packages/79/48/bb8485687f5a64d52ac68cfcb02e9b8e46a9e107f380c54d484b6632c87e/cmake-3.31.4-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:225d9a643b0b60ffce0399ff0cabd7a4820e0dbcb794e97d3aacfcf7c0589ae6", size = 26908885 },
+    { url = "https://files.pythonhosted.org/packages/e5/9e/2594d7fa8b263296497bf044469b4ab4797c51675ea629f9672011cdfe09/cmake-3.31.4-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:89143a5e2a5916061f2cfc5012e9fe6281aaf7c0dae7930bdc68d105d22ddc39", size = 27784555 },
+    { url = "https://files.pythonhosted.org/packages/95/16/5b1989f1d2287b05cd68792c0a48b721c060f728506d719fcf0e3b80ceb2/cmake-3.31.4-py3-none-manylinux_2_31_armv7l.whl", hash = "sha256:f96127bf663168accd29d5a50ee68ea80f26bcd37f96c7a14ef2378781f19936", size = 24965366 },
+    { url = "https://files.pythonhosted.org/packages/5a/4c/289fb0986c6ff63583383eca0c9479147f362330938856a9b5201c84cee8/cmake-3.31.4-py3-none-musllinux_1_1_aarch64.whl", hash = "sha256:25c5094394f0cee21130b5678e5b4552f72470e266df6d6fb1d5c505100f0eaa", size = 27824887 },
+    { url = "https://files.pythonhosted.org/packages/3c/f3/d45ba2b5bb54f4ef615a6a24cf6258600eec790a9d5017c9584107b445b9/cmake-3.31.4-py3-none-musllinux_1_1_i686.whl", hash = "sha256:466c9295af440bb4a47cc5e1af10576cf2227620528afd0fd0b3effa1d513b49", size = 31368421 },
+    { url = "https://files.pythonhosted.org/packages/34/3d/f6b712241ede5fb8e32c13e119c06e142f3f12ead1656721b1f67756106b/cmake-3.31.4-py3-none-musllinux_1_1_ppc64le.whl", hash = "sha256:f6af3b83a1b1fc1d990d18b6a566ee9c95c0393f986c6df15f2505dda8ad1bcc", size = 32074545 },
+    { url = "https://files.pythonhosted.org/packages/f0/23/48cd0404d7238d703a4cd4d7434eeaf12e8fbe68160d52f1489f55f582df/cmake-3.31.4-py3-none-musllinux_1_1_s390x.whl", hash = "sha256:23781e17563693a68b0cef85749746894b8a61488e56e96fc6649b73652e8236", size = 27946950 },
+    { url = "https://files.pythonhosted.org/packages/21/03/014d9710bccf5a7e04c6f6ee27bfaba1220e79ee145d7b95f84e7843729b/cmake-3.31.4-py3-none-musllinux_1_1_x86_64.whl", hash = "sha256:838a388b559137f3654d8cf30f62bbdec10f8d1c3624f0d289614d33cdf4fba1", size = 29473412 },
+    { url = "https://files.pythonhosted.org/packages/23/de/5a8142732f0a52dedac2887e0c105c9bbb449e517ade500e56bf2af520d1/cmake-3.31.4-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:a6a3b0b9557f41c955a6b25c94205f2ca9c3a46edca809ad87507c5ef6bc4274", size = 32971081 },
+    { url = "https://files.pythonhosted.org/packages/a5/a1/50c11f0b110986c753592f025970094030b25748df126abe8e38265be722/cmake-3.31.4-py3-none-win32.whl", hash = "sha256:d378c9e58eac906bddafd673c7571262dcd5a9946bb1e8f9e3902572a8fa95ca", size = 33351393 },
+    { url = "https://files.pythonhosted.org/packages/0c/7f/331d181b6b1b8942ec5fad23e98fff85218485f29f62f6bc60663d424df8/cmake-3.31.4-py3-none-win_amd64.whl", hash = "sha256:20be7cdb41903edf85e8a498c4beff8d6854acbb087abfb07c362c738bdf0018", size = 36496715 },
+    { url = "https://files.pythonhosted.org/packages/65/26/11a78723364716004928b7bea7d96cf2c72dc3abfaa7c163159110fcb649/cmake-3.31.4-py3-none-win_arm64.whl", hash = "sha256:9479a9255197c49e135df039d8484c69aa63158a06ae9c2d0eb939da2f0f7dff", size = 35559239 },
+]
+
+[[package]]
+name = "colorama"
+version = "0.4.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d8/53/6f443c9a4a8358a93a6792e2acffb9d9d5cb0a5cfd8802644b7b1c9a02e4/colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44", size = 27697 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6", size = 25335 },
+]
+
+[[package]]
+name = "coverage"
+version = "7.6.10"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/84/ba/ac14d281f80aab516275012e8875991bb06203957aa1e19950139238d658/coverage-7.6.10.tar.gz", hash = "sha256:7fb105327c8f8f0682e29843e2ff96af9dcbe5bab8eeb4b398c6a33a16d80a23", size = 803868 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c5/12/2a2a923edf4ddabdffed7ad6da50d96a5c126dae7b80a33df7310e329a1e/coverage-7.6.10-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:5c912978f7fbf47ef99cec50c4401340436d200d41d714c7a4766f377c5b7b78", size = 207982 },
+    { url = "https://files.pythonhosted.org/packages/ca/49/6985dbca9c7be3f3cb62a2e6e492a0c88b65bf40579e16c71ae9c33c6b23/coverage-7.6.10-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a01ec4af7dfeb96ff0078ad9a48810bb0cc8abcb0115180c6013a6b26237626c", size = 208414 },
+    { url = "https://files.pythonhosted.org/packages/35/93/287e8f1d1ed2646f4e0b2605d14616c9a8a2697d0d1b453815eb5c6cebdb/coverage-7.6.10-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a3b204c11e2b2d883946fe1d97f89403aa1811df28ce0447439178cc7463448a", size = 236860 },
+    { url = "https://files.pythonhosted.org/packages/de/e1/cfdb5627a03567a10031acc629b75d45a4ca1616e54f7133ca1fa366050a/coverage-7.6.10-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:32ee6d8491fcfc82652a37109f69dee9a830e9379166cb73c16d8dc5c2915165", size = 234758 },
+    { url = "https://files.pythonhosted.org/packages/6d/85/fc0de2bcda3f97c2ee9fe8568f7d48f7279e91068958e5b2cc19e0e5f600/coverage-7.6.10-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:675cefc4c06e3b4c876b85bfb7c59c5e2218167bbd4da5075cbe3b5790a28988", size = 235920 },
+    { url = "https://files.pythonhosted.org/packages/79/73/ef4ea0105531506a6f4cf4ba571a214b14a884630b567ed65b3d9c1975e1/coverage-7.6.10-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:f4f620668dbc6f5e909a0946a877310fb3d57aea8198bde792aae369ee1c23b5", size = 234986 },
+    { url = "https://files.pythonhosted.org/packages/c6/4d/75afcfe4432e2ad0405c6f27adeb109ff8976c5e636af8604f94f29fa3fc/coverage-7.6.10-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:4eea95ef275de7abaef630c9b2c002ffbc01918b726a39f5a4353916ec72d2f3", size = 233446 },
+    { url = "https://files.pythonhosted.org/packages/86/5b/efee56a89c16171288cafff022e8af44f8f94075c2d8da563c3935212871/coverage-7.6.10-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:e2f0280519e42b0a17550072861e0bc8a80a0870de260f9796157d3fca2733c5", size = 234566 },
+    { url = "https://files.pythonhosted.org/packages/f2/db/67770cceb4a64d3198bf2aa49946f411b85ec6b0a9b489e61c8467a4253b/coverage-7.6.10-cp310-cp310-win32.whl", hash = "sha256:bc67deb76bc3717f22e765ab3e07ee9c7a5e26b9019ca19a3b063d9f4b874244", size = 210675 },
+    { url = "https://files.pythonhosted.org/packages/8d/27/e8bfc43f5345ec2c27bc8a1fa77cdc5ce9dcf954445e11f14bb70b889d14/coverage-7.6.10-cp310-cp310-win_amd64.whl", hash = "sha256:0f460286cb94036455e703c66988851d970fdfd8acc2a1122ab7f4f904e4029e", size = 211518 },
+    { url = "https://files.pythonhosted.org/packages/85/d2/5e175fcf6766cf7501a8541d81778fd2f52f4870100e791f5327fd23270b/coverage-7.6.10-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:ea3c8f04b3e4af80e17bab607c386a830ffc2fb88a5484e1df756478cf70d1d3", size = 208088 },
+    { url = "https://files.pythonhosted.org/packages/4b/6f/06db4dc8fca33c13b673986e20e466fd936235a6ec1f0045c3853ac1b593/coverage-7.6.10-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:507a20fc863cae1d5720797761b42d2d87a04b3e5aeb682ef3b7332e90598f43", size = 208536 },
+    { url = "https://files.pythonhosted.org/packages/0d/62/c6a0cf80318c1c1af376d52df444da3608eafc913b82c84a4600d8349472/coverage-7.6.10-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d37a84878285b903c0fe21ac8794c6dab58150e9359f1aaebbeddd6412d53132", size = 240474 },
+    { url = "https://files.pythonhosted.org/packages/a3/59/750adafc2e57786d2e8739a46b680d4fb0fbc2d57fbcb161290a9f1ecf23/coverage-7.6.10-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a534738b47b0de1995f85f582d983d94031dffb48ab86c95bdf88dc62212142f", size = 237880 },
+    { url = "https://files.pythonhosted.org/packages/2c/f8/ef009b3b98e9f7033c19deb40d629354aab1d8b2d7f9cfec284dbedf5096/coverage-7.6.10-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0d7a2bf79378d8fb8afaa994f91bfd8215134f8631d27eba3e0e2c13546ce994", size = 239750 },
+    { url = "https://files.pythonhosted.org/packages/a6/e2/6622f3b70f5f5b59f705e680dae6db64421af05a5d1e389afd24dae62e5b/coverage-7.6.10-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:6713ba4b4ebc330f3def51df1d5d38fad60b66720948112f114968feb52d3f99", size = 238642 },
+    { url = "https://files.pythonhosted.org/packages/2d/10/57ac3f191a3c95c67844099514ff44e6e19b2915cd1c22269fb27f9b17b6/coverage-7.6.10-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:ab32947f481f7e8c763fa2c92fd9f44eeb143e7610c4ca9ecd6a36adab4081bd", size = 237266 },
+    { url = "https://files.pythonhosted.org/packages/ee/2d/7016f4ad9d553cabcb7333ed78ff9d27248ec4eba8dd21fa488254dff894/coverage-7.6.10-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:7bbd8c8f1b115b892e34ba66a097b915d3871db7ce0e6b9901f462ff3a975377", size = 238045 },
+    { url = "https://files.pythonhosted.org/packages/a7/fe/45af5c82389a71e0cae4546413266d2195c3744849669b0bab4b5f2c75da/coverage-7.6.10-cp311-cp311-win32.whl", hash = "sha256:299e91b274c5c9cdb64cbdf1b3e4a8fe538a7a86acdd08fae52301b28ba297f8", size = 210647 },
+    { url = "https://files.pythonhosted.org/packages/db/11/3f8e803a43b79bc534c6a506674da9d614e990e37118b4506faf70d46ed6/coverage-7.6.10-cp311-cp311-win_amd64.whl", hash = "sha256:489a01f94aa581dbd961f306e37d75d4ba16104bbfa2b0edb21d29b73be83609", size = 211508 },
+    { url = "https://files.pythonhosted.org/packages/86/77/19d09ea06f92fdf0487499283b1b7af06bc422ea94534c8fe3a4cd023641/coverage-7.6.10-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:27c6e64726b307782fa5cbe531e7647aee385a29b2107cd87ba7c0105a5d3853", size = 208281 },
+    { url = "https://files.pythonhosted.org/packages/b6/67/5479b9f2f99fcfb49c0d5cf61912a5255ef80b6e80a3cddba39c38146cf4/coverage-7.6.10-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:c56e097019e72c373bae32d946ecf9858fda841e48d82df7e81c63ac25554078", size = 208514 },
+    { url = "https://files.pythonhosted.org/packages/15/d1/febf59030ce1c83b7331c3546d7317e5120c5966471727aa7ac157729c4b/coverage-7.6.10-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c7827a5bc7bdb197b9e066cdf650b2887597ad124dd99777332776f7b7c7d0d0", size = 241537 },
+    { url = "https://files.pythonhosted.org/packages/4b/7e/5ac4c90192130e7cf8b63153fe620c8bfd9068f89a6d9b5f26f1550f7a26/coverage-7.6.10-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:204a8238afe787323a8b47d8be4df89772d5c1e4651b9ffa808552bdf20e1d50", size = 238572 },
+    { url = "https://files.pythonhosted.org/packages/dc/03/0334a79b26ecf59958f2fe9dd1f5ab3e2f88db876f5071933de39af09647/coverage-7.6.10-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e67926f51821b8e9deb6426ff3164870976fe414d033ad90ea75e7ed0c2e5022", size = 240639 },
+    { url = "https://files.pythonhosted.org/packages/d7/45/8a707f23c202208d7b286d78ad6233f50dcf929319b664b6cc18a03c1aae/coverage-7.6.10-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e78b270eadb5702938c3dbe9367f878249b5ef9a2fcc5360ac7bff694310d17b", size = 240072 },
+    { url = "https://files.pythonhosted.org/packages/66/02/603ce0ac2d02bc7b393279ef618940b4a0535b0868ee791140bda9ecfa40/coverage-7.6.10-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:714f942b9c15c3a7a5fe6876ce30af831c2ad4ce902410b7466b662358c852c0", size = 238386 },
+    { url = "https://files.pythonhosted.org/packages/04/62/4e6887e9be060f5d18f1dd58c2838b2d9646faf353232dec4e2d4b1c8644/coverage-7.6.10-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:abb02e2f5a3187b2ac4cd46b8ced85a0858230b577ccb2c62c81482ca7d18852", size = 240054 },
+    { url = "https://files.pythonhosted.org/packages/5c/74/83ae4151c170d8bd071924f212add22a0e62a7fe2b149edf016aeecad17c/coverage-7.6.10-cp312-cp312-win32.whl", hash = "sha256:55b201b97286cf61f5e76063f9e2a1d8d2972fc2fcfd2c1272530172fd28c359", size = 210904 },
+    { url = "https://files.pythonhosted.org/packages/c3/54/de0893186a221478f5880283119fc40483bc460b27c4c71d1b8bba3474b9/coverage-7.6.10-cp312-cp312-win_amd64.whl", hash = "sha256:e4ae5ac5e0d1e4edfc9b4b57b4cbecd5bc266a6915c500f358817a8496739247", size = 211692 },
+    { url = "https://files.pythonhosted.org/packages/25/6d/31883d78865529257bf847df5789e2ae80e99de8a460c3453dbfbe0db069/coverage-7.6.10-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:05fca8ba6a87aabdd2d30d0b6c838b50510b56cdcfc604d40760dae7153b73d9", size = 208308 },
+    { url = "https://files.pythonhosted.org/packages/70/22/3f2b129cc08de00c83b0ad6252e034320946abfc3e4235c009e57cfeee05/coverage-7.6.10-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:9e80eba8801c386f72e0712a0453431259c45c3249f0009aff537a517b52942b", size = 208565 },
+    { url = "https://files.pythonhosted.org/packages/97/0a/d89bc2d1cc61d3a8dfe9e9d75217b2be85f6c73ebf1b9e3c2f4e797f4531/coverage-7.6.10-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a372c89c939d57abe09e08c0578c1d212e7a678135d53aa16eec4430adc5e690", size = 241083 },
+    { url = "https://files.pythonhosted.org/packages/4c/81/6d64b88a00c7a7aaed3a657b8eaa0931f37a6395fcef61e53ff742b49c97/coverage-7.6.10-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ec22b5e7fe7a0fa8509181c4aac1db48f3dd4d3a566131b313d1efc102892c18", size = 238235 },
+    { url = "https://files.pythonhosted.org/packages/9a/0b/7797d4193f5adb4b837207ed87fecf5fc38f7cc612b369a8e8e12d9fa114/coverage-7.6.10-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:26bcf5c4df41cad1b19c84af71c22cbc9ea9a547fc973f1f2cc9a290002c8b3c", size = 240220 },
+    { url = "https://files.pythonhosted.org/packages/65/4d/6f83ca1bddcf8e51bf8ff71572f39a1c73c34cf50e752a952c34f24d0a60/coverage-7.6.10-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4e4630c26b6084c9b3cb53b15bd488f30ceb50b73c35c5ad7871b869cb7365fd", size = 239847 },
+    { url = "https://files.pythonhosted.org/packages/30/9d/2470df6aa146aff4c65fee0f87f58d2164a67533c771c9cc12ffcdb865d5/coverage-7.6.10-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:2396e8116db77789f819d2bc8a7e200232b7a282c66e0ae2d2cd84581a89757e", size = 237922 },
+    { url = "https://files.pythonhosted.org/packages/08/dd/723fef5d901e6a89f2507094db66c091449c8ba03272861eaefa773ad95c/coverage-7.6.10-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:79109c70cc0882e4d2d002fe69a24aa504dec0cc17169b3c7f41a1d341a73694", size = 239783 },
+    { url = "https://files.pythonhosted.org/packages/3d/f7/64d3298b2baf261cb35466000628706ce20a82d42faf9b771af447cd2b76/coverage-7.6.10-cp313-cp313-win32.whl", hash = "sha256:9e1747bab246d6ff2c4f28b4d186b205adced9f7bd9dc362051cc37c4a0c7bd6", size = 210965 },
+    { url = "https://files.pythonhosted.org/packages/d5/58/ec43499a7fc681212fe7742fe90b2bc361cdb72e3181ace1604247a5b24d/coverage-7.6.10-cp313-cp313-win_amd64.whl", hash = "sha256:254f1a3b1eef5f7ed23ef265eaa89c65c8c5b6b257327c149db1ca9d4a35f25e", size = 211719 },
+    { url = "https://files.pythonhosted.org/packages/ab/c9/f2857a135bcff4330c1e90e7d03446b036b2363d4ad37eb5e3a47bbac8a6/coverage-7.6.10-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:2ccf240eb719789cedbb9fd1338055de2761088202a9a0b73032857e53f612fe", size = 209050 },
+    { url = "https://files.pythonhosted.org/packages/aa/b3/f840e5bd777d8433caa9e4a1eb20503495709f697341ac1a8ee6a3c906ad/coverage-7.6.10-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:0c807ca74d5a5e64427c8805de15b9ca140bba13572d6d74e262f46f50b13273", size = 209321 },
+    { url = "https://files.pythonhosted.org/packages/85/7d/125a5362180fcc1c03d91850fc020f3831d5cda09319522bcfa6b2b70be7/coverage-7.6.10-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2bcfa46d7709b5a7ffe089075799b902020b62e7ee56ebaed2f4bdac04c508d8", size = 252039 },
+    { url = "https://files.pythonhosted.org/packages/a9/9c/4358bf3c74baf1f9bddd2baf3756b54c07f2cfd2535f0a47f1e7757e54b3/coverage-7.6.10-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4e0de1e902669dccbf80b0415fb6b43d27edca2fbd48c74da378923b05316098", size = 247758 },
+    { url = "https://files.pythonhosted.org/packages/cf/c7/de3eb6fc5263b26fab5cda3de7a0f80e317597a4bad4781859f72885f300/coverage-7.6.10-cp313-cp313t-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3f7b444c42bbc533aaae6b5a2166fd1a797cdb5eb58ee51a92bee1eb94a1e1cb", size = 250119 },
+    { url = "https://files.pythonhosted.org/packages/3e/e6/43de91f8ba2ec9140c6a4af1102141712949903dc732cf739167cfa7a3bc/coverage-7.6.10-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:b330368cb99ef72fcd2dc3ed260adf67b31499584dc8a20225e85bfe6f6cfed0", size = 249597 },
+    { url = "https://files.pythonhosted.org/packages/08/40/61158b5499aa2adf9e37bc6d0117e8f6788625b283d51e7e0c53cf340530/coverage-7.6.10-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:9a7cfb50515f87f7ed30bc882f68812fd98bc2852957df69f3003d22a2aa0abf", size = 247473 },
+    { url = "https://files.pythonhosted.org/packages/50/69/b3f2416725621e9f112e74e8470793d5b5995f146f596f133678a633b77e/coverage-7.6.10-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:6f93531882a5f68c28090f901b1d135de61b56331bba82028489bc51bdd818d2", size = 248737 },
+    { url = "https://files.pythonhosted.org/packages/3c/6e/fe899fb937657db6df31cc3e61c6968cb56d36d7326361847440a430152e/coverage-7.6.10-cp313-cp313t-win32.whl", hash = "sha256:89d76815a26197c858f53c7f6a656686ec392b25991f9e409bcef020cd532312", size = 211611 },
+    { url = "https://files.pythonhosted.org/packages/1c/55/52f5e66142a9d7bc93a15192eba7a78513d2abf6b3558d77b4ca32f5f424/coverage-7.6.10-cp313-cp313t-win_amd64.whl", hash = "sha256:54a5f0f43950a36312155dae55c505a76cd7f2b12d26abeebbe7a0b36dbc868d", size = 212781 },
+    { url = "https://files.pythonhosted.org/packages/40/41/473617aadf9a1c15bc2d56be65d90d7c29bfa50a957a67ef96462f7ebf8e/coverage-7.6.10-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:656c82b8a0ead8bba147de9a89bda95064874c91a3ed43a00e687f23cc19d53a", size = 207978 },
+    { url = "https://files.pythonhosted.org/packages/10/f6/480586607768b39a30e6910a3c4522139094ac0f1677028e1f4823688957/coverage-7.6.10-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:ccc2b70a7ed475c68ceb548bf69cec1e27305c1c2606a5eb7c3afff56a1b3b27", size = 208415 },
+    { url = "https://files.pythonhosted.org/packages/f1/af/439bb760f817deff6f4d38fe7da08d9dd7874a560241f1945bc3b4446550/coverage-7.6.10-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a5e37dc41d57ceba70956fa2fc5b63c26dba863c946ace9705f8eca99daecdc4", size = 236452 },
+    { url = "https://files.pythonhosted.org/packages/d0/13/481f4ceffcabe29ee2332e60efb52e4694f54a402f3ada2bcec10bb32e43/coverage-7.6.10-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0aa9692b4fdd83a4647eeb7db46410ea1322b5ed94cd1715ef09d1d5922ba87f", size = 234374 },
+    { url = "https://files.pythonhosted.org/packages/c5/59/4607ea9d6b1b73e905c7656da08d0b00cdf6e59f2293ec259e8914160025/coverage-7.6.10-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:aa744da1820678b475e4ba3dfd994c321c5b13381d1041fe9c608620e6676e25", size = 235505 },
+    { url = "https://files.pythonhosted.org/packages/85/60/d66365723b9b7f29464b11d024248ed3523ce5aab958e4ad8c43f3f4148b/coverage-7.6.10-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:c0b1818063dc9e9d838c09e3a473c1422f517889436dd980f5d721899e66f315", size = 234616 },
+    { url = "https://files.pythonhosted.org/packages/74/f8/2cf7a38e7d81b266f47dfcf137fecd8fa66c7bdbd4228d611628d8ca3437/coverage-7.6.10-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:59af35558ba08b758aec4d56182b222976330ef8d2feacbb93964f576a7e7a90", size = 233099 },
+    { url = "https://files.pythonhosted.org/packages/50/2b/bff6c1c6b63c4396ea7ecdbf8db1788b46046c681b8fcc6ec77db9f4ea49/coverage-7.6.10-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:7ed2f37cfce1ce101e6dffdfd1c99e729dd2ffc291d02d3e2d0af8b53d13840d", size = 234089 },
+    { url = "https://files.pythonhosted.org/packages/bf/b5/baace1c754d546a67779358341aa8d2f7118baf58cac235db457e1001d1b/coverage-7.6.10-cp39-cp39-win32.whl", hash = "sha256:4bcc276261505d82f0ad426870c3b12cb177752834a633e737ec5ee79bbdff18", size = 210701 },
+    { url = "https://files.pythonhosted.org/packages/b1/bf/9e1e95b8b20817398ecc5a1e8d3e05ff404e1b9fb2185cd71561698fe2a2/coverage-7.6.10-cp39-cp39-win_amd64.whl", hash = "sha256:457574f4599d2b00f7f637a0700a6422243b3565509457b2dbd3f50703e11f59", size = 211482 },
+    { url = "https://files.pythonhosted.org/packages/a1/70/de81bfec9ed38a64fc44a77c7665e20ca507fc3265597c28b0d989e4082e/coverage-7.6.10-pp39.pp310-none-any.whl", hash = "sha256:fd34e7b3405f0cc7ab03d54a334c17a9e802897580d964bd8c2001f4b9fd488f", size = 200223 },
+]
+
+[package.optional-dependencies]
+toml = [
+    { name = "tomli", marker = "python_full_version <= '3.11'" },
+]
+
+[[package]]
+name = "cramjam"
+version = "2.9.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/c9/68/09b6b5603d21a0c7d4362d513217a5079c47b1b7a88967c52dbef13db183/cramjam-2.9.1.tar.gz", hash = "sha256:336cc591d86cbd225d256813779f46624f857bc9c779db126271eff9ddc524ae", size = 47892 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/27/5d/0b03115fa6a95a6dd9be344cd186879b763f1a6fab57ae55ffe2777aa0a7/cramjam-2.9.1-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:8e82464d1e00fbbb12958999b8471ba5e9f3d9711954505a0a7b378762332e6f", size = 2136622 },
+    { url = "https://files.pythonhosted.org/packages/6f/ac/a17644e182ede7e8e24fb3af038bc2c1cf3dd0447c935cb10409f21d099b/cramjam-2.9.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6d2df8a6511cc08ef1fccd2e0c65e2ebc9f57574ec8376052a76851af5398810", size = 1927947 },
+    { url = "https://files.pythonhosted.org/packages/9e/1e/e6c4f9695e4ba7b9c63160dcbfa76428bd3221930eedeb8f16364ab6f642/cramjam-2.9.1-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:21ea784e6c3f1843d3523ae0f03651dd06058b39eeb64beb82ee3b100fa83662", size = 2268766 },
+    { url = "https://files.pythonhosted.org/packages/ab/37/4c81e5d039bdfc75a695abd426e6cdd9ab18a87f65d57837d78936cfa226/cramjam-2.9.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e0c5d98a4e791f0bbd0ffcb7dae879baeb2dcc357348a8dc2be0a8c10403a2a", size = 2108762 },
+    { url = "https://files.pythonhosted.org/packages/b9/bb/3bf3a8877b9a4105b625d710410bd2bc83ef38d4a7fe4eaeb3895d997b2d/cramjam-2.9.1-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e076fd87089197cb61117c63dbe7712ad5eccb93968860eb3bae09b767bac813", size = 2086694 },
+    { url = "https://files.pythonhosted.org/packages/c3/78/317b7ab6a9b0f24c45d56305a8288cdb6408f855034dc80530ed16a5cc6c/cramjam-2.9.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6d86b44933aea0151e4a2e1e6935448499849045c38167d288ca4c59d5b8cd4e", size = 2441698 },
+    { url = "https://files.pythonhosted.org/packages/c5/2d/bc98992c29eb8647196b3bda814fd7ecfba6aff85177d44180be2aa320e8/cramjam-2.9.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7eb032549dec897b942ddcf80c1cdccbcb40629f15fc902731dbe6362da49326", size = 2759280 },
+    { url = "https://files.pythonhosted.org/packages/dd/64/a4e54d74110c22477e467586935167d61fc7bae5284d393e76779b214a3e/cramjam-2.9.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cf29b4def86ec503e329fe138842a9b79a997e3beb6c7809b05665a0d291edff", size = 2385128 },
+    { url = "https://files.pythonhosted.org/packages/b0/1a/6ee093bf8a41cf31980175310abbbcdd1a39dadadbe96843112f42cef0fe/cramjam-2.9.1-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:a36adf7d13b7accfa206e1c917f08924eb905b45aa8e62176509afa7b14db71e", size = 2373494 },
+    { url = "https://files.pythonhosted.org/packages/9d/a6/1ae1f1a8ef559c2fab9d6d7f09b19995684e6727e617bf1b73967ee1c6be/cramjam-2.9.1-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:cf4ea758d98b6fad1b4b2d808d0de690d3162ac56c26968aea0af6524e3eb736", size = 2386900 },
+    { url = "https://files.pythonhosted.org/packages/d9/e6/cf18deeaa0a96e7fc87f0eacde3c97e2893b573ac148ec746655570c18fc/cramjam-2.9.1-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:4826d6d81ea490fa7a3ae7a4b9729866a945ffac1f77fe57b71e49d6e1b21efd", size = 2400609 },
+    { url = "https://files.pythonhosted.org/packages/90/97/98a8fa24249dc72a936a9a51a81407a399070ba4ceb528d0af291c760eff/cramjam-2.9.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:335103317475bf992953c58838152a4761fc3c87354000edbfc4d7e57cf05909", size = 2553159 },
+    { url = "https://files.pythonhosted.org/packages/ae/6b/4f71f72bc3405f221ec8bd2ba869e324d5f87ddd58c14bf59f7937ea37ab/cramjam-2.9.1-cp310-cp310-win32.whl", hash = "sha256:258120cb1e3afc3443f756f9de161ed63eed56a2c31f6093e81c571c0f2dc9f6", size = 1817873 },
+    { url = "https://files.pythonhosted.org/packages/8e/f4/32639916897d59e94d286b5b22263ce8c2903ecc93a868ebe9443ece8f12/cramjam-2.9.1-cp310-cp310-win_amd64.whl", hash = "sha256:c60e5996aa02547d12bc2740d44e90e006b0f93100f53206f7abe6732ad56e69", size = 2092168 },
+    { url = "https://files.pythonhosted.org/packages/6c/28/dd2b62be30ffe1fa8df10c99ba7b46abfbfb2fc6ace6acbbf9264a1a6b48/cramjam-2.9.1-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:b9db1debe48060e41a5b91af9193c524e473c57f6105462c5524a41f5aabdb88", size = 2136699 },
+    { url = "https://files.pythonhosted.org/packages/03/c9/fcebeb6f06879af4226337715fbc42ffe543158bcba8c244bba144767897/cramjam-2.9.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:f6f18f0242212d3409d26ce3874937b5b979cebd61f08b633a6ea893c32fc7b6", size = 1927934 },
+    { url = "https://files.pythonhosted.org/packages/e8/f3/77032e4f5db4dfcc2b0365f92655b7d6f3fc1527ea5b637f9fb9f8156a65/cramjam-2.9.1-cp311-cp311-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b5b1cd7d39242b2b903cf09cd4696b3a6e04dc537ffa9f3ac8668edae76eecb6", size = 2268584 },
+    { url = "https://files.pythonhosted.org/packages/38/16/52175e94390f57196382783a3386c122ace7656b57339abaacdc9433b609/cramjam-2.9.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a47de0a68f5f4d9951250ef5af31f2a7228132caa9ed60994234f7eb98090d33", size = 2108599 },
+    { url = "https://files.pythonhosted.org/packages/99/25/5f7476d127a8d18cd19a2f3fd25c0fe09ef7848069d23aac70bc96385eb6/cramjam-2.9.1-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e13c9a697881e5e38148958612dc6856967f5ff8cd7bba5ff751f2d6ac020aa4", size = 2086632 },
+    { url = "https://files.pythonhosted.org/packages/7b/97/76ff3e1209add6acb7e2aa7997be48dc1f92ad66ee3e8fa1179eb2bb9b44/cramjam-2.9.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ba560244bc1335b420b74e91e35f9d4e7f307a3be3a4603ce0f0d7e15a0acdf0", size = 2441757 },
+    { url = "https://files.pythonhosted.org/packages/69/c4/228e74c30576556d11e54d86f356955cd86ff5e11bbfec74b66ed0dd237d/cramjam-2.9.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8d47fd41ce260cf4f0ff0e788de961fab9e9c6844a05ce55d06ce31e06107bdc", size = 2758144 },
+    { url = "https://files.pythonhosted.org/packages/4b/e7/0fd22e12c6a2879abc501979779d4b8cfe8fe692c708c2c0d1664e88fd79/cramjam-2.9.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:84d154fbadece82935396eb6bcb502085d944d2fd13b07a94348364344370c2c", size = 2385062 },
+    { url = "https://files.pythonhosted.org/packages/dd/9c/845592ddf9eb7130ae8bc5958a01d469304a43f8071effe164e2d239e3fa/cramjam-2.9.1-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:038df668ffb94d64d67b6ecc59cbd206745a425ffc0402897dde12d89fa6a870", size = 2373473 },
+    { url = "https://files.pythonhosted.org/packages/10/c2/287cc94b7f8e87e3b0c21819d3a5deead99ebfdcb2b2d85cd04011b37292/cramjam-2.9.1-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:4125d8cd86fa08495d310e80926c2f0563f157b76862e7479f9b2cf94823ea0c", size = 2386816 },
+    { url = "https://files.pythonhosted.org/packages/7c/22/869a1eeea53db4d9fbde6693a2465909762bffeab1a671e193c95b26f99f/cramjam-2.9.1-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:4206ebdd1d1ef0f3f86c8c2f7c426aa4af6094f4f41e274601fd4c4569f37454", size = 2400713 },
+    { url = "https://files.pythonhosted.org/packages/3f/89/ff988bd6427f01041ccb1a9104c05b6373ae476682d317b6844f4b40af92/cramjam-2.9.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:ab687bef5c493732b9a4ab870542ee43f5eae0025f9c684c7cb399c3a85cb380", size = 2553081 },
+    { url = "https://files.pythonhosted.org/packages/2e/68/13fa8561335de609f3cd40b132c1a3abbaf26d3c277e8b8a7446de34ef2c/cramjam-2.9.1-cp311-cp311-win32.whl", hash = "sha256:dda7698b6d7caeae1047adafebc4b43b2a82478234f6c2b45bc3edad854e0600", size = 1817782 },
+    { url = "https://files.pythonhosted.org/packages/94/75/f3506ee802460e3b86a91e53bba1f67cf457fa04e4316fe7d5823ba5d28b/cramjam-2.9.1-cp311-cp311-win_amd64.whl", hash = "sha256:872b00ff83e84bcbdc7e951af291ebe65eed20b09c47e7c4af21c312f90b796f", size = 2092227 },
+    { url = "https://files.pythonhosted.org/packages/56/66/69a1c17331e38b02c78c923262fc315272de7c2618ef7eac8b3358969d90/cramjam-2.9.1-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:79417957972553502b217a0093532e48893c8b4ca30ccc941cefe9c72379df7c", size = 2132273 },
+    { url = "https://files.pythonhosted.org/packages/3d/17/23d0b1d3301480e924545cdd27f2b949c50438949f64c74e800a09c12c37/cramjam-2.9.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce2b94117f373defc876f88e74e44049a9969223dbca3240415b71752d0422fb", size = 1926919 },
+    { url = "https://files.pythonhosted.org/packages/8e/da/e9565f4abbbaa14645ccd7ce83f9631e90955454b87dc3ef9208aebc72e6/cramjam-2.9.1-cp312-cp312-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:67040e0fd84404885ec716a806bee6110f9960c3647e0ef1670aab3b7375a70a", size = 2271776 },
+    { url = "https://files.pythonhosted.org/packages/88/ac/e6e0794ac01deb52e7a6a3e59720699abdee08d9b9c63a8d8874201d8155/cramjam-2.9.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0bedb84e068b53c944bd08dcb501fd00d67daa8a917922356dd559b484ce7eab", size = 2109248 },
+    { url = "https://files.pythonhosted.org/packages/22/0f/c3724b2dcdfbe7e07917803cf7a6db4a874818a6f8d2b95ca1ceaf177170/cramjam-2.9.1-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:06e3f97a379386d97debf08638a78b3d3850fdf6124755eb270b54905a169930", size = 2088611 },
+    { url = "https://files.pythonhosted.org/packages/ce/16/929a5ae899ad6298f58e66622dc223476fe8e1d4e8dae608f4e1a34bfd09/cramjam-2.9.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:11118675e9c7952ececabc62f023290ee4f8ecf0bee0d2c7eb8d1c402ee9769d", size = 2438373 },
+    { url = "https://files.pythonhosted.org/packages/2a/2a/ad473f1ca65d3285e8c1d99fc0289f5856224c0d452dabcf856fd4dcdd77/cramjam-2.9.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6b7de6b61b11545570e4d6033713f3599525efc615ee353a822be8f6b0c65b77", size = 2836669 },
+    { url = "https://files.pythonhosted.org/packages/9b/5a/e9b4868ee27099a2a21646cf5ea5cf08c660eae90b55a395ada974dcf3fb/cramjam-2.9.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:57ca8f3775324a9de3ee6f05ca172687ba258c0dea79f7e3a6b4112834982f2a", size = 2343995 },
+    { url = "https://files.pythonhosted.org/packages/5f/c4/870a9b4524107bf85a207b82a42613318881238b20f2d237e62815af646a/cramjam-2.9.1-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:9847dd6f288f1c56359f52acb48ff2df848ff3e3bff34d23855bbcf7016427cc", size = 2374270 },
+    { url = "https://files.pythonhosted.org/packages/70/4b/b69e8e3951b7cec5e7da2539b7573bb396bed66af07d760b1878b00fd120/cramjam-2.9.1-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:8d1248dfa7f151e893ce819670f00879e4b7650b8d4c01279ce4f12140d68dd2", size = 2388789 },
+    { url = "https://files.pythonhosted.org/packages/05/1a/af02f6192060413314735c0db61259d7279b0d8d99eee29eff2af09c5892/cramjam-2.9.1-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:9da6d970281083bae91b914362de325414aa03c01fc806f6bb2cc006322ec834", size = 2402459 },
+    { url = "https://files.pythonhosted.org/packages/20/9a/a4ab3e90d72eb4f2c1b983fa32b4050ba676f533ba15bd78158f0632295a/cramjam-2.9.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:1c33bc095db5733c841a102b8693062be5db8cdac17b9782ebc00577c6a94480", size = 2518440 },
+    { url = "https://files.pythonhosted.org/packages/35/3b/e632dd7e2c5c8a2af2d83144b00d6840f1afcf9c6959ed59ec5b0f925288/cramjam-2.9.1-cp312-cp312-win32.whl", hash = "sha256:9e9193cd4bb57e7acd3af24891526299244bfed88168945efdaa09af4e50720f", size = 1822630 },
+    { url = "https://files.pythonhosted.org/packages/0e/a2/d1c46618b81b83578d58a62f3709046c4f3b4ddba10df4b9797cfe096b98/cramjam-2.9.1-cp312-cp312-win_amd64.whl", hash = "sha256:15955dd75e80f66c1ea271167a5347661d9bdc365f894a57698c383c9b7d465c", size = 2094684 },
+    { url = "https://files.pythonhosted.org/packages/85/45/f1d1e6ffdceb3b0c18511df2f8e779e03972459fb71d7c1ab0f6a5c063a3/cramjam-2.9.1-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:5a7797a2fff994fc5e323f7a967a35a3e37e3006ed21d64dcded086502f482af", size = 2131814 },
+    { url = "https://files.pythonhosted.org/packages/3a/96/36bbd431fbf0fa2ff51fd2db4c3bead66e9e373693a8455d411d45125a68/cramjam-2.9.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d51b9b140b1df39a44bff7896d98a10da345b7d5f5ce92368d328c1c2c829167", size = 1926380 },
+    { url = "https://files.pythonhosted.org/packages/67/c4/99b6507ec697d5f56d32c9c04614775004b05b7fa870725a492dc6b639eb/cramjam-2.9.1-cp313-cp313-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:07ac76b7f992556e7aa910244be11ece578cdf84f4d5d5297461f9a895e18312", size = 2271581 },
+    { url = "https://files.pythonhosted.org/packages/cb/1b/6d55dff244fb22c0b686dd5a96a754c0638f8a94056beb27c457c6035cc5/cramjam-2.9.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d90a72608c7550cd7eba914668f6277bfb0b24f074d1f1bd9d061fcb6f2adbd6", size = 2109255 },
+    { url = "https://files.pythonhosted.org/packages/ca/fb/b9fcf492a21a8d978c6f999025fce2c6656399448c017ed2fc859425f37f/cramjam-2.9.1-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:56495975401b1821dbe1f29cf222e23556232209a2fdb809fe8156d120ca9c7f", size = 2088323 },
+    { url = "https://files.pythonhosted.org/packages/88/1f/69b523395aeaa201dbd53d203453288205a0c651e7c910161892d694eb4d/cramjam-2.9.1-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3b695259e71fde6d5be66b77a4474523ced9ffe9fe8a34cb9b520ec1241a14d3", size = 2437930 },
+    { url = "https://files.pythonhosted.org/packages/b0/2c/d07e802f1786c4082e8286db1087563e4fab31cd6534ed31523f1f9584d1/cramjam-2.9.1-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ab1e69dc4831bbb79b6d547077aae89074c83e8ad94eba1a3d80e94d2424fd02", size = 2836655 },
+    { url = "https://files.pythonhosted.org/packages/1f/f5/6b425e82395c078bc95a7437b685e6bdba39d28c2b2986d79374fc1681aa/cramjam-2.9.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:440b489902bfb7a26d3fec1ca888007615336ff763d2a32a2fc40586548a0dbf", size = 2387107 },
+    { url = "https://files.pythonhosted.org/packages/33/65/7bf97d89ba7607aaea5464af6f249e3d94c291acf73d72768367a3e361c0/cramjam-2.9.1-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:217fe22b41f8c3dce03852f828b059abfad11d1344a1df2f43d3eb8634b18d75", size = 2374006 },
+    { url = "https://files.pythonhosted.org/packages/29/11/8b6c82eda6d0affbc15d7ab4dc758856eb4308e8ddae73300c1648f5aa0f/cramjam-2.9.1-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:95f3646ddc98af25af25d5692ae65966488a283813336ea9cf41b22e542e7c0d", size = 2388731 },
+    { url = "https://files.pythonhosted.org/packages/48/25/6cdd57c0b1a83c98aec9029310d09a6c1a31e9e9fb8efd9001bd0cbea992/cramjam-2.9.1-cp313-cp313-musllinux_1_1_i686.whl", hash = "sha256:6b19fc60ead1cae9795a5b359599da3a1c95d38f869bdfb51c441fd76b04e926", size = 2402131 },
+    { url = "https://files.pythonhosted.org/packages/b4/e7/cbf80c9647fa582432aa833c4bdd20cf437917c8066ce653e3b78deff658/cramjam-2.9.1-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:8dc5207567459d049696f62a1fdfb220f3fe6aa0d722285d44753e12504dac6c", size = 2555296 },
+    { url = "https://files.pythonhosted.org/packages/18/a6/fabe1959a980f5d2783a6c138311509dd168bd76e62018624a91cd1cbb41/cramjam-2.9.1-cp313-cp313-win32.whl", hash = "sha256:fbfe35929a61b914de9e5dbacde0cfbba86cbf5122f9285a24c14ed0b645490b", size = 1822484 },
+    { url = "https://files.pythonhosted.org/packages/55/d5/24e4562771711711c466768c92097640ed97b0283abe9043ffb6c6d4cf04/cramjam-2.9.1-cp313-cp313-win_amd64.whl", hash = "sha256:06068bd191a82ad4fc1ac23d6f8627fb5e37ec4be0431711b9a2dbacaccfeddb", size = 2094445 },
+    { url = "https://files.pythonhosted.org/packages/c7/5a/50523fd478390acb6ca8e57239f7cf79f7260dc0d16be89137d47823e50a/cramjam-2.9.1-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:af39006faddfc6253beb93ca821d544931cfee7f0177b99ff106dfd8fd6a2cd8", size = 2137158 },
+    { url = "https://files.pythonhosted.org/packages/df/83/54eca302e431d51149074d8aad6ec588870c5797060e2142dfe6ca3599a8/cramjam-2.9.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:b3291be0d3f73d5774d69013be4ab33978c777363b5312d14f62f77817c2f75a", size = 1927910 },
+    { url = "https://files.pythonhosted.org/packages/6d/e9/5d38ffa5376c5bffcbd16545707d9dac6beffccd00410f0cc19d83d85ef7/cramjam-2.9.1-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:1539fd758f0e57fad7913cebff8baaee871bb561ddf6fa710a427b74da6b6778", size = 2269458 },
+    { url = "https://files.pythonhosted.org/packages/15/f3/99fedc4210db1967256e602fdcb60947585421fd659f8baeeeb4ea16e4c7/cramjam-2.9.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ff362f68bd68ac0eccb445209238d589bba728fb6d7f2e9dc199e0ec3a61d6e0", size = 2109406 },
+    { url = "https://files.pythonhosted.org/packages/f2/e9/f380e0c1bd03046c522da4fd6d43ea897ba0b832c78fc4ea5708d8c35c21/cramjam-2.9.1-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:23b9786d1d17686fb8d600ade2a19374c7188d4b8867efa9af0d8274a220aec7", size = 2086677 },
+    { url = "https://files.pythonhosted.org/packages/13/a7/3ae887753f6d41f6e4af8e25654d103c56e13dda2f4b4d13acac570c65c1/cramjam-2.9.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8bc9c2c748aaf91863d89c4583f529c1c709485c94f8dfeb3ee48662d88e3258", size = 2442136 },
+    { url = "https://files.pythonhosted.org/packages/de/a2/763fd98340936057e44ea0b870c9cdb87ad5f90d49e492e8a11cf74e7b29/cramjam-2.9.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fd0fa9a0e7f18224b6d2d1d69dbdc3aecec80ef1393c59244159b131604a4395", size = 2754985 },
+    { url = "https://files.pythonhosted.org/packages/33/31/7c8cdf6b16fcd46bad4a307c8203a58b7a2fddf6cb3aad9dc441c050f62f/cramjam-2.9.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2ceef6e09ee22457997370882aa3c69de01e6dd0aaa2f953e1e87ad11641d042", size = 2385597 },
+    { url = "https://files.pythonhosted.org/packages/dd/ba/ec0f3b5a3a90721bdb42f4f4989b60adf823d137f40365e83df0cd299378/cramjam-2.9.1-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:1376f6fdbf0b30712413a0b4e51663a4938ae2f6b449f8e4635dbb3694db83cf", size = 2374339 },
+    { url = "https://files.pythonhosted.org/packages/ff/0a/f5bccdc8d12821aed4473a427e9eb8282a38c9337a30e02ed102b18941bf/cramjam-2.9.1-cp39-cp39-musllinux_1_1_armv7l.whl", hash = "sha256:342fb946f8d3e9e35b837288b03ab23cfbe0bb5a30e582ed805ef79706823a96", size = 2386933 },
+    { url = "https://files.pythonhosted.org/packages/a0/6e/ce3ffad2b3b8cb73156a19345e27a2e27fb5be79b64f2c81b0c6d6e16c57/cramjam-2.9.1-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:a237064a6e2c2256c9a1cf2beb7c971382190c0f1eb2e810e02e971881756132", size = 2400860 },
+    { url = "https://files.pythonhosted.org/packages/32/a9/e4509e5dfc8f41d9e7f9fdddbf567967937303621d410197c86b11d6a3e4/cramjam-2.9.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:53145fc9f2319c1245d4329e1da8cfacd6e35e27090c07c0b9d453ae2bbdac3e", size = 2553681 },
+    { url = "https://files.pythonhosted.org/packages/0a/83/52401c5c654ddff2850d890b0f1cfc355ff6887c6def420d0c8d8178ff97/cramjam-2.9.1-cp39-cp39-win32.whl", hash = "sha256:8a9f52c27292c21457f43c4ce124939302a9acfb62295e7cda8667310563a5a3", size = 1818130 },
+    { url = "https://files.pythonhosted.org/packages/93/b3/1645986d8b915fd0426a7224cd00c2c17c32b4d69bc5faad3fb3f5fd5081/cramjam-2.9.1-cp39-cp39-win_amd64.whl", hash = "sha256:8097ee39b61c86848a443c0b25b2df1de6b331fd512b20836a4f5cfde51ab255", size = 2092440 },
+    { url = "https://files.pythonhosted.org/packages/bc/91/3f7884172573072a4280bc8bc19b7562b2cd66d2a65576b11e72115cd5fe/cramjam-2.9.1-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:86824c695688fcd06c5ac9bbd3fea9bdfb4cca194b1e706fbf11a629df48d2b4", size = 2159537 },
+    { url = "https://files.pythonhosted.org/packages/ef/49/a0a89e9c45413e89a1e408d4ab416c0f88f19f6db7571fd5c517e429e276/cramjam-2.9.1-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:27571bfa5a5d618604696747d0dc1d2a99b5906c967c8dee53c13a7107edfde6", size = 1936244 },
+    { url = "https://files.pythonhosted.org/packages/26/f7/6422b9e4d148f1a351c0358a95d59023f25cab76609b180804f6a3ed17e9/cramjam-2.9.1-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fb01f6e38719818778144d3165a89ea1ad9dc58c6342b7f20aa194c70f34cbd1", size = 2119487 },
+    { url = "https://files.pythonhosted.org/packages/b5/59/6fc930217f7ae085eca6d22d3477cd0145a105cdc39e63b834cb0c1b25e3/cramjam-2.9.1-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b5cef5cf40725fe64592af9ec163e7389855077700678a1d94bec549403a74d", size = 2400910 },
+    { url = "https://files.pythonhosted.org/packages/2d/36/7e53cf5aaed4b446490e298f7571e69ce15d0dfb148feabe8bf02e58827f/cramjam-2.9.1-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:ac48b978aa0675f62b642750e798c394a64d25ce852e4e541f69bef9a564c2f0", size = 2100860 },
+]
+
+[[package]]
+name = "datasets"
+version = "3.2.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "aiohttp" },
+    { name = "dill" },
+    { name = "filelock" },
+    { name = "fsspec", extra = ["http"] },
+    { name = "huggingface-hub" },
+    { name = "multiprocess" },
+    { name = "numpy" },
+    { name = "packaging" },
+    { name = "pandas" },
+    { name = "pyarrow" },
+    { name = "pyyaml" },
+    { name = "requests" },
+    { name = "tqdm" },
+    { name = "xxhash" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fc/48/744286c044e2b942d4fa67f92816126522ad1f0675def0ea3264e6242005/datasets-3.2.0.tar.gz", hash = "sha256:9a6e1a356052866b5dbdd9c9eedb000bf3fc43d986e3584d9b028f4976937229", size = 558366 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d7/84/0df6c5981f5fc722381662ff8cfbdf8aad64bec875f75d80b55bfef394ce/datasets-3.2.0-py3-none-any.whl", hash = "sha256:f3d2ba2698b7284a4518019658596a6a8bc79f31e51516524249d6c59cf0fe2a", size = 480647 },
+]
+
+[[package]]
+name = "diff-cover"
+version = "9.2.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.9.17'",
+]
+dependencies = [
+    { name = "chardet", marker = "python_full_version < '3.9.17'" },
+    { name = "jinja2", marker = "python_full_version < '3.9.17'" },
+    { name = "pluggy", marker = "python_full_version < '3.9.17'" },
+    { name = "pygments", marker = "python_full_version < '3.9.17'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/44/3a/e49ccba052a4dda264fbad4f467739ecc63498f7223bfc03d4bfac23ea95/diff_cover-9.2.0.tar.gz", hash = "sha256:85a0b353ebbb678f9e87ea303f75b545bd0baca38f563219bb72f2ae862bba36", size = 94857 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b4/8a/bddb8e4aea550066559144e72d3566e9ae2f757b8ac154e769c563f48177/diff_cover-9.2.0-py3-none-any.whl", hash = "sha256:1e24edc51c39e810c47dd9986e76c333ed95859655c091f572e590c39cabbdbe", size = 52561 },
+]
+
+[[package]]
+name = "diff-cover"
+version = "9.2.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.13'",
+    "python_full_version == '3.12.*'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "chardet", marker = "python_full_version >= '3.9.17'" },
+    { name = "jinja2", marker = "python_full_version >= '3.9.17'" },
+    { name = "pluggy", marker = "python_full_version >= '3.9.17'" },
+    { name = "pygments", marker = "python_full_version >= '3.9.17'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/36/b0/f3ccf97926f6e5cc76d5ece42f3c685d75673d1886fcec62886b8b00c51a/diff_cover-9.2.1.tar.gz", hash = "sha256:5fa5b2d71ccf5d16cd222a71c2ca069d9bf5fa3d657f6fac9b4d9c23379323bf", size = 94964 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e2/67/5785b653e392a17bec8a340fc5f4f9278ff3588098cc96b1b3b8f9c3f167/diff_cover-9.2.1-py3-none-any.whl", hash = "sha256:97393d4acaa7545eae86f66e287ad96ffbeef70c188cca5a7f433983363e3849", size = 52560 },
+]
+
+[[package]]
+name = "dill"
+version = "0.3.8"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/17/4d/ac7ffa80c69ea1df30a8aa11b3578692a5118e7cd1aa157e3ef73b092d15/dill-0.3.8.tar.gz", hash = "sha256:3ebe3c479ad625c4553aca177444d89b486b1d84982eeacded644afc0cf797ca", size = 184847 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c9/7a/cef76fd8438a42f96db64ddaa85280485a9c395e7df3db8158cfec1eee34/dill-0.3.8-py3-none-any.whl", hash = "sha256:c36ca9ffb54365bdd2f8eb3eff7d2a21237f8452b57ace88b1ac615b7e815bd7", size = 116252 },
+]
+
+[[package]]
+name = "diskcache"
+version = "5.6.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3f/21/1c1ffc1a039ddcc459db43cc108658f32c57d271d7289a2794e401d0fdb6/diskcache-5.6.3.tar.gz", hash = "sha256:2c3a3fa2743d8535d832ec61c2054a1641f41775aa7c556758a109941e33e4fc", size = 67916 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3f/27/4570e78fc0bf5ea0ca45eb1de3818a23787af9b390c0b0a0033a1b8236f9/diskcache-5.6.3-py3-none-any.whl", hash = "sha256:5e31b2d5fbad117cc363ebaf6b689474db18a1f6438bc82358b024abd4c2ca19", size = 45550 },
+]
+
+[[package]]
+name = "distlib"
+version = "0.3.9"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0d/dd/1bec4c5ddb504ca60fc29472f3d27e8d4da1257a854e1d96742f15c1d02d/distlib-0.3.9.tar.gz", hash = "sha256:a60f20dea646b8a33f3e7772f74dc0b2d0772d2837ee1342a00645c81edf9403", size = 613923 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/91/a1/cf2472db20f7ce4a6be1253a81cfdf85ad9c7885ffbed7047fb72c24cf87/distlib-0.3.9-py2.py3-none-any.whl", hash = "sha256:47f8c22fd27c27e25a65601af709b38e4f0a45ea4fc2e710f65755fa8caaaf87", size = 468973 },
+]
+
+[[package]]
+name = "distro"
+version = "1.9.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/fc/f8/98eea607f65de6527f8a2e8885fc8015d3e6f5775df186e443e0964a11c3/distro-1.9.0.tar.gz", hash = "sha256:2fa77c6fd8940f116ee1d6b94a2f90b13b5ea8d019b98bc8bafdcabcdd9bdbed", size = 60722 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/12/b3/231ffd4ab1fc9d679809f356cebee130ac7daa00d6d6f3206dd4fd137e9e/distro-1.9.0-py3-none-any.whl", hash = "sha256:7bffd925d65168f85027d8da9af6bddab658135b840670a223589bc0c8ef02b2", size = 20277 },
+]
+
+[[package]]
+name = "exceptiongroup"
+version = "1.2.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/09/35/2495c4ac46b980e4ca1f6ad6db102322ef3ad2410b79fdde159a4b0f3b92/exceptiongroup-1.2.2.tar.gz", hash = "sha256:47c2edf7c6738fafb49fd34290706d1a1a2f4d1c6df275526b62cbb4aa5393cc", size = 28883 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/02/cc/b7e31358aac6ed1ef2bb790a9746ac2c69bcb3c8588b41616914eb106eaf/exceptiongroup-1.2.2-py3-none-any.whl", hash = "sha256:3111b9d131c238bec2f8f516e123e14ba243563fb135d3fe885990585aa7795b", size = 16453 },
+]
+
+[[package]]
+name = "exllamav2"
+version = "0.2.7"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "fastparquet" },
+    { name = "ninja" },
+    { name = "numpy" },
+    { name = "pandas" },
+    { name = "pillow" },
+    { name = "pygments" },
+    { name = "regex" },
+    { name = "rich" },
+    { name = "safetensors" },
+    { name = "sentencepiece" },
+    { name = "torch" },
+    { name = "websockets" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b5/04/300012908311c9a904d44c74a037821e7cde530c648afff4ad6bed4e22c3/exllamav2-0.2.7-py3-none-any.whl", hash = "sha256:410d1a69fed329f6f1c10de6988a3b6b33e1fa715aa4fe32c545edd1abd1254b", size = 1512625 },
+]
+
+[[package]]
+name = "fastapi"
+version = "0.115.6"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+    { name = "starlette" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/93/72/d83b98cd106541e8f5e5bfab8ef2974ab45a62e8a6c5b5e6940f26d2ed4b/fastapi-0.115.6.tar.gz", hash = "sha256:9ec46f7addc14ea472958a96aae5b5de65f39721a46aaf5705c480d9a8b76654", size = 301336 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/52/b3/7e4df40e585df024fac2f80d1a2d579c854ac37109675db2b0cc22c0bb9e/fastapi-0.115.6-py3-none-any.whl", hash = "sha256:e9240b29e36fa8f4bb7290316988e90c381e5092e0cbe84e7818cc3713bcf305", size = 94843 },
+]
+
+[[package]]
+name = "fastparquet"
+version = "2024.11.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cramjam" },
+    { name = "fsspec" },
+    { name = "numpy" },
+    { name = "packaging" },
+    { name = "pandas" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b4/66/862da14f5fde4eff2cedc0f51a8dc34ba145088e5041b45b2d57ac54f922/fastparquet-2024.11.0.tar.gz", hash = "sha256:e3b1fc73fd3e1b70b0de254bae7feb890436cb67e99458b88cb9bd3cc44db419", size = 467192 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3d/56/476f5b83476a256489879b78513bee737691a80905e246a2daa30ebcc362/fastparquet-2024.11.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:60ccf587410f0979105e17036df61bb60e1c2b81880dc91895cdb4ee65b71e7f", size = 910272 },
+    { url = "https://files.pythonhosted.org/packages/3b/ad/4ce73440df874479f7205fe5445090f71ed4e9bd77fdb3b740253ce82703/fastparquet-2024.11.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a5ad5fc14b0567e700bea3cd528a0bd45a6f9371370b49de8889fb3d10a6574a", size = 684095 },
+    { url = "https://files.pythonhosted.org/packages/20/37/c3164261d6183d529a59afef2749821b262c8581d837faa91043837c6f76/fastparquet-2024.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0b74333914f454344458dab9d1432fda9b70d62e28dc7acb1512d937ef1424ee", size = 1700355 },
+    { url = "https://files.pythonhosted.org/packages/e6/95/cf4b175c22160ec21e4664830763bfaa80b2cf05133ef854c3f436d01c16/fastparquet-2024.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41d1610130b5cb1ce36467766191c5418cba8631e2bfe3affffaf13f9be4e7a8", size = 1714663 },
+    { url = "https://files.pythonhosted.org/packages/2c/31/b6c8cdb6d5df964a192e4e8c8ecd979718afb9ca7e2dc9243a4368b370e9/fastparquet-2024.11.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d281edd625c33628ba028d3221180283d6161bc5ceb55eae1f0ca1678f864f26", size = 1666729 },
+    { url = "https://files.pythonhosted.org/packages/31/e5/8a0575c46a7973849f8f2a88af16618b9c7efe98f249f03e3e3de69c2b86/fastparquet-2024.11.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fa56b19a29008c34cfe8831e810f770080debcbffc69aabd1df4d47572181f9c", size = 1741669 },
+    { url = "https://files.pythonhosted.org/packages/bb/6a/669f8c9cf2fc6e30c9353832f870e5a2e170b458d12c5080837f742d963d/fastparquet-2024.11.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5914ecfa766b7763201b9f49d832a5e89c2dccad470ca4f9c9b228d9a8349756", size = 1782359 },
+    { url = "https://files.pythonhosted.org/packages/70/c0/1374cb43924739f4542e39d972481c1f4c7dd96808a1947450808e4e7df7/fastparquet-2024.11.0-cp310-cp310-win_amd64.whl", hash = "sha256:561202e8f0e859ccc1aa77c4aaad1d7901b2d50fd6f624ca018bae4c3c7a62ce", size = 670700 },
+    { url = "https://files.pythonhosted.org/packages/7c/51/e0d6e702523ac923ede6c05e240f4a02533ccf2cea9fec7a43491078e920/fastparquet-2024.11.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:374cdfa745aa7d5188430528d5841cf823eb9ad16df72ad6dadd898ccccce3be", size = 909934 },
+    { url = "https://files.pythonhosted.org/packages/0a/c8/5c0fb644c19a8d80b2ae4d8aa7d90c2d85d0bd4a948c5c700bea5c2802ea/fastparquet-2024.11.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4c8401bfd86cccaf0ab7c0ade58c91ae19317ff6092e1d4ad96c2178197d8124", size = 683844 },
+    { url = "https://files.pythonhosted.org/packages/33/4a/1e532fd1a0d4d8af7ffc7e3a8106c0bcd13ed914a93a61e299b3832dd3d2/fastparquet-2024.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f9cca4c6b5969df5561c13786f9d116300db1ec22c7941e237cfca4ce602f59b", size = 1791698 },
+    { url = "https://files.pythonhosted.org/packages/8d/e8/e1ede861bea68394a755d8be1aa2e2d60a3b9f6b551bfd56aeca74987e2e/fastparquet-2024.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9a9387e77ac608d8978774caaf1e19de67eaa1386806e514dcb19f741b19cfe5", size = 1804289 },
+    { url = "https://files.pythonhosted.org/packages/4f/1e/957090cccaede805583ca3f3e46e2762d0f9bf8860ecbce65197e47d84c1/fastparquet-2024.11.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6595d3771b3d587a31137e985f751b4d599d5c8e9af9c4858e373fdf5c3f8720", size = 1753638 },
+    { url = "https://files.pythonhosted.org/packages/85/72/344787c685fd1531f07ae712a855a7c34d13deaa26c3fd4a9231bea7dbab/fastparquet-2024.11.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:053695c2f730b78a2d3925df7cd5c6444d6c1560076af907993361cc7accf3e2", size = 1814407 },
+    { url = "https://files.pythonhosted.org/packages/6c/ec/ab9d5685f776a1965797eb68c4364c72edf57cd35beed2df49b34425d1df/fastparquet-2024.11.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0a52eecc6270ae15f0d51347c3f762703dd667ca486f127dc0a21e7e59856ae5", size = 1874462 },
+    { url = "https://files.pythonhosted.org/packages/90/4f/7a4ea9a7ddf0a3409873f0787f355806f9e0b73f42f2acecacdd9a8eff0a/fastparquet-2024.11.0-cp311-cp311-win_amd64.whl", hash = "sha256:e29ff7a367fafa57c6896fb6abc84126e2466811aefd3e4ad4070b9e18820e54", size = 671023 },
+    { url = "https://files.pythonhosted.org/packages/08/76/068ac7ec9b4fc783be21a75a6a90b8c0654da4d46934d969e524ce287787/fastparquet-2024.11.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:dbad4b014782bd38b58b8e9f514fe958cfa7a6c4e187859232d29fd5c5ddd849", size = 915968 },
+    { url = "https://files.pythonhosted.org/packages/c7/9e/6d3b4188ad64ed51173263c07109a5f18f9c84a44fa39ab524fca7420cda/fastparquet-2024.11.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:403d31109d398b6be7ce84fa3483fc277c6a23f0b321348c0a505eb098a041cb", size = 685399 },
+    { url = "https://files.pythonhosted.org/packages/8f/6c/809220bc9fbe83d107df2d664c3fb62fb81867be8f5218ac66c2e6b6a358/fastparquet-2024.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cbbb9057a26acf0abad7adf58781ee357258b7708ee44a289e3bee97e2f55d42", size = 1758557 },
+    { url = "https://files.pythonhosted.org/packages/e0/2c/b3b3e6ca2e531484289024138cd4709c22512b3fe68066d7f9849da4a76c/fastparquet-2024.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:63e0e416e25c15daa174aad8ba991c2e9e5b0dc347e5aed5562124261400f87b", size = 1781052 },
+    { url = "https://files.pythonhosted.org/packages/21/fe/97ed45092d0311c013996dae633122b7a51c5d9fe8dcbc2c840dc491201e/fastparquet-2024.11.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0e2d7f02f57231e6c86d26e9ea71953737202f20e948790e5d4db6d6a1a150dc", size = 1715797 },
+    { url = "https://files.pythonhosted.org/packages/24/df/02fa6aee6c0d53d1563b5bc22097076c609c4c5baa47056b0b4bed456fcf/fastparquet-2024.11.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:fbe4468146b633d8f09d7b196fea0547f213cb5ce5f76e9d1beb29eaa9593a93", size = 1795682 },
+    { url = "https://files.pythonhosted.org/packages/b0/25/f4f87557589e1923ee0e3bebbc84f08b7c56962bf90f51b116ddc54f2c9f/fastparquet-2024.11.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:29d5c718817bcd765fc519b17f759cad4945974421ecc1931d3bdc3e05e57fa9", size = 1857842 },
+    { url = "https://files.pythonhosted.org/packages/b1/f9/98cd0c39115879be1044d59c9b76e8292776e99bb93565bf990078fd11c4/fastparquet-2024.11.0-cp312-cp312-win_amd64.whl", hash = "sha256:74a0b3c40ab373442c0fda96b75a36e88745d8b138fcc3a6143e04682cbbb8ca", size = 673269 },
+    { url = "https://files.pythonhosted.org/packages/47/e3/e7db38704be5db787270d43dde895eaa1a825ab25dc245e71df70860ec12/fastparquet-2024.11.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:59e5c5b51083d5b82572cdb7aed0346e3181e3ac9d2e45759da2e804bdafa7ee", size = 912523 },
+    { url = "https://files.pythonhosted.org/packages/d3/66/e3387c99293dae441634e7724acaa425b27de19a00ee3d546775dace54a9/fastparquet-2024.11.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:bdadf7b6bad789125b823bfc5b0a719ba5c4a2ef965f973702d3ea89cff057f6", size = 683779 },
+    { url = "https://files.pythonhosted.org/packages/0a/21/d112d0573d086b578bf04302a502e9a7605ea8f1244a7b8577cd945eec78/fastparquet-2024.11.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46b2db02fc2a1507939d35441c8ab211d53afd75d82eec9767d1c3656402859b", size = 1751113 },
+    { url = "https://files.pythonhosted.org/packages/6b/a7/040507cee3a7798954e8fdbca21d2dbc532774b02b882d902b8a4a6849ef/fastparquet-2024.11.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a3afdef2895c9f459135a00a7ed3ceafebfbce918a9e7b5d550e4fae39c1b64d", size = 1780496 },
+    { url = "https://files.pythonhosted.org/packages/bc/75/d0d9f7533d780ec167eede16ad88073ee71696150511126c31940e7f73aa/fastparquet-2024.11.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:36b5c9bd2ffaaa26ff45d59a6cefe58503dd748e0c7fad80dd905749da0f2b9e", size = 1713608 },
+    { url = "https://files.pythonhosted.org/packages/30/fa/1d95bc86e45e80669c4f374b2ca26a9e5895a1011bb05d6341b4a7414693/fastparquet-2024.11.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:6b7df5d3b61a19d76e209fe8d3133759af1c139e04ebc6d43f3cc2d8045ef338", size = 1792779 },
+    { url = "https://files.pythonhosted.org/packages/13/3d/c076beeb926c79593374c04662a9422a76650eef17cd1c8e10951340764a/fastparquet-2024.11.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:8b35823ac7a194134e5f82fa4a9659e42e8f9ad1f2d22a55fbb7b9e4053aabbb", size = 1851322 },
+    { url = "https://files.pythonhosted.org/packages/09/5a/1d0d47e64816002824d4a876644e8c65540fa23f91b701f0daa726931545/fastparquet-2024.11.0-cp313-cp313-win_amd64.whl", hash = "sha256:d20632964e65530374ff7cddd42cc06aa0a1388934903693d6d22592a5ba827b", size = 673266 },
+    { url = "https://files.pythonhosted.org/packages/a2/3a/607301b22fd50065503a8be7ea0068ae61d9b7f63be44a4bad68a286f712/fastparquet-2024.11.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:1ae953c0e3832ae3936b6d92fde493ac7d8b775d7d59d02f7f46f67e1c21ed24", size = 911743 },
+    { url = "https://files.pythonhosted.org/packages/e7/a7/a9ff37efc9cf5164e372a813630488c0b3600e35c6bc1a7add9b9d17296f/fastparquet-2024.11.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:869e167a4067116b4a27eb7adbe597130b2e2e9cfc0f3e84f60e2e182a933f23", size = 684662 },
+    { url = "https://files.pythonhosted.org/packages/2c/76/dd9456425c312e8629e969c2099c4bc538fbf433542752395acb22c65559/fastparquet-2024.11.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:eb3356862fba2f9b2ea8e679d66901f466c92be8e023439fe854bc392fbf40a6", size = 1701403 },
+    { url = "https://files.pythonhosted.org/packages/b3/9d/17a296e458b62e7b9af301c2e7e77a21af37bcf29030a20f56d5d6757ae3/fastparquet-2024.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dc475993232c6a64f350aeb928013a807eb93f78675810fd019cbcff39f6baf3", size = 1715213 },
+    { url = "https://files.pythonhosted.org/packages/d2/05/2b5722eb6e24249706886423fbab2c42a5d60b0fc0f1a27e955ff2b829e8/fastparquet-2024.11.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d24c923a2d9d22a5e7564245f856e6462d524d57982ac8f7479cde991ff73362", size = 1668310 },
+    { url = "https://files.pythonhosted.org/packages/4c/9a/8a53dfa0c564f0683b693aa84878074d82ce2516e0f599d2a19ff3a0b073/fastparquet-2024.11.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6b936dcf40ca5fff9e70383d48811b1482b871ff74af857cb4db5f4d072f01ab", size = 1743186 },
+    { url = "https://files.pythonhosted.org/packages/cd/f6/61986beb1a894b40e7b9e9f5d0228ca068e005e66576af0b4fa94481ffbc/fastparquet-2024.11.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:4abd3426607335e5ad09be29ef4eeccdf097710e44420deac16893cee64ea0d8", size = 1784121 },
+    { url = "https://files.pythonhosted.org/packages/91/71/2cf345d0ccf1239108190bcd838a19b289fa1bac5676e8ae701ae24b602f/fastparquet-2024.11.0-cp39-cp39-win_amd64.whl", hash = "sha256:6ec7b398a86432993441d0a08dfae59e29649c803ed64ec4b1d7c3e0855b14cb", size = 671248 },
+]
+
+[[package]]
+name = "filelock"
+version = "3.16.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/9d/db/3ef5bb276dae18d6ec2124224403d1d67bccdbefc17af4cc8f553e341ab1/filelock-3.16.1.tar.gz", hash = "sha256:c249fbfcd5db47e5e2d6d62198e565475ee65e4831e2561c8e313fa7eb961435", size = 18037 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b9/f8/feced7779d755758a52d1f6635d990b8d98dc0a29fa568bbe0625f18fdf3/filelock-3.16.1-py3-none-any.whl", hash = "sha256:2082e5703d51fbf98ea75855d9d5527e33d8ff23099bec374a134febee6946b0", size = 16163 },
+]
+
+[[package]]
+name = "frozenlist"
+version = "1.5.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8f/ed/0f4cec13a93c02c47ec32d81d11c0c1efbadf4a471e3f3ce7cad366cbbd3/frozenlist-1.5.0.tar.gz", hash = "sha256:81d5af29e61b9c8348e876d442253723928dce6433e0e76cd925cd83f1b4b817", size = 39930 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/54/79/29d44c4af36b2b240725dce566b20f63f9b36ef267aaaa64ee7466f4f2f8/frozenlist-1.5.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:5b6a66c18b5b9dd261ca98dffcb826a525334b2f29e7caa54e182255c5f6a65a", size = 94451 },
+    { url = "https://files.pythonhosted.org/packages/47/47/0c999aeace6ead8a44441b4f4173e2261b18219e4ad1fe9a479871ca02fc/frozenlist-1.5.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:d1b3eb7b05ea246510b43a7e53ed1653e55c2121019a97e60cad7efb881a97bb", size = 54301 },
+    { url = "https://files.pythonhosted.org/packages/8d/60/107a38c1e54176d12e06e9d4b5d755b677d71d1219217cee063911b1384f/frozenlist-1.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:15538c0cbf0e4fa11d1e3a71f823524b0c46299aed6e10ebb4c2089abd8c3bec", size = 52213 },
+    { url = "https://files.pythonhosted.org/packages/17/62/594a6829ac5679c25755362a9dc93486a8a45241394564309641425d3ff6/frozenlist-1.5.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e79225373c317ff1e35f210dd5f1344ff31066ba8067c307ab60254cd3a78ad5", size = 240946 },
+    { url = "https://files.pythonhosted.org/packages/7e/75/6c8419d8f92c80dd0ee3f63bdde2702ce6398b0ac8410ff459f9b6f2f9cb/frozenlist-1.5.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9272fa73ca71266702c4c3e2d4a28553ea03418e591e377a03b8e3659d94fa76", size = 264608 },
+    { url = "https://files.pythonhosted.org/packages/88/3e/82a6f0b84bc6fb7e0be240e52863c6d4ab6098cd62e4f5b972cd31e002e8/frozenlist-1.5.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:498524025a5b8ba81695761d78c8dd7382ac0b052f34e66939c42df860b8ff17", size = 261361 },
+    { url = "https://files.pythonhosted.org/packages/fd/85/14e5f9ccac1b64ff2f10c927b3ffdf88772aea875882406f9ba0cec8ad84/frozenlist-1.5.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:92b5278ed9d50fe610185ecd23c55d8b307d75ca18e94c0e7de328089ac5dcba", size = 231649 },
+    { url = "https://files.pythonhosted.org/packages/ee/59/928322800306f6529d1852323014ee9008551e9bb027cc38d276cbc0b0e7/frozenlist-1.5.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7f3c8c1dacd037df16e85227bac13cca58c30da836c6f936ba1df0c05d046d8d", size = 241853 },
+    { url = "https://files.pythonhosted.org/packages/7d/bd/e01fa4f146a6f6c18c5d34cab8abdc4013774a26c4ff851128cd1bd3008e/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:f2ac49a9bedb996086057b75bf93538240538c6d9b38e57c82d51f75a73409d2", size = 243652 },
+    { url = "https://files.pythonhosted.org/packages/a5/bd/e4771fd18a8ec6757033f0fa903e447aecc3fbba54e3630397b61596acf0/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e66cc454f97053b79c2ab09c17fbe3c825ea6b4de20baf1be28919460dd7877f", size = 241734 },
+    { url = "https://files.pythonhosted.org/packages/21/13/c83821fa5544af4f60c5d3a65d054af3213c26b14d3f5f48e43e5fb48556/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:5a3ba5f9a0dfed20337d3e966dc359784c9f96503674c2faf015f7fe8e96798c", size = 260959 },
+    { url = "https://files.pythonhosted.org/packages/71/f3/1f91c9a9bf7ed0e8edcf52698d23f3c211d8d00291a53c9f115ceb977ab1/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:6321899477db90bdeb9299ac3627a6a53c7399c8cd58d25da094007402b039ab", size = 262706 },
+    { url = "https://files.pythonhosted.org/packages/4c/22/4a256fdf5d9bcb3ae32622c796ee5ff9451b3a13a68cfe3f68e2c95588ce/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:76e4753701248476e6286f2ef492af900ea67d9706a0155335a40ea21bf3b2f5", size = 250401 },
+    { url = "https://files.pythonhosted.org/packages/af/89/c48ebe1f7991bd2be6d5f4ed202d94960c01b3017a03d6954dd5fa9ea1e8/frozenlist-1.5.0-cp310-cp310-win32.whl", hash = "sha256:977701c081c0241d0955c9586ffdd9ce44f7a7795df39b9151cd9a6fd0ce4cfb", size = 45498 },
+    { url = "https://files.pythonhosted.org/packages/28/2f/cc27d5f43e023d21fe5c19538e08894db3d7e081cbf582ad5ed366c24446/frozenlist-1.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:189f03b53e64144f90990d29a27ec4f7997d91ed3d01b51fa39d2dbe77540fd4", size = 51622 },
+    { url = "https://files.pythonhosted.org/packages/79/43/0bed28bf5eb1c9e4301003b74453b8e7aa85fb293b31dde352aac528dafc/frozenlist-1.5.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:fd74520371c3c4175142d02a976aee0b4cb4a7cc912a60586ffd8d5929979b30", size = 94987 },
+    { url = "https://files.pythonhosted.org/packages/bb/bf/b74e38f09a246e8abbe1e90eb65787ed745ccab6eaa58b9c9308e052323d/frozenlist-1.5.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:2f3f7a0fbc219fb4455264cae4d9f01ad41ae6ee8524500f381de64ffaa077d5", size = 54584 },
+    { url = "https://files.pythonhosted.org/packages/2c/31/ab01375682f14f7613a1ade30149f684c84f9b8823a4391ed950c8285656/frozenlist-1.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:f47c9c9028f55a04ac254346e92977bf0f166c483c74b4232bee19a6697e4778", size = 52499 },
+    { url = "https://files.pythonhosted.org/packages/98/a8/d0ac0b9276e1404f58fec3ab6e90a4f76b778a49373ccaf6a563f100dfbc/frozenlist-1.5.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0996c66760924da6e88922756d99b47512a71cfd45215f3570bf1e0b694c206a", size = 276357 },
+    { url = "https://files.pythonhosted.org/packages/ad/c9/c7761084fa822f07dac38ac29f841d4587570dd211e2262544aa0b791d21/frozenlist-1.5.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a2fe128eb4edeabe11896cb6af88fca5346059f6c8d807e3b910069f39157869", size = 287516 },
+    { url = "https://files.pythonhosted.org/packages/a1/ff/cd7479e703c39df7bdab431798cef89dc75010d8aa0ca2514c5b9321db27/frozenlist-1.5.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1a8ea951bbb6cacd492e3948b8da8c502a3f814f5d20935aae74b5df2b19cf3d", size = 283131 },
+    { url = "https://files.pythonhosted.org/packages/59/a0/370941beb47d237eca4fbf27e4e91389fd68699e6f4b0ebcc95da463835b/frozenlist-1.5.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:de537c11e4aa01d37db0d403b57bd6f0546e71a82347a97c6a9f0dcc532b3a45", size = 261320 },
+    { url = "https://files.pythonhosted.org/packages/b8/5f/c10123e8d64867bc9b4f2f510a32042a306ff5fcd7e2e09e5ae5100ee333/frozenlist-1.5.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9c2623347b933fcb9095841f1cc5d4ff0b278addd743e0e966cb3d460278840d", size = 274877 },
+    { url = "https://files.pythonhosted.org/packages/fa/79/38c505601ae29d4348f21706c5d89755ceded02a745016ba2f58bd5f1ea6/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:cee6798eaf8b1416ef6909b06f7dc04b60755206bddc599f52232606e18179d3", size = 269592 },
+    { url = "https://files.pythonhosted.org/packages/19/e2/39f3a53191b8204ba9f0bb574b926b73dd2efba2a2b9d2d730517e8f7622/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:f5f9da7f5dbc00a604fe74aa02ae7c98bcede8a3b8b9666f9f86fc13993bc71a", size = 265934 },
+    { url = "https://files.pythonhosted.org/packages/d5/c9/3075eb7f7f3a91f1a6b00284af4de0a65a9ae47084930916f5528144c9dd/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:90646abbc7a5d5c7c19461d2e3eeb76eb0b204919e6ece342feb6032c9325ae9", size = 283859 },
+    { url = "https://files.pythonhosted.org/packages/05/f5/549f44d314c29408b962fa2b0e69a1a67c59379fb143b92a0a065ffd1f0f/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:bdac3c7d9b705d253b2ce370fde941836a5f8b3c5c2b8fd70940a3ea3af7f4f2", size = 287560 },
+    { url = "https://files.pythonhosted.org/packages/9d/f8/cb09b3c24a3eac02c4c07a9558e11e9e244fb02bf62c85ac2106d1eb0c0b/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:03d33c2ddbc1816237a67f66336616416e2bbb6beb306e5f890f2eb22b959cdf", size = 277150 },
+    { url = "https://files.pythonhosted.org/packages/37/48/38c2db3f54d1501e692d6fe058f45b6ad1b358d82cd19436efab80cfc965/frozenlist-1.5.0-cp311-cp311-win32.whl", hash = "sha256:237f6b23ee0f44066219dae14c70ae38a63f0440ce6750f868ee08775073f942", size = 45244 },
+    { url = "https://files.pythonhosted.org/packages/ca/8c/2ddffeb8b60a4bce3b196c32fcc30d8830d4615e7b492ec2071da801b8ad/frozenlist-1.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:0cc974cc93d32c42e7b0f6cf242a6bd941c57c61b618e78b6c0a96cb72788c1d", size = 51634 },
+    { url = "https://files.pythonhosted.org/packages/79/73/fa6d1a96ab7fd6e6d1c3500700963eab46813847f01ef0ccbaa726181dd5/frozenlist-1.5.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:31115ba75889723431aa9a4e77d5f398f5cf976eea3bdf61749731f62d4a4a21", size = 94026 },
+    { url = "https://files.pythonhosted.org/packages/ab/04/ea8bf62c8868b8eada363f20ff1b647cf2e93377a7b284d36062d21d81d1/frozenlist-1.5.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:7437601c4d89d070eac8323f121fcf25f88674627505334654fd027b091db09d", size = 54150 },
+    { url = "https://files.pythonhosted.org/packages/d0/9a/8e479b482a6f2070b26bda572c5e6889bb3ba48977e81beea35b5ae13ece/frozenlist-1.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:7948140d9f8ece1745be806f2bfdf390127cf1a763b925c4a805c603df5e697e", size = 51927 },
+    { url = "https://files.pythonhosted.org/packages/e3/12/2aad87deb08a4e7ccfb33600871bbe8f0e08cb6d8224371387f3303654d7/frozenlist-1.5.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:feeb64bc9bcc6b45c6311c9e9b99406660a9c05ca8a5b30d14a78555088b0b3a", size = 282647 },
+    { url = "https://files.pythonhosted.org/packages/77/f2/07f06b05d8a427ea0060a9cef6e63405ea9e0d761846b95ef3fb3be57111/frozenlist-1.5.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:683173d371daad49cffb8309779e886e59c2f369430ad28fe715f66d08d4ab1a", size = 289052 },
+    { url = "https://files.pythonhosted.org/packages/bd/9f/8bf45a2f1cd4aa401acd271b077989c9267ae8463e7c8b1eb0d3f561b65e/frozenlist-1.5.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7d57d8f702221405a9d9b40f9da8ac2e4a1a8b5285aac6100f3393675f0a85ee", size = 291719 },
+    { url = "https://files.pythonhosted.org/packages/41/d1/1f20fd05a6c42d3868709b7604c9f15538a29e4f734c694c6bcfc3d3b935/frozenlist-1.5.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:30c72000fbcc35b129cb09956836c7d7abf78ab5416595e4857d1cae8d6251a6", size = 267433 },
+    { url = "https://files.pythonhosted.org/packages/af/f2/64b73a9bb86f5a89fb55450e97cd5c1f84a862d4ff90d9fd1a73ab0f64a5/frozenlist-1.5.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:000a77d6034fbad9b6bb880f7ec073027908f1b40254b5d6f26210d2dab1240e", size = 283591 },
+    { url = "https://files.pythonhosted.org/packages/29/e2/ffbb1fae55a791fd6c2938dd9ea779509c977435ba3940b9f2e8dc9d5316/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5d7f5a50342475962eb18b740f3beecc685a15b52c91f7d975257e13e029eca9", size = 273249 },
+    { url = "https://files.pythonhosted.org/packages/2e/6e/008136a30798bb63618a114b9321b5971172a5abddff44a100c7edc5ad4f/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:87f724d055eb4785d9be84e9ebf0f24e392ddfad00b3fe036e43f489fafc9039", size = 271075 },
+    { url = "https://files.pythonhosted.org/packages/ae/f0/4e71e54a026b06724cec9b6c54f0b13a4e9e298cc8db0f82ec70e151f5ce/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:6e9080bb2fb195a046e5177f10d9d82b8a204c0736a97a153c2466127de87784", size = 285398 },
+    { url = "https://files.pythonhosted.org/packages/4d/36/70ec246851478b1c0b59f11ef8ade9c482ff447c1363c2bd5fad45098b12/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:9b93d7aaa36c966fa42efcaf716e6b3900438632a626fb09c049f6a2f09fc631", size = 294445 },
+    { url = "https://files.pythonhosted.org/packages/37/e0/47f87544055b3349b633a03c4d94b405956cf2437f4ab46d0928b74b7526/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:52ef692a4bc60a6dd57f507429636c2af8b6046db8b31b18dac02cbc8f507f7f", size = 280569 },
+    { url = "https://files.pythonhosted.org/packages/f9/7c/490133c160fb6b84ed374c266f42800e33b50c3bbab1652764e6e1fc498a/frozenlist-1.5.0-cp312-cp312-win32.whl", hash = "sha256:29d94c256679247b33a3dc96cce0f93cbc69c23bf75ff715919332fdbb6a32b8", size = 44721 },
+    { url = "https://files.pythonhosted.org/packages/b1/56/4e45136ffc6bdbfa68c29ca56ef53783ef4c2fd395f7cbf99a2624aa9aaa/frozenlist-1.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:8969190d709e7c48ea386db202d708eb94bdb29207a1f269bab1196ce0dcca1f", size = 51329 },
+    { url = "https://files.pythonhosted.org/packages/da/3b/915f0bca8a7ea04483622e84a9bd90033bab54bdf485479556c74fd5eaf5/frozenlist-1.5.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:7a1a048f9215c90973402e26c01d1cff8a209e1f1b53f72b95c13db61b00f953", size = 91538 },
+    { url = "https://files.pythonhosted.org/packages/c7/d1/a7c98aad7e44afe5306a2b068434a5830f1470675f0e715abb86eb15f15b/frozenlist-1.5.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:dd47a5181ce5fcb463b5d9e17ecfdb02b678cca31280639255ce9d0e5aa67af0", size = 52849 },
+    { url = "https://files.pythonhosted.org/packages/3a/c8/76f23bf9ab15d5f760eb48701909645f686f9c64fbb8982674c241fbef14/frozenlist-1.5.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:1431d60b36d15cda188ea222033eec8e0eab488f39a272461f2e6d9e1a8e63c2", size = 50583 },
+    { url = "https://files.pythonhosted.org/packages/1f/22/462a3dd093d11df623179d7754a3b3269de3b42de2808cddef50ee0f4f48/frozenlist-1.5.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6482a5851f5d72767fbd0e507e80737f9c8646ae7fd303def99bfe813f76cf7f", size = 265636 },
+    { url = "https://files.pythonhosted.org/packages/80/cf/e075e407fc2ae7328155a1cd7e22f932773c8073c1fc78016607d19cc3e5/frozenlist-1.5.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:44c49271a937625619e862baacbd037a7ef86dd1ee215afc298a417ff3270608", size = 270214 },
+    { url = "https://files.pythonhosted.org/packages/a1/58/0642d061d5de779f39c50cbb00df49682832923f3d2ebfb0fedf02d05f7f/frozenlist-1.5.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:12f78f98c2f1c2429d42e6a485f433722b0061d5c0b0139efa64f396efb5886b", size = 273905 },
+    { url = "https://files.pythonhosted.org/packages/ab/66/3fe0f5f8f2add5b4ab7aa4e199f767fd3b55da26e3ca4ce2cc36698e50c4/frozenlist-1.5.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ce3aa154c452d2467487765e3adc730a8c153af77ad84096bc19ce19a2400840", size = 250542 },
+    { url = "https://files.pythonhosted.org/packages/f6/b8/260791bde9198c87a465224e0e2bb62c4e716f5d198fc3a1dacc4895dbd1/frozenlist-1.5.0-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9b7dc0c4338e6b8b091e8faf0db3168a37101943e687f373dce00959583f7439", size = 267026 },
+    { url = "https://files.pythonhosted.org/packages/2e/a4/3d24f88c527f08f8d44ade24eaee83b2627793fa62fa07cbb7ff7a2f7d42/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:45e0896250900b5aa25180f9aec243e84e92ac84bd4a74d9ad4138ef3f5c97de", size = 257690 },
+    { url = "https://files.pythonhosted.org/packages/de/9a/d311d660420b2beeff3459b6626f2ab4fb236d07afbdac034a4371fe696e/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:561eb1c9579d495fddb6da8959fd2a1fca2c6d060d4113f5844b433fc02f2641", size = 253893 },
+    { url = "https://files.pythonhosted.org/packages/c6/23/e491aadc25b56eabd0f18c53bb19f3cdc6de30b2129ee0bc39cd387cd560/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:df6e2f325bfee1f49f81aaac97d2aa757c7646534a06f8f577ce184afe2f0a9e", size = 267006 },
+    { url = "https://files.pythonhosted.org/packages/08/c4/ab918ce636a35fb974d13d666dcbe03969592aeca6c3ab3835acff01f79c/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:140228863501b44b809fb39ec56b5d4071f4d0aa6d216c19cbb08b8c5a7eadb9", size = 276157 },
+    { url = "https://files.pythonhosted.org/packages/c0/29/3b7a0bbbbe5a34833ba26f686aabfe982924adbdcafdc294a7a129c31688/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:7707a25d6a77f5d27ea7dc7d1fc608aa0a478193823f88511ef5e6b8a48f9d03", size = 264642 },
+    { url = "https://files.pythonhosted.org/packages/ab/42/0595b3dbffc2e82d7fe658c12d5a5bafcd7516c6bf2d1d1feb5387caa9c1/frozenlist-1.5.0-cp313-cp313-win32.whl", hash = "sha256:31a9ac2b38ab9b5a8933b693db4939764ad3f299fcaa931a3e605bc3460e693c", size = 44914 },
+    { url = "https://files.pythonhosted.org/packages/17/c4/b7db1206a3fea44bf3b838ca61deb6f74424a8a5db1dd53ecb21da669be6/frozenlist-1.5.0-cp313-cp313-win_amd64.whl", hash = "sha256:11aabdd62b8b9c4b84081a3c246506d1cddd2dd93ff0ad53ede5defec7886b28", size = 51167 },
+    { url = "https://files.pythonhosted.org/packages/da/4d/d94ff0fb0f5313902c132817c62d19cdc5bdcd0c195d392006ef4b779fc6/frozenlist-1.5.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:9bbcdfaf4af7ce002694a4e10a0159d5a8d20056a12b05b45cea944a4953f972", size = 95319 },
+    { url = "https://files.pythonhosted.org/packages/8c/1b/d90e554ca2b483d31cb2296e393f72c25bdc38d64526579e95576bfda587/frozenlist-1.5.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1893f948bf6681733aaccf36c5232c231e3b5166d607c5fa77773611df6dc336", size = 54749 },
+    { url = "https://files.pythonhosted.org/packages/f8/66/7fdecc9ef49f8db2aa4d9da916e4ecf357d867d87aea292efc11e1b2e932/frozenlist-1.5.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2b5e23253bb709ef57a8e95e6ae48daa9ac5f265637529e4ce6b003a37b2621f", size = 52718 },
+    { url = "https://files.pythonhosted.org/packages/08/04/e2fddc92135276e07addbc1cf413acffa0c2d848b3e54cacf684e146df49/frozenlist-1.5.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0f253985bb515ecd89629db13cb58d702035ecd8cfbca7d7a7e29a0e6d39af5f", size = 241756 },
+    { url = "https://files.pythonhosted.org/packages/c6/52/be5ff200815d8a341aee5b16b6b707355e0ca3652953852238eb92b120c2/frozenlist-1.5.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:04a5c6babd5e8fb7d3c871dc8b321166b80e41b637c31a995ed844a6139942b6", size = 267718 },
+    { url = "https://files.pythonhosted.org/packages/88/be/4bd93a58be57a3722fc544c36debdf9dcc6758f761092e894d78f18b8f20/frozenlist-1.5.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a9fe0f1c29ba24ba6ff6abf688cb0b7cf1efab6b6aa6adc55441773c252f7411", size = 263494 },
+    { url = "https://files.pythonhosted.org/packages/32/ba/58348b90193caa096ce9e9befea6ae67f38dabfd3aacb47e46137a6250a8/frozenlist-1.5.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:226d72559fa19babe2ccd920273e767c96a49b9d3d38badd7c91a0fdeda8ea08", size = 232838 },
+    { url = "https://files.pythonhosted.org/packages/f6/33/9f152105227630246135188901373c4f322cc026565ca6215b063f4c82f4/frozenlist-1.5.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:15b731db116ab3aedec558573c1a5eec78822b32292fe4f2f0345b7f697745c2", size = 242912 },
+    { url = "https://files.pythonhosted.org/packages/a0/10/3db38fb3ccbafadd80a1b0d6800c987b0e3fe3ef2d117c6ced0246eea17a/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:366d8f93e3edfe5a918c874702f78faac300209a4d5bf38352b2c1bdc07a766d", size = 244763 },
+    { url = "https://files.pythonhosted.org/packages/e2/cd/1df468fdce2f66a4608dffe44c40cdc35eeaa67ef7fd1d813f99a9a37842/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:1b96af8c582b94d381a1c1f51ffaedeb77c821c690ea5f01da3d70a487dd0a9b", size = 242841 },
+    { url = "https://files.pythonhosted.org/packages/ee/5f/16097a5ca0bb6b6779c02cc9379c72fe98d56115d4c54d059fb233168fb6/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:c03eff4a41bd4e38415cbed054bbaff4a075b093e2394b6915dca34a40d1e38b", size = 263407 },
+    { url = "https://files.pythonhosted.org/packages/0f/f7/58cd220ee1c2248ee65a32f5b4b93689e3fe1764d85537eee9fc392543bc/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:50cf5e7ee9b98f22bdecbabf3800ae78ddcc26e4a435515fc72d97903e8488e0", size = 265083 },
+    { url = "https://files.pythonhosted.org/packages/62/b8/49768980caabf81ac4a2d156008f7cbd0107e6b36d08a313bb31035d9201/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1e76bfbc72353269c44e0bc2cfe171900fbf7f722ad74c9a7b638052afe6a00c", size = 251564 },
+    { url = "https://files.pythonhosted.org/packages/cb/83/619327da3b86ef957ee7a0cbf3c166a09ed1e87a3f7f1ff487d7d0284683/frozenlist-1.5.0-cp39-cp39-win32.whl", hash = "sha256:666534d15ba8f0fda3f53969117383d5dc021266b3c1a42c9ec4855e4b58b9d3", size = 45691 },
+    { url = "https://files.pythonhosted.org/packages/8b/28/407bc34a745151ed2322c690b6e7d83d7101472e81ed76e1ebdac0b70a78/frozenlist-1.5.0-cp39-cp39-win_amd64.whl", hash = "sha256:5c28f4b5dbef8a0d8aad0d4de24d1e9e981728628afaf4ea0792f5d0939372f0", size = 51767 },
+    { url = "https://files.pythonhosted.org/packages/c6/c8/a5be5b7550c10858fcf9b0ea054baccab474da77d37f1e828ce043a3a5d4/frozenlist-1.5.0-py3-none-any.whl", hash = "sha256:d994863bba198a4a518b467bb971c56e1db3f180a25c6cf7bb1949c267f748c3", size = 11901 },
+]
+
+[[package]]
+name = "fsspec"
+version = "2024.9.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/62/7c/12b0943011daaaa9c35c2a2e22e5eb929ac90002f08f1259d69aedad84de/fsspec-2024.9.0.tar.gz", hash = "sha256:4b0afb90c2f21832df142f292649035d80b421f60a9e1c027802e5a0da2b04e8", size = 286206 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1d/a0/6aaea0c2fbea2f89bfd5db25fb1e3481896a423002ebe4e55288907a97a3/fsspec-2024.9.0-py3-none-any.whl", hash = "sha256:a0947d552d8a6efa72cc2c730b12c41d043509156966cca4fb157b0f2a0c574b", size = 179253 },
+]
+
+[package.optional-dependencies]
+http = [
+    { name = "aiohttp" },
+]
+
+[[package]]
+name = "genson"
+version = "1.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/c5/cf/2303c8ad276dcf5ee2ad6cf69c4338fd86ef0f471a5207b069adf7a393cf/genson-1.3.0.tar.gz", hash = "sha256:e02db9ac2e3fd29e65b5286f7135762e2cd8a986537c075b06fc5f1517308e37", size = 34919 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f8/5c/e226de133afd8bb267ec27eead9ae3d784b95b39a287ed404caab39a5f50/genson-1.3.0-py3-none-any.whl", hash = "sha256:468feccd00274cc7e4c09e84b08704270ba8d95232aa280f65b986139cec67f7", size = 21470 },
+]
+
+[[package]]
+name = "h11"
+version = "0.14.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f5/38/3af3d3633a34a3316095b39c8e8fb4853a28a536e55d347bd8d8e9a14b03/h11-0.14.0.tar.gz", hash = "sha256:8f19fbbe99e72420ff35c00b27a34cb9937e902a8b810e2c88300c6f0a3b699d", size = 100418 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/95/04/ff642e65ad6b90db43e668d70ffb6736436c7ce41fcc549f4e9472234127/h11-0.14.0-py3-none-any.whl", hash = "sha256:e3fe4ac4b851c468cc8363d500db52c2ead036020723024a109d37346efaa761", size = 58259 },
+]
+
+[[package]]
+name = "httpcore"
+version = "1.0.7"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "certifi" },
+    { name = "h11" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6a/41/d7d0a89eb493922c37d343b607bc1b5da7f5be7e383740b4753ad8943e90/httpcore-1.0.7.tar.gz", hash = "sha256:8551cb62a169ec7162ac7be8d4817d561f60e08eaa485234898414bb5a8a0b4c", size = 85196 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/87/f5/72347bc88306acb359581ac4d52f23c0ef445b57157adedb9aee0cd689d2/httpcore-1.0.7-py3-none-any.whl", hash = "sha256:a3fff8f43dc260d5bd363d9f9cf1830fa3a458b332856f34282de498ed420edd", size = 78551 },
+]
+
+[[package]]
+name = "httptools"
+version = "0.6.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/a7/9a/ce5e1f7e131522e6d3426e8e7a490b3a01f39a6696602e1c4f33f9e94277/httptools-0.6.4.tar.gz", hash = "sha256:4e93eee4add6493b59a5c514da98c939b244fce4a0d8879cd3f466562f4b7d5c", size = 240639 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3b/6f/972f8eb0ea7d98a1c6be436e2142d51ad2a64ee18e02b0e7ff1f62171ab1/httptools-0.6.4-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:3c73ce323711a6ffb0d247dcd5a550b8babf0f757e86a52558fe5b86d6fefcc0", size = 198780 },
+    { url = "https://files.pythonhosted.org/packages/6a/b0/17c672b4bc5c7ba7f201eada4e96c71d0a59fbc185e60e42580093a86f21/httptools-0.6.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:345c288418f0944a6fe67be8e6afa9262b18c7626c3ef3c28adc5eabc06a68da", size = 103297 },
+    { url = "https://files.pythonhosted.org/packages/92/5e/b4a826fe91971a0b68e8c2bd4e7db3e7519882f5a8ccdb1194be2b3ab98f/httptools-0.6.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:deee0e3343f98ee8047e9f4c5bc7cedbf69f5734454a94c38ee829fb2d5fa3c1", size = 443130 },
+    { url = "https://files.pythonhosted.org/packages/b0/51/ce61e531e40289a681a463e1258fa1e05e0be54540e40d91d065a264cd8f/httptools-0.6.4-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ca80b7485c76f768a3bc83ea58373f8db7b015551117375e4918e2aa77ea9b50", size = 442148 },
+    { url = "https://files.pythonhosted.org/packages/ea/9e/270b7d767849b0c96f275c695d27ca76c30671f8eb8cc1bab6ced5c5e1d0/httptools-0.6.4-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:90d96a385fa941283ebd231464045187a31ad932ebfa541be8edf5b3c2328959", size = 415949 },
+    { url = "https://files.pythonhosted.org/packages/81/86/ced96e3179c48c6f656354e106934e65c8963d48b69be78f355797f0e1b3/httptools-0.6.4-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:59e724f8b332319e2875efd360e61ac07f33b492889284a3e05e6d13746876f4", size = 417591 },
+    { url = "https://files.pythonhosted.org/packages/75/73/187a3f620ed3175364ddb56847d7a608a6fc42d551e133197098c0143eca/httptools-0.6.4-cp310-cp310-win_amd64.whl", hash = "sha256:c26f313951f6e26147833fc923f78f95604bbec812a43e5ee37f26dc9e5a686c", size = 88344 },
+    { url = "https://files.pythonhosted.org/packages/7b/26/bb526d4d14c2774fe07113ca1db7255737ffbb119315839af2065abfdac3/httptools-0.6.4-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:f47f8ed67cc0ff862b84a1189831d1d33c963fb3ce1ee0c65d3b0cbe7b711069", size = 199029 },
+    { url = "https://files.pythonhosted.org/packages/a6/17/3e0d3e9b901c732987a45f4f94d4e2c62b89a041d93db89eafb262afd8d5/httptools-0.6.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:0614154d5454c21b6410fdf5262b4a3ddb0f53f1e1721cfd59d55f32138c578a", size = 103492 },
+    { url = "https://files.pythonhosted.org/packages/b7/24/0fe235d7b69c42423c7698d086d4db96475f9b50b6ad26a718ef27a0bce6/httptools-0.6.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f8787367fbdfccae38e35abf7641dafc5310310a5987b689f4c32cc8cc3ee975", size = 462891 },
+    { url = "https://files.pythonhosted.org/packages/b1/2f/205d1f2a190b72da6ffb5f41a3736c26d6fa7871101212b15e9b5cd8f61d/httptools-0.6.4-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:40b0f7fe4fd38e6a507bdb751db0379df1e99120c65fbdc8ee6c1d044897a636", size = 459788 },
+    { url = "https://files.pythonhosted.org/packages/6e/4c/d09ce0eff09057a206a74575ae8f1e1e2f0364d20e2442224f9e6612c8b9/httptools-0.6.4-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:40a5ec98d3f49904b9fe36827dcf1aadfef3b89e2bd05b0e35e94f97c2b14721", size = 433214 },
+    { url = "https://files.pythonhosted.org/packages/3e/d2/84c9e23edbccc4a4c6f96a1b8d99dfd2350289e94f00e9ccc7aadde26fb5/httptools-0.6.4-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:dacdd3d10ea1b4ca9df97a0a303cbacafc04b5cd375fa98732678151643d4988", size = 434120 },
+    { url = "https://files.pythonhosted.org/packages/d0/46/4d8e7ba9581416de1c425b8264e2cadd201eb709ec1584c381f3e98f51c1/httptools-0.6.4-cp311-cp311-win_amd64.whl", hash = "sha256:288cd628406cc53f9a541cfaf06041b4c71d751856bab45e3702191f931ccd17", size = 88565 },
+    { url = "https://files.pythonhosted.org/packages/bb/0e/d0b71465c66b9185f90a091ab36389a7352985fe857e352801c39d6127c8/httptools-0.6.4-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:df017d6c780287d5c80601dafa31f17bddb170232d85c066604d8558683711a2", size = 200683 },
+    { url = "https://files.pythonhosted.org/packages/e2/b8/412a9bb28d0a8988de3296e01efa0bd62068b33856cdda47fe1b5e890954/httptools-0.6.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:85071a1e8c2d051b507161f6c3e26155b5c790e4e28d7f236422dbacc2a9cc44", size = 104337 },
+    { url = "https://files.pythonhosted.org/packages/9b/01/6fb20be3196ffdc8eeec4e653bc2a275eca7f36634c86302242c4fbb2760/httptools-0.6.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:69422b7f458c5af875922cdb5bd586cc1f1033295aa9ff63ee196a87519ac8e1", size = 508796 },
+    { url = "https://files.pythonhosted.org/packages/f7/d8/b644c44acc1368938317d76ac991c9bba1166311880bcc0ac297cb9d6bd7/httptools-0.6.4-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:16e603a3bff50db08cd578d54f07032ca1631450ceb972c2f834c2b860c28ea2", size = 510837 },
+    { url = "https://files.pythonhosted.org/packages/52/d8/254d16a31d543073a0e57f1c329ca7378d8924e7e292eda72d0064987486/httptools-0.6.4-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:ec4f178901fa1834d4a060320d2f3abc5c9e39766953d038f1458cb885f47e81", size = 485289 },
+    { url = "https://files.pythonhosted.org/packages/5f/3c/4aee161b4b7a971660b8be71a92c24d6c64372c1ab3ae7f366b3680df20f/httptools-0.6.4-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:f9eb89ecf8b290f2e293325c646a211ff1c2493222798bb80a530c5e7502494f", size = 489779 },
+    { url = "https://files.pythonhosted.org/packages/12/b7/5cae71a8868e555f3f67a50ee7f673ce36eac970f029c0c5e9d584352961/httptools-0.6.4-cp312-cp312-win_amd64.whl", hash = "sha256:db78cb9ca56b59b016e64b6031eda5653be0589dba2b1b43453f6e8b405a0970", size = 88634 },
+    { url = "https://files.pythonhosted.org/packages/94/a3/9fe9ad23fd35f7de6b91eeb60848986058bd8b5a5c1e256f5860a160cc3e/httptools-0.6.4-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ade273d7e767d5fae13fa637f4d53b6e961fb7fd93c7797562663f0171c26660", size = 197214 },
+    { url = "https://files.pythonhosted.org/packages/ea/d9/82d5e68bab783b632023f2fa31db20bebb4e89dfc4d2293945fd68484ee4/httptools-0.6.4-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:856f4bc0478ae143bad54a4242fccb1f3f86a6e1be5548fecfd4102061b3a083", size = 102431 },
+    { url = "https://files.pythonhosted.org/packages/96/c1/cb499655cbdbfb57b577734fde02f6fa0bbc3fe9fb4d87b742b512908dff/httptools-0.6.4-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:322d20ea9cdd1fa98bd6a74b77e2ec5b818abdc3d36695ab402a0de8ef2865a3", size = 473121 },
+    { url = "https://files.pythonhosted.org/packages/af/71/ee32fd358f8a3bb199b03261f10921716990808a675d8160b5383487a317/httptools-0.6.4-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4d87b29bd4486c0093fc64dea80231f7c7f7eb4dc70ae394d70a495ab8436071", size = 473805 },
+    { url = "https://files.pythonhosted.org/packages/8a/0a/0d4df132bfca1507114198b766f1737d57580c9ad1cf93c1ff673e3387be/httptools-0.6.4-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:342dd6946aa6bda4b8f18c734576106b8a31f2fe31492881a9a160ec84ff4bd5", size = 448858 },
+    { url = "https://files.pythonhosted.org/packages/1e/6a/787004fdef2cabea27bad1073bf6a33f2437b4dbd3b6fb4a9d71172b1c7c/httptools-0.6.4-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:4b36913ba52008249223042dca46e69967985fb4051951f94357ea681e1f5dc0", size = 452042 },
+    { url = "https://files.pythonhosted.org/packages/4d/dc/7decab5c404d1d2cdc1bb330b1bf70e83d6af0396fd4fc76fc60c0d522bf/httptools-0.6.4-cp313-cp313-win_amd64.whl", hash = "sha256:28908df1b9bb8187393d5b5db91435ccc9c8e891657f9cbb42a2541b44c82fc8", size = 87682 },
+    { url = "https://files.pythonhosted.org/packages/51/b1/4fc6f52afdf93b7c4304e21f6add9e981e4f857c2fa622a55dfe21b6059e/httptools-0.6.4-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:85797e37e8eeaa5439d33e556662cc370e474445d5fab24dcadc65a8ffb04003", size = 201123 },
+    { url = "https://files.pythonhosted.org/packages/c2/01/e6ecb40ac8fdfb76607c7d3b74a41b464458d5c8710534d8f163b0c15f29/httptools-0.6.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:db353d22843cf1028f43c3651581e4bb49374d85692a85f95f7b9a130e1b2cab", size = 104507 },
+    { url = "https://files.pythonhosted.org/packages/dc/24/c70c34119d209bf08199d938dc9c69164f585ed3029237b4bdb90f673cb9/httptools-0.6.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d1ffd262a73d7c28424252381a5b854c19d9de5f56f075445d33919a637e3547", size = 449615 },
+    { url = "https://files.pythonhosted.org/packages/2b/62/e7f317fed3703bd81053840cacba4e40bcf424b870e4197f94bd1cf9fe7a/httptools-0.6.4-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:703c346571fa50d2e9856a37d7cd9435a25e7fd15e236c397bf224afaa355fe9", size = 448819 },
+    { url = "https://files.pythonhosted.org/packages/2a/13/68337d3be6b023260139434c49d7aa466aaa98f9aee7ed29270ac7dde6a2/httptools-0.6.4-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:aafe0f1918ed07b67c1e838f950b1c1fabc683030477e60b335649b8020e1076", size = 422093 },
+    { url = "https://files.pythonhosted.org/packages/fc/b3/3a1bc45be03dda7a60c7858e55b6cd0489a81613c1908fb81cf21d34ae50/httptools-0.6.4-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:0e563e54979e97b6d13f1bbc05a96109923e76b901f786a5eae36e99c01237bd", size = 423898 },
+    { url = "https://files.pythonhosted.org/packages/05/72/2ddc2ae5f7ace986f7e68a326215b2e7c32e32fd40e6428fa8f1d8065c7e/httptools-0.6.4-cp39-cp39-win_amd64.whl", hash = "sha256:b799de31416ecc589ad79dd85a0b2657a8fe39327944998dea368c1d4c9e55e6", size = 89552 },
+]
+
+[[package]]
+name = "httpx"
+version = "0.28.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+    { name = "certifi" },
+    { name = "httpcore" },
+    { name = "idna" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b1/df/48c586a5fe32a0f01324ee087459e112ebb7224f646c0b5023f5e79e9956/httpx-0.28.1.tar.gz", hash = "sha256:75e98c5f16b0f35b567856f597f06ff2270a374470a5c2392242528e3e3e42fc", size = 141406 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2a/39/e50c7c3a983047577ee07d2a9e53faf5a69493943ec3f6a384bdc792deb2/httpx-0.28.1-py3-none-any.whl", hash = "sha256:d909fcccc110f8c7faf814ca82a9a4d816bc5a6dbfea25d6591d6985b8ba59ad", size = 73517 },
+]
+
+[[package]]
+name = "huggingface-hub"
+version = "0.27.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "filelock" },
+    { name = "fsspec" },
+    { name = "packaging" },
+    { name = "pyyaml" },
+    { name = "requests" },
+    { name = "tqdm" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/e1/d2/d6976de7542792fc077b498d64af64882b6d8bb40679284ec0bff77d5929/huggingface_hub-0.27.1.tar.gz", hash = "sha256:c004463ca870283909d715d20f066ebd6968c2207dae9393fdffb3c1d4d8f98b", size = 379407 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6c/3f/50f6b25fafdcfb1c089187a328c95081abf882309afd86f4053951507cd1/huggingface_hub-0.27.1-py3-none-any.whl", hash = "sha256:1c5155ca7d60b60c2e2fc38cbb3ffb7f7c3adf48f824015b219af9061771daec", size = 450658 },
+]
+
+[[package]]
+name = "identify"
+version = "2.6.5"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/cf/92/69934b9ef3c31ca2470980423fda3d00f0460ddefdf30a67adf7f17e2e00/identify-2.6.5.tar.gz", hash = "sha256:c10b33f250e5bba374fae86fb57f3adcebf1161bce7cdf92031915fd480c13bc", size = 99213 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ec/fa/dce098f4cdf7621aa8f7b4f919ce545891f489482f0bfa5102f3eca8608b/identify-2.6.5-py2.py3-none-any.whl", hash = "sha256:14181a47091eb75b337af4c23078c9d09225cd4c48929f521f3bf16b09d02566", size = 99078 },
+]
+
+[[package]]
+name = "idna"
+version = "3.10"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f1/70/7703c29685631f5a7590aa73f1f1d3fa9a380e654b86af429e0934a32f7d/idna-3.10.tar.gz", hash = "sha256:12f65c9b470abda6dc35cf8e63cc574b1c52b11df2c86030af0ac09b01b13ea9", size = 190490 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/76/c6/c88e154df9c4e1a2a66ccf0005a88dfb2650c1dffb6f5ce603dfbd452ce3/idna-3.10-py3-none-any.whl", hash = "sha256:946d195a0d259cbba61165e88e65941f16e9b36ea6ddb97f00452bae8b1287d3", size = 70442 },
+]
+
+[[package]]
+name = "importlib-metadata"
+version = "8.5.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "zipp", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/cd/12/33e59336dca5be0c398a7482335911a33aa0e20776128f038019f1a95f1b/importlib_metadata-8.5.0.tar.gz", hash = "sha256:71522656f0abace1d072b9e5481a48f07c138e00f079c38c8f883823f9c26bd7", size = 55304 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a0/d9/a1e041c5e7caa9a05c925f4bdbdfb7f006d1f74996af53467bc394c97be7/importlib_metadata-8.5.0-py3-none-any.whl", hash = "sha256:45e54197d28b7a7f1559e60b95e7c567032b602131fbd588f1497f47880aa68b", size = 26514 },
+]
+
+[[package]]
+name = "iniconfig"
+version = "2.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d7/4b/cbd8e699e64a6f16ca3a8220661b5f83792b3017d0f79807cb8708d33913/iniconfig-2.0.0.tar.gz", hash = "sha256:2d91e135bf72d31a410b17c16da610a82cb55f6b0477d1a902134b24a455b8b3", size = 4646 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ef/a6/62565a6e1cf69e10f5727360368e451d4b7f58beeac6173dc9db836a5b46/iniconfig-2.0.0-py3-none-any.whl", hash = "sha256:b6a85871a79d2e3b22d2d1b94ac2824226a63c6b741c88f7ae975f18b6778374", size = 5892 },
+]
+
+[[package]]
+name = "intel-openmp"
+version = "2021.4.0"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/45/18/527f247d673ff84c38e0b353b6901539b99e83066cd505be42ad341ab16d/intel_openmp-2021.4.0-py2.py3-none-win32.whl", hash = "sha256:6e863d8fd3d7e8ef389d52cf97a50fe2afe1a19247e8c0d168ce021546f96fc9", size = 1860605 },
+    { url = "https://files.pythonhosted.org/packages/6f/21/b590c0cc3888b24f2ac9898c41d852d7454a1695fbad34bee85dba6dc408/intel_openmp-2021.4.0-py2.py3-none-win_amd64.whl", hash = "sha256:eef4c8bcc8acefd7f5cd3b9384dbf73d59e2c99fc56545712ded913f43c4a94f", size = 3516906 },
+]
+
+[[package]]
+name = "interegular"
+version = "0.3.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/dc/9d/8b6dde58a028a3962ce17e84d5fe73758df61378e00ef8ac3d85da34b0ff/interegular-0.3.3.tar.gz", hash = "sha256:d9b697b21b34884711399ba0f0376914b81899ce670032486d0d048344a76600", size = 24705 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c4/01/72d6472f80651673716d1deda2a5bbb633e563ecf94f4479da5519d69d25/interegular-0.3.3-py37-none-any.whl", hash = "sha256:b0c07007d48c89d6d19f7204972d369b2a77222722e126b6aa63aa721dc3b19c", size = 23635 },
+]
+
+[[package]]
+name = "jax"
+version = "0.4.30"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
+    "python_full_version < '3.9.17'",
+]
+dependencies = [
+    { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
+    { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", marker = "python_full_version < '3.10'" },
+    { name = "opt-einsum", marker = "python_full_version < '3.10'" },
+    { name = "scipy", version = "1.13.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/15/41/d6dbafc31d6bd93eeec2e1c709adfa454266e83714ebeeed9de52a6ad881/jax-0.4.30.tar.gz", hash = "sha256:94d74b5b2db0d80672b61d83f1f63ebf99d2ab7398ec12b2ca0c9d1e97afe577", size = 1715462 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fd/f2/9dbb75de3058acfd1600cf0839bcce7ea391148c9d2b4fa5f5666e66f09e/jax-0.4.30-py3-none-any.whl", hash = "sha256:289b30ae03b52f7f4baf6ef082a9f4e3e29c1080e22d13512c5ecf02d5f1a55b", size = 2009197 },
+]
+
+[[package]]
+name = "jax"
+version = "0.5.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.13'",
+    "python_full_version == '3.12.*'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "jaxlib", version = "0.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.13' or (python_full_version >= '3.10' and python_full_version < '3.12')" },
+    { name = "ml-dtypes", version = "0.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.12.*'" },
+    { name = "numpy", marker = "python_full_version >= '3.10'" },
+    { name = "opt-einsum", marker = "python_full_version >= '3.10'" },
+    { name = "scipy", version = "1.15.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4a/cb/22d62b26284f08e62d6eb64603d3b010004cfdb7a97ce6cca5c6cf86edab/jax-0.5.0.tar.gz", hash = "sha256:49df70bf293a345a7fb519f71193506d37a024c4f850b358042eb32d502c81c8", size = 1959707 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f4/58/cc0721a1030fcbab0984beea0bf3c4610ec103f738423cdfa9c4ceb40598/jax-0.5.0-py3-none-any.whl", hash = "sha256:b3907aa87ae2c340b39cdbf80c07a74550369cafcaf7398fb60ba58d167345ab", size = 2270365 },
+]
+
+[[package]]
+name = "jaxlib"
+version = "0.4.30"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
+    "python_full_version < '3.9.17'",
+]
+dependencies = [
+    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", marker = "python_full_version < '3.10'" },
+    { name = "scipy", version = "1.13.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f3/18/ff7f2f6d6195853ed55c5b5d835f5c8c3c8b190c7221cb04a0cb81f5db10/jaxlib-0.4.30-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:c40856e28f300938c6824ab1a615166193d6997dec946578823f6d402ad454e5", size = 83542097 },
+    { url = "https://files.pythonhosted.org/packages/d4/c0/ff65503ecfed3aee11e4abe4c4e9e8a3513f072e0b595f8247b9989d1510/jaxlib-0.4.30-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:4bdfda6a3c7a2b0cc0a7131009eb279e98ca4a6f25679fabb5302dd135a5e349", size = 66694495 },
+    { url = "https://files.pythonhosted.org/packages/b9/d7/82df748a31a1cfbd531a12979ea846d6b676d4adfa1e91114b848665b2aa/jaxlib-0.4.30-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:28e032c9b394ab7624d89b0d9d3bbcf4d1d71694fe8b3e09d3fe64122eda7b0c", size = 67781242 },
+    { url = "https://files.pythonhosted.org/packages/4a/ca/561aabed63007bb2621a62f0d816aa2f68cfe947859c8b4e61519940344b/jaxlib-0.4.30-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:d83f36ef42a403bbf7c7f2da526b34ba286988e170f4df5e58b3bb735417868c", size = 79640266 },
+    { url = "https://files.pythonhosted.org/packages/b0/90/8e5347eda95d3cb695cd5ebb82f850fa7866078a6a7a0568549e34125a82/jaxlib-0.4.30-cp310-cp310-win_amd64.whl", hash = "sha256:a56678b28f96b524ded6da8ef4b38e72a532356d139cfd434da804abf4234e14", size = 51945307 },
+    { url = "https://files.pythonhosted.org/packages/33/2d/b6078f5d173d3087d32b1b49e5f65d406985fb3894ff1d21905972b9c89d/jaxlib-0.4.30-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:bfb5d85b69c29c3c6e8051a0ea715ac1e532d6e54494c8d9c3813dcc00deac30", size = 83539315 },
+    { url = "https://files.pythonhosted.org/packages/12/95/399da9204c3b13696baefb93468402f3389416b0caecfd9126aa94742bf2/jaxlib-0.4.30-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:974998cd8a78550402e6c09935c1f8d850cad9cc19ccd7488bde45b6f7f99c12", size = 66690971 },
+    { url = "https://files.pythonhosted.org/packages/a4/f8/b85a46cb0cc4bc228cea4366b0d15caf42656c6d43cf8c91d90f7399aa4d/jaxlib-0.4.30-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:e93eb0646b41ba213252b51b0b69096b9cd1d81a35ea85c9d06663b5d11efe45", size = 67780747 },
+    { url = "https://files.pythonhosted.org/packages/a6/a3/951da3d1487b2f8995a2a14cc7e9496c9a7c93aa1f1d0b33e833e24dee92/jaxlib-0.4.30-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:16b2ab18ea90d2e15941bcf45de37afc2f289a029129c88c8d7aba0404dd0043", size = 79640352 },
+    { url = "https://files.pythonhosted.org/packages/bb/1a/8f45ea28a5ca67e4d23ebd70fc78ea94be6fa20323f983c7607c32c6f9a5/jaxlib-0.4.30-cp311-cp311-win_amd64.whl", hash = "sha256:3a2e2c11c179f8851a72249ba1ae40ae817dfaee9877d23b3b8f7c6b7a012f76", size = 51943960 },
+    { url = "https://files.pythonhosted.org/packages/19/40/ae943d3c1fc8b50947aebbaa3bad2842759e43bc9fc91e1758c1c20a81ab/jaxlib-0.4.30-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:7704db5962b32a2be3cc07185433cbbcc94ed90ee50c84021a3f8a1ecfd66ee3", size = 83587124 },
+    { url = "https://files.pythonhosted.org/packages/c6/e3/97f8edff6f64245a500415be021869522b235e8b38cd930d358b91243583/jaxlib-0.4.30-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:57090d33477fd0f0c99dc686274882ea75c44c7d712ae42dd2460b10f896131d", size = 66724768 },
+    { url = "https://files.pythonhosted.org/packages/4c/c7/ee1f48f8daa409d0ed039e0d8b5ae1a447e53db3acb2ff06239828ad96d5/jaxlib-0.4.30-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:0a3850e76278038e21685975a62b622bcf3708485f13125757a0561ee4512940", size = 67800348 },
+    { url = "https://files.pythonhosted.org/packages/f2/fa/a2dddea0d6965b8e433bb99aeedbe5c8a9b47110c1c4f197a7b6239daf44/jaxlib-0.4.30-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:c58a8071c4e00898282118169f6a5a97eb15a79c2897858f3a732b17891c99ab", size = 79674030 },
+    { url = "https://files.pythonhosted.org/packages/db/31/3500633d61b20b882a0fbcf8100013195c31b51f71249b0b38737851fc9a/jaxlib-0.4.30-cp312-cp312-win_amd64.whl", hash = "sha256:b7079a5b1ab6864a7d4f2afaa963841451186d22c90f39719a3ff85735ce3915", size = 51965689 },
+    { url = "https://files.pythonhosted.org/packages/46/12/9de601dbae3c66666eeaaf5a28683d947909c046880baef390b7cd1d4b1d/jaxlib-0.4.30-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:ea3a00005faafbe3c18b178d3b534208b3b4027b2be6230227e7b87ce399fc29", size = 83544602 },
+    { url = "https://files.pythonhosted.org/packages/f3/1d/2d417a1445d5e696bb44d564c7519d4a6761db4d3e31712620c510ed0127/jaxlib-0.4.30-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3d31e01191ce8052bd611aaf16ff967d8d0ec0b63f1ea4b199020cecb248d667", size = 66695975 },
+    { url = "https://files.pythonhosted.org/packages/e4/f9/e29370046f4648bd464df7eceaebbbaefd091cc88c77da4a6e3a5f1a00d7/jaxlib-0.4.30-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:11602d5556e8baa2f16314c36518e9be4dfae0c2c256a361403fb29dc9dc79a4", size = 67784388 },
+    { url = "https://files.pythonhosted.org/packages/07/3b/a596036325666624ca084df554636fb3777e78e9386b52476d96fa14394e/jaxlib-0.4.30-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:f74a6b0e09df4b5e2ee399ebb9f0e01190e26e84ccb0a758fadb516415c07f18", size = 79643370 },
+    { url = "https://files.pythonhosted.org/packages/8a/a3/7342ceb02e49803af9a42ab3ad9b6c272cf7b2a83163e3a06859360012d5/jaxlib-0.4.30-cp39-cp39-win_amd64.whl", hash = "sha256:54987e97a22db70f3829b437b9329e4799d653634bacc8b398554d3b90c76b2a", size = 51946140 },
+]
+
+[[package]]
+name = "jaxlib"
+version = "0.5.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.13'",
+    "python_full_version == '3.12.*'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.13' or (python_full_version >= '3.10' and python_full_version < '3.12')" },
+    { name = "ml-dtypes", version = "0.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.12.*'" },
+    { name = "numpy", marker = "python_full_version >= '3.10'" },
+    { name = "scipy", version = "1.15.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c8/41/3e4ac64df72c4da126df3fd66a2214025a46b6263f7be266728e7b8e473e/jaxlib-0.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:1b8a6c4345f137f387650de2dbc488c20251b7412b55dd648e1a4f13bcf507fb", size = 79248968 },
+    { url = "https://files.pythonhosted.org/packages/1e/5f/2a16e61f1d54ae5f55fbf3cb3e22ef5bb01bf9d7d6474e0d34fedba19c4d/jaxlib-0.5.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:5b2efe3dfebf18a84c451d3803ac884ee242021c1113b279c13f4bbc378c3dc0", size = 93181077 },
+    { url = "https://files.pythonhosted.org/packages/08/c3/573e2f01b99f1247e8fbe1aa46b95a0faa68ef208f9a8e8ef775d607b3e6/jaxlib-0.5.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:74440b632107336400d4f97a16481d767f13ea914c53ba14e544c6fda54819b3", size = 101969119 },
+    { url = "https://files.pythonhosted.org/packages/6e/38/512f61ea13da41ca47f2411d7c05af0cf74a37f225e16725ed0e6fb58893/jaxlib-0.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:53478a28eee6c2ef01759b05a9491702daef9268c3ed013d6f8e2e5f5cae0887", size = 63883394 },
+    { url = "https://files.pythonhosted.org/packages/92/4b/8875870ff52ad3fbea876c905228f691f05c8dc8556b226cbfaf0fba7f62/jaxlib-0.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6cd762ed1623132499fa701c4203446102e0a9c82ca23194b87288f746d12a29", size = 79242870 },
+    { url = "https://files.pythonhosted.org/packages/a0/0f/00cdfa411d7218e4696c10c5867f7d3c396219adbcaeb02e95108ca802de/jaxlib-0.5.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:63088dbfaa85bb56cd521a925a3472fd7328b18ec93c2d8ffa85af331095c995", size = 93181807 },
+    { url = "https://files.pythonhosted.org/packages/58/8e/a5c29db03d5a93b0326e297b556d0e0a9805e9c9c1ae5f82f69557273faa/jaxlib-0.5.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:09113ef1582ba34d7cbc440fedb318f4855b59b776711a8aba2473c9727d3025", size = 101969212 },
+    { url = "https://files.pythonhosted.org/packages/70/86/ceae20e4f37fa07f1cc95551cc0f49170d0db46d2e82fdf511d26bffd801/jaxlib-0.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:78289fc3ddc1e4e9510de2536a6375df9fe1c50de0ac60826c286b7a5c5090fe", size = 63881994 },
+    { url = "https://files.pythonhosted.org/packages/57/d6/d971b40cb156e0637aa3c1522a1e803b641142e9a8f3ade6a574711bb073/jaxlib-0.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:73e335715760c56e635109d61426435a5d7f46f3363a115daea09427d5cd0efd", size = 79246087 },
+    { url = "https://files.pythonhosted.org/packages/41/2e/ba9770330077c3e4082cd0353e6a61419f79bff3e2197f904ce70167b9ad/jaxlib-0.5.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:4b4b01afb0ddec96c08356bff2bb685ddbe97fdffe4ed6e2d834b30aba972f22", size = 93179593 },
+    { url = "https://files.pythonhosted.org/packages/66/e9/211ba3e46ec22c722c4d61a739cfccf79b0618006d6f5fa53eb4eb93ed6d/jaxlib-0.5.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:f980c733e98c998a8da87c9a8cc61b6726d0be667a58bd664c1d717b4b4eae75", size = 101984785 },
+    { url = "https://files.pythonhosted.org/packages/2d/cb/11bb92324afb6ba678f388e10b78d6b02196bc8887eb5aa0d85ce398edf9/jaxlib-0.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:5baedbeeb60fa493c7528783254f04c6e986a2826266b198ed37e9336af2ef8c", size = 63899871 },
+    { url = "https://files.pythonhosted.org/packages/22/ac/e400473e6a2f405fd6e4dc40a713bb9a3868a3f76a8ffc5eb66f6e686002/jaxlib-0.5.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:ed18ea7161d03aa8fd4d1b55494882f21420efdfea68e5f298c4aebcf2ac3f34", size = 79245359 },
+    { url = "https://files.pythonhosted.org/packages/44/2d/c210abf4a9b2ce2e0858fcd3567c8773a739114e37d751af6c228901af57/jaxlib-0.5.0-cp313-cp313-manylinux2014_aarch64.whl", hash = "sha256:7d9b17a7ea19355d45ecdb2ff0db5d707a86f0c5a862d94b89b4568d6c45311a", size = 93180025 },
+    { url = "https://files.pythonhosted.org/packages/30/f8/316f7b4797c5eb50c6d70e461724a7cbe08b4505ca4da1bfd260c135895a/jaxlib-0.5.0-cp313-cp313-manylinux2014_x86_64.whl", hash = "sha256:11eef01d37c0f1c5306265b76f207f1002d13480ded2e31fd63ec76912c93ca2", size = 101982281 },
+    { url = "https://files.pythonhosted.org/packages/4c/f2/cfa012a0417c9b13b44c8e1d3ebf5fd04e8bb738b5c93e20c9fc97919880/jaxlib-0.5.0-cp313-cp313-win_amd64.whl", hash = "sha256:61b4d26cd6a0c49ba0b1e4340c7d29198913ee2dc70b65ee90752717d22305bb", size = 63900219 },
+]
+
+[[package]]
+name = "jinja2"
+version = "3.1.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "markupsafe" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/af/92/b3130cbbf5591acf9ade8708c365f3238046ac7cb8ccba6e81abccb0ccff/jinja2-3.1.5.tar.gz", hash = "sha256:8fefff8dc3034e27bb80d67c671eb8a9bc424c0ef4c0826edbff304cceff43bb", size = 244674 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/bd/0f/2ba5fbcd631e3e88689309dbe978c5769e883e4b84ebfe7da30b43275c5a/jinja2-3.1.5-py3-none-any.whl", hash = "sha256:aba0f4dc9ed8013c424088f68a5c226f7d6097ed89b246d7749c2ec4175c6adb", size = 134596 },
+]
+
+[[package]]
+name = "jiter"
+version = "0.8.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f8/70/90bc7bd3932e651486861df5c8ffea4ca7c77d28e8532ddefe2abc561a53/jiter-0.8.2.tar.gz", hash = "sha256:cd73d3e740666d0e639f678adb176fad25c1bcbdae88d8d7b857e1783bb4212d", size = 163007 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f2/f3/8c11e0e87bd5934c414f9b1cfae3cbfd4a938d4669d57cb427e1c4d11a7f/jiter-0.8.2-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:ca8577f6a413abe29b079bc30f907894d7eb07a865c4df69475e868d73e71c7b", size = 303381 },
+    { url = "https://files.pythonhosted.org/packages/ea/28/4cd3f0bcbf40e946bc6a62a82c951afc386a25673d3d8d5ee461f1559bbe/jiter-0.8.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:b25bd626bde7fb51534190c7e3cb97cee89ee76b76d7585580e22f34f5e3f393", size = 311718 },
+    { url = "https://files.pythonhosted.org/packages/0d/17/57acab00507e60bd954eaec0837d9d7b119b4117ff49b8a62f2b646f32ed/jiter-0.8.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d5c826a221851a8dc028eb6d7d6429ba03184fa3c7e83ae01cd6d3bd1d4bd17d", size = 335465 },
+    { url = "https://files.pythonhosted.org/packages/74/b9/1a3ddd2bc95ae17c815b021521020f40c60b32137730126bada962ef32b4/jiter-0.8.2-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:d35c864c2dff13dfd79fb070fc4fc6235d7b9b359efe340e1261deb21b9fcb66", size = 355570 },
+    { url = "https://files.pythonhosted.org/packages/78/69/6d29e2296a934199a7d0dde673ecccf98c9c8db44caf0248b3f2b65483cb/jiter-0.8.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f557c55bc2b7676e74d39d19bcb8775ca295c7a028246175d6a8b431e70835e5", size = 381383 },
+    { url = "https://files.pythonhosted.org/packages/22/d7/fbc4c3fb1bf65f9be22a32759b539f88e897aeb13fe84ab0266e4423487a/jiter-0.8.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:580ccf358539153db147e40751a0b41688a5ceb275e6f3e93d91c9467f42b2e3", size = 390454 },
+    { url = "https://files.pythonhosted.org/packages/4d/a0/3993cda2e267fe679b45d0bcc2cef0b4504b0aa810659cdae9737d6bace9/jiter-0.8.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:af102d3372e917cffce49b521e4c32c497515119dc7bd8a75665e90a718bbf08", size = 345039 },
+    { url = "https://files.pythonhosted.org/packages/b9/ef/69c18562b4c09ce88fab5df1dcaf643f6b1a8b970b65216e7221169b81c4/jiter-0.8.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:cadcc978f82397d515bb2683fc0d50103acff2a180552654bb92d6045dec2c49", size = 376200 },
+    { url = "https://files.pythonhosted.org/packages/4d/17/0b5a8de46a6ab4d836f70934036278b49b8530c292b29dde3483326d4555/jiter-0.8.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:ba5bdf56969cad2019d4e8ffd3f879b5fdc792624129741d3d83fc832fef8c7d", size = 511158 },
+    { url = "https://files.pythonhosted.org/packages/6c/b2/c401a0a2554b36c9e6d6e4876b43790d75139cf3936f0222e675cbc23451/jiter-0.8.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:3b94a33a241bee9e34b8481cdcaa3d5c2116f575e0226e421bed3f7a6ea71cff", size = 503956 },
+    { url = "https://files.pythonhosted.org/packages/d4/02/a0291ed7d72c0ac130f172354ee3cf0b2556b69584de391463a8ee534f40/jiter-0.8.2-cp310-cp310-win32.whl", hash = "sha256:6e5337bf454abddd91bd048ce0dca5134056fc99ca0205258766db35d0a2ea43", size = 202846 },
+    { url = "https://files.pythonhosted.org/packages/ad/20/8c988831ae4bf437e29f1671e198fc99ba8fe49f2895f23789acad1d1811/jiter-0.8.2-cp310-cp310-win_amd64.whl", hash = "sha256:4a9220497ca0cb1fe94e3f334f65b9b5102a0b8147646118f020d8ce1de70105", size = 204414 },
+    { url = "https://files.pythonhosted.org/packages/cb/b0/c1a7caa7f9dc5f1f6cfa08722867790fe2d3645d6e7170ca280e6e52d163/jiter-0.8.2-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:2dd61c5afc88a4fda7d8b2cf03ae5947c6ac7516d32b7a15bf4b49569a5c076b", size = 303666 },
+    { url = "https://files.pythonhosted.org/packages/f5/97/0468bc9eeae43079aaa5feb9267964e496bf13133d469cfdc135498f8dd0/jiter-0.8.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a6c710d657c8d1d2adbbb5c0b0c6bfcec28fd35bd6b5f016395f9ac43e878a15", size = 311934 },
+    { url = "https://files.pythonhosted.org/packages/e5/69/64058e18263d9a5f1e10f90c436853616d5f047d997c37c7b2df11b085ec/jiter-0.8.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a9584de0cd306072635fe4b89742bf26feae858a0683b399ad0c2509011b9dc0", size = 335506 },
+    { url = "https://files.pythonhosted.org/packages/9d/14/b747f9a77b8c0542141d77ca1e2a7523e854754af2c339ac89a8b66527d6/jiter-0.8.2-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5a90a923338531b7970abb063cfc087eebae6ef8ec8139762007188f6bc69a9f", size = 355849 },
+    { url = "https://files.pythonhosted.org/packages/53/e2/98a08161db7cc9d0e39bc385415890928ff09709034982f48eccfca40733/jiter-0.8.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d21974d246ed0181558087cd9f76e84e8321091ebfb3a93d4c341479a736f099", size = 381700 },
+    { url = "https://files.pythonhosted.org/packages/7a/38/1674672954d35bce3b1c9af99d5849f9256ac8f5b672e020ac7821581206/jiter-0.8.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:32475a42b2ea7b344069dc1e81445cfc00b9d0e3ca837f0523072432332e9f74", size = 389710 },
+    { url = "https://files.pythonhosted.org/packages/f8/9b/92f9da9a9e107d019bcf883cd9125fa1690079f323f5a9d5c6986eeec3c0/jiter-0.8.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8b9931fd36ee513c26b5bf08c940b0ac875de175341cbdd4fa3be109f0492586", size = 345553 },
+    { url = "https://files.pythonhosted.org/packages/44/a6/6d030003394e9659cd0d7136bbeabd82e869849ceccddc34d40abbbbb269/jiter-0.8.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:ce0820f4a3a59ddced7fce696d86a096d5cc48d32a4183483a17671a61edfddc", size = 376388 },
+    { url = "https://files.pythonhosted.org/packages/ad/8d/87b09e648e4aca5f9af89e3ab3cfb93db2d1e633b2f2931ede8dabd9b19a/jiter-0.8.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:8ffc86ae5e3e6a93765d49d1ab47b6075a9c978a2b3b80f0f32628f39caa0c88", size = 511226 },
+    { url = "https://files.pythonhosted.org/packages/77/95/8008ebe4cdc82eac1c97864a8042ca7e383ed67e0ec17bfd03797045c727/jiter-0.8.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:5127dc1abd809431172bc3fbe8168d6b90556a30bb10acd5ded41c3cfd6f43b6", size = 504134 },
+    { url = "https://files.pythonhosted.org/packages/26/0d/3056a74de13e8b2562e4d526de6dac2f65d91ace63a8234deb9284a1d24d/jiter-0.8.2-cp311-cp311-win32.whl", hash = "sha256:66227a2c7b575720c1871c8800d3a0122bb8ee94edb43a5685aa9aceb2782d44", size = 203103 },
+    { url = "https://files.pythonhosted.org/packages/4e/1e/7f96b798f356e531ffc0f53dd2f37185fac60fae4d6c612bbbd4639b90aa/jiter-0.8.2-cp311-cp311-win_amd64.whl", hash = "sha256:cde031d8413842a1e7501e9129b8e676e62a657f8ec8166e18a70d94d4682855", size = 206717 },
+    { url = "https://files.pythonhosted.org/packages/a1/17/c8747af8ea4e045f57d6cfd6fc180752cab9bc3de0e8a0c9ca4e8af333b1/jiter-0.8.2-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:e6ec2be506e7d6f9527dae9ff4b7f54e68ea44a0ef6b098256ddf895218a2f8f", size = 302027 },
+    { url = "https://files.pythonhosted.org/packages/3c/c1/6da849640cd35a41e91085723b76acc818d4b7d92b0b6e5111736ce1dd10/jiter-0.8.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:76e324da7b5da060287c54f2fabd3db5f76468006c811831f051942bf68c9d44", size = 310326 },
+    { url = "https://files.pythonhosted.org/packages/06/99/a2bf660d8ccffee9ad7ed46b4f860d2108a148d0ea36043fd16f4dc37e94/jiter-0.8.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:180a8aea058f7535d1c84183c0362c710f4750bef66630c05f40c93c2b152a0f", size = 334242 },
+    { url = "https://files.pythonhosted.org/packages/a7/5f/cea1c17864828731f11427b9d1ab7f24764dbd9aaf4648a7f851164d2718/jiter-0.8.2-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:025337859077b41548bdcbabe38698bcd93cfe10b06ff66617a48ff92c9aec60", size = 356654 },
+    { url = "https://files.pythonhosted.org/packages/e9/13/62774b7e5e7f5d5043efe1d0f94ead66e6d0f894ae010adb56b3f788de71/jiter-0.8.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ecff0dc14f409599bbcafa7e470c00b80f17abc14d1405d38ab02e4b42e55b57", size = 379967 },
+    { url = "https://files.pythonhosted.org/packages/ec/fb/096b34c553bb0bd3f2289d5013dcad6074948b8d55212aa13a10d44c5326/jiter-0.8.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ffd9fee7d0775ebaba131f7ca2e2d83839a62ad65e8e02fe2bd8fc975cedeb9e", size = 389252 },
+    { url = "https://files.pythonhosted.org/packages/17/61/beea645c0bf398ced8b199e377b61eb999d8e46e053bb285c91c3d3eaab0/jiter-0.8.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:14601dcac4889e0a1c75ccf6a0e4baf70dbc75041e51bcf8d0e9274519df6887", size = 345490 },
+    { url = "https://files.pythonhosted.org/packages/d5/df/834aa17ad5dcc3cf0118821da0a0cf1589ea7db9832589278553640366bc/jiter-0.8.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:92249669925bc1c54fcd2ec73f70f2c1d6a817928480ee1c65af5f6b81cdf12d", size = 376991 },
+    { url = "https://files.pythonhosted.org/packages/67/80/87d140399d382fb4ea5b3d56e7ecaa4efdca17cd7411ff904c1517855314/jiter-0.8.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:e725edd0929fa79f8349ab4ec7f81c714df51dc4e991539a578e5018fa4a7152", size = 510822 },
+    { url = "https://files.pythonhosted.org/packages/5c/37/3394bb47bac1ad2cb0465601f86828a0518d07828a650722e55268cdb7e6/jiter-0.8.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:bf55846c7b7a680eebaf9c3c48d630e1bf51bdf76c68a5f654b8524335b0ad29", size = 503730 },
+    { url = "https://files.pythonhosted.org/packages/f9/e2/253fc1fa59103bb4e3aa0665d6ceb1818df1cd7bf3eb492c4dad229b1cd4/jiter-0.8.2-cp312-cp312-win32.whl", hash = "sha256:7efe4853ecd3d6110301665a5178b9856be7e2a9485f49d91aa4d737ad2ae49e", size = 203375 },
+    { url = "https://files.pythonhosted.org/packages/41/69/6d4bbe66b3b3b4507e47aa1dd5d075919ad242b4b1115b3f80eecd443687/jiter-0.8.2-cp312-cp312-win_amd64.whl", hash = "sha256:83c0efd80b29695058d0fd2fa8a556490dbce9804eac3e281f373bbc99045f6c", size = 204740 },
+    { url = "https://files.pythonhosted.org/packages/6c/b0/bfa1f6f2c956b948802ef5a021281978bf53b7a6ca54bb126fd88a5d014e/jiter-0.8.2-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:ca1f08b8e43dc3bd0594c992fb1fd2f7ce87f7bf0d44358198d6da8034afdf84", size = 301190 },
+    { url = "https://files.pythonhosted.org/packages/a4/8f/396ddb4e292b5ea57e45ade5dc48229556b9044bad29a3b4b2dddeaedd52/jiter-0.8.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:5672a86d55416ccd214c778efccf3266b84f87b89063b582167d803246354be4", size = 309334 },
+    { url = "https://files.pythonhosted.org/packages/7f/68/805978f2f446fa6362ba0cc2e4489b945695940656edd844e110a61c98f8/jiter-0.8.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:58dc9bc9767a1101f4e5e22db1b652161a225874d66f0e5cb8e2c7d1c438b587", size = 333918 },
+    { url = "https://files.pythonhosted.org/packages/b3/99/0f71f7be667c33403fa9706e5b50583ae5106d96fab997fa7e2f38ee8347/jiter-0.8.2-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:37b2998606d6dadbb5ccda959a33d6a5e853252d921fec1792fc902351bb4e2c", size = 356057 },
+    { url = "https://files.pythonhosted.org/packages/8d/50/a82796e421a22b699ee4d2ce527e5bcb29471a2351cbdc931819d941a167/jiter-0.8.2-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4ab9a87f3784eb0e098f84a32670cfe4a79cb6512fd8f42ae3d0709f06405d18", size = 379790 },
+    { url = "https://files.pythonhosted.org/packages/3c/31/10fb012b00f6d83342ca9e2c9618869ab449f1aa78c8f1b2193a6b49647c/jiter-0.8.2-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:79aec8172b9e3c6d05fd4b219d5de1ac616bd8da934107325a6c0d0e866a21b6", size = 388285 },
+    { url = "https://files.pythonhosted.org/packages/c8/81/f15ebf7de57be488aa22944bf4274962aca8092e4f7817f92ffa50d3ee46/jiter-0.8.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:711e408732d4e9a0208008e5892c2966b485c783cd2d9a681f3eb147cf36c7ef", size = 344764 },
+    { url = "https://files.pythonhosted.org/packages/b3/e8/0cae550d72b48829ba653eb348cdc25f3f06f8a62363723702ec18e7be9c/jiter-0.8.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:653cf462db4e8c41995e33d865965e79641ef45369d8a11f54cd30888b7e6ff1", size = 376620 },
+    { url = "https://files.pythonhosted.org/packages/b8/50/e5478ff9d82534a944c03b63bc217c5f37019d4a34d288db0f079b13c10b/jiter-0.8.2-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:9c63eaef32b7bebac8ebebf4dabebdbc6769a09c127294db6babee38e9f405b9", size = 510402 },
+    { url = "https://files.pythonhosted.org/packages/8e/1e/3de48bbebbc8f7025bd454cedc8c62378c0e32dd483dece5f4a814a5cb55/jiter-0.8.2-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:eb21aaa9a200d0a80dacc7a81038d2e476ffe473ffdd9c91eb745d623561de05", size = 503018 },
+    { url = "https://files.pythonhosted.org/packages/d5/cd/d5a5501d72a11fe3e5fd65c78c884e5164eefe80077680533919be22d3a3/jiter-0.8.2-cp313-cp313-win32.whl", hash = "sha256:789361ed945d8d42850f919342a8665d2dc79e7e44ca1c97cc786966a21f627a", size = 203190 },
+    { url = "https://files.pythonhosted.org/packages/51/bf/e5ca301245ba951447e3ad677a02a64a8845b185de2603dabd83e1e4b9c6/jiter-0.8.2-cp313-cp313-win_amd64.whl", hash = "sha256:ab7f43235d71e03b941c1630f4b6e3055d46b6cb8728a17663eaac9d8e83a865", size = 203551 },
+    { url = "https://files.pythonhosted.org/packages/2f/3c/71a491952c37b87d127790dd7a0b1ebea0514c6b6ad30085b16bbe00aee6/jiter-0.8.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:b426f72cd77da3fec300ed3bc990895e2dd6b49e3bfe6c438592a3ba660e41ca", size = 308347 },
+    { url = "https://files.pythonhosted.org/packages/a0/4c/c02408042e6a7605ec063daed138e07b982fdb98467deaaf1c90950cf2c6/jiter-0.8.2-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b2dd880785088ff2ad21ffee205e58a8c1ddabc63612444ae41e5e4b321b39c0", size = 342875 },
+    { url = "https://files.pythonhosted.org/packages/91/61/c80ef80ed8a0a21158e289ef70dac01e351d929a1c30cb0f49be60772547/jiter-0.8.2-cp313-cp313t-win_amd64.whl", hash = "sha256:3ac9f578c46f22405ff7f8b1f5848fb753cc4b8377fbec8470a7dc3997ca7566", size = 202374 },
+    { url = "https://files.pythonhosted.org/packages/c9/b2/ed7fbabd21c3cf556d6ea849cee35c74f13a509e668baad8323091e2867e/jiter-0.8.2-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:e41e75344acef3fc59ba4765df29f107f309ca9e8eace5baacabd9217e52a5ee", size = 304502 },
+    { url = "https://files.pythonhosted.org/packages/75/6e/1386857ac9165c1e9c71031566e7884d8a4f63724ce29ad1ace5bfe1351c/jiter-0.8.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:7f22b16b35d5c1df9dfd58843ab2cd25e6bf15191f5a236bed177afade507bfc", size = 300982 },
+    { url = "https://files.pythonhosted.org/packages/56/4c/b413977c20bbb359b4d6c91d04f7f36fc525af0b7778119815477fc97242/jiter-0.8.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f7200b8f7619d36aa51c803fd52020a2dfbea36ffec1b5e22cab11fd34d95a6d", size = 335344 },
+    { url = "https://files.pythonhosted.org/packages/b0/59/51b080519938192edd33b4e8d48adb7e9bf9e0d699ec8b91119b9269fc75/jiter-0.8.2-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:70bf4c43652cc294040dbb62256c83c8718370c8b93dd93d934b9a7bf6c4f53c", size = 356298 },
+    { url = "https://files.pythonhosted.org/packages/72/bb/828db5ea406916d7b2232be31393f782b0f71bcb0b128750c4a028157565/jiter-0.8.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f9d471356dc16f84ed48768b8ee79f29514295c7295cb41e1133ec0b2b8d637d", size = 381703 },
+    { url = "https://files.pythonhosted.org/packages/c0/88/45d33a8728733e161e9783c54d8ecca0fc4c1aa74b1cebea1d97917eddc3/jiter-0.8.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:859e8eb3507894093d01929e12e267f83b1d5f6221099d3ec976f0c995cb6bd9", size = 391281 },
+    { url = "https://files.pythonhosted.org/packages/45/3e/142712e0f45c28ad8a678dc8732a78294ce5a36fc694141f772bb827a8f2/jiter-0.8.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:eaa58399c01db555346647a907b4ef6d4f584b123943be6ed5588c3f2359c9f4", size = 345553 },
+    { url = "https://files.pythonhosted.org/packages/36/42/9b463b59fd22687b6da1afcad6c9adc870464a808208651de73f1dbeda09/jiter-0.8.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:8f2d5ed877f089862f4c7aacf3a542627c1496f972a34d0474ce85ee7d939c27", size = 377063 },
+    { url = "https://files.pythonhosted.org/packages/83/b3/44b1f5cd2e4eb15757eec341b25399da4c90515bb881ef6636b50a8c08a5/jiter-0.8.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:03c9df035d4f8d647f8c210ddc2ae0728387275340668fb30d2421e17d9a0841", size = 512543 },
+    { url = "https://files.pythonhosted.org/packages/46/4e/c695c803aa2b668c057b2dea1cdd7a884d1a819ce610cec0be9666210bfd/jiter-0.8.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:8bd2a824d08d8977bb2794ea2682f898ad3d8837932e3a74937e93d62ecbb637", size = 505141 },
+    { url = "https://files.pythonhosted.org/packages/8e/51/e805b837db056f872db0b7a7a3610b7d764392be696dbe47afa0bea05bf2/jiter-0.8.2-cp39-cp39-win32.whl", hash = "sha256:ca29b6371ebc40e496995c94b988a101b9fbbed48a51190a4461fcb0a68b4a36", size = 203529 },
+    { url = "https://files.pythonhosted.org/packages/32/b7/a3cde72c644fd1caf9da07fb38cf2c130f43484d8f91011940b7c4f42c8f/jiter-0.8.2-cp39-cp39-win_amd64.whl", hash = "sha256:1c0dfbd1be3cbefc7510102370d86e35d1d53e5a93d48519688b1bf0f761160a", size = 207527 },
+]
+
+[[package]]
+name = "jsonschema"
+version = "4.23.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "attrs" },
+    { name = "jsonschema-specifications" },
+    { name = "referencing" },
+    { name = "rpds-py" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/38/2e/03362ee4034a4c917f697890ccd4aec0800ccf9ded7f511971c75451deec/jsonschema-4.23.0.tar.gz", hash = "sha256:d71497fef26351a33265337fa77ffeb82423f3ea21283cd9467bb03999266bc4", size = 325778 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/69/4a/4f9dbeb84e8850557c02365a0eee0649abe5eb1d84af92a25731c6c0f922/jsonschema-4.23.0-py3-none-any.whl", hash = "sha256:fbadb6f8b144a8f8cf9f0b89ba94501d143e50411a1278633f56a7acf7fd5566", size = 88462 },
+]
+
+[[package]]
+name = "jsonschema-specifications"
+version = "2024.10.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "referencing" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/10/db/58f950c996c793472e336ff3655b13fbcf1e3b359dcf52dcf3ed3b52c352/jsonschema_specifications-2024.10.1.tar.gz", hash = "sha256:0f38b83639958ce1152d02a7f062902c41c8fd20d558b0c34344292d417ae272", size = 15561 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d1/0f/8910b19ac0670a0f80ce1008e5e751c4a57e14d2c4c13a482aa6079fa9d6/jsonschema_specifications-2024.10.1-py3-none-any.whl", hash = "sha256:a09a0680616357d9a0ecf05c12ad234479f549239d0f5b55f3deea67475da9bf", size = 18459 },
+]
+
+[[package]]
+name = "lark"
+version = "1.2.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/af/60/bc7622aefb2aee1c0b4ba23c1446d3e30225c8770b38d7aedbfb65ca9d5a/lark-1.2.2.tar.gz", hash = "sha256:ca807d0162cd16cef15a8feecb862d7319e7a09bdb13aef927968e45040fed80", size = 252132 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2d/00/d90b10b962b4277f5e64a78b6609968859ff86889f5b898c1a778c06ec00/lark-1.2.2-py3-none-any.whl", hash = "sha256:c2276486b02f0f1b90be155f2c8ba4a8e194d42775786db622faccd652d8e80c", size = 111036 },
+]
+
+[[package]]
+name = "llama-cpp-python"
+version = "0.3.6"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "diskcache" },
+    { name = "jinja2" },
+    { name = "numpy" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/72/cd/efae0a18621ba66baf19d19507bb17f1aa2dd5e7f0d5dd4aa008c9bc7091/llama_cpp_python-0.3.6.tar.gz", hash = "sha256:86e35a8888274466958e24201b856cd71c8def0ea72e14312be13da96c15c7a4", size = 66893177 }
+
+[[package]]
+name = "lm-format-enforcer"
+version = "0.10.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "interegular" },
+    { name = "packaging" },
+    { name = "pydantic" },
+    { name = "pyyaml" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/bb/ac/11fb977637125ed92adb1de4138b0e365918465e3e4e39a437fdb8156746/lm_format_enforcer-0.10.1.tar.gz", hash = "sha256:23e65a4199714fca348063e8c906838622619f905a673c4d6d428eee7e7d2095", size = 38781 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2d/fd/e21d1f0608da7696746c5d4cdf2f944d714daf63a97def416e0f3c154f98/lm_format_enforcer-0.10.1-py3-none-any.whl", hash = "sha256:5520004af248d787930327ead052aeff75e21fad595f388e5eade9f062ffddda", size = 42929 },
+]
+
+[[package]]
+name = "markdown-it-py"
+version = "3.0.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "mdurl" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/38/71/3b932df36c1a044d397a1f92d1cf91ee0a503d91e470cbd670aa66b07ed0/markdown-it-py-3.0.0.tar.gz", hash = "sha256:e3f60a94fa066dc52ec76661e37c851cb232d92f9886b15cb560aaada2df8feb", size = 74596 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/42/d7/1ec15b46af6af88f19b8e5ffea08fa375d433c998b8a7639e76935c14f1f/markdown_it_py-3.0.0-py3-none-any.whl", hash = "sha256:355216845c60bd96232cd8d8c40e8f9765cc86f46880e43a8fd22dc1a1a8cab1", size = 87528 },
+]
+
+[[package]]
+name = "markupsafe"
+version = "3.0.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/b2/97/5d42485e71dfc078108a86d6de8fa46db44a1a9295e89c5d6d4a06e23a62/markupsafe-3.0.2.tar.gz", hash = "sha256:ee55d3edf80167e48ea11a923c7386f4669df67d7994554387f84e7d8b0a2bf0", size = 20537 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/04/90/d08277ce111dd22f77149fd1a5d4653eeb3b3eaacbdfcbae5afb2600eebd/MarkupSafe-3.0.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7e94c425039cde14257288fd61dcfb01963e658efbc0ff54f5306b06054700f8", size = 14357 },
+    { url = "https://files.pythonhosted.org/packages/04/e1/6e2194baeae0bca1fae6629dc0cbbb968d4d941469cbab11a3872edff374/MarkupSafe-3.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9e2d922824181480953426608b81967de705c3cef4d1af983af849d7bd619158", size = 12393 },
+    { url = "https://files.pythonhosted.org/packages/1d/69/35fa85a8ece0a437493dc61ce0bb6d459dcba482c34197e3efc829aa357f/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:38a9ef736c01fccdd6600705b09dc574584b89bea478200c5fbf112a6b0d5579", size = 21732 },
+    { url = "https://files.pythonhosted.org/packages/22/35/137da042dfb4720b638d2937c38a9c2df83fe32d20e8c8f3185dbfef05f7/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bbcb445fa71794da8f178f0f6d66789a28d7319071af7a496d4d507ed566270d", size = 20866 },
+    { url = "https://files.pythonhosted.org/packages/29/28/6d029a903727a1b62edb51863232152fd335d602def598dade38996887f0/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:57cb5a3cf367aeb1d316576250f65edec5bb3be939e9247ae594b4bcbc317dfb", size = 20964 },
+    { url = "https://files.pythonhosted.org/packages/cc/cd/07438f95f83e8bc028279909d9c9bd39e24149b0d60053a97b2bc4f8aa51/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:3809ede931876f5b2ec92eef964286840ed3540dadf803dd570c3b7e13141a3b", size = 21977 },
+    { url = "https://files.pythonhosted.org/packages/29/01/84b57395b4cc062f9c4c55ce0df7d3108ca32397299d9df00fedd9117d3d/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e07c3764494e3776c602c1e78e298937c3315ccc9043ead7e685b7f2b8d47b3c", size = 21366 },
+    { url = "https://files.pythonhosted.org/packages/bd/6e/61ebf08d8940553afff20d1fb1ba7294b6f8d279df9fd0c0db911b4bbcfd/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b424c77b206d63d500bcb69fa55ed8d0e6a3774056bdc4839fc9298a7edca171", size = 21091 },
+    { url = "https://files.pythonhosted.org/packages/11/23/ffbf53694e8c94ebd1e7e491de185124277964344733c45481f32ede2499/MarkupSafe-3.0.2-cp310-cp310-win32.whl", hash = "sha256:fcabf5ff6eea076f859677f5f0b6b5c1a51e70a376b0579e0eadef8db48c6b50", size = 15065 },
+    { url = "https://files.pythonhosted.org/packages/44/06/e7175d06dd6e9172d4a69a72592cb3f7a996a9c396eee29082826449bbc3/MarkupSafe-3.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:6af100e168aa82a50e186c82875a5893c5597a0c1ccdb0d8b40240b1f28b969a", size = 15514 },
+    { url = "https://files.pythonhosted.org/packages/6b/28/bbf83e3f76936960b850435576dd5e67034e200469571be53f69174a2dfd/MarkupSafe-3.0.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:9025b4018f3a1314059769c7bf15441064b2207cb3f065e6ea1e7359cb46db9d", size = 14353 },
+    { url = "https://files.pythonhosted.org/packages/6c/30/316d194b093cde57d448a4c3209f22e3046c5bb2fb0820b118292b334be7/MarkupSafe-3.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:93335ca3812df2f366e80509ae119189886b0f3c2b81325d39efdb84a1e2ae93", size = 12392 },
+    { url = "https://files.pythonhosted.org/packages/f2/96/9cdafba8445d3a53cae530aaf83c38ec64c4d5427d975c974084af5bc5d2/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2cb8438c3cbb25e220c2ab33bb226559e7afb3baec11c4f218ffa7308603c832", size = 23984 },
+    { url = "https://files.pythonhosted.org/packages/f1/a4/aefb044a2cd8d7334c8a47d3fb2c9f328ac48cb349468cc31c20b539305f/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a123e330ef0853c6e822384873bef7507557d8e4a082961e1defa947aa59ba84", size = 23120 },
+    { url = "https://files.pythonhosted.org/packages/8d/21/5e4851379f88f3fad1de30361db501300d4f07bcad047d3cb0449fc51f8c/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1e084f686b92e5b83186b07e8a17fc09e38fff551f3602b249881fec658d3eca", size = 23032 },
+    { url = "https://files.pythonhosted.org/packages/00/7b/e92c64e079b2d0d7ddf69899c98842f3f9a60a1ae72657c89ce2655c999d/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d8213e09c917a951de9d09ecee036d5c7d36cb6cb7dbaece4c71a60d79fb9798", size = 24057 },
+    { url = "https://files.pythonhosted.org/packages/f9/ac/46f960ca323037caa0a10662ef97d0a4728e890334fc156b9f9e52bcc4ca/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:5b02fb34468b6aaa40dfc198d813a641e3a63b98c2b05a16b9f80b7ec314185e", size = 23359 },
+    { url = "https://files.pythonhosted.org/packages/69/84/83439e16197337b8b14b6a5b9c2105fff81d42c2a7c5b58ac7b62ee2c3b1/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0bff5e0ae4ef2e1ae4fdf2dfd5b76c75e5c2fa4132d05fc1b0dabcd20c7e28c4", size = 23306 },
+    { url = "https://files.pythonhosted.org/packages/9a/34/a15aa69f01e2181ed8d2b685c0d2f6655d5cca2c4db0ddea775e631918cd/MarkupSafe-3.0.2-cp311-cp311-win32.whl", hash = "sha256:6c89876f41da747c8d3677a2b540fb32ef5715f97b66eeb0c6b66f5e3ef6f59d", size = 15094 },
+    { url = "https://files.pythonhosted.org/packages/da/b8/3a3bd761922d416f3dc5d00bfbed11f66b1ab89a0c2b6e887240a30b0f6b/MarkupSafe-3.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:70a87b411535ccad5ef2f1df5136506a10775d267e197e4cf531ced10537bd6b", size = 15521 },
+    { url = "https://files.pythonhosted.org/packages/22/09/d1f21434c97fc42f09d290cbb6350d44eb12f09cc62c9476effdb33a18aa/MarkupSafe-3.0.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:9778bd8ab0a994ebf6f84c2b949e65736d5575320a17ae8984a77fab08db94cf", size = 14274 },
+    { url = "https://files.pythonhosted.org/packages/6b/b0/18f76bba336fa5aecf79d45dcd6c806c280ec44538b3c13671d49099fdd0/MarkupSafe-3.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:846ade7b71e3536c4e56b386c2a47adf5741d2d8b94ec9dc3e92e5e1ee1e2225", size = 12348 },
+    { url = "https://files.pythonhosted.org/packages/e0/25/dd5c0f6ac1311e9b40f4af06c78efde0f3b5cbf02502f8ef9501294c425b/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1c99d261bd2d5f6b59325c92c73df481e05e57f19837bdca8413b9eac4bd8028", size = 24149 },
+    { url = "https://files.pythonhosted.org/packages/f3/f0/89e7aadfb3749d0f52234a0c8c7867877876e0a20b60e2188e9850794c17/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e17c96c14e19278594aa4841ec148115f9c7615a47382ecb6b82bd8fea3ab0c8", size = 23118 },
+    { url = "https://files.pythonhosted.org/packages/d5/da/f2eeb64c723f5e3777bc081da884b414671982008c47dcc1873d81f625b6/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:88416bd1e65dcea10bc7569faacb2c20ce071dd1f87539ca2ab364bf6231393c", size = 22993 },
+    { url = "https://files.pythonhosted.org/packages/da/0e/1f32af846df486dce7c227fe0f2398dc7e2e51d4a370508281f3c1c5cddc/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2181e67807fc2fa785d0592dc2d6206c019b9502410671cc905d132a92866557", size = 24178 },
+    { url = "https://files.pythonhosted.org/packages/c4/f6/bb3ca0532de8086cbff5f06d137064c8410d10779c4c127e0e47d17c0b71/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:52305740fe773d09cffb16f8ed0427942901f00adedac82ec8b67752f58a1b22", size = 23319 },
+    { url = "https://files.pythonhosted.org/packages/a2/82/8be4c96ffee03c5b4a034e60a31294daf481e12c7c43ab8e34a1453ee48b/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ad10d3ded218f1039f11a75f8091880239651b52e9bb592ca27de44eed242a48", size = 23352 },
+    { url = "https://files.pythonhosted.org/packages/51/ae/97827349d3fcffee7e184bdf7f41cd6b88d9919c80f0263ba7acd1bbcb18/MarkupSafe-3.0.2-cp312-cp312-win32.whl", hash = "sha256:0f4ca02bea9a23221c0182836703cbf8930c5e9454bacce27e767509fa286a30", size = 15097 },
+    { url = "https://files.pythonhosted.org/packages/c1/80/a61f99dc3a936413c3ee4e1eecac96c0da5ed07ad56fd975f1a9da5bc630/MarkupSafe-3.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:8e06879fc22a25ca47312fbe7c8264eb0b662f6db27cb2d3bbbc74b1df4b9b87", size = 15601 },
+    { url = "https://files.pythonhosted.org/packages/83/0e/67eb10a7ecc77a0c2bbe2b0235765b98d164d81600746914bebada795e97/MarkupSafe-3.0.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ba9527cdd4c926ed0760bc301f6728ef34d841f405abf9d4f959c478421e4efd", size = 14274 },
+    { url = "https://files.pythonhosted.org/packages/2b/6d/9409f3684d3335375d04e5f05744dfe7e9f120062c9857df4ab490a1031a/MarkupSafe-3.0.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f8b3d067f2e40fe93e1ccdd6b2e1d16c43140e76f02fb1319a05cf2b79d99430", size = 12352 },
+    { url = "https://files.pythonhosted.org/packages/d2/f5/6eadfcd3885ea85fe2a7c128315cc1bb7241e1987443d78c8fe712d03091/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:569511d3b58c8791ab4c2e1285575265991e6d8f8700c7be0e88f86cb0672094", size = 24122 },
+    { url = "https://files.pythonhosted.org/packages/0c/91/96cf928db8236f1bfab6ce15ad070dfdd02ed88261c2afafd4b43575e9e9/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:15ab75ef81add55874e7ab7055e9c397312385bd9ced94920f2802310c930396", size = 23085 },
+    { url = "https://files.pythonhosted.org/packages/c2/cf/c9d56af24d56ea04daae7ac0940232d31d5a8354f2b457c6d856b2057d69/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f3818cb119498c0678015754eba762e0d61e5b52d34c8b13d770f0719f7b1d79", size = 22978 },
+    { url = "https://files.pythonhosted.org/packages/2a/9f/8619835cd6a711d6272d62abb78c033bda638fdc54c4e7f4272cf1c0962b/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:cdb82a876c47801bb54a690c5ae105a46b392ac6099881cdfb9f6e95e4014c6a", size = 24208 },
+    { url = "https://files.pythonhosted.org/packages/f9/bf/176950a1792b2cd2102b8ffeb5133e1ed984547b75db47c25a67d3359f77/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:cabc348d87e913db6ab4aa100f01b08f481097838bdddf7c7a84b7575b7309ca", size = 23357 },
+    { url = "https://files.pythonhosted.org/packages/ce/4f/9a02c1d335caabe5c4efb90e1b6e8ee944aa245c1aaaab8e8a618987d816/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:444dcda765c8a838eaae23112db52f1efaf750daddb2d9ca300bcae1039adc5c", size = 23344 },
+    { url = "https://files.pythonhosted.org/packages/ee/55/c271b57db36f748f0e04a759ace9f8f759ccf22b4960c270c78a394f58be/MarkupSafe-3.0.2-cp313-cp313-win32.whl", hash = "sha256:bcf3e58998965654fdaff38e58584d8937aa3096ab5354d493c77d1fdd66d7a1", size = 15101 },
+    { url = "https://files.pythonhosted.org/packages/29/88/07df22d2dd4df40aba9f3e402e6dc1b8ee86297dddbad4872bd5e7b0094f/MarkupSafe-3.0.2-cp313-cp313-win_amd64.whl", hash = "sha256:e6a2a455bd412959b57a172ce6328d2dd1f01cb2135efda2e4576e8a23fa3b0f", size = 15603 },
+    { url = "https://files.pythonhosted.org/packages/62/6a/8b89d24db2d32d433dffcd6a8779159da109842434f1dd2f6e71f32f738c/MarkupSafe-3.0.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:b5a6b3ada725cea8a5e634536b1b01c30bcdcd7f9c6fff4151548d5bf6b3a36c", size = 14510 },
+    { url = "https://files.pythonhosted.org/packages/7a/06/a10f955f70a2e5a9bf78d11a161029d278eeacbd35ef806c3fd17b13060d/MarkupSafe-3.0.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:a904af0a6162c73e3edcb969eeeb53a63ceeb5d8cf642fade7d39e7963a22ddb", size = 12486 },
+    { url = "https://files.pythonhosted.org/packages/34/cf/65d4a571869a1a9078198ca28f39fba5fbb910f952f9dbc5220afff9f5e6/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4aa4e5faecf353ed117801a068ebab7b7e09ffb6e1d5e412dc852e0da018126c", size = 25480 },
+    { url = "https://files.pythonhosted.org/packages/0c/e3/90e9651924c430b885468b56b3d597cabf6d72be4b24a0acd1fa0e12af67/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0ef13eaeee5b615fb07c9a7dadb38eac06a0608b41570d8ade51c56539e509d", size = 23914 },
+    { url = "https://files.pythonhosted.org/packages/66/8c/6c7cf61f95d63bb866db39085150df1f2a5bd3335298f14a66b48e92659c/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d16a81a06776313e817c951135cf7340a3e91e8c1ff2fac444cfd75fffa04afe", size = 23796 },
+    { url = "https://files.pythonhosted.org/packages/bb/35/cbe9238ec3f47ac9a7c8b3df7a808e7cb50fe149dc7039f5f454b3fba218/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:6381026f158fdb7c72a168278597a5e3a5222e83ea18f543112b2662a9b699c5", size = 25473 },
+    { url = "https://files.pythonhosted.org/packages/e6/32/7621a4382488aa283cc05e8984a9c219abad3bca087be9ec77e89939ded9/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:3d79d162e7be8f996986c064d1c7c817f6df3a77fe3d6859f6f9e7be4b8c213a", size = 24114 },
+    { url = "https://files.pythonhosted.org/packages/0d/80/0985960e4b89922cb5a0bac0ed39c5b96cbc1a536a99f30e8c220a996ed9/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:131a3c7689c85f5ad20f9f6fb1b866f402c445b220c19fe4308c0b147ccd2ad9", size = 24098 },
+    { url = "https://files.pythonhosted.org/packages/82/78/fedb03c7d5380df2427038ec8d973587e90561b2d90cd472ce9254cf348b/MarkupSafe-3.0.2-cp313-cp313t-win32.whl", hash = "sha256:ba8062ed2cf21c07a9e295d5b8a2a5ce678b913b45fdf68c32d95d6c1291e0b6", size = 15208 },
+    { url = "https://files.pythonhosted.org/packages/4f/65/6079a46068dfceaeabb5dcad6d674f5f5c61a6fa5673746f42a9f4c233b3/MarkupSafe-3.0.2-cp313-cp313t-win_amd64.whl", hash = "sha256:e444a31f8db13eb18ada366ab3cf45fd4b31e4db1236a4448f68778c1d1a5a2f", size = 15739 },
+    { url = "https://files.pythonhosted.org/packages/a7/ea/9b1530c3fdeeca613faeb0fb5cbcf2389d816072fab72a71b45749ef6062/MarkupSafe-3.0.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:eaa0a10b7f72326f1372a713e73c3f739b524b3af41feb43e4921cb529f5929a", size = 14344 },
+    { url = "https://files.pythonhosted.org/packages/4b/c2/fbdbfe48848e7112ab05e627e718e854d20192b674952d9042ebd8c9e5de/MarkupSafe-3.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:48032821bbdf20f5799ff537c7ac3d1fba0ba032cfc06194faffa8cda8b560ff", size = 12389 },
+    { url = "https://files.pythonhosted.org/packages/f0/25/7a7c6e4dbd4f867d95d94ca15449e91e52856f6ed1905d58ef1de5e211d0/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1a9d3f5f0901fdec14d8d2f66ef7d035f2157240a433441719ac9a3fba440b13", size = 21607 },
+    { url = "https://files.pythonhosted.org/packages/53/8f/f339c98a178f3c1e545622206b40986a4c3307fe39f70ccd3d9df9a9e425/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88b49a3b9ff31e19998750c38e030fc7bb937398b1f78cfa599aaef92d693144", size = 20728 },
+    { url = "https://files.pythonhosted.org/packages/1a/03/8496a1a78308456dbd50b23a385c69b41f2e9661c67ea1329849a598a8f9/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:cfad01eed2c2e0c01fd0ecd2ef42c492f7f93902e39a42fc9ee1692961443a29", size = 20826 },
+    { url = "https://files.pythonhosted.org/packages/e6/cf/0a490a4bd363048c3022f2f475c8c05582179bb179defcee4766fb3dcc18/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1225beacc926f536dc82e45f8a4d68502949dc67eea90eab715dea3a21c1b5f0", size = 21843 },
+    { url = "https://files.pythonhosted.org/packages/19/a3/34187a78613920dfd3cdf68ef6ce5e99c4f3417f035694074beb8848cd77/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:3169b1eefae027567d1ce6ee7cae382c57fe26e82775f460f0b2778beaad66c0", size = 21219 },
+    { url = "https://files.pythonhosted.org/packages/17/d8/5811082f85bb88410ad7e452263af048d685669bbbfb7b595e8689152498/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:eb7972a85c54febfb25b5c4b4f3af4dcc731994c7da0d8a0b4a6eb0640e1d178", size = 20946 },
+    { url = "https://files.pythonhosted.org/packages/7c/31/bd635fb5989440d9365c5e3c47556cfea121c7803f5034ac843e8f37c2f2/MarkupSafe-3.0.2-cp39-cp39-win32.whl", hash = "sha256:8c4e8c3ce11e1f92f6536ff07154f9d49677ebaaafc32db9db4620bc11ed480f", size = 15063 },
+    { url = "https://files.pythonhosted.org/packages/b3/73/085399401383ce949f727afec55ec3abd76648d04b9f22e1c0e99cb4bec3/MarkupSafe-3.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:6e296a513ca3d94054c2c881cc913116e90fd030ad1c656b3869762b754f5f8a", size = 15506 },
+]
+
+[[package]]
+name = "mdurl"
+version = "0.1.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d6/54/cfe61301667036ec958cb99bd3efefba235e65cdeb9c84d24a8293ba1d90/mdurl-0.1.2.tar.gz", hash = "sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba", size = 8729 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b3/38/89ba8ad64ae25be8de66a6d463314cf1eb366222074cfda9ee839c56a4b4/mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8", size = 9979 },
+]
+
+[[package]]
+name = "mkl"
+version = "2021.4.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "intel-openmp" },
+    { name = "tbb" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ce/c6/892fe3bc91e811b78e4f85653864f2d92541d5e5c306b0cb3c2311e9ca64/mkl-2021.4.0-py2.py3-none-win32.whl", hash = "sha256:439c640b269a5668134e3dcbcea4350459c4a8bc46469669b2d67e07e3d330e8", size = 129048357 },
+    { url = "https://files.pythonhosted.org/packages/fe/1c/5f6dbf18e8b73e0a5472466f0ea8d48ce9efae39bd2ff38cebf8dce61259/mkl-2021.4.0-py2.py3-none-win_amd64.whl", hash = "sha256:ceef3cafce4c009dd25f65d7ad0d833a0fbadc3d8903991ec92351fe5de1e718", size = 228499609 },
+]
+
+[[package]]
+name = "ml-dtypes"
+version = "0.4.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.13'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
+    "python_full_version < '3.9.17'",
+]
+dependencies = [
+    { name = "numpy", marker = "python_full_version != '3.12.*'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fd/15/76f86faa0902836cc133939732f7611ace68cf54148487a99c539c272dc8/ml_dtypes-0.4.1.tar.gz", hash = "sha256:fad5f2de464fd09127e49b7fd1252b9006fb43d2edc1ff112d390c324af5ca7a", size = 692594 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/56/9e/76b84f77c7afee3b116dc8407903a2d5004ba3059a8f3dcdcfa6ebf33fff/ml_dtypes-0.4.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:1fe8b5b5e70cd67211db94b05cfd58dace592f24489b038dc6f9fe347d2e07d5", size = 397975 },
+    { url = "https://files.pythonhosted.org/packages/03/7b/32650e1b2a2713a5923a0af2a8503d0d4a8fc99d1e1e0a1c40e996634460/ml_dtypes-0.4.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8c09a6d11d8475c2a9fd2bc0695628aec105f97cab3b3a3fb7c9660348ff7d24", size = 2182570 },
+    { url = "https://files.pythonhosted.org/packages/16/86/a9f7569e7e4f5395f927de38a13b92efa73f809285d04f2923b291783dd2/ml_dtypes-0.4.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9f5e8f75fa371020dd30f9196e7d73babae2abd51cf59bdd56cb4f8de7e13354", size = 2160365 },
+    { url = "https://files.pythonhosted.org/packages/04/1b/9a3afb437702503514f3934ec8d7904270edf013d28074f3e700e5dfbb0f/ml_dtypes-0.4.1-cp310-cp310-win_amd64.whl", hash = "sha256:15fdd922fea57e493844e5abb930b9c0bd0af217d9edd3724479fc3d7ce70e3f", size = 126633 },
+    { url = "https://files.pythonhosted.org/packages/d1/76/9835c8609c29f2214359e88f29255fc4aad4ea0f613fb48aa8815ceda1b6/ml_dtypes-0.4.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:2d55b588116a7085d6e074cf0cdb1d6fa3875c059dddc4d2c94a4cc81c23e975", size = 397973 },
+    { url = "https://files.pythonhosted.org/packages/7e/99/e68c56fac5de973007a10254b6e17a0362393724f40f66d5e4033f4962c2/ml_dtypes-0.4.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e138a9b7a48079c900ea969341a5754019a1ad17ae27ee330f7ebf43f23877f9", size = 2185134 },
+    { url = "https://files.pythonhosted.org/packages/28/bc/6a2344338ea7b61cd7b46fb24ec459360a5a0903b57c55b156c1e46c644a/ml_dtypes-0.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:74c6cfb5cf78535b103fde9ea3ded8e9f16f75bc07789054edc7776abfb3d752", size = 2163661 },
+    { url = "https://files.pythonhosted.org/packages/e8/d3/ddfd9878b223b3aa9a930c6100a99afca5cfab7ea703662e00323acb7568/ml_dtypes-0.4.1-cp311-cp311-win_amd64.whl", hash = "sha256:274cc7193dd73b35fb26bef6c5d40ae3eb258359ee71cd82f6e96a8c948bdaa6", size = 126727 },
+    { url = "https://files.pythonhosted.org/packages/ba/1a/99e924f12e4b62139fbac87419698c65f956d58de0dbfa7c028fa5b096aa/ml_dtypes-0.4.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:827d3ca2097085cf0355f8fdf092b888890bb1b1455f52801a2d7756f056f54b", size = 405077 },
+    { url = "https://files.pythonhosted.org/packages/8f/8c/7b610bd500617854c8cc6ed7c8cfb9d48d6a5c21a1437a36a4b9bc8a3598/ml_dtypes-0.4.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:772426b08a6172a891274d581ce58ea2789cc8abc1c002a27223f314aaf894e7", size = 2181554 },
+    { url = "https://files.pythonhosted.org/packages/c7/c6/f89620cecc0581dc1839e218c4315171312e46c62a62da6ace204bda91c0/ml_dtypes-0.4.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:126e7d679b8676d1a958f2651949fbfa182832c3cd08020d8facd94e4114f3e9", size = 2160488 },
+    { url = "https://files.pythonhosted.org/packages/ae/11/a742d3c31b2cc8557a48efdde53427fd5f9caa2fa3c9c27d826e78a66f51/ml_dtypes-0.4.1-cp312-cp312-win_amd64.whl", hash = "sha256:df0fb650d5c582a9e72bb5bd96cfebb2cdb889d89daff621c8fbc60295eba66c", size = 127462 },
+    { url = "https://files.pythonhosted.org/packages/8f/d7/6e1372052fe95c0cacfdb9718dba04726203885ffddb0cfddd8f8aa89a3b/ml_dtypes-0.4.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:e35e486e97aee577d0890bc3bd9e9f9eece50c08c163304008587ec8cfe7575b", size = 396578 },
+    { url = "https://files.pythonhosted.org/packages/1a/f6/ad0bd2735b9570ebf9c113f024b4f2b34f2331f16197c60babdc168b22d5/ml_dtypes-0.4.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:560be16dc1e3bdf7c087eb727e2cf9c0e6a3d87e9f415079d2491cc419b3ebf5", size = 2181057 },
+    { url = "https://files.pythonhosted.org/packages/3e/55/b9711de47135d4d8766ff7907fe54c8bffff545fd646817c352de37b0ad5/ml_dtypes-0.4.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ad0b757d445a20df39035c4cdeed457ec8b60d236020d2560dbc25887533cf50", size = 2156131 },
+    { url = "https://files.pythonhosted.org/packages/4b/f3/e5ff8dd27f66c8b80f97f0f89bb0b74e4a7005e5ff5f8f4237126c827911/ml_dtypes-0.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:ef0d7e3fece227b49b544fa69e50e607ac20948f0043e9f76b44f35f229ea450", size = 126727 },
+]
+
+[[package]]
+name = "ml-dtypes"
+version = "0.5.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version == '3.12.*'",
+]
+dependencies = [
+    { name = "numpy", marker = "python_full_version == '3.12.*'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/32/49/6e67c334872d2c114df3020e579f3718c333198f8312290e09ec0216703a/ml_dtypes-0.5.1.tar.gz", hash = "sha256:ac5b58559bb84a95848ed6984eb8013249f90b6bab62aa5acbad876e256002c9", size = 698772 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f4/88/11ebdbc75445eeb5b6869b708a0d787d1ed812ff86c2170bbfb95febdce1/ml_dtypes-0.5.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:bd73f51957949069573ff783563486339a9285d72e2f36c18e0c1aa9ca7eb190", size = 671450 },
+    { url = "https://files.pythonhosted.org/packages/a4/a4/9321cae435d6140f9b0e7af8334456a854b60e3a9c6101280a16e3594965/ml_dtypes-0.5.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:810512e2eccdfc3b41eefa3a27402371a3411453a1efc7e9c000318196140fed", size = 4621075 },
+    { url = "https://files.pythonhosted.org/packages/16/d8/4502e12c6a10d42e13a552e8d97f20198e3cf82a0d1411ad50be56a5077c/ml_dtypes-0.5.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:141b2ea2f20bb10802ddca55d91fe21231ef49715cfc971998e8f2a9838f3dbe", size = 4738414 },
+    { url = "https://files.pythonhosted.org/packages/6b/7e/bc54ae885e4d702e60a4bf50aa9066ff35e9c66b5213d11091f6bffb3036/ml_dtypes-0.5.1-cp310-cp310-win_amd64.whl", hash = "sha256:26ebcc69d7b779c8f129393e99732961b5cc33fcff84090451f448c89b0e01b4", size = 209718 },
+    { url = "https://files.pythonhosted.org/packages/c9/fd/691335926126bb9beeb030b61a28f462773dcf16b8e8a2253b599013a303/ml_dtypes-0.5.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:023ce2f502efd4d6c1e0472cc58ce3640d051d40e71e27386bed33901e201327", size = 671448 },
+    { url = "https://files.pythonhosted.org/packages/ff/a6/63832d91f2feb250d865d069ba1a5d0c686b1f308d1c74ce9764472c5e22/ml_dtypes-0.5.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7000b6e4d8ef07542c05044ec5d8bbae1df083b3f56822c3da63993a113e716f", size = 4625792 },
+    { url = "https://files.pythonhosted.org/packages/cc/2a/5421fd3dbe6eef9b844cc9d05f568b9fb568503a2e51cb1eb4443d9fc56b/ml_dtypes-0.5.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c09526488c3a9e8b7a23a388d4974b670a9a3dd40c5c8a61db5593ce9b725bab", size = 4743893 },
+    { url = "https://files.pythonhosted.org/packages/60/30/d3f0fc9499a22801219679a7f3f8d59f1429943c6261f445fb4bfce20718/ml_dtypes-0.5.1-cp311-cp311-win_amd64.whl", hash = "sha256:15ad0f3b0323ce96c24637a88a6f44f6713c64032f27277b069f285c3cf66478", size = 209712 },
+    { url = "https://files.pythonhosted.org/packages/47/56/1bb21218e1e692506c220ffabd456af9733fba7aa1b14f73899979f4cc20/ml_dtypes-0.5.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:6f462f5eca22fb66d7ff9c4744a3db4463af06c49816c4b6ac89b16bfcdc592e", size = 670372 },
+    { url = "https://files.pythonhosted.org/packages/20/95/d8bd96a3b60e00bf31bd78ca4bdd2d6bbaf5acb09b42844432d719d34061/ml_dtypes-0.5.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6f76232163b5b9c34291b54621ee60417601e2e4802a188a0ea7157cd9b323f4", size = 4635946 },
+    { url = "https://files.pythonhosted.org/packages/08/57/5d58fad4124192b1be42f68bd0c0ddaa26e44a730ff8c9337adade2f5632/ml_dtypes-0.5.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ad4953c5eb9c25a56d11a913c2011d7e580a435ef5145f804d98efa14477d390", size = 4694804 },
+    { url = "https://files.pythonhosted.org/packages/38/bc/c4260e4a6c6bf684d0313308de1c860467275221d5e7daf69b3fcddfdd0b/ml_dtypes-0.5.1-cp312-cp312-win_amd64.whl", hash = "sha256:9626d0bca1fb387d5791ca36bacbba298c5ef554747b7ebeafefb4564fc83566", size = 210853 },
+    { url = "https://files.pythonhosted.org/packages/0f/92/bb6a3d18e16fddd18ce6d5f480e1919b33338c70e18cba831c6ae59812ee/ml_dtypes-0.5.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:12651420130ee7cc13059fc56dac6ad300c3af3848b802d475148c9defd27c23", size = 667696 },
+    { url = "https://files.pythonhosted.org/packages/6d/29/cfc89d842767e9a51146043b0fa18332c2b38f8831447e6cb1160e3c6102/ml_dtypes-0.5.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c9945669d3dadf8acb40ec2e57d38c985d8c285ea73af57fc5b09872c516106d", size = 4638365 },
+    { url = "https://files.pythonhosted.org/packages/be/26/adc36e3ea09603d9f6d114894e1c1b7b8e8a9ef6d0b031cc270c6624a37c/ml_dtypes-0.5.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bf9975bda82a99dc935f2ae4c83846d86df8fd6ba179614acac8e686910851da", size = 4702722 },
+    { url = "https://files.pythonhosted.org/packages/da/8a/a2b9375c94077e5a488a624a195621407846f504068ce22ccf805c674156/ml_dtypes-0.5.1-cp313-cp313-win_amd64.whl", hash = "sha256:fd918d4e6a4e0c110e2e05be7a7814d10dc1b95872accbf6512b80a109b71ae1", size = 210850 },
+    { url = "https://files.pythonhosted.org/packages/52/38/703169100fdde27957f061d4d0ea3e00525775a09acaccf7e655d9609d55/ml_dtypes-0.5.1-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:05f23447a1c20ddf4dc7c2c661aa9ed93fcb2658f1017c204d1e758714dc28a8", size = 693043 },
+    { url = "https://files.pythonhosted.org/packages/28/ff/4e234c9c23e0d456f5da5a326c103bf890c746d93351524d987e41f438b3/ml_dtypes-0.5.1-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1b7fbe5571fdf28fd3aaab3ef4aafc847de9ebf263be959958c1ca58ec8eadf5", size = 4903946 },
+    { url = "https://files.pythonhosted.org/packages/b7/45/c1a1ccfdd02bc4173ca0f4a2d327683a27df85797b885eb1da1ca325b85c/ml_dtypes-0.5.1-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d13755f8e8445b3870114e5b6240facaa7cb0c3361e54beba3e07fa912a6e12b", size = 5052731 },
+    { url = "https://files.pythonhosted.org/packages/4c/17/b8f22639eb0cf52aee86c38c6e4cf1586e99445ef8e87bc1d30237bd5a1a/ml_dtypes-0.5.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:b8a9d46b4df5ae2135a8e8e72b465448ebbc1559997f4f9304a9ecc3413efb5b", size = 667609 },
+    { url = "https://files.pythonhosted.org/packages/95/87/4a1c91ea325ec4708293ae270b375fbdc53917031a64be9d5e3908ac0185/ml_dtypes-0.5.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:afb2009ac98da274e893e03162f6269398b2b00d947e7057ee2469a921d58135", size = 4615607 },
+    { url = "https://files.pythonhosted.org/packages/e7/6d/f53c438cbdf753164497958f4ff12a6844ae66a00edcabebcde1dc575ab7/ml_dtypes-0.5.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:aefedc579ece2f8fb38f876aa7698204ee4c372d0e54f1c1ffa8ca580b54cc60", size = 4729131 },
+    { url = "https://files.pythonhosted.org/packages/aa/10/03107a1ff3b03d6515d7365d9e356bc612b0750edbd5f4327a1db44404bc/ml_dtypes-0.5.1-cp39-cp39-win_amd64.whl", hash = "sha256:8f2c028954f16ede77902b223a8da2d9cbb3892375b85809a5c3cfb1587960c4", size = 209426 },
+]
+
+[[package]]
+name = "mlx"
+version = "0.22.0"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/41/3b/5160be3949dbed395c73646948f45d792a3de0dd6d8de213f26f2ea8d428/mlx-0.22.0-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:8082581e05b09d576cdf241184992fe0f96dcae1a857de38a562b2439d4e3333", size = 27110425 },
+    { url = "https://files.pythonhosted.org/packages/ec/e1/5d52b4144a946f2d66261df576b1e5f34845d781a2eba06d28ea0e554c51/mlx-0.22.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:b45716a948df0727e4e3525ca22e3f590323842020b7505e74491120a3a06d47", size = 26853548 },
+    { url = "https://files.pythonhosted.org/packages/b4/e7/2b4b551cff66370e0ce5d37dfb107122688d2b981eced38438e595ef98da/mlx-0.22.0-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:564ba1ca7711fc7ca5a1a17f43fc1859aef164229ee830050c66ef7453a9109d", size = 27110977 },
+    { url = "https://files.pythonhosted.org/packages/69/33/16e3a1451f6b9d9d4252d48453470d907c20f13a39915fd3e18d7a049570/mlx-0.22.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:03326dde450b9a8d75933514d3d4dd86163d2296bf70ff5d37befed3d6ba4144", size = 26854060 },
+    { url = "https://files.pythonhosted.org/packages/bc/33/dfab1ea36fc9b11e310922dceaeee210fc589efd277c409b0dc51bd38b79/mlx-0.22.0-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:56dec470a1a75010d1888a2e8f9cf24780980ff163eb6b03e9871bf440642a5b", size = 27107090 },
+    { url = "https://files.pythonhosted.org/packages/32/2b/756b04394e41dacd04ccb6fde0fede837590ba752229d2c27e6c442929ce/mlx-0.22.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:1dface82280483f14f2557da2afe8c965f7b09caa4bdae962cfa78897d0bed73", size = 26850177 },
+    { url = "https://files.pythonhosted.org/packages/62/2b/427896261bc8d940eff561e6199d1aee9dbdc7caa117486654a44d7d793c/mlx-0.22.0-cp313-cp313-macosx_13_0_arm64.whl", hash = "sha256:50d0d76826cfe939025791ce2c014e743ec7aff7aa67194ffaef40c40e574ef4", size = 27106970 },
+    { url = "https://files.pythonhosted.org/packages/6e/5a/314451ba1600a57619643bfc61c65e8ce34fed0ad011488ad3bc0fe3e739/mlx-0.22.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:bff0785c1be9ebc323d733c72f8a0683e757210221a9b8d53db3ac40ac7b75cf", size = 26850056 },
+    { url = "https://files.pythonhosted.org/packages/57/51/19deaecfe6b34c6ca30b4e966c8224145664747330143463f76b4c218a55/mlx-0.22.0-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:121d23087ea706b216c9f43f684c8068a6803d4aec1ddce1e3f6a66e99e797fa", size = 27110566 },
+    { url = "https://files.pythonhosted.org/packages/e4/88/56d767e752fea62e44704c9960fdf45a164b35866e6bc94f4a245122d8cc/mlx-0.22.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:d287de53d69af9129d8b59e13071c9c0711d729ddfc749f3187b0c894bf85d6b", size = 26853719 },
+]
+
+[[package]]
+name = "mlx-lm"
+version = "0.21.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "jinja2" },
+    { name = "mlx" },
+    { name = "numpy" },
+    { name = "protobuf" },
+    { name = "pyyaml" },
+    { name = "transformers", extra = ["sentencepiece"] },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/06/3a/125adb6e677795129ee932d189bac031253ebfcebaa1df559bdd4f0671b6/mlx_lm-0.21.1.tar.gz", hash = "sha256:701d4d1de92785cc24e651a28d46894df567187711cc994c88cb7247f3d1cde0", size = 106147 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9c/8f/41cd35d3ef6117a6831a02b27633b7a7dd8798c6d1a435a864a9db6ecb84/mlx_lm-0.21.1-py3-none-any.whl", hash = "sha256:fcc4a37c08cd3d6855f1fc5197cf4c8ebf790e1c778bce19f5c2c742096708c9", size = 140948 },
+]
+
+[[package]]
+name = "mpmath"
+version = "1.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f", size = 508106 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", size = 536198 },
+]
+
+[[package]]
+name = "msgpack"
+version = "1.1.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/cb/d0/7555686ae7ff5731205df1012ede15dd9d927f6227ea151e901c7406af4f/msgpack-1.1.0.tar.gz", hash = "sha256:dd432ccc2c72b914e4cb77afce64aab761c1137cc698be3984eee260bcb2896e", size = 167260 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/4b/f9/a892a6038c861fa849b11a2bb0502c07bc698ab6ea53359e5771397d883b/msgpack-1.1.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7ad442d527a7e358a469faf43fda45aaf4ac3249c8310a82f0ccff9164e5dccd", size = 150428 },
+    { url = "https://files.pythonhosted.org/packages/df/7a/d174cc6a3b6bb85556e6a046d3193294a92f9a8e583cdbd46dc8a1d7e7f4/msgpack-1.1.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:74bed8f63f8f14d75eec75cf3d04ad581da6b914001b474a5d3cd3372c8cc27d", size = 84131 },
+    { url = "https://files.pythonhosted.org/packages/08/52/bf4fbf72f897a23a56b822997a72c16de07d8d56d7bf273242f884055682/msgpack-1.1.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:914571a2a5b4e7606997e169f64ce53a8b1e06f2cf2c3a7273aa106236d43dd5", size = 81215 },
+    { url = "https://files.pythonhosted.org/packages/02/95/dc0044b439b518236aaf012da4677c1b8183ce388411ad1b1e63c32d8979/msgpack-1.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c921af52214dcbb75e6bdf6a661b23c3e6417f00c603dd2070bccb5c3ef499f5", size = 371229 },
+    { url = "https://files.pythonhosted.org/packages/ff/75/09081792db60470bef19d9c2be89f024d366b1e1973c197bb59e6aabc647/msgpack-1.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d8ce0b22b890be5d252de90d0e0d119f363012027cf256185fc3d474c44b1b9e", size = 378034 },
+    { url = "https://files.pythonhosted.org/packages/32/d3/c152e0c55fead87dd948d4b29879b0f14feeeec92ef1fd2ec21b107c3f49/msgpack-1.1.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:73322a6cc57fcee3c0c57c4463d828e9428275fb85a27aa2aa1a92fdc42afd7b", size = 363070 },
+    { url = "https://files.pythonhosted.org/packages/d9/2c/82e73506dd55f9e43ac8aa007c9dd088c6f0de2aa19e8f7330e6a65879fc/msgpack-1.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:e1f3c3d21f7cf67bcf2da8e494d30a75e4cf60041d98b3f79875afb5b96f3a3f", size = 359863 },
+    { url = "https://files.pythonhosted.org/packages/cb/a0/3d093b248837094220e1edc9ec4337de3443b1cfeeb6e0896af8ccc4cc7a/msgpack-1.1.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:64fc9068d701233effd61b19efb1485587560b66fe57b3e50d29c5d78e7fef68", size = 368166 },
+    { url = "https://files.pythonhosted.org/packages/e4/13/7646f14f06838b406cf5a6ddbb7e8dc78b4996d891ab3b93c33d1ccc8678/msgpack-1.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:42f754515e0f683f9c79210a5d1cad631ec3d06cea5172214d2176a42e67e19b", size = 370105 },
+    { url = "https://files.pythonhosted.org/packages/67/fa/dbbd2443e4578e165192dabbc6a22c0812cda2649261b1264ff515f19f15/msgpack-1.1.0-cp310-cp310-win32.whl", hash = "sha256:3df7e6b05571b3814361e8464f9304c42d2196808e0119f55d0d3e62cd5ea044", size = 68513 },
+    { url = "https://files.pythonhosted.org/packages/24/ce/c2c8fbf0ded750cb63cbcbb61bc1f2dfd69e16dca30a8af8ba80ec182dcd/msgpack-1.1.0-cp310-cp310-win_amd64.whl", hash = "sha256:685ec345eefc757a7c8af44a3032734a739f8c45d1b0ac45efc5d8977aa4720f", size = 74687 },
+    { url = "https://files.pythonhosted.org/packages/b7/5e/a4c7154ba65d93be91f2f1e55f90e76c5f91ccadc7efc4341e6f04c8647f/msgpack-1.1.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:3d364a55082fb2a7416f6c63ae383fbd903adb5a6cf78c5b96cc6316dc1cedc7", size = 150803 },
+    { url = "https://files.pythonhosted.org/packages/60/c2/687684164698f1d51c41778c838d854965dd284a4b9d3a44beba9265c931/msgpack-1.1.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:79ec007767b9b56860e0372085f8504db5d06bd6a327a335449508bbee9648fa", size = 84343 },
+    { url = "https://files.pythonhosted.org/packages/42/ae/d3adea9bb4a1342763556078b5765e666f8fdf242e00f3f6657380920972/msgpack-1.1.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6ad622bf7756d5a497d5b6836e7fc3752e2dd6f4c648e24b1803f6048596f701", size = 81408 },
+    { url = "https://files.pythonhosted.org/packages/dc/17/6313325a6ff40ce9c3207293aee3ba50104aed6c2c1559d20d09e5c1ff54/msgpack-1.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e59bca908d9ca0de3dc8684f21ebf9a690fe47b6be93236eb40b99af28b6ea6", size = 396096 },
+    { url = "https://files.pythonhosted.org/packages/a8/a1/ad7b84b91ab5a324e707f4c9761633e357820b011a01e34ce658c1dda7cc/msgpack-1.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5e1da8f11a3dd397f0a32c76165cf0c4eb95b31013a94f6ecc0b280c05c91b59", size = 403671 },
+    { url = "https://files.pythonhosted.org/packages/bb/0b/fd5b7c0b308bbf1831df0ca04ec76fe2f5bf6319833646b0a4bd5e9dc76d/msgpack-1.1.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:452aff037287acb1d70a804ffd022b21fa2bb7c46bee884dbc864cc9024128a0", size = 387414 },
+    { url = "https://files.pythonhosted.org/packages/f0/03/ff8233b7c6e9929a1f5da3c7860eccd847e2523ca2de0d8ef4878d354cfa/msgpack-1.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8da4bf6d54ceed70e8861f833f83ce0814a2b72102e890cbdfe4b34764cdd66e", size = 383759 },
+    { url = "https://files.pythonhosted.org/packages/1f/1b/eb82e1fed5a16dddd9bc75f0854b6e2fe86c0259c4353666d7fab37d39f4/msgpack-1.1.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:41c991beebf175faf352fb940bf2af9ad1fb77fd25f38d9142053914947cdbf6", size = 394405 },
+    { url = "https://files.pythonhosted.org/packages/90/2e/962c6004e373d54ecf33d695fb1402f99b51832631e37c49273cc564ffc5/msgpack-1.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a52a1f3a5af7ba1c9ace055b659189f6c669cf3657095b50f9602af3a3ba0fe5", size = 396041 },
+    { url = "https://files.pythonhosted.org/packages/f8/20/6e03342f629474414860c48aeffcc2f7f50ddaf351d95f20c3f1c67399a8/msgpack-1.1.0-cp311-cp311-win32.whl", hash = "sha256:58638690ebd0a06427c5fe1a227bb6b8b9fdc2bd07701bec13c2335c82131a88", size = 68538 },
+    { url = "https://files.pythonhosted.org/packages/aa/c4/5a582fc9a87991a3e6f6800e9bb2f3c82972912235eb9539954f3e9997c7/msgpack-1.1.0-cp311-cp311-win_amd64.whl", hash = "sha256:fd2906780f25c8ed5d7b323379f6138524ba793428db5d0e9d226d3fa6aa1788", size = 74871 },
+    { url = "https://files.pythonhosted.org/packages/e1/d6/716b7ca1dbde63290d2973d22bbef1b5032ca634c3ff4384a958ec3f093a/msgpack-1.1.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:d46cf9e3705ea9485687aa4001a76e44748b609d260af21c4ceea7f2212a501d", size = 152421 },
+    { url = "https://files.pythonhosted.org/packages/70/da/5312b067f6773429cec2f8f08b021c06af416bba340c912c2ec778539ed6/msgpack-1.1.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:5dbad74103df937e1325cc4bfeaf57713be0b4f15e1c2da43ccdd836393e2ea2", size = 85277 },
+    { url = "https://files.pythonhosted.org/packages/28/51/da7f3ae4462e8bb98af0d5bdf2707f1b8c65a0d4f496e46b6afb06cbc286/msgpack-1.1.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:58dfc47f8b102da61e8949708b3eafc3504509a5728f8b4ddef84bd9e16ad420", size = 82222 },
+    { url = "https://files.pythonhosted.org/packages/33/af/dc95c4b2a49cff17ce47611ca9ba218198806cad7796c0b01d1e332c86bb/msgpack-1.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4676e5be1b472909b2ee6356ff425ebedf5142427842aa06b4dfd5117d1ca8a2", size = 392971 },
+    { url = "https://files.pythonhosted.org/packages/f1/54/65af8de681fa8255402c80eda2a501ba467921d5a7a028c9c22a2c2eedb5/msgpack-1.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:17fb65dd0bec285907f68b15734a993ad3fc94332b5bb21b0435846228de1f39", size = 401403 },
+    { url = "https://files.pythonhosted.org/packages/97/8c/e333690777bd33919ab7024269dc3c41c76ef5137b211d776fbb404bfead/msgpack-1.1.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a51abd48c6d8ac89e0cfd4fe177c61481aca2d5e7ba42044fd218cfd8ea9899f", size = 385356 },
+    { url = "https://files.pythonhosted.org/packages/57/52/406795ba478dc1c890559dd4e89280fa86506608a28ccf3a72fbf45df9f5/msgpack-1.1.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2137773500afa5494a61b1208619e3871f75f27b03bcfca7b3a7023284140247", size = 383028 },
+    { url = "https://files.pythonhosted.org/packages/e7/69/053b6549bf90a3acadcd8232eae03e2fefc87f066a5b9fbb37e2e608859f/msgpack-1.1.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:398b713459fea610861c8a7b62a6fec1882759f308ae0795b5413ff6a160cf3c", size = 391100 },
+    { url = "https://files.pythonhosted.org/packages/23/f0/d4101d4da054f04274995ddc4086c2715d9b93111eb9ed49686c0f7ccc8a/msgpack-1.1.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:06f5fd2f6bb2a7914922d935d3b8bb4a7fff3a9a91cfce6d06c13bc42bec975b", size = 394254 },
+    { url = "https://files.pythonhosted.org/packages/1c/12/cf07458f35d0d775ff3a2dc5559fa2e1fcd06c46f1ef510e594ebefdca01/msgpack-1.1.0-cp312-cp312-win32.whl", hash = "sha256:ad33e8400e4ec17ba782f7b9cf868977d867ed784a1f5f2ab46e7ba53b6e1e1b", size = 69085 },
+    { url = "https://files.pythonhosted.org/packages/73/80/2708a4641f7d553a63bc934a3eb7214806b5b39d200133ca7f7afb0a53e8/msgpack-1.1.0-cp312-cp312-win_amd64.whl", hash = "sha256:115a7af8ee9e8cddc10f87636767857e7e3717b7a2e97379dc2054712693e90f", size = 75347 },
+    { url = "https://files.pythonhosted.org/packages/c8/b0/380f5f639543a4ac413e969109978feb1f3c66e931068f91ab6ab0f8be00/msgpack-1.1.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:071603e2f0771c45ad9bc65719291c568d4edf120b44eb36324dcb02a13bfddf", size = 151142 },
+    { url = "https://files.pythonhosted.org/packages/c8/ee/be57e9702400a6cb2606883d55b05784fada898dfc7fd12608ab1fdb054e/msgpack-1.1.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:0f92a83b84e7c0749e3f12821949d79485971f087604178026085f60ce109330", size = 84523 },
+    { url = "https://files.pythonhosted.org/packages/7e/3a/2919f63acca3c119565449681ad08a2f84b2171ddfcff1dba6959db2cceb/msgpack-1.1.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:4a1964df7b81285d00a84da4e70cb1383f2e665e0f1f2a7027e683956d04b734", size = 81556 },
+    { url = "https://files.pythonhosted.org/packages/7c/43/a11113d9e5c1498c145a8925768ea2d5fce7cbab15c99cda655aa09947ed/msgpack-1.1.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:59caf6a4ed0d164055ccff8fe31eddc0ebc07cf7326a2aaa0dbf7a4001cd823e", size = 392105 },
+    { url = "https://files.pythonhosted.org/packages/2d/7b/2c1d74ca6c94f70a1add74a8393a0138172207dc5de6fc6269483519d048/msgpack-1.1.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0907e1a7119b337971a689153665764adc34e89175f9a34793307d9def08e6ca", size = 399979 },
+    { url = "https://files.pythonhosted.org/packages/82/8c/cf64ae518c7b8efc763ca1f1348a96f0e37150061e777a8ea5430b413a74/msgpack-1.1.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:65553c9b6da8166e819a6aa90ad15288599b340f91d18f60b2061f402b9a4915", size = 383816 },
+    { url = "https://files.pythonhosted.org/packages/69/86/a847ef7a0f5ef3fa94ae20f52a4cacf596a4e4a010197fbcc27744eb9a83/msgpack-1.1.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:7a946a8992941fea80ed4beae6bff74ffd7ee129a90b4dd5cf9c476a30e9708d", size = 380973 },
+    { url = "https://files.pythonhosted.org/packages/aa/90/c74cf6e1126faa93185d3b830ee97246ecc4fe12cf9d2d31318ee4246994/msgpack-1.1.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:4b51405e36e075193bc051315dbf29168d6141ae2500ba8cd80a522964e31434", size = 387435 },
+    { url = "https://files.pythonhosted.org/packages/7a/40/631c238f1f338eb09f4acb0f34ab5862c4e9d7eda11c1b685471a4c5ea37/msgpack-1.1.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:b4c01941fd2ff87c2a934ee6055bda4ed353a7846b8d4f341c428109e9fcde8c", size = 399082 },
+    { url = "https://files.pythonhosted.org/packages/e9/1b/fa8a952be252a1555ed39f97c06778e3aeb9123aa4cccc0fd2acd0b4e315/msgpack-1.1.0-cp313-cp313-win32.whl", hash = "sha256:7c9a35ce2c2573bada929e0b7b3576de647b0defbd25f5139dcdaba0ae35a4cc", size = 69037 },
+    { url = "https://files.pythonhosted.org/packages/b6/bc/8bd826dd03e022153bfa1766dcdec4976d6c818865ed54223d71f07862b3/msgpack-1.1.0-cp313-cp313-win_amd64.whl", hash = "sha256:bce7d9e614a04d0883af0b3d4d501171fbfca038f12c77fa838d9f198147a23f", size = 75140 },
+    { url = "https://files.pythonhosted.org/packages/f7/3b/544a5c5886042b80e1f4847a4757af3430f60d106d8d43bb7be72c9e9650/msgpack-1.1.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:53258eeb7a80fc46f62fd59c876957a2d0e15e6449a9e71842b6d24419d88ca1", size = 150713 },
+    { url = "https://files.pythonhosted.org/packages/93/af/d63f25bcccd3d6f06fd518ba4a321f34a4370c67b579ca5c70b4a37721b4/msgpack-1.1.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7e7b853bbc44fb03fbdba34feb4bd414322180135e2cb5164f20ce1c9795ee48", size = 84277 },
+    { url = "https://files.pythonhosted.org/packages/92/9b/5c0dfb0009b9f96328664fecb9f8e4e9c8a1ae919e6d53986c1b813cb493/msgpack-1.1.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:f3e9b4936df53b970513eac1758f3882c88658a220b58dcc1e39606dccaaf01c", size = 81357 },
+    { url = "https://files.pythonhosted.org/packages/d1/7c/3a9ee6ec9fc3e47681ad39b4d344ee04ff20a776b594fba92d88d8b68356/msgpack-1.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46c34e99110762a76e3911fc923222472c9d681f1094096ac4102c18319e6468", size = 371256 },
+    { url = "https://files.pythonhosted.org/packages/f7/0a/8a213cecea7b731c540f25212ba5f9a818f358237ac51a44d448bd753690/msgpack-1.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a706d1e74dd3dea05cb54580d9bd8b2880e9264856ce5068027eed09680aa74", size = 377868 },
+    { url = "https://files.pythonhosted.org/packages/1b/94/a82b0db0981e9586ed5af77d6cfb343da05d7437dceaae3b35d346498110/msgpack-1.1.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:534480ee5690ab3cbed89d4c8971a5c631b69a8c0883ecfea96c19118510c846", size = 363370 },
+    { url = "https://files.pythonhosted.org/packages/93/fc/6c7f0dcc1c913e14861e16eaf494c07fc1dde454ec726ff8cebcf348ae53/msgpack-1.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:8cf9e8c3a2153934a23ac160cc4cba0ec035f6867c8013cc6077a79823370346", size = 358970 },
+    { url = "https://files.pythonhosted.org/packages/1f/c6/e4a04c0089deace870dabcdef5c9f12798f958e2e81d5012501edaff342f/msgpack-1.1.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:3180065ec2abbe13a4ad37688b61b99d7f9e012a535b930e0e683ad6bc30155b", size = 366358 },
+    { url = "https://files.pythonhosted.org/packages/b6/54/7d8317dac590cf16b3e08e3fb74d2081e5af44eb396f0effa13f17777f30/msgpack-1.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c5a91481a3cc573ac8c0d9aace09345d989dc4a0202b7fcb312c88c26d4e71a8", size = 370336 },
+    { url = "https://files.pythonhosted.org/packages/dc/6f/a5a1f43b6566831e9630e5bc5d86034a8884386297302be128402555dde1/msgpack-1.1.0-cp39-cp39-win32.whl", hash = "sha256:f80bc7d47f76089633763f952e67f8214cb7b3ee6bfa489b3cb6a84cfac114cd", size = 68683 },
+    { url = "https://files.pythonhosted.org/packages/5f/e8/2162621e18dbc36e2bc8492fd0e97b3975f5d89fe0472ae6d5f7fbdd8cf7/msgpack-1.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:4d1b7ff2d6146e16e8bd665ac726a89c74163ef8cd39fa8c1087d4e52d3a2325", size = 74787 },
+]
+
+[[package]]
+name = "multidict"
+version = "6.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/d6/be/504b89a5e9ca731cd47487e91c469064f8ae5af93b7259758dcfc2b9c848/multidict-6.1.0.tar.gz", hash = "sha256:22ae2ebf9b0c69d206c003e2f6a914ea33f0a932d4aa16f236afc049d9958f4a", size = 64002 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/29/68/259dee7fd14cf56a17c554125e534f6274c2860159692a414d0b402b9a6d/multidict-6.1.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:3380252550e372e8511d49481bd836264c009adb826b23fefcc5dd3c69692f60", size = 48628 },
+    { url = "https://files.pythonhosted.org/packages/50/79/53ba256069fe5386a4a9e80d4e12857ced9de295baf3e20c68cdda746e04/multidict-6.1.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:99f826cbf970077383d7de805c0681799491cb939c25450b9b5b3ced03ca99f1", size = 29327 },
+    { url = "https://files.pythonhosted.org/packages/ff/10/71f1379b05b196dae749b5ac062e87273e3f11634f447ebac12a571d90ae/multidict-6.1.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a114d03b938376557927ab23f1e950827c3b893ccb94b62fd95d430fd0e5cf53", size = 29689 },
+    { url = "https://files.pythonhosted.org/packages/71/45/70bac4f87438ded36ad4793793c0095de6572d433d98575a5752629ef549/multidict-6.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b1c416351ee6271b2f49b56ad7f308072f6f44b37118d69c2cad94f3fa8a40d5", size = 126639 },
+    { url = "https://files.pythonhosted.org/packages/80/cf/17f35b3b9509b4959303c05379c4bfb0d7dd05c3306039fc79cf035bbac0/multidict-6.1.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6b5d83030255983181005e6cfbac1617ce9746b219bc2aad52201ad121226581", size = 134315 },
+    { url = "https://files.pythonhosted.org/packages/ef/1f/652d70ab5effb33c031510a3503d4d6efc5ec93153562f1ee0acdc895a57/multidict-6.1.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3e97b5e938051226dc025ec80980c285b053ffb1e25a3db2a3aa3bc046bf7f56", size = 129471 },
+    { url = "https://files.pythonhosted.org/packages/a6/64/2dd6c4c681688c0165dea3975a6a4eab4944ea30f35000f8b8af1df3148c/multidict-6.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d618649d4e70ac6efcbba75be98b26ef5078faad23592f9b51ca492953012429", size = 124585 },
+    { url = "https://files.pythonhosted.org/packages/87/56/e6ee5459894c7e554b57ba88f7257dc3c3d2d379cb15baaa1e265b8c6165/multidict-6.1.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:10524ebd769727ac77ef2278390fb0068d83f3acb7773792a5080f2b0abf7748", size = 116957 },
+    { url = "https://files.pythonhosted.org/packages/36/9e/616ce5e8d375c24b84f14fc263c7ef1d8d5e8ef529dbc0f1df8ce71bb5b8/multidict-6.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ff3827aef427c89a25cc96ded1759271a93603aba9fb977a6d264648ebf989db", size = 128609 },
+    { url = "https://files.pythonhosted.org/packages/8c/4f/4783e48a38495d000f2124020dc96bacc806a4340345211b1ab6175a6cb4/multidict-6.1.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:06809f4f0f7ab7ea2cabf9caca7d79c22c0758b58a71f9d32943ae13c7ace056", size = 123016 },
+    { url = "https://files.pythonhosted.org/packages/3e/b3/4950551ab8fc39862ba5e9907dc821f896aa829b4524b4deefd3e12945ab/multidict-6.1.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:f179dee3b863ab1c59580ff60f9d99f632f34ccb38bf67a33ec6b3ecadd0fd76", size = 133542 },
+    { url = "https://files.pythonhosted.org/packages/96/4d/f0ce6ac9914168a2a71df117935bb1f1781916acdecbb43285e225b484b8/multidict-6.1.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:aaed8b0562be4a0876ee3b6946f6869b7bcdb571a5d1496683505944e268b160", size = 130163 },
+    { url = "https://files.pythonhosted.org/packages/be/72/17c9f67e7542a49dd252c5ae50248607dfb780bcc03035907dafefb067e3/multidict-6.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:3c8b88a2ccf5493b6c8da9076fb151ba106960a2df90c2633f342f120751a9e7", size = 126832 },
+    { url = "https://files.pythonhosted.org/packages/71/9f/72d719e248cbd755c8736c6d14780533a1606ffb3fbb0fbd77da9f0372da/multidict-6.1.0-cp310-cp310-win32.whl", hash = "sha256:4a9cb68166a34117d6646c0023c7b759bf197bee5ad4272f420a0141d7eb03a0", size = 26402 },
+    { url = "https://files.pythonhosted.org/packages/04/5a/d88cd5d00a184e1ddffc82aa2e6e915164a6d2641ed3606e766b5d2f275a/multidict-6.1.0-cp310-cp310-win_amd64.whl", hash = "sha256:20b9b5fbe0b88d0bdef2012ef7dee867f874b72528cf1d08f1d59b0e3850129d", size = 28800 },
+    { url = "https://files.pythonhosted.org/packages/93/13/df3505a46d0cd08428e4c8169a196131d1b0c4b515c3649829258843dde6/multidict-6.1.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:3efe2c2cb5763f2f1b275ad2bf7a287d3f7ebbef35648a9726e3b69284a4f3d6", size = 48570 },
+    { url = "https://files.pythonhosted.org/packages/f0/e1/a215908bfae1343cdb72f805366592bdd60487b4232d039c437fe8f5013d/multidict-6.1.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c7053d3b0353a8b9de430a4f4b4268ac9a4fb3481af37dfe49825bf45ca24156", size = 29316 },
+    { url = "https://files.pythonhosted.org/packages/70/0f/6dc70ddf5d442702ed74f298d69977f904960b82368532c88e854b79f72b/multidict-6.1.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:27e5fc84ccef8dfaabb09d82b7d179c7cf1a3fbc8a966f8274fcb4ab2eb4cadb", size = 29640 },
+    { url = "https://files.pythonhosted.org/packages/d8/6d/9c87b73a13d1cdea30b321ef4b3824449866bd7f7127eceed066ccb9b9ff/multidict-6.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0e2b90b43e696f25c62656389d32236e049568b39320e2735d51f08fd362761b", size = 131067 },
+    { url = "https://files.pythonhosted.org/packages/cc/1e/1b34154fef373371fd6c65125b3d42ff5f56c7ccc6bfff91b9b3c60ae9e0/multidict-6.1.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d83a047959d38a7ff552ff94be767b7fd79b831ad1cd9920662db05fec24fe72", size = 138507 },
+    { url = "https://files.pythonhosted.org/packages/fb/e0/0bc6b2bac6e461822b5f575eae85da6aae76d0e2a79b6665d6206b8e2e48/multidict-6.1.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:d1a9dd711d0877a1ece3d2e4fea11a8e75741ca21954c919406b44e7cf971304", size = 133905 },
+    { url = "https://files.pythonhosted.org/packages/ba/af/73d13b918071ff9b2205fcf773d316e0f8fefb4ec65354bbcf0b10908cc6/multidict-6.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ec2abea24d98246b94913b76a125e855eb5c434f7c46546046372fe60f666351", size = 129004 },
+    { url = "https://files.pythonhosted.org/packages/74/21/23960627b00ed39643302d81bcda44c9444ebcdc04ee5bedd0757513f259/multidict-6.1.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4867cafcbc6585e4b678876c489b9273b13e9fff9f6d6d66add5e15d11d926cb", size = 121308 },
+    { url = "https://files.pythonhosted.org/packages/8b/5c/cf282263ffce4a596ed0bb2aa1a1dddfe1996d6a62d08842a8d4b33dca13/multidict-6.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:5b48204e8d955c47c55b72779802b219a39acc3ee3d0116d5080c388970b76e3", size = 132608 },
+    { url = "https://files.pythonhosted.org/packages/d7/3e/97e778c041c72063f42b290888daff008d3ab1427f5b09b714f5a8eff294/multidict-6.1.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:d8fff389528cad1618fb4b26b95550327495462cd745d879a8c7c2115248e399", size = 127029 },
+    { url = "https://files.pythonhosted.org/packages/47/ac/3efb7bfe2f3aefcf8d103e9a7162572f01936155ab2f7ebcc7c255a23212/multidict-6.1.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a7a9541cd308eed5e30318430a9c74d2132e9a8cb46b901326272d780bf2d423", size = 137594 },
+    { url = "https://files.pythonhosted.org/packages/42/9b/6c6e9e8dc4f915fc90a9b7798c44a30773dea2995fdcb619870e705afe2b/multidict-6.1.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:da1758c76f50c39a2efd5e9859ce7d776317eb1dd34317c8152ac9251fc574a3", size = 134556 },
+    { url = "https://files.pythonhosted.org/packages/1d/10/8e881743b26aaf718379a14ac58572a240e8293a1c9d68e1418fb11c0f90/multidict-6.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:c943a53e9186688b45b323602298ab727d8865d8c9ee0b17f8d62d14b56f0753", size = 130993 },
+    { url = "https://files.pythonhosted.org/packages/45/84/3eb91b4b557442802d058a7579e864b329968c8d0ea57d907e7023c677f2/multidict-6.1.0-cp311-cp311-win32.whl", hash = "sha256:90f8717cb649eea3504091e640a1b8568faad18bd4b9fcd692853a04475a4b80", size = 26405 },
+    { url = "https://files.pythonhosted.org/packages/9f/0b/ad879847ecbf6d27e90a6eabb7eff6b62c129eefe617ea45eae7c1f0aead/multidict-6.1.0-cp311-cp311-win_amd64.whl", hash = "sha256:82176036e65644a6cc5bd619f65f6f19781e8ec2e5330f51aa9ada7504cc1926", size = 28795 },
+    { url = "https://files.pythonhosted.org/packages/fd/16/92057c74ba3b96d5e211b553895cd6dc7cc4d1e43d9ab8fafc727681ef71/multidict-6.1.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:b04772ed465fa3cc947db808fa306d79b43e896beb677a56fb2347ca1a49c1fa", size = 48713 },
+    { url = "https://files.pythonhosted.org/packages/94/3d/37d1b8893ae79716179540b89fc6a0ee56b4a65fcc0d63535c6f5d96f217/multidict-6.1.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:6180c0ae073bddeb5a97a38c03f30c233e0a4d39cd86166251617d1bbd0af436", size = 29516 },
+    { url = "https://files.pythonhosted.org/packages/a2/12/adb6b3200c363062f805275b4c1e656be2b3681aada66c80129932ff0bae/multidict-6.1.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:071120490b47aa997cca00666923a83f02c7fbb44f71cf7f136df753f7fa8761", size = 29557 },
+    { url = "https://files.pythonhosted.org/packages/47/e9/604bb05e6e5bce1e6a5cf80a474e0f072e80d8ac105f1b994a53e0b28c42/multidict-6.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:50b3a2710631848991d0bf7de077502e8994c804bb805aeb2925a981de58ec2e", size = 130170 },
+    { url = "https://files.pythonhosted.org/packages/7e/13/9efa50801785eccbf7086b3c83b71a4fb501a4d43549c2f2f80b8787d69f/multidict-6.1.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b58c621844d55e71c1b7f7c498ce5aa6985d743a1a59034c57a905b3f153c1ef", size = 134836 },
+    { url = "https://files.pythonhosted.org/packages/bf/0f/93808b765192780d117814a6dfcc2e75de6dcc610009ad408b8814dca3ba/multidict-6.1.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:55b6d90641869892caa9ca42ff913f7ff1c5ece06474fbd32fb2cf6834726c95", size = 133475 },
+    { url = "https://files.pythonhosted.org/packages/d3/c8/529101d7176fe7dfe1d99604e48d69c5dfdcadb4f06561f465c8ef12b4df/multidict-6.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b820514bfc0b98a30e3d85462084779900347e4d49267f747ff54060cc33925", size = 131049 },
+    { url = "https://files.pythonhosted.org/packages/ca/0c/fc85b439014d5a58063e19c3a158a889deec399d47b5269a0f3b6a2e28bc/multidict-6.1.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:10a9b09aba0c5b48c53761b7c720aaaf7cf236d5fe394cd399c7ba662d5f9966", size = 120370 },
+    { url = "https://files.pythonhosted.org/packages/db/46/d4416eb20176492d2258fbd47b4abe729ff3b6e9c829ea4236f93c865089/multidict-6.1.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:1e16bf3e5fc9f44632affb159d30a437bfe286ce9e02754759be5536b169b305", size = 125178 },
+    { url = "https://files.pythonhosted.org/packages/5b/46/73697ad7ec521df7de5531a32780bbfd908ded0643cbe457f981a701457c/multidict-6.1.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:76f364861c3bfc98cbbcbd402d83454ed9e01a5224bb3a28bf70002a230f73e2", size = 119567 },
+    { url = "https://files.pythonhosted.org/packages/cd/ed/51f060e2cb0e7635329fa6ff930aa5cffa17f4c7f5c6c3ddc3500708e2f2/multidict-6.1.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:820c661588bd01a0aa62a1283f20d2be4281b086f80dad9e955e690c75fb54a2", size = 129822 },
+    { url = "https://files.pythonhosted.org/packages/df/9e/ee7d1954b1331da3eddea0c4e08d9142da5f14b1321c7301f5014f49d492/multidict-6.1.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:0e5f362e895bc5b9e67fe6e4ded2492d8124bdf817827f33c5b46c2fe3ffaca6", size = 128656 },
+    { url = "https://files.pythonhosted.org/packages/77/00/8538f11e3356b5d95fa4b024aa566cde7a38aa7a5f08f4912b32a037c5dc/multidict-6.1.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:3ec660d19bbc671e3a6443325f07263be452c453ac9e512f5eb935e7d4ac28b3", size = 125360 },
+    { url = "https://files.pythonhosted.org/packages/be/05/5d334c1f2462d43fec2363cd00b1c44c93a78c3925d952e9a71caf662e96/multidict-6.1.0-cp312-cp312-win32.whl", hash = "sha256:58130ecf8f7b8112cdb841486404f1282b9c86ccb30d3519faf301b2e5659133", size = 26382 },
+    { url = "https://files.pythonhosted.org/packages/a3/bf/f332a13486b1ed0496d624bcc7e8357bb8053823e8cd4b9a18edc1d97e73/multidict-6.1.0-cp312-cp312-win_amd64.whl", hash = "sha256:188215fc0aafb8e03341995e7c4797860181562380f81ed0a87ff455b70bf1f1", size = 28529 },
+    { url = "https://files.pythonhosted.org/packages/22/67/1c7c0f39fe069aa4e5d794f323be24bf4d33d62d2a348acdb7991f8f30db/multidict-6.1.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d569388c381b24671589335a3be6e1d45546c2988c2ebe30fdcada8457a31008", size = 48771 },
+    { url = "https://files.pythonhosted.org/packages/3c/25/c186ee7b212bdf0df2519eacfb1981a017bda34392c67542c274651daf23/multidict-6.1.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:052e10d2d37810b99cc170b785945421141bf7bb7d2f8799d431e7db229c385f", size = 29533 },
+    { url = "https://files.pythonhosted.org/packages/67/5e/04575fd837e0958e324ca035b339cea174554f6f641d3fb2b4f2e7ff44a2/multidict-6.1.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f90c822a402cb865e396a504f9fc8173ef34212a342d92e362ca498cad308e28", size = 29595 },
+    { url = "https://files.pythonhosted.org/packages/d3/b2/e56388f86663810c07cfe4a3c3d87227f3811eeb2d08450b9e5d19d78876/multidict-6.1.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b225d95519a5bf73860323e633a664b0d85ad3d5bede6d30d95b35d4dfe8805b", size = 130094 },
+    { url = "https://files.pythonhosted.org/packages/6c/ee/30ae9b4186a644d284543d55d491fbd4239b015d36b23fea43b4c94f7052/multidict-6.1.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:23bfd518810af7de1116313ebd9092cb9aa629beb12f6ed631ad53356ed6b86c", size = 134876 },
+    { url = "https://files.pythonhosted.org/packages/84/c7/70461c13ba8ce3c779503c70ec9d0345ae84de04521c1f45a04d5f48943d/multidict-6.1.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5c09fcfdccdd0b57867577b719c69e347a436b86cd83747f179dbf0cc0d4c1f3", size = 133500 },
+    { url = "https://files.pythonhosted.org/packages/4a/9f/002af221253f10f99959561123fae676148dd730e2daa2cd053846a58507/multidict-6.1.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bf6bea52ec97e95560af5ae576bdac3aa3aae0b6758c6efa115236d9e07dae44", size = 131099 },
+    { url = "https://files.pythonhosted.org/packages/82/42/d1c7a7301d52af79d88548a97e297f9d99c961ad76bbe6f67442bb77f097/multidict-6.1.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:57feec87371dbb3520da6192213c7d6fc892d5589a93db548331954de8248fd2", size = 120403 },
+    { url = "https://files.pythonhosted.org/packages/68/f3/471985c2c7ac707547553e8f37cff5158030d36bdec4414cb825fbaa5327/multidict-6.1.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:0c3f390dc53279cbc8ba976e5f8035eab997829066756d811616b652b00a23a3", size = 125348 },
+    { url = "https://files.pythonhosted.org/packages/67/2c/e6df05c77e0e433c214ec1d21ddd203d9a4770a1f2866a8ca40a545869a0/multidict-6.1.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:59bfeae4b25ec05b34f1956eaa1cb38032282cd4dfabc5056d0a1ec4d696d3aa", size = 119673 },
+    { url = "https://files.pythonhosted.org/packages/c5/cd/bc8608fff06239c9fb333f9db7743a1b2eafe98c2666c9a196e867a3a0a4/multidict-6.1.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:b2f59caeaf7632cc633b5cf6fc449372b83bbdf0da4ae04d5be36118e46cc0aa", size = 129927 },
+    { url = "https://files.pythonhosted.org/packages/44/8e/281b69b7bc84fc963a44dc6e0bbcc7150e517b91df368a27834299a526ac/multidict-6.1.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:37bb93b2178e02b7b618893990941900fd25b6b9ac0fa49931a40aecdf083fe4", size = 128711 },
+    { url = "https://files.pythonhosted.org/packages/12/a4/63e7cd38ed29dd9f1881d5119f272c898ca92536cdb53ffe0843197f6c85/multidict-6.1.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:4e9f48f58c2c523d5a06faea47866cd35b32655c46b443f163d08c6d0ddb17d6", size = 125519 },
+    { url = "https://files.pythonhosted.org/packages/38/e0/4f5855037a72cd8a7a2f60a3952d9aa45feedb37ae7831642102604e8a37/multidict-6.1.0-cp313-cp313-win32.whl", hash = "sha256:3a37ffb35399029b45c6cc33640a92bef403c9fd388acce75cdc88f58bd19a81", size = 26426 },
+    { url = "https://files.pythonhosted.org/packages/7e/a5/17ee3a4db1e310b7405f5d25834460073a8ccd86198ce044dfaf69eac073/multidict-6.1.0-cp313-cp313-win_amd64.whl", hash = "sha256:e9aa71e15d9d9beaad2c6b9319edcdc0a49a43ef5c0a4c8265ca9ee7d6c67774", size = 28531 },
+    { url = "https://files.pythonhosted.org/packages/e7/c9/9e153a6572b38ac5ff4434113af38acf8d5e9957897cdb1f513b3d6614ed/multidict-6.1.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:4e18b656c5e844539d506a0a06432274d7bd52a7487e6828c63a63d69185626c", size = 48550 },
+    { url = "https://files.pythonhosted.org/packages/76/f5/79565ddb629eba6c7f704f09a09df085c8dc04643b12506f10f718cee37a/multidict-6.1.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a185f876e69897a6f3325c3f19f26a297fa058c5e456bfcff8015e9a27e83ae1", size = 29298 },
+    { url = "https://files.pythonhosted.org/packages/60/1b/9851878b704bc98e641a3e0bce49382ae9e05743dac6d97748feb5b7baba/multidict-6.1.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:ab7c4ceb38d91570a650dba194e1ca87c2b543488fe9309b4212694174fd539c", size = 29641 },
+    { url = "https://files.pythonhosted.org/packages/89/87/d451d45aab9e422cb0fb2f7720c31a4c1d3012c740483c37f642eba568fb/multidict-6.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e617fb6b0b6953fffd762669610c1c4ffd05632c138d61ac7e14ad187870669c", size = 126202 },
+    { url = "https://files.pythonhosted.org/packages/fa/b4/27cbe9f3e2e469359887653f2e45470272eef7295139916cc21107c6b48c/multidict-6.1.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:16e5f4bf4e603eb1fdd5d8180f1a25f30056f22e55ce51fb3d6ad4ab29f7d96f", size = 133925 },
+    { url = "https://files.pythonhosted.org/packages/4d/a3/afc841899face8adfd004235ce759a37619f6ec99eafd959650c5ce4df57/multidict-6.1.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f4c035da3f544b1882bac24115f3e2e8760f10a0107614fc9839fd232200b875", size = 129039 },
+    { url = "https://files.pythonhosted.org/packages/5e/41/0d0fb18c1ad574f807196f5f3d99164edf9de3e169a58c6dc2d6ed5742b9/multidict-6.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:957cf8e4b6e123a9eea554fa7ebc85674674b713551de587eb318a2df3e00255", size = 124072 },
+    { url = "https://files.pythonhosted.org/packages/00/22/defd7a2e71a44e6e5b9a5428f972e5b572e7fe28e404dfa6519bbf057c93/multidict-6.1.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:483a6aea59cb89904e1ceabd2b47368b5600fb7de78a6e4a2c2987b2d256cf30", size = 116532 },
+    { url = "https://files.pythonhosted.org/packages/91/25/f7545102def0b1d456ab6449388eed2dfd822debba1d65af60194904a23a/multidict-6.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:87701f25a2352e5bf7454caa64757642734da9f6b11384c1f9d1a8e699758057", size = 128173 },
+    { url = "https://files.pythonhosted.org/packages/45/79/3dbe8d35fc99f5ea610813a72ab55f426cb9cf482f860fa8496e5409be11/multidict-6.1.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:682b987361e5fd7a139ed565e30d81fd81e9629acc7d925a205366877d8c8657", size = 122654 },
+    { url = "https://files.pythonhosted.org/packages/97/cb/209e735eeab96e1b160825b5d0b36c56d3862abff828fc43999bb957dcad/multidict-6.1.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:ce2186a7df133a9c895dea3331ddc5ddad42cdd0d1ea2f0a51e5d161e4762f28", size = 133197 },
+    { url = "https://files.pythonhosted.org/packages/e4/3a/a13808a7ada62808afccea67837a79d00ad6581440015ef00f726d064c2d/multidict-6.1.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:9f636b730f7e8cb19feb87094949ba54ee5357440b9658b2a32a5ce4bce53972", size = 129754 },
+    { url = "https://files.pythonhosted.org/packages/77/dd/8540e139eafb240079242da8f8ffdf9d3f4b4ad1aac5a786cd4050923783/multidict-6.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:73eae06aa53af2ea5270cc066dcaf02cc60d2994bbb2c4ef5764949257d10f43", size = 126402 },
+    { url = "https://files.pythonhosted.org/packages/86/99/e82e1a275d8b1ea16d3a251474262258dbbe41c05cce0c01bceda1fc8ea5/multidict-6.1.0-cp39-cp39-win32.whl", hash = "sha256:1ca0083e80e791cffc6efce7660ad24af66c8d4079d2a750b29001b53ff59ada", size = 26421 },
+    { url = "https://files.pythonhosted.org/packages/86/1c/9fa630272355af7e4446a2c7550c259f11ee422ab2d30ff90a0a71cf3d9e/multidict-6.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:aa466da5b15ccea564bdab9c89175c762bc12825f4659c11227f515cee76fa4a", size = 28791 },
+    { url = "https://files.pythonhosted.org/packages/99/b7/b9e70fde2c0f0c9af4cc5277782a89b66d35948ea3369ec9f598358c3ac5/multidict-6.1.0-py3-none-any.whl", hash = "sha256:48e171e52d1c4d33888e529b999e5900356b9ae588c2f09a52dcefb158b27506", size = 10051 },
+]
+
+[[package]]
+name = "multiprocess"
+version = "0.70.16"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "dill" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b5/ae/04f39c5d0d0def03247c2893d6f2b83c136bf3320a2154d7b8858f2ba72d/multiprocess-0.70.16.tar.gz", hash = "sha256:161af703d4652a0e1410be6abccecde4a7ddffd19341be0a7011b94aeb171ac1", size = 1772603 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ef/76/6e712a2623d146d314f17598df5de7224c85c0060ef63fd95cc15a25b3fa/multiprocess-0.70.16-pp310-pypy310_pp73-macosx_10_13_x86_64.whl", hash = "sha256:476887be10e2f59ff183c006af746cb6f1fd0eadcfd4ef49e605cbe2659920ee", size = 134980 },
+    { url = "https://files.pythonhosted.org/packages/0f/ab/1e6e8009e380e22254ff539ebe117861e5bdb3bff1fc977920972237c6c7/multiprocess-0.70.16-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:d951bed82c8f73929ac82c61f01a7b5ce8f3e5ef40f5b52553b4f547ce2b08ec", size = 134982 },
+    { url = "https://files.pythonhosted.org/packages/d8/94/8638a89f93c80df329116e6781a060506c7e91e1f4370dc831e9d17a041d/multiprocess-0.70.16-pp39-pypy39_pp73-macosx_10_13_x86_64.whl", hash = "sha256:0dfd078c306e08d46d7a8d06fb120313d87aa43af60d66da43ffff40b44d2f41", size = 133497 },
+    { url = "https://files.pythonhosted.org/packages/89/21/222066f6bb8d8af287923ae3bd26cf4699a9ce020228ac273caca1de8250/multiprocess-0.70.16-pp39-pypy39_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:e7b9d0f307cd9bd50851afaac0dba2cb6c44449efff697df7c7645f7d3f2be3a", size = 133498 },
+    { url = "https://files.pythonhosted.org/packages/bc/f7/7ec7fddc92e50714ea3745631f79bd9c96424cb2702632521028e57d3a36/multiprocess-0.70.16-py310-none-any.whl", hash = "sha256:c4a9944c67bd49f823687463660a2d6daae94c289adff97e0f9d696ba6371d02", size = 134824 },
+    { url = "https://files.pythonhosted.org/packages/50/15/b56e50e8debaf439f44befec5b2af11db85f6e0f344c3113ae0be0593a91/multiprocess-0.70.16-py311-none-any.whl", hash = "sha256:af4cabb0dac72abfb1e794fa7855c325fd2b55a10a44628a3c1ad3311c04127a", size = 143519 },
+    { url = "https://files.pythonhosted.org/packages/0a/7d/a988f258104dcd2ccf1ed40fdc97e26c4ac351eeaf81d76e266c52d84e2f/multiprocess-0.70.16-py312-none-any.whl", hash = "sha256:fc0544c531920dde3b00c29863377f87e1632601092ea2daca74e4beb40faa2e", size = 146741 },
+    { url = "https://files.pythonhosted.org/packages/ea/89/38df130f2c799090c978b366cfdf5b96d08de5b29a4a293df7f7429fa50b/multiprocess-0.70.16-py38-none-any.whl", hash = "sha256:a71d82033454891091a226dfc319d0cfa8019a4e888ef9ca910372a446de4435", size = 132628 },
+    { url = "https://files.pythonhosted.org/packages/da/d9/f7f9379981e39b8c2511c9e0326d212accacb82f12fbfdc1aa2ce2a7b2b6/multiprocess-0.70.16-py39-none-any.whl", hash = "sha256:a0bafd3ae1b732eac64be2e72038231c1ba97724b60b09400d68f229fcc2fbf3", size = 133351 },
+]
+
+[[package]]
+name = "nest-asyncio"
+version = "1.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/83/f8/51569ac65d696c8ecbee95938f89d4abf00f47d58d48f6fbabfe8f0baefe/nest_asyncio-1.6.0.tar.gz", hash = "sha256:6f172d5449aca15afd6c646851f4e31e02c598d553a667e38cafa997cfec55fe", size = 7418 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a0/c4/c2971a3ba4c6103a3d10c4b0f24f461ddc027f0f09763220cf35ca1401b3/nest_asyncio-1.6.0-py3-none-any.whl", hash = "sha256:87af6efd6b5e897c81050477ef65c62e2b2f35d51703cae01aff2905b1852e1c", size = 5195 },
+]
+
+[[package]]
+name = "networkx"
+version = "3.2.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
+    "python_full_version < '3.9.17'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/c4/80/a84676339aaae2f1cfdf9f418701dd634aef9cc76f708ef55c36ff39c3ca/networkx-3.2.1.tar.gz", hash = "sha256:9f1bb5cf3409bf324e0a722c20bdb4c20ee39bf1c30ce8ae499c8502b0b5e0c6", size = 2073928 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d5/f0/8fbc882ca80cf077f1b246c0e3c3465f7f415439bdea6b899f6b19f61f70/networkx-3.2.1-py3-none-any.whl", hash = "sha256:f18c69adc97877c42332c170849c96cefa91881c99a7cb3e95b7c659ebdc1ec2", size = 1647772 },
+]
+
+[[package]]
+name = "networkx"
+version = "3.4.2"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.13'",
+    "python_full_version == '3.12.*'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fd/1d/06475e1cd5264c0b870ea2cc6fdb3e37177c1e565c43f56ff17a10e3937f/networkx-3.4.2.tar.gz", hash = "sha256:307c3669428c5362aab27c8a1260aa8f47c4e91d3891f48be0141738d8d053e1", size = 2151368 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b9/54/dd730b32ea14ea797530a4479b2ed46a6fb250f682a9cfb997e968bf0261/networkx-3.4.2-py3-none-any.whl", hash = "sha256:df5d4365b724cf81b8c6a7312509d0c22386097011ad1abe274afd5e9d3bbc5f", size = 1723263 },
+]
+
+[[package]]
+name = "ninja"
+version = "1.11.1.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/bd/8f/21a2701f95b7d0d5137736561b3427ece0c4a1e085d4a223b92d16ab7d8b/ninja-1.11.1.3.tar.gz", hash = "sha256:edfa0d2e9d7ead1635b03e40a32ad56cc8f56798b6e2e9848d8300b174897076", size = 129532 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ea/ba/0069cd4a83d68f7b0308be70e219b15d675e50c8ea28763a3f0373c45bfc/ninja-1.11.1.3-py3-none-macosx_10_9_universal2.whl", hash = "sha256:2b4879ea3f1169f3d855182c57dcc84d1b5048628c8b7be0d702b81882a37237", size = 279132 },
+    { url = "https://files.pythonhosted.org/packages/72/6b/3805be87df8417a0c7b21078c8045f2a1e59b34f371bfe4cb4fb0d6df7f2/ninja-1.11.1.3-py3-none-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:bc3ebc8b2e47716149f3541742b5cd8e0b08f51013b825c05baca3e34854370d", size = 472101 },
+    { url = "https://files.pythonhosted.org/packages/6b/35/a8e38d54768e67324e365e2a41162be298f51ec93e6bd4b18d237d7250d8/ninja-1.11.1.3-py3-none-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:a27e78ca71316c8654965ee94b286a98c83877bfebe2607db96897bbfe458af0", size = 422884 },
+    { url = "https://files.pythonhosted.org/packages/2f/99/7996457319e139c02697fb2aa28e42fe32bb0752cef492edc69d56a3552e/ninja-1.11.1.3-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2883ea46b3c5079074f56820f9989c6261fcc6fd873d914ee49010ecf283c3b2", size = 157046 },
+    { url = "https://files.pythonhosted.org/packages/6d/8b/93f38e5cddf76ccfdab70946515b554f25d2b4c95ef9b2f9cfbc43fa7cc1/ninja-1.11.1.3-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8c4bdb9fd2d0c06501ae15abfd23407660e95659e384acd36e013b6dd7d8a8e4", size = 180014 },
+    { url = "https://files.pythonhosted.org/packages/7d/1d/713884d0fa3c972164f69d552e0701d30e2bf25eba9ef160bfb3dc69926a/ninja-1.11.1.3-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:114ed5c61c8474df6a69ab89097a20749b769e2c219a452cb2fadc49b0d581b0", size = 157098 },
+    { url = "https://files.pythonhosted.org/packages/c7/22/ecb0f70e77c9e22ee250aa717a608a142756833a34d43943d7d658ee0e56/ninja-1.11.1.3-py3-none-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:7fa2247fce98f683bc712562d82b22b8a0a5c000738a13147ca2d1b68c122298", size = 130089 },
+    { url = "https://files.pythonhosted.org/packages/ec/a6/3ee846c20ab6ad95b90c5c8703c76cb1f39cc8ce2d1ae468956e3b1b2581/ninja-1.11.1.3-py3-none-musllinux_1_1_aarch64.whl", hash = "sha256:a38c6c6c8032bed68b70c3b065d944c35e9f903342875d3a3218c1607987077c", size = 372508 },
+    { url = "https://files.pythonhosted.org/packages/95/0d/aa44abe4141f29148ce671ac8c92045878906b18691c6f87a29711c2ff1c/ninja-1.11.1.3-py3-none-musllinux_1_1_i686.whl", hash = "sha256:56ada5d33b8741d298836644042faddebc83ee669782d661e21563034beb5aba", size = 419369 },
+    { url = "https://files.pythonhosted.org/packages/f7/ec/48bf5105568ac9bd2016b701777bdd5000cc09a14ac837fef9f15e8d634e/ninja-1.11.1.3-py3-none-musllinux_1_1_ppc64le.whl", hash = "sha256:53409151da081f3c198bb0bfc220a7f4e821e022c5b7d29719adda892ddb31bb", size = 420304 },
+    { url = "https://files.pythonhosted.org/packages/18/e5/69df63976cf971a03379899f8520a036c9dbab26330b37197512aed5b3df/ninja-1.11.1.3-py3-none-musllinux_1_1_s390x.whl", hash = "sha256:1ad2112c2b0159ed7c4ae3731595191b1546ba62316fc40808edecd0306fefa3", size = 416056 },
+    { url = "https://files.pythonhosted.org/packages/6f/4f/bdb401af7ed0e24a3fef058e13a149f2de1ce4b176699076993615d55610/ninja-1.11.1.3-py3-none-musllinux_1_1_x86_64.whl", hash = "sha256:28aea3c1c280cba95b8608d50797169f3a34280e3e9a6379b6e340f0c9eaeeb0", size = 379725 },
+    { url = "https://files.pythonhosted.org/packages/bd/68/05e7863bf13128c61652eeb3ec7096c3d3a602f32f31752dbfb034e3fa07/ninja-1.11.1.3-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:b6966f83064a88a51693073eea3decd47e08c3965241e09578ef7aa3a7738329", size = 434881 },
+    { url = "https://files.pythonhosted.org/packages/bd/ad/edc0d1efe77f29f45bbca2e1dab07ef597f61a88de6e4bccffc0aec2256c/ninja-1.11.1.3-py3-none-win32.whl", hash = "sha256:a4a3b71490557e18c010cbb26bd1ea9a0c32ee67e8f105e9731515b6e0af792e", size = 255988 },
+    { url = "https://files.pythonhosted.org/packages/03/93/09a9f7672b4f97438aca6217ac54212a63273f1cd3b46b731d0bb22c53e7/ninja-1.11.1.3-py3-none-win_amd64.whl", hash = "sha256:04d48d14ea7ba11951c156599ab526bdda575450797ff57c6fdf99b2554d09c7", size = 296502 },
+    { url = "https://files.pythonhosted.org/packages/d9/9d/0cc1e82849070ff3cbee69f326cb48a839407bcd15d8844443c30a5e7509/ninja-1.11.1.3-py3-none-win_arm64.whl", hash = "sha256:17978ad611d8ead578d83637f5ae80c2261b033db0b493a7ce94f88623f29e1b", size = 270571 },
+]
+
+[[package]]
+name = "nodeenv"
+version = "1.9.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/43/16/fc88b08840de0e0a72a2f9d8c6bae36be573e475a6326ae854bcc549fc45/nodeenv-1.9.1.tar.gz", hash = "sha256:6ec12890a2dab7946721edbfbcd91f3319c6ccc9aec47be7c7e6b7011ee6645f", size = 47437 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d2/1d/1b658dbd2b9fa9c4c9f32accbfc0205d532c8c6194dc0f2a4c0428e7128a/nodeenv-1.9.1-py2.py3-none-any.whl", hash = "sha256:ba11c9782d29c27c70ffbdda2d7415098754709be8a7056d79a737cd901155c9", size = 22314 },
+]
+
+[[package]]
+name = "numpy"
+version = "1.26.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/65/6e/09db70a523a96d25e115e71cc56a6f9031e7b8cd166c1ac8438307c14058/numpy-1.26.4.tar.gz", hash = "sha256:2a02aba9ed12e4ac4eb3ea9421c420301a0c6460d9830d74a9df87efa4912010", size = 15786129 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a7/94/ace0fdea5241a27d13543ee117cbc65868e82213fb31a8eb7fe9ff23f313/numpy-1.26.4-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:9ff0f4f29c51e2803569d7a51c2304de5554655a60c5d776e35b4a41413830d0", size = 20631468 },
+    { url = "https://files.pythonhosted.org/packages/20/f7/b24208eba89f9d1b58c1668bc6c8c4fd472b20c45573cb767f59d49fb0f6/numpy-1.26.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:2e4ee3380d6de9c9ec04745830fd9e2eccb3e6cf790d39d7b98ffd19b0dd754a", size = 13966411 },
+    { url = "https://files.pythonhosted.org/packages/fc/a5/4beee6488160798683eed5bdb7eead455892c3b4e1f78d79d8d3f3b084ac/numpy-1.26.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d209d8969599b27ad20994c8e41936ee0964e6da07478d6c35016bc386b66ad4", size = 14219016 },
+    { url = "https://files.pythonhosted.org/packages/4b/d7/ecf66c1cd12dc28b4040b15ab4d17b773b87fa9d29ca16125de01adb36cd/numpy-1.26.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ffa75af20b44f8dba823498024771d5ac50620e6915abac414251bd971b4529f", size = 18240889 },
+    { url = "https://files.pythonhosted.org/packages/24/03/6f229fe3187546435c4f6f89f6d26c129d4f5bed40552899fcf1f0bf9e50/numpy-1.26.4-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:62b8e4b1e28009ef2846b4c7852046736bab361f7aeadeb6a5b89ebec3c7055a", size = 13876746 },
+    { url = "https://files.pythonhosted.org/packages/39/fe/39ada9b094f01f5a35486577c848fe274e374bbf8d8f472e1423a0bbd26d/numpy-1.26.4-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:a4abb4f9001ad2858e7ac189089c42178fcce737e4169dc61321660f1a96c7d2", size = 18078620 },
+    { url = "https://files.pythonhosted.org/packages/d5/ef/6ad11d51197aad206a9ad2286dc1aac6a378059e06e8cf22cd08ed4f20dc/numpy-1.26.4-cp310-cp310-win32.whl", hash = "sha256:bfe25acf8b437eb2a8b2d49d443800a5f18508cd811fea3181723922a8a82b07", size = 5972659 },
+    { url = "https://files.pythonhosted.org/packages/19/77/538f202862b9183f54108557bfda67e17603fc560c384559e769321c9d92/numpy-1.26.4-cp310-cp310-win_amd64.whl", hash = "sha256:b97fe8060236edf3662adfc2c633f56a08ae30560c56310562cb4f95500022d5", size = 15808905 },
+    { url = "https://files.pythonhosted.org/packages/11/57/baae43d14fe163fa0e4c47f307b6b2511ab8d7d30177c491960504252053/numpy-1.26.4-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4c66707fabe114439db9068ee468c26bbdf909cac0fb58686a42a24de1760c71", size = 20630554 },
+    { url = "https://files.pythonhosted.org/packages/1a/2e/151484f49fd03944c4a3ad9c418ed193cfd02724e138ac8a9505d056c582/numpy-1.26.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:edd8b5fe47dab091176d21bb6de568acdd906d1887a4584a15a9a96a1dca06ef", size = 13997127 },
+    { url = "https://files.pythonhosted.org/packages/79/ae/7e5b85136806f9dadf4878bf73cf223fe5c2636818ba3ab1c585d0403164/numpy-1.26.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7ab55401287bfec946ced39700c053796e7cc0e3acbef09993a9ad2adba6ca6e", size = 14222994 },
+    { url = "https://files.pythonhosted.org/packages/3a/d0/edc009c27b406c4f9cbc79274d6e46d634d139075492ad055e3d68445925/numpy-1.26.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:666dbfb6ec68962c033a450943ded891bed2d54e6755e35e5835d63f4f6931d5", size = 18252005 },
+    { url = "https://files.pythonhosted.org/packages/09/bf/2b1aaf8f525f2923ff6cfcf134ae5e750e279ac65ebf386c75a0cf6da06a/numpy-1.26.4-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:96ff0b2ad353d8f990b63294c8986f1ec3cb19d749234014f4e7eb0112ceba5a", size = 13885297 },
+    { url = "https://files.pythonhosted.org/packages/df/a0/4e0f14d847cfc2a633a1c8621d00724f3206cfeddeb66d35698c4e2cf3d2/numpy-1.26.4-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:60dedbb91afcbfdc9bc0b1f3f402804070deed7392c23eb7a7f07fa857868e8a", size = 18093567 },
+    { url = "https://files.pythonhosted.org/packages/d2/b7/a734c733286e10a7f1a8ad1ae8c90f2d33bf604a96548e0a4a3a6739b468/numpy-1.26.4-cp311-cp311-win32.whl", hash = "sha256:1af303d6b2210eb850fcf03064d364652b7120803a0b872f5211f5234b399f20", size = 5968812 },
+    { url = "https://files.pythonhosted.org/packages/3f/6b/5610004206cf7f8e7ad91c5a85a8c71b2f2f8051a0c0c4d5916b76d6cbb2/numpy-1.26.4-cp311-cp311-win_amd64.whl", hash = "sha256:cd25bcecc4974d09257ffcd1f098ee778f7834c3ad767fe5db785be9a4aa9cb2", size = 15811913 },
+    { url = "https://files.pythonhosted.org/packages/95/12/8f2020a8e8b8383ac0177dc9570aad031a3beb12e38847f7129bacd96228/numpy-1.26.4-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:b3ce300f3644fb06443ee2222c2201dd3a89ea6040541412b8fa189341847218", size = 20335901 },
+    { url = "https://files.pythonhosted.org/packages/75/5b/ca6c8bd14007e5ca171c7c03102d17b4f4e0ceb53957e8c44343a9546dcc/numpy-1.26.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:03a8c78d01d9781b28a6989f6fa1bb2c4f2d51201cf99d3dd875df6fbd96b23b", size = 13685868 },
+    { url = "https://files.pythonhosted.org/packages/79/f8/97f10e6755e2a7d027ca783f63044d5b1bc1ae7acb12afe6a9b4286eac17/numpy-1.26.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9fad7dcb1aac3c7f0584a5a8133e3a43eeb2fe127f47e3632d43d677c66c102b", size = 13925109 },
+    { url = "https://files.pythonhosted.org/packages/0f/50/de23fde84e45f5c4fda2488c759b69990fd4512387a8632860f3ac9cd225/numpy-1.26.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:675d61ffbfa78604709862923189bad94014bef562cc35cf61d3a07bba02a7ed", size = 17950613 },
+    { url = "https://files.pythonhosted.org/packages/4c/0c/9c603826b6465e82591e05ca230dfc13376da512b25ccd0894709b054ed0/numpy-1.26.4-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:ab47dbe5cc8210f55aa58e4805fe224dac469cde56b9f731a4c098b91917159a", size = 13572172 },
+    { url = "https://files.pythonhosted.org/packages/76/8c/2ba3902e1a0fc1c74962ea9bb33a534bb05984ad7ff9515bf8d07527cadd/numpy-1.26.4-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:1dda2e7b4ec9dd512f84935c5f126c8bd8b9f2fc001e9f54af255e8c5f16b0e0", size = 17786643 },
+    { url = "https://files.pythonhosted.org/packages/28/4a/46d9e65106879492374999e76eb85f87b15328e06bd1550668f79f7b18c6/numpy-1.26.4-cp312-cp312-win32.whl", hash = "sha256:50193e430acfc1346175fcbdaa28ffec49947a06918b7b92130744e81e640110", size = 5677803 },
+    { url = "https://files.pythonhosted.org/packages/16/2e/86f24451c2d530c88daf997cb8d6ac622c1d40d19f5a031ed68a4b73a374/numpy-1.26.4-cp312-cp312-win_amd64.whl", hash = "sha256:08beddf13648eb95f8d867350f6a018a4be2e5ad54c8d8caed89ebca558b2818", size = 15517754 },
+    { url = "https://files.pythonhosted.org/packages/7d/24/ce71dc08f06534269f66e73c04f5709ee024a1afe92a7b6e1d73f158e1f8/numpy-1.26.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7349ab0fa0c429c82442a27a9673fc802ffdb7c7775fad780226cb234965e53c", size = 20636301 },
+    { url = "https://files.pythonhosted.org/packages/ae/8c/ab03a7c25741f9ebc92684a20125fbc9fc1b8e1e700beb9197d750fdff88/numpy-1.26.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:52b8b60467cd7dd1e9ed082188b4e6bb35aa5cdd01777621a1658910745b90be", size = 13971216 },
+    { url = "https://files.pythonhosted.org/packages/6d/64/c3bcdf822269421d85fe0d64ba972003f9bb4aa9a419da64b86856c9961f/numpy-1.26.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d5241e0a80d808d70546c697135da2c613f30e28251ff8307eb72ba696945764", size = 14226281 },
+    { url = "https://files.pythonhosted.org/packages/54/30/c2a907b9443cf42b90c17ad10c1e8fa801975f01cb9764f3f8eb8aea638b/numpy-1.26.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f870204a840a60da0b12273ef34f7051e98c3b5961b61b0c2c1be6dfd64fbcd3", size = 18249516 },
+    { url = "https://files.pythonhosted.org/packages/43/12/01a563fc44c07095996d0129b8899daf89e4742146f7044cdbdb3101c57f/numpy-1.26.4-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:679b0076f67ecc0138fd2ede3a8fd196dddc2ad3254069bcb9faf9a79b1cebcd", size = 13882132 },
+    { url = "https://files.pythonhosted.org/packages/16/ee/9df80b06680aaa23fc6c31211387e0db349e0e36d6a63ba3bd78c5acdf11/numpy-1.26.4-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:47711010ad8555514b434df65f7d7b076bb8261df1ca9bb78f53d3b2db02e95c", size = 18084181 },
+    { url = "https://files.pythonhosted.org/packages/28/7d/4b92e2fe20b214ffca36107f1a3e75ef4c488430e64de2d9af5db3a4637d/numpy-1.26.4-cp39-cp39-win32.whl", hash = "sha256:a354325ee03388678242a4d7ebcd08b5c727033fcff3b2f536aea978e15ee9e6", size = 5976360 },
+    { url = "https://files.pythonhosted.org/packages/b5/42/054082bd8220bbf6f297f982f0a8f5479fcbc55c8b511d928df07b965869/numpy-1.26.4-cp39-cp39-win_amd64.whl", hash = "sha256:3373d5d70a5fe74a2c1bb6d2cfd9609ecf686d47a2d7b1d37a8f3b6bf6003aea", size = 15814633 },
+    { url = "https://files.pythonhosted.org/packages/3f/72/3df6c1c06fc83d9cfe381cccb4be2532bbd38bf93fbc9fad087b6687f1c0/numpy-1.26.4-pp39-pypy39_pp73-macosx_10_9_x86_64.whl", hash = "sha256:afedb719a9dcfc7eaf2287b839d8198e06dcd4cb5d276a3df279231138e83d30", size = 20455961 },
+    { url = "https://files.pythonhosted.org/packages/8e/02/570545bac308b58ffb21adda0f4e220ba716fb658a63c151daecc3293350/numpy-1.26.4-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:95a7476c59002f2f6c590b9b7b998306fba6a5aa646b1e22ddfeaf8f78c3a29c", size = 18061071 },
+    { url = "https://files.pythonhosted.org/packages/f4/5f/fafd8c51235f60d49f7a88e2275e13971e90555b67da52dd6416caec32fe/numpy-1.26.4-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:7e50d0a0cc3189f9cb0aeb3a6a6af18c16f59f004b866cd2be1c14b36134a4a0", size = 15709730 },
+]
+
+[[package]]
+name = "nvidia-cublas-cu12"
+version = "12.1.3.1"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/37/6d/121efd7382d5b0284239f4ab1fc1590d86d34ed4a4a2fdb13b30ca8e5740/nvidia_cublas_cu12-12.1.3.1-py3-none-manylinux1_x86_64.whl", hash = "sha256:ee53ccca76a6fc08fb9701aa95b6ceb242cdaab118c3bb152af4e579af792728", size = 410594774 },
+]
+
+[[package]]
+name = "nvidia-cuda-cupti-cu12"
+version = "12.1.105"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7e/00/6b218edd739ecfc60524e585ba8e6b00554dd908de2c9c66c1af3e44e18d/nvidia_cuda_cupti_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:e54fde3983165c624cb79254ae9818a456eb6e87a7fd4d56a2352c24ee542d7e", size = 14109015 },
+]
+
+[[package]]
+name = "nvidia-cuda-nvrtc-cu12"
+version = "12.1.105"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b6/9f/c64c03f49d6fbc56196664d05dba14e3a561038a81a638eeb47f4d4cfd48/nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:339b385f50c309763ca65456ec75e17bbefcbbf2893f462cb8b90584cd27a1c2", size = 23671734 },
+]
+
+[[package]]
+name = "nvidia-cuda-runtime-cu12"
+version = "12.1.105"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/eb/d5/c68b1d2cdfcc59e72e8a5949a37ddb22ae6cade80cd4a57a84d4c8b55472/nvidia_cuda_runtime_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:6e258468ddf5796e25f1dc591a31029fa317d97a0a94ed93468fc86301d61e40", size = 823596 },
+]
+
+[[package]]
+name = "nvidia-cudnn-cu12"
+version = "8.9.2.26"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-cublas-cu12" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ff/74/a2e2be7fb83aaedec84f391f082cf765dfb635e7caa9b49065f73e4835d8/nvidia_cudnn_cu12-8.9.2.26-py3-none-manylinux1_x86_64.whl", hash = "sha256:5ccb288774fdfb07a7e7025ffec286971c06d8d7b4fb162525334616d7629ff9", size = 731725872 },
+]
+
+[[package]]
+name = "nvidia-cufft-cu12"
+version = "11.0.2.54"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/86/94/eb540db023ce1d162e7bea9f8f5aa781d57c65aed513c33ee9a5123ead4d/nvidia_cufft_cu12-11.0.2.54-py3-none-manylinux1_x86_64.whl", hash = "sha256:794e3948a1aa71fd817c3775866943936774d1c14e7628c74f6f7417224cdf56", size = 121635161 },
+]
+
+[[package]]
+name = "nvidia-curand-cu12"
+version = "10.3.2.106"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/44/31/4890b1c9abc496303412947fc7dcea3d14861720642b49e8ceed89636705/nvidia_curand_cu12-10.3.2.106-py3-none-manylinux1_x86_64.whl", hash = "sha256:9d264c5036dde4e64f1de8c50ae753237c12e0b1348738169cd0f8a536c0e1e0", size = 56467784 },
+]
+
+[[package]]
+name = "nvidia-cusolver-cu12"
+version = "11.4.5.107"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-cublas-cu12" },
+    { name = "nvidia-cusparse-cu12" },
+    { name = "nvidia-nvjitlink-cu12" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/bc/1d/8de1e5c67099015c834315e333911273a8c6aaba78923dd1d1e25fc5f217/nvidia_cusolver_cu12-11.4.5.107-py3-none-manylinux1_x86_64.whl", hash = "sha256:8a7ec542f0412294b15072fa7dab71d31334014a69f953004ea7a118206fe0dd", size = 124161928 },
+]
+
+[[package]]
+name = "nvidia-cusparse-cu12"
+version = "12.1.0.106"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-nvjitlink-cu12" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/65/5b/cfaeebf25cd9fdec14338ccb16f6b2c4c7fa9163aefcf057d86b9cc248bb/nvidia_cusparse_cu12-12.1.0.106-py3-none-manylinux1_x86_64.whl", hash = "sha256:f3b50f42cf363f86ab21f720998517a659a48131e8d538dc02f8768237bd884c", size = 195958278 },
+]
+
+[[package]]
+name = "nvidia-ml-py"
+version = "12.560.30"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/53/10/5f34de4a71db8b2b7ec4269f4a33287f24c23e2857ea3187c977b7bc3604/nvidia-ml-py-12.560.30.tar.gz", hash = "sha256:f0254dc7400647680a072ee02509bfd46102b60bdfeca321576d4d4817e7fe97", size = 39194 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b7/f3/a69ce0b1a1e12fbf6b2ad9f4c14c9999fdbdf15f2478d210f0fd501ddc98/nvidia_ml_py-12.560.30-py3-none-any.whl", hash = "sha256:fea371c94d63e38a611c17bbb85fe400e9c8ddb9e8684a9cd0e47786a4bc3c73", size = 40526 },
+]
+
+[[package]]
+name = "nvidia-nccl-cu12"
+version = "2.20.5"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/4b/2a/0a131f572aa09f741c30ccd45a8e56316e8be8dfc7bc19bf0ab7cfef7b19/nvidia_nccl_cu12-2.20.5-py3-none-manylinux2014_x86_64.whl", hash = "sha256:057f6bf9685f75215d0c53bf3ac4a10b3e6578351de307abad9e18a99182af56", size = 176249402 },
+]
+
+[[package]]
+name = "nvidia-nvjitlink-cu12"
+version = "12.6.85"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9d/d7/c5383e47c7e9bf1c99d5bd2a8c935af2b6d705ad831a7ec5c97db4d82f4f/nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:eedc36df9e88b682efe4309aa16b5b4e78c2407eac59e8c10a6a47535164369a", size = 19744971 },
+]
+
+[[package]]
+name = "nvidia-nvtx-cu12"
+version = "12.1.105"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/da/d3/8057f0587683ed2fcd4dbfbdfdfa807b9160b809976099d36b8f60d08f03/nvidia_nvtx_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:dc21cf308ca5691e7c04d962e213f8a4aa9bbfa23d95412f452254c2caeb09e5", size = 99138 },
+]
+
+[[package]]
+name = "openai"
+version = "1.59.8"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+    { name = "distro" },
+    { name = "httpx" },
+    { name = "jiter" },
+    { name = "pydantic" },
+    { name = "sniffio" },
+    { name = "tqdm" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/e0/c4/b4482784de63c7158f6c0afcb07fd66450ea6c912d6bddf9d7599f2eda25/openai-1.59.8.tar.gz", hash = "sha256:ac4bda5fa9819fdc6127e8ea8a63501f425c587244bc653c7c11a8ad84f953e1", size = 346775 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8c/cf/5b235e12ead3cd2098f9792776c966994c1bc558cba5799e12f3045227df/openai-1.59.8-py3-none-any.whl", hash = "sha256:a8b8ee35c4083b88e6da45406d883cf6bd91a98ab7dd79178b8bc24c8bfb09d9", size = 455567 },
+]
+
+[[package]]
+name = "opt-einsum"
+version = "3.4.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8c/b9/2ac072041e899a52f20cf9510850ff58295003aa75525e58343591b0cbfb/opt_einsum-3.4.0.tar.gz", hash = "sha256:96ca72f1b886d148241348783498194c577fa30a8faac108586b14f1ba4473ac", size = 63004 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/23/cd/066e86230ae37ed0be70aae89aabf03ca8d9f39c8aea0dec8029455b5540/opt_einsum-3.4.0-py3-none-any.whl", hash = "sha256:69bb92469f86a1565195ece4ac0323943e83477171b91d24c35afe028a90d7cd", size = 71932 },
+]
+
+[[package]]
+name = "outlines"
+source = { editable = "." }
+dependencies = [
+    { name = "airportsdata" },
+    { name = "cloudpickle" },
+    { name = "diskcache" },
+    { name = "genson" },
+    { name = "interegular" },
+    { name = "jinja2" },
+    { name = "jsonschema" },
+    { name = "lark" },
+    { name = "nest-asyncio" },
+    { name = "numpy" },
+    { name = "outlines-core" },
+    { name = "pre-commit" },
+    { name = "pycountry" },
+    { name = "pydantic" },
+    { name = "referencing" },
+    { name = "requests" },
+    { name = "torch" },
+    { name = "tqdm" },
+    { name = "typing-extensions" },
+]
+
+[package.optional-dependencies]
+exllamav2 = [
+    { name = "exllamav2" },
+]
+llamacpp = [
+    { name = "datasets" },
+    { name = "llama-cpp-python" },
+    { name = "numpy" },
+    { name = "transformers" },
+]
+mlxlm = [
+    { name = "datasets" },
+    { name = "mlx-lm" },
+]
+openai = [
+    { name = "openai" },
+]
+serve = [
+    { name = "fastapi" },
+    { name = "pydantic" },
+    { name = "uvicorn" },
+    { name = "vllm" },
+]
+test = [
+    { name = "accelerate" },
+    { name = "beartype" },
+    { name = "coverage", extra = ["toml"] },
+    { name = "datasets" },
+    { name = "diff-cover", version = "9.2.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.9.17'" },
+    { name = "diff-cover", version = "9.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.9.17'" },
+    { name = "exllamav2" },
+    { name = "huggingface-hub" },
+    { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "jax", version = "0.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "llama-cpp-python" },
+    { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
+    { name = "openai" },
+    { name = "pillow" },
+    { name = "pre-commit" },
+    { name = "pytest" },
+    { name = "pytest-benchmark" },
+    { name = "pytest-cov" },
+    { name = "pytest-mock" },
+    { name = "responses" },
+    { name = "transformers" },
+    { name = "vllm", marker = "sys_platform == 'linux'" },
+]
+transformers = [
+    { name = "accelerate" },
+    { name = "datasets" },
+    { name = "numpy" },
+    { name = "transformers" },
+]
+vllm = [
+    { name = "numpy" },
+    { name = "transformers" },
+    { name = "vllm" },
+]
+
+[package.metadata]
+requires-dist = [
+    { name = "accelerate", marker = "extra == 'test'" },
+    { name = "accelerate", marker = "extra == 'transformers'" },
+    { name = "airportsdata" },
+    { name = "beartype", marker = "extra == 'test'", specifier = "<0.16.0" },
+    { name = "cloudpickle" },
+    { name = "coverage", extras = ["toml"], marker = "extra == 'test'", specifier = ">=5.1" },
+    { name = "datasets", marker = "extra == 'llamacpp'" },
+    { name = "datasets", marker = "extra == 'mlxlm'" },
+    { name = "datasets", marker = "extra == 'test'" },
+    { name = "datasets", marker = "extra == 'transformers'" },
+    { name = "diff-cover", marker = "extra == 'test'" },
+    { name = "diskcache" },
+    { name = "exllamav2", marker = "extra == 'exllamav2'" },
+    { name = "exllamav2", marker = "extra == 'test'" },
+    { name = "fastapi", marker = "extra == 'serve'" },
+    { name = "genson" },
+    { name = "huggingface-hub", marker = "extra == 'test'" },
+    { name = "interegular" },
+    { name = "jax", marker = "extra == 'test'" },
+    { name = "jinja2" },
+    { name = "jsonschema" },
+    { name = "lark" },
+    { name = "llama-cpp-python", marker = "extra == 'llamacpp'" },
+    { name = "llama-cpp-python", marker = "extra == 'test'" },
+    { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin' and extra == 'test'", specifier = ">=0.19.2" },
+    { name = "mlx-lm", marker = "extra == 'mlxlm'" },
+    { name = "nest-asyncio" },
+    { name = "numpy" },
+    { name = "numpy", marker = "extra == 'llamacpp'", specifier = "<2" },
+    { name = "numpy", marker = "extra == 'transformers'", specifier = "<2" },
+    { name = "numpy", marker = "extra == 'vllm'", specifier = "<2" },
+    { name = "openai", marker = "extra == 'openai'" },
+    { name = "openai", marker = "extra == 'test'", specifier = ">=1.0.0" },
+    { name = "outlines-core", specifier = "==0.1.26" },
+    { name = "pillow", marker = "extra == 'test'" },
+    { name = "pre-commit", specifier = ">=4.0.1" },
+    { name = "pre-commit", marker = "extra == 'test'" },
+    { name = "pycountry" },
+    { name = "pydantic", specifier = ">=2.0" },
+    { name = "pydantic", marker = "extra == 'serve'", specifier = ">=2.0" },
+    { name = "pytest", marker = "extra == 'test'" },
+    { name = "pytest-benchmark", marker = "extra == 'test'" },
+    { name = "pytest-cov", marker = "extra == 'test'" },
+    { name = "pytest-mock", marker = "extra == 'test'" },
+    { name = "referencing" },
+    { name = "requests" },
+    { name = "responses", marker = "extra == 'test'" },
+    { name = "torch" },
+    { name = "tqdm" },
+    { name = "transformers", marker = "extra == 'llamacpp'" },
+    { name = "transformers", marker = "extra == 'test'" },
+    { name = "transformers", marker = "extra == 'transformers'" },
+    { name = "transformers", marker = "extra == 'vllm'" },
+    { name = "typing-extensions" },
+    { name = "uvicorn", marker = "extra == 'serve'" },
+    { name = "vllm", marker = "sys_platform == 'linux' and extra == 'test'" },
+    { name = "vllm", marker = "extra == 'serve'", specifier = ">=0.3.0" },
+    { name = "vllm", marker = "extra == 'vllm'" },
+]
+
+[[package]]
+name = "outlines-core"
+version = "0.1.26"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "interegular" },
+    { name = "jsonschema" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/d3/f3/274d07f4702728b43581235a77e545ec602b25f9b0098b288a0f3052521d/outlines_core-0.1.26.tar.gz", hash = "sha256:481c4301341e77cc8f1832d616784adb4d461b4fec65878e7c0d2cba7163a189", size = 75139 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e2/df/e9ff00f1dcf671cb8c4c20abcfd53406328b344cafa689a2832e8059c0b4/outlines_core-0.1.26-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:6a962a7452e7ac170fa04d405342cadae2d28fafa5b1830cef7aa610257ed32f", size = 322602 },
+    { url = "https://files.pythonhosted.org/packages/3c/f1/e9064f18c462a61f4abbe73b24f25e36d8abef19c593416fa69dce6a83c0/outlines_core-0.1.26-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:15a3684fa29564da2db03934cf0097bef3e871f70d3af0ef2b52fdb886da2e09", size = 301929 },
+    { url = "https://files.pythonhosted.org/packages/76/c3/6bc82db40b4818421e573237f43d4026c40a3305fa2558eb0aa1a7aa08f7/outlines_core-0.1.26-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:64e01c0cfa9ba371634d7c3f6ea1862397cef98e4509fe98e3f57faa721a72d6", size = 321355 },
+    { url = "https://files.pythonhosted.org/packages/c9/c2/1d85bfeaee3a83327e0d162bee4bdc7d7889bea5998e44fcc66c924dc1fd/outlines_core-0.1.26-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a3c4196148e47f455f1ace78e329d5b97e531cbc406456d681592952adae7e17", size = 343552 },
+    { url = "https://files.pythonhosted.org/packages/45/da/1e61d3d997ba1858fb8e71c3127f24a95c30575559da012ea5b45b147ad3/outlines_core-0.1.26-cp310-cp310-win32.whl", hash = "sha256:f38d290a7f6e5e12cbfcaee03269dfc0dbda49b360024b4279d1aba251fdc346", size = 234750 },
+    { url = "https://files.pythonhosted.org/packages/1c/04/6d7968019a81df235ad6bc7405eefe32be8da4c4153792655e7490d06c8d/outlines_core-0.1.26-cp310-cp310-win_amd64.whl", hash = "sha256:11ff56af56cb54c563b7f25d86cd9ee77f3fed825f1d4dccd9449bb1e4e89538", size = 243713 },
+    { url = "https://files.pythonhosted.org/packages/17/94/19d5c50c303ba71f3465c81620ca9b5af4db07fd8922dfe59ae5a9ae61d1/outlines_core-0.1.26-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:b6787b07b7c673fc3087d2b537719ecac8e03b10a47d032dd1926985c32885b0", size = 322344 },
+    { url = "https://files.pythonhosted.org/packages/f2/ea/f44beea7f610f2737ebb908c8dfa37d8324e92ca529468a56b00a77af199/outlines_core-0.1.26-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1e0ea28a76da31d25b6f53242bf13e1b59a0241badf82353c88f55e1cf81b128", size = 301670 },
+    { url = "https://files.pythonhosted.org/packages/6a/a6/ceac3760e1feb898b4047aeb54e0a3de975b59e87a17d6ba0a04dec5eaed/outlines_core-0.1.26-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a8932044a3d9329be53a226118850638f85b4d7842f9b863d0a123f23de220cd", size = 321067 },
+    { url = "https://files.pythonhosted.org/packages/92/f0/ad0074d6726fed86bb0bba1b9307cbbd67a2af5debd3540d66c69298a001/outlines_core-0.1.26-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a84b7cd2fb6268bf990dd3d479ffb4fa0bace6f571cb85b15b6cdb44b84f5b69", size = 343264 },
+    { url = "https://files.pythonhosted.org/packages/e6/bd/198c9a73d5f36e2ecad558a26359af3f0dbe4f5ba11c4629e46fccdfe2d6/outlines_core-0.1.26-cp311-cp311-win32.whl", hash = "sha256:f19765c151abfc970996368080aeea6d2a19e927817fe4e2af6726e639be3de4", size = 234529 },
+    { url = "https://files.pythonhosted.org/packages/b9/27/354b484045e6368c92f688d954124064ec2ce961681e56711852904e1ec2/outlines_core-0.1.26-cp311-cp311-win_amd64.whl", hash = "sha256:3f59aeccea21ed6ff3cf52102fd163f26d279821c20e5127ddd18d4ea4d0c8d2", size = 243457 },
+    { url = "https://files.pythonhosted.org/packages/c6/86/0fb40746e579db38d89f127122a3900d9e0350f76aae8cb61adeaff44cc2/outlines_core-0.1.26-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f54633bca50055d42ea4d94ae06dcbe52d3d76a9b621b75723b1177d0d952953", size = 321874 },
+    { url = "https://files.pythonhosted.org/packages/ab/0c/b91f7bc03843796c1d643ee030b6cd8fd5a8ba2cd4856c855f140c878976/outlines_core-0.1.26-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:9525321b48700dcaaabf60bcdc951e45f9357ba3fb3e1bfc81b662d7d4170e7c", size = 301995 },
+    { url = "https://files.pythonhosted.org/packages/ad/db/fa91a2d54288b900de82d86eda3adb2417b3b5b2db6256854a5e8bc85c32/outlines_core-0.1.26-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:00f409f72c11f6ffadb57066950dd384d5388015028c1a1a615c9a64988dae3e", size = 321050 },
+    { url = "https://files.pythonhosted.org/packages/e2/1d/a36292b6198986bd9c3ff8c24355deb82ed5475403379ee40b5b5473e2e3/outlines_core-0.1.26-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e86a1bb46adc5cbf6dfd7a7fe4105e0e2a4c6e041732a053126b41c521a1f223", size = 343201 },
+    { url = "https://files.pythonhosted.org/packages/08/63/5dd2b5a364412f674b6edcb59b0c21513bdb07cdcc7613b064c1a0660d01/outlines_core-0.1.26-cp312-cp312-win32.whl", hash = "sha256:19f462f6b00935708677ad27cb4df55e0e17f6ffe713ab750f5f2683b090f95d", size = 233970 },
+    { url = "https://files.pythonhosted.org/packages/a5/56/8adf0b7446d1e975c2314454813c59eb7b195889908a2932ed34148c113c/outlines_core-0.1.26-cp312-cp312-win_amd64.whl", hash = "sha256:9b36bff12779e58883747116893a17b3551bbd10865878b951b03a44d112229a", size = 243578 },
+    { url = "https://files.pythonhosted.org/packages/5b/b8/d38f13e417f2af82f6cd588df5a26ff8ffeb9a3a15b8a1a883d5922048ac/outlines_core-0.1.26-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:7b7849cf40028319ebb9d8ba0fe4c590ef5888eebe524a81b3af30aaa06ea21c", size = 323047 },
+    { url = "https://files.pythonhosted.org/packages/3b/59/43f5c7b8ad0e98a663db8f24cbff0421a58e946934e3f3751a17a06ae84d/outlines_core-0.1.26-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2f8641aab4a6bd84516907492ce82099503129da01b3c29c1dc9ad50320bae77", size = 302547 },
+    { url = "https://files.pythonhosted.org/packages/23/30/d9eb8202cbc994536a1fefa60e87c582a714ad1b46367c43af299f3889ba/outlines_core-0.1.26-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bba56604efdbc5932c7a8a88c2b8b0d0c740ab883b0012fb5464a9736796802b", size = 321817 },
+    { url = "https://files.pythonhosted.org/packages/80/c7/bd718f6329f5c65b936edff532e40614b66032b153ab9ac89e6cb6d2fcff/outlines_core-0.1.26-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8cc8c87d89bd267356f8149c9066cbb98970425ec162997fbf195c3f1feb7009", size = 344102 },
+    { url = "https://files.pythonhosted.org/packages/cb/3e/69ca7394fc89c98c763f2d475c08206fa56f2ffeec210833c1db0195ba51/outlines_core-0.1.26-cp39-cp39-win32.whl", hash = "sha256:9d792a43ed9d8a4e1b38f4d83fe99db442d57aad4404c2edf98b710892eda47e", size = 235081 },
+    { url = "https://files.pythonhosted.org/packages/56/7f/38decd8d63f6b2d513c372186c68337455efba3fb45d0ff722ac268f05d2/outlines_core-0.1.26-cp39-cp39-win_amd64.whl", hash = "sha256:ad8564ecd7b64bcb840596c5049ff1c1a96346de494302ffcc0f2b188c15675e", size = 244007 },
+]
+
+[[package]]
+name = "packaging"
+version = "24.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/d0/63/68dbb6eb2de9cb10ee4c9c14a0148804425e13c4fb20d61cce69f53106da/packaging-24.2.tar.gz", hash = "sha256:c228a6dc5e932d346bc5739379109d49e8853dd8223571c7c5b55260edc0b97f", size = 163950 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/88/ef/eb23f262cca3c0c4eb7ab1933c3b1f03d021f2c48f54763065b6f0e321be/packaging-24.2-py3-none-any.whl", hash = "sha256:09abb1bccd265c01f4a3aa3f7a7db064b36514d2cba19a2f694fe6150451a759", size = 65451 },
+]
+
+[[package]]
+name = "pandas"
+version = "2.2.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy" },
+    { name = "python-dateutil" },
+    { name = "pytz" },
+    { name = "tzdata" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9c/d6/9f8431bacc2e19dca897724cd097b1bb224a6ad5433784a44b587c7c13af/pandas-2.2.3.tar.gz", hash = "sha256:4f18ba62b61d7e192368b84517265a99b4d7ee8912f8708660fb4a366cc82667", size = 4399213 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/aa/70/c853aec59839bceed032d52010ff5f1b8d87dc3114b762e4ba2727661a3b/pandas-2.2.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:1948ddde24197a0f7add2bdc4ca83bf2b1ef84a1bc8ccffd95eda17fd836ecb5", size = 12580827 },
+    { url = "https://files.pythonhosted.org/packages/99/f2/c4527768739ffa4469b2b4fff05aa3768a478aed89a2f271a79a40eee984/pandas-2.2.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:381175499d3802cde0eabbaf6324cce0c4f5d52ca6f8c377c29ad442f50f6348", size = 11303897 },
+    { url = "https://files.pythonhosted.org/packages/ed/12/86c1747ea27989d7a4064f806ce2bae2c6d575b950be087837bdfcabacc9/pandas-2.2.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d9c45366def9a3dd85a6454c0e7908f2b3b8e9c138f5dc38fed7ce720d8453ed", size = 66480908 },
+    { url = "https://files.pythonhosted.org/packages/44/50/7db2cd5e6373ae796f0ddad3675268c8d59fb6076e66f0c339d61cea886b/pandas-2.2.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86976a1c5b25ae3f8ccae3a5306e443569ee3c3faf444dfd0f41cda24667ad57", size = 13064210 },
+    { url = "https://files.pythonhosted.org/packages/61/61/a89015a6d5536cb0d6c3ba02cebed51a95538cf83472975275e28ebf7d0c/pandas-2.2.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:b8661b0238a69d7aafe156b7fa86c44b881387509653fdf857bebc5e4008ad42", size = 16754292 },
+    { url = "https://files.pythonhosted.org/packages/ce/0d/4cc7b69ce37fac07645a94e1d4b0880b15999494372c1523508511b09e40/pandas-2.2.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:37e0aced3e8f539eccf2e099f65cdb9c8aa85109b0be6e93e2baff94264bdc6f", size = 14416379 },
+    { url = "https://files.pythonhosted.org/packages/31/9e/6ebb433de864a6cd45716af52a4d7a8c3c9aaf3a98368e61db9e69e69a9c/pandas-2.2.3-cp310-cp310-win_amd64.whl", hash = "sha256:56534ce0746a58afaf7942ba4863e0ef81c9c50d3f0ae93e9497d6a41a057645", size = 11598471 },
+    { url = "https://files.pythonhosted.org/packages/a8/44/d9502bf0ed197ba9bf1103c9867d5904ddcaf869e52329787fc54ed70cc8/pandas-2.2.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:66108071e1b935240e74525006034333f98bcdb87ea116de573a6a0dccb6c039", size = 12602222 },
+    { url = "https://files.pythonhosted.org/packages/52/11/9eac327a38834f162b8250aab32a6781339c69afe7574368fffe46387edf/pandas-2.2.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7c2875855b0ff77b2a64a0365e24455d9990730d6431b9e0ee18ad8acee13dbd", size = 11321274 },
+    { url = "https://files.pythonhosted.org/packages/45/fb/c4beeb084718598ba19aa9f5abbc8aed8b42f90930da861fcb1acdb54c3a/pandas-2.2.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:cd8d0c3be0515c12fed0bdbae072551c8b54b7192c7b1fda0ba56059a0179698", size = 15579836 },
+    { url = "https://files.pythonhosted.org/packages/cd/5f/4dba1d39bb9c38d574a9a22548c540177f78ea47b32f99c0ff2ec499fac5/pandas-2.2.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c124333816c3a9b03fbeef3a9f230ba9a737e9e5bb4060aa2107a86cc0a497fc", size = 13058505 },
+    { url = "https://files.pythonhosted.org/packages/b9/57/708135b90391995361636634df1f1130d03ba456e95bcf576fada459115a/pandas-2.2.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:63cc132e40a2e084cf01adf0775b15ac515ba905d7dcca47e9a251819c575ef3", size = 16744420 },
+    { url = "https://files.pythonhosted.org/packages/86/4a/03ed6b7ee323cf30404265c284cee9c65c56a212e0a08d9ee06984ba2240/pandas-2.2.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:29401dbfa9ad77319367d36940cd8a0b3a11aba16063e39632d98b0e931ddf32", size = 14440457 },
+    { url = "https://files.pythonhosted.org/packages/ed/8c/87ddf1fcb55d11f9f847e3c69bb1c6f8e46e2f40ab1a2d2abadb2401b007/pandas-2.2.3-cp311-cp311-win_amd64.whl", hash = "sha256:3fc6873a41186404dad67245896a6e440baacc92f5b716ccd1bc9ed2995ab2c5", size = 11617166 },
+    { url = "https://files.pythonhosted.org/packages/17/a3/fb2734118db0af37ea7433f57f722c0a56687e14b14690edff0cdb4b7e58/pandas-2.2.3-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:b1d432e8d08679a40e2a6d8b2f9770a5c21793a6f9f47fdd52c5ce1948a5a8a9", size = 12529893 },
+    { url = "https://files.pythonhosted.org/packages/e1/0c/ad295fd74bfac85358fd579e271cded3ac969de81f62dd0142c426b9da91/pandas-2.2.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a5a1595fe639f5988ba6a8e5bc9649af3baf26df3998a0abe56c02609392e0a4", size = 11363475 },
+    { url = "https://files.pythonhosted.org/packages/c6/2a/4bba3f03f7d07207481fed47f5b35f556c7441acddc368ec43d6643c5777/pandas-2.2.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5de54125a92bb4d1c051c0659e6fcb75256bf799a732a87184e5ea503965bce3", size = 15188645 },
+    { url = "https://files.pythonhosted.org/packages/38/f8/d8fddee9ed0d0c0f4a2132c1dfcf0e3e53265055da8df952a53e7eaf178c/pandas-2.2.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fffb8ae78d8af97f849404f21411c95062db1496aeb3e56f146f0355c9989319", size = 12739445 },
+    { url = "https://files.pythonhosted.org/packages/20/e8/45a05d9c39d2cea61ab175dbe6a2de1d05b679e8de2011da4ee190d7e748/pandas-2.2.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:6dfcb5ee8d4d50c06a51c2fffa6cff6272098ad6540aed1a76d15fb9318194d8", size = 16359235 },
+    { url = "https://files.pythonhosted.org/packages/1d/99/617d07a6a5e429ff90c90da64d428516605a1ec7d7bea494235e1c3882de/pandas-2.2.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:062309c1b9ea12a50e8ce661145c6aab431b1e99530d3cd60640e255778bd43a", size = 14056756 },
+    { url = "https://files.pythonhosted.org/packages/29/d4/1244ab8edf173a10fd601f7e13b9566c1b525c4f365d6bee918e68381889/pandas-2.2.3-cp312-cp312-win_amd64.whl", hash = "sha256:59ef3764d0fe818125a5097d2ae867ca3fa64df032331b7e0917cf5d7bf66b13", size = 11504248 },
+    { url = "https://files.pythonhosted.org/packages/64/22/3b8f4e0ed70644e85cfdcd57454686b9057c6c38d2f74fe4b8bc2527214a/pandas-2.2.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:f00d1345d84d8c86a63e476bb4955e46458b304b9575dcf71102b5c705320015", size = 12477643 },
+    { url = "https://files.pythonhosted.org/packages/e4/93/b3f5d1838500e22c8d793625da672f3eec046b1a99257666c94446969282/pandas-2.2.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:3508d914817e153ad359d7e069d752cdd736a247c322d932eb89e6bc84217f28", size = 11281573 },
+    { url = "https://files.pythonhosted.org/packages/f5/94/6c79b07f0e5aab1dcfa35a75f4817f5c4f677931d4234afcd75f0e6a66ca/pandas-2.2.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:22a9d949bfc9a502d320aa04e5d02feab689d61da4e7764b62c30b991c42c5f0", size = 15196085 },
+    { url = "https://files.pythonhosted.org/packages/e8/31/aa8da88ca0eadbabd0a639788a6da13bb2ff6edbbb9f29aa786450a30a91/pandas-2.2.3-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3a255b2c19987fbbe62a9dfd6cff7ff2aa9ccab3fc75218fd4b7530f01efa24", size = 12711809 },
+    { url = "https://files.pythonhosted.org/packages/ee/7c/c6dbdb0cb2a4344cacfb8de1c5808ca885b2e4dcfde8008266608f9372af/pandas-2.2.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:800250ecdadb6d9c78eae4990da62743b857b470883fa27f652db8bdde7f6659", size = 16356316 },
+    { url = "https://files.pythonhosted.org/packages/57/b7/8b757e7d92023b832869fa8881a992696a0bfe2e26f72c9ae9f255988d42/pandas-2.2.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:6374c452ff3ec675a8f46fd9ab25c4ad0ba590b71cf0656f8b6daa5202bca3fb", size = 14022055 },
+    { url = "https://files.pythonhosted.org/packages/3b/bc/4b18e2b8c002572c5a441a64826252ce5da2aa738855747247a971988043/pandas-2.2.3-cp313-cp313-win_amd64.whl", hash = "sha256:61c5ad4043f791b61dd4752191d9f07f0ae412515d59ba8f005832a532f8736d", size = 11481175 },
+    { url = "https://files.pythonhosted.org/packages/76/a3/a5d88146815e972d40d19247b2c162e88213ef51c7c25993942c39dbf41d/pandas-2.2.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:3b71f27954685ee685317063bf13c7709a7ba74fc996b84fc6821c59b0f06468", size = 12615650 },
+    { url = "https://files.pythonhosted.org/packages/9c/8c/f0fd18f6140ddafc0c24122c8a964e48294acc579d47def376fef12bcb4a/pandas-2.2.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:38cf8125c40dae9d5acc10fa66af8ea6fdf760b2714ee482ca691fc66e6fcb18", size = 11290177 },
+    { url = "https://files.pythonhosted.org/packages/ed/f9/e995754eab9c0f14c6777401f7eece0943840b7a9fc932221c19d1abee9f/pandas-2.2.3-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:ba96630bc17c875161df3818780af30e43be9b166ce51c9a18c1feae342906c2", size = 14651526 },
+    { url = "https://files.pythonhosted.org/packages/25/b0/98d6ae2e1abac4f35230aa756005e8654649d305df9a28b16b9ae4353bff/pandas-2.2.3-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1db71525a1538b30142094edb9adc10be3f3e176748cd7acc2240c2f2e5aa3a4", size = 11871013 },
+    { url = "https://files.pythonhosted.org/packages/cc/57/0f72a10f9db6a4628744c8e8f0df4e6e21de01212c7c981d31e50ffc8328/pandas-2.2.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:15c0e1e02e93116177d29ff83e8b1619c93ddc9c49083f237d4312337a61165d", size = 15711620 },
+    { url = "https://files.pythonhosted.org/packages/ab/5f/b38085618b950b79d2d9164a711c52b10aefc0ae6833b96f626b7021b2ed/pandas-2.2.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:ad5b65698ab28ed8d7f18790a0dc58005c7629f227be9ecc1072aa74c0c1d43a", size = 13098436 },
+    { url = "https://files.pythonhosted.org/packages/ca/8c/8848a4c9b8fdf5a534fe2077af948bf53cd713d77ffbcd7bd15710348fd7/pandas-2.2.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bc6b93f9b966093cb0fd62ff1a7e4c09e6d546ad7c1de191767baffc57628f39", size = 12595535 },
+    { url = "https://files.pythonhosted.org/packages/9c/b9/5cead4f63b6d31bdefeb21a679bc5a7f4aaf262ca7e07e2bc1c341b68470/pandas-2.2.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:5dbca4c1acd72e8eeef4753eeca07de9b1db4f398669d5994086f788a5d7cc30", size = 11319822 },
+    { url = "https://files.pythonhosted.org/packages/31/af/89e35619fb573366fa68dc26dad6ad2c08c17b8004aad6d98f1a31ce4bb3/pandas-2.2.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:8cd6d7cc958a3910f934ea8dbdf17b2364827bb4dafc38ce6eef6bb3d65ff09c", size = 15625439 },
+    { url = "https://files.pythonhosted.org/packages/3d/dd/bed19c2974296661493d7acc4407b1d2db4e2a482197df100f8f965b6225/pandas-2.2.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:99df71520d25fade9db7c1076ac94eb994f4d2673ef2aa2e86ee039b6746d20c", size = 13068928 },
+    { url = "https://files.pythonhosted.org/packages/31/a3/18508e10a31ea108d746c848b5a05c0711e0278fa0d6f1c52a8ec52b80a5/pandas-2.2.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:31d0ced62d4ea3e231a9f228366919a5ea0b07440d9d4dac345376fd8e1477ea", size = 16783266 },
+    { url = "https://files.pythonhosted.org/packages/c4/a5/3429bd13d82bebc78f4d78c3945efedef63a7cd0c15c17b2eeb838d1121f/pandas-2.2.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:7eee9e7cea6adf3e3d24e304ac6b8300646e2a5d1cd3a3c2abed9101b0846761", size = 14450871 },
+    { url = "https://files.pythonhosted.org/packages/2f/49/5c30646e96c684570925b772eac4eb0a8cb0ca590fa978f56c5d3ae73ea1/pandas-2.2.3-cp39-cp39-win_amd64.whl", hash = "sha256:4850ba03528b6dd51d6c5d273c46f183f39a9baf3f0143e566b89450965b105e", size = 11618011 },
+]
+
+[[package]]
+name = "pillow"
+version = "11.1.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/af/c097e544e7bd278333db77933e535098c259609c4eb3b85381109602fb5b/pillow-11.1.0.tar.gz", hash = "sha256:368da70808b36d73b4b390a8ffac11069f8a5c85f29eff1f1b01bcf3ef5b2a20", size = 46742715 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/50/1c/2dcea34ac3d7bc96a1fd1bd0a6e06a57c67167fec2cff8d95d88229a8817/pillow-11.1.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:e1abe69aca89514737465752b4bcaf8016de61b3be1397a8fc260ba33321b3a8", size = 3229983 },
+    { url = "https://files.pythonhosted.org/packages/14/ca/6bec3df25e4c88432681de94a3531cc738bd85dea6c7aa6ab6f81ad8bd11/pillow-11.1.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c640e5a06869c75994624551f45e5506e4256562ead981cce820d5ab39ae2192", size = 3101831 },
+    { url = "https://files.pythonhosted.org/packages/d4/2c/668e18e5521e46eb9667b09e501d8e07049eb5bfe39d56be0724a43117e6/pillow-11.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a07dba04c5e22824816b2615ad7a7484432d7f540e6fa86af60d2de57b0fcee2", size = 4314074 },
+    { url = "https://files.pythonhosted.org/packages/02/80/79f99b714f0fc25f6a8499ecfd1f810df12aec170ea1e32a4f75746051ce/pillow-11.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e267b0ed063341f3e60acd25c05200df4193e15a4a5807075cd71225a2386e26", size = 4394933 },
+    { url = "https://files.pythonhosted.org/packages/81/aa/8d4ad25dc11fd10a2001d5b8a80fdc0e564ac33b293bdfe04ed387e0fd95/pillow-11.1.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:bd165131fd51697e22421d0e467997ad31621b74bfc0b75956608cb2906dda07", size = 4353349 },
+    { url = "https://files.pythonhosted.org/packages/84/7a/cd0c3eaf4a28cb2a74bdd19129f7726277a7f30c4f8424cd27a62987d864/pillow-11.1.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:abc56501c3fd148d60659aae0af6ddc149660469082859fa7b066a298bde9482", size = 4476532 },
+    { url = "https://files.pythonhosted.org/packages/8f/8b/a907fdd3ae8f01c7670dfb1499c53c28e217c338b47a813af8d815e7ce97/pillow-11.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:54ce1c9a16a9561b6d6d8cb30089ab1e5eb66918cb47d457bd996ef34182922e", size = 4279789 },
+    { url = "https://files.pythonhosted.org/packages/6f/9a/9f139d9e8cccd661c3efbf6898967a9a337eb2e9be2b454ba0a09533100d/pillow-11.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:73ddde795ee9b06257dac5ad42fcb07f3b9b813f8c1f7f870f402f4dc54b5269", size = 4413131 },
+    { url = "https://files.pythonhosted.org/packages/a8/68/0d8d461f42a3f37432203c8e6df94da10ac8081b6d35af1c203bf3111088/pillow-11.1.0-cp310-cp310-win32.whl", hash = "sha256:3a5fe20a7b66e8135d7fd617b13272626a28278d0e578c98720d9ba4b2439d49", size = 2291213 },
+    { url = "https://files.pythonhosted.org/packages/14/81/d0dff759a74ba87715509af9f6cb21fa21d93b02b3316ed43bda83664db9/pillow-11.1.0-cp310-cp310-win_amd64.whl", hash = "sha256:b6123aa4a59d75f06e9dd3dac5bf8bc9aa383121bb3dd9a7a612e05eabc9961a", size = 2625725 },
+    { url = "https://files.pythonhosted.org/packages/ce/1f/8d50c096a1d58ef0584ddc37e6f602828515219e9d2428e14ce50f5ecad1/pillow-11.1.0-cp310-cp310-win_arm64.whl", hash = "sha256:a76da0a31da6fcae4210aa94fd779c65c75786bc9af06289cd1c184451ef7a65", size = 2375213 },
+    { url = "https://files.pythonhosted.org/packages/dd/d6/2000bfd8d5414fb70cbbe52c8332f2283ff30ed66a9cde42716c8ecbe22c/pillow-11.1.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:e06695e0326d05b06833b40b7ef477e475d0b1ba3a6d27da1bb48c23209bf457", size = 3229968 },
+    { url = "https://files.pythonhosted.org/packages/d9/45/3fe487010dd9ce0a06adf9b8ff4f273cc0a44536e234b0fad3532a42c15b/pillow-11.1.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:96f82000e12f23e4f29346e42702b6ed9a2f2fea34a740dd5ffffcc8c539eb35", size = 3101806 },
+    { url = "https://files.pythonhosted.org/packages/e3/72/776b3629c47d9d5f1c160113158a7a7ad177688d3a1159cd3b62ded5a33a/pillow-11.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a3cd561ded2cf2bbae44d4605837221b987c216cff94f49dfeed63488bb228d2", size = 4322283 },
+    { url = "https://files.pythonhosted.org/packages/e4/c2/e25199e7e4e71d64eeb869f5b72c7ddec70e0a87926398785ab944d92375/pillow-11.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f189805c8be5ca5add39e6f899e6ce2ed824e65fb45f3c28cb2841911da19070", size = 4402945 },
+    { url = "https://files.pythonhosted.org/packages/c1/ed/51d6136c9d5911f78632b1b86c45241c712c5a80ed7fa7f9120a5dff1eba/pillow-11.1.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:dd0052e9db3474df30433f83a71b9b23bd9e4ef1de13d92df21a52c0303b8ab6", size = 4361228 },
+    { url = "https://files.pythonhosted.org/packages/48/a4/fbfe9d5581d7b111b28f1d8c2762dee92e9821bb209af9fa83c940e507a0/pillow-11.1.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:837060a8599b8f5d402e97197d4924f05a2e0d68756998345c829c33186217b1", size = 4484021 },
+    { url = "https://files.pythonhosted.org/packages/39/db/0b3c1a5018117f3c1d4df671fb8e47d08937f27519e8614bbe86153b65a5/pillow-11.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:aa8dd43daa836b9a8128dbe7d923423e5ad86f50a7a14dc688194b7be5c0dea2", size = 4287449 },
+    { url = "https://files.pythonhosted.org/packages/d9/58/bc128da7fea8c89fc85e09f773c4901e95b5936000e6f303222490c052f3/pillow-11.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0a2f91f8a8b367e7a57c6e91cd25af510168091fb89ec5146003e424e1558a96", size = 4419972 },
+    { url = "https://files.pythonhosted.org/packages/5f/bb/58f34379bde9fe197f51841c5bbe8830c28bbb6d3801f16a83b8f2ad37df/pillow-11.1.0-cp311-cp311-win32.whl", hash = "sha256:c12fc111ef090845de2bb15009372175d76ac99969bdf31e2ce9b42e4b8cd88f", size = 2291201 },
+    { url = "https://files.pythonhosted.org/packages/3a/c6/fce9255272bcf0c39e15abd2f8fd8429a954cf344469eaceb9d0d1366913/pillow-11.1.0-cp311-cp311-win_amd64.whl", hash = "sha256:fbd43429d0d7ed6533b25fc993861b8fd512c42d04514a0dd6337fb3ccf22761", size = 2625686 },
+    { url = "https://files.pythonhosted.org/packages/c8/52/8ba066d569d932365509054859f74f2a9abee273edcef5cd75e4bc3e831e/pillow-11.1.0-cp311-cp311-win_arm64.whl", hash = "sha256:f7955ecf5609dee9442cbface754f2c6e541d9e6eda87fad7f7a989b0bdb9d71", size = 2375194 },
+    { url = "https://files.pythonhosted.org/packages/95/20/9ce6ed62c91c073fcaa23d216e68289e19d95fb8188b9fb7a63d36771db8/pillow-11.1.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:2062ffb1d36544d42fcaa277b069c88b01bb7298f4efa06731a7fd6cc290b81a", size = 3226818 },
+    { url = "https://files.pythonhosted.org/packages/b9/d8/f6004d98579a2596c098d1e30d10b248798cceff82d2b77aa914875bfea1/pillow-11.1.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a85b653980faad27e88b141348707ceeef8a1186f75ecc600c395dcac19f385b", size = 3101662 },
+    { url = "https://files.pythonhosted.org/packages/08/d9/892e705f90051c7a2574d9f24579c9e100c828700d78a63239676f960b74/pillow-11.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9409c080586d1f683df3f184f20e36fb647f2e0bc3988094d4fd8c9f4eb1b3b3", size = 4329317 },
+    { url = "https://files.pythonhosted.org/packages/8c/aa/7f29711f26680eab0bcd3ecdd6d23ed6bce180d82e3f6380fb7ae35fcf3b/pillow-11.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7fdadc077553621911f27ce206ffcbec7d3f8d7b50e0da39f10997e8e2bb7f6a", size = 4412999 },
+    { url = "https://files.pythonhosted.org/packages/c8/c4/8f0fe3b9e0f7196f6d0bbb151f9fba323d72a41da068610c4c960b16632a/pillow-11.1.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:93a18841d09bcdd774dcdc308e4537e1f867b3dec059c131fde0327899734aa1", size = 4368819 },
+    { url = "https://files.pythonhosted.org/packages/38/0d/84200ed6a871ce386ddc82904bfadc0c6b28b0c0ec78176871a4679e40b3/pillow-11.1.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:9aa9aeddeed452b2f616ff5507459e7bab436916ccb10961c4a382cd3e03f47f", size = 4496081 },
+    { url = "https://files.pythonhosted.org/packages/84/9c/9bcd66f714d7e25b64118e3952d52841a4babc6d97b6d28e2261c52045d4/pillow-11.1.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3cdcdb0b896e981678eee140d882b70092dac83ac1cdf6b3a60e2216a73f2b91", size = 4296513 },
+    { url = "https://files.pythonhosted.org/packages/db/61/ada2a226e22da011b45f7104c95ebda1b63dcbb0c378ad0f7c2a710f8fd2/pillow-11.1.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:36ba10b9cb413e7c7dfa3e189aba252deee0602c86c309799da5a74009ac7a1c", size = 4431298 },
+    { url = "https://files.pythonhosted.org/packages/e7/c4/fc6e86750523f367923522014b821c11ebc5ad402e659d8c9d09b3c9d70c/pillow-11.1.0-cp312-cp312-win32.whl", hash = "sha256:cfd5cd998c2e36a862d0e27b2df63237e67273f2fc78f47445b14e73a810e7e6", size = 2291630 },
+    { url = "https://files.pythonhosted.org/packages/08/5c/2104299949b9d504baf3f4d35f73dbd14ef31bbd1ddc2c1b66a5b7dfda44/pillow-11.1.0-cp312-cp312-win_amd64.whl", hash = "sha256:a697cd8ba0383bba3d2d3ada02b34ed268cb548b369943cd349007730c92bddf", size = 2626369 },
+    { url = "https://files.pythonhosted.org/packages/37/f3/9b18362206b244167c958984b57c7f70a0289bfb59a530dd8af5f699b910/pillow-11.1.0-cp312-cp312-win_arm64.whl", hash = "sha256:4dd43a78897793f60766563969442020e90eb7847463eca901e41ba186a7d4a5", size = 2375240 },
+    { url = "https://files.pythonhosted.org/packages/b3/31/9ca79cafdce364fd5c980cd3416c20ce1bebd235b470d262f9d24d810184/pillow-11.1.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ae98e14432d458fc3de11a77ccb3ae65ddce70f730e7c76140653048c71bfcbc", size = 3226640 },
+    { url = "https://files.pythonhosted.org/packages/ac/0f/ff07ad45a1f172a497aa393b13a9d81a32e1477ef0e869d030e3c1532521/pillow-11.1.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:cc1331b6d5a6e144aeb5e626f4375f5b7ae9934ba620c0ac6b3e43d5e683a0f0", size = 3101437 },
+    { url = "https://files.pythonhosted.org/packages/08/2f/9906fca87a68d29ec4530be1f893149e0cb64a86d1f9f70a7cfcdfe8ae44/pillow-11.1.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:758e9d4ef15d3560214cddbc97b8ef3ef86ce04d62ddac17ad39ba87e89bd3b1", size = 4326605 },
+    { url = "https://files.pythonhosted.org/packages/b0/0f/f3547ee15b145bc5c8b336401b2d4c9d9da67da9dcb572d7c0d4103d2c69/pillow-11.1.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b523466b1a31d0dcef7c5be1f20b942919b62fd6e9a9be199d035509cbefc0ec", size = 4411173 },
+    { url = "https://files.pythonhosted.org/packages/b1/df/bf8176aa5db515c5de584c5e00df9bab0713548fd780c82a86cba2c2fedb/pillow-11.1.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:9044b5e4f7083f209c4e35aa5dd54b1dd5b112b108648f5c902ad586d4f945c5", size = 4369145 },
+    { url = "https://files.pythonhosted.org/packages/de/7c/7433122d1cfadc740f577cb55526fdc39129a648ac65ce64db2eb7209277/pillow-11.1.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:3764d53e09cdedd91bee65c2527815d315c6b90d7b8b79759cc48d7bf5d4f114", size = 4496340 },
+    { url = "https://files.pythonhosted.org/packages/25/46/dd94b93ca6bd555588835f2504bd90c00d5438fe131cf01cfa0c5131a19d/pillow-11.1.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:31eba6bbdd27dde97b0174ddf0297d7a9c3a507a8a1480e1e60ef914fe23d352", size = 4296906 },
+    { url = "https://files.pythonhosted.org/packages/a8/28/2f9d32014dfc7753e586db9add35b8a41b7a3b46540e965cb6d6bc607bd2/pillow-11.1.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:b5d658fbd9f0d6eea113aea286b21d3cd4d3fd978157cbf2447a6035916506d3", size = 4431759 },
+    { url = "https://files.pythonhosted.org/packages/33/48/19c2cbe7403870fbe8b7737d19eb013f46299cdfe4501573367f6396c775/pillow-11.1.0-cp313-cp313-win32.whl", hash = "sha256:f86d3a7a9af5d826744fabf4afd15b9dfef44fe69a98541f666f66fbb8d3fef9", size = 2291657 },
+    { url = "https://files.pythonhosted.org/packages/3b/ad/285c556747d34c399f332ba7c1a595ba245796ef3e22eae190f5364bb62b/pillow-11.1.0-cp313-cp313-win_amd64.whl", hash = "sha256:593c5fd6be85da83656b93ffcccc2312d2d149d251e98588b14fbc288fd8909c", size = 2626304 },
+    { url = "https://files.pythonhosted.org/packages/e5/7b/ef35a71163bf36db06e9c8729608f78dedf032fc8313d19bd4be5c2588f3/pillow-11.1.0-cp313-cp313-win_arm64.whl", hash = "sha256:11633d58b6ee5733bde153a8dafd25e505ea3d32e261accd388827ee987baf65", size = 2375117 },
+    { url = "https://files.pythonhosted.org/packages/79/30/77f54228401e84d6791354888549b45824ab0ffde659bafa67956303a09f/pillow-11.1.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:70ca5ef3b3b1c4a0812b5c63c57c23b63e53bc38e758b37a951e5bc466449861", size = 3230060 },
+    { url = "https://files.pythonhosted.org/packages/ce/b1/56723b74b07dd64c1010fee011951ea9c35a43d8020acd03111f14298225/pillow-11.1.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:8000376f139d4d38d6851eb149b321a52bb8893a88dae8ee7d95840431977081", size = 3106192 },
+    { url = "https://files.pythonhosted.org/packages/e1/cd/7bf7180e08f80a4dcc6b4c3a0aa9e0b0ae57168562726a05dc8aa8fa66b0/pillow-11.1.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9ee85f0696a17dd28fbcfceb59f9510aa71934b483d1f5601d1030c3c8304f3c", size = 4446805 },
+    { url = "https://files.pythonhosted.org/packages/97/42/87c856ea30c8ed97e8efbe672b58c8304dee0573f8c7cab62ae9e31db6ae/pillow-11.1.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:dd0e081319328928531df7a0e63621caf67652c8464303fd102141b785ef9547", size = 4530623 },
+    { url = "https://files.pythonhosted.org/packages/ff/41/026879e90c84a88e33fb00cc6bd915ac2743c67e87a18f80270dfe3c2041/pillow-11.1.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:e63e4e5081de46517099dc30abe418122f54531a6ae2ebc8680bcd7096860eab", size = 4465191 },
+    { url = "https://files.pythonhosted.org/packages/e5/fb/a7960e838bc5df57a2ce23183bfd2290d97c33028b96bde332a9057834d3/pillow-11.1.0-cp313-cp313t-win32.whl", hash = "sha256:dda60aa465b861324e65a78c9f5cf0f4bc713e4309f83bc387be158b077963d9", size = 2295494 },
+    { url = "https://files.pythonhosted.org/packages/d7/6c/6ec83ee2f6f0fda8d4cf89045c6be4b0373ebfc363ba8538f8c999f63fcd/pillow-11.1.0-cp313-cp313t-win_amd64.whl", hash = "sha256:ad5db5781c774ab9a9b2c4302bbf0c1014960a0a7be63278d13ae6fdf88126fe", size = 2631595 },
+    { url = "https://files.pythonhosted.org/packages/cf/6c/41c21c6c8af92b9fea313aa47c75de49e2f9a467964ee33eb0135d47eb64/pillow-11.1.0-cp313-cp313t-win_arm64.whl", hash = "sha256:67cd427c68926108778a9005f2a04adbd5e67c442ed21d95389fe1d595458756", size = 2377651 },
+    { url = "https://files.pythonhosted.org/packages/9a/1f/9df5ac77491fddd2e36c352d16976dc11fbe6ab842f5df85fd7e31b847b9/pillow-11.1.0-cp39-cp39-macosx_10_10_x86_64.whl", hash = "sha256:bf902d7413c82a1bfa08b06a070876132a5ae6b2388e2712aab3a7cbc02205c6", size = 3229995 },
+    { url = "https://files.pythonhosted.org/packages/a6/62/c7b359e924dca274173b04922ac06aa63614f7e934d132f2fe1d852509aa/pillow-11.1.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:c1eec9d950b6fe688edee07138993e54ee4ae634c51443cfb7c1e7613322718e", size = 3101890 },
+    { url = "https://files.pythonhosted.org/packages/7b/63/136f21340a434de895b62bcf2c386005a8aa24066c4facd619f5e0e9f283/pillow-11.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e275ee4cb11c262bd108ab2081f750db2a1c0b8c12c1897f27b160c8bd57bbc", size = 4310366 },
+    { url = "https://files.pythonhosted.org/packages/f6/46/0bd0ca03d9d1164a7fa33d285ef6d1c438e963d0c8770e4c5b3737ef5abe/pillow-11.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4db853948ce4e718f2fc775b75c37ba2efb6aaea41a1a5fc57f0af59eee774b2", size = 4391582 },
+    { url = "https://files.pythonhosted.org/packages/0c/55/f182db572b28bd833b8e806f933f782ceb2df64c40e4d8bd3d4226a46eca/pillow-11.1.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:ab8a209b8485d3db694fa97a896d96dd6533d63c22829043fd9de627060beade", size = 4350278 },
+    { url = "https://files.pythonhosted.org/packages/75/fb/e330fdbbcbc4744214b5f53b84d9d8a9f4ffbebc2e9c2ac10475386e3296/pillow-11.1.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:54251ef02a2309b5eec99d151ebf5c9904b77976c8abdcbce7891ed22df53884", size = 4471768 },
+    { url = "https://files.pythonhosted.org/packages/eb/51/20ee6c4da4448d7a67ffb720a5fcdb965115a78e211a1f58f9845ae15f86/pillow-11.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:5bb94705aea800051a743aa4874bb1397d4695fb0583ba5e425ee0328757f196", size = 4276549 },
+    { url = "https://files.pythonhosted.org/packages/37/f2/a25c0bdaa6d6fd5cc3d4a6f65b5a7ea46e7af58bee00a98efe0a5af79c58/pillow-11.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:89dbdb3e6e9594d512780a5a1c42801879628b38e3efc7038094430844e271d8", size = 4409350 },
+    { url = "https://files.pythonhosted.org/packages/12/a7/06687947604cd3e47abeea1b78b65d34ffce7feab03cfe0dd985f115dca3/pillow-11.1.0-cp39-cp39-win32.whl", hash = "sha256:e5449ca63da169a2e6068dd0e2fcc8d91f9558aba89ff6d02121ca8ab11e79e5", size = 2291271 },
+    { url = "https://files.pythonhosted.org/packages/21/a6/f51d47675940b5c63b08ff0575b3518428b4acb891f88526fa4ee1edab6f/pillow-11.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:3362c6ca227e65c54bf71a5f88b3d4565ff1bcbc63ae72c34b07bbb1cc59a43f", size = 2625783 },
+    { url = "https://files.pythonhosted.org/packages/95/56/97750bd33e68648fa432dfadcb8ede7624bd905822d42262d34bcebdd9d7/pillow-11.1.0-cp39-cp39-win_arm64.whl", hash = "sha256:b20be51b37a75cc54c2c55def3fa2c65bb94ba859dde241cd0a4fd302de5ae0a", size = 2375193 },
+    { url = "https://files.pythonhosted.org/packages/fa/c5/389961578fb677b8b3244fcd934f720ed25a148b9a5cc81c91bdf59d8588/pillow-11.1.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:8c730dc3a83e5ac137fbc92dfcfe1511ce3b2b5d7578315b63dbbb76f7f51d90", size = 3198345 },
+    { url = "https://files.pythonhosted.org/packages/c4/fa/803c0e50ffee74d4b965229e816af55276eac1d5806712de86f9371858fd/pillow-11.1.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:7d33d2fae0e8b170b6a6c57400e077412240f6f5bb2a342cf1ee512a787942bb", size = 3072938 },
+    { url = "https://files.pythonhosted.org/packages/dc/67/2a3a5f8012b5d8c63fe53958ba906c1b1d0482ebed5618057ef4d22f8076/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a8d65b38173085f24bc07f8b6c505cbb7418009fa1a1fcb111b1f4961814a442", size = 3400049 },
+    { url = "https://files.pythonhosted.org/packages/e5/a0/514f0d317446c98c478d1872497eb92e7cde67003fed74f696441e647446/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:015c6e863faa4779251436db398ae75051469f7c903b043a48f078e437656f83", size = 3422431 },
+    { url = "https://files.pythonhosted.org/packages/cd/00/20f40a935514037b7d3f87adfc87d2c538430ea625b63b3af8c3f5578e72/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:d44ff19eea13ae4acdaaab0179fa68c0c6f2f45d66a4d8ec1eda7d6cecbcc15f", size = 3446208 },
+    { url = "https://files.pythonhosted.org/packages/28/3c/7de681727963043e093c72e6c3348411b0185eab3263100d4490234ba2f6/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:d3d8da4a631471dfaf94c10c85f5277b1f8e42ac42bade1ac67da4b4a7359b73", size = 3509746 },
+    { url = "https://files.pythonhosted.org/packages/41/67/936f9814bdd74b2dfd4822f1f7725ab5d8ff4103919a1664eb4874c58b2f/pillow-11.1.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:4637b88343166249fe8aa94e7c4a62a180c4b3898283bb5d3d2fd5fe10d8e4e0", size = 2626353 },
+]
+
+[[package]]
+name = "platformdirs"
+version = "4.3.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/13/fc/128cc9cb8f03208bdbf93d3aa862e16d376844a14f9a0ce5cf4507372de4/platformdirs-4.3.6.tar.gz", hash = "sha256:357fb2acbc885b0419afd3ce3ed34564c13c9b95c89360cd9563f73aa5e2b907", size = 21302 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3c/a6/bc1012356d8ece4d66dd75c4b9fc6c1f6650ddd5991e421177d9f8f671be/platformdirs-4.3.6-py3-none-any.whl", hash = "sha256:73e575e1408ab8103900836b97580d5307456908a03e92031bab39e4554cc3fb", size = 18439 },
+]
+
+[[package]]
+name = "pluggy"
+version = "1.5.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/96/2d/02d4312c973c6050a18b314a5ad0b3210edb65a906f868e31c111dede4a6/pluggy-1.5.0.tar.gz", hash = "sha256:2cffa88e94fdc978c4c574f15f9e59b7f4201d439195c3715ca9e2486f1d0cf1", size = 67955 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/88/5f/e351af9a41f866ac3f1fac4ca0613908d9a41741cfcf2228f4ad853b697d/pluggy-1.5.0-py3-none-any.whl", hash = "sha256:44e1ad92c8ca002de6377e165f3e0f1be63266ab4d554740532335b9d75ea669", size = 20556 },
+]
+
+[[package]]
+name = "pre-commit"
+version = "4.0.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cfgv" },
+    { name = "identify" },
+    { name = "nodeenv" },
+    { name = "pyyaml" },
+    { name = "virtualenv" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/2e/c8/e22c292035f1bac8b9f5237a2622305bc0304e776080b246f3df57c4ff9f/pre_commit-4.0.1.tar.gz", hash = "sha256:80905ac375958c0444c65e9cebebd948b3cdb518f335a091a670a89d652139d2", size = 191678 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/16/8f/496e10d51edd6671ebe0432e33ff800aa86775d2d147ce7d43389324a525/pre_commit-4.0.1-py2.py3-none-any.whl", hash = "sha256:efde913840816312445dc98787724647c65473daefe420785f885e8ed9a06878", size = 218713 },
+]
+
+[[package]]
+name = "prometheus-client"
+version = "0.21.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/62/14/7d0f567991f3a9af8d1cd4f619040c93b68f09a02b6d0b6ab1b2d1ded5fe/prometheus_client-0.21.1.tar.gz", hash = "sha256:252505a722ac04b0456be05c05f75f45d760c2911ffc45f2a06bcaed9f3ae3fb", size = 78551 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ff/c2/ab7d37426c179ceb9aeb109a85cda8948bb269b7561a0be870cc656eefe4/prometheus_client-0.21.1-py3-none-any.whl", hash = "sha256:594b45c410d6f4f8888940fe80b5cc2521b305a1fafe1c58609ef715a001f301", size = 54682 },
+]
+
+[[package]]
+name = "prometheus-fastapi-instrumentator"
+version = "7.0.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "prometheus-client" },
+    { name = "starlette" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/c4/01/e3ccb464ba9bf6c001ad85652e6fc7e63098c2685275a682cabc8e7871b8/prometheus_fastapi_instrumentator-7.0.2.tar.gz", hash = "sha256:8a4d8fb13dbe19d2882ac6af9ce236e4e1f98dc48e3fa44fe88d8e23ac3c953f", size = 19844 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f1/f7/a67e804853d05b3f1f5def0ebd662b9f48dbcfe0b452f0aa5d4c183a6f86/prometheus_fastapi_instrumentator-7.0.2-py3-none-any.whl", hash = "sha256:975e39992acb7a112758ff13ba95317e6c54d1bbf605f9156f31ac9f2800c32d", size = 18998 },
+]
+
+[[package]]
+name = "propcache"
+version = "0.2.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/20/c8/2a13f78d82211490855b2fb303b6721348d0787fdd9a12ac46d99d3acde1/propcache-0.2.1.tar.gz", hash = "sha256:3f77ce728b19cb537714499928fe800c3dda29e8d9428778fc7c186da4c09a64", size = 41735 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a7/a5/0ea64c9426959ef145a938e38c832fc551843481d356713ececa9a8a64e8/propcache-0.2.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:6b3f39a85d671436ee3d12c017f8fdea38509e4f25b28eb25877293c98c243f6", size = 79296 },
+    { url = "https://files.pythonhosted.org/packages/76/5a/916db1aba735f55e5eca4733eea4d1973845cf77dfe67c2381a2ca3ce52d/propcache-0.2.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:39d51fbe4285d5db5d92a929e3e21536ea3dd43732c5b177c7ef03f918dff9f2", size = 45622 },
+    { url = "https://files.pythonhosted.org/packages/2d/62/685d3cf268b8401ec12b250b925b21d152b9d193b7bffa5fdc4815c392c2/propcache-0.2.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6445804cf4ec763dc70de65a3b0d9954e868609e83850a47ca4f0cb64bd79fea", size = 45133 },
+    { url = "https://files.pythonhosted.org/packages/4d/3d/31c9c29ee7192defc05aa4d01624fd85a41cf98e5922aaed206017329944/propcache-0.2.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f9479aa06a793c5aeba49ce5c5692ffb51fcd9a7016e017d555d5e2b0045d212", size = 204809 },
+    { url = "https://files.pythonhosted.org/packages/10/a1/e4050776f4797fc86140ac9a480d5dc069fbfa9d499fe5c5d2fa1ae71f07/propcache-0.2.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d9631c5e8b5b3a0fda99cb0d29c18133bca1e18aea9effe55adb3da1adef80d3", size = 219109 },
+    { url = "https://files.pythonhosted.org/packages/c9/c0/e7ae0df76343d5e107d81e59acc085cea5fd36a48aa53ef09add7503e888/propcache-0.2.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3156628250f46a0895f1f36e1d4fbe062a1af8718ec3ebeb746f1d23f0c5dc4d", size = 217368 },
+    { url = "https://files.pythonhosted.org/packages/fc/e1/e0a2ed6394b5772508868a977d3238f4afb2eebaf9976f0b44a8d347ad63/propcache-0.2.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b6fb63ae352e13748289f04f37868099e69dba4c2b3e271c46061e82c745634", size = 205124 },
+    { url = "https://files.pythonhosted.org/packages/50/c1/e388c232d15ca10f233c778bbdc1034ba53ede14c207a72008de45b2db2e/propcache-0.2.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:887d9b0a65404929641a9fabb6452b07fe4572b269d901d622d8a34a4e9043b2", size = 195463 },
+    { url = "https://files.pythonhosted.org/packages/0a/fd/71b349b9def426cc73813dbd0f33e266de77305e337c8c12bfb0a2a82bfb/propcache-0.2.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:a96dc1fa45bd8c407a0af03b2d5218392729e1822b0c32e62c5bf7eeb5fb3958", size = 198358 },
+    { url = "https://files.pythonhosted.org/packages/02/f2/d7c497cd148ebfc5b0ae32808e6c1af5922215fe38c7a06e4e722fe937c8/propcache-0.2.1-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:a7e65eb5c003a303b94aa2c3852ef130230ec79e349632d030e9571b87c4698c", size = 195560 },
+    { url = "https://files.pythonhosted.org/packages/bb/57/f37041bbe5e0dfed80a3f6be2612a3a75b9cfe2652abf2c99bef3455bbad/propcache-0.2.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:999779addc413181912e984b942fbcc951be1f5b3663cd80b2687758f434c583", size = 196895 },
+    { url = "https://files.pythonhosted.org/packages/83/36/ae3cc3e4f310bff2f064e3d2ed5558935cc7778d6f827dce74dcfa125304/propcache-0.2.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:19a0f89a7bb9d8048d9c4370c9c543c396e894c76be5525f5e1ad287f1750ddf", size = 207124 },
+    { url = "https://files.pythonhosted.org/packages/8c/c4/811b9f311f10ce9d31a32ff14ce58500458443627e4df4ae9c264defba7f/propcache-0.2.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:1ac2f5fe02fa75f56e1ad473f1175e11f475606ec9bd0be2e78e4734ad575034", size = 210442 },
+    { url = "https://files.pythonhosted.org/packages/18/dd/a1670d483a61ecac0d7fc4305d91caaac7a8fc1b200ea3965a01cf03bced/propcache-0.2.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:574faa3b79e8ebac7cb1d7930f51184ba1ccf69adfdec53a12f319a06030a68b", size = 203219 },
+    { url = "https://files.pythonhosted.org/packages/f9/2d/30ced5afde41b099b2dc0c6573b66b45d16d73090e85655f1a30c5a24e07/propcache-0.2.1-cp310-cp310-win32.whl", hash = "sha256:03ff9d3f665769b2a85e6157ac8b439644f2d7fd17615a82fa55739bc97863f4", size = 40313 },
+    { url = "https://files.pythonhosted.org/packages/23/84/bd9b207ac80da237af77aa6e153b08ffa83264b1c7882495984fcbfcf85c/propcache-0.2.1-cp310-cp310-win_amd64.whl", hash = "sha256:2d3af2e79991102678f53e0dbf4c35de99b6b8b58f29a27ca0325816364caaba", size = 44428 },
+    { url = "https://files.pythonhosted.org/packages/bc/0f/2913b6791ebefb2b25b4efd4bb2299c985e09786b9f5b19184a88e5778dd/propcache-0.2.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:1ffc3cca89bb438fb9c95c13fc874012f7b9466b89328c3c8b1aa93cdcfadd16", size = 79297 },
+    { url = "https://files.pythonhosted.org/packages/cf/73/af2053aeccd40b05d6e19058419ac77674daecdd32478088b79375b9ab54/propcache-0.2.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:f174bbd484294ed9fdf09437f889f95807e5f229d5d93588d34e92106fbf6717", size = 45611 },
+    { url = "https://files.pythonhosted.org/packages/3c/09/8386115ba7775ea3b9537730e8cf718d83bbf95bffe30757ccf37ec4e5da/propcache-0.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:70693319e0b8fd35dd863e3e29513875eb15c51945bf32519ef52927ca883bc3", size = 45146 },
+    { url = "https://files.pythonhosted.org/packages/03/7a/793aa12f0537b2e520bf09f4c6833706b63170a211ad042ca71cbf79d9cb/propcache-0.2.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b480c6a4e1138e1aa137c0079b9b6305ec6dcc1098a8ca5196283e8a49df95a9", size = 232136 },
+    { url = "https://files.pythonhosted.org/packages/f1/38/b921b3168d72111769f648314100558c2ea1d52eb3d1ba7ea5c4aa6f9848/propcache-0.2.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d27b84d5880f6d8aa9ae3edb253c59d9f6642ffbb2c889b78b60361eed449787", size = 239706 },
+    { url = "https://files.pythonhosted.org/packages/14/29/4636f500c69b5edea7786db3c34eb6166f3384b905665ce312a6e42c720c/propcache-0.2.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:857112b22acd417c40fa4595db2fe28ab900c8c5fe4670c7989b1c0230955465", size = 238531 },
+    { url = "https://files.pythonhosted.org/packages/85/14/01fe53580a8e1734ebb704a3482b7829a0ef4ea68d356141cf0994d9659b/propcache-0.2.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cf6c4150f8c0e32d241436526f3c3f9cbd34429492abddbada2ffcff506c51af", size = 231063 },
+    { url = "https://files.pythonhosted.org/packages/33/5c/1d961299f3c3b8438301ccfbff0143b69afcc30c05fa28673cface692305/propcache-0.2.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:66d4cfda1d8ed687daa4bc0274fcfd5267873db9a5bc0418c2da19273040eeb7", size = 220134 },
+    { url = "https://files.pythonhosted.org/packages/00/d0/ed735e76db279ba67a7d3b45ba4c654e7b02bc2f8050671ec365d8665e21/propcache-0.2.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:c2f992c07c0fca81655066705beae35fc95a2fa7366467366db627d9f2ee097f", size = 220009 },
+    { url = "https://files.pythonhosted.org/packages/75/90/ee8fab7304ad6533872fee982cfff5a53b63d095d78140827d93de22e2d4/propcache-0.2.1-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:4a571d97dbe66ef38e472703067021b1467025ec85707d57e78711c085984e54", size = 212199 },
+    { url = "https://files.pythonhosted.org/packages/eb/ec/977ffaf1664f82e90737275873461695d4c9407d52abc2f3c3e24716da13/propcache-0.2.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:bb6178c241278d5fe853b3de743087be7f5f4c6f7d6d22a3b524d323eecec505", size = 214827 },
+    { url = "https://files.pythonhosted.org/packages/57/48/031fb87ab6081764054821a71b71942161619549396224cbb242922525e8/propcache-0.2.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:ad1af54a62ffe39cf34db1aa6ed1a1873bd548f6401db39d8e7cd060b9211f82", size = 228009 },
+    { url = "https://files.pythonhosted.org/packages/1a/06/ef1390f2524850838f2390421b23a8b298f6ce3396a7cc6d39dedd4047b0/propcache-0.2.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:e7048abd75fe40712005bcfc06bb44b9dfcd8e101dda2ecf2f5aa46115ad07ca", size = 231638 },
+    { url = "https://files.pythonhosted.org/packages/38/2a/101e6386d5a93358395da1d41642b79c1ee0f3b12e31727932b069282b1d/propcache-0.2.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:160291c60081f23ee43d44b08a7e5fb76681221a8e10b3139618c5a9a291b84e", size = 222788 },
+    { url = "https://files.pythonhosted.org/packages/db/81/786f687951d0979007e05ad9346cd357e50e3d0b0f1a1d6074df334b1bbb/propcache-0.2.1-cp311-cp311-win32.whl", hash = "sha256:819ce3b883b7576ca28da3861c7e1a88afd08cc8c96908e08a3f4dd64a228034", size = 40170 },
+    { url = "https://files.pythonhosted.org/packages/cf/59/7cc7037b295d5772eceb426358bb1b86e6cab4616d971bd74275395d100d/propcache-0.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:edc9fc7051e3350643ad929df55c451899bb9ae6d24998a949d2e4c87fb596d3", size = 44404 },
+    { url = "https://files.pythonhosted.org/packages/4c/28/1d205fe49be8b1b4df4c50024e62480a442b1a7b818e734308bb0d17e7fb/propcache-0.2.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:081a430aa8d5e8876c6909b67bd2d937bfd531b0382d3fdedb82612c618bc41a", size = 79588 },
+    { url = "https://files.pythonhosted.org/packages/21/ee/fc4d893f8d81cd4971affef2a6cb542b36617cd1d8ce56b406112cb80bf7/propcache-0.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:d2ccec9ac47cf4e04897619c0e0c1a48c54a71bdf045117d3a26f80d38ab1fb0", size = 45825 },
+    { url = "https://files.pythonhosted.org/packages/4a/de/bbe712f94d088da1d237c35d735f675e494a816fd6f54e9db2f61ef4d03f/propcache-0.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:14d86fe14b7e04fa306e0c43cdbeebe6b2c2156a0c9ce56b815faacc193e320d", size = 45357 },
+    { url = "https://files.pythonhosted.org/packages/7f/14/7ae06a6cf2a2f1cb382586d5a99efe66b0b3d0c6f9ac2f759e6f7af9d7cf/propcache-0.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:049324ee97bb67285b49632132db351b41e77833678432be52bdd0289c0e05e4", size = 241869 },
+    { url = "https://files.pythonhosted.org/packages/cc/59/227a78be960b54a41124e639e2c39e8807ac0c751c735a900e21315f8c2b/propcache-0.2.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1cd9a1d071158de1cc1c71a26014dcdfa7dd3d5f4f88c298c7f90ad6f27bb46d", size = 247884 },
+    { url = "https://files.pythonhosted.org/packages/84/58/f62b4ffaedf88dc1b17f04d57d8536601e4e030feb26617228ef930c3279/propcache-0.2.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:98110aa363f1bb4c073e8dcfaefd3a5cea0f0834c2aab23dda657e4dab2f53b5", size = 248486 },
+    { url = "https://files.pythonhosted.org/packages/1c/07/ebe102777a830bca91bbb93e3479cd34c2ca5d0361b83be9dbd93104865e/propcache-0.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:647894f5ae99c4cf6bb82a1bb3a796f6e06af3caa3d32e26d2350d0e3e3faf24", size = 243649 },
+    { url = "https://files.pythonhosted.org/packages/ed/bc/4f7aba7f08f520376c4bb6a20b9a981a581b7f2e385fa0ec9f789bb2d362/propcache-0.2.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bfd3223c15bebe26518d58ccf9a39b93948d3dcb3e57a20480dfdd315356baff", size = 229103 },
+    { url = "https://files.pythonhosted.org/packages/fe/d5/04ac9cd4e51a57a96f78795e03c5a0ddb8f23ec098b86f92de028d7f2a6b/propcache-0.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d71264a80f3fcf512eb4f18f59423fe82d6e346ee97b90625f283df56aee103f", size = 226607 },
+    { url = "https://files.pythonhosted.org/packages/e3/f0/24060d959ea41d7a7cc7fdbf68b31852331aabda914a0c63bdb0e22e96d6/propcache-0.2.1-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:e73091191e4280403bde6c9a52a6999d69cdfde498f1fdf629105247599b57ec", size = 221153 },
+    { url = "https://files.pythonhosted.org/packages/77/a7/3ac76045a077b3e4de4859a0753010765e45749bdf53bd02bc4d372da1a0/propcache-0.2.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:3935bfa5fede35fb202c4b569bb9c042f337ca4ff7bd540a0aa5e37131659348", size = 222151 },
+    { url = "https://files.pythonhosted.org/packages/e7/af/5e29da6f80cebab3f5a4dcd2a3240e7f56f2c4abf51cbfcc99be34e17f0b/propcache-0.2.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:f508b0491767bb1f2b87fdfacaba5f7eddc2f867740ec69ece6d1946d29029a6", size = 233812 },
+    { url = "https://files.pythonhosted.org/packages/8c/89/ebe3ad52642cc5509eaa453e9f4b94b374d81bae3265c59d5c2d98efa1b4/propcache-0.2.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:1672137af7c46662a1c2be1e8dc78cb6d224319aaa40271c9257d886be4363a6", size = 238829 },
+    { url = "https://files.pythonhosted.org/packages/e9/2f/6b32f273fa02e978b7577159eae7471b3cfb88b48563b1c2578b2d7ca0bb/propcache-0.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b74c261802d3d2b85c9df2dfb2fa81b6f90deeef63c2db9f0e029a3cac50b518", size = 230704 },
+    { url = "https://files.pythonhosted.org/packages/5c/2e/f40ae6ff5624a5f77edd7b8359b208b5455ea113f68309e2b00a2e1426b6/propcache-0.2.1-cp312-cp312-win32.whl", hash = "sha256:d09c333d36c1409d56a9d29b3a1b800a42c76a57a5a8907eacdbce3f18768246", size = 40050 },
+    { url = "https://files.pythonhosted.org/packages/3b/77/a92c3ef994e47180862b9d7d11e37624fb1c00a16d61faf55115d970628b/propcache-0.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:c214999039d4f2a5b2073ac506bba279945233da8c786e490d411dfc30f855c1", size = 44117 },
+    { url = "https://files.pythonhosted.org/packages/0f/2a/329e0547cf2def8857157f9477669043e75524cc3e6251cef332b3ff256f/propcache-0.2.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:aca405706e0b0a44cc6bfd41fbe89919a6a56999157f6de7e182a990c36e37bc", size = 77002 },
+    { url = "https://files.pythonhosted.org/packages/12/2d/c4df5415e2382f840dc2ecbca0eeb2293024bc28e57a80392f2012b4708c/propcache-0.2.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:12d1083f001ace206fe34b6bdc2cb94be66d57a850866f0b908972f90996b3e9", size = 44639 },
+    { url = "https://files.pythonhosted.org/packages/d0/5a/21aaa4ea2f326edaa4e240959ac8b8386ea31dedfdaa636a3544d9e7a408/propcache-0.2.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d93f3307ad32a27bda2e88ec81134b823c240aa3abb55821a8da553eed8d9439", size = 44049 },
+    { url = "https://files.pythonhosted.org/packages/4e/3e/021b6cd86c0acc90d74784ccbb66808b0bd36067a1bf3e2deb0f3845f618/propcache-0.2.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ba278acf14471d36316159c94a802933d10b6a1e117b8554fe0d0d9b75c9d536", size = 224819 },
+    { url = "https://files.pythonhosted.org/packages/3c/57/c2fdeed1b3b8918b1770a133ba5c43ad3d78e18285b0c06364861ef5cc38/propcache-0.2.1-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4e6281aedfca15301c41f74d7005e6e3f4ca143584ba696ac69df4f02f40d629", size = 229625 },
+    { url = "https://files.pythonhosted.org/packages/9d/81/70d4ff57bf2877b5780b466471bebf5892f851a7e2ca0ae7ffd728220281/propcache-0.2.1-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5b750a8e5a1262434fb1517ddf64b5de58327f1adc3524a5e44c2ca43305eb0b", size = 232934 },
+    { url = "https://files.pythonhosted.org/packages/3c/b9/bb51ea95d73b3fb4100cb95adbd4e1acaf2cbb1fd1083f5468eeb4a099a8/propcache-0.2.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bf72af5e0fb40e9babf594308911436c8efde3cb5e75b6f206c34ad18be5c052", size = 227361 },
+    { url = "https://files.pythonhosted.org/packages/f1/20/3c6d696cd6fd70b29445960cc803b1851a1131e7a2e4ee261ee48e002bcd/propcache-0.2.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b2d0a12018b04f4cb820781ec0dffb5f7c7c1d2a5cd22bff7fb055a2cb19ebce", size = 213904 },
+    { url = "https://files.pythonhosted.org/packages/a1/cb/1593bfc5ac6d40c010fa823f128056d6bc25b667f5393781e37d62f12005/propcache-0.2.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:e800776a79a5aabdb17dcc2346a7d66d0777e942e4cd251defeb084762ecd17d", size = 212632 },
+    { url = "https://files.pythonhosted.org/packages/6d/5c/e95617e222be14a34c709442a0ec179f3207f8a2b900273720501a70ec5e/propcache-0.2.1-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:4160d9283bd382fa6c0c2b5e017acc95bc183570cd70968b9202ad6d8fc48dce", size = 207897 },
+    { url = "https://files.pythonhosted.org/packages/8e/3b/56c5ab3dc00f6375fbcdeefdede5adf9bee94f1fab04adc8db118f0f9e25/propcache-0.2.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:30b43e74f1359353341a7adb783c8f1b1c676367b011709f466f42fda2045e95", size = 208118 },
+    { url = "https://files.pythonhosted.org/packages/86/25/d7ef738323fbc6ebcbce33eb2a19c5e07a89a3df2fded206065bd5e868a9/propcache-0.2.1-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:58791550b27d5488b1bb52bc96328456095d96206a250d28d874fafe11b3dfaf", size = 217851 },
+    { url = "https://files.pythonhosted.org/packages/b3/77/763e6cef1852cf1ba740590364ec50309b89d1c818e3256d3929eb92fabf/propcache-0.2.1-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:0f022d381747f0dfe27e99d928e31bc51a18b65bb9e481ae0af1380a6725dd1f", size = 222630 },
+    { url = "https://files.pythonhosted.org/packages/4f/e9/0f86be33602089c701696fbed8d8c4c07b6ee9605c5b7536fd27ed540c5b/propcache-0.2.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:297878dc9d0a334358f9b608b56d02e72899f3b8499fc6044133f0d319e2ec30", size = 216269 },
+    { url = "https://files.pythonhosted.org/packages/cc/02/5ac83217d522394b6a2e81a2e888167e7ca629ef6569a3f09852d6dcb01a/propcache-0.2.1-cp313-cp313-win32.whl", hash = "sha256:ddfab44e4489bd79bda09d84c430677fc7f0a4939a73d2bba3073036f487a0a6", size = 39472 },
+    { url = "https://files.pythonhosted.org/packages/f4/33/d6f5420252a36034bc8a3a01171bc55b4bff5df50d1c63d9caa50693662f/propcache-0.2.1-cp313-cp313-win_amd64.whl", hash = "sha256:556fc6c10989f19a179e4321e5d678db8eb2924131e64652a51fe83e4c3db0e1", size = 43363 },
+    { url = "https://files.pythonhosted.org/packages/0a/08/6ab7f65240a16fa01023125e65258acf7e4884f483f267cdd6fcc48f37db/propcache-0.2.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:6a9a8c34fb7bb609419a211e59da8887eeca40d300b5ea8e56af98f6fbbb1541", size = 80403 },
+    { url = "https://files.pythonhosted.org/packages/34/fe/e7180285e21b4e6dff7d311fdf22490c9146a09a02834b5232d6248c6004/propcache-0.2.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ae1aa1cd222c6d205853b3013c69cd04515f9d6ab6de4b0603e2e1c33221303e", size = 46152 },
+    { url = "https://files.pythonhosted.org/packages/9c/36/aa74d884af826030ba9cee2ac109b0664beb7e9449c315c9c44db99efbb3/propcache-0.2.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:accb6150ce61c9c4b7738d45550806aa2b71c7668c6942f17b0ac182b6142fd4", size = 45674 },
+    { url = "https://files.pythonhosted.org/packages/22/59/6fe80a3fe7720f715f2c0f6df250dacbd7cad42832410dbd84c719c52f78/propcache-0.2.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5eee736daafa7af6d0a2dc15cc75e05c64f37fc37bafef2e00d77c14171c2097", size = 207792 },
+    { url = "https://files.pythonhosted.org/packages/4a/68/584cd51dd8f4d0f5fff5b128ce0cdb257cde903898eecfb92156bbc2c780/propcache-0.2.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f7a31fc1e1bd362874863fdeed71aed92d348f5336fd84f2197ba40c59f061bd", size = 223280 },
+    { url = "https://files.pythonhosted.org/packages/85/cb/4c3528460c41e61b06ec3f970c0f89f87fa21f63acac8642ed81a886c164/propcache-0.2.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cba4cfa1052819d16699e1d55d18c92b6e094d4517c41dd231a8b9f87b6fa681", size = 221293 },
+    { url = "https://files.pythonhosted.org/packages/69/c0/560e050aa6d31eeece3490d1174da508f05ab27536dfc8474af88b97160a/propcache-0.2.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f089118d584e859c62b3da0892b88a83d611c2033ac410e929cb6754eec0ed16", size = 208259 },
+    { url = "https://files.pythonhosted.org/packages/0c/87/d6c86a77632eb1ba86a328e3313159f246e7564cb5951e05ed77555826a0/propcache-0.2.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:781e65134efaf88feb447e8c97a51772aa75e48b794352f94cb7ea717dedda0d", size = 198632 },
+    { url = "https://files.pythonhosted.org/packages/3a/2b/3690ea7b662dc762ab7af5f3ef0e2d7513c823d193d7b2a1b4cda472c2be/propcache-0.2.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:31f5af773530fd3c658b32b6bdc2d0838543de70eb9a2156c03e410f7b0d3aae", size = 203516 },
+    { url = "https://files.pythonhosted.org/packages/4d/b5/afe716c16c23c77657185c257a41918b83e03993b6ccdfa748e5e7d328e9/propcache-0.2.1-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:a7a078f5d37bee6690959c813977da5291b24286e7b962e62a94cec31aa5188b", size = 199402 },
+    { url = "https://files.pythonhosted.org/packages/a4/c0/2d2df3aa7f8660d0d4cc4f1e00490c48d5958da57082e70dea7af366f876/propcache-0.2.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:cea7daf9fc7ae6687cf1e2c049752f19f146fdc37c2cc376e7d0032cf4f25347", size = 200528 },
+    { url = "https://files.pythonhosted.org/packages/21/c8/65ac9142f5e40c8497f7176e71d18826b09e06dd4eb401c9a4ee41aa9c74/propcache-0.2.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:8b3489ff1ed1e8315674d0775dc7d2195fb13ca17b3808721b54dbe9fd020faf", size = 211254 },
+    { url = "https://files.pythonhosted.org/packages/09/e4/edb70b447a1d8142df51ec7511e84aa64d7f6ce0a0fdf5eb55363cdd0935/propcache-0.2.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:9403db39be1393618dd80c746cb22ccda168efce239c73af13c3763ef56ffc04", size = 214589 },
+    { url = "https://files.pythonhosted.org/packages/cb/02/817f309ec8d8883287781d6d9390f80b14db6e6de08bc659dfe798a825c2/propcache-0.2.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:5d97151bc92d2b2578ff7ce779cdb9174337390a535953cbb9452fb65164c587", size = 207283 },
+    { url = "https://files.pythonhosted.org/packages/d7/fe/2d18612096ed2212cfef821b6fccdba5d52efc1d64511c206c5c16be28fd/propcache-0.2.1-cp39-cp39-win32.whl", hash = "sha256:9caac6b54914bdf41bcc91e7eb9147d331d29235a7c967c150ef5df6464fd1bb", size = 40866 },
+    { url = "https://files.pythonhosted.org/packages/24/2e/b5134802e7b57c403c7b73c7a39374e7a6b7f128d1968b4a4b4c0b700250/propcache-0.2.1-cp39-cp39-win_amd64.whl", hash = "sha256:92fc4500fcb33899b05ba73276dfb684a20d31caa567b7cb5252d48f896a91b1", size = 44975 },
+    { url = "https://files.pythonhosted.org/packages/41/b6/c5319caea262f4821995dca2107483b94a3345d4607ad797c76cb9c36bcc/propcache-0.2.1-py3-none-any.whl", hash = "sha256:52277518d6aae65536e9cea52d4e7fd2f7a66f4aa2d30ed3f2fcea620ace3c54", size = 11818 },
+]
+
+[[package]]
+name = "protobuf"
+version = "5.29.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f7/d1/e0a911544ca9993e0f17ce6d3cc0932752356c1b0a834397f28e63479344/protobuf-5.29.3.tar.gz", hash = "sha256:5da0f41edaf117bde316404bad1a486cb4ededf8e4a54891296f648e8e076620", size = 424945 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/dc/7a/1e38f3cafa022f477ca0f57a1f49962f21ad25850c3ca0acd3b9d0091518/protobuf-5.29.3-cp310-abi3-win32.whl", hash = "sha256:3ea51771449e1035f26069c4c7fd51fba990d07bc55ba80701c78f886bf9c888", size = 422708 },
+    { url = "https://files.pythonhosted.org/packages/61/fa/aae8e10512b83de633f2646506a6d835b151edf4b30d18d73afd01447253/protobuf-5.29.3-cp310-abi3-win_amd64.whl", hash = "sha256:a4fa6f80816a9a0678429e84973f2f98cbc218cca434abe8db2ad0bffc98503a", size = 434508 },
+    { url = "https://files.pythonhosted.org/packages/dd/04/3eaedc2ba17a088961d0e3bd396eac764450f431621b58a04ce898acd126/protobuf-5.29.3-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:a8434404bbf139aa9e1300dbf989667a83d42ddda9153d8ab76e0d5dcaca484e", size = 417825 },
+    { url = "https://files.pythonhosted.org/packages/4f/06/7c467744d23c3979ce250397e26d8ad8eeb2bea7b18ca12ad58313c1b8d5/protobuf-5.29.3-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:daaf63f70f25e8689c072cfad4334ca0ac1d1e05a92fc15c54eb9cf23c3efd84", size = 319573 },
+    { url = "https://files.pythonhosted.org/packages/a8/45/2ebbde52ad2be18d3675b6bee50e68cd73c9e0654de77d595540b5129df8/protobuf-5.29.3-cp38-abi3-manylinux2014_x86_64.whl", hash = "sha256:c027e08a08be10b67c06bf2370b99c811c466398c357e615ca88c91c07f0910f", size = 319672 },
+    { url = "https://files.pythonhosted.org/packages/85/a6/bf65a38f8be5ab8c3b575822acfd338702fdf7ac9abd8c81630cc7c9f4bd/protobuf-5.29.3-cp39-cp39-win32.whl", hash = "sha256:0eb32bfa5219fc8d4111803e9a690658aa2e6366384fd0851064b963b6d1f2a7", size = 422676 },
+    { url = "https://files.pythonhosted.org/packages/ac/e2/48d46adc86369ff092eaece3e537f76b3baaab45ca3dde257838cde831d2/protobuf-5.29.3-cp39-cp39-win_amd64.whl", hash = "sha256:6ce8cc3389a20693bfde6c6562e03474c40851b44975c9b2bf6df7d8c4f864da", size = 434593 },
+    { url = "https://files.pythonhosted.org/packages/fd/b2/ab07b09e0f6d143dfb839693aa05765257bceaa13d03bf1a696b78323e7a/protobuf-5.29.3-py3-none-any.whl", hash = "sha256:0a18ed4a24198528f2333802eb075e59dea9d679ab7a6c5efb017a59004d849f", size = 172550 },
+]
+
+[[package]]
+name = "psutil"
+version = "6.1.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/1f/5a/07871137bb752428aa4b659f910b399ba6f291156bdea939be3e96cae7cb/psutil-6.1.1.tar.gz", hash = "sha256:cf8496728c18f2d0b45198f06895be52f36611711746b7f30c464b422b50e2f5", size = 508502 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/61/99/ca79d302be46f7bdd8321089762dd4476ee725fce16fc2b2e1dbba8cac17/psutil-6.1.1-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:fc0ed7fe2231a444fc219b9c42d0376e0a9a1a72f16c5cfa0f68d19f1a0663e8", size = 247511 },
+    { url = "https://files.pythonhosted.org/packages/0b/6b/73dbde0dd38f3782905d4587049b9be64d76671042fdcaf60e2430c6796d/psutil-6.1.1-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:0bdd4eab935276290ad3cb718e9809412895ca6b5b334f5a9111ee6d9aff9377", size = 248985 },
+    { url = "https://files.pythonhosted.org/packages/17/38/c319d31a1d3f88c5b79c68b3116c129e5133f1822157dd6da34043e32ed6/psutil-6.1.1-cp36-abi3-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b6e06c20c05fe95a3d7302d74e7097756d4ba1247975ad6905441ae1b5b66003", size = 284488 },
+    { url = "https://files.pythonhosted.org/packages/9c/39/0f88a830a1c8a3aba27fededc642da37613c57cbff143412e3536f89784f/psutil-6.1.1-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:97f7cb9921fbec4904f522d972f0c0e1f4fabbdd4e0287813b21215074a0f160", size = 287477 },
+    { url = "https://files.pythonhosted.org/packages/47/da/99f4345d4ddf2845cb5b5bd0d93d554e84542d116934fde07a0c50bd4e9f/psutil-6.1.1-cp36-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:33431e84fee02bc84ea36d9e2c4a6d395d479c9dd9bba2376c1f6ee8f3a4e0b3", size = 289017 },
+    { url = "https://files.pythonhosted.org/packages/38/53/bd755c2896f4461fd4f36fa6a6dcb66a88a9e4b9fd4e5b66a77cf9d4a584/psutil-6.1.1-cp37-abi3-win32.whl", hash = "sha256:eaa912e0b11848c4d9279a93d7e2783df352b082f40111e078388701fd479e53", size = 250602 },
+    { url = "https://files.pythonhosted.org/packages/7b/d7/7831438e6c3ebbfa6e01a927127a6cb42ad3ab844247f3c5b96bea25d73d/psutil-6.1.1-cp37-abi3-win_amd64.whl", hash = "sha256:f35cfccb065fff93529d2afb4a2e89e363fe63ca1e4a5da22b603a85833c2649", size = 254444 },
+]
+
+[[package]]
+name = "py-cpuinfo"
+version = "9.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/37/a8/d832f7293ebb21690860d2e01d8115e5ff6f2ae8bbdc953f0eb0fa4bd2c7/py-cpuinfo-9.0.0.tar.gz", hash = "sha256:3cdbbf3fac90dc6f118bfd64384f309edeadd902d7c8fb17f02ffa1fc3f49690", size = 104716 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e0/a9/023730ba63db1e494a271cb018dcd361bd2c917ba7004c3e49d5daf795a2/py_cpuinfo-9.0.0-py3-none-any.whl", hash = "sha256:859625bc251f64e21f077d099d4162689c762b5d6a4c3c97553d56241c9674d5", size = 22335 },
+]
+
+[[package]]
+name = "pyarrow"
+version = "19.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7b/01/fe1fd04744c2aa038e5a11c7a4adb3d62bce09798695e54f7274b5977134/pyarrow-19.0.0.tar.gz", hash = "sha256:8d47c691765cf497aaeed4954d226568563f1b3b74ff61139f2d77876717084b", size = 1129096 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1c/02/1ad80ffd3c558916858a49c83b6e494a9d93009bbebc603cf0cb8263bea7/pyarrow-19.0.0-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:c318eda14f6627966997a7d8c374a87d084a94e4e38e9abbe97395c215830e0c", size = 30686262 },
+    { url = "https://files.pythonhosted.org/packages/1b/f0/adab5f142eb8203db8bfbd3a816816e37a85423ae684567e7f3555658315/pyarrow-19.0.0-cp310-cp310-macosx_12_0_x86_64.whl", hash = "sha256:62ef8360ff256e960f57ce0299090fb86423afed5e46f18f1225f960e05aae3d", size = 32100005 },
+    { url = "https://files.pythonhosted.org/packages/94/8b/e674083610e5efc48d2f205c568d842cdfdf683d12f9ff0d546e38757722/pyarrow-19.0.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2795064647add0f16563e57e3d294dbfc067b723f0fd82ecd80af56dad15f503", size = 41144815 },
+    { url = "https://files.pythonhosted.org/packages/d5/fb/2726241a792b7f8a58789e5a63d1be9a5a4059206318fd0ff9485a578952/pyarrow-19.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a218670b26fb1bc74796458d97bcab072765f9b524f95b2fccad70158feb8b17", size = 42180380 },
+    { url = "https://files.pythonhosted.org/packages/7d/09/7aef12446d8e7002dfc07bb7bc71f594c1d5844ca78b364a49f07efb65b1/pyarrow-19.0.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:66732e39eaa2247996a6b04c8aa33e3503d351831424cdf8d2e9a0582ac54b34", size = 40515021 },
+    { url = "https://files.pythonhosted.org/packages/31/55/f05fc5608cc96060c2b24de505324d641888bd62d4eed2fa1dacd872a1e1/pyarrow-19.0.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:e675a3ad4732b92d72e4d24009707e923cab76b0d088e5054914f11a797ebe44", size = 42067488 },
+    { url = "https://files.pythonhosted.org/packages/f0/01/097653cec7a944c16313cb748a326771133c142034b252076bd84743b98d/pyarrow-19.0.0-cp310-cp310-win_amd64.whl", hash = "sha256:f094742275586cdd6b1a03655ccff3b24b2610c3af76f810356c4c71d24a2a6c", size = 25276726 },
+    { url = "https://files.pythonhosted.org/packages/82/42/fba3a35bef5833bf88ed35e6a810dc1781236e1d4f808d2df824a7d21819/pyarrow-19.0.0-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:8e3a839bf36ec03b4315dc924d36dcde5444a50066f1c10f8290293c0427b46a", size = 30711936 },
+    { url = "https://files.pythonhosted.org/packages/88/7a/0da93a3eaaf251a30e32f3221e874263cdcd366c2cd6b7c05293aad91152/pyarrow-19.0.0-cp311-cp311-macosx_12_0_x86_64.whl", hash = "sha256:ce42275097512d9e4e4a39aade58ef2b3798a93aa3026566b7892177c266f735", size = 32133182 },
+    { url = "https://files.pythonhosted.org/packages/2f/df/fe43b1c50d3100d0de53f988344118bc20362d0de005f8a407454fa565f8/pyarrow-19.0.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9348a0137568c45601b031a8d118275069435f151cbb77e6a08a27e8125f59d4", size = 41145489 },
+    { url = "https://files.pythonhosted.org/packages/45/bb/6f73b41b342a0342f2516a02db4aa97a4f9569cc35482a5c288090140cd4/pyarrow-19.0.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2a0144a712d990d60f7f42b7a31f0acaccf4c1e43e957f7b1ad58150d6f639c1", size = 42177823 },
+    { url = "https://files.pythonhosted.org/packages/23/7b/f038a96f421e453a71bd7a0f78d62b1b2ae9bcac06ed51179ca532e6a0a2/pyarrow-19.0.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:2a1a109dfda558eb011e5f6385837daffd920d54ca00669f7a11132d0b1e6042", size = 40530609 },
+    { url = "https://files.pythonhosted.org/packages/b8/39/a2a6714b471c000e6dd6af4495dce00d7d1332351b8e3170dfb9f91dad1f/pyarrow-19.0.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:be686bf625aa7b9bada18defb3a3ea3981c1099697239788ff111d87f04cd263", size = 42081534 },
+    { url = "https://files.pythonhosted.org/packages/6c/a3/8396fb06ca05d807e89980c177be26617aad15211ece3184e0caa730b8a6/pyarrow-19.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:239ca66d9a05844bdf5af128861af525e14df3c9591bcc05bac25918e650d3a2", size = 25281090 },
+    { url = "https://files.pythonhosted.org/packages/bc/2e/152885f5ef421e80dae68b9c133ab261934f93a6d5e16b61d79c0ed597fb/pyarrow-19.0.0-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:a7bbe7109ab6198688b7079cbad5a8c22de4d47c4880d8e4847520a83b0d1b68", size = 30667964 },
+    { url = "https://files.pythonhosted.org/packages/80/c2/08bbee9a8610a47c9a1466845f405baf53a639ddd947c5133d8ba13544b6/pyarrow-19.0.0-cp312-cp312-macosx_12_0_x86_64.whl", hash = "sha256:4624c89d6f777c580e8732c27bb8e77fd1433b89707f17c04af7635dd9638351", size = 32125039 },
+    { url = "https://files.pythonhosted.org/packages/d2/56/06994df823212f5688d3c8bf4294928b12c9be36681872853655724d28c6/pyarrow-19.0.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2b6d3ce4288793350dc2d08d1e184fd70631ea22a4ff9ea5c4ff182130249d9b", size = 41140729 },
+    { url = "https://files.pythonhosted.org/packages/94/65/38ad577c98140a9db71e9e1e594b6adb58a7478a5afec6456a8ca2df7f70/pyarrow-19.0.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:450a7d27e840e4d9a384b5c77199d489b401529e75a3b7a3799d4cd7957f2f9c", size = 42202267 },
+    { url = "https://files.pythonhosted.org/packages/b6/1f/966b722251a7354114ccbb71cf1a83922023e69efd8945ebf628a851ec4c/pyarrow-19.0.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:a08e2a8a039a3f72afb67a6668180f09fddaa38fe0d21f13212b4aba4b5d2451", size = 40505858 },
+    { url = "https://files.pythonhosted.org/packages/3b/5e/6bc81aa7fc9affc7d1c03b912fbcc984ca56c2a18513684da267715dab7b/pyarrow-19.0.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:f43f5aef2a13d4d56adadae5720d1fed4c1356c993eda8b59dace4b5983843c1", size = 42084973 },
+    { url = "https://files.pythonhosted.org/packages/53/c3/2f56da818b6a4758cbd514957c67bd0f078ebffa5390ee2e2bf0f9e8defc/pyarrow-19.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:2f672f5364b2d7829ef7c94be199bb88bf5661dd485e21d2d37de12ccb78a136", size = 25241976 },
+    { url = "https://files.pythonhosted.org/packages/f5/b9/ba07ed3dd6b6e4f379b78e9c47c50c8886e07862ab7fa6339ac38622d755/pyarrow-19.0.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:cf3bf0ce511b833f7bc5f5bb3127ba731e97222023a444b7359f3a22e2a3b463", size = 30651291 },
+    { url = "https://files.pythonhosted.org/packages/ad/10/0d304243c8277035298a68a70807efb76199c6c929bb3363c92ac9be6a0d/pyarrow-19.0.0-cp313-cp313-macosx_12_0_x86_64.whl", hash = "sha256:4d8b0c0de0a73df1f1bf439af1b60f273d719d70648e898bc077547649bb8352", size = 32100461 },
+    { url = "https://files.pythonhosted.org/packages/8a/61/bcfc5182e11831bca3f849945b9b106e09fd10ded773dff466658e972a45/pyarrow-19.0.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a92aff08e23d281c69835e4a47b80569242a504095ef6a6223c1f6bb8883431d", size = 41132491 },
+    { url = "https://files.pythonhosted.org/packages/8e/87/2915a29049ec352dc69a967fbcbd76b0180319233de0daf8bd368df37099/pyarrow-19.0.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3b78eff5968a1889a0f3bc81ca57e1e19b75f664d9c61a42a604bf9d8402aae", size = 42192529 },
+    { url = "https://files.pythonhosted.org/packages/48/18/44e5542b2707a8afaf78b5b88c608f261871ae77787eac07b7c679ca6f0f/pyarrow-19.0.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:b34d3bde38eba66190b215bae441646330f8e9da05c29e4b5dd3e41bde701098", size = 40495363 },
+    { url = "https://files.pythonhosted.org/packages/ba/d6/5096deb7599bbd20bc2768058fe23bc725b88eb41bee58303293583a2935/pyarrow-19.0.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:5418d4d0fab3a0ed497bad21d17a7973aad336d66ad4932a3f5f7480d4ca0c04", size = 42074075 },
+    { url = "https://files.pythonhosted.org/packages/2c/df/e3c839c04c284c9ec3d62b02a8c452b795d9b07b04079ab91ce33484d4c5/pyarrow-19.0.0-cp313-cp313-win_amd64.whl", hash = "sha256:e82c3d5e44e969c217827b780ed8faf7ac4c53f934ae9238872e749fa531f7c9", size = 25239803 },
+    { url = "https://files.pythonhosted.org/packages/6a/d3/a6d4088e906c7b5d47792256212606d2ae679046dc750eee0ae167338e5c/pyarrow-19.0.0-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:f208c3b58a6df3b239e0bb130e13bc7487ed14f39a9ff357b6415e3f6339b560", size = 30695401 },
+    { url = "https://files.pythonhosted.org/packages/94/25/70040fd0e397dd1b937f459eaeeec942a76027357491dca0ada09d1322af/pyarrow-19.0.0-cp313-cp313t-macosx_12_0_x86_64.whl", hash = "sha256:c751c1c93955b7a84c06794df46f1cec93e18610dcd5ab7d08e89a81df70a849", size = 32104680 },
+    { url = "https://files.pythonhosted.org/packages/4e/f9/92783290cc0d80ca16d34b0c126305bfacca4b87dd889c8f16c6ef2a8fd7/pyarrow-19.0.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b903afaa5df66d50fc38672ad095806443b05f202c792694f3a604ead7c6ea6e", size = 41076754 },
+    { url = "https://files.pythonhosted.org/packages/05/46/2c9870f50a495c72e2b8982ae29a9b1680707ea936edc0de444cec48f875/pyarrow-19.0.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a22a4bc0937856263df8b94f2f2781b33dd7f876f787ed746608e06902d691a5", size = 42163133 },
+    { url = "https://files.pythonhosted.org/packages/7b/2f/437922b902549228fb15814e8a26105bff2787ece466a8d886eb6699efad/pyarrow-19.0.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:5e8a28b918e2e878c918f6d89137386c06fe577cd08d73a6be8dafb317dc2d73", size = 40452210 },
+    { url = "https://files.pythonhosted.org/packages/36/ef/1d7975053af9d106da973bac142d0d4da71b7550a3576cc3e0b3f444d21a/pyarrow-19.0.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:29cd86c8001a94f768f79440bf83fee23963af5e7bc68ce3a7e5f120e17edf89", size = 42077618 },
+    { url = "https://files.pythonhosted.org/packages/59/13/e39417005ee632e131d0246cf5c1149618a55554ccdf2a4d887065e672a7/pyarrow-19.0.0-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:c0423393e4a07ff6fea08feb44153302dd261d0551cc3b538ea7a5dc853af43a", size = 30698254 },
+    { url = "https://files.pythonhosted.org/packages/06/87/1f9d7df296dd5c065e52ae3e9070dfe611f6bd97e90f28b6a45c410dcb67/pyarrow-19.0.0-cp39-cp39-macosx_12_0_x86_64.whl", hash = "sha256:718947fb6d82409013a74b176bf93e0f49ef952d8a2ecd068fecd192a97885b7", size = 32114467 },
+    { url = "https://files.pythonhosted.org/packages/b2/b1/9e7babf5d469bd35f1d062a04721ead72456101f6d851a2e8a43bb07a580/pyarrow-19.0.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3c1c162c4660e0978411a4761f91113dde8da3433683efa473501254563dcbe8", size = 41157781 },
+    { url = "https://files.pythonhosted.org/packages/9f/25/fa8e882a6c06e6d8dd640d3acce3912d7c39358940eb0c7b3c8b962457d0/pyarrow-19.0.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c73268cf557e688efb60f1ccbc7376f7e18cd8e2acae9e663e98b194c40c1a2d", size = 42190773 },
+    { url = "https://files.pythonhosted.org/packages/b8/d6/7fd60aa79cada815306d9804403386b06893ef63e73876174717a62002c4/pyarrow-19.0.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:edfe6d3916e915ada9acc4e48f6dafca7efdbad2e6283db6fd9385a1b23055f1", size = 40528310 },
+    { url = "https://files.pythonhosted.org/packages/0e/f1/c1ec7620a5768b8c7f9083572fda7b05b77ea083c3400fbd9e4ae40e63bd/pyarrow-19.0.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:da410b70a7ab8eb524112f037a7a35da7128b33d484f7671a264a4c224ac131d", size = 42080774 },
+    { url = "https://files.pythonhosted.org/packages/c4/28/c51c9af2703b5a592d1b66546611b24de8ca01e04c3f5da769c3318bca6c/pyarrow-19.0.0-cp39-cp39-win_amd64.whl", hash = "sha256:597360ffc71fc8cceea1aec1fb60cb510571a744fffc87db33d551d5de919bec", size = 25464978 },
+]
+
+[[package]]
+name = "pycountry"
+version = "24.6.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/76/57/c389fa68c50590881a75b7883eeb3dc15e9e73a0fdc001cdd45c13290c92/pycountry-24.6.1.tar.gz", hash = "sha256:b61b3faccea67f87d10c1f2b0fc0be714409e8fcdcc1315613174f6466c10221", size = 6043910 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b1/ec/1fb891d8a2660716aadb2143235481d15ed1cbfe3ad669194690b0604492/pycountry-24.6.1-py3-none-any.whl", hash = "sha256:f1a4fb391cd7214f8eefd39556d740adcc233c778a27f8942c8dca351d6ce06f", size = 6335189 },
+]
+
+[[package]]
+name = "pydantic"
+version = "2.10.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "annotated-types" },
+    { name = "pydantic-core" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6a/c7/ca334c2ef6f2e046b1144fe4bb2a5da8a4c574e7f2ebf7e16b34a6a2fa92/pydantic-2.10.5.tar.gz", hash = "sha256:278b38dbbaec562011d659ee05f63346951b3a248a6f3642e1bc68894ea2b4ff", size = 761287 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/58/26/82663c79010b28eddf29dcdd0ea723439535fa917fce5905885c0e9ba562/pydantic-2.10.5-py3-none-any.whl", hash = "sha256:4dd4e322dbe55472cb7ca7e73f4b63574eecccf2835ffa2af9021ce113c83c53", size = 431426 },
+]
+
+[[package]]
+name = "pydantic-core"
+version = "2.27.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fc/01/f3e5ac5e7c25833db5eb555f7b7ab24cd6f8c322d3a3ad2d67a952dc0abc/pydantic_core-2.27.2.tar.gz", hash = "sha256:eb026e5a4c1fee05726072337ff51d1efb6f59090b7da90d30ea58625b1ffb39", size = 413443 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3a/bc/fed5f74b5d802cf9a03e83f60f18864e90e3aed7223adaca5ffb7a8d8d64/pydantic_core-2.27.2-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:2d367ca20b2f14095a8f4fa1210f5a7b78b8a20009ecced6b12818f455b1e9fa", size = 1895938 },
+    { url = "https://files.pythonhosted.org/packages/71/2a/185aff24ce844e39abb8dd680f4e959f0006944f4a8a0ea372d9f9ae2e53/pydantic_core-2.27.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:491a2b73db93fab69731eaee494f320faa4e093dbed776be1a829c2eb222c34c", size = 1815684 },
+    { url = "https://files.pythonhosted.org/packages/c3/43/fafabd3d94d159d4f1ed62e383e264f146a17dd4d48453319fd782e7979e/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7969e133a6f183be60e9f6f56bfae753585680f3b7307a8e555a948d443cc05a", size = 1829169 },
+    { url = "https://files.pythonhosted.org/packages/a2/d1/f2dfe1a2a637ce6800b799aa086d079998959f6f1215eb4497966efd2274/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:3de9961f2a346257caf0aa508a4da705467f53778e9ef6fe744c038119737ef5", size = 1867227 },
+    { url = "https://files.pythonhosted.org/packages/7d/39/e06fcbcc1c785daa3160ccf6c1c38fea31f5754b756e34b65f74e99780b5/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e2bb4d3e5873c37bb3dd58714d4cd0b0e6238cebc4177ac8fe878f8b3aa8e74c", size = 2037695 },
+    { url = "https://files.pythonhosted.org/packages/7a/67/61291ee98e07f0650eb756d44998214231f50751ba7e13f4f325d95249ab/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:280d219beebb0752699480fe8f1dc61ab6615c2046d76b7ab7ee38858de0a4e7", size = 2741662 },
+    { url = "https://files.pythonhosted.org/packages/32/90/3b15e31b88ca39e9e626630b4c4a1f5a0dfd09076366f4219429e6786076/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:47956ae78b6422cbd46f772f1746799cbb862de838fd8d1fbd34a82e05b0983a", size = 1993370 },
+    { url = "https://files.pythonhosted.org/packages/ff/83/c06d333ee3a67e2e13e07794995c1535565132940715931c1c43bfc85b11/pydantic_core-2.27.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:14d4a5c49d2f009d62a2a7140d3064f686d17a5d1a268bc641954ba181880236", size = 1996813 },
+    { url = "https://files.pythonhosted.org/packages/7c/f7/89be1c8deb6e22618a74f0ca0d933fdcb8baa254753b26b25ad3acff8f74/pydantic_core-2.27.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:337b443af21d488716f8d0b6164de833e788aa6bd7e3a39c005febc1284f4962", size = 2005287 },
+    { url = "https://files.pythonhosted.org/packages/b7/7d/8eb3e23206c00ef7feee17b83a4ffa0a623eb1a9d382e56e4aa46fd15ff2/pydantic_core-2.27.2-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:03d0f86ea3184a12f41a2d23f7ccb79cdb5a18e06993f8a45baa8dfec746f0e9", size = 2128414 },
+    { url = "https://files.pythonhosted.org/packages/4e/99/fe80f3ff8dd71a3ea15763878d464476e6cb0a2db95ff1c5c554133b6b83/pydantic_core-2.27.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:7041c36f5680c6e0f08d922aed302e98b3745d97fe1589db0a3eebf6624523af", size = 2155301 },
+    { url = "https://files.pythonhosted.org/packages/2b/a3/e50460b9a5789ca1451b70d4f52546fa9e2b420ba3bfa6100105c0559238/pydantic_core-2.27.2-cp310-cp310-win32.whl", hash = "sha256:50a68f3e3819077be2c98110c1f9dcb3817e93f267ba80a2c05bb4f8799e2ff4", size = 1816685 },
+    { url = "https://files.pythonhosted.org/packages/57/4c/a8838731cb0f2c2a39d3535376466de6049034d7b239c0202a64aaa05533/pydantic_core-2.27.2-cp310-cp310-win_amd64.whl", hash = "sha256:e0fd26b16394ead34a424eecf8a31a1f5137094cabe84a1bcb10fa6ba39d3d31", size = 1982876 },
+    { url = "https://files.pythonhosted.org/packages/c2/89/f3450af9d09d44eea1f2c369f49e8f181d742f28220f88cc4dfaae91ea6e/pydantic_core-2.27.2-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:8e10c99ef58cfdf2a66fc15d66b16c4a04f62bca39db589ae8cba08bc55331bc", size = 1893421 },
+    { url = "https://files.pythonhosted.org/packages/9e/e3/71fe85af2021f3f386da42d291412e5baf6ce7716bd7101ea49c810eda90/pydantic_core-2.27.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:26f32e0adf166a84d0cb63be85c562ca8a6fa8de28e5f0d92250c6b7e9e2aff7", size = 1814998 },
+    { url = "https://files.pythonhosted.org/packages/a6/3c/724039e0d848fd69dbf5806894e26479577316c6f0f112bacaf67aa889ac/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8c19d1ea0673cd13cc2f872f6c9ab42acc4e4f492a7ca9d3795ce2b112dd7e15", size = 1826167 },
+    { url = "https://files.pythonhosted.org/packages/2b/5b/1b29e8c1fb5f3199a9a57c1452004ff39f494bbe9bdbe9a81e18172e40d3/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5e68c4446fe0810e959cdff46ab0a41ce2f2c86d227d96dc3847af0ba7def306", size = 1865071 },
+    { url = "https://files.pythonhosted.org/packages/89/6c/3985203863d76bb7d7266e36970d7e3b6385148c18a68cc8915fd8c84d57/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d9640b0059ff4f14d1f37321b94061c6db164fbe49b334b31643e0528d100d99", size = 2036244 },
+    { url = "https://files.pythonhosted.org/packages/0e/41/f15316858a246b5d723f7d7f599f79e37493b2e84bfc789e58d88c209f8a/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:40d02e7d45c9f8af700f3452f329ead92da4c5f4317ca9b896de7ce7199ea459", size = 2737470 },
+    { url = "https://files.pythonhosted.org/packages/a8/7c/b860618c25678bbd6d1d99dbdfdf0510ccb50790099b963ff78a124b754f/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1c1fd185014191700554795c99b347d64f2bb637966c4cfc16998a0ca700d048", size = 1992291 },
+    { url = "https://files.pythonhosted.org/packages/bf/73/42c3742a391eccbeab39f15213ecda3104ae8682ba3c0c28069fbcb8c10d/pydantic_core-2.27.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:d81d2068e1c1228a565af076598f9e7451712700b673de8f502f0334f281387d", size = 1994613 },
+    { url = "https://files.pythonhosted.org/packages/94/7a/941e89096d1175d56f59340f3a8ebaf20762fef222c298ea96d36a6328c5/pydantic_core-2.27.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:1a4207639fb02ec2dbb76227d7c751a20b1a6b4bc52850568e52260cae64ca3b", size = 2002355 },
+    { url = "https://files.pythonhosted.org/packages/6e/95/2359937a73d49e336a5a19848713555605d4d8d6940c3ec6c6c0ca4dcf25/pydantic_core-2.27.2-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:3de3ce3c9ddc8bbd88f6e0e304dea0e66d843ec9de1b0042b0911c1663ffd474", size = 2126661 },
+    { url = "https://files.pythonhosted.org/packages/2b/4c/ca02b7bdb6012a1adef21a50625b14f43ed4d11f1fc237f9d7490aa5078c/pydantic_core-2.27.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:30c5f68ded0c36466acede341551106821043e9afaad516adfb6e8fa80a4e6a6", size = 2153261 },
+    { url = "https://files.pythonhosted.org/packages/72/9d/a241db83f973049a1092a079272ffe2e3e82e98561ef6214ab53fe53b1c7/pydantic_core-2.27.2-cp311-cp311-win32.whl", hash = "sha256:c70c26d2c99f78b125a3459f8afe1aed4d9687c24fd677c6a4436bc042e50d6c", size = 1812361 },
+    { url = "https://files.pythonhosted.org/packages/e8/ef/013f07248041b74abd48a385e2110aa3a9bbfef0fbd97d4e6d07d2f5b89a/pydantic_core-2.27.2-cp311-cp311-win_amd64.whl", hash = "sha256:08e125dbdc505fa69ca7d9c499639ab6407cfa909214d500897d02afb816e7cc", size = 1982484 },
+    { url = "https://files.pythonhosted.org/packages/10/1c/16b3a3e3398fd29dca77cea0a1d998d6bde3902fa2706985191e2313cc76/pydantic_core-2.27.2-cp311-cp311-win_arm64.whl", hash = "sha256:26f0d68d4b235a2bae0c3fc585c585b4ecc51382db0e3ba402a22cbc440915e4", size = 1867102 },
+    { url = "https://files.pythonhosted.org/packages/d6/74/51c8a5482ca447871c93e142d9d4a92ead74de6c8dc5e66733e22c9bba89/pydantic_core-2.27.2-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:9e0c8cfefa0ef83b4da9588448b6d8d2a2bf1a53c3f1ae5fca39eb3061e2f0b0", size = 1893127 },
+    { url = "https://files.pythonhosted.org/packages/d3/f3/c97e80721735868313c58b89d2de85fa80fe8dfeeed84dc51598b92a135e/pydantic_core-2.27.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:83097677b8e3bd7eaa6775720ec8e0405f1575015a463285a92bfdfe254529ef", size = 1811340 },
+    { url = "https://files.pythonhosted.org/packages/9e/91/840ec1375e686dbae1bd80a9e46c26a1e0083e1186abc610efa3d9a36180/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:172fce187655fece0c90d90a678424b013f8fbb0ca8b036ac266749c09438cb7", size = 1822900 },
+    { url = "https://files.pythonhosted.org/packages/f6/31/4240bc96025035500c18adc149aa6ffdf1a0062a4b525c932065ceb4d868/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:519f29f5213271eeeeb3093f662ba2fd512b91c5f188f3bb7b27bc5973816934", size = 1869177 },
+    { url = "https://files.pythonhosted.org/packages/fa/20/02fbaadb7808be578317015c462655c317a77a7c8f0ef274bc016a784c54/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:05e3a55d124407fffba0dd6b0c0cd056d10e983ceb4e5dbd10dda135c31071d6", size = 2038046 },
+    { url = "https://files.pythonhosted.org/packages/06/86/7f306b904e6c9eccf0668248b3f272090e49c275bc488a7b88b0823444a4/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9c3ed807c7b91de05e63930188f19e921d1fe90de6b4f5cd43ee7fcc3525cb8c", size = 2685386 },
+    { url = "https://files.pythonhosted.org/packages/8d/f0/49129b27c43396581a635d8710dae54a791b17dfc50c70164866bbf865e3/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6fb4aadc0b9a0c063206846d603b92030eb6f03069151a625667f982887153e2", size = 1997060 },
+    { url = "https://files.pythonhosted.org/packages/0d/0f/943b4af7cd416c477fd40b187036c4f89b416a33d3cc0ab7b82708a667aa/pydantic_core-2.27.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:28ccb213807e037460326424ceb8b5245acb88f32f3d2777427476e1b32c48c4", size = 2004870 },
+    { url = "https://files.pythonhosted.org/packages/35/40/aea70b5b1a63911c53a4c8117c0a828d6790483f858041f47bab0b779f44/pydantic_core-2.27.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:de3cd1899e2c279b140adde9357c4495ed9d47131b4a4eaff9052f23398076b3", size = 1999822 },
+    { url = "https://files.pythonhosted.org/packages/f2/b3/807b94fd337d58effc5498fd1a7a4d9d59af4133e83e32ae39a96fddec9d/pydantic_core-2.27.2-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:220f892729375e2d736b97d0e51466252ad84c51857d4d15f5e9692f9ef12be4", size = 2130364 },
+    { url = "https://files.pythonhosted.org/packages/fc/df/791c827cd4ee6efd59248dca9369fb35e80a9484462c33c6649a8d02b565/pydantic_core-2.27.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:a0fcd29cd6b4e74fe8ddd2c90330fd8edf2e30cb52acda47f06dd615ae72da57", size = 2158303 },
+    { url = "https://files.pythonhosted.org/packages/9b/67/4e197c300976af185b7cef4c02203e175fb127e414125916bf1128b639a9/pydantic_core-2.27.2-cp312-cp312-win32.whl", hash = "sha256:1e2cb691ed9834cd6a8be61228471d0a503731abfb42f82458ff27be7b2186fc", size = 1834064 },
+    { url = "https://files.pythonhosted.org/packages/1f/ea/cd7209a889163b8dcca139fe32b9687dd05249161a3edda62860430457a5/pydantic_core-2.27.2-cp312-cp312-win_amd64.whl", hash = "sha256:cc3f1a99a4f4f9dd1de4fe0312c114e740b5ddead65bb4102884b384c15d8bc9", size = 1989046 },
+    { url = "https://files.pythonhosted.org/packages/bc/49/c54baab2f4658c26ac633d798dab66b4c3a9bbf47cff5284e9c182f4137a/pydantic_core-2.27.2-cp312-cp312-win_arm64.whl", hash = "sha256:3911ac9284cd8a1792d3cb26a2da18f3ca26c6908cc434a18f730dc0db7bfa3b", size = 1885092 },
+    { url = "https://files.pythonhosted.org/packages/41/b1/9bc383f48f8002f99104e3acff6cba1231b29ef76cfa45d1506a5cad1f84/pydantic_core-2.27.2-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:7d14bd329640e63852364c306f4d23eb744e0f8193148d4044dd3dacdaacbd8b", size = 1892709 },
+    { url = "https://files.pythonhosted.org/packages/10/6c/e62b8657b834f3eb2961b49ec8e301eb99946245e70bf42c8817350cbefc/pydantic_core-2.27.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:82f91663004eb8ed30ff478d77c4d1179b3563df6cdb15c0817cd1cdaf34d154", size = 1811273 },
+    { url = "https://files.pythonhosted.org/packages/ba/15/52cfe49c8c986e081b863b102d6b859d9defc63446b642ccbbb3742bf371/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:71b24c7d61131bb83df10cc7e687433609963a944ccf45190cfc21e0887b08c9", size = 1823027 },
+    { url = "https://files.pythonhosted.org/packages/b1/1c/b6f402cfc18ec0024120602bdbcebc7bdd5b856528c013bd4d13865ca473/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:fa8e459d4954f608fa26116118bb67f56b93b209c39b008277ace29937453dc9", size = 1868888 },
+    { url = "https://files.pythonhosted.org/packages/bd/7b/8cb75b66ac37bc2975a3b7de99f3c6f355fcc4d89820b61dffa8f1e81677/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ce8918cbebc8da707ba805b7fd0b382816858728ae7fe19a942080c24e5b7cd1", size = 2037738 },
+    { url = "https://files.pythonhosted.org/packages/c8/f1/786d8fe78970a06f61df22cba58e365ce304bf9b9f46cc71c8c424e0c334/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eda3f5c2a021bbc5d976107bb302e0131351c2ba54343f8a496dc8783d3d3a6a", size = 2685138 },
+    { url = "https://files.pythonhosted.org/packages/a6/74/d12b2cd841d8724dc8ffb13fc5cef86566a53ed358103150209ecd5d1999/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bd8086fa684c4775c27f03f062cbb9eaa6e17f064307e86b21b9e0abc9c0f02e", size = 1997025 },
+    { url = "https://files.pythonhosted.org/packages/a0/6e/940bcd631bc4d9a06c9539b51f070b66e8f370ed0933f392db6ff350d873/pydantic_core-2.27.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:8d9b3388db186ba0c099a6d20f0604a44eabdeef1777ddd94786cdae158729e4", size = 2004633 },
+    { url = "https://files.pythonhosted.org/packages/50/cc/a46b34f1708d82498c227d5d80ce615b2dd502ddcfd8376fc14a36655af1/pydantic_core-2.27.2-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:7a66efda2387de898c8f38c0cf7f14fca0b51a8ef0b24bfea5849f1b3c95af27", size = 1999404 },
+    { url = "https://files.pythonhosted.org/packages/ca/2d/c365cfa930ed23bc58c41463bae347d1005537dc8db79e998af8ba28d35e/pydantic_core-2.27.2-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:18a101c168e4e092ab40dbc2503bdc0f62010e95d292b27827871dc85450d7ee", size = 2130130 },
+    { url = "https://files.pythonhosted.org/packages/f4/d7/eb64d015c350b7cdb371145b54d96c919d4db516817f31cd1c650cae3b21/pydantic_core-2.27.2-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:ba5dd002f88b78a4215ed2f8ddbdf85e8513382820ba15ad5ad8955ce0ca19a1", size = 2157946 },
+    { url = "https://files.pythonhosted.org/packages/a4/99/bddde3ddde76c03b65dfd5a66ab436c4e58ffc42927d4ff1198ffbf96f5f/pydantic_core-2.27.2-cp313-cp313-win32.whl", hash = "sha256:1ebaf1d0481914d004a573394f4be3a7616334be70261007e47c2a6fe7e50130", size = 1834387 },
+    { url = "https://files.pythonhosted.org/packages/71/47/82b5e846e01b26ac6f1893d3c5f9f3a2eb6ba79be26eef0b759b4fe72946/pydantic_core-2.27.2-cp313-cp313-win_amd64.whl", hash = "sha256:953101387ecf2f5652883208769a79e48db18c6df442568a0b5ccd8c2723abee", size = 1990453 },
+    { url = "https://files.pythonhosted.org/packages/51/b2/b2b50d5ecf21acf870190ae5d093602d95f66c9c31f9d5de6062eb329ad1/pydantic_core-2.27.2-cp313-cp313-win_arm64.whl", hash = "sha256:ac4dbfd1691affb8f48c2c13241a2e3b60ff23247cbcf981759c768b6633cf8b", size = 1885186 },
+    { url = "https://files.pythonhosted.org/packages/27/97/3aef1ddb65c5ccd6eda9050036c956ff6ecbfe66cb7eb40f280f121a5bb0/pydantic_core-2.27.2-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:c10eb4f1659290b523af58fa7cffb452a61ad6ae5613404519aee4bfbf1df993", size = 1896475 },
+    { url = "https://files.pythonhosted.org/packages/ad/d3/5668da70e373c9904ed2f372cb52c0b996426f302e0dee2e65634c92007d/pydantic_core-2.27.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:ef592d4bad47296fb11f96cd7dc898b92e795032b4894dfb4076cfccd43a9308", size = 1772279 },
+    { url = "https://files.pythonhosted.org/packages/8a/9e/e44b8cb0edf04a2f0a1f6425a65ee089c1d6f9c4c2dcab0209127b6fdfc2/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c61709a844acc6bf0b7dce7daae75195a10aac96a596ea1b776996414791ede4", size = 1829112 },
+    { url = "https://files.pythonhosted.org/packages/1c/90/1160d7ac700102effe11616e8119e268770f2a2aa5afb935f3ee6832987d/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:42c5f762659e47fdb7b16956c71598292f60a03aa92f8b6351504359dbdba6cf", size = 1866780 },
+    { url = "https://files.pythonhosted.org/packages/ee/33/13983426df09a36d22c15980008f8d9c77674fc319351813b5a2739b70f3/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4c9775e339e42e79ec99c441d9730fccf07414af63eac2f0e48e08fd38a64d76", size = 2037943 },
+    { url = "https://files.pythonhosted.org/packages/01/d7/ced164e376f6747e9158c89988c293cd524ab8d215ae4e185e9929655d5c/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:57762139821c31847cfb2df63c12f725788bd9f04bc2fb392790959b8f70f118", size = 2740492 },
+    { url = "https://files.pythonhosted.org/packages/8b/1f/3dc6e769d5b7461040778816aab2b00422427bcaa4b56cc89e9c653b2605/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0d1e85068e818c73e048fe28cfc769040bb1f475524f4745a5dc621f75ac7630", size = 1995714 },
+    { url = "https://files.pythonhosted.org/packages/07/d7/a0bd09bc39283530b3f7c27033a814ef254ba3bd0b5cfd040b7abf1fe5da/pydantic_core-2.27.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:097830ed52fd9e427942ff3b9bc17fab52913b2f50f2880dc4a5611446606a54", size = 1997163 },
+    { url = "https://files.pythonhosted.org/packages/2d/bb/2db4ad1762e1c5699d9b857eeb41959191980de6feb054e70f93085e1bcd/pydantic_core-2.27.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:044a50963a614ecfae59bb1eaf7ea7efc4bc62f49ed594e18fa1e5d953c40e9f", size = 2005217 },
+    { url = "https://files.pythonhosted.org/packages/53/5f/23a5a3e7b8403f8dd8fc8a6f8b49f6b55c7d715b77dcf1f8ae919eeb5628/pydantic_core-2.27.2-cp39-cp39-musllinux_1_1_armv7l.whl", hash = "sha256:4e0b4220ba5b40d727c7f879eac379b822eee5d8fff418e9d3381ee45b3b0362", size = 2127899 },
+    { url = "https://files.pythonhosted.org/packages/c2/ae/aa38bb8dd3d89c2f1d8362dd890ee8f3b967330821d03bbe08fa01ce3766/pydantic_core-2.27.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:5e4f4bb20d75e9325cc9696c6802657b58bc1dbbe3022f32cc2b2b632c3fbb96", size = 2155726 },
+    { url = "https://files.pythonhosted.org/packages/98/61/4f784608cc9e98f70839187117ce840480f768fed5d386f924074bf6213c/pydantic_core-2.27.2-cp39-cp39-win32.whl", hash = "sha256:cca63613e90d001b9f2f9a9ceb276c308bfa2a43fafb75c8031c4f66039e8c6e", size = 1817219 },
+    { url = "https://files.pythonhosted.org/packages/57/82/bb16a68e4a1a858bb3768c2c8f1ff8d8978014e16598f001ea29a25bf1d1/pydantic_core-2.27.2-cp39-cp39-win_amd64.whl", hash = "sha256:77d1bca19b0f7021b3a982e6f903dcd5b2b06076def36a652e3907f596e29f67", size = 1985382 },
+    { url = "https://files.pythonhosted.org/packages/46/72/af70981a341500419e67d5cb45abe552a7c74b66326ac8877588488da1ac/pydantic_core-2.27.2-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:2bf14caea37e91198329b828eae1618c068dfb8ef17bb33287a7ad4b61ac314e", size = 1891159 },
+    { url = "https://files.pythonhosted.org/packages/ad/3d/c5913cccdef93e0a6a95c2d057d2c2cba347815c845cda79ddd3c0f5e17d/pydantic_core-2.27.2-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:b0cb791f5b45307caae8810c2023a184c74605ec3bcbb67d13846c28ff731ff8", size = 1768331 },
+    { url = "https://files.pythonhosted.org/packages/f6/f0/a3ae8fbee269e4934f14e2e0e00928f9346c5943174f2811193113e58252/pydantic_core-2.27.2-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:688d3fd9fcb71f41c4c015c023d12a79d1c4c0732ec9eb35d96e3388a120dcf3", size = 1822467 },
+    { url = "https://files.pythonhosted.org/packages/d7/7a/7bbf241a04e9f9ea24cd5874354a83526d639b02674648af3f350554276c/pydantic_core-2.27.2-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3d591580c34f4d731592f0e9fe40f9cc1b430d297eecc70b962e93c5c668f15f", size = 1979797 },
+    { url = "https://files.pythonhosted.org/packages/4f/5f/4784c6107731f89e0005a92ecb8a2efeafdb55eb992b8e9d0a2be5199335/pydantic_core-2.27.2-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:82f986faf4e644ffc189a7f1aafc86e46ef70372bb153e7001e8afccc6e54133", size = 1987839 },
+    { url = "https://files.pythonhosted.org/packages/6d/a7/61246562b651dff00de86a5f01b6e4befb518df314c54dec187a78d81c84/pydantic_core-2.27.2-pp310-pypy310_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:bec317a27290e2537f922639cafd54990551725fc844249e64c523301d0822fc", size = 1998861 },
+    { url = "https://files.pythonhosted.org/packages/86/aa/837821ecf0c022bbb74ca132e117c358321e72e7f9702d1b6a03758545e2/pydantic_core-2.27.2-pp310-pypy310_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:0296abcb83a797db256b773f45773da397da75a08f5fcaef41f2044adec05f50", size = 2116582 },
+    { url = "https://files.pythonhosted.org/packages/81/b0/5e74656e95623cbaa0a6278d16cf15e10a51f6002e3ec126541e95c29ea3/pydantic_core-2.27.2-pp310-pypy310_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:0d75070718e369e452075a6017fbf187f788e17ed67a3abd47fa934d001863d9", size = 2151985 },
+    { url = "https://files.pythonhosted.org/packages/63/37/3e32eeb2a451fddaa3898e2163746b0cffbbdbb4740d38372db0490d67f3/pydantic_core-2.27.2-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:7e17b560be3c98a8e3aa66ce828bdebb9e9ac6ad5466fba92eb74c4c95cb1151", size = 2004715 },
+    { url = "https://files.pythonhosted.org/packages/29/0e/dcaea00c9dbd0348b723cae82b0e0c122e0fa2b43fa933e1622fd237a3ee/pydantic_core-2.27.2-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:c33939a82924da9ed65dab5a65d427205a73181d8098e79b6b426bdf8ad4e656", size = 1891733 },
+    { url = "https://files.pythonhosted.org/packages/86/d3/e797bba8860ce650272bda6383a9d8cad1d1c9a75a640c9d0e848076f85e/pydantic_core-2.27.2-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:00bad2484fa6bda1e216e7345a798bd37c68fb2d97558edd584942aa41b7d278", size = 1768375 },
+    { url = "https://files.pythonhosted.org/packages/41/f7/f847b15fb14978ca2b30262548f5fc4872b2724e90f116393eb69008299d/pydantic_core-2.27.2-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c817e2b40aba42bac6f457498dacabc568c3b7a986fc9ba7c8d9d260b71485fb", size = 1822307 },
+    { url = "https://files.pythonhosted.org/packages/9c/63/ed80ec8255b587b2f108e514dc03eed1546cd00f0af281e699797f373f38/pydantic_core-2.27.2-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:251136cdad0cb722e93732cb45ca5299fb56e1344a833640bf93b2803f8d1bfd", size = 1979971 },
+    { url = "https://files.pythonhosted.org/packages/a9/6d/6d18308a45454a0de0e975d70171cadaf454bc7a0bf86b9c7688e313f0bb/pydantic_core-2.27.2-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:d2088237af596f0a524d3afc39ab3b036e8adb054ee57cbb1dcf8e09da5b29cc", size = 1987616 },
+    { url = "https://files.pythonhosted.org/packages/82/8a/05f8780f2c1081b800a7ca54c1971e291c2d07d1a50fb23c7e4aef4ed403/pydantic_core-2.27.2-pp39-pypy39_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:d4041c0b966a84b4ae7a09832eb691a35aec90910cd2dbe7a208de59be77965b", size = 1998943 },
+    { url = "https://files.pythonhosted.org/packages/5e/3e/fe5b6613d9e4c0038434396b46c5303f5ade871166900b357ada4766c5b7/pydantic_core-2.27.2-pp39-pypy39_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:8083d4e875ebe0b864ffef72a4304827015cff328a1be6e22cc850753bfb122b", size = 2116654 },
+    { url = "https://files.pythonhosted.org/packages/db/ad/28869f58938fad8cc84739c4e592989730bfb69b7c90a8fff138dff18e1e/pydantic_core-2.27.2-pp39-pypy39_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:f141ee28a0ad2123b6611b6ceff018039df17f32ada8b534e6aa039545a3efb2", size = 2152292 },
+    { url = "https://files.pythonhosted.org/packages/a1/0c/c5c5cd3689c32ed1fe8c5d234b079c12c281c051759770c05b8bed6412b5/pydantic_core-2.27.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:7d0c8399fcc1848491f00e0314bd59fb34a9c008761bcb422a057670c3f65e35", size = 2004961 },
+]
+
+[[package]]
+name = "pygments"
+version = "2.19.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7c/2d/c3338d48ea6cc0feb8446d8e6937e1408088a72a39937982cc6111d17f84/pygments-2.19.1.tar.gz", hash = "sha256:61c16d2a8576dc0649d9f39e089b5f02bcd27fba10d8fb4dcc28173f7a45151f", size = 4968581 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8a/0b/9fcc47d19c48b59121088dd6da2488a49d5f72dacf8262e2790a1d2c7d15/pygments-2.19.1-py3-none-any.whl", hash = "sha256:9ea1544ad55cecf4b8242fab6dd35a93bbce657034b0611ee383099054ab6d8c", size = 1225293 },
+]
+
+[[package]]
+name = "pytest"
+version = "8.3.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "exceptiongroup", marker = "python_full_version < '3.11'" },
+    { name = "iniconfig" },
+    { name = "packaging" },
+    { name = "pluggy" },
+    { name = "tomli", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/05/35/30e0d83068951d90a01852cb1cef56e5d8a09d20c7f511634cc2f7e0372a/pytest-8.3.4.tar.gz", hash = "sha256:965370d062bce11e73868e0335abac31b4d3de0e82f4007408d242b4f8610761", size = 1445919 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/11/92/76a1c94d3afee238333bc0a42b82935dd8f9cf8ce9e336ff87ee14d9e1cf/pytest-8.3.4-py3-none-any.whl", hash = "sha256:50e16d954148559c9a74109af1eaf0c945ba2d8f30f0a3d3335edde19788b6f6", size = 343083 },
+]
+
+[[package]]
+name = "pytest-benchmark"
+version = "5.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "py-cpuinfo" },
+    { name = "pytest" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/39/d0/a8bd08d641b393db3be3819b03e2d9bb8760ca8479080a26a5f6e540e99c/pytest-benchmark-5.1.0.tar.gz", hash = "sha256:9ea661cdc292e8231f7cd4c10b0319e56a2118e2c09d9f50e1b3d150d2aca105", size = 337810 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9e/d6/b41653199ea09d5969d4e385df9bbfd9a100f28ca7e824ce7c0a016e3053/pytest_benchmark-5.1.0-py3-none-any.whl", hash = "sha256:922de2dfa3033c227c96da942d1878191afa135a29485fb942e85dff1c592c89", size = 44259 },
+]
+
+[[package]]
+name = "pytest-cov"
+version = "6.0.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "coverage", extra = ["toml"] },
+    { name = "pytest" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/be/45/9b538de8cef30e17c7b45ef42f538a94889ed6a16f2387a6c89e73220651/pytest-cov-6.0.0.tar.gz", hash = "sha256:fde0b595ca248bb8e2d76f020b465f3b107c9632e6a1d1705f17834c89dcadc0", size = 66945 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/36/3b/48e79f2cd6a61dbbd4807b4ed46cb564b4fd50a76166b1c4ea5c1d9e2371/pytest_cov-6.0.0-py3-none-any.whl", hash = "sha256:eee6f1b9e61008bd34975a4d5bab25801eb31898b032dd55addc93e96fcaaa35", size = 22949 },
+]
+
+[[package]]
+name = "pytest-mock"
+version = "3.14.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pytest" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/c6/90/a955c3ab35ccd41ad4de556596fa86685bf4fc5ffcc62d22d856cfd4e29a/pytest-mock-3.14.0.tar.gz", hash = "sha256:2719255a1efeceadbc056d6bf3df3d1c5015530fb40cf347c0f9afac88410bd0", size = 32814 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f2/3b/b26f90f74e2986a82df6e7ac7e319b8ea7ccece1caec9f8ab6104dc70603/pytest_mock-3.14.0-py3-none-any.whl", hash = "sha256:0b72c38033392a5f4621342fe11e9219ac11ec9d375f8e2a0c164539e0d70f6f", size = 9863 },
+]
+
+[[package]]
+name = "python-dateutil"
+version = "2.9.0.post0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "six" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/66/c0/0c8b6ad9f17a802ee498c46e004a0eb49bc148f2fd230864601a86dcf6db/python-dateutil-2.9.0.post0.tar.gz", hash = "sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3", size = 342432 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892 },
+]
+
+[[package]]
+name = "python-dotenv"
+version = "1.0.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/bc/57/e84d88dfe0aec03b7a2d4327012c1627ab5f03652216c63d49846d7a6c58/python-dotenv-1.0.1.tar.gz", hash = "sha256:e324ee90a023d808f1959c46bcbc04446a10ced277783dc6ee09987c37ec10ca", size = 39115 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6a/3e/b68c118422ec867fa7ab88444e1274aa40681c606d59ac27de5a5588f082/python_dotenv-1.0.1-py3-none-any.whl", hash = "sha256:f7b63ef50f1b690dddf550d03497b66d609393b40b564ed0d674909a68ebf16a", size = 19863 },
+]
+
+[[package]]
+name = "pytz"
+version = "2024.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3a/31/3c70bf7603cc2dca0f19bdc53b4537a797747a58875b552c8c413d963a3f/pytz-2024.2.tar.gz", hash = "sha256:2aa355083c50a0f93fa581709deac0c9ad65cca8a9e9beac660adcbd493c798a", size = 319692 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/11/c3/005fcca25ce078d2cc29fd559379817424e94885510568bc1bc53d7d5846/pytz-2024.2-py2.py3-none-any.whl", hash = "sha256:31c7c1817eb7fae7ca4b8c7ee50c72f93aa2dd863de768e1ef4245d426aa0725", size = 508002 },
+]
+
+[[package]]
+name = "pyyaml"
+version = "6.0.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/54/ed/79a089b6be93607fa5cdaedf301d7dfb23af5f25c398d5ead2525b063e17/pyyaml-6.0.2.tar.gz", hash = "sha256:d584d9ec91ad65861cc08d42e834324ef890a082e591037abe114850ff7bbc3e", size = 130631 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9b/95/a3fac87cb7158e231b5a6012e438c647e1a87f09f8e0d123acec8ab8bf71/PyYAML-6.0.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0a9a2848a5b7feac301353437eb7d5957887edbf81d56e903999a75a3d743086", size = 184199 },
+    { url = "https://files.pythonhosted.org/packages/c7/7a/68bd47624dab8fd4afbfd3c48e3b79efe09098ae941de5b58abcbadff5cb/PyYAML-6.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:29717114e51c84ddfba879543fb232a6ed60086602313ca38cce623c1d62cfbf", size = 171758 },
+    { url = "https://files.pythonhosted.org/packages/49/ee/14c54df452143b9ee9f0f29074d7ca5516a36edb0b4cc40c3f280131656f/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8824b5a04a04a047e72eea5cec3bc266db09e35de6bdfe34c9436ac5ee27d237", size = 718463 },
+    { url = "https://files.pythonhosted.org/packages/4d/61/de363a97476e766574650d742205be468921a7b532aa2499fcd886b62530/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7c36280e6fb8385e520936c3cb3b8042851904eba0e58d277dca80a5cfed590b", size = 719280 },
+    { url = "https://files.pythonhosted.org/packages/6b/4e/1523cb902fd98355e2e9ea5e5eb237cbc5f3ad5f3075fa65087aa0ecb669/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ec031d5d2feb36d1d1a24380e4db6d43695f3748343d99434e6f5f9156aaa2ed", size = 751239 },
+    { url = "https://files.pythonhosted.org/packages/b7/33/5504b3a9a4464893c32f118a9cc045190a91637b119a9c881da1cf6b7a72/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:936d68689298c36b53b29f23c6dbb74de12b4ac12ca6cfe0e047bedceea56180", size = 695802 },
+    { url = "https://files.pythonhosted.org/packages/5c/20/8347dcabd41ef3a3cdc4f7b7a2aff3d06598c8779faa189cdbf878b626a4/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:23502f431948090f597378482b4812b0caae32c22213aecf3b55325e049a6c68", size = 720527 },
+    { url = "https://files.pythonhosted.org/packages/be/aa/5afe99233fb360d0ff37377145a949ae258aaab831bde4792b32650a4378/PyYAML-6.0.2-cp310-cp310-win32.whl", hash = "sha256:2e99c6826ffa974fe6e27cdb5ed0021786b03fc98e5ee3c5bfe1fd5015f42b99", size = 144052 },
+    { url = "https://files.pythonhosted.org/packages/b5/84/0fa4b06f6d6c958d207620fc60005e241ecedceee58931bb20138e1e5776/PyYAML-6.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:a4d3091415f010369ae4ed1fc6b79def9416358877534caf6a0fdd2146c87a3e", size = 161774 },
+    { url = "https://files.pythonhosted.org/packages/f8/aa/7af4e81f7acba21a4c6be026da38fd2b872ca46226673c89a758ebdc4fd2/PyYAML-6.0.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:cc1c1159b3d456576af7a3e4d1ba7e6924cb39de8f67111c735f6fc832082774", size = 184612 },
+    { url = "https://files.pythonhosted.org/packages/8b/62/b9faa998fd185f65c1371643678e4d58254add437edb764a08c5a98fb986/PyYAML-6.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1e2120ef853f59c7419231f3bf4e7021f1b936f6ebd222406c3b60212205d2ee", size = 172040 },
+    { url = "https://files.pythonhosted.org/packages/ad/0c/c804f5f922a9a6563bab712d8dcc70251e8af811fce4524d57c2c0fd49a4/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5d225db5a45f21e78dd9358e58a98702a0302f2659a3c6cd320564b75b86f47c", size = 736829 },
+    { url = "https://files.pythonhosted.org/packages/51/16/6af8d6a6b210c8e54f1406a6b9481febf9c64a3109c541567e35a49aa2e7/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5ac9328ec4831237bec75defaf839f7d4564be1e6b25ac710bd1a96321cc8317", size = 764167 },
+    { url = "https://files.pythonhosted.org/packages/75/e4/2c27590dfc9992f73aabbeb9241ae20220bd9452df27483b6e56d3975cc5/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3ad2a3decf9aaba3d29c8f537ac4b243e36bef957511b4766cb0057d32b0be85", size = 762952 },
+    { url = "https://files.pythonhosted.org/packages/9b/97/ecc1abf4a823f5ac61941a9c00fe501b02ac3ab0e373c3857f7d4b83e2b6/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:ff3824dc5261f50c9b0dfb3be22b4567a6f938ccce4587b38952d85fd9e9afe4", size = 735301 },
+    { url = "https://files.pythonhosted.org/packages/45/73/0f49dacd6e82c9430e46f4a027baa4ca205e8b0a9dce1397f44edc23559d/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:797b4f722ffa07cc8d62053e4cff1486fa6dc094105d13fea7b1de7d8bf71c9e", size = 756638 },
+    { url = "https://files.pythonhosted.org/packages/22/5f/956f0f9fc65223a58fbc14459bf34b4cc48dec52e00535c79b8db361aabd/PyYAML-6.0.2-cp311-cp311-win32.whl", hash = "sha256:11d8f3dd2b9c1207dcaf2ee0bbbfd5991f571186ec9cc78427ba5bd32afae4b5", size = 143850 },
+    { url = "https://files.pythonhosted.org/packages/ed/23/8da0bbe2ab9dcdd11f4f4557ccaf95c10b9811b13ecced089d43ce59c3c8/PyYAML-6.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:e10ce637b18caea04431ce14fabcf5c64a1c61ec9c56b071a4b7ca131ca52d44", size = 161980 },
+    { url = "https://files.pythonhosted.org/packages/86/0c/c581167fc46d6d6d7ddcfb8c843a4de25bdd27e4466938109ca68492292c/PyYAML-6.0.2-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:c70c95198c015b85feafc136515252a261a84561b7b1d51e3384e0655ddf25ab", size = 183873 },
+    { url = "https://files.pythonhosted.org/packages/a8/0c/38374f5bb272c051e2a69281d71cba6fdb983413e6758b84482905e29a5d/PyYAML-6.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce826d6ef20b1bc864f0a68340c8b3287705cae2f8b4b1d932177dcc76721725", size = 173302 },
+    { url = "https://files.pythonhosted.org/packages/c3/93/9916574aa8c00aa06bbac729972eb1071d002b8e158bd0e83a3b9a20a1f7/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1f71ea527786de97d1a0cc0eacd1defc0985dcf6b3f17bb77dcfc8c34bec4dc5", size = 739154 },
+    { url = "https://files.pythonhosted.org/packages/95/0f/b8938f1cbd09739c6da569d172531567dbcc9789e0029aa070856f123984/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9b22676e8097e9e22e36d6b7bda33190d0d400f345f23d4065d48f4ca7ae0425", size = 766223 },
+    { url = "https://files.pythonhosted.org/packages/b9/2b/614b4752f2e127db5cc206abc23a8c19678e92b23c3db30fc86ab731d3bd/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:80bab7bfc629882493af4aa31a4cfa43a4c57c83813253626916b8c7ada83476", size = 767542 },
+    { url = "https://files.pythonhosted.org/packages/d4/00/dd137d5bcc7efea1836d6264f049359861cf548469d18da90cd8216cf05f/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:0833f8694549e586547b576dcfaba4a6b55b9e96098b36cdc7ebefe667dfed48", size = 731164 },
+    { url = "https://files.pythonhosted.org/packages/c9/1f/4f998c900485e5c0ef43838363ba4a9723ac0ad73a9dc42068b12aaba4e4/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:8b9c7197f7cb2738065c481a0461e50ad02f18c78cd75775628afb4d7137fb3b", size = 756611 },
+    { url = "https://files.pythonhosted.org/packages/df/d1/f5a275fdb252768b7a11ec63585bc38d0e87c9e05668a139fea92b80634c/PyYAML-6.0.2-cp312-cp312-win32.whl", hash = "sha256:ef6107725bd54b262d6dedcc2af448a266975032bc85ef0172c5f059da6325b4", size = 140591 },
+    { url = "https://files.pythonhosted.org/packages/0c/e8/4f648c598b17c3d06e8753d7d13d57542b30d56e6c2dedf9c331ae56312e/PyYAML-6.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:7e7401d0de89a9a855c839bc697c079a4af81cf878373abd7dc625847d25cbd8", size = 156338 },
+    { url = "https://files.pythonhosted.org/packages/ef/e3/3af305b830494fa85d95f6d95ef7fa73f2ee1cc8ef5b495c7c3269fb835f/PyYAML-6.0.2-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:efdca5630322a10774e8e98e1af481aad470dd62c3170801852d752aa7a783ba", size = 181309 },
+    { url = "https://files.pythonhosted.org/packages/45/9f/3b1c20a0b7a3200524eb0076cc027a970d320bd3a6592873c85c92a08731/PyYAML-6.0.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:50187695423ffe49e2deacb8cd10510bc361faac997de9efef88badc3bb9e2d1", size = 171679 },
+    { url = "https://files.pythonhosted.org/packages/7c/9a/337322f27005c33bcb656c655fa78325b730324c78620e8328ae28b64d0c/PyYAML-6.0.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0ffe8360bab4910ef1b9e87fb812d8bc0a308b0d0eef8c8f44e0254ab3b07133", size = 733428 },
+    { url = "https://files.pythonhosted.org/packages/a3/69/864fbe19e6c18ea3cc196cbe5d392175b4cf3d5d0ac1403ec3f2d237ebb5/PyYAML-6.0.2-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:17e311b6c678207928d649faa7cb0d7b4c26a0ba73d41e99c4fff6b6c3276484", size = 763361 },
+    { url = "https://files.pythonhosted.org/packages/04/24/b7721e4845c2f162d26f50521b825fb061bc0a5afcf9a386840f23ea19fa/PyYAML-6.0.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:70b189594dbe54f75ab3a1acec5f1e3faa7e8cf2f1e08d9b561cb41b845f69d5", size = 759523 },
+    { url = "https://files.pythonhosted.org/packages/2b/b2/e3234f59ba06559c6ff63c4e10baea10e5e7df868092bf9ab40e5b9c56b6/PyYAML-6.0.2-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:41e4e3953a79407c794916fa277a82531dd93aad34e29c2a514c2c0c5fe971cc", size = 726660 },
+    { url = "https://files.pythonhosted.org/packages/fe/0f/25911a9f080464c59fab9027482f822b86bf0608957a5fcc6eaac85aa515/PyYAML-6.0.2-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:68ccc6023a3400877818152ad9a1033e3db8625d899c72eacb5a668902e4d652", size = 751597 },
+    { url = "https://files.pythonhosted.org/packages/14/0d/e2c3b43bbce3cf6bd97c840b46088a3031085179e596d4929729d8d68270/PyYAML-6.0.2-cp313-cp313-win32.whl", hash = "sha256:bc2fa7c6b47d6bc618dd7fb02ef6fdedb1090ec036abab80d4681424b84c1183", size = 140527 },
+    { url = "https://files.pythonhosted.org/packages/fa/de/02b54f42487e3d3c6efb3f89428677074ca7bf43aae402517bc7cca949f3/PyYAML-6.0.2-cp313-cp313-win_amd64.whl", hash = "sha256:8388ee1976c416731879ac16da0aff3f63b286ffdd57cdeb95f3f2e085687563", size = 156446 },
+    { url = "https://files.pythonhosted.org/packages/65/d8/b7a1db13636d7fb7d4ff431593c510c8b8fca920ade06ca8ef20015493c5/PyYAML-6.0.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:688ba32a1cffef67fd2e9398a2efebaea461578b0923624778664cc1c914db5d", size = 184777 },
+    { url = "https://files.pythonhosted.org/packages/0a/02/6ec546cd45143fdf9840b2c6be8d875116a64076218b61d68e12548e5839/PyYAML-6.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a8786accb172bd8afb8be14490a16625cbc387036876ab6ba70912730faf8e1f", size = 172318 },
+    { url = "https://files.pythonhosted.org/packages/0e/9a/8cc68be846c972bda34f6c2a93abb644fb2476f4dcc924d52175786932c9/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d8e03406cac8513435335dbab54c0d385e4a49e4945d2909a581c83647ca0290", size = 720891 },
+    { url = "https://files.pythonhosted.org/packages/e9/6c/6e1b7f40181bc4805e2e07f4abc10a88ce4648e7e95ff1abe4ae4014a9b2/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f753120cb8181e736c57ef7636e83f31b9c0d1722c516f7e86cf15b7aa57ff12", size = 722614 },
+    { url = "https://files.pythonhosted.org/packages/3d/32/e7bd8535d22ea2874cef6a81021ba019474ace0d13a4819c2a4bce79bd6a/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3b1fdb9dc17f5a7677423d508ab4f243a726dea51fa5e70992e59a7411c89d19", size = 737360 },
+    { url = "https://files.pythonhosted.org/packages/d7/12/7322c1e30b9be969670b672573d45479edef72c9a0deac3bb2868f5d7469/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:0b69e4ce7a131fe56b7e4d770c67429700908fc0752af059838b1cfb41960e4e", size = 699006 },
+    { url = "https://files.pythonhosted.org/packages/82/72/04fcad41ca56491995076630c3ec1e834be241664c0c09a64c9a2589b507/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a9f8c2e67970f13b16084e04f134610fd1d374bf477b17ec1599185cf611d725", size = 723577 },
+    { url = "https://files.pythonhosted.org/packages/ed/5e/46168b1f2757f1fcd442bc3029cd8767d88a98c9c05770d8b420948743bb/PyYAML-6.0.2-cp39-cp39-win32.whl", hash = "sha256:6395c297d42274772abc367baaa79683958044e5d3835486c16da75d2a694631", size = 144593 },
+    { url = "https://files.pythonhosted.org/packages/19/87/5124b1c1f2412bb95c59ec481eaf936cd32f0fe2a7b16b97b81c4c017a6a/PyYAML-6.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:39693e1f8320ae4f43943590b49779ffb98acb81f788220ea932a6b6c51004d8", size = 162312 },
+]
+
+[[package]]
+name = "ray"
+version = "2.40.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "aiosignal" },
+    { name = "click" },
+    { name = "filelock" },
+    { name = "frozenlist" },
+    { name = "jsonschema" },
+    { name = "msgpack" },
+    { name = "packaging" },
+    { name = "protobuf" },
+    { name = "pyyaml" },
+    { name = "requests" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ca/42/492dc35c112c5adbcf258066d739f897f484b6e2aff29b28dd9ebc9832d4/ray-2.40.0-cp310-cp310-macosx_10_15_x86_64.whl", hash = "sha256:064af8bc52cc988c82470b8e76e5df417737fa7c1d87f597a892c69eb4ec3caa", size = 67061821 },
+    { url = "https://files.pythonhosted.org/packages/07/c2/0847df9d81524ceafe0fafcddce9e4e4799501807687759d2c48f1b34f43/ray-2.40.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:45beb4019cd20b6cb10572d8012c771bccd623f544a669da6797ccf993c4bb33", size = 64422550 },
+    { url = "https://files.pythonhosted.org/packages/62/e7/cf468bfb109d904cc5b7650d890f7c4284842d72acf65b1ae05a5c02c2e5/ray-2.40.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:6cede5fbf7de4fae22cebe2c6977aaf3c85fde6f7de2aa10c46992cf24ea8bda", size = 65916634 },
+    { url = "https://files.pythonhosted.org/packages/31/73/6763d7f87756816698fa5f841ba590a869e821a46c0a98d848055ea831bb/ray-2.40.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:f6eab11dc8490f88e78e06aa645905b259cde1fa03b15e8426155c4782ba0bbe", size = 66845129 },
+    { url = "https://files.pythonhosted.org/packages/1f/cd/0107c782d8511ae100a49da95dc1951c1da67a467e204671ae730f6a5401/ray-2.40.0-cp310-cp310-win_amd64.whl", hash = "sha256:f83cda1ecceb7abe021cd377f0c503596f26d2d66cdff13c1089a06c8b780c23", size = 25295665 },
+    { url = "https://files.pythonhosted.org/packages/a9/fc/811e88c982ea755c6351bd5f2c410cd8526695069a708d9aa3c9410a8dc3/ray-2.40.0-cp311-cp311-macosx_10_15_x86_64.whl", hash = "sha256:dac89bb2cb889c19549a4ac0383492e7550f3e63b78b629a3118e8b91e4e82f3", size = 67000351 },
+    { url = "https://files.pythonhosted.org/packages/72/fb/f048a8580d97429ab3b60b13844500132e756ceaa5bb643087a0d46d9c9b/ray-2.40.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3e4efdf8aebff6e71391c2d5dd66bb45835f2d6d629ac03a3e21e2d4283e2311", size = 64359351 },
+    { url = "https://files.pythonhosted.org/packages/99/34/e16767825432cf7553badba349c80344f4a86b227e5ff2ef4ae20ea40b6c/ray-2.40.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:c776f131e5d0a169a98ab8021c5796f52bf48fcfc6c44ffbd2a9d090fe10748a", size = 66058000 },
+    { url = "https://files.pythonhosted.org/packages/55/4e/a46f514a574c33f699f729e26866475bd725287cf9ba3adf55883d6f809d/ray-2.40.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:71711cbf2c156213fd49b0f9cc93180a7ba424110070a34bdea3dc09527f31df", size = 66967160 },
+    { url = "https://files.pythonhosted.org/packages/9b/fe/01adb0316658feab19ee48badb3dd232ae681633f47b2d6c191c4df14ebe/ray-2.40.0-cp311-cp311-win_amd64.whl", hash = "sha256:532321132618983366e39aeb4cc7867cf7241b0b1e49ee44b01d2aee9923e422", size = 25236232 },
+    { url = "https://files.pythonhosted.org/packages/86/01/994daffa3516e582cd8eacbe958938d2128b79284f2387ff6ebc53b970e2/ray-2.40.0-cp312-cp312-macosx_10_15_x86_64.whl", hash = "sha256:6992922fe91a90b5cc97d9f05ca51b64d72cd644db7ad55caa936be9a6098cce", size = 66981507 },
+    { url = "https://files.pythonhosted.org/packages/46/ce/bcf8416ce4137ce4bc7e0ebdedb2200a598f7beeca1a1004bc45478bb4b0/ray-2.40.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:28329e7a7471610a475d3bb09a4c1b31abcf3596cee25c4254f8d01ad161ba84", size = 64345984 },
+    { url = "https://files.pythonhosted.org/packages/8a/96/dbcb31dd8ff74b81dbfe9332f49941c716daffaffdc3efb1c30ca5a1b4cd/ray-2.40.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:8ea05221fa48e32c652c29498d320e90134b3a012421006af98965097dd1cc3b", size = 66071071 },
+    { url = "https://files.pythonhosted.org/packages/82/ad/2eaf308c64e39704629863720c59b82e1ceed814fca319b8585deb92bb74/ray-2.40.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:674755814f5692306c554cadbc24015af823dc0516e34bdef24ccac9d7a656e3", size = 67018887 },
+    { url = "https://files.pythonhosted.org/packages/21/c4/23616495341e01c97d5b72d07555ab8908e87190cf2f53b4f029cb65bcf0/ray-2.40.0-cp312-cp312-win_amd64.whl", hash = "sha256:bbc01d773cbc43e3efa462ec28ee4c0cacc50f098078332fb45b1ab38eaf9b5d", size = 25223820 },
+    { url = "https://files.pythonhosted.org/packages/44/6f/ecaf4363063c1a1f7e5d611083e459919c33815b32008c25eeb853b8da1f/ray-2.40.0-cp39-cp39-macosx_10_15_x86_64.whl", hash = "sha256:27292bf8921dd69757e7581644afcd3ccae13d6f10f3841f5523ae82b6612f4b", size = 67076882 },
+    { url = "https://files.pythonhosted.org/packages/5c/7f/d6da83e7e2357770f482f9f521d15b46ea13116aeb0ba0569e3986d11658/ray-2.40.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2b74ca43d0c4ccdcaefbf1e7d26aabb1c0d20f825688a9fd7134ba918bda8442", size = 64436021 },
+    { url = "https://files.pythonhosted.org/packages/87/79/95634cfe3fd0457a662e24967391597a5074b5aecfab8c453221481c5dfa/ray-2.40.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:5eb7a203f58defedff0dc53f78a4e1431d040b2b8458548704979c0113f3b892", size = 65938206 },
+    { url = "https://files.pythonhosted.org/packages/1f/c9/09de884f8e62efe0251f8915380f322d92e49f7f7e471c2f15471c40b831/ray-2.40.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:a36a20a3b936b36d14fab031222f92e3c5e731d7db6bb183ca4fba6d0ce3f52a", size = 66864869 },
+    { url = "https://files.pythonhosted.org/packages/ae/57/6742621756a7bce8ac99f9d30e2e7461e1d58716d5fd274a0a88657abd60/ray-2.40.0-cp39-cp39-win_amd64.whl", hash = "sha256:fbe9cd3e076dea676afd57caf19b2897a67ecdf14a542c03864800966cf2aec9", size = 25303127 },
+]
+
+[[package]]
+name = "referencing"
+version = "0.36.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "attrs" },
+    { name = "rpds-py" },
+    { name = "typing-extensions", marker = "python_full_version < '3.13'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/27/32/fd98246df7a0f309b58cae68b10b6b219ef2eb66747f00dfb34422687087/referencing-0.36.1.tar.gz", hash = "sha256:ca2e6492769e3602957e9b831b94211599d2aade9477f5d44110d2530cf9aade", size = 74661 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cc/fa/9f193ef0c9074b659009f06d7cbacc6f25b072044815bcf799b76533dbb8/referencing-0.36.1-py3-none-any.whl", hash = "sha256:363d9c65f080d0d70bc41c721dce3c7f3e77fc09f269cd5c8813da18069a6794", size = 26777 },
+]
+
+[[package]]
+name = "regex"
+version = "2024.11.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8e/5f/bd69653fbfb76cf8604468d3b4ec4c403197144c7bfe0e6a5fc9e02a07cb/regex-2024.11.6.tar.gz", hash = "sha256:7ab159b063c52a0333c884e4679f8d7a85112ee3078fe3d9004b2dd875585519", size = 399494 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/95/3c/4651f6b130c6842a8f3df82461a8950f923925db8b6961063e82744bddcc/regex-2024.11.6-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:ff590880083d60acc0433f9c3f713c51f7ac6ebb9adf889c79a261ecf541aa91", size = 482674 },
+    { url = "https://files.pythonhosted.org/packages/15/51/9f35d12da8434b489c7b7bffc205c474a0a9432a889457026e9bc06a297a/regex-2024.11.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:658f90550f38270639e83ce492f27d2c8d2cd63805c65a13a14d36ca126753f0", size = 287684 },
+    { url = "https://files.pythonhosted.org/packages/bd/18/b731f5510d1b8fb63c6b6d3484bfa9a59b84cc578ac8b5172970e05ae07c/regex-2024.11.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:164d8b7b3b4bcb2068b97428060b2a53be050085ef94eca7f240e7947f1b080e", size = 284589 },
+    { url = "https://files.pythonhosted.org/packages/78/a2/6dd36e16341ab95e4c6073426561b9bfdeb1a9c9b63ab1b579c2e96cb105/regex-2024.11.6-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d3660c82f209655a06b587d55e723f0b813d3a7db2e32e5e7dc64ac2a9e86fde", size = 782511 },
+    { url = "https://files.pythonhosted.org/packages/1b/2b/323e72d5d2fd8de0d9baa443e1ed70363ed7e7b2fb526f5950c5cb99c364/regex-2024.11.6-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d22326fcdef5e08c154280b71163ced384b428343ae16a5ab2b3354aed12436e", size = 821149 },
+    { url = "https://files.pythonhosted.org/packages/90/30/63373b9ea468fbef8a907fd273e5c329b8c9535fee36fc8dba5fecac475d/regex-2024.11.6-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f1ac758ef6aebfc8943560194e9fd0fa18bcb34d89fd8bd2af18183afd8da3a2", size = 809707 },
+    { url = "https://files.pythonhosted.org/packages/f2/98/26d3830875b53071f1f0ae6d547f1d98e964dd29ad35cbf94439120bb67a/regex-2024.11.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:997d6a487ff00807ba810e0f8332c18b4eb8d29463cfb7c820dc4b6e7562d0cf", size = 781702 },
+    { url = "https://files.pythonhosted.org/packages/87/55/eb2a068334274db86208ab9d5599ffa63631b9f0f67ed70ea7c82a69bbc8/regex-2024.11.6-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:02a02d2bb04fec86ad61f3ea7f49c015a0681bf76abb9857f945d26159d2968c", size = 771976 },
+    { url = "https://files.pythonhosted.org/packages/74/c0/be707bcfe98254d8f9d2cff55d216e946f4ea48ad2fd8cf1428f8c5332ba/regex-2024.11.6-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f02f93b92358ee3f78660e43b4b0091229260c5d5c408d17d60bf26b6c900e86", size = 697397 },
+    { url = "https://files.pythonhosted.org/packages/49/dc/bb45572ceb49e0f6509f7596e4ba7031f6819ecb26bc7610979af5a77f45/regex-2024.11.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:06eb1be98df10e81ebaded73fcd51989dcf534e3c753466e4b60c4697a003b67", size = 768726 },
+    { url = "https://files.pythonhosted.org/packages/5a/db/f43fd75dc4c0c2d96d0881967897926942e935d700863666f3c844a72ce6/regex-2024.11.6-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:040df6fe1a5504eb0f04f048e6d09cd7c7110fef851d7c567a6b6e09942feb7d", size = 775098 },
+    { url = "https://files.pythonhosted.org/packages/99/d7/f94154db29ab5a89d69ff893159b19ada89e76b915c1293e98603d39838c/regex-2024.11.6-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:fdabbfc59f2c6edba2a6622c647b716e34e8e3867e0ab975412c5c2f79b82da2", size = 839325 },
+    { url = "https://files.pythonhosted.org/packages/f7/17/3cbfab1f23356fbbf07708220ab438a7efa1e0f34195bf857433f79f1788/regex-2024.11.6-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:8447d2d39b5abe381419319f942de20b7ecd60ce86f16a23b0698f22e1b70008", size = 843277 },
+    { url = "https://files.pythonhosted.org/packages/7e/f2/48b393b51900456155de3ad001900f94298965e1cad1c772b87f9cfea011/regex-2024.11.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:da8f5fc57d1933de22a9e23eec290a0d8a5927a5370d24bda9a6abe50683fe62", size = 773197 },
+    { url = "https://files.pythonhosted.org/packages/45/3f/ef9589aba93e084cd3f8471fded352826dcae8489b650d0b9b27bc5bba8a/regex-2024.11.6-cp310-cp310-win32.whl", hash = "sha256:b489578720afb782f6ccf2840920f3a32e31ba28a4b162e13900c3e6bd3f930e", size = 261714 },
+    { url = "https://files.pythonhosted.org/packages/42/7e/5f1b92c8468290c465fd50c5318da64319133231415a8aa6ea5ab995a815/regex-2024.11.6-cp310-cp310-win_amd64.whl", hash = "sha256:5071b2093e793357c9d8b2929dfc13ac5f0a6c650559503bb81189d0a3814519", size = 274042 },
+    { url = "https://files.pythonhosted.org/packages/58/58/7e4d9493a66c88a7da6d205768119f51af0f684fe7be7bac8328e217a52c/regex-2024.11.6-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:5478c6962ad548b54a591778e93cd7c456a7a29f8eca9c49e4f9a806dcc5d638", size = 482669 },
+    { url = "https://files.pythonhosted.org/packages/34/4c/8f8e631fcdc2ff978609eaeef1d6994bf2f028b59d9ac67640ed051f1218/regex-2024.11.6-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:2c89a8cc122b25ce6945f0423dc1352cb9593c68abd19223eebbd4e56612c5b7", size = 287684 },
+    { url = "https://files.pythonhosted.org/packages/c5/1b/f0e4d13e6adf866ce9b069e191f303a30ab1277e037037a365c3aad5cc9c/regex-2024.11.6-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:94d87b689cdd831934fa3ce16cc15cd65748e6d689f5d2b8f4f4df2065c9fa20", size = 284589 },
+    { url = "https://files.pythonhosted.org/packages/25/4d/ab21047f446693887f25510887e6820b93f791992994f6498b0318904d4a/regex-2024.11.6-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1062b39a0a2b75a9c694f7a08e7183a80c63c0d62b301418ffd9c35f55aaa114", size = 792121 },
+    { url = "https://files.pythonhosted.org/packages/45/ee/c867e15cd894985cb32b731d89576c41a4642a57850c162490ea34b78c3b/regex-2024.11.6-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:167ed4852351d8a750da48712c3930b031f6efdaa0f22fa1933716bfcd6bf4a3", size = 831275 },
+    { url = "https://files.pythonhosted.org/packages/b3/12/b0f480726cf1c60f6536fa5e1c95275a77624f3ac8fdccf79e6727499e28/regex-2024.11.6-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2d548dafee61f06ebdb584080621f3e0c23fff312f0de1afc776e2a2ba99a74f", size = 818257 },
+    { url = "https://files.pythonhosted.org/packages/bf/ce/0d0e61429f603bac433910d99ef1a02ce45a8967ffbe3cbee48599e62d88/regex-2024.11.6-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f2a19f302cd1ce5dd01a9099aaa19cae6173306d1302a43b627f62e21cf18ac0", size = 792727 },
+    { url = "https://files.pythonhosted.org/packages/e4/c1/243c83c53d4a419c1556f43777ccb552bccdf79d08fda3980e4e77dd9137/regex-2024.11.6-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bec9931dfb61ddd8ef2ebc05646293812cb6b16b60cf7c9511a832b6f1854b55", size = 780667 },
+    { url = "https://files.pythonhosted.org/packages/c5/f4/75eb0dd4ce4b37f04928987f1d22547ddaf6c4bae697623c1b05da67a8aa/regex-2024.11.6-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:9714398225f299aa85267fd222f7142fcb5c769e73d7733344efc46f2ef5cf89", size = 776963 },
+    { url = "https://files.pythonhosted.org/packages/16/5d/95c568574e630e141a69ff8a254c2f188b4398e813c40d49228c9bbd9875/regex-2024.11.6-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:202eb32e89f60fc147a41e55cb086db2a3f8cb82f9a9a88440dcfc5d37faae8d", size = 784700 },
+    { url = "https://files.pythonhosted.org/packages/8e/b5/f8495c7917f15cc6fee1e7f395e324ec3e00ab3c665a7dc9d27562fd5290/regex-2024.11.6-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:4181b814e56078e9b00427ca358ec44333765f5ca1b45597ec7446d3a1ef6e34", size = 848592 },
+    { url = "https://files.pythonhosted.org/packages/1c/80/6dd7118e8cb212c3c60b191b932dc57db93fb2e36fb9e0e92f72a5909af9/regex-2024.11.6-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:068376da5a7e4da51968ce4c122a7cd31afaaec4fccc7856c92f63876e57b51d", size = 852929 },
+    { url = "https://files.pythonhosted.org/packages/11/9b/5a05d2040297d2d254baf95eeeb6df83554e5e1df03bc1a6687fc4ba1f66/regex-2024.11.6-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ac10f2c4184420d881a3475fb2c6f4d95d53a8d50209a2500723d831036f7c45", size = 781213 },
+    { url = "https://files.pythonhosted.org/packages/26/b7/b14e2440156ab39e0177506c08c18accaf2b8932e39fb092074de733d868/regex-2024.11.6-cp311-cp311-win32.whl", hash = "sha256:c36f9b6f5f8649bb251a5f3f66564438977b7ef8386a52460ae77e6070d309d9", size = 261734 },
+    { url = "https://files.pythonhosted.org/packages/80/32/763a6cc01d21fb3819227a1cc3f60fd251c13c37c27a73b8ff4315433a8e/regex-2024.11.6-cp311-cp311-win_amd64.whl", hash = "sha256:02e28184be537f0e75c1f9b2f8847dc51e08e6e171c6bde130b2687e0c33cf60", size = 274052 },
+    { url = "https://files.pythonhosted.org/packages/ba/30/9a87ce8336b172cc232a0db89a3af97929d06c11ceaa19d97d84fa90a8f8/regex-2024.11.6-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:52fb28f528778f184f870b7cf8f225f5eef0a8f6e3778529bdd40c7b3920796a", size = 483781 },
+    { url = "https://files.pythonhosted.org/packages/01/e8/00008ad4ff4be8b1844786ba6636035f7ef926db5686e4c0f98093612add/regex-2024.11.6-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:fdd6028445d2460f33136c55eeb1f601ab06d74cb3347132e1c24250187500d9", size = 288455 },
+    { url = "https://files.pythonhosted.org/packages/60/85/cebcc0aff603ea0a201667b203f13ba75d9fc8668fab917ac5b2de3967bc/regex-2024.11.6-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:805e6b60c54bf766b251e94526ebad60b7de0c70f70a4e6210ee2891acb70bf2", size = 284759 },
+    { url = "https://files.pythonhosted.org/packages/94/2b/701a4b0585cb05472a4da28ee28fdfe155f3638f5e1ec92306d924e5faf0/regex-2024.11.6-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b85c2530be953a890eaffde05485238f07029600e8f098cdf1848d414a8b45e4", size = 794976 },
+    { url = "https://files.pythonhosted.org/packages/4b/bf/fa87e563bf5fee75db8915f7352e1887b1249126a1be4813837f5dbec965/regex-2024.11.6-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bb26437975da7dc36b7efad18aa9dd4ea569d2357ae6b783bf1118dabd9ea577", size = 833077 },
+    { url = "https://files.pythonhosted.org/packages/a1/56/7295e6bad94b047f4d0834e4779491b81216583c00c288252ef625c01d23/regex-2024.11.6-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:abfa5080c374a76a251ba60683242bc17eeb2c9818d0d30117b4486be10c59d3", size = 823160 },
+    { url = "https://files.pythonhosted.org/packages/fb/13/e3b075031a738c9598c51cfbc4c7879e26729c53aa9cca59211c44235314/regex-2024.11.6-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:70b7fa6606c2881c1db9479b0eaa11ed5dfa11c8d60a474ff0e095099f39d98e", size = 796896 },
+    { url = "https://files.pythonhosted.org/packages/24/56/0b3f1b66d592be6efec23a795b37732682520b47c53da5a32c33ed7d84e3/regex-2024.11.6-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0c32f75920cf99fe6b6c539c399a4a128452eaf1af27f39bce8909c9a3fd8cbe", size = 783997 },
+    { url = "https://files.pythonhosted.org/packages/f9/a1/eb378dada8b91c0e4c5f08ffb56f25fcae47bf52ad18f9b2f33b83e6d498/regex-2024.11.6-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:982e6d21414e78e1f51cf595d7f321dcd14de1f2881c5dc6a6e23bbbbd68435e", size = 781725 },
+    { url = "https://files.pythonhosted.org/packages/83/f2/033e7dec0cfd6dda93390089864732a3409246ffe8b042e9554afa9bff4e/regex-2024.11.6-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:a7c2155f790e2fb448faed6dd241386719802296ec588a8b9051c1f5c481bc29", size = 789481 },
+    { url = "https://files.pythonhosted.org/packages/83/23/15d4552ea28990a74e7696780c438aadd73a20318c47e527b47a4a5a596d/regex-2024.11.6-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:149f5008d286636e48cd0b1dd65018548944e495b0265b45e1bffecce1ef7f39", size = 852896 },
+    { url = "https://files.pythonhosted.org/packages/e3/39/ed4416bc90deedbfdada2568b2cb0bc1fdb98efe11f5378d9892b2a88f8f/regex-2024.11.6-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:e5364a4502efca094731680e80009632ad6624084aff9a23ce8c8c6820de3e51", size = 860138 },
+    { url = "https://files.pythonhosted.org/packages/93/2d/dd56bb76bd8e95bbce684326302f287455b56242a4f9c61f1bc76e28360e/regex-2024.11.6-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:0a86e7eeca091c09e021db8eb72d54751e527fa47b8d5787caf96d9831bd02ad", size = 787692 },
+    { url = "https://files.pythonhosted.org/packages/0b/55/31877a249ab7a5156758246b9c59539abbeba22461b7d8adc9e8475ff73e/regex-2024.11.6-cp312-cp312-win32.whl", hash = "sha256:32f9a4c643baad4efa81d549c2aadefaeba12249b2adc5af541759237eee1c54", size = 262135 },
+    { url = "https://files.pythonhosted.org/packages/38/ec/ad2d7de49a600cdb8dd78434a1aeffe28b9d6fc42eb36afab4a27ad23384/regex-2024.11.6-cp312-cp312-win_amd64.whl", hash = "sha256:a93c194e2df18f7d264092dc8539b8ffb86b45b899ab976aa15d48214138e81b", size = 273567 },
+    { url = "https://files.pythonhosted.org/packages/90/73/bcb0e36614601016552fa9344544a3a2ae1809dc1401b100eab02e772e1f/regex-2024.11.6-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a6ba92c0bcdf96cbf43a12c717eae4bc98325ca3730f6b130ffa2e3c3c723d84", size = 483525 },
+    { url = "https://files.pythonhosted.org/packages/0f/3f/f1a082a46b31e25291d830b369b6b0c5576a6f7fb89d3053a354c24b8a83/regex-2024.11.6-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:525eab0b789891ac3be914d36893bdf972d483fe66551f79d3e27146191a37d4", size = 288324 },
+    { url = "https://files.pythonhosted.org/packages/09/c9/4e68181a4a652fb3ef5099e077faf4fd2a694ea6e0f806a7737aff9e758a/regex-2024.11.6-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:086a27a0b4ca227941700e0b31425e7a28ef1ae8e5e05a33826e17e47fbfdba0", size = 284617 },
+    { url = "https://files.pythonhosted.org/packages/fc/fd/37868b75eaf63843165f1d2122ca6cb94bfc0271e4428cf58c0616786dce/regex-2024.11.6-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bde01f35767c4a7899b7eb6e823b125a64de314a8ee9791367c9a34d56af18d0", size = 795023 },
+    { url = "https://files.pythonhosted.org/packages/c4/7c/d4cd9c528502a3dedb5c13c146e7a7a539a3853dc20209c8e75d9ba9d1b2/regex-2024.11.6-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b583904576650166b3d920d2bcce13971f6f9e9a396c673187f49811b2769dc7", size = 833072 },
+    { url = "https://files.pythonhosted.org/packages/4f/db/46f563a08f969159c5a0f0e722260568425363bea43bb7ae370becb66a67/regex-2024.11.6-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1c4de13f06a0d54fa0d5ab1b7138bfa0d883220965a29616e3ea61b35d5f5fc7", size = 823130 },
+    { url = "https://files.pythonhosted.org/packages/db/60/1eeca2074f5b87df394fccaa432ae3fc06c9c9bfa97c5051aed70e6e00c2/regex-2024.11.6-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3cde6e9f2580eb1665965ce9bf17ff4952f34f5b126beb509fee8f4e994f143c", size = 796857 },
+    { url = "https://files.pythonhosted.org/packages/10/db/ac718a08fcee981554d2f7bb8402f1faa7e868c1345c16ab1ebec54b0d7b/regex-2024.11.6-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0d7f453dca13f40a02b79636a339c5b62b670141e63efd511d3f8f73fba162b3", size = 784006 },
+    { url = "https://files.pythonhosted.org/packages/c2/41/7da3fe70216cea93144bf12da2b87367590bcf07db97604edeea55dac9ad/regex-2024.11.6-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:59dfe1ed21aea057a65c6b586afd2a945de04fc7db3de0a6e3ed5397ad491b07", size = 781650 },
+    { url = "https://files.pythonhosted.org/packages/a7/d5/880921ee4eec393a4752e6ab9f0fe28009435417c3102fc413f3fe81c4e5/regex-2024.11.6-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b97c1e0bd37c5cd7902e65f410779d39eeda155800b65fc4d04cc432efa9bc6e", size = 789545 },
+    { url = "https://files.pythonhosted.org/packages/dc/96/53770115e507081122beca8899ab7f5ae28ae790bfcc82b5e38976df6a77/regex-2024.11.6-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:f9d1e379028e0fc2ae3654bac3cbbef81bf3fd571272a42d56c24007979bafb6", size = 853045 },
+    { url = "https://files.pythonhosted.org/packages/31/d3/1372add5251cc2d44b451bd94f43b2ec78e15a6e82bff6a290ef9fd8f00a/regex-2024.11.6-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:13291b39131e2d002a7940fb176e120bec5145f3aeb7621be6534e46251912c4", size = 860182 },
+    { url = "https://files.pythonhosted.org/packages/ed/e3/c446a64984ea9f69982ba1a69d4658d5014bc7a0ea468a07e1a1265db6e2/regex-2024.11.6-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:4f51f88c126370dcec4908576c5a627220da6c09d0bff31cfa89f2523843316d", size = 787733 },
+    { url = "https://files.pythonhosted.org/packages/2b/f1/e40c8373e3480e4f29f2692bd21b3e05f296d3afebc7e5dcf21b9756ca1c/regex-2024.11.6-cp313-cp313-win32.whl", hash = "sha256:63b13cfd72e9601125027202cad74995ab26921d8cd935c25f09c630436348ff", size = 262122 },
+    { url = "https://files.pythonhosted.org/packages/45/94/bc295babb3062a731f52621cdc992d123111282e291abaf23faa413443ea/regex-2024.11.6-cp313-cp313-win_amd64.whl", hash = "sha256:2b3361af3198667e99927da8b84c1b010752fa4b1115ee30beaa332cabc3ef1a", size = 273545 },
+    { url = "https://files.pythonhosted.org/packages/89/23/c4a86df398e57e26f93b13ae63acce58771e04bdde86092502496fa57f9c/regex-2024.11.6-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:5704e174f8ccab2026bd2f1ab6c510345ae8eac818b613d7d73e785f1310f839", size = 482682 },
+    { url = "https://files.pythonhosted.org/packages/3c/8b/45c24ab7a51a1658441b961b86209c43e6bb9d39caf1e63f46ce6ea03bc7/regex-2024.11.6-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:220902c3c5cc6af55d4fe19ead504de80eb91f786dc102fbd74894b1551f095e", size = 287679 },
+    { url = "https://files.pythonhosted.org/packages/7a/d1/598de10b17fdafc452d11f7dada11c3be4e379a8671393e4e3da3c4070df/regex-2024.11.6-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:5e7e351589da0850c125f1600a4c4ba3c722efefe16b297de54300f08d734fbf", size = 284578 },
+    { url = "https://files.pythonhosted.org/packages/49/70/c7eaa219efa67a215846766fde18d92d54cb590b6a04ffe43cef30057622/regex-2024.11.6-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5056b185ca113c88e18223183aa1a50e66507769c9640a6ff75859619d73957b", size = 782012 },
+    { url = "https://files.pythonhosted.org/packages/89/e5/ef52c7eb117dd20ff1697968219971d052138965a4d3d9b95e92e549f505/regex-2024.11.6-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:2e34b51b650b23ed3354b5a07aab37034d9f923db2a40519139af34f485f77d0", size = 820580 },
+    { url = "https://files.pythonhosted.org/packages/5f/3f/9f5da81aff1d4167ac52711acf789df13e789fe6ac9545552e49138e3282/regex-2024.11.6-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5670bce7b200273eee1840ef307bfa07cda90b38ae56e9a6ebcc9f50da9c469b", size = 809110 },
+    { url = "https://files.pythonhosted.org/packages/86/44/2101cc0890c3621b90365c9ee8d7291a597c0722ad66eccd6ffa7f1bcc09/regex-2024.11.6-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:08986dce1339bc932923e7d1232ce9881499a0e02925f7402fb7c982515419ef", size = 780919 },
+    { url = "https://files.pythonhosted.org/packages/ce/2e/3e0668d8d1c7c3c0d397bf54d92fc182575b3a26939aed5000d3cc78760f/regex-2024.11.6-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:93c0b12d3d3bc25af4ebbf38f9ee780a487e8bf6954c115b9f015822d3bb8e48", size = 771515 },
+    { url = "https://files.pythonhosted.org/packages/a6/49/1bc4584254355e3dba930a3a2fd7ad26ccba3ebbab7d9100db0aff2eedb0/regex-2024.11.6-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:764e71f22ab3b305e7f4c21f1a97e1526a25ebdd22513e251cf376760213da13", size = 696957 },
+    { url = "https://files.pythonhosted.org/packages/c8/dd/42879c1fc8a37a887cd08e358af3d3ba9e23038cd77c7fe044a86d9450ba/regex-2024.11.6-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:f056bf21105c2515c32372bbc057f43eb02aae2fda61052e2f7622c801f0b4e2", size = 768088 },
+    { url = "https://files.pythonhosted.org/packages/89/96/c05a0fe173cd2acd29d5e13c1adad8b706bcaa71b169e1ee57dcf2e74584/regex-2024.11.6-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:69ab78f848845569401469da20df3e081e6b5a11cb086de3eed1d48f5ed57c95", size = 774752 },
+    { url = "https://files.pythonhosted.org/packages/b5/f3/a757748066255f97f14506483436c5f6aded7af9e37bca04ec30c90ca683/regex-2024.11.6-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:86fddba590aad9208e2fa8b43b4c098bb0ec74f15718bb6a704e3c63e2cef3e9", size = 838862 },
+    { url = "https://files.pythonhosted.org/packages/5c/93/c6d2092fd479dcaeea40fc8fa673822829181ded77d294a7f950f1dda6e2/regex-2024.11.6-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:684d7a212682996d21ca12ef3c17353c021fe9de6049e19ac8481ec35574a70f", size = 842622 },
+    { url = "https://files.pythonhosted.org/packages/ff/9c/daa99532c72f25051a90ef90e1413a8d54413a9e64614d9095b0c1c154d0/regex-2024.11.6-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:a03e02f48cd1abbd9f3b7e3586d97c8f7a9721c436f51a5245b3b9483044480b", size = 772713 },
+    { url = "https://files.pythonhosted.org/packages/13/5d/61a533ccb8c231b474ac8e3a7d70155b00dfc61af6cafdccd1947df6d735/regex-2024.11.6-cp39-cp39-win32.whl", hash = "sha256:41758407fc32d5c3c5de163888068cfee69cb4c2be844e7ac517a52770f9af57", size = 261756 },
+    { url = "https://files.pythonhosted.org/packages/dc/7b/e59b7f7c91ae110d154370c24133f947262525b5d6406df65f23422acc17/regex-2024.11.6-cp39-cp39-win_amd64.whl", hash = "sha256:b2837718570f95dd41675328e111345f9b7095d821bac435aac173ac80b19983", size = 274110 },
+]
+
+[[package]]
+name = "requests"
+version = "2.32.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "certifi" },
+    { name = "charset-normalizer" },
+    { name = "idna" },
+    { name = "urllib3" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/63/70/2bf7780ad2d390a8d301ad0b550f1581eadbd9a20f896afe06353c2a2913/requests-2.32.3.tar.gz", hash = "sha256:55365417734eb18255590a9ff9eb97e9e1da868d4ccd6402399eaf68af20a760", size = 131218 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f9/9b/335f9764261e915ed497fcdeb11df5dfd6f7bf257d4a6a2a686d80da4d54/requests-2.32.3-py3-none-any.whl", hash = "sha256:70761cfe03c773ceb22aa2f671b4757976145175cdfca038c02654d061d6dcc6", size = 64928 },
+]
+
+[[package]]
+name = "responses"
+version = "0.25.6"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pyyaml" },
+    { name = "requests" },
+    { name = "urllib3" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/97/63/759996eea0f17e8dc4c9ea9c60765292d28a7750bdbee073ad55d83caa57/responses-0.25.6.tar.gz", hash = "sha256:eae7ce61a9603004e76c05691e7c389e59652d91e94b419623c12bbfb8e331d8", size = 79145 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/92/c4/8d23584b3a3471ea6f5a18cfb035e11eeb9fa9b3112d901477c6ad10cc4e/responses-0.25.6-py3-none-any.whl", hash = "sha256:9cac8f21e1193bb150ec557875377e41ed56248aed94e4567ed644db564bacf1", size = 34730 },
+]
+
+[[package]]
+name = "rich"
+version = "13.9.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "markdown-it-py" },
+    { name = "pygments" },
+    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ab/3a/0316b28d0761c6734d6bc14e770d85506c986c85ffb239e688eeaab2c2bc/rich-13.9.4.tar.gz", hash = "sha256:439594978a49a09530cff7ebc4b5c7103ef57baf48d5ea3184f21d9a2befa098", size = 223149 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/19/71/39c7c0d87f8d4e6c020a393182060eaefeeae6c01dab6a84ec346f2567df/rich-13.9.4-py3-none-any.whl", hash = "sha256:6049d5e6ec054bf2779ab3358186963bac2ea89175919d699e378b99738c2a90", size = 242424 },
+]
+
+[[package]]
+name = "rpds-py"
+version = "0.22.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/01/80/cce854d0921ff2f0a9fa831ba3ad3c65cee3a46711addf39a2af52df2cfd/rpds_py-0.22.3.tar.gz", hash = "sha256:e32fee8ab45d3c2db6da19a5323bc3362237c8b653c70194414b892fd06a080d", size = 26771 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/42/2a/ead1d09e57449b99dcc190d8d2323e3a167421d8f8fdf0f217c6f6befe47/rpds_py-0.22.3-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:6c7b99ca52c2c1752b544e310101b98a659b720b21db00e65edca34483259967", size = 359514 },
+    { url = "https://files.pythonhosted.org/packages/8f/7e/1254f406b7793b586c68e217a6a24ec79040f85e030fff7e9049069284f4/rpds_py-0.22.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:be2eb3f2495ba669d2a985f9b426c1797b7d48d6963899276d22f23e33d47e37", size = 349031 },
+    { url = "https://files.pythonhosted.org/packages/aa/da/17c6a2c73730d426df53675ff9cc6653ac7a60b6438d03c18e1c822a576a/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:70eb60b3ae9245ddea20f8a4190bd79c705a22f8028aaf8bbdebe4716c3fab24", size = 381485 },
+    { url = "https://files.pythonhosted.org/packages/aa/13/2dbacd820466aa2a3c4b747afb18d71209523d353cf865bf8f4796c969ea/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:4041711832360a9b75cfb11b25a6a97c8fb49c07b8bd43d0d02b45d0b499a4ff", size = 386794 },
+    { url = "https://files.pythonhosted.org/packages/6d/62/96905d0a35ad4e4bc3c098b2f34b2e7266e211d08635baa690643d2227be/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:64607d4cbf1b7e3c3c8a14948b99345eda0e161b852e122c6bb71aab6d1d798c", size = 423523 },
+    { url = "https://files.pythonhosted.org/packages/eb/1b/d12770f2b6a9fc2c3ec0d810d7d440f6d465ccd8b7f16ae5385952c28b89/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:81e69b0a0e2537f26d73b4e43ad7bc8c8efb39621639b4434b76a3de50c6966e", size = 446695 },
+    { url = "https://files.pythonhosted.org/packages/4d/cf/96f1fd75512a017f8e07408b6d5dbeb492d9ed46bfe0555544294f3681b3/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc27863442d388870c1809a87507727b799c8460573cfbb6dc0eeaef5a11b5ec", size = 381959 },
+    { url = "https://files.pythonhosted.org/packages/ab/f0/d1c5b501c8aea85aeb938b555bfdf7612110a2f8cdc21ae0482c93dd0c24/rpds_py-0.22.3-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:e79dd39f1e8c3504be0607e5fc6e86bb60fe3584bec8b782578c3b0fde8d932c", size = 410420 },
+    { url = "https://files.pythonhosted.org/packages/33/3b/45b6c58fb6aad5a569ae40fb890fc494c6b02203505a5008ee6dc68e65f7/rpds_py-0.22.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:e0fa2d4ec53dc51cf7d3bb22e0aa0143966119f42a0c3e4998293a3dd2856b09", size = 557620 },
+    { url = "https://files.pythonhosted.org/packages/83/62/3fdd2d3d47bf0bb9b931c4c73036b4ab3ec77b25e016ae26fab0f02be2af/rpds_py-0.22.3-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fda7cb070f442bf80b642cd56483b5548e43d366fe3f39b98e67cce780cded00", size = 584202 },
+    { url = "https://files.pythonhosted.org/packages/04/f2/5dced98b64874b84ca824292f9cee2e3f30f3bcf231d15a903126684f74d/rpds_py-0.22.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:cff63a0272fcd259dcc3be1657b07c929c466b067ceb1c20060e8d10af56f5bf", size = 552787 },
+    { url = "https://files.pythonhosted.org/packages/67/13/2273dea1204eda0aea0ef55145da96a9aa28b3f88bb5c70e994f69eda7c3/rpds_py-0.22.3-cp310-cp310-win32.whl", hash = "sha256:9bd7228827ec7bb817089e2eb301d907c0d9827a9e558f22f762bb690b131652", size = 220088 },
+    { url = "https://files.pythonhosted.org/packages/4e/80/8c8176b67ad7f4a894967a7a4014ba039626d96f1d4874d53e409b58d69f/rpds_py-0.22.3-cp310-cp310-win_amd64.whl", hash = "sha256:9beeb01d8c190d7581a4d59522cd3d4b6887040dcfc744af99aa59fef3e041a8", size = 231737 },
+    { url = "https://files.pythonhosted.org/packages/15/ad/8d1ddf78f2805a71253fcd388017e7b4a0615c22c762b6d35301fef20106/rpds_py-0.22.3-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:d20cfb4e099748ea39e6f7b16c91ab057989712d31761d3300d43134e26e165f", size = 359773 },
+    { url = "https://files.pythonhosted.org/packages/c8/75/68c15732293a8485d79fe4ebe9045525502a067865fa4278f178851b2d87/rpds_py-0.22.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:68049202f67380ff9aa52f12e92b1c30115f32e6895cd7198fa2a7961621fc5a", size = 349214 },
+    { url = "https://files.pythonhosted.org/packages/3c/4c/7ce50f3070083c2e1b2bbd0fb7046f3da55f510d19e283222f8f33d7d5f4/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fb4f868f712b2dd4bcc538b0a0c1f63a2b1d584c925e69a224d759e7070a12d5", size = 380477 },
+    { url = "https://files.pythonhosted.org/packages/9a/e9/835196a69cb229d5c31c13b8ae603bd2da9a6695f35fe4270d398e1db44c/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:bc51abd01f08117283c5ebf64844a35144a0843ff7b2983e0648e4d3d9f10dbb", size = 386171 },
+    { url = "https://files.pythonhosted.org/packages/f9/8e/33fc4eba6683db71e91e6d594a2cf3a8fbceb5316629f0477f7ece5e3f75/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0f3cec041684de9a4684b1572fe28c7267410e02450f4561700ca5a3bc6695a2", size = 422676 },
+    { url = "https://files.pythonhosted.org/packages/37/47/2e82d58f8046a98bb9497a8319604c92b827b94d558df30877c4b3c6ccb3/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7ef9d9da710be50ff6809fed8f1963fecdfecc8b86656cadfca3bc24289414b0", size = 446152 },
+    { url = "https://files.pythonhosted.org/packages/e1/78/79c128c3e71abbc8e9739ac27af11dc0f91840a86fce67ff83c65d1ba195/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:59f4a79c19232a5774aee369a0c296712ad0e77f24e62cad53160312b1c1eaa1", size = 381300 },
+    { url = "https://files.pythonhosted.org/packages/c9/5b/2e193be0e8b228c1207f31fa3ea79de64dadb4f6a4833111af8145a6bc33/rpds_py-0.22.3-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1a60bce91f81ddaac922a40bbb571a12c1070cb20ebd6d49c48e0b101d87300d", size = 409636 },
+    { url = "https://files.pythonhosted.org/packages/c2/3f/687c7100b762d62186a1c1100ffdf99825f6fa5ea94556844bbbd2d0f3a9/rpds_py-0.22.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:e89391e6d60251560f0a8f4bd32137b077a80d9b7dbe6d5cab1cd80d2746f648", size = 556708 },
+    { url = "https://files.pythonhosted.org/packages/8c/a2/c00cbc4b857e8b3d5e7f7fc4c81e23afd8c138b930f4f3ccf9a41a23e9e4/rpds_py-0.22.3-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:e3fb866d9932a3d7d0c82da76d816996d1667c44891bd861a0f97ba27e84fc74", size = 583554 },
+    { url = "https://files.pythonhosted.org/packages/d0/08/696c9872cf56effdad9ed617ac072f6774a898d46b8b8964eab39ec562d2/rpds_py-0.22.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:1352ae4f7c717ae8cba93421a63373e582d19d55d2ee2cbb184344c82d2ae55a", size = 552105 },
+    { url = "https://files.pythonhosted.org/packages/18/1f/4df560be1e994f5adf56cabd6c117e02de7c88ee238bb4ce03ed50da9d56/rpds_py-0.22.3-cp311-cp311-win32.whl", hash = "sha256:b0b4136a252cadfa1adb705bb81524eee47d9f6aab4f2ee4fa1e9d3cd4581f64", size = 220199 },
+    { url = "https://files.pythonhosted.org/packages/b8/1b/c29b570bc5db8237553002788dc734d6bd71443a2ceac2a58202ec06ef12/rpds_py-0.22.3-cp311-cp311-win_amd64.whl", hash = "sha256:8bd7c8cfc0b8247c8799080fbff54e0b9619e17cdfeb0478ba7295d43f635d7c", size = 231775 },
+    { url = "https://files.pythonhosted.org/packages/75/47/3383ee3bd787a2a5e65a9b9edc37ccf8505c0a00170e3a5e6ea5fbcd97f7/rpds_py-0.22.3-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:27e98004595899949bd7a7b34e91fa7c44d7a97c40fcaf1d874168bb652ec67e", size = 352334 },
+    { url = "https://files.pythonhosted.org/packages/40/14/aa6400fa8158b90a5a250a77f2077c0d0cd8a76fce31d9f2b289f04c6dec/rpds_py-0.22.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1978d0021e943aae58b9b0b196fb4895a25cc53d3956b8e35e0b7682eefb6d56", size = 342111 },
+    { url = "https://files.pythonhosted.org/packages/7d/06/395a13bfaa8a28b302fb433fb285a67ce0ea2004959a027aea8f9c52bad4/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:655ca44a831ecb238d124e0402d98f6212ac527a0ba6c55ca26f616604e60a45", size = 384286 },
+    { url = "https://files.pythonhosted.org/packages/43/52/d8eeaffab047e6b7b7ef7f00d5ead074a07973968ffa2d5820fa131d7852/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:feea821ee2a9273771bae61194004ee2fc33f8ec7db08117ef9147d4bbcbca8e", size = 391739 },
+    { url = "https://files.pythonhosted.org/packages/83/31/52dc4bde85c60b63719610ed6f6d61877effdb5113a72007679b786377b8/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:22bebe05a9ffc70ebfa127efbc429bc26ec9e9b4ee4d15a740033efda515cf3d", size = 427306 },
+    { url = "https://files.pythonhosted.org/packages/70/d5/1bab8e389c2261dba1764e9e793ed6830a63f830fdbec581a242c7c46bda/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3af6e48651c4e0d2d166dc1b033b7042ea3f871504b6805ba5f4fe31581d8d38", size = 442717 },
+    { url = "https://files.pythonhosted.org/packages/82/a1/a45f3e30835b553379b3a56ea6c4eb622cf11e72008229af840e4596a8ea/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e67ba3c290821343c192f7eae1d8fd5999ca2dc99994114643e2f2d3e6138b15", size = 385721 },
+    { url = "https://files.pythonhosted.org/packages/a6/27/780c942de3120bdd4d0e69583f9c96e179dfff082f6ecbb46b8d6488841f/rpds_py-0.22.3-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:02fbb9c288ae08bcb34fb41d516d5eeb0455ac35b5512d03181d755d80810059", size = 415824 },
+    { url = "https://files.pythonhosted.org/packages/94/0b/aa0542ca88ad20ea719b06520f925bae348ea5c1fdf201b7e7202d20871d/rpds_py-0.22.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:f56a6b404f74ab372da986d240e2e002769a7d7102cc73eb238a4f72eec5284e", size = 561227 },
+    { url = "https://files.pythonhosted.org/packages/0d/92/3ed77d215f82c8f844d7f98929d56cc321bb0bcfaf8f166559b8ec56e5f1/rpds_py-0.22.3-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:0a0461200769ab3b9ab7e513f6013b7a97fdeee41c29b9db343f3c5a8e2b9e61", size = 587424 },
+    { url = "https://files.pythonhosted.org/packages/09/42/cacaeb047a22cab6241f107644f230e2935d4efecf6488859a7dd82fc47d/rpds_py-0.22.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:8633e471c6207a039eff6aa116e35f69f3156b3989ea3e2d755f7bc41754a4a7", size = 555953 },
+    { url = "https://files.pythonhosted.org/packages/e6/52/c921dc6d5f5d45b212a456c1f5b17df1a471127e8037eb0972379e39dff4/rpds_py-0.22.3-cp312-cp312-win32.whl", hash = "sha256:593eba61ba0c3baae5bc9be2f5232430453fb4432048de28399ca7376de9c627", size = 221339 },
+    { url = "https://files.pythonhosted.org/packages/f2/c7/f82b5be1e8456600395366f86104d1bd8d0faed3802ad511ef6d60c30d98/rpds_py-0.22.3-cp312-cp312-win_amd64.whl", hash = "sha256:d115bffdd417c6d806ea9069237a4ae02f513b778e3789a359bc5856e0404cc4", size = 235786 },
+    { url = "https://files.pythonhosted.org/packages/d0/bf/36d5cc1f2c609ae6e8bf0fc35949355ca9d8790eceb66e6385680c951e60/rpds_py-0.22.3-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:ea7433ce7e4bfc3a85654aeb6747babe3f66eaf9a1d0c1e7a4435bbdf27fea84", size = 351657 },
+    { url = "https://files.pythonhosted.org/packages/24/2a/f1e0fa124e300c26ea9382e59b2d582cba71cedd340f32d1447f4f29fa4e/rpds_py-0.22.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:6dd9412824c4ce1aca56c47b0991e65bebb7ac3f4edccfd3f156150c96a7bf25", size = 341829 },
+    { url = "https://files.pythonhosted.org/packages/cf/c2/0da1231dd16953845bed60d1a586fcd6b15ceaeb965f4d35cdc71f70f606/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:20070c65396f7373f5df4005862fa162db5d25d56150bddd0b3e8214e8ef45b4", size = 384220 },
+    { url = "https://files.pythonhosted.org/packages/c7/73/a4407f4e3a00a9d4b68c532bf2d873d6b562854a8eaff8faa6133b3588ec/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:0b09865a9abc0ddff4e50b5ef65467cd94176bf1e0004184eb915cbc10fc05c5", size = 391009 },
+    { url = "https://files.pythonhosted.org/packages/a9/c3/04b7353477ab360fe2563f5f0b176d2105982f97cd9ae80a9c5a18f1ae0f/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3453e8d41fe5f17d1f8e9c383a7473cd46a63661628ec58e07777c2fff7196dc", size = 426989 },
+    { url = "https://files.pythonhosted.org/packages/8d/e6/e4b85b722bcf11398e17d59c0f6049d19cd606d35363221951e6d625fcb0/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f5d36399a1b96e1a5fdc91e0522544580dbebeb1f77f27b2b0ab25559e103b8b", size = 441544 },
+    { url = "https://files.pythonhosted.org/packages/27/fc/403e65e56f65fff25f2973216974976d3f0a5c3f30e53758589b6dc9b79b/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:009de23c9c9ee54bf11303a966edf4d9087cd43a6003672e6aa7def643d06518", size = 385179 },
+    { url = "https://files.pythonhosted.org/packages/57/9b/2be9ff9700d664d51fd96b33d6595791c496d2778cb0b2a634f048437a55/rpds_py-0.22.3-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1aef18820ef3e4587ebe8b3bc9ba6e55892a6d7b93bac6d29d9f631a3b4befbd", size = 415103 },
+    { url = "https://files.pythonhosted.org/packages/bb/a5/03c2ad8ca10994fcf22dd2150dd1d653bc974fa82d9a590494c84c10c641/rpds_py-0.22.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f60bd8423be1d9d833f230fdbccf8f57af322d96bcad6599e5a771b151398eb2", size = 560916 },
+    { url = "https://files.pythonhosted.org/packages/ba/2e/be4fdfc8b5b576e588782b56978c5b702c5a2307024120d8aeec1ab818f0/rpds_py-0.22.3-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:62d9cfcf4948683a18a9aff0ab7e1474d407b7bab2ca03116109f8464698ab16", size = 587062 },
+    { url = "https://files.pythonhosted.org/packages/67/e0/2034c221937709bf9c542603d25ad43a68b4b0a9a0c0b06a742f2756eb66/rpds_py-0.22.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:9253fc214112405f0afa7db88739294295f0e08466987f1d70e29930262b4c8f", size = 555734 },
+    { url = "https://files.pythonhosted.org/packages/ea/ce/240bae07b5401a22482b58e18cfbabaa392409b2797da60223cca10d7367/rpds_py-0.22.3-cp313-cp313-win32.whl", hash = "sha256:fb0ba113b4983beac1a2eb16faffd76cb41e176bf58c4afe3e14b9c681f702de", size = 220663 },
+    { url = "https://files.pythonhosted.org/packages/cb/f0/d330d08f51126330467edae2fa4efa5cec8923c87551a79299380fdea30d/rpds_py-0.22.3-cp313-cp313-win_amd64.whl", hash = "sha256:c58e2339def52ef6b71b8f36d13c3688ea23fa093353f3a4fee2556e62086ec9", size = 235503 },
+    { url = "https://files.pythonhosted.org/packages/f7/c4/dbe1cc03df013bf2feb5ad00615038050e7859f381e96fb5b7b4572cd814/rpds_py-0.22.3-cp313-cp313t-macosx_10_12_x86_64.whl", hash = "sha256:f82a116a1d03628a8ace4859556fb39fd1424c933341a08ea3ed6de1edb0283b", size = 347698 },
+    { url = "https://files.pythonhosted.org/packages/a4/3a/684f66dd6b0f37499cad24cd1c0e523541fd768576fa5ce2d0a8799c3cba/rpds_py-0.22.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:3dfcbc95bd7992b16f3f7ba05af8a64ca694331bd24f9157b49dadeeb287493b", size = 337330 },
+    { url = "https://files.pythonhosted.org/packages/82/eb/e022c08c2ce2e8f7683baa313476492c0e2c1ca97227fe8a75d9f0181e95/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:59259dc58e57b10e7e18ce02c311804c10c5a793e6568f8af4dead03264584d1", size = 380022 },
+    { url = "https://files.pythonhosted.org/packages/e4/21/5a80e653e4c86aeb28eb4fea4add1f72e1787a3299687a9187105c3ee966/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5725dd9cc02068996d4438d397e255dcb1df776b7ceea3b9cb972bdb11260a83", size = 390754 },
+    { url = "https://files.pythonhosted.org/packages/37/a4/d320a04ae90f72d080b3d74597074e62be0a8ecad7d7321312dfe2dc5a6a/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:99b37292234e61325e7a5bb9689e55e48c3f5f603af88b1642666277a81f1fbd", size = 423840 },
+    { url = "https://files.pythonhosted.org/packages/87/70/674dc47d93db30a6624279284e5631be4c3a12a0340e8e4f349153546728/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:27b1d3b3915a99208fee9ab092b8184c420f2905b7d7feb4aeb5e4a9c509b8a1", size = 438970 },
+    { url = "https://files.pythonhosted.org/packages/3f/64/9500f4d66601d55cadd21e90784cfd5d5f4560e129d72e4339823129171c/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f612463ac081803f243ff13cccc648578e2279295048f2a8d5eb430af2bae6e3", size = 383146 },
+    { url = "https://files.pythonhosted.org/packages/4d/45/630327addb1d17173adcf4af01336fd0ee030c04798027dfcb50106001e0/rpds_py-0.22.3-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f73d3fef726b3243a811121de45193c0ca75f6407fe66f3f4e183c983573e130", size = 408294 },
+    { url = "https://files.pythonhosted.org/packages/5f/ef/8efb3373cee54ea9d9980b772e5690a0c9e9214045a4e7fa35046e399fee/rpds_py-0.22.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:3f21f0495edea7fdbaaa87e633a8689cd285f8f4af5c869f27bc8074638ad69c", size = 556345 },
+    { url = "https://files.pythonhosted.org/packages/54/01/151d3b9ef4925fc8f15bfb131086c12ec3c3d6dd4a4f7589c335bf8e85ba/rpds_py-0.22.3-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:1e9663daaf7a63ceccbbb8e3808fe90415b0757e2abddbfc2e06c857bf8c5e2b", size = 582292 },
+    { url = "https://files.pythonhosted.org/packages/30/89/35fc7a6cdf3477d441c7aca5e9bbf5a14e0f25152aed7f63f4e0b141045d/rpds_py-0.22.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:a76e42402542b1fae59798fab64432b2d015ab9d0c8c47ba7addddbaf7952333", size = 553855 },
+    { url = "https://files.pythonhosted.org/packages/8f/e0/830c02b2457c4bd20a8c5bb394d31d81f57fbefce2dbdd2e31feff4f7003/rpds_py-0.22.3-cp313-cp313t-win32.whl", hash = "sha256:69803198097467ee7282750acb507fba35ca22cc3b85f16cf45fb01cb9097730", size = 219100 },
+    { url = "https://files.pythonhosted.org/packages/f8/30/7ac943f69855c2db77407ae363484b915d861702dbba1aa82d68d57f42be/rpds_py-0.22.3-cp313-cp313t-win_amd64.whl", hash = "sha256:f5cf2a0c2bdadf3791b5c205d55a37a54025c6e18a71c71f82bb536cf9a454bf", size = 233794 },
+    { url = "https://files.pythonhosted.org/packages/db/0f/a8ad17ddac7c880f48d5da50733dd25bfc35ba2be1bec9f23453e8c7a123/rpds_py-0.22.3-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:378753b4a4de2a7b34063d6f95ae81bfa7b15f2c1a04a9518e8644e81807ebea", size = 359735 },
+    { url = "https://files.pythonhosted.org/packages/0c/41/430903669397ea3ee76865e0b53ea236e8dc0ffbecde47b2c4c783ad6759/rpds_py-0.22.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3445e07bf2e8ecfeef6ef67ac83de670358abf2996916039b16a218e3d95e97e", size = 348724 },
+    { url = "https://files.pythonhosted.org/packages/c9/5c/3496f4f0ee818297544f2d5f641c49dde8ae156392e6834b79c0609ba006/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7b2513ba235829860b13faa931f3b6846548021846ac808455301c23a101689d", size = 381782 },
+    { url = "https://files.pythonhosted.org/packages/b6/dc/db0523ce0cd16ce579185cc9aa9141992de956d0a9c469ecfd1fb5d54ddc/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:eaf16ae9ae519a0e237a0f528fd9f0197b9bb70f40263ee57ae53c2b8d48aeb3", size = 387036 },
+    { url = "https://files.pythonhosted.org/packages/85/2a/9525c2427d2c257f877348918136a5d4e1b945c205a256e53bec61e54551/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:583f6a1993ca3369e0f80ba99d796d8e6b1a3a2a442dd4e1a79e652116413091", size = 424566 },
+    { url = "https://files.pythonhosted.org/packages/b9/1c/f8c012a39794b84069635709f559c0309103d5d74b3f5013916e6ca4f174/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:4617e1915a539a0d9a9567795023de41a87106522ff83fbfaf1f6baf8e85437e", size = 447203 },
+    { url = "https://files.pythonhosted.org/packages/93/f5/c1c772364570d35b98ba64f36ec90c3c6d0b932bc4d8b9b4efef6dc64b07/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0c150c7a61ed4a4f4955a96626574e9baf1adf772c2fb61ef6a5027e52803543", size = 382283 },
+    { url = "https://files.pythonhosted.org/packages/10/06/f94f61313f94fc75c3c3aa74563f80bbd990e5b25a7c1a38cee7d5d0309b/rpds_py-0.22.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2fa4331c200c2521512595253f5bb70858b90f750d39b8cbfd67465f8d1b596d", size = 410022 },
+    { url = "https://files.pythonhosted.org/packages/3f/b0/37ab416a9528419920dfb64886c220f58fcbd66b978e0a91b66e9ee9a993/rpds_py-0.22.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:214b7a953d73b5e87f0ebece4a32a5bd83c60a3ecc9d4ec8f1dca968a2d91e99", size = 557817 },
+    { url = "https://files.pythonhosted.org/packages/2c/5d/9daa18adcd676dd3b2817c8a7cec3f3ebeeb0ce0d05a1b63bf994fc5114f/rpds_py-0.22.3-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:f47ad3d5f3258bd7058d2d506852217865afefe6153a36eb4b6928758041d831", size = 585099 },
+    { url = "https://files.pythonhosted.org/packages/41/3f/ad4e58035d3f848410aa3d59857b5f238bafab81c8b4a844281f80445d62/rpds_py-0.22.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:f276b245347e6e36526cbd4a266a417796fc531ddf391e43574cf6466c492520", size = 552818 },
+    { url = "https://files.pythonhosted.org/packages/b8/19/123acae8f4cab3c9463097c3ced3cc87c46f405056e249c874940e045309/rpds_py-0.22.3-cp39-cp39-win32.whl", hash = "sha256:bbb232860e3d03d544bc03ac57855cd82ddf19c7a07651a7c0fdb95e9efea8b9", size = 220246 },
+    { url = "https://files.pythonhosted.org/packages/8b/8d/9db93e48d96ace1f6713c71ce72e2d94b71d82156c37b6a54e0930486f00/rpds_py-0.22.3-cp39-cp39-win_amd64.whl", hash = "sha256:cfbc454a2880389dbb9b5b398e50d439e2e58669160f27b60e5eca11f68ae17c", size = 231932 },
+    { url = "https://files.pythonhosted.org/packages/8b/63/e29f8ee14fcf383574f73b6bbdcbec0fbc2e5fc36b4de44d1ac389b1de62/rpds_py-0.22.3-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:d48424e39c2611ee1b84ad0f44fb3b2b53d473e65de061e3f460fc0be5f1939d", size = 360786 },
+    { url = "https://files.pythonhosted.org/packages/d3/e0/771ee28b02a24e81c8c0e645796a371350a2bb6672753144f36ae2d2afc9/rpds_py-0.22.3-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:24e8abb5878e250f2eb0d7859a8e561846f98910326d06c0d51381fed59357bd", size = 350589 },
+    { url = "https://files.pythonhosted.org/packages/cf/49/abad4c4a1e6f3adf04785a99c247bfabe55ed868133e2d1881200aa5d381/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4b232061ca880db21fa14defe219840ad9b74b6158adb52ddf0e87bead9e8493", size = 381848 },
+    { url = "https://files.pythonhosted.org/packages/3a/7d/f4bc6d6fbe6af7a0d2b5f2ee77079efef7c8528712745659ec0026888998/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ac0a03221cdb5058ce0167ecc92a8c89e8d0decdc9e99a2ec23380793c4dcb96", size = 387879 },
+    { url = "https://files.pythonhosted.org/packages/13/b0/575c797377fdcd26cedbb00a3324232e4cb2c5d121f6e4b0dbf8468b12ef/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:eb0c341fa71df5a4595f9501df4ac5abfb5a09580081dffbd1ddd4654e6e9123", size = 423916 },
+    { url = "https://files.pythonhosted.org/packages/54/78/87157fa39d58f32a68d3326f8a81ad8fb99f49fe2aa7ad9a1b7d544f9478/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bf9db5488121b596dbfc6718c76092fda77b703c1f7533a226a5a9f65248f8ad", size = 448410 },
+    { url = "https://files.pythonhosted.org/packages/59/69/860f89996065a88be1b6ff2d60e96a02b920a262d8aadab99e7903986597/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0b8db6b5b2d4491ad5b6bdc2bc7c017eec108acbf4e6785f42a9eb0ba234f4c9", size = 382841 },
+    { url = "https://files.pythonhosted.org/packages/bd/d7/bc144e10d27e3cb350f98df2492a319edd3caaf52ddfe1293f37a9afbfd7/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:b3d504047aba448d70cf6fa22e06cb09f7cbd761939fdd47604f5e007675c24e", size = 409662 },
+    { url = "https://files.pythonhosted.org/packages/14/2a/6bed0b05233c291a94c7e89bc76ffa1c619d4e1979fbfe5d96024020c1fb/rpds_py-0.22.3-pp310-pypy310_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:e61b02c3f7a1e0b75e20c3978f7135fd13cb6cf551bf4a6d29b999a88830a338", size = 558221 },
+    { url = "https://files.pythonhosted.org/packages/11/23/cd8f566de444a137bc1ee5795e47069a947e60810ba4152886fe5308e1b7/rpds_py-0.22.3-pp310-pypy310_pp73-musllinux_1_2_i686.whl", hash = "sha256:e35ba67d65d49080e8e5a1dd40101fccdd9798adb9b050ff670b7d74fa41c566", size = 583780 },
+    { url = "https://files.pythonhosted.org/packages/8d/63/79c3602afd14d501f751e615a74a59040328da5ef29ed5754ae80d236b84/rpds_py-0.22.3-pp310-pypy310_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:26fd7cac7dd51011a245f29a2cc6489c4608b5a8ce8d75661bb4a1066c52dfbe", size = 553619 },
+    { url = "https://files.pythonhosted.org/packages/9f/2e/c5c1689e80298d4e94c75b70faada4c25445739d91b94c211244a3ed7ed1/rpds_py-0.22.3-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:177c7c0fce2855833819c98e43c262007f42ce86651ffbb84f37883308cb0e7d", size = 233338 },
+    { url = "https://files.pythonhosted.org/packages/bc/b7/d2c205723e3b4d75b03215694f0297a1b4b395bf834cb5896ad9bbb90f90/rpds_py-0.22.3-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:bb47271f60660803ad11f4c61b42242b8c1312a31c98c578f79ef9387bbde21c", size = 360594 },
+    { url = "https://files.pythonhosted.org/packages/d8/8f/c3515f5234cf6055046d4cfe9c80a3742a20acfa7d0b1b290f0d7f56a8db/rpds_py-0.22.3-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:70fb28128acbfd264eda9bf47015537ba3fe86e40d046eb2963d75024be4d055", size = 349594 },
+    { url = "https://files.pythonhosted.org/packages/6b/98/5b487cb06afc484befe350c87fda37f4ce11333f04f3380aba43dcf5bce2/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:44d61b4b7d0c2c9ac019c314e52d7cbda0ae31078aabd0f22e583af3e0d79723", size = 381138 },
+    { url = "https://files.pythonhosted.org/packages/5e/3a/12308d2c51b3fdfc173619943b7dc5ba41b4850c47112eeda38d9c54ed12/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5f0e260eaf54380380ac3808aa4ebe2d8ca28b9087cf411649f96bad6900c728", size = 387828 },
+    { url = "https://files.pythonhosted.org/packages/17/b2/c242241ab5a2a206e093f24ccbfa519c4bbf10a762ac90bffe1766c225e0/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b25bc607423935079e05619d7de556c91fb6adeae9d5f80868dde3468657994b", size = 424634 },
+    { url = "https://files.pythonhosted.org/packages/d5/c7/52a1b15012139f3ba740f291f1d03c6b632938ba61bc605f24c101952493/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fb6116dfb8d1925cbdb52595560584db42a7f664617a1f7d7f6e32f138cdf37d", size = 447862 },
+    { url = "https://files.pythonhosted.org/packages/55/3e/4d3ed8fd01bad77e8ed101116fe63b03f1011940d9596a8f4d82ac80cacd/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a63cbdd98acef6570c62b92a1e43266f9e8b21e699c363c0fef13bd530799c11", size = 382506 },
+    { url = "https://files.pythonhosted.org/packages/30/78/df59d6f92470a84369a3757abeae1cfd7f7239c8beb6d948949bf78317d2/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2b8f60e1b739a74bab7e01fcbe3dddd4657ec685caa04681df9d562ef15b625f", size = 410534 },
+    { url = "https://files.pythonhosted.org/packages/38/97/ea45d1edd9b753b20084b52dd5db6ee5e1ac3e036a27149972398a413858/rpds_py-0.22.3-pp39-pypy39_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:2e8b55d8517a2fda8d95cb45d62a5a8bbf9dd0ad39c5b25c8833efea07b880ca", size = 557453 },
+    { url = "https://files.pythonhosted.org/packages/08/cd/3a1b35eb9da27ffbb981cfffd32a01c7655c4431ccb278cb3064f8887462/rpds_py-0.22.3-pp39-pypy39_pp73-musllinux_1_2_i686.whl", hash = "sha256:2de29005e11637e7a2361fa151f780ff8eb2543a0da1413bb951e9f14b699ef3", size = 584412 },
+    { url = "https://files.pythonhosted.org/packages/87/91/31d1c5aeb1606f71188259e0ba6ed6f5c21a3c72f58b51db6a8bd0aa2b5d/rpds_py-0.22.3-pp39-pypy39_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:666ecce376999bf619756a24ce15bb14c5bfaf04bf00abc7e663ce17c3f34fe7", size = 553446 },
+    { url = "https://files.pythonhosted.org/packages/e7/ad/03b5ccd1ab492c9dece85b3bf1c96453ab8c47983936fae6880f688f60b3/rpds_py-0.22.3-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:5246b14ca64a8675e0a7161f7af68fe3e910e6b90542b4bfb5439ba752191df6", size = 233013 },
+]
+
+[[package]]
+name = "safetensors"
+version = "0.5.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f4/4f/2ef9ef1766f8c194b01b67a63a444d2e557c8fe1d82faf3ebd85f370a917/safetensors-0.5.2.tar.gz", hash = "sha256:cb4a8d98ba12fa016f4241932b1fc5e702e5143f5374bba0bbcf7ddc1c4cf2b8", size = 66957 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/96/d1/017e31e75e274492a11a456a9e7c171f8f7911fe50735b4ec6ff37221220/safetensors-0.5.2-cp38-abi3-macosx_10_12_x86_64.whl", hash = "sha256:45b6092997ceb8aa3801693781a71a99909ab9cc776fbc3fa9322d29b1d3bef2", size = 427067 },
+    { url = "https://files.pythonhosted.org/packages/24/84/e9d3ff57ae50dd0028f301c9ee064e5087fe8b00e55696677a0413c377a7/safetensors-0.5.2-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:6d0d6a8ee2215a440e1296b843edf44fd377b055ba350eaba74655a2fe2c4bae", size = 408856 },
+    { url = "https://files.pythonhosted.org/packages/f1/1d/fe95f5dd73db16757b11915e8a5106337663182d0381811c81993e0014a9/safetensors-0.5.2-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:86016d40bcaa3bcc9a56cd74d97e654b5f4f4abe42b038c71e4f00a089c4526c", size = 450088 },
+    { url = "https://files.pythonhosted.org/packages/cf/21/e527961b12d5ab528c6e47b92d5f57f33563c28a972750b238b871924e49/safetensors-0.5.2-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:990833f70a5f9c7d3fc82c94507f03179930ff7d00941c287f73b6fcbf67f19e", size = 458966 },
+    { url = "https://files.pythonhosted.org/packages/a5/8b/1a037d7a57f86837c0b41905040369aea7d8ca1ec4b2a77592372b2ec380/safetensors-0.5.2-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3dfa7c2f3fe55db34eba90c29df94bcdac4821043fc391cb5d082d9922013869", size = 509915 },
+    { url = "https://files.pythonhosted.org/packages/61/3d/03dd5cfd33839df0ee3f4581a20bd09c40246d169c0e4518f20b21d5f077/safetensors-0.5.2-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:46ff2116150ae70a4e9c490d2ab6b6e1b1b93f25e520e540abe1b81b48560c3a", size = 527664 },
+    { url = "https://files.pythonhosted.org/packages/c5/dc/8952caafa9a10a3c0f40fa86bacf3190ae7f55fa5eef87415b97b29cb97f/safetensors-0.5.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3ab696dfdc060caffb61dbe4066b86419107a24c804a4e373ba59be699ebd8d5", size = 461978 },
+    { url = "https://files.pythonhosted.org/packages/60/da/82de1fcf1194e3dbefd4faa92dc98b33c06bed5d67890e0962dd98e18287/safetensors-0.5.2-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:03c937100f38c9ff4c1507abea9928a6a9b02c9c1c9c3609ed4fb2bf413d4975", size = 491253 },
+    { url = "https://files.pythonhosted.org/packages/5a/9a/d90e273c25f90c3ba1b0196a972003786f04c39e302fbd6649325b1272bb/safetensors-0.5.2-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:a00e737948791b94dad83cf0eafc09a02c4d8c2171a239e8c8572fe04e25960e", size = 628644 },
+    { url = "https://files.pythonhosted.org/packages/70/3c/acb23e05aa34b4f5edd2e7f393f8e6480fbccd10601ab42cd03a57d4ab5f/safetensors-0.5.2-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:d3a06fae62418ec8e5c635b61a8086032c9e281f16c63c3af46a6efbab33156f", size = 721648 },
+    { url = "https://files.pythonhosted.org/packages/71/45/eaa3dba5253a7c6931230dc961641455710ab231f8a89cb3c4c2af70f8c8/safetensors-0.5.2-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:1506e4c2eda1431099cebe9abf6c76853e95d0b7a95addceaa74c6019c65d8cf", size = 659588 },
+    { url = "https://files.pythonhosted.org/packages/b0/71/2f9851164f821064d43b481ddbea0149c2d676c4f4e077b178e7eeaa6660/safetensors-0.5.2-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:5c5b5d9da594f638a259fca766046f44c97244cc7ab8bef161b3e80d04becc76", size = 632533 },
+    { url = "https://files.pythonhosted.org/packages/00/f1/5680e2ef61d9c61454fad82c344f0e40b8741a9dbd1e31484f0d31a9b1c3/safetensors-0.5.2-cp38-abi3-win32.whl", hash = "sha256:fe55c039d97090d1f85277d402954dd6ad27f63034fa81985a9cc59655ac3ee2", size = 291167 },
+    { url = "https://files.pythonhosted.org/packages/86/ca/aa489392ec6fb59223ffce825461e1f811a3affd417121a2088be7a5758b/safetensors-0.5.2-cp38-abi3-win_amd64.whl", hash = "sha256:78abdddd03a406646107f973c7843276e7b64e5e32623529dc17f3d94a20f589", size = 303756 },
+]
+
+[[package]]
+name = "scipy"
+version = "1.13.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
+    "python_full_version < '3.9.17'",
+]
+dependencies = [
+    { name = "numpy", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ae/00/48c2f661e2816ccf2ecd77982f6605b2950afe60f60a52b4cbbc2504aa8f/scipy-1.13.1.tar.gz", hash = "sha256:095a87a0312b08dfd6a6155cbbd310a8c51800fc931b8c0b84003014b874ed3c", size = 57210720 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/33/59/41b2529908c002ade869623b87eecff3e11e3ce62e996d0bdcb536984187/scipy-1.13.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:20335853b85e9a49ff7572ab453794298bcf0354d8068c5f6775a0eabf350aca", size = 39328076 },
+    { url = "https://files.pythonhosted.org/packages/d5/33/f1307601f492f764062ce7dd471a14750f3360e33cd0f8c614dae208492c/scipy-1.13.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:d605e9c23906d1994f55ace80e0125c587f96c020037ea6aa98d01b4bd2e222f", size = 30306232 },
+    { url = "https://files.pythonhosted.org/packages/c0/66/9cd4f501dd5ea03e4a4572ecd874936d0da296bd04d1c45ae1a4a75d9c3a/scipy-1.13.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cfa31f1def5c819b19ecc3a8b52d28ffdcc7ed52bb20c9a7589669dd3c250989", size = 33743202 },
+    { url = "https://files.pythonhosted.org/packages/a3/ba/7255e5dc82a65adbe83771c72f384d99c43063648456796436c9a5585ec3/scipy-1.13.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f26264b282b9da0952a024ae34710c2aff7d27480ee91a2e82b7b7073c24722f", size = 38577335 },
+    { url = "https://files.pythonhosted.org/packages/49/a5/bb9ded8326e9f0cdfdc412eeda1054b914dfea952bda2097d174f8832cc0/scipy-1.13.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:eccfa1906eacc02de42d70ef4aecea45415f5be17e72b61bafcfd329bdc52e94", size = 38820728 },
+    { url = "https://files.pythonhosted.org/packages/12/30/df7a8fcc08f9b4a83f5f27cfaaa7d43f9a2d2ad0b6562cced433e5b04e31/scipy-1.13.1-cp310-cp310-win_amd64.whl", hash = "sha256:2831f0dc9c5ea9edd6e51e6e769b655f08ec6db6e2e10f86ef39bd32eb11da54", size = 46210588 },
+    { url = "https://files.pythonhosted.org/packages/b4/15/4a4bb1b15bbd2cd2786c4f46e76b871b28799b67891f23f455323a0cdcfb/scipy-1.13.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:27e52b09c0d3a1d5b63e1105f24177e544a222b43611aaf5bc44d4a0979e32f9", size = 39333805 },
+    { url = "https://files.pythonhosted.org/packages/ba/92/42476de1af309c27710004f5cdebc27bec62c204db42e05b23a302cb0c9a/scipy-1.13.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:54f430b00f0133e2224c3ba42b805bfd0086fe488835effa33fa291561932326", size = 30317687 },
+    { url = "https://files.pythonhosted.org/packages/80/ba/8be64fe225360a4beb6840f3cbee494c107c0887f33350d0a47d55400b01/scipy-1.13.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e89369d27f9e7b0884ae559a3a956e77c02114cc60a6058b4e5011572eea9299", size = 33694638 },
+    { url = "https://files.pythonhosted.org/packages/36/07/035d22ff9795129c5a847c64cb43c1fa9188826b59344fee28a3ab02e283/scipy-1.13.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a78b4b3345f1b6f68a763c6e25c0c9a23a9fd0f39f5f3d200efe8feda560a5fa", size = 38569931 },
+    { url = "https://files.pythonhosted.org/packages/d9/10/f9b43de37e5ed91facc0cfff31d45ed0104f359e4f9a68416cbf4e790241/scipy-1.13.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:45484bee6d65633752c490404513b9ef02475b4284c4cfab0ef946def50b3f59", size = 38838145 },
+    { url = "https://files.pythonhosted.org/packages/4a/48/4513a1a5623a23e95f94abd675ed91cfb19989c58e9f6f7d03990f6caf3d/scipy-1.13.1-cp311-cp311-win_amd64.whl", hash = "sha256:5713f62f781eebd8d597eb3f88b8bf9274e79eeabf63afb4a737abc6c84ad37b", size = 46196227 },
+    { url = "https://files.pythonhosted.org/packages/f2/7b/fb6b46fbee30fc7051913068758414f2721003a89dd9a707ad49174e3843/scipy-1.13.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:5d72782f39716b2b3509cd7c33cdc08c96f2f4d2b06d51e52fb45a19ca0c86a1", size = 39357301 },
+    { url = "https://files.pythonhosted.org/packages/dc/5a/2043a3bde1443d94014aaa41e0b50c39d046dda8360abd3b2a1d3f79907d/scipy-1.13.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:017367484ce5498445aade74b1d5ab377acdc65e27095155e448c88497755a5d", size = 30363348 },
+    { url = "https://files.pythonhosted.org/packages/e7/cb/26e4a47364bbfdb3b7fb3363be6d8a1c543bcd70a7753ab397350f5f189a/scipy-1.13.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:949ae67db5fa78a86e8fa644b9a6b07252f449dcf74247108c50e1d20d2b4627", size = 33406062 },
+    { url = "https://files.pythonhosted.org/packages/88/ab/6ecdc526d509d33814835447bbbeedbebdec7cca46ef495a61b00a35b4bf/scipy-1.13.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de3ade0e53bc1f21358aa74ff4830235d716211d7d077e340c7349bc3542e884", size = 38218311 },
+    { url = "https://files.pythonhosted.org/packages/0b/00/9f54554f0f8318100a71515122d8f4f503b1a2c4b4cfab3b4b68c0eb08fa/scipy-1.13.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:2ac65fb503dad64218c228e2dc2d0a0193f7904747db43014645ae139c8fad16", size = 38442493 },
+    { url = "https://files.pythonhosted.org/packages/3e/df/963384e90733e08eac978cd103c34df181d1fec424de383cdc443f418dd4/scipy-1.13.1-cp312-cp312-win_amd64.whl", hash = "sha256:cdd7dacfb95fea358916410ec61bbc20440f7860333aee6d882bb8046264e949", size = 45910955 },
+    { url = "https://files.pythonhosted.org/packages/7f/29/c2ea58c9731b9ecb30b6738113a95d147e83922986b34c685b8f6eefde21/scipy-1.13.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:436bbb42a94a8aeef855d755ce5a465479c721e9d684de76bf61a62e7c2b81d5", size = 39352927 },
+    { url = "https://files.pythonhosted.org/packages/5c/c0/e71b94b20ccf9effb38d7147c0064c08c622309fd487b1b677771a97d18c/scipy-1.13.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:8335549ebbca860c52bf3d02f80784e91a004b71b059e3eea9678ba994796a24", size = 30324538 },
+    { url = "https://files.pythonhosted.org/packages/6d/0f/aaa55b06d474817cea311e7b10aab2ea1fd5d43bc6a2861ccc9caec9f418/scipy-1.13.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d533654b7d221a6a97304ab63c41c96473ff04459e404b83275b60aa8f4b7004", size = 33732190 },
+    { url = "https://files.pythonhosted.org/packages/35/f5/d0ad1a96f80962ba65e2ce1de6a1e59edecd1f0a7b55990ed208848012e0/scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:637e98dcf185ba7f8e663e122ebf908c4702420477ae52a04f9908707456ba4d", size = 38612244 },
+    { url = "https://files.pythonhosted.org/packages/8d/02/1165905f14962174e6569076bcc3315809ae1291ed14de6448cc151eedfd/scipy-1.13.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a014c2b3697bde71724244f63de2476925596c24285c7a637364761f8710891c", size = 38845637 },
+    { url = "https://files.pythonhosted.org/packages/3e/77/dab54fe647a08ee4253963bcd8f9cf17509c8ca64d6335141422fe2e2114/scipy-1.13.1-cp39-cp39-win_amd64.whl", hash = "sha256:392e4ec766654852c25ebad4f64e4e584cf19820b980bc04960bca0b0cd6eaa2", size = 46227440 },
+]
+
+[[package]]
+name = "scipy"
+version = "1.15.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.13'",
+    "python_full_version == '3.12.*'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "numpy", marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/76/c6/8eb0654ba0c7d0bb1bf67bf8fbace101a8e4f250f7722371105e8b6f68fc/scipy-1.15.1.tar.gz", hash = "sha256:033a75ddad1463970c96a88063a1df87ccfddd526437136b6ee81ff0312ebdf6", size = 59407493 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/86/53/b204ce5a4433f1864001b9d16f103b9c25f5002a602ae83585d0ea5f9c4a/scipy-1.15.1-cp310-cp310-macosx_10_13_x86_64.whl", hash = "sha256:c64ded12dcab08afff9e805a67ff4480f5e69993310e093434b10e85dc9d43e1", size = 41414518 },
+    { url = "https://files.pythonhosted.org/packages/c7/fc/54ffa7a8847f7f303197a6ba65a66104724beba2e38f328135a78f0dc480/scipy-1.15.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:5b190b935e7db569960b48840e5bef71dc513314cc4e79a1b7d14664f57fd4ff", size = 32519265 },
+    { url = "https://files.pythonhosted.org/packages/f1/77/a98b8ba03d6f371dc31a38719affd53426d4665729dcffbed4afe296784a/scipy-1.15.1-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:4b17d4220df99bacb63065c76b0d1126d82bbf00167d1730019d2a30d6ae01ea", size = 24792859 },
+    { url = "https://files.pythonhosted.org/packages/a7/78/70bb9f0df7444b18b108580934bfef774822e28fd34a68e5c263c7d2828a/scipy-1.15.1-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:63b9b6cd0333d0eb1a49de6f834e8aeaefe438df8f6372352084535ad095219e", size = 27886506 },
+    { url = "https://files.pythonhosted.org/packages/14/a7/f40f6033e06de4176ddd6cc8c3ae9f10a226c3bca5d6b4ab883bc9914a14/scipy-1.15.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9f151e9fb60fbf8e52426132f473221a49362091ce7a5e72f8aa41f8e0da4f25", size = 38375041 },
+    { url = "https://files.pythonhosted.org/packages/17/03/390a1c5c61fd76b0fa4b3c5aa3bdd7e60f6c46f712924f1a9df5705ec046/scipy-1.15.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:21e10b1dd56ce92fba3e786007322542361984f8463c6d37f6f25935a5a6ef52", size = 40597556 },
+    { url = "https://files.pythonhosted.org/packages/4e/70/fa95b3ae026b97eeca58204a90868802e5155ac71b9d7bdee92b68115dd3/scipy-1.15.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5dff14e75cdbcf07cdaa1c7707db6017d130f0af9ac41f6ce443a93318d6c6e0", size = 42938505 },
+    { url = "https://files.pythonhosted.org/packages/d6/07/427859116bdd71847c898180f01802691f203c3e2455a1eb496130ff07c5/scipy-1.15.1-cp310-cp310-win_amd64.whl", hash = "sha256:f82fcf4e5b377f819542fbc8541f7b5fbcf1c0017d0df0bc22c781bf60abc4d8", size = 43909663 },
+    { url = "https://files.pythonhosted.org/packages/8e/2e/7b71312da9c2dabff53e7c9a9d08231bc34d9d8fdabe88a6f1155b44591c/scipy-1.15.1-cp311-cp311-macosx_10_13_x86_64.whl", hash = "sha256:5bd8d27d44e2c13d0c1124e6a556454f52cd3f704742985f6b09e75e163d20d2", size = 41424362 },
+    { url = "https://files.pythonhosted.org/packages/81/8c/ab85f1aa1cc200c796532a385b6ebf6a81089747adc1da7482a062acc46c/scipy-1.15.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:be3deeb32844c27599347faa077b359584ba96664c5c79d71a354b80a0ad0ce0", size = 32535910 },
+    { url = "https://files.pythonhosted.org/packages/3b/9c/6f4b787058daa8d8da21ddff881b4320e28de4704a65ec147adb50cb2230/scipy-1.15.1-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:5eb0ca35d4b08e95da99a9f9c400dc9f6c21c424298a0ba876fdc69c7afacedf", size = 24809398 },
+    { url = "https://files.pythonhosted.org/packages/16/2b/949460a796df75fc7a1ee1becea202cf072edbe325ebe29f6d2029947aa7/scipy-1.15.1-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:74bb864ff7640dea310a1377d8567dc2cb7599c26a79ca852fc184cc851954ac", size = 27918045 },
+    { url = "https://files.pythonhosted.org/packages/5f/36/67fe249dd7ccfcd2a38b25a640e3af7e59d9169c802478b6035ba91dfd6d/scipy-1.15.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:667f950bf8b7c3a23b4199db24cb9bf7512e27e86d0e3813f015b74ec2c6e3df", size = 38332074 },
+    { url = "https://files.pythonhosted.org/packages/fc/da/452e1119e6f720df3feb588cce3c42c5e3d628d4bfd4aec097bd30b7de0c/scipy-1.15.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:395be70220d1189756068b3173853029a013d8c8dd5fd3d1361d505b2aa58fa7", size = 40588469 },
+    { url = "https://files.pythonhosted.org/packages/7f/71/5f94aceeac99a4941478af94fe9f459c6752d497035b6b0761a700f5f9ff/scipy-1.15.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ce3a000cd28b4430426db2ca44d96636f701ed12e2b3ca1f2b1dd7abdd84b39a", size = 42965214 },
+    { url = "https://files.pythonhosted.org/packages/af/25/caa430865749d504271757cafd24066d596217e83326155993980bc22f97/scipy-1.15.1-cp311-cp311-win_amd64.whl", hash = "sha256:3fe1d95944f9cf6ba77aa28b82dd6bb2a5b52f2026beb39ecf05304b8392864b", size = 43896034 },
+    { url = "https://files.pythonhosted.org/packages/d8/6e/a9c42d0d39e09ed7fd203d0ac17adfea759cba61ab457671fe66e523dbec/scipy-1.15.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c09aa9d90f3500ea4c9b393ee96f96b0ccb27f2f350d09a47f533293c78ea776", size = 41478318 },
+    { url = "https://files.pythonhosted.org/packages/04/ee/e3e535c81828618878a7433992fecc92fa4df79393f31a8fea1d05615091/scipy-1.15.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:0ac102ce99934b162914b1e4a6b94ca7da0f4058b6d6fd65b0cef330c0f3346f", size = 32596696 },
+    { url = "https://files.pythonhosted.org/packages/c4/5e/b1b0124be8e76f87115f16b8915003eec4b7060298117715baf13f51942c/scipy-1.15.1-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:09c52320c42d7f5c7748b69e9f0389266fd4f82cf34c38485c14ee976cb8cb04", size = 24870366 },
+    { url = "https://files.pythonhosted.org/packages/14/36/c00cb73eefda85946172c27913ab995c6ad4eee00fa4f007572e8c50cd51/scipy-1.15.1-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:cdde8414154054763b42b74fe8ce89d7f3d17a7ac5dd77204f0e142cdc9239e9", size = 28007461 },
+    { url = "https://files.pythonhosted.org/packages/68/94/aff5c51b3799349a9d1e67a056772a0f8a47db371e83b498d43467806557/scipy-1.15.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4c9d8fc81d6a3b6844235e6fd175ee1d4c060163905a2becce8e74cb0d7554ce", size = 38068174 },
+    { url = "https://files.pythonhosted.org/packages/b0/3c/0de11ca154e24a57b579fb648151d901326d3102115bc4f9a7a86526ce54/scipy-1.15.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0fb57b30f0017d4afa5fe5f5b150b8f807618819287c21cbe51130de7ccdaed2", size = 40249869 },
+    { url = "https://files.pythonhosted.org/packages/15/09/472e8d0a6b33199d1bb95e49bedcabc0976c3724edd9b0ef7602ccacf41e/scipy-1.15.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:491d57fe89927fa1aafbe260f4cfa5ffa20ab9f1435025045a5315006a91b8f5", size = 42629068 },
+    { url = "https://files.pythonhosted.org/packages/ff/ba/31c7a8131152822b3a2cdeba76398ffb404d81d640de98287d236da90c49/scipy-1.15.1-cp312-cp312-win_amd64.whl", hash = "sha256:900f3fa3db87257510f011c292a5779eb627043dd89731b9c461cd16ef76ab3d", size = 43621992 },
+    { url = "https://files.pythonhosted.org/packages/2b/bf/dd68965a4c5138a630eeed0baec9ae96e5d598887835bdde96cdd2fe4780/scipy-1.15.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:100193bb72fbff37dbd0bf14322314fc7cbe08b7ff3137f11a34d06dc0ee6b85", size = 41441136 },
+    { url = "https://files.pythonhosted.org/packages/ef/5e/4928581312922d7e4d416d74c416a660addec4dd5ea185401df2269ba5a0/scipy-1.15.1-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:2114a08daec64980e4b4cbdf5bee90935af66d750146b1d2feb0d3ac30613692", size = 32533699 },
+    { url = "https://files.pythonhosted.org/packages/32/90/03f99c43041852837686898c66767787cd41c5843d7a1509c39ffef683e9/scipy-1.15.1-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:6b3e71893c6687fc5e29208d518900c24ea372a862854c9888368c0b267387ab", size = 24807289 },
+    { url = "https://files.pythonhosted.org/packages/9d/52/bfe82b42ae112eaba1af2f3e556275b8727d55ac6e4932e7aef337a9d9d4/scipy-1.15.1-cp313-cp313-macosx_14_0_x86_64.whl", hash = "sha256:837299eec3d19b7e042923448d17d95a86e43941104d33f00da7e31a0f715d3c", size = 27929844 },
+    { url = "https://files.pythonhosted.org/packages/f6/77/54ff610bad600462c313326acdb035783accc6a3d5f566d22757ad297564/scipy-1.15.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:82add84e8a9fb12af5c2c1a3a3f1cb51849d27a580cb9e6bd66226195142be6e", size = 38031272 },
+    { url = "https://files.pythonhosted.org/packages/f1/26/98585cbf04c7cf503d7eb0a1966df8a268154b5d923c5fe0c1ed13154c49/scipy-1.15.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:070d10654f0cb6abd295bc96c12656f948e623ec5f9a4eab0ddb1466c000716e", size = 40210217 },
+    { url = "https://files.pythonhosted.org/packages/fd/3f/3d2285eb6fece8bc5dbb2f9f94d61157d61d155e854fd5fea825b8218f12/scipy-1.15.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:55cc79ce4085c702ac31e49b1e69b27ef41111f22beafb9b49fea67142b696c4", size = 42587785 },
+    { url = "https://files.pythonhosted.org/packages/48/7d/5b5251984bf0160d6533695a74a5fddb1fa36edd6f26ffa8c871fbd4782a/scipy-1.15.1-cp313-cp313-win_amd64.whl", hash = "sha256:c352c1b6d7cac452534517e022f8f7b8d139cd9f27e6fbd9f3cbd0bfd39f5bef", size = 43640439 },
+    { url = "https://files.pythonhosted.org/packages/e7/b8/0e092f592d280496de52e152582030f8a270b194f87f890e1a97c5599b81/scipy-1.15.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:0458839c9f873062db69a03de9a9765ae2e694352c76a16be44f93ea45c28d2b", size = 41619862 },
+    { url = "https://files.pythonhosted.org/packages/f6/19/0b6e1173aba4db9e0b7aa27fe45019857fb90d6904038b83927cbe0a6c1d/scipy-1.15.1-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:af0b61c1de46d0565b4b39c6417373304c1d4f5220004058bdad3061c9fa8a95", size = 32610387 },
+    { url = "https://files.pythonhosted.org/packages/e7/02/754aae3bd1fa0f2479ade3cfdf1732ecd6b05853f63eee6066a32684563a/scipy-1.15.1-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:71ba9a76c2390eca6e359be81a3e879614af3a71dfdabb96d1d7ab33da6f2364", size = 24883814 },
+    { url = "https://files.pythonhosted.org/packages/1f/ac/d7906201604a2ea3b143bb0de51b3966f66441ba50b7dc182c4505b3edf9/scipy-1.15.1-cp313-cp313t-macosx_14_0_x86_64.whl", hash = "sha256:14eaa373c89eaf553be73c3affb11ec6c37493b7eaaf31cf9ac5dffae700c2e0", size = 27944865 },
+    { url = "https://files.pythonhosted.org/packages/84/9d/8f539002b5e203723af6a6f513a45e0a7671e9dabeedb08f417ac17e4edc/scipy-1.15.1-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f735bc41bd1c792c96bc426dece66c8723283695f02df61dcc4d0a707a42fc54", size = 39883261 },
+    { url = "https://files.pythonhosted.org/packages/97/c0/62fd3bab828bcccc9b864c5997645a3b86372a35941cdaf677565c25c98d/scipy-1.15.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:2722a021a7929d21168830790202a75dbb20b468a8133c74a2c0230c72626b6c", size = 42093299 },
+    { url = "https://files.pythonhosted.org/packages/e4/1f/5d46a8d94e9f6d2c913cbb109e57e7eed914de38ea99e2c4d69a9fc93140/scipy-1.15.1-cp313-cp313t-win_amd64.whl", hash = "sha256:bc7136626261ac1ed988dca56cfc4ab5180f75e0ee52e58f1e6aa74b5f3eacd5", size = 43181730 },
+]
+
+[[package]]
+name = "sentencepiece"
+version = "0.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/c9/d2/b9c7ca067c26d8ff085d252c89b5f69609ca93fb85a00ede95f4857865d4/sentencepiece-0.2.0.tar.gz", hash = "sha256:a52c19171daaf2e697dc6cbe67684e0fa341b1248966f6aebb541de654d15843", size = 2632106 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f6/71/98648c3b64b23edb5403f74bcc906ad21766872a6e1ada26ea3f1eb941ab/sentencepiece-0.2.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:188779e1298a1c8b8253c7d3ad729cb0a9891e5cef5e5d07ce4592c54869e227", size = 2408979 },
+    { url = "https://files.pythonhosted.org/packages/77/9f/7efbaa6d4c0c718a9affbecc536b03ca62f99f421bdffb531c16030e2d2b/sentencepiece-0.2.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:bed9cf85b296fa2b76fc2547b9cbb691a523864cebaee86304c43a7b4cb1b452", size = 1238845 },
+    { url = "https://files.pythonhosted.org/packages/1c/e4/c2541027a43ec6962ba9b601805d17ba3f86b38bdeae0e8ac65a2981e248/sentencepiece-0.2.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d7b67e724bead13f18db6e1d10b6bbdc454af574d70efbb36f27d90387be1ca3", size = 1181472 },
+    { url = "https://files.pythonhosted.org/packages/fd/46/316c1ba6c52b97de76aff7b9da678f7afbb52136afb2987c474d95630e65/sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2fde4b08cfe237be4484c6c7c2e2c75fb862cfeab6bd5449ce4caeafd97b767a", size = 1259151 },
+    { url = "https://files.pythonhosted.org/packages/aa/5a/3c48738a0835d76dd06c62b6ac48d39c923cde78dd0f587353bdcbb99851/sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4c378492056202d1c48a4979650981635fd97875a00eabb1f00c6a236b013b5e", size = 1355931 },
+    { url = "https://files.pythonhosted.org/packages/a6/27/33019685023221ca8ed98e8ceb7ae5e166032686fa3662c68f1f1edf334e/sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1380ce6540a368de2ef6d7e6ba14ba8f3258df650d39ba7d833b79ee68a52040", size = 1301537 },
+    { url = "https://files.pythonhosted.org/packages/ca/e4/55f97cef14293171fef5f96e96999919ab5b4d1ce95b53547ad653d7e3bf/sentencepiece-0.2.0-cp310-cp310-win32.whl", hash = "sha256:a1151d6a6dd4b43e552394aed0edfe9292820272f0194bd56c7c1660a0c06c3d", size = 936747 },
+    { url = "https://files.pythonhosted.org/packages/85/f4/4ef1a6e0e9dbd8a60780a91df8b7452ada14cfaa0e17b3b8dfa42cecae18/sentencepiece-0.2.0-cp310-cp310-win_amd64.whl", hash = "sha256:d490142b0521ef22bc1085f061d922a2a6666175bb6b42e588ff95c0db6819b2", size = 991525 },
+    { url = "https://files.pythonhosted.org/packages/32/43/8f8885168a47a02eba1455bd3f4f169f50ad5b8cebd2402d0f5e20854d04/sentencepiece-0.2.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:17982700c4f6dbb55fa3594f3d7e5dd1c8659a274af3738e33c987d2a27c9d5c", size = 2409036 },
+    { url = "https://files.pythonhosted.org/packages/0f/35/e63ba28062af0a3d688a9f128e407a1a2608544b2f480cb49bf7f4b1cbb9/sentencepiece-0.2.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:7c867012c0e8bcd5bdad0f791609101cb5c66acb303ab3270218d6debc68a65e", size = 1238921 },
+    { url = "https://files.pythonhosted.org/packages/de/42/ae30952c4a0bd773e90c9bf2579f5533037c886dfc8ec68133d5694f4dd2/sentencepiece-0.2.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7fd6071249c74f779c5b27183295b9202f8dedb68034e716784364443879eaa6", size = 1181477 },
+    { url = "https://files.pythonhosted.org/packages/e3/ac/2f2ab1d60bb2d795d054eebe5e3f24b164bc21b5a9b75fba7968b3b91b5a/sentencepiece-0.2.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:27f90c55a65013cbb8f4d7aab0599bf925cde4adc67ae43a0d323677b5a1c6cb", size = 1259182 },
+    { url = "https://files.pythonhosted.org/packages/45/fb/14633c6ecf262c468759ffcdb55c3a7ee38fe4eda6a70d75ee7c7d63c58b/sentencepiece-0.2.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b293734059ef656dcd65be62ff771507bea8fed0a711b6733976e1ed3add4553", size = 1355537 },
+    { url = "https://files.pythonhosted.org/packages/fb/12/2f5c8d4764b00033cf1c935b702d3bb878d10be9f0b87f0253495832d85f/sentencepiece-0.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e58b47f933aca74c6a60a79dcb21d5b9e47416256c795c2d58d55cec27f9551d", size = 1301464 },
+    { url = "https://files.pythonhosted.org/packages/4e/b1/67afc0bde24f6dcb3acdea0dd8dcdf4b8b0db240f6bacd39378bd32d09f8/sentencepiece-0.2.0-cp311-cp311-win32.whl", hash = "sha256:c581258cf346b327c62c4f1cebd32691826306f6a41d8c4bec43b010dee08e75", size = 936749 },
+    { url = "https://files.pythonhosted.org/packages/a2/f6/587c62fd21fc988555b85351f50bbde43a51524caafd63bc69240ded14fd/sentencepiece-0.2.0-cp311-cp311-win_amd64.whl", hash = "sha256:0993dbc665f4113017892f1b87c3904a44d0640eda510abcacdfb07f74286d36", size = 991520 },
+    { url = "https://files.pythonhosted.org/packages/27/5a/141b227ed54293360a9ffbb7bf8252b4e5efc0400cdeac5809340e5d2b21/sentencepiece-0.2.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:ea5f536e32ea8ec96086ee00d7a4a131ce583a1b18d130711707c10e69601cb2", size = 2409370 },
+    { url = "https://files.pythonhosted.org/packages/2e/08/a4c135ad6fc2ce26798d14ab72790d66e813efc9589fd30a5316a88ca8d5/sentencepiece-0.2.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:d0cb51f53b6aae3c36bafe41e86167c71af8370a039f542c43b0cce5ef24a68c", size = 1239288 },
+    { url = "https://files.pythonhosted.org/packages/49/0a/2fe387f825ac5aad5a0bfe221904882106cac58e1b693ba7818785a882b6/sentencepiece-0.2.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3212121805afc58d8b00ab4e7dd1f8f76c203ddb9dc94aa4079618a31cf5da0f", size = 1181597 },
+    { url = "https://files.pythonhosted.org/packages/cc/38/e4698ee2293fe4835dc033c49796a39b3eebd8752098f6bd0aa53a14af1f/sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2a3149e3066c2a75e0d68a43eb632d7ae728c7925b517f4c05c40f6f7280ce08", size = 1259220 },
+    { url = "https://files.pythonhosted.org/packages/12/24/fd7ef967c9dad2f6e6e5386d0cadaf65cda8b7be6e3861a9ab3121035139/sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:632f3594d3e7ac8b367bca204cb3fd05a01d5b21455acd097ea4c0e30e2f63d7", size = 1355962 },
+    { url = "https://files.pythonhosted.org/packages/4f/d2/18246f43ca730bb81918f87b7e886531eda32d835811ad9f4657c54eee35/sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f295105c6bdbb05bd5e1b0cafbd78ff95036f5d3641e7949455a3f4e5e7c3109", size = 1301706 },
+    { url = "https://files.pythonhosted.org/packages/8a/47/ca237b562f420044ab56ddb4c278672f7e8c866e183730a20e413b38a989/sentencepiece-0.2.0-cp312-cp312-win32.whl", hash = "sha256:fb89f811e5efd18bab141afc3fea3de141c3f69f3fe9e898f710ae7fe3aab251", size = 936941 },
+    { url = "https://files.pythonhosted.org/packages/c6/97/d159c32642306ee2b70732077632895438867b3b6df282354bd550cf2a67/sentencepiece-0.2.0-cp312-cp312-win_amd64.whl", hash = "sha256:7a673a72aab81fef5ebe755c6e0cc60087d1f3a4700835d40537183c1703a45f", size = 991994 },
+    { url = "https://files.pythonhosted.org/packages/e9/18/eb620d94d63f62ca69cecccf4459529864ac3fbb35ec123190bd58dadb46/sentencepiece-0.2.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:1e0f9c4d0a6b0af59b613175f019916e28ade076e21242fd5be24340d8a2f64a", size = 2409003 },
+    { url = "https://files.pythonhosted.org/packages/6e/a6/df28bc0b6a2a86416232c0a5f0d69a9cb7244bb95cb5dcdfcbf01cced8a6/sentencepiece-0.2.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:298f21cc1366eb60311aedba3169d30f885c363ddbf44214b0a587d2908141ad", size = 1238898 },
+    { url = "https://files.pythonhosted.org/packages/79/91/b54a528e0789cd7986341ed3909bec56365c3b672daef8b10aa4098238f0/sentencepiece-0.2.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3f1ec95aa1e5dab11f37ac7eff190493fd87770f7a8b81ebc9dd768d1a3c8704", size = 1181534 },
+    { url = "https://files.pythonhosted.org/packages/a3/69/e96ef68261fa5b82379fdedb325ceaf1d353c6e839ec346d8244e0da5f2f/sentencepiece-0.2.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7b06b70af54daa4b4904cbb90b4eb6d35c9f3252fdc86c9c32d5afd4d30118d8", size = 1259161 },
+    { url = "https://files.pythonhosted.org/packages/45/de/461d15856c29ba1ce778cf76e0462572661f647abc8a5373690c52e98a00/sentencepiece-0.2.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:22e37bac44dd6603388cb598c64ff7a76e41ca774646f21c23aadfbf5a2228ab", size = 1355945 },
+    { url = "https://files.pythonhosted.org/packages/5f/01/c95e42eb86282b2c79305d3e0b0ca5a743f85a61262bb7130999c70b9374/sentencepiece-0.2.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0461324897735512a32d222e3d886e24ad6a499761952b6bda2a9ee6e4313ea5", size = 1301596 },
+    { url = "https://files.pythonhosted.org/packages/be/47/e16f368fe6327e873e8029aa539115025e9f61a4e8ca8f0f8eaf8e6a4c1c/sentencepiece-0.2.0-cp39-cp39-win32.whl", hash = "sha256:38aed822fb76435fa1f12185f10465a94ab9e51d5e8a9159e9a540ce926f0ffd", size = 936757 },
+    { url = "https://files.pythonhosted.org/packages/4b/36/497e6407700efd6b97f81bc160913a70d33b9b09227429f68fc86f387bbe/sentencepiece-0.2.0-cp39-cp39-win_amd64.whl", hash = "sha256:d8cf876516548b5a1d6ac4745d8b554f5c07891d55da557925e5c13ff0b4e6ad", size = 991541 },
+]
+
+[[package]]
+name = "six"
+version = "1.17.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/94/e7/b2c673351809dca68a0e064b6af791aa332cf192da575fd474ed7d6f16a2/six-1.17.0.tar.gz", hash = "sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81", size = 34031 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b7/ce/149a00dd41f10bc29e5921b496af8b574d8413afcd5e30dfa0ed46c2cc5e/six-1.17.0-py2.py3-none-any.whl", hash = "sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274", size = 11050 },
+]
+
+[[package]]
+name = "sniffio"
+version = "1.3.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/a2/87/a6771e1546d97e7e041b6ae58d80074f81b7d5121207425c964ddf5cfdbd/sniffio-1.3.1.tar.gz", hash = "sha256:f4324edc670a0f49750a81b895f35c3adb843cca46f0530f79fc1babb23789dc", size = 20372 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2", size = 10235 },
+]
+
+[[package]]
+name = "starlette"
+version = "0.41.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/1a/4c/9b5764bd22eec91c4039ef4c55334e9187085da2d8a2df7bd570869aae18/starlette-0.41.3.tar.gz", hash = "sha256:0e4ab3d16522a255be6b28260b938eae2482f98ce5cc934cb08dce8dc3ba5835", size = 2574159 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/96/00/2b325970b3060c7cecebab6d295afe763365822b1306a12eeab198f74323/starlette-0.41.3-py3-none-any.whl", hash = "sha256:44cedb2b7c77a9de33a8b74b2b90e9f50d11fcf25d8270ea525ad71a25374ff7", size = 73225 },
+]
+
+[[package]]
+name = "sympy"
+version = "1.13.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "mpmath" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/11/8a/5a7fd6284fa8caac23a26c9ddf9c30485a48169344b4bd3b0f02fef1890f/sympy-1.13.3.tar.gz", hash = "sha256:b27fd2c6530e0ab39e275fc9b683895367e51d5da91baa8d3d64db2565fec4d9", size = 7533196 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/99/ff/c87e0622b1dadea79d2fb0b25ade9ed98954c9033722eb707053d310d4f3/sympy-1.13.3-py3-none-any.whl", hash = "sha256:54612cf55a62755ee71824ce692986f23c88ffa77207b30c1368eda4a7060f73", size = 6189483 },
+]
+
+[[package]]
+name = "tbb"
+version = "2021.13.1"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fb/8a/5062b00c378c051e26507e5eca8d3b5c91ed63f8a2139f6f0f422be84b02/tbb-2021.13.1-py3-none-win32.whl", hash = "sha256:00f5e5a70051650ddd0ab6247c0549521968339ec21002e475cd23b1cbf46d66", size = 248994 },
+    { url = "https://files.pythonhosted.org/packages/9b/24/84ce997e8ae6296168a74d0d9c4dde572d90fb23fd7c0b219c30ff71e00e/tbb-2021.13.1-py3-none-win_amd64.whl", hash = "sha256:cbf024b2463fdab3ebe3fa6ff453026358e6b903839c80d647e08ad6d0796ee9", size = 286908 },
+]
+
+[[package]]
+name = "tiktoken"
+version = "0.8.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "regex" },
+    { name = "requests" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/37/02/576ff3a6639e755c4f70997b2d315f56d6d71e0d046f4fb64cb81a3fb099/tiktoken-0.8.0.tar.gz", hash = "sha256:9ccbb2740f24542534369c5635cfd9b2b3c2490754a78ac8831d99f89f94eeb2", size = 35107 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c9/ba/a35fad753bbca8ba0cc1b0f3402a70256a110ced7ac332cf84ba89fc87ab/tiktoken-0.8.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b07e33283463089c81ef1467180e3e00ab00d46c2c4bbcef0acab5f771d6695e", size = 1039905 },
+    { url = "https://files.pythonhosted.org/packages/91/05/13dab8fd7460391c387b3e69e14bf1e51ff71fe0a202cd2933cc3ea93fb6/tiktoken-0.8.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9269348cb650726f44dd3bbb3f9110ac19a8dcc8f54949ad3ef652ca22a38e21", size = 982417 },
+    { url = "https://files.pythonhosted.org/packages/e9/98/18ec4a8351a6cf4537e40cd6e19a422c10cce1ef00a2fcb716e0a96af58b/tiktoken-0.8.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:25e13f37bc4ef2d012731e93e0fef21dc3b7aea5bb9009618de9a4026844e560", size = 1144915 },
+    { url = "https://files.pythonhosted.org/packages/2e/28/cf3633018cbcc6deb7805b700ccd6085c9a5a7f72b38974ee0bffd56d311/tiktoken-0.8.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f13d13c981511331eac0d01a59b5df7c0d4060a8be1e378672822213da51e0a2", size = 1177221 },
+    { url = "https://files.pythonhosted.org/packages/57/81/8a5be305cbd39d4e83a794f9e80c7f2c84b524587b7feb27c797b2046d51/tiktoken-0.8.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:6b2ddbc79a22621ce8b1166afa9f9a888a664a579350dc7c09346a3b5de837d9", size = 1237398 },
+    { url = "https://files.pythonhosted.org/packages/dc/da/8d1cc3089a83f5cf11c2e489332752981435280285231924557350523a59/tiktoken-0.8.0-cp310-cp310-win_amd64.whl", hash = "sha256:d8c2d0e5ba6453a290b86cd65fc51fedf247e1ba170191715b049dac1f628005", size = 884215 },
+    { url = "https://files.pythonhosted.org/packages/f6/1e/ca48e7bfeeccaf76f3a501bd84db1fa28b3c22c9d1a1f41af9fb7579c5f6/tiktoken-0.8.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:d622d8011e6d6f239297efa42a2657043aaed06c4f68833550cac9e9bc723ef1", size = 1039700 },
+    { url = "https://files.pythonhosted.org/packages/8c/f8/f0101d98d661b34534769c3818f5af631e59c36ac6d07268fbfc89e539ce/tiktoken-0.8.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2efaf6199717b4485031b4d6edb94075e4d79177a172f38dd934d911b588d54a", size = 982413 },
+    { url = "https://files.pythonhosted.org/packages/ac/3c/2b95391d9bd520a73830469f80a96e3790e6c0a5ac2444f80f20b4b31051/tiktoken-0.8.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5637e425ce1fc49cf716d88df3092048359a4b3bbb7da762840426e937ada06d", size = 1144242 },
+    { url = "https://files.pythonhosted.org/packages/01/c4/c4a4360de845217b6aa9709c15773484b50479f36bb50419c443204e5de9/tiktoken-0.8.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9fb0e352d1dbe15aba082883058b3cce9e48d33101bdaac1eccf66424feb5b47", size = 1176588 },
+    { url = "https://files.pythonhosted.org/packages/f8/a3/ef984e976822cd6c2227c854f74d2e60cf4cd6fbfca46251199914746f78/tiktoken-0.8.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:56edfefe896c8f10aba372ab5706b9e3558e78db39dd497c940b47bf228bc419", size = 1237261 },
+    { url = "https://files.pythonhosted.org/packages/1e/86/eea2309dc258fb86c7d9b10db536434fc16420feaa3b6113df18b23db7c2/tiktoken-0.8.0-cp311-cp311-win_amd64.whl", hash = "sha256:326624128590def898775b722ccc327e90b073714227175ea8febbc920ac0a99", size = 884537 },
+    { url = "https://files.pythonhosted.org/packages/c1/22/34b2e136a6f4af186b6640cbfd6f93400783c9ef6cd550d9eab80628d9de/tiktoken-0.8.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:881839cfeae051b3628d9823b2e56b5cc93a9e2efb435f4cf15f17dc45f21586", size = 1039357 },
+    { url = "https://files.pythonhosted.org/packages/04/d2/c793cf49c20f5855fd6ce05d080c0537d7418f22c58e71f392d5e8c8dbf7/tiktoken-0.8.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:fe9399bdc3f29d428f16a2f86c3c8ec20be3eac5f53693ce4980371c3245729b", size = 982616 },
+    { url = "https://files.pythonhosted.org/packages/b3/a1/79846e5ef911cd5d75c844de3fa496a10c91b4b5f550aad695c5df153d72/tiktoken-0.8.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9a58deb7075d5b69237a3ff4bb51a726670419db6ea62bdcd8bd80c78497d7ab", size = 1144011 },
+    { url = "https://files.pythonhosted.org/packages/26/32/e0e3a859136e95c85a572e4806dc58bf1ddf651108ae8b97d5f3ebe1a244/tiktoken-0.8.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d2908c0d043a7d03ebd80347266b0e58440bdef5564f84f4d29fb235b5df3b04", size = 1175432 },
+    { url = "https://files.pythonhosted.org/packages/c7/89/926b66e9025b97e9fbabeaa59048a736fe3c3e4530a204109571104f921c/tiktoken-0.8.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:294440d21a2a51e12d4238e68a5972095534fe9878be57d905c476017bff99fc", size = 1236576 },
+    { url = "https://files.pythonhosted.org/packages/45/e2/39d4aa02a52bba73b2cd21ba4533c84425ff8786cc63c511d68c8897376e/tiktoken-0.8.0-cp312-cp312-win_amd64.whl", hash = "sha256:d8f3192733ac4d77977432947d563d7e1b310b96497acd3c196c9bddb36ed9db", size = 883824 },
+    { url = "https://files.pythonhosted.org/packages/e3/38/802e79ba0ee5fcbf240cd624143f57744e5d411d2e9d9ad2db70d8395986/tiktoken-0.8.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:02be1666096aff7da6cbd7cdaa8e7917bfed3467cd64b38b1f112e96d3b06a24", size = 1039648 },
+    { url = "https://files.pythonhosted.org/packages/b1/da/24cdbfc302c98663fbea66f5866f7fa1048405c7564ab88483aea97c3b1a/tiktoken-0.8.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:c94ff53c5c74b535b2cbf431d907fc13c678bbd009ee633a2aca269a04389f9a", size = 982763 },
+    { url = "https://files.pythonhosted.org/packages/e4/f0/0ecf79a279dfa41fc97d00adccf976ecc2556d3c08ef3e25e45eb31f665b/tiktoken-0.8.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6b231f5e8982c245ee3065cd84a4712d64692348bc609d84467c57b4b72dcbc5", size = 1144417 },
+    { url = "https://files.pythonhosted.org/packages/ab/d3/155d2d4514f3471a25dc1d6d20549ef254e2aa9bb5b1060809b1d3b03d3a/tiktoken-0.8.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4177faa809bd55f699e88c96d9bb4635d22e3f59d635ba6fd9ffedf7150b9953", size = 1175108 },
+    { url = "https://files.pythonhosted.org/packages/19/eb/5989e16821ee8300ef8ee13c16effc20dfc26c777d05fbb6825e3c037b81/tiktoken-0.8.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5376b6f8dc4753cd81ead935c5f518fa0fbe7e133d9e25f648d8c4dabdd4bad7", size = 1236520 },
+    { url = "https://files.pythonhosted.org/packages/40/59/14b20465f1d1cb89cfbc96ec27e5617b2d41c79da12b5e04e96d689be2a7/tiktoken-0.8.0-cp313-cp313-win_amd64.whl", hash = "sha256:18228d624807d66c87acd8f25fc135665617cab220671eb65b50f5d70fa51f69", size = 883849 },
+    { url = "https://files.pythonhosted.org/packages/08/f3/8a8ba9329e6b426d822c974d58fc6477f3f7b3b8deef651813d275cbe75f/tiktoken-0.8.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7e17807445f0cf1f25771c9d86496bd8b5c376f7419912519699f3cc4dc5c12e", size = 1040915 },
+    { url = "https://files.pythonhosted.org/packages/42/7a/914bd98100449422778f9222d00b3a4ee654211c40784e57541fa46311ab/tiktoken-0.8.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:886f80bd339578bbdba6ed6d0567a0d5c6cfe198d9e587ba6c447654c65b8edc", size = 983753 },
+    { url = "https://files.pythonhosted.org/packages/f7/01/1483856d84827c5fe541cb160f07914c6b063b8d961146e9c3557c4730c0/tiktoken-0.8.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6adc8323016d7758d6de7313527f755b0fc6c72985b7d9291be5d96d73ecd1e1", size = 1145913 },
+    { url = "https://files.pythonhosted.org/packages/c2/e1/6c7a772e0200131e960e3381f1d7b26406bc5612c70677989c1498af2a60/tiktoken-0.8.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b591fb2b30d6a72121a80be24ec7a0e9eb51c5500ddc7e4c2496516dd5e3816b", size = 1178505 },
+    { url = "https://files.pythonhosted.org/packages/3e/6b/3ae00f0bff5d0b6925bf6370cf0ff606f56daed76210c2b0a156017b78dc/tiktoken-0.8.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:845287b9798e476b4d762c3ebda5102be87ca26e5d2c9854002825d60cdb815d", size = 1239111 },
+    { url = "https://files.pythonhosted.org/packages/d5/3b/7c8812952ca55e1bab08afc1dda3c5991804c71b550b9402e82a082ab795/tiktoken-0.8.0-cp39-cp39-win_amd64.whl", hash = "sha256:1473cfe584252dc3fa62adceb5b1c763c1874e04511b197da4e6de51d6ce5a02", size = 884803 },
+]
+
+[[package]]
+name = "tokenizers"
+version = "0.21.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "huggingface-hub" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/20/41/c2be10975ca37f6ec40d7abd7e98a5213bb04f284b869c1a24e6504fd94d/tokenizers-0.21.0.tar.gz", hash = "sha256:ee0894bf311b75b0c03079f33859ae4b2334d675d4e93f5a4132e1eae2834fe4", size = 343021 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b0/5c/8b09607b37e996dc47e70d6a7b6f4bdd4e4d5ab22fe49d7374565c7fefaf/tokenizers-0.21.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:3c4c93eae637e7d2aaae3d376f06085164e1660f89304c0ab2b1d08a406636b2", size = 2647461 },
+    { url = "https://files.pythonhosted.org/packages/22/7a/88e58bb297c22633ed1c9d16029316e5b5ac5ee44012164c2edede599a5e/tokenizers-0.21.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:f53ea537c925422a2e0e92a24cce96f6bc5046bbef24a1652a5edc8ba975f62e", size = 2563639 },
+    { url = "https://files.pythonhosted.org/packages/f7/14/83429177c19364df27d22bc096d4c2e431e0ba43e56c525434f1f9b0fd00/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6b177fb54c4702ef611de0c069d9169f0004233890e0c4c5bd5508ae05abf193", size = 2903304 },
+    { url = "https://files.pythonhosted.org/packages/7e/db/3433eab42347e0dc5452d8fcc8da03f638c9accffefe5a7c78146666964a/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:6b43779a269f4629bebb114e19c3fca0223296ae9fea8bb9a7a6c6fb0657ff8e", size = 2804378 },
+    { url = "https://files.pythonhosted.org/packages/57/8b/7da5e6f89736c2ade02816b4733983fca1c226b0c42980b1ae9dc8fcf5cc/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9aeb255802be90acfd363626753fda0064a8df06031012fe7d52fd9a905eb00e", size = 3095488 },
+    { url = "https://files.pythonhosted.org/packages/4d/f6/5ed6711093dc2c04a4e03f6461798b12669bc5a17c8be7cce1240e0b5ce8/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d8b09dbeb7a8d73ee204a70f94fc06ea0f17dcf0844f16102b9f414f0b7463ba", size = 3121410 },
+    { url = "https://files.pythonhosted.org/packages/81/42/07600892d48950c5e80505b81411044a2d969368cdc0d929b1c847bf6697/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:400832c0904f77ce87c40f1a8a27493071282f785724ae62144324f171377273", size = 3388821 },
+    { url = "https://files.pythonhosted.org/packages/22/06/69d7ce374747edaf1695a4f61b83570d91cc8bbfc51ccfecf76f56ab4aac/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e84ca973b3a96894d1707e189c14a774b701596d579ffc7e69debfc036a61a04", size = 3008868 },
+    { url = "https://files.pythonhosted.org/packages/c8/69/54a0aee4d576045b49a0eb8bffdc495634309c823bf886042e6f46b80058/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:eb7202d231b273c34ec67767378cd04c767e967fda12d4a9e36208a34e2f137e", size = 8975831 },
+    { url = "https://files.pythonhosted.org/packages/f7/f3/b776061e4f3ebf2905ba1a25d90380aafd10c02d406437a8ba22d1724d76/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:089d56db6782a73a27fd8abf3ba21779f5b85d4a9f35e3b493c7bbcbbf0d539b", size = 8920746 },
+    { url = "https://files.pythonhosted.org/packages/d8/ee/ce83d5ec8b6844ad4c3ecfe3333d58ecc1adc61f0878b323a15355bcab24/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:c87ca3dc48b9b1222d984b6b7490355a6fdb411a2d810f6f05977258400ddb74", size = 9161814 },
+    { url = "https://files.pythonhosted.org/packages/18/07/3e88e65c0ed28fa93aa0c4d264988428eef3df2764c3126dc83e243cb36f/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:4145505a973116f91bc3ac45988a92e618a6f83eb458f49ea0790df94ee243ff", size = 9357138 },
+    { url = "https://files.pythonhosted.org/packages/15/b0/dc4572ca61555fc482ebc933f26cb407c6aceb3dc19c301c68184f8cad03/tokenizers-0.21.0-cp39-abi3-win32.whl", hash = "sha256:eb1702c2f27d25d9dd5b389cc1f2f51813e99f8ca30d9e25348db6585a97e24a", size = 2202266 },
+    { url = "https://files.pythonhosted.org/packages/44/69/d21eb253fa91622da25585d362a874fa4710be600f0ea9446d8d0217cec1/tokenizers-0.21.0-cp39-abi3-win_amd64.whl", hash = "sha256:87841da5a25a3a5f70c102de371db120f41873b854ba65e52bccd57df5a3780c", size = 2389192 },
+]
+
+[[package]]
+name = "tomli"
+version = "2.2.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/18/87/302344fed471e44a87289cf4967697d07e532f2421fdaf868a303cbae4ff/tomli-2.2.1.tar.gz", hash = "sha256:cd45e1dc79c835ce60f7404ec8119f2eb06d38b1deba146f07ced3bbc44505ff", size = 17175 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/ca/75707e6efa2b37c77dadb324ae7d9571cb424e61ea73fad7c56c2d14527f/tomli-2.2.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:678e4fa69e4575eb77d103de3df8a895e1591b48e740211bd1067378c69e8249", size = 131077 },
+    { url = "https://files.pythonhosted.org/packages/c7/16/51ae563a8615d472fdbffc43a3f3d46588c264ac4f024f63f01283becfbb/tomli-2.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:023aa114dd824ade0100497eb2318602af309e5a55595f76b626d6d9f3b7b0a6", size = 123429 },
+    { url = "https://files.pythonhosted.org/packages/f1/dd/4f6cd1e7b160041db83c694abc78e100473c15d54620083dbd5aae7b990e/tomli-2.2.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ece47d672db52ac607a3d9599a9d48dcb2f2f735c6c2d1f34130085bb12b112a", size = 226067 },
+    { url = "https://files.pythonhosted.org/packages/a9/6b/c54ede5dc70d648cc6361eaf429304b02f2871a345bbdd51e993d6cdf550/tomli-2.2.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6972ca9c9cc9f0acaa56a8ca1ff51e7af152a9f87fb64623e31d5c83700080ee", size = 236030 },
+    { url = "https://files.pythonhosted.org/packages/1f/47/999514fa49cfaf7a92c805a86c3c43f4215621855d151b61c602abb38091/tomli-2.2.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c954d2250168d28797dd4e3ac5cf812a406cd5a92674ee4c8f123c889786aa8e", size = 240898 },
+    { url = "https://files.pythonhosted.org/packages/73/41/0a01279a7ae09ee1573b423318e7934674ce06eb33f50936655071d81a24/tomli-2.2.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8dd28b3e155b80f4d54beb40a441d366adcfe740969820caf156c019fb5c7ec4", size = 229894 },
+    { url = "https://files.pythonhosted.org/packages/55/18/5d8bc5b0a0362311ce4d18830a5d28943667599a60d20118074ea1b01bb7/tomli-2.2.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:e59e304978767a54663af13c07b3d1af22ddee3bb2fb0618ca1593e4f593a106", size = 245319 },
+    { url = "https://files.pythonhosted.org/packages/92/a3/7ade0576d17f3cdf5ff44d61390d4b3febb8a9fc2b480c75c47ea048c646/tomli-2.2.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:33580bccab0338d00994d7f16f4c4ec25b776af3ffaac1ed74e0b3fc95e885a8", size = 238273 },
+    { url = "https://files.pythonhosted.org/packages/72/6f/fa64ef058ac1446a1e51110c375339b3ec6be245af9d14c87c4a6412dd32/tomli-2.2.1-cp311-cp311-win32.whl", hash = "sha256:465af0e0875402f1d226519c9904f37254b3045fc5084697cefb9bdde1ff99ff", size = 98310 },
+    { url = "https://files.pythonhosted.org/packages/6a/1c/4a2dcde4a51b81be3530565e92eda625d94dafb46dbeb15069df4caffc34/tomli-2.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:2d0f2fdd22b02c6d81637a3c95f8cd77f995846af7414c5c4b8d0545afa1bc4b", size = 108309 },
+    { url = "https://files.pythonhosted.org/packages/52/e1/f8af4c2fcde17500422858155aeb0d7e93477a0d59a98e56cbfe75070fd0/tomli-2.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:4a8f6e44de52d5e6c657c9fe83b562f5f4256d8ebbfe4ff922c495620a7f6cea", size = 132762 },
+    { url = "https://files.pythonhosted.org/packages/03/b8/152c68bb84fc00396b83e7bbddd5ec0bd3dd409db4195e2a9b3e398ad2e3/tomli-2.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8d57ca8095a641b8237d5b079147646153d22552f1c637fd3ba7f4b0b29167a8", size = 123453 },
+    { url = "https://files.pythonhosted.org/packages/c8/d6/fc9267af9166f79ac528ff7e8c55c8181ded34eb4b0e93daa767b8841573/tomli-2.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e340144ad7ae1533cb897d406382b4b6fede8890a03738ff1683af800d54192", size = 233486 },
+    { url = "https://files.pythonhosted.org/packages/5c/51/51c3f2884d7bab89af25f678447ea7d297b53b5a3b5730a7cb2ef6069f07/tomli-2.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:db2b95f9de79181805df90bedc5a5ab4c165e6ec3fe99f970d0e302f384ad222", size = 242349 },
+    { url = "https://files.pythonhosted.org/packages/ab/df/bfa89627d13a5cc22402e441e8a931ef2108403db390ff3345c05253935e/tomli-2.2.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:40741994320b232529c802f8bc86da4e1aa9f413db394617b9a256ae0f9a7f77", size = 252159 },
+    { url = "https://files.pythonhosted.org/packages/9e/6e/fa2b916dced65763a5168c6ccb91066f7639bdc88b48adda990db10c8c0b/tomli-2.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:400e720fe168c0f8521520190686ef8ef033fb19fc493da09779e592861b78c6", size = 237243 },
+    { url = "https://files.pythonhosted.org/packages/b4/04/885d3b1f650e1153cbb93a6a9782c58a972b94ea4483ae4ac5cedd5e4a09/tomli-2.2.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:02abe224de6ae62c19f090f68da4e27b10af2b93213d36cf44e6e1c5abd19fdd", size = 259645 },
+    { url = "https://files.pythonhosted.org/packages/9c/de/6b432d66e986e501586da298e28ebeefd3edc2c780f3ad73d22566034239/tomli-2.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b82ebccc8c8a36f2094e969560a1b836758481f3dc360ce9a3277c65f374285e", size = 244584 },
+    { url = "https://files.pythonhosted.org/packages/1c/9a/47c0449b98e6e7d1be6cbac02f93dd79003234ddc4aaab6ba07a9a7482e2/tomli-2.2.1-cp312-cp312-win32.whl", hash = "sha256:889f80ef92701b9dbb224e49ec87c645ce5df3fa2cc548664eb8a25e03127a98", size = 98875 },
+    { url = "https://files.pythonhosted.org/packages/ef/60/9b9638f081c6f1261e2688bd487625cd1e660d0a85bd469e91d8db969734/tomli-2.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:7fc04e92e1d624a4a63c76474610238576942d6b8950a2d7f908a340494e67e4", size = 109418 },
+    { url = "https://files.pythonhosted.org/packages/04/90/2ee5f2e0362cb8a0b6499dc44f4d7d48f8fff06d28ba46e6f1eaa61a1388/tomli-2.2.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:f4039b9cbc3048b2416cc57ab3bda989a6fcf9b36cf8937f01a6e731b64f80d7", size = 132708 },
+    { url = "https://files.pythonhosted.org/packages/c0/ec/46b4108816de6b385141f082ba99e315501ccd0a2ea23db4a100dd3990ea/tomli-2.2.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:286f0ca2ffeeb5b9bd4fcc8d6c330534323ec51b2f52da063b11c502da16f30c", size = 123582 },
+    { url = "https://files.pythonhosted.org/packages/a0/bd/b470466d0137b37b68d24556c38a0cc819e8febe392d5b199dcd7f578365/tomli-2.2.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a92ef1a44547e894e2a17d24e7557a5e85a9e1d0048b0b5e7541f76c5032cb13", size = 232543 },
+    { url = "https://files.pythonhosted.org/packages/d9/e5/82e80ff3b751373f7cead2815bcbe2d51c895b3c990686741a8e56ec42ab/tomli-2.2.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9316dc65bed1684c9a98ee68759ceaed29d229e985297003e494aa825ebb0281", size = 241691 },
+    { url = "https://files.pythonhosted.org/packages/05/7e/2a110bc2713557d6a1bfb06af23dd01e7dde52b6ee7dadc589868f9abfac/tomli-2.2.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e85e99945e688e32d5a35c1ff38ed0b3f41f43fad8df0bdf79f72b2ba7bc5272", size = 251170 },
+    { url = "https://files.pythonhosted.org/packages/64/7b/22d713946efe00e0adbcdfd6d1aa119ae03fd0b60ebed51ebb3fa9f5a2e5/tomli-2.2.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ac065718db92ca818f8d6141b5f66369833d4a80a9d74435a268c52bdfa73140", size = 236530 },
+    { url = "https://files.pythonhosted.org/packages/38/31/3a76f67da4b0cf37b742ca76beaf819dca0ebef26d78fc794a576e08accf/tomli-2.2.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:d920f33822747519673ee656a4b6ac33e382eca9d331c87770faa3eef562aeb2", size = 258666 },
+    { url = "https://files.pythonhosted.org/packages/07/10/5af1293da642aded87e8a988753945d0cf7e00a9452d3911dd3bb354c9e2/tomli-2.2.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a198f10c4d1b1375d7687bc25294306e551bf1abfa4eace6650070a5c1ae2744", size = 243954 },
+    { url = "https://files.pythonhosted.org/packages/5b/b9/1ed31d167be802da0fc95020d04cd27b7d7065cc6fbefdd2f9186f60d7bd/tomli-2.2.1-cp313-cp313-win32.whl", hash = "sha256:d3f5614314d758649ab2ab3a62d4f2004c825922f9e370b29416484086b264ec", size = 98724 },
+    { url = "https://files.pythonhosted.org/packages/c7/32/b0963458706accd9afcfeb867c0f9175a741bf7b19cd424230714d722198/tomli-2.2.1-cp313-cp313-win_amd64.whl", hash = "sha256:a38aa0308e754b0e3c67e344754dff64999ff9b513e691d0e786265c93583c69", size = 109383 },
+    { url = "https://files.pythonhosted.org/packages/6e/c2/61d3e0f47e2b74ef40a68b9e6ad5984f6241a942f7cd3bbfbdbd03861ea9/tomli-2.2.1-py3-none-any.whl", hash = "sha256:cb55c73c5f4408779d0cf3eef9f762b9c9f147a77de7b258bef0a5628adc85cc", size = 14257 },
+]
+
+[[package]]
+name = "torch"
+version = "2.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "filelock" },
+    { name = "fsspec" },
+    { name = "jinja2" },
+    { name = "mkl", marker = "sys_platform == 'win32'" },
+    { name = "networkx", version = "3.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "networkx", version = "3.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "nvidia-cublas-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-cupti-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-nvrtc-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-runtime-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cudnn-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufft-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-curand-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusolver-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparse-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nccl-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvtx-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "sympy" },
+    { name = "triton", marker = "python_full_version < '3.12' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "typing-extensions" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/e5/2ddae60ae999b224aceb74490abeb885ee118227f866cb12046f0481d4c9/torch-2.3.0-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:d8ea5a465dbfd8501f33c937d1f693176c9aef9d1c1b0ca1d44ed7b0a18c52ac", size = 779141888 },
+    { url = "https://files.pythonhosted.org/packages/dc/82/7ab793eba3f4c3c84244e4c578ea838536d526db01c787448e8177404598/torch-2.3.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:09c81c5859a5b819956c6925a405ef1cdda393c9d8a01ce3851453f699d3358c", size = 88506981 },
+    { url = "https://files.pythonhosted.org/packages/2e/f7/503bab04f4e7a0a43f2ff05c3635cee9dfcf2e09656020d29502b87a94a3/torch-2.3.0-cp310-cp310-win_amd64.whl", hash = "sha256:1bf023aa20902586f614f7682fedfa463e773e26c58820b74158a72470259459", size = 159799740 },
+    { url = "https://files.pythonhosted.org/packages/01/c1/c6b42224122989ec95a820974aee92bdd4308380a7bb6ffa9a9d2429765d/torch-2.3.0-cp310-none-macosx_11_0_arm64.whl", hash = "sha256:758ef938de87a2653bba74b91f703458c15569f1562bf4b6c63c62d9c5a0c1f5", size = 60993271 },
+    { url = "https://files.pythonhosted.org/packages/35/3a/a39f354fa3119785be87e2f94ffa2620f8a270c8560f7356358ee62fb4c5/torch-2.3.0-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:493d54ee2f9df100b5ce1d18c96dbb8d14908721f76351e908c9d2622773a788", size = 779160265 },
+    { url = "https://files.pythonhosted.org/packages/91/3c/98a9b901b40278b40a9ff5b9c6be387b20997269f929f2182d8845c94085/torch-2.3.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:bce43af735c3da16cc14c7de2be7ad038e2fbf75654c2e274e575c6c05772ace", size = 88536251 },
+    { url = "https://files.pythonhosted.org/packages/2a/b7/a3cf5fd40334b9785cc83ee0c96b50603026eb3aa70210a33729018e7029/torch-2.3.0-cp311-cp311-win_amd64.whl", hash = "sha256:729804e97b7cf19ae9ab4181f91f5e612af07956f35c8b2c8e9d9f3596a8e877", size = 159803952 },
+    { url = "https://files.pythonhosted.org/packages/ad/08/c5e41eb22323db4a52260607598a207a2e1918916ae8201aa7a8ae005fcd/torch-2.3.0-cp311-none-macosx_11_0_arm64.whl", hash = "sha256:d24e328226d8e2af7cf80fcb1d2f1d108e0de32777fab4aaa2b37b9765d8be73", size = 60998957 },
+    { url = "https://files.pythonhosted.org/packages/51/03/1abad10990c76bee3703857b1617563b241f87d297ee466dbad922b0c308/torch-2.3.0-cp312-cp312-manylinux1_x86_64.whl", hash = "sha256:b0de2bdc0486ea7b14fc47ff805172df44e421a7318b7c4d92ef589a75d27410", size = 779062531 },
+    { url = "https://files.pythonhosted.org/packages/f1/9d/dfe273e19b7165148208bd182fac87488c5a0f7a3318d1646f5f37af872f/torch-2.3.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:a306c87a3eead1ed47457822c01dfbd459fe2920f2d38cbdf90de18f23f72542", size = 88437165 },
+    { url = "https://files.pythonhosted.org/packages/37/04/a5cd83baccbf2d4329990ec06b8abf3a644e1559a7b1f764f42d2cb77d51/torch-2.3.0-cp312-cp312-win_amd64.whl", hash = "sha256:f9b98bf1a3c8af2d4c41f0bf1433920900896c446d1ddc128290ff146d1eb4bd", size = 159749140 },
+    { url = "https://files.pythonhosted.org/packages/55/51/4bdee83e6fa9cca8e3a6cdf81a2695ede9d3fd7148e4fd4188dff142d7b0/torch-2.3.0-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:dca986214267b34065a79000cee54232e62b41dff1ec2cab9abc3fc8b3dee0ad", size = 60968873 },
+    { url = "https://files.pythonhosted.org/packages/da/af/abfe44a9ac7fd9705f145483e9e6c73a23d0533223082468eb2d978a494c/torch-2.3.0-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:cd0dc498b961ab19cb3f8dbf0c6c50e244f2f37dbfa05754ab44ea057c944ef9", size = 779140324 },
+    { url = "https://files.pythonhosted.org/packages/e6/b3/0ba7ab81889a964b027f9f46f7e1ddc49ac5532d5f7f09fc0730a9295bde/torch-2.3.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:e05f836559251e4096f3786ee99f4a8cbe67bc7fbedba8ad5e799681e47c5e80", size = 88508656 },
+    { url = "https://files.pythonhosted.org/packages/af/73/9fbe55b0db8e329b124df415bbdac760f4592e030d2cf20b05fd9aec68f5/torch-2.3.0-cp39-cp39-win_amd64.whl", hash = "sha256:4fb27b35dbb32303c2927da86e27b54a92209ddfb7234afb1949ea2b3effffea", size = 159734927 },
+    { url = "https://files.pythonhosted.org/packages/5b/e0/12a8ef2d7a00060ca5cbacef7b12f2b3e35fe3472973882e4a87165fea67/torch-2.3.0-cp39-none-macosx_11_0_arm64.whl", hash = "sha256:760f8bedff506ce9e6e103498f9b1e9e15809e008368594c3a66bf74a8a51380", size = 60993508 },
+]
+
+[[package]]
+name = "torchvision"
+version = "0.18.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy" },
+    { name = "pillow" },
+    { name = "torch" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c5/e7/19e06d609f44f2c670db1056817deb928412ed329ff3f63cc2a27231b029/torchvision-0.18.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:dd61628a3d189c6852a12dc5ed4cd2eece66d2d67f35a866cb16f1dcb06c8c62", size = 1554228 },
+    { url = "https://files.pythonhosted.org/packages/d4/7e/d41b771dbffa927b9cc37372b1e18c881348cd18a0e4ad73f2c6bdf56c0e/torchvision-0.18.0-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:493c45f9937dad37aa1b64b14da17c7a589c72b91adc4837d431009cfe29bd53", size = 6954946 },
+    { url = "https://files.pythonhosted.org/packages/b2/dd/e5d39496413a5e5c2ca69d333bc241e7c8e8e412778c8309d54ce27cb9ec/torchvision-0.18.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:5337f6acfa1fe959d5cb340d01a00614d6b31ce7a4824ccb95435a85c5273b95", size = 13995854 },
+    { url = "https://files.pythonhosted.org/packages/f3/ef/f9559f3fc09362eafea5937521a2013a3ae67e38a7c6c3c9b51b3751c320/torchvision-0.18.0-cp310-cp310-win_amd64.whl", hash = "sha256:bd8e6f3b5beb49965f15c461302488edfa3d8c2d01d3bb79b150d6fb62711e3a", size = 1174147 },
+    { url = "https://files.pythonhosted.org/packages/b5/14/c05da13c98f528ba5fd99897320a7684df5dd136ec6faa6a5766f25e4a7e/torchvision-0.18.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6896a52168befe1105fb3c9335287390ed227e71d1e4ec4d68b62e8a3099fc09", size = 1554225 },
+    { url = "https://files.pythonhosted.org/packages/6e/7d/bc67ec2d1011226e75cdcc45a2c85d97b8eaac32a7d648b71c432d584367/torchvision-0.18.0-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:3d7955398d4ceaad77c487c2c44f6f7813112402c9bab8cd906d346005891048", size = 6955003 },
+    { url = "https://files.pythonhosted.org/packages/70/1d/107894816bf5ebecbc5a8556743fd89c0a1dd6da82b1c562d6becd5a108a/torchvision-0.18.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:e5a24d620cea14a4bb89f24aa2b506230c0a16a3ada57fc53ad80cfd256a2128", size = 13995855 },
+    { url = "https://files.pythonhosted.org/packages/12/c2/7c89c62f2b0a606070aa7cdb8af8af0c905562942778ebdd77600642c3b9/torchvision-0.18.0-cp311-cp311-win_amd64.whl", hash = "sha256:6ad70ddfa879bda5ed886b2518fe562640e0059787cbd65cb2bffa7674541410", size = 1174125 },
+    { url = "https://files.pythonhosted.org/packages/7c/12/49d4fd4ae7a48c6d33babb01a523594aca38365d378518320a10f9a5baa1/torchvision-0.18.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:eb9d83c0e1dbb54ecb0fb04c87f786333e3a6fb8b9c400aca7c31081f9aa5707", size = 1554232 },
+    { url = "https://files.pythonhosted.org/packages/6d/fe/729256fec03403b0bfdc229d4350936e29020d618f1740009c6a4f995b06/torchvision-0.18.0-cp312-cp312-manylinux1_x86_64.whl", hash = "sha256:b657d052d146f24cb3b2a78219bfc82ae70a9706671c50f632528907d10cccec", size = 6955047 },
+    { url = "https://files.pythonhosted.org/packages/51/2d/30883e9c6734546f9e7e0c429b76bd2b651aa25f6ced087c9c11cbf2ef41/torchvision-0.18.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:a964afbc7ddf50a46b941477f6c35729b416deedd139756befd488245e2e226d", size = 13996688 },
+    { url = "https://files.pythonhosted.org/packages/53/8a/864c3969af219a95213a5065d453313a96598e7c744b859e99b6ac134e16/torchvision-0.18.0-cp312-cp312-win_amd64.whl", hash = "sha256:7c770f0f748e0b17f57c0297508d7254f686cdf03fc2e2949f422b20574f4c0f", size = 1174123 },
+    { url = "https://files.pythonhosted.org/packages/d1/dc/bace7633ce37bc165fb6509e741d3cd73535f36e86b3ee5a95b2b647af92/torchvision-0.18.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:75e22ecf44a13b8f95b8ad421c0261282d859c61816badaca1959e073ccdd691", size = 1554270 },
+    { url = "https://files.pythonhosted.org/packages/54/20/309df7711dd17c399b7cd233e78bcee0f6fcdfbef5f2ec0930910c1884df/torchvision-0.18.0-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:4c334b3e719ba0a9ba6e15d4aff1178f5e6d029174f346163fed525f0ccfffd3", size = 6956173 },
+    { url = "https://files.pythonhosted.org/packages/72/73/b50f27719a9d0d9decae9f4264967a5c54a84046d1986772316d2471b6a2/torchvision-0.18.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:36efd87001c6bee2383e043e46a025affb03179747c8f4777b9918527ffce756", size = 1626203 },
+    { url = "https://files.pythonhosted.org/packages/db/4a/33a91dfcce5d62f826b15691a3bda2cdf62e4fbaf164f469af6a6756ebb4/torchvision-0.18.0-cp39-cp39-win_amd64.whl", hash = "sha256:ccc292e093771d5baacf5535ac4416306b6b5f15676341cd4d010d8542eace25", size = 1174149 },
+]
+
+[[package]]
+name = "tqdm"
+version = "4.67.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a8/4b/29b4ef32e036bb34e4ab51796dd745cdba7ed47ad142a9f4a1eb8e0c744d/tqdm-4.67.1.tar.gz", hash = "sha256:f8aef9c52c08c13a65f30ea34f4e5aac3fd1a34959879d7e59e63027286627f2", size = 169737 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d0/30/dc54f88dd4a2b5dc8a0279bdd7270e735851848b762aeb1c1184ed1f6b14/tqdm-4.67.1-py3-none-any.whl", hash = "sha256:26445eca388f82e72884e0d580d5464cd801a3ea01e63e5601bdff9ba6a48de2", size = 78540 },
+]
+
+[[package]]
+name = "transformers"
+version = "4.48.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "filelock" },
+    { name = "huggingface-hub" },
+    { name = "numpy" },
+    { name = "packaging" },
+    { name = "pyyaml" },
+    { name = "regex" },
+    { name = "requests" },
+    { name = "safetensors" },
+    { name = "tokenizers" },
+    { name = "tqdm" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ea/71/93a6331682d6f15adf7d646956db0c43e5f1759bbbd05f2ef53029bae107/transformers-4.48.0.tar.gz", hash = "sha256:03fdfcbfb8b0367fb6c9fbe9d1c9aa54dfd847618be9b52400b2811d22799cb1", size = 8372101 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/45/d6/a69764e89fc5c2c957aa473881527c8c35521108d553df703e9ba703daeb/transformers-4.48.0-py3-none-any.whl", hash = "sha256:6d3de6d71cb5f2a10f9775ccc17abce9620195caaf32ec96542bd2a6937f25b0", size = 9673380 },
+]
+
+[package.optional-dependencies]
+sentencepiece = [
+    { name = "protobuf" },
+    { name = "sentencepiece" },
+]
+
+[[package]]
+name = "triton"
+version = "2.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "filelock", marker = "python_full_version < '3.12'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/db/ee/8d50d44ed5b63677bb387f4ee67a7dbaaded0189b320ffe82685a6827728/triton-2.3.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5ce4b8ff70c48e47274c66f269cce8861cf1dc347ceeb7a67414ca151b1822d8", size = 168055246 },
+    { url = "https://files.pythonhosted.org/packages/3c/00/84e0006f2025260fa111ddfc66194bd1af731b3ee18e2fd611a00f290b5e/triton-2.3.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3c3d9607f85103afdb279938fc1dd2a66e4f5999a58eb48a346bd42738f986dd", size = 168079300 },
+    { url = "https://files.pythonhosted.org/packages/90/2f/3e8f0ea4ef7ac6d8720a48b9b9700fd32c9909ee83b2eb1f25209ace0767/triton-2.3.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:218d742e67480d9581bafb73ed598416cc8a56f6316152e5562ee65e33de01c0", size = 168091361 },
+    { url = "https://files.pythonhosted.org/packages/8e/7b/8557e97f0f39992eaf390aacdf36c38a4de059caa9eda99321f7d3576d82/triton-2.3.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6d8f636e0341ac348899a47a057c3daea99ea7db31528a225a3ba4ded28ccc65", size = 168051036 },
+]
+
+[[package]]
+name = "typing-extensions"
+version = "4.12.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/df/db/f35a00659bc03fec321ba8bce9420de607a1d37f8342eee1863174c69557/typing_extensions-4.12.2.tar.gz", hash = "sha256:1a7ead55c7e559dd4dee8856e3a88b41225abfe1ce8df57b7c13915fe121ffb8", size = 85321 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/26/9f/ad63fc0248c5379346306f8668cda6e2e2e9c95e01216d2b8ffd9ff037d0/typing_extensions-4.12.2-py3-none-any.whl", hash = "sha256:04e5ca0351e0f3f85c6853954072df659d0d13fac324d0072316b67d7794700d", size = 37438 },
+]
+
+[[package]]
+name = "tzdata"
+version = "2024.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e1/34/943888654477a574a86a98e9896bae89c7aa15078ec29f490fef2f1e5384/tzdata-2024.2.tar.gz", hash = "sha256:7d85cc416e9382e69095b7bdf4afd9e3880418a2413feec7069d533d6b4e31cc", size = 193282 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a6/ab/7e5f53c3b9d14972843a647d8d7a853969a58aecc7559cb3267302c94774/tzdata-2024.2-py2.py3-none-any.whl", hash = "sha256:a48093786cdcde33cad18c2555e8532f34422074448fbc874186f0abd79565cd", size = 346586 },
+]
+
+[[package]]
+name = "urllib3"
+version = "2.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/aa/63/e53da845320b757bf29ef6a9062f5c669fe997973f966045cb019c3f4b66/urllib3-2.3.0.tar.gz", hash = "sha256:f8c5449b3cf0861679ce7e0503c7b44b5ec981bec0d1d3795a07f1ba96f0204d", size = 307268 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c8/19/4ec628951a74043532ca2cf5d97b7b14863931476d117c471e8e2b1eb39f/urllib3-2.3.0-py3-none-any.whl", hash = "sha256:1cee9ad369867bfdbbb48b7dd50374c0967a0bb7710050facf0dd6911440e3df", size = 128369 },
+]
+
+[[package]]
+name = "uvicorn"
+version = "0.34.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "click" },
+    { name = "h11" },
+    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4b/4d/938bd85e5bf2edeec766267a5015ad969730bb91e31b44021dfe8b22df6c/uvicorn-0.34.0.tar.gz", hash = "sha256:404051050cd7e905de2c9a7e61790943440b3416f49cb409f965d9dcd0fa73e9", size = 76568 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/61/14/33a3a1352cfa71812a3a21e8c9bfb83f60b0011f5e36f2b1399d51928209/uvicorn-0.34.0-py3-none-any.whl", hash = "sha256:023dc038422502fa28a09c7a30bf2b6991512da7dcdb8fd35fe57cfc154126f4", size = 62315 },
+]
+
+[package.optional-dependencies]
+standard = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "httptools" },
+    { name = "python-dotenv" },
+    { name = "pyyaml" },
+    { name = "uvloop", marker = "platform_python_implementation != 'PyPy' and sys_platform != 'cygwin' and sys_platform != 'win32'" },
+    { name = "watchfiles" },
+    { name = "websockets" },
+]
+
+[[package]]
+name = "uvloop"
+version = "0.21.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/af/c0/854216d09d33c543f12a44b393c402e89a920b1a0a7dc634c42de91b9cf6/uvloop-0.21.0.tar.gz", hash = "sha256:3bf12b0fda68447806a7ad847bfa591613177275d35b6724b1ee573faa3704e3", size = 2492741 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3d/76/44a55515e8c9505aa1420aebacf4dd82552e5e15691654894e90d0bd051a/uvloop-0.21.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:ec7e6b09a6fdded42403182ab6b832b71f4edaf7f37a9a0e371a01db5f0cb45f", size = 1442019 },
+    { url = "https://files.pythonhosted.org/packages/35/5a/62d5800358a78cc25c8a6c72ef8b10851bdb8cca22e14d9c74167b7f86da/uvloop-0.21.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:196274f2adb9689a289ad7d65700d37df0c0930fd8e4e743fa4834e850d7719d", size = 801898 },
+    { url = "https://files.pythonhosted.org/packages/f3/96/63695e0ebd7da6c741ccd4489b5947394435e198a1382349c17b1146bb97/uvloop-0.21.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f38b2e090258d051d68a5b14d1da7203a3c3677321cf32a95a6f4db4dd8b6f26", size = 3827735 },
+    { url = "https://files.pythonhosted.org/packages/61/e0/f0f8ec84979068ffae132c58c79af1de9cceeb664076beea86d941af1a30/uvloop-0.21.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:87c43e0f13022b998eb9b973b5e97200c8b90823454d4bc06ab33829e09fb9bb", size = 3825126 },
+    { url = "https://files.pythonhosted.org/packages/bf/fe/5e94a977d058a54a19df95f12f7161ab6e323ad49f4dabc28822eb2df7ea/uvloop-0.21.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:10d66943def5fcb6e7b37310eb6b5639fd2ccbc38df1177262b0640c3ca68c1f", size = 3705789 },
+    { url = "https://files.pythonhosted.org/packages/26/dd/c7179618e46092a77e036650c1f056041a028a35c4d76945089fcfc38af8/uvloop-0.21.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:67dd654b8ca23aed0a8e99010b4c34aca62f4b7fce88f39d452ed7622c94845c", size = 3800523 },
+    { url = "https://files.pythonhosted.org/packages/57/a7/4cf0334105c1160dd6819f3297f8700fda7fc30ab4f61fbf3e725acbc7cc/uvloop-0.21.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:c0f3fa6200b3108919f8bdabb9a7f87f20e7097ea3c543754cabc7d717d95cf8", size = 1447410 },
+    { url = "https://files.pythonhosted.org/packages/8c/7c/1517b0bbc2dbe784b563d6ab54f2ef88c890fdad77232c98ed490aa07132/uvloop-0.21.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:0878c2640cf341b269b7e128b1a5fed890adc4455513ca710d77d5e93aa6d6a0", size = 805476 },
+    { url = "https://files.pythonhosted.org/packages/ee/ea/0bfae1aceb82a503f358d8d2fa126ca9dbdb2ba9c7866974faec1cb5875c/uvloop-0.21.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b9fb766bb57b7388745d8bcc53a359b116b8a04c83a2288069809d2b3466c37e", size = 3960855 },
+    { url = "https://files.pythonhosted.org/packages/8a/ca/0864176a649838b838f36d44bf31c451597ab363b60dc9e09c9630619d41/uvloop-0.21.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a375441696e2eda1c43c44ccb66e04d61ceeffcd76e4929e527b7fa401b90fb", size = 3973185 },
+    { url = "https://files.pythonhosted.org/packages/30/bf/08ad29979a936d63787ba47a540de2132169f140d54aa25bc8c3df3e67f4/uvloop-0.21.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:baa0e6291d91649c6ba4ed4b2f982f9fa165b5bbd50a9e203c416a2797bab3c6", size = 3820256 },
+    { url = "https://files.pythonhosted.org/packages/da/e2/5cf6ef37e3daf2f06e651aae5ea108ad30df3cb269102678b61ebf1fdf42/uvloop-0.21.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4509360fcc4c3bd2c70d87573ad472de40c13387f5fda8cb58350a1d7475e58d", size = 3937323 },
+    { url = "https://files.pythonhosted.org/packages/8c/4c/03f93178830dc7ce8b4cdee1d36770d2f5ebb6f3d37d354e061eefc73545/uvloop-0.21.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:359ec2c888397b9e592a889c4d72ba3d6befba8b2bb01743f72fffbde663b59c", size = 1471284 },
+    { url = "https://files.pythonhosted.org/packages/43/3e/92c03f4d05e50f09251bd8b2b2b584a2a7f8fe600008bcc4523337abe676/uvloop-0.21.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f7089d2dc73179ce5ac255bdf37c236a9f914b264825fdaacaded6990a7fb4c2", size = 821349 },
+    { url = "https://files.pythonhosted.org/packages/a6/ef/a02ec5da49909dbbfb1fd205a9a1ac4e88ea92dcae885e7c961847cd51e2/uvloop-0.21.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:baa4dcdbd9ae0a372f2167a207cd98c9f9a1ea1188a8a526431eef2f8116cc8d", size = 4580089 },
+    { url = "https://files.pythonhosted.org/packages/06/a7/b4e6a19925c900be9f98bec0a75e6e8f79bb53bdeb891916609ab3958967/uvloop-0.21.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86975dca1c773a2c9864f4c52c5a55631038e387b47eaf56210f873887b6c8dc", size = 4693770 },
+    { url = "https://files.pythonhosted.org/packages/ce/0c/f07435a18a4b94ce6bd0677d8319cd3de61f3a9eeb1e5f8ab4e8b5edfcb3/uvloop-0.21.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:461d9ae6660fbbafedd07559c6a2e57cd553b34b0065b6550685f6653a98c1cb", size = 4451321 },
+    { url = "https://files.pythonhosted.org/packages/8f/eb/f7032be105877bcf924709c97b1bf3b90255b4ec251f9340cef912559f28/uvloop-0.21.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:183aef7c8730e54c9a3ee3227464daed66e37ba13040bb3f350bc2ddc040f22f", size = 4659022 },
+    { url = "https://files.pythonhosted.org/packages/3f/8d/2cbef610ca21539f0f36e2b34da49302029e7c9f09acef0b1c3b5839412b/uvloop-0.21.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:bfd55dfcc2a512316e65f16e503e9e450cab148ef11df4e4e679b5e8253a5281", size = 1468123 },
+    { url = "https://files.pythonhosted.org/packages/93/0d/b0038d5a469f94ed8f2b2fce2434a18396d8fbfb5da85a0a9781ebbdec14/uvloop-0.21.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:787ae31ad8a2856fc4e7c095341cccc7209bd657d0e71ad0dc2ea83c4a6fa8af", size = 819325 },
+    { url = "https://files.pythonhosted.org/packages/50/94/0a687f39e78c4c1e02e3272c6b2ccdb4e0085fda3b8352fecd0410ccf915/uvloop-0.21.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5ee4d4ef48036ff6e5cfffb09dd192c7a5027153948d85b8da7ff705065bacc6", size = 4582806 },
+    { url = "https://files.pythonhosted.org/packages/d2/19/f5b78616566ea68edd42aacaf645adbf71fbd83fc52281fba555dc27e3f1/uvloop-0.21.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3df876acd7ec037a3d005b3ab85a7e4110422e4d9c1571d4fc89b0fc41b6816", size = 4701068 },
+    { url = "https://files.pythonhosted.org/packages/47/57/66f061ee118f413cd22a656de622925097170b9380b30091b78ea0c6ea75/uvloop-0.21.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:bd53ecc9a0f3d87ab847503c2e1552b690362e005ab54e8a48ba97da3924c0dc", size = 4454428 },
+    { url = "https://files.pythonhosted.org/packages/63/9a/0962b05b308494e3202d3f794a6e85abe471fe3cafdbcf95c2e8c713aabd/uvloop-0.21.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a5c39f217ab3c663dc699c04cbd50c13813e31d917642d459fdcec07555cc553", size = 4660018 },
+    { url = "https://files.pythonhosted.org/packages/3c/a4/646a9d0edff7cde25fc1734695d3dfcee0501140dd0e723e4df3f0a50acb/uvloop-0.21.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:c097078b8031190c934ed0ebfee8cc5f9ba9642e6eb88322b9958b649750f72b", size = 1439646 },
+    { url = "https://files.pythonhosted.org/packages/01/2e/e128c66106af9728f86ebfeeb52af27ecd3cb09336f3e2f3e06053707a15/uvloop-0.21.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:46923b0b5ee7fc0020bef24afe7836cb068f5050ca04caf6b487c513dc1a20b2", size = 800931 },
+    { url = "https://files.pythonhosted.org/packages/2d/1a/9fbc2b1543d0df11f7aed1632f64bdf5ecc4053cf98cdc9edb91a65494f9/uvloop-0.21.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:53e420a3afe22cdcf2a0f4846e377d16e718bc70103d7088a4f7623567ba5fb0", size = 3829660 },
+    { url = "https://files.pythonhosted.org/packages/b8/c0/392e235e4100ae3b95b5c6dac77f82b529d2760942b1e7e0981e5d8e895d/uvloop-0.21.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88cb67cdbc0e483da00af0b2c3cdad4b7c61ceb1ee0f33fe00e09c81e3a6cb75", size = 3827185 },
+    { url = "https://files.pythonhosted.org/packages/e1/24/a5da6aba58f99aed5255eca87d58d1760853e8302d390820cc29058408e3/uvloop-0.21.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:221f4f2a1f46032b403bf3be628011caf75428ee3cc204a22addf96f586b19fd", size = 3705833 },
+    { url = "https://files.pythonhosted.org/packages/1a/5c/6ba221bb60f1e6474474102e17e38612ec7a06dc320e22b687ab563d877f/uvloop-0.21.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:2d1f581393673ce119355d56da84fe1dd9d2bb8b3d13ce792524e1607139feff", size = 3804696 },
+]
+
+[[package]]
+name = "virtualenv"
+version = "20.29.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "distlib" },
+    { name = "filelock" },
+    { name = "platformdirs" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a7/ca/f23dcb02e161a9bba141b1c08aa50e8da6ea25e6d780528f1d385a3efe25/virtualenv-20.29.1.tar.gz", hash = "sha256:b8b8970138d32fb606192cb97f6cd4bb644fa486be9308fb9b63f81091b5dc35", size = 7658028 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/89/9b/599bcfc7064fbe5740919e78c5df18e5dceb0887e676256a1061bb5ae232/virtualenv-20.29.1-py3-none-any.whl", hash = "sha256:4e4cb403c0b0da39e13b46b1b2476e505cb0046b25f242bee80f62bf990b2779", size = 4282379 },
+]
+
+[[package]]
+name = "vllm"
+version = "0.5.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "aiohttp" },
+    { name = "cmake" },
+    { name = "fastapi" },
+    { name = "filelock" },
+    { name = "lm-format-enforcer" },
+    { name = "ninja" },
+    { name = "numpy" },
+    { name = "nvidia-ml-py" },
+    { name = "openai" },
+    { name = "outlines" },
+    { name = "pillow" },
+    { name = "prometheus-client" },
+    { name = "prometheus-fastapi-instrumentator" },
+    { name = "psutil" },
+    { name = "py-cpuinfo" },
+    { name = "pydantic" },
+    { name = "ray" },
+    { name = "requests" },
+    { name = "sentencepiece" },
+    { name = "tiktoken" },
+    { name = "tokenizers" },
+    { name = "torch" },
+    { name = "torchvision" },
+    { name = "tqdm" },
+    { name = "transformers" },
+    { name = "typing-extensions" },
+    { name = "uvicorn", extra = ["standard"] },
+    { name = "vllm-flash-attn" },
+    { name = "xformers" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fc/d2/38993991ee61bacaccd49182ce597bf4496bc4ac96e92a3b011fcde90bb7/vllm-0.5.1.tar.gz", hash = "sha256:c7b6d01ec0644dd251e00a52bd08f9a172da2185dce87538deb56c577120e0eb", size = 790562 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/82/43/a0fbf45a2ccf038ea05774fba715633c70f7a26c7b3270c337038f9f77c0/vllm-0.5.1-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:562e3dd1d54ad0afb299e64412322b372976faaebb70a7db8e4a6c4ef5ed67c4", size = 146906281 },
+    { url = "https://files.pythonhosted.org/packages/13/79/e212b0eae62716be0d75994b71b7f97f4a45a7ffd8533408bcd3cde6a7e0/vllm-0.5.1-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:8b7483adcfc115dc7237663636a0976943fa65cde6f616a3839f0de69a419b5c", size = 146906182 },
+    { url = "https://files.pythonhosted.org/packages/72/89/671a176610d4ffe04216dccde6bffd07a8c67f4c0c45ce8be60266d7f574/vllm-0.5.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:6e6cad87335713cfc6710bc44c2688a6158abc3184d3530731e89b948617b2c7", size = 146906268 },
+]
+
+[[package]]
+name = "vllm-flash-attn"
+version = "2.5.9"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "torch" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2c/35/b00f367f3655bfed85a0df3c32fe6af34c0fd38b28ec7f0f9363991615f1/vllm_flash_attn-2.5.9-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:3e33af65880e99c47ff636f465d45c08e5e4c2eedbc06c9970599a6408768944", size = 37095920 },
+    { url = "https://files.pythonhosted.org/packages/36/42/307d9afa8c884efd369f7e402e69a588dea22b23fb23ce540b9ef6a284b0/vllm_flash_attn-2.5.9-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:88a63365db204859131fd8b3a0e85ea3b8b726f4be5ddbf2c6211a94a2cf4258", size = 37112678 },
+    { url = "https://files.pythonhosted.org/packages/f4/f2/155a33878f747b09250cf5a73e0449f52d23e723c770913952aa5340d66a/vllm_flash_attn-2.5.9-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:384845ae339946cb46e11cdfb5fadda1c0e519d2312b595e04da0df67a6d5383", size = 37092795 },
+]
+
+[[package]]
+name = "watchfiles"
+version = "1.0.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f5/26/c705fc77d0a9ecdb9b66f1e2976d95b81df3cae518967431e7dbf9b5e219/watchfiles-1.0.4.tar.gz", hash = "sha256:6ba473efd11062d73e4f00c2b730255f9c1bdd73cd5f9fe5b5da8dbd4a717205", size = 94625 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/14/02/22fcaed0396730b0d362bc8d1ffb3be2658fd473eecbb2ba84243e157f11/watchfiles-1.0.4-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:ba5bb3073d9db37c64520681dd2650f8bd40902d991e7b4cfaeece3e32561d08", size = 395212 },
+    { url = "https://files.pythonhosted.org/packages/e9/3d/ec5a2369a46edf3ebe092c39d9ae48e8cb6dacbde51c4b4f98936c524269/watchfiles-1.0.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9f25d0ba0fe2b6d2c921cf587b2bf4c451860086534f40c384329fb96e2044d1", size = 384815 },
+    { url = "https://files.pythonhosted.org/packages/df/b4/898991cececbe171e67142c31905510203649569d9817848f47c4177ee42/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:47eb32ef8c729dbc4f4273baece89398a4d4b5d21a1493efea77a17059f4df8a", size = 450680 },
+    { url = "https://files.pythonhosted.org/packages/58/f7/d4aa3000e812cfb5e5c2c6c0a3ec9d0a46a42489a8727edd160631c4e210/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:076f293100db3b0b634514aa0d294b941daa85fc777f9c698adb1009e5aca0b1", size = 455923 },
+    { url = "https://files.pythonhosted.org/packages/dd/95/7e2e4c6aba1b02fb5c76d2f6a450b85215921ec5f8f7ad5efd075369563f/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1eacd91daeb5158c598fe22d7ce66d60878b6294a86477a4715154990394c9b3", size = 482339 },
+    { url = "https://files.pythonhosted.org/packages/bb/67/4265b0fabcc2ef2c9e3e8802ba7908cf718a357ebfb49c72e53787156a48/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:13c2ce7b72026cfbca120d652f02c7750f33b4c9395d79c9790b27f014c8a5a2", size = 519908 },
+    { url = "https://files.pythonhosted.org/packages/0d/96/b57802d5f8164bdf070befb4fd3dec4edba5a364ec0670965a97eb8098ce/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:90192cdc15ab7254caa7765a98132a5a41471cf739513cc9bcf7d2ffcc0ec7b2", size = 501410 },
+    { url = "https://files.pythonhosted.org/packages/8b/18/6db0de4e8911ba14e31853201b40c0fa9fea5ecf3feb86b0ad58f006dfc3/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:278aaa395f405972e9f523bd786ed59dfb61e4b827856be46a42130605fd0899", size = 452876 },
+    { url = "https://files.pythonhosted.org/packages/df/df/092a961815edf723a38ba2638c49491365943919c3526cc9cf82c42786a6/watchfiles-1.0.4-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:a462490e75e466edbb9fc4cd679b62187153b3ba804868452ef0577ec958f5ff", size = 615353 },
+    { url = "https://files.pythonhosted.org/packages/f3/cf/b85fe645de4ff82f3f436c5e9032379fce37c303f6396a18f9726cc34519/watchfiles-1.0.4-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:8d0d0630930f5cd5af929040e0778cf676a46775753e442a3f60511f2409f48f", size = 613187 },
+    { url = "https://files.pythonhosted.org/packages/f6/d4/a9fea27aef4dd69689bc3556718c1157a7accb72aa035ece87c1fa8483b5/watchfiles-1.0.4-cp310-cp310-win32.whl", hash = "sha256:cc27a65069bcabac4552f34fd2dce923ce3fcde0721a16e4fb1b466d63ec831f", size = 270799 },
+    { url = "https://files.pythonhosted.org/packages/df/02/dbe9d4439f15dd4ad0720b6e039bde9d66d1f830331f34c18eb70fa6608e/watchfiles-1.0.4-cp310-cp310-win_amd64.whl", hash = "sha256:8b1f135238e75d075359cf506b27bf3f4ca12029c47d3e769d8593a2024ce161", size = 284145 },
+    { url = "https://files.pythonhosted.org/packages/0f/bb/8461adc4b1fed009546fb797fc0d5698dcfe5e289cb37e1b8f16a93cdc30/watchfiles-1.0.4-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:2a9f93f8439639dc244c4d2902abe35b0279102bca7bbcf119af964f51d53c19", size = 394869 },
+    { url = "https://files.pythonhosted.org/packages/55/88/9ebf36b3547176d1709c320de78c1fa3263a46be31b5b1267571d9102686/watchfiles-1.0.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:9eea33ad8c418847dd296e61eb683cae1c63329b6d854aefcd412e12d94ee235", size = 384905 },
+    { url = "https://files.pythonhosted.org/packages/03/8a/04335ce23ef78d8c69f0913e8b20cf7d9233e3986543aeef95ef2d6e43d2/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:31f1a379c9dcbb3f09cf6be1b7e83b67c0e9faabed0471556d9438a4a4e14202", size = 449944 },
+    { url = "https://files.pythonhosted.org/packages/17/4e/c8d5dcd14fe637f4633616dabea8a4af0a10142dccf3b43e0f081ba81ab4/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ab594e75644421ae0a2484554832ca5895f8cab5ab62de30a1a57db460ce06c6", size = 456020 },
+    { url = "https://files.pythonhosted.org/packages/5e/74/3e91e09e1861dd7fbb1190ce7bd786700dc0fbc2ccd33bb9fff5de039229/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fc2eb5d14a8e0d5df7b36288979176fbb39672d45184fc4b1c004d7c3ce29317", size = 482983 },
+    { url = "https://files.pythonhosted.org/packages/a1/3d/e64de2d1ce4eb6a574fd78ce3a28c279da263be9ef3cfcab6f708df192f2/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3f68d8e9d5a321163ddacebe97091000955a1b74cd43724e346056030b0bacee", size = 520320 },
+    { url = "https://files.pythonhosted.org/packages/2c/bd/52235f7063b57240c66a991696ed27e2a18bd6fcec8a1ea5a040b70d0611/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f9ce064e81fe79faa925ff03b9f4c1a98b0bbb4a1b8c1b015afa93030cb21a49", size = 500988 },
+    { url = "https://files.pythonhosted.org/packages/3a/b0/ff04194141a5fe650c150400dd9e42667916bc0f52426e2e174d779b8a74/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b77d5622ac5cc91d21ae9c2b284b5d5c51085a0bdb7b518dba263d0af006132c", size = 452573 },
+    { url = "https://files.pythonhosted.org/packages/3d/9d/966164332c5a178444ae6d165082d4f351bd56afd9c3ec828eecbf190e6a/watchfiles-1.0.4-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:1941b4e39de9b38b868a69b911df5e89dc43767feeda667b40ae032522b9b5f1", size = 615114 },
+    { url = "https://files.pythonhosted.org/packages/94/df/f569ae4c1877f96ad4086c153a8eee5a19a3b519487bf5c9454a3438c341/watchfiles-1.0.4-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:4f8c4998506241dedf59613082d1c18b836e26ef2a4caecad0ec41e2a15e4226", size = 613076 },
+    { url = "https://files.pythonhosted.org/packages/15/ae/8ce5f29e65d5fa5790e3c80c289819c55e12be2e1b9f5b6a0e55e169b97d/watchfiles-1.0.4-cp311-cp311-win32.whl", hash = "sha256:4ebbeca9360c830766b9f0df3640b791be569d988f4be6c06d6fae41f187f105", size = 271013 },
+    { url = "https://files.pythonhosted.org/packages/a4/c6/79dc4a7c598a978e5fafa135090aaf7bbb03b8dec7bada437dfbe578e7ed/watchfiles-1.0.4-cp311-cp311-win_amd64.whl", hash = "sha256:05d341c71f3d7098920f8551d4df47f7b57ac5b8dad56558064c3431bdfc0b74", size = 284229 },
+    { url = "https://files.pythonhosted.org/packages/37/3d/928633723211753f3500bfb138434f080363b87a1b08ca188b1ce54d1e05/watchfiles-1.0.4-cp311-cp311-win_arm64.whl", hash = "sha256:32b026a6ab64245b584acf4931fe21842374da82372d5c039cba6bf99ef722f3", size = 276824 },
+    { url = "https://files.pythonhosted.org/packages/5b/1a/8f4d9a1461709756ace48c98f07772bc6d4519b1e48b5fa24a4061216256/watchfiles-1.0.4-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:229e6ec880eca20e0ba2f7e2249c85bae1999d330161f45c78d160832e026ee2", size = 391345 },
+    { url = "https://files.pythonhosted.org/packages/bc/d2/6750b7b3527b1cdaa33731438432e7238a6c6c40a9924049e4cebfa40805/watchfiles-1.0.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:5717021b199e8353782dce03bd8a8f64438832b84e2885c4a645f9723bf656d9", size = 381515 },
+    { url = "https://files.pythonhosted.org/packages/4e/17/80500e42363deef1e4b4818729ed939aaddc56f82f4e72b2508729dd3c6b/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0799ae68dfa95136dde7c472525700bd48777875a4abb2ee454e3ab18e9fc712", size = 449767 },
+    { url = "https://files.pythonhosted.org/packages/10/37/1427fa4cfa09adbe04b1e97bced19a29a3462cc64c78630787b613a23f18/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:43b168bba889886b62edb0397cab5b6490ffb656ee2fcb22dec8bfeb371a9e12", size = 455677 },
+    { url = "https://files.pythonhosted.org/packages/c5/7a/39e9397f3a19cb549a7d380412fd9e507d4854eddc0700bfad10ef6d4dba/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fb2c46e275fbb9f0c92e7654b231543c7bbfa1df07cdc4b99fa73bedfde5c844", size = 482219 },
+    { url = "https://files.pythonhosted.org/packages/45/2d/7113931a77e2ea4436cad0c1690c09a40a7f31d366f79c6f0a5bc7a4f6d5/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:857f5fc3aa027ff5e57047da93f96e908a35fe602d24f5e5d8ce64bf1f2fc733", size = 518830 },
+    { url = "https://files.pythonhosted.org/packages/f9/1b/50733b1980fa81ef3c70388a546481ae5fa4c2080040100cd7bf3bf7b321/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:55ccfd27c497b228581e2838d4386301227fc0cb47f5a12923ec2fe4f97b95af", size = 497997 },
+    { url = "https://files.pythonhosted.org/packages/2b/b4/9396cc61b948ef18943e7c85ecfa64cf940c88977d882da57147f62b34b1/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5c11ea22304d17d4385067588123658e9f23159225a27b983f343fcffc3e796a", size = 452249 },
+    { url = "https://files.pythonhosted.org/packages/fb/69/0c65a5a29e057ad0dc691c2fa6c23b2983c7dabaa190ba553b29ac84c3cc/watchfiles-1.0.4-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:74cb3ca19a740be4caa18f238298b9d472c850f7b2ed89f396c00a4c97e2d9ff", size = 614412 },
+    { url = "https://files.pythonhosted.org/packages/7f/b9/319fcba6eba5fad34327d7ce16a6b163b39741016b1996f4a3c96b8dd0e1/watchfiles-1.0.4-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:c7cce76c138a91e720d1df54014a047e680b652336e1b73b8e3ff3158e05061e", size = 611982 },
+    { url = "https://files.pythonhosted.org/packages/f1/47/143c92418e30cb9348a4387bfa149c8e0e404a7c5b0585d46d2f7031b4b9/watchfiles-1.0.4-cp312-cp312-win32.whl", hash = "sha256:b045c800d55bc7e2cadd47f45a97c7b29f70f08a7c2fa13241905010a5493f94", size = 271822 },
+    { url = "https://files.pythonhosted.org/packages/ea/94/b0165481bff99a64b29e46e07ac2e0df9f7a957ef13bec4ceab8515f44e3/watchfiles-1.0.4-cp312-cp312-win_amd64.whl", hash = "sha256:c2acfa49dd0ad0bf2a9c0bb9a985af02e89345a7189be1efc6baa085e0f72d7c", size = 285441 },
+    { url = "https://files.pythonhosted.org/packages/11/de/09fe56317d582742d7ca8c2ca7b52a85927ebb50678d9b0fa8194658f536/watchfiles-1.0.4-cp312-cp312-win_arm64.whl", hash = "sha256:22bb55a7c9e564e763ea06c7acea24fc5d2ee5dfc5dafc5cfbedfe58505e9f90", size = 277141 },
+    { url = "https://files.pythonhosted.org/packages/08/98/f03efabec64b5b1fa58c0daab25c68ef815b0f320e54adcacd0d6847c339/watchfiles-1.0.4-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:8012bd820c380c3d3db8435e8cf7592260257b378b649154a7948a663b5f84e9", size = 390954 },
+    { url = "https://files.pythonhosted.org/packages/16/09/4dd49ba0a32a45813debe5fb3897955541351ee8142f586303b271a02b40/watchfiles-1.0.4-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:aa216f87594f951c17511efe5912808dfcc4befa464ab17c98d387830ce07b60", size = 381133 },
+    { url = "https://files.pythonhosted.org/packages/76/59/5aa6fc93553cd8d8ee75c6247763d77c02631aed21551a97d94998bf1dae/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:62c9953cf85529c05b24705639ffa390f78c26449e15ec34d5339e8108c7c407", size = 449516 },
+    { url = "https://files.pythonhosted.org/packages/4c/aa/df4b6fe14b6317290b91335b23c96b488d365d65549587434817e06895ea/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7cf684aa9bba4cd95ecb62c822a56de54e3ae0598c1a7f2065d51e24637a3c5d", size = 454820 },
+    { url = "https://files.pythonhosted.org/packages/5e/71/185f8672f1094ce48af33252c73e39b48be93b761273872d9312087245f6/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f44a39aee3cbb9b825285ff979ab887a25c5d336e5ec3574f1506a4671556a8d", size = 481550 },
+    { url = "https://files.pythonhosted.org/packages/85/d7/50ebba2c426ef1a5cb17f02158222911a2e005d401caf5d911bfca58f4c4/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a38320582736922be8c865d46520c043bff350956dfc9fbaee3b2df4e1740a4b", size = 518647 },
+    { url = "https://files.pythonhosted.org/packages/f0/7a/4c009342e393c545d68987e8010b937f72f47937731225b2b29b7231428f/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:39f4914548b818540ef21fd22447a63e7be6e24b43a70f7642d21f1e73371590", size = 497547 },
+    { url = "https://files.pythonhosted.org/packages/0f/7c/1cf50b35412d5c72d63b2bf9a4fffee2e1549a245924960dd087eb6a6de4/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f12969a3765909cf5dc1e50b2436eb2c0e676a3c75773ab8cc3aa6175c16e902", size = 452179 },
+    { url = "https://files.pythonhosted.org/packages/d6/a9/3db1410e1c1413735a9a472380e4f431ad9a9e81711cda2aaf02b7f62693/watchfiles-1.0.4-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:0986902677a1a5e6212d0c49b319aad9cc48da4bd967f86a11bde96ad9676ca1", size = 614125 },
+    { url = "https://files.pythonhosted.org/packages/f2/e1/0025d365cf6248c4d1ee4c3d2e3d373bdd3f6aff78ba4298f97b4fad2740/watchfiles-1.0.4-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:308ac265c56f936636e3b0e3f59e059a40003c655228c131e1ad439957592303", size = 611911 },
+    { url = "https://files.pythonhosted.org/packages/55/55/035838277d8c98fc8c917ac9beeb0cd6c59d675dc2421df5f9fcf44a0070/watchfiles-1.0.4-cp313-cp313-win32.whl", hash = "sha256:aee397456a29b492c20fda2d8961e1ffb266223625346ace14e4b6d861ba9c80", size = 271152 },
+    { url = "https://files.pythonhosted.org/packages/f0/e5/96b8e55271685ddbadc50ce8bc53aa2dff278fb7ac4c2e473df890def2dc/watchfiles-1.0.4-cp313-cp313-win_amd64.whl", hash = "sha256:d6097538b0ae5c1b88c3b55afa245a66793a8fec7ada6755322e465fb1a0e8cc", size = 285216 },
+    { url = "https://files.pythonhosted.org/packages/15/81/54484fc2fa715abe79694b975692af963f0878fb9d72b8251aa542bf3f10/watchfiles-1.0.4-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:d3452c1ec703aa1c61e15dfe9d482543e4145e7c45a6b8566978fbb044265a21", size = 394967 },
+    { url = "https://files.pythonhosted.org/packages/14/b3/557f0cd90add86586fe3deeebd11e8299db6bc3452b44a534f844c6ab831/watchfiles-1.0.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:7b75fee5a16826cf5c46fe1c63116e4a156924d668c38b013e6276f2582230f0", size = 384707 },
+    { url = "https://files.pythonhosted.org/packages/03/a3/34638e1bffcb85a405e7b005e30bb211fd9be2ab2cb1847f2ceb81bef27b/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e997802d78cdb02623b5941830ab06f8860038faf344f0d288d325cc9c5d2ff", size = 450442 },
+    { url = "https://files.pythonhosted.org/packages/8f/9f/6a97460dd11a606003d634c7158d9fea8517e98daffc6f56d0f5fde2e86a/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e0611d244ce94d83f5b9aff441ad196c6e21b55f77f3c47608dcf651efe54c4a", size = 455959 },
+    { url = "https://files.pythonhosted.org/packages/9d/bb/e0648c6364e4d37ec692bc3f0c77507d17d8bb8f75689148819142010bbf/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9745a4210b59e218ce64c91deb599ae8775c8a9da4e95fb2ee6fe745fc87d01a", size = 483187 },
+    { url = "https://files.pythonhosted.org/packages/dd/ad/d9290586a25288a81dfa8ad6329cf1de32aa1a9798ace45259eb95dcfb37/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4810ea2ae622add560f4aa50c92fef975e475f7ac4900ce5ff5547b2434642d8", size = 519733 },
+    { url = "https://files.pythonhosted.org/packages/4e/a9/150c1666825cc9637093f8cae7fc6f53b3296311ab8bd65f1389acb717cb/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:740d103cd01458f22462dedeb5a3382b7f2c57d07ff033fbc9465919e5e1d0f3", size = 502275 },
+    { url = "https://files.pythonhosted.org/packages/44/dc/5bfd21e20a330aca1706ac44713bc322838061938edf4b53130f97a7b211/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cdbd912a61543a36aef85e34f212e5d2486e7c53ebfdb70d1e0b060cc50dd0bf", size = 452907 },
+    { url = "https://files.pythonhosted.org/packages/50/fe/8f4fc488f1699f564687b697456eb5c0cb8e2b0b8538150511c234c62094/watchfiles-1.0.4-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:0bc80d91ddaf95f70258cf78c471246846c1986bcc5fd33ccc4a1a67fcb40f9a", size = 615927 },
+    { url = "https://files.pythonhosted.org/packages/ad/19/2e45f6f6eec89dd97a4d281635e3d73c17e5f692e7432063bdfdf9562c89/watchfiles-1.0.4-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:ab0311bb2ffcd9f74b6c9de2dda1612c13c84b996d032cd74799adb656af4e8b", size = 613435 },
+    { url = "https://files.pythonhosted.org/packages/91/17/dc5ac62ca377827c24321d68050efc2eaee2ebaf3f21d055bbce2206d309/watchfiles-1.0.4-cp39-cp39-win32.whl", hash = "sha256:02a526ee5b5a09e8168314c905fc545c9bc46509896ed282aeb5a8ba9bd6ca27", size = 270810 },
+    { url = "https://files.pythonhosted.org/packages/82/2b/dad851342492d538e7ffe72a8c756f747dd147988abb039ac9d6577d2235/watchfiles-1.0.4-cp39-cp39-win_amd64.whl", hash = "sha256:a5ae5706058b27c74bac987d615105da17724172d5aaacc6c362a40599b6de43", size = 284866 },
+    { url = "https://files.pythonhosted.org/packages/6f/06/175d5ac6b838fb319008c0cd981d7bf289317c510154d411d3584ca2b67b/watchfiles-1.0.4-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:cdcc92daeae268de1acf5b7befcd6cfffd9a047098199056c72e4623f531de18", size = 396269 },
+    { url = "https://files.pythonhosted.org/packages/86/ee/5db93b0b57dc0587abdbac4149296ee73275f615d790a82cb5598af0557f/watchfiles-1.0.4-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:d8d3d9203705b5797f0af7e7e5baa17c8588030aaadb7f6a86107b7247303817", size = 386010 },
+    { url = "https://files.pythonhosted.org/packages/75/61/fe0dc5fedf152bfc085a53711f740701f6bdb8ab6b5c950402b681d4858b/watchfiles-1.0.4-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bdef5a1be32d0b07dcea3318a0be95d42c98ece24177820226b56276e06b63b0", size = 450913 },
+    { url = "https://files.pythonhosted.org/packages/9f/dd/3c7731af3baf1a9957afc643d176f94480921a690ec3237c9f9d11301c08/watchfiles-1.0.4-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:342622287b5604ddf0ed2d085f3a589099c9ae8b7331df3ae9845571586c4f3d", size = 453474 },
+    { url = "https://files.pythonhosted.org/packages/6b/b4/c3998f54c91a35cee60ee6d3a855a069c5dff2bae6865147a46e9090dccd/watchfiles-1.0.4-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:9fe37a2de80aa785d340f2980276b17ef697ab8db6019b07ee4fd28a8359d2f3", size = 395565 },
+    { url = "https://files.pythonhosted.org/packages/3f/05/ac1a4d235beb9ddfb8ac26ce93a00ba6bd1b1b43051ef12d7da957b4a9d1/watchfiles-1.0.4-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:9d1ef56b56ed7e8f312c934436dea93bfa3e7368adfcf3df4c0da6d4de959a1e", size = 385406 },
+    { url = "https://files.pythonhosted.org/packages/4c/ea/36532e7d86525f4e52a10efed182abf33efb106a93d49f5fbc994b256bcd/watchfiles-1.0.4-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:95b42cac65beae3a362629950c444077d1b44f1790ea2772beaea95451c086bb", size = 450424 },
+    { url = "https://files.pythonhosted.org/packages/7a/e9/3cbcf4d70cd0b6d3f30631deae1bf37cc0be39887ca327a44462fe546bf5/watchfiles-1.0.4-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5e0227b8ed9074c6172cf55d85b5670199c99ab11fd27d2c473aa30aec67ee42", size = 452488 },
+]
+
+[[package]]
+name = "websockets"
+version = "14.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/94/54/8359678c726243d19fae38ca14a334e740782336c9f19700858c4eb64a1e/websockets-14.2.tar.gz", hash = "sha256:5059ed9c54945efb321f097084b4c7e52c246f2c869815876a69d1efc4ad6eb5", size = 164394 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/28/fa/76607eb7dcec27b2d18d63f60a32e60e2b8629780f343bb83a4dbb9f4350/websockets-14.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e8179f95323b9ab1c11723e5d91a89403903f7b001828161b480a7810b334885", size = 163089 },
+    { url = "https://files.pythonhosted.org/packages/9e/00/ad2246b5030575b79e7af0721810fdaecaf94c4b2625842ef7a756fa06dd/websockets-14.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0d8c3e2cdb38f31d8bd7d9d28908005f6fa9def3324edb9bf336d7e4266fd397", size = 160741 },
+    { url = "https://files.pythonhosted.org/packages/72/f7/60f10924d333a28a1ff3fcdec85acf226281331bdabe9ad74947e1b7fc0a/websockets-14.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:714a9b682deb4339d39ffa674f7b674230227d981a37d5d174a4a83e3978a610", size = 160996 },
+    { url = "https://files.pythonhosted.org/packages/63/7c/c655789cf78648c01ac6ecbe2d6c18f91b75bdc263ffee4d08ce628d12f0/websockets-14.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f2e53c72052f2596fb792a7acd9704cbc549bf70fcde8a99e899311455974ca3", size = 169974 },
+    { url = "https://files.pythonhosted.org/packages/fb/5b/013ed8b4611857ac92ac631079c08d9715b388bd1d88ec62e245f87a39df/websockets-14.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e3fbd68850c837e57373d95c8fe352203a512b6e49eaae4c2f4088ef8cf21980", size = 168985 },
+    { url = "https://files.pythonhosted.org/packages/cd/33/aa3e32fd0df213a5a442310754fe3f89dd87a0b8e5b4e11e0991dd3bcc50/websockets-14.2-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b27ece32f63150c268593d5fdb82819584831a83a3f5809b7521df0685cd5d8", size = 169297 },
+    { url = "https://files.pythonhosted.org/packages/93/17/dae0174883d6399f57853ac44abf5f228eaba86d98d160f390ffabc19b6e/websockets-14.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4daa0faea5424d8713142b33825fff03c736f781690d90652d2c8b053345b0e7", size = 169677 },
+    { url = "https://files.pythonhosted.org/packages/42/e2/0375af7ac00169b98647c804651c515054b34977b6c1354f1458e4116c1e/websockets-14.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:bc63cee8596a6ec84d9753fd0fcfa0452ee12f317afe4beae6b157f0070c6c7f", size = 169089 },
+    { url = "https://files.pythonhosted.org/packages/73/8d/80f71d2a351a44b602859af65261d3dde3a0ce4e76cf9383738a949e0cc3/websockets-14.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:7a570862c325af2111343cc9b0257b7119b904823c675b22d4ac547163088d0d", size = 169026 },
+    { url = "https://files.pythonhosted.org/packages/48/97/173b1fa6052223e52bb4054a141433ad74931d94c575e04b654200b98ca4/websockets-14.2-cp310-cp310-win32.whl", hash = "sha256:75862126b3d2d505e895893e3deac0a9339ce750bd27b4ba515f008b5acf832d", size = 163967 },
+    { url = "https://files.pythonhosted.org/packages/c0/5b/2fcf60f38252a4562b28b66077e0d2b48f91fef645d5f78874cd1dec807b/websockets-14.2-cp310-cp310-win_amd64.whl", hash = "sha256:cc45afb9c9b2dc0852d5c8b5321759cf825f82a31bfaf506b65bf4668c96f8b2", size = 164413 },
+    { url = "https://files.pythonhosted.org/packages/15/b6/504695fb9a33df0ca56d157f5985660b5fc5b4bf8c78f121578d2d653392/websockets-14.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:3bdc8c692c866ce5fefcaf07d2b55c91d6922ac397e031ef9b774e5b9ea42166", size = 163088 },
+    { url = "https://files.pythonhosted.org/packages/81/26/ebfb8f6abe963c795122439c6433c4ae1e061aaedfc7eff32d09394afbae/websockets-14.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c93215fac5dadc63e51bcc6dceca72e72267c11def401d6668622b47675b097f", size = 160745 },
+    { url = "https://files.pythonhosted.org/packages/a1/c6/1435ad6f6dcbff80bb95e8986704c3174da8866ddb751184046f5c139ef6/websockets-14.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1c9b6535c0e2cf8a6bf938064fb754aaceb1e6a4a51a80d884cd5db569886910", size = 160995 },
+    { url = "https://files.pythonhosted.org/packages/96/63/900c27cfe8be1a1f2433fc77cd46771cf26ba57e6bdc7cf9e63644a61863/websockets-14.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0a52a6d7cf6938e04e9dceb949d35fbdf58ac14deea26e685ab6368e73744e4c", size = 170543 },
+    { url = "https://files.pythonhosted.org/packages/00/8b/bec2bdba92af0762d42d4410593c1d7d28e9bfd952c97a3729df603dc6ea/websockets-14.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9f05702e93203a6ff5226e21d9b40c037761b2cfb637187c9802c10f58e40473", size = 169546 },
+    { url = "https://files.pythonhosted.org/packages/6b/a9/37531cb5b994f12a57dec3da2200ef7aadffef82d888a4c29a0d781568e4/websockets-14.2-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:22441c81a6748a53bfcb98951d58d1af0661ab47a536af08920d129b4d1c3473", size = 169911 },
+    { url = "https://files.pythonhosted.org/packages/60/d5/a6eadba2ed9f7e65d677fec539ab14a9b83de2b484ab5fe15d3d6d208c28/websockets-14.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:efd9b868d78b194790e6236d9cbc46d68aba4b75b22497eb4ab64fa640c3af56", size = 170183 },
+    { url = "https://files.pythonhosted.org/packages/76/57/a338ccb00d1df881c1d1ee1f2a20c9c1b5b29b51e9e0191ee515d254fea6/websockets-14.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:1a5a20d5843886d34ff8c57424cc65a1deda4375729cbca4cb6b3353f3ce4142", size = 169623 },
+    { url = "https://files.pythonhosted.org/packages/64/22/e5f7c33db0cb2c1d03b79fd60d189a1da044e2661f5fd01d629451e1db89/websockets-14.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:34277a29f5303d54ec6468fb525d99c99938607bc96b8d72d675dee2b9f5bf1d", size = 169583 },
+    { url = "https://files.pythonhosted.org/packages/aa/2e/2b4662237060063a22e5fc40d46300a07142afe30302b634b4eebd717c07/websockets-14.2-cp311-cp311-win32.whl", hash = "sha256:02687db35dbc7d25fd541a602b5f8e451a238ffa033030b172ff86a93cb5dc2a", size = 163969 },
+    { url = "https://files.pythonhosted.org/packages/94/a5/0cda64e1851e73fc1ecdae6f42487babb06e55cb2f0dc8904b81d8ef6857/websockets-14.2-cp311-cp311-win_amd64.whl", hash = "sha256:862e9967b46c07d4dcd2532e9e8e3c2825e004ffbf91a5ef9dde519ee2effb0b", size = 164408 },
+    { url = "https://files.pythonhosted.org/packages/c1/81/04f7a397653dc8bec94ddc071f34833e8b99b13ef1a3804c149d59f92c18/websockets-14.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:1f20522e624d7ffbdbe259c6b6a65d73c895045f76a93719aa10cd93b3de100c", size = 163096 },
+    { url = "https://files.pythonhosted.org/packages/ec/c5/de30e88557e4d70988ed4d2eabd73fd3e1e52456b9f3a4e9564d86353b6d/websockets-14.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:647b573f7d3ada919fd60e64d533409a79dcf1ea21daeb4542d1d996519ca967", size = 160758 },
+    { url = "https://files.pythonhosted.org/packages/e5/8c/d130d668781f2c77d106c007b6c6c1d9db68239107c41ba109f09e6c218a/websockets-14.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6af99a38e49f66be5a64b1e890208ad026cda49355661549c507152113049990", size = 160995 },
+    { url = "https://files.pythonhosted.org/packages/a6/bc/f6678a0ff17246df4f06765e22fc9d98d1b11a258cc50c5968b33d6742a1/websockets-14.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:091ab63dfc8cea748cc22c1db2814eadb77ccbf82829bac6b2fbe3401d548eda", size = 170815 },
+    { url = "https://files.pythonhosted.org/packages/d8/b2/8070cb970c2e4122a6ef38bc5b203415fd46460e025652e1ee3f2f43a9a3/websockets-14.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b374e8953ad477d17e4851cdc66d83fdc2db88d9e73abf755c94510ebddceb95", size = 169759 },
+    { url = "https://files.pythonhosted.org/packages/81/da/72f7caabd94652e6eb7e92ed2d3da818626e70b4f2b15a854ef60bf501ec/websockets-14.2-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a39d7eceeea35db85b85e1169011bb4321c32e673920ae9c1b6e0978590012a3", size = 170178 },
+    { url = "https://files.pythonhosted.org/packages/31/e0/812725b6deca8afd3a08a2e81b3c4c120c17f68c9b84522a520b816cda58/websockets-14.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:0a6f3efd47ffd0d12080594f434faf1cd2549b31e54870b8470b28cc1d3817d9", size = 170453 },
+    { url = "https://files.pythonhosted.org/packages/66/d3/8275dbc231e5ba9bb0c4f93144394b4194402a7a0c8ffaca5307a58ab5e3/websockets-14.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:065ce275e7c4ffb42cb738dd6b20726ac26ac9ad0a2a48e33ca632351a737267", size = 169830 },
+    { url = "https://files.pythonhosted.org/packages/a3/ae/e7d1a56755ae15ad5a94e80dd490ad09e345365199600b2629b18ee37bc7/websockets-14.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e9d0e53530ba7b8b5e389c02282f9d2aa47581514bd6049d3a7cffe1385cf5fe", size = 169824 },
+    { url = "https://files.pythonhosted.org/packages/b6/32/88ccdd63cb261e77b882e706108d072e4f1c839ed723bf91a3e1f216bf60/websockets-14.2-cp312-cp312-win32.whl", hash = "sha256:20e6dd0984d7ca3037afcb4494e48c74ffb51e8013cac71cf607fffe11df7205", size = 163981 },
+    { url = "https://files.pythonhosted.org/packages/b3/7d/32cdb77990b3bdc34a306e0a0f73a1275221e9a66d869f6ff833c95b56ef/websockets-14.2-cp312-cp312-win_amd64.whl", hash = "sha256:44bba1a956c2c9d268bdcdf234d5e5ff4c9b6dc3e300545cbe99af59dda9dcce", size = 164421 },
+    { url = "https://files.pythonhosted.org/packages/82/94/4f9b55099a4603ac53c2912e1f043d6c49d23e94dd82a9ce1eb554a90215/websockets-14.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:6f1372e511c7409a542291bce92d6c83320e02c9cf392223272287ce55bc224e", size = 163102 },
+    { url = "https://files.pythonhosted.org/packages/8e/b7/7484905215627909d9a79ae07070057afe477433fdacb59bf608ce86365a/websockets-14.2-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:4da98b72009836179bb596a92297b1a61bb5a830c0e483a7d0766d45070a08ad", size = 160766 },
+    { url = "https://files.pythonhosted.org/packages/a3/a4/edb62efc84adb61883c7d2c6ad65181cb087c64252138e12d655989eec05/websockets-14.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f8a86a269759026d2bde227652b87be79f8a734e582debf64c9d302faa1e9f03", size = 160998 },
+    { url = "https://files.pythonhosted.org/packages/f5/79/036d320dc894b96af14eac2529967a6fc8b74f03b83c487e7a0e9043d842/websockets-14.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:86cf1aaeca909bf6815ea714d5c5736c8d6dd3a13770e885aafe062ecbd04f1f", size = 170780 },
+    { url = "https://files.pythonhosted.org/packages/63/75/5737d21ee4dd7e4b9d487ee044af24a935e36a9ff1e1419d684feedcba71/websockets-14.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a9b0f6c3ba3b1240f602ebb3971d45b02cc12bd1845466dd783496b3b05783a5", size = 169717 },
+    { url = "https://files.pythonhosted.org/packages/2c/3c/bf9b2c396ed86a0b4a92ff4cdaee09753d3ee389be738e92b9bbd0330b64/websockets-14.2-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:669c3e101c246aa85bc8534e495952e2ca208bd87994650b90a23d745902db9a", size = 170155 },
+    { url = "https://files.pythonhosted.org/packages/75/2d/83a5aca7247a655b1da5eb0ee73413abd5c3a57fc8b92915805e6033359d/websockets-14.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:eabdb28b972f3729348e632ab08f2a7b616c7e53d5414c12108c29972e655b20", size = 170495 },
+    { url = "https://files.pythonhosted.org/packages/79/dd/699238a92761e2f943885e091486378813ac8f43e3c84990bc394c2be93e/websockets-14.2-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:2066dc4cbcc19f32c12a5a0e8cc1b7ac734e5b64ac0a325ff8353451c4b15ef2", size = 169880 },
+    { url = "https://files.pythonhosted.org/packages/c8/c9/67a8f08923cf55ce61aadda72089e3ed4353a95a3a4bc8bf42082810e580/websockets-14.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:ab95d357cd471df61873dadf66dd05dd4709cae001dd6342edafc8dc6382f307", size = 169856 },
+    { url = "https://files.pythonhosted.org/packages/17/b1/1ffdb2680c64e9c3921d99db460546194c40d4acbef999a18c37aa4d58a3/websockets-14.2-cp313-cp313-win32.whl", hash = "sha256:a9e72fb63e5f3feacdcf5b4ff53199ec8c18d66e325c34ee4c551ca748623bbc", size = 163974 },
+    { url = "https://files.pythonhosted.org/packages/14/13/8b7fc4cb551b9cfd9890f0fd66e53c18a06240319915533b033a56a3d520/websockets-14.2-cp313-cp313-win_amd64.whl", hash = "sha256:b439ea828c4ba99bb3176dc8d9b933392a2413c0f6b149fdcba48393f573377f", size = 164420 },
+    { url = "https://files.pythonhosted.org/packages/6f/eb/367e0ed7b8a960b4fc12c7c6bf3ebddf06875037de641637994849560d47/websockets-14.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:7cd5706caec1686c5d233bc76243ff64b1c0dc445339bd538f30547e787c11fe", size = 163087 },
+    { url = "https://files.pythonhosted.org/packages/96/f7/1f18d028ec4a2c14598dfec6a73381a915c27464b693873198c1de872095/websockets-14.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ec607328ce95a2f12b595f7ae4c5d71bf502212bddcea528290b35c286932b12", size = 160740 },
+    { url = "https://files.pythonhosted.org/packages/5c/db/b4b353fb9c3f0eaa8138ea4c76e6fa555b6d2821ed2d51d0ac3c320bc57e/websockets-14.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:da85651270c6bfb630136423037dd4975199e5d4114cae6d3066641adcc9d1c7", size = 160992 },
+    { url = "https://files.pythonhosted.org/packages/b9/b1/9149e420c61f375e432654d5c1545e563b90ac1f829ee1a8d1dccaf0869d/websockets-14.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c3ecadc7ce90accf39903815697917643f5b7cfb73c96702318a096c00aa71f5", size = 169757 },
+    { url = "https://files.pythonhosted.org/packages/2b/33/0bb58204191e113212360f1392b6b1e9f85f62c7ca5b3b15f52f2f835516/websockets-14.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1979bee04af6a78608024bad6dfcc0cc930ce819f9e10342a29a05b5320355d0", size = 168762 },
+    { url = "https://files.pythonhosted.org/packages/be/3d/c3c192f16210d7b7535fbf4ee9a299612f4dccff665587617b13fa0a6aa3/websockets-14.2-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2dddacad58e2614a24938a50b85969d56f88e620e3f897b7d80ac0d8a5800258", size = 169060 },
+    { url = "https://files.pythonhosted.org/packages/a6/73/75efa8d9e4b1b257818a7b7a0b9ac84a07c91120b52148941370ef2c8f16/websockets-14.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:89a71173caaf75fa71a09a5f614f450ba3ec84ad9fca47cb2422a860676716f0", size = 169457 },
+    { url = "https://files.pythonhosted.org/packages/a4/11/300cf36cfd6990ffb218394862f0513be8c21917c9ff5e362f94599caedd/websockets-14.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6af6a4b26eea4fc06c6818a6b962a952441e0e39548b44773502761ded8cc1d4", size = 168860 },
+    { url = "https://files.pythonhosted.org/packages/c0/3d/5fd82500714ab7c09f003bde671dad1a3a131ac77b6b11ada72e466de4f6/websockets-14.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:80c8efa38957f20bba0117b48737993643204645e9ec45512579132508477cfc", size = 168825 },
+    { url = "https://files.pythonhosted.org/packages/88/16/715580eb6caaacc232f303e9619103a42dcd354b0854baa5ed26aacaf828/websockets-14.2-cp39-cp39-win32.whl", hash = "sha256:2e20c5f517e2163d76e2729104abc42639c41cf91f7b1839295be43302713661", size = 163960 },
+    { url = "https://files.pythonhosted.org/packages/63/a7/a1035cb198eaa12eaa9621aaaa3ec021b0e3bac96e1df9ceb6bfe5e53e5f/websockets-14.2-cp39-cp39-win_amd64.whl", hash = "sha256:b4c8cef610e8d7c70dea92e62b6814a8cd24fbd01d7103cc89308d2bfe1659ef", size = 164424 },
+    { url = "https://files.pythonhosted.org/packages/10/3d/91d3d2bb1325cd83e8e2c02d0262c7d4426dc8fa0831ef1aa4d6bf2041af/websockets-14.2-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:d7d9cafbccba46e768be8a8ad4635fa3eae1ffac4c6e7cb4eb276ba41297ed29", size = 160773 },
+    { url = "https://files.pythonhosted.org/packages/33/7c/cdedadfef7381939577858b1b5718a4ab073adbb584e429dd9d9dc9bfe16/websockets-14.2-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:c76193c1c044bd1e9b3316dcc34b174bbf9664598791e6fb606d8d29000e070c", size = 161007 },
+    { url = "https://files.pythonhosted.org/packages/ca/35/7a20a3c450b27c04e50fbbfc3dfb161ed8e827b2a26ae31c4b59b018b8c6/websockets-14.2-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fd475a974d5352390baf865309fe37dec6831aafc3014ffac1eea99e84e83fc2", size = 162264 },
+    { url = "https://files.pythonhosted.org/packages/e8/9c/e3f9600564b0c813f2448375cf28b47dc42c514344faed3a05d71fb527f9/websockets-14.2-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2c6c0097a41968b2e2b54ed3424739aab0b762ca92af2379f152c1aef0187e1c", size = 161873 },
+    { url = "https://files.pythonhosted.org/packages/3f/37/260f189b16b2b8290d6ae80c9f96d8b34692cf1bb3475df54c38d3deb57d/websockets-14.2-pp310-pypy310_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6d7ff794c8b36bc402f2e07c0b2ceb4a2424147ed4785ff03e2a7af03711d60a", size = 161818 },
+    { url = "https://files.pythonhosted.org/packages/ff/1e/e47dedac8bf7140e59aa6a679e850c4df9610ae844d71b6015263ddea37b/websockets-14.2-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:dec254fcabc7bd488dab64846f588fc5b6fe0d78f641180030f8ea27b76d72c3", size = 164465 },
+    { url = "https://files.pythonhosted.org/packages/f7/c0/8e9325c4987dcf66d4a0d63ec380d4aefe8dcc1e521af71ad17adf2c1ae2/websockets-14.2-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:bbe03eb853e17fd5b15448328b4ec7fb2407d45fb0245036d06a3af251f8e48f", size = 160773 },
+    { url = "https://files.pythonhosted.org/packages/5a/6e/c9a7f2edd4afddc4f8cccfc4e12468b7f6ec40f28d1b1e966a8d0298b875/websockets-14.2-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:a3c4aa3428b904d5404a0ed85f3644d37e2cb25996b7f096d77caeb0e96a3b42", size = 161006 },
+    { url = "https://files.pythonhosted.org/packages/f3/10/b90ece894828c954e674a81cb0db250e6c324c54db30a8b19e96431f928f/websockets-14.2-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:577a4cebf1ceaf0b65ffc42c54856214165fb8ceeba3935852fc33f6b0c55e7f", size = 162260 },
+    { url = "https://files.pythonhosted.org/packages/52/93/1147b6b5464a5fb6e8987da3ec7991dcc44f9090f67d9c841d7382fed429/websockets-14.2-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ad1c1d02357b7665e700eca43a31d52814ad9ad9b89b58118bdabc365454b574", size = 161868 },
+    { url = "https://files.pythonhosted.org/packages/32/ab/f7d80b4049bff0aa617507330db3a27389d0e70df54e29f7a3d76bbd2086/websockets-14.2-pp39-pypy39_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f390024a47d904613577df83ba700bd189eedc09c57af0a904e5c39624621270", size = 161813 },
+    { url = "https://files.pythonhosted.org/packages/cd/cc/adc9fb85f031b8df8e9f3d96cc004df25d2643e503953af5223c5b6825b7/websockets-14.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:3c1426c021c38cf92b453cdf371228d3430acd775edee6bac5a4d577efc72365", size = 164457 },
+    { url = "https://files.pythonhosted.org/packages/7b/c8/d529f8a32ce40d98309f4470780631e971a5a842b60aec864833b3615786/websockets-14.2-py3-none-any.whl", hash = "sha256:7a6ceec4ea84469f15cf15807a747e9efe57e369c384fa86e022b3bea679b79b", size = 157416 },
+]
+
+[[package]]
+name = "xformers"
+version = "0.0.26.post1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy" },
+    { name = "torch" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/fc/d3/5eab1cb3171fc9d4e5613025e965b72727f6d12ae820d2d7ab9c37320eb6/xformers-0.0.26.post1.tar.gz", hash = "sha256:1d14b5f999ede649198379b0470ebdd25007ba224ae336ef958124158a6de8b1", size = 4087142 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/32/e7/27003645ef99e7571fb6964cd2f39da3f1b3f3011aa00bb2d3ac9b790757/xformers-0.0.26.post1-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:1fe0f9a3dddb7f6175c2e34de2f318d6742eec28c068b8334b093016b2c9c2c8", size = 222745400 },
+    { url = "https://files.pythonhosted.org/packages/7a/53/f4a36d24f6c2d06fccc356986bb45c2cb2fd06609986d44a925d57f38070/xformers-0.0.26.post1-cp310-cp310-win_amd64.whl", hash = "sha256:e34b8dd6982077bee0c8eb2db8bc1513177201bfe0af890a4db42d8d31c966a5", size = 208730032 },
+    { url = "https://files.pythonhosted.org/packages/ce/72/58e13d633efada336f65e23290dfc0bc9497c60fd9b8fdaab59b555b5f28/xformers-0.0.26.post1-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:0d35615870b9237077aec51802a5a821f9e9d2708730c8914d5cff301fb29558", size = 222780107 },
+    { url = "https://files.pythonhosted.org/packages/bb/6b/ac99df9545d965c11172ec3613b4652d6f0f9098328551cfe0bbbfa634fe/xformers-0.0.26.post1-cp311-cp311-win_amd64.whl", hash = "sha256:d05c547b4ba603fc8e21fad03a342138eaaece35fe0a1692e6ee0d061ddd21ac", size = 208727774 },
+    { url = "https://files.pythonhosted.org/packages/1d/42/7728b1d16425d3f5505eeb70749e15a71257e126a90ab44184d231d00341/xformers-0.0.26.post1-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:16dcfa801450a03f148d6d5f7ba2b5540a27c52517946570fe0e1f75238d73a1", size = 222743742 },
+    { url = "https://files.pythonhosted.org/packages/28/f7/452f15a1a69ef553f05119e03f4c26fa683793491520c88d0055ca90c9ec/xformers-0.0.26.post1-cp39-cp39-win_amd64.whl", hash = "sha256:cf267bac8f4454db1056e4e6ff9e4df1e02b2b31d8116d50913af15fa6ae7132", size = 208757765 },
+]
+
+[[package]]
+name = "xxhash"
+version = "3.5.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/00/5e/d6e5258d69df8b4ed8c83b6664f2b47d30d2dec551a29ad72a6c69eafd31/xxhash-3.5.0.tar.gz", hash = "sha256:84f2caddf951c9cbf8dc2e22a89d4ccf5d86391ac6418fe81e3c67d0cf60b45f", size = 84241 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/bb/8a/0e9feca390d512d293afd844d31670e25608c4a901e10202aa98785eab09/xxhash-3.5.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:ece616532c499ee9afbb83078b1b952beffef121d989841f7f4b3dc5ac0fd212", size = 31970 },
+    { url = "https://files.pythonhosted.org/packages/16/e6/be5aa49580cd064a18200ab78e29b88b1127e1a8c7955eb8ecf81f2626eb/xxhash-3.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:3171f693dbc2cef6477054a665dc255d996646b4023fe56cb4db80e26f4cc520", size = 30801 },
+    { url = "https://files.pythonhosted.org/packages/20/ee/b8a99ebbc6d1113b3a3f09e747fa318c3cde5b04bd9c197688fadf0eeae8/xxhash-3.5.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7c5d3e570ef46adaf93fc81b44aca6002b5a4d8ca11bd0580c07eac537f36680", size = 220927 },
+    { url = "https://files.pythonhosted.org/packages/58/62/15d10582ef159283a5c2b47f6d799fc3303fe3911d5bb0bcc820e1ef7ff4/xxhash-3.5.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7cb29a034301e2982df8b1fe6328a84f4b676106a13e9135a0d7e0c3e9f806da", size = 200360 },
+    { url = "https://files.pythonhosted.org/packages/23/41/61202663ea9b1bd8e53673b8ec9e2619989353dba8cfb68e59a9cbd9ffe3/xxhash-3.5.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5d0d307d27099bb0cbeea7260eb39ed4fdb99c5542e21e94bb6fd29e49c57a23", size = 428528 },
+    { url = "https://files.pythonhosted.org/packages/f2/07/d9a3059f702dec5b3b703737afb6dda32f304f6e9da181a229dafd052c29/xxhash-3.5.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0342aafd421795d740e514bc9858ebddfc705a75a8c5046ac56d85fe97bf196", size = 194149 },
+    { url = "https://files.pythonhosted.org/packages/eb/58/27caadf78226ecf1d62dbd0c01d152ed381c14c1ee4ad01f0d460fc40eac/xxhash-3.5.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3dbbd9892c5ebffeca1ed620cf0ade13eb55a0d8c84e0751a6653adc6ac40d0c", size = 207703 },
+    { url = "https://files.pythonhosted.org/packages/b1/08/32d558ce23e1e068453c39aed7b3c1cdc690c177873ec0ca3a90d5808765/xxhash-3.5.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4cc2d67fdb4d057730c75a64c5923abfa17775ae234a71b0200346bfb0a7f482", size = 216255 },
+    { url = "https://files.pythonhosted.org/packages/3f/d4/2b971e2d2b0a61045f842b622ef11e94096cf1f12cd448b6fd426e80e0e2/xxhash-3.5.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:ec28adb204b759306a3d64358a5e5c07d7b1dd0ccbce04aa76cb9377b7b70296", size = 202744 },
+    { url = "https://files.pythonhosted.org/packages/19/ae/6a6438864a8c4c39915d7b65effd85392ebe22710412902487e51769146d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:1328f6d8cca2b86acb14104e381225a3d7b42c92c4b86ceae814e5c400dbb415", size = 210115 },
+    { url = "https://files.pythonhosted.org/packages/48/7d/b3c27c27d1fc868094d02fe4498ccce8cec9fcc591825c01d6bcb0b4fc49/xxhash-3.5.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:8d47ebd9f5d9607fd039c1fbf4994e3b071ea23eff42f4ecef246ab2b7334198", size = 414247 },
+    { url = "https://files.pythonhosted.org/packages/a1/05/918f9e7d2fbbd334b829997045d341d6239b563c44e683b9a7ef8fe50f5d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b96d559e0fcddd3343c510a0fe2b127fbff16bf346dd76280b82292567523442", size = 191419 },
+    { url = "https://files.pythonhosted.org/packages/08/29/dfe393805b2f86bfc47c290b275f0b7c189dc2f4e136fd4754f32eb18a8d/xxhash-3.5.0-cp310-cp310-win32.whl", hash = "sha256:61c722ed8d49ac9bc26c7071eeaa1f6ff24053d553146d5df031802deffd03da", size = 30114 },
+    { url = "https://files.pythonhosted.org/packages/7b/d7/aa0b22c4ebb7c3ccb993d4c565132abc641cd11164f8952d89eb6a501909/xxhash-3.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:9bed5144c6923cc902cd14bb8963f2d5e034def4486ab0bbe1f58f03f042f9a9", size = 30003 },
+    { url = "https://files.pythonhosted.org/packages/69/12/f969b81541ee91b55f1ce469d7ab55079593c80d04fd01691b550e535000/xxhash-3.5.0-cp310-cp310-win_arm64.whl", hash = "sha256:893074d651cf25c1cc14e3bea4fceefd67f2921b1bb8e40fcfeba56820de80c6", size = 26773 },
+    { url = "https://files.pythonhosted.org/packages/b8/c7/afed0f131fbda960ff15eee7f304fa0eeb2d58770fade99897984852ef23/xxhash-3.5.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:02c2e816896dc6f85922ced60097bcf6f008dedfc5073dcba32f9c8dd786f3c1", size = 31969 },
+    { url = "https://files.pythonhosted.org/packages/8c/0c/7c3bc6d87e5235672fcc2fb42fd5ad79fe1033925f71bf549ee068c7d1ca/xxhash-3.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6027dcd885e21581e46d3c7f682cfb2b870942feeed58a21c29583512c3f09f8", size = 30800 },
+    { url = "https://files.pythonhosted.org/packages/04/9e/01067981d98069eec1c20201f8c145367698e9056f8bc295346e4ea32dd1/xxhash-3.5.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1308fa542bbdbf2fa85e9e66b1077eea3a88bef38ee8a06270b4298a7a62a166", size = 221566 },
+    { url = "https://files.pythonhosted.org/packages/d4/09/d4996de4059c3ce5342b6e1e6a77c9d6c91acce31f6ed979891872dd162b/xxhash-3.5.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c28b2fdcee797e1c1961cd3bcd3d545cab22ad202c846235197935e1df2f8ef7", size = 201214 },
+    { url = "https://files.pythonhosted.org/packages/62/f5/6d2dc9f8d55a7ce0f5e7bfef916e67536f01b85d32a9fbf137d4cadbee38/xxhash-3.5.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:924361811732ddad75ff23e90efd9ccfda4f664132feecb90895bade6a1b4623", size = 429433 },
+    { url = "https://files.pythonhosted.org/packages/d9/72/9256303f10e41ab004799a4aa74b80b3c5977d6383ae4550548b24bd1971/xxhash-3.5.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:89997aa1c4b6a5b1e5b588979d1da048a3c6f15e55c11d117a56b75c84531f5a", size = 194822 },
+    { url = "https://files.pythonhosted.org/packages/34/92/1a3a29acd08248a34b0e6a94f4e0ed9b8379a4ff471f1668e4dce7bdbaa8/xxhash-3.5.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:685c4f4e8c59837de103344eb1c8a3851f670309eb5c361f746805c5471b8c88", size = 208538 },
+    { url = "https://files.pythonhosted.org/packages/53/ad/7fa1a109663366de42f724a1cdb8e796a260dbac45047bce153bc1e18abf/xxhash-3.5.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dbd2ecfbfee70bc1a4acb7461fa6af7748ec2ab08ac0fa298f281c51518f982c", size = 216953 },
+    { url = "https://files.pythonhosted.org/packages/35/02/137300e24203bf2b2a49b48ce898ecce6fd01789c0fcd9c686c0a002d129/xxhash-3.5.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:25b5a51dc3dfb20a10833c8eee25903fd2e14059e9afcd329c9da20609a307b2", size = 203594 },
+    { url = "https://files.pythonhosted.org/packages/23/03/aeceb273933d7eee248c4322b98b8e971f06cc3880e5f7602c94e5578af5/xxhash-3.5.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a8fb786fb754ef6ff8c120cb96629fb518f8eb5a61a16aac3a979a9dbd40a084", size = 210971 },
+    { url = "https://files.pythonhosted.org/packages/e3/64/ed82ec09489474cbb35c716b189ddc1521d8b3de12b1b5ab41ce7f70253c/xxhash-3.5.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:a905ad00ad1e1c34fe4e9d7c1d949ab09c6fa90c919860c1534ff479f40fd12d", size = 415050 },
+    { url = "https://files.pythonhosted.org/packages/71/43/6db4c02dcb488ad4e03bc86d70506c3d40a384ee73c9b5c93338eb1f3c23/xxhash-3.5.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:963be41bcd49f53af6d795f65c0da9b4cc518c0dd9c47145c98f61cb464f4839", size = 192216 },
+    { url = "https://files.pythonhosted.org/packages/22/6d/db4abec29e7a567455344433d095fdb39c97db6955bb4a2c432e486b4d28/xxhash-3.5.0-cp311-cp311-win32.whl", hash = "sha256:109b436096d0a2dd039c355fa3414160ec4d843dfecc64a14077332a00aeb7da", size = 30120 },
+    { url = "https://files.pythonhosted.org/packages/52/1c/fa3b61c0cf03e1da4767213672efe186b1dfa4fc901a4a694fb184a513d1/xxhash-3.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:b702f806693201ad6c0a05ddbbe4c8f359626d0b3305f766077d51388a6bac58", size = 30003 },
+    { url = "https://files.pythonhosted.org/packages/6b/8e/9e6fc572acf6e1cc7ccb01973c213f895cb8668a9d4c2b58a99350da14b7/xxhash-3.5.0-cp311-cp311-win_arm64.whl", hash = "sha256:c4dcb4120d0cc3cc448624147dba64e9021b278c63e34a38789b688fd0da9bf3", size = 26777 },
+    { url = "https://files.pythonhosted.org/packages/07/0e/1bfce2502c57d7e2e787600b31c83535af83746885aa1a5f153d8c8059d6/xxhash-3.5.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:14470ace8bd3b5d51318782cd94e6f94431974f16cb3b8dc15d52f3b69df8e00", size = 31969 },
+    { url = "https://files.pythonhosted.org/packages/3f/d6/8ca450d6fe5b71ce521b4e5db69622383d039e2b253e9b2f24f93265b52c/xxhash-3.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:59aa1203de1cb96dbeab595ded0ad0c0056bb2245ae11fac11c0ceea861382b9", size = 30787 },
+    { url = "https://files.pythonhosted.org/packages/5b/84/de7c89bc6ef63d750159086a6ada6416cc4349eab23f76ab870407178b93/xxhash-3.5.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:08424f6648526076e28fae6ea2806c0a7d504b9ef05ae61d196d571e5c879c84", size = 220959 },
+    { url = "https://files.pythonhosted.org/packages/fe/86/51258d3e8a8545ff26468c977101964c14d56a8a37f5835bc0082426c672/xxhash-3.5.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:61a1ff00674879725b194695e17f23d3248998b843eb5e933007ca743310f793", size = 200006 },
+    { url = "https://files.pythonhosted.org/packages/02/0a/96973bd325412feccf23cf3680fd2246aebf4b789122f938d5557c54a6b2/xxhash-3.5.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f2f2c61bee5844d41c3eb015ac652a0229e901074951ae48581d58bfb2ba01be", size = 428326 },
+    { url = "https://files.pythonhosted.org/packages/11/a7/81dba5010f7e733de88af9555725146fc133be97ce36533867f4c7e75066/xxhash-3.5.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9d32a592cac88d18cc09a89172e1c32d7f2a6e516c3dfde1b9adb90ab5df54a6", size = 194380 },
+    { url = "https://files.pythonhosted.org/packages/fb/7d/f29006ab398a173f4501c0e4977ba288f1c621d878ec217b4ff516810c04/xxhash-3.5.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:70dabf941dede727cca579e8c205e61121afc9b28516752fd65724be1355cc90", size = 207934 },
+    { url = "https://files.pythonhosted.org/packages/8a/6e/6e88b8f24612510e73d4d70d9b0c7dff62a2e78451b9f0d042a5462c8d03/xxhash-3.5.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e5d0ddaca65ecca9c10dcf01730165fd858533d0be84c75c327487c37a906a27", size = 216301 },
+    { url = "https://files.pythonhosted.org/packages/af/51/7862f4fa4b75a25c3b4163c8a873f070532fe5f2d3f9b3fc869c8337a398/xxhash-3.5.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:3e5b5e16c5a480fe5f59f56c30abdeba09ffd75da8d13f6b9b6fd224d0b4d0a2", size = 203351 },
+    { url = "https://files.pythonhosted.org/packages/22/61/8d6a40f288f791cf79ed5bb113159abf0c81d6efb86e734334f698eb4c59/xxhash-3.5.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:149b7914451eb154b3dfaa721315117ea1dac2cc55a01bfbd4df7c68c5dd683d", size = 210294 },
+    { url = "https://files.pythonhosted.org/packages/17/02/215c4698955762d45a8158117190261b2dbefe9ae7e5b906768c09d8bc74/xxhash-3.5.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:eade977f5c96c677035ff39c56ac74d851b1cca7d607ab3d8f23c6b859379cab", size = 414674 },
+    { url = "https://files.pythonhosted.org/packages/31/5c/b7a8db8a3237cff3d535261325d95de509f6a8ae439a5a7a4ffcff478189/xxhash-3.5.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:fa9f547bd98f5553d03160967866a71056a60960be00356a15ecc44efb40ba8e", size = 192022 },
+    { url = "https://files.pythonhosted.org/packages/78/e3/dd76659b2811b3fd06892a8beb850e1996b63e9235af5a86ea348f053e9e/xxhash-3.5.0-cp312-cp312-win32.whl", hash = "sha256:f7b58d1fd3551b8c80a971199543379be1cee3d0d409e1f6d8b01c1a2eebf1f8", size = 30170 },
+    { url = "https://files.pythonhosted.org/packages/d9/6b/1c443fe6cfeb4ad1dcf231cdec96eb94fb43d6498b4469ed8b51f8b59a37/xxhash-3.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:fa0cafd3a2af231b4e113fba24a65d7922af91aeb23774a8b78228e6cd785e3e", size = 30040 },
+    { url = "https://files.pythonhosted.org/packages/0f/eb/04405305f290173acc0350eba6d2f1a794b57925df0398861a20fbafa415/xxhash-3.5.0-cp312-cp312-win_arm64.whl", hash = "sha256:586886c7e89cb9828bcd8a5686b12e161368e0064d040e225e72607b43858ba2", size = 26796 },
+    { url = "https://files.pythonhosted.org/packages/c9/b8/e4b3ad92d249be5c83fa72916c9091b0965cb0faeff05d9a0a3870ae6bff/xxhash-3.5.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:37889a0d13b0b7d739cfc128b1c902f04e32de17b33d74b637ad42f1c55101f6", size = 31795 },
+    { url = "https://files.pythonhosted.org/packages/fc/d8/b3627a0aebfbfa4c12a41e22af3742cf08c8ea84f5cc3367b5de2d039cce/xxhash-3.5.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:97a662338797c660178e682f3bc180277b9569a59abfb5925e8620fba00b9fc5", size = 30792 },
+    { url = "https://files.pythonhosted.org/packages/c3/cc/762312960691da989c7cd0545cb120ba2a4148741c6ba458aa723c00a3f8/xxhash-3.5.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7f85e0108d51092bdda90672476c7d909c04ada6923c14ff9d913c4f7dc8a3bc", size = 220950 },
+    { url = "https://files.pythonhosted.org/packages/fe/e9/cc266f1042c3c13750e86a535496b58beb12bf8c50a915c336136f6168dc/xxhash-3.5.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cd2fd827b0ba763ac919440042302315c564fdb797294d86e8cdd4578e3bc7f3", size = 199980 },
+    { url = "https://files.pythonhosted.org/packages/bf/85/a836cd0dc5cc20376de26b346858d0ac9656f8f730998ca4324921a010b9/xxhash-3.5.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:82085c2abec437abebf457c1d12fccb30cc8b3774a0814872511f0f0562c768c", size = 428324 },
+    { url = "https://files.pythonhosted.org/packages/b4/0e/15c243775342ce840b9ba34aceace06a1148fa1630cd8ca269e3223987f5/xxhash-3.5.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:07fda5de378626e502b42b311b049848c2ef38784d0d67b6f30bb5008642f8eb", size = 194370 },
+    { url = "https://files.pythonhosted.org/packages/87/a1/b028bb02636dfdc190da01951d0703b3d904301ed0ef6094d948983bef0e/xxhash-3.5.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c279f0d2b34ef15f922b77966640ade58b4ccdfef1c4d94b20f2a364617a493f", size = 207911 },
+    { url = "https://files.pythonhosted.org/packages/80/d5/73c73b03fc0ac73dacf069fdf6036c9abad82de0a47549e9912c955ab449/xxhash-3.5.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:89e66ceed67b213dec5a773e2f7a9e8c58f64daeb38c7859d8815d2c89f39ad7", size = 216352 },
+    { url = "https://files.pythonhosted.org/packages/b6/2a/5043dba5ddbe35b4fe6ea0a111280ad9c3d4ba477dd0f2d1fe1129bda9d0/xxhash-3.5.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:bcd51708a633410737111e998ceb3b45d3dbc98c0931f743d9bb0a209033a326", size = 203410 },
+    { url = "https://files.pythonhosted.org/packages/a2/b2/9a8ded888b7b190aed75b484eb5c853ddd48aa2896e7b59bbfbce442f0a1/xxhash-3.5.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:3ff2c0a34eae7df88c868be53a8dd56fbdf592109e21d4bfa092a27b0bf4a7bf", size = 210322 },
+    { url = "https://files.pythonhosted.org/packages/98/62/440083fafbc917bf3e4b67c2ade621920dd905517e85631c10aac955c1d2/xxhash-3.5.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:4e28503dccc7d32e0b9817aa0cbfc1f45f563b2c995b7a66c4c8a0d232e840c7", size = 414725 },
+    { url = "https://files.pythonhosted.org/packages/75/db/009206f7076ad60a517e016bb0058381d96a007ce3f79fa91d3010f49cc2/xxhash-3.5.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a6c50017518329ed65a9e4829154626f008916d36295b6a3ba336e2458824c8c", size = 192070 },
+    { url = "https://files.pythonhosted.org/packages/1f/6d/c61e0668943a034abc3a569cdc5aeae37d686d9da7e39cf2ed621d533e36/xxhash-3.5.0-cp313-cp313-win32.whl", hash = "sha256:53a068fe70301ec30d868ece566ac90d873e3bb059cf83c32e76012c889b8637", size = 30172 },
+    { url = "https://files.pythonhosted.org/packages/96/14/8416dce965f35e3d24722cdf79361ae154fa23e2ab730e5323aa98d7919e/xxhash-3.5.0-cp313-cp313-win_amd64.whl", hash = "sha256:80babcc30e7a1a484eab952d76a4f4673ff601f54d5142c26826502740e70b43", size = 30041 },
+    { url = "https://files.pythonhosted.org/packages/27/ee/518b72faa2073f5aa8e3262408d284892cb79cf2754ba0c3a5870645ef73/xxhash-3.5.0-cp313-cp313-win_arm64.whl", hash = "sha256:4811336f1ce11cac89dcbd18f3a25c527c16311709a89313c3acaf771def2d4b", size = 26801 },
+    { url = "https://files.pythonhosted.org/packages/d4/f6/531dd6858adf8877675270b9d6989b6dacfd1c2d7135b17584fc29866df3/xxhash-3.5.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bfc8cdd7f33d57f0468b0614ae634cc38ab9202c6957a60e31d285a71ebe0301", size = 31971 },
+    { url = "https://files.pythonhosted.org/packages/7c/a8/b2a42b6c9ae46e233f474f3d307c2e7bca8d9817650babeca048d2ad01d6/xxhash-3.5.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e0c48b6300cd0b0106bf49169c3e0536408dfbeb1ccb53180068a18b03c662ab", size = 30801 },
+    { url = "https://files.pythonhosted.org/packages/b4/92/9ac297e3487818f429bcf369c1c6a097edf5b56ed6fc1feff4c1882e87ef/xxhash-3.5.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fe1a92cfbaa0a1253e339ccec42dbe6db262615e52df591b68726ab10338003f", size = 220644 },
+    { url = "https://files.pythonhosted.org/packages/86/48/c1426dd3c86fc4a52f983301867463472f6a9013fb32d15991e60c9919b6/xxhash-3.5.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:33513d6cc3ed3b559134fb307aae9bdd94d7e7c02907b37896a6c45ff9ce51bd", size = 200021 },
+    { url = "https://files.pythonhosted.org/packages/f3/de/0ab8c79993765c94fc0d0c1a22b454483c58a0161e1b562f58b654f47660/xxhash-3.5.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eefc37f6138f522e771ac6db71a6d4838ec7933939676f3753eafd7d3f4c40bc", size = 428217 },
+    { url = "https://files.pythonhosted.org/packages/b4/b4/332647451ed7d2c021294b7c1e9c144dbb5586b1fb214ad4f5a404642835/xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a606c8070ada8aa2a88e181773fa1ef17ba65ce5dd168b9d08038e2a61b33754", size = 193868 },
+    { url = "https://files.pythonhosted.org/packages/f4/1c/a42c0a6cac752f84f7b44a90d1a9fa9047cf70bdba5198a304fde7cc471f/xxhash-3.5.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:42eca420c8fa072cc1dd62597635d140e78e384a79bb4944f825fbef8bfeeef6", size = 207403 },
+    { url = "https://files.pythonhosted.org/packages/c4/d7/04e1b0daae9dc9b02c73c1664cc8aa527498c3f66ccbc586eeb25bbe9f14/xxhash-3.5.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:604253b2143e13218ff1ef0b59ce67f18b8bd1c4205d2ffda22b09b426386898", size = 215978 },
+    { url = "https://files.pythonhosted.org/packages/c4/f4/05e15e67505228fc19ee98a79e427b3a0b9695f5567cd66ced5d66389883/xxhash-3.5.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6e93a5ad22f434d7876665444a97e713a8f60b5b1a3521e8df11b98309bff833", size = 202416 },
+    { url = "https://files.pythonhosted.org/packages/94/fb/e9028d3645bba5412a09de13ee36df276a567e60bdb31d499dafa46d76ae/xxhash-3.5.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:7a46e1d6d2817ba8024de44c4fd79913a90e5f7265434cef97026215b7d30df6", size = 209853 },
+    { url = "https://files.pythonhosted.org/packages/02/2c/18c6a622429368274739372d2f86c8125413ec169025c7d8ffb051784bba/xxhash-3.5.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:30eb2efe6503c379b7ab99c81ba4a779748e3830241f032ab46bd182bf5873af", size = 413926 },
+    { url = "https://files.pythonhosted.org/packages/72/bb/5b55c391084a0321c3809632a018b9b657e59d5966289664f85a645942ac/xxhash-3.5.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c8aa771ff2c13dd9cda8166d685d7333d389fae30a4d2bb39d63ab5775de8606", size = 191156 },
+    { url = "https://files.pythonhosted.org/packages/86/2b/915049db13401792fec159f57e4f4a5ca7a9768e83ef71d6645b9d0cd749/xxhash-3.5.0-cp39-cp39-win32.whl", hash = "sha256:5ed9ebc46f24cf91034544b26b131241b699edbfc99ec5e7f8f3d02d6eb7fba4", size = 30122 },
+    { url = "https://files.pythonhosted.org/packages/d5/87/382ef7b24917d7cf4c540ee30f29b283bc87ac5893d2f89b23ea3cdf7d77/xxhash-3.5.0-cp39-cp39-win_amd64.whl", hash = "sha256:220f3f896c6b8d0316f63f16c077d52c412619e475f9372333474ee15133a558", size = 30021 },
+    { url = "https://files.pythonhosted.org/packages/e2/47/d06b24e2d9c3dcabccfd734d11b5bbebfdf59ceac2c61509d8205dd20ac6/xxhash-3.5.0-cp39-cp39-win_arm64.whl", hash = "sha256:a7b1d8315d9b5e9f89eb2933b73afae6ec9597a258d52190944437158b49d38e", size = 26780 },
+    { url = "https://files.pythonhosted.org/packages/ab/9a/233606bada5bd6f50b2b72c45de3d9868ad551e83893d2ac86dc7bb8553a/xxhash-3.5.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:2014c5b3ff15e64feecb6b713af12093f75b7926049e26a580e94dcad3c73d8c", size = 29732 },
+    { url = "https://files.pythonhosted.org/packages/0c/67/f75276ca39e2c6604e3bee6c84e9db8a56a4973fde9bf35989787cf6e8aa/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fab81ef75003eda96239a23eda4e4543cedc22e34c373edcaf744e721a163986", size = 36214 },
+    { url = "https://files.pythonhosted.org/packages/0f/f8/f6c61fd794229cc3848d144f73754a0c107854372d7261419dcbbd286299/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4e2febf914ace002132aa09169cc572e0d8959d0f305f93d5828c4836f9bc5a6", size = 32020 },
+    { url = "https://files.pythonhosted.org/packages/79/d3/c029c99801526f859e6b38d34ab87c08993bf3dcea34b11275775001638a/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5d3a10609c51da2a1c0ea0293fc3968ca0a18bd73838455b5bca3069d7f8e32b", size = 40515 },
+    { url = "https://files.pythonhosted.org/packages/62/e3/bef7b82c1997579c94de9ac5ea7626d01ae5858aa22bf4fcb38bf220cb3e/xxhash-3.5.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:5a74f23335b9689b66eb6dbe2a931a88fcd7a4c2cc4b1cb0edba8ce381c7a1da", size = 30064 },
+    { url = "https://files.pythonhosted.org/packages/c2/56/30d3df421814947f9d782b20c9b7e5e957f3791cbd89874578011daafcbd/xxhash-3.5.0-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:531af8845aaadcadf951b7e0c1345c6b9c68a990eeb74ff9acd8501a0ad6a1c9", size = 29734 },
+    { url = "https://files.pythonhosted.org/packages/82/dd/3c42a1f022ad0d82c852d3cb65493ebac03dcfa8c994465a5fb052b00e3c/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7ce379bcaa9fcc00f19affa7773084dd09f5b59947b3fb47a1ceb0179f91aaa1", size = 36216 },
+    { url = "https://files.pythonhosted.org/packages/b2/40/8f902ab3bebda228a9b4de69eba988280285a7f7f167b942bc20bb562df9/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1b2281d01723f076df3c8188f43f2472248a6b63118b036e641243656b1b0f", size = 32042 },
+    { url = "https://files.pythonhosted.org/packages/db/87/bd06beb8ccaa0e9e577c9b909a49cfa5c5cd2ca46034342d72dd9ce5bc56/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9c770750cc80e8694492244bca7251385188bc5597b6a39d98a9f30e8da984e0", size = 40516 },
+    { url = "https://files.pythonhosted.org/packages/bb/f8/505385e2fbd753ddcaafd5550eabe86f6232cbebabad3b2508d411b19153/xxhash-3.5.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:b150b8467852e1bd844387459aa6fbe11d7f38b56e901f9f3b3e6aba0d660240", size = 30108 },
+]
+
+[[package]]
+name = "yarl"
+version = "1.18.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "idna" },
+    { name = "multidict" },
+    { name = "propcache" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b7/9d/4b94a8e6d2b51b599516a5cb88e5bc99b4d8d4583e468057eaa29d5f0918/yarl-1.18.3.tar.gz", hash = "sha256:ac1801c45cbf77b6c99242eeff4fffb5e4e73a800b5c4ad4fc0be5def634d2e1", size = 181062 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d2/98/e005bc608765a8a5569f58e650961314873c8469c333616eb40bff19ae97/yarl-1.18.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7df647e8edd71f000a5208fe6ff8c382a1de8edfbccdbbfe649d263de07d8c34", size = 141458 },
+    { url = "https://files.pythonhosted.org/packages/df/5d/f8106b263b8ae8a866b46d9be869ac01f9b3fb7f2325f3ecb3df8003f796/yarl-1.18.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:c69697d3adff5aa4f874b19c0e4ed65180ceed6318ec856ebc423aa5850d84f7", size = 94365 },
+    { url = "https://files.pythonhosted.org/packages/56/3e/d8637ddb9ba69bf851f765a3ee288676f7cf64fb3be13760c18cbc9d10bd/yarl-1.18.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:602d98f2c2d929f8e697ed274fbadc09902c4025c5a9963bf4e9edfc3ab6f7ed", size = 92181 },
+    { url = "https://files.pythonhosted.org/packages/76/f9/d616a5c2daae281171de10fba41e1c0e2d8207166fc3547252f7d469b4e1/yarl-1.18.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c654d5207c78e0bd6d749f6dae1dcbbfde3403ad3a4b11f3c5544d9906969dde", size = 315349 },
+    { url = "https://files.pythonhosted.org/packages/bb/b4/3ea5e7b6f08f698b3769a06054783e434f6d59857181b5c4e145de83f59b/yarl-1.18.3-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5094d9206c64181d0f6e76ebd8fb2f8fe274950a63890ee9e0ebfd58bf9d787b", size = 330494 },
+    { url = "https://files.pythonhosted.org/packages/55/f1/e0fc810554877b1b67420568afff51b967baed5b53bcc983ab164eebf9c9/yarl-1.18.3-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:35098b24e0327fc4ebdc8ffe336cee0a87a700c24ffed13161af80124b7dc8e5", size = 326927 },
+    { url = "https://files.pythonhosted.org/packages/a9/42/b1753949b327b36f210899f2dd0a0947c0c74e42a32de3f8eb5c7d93edca/yarl-1.18.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3236da9272872443f81fedc389bace88408f64f89f75d1bdb2256069a8730ccc", size = 319703 },
+    { url = "https://files.pythonhosted.org/packages/f0/6d/e87c62dc9635daefb064b56f5c97df55a2e9cc947a2b3afd4fd2f3b841c7/yarl-1.18.3-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e2c08cc9b16f4f4bc522771d96734c7901e7ebef70c6c5c35dd0f10845270bcd", size = 310246 },
+    { url = "https://files.pythonhosted.org/packages/e3/ef/e2e8d1785cdcbd986f7622d7f0098205f3644546da7919c24b95790ec65a/yarl-1.18.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:80316a8bd5109320d38eef8833ccf5f89608c9107d02d2a7f985f98ed6876990", size = 319730 },
+    { url = "https://files.pythonhosted.org/packages/fc/15/8723e22345bc160dfde68c4b3ae8b236e868f9963c74015f1bc8a614101c/yarl-1.18.3-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:c1e1cc06da1491e6734f0ea1e6294ce00792193c463350626571c287c9a704db", size = 321681 },
+    { url = "https://files.pythonhosted.org/packages/86/09/bf764e974f1516efa0ae2801494a5951e959f1610dd41edbfc07e5e0f978/yarl-1.18.3-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fea09ca13323376a2fdfb353a5fa2e59f90cd18d7ca4eaa1fd31f0a8b4f91e62", size = 324812 },
+    { url = "https://files.pythonhosted.org/packages/f6/4c/20a0187e3b903c97d857cf0272d687c1b08b03438968ae8ffc50fe78b0d6/yarl-1.18.3-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:e3b9fd71836999aad54084906f8663dffcd2a7fb5cdafd6c37713b2e72be1760", size = 337011 },
+    { url = "https://files.pythonhosted.org/packages/c9/71/6244599a6e1cc4c9f73254a627234e0dad3883ece40cc33dce6265977461/yarl-1.18.3-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:757e81cae69244257d125ff31663249b3013b5dc0a8520d73694aed497fb195b", size = 338132 },
+    { url = "https://files.pythonhosted.org/packages/af/f5/e0c3efaf74566c4b4a41cb76d27097df424052a064216beccae8d303c90f/yarl-1.18.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b1771de9944d875f1b98a745bc547e684b863abf8f8287da8466cf470ef52690", size = 331849 },
+    { url = "https://files.pythonhosted.org/packages/8a/b8/3d16209c2014c2f98a8f658850a57b716efb97930aebf1ca0d9325933731/yarl-1.18.3-cp310-cp310-win32.whl", hash = "sha256:8874027a53e3aea659a6d62751800cf6e63314c160fd607489ba5c2edd753cf6", size = 84309 },
+    { url = "https://files.pythonhosted.org/packages/fd/b7/2e9a5b18eb0fe24c3a0e8bae994e812ed9852ab4fd067c0107fadde0d5f0/yarl-1.18.3-cp310-cp310-win_amd64.whl", hash = "sha256:93b2e109287f93db79210f86deb6b9bbb81ac32fc97236b16f7433db7fc437d8", size = 90484 },
+    { url = "https://files.pythonhosted.org/packages/40/93/282b5f4898d8e8efaf0790ba6d10e2245d2c9f30e199d1a85cae9356098c/yarl-1.18.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:8503ad47387b8ebd39cbbbdf0bf113e17330ffd339ba1144074da24c545f0069", size = 141555 },
+    { url = "https://files.pythonhosted.org/packages/6d/9c/0a49af78df099c283ca3444560f10718fadb8a18dc8b3edf8c7bd9fd7d89/yarl-1.18.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:02ddb6756f8f4517a2d5e99d8b2f272488e18dd0bfbc802f31c16c6c20f22193", size = 94351 },
+    { url = "https://files.pythonhosted.org/packages/5a/a1/205ab51e148fdcedad189ca8dd587794c6f119882437d04c33c01a75dece/yarl-1.18.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:67a283dd2882ac98cc6318384f565bffc751ab564605959df4752d42483ad889", size = 92286 },
+    { url = "https://files.pythonhosted.org/packages/ed/fe/88b690b30f3f59275fb674f5f93ddd4a3ae796c2b62e5bb9ece8a4914b83/yarl-1.18.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d980e0325b6eddc81331d3f4551e2a333999fb176fd153e075c6d1c2530aa8a8", size = 340649 },
+    { url = "https://files.pythonhosted.org/packages/07/eb/3b65499b568e01f36e847cebdc8d7ccb51fff716dbda1ae83c3cbb8ca1c9/yarl-1.18.3-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b643562c12680b01e17239be267bc306bbc6aac1f34f6444d1bded0c5ce438ca", size = 356623 },
+    { url = "https://files.pythonhosted.org/packages/33/46/f559dc184280b745fc76ec6b1954de2c55595f0ec0a7614238b9ebf69618/yarl-1.18.3-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c017a3b6df3a1bd45b9fa49a0f54005e53fbcad16633870104b66fa1a30a29d8", size = 354007 },
+    { url = "https://files.pythonhosted.org/packages/af/ba/1865d85212351ad160f19fb99808acf23aab9a0f8ff31c8c9f1b4d671fc9/yarl-1.18.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:75674776d96d7b851b6498f17824ba17849d790a44d282929c42dbb77d4f17ae", size = 344145 },
+    { url = "https://files.pythonhosted.org/packages/94/cb/5c3e975d77755d7b3d5193e92056b19d83752ea2da7ab394e22260a7b824/yarl-1.18.3-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ccaa3a4b521b780a7e771cc336a2dba389a0861592bbce09a476190bb0c8b4b3", size = 336133 },
+    { url = "https://files.pythonhosted.org/packages/19/89/b77d3fd249ab52a5c40859815765d35c91425b6bb82e7427ab2f78f5ff55/yarl-1.18.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:2d06d3005e668744e11ed80812e61efd77d70bb7f03e33c1598c301eea20efbb", size = 347967 },
+    { url = "https://files.pythonhosted.org/packages/35/bd/f6b7630ba2cc06c319c3235634c582a6ab014d52311e7d7c22f9518189b5/yarl-1.18.3-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:9d41beda9dc97ca9ab0b9888cb71f7539124bc05df02c0cff6e5acc5a19dcc6e", size = 346397 },
+    { url = "https://files.pythonhosted.org/packages/18/1a/0b4e367d5a72d1f095318344848e93ea70da728118221f84f1bf6c1e39e7/yarl-1.18.3-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:ba23302c0c61a9999784e73809427c9dbedd79f66a13d84ad1b1943802eaaf59", size = 350206 },
+    { url = "https://files.pythonhosted.org/packages/b5/cf/320fff4367341fb77809a2d8d7fe75b5d323a8e1b35710aafe41fdbf327b/yarl-1.18.3-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:6748dbf9bfa5ba1afcc7556b71cda0d7ce5f24768043a02a58846e4a443d808d", size = 362089 },
+    { url = "https://files.pythonhosted.org/packages/57/cf/aadba261d8b920253204085268bad5e8cdd86b50162fcb1b10c10834885a/yarl-1.18.3-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:0b0cad37311123211dc91eadcb322ef4d4a66008d3e1bdc404808992260e1a0e", size = 366267 },
+    { url = "https://files.pythonhosted.org/packages/54/58/fb4cadd81acdee6dafe14abeb258f876e4dd410518099ae9a35c88d8097c/yarl-1.18.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0fb2171a4486bb075316ee754c6d8382ea6eb8b399d4ec62fde2b591f879778a", size = 359141 },
+    { url = "https://files.pythonhosted.org/packages/9a/7a/4c571597589da4cd5c14ed2a0b17ac56ec9ee7ee615013f74653169e702d/yarl-1.18.3-cp311-cp311-win32.whl", hash = "sha256:61b1a825a13bef4a5f10b1885245377d3cd0bf87cba068e1d9a88c2ae36880e1", size = 84402 },
+    { url = "https://files.pythonhosted.org/packages/ae/7b/8600250b3d89b625f1121d897062f629883c2f45339623b69b1747ec65fa/yarl-1.18.3-cp311-cp311-win_amd64.whl", hash = "sha256:b9d60031cf568c627d028239693fd718025719c02c9f55df0a53e587aab951b5", size = 91030 },
+    { url = "https://files.pythonhosted.org/packages/33/85/bd2e2729752ff4c77338e0102914897512e92496375e079ce0150a6dc306/yarl-1.18.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:1dd4bdd05407ced96fed3d7f25dbbf88d2ffb045a0db60dbc247f5b3c5c25d50", size = 142644 },
+    { url = "https://files.pythonhosted.org/packages/ff/74/1178322cc0f10288d7eefa6e4a85d8d2e28187ccab13d5b844e8b5d7c88d/yarl-1.18.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:7c33dd1931a95e5d9a772d0ac5e44cac8957eaf58e3c8da8c1414de7dd27c576", size = 94962 },
+    { url = "https://files.pythonhosted.org/packages/be/75/79c6acc0261e2c2ae8a1c41cf12265e91628c8c58ae91f5ff59e29c0787f/yarl-1.18.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:25b411eddcfd56a2f0cd6a384e9f4f7aa3efee14b188de13048c25b5e91f1640", size = 92795 },
+    { url = "https://files.pythonhosted.org/packages/6b/32/927b2d67a412c31199e83fefdce6e645247b4fb164aa1ecb35a0f9eb2058/yarl-1.18.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:436c4fc0a4d66b2badc6c5fc5ef4e47bb10e4fd9bf0c79524ac719a01f3607c2", size = 332368 },
+    { url = "https://files.pythonhosted.org/packages/19/e5/859fca07169d6eceeaa4fde1997c91d8abde4e9a7c018e371640c2da2b71/yarl-1.18.3-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e35ef8683211db69ffe129a25d5634319a677570ab6b2eba4afa860f54eeaf75", size = 342314 },
+    { url = "https://files.pythonhosted.org/packages/08/75/76b63ccd91c9e03ab213ef27ae6add2e3400e77e5cdddf8ed2dbc36e3f21/yarl-1.18.3-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:84b2deecba4a3f1a398df819151eb72d29bfeb3b69abb145a00ddc8d30094512", size = 341987 },
+    { url = "https://files.pythonhosted.org/packages/1a/e1/a097d5755d3ea8479a42856f51d97eeff7a3a7160593332d98f2709b3580/yarl-1.18.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:00e5a1fea0fd4f5bfa7440a47eff01d9822a65b4488f7cff83155a0f31a2ecba", size = 336914 },
+    { url = "https://files.pythonhosted.org/packages/0b/42/e1b4d0e396b7987feceebe565286c27bc085bf07d61a59508cdaf2d45e63/yarl-1.18.3-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d0e883008013c0e4aef84dcfe2a0b172c4d23c2669412cf5b3371003941f72bb", size = 325765 },
+    { url = "https://files.pythonhosted.org/packages/7e/18/03a5834ccc9177f97ca1bbb245b93c13e58e8225276f01eedc4cc98ab820/yarl-1.18.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5a3f356548e34a70b0172d8890006c37be92995f62d95a07b4a42e90fba54272", size = 344444 },
+    { url = "https://files.pythonhosted.org/packages/c8/03/a713633bdde0640b0472aa197b5b86e90fbc4c5bc05b727b714cd8a40e6d/yarl-1.18.3-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:ccd17349166b1bee6e529b4add61727d3f55edb7babbe4069b5764c9587a8cc6", size = 340760 },
+    { url = "https://files.pythonhosted.org/packages/eb/99/f6567e3f3bbad8fd101886ea0276c68ecb86a2b58be0f64077396cd4b95e/yarl-1.18.3-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:b958ddd075ddba5b09bb0be8a6d9906d2ce933aee81100db289badbeb966f54e", size = 346484 },
+    { url = "https://files.pythonhosted.org/packages/8e/a9/84717c896b2fc6cb15bd4eecd64e34a2f0a9fd6669e69170c73a8b46795a/yarl-1.18.3-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:c7d79f7d9aabd6011004e33b22bc13056a3e3fb54794d138af57f5ee9d9032cb", size = 359864 },
+    { url = "https://files.pythonhosted.org/packages/1e/2e/d0f5f1bef7ee93ed17e739ec8dbcb47794af891f7d165fa6014517b48169/yarl-1.18.3-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:4891ed92157e5430874dad17b15eb1fda57627710756c27422200c52d8a4e393", size = 364537 },
+    { url = "https://files.pythonhosted.org/packages/97/8a/568d07c5d4964da5b02621a517532adb8ec5ba181ad1687191fffeda0ab6/yarl-1.18.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ce1af883b94304f493698b00d0f006d56aea98aeb49d75ec7d98cd4a777e9285", size = 357861 },
+    { url = "https://files.pythonhosted.org/packages/7d/e3/924c3f64b6b3077889df9a1ece1ed8947e7b61b0a933f2ec93041990a677/yarl-1.18.3-cp312-cp312-win32.whl", hash = "sha256:f91c4803173928a25e1a55b943c81f55b8872f0018be83e3ad4938adffb77dd2", size = 84097 },
+    { url = "https://files.pythonhosted.org/packages/34/45/0e055320daaabfc169b21ff6174567b2c910c45617b0d79c68d7ab349b02/yarl-1.18.3-cp312-cp312-win_amd64.whl", hash = "sha256:7e2ee16578af3b52ac2f334c3b1f92262f47e02cc6193c598502bd46f5cd1477", size = 90399 },
+    { url = "https://files.pythonhosted.org/packages/30/c7/c790513d5328a8390be8f47be5d52e141f78b66c6c48f48d241ca6bd5265/yarl-1.18.3-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:90adb47ad432332d4f0bc28f83a5963f426ce9a1a8809f5e584e704b82685dcb", size = 140789 },
+    { url = "https://files.pythonhosted.org/packages/30/aa/a2f84e93554a578463e2edaaf2300faa61c8701f0898725842c704ba5444/yarl-1.18.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:913829534200eb0f789d45349e55203a091f45c37a2674678744ae52fae23efa", size = 94144 },
+    { url = "https://files.pythonhosted.org/packages/c6/fc/d68d8f83714b221a85ce7866832cba36d7c04a68fa6a960b908c2c84f325/yarl-1.18.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:ef9f7768395923c3039055c14334ba4d926f3baf7b776c923c93d80195624782", size = 91974 },
+    { url = "https://files.pythonhosted.org/packages/56/4e/d2563d8323a7e9a414b5b25341b3942af5902a2263d36d20fb17c40411e2/yarl-1.18.3-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:88a19f62ff30117e706ebc9090b8ecc79aeb77d0b1f5ec10d2d27a12bc9f66d0", size = 333587 },
+    { url = "https://files.pythonhosted.org/packages/25/c9/cfec0bc0cac8d054be223e9f2c7909d3e8442a856af9dbce7e3442a8ec8d/yarl-1.18.3-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e17c9361d46a4d5addf777c6dd5eab0715a7684c2f11b88c67ac37edfba6c482", size = 344386 },
+    { url = "https://files.pythonhosted.org/packages/ab/5d/4c532190113b25f1364d25f4c319322e86232d69175b91f27e3ebc2caf9a/yarl-1.18.3-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1a74a13a4c857a84a845505fd2d68e54826a2cd01935a96efb1e9d86c728e186", size = 345421 },
+    { url = "https://files.pythonhosted.org/packages/23/d1/6cdd1632da013aa6ba18cee4d750d953104a5e7aac44e249d9410a972bf5/yarl-1.18.3-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41f7ce59d6ee7741af71d82020346af364949314ed3d87553763a2df1829cc58", size = 339384 },
+    { url = "https://files.pythonhosted.org/packages/9a/c4/6b3c39bec352e441bd30f432cda6ba51681ab19bb8abe023f0d19777aad1/yarl-1.18.3-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f52a265001d830bc425f82ca9eabda94a64a4d753b07d623a9f2863fde532b53", size = 326689 },
+    { url = "https://files.pythonhosted.org/packages/23/30/07fb088f2eefdc0aa4fc1af4e3ca4eb1a3aadd1ce7d866d74c0f124e6a85/yarl-1.18.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:82123d0c954dc58db301f5021a01854a85bf1f3bb7d12ae0c01afc414a882ca2", size = 345453 },
+    { url = "https://files.pythonhosted.org/packages/63/09/d54befb48f9cd8eec43797f624ec37783a0266855f4930a91e3d5c7717f8/yarl-1.18.3-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:2ec9bbba33b2d00999af4631a3397d1fd78290c48e2a3e52d8dd72db3a067ac8", size = 341872 },
+    { url = "https://files.pythonhosted.org/packages/91/26/fd0ef9bf29dd906a84b59f0cd1281e65b0c3e08c6aa94b57f7d11f593518/yarl-1.18.3-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:fbd6748e8ab9b41171bb95c6142faf068f5ef1511935a0aa07025438dd9a9bc1", size = 347497 },
+    { url = "https://files.pythonhosted.org/packages/d9/b5/14ac7a256d0511b2ac168d50d4b7d744aea1c1aa20c79f620d1059aab8b2/yarl-1.18.3-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:877d209b6aebeb5b16c42cbb377f5f94d9e556626b1bfff66d7b0d115be88d0a", size = 359981 },
+    { url = "https://files.pythonhosted.org/packages/ca/b3/d493221ad5cbd18bc07e642894030437e405e1413c4236dd5db6e46bcec9/yarl-1.18.3-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:b464c4ab4bfcb41e3bfd3f1c26600d038376c2de3297760dfe064d2cb7ea8e10", size = 366229 },
+    { url = "https://files.pythonhosted.org/packages/04/56/6a3e2a5d9152c56c346df9b8fb8edd2c8888b1e03f96324d457e5cf06d34/yarl-1.18.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:8d39d351e7faf01483cc7ff7c0213c412e38e5a340238826be7e0e4da450fdc8", size = 360383 },
+    { url = "https://files.pythonhosted.org/packages/fd/b7/4b3c7c7913a278d445cc6284e59b2e62fa25e72758f888b7a7a39eb8423f/yarl-1.18.3-cp313-cp313-win32.whl", hash = "sha256:61ee62ead9b68b9123ec24bc866cbef297dd266175d53296e2db5e7f797f902d", size = 310152 },
+    { url = "https://files.pythonhosted.org/packages/f5/d5/688db678e987c3e0fb17867970700b92603cadf36c56e5fb08f23e822a0c/yarl-1.18.3-cp313-cp313-win_amd64.whl", hash = "sha256:578e281c393af575879990861823ef19d66e2b1d0098414855dd367e234f5b3c", size = 315723 },
+    { url = "https://files.pythonhosted.org/packages/6a/3b/fec4b08f5e88f68e56ee698a59284a73704df2e0e0b5bdf6536c86e76c76/yarl-1.18.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:61e5e68cb65ac8f547f6b5ef933f510134a6bf31bb178be428994b0cb46c2a04", size = 142780 },
+    { url = "https://files.pythonhosted.org/packages/ed/85/796b0d6a22d536ec8e14bdbb86519250bad980cec450b6e299b1c2a9079e/yarl-1.18.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:fe57328fbc1bfd0bd0514470ac692630f3901c0ee39052ae47acd1d90a436719", size = 94981 },
+    { url = "https://files.pythonhosted.org/packages/ee/0e/a830fd2238f7a29050f6dd0de748b3d6f33a7dbb67dbbc081a970b2bbbeb/yarl-1.18.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a440a2a624683108a1b454705ecd7afc1c3438a08e890a1513d468671d90a04e", size = 92789 },
+    { url = "https://files.pythonhosted.org/packages/0f/4f/438c9fd668954779e48f08c0688ee25e0673380a21bb1e8ccc56de5b55d7/yarl-1.18.3-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:09c7907c8548bcd6ab860e5f513e727c53b4a714f459b084f6580b49fa1b9cee", size = 317327 },
+    { url = "https://files.pythonhosted.org/packages/bd/79/a78066f06179b4ed4581186c136c12fcfb928c475cbeb23743e71a991935/yarl-1.18.3-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b4f6450109834af88cb4cc5ecddfc5380ebb9c228695afc11915a0bf82116789", size = 336999 },
+    { url = "https://files.pythonhosted.org/packages/55/02/527963cf65f34a06aed1e766ff9a3b3e7d0eaa1c90736b2948a62e528e1d/yarl-1.18.3-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a9ca04806f3be0ac6d558fffc2fdf8fcef767e0489d2684a21912cc4ed0cd1b8", size = 331693 },
+    { url = "https://files.pythonhosted.org/packages/a2/2a/167447ae39252ba624b98b8c13c0ba35994d40d9110e8a724c83dbbb5822/yarl-1.18.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:77a6e85b90a7641d2e07184df5557132a337f136250caafc9ccaa4a2a998ca2c", size = 321473 },
+    { url = "https://files.pythonhosted.org/packages/55/03/07955fabb20082373be311c91fd78abe458bc7ff9069d34385e8bddad20e/yarl-1.18.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6333c5a377c8e2f5fae35e7b8f145c617b02c939d04110c76f29ee3676b5f9a5", size = 313571 },
+    { url = "https://files.pythonhosted.org/packages/95/e2/67c8d3ec58a8cd8ddb1d63bd06eb7e7b91c9f148707a3eeb5a7ed87df0ef/yarl-1.18.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:0b3c92fa08759dbf12b3a59579a4096ba9af8dd344d9a813fc7f5070d86bbab1", size = 325004 },
+    { url = "https://files.pythonhosted.org/packages/06/43/51ceb3e427368fe6ccd9eccd162be227fd082523e02bad1fd3063daf68da/yarl-1.18.3-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:4ac515b860c36becb81bb84b667466885096b5fc85596948548b667da3bf9f24", size = 322677 },
+    { url = "https://files.pythonhosted.org/packages/e4/0e/7ef286bfb23267739a703f7b967a858e2128c10bea898de8fa027e962521/yarl-1.18.3-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:045b8482ce9483ada4f3f23b3774f4e1bf4f23a2d5c912ed5170f68efb053318", size = 332806 },
+    { url = "https://files.pythonhosted.org/packages/c8/94/2d1f060f4bfa47c8bd0bcb652bfe71fba881564bcac06ebb6d8ced9ac3bc/yarl-1.18.3-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:a4bb030cf46a434ec0225bddbebd4b89e6471814ca851abb8696170adb163985", size = 339919 },
+    { url = "https://files.pythonhosted.org/packages/8e/8d/73b5f9a6ab69acddf1ca1d5e7bc92f50b69124512e6c26b36844531d7f23/yarl-1.18.3-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:54d6921f07555713b9300bee9c50fb46e57e2e639027089b1d795ecd9f7fa910", size = 340960 },
+    { url = "https://files.pythonhosted.org/packages/41/13/ce6bc32be4476b60f4f8694831f49590884b2c975afcffc8d533bf2be7ec/yarl-1.18.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1d407181cfa6e70077df3377938c08012d18893f9f20e92f7d2f314a437c30b1", size = 336592 },
+    { url = "https://files.pythonhosted.org/packages/81/d5/6e0460292d6299ac3919945f912b16b104f4e81ab20bf53e0872a1296daf/yarl-1.18.3-cp39-cp39-win32.whl", hash = "sha256:ac36703a585e0929b032fbaab0707b75dc12703766d0b53486eabd5139ebadd5", size = 84833 },
+    { url = "https://files.pythonhosted.org/packages/b2/fc/a8aef69156ad5508165d8ae956736d55c3a68890610834bd985540966008/yarl-1.18.3-cp39-cp39-win_amd64.whl", hash = "sha256:ba87babd629f8af77f557b61e49e7c7cac36f22f871156b91e10a6e9d4f829e9", size = 90968 },
+    { url = "https://files.pythonhosted.org/packages/f5/4b/a06e0ec3d155924f77835ed2d167ebd3b211a7b0853da1cf8d8414d784ef/yarl-1.18.3-py3-none-any.whl", hash = "sha256:b57f4f58099328dfb26c6a771d09fb20dbbae81d20cfb66141251ea063bd101b", size = 45109 },
+]
+
+[[package]]
+name = "zipp"
+version = "3.21.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3f/50/bad581df71744867e9468ebd0bcd6505de3b275e06f202c2cb016e3ff56f/zipp-3.21.0.tar.gz", hash = "sha256:2c9958f6430a2040341a52eb608ed6dd93ef4392e02ffe219417c1b28b5dd1f4", size = 24545 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b7/1a/7e4798e9339adc931158c9d69ecc34f5e6791489d469f5e50ec15e35f458/zipp-3.21.0-py3-none-any.whl", hash = "sha256:ac1bbe05fd2991f160ebce24ffbac5f6d11d83dc90891255885223d42b3cd931", size = 9630 },
+]

From 9fc6df9fcd6d2d7675f02b23ccfd96e5a0f6985e Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Wed, 19 Feb 2025 07:57:53 -0800
Subject: [PATCH 163/505] Fix documentation error #1426

---
 docs/reference/models/openai.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
index 638107568..0777ad8b7 100644
--- a/docs/reference/models/openai.md
+++ b/docs/reference/models/openai.md
@@ -131,7 +131,7 @@ class Person(BaseModel):
     last_name: str
     age: int
 
-generate.json(model, Person)
+generator = generate.json(model, Person)
 generator("current indian prime minister on january 1st 2023")
 # Person(first_name='Narendra', last_name='Modi', age=72)
 

From 7085885d975d0bb0d7b29927df41260ad5cfa48d Mon Sep 17 00:00:00 2001
From: Pedro Larroy <plarroy@nvidia.com>
Date: Tue, 18 Feb 2025 16:45:29 -0800
Subject: [PATCH 164/505] Make cache dir more robust inside docker containers
 and try to use XDG_CACHE_HOME if available.

---
 outlines/caching.py | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/outlines/caching.py b/outlines/caching.py
index 6fdda6214..04c655943 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -2,6 +2,7 @@
 import contextlib
 import functools
 import os
+import tempfile
 from typing import Callable, Optional
 
 import cloudpickle
@@ -50,8 +51,17 @@ def get_cache():
     """
     from outlines._version import __version__ as outlines_version  # type: ignore
 
-    home_dir = os.path.expanduser("~")
-    cache_dir = os.environ.get("OUTLINES_CACHE_DIR", f"{home_dir}/.cache/outlines")
+    xdg_cache_home = os.environ.get("XDG_CACHE_HOME")
+    if xdg_cache_home:
+        cache_dir = os.path.join(xdg_cache_home, ".cache", "outlines")
+    else:
+        home_dir = os.path.normpath(os.path.expanduser("~"))
+        if home_dir != "/":
+            cache_dir = os.environ.get("OUTLINES_CACHE_DIR", f"{home_dir}/.cache/outlines")
+        else:
+            tempdir = tempfile.gettempdir()
+            cache_dir = os.path.join(tempdir, ".cache", "outlines")
+
     memory = Cache(
         cache_dir,
         eviction_policy="none",

From d8ccb844e55b653089458f22bc15474e0e5546c2 Mon Sep 17 00:00:00 2001
From: Pedro Larroy <plarroy@nvidia.com>
Date: Tue, 18 Feb 2025 16:50:25 -0800
Subject: [PATCH 165/505] Make cache dir more robust inside docker containers
 and try to use XDG_CACHE_HOME if available.

---
 outlines/caching.py | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/outlines/caching.py b/outlines/caching.py
index 04c655943..0831c40bb 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -51,16 +51,20 @@ def get_cache():
     """
     from outlines._version import __version__ as outlines_version  # type: ignore
 
+    outlines_cache_dir = os.environ.get('OUTLINES_CACHE_DIR')
     xdg_cache_home = os.environ.get("XDG_CACHE_HOME")
-    if xdg_cache_home:
+    home_dir = os.path.normpath(os.path.expanduser("~"))
+    if outlines_cache_dir:
+        # OUTLINES_CACHE_DIR takes precendence
+        cache_dir = outlines_cache_dir
+    elif xdg_cache_home:
         cache_dir = os.path.join(xdg_cache_home, ".cache", "outlines")
+    elif home_dir != "/":
+        cache_dir = os.path.join(home_dir, ".cache", "outlines")
     else:
-        home_dir = os.path.normpath(os.path.expanduser("~"))
-        if home_dir != "/":
-            cache_dir = os.environ.get("OUTLINES_CACHE_DIR", f"{home_dir}/.cache/outlines")
-        else:
-            tempdir = tempfile.gettempdir()
-            cache_dir = os.path.join(tempdir, ".cache", "outlines")
+        # home_dir may be / inside a docker container without existing user
+        tempdir = tempfile.gettempdir()
+        cache_dir = os.path.join(tempdir, ".cache", "outlines")
 
     memory = Cache(
         cache_dir,

From da1b0db5658260f92d8b5529c9a600933e36f4bc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 26 Jan 2025 19:47:29 +0100
Subject: [PATCH 166/505] Add regex DSL and re-organize custom types

---
 docs/reference/functions.md       |   1 -
 docs/reference/regex_dsl.md       | 316 ++++++++++++++++++++++++
 mkdocs.yml                        |   1 +
 outlines/fsm/types.py             |  40 ++-
 outlines/generate/format.py       |   1 +
 outlines/generate/regex.py        |  11 +-
 outlines/types/__init__.py        |  43 +++-
 outlines/types/airports.py        |   2 -
 outlines/types/dsl.py             | 395 ++++++++++++++++++++++++++++++
 outlines/types/email.py           |  11 -
 outlines/types/isbn.py            |  12 -
 outlines/types/locale/__init__.py |   1 +
 outlines/types/locale/us.py       |   4 +
 outlines/types/locales.py         |  21 --
 outlines/types/phone_numbers.py   |  16 --
 outlines/types/zip_codes.py       |  13 -
 pyproject.toml                    |   1 +
 tests/fsm/test_types.py           |  31 +--
 tests/test_types.py               | 103 --------
 tests/types/test_custom_types.py  | 111 +++++++++
 tests/types/test_dsl.py           | 256 +++++++++++++++++++
 tests/types/test_to_regex.py      |  71 ++++++
 22 files changed, 1233 insertions(+), 228 deletions(-)
 delete mode 100644 docs/reference/functions.md
 create mode 100644 docs/reference/regex_dsl.md
 create mode 100644 outlines/types/dsl.py
 delete mode 100644 outlines/types/email.py
 delete mode 100644 outlines/types/isbn.py
 create mode 100644 outlines/types/locale/__init__.py
 create mode 100644 outlines/types/locale/us.py
 delete mode 100644 outlines/types/locales.py
 delete mode 100644 outlines/types/phone_numbers.py
 delete mode 100644 outlines/types/zip_codes.py
 delete mode 100644 tests/test_types.py
 create mode 100644 tests/types/test_custom_types.py
 create mode 100644 tests/types/test_dsl.py
 create mode 100644 tests/types/test_to_regex.py

diff --git a/docs/reference/functions.md b/docs/reference/functions.md
deleted file mode 100644
index d29213a89..000000000
--- a/docs/reference/functions.md
+++ /dev/null
@@ -1 +0,0 @@
-# Outlines functions
diff --git a/docs/reference/regex_dsl.md b/docs/reference/regex_dsl.md
new file mode 100644
index 000000000..aa04303cf
--- /dev/null
+++ b/docs/reference/regex_dsl.md
@@ -0,0 +1,316 @@
+# DSL to express constraints
+
+This library provides a Domain-Specific Language (DSL) to construct regular expressions in a more intuitive and modular way. It allows you to create complex regexes using simple building blocks that represent literal strings, patterns, and various quantifiers. Additionally, these custom regex types can be used directly as types in [Pydantic](https://pydantic-docs.helpmanual.io/) schemas to enforce pattern constraints during text generation.
+
+---
+
+## Why Use This DSL?
+
+1. **Modularity & Readability**: Instead of writing cryptic regular expression strings, you compose a regex as a tree of objects.
+2. **Enhanced Debugging**: Each expression can be visualized as an ASCII tree, making it easier to understand and debug complex regexes.
+3. **Pydantic Integration**: Use your DSL-defined regex as types in Pydantic models. The DSL seamlessly converts to JSON Schema with proper pattern constraints.
+4. **Extensibility**: Easily add or modify quantifiers and other regex components by extending the provided classes.
+
+---
+
+## Building Blocks
+
+
+Every regex component in this DSL is a **Term**. Here are two primary types:
+
+- **`String`**: Represents a literal string. It escapes the characters that have a special meaning in regular expressions.
+- **`Regex`**: Represents an existing regex pattern string.
+
+```python
+from outlines.types import String, Regex
+
+# A literal string "hello"
+literal = String("hello")   # Internally represents "hello"
+
+# A regex pattern to match one or more digits
+digit = Regex(r"[0-9]+")     # Internally represents the pattern [0-9]+
+
+# Converting to standard regex strings:
+from outlines.types.dsl import to_regex
+
+print(to_regex(literal))  # Output: hello
+print(to_regex(digit))    # Output: [0-9]+
+```
+
+---
+
+## Early Introduction to Quantifiers & Operators
+
+The DSL supports common regex quantifiers as methods on every `Term`. These methods allow you to specify how many times a pattern should be matched. They include:
+
+- **`times(count)`**: Matches the term exactly `count` times.
+- **`optional()`**: Matches the term zero or one time.
+- **`one_or_more()`**: Matches the term one or more times (Kleene Plus).
+- **`zero_or_more()`**: Matches the term zero or more times (Kleene Star).
+- **`repeat(min_count, max_count)`**: Matches the term between `min_count` and `max_count` times (or open-ended if one value is omitted).
+
+Let’s see these quantifiers side by side with examples.
+
+### Quantifiers in Action
+
+#### `times(count)`
+
+This method restricts the term to appear exactly `count` times.
+
+```python
+# Example: exactly 5 digits
+five_digits = Regex(r"\d").times(5)
+print(to_regex(five_digits))  # Output: (\d){5}
+```
+
+You can also use the `times` function:
+
+```python
+from outlines.types import times
+
+# Example: exactly 5 digits
+five_digits = times(Regex(r"\d"), 5)
+print(to_regex(five_digits))  # Output: (\d){5}
+```
+
+#### `optional()`
+
+The `optional()` method makes a term optional, meaning it may occur zero or one time.
+
+```python
+# Example: an optional "s" at the end of a word
+maybe_s = String("s").optional()
+print(to_regex(maybe_s))  # Output: (s)?
+```
+
+You can also use the `optional` function, the string will automatically be converted to a `String` object:
+
+```python
+from outlines.types import optional
+
+# Example: an optional "s" at the end of a word
+maybe_s = optional("s")
+print(to_regex(maybe_s))  # Output: (s)?
+```
+
+#### `one_or_more()`
+
+This method indicates that the term must appear at least once.
+
+```python
+# Example: one or more alphabetic characters
+letters = Regex(r"[A-Za-z]").one_or_more()
+print(to_regex(letters))  # Output: ([A-Za-z])+
+```
+
+You can also use the `one_or_more` function:
+
+```python
+from outlines.types import one_or_more
+
+# Example: one or more alphabetic characters
+letters = one_or_more(Regex(r"[A-Za-z]"))
+print(to_regex(letters))  # Output: ([A-Za-z])+
+
+```
+
+#### `zero_or_more()`
+
+This method means that the term can occur zero or more times.
+
+```python
+# Example: zero or more spaces
+spaces = String(" ").zero_or_more()
+print(to_regex(spaces))  # Output: ( )*
+```
+
+You can also use the `zero_or_more` function, the string will automatically be converted to a `String` instance:
+
+```python
+from outlines.types import zero_or_more
+
+# Example: zero or more spaces
+spaces = zero_or_more(" ")
+print(to_regex(spaces))  # Output: ( )*
+```
+
+#### `repeat(min_count, max_count)`
+
+The `repeat` method provides flexibility to set a lower and/or upper bound on the number of occurrences.
+
+```python
+# Example: Between 2 and 4 word characters
+word_chars = Regex(r"\w").repeat(2, 4)
+print(to_regex(word_chars))  # Output: (\w){2,4}
+
+# Example: At least 3 digits (min specified, max left open)
+at_least_three = Regex(r"\d").repeat(3, None)
+print(to_regex(at_least_three))  # Output: (\d){3,}
+
+# Example: Up to 2 punctuation marks (max specified, min omitted)
+up_to_two = Regex(r"[,.]").repeat(None, 2)
+print(to_regex(up_to_two))  # Output: ([,.]){,2}
+```
+
+You can also use the `repeat` function:
+
+```python
+from outlines import repeat
+
+# Example: Between 2 and 4 word characters
+word_chars = repeat(Regex(r"\w"), 2, 4)
+print(to_regex(word_chars))  # Output: (\w){2,4}
+
+# Example: At least 3 digits (min specified, max left open)
+at_least_three = repeat(Regex(r"\d"), 3, None)
+print(to_regex(at_least_three))  # Output: (\d){3,}
+
+# Example: Up to 2 punctuation marks (max specified, min omitted)
+up_to_two = repeat(Regex(r"[,.]"), None, 2)
+print(to_regex(up_to_two))  # Output: ([,.]){,2}
+```
+
+---
+
+## Combining Terms
+
+The DSL allows you to combine basic terms into more complex patterns using concatenation and alternation.
+
+### Concatenation (`+`)
+
+The `+` operator (and its reflected variant) concatenates terms, meaning that the terms are matched in sequence.
+
+```python
+# Example: Match "hello world"
+pattern = String("hello") + " " + Regex(r"\w+")
+print(to_regex(pattern))  # Output: hello\ (\w+)
+```
+
+### Alternation (`|`)
+
+The `|` operator creates alternatives, allowing a match for one of several patterns.
+
+```python
+# Example: Match either "cat" or "dog"
+animal = String("cat") | "dog"
+print(to_regex(animal))  # Output: (cat|dog)
+```
+
+*Note:* When using operators with plain strings (such as `"dog"`), the DSL automatically wraps them in a `String` object and escapes the characters that have a special meaning in regular expressions.
+
+---
+
+## Custom types
+
+The DSL comes "batteries included" with types that represent common text constructs:
+
+- `integer` represents an integer number as recognized by `int`
+- `boolean` represents a boolean, "True" or "False" as recognized by `bool`
+- `number` represents a floating-point number recognize by Python's `float`
+- `date` represents a date as understood by `datetime.date`
+- `time` represents a time as undestood by `datetime.time`
+- `datetime` represents a time as understoof by `datetime.datetime`
+- `digit` represents a single digit
+- `char` represents a single character
+- `newline` represents a new line character
+- `whitespace` represents a white space
+- `sentence` represents a sentence
+- `paragraph` reprensents a pagraph (one or more sentences separated by one or more ilne breaks)
+
+
+For instance you can describe the answers in the GSM8K dataset using the following pattern:
+
+```python
+from outlines.types import sentence, digit
+
+answer = "A: " + sentence.repeat(2,4) + " So the answer is: " + digit.repeat(1,4)
+```
+
+---
+
+## Practical Examples
+
+### Example 1: Matching a Custom ID Format
+
+Suppose you want to create a regex that matches an ID format like "ID-12345", where:
+- The literal "ID-" must be at the start.
+- Followed by exactly 5 digits.
+
+```python
+id_pattern = "ID-" + Regex(r"\d").times(5)
+print(to_regex(id_pattern))  # Output: ID-(\d){5}
+```
+
+### Example 2: Email Validation with Pydantic
+
+You can define a regex for email validation and use it as a type in a Pydantic model.
+
+```python
+from pydantic import BaseModel, ValidationError
+
+# Define an email regex term (this is a simplified version)
+email_regex = Regex(r"[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+")
+
+class User(BaseModel):
+    name: str
+    email: email_regex  # Use our DSL regex as a field type
+
+# Valid input
+user = User(name="Alice", email="alice@example.com")
+print(user)
+
+# Invalid input (raises a ValidationError)
+try:
+    User(name="Bob", email="not-an-email")
+except ValidationError as e:
+    print(e)
+```
+
+When used in a Pydantic model, the email field is automatically validated against the regex pattern and its JSON Schema includes the `pattern` constraint.
+
+### Example 3: Building a Complex Pattern
+
+Consider a pattern to match a simple date format: `YYYY-MM-DD`.
+
+```python
+year = Regex(r"\d").times(4)         # Four digits for the year
+month = Regex(r"\d").times(2)         # Two digits for the month
+day = Regex(r"\d").times(2)           # Two digits for the day
+
+# Combine with literal hyphens
+date_pattern = year + "-" + month + "-" + day
+print(to_regex(date_pattern))
+# Output: (\d){4}\-(\d){2}\-(\d){2}
+```
+
+---
+
+## Visualizing Your Pattern
+
+One of the unique features of this DSL is that each term can print its underlying structure as an ASCII tree. This visualization can be particularly helpful when dealing with complex expressions.
+
+```python
+# A composite pattern using concatenation and quantifiers
+pattern = "a" + String("b").one_or_more() + "c"
+print(pattern)
+```
+
+*Expected Output:*
+
+```
+└── Sequence
+    ├── String('a')
+    ├── KleenePlus(+)
+    │   └── String('b')
+    └── String('c')
+```
+
+This tree representation makes it easy to see the hierarchy and order of operations in your regular expression.
+
+---
+
+## Final Words
+
+This DSL is designed to simplify the creation and management of regular expressions—whether you're validating inputs in a web API, constraining the output of an LLM, or just experimenting with regex patterns. With intuitive methods for common quantifiers and operators, clear visual feedback, and built-in integration with Pydantic, you can build robust and maintainable regex-based validations with ease.
+
+Feel free to explore the library further and adapt the examples to your use cases. Happy regexing!
diff --git a/mkdocs.yml b/mkdocs.yml
index 8d051d1af..2c31494da 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -133,6 +133,7 @@ nav:
       - Generation:
           - Overview: reference/generation/generation.md
           - Chat templating: reference/chat_templating.md
+          - Regex DSL: reference/regex_dsl.md
           - Text: reference/text.md
           - Samplers: reference/samplers.md
           - Structured generation:
diff --git a/outlines/fsm/types.py b/outlines/fsm/types.py
index cdb1d2115..e5a1f8f47 100644
--- a/outlines/fsm/types.py
+++ b/outlines/fsm/types.py
@@ -2,14 +2,9 @@
 from enum import EnumMeta
 from typing import Any, Protocol, Tuple, Type
 
-from typing_extensions import _AnnotatedAlias, get_args
-
-INTEGER = r"[+-]?(0|[1-9][0-9]*)"
-BOOLEAN = "(True|False)"
-FLOAT = rf"{INTEGER}(\.[0-9]+)?([eE][+-][0-9]+)?"
-DATE = r"(\d{4})-(0[1-9]|1[0-2])-([0-2][0-9]|3[0-1])"
-TIME = r"([0-1][0-9]|2[0-3]):([0-5][0-9]):([0-5][0-9])"
-DATETIME = rf"({DATE})(\s)({TIME})"
+from outlines.types import Regex, boolean as boolean_regex, date as date_regex
+from outlines.types import datetime as datetime_regex
+from outlines.types import integer as integer_regex, number as number_regex, time as time_regex
 
 
 class FormatFunction(Protocol):
@@ -17,18 +12,15 @@ def __call__(self, sequence: str) -> Any:
         ...
 
 
-def python_types_to_regex(python_type: Type) -> Tuple[str, FormatFunction]:
+def python_types_to_regex(python_type: Type) -> Tuple[Regex, FormatFunction]:
     # If it is a custom type
-    if isinstance(python_type, _AnnotatedAlias):
-        json_schema = get_args(python_type)[1].json_schema
-        type_class = get_args(python_type)[0]
-
-        custom_regex_str = json_schema["pattern"]
+    if isinstance(python_type, Regex):
+        custom_regex_str = python_type.pattern
 
-        def custom_format_fn(sequence: str) -> Any:
-            return type_class(sequence)
+        def custom_format_fn(sequence: str) -> str:
+            return str(sequence)
 
-        return custom_regex_str, custom_format_fn
+        return Regex(custom_regex_str), custom_format_fn
 
     if isinstance(python_type, EnumMeta):
         values = python_type.__members__.keys()
@@ -37,44 +29,44 @@ def custom_format_fn(sequence: str) -> Any:
         def enum_format_fn(sequence: str) -> str:
             return str(sequence)
 
-        return enum_regex_str, enum_format_fn
+        return Regex(enum_regex_str), enum_format_fn
 
     if python_type is float:
 
         def float_format_fn(sequence: str) -> float:
             return float(sequence)
 
-        return FLOAT, float_format_fn
+        return number_regex, float_format_fn
     elif python_type is int:
 
         def int_format_fn(sequence: str) -> int:
             return int(sequence)
 
-        return INTEGER, int_format_fn
+        return integer_regex, int_format_fn
     elif python_type is bool:
 
         def bool_format_fn(sequence: str) -> bool:
             return bool(sequence)
 
-        return BOOLEAN, bool_format_fn
+        return boolean_regex, bool_format_fn
     elif python_type == datetime.date:
 
         def date_format_fn(sequence: str) -> datetime.date:
             return datetime.datetime.strptime(sequence, "%Y-%m-%d").date()
 
-        return DATE, date_format_fn
+        return date_regex, date_format_fn
     elif python_type == datetime.time:
 
         def time_format_fn(sequence: str) -> datetime.time:
             return datetime.datetime.strptime(sequence, "%H:%M:%S").time()
 
-        return TIME, time_format_fn
+        return time_regex, time_format_fn
     elif python_type == datetime.datetime:
 
         def datetime_format_fn(sequence: str) -> datetime.datetime:
             return datetime.datetime.strptime(sequence, "%Y-%m-%d %H:%M:%S")
 
-        return DATETIME, datetime_format_fn
+        return datetime_regex, datetime_format_fn
     else:
         raise NotImplementedError(
             f"The Python type {python_type} is not supported. Please open an issue."
diff --git a/outlines/generate/format.py b/outlines/generate/format.py
index 88acec75f..56ed10ac1 100644
--- a/outlines/generate/format.py
+++ b/outlines/generate/format.py
@@ -33,6 +33,7 @@ def format(
 
     """
     regex_str, format_fn = python_types_to_regex(python_type)
+    regex_str = regex_str.pattern
     generator = regex(model, regex_str, sampler)
     generator.format_sequence = format_fn
 
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
index 673880e49..326701f4b 100644
--- a/outlines/generate/regex.py
+++ b/outlines/generate/regex.py
@@ -6,10 +6,11 @@
 )
 from outlines.models import OpenAI, TransformersVision
 from outlines.samplers import Sampler, multinomial
+from outlines.types import Regex
 
 
 @singledispatch
-def regex(model, regex_str: str, sampler: Sampler = multinomial()):
+def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
     """Generate structured text in the language of a regular expression.
 
     Parameters
@@ -31,6 +32,9 @@ def regex(model, regex_str: str, sampler: Sampler = multinomial()):
     """
     from outlines.processors import RegexLogitsProcessor
 
+    if isinstance(regex_str, Regex):
+        regex_str = regex_str.pattern
+
     logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
 
@@ -38,11 +42,14 @@ def regex(model, regex_str: str, sampler: Sampler = multinomial()):
 @regex.register(TransformersVision)
 def regex_vision(
     model,
-    regex_str: str,
+    regex_str: str | Regex,
     sampler: Sampler = multinomial(),
 ):
     from outlines.processors import RegexLogitsProcessor
 
+    if isinstance(regex_str, Regex):
+        regex_str = regex_str.pattern
+
     logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
     return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
 
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index f4d2b8cd3..9511720af 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,4 +1,39 @@
-from . import airports, countries
-from .email import Email
-from .isbn import ISBN
-from .locales import locale
+from enum import Enum
+
+from . import airports, countries, locale
+from outlines.types.dsl import Regex, json_schema, one_or_more, optional, regex, repeat, zero_or_more, times
+
+# Python types
+integer = Regex(r"[+-]?(0|[1-9][0-9]*)")
+boolean = Regex("(True|False)")
+number = Regex(rf"{integer.pattern}(\.[0-9]+)?([eE][+-][0-9]+)?")
+date = Regex(r"(\d{4})-(0[1-9]|1[0-2])-([0-2][0-9]|3[0-1])")
+time = Regex(r"([0-1][0-9]|2[0-3]):([0-5][0-9]):([0-5][0-9])")
+datetime = Regex(rf"({date.pattern})(\s)({time.pattern})")
+
+# Basic regex types
+digit = Regex(r"\d")
+char = Regex(r"\w")
+newline = Regex(r"(\r\n|\r|\n)")  # Matched new lines on Linux, Windows & MacOS
+whitespace = Regex(r"\s")
+
+# Document-specific types
+sentence = Regex(r"[A-Z].*\s*[.!?]")
+paragraph = Regex(rf"{sentence.pattern}(?:\s+{sentence.pattern})*\n+")
+
+
+# The following regex is FRC 5322 compliant and was found at:
+# https://emailregex.com/
+email = Regex(
+    r"""(?:[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*|"(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21\x23-\x5b\x5d-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])*")@(?:(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?|\[(?:(?:(2(5[0-5]|[0-4][0-9])|1[0-9][0-9]|[1-9]?[0-9]))\.){3}(?:(2(5[0-5]|[0-4][0-9])|1[0-9][0-9]|[1-9]?[0-9])|[a-z0-9-]*[a-z0-9]:(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21-\x5a\x53-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])+)\])"""
+)
+
+# Matches any ISBN number. Note that this is not completely correct as not all
+# 10 or 13 digits numbers are valid ISBNs. See https://en.wikipedia.org/wiki/ISBN
+# Taken from O'Reilly's Regular Expression Cookbook:
+# https://www.oreilly.com/library/view/regular-expressions-cookbook/9781449327453/ch04s13.html
+#
+# TODO: The check digit can only be computed by calling a function to compute it dynamically
+isbn = Regex(
+    r"(?:ISBN(?:-1[03])?:? )?(?=[0-9X]{10}$|(?=(?:[0-9]+[- ]){3})[- 0-9X]{13}$|97[89][0-9]{10}$|(?=(?:[0-9]+[- ]){4})[- 0-9]{17}$)(?:97[89][- ]?)?[0-9]{1,5}[- ]?[0-9]+[- ]?[0-9]+[- ]?[0-9X]"
+)
diff --git a/outlines/types/airports.py b/outlines/types/airports.py
index 934ae1844..ec0ef72bd 100644
--- a/outlines/types/airports.py
+++ b/outlines/types/airports.py
@@ -6,6 +6,4 @@
 AIRPORT_IATA_LIST = [
     (v["iata"], v["iata"]) for v in airportsdata.load().values() if v["iata"]
 ]
-
-
 IATA = Enum("Airport", AIRPORT_IATA_LIST)  # type:ignore
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
new file mode 100644
index 000000000..86feabca0
--- /dev/null
+++ b/outlines/types/dsl.py
@@ -0,0 +1,395 @@
+import json as json
+import re
+from dataclasses import dataclass
+from typing import Any, List, Union
+
+from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler
+from pydantic.json_schema import JsonSchemaValue
+from pydantic_core import core_schema as cs
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+
+class Term:
+    """Represents types defined with a regular expression.
+
+    `Regex` instances can be used as a type in a Pydantic model definittion.
+    They will be translated to JSON Schema as a "string" field with the
+    "pattern" keyword set to the regular expression this class represents. The
+    class also handles validation.
+
+    Examples
+    --------
+
+    >>> from outlines.types import Regex
+    >>> from pydantic import BaseModel
+    >>>
+    >>> age_type = Regex("[0-9]+")
+    >>>
+    >>> class User(BaseModel):
+    >>>     name: str
+    >>>     age: age_type
+
+    """
+
+    def __add__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
+        if isinstance(other, str):
+            other = String(other)
+
+        return Sequence([self, other])
+
+    def __radd__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
+        if isinstance(other, str):
+            other = String(other)
+
+        return Sequence([other, self])
+
+    def __or__(self: "Term", other: Union[str, "Term"]) -> "Alternatives":
+        if isinstance(other, str):
+            other = String(other)
+
+        return Alternatives([self, other])
+
+    def __ror__(self: "Term", other: Union[str, "Term"]) -> "Alternatives":
+        if isinstance(other, str):
+            other = String(other)
+
+        return Alternatives([other, self])
+
+    def __get_validator__(self, _core_schema):
+        def validate(input_value):
+            return self.validate(input_value)
+
+        return validate
+
+    def __get_pydantic_core_schema__(
+        self, source_type: Any, handler: GetCoreSchemaHandler
+    ) -> cs.CoreSchema:
+        return cs.no_info_plain_validator_function(lambda value: self.validate(value))
+
+    def __get_pydantic_json_schema__(
+        self, core_schema: cs.CoreSchema, handler: GetJsonSchemaHandler
+    ) -> JsonSchemaValue:
+
+        return {"type": "string", "pattern": to_regex(self)}
+
+    def validate(self, value: str) -> str:
+        pattern = to_regex(self)
+        compiled = re.compile(pattern)
+        if not compiled.fullmatch(str(value)):
+            raise ValueError(
+                f"Input should be in the language of the regular expression {pattern}"
+            )
+        return value
+
+    def matches(self, value: str) -> bool:
+        """Check that a given value is in the language defined by the Term.
+
+        We make the assumption that the language defined by the term can
+        be defined with a regular expression.
+
+        """
+        pattern = to_regex(self)
+        compiled = re.compile(pattern)
+        if compiled.fullmatch(str(value)):
+            return True
+        return False
+
+    def display_ascii_tree(self, indent="", is_last=True) -> str:
+        """Display the regex tree in ASCII format."""
+        branch = "└── " if is_last else "├── "
+        result = indent + branch + self._display_node() + "\n"
+
+        # Calculate the new indent for children
+        new_indent = indent + ("    " if is_last else "│   ")
+
+        # Let each subclass handle its children
+        result += self._display_children(new_indent)
+        return result
+
+    def _display_node(self):
+        raise NotImplementedError
+
+    def _display_children(self, indent: str) -> str:
+        """Display the children of this node. Override in subclasses with children."""
+        return ""
+
+    def __str__(self):
+        return self.display_ascii_tree()
+
+
+@dataclass
+class String(Term):
+    value: str
+
+    def _display_node(self) -> str:
+        return f"String('{self.value}')"
+
+    def __repr__(self):
+        return f"String(value='{self.value}')"
+
+
+@dataclass
+class Regex(Term):
+    pattern: str
+
+    def _display_node(self) -> str:
+        return f"Regex('{self.pattern}')"
+
+    def __repr__(self):
+        return f"Regex(pattern='{self.pattern}')"
+
+
+class JsonSchema(Term):
+    def __init__(self, schema: Union[dict, str, type[BaseModel]]):
+        if isinstance(schema, dict):
+            schema_str = json.dumps(schema)
+        elif isinstance(schema, str):
+            schema_str = schema
+        elif issubclass(schema, BaseModel):
+            schema_str = json.dumps(schema.model_json_schema())
+        else:
+            raise ValueError(
+                f"Cannot parse schema {json_schema}. The schema must be either "
+                + "a Pydantic class, a dictionary or a string that contains the JSON "
+                + "schema specification"
+            )
+
+        self.schema = schema_str
+
+    def _display_node(self) -> str:
+        return f"JsonSchema('{self.schema}')"
+
+    def __repr__(self):
+        return f"JsonSchema(schema='{self.schema}')"
+
+
+@dataclass
+class KleeneStar(Term):
+    term: Term
+
+    def _display_node(self) -> str:
+        return "KleeneStar(*)"
+
+    def _display_children(self, indent: str) -> str:
+        return self.term.display_ascii_tree(indent, True)
+
+    def __repr__(self):
+        return f"KleeneStar(term={repr(self.term)})"
+
+
+@dataclass
+class KleenePlus(Term):
+    term: Term
+
+    def _display_node(self) -> str:
+        return "KleenePlus(+)"
+
+    def _display_children(self, indent: str) -> str:
+        return self.term.display_ascii_tree(indent, True)
+
+    def __repr__(self):
+        return f"KleenePlus(term={repr(self.term)})"
+
+
+@dataclass
+class Optional(Term):
+    term: Term
+
+    def _display_node(self) -> str:
+        return "Optional(?)"
+
+    def _display_children(self, indent: str) -> str:
+        return self.term.display_ascii_tree(indent, True)
+
+    def __repr__(self):
+        return f"Optional(term={repr(self.term)})"
+
+
+@dataclass
+class Alternatives(Term):
+    terms: List[Term]
+
+    def _display_node(self) -> str:
+        return "Alternatives(|)"
+
+    def _display_children(self, indent: str) -> str:
+        return "".join(
+            term.display_ascii_tree(indent, i == len(self.terms) - 1)
+            for i, term in enumerate(self.terms)
+        )
+
+    def __repr__(self):
+        return f"Alternatives(terms={repr(self.terms)})"
+
+
+@dataclass
+class Sequence(Term):
+    terms: List[Term]
+
+    def _display_node(self) -> str:
+        return "Sequence"
+
+    def _display_children(self, indent: str) -> str:
+        return "".join(
+            term.display_ascii_tree(indent, i == len(self.terms) - 1)
+            for i, term in enumerate(self.terms)
+        )
+
+    def __repr__(self):
+        return f"Sequence(terms={repr(self.terms)})"
+
+
+@dataclass
+class QuantifyExact(Term):
+    term: Term
+    count: int
+
+    def _display_node(self) -> str:
+        return f"Quantify({{{self.count}}})"
+
+    def _display_children(self, indent: str) -> str:
+        return self.term.display_ascii_tree(indent, True)
+
+    def __repr__(self):
+        return f"QuantifyExact(term={repr(self.term)}, count={repr(self.count)})"
+
+
+@dataclass
+class QuantifyMinimum(Term):
+    term: Term
+    min_count: int
+
+    def _display_node(self) -> str:
+        return f"Quantify({{{self.min_count},}})"
+
+    def _display_children(self, indent: str) -> str:
+        return self.term.display_ascii_tree(indent, True)
+
+    def __repr__(self):
+        return (
+            f"QuantifyMinimum(term={repr(self.term)}, min_count={repr(self.min_count)})"
+        )
+
+
+@dataclass
+class QuantifyMaximum(Term):
+    term: Term
+    max_count: int
+
+    def _display_node(self) -> str:
+        return f"Quantify({{,{self.max_count}}})"
+
+    def _display_children(self, indent: str) -> str:
+        return self.term.display_ascii_tree(indent, True)
+
+    def __repr__(self):
+        return (
+            f"QuantifyMaximum(term={repr(self.term)}, max_count={repr(self.max_count)})"
+        )
+
+
+@dataclass
+class QuantifyBetween(Term):
+    term: Term
+    min_count: int
+    max_count: int
+
+    def __post_init__(self):
+        if self.min_count > self.max_count:
+            raise ValueError(
+                "QuantifyBetween: `max_count` must be greater than `min_count`."
+            )
+
+    def _display_node(self) -> str:
+        return f"Quantify({{{self.min_count},{self.max_count}}})"
+
+    def _display_children(self, indent: str) -> str:
+        return self.term.display_ascii_tree(indent, True)
+
+    def __repr__(self):
+        return f"QuantifyBetween(term={repr(self.term)}, min_count={repr(self.min_count)}, max_count={repr(self.max_count)})"
+
+
+def optional(self: Term) -> Optional:
+    return Optional(self)
+
+
+def one_or_more(self: Term) -> KleenePlus:
+    return KleenePlus(self)
+
+
+def repeat(self: Term, min_count: int, max_count: int) -> QuantifyBetween:
+    match (min_count, max_count):
+        case (None, None):
+            raise ValueError(
+                "repeat: you must provide a value for at least `min_count` or `max_count`"
+            )
+        case (_, None):
+            return QuantifyMinimum(self, min_count)
+        case (None, _):
+            return QuantifyMaximum(self, max_count)
+        case _:
+            return QuantifyBetween(self, min_count, max_count)
+
+
+def times(self: Term, count: int = 0) -> QuantifyExact:
+    return QuantifyExact(self, count)
+
+
+def zero_or_more(self: Term) -> KleeneStar:
+    return KleeneStar(self)
+
+
+Term.one_or_more = one_or_more  # type: ignore
+Term.optional = optional  # type: ignore
+Term.repeat = repeat  # type: ignore
+Term.times = times  # type: ignore
+Term.zero_or_more = zero_or_more  # type: ignore
+
+
+def regex(pattern: str):
+    return Regex(pattern)
+
+
+def json_schema(schema: Union[str, dict, type[BaseModel]]):
+    return JsonSchema(schema)
+
+
+def to_regex(term: Term) -> str:
+    """Convert a term to a regular expression.
+
+    We only consider self-contained terms that do not refer to another rule.
+
+    """
+    match term:
+        case String():
+            return re.escape(term.value)
+        case Regex():
+            return f"({term.pattern})"
+        case JsonSchema():
+            regex_str = build_regex_from_schema(term.schema)
+            return f"({regex_str})"
+        case KleeneStar():
+            return f"({to_regex(term.term)})*"
+        case KleenePlus():
+            return f"({to_regex(term.term)})+"
+        case Optional():
+            return f"({to_regex(term.term)})?"
+        case Alternatives():
+            regexes = [to_regex(subterm) for subterm in term.terms]
+            return f"({'|'.join(regexes)})"
+        case Sequence():
+            regexes = [to_regex(subterm) for subterm in term.terms]
+            return f"{''.join(regexes)}"
+        case QuantifyExact():
+            return f"({to_regex(term.term)}){{{term.count}}}"
+        case QuantifyMinimum():
+            return f"({to_regex(term.term)}){{{term.min_count},}}"
+        case QuantifyMaximum():
+            return f"({to_regex(term.term)}){{,{term.max_count}}}"
+        case QuantifyBetween():
+            return f"({to_regex(term.term)}){{{term.min_count},{term.max_count}}}"
+        case _:
+            raise TypeError(
+                f"Cannot convert object {repr(term)} to a regular expression."
+            )
diff --git a/outlines/types/email.py b/outlines/types/email.py
deleted file mode 100644
index 45f8c4b2c..000000000
--- a/outlines/types/email.py
+++ /dev/null
@@ -1,11 +0,0 @@
-"""Email Address types."""
-from pydantic import WithJsonSchema
-from typing_extensions import Annotated
-
-# Taken from StackOverflow
-# https://stackoverflow.com/a/201378/14773537
-EMAIL_REGEX = r"""(?:[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*|"(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21\x23-\x5b\x5d-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])*")@(?:(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?|\[(?:(?:(2(5[0-5]|[0-4][0-9])|1[0-9][0-9]|[1-9]?[0-9]))\.){3}(?:(2(5[0-5]|[0-4][0-9])|1[0-9][0-9]|[1-9]?[0-9])|[a-z0-9-]*[a-z0-9]:(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21-\x5a\x53-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])+)\])"""
-Email = Annotated[
-    str,
-    WithJsonSchema({"type": "string", "pattern": EMAIL_REGEX}),
-]
diff --git a/outlines/types/isbn.py b/outlines/types/isbn.py
deleted file mode 100644
index 5aebb067e..000000000
--- a/outlines/types/isbn.py
+++ /dev/null
@@ -1,12 +0,0 @@
-"""ISBN type"""
-from pydantic import WithJsonSchema
-from typing_extensions import Annotated
-
-# Matches any ISBN number.  Note that this is not completely correct as not all
-# 10 or 13 digits numbers are valid ISBNs. See https://en.wikipedia.org/wiki/ISBN
-# Taken from O'Reilly's Regular Expression Cookbook:
-# https://www.oreilly.com/library/view/regular-expressions-cookbook/9781449327453/ch04s13.html
-# TODO: Can this be represented by a grammar or do we need semantic checks?
-ISBN_REGEX = r"(?:ISBN(?:-1[03])?:? )?(?=[0-9X]{10}$|(?=(?:[0-9]+[- ]){3})[- 0-9X]{13}$|97[89][0-9]{10}$|(?=(?:[0-9]+[- ]){4})[- 0-9]{17}$)(?:97[89][- ]?)?[0-9]{1,5}[- ]?[0-9]+[- ]?[0-9]+[- ]?[0-9X]"
-
-ISBN = Annotated[str, WithJsonSchema({"type": "string", "pattern": ISBN_REGEX})]
diff --git a/outlines/types/locale/__init__.py b/outlines/types/locale/__init__.py
new file mode 100644
index 000000000..511631d84
--- /dev/null
+++ b/outlines/types/locale/__init__.py
@@ -0,0 +1 @@
+from . import us
diff --git a/outlines/types/locale/us.py b/outlines/types/locale/us.py
new file mode 100644
index 000000000..0bda82d44
--- /dev/null
+++ b/outlines/types/locale/us.py
@@ -0,0 +1,4 @@
+from outlines.types.dsl import Regex
+
+zip_code = Regex(r"\d{5}(?:-\d{4})?")
+phone_number = Regex(r"(\([0-9]{3}\) |[0-9]{3}-)[0-9]{3}-[0-9]{4}")
diff --git a/outlines/types/locales.py b/outlines/types/locales.py
deleted file mode 100644
index c5d251bae..000000000
--- a/outlines/types/locales.py
+++ /dev/null
@@ -1,21 +0,0 @@
-from dataclasses import dataclass
-
-from outlines.types.phone_numbers import USPhoneNumber
-from outlines.types.zip_codes import USZipCode
-
-
-@dataclass
-class US:
-    ZipCode = USZipCode
-    PhoneNumber = USPhoneNumber
-
-
-def locale(locale_str: str):
-    locales = {"us": US}
-
-    if locale_str not in locales:
-        raise NotImplementedError(
-            f"The locale {locale_str} is not supported yet. Please don't hesitate to create custom types for you locale and open a Pull Request."
-        )
-
-    return locales[locale_str]
diff --git a/outlines/types/phone_numbers.py b/outlines/types/phone_numbers.py
deleted file mode 100644
index 618687e75..000000000
--- a/outlines/types/phone_numbers.py
+++ /dev/null
@@ -1,16 +0,0 @@
-"""Phone number types.
-
-We currently only support US phone numbers. We can however imagine having custom types
-for each country, for instance leveraging the `phonenumbers` library.
-
-"""
-from pydantic import WithJsonSchema
-from typing_extensions import Annotated
-
-US_PHONE_NUMBER = r"(\([0-9]{3}\) |[0-9]{3}-)[0-9]{3}-[0-9]{4}"
-
-
-USPhoneNumber = Annotated[
-    str,
-    WithJsonSchema({"type": "string", "pattern": US_PHONE_NUMBER}),
-]
diff --git a/outlines/types/zip_codes.py b/outlines/types/zip_codes.py
deleted file mode 100644
index 67d994d5c..000000000
--- a/outlines/types/zip_codes.py
+++ /dev/null
@@ -1,13 +0,0 @@
-"""Zip code types.
-
-We currently only support US Zip Codes.
-
-"""
-from pydantic import WithJsonSchema
-from typing_extensions import Annotated
-
-# This matches Zip and Zip+4 codes
-US_ZIP_CODE = r"\d{5}(?:-\d{4})?"
-
-
-USZipCode = Annotated[str, WithJsonSchema({"type": "string", "pattern": US_ZIP_CODE})]
diff --git a/pyproject.toml b/pyproject.toml
index f1ff57323..2762db439 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -132,6 +132,7 @@ module = [
     "cloudpickle.*",
     "diskcache.*",
     "pydantic.*",
+    "pydantic_core.*",
     "pytest",
     "referencing.*",
     "torch.*",
diff --git a/tests/fsm/test_types.py b/tests/fsm/test_types.py
index d5450434c..4017a817c 100644
--- a/tests/fsm/test_types.py
+++ b/tests/fsm/test_types.py
@@ -1,29 +1,22 @@
-import datetime
+import datetime as pydatetime
 
 import pytest
 
-from outlines.fsm.types import (
-    BOOLEAN,
-    DATE,
-    DATETIME,
-    FLOAT,
-    INTEGER,
-    TIME,
-    python_types_to_regex,
-)
+from outlines.fsm.types import python_types_to_regex
+from outlines import types
 
 
 @pytest.mark.parametrize(
-    "python_type,regex",
+    "python_type,custom_type",
     [
-        (int, INTEGER),
-        (float, FLOAT),
-        (bool, BOOLEAN),
-        (datetime.date, DATE),
-        (datetime.time, TIME),
-        (datetime.datetime, DATETIME),
+        (int, types.integer),
+        (float, types.number),
+        (bool, types.boolean),
+        (pydatetime.date, types.date),
+        (pydatetime.time, types.time),
+        (pydatetime.datetime, types.datetime),
     ],
 )
-def test_python_types(python_type, regex):
+def test_python_types(python_type, custom_type):
     test_regex, _ = python_types_to_regex(python_type)
-    assert regex == test_regex
+    assert custom_type.pattern == test_regex.pattern
diff --git a/tests/test_types.py b/tests/test_types.py
deleted file mode 100644
index 5e60348b2..000000000
--- a/tests/test_types.py
+++ /dev/null
@@ -1,103 +0,0 @@
-import re
-
-import pytest
-from pydantic import BaseModel
-
-from outlines import types
-from outlines.fsm.types import python_types_to_regex
-
-
-@pytest.mark.parametrize(
-    "custom_type,test_string,should_match",
-    [
-        (types.phone_numbers.USPhoneNumber, "12", False),
-        (types.phone_numbers.USPhoneNumber, "(123) 123-1234", True),
-        (types.phone_numbers.USPhoneNumber, "123-123-1234", True),
-        (types.zip_codes.USZipCode, "12", False),
-        (types.zip_codes.USZipCode, "12345", True),
-        (types.zip_codes.USZipCode, "12345-1234", True),
-        (types.ISBN, "ISBN 0-1-2-3-4-5", False),
-        (types.ISBN, "ISBN 978-0-596-52068-7", True),
-        # (types.ISBN, "ISBN 978-0-596-52068-1", True), wrong check digit
-        (types.ISBN, "ISBN-13: 978-0-596-52068-7", True),
-        (types.ISBN, "978 0 596 52068 7", True),
-        (types.ISBN, "9780596520687", True),
-        (types.ISBN, "ISBN-10: 0-596-52068-9", True),
-        (types.ISBN, "0-596-52068-9", True),
-        (types.Email, "eitan@gmail.com", True),
-        (types.Email, "99@yahoo.com", True),
-        (types.Email, "eitan@.gmail.com", False),
-        (types.Email, "myemail", False),
-        (types.Email, "eitan@gmail", False),
-        (types.Email, "eitan@my.custom.domain", True),
-    ],
-)
-def test_type_regex(custom_type, test_string, should_match):
-    class Model(BaseModel):
-        attr: custom_type
-
-    schema = Model.model_json_schema()
-    assert schema["properties"]["attr"]["type"] == "string"
-    regex_str = schema["properties"]["attr"]["pattern"]
-    does_match = re.match(regex_str, test_string) is not None
-    assert does_match is should_match
-
-    regex_str, format_fn = python_types_to_regex(custom_type)
-    assert isinstance(format_fn(1), str)
-    does_match = re.match(regex_str, test_string) is not None
-    assert does_match is should_match
-
-
-def test_locale_not_implemented():
-    with pytest.raises(NotImplementedError):
-        types.locale("fr")
-
-
-@pytest.mark.parametrize(
-    "locale_str,base_types,locale_types",
-    [
-        (
-            "us",
-            ["ZipCode", "PhoneNumber"],
-            [types.zip_codes.USZipCode, types.phone_numbers.USPhoneNumber],
-        )
-    ],
-)
-def test_locale(locale_str, base_types, locale_types):
-    for base_type, locale_type in zip(base_types, locale_types):
-        type = getattr(types.locale(locale_str), base_type)
-        assert type == locale_type
-
-
-@pytest.mark.parametrize(
-    "custom_type,test_string,should_match",
-    [
-        (types.airports.IATA, "CDG", True),
-        (types.airports.IATA, "XXX", False),
-        (types.countries.Alpha2, "FR", True),
-        (types.countries.Alpha2, "XX", False),
-        (types.countries.Alpha3, "UKR", True),
-        (types.countries.Alpha3, "XXX", False),
-        (types.countries.Numeric, "004", True),
-        (types.countries.Numeric, "900", False),
-        (types.countries.Name, "Ukraine", True),
-        (types.countries.Name, "Wonderland", False),
-        (types.countries.Flag, "🇿🇼", True),
-        (types.countries.Flag, "🤗", False),
-    ],
-)
-def test_type_enum(custom_type, test_string, should_match):
-    type_name = custom_type.__name__
-
-    class Model(BaseModel):
-        attr: custom_type
-
-    schema = Model.model_json_schema()
-    assert isinstance(schema["$defs"][type_name]["enum"], list)
-    does_match = test_string in schema["$defs"][type_name]["enum"]
-    assert does_match is should_match
-
-    regex_str, format_fn = python_types_to_regex(custom_type)
-    assert isinstance(format_fn(1), str)
-    does_match = re.match(regex_str, test_string) is not None
-    assert does_match is should_match
diff --git a/tests/types/test_custom_types.py b/tests/types/test_custom_types.py
new file mode 100644
index 000000000..46df9ba4a
--- /dev/null
+++ b/tests/types/test_custom_types.py
@@ -0,0 +1,111 @@
+import re
+
+import pytest
+from pydantic import BaseModel
+
+from outlines import types
+from outlines.fsm.types import python_types_to_regex
+from outlines.types.dsl import to_regex
+
+
+@pytest.mark.parametrize(
+    "custom_type,test_string,should_match",
+    [
+        (types.locale.us.phone_number, "12", False),
+        (types.locale.us.phone_number, "(123) 123-1234", True),
+        (types.locale.us.phone_number, "123-123-1234", True),
+        (types.locale.us.zip_code, "12", False),
+        (types.locale.us.zip_code, "12345", True),
+        (types.locale.us.zip_code, "12345-1234", True),
+        (types.isbn, "ISBN 0-1-2-3-4-5", False),
+        (types.isbn, "ISBN 978-0-596-52068-7", True),
+        (types.isbn, "ISBN-13: 978-0-596-52068-7", True),
+        (types.isbn, "978 0 596 52068 7", True),
+        (types.isbn, "9780596520687", True),
+        (types.isbn, "ISBN-10: 0-596-52068-9", True),
+        (types.isbn, "0-596-52068-9", True),
+        (types.email, "eitan@gmail.com", True),
+        (types.email, "99@yahoo.com", True),
+        (types.email, "eitan@.gmail.com", False),
+        (types.email, "myemail", False),
+        (types.email, "eitan@gmail", False),
+        (types.email, "eitan@my.custom.domain", True),
+        (types.integer, "-19", True),
+        (types.integer, "19", True),
+        (types.integer, "019", False),
+        (types.integer, "1.9", False),
+        (types.integer, "a", False),
+        (types.boolean, "True", True),
+        (types.boolean, "False", True),
+        (types.boolean, "true", False),
+        (types.number, "10", True),
+        (types.number, "10.9", True),
+        (types.number, "10.9e+3", True),
+        (types.number, "10.9e-3", True),
+        (types.number, "a", False),
+        (types.date, "2022-03-23", True),
+        (types.date, "2022-03-32", False),
+        (types.date, "2022-13-23", False),
+        (types.date, "32-03-2022", False),
+        (types.time, "01:23:59", True),
+        (types.time, "01:23:61", False),
+        (types.time, "01:61:59", False),
+        (types.time, "24:23:59", False),
+        (types.sentence, "The temperature is 23.5 degrees !", True),
+        (types.sentence, "Did you earn $1,234.56 last month  ?", True),
+        (types.sentence, "The #1 player scored 100 points .", True),
+        (types.sentence, "Hello @world, this is a test!", True),
+        (types.sentence, "invalid sentence.", False),
+        (types.sentence, "Invalid sentence", False),
+        (types.paragraph, "This is a paragraph!\n", True),
+        (types.paragraph, "Line1\nLine2", False),
+        (types.paragraph, "One sentence. Two sentences.\n\n", True),
+        (types.paragraph, "One sentence. invalid sentence.", False),
+        (types.paragraph, "One sentence. Invalid sentence\n", False),
+    ],
+)
+def test_type_regex(custom_type, test_string, should_match):
+    class Model(BaseModel):
+        attr: custom_type
+
+    schema = Model.model_json_schema()
+    assert schema["properties"]["attr"]["type"] == "string"
+    regex_str = schema["properties"]["attr"]["pattern"]
+    does_match = re.fullmatch(regex_str, test_string) is not None
+    assert does_match is should_match
+
+    regex_str = to_regex(custom_type)
+    does_match = re.fullmatch(regex_str, test_string) is not None
+    assert does_match is should_match
+
+
+@pytest.mark.parametrize(
+    "custom_type,test_string,should_match",
+    [
+        (types.airports.IATA, "CDG", True),
+        (types.airports.IATA, "XXX", False),
+        (types.countries.Alpha2, "FR", True),
+        (types.countries.Alpha2, "XX", False),
+        (types.countries.Alpha3, "UKR", True),
+        (types.countries.Alpha3, "XXX", False),
+        (types.countries.Numeric, "004", True),
+        (types.countries.Numeric, "900", False),
+        (types.countries.Name, "Ukraine", True),
+        (types.countries.Name, "Wonderland", False),
+        (types.countries.Flag, "🇿🇼", True),
+        (types.countries.Flag, "🤗", False),
+    ],
+)
+def test_type_enum(custom_type, test_string, should_match):
+    type_name = custom_type.__name__
+
+    class Model(BaseModel):
+        attr: custom_type
+
+    schema = Model.model_json_schema()
+    assert isinstance(schema["$defs"][type_name]["enum"], list)
+    does_match = test_string in schema["$defs"][type_name]["enum"]
+    assert does_match is should_match
+
+    does_match = test_string in custom_type.__members__
+    assert does_match is should_match
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
new file mode 100644
index 000000000..dc2779ef1
--- /dev/null
+++ b/tests/types/test_dsl.py
@@ -0,0 +1,256 @@
+import pytest
+from pydantic import BaseModel
+
+from outlines.types.dsl import (
+    Alternatives,
+    JsonSchema,
+    KleenePlus,
+    KleeneStar,
+    Optional,
+    QuantifyBetween,
+    QuantifyExact,
+    QuantifyMaximum,
+    QuantifyMinimum,
+    Regex,
+    Sequence,
+    String,
+    Term,
+    one_or_more,
+    optional,
+    repeat,
+    times,
+    regex,
+    json_schema,
+    zero_or_more,
+)
+
+
+def test_dsl_init():
+    string = String("test")
+    assert string.value == "test"
+    assert repr(string) == "String(value='test')"
+
+    regex = Regex("[0-9]")
+    assert regex.pattern == "[0-9]"
+    assert repr(regex) == "Regex(pattern='[0-9]')"
+
+    schema = JsonSchema('{ "type": "string" }')
+    assert schema.schema == '{ "type": "string" }'
+    assert repr(schema) == 'JsonSchema(schema=\'{ "type": "string" }\')'
+
+    kleene_star = KleeneStar(string)
+    assert kleene_star.term == string
+    assert repr(kleene_star) == "KleeneStar(term=String(value='test'))"
+
+    kleene_plus = KleenePlus(string)
+    assert kleene_plus.term == string
+    assert repr(kleene_plus) == "KleenePlus(term=String(value='test'))"
+
+    optional = Optional(string)
+    assert optional.term == string
+    assert repr(optional) == "Optional(term=String(value='test'))"
+
+    alternatives = Alternatives([string, regex])
+    assert alternatives.terms[0] == string
+    assert alternatives.terms[1] == regex
+    assert (
+        repr(alternatives)
+        == "Alternatives(terms=[String(value='test'), Regex(pattern='[0-9]')])"
+    )
+
+    sequence = Sequence([string, regex])
+    assert sequence.terms[0] == string
+    assert sequence.terms[1] == regex
+    assert (
+        repr(sequence)
+        == "Sequence(terms=[String(value='test'), Regex(pattern='[0-9]')])"
+    )
+
+    exact = QuantifyExact(string, 3)
+    assert exact.term == string
+    assert exact.count == 3
+    assert repr(exact) == "QuantifyExact(term=String(value='test'), count=3)"
+
+    minimum = QuantifyMinimum(string, 3)
+    assert minimum.term == string
+    assert minimum.min_count == 3
+    assert repr(minimum) == "QuantifyMinimum(term=String(value='test'), min_count=3)"
+
+    maximum = QuantifyMaximum(string, 3)
+    assert maximum.term == string
+    assert maximum.max_count == 3
+    assert repr(maximum) == "QuantifyMaximum(term=String(value='test'), max_count=3)"
+
+    between = QuantifyBetween(string, 1, 3)
+    assert between.term == string
+    assert between.min_count == 1
+    assert between.max_count == 3
+    assert (
+        repr(between)
+        == "QuantifyBetween(term=String(value='test'), min_count=1, max_count=3)"
+    )
+
+    with pytest.raises(
+        ValueError, match="`max_count` must be greater than `min_count`"
+    ):
+        QuantifyBetween(string, 3, 1)
+
+
+def test_dsl_operations():
+    a = String("a")
+    b = String("b")
+    assert isinstance(a + b, Sequence)
+    assert (a + b).terms[0] == a
+    assert (a + b).terms[1] == b
+
+    assert isinstance(a | b, Alternatives)
+    assert (a | b).terms[0] == a
+    assert (a | b).terms[1] == b
+
+
+def test_dsl_operations_string_conversion():
+    b = String("b")
+    sequence = "a" + b
+    assert isinstance(sequence, Sequence)
+    assert isinstance(sequence.terms[0], String)
+    assert sequence.terms[0].value == "a"
+    assert sequence.terms[1].value == "b"
+
+    sequence = b + "a"
+    assert isinstance(sequence, Sequence)
+    assert isinstance(sequence.terms[0], String)
+    assert sequence.terms[0].value == "b"
+    assert sequence.terms[1].value == "a"
+
+    alternative = "a" | b
+    assert isinstance(alternative, Alternatives)
+    assert isinstance(alternative.terms[0], String)
+    assert alternative.terms[0].value == "a"
+    assert alternative.terms[1].value == "b"
+
+    alternative = b | "a"
+    assert isinstance(alternative, Alternatives)
+    assert isinstance(alternative.terms[0], String)
+    assert alternative.terms[0].value == "b"
+    assert alternative.terms[1].value == "a"
+
+
+def test_dsl_aliases():
+    test = regex("[0-9]")
+    assert isinstance(test, Regex)
+
+    test = json_schema('{"type": "string"}')
+    assert isinstance(test, JsonSchema)
+
+    test = String("test")
+
+    assert isinstance(test.times(3), QuantifyExact)
+    assert test.times(3).count == 3
+    assert test.times(3).term == test
+
+    assert isinstance(times(test, 3), QuantifyExact)
+    assert times(test, 3).count == 3
+    assert times(test, 3).term == test
+
+    assert isinstance(test.one_or_more(), KleenePlus)
+    assert test.one_or_more().term == test
+
+    assert isinstance(one_or_more(test), KleenePlus)
+    assert one_or_more(test).term == test
+
+    assert isinstance(test.zero_or_more(), KleeneStar)
+    assert test.zero_or_more().term == test
+
+    assert isinstance(zero_or_more(test), KleeneStar)
+    assert zero_or_more(test).term == test
+
+    assert isinstance(test.optional(), Optional)
+    assert test.optional().term == test
+
+    assert isinstance(optional(test), Optional)
+    assert optional(test).term == test
+
+    rep_min = test.repeat(2, None)
+    assert isinstance(rep_min, QuantifyMinimum)
+    assert rep_min.min_count == 2
+
+    rep_min = repeat(test, 2, None)
+    assert isinstance(rep_min, QuantifyMinimum)
+    assert rep_min.min_count == 2
+
+    rep_max = test.repeat(None, 2)
+    assert isinstance(rep_max, QuantifyMaximum)
+    assert rep_max.max_count == 2
+
+    rep_max = repeat(test, None, 2)
+    assert isinstance(rep_max, QuantifyMaximum)
+    assert rep_max.max_count == 2
+
+    rep_between = test.repeat(1, 2)
+    assert isinstance(rep_between, QuantifyBetween)
+    assert rep_between.min_count == 1
+    assert rep_between.max_count == 2
+
+    rep_between = repeat(test, 1, 2)
+    assert isinstance(rep_between, QuantifyBetween)
+    assert rep_between.min_count == 1
+    assert rep_between.max_count == 2
+
+    with pytest.raises(ValueError, match="QuantifyBetween: `max_count` must be"):
+        test.repeat(2, 1)
+
+    with pytest.raises(ValueError, match="QuantifyBetween: `max_count` must be"):
+        repeat(test, 2, 1)
+
+    with pytest.raises(ValueError, match="repeat: you must provide"):
+        test.repeat(None, None)
+
+    with pytest.raises(ValueError, match="repeat: you must provide"):
+        repeat(test, None, None)
+
+
+def test_dsl_term_pydantic_simple():
+    a = String("a")
+
+    class Model(BaseModel):
+        field: a
+
+    schema = Model.model_json_schema()
+    assert schema == {
+        "properties": {"field": {"pattern": "a", "title": "Field", "type": "string"}},
+        "required": ["field"],
+        "title": "Model",
+        "type": "object",
+    }
+
+
+def test_dsl_term_pydantic_combination():
+    a = String("a")
+    b = String("b")
+    c = String("c")
+
+    class Model(BaseModel):
+        field: (a + b) | c
+
+    schema = Model.model_json_schema()
+    assert schema == {
+        "properties": {
+            "field": {"pattern": "(ab|c)", "title": "Field", "type": "string"}
+        },
+        "required": ["field"],
+        "title": "Model",
+        "type": "object",
+    }
+
+
+def test_dsl_display():
+    a = String("a")
+    b = String("b")
+    c = Regex("[0-9]")
+    d = KleeneStar(a | b) + c
+
+    tree = str(d)
+    assert (
+        tree
+        == "└── Sequence\n    ├── KleeneStar(*)\n    │   └── Alternatives(|)\n    │       ├── String('a')\n    │       └── String('b')\n    └── Regex('[0-9]')\n"
+    )
diff --git a/tests/types/test_to_regex.py b/tests/types/test_to_regex.py
new file mode 100644
index 000000000..6cb566fc5
--- /dev/null
+++ b/tests/types/test_to_regex.py
@@ -0,0 +1,71 @@
+import pytest
+
+
+from outlines.types.dsl import String, Regex, JsonSchema, KleeneStar, KleenePlus, QuantifyBetween, QuantifyExact, QuantifyMaximum, QuantifyMinimum, Sequence, Alternatives, Optional, Term, to_regex
+
+
+def test_to_regex_simple():
+    a = String("a")
+    assert to_regex(a) == "a"
+    assert a.matches("a") is True
+
+    a = Regex("[0-9]")
+    assert to_regex(a) == "([0-9])"
+    assert a.matches(0) is True
+    assert a.matches(10) is False
+    assert a.matches("a") is False
+
+    a = JsonSchema({"type": "integer"})
+    assert to_regex(a) == r"((-)?(0|[1-9][0-9]*))"
+    assert a.matches(1) is True
+    assert a.matches("1") is True
+    assert a.matches("a") is False
+
+    a = Optional(String("a"))
+    assert to_regex(a) == "(a)?"
+    assert a.matches("") is True
+    assert a.matches("a") is True
+
+    a = KleeneStar(String("a"))
+    assert to_regex(a) == "(a)*"
+    assert a.matches("") is True
+    assert a.matches("a") is True
+    assert a.matches("aaaaa") is True
+
+    a = KleenePlus(String("a"))
+    assert to_regex(a) == "(a)+"
+    assert a.matches("") is False
+    assert a.matches("a") is True
+    assert a.matches("aaaaa") is True
+
+    a = QuantifyExact(String("a"), 2)
+    assert to_regex(a) == "(a){2}"
+    assert a.matches("a") is False
+    assert a.matches("aa") is True
+    assert a.matches("aaa") is False
+
+    a = QuantifyMinimum(String("a"), 2)
+    assert to_regex(a) == "(a){2,}"
+    assert a.matches("a") is False
+    assert a.matches("aa") is True
+    assert a.matches("aaa") is True
+
+    a = QuantifyMaximum(String("a"), 2)
+    assert to_regex(a) == "(a){,2}"
+    assert a.matches("aa") is True
+    assert a.matches("aaa") is False
+
+    a = QuantifyBetween(String("a"), 1, 2)
+    assert to_regex(a) == "(a){1,2}"
+    assert a.matches("") is False
+    assert a.matches("a") is True
+    assert a.matches("aa") is True
+    assert a.matches("aaa") is False
+
+    with pytest.raises(TypeError, match="Cannot convert"):
+        to_regex(Term())
+
+
+def test_to_regex_combinations():
+    a = Sequence([Regex("dog|cat"), String("fish")])
+    assert to_regex(a) == "(dog|cat)fish"

From 9c27efb9be33bb86862cbef341ce2066ece8fe5a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Thu, 13 Feb 2025 18:54:43 +0100
Subject: [PATCH 167/505] Use `uv` in CI

---
 .github/workflows/tests.yml | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index f54a888fa..46d3d5ff6 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -32,7 +32,9 @@ jobs:
     - name: Set up test environment
       run: |
         python -m pip install --upgrade pip
-        pip install -e .[test]
+        pip install uv
+        uv venv
+        uv pip install -e .[test]
     - name: Create matrix id
       id: matrix-id
       env:
@@ -44,7 +46,7 @@ jobs:
         echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
-        pytest -x --cov=outlines
+        uv run pytest -x --cov=outlines
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
     - name: Upload coverage data

From 116ebff8a5dbd26825075c847a8790b710c43da1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 17 Feb 2025 22:58:42 +0100
Subject: [PATCH 168/505] Remove vLLM test dependency

vLLM can only currently be run on GPU (unless you want to go to extreme
lengths to make it work on CPU), and we thus cannot run the tests in CI.
We thus separate the test dependencies in "with GPU" and "without
GPU" (a subset of "with GPU") that the user has to pick manually.
---
 docs/community/contribute.md | 44 ++++++++++++++++++++++++------------
 pyproject.toml               |  2 +-
 2 files changed, 30 insertions(+), 16 deletions(-)

diff --git a/docs/community/contribute.md b/docs/community/contribute.md
index 6867263b4..21dc4581e 100644
--- a/docs/community/contribute.md
+++ b/docs/community/contribute.md
@@ -18,7 +18,7 @@ Note that the [issue tracker][issues] is only intended for actionable items. In
 
 First, [fork the repository on GitHub](https://github.com/dottxt-ai/outlines/fork) and clone the fork locally:
 
-```bash
+```shell
 git clone git@github.com/YourUserName/outlines.git
 cd outlines
 ```
@@ -27,7 +27,7 @@ Create a new virtual environment:
 
 *If you are using `uv`*:
 
-```bash
+```shell
 uv venv
 source .venv/bin/activate
 alias pip="uv pip" # ... or just remember to prepend any pip command with uv in the rest of this guide
@@ -35,27 +35,41 @@ alias pip="uv pip" # ... or just remember to prepend any pip command with uv in
 
 *If you are using `venv`*:
 
-```bash
+```shell
 python -m venv .venv
 source .venv/bin/activate
 ```
 
 *If you are using `conda`*:
 
-```bash
+```shell
 conda env create -f environment.yml
 ```
 
 Then install the dependencies in editable mode, and install the `pre-commit` hooks:
 
-```bash
+```shell
+python -m venv .venv
+source .venv/bin/activate
+```
+
+Then install the dependencies in editable mode, and install the pre-commit hooks:
+
+```shell
 pip install -e ".[test]"
 pre-commit install
 ```
+If you own a GPU and want to run the vLLM tests you will have to run:
+
+```shell
+pip install -e ".[test-gpu]"
+```
+
+instead.
 
 Outlines provides optional dependencies for different supported backends, which you can install with
 
-```bash
+```shell
 pip install ".[vllm]"
 ```
 
@@ -85,13 +99,13 @@ You will not have access to a GPU, but you'll be able to make basic contribution
 
 Run the tests:
 
-```bash
+```shell
 pytest
 ```
 
 And run the code style checks:
 
-```bash
+```shell
 pre-commit run --all-files
 ```
 
@@ -101,7 +115,7 @@ Outlines uses [asv](https://asv.readthedocs.io) for automated benchmark testing.
 
 You can run the benchmark test suite locally with the following command:
 
-```bash
+```shell
 asv run --config benchmarks/asv.conf.json
 ```
 
@@ -112,19 +126,19 @@ Caveats:
 
 #### Run a specific test:
 
-```bash
+```shell
 asv run --config benchmarks/asv.conf.json -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
 ```
 
 #### Profile a specific test:
 
-```bash
+```shell
 asv run --config benchmarks/asv.conf.json --profile -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
 ```
 
 #### Compare to `origin/main`
 
-```bash
+```shell
 get fetch origin
 asv continuous origin/main HEAD --config benchmarks/asv.conf.json
 ```
@@ -140,13 +154,13 @@ asv continuous origin/main HEAD --config benchmarks/asv.conf.json
 
 To work on the *documentation* you will need to install the related dependencies:
 
-```bash
+```shell
 pip install -r requirements-doc.txt
 ```
 
 To build the documentation and serve it locally, run the following command in the repository's root folder:
 
-```bash
+```shell
 mkdocs serve
 ```
 
@@ -157,7 +171,7 @@ It will be updated every time you make a change.
 
 Create a new branch on your fork, commit and push the changes:
 
-```bash
+```shell
 git checkout -b new-branch
 git add .
 git commit -m "Changes I made"
diff --git a/pyproject.toml b/pyproject.toml
index 2762db439..969290b07 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -69,12 +69,12 @@ test = [
     "huggingface_hub",
     "openai>=1.0.0",
     "datasets",
-    "vllm; sys_platform == 'linux'",
     "transformers",
     "pillow",
     "exllamav2",
     "jax",
 ]
+test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
 serve = [
     "vllm>=0.3.0",
     "uvicorn",

From f9e3af00ce96f13a26e2b955be7d8cd54b364489 Mon Sep 17 00:00:00 2001
From: Edoardo Abati <29585319+EdAbati@users.noreply.github.com>
Date: Thu, 20 Feb 2025 20:09:38 +0100
Subject: [PATCH 169/505] Various docs fixes (#1429)

Follow up of #1343

In this PR I added:
- all models to the API reference (we had only openai and transformers
before)
- fixes to the docstrings that were not following the numpy style
- fixes to docstrings wrongly rendered by mkdocstring

Thanks :)
---
 docs/api/models.md                     |  4 +---
 docs/community/contribute.md           |  7 -------
 outlines/models/exllamav2.py           |  8 +++++---
 outlines/models/llamacpp.py            | 12 ++++++------
 outlines/models/mlxlm.py               | 25 +++++++++++++++----------
 outlines/models/openai.py              |  5 ++---
 outlines/models/transformers.py        |  6 +++---
 outlines/models/transformers_vision.py |  4 ++--
 outlines/models/vllm.py                |  6 +++---
 outlines/prompts.py                    | 14 ++++++++++----
 10 files changed, 47 insertions(+), 44 deletions(-)

diff --git a/docs/api/models.md b/docs/api/models.md
index 27ad297fc..cb9497be1 100644
--- a/docs/api/models.md
+++ b/docs/api/models.md
@@ -1,3 +1 @@
-::: outlines.models.transformers
-
-::: outlines.models.openai
+::: outlines.models
diff --git a/docs/community/contribute.md b/docs/community/contribute.md
index 21dc4581e..78c713b18 100644
--- a/docs/community/contribute.md
+++ b/docs/community/contribute.md
@@ -48,13 +48,6 @@ conda env create -f environment.yml
 
 Then install the dependencies in editable mode, and install the `pre-commit` hooks:
 
-```shell
-python -m venv .venv
-source .venv/bin/activate
-```
-
-Then install the dependencies in editable mode, and install the pre-commit hooks:
-
 ```shell
 pip install -e ".[test]"
 pre-commit install
diff --git a/outlines/models/exllamav2.py b/outlines/models/exllamav2.py
index d2aa84b0d..efd70f96a 100644
--- a/outlines/models/exllamav2.py
+++ b/outlines/models/exllamav2.py
@@ -118,9 +118,11 @@ def reformat_output(
         self, output: Union[str, List[str]], sampling_parameters: SamplingParameters
     ):
         """
-        The purpose of this function is to reformat the output from exllamav2's output format to outline's output format
-        For exllamav2, it mainly accepts only a list or a string(they also do cfg sampling with tuples but we will ignore this for now)
-        The exllamav2's logic is
+        The purpose of this function is to reformat the output from exllamav2's output format to outline's output format.
+
+        For exllamav2, it mainly accepts only a list or a string(they also do cfg sampling with tuples but we will ignore this for now).
+        The exllamav2's logic is:
+
         1. If the prompt is a string, return a string. This is the same as outlines
         2. If a prompt is a list, return a list. This is not the same as outlines output in that if the list is only one element, the string is expected to be outputted.
         3. There is no such thing as num_samples, so the prompts had to be duplicated by num_samples times. Then, we had the function output a list of lists
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 904b193c4..7c7ad64df 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -248,8 +248,8 @@ def generate(
     ) -> str:
         """Generate text using `llama-cpp-python`.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         prompts
             A prompt or list of prompts.
         generation_parameters
@@ -302,8 +302,8 @@ def stream(
     ) -> Iterator[str]:
         """Stream text using `llama-cpp-python`.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         prompts
             A prompt or list of prompts.
         generation_parameters
@@ -372,8 +372,8 @@ def llamacpp(
     a path to the downloaded model. One can still load a local model
     by initializing `llama_cpp.Llama` directly.
 
-    Arguments
-    ---------
+    Parameters
+    ----------
     repo_id
         The name of the model repository.
     filename:
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index d8b7e032c..843107d66 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -49,8 +49,8 @@ def stream(
     ) -> Iterator[str]:
         """Generate text using `mlx_lm`.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         prompts
             A prompt or list of prompts.
         generation_parameters
@@ -63,6 +63,7 @@ def stream(
             An instance of `SamplingParameters`, a dataclass that contains
             the name of the sampler to use and related parameters as available
             in Outlines.
+
         Returns
         -------
         The generated text.
@@ -135,14 +136,18 @@ def generate_step(
 
         A generator producing token ids based on the given prompt from the model.
 
-            Args:
-                prompt (mx.array): The input prompt.
-                temp (float): The temperature for sampling, if 0 the argmax is used.
-                  Default: ``0``.
-                top_p (float, optional): Nulceus sampling, higher means model considers
-                  more less likely words.
-                sampler (str): The sampler string defined by SequenceGeneratorAdapter
-                logits_processor (OutlinesLogitsProcessor): Augment logits before sampling.
+        Parameters
+        ----------
+        prompt
+            The input prompt.
+        temp
+            The temperature for sampling, if 0 the argmax is used.
+        top_p
+            Nulceus sampling, higher means model considers more less likely words.
+        sampler
+            The sampler string defined by SequenceGeneratorAdapter
+        logits_processor
+            Augment logits before sampling.
         """
         import mlx.core as mx
         import mlx_lm
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 89c26f217..40ade1c25 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -20,11 +20,11 @@ class OpenAIConfig:
     properties that are specific to the OpenAI API. Not all these properties are
     supported by Outlines.
 
-    Properties
+    Parameters
     ----------
     model
         The name of the model. Available models can be found on OpenAI's website.
-    frequence_penalty
+    frequency_penalty
         Number between 2.0 and -2.0. Positive values penalize new tokens based on
         their existing frequency in the text,
     logit_bias
@@ -49,7 +49,6 @@ class OpenAIConfig:
         Number between 0 and 1. Parameter for nucleus sampling.
     user
         A unique identifier for the end-user.
-
     """
 
     model: str = ""
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 444492500..6b204ec6e 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -203,8 +203,8 @@ def generate(
     ) -> Union[str, List[str], List[List[str]]]:
         """Generate text using `transformers`.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         prompts
             A prompt or list of prompts.
         generation_parameters
@@ -304,7 +304,7 @@ def _get_generation_kwargs(
         sampling_parameters: SamplingParameters,
     ) -> dict:
         """
-        Conert outlines generation parameters into model.generate kwargs
+        Convert outlines generation parameters into model.generate kwargs
         """
         from transformers import GenerationConfig, LogitsProcessorList, set_seed
 
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
index 772645b80..c8a86536e 100644
--- a/outlines/models/transformers_vision.py
+++ b/outlines/models/transformers_vision.py
@@ -22,8 +22,8 @@ def generate(  # type: ignore
     ) -> Union[str, List[str], List[List[str]]]:
         """Generate text using `transformers`.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         prompts
             A prompt or list of prompts.
         media
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 778c27c6f..b9b035d1f 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -52,8 +52,8 @@ def generate(
     ):
         """Generate text using vLLM.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         prompts
             A prompt or list of prompts.
         generation_parameters
@@ -171,7 +171,7 @@ def load_lora(self, adapter_path: Optional[str]):
 def vllm(model_name: str, **vllm_model_params):
     """Load a vLLM model.
 
-    Arguments
+    Parameters
     ---------
     model_name
         The name of the model to load from the HuggingFace hub.
diff --git a/outlines/prompts.py b/outlines/prompts.py
index 1cc264226..b04f91856 100644
--- a/outlines/prompts.py
+++ b/outlines/prompts.py
@@ -125,6 +125,7 @@ def prompt(
     manipulation by providing some degree of encapsulation. It uses the `render`
     function internally to render templates.
 
+    ```pycon
     >>> import outlines
     >>>
     >>> @outlines.prompt
@@ -132,35 +133,40 @@ def prompt(
     ...    "I have a ${question}"
     ...
     >>> prompt = build_prompt("How are you?")
+    ```
 
     This API can also be helpful in an "agent" context where parts of the prompt
     are set when the agent is initialized and never modified later. In this situation
     we can partially apply the prompt function at initialization.
 
+    ```pycon
     >>> import outlines
     >>> import functools as ft
     ...
     >>> @outlines.prompt
     ... def solve_task(name: str, objective: str, task: str):
-    ...     '''Your name is {{name}}.
-    ..      Your overall objective is to {{objective}}.
+    ...     \"""Your name is {{name}}.
+    ...     Your overall objective is to {{objective}}.
     ...     Please solve the following task: {{task}}
-    ...     '''
+    ...     \"""
     ...
     >>> hal = ft.partial(solve_task, "HAL", "Travel to Jupiter")
+    ```
 
     Additional Jinja2 filters can be provided as keyword arguments to the decorator.
 
+    ```pycon
     >>> def reverse(s: str) -> str:
     ...     return s[::-1]
     ...
     >>> @outlines.prompt(filters={ 'reverse': reverse })
     ... def reverse_prompt(text):
-    ...     '''{{ text | reverse }}'''
+    ...     \"""{{ text | reverse }}\"""
     ...
     >>> prompt = reverse_prompt("Hello")
     >>> print(prompt)
     ... "olleH"
+    ```
 
     Returns
     -------

From 9c98de767c06fe27f0a0d544eb0ea0261fcef8fe Mon Sep 17 00:00:00 2001
From: deepybee <dpb@dottxt.co>
Date: Fri, 21 Feb 2025 16:27:49 +0100
Subject: [PATCH 170/505] Add Bluesky social media profile link to docs

This commit adds a link in the docs to the social media profile of dottxt on
the Bluesky social network alongside the existing link to the profile on X.
---
 docs/overrides/home.html | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/docs/overrides/home.html b/docs/overrides/home.html
index 6525ab062..de5b2258a 100644
--- a/docs/overrides/home.html
+++ b/docs/overrides/home.html
@@ -135,8 +135,11 @@ <h2 class="subtitle" style="font-weight: 500; padding-top: 1rem;">
                 </div>
                 <div class="md-buttons">
                     <p>
-                        <a href="https://twitter.com/dottxtai" title="Follow us on X" target="_blank">
-                            Follow us on X
+                        Follow us on <a href="https://twitter.com/dottxtai" title="Follow us on X" target="_blank">
+                            X
+                        </a> and
+                        <a href="https://bsky.app/profile/dottxtai.bsky.social" title="Follow us on Bluesky" target="_blank">
+                            Bluesky
                         </a>
                     </p>
                 </div>

From f25d3ab336fa1036af75d7ecc678475735bf1df7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:24:46 +0100
Subject: [PATCH 171/505] Misc format fixes

---
 outlines/caching.py               |  2 +-
 outlines/fsm/types.py             |  9 ++++++---
 outlines/models/openai.py         |  9 ++++++++-
 outlines/processors/structured.py |  1 +
 outlines/samplers.py              |  3 +--
 outlines/serve/serve.py           |  3 ++-
 outlines/types/__init__.py        | 11 ++++++++++-
 outlines/types/airports.py        |  1 +
 outlines/types/countries.py       |  1 +
 outlines/types/dsl.py             |  1 -
 tests/types/test_to_regex.py      | 17 ++++++++++++++++-
 11 files changed, 47 insertions(+), 11 deletions(-)

diff --git a/outlines/caching.py b/outlines/caching.py
index 0831c40bb..6882bac6b 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -51,7 +51,7 @@ def get_cache():
     """
     from outlines._version import __version__ as outlines_version  # type: ignore
 
-    outlines_cache_dir = os.environ.get('OUTLINES_CACHE_DIR')
+    outlines_cache_dir = os.environ.get("OUTLINES_CACHE_DIR")
     xdg_cache_home = os.environ.get("XDG_CACHE_HOME")
     home_dir = os.path.normpath(os.path.expanduser("~"))
     if outlines_cache_dir:
diff --git a/outlines/fsm/types.py b/outlines/fsm/types.py
index e5a1f8f47..f6409aa66 100644
--- a/outlines/fsm/types.py
+++ b/outlines/fsm/types.py
@@ -4,12 +4,15 @@
 
 from outlines.types import Regex, boolean as boolean_regex, date as date_regex
 from outlines.types import datetime as datetime_regex
-from outlines.types import integer as integer_regex, number as number_regex, time as time_regex
+from outlines.types import (
+    integer as integer_regex,
+    number as number_regex,
+    time as time_regex,
+)
 
 
 class FormatFunction(Protocol):
-    def __call__(self, sequence: str) -> Any:
-        ...
+    def __call__(self, sequence: str) -> Any: ...
 
 
 def python_types_to_regex(python_type: Type) -> Tuple[Regex, FormatFunction]:
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 40ade1c25..e46107bd4 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,4 +1,5 @@
 """Integration with OpenAI's API."""
+
 import copy
 import functools
 from dataclasses import asdict, dataclass, field, replace
@@ -139,7 +140,13 @@ def __call__(
         if samples is None:
             samples = self.config.n
 
-        config = replace(self.config, max_tokens=max_tokens, temperature=temperature, n=samples, stop=stop_at)  # type: ignore
+        config = replace(
+            self.config,
+            max_tokens=max_tokens,
+            temperature=temperature,
+            n=samples,
+            stop=stop_at,
+        )  # type: ignore
 
         response, prompt_tokens, completion_tokens = generate_chat(
             prompt, system_prompt, self.client, config
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 583fcc98f..8ce69c32a 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -23,6 +23,7 @@
 See the License for the specific language governing permissions and
 limitations under the License.
 """
+
 import math
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
 
diff --git a/outlines/samplers.py b/outlines/samplers.py
index 3ab1728fc..3fef673b1 100644
--- a/outlines/samplers.py
+++ b/outlines/samplers.py
@@ -14,8 +14,7 @@ def __call__(
         next_token_logits: "torch.DoubleTensor",
         sequence_weights: "torch.DoubleTensor",
         rng: "torch.Generator",
-    ) -> "torch.DoubleTensor":
-        ...
+    ) -> "torch.DoubleTensor": ...
 
 
 @dataclass(frozen=True)
diff --git a/outlines/serve/serve.py b/outlines/serve/serve.py
index 998fbc459..61d3ed7af 100644
--- a/outlines/serve/serve.py
+++ b/outlines/serve/serve.py
@@ -78,7 +78,8 @@ async def generate(request: Request) -> Response:
         logits_processors = []
 
     sampling_params = SamplingParams(
-        **request_dict, logits_processors=logits_processors  # type: ignore
+        **request_dict,
+        logits_processors=logits_processors,  # type: ignore
     )
     request_id = random_uuid()
 
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 9511720af..dbd8f507b 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,7 +1,16 @@
 from enum import Enum
 
 from . import airports, countries, locale
-from outlines.types.dsl import Regex, json_schema, one_or_more, optional, regex, repeat, zero_or_more, times
+from outlines.types.dsl import (
+    Regex,
+    json_schema,
+    one_or_more,
+    optional,
+    regex,
+    repeat,
+    zero_or_more,
+    times,
+)
 
 # Python types
 integer = Regex(r"[+-]?(0|[1-9][0-9]*)")
diff --git a/outlines/types/airports.py b/outlines/types/airports.py
index ec0ef72bd..6e3d011b4 100644
--- a/outlines/types/airports.py
+++ b/outlines/types/airports.py
@@ -1,4 +1,5 @@
 """Generate valid airport codes."""
+
 from enum import Enum
 
 import airportsdata
diff --git a/outlines/types/countries.py b/outlines/types/countries.py
index c612640b4..96be735d3 100644
--- a/outlines/types/countries.py
+++ b/outlines/types/countries.py
@@ -1,4 +1,5 @@
 """Generate valid country codes and names."""
+
 from enum import Enum
 
 from iso3166 import countries
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 86feabca0..94b4d26d9 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -69,7 +69,6 @@ def __get_pydantic_core_schema__(
     def __get_pydantic_json_schema__(
         self, core_schema: cs.CoreSchema, handler: GetJsonSchemaHandler
     ) -> JsonSchemaValue:
-
         return {"type": "string", "pattern": to_regex(self)}
 
     def validate(self, value: str) -> str:
diff --git a/tests/types/test_to_regex.py b/tests/types/test_to_regex.py
index 6cb566fc5..4b0403ac6 100644
--- a/tests/types/test_to_regex.py
+++ b/tests/types/test_to_regex.py
@@ -1,7 +1,22 @@
 import pytest
 
 
-from outlines.types.dsl import String, Regex, JsonSchema, KleeneStar, KleenePlus, QuantifyBetween, QuantifyExact, QuantifyMaximum, QuantifyMinimum, Sequence, Alternatives, Optional, Term, to_regex
+from outlines.types.dsl import (
+    String,
+    Regex,
+    JsonSchema,
+    KleeneStar,
+    KleenePlus,
+    QuantifyBetween,
+    QuantifyExact,
+    QuantifyMaximum,
+    QuantifyMinimum,
+    Sequence,
+    Alternatives,
+    Optional,
+    Term,
+    to_regex,
+)
 
 
 def test_to_regex_simple():

From 41eacc5db717e985f9d276468c74cbb005b150c2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:25:08 +0100
Subject: [PATCH 172/505] Refactor the `prompts.py` file

I replace a couple `classmethod`s with simple function, and added
the default filter to the Jinja environment built when reading
the template from a file.
---
 outlines/prompts.py   | 108 +++++++++++++++++++++++++-----------------
 tests/test_prompts.py |   6 +--
 2 files changed, 68 insertions(+), 46 deletions(-)

diff --git a/outlines/prompts.py b/outlines/prompts.py
index b04f91856..257dc3007 100644
--- a/outlines/prompts.py
+++ b/outlines/prompts.py
@@ -14,10 +14,10 @@
 
 @dataclass
 class Prompt:
-    """Represents a prompt function.
+    """Represents a prompt template.
 
     We return a `Prompt` class instead of a simple function so the
-    template defined in prompt functions can be accessed.
+    template can be accessed by callers.
 
     """
 
@@ -41,8 +41,7 @@ def __call__(self, *args, **kwargs) -> str:
 
     @classmethod
     def from_str(cls, content: str, filters: Dict[str, Callable] = {}):
-        """
-        Create an instance of the class from a string.
+        """Create a `Prompt` instance from a string containing a Jinja template.
 
         Parameters
         ----------
@@ -53,12 +52,11 @@ def from_str(cls, content: str, filters: Dict[str, Callable] = {}):
         -------
         An instance of the class with the provided content as a template.
         """
-        return cls(cls._template_from_str(content, filters), None)
+        return cls(build_template_from_str(content, filters), None)
 
     @classmethod
     def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
-        """
-        Create a Prompt instance from a file containing a Jinja template.
+        """Create a `Prompt` instance from a file containing a Jinja template.
 
         Note: This method does not allow to include and inheritance to reference files
         that are outside the folder or subfolders of the file given to `from_file`.
@@ -75,44 +73,38 @@ def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
         """
         # We don't use a `Signature` here because it seems not feasible to infer one from a Jinja2 environment that is
         # split across multiple files (since e.g. we support features like Jinja2 includes and template inheritance)
-        return cls(cls._template_from_file(path, filters), None)
+        return cls(build_template_from_file(path, filters), None)
 
-    @classmethod
-    def _template_from_str(
-        _, content: str, filters: Dict[str, Callable] = {}
-    ) -> jinja2.Template:
-        # Dedent, and remove extra linebreak
-        cleaned_template = inspect.cleandoc(content)
-
-        # Add linebreak if there were any extra linebreaks that
-        # `cleandoc` would have removed
-        ends_with_linebreak = content.replace(" ", "").endswith("\n\n")
-        if ends_with_linebreak:
-            cleaned_template += "\n"
-
-        # Remove extra whitespaces, except those that immediately follow a newline symbol.
-        # This is necessary to avoid introducing whitespaces after backslash `\` characters
-        # used to continue to the next line without linebreak.
-        cleaned_template = re.sub(r"(?![\r\n])(\b\s+)", " ", cleaned_template)
-
-        env = create_jinja_env(None, filters)
-        env.filters["name"] = get_fn_name
-        env.filters["description"] = get_fn_description
-        env.filters["source"] = get_fn_source
-        env.filters["signature"] = get_fn_signature
-        env.filters["schema"] = get_schema
-        env.filters["args"] = get_fn_args
-
-        return env.from_string(cleaned_template)
 
-    @classmethod
-    def _template_from_file(
-        _, path: Path, filters: Dict[str, Callable] = {}
-    ) -> jinja2.Template:
-        file_directory = os.path.dirname(os.path.abspath(path))
-        env = create_jinja_env(jinja2.FileSystemLoader(file_directory), filters)
+def build_template_from_str(
+    content: str, filters: Dict[str, Callable] = {}
+) -> jinja2.Template:
+    # Dedent, and remove extra linebreak
+    cleaned_template = inspect.cleandoc(content)
+
+    # Add linebreak if there were any extra linebreaks that
+    # `cleandoc` would have removed
+    ends_with_linebreak = content.replace(" ", "").endswith("\n\n")
+    if ends_with_linebreak:
+        cleaned_template += "\n"
+
+    # Remove extra whitespaces, except those that immediately follow a newline symbol.
+    # This is necessary to avoid introducing whitespaces after backslash `\` characters
+    # used to continue to the next line without linebreak.
+    cleaned_template = re.sub(r"(?![\r\n])(\b\s+)", " ", cleaned_template)
+
+    env = create_jinja_env(None, filters)
 
-        return env.get_template(os.path.basename(path))
+    return env.from_string(cleaned_template)
+
+
+def build_template_from_file(
+    path: Path, filters: Dict[str, Callable] = {}
+) -> jinja2.Template:
+    file_directory = os.path.dirname(os.path.abspath(path))
+    env = create_jinja_env(jinja2.FileSystemLoader(file_directory), filters)
+
+    return env.get_template(os.path.basename(path))
 
 
 def prompt(
@@ -184,7 +176,7 @@ def prompt(
     if docstring is None:
         raise TypeError("Could not find a template in the function's docstring.")
 
-    template = Prompt._template_from_str(cast(str, docstring), filters)
+    template = build_template_from_str(cast(str, docstring), filters)
 
     return Prompt(template, signature)
 
@@ -192,6 +184,27 @@ def prompt(
 def create_jinja_env(
     loader: Optional[jinja2.BaseLoader], filters: Dict[str, Callable]
 ) -> jinja2.Environment:
+    """Create a new Jinja environment.
+
+    The Jinja environment is loaded with a set of pre-defined filters:
+    - `name`: get the name of a function
+    - `description`: get a function's docstring
+    - `source`: get a function's source code
+    - `signature`: get a function's signature
+    - `args`: get a function's arguments
+    - `schema`: isplay a JSON Schema
+
+    Users may pass additional filters, and/or override existing ones.
+
+    Arguments
+    ---------
+    loader
+       An optional `BaseLoader` instance
+    filters
+       A dictionary of filters, map between the filter's name and the
+       corresponding function.
+
+    """
     env = jinja2.Environment(
         loader=loader,
         trim_blocks=True,
@@ -200,6 +213,15 @@ def create_jinja_env(
         undefined=jinja2.StrictUndefined,
     )
 
+    env.filters["name"] = get_fn_name
+    env.filters["description"] = get_fn_description
+    env.filters["source"] = get_fn_source
+    env.filters["signature"] = get_fn_signature
+    env.filters["schema"] = get_schema
+    env.filters["args"] = get_fn_args
+
+    # The filters passed by the user may override the
+    # pre-defined filters.
     for name, filter_fn in filters.items():
         env.filters[name] = filter_fn
 
diff --git a/tests/test_prompts.py b/tests/test_prompts.py
index f59c04ac0..c29348b8e 100644
--- a/tests/test_prompts.py
+++ b/tests/test_prompts.py
@@ -6,11 +6,11 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.prompts import Prompt
+from outlines.prompts import Prompt, build_template_from_str
 
 
 def render(content: str, **kwargs):
-    template = Prompt._template_from_str(content)
+    template = build_template_from_str(content)
     return template.render(kwargs)
 
 
@@ -418,5 +418,5 @@ def test_template_from_str_with_extra_linebreaks():
 
 
     """
-    template = Prompt._template_from_str(content)
+    template = build_template_from_str(content)
     assert template.render(name="World") == "Hello, World!\n"

From f35b57956094b08e52bf55d52d345a034f62a6c5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:50:04 +0100
Subject: [PATCH 173/505] Rename `Prompt` to `Template`

The `Prompt` class really represents a prompt *template* and not a
prompt. I thus renamed it to `Template` to make it less confusing to the
user.
---
 README.md                                    | 21 ++++++++++++++++++++
 docs/api/prompts.md                          |  1 -
 docs/api/templates.md                        |  1 +
 outlines/__init__.py                         |  2 +-
 outlines/function.py                         |  4 ++--
 outlines/{prompts.py => templates.py}        | 16 +++++++--------
 tests/{test_prompts.py => test_templates.py} |  6 +++---
 7 files changed, 36 insertions(+), 15 deletions(-)
 delete mode 100644 docs/api/prompts.md
 create mode 100644 docs/api/templates.md
 rename outlines/{prompts.py => templates.py} (95%)
 rename tests/{test_prompts.py => test_templates.py} (98%)

diff --git a/README.md b/README.md
index b3cde1dd1..83740fcec 100644
--- a/README.md
+++ b/README.md
@@ -412,6 +412,27 @@ prompt = labelling("Just awesome", examples)
 answer = outlines.generate.text(model)(prompt, max_tokens=100)
 ```
 
+You can also load a template from a text file. Assume you have the following prompt template defined in `prompt.txt`:
+
+``` text
+You are a sentiment-labelling assistant.
+
+{% for example in examples %}
+{{ example[0] }} // {{ example[1] }}
+{% endfor %}
+{{ to_label }} //
+```
+
+You can load it with:
+
+``` python
+import outlines
+
+labelling = outlines.Template.from_file("prompt.txt")
+prompt = labelling("Just awesome", examples)
+
+```
+
 ## Join us
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
diff --git a/docs/api/prompts.md b/docs/api/prompts.md
deleted file mode 100644
index 9d28f838c..000000000
--- a/docs/api/prompts.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.prompts
diff --git a/docs/api/templates.md b/docs/api/templates.md
new file mode 100644
index 000000000..6150512d4
--- /dev/null
+++ b/docs/api/templates.md
@@ -0,0 +1 @@
+::: outlines.templates
diff --git a/outlines/__init__.py b/outlines/__init__.py
index eeba78dc7..628c88f0e 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -8,7 +8,7 @@
 from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function
-from outlines.prompts import Prompt, prompt
+from outlines.templates import Template, prompt
 
 __all__ = [
     "clear_cache",
diff --git a/outlines/function.py b/outlines/function.py
index 48577be8f..aff21d68f 100644
--- a/outlines/function.py
+++ b/outlines/function.py
@@ -8,7 +8,7 @@
 
 if TYPE_CHECKING:
     from outlines.generate.api import SequenceGenerator
-    from outlines.prompts import Prompt
+    from outlines.templates import Template
 
 
 @dataclass
@@ -22,7 +22,7 @@ class Function:
 
     """
 
-    prompt_template: "Prompt"
+    prompt_template: "Template"
     schema: Union[str, Callable, object]
     model_name: str
     generator: Optional["SequenceGenerator"] = None
diff --git a/outlines/prompts.py b/outlines/templates.py
similarity index 95%
rename from outlines/prompts.py
rename to outlines/templates.py
index 257dc3007..56c4bc7c0 100644
--- a/outlines/prompts.py
+++ b/outlines/templates.py
@@ -13,10 +13,10 @@
 
 
 @dataclass
-class Prompt:
+class Template:
     """Represents a prompt template.
 
-    We return a `Prompt` class instead of a simple function so the
+    We return a `Template` class instead of a simple function so the
     template can be accessed by callers.
 
     """
@@ -41,7 +41,7 @@ def __call__(self, *args, **kwargs) -> str:
 
     @classmethod
     def from_str(cls, content: str, filters: Dict[str, Callable] = {}):
-        """Create a `Prompt` instance from a string containing a Jinja template.
+        """Create a `Template` instance from a string containing a Jinja template.
 
         Parameters
         ----------
@@ -56,7 +56,7 @@ def from_str(cls, content: str, filters: Dict[str, Callable] = {}):
 
     @classmethod
     def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
-        """Create a `Prompt` instance from a file containing a Jinja template.
+        """Create a `Template` instance from a file containing a Jinja template.
 
         Note: This method does not allow to include and inheritance to reference files
         that are outside the folder or subfolders of the file given to `from_file`.
@@ -68,8 +68,8 @@ def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
 
         Returns
         -------
-        Prompt
-            An instance of the Prompt class with the template loaded from the file.
+        Template
+            An instance of the Template class with the template loaded from the file.
         """
         # We don't use a `Signature` here because it seems not feasible to infer one from a Jinja2 environment that is
         # split across multiple files (since e.g. we support features like Jinja2 includes and template inheritance)
@@ -162,7 +162,7 @@ def prompt(
 
     Returns
     -------
-    A `Prompt` callable class which will render the template when called.
+    A `Template` callable class which will render the template when called.
 
     """
     if fn is None:
@@ -178,7 +178,7 @@ def prompt(
 
     template = build_template_from_str(cast(str, docstring), filters)
 
-    return Prompt(template, signature)
+    return Template(template, signature)
 
 
 def create_jinja_env(
diff --git a/tests/test_prompts.py b/tests/test_templates.py
similarity index 98%
rename from tests/test_prompts.py
rename to tests/test_templates.py
index c29348b8e..1b16b0157 100644
--- a/tests/test_prompts.py
+++ b/tests/test_templates.py
@@ -6,7 +6,7 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.prompts import Prompt, build_template_from_str
+from outlines.templates import Template, build_template_from_str
 
 
 def render(content: str, **kwargs):
@@ -380,7 +380,7 @@ def temp_prompt_file():
 
 
 def test_prompt_from_file(temp_prompt_file):
-    prompt = Prompt.from_file(temp_prompt_file)
+    prompt = Template.from_file(temp_prompt_file)
     assert prompt.signature is None
     examples = [
         {"question": "What is the capital of France?", "answer": "Paris"},
@@ -407,7 +407,7 @@ def test_prompt_from_str():
     content = """
     Hello, {{ name }}!
     """
-    prompt = Prompt.from_str(content)
+    prompt = Template.from_str(content)
     assert prompt.signature is None
     assert prompt(name="World") == "Hello, World!"
 

From 2c8b40db64c77f542de99fbe24190beb1cc57824 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:58:57 +0100
Subject: [PATCH 174/505] Add deprecation warning for the `prompt` decorator

---
 README.md               |  55 ++++------
 outlines/templates.py   |  11 ++
 tests/test_function.py  |  21 ++--
 tests/test_templates.py | 236 +++++++++++++++++++++-------------------
 4 files changed, 166 insertions(+), 157 deletions(-)

diff --git a/README.md b/README.md
index 83740fcec..cb57be0ae 100644
--- a/README.md
+++ b/README.md
@@ -379,40 +379,12 @@ print(result)
 ## Prompting
 
 Building prompts can get messy. **Outlines** makes it easier to write and manage
-prompts by encapsulating templates inside "template functions".
+prompts by encapsulating templates inside "template functions". Template
+functions use the Jinja2 templating engine to help build complex prompts in a
+concise manner.
 
-These functions make it possible to neatly separate the prompt logic from the
-general program logic; they can be imported from other modules and libraries.
-
-Template functions require no superfluous abstraction, they use the Jinja2
-templating engine to help build complex prompts in a concise manner:
-
-``` python
-import outlines
-
-examples = [
-    ("The food was disgusting", "Negative"),
-    ("We had a fantastic night", "Positive"),
-    ("Recommended", "Positive"),
-    ("The waiter was rude", "Negative")
-]
-
-@outlines.prompt
-def labelling(to_label, examples):
-    """You are a sentiment-labelling assistant.
-
-    {% for example in examples %}
-    {{ example[0] }} // {{ example[1] }}
-    {% endfor %}
-    {{ to_label }} //
-    """
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
-prompt = labelling("Just awesome", examples)
-answer = outlines.generate.text(model)(prompt, max_tokens=100)
-```
-
-You can also load a template from a text file. Assume you have the following prompt template defined in `prompt.txt`:
+Template functions are created by loading a Jinja2 template from a text file.
+Assume you have the following prompt template defined in `prompt.txt`:
 
 ``` text
 You are a sentiment-labelling assistant.
@@ -423,16 +395,29 @@ You are a sentiment-labelling assistant.
 {{ to_label }} //
 ```
 
-You can load it with:
+You can then load it and call it with:
 
 ``` python
 import outlines
 
+examples = [
+    ("The food was disgusting", "Negative"),
+    ("We had a fantastic night", "Positive"),
+    ("Recommended", "Positive"),
+    ("The waiter was rude", "Negative")
+]
+
 labelling = outlines.Template.from_file("prompt.txt")
 prompt = labelling("Just awesome", examples)
-
 ```
 
+This helps:
+
+- Keep content separate from the code
+- Design "white space perfect" prompts
+
+It is more maintainable and means prompts can be versioned separately from the code.
+
 ## Join us
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
diff --git a/outlines/templates.py b/outlines/templates.py
index 56c4bc7c0..a1fbc1932 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -7,6 +7,7 @@
 from dataclasses import dataclass
 from pathlib import Path
 from typing import Any, Callable, Dict, Optional, Type, cast
+import warnings
 
 import jinja2
 import pydantic
@@ -165,6 +166,16 @@ def prompt(
     A `Template` callable class which will render the template when called.
 
     """
+    warnings.warn(
+        "The @prompt decorator is deprecated and will be removed in outlines 1.1.0. "
+        "Instead of using docstring templates, please use Template.from_file() to "
+        "load your prompts from separate template files, or a simple Python function "
+        "that returns text. This helps keep prompt content separate from code and is "
+        "more maintainable.",
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
     if fn is None:
         return lambda fn: prompt(fn, cast(Dict[str, Callable], filters))
 
diff --git a/tests/test_function.py b/tests/test_function.py
index 62f7ea29f..d66f3911c 100644
--- a/tests/test_function.py
+++ b/tests/test_function.py
@@ -8,9 +8,10 @@
 
 
 def test_function_basic():
-    @outlines.prompt
-    def test_template(text: str):
-        """{{ text }}"""
+    with pytest.deprecated_call(match="The @prompt decorator"):
+        @outlines.prompt
+        def test_template(text: str):
+            """{{ text }}"""
 
     class Foo(BaseModel):
         id: int
@@ -102,10 +103,11 @@ class User(BaseModel):
 )
     """
 
-    fn = extract_function_from_file(content, "function")
-    assert (
-        str(type(fn)) == "<class 'outlines.function.Function'>"
-    )  # because imported via `exec`
+    with pytest.deprecated_call(match="The @prompt decorator"):
+        fn = extract_function_from_file(content, "function")
+        assert (
+            str(type(fn)) == "<class 'outlines.function.Function'>"
+        )  # because imported via `exec`
 
 
 def test_extract_function_from_file_no_function():
@@ -129,5 +131,6 @@ class User(BaseModel):
 )
     """
 
-    with pytest.raises(AttributeError, match="Could not find"):
-        extract_function_from_file(content, "function")
+    with pytest.deprecated_call(match="The @prompt decorator"):
+        with pytest.raises(AttributeError, match="Could not find"):
+            extract_function_from_file(content, "function")
diff --git a/tests/test_templates.py b/tests/test_templates.py
index 1b16b0157..06727eab1 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -113,60 +113,63 @@ def test_render_jinja():
 
 
 def test_prompt_basic():
-    @outlines.prompt
-    def test_tpl(variable):
-        """{{variable}} test"""
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def test_tpl(variable):
+            """{{variable}} test"""
 
-    assert list(test_tpl.signature.parameters) == ["variable"]
+        assert list(test_tpl.signature.parameters) == ["variable"]
 
-    with pytest.raises(TypeError):
-        test_tpl(v="test")
+        with pytest.raises(TypeError):
+            test_tpl(v="test")
 
-    p = test_tpl("test")
-    assert p == "test test"
+        p = test_tpl("test")
+        assert p == "test test"
 
-    p = test_tpl(variable="test")
-    assert p == "test test"
+        p = test_tpl(variable="test")
+        assert p == "test test"
 
-    @outlines.prompt
-    def test_single_quote_tpl(variable):
-        "${variable} test"
+        @outlines.prompt
+        def test_single_quote_tpl(variable):
+            "${variable} test"
 
-    assert list(test_single_quote_tpl.signature.parameters) == ["variable"]
+        assert list(test_single_quote_tpl.signature.parameters) == ["variable"]
 
-    p = test_tpl("test")
-    assert p == "test test"
+        p = test_tpl("test")
+        assert p == "test test"
 
 
 def test_prompt_kwargs():
-    @outlines.prompt
-    def test_kwarg_tpl(var, other_var="other"):
-        """{{var}} and {{other_var}}"""
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def test_kwarg_tpl(var, other_var="other"):
+            """{{var}} and {{other_var}}"""
 
-    assert list(test_kwarg_tpl.signature.parameters) == ["var", "other_var"]
+        assert list(test_kwarg_tpl.signature.parameters) == ["var", "other_var"]
 
-    p = test_kwarg_tpl("test")
-    assert p == "test and other"
+        p = test_kwarg_tpl("test")
+        assert p == "test and other"
 
-    p = test_kwarg_tpl("test", other_var="kwarg")
-    assert p == "test and kwarg"
+        p = test_kwarg_tpl("test", other_var="kwarg")
+        assert p == "test and kwarg"
 
-    p = test_kwarg_tpl("test", "test")
-    assert p == "test and test"
+        p = test_kwarg_tpl("test", "test")
+        assert p == "test and test"
 
 
 def test_no_prompt():
-    with pytest.raises(TypeError, match="template"):
+    with pytest.deprecated_call():
+        with pytest.raises(TypeError, match="template"):
 
-        @outlines.prompt
-        def test_empty(variable):
-            pass
+            @outlines.prompt
+            def test_empty(variable):
+                pass
 
-    with pytest.raises(TypeError, match="template"):
+        with pytest.raises(TypeError, match="template"):
 
-        @outlines.prompt
-        def test_only_code(variable):
-            return variable
+            @outlines.prompt
+            def test_only_code(variable):
+                return variable
 
 
 def test_prompt_function():
@@ -180,41 +183,44 @@ def with_description():
         """
         pass
 
-    @outlines.prompt
-    def name_description_ppt(fn):
-        """
-        {{fn|name}}: {{fn|description}}
-        """
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def name_description_ppt(fn):
+            """
+            {{fn|name}}: {{fn|description}}
+            """
 
-    rendered = name_description_ppt(empty_fn)
-    assert rendered == "empty_fn: "
+        rendered = name_description_ppt(empty_fn)
+        assert rendered == "empty_fn: "
 
-    rendered = name_description_ppt(with_description)
-    assert rendered == "with_description: A description."
+        rendered = name_description_ppt(with_description)
+        assert rendered == "with_description: A description."
 
     def with_signature(one: int, two: List[str], three: float = 1.0):
         pass
 
-    @outlines.prompt
-    def name_signature_ppt(fn):
-        """
-        {{fn|name}}: {{fn|signature}}
-        """
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def name_signature_ppt(fn):
+            """
+            {{fn|name}}: {{fn|signature}}
+            """
 
-    rendered = name_signature_ppt(with_signature)
-    assert rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
+        rendered = name_signature_ppt(with_signature)
+        assert rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
 
     def test_function_call(one, two=2):
         return one + two
 
-    @outlines.prompt
-    def source_ppt(fn):
-        """
-        {{fn|source}}
-        """
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def source_ppt(fn):
+            """
+            {{fn|source}}
+            """
 
-    rendered = source_ppt(test_function_call)
-    assert rendered == "def test_function_call(one, two=2):\n    return one + two\n"
+        rendered = source_ppt(test_function_call)
+        assert rendered == "def test_function_call(one, two=2):\n    return one + two\n"
 
 
 def test_prompt_pydantic_response():
@@ -222,43 +228,45 @@ class SimpleResponse(BaseModel):
         one: str = Field(description="a description")
         two: str
 
-    @outlines.prompt
-    def source_ppt(model):
-        "{{model | schema }}"
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def source_ppt(model):
+            "{{model | schema }}"
 
-    prompt = source_ppt(SimpleResponse)
-    assert prompt == '{\n  "one": "a description",\n  "two": "<two>"\n}'
+        prompt = source_ppt(SimpleResponse)
+        assert prompt == '{\n  "one": "a description",\n  "two": "<two>"\n}'
 
-    class NestedResponse(BaseModel):
-        answer: str
-        thought: SimpleResponse
+        class NestedResponse(BaseModel):
+            answer: str
+            thought: SimpleResponse
 
-    prompt = source_ppt(NestedResponse)
-    assert (
-        prompt
-        == '{\n  "answer": "<answer>",\n  "thought": {\n    "one": "a description",\n    "two": "<two>"\n  }\n}'
-    )
+        prompt = source_ppt(NestedResponse)
+        assert (
+            prompt
+            == '{\n  "answer": "<answer>",\n  "thought": {\n    "one": "a description",\n    "two": "<two>"\n  }\n}'
+        )
 
-    class ConvolutedResponse(BaseModel):
-        part_one: NestedResponse
-        part_two: SimpleResponse
+        class ConvolutedResponse(BaseModel):
+            part_one: NestedResponse
+            part_two: SimpleResponse
 
-    prompt = source_ppt(ConvolutedResponse)
-    assert (
-        prompt
-        == '{\n  "part_one": {\n    "answer": "<answer>",\n    "thought": {\n      "one": "a description",\n      "two": "<two>"\n    }\n  },\n  "part_two": {\n    "one": "a description",\n    "two": "<two>"\n  }\n}'
-    )
+        prompt = source_ppt(ConvolutedResponse)
+        assert (
+            prompt
+            == '{\n  "part_one": {\n    "answer": "<answer>",\n    "thought": {\n      "one": "a description",\n      "two": "<two>"\n    }\n  },\n  "part_two": {\n    "one": "a description",\n    "two": "<two>"\n  }\n}'
+        )
 
 
 def test_prompt_dict_response():
     response = {"one": "a description", "two": ""}
 
-    @outlines.prompt
-    def source_ppt(model):
-        "{{model | schema }}"
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def source_ppt(model):
+            "{{model | schema }}"
 
-    prompt = source_ppt(response)
-    assert prompt == '{\n  "one": "a description",\n  "two": ""\n}'
+        prompt = source_ppt(response)
+        assert prompt == '{\n  "one": "a description",\n  "two": ""\n}'
 
 
 def test_prompt_args():
@@ -297,45 +305,47 @@ def with_all(
     ):
         pass
 
-    @outlines.prompt
-    def args_prompt(fn):
-        """args: {{ fn | args }}"""
-
-    assert args_prompt(no_args) == "args: "
-    assert args_prompt(with_args) == "args: x, y, z"
-    assert (
-        args_prompt(with_annotations)
-        == "args: x: bool, y: str, z: Dict[int, List[str]]"
-    )
-    assert (
-        args_prompt(with_defaults)
-        == "args: x=True, y='Hi', z={4: ['I', 'love', 'outlines']}"
-    )
-    assert (
-        args_prompt(with_annotations_and_defaults)
-        == "args: x: bool = True, y: str = 'Hi', z: Dict[int, List[str]] = {4: ['I', 'love', 'outlines']}"
-    )
-    assert (
-        args_prompt(with_all)
-        == "args: x1, y1, z1, x2: bool, y2: str, z2: Dict[int, List[str]], x3=True, y3='Hi', z3={4: ['I', 'love', 'outlines']}, x4: bool = True, y4: str = 'Hi', z4: Dict[int, List[str]] = {4: ['I', 'love', 'outlines']}"
-    )
+    with pytest.deprecated_call():
+        @outlines.prompt
+        def args_prompt(fn):
+            """args: {{ fn | args }}"""
+
+        assert args_prompt(no_args) == "args: "
+        assert args_prompt(with_args) == "args: x, y, z"
+        assert (
+            args_prompt(with_annotations)
+            == "args: x: bool, y: str, z: Dict[int, List[str]]"
+        )
+        assert (
+            args_prompt(with_defaults)
+            == "args: x=True, y='Hi', z={4: ['I', 'love', 'outlines']}"
+        )
+        assert (
+            args_prompt(with_annotations_and_defaults)
+            == "args: x: bool = True, y: str = 'Hi', z: Dict[int, List[str]] = {4: ['I', 'love', 'outlines']}"
+        )
+        assert (
+            args_prompt(with_all)
+            == "args: x1, y1, z1, x2: bool, y2: str, z2: Dict[int, List[str]], x3=True, y3='Hi', z3={4: ['I', 'love', 'outlines']}, x4: bool = True, y4: str = 'Hi', z4: Dict[int, List[str]] = {4: ['I', 'love', 'outlines']}"
+        )
 
 
 def test_prompt_with_additional_filters():
     def reverse(s: str) -> str:
         return s[::-1]
 
-    @outlines.prompt(filters=dict(reverse=reverse))
-    def test_tpl(variable):
-        """{{ variable | reverse }} test"""
+    with pytest.deprecated_call():
+        @outlines.prompt(filters=dict(reverse=reverse))
+        def test_tpl(variable):
+            """{{ variable | reverse }} test"""
 
-    assert list(test_tpl.signature.parameters) == ["variable"]
+        assert list(test_tpl.signature.parameters) == ["variable"]
 
-    p = test_tpl("test")
-    assert p == "tset test"
+        p = test_tpl("test")
+        assert p == "tset test"
 
-    p = test_tpl(variable="example")
-    assert p == "elpmaxe test"
+        p = test_tpl(variable="example")
+        assert p == "elpmaxe test"
 
 
 @pytest.fixture

From ce27432021b502a4db1ec8740caac0b5c3c8e278 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 24 Feb 2025 09:42:11 +0100
Subject: [PATCH 175/505] Improve the README

---
 README.md | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/README.md b/README.md
index cb57be0ae..01d7b9c98 100644
--- a/README.md
+++ b/README.md
@@ -26,17 +26,12 @@ First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/la
 
 ## Features
 
-- [x] 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
-- [x] 🖍️ Simple and powerful prompting primitives based on the [Jinja templating engine](https://jinja.palletsprojects.com/)
-- [x] 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
-- [x] ⚡ Fast [regex-structured generation](#efficient-regex-structured-generation)
-- [x] 🔥 Fast [JSON generation](#efficient-json-generation-following-a-pydantic-model) following a JSON schema or a Pydantic model
-- [x] 📝 [Grammar-structured generation](#using-context-free-grammars-to-guide-generation)
-- [x] 🐍 Interleave completions with loops, conditionals, and custom Python functions
-- [x] 💾 Caching of generations
-- [x] 🗂️ Batch inference
-- [x] 🎲 Sample with the greedy, multinomial and beam search algorithms (and more to come!)
-- [x] 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/latest/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
+- 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
+- 🔥 Fast [JSON generation](#efficient-json-generation-following-a-pydantic-model) following a JSON schema or a Pydantic model
+- 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
+- 📝 Generate text that follows a [regex](#efficient-regex-structured-generation) or a [context-free grammar](#using-context-free-grammars-to-guide-generation)
+- 🖍️ Simple and powerful prompting primitives based on the [Jinja templating engine](https://jinja.palletsprojects.com/)
+- 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/latest/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
 
 
 Outlines  has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!

From eced2ceb275f2edfffc8ace64b70b5d590d38fcd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 24 Feb 2025 12:38:44 +0100
Subject: [PATCH 176/505] Fix documentation ToC

---
 mkdocs.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mkdocs.yml b/mkdocs.yml
index 2c31494da..b31fc1f54 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -167,7 +167,7 @@ nav:
   - API Reference:
       - api/index.md
       - api/models.md
-      - api/prompts.md
+      - api/templates.md
       - api/json_schema.md
       - api/guide.md
       - api/parsing.md

From 5c8ff7e1b3f501364ce4cb5b262efdf4f0c64433 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Tue, 25 Feb 2025 12:28:13 +0100
Subject: [PATCH 177/505] s/from_str/from_string

---
 outlines/templates.py   | 8 ++++----
 tests/test_templates.py | 8 ++++----
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/outlines/templates.py b/outlines/templates.py
index a1fbc1932..16588e813 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -41,7 +41,7 @@ def __call__(self, *args, **kwargs) -> str:
             return self.template.render(**kwargs)
 
     @classmethod
-    def from_str(cls, content: str, filters: Dict[str, Callable] = {}):
+    def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
         """Create a `Template` instance from a string containing a Jinja template.
 
         Parameters
@@ -53,7 +53,7 @@ def from_str(cls, content: str, filters: Dict[str, Callable] = {}):
         -------
         An instance of the class with the provided content as a template.
         """
-        return cls(build_template_from_str(content, filters), None)
+        return cls(build_template_from_string(content, filters), None)
 
     @classmethod
     def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
@@ -77,7 +77,7 @@ def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
         return cls(build_template_from_file(path, filters), None)
 
 
-def build_template_from_str(
+def build_template_from_string(
     content: str, filters: Dict[str, Callable] = {}
 ) -> jinja2.Template:
     # Dedent, and remove extra linebreak
@@ -187,7 +187,7 @@ def prompt(
     if docstring is None:
         raise TypeError("Could not find a template in the function's docstring.")
 
-    template = build_template_from_str(cast(str, docstring), filters)
+    template = build_template_from_string(cast(str, docstring), filters)
 
     return Template(template, signature)
 
diff --git a/tests/test_templates.py b/tests/test_templates.py
index 06727eab1..aa9939278 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -6,11 +6,11 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.templates import Template, build_template_from_str
+from outlines.templates import Template, build_template_from_string
 
 
 def render(content: str, **kwargs):
-    template = build_template_from_str(content)
+    template = build_template_from_string(content)
     return template.render(kwargs)
 
 
@@ -417,7 +417,7 @@ def test_prompt_from_str():
     content = """
     Hello, {{ name }}!
     """
-    prompt = Template.from_str(content)
+    prompt = Template.from_string(content)
     assert prompt.signature is None
     assert prompt(name="World") == "Hello, World!"
 
@@ -428,5 +428,5 @@ def test_template_from_str_with_extra_linebreaks():
 
 
     """
-    template = build_template_from_str(content)
+    template = build_template_from_string(content)
     assert template.render(name="World") == "Hello, World!\n"

From 4abe628a2961c87c437b52e7bf0da87f5ed924d5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Tue, 25 Feb 2025 12:26:39 +0100
Subject: [PATCH 178/505] Update documentation and examples following
 `outlines.prompt` deprecation

---
 docs/cookbook/chain_of_density.md |   7 +-
 docs/cookbook/classification.md   |   7 +-
 docs/cookbook/dating_profiles.md  |   5 +-
 docs/cookbook/extraction.md       |   7 +-
 docs/cookbook/simtom.md           |  14 +--
 docs/quickstart.md                |  10 +-
 docs/reference/prompting.md       | 182 ++++++++++++++----------------
 examples/babyagi.py               |  64 +----------
 examples/dating_profile.py        |  20 +---
 examples/math_generate_code.py    |   6 +-
 examples/meta_prompting.py        |  32 +++---
 examples/pick_odd_one_out.py      |   5 +-
 examples/react.py                 |  39 +++----
 examples/self_consistency.py      |   6 +-
 14 files changed, 169 insertions(+), 235 deletions(-)

diff --git a/docs/cookbook/chain_of_density.md b/docs/cookbook/chain_of_density.md
index 2a6b4eb39..10bda9771 100644
--- a/docs/cookbook/chain_of_density.md
+++ b/docs/cookbook/chain_of_density.md
@@ -29,10 +29,10 @@ The prompt also asks the model to return a list of JSON objects that contain the
 We can now implement the prompt provided in the paper:
 
 ```python
-import outlines
+from outlines import Template
 
-@outlines.prompt
-def chain_of_density(article):
+
+chain_of_density = Template.from_string(
     """Article: {{ article }}
 
     You will generate increasingly concise, entity-dense summaries of the above Article.
@@ -61,6 +61,7 @@ def chain_of_density(article):
 
     Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
     """
+)
 ```
 
 ??? Note
diff --git a/docs/cookbook/classification.md b/docs/cookbook/classification.md
index c56096318..023e4543a 100644
--- a/docs/cookbook/classification.md
+++ b/docs/cookbook/classification.md
@@ -16,8 +16,10 @@ model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device=
 We will use the following prompt template:
 
 ```python
-@outlines.prompt
-def customer_support(request):
+from outlines import Template
+
+
+customer_support = Template.from_string(
     """You are an experienced customer success manager.
 
     Given a request from a client, you need to determine when the
@@ -36,6 +38,7 @@ def customer_support(request):
 
     Request: {{ request }}
     Label: """
+)
 ```
 
 ## Choosing between multiple choices
diff --git a/docs/cookbook/dating_profiles.md b/docs/cookbook/dating_profiles.md
index d0fb9b576..1cbea556c 100644
--- a/docs/cookbook/dating_profiles.md
+++ b/docs/cookbook/dating_profiles.md
@@ -57,9 +57,9 @@ class Example:
 We will use Outlines' prompt templating abilities to generate the prompt for us. This help clearly separate the general prompting logic from what is specific to an example.
 
 ```python
+from outlines import Template
 
-@outlines.prompt
-def dating_profile_prompt(description: str, examples: list[Example]):
+dating_profile_prompt = Template.from_string(
     """
     You are a world-renowned matchmaker who understands the modern dating
     market. Your job is to generate dating app profiles for male clients
@@ -79,6 +79,7 @@ def dating_profile_prompt(description: str, examples: list[Example]):
     Description: {{ description }}
     Profile:
     """
+)
 ```
 
 We will provide the model with several few-shot examples:
diff --git a/docs/cookbook/extraction.md b/docs/cookbook/extraction.md
index 28317b6b0..dfcca0cfd 100644
--- a/docs/cookbook/extraction.md
+++ b/docs/cookbook/extraction.md
@@ -15,8 +15,10 @@ model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device=
 And we will be using the following prompt template:
 
 ```python
-@outlines.prompt
-def take_order(order):
+from outlines import Template
+
+
+take_order = Template.from_string(
     """You are the owner of a pizza parlor. Customers \
     send you orders from which you need to extract:
 
@@ -42,6 +44,7 @@ def take_order(order):
 
     ORDER: {{ order }}
     RESULT: """
+)
 ```
 
 We now define our data model using Pydantic:
diff --git a/docs/cookbook/simtom.md b/docs/cookbook/simtom.md
index 4ad78846b..a80ebc24f 100644
--- a/docs/cookbook/simtom.md
+++ b/docs/cookbook/simtom.md
@@ -23,18 +23,15 @@ To implement SimToM with Outlines, we will need to:
 
 Let's dive into it!
 
-### Using Prompt Functions
-
-With Outlines, you can write your prompts as Python functions by adding the `@outlines.prompt` decorator. The prompt template is contained in their docstring, and their arguments correspond to variables used in the prompt.
+### Using Prompt Templates
 
 The authors have shared their code, prompts and data in [this GitHub repository](https://github.com/shawnsihyunlee/simulatedtom). Below, we define in Outlines the prompts they used for the ToMI dataset:
 
 ```python
-import outlines
+from outlines import Template
 
 
-@outlines.prompt
-def perspective_taking(story: str, character: str) -> None:
+perspective_taking = Template.from_string(
     """<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
     Your job is to output only the events that the specified character, {{character}}, knows about.
 
@@ -45,9 +42,9 @@ def perspective_taking(story: str, character: str) -> None:
 
     Story: {{story}}
     What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]""" # noqa
+)
 
-@outlines.prompt
-def simulation(events: list, name: str, question: str) -> None:
+simulation = Template.from_string(
     """<s>[INST] {% for event in events %}
     {{event}}
     {% endfor %}
@@ -55,6 +52,7 @@ def simulation(events: list, name: str, question: str) -> None:
     Based on the above information, answer the following question:
     {{question}}
     You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]""" # noqa
+)
 ```
 
 ### JSON Structured Generation
diff --git a/docs/quickstart.md b/docs/quickstart.md
index 81a067ad6..ec445dcba 100644
--- a/docs/quickstart.md
+++ b/docs/quickstart.md
@@ -115,10 +115,9 @@ Or use the [requests][requests]{:target="_blank"} library from another python pr
 Prompting can lead to messy code. Outlines' prompt functions are python functions that contain a template for the prompt in their docstring. We use a powerful templating language to allow you to loop over lists, dictionaries, add conditionals, etc. directly from the prompt. When called, a prompt function returns the rendered template:
 
 ```python
-import outlines
+from outlines import Template
 
-@outlines.prompt
-def few_shots(instructions, examples, question):
+few_shots = Template.from_string(
     """{{ instructions }}
 
     Examples
@@ -135,6 +134,7 @@ def few_shots(instructions, examples, question):
     Q: {{ question }}
     A:
     """
+)
 
 instructions = "Please answer the following question following the examples"
 examples = [
@@ -175,9 +175,9 @@ Once you are done experimenting with a prompt and an output structure, it is use
     import outlines
 
 
-    @outlines.prompt
-    def tell_a_joke(topic):
+    tell_a_joke = outlines.Template.from_string(
         """Tell me a joke about {{ topic }}."""
+    )
 
     class Joke(BaseModel):
         setup: str
diff --git a/docs/reference/prompting.md b/docs/reference/prompting.md
index 8ea9c0243..986ca82fc 100644
--- a/docs/reference/prompting.md
+++ b/docs/reference/prompting.md
@@ -33,15 +33,13 @@ will pass to the prompt function.
 
 === "Code"
 
-    ```python
-    import outlines
+    ```python title="greetings.py"
+    from outlines import Template
 
-    @outlines.prompt
-    def greetings(name, question):
-        """Hello, {{ name }}!
-        {{ question }}
-        """
+    prompt = """Hello, {{ name }}!
+    {{ question }}"""
 
+    greetings = Template.from_string(prompt)
     prompt = greetings("user", "How are you?")
     print(prompt)
     ```
@@ -58,12 +56,10 @@ If a variable is missing in the function's arguments, Jinja2 will throw an `Unde
 === "Code"
 
     ```python
-    import outlines
-
-    @outlines.prompt
-    def greetings(name):
-        """Hello, {{ surname }}!"""
+    from outlines import Template
 
+    prompt = """Hello, {{ surname }}!"""
+    greetings = Template.from_string(prompt)
     prompt = greetings("user")
     ```
 
@@ -72,9 +68,9 @@ If a variable is missing in the function's arguments, Jinja2 will throw an `Unde
     ```text
     Traceback (most recent call last):
       File "<stdin>", line 9, in <module>
-      File "/home/remi/projects/normal/outlines/outlines/prompts.py", line 38, in __call__
+      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 38, in __call__
           return render(self.template, **bound_arguments.arguments)
-      File "/home/remi/projects/normal/outlines/outlines/prompts.py", line 213, in render
+      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 213, in render
           return jinja_template.render(**values)
       File "/home/remi/micromamba/envs/outlines/lib/python3.9/site-packages/jinja2/environment.py", line 1301, in render
           self.environment.handle_exception()
@@ -84,26 +80,23 @@ If a variable is missing in the function's arguments, Jinja2 will throw an `Unde
       jinja2.exceptions.UndefinedError: 'surname' is undefined
     ```
 
-## Importing prompt functions
+## Importing prompts from files
 
-Prompt functions are functions, and thus can be imported from other modules:
-
-=== "prompts.py"
-    ```python
-    import outlines
+Outlines allows you to read a prompt template from a text file. This way you can build "white space perfect" prompts, and version them independently from your code. We have found ourselves gravitating around this pattern a lot since Outlines came out:
 
-    @outlines.prompt
-    def greetings(name, question):
-        """Hello, {{ name }}!
-        {{ question }}
-        """
+=== "prompt.txt"
+    ```text
+    """Hello, {{ name }}!
+    {{ question }}
+    """
     ```
 
 === "generate.py"
 
     ```python
-    from .prompts import greetings
+    from outlines import Template
 
+    greetings = Template.from_file("prompt.txt")
     prompt = greetings("John Doe", "How are you today?")
     ```
 
@@ -123,27 +116,26 @@ keys `question` and `answer` to the prompt function:
 
 === "Code"
 
-    ```python
-    import outlines
+    ```text title="prompt.txt"
+    {{ instructions }}
 
-    @outlines.prompt
-    def few_shots(instructions, examples, question):
-        """{{ instructions }}
+    Examples
+    --------
 
-        Examples
-        --------
+    {% for example in examples %}
+    Q: {{ example.question }}
+    A: {{ example.answer }}
 
-        {% for example in examples %}
-        Q: {{ example.question }}
-        A: {{ example.answer }}
+    {% endfor %}
+    Question
+    --------
 
-        {% endfor %}
-        Question
-        --------
+    Q: {{ question }}
+    A:
+    ```
 
-        Q: {{ question }}
-        A:
-        """
+    ```python title="render.py"
+    from outlines import Template
 
     instructions = "Please answer the following question following the examples"
     examples = [
@@ -152,6 +144,7 @@ keys `question` and `answer` to the prompt function:
     ]
     question = "4+4 = ?"
 
+    few_shots = Template.from_file("prompt.txt")
     prompt = few_shots(instructions, examples, question)
     print(prompt)
     ```
@@ -194,7 +187,7 @@ Several projects (e.g.[Toolformer](https://arxiv.org/abs/2302.04761), [ViperGPT]
 === "Code"
 
     ```python
-    import outlines
+    from outlines import Template
 
     def my_tool(arg1: str, arg2: int):
         """Tool description.
@@ -203,16 +196,15 @@ Several projects (e.g.[Toolformer](https://arxiv.org/abs/2302.04761), [ViperGPT]
         """
         pass
 
-    @outlines.prompt
-    def tool_prompt(question, tool):
-        """{{ question }}
+    prompt = """{{ question }}
 
-        COMMANDS
-        1. {{ tool | name }}: {{ tool | description }}, args: {{ tool | args }}
+    COMMANDS
+    1. {{ tool | name }}: {{ tool | description }}, args: {{ tool | args }}
 
-        {{ tool | source }}
-        """
+    {{ tool | source }}
+    """
 
+    tool_prompt = Template.from_string(prompt)
     prompt = tool_prompt("Can you do something?", my_tool)
     print(prompt)
     ```
@@ -250,16 +242,14 @@ pretty print a dictionary from within an Outlines prompt function
     ```python
     from pydantic import BaseModel, Field
 
-    import outlines
+   from outlines import Template
 
     class MyResponse(BaseModel):
         field1: int = Field(description="an int")
         field2: str
 
-    @outlines.prompt
-    def my_prompt(response_model):
-        """{{ response_model | schema }}"""
 
+    my_prompt = Template.from_string("""{{ response_model | schema }}""")
     prompt = my_prompt(MyResponse)
     print(prompt)
     # {
@@ -285,8 +275,9 @@ pretty print a dictionary from within an Outlines prompt function
 
 ## Formatting conventions
 
-Prompt functions are opinionated when it comes to rendering, and these opinions
-are meant to avoid prompting mistakes and help with formatting.
+Prompt templates are opinionated when it comes to rendering a template read from
+a string, and these opinions are meant to avoid prompting mistakes and help with
+formatting.
 
 ### Whitespaces
 
@@ -300,18 +291,19 @@ below does not matter for formatting:
 === "Code"
 
     ```python
-    import outlines
+    from outlines import Template
 
-    @outlines.prompt
-    def prompt1():
-        """My prompt
-        """
 
-    @outlines.prompt
-    def prompt2():
-        """
-        My prompt
-        """
+    prompt1 = Template.from_string(
+    """My prompt
+    """
+    )
+
+    prompt2 = Template.from_string(
+    """
+    My prompt
+    """
+    )
 
     print(prompt1())
     print(prompt2())
@@ -329,27 +321,27 @@ Indentation is relative to the second line of the docstring, and leading spaces
 === "Code"
 
     ```python
-    import outlines
+    from outlines import Template
+
+    example1 = Template.from_string(
+    """First line
+    Second line
+    """
+    )
 
-    @outlines.prompt
-    def example1():
-        """First line
+    example2 = Template.from_string(
+    """
         Second line
-        """
+        Third line
+    """
+    )
 
-    @outlines.prompt
-    def example2():
-        """
-          Second line
+    example3 = Template.from_string(
+    """
+        Second line
           Third line
-        """
-
-    @outlines.prompt
-    def example3():
-        """
-          Second line
-            Third line
-        """
+    """
+    )
 
     print(example1())
     print(example2())
@@ -378,18 +370,18 @@ You can use the backslash `\` to break a long line of text. It will render as a
 === "Code"
 
     ```python
-    import outlines
-
-    @outlines.prompt
-    def example():
-       """
-       Break in \
-       several lines \
-       But respect the indentation
-           on line breaks.
-       And after everything \
-       Goes back to normal
-       """
+    from outlines import Template
+
+    example = Template.from_string(
+    """
+    Break in \
+    several lines \
+    But respect the indentation
+        on line breaks.
+    And after everything \
+    Goes back to normal
+    """
+    )
 
     print(example())
     ```
diff --git a/examples/babyagi.py b/examples/babyagi.py
index 0a7a0b13b..314f9dbca 100644
--- a/examples/babyagi.py
+++ b/examples/babyagi.py
@@ -9,48 +9,16 @@
 
 import outlines
 import outlines.models as models
+from outlines import Template
+
 
 model = models.openai("gpt-4o-mini")
 complete = outlines.generate.text(model)
 
-
-#################
-# Perform tasks #
-#################
-
-
-@outlines.prompt
-def perform_task_ppt(objective: str, task: str):
-    """You are an AI who performs one task based on the following objective: {{objective}}.
-
-    Your task: {{task.task_name}}
-
-    Response:
-    """
-
-
-#####################
-# Create a new task #
-#####################
-
-
-@outlines.prompt
-def create_tasks_ppt(
-    objective: str, previous_task: str, result: str, task_list: List[str]
-):
-    """You are an task creation AI that uses the result of an execution agent to \
-    create new tasks with the following objective: {{objective}}.
-
-    The last completed task has the result: {{result}}.
-
-    This result was based on this task description: {{previous_task}}. These are \
-    incomplete tasks: {{task_list | join(task_list)}}.
-
-    Based on the result, create new tasks to be completed by the AI system that \
-    do not overlap with incomplete tasks.
-
-    Return the tasks as an array.
-    """
+## Load the prompts
+perform_task_ppt = Template.from_file("prompts/babyagi_perform_task.txt")
+create_tasks_ppt = Template.from_file("prompts/babyagi_create_task.txt")
+prioritize_tasks_ppt = Template.from_file("prompts/babyagi_prioritize_task.txt")
 
 
 def create_tasks_fmt(result: str) -> List[str]:
@@ -65,26 +33,6 @@ def create_tasks_fmt(result: str) -> List[str]:
     return task_list
 
 
-########################
-# Prioritize new tasks #
-########################
-
-
-@outlines.prompt
-def prioritize_tasks_ppt(objective: str, task_names: List[str], next_task_id: int):
-    """You are a task prioritization AI tasked with cleaning the formatting of \
-    and reprioritizing the following tasks: {{task_names}}.
-
-    Consider the ultimate objective of your team: {{objective}}.
-
-    Do not remove any tasks. Return the result as a numbered list, like:
-    #. First task
-    #. Second task
-
-    Start the tasks list with the number {{next_task_id}}.
-    """
-
-
 def prioritize_tasks_fmt(result: str):
     new_tasks = result.split("\n")
 
diff --git a/examples/dating_profile.py b/examples/dating_profile.py
index 504ec943d..d26984223 100644
--- a/examples/dating_profile.py
+++ b/examples/dating_profile.py
@@ -6,7 +6,7 @@
 from pydantic import BaseModel, conlist
 
 import outlines
-from outlines import models
+from outlines import models, Template
 
 
 class QuestionChoice(str, Enum):
@@ -41,23 +41,6 @@ class Example:
     profile: DatingProfile
 
 
-@outlines.prompt
-def dating_profile_prompt(description: str, examples: list[Example]):
-    """
-    You are a world-renowned matchmaker who understands the modern dating market. Your job is to generate dating app profiles for male clients interested in women based on a provided description. The profiles should be authentic, show off their strengths, and maximize their likelihood of getting matches on dating apps.
-    Here are some examples of past clients that you have successfully created profiles for:
-    {% for example in examples %}
-    Description:
-    {{ example.description }}
-    Profile:
-    {{ example.profile }}
-    {% endfor %}
-    Here is the new client who you need to create a profile for:
-    Description: {{ description }}
-    Profile:
-    """
-
-
 samples: list[Example] = [
     Example(
         description="I'm an author and former professional soccer player living in Seattle who publishes popular fiction books. A typical day for me starts by hanging out with my cat, drinking a coffee, and reading as much as I can in a few hours. Then, I'll prepare a quick smoothie before starting to write for a few hours, take a break with soccer or running a few miles, and finally meet friends for dinner at a new, hip restaurant in the evening. Sometimes we go axe-throwing afterwards, or play poker, or watch a comedy show, or visit a dive bar. On my vacations, I travel extensively to countries South America, Europe, and Asia, with the goal of visiting them all!",
@@ -120,6 +103,7 @@ def dating_profile_prompt(description: str, examples: list[Example]):
 
 new_description = "I'm a laid-back lawyer who spends a lot of his free-time gaming. I work in a corporate office, but ended up here after the start-up I cofounded got acquired, so still play ping pong with my cool coworkers every day. I have a bar at home where I make cocktails, which is great for entertaining friends. I secretly like to wear suits and get a new one tailored every few months. I also like weddings because I get to wear those suits, and it's a good excuse for a date. I watch the latest series because I'm paying, with my hard-earned money, for every streaming service."
 
+dating_profile_prompt = Template.from_file("prompts/dating_profile.txt")
 prompt = dating_profile_prompt(description=new_description, examples=samples)
 profile = outlines.generate.json(model, DatingProfile)(prompt)  # type: ignore
 print(profile)
diff --git a/examples/math_generate_code.py b/examples/math_generate_code.py
index 7eb1651a7..6d2ba8ae6 100644
--- a/examples/math_generate_code.py
+++ b/examples/math_generate_code.py
@@ -1,6 +1,7 @@
 """Example from https://dust.tt/spolu/a/d12ac33169"""
 import outlines
 import outlines.models as models
+from outlines import Template
 
 examples = [
     {"question": "What is 37593 * 67?", "code": "37593 * 67"},
@@ -16,9 +17,7 @@
 
 question = "Carla is downloading a 200 GB file. She can download 2 GB/minute, but 40% of the way through the download, the download fails. Then Carla has to restart the download from the beginning. How load did it take her to download the file in minutes?"
 
-
-@outlines.prompt
-def answer_with_code_prompt(question, examples):
+answer_with_code_prompt = Template.from_string(
     """
     {% for example in examples %}
     QUESTION: {{example.question}}
@@ -27,6 +26,7 @@ def answer_with_code_prompt(question, examples):
     {% endfor %}
     QUESTION: {{question}}
     CODE:"""
+)
 
 
 def execute_code(code):
diff --git a/examples/meta_prompting.py b/examples/meta_prompting.py
index cba18b5fe..85f512bf6 100644
--- a/examples/meta_prompting.py
+++ b/examples/meta_prompting.py
@@ -13,14 +13,16 @@
 
 import outlines
 import outlines.models as models
+from outlines import Template
 
 
 def split_into_steps(question, model_name: str):
-    @outlines.prompt
-    def solve(question):
+
+    solve = Template.from_string(
         """{{question}}
         Rephrase : : as a true or false statement, identify an Object, relationship and subject
         """
+    )
 
     model = models.openai(model_name)
     generator = outlines.generate.text(model)
@@ -38,16 +40,17 @@ def solve(question):
 
 
 def fill_in_the_blanks(question, model_name: str):
-    @outlines.prompt
-    def determine_goal(question):
+
+    determine_goal = Template.from_string(
         """{{question}}
 
         In order to solve this problem, we will analyze each of the options and determine
         """
+    )
 
-    @outlines.prompt
-    def solve(memory):
+    solve = Template.from_string(
         """{{memory}}. Let's begin."""
+    )
 
     model = models.openai(model_name)
     generator = outlines.generate.text(model)
@@ -62,8 +65,8 @@ def solve(memory):
 
 
 def ask_an_expert(question, model_name: str):
-    @outlines.prompt
-    def find_expert(question):
+
+    find_expert = Template.from_string(
         """
         {{question}}
         I entered my question into the Expert Generator \
@@ -79,15 +82,16 @@ def find_expert(question):
         found the most qualified expert. The name displayed \
         on the screen: "
         """
+    )
 
-    @outlines.prompt
-    def get_answer(question, expert, memory):
+    get_answer = Template.from_string(
         """
         {{memory}}".
         I am ready to ask my question.
         "{{expert}}" I say,
         {{question}}
         """
+    )
 
     model = models.openai(model_name)
     generator = outlines.generate.text(model)
@@ -102,20 +106,20 @@ def get_answer(question, expert, memory):
 
 
 def ask_an_expert_simple(question, model_name: str):
-    @outlines.prompt
-    def find_expert(question):
+    find_expert = Template.from_string(
         """
         Q: {{question}}
         A: A good person to answer this question would be
         """
+    )
 
-    @outlines.prompt
-    def get_answer(expert, memory):
+    get_answer = Template.from_string(
         """
         {{memory}}.
 
         For instance, {{expert}} would answer
         """
+    )
 
     model = models.openai(model_name)
     generator = outlines.generate.text(model)
diff --git a/examples/pick_odd_one_out.py b/examples/pick_odd_one_out.py
index 6cd9f1daf..834c40824 100644
--- a/examples/pick_odd_one_out.py
+++ b/examples/pick_odd_one_out.py
@@ -13,8 +13,7 @@
 import outlines.models as models
 
 
-@outlines.prompt
-def build_ooo_prompt(options):
+build_ooo_prompt = outlines.Template.from_string(
     """
     Pick the odd word out: skirt, dress, pen, jacket.
     skirt is clothing, dress is clothing, pen is an object, jacket is clothing.
@@ -27,7 +26,7 @@ def build_ooo_prompt(options):
     Pick the odd word out: {{ options | join(", ") }}.
 
     """
-
+)
 
 options = ["sea", "mountains", "plains", "sock"]
 
diff --git a/examples/react.py b/examples/react.py
index 34b3c6eb2..9b876eb49 100644
--- a/examples/react.py
+++ b/examples/react.py
@@ -13,29 +13,30 @@
 import requests  # type: ignore
 
 import outlines
+from outlines import Template
 import outlines.generate as generate
 import outlines.models as models
 
 
-@outlines.prompt
-def build_reAct_prompt(question):
-    """What is the elevation range for the area that the eastern sector of the Colorado orogeny extends into?
-    Tho 1: I need to search Colorado orogeny, find the area that the eastern sector of the Colorado ...
-    Act 2: Search 'Colorado orogeny'
-    Obs 2: The Colorado orogeny was an episode of mountain building (an orogeny) ...
-    Tho 3: It does not mention the eastern sector. So I need to look up eastern sector.
-    ...
-    Tho 4: High Plains rise in elevation from around 1,800 to 7,000 ft, so the answer is 1,800 to 7,000 ft.
-    Act 5: Finish '1,800 to 7,000 ft'
-    {{ question }}
-    """
-
-
-@outlines.prompt
-def add_mode(i, mode, result, prompt):
-    """{{ prompt }}
-    {{ mode }} {{ i }}: {{ result }}
-    """
+build_reAct_prompt = Template.from_string(
+"""What is the elevation range for the area that the eastern sector of the Colorado orogeny extends into?
+Tho 1: I need to search Colorado orogeny, find the area that the eastern sector of the Colorado ...
+Act 2: Search 'Colorado orogeny'
+Obs 2: The Colorado orogeny was an episode of mountain building (an orogeny) ...
+Tho 3: It does not mention the eastern sector. So I need to look up eastern sector.
+...
+Tho 4: High Plains rise in elevation from around 1,800 to 7,000 ft, so the answer is 1,800 to 7,000 ft.
+Act 5: Finish '1,800 to 7,000 ft'
+{{ question }}
+"""
+)
+
+
+add_mode = Template.from_string(
+"""{{ prompt }}
+{{ mode }} {{ i }}: {{ result }}
+"""
+)
 
 
 def search_wikipedia(query: str):
diff --git a/examples/self_consistency.py b/examples/self_consistency.py
index f1bbe2a18..50f12acf1 100644
--- a/examples/self_consistency.py
+++ b/examples/self_consistency.py
@@ -4,6 +4,7 @@
 
 import outlines
 import outlines.models as models
+from outlines import Template
 
 examples = [
     {
@@ -43,8 +44,7 @@
 question = "When I was 6 my sister was half my age. Now I’m 70 how old is my sister?"
 
 
-@outlines.prompt
-def few_shots(question, examples):
+few_shots = Template.from_string(
     """
     {% for example in examples %}
     Q: {{ example.question }}
@@ -53,7 +53,7 @@ def few_shots(question, examples):
     Q: {{ question }}
     A:
     """
-
+)
 
 model = models.openai("gpt-4o-mini")
 generator = outlines.generate.text(model)

From 8a06251aa91b6923116a6253d1c902a9d472a05c Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Fri, 28 Feb 2025 19:39:10 +0100
Subject: [PATCH 179/505] Format files and fix numerous `mypy` warnings, thanks
 to `ruff`

Left to fix:

    outlines/models/exllamav2.py:356: error: Argument 3 to "ExLlamaV2Model" has incompatible type "int | None"; expected "int"  [arg-type]
    outlines/generate/format.py:36: error: Incompatible types in assignment (expression has type "str", variable has type "Regex")  [assignment]
---
 examples/babyagi.py                       |  1 +
 examples/math_generate_code.py            |  1 +
 examples/meta_prompting.py                |  8 ++------
 examples/parsing.py                       |  1 +
 examples/pick_odd_one_out.py              |  1 +
 examples/react.py                         |  5 +++--
 examples/sampling.ipynb                   | 25 ++++++++++++++++-------
 examples/self_consistency.py              |  2 +-
 examples/simulation_based_inference.ipynb | 10 ++++++---
 tests/test_function.py                    |  1 +
 tests/test_templates.py                   | 13 +++++++++++-
 11 files changed, 48 insertions(+), 20 deletions(-)

diff --git a/examples/babyagi.py b/examples/babyagi.py
index 314f9dbca..a8f1de8ba 100644
--- a/examples/babyagi.py
+++ b/examples/babyagi.py
@@ -4,6 +4,7 @@
 
 The original repo can be found at https://github.com/yoheinakajima/babyagi
 """
+
 from collections import deque
 from typing import Deque, List
 
diff --git a/examples/math_generate_code.py b/examples/math_generate_code.py
index 6d2ba8ae6..c65537ddb 100644
--- a/examples/math_generate_code.py
+++ b/examples/math_generate_code.py
@@ -1,4 +1,5 @@
 """Example from https://dust.tt/spolu/a/d12ac33169"""
+
 import outlines
 import outlines.models as models
 from outlines import Template
diff --git a/examples/meta_prompting.py b/examples/meta_prompting.py
index 85f512bf6..18f1ad41a 100644
--- a/examples/meta_prompting.py
+++ b/examples/meta_prompting.py
@@ -9,6 +9,7 @@
        https://arxiv.org/abs/2102.07350.
 
 """
+
 import argparse
 
 import outlines
@@ -17,7 +18,6 @@
 
 
 def split_into_steps(question, model_name: str):
-
     solve = Template.from_string(
         """{{question}}
         Rephrase : : as a true or false statement, identify an Object, relationship and subject
@@ -40,7 +40,6 @@ def split_into_steps(question, model_name: str):
 
 
 def fill_in_the_blanks(question, model_name: str):
-
     determine_goal = Template.from_string(
         """{{question}}
 
@@ -48,9 +47,7 @@ def fill_in_the_blanks(question, model_name: str):
         """
     )
 
-    solve = Template.from_string(
-        """{{memory}}. Let's begin."""
-    )
+    solve = Template.from_string("""{{memory}}. Let's begin.""")
 
     model = models.openai(model_name)
     generator = outlines.generate.text(model)
@@ -65,7 +62,6 @@ def fill_in_the_blanks(question, model_name: str):
 
 
 def ask_an_expert(question, model_name: str):
-
     find_expert = Template.from_string(
         """
         {{question}}
diff --git a/examples/parsing.py b/examples/parsing.py
index a10da4ebe..9e08eae14 100644
--- a/examples/parsing.py
+++ b/examples/parsing.py
@@ -1,4 +1,5 @@
 """An example illustrating parser-based masking."""
+
 import math
 import time
 from copy import copy
diff --git a/examples/pick_odd_one_out.py b/examples/pick_odd_one_out.py
index 834c40824..f1a992411 100644
--- a/examples/pick_odd_one_out.py
+++ b/examples/pick_odd_one_out.py
@@ -9,6 +9,7 @@
        arXiv preprint arXiv:2212.06094.
 
 """
+
 import outlines
 import outlines.models as models
 
diff --git a/examples/react.py b/examples/react.py
index 9b876eb49..9c1954e88 100644
--- a/examples/react.py
+++ b/examples/react.py
@@ -10,6 +10,7 @@
 .. [2] Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, K., & Cao, Y. (2022). React: Synergizing reasoning and acting in language models. arXiv preprint arXiv:2210.03629.
 
 """
+
 import requests  # type: ignore
 
 import outlines
@@ -19,7 +20,7 @@
 
 
 build_reAct_prompt = Template.from_string(
-"""What is the elevation range for the area that the eastern sector of the Colorado orogeny extends into?
+    """What is the elevation range for the area that the eastern sector of the Colorado orogeny extends into?
 Tho 1: I need to search Colorado orogeny, find the area that the eastern sector of the Colorado ...
 Act 2: Search 'Colorado orogeny'
 Obs 2: The Colorado orogeny was an episode of mountain building (an orogeny) ...
@@ -33,7 +34,7 @@
 
 
 add_mode = Template.from_string(
-"""{{ prompt }}
+    """{{ prompt }}
 {{ mode }} {{ i }}: {{ result }}
 """
 )
diff --git a/examples/sampling.ipynb b/examples/sampling.ipynb
index bcbcca1e2..d6b50e33d 100644
--- a/examples/sampling.ipynb
+++ b/examples/sampling.ipynb
@@ -85,6 +85,7 @@
     "    },\n",
     "]\n",
     "\n",
+    "\n",
     "@text.prompt\n",
     "def few_shot_prompt(question, examples):\n",
     "    \"\"\"\n",
@@ -96,6 +97,7 @@
     "    A:\n",
     "    \"\"\"\n",
     "\n",
+    "\n",
     "# Prompt functions can be partially evaluated like any other function\n",
     "gsm8k_prompt = ft.partial(few_shot_prompt, examples=examples)"
    ]
@@ -155,31 +157,40 @@
     "                digits.append(digit)\n",
     "        except AttributeError:\n",
     "            print(f\"Could not parse the completion: '{answer}'\")\n",
-    "    \n",
+    "\n",
     "    unique_digits, counts = np.unique(digits, return_counts=True)\n",
     "    return {d: c for d, c in zip(unique_digits, counts)}\n",
     "\n",
+    "\n",
     "def plot_counts(counts):\n",
-    "    fig = plt.figure(figsize=(12,8))\n",
+    "    fig = plt.figure(figsize=(12, 8))\n",
     "    ax = fig.add_subplot(111)\n",
-    "    \n",
+    "\n",
     "    bar = ax.bar(counts.keys(), counts.values())\n",
     "    ax.spines[[\"right\", \"top\", \"left\"]].set_visible(False)\n",
     "    ax.get_yaxis().set_visible(False)\n",
     "    ax.get_yaxis().set_visible(False)\n",
-    "    \n",
+    "\n",
     "    for rect in bar:\n",
     "        height = rect.get_height()\n",
-    "        plt.text(rect.get_x() + rect.get_width() / 2.0, height, f'{height:.0f}', ha='center', va='bottom', fontsize=20)\n",
-    "    \n",
+    "        plt.text(\n",
+    "            rect.get_x() + rect.get_width() / 2.0,\n",
+    "            height,\n",
+    "            f\"{height:.0f}\",\n",
+    "            ha=\"center\",\n",
+    "            va=\"bottom\",\n",
+    "            fontsize=20,\n",
+    "        )\n",
+    "\n",
     "    ax.set_xticks(list(counts.keys()))\n",
     "    ax.set_xlabel(\"Answer\")\n",
     "\n",
+    "\n",
     "def entropy(counts):\n",
     "    counts = np.array(list(counts.values()))\n",
     "    probs = counts / np.sum(counts)\n",
     "    log_probs = np.log(probs)\n",
-    "    return - np.sum(probs * log_probs)"
+    "    return -np.sum(probs * log_probs)"
    ]
   },
   {
diff --git a/examples/self_consistency.py b/examples/self_consistency.py
index 50f12acf1..4acba9d58 100644
--- a/examples/self_consistency.py
+++ b/examples/self_consistency.py
@@ -78,5 +78,5 @@
 answer_value = [key for key, value in results.items() if value == max_count][0]
 total_count = sum(results.values())
 print(
-    f"The most likely answer is {answer_value} ({max_count/total_count*100}% consensus)"
+    f"The most likely answer is {answer_value} ({max_count / total_count * 100}% consensus)"
 )
diff --git a/examples/simulation_based_inference.ipynb b/examples/simulation_based_inference.ipynb
index e6b999582..6ad52e58e 100644
--- a/examples/simulation_based_inference.ipynb
+++ b/examples/simulation_based_inference.ipynb
@@ -61,7 +61,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "result = requests.get(\"https://raw.githubusercontent.com/openai/grade-school-math/master/grade_school_math/data/train.jsonl\")\n",
+    "result = requests.get(\n",
+    "    \"https://raw.githubusercontent.com/openai/grade-school-math/master/grade_school_math/data/train.jsonl\"\n",
+    ")\n",
     "lines = result.iter_lines()"
    ]
   },
@@ -121,8 +123,10 @@
     "    A:\n",
     "    \"\"\"\n",
     "\n",
+    "\n",
     "model = models.text_completion.openai(\"text-davinci-003\", max_tokens=128)\n",
     "\n",
+    "\n",
     "# TODO: This could largely benefit from vectorization in #52\n",
     "def one_train_example(problem, example_set):\n",
     "    example_ids = random.choices(range(0, len(example_set)), k=5)\n",
@@ -216,7 +220,7 @@
     "\n",
     "example_ids, counts = np.unique(samples, return_counts=True)\n",
     "\n",
-    "fig = plt.figure(figsize=(12,8))\n",
+    "fig = plt.figure(figsize=(12, 8))\n",
     "ax = fig.add_subplot(111)\n",
     "ax.bar(example_ids, counts)\n",
     "\n",
@@ -224,7 +228,7 @@
     "\n",
     "ax.set_xticks(range(10))\n",
     "ax.set_xlabel(\"Example #\")\n",
-    "ax.set_ylabel(\"Counts\")\n"
+    "ax.set_ylabel(\"Counts\")"
    ]
   },
   {
diff --git a/tests/test_function.py b/tests/test_function.py
index d66f3911c..8bb12976e 100644
--- a/tests/test_function.py
+++ b/tests/test_function.py
@@ -9,6 +9,7 @@
 
 def test_function_basic():
     with pytest.deprecated_call(match="The @prompt decorator"):
+
         @outlines.prompt
         def test_template(text: str):
             """{{ text }}"""
diff --git a/tests/test_templates.py b/tests/test_templates.py
index aa9939278..2fda3caa3 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -114,6 +114,7 @@ def test_render_jinja():
 
 def test_prompt_basic():
     with pytest.deprecated_call():
+
         @outlines.prompt
         def test_tpl(variable):
             """{{variable}} test"""
@@ -141,6 +142,7 @@ def test_single_quote_tpl(variable):
 
 def test_prompt_kwargs():
     with pytest.deprecated_call():
+
         @outlines.prompt
         def test_kwarg_tpl(var, other_var="other"):
             """{{var}} and {{other_var}}"""
@@ -184,6 +186,7 @@ def with_description():
         pass
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def name_description_ppt(fn):
             """
@@ -200,6 +203,7 @@ def with_signature(one: int, two: List[str], three: float = 1.0):
         pass
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def name_signature_ppt(fn):
             """
@@ -207,12 +211,15 @@ def name_signature_ppt(fn):
             """
 
         rendered = name_signature_ppt(with_signature)
-        assert rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
+        assert (
+            rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
+        )
 
     def test_function_call(one, two=2):
         return one + two
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def source_ppt(fn):
             """
@@ -229,6 +236,7 @@ class SimpleResponse(BaseModel):
         two: str
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def source_ppt(model):
             "{{model | schema }}"
@@ -261,6 +269,7 @@ def test_prompt_dict_response():
     response = {"one": "a description", "two": ""}
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def source_ppt(model):
             "{{model | schema }}"
@@ -306,6 +315,7 @@ def with_all(
         pass
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def args_prompt(fn):
             """args: {{ fn | args }}"""
@@ -335,6 +345,7 @@ def reverse(s: str) -> str:
         return s[::-1]
 
     with pytest.deprecated_call():
+
         @outlines.prompt(filters=dict(reverse=reverse))
         def test_tpl(variable):
             """{{ variable | reverse }} test"""

From e91a64c40bef28da70bd183736458c717d58a050 Mon Sep 17 00:00:00 2001
From: Yvan Sraka <yvan@sraka.xyz>
Date: Mon, 3 Mar 2025 15:01:00 +0100
Subject: [PATCH 180/505] Remove `pre-commit` runtime dependency

---
 pyproject.toml | 1 -
 1 file changed, 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 969290b07..dfed39973 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -42,7 +42,6 @@ dependencies = [
    "torch",
    "outlines_core==0.1.26",
    "genson",
-   "pre-commit>=4.0.1",
 ]
 dynamic = ["version"]
 

From 4701ecd6b94d48e431fb022e1fdc6d7b81eaebb3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 28 Feb 2025 15:02:13 +0100
Subject: [PATCH 181/505] Remove `|` and us `either` factory function instead

`|` is not defined for strings, so using it has the potential to lead to
confusing errors. We thus replace the operator overloading with a more
explicit `either` factory function which converts string arguments to
`String` objects automatically.
---
 outlines/types/dsl.py | 22 ++++++++++------------
 1 file changed, 10 insertions(+), 12 deletions(-)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 94b4d26d9..e5163b0ce 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -43,18 +43,6 @@ def __radd__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
 
         return Sequence([other, self])
 
-    def __or__(self: "Term", other: Union[str, "Term"]) -> "Alternatives":
-        if isinstance(other, str):
-            other = String(other)
-
-        return Alternatives([self, other])
-
-    def __ror__(self: "Term", other: Union[str, "Term"]) -> "Alternatives":
-        if isinstance(other, str):
-            other = String(other)
-
-        return Alternatives([other, self])
-
     def __get_validator__(self, _core_schema):
         def validate(input_value):
             return self.validate(input_value)
@@ -310,6 +298,16 @@ def __repr__(self):
 
 
 def optional(self: Term) -> Optional:
+def either(*args: Union[str, Term]):
+    """Represents an alternative between different terms or strings.
+
+    This factory function automatically translates string arguments
+    into `String` objects.
+    """
+    args = [String(arg) if arg is str else arg for arg in args]
+    return Alternatives(args)
+
+
     return Optional(self)
 
 

From cf2eaf7c3272b0806a4b1c7b3f4ae4bec1d52a29 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 28 Feb 2025 15:04:59 +0100
Subject: [PATCH 182/505] Convert string inputs into `String` objects
 automatically

---
 outlines/types/dsl.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index e5163b0ce..0246b28ef 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -297,7 +297,6 @@ def __repr__(self):
         return f"QuantifyBetween(term={repr(self.term)}, min_count={repr(self.min_count)}, max_count={repr(self.max_count)})"
 
 
-def optional(self: Term) -> Optional:
 def either(*args: Union[str, Term]):
     """Represents an alternative between different terms or strings.
 
@@ -308,14 +307,18 @@ def either(*args: Union[str, Term]):
     return Alternatives(args)
 
 
+def optional(self: Union[Term, str]) -> Optional:
+    self = String(self) if self is str else self
     return Optional(self)
 
 
 def one_or_more(self: Term) -> KleenePlus:
+    self = String(self) if self is str else self
     return KleenePlus(self)
 
 
-def repeat(self: Term, min_count: int, max_count: int) -> QuantifyBetween:
+def between(self: Term, min_count: int, max_count: int) -> QuantifyBetween:
+    self = String(self) if self is str else self
     match (min_count, max_count):
         case (None, None):
             raise ValueError(
@@ -330,10 +333,12 @@ def repeat(self: Term, min_count: int, max_count: int) -> QuantifyBetween:
 
 
 def times(self: Term, count: int = 0) -> QuantifyExact:
+    self = String(self) if self is str else self
     return QuantifyExact(self, count)
 
 
 def zero_or_more(self: Term) -> KleeneStar:
+    self = String(self) if self is str else self
     return KleeneStar(self)
 
 

From 3ca64b325171e99a85c9ef98af53d8666397195c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 28 Feb 2025 15:15:20 +0100
Subject: [PATCH 183/505] Update the quantifiers factory functions

---
 outlines/types/__init__.py |  11 +-
 outlines/types/dsl.py      |  95 ++++++++++-------
 tests/types/test_dsl.py    | 205 ++++++++++++++++++++++---------------
 3 files changed, 187 insertions(+), 124 deletions(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index dbd8f507b..d7f40309a 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -4,12 +4,15 @@
 from outlines.types.dsl import (
     Regex,
     json_schema,
-    one_or_more,
-    optional,
     regex,
-    repeat,
+    either,
+    optional,
+    exactly,
+    at_least,
+    at_most,
+    between,
+    one_or_more,
     zero_or_more,
-    times,
 )
 
 # Python types
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 0246b28ef..73de3876c 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -103,6 +103,27 @@ def _display_children(self, indent: str) -> str:
     def __str__(self):
         return self.display_ascii_tree()
 
+    def optional(self) -> "Optional":
+        return optional(self)
+
+    def exactly(self, count: int) -> "QuantifyExact":
+        return exactly(count, self)
+
+    def at_least(self, count: int) -> "QuantifyMinimum":
+        return at_least(count, self)
+
+    def at_most(self, count: int) -> "QuantifyMaximum":
+        return at_most(count, self)
+
+    def between(self, min_count: int, max_count: int) -> "QuantifyBetween":
+        return between(min_count, max_count, self)
+
+    def one_or_more(self) -> "KleenePlus":
+        return one_or_more(self)
+
+    def zero_or_more(self) -> "KleeneStar":
+        return zero_or_more(self)
+
 
 @dataclass
 class String(Term):
@@ -297,64 +318,60 @@ def __repr__(self):
         return f"QuantifyBetween(term={repr(self.term)}, min_count={repr(self.min_count)}, max_count={repr(self.max_count)})"
 
 
-def either(*args: Union[str, Term]):
+def regex(pattern: str):
+    return Regex(pattern)
+
+
+def json_schema(schema: Union[str, dict, type[BaseModel]]):
+    return JsonSchema(schema)
+
+
+def either(*terms: Union[str, Term]):
     """Represents an alternative between different terms or strings.
 
     This factory function automatically translates string arguments
     into `String` objects.
     """
-    args = [String(arg) if arg is str else arg for arg in args]
-    return Alternatives(args)
+    terms = [String(arg) if isinstance(arg, str) else arg for arg in terms]
+    return Alternatives(terms)
 
 
-def optional(self: Union[Term, str]) -> Optional:
-    self = String(self) if self is str else self
-    return Optional(self)
+def optional(term: Union[Term, str]) -> Optional:
+    term = String(term) if isinstance(term, str) else term
+    return Optional(term)
 
 
-def one_or_more(self: Term) -> KleenePlus:
-    self = String(self) if self is str else self
-    return KleenePlus(self)
-
-
-def between(self: Term, min_count: int, max_count: int) -> QuantifyBetween:
-    self = String(self) if self is str else self
-    match (min_count, max_count):
-        case (None, None):
-            raise ValueError(
-                "repeat: you must provide a value for at least `min_count` or `max_count`"
-            )
-        case (_, None):
-            return QuantifyMinimum(self, min_count)
-        case (None, _):
-            return QuantifyMaximum(self, max_count)
-        case _:
-            return QuantifyBetween(self, min_count, max_count)
+def exactly(count: int, term: Union[Term, str]) -> QuantifyExact:
+    """Repeat the term exactly `count` times."""
+    term = String(term) if isinstance(term, str) else term
+    return QuantifyExact(term, count)
 
 
-def times(self: Term, count: int = 0) -> QuantifyExact:
-    self = String(self) if self is str else self
-    return QuantifyExact(self, count)
+def at_least(count: int, term: Union[Term, str]) -> QuantifyMinimum:
+    """Repeat the term at least `count` times."""
+    term = String(term) if isinstance(term, str) else term
+    return QuantifyMinimum(term, count)
 
 
-def zero_or_more(self: Term) -> KleeneStar:
-    self = String(self) if self is str else self
-    return KleeneStar(self)
+def at_most(count: int, term: Union[Term, str]) -> QuantifyMaximum:
+    """Repeat the term exactly `count` times."""
+    term = String(term) if isinstance(term, str) else term
+    return QuantifyMaximum(term, count)
 
 
-Term.one_or_more = one_or_more  # type: ignore
-Term.optional = optional  # type: ignore
-Term.repeat = repeat  # type: ignore
-Term.times = times  # type: ignore
-Term.zero_or_more = zero_or_more  # type: ignore
+def between(min_count: int, max_count: int, term: Union[Term, str]) -> QuantifyBetween:
+    term = String(term) if isinstance(term, str) else term
+    return QuantifyBetween(term, min_count, max_count)
 
 
-def regex(pattern: str):
-    return Regex(pattern)
+def zero_or_more(term: Union[Term, str]) -> KleeneStar:
+    term = String(term) if isinstance(term, str) else term
+    return KleeneStar(term)
 
 
-def json_schema(schema: Union[str, dict, type[BaseModel]]):
-    return JsonSchema(schema)
+def one_or_more(term: Union[Term, str]) -> KleenePlus:
+    term = String(term) if isinstance(term, str) else term
+    return KleenePlus(term)
 
 
 def to_regex(term: Term) -> str:
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index dc2779ef1..8351ef319 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -15,13 +15,16 @@
     Sequence,
     String,
     Term,
+    either,
     one_or_more,
+    zero_or_more,
     optional,
-    repeat,
-    times,
+    between,
+    at_most,
+    at_least,
+    exactly,
     regex,
     json_schema,
-    zero_or_more,
 )
 
 
@@ -96,117 +99,157 @@ def test_dsl_init():
         QuantifyBetween(string, 3, 1)
 
 
-def test_dsl_operations():
+def test_dsl_sequence():
     a = String("a")
     b = String("b")
-    assert isinstance(a + b, Sequence)
-    assert (a + b).terms[0] == a
-    assert (a + b).terms[1] == b
-
-    assert isinstance(a | b, Alternatives)
-    assert (a | b).terms[0] == a
-    assert (a | b).terms[1] == b
 
+    sequence = a + b
+    assert isinstance(sequence, Sequence)
+    assert sequence.terms[0] == a
+    assert sequence.terms[1] == b
 
-def test_dsl_operations_string_conversion():
-    b = String("b")
     sequence = "a" + b
     assert isinstance(sequence, Sequence)
     assert isinstance(sequence.terms[0], String)
     assert sequence.terms[0].value == "a"
     assert sequence.terms[1].value == "b"
 
-    sequence = b + "a"
+    sequence = a + "b"
     assert isinstance(sequence, Sequence)
-    assert isinstance(sequence.terms[0], String)
-    assert sequence.terms[0].value == "b"
-    assert sequence.terms[1].value == "a"
+    assert isinstance(sequence.terms[1], String)
+    assert sequence.terms[0].value == "a"
+    assert sequence.terms[1].value == "b"
 
-    alternative = "a" | b
-    assert isinstance(alternative, Alternatives)
-    assert isinstance(alternative.terms[0], String)
-    assert alternative.terms[0].value == "a"
-    assert alternative.terms[1].value == "b"
 
-    alternative = b | "a"
-    assert isinstance(alternative, Alternatives)
-    assert isinstance(alternative.terms[0], String)
-    assert alternative.terms[0].value == "b"
-    assert alternative.terms[1].value == "a"
+def test_dsl_alternatives():
+    a = String("a")
+    b = String("b")
 
+    alt = either(a, b)
+    assert isinstance(alt, Alternatives)
+    assert isinstance(alt.terms[0], String)
+    assert isinstance(alt.terms[1], String)
 
-def test_dsl_aliases():
-    test = regex("[0-9]")
-    assert isinstance(test, Regex)
+    alt = either("a", "b")
+    assert isinstance(alt, Alternatives)
+    assert isinstance(alt.terms[0], String)
+    assert isinstance(alt.terms[1], String)
 
-    test = json_schema('{"type": "string"}')
-    assert isinstance(test, JsonSchema)
+    alt = either("a", b)
+    assert isinstance(alt, Alternatives)
+    assert isinstance(alt.terms[0], String)
+    assert isinstance(alt.terms[1], String)
+
+
+def test_dsl_optional():
+    a = String("a")
+
+    opt = optional(a)
+    assert isinstance(opt, Optional)
+
+    opt = optional("a")
+    assert isinstance(opt, Optional)
+    assert isinstance(opt.term, String)
+
+    opt = a.optional()
+    assert isinstance(opt, Optional)
+
+
+def test_dsl_exactly():
+    a = String("a")
+
+    rep = exactly(2, a)
+    assert isinstance(rep, QuantifyExact)
+    assert rep.count == 2
+
+    rep = exactly(2, "a")
+    assert isinstance(rep, QuantifyExact)
+    assert isinstance(rep.term, String)
+
+    rep = a.exactly(2)
+    assert isinstance(rep, QuantifyExact)
+
+
+def test_dsl_at_least():
+    a = String("a")
+
+    rep = at_least(2, a)
+    assert isinstance(rep, QuantifyMinimum)
+    assert rep.min_count == 2
 
-    test = String("test")
+    rep = at_least(2, "a")
+    assert isinstance(rep, QuantifyMinimum)
+    assert isinstance(rep.term, String)
 
-    assert isinstance(test.times(3), QuantifyExact)
-    assert test.times(3).count == 3
-    assert test.times(3).term == test
+    rep = a.at_least(2)
+    assert isinstance(rep, QuantifyMinimum)
 
-    assert isinstance(times(test, 3), QuantifyExact)
-    assert times(test, 3).count == 3
-    assert times(test, 3).term == test
 
-    assert isinstance(test.one_or_more(), KleenePlus)
-    assert test.one_or_more().term == test
+def test_dsl_at_most():
+    a = String("a")
 
-    assert isinstance(one_or_more(test), KleenePlus)
-    assert one_or_more(test).term == test
+    rep = at_most(2, a)
+    assert isinstance(rep, QuantifyMaximum)
+    assert rep.max_count == 2
 
-    assert isinstance(test.zero_or_more(), KleeneStar)
-    assert test.zero_or_more().term == test
+    rep = at_most(2, "a")
+    assert isinstance(rep, QuantifyMaximum)
+    assert isinstance(rep.term, String)
 
-    assert isinstance(zero_or_more(test), KleeneStar)
-    assert zero_or_more(test).term == test
+    rep = a.at_most(2)
+    assert isinstance(rep, QuantifyMaximum)
 
-    assert isinstance(test.optional(), Optional)
-    assert test.optional().term == test
 
-    assert isinstance(optional(test), Optional)
-    assert optional(test).term == test
+def test_between():
+    a = String("a")
 
-    rep_min = test.repeat(2, None)
-    assert isinstance(rep_min, QuantifyMinimum)
-    assert rep_min.min_count == 2
+    rep = between(1, 2, a)
+    assert isinstance(rep, QuantifyBetween)
+    assert rep.min_count == 1
+    assert rep.max_count == 2
 
-    rep_min = repeat(test, 2, None)
-    assert isinstance(rep_min, QuantifyMinimum)
-    assert rep_min.min_count == 2
+    rep = between(1, 2, "a")
+    assert isinstance(rep, QuantifyBetween)
+    assert isinstance(rep.term, String)
 
-    rep_max = test.repeat(None, 2)
-    assert isinstance(rep_max, QuantifyMaximum)
-    assert rep_max.max_count == 2
+    rep = a.between(1, 2)
+    assert isinstance(rep, QuantifyBetween)
 
-    rep_max = repeat(test, None, 2)
-    assert isinstance(rep_max, QuantifyMaximum)
-    assert rep_max.max_count == 2
 
-    rep_between = test.repeat(1, 2)
-    assert isinstance(rep_between, QuantifyBetween)
-    assert rep_between.min_count == 1
-    assert rep_between.max_count == 2
+def test_dsl_zero_or_more():
+    a = String("a")
 
-    rep_between = repeat(test, 1, 2)
-    assert isinstance(rep_between, QuantifyBetween)
-    assert rep_between.min_count == 1
-    assert rep_between.max_count == 2
+    rep = zero_or_more(a)
+    assert isinstance(rep, KleeneStar)
 
-    with pytest.raises(ValueError, match="QuantifyBetween: `max_count` must be"):
-        test.repeat(2, 1)
+    rep = zero_or_more("a")
+    assert isinstance(rep, KleeneStar)
+    assert isinstance(rep.term, String)
 
-    with pytest.raises(ValueError, match="QuantifyBetween: `max_count` must be"):
-        repeat(test, 2, 1)
+    rep = a.zero_or_more()
+    assert isinstance(rep, KleeneStar)
 
-    with pytest.raises(ValueError, match="repeat: you must provide"):
-        test.repeat(None, None)
 
-    with pytest.raises(ValueError, match="repeat: you must provide"):
-        repeat(test, None, None)
+def test_dsl_one_or_more():
+    a = String("a")
+
+    rep = one_or_more(a)
+    assert isinstance(rep, KleenePlus)
+
+    rep = one_or_more("a")
+    assert isinstance(rep, KleenePlus)
+    assert isinstance(rep.term, String)
+
+    rep = a.zero_or_more()
+    assert isinstance(rep, KleeneStar)
+
+
+def test_dsl_aliases():
+    test = regex("[0-9]")
+    assert isinstance(test, Regex)
+
+    test = json_schema('{"type": "string"}')
+    assert isinstance(test, JsonSchema)
 
 
 def test_dsl_term_pydantic_simple():
@@ -230,7 +273,7 @@ def test_dsl_term_pydantic_combination():
     c = String("c")
 
     class Model(BaseModel):
-        field: (a + b) | c
+        field: either((a + b), c)
 
     schema = Model.model_json_schema()
     assert schema == {
@@ -247,7 +290,7 @@ def test_dsl_display():
     a = String("a")
     b = String("b")
     c = Regex("[0-9]")
-    d = KleeneStar(a | b) + c
+    d = Sequence([KleeneStar(Alternatives([a, b])), c])
 
     tree = str(d)
     assert (

From fde48c850bf5a7fa39053c033e6e36e2993edc15 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Tue, 4 Mar 2025 15:20:34 +0100
Subject: [PATCH 184/505] Update the documentation of the regex dsl

---
 docs/reference/regex_dsl.md | 114 ++++++++++++++++++++++--------------
 1 file changed, 71 insertions(+), 43 deletions(-)

diff --git a/docs/reference/regex_dsl.md b/docs/reference/regex_dsl.md
index aa04303cf..f42b6dd34 100644
--- a/docs/reference/regex_dsl.md
+++ b/docs/reference/regex_dsl.md
@@ -39,43 +39,47 @@ print(to_regex(digit))    # Output: [0-9]+
 
 ---
 
-## Early Introduction to Quantifiers & Operators
+## Early Introduction to Quantifiers & Combining Terms
 
 The DSL supports common regex quantifiers as methods on every `Term`. These methods allow you to specify how many times a pattern should be matched. They include:
 
-- **`times(count)`**: Matches the term exactly `count` times.
+- **`exactly(count)`**: Matches the term exactly `count` times.
 - **`optional()`**: Matches the term zero or one time.
 - **`one_or_more()`**: Matches the term one or more times (Kleene Plus).
 - **`zero_or_more()`**: Matches the term zero or more times (Kleene Star).
-- **`repeat(min_count, max_count)`**: Matches the term between `min_count` and `max_count` times (or open-ended if one value is omitted).
+- **`between(min_count, max_count)`**: Matches the term between `min_count` and `max_count` times (inclusive).
+- **`at_least(count)`**: Matches the term at least `count` times.
+- **`at_most(count)`**: Matches the term up to `count` times.
 
-Let’s see these quantifiers side by side with examples.
+These quantifiers can also be used as functions that take the `Term` as an argument. If the term is a plain string, it will be automatically converted to a `String` object. Thus `String("foo").optional()` is equivalent to `optional("foo")`.
+
+Let's see these quantifiers side by side with examples.
 
 ### Quantifiers in Action
 
-#### `times(count)`
+#### `exactly(count)`
 
 This method restricts the term to appear exactly `count` times.
 
 ```python
 # Example: exactly 5 digits
-five_digits = Regex(r"\d").times(5)
+five_digits = Regex(r"\d").exactly(5)
 print(to_regex(five_digits))  # Output: (\d){5}
 ```
 
-You can also use the `times` function:
+You can also use the `exactly` function:
 
 ```python
-from outlines.types import times
+from outlines.types import exactly
 
 # Example: exactly 5 digits
-five_digits = times(Regex(r"\d"), 5)
+five_digits = exactly(Regex(r"\d"), 5)
 print(to_regex(five_digits))  # Output: (\d){5}
 ```
 
 #### `optional()`
 
-The `optional()` method makes a term optional, meaning it may occur zero or one time.
+This method makes a term optional, meaning it may occur zero or one time.
 
 ```python
 # Example: an optional "s" at the end of a word
@@ -83,7 +87,7 @@ maybe_s = String("s").optional()
 print(to_regex(maybe_s))  # Output: (s)?
 ```
 
-You can also use the `optional` function, the string will automatically be converted to a `String` object:
+You can also use the `optional` function:
 
 ```python
 from outlines.types import optional
@@ -116,7 +120,7 @@ print(to_regex(letters))  # Output: ([A-Za-z])+
 
 #### `zero_or_more()`
 
-This method means that the term can occur zero or more times.
+This method indicates that the term can occur zero or more times.
 
 ```python
 # Example: zero or more spaces
@@ -124,7 +128,7 @@ spaces = String(" ").zero_or_more()
 print(to_regex(spaces))  # Output: ( )*
 ```
 
-You can also use the `zero_or_more` function, the string will automatically be converted to a `String` instance:
+You can also use the `zero_or_more` function:
 
 ```python
 from outlines.types import zero_or_more
@@ -134,40 +138,64 @@ spaces = zero_or_more(" ")
 print(to_regex(spaces))  # Output: ( )*
 ```
 
-#### `repeat(min_count, max_count)`
+#### `between(min_count, max_count)`
 
-The `repeat` method provides flexibility to set a lower and/or upper bound on the number of occurrences.
+This method indicates that the term can appear any number of times between `min_count` and `max_count` (inclusive).
 
 ```python
 # Example: Between 2 and 4 word characters
-word_chars = Regex(r"\w").repeat(2, 4)
+word_chars = Regex(r"\w").between(2, 4)
 print(to_regex(word_chars))  # Output: (\w){2,4}
-
-# Example: At least 3 digits (min specified, max left open)
-at_least_three = Regex(r"\d").repeat(3, None)
-print(to_regex(at_least_three))  # Output: (\d){3,}
-
-# Example: Up to 2 punctuation marks (max specified, min omitted)
-up_to_two = Regex(r"[,.]").repeat(None, 2)
-print(to_regex(up_to_two))  # Output: ([,.]){,2}
 ```
 
-You can also use the `repeat` function:
+You can also use the `between` function:
 
 ```python
-from outlines import repeat
+from outlines.types import between
 
 # Example: Between 2 and 4 word characters
-word_chars = repeat(Regex(r"\w"), 2, 4)
+word_chars = between(Regex(r"\w"), 2, 4)
 print(to_regex(word_chars))  # Output: (\w){2,4}
+```
+
+#### `at_least(count)`
 
-# Example: At least 3 digits (min specified, max left open)
-at_least_three = repeat(Regex(r"\d"), 3, None)
+This method indicates that the term must appear at least `count` times.
+
+```python
+# Example: At least 3 digits
+at_least_three = Regex(r"\d").at_least(3)
 print(to_regex(at_least_three))  # Output: (\d){3,}
+```
+
+You can also use the `at_least` function:
+
+```python
+from outlines.types import at_least
+
+# Example: At least 3 digits
+at_least_three = at_least(Regex(r"\d"), 3)
+print(to_regex(at_least_three))  # Output: (\d){3,}
+```
+
+#### `at_most(count)`
+
+This method indicates that the term can appear at most `count` times.
+
+```python
+# Example: At most 3 digits
+up_to_three = Regex(r"\d").at_most(3)
+print(to_regex(up_to_three))  # Output: (\d){0,3}
+```
+
+You can also use the `at_most` function:
+
+```python
+from outlines.types import at_most
 
-# Example: Up to 2 punctuation marks (max specified, min omitted)
-up_to_two = repeat(Regex(r"[,.]"), None, 2)
-print(to_regex(up_to_two))  # Output: ([,.]){,2}
+# Example: At most 3 digits
+up_to_three = at_most(Regex(r"\d"), 3)
+print(to_regex(up_to_three))  # Output: (\d){0,3}
 ```
 
 ---
@@ -186,17 +214,17 @@ pattern = String("hello") + " " + Regex(r"\w+")
 print(to_regex(pattern))  # Output: hello\ (\w+)
 ```
 
-### Alternation (`|`)
+### Alternation (`either()`)
 
-The `|` operator creates alternatives, allowing a match for one of several patterns.
+The `either()` function creates alternatives, allowing a match for one of several patterns. You can provide as many terms as you want.
 
 ```python
-# Example: Match either "cat" or "dog"
-animal = String("cat") | "dog"
-print(to_regex(animal))  # Output: (cat|dog)
+# Example: Match either "cat" or "dog" or "mouse"
+animal = either(String("cat"), "dog", "mouse")
+print(to_regex(animal))  # Output: (cat|dog|mouse)
 ```
 
-*Note:* When using operators with plain strings (such as `"dog"`), the DSL automatically wraps them in a `String` object and escapes the characters that have a special meaning in regular expressions.
+*Note:* When using `either()` with plain strings (such as `"dog"`), the DSL automatically wraps them in a `String` object that escapes the characters that have a special meaning in regular expressions, just like with quantifier functions.
 
 ---
 
@@ -223,7 +251,7 @@ For instance you can describe the answers in the GSM8K dataset using the followi
 ```python
 from outlines.types import sentence, digit
 
-answer = "A: " + sentence.repeat(2,4) + " So the answer is: " + digit.repeat(1,4)
+answer = "A: " + sentence.between(2,4) + " So the answer is: " + digit.between(1,4)
 ```
 
 ---
@@ -237,7 +265,7 @@ Suppose you want to create a regex that matches an ID format like "ID-12345", wh
 - Followed by exactly 5 digits.
 
 ```python
-id_pattern = "ID-" + Regex(r"\d").times(5)
+id_pattern = "ID-" + Regex(r"\d").exactly(5)
 print(to_regex(id_pattern))  # Output: ID-(\d){5}
 ```
 
@@ -273,9 +301,9 @@ When used in a Pydantic model, the email field is automatically validated agains
 Consider a pattern to match a simple date format: `YYYY-MM-DD`.
 
 ```python
-year = Regex(r"\d").times(4)         # Four digits for the year
-month = Regex(r"\d").times(2)         # Two digits for the month
-day = Regex(r"\d").times(2)           # Two digits for the day
+year = Regex(r"\d").exactly(4)         # Four digits for the year
+month = Regex(r"\d").exactly(2)        # Two digits for the month
+day = Regex(r"\d").exactly(2)          # Two digits for the day
 
 # Combine with literal hyphens
 date_pattern = year + "-" + month + "-" + day

From 9dd927d5fdb3b39560092d2a9881a7bdc7dc374f Mon Sep 17 00:00:00 2001
From: Frederik Fix <ich@derfred.com>
Date: Tue, 4 Mar 2025 17:08:35 +0100
Subject: [PATCH 185/505] fix schema schema issue

---
 outlines/generate/json.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/outlines/generate/json.py b/outlines/generate/json.py
index 008662206..ab05df1f4 100644
--- a/outlines/generate/json.py
+++ b/outlines/generate/json.py
@@ -92,7 +92,9 @@ def json_openai(
         )
 
     if isinstance(schema_object, type(BaseModel)):
-        schema = pyjson.dumps(schema_object.model_json_schema())
+        schema = schema_object.model_json_schema()
+        schema["additionalProperties"] = False
+        schema = pyjson.dumps(schema)
         format_sequence = lambda x: schema_object.parse_raw(x)
     elif isinstance(schema_object, str):
         schema = schema_object

From d1a9adaddc2e4bf254fed695408e9c362d705fbf Mon Sep 17 00:00:00 2001
From: Nicholas Chammas <nicholas.chammas@gmail.com>
Date: Wed, 5 Mar 2025 12:08:43 -0500
Subject: [PATCH 186/505] Tweak version commands in bug_report.yml

The way these commands show up when a user fills out a new bug report is misleading, as they all get mashed into one line.

Copy-pasting this results in only the first command running, which is presumably not what the template author intended.
---
 .github/ISSUE_TEMPLATE/bug_report.yml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/.github/ISSUE_TEMPLATE/bug_report.yml b/.github/ISSUE_TEMPLATE/bug_report.yml
index af90ed988..cbc67f624 100644
--- a/.github/ISSUE_TEMPLATE/bug_report.yml
+++ b/.github/ISSUE_TEMPLATE/bug_report.yml
@@ -63,9 +63,9 @@ body:
       label: "Outlines/Python version information:"
       description: |
           Please run the following code and paste the output here.
-          python -c "from outlines import _version; print(_version.__version__)"
-          python -c "import sys; print('Python', sys.version)"
-          pip freeze
+          python -c "from outlines import _version; print(_version.__version__)";
+          python -c "import sys; print('Python', sys.version)";
+          pip freeze;
       value: |
           Version information
           <details>

From 433581ace44a045faa47c2fea3ecb05d8f1063f8 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Mon, 10 Mar 2025 11:31:32 -0700
Subject: [PATCH 187/505] Fix capitalization in installation.md

---
 docs/installation.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/installation.md b/docs/installation.md
index 5f8bb2d13..f88bc7125 100644
--- a/docs/installation.md
+++ b/docs/installation.md
@@ -1,5 +1,5 @@
 ---
-title: installation
+title: Installation
 ---
 
 # Installation

From fc6192f5d6ec67c4cb0ce9b3f5b96e04aa5f8931 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 15 Mar 2025 11:37:52 +0100
Subject: [PATCH 188/505] Remove all issue templates except bug reports

---
 .github/ISSUE_TEMPLATE/doc_request.md     | 21 --------------
 .github/ISSUE_TEMPLATE/feature_request.md | 27 ------------------
 .github/ISSUE_TEMPLATE/improvement.yml    | 34 -----------------------
 3 files changed, 82 deletions(-)
 delete mode 100644 .github/ISSUE_TEMPLATE/doc_request.md
 delete mode 100644 .github/ISSUE_TEMPLATE/feature_request.md
 delete mode 100644 .github/ISSUE_TEMPLATE/improvement.yml

diff --git a/.github/ISSUE_TEMPLATE/doc_request.md b/.github/ISSUE_TEMPLATE/doc_request.md
deleted file mode 100644
index f9dd65f92..000000000
--- a/.github/ISSUE_TEMPLATE/doc_request.md
+++ /dev/null
@@ -1,21 +0,0 @@
----
-name: 📚 Documentation
-about: Help with documentation improvements.
-title: ''
-labels: ''
-assignees: ''
----
-
-### Documentation request
-
-Documentation is a high priority for outlines. If you would like to see something added to the documentation, please help us out by:
-
-- Checking to see if the documentation does not already exist.
-- Providing a detailed explanation of what you'd like to understand better.
-- Highlighting any specific difficulties you ran into.
-
-### Are you willing to open a PR?
-
-Tell us whether you are willing to submit a PR for this yourself, we'd greatly appreciate it!
-
-Thanks for contributing!
diff --git a/.github/ISSUE_TEMPLATE/feature_request.md b/.github/ISSUE_TEMPLATE/feature_request.md
deleted file mode 100644
index 702a41bd1..000000000
--- a/.github/ISSUE_TEMPLATE/feature_request.md
+++ /dev/null
@@ -1,27 +0,0 @@
----
-name: 🚀 New feature
-about: Request a new feature.
-title: ''
-labels: ''
-assignees: ''
----
-
-Before suggesting a new feature, please make sure this hasn't already been suggested by searching through the past issues and the PR tracker.
-
-### Presentation of the new feature
-
-Assume that we know nothing about this feature. Please give us as much information as possible so we can judge it fairly. That includes (but is not limited to):
-- Academic article
-- Blog posts
-- Implementations
-- Personnal experience with the feature
-
-### Where does it fit in Outlines?
-
-Please explain to us why you are suggesting this feature for integration in Outlines.
-
-### Are you willing to open a PR?
-
-Tell us whether you are willing to add the feature yourself, and if so if you can share a design plan. **You may be challenged.**
-
-Thanks for contributing!
diff --git a/.github/ISSUE_TEMPLATE/improvement.yml b/.github/ISSUE_TEMPLATE/improvement.yml
deleted file mode 100644
index 64212eedd..000000000
--- a/.github/ISSUE_TEMPLATE/improvement.yml
+++ /dev/null
@@ -1,34 +0,0 @@
-name: ✨ Improvement
-description: Propose an improvement to Outlines.
-title: "<Description of the proposed change>"
-labels: "enhancement"
-
-body:
-  - type: markdown
-    attributes:
-      value: >-
-        Before suggesting an improvement, please make sure this hasn't already been suggested by searching through the past issues and the PR tracker.
-
-        ## Current behavior
-
-  - type: textarea
-    attributes:
-      label: "What behavior of the library made you think about the improvement?"
-      placeholder: |
-        Sample code with the current behavior
-
-  - type: markdown
-    attributes:
-      value: >-
-        ## Desired behavior
-
-  - type: textarea
-    attributes:
-      label: "How would you like it to behave?"
-      placeholder: |
-        Sample code with the desired behavior
-
-  - type: markdown
-    attributes:
-      value: >-
-        **Be aware that your proposal may be challenged.** Outlines has strong design principles that we are committed to stick to. If there is an alternative way to achieve what you would like, we will let you know.

From 05d775fe628ec9929e65ee150bc5322b17dd384f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 3 Apr 2025 12:01:49 +0200
Subject: [PATCH 189/505] Update the logo

---
 README.md                              |   8 +-
 docs/assets/images/dottxt-dark.svg     |   1 +
 docs/assets/images/dottxt-light.svg    |  32 ++++
 docs/assets/images/logo-dark-mode.png  | Bin 0 -> 7425 bytes
 docs/assets/images/logo-dark-mode.svg  | 207 ++++++++++++++++++++++
 docs/assets/images/logo-light-mode.png | Bin 0 -> 9174 bytes
 docs/assets/images/logo-light-mode.svg |   1 +
 docs/assets/images/logo-simple.png     | Bin 2196 -> 3323 bytes
 docs/assets/images/logo-square.png     | Bin 2233 -> 3323 bytes
 docs/assets/images/logo-square.svg     | 229 +++++++++++++------------
 docs/assets/images/logo.png            | Bin 6386 -> 0 bytes
 docs/assets/images/logo.svg            | 134 ---------------
 docs/overrides/home.html               |   3 +-
 13 files changed, 367 insertions(+), 248 deletions(-)
 create mode 100644 docs/assets/images/dottxt-dark.svg
 create mode 100644 docs/assets/images/dottxt-light.svg
 create mode 100644 docs/assets/images/logo-dark-mode.png
 create mode 100644 docs/assets/images/logo-dark-mode.svg
 create mode 100644 docs/assets/images/logo-light-mode.png
 create mode 100644 docs/assets/images/logo-light-mode.svg
 delete mode 100644 docs/assets/images/logo.png
 delete mode 100644 docs/assets/images/logo.svg

diff --git a/README.md b/README.md
index 01d7b9c98..419282561 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,7 @@
 <div align="center" style="margin-bottom: 1em;">
 
-<img src="./docs/assets/images/logo.png" alt="Outlines Logo" width=500></img>
+<img src="./docs/assets/images/logo-light-mode.svg#gh-light-mode-only" alt="Outlines Logo" width=500></img>
+<img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=500></img>
 
 
  🗒️ *Make LLMs speak the language of every application.* 🗒️
@@ -48,10 +49,11 @@ Outlines  has new releases and features coming every week. Make sure to ⭐ star
 ## .txt company
 
 <div align="center">
-<img src="./docs/assets/images/dottxt.png" alt="Outlines Logo" width=100></img>
+<img src="./docs/assets/images/dottxt-light.svg#gh-light-mode-only" alt="dottxt logo" width=100></img>
+<img src="./docs/assets/images/dottxt-dark.svg#gh-dark-mode-only" alt="dottxt logo" width=100></img>
 </div>
 
-We started a company to keep pushing the boundaries of structured generation. Learn more about [.txt](https://twitter.com/dottxtai), and  [give our .json API a try](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF) if you need a hosted solution ✨
+We started a company to keep pushing the boundaries of structured generation. Learn more about [.txt](https://dottxt.ai), and  [license our commercial product](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF) if you need SOTA performance and quality ✨
 
 ## Structured generation
 
diff --git a/docs/assets/images/dottxt-dark.svg b/docs/assets/images/dottxt-dark.svg
new file mode 100644
index 000000000..a95ff7d32
--- /dev/null
+++ b/docs/assets/images/dottxt-dark.svg
@@ -0,0 +1 @@
+<?xml version="1.0" encoding="UTF-8"?><svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 129.09 196.26"><defs><style>.cls-1{fill:#fff;}</style></defs><g id="Layer_1-2"><rect class="cls-1" x="39.12" y="3.91" width="3.91" height="3.91"/><polygon class="cls-1" points="46.94 7.82 50.86 7.82 54.77 7.82 54.77 3.91 50.86 3.91 50.86 0 46.94 0 43.03 0 43.03 3.91 46.94 3.91 46.94 7.82"/><polygon class="cls-1" points="58.68 7.82 62.59 7.82 66.5 7.82 66.5 3.91 62.59 3.91 62.59 0 58.68 0 54.77 0 54.77 3.91 58.68 3.91 58.68 7.82"/><polygon class="cls-1" points="70.42 7.82 74.33 7.82 78.24 7.82 78.24 3.91 74.33 3.91 74.33 0 70.42 0 66.5 0 66.5 3.91 70.42 3.91 70.42 7.82"/><polygon class="cls-1" points="82.15 7.82 86.06 7.82 86.06 11.74 89.98 11.74 93.89 11.74 97.8 11.74 97.8 7.82 101.71 7.82 101.71 3.91 97.8 3.91 97.8 0 93.89 0 89.98 0 89.98 3.91 93.89 3.91 93.89 7.82 89.98 7.82 89.98 3.91 86.06 3.91 86.06 0 82.15 0 78.24 0 78.24 3.91 82.15 3.91 82.15 7.82"/><polygon class="cls-1" points="105.62 7.82 109.53 7.82 113.45 7.82 113.45 11.74 117.36 11.74 121.27 11.74 121.27 7.82 117.36 7.82 117.36 3.91 113.45 3.91 109.53 3.91 109.53 0 105.62 0 101.71 0 101.71 3.91 105.62 3.91 105.62 7.82"/><rect class="cls-1" x="43.03" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="54.77" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="66.5" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="78.24" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="101.71" y="7.82" width="3.91" height="3.91"/><polygon class="cls-1" points="39.12 39.12 43.03 39.12 46.94 39.12 50.86 39.12 54.77 39.12 58.68 39.12 62.59 39.12 66.5 39.12 70.42 39.12 74.33 39.12 78.24 39.12 78.24 35.21 74.33 35.21 70.42 35.21 66.5 35.21 62.59 35.21 58.68 35.21 54.77 35.21 50.86 35.21 46.94 35.21 43.03 35.21 39.12 35.21 39.12 39.12"/><polygon class="cls-1" points="66.5 50.86 62.59 50.86 58.68 50.86 54.77 50.86 50.86 50.86 46.94 50.86 43.03 50.86 39.12 50.86 35.21 50.86 31.3 50.86 31.3 54.77 35.21 54.77 39.12 54.77 43.03 54.77 46.94 54.77 50.86 54.77 54.77 54.77 58.68 54.77 62.59 54.77 66.5 54.77 70.42 54.77 74.33 54.77 74.33 50.86 70.42 50.86 66.5 50.86"/><polygon class="cls-1" points="43.03 101.71 46.94 101.71 50.86 101.71 54.77 101.71 58.68 101.71 62.59 101.71 66.5 101.71 70.42 101.71 74.33 101.71 78.24 101.71 82.15 101.71 86.06 101.71 89.98 101.71 93.89 101.71 97.8 101.71 101.71 101.71 105.62 101.71 109.53 101.71 109.53 97.8 105.62 97.8 101.71 97.8 97.8 97.8 93.89 97.8 89.98 97.8 86.06 97.8 82.15 97.8 78.24 97.8 74.33 97.8 70.42 97.8 66.5 97.8 62.59 97.8 58.68 97.8 54.77 97.8 50.86 97.8 46.94 97.8 43.03 97.8 43.03 93.89 43.03 89.97 46.94 89.97 50.86 89.97 54.77 89.97 58.68 89.97 62.59 89.97 66.5 89.97 70.42 89.97 74.33 89.97 78.24 89.97 78.24 86.06 82.15 86.06 86.06 86.06 89.98 86.06 93.89 86.06 97.8 86.06 101.71 86.06 105.62 86.06 109.53 86.06 109.53 82.15 105.62 82.15 101.71 82.15 97.8 82.15 93.89 82.15 89.98 82.15 86.06 82.15 82.15 82.15 82.15 78.24 82.15 74.33 86.06 74.33 86.06 70.42 89.98 70.42 93.89 70.42 97.8 70.42 101.71 70.42 105.62 70.42 109.53 70.42 109.53 66.5 105.62 66.5 101.71 66.5 97.8 66.5 93.89 66.5 89.98 66.5 89.98 62.59 89.98 58.68 93.89 58.68 93.89 54.77 97.8 54.77 101.71 54.77 105.62 54.77 109.53 54.77 109.53 50.86 105.62 50.86 101.71 50.86 97.8 50.86 97.8 46.94 97.8 43.03 101.71 43.03 101.71 39.12 105.62 39.12 109.53 39.12 109.53 35.21 105.62 35.21 105.62 31.3 109.53 31.3 109.53 27.38 109.53 23.47 109.53 19.56 113.45 19.56 113.45 15.65 113.45 11.74 109.53 11.74 105.62 11.74 105.62 15.65 105.62 19.56 101.71 19.56 101.71 23.47 101.71 27.38 97.8 27.38 97.8 31.3 97.8 35.21 93.89 35.21 93.89 39.12 93.89 43.03 89.98 43.03 89.98 46.94 89.98 50.86 86.06 50.86 86.06 54.77 86.06 58.68 82.15 58.68 82.15 62.59 82.15 66.5 78.24 66.5 78.24 70.42 78.24 74.33 74.33 74.33 74.33 78.24 74.33 82.15 70.42 82.15 66.5 82.15 62.59 82.15 58.68 82.15 54.77 82.15 50.86 82.15 46.94 82.15 43.03 82.15 39.12 82.15 35.21 82.15 31.3 82.15 27.38 82.15 23.47 82.15 19.56 82.15 15.65 82.15 11.74 82.15 7.82 82.15 7.82 78.24 7.82 74.33 7.82 70.42 11.74 70.42 11.74 66.5 11.74 62.59 15.65 62.59 15.65 58.68 15.65 54.77 19.56 54.77 19.56 50.86 19.56 46.94 23.47 46.94 23.47 43.03 23.47 39.12 27.38 39.12 27.38 35.21 27.38 31.3 31.3 31.3 31.3 27.38 31.3 23.47 35.21 23.47 35.21 19.56 35.21 15.65 39.12 15.65 39.12 11.74 39.12 7.82 35.21 7.82 31.3 7.82 31.3 11.74 31.3 15.65 27.38 15.65 27.38 19.56 27.38 23.47 23.47 23.47 23.47 27.38 23.47 31.3 19.56 31.3 19.56 35.21 19.56 39.12 15.65 39.12 15.65 43.03 15.65 46.94 11.74 46.94 11.74 50.86 11.74 54.77 7.82 54.77 7.82 58.68 7.82 62.59 3.91 62.59 3.91 66.5 3.91 70.42 0 70.42 0 74.33 0 78.24 0 82.15 0 86.06 3.91 86.06 7.82 86.06 7.82 89.97 11.74 89.97 15.65 89.97 19.56 89.97 23.47 89.97 27.38 89.97 31.3 89.97 35.21 89.97 35.21 93.89 35.21 97.8 35.21 101.71 35.21 105.62 35.21 109.53 39.12 109.53 43.03 109.53 43.03 105.62 43.03 101.71"/><polygon class="cls-1" points="125.18 11.74 121.27 11.74 121.27 15.65 121.27 19.56 121.27 23.47 121.27 27.38 121.27 31.3 121.27 35.21 121.27 39.12 121.27 43.03 121.27 46.94 121.27 50.86 121.27 54.77 121.27 58.68 121.27 62.59 121.27 66.5 121.27 70.42 121.27 74.33 121.27 78.24 121.27 82.15 121.27 86.06 121.27 89.97 121.27 93.89 121.27 97.8 121.27 101.71 121.27 105.62 121.27 109.53 117.36 109.53 117.36 113.45 113.45 113.45 109.53 113.45 105.62 113.45 101.71 113.45 97.8 113.45 93.89 113.45 89.98 113.45 86.06 113.45 82.15 113.45 78.24 113.45 74.33 113.45 70.42 113.45 66.5 113.45 62.59 113.45 58.68 113.45 54.77 113.45 50.86 113.45 50.86 109.53 46.94 109.53 43.03 109.53 43.03 113.45 43.03 117.36 46.94 117.36 50.86 117.36 50.86 121.27 54.77 121.27 58.68 121.27 62.59 121.27 66.5 121.27 70.42 121.27 74.33 121.27 78.24 121.27 82.15 121.27 86.06 121.27 89.98 121.27 93.89 121.27 97.8 121.27 101.71 121.27 105.62 121.27 109.53 121.27 113.45 121.27 117.36 121.27 117.36 117.36 121.27 117.36 125.18 117.36 125.18 113.45 125.18 109.53 129.09 109.53 129.09 105.62 129.09 101.71 129.09 97.8 129.09 93.89 129.09 89.97 129.09 86.06 129.09 82.15 129.09 78.24 129.09 74.33 129.09 70.42 129.09 66.5 129.09 62.59 129.09 58.68 129.09 54.77 129.09 50.86 129.09 46.94 129.09 43.03 129.09 39.12 129.09 35.21 129.09 31.3 129.09 27.38 129.09 23.47 129.09 19.56 129.09 15.65 129.09 11.74 125.18 11.74"/><path class="cls-1" d="M.31,196.26v-7.35h7.35v7.35H.31Z"/><path class="cls-1" d="M19.53,196.26v-44.17H4.81v-7.37h36.79v7.37h-14.72v44.17h-7.35Z"/><path class="cls-1" d="M46.63,196.26v-11.02h3.67v-3.7h3.67v-3.67h3.67v-3.67h3.7v-7.37h-3.7v-3.67h-3.67v-3.67h-3.67v-3.7h-3.67v-11.05h7.35v7.37h3.67v3.67h3.7v3.7h3.67v3.67h3.67v-3.67h3.7v-3.7h3.67v-3.67h3.67v-7.37h7.37v11.05h-3.7v3.7h-3.67v3.67h-3.67v3.67h-3.67v7.37h3.67v3.67h3.67v3.67h3.67v3.7h3.7v11.02h-7.37v-7.35h-3.67v-3.67h-3.67v-3.7h-3.7v-3.67h-3.67v3.67h-3.67v3.7h-3.7v3.67h-3.67v7.35h-7.35Z"/><path class="cls-1" d="M106.9,196.26v-44.17h-14.72v-7.37h36.79v7.37h-14.72v44.17h-7.35Z"/></g></svg>
diff --git a/docs/assets/images/dottxt-light.svg b/docs/assets/images/dottxt-light.svg
new file mode 100644
index 000000000..567e239ee
--- /dev/null
+++ b/docs/assets/images/dottxt-light.svg
@@ -0,0 +1,32 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" version="1.1" viewBox="0 0 129.1 196.3">
+  <!-- Generator: Adobe Illustrator 29.0.1, SVG Export Plug-In . SVG Version: 2.1.0 Build 192)  -->
+  <defs>
+    <style>
+      .st0 {
+        fill: #1d1d1b;
+      }
+    </style>
+  </defs>
+  <g id="Layer_1-2">
+    <rect class="st0" x="39.1" y="3.9" width="3.9" height="3.9"/>
+    <path class="st0" d="M46.9,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M58.7,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M70.4,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M82.2,7.8h3.9v3.9h11.7v-3.9h3.9v-3.9h-3.9V0h-7.8v3.9h3.9v3.9h-3.9v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M105.6,7.8h7.8v3.9h7.8v-3.9h-3.9v-3.9h-7.8V0h-7.8v3.9h3.9v3.9Z"/>
+    <rect class="st0" x="43" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="54.8" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="66.5" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="78.2" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="101.7" y="7.8" width="3.9" height="3.9"/>
+    <path class="st0" d="M39.1,39.1h39.1v-3.9h-39.1v3.9Z"/>
+    <path class="st0" d="M31.3,50.9v3.9h43v-3.9H31.3Z"/>
+    <path class="st0" d="M43,101.7h66.5v-3.9H43v-7.8h35.2v-3.9h31.3v-3.9h-27.4v-7.8h3.9v-3.9h23.5v-3.9h-19.5v-7.8h3.9v-3.9h15.6v-3.9h-11.7v-7.8h3.9v-3.9h7.8v-3.9h-3.9v-3.9h3.9v-11.7h3.9v-7.8h-7.8v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8H7.8v-11.7h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h-7.8v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8H0v15.6h7.8v3.9h27.4v19.6h7.8v-7.8Z"/>
+    <path class="st0" d="M125.2,11.7h-3.9v97.8h-3.9v3.9H50.9v-3.9h-7.8v7.8h7.8v3.9h66.5v-3.9h7.8v-7.8h3.9V11.7h-3.9Z"/>
+    <path class="st0" d="M.3,196.3v-7.4h7.3v7.4H.3Z"/>
+    <path class="st0" d="M19.5,196.3v-44.2H4.8v-7.4h36.8v7.4h-14.7v44.2h-7.4Z"/>
+    <path class="st0" d="M46.6,196.3v-11h3.7v-3.7h3.7v-3.7h3.7v-3.7h3.7v-7.4h-3.7v-3.7h-3.7v-3.7h-3.7v-3.7h-3.7v-11.1h7.3v7.4h3.7v3.7h3.7v3.7h3.7v3.7h3.7v-3.7h3.7v-3.7h3.7v-3.7h3.7v-7.4h7.4v11.1h-3.7v3.7h-3.7v3.7h-3.7v3.7h-3.7v7.4h3.7v3.7h3.7v3.7h3.7v3.7h3.7v11h-7.4v-7.4h-3.7v-3.7h-3.7v-3.7h-3.7v-3.7h-3.7v3.7h-3.7v3.7h-3.7v3.7h-3.7v7.4h-7.4Z"/>
+    <path class="st0" d="M106.9,196.3v-44.2h-14.7v-7.4h36.8v7.4h-14.7v44.2h-7.3Z"/>
+  </g>
+</svg>
diff --git a/docs/assets/images/logo-dark-mode.png b/docs/assets/images/logo-dark-mode.png
new file mode 100644
index 0000000000000000000000000000000000000000..24c5f3bc36cd271902376dcaebc7c667f9e909a4
GIT binary patch
literal 7425
zcmZvhcRX8v+sCP*cJ0ltTB}y6J^M3CQLFZficwNxC#0xq)d;G#7FD}NY&ApeQKQsM
z#HiX5lpw;B-oN|xyzb}q+<)XG=OpKxoa_6&KJU+UedEkb^ciS*XvxUP7#<ktT9A>E
z|0aEoy+%daGXKr}lJrL7XJ8jdM#irA=S7}pUXw~X$Q`6-8}tb55d?W0;7$gCK%~8V
zL4m->e(uuX0MB=O>O5p*d`S;<|F(p_-$ldT^I7HM_s7cfxLFLYM7j8Cpo~kXuNesF
zz7eZV$V{iyv#xVfH9t0gc@(2$4aG(2`#Et?vgKRZG7EW#$kJUI)=M#Bvz^s@Gwdp-
zu`HCG_tKB<&70wopj=|%r^z2bnzpj{HVY2Jk<LWJ8H7o0pohp^WYwcm6K;@H;SbCY
z|6<u+5^F<W`ZR)&?jReiy2#wVRCIF@N_Kkr%Q2ux;9C75q%UiGNPr??zV(qr?ZT$p
z%B`iO13;99=pq_o6#^ozj8axLFlA`HOjyw<Pk=PmxsCFnX6Id)`-3?~>^puH9QY)}
z#=`R^HyC4MV+S_G3rM^2<ibJB%qb!+x1DrMWi(IjR1hMEtzA*CC4Uv?E@Tla*IG}D
zcbr@n>FdwM-i+=l!YH(8t=eg!!jNmpz>&LiJ~Oucw$?9`PE9{(jkhvT9aOz`*cp31
z7&Xq?Il7Hu42cGvAkO#RNkl=e3PoNC8Kt??*XHdnIG`0%*fm055A%dUFc!MHy51G8
z+JJoHXPvKFhdFAG1y?k}F$<w1ADa)I`cHwc8ZT8Ab5&&^Bkx0I;wnG;jWq1<O=YV!
zapc&xGyuPRm4)=RK8ob0FG`KejPfdGyn~zF*_&$B=~t)Yvn<K=Tt%0wblr=rJ(P_m
z0-|yd@eq98qXMLmkdW2no7UJ^16DI_PG)AWE0j8W+nnEW^Zi{3l(cPh-A#u~brA01
zq+Ex-=9#Z_QAAdL5NEm%*LoMc0KCrZHt#wuDGh;x-s?w|)-Kf8_66S^eQQ<&e16<{
ze3Q}fjSUm(+79=%H76bFy2Y>=Y#=ib(I<EF`j*1<uQRMN!gd(B9+`L?u#_7$zRMBl
z^I^2>q;?9C5MD|8Z$F;UP&@XYe(L)5wn$w4$pB;IS!dq5qDK?)m3su2*g?-uUdQPg
zj|oJ0^sb*#6}M;iPTo{SZNH?+`?zoXs~+#Wpb+mSm+UAv*XCD~6a{$8_Y}|DJ668R
zEtgYetkuEJLZl_@sz7@MN&b1#7)DXvoxvzoXX+Ab<@>t2MM4bC|1n2Cyp$Fa`hL~^
z``~Dkc1`2>aeHMAjUXM>ic--*(wsW#_*=M0Pe(d>kpp)3QTcLWswhAJqgWq!BfDT1
z(VU(X$28Is0VIGl=6oP45|m04O+6ydLiGD0IxL+2j(*qZ!{9rX^DYd!TKC%YHpDp0
zb=yU~x*c=iX$^L?xX5}8A!O>%?d`B%T&O7qo>k;EH0kXUpcDB!=UH7<Pb~@40TD3M
zt@!&5GD6><NVz~{06lIc7@V8(=DLd8+GKR6ivojS=(j?hPBVr6YVRMoAhFgraLi0C
zv2sIb+XP|O|J$$A-eEKAqBpMVt4ho%CUVA~R^)aEKNpHC3-`LK2859X4`biO8dN?R
za3aNrM(jUx=)-oBMMG3Czo!F8GfPmxos;sD+Kkg%6Op9R1n#^f86&5u4+0A5LA9&$
zcLv4W9b!;*o)wOS0IB(1tl|53ojEJfI=RYdPkuk*A;RW%(~Z!9i*AuXyEIk)8OOV=
zrT7jpcRikmG)|v9DpgpND<_Bpz!7y9XXlZeaGuHRU%)%LiYsH!U=?fI_a3se0F`^}
zIqr*tVnib-%d*CuMC&sB3x9akKqw414jMn)fc(k(e~)R}xSp<V&hx4QBRh)09u_$I
zCRbfnbb|UzL>fqfvwR+NeN#Mmuj0?cy)FdPH#)7$#M~14E}u4DEJ>H?dgxhglQI$q
zDIKlw&oqDp^F_-tv$SM`q_k-CMH^I+MMLQfdb)jT|7|)2F*uRgw);6Rq@f$JoJqW{
zH7`7;INTdwd~-1FxmU0A*`13_<n_l#QR3!|;NV97EN3}U@I*E`#2D(2NjC+Tn=`rd
z@e0&N*JOp>FtppQZeAoZm$M58k-HcU`-%GQ4Em`D+;2Rtf|`08TPG(#@I))+3};j*
z_jDfb+6jDHog-=Y39s=L4TX(7cpw3>;XuJRz7Y{SW$(mtcBU6Nf43%sZp$3<L(;Sn
zg69|sV&vJ_a(t4gAq|No29neZAMO5qut%@6BWgUCBucpryA>pT9hlmrM;qUr&a=vI
z`3#h8`g^(3D3lMTo-yXB#PExGiBMpXe}z%VX2a?p4QVd*Q9g`k{i8pI8>GB{_(57{
zuWGV|@?Gc$fnb|4O{emgbYt8j-E%&`r+~=X)h}i>qV%JEjtv!`odhu<7x{_3zO+=X
zI?-VHPj|_5_LkP9q*#3i=opy<-Qmqu8M{TCg3q|7odk+2WRKhChR=Kv?u5|r2O_>#
zSkl$LKH7ILuL{sR;}pG^w(W2#G5D!J(Wt!FzUk&#W}%Yc^kWf?$D40grbkUO?N!U-
zZ6*vC=vL9;DMKZ#EO_8-YruoUZ72l1cy%<BbM#$(Vep1l;ZeE8(fOSY$)m}fDXZcj
zoAqU`)&R-*0Ql)hg?sy@EQiml;?!IW1<Vl5E2l09$T%~YS_yuzvA_2N6y>MQ*EFA5
zEKMyv=8FKJnmaqAyW|vCN8=D1;UObxN&?)|O_=JI73$)eAeXT~-l)Vb+pxi|Nk8<w
zV%p$N5|A;`Vyb1-xxuHc)A2(9KoV?qd&Q;H-gu3PdV<fg`5M>AJWqp7t~Er+_a&D1
zxdJmw{g~gKJCy-7UsK`Vo+dO*uvD6+Z#9LcLi#g;MP0p*FUu<63aNb`b6Gr?*Sdc{
zv2*w#))IsISLU#^{0BmUn`0I)%XNNnD+NL{wyH9Mn?JXU2H$U%qexIg0-ajx;2@}+
z2kN*@JLzyZ$j0?D_vkD8u12|pH_D0GD}dt{9un-x4d0!f{_X*)H>;%+J%R-Pbkc{w
z?UY*8g5Ar3_l0te-(US{IvX(Z+rH5}33xZ(RFqY8owAbX)K-cuVR;QVP|g0rHhYsT
z0wT2__HE~#iAV%nQlBIyZ!R@Vm=9s3oy@?{(2{hwkTk!(>1I#4mbzx;q61u9j%yn>
zUE+e|WGMpN;g837jxBUU`1$*?l3YGm5At<(UV+=avC&eeosusAD)S6;g6uunD`W+w
zuJQC!dgqKsjV)_Pd}Ka;@fEvYy?sF@X0X8%@Gr84DJR6@`dk)*S!b;eF3uVoGZ2rP
z?^w9^sXX)Zt6ouMU@i4UKoKpmx;FVY@YZT?{7>(b!k4p_ZKwxd^8j5*eiUQ}u#Gk9
z+TjcF|8RqjCw8j)hD5%=RVt%hK?9KxU4C|f(dyF=>0^pP1vgTh%t;CSn>Q~su2WlT
z6<x76rO~OeHYD%8Cd@o1Oz$v)_^dJDl1yD<DeZ;lvz^+XS>E8yZfqqo1-c)}JZRH`
zYEvxPs(y?ZsPjt}Ka8rmq60l}+6-ab0esJf_@D=?-9t>V;ps_End;9z`BPy4Mf=kS
zJ~;~?e^&|t*SUtf%;QyM*LyC6b3^HvcdN58UL<N1?Kl3B5tD;Yf8`phGxC`?7iEL^
zX)|sts@1LXcMh{Bhcxr;n0MJK9G=X(W*0?%v}yQ3x8Cs3i+`TExS}iXh0)^Mis{*g
zd-`!YA@SKhG~238)(gm#b5#5TQ8gsZ@pmzIDMujcJpZ2LDqq0^|37;2FOB(MUODM-
zGIKj{e`W2VchI*^pQ8$K5Sukzy=ZiB`SRBnG#tA5-sh3%thhr=JbGpJh0^0Haa+)9
z_>J>A_)~nN@qLnDxzTff@ehG){kETC$L?t;@@-=cpUrTMBJ_wqYP0$mT4Bi*?z#HT
zha1<&Ig(0K<J6v~de2$kJT;0-_Ez!#ZzT9PghYDdVhtX0?VzIER+@lQsB?SkMh0us
z`d>YMC>|Kyo#rY3O}CFLwY4c+8Ob4W`z^PAb7t_#)%6Bz0g_w#8HPWJZBrz~?r4iR
zoVXOK$+84yRED#PoxIC45ux?%raMi@3F>T!<nm!BX-#Ff&RvqhvewuVP~vlpKgW@>
zndk4Be{lFuQkM<CA-qvo0KWO8k6w+bJRkU*nK=|zmnD%{(`qB>`K^EoW>_2`%pEw=
zxuRRdei36&9U0JhG=>yNnh`F*B0qu|#_f_@NEhxigh8;v=0haad*JeuhCit;WA!5&
z+y`Vw!*(m^-yuQ2Cb_JJ14|Jg=wK<^URJiQa3Y*b;tv)qBo>H9Ro}@BH!KvQ=a7-+
z>#c_`6zj^vd*?lQr-Z3E#u8hzXW=d*vc6w)N<tlLW8t0|Hi^qM^On6gI(yzyy4Hje
z6I4s7bau0wE{a}CTh&cFyYNX5FsUk0lq!_yyTpahDIcTGv)(qx!V{sUSnzYq)W@nI
zxqUW2+j1fm%m_+_It;<zlovPo|Muq}dr10<EEb&Ad^SetP`}Fu_TmlW8-}~~|CD_%
z|HinDi{36eQO1aUO2vM&D#@xZD_Upyfq627Z&Kw;;aXzbk^q>!qD9`j&pF*h<U)n5
zon-sZTorzthc35qOI7OhdW>}df9A55C!8)LDadquqH}vQZYK)0SBHP}LK9C@LI!y9
zNH>rwv-EX&u;TOurSLuE4oV(ow?%hTvx6#aytj^I{m9L530Wp0s91#^kH=xoJ<}fm
zy$>axe5UR8T`;1qc;@{`O3m`C<z+gFZS#iqB8<IU9@dqr!WR0Ky%9<di}S#q!mgU~
zs==q+r&zV=4({SCwWs!pK>dlNIZ#jkv;j$awq701A7|ZBZxskm=2UD1P8w+W$H)HT
zE1YZiU$=0tCm)r;P<{*2RMOr-N~9v_Ug`>IoAZKX#3+xAY(;-dy5R3}1!|`iP7Wry
zD`4Ns?#BwlYJLfJcS&{cIo$MZkvfL+*Ba46W<N)1+_rX6DM3oq)PGWBV9)PW&#PD7
zOW$<U@5R<M{7U`<n7)sua)#56*No{XO}laX2K`NpPPrx_^hrft-9@&)zaL9iq~ewO
z`;3b~8zPejd>&tBvB$|#gY3!u0uHT6NJ%@&8Fbpx)y4+zb@kaM*}n=X1cUfm6C~%a
z?x$VPV_e&)@hN7G3%gmlNgY3@1vtEIX68vkQM?!pKcjpM1!%6+f5TEpvK2%RBY7P#
z#YJsN==TKHz#3*&)n{@;mvOG2-Z^X+-9J+k1f2c`v0b7&-d>|Ye<gQe<a)4R6^M3N
z)yvB@&wS}y^7LW$w^l)iKS{3>S=joQkkDJJ%$J&nIZ>6J-A82(q)^g&S-k&4=5d9f
z@|=d_8*neWK6nE`FIUWliAo@F*)&uZGy}II#H>|)15}Q75}rqUn-Q!N+H3xzExqpi
zrNkK?K4c%LQ$}o$O6@NS=T6S6zJ8%4%b@Hf$<yS=RINF6aWnXd_p4Xp1tA<{*a{to
z$6)&}!hBCbb?*|`t+g{hHQ;|cF*ypti&ObanL|H1?%DyZ8)dfJE6jH}9S$bm?R$H_
z8Q<OBMVFg$PUl18Weqp#GLO)VC;@mSKpxVut!%i_T1Kh~zjb~ic^vcp31tss7FA^o
z80&oZpxFLqJfyUPE$K*<RZ9u@c6_(*c(3CHhG-%G^IzrKzg6`gE&CG&24Lgkx$&}J
zm7pg1PrKhoM7W6f9#`7D7wa%_3t~*B{=#clxyni%kyP&e8BZyMwKPx(Xk4TylGSCL
z7QnsDsG^**RX@A&nbw{l)JLx;c4GfiRll?Io$h(0(ui+Lho`y9?BAJBv=9?=Ay9&C
z{iAxH9g)7JZTYN3t8^@OtE~EZ@~KZ-E55sMGYv}eoE^TB7#W8FzZ|VO(vaPf`)_dA
zk`tG1|I^XP0BxJ&p^yDDNfedx3DP2+@xg=!&8?W~)GzP{0Cd0+EH9dxRNYAap9`6l
zzH;A@u#K?KZ|SG*Z%I~cVIq;yjs$_a46R9~XH^2hq+-|L#=lRhT$BE`^~5FT@<PSs
zAq9b?z7WHWw+htITh+vfB~ykiBPmDP_e)45-^u;e79n!MD~r3xNRAVC(r>?8ri}|r
zP#H7$7`fB=p=AizrWNH#n}T!X8(x(YaX(XZSO%EJulDi+8HI(Lw>~Z<1N_EZQV9p8
z<d|!7awi&lM8fSh!0j}ah8u#ED;ESSxRZ81lN+I#DbZ*{$U>Ba1G$W8^2py9gpZrP
zZcZX;eBs5|QyJ!RS#+bwpE^A#MtZHDEaCaBJDgcppLVwH!5B_;N)Gou?syXEp~c2%
z!fCK4k#}*|-5kwW&k={AIj@cn3&T~3XCCk`MZGB7zL4nSdnr3fF`-ed2ziLg3dHTP
zofs`$o=k*Otc{UtE+ZqaeHo9u40MUPywHaNK*AXEqyr3w_M8h#I0R`aDw&07p3dxP
zU7^=|a&z1A!{?C;g-$);<8bnmFP9h00T_}E5f%;-cjv*umS)Y)`8!*m%aCJ)l%Ad4
z#wGvs=m(cRb0uPaUTh#K&u_UROUPLFHPzXSHe|>rjdtjAuI`Y(kHssbAAs)-A(S?_
zDl{b75+3iysWrjph9i|G!J-wDk>pPqdXlM+@fZyCc^4SNQ4YLPEsAq)V`3x=zd)MA
zh2R-`3zMc)A{-8*Of$ss*KrhW2rvC70M8G*-kQ3x29!JjnHE)|nIiy70g5vU!ou_6
z=ZD1I;7%|FA=9!<E9<)N*LODrdeb`-(oAL(h(x;EQ-KA5Dn~hL>~w9g-U`#>5j3>v
zxjO@J1g!?B9!v(4?HvQUt~QBWk5otOp?YeJh!G~YCy<lp#7G+Y)o<^|M1qveF6p9x
z&U8K4<6%A71+C++v};L6!{Mr~?Lrn5!78M7-TS5{FJ9GR5?**?L_IErj@$}c&GiZ7
z4$tdwd+ehp`>VLO#J?|&d!cw%tt2J-OJ=QzTYg-pG&0+Fmfmz8f4@fbqBg-Nf7_zi
z6SKfm4^7p{_)#@4dp(z5%!ZDu?fc47fTxxnmmbs#+VibHC39?|1GW$Y9Atm#u(gau
zM&S#@U<>nBoj>%3Aj?uafsucOFjn)NSp5|u?WgNPkP7hJT*@W)Bm`38tS*VGKLG-9
zD{Kdm9%M=@!h3$0Izi1v>?MqU@ihZ;d11?~Akju(<cz8jDZ!NkWETZbJf<m89P#Mn
z%scs4MOKM|XCltE?flrMAzH|%EXoLjb~*|ypp$nV;l{Wadymv{oYAfsGuJE_le8GC
z)7v`bYJR!6A{Z!jT6p~^?JHr)pK384lDQNP9TW-DxtQ6fIIubU+<-`N2$k7pSt>W%
z>Hf)b=6ZQVM-t5~6P-PaAtIZ|Va56&QZwTAVgrhPaSk@k5|2>E$8{2lbp;O?RB2CM
z3+=zdRqfvoZ`%cs=S#5w)@A}{uAN!a$t*uiqBBufX9V}GBE^l1@Z`hWv-x$njHe78
z@o4=B+Qsv0%-ledn;WWRKc>#yK_#NDA1nME?ej>{x=O!z-9JPqt5*%i8&^7jlzxA7
zdoi1k5qnfYl_$;;If+<5Vm_)0xex$%d{t~?P6>_Tgl5vt$11W$$e-bUnO_AEmYl+9
zIu@@e&NRyiS?D^&peliq-!yxE?X{JlDsF3Jxic(YMWD}jLoP84P@QhU!=}uohiyZ&
zCNTt!MZc!BjD3K_KFcox1;Lt(Y4Xe53WV&nQkf4kO}UYFJ{pa`nhxfDyI=Xe;t&oH
zfdPVXUD7jej5bP_Ney)k5S_^HC0tt?glZ8lpEaHe-wEh7*nO~9MTPry^JO?Ec7#|V
zyO|QZQ$!dqPD9^Jo4bkAR2~WATns*@Aj*|4J27d#MO_hf7|5@YT0?95);^xjg%<K%
z%HI#qdZ9#K`(AJreS_4$lItV2kPNdO9$%9G6eTi5;iVVhxQZQ(&91t`bY0|VyTm&<
z1abZvC$dQy&ZRt$!C0?H{3c_fY8>k_LM_kqsp55X4b8MOEi@e;UzSJVxoN@KXWY%8
z*2pMG2zMS@lLvcKsag67)#6i}#?9$Nfp`G*Vn4zYBSh%?z6N~|Ahf7~VY=$P#Lx{c
zIv4}hd%V@s#<46~v@g%V_V2suy1_|>pP>pj7LMB&A{<}sV4l6;$K5X*x!?A9#dIay
z@0@&*n-5wyB<L2j*Q+*sxnOo)2%W(dwlITd&t96HheL_rG&?AMdyK9}ppVe}93mKo
zI}1Hl**4D{ugqM7N({vfa6KYDf<2C9UVjj9&r`%%h~Z*F`bc>MG{8_7seb)ZrFTYH
zrj%2e6|6FWKdmVo5S*truq-RZAl6xbj(W4|ZPDZ<GOnUEryUCR;=cG+3G|<e4vnn<
z;1De(Xwv}kva<uSA{F@VNbdkkXoLd#Vds%r3HXBBhT2+L=Lr1MHbf2oOuO><CE{(X
z50<u#om}RqNh?9=fThtUB`X$AT5Aw@S97+2(SxCQ?zZI}rX?w%)P}V02?*3{$Yp^3
zU%##>!9Qs6;@*zK28Dl9M2%V6jfo)h7z1d+@4pI@V-q=dwbwavG(COU@f?+N*py<l
z0Wnj4y-Y=m#ddID$;CX+7E*EdEi)c{;c*_XFw;Ul&m6~hW!=59Pd6a^djY2z<i>gy
z#}<i8h?S2exoq)#aw?p4m>M}2zT<X~8mhGBdDnr~vle5t|2kLqNTm&iF+#(_`4>md
z2u3?&s-$4`(yq1OL`v8>6Dm9XL%OptmSv*be+|hzr1KMNetF@J%rE8u*)3y{bXd>e
ztIZ1JFGi1s(_uwE!bGSJaZMm}{(<kHBQv*-lr0SC+>=!}pwYryk*oP;X9Q-s()cht
zGCy#A9tUI=c5|}nrv~=XF#a7Q3Sa>GY^YXpmqCz*v&I(`h8GkF<oYrit3I5}wCohv
z5&zw&o9b1*e@BpI`-oJ1z|Gzn)f1Vvx@p3I#zFmT4q-0L_r|5bS@pUY(l3FI-gTGA
z#hTXukPKy9PE<s3#b4gQT_^^m?uSKn#qYunxoSNpli9^HiDBSkzppMKDMb%$-#L(v
zXZZ-vJL)WVKdn_<Nj#^-Q*%BJgQIRoo2HfN=-%(NeHSTyH`f?-@0C7)ajt-y`dGbm
z;n8gaQC3rn2;oYp{XNUgI^@2)++z3uZ@Djfr$iLu1dHixr2^bGC$&wlZSu6}He_$j
zY-qyIsq-cE1|4;aq)COwwB{=%LMYEE>qg@;!(>9N0l!#*Q6+y;(;js8l+D?clR1Tl
z+guy0P)Xc$D_#hH<#b5#2x=Rl;YZ^aOmQY8R6ES&^>re3swfS-DrKX}k99sivP8zh
zV`DA7Y}BLE&K6z>v<V5xmDE?0jv26uDeFKS*S%+EF&P>|vyok`qIN^Fj&RU)#=H-|
zgKrlL$zcxAa%9AjdT~#CU{rN!%U?!c*bUF9Pm#LEu?C0#cU|db&YG`erd@ZDJ)~sr
tDX<JV5q_OH#i^S#3rV(f;N(E7$@ulK__9wEY59`ufu4zOmA1=^{{gxFrcnR@

literal 0
HcmV?d00001

diff --git a/docs/assets/images/logo-dark-mode.svg b/docs/assets/images/logo-dark-mode.svg
new file mode 100644
index 000000000..004bab98d
--- /dev/null
+++ b/docs/assets/images/logo-dark-mode.svg
@@ -0,0 +1,207 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="348"
+   height="95"
+   viewBox="0 0 348 95"
+   fill="none"
+   version="1.1"
+   id="svg39"
+   sodipodi:docname="logo-dark-mode.svg"
+   inkscape:export-filename="../projects/outlines/docs/assets/images/logo-light.png"
+   inkscape:export-xdpi="96"
+   inkscape:export-ydpi="96"
+   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg">
+  <sodipodi:namedview
+     id="namedview39"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     inkscape:zoom="0.27368421"
+     inkscape:cx="283.17308"
+     inkscape:cy="45.673077"
+     inkscape:window-width="1416"
+     inkscape:window-height="818"
+     inkscape:window-x="30"
+     inkscape:window-y="38"
+     inkscape:window-maximized="0"
+     inkscape:current-layer="svg39" />
+  <g
+     clip-path="url(#clip0_56_2)"
+     id="g39">
+    <path
+       d="M101.92 79.3501C101.92 83.0601 99.78 85.3201 96.88 85.3201C95.13 85.3201 94.03 84.4601 93.43 83.6801H93.39V84.9801H91.07V69.1801H93.39V75.0601H93.43C94.03 74.2001 95.13 73.3801 96.88 73.3801C99.77 73.3801 101.92 75.6301 101.92 79.3501ZM99.6 79.3501C99.6 77.1401 98.52 75.2801 96.48 75.2801C94.58 75.2801 93.41 76.8301 93.41 79.3501C93.41 81.8701 94.58 83.4201 96.48 83.4201C98.51 83.4201 99.6 81.5601 99.6 79.3501Z"
+       fill="white"
+       id="path1" />
+    <path
+       d="M108.52 85.69C107.88 87.44 107.15 88.83 104.98 88.83C104.32 88.83 103.76 88.72 103.32 88.61V86.84H103.52C104.49 86.97 105.22 86.91 105.73 86.29C106.15 85.78 106.46 84.9 106.04 83.84L101.95 73.7H104.45L107.61 82.34H107.65L110.55 73.7H113.02L108.51 85.7L108.52 85.69Z"
+       fill="white"
+       id="path2" />
+    <path
+       d="M121.12 84.98V83.84H119.98V82.7H118.84V75.82H119.98V74.68H121.12V73.54H125.7V74.68H126.85V68.95H129.14V84.99H126.85V83.85H125.7V84.99H121.12V84.98ZM125.7 82.7V81.56H126.85V76.97H125.7V75.83H122.26V76.97H121.12V81.56H122.26V82.7H125.7Z"
+       fill="white"
+       id="path3" />
+    <path
+       d="M133.7 84.98V83.84H132.56V82.7H131.42V75.82H132.56V74.68H133.7V73.54H139.43V74.68H140.57V75.82H141.71V82.7H140.57V83.84H139.43V84.98H133.7ZM138.29 82.7V81.56H139.44V76.97H138.29V75.83H134.85V76.97H133.71V81.56H134.85V82.7H138.29Z"
+       fill="white"
+       id="path4" />
+    <path
+       d="M146.36 84.98V83.84H145.22V75.25H143.5V73.53H145.22V70.09H147.51V73.53H149.8V75.25H147.51V82.7H149.8V84.99H146.36V84.98Z"
+       fill="#FFFDFD"
+       id="path5" />
+    <path
+       d="M154.34 84.98V83.84H153.2V75.25H151.48V73.53H153.2V70.09H155.49V73.53H157.78V75.25H155.49V82.7H157.78V84.99H154.34V84.98Z"
+       fill="white"
+       id="path6" />
+    <path
+       d="M159.47 84.98V83.26H160.61V82.12H161.75V80.97H162.89V79.83H164.04V78.69H162.89V77.55H161.75V76.4H160.61V75.26H159.47V73.54H161.76V74.68H162.9V75.82H164.05V76.96H166.34V75.82H167.49V74.68H168.63V73.54H170.92V75.26H169.78V76.4H168.64V77.55H167.5V78.69H166.35V79.83H167.5V80.97H168.64V82.12H169.78V83.26H170.92V84.98H168.63V83.84H167.49V82.7H166.34V81.56H164.05V82.7H162.9V83.84H161.76V84.98H159.47Z"
+       fill="white"
+       id="path7" />
+    <path
+       d="M175.45 84.98V83.84H174.31V75.25H172.59V73.53H174.31V70.09H176.6V73.53H178.89V75.25H176.6V82.7H178.89V84.99H175.45V84.98Z"
+       fill="white"
+       id="path8" />
+    <path
+       d="M180.83 84.9799V82.6899H183.12V84.9799H180.83Z"
+       fill="#FFFEFE"
+       id="path9" />
+    <path
+       d="M187.59 84.9801V83.8401H186.45V82.7001H185.31V80.4101H186.45V79.2701H188.74V78.1301H193.33V76.9801H192.18V75.8401H188.74V76.9801H185.31V75.8401H186.45V74.7001H187.59V73.5601H193.32V74.7001H194.46V75.8401H195.6V85.0001H193.31V83.8601H192.16V85.0001H187.58L187.59 84.9801ZM192.17 82.7001V81.5601H193.32V79.2701H192.17V80.4101H188.16V82.7001H192.17Z"
+       fill="white"
+       id="path10" />
+    <path
+       d="M197.96 71.81V69.52H200.25V71.81H197.96ZM197.96 84.98V73.53H200.25V84.98H197.96Z"
+       fill="white"
+       id="path11" />
+    <path
+       d="M15.18 39.46V42.5H54.64V39.46H15.18Z"
+       fill="#E35A26"
+       id="path12" />
+    <path
+       d="M39.46 30.3501V27.3101H15.18V30.3501H39.46Z"
+       fill="#E35A26"
+       id="path13" />
+    <path
+       d="M48.57 27.3199H45.53V30.3599H48.57V27.3199Z"
+       fill="#E35A26"
+       id="path14" />
+    <path
+       d="M51.6 24.28H48.56V27.32H51.6V24.28Z"
+       fill="#E35A26"
+       id="path15" />
+    <path
+       d="M54.64 27.3199H51.6V30.3599H54.64V27.3199Z"
+       fill="#E35A26"
+       id="path16" />
+    <path
+       d="M48.57 21.25H45.53V24.29H48.57V21.25Z"
+       fill="#E35A26"
+       id="path17" />
+    <path
+       d="M54.64 21.25H51.6V24.29H54.64V21.25Z"
+       fill="#E35A26"
+       id="path18" />
+    <path
+       d="M15.18 51.6V54.64H54.64V51.6H15.18Z"
+       fill="#E35A26"
+       id="path19" />
+    <path
+       d="M39.46 24.28V21.24H15.18V24.28H39.46Z"
+       fill="#E35A26"
+       id="path20" />
+    <path
+       d="M15.18 63.74V66.78H54.64V63.74H15.18Z"
+       fill="#E35A26"
+       id="path21" />
+    <path
+       d="M15.18 75.88V78.92H54.64V75.88H15.18Z"
+       fill="#E35A26"
+       id="path22" />
+    <path
+       d="M15.18 6.07H21.25V3.03H18.21V0H12.14V3.04H15.18V6.08V6.07Z"
+       fill="#E35A26"
+       id="path23" />
+    <path
+       d="M24.28 6.07H30.35V3.03H27.31V0H21.24V3.04H24.28V6.08V6.07Z"
+       fill="#E35A26"
+       id="path24" />
+    <path
+       d="M33.39 6.07H36.43V9.11H45.54V6.07H48.58V3.03H45.54V0H39.47V3.04H42.51V6.08H39.47V3.04H36.43V0H30.36V3.04H33.4V6.08L33.39 6.07Z"
+       fill="#E35A26"
+       id="path25" />
+    <path
+       d="M51.6 6.07H48.56V9.11H54.63V6.07H57.67V3.03H54.63V0H48.56V3.04H51.6V6.08V6.07Z"
+       fill="#E35A26"
+       id="path26" />
+    <path
+       d="M60.71 0H57.67V3.04H60.71V0Z"
+       fill="#E35A26"
+       id="path27" />
+    <path
+       d="M15.18 6.06995H12.14V9.10995H15.18V6.06995Z"
+       fill="#E35A26"
+       id="path28" />
+    <path
+       d="M24.29 6.06995H21.25V9.10995H24.29V6.06995Z"
+       fill="#E35A26"
+       id="path29" />
+    <path
+       d="M33.39 6.06995H30.35V9.10995H33.39V6.06995Z"
+       fill="#E35A26"
+       id="path30" />
+    <path
+       d="M66.77 9.11003V6.07003H63.73V3.03003H60.69V6.07003H57.65V9.11003H63.72V84.99H60.68V88.03H9.11V84.99H6.07V9.11003H9.11V6.07003H12.15V3.03003H6.08V6.07003H3.04V9.11003H0V84.99H3.04V91.06H9.11V94.1H60.71V91.06H66.78V84.99H69.82V9.11003H66.78H66.77Z"
+       fill="#E35A26"
+       id="path31" />
+    <path
+       d="M102.09 60.7V57.02H98.41V53.34H94.73V49.66H91.05V20.16H94.73V16.48H98.41V12.8H102.09V9.09998H124.21V12.8H127.89V16.48H131.59V20.16H135.27V49.66H131.59V53.34H127.89V57.02H124.21V60.7H102.09ZM120.53 53.34V49.66H124.21V45.96H127.89V23.87H124.21V20.17H120.53V16.49H105.79V20.17H102.09V23.87H98.41V45.96H102.09V49.66H105.79V53.34H120.53Z"
+       fill="#E35A26"
+       id="path32" />
+    <path
+       d="M149.89 60.7V57.02H146.21V53.34H142.53V23.87H149.89V49.67H153.57V53.35H162.79V49.67H166.47V23.87H173.85V60.7H166.47V57.02H162.79V60.7H149.89Z"
+       fill="#E35A26"
+       id="path33" />
+    <path
+       d="M189.18 60.7001V57.0201H185.5V29.38H179.98V23.86H185.5V12.8H192.88V23.86H200.24V29.38H192.88V53.34H200.24V60.7001H189.18Z"
+       fill="#E35A26"
+       id="path34" />
+    <path
+       d="M206.38 60.7001V12.8101H213.74V60.7001H206.38Z"
+       fill="#E35A26"
+       id="path35" />
+    <path
+       d="M221.2 20.1701V12.8101H228.56V20.1701H221.2ZM221.2 60.7001V23.8701H228.56V60.7001H221.2Z"
+       fill="#E35A26"
+       id="path36" />
+    <path
+       d="M236 60.7V23.87H243.36V27.55H247.04V23.87H259.94V27.55H263.64V31.23H267.32V60.7H259.94V34.9H256.26V31.22H247.04V34.9H243.36V60.7H236Z"
+       fill="#E35A26"
+       id="path37" />
+    <path
+       d="M281.73 60.7V57.02H278.05V53.34H274.37V31.22H278.05V27.54H281.73V23.86H300.17V27.54H303.85V31.22H307.53V44.12H281.73V49.66H285.41V53.34H296.47V49.66H307.53V53.34H303.85V57.02H300.17V60.7H281.73ZM300.18 38.61V34.91H296.48V31.23H285.42V34.91H281.74V38.61H300.18Z"
+       fill="#E35A26"
+       id="path38" />
+    <path
+       d="M321.81 60.7V57.02H318.13V53.34H314.45V49.66H325.48V53.34H338.38V45.96H325.48V42.28H318.12V38.6H314.44V31.22H318.12V27.54H321.8V23.86H340.24V27.54H343.92V31.22H347.6V34.9H336.54V31.22H323.64V38.6H336.54V42.28H343.92V45.96H347.6V53.34H343.92V57.02H340.24V60.7H321.8H321.81Z"
+       fill="#E35A26"
+       id="path39" />
+  </g>
+  <defs
+     id="defs39">
+    <clipPath
+       id="clip0_56_2">
+      <rect
+         width="347.6"
+         height="94.09"
+         fill="white"
+         id="rect39" />
+    </clipPath>
+  </defs>
+</svg>
diff --git a/docs/assets/images/logo-light-mode.png b/docs/assets/images/logo-light-mode.png
new file mode 100644
index 0000000000000000000000000000000000000000..ac04c20f52f01fcbdca2913c22df1e0e9f63d13a
GIT binary patch
literal 9174
zcmb_?2UL?=({2bL2q+ec0zyEl^e!Di=|w^59YPU8?>%~yqcn*adQqAP1PQ%IJ(N%d
zga|>vgMf7DC6xO{Dc}FEyVkeXT^AO~o5}3iduGo(GqX3(?&_(Zqhg{0fk5XpHEtV#
zK;RnSxsmb=@M?lMUkSXC_!y|GfKWXwv%rVHpeCA*y1Jknz%wO?oa7RSl*j`7f=HM^
z<iDRmAT1K+lV<}G-Xk6`2o&W4B0J(S1Kx>0FMvN_^wB$MHp!{PZ16ERxF(zQ_!-nl
z95yV<pbxyA_0%x+0f88}h<_xS2AsbDX{IjsO?*vsb!6>5U_!PI9(GWnK$s_S6i7Z$
z7I=g~eQhCuFgJG}*+2!3-yE{QGm$LJ0r}11>#D$EqI(yj>fsHANC}AuiEt=VK_C!$
zZwE(NgWGCH!+}o<9L~PJp0dKi0RaI*0pda)-cG`zGBPs4B4WZ~VuApNpihvyuWg{9
zyU+DMLXPF!hWglhyLkG#c(_A|a&7HA{CpKSIEWLS{QOy`uZ!d9Ozu8MY5@ug6TcA_
z6%r9Xkqr!$CsJkgy<MQd%tZN$qVm5v{|D`89(iHn<o~glKbrog0;(!f$qS#DO_A!V
z)prUIi2bSN?OXQ)Nfy5;_>8KhV!UkR;o{;A7>igizN2eFMJ3CI-o4Bs4=W@uTOX6N
zAm;Nan!$VuDwC*0^7_}Cv~CA}5jU9U2Xf!(NV26M4Ggtpy=Y)W%1*!o*E9QLN-w_T
z^yv(21?w$02Qvg}dRw}tw*0%ZQtffQyq8ekYgk^qJ|fe@DCQq*HnR1p!??{@$z4%k
zRo*lnX_-8=<a*e>GDr(IwT@uveYvq6-^y3+Z<1{984ChaLLz~eE=wEHt#j|g$u1~^
zNdCM$2b0y@IhT2z`1Kzit4L75X7!76zd!wcp(CgCjAoKoKOS=-xe-LNwBW(cd?Mxo
zFO~c*XLwovW&@VzKw*chSq7Pv&*PFBZjVf|Bs3lR%@OeGu!oub5c|EPn2m2aUP}n=
z-mRw4`%5=88k<Ad1be#Y)`d(7O<Vl}0ak>uh>(MSmAZDh)uiSLUY2U<C>ylTu_EG(
zCg;n-=M${06<)rqo6FeE`%vOnT;A&nQw6sjDjeWx6NYE^hZ&{OyyaiR775aE8)Qp*
z$kZ=~i`yPZ`#z>C!$UdZ9ZEU^mks|qR(v4~<Z<t7b4W+rv31~&86X&ipZMPCi${vW
z_b)>*-~KC)XXYD`c9*lr0eY=x*LX;7>Kl8WLvkPTWpNRV%j?Pq=)2%+uDC*KmtCHn
z34HTZ_YDQJ1*VL;iPV?@dJD0{qus##LoTau=fmdNgDox|lCL94LA82d&kWQAleV7r
zcDKKoic<l94foRA7xJs)j>e9QW|yY^Y$7BQ({%P1!;`T3oCS)boxX6>lLS0UdI+{5
z{0Gy8i6#6ssL}}qPJLm5b(;}{KiTa~U7MZGVJFvd^I|p+0ZRp;df^?@1Db1o@y|Ux
z8Ut&8St-18HnO><KJ!-GVvg(e@=jllVsy)-n9C2hL+8j<@Kv6ws?R|$|1o<|@c#A2
z6Ti6nAFJPu86!;QL$Esq#>-V~NY<V~r5;5&&Bm6gbpOzwTW)$=&nr2IKIwaX0;hDo
zE5avqL-lj_dW=|A<mh3=5Yzd+Z4sG5ju#Oz7C15H`W5(idD%`18&OxAK$5XHA`LR)
zzXkmE;B{W*Nc?pn)l)a85mt_r`+wkhzu@KeRLuqnrd6NW;m7|wQ138Bs7jqhVpDP6
zW@<K;uUYN|>@DKk&D6galc6swK2nQj{WYJxKyby4c)Yx`y<T6vg7UB0UHAuL{=mg_
zE^mbmHTEPg_U&yD9DDY{gx(VJeRdf+YtIuseb{`mxo1(x3(4k+#i^!+?|W12rur?R
zWC+d|Emjg{zap(EpySr{LroUJ2^G%*Dt*34qZ2lo5>`~&lIgBTVs4WMMt{8RLDQ*n
z61vXZfF-(Ot&TSw1inlU!DRf5wKY8@b(2wgrq%XQAA63n%^LtFE;ot_oRYFrKqT(s
z(!uN}$f0vT(*7fNu)T><GuWHkaW3<9ZpXLTrq-?fQp54_CFdcR%1j^gU^<f>H%7bf
z@`|(Db8RUZyXH-%4}ZtkdDTeJmdKKLJ&T6&WBls~uSaW1H)$%I@^AfKz0E^#!yVg)
z0e@|rB2&dB2nN=o?sNiZ@Is;q7vtWuo|24!wU=!Fj5!W;h=?Q*2!-D+4n000eb$OJ
zq%&LT3wj4`P7kZvY-|~6Y?@YwW0OlQCdbBt%ZgCdt1TR>h+Mb88BrT(M#U0<F9}7l
z%Um;AR*lT-8WK1&UiIupk!ysiVvjj4FH;3SN{EOqFq6v|qtT!NW^3=&OkROtRGEKL
zxvlij^8~7Ldn^ZT9Xc~wx4Tz93QHDJl-+fb%VZaU!3)F2^~|FO*+nAYec;iogLSt`
z3(k3&%lTGJhu`NcP5fwyDWaVp1|J@i)JwUjRIGakb%Ih9#6eU2=0VSJ+s%3fo}^RF
znDKVmj~%W-l<sVc!;(Gcb9ASIT>d`RpXCLJO-wF3t=s5ChUx_X`~Ir$Vm+ad0>)PO
zUp!Y96>6d)I{~JNGR%Gy&xXSGIRxvcO~A{&9-}h+u6pnLO=~S9UM)-G9aDpg2(%xh
zeumfY!3YvD9!$R1Dzpdocd*}d2PS{bjVfKvS#WLau9=pH;-Id)^U_9^10ka`!R3bT
zyXkMr%L;ztre`{xNl*u$HP{q$2Zy__hG*vvUCj<QQ`v-z`4k@VsLQ~*K1%2w-ifR4
zr$r|`r4OvsFSy@sm2TVnNv3)!2eWI_tfY|lokAxmnjrqSbXalGT(T;vxyM5HvH7>^
z^)No3hKAyYcipNR_P1*-)LYP+iu#Co?)i>`OsdrN5}(7amCmLie2i<4p}%F-Q*?OF
zpkzavLMd)%t$HSCo>u8rvy_sx8ntJ6#6^mYWlU&@gZlF))929>n=7+at7t7#PrJBJ
zl1IrF5*azO!fMe(b)0fY$Tqcr{9EeIllI5!iLMX9Ou3!}u(C}l5S?3d{^+ZxF5ai;
z;2+V2<uL8tVfa}{IZV-Z12%a;85d=ZbFDhOuh_P?T|uU_|JA?lk^fkM{!ZEzixVh>
zkpZhs3O?`nDd*9MTTN%vaDuh%lqlOHym|qbvu6kZyf$$@Y!?c8a$r)9A%DDVQ*A=U
zYU2G~nCQ;~Eu6?YpAzbF)EW9hnUrD~!CA4N35mneZaA*_6suP(mfTNKey?^{7|0o!
zPx2E8Tuukz&J@hP?-X~0NWd0JH2;H#=wRhv?1)U5Nbqo!ru2h+x1EE3`nS*zQnI9o
z<V3PS@&K)X$Cdwq3k+8l=|?9$#?{cl%^T+(C6qIa?S0LooF#JbSc4;v20$%`yyo(T
z%@OkHSWXxRg*oH9IR*yuCmlSZ%lw;l8NLrJkF^eUAU<06D4G*8o6*uMq`mxdew}0d
z{QIU||MoM)-WYi08`A%_Xof=QQ{BwM%vw@<^>SVSwxe_d{bnlkAoNWSD`90|>f(gI
z|H~_szfl#=PPSLFxUkMGt{pje(=KsG$ik<?bSvRY3o6*SHeI+G3EBK&HSSnw#p}pS
zsOi?mH*W2RXALzDDLN^QKU!-DJG<;kUcF15$C}lfr9UueZyjEzIQAGnXlciC>KK}o
z037}+^`7#SgaX{B9^vr+h3O`DF~4T(A5YDegWn1dL+^n(X1?EB?rr*eRNddNy01sg
zOd!*#8B$oz9&qmdCUrVm+@&-i)~kI38V<7zB(E-k>>oo62^lvRkd_PaAM8#8Vub3#
zL=}Cm(n-i*asdr^{Nzk%PDsk;0M1rDT0bdTE+i)cC2FoaMf7PHIrC#gZDHiI@kgjP
zNj=Jy<Xth`y-kfZ@=Rh0>QVg~Y&K<gF-X(yn!Ik~tK|^kl2^!&QTG1|&}T4k;U|j%
zW-`vFoRcz(Bn$-mRR9o?8?SrqW3p=2MgV30!o}a!M=Q8#D=!@{mvN5f+Gs~l`Bw;<
z-C<_c=)7l?3~6n%^raYvi7|?5&5tDQRiz&vLa&F;-#5Qfa;z|1lzf*uqivMl=OjWY
zS&9K=K8yuR#T^6oLIFStA2arUa!mQZ@nD?U#PIrhr%XEf82u%81CN#b;g0@S>mmKc
z9;T(f8wU$(x~9}l87$9?{NH?WCRtistu9H={WV(adE<CpBxJG|x7X*s8J(d@r9qOv
zX4KtX`k8afr`(IJhY)Kq8P71Gc+OrSX1G=a_lWOwBb2jsBDW@xvv;G8R|*l4gZEi|
z$jG3hR{H~Ma4c1O>6`tcy&EHeXzJp!_g}O(?bsx%MuQB&qdl5eFB}yjPZ@qGnf(Q2
zlVu84=#oYO_4*ZD*6^wK$TCdOkH+1)>V5!T9k(djvR0tBkW0bylR-mPmJru_0~Qm#
z{{%ul-~0LHhs4}0Z6x_7vLR|dFNsrF8!1xyLPQgl5t4$_OUGw}iSuSF!6%E@q{Zr{
ztp*6?!W_NY`b9cuVD#9VyLS?Z2y~=a`9%QGx=ZX{pOPK`o?67)cjjpQKy4}o6eH!e
z4+Q8=NZHSULW-Vi<U|f-ml}`)q^h4kIW1uCUz|vOe?zMxB@1JB5qr-XI^V=wBv2G`
zv-O{`k6VZ}{n9bmV&#6~f%oR{$QFa^*3ZKFhXLLE#xL>sd*NKuW_zTBRK()SV7*U>
zWn(`&L_qy(@m#E$E7^rBw<m;Z6PLd+<1JQ;JU%H|G*ZL-U>*L4dobDH@+>DqbNzvD
z0e$hHYU(?!$?3|QomV(;AIGx7#%^K+<#L2ALw$dwmX~^uuNgyKABubUZ`!7@*DVol
zZ{8S2*6q<2KAaI5F}igXOc}#D5n|oZOELdr)kh3%R3WgiJmc|a2JJDUZ)s<%QV~6;
z#-m*40czDO!Rx1i5CahzF{eB5+8oxL6BJVax@M&l1mvgpjLf!|j|VEZ5n+96`GF2%
zAa3iE=P?}wrVzrI)1BX=j$AyENZ=M3;jbZ_!G=$BPc(a#-c(4a(;wYfN?jL9urF4m
z9TnVty8=}t@I>)<ZI;jMcs=pYkko9HK-zbA>@gqo>eXE$=sHm=yoK*&7k-1GDYL@2
zq!f9Dr0_Y)=nZ+CtH+l5x`K;T&|2xr>dx-&PSN=d!`WV(gu?2Qe(GB#)ZR7m&%HRa
zHWB2({+&qt55)neO)zxXHvA@O<*?|zrsYN!Rvwr*l$Rk*iuB83TE?BLb-S9`gTAV}
zjN?ySEjl{MmYT=(JI!k5klMnCx~~uPWx#%99X;rtgNsAILK=2(fjhect8=R(JrM`U
z{m`((!E)uEsAeH+MET~kFA^02al=(_zVvSN=2pGz|J^_x1?4?5wGeXS2<m?C?~P1Q
zoX+!;0DfOhH1Tjql~f3SXX`0Ks9l^8q{?|O&UXsT)_}i<Va0?_C>UOV^1pd<8Y!wv
z&xs0wVdSC0GR=p&AxQY1go5V4I<xkmLv-l)q3lN3e9H4(+g_|-Xj$Nr#ek*v@E(i$
z%-i@I9bN0ilU{L;Ln^vxsn+F37TQs?9UFFW*46P+w?2HJP7I43pOjDn7b)QpSx%~l
zdySR{RhDDn3kN&0x0=yeE%kepo&*Bg@ka_(Xz)f46V06e50+ap%XNdbQICs&5(wz$
zoPRxI)-$OxlF8Z5ZxOD1+wewATqault^DW6bLF&4tyO$8+kUQi3TTwz-Azf9*lMGU
zOg0*rPI%MhN@s5`!KbLiP`|Hq!Kuve@)H#RJ1`TcX2R?NM=UcDw8JDPc2d3-3963#
zV0fCicYxNs=a3amdP;f(KnKa=f>Re?W1OISvNNvgFOnU}s*Sb?EIpHs-nd^MXW*jn
zA(AxL>e6l|+lB(PaG03p`rXcs1pM`71_vvA1ucaw*?y~ynPm<e?+CWHPoGN?-!tQL
zcmF5kfrVPeOze_Z#@Wq=#UFbkp}hoefq&H8zuwtMn{JkA+GT$`mWecEctNmU?&6mj
z@X%_@)EP!r9<_#;o5aSACBX5{NfG-~6Ht!F<moBpiaLOK|CPt4!jLX}cc6UGJE6F3
zcyrH5p=sAk{Lu3KiA}OYj(Z=?i~;7l@x3nf(XP;affL96c_*@(cl~2W_NmeAfu1pr
ztGD5l(g>7IoBx$3W|Li9skR<CwMet_Z}|^_6#ODUi7-1@U-FiMg1@)*G(!awBc$=Y
z81KnGV@!Z}u*ZrEpHP&A!8wh0UMGDM0l4nd-%NO_t~S|TGkxiI%7B{!^-#}hRoPD}
zu&YZ>ZL&)3XPwms?OQh`A)%A0+JBtGkq>}6#~{3kFFE{GMIQ*2Tp0tLbaj(U^-D^6
zLvWk`n~ec5^UJzx0%w4OE)mj#xA|)L1IYr{(NT*6tCzo{S{A6$7cabgeH!Aej1bJV
zL)Q1;Qztx6zWu^9Dw&2@iY(MB3lPL3NT<3pms1wmBQZ39V><;7n1})Di6E4D*B){c
z-Qpz!V277Y3IgX(^B@XTIx$3_v)^<!x%n3adF~G<uqhZFa-hP`myUy0U-~>gk=OiF
zd0cK=2~kSD!xO}+J*F|_?`Wod%faDVk)E0MWP7_`+lwW(TxFD7n|=y(1f?t1#$P|S
zkgPEuz27}upw?J*jw)7awW0H1dizPCR@RwTS>N>7AM)g@xo#OFC=-uO+%7lY3%%(Y
zC@ziar8BcnOty+rDltJnrus#-ls&2=MFLkI`enAt!{r*p;m3&R<!XMxqS<38ps%+U
zy>PEo8sUMSx1^O_N>a<gwMr%(xz^{qk;b0PANycPFuZuz%-ht{d^EXgU~^u@mb3TJ
zs-`CkwZ<dIYHZ+frh0_yixN|A<@z&DYb_L_!jf|*S8KAk+l)0wpi_0K0*SnboU&!$
z+cGsaDXDtKX<V6<|2Vg@>yjp_9+kiEaN`ou;({bVp*6H`Be_qNT-q?_rlcLm#im(B
zo@ubKe`i&yGq?o)7)d{1CiISsPWX8SdS<+HGB&S4t51t*cynrQek7-;5n0NKOHm~_
z2P38k)SYZ)iS+U5Vl&$h=Bm}2-xwP2$2-K0az(_%1i_N{`6O_~hueLUt%$W;9F*|I
zza_XaZX^Dw3)g;^--D|Uar^JY&cFb7f@@L&b?`xj)ips6CB>A#KGHCH5n=XeXJ$V_
z@~wOl3nN1%VkQZ3Xu{i>^>mf<FgQ%GxAGegAQ+x7<go^Bn;bD<HS6%FFUk9$<qa9u
z33>rV&iiFxoCFt*5u{khT~jK%a*_I~?lSD0N>B`S?}n_t%K>M|{Z0%oztvOk*E0IL
z1PU>XpQ6U*TPQQO?8jYi(UK`UE}0i@`96Gq$Ef(QP0LmW*{r*G*J5SKkv*u=XV+cQ
zC^mwsU&#+UFz(p<V@ZitrX^wyVM0o?tNH%)YpK`=y8|Yz0>&%GrW+&ZRg(@YjUmVs
zGqxr77RJn*cSmj26)Vie2w1(Ml)ohRBMo=>E3s=u?q=FihUz?)sz2VAlN-n7ZxQ^u
zFKQLU=H$TQqKjdi@m`yu9kg{G-5#7#h-eo<Ml<QkL}vWZ)NU$e?n-6f_h%L}_OKXP
zApmaRrj!DXWm*i+lmo!KT0_BYo=FCWjd{9dSyiF~)a96(s=+8VW}zs?za6ypDRLIs
zV$}^;ub6grC2qxcQT^%hpE44g=<n4XF_fd!ZZ~Apwd<9bVER*LkJFg3=2#{HM(cz@
z<0dPOaKT~Z5c1<^gQ1FchU^skDHuoL=lw6by?Lq?Ifi%{e;Os{%CYhJTTX6@rb|{%
ze`zqy?5_vCEIm7`Mm(9!=fd#U4Aij|=Uc;NJt+Q$@-bmjXLIj*ng(Vn4T5k;BLfqf
zr<JjB&4WC-SW0J5>9GHD_nn7n4UML4zmHNde}fwy$#Eze(hA*`PL6SS;&_)yBL(Ri
zEVyH^$`hQB$G95Exx+2Yo5*}8ehlw2<fG7_oT%$Eri9n3&>b5yQp!W$n3qrKe`HDx
zZ0Kj80I!Z^4V&S9Us5{}s2+08(@Hzs)Vbip5;fNLwN4f)*0qsF5K5=3s=Tbr3HZxF
z^*%FSG-n%P2GOO^1^rfh8NfAXP9P1WUGtfeHj~V_;LM#(;*wEO@_E}}gpSl=(e6wQ
zKcBeu*#oPNIh1<ci!+xf%Y&p=ozFxiB)aU}H-u0RSV+2HP_wtaS0b)2QSXofB9n74
zKzqhgX0rC-EX#whD@G$fLqwxV7TBhOxnz*Vvn`)KQA9jtv+xp?gjWhU|8`|%Dk+3L
z_Vtg#JY@#od-nH6tPDAaX{IaFiB}(8dXcm#-Al-_2?p>k)q{-tmHT<n&vtj~rRE}D
z&EwRrCJkIN>GYP+s2~qD%lC7|8fw7>)Yi1b&H}5s=?t{pkDK->s(0RzxSqFnWt(~R
zME#tRq_beYpMM#m?cbt$GnP3vPBK}B8sv&ZvqbC0RGk2GckOLp+oy>~@P{Go=igDw
zC*&8+buU|=6HoK#2G=$v6yI_eb-5eD5o9UQl^KN;d!En!qC-p~%g7|I3X3;(*1Js1
zKk(N;e-(vr%q*lV<`=q+53{QnG<JcJprq<0M&_nuJ9p*vcf&1lSAx>=<AdJQC(^vA
z?wcsqtV8O_w&v&h;AOO)vNmWf<9x2W4mag0OQH3*T&<U=34mxMIJCW(3-Zm0$CI5u
zT(DN4T^_Nk-Pxm}ZEBS}=`vv$D)-S0)eVtJ6zi`b({qriB9s(0B(TiBROuHJzhnBL
zR^Gi^>nM_dBlc@UX~X&K7qDVrY?ZPD*HkH28oZEB<}svXz?W}Fc&nH7;C<{xhB>_a
zpce-g{=*P}i}@Np28mEst!51X-_q?fUUjPGt>18^H-5PP`R8p`j-35k3y)Ya9CYwN
zT`UqfTQFeIjN?k<OMC<jkXcSrnlsTgaPXy?GV{q4ntHZWko@T1C-S|qmGai{QqR_D
z)gRVbwRr+J1t9PMaLSm`8fHWPiOP9mM+tFwW2f>5Gj^A8pBlg_e1BG+NYw+n4wayA
zzF~&156hRi<}EC9MVK9~$G6z1MK^y@lGU)4YE@(HmdVsenSSSDSZVt@LpF77F6>>Z
z{|2oCq?;e>_&nFfp1~1kyIGT<?i1$9N6o2jij302vhaunXS7!iw>qy9f~D2)mit{^
zHN?XKN1~4q%-heHQ8pwtr3P??&41WIshNiH#vd*@7x`Hhn4I}y%q;n<=+f6YD|b#;
zOfHGdD?07#8LEaEHnUDHY_0{*`Lc3)MxLwU7TGjrqW=Z0%5Oc3^~UE_^~pqUwv-lV
z%(?Md&uHDq`pQ%x5S`ati$b%nZAkbDGpETC%OpB@3vam?A9#{-&osaLP^`2L3}d?w
z6IfJH$6M#^Rpy{{y?8T7u``GwP8g%cn2}3*<=H9GcC`F`z1I~qXaQO|Lx`XR&NNNp
zP=S0^$7kT6e`D-k8ztAW)@un$&LWo#It}~o*FO{ucpslZ6rr^758WrZziARnA73(t
z0aGgeTiF3{<`ni9ds^r-=}^u2N=t|-;ejp+!d@&3T0APxvNIwN2eL9tklG&5^DoD^
zT*ph-$E%wIfFil?aQ_j{%S(X~n?Ytyx0;qCGIkGgY!>}@@IZTlJG(ungNNc@(d5@A
z+tq$9p4C`2^M;sa&F~?45`N$iX$wAF`{4+;`~rguX-}?M&9my=h^c-hRVNG!aj1xo
zam7Pz_<a*w3{bW1=n_FTdwzZQRq8hT!2|x;(%R&AG1hLSz44s$=`D1fPo~io2(9ia
zI_=lf^23{SQ?KIG@pW#xYhLfV$`B+Hi~~kO?$618-!`{0{IaBe5$Sy0j}aAYE<=LS
zg)%|?NG&|d0>g>L0#i}e;Ef(=<@go;gL(_sU&=@KTyPRH2GXt%bd)U5BhQpEcRC6K
zSbfxuDC>jM-+aC-@QL*=jG6ycC%G{8cvtK+mMko+Mj#DIaa7le0ClVGMbyQ&Lhq6J
z1#+qLMb*-BBX^H)ps!H@x2xwxuTqhyDvyWMi+0oy5se5*ndGuz<kg!LDZyhbpRS9%
z!7!(ok5`?n3{Eb4&XdAQZh&2+N|~4Xb2S?aJ;SBWH4Sgxl;f-D>p=Gr7uk~-a{WSw
zp2VL@9PmRGa05kviuB{W)03hKpc;pHv{N3Hmp~j}1}<qbK6l&jost&G)0aOq0w>wt
zL5H>}TIJg=!HE3A9eyqTjl6q@;6Kh#HK$YEIew$MM3(#5oj`Elj2guq1RNM(v=Q^D
zR{Qiga)#)cw(+J|S=LwIkGzg}Z3!3y`|T3Lu>z+V^tAKg1`c!NYRx}BIvz$sHUj8b
z)t7VH`CxvKr5Esiu`CRq3lAc!tkvy1E5(2cIA|uj1dj9(X~G<m%1L`AZFc7VP5kE;
N%{zLxQ7Sf%{trs+hC2WN

literal 0
HcmV?d00001

diff --git a/docs/assets/images/logo-light-mode.svg b/docs/assets/images/logo-light-mode.svg
new file mode 100644
index 000000000..9dc4ae1de
--- /dev/null
+++ b/docs/assets/images/logo-light-mode.svg
@@ -0,0 +1 @@
+<?xml version="1.0" encoding="UTF-8"?><svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 347.6 94.09"><defs><style>.cls-1{fill:#e35a26;}</style></defs><g id="Layer_1-2"><path d="M101.92,79.35c0,3.71-2.14,5.97-5.04,5.97-1.75,0-2.85-.86-3.45-1.64h-.04v1.3h-2.32v-15.8h2.32v5.88h.04c.6-.86,1.7-1.68,3.45-1.68,2.89,0,5.04,2.25,5.04,5.97ZM99.6,79.35c0-2.21-1.08-4.07-3.12-4.07-1.9,0-3.07,1.55-3.07,4.07s1.17,4.07,3.07,4.07c2.03,0,3.12-1.86,3.12-4.07Z"/><path d="M108.52,85.69c-.64,1.75-1.37,3.14-3.54,3.14-.66,0-1.22-.11-1.66-.22v-1.77h.2c.97.13,1.7.07,2.21-.55.42-.51.73-1.39.31-2.45l-4.09-10.14h2.5l3.16,8.64h.04l2.9-8.64h2.47l-4.51,12Z"/><path d="M121.12,84.98v-1.14h-1.14v-1.14h-1.14v-6.88h1.14v-1.14h1.14v-1.14h4.58v1.14h1.15v-5.73h2.29v16.04h-2.29v-1.14h-1.15v1.14h-4.58ZM125.7,82.7v-1.14h1.15v-4.59h-1.15v-1.14h-3.44v1.14h-1.14v4.59h1.14v1.14h3.44Z"/><path d="M133.7,84.98v-1.14h-1.14v-1.14h-1.14v-6.88h1.14v-1.14h1.14v-1.14h5.73v1.14h1.14v1.14h1.14v6.88h-1.14v1.14h-1.14v1.14h-5.73ZM138.29,82.7v-1.14h1.15v-4.59h-1.15v-1.14h-3.44v1.14h-1.14v4.59h1.14v1.14h3.44Z"/><path d="M146.36,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M154.34,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M159.47,84.98v-1.72h1.14v-1.14h1.14v-1.15h1.14v-1.14h1.15v-1.14h-1.15v-1.14h-1.14v-1.15h-1.14v-1.14h-1.14v-1.72h2.29v1.14h1.14v1.14h1.15v1.14h2.29v-1.14h1.15v-1.14h1.14v-1.14h2.29v1.72h-1.14v1.14h-1.14v1.15h-1.14v1.14h-1.15v1.14h1.15v1.14h1.14v1.15h1.14v1.14h1.14v1.72h-2.29v-1.14h-1.14v-1.14h-1.15v-1.14h-2.29v1.14h-1.15v1.14h-1.14v1.14h-2.29Z"/><path d="M175.45,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M180.83,84.98v-2.29h2.29v2.29h-2.29Z"/><path d="M187.59,84.98v-1.14h-1.14v-1.14h-1.14v-2.29h1.14v-1.14h2.29v-1.14h4.59v-1.15h-1.15v-1.14h-3.44v1.14h-3.43v-1.14h1.14v-1.14h1.14v-1.14h5.73v1.14h1.14v1.14h1.14v9.16h-2.29v-1.14h-1.15v1.14h-4.58ZM192.17,82.7v-1.14h1.15v-2.29h-1.15v1.14h-4.01v2.29h4.01Z"/><path d="M197.96,71.81v-2.29h2.29v2.29h-2.29ZM197.96,84.98v-11.45h2.29v11.45h-2.29Z"/><path class="cls-1" d="M15.18,39.46v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M39.46,30.35v-3.04H15.18v3.04h24.28Z"/><rect class="cls-1" x="45.53" y="27.32" width="3.04" height="3.04"/><rect class="cls-1" x="48.56" y="24.28" width="3.04" height="3.04"/><rect class="cls-1" x="51.6" y="27.32" width="3.04" height="3.04"/><rect class="cls-1" x="45.53" y="21.25" width="3.04" height="3.04"/><rect class="cls-1" x="51.6" y="21.25" width="3.04" height="3.04"/><path class="cls-1" d="M15.18,51.6v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M39.46,24.28v-3.04H15.18v3.04h24.28Z"/><path class="cls-1" d="M15.18,63.74v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M15.18,75.88v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M15.18,6.07h6.07v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M24.28,6.07h6.07v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M33.39,6.07h3.04v3.04h9.11v-3.04h3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04h-3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M51.6,6.07h-3.04v3.04h6.07v-3.04h3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><rect class="cls-1" x="57.67" width="3.04" height="3.04"/><rect class="cls-1" x="12.14" y="6.07" width="3.04" height="3.04"/><rect class="cls-1" x="21.25" y="6.07" width="3.04" height="3.04"/><rect class="cls-1" x="30.35" y="6.07" width="3.04" height="3.04"/><path class="cls-1" d="M66.77,9.11v-3.04h-3.04v-3.04h-3.04v3.04h-3.04v3.04h6.07v75.88h-3.04v3.04H9.11v-3.04h-3.04V9.11h3.04v-3.04h3.04v-3.04h-6.07v3.04h-3.04v3.04H0v75.88h3.04v6.07h6.07v3.04h51.6v-3.04h6.07v-6.07h3.04V9.11h-3.04Z"/><path class="cls-1" d="M102.09,60.7v-3.68h-3.68v-3.68h-3.68v-3.68h-3.68v-29.5h3.68v-3.68h3.68v-3.68h3.68v-3.7h22.12v3.7h3.68v3.68h3.7v3.68h3.68v29.5h-3.68v3.68h-3.7v3.68h-3.68v3.68h-22.12ZM120.53,53.34v-3.68h3.68v-3.7h3.68v-22.09h-3.68v-3.7h-3.68v-3.68h-14.74v3.68h-3.7v3.7h-3.68v22.09h3.68v3.7h3.7v3.68h14.74Z"/><path class="cls-1" d="M149.89,60.7v-3.68h-3.68v-3.68h-3.68v-29.47h7.36v25.8h3.68v3.68h9.22v-3.68h3.68v-25.8h7.38v36.83h-7.38v-3.68h-3.68v3.68h-12.9Z"/><path class="cls-1" d="M189.18,60.7v-3.68h-3.68v-27.64h-5.52v-5.52h5.52v-11.06h7.38v11.06h7.36v5.52h-7.36v23.96h7.36v7.36h-11.06Z"/><path class="cls-1" d="M206.38,60.7V12.81h7.36v47.89h-7.36Z"/><path class="cls-1" d="M221.2,20.17v-7.36h7.36v7.36h-7.36ZM221.2,60.7V23.87h7.36v36.83h-7.36Z"/><path class="cls-1" d="M236,60.7V23.87h7.36v3.68h3.68v-3.68h12.9v3.68h3.7v3.68h3.68v29.47h-7.38v-25.8h-3.68v-3.68h-9.22v3.68h-3.68v25.8h-7.36Z"/><path class="cls-1" d="M281.73,60.7v-3.68h-3.68v-3.68h-3.68v-22.12h3.68v-3.68h3.68v-3.68h18.44v3.68h3.68v3.68h3.68v12.9h-25.8v5.54h3.68v3.68h11.06v-3.68h11.06v3.68h-3.68v3.68h-3.68v3.68h-18.44ZM300.18,38.61v-3.7h-3.7v-3.68h-11.06v3.68h-3.68v3.7h18.44Z"/><path class="cls-1" d="M321.81,60.7v-3.68h-3.68v-3.68h-3.68v-3.68h11.03v3.68h12.9v-7.38h-12.9v-3.68h-7.36v-3.68h-3.68v-7.38h3.68v-3.68h3.68v-3.68h18.44v3.68h3.68v3.68h3.68v3.68h-11.06v-3.68h-12.9v7.38h12.9v3.68h7.38v3.68h3.68v7.38h-3.68v3.68h-3.68v3.68h-18.44Z"/></g></svg>
diff --git a/docs/assets/images/logo-simple.png b/docs/assets/images/logo-simple.png
index 2c14b4d9f9d8e38431c44c004594d83ad6a111e8..9bbb39243415b826a71ba65121b20ada5a7b9503 100644
GIT binary patch
literal 3323
zcmc&%X;4#H8V#<vfo)fVB9L~likU${76E068zM47+wj;43J5eIgguf#5NSc1B~AAz
zvg<H(h8MyzAwh%$Xl1bt1SJ6_K%xyqLPFS*K*EEGT~^OHwk=a-)%>{is_w1(z5AW-
zob%m#;|n+E9b5Ib!eFo+E@w`C34>`IP>;WAZc@(>$KYu7XG_c(-*^}dfq6YNt}w}f
zdU8*KlTU&NCM*GaG42u!i^ZCL6^)7yxfpZF6cZO-JZ-NBgKfX=a_YF}<r2wA(DiX)
zVArfPej>`+eAUmyrKmLb{CB12HfHM?{V9U{^z0`Gzi9{!uN0hf+E4jogvp<;o;har
zjV9&Jt#3=6Dv(xJzd$v3?(?Mh&?eoi>9-4aoIvl|I|$jt8=8GLfDA3NUvJq8MIp^@
zH!sblC}Mi?a_lUtVH6+eVx3oOKZ+Zn4GI@T7{?a*Lu2PKYj1Bu3J8Bzagzypk6<sO
zA>!$<UVZ;1$*4%Eh$PiU%G%~`$P~n!k|TsM5*`B|BHvsIQc=fg9td-b`dD#yDuwnv
zp7z9h0c<_mLSGrkFU<Yfgg=6VAi;Te(^?Auh%FTp=3xe#IlDz?=gGP&yRUA1xbwj0
znHMz7H*CAU&*#|o_I1IUzb-|n5|@(W6UX^vk(}X3GMm6O)g{bY8d-WPo_CQ&W$hI@
z8T=s1!O_GTk(qN?jc-F14uv>+*)wf+>a${62jmYG-F0oDm@c0nE8F_Q03lg4EMPry
zGFZNupY9vDJPWR<GK4u(l3F&Ko)#54CUU}~XmV1sE&7$ipda+0$*%<+I-%@nZ#JX3
z!r}g8$7Xe6k_^jc3qJ8AvcXd|i@0A&na7scLgV@QJ=_XjM!6IWT=v&NPrfwJF4Y@G
zUqUUEfRSbdMlfSYjz+D{iMv^yq9^msmq%&{>Vr;k3p%m>N9-2XiM^&$7ont-@wBcW
zI;fYzmLXlS)wvS+!fb^usNLFFu&Fg|(0<pt-$d&zAaeo(38MiHN;m*@w%6~CoFVX@
zM$HE5PWYF@3k!o<L?Hmp!m2-9YMVB*+;?qq#5QIY)RuvbMAiX61~yjk()o4GZwK1X
zOi?4HhDC)vgb-A$Ksz&R=$Znk4Km=>Fav<q!?y8(z=)Ma9~nxnC3hk-2-8{Q9s)Lk
zsyjV{w6l7mGNx-G_3no<5@J^-yr`SAdXrxu1&JA!*&vP(dZ!VmoVg*h*<MEFJ!V6l
zpx%ywI!Q;m5VP7nDLfsfCTaj*`7AmcSrEP2t()2VJU5qHF-ULkGn0uLC0)^7M8j%F
zIxiS+-YOY;Ryo!aVj}D$L;>7w2yyi8GCdAxYG}v^O%-=1@v_(_Flf|*18#z8<ey4c
zG;_DPN1hZmAj7fvoyRu3MZLGce8b8RX4CySj_E|EG?6Can-^mcmKO6$QgkYAp~T5w
z*-pM?s=vE1Sw4JtHF@gQ@#aHPh$Lzou*qSul!-w@<W#1Y)SHp!h)@+9S%x=;KfvlY
zY9W_r#4iDDVy`4iLWo@?@RpTSIAz{DZ5}C}rwuhde--q&#??Crw;%|n3!7!GR2$PF
zaejM!F}A;qt`;40nW9{(>^RqGpQ6lG#p=|hsg*ok1-eVRd=0mxe?VftmFI_r`O$aX
zc4xGW3s|BjhyYJUtjOW=S5z-;gQKyc418*DrzjA?QNAp24#RQz298ZD>G_2{s=O3;
zn%v*Wk|Im9Ymh%==dc29Y#@U;?h`M_jx?HKu%LWM(6=Oa^f0)JymK^RrnX-s-zQaE
zg0w8>&YjWR@X7m~=S_mc?jTtCT+TMtakX({Y&JvJGawG3q+ckD;7ubetf*WqjG^6}
zO@=dsD(3bbLF45t!5;7Z?I1XaNKlslFq%Gj5OI(21K69xiqf}>3!=$y-OcbTMI;16
zD|h#RKN;=YdFq2c7*kLOS`%xR-DFC7l|oQS8(1CdyJ76uGAJzZbx=atVENv{NuZqU
zi|^`YMb*oS)FL6~uo{?)!*h>}39+FQXZQO60AnPOD=o$CNb2Jur2WUU{Bt3lYmwBW
z%Y9XMzl|~sO2$uxw0`!tA#rggb9L)@{?2zT`JKw&bM*ljI;|W({0Ij>^Hhgg*{_Q2
z9r#kktE4C&<rjMK_ErcNB=W+6&fZB%YV1OZ$$6e_(%+?B$LC$Mi=&el8*~k74i%G4
z6rNm?{O1G2um7tHyz|R{H;9kE^oEDC71AjN^ODHnf-!}wt5s?%Yp+t=F1i2!QduuZ
zkaF;CZa54P;F2AE)~2!MuK}*#D#J;dW-c?V``(wFTZO%+ujvIC9IamwaO_}aY&8}B
z56XK3;e>}-%#mld%()uhT7S~$`lNqG$QE9FI+bQGGxYeh8jNHx8ogVwzaY}s1&RRp
z+6wg`@JA3Fg;#kI-5Hz2|4<l{kT2ps_W@`raZq{{7KSWyKJniE|Iy3)PCyn3DTu49
z$Bn7{P_@N4?Po4h$Te%QY;lZ(|I4{L5af!}MbK!K;!0^lk%Je2DZ&w}lJ{@;k4wrs
zG2hU0dY2bKTOo3%hB>IVJik>j$0R^YZBVz|tN93-R@>wGGn&75`FUU5-@h`gbyv<3
zXIp@lwKqQ|oJ>KzF8EUvAt_lh+wUPYKdkmIup@&p#PGmq*!kQ4l6Ut1kT<}u2HEb6
zjpAyI8>4;vSRELdjD|`*q_7tIkFIa?S};7xTrmVj8m;G1mr(2$RZT}X>n_GY76tji
zi!_p?{n92?+d(0F^pJ%m`Tj*+_){dn&eNies%Bi&&cg<{xeqcP+{!yp>p;Ls*%}~d
zoyT9I4}`;cSrsdI_1O<soLrO8J5xj(nPn0QWr@U(SEQ?n<DFfMqFc$dg4mY~sw*qy
zi#@CJ%N!J}V9T122x-J_X(UG?R{lV}^WzGS$d=2aGm~Tm<%-7>qMs(g`ynC+0WkJx
ooF=m)A%l%qHm_eBt7g+qwHY~_RyAv@A1*K#C%02ICoWw57oI%f82|tP

literal 2196
zcmb_edsNa_9Ea7cY-(l8=XAB1bt<(qEfs4EmmU{A2rTe<GGL<Ot28OJJhRlum7B#3
zoKij@P*d=Mg3G2k#fE|S#7f0PiC7UnLiclZoSkFY&gqYP?m74OyXStspU>y}xi|Fy
zY_Ea7xjqO4GVu0-9s+^D2Ef?8R1a9m7jLWvre%>{M`J*sjTtioEMB`E_yc4X>v<#=
zPQb**1xFtT#l^+h{SbkVITjpw+>Q{9&3@``4gxJ*>ka+paQt;;uP~j8HEUFlkTB=W
zS=G1G?0H=;ojO{)hnx<JsDIsDo@;m4<bZFgh0fTo6Op`nlm5EG-byy1@~5BWHyn*m
zxRO!+<=LnetJ6@9&wC=OH^P3~C2WdmBYBKeRTg`W$$QeX3pBD3;b;&)zKS99&{#R7
zmfpwu+No^5s%M-8%dSBeUAUmPVx_^C>&$mUosur#Kr2O{b$MXu>kpmR6V6%;D~2a`
z`B^BrxWbCsP+f?FI*U<l>q=J;5SpP#JtUh$*7`$YoF~-JQqmea6g){|hUXsXR8)6Z
z;`E?&a_D{pwU%C?HtNJEP0D(Ga#}G#8bf({t!yNATYthQZ_aurh@;&Vwv*`eB`LWu
z8%LI;m?0@^YDx7CsLBf0BShdW$hn&HjEb11?+MEOGOCXtKmtn|S+-);yNY^!&_I3-
zq_Yp@Ub0m@lu0Q~=`&y@+PLEA(_1Hqiia(avsns%raN8Rm7}z?_Qn{>yN$T*9vlWW
z0xYN@V#mkCK9)oI&|^3qTh@IwkD>^qDmPoBC9oQX6ibvSH+c-EGSc!6)^ZaNF9>q-
z&1oj1eVaMkc0Qix)_!?R*`!kS@}J}AItR+Igv1x6OkSY~r&L2_!m+@M;bQ|BN$M}O
zgVHH0BPy=X;!;WuJBW(D^Qd90P3-2k-;zv6+A>3FE#(1%D!jIc`eHLuC2oZas0ru$
zTe6x>n2JA&v=0SUw712c3*wCnoQ>q^R{K6v{uvh6i7YoNk-uu2<BoYY0Jgc?2Pr}l
zcnR`R%IBleK2O3=$03yK9a$4O$nw_d@adM$jGl`bHo}u9X+9XkHR)Qzz4ep6oCnoz
zJ*I)K96!E;#;UQ2E*fbqZ7A?9Y2K0tap|s&Z4{$>HtexZbVl~_yWq~WLdFvMWxOu^
zn1!J9*E8*XksbDjCF4cf9hU&@=SMc{^sf_%hcwc5g{{l!O!VE3$-sQsqm)ohGE&1v
zSc!<r#yzZ;D5mNbz$%otL!&2sK<n&k<(D8&9Ov?SP)*Kt`%X9njv^$(Aq|?v@9m%k
zkj&F_A8SV#thP=wYQM(1$K71!2(<o>^<?!;c=O<LxGL&8hujd^A@FFAk5E8@^y`BK
z2M|LAvi=<(;7`sOVjGHg1VNxC7+w4BUw+ECqDr7A;t>!aW9sl9;H(9LXO@I}U8K#!
zL3lMMR6{A!CV9V@5QP!OpNa97*l_wmR$?JTSP-yiy=bC~lqFhNi`!hnehlyGecu_Y
zfC8HB-yahHN1mhI88%A77Aw;f*X=WGN8QrFiOse%q|Rm(Tm^=kq5!V(W(HE@cZBYo
z+W8PO;-uO^JS8U9t!^k~DFr$}IHvup_P>8Otf72#p7}a2gqsV1o=ghn631VO(Uy%1
zQ*^J34=Iw-ijh2cF*x&ya=gg5k_O-ego;E!+&{7-yoeaSZZPCwA;X#Z)Cbd0>R*FZ
zwQ1?Xq<fBNZDNG!#NJFwJ)^+oPK-0wt-0~a+2jY^=rg!J7zU*gGFBNJ$wlgnYo#Vv
zDCE*zuDs||_LL|H_`geUmjAbI!?65CD)&>570tyLMadjGsd6R(ZeYR1fyHVOnLEx3
zk$+)(o;kcT%E{x|K<4x<#s<)m=Sm_F#I#RqNGy(;O)Wr9<&-TPPXoK`aGT5*Q{+2=
P7X`@M69#2}8+7g;FTJ7w

diff --git a/docs/assets/images/logo-square.png b/docs/assets/images/logo-square.png
index 0592b8fa9a2d43e08f7367e187514beabc685e2f..9bbb39243415b826a71ba65121b20ada5a7b9503 100644
GIT binary patch
literal 3323
zcmc&%X;4#H8V#<vfo)fVB9L~likU${76E068zM47+wj;43J5eIgguf#5NSc1B~AAz
zvg<H(h8MyzAwh%$Xl1bt1SJ6_K%xyqLPFS*K*EEGT~^OHwk=a-)%>{is_w1(z5AW-
zob%m#;|n+E9b5Ib!eFo+E@w`C34>`IP>;WAZc@(>$KYu7XG_c(-*^}dfq6YNt}w}f
zdU8*KlTU&NCM*GaG42u!i^ZCL6^)7yxfpZF6cZO-JZ-NBgKfX=a_YF}<r2wA(DiX)
zVArfPej>`+eAUmyrKmLb{CB12HfHM?{V9U{^z0`Gzi9{!uN0hf+E4jogvp<;o;har
zjV9&Jt#3=6Dv(xJzd$v3?(?Mh&?eoi>9-4aoIvl|I|$jt8=8GLfDA3NUvJq8MIp^@
zH!sblC}Mi?a_lUtVH6+eVx3oOKZ+Zn4GI@T7{?a*Lu2PKYj1Bu3J8Bzagzypk6<sO
zA>!$<UVZ;1$*4%Eh$PiU%G%~`$P~n!k|TsM5*`B|BHvsIQc=fg9td-b`dD#yDuwnv
zp7z9h0c<_mLSGrkFU<Yfgg=6VAi;Te(^?Auh%FTp=3xe#IlDz?=gGP&yRUA1xbwj0
znHMz7H*CAU&*#|o_I1IUzb-|n5|@(W6UX^vk(}X3GMm6O)g{bY8d-WPo_CQ&W$hI@
z8T=s1!O_GTk(qN?jc-F14uv>+*)wf+>a${62jmYG-F0oDm@c0nE8F_Q03lg4EMPry
zGFZNupY9vDJPWR<GK4u(l3F&Ko)#54CUU}~XmV1sE&7$ipda+0$*%<+I-%@nZ#JX3
z!r}g8$7Xe6k_^jc3qJ8AvcXd|i@0A&na7scLgV@QJ=_XjM!6IWT=v&NPrfwJF4Y@G
zUqUUEfRSbdMlfSYjz+D{iMv^yq9^msmq%&{>Vr;k3p%m>N9-2XiM^&$7ont-@wBcW
zI;fYzmLXlS)wvS+!fb^usNLFFu&Fg|(0<pt-$d&zAaeo(38MiHN;m*@w%6~CoFVX@
zM$HE5PWYF@3k!o<L?Hmp!m2-9YMVB*+;?qq#5QIY)RuvbMAiX61~yjk()o4GZwK1X
zOi?4HhDC)vgb-A$Ksz&R=$Znk4Km=>Fav<q!?y8(z=)Ma9~nxnC3hk-2-8{Q9s)Lk
zsyjV{w6l7mGNx-G_3no<5@J^-yr`SAdXrxu1&JA!*&vP(dZ!VmoVg*h*<MEFJ!V6l
zpx%ywI!Q;m5VP7nDLfsfCTaj*`7AmcSrEP2t()2VJU5qHF-ULkGn0uLC0)^7M8j%F
zIxiS+-YOY;Ryo!aVj}D$L;>7w2yyi8GCdAxYG}v^O%-=1@v_(_Flf|*18#z8<ey4c
zG;_DPN1hZmAj7fvoyRu3MZLGce8b8RX4CySj_E|EG?6Can-^mcmKO6$QgkYAp~T5w
z*-pM?s=vE1Sw4JtHF@gQ@#aHPh$Lzou*qSul!-w@<W#1Y)SHp!h)@+9S%x=;KfvlY
zY9W_r#4iDDVy`4iLWo@?@RpTSIAz{DZ5}C}rwuhde--q&#??Crw;%|n3!7!GR2$PF
zaejM!F}A;qt`;40nW9{(>^RqGpQ6lG#p=|hsg*ok1-eVRd=0mxe?VftmFI_r`O$aX
zc4xGW3s|BjhyYJUtjOW=S5z-;gQKyc418*DrzjA?QNAp24#RQz298ZD>G_2{s=O3;
zn%v*Wk|Im9Ymh%==dc29Y#@U;?h`M_jx?HKu%LWM(6=Oa^f0)JymK^RrnX-s-zQaE
zg0w8>&YjWR@X7m~=S_mc?jTtCT+TMtakX({Y&JvJGawG3q+ckD;7ubetf*WqjG^6}
zO@=dsD(3bbLF45t!5;7Z?I1XaNKlslFq%Gj5OI(21K69xiqf}>3!=$y-OcbTMI;16
zD|h#RKN;=YdFq2c7*kLOS`%xR-DFC7l|oQS8(1CdyJ76uGAJzZbx=atVENv{NuZqU
zi|^`YMb*oS)FL6~uo{?)!*h>}39+FQXZQO60AnPOD=o$CNb2Jur2WUU{Bt3lYmwBW
z%Y9XMzl|~sO2$uxw0`!tA#rggb9L)@{?2zT`JKw&bM*ljI;|W({0Ij>^Hhgg*{_Q2
z9r#kktE4C&<rjMK_ErcNB=W+6&fZB%YV1OZ$$6e_(%+?B$LC$Mi=&el8*~k74i%G4
z6rNm?{O1G2um7tHyz|R{H;9kE^oEDC71AjN^ODHnf-!}wt5s?%Yp+t=F1i2!QduuZ
zkaF;CZa54P;F2AE)~2!MuK}*#D#J;dW-c?V``(wFTZO%+ujvIC9IamwaO_}aY&8}B
z56XK3;e>}-%#mld%()uhT7S~$`lNqG$QE9FI+bQGGxYeh8jNHx8ogVwzaY}s1&RRp
z+6wg`@JA3Fg;#kI-5Hz2|4<l{kT2ps_W@`raZq{{7KSWyKJniE|Iy3)PCyn3DTu49
z$Bn7{P_@N4?Po4h$Te%QY;lZ(|I4{L5af!}MbK!K;!0^lk%Je2DZ&w}lJ{@;k4wrs
zG2hU0dY2bKTOo3%hB>IVJik>j$0R^YZBVz|tN93-R@>wGGn&75`FUU5-@h`gbyv<3
zXIp@lwKqQ|oJ>KzF8EUvAt_lh+wUPYKdkmIup@&p#PGmq*!kQ4l6Ut1kT<}u2HEb6
zjpAyI8>4;vSRELdjD|`*q_7tIkFIa?S};7xTrmVj8m;G1mr(2$RZT}X>n_GY76tji
zi!_p?{n92?+d(0F^pJ%m`Tj*+_){dn&eNies%Bi&&cg<{xeqcP+{!yp>p;Ls*%}~d
zoyT9I4}`;cSrsdI_1O<soLrO8J5xj(nPn0QWr@U(SEQ?n<DFfMqFc$dg4mY~sw*qy
zi#@CJ%N!J}V9T122x-J_X(UG?R{lV}^WzGS$d=2aGm~Tm<%-7>qMs(g`ynC+0WkJx
ooF=m)A%l%qHm_eBt7g+qwHY~_RyAv@A1*K#C%02ICoWw57oI%f82|tP

literal 2233
zcmcgue>9VO7=PXM8_DUO${Zw>>J%v?8>;KK?o<p-k(Mas$8e|ld9J&pj4P!(W4j2M
zm<WY6+G)EYn%`%b7Ny24v5n2RukJbBQqmvy+&{kOJ?DMi_kBLk=lML(=S@6<uv1+H
zT?GJus)N1tF#u3f1<!#M%fLJN?c1xt%gSK;<DmesJ!S4Fwd<OIPXHQWHXdOvL1)9T
zJ{VsBi^Uq83k(QF`vm(M1Y!J0qi`qytgv;kKIj^eDeP-WAtN=|d^udCvcoYr-k+#z
zaJy~CQPlBlPS#4N_0Xp`CiRZG7i%iko@ltrT4R*4!Q9p30n0nYN3d$!`L*VG&^8U|
zMx)f<wwYxOJ$SAfk}+{DT3(AEMzv&x8fG~6mz!NeiRip%m0ew4G^*5@zWlJy1~yZU
zphs-YAa?DkgQfWc>KtHGj*_)C<jR$0Yt&UYZh{`LHjcaUPr%8=FnAYzqOM+1Q(nD=
zNJM676!k?_#P(Ae1yoD*5{cAHXvkM->TBK|HTY6i|Ax(k?SPtS#=>2rO*I~#^P5WV
zA)Cu(9Wx4vj4h&aEp&t(a+*7(d-|9$q^n*c;}WT3<TEVslCP!z6jxbkkh6i-{s?a!
zR!tf)WQ1|CnErcHtwctGRrwax*Ki-(qB99<t~4~~`7wF#(*Y8L)aH2LAsNv9R)t`S
zlE$7iC%e_WeU)QH4NXUw6r3jGX48q7(5WZ>M&~0Hb<!fK=Mc_**w5Xx+3I@I1)Y;M
z43_L=ts5#V`cD5{a&HtyU?IVr%(({}UXG##ZQhF(#zguKOryuSiL?G4I)dUribz-Z
zmU=aZgVPf1lJ6Twv{HF@sQ9GI!z;gF{nN6+c5IG`#JtUh1k|7{Qk(8my$L40q^92c
zp;M<THG=F!LS=M7<2W4c52NMng~1xFG}5uLMk&#OOPp$8Gz6V_b5O0*K*a04%n#nv
zN>%Wi3G2y(meZqdWHTwd5Zm2oM=_UUkaW0Md)lyT?dJ-ZhXs_zU#$`uDrM_s{?tvv
zGt+|TxTQn4zEsY<_cqUtI&{blN_3_tO!^0x8EMXn>q9(Z42b87XW+Ex8Sfr7?a9Gj
zt90X$+RFEnc;$vBZ$^|fA>maY>~%$3jwclB)rS>4?re`7c_p+)U%WF$40zO_nxp2l
z-fdrVpNUwz%P~N4>nCv!*eQy|Nu*R?udxy4EkgE_+H94BGM^z_JH`s7-PaZ;d{9_x
zG36<iP0c~&pdx(*S|xd2!ZCrOb3X-RO7;)WXURD5%7^L`ZUqHAz9Y9YUS4i^7iqyL
zyLZXopa<yc_@L}+ePW1((!%34kEDO+?*9nP&GwdC$x#=??@On%3#D<}cM@Q<QCPT7
z0I24&QqBgya?$E&xT#osH$KD^b`;4j;Gqpm^gr1TU&Wb<kB)3DLR5K)vT$mD9104@
zf&88E$-Rfej5knm;H<<|Vg5T>Qt=9p)=t@s$W_%Yp9je|BBT1uX5Y1Yj`;fIa!@tp
zI4e-uyAwOtNy?MH4}BR1GaiA|qTtkC6i>H)V?Tbi{slMc2c&s=%SFS?2x)s{-H%+e
zOi<MLerGz{Yn>@{;6l`&I4#l~X%vD!^}KE_ANRuooB3nr<K0#I%+Yt{mPvjEVD-;`
zNx&<r7^R#^kun1xRSFTEUA*Q65zh~lmFhTgEvqH=V)Hp!B)aLlAp*52I0Ah#Zy^u)
z3@KfD4ap335h<m;8F*6Y{@^;Ez!I>g5{d-;LMgy|y0{N~=1(8LJ>R&%09O((XAoD5
zy`;7YD&ex^G?LZNElMRkC%djkt!&xhJ#8q9G;(<uoAnyqP?x8N)e<n@G2h(arBfY^
zPmyu*sEV12Gq(u?xgQMkdnS2j9%NutbBM)<aN;X0lM*UB)!-!Z(i3!3m6T-k>_OxE
z@|4iG`$LS~$XuBp(nhJ<9QVSdSjL~&`9dJJOvVOglH{??i#IGa7NYgUCOf3JwIBTL
O032)()+L8d{rWfXa;-}M

diff --git a/docs/assets/images/logo-square.svg b/docs/assets/images/logo-square.svg
index 29b1c20f2..efc68a642 100644
--- a/docs/assets/images/logo-square.svg
+++ b/docs/assets/images/logo-square.svg
@@ -1,23 +1,22 @@
 <?xml version="1.0" encoding="UTF-8" standalone="no"?>
 <svg
-   width="95.38842"
-   height="113.55765"
-   viewBox="0 0 95.388419 113.55765"
+   width="69.82"
+   height="94.089996"
+   viewBox="0 0 69.82 94.089996"
    fill="none"
    version="1.1"
-   id="svg314"
-   sodipodi:docname="logo-square.svg"
-   xml:space="preserve"
-   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
-   inkscape:export-filename="logo.png"
-   inkscape:export-xdpi="270.25565"
-   inkscape:export-ydpi="270.25565"
+   id="svg39"
+   sodipodi:docname="Outlines_white dottxt.ai.svg"
+   inkscape:export-filename="../projects/outlines/docs/assets/images/logo-dark-mode.png"
+   inkscape:export-xdpi="137.93103"
+   inkscape:export-ydpi="137.93103"
+   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
    xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
    xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
    xmlns="http://www.w3.org/2000/svg"
-   xmlns:svg="http://www.w3.org/2000/svg"><defs
-     id="defs314" /><sodipodi:namedview
-     id="namedview314"
+   xmlns:svg="http://www.w3.org/2000/svg">
+  <sodipodi:namedview
+     id="namedview39"
      pagecolor="#ffffff"
      bordercolor="#000000"
      borderopacity="0.25"
@@ -25,100 +24,110 @@
      inkscape:pageopacity="0.0"
      inkscape:pagecheckerboard="0"
      inkscape:deskcolor="#d1d1d1"
-     showgrid="false"
-     showguides="true"
-     inkscape:zoom="1.0083333"
-     inkscape:cx="280.66117"
-     inkscape:cy="25.785125"
-     inkscape:window-width="3440"
-     inkscape:window-height="1416"
-     inkscape:window-x="0"
-     inkscape:window-y="0"
-     inkscape:window-maximized="1"
-     inkscape:current-layer="svg314"><sodipodi:guide
-       position="-146.75933,64.2455"
-       orientation="0,-1"
-       id="guide314"
-       inkscape:locked="false" /><sodipodi:guide
-       position="275.80377,3.1942902"
-       orientation="1,0"
-       id="guide315"
-       inkscape:locked="false" /></sodipodi:namedview><g
-     id="g13"
-     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
-     style="fill:#e35a26;fill-opacity:1"><g
-       id="g314"
-       transform="translate(1367.0913,152.50405)"
-       style="fill:#e35a26;fill-opacity:1"><rect
-         x="28"
-         y="644"
-         width="28"
-         height="28"
-         id="rect1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
-         id="path1-7"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="28"
-         y="112"
-         width="28"
-         height="28"
-         id="rect2"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,560 v 28 h 420 v -28 z"
-         id="path2-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,476 v 28 h 420 v -28 z"
-         id="path3-4"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,392 v 28 h 420 v -28 z"
-         id="path4-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,308 v 28 h 420 v -28 z"
-         id="path5-8"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,168 v 28 h 140 v -28 z"
-         id="path6-5"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 336,196 v -28 h -56 v 28 z"
-         id="path7-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 336,112 V 84 h -56 v 28 z"
-         id="path8-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,84 v 28 H 504 V 84 Z"
-         id="path9-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 56,672 v 28 h 476 v -28 z"
-         id="path10-2"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="56"
-         y="84"
-         width="28"
-         height="28"
-         id="rect10"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="112"
-         y="28"
-         width="28"
-         height="28"
-         id="rect11"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="84"
-         y="56"
-         width="28"
-         height="28"
-         id="rect12"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
-         id="path12-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 560,28 v 616 h 28 V 28 Z"
-         id="path13-9"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="532"
-         y="644"
-         width="28"
-         height="28"
-         id="rect13"
-         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
+     inkscape:zoom="0.27368421"
+     inkscape:cx="171.73077"
+     inkscape:cy="45.673077"
+     inkscape:window-width="1416"
+     inkscape:window-height="820"
+     inkscape:window-x="30"
+     inkscape:window-y="38"
+     inkscape:window-maximized="0"
+     inkscape:current-layer="g39" />
+  <g
+     clip-path="url(#clip0_56_2)"
+     id="g39">
+    <path
+       d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
+       fill="#e35a26"
+       id="path12" />
+    <path
+       d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
+       fill="#e35a26"
+       id="path13" />
+    <path
+       d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path14" />
+    <path
+       d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path15" />
+    <path
+       d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path16" />
+    <path
+       d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path17" />
+    <path
+       d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path18" />
+    <path
+       d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
+       fill="#e35a26"
+       id="path19" />
+    <path
+       d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
+       fill="#e35a26"
+       id="path20" />
+    <path
+       d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
+       fill="#e35a26"
+       id="path21" />
+    <path
+       d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
+       fill="#e35a26"
+       id="path22" />
+    <path
+       d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path23" />
+    <path
+       d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path24" />
+    <path
+       d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path25" />
+    <path
+       d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path26" />
+    <path
+       d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path27" />
+    <path
+       d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path28" />
+    <path
+       d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path29" />
+    <path
+       d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path30" />
+    <path
+       d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
+       fill="#e35a26"
+       id="path31" />
+  </g>
+  <defs
+     id="defs39">
+    <clipPath
+       id="clip0_56_2">
+      <rect
+         width="347.60001"
+         height="94.089996"
+         fill="#ffffff"
+         id="rect39"
+         x="0"
+         y="0" />
+    </clipPath>
+  </defs>
+</svg>
diff --git a/docs/assets/images/logo.png b/docs/assets/images/logo.png
deleted file mode 100644
index 005b7c6d2195cf0af295c15fd2e167f3f30b1fad..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 6386
zcmcgw2~^T~yQb+Iota#QoE9~8(oD-yGjheWnMlngQ!8_W%F+}G6&1~enx+YJD?@QZ
zO0(4^H`G+riL4MY_bnwcB{v8a6<PfL^v!qfnR`3k@1E)0!-4nkUw+H`d*1hbo;UI6
z5qrh&)V>3OK#GSQVD2E$Kj0wHx&z<H0avcxuB`;jx1kPR7!YVzn(Vu-XS=B)a8Whl
z(8&l7bU+05boe<C7K=3sx`4#^pAJ1|gbokP<yxzOKwEK#VLu;>&YK&Fzd;I&(RwS2
z>F^r-#o|D-N&1Oaxx)r+J0n*1MyNGscd6#xaCUfh?Zn>N@AbQ1XW#htp#0E*so%0Q
z<D4BFHI!S*tHRHuE0k<FokNK`yYh|yZN^qFgUc;L8K}z#^9BaeqdG*3X}|AUe^a74
zH!8}{D4pSn>R9B66Z4(uWlN*#J_`h?mg|+9X{V<(4V7Pf_j>V-I3xZVuoDZSX}mmO
zK^C67fW3o2>H2@q5WJBr)f4UIjnF82jSI6@P1}cB+zHCb3&|S=d~i9Y0TNb_5+0Wa
zxltwwNNFic!<^Yom*jGkMo1aofIx43A+ylt;NakeF?<xZUpS$!jLwBN1my+(#I0xR
zc2dY&InIU?lPGzmCKNEp+(5pDr&&>>e;x!ya{-6NAJGPZE@9Fh+sQU__y^-JVSbO7
zx489uKNRlk2!V_~4EO0-&E}2^da}x8qg&+;u#4<4@vw#{U=bRAky9;&2$$>s0B-Hs
z?3p5a<nKE5&ofczXKzmul|~;<9@;-n3T`@DAe=pEPj8I{KEgR^Kkw=LcRB*|3UW6r
zb~y8jJzAZ^VCR$2iTWK3`o1a0;Rf`K)=11<-Zo%O#f+REjH@Z0nH!rdfIUh2>V&!O
zUxC5k!lH>3H#vW*(KxcOPEWgaH@tvAJVa}A|Egqt<ZdOq-`HP+K&MEUUf{SuKPJll
z{cZCq^v(JgOWXm!=~HVKfjRnG8i!)I7J+ilRwhrjY#r9!FigdSw8i^38U+@bP2`s=
zn-=0mC?2j3_)4P?Ytt+qbUm;N1M?`pd%@tb0-IRIL2}b#^O$j}=%@5}wn`h(&iUQU
z-o`Z~V&>zc#e^@5kfcgTWZrsR_R>_HFnyEEGjDyz^!{Eb{WTI5I@pZy)OI*O@3?El
zk$tEbcNCr2YUvSNpOq1KsI)F18gNhtJ;518-|sq%dp{b+{E0hEZA9JAzmXBth4NKn
zAmU~;lm#TfRhKd>bJ^#xDsyv>^UpS(r{AZvYuM;rzppa9IiX5baURiTHmU*Usqq+Z
zu{q}vndyNV7Lh;us({I}P+7V#7=Hu?%z=&1H+%t;KIWAtjOtl=`AE-8FW7@4Vm|Sf
z^LS>TnK_d(*b!79VGmpjmMq{avc3=W({~DUWvK$Qo~*0f1?T45Sgf@kcl{$LmM*Vs
zyA`Y=^gg0xJujs~fJd4g>w;x*cmu}&N@m`Ll&Y-*9SB^eE~yezGw+Y)R5t~+mH1q!
zTd4FBtvjVF%NKXm43)`hZmjg`CZ%Dh&M*A0nqlFe1G{qn*Pu)G<xj92bcVP4s5udD
z974)vk;~ODy}BK}4!GwwG(V|tl2Klt8GQx@94qs*Y#kub4z~v2_rDwd*i#TFk((f3
z8pvIl=N3!mZV%$7+OeaV)mosFs({N66a;*~#h%s?IA>hgxkVdTB1naPTn8BH(vIs}
z{-4Y<DgdPB9rdC9bt{)W3e(GDEvn<J{Eb1k4$8cp#@?l6Eq?2<s<f{8>BN_W%jZCs
zewww8`EGP@#MS5eDuIeQpk-#HIB853q+VNP{=7yl=S&fu6?2kSndO`13R~9aDl=Iu
zY-%OH8j1A{9lJF{DVFFVg?&rw?bnQuI4v$tQkrxmA5PtYTst-x!P7OzP1GBVU_zoM
zr(7W3W9E<Kd`{$ycM{z58*0&_rD93<a8rV(*c(;3R3^WnY?Pn8_qbClx5!QOc0!(K
zb+a6*eP+zGo<bO#AKx3LJf!F(t9>JpQd(rkP}Q-fdP|+Q;2ax)08J~2?7^lUCI|yh
zcm1SnpG1s*udXLzcHWPLSH}B5lQ+orVDx5>HJ_lYnXm`BM%!%M7P-j1M?;F3>*D__
zUM0D~_eQs%W~ZPpyiZ=3=D}#j%cSuT-w(gFKnhE685CQ(@gbUg#nSTFyN4@-53blU
zhMIhjH15`xo-K;a!LO;zo$7Xsy1TZasCBqN34oz=#jidg1W7{#&LNZ?yW>oUvgiZ*
zdli)IXw>8pFu15k(y0Uo5MgPDwWoF0Gs&fT2=Ad5a==+H7gn-C4SaJLf@QhWO?4j@
zrmimAh}lnyh#$b@DX2X;SoO_jlyxO<8*=PsH~)A1%F(`rFH=VqL`tAg8?U^PMx8Ul
zY(I_@>K^i8_RXDk56_9%vrNwjl9qyf4O;rFOskFYkLJRQZ!B<VqJBq*LZ!EydS;nl
z-iYu3ad3zbH7pgToeiXivm!kkqGR@54DT1J746Etywf3Nwcz<6i5Nj5!>v{k<9$|i
zMwR=9vO&R{t}0>_sjNu+QaFE+=uNJ)s9!(>u7Rsb%I>pP1<Zw8J`IG`R}Us4SSGiX
z&+Xz)LDyRbLoUVywc)0%arp!z9B&98j_MzyXu!3==6aLs-ntR+NjmgvB4+uJV~M$4
zgqRcHV+StZZhEw!^8wYoz!yMPoX|cR1}oufwF)<}M0X9hpi-kPOuFcb)5eDkMNd=7
zu4UMsXbda{)6d@^-;U8?xA}4ODA9vv*mw8vwGTO4Wl1WqJ@8Y>Dtn*<NOA9dQfkY0
zg^hiO4M|nAt@qAOIGQ+Lr!O$@5?zaY_NsUUaQji<^}H5Joe<$Qm3noMWW3w0>pIbF
zC@lmU`Ys+wzn3z?{+Bd+@^XFIaHQ~dN!g4QA~@z}So?-!?B!rX7Blt$)}?2}l%3%v
zLaN)=)M6)9{W!#-qO~|-OZ}0v6YJiZjCd7_`QElwFj5L$<8XcM^F|1dYmFFR0hkY~
z<K;$hSR9K<epGGUv~XFi+bz7J2amQVu-XP~sQ3DCa3|l4YMk)Sizfdt9wWSF>f+ks
zWgQ-33X;Yf;s2`E-u9GfZkk`coA-;8_=R`dv&ZGTom9eHQdZf;5>&Ui4VZ2@lH}O{
zDsKL|ur|b!O&W3Ck!b|(q|7N2HLX{Bd#*)P0g%`P)-8!n#FG=%Y`Y_)cbkcWWEEr8
zo&A!s{u-;YFU>EH6mxpVCS$N$LF$6dj~CMLRqP_v_<cuOu9cR<Za-;)K+=stu3aR*
z`eQWPz^UB;?noQ1Z)!KN*FN+7xs&0Q$Au-NCh3V8zxN*^M7gewg7>hvLZ0*HWu`)*
zoi~3Y7`>Ha-t%SZiL6S52AHyI3U`sGt=4K4ulQ%jw3-C}PDU0;>Pi=iN{rL+xSLzQ
zy>$E-Qg}jv?E-|Q53h}ZY2{=tv1-nJKv;F8u@AaUt3V(n2*H-%?w_#PdBNi?v-|v2
zGnRW#Z=1D(53P>k2Qsqouq$R4yfg{MOQi8wpq8r?!-QRn(r0TWylsKwK_cZ5bug*Y
zm^&O9Q)Oxb20eb%V^5hQlULU4Q2Rp{7seUJOs-W%#<@nQ&|FYBY#SBrCzXcOPkfjg
zXK0a0!s9&0zOr_Cj<P*%jEaf-x6IhNqzB&|7&v;ou%Uss?e#Uai4-Dp#L~o84IoR+
zgV}&U=&LOztj@{`lB0*w^uOpT=l@z2Wz`J`#jH4Wa*ud|KbSp)O|5=yH@iFjTC{>w
zVK;7H$UU3kNNUdt792L`NQUO9#fevD8S{?IXM>&UP7hyxhI@@o<$a_bAF_gJm{sPJ
z6FVCt67jX-w|;usY*@eWY9j>acjAuOW4E<Vdd-f<e3mHngT&WV#-JPufM5sA-g3_r
zibfvJ57ton@ATmMsr2wkg7fD^)g=vJEYQp9xy76PTGtI&S>>XzT<WlbD|+QZE4OkH
zy=9rvpo5H2rS<wOci@U<8a|2*W@JrYCl~KV=Xz$guzQ$<-oQeb1}yRCA?ZX^@$kV<
zax1>7ObS|4yfxoL|ERkSlMGWKPbeStq4m96%E^?FwSi8Lc+Mq3>*s)P<13?A7O|do
zDLP@bysc>uGw<6yboHIeiF)lDwfq>HKWkO?(MQbDvfd~wHl6Bko%wt)TV0#2XaGMp
z?kabsE9TATrce=$Jxz*@YN#ay+>?Q6A|Q0Qdr9kQ&7ZOjDD$Q{Gdk>QV6ip|1)7Q9
z$Vhmz)#uQZo+t_R18E2MjOdjP3vP|L`hM^v0zClUr0z7A>dCFG^|go&;H<CaIwV(?
zX5h2Ks7jh4Kq<5hUF7YBNN%h2??-yj9zL#T_CKh13iuIpV&rRAHR?NBWPRpShEYx7
zwHi%}RpG3&pP_G!=#gTl^d!qt`vh-!GQU5BWtOFOgeJOgb$zv`?iu+#^u=BG29WcU
zjr^q|xA@q8A;m^K4(9jL97u+)8DzYhB$Q>zR1_FG6-GPtPo<bj55wY&OkOKnYB&*z
zZjz+Z1;;Iy1ovU0@*=nXsYxcjcDD&LdZ-gh3>N>wGLe!e9bmrYU50a^E7zlr;V_%N
zD)BW$!JGve{_cNO1b+&HznZgW*2?U3hm<qU+YDos-aiW`2tUe+C*dkS4bEEDs}J!j
z$5@s!@G>($sUvqTLExaiD-!`RcPLj_Tq{=3{H*tMQpNnfqQROyP>eQmEt;N2WC%>9
z_J&~(W}>0zsiaLss>xg=vRgQF8&<cwK-yhwU9AAG-UD6uwOF!t!=3KYKOYlE?h@1f
zm}TfACFh?_Df=$nLFX}sl#BB>@@<hc9e+lXm3F#RYGnHqc(2W}{zX4*(5-NC|7`-R
zq}m9pmNH+`8+{fvY-~>m8@(p2^q%JSMlUJvs&?+0iQnmB=-FHG9YS7rGB}0FGu&hh
zGT0*%%}IE~lvUFc^khj<N)=mIJHfPiHtWZ8-`6<v$W3*;ZD6pE@;1L@6ROhY<@Sc{
zLk<RT?P;Gr@yzX=)a(hj3~!v3c7hc^p<~C|!FQrBN1nRKeMvq<BS(^KZR%&Png+Y`
z-HmsFd2NIHsXbk~+H$#?V)I<Ig*0^fU_vvcPKTU|cg2zr15k{|!<!xrFv8)BjU8u?
zI}-8d?dH&^x{3EhXh#{{LzbRaw*O5LBr95ZuE8kV(S7X_l5*8Vis|$Q6&;sUZ)t(T
zbW7>@oVzHSVexRN<-Q~2DRm|qHU1P|yVInT-fIDu<1<UgQHE=^J*!Cow|pM%m(SXh
zH7&i^A05;)a4Y-UAGHfISM%N3?9v64x_RcpJJyXN0wT>lEsDevn=j|l8I@{sZWoGO
zBrHD;(q~k~?LGwJa>G{NMSQM*ENbf4`WIdaaPggWSw9dpEiUS7D@BqJZ=efZ55nL;
zODBkzH;Q{VNyQXjovJ%%uT!v)TQ?pXSXvrckwkQ-ulHHj;)+)dxg))Hv>3Mr--HjP
zD&S7ZazIqSiq0eU*AC|Kr=%S(vJrWty6pyScXxrcSQWSgNy06T){}|mu?`rFgXre(
zhoDPznMv#-vx&-X|AsVG;wK~Fa~-&keYLCmA?6ty%jST-WI<4A_9%{{CrSlQv6;S-
z_!lXw65IRDC49>%p~l?&TXR4&?39}rj<?V9or`rMsEdy9&pUc(ss>U^)NQQrw>n<Y
zG1H6l<`q$a>OB{q7z(=MIV<@kY1<M`6+z|FaPL!>?zjY%VfS6RS_LQ;w|qo_i}3ez
zkgf50{=DTLO1wWrf6KYq3r_n<<uz!OMq|WL`Z4-^=i4ztBeE#-vC}JI<(?q9_N(?H
z=Iffh`eoNCBLmn})3D&_las3zPXYrq78rBgn1v3f!wd0ots6MirYp%r|JW$sLBeEe
z7?Xg{r7+tO<cx1+y&?>}Dnph};$<xPe{=*sSw(^J`PSJn5#6VckTrGhQ23|*?29BC
z!tZe=S)0JTX-iovyV>iDHtW7j%kPt;ym|rydcEtbVSaBQ3nG6~;%XvFKJCq3_`Xxy
zfEF7sO>IjA3GKK*Dqj=#M<s~Q0=wgsjR6e4+>rnfmktBnR;zA8#Sh@?TV4uQ^~=MO
zg$I2&lV#3MvaZ*^5bE_gO#bKhe)rcJSbXhC)LgyENyP+wFNoPtSx>d#R$k(oRS9xD
z+E3LjT&@s1oY@8}QoEqXyIZT`33d{Zq1Kcu{m~*#QFHC}y>|0$xZ00}U5xuAZ~5fW
e{|6U^O+Qpov+E8&+y~4KI(+B|tn`;tzy1^S?}<qO

diff --git a/docs/assets/images/logo.svg b/docs/assets/images/logo.svg
deleted file mode 100644
index 440b3f0dc..000000000
--- a/docs/assets/images/logo.svg
+++ /dev/null
@@ -1,134 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<svg
-   width="397.00375"
-   height="113.55765"
-   viewBox="0 0 397.00375 113.55765"
-   fill="none"
-   version="1.1"
-   id="svg314"
-   sodipodi:docname="logo.svg"
-   xml:space="preserve"
-   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
-   inkscape:export-filename="../../../../../Downloads/logo-square-side.png"
-   inkscape:export-xdpi="270.25565"
-   inkscape:export-ydpi="270.25565"
-   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
-   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
-   xmlns="http://www.w3.org/2000/svg"
-   xmlns:svg="http://www.w3.org/2000/svg"><defs
-     id="defs314" /><sodipodi:namedview
-     id="namedview314"
-     pagecolor="#ffffff"
-     bordercolor="#000000"
-     borderopacity="0.25"
-     inkscape:showpageshadow="2"
-     inkscape:pageopacity="0.0"
-     inkscape:pagecheckerboard="0"
-     inkscape:deskcolor="#d1d1d1"
-     showgrid="false"
-     showguides="true"
-     inkscape:zoom="1.0083333"
-     inkscape:cx="280.66117"
-     inkscape:cy="25.785125"
-     inkscape:window-width="3440"
-     inkscape:window-height="1416"
-     inkscape:window-x="0"
-     inkscape:window-y="0"
-     inkscape:window-maximized="1"
-     inkscape:current-layer="svg314"><sodipodi:guide
-       position="-146.75933,121.72017"
-       orientation="0,-1"
-       id="guide314"
-       inkscape:locked="false" /><sodipodi:guide
-       position="275.80377,60.668961"
-       orientation="1,0"
-       id="guide315"
-       inkscape:locked="false" /></sodipodi:namedview><text
-     xml:space="preserve"
-     style="font-size:117.641px;font-family:'PP Neue Montreal Mono';-inkscape-font-specification:'PP Neue Montreal Mono';opacity:0.99;fill:#e35a26;fill-opacity:1;stroke-width:0.612713"
-     x="115.73952"
-     y="91.977005"
-     id="text314"><tspan
-       sodipodi:role="line"
-       id="tspan314"
-       x="115.73952"
-       y="91.977005"
-       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:'PP NeueBit';-inkscape-font-specification:'PP NeueBit Bold';fill:#e35a26;fill-opacity:1;stroke-width:0.612713">outlines</tspan></text><g
-     id="g13"
-     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
-     style="fill:#e35a26;fill-opacity:1"><g
-       id="g314"
-       transform="translate(1367.0913,152.50405)"
-       style="fill:#e35a26;fill-opacity:1"><rect
-         x="28"
-         y="644"
-         width="28"
-         height="28"
-         id="rect1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
-         id="path1-7"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="28"
-         y="112"
-         width="28"
-         height="28"
-         id="rect2"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,560 v 28 h 420 v -28 z"
-         id="path2-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,476 v 28 h 420 v -28 z"
-         id="path3-4"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,392 v 28 h 420 v -28 z"
-         id="path4-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,308 v 28 h 420 v -28 z"
-         id="path5-8"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,168 v 28 h 140 v -28 z"
-         id="path6-5"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 336,196 v -28 h -56 v 28 z"
-         id="path7-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 336,112 V 84 h -56 v 28 z"
-         id="path8-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,84 v 28 H 504 V 84 Z"
-         id="path9-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 56,672 v 28 h 476 v -28 z"
-         id="path10-2"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="56"
-         y="84"
-         width="28"
-         height="28"
-         id="rect10"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="112"
-         y="28"
-         width="28"
-         height="28"
-         id="rect11"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="84"
-         y="56"
-         width="28"
-         height="28"
-         id="rect12"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
-         id="path12-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 560,28 v 616 h 28 V 28 Z"
-         id="path13-9"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="532"
-         y="644"
-         width="28"
-         height="28"
-         id="rect13"
-         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
diff --git a/docs/overrides/home.html b/docs/overrides/home.html
index de5b2258a..9e7fdaf0b 100644
--- a/docs/overrides/home.html
+++ b/docs/overrides/home.html
@@ -119,7 +119,8 @@
     <div class="md-grid md-typeset">
         <div class="mdx-hero">
             <div class="mdx-hero__image">
-                <img src="assets/images/logo.png" alt="Outlines Logo" width="600" draggable="false">
+                <img src="assets/images/logo-light-mode.svg#only-light" alt="Outlines Logo" width="600" draggable="false">
+                <img src="assets/images/logo-dark-mode.svg#only-dark" alt="Outlines Logo" width="600" draggable="false">
             </div>
             <div class="mdx-hero__content">
                 <h2 class="subtitle" style="font-weight: 500; padding-top: 1rem;">

From 837c814497fbdcf9018d229227b4f0aea6b9c427 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Thu, 24 Apr 2025 11:13:05 -0700
Subject: [PATCH 190/505] Add commerical/contact info for .txt

---
 README.md | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 419282561..b20fc821c 100644
--- a/README.md
+++ b/README.md
@@ -23,7 +23,9 @@ Made with ❤👷️ by the team at [.txt](https://dottxt.co).
 pip install outlines
 ```
 
-First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/latest/welcome/)
+First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/latest/welcome/)/
+
+Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales) with us.
 
 ## Features
 

From 2fd02438846d84fcdaf969c88a851627ecdc9c3b Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Wed, 30 Apr 2025 10:54:56 -0700
Subject: [PATCH 191/505] Make docs "blog" link to blog.dottxt.co

---
 mkdocs.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mkdocs.yml b/mkdocs.yml
index b31fc1f54..2e77a0440 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -180,4 +180,4 @@ nav:
       - How to contribute 🏗️: community/contribute.md
       - Your projects 👏: community/examples.md
       - Versioning Guide 📌: community/versioning.md
-  - Blog: blog/index.md
+  - Blog: https://blog.dottxt.co

From 7fc5f1cf031c1e31bbc69cf89dfc572625e107cf Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Wed, 18 Jun 2025 12:07:24 +0200
Subject: [PATCH 192/505] Merge v1.0 into main (#1619)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Co-authored-by: Rémi Louf <remi@thetypicalset.com>
Co-authored-by: Rémi Louf <remilouf@protonmail.com>
Co-authored-by: Victoria Terenina <torymur@gmail.com>
Co-authored-by: Robin Picard <robinpicard@device-36.home>
Co-authored-by: Robin Picard <robin@dottxt.co>
Co-authored-by: Cameron Pfiffer <cameron@pfiffer.org>
---
 .github/workflows/tests.yml                   |  29 +-
 .github/workflows/tests_nightly.yml           |  60 ++
 Dockerfile                                    |  23 -
 README.md                                     | 802 +++++++++++-------
 benchmarks/bench_cfg_guide.py                 |   2 +-
 benchmarks/bench_json_schema.py               |   2 +-
 benchmarks/bench_regex_guide.py               |   2 +-
 docs/api/guide.md                             |   1 -
 docs/api/json_schema.md                       |   1 -
 docs/api/models.md                            |   1 -
 docs/api/parsing.md                           |   1 -
 docs/api/regex.md                             |   1 -
 docs/api/samplers.md                          |   1 -
 docs/api/templates.md                         |   1 -
 docs/{api => api_reference}/index.md          |   0
 docs/assets/images/install.png                | Bin 0 -> 12669 bytes
 docs/assets/images/logo-square.png            | Bin 3323 -> 13864 bytes
 docs/assets/images/logo-square.svg            | 192 +++--
 docs/assets/images/readme-dark.png            | Bin 0 -> 38912 bytes
 docs/assets/images/readme-light.png           | Bin 0 -> 38943 bytes
 docs/assets/images/use_philosophy.png         | Bin 0 -> 16463 bytes
 docs/blog/.authors.yml                        |   5 -
 docs/blog/assets/4000_stars.png               | Bin 57307 -> 0 bytes
 docs/blog/posts/roadmap-2024.md               |  91 --
 docs/community/belonging.png                  | Bin 142520 -> 0 bytes
 docs/community/index.md                       |   3 -
 docs/community/versioning.md                  |   4 +
 docs/cookbook/classification.md               | 106 ---
 docs/cookbook/extract_event_details.py        |  46 -
 docs/core_concepts.md                         |  13 +
 docs/{cookbook => examples}/atomic_caption.md |  53 +-
 .../chain_of_density.md                       |  51 +-
 .../chain_of_thought.md                       |  87 +-
 docs/examples/classification.md               |  85 ++
 .../{cookbook => examples}/dating_profiles.md |  30 +-
 .../deploy-using-bentoml.md                   |  22 +-
 .../deploy-using-cerebrium.md                 |  11 +-
 .../deploy-using-modal.md                     |  16 +-
 .../earnings-reports.md                       |  24 +-
 .../extract_event_details.md                  |   2 +-
 docs/examples/extract_event_details.py        |  54 ++
 docs/{cookbook => examples}/extraction.md     |  16 +-
 .../images/chain_of_density.png               | Bin
 .../images/coding_structure_diagram.png       | Bin
 .../images/knowledge-graph-extraction.png     | Bin
 .../images/nvidia-income.png                  | Bin
 docs/{cookbook => examples}/images/simtom.png | Bin
 .../images/trader-joes-receipt.jpg            | Bin
 docs/{cookbook => examples}/index.md          |  14 +-
 .../knowledge_graph_extraction.md             | 102 ++-
 .../models_playing_chess.md                   |  28 +-
 .../prompt_templates/chain_of_density.txt     |  27 +
 .../prompt_templates/classification.txt       |  18 +
 .../examples/prompt_templates/react_agent.txt |  27 +
 .../simtom_prospective_taking.txt             |  10 +
 .../prompt_templates/simtom_simulation.txt    |   7 +
 .../qa-with-citations.md                      | 125 +--
 docs/{cookbook => examples}/react_agent.md    |  90 +-
 docs/{cookbook => examples}/read-pdfs.md      |  79 +-
 .../receipt-digitization.md                   |  99 +--
 docs/{cookbook => examples}/simtom.md         |  52 +-
 .../structured_generation_workflow.md         |  62 +-
 docs/features/advanced/logits_processors.md   |  97 +++
 docs/features/core/generator.md               | 122 +++
 docs/features/core/output_types.md            | 205 +++++
 docs/features/index.md                        |  19 +
 docs/features/models/anthropic.md             |  98 +++
 docs/features/models/dottxt.md                |  74 ++
 docs/features/models/gemini.md                | 148 ++++
 docs/features/models/index.md                 | 107 +++
 docs/features/models/llamacpp.md              | 168 ++++
 docs/features/models/mlxlm.md                 | 189 +++++
 docs/features/models/ollama.md                | 103 +++
 docs/features/models/openai.md                | 134 +++
 docs/features/models/openai_compatible.md     |  87 ++
 docs/features/models/sglang.md                | 171 ++++
 docs/features/models/tgi.md                   | 137 +++
 docs/features/models/transformers.md          | 207 +++++
 .../models/transformers_multimodal.md         |  89 ++
 docs/features/models/vllm.md                  | 186 ++++
 docs/features/models/vllm_offline.md          | 176 ++++
 docs/features/utility/application.md          |  65 ++
 .../utility}/regex_dsl.md                     |   6 +-
 docs/features/utility/template.md             |  85 ++
 docs/{reference => guide}/chat_templating.md  |   0
 docs/guide/core_concepts.md                   |  13 +
 docs/guide/getting_started.md                 | 363 ++++++++
 docs/guide/installation.md                    |  73 ++
 docs/guide/migration.md                       | 246 ++++++
 docs/guide/selecting_an_inference_backend.md  |   5 +
 docs/index.md                                 | 162 +++-
 docs/installation.md                          |  49 --
 docs/logos/cohere.png                         | Bin 0 -> 113536 bytes
 docs/logos/vllm.png                           | Bin 0 -> 89976 bytes
 docs/quickstart.md                            | 228 -----
 docs/reference/generation/cfg.md              | 154 ----
 docs/reference/generation/choices.md          |  16 -
 .../reference/generation/creating_grammars.md |  99 ---
 docs/reference/generation/custom_fsm_ops.md   |  37 -
 docs/reference/generation/format.md           |  23 -
 docs/reference/generation/generation.md       | 216 -----
 docs/reference/generation/json.md             | 178 ----
 docs/reference/generation/regex.md            |  27 -
 .../structured_generation_explanation.md      |  54 --
 docs/reference/generation/types.md            |  83 --
 docs/reference/index.md                       |  15 -
 docs/reference/models/exllamav2.md            |  15 -
 docs/reference/models/llamacpp.md             | 230 -----
 docs/reference/models/mlxlm.md                |  93 --
 docs/reference/models/models.md               |  56 --
 docs/reference/models/openai.md               | 204 -----
 docs/reference/models/tgi.md                  |   3 -
 docs/reference/models/transformers.md         | 148 ----
 docs/reference/models/transformers_vision.md  | 115 ---
 docs/reference/models/vllm.md                 | 235 -----
 docs/reference/prompting.md                   | 395 ---------
 docs/reference/samplers.md                    | 178 ----
 docs/reference/serve/lmstudio.md              |  89 --
 docs/reference/serve/vllm.md                  |  73 --
 docs/reference/text.md                        |  86 --
 docs/stylesheets/extra.css                    | 161 +---
 docs/welcome.md                               | 183 ----
 examples/babyagi.py                           |  18 +-
 examples/beam-cloud/app.py                    |  11 +-
 examples/bentoml/service.py                   |  16 +-
 examples/cerebrium/main.py                    |  16 +-
 examples/cfg.py                               |  22 +-
 examples/dating_profile.py                    |  21 +-
 examples/llamacpp_example.py                  |   5 +-
 examples/llamacpp_processor.py                |   4 +-
 examples/math_generate_code.py                |  10 +-
 examples/meta_prompting.py                    |  48 +-
 examples/modal_example.py                     |  23 +-
 examples/parsing.py                           |   1 -
 examples/pick_odd_one_out.py                  |  47 +-
 examples/prompts/babyagi_create_task.txt      |   9 +
 examples/prompts/babyagi_perform_task.txt     |   5 +
 examples/prompts/babyagi_prioritize_task.txt  |   7 +
 examples/prompts/dating_profile.txt           |  11 +
 examples/prompts/pick_odd_one_out.txt         |   9 +
 examples/prompts/self_consistency.txt         |   6 +
 examples/react.py                             |  56 +-
 examples/sampling.ipynb                       |  48 +-
 examples/self_consistency.py                  |  23 +-
 examples/simulation_based_inference.ipynb     |  18 +-
 examples/transformers_integration.py          |  25 -
 ...gration.py => vllm_offline_integration.py} |   7 +-
 mkdocs.yml                                    | 366 ++++----
 outlines/__init__.py                          |  77 +-
 outlines/applications.py                      | 107 +++
 outlines/caching.py                           |  10 +-
 outlines/fsm/__init__.py                      |   1 +
 outlines/fsm/json_schema.py                   |  83 --
 outlines/fsm/parsing.py                       |   2 +
 outlines/fsm/types.py                         |  76 --
 outlines/generate/api.py                      | 609 -------------
 outlines/generate/cfg.py                      |  54 --
 outlines/generate/choice.py                   |  59 --
 outlines/generate/format.py                   |  48 --
 outlines/generate/fsm.py                      |  31 -
 outlines/generate/generator.py                | 312 -------
 outlines/generate/json.py                     | 123 ---
 outlines/generate/regex.py                    |  66 --
 outlines/generate/text.py                     |  50 --
 outlines/generator.py                         | 347 ++++++++
 outlines/grammars.py                          |  23 +-
 outlines/models/__init__.py                   |  43 +-
 outlines/models/anthropic.py                  | 227 +++++
 outlines/models/base.py                       | 377 ++++++++
 outlines/models/dottxt.py                     | 215 +++++
 outlines/models/gemini.py                     | 296 +++++++
 outlines/models/llamacpp.py                   | 393 ++++-----
 outlines/models/mlxlm.py                      | 340 +++-----
 outlines/models/ollama.py                     | 213 +++++
 outlines/models/openai.py                     | 483 +++++++++--
 outlines/models/sglang.py                     | 372 ++++++++
 outlines/models/tgi.py                        | 340 ++++++++
 outlines/models/tokenizer.py                  |  12 +-
 outlines/models/transformers.py               | 555 ++++++------
 outlines/models/transformers_vision.py        | 138 ---
 outlines/models/vllm.py                       | 483 +++++++----
 outlines/models/vllm_offline.py               | 206 +++++
 outlines/processors/__init__.py               |   2 +
 outlines/processors/base_logits_processor.py  | 246 +++---
 outlines/{fsm => processors}/guide.py         | 236 +++++-
 outlines/processors/structured.py             | 218 +++--
 .../processors/tensor_adapters/__init__.py    |  26 +
 outlines/processors/tensor_adapters/base.py   | 257 ++++++
 outlines/processors/tensor_adapters/jax.py    |  50 ++
 outlines/processors/tensor_adapters/mlx.py    |  60 ++
 outlines/processors/tensor_adapters/numpy.py  |  50 ++
 .../processors/tensor_adapters/tensorflow.py  |  50 ++
 outlines/processors/tensor_adapters/torch.py  |  48 ++
 outlines/release_note.md                      | 392 +++++++++
 outlines/samplers.py                          | 353 --------
 outlines/serve/__init__.py                    |   0
 outlines/serve/serve.py                       | 140 ---
 outlines/templates.py                         |  85 +-
 outlines/types/__init__.py                    |  27 +-
 outlines/types/dsl.py                         | 567 ++++++++++++-
 outlines/types/locale/__init__.py             |   2 +
 outlines/types/locale/us.py                   |   2 +
 outlines/types/utils.py                       | 206 +++++
 outlines/{ => v0_legacy}/base.py              | 141 ++-
 outlines/{ => v0_legacy}/function.py          |  75 +-
 outlines/{ => v0_legacy}/generate/__init__.py |  11 +-
 outlines/v0_legacy/generate/api.py            | 175 ++++
 outlines/v0_legacy/generate/cfg.py            |  80 ++
 outlines/v0_legacy/generate/choice.py         | 126 +++
 outlines/v0_legacy/generate/format.py         |  70 ++
 outlines/v0_legacy/generate/fsm.py            |  77 ++
 outlines/v0_legacy/generate/json.py           | 145 ++++
 outlines/v0_legacy/generate/regex.py          |  67 ++
 outlines/v0_legacy/generate/text.py           |  64 ++
 outlines/v0_legacy/models/__init__.py         |  21 +
 outlines/{ => v0_legacy}/models/exllamav2.py  | 141 +--
 outlines/v0_legacy/models/llamacpp.py         | 147 ++++
 outlines/v0_legacy/models/mlxlm.py            | 153 ++++
 outlines/v0_legacy/models/openai.py           |  64 ++
 outlines/v0_legacy/models/transformers.py     | 196 +++++
 .../v0_legacy/models/transformers_vision.py   | 107 +++
 outlines/v0_legacy/models/vllm_offline.py     | 102 +++
 outlines/v0_legacy/samplers.py                | 118 +++
 pyproject.toml                                |  77 +-
 requirements-doc.txt                          |   3 +
 scripts/gen_ref_pages.py                      |  53 ++
 tests/fsm/test_json_schema.py                 |  78 --
 tests/fsm/test_types.py                       |  22 -
 tests/generate/__init__.py                    |   0
 tests/generate/conftest.py                    |  66 --
 tests/generate/test_api.py                    |  33 -
 tests/generate/test_generate.py               | 439 ----------
 tests/generate/test_generator.py              | 497 -----------
 tests/generate/test_integration_exllamav2.py  | 363 --------
 tests/generate/test_integration_llamacpp.py   | 374 --------
 .../generate/test_integration_transformers.py | 569 -------------
 .../test_integration_transformers_vision.py   | 116 ---
 tests/generate/test_integration_vllm.py       | 243 ------
 tests/models/test_anthropic.py                | 106 +++
 tests/models/test_dottxt.py                   | 131 +++
 tests/models/test_dottxt_type_adapter.py      | 137 +++
 tests/models/test_gemini.py                   | 221 +++++
 tests/models/test_gemini_type_adapter.py      | 204 +++++
 tests/models/test_llamacpp.py                 | 134 +++
 tests/models/test_llamacpp_tokenizer.py       | 133 +++
 tests/models/test_llamacpp_type_adapter.py    |  41 +
 tests/models/test_mlxlm.py                    | 133 +--
 tests/models/test_ollama.py                   | 101 +++
 tests/models/test_ollama_type_adapter.py      | 131 +++
 tests/models/test_openai.py                   | 292 ++++---
 tests/models/test_openai_type_adapter.py      | 195 +++++
 tests/models/test_sglang.py                   | 272 ++++++
 tests/models/test_sglang_type_adapter.py      | 145 ++++
 tests/models/test_tgi.py                      | 180 ++++
 tests/models/test_tgi_model_adapter.py        |  93 ++
 tests/models/test_transformers.py             | 283 +++---
 tests/models/test_transformers_multimodal.py  | 194 +++++
 ...st_transformers_multimodal_type_adapter.py |  48 ++
 tests/models/test_transformers_tokenizer.py   | 142 ++++
 .../models/test_transformers_type_adapter.py  |  50 ++
 tests/models/test_vllm.py                     | 254 ++++++
 tests/models/test_vllm_offline.py             | 117 +++
 .../models/test_vllm_offline_type_adapter.py  |  84 ++
 tests/models/test_vllm_type_adapter.py        | 121 +++
 tests/processors/test_base_processor.py       | 148 ++--
 tests/{fsm => processors}/test_cfg_guide.py   |  15 +-
 tests/{fsm => processors}/test_guide.py       |  82 +-
 tests/processors/test_structured.py           | 135 +++
 tests/processors/test_tensor_adapters.py      | 276 ++++++
 tests/test_applications.py                    |  93 ++
 tests/test_cache.py                           |  31 +
 tests/test_generator.py                       | 255 ++++++
 tests/test_grammars.py                        |   2 +-
 tests/test_samplers.py                        | 277 ------
 tests/test_templates.py                       | 122 ++-
 tests/test_utils/mock_openai_client.py        | 117 +++
 tests/test_utils/mock_tgi_client.py           |  63 ++
 tests/test_utils/utils.py                     |  23 +
 tests/types/test_custom_types.py              |   1 -
 tests/types/test_dsl.py                       | 508 ++++++++++-
 tests/types/test_types_utils.py               | 409 +++++++++
 tests/v0_legacy/generate/test_api_legacy.py   | 186 ++++
 tests/v0_legacy/generate/test_cfg_legacy.py   | 111 +++
 .../v0_legacy/generate/test_choice_legacy.py  |  97 +++
 .../v0_legacy/generate/test_format_legacy.py  |  93 ++
 tests/v0_legacy/generate/test_fsm_legacy.py   | 104 +++
 tests/v0_legacy/generate/test_json_legacy.py  | 200 +++++
 tests/v0_legacy/generate/test_regex_legacy.py |  90 ++
 tests/v0_legacy/generate/test_text_legacy.py  |  83 ++
 .../v0_legacy/models/test_exllamav2_legacy.py |  76 ++
 .../v0_legacy/models/test_llamacpp_legacy.py  | 169 ++++
 tests/v0_legacy/models/test_mlxlm_legacy.py   | 193 +++++
 tests/v0_legacy/models/test_openai_legacy.py  | 159 ++++
 .../models/test_transformers_legacy.py        | 172 ++++
 .../models/test_transformers_vision_legacy.py | 110 +++
 .../models/test_vllm_offline_legacy.py        | 181 ++++
 .../test_base_legacy.py}                      |   2 +-
 .../test_function_legacy.py}                  |  85 +-
 tests/v0_legacy/test_samplers_legacy.py       |  60 ++
 uv.lock                                       |  28 -
 300 files changed, 21850 insertions(+), 12037 deletions(-)
 create mode 100644 .github/workflows/tests_nightly.yml
 delete mode 100644 Dockerfile
 delete mode 100644 docs/api/guide.md
 delete mode 100644 docs/api/json_schema.md
 delete mode 100644 docs/api/models.md
 delete mode 100644 docs/api/parsing.md
 delete mode 100644 docs/api/regex.md
 delete mode 100644 docs/api/samplers.md
 delete mode 100644 docs/api/templates.md
 rename docs/{api => api_reference}/index.md (100%)
 create mode 100644 docs/assets/images/install.png
 create mode 100644 docs/assets/images/readme-dark.png
 create mode 100644 docs/assets/images/readme-light.png
 create mode 100644 docs/assets/images/use_philosophy.png
 delete mode 100644 docs/blog/.authors.yml
 delete mode 100644 docs/blog/assets/4000_stars.png
 delete mode 100644 docs/blog/posts/roadmap-2024.md
 delete mode 100644 docs/community/belonging.png
 delete mode 100644 docs/cookbook/classification.md
 delete mode 100644 docs/cookbook/extract_event_details.py
 create mode 100644 docs/core_concepts.md
 rename docs/{cookbook => examples}/atomic_caption.md (75%)
 rename docs/{cookbook => examples}/chain_of_density.md (74%)
 rename docs/{cookbook => examples}/chain_of_thought.md (63%)
 create mode 100644 docs/examples/classification.md
 rename docs/{cookbook => examples}/dating_profiles.md (92%)
 rename docs/{cookbook => examples}/deploy-using-bentoml.md (91%)
 rename docs/{cookbook => examples}/deploy-using-cerebrium.md (90%)
 rename docs/{cookbook => examples}/deploy-using-modal.md (93%)
 rename docs/{cookbook => examples}/earnings-reports.md (94%)
 rename docs/{cookbook => examples}/extract_event_details.md (94%)
 create mode 100644 docs/examples/extract_event_details.py
 rename docs/{cookbook => examples}/extraction.md (85%)
 rename docs/{cookbook => examples}/images/chain_of_density.png (100%)
 rename docs/{cookbook => examples}/images/coding_structure_diagram.png (100%)
 rename docs/{cookbook => examples}/images/knowledge-graph-extraction.png (100%)
 rename docs/{cookbook => examples}/images/nvidia-income.png (100%)
 rename docs/{cookbook => examples}/images/simtom.png (100%)
 rename docs/{cookbook => examples}/images/trader-joes-receipt.jpg (100%)
 rename docs/{cookbook => examples}/index.md (82%)
 rename docs/{cookbook => examples}/knowledge_graph_extraction.md (60%)
 rename docs/{cookbook => examples}/models_playing_chess.md (77%)
 create mode 100644 docs/examples/prompt_templates/chain_of_density.txt
 create mode 100644 docs/examples/prompt_templates/classification.txt
 create mode 100644 docs/examples/prompt_templates/react_agent.txt
 create mode 100644 docs/examples/prompt_templates/simtom_prospective_taking.txt
 create mode 100644 docs/examples/prompt_templates/simtom_simulation.txt
 rename docs/{cookbook => examples}/qa-with-citations.md (68%)
 rename docs/{cookbook => examples}/react_agent.md (71%)
 rename docs/{cookbook => examples}/read-pdfs.md (79%)
 rename docs/{cookbook => examples}/receipt-digitization.md (71%)
 rename docs/{cookbook => examples}/simtom.md (68%)
 rename docs/{cookbook => examples}/structured_generation_workflow.md (79%)
 create mode 100644 docs/features/advanced/logits_processors.md
 create mode 100644 docs/features/core/generator.md
 create mode 100644 docs/features/core/output_types.md
 create mode 100644 docs/features/index.md
 create mode 100644 docs/features/models/anthropic.md
 create mode 100644 docs/features/models/dottxt.md
 create mode 100644 docs/features/models/gemini.md
 create mode 100644 docs/features/models/index.md
 create mode 100644 docs/features/models/llamacpp.md
 create mode 100644 docs/features/models/mlxlm.md
 create mode 100644 docs/features/models/ollama.md
 create mode 100644 docs/features/models/openai.md
 create mode 100644 docs/features/models/openai_compatible.md
 create mode 100644 docs/features/models/sglang.md
 create mode 100644 docs/features/models/tgi.md
 create mode 100644 docs/features/models/transformers.md
 create mode 100644 docs/features/models/transformers_multimodal.md
 create mode 100644 docs/features/models/vllm.md
 create mode 100644 docs/features/models/vllm_offline.md
 create mode 100644 docs/features/utility/application.md
 rename docs/{reference => features/utility}/regex_dsl.md (99%)
 create mode 100644 docs/features/utility/template.md
 rename docs/{reference => guide}/chat_templating.md (100%)
 create mode 100644 docs/guide/core_concepts.md
 create mode 100644 docs/guide/getting_started.md
 create mode 100644 docs/guide/installation.md
 create mode 100644 docs/guide/migration.md
 create mode 100644 docs/guide/selecting_an_inference_backend.md
 delete mode 100644 docs/installation.md
 create mode 100644 docs/logos/cohere.png
 create mode 100644 docs/logos/vllm.png
 delete mode 100644 docs/quickstart.md
 delete mode 100644 docs/reference/generation/cfg.md
 delete mode 100644 docs/reference/generation/choices.md
 delete mode 100644 docs/reference/generation/creating_grammars.md
 delete mode 100644 docs/reference/generation/custom_fsm_ops.md
 delete mode 100644 docs/reference/generation/format.md
 delete mode 100644 docs/reference/generation/generation.md
 delete mode 100644 docs/reference/generation/json.md
 delete mode 100644 docs/reference/generation/regex.md
 delete mode 100644 docs/reference/generation/structured_generation_explanation.md
 delete mode 100644 docs/reference/generation/types.md
 delete mode 100644 docs/reference/index.md
 delete mode 100644 docs/reference/models/exllamav2.md
 delete mode 100644 docs/reference/models/llamacpp.md
 delete mode 100644 docs/reference/models/mlxlm.md
 delete mode 100644 docs/reference/models/models.md
 delete mode 100644 docs/reference/models/openai.md
 delete mode 100644 docs/reference/models/tgi.md
 delete mode 100644 docs/reference/models/transformers.md
 delete mode 100644 docs/reference/models/transformers_vision.md
 delete mode 100644 docs/reference/models/vllm.md
 delete mode 100644 docs/reference/prompting.md
 delete mode 100644 docs/reference/samplers.md
 delete mode 100644 docs/reference/serve/lmstudio.md
 delete mode 100644 docs/reference/serve/vllm.md
 delete mode 100644 docs/reference/text.md
 delete mode 100644 docs/welcome.md
 create mode 100644 examples/prompts/babyagi_create_task.txt
 create mode 100644 examples/prompts/babyagi_perform_task.txt
 create mode 100644 examples/prompts/babyagi_prioritize_task.txt
 create mode 100644 examples/prompts/dating_profile.txt
 create mode 100644 examples/prompts/pick_odd_one_out.txt
 create mode 100644 examples/prompts/self_consistency.txt
 delete mode 100644 examples/transformers_integration.py
 rename examples/{vllm_integration.py => vllm_offline_integration.py} (80%)
 create mode 100644 outlines/applications.py
 delete mode 100644 outlines/fsm/json_schema.py
 delete mode 100644 outlines/fsm/types.py
 delete mode 100644 outlines/generate/api.py
 delete mode 100644 outlines/generate/cfg.py
 delete mode 100644 outlines/generate/choice.py
 delete mode 100644 outlines/generate/format.py
 delete mode 100644 outlines/generate/fsm.py
 delete mode 100644 outlines/generate/generator.py
 delete mode 100644 outlines/generate/json.py
 delete mode 100644 outlines/generate/regex.py
 delete mode 100644 outlines/generate/text.py
 create mode 100644 outlines/generator.py
 create mode 100644 outlines/models/anthropic.py
 create mode 100644 outlines/models/base.py
 create mode 100644 outlines/models/dottxt.py
 create mode 100644 outlines/models/gemini.py
 create mode 100644 outlines/models/ollama.py
 create mode 100644 outlines/models/sglang.py
 create mode 100644 outlines/models/tgi.py
 delete mode 100644 outlines/models/transformers_vision.py
 create mode 100644 outlines/models/vllm_offline.py
 rename outlines/{fsm => processors}/guide.py (58%)
 create mode 100644 outlines/processors/tensor_adapters/__init__.py
 create mode 100644 outlines/processors/tensor_adapters/base.py
 create mode 100644 outlines/processors/tensor_adapters/jax.py
 create mode 100644 outlines/processors/tensor_adapters/mlx.py
 create mode 100644 outlines/processors/tensor_adapters/numpy.py
 create mode 100644 outlines/processors/tensor_adapters/tensorflow.py
 create mode 100644 outlines/processors/tensor_adapters/torch.py
 create mode 100644 outlines/release_note.md
 delete mode 100644 outlines/samplers.py
 delete mode 100644 outlines/serve/__init__.py
 delete mode 100644 outlines/serve/serve.py
 create mode 100644 outlines/types/utils.py
 rename outlines/{ => v0_legacy}/base.py (67%)
 rename outlines/{ => v0_legacy}/function.py (58%)
 rename outlines/{ => v0_legacy}/generate/__init__.py (62%)
 create mode 100644 outlines/v0_legacy/generate/api.py
 create mode 100644 outlines/v0_legacy/generate/cfg.py
 create mode 100644 outlines/v0_legacy/generate/choice.py
 create mode 100644 outlines/v0_legacy/generate/format.py
 create mode 100644 outlines/v0_legacy/generate/fsm.py
 create mode 100644 outlines/v0_legacy/generate/json.py
 create mode 100644 outlines/v0_legacy/generate/regex.py
 create mode 100644 outlines/v0_legacy/generate/text.py
 create mode 100644 outlines/v0_legacy/models/__init__.py
 rename outlines/{ => v0_legacy}/models/exllamav2.py (79%)
 create mode 100644 outlines/v0_legacy/models/llamacpp.py
 create mode 100644 outlines/v0_legacy/models/mlxlm.py
 create mode 100644 outlines/v0_legacy/models/openai.py
 create mode 100644 outlines/v0_legacy/models/transformers.py
 create mode 100644 outlines/v0_legacy/models/transformers_vision.py
 create mode 100644 outlines/v0_legacy/models/vllm_offline.py
 create mode 100644 outlines/v0_legacy/samplers.py
 create mode 100644 scripts/gen_ref_pages.py
 delete mode 100644 tests/fsm/test_json_schema.py
 delete mode 100644 tests/fsm/test_types.py
 delete mode 100644 tests/generate/__init__.py
 delete mode 100644 tests/generate/conftest.py
 delete mode 100644 tests/generate/test_api.py
 delete mode 100644 tests/generate/test_generate.py
 delete mode 100644 tests/generate/test_generator.py
 delete mode 100644 tests/generate/test_integration_exllamav2.py
 delete mode 100644 tests/generate/test_integration_llamacpp.py
 delete mode 100644 tests/generate/test_integration_transformers.py
 delete mode 100644 tests/generate/test_integration_transformers_vision.py
 delete mode 100644 tests/generate/test_integration_vllm.py
 create mode 100644 tests/models/test_anthropic.py
 create mode 100644 tests/models/test_dottxt.py
 create mode 100644 tests/models/test_dottxt_type_adapter.py
 create mode 100644 tests/models/test_gemini.py
 create mode 100644 tests/models/test_gemini_type_adapter.py
 create mode 100644 tests/models/test_llamacpp.py
 create mode 100644 tests/models/test_llamacpp_tokenizer.py
 create mode 100644 tests/models/test_llamacpp_type_adapter.py
 create mode 100644 tests/models/test_ollama.py
 create mode 100644 tests/models/test_ollama_type_adapter.py
 create mode 100644 tests/models/test_openai_type_adapter.py
 create mode 100644 tests/models/test_sglang.py
 create mode 100644 tests/models/test_sglang_type_adapter.py
 create mode 100644 tests/models/test_tgi.py
 create mode 100644 tests/models/test_tgi_model_adapter.py
 create mode 100644 tests/models/test_transformers_multimodal.py
 create mode 100644 tests/models/test_transformers_multimodal_type_adapter.py
 create mode 100644 tests/models/test_transformers_tokenizer.py
 create mode 100644 tests/models/test_transformers_type_adapter.py
 create mode 100644 tests/models/test_vllm.py
 create mode 100644 tests/models/test_vllm_offline.py
 create mode 100644 tests/models/test_vllm_offline_type_adapter.py
 create mode 100644 tests/models/test_vllm_type_adapter.py
 rename tests/{fsm => processors}/test_cfg_guide.py (97%)
 rename tests/{fsm => processors}/test_guide.py (85%)
 create mode 100644 tests/processors/test_structured.py
 create mode 100644 tests/processors/test_tensor_adapters.py
 create mode 100644 tests/test_applications.py
 create mode 100644 tests/test_generator.py
 delete mode 100644 tests/test_samplers.py
 create mode 100644 tests/test_utils/mock_openai_client.py
 create mode 100644 tests/test_utils/mock_tgi_client.py
 create mode 100644 tests/test_utils/utils.py
 create mode 100644 tests/types/test_types_utils.py
 create mode 100644 tests/v0_legacy/generate/test_api_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_cfg_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_choice_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_format_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_fsm_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_json_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_regex_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_text_legacy.py
 create mode 100644 tests/v0_legacy/models/test_exllamav2_legacy.py
 create mode 100644 tests/v0_legacy/models/test_llamacpp_legacy.py
 create mode 100644 tests/v0_legacy/models/test_mlxlm_legacy.py
 create mode 100644 tests/v0_legacy/models/test_openai_legacy.py
 create mode 100644 tests/v0_legacy/models/test_transformers_legacy.py
 create mode 100644 tests/v0_legacy/models/test_transformers_vision_legacy.py
 create mode 100644 tests/v0_legacy/models/test_vllm_offline_legacy.py
 rename tests/{test_base.py => v0_legacy/test_base_legacy.py} (99%)
 rename tests/{test_function.py => v0_legacy/test_function_legacy.py} (59%)
 create mode 100644 tests/v0_legacy/test_samplers_legacy.py

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 46d3d5ff6..c6e37d684 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -2,7 +2,7 @@ name: Tests
 
 on:
   pull_request:
-    branches: [main]
+    branches: [main,v1.0]
   push:
     branches: [main]
 
@@ -29,6 +29,11 @@ jobs:
       uses: actions/setup-python@v4
       with:
         python-version: ${{ matrix.python-version }}
+    - name: Install Ollama
+      run: |
+        curl -fsSL https://ollama.com/install.sh | sh
+        ollama --version
+        ollama pull tinyllama
     - name: Set up test environment
       run: |
         python -m pip install --upgrade pip
@@ -46,7 +51,7 @@ jobs:
         echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
-        uv run pytest -x --cov=outlines
+        uv run pytest -x --cov=outlines -m 'not api_call'
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
     - name: Upload coverage data
@@ -82,16 +87,30 @@ jobs:
         with:
           name: coverage-data
 
-      - name: Fetch main for coverage diff
+      - name: Determine base branch for comparison
+        id: base-branch
+        run: |
+          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
+            # For PRs, use the remote tracking branch
+            COMPARE_BRANCH="origin/${{ github.base_ref }}"
+            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
+          else
+            # For push events, compare against the parent commit
+            COMPARE_BRANCH="${{ github.event.before }}"
+            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
+          fi
+          echo "Using $COMPARE_BRANCH for coverage comparison"
+
+      - name: Fetch base branch for coverage diff
         run: |
-          git fetch --no-tags --prune origin main
+          git fetch --no-tags --prune origin ${COMPARE_BRANCH#origin/}
 
       - name: Combine coverage & fail if it's <100%.
         run: |
           python -m coverage combine
           python -m coverage html --skip-covered --skip-empty
           python -m coverage xml
-          diff-cover coverage.xml --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
+          diff-cover coverage.xml --compare-branch=$COMPARE_BRANCH --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
 
       - name: Upload HTML report if check failed.
         uses: actions/upload-artifact@v4
diff --git a/.github/workflows/tests_nightly.yml b/.github/workflows/tests_nightly.yml
new file mode 100644
index 000000000..c34a707fa
--- /dev/null
+++ b/.github/workflows/tests_nightly.yml
@@ -0,0 +1,60 @@
+name: Nightly Tests
+
+# Run every day at 6:00 AM UTC
+on:
+  schedule:
+    - cron: "0 6 * * *"
+
+jobs:
+  tests:
+    name: Run the tests
+    runs-on: ubuntu-latest
+    env:
+      OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+      ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
+      GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
+      DOTTXT_API_KEY: ${{ secrets.DOTTXT_API_KEY }}
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.10"]
+
+    steps:
+    - uses: actions/checkout@v3
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v4
+      with:
+        python-version: ${{ matrix.python-version }}
+    - name: Install Ollama
+      run: |
+        curl -fsSL https://ollama.com/install.sh | sh
+        ollama --version
+        ollama pull tinyllama
+    - name: Set up test environment
+      run: |
+        python -m pip install --upgrade pip
+        pip install uv
+        uv venv
+        uv pip install -e .[test]
+    - name: Create matrix id
+      id: matrix-id
+      env:
+        MATRIX_CONTEXT: ${{ toJson(matrix) }}
+      run: |
+        echo $MATRIX_CONTEXT
+        export MATRIX_ID=`echo $MATRIX_CONTEXT | md5sum | cut -c 1-32`
+        echo $MATRIX_ID
+        echo "::set-output name=id::$MATRIX_ID"
+    - name: Run tests
+      run: |
+        uv run pytest -x --cov=outlines
+      env:
+        COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
+    - name: Slack notification on failure
+      if: failure()
+      uses: 8398a7/action-slack@v3
+      with:
+        status: ${{ job.status }}
+        fields: commit,author,ref
+      env:
+        SLACK_WEBHOOK_URL: ${{ secrets.SLACK_TESTS_ERRORS_WEBHOOK_URL }}
diff --git a/Dockerfile b/Dockerfile
deleted file mode 100644
index 7e6f2e68a..000000000
--- a/Dockerfile
+++ /dev/null
@@ -1,23 +0,0 @@
-### Build stage
-FROM python:3.10 AS builder
-
-WORKDIR /outlines
-
-RUN pip install --upgrade pip
-
-# Copy necessary build components
-COPY pyproject.toml .
-COPY outlines ./outlines
-
-# Install outlines and outlines[serve]
-# .git required by setuptools-scm
-RUN --mount=source=.git,target=.git,type=bind \
-    pip install --no-cache-dir .[serve]
-
-### Runtime stage
-FROM python:3.10
-WORKDIR /outlines
-COPY --from=builder /outlines /outlines
-
-# https://dottxt-ai.github.io/outlines/reference/vllm/
-ENTRYPOINT ["python3", "-m", "outlines.serve.serve"]
diff --git a/README.md b/README.md
index b20fc821c..0d85aebf1 100644
--- a/README.md
+++ b/README.md
@@ -1,449 +1,669 @@
 <div align="center" style="margin-bottom: 1em;">
 
-<img src="./docs/assets/images/logo-light-mode.svg#gh-light-mode-only" alt="Outlines Logo" width=500></img>
-<img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=500></img>
+<img src="./docs/assets/images/logo-light-mode.svg#gh-light-mode-only" alt="Outlines Logo" width=300></img>
+<img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=300></img>
 
 
- 🗒️ *Make LLMs speak the language of every application.* 🗒️
+ 🗒️ *Structured outputs for LLMs* 🗒️
 
-Made with ❤👷️ by the team at [.txt](https://dottxt.co).
+Made with ❤👷️ by the team at [.txt](https://dottxt.co)
+<br>Trusted by NVIDIA, Cohere, HuggingFace, vLLM, etc.
 
-[![Documentation][documentation-badge]][documentation]
-[![Contributors][contributors-badge]][contributors]
+<!-- Project Badges -->
+[![PyPI Version][pypi-version-badge]][pypi]
 [![Downloads][downloads-badge]][pypistats]
-[![Discord][discord-badge]][discord]
-
-[Youtube channel][youtube-dottxt] | [.txt blog][blog-dottxt] | [Twitter][dottxt-twitter]
+[![Stars][stars-badge]][stars]
 
+<!-- Community Badges -->
+[![Discord][discord-badge]][discord]
+[![Blog][dottxt-blog-badge]][dottxt-blog]
+[![Twitter][twitter-badge]][twitter]
 
 </div>
 
+Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales).
 
-``` bash
-pip install outlines
-```
+## Table of Contents
 
-First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/latest/welcome/)/
+- [Why Outlines?](#why-outlines)
+- [Quickstart](#quickstart)
+- [Real-World Examples](#real-world-examples)
+  - [🙋‍♂️ Customer Support Triage](#customer-support-triage)
+  - [📦 E-commerce Product Categorization](#e-commerce-product-categorization)
+  - [💊 Pull Medication Data from Unstructured Notes](#pull-medication-data-from-unstructured-notes)
+  - [📊 Parse Event Details with Incomplete Data](#parse-event-details-with-incomplete-data)
+  - [🗂️ Categorize Documents into Predefined Types](#categorize-documents-into-predefined-types)
+  - [📅 Schedule a Meeting with Function Calling](#schedule-a-meeting-with-function-calling)
+  - [📝 Dynamically Generate Prompts with Re-usable Templates](#dynamically-generate-prompts-with-re-usable-templates)
+- [They Use Outlines](#they-use-outlines)
+- [Model Integrations](#model-integrations)
+- [Core Features](#core-features)
+- [Other Features](#other-features)
+- [About .txt](#about-txt)
+- [Community](#community)
 
-Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales) with us.
+<div align="center"><img src="./docs/assets/images/install.png" width=300></img></div>
 
-## Features
+## Why Outlines?
 
-- 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
-- 🔥 Fast [JSON generation](#efficient-json-generation-following-a-pydantic-model) following a JSON schema or a Pydantic model
-- 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
-- 📝 Generate text that follows a [regex](#efficient-regex-structured-generation) or a [context-free grammar](#using-context-free-grammars-to-guide-generation)
-- 🖍️ Simple and powerful prompting primitives based on the [Jinja templating engine](https://jinja.palletsprojects.com/)
-- 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/latest/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
+LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
 
+Outlines guarantees structured outputs during generation — directly from any LLM.
 
-Outlines  has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!
+- **Works with any model** - Same code runs across OpenAI, Ollama, vLLM, and more
+- **Simple integration** - Just pass your desired output type: `model(prompt, output_type)`
+- **Guaranteed valid structure** - No more parsing headaches or broken JSON
+- **Provider independence** - Switch models without changing code
 
-## Why should I use structured generation?
 
-* It doesn't add any overhead during inference (cost-free)
-* It allows Open Source models to beat closed source models ([Mistral](https://x.com/dottxtai/status/1797692104023363765), [GPT-4](https://x.com/dottxtai/status/1798443290913853770))
-* [It speeds up inference](http://blog.dottxt.co/coalescence.html)
-* [It improves the performance of base models (GSM8K)](http://blog.dottxt.co/performance-gsm8k.html)
-* [It improves the performance of finetuned models (CoNNL)](https://predibase.com/blog/lorax-outlines-better-json-extraction-with-structured-generation-and-lora)
-* [It improves model efficiency (less examples needed)](https://huggingface.co/blog/evaluation-structured-outputs)
+### The Outlines Philosophy
 
-## .txt company
+<div align="center"><img src="./docs/assets/images/use_philosophy.png" width=300></img></div>
 
-<div align="center">
-<img src="./docs/assets/images/dottxt-light.svg#gh-light-mode-only" alt="dottxt logo" width=100></img>
-<img src="./docs/assets/images/dottxt-dark.svg#gh-dark-mode-only" alt="dottxt logo" width=100></img>
-</div>
-
-We started a company to keep pushing the boundaries of structured generation. Learn more about [.txt](https://dottxt.ai), and  [license our commercial product](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF) if you need SOTA performance and quality ✨
-
-## Structured generation
-
-The first step towards reliability of systems that include large language models
-is to ensure that there is a well-defined interface between their output and
-user-defined code. **Outlines** provides ways to control the generation of
-language models to make their output more predictable.
+Outlines follows a simple pattern that mirrors Python's own type system. Simply specify the desired output type, and Outlines will ensure your data matches that structure exactly:
 
-The following methods of structured generation are supported:
+- For a yes/no response, use `Literal["Yes", "No"]`
+- For numerical values, use `int`
+- For complex objects, define a structure with a [Pydantic model](https://docs.pydantic.dev/latest/)
 
-- [Multiple choices](#multiple-choices)
-- [Type constraints](#type-constraint)
-- [Efficient regex-structured generation](#efficient-regex-structured-generation)
-- [Efficient JSON generation following a Pydantic model](#efficient-json-generation-following-a-pydantic-model)
-- [Using context-free grammars to guide generation](#using-context-free-grammars-to-guide-generation)
-- [Open functions](#open-functions)
+## Quickstart
 
-### Chat template tokens
+Getting started with outlines is simple:
 
-Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself. Chat template tokens are not needed for base models.
+### 1. Install outlines
 
-Please see [the documentation](https://dottxt-ai.github.io/outlines/latest/reference/chat_templating) on chat templating for more.
-
-### Multiple choices
+``` shell
+pip install outlines
+```
 
-You can reduce the completion to a choice between multiple possibilities:
+### 2. Connect to your preferred model
 
 ``` python
 import outlines
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-model_name = "HuggingFaceTB/SmolLM2-360M-Instruct"
-model = outlines.models.transformers(model_name)
 
-# You must apply the chat template tokens to the prompt!
-# See below for an example.
-prompt = """
-<|im_start|>system
-You extract information from text.
-<|im_end|>
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
+```
 
-<|im_start|>user
-What food does the following text describe?
+### 3. Start with simple structured outputs
+
+``` python
+from typing import Literal
+from pydantic import BaseModel
 
-Text: I really really really want pizza.
-<|im_end|>
-<|im_start|>assistant
-"""
 
-generator = outlines.generate.choice(model, ["Pizza", "Pasta", "Salad", "Dessert"])
-answer = generator(prompt)
+# Simple classification
+sentiment = model(
+    "Analyze: 'This product completely changed my life!'",
+    Literal["Positive", "Negative", "Neutral"]
+)
+print(sentiment)  # "Positive"
 
-# Likely answer: Pizza
+# Extract specific types
+temperature = model("What's the boiling point of water in Celsius?", int)
+print(temperature)  # 100
 ```
 
-You can also pass in choices with an `Enum`:
+### 4. Create complex structures
 
-````python
+``` python
+from pydantic import BaseModel
 from enum import Enum
 
-class Food(str, Enum):
-    pizza = "Pizza"
-    pasta = "Pasta"
-    salad = "Salad"
-    dessert = "Dessert"
+class Rating(Enum):
+    poor = 1
+    fair = 2
+    good = 3
+    excellent = 4
+
+class ProductReview(BaseModel):
+    rating: Rating
+    pros: list[str]
+    cons: list[str]
+    summary: str
+
+review = model(
+    "Review: The XPS 13 has great battery life and a stunning display, but it runs hot and the webcam is poor quality.",
+    ProductReview,
+    max_new_tokens=200,
+)
 
-generator = outlines.generate.choice(model, Food)
-answer = generator(prompt)
-# Likely answer: Pizza
-````
+review = ProductReview.model_validate_json(review)
+print(f"Rating: {review.rating.name}")  # "Rating: good"
+print(f"Pros: {review.pros}")           # "Pros: ['great battery life', 'stunning display']"
+print(f"Summary: {review.summary}")     # "Summary: Good laptop with great display but thermal issues"t(result)
+```
 
-### Type constraints
+## Real-world examples
 
-You can instruct the model to only return integers or floats:
+Here are production-ready examples showing how Outlines solves common problems:
 
+<details id="customer-support-triage"><summary><b>🙋‍♂️ Customer Support Triage</b>
+<br>This example shows how to convert a free-form customer email into a structured service ticket. By parsing attributes like priority, category, and escalation flags, the code enables automated routing and handling of support issues.
+</summary>
 
 ``` python
 import outlines
+from enum import Enum
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from typing import List
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
 
-prompt = "<s>result of 9 + 9 = 18</s><s>result of 1 + 2 = "
-answer = outlines.generate.format(model, int)(prompt)
-print(answer)
-# 3
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 
-prompt = "sqrt(2)="
-generator = outlines.generate.format(model, float)
-answer = generator(prompt, max_tokens=10)
-print(answer)
-# 1.41421356
-```
 
-### Efficient regex-structured generation
+def alert_manager(ticket):
+    print("Alert!", ticket)
 
-Outlines also comes with fast regex-structured generation. In fact, the `choice` and
-`format` functions above all use regex-structured generation under the
-hood:
 
-``` python
-import outlines
+class TicketPriority(str, Enum):
+    low = "low"
+    medium = "medium"
+    high = "high"
+    urgent = "urgent"
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+class ServiceTicket(BaseModel):
+    priority: TicketPriority
+    category: str
+    requires_manager: bool
+    summary: str
+    action_items: List[str]
 
-prompt = """
-<|im_start|>system You are a helpful assistant.
-<|im_end|>
 
+customer_email = """
+Subject: URGENT - Cannot access my account after payment
+
+I paid for the premium plan 3 hours ago and still can't access any features.
+I've tried logging out and back in multiple times. This is unacceptable as I
+have a client presentation in an hour and need the analytics dashboard.
+Please fix this immediately or refund my payment.
+"""
+
+prompt = f"""
 <|im_start|>user
-What is an IP address of the Google DNS servers?
+Analyze this customer email:
+
+{customer_email}
 <|im_end|>
 <|im_start|>assistant
-The IP address of a Google DNS server is
-
 """
 
-generator = outlines.generate.text(model)
-unstructured = generator(prompt, max_tokens=30)
+ticket = model(
+    prompt,
+    ServiceTicket,
+    max_new_tokens=500
+)
+
+# Use structured data to route the ticket
+ticket = ServiceTicket.model_validate_json(ticket)
+if ticket.priority == "urgent" or ticket.requires_manager:
+    alert_manager(ticket)
+```
+</details>
+
+<details id="e-commerce-product-categorization"><summary><b>📦 E-commerce product categorization</b>
+<br>This use case demonstrates how outlines can transform product descriptions into structured categorization data (e.g., main category, sub-category, and attributes) to streamline tasks such as inventory management. Each product description is processed automatically, reducing manual categorization overhead.
+</summary>
+
+```python
+import outlines
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from typing import List, Optional
 
-generator = outlines.generate.regex(
-    model,
-    r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)",
-    sampler=outlines.samplers.greedy(),
+
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
 )
-structured = generator(prompt, max_tokens=30)
 
-print(unstructured)
-# 8.8.8.8
-#
-# <|im_end|>
 
-print(structured)
-# 8.8.8.8
-```
+def update_inventory(product, category, sub_category):
+    print(f"Updated {product.split(',')[0]} in category {category}/{sub_category}")
 
-Unlike other libraries, regex-structured generation in Outlines is almost as fast
-as non-structured generation.
 
-### Efficient JSON generation following a Pydantic model
+class ProductCategory(BaseModel):
+    main_category: str
+    sub_category: str
+    attributes: List[str]
+    brand_match: Optional[str]
 
-Outlines users can guide the generation process so the output is *guaranteed* to follow a [JSON schema](https://json-schema.org/) or [Pydantic model](https://docs.pydantic.dev/latest/):
+# Process product descriptions in batches
+product_descriptions = [
+    "Apple iPhone 15 Pro Max 256GB Titanium, 6.7-inch Super Retina XDR display with ProMotion",
+    "Organic Cotton T-Shirt, Men's Medium, Navy Blue, 100% Sustainable Materials",
+    "KitchenAid Stand Mixer, 5 Quart, Red, 10-Speed Settings with Dough Hook Attachment"
+]
+
+template = outlines.Template.from_string("""
+<|im_start|>user
+Categorize this product:
+
+{{ description }}
+<|im_end|>
+<|im_start|>assistant
+""")
+
+# Get structured categorization for all products
+categories = model(
+    [template(description=desc) for desc in product_descriptions],
+    ProductCategory,
+    max_new_tokens=200
+)
+
+# Use categorization for inventory management
+categories = [
+    ProductCategory.model_validate_json(category) for category in categories
+]
+for product, category in zip(product_descriptions, categories):
+    update_inventory(product, category.main_category, category.sub_category)
+```
+</details>
+
+<details id="parse-event-details-with-incomplete-data"><summary><b>📊 Parse event details with incomplete data</b>
+<br>This example uses outlines to parse event descriptions into structured information (like event name, date, location, type, and topics), even handling cases where the data is incomplete. It leverages union types to return either structured event data or a fallback “I don’t know” answer, ensuring robust extraction in varying scenarios.
+</summary>
 
 ```python
+import outlines
+from typing import Union, List, Literal
+from pydantic import BaseModel
 from enum import Enum
-from pydantic import BaseModel, constr
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-import outlines
 
-class Weapon(str, Enum):
-    sword = "sword"
-    axe = "axe"
-    mace = "mace"
-    spear = "spear"
-    bow = "bow"
-    crossbow = "crossbow"
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 
+class EventType(str, Enum):
+    conference = "conference"
+    webinar = "webinar"
+    workshop = "workshop"
+    meetup = "meetup"
+    other = "other"
+
+
+class EventInfo(BaseModel):
+    """Structured information about a tech event"""
+    name: str
+    date: str
+    location: str
+    event_type: EventType
+    topics: List[str]
+    registration_required: bool
+
+# Create a union type that can either be a structured EventInfo or "I don't know"
+EventResponse = Union[EventInfo, Literal["I don't know"]]
+
+# Sample event descriptions
+event_descriptions = [
+    # Complete information
+    """
+    Join us for DevCon 2023, the premier developer conference happening on November 15-17, 2023
+    at the San Francisco Convention Center. Topics include AI/ML, cloud infrastructure, and web3.
+    Registration is required.
+    """,
+
+    # Insufficient information
+    """
+    Tech event next week. More details coming soon!
+    """
+]
 
-class Armor(str, Enum):
-    leather = "leather"
-    chainmail = "chainmail"
-    plate = "plate"
+# Process events
+results = []
+for description in event_descriptions:
+    prompt = f"""
+<|im_start>system
+You are a helpful assistant
+<|im_end|>
+<|im_start>user
+Extract structured information about this tech event:
 
+{description}
 
-class Character(BaseModel):
-    name: constr(max_length=10)
-    age: int
-    armor: Armor
-    weapon: Weapon
-    strength: int
+If there is enough information, return a JSON object with the following fields:
 
+- name: The name of the event
+- date: The date where the event is taking place
+- location: Where the event is taking place
+- event_type: either 'conference', 'webinar', 'workshop', 'meetup' or 'other'
+- topics: a list of topics of the conference
+- registration_required: a boolean that indicates whether registration is required
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+If the information available does not allow you to fill this JSON, and only then, answer 'I don't know'.
+<|im_end|>
+<|im_start|>assistant
+"""
+    # Union type allows the model to return structured data or "I don't know"
+    result = model(prompt, EventResponse, max_new_tokens=200)
+    results.append(result)
+
+# Display results
+for i, result in enumerate(results):
+    print(f"Event {i+1}:")
+    if isinstance(result, str):
+        print(f"  {result}")
+    else:
+        # It's an EventInfo object
+        print(f"  Name: {result.name}")
+        print(f"  Type: {result.event_type}")
+        print(f"  Date: {result.date}")
+        print(f"  Topics: {', '.join(result.topics)}")
+    print()
+
+# Use structured data in downstream processing
+structured_count = sum(1 for r in results if isinstance(r, EventInfo))
+print(f"Successfully extracted data for {structured_count} of {len(results)} events")
+```
+</details>
 
-# Construct structured sequence generator
-generator = outlines.generate.json(model, Character)
+<details id="categorize-documents-into-predefined-types"><summary><b>🗂️ Categorize documents into predefined types</b>
+<br>In this case, outlines classifies documents into predefined categories (e.g., “Financial Report,” “Legal Contract”) using a literal type specification. The resulting classifications are displayed in both a table format and through a category distribution summary, illustrating how structured outputs can simplify content management.
+</summary>
 
-# Draw a sample
-seed = 789001
+```python
+import outlines
+from typing import Literal, List
+import pandas as pd
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-character = generator("Give me a character description", seed=seed)
 
-print(repr(character))
-# Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 
-character = generator("Give me an interesting character description")
 
-print(repr(character))
-# Character(name='Vivian Thr', age=44, armor=<Armor.plate: 'plate'>, weapon=<Weapon.crossbow: 'crossbow'>, strength=125)
-```
+# Define classification categories using Literal
+DocumentCategory = Literal[
+    "Financial Report",
+    "Legal Contract",
+    "Technical Documentation",
+    "Marketing Material",
+    "Personal Correspondence"
+]
 
-The method works with union types, optional types, arrays, nested schemas, etc. Some field constraints are [not supported yet](https://github.com/dottxt-ai/outlines/issues/215), but everything else should work.
+# Sample documents to classify
+documents = [
+    "Q3 Financial Summary: Revenue increased by 15% year-over-year to $12.4M. EBITDA margin improved to 23% compared to 19% in Q3 last year. Operating expenses...",
 
-### Efficient JSON generation following a JSON Schema
+    "This agreement is made between Party A and Party B, hereinafter referred to as 'the Parties', on this day of...",
 
-Sometimes you just want to be able to pass a JSON Schema instead of a Pydantic model. We've got you covered:
+    "The API accepts POST requests with JSON payloads. Required parameters include 'user_id' and 'transaction_type'. The endpoint returns a 200 status code on success."
+]
 
-``` python
-import outlines
+template = outlines.Template.from_string("""
+<|im_start|>user
+Classify the following document into exactly one category among the following categories:
+- Financial Report
+- Legal Contract
+- Technical Documentation
+- Marketing Material
+- Personal Correspondence
+
+Document:
+{{ document }}
+<|im_end|>
+<|im_start|>assistant
+""")
 
-schema = '''{
-    "title": "Character",
-    "type": "object",
-    "properties": {
-        "name": {
-            "title": "Name",
-            "maxLength": 10,
-            "type": "string"
-        },
-        "age": {
-            "title": "Age",
-            "type": "integer"
-        },
-        "armor": {"$ref": "#/definitions/Armor"},
-        "weapon": {"$ref": "#/definitions/Weapon"},
-        "strength": {
-            "title": "Strength",
-            "type": "integer"
-        }
-    },
-    "required": ["name", "age", "armor", "weapon", "strength"],
-    "definitions": {
-        "Armor": {
-            "title": "Armor",
-            "description": "An enumeration.",
-            "enum": ["leather", "chainmail", "plate"],
-            "type": "string"
-        },
-        "Weapon": {
-            "title": "Weapon",
-            "description": "An enumeration.",
-            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
-            "type": "string"
-        }
-    }
-}'''
+# Classify documents
+def classify_documents(texts: List[str]) -> List[DocumentCategory]:
+    results = []
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = outlines.generate.json(model, schema)
-character = generator("Give me a character description")
-```
+    for text in texts:
+        prompt = template(document=text)
+        # The model must return one of the predefined categories
+        category = model(prompt, DocumentCategory, max_new_tokens=200)
+        results.append(category)
 
-### Using context-free grammars to guide generation
+    return results
 
-Formal grammars rule the world, and Outlines makes them rule LLMs too. You can pass any context-free grammar in the EBNF format and Outlines will generate an output that is valid to this grammar:
+# Perform classification
+classifications = classify_documents(documents)
 
-``` python
+# Create a simple results table
+results_df = pd.DataFrame({
+    "Document": [doc[:50] + "..." for doc in documents],
+    "Classification": classifications
+})
+
+print(results_df)
+
+# Count documents by category
+category_counts = pd.Series(classifications).value_counts()
+print("\nCategory Distribution:")
+print(category_counts)
+```
+</details>
+
+<details>
+<summary id="schedule-a-meeting-with-function-calling"><b>📅 Schedule a meeting from requests with Function Calling</b>
+<br>This example demonstrates how outlines can interpret a natural language meeting request and translate it into a structured format matching a predefined function’s parameters. Once the meeting details are extracted (e.g., title, date, duration, attendees), they are used to automatically schedule the meeting.
+</summary>
+
+```python
 import outlines
+import json
+from typing import List, Optional
+from datetime import date
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-arithmetic_grammar = """
-    ?start: expression
 
-    ?expression: term (("+" | "-") term)*
+MODEL_NAME = "microsoft/phi-4"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 
-    ?term: factor (("*" | "/") factor)*
 
-    ?factor: NUMBER
-           | "-" factor
-           | "(" expression ")"
+# Define a function with typed parameters
+def schedule_meeting(
+    title: str,
+    date: date,
+    duration_minutes: int,
+    attendees: List[str],
+    location: Optional[str] = None,
+    agenda_items: Optional[List[str]] = None
+):
+    """Schedule a meeting with the specified details"""
+    # In a real app, this would create the meeting
+    meeting = {
+        "title": title,
+        "date": date,
+        "duration_minutes": duration_minutes,
+        "attendees": attendees,
+        "location": location,
+        "agenda_items": agenda_items
+    }
+    return f"Meeting '{title}' scheduled for {date} with {len(attendees)} attendees"
 
-    %import common.NUMBER
+# Natural language request
+user_request = """
+I need to set up a product roadmap review with the engineering team for next
+Tuesday at 2pm. It should last 90 minutes. Please invite john@example.com,
+sarah@example.com, and the product team at product@example.com.
 """
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = outlines.generate.cfg(model, arithmetic_grammar)
-sequence = generator("Alice had 4 apples and Bob ate 2. Write an expression for Alice's apples:")
+# Outlines automatically infers the required structure from the function signature
+prompt = f"""
+<|im_start|>user
+Extract the meeting details from this request:
 
-print(sequence)
-# (8-2)
-```
+{user_request}
+<|im_end|>
+<|im_start|>assistant
+"""
+meeting_params = model(prompt, schedule_meeting, max_new_tokens=200)
 
-This was a very simple grammar, and you can use `outlines.generate.cfg` to generate syntactically valid Python, SQL, and much more than this. Any kind of structured text, really. All you have to do is search for "X EBNF grammar" on the web, and take a look at the [Outlines `grammars` module](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
+# The result is a dictionary matching the function parameters
+meeting_params = json.loads(meeting_params)
+print(meeting_params)
 
-### Open functions
+# Call the function with the extracted parameters
+result = schedule_meeting(**meeting_params)
+print(result)
+# "Meeting 'Product Roadmap Review' scheduled for 2023-10-17 with 3 attendees"
+```
+</details>
 
-Outlines can infer the structure of the output from the signature of a function. The result is a dictionary, and can be passed directly to the function using the usual dictionary expansion syntax `**`:
+<details>
+<summary id="dynamically-generate-prompts-with-re-usable-templates"><b>📝 Dynamically generate prompts with re-usable templates</b>
+<br>Using Jinja-based templates, this example shows how to generate dynamic prompts for tasks like sentiment analysis. It illustrates how to easily re-use and customize prompts—including few-shot learning strategies—for different content types while ensuring the outputs remain structured.
+</summary>
 
 ```python
 import outlines
+from typing import List, Literal
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
 
-def add(a: int, b: int):
-    return a + b
+MODEL_NAME = "microsoft/phi-4"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = outlines.generate.json(model, add)
-result = generator("Return json with two integers named a and b respectively. a is odd and b even.")
 
-print(add(**result))
-# 3
-```
+# 1. Create a reusable template with Jinja syntax
+sentiment_template = outlines.Template.from_string("""
+<|im_start>user
+Analyze the sentiment of the following {{ content_type }}:
 
-A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
+{{ text }}
 
-You can also embed various functions into an enum to generate params:
+Provide your analysis as either "Positive", "Negative", or "Neutral".
+<|im_end>
+<|im_start>assistant
+"""
 
-```python
-from enum import Enum
-from functools import partial
+# 2. Generate prompts with different parameters
+review = "This restaurant exceeded all my expectations. Fantastic service!"
+prompt = sentiment_template(content_type="review", text=review)
 
-import outlines
+# 3. Use the templated prompt with structured generation
+result = model(prompt, Literal["Positive", "Negative", "Neutral"])
+print(result)  # "Positive"
 
+# Templates can also be loaded from files
+example_template = outlines.Template.from_file("templates/few_shot.txt")
 
-def add(a: int, b: int) -> int:
-    return a + b
+# Use with examples for few-shot learning
+examples = [
+    ("The food was cold", "Negative"),
+    ("The staff was friendly", "Positive")
+]
+few_shot_prompt = example_template(examples=examples, query="Service was slow")
+print(few_shot_prompt)
+```
+</details>
 
-def mul(c: float, d: float) -> float:
-    return c * d
+## They use outlines
 
-class Operation(Enum):
-    add = partial(add)
-    mul = partial(mul)
+<div align="center">
+<img src="./docs/assets/images/readme-light.png#gh-light-mode-only" alt="Users Logo"></img>
+<img src="./docs/assets/images/readme-dark.png#gh-dark-mode-only" alt="Users Logo"></img>
+</div>
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = outlines.generate.json(model, Operation)
-result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
+## Model Integrations
 
-print(result)
-# {'c': -3.14, 'd': 1.5}
-```
+| Model type | Description | Documentation |
+|---------|-------------|:-------------:|
+| **Server Support** | vLLM and Ollama | [Server Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
+| **Local Model Support** | transformers and llama.cpp | [Model Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
+| **API Support** | OpenAI and Gemini | [API Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
 
-## Prompting
+## Core Features
 
-Building prompts can get messy. **Outlines** makes it easier to write and manage
-prompts by encapsulating templates inside "template functions". Template
-functions use the Jinja2 templating engine to help build complex prompts in a
-concise manner.
+| Feature | Description | Documentation |
+|---------|-------------|:-------------:|
+| **Multiple Choices** | Constrain outputs to predefined options | [Multiple Choices →](https://dottxt-ai.github.io/outlines/latest/guides/choice_generation) |
+| **️Function Calls** | Infer structure from function signatures | [Function Guide →](https://dottxt-ai.github.io/outlines/latest/guides/function_generation) |
+| **JSON/Pydantic** | Generate outputs matching JSON schemas | [JSON Guide →](https://dottxt-ai.github.io/outlines/latest/guides/json_generation) |
+| **Regular Expressions** | Generate text following a regex pattern | [Regex Guide →](https://dottxt-ai.github.io/outlines/latest/guides/regex_generation) |
+| **Grammars** | Enforce complex output structures | [Grammar Guide →](https://dottxt-ai.github.io/outlines/latest/guides/cfg_generation) |
 
-Template functions are created by loading a Jinja2 template from a text file.
-Assume you have the following prompt template defined in `prompt.txt`:
+## Other Features
 
-``` text
-You are a sentiment-labelling assistant.
+| Feature | Description | Documentation |
+|---------|-------------|:-------------:|
+| **Prompt templates** | Separate complex prompts from code | |
+| **Custome types** | Intuitive interface to build complex types | |
+| **Applications** | Encapsulate templates and types into functions | |
 
-{% for example in examples %}
-{{ example[0] }} // {{ example[1] }}
-{% endfor %}
-{{ to_label }} //
-```
+## About .txt
 
-You can then load it and call it with:
+<div align="center">
+<img src="./docs/assets/images/dottxt-light.svg#gh-light-mode-only" alt="dottxt logo" width=100></img>
+<img src="./docs/assets/images/dottxt-dark.svg#gh-dark-mode-only" alt="dottxt logo" width=100></img>
+</div>
 
-``` python
-import outlines
+Outlines is developed and maintained by [.txt](https://dottxt.co), a company dedicated to making LLMs more reliable for production applications.
 
-examples = [
-    ("The food was disgusting", "Negative"),
-    ("We had a fantastic night", "Positive"),
-    ("Recommended", "Positive"),
-    ("The waiter was rude", "Negative")
-]
+Our focus is on advancing structured generation technology through:
 
-labelling = outlines.Template.from_file("prompt.txt")
-prompt = labelling("Just awesome", examples)
-```
+- 🧪 **Cutting-edge Research**: We publish our findings on [structured generation](http://blog.dottxt.co/performance-gsm8k.html)
+- 🚀 **Enterprise-grade solutions**: You can license [our enterprise-grade libraries](https://docs.dottxt.co).
+- 🧩 **Open Source Collaboration**: We believe in building in public and contributing to the community
+
+Follow us on [Twitter](https://twitter.com/dottxtai) or check out our [blog](https://blog.dottxt.co/) to stay updated on our latest work in making LLMs more reliable.
 
-This helps:
+## Community
 
-- Keep content separate from the code
-- Design "white space perfect" prompts
+<div align="center" style="margin-bottom: 1em;">
 
-It is more maintainable and means prompts can be versioned separately from the code.
+[![Contributors][contributors-badge]][contributors]
+[![Stars][stars-badge]][stars]
+[![Downloads][downloads-badge]][pypistats]
+[![Discord badge][discord-badge]][discord]
 
-## Join us
+</div>
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
-- 🔨 **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/latest/community/contribute/).
 - 🐞 **Found a bug?** Open an [issue](https://github.com/dottxt-ai/outlines/issues)
+- 🧩  **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/latest/community/contribute/).
 
 
 ## Cite Outlines
 
 ```
 @article{willard2023efficient,
-  title={Efficient Guided Generation for LLMs},
+  title={Efficient Guided Generation for Large Language Models},
   author={Willard, Brandon T and Louf, R{\'e}mi},
   journal={arXiv preprint arXiv:2307.09702},
   year={2023}
 }
 ```
 
-[documentation]: https://dottxt-ai.github.io/outlines/latest/welcome/
-[documentation-badge]: https://img.shields.io/readthedocs/outlines
 [contributors]: https://github.com/dottxt-ai/outlines/graphs/contributors
 [contributors-badge]: https://img.shields.io/github/contributors/dottxt-ai/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
+[dottxt-blog]: https://blog.dottxt.co/
+[dottxt-blog-badge]: https://img.shields.io/badge/dottxt%20blog-a6b4a3
 [dottxt-twitter]: https://twitter.com/dottxtai
+[dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
 [discord]: https://discord.gg/R9DSu34mGd
-[discord-badge]: https://img.shields.io/discord/1182316225284554793?color=81A1C1&logo=discord&logoColor=white&style=flat-square
-[downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=89AC6B&logo=python&logoColor=white&style=flat-square
+[discord-badge]: https://img.shields.io/discord/1182316225284554793?color=ddb8ca&logo=discord&logoColor=white&style=flat-square
+[downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=A6B4A3&logo=python&logoColor=white&style=flat-square
 [pypistats]: https://pypistats.org/packages/outlines
-[dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
-[youtube-dottxt]: https://www.youtube.com/@dottxt-ai
-[blog-dottxt]: https://blog.dottxt.co/
+[pypi-version-badge]: https://img.shields.io/pypi/v/outlines?style=flat-square&logoColor=white&color=ddb8ca
+[pypi]: https://pypi.org/project/outlines/
+[stars]: https://github.com/dottxt-ai/outlines/stargazers
+[stars-badge]: https://img.shields.io/github/stars/dottxt-ai/outlines?style=flat-square&logo=github&color=BD932F&logoColor=white
+[twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=flat-square&logo=x&logoColor=white&color=bd932f
+[twitter]: https://x.com/dottxtai
diff --git a/benchmarks/bench_cfg_guide.py b/benchmarks/bench_cfg_guide.py
index 477104435..a4d288700 100644
--- a/benchmarks/bench_cfg_guide.py
+++ b/benchmarks/bench_cfg_guide.py
@@ -4,7 +4,7 @@
 
 import outlines.grammars
 from outlines.caching import cache_disabled
-from outlines.fsm.guide import CFGGuide
+from outlines.processors.guide import CFGGuide
 from outlines.models.transformers import TransformerTokenizer
 
 random.seed(42)
diff --git a/benchmarks/bench_json_schema.py b/benchmarks/bench_json_schema.py
index 3a1f72cb6..6ed223ef4 100644
--- a/benchmarks/bench_json_schema.py
+++ b/benchmarks/bench_json_schema.py
@@ -1,7 +1,7 @@
 from outlines_core.fsm.json_schema import build_regex_from_schema
 
 from outlines.caching import cache_disabled
-from outlines.fsm.guide import RegexGuide
+from outlines.processors.guide import RegexGuide
 
 from .common import setup_tokenizer  # noqa: E402
 
diff --git a/benchmarks/bench_regex_guide.py b/benchmarks/bench_regex_guide.py
index fa23a724f..14b15da89 100644
--- a/benchmarks/bench_regex_guide.py
+++ b/benchmarks/bench_regex_guide.py
@@ -1,5 +1,5 @@
 from outlines.caching import cache_disabled
-from outlines.fsm.guide import RegexGuide
+from outlines.processors.guide import RegexGuide
 
 from .common import setup_tokenizer
 
diff --git a/docs/api/guide.md b/docs/api/guide.md
deleted file mode 100644
index 1c3160c8c..000000000
--- a/docs/api/guide.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.fsm.guide
diff --git a/docs/api/json_schema.md b/docs/api/json_schema.md
deleted file mode 100644
index 471cb3a80..000000000
--- a/docs/api/json_schema.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.fsm.json_schema
diff --git a/docs/api/models.md b/docs/api/models.md
deleted file mode 100644
index cb9497be1..000000000
--- a/docs/api/models.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.models
diff --git a/docs/api/parsing.md b/docs/api/parsing.md
deleted file mode 100644
index e9662999c..000000000
--- a/docs/api/parsing.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.fsm.parsing
diff --git a/docs/api/regex.md b/docs/api/regex.md
deleted file mode 100644
index 5ef91db46..000000000
--- a/docs/api/regex.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.generate.regex
diff --git a/docs/api/samplers.md b/docs/api/samplers.md
deleted file mode 100644
index 2b9b34234..000000000
--- a/docs/api/samplers.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.samplers
diff --git a/docs/api/templates.md b/docs/api/templates.md
deleted file mode 100644
index 6150512d4..000000000
--- a/docs/api/templates.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.templates
diff --git a/docs/api/index.md b/docs/api_reference/index.md
similarity index 100%
rename from docs/api/index.md
rename to docs/api_reference/index.md
diff --git a/docs/assets/images/install.png b/docs/assets/images/install.png
new file mode 100644
index 0000000000000000000000000000000000000000..2b89b72680c6dff599fc96dc5a2cbfe014a48903
GIT binary patch
literal 12669
zcmeHu^;?u(*Y-8EAYmenq=<mh4GKs%(y4&bokI*N;H}an4bl=r4-JZdbTc$acg!%r
zFvK_NgZFnF??3P!-}msthHLL@*V=2HYwhb?{G*{JPfAQp3;+PBqJpdz0N_Bde;*JL
zU`vvp2{YJVcy3P=b%?M(engh<u+JbjE%_%vNgwSx05AfIvX6DVO}A$V{dBCFF81eC
zsBPZ99l%?^_nPr_K7{it@rEHFrvVSCgS;o*ZN$u7%wmiBqo!<qpsein_mC3ODMNr&
zP~#?Enbys_l@vn?iHR3TG2yLjm1vI8o+ZD&&1GdX>y^aShIBr@IiE?FD}*>ij1b^g
z0m;HM4i`02yA8slkOqDg(nJtryuZz@ctj8~J7+b@%XUV<mGnP-0UWX$jDTp-R}quT
z&Jf^-z@LG!qYJA*fVEPBRrhOuMg~AQjPT#H{wn4lf^afgsZ)mKeou=nuoL6{KR&S$
z&V0sWABU#n=V-}L1J}CC?Mmz5Z*}QtUV-Gg&PXUW@u^z)EO?MoChQVj({t6qzjyUg
z{LU#5sFq)e+4pVC0DHN3W*4e@-ExYSLq--<YS!HF@w?94tp!&T>-)S+LtT4*m*)Zm
z5#S@*R*j5Jb_^>mS`pFC#ox#v@)38U5{;vdUnbNzk{q@9(En8n=`|e5{oVqVf^(!0
z8r_yDHT7{g1j(|ER#rfsQU2til}LMYd6Kf+8FvS?S5Eq!eQ{ceA-JI*Q=W(T??yCi
z#Qs$UMv7`kzmUiFOU~PGPfjCoGq{iU*t%j?f<GZCzX(RiU#h@o&dA-8_-+}KJJ($S
z+K28-9^C1hnQLRIL;ulE?2F&7?bZ6hp+;2;KQ+>5u&)WbFFmiI_0hV=^X<i7Rl~Et
z?u<?|thWu``7yM_d;kcuIK>!<4mEZ{TX6q<YwN~d=So#N*Y340lD9Vj4<TEp&rpbJ
zn#|pmvdhDPq~bJnRTb)GzeNz=#h2#tp1<N_&1_5duW<FR>DT-EC?s-c6W#z`<RljS
zHLyJ%tkUb@UJCaAOQr81XkkfEn8S<Wz9QM_!C;S?F^x)>;GS5!R)7%Ljf=G07=>x)
zxc8MrPhZu(TWqHI+MZJ-I&3N?r)#9bAVM}>$D`!BdboT;G5J`3qtxT;hinQ%)htmP
z)+J^Y3wsNH&gpwKWD#c`A{+IO4b$0-X@P=5Y)>qUO`Lgjh6tIrxR$XdD@U5G!LCR5
zuocth7jh?DvNf%{328bH&P3@Vm7HE%X4kO@ESrD258YE#tW}Smp`R#ij1xyH<`!W@
zmP>U@zZm7jM%+tK((ada`dp(OX43Uos^FnT=`{1`j;*2~N~u>pE}N?}v?`$C&VLS^
zpL_lj@M-a|GnI;aJ#H;pkjf@HIgQQH>#R}76sqo-5Z~Dp&5O>QWcP|YwuHiG>5Eg`
zBYN&=-;<cqQ}Wzov4OX9IqF58p#942()Eas^g1KNVysqZCdy$4aEdYS<WLlmQ?9&H
zKTp(`@X_d4tJfx;?`x);JyXZ-53Zt=o>>Iz+=^2k3YS$3WR;n2I~?OFidj(nBQjx~
zmTMpl@Y(sNpE3M5PU@;VuND-*b~xjCZxCD<E|eDw{c3ccbyQSWj&`I|v62z4zNw>Z
zV-R=hU{m7I7&N&F3$%}7$;(r6%{!p5W!I{0)VZ(Ror97pUn9M$7*5Fi8v)GIr9Y5E
zsL#)ne#Uu<IIu`mJ=2fQSthiE&-x=eu6vEhWP~&~8`mgL!c?O^R_^(BDs_pMyQv4C
zuIG4ak_S<rzh~Qu8JT{ds6y@}467TtO0&xyUG#E_0;jbO$LIPVVGrJVH3-rO^{1@7
zEyByrxwW^tI!19kJj;|=ZuVJ_wy3oU2_3&{BE}L*cYcZ*rl_gP(8}f7xKs7hobHe-
zr9sco!<NqT7ztOGnGf2=uwtTUL*8Uz=yUo182Xdj*zxxQRgX8*6R8oUP7mei$V|<Q
zxi1`_nDd8UP&FH!c%eeXw`jJiJcPZ^bYZ6*$lAgDw3}wSEJMt9mn`PYc}iNpmUl8I
zeyI4YB}M%B59$T&;ThTkBbjqjRi2%isguF%H<k?M9OqaACw4!0-_s8bxbQxwU!US?
zuvj-|uX5=Ip*J}lnCXefGUpcd+oEUr*uMniNbSV+tnyJ2hkv;reD}B1DUzf=kU|3a
z8UtmfvXwDC1Gl;O+hG+oI=b7OaZ<A5B*$}$p4|hIlIxLlvklGWs0a5Betg+AMg*F^
zE24D1$YydusYWkV7lcK<Jvi!;SYDXYXDv?uO!<&SnEvB$^q0PgZ)gdOsHUa&Bb4B3
zbgvyz*Gn=TuT^+}gFf=OhBg{hIwDS|V;zPXm2C%#O}~|-PcT^{`<GoiU-$iVy6qt7
zGwVOa-6>q|@T;{r-SL_>Vwk=&^@XqX!~;>|r)i;o=8?tW;srMLb}xEey9eo*H}Of>
zFumb3h>Q0P7UZ}LwC6K+A%WWy-+dP2iTSRk+&Fw#N_G)%YRpt1QO;af^ID?TPdhW4
zr_=n-_-zy7KM_ZUOVDj9pd<TuX{EY&>76||ges8ZK{V!&Hd5&Rw<%iR1IYE-^`h(R
zLy`>+&#3Lj(!F%MrUR{iByaLkgI_fUN;x;w4tMy(MGp;r%$jo6*&65RRF26RO<E)4
znbLUnUiSC-X2=m&{s6q?zNnXUswSk2{lw2`H}u^-zwTqMS0D~%YyiB`rA)QfBOse(
ztDQu<*0Q{OC3<i6XK5Jz20%}+T=q*pzn6D+_|B=~MS<?kc%=mzZpCg`o*#_@$MxTF
zLC|ehY2a#>$Q6i12EKSOa3_(W|LO0^8MS`xkqXC}NO(wyp#P7b&0r<LuIL97#qFTV
zB_XcWkUuic{w3pb(yLP~vM25H=syrqY58k_s;}cjqUBu9<3x3=&D$bEzYQXfq;eei
zQ9REy^&-;ftH@ba5c2o<uR3wK1OeNuGw1|<PN4M{cZV3NIKk;UsQwsbLU*&gH;r(h
z)%nX)$Mv~U&Kc29sfAh2_8j@~D5Dt!L%;kOW)-TmZzabhah;`B+r8SZVA}NlZ=>0s
zNPk59_GZxU<wBV84Bdeq7mQ`ll_2>QOGRJX!gth_m8X!?N|-|Kr#%>)X7)U=iZe21
zCkm~%6<I<T&C1MG4#r7{^|7>joTFkiqInqqR5!jkZB`}XrrIoLv1nEN_k$mT#U<Ru
zvA9!5%7@G%3*6yqXH2#lX@3+R@&+F}>xk!Q=6tiYcaDi#a_KMZG65fTGAK`5%$;+M
z3*^FrG2OxS+5S_pzZ+A*_m|Xpsz@-0PG2|jWWLJXI8?NC)E&R8(Q1a?Ve0g)CWQsM
z?S+!&h5y<ag%}Q(Byb`#&%oyw9iMg^;ap0FiI0o#G*dR#Bo93Amc};qO>wQ;DBdL}
z+$?qa1ILadt?A7H)Hp{GT+vGhBaxwRJjP+=mBxk_%pdkB#dg8AuYZj`?+lX1lXiGP
zN7lsG%G!9{WQdJf)m8P5d;KOu7s}VaeClnJKdTi&rb>P1{=S^UxX?H%xAiMu`%R_(
z!?@@fQN_tBmk(37vI8X3w<cHPzMVZoEcRI@qMv$>g!sQ1RS$imeCdng?ki4kEV$Nh
z`q%Vt#(nulTyXz>LN?t$(dSA8txaSX89xtx2V&2aTONV-G@O#DuBUynz8zZAs1+4?
zFOfrW=>E_7luBIsh?{ykitjrf<nGliuk)4_!+FN7G8%*ar|+hC@IQRFLt_9|>Wx9m
z%w0J{BYcC&_jYt%Bp^0ZnusI3yCt-?OX8Q(wOIt_@Y=Yddg(qE!$_+B0CR>FkuW!K
zlB_E9VaLu@IRP)d^1Ld`p;>+DWrVVzuk4qeK%v#Be}k8?0@K3#nC_YK;)jZYj_UFE
zwfn6luKR0^)+&jMupQ*k6{Xkra9si4V6icZR`B41Lk(QKUH<r%=SD4Mpoi7T!IOWm
z8w=ovS15q(iX6+!D3g&5p9u13;(-qF<z1ju9Rj>c=VJZiWU*ZyaYzFnJFPRd{zk6;
zIC#<+dk(70F7FgUIOZyXBymsv_#P167WN7^`BJ&_vWv6|F$M6V+#vtve?k~W^K<|-
zAZeuZ2T`#7o@+sXJ2Aaaa{k8T80_8ca6*gsvI|8F4qnSA``h<}f7>dy7-yFTD*NN3
zFC*`uPd6EXcFPJ4j?4YENMi4U{?)O63=P}mdo}=i$d(+e^k0!&O7XuU`QN1_AvVd8
zk@sG7n6fOkhsC|`A~tZoz(wqLFoYTyA|<&bBusKuliAo)9?E)djdSz8?2AtiI4SVq
zy!ExvZi<DKwJQjhkm+8|y1K(~ZrHR-$F}Q;ru?YSP<t4S)!~+imX6Mdd60kj2D8_+
zXT9aBCERB^*{P9LUHrU8M-QB2%0W*xj8kk_75(Ll?Y{RHn&Lb<_|pXL)LsE_D6Px)
zYY6nq^zH^UGCMoFa<Fa>z@vC2g0;+tQ@oEJJr3|%j^jPqswgH$$X}d$`5ZMKeP2Ze
z8$3r!aB*@fJ2syQu1~?BH+Q&V4cx4|z2*Yoez%B`aWUG_R>!43d94p#p?|#H*E98b
z)?kW}xdGZUBaM%qRC;P6Xex@@VY9+0FU`m)TOO%5kJt=m>vFhmuu>|(1N@hy6%!j^
z$M_uMl>$BS5$Sn0d!iSoB-z?|;k<$8IGcoeFGA@+rFu|P*VT_yW8~pZ&1bsSy-8e{
z!)?2w`=@juJi|yVHuSVt;>_;`uxF``4*)gJKbgzzClJD<kYZJ!;mi!=eE0ypQ@`69
zYtxiRO3km3JLeaZotx7c&8ocAMrkJb091NN_0rn9YdcrIbna1yf;JeiOBl?0Nsfkz
z?Z&30(EgK0FO6fqw`m4G%T+-)UZoLK2{`vjp7wzEu;ELKm4ul}V5dTw1E(uAq>u{g
z?HtD@FF`04!90m50A`I*^8ON&`uBNth5@H8F|l!RavFw6iP2@(MEZZ!hRR+La&U9U
zt0oJ!Eju%VO3@{EFF0=ii_E|V|35o3^@_T>b`!O3dG=F>%Ess34(FU{)Ql9(*z-Gf
z5VJK^h5+NI33#A~k~ro_Gma4~Fee#d>!!I{Hsw{Z2rbZ6O2PzO2U?rFy_s-`Xmx*i
zewf(TqOs9i^@}~u@<CPiD`Y~$T~F5Y6nAGES-H4Isc{T2%Z5H@VQ{OX9SM}+Yi`~e
zM}!Bj>r#Qh`Rs@sqm8v=Qb{$P=m8#jtD@iXCo{?NQjB_hOlazZnL5vkuIUD=3O__A
zI|3_tMIWD92TB=7N25fSj;n*U)<N>xltBlZbLx)uC|yNG8&-13V@Q$dSz?N77g1+-
zH=m61OwQ-e-+BTu?I|5%h&D<)nv@R-&u`<UxJ<6|+8;kGG;t?Oo!XA>>QX|tt7w?P
zJd2BZ`V(gAJ>mI<eK|>@R`1u4$*ybnJHR`&@?%K|TLwDO%7nSo6(L^lJqqnmvgoKN
z?q}Ao0y_@35v>tYKj1WihJ_8u<>jA5AZ1lVECK@05KqHmGiB6bxlQU7bT`M!Va`zd
ziSlQX=W8#}d#&WfcEyL+jsxb(vRZm3>brYpFtzQiI8sJky*DGlyUBJXM?6Ix*)J`B
z6c6Whw=5C}E+Z=GW14x*=>66*37{3utDz@z6~(=1LV8sbyz9M34Rxfn!qRFID4k8@
zlf3j-G;2G!W=_Y3@lgG{&8J#k`x@qrZsl%E4amMY5H8{SfV6c-0^ggI2|rFyW}Qj>
z&5`PQ{`YfoiX0sg_&;#zSm$ein!t;I%@2Mf`NJySiY3G2Cm?>f-#e$-@|>?cZ(F27
zZU91wB5zJWLiWeboLeK3lIR0%pQWFnz2CRas*NKDi6QIjHi&2g`@t-Eaw>64=hV6{
zy+~|ir)TD0HR}^B`Y2%NyYljbiloVHPeKW}*Q~!DJ@}xE+H7Qy2qJp%$jLIz7(7!I
zof3QV0OjFvM9ARB7$I?}pEYHLjW5^%dY`2z*X<>{Vtf%4-DF9jZgKf!)gieQj=Aci
zK;Ywn1oZK?7BWo=`H}B+)2gzvlI38GdKY4O7miuflxogZO=UZa*k(A8FmecJk+7Xa
zrNZVc0<yT~m~K+<;)IhxsCgSqp(qmYa#x)7etSf;y>A&oqFL0Na0W|%#$tntr#sCT
z=TBSVPHG}@j>l)D3$C@vm82@><QZ_VbNwd?uxB@BH;_?`*i*qXnU=~Ep{NT@8PkM4
z^4Y0<>0lO_dE*;Ty?r#wPIQD`dRuiMGASy`db)|fG2iSx+^^tJNaC$bC|TV1@3Iv`
zWzs7>X(sGC^&e3p*0)gE-P{$2+no2tIh@XrT!nB|axP5GVtCBg4)S<%gx=T{NY$L*
z&#NFpf))wzp?>2Ve0+SZN~nTiBA04iqMW2m&-ptCw@E_>#KH%jP7)sRIrC(!P_e*q
zC@z%v>?dxx#8{n<d)4N~XkpQ|+9b@$3DsmLJ+gRqG$+J~zA(*#OAG<}n3In=&su!L
zY{{Dl@-@H!vrj)eJ9~GwVocG@je}6=;YQ%WzB~e{J|&w+ey~2%eKM_4=pP)+>$M|p
za5SnXa1Z4KH+Bo7a=2r27HDGAyw$j@Wn4aYa_X2cr52JVtmBK?;8K6!`x>~j%MpAs
z#Ww!~(AU(o3D}pqzuO4g8sL$kV;dsBMK?Rz%wt@mq5H|=UC1Ho0}EjL!8W3G=4jgF
zMSo+s^|8xI@vt1sru01Eh;|BQ`42x}Xn|VRTS8Ns3XjO-nWgEO_{62Sw6C*0c-DPG
z{Q>XKo%9912oysQjmBo^nZ(fyJI$C=zFXNQ^sH#i!E$(|mz0b(^ag-xbH}eFL!a-p
z=K6NSOCO%CB+DJW7`A9NsIWG)5Ey?9*q;OWU(XMQ)nW=uY9}`)M%U?(EB(+icxp9<
z3#J0|g;sUgl%VS!-)~6~K?I<CEq9lk=a8ajR$Z|L)-{5?CqL_74KfXCQe9+j=YsY^
zXdEAEKju{-pLaEQb*Nt~tl_fuC@5Kf)AQL0l|lWHZlE#!M9zna&xK0W7bkCqs+}~6
zpE9c9(^3IaLK1$+)m~V&_!aT+N3F83Pa<M<jon|K6&9Pegseo!UI$RP2g`b<@vQw^
z1%t`i43L*;KDov=f{AJh&dIa6f%+HYwV5v2AdOYI84q}oC^(5O@O0l5sVZdXmO1OU
zNQ$)c(+qPP<+^u@ArH7XBj>eSd0^+LTO(fsDlLN#qm?L4y_<|(J6Wv@(_L#<P5B}Q
z@Q88DO@j0d@U9;(ek1lmu5_-ZLkl#2PTtiFC1&9BGT!eeWB~}sAwL^dt5kD8iPW<v
z&5HP+r---}ol)2^Oe%Y$M)i!`Dm{`u?j+n1W_ajKE|bc6ww4t!CKg&zb|ere-?IM8
zq9^D3B{4+}r>-;gvTFnF#jxq;q#xD@!yUs7=LlmjKL3NA2BS4p?m^_;zU`&pSZ{3S
z(&jhJXR#`~qJTU9{?=|cvo-Tpi`lv_y=W~)ts=4PaDMdY4&C*1QI9XE;EYD#g!OEh
z(k$dGt&$V4)xf&Yo*NCG<=ztX274_y70&a_IOY;KW!HhKgjXT8!@-A$b*3}}uikhG
z@Fw@pkiR%8{#6~(&t#{jvfJ;_yYC(YSLuOPpZ?9&uOF$r6zV;<mr}iYQcU@zpiON;
zlVGnMkrzca9LvkMtSx#Q1tcjC4m}LjR{K)rR<-6FZhLNzmoabklcv3cXyemT0v?**
za92pgr7l|Qn#FN-(Hjlb^}9_*g7*1hEL34YjUAjj(I#MG3^vZ<9264Z(<^wiGUKq4
zH{pR*4UZbP^`RWZh<5)-Ewms&V{?Y!V2;86AnJkso^;q~*vaF-LxY!{xAcrHdO;)f
z6MhwwdaDOpd`EqwKXpmFR3(_MKt%D)-x{=u4;3kmJZmZTV0C@+BsHOgUq*>s;y6;^
z<%98S{W2~8<kC%g?yx8py9epav?&Iy7BvoGJh#dn(JNBVSpj<ggTb+FQV;qJ?tDY%
zi`TwFw(!}2M#2iRU8kw)-f2ScSmrDnj`<nr;q=kzKkRAugl2_%ljh(fGJ!g?&E`QU
z^iQculM;bpkEdbOR%bkP3!~_1BSELCCr#W-g*^I#C;=Z@9JOoC#=uA$gJ_vP%%Z>_
zZ+2GoO^5%AOC2N2_%%iH46>lL1XX{eHcseMRI3WZ%T=Cf1swCIPjlB*{U;vi8fZBI
z2NzmU9^gwM0d#+{qw~|F7e)BY0=OaYo^Fh$DcGT4g8JpNRD>N&8EYIb`_ruZ22o?i
zrR`*-5E2H{SJeeX0aMN1WDtYx$u25AwaBsBi7?Givl>vp)1hR<_684GZ`>&G$qOwk
z@>%c5x{-3P1%$hFyW+>=RnHCaa`~Q=StULg>@YL1x$WSDoc}|MyJ=qE?g#_DF9opq
z$DSztmwrU?qU$<IORAwWlA;XtO$=M2JVzUa(CE7|AtN^0{(W^Gn@ep8rm?%dv|qg&
zq`+BDlJpSio7gKm=$$UW->8|TddFr5@Sbvj{nMvtzYfwsSFx8u*UP9RpVw49*ACr2
zAZZPJwdOlJJ*5YxE-aLI%!6^p$q1>k$<=#anh4G3tY{XLXksWjZ87xLHV=tAsKO0}
z_%AR6=aV>^q!4vU9{l9M_uu3vzcTRyYC>Q8nj<p`K+Y>YiRB)gpwgz0Jsun^rn4m-
zVYJnyn4w8WZu6f@HE7IMiEV+AG%Qw0gTc7ubX|e-UkQ5yAa+DSnF|PS`Cs2>LU{3t
zVq%-mPc}0spRh<kROxZfA5+SYNCk2j>b-J11jf9`3=9lhhZNZjXp|X7^CARp7MhAm
zDZ{p}c)rDY01=_&TTO)E_qzHwy6l`<ia{PdYZbj3sO5!6cgt9hfTzBytajw{-$`<i
znvL1h_CGK?pIOW8>}E*)V|rF*nAwdP<>`m>zIFr~G0+XAx?r}4*u?5PWm-h%^eAb!
z#$B9gKIa|)8{pOGD~rpokvy@8@*!%P$M6!c@&o}}3Xp&MBrV_|(;EP;(frB<NE3AU
zG7%`yHc5G~o|o=Wsn=Yvz1y5#8^D%sSv|DzTo>5fAFV^>s_xjhD?@Xe$RcDiSg&A3
zVY?_`bxmO#EsYX5AsT*x!py0c>26ByAKxH^i97<&<BQ=Of$TS!C``hNK1g@<m=J7<
z9w^%{e7}O2nmgBh-}C%q{p$3L`QE6n>=4bs2y$Rsj5}3WvrlA|b{MJ0N}e26=CS$C
zy#4%Po|mV=bKk3G{4_d3mA-~=X4VE@8Z<VYw3??GP@NYf^5|j)e2ra}cZ<;PPilg#
zzjl{fk&=TjNBYBQ-wi5W8elj>T=BRw3Y>0iX>^IEM5ywb?c{xw(yJpEeo7V_+X?<)
z7WrD05Ilkt5nHLt>$)H}RI9Ge!7<IgE8Zka?1eXH$Ufq>HeMDZaG^u_=5(MXBELnB
zakJUbiR<2;lk#T?6+IIVXHI5^=0t06eZ4}sF!>EjXtMZ3T~=^Y9_}ulR()R;ZpNWt
zNQImte+jAG+SJu3JN1IJ;Vpx(iyznkl%b|Z6*e{K;PoVNacaa)I#7D!Ji^zpzE`sF
z1h+1eAzYO!q2unr5Z*6j&Kk(I#~qFxh54?2NOo)(OzJp{2yA$@<GugnO;Fa#(*e~f
zf)zO~xIOOTzRr~A_PQWDrHOlxvEP21mZpJ)Llkm7uaMfWQK9*2k}t0>STHsHr9-vD
zp`#zXi2mwrzynoH0KHnVtx_7y5d<v+?ZH-4U(WoK{rL;;nvd7+n5v@y39+ch)~jus
z<Kq_*-BNP_Vy`4BQ=RzXZC9*u*K<!J^ZS{cL1ERwj*6D65s=(!`4<t&4~Orm-Epfs
zpn0StZhb6%(I&LiAz(IIrX*wpqy{Xv4IRD<FZaTz`5gVTszUL=w&<Ac!URS?#P!ad
zlc^DZzQM<hW{x0ttV`$2g5TmBtLs<%Rih6HD~ZoGPToA&8}Ju<`6;7-MoH*4fQk=}
z=~Z%_vC|Qf9y975g%K5%#e8h9%1k%V&QmM%UFLsld&l8b2X3);_O6dZjE(hZa}$sI
z*Mz>d!#P1{nH$;i?%2EbW$PChxBe=@n6W6_#vB$b(LQq18OuQ_ehU<;KPOaMBA+HQ
zY#m!*PmhN8%zBXaY7Uw^hfah`v$JeAIUn_=8okgSo}@G~2S%FL6Iy1h)oXW~&UOZ5
zio-H`S1KEQQRQ8AM`_^At+!4ca`q_e**&l#ZzH{OLeu?zaiq!{<Tle8?||C?&v&wI
zu}{wure4po?QGbUL~HtFkvWbS^_w~d-45E$l(>qG6Ych@9YpQ3SN-3l!Pb|-o4rC-
z`@`y^`VTU|3G4T*W(b}#K&M(B1Z>zCoJ`t1hLWZ#i`ooXx$e}i>RYW>KVnD(_l59M
zNjNJrHO6$=4}VxQOEH(n$QP;zY>4DHUu6K7ti6A}eP*;j&8bCAvBS?o_JfA|jha|n
z(nalhH>w;F49~7Bzi&8N+SNFJI@gtRgH2zXh<b|1P%nAI?@=R9X>;m<v6`4etg(Y`
ziY?Dx`Eqwreef}5%qy`7Ix)3&Q};|1br^n2OWGmsEyAU9G)WOAAk<fOagwO$Lt6hE
za*0>l9|r~msKbsL9MRce&<4hKI9FAdsK`Ii3bofy(%0Kt^$|x8WlNqO5Wn%Q=47KV
zy`!!)4^~^KP<lozBe1$pR4EIRCy;hfn9<vHsMy~;nhSi1x|8bCo;XGrLN98ZVbsF!
z4kPqWAzNPyM^Q6XzEqHsh8LlyPfCZP=AR%p-H<C>p}zQbgwkBCb!UlJ=<&r!ing`-
z1l@XFr&bJ61C7-O;VhFF^P|kK+O0)o|E%|p%=ybZ1U!U!AoXF=nB|0R0bM5f#ct^%
zk%r-|(Xr8kv?77R!vp%>^PjMrZs4YvUgNY(_7rpf+Kuguo%$n5z65q}o~V+vv;ZWR
z>lWmEOk;a0H^H$W(QahmwrJhFER19jImQ#yui`ucKG#j8|7z@}%+WX#TXOx7;p9%E
z+#Fv&zj&Y`d){U|<%wJ#eAo`!=(uv5p0*vz|M+D^ZK2{k9fd6nI?5?__@vhR)SJ_L
zW8k1G4y~(RYHBn|lW&AK3M(JovCAn#X!p|5BDfo+MU&!MeAkRyifNkSB&IwSl5DcL
z<&P?>lLnDr?g_yx`9b@6VS{GH1@MDj>la=Z7(<x5p~VhYA?)h#(fIm`Y2{3vN7Y8Y
z1eiPt(63i@=T0oKgm22uE!HKO$P*q24^EdYELpH8B9wB%uCRvTOH0gY_s~{tLYrsn
z$ZNUxe6b&sicL#HZVOF#IM52M2_pj%9eV>u{nh6l?fNv5el}gKJGS{y`xJLqjvlKr
z=L1cLlXgXS-6Y&!oi$o@7d^}$7H}VLG=lB8p3X1I1rChDz!2`6!#_8_XXBxl7S+S%
zl4L-dl&{zPXE6us4qO|)XDIkG)r+FlRxC3p@H<X^r_D~V>Gx;`TymIc29Qe}M52NR
zV(DpS;sQ>%gc8Fn#YQ=lLz7l#g^mW;`hq~t*pR!t!aU)Y?uMBAD~0qzv+JRYrA;YM
zKStWIo7tV!k%=F?sXWx>89e6QBa}`Z9vR3am^Atvl$R;fZ*jb5F&5H$l7sbmnJwTR
zTAs!^=*7V5RT|H7l{7z-#{R($8|YkUE+N!wcXlk~E9o}pDvusmh}~3lu;-C@eIMzr
zM^vZru+bLz^{;)3yHet)_JYtIE1KgP=g@faj;d_vV3QrCh{pncbfSrymG%54xaa*&
z|C}K=jpXUbq*i;l4CR|?%h%riyXqI4>*REFv!o4EZ!erCwR4A=+aqZ7cAFxpXAey4
zC{6gs-tOa)_JKxt1u|X%BNhWE;eIy|4E*U*VP|U}uyAZ)*MJUkX<FwPEqD|ZdiZXm
z7>NVEm$jb;hYy&SOnqG&6C<TNK^J<RcyGB>Sh%;`mc(9>IU8t1-8a57M~G`J9tL_E
zzli-ux`7xQl*A{e#rPn16ZnzJ2SpYA@!Y8qrH;>X8$D0j&cD>=IpiZ<58%1nx@&V#
zvoP|Z5NupYeFmWu_b$E4CzP}MY9(Qvu$*Avs3&xm#;K7C@hwP!=gGE*D0=BeZ^E~>
zu_d1i3q4?DFUKv9-<#}?aLWd0lpaUopw%qYJ3et1w4FGxHx?_?y){~xhyF(I)J3b$
zyacBN&a7-B83Ol(k*)K>75S@XQdjx&?IO|-)olCQN-bKG+!O9_>GjKn26;rt%s93O
z@(TbPivld_HI}b$X`xQ`7|Q5@sx-$6;f;5*64`Sc?1mVejnXvwy0sUrp@qwsQ8Pmc
zh$jPeyqXY+2m*m9svkNgr=uTf6R%S@-=2CxFs3-QPwV!Q*g>8ypvt#H&>?<#7aAGI
zP*fS9q9)LqRHsppDaUxuTP5B&v-k>$&f2E&#xCON=Hx)#M=HL_Q@8L}<aHONe87g?
zT2)cHzmK(3%L_tUqs7Uw>f?cfN^T(sWY#)Wkp!uH=xoF~_+ttpSzATjjG@8>+#$2;
zc2;(ndgZ{W2^~_z3yMsWs-H87W$4EVzY0-BM^<NP;(`J4j`IM|6kIE&-w7oU=~EI4
zfcyo`b%2~%kmr)ydyM5JL-!XL0s5!*rkCU-5oEF5gk`Zu)-e_L2gSo(qW{o*&gFmv
zB(w~*4if)3-f+n=4&W+~Gw$MCQi1as5N2P%#{Tjd2nxiKp|#Y1`NmM}Bi-113il=9
zC=D!8UnZI2(XvSc?)q=s{$^JKph&w*zLUcF6PC`jAnjZK&!zw@MOt#nk4m>#U>`#T
zg!4XJa&Fj(3I03re?|Qt6`^G_w*zST??J<a2K5*hlsrFi{;MnB^C3{z?TuR;nL2&9
zy56A5a6^|It|z#YqkP2a|J5-9P`bS-W&%N<8B6O6(2Ba29UI3y@wrC-SLqndV<6o}
zBa_cKxk+*kB5~&-naof(v@v-wf=T?}T<sQmP<pr5S=}>Eda_@yWh4@g&m`YdPokv4
z=CkhW|7$?`Bun*$%Ltw({U(vvXc;ny;c?4d=Dc^^9)b6GzW-lov!5T#BUMc>6Yude
z<EeFlEjSMk^WEMfGj5le+Z!YZ?|+B<f$z7FyvX5pwi-g(t^9<Gs|d*FaXol<EwS^~
zKY?s`AxeFVf9d`oc>Jf&a&<1y$)~3a&Pv2$z{Up~b@S(6c8U3Di#XLk-MFJhvPXM4
zdd4~qdI8B*!tKRHoVVJ4!Fvr~I`D4Y!|yRmV#7D%*-hcjF6eOcinUxAOT?c%E3AQU
zQ0_fT+CB=n2HTZM|ABaWA3l}IeVh-S&c~BZ`kO*6%@~0F_OU;#J96{#NZ!52(uwy!
z-u#^+^9$>~M_?-gCf_Ih1qsb-jKI_2s3(_D5Oj>KXjeDV)V^FZ<9#eOm;X?U=n|AD
zq<(RAHk`KaFR?^c4$I5kdU~1p2f|yB$5u!=$f^DnR6dqD%=)Hq_isMNDYjxUuQ;FM
z@36P9IUG87IPm_io&j5NT6s(0Zzla`CTt=IF>~<WME!kcY(;wY=iI-;>R@G{X^~Cw
hclH0fg>o90?Hy*n8bj!a{r?F-QBF;^<cay~{{vwg+7|!-

literal 0
HcmV?d00001

diff --git a/docs/assets/images/logo-square.png b/docs/assets/images/logo-square.png
index 9bbb39243415b826a71ba65121b20ada5a7b9503..ebf8cb8cb9e1d7d59f427a2ddda99de5d12e5ba7 100644
GIT binary patch
literal 13864
zcmeHtbySqy+CE5#AT3D8&?22fcS$z_!Y~ZoGc+PCDGky}!w^!^B?uxRjdX)FGSUdY
z(Kn9oI_Ep*_pbGQ|DCDz?7i>%+WWrtb?^N=Yfprxx*{$%1vUx_3a+w}oHhyyYTS=6
z7ACSLpj~hP1%>jXkFFt18{|pn;^u5+>j0sHdAmU9AYQgsC@5Yt6B$MxRLuzvH$0IP
zGOuoj1$&gIcT)O0%F1ao%5R6C&gHPE4CpE_E*<vF%>{Dr+<ZR}yy?nGFyW=&NU&46
z|4asVLSP|j=Xz|&**6q%c~w&Bd9u2>u@|&Df!IF4VVbWzIX*{(PQe{pq7E=0tTqly
zJ}F-IcV8cq`Xu*+2)HniD)&ihVgCLupHy?Id%@$QOQvt`?GlUa6CILyX0}aUm^+_R
zCBI8r)(SBoyf#znuUG0%V^N;m<bI72iddXvxOaea@clewF_w{aja|I;yYz_Fqw&La
zHQF+_r{MA_QN;A)PVxFP4Hb3gN;7{ahpU+fipK|5c<X~|5`Ni1&9|E*Y|U;?vLQJF
zHIbx5rRtw@pF9E{G~h$g)lBc^7O*3@c;zG{lXHR$@J&EZR=2P28kcvO%wmc<*34EQ
zM0TaRmr`8Qgobc8KRthMZ7|>AGkN>_=e_Dme+#wEULN1WeKk_w!tyC@>$mR)#@ygF
zcdgDp#y%U+Xn#{L(>Vig*}`tjOspo9O&N1`>H>I3s>CRER_1(qW@!+Mkyl$rP{R6l
zQTCh7Gx|Wuu#-75?^f{e#`)cV$d@93=wm|^xDl;mwSfnxaTEXt(=L49&xIQ2GugbH
z(*1C92c68q4fg45V^f%)!x<HK7g7A#$IpcqVO28WXBv1Iw00xC1YlL3ULt#kzM?n_
z6*iVYi++jQiC}f$W`e!b`&3N}Wu{bZJKG68e4CzhyNZIz849SNf|FmftE~OJ`=V@Y
zVNtg4R$YDIWi<DS5<@KSw#}HDWM^!eRW@XDI>oeM6q~)G_UHk;;;n1l-XNanel3Et
zdH=+!uD7iH`&EbCNmaUN%4p2_pq*bj!XArVIHH;y6UuD(FdK4<YNpT1po(7DNL{eG
zl;J_H><88iHTEb_=bN&q*)|93IY%v{eY5GowZl%QeaX7a_$_~%#>Ksp<cY%3VPVqd
zJWZ8Wv60?my>xc9zJ34s?|$P=S<DSu!?v{Iv4eYjy~PHz7W-2$TNcHD0&2Uoni*<Y
z&th!%W|~mDDJKsmo$pqr^gI1t(qd$k6huQNJ$6yb7CY?no^>-gOMM2FaikY!Pj$V}
z^3=!lzlCqYqa<eE?3BeYiAY*+mwo=U;PcXe*8VV*cQmpu_a(fH26Zfq6I{kz>%TPy
zntkDD=2|hW(5>7Qno(vct8jD=+{M(Nt&{*?<7y4n<jY&DD+ftXJ;XcMX7$hON>)S(
zdbr^T<*L`pIJkayzq)t4%d_mh+iaRwUCnxK5_OUD7(TbH+>$v9cS}Q@beg7pl3MU1
zYQ3PXNX8}m=q&AU2H^so@wTUY6ZF0kdE*Us^HU!7gP}5~hFQ*`N@cLjM%JN}RN5ik
zE;}+lV2AbNQ7tp3^YG^h@AV&+?@APA_Q#T^euAu2#DpHiIoOJtG_Ng0X7WvByLw$Q
zIzr=%FkK@rjJsGHrv!<}_?QJv_Sd?#vt|R(c?MhIRpWB^yxNUpyJuC=Lhe2?GfgSB
zINN_Jir4tAHX{6SAS+j%?n*sNAtiD96d)oVJ|k?`%X3$YGhYd`%l*9Gbx6!_yP%;2
z)mKWd?z|dWQ8ge?5z@HcM$BQ{oYGaEvi#26ozW>XUtP|uZ3aVERjjD)OF&>PINN`H
zt<ifU8u5BepR}!3ks+1~Y|u^E`SOm+@e2y-AfTswCF?hTU59BbL(jp84N5Wr4PF_a
z@h@njv!JVV;TxrC^{GcTa8Iuuf*l)^BaleA+M}9#E!kqGn_u9!cT=!TwVqSIb5aN+
zH+-uwrG$mS{w*bf&IM|gXY5*{-B$@ep>Vm{?hPFb3~iAU-ngr3U-rJD3o!wa{WyeX
zy;s~wa5J}`ddq}(6_Y+%>Wb@LiFt2mO|1LFcC~K$8HY0i_?g-P(>l*#A}&@v%EG#r
zyO4Z}$V3kcIde*^V6&ho=_-_Nj(~+*K(ma<H4Z-RYr00Du+<``>ycb~MmIp52a~#t
z^oG48Cy0Q0w3w$i0yy6~1kN-*d6fC|t~r}R7)HVpmedj%r(5VST4yE3p?q^Oo=jU)
z%sFeb`%zDuA;#FI_LNddMfhXtbIe3e*S^6f5OwP(N5s1r-fcEl+pML-bY=*^|C=&z
zHxw)GZktsx2Ymt?l{S;oo@#Hz4!IyEK+>i$)c~edv8=fD82v1;@|6rA@iW=Sm?b_5
z6w7?843>1jzRuU^2td($pK0T`@>J9$6HG3=xXDLAv=N?hx)vN;E`5BIgc1fqHL-US
zmOG6m@ro<z-JY(TLoT?C017G;e^+UHBd_GaHC!B<_>RLHEozk(vKI&ZfJjv>rQueY
z)Iet`Jox*_JE_h#+@2ql7){YuikBm!CM^Bn)aGn_3FZX&1-vEJbP_S7=<n4Ay<;An
zNwjS=DqC?Li1jI}aJA+vp9lj(mRmVHh$w(wOVfCXsAOl+^RL*PQdmj?Y^ykxckL(@
zTRx=a5ugFV_iqPT0GCBzU~{oVLoVI~$kA>_=3W0q0iyLgG-mHyOy&c~HDiu@J`B5l
z9;9f#Rc)lX{C1f(D1jqFU{Ijt`qSfpfqq&xC0|P7ZG;qF+N|umD-M(n1H2sqgyPZU
z+Csk0?kGuOTJ0^elZY3c5~Gp?ckzQ}3nUEHBpgtqHxTp90*bEdGFs$AA-OxnnmzL)
zy(Mv0_%sidO0yGYLvCZ6#4#lWmU7;!mSL<+MqA4mXVjKmvGf}Ct10@71~s>Mh;l;~
zQi3soQB8)mLE!Evef;LJa8KENjF-BMsq~Pc@EgMoGT8%lX`qK3Ost&X`!ITb`!I&n
z=QlN-{MjbY*%@>d>+Mm(Jyc<nJ{~W<E2D*T>6ZG_-f~~~O=_Cb&e@VG>z4(8d6O|f
zje^-=uq8$G@i_?^@i0qE>Fem)ueXD!;MzM6A9isM2*eaKJP|>O>cykGW8d=qWk@5%
zmk=SVjC%W|a@=d;_q3JF6iQ`;{R?H?FIlevF+sFsY)#k|c|GECXywu0ZAKF|&db2H
zt9s`Aa~NTs7_x5+4e9&SITNo=siI#E$kp!VUo}bH7xOS97Y5u!(s7SStR<%@4esy`
zXgc0@mk%IeO25Nfhhl2x1ib1yCOg105N1$eG7ZelRit!yU0A0`*PPT)zVeZ_vJYw-
zDRW(WcLX#WLxn@K=R)d-9oAqS{y?oj7QssBCSR2pX~1ZJogo^xbH~!g7<Ttopph+c
zjLkho45?HLf7zZ?;9~~9Z-=jNb1mdgLpWy`H3OLEKnqDUY*#_DbfgcIvgG&_F|WSM
zm}8NL%U#Gie>Kmr)L&f@UFw4Q(8fPc=q4{E$Ic`6>L|+B+MZ5~Am&UvxqqJ|anr4E
z?rZbsxp4oL0%s+Fjo&9yc6yhEHiC2KCxZwDom4*!YA4lfHmbf2hUw=1PwyGXQ-ba2
z2)DQ_%!NgiT6fz>(0OD=8b5on15%<}Z(eaDq``}<yJ{?E@LdyrTK*Fw5aF#iC>l)i
znrJ@N!R;t$cA*3_ngAG3{-JYjmau9Fam2YHi41OK7oTZB)mgN&wz|A|x!|kYHm};E
z*7X*(BD5Bd9a#?_(#hLmc$t_FMZAqt$xLFR+8<w+<4H=SFaD53@|{ZRzL(nJLk?$>
z)qt<7V;>*kyi*H=>o%fot)Rn;H%&7y>b1T*gVQ~O7r-f^W39A0UM>o)y;p8>h~$?!
z`T4pA3z?|j!qVS`f4ntFG#z_K;URi0Us9L>=IYZxYSf1q1)<mnfez*)BPMuBI!QQz
z=t3ownm{pvw>i)gwuJg(pY${a7@0|Bk+b~Q%_VTsu=>Ntav9YUU2npd_jF$KM{LVW
z@!b!ls4(vP*bt?x1L+DE9JI$PCLBY*<s|xLf-Jb3?0ZOO`Eto>_2~nPL}5nfV#ytn
z4i#rzf4Iy0%Z@9;#~95_rmq{)*ON$#Pt&7D(ao|v7W^YUX(Y>!#$>csQ>jRffFBI*
z83GG-lw9K|t<+u$@Jf&;m|$_^H*UYj_-v8Ghu7f349PKcyLYzdg|ir1SZ;jWBrDXy
z`K$r!D<YCC6BHuztt<O#UOZ4Zo_h(Tegj=%k9N3!T2gm3Y0-yh#OH)y&1dOlMCo+q
zjlCvX-Ae9oRJS-=IfA;%hVkl9h6I;)$hm}JWCV<L7*V78Jj9CHD9WX!#ZJ$(eB{Pt
z8(Fp?ay+HkYYWE}Pi#^ToDnLS?KU0%U`k1G52v)hgX7Ft;q(PjY4NMqSnJePx1u8G
z5j_ng(8r7j3=GNqbpU;YR)@|;mhBX7o8<(t@Nmj9YJ|(Yw3<bK2iCo(w&8I7?JZ(E
zA)Jr-^J)~r!Ot34#bb$Q2nxmunl6^(Je}H$kCzB^WiOpV0+R<wU#_WP@esW3Ae0$?
z*L31?cncbZ@2&IVgU}t$xcFvTRQNUn6>L<9t>d^g_jL|V5Sc0O46aYb+Y(P4EThrH
z2LlWly(gq<BoW`F1M4qd91Y}j`zM{=#+??LPv=UaY+k^H@$V+<E#zQyHN0Jds5qDo
z*CG<`4&gIYVD~X;1O~Onw>q>H2e2)^&Z}tVXEjbV&=4>b7O^}fTE5I@fpV>m=%m*B
z8oK47guW*7=-`^8wM3_={=(u;zqBh}C&H+^cK7Vb$Qc&5@P}0K-Cb=s7QTKCGu;v_
zj<2U2{<Xj`@lCo3TSr+IFKAg-_*&3hATN<D9+l;GwjarW?*n0Y7z5c><p}dik`hxA
zn<Z->A3?5FqUD1w4>ZF-hPCA7ajNm>#9>njpD<peH*crtF?Bm(YY0quJlHPbZ@b$N
z8g^@NgeHyVz_2)uN{oICSf+fel7`bT0XKFhwV;@MjS8S`bCFI|W=E|HKC>jO(j|KP
zpy(ww@VT?5gwBv5H(m1+*5KDfc0_=B1KLn*{GBHsl=l?aFwnK6?cgJ&2UzCXWLmc$
zTC$826vv~|b1^dHyaJ5R^-WkB?>v1FTl^XoSY%aDtjwaxpp2E?PC7Ksz=X-Iri?mB
z&UaZYR#t$gF9Z#hV=8&{Il=%u)hA^;MS{IwsIO_?vG__(Wi7Lb;SQ^SfON1=Ud&<O
z)==D%e7;FzJS?clk#FDO$g#r9^on^b-+-4#@6GdE(#k3S>RWeF-bcgd`{eUwKlqRK
z9-ouU7mN<{$vj!W;OJyUq?AoEr0IdMQ=6F(2qvTQSj!VmD3wEquS+}j>1#oj2(7c?
zg0bSAu$du)p3(9iY9IX~>)Ir{fR7=X9a)bB9nh#lzrbheL&Q0WU7Z%g3EHyhL_K*M
zSUvC<`99qFIzKAi_b4Na45!rgWRCa&6E=>yPGJ>)cz(JLI#>^++A4~7h&nVJfxxT2
zrM9rWf}P1(>JVs;Ykll}d(ODUYt<qjrXN%EwvJmH(vytOyPq0YAFe0`mA&F%c=>?s
z*_{%*B{hW`qwdrIMKNgq?LIfd_it;*8wx2@UyBIQDaWay_sNG+U6Bvj^tGdYxFz>;
zpc%CWK64)g%4*SgA+w$+a>Hn`#~O%I(h(4p&T<EiLNlZEZV}6=G_6$IF8@p9*AU%<
z1+yhKd$Z&O(uMmIn*-eOTAE4@(v-vtJsTM_qr+_yY^7*=(nphp{m~sJN&QQbRD22q
ziGJu6rT3fjq9p8$RYkGtP||iDk9kj`6wEjYat3{wB=3eFhA4B0mmIurIAl*=$m+y#
zsKyPaCF8a}mz}q2E8USudb)q7|M|+(-opV6wgu0-pRSTto?3aOzcNV-RC>rO*NwmA
z)oB6XIyLw6GAc2b6w7_JL9>1zb!k8^-gxwohoHqh-!4T=FIh^zl=?VH;5^CM9<Fy3
zh4Iu?tM&H!r8usDvLTquhLI{RC@ypRCOB}gIbQEor$Pm7^u@)3%LYF5+C~>)@c1lB
zYw+~lwW|K=P1gGb1%dWE>~B6|oG6K>tZGh{<2Zz4R4hNl`@(e{6!?VNdJ0vE05`WF
zWUKB{f#O)DEc9L6JCAEt_w-MVHFYT<cLc=_!D;(o)&_nbB5Lm3)MvM|rC_)cvE$hY
zjlUCkA09K0!BuNFIf(ftf$Ubkb7-xCg*+b`>Aj^gf;q<erb2N?m)g#0b5`}w0@md_
zFv@7sh{k=({O|qB^n=gZF2r;kW?2_f5BQa*gK51+5rU?n2~7p>sWB^SP@2*lUNa{~
z^@2aFUT1vLqMw;7Om_c-tqT$|WV;jOZSEb6uiH+`)J|NrKv^04<r4{FYcyWsin)cN
z&Q)`{n7#Dmyv@D^AmMjnA3gK+*d_YdYp61cwgMz<BD05C5YZs;unnivl(m3nplyZi
z;)G4{1*4`i6UnQz6;?k4u(pHtowR^j>jq2^MPIGpbW~Kkv3d5|=yW@+YJ3s)%yh$j
zS?jv|(pf;V8-ujb5?5GS2Y#{Zc?#N?R4_d?`Qom*+!02{ttEJeXF2hV#*)*9lHo0f
zb@#L0=o8s(0ZHeQ3zp-iZ`t@6H<<*_qHPg5#4E8pjuM!4>b{u`Z;RurWn4Lub7-Er
zOSg8GefRnRa95vD2o$U`6;KZ;V9iIA%ZCc5S8NAlefi=U!|}FQ^T-ad5PEURzA?pQ
z_2SVbOoB^mpqfJ8>Bu}am4UnGYrMax0TIglJ^j|r)#>jd*T-9<JJ`&LsKgZ^&r1lU
zh?cTgA`ZXCaP*5+aj;FP^~ADSZW|PBG~3gz3E2fTR1Ckm5RD-@(C}(|?<w9C5&lm6
z^vU2>jbDqEgh_sJ>e*8`4OX?jZa_%+o`_UkCd?<ucULr=i9=5P<AMnJxAz5`9|y4g
zScgqN7aPEg15OVd%oivF>mE%t$=^J+4(6)xd(#QT{mvUqwcJX-uVmmTj);KK$Ebs(
z(3iKD2FJZgQZqhc=&V1*cAHBK$;112=g7~R$}w^3sF$t{Jm=S6DT#hAS>%LR<-TE#
z8?02i{-k}n4w2##9G%(TG$hW<R_P044uq<wp1jyI)OUO-n;Y!$*l}#%Bh+rziOoBl
zWt%opRBtu$)@F^X_5@fv!rPNku_5v7>3L%@Q6ts!B{!${ABJ@1<1?JVn#|AJr)lw=
z+i1+DQw9X&fs(P`>nu0?Ca%ZkZny;(1#T3$4tfQjv7AZvJ6@|FK1-Rs%35mgp$VWN
zaRcPv0`{m;SC!gRK&~F!<!bjDC)f1T6&9m?VjEfrK!LZ<ZduM?;b3QF;*@?sxH|)B
zi}yr~J+_B@QVWjR9_^{sWmCeizRl%Q&cI{W7av$5%%<<5atsJ%0G^elE`HqdB&j(s
zw``MZPH>*l%d<i~>3JnsGPw1jx{hR{>!qU)wP>>vW}Dea2C4XRWw2^T=B)!;pzF?<
zTV&BX{~oG~yTiC^LfN)32lbm-U3E-IOxs(H@Dd%FWJZ2<q3sFSXrl?!yEOGNwtQ>P
zHK*`J)1v*D8TD~n??TZn?|RV-XD#ufS6$>qjvhkUPCV`RM#Z#M6xa!A?^JSOh*93D
z4^Z`%OQL+Fv?pUTCEMD(FBy4yd)a>~_dJmTxkZ?X@ZaQjw~vE(?6kc_vtrfN9=IpT
z;(|q}8|onU;JDs)OD#UE(%syt3-%+*{KpbSERr)e=FinBszvC672BzEtFIkI=Tbkp
zzJvx3J;*X{pl{hWOE$WR8LHan`|<0<vsj=xVPmMh8DX^W)}z!3h4r#n#8M&r3EN{S
zo-f5tLOQIueDC+a5jG)g8`e^?>GOoE@XYT~%~-dqnWF90n<94WzxrmXS9_a|6n^$Q
zi65GfVDEo^QZ7v{VPurQ(aJtuJH#%0tF+_uOZn{Y^P20v<oiZ)E!u;d!`WI~r*LwA
zk-&4OupD)B|E7=w=Cd19#V5vk=N{VZEKGcd!HHJT7=D>kL3y)7?IVu73(>v`qDq#&
zF!z{-wIY!+Tw%Enz>$kd1h3>Gm)*_;!bc->upRG`WqVLksP+1VaK_@hY@_rd%dhB$
zw;!N-vf>>RJjk(o-EBUBAE;t^-^ZhGmZIrSfxEaWl)F77K~{UiSw~Ter1%ps2a=Vg
zuTj>Jz0z6<y}Mve-02zY!F<+ia4)W&Y$dBaY%FcV@1TU0nkG)XC|TWOY<iP2q3%;d
zozB>&Duy`mP4m>k&q>hM4nfQL1Bf>j;z<ahymz^53}Y6jNOi*ZgrznGBJzSHJ&HLl
zEgI=0YiH%?$l(LhD(31q!RFA{T8A8MhgpJyX2mtTAyoNeTuJB5q29!+KoQ3qx@)0G
z2OqlLX!#_cN<~_>*V>-4ijNuYaW-|h=>X^zjyQ3~R{=hmk${D#cnpYOVhJCz9dx5h
zi__jKQO?V#kLk~PA<wjo^3Wt0P6^$=B~#q$k#s*&`MNr(m-~XLl$uY>BWS?fWgoA~
z)4^nJI1O)jP|9!vU7AzCctyBZ%t<|^D%32`>V_%<NGfL-cOdp9r1viMyR7Eoz&D<c
z{c0pO^*<&@HL*Ko$#Sq`A9A$-LZ9fNY!MvK$lsoWU*y(oXYy8gs4sjJorey~!S<Q5
z$yNvtTfORC+U1DH;awH#33zlXes+gn3<2)C{&EB{Pwxc2m3WjcJ*o48Jd0U_E|Ug~
zT+D$2-OrxSh?;#V@?&CeA0NId;OqPP2xL~gBgS@J_~vvpT|?y36k?~mf+1ekWb)iJ
zeo7{K$~=thCEM-e(xGLYZSq+ZCvj@>$&VyM4{pLiVbic-#Odb?s>d4|MVl*d&9a*o
zd+;@HFroON%&-k8lWc0jU3*?7!^e`Tq&R|dXd1?QiS1sI_4U5^-Di>?CPv(M%MzJR
z;~qUhm8$l6@!o1cNMrl}@~vBbc(qFpAIEd=)l&qwLXJ9r9aOfmta*F-vAAK4@ofd7
zXu4CpFEN)eZl<Wx=aru@pWWE)3;FcSi;C9;>7=aO!U{a|y9lRcydI_}J{FAK6AAZm
z5YkOZd+X@n$F^<bGDt)*ZrE{?G+eCCTeN?CrX?1O-{)U+8I{6N5y>8;*q)6+_obrZ
zxw!t`<{B6Ov#qNG@9GgX@s9x(i*J&=T|!f12Ccz4es134UYg@yW-6ashqH)_6H@iP
zU_G>xv_s2l;uc*xm<aD8a}jCF&BC2~_yrWne}sWBla4jFa0Nsp6UR*LWxwey0`gVd
z+vlLYH!(4cQ_x$`djH6{mUL$ub=l^)ww(qqcsr)FcO#s4XEamzKKV|NNqPbm<du#}
z%`!8H-Dhq&S$S@;mG4<YbPx5n&gDpsn#Wm7g>NW*U`}4v?vuIOpWB1R{90FhJEJa-
zhhgJGp}Ss82OKjnw9qVQyR7`)WS;ld!u}|^ynN?mq8S3v*9(hL*NgIf5=HQiUEzs<
zl|WboiLr*gFc>AME|(z(_UayMBszi7i8B1nhp(7-8o1lcukm%G5TX(f2_}$7d?B{7
zvYN`WvVWZFA<yzMpT>(TbxBdZH`A#Da8U2kjnbRDSJORMDb{0^3Dsn;-8OgIdOn;_
zuk7l2&jbh0|Dis$Nv6BI3kiBBK7nVa@`$|3m6^5;;ZCQk<ifn~Tq|0LPnxa%6xpk-
zw`uCk-40IjcUy9A(}(bc6c0QE>nCV`5pLUGUfnUxw04&`-NY*KjzX1)_X;B~RBKTe
ztFxudJdI!|V}y5acIPph#+}9QMD{voMsebZ@mW}nLc0kqsy_FqP}U@r5NZ(h>Z%p_
z(eG4!^p2|0uc+P-KgS97e(D{u!=I;Qff^M@CnHTl$=aomsh@QZ$96^oc8K5PA7O&s
zDA@6VCl;IkLj~;>zQ3nE33E?tzWJ>${;rv^Ai6g={B|!8Pj`i5h&yiXnrv}AE5K8I
zmhi;hwu*tS1=qx+o*#nWFo%kneRn#^v`<5a^<Ka!-$fzS@H3om<vy(fr>++$mc{UK
zTo)It=zDCw<z3Cbv%Z(7sx|GjwzO_}I<In|$kSH}TjVLLp_-})*x8X2Wa(@H;q-EJ
zL7u{*pomF&xq!g-5Ez{W#M;(L9B|Ow4xqEO6bBdxsBx>g$U<yvm3-VFIzH;UU>|!h
z&=Mdifi31Gf&_4cz(90fjt)*x5ifDTPh1h?^^a;U0Nqa$n7uf_P)(Cg*4Yh0$Ir>n
z$;~0}W$VEUkie!BbF;J((Uw#A9RgVr2iU-1E+Sl9o}QkZo_w6nZq{5pKp>Edo0p51
zmjh|R0rhr*fxI}JpbS4Ce#4N1K*4UdE-+hXC%PY)APZ-Am^c7{Y^VEUe2y+^YJb8z
zL4Rif$p@Dg$c2lClbg%Yk?U6vC`{f13G(}Z{-p<07kOCCr44~PySsrQ@*WT;7{jj+
zmf%1AUEJLqeuiTS=7Kmt9Fe9_WUoAb=~7WyP4iEW9~4;II=cMyLX!O#OPH<IKgjya
z*nYJ94CmK@Al?7O{fqS<vHvtiTB)gt$T@@Ee+*AqP8{%~e-TS(u&t%Y&r2{bFSmuI
zkQE1nSBQ^;U(m{e1IW)O#9<*QY-J&AX(4FE5BUX3*$D~*Ie{TRppf94wn!X)D<M8^
zK??y6K^`kh4t^dWghQAg4Cde!6ymcIwtxr<S_=LGq2XqWR3*sa*HQg|vP43GEV%_O
zL6$rmAR%rG4t^mEK@JO`pfCqFzXcz+6>>nlV7{MFmS7PDXE#R>ayo4tLDmp17bokV
z4L=AMk=9fe2k>%o|D#6J0R*!`I*0>QZJpe`{?VXo>j=?-fqt;bBg79B<`o74h4_Sp
z1$cS?VWbangCdpq1CxiFlkYd&AJZa&3<gOo=!Z^`06#sDv53gJK|nBPH(h6E2XVlU
zA<_Nt{3&lbvEL>|$rg&V@cyCr-!-oTas93ITM9VX{%oS7`zc!y5cs!Fpdb&3<<E|g
ze!n$=Z9q=e5M+M;E~r21ZU4n&frWX2Ag~oL2d^bikb_@X0LWnh0`hTytUwR}9&Qkb
z-|Dx(|3rs6Tfsa*ZV+i}WTePwkOKM{4IS%GD%t)z7Ec?<51zPr1vt0`Ie7SVxrIge
z1VxaF;uaF&<_2*69x&IBto|cnF|L2&MC_-*FUtVZ@3%VS@`7Bgxc*$Oe&_55jsJ(Q
z-{<20um>dce>?eG{Qj4&|I+oh82DSp|5ew2>H1p?{4L}Es_TD^F6@6i@Iai9cR`-W
z$C**0(-+7`Axw)$igGA7Ki;oeixQDFcU+W=kk3z&&Og4W0c~(SWaVv`vYPzuElgZ|
zMkXGN_D3ivIBd#t(z;$Vdvj*t1rX|vuZz&3w0Qw{13rGoPsPibBSHE6;LuTQ+=v9g
zc-reAPDa2uD{D+HolG%Kgj6v0U{=k^flP4601-}#3V>(+WS~%>hBI)>$!Tn0C=(%J
zS}QaX^ND}+&}{<|*y`uCa6ZU)a5{gwq9)D3G31`D4CEv~kpLd)CC-QTKAL<u=s^<`
z?umOdV_{W2>-)-ba5-&g3L2?_|NVZ_%Fv15!AY)tM#hEL(ANHPPWHD|kJF{+kwdcs
z<EknpqCtyWpIDk(?Xho27t-CX-nzhL_ne$RW?42+D4+xcpwQ8wU}2%$x`q7wTDoB(
zmrg5~9?D@@2?8A?!Y3zK((N@;%BJHfW97O@LXGv_KWEkxs*t-Xvm9j8YmzMbc0J0I
zd3E|RCtTb8kz~#&$0dXC*R*@CipzE73f(Hp=}siG?V2Yc^(UR-S9`B%cQgH~X2|?r
zeQD9#N>|!ar2Jg|K-Bqbb8>M7hY{4L?)%Yk?X}XfyZHS6$-)}v>BqY0SB6l=8>syW
z@;Cy&)r7QuOM3achwKebxi-|%?x6FX{mwU`vjClMafuALLkAM>wU&b`yWh4kH{PzU
zwFlYv4<6a?Jd9mq<61ZITi6h}c|dD?p4Jac*O{-j_w{$WkXu`+nTPpO9u1<QprfOp
zqDo1j5D@&h{A=lYD&_j+%INMlGlrvzcM#D>o0AT@saZD1IUGh~k7rL@<W%pM_r>fO
z)K^LXcN^8p9GgWIO$@3FRb1@_s!o=MbHXQ?+vdW*jpPc%^cY-BbbCfl<`dV;zu+Q$
zvd~ZX_4J)Uk$m?w6OLPc&e~^v!Qhso1*I5g(kI630CkSX@d)tuy&ZX4M~8~K3YC$D
z%EOs-7Hb2TD#MD`NNnQDhc&+gT;pS!ruUMCFFobED?kUl#mq0vybfN2ne0h7MT&|{
zJRJ3293%hB++PK{5sRuBD=K_<C<t9cD7rdA$7UH)O^pSLA8#J8Uw&8~Rkeno#hQs~
zTvMLtEYDaX=gPVe(-!?14zR#~i(K4qOjlAU?3v@&93fu$%Y3SS<8!=i64GB4^;2Sh
z>I+%QFn4GRt>d3Hn(L@uIUp+d;Otb3$(39N^!(bwKzKXyq(mDrKby|NtTIVu3S4()
z0P+K$J!HP9eBh(qSJWyTY2WWXyH`GWWPnT%e2VvrU$&jinv1I5?%j63GA3KATRGsX
zQ#bSH*7wYm+_iti9+NyOup>H2JO4x=%-_+w)zvnc0-uCa)SBTm+oN6xh<7?)bxv_G
z>cA4aw$b<`>*+o|e(TaST2KF|!)9do*&yqQou@C@_23vA=YLklUv=`oQ2I@Ne<<;v
zN`i|0SUtVV(_07ngsJIrYB`hIbM3y@OQu~5)~+Wjr}-vN+bq>yoE@*VEl@Mrr#Pki
z80UYhOAugNm)zgy^m}JZKw!tLXp#wNf6r)liR0tI_&RbsGjnAR0w1kfiAmp@i0?n?
z8*%wWUi2KmfbwTTprZat!q3vu&1N${IDeYVuevn7Sv5DiwA0-6-sPIUSUTqUQ+fM~
zl2(4c<Nj(|R>@=vty&EV{wHok=phs$&Ihj-Gj8o29B0RdZz892NOe$7#o)eUl7;)J
z-vKG{VpGezTmP<SbK*17Niw7K9X|%QmJ!Bnwt}m!c?8bsUEiFpnz4g2fP`K(^X9e`
z|LC#2MvmT1^!w?p=aE)%aJ0RrI63YV=pjvgTSwt7y>xb$m*>d;PT3W)UaC0dE2!>^
z8^XY5={qK=Zk+R*0(kGv4*&i12dFI>8-f`Xj~kOljhMAG^zs_HBcEOjeuG;{N|;nL
zzc(*AYr!2&WJ}&l)bZ(5M$&In-F(=<=F<87c1~3IW|;p1jj$IZ;3WNQa{WIf(JxE(
zKPA}TX_Jbni#GS8%Fbxx*lZYPEQTkoG|nS#AMoBmeQo9Tf)a2<llZF5ntv}A3zeMs
z2?+I%O#_+He<t<M4w0odcl>bdqw|y6_13R+5?{U5C}T@em6j@gJ&$O;R7EA5MhhhT
zbJzJ-0RImBZ?%mZ{78-!lg%E57YNm$rc!OHw9JF2i;U<Y>B_+QbcO?-J9d5wr-~r~
z(RV7}{Jzxx%(Y+ChP*_6KBKpV3aGpJvucZh3C7Q&X?W35Ux&?H`go?_;(qWSDEV8X
z{$nQpiC84cxZ%%;cE?4vRjkv{-IJmIu0PbHjD;nTnCoi2*3!#Z{FQv3me3%*+w1>F
zs;i9aEP&ofY0Xu>i&}al%y{;}yajegu~#U@FC+MME&r!`g<quh_uDkV91LW%k7`m?
z3%=yo5smA;m~?CQHSa`D(oD&IfBeZyGy#dcONJE{C>j2L;Q6QY%GSCFzm@?t5qyyW
znnHU{2bel23h&t;{<??%(>nUgO8q~!eYSfH{#Mgc8z^9{AoswC+opC%>?%GAo*+=^
z`#tAHA(#AmL#K_HD*{4XX==J)RLy)0&!RkI7$kIs!H0m3>#}e8XMv=I&0+02HDmiU
zhKJU;c~!!mBXw*-G&G(1H#VNW)(#Uu-+#}12eS=%=OHfU_eo+QGeHm$H-+uF*VsLD
z%ZL0KTWQd1?>V}Fs+Iq+a%7mppW)aO5GHGO86r2&6LWTaVIa7x9=RchhexU2QrOUJ
sp3UIJ)ad!`?(na9`kNZKRc<hIAT+s3$>CAR`)Cwpd3Cu88S|k32Lf%CzW@LL

literal 3323
zcmc&%X;4#H8V#<vfo)fVB9L~likU${76E068zM47+wj;43J5eIgguf#5NSc1B~AAz
zvg<H(h8MyzAwh%$Xl1bt1SJ6_K%xyqLPFS*K*EEGT~^OHwk=a-)%>{is_w1(z5AW-
zob%m#;|n+E9b5Ib!eFo+E@w`C34>`IP>;WAZc@(>$KYu7XG_c(-*^}dfq6YNt}w}f
zdU8*KlTU&NCM*GaG42u!i^ZCL6^)7yxfpZF6cZO-JZ-NBgKfX=a_YF}<r2wA(DiX)
zVArfPej>`+eAUmyrKmLb{CB12HfHM?{V9U{^z0`Gzi9{!uN0hf+E4jogvp<;o;har
zjV9&Jt#3=6Dv(xJzd$v3?(?Mh&?eoi>9-4aoIvl|I|$jt8=8GLfDA3NUvJq8MIp^@
zH!sblC}Mi?a_lUtVH6+eVx3oOKZ+Zn4GI@T7{?a*Lu2PKYj1Bu3J8Bzagzypk6<sO
zA>!$<UVZ;1$*4%Eh$PiU%G%~`$P~n!k|TsM5*`B|BHvsIQc=fg9td-b`dD#yDuwnv
zp7z9h0c<_mLSGrkFU<Yfgg=6VAi;Te(^?Auh%FTp=3xe#IlDz?=gGP&yRUA1xbwj0
znHMz7H*CAU&*#|o_I1IUzb-|n5|@(W6UX^vk(}X3GMm6O)g{bY8d-WPo_CQ&W$hI@
z8T=s1!O_GTk(qN?jc-F14uv>+*)wf+>a${62jmYG-F0oDm@c0nE8F_Q03lg4EMPry
zGFZNupY9vDJPWR<GK4u(l3F&Ko)#54CUU}~XmV1sE&7$ipda+0$*%<+I-%@nZ#JX3
z!r}g8$7Xe6k_^jc3qJ8AvcXd|i@0A&na7scLgV@QJ=_XjM!6IWT=v&NPrfwJF4Y@G
zUqUUEfRSbdMlfSYjz+D{iMv^yq9^msmq%&{>Vr;k3p%m>N9-2XiM^&$7ont-@wBcW
zI;fYzmLXlS)wvS+!fb^usNLFFu&Fg|(0<pt-$d&zAaeo(38MiHN;m*@w%6~CoFVX@
zM$HE5PWYF@3k!o<L?Hmp!m2-9YMVB*+;?qq#5QIY)RuvbMAiX61~yjk()o4GZwK1X
zOi?4HhDC)vgb-A$Ksz&R=$Znk4Km=>Fav<q!?y8(z=)Ma9~nxnC3hk-2-8{Q9s)Lk
zsyjV{w6l7mGNx-G_3no<5@J^-yr`SAdXrxu1&JA!*&vP(dZ!VmoVg*h*<MEFJ!V6l
zpx%ywI!Q;m5VP7nDLfsfCTaj*`7AmcSrEP2t()2VJU5qHF-ULkGn0uLC0)^7M8j%F
zIxiS+-YOY;Ryo!aVj}D$L;>7w2yyi8GCdAxYG}v^O%-=1@v_(_Flf|*18#z8<ey4c
zG;_DPN1hZmAj7fvoyRu3MZLGce8b8RX4CySj_E|EG?6Can-^mcmKO6$QgkYAp~T5w
z*-pM?s=vE1Sw4JtHF@gQ@#aHPh$Lzou*qSul!-w@<W#1Y)SHp!h)@+9S%x=;KfvlY
zY9W_r#4iDDVy`4iLWo@?@RpTSIAz{DZ5}C}rwuhde--q&#??Crw;%|n3!7!GR2$PF
zaejM!F}A;qt`;40nW9{(>^RqGpQ6lG#p=|hsg*ok1-eVRd=0mxe?VftmFI_r`O$aX
zc4xGW3s|BjhyYJUtjOW=S5z-;gQKyc418*DrzjA?QNAp24#RQz298ZD>G_2{s=O3;
zn%v*Wk|Im9Ymh%==dc29Y#@U;?h`M_jx?HKu%LWM(6=Oa^f0)JymK^RrnX-s-zQaE
zg0w8>&YjWR@X7m~=S_mc?jTtCT+TMtakX({Y&JvJGawG3q+ckD;7ubetf*WqjG^6}
zO@=dsD(3bbLF45t!5;7Z?I1XaNKlslFq%Gj5OI(21K69xiqf}>3!=$y-OcbTMI;16
zD|h#RKN;=YdFq2c7*kLOS`%xR-DFC7l|oQS8(1CdyJ76uGAJzZbx=atVENv{NuZqU
zi|^`YMb*oS)FL6~uo{?)!*h>}39+FQXZQO60AnPOD=o$CNb2Jur2WUU{Bt3lYmwBW
z%Y9XMzl|~sO2$uxw0`!tA#rggb9L)@{?2zT`JKw&bM*ljI;|W({0Ij>^Hhgg*{_Q2
z9r#kktE4C&<rjMK_ErcNB=W+6&fZB%YV1OZ$$6e_(%+?B$LC$Mi=&el8*~k74i%G4
z6rNm?{O1G2um7tHyz|R{H;9kE^oEDC71AjN^ODHnf-!}wt5s?%Yp+t=F1i2!QduuZ
zkaF;CZa54P;F2AE)~2!MuK}*#D#J;dW-c?V``(wFTZO%+ujvIC9IamwaO_}aY&8}B
z56XK3;e>}-%#mld%()uhT7S~$`lNqG$QE9FI+bQGGxYeh8jNHx8ogVwzaY}s1&RRp
z+6wg`@JA3Fg;#kI-5Hz2|4<l{kT2ps_W@`raZq{{7KSWyKJniE|Iy3)PCyn3DTu49
z$Bn7{P_@N4?Po4h$Te%QY;lZ(|I4{L5af!}MbK!K;!0^lk%Je2DZ&w}lJ{@;k4wrs
zG2hU0dY2bKTOo3%hB>IVJik>j$0R^YZBVz|tN93-R@>wGGn&75`FUU5-@h`gbyv<3
zXIp@lwKqQ|oJ>KzF8EUvAt_lh+wUPYKdkmIup@&p#PGmq*!kQ4l6Ut1kT<}u2HEb6
zjpAyI8>4;vSRELdjD|`*q_7tIkFIa?S};7xTrmVj8m;G1mr(2$RZT}X>n_GY76tji
zi!_p?{n92?+d(0F^pJ%m`Tj*+_){dn&eNies%Bi&&cg<{xeqcP+{!yp>p;Ls*%}~d
zoyT9I4}`;cSrsdI_1O<soLrO8J5xj(nPn0QWr@U(SEQ?n<DFfMqFc$dg4mY~sw*qy
zi#@CJ%N!J}V9T122x-J_X(UG?R{lV}^WzGS$d=2aGm~Tm<%-7>qMs(g`ynC+0WkJx
ooF=m)A%l%qHm_eBt7g+qwHY~_RyAv@A1*K#C%02ICoWw57oI%f82|tP

diff --git a/docs/assets/images/logo-square.svg b/docs/assets/images/logo-square.svg
index efc68a642..2c1b9321e 100644
--- a/docs/assets/images/logo-square.svg
+++ b/docs/assets/images/logo-square.svg
@@ -1,16 +1,16 @@
 <?xml version="1.0" encoding="UTF-8" standalone="no"?>
 <svg
-   width="69.82"
-   height="94.089996"
-   viewBox="0 0 69.82 94.089996"
+   width="100"
+   height="100"
+   viewBox="0 0 100 100"
    fill="none"
    version="1.1"
    id="svg39"
-   sodipodi:docname="Outlines_white dottxt.ai.svg"
+   sodipodi:docname="logo-square.svg"
    inkscape:export-filename="../projects/outlines/docs/assets/images/logo-dark-mode.png"
    inkscape:export-xdpi="137.93103"
    inkscape:export-ydpi="137.93103"
-   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
+   inkscape:version="1.1.2 (0a00cf5339, 2022-02-04)"
    xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
    xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
    xmlns="http://www.w3.org/2000/svg"
@@ -24,98 +24,104 @@
      inkscape:pageopacity="0.0"
      inkscape:pagecheckerboard="0"
      inkscape:deskcolor="#d1d1d1"
-     inkscape:zoom="0.27368421"
-     inkscape:cx="171.73077"
-     inkscape:cy="45.673077"
-     inkscape:window-width="1416"
-     inkscape:window-height="820"
-     inkscape:window-x="30"
-     inkscape:window-y="38"
+     inkscape:zoom="3.3793722"
+     inkscape:cx="30.479034"
+     inkscape:cy="41.279857"
+     inkscape:window-width="1424"
+     inkscape:window-height="875"
+     inkscape:window-x="8"
+     inkscape:window-y="40"
      inkscape:window-maximized="0"
-     inkscape:current-layer="g39" />
+     inkscape:current-layer="g39"
+     inkscape:pageshadow="2"
+     showgrid="false" />
   <g
      clip-path="url(#clip0_56_2)"
      id="g39">
-    <path
-       d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
-       fill="#e35a26"
-       id="path12" />
-    <path
-       d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
-       fill="#e35a26"
-       id="path13" />
-    <path
-       d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path14" />
-    <path
-       d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path15" />
-    <path
-       d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path16" />
-    <path
-       d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path17" />
-    <path
-       d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path18" />
-    <path
-       d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
-       fill="#e35a26"
-       id="path19" />
-    <path
-       d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
-       fill="#e35a26"
-       id="path20" />
-    <path
-       d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
-       fill="#e35a26"
-       id="path21" />
-    <path
-       d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
-       fill="#e35a26"
-       id="path22" />
-    <path
-       d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path23" />
-    <path
-       d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path24" />
-    <path
-       d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path25" />
-    <path
-       d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path26" />
-    <path
-       d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path27" />
-    <path
-       d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path28" />
-    <path
-       d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path29" />
-    <path
-       d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path30" />
-    <path
-       d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
-       fill="#e35a26"
-       id="path31" />
+    <g
+       id="g863"
+       transform="translate(15.09,2.95)">
+      <path
+         d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
+         fill="#e35a26"
+         id="path12" />
+      <path
+         d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
+         fill="#e35a26"
+         id="path13" />
+      <path
+         d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path14" />
+      <path
+         d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path15" />
+      <path
+         d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path16" />
+      <path
+         d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path17" />
+      <path
+         d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path18" />
+      <path
+         d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
+         fill="#e35a26"
+         id="path19" />
+      <path
+         d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
+         fill="#e35a26"
+         id="path20" />
+      <path
+         d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
+         fill="#e35a26"
+         id="path21" />
+      <path
+         d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
+         fill="#e35a26"
+         id="path22" />
+      <path
+         d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path23" />
+      <path
+         d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path24" />
+      <path
+         d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path25" />
+      <path
+         d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path26" />
+      <path
+         d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path27" />
+      <path
+         d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path28" />
+      <path
+         d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path29" />
+      <path
+         d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path30" />
+      <path
+         d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
+         fill="#e35a26"
+         id="path31" />
+    </g>
   </g>
   <defs
      id="defs39">
diff --git a/docs/assets/images/readme-dark.png b/docs/assets/images/readme-dark.png
new file mode 100644
index 0000000000000000000000000000000000000000..d4526256d462dde768cbcbde886afe07bea2769a
GIT binary patch
literal 38912
zcmdpdV|Qg;&~4aJ$4)x7ZT9KdIk7vo?T&359d~Towr$&d&vVCp?~k}^jLosX>{?Z|
zs^*+4OhHZ@5grd73=9lWQbGg>1_luiy7!0u20B0VP|t%daCQ<Jj$mL&eg7TcL$)PG
zpqIE#qUugcwkA%l`VPimuCA_(<~Ej&hWd8KjJ6J@8JGNcU|_^xk|IBq-7?R!T-`8L
zUWbHC4;R<(rT&1y5fh7wKw*j~r{_|a8(dX-+Z$t68)zf^wmn;|uKFm>uMebj;RO~@
z-jf#`Q8rCwZV{yfl9Pj@1K{u6v-y!*91k<OQanvN@3Na<X(+z^zu}8nSc=H}zr!pI
z2Qb_He@CwF|0fT*ai}ohrw@+T+K`|Tfc~esxw*Qiai@y7AW_yj!0+e3Jnj1=-zuPz
z(!bE|D10u$2I}*bI@2Vkqk8kX{T>eP=HQzH4rXR<Mp%-54yG$YZOv-isinEe<(2tI
zc2|)q-Q?xW+0^y*EDdos!B(AHjJhz=oRsrayN3jbV!;+>%6-`u{eY?(YisJ@SaxPJ
zVn8VE)?bGkbw)f#M@PE;zjG)279_Y>SgI`M+S-|cWxKW7>JWn|%iZ0%GtI9-<KW5F
zsmgijQ)^#$zUUo%CT+>|8oTBeZRvC@EINHPuQ>~-H$iV~Og!n125S^u_c7A>REHQA
zCGhnzY;_L*_&oKpeMb&AI-S7>p5#V4z1^d%J-wL&jUdx98O0!@iyHUT^I)ZlkP{Hn
zSEG}KjlnfE_>&Qa9A=(J>8b*quQof1H*Bsdsinfp$}ax65d#%5_;Mk5w`ZI?poZt1
zy?u#MmQSAEnoLYhtB8bryecP55>PNwlY)#Xeukm+UMu6_;QT<R99Q4sBd0yFPvE#G
z4OG}@(W-Sgl+`8=R8Pvz{xncpTUv*eKF7m0PmynkQL4gaBOZVB@v!&Nwckb-g_TWt
zHK94+8sKTePSN}Ocg?9)3R{Yg<o*JPz0u}c&7Tx+rvJc+JYqfWB*Q#MH4rRB{9+_&
zKa<nu?TTe;>I3Z5HOi11fj0<O-rvkGU&aMi-g6CXS;LNeW2KNa0vnOF9;yLIaZdc8
z@#JhEnD_zS5HW<U>qb)R`}KLGNuPX&A&N^Lc#M>aSDg*z^7iq$<)ot*pOBCs==1(8
zU?-uR>`7ZJn?V^>VGDAzVCKs(J>HO^Z1xGGh-LtFc+2}kK1Ut($}@C&5u$dz+01QG
zeN78$dTji3al*k;S4_0bBrR;U#lH|6NvziThV5z^?hW2EHUY1<y1OTXl+1u>u2bNO
zr-kWR?`BrHzHS#eC!o;KKdnRolO%V#w9ce0Ba6p}k)JHGDmm%+pQ{@a)9C0kTCU;I
zq!7QI@p-I}tLF=I$8n{9dDfSCcvlAr7Z)fT!X1!}oWW0=cn1>X1#w*_d5*NLI5U{k
zRGg`Q3nWWJ5=@vlif`W{5%3<<sg|qnJ-u}e4-Y?|r(0I8q-12!?d|UF2g0HE8^#Dw
zBEw7Rhf0F}{_nx=e9oF!zpFQS*@rn7P521#qnXMQ5je{KZ=to4=G@YPbgf`Lzsw<U
z{=gGjgYQUHa8Xdqdp9~FKmKX|Tb4g$+uF$RZB6TVzyiJ9GG_3-%k&rdTes^WHllh{
zpQDn!>fr5ge`3Zt*3Am$FS5ky*x#RZMbd{=)D%@BeFrl%jC1a<3H+BITJ;7j5l#7P
zltY?MPa)B;Ar<|FdpD8=-7fE2am2ymwSG8+Ba8An5Vsk&ohR9;sctD{Wolrk9CF)q
z!T!z)^>CTE3XL%9Fx`6J!I5(VnEhxltQ3FSLgo>WypxBXBF1g_4_OXYnXk6b&d#dA
zAmZHzfZU&khKBdY>LX`k8{+rxrv_H1v&+NHmrohzajrCFErSXS8(QO=Rh42EeqYk^
z@?UE=jWwlVW(oXuZbC!0doUq)uddM9&Yn_hd`}m2kx~*TG|KK9Nk_q7XNedY%Qag=
z@0f-4Q5-BKrm)tAG9ui{jh7EMbHS|iCPBt$W4fbj881^H`@m(P%i2br6XAW1iE(wr
zSA4(3^e+R0RfaAI9JeB3@>LQJ;bu!<jM9JeAri86a`#y|eiRR))!#u%8vn@FebH#+
z<Q(ucbR$RnZjq-goE$Mc3gPhi{!%O=Dk@M=cB)>MG6SF3^Eps$3vTp<f)NX>>LNL0
z4#fwvo=U0j0?U_2g>hX1A1^D5jtULMX|vgAO=3(rD`lJFcbx2TlM>+INE#yaEPuaZ
zfii|fL8mOaIQNsNn<6NpvxuO{GP$J#cTA;Tf1hlcENUFcb!YL|9onhU36Y637y2D<
z(I~dSv6xUt=!B?aI&xE6Rf-Cr27Y7O4DTkiP^;F2;*6*+O+WGLF?GV<s!QN#kq?jJ
zRHXC*m-I>u>*@HloN;#aH)vp;l{X%uc(9XS=;+ZoSK18SxV6|bhOr}PyC_n|cgS7y
z#IPwkEWMd;iD@%SFK~er^hIw0d@TY*`O0OHv=#z(HmaCda?CDk)J;I2_0f~_vO_q8
zQ)E-@tNx`<ch*6pc*6xdVLy+k1|O9qO@o+sPUM?ZA2WY^MZ&4>aG`Vxtz5B^*Za-T
zMVr&fZ8OOK&}8D_NFl442pWCCJUd~0&G&GzH#OoN3UkLEP&N;$`+%c4ptYkol2E|I
z{QyU*kI(%cmP91v-yk_1Ff$<5l$Zxv0vrIwmXzp)APmz^T!&H)phc)Ty<^?VO4yhW
z#$`wRE8b-gh3kBCK`DN(ykKSP$bN)D@0&DR;aISAcJb1OEcCvz%*n~Y5=J1uSa{o&
zpDFiIsEvjK2p#U1({r&j#U;Wv-<@aVl(R4;DW?9jMz@l*XWA!+obmL0>>D4|nPp8K
z{@b)M#x7(Ox;TWC!U+4@@F;RM)7Df|7xz<%4-Ye!tLfrMhn$CzvhUCyq~|iKu%D8l
zZDr2WY&h<6-{f{S3B86|^4*@t62^9>LYu>$64$_9HFS>olBtKA>(3juHmk;vgJ8tY
zl72ZRd2Xs*BD$tlQ`;5n<_2~uvf`Rex?d8UD8D2crD)^LM>}GRvpvdQPKt9n*9>a-
zX#-&^a88)~CTx2Tzy7e@ach+V*7T<>;|n(I6lK@Ue7bxW!!hXyv$^dJ^j}<_j&gi;
zFY!1!pE;&aRu#j{ivD3V14!k;^#yM=7*PBWMt+DcWv|?vv}4LzNB7+1eOcS4w>@xw
zSxB3$8<q{s+?2>C#(!*mWC@RS3Y?sr5NN&inaRj3*>Q-xm3qb(J`wSK6k}g()v&h8
z9&N_$(0jkPKAR;-R=@qpc{hkU?*;qU)Ka5?;X-g-b#Li~Z-b3Y)l{SY%L>9*I35La
z{r$r_VNFBfv`b0cMwMfL6q9kLKwT#@>!lP0Akui^)X_(2BOl4~#QnET;9Xepk01>v
zTE2`|YyEQ*$Jz%v)a!>YAKy91f*#a&$9@T`8ZIjpq$iPtB*_YGy-!_yO<&LAD+s`y
zr8dS$T}%Td5&crt*A=v*ae*|xZ2G5aLh0`$E5e}T6lI|!cFXtcL(<;fzD`+%K!Jr;
z;TyhQFyfSd(>5#{1SeaddbbJ0Ke%z)5_oEuha4z&0tnt`_Umum7ceQv@8UIds9a#L
z0`>|DgaE0QsS~G7mqiWJ!l8kIlVI0cllZD1OK|TAsM?VAC6#E1KcyhTx9k2WQ(b-T
zuYJf`$NbzQ1}Y`d^;P1T!bM1PX!L#FJ_snJPkh?iik+^)fNa0S6&xii3k1e@$TVeA
zI_p5c^bddP2guTTeQBLQ0%9>y%lLg)R(596k@VgbcAY)$Z3AP+;==zb;m_@*<e>Mw
zCl!{D*~O$#f4n)>@^hcwE6AMYN#FW|UT+?H%Te!Uv|81q?u)zXU~DL==<DhJS?S_U
z;nC47%P;GM<*U2I!QXJ!%g#Zbjo3mDzm2XuxfrL^Z9ecX=^s`|!qi%)*Hn7EOru&G
za*?X;$Kd6#Qteq1;?q;Z_}II3R?ZpbUiCk5)`uPG>h^XN6<66|FP%k@R0Ed;-b!@K
z1W!aT5i6=l!kXlh*Dp}ez<Lm$P;FyCg&6m6aLFl?Xt>k0IneLd8jn>X!oj)2aUTaI
zR___AvtVk!S>#Mx_)GAKEECVeBf^#5ear2K9I&B&{G&VM+W0|@^Gqt@^v_LFiU!iJ
zTbilkezwo?aQp_(;1*@Wwv&7OXr@SNM4&$3qxykZTydX)PcZ$?TM6I5&Z@=WH5m2r
z*d^v)W3}Hy)G+l%=O6zqU6G{+OKe8%=NT5x8V!_9#zL*;$wS5lD$YjQ+8J)*hBky+
zm(IoM7P&~*^6d4us8IH`X0RYdh)ES$a_~^8CLNVp(?j=KO3|9KTc)RCxR5e7oFs$-
zqaV|GEY@7y9u{bEhJ%qcn~qZgQ|S|jYjodu*|v|g^uO}csen-Z+{BBh7%@Ze|KN53
z0Yd3WK{mq-LBOjW!>7rR@=m+aU(=@W)Fy!_y;huGcY0s1YF#^hA$rbt6I?{wze%GL
zZvz?76CEB>Nr2%^ewDJ;(rm;(Pj}Du{|(7(zI3%PWrya3EDi>yoqb#GTN64lTl?_x
zGgFYMqVi;b%8}G^&X>i{bNJ_Bz%M_!ATXYE8NP+@Km@AIIVdO57g2&DoO?KuQN-*b
zPr(J*0LkD8f1pcC4dUNqzMa|Cp=by9`_L~kKc6XM$Km4Qjz;3L^@bo3yq1=g_0fL2
ze1`SuK=7gnkv9MmlYHOp^Z&AuQKq0)HY89u3^+%lc2bFQo9C~7MO|it3HG6flSYO4
zS|ss)z1MhrSklwuw^?syXlZGQ|4ljwM?ooCrod~1>sTesKT=ciw2~k98xU@qQ1!za
zBHP@;^ZV%#rhLV@anxxlUb5XSH%a0*svt?7zKl9TnuDAOY$h9(9prpVHA^G&-${uM
z`TjU!5v2omRww|OaGN>zG>lhER5$`5u+7RPeyZV)z0h?oH78}+^`LDr@)~i^?EHlH
zd+L_D&ND%7`F?X+R|s^!<JU>Ya_+dsMoY9f+l~iS?X6C4Qv?z@Na4F)TXg-gj1c);
zSTl4a>t4J@=M4pI!DS_k_0P)SdJ&Sz7vXp({R4a{8#%d2QILOV5MW_Zf=pMIL+#x*
zCGey@H`y4y6#Jf)oaFA!KO?N0@zTYU^TWO5r{m7-d=Bu(%?2UI3RxvuxW{;W|9eoH
z(L@V6&A~rVZu#O?z2k%lM9G&0I_Q@Cs-YN2zo6iQFo?(V*hkj_cXUQxHK+7>{yUCM
zq+e@v_*_LlIr&~M=$Lg4a2N&5{5s{QhDsqTy&I#-gBr1bI`gvznpWl8q5Hc?$N+ay
z{k?uSC0YTY=*|bqGhNc|2}Ytz5kv#$pRxTV+4aj8!$`?N_PA=llk>_E^gcu&@c`eR
z2ngCEIU4Gt3a$P%RI=UZ=$O|H-G+cHKZy!|pF?uutruh$55ywmw)=gb%3`Rhf|XFg
zV^aHN-IR8h^(F{G$=<-g@CPl=0+LlQvk8P5u#{v(5NRf&%v3G~n0W4205}NlE-5F4
zx{B-;GLBmXZFYKtoGHWV*-3ZXoN$H6jn1Y8k#2IX;HMu?G9>`4e>)~)H<4Oht63i^
zGE?fFjMkNMTGD59uAgtaN#jU4fPQq)B@NH|<kgzJTmWWYQ&daAu5t_gDH`)#9wTPU
zHV2Gwl%9kQ)I=FcbaQGiG~n=4L)hp^pApOTabn@FtQUX8Z;09JRCcTdLsC7rouFYs
z@7M84l(@n+5^D?$6}prn*kcsTNHr@Pkr3f#L>EjbQDYup2<8)lwkHK-Ha09YiHNe7
z*ZWhQ=@g-$%lR@QK9>y!$SOcs5?)>R9p69_(wJns($X)?($b?b^SNmE7L<Ts#~nz=
z{QR}242(7RX`!Hzbu(;o`A0mkm=Pn*MRiMdOx1|bdr)0NN-sh(U#T!2_Mo`cI5ibT
z#t;}neScxl#zR@NhyP#Q{juOK$4$cHoEI7!FCM!tdvT)}Bw<qO-;{#~>fDS3;C8wT
zkF`&ui7hSCpJO(9|IG!D=BE!%;m|%osLALM3nu!1L0sV1cK(bC9VK-*p3ZBmo8&!H
zA`uY+8+GYxzw2)Kd#b_7<`f3)$;QZF2F!J&4@~;jk&yJTJ(vQOjy>rW6=1+h$-;88
zrcq~-<v1tHbx{a}kV{zEl5lD@7&nS-Vs9W|h|D8sq9=m!T9^;UQ8u10V9H1LjC@sU
z2=3=P<OL#ckF5Mjbhvnbxr<RaDmO-rOJ)5n!Sq(Bs874~aaz!P-;Wjt<_#=wzm)}l
z(!uH*{>+O`(%L`<SNj4PAaxo@5}2CKSQ*Eg5%(UEE>b{@oEjMt%_cVbhDS^KkBmgI
zw@XKmwWuKMO^GW^a|~RTh~|iz<R0~Upr*d(5A8=Ss7y|RA@aHwRZ*FO?9M{Gn-tO(
z30~XY4?JWbr9&rz-`L&P9}&v<LKyo}A;Abjk?in%`h34^yK*ryGNOhNNbpz15L!Mt
zMVP|$r}_^d2^DsiC@9KiehNzlhya>?Mlq0rd%0)Cfpc~D4(gt{5x2TD3^rcOlkv7X
z2KxWt02K!=2hS^wTSE_Iq<GF3htEzv4lr~n-{JS}h@_5Fe;0|1pH;hDtOkjYy9f%9
zc8={>&sppQUQ2A|P{tS2Z+8J%^OH$@pqSo9-FAbdEgmrl4fFq`8GE*v-xAuNjmOmn
z6Z5-yRMSII%E+R-O1OcO_0|7Wj*g(mjvGs1>YB0Q-1-i+hF28j7k{B5ABp6bh=Zs8
zbMG151pQtj)=H4LtY6l~OA%$Zk`(0MGheHgRo*u%ra2uPTwH2a>&>YCOC7hs6NcnV
zF1;U=<X5BU`<kxjam7Ve8qQePN=jM|^b#EvzQe{x$I|0vkqd>)l3`_MCF#-d)WG~>
zw4mQfT+m^$Fux)TxH*y4-fo{=&b|LtSq=Y4u5P}I31Sf6{k=l;N7Ev^;YT2fl-Jw7
z&|oH}4^U3AyU9I<ONeE)f6Wo0#lTD?E5((7I*#{kM!n3FN2z8O4pRk+%ex%#`)4kg
zAo&YS#!xD%TdY+wIMxXvqtb;)-n3ZBvgHf6xp4Bl9dT;0r^27j@H7dG`+z$wRC0Js
zM9IAQIE5M?miVMu@zO-?yj3``XkNiqZ@xA4*2?e2(o}^}6moODE>9O$5gHvE3}kDC
zk+iERaV`Z<jVgNDK_~{d5TBLZb+wH)nEo7qqrnrh{no!}j;8?BpMTg2r1VPXbo@M3
z(9=A?Qke5{Q8E9>3G0#Igirnn`0C=Imvx1bE-DH$g*SA&kD4p!DzcvwA6_ah3Ug*{
z9&l3jRzyT9cF-=EJHaqFHy8K?0AwQ*&g-u@$9A=zO0pnHM4PvK@?>aC3xIPTDa-b1
z@Q9+=#~1XegXfl=K6HJ)p4Ean?K?7mt@V5$rEA2%*$mL{uPwBx9S+)ulxA3L=HsB$
zQZVuxLwGm}GJ6u@8-k2?Fe3hAm1qQ(fL;zW8cAUDPE>DR)bCY&x7tPGLMJRNEJh*1
zGY(3ilt81-?EqlsDMMT|XzFDnjhZ5{*?Dy^RlUBbe9>`hX>=HNp$9___$F7N&6#a1
z^a-$+gbUA)U8eMVI9o&?UjHQ$Q&}&65=aV_UWAT{I+;omF#Pq*=0iu_y}LHjnZ-y*
z(LT5~ik*(|VT69h`cXp3r{{Ak*u|hHX>uq>uZn=Wv%{^!xaoG}&Fg$3^Ry{|3dl9b
z!-@!!BjB^E9O?HY^6oWUX$y-|;HDaq@jbXkoYv!N!JXca>#B5VJbyqtQRvSy{A+-O
zVKJ_!!FymPLSDaHe({zZuD&qECo#C_r0tL4fAlNd?OH-(A)yodVqCS7SY5)4I0@=G
zZuCn=lYTV~avf|PntZV}bernrs5b4du}g*ToWYFZ4VxOQT;<qi@N{@xqv*l+iwMTn
z;n(B8F`v|d#|Yg-bNa>&p3(&bt^Ez!s2aFULe!LyNxOgO<CtA|0@ddC7$=8^ahUac
z1IH5+F*@4c5E<s_R$jtcs7}Jwkte$TDzljmP1-oh#m_~1$7~jDClq^7WqQ}T|C^9;
zQz{qot%!O`;Z<+8%hSy195STzouSMd67|GfDCZ3JJ~-|KK6*QIcy#j#2?%`n4{D#e
z#V5w!4jL{EY_-}s;~IVoyj?-2xIg}VumK(9G%w3xV%??BxVO)CuKoI?=JwWs{IdVi
zgoLjI+}$sC6t8`JGES8Vc{{f0G$cj8YEF8}?n>VJw6k}$k%($aZmMTxY^Q}-`Q%N0
zzeacL{45{rGMrzQcE4-DK8ba5!yE9wAHYaStp0t8eIK?lSQI10BVO8?$D{Ui9k3v7
zP#*?DR#rN<LNtPGBvq3?cO1Oi3hBj6ObB(AD!*wOpO4MI7*wAc%3716g-OS0vluqF
zS23dU@aTL`&!qgEveKuDH}(7ZF3NOycQoS7CN@m<q3?us4W^(OAsY}Bwsj@Xm&&0H
zIh9T2%Cl_+*Y$;oFXt4<U2h%-!Z`T}W8Aj=U_MnyNRi*a{UfUIi0?y_(KJUSMf0!8
z(%CJunR*M4m9F~M(ThqrQ)4)k5)l!Rbn7Nr^7`38?i3WlEV5sd>v)@7V=rsnU$Z+3
zvvh=!ILrlFOkUA84-*~nv_v4>9Jxq1h$<>FqUQR_&i%nOpusy|O9RW_!LAB1?LSK!
zs5<|-{FF?SyV4b?PPI%1+u6}~I$JFiLxtOR0l9co&&2`@9Yn>T;%{__Vy$A&rUW}~
zfJH;8&RS#OvgM_5{uys;B|2_KU?ogbhoyn?mjV(7^s|8m3kWlLG5?!UjS|GrCXf-W
zDgvOxh*z5J7b&eyU9HW6c5mHoKOs9fYM5Z*(Yeoxr<@Jkrvl&O{zd{el|!=u%2+4i
zL7T!+EOEjx&_1cP3Ym&){BvTX5IeuQYn`n)2$0Kc&8O1Xit7dp(jbJp5yvX4=!?3`
zzlo~#WdPWjddsIt%EDb0_bdK9%+a-zg1|!UHkS*xm?1I;M@LZ$3yX;LhGVafWh7!e
zj1Ut$SegeyS?j(UBb}np@+GEBS!;3c>SfX?0m3s-jsnXyA62xrZNw-z{kk1`s1%U2
zrv#M4^N505ne_azQ4!V1o_7~bzRmB5<ji7thCpjt3z$+fgu(*@l<ELi$~oz=5Wr<;
zD=X4m$#mMObe?06yRlm{Vt#FuY$%UgTyOq0Zy1rCc0#OcIoy|5QMqyBS0AU`B~#%f
zw`jL?t^rh?a@|?R*LeiT>g2NIW)R3GSgjr`)WJXCHZW>KR)|Ry*`Ottn(q>9Q<M}|
zHxrAeWWJ5VYWx*vaxR6<N~hEnt3|`n9s6OkEl%SuFSWU-EZ1dQ!+nZVUK}!-MS?Y%
z5T9a!i^kxDHBUil`aNC_fXncGY9E#6sVK?R+>@Vhrqlbac>A~3sI$EA$YL24g?K>W
z4AFB&@Uy~CCkYkf=10jwhmP^s*w|t!xlA)x-mLsW6MIX+4=jr|B%F8S0zeyxA5(X@
zsCS0Bm43EWT2mwnv><fZS40%u9np<QWVJn+z5hE(WHu>6K_wpk`7r@D;Q5GAzN+9c
zSSzGWCyxBx_%;xSUvOk>YhR7WwBg=<vV^&46K|;SSMQ9|^@~>^H(+?ZB|^_v?`ImF
zdKH&Y7xA%uWEf?x`T6kjNEyeOdW(nbfS2k&Xcd*~SMroh%t4jCd^;*QB#A7mQWDS7
zbuY*7Tv5>gDTP4)3A}QqF1Hib&JQ2PEibN+_V?T@n<o5(k-)VJ{hzP9w3cb(lukP0
zu+VKPJ49;zc54+i?>1Gm{=O^w20f#|Z;#b%pc>7?xpvg`5HphgKiWhXHczbaB$YHi
z;f&sCi#Bi{(osZtg%C@kxinf|6kxi~Mz$ZEjwGi+{H?jYJn0%4JfswsWEp<lKO+#A
zY(F-Q>>!Fwe6$N;<U2?AB#pVjT4K;{9*s~$dV&Wni*?EO@gFo_G*C(?AwxNmu;QkZ
z^U(Q=>I1j|1xkvz0sY7PiZDra^(l-xv>>j}A-Jpz8&aVbk&|ETA@I5S_fq<z%#fL)
zD7B^T^GEh5g&(i3=S4$Pk|#w*Z+|dxAPT@vMQLrOHFl`oP)e!JE8>c$Drc3AKApty
z>qQi10c*9TW@$$I1xo4>jNW|2**FM=wA1^x+%M}K+PfkCLosaR2yEG^*3Rf_nmj{e
zCqplD-CL5~@SAFf^Annhj}+5Ei-vo9jw%!y003Z4lWD(s^o;B@g6Mk3yFZXgzeOSP
zzJ<<1f=Nqj<Mv!O+x$94&rbdPk<nxY*=@zDNJw$N(`z8;(e}rt4HGR$%3s0WYTEsy
zkU#s499cF+f~H0@AdV*@nSL*BzE~l!x1=WJ_UN#JIy`)&VAr9gdFRTt*61WJO0I51
zL_z`<C<QOY4&37DuKSueP<CCISh$o_C2VKKYEm-=5gTOk$e3{?7UF+Op7%Q?X=wPS
zPEVHOawBvUZ$`W=J#)0$V>9c8Mb1ooQ2g^6kkFo;M~vsh7c<iYC}h_of8+Z1b>*Vv
zB=Rfc$X0>omXB?4bhYtUYwjJCu~jn&^ZGEe$`RC_6p7ay+cBeg{_JtjUtBIIo=kc1
zX-2VI;5-y5s^ywJaSCVDX{DZ-nL$>%)Y0PMpw1tyNzL`%TST+ON=@u*Y+gxWT;}Vl
zsq&lpUoF5cN85(rNe>K>lk%ffW1{+qoLZD1w1l|@-}iK`CjJ&ds;!mU_IUlQ*DNZL
zy{$$8p0l{t;L3!};|LB?MQgtAcx=q)gIYr(VPPpELz*(XDl`9522f4B?Idzy19P~V
zA^tVB5xMXDdmqB8r1=MYs;E$CQF66QVGG=7^CCvF=8*R^gJ)rx5fOvB!XC-ubr@M@
z0du;D2P`qt$fUbb>3QD@>hMxQqD-?FBocR@%r+|`?(h$UOHDg+h}GDWk%*YUAXveY
zOyB>WpSivZnXUPSX2{h3XXdY1_5D!JGzX%nVt|W?80-G|&!lqflCp52+@0IgOjQGu
z{H>l`3plO}7xyfA@A$8ETi;gO=hLBhJyL*j@FRk3l!R2c7=^Q&kocPxW2ixI?6q4d
zYhmM4%K*X%tPmlbq;pgp?-V~9a8X^sc@*M?+rO!#Ma&S1_6QVuIb{{>@`z5%H1uA-
z^SB=*AeB<!@uwRW!O=3bO8ErmNSnb*1W2Et{+qbHn@|u%^Qff4(aez8h3Dd<^AK1d
zp`i7ywtE~3k1DeIFMLD=c}yzH4s9?B-&M)D%1GK?2DYDBb6M5PQGDjC(rN~%(c*X1
z(W!|>oIz4Xan5aJsa7wVn9C$cct&8EkCwnE+925}E?yD3r_tURnO7WmAI?af#dcd?
zT?%%m#-m5Q-gOt&TRuMjPU<@`(jpux`cnnUqz*Ts%%Ewlc|U(<Tpl@j>E6S;e_(L?
z0lJGa9OeD}?CSk!!j(4OR%&c?8X5kV6Z3*Y7!Wr22thQLFtTOdnIMELP?)&;6*1#@
z?jO~sFxbKGA6*kB0|q=8Bj~WtiWtycLVxrU8ccp1Vp}2~U5zzYG1J|BFnIDWeb)CJ
z5W1B8)hXX#_5JPob{xfr%%chAM!HI_hntk@<t@wEWLBg@tpne#wS+@rQAx(dqvIrc
zEg>sV5%fC;BU8XI_5Gr1m7GL->E(2@khHdu<*7K#EhR>DelvQQjy$D<8ECS=IkeAi
z|I!0E_$J<c_7ML3{M<%^jboXUE2Wa5U|D_bCJ>9J`fk1#G?njmo`tiSN$~UV(-1$q
z_~g1zlW*#sDSgOf;QBHXw)Pf36>?FqwrqV=udv!xcc6+h<6`n>z_&EbbTy#Vq=f~(
zCO!bQK{rH`m1{!1LuhIw1EdI;Ar>ji*R=YD=S@<&H}MHcv%+FD96K(iSA`l2yqc{`
zco;ZFIVf%E;uG?=w8`uA$QWAr8<2DsERM+6CLe0v6Am<W$@AhU(kV9ecsy;D6zA7o
z^9lzOMOCi|d6)HVGosJ&e&HxIHVzDdX&UI*%)Op<mP{K9TJk)8kzR6QmlX#CROxt0
z3eQnUQB2iM9U1QI5*=^&qNGIQkL9w#pus}XMz2{n{vySsW?=ZG2&J1tBn!jTz{*}|
znZj_9prnc*YW?+bH8~ZCvd&BSdXFH7%yBVsXUXK7#nh#)T02+(Yi*Vj`&A0~kt{Sz
z3m=b=l~UNdF;A#1B<SoHn#~DRJ;CI4Y+mPd_ghw0(h;N4U~>#yUj(a*XYk1`4sM=!
zn-gJzHh~=B$fEv9+TuJ&tzl{MI3HvX`3Jd#Bzumyx#DBlFh<45qU&kXlf|U&#a6i5
zapH1CMcz!Ui@`cYe5Z%Y;Kg2Np?S2)RNSHO*x(XJZGJnL`(=RksRL_>Xzd^G#8xIj
zj}R4x!^_Kqj9^0sv?2*FbmcF#@~A1vc(Ig}56LtZ!@=sF0GKJE<0LDZ9bhqg19%0Z
z&Qc?hyYiTk3gy&5pku~F<lj>3(ughryF<zF6<tS;c*D@GmK<RXB;pjA49*JCMI?sN
z&4%m9xweWdFZXftDWABm?WcxKvyi?ct$MKy*Q+gqd?z@$CPLo&7;iPJ=FyV5Uuhrz
z>V|^#HaI4asGi3eL=@)1zw>{vzxxP%HM_!df$&726a-9DC#2>dIwpL1b@jq#Y4Fxc
z=%#Y-20d{BU_Pv$0+f&M_O=&Z-=qLc!~hVLOXEe8KCpz89#iG#F+|Q(2y=>QGND|C
z*<ABbWlZ16touQRKT+_LW3_Qi3W#WZ?_7x*tI%t*>2kH{W<swLVsa2bgW%RH7nF~9
zl7RR@5Bu508@#M@{gjRb%iUgU7K{4Ch|++KA_QNs9}I&Po|mMT(JmYb^w%M^z=(sf
z_Aok>c=>5P1cD+E3MNx<IBqQp-=$SMD>kb1y2fab^_Fibhoy`$e<c5Ay70K3&IpN+
zfb~GxdJ**hD&#6ZtSRd2U>YK#1R-NSJ#1eQl-BN$m%i3yWuxomh5ow}m|QW_)6-||
zx>eLqQj5NMd%tD_;an`}K8zLgYF%A~3|*D*XVll(tsg$y4HVbBb60-L@}Fpv;B`@I
zxNC{f7)PuxGMdU~`>J^rgpqgyY#8assW<9OW7evmU7w*)h|JO6Zr@8_bXXchIV8KQ
zq{-2`gcNp+l`(>93>k^NZ_vIW9miCjVUo`6-NdI5-D`QjJe-GzfdWp^pKxTgTJ@IS
ztv$pmjD*A`+v6QI+xv!NRGV?!#Bsp8vbqFTmXEy@1;P)k^zYbe(+p&5!FpVp4$koV
zgZ$HE|8cJG%u$Mj=!#Zqn#MfOqQ}Ry(hmz8`#-|M8(QAkswCv#1%q9Ft9x!x=*WZQ
zs@@t0I=Mxkt5^}+mBGv8!pFQ-n~hSb;6?N(t_VG@!WEZHa8r{wBst|5&xuYfs=Z6X
z|Hkt97ZW|nI&`b5SFF&%roF2>(#7Pt4f4IC%2zCmS5{lu+PB_0AyS$?mp+5&epro{
zf4KjYayuB4IK7~*&w$=({ADUl&^S@Xyllj(!we$vLPwn`GVUH54O}P;+)iR>Ah1TT
zPSbcblPfXVIzIQZ>YtbO@T?mH2~u9;Hp?5p1ZC%v31c4cB9Ly9?fyK&Dgo5n5+@}k
zB_A7S4w1n0KGXqH`9W_?5ylIuF`rj4w=Ujyb$4GDSu#*r=*Gszm}^U5YI^RtTsppp
zoc}Oo=Rl3Y`P69dg$vtD-q7<}($d{WJ9H0Cps-Y!B&M^N9UlrTYGeexp5+L#(9KY=
zY+z83!<EqGc0h~}Gz}aSI~J2paAJIp4-ZCAu{j=W^;)@`7w@NLdcpywxt_c?pXnai
z6#@75QoW_`$jzhM=f8JbhX@FeUW}|@6M<MzjGzQoB9g^W<>C1Bt{#ts=MJI;QqlCA
zu2xEn*1s-p?^OD4NRA7H`<HK>*fogk6{bdn2}K1FYAEdhf*mLbT{jc*1WP_a3R&%_
zeQ681@2uQ`C7DKdATF&58}J>B*b=VU5h;qD21TW7d03C@M~jR?sn$-+#lzBjSGmwF
z-rJqb`KXPsjEQzbG!kwOG_-MS?A?@*Z}XC;1R~ebg>@cHTz&qT=ACd=i?|luZVprV
z_$UL`5Qpk*%}}(@c_3UVR-PuFfFSS`FB{UAm?8tzJOC}*HV@_g)4pva&<B9kB0y57
zw+~fhP-buNk>97tGOiA#h1pTJ-#ZJgJ-u#f*11R*3k#=d&Lzo$sjB3jpnz7G#wzgF
z$CHV@7$~hXi}c{7v%l5`q#29vgQi3iyL(TY`x{)k(82=8x1gX<=uwMkMmO=xhgzb=
zj-QhK;-d4_l@W2{PSN9y=KJ(PC#=HG*&1^H6>@#Fx|POnZZPn}$CsIF*`nb$3!IdN
zx-<kJw$pMbT30Z~rBXxXNW`P1gIG=bYw|sx;KE?yUZzntT;|-({Lts>M7OKJ))XJj
z{?|i`<^r^uirbIjD#PJlrFZ>1o>7F^Fpo>jf&h-Yt8e&}Zg?Xb&WLf>{gH9+{~C$w
zK^?fDKTDiA@cR`c(UNk`tw^C#u^buASNT=><1{X&!#)xxBNIl7UvXJqTl4vGr=U1r
zCCSku_v8`ss?K4aT#9i)$7!eBiOtW?^wGjUj9_MPYij^AlgP*mLrF0w_1X6#&Rw_b
zHO>o-n5wj7;2Gbm(Snt{*}3t15Z`o%T<#$)6@%aO^FAkpO*oPg!wUr-1!EQh4DaQS
zS*rcgW{ISQHU1L#f-1(nbbVA}|As<Ll_$$2n~Z_u_Og`Q;D*RlZgjst{RTDjcy~6O
z8UNhqFiEO4Dli+!7?QUq<12q{UAmZ1xVa5z59*fYLDt=~>PFa+c$W2ex$9JsYW~~f
zRgZ0_^xtrWvS|XvunCe=@Y3GKcyfq=Fr<VNZuy|XZza}2M0n&ZH_3Yh#hTl?3eOJt
zuj<N3xL>_|u5$H0_7l|-;;&`E-g9`$1Xj*gouQtjB})~+ZRwqHCx%Rgp+vj%&de#(
zBo18`U}qC=3T55HH=Vo6uV}&9@D8$R;s*&)sieq%$*#B{Vu^zP*T$C~FE27E*V(4B
z0Cay<vrQ2pvb7xmt+n2^zj6v3R9O{Viy9gAZ3)lE0w6`Jsjwa3>5!UY5axYsD571o
zqmBlNUbMmM|8<J_z@77=)O-5+OuJMhZ%B&5l3J?>d3qkHzRwx4X5nLF7nf@`NP)3G
zU04Sgh9-XLz6|1vYrU3>I=-P>J{=I4R_)IiAX+=>2Gl)9``<oG<PN|~){PD`hrHU!
z>Hi7(Cygt_9T_nNUp&u{q;VH9O!GZf3Fo(24$3=tn4}6CI0VpH<q^$h8q$gI_g?ne
zUbDRsIi*UENnL(f!x44cItSGm7$)u(xGZZ)aTcApSnT@Rs0nU2jk6a4#$^X%GTkJi
zSZA6tdS3ME-qr*qI;Z7CMnZZ~9yNptI0(sD-HBUg=Zv@B3AqFp6#m%^M%qEzV<G13
z?sVBkZ=$Rnc+!)@*_zswpQ!lp>m<vnZ!f`W<!qo}mSNl^UDv%sm4Y1nM;Eg+eDWf%
zF!RmlOTPR=N-=Um4kU;54mG?N`-$fa4nr>zHM7Matj6BJSr3t@wvix0?7B~&yOd(H
z)gbkY(+dJy@R;aF#0eF76GTO=lejD_$*9&VI10$TC1rT@FiN$EqmMuE%4vH(=Nc5Z
zp<Ek!;V2mGzA<6-$+_xtsx0Vzw7R>!KJr)oI3%Z)`RfP*q<I$<y>?)=?Zj})AU613
zo1bn0gq?yp*Rf^()xOBcHK<A1+8dp@7fCm-Bu&kaL1nIYE565qq=9B{pkBG){(L3j
zaPve=(cW(-N(f8h*}VAhxw`a%JG=ay?K-m<*QHI7{&7!P$bkWk-=^s;Wv@opkT|H;
z{ya_JYVGK6i40YL*)}p1U%R%W50n6}I!|#-0L9s;8NMMUt8x@XXoDrirPO35MqaH&
zONO`!!FIqsX?b#PirrP`$N6Sm%KJOS?d>fM1L9(cvPzO10WVljYkXm0VPbO9jzimv
z<Cv<?hc~pw3XcR_m5)~USBF-pXJO9I@F3VcyIY$eHN6SrLveQr|9q#d&JQmkSMZD`
z8^f{23#;{eDtyilNmUml&Wxx(MR6^T*lB#O_QYH%{=EQ~tCg`}S>IaEDP6!h7Hge1
z|GNiLRae~9%0LJw`v%ZIwkN}1%qUfdqBpNrzc_`DF<cH;fH*N!0W(JEIr%@RU#Jct
z%}w%qYhsAU>8#Y~m$Jz!d#QXjmCa9ML<J?RAPvP-=QUb;lLyj0zz!#QvDAp<!lWd-
zRw|phANSRZ1x4U#MBo4x`jUWs_PW~K4JgOb69{68Q#T?R!H+bb^dViXI<6Q4+rDo~
z>Cg(=yQf9|Mq|H1!yXF1eIAYxe5HniHQ73P*LGGYN#f19)yZDfL#|7}LEnCq4tGBo
z``t2K@C?SZnGn%e#AYCj`=qP*r!s=<GIL3E%7|&~vkpJwAeK`qbSCu~zz+m8|1!N@
zSP+bFSk(#1t;lZ3ZX!)hEr4n=L_<T9d-CX58*6IHup}Uc)}TgBw+dQn+>`0Vf?cN;
z`f7jdXiyzyU|YP<`PekKbq!5;Vt>>W$ERe~2b57b?RVjIaF$)V?v1TM06!n+<b@XM
ziqz-+BAnd6QlK-1nH(OyoKHO5&52>BAR|(-Di4^Tw&&>fKlki3^YcPa3TN~q(NuZ7
zv(EY3ouX8Ng#ixqQG*Wyn;u(21y%<6u6)O%mUIV{*m9_ty1-~C%F6Po;pK`GRQ@fo
zdX2m78W>)`e5hAoopa(#N?xRW_OlcET2Nk8Kxk05C{}pT5$3Dol2&;6bVrT-y?f25
zYd#n}qh_SpiKyz#D<4;>SxmNGdp>A7_};dSCU>z0BdTcEqAfua79LrkHbwYeqfgfL
z^~!dXS?*Yf)=QIbp{z}V7AA2V8(%8||8WFTBL0O5taY=-HBmvaP{R@8&Rptb$W^1l
z-}*S*kB-PUBeKk&S8C|f`*$bN+tZIPse|o5M|y{BEBjv!m5gT?x4&RFNn_G8`*s+P
z7m-24;ZM5FI8^n$N>aq<CP<;UC<2CpUH~4IM^A;Pqj53jKn9;y+B7mT=36dI{1lHL
zKWA82Sh4bX?_ou`2lGAbLPDZuCt;Fqm%bAVvQCbCjGSQ67Ai*EK?!__JN?-&V^lah
zRVFUiuZqym4S{wJij2g@1s6igI<1PmPGeQ*MybJX>;An_m|Y;Df2W^tu7)QWIOz{b
z9^kxUN6+>!tTsTca`o=zT~&*(mND6ODf{NR#5?Aeo6Ix%xGMD`YO`%2r3-rF%Hqb@
z7c9R|`CK&!#EP<7{zUcrQ{cF^{UAs-7K0LtQh56?B{zAB%X(QUoqxreCX!0jvZZqI
zSOQUoWRFT~JLQqXu+}(NsiOs|w+2Qw+Tzwr$u{Ip0ajMut^ZQM*~eurUSY@y6msdb
zIZ<*_2G7m{u?iUa&u9uA^P!8IqDf&PcFAO$Z@KQ)<nPp5?Ksoc7UMfy+0xd&i1{qJ
z;!sgDXEwt7c7L7XVR8JKuo$$c*>-4A?{vTnf`(S7&g(8k1^#l<Vc`XuoxM3X<hV~g
zr-pnEUsqEtO<nsS2}&^K>y*QZb`O{a-JD~)XU=Z|Vvtlpzk57gHW05_7z>rk0_>D#
zz9}#vS1MveF+}3CO$3=1T39xFXtOL#MQN@d^C1-V4%SUH#VLaY@M*fr?!HWAKI%1N
zZ45P5urGNkQ5Kgg*;5~j8tH!OrJ?4ah?$K~@-&V^!<52|HT)6r*n~qnB(7Ppc!>!E
z&!tRo%<Uf=S<>T92NkFm_3wwT91W2`y698%kbKuVb89LW<<WGZH{y<^{$u)p(KE;Z
zBnz7{y$_gb$_XR$9k+R-JCGA$Nw)1f;<Kb}_Us7cj3fh?xH*$@mKI9s;cGeSZCApm
zS0Js+!Lk8wrZ4~EjZp<S@wERQ8Mv~JZN~awWRc{mNJeOc0#Dag-Y5py8y~D${8AU)
ze&Qsj4G!u^`#~XNjNS)w{6<j7R{9OEGr3*giwOX}Yp7w*7sL7m$9gM8{myi6>w`xc
zo9Xv#Aq({Y-HZL@8HVjg>ia~__QoBe$D*s-qnCJ0Y6obEBHON9>$%?Zz=mvH3qxD0
z_FUS#wj~E_$@rV&BRyTw>)Z=~)Fo6rtSRFUj?`;)U<_1<AHsO7T!<7*QOr+Mnd>v7
zAUqZJ-%klFsST~n1^0`Xvkjubz|wU##`Jp?WBRS#!+RT-U9x}-UV_~Tk)YDZbE5FI
zVlt@vQ=<yB)1|T|-*^3WeV17T?u+UG2vK;!^j6{T*@8fuaxb$%yB3F2Z5HM97bJIe
z&jB2`92wniAwmi4DbfIEAxm;~P78!<7Vyn%7LX=NPJb4=*CzTmE4&hqgZHoiY-JXY
z_NQH-eM7#)qr^`d6tyj$bUL6ErA(^AO<P%67M%~xCeXS10~YRD-({LCAdD^@$ppP$
z^v}a`qY(dLs<Ih2x=j7eYd}6LE&H<)_QF<-)KyA8^$1v!hb@;@Rb!$-14^wah#!F%
z!3uBZ-`l%N;b<U{otVmKz`cn(o#2$Nko_5rAHL$EeJbuzyF^k*KRPz4>&(As$zFhF
zQb%1<_Ittw87Tt*;-%lwqJ|ZM8d*%jrJOH2xzk}o049ww<?nyK#xIjawVpiqe%M)(
zhm{qY4N3CR9kyN*-;Dj9xn_o<6c*hUL!*R7P))8iE{?H*!MNpeLlK7ZXC=q$=^sXc
zLKnSQ+C{*`PxJjoy|({Xdlp`pOr2E%l2<w@rE_#>2m3oUTVK)dLWwXDs?;8Q;$s?q
zd3A}B$EQRR#b1Fhk8}u+hgd<!cWTKFAJ-B#KQKqyx1{a00_aRJp(G+XXTHe!a5V(S
zDStkZ)0rdy2Qp^Mg7)uBrJnjC1_nOCx)ha~5F`sQ#<N-#G6lR-%V%5kU(H%Rdx!>3
z(leg?5G+zZ>2+>>AzmS@{jzz<sdG=`NiiBr;k$&qJI6~VI+X)~$&&|%u-OICa$HI0
zA0qa4OLGCnJE%j8p(sjNcG&)D^P3!{qDmV|MbuK)Ujo|@PkvG`A__Q!VX)%Y;wj>c
z=<}J?8o9r`rfn2m!Pnu-L2)~qmE>!w=WVG4J40_s5j5@l&E6U`mHx5B$;6Ocrth|!
zlSQ0GJl8R}XOERWU0Zwv24{?Gi_4Mimm1=0d9K+r%V)=%XrLAh%xAS)|5wC!MAZB)
z9)6Op99kowWSW>;Xz!`DO)tSu)CRgCarMwBCSUp;y_CUi1PQ58C!ekJ!_F}SQc%2;
z>SpK&KSuWn#9stT@hLtrIC85lGCr}Xs!J%IGF1Rj?lBI@bn7C*ok9Afsv&P0X$vU!
zdhm^NI8>yhV!;te@U_QR90BIUdX>WR4OEQY8dO@HH4V3qHV7L+OlV;_{cg0hR0kTM
zmQG>^-Cq?1$v;$+32&G3=-AZN7@?*55-_E8W~TsqA7_@A`4v_*;z1d?J>G5e(F9If
z67m6g7(<LB6(LpprVg5MMS>O?Y<}Y0CB@8IqGroMiI<6)O2sUiJ5eqSU2!|@Y?}=K
zfR6?`G;1&n(J+|qELlvdbd3+og>PE%q2=RxHEX~MlE^oO+Afq{LB~UxA<%9y4Tng2
zOLhow>$nl4S!Uk#ABZo&aPyH3=`D3uYsrr@<pzMynn^k?5o+^@Ns$i$LNs`s_h4pe
zdU@36^wo$)%siw7%^bdG#UL>a%}92l)#fx61^EDWtdYr^Ig#n00a||`3=<?z5S|8f
zVI)3Aa9Q`O4D!*veJ=dZ3^Vkxre;XXMo+?vsF%{$<=;P<g?W5*BFIEP$k3l8qgZ1;
z&xazyr`nxm=BIshtc;FeLhY!_*H2CQ&GWn(>r*#JI2$h>?17#L1ID~zER^fI1SI>_
zam75r7lPY!WT5S7n-+%Tba&&hEkEa4{-kS>Ksd$m3d!jZBdsANNbnd2dg+B!LRrc#
z@_~K$LFbPpvfrp1QLz?m@BKxI1K&0V)$)<359o19*x{ix2>nhNi$BjW<2dZ61F*aT
zZ&eZ8IkHfQKDvS>D^7LFU$es!6BA2rk+>#WnBAErMKY^7lw;yq1Kl*r{5JW7UM@L)
zi4aa^$<uE+=EiEsng=Wpu1w<%ii^5`NXX7<8s*D^nAmO;)#RKtuD~^>NYXCyXM4hB
zGN=&Qyb@(Y-%g@BNfd1T-MxW<d!do&xe9a(wdJ&nqk#H+jb2L+>y&xT4_Ez^c?FcN
z%9X>`x4A+J%_%w6MPoOKP}aD#3)3fXp}%s-#uF}FT2HUcrT!pCR+sTV&|25#;*YN}
z4brcp<-_Q=t=J^fn5_G|A5L>(z|DJyGC%)V9f0x9PobmDW!Rjl^RdlFW`ui98k%yM
zGld)8i?eOj5$zI`a1_o-egA963jA-ucH-Z4BdFAWeTk-G!-(R99^9qUK@$fDE>OtC
zp;c2qD!@rZrZcvgDm;Wa0{8Vx7SM*@RmM(&f`U{=7R`g(RzpzQWA^8ZIRCzotmpcc
z9m{OpoIGh|tfn7Mjxvn0-6(n_xKm&zJUo-H8dZTWLt6bG8cwxR`ve=HEnw+`{nn>8
zq?3tZY6430Gdr}cg!c-SW7o_!cHDQ#DQK8wyVW$~I=WIrZfv#_bmKQ|5T!3KxD<-k
z@m%}T9kw!%{5}&|-*KlusJr;;Sdm01irN!#tkdhAb7KBb@_Z@7O!=lw1*i=G599W+
zV8{q$Xq(%4J&~UO>occ(scAG*cJNhWI}D%T$jC`jMI80gxwSCIf5rp*Vqs}sOdQhP
zRC*y{1`Ih0yi72<Ai5xAB(eEG5?E$papAP@@B_@b70l?-{Nz7kN{FF>5?CkkL#EM*
zLW@VXGCH(?F|T)>Zg1GtMbqfTz{FYlK-fi*av8FhIZh+>mCmX0?M)j4vKVW)&Un@!
zqNSCvh<O}42Zjbxe82+$YGIQtVZ<o-_rHG7Hfob{Qj+Z_Uo+M;pr061#w@@NG6$A}
znQgDnT`QV<73_~zan*XqYlqU-1q#(A@1|u;*V|cY$($I`BLAK`I`tTZLM^h+lCOGg
z!GT9nVWNkK*U{Tim7Om8<G9^H!wU5Gv!m{l#Sxl$<_xHo0dSMA)BVznJsLX;`nSK|
z^Fjq+ZW|)t9bHqJBtP{tzTZco4%ueBx&f)3Z>PVUOiT&Es7EI=-@66~CiGn%E?dT=
z2bS6TKyCbPK)nIr7sliMtG&CkOxbo{(15m+7|z=rB)zWEqh?QM<i>RP*w_4E9ofDK
zM>j5Tas5LK%C`6mrTX*JiAC;5DU*r<^&of$`FE&qU;=+Y0_$5?H*zni@2P?Sq^g)u
zFK$_KSy^=Xl$T|SZFi9T>!-i}5x0QO=UWHP=ZDRAY*`!e({db_BX<5iz9xx-uNA(l
zXX>&JWFgbWVlIh1ioi|L9te15AG3=8)dIMn4(Kv2qYP|TlIkjwJb7F-bUx~$h|(`J
z5@7rc|Dzm7NDNo-)M;%p`8HHw!$~d-A>uBWBkXcGKEASXe*W}+I;I!scw$K4e%MD%
z==+94vG(?Uk|9&{3qy#nhWq&;A3*c6?KZUsmHT~!FPsska6FxBry!Ky%V&l5>+E(K
zp04Sa;n})lDsx&!*{=P(vIbckkkR_WvVoJL<t^TK6ZoQIUO4zb26(n`r&$T2J$cb3
z^_TUk&P}}}sDRyWxI!3Vm{~N&E?dGpIIk)lpr7_#rO*<Sf1{&tBCp?%a!pZ=t8MI0
zNCYa$oFWv;eQdN?j*cZ;mu`;>f?I7K?KbJiL0G}JA-SnHRK)9MmG_ILzid?%7S2|5
zsdz_sC5qOje?`W-OO6c$B@uqY@ki(6dV#|MZa4O%)!zb1v)A<c+pDS)7|ac3{Tk3Z
z=l=&nLA|~`eDTs{(q9quLvsCWVd@&nMG__gmjUJzHgN6uVYG8G`P`57)-kA68aY0A
zRKc88K`dAk$VEddxn)!--8&v$GGuzLnVCyMn6oknx3~74b-D$TRxqj>zUtc3`n+Zy
zkLSpHJst-vn*M@nR&EC(j`v)SJt9{iKkz8}yXi9*q#y6uv}GRozbzS?J~`#1mbOYZ
z)o#=h6nRYC+HNC0QUjJ9i|e&Et}^Yx4;B=kcHb2d@wZ<R#!Isc5CVqwbl^?*Q0~-u
zW=5_wn}oP=WXS31dEQ^cV!Ga4oW1`804wfgJEh5~f#XQB^>U?7E*hm$@-;4cOB1~@
z6CxC&^`Vqhprc}BX!jQu_DYpfBnn3TKZg+L5}Zgfq(*vFD3$pIoKj^a9Vjex7Dtv#
zxJ`PIQp~S@92S#!L6_)&xx@Pf*ni$!!OW#4T9;^xDI`F_Y8G^EZ(+ud#f%sbz=#2X
z%vn>WuHWOnvp6rcCL8lFDI@`=j+83!<P$%M1ybW9M1!`7)tn)9euxj{;N-<(7AAfO
zFtefR_+}$P3P_RgD}tq7oyD(QERU!M_M|1Ep9#h#gmRPgSJvnxbtwbdVF{_fRcjsN
zT11FOW`6C*O4S4l)&#O}Z6K%ibaC0R3Wg7G5+3Y-O;YYuShLN+{9glE{f7hH&osk>
z_15vst9t#@T0-kRPD@Mk2{b*iN&b34Ue~dgO_T8Y@t(^uLITMM`IZi!rF}l{2+M?>
z#Fq`bOfR$#)o(vmbtQ!m;<opTN3{Bdq>d|RD^kk6OP4N9YxV0L1M>1xvx3?uocVv>
z<p2BZegT(71e@B21}dD<+d)C8#+LoAzbisqdeb?<7mOGX2%x0p<W2^|gOQz`T@fCa
zuw<`PujnZbUn0G2s^FZ{16jGt%9vpnFm=k~{IYU2PNxfp!%j$WFoA&qXqraafpngI
z?j`<q{q>AGXE=&Up}4r1(z0@p5}VCNubw^V)2jy;j=q{*mE~M}hMVu^l<=_DL|tG%
zc#!Or-lCFrfwgt0h5pFc^CIlsJ6P!uW#*M{3$qU7dOriseCVo(kxq{!qrgj1nO^ka
z!lE^6cUAnSI4iAQzu~{EkgVpp3*@S#<zirko5vM}NmAnX-4W0U!DMtMS9&OUVu@HU
zVVe-l(Is&Lf(4i*s7NYNC=`-xBnNbbJ!XY%htq||mnWBsKAOrk2(DA0{|OzSb^@X6
zdV$Gg>WbaR9qer#A(;^CW!wIS`BzfHhFx|x?6PB0AR*36Vh0bAAsRwJQJIO9EDL+n
zEqI&gM3~^)60xQ5`iNV5R9y?D{{8pgw>e>wl2!3ed~D#gz&XczF0F+GRUlVOcthAK
z=ci0xcFf(de|C&m$x}b>+)f^oSjjQ>e-W!#?9C|+cT{+b)6gmlky5tlZZ^J@=cZ%>
z6NC_=N3U)rPyQ_;xGF|c8`{J1)R}!95y|O>hp5u6Q|{91c?5+nzuti7ik_y!mq@V@
zW^5M0dvCtVTkn2Ar;Z)C>2KFJI)34jpO`q|R=Os{H##OGGm8QJ`_ivZZ%TH)N2o<H
zY{-YnAG#Me{2=Sp3KfulFwaZ(4p#C@G|uef=r}0R{*U+P7nRtp3Vpj+$t_a(VO8n<
zh1sdI{GInNFC+n`j+8FopN}jSA6SgqjuYUJS$d%oE&wjjBwb&;NVa%Owg(pqn?o=h
z6Qh93k@RS-1d%^8QdU`g9qmD})+UgsTB{TpAJpB=+|`F(DQY_GO*cg2kzS)!@c7)B
z0y@^bI8w@jdGqFZ0|EjF4i1)m`}S=UYZ<w0nc8*S=v$-|>-qC_aw-*q**d(FHfzDA
z6R|k_pNS2-#0%}i<oIKe!5x@5bc9gMtzHM!mHj(aVdQ@bAsG?Z^=yZYDVm1M>(!=6
zDYr+&CFFZG8EiHSqQlLmtA+<he6g&w{Ez*vpZ8^Y&M7F<(u*s#d+MIYtXIz2B9zkW
zkRjH?_~O+8hgFbWs1hCB4v(kla3;6gjmdQAx7zFVVm6xz4Gp142P}dzmyQj=;mmKH
z^dhC4S$CUa@=lrcb7|LA+bRcU<a<{I*p$#N(U!CByfi$tuuP+S2RN9gqDv(&JT@U-
zQFz{NQTkQ7^u@(Fk&~s|;5{iM>rmxte(=eK;$NDHi5P;i>5elkK+<wyuJkfUN=6De
zH^9s}QWA9HHlk}AXquL0Hk(11AgBwJCL22>;8G!=W!oJNALv9;@avLpDMk#cY@HLk
zd@P+Ii8U5UUDwkTMUgg}t>!OM2ytSU0IFjD=kwFz$6a=VFiL+lTCS@Nk&NIYsi>c(
ze749~D1CmMXY?0f;&U4gZ2b7tPr5|Qdt2gsWo2cJDgcW;=oqy{+zg;=nzsB{T(32W
zjO-rLJ;5^f;cFs8JzmLkU*-)AkBk4i&_!}ik(v-6ZRMKbA-wi|;r{GG_57l&l+?z$
zbLHmdmWRj1f4xQb-qD;Q<MVVEJ3}=>gB6DM3m`eegV}7xY%#uHG|lKyttdv)r`zqu
zW;1R@UauEbJ#?g;*=)vDb&Qm7)`&oUOmmZ*;$}iw08!0ICM}XWJEeDjNlx1Gy4wrV
z)6;>og%FCAQWrvqnH63BdFSYer}}g;=9pdRl_**$D0A)~6VbnKyp6Z!6!qCkK`QY2
zNg?Tn|BRA(z`v7{#D6Q!V3-hGjbs#po-K0)Ato&$UeP%Ph0`RQE?@|TdmJ!ap(i;y
z%x1H6yWPpvA6UYA*9u7hrw8n2$SH%uN^q-&g)-0zk%15%aM&3wQK1^c`a4<qd+UT!
z_2}fGSC^ycU{>Oi<d;LaTcGM-HAA=qVnU$8TR+(*w0F(#nx-ZDo~kW~$ZAqQpSLW2
z?C5a<F_WhI*`J8G1^lE--m$dPe@cFRL7Pf^`m?Zlk9a;hoZIbI*{Xj-La|)Xn&$_Q
znj3fCPNbCUmn~b?ShJ3{gVocid&iJan?+!=2)aaD<If))^xe`;<yvvMj);NeOb=NF
z>XXIUDX9Q;MWnB&l*o5iNcSB>X<A6GlO7(rJd(&zGj@wGgrvztfW!D6c>Dbi`QN;S
zm=pz@&Bo4Md)Tpg4Hk=qtn3^%ZP`j<*IJv^^iMx$)8^k%HH~*(e;J3vj;5+uEfzE>
z>D1nW5JoTTf-;?`W)$FAB|T{HzLd48<U$BDpi3$BI@^`voV28+n{BtB(c2LR;Py&!
z)v(a77yec*x^=K(HVOK6v9Wc(vm2arZ6A`UQ>6~9tonIm@{b}8jWZAo5`tb563hx=
z4=xn8$lvf9<Aq3Cj>%;vSaTAELX4u)359rFr@Ju3XdeYatKa*R9_!XYRaH?|R<=DP
zq$<%Sau6J-Q)|B13E8DkS;rEmRc$YJ8iKNGB-nkdmR=1iJcinLSQ_lVWmGAvwmKSJ
z4A_8kB<vATH@f{4jADO>stA1RrHynbbgRbm5Eygv?VIh<Af*6GUG@LmTwciXLO~^H
zKFKzbgK7`=c(wyncea7HW09;Tb=LfE<HwFhNbxD+__e(|OfrznRHXcP-<R`G77#zq
zJowy(0~@Cgp5CQB_cT5F$Qb?DcU@xpk7;#Jn5SSbx~J?<QT7L*F#EvnwL20wca624
z;_^tgrnq(sL6=9bi>eX*nFT6RS2u>mcAey~iNV=Hj$7rnZJzMBt~ZCq#7_ZwRgW#s
zOxqY9({*)<q7F-FBI#zV;$Ys)67r=+!XP6KUQrZmHbaC41_tuOS2Gzmej?XheFb*A
z4U5G>MrIZnnOV$Tu!!rfz5;+1tA1t6)@@7}e>2mbeI6+#4!Z*om`o;g9p3n+fYXXC
zTp}zaHX|X(gJhNNuFw_FM|HP{#U%U#sb3ru6LWEPc6LR0Y{G3F61qN_zJFhjjtPm|
zsVXBl>G~+v?Q#lDCvC=(;x1?O31I2wa+fZ}=WtR;9v07EC^LcYs(t{NyhQAGn24`D
z6ZhmLLPMaSW5ZJweWw}e+Q^3Wo=?*HEL^y7`-BM-yg~>IQK#46S=fY&xWWTFbD&3i
zLsY6d94Laqih6Ck3A93&2;+ZXa1R&fpH|6F>l<p7izGZEp*=<l|5yb)KsVqm9jw5w
z0-hI;;dAW3Zf-6b;!`8vr9xUEl=|dXggPKT+^{&hRoImao<q1#jD59#p#gCa*s11W
zadGixT{nbgSy`FbzJ0r2m(Yn$>a6+SCXBr_ACy^~xV2V25>_b44=T;-{OnH``)Q1D
zlH<k3-O6*(VfuKxwRJ4pp&bUDcPzdvs@o#+Cy9@gQmPRViRZrgL%}#HaTPnnEcece
zu9*eu!pS#8bq26o1d7Rgw!<#e2^WXm_}2dl(;{QLd^Wgy!28i*W<R+O^2H^(d)Qq~
zB;E8D7J3T{I`D3ulMXQ!gb;YET-LYWG7bQz%Y_hvYp*gy<Dwsz(zj19R<2ruuIng@
zLRNMTJ9q6SF(H8;zBa-c%@#8jiy5H^bX^kQ7M!m%KT2nf?x7rgdTB;Ry`-YJD5dzy
zGsA{`vZUn2@b(Gs59#B0KQ7Xg8X4QAPk5*$zIxoUEfsRXZ)FopauPn-q^4ZIagVFL
zW>yxIW~c0dlR~nk5LCO%TBX3>Nt%O{lBA?0z_6UDsw%F!>Z*${w3>zJvp_L}ei)UT
zR2`D@pj$h`qURe&2!`O?m1B$vHU11Hk1OQ&WE*+KMi8~Y5Rl{QNUsE7yo65#Osmst
zbC`sg5*|^Y`c@z4(3yIB0ZLsDMx{rxJJ)bg`%>bB9UkZw1t|p*PZO!&zQok8Jkq{>
z`<uV|>Z`ps-+c1{pOIC|q_#uKtOdU$UO(nc-J*Sf6c-)u`P3m(B0fi#)6>41zw5As
zdJ@XU(3Q~=(NjT*M`-ON1c9g&$U>>hG)$&lWkuf1kyA&u+3qDL0u!IxaA4yngForg
zj=Njq{ANecF=~sr1u12j$z=NNSX}Q|P*}LH1bAC??`{jL^!6!?4BjsD%f`)LSGMAs
zvqOZUfWs<Y>fFx!(kb2Sbnj?=M3?fSyGDoG7yMRUQzNcAUa&YPb?zTA@mtfSHlSlO
z>NX0DE-$94yRGW2|50<h>PMvP*twfCh7MuIo-f$2X$wP78_ZQ>Ml)yLLWZ9|ikF}J
z7Z;p25~)ipRq>WOU^|Q?Q7@7orQ0JE(~AJ2Uq5Sg*0#$fBTnt%pu#05e)?nah^vPO
zTXv?o1EesG3$!WSRZWspq_TKp*@q=LX&;BRSNbaooAWbMcGkVdU#_#wN&d8mq@*OV
zefxG%R#qmaOqt?HO--E{85#Mh5W?sYZ4HB{)BGPDVuw>YLcdsu@UbL%HQ1HyUqn`;
zOW2=ZlldGGqVdXuxda6mBJXw|5n0Wh66WYeE2MhU0bbJ~9<Z7v&)t(xmzcUPcpdiU
z<5xu1o%hZhaQTSHP&@RAfx#UivJnq&hx+=;YPH@tY}l~%n>TO1)a7!Cyu3V-l$0dE
zh%IV-5+^46zMgX+GAQRVkaq()$8joNAy|qOH#*JU9;vhDKA!sJyq(Pvk+ZL#ZQAh4
z%Ap%yTlLb$uvJ?TV!t9-hR~;jctIeZ5%3&{_l2S?7E<mF4m0oC{QBy*H@?35+>P&V
zY}rh+zYyzp%jb`_+l|ZRYVf%cB1uuqwpOnnse7Gx3U&jmTfF$sXYmfL@bpV)Z{2iR
ztjw~R&6oYQ%el|vm1GsDj5#yJ92so744`*soAr{@f=@A<1i=oaLqu%+sDz%~ZtdNt
z+eR?jE=q-|bM@xG)rz3Tn51d-D`7|^PN$QqbMw>B%UQO31wny<%>M2N0IXIkGpE1D
zV-MfY%@gin#j0O1nG`H$v*Fo<`UZt2SgCs`lJbMXjC}_H!U`&$7&)xh{E!Y^dq#$u
zZw46NFTj-8!P=*Nn3=F(Q(Ro6Y1kEK1+#Xilizl`&^1vpa#+tj7Yzy6qU++}2G6<q
z(>cjWPE4wW%=YcuMP6Q>xbemttvBC%^Oq)*>1qI^1_%BJYaawxBc(he1Hr5q@r8*|
z5a-jP+4pPRVX3h`)}ZQ;UK_LCJKn>)kL7SyxW>drWPt#p8hKtpz%+->D-Y#zRzGL`
zomr(4w~s&!wn0!6?v8drwXS-T0#S{WDoXhpe!U0we+ingiWWkIgoTBDv1ZMh(Rq1!
z%J%KsMN(3dXoDVZ05-h0!J9H`{(DxGejuI$3XkFCvjx1Yqnwg5YyO2PU(Efgxi0K2
zuPzVU@Y?DL-G;@jQ53O8z@q~Cil&o@ECLdxl=leaQsH!N+4TCViECb2b4-eC|CzY^
zznfF@ikaTh=e_3Xb9$RB0UcWUyuP^ewj(t!z>?O!*0Dy2bIm1L2ev2*dvCiqEaItg
zQIpMP(QWa@a{6?!(LUT15*2Db3!tb>qkn>ptU{IEoo(2yrXL=-JmQnyU2J`XIL!i3
zoSFK+?b5quPcwRuNI(}=ZW<}keOe?cih{eUB+%~Uy<B?ng>>%RiN_wfpJl%oosrl6
zwt>{N1Kd3J2JRStGppCEtA1igB#I*NXwAFlRw7xfyUQ$^e71U<-9qd=V<IjcnrPe6
zy@NF|uUN&aNxH{bDXGv%h_Q<HVdfG2;%#h7c9NXoX5#4Zhab2yGBG09gcN)sgb*<?
zF@ZJDt68%-$w^LNs$AM4FE3Bz<>e_4KKS4>CX?x60Ll-*+%w>(D`CeQ^|Q9?pQQfd
z4xtbg=znam9s0%?f~<3R{e6X%^$L6Uj`#5Tv`oyyDh}_{CV)Xs`X^;^Mqi)Ru#}QK
zpC&W0A<!!tdPg_3s8$ndH)1(`PUd%P`a;<CE-e2W{4fxTcU14O1_uXEpEGAp{Qmv>
zYs6$*v`SX9d)A!%lv#8CVOI5C5*`Eopk>?JR^yWJlMwP2v!eIhKWpA&X|v{RZ>|e_
z-ODTcY<zvy%epY{R*)Y`<PeVZ9xf3;A0+Qt6>a^-S5{m^8-5Ru2kZ987u;T)O^<9o
z_NZ<ng=pd8wtjGB8U3|YRsB=6MXV;gL;NWbv8g+PJ9JG<&hS2+lI0;b!px)BM)K39
z3VL_8(W{dUt*S+?UX{DBFVjO-q003mLabe)Em%xg;^ML_U}Sa+T|WA)=4^K7Y$2+u
zTGaVumz9^3pI<;$b`DZXW_<Pqy?gZ_EHsocqeqdFdVqNg7W2op9o+NKV=P?!6I*`&
zgL6lmjizbnx{j0*i>YDnh{jl;yQmP_^SMTWb^xqalS2_O{^BsDuuLO3z{JKqPP%ol
zvM<X^QJKcTu6E2Ofs}A&?*LBkZLBrd?{O{;P&~gy#wL8KTLNCIKIcgxd6G7b5Tc$-
zd%=PQ-K|#ZBLJx7u;dcR-_Z1lO8mGkx<nY3M_=MXWKA%9ba4GKQW_c1b>|qY!=3r?
zY$jY(3hfWEvJKbZ+;Vs(3m(peC_iJ1ow8m#LyM`t1=ZJ@3Sc!s*Qkc^yT#3L*_7;t
zMHhmz#seJMrAwFRs$JaBwn_mtM)KDS@=|BbdpR|$VzAJ~U|=e+0w_7cQ|cfN0<)!L
zf~tycsk7!?xc~F{pC!*+&`A1yeU>LJw{LlU<rSM=U%l93RyGUC-!XEvS_}bwg`&*g
z^xDc-PFg3M&)xssoSI$CjFvp7*&1++>PK!tN|~FNm$&s;T<=(+>x$*?*G6_9*2DI@
z!=_C9eZO&kt90qaM_Y&pHX(!&8I+viA=sgCFi*whku3kcf+0Qa{IsbYi&^0D1lI&b
z7UZO^Nk^NpLO*oXErF4E;#H+_PposlEhs4D+N;J8-=#ApC8Y!f1=6=qFG|bGP!xr!
zPd-k1MkZH|8O>`i{+EobES{Y<g)@ephR5qg)3lmpzupXsWQNpOp?mfg*~H6rx7*CZ
zoRP1x<A9sQI4fKBxk%3R5Ep4CyHI7`+A{ja8=kGWNDE;>3MCa9xy33O`D%|Y9fDFP
z-4J=x(4Mspkdw(aM~T##A{2s;_dHq*g*Md&uR6gdFE3BHTrLqC8+#96D7&pMK=IB)
zpP-teN+Urfw9faaB8>7sJ96q9?gFg!<ygBQBb&RLoKk2X)Uaz#oEOFx89!wv$TdTS
zVys#NbZDnSSTmNIS*7(CfRO?9ujQ~9kM%lpj*m24_jRfC=;W3+Z!uY!2^*h;;YRfc
zlgTt@&YU@2uD|~J)GDha)22<MEoO52B9|>wf#0is{+TfTsxaN-O;jZPq$JU3mntJc
z1VeT2W$6M+B_bO^DpKw?>HMKPDh?b;%fiiXu8xzMep^VG`F35BcntEH%t#)xDcXR|
zZ>+rGln2hvYwZjE8CbX1{GaGBb-dkj_=WE4meK;a^wKwM(O2Kr>a9FAXA$1wG=N{W
z)oGRYXTanZyGzP7&z+Zs+cjODJ7-N907c~*B^5?d@XvlKX6$*P{NFbPoY6ag>_U|l
zTPrwsP#{Y;S1`1fgLONd=(<+#{ZXZ9@;u#j*+5~53U8zUxDM!wVKMZ*6UKMx%(SPT
zsChhP@}od?%%j4C_uW;quU%9mJ$rPkJ+G#tX<Ch#lqSg4bOMDS&`)-+4$RUkrB}%N
zWlBoD`;W~6o!eXZd2<EVjR@i8Zwj!Q1z~|E7OgMmj!|K(+E&Twy&MEN6gKQ}(Lce?
zuiGnm=B9QD(cxx1s^q=7p5Lp_c~VFoB`;hezc}7=_$yP_8tHbsh1qN#W$cl#=fnDg
z!yFJ9AgS~U3Mv}51l(T$y<?0b<<$gOAua?m>KLVb?+}rJ#(u5J@!`=RzZ}{hVto;;
z89#?hu~Smvsl9Uupq<ZkWX-?ExOPDe<2k*WWN*G`@LbG_@q7ZUMoFRkI$?YJzJN1c
z2Xi2RDK<9t5|_*MUX@j{%?rrJNSXe9F|e`fr`4D}d$uVtJMJtbcOm7KAR39SXjRUY
zy7+11t1B-#_2IL#j>Gla6!$*++riB<27lZs>gelA2Aw&ypULWI^){aFoUI@=bY1-1
z+SfYP6c+9)x#avlOU~#MaG5~h@k(?F{Sxdf*=XF$N-K2i7KOVm3+L<QrF3j(A-hQB
z)E*9A{wANx&Im?Pz+x$@cY`j=-kTc{7ys}qy>jNgrl5NAZKL$CNO2PqqR_XagMv~O
zkj=U)+HAHaJKLDdCR9yBRaJBeXPy<vf^-jQnI5jFv@=N5_niH#yGWJVjFPOB#lGiK
z6r)RbNT9;518(9X%_L`d_;FJ?H;xQpKvz54Qr)PU@!Fl$*H|m3=c}C3&G;VKu*c~w
zEq2$GK|JZ&K1rL!H!kn#r=JcILUaZUH*95{??JO7NsNMsKtaba=pSdqH=62v1XLZ8
za~lSr#D>&&Usw7p;?0U7*6Na3dgzSmBSA`8H<<@8y6@P1)`KpujxLSY#%h92VfFXx
zx+FQTezJ}tjJVJNRj*5Apkc+Vdv47MCAH<!6-DXecDqHLx{@vOH2-Ssczb!!=GRtF
z=$aYx8@gPMl;eQoAtI|8AcXjS^$V;2Jf4<TWK%Lnoxu%Y3yV4LSk#WIYKtgU)%~X^
zMDEMlU9LB_?su^-%fsfqE(%Ij1P~gipi8h?1iRDSv<o$JR^I>?tuLoroR#cC6}P9h
zv5mvw&;UAh>f{KEiN7N(rt2LJ;r)JxRKN3eaWB)|jK40N$8K%MrKbngL_X=wxF=d|
zR`e!%>IxySn9YXZlyJ^~0LGpd${GCv$d&cKuQH^@eBFIOTJ^t&$HiYA9+U7;`}PSj
zHBU^U0ywM+>-V@A*ws#9na069m9zQ=pzE-8zl)GS6D1WI9ixozzKnb?^VgN*@<^6!
zsQ4y7|6on2<C7N2leBrN-P)ZycXm+oH|_HQmOgiAMHumo5dn}|Xmqjxqrms>TqAL%
z+7&DW#77vVbO3V8>$|`KVuK;Cyw)n|)gYtTPjbsqkX<HmdTW)hs1kp^<d)Sw`A$A!
zS(mUUPp86DTNEwW1__Z3yWtqruZ&vfWmOdAc#562&+OTzO&J|-MwpX8b`~dA0ov0Y
zcHT)!O1fq0)Tyo5Ilk>P;n_`Tn`aE3e)v|&?)4rimSa#qQtG&}w<4uHxMIbMy{&$|
z<3Mr7{w3XebzA+!4N=1qVytZ4@8<kd%ye#NA*V>CeVCbVSC;aR8>1MMXeT1XM6XUZ
zb{{mvC4fgoOHGZ`0CcnEt9vd9yIcaF{aA8^=5>w#UKc~V33JB|RkbC@C=Ql<Q9`8&
z;!m|<w<(mCm9cI64yJ$lIo-P@q9_WA5Lj(Cq^{#~yNQg9prD`-o6W|mU)Q3lsxhwW
zM+iY-Q4w?JFJfOxDyI)Q6-9v$<`rSvs}m<ooLQ$Zxf{vHT7~98-lBNLJ=Y8m{dbpW
z3$uPMd89*!SX4(o001BWNkl<Z4t+8*GAf-O+yJU3sdVe~>|{gN;ljbj?fl10<y5*Q
zL3RbNDhaeJlvV0v6{_5JaTqgxELJn~J^!lvnx7PsCuy_PEkF*iZr!@F5hF%;gb-F7
zG2n=<O%Ty^NH3Oj4zK@hZ81Ujc7`}SSY!w-r?*N>wyzaoK{g|cJHNsZo%t2@v%1YD
z!v%h@j>Ryy9Kr)6A@-wJ2u`mgv!woc+6Ogk-&p3-$*YQc47JzK;MS#NZyuDnYpbNk
zg+gqoe?v%iiHb|Dy+@er5L(-AUf1<PRaHfGJ-no(B++K#B<~s%Kv>W)6wXB#13`8M
z90;K!$VJNCpsYn`Kl0x%Qd|1+>)u%2(W=LOEEIWx^j+kfFvzi^!_Hp<e0scRa-y>-
zndjSuaKzniE`9I~Tg;%_&ur;=`J$F(p;qJo{M05sr$<0!$L<|<O-p|D`+{Ku66`Eo
zUruC*iJqNoY)f&G5Mv=Z!_DH26<j(rh<@=l!h%hBygHrQn|W_R1?5_}sa>b64}xR5
zJwCL%?UErq4a322JDry<Tiv<X6q9<>$6Do2lgvS8JAxj<OpmJWHy@=s>6B>Y)i>T{
z?TQ~+y=EO-w{6Gea--`yLI{KqC_->BBa;=Yer41Z*D>zK>#$j^M#&%vs-|I56vD&9
zxckoA@p?VnanD2Cc$UDOuXCHJVReqLG$+~U|0>E#`K5c$#OFs1H9Tc!_6g|n!wT=9
zu$Y8S;b9>cy}QsufLUWef{h@%LXbltA=XM>iORNAH(g^blvV141S)J#b+dGH1)bVi
z_+fP!8}>L4YPuKz)FdFB6p|<TQ{Z;H#hY)w>6tWX(taUCFA&iERwLo2nw&DD3rv@Y
z+SUWrv=1`k0{7)t^{w{6?mXz(u2xKT41*%4AxcvVjabRLF74=GLtr}V6b?@*fIcx2
zi|+*w16`Nw%csU&4Tb>rwMG^@b@mr1h7D2FFHIz+WM7^}iOcxRn@x}yX%u4jMM}xR
zVx63_`uB(UYrs<bd|fVA&5fkZ#7R!Nr9CF?A;N>l12IUd41smt9!Yl)!$97SR9rkU
zZW&U%OhET#w19PP)9b5FwMg^#5;`^7ko`nJA`~fK-u%YOg(s754$XvTH>LePWAG;(
zqMG0BMpd0Pe{_NEP9Lh6?IDL6-x9cTeg~<e>*7z*7SSKN>U}Og);jjVD<cRAFmd7F
zAa<v_nY*TptRj^OmxM8}tDUqQ51I~9AtquY%<RqZ&~xZG?wW8LQBje~`i+}!9eve}
zv3oP6X-~SxxP@kUR!g}|1%3&O?fT@WTICCOn(C*!=vC)juyzHbMqffq`v^jV9GFZB
zss^eCQYe5REu)Z;0|Q9O$>XP=mJsAHVYQl3guw0bqH7wOt}*uqaM%=L0}JUG6+(GM
z!`sj=x|_Aqv#=yP^;v*|Qf<ZR?Ukp;M_bsi$BAy@_WLK^G2!VaCy}3Dz}j`|dFPFP
zai&LSP**#OfM2#&GHOT=sW~2IuP7lsUnM-q#7!535)o`dN{9$HCCpu0_CjRG?yC#Z
zlhfg3jO0n$IMs<Z0E-td?x`qBA^@rra(-=iKtTm;&uZA48Vn)XC)#lJ0#vwRPhM^I
zv!%-F2!@DEErguXS|w2J0<U{l@o0kNJZcJ;9WGMBz5;N1>kH8?5%pcx1r-{}c}k6I
zy{@%VO3A(gNpY2k46qo_)1Qc}^l0qJk>r#$><3n`8Qir6-2(#yFICmUY`u8ANJ>)p
z$4&R5s5^ltfmoUskY)!)A><OAd%okbab1pP8e3lbb%=l;flf^hJ`oTiO4r5Y<2{YG
zz=mX=bE~5Hv6At%zuF@^p4aj-YDiVtAL;eFmLH4j9S@4KQ*Sz$<LUeIHw8<w3sph_
z6^8V5aQ~GN+<r+IU;k3d_B1y`dOAqU@o>T5ApS^k(`(p7rc8c}>u<iDoZLJH^zVxh
zoPT<6hw}Va`O9ATx?pON+kI8_bxN{RUfQC2XU+5JdQ)Di5nxxi={y^k4D`?=B8zab
zk7&zYVyt_KRQ3=qcVTyJ=V01it{xsjzb*=qru`%Yr_#x>mkzdMx&)@sIp_eLgEEM<
zrxUImAU?*-!r#iV`sC*9k}BWo&OKQ5aZ(yVlSSE4Q>HH}y8oF^@_rrC(*aW4e8uIX
z&^3)m9)FUUm}p*o<5OlV6I^qS@x7)9*pu#I{caaMI@);Vrgog&KahxE6JAyF@uH#w
zbAK(lyeKO*zFI`WNg;WXKM8yH?iJm;cYlED(mz|0AZL~T!S1R!$drPHi8W>u^lWb=
zn*fx%jaG!!GO&Gc{aC_-C8A!LogiDoI*t_{L2{nftjO!SNML_~q}W+s@U#npaGw~>
zgT=LD%u4+^q!sBDRaUvX9Y*}8Z#<-?OAZui?8s55^!UFo6z_nwkJMhqY`*-*AAjst
zw~WaN=`EW)A(+ZN-y-Fk2yvuAs|s)hX0evZV=p|C6Kr^ORaYJTe^L_O^ym`@0e5YD
zd3j?Em`+l~gl9LU<rm9Onmaze&hLuZ(dzXhrLN-6`2!)u)|D$)W;cCA+a*6Ub?5eN
zdq=%BxA2|pLY4dyjZYRAV>1gLy*85eVP+~^I$l-s<?>R_x%}@u`p|t?EM@`&1KG0m
z58nUq6F?U2&+;OKTw0Ws^3T%r^qQ5llu`y;)H@dIuC;6R+62r|s-4aMD`)BKQZ{^F
z&b9>=9Q?^i%6umU3tg1`?51+L8}k~KA#MvF&o3e^*@Jhzm&`>@vKKfxu+T~RVi$Qo
zx+q@i!uhiY_fijHg_lm7Bopd{r)NrykF<(Hk0Qntr==Cu&ToUaFe~*fp~wX}MJgb-
zj-PnHzHa?SI>g43*foKEeS7ovyPq+0g_)u<jUFAWY)^A>&*kBq*3-evrNwMYcB1K$
z>5EHt{ql$N^up}4dA?y0Co?Ti(q>6YN>a9O-!As=->+D$R`crBs}BkxBGKINMGScA
z#4B1%#*fQezrLbbfevAYIP<kn9VmhvqYfEyp|!5$YH{h+4B^--y8f{e*q3Lx!Rrzp
zpwlg?`HQ!@*RLkXX2fp#HX+%UFDa}Pm=z;&Cag+m=9cTE6&lx!41|P;hQ}6EXdEmS
zc(wY8JPwQDhpAg8Nio9>=|=o1;BvX1>fgWrOQE5mde^R9^{>DFTIWP^aVL!rqtdeg
zK}*J;IT7+^{yXd2<{q$P_71Bnv+x(-j24{7F+~1-@Iz<6a=d48qVs;zX`M$8lDlnY
z*=VffqAhl2EDxljq9WkeUw?HDpYTF(k6{nwDrV~u#aA{LPvKs&7d(KJ@|7Qd{PFLt
zd94!#A%ywVQ+lla$KRrcR=RcGo?FDFLxSiVZ==$sGkawPH{AXVV{f>&=9uE*5@vq+
zH4)+A^zGA&J<0pn`o}iDnEl<)dAY@Rm*noN*B&i6rdxYE`l>t40p0tF!*5^jLSjZ7
zQJi6+)cHTH+*vH-#f6#sSNhH?G|2wi)mMzZZp7KcOzongC@U}LV0s2uT|S!V=-R#p
zsc8q9IPq4l8B#$+h?z>4&RcT|xq5gAJvv%hw7y)My}azyqMZG|H<$z`Es`f`qg0pv
zsm|D*K7D#;Aw(o#SkCH-NUI6@#~N)7Pw8kB84t7?`&1p$3Snn9RQQzj=^Spjs!fU^
z=&I#WkPQaL!N3kiOF~}(Dlodp1pCw|E~-?>DQjNw$fC-IBGO?o5`+A8vP+B-4C-La
zDXfmQGN(+S2qV6;Yh?XnU0$8txf=TmOf@3XZZX{Er*tr0uK{sJEzN3l74p~DJT#ll
zr=GCx?WF6iI6Ysru!sb3AjJ%x7}xCL(Ulnm57TBvBnTmHJl=C?3rw84;Xq!gJggqv
z9ea&Av$aJe)V4*`H2#!jfBYe(RQG4OlU8l3B+#yK@8#iqzq*vtN}XBDE1B@%^Yug|
zKv-BP_uqXdeR}oeU;llHO`Csb=k8?gzGM7(XAd9pTWD<Jkm|7!5s9H@wD(jYCw`<?
zWF$AEAGa^|BNUOP=)9>_c#F{QmAd?UNQZ=8b?4PBzQgior#^XOVPUap-h#z^_{nED
z>~`+HV|+ak2@v0<6Q6uCi|;pukz1_t?W$6)JS&(U9jz3WY5csU;+w`qB($Amo3X&?
zW{#DB3n&0mfDOPV{P)hi3b@?Q_H2wes;0OXIL*)Y_keXk0Pv=t{aL`2I*)GvE^gvl
zI^f1=*OrNqG4&fzcBq36aqbU*HBB4~5DvU}i2WMy5FC9SwZp+TV>>i7w5CoLUY|bH
zR&#AUp&|^+Tx6hOF-$KutdtcVLrg}87?#KCPBHC*jLc`RUSBW@VF=X5P=q1+cIVa?
zsOiND;SL?EZ=0DqXgZ|%G#Ij(4VU{N4#VPSK0VqAOiKOZB#=|;C8NZQE{z>llVO>S
z3~cy3?dJNmGreVXc8ln`mrF`Y5>uy6JuxhjF6%QOFKVT;2O^otq+3qo#ZPk&HE`pr
zD`QQHc%m&kf20(rZF_B9biapA&uMz}Uz;5Vc)ooww^=PTtKFm}AncgMM*@{ns;iI1
z^-dHDvr-qtc1f75>GHV|0|Lc;V<MQf+{NQhzs?2cj_`X-dPXMy`tOU(`0#Clf`TY0
zDCBRq-pNx>JQ|krz@u*gLjka;{RVdrxN>WXtEWrlns;Tz|L!pbMt2jfE3L|qysnmO
zIYLfQq_}lxAN%=hce=g@b^+)%{jNtJy07<)nP2ebrymg+8G+mF;fbfG;qiF5<f022
zoHH&qhR;5o#rO%gaZW#-ZgEypvORpVr1+PTLi3%6nonC1l7leB<dKpNyaBw1A;zjH
z2FCjtuUCyd)J;<Xo(5ukwrkXiP-`RS-$i{PIi!(sP2mRS1OLVl%l>HW+&|MiAvq41
z(8QSKz%lLcBZOeuv}rYubzQGbl@9jFhp6xv?&u&N3t_ke!t9V!4w+S*X|hX=M4awX
z606yeR3@c<VoP%|DbTGQ?8vT-#nh!Bvs4m)h)})U^7^rx)uPm&P;3g{SV&cq?9G>y
zxh*y4jtMb}uKTuh1f#B|Zy)lYgqmNaG?w7iC4f#C9v?Rdg~wa<8iG!k)c?}d=R=5l
z+ge;?L!{!7?qR@pO%DBwGj__Rv_EDJ{wOw_`<pY~WU)2;R&QaX=E0r29Vz9eWy_Wo
zH+{sPi>$Pi7s6r^(pLT9d^IE_DCR#eyvJE*4D~y<veL=Jlm5XgFaC$1pdfx)_6zTQ
zIGqW%-NflbPoqbV?gPSNx^ykcIglC>8z0=cy@i2^_TJN%6y4$0l?&g|Dt^4n6d2v3
zmBeHTI&Y}uS{Cwl(A$kZBjo((a1-lxIs=6eA~>pRcf~AZaBwiAM_tUl4?o86v(Dgw
zdnWSYfBwm>cic<&ZiyszO=xiLc2SXhHe)8^C*010b=zHkq<G#gw8<$lFZqy$qQ8o3
zyCd*C@GJF&-%7xrH9-NIvSge}LlFtE8pEP_EU*DrV^}7qG%@x_xuY3l4!K%sEw*pp
zUL#&MY}k-WwevZ!6GWa`J5Wj^`?_wsziNblm|(;8U7cW4>4BX&qCq0gk>IbhOhJV>
zWGjO_AB$>-(E6Qqjuc%=cI85u+xXmvIiO!`eG6<|ik*^OIkiRB!FX+>YF}Gj*Y}-}
zu5Cp;OLgK^t0Knn#7+Gh+q*nzxm|(VT7MqL0%gGQo<Un-%Rc2rk4ojC#&|p)G3K#(
zQ6@`3zt*0QJ8vgab0ekrv9+&tqEeil@>O<;^7AWiOm9>~>bm6NN&n!{hwi6yr;e;y
zx1L#F&Su63@9@Li`KA3%J!|>0l`9_#3o^ekXi)dfCMmn+mud{}7r^koffJ;nJ?KKN
zep{<Z-zvR+uW>ZwOPyD=a!(f8LV>)jR|k92n6rW@F4xhe>@a9Rx8GtSmCu6B-Z#hJ
zdEc8~eErS-|9k%5D2l?rpMM#%*~}|1{)fp=B~f1C=lc){4+~|+^y!pn(fbRtQYQKK
z>h{N9MMzdN7?_7)DFje~IgO0Hq=|D~*2wnbv0^s?&(b`1{NXSZ7}$()BY_?*eC@;H
zzWeSg(sjMcDl4G-ZS@Cex{;;5BO3}U8}`e#n2Z?0-qD8T(xX9zyJeh_k%9GF4eC-*
z=Jr3<={2&-t1&Bw?%91fIK2X=SKzQfk9LNY)7N4tffAP?IFs`lis{60pw#NKva+&e
zEi<_mWb%!@K{oaIZ7FQPeU0r54L1#wf)2-MF0D=1<2`}4!p&1R94IQ|<EF+1#U>6H
zCbO-naYsv)TN@$dpRT|5;V}4=0V7Ts*!X(rh5vhvbB3SAnWqn7@4o%K^7`An`O1r=
zrlqlH$<lAK^NMdMu`$DBB5U&XQ3Foz9q@X;cpIDdy10Hsh=>X?-*Lmp&?{6$yYWM<
zV)ts@)9AGhmjhDcWv#3tUzgX8IV-qV*Es8Se>*=UXj8J2zVSBtC)l6-$JlnKIcy^L
z;K75=qU@B1ckkW%_;)|d<-WV_z-qPf&ifzJzFibgJo*riJ@Is-RW?FHLKr>jqCO$*
zx+XSfZhti)35F%HZY!(Cgx^IN0<AIp2}%7B=Q|QAFx>COKqXCQ!~PThF`Gw<KWpTA
z8pcD4M#dgfq?ENTo78U_4iFe_?tS^NHB*vNQeU)zAtHN6L!TH3akQMJEzBoes?zKK
zz4Yzl9b)%yEj$tetj2Te8{4qOqgRt;l}ff{!k%0~nah8b#nf*>)><f4RXqXSWTwAy
z=+ABoBd=uAE&kg@N^wDh_MB}Mg?7h#3T=svY5H?s73ZPGSR4_p-xd+0zqXXJSXI>x
z$KrY?DyphAnho>A-1(H3S8&tct|z~sfJskI<;_=LARxd&?_NE5{@H1_-+t3|Ib%m&
zSW@Y+T$fp>5+7p`F%f2TT~bn^bMC-EE;=>PG-^o5Gg2{K_=Q&a>;H7;QJ1m!U3z&{
zE6*vG^8CR)>}QWUEx6mzo(>{IOca%Cv=1{A6K)o&2I+a;Q~IAeZ2gV5+);7s*c)e$
zz0pV}n)Jv6>`zVQhq((l?bK5^eaNZ2`R)geUbCd6RJWlwm4gAaofyf+_yQw7ueu+w
zz<5qwfop-zjU0C$@Oss8KVt|rAO9m!81aQ`{T$x`IK|KQ^}xZyc>+x^4)|Zy?*fPf
z&H)|<&h|6jgb}xSMGIUz81R>=`9{_GukqP_EASN1a!ar#m_2(oBS(&`d8{ak;jd6D
zAg4||xxk2347P*QW4O2rDg>D&kW~U9c1d(Fgw_#`fps#KTY`qMivGu`J|1uX{jZM<
zLBlxFa<@))nV{Gykh1EpEYN8G7-+3+7g=30sWchNQy|z_s0IfI4*^cBE*YXpZ$e8G
zjYS`X5!O-<DdkYVS@vxay?MzaCo#BX%9f0svrc`#U8sD>-+q_Nr6?A=|7?#IM0c0s
zE!>X~V&$@B%Z_PM<cY?{En61t+LL@&uO9XLoNw8>jVGV^7fa@ShtuWazK169+`p#M
zzMZjl5<)O>!mZWAt;u`$Ss#C5GJpKBg`OR4^o_Uk@uDKSb+D3Cq|&d8?V{v#_Xo)Z
z+UO-t&zo6NA9t%M0LNjjDJ8H-cQIdgZIy6EWQggG^G^wwv^Ud()gt(AWf{Z!2Oy>7
z_x&z3sqoPK_jGvTu}3;stqmi4p8MBS?zrb6Vq>BidCqVyzjiDG`t@bR*~99MOFM9o
zx$_q;%FRt_R+wH}w@9uA-p9y5ejg)S`Yqs5pcjo>2fG8^s(!CdSn<cGMz)W}zrCh4
z-P#h!ue!!aps|{6mtqLcqvb{nF<F<=s_TE|XMB5FM!+_kY15|FEcE{U?|%mfA&mZg
zo>Hh=hQVYqEP^Is2&wo8=pR?x-mt_a*pnyNl_MKiC|e56XH6;^Q5U|}NA0LP@-;Fa
zU6Pz9*_oqIRB4FFhydspXSn!7>_)OskgZ`qZ+C4*xyfXzm$iM|Gq(SJy#r|Vn6m~*
z={M*~isKvd(UJ&p0@Vj?BWz32Ui51C4MyvD%7hrjZ13CJb8zKuL+W0nlqX&o$}z_2
z)~8Oq_o0JX*|qv8CnuNj6YtxZlbL<i4L9AM_P`^L^T53m>Di-OgX6!SGmkav*3-RP
zBHw)VHItushItz#S%qGTD|Aw`y+ntbnX|T>znvR)aeJF-dT~y~_^nc%^rBj>X2|Bp
z-B%)cU#nog?)uuSRSdIQ#I?6v5ISZ4x^g<Tw@_53vGbst=rA+ub~%ZPO=R*vKjxVy
zAIG1F1TdLQy!pxty!guNjK2EDt(oZu```1xqrascsI`A(XXkM9gu61!$}5}Q?z626
z$wT1;ZU>4Q**~bx?~Mt-#w;8D+}aXoc7fufL<c4ToBixF11+)2_2;Udjp1%zh#_iy
zIW9>z02?-J@KWviw?~1cjs{jWBNMtVR+Awp`^MI~oy**UgRSJLE%%9uY|PT=yS==X
z47?d7)O5@Vu^TS$#HjkQoM4nls_SkSCJ3r$=y#jV<{E*Kl$3OQ&2MH)M`~lBj~$W*
zu|%n#EVqMb_5R7n3MnZ(-g9VcTsLJ)Mp>nN-`~FA-or);)1m5lwG7qqk*ccxWfN_x
zDbLxTvvu2!5ocd8dg=WSPtvD7`#htryk1K=aPZZ_+_ZJO_av{r=z{YYewP1|uZxx}
z<>{pVuxIZ+OeTdp$KS$Nvlo+RAC0P;sdVe8n#5)ntlQ<}fvX~~Oz2+xfk1wfjrPGy
zT3Oag-QDDhY@bwlPAl{LF1=p~!E~=F@S5=#g}tEZkecHq*r8Bdp%LHS!pBP`{my-a
zUw)j==|h^1Yzqtw;LTTFz-G1DHH*y5%*uWGyi3NE|KsU@GWB22alz;-H*Wo7``HB<
zdz<ZD{WA+mFlsK&Yh+*D0Qg1VH^c5!Xb>0V0Q}Yc`W@JJxDz}IGz=@OKL$0jy}1ma
zx=vPoKB@ZO1AOgg|M}4H3$-=y4HEwT`|rI{%2dEe2I=f$98D`UYDW6Dysj?tU7fvL
zZKZ5UyjsKF-XV>|*9Thb7Y(oU81ChkCL%x(7iz?U_H5tK<t;HvA?@>7FT{)hdu=gW
zuh+W=z;3sn7}hGY9>Fn<kE}w{poL&*euRnU_495MNMy$Go<mz?*8%a8s)+`%X10K+
z)~_EaC2pg?cJ}h+%eS}s_1XZ%*{Lb{1x4?D{q46I%N9=862kKt5O2FXET+qp$&Woe
z_WEnC^n2W|Yky<T{6)O`#{bcyoAEWbYj-kX;bA=Y>@?=iUBVe7Z{WklP72C2wxzg;
z3N<mjU%;(#k>;(pUmSLi8(#ZDcP@BOtDw|RX=@djtGjqjuShP?^c&6_6fkP!z`(hQ
zvDOcU_Hu}IyPP<zf}giku)iRLQP)1q4|9HE;_bJRo14evr>1enh>N)3(kuCF)>n-t
z)5Jta^XW(L^oR(I{3b5aH0G3euYJqvFSXA;`_P@9nX|6Mp=2J;Yx-vwlF)d4#k#@q
zl^9~u7ZVN30G;b|Y&FNq8d;ad>OgjBWP5Y*XN^pzehipD_qIaQH78q?cJ10p0JGUF
zJswZ(ihFQUeXAruRw-=Hg2GDKU`gn(7;O+kTcN>3R%v|!>#*q9EdD=xCZ*vVGD}*1
zPC+(9IQxqQEph7X$cBT(^|RX*3!L`fx+C>Hd-l}pl5yN!+s6yP?`HOi)GYF-Q84X1
z-g9V+jCp)hW`#?Cr@;=Bxw%a?ju^Ev4{vD(z*67-$bVHtLV{fpe0<aSp>cO#7Gdfa
zZ^tCW3zuAY!BdYt?7#G<S5=wx<TE^Q&z*EjtgXMeJ9#go#$3yeUAyVju>&vsCyB4W
zTh6@?JVCn-J(%@#32v`sSl@u)Gy6EOn#H)1?9|J);=SitwXAHd?y0#&q)L_N)lz+-
z?s`X2%LhwAuU?&OAs3z+XbyA;%Wu1!q~}Xc9eNHI-uNFjty#eX_fEv&aPa;|GnoGA
z=ZwAKI%a(E7MEUpA&*a)#y_6^XQQt}|Gs^gI%V>?xuxp%$PiQJ_)Ef+2d|8<#Dpt%
zhq=Xdhx3~L6@+9~Binr~1<iGDH}>H_9%X+TE0`X76z6mRH~HB<gZf{K>wrCe#@)s-
zsNYvN#IV{LmY$ye0fwa(g1f@VFTvzvX>+NtFJG`N6LQOBtHmHzQ|W2wR@ZdFzI;h)
zq2xf3<UkP|D3YWUNRsow>66i2SP8|4O#VE2bScR%mu%0L?8#FaWRD9I40|8i)f%c*
zRb6z&6<4I1&1M-K94!0x?R)ZzstyAlyQl(j=vsnp2~~Li(iy);LuQV6Nte{D2u+S~
z>>+Wj9`@kQ*$Pre*Yy)CA?`RLQ&L<~%Kp<A6-}J{e$J+>0<XR&-JL(;oMG0+Lf7B>
za5^O=rKF^$v3t*60K8r=&%N|2=L|oK9^Gr>DMLd;xciRrEL%LEm)@SkS(n^Tsbv5k
zEOMdivQKz)m%o9wp-jq)KGQ0Hd{rxFrdG+zTKS$Vsa}E9U-Fs)h6ur>E4Ehh{dz0i
z2VTs;i>9!0#jkw%*~eUQ*`-)47V`58c;cyPy!GCPO#1tS4D8>Ro;|wr_lNG|%TGV#
zqZu>#dHG7e&k5l2(HA47GBh&8^xLyD^9!ean)~0ZLTy-aR?6px^P0BxtnKFDzv^Z~
z=3^w@`1%|##t_^(hWPX+BzIK(9BUd|2po+&?HXv%g+l{A_1PwYnHVL9>Jp35cCuB4
zb@l4{`s=S3U|1wajvTpT@7}#1*=)9n0Cc((F8K+TkA(`Ka1WOXX+=g%WO#s)jVtMQ
z000+VNkl<ZsT^YS-^cn$@o2C&zhTWvk0vN=pf5xtzZ;|!K(FYQDNJ6C1j^l#!U`y;
z6pe~%dlU@+2s)3cyVaSTocy%eY?d~g?WBNg1Yv}`L^nC`ghv((i5t<FcjqKcaP|Y6
z^3uK={6?USrvU6R16%f9(+E}Zky6+7<*k0PHbhZYN-FTN=+|%N>=mW|K*(UH)Bj>U
zBQuNVUw(}b-+7%gPand|ufB=RX2WbY<954w?BV+{n`)zJo|^V=?!9XQ9pYl?(Jhf4
z-4c1|zPs?MDtq?qHSO57`}364G%~ZYDJm}3D=I3{b)5^Xmc*dIpgHZLB1OlJadhk2
zmA-v?(Xm5Zz4@f4XK)}rgHs0&A}T79e?2pW<ttY6#@p{x*P=S&>|xwF{uVy?cn0T<
z^jG8stQIQ*tnbV%`a?>6a&dOr{v&xk{|Z8~xhAJ#U@1n{b6whD)Moq*=#PK5abu#X
zwZC9HHL|_Dk?qaFU&74K_%CZT6G)<=@VS#?WTEZbx3A30%abmbORB0W-+c4UDGxvV
z@F0`PWSGvPhx7mKop*dw<^IRt&q>lGEv2l=6rqS9vMb_cPf-*F#j2oQC#WdH1u`r+
zKtz@ba(`UB0!0KBMN|-FiR_J_MW7TYvy|>Ow8?pXf1IW*Y0}b2OOfaOdgb+MPR@Dq
zO`7C<&UeyhFYJtj=vm;|AXb-<kZ#<}no)RDArOMCh0nDvilj`4J<nyCjsr$ppg+$*
z?TSWg9)EG^hImZMG;GXr`97No4eC7#t=<EB_56>>#Kgp*-MV#)4h#&GHk(blQw)R<
zKJ#))2a$~NoiE*Jixl3gv;@15w56SASQA0FfDu7JI!F@)LlvpgL5fBxqSB;;bm;-<
zy-Jl5gFpyXFo1vr5b2$OAXS>wP(m*u^j_|IzwgKWfA`<)&dff~o;lB%GkeZ?`SncH
z2|=lQ_u=d4Ku8kyy+#E|ni;{v$>K|l#L>$q8!{UBr|K+CfN`<rxsEX4aGz@yw8*5X
zF=h-pg!GAP%Cj@FWy1MH@|UV$uEV(yR)0@%9pkP9zYh#bWms5Oj-}gZ;e@kak($R>
z#=YNLnton}*}$R4qAhSfVgDETs7AuETPWe~iZtjNpFB9T{qw-H`FfX6|GQ}?aQnv|
zT+)zGF!1^i>?|elc`hI$nj<9Y9N!RKUhm}&Wj&Y?_iSl^4vx3%^p*|fn`ZlNsOf6s
z^(;-vb8*=Dmlw6u1S$CrQWi^2{^&X6Nt7>M^ArKSgj&QS4v>$vx-?mZSwKWE7)(q;
zLU+np%0Vh)`KQCkFq5F+`<9XQzFrQ>Zv-`((dWl)uQfmXR%J)Nk@kPZ{xKTkR(|~x
zbZfAHMQ=PZ796X6$Afn)E!#u1dHyOd&0IQ8bpaY6H;6BcZ|ptz9zHt|cMM0Gf9fb?
z$=z4dK%?k)3>xR1(|7neCB-8Yh>md)R2~Y-NO69ChqZp`xyJkV??0(AF0He~b$6@z
zb?%p&9?>()Zsy0cZ!U)EDOQ{7sdZeVeo`7detm{}J?aJFBz4w%89Zg<t<Hvu_aJb}
zPv*%)dw6jzA8p-?zS;=z0diEAl`dcDr!KrdS!uu<wOEQ<t5UbvUq{9@7Z?maUZ0i3
z-)Rg+^X6U4#k?3Hy-m@FajBX1Jv(jdJHH^`#!W*%L|aePLr4TdjL-!_)9xaT^aDAo
zYFpX^zq)Rg&&P>)i~~V0617E&CF55;U&1u^aoE!iw%B)r_;Z4_KPXBd23=6Jw7@oD
z!_)`UdDsqs8+f=y55FRwC$e4}VT-tOt}(}W;~h1|XuPNfI2+%a#hfke3f>bfZ-~I{
zsmRU~b3ji`vW_KY(O8dNA>O99$_V$rcNYuq!JjAE)5_NQ`M-)^Na2ul96jyNke*X`
z*OR;}l>s5xCNJ1K%*BK({M~(TLB*DNlV9;Rjz=Z<s~zeKZ@k+_-hSUg?x@yYcLw2&
zcJ@c@XcoTv3RG-%AORLrDr;IxLJt~4fv~2#$qhHQ#@qgmTh5U%l`VHmJ^a%~3N?Ba
zb*y_I#thnUhCllF2%5KO|FmlUSIDyub_uJ;PXJ5W1?AxQQ}4Cu#Tch7ozeeqv57pM
zRkOfm>Z%+lh0-cKP(W7qZ`x)ARdx3rNbSr!TD?4(uO?0}75E4(m$kDoDhp{dMNQ2K
z#R$^*Omo0X;wdQ}eC5uL!7$vND%ZZ|%MSYPzT86_gvVnR#&ShIQkK9!XJw7cS}%(@
z8CH4U%NSoC^Zo_M!etaZ%XTUUs3RTCCxedK<6|ZJu^+@}GykyBQ3+nn^E7HcO{7<}
z%f<PS>WRUoJ^>L4iN|;lt>CE@`{d+gJhOP>v`I~#8AdLx-_}^d32caw3v3DGp`|`3
zu!FkPn)(fVteItdu}OtF+1qrDU32mU9bYs<AG>#;l9L8kL~MS2pKR&sV8?KK1*^$<
z*i0_ho*01ltgw&dowKu>^_geoT!q|Hwa%&#Am%qgfsxfA`_}yg^KO9b_htR)o(00c
z`8-f<?2d^crDbEwWW}0QJ0q-rC-IcI%r^Qt;HIkh9Me-*3CrJ-^v{MCpcl4o$utrq
z29qP29pQVqz2Vjj<j*RfRBHbTj9-8WhJ1Bv&~*vm2;Xe29&cOEcX_7X?Owl1#-OzR
z@C)bYHu2%v(?e90^pE>SRMFN007w36pc}^ZS%@*trE>D%(<d4Q2?>dDR->qg)gc8E
zkx1@X!Cop?>PTu*%~{#Y8s*kjoyyf8TwJHBV>#xs#j_QBL4#L0Q8t+qv^>a4BL9SC
zn470UD{<j|INp)GQ@Nu{wUAIe3;`NLyO<m;?p^nty;t|}G5`JAr25g;A&i3Mz|>{i
z%3b~TAa!u0{({itaaT?<dHu;;z4f?DyUE=mitU-6Py+HM|09dVl~RhJp|$fS>n>sc
z!>y;K>5?)#<Deem6zkzpXxibz(o7Y*e>rtr^I233i9Z*O1SE5OuvvXsJ^r3#Cd_&+
z7m_h8>n)$i^Af(Rv1C7xnXp<V{YxhGa%Ut4DPF;0w%rY*OCmmv3I`7py=rSXF6Bl~
zjdgGK0`X_<_EJ}#1J<(>2_ct)_|7y9kGgNMFuL+BF_aNOje+|vw$uOl^N3q(fR*Cq
z<8vZ_Ojlo40k~S+>Oj_XA+Vy$07*(8KDXDZYI1A1<dSKiO$?$g)PIV_f4Qcl6ztPF
zXSY9P)V+9|uF~57;mhB*<xYZZMieM=W%lI&SxWF-aDwprO9d5kHS^ry+D|DdRrbf*
zm^a&XtJ!Ek8Ed9-^}rSi1OoXR8m4VC!rryBuB-eYZPQX0@=_JF+qe$&{vnM(VBPnK
zh{;b;w^Dzy4F@l$NMUn24lET?0kH6?a>>}9(F-Mb^42+k3Rrx3=64m#o%3<h|MGCW
z#;<&iUfeA3YLd^4T}(~Fbt)~G`mq0SPc}>AR*P78(dS?EI?eJNug=rlmxTu`=bAm3
zn($4mZ^w;lg1PADVu|Ka>==Z&&kd^Rn?NWk{AH>bx2o<HOHgRV*XaEzR@c4pTy$NK
z!1+-#+zsD6_$9Z@ZAXhjlgbwcEzm&fD1TL~zD84v*}2>~KE!niJTtdxH-P~;KMbxd
z(tdBi7YI8#O%5Sgo&^s#`5vWcLO37Db&(EgeP<#0PUVG+ZBowQa%nd&tkGyLQu{;r
zrPPgETJ2yl63s_svda1zg>SYdE4wC<FFuz&)csa&bGR|~wph1FtktSw1{BeKe&NLU
zOjoa$<#uuk>F)G>OTNrG^4~+B{E2=9t`pw~8_RVkM^_rGUJ))>@-Ih5L`3BB^YfE@
z2oFb*lai9sf{D^!@rutU>DbD#C5{3oARa|UMZ1)&Qh|fcQa;4--YWN2o|e3L?@5+<
z!ZLshjhER;xw0XPa`(EAyrJuOJH_NKF2Ffk;YeZj6~CsfjZNQlt=nSZ<@vEvzC-#?
z<I;c9fBN)keZTEzv*$J}+OXCRViT4jSMiwP2kvelQx2x;XO&NJS5vY<m(Db^8KWuM
zM44^eW?(-qk;m!32tgf=1kqoYewmvHriQ#aFA`|mUF;GMD||RgagLpSrxYx+S7iy*
zfxi-CK_J*3vv*C?^A0sR+1-_NCV0|y1evH17a0+fm5oeGGYmFuPWnBfXuGvWE8#Fh
z8anL)d13J-3($JEfK7d5exb)R2v_%s>cI{6&0`wqbNSYDS%yV|++4l2iDn^6m*T<E
z7mF;4Vo67D1H>T1zV$x1_yPxtkawhS<-9F`A2tUY=kMDKN~cx%az17Wy?eAG&PWf4
zoi2KaYu6lJj;9lDde|sonC2lM>=ou>Rvi+Cm+!mKZC@K-pTwMLP>48Ya!4`2FHV!(
zudgmSd3kR|qe1_QZOE`0J31CSPySdIsYjKiD}@ACNkOKb1KCK*OA$>XROBSO5$Qh6
zH5Zz&tXqk`-5SIOskFDNiCFy-Qoj)PE&+`BHgq&Kwf#szzxV4Acg;TD&-}-(aDlQD
z<#Df}YRGi6wzf8q_YfX6e}$SV0i#zFSeX~L`$j%x5u%($jap3yXyQ`$;4@aw)e)&#
zSxxdF&_W!OnXly`z{AZvB}Pt@|MP?QqbqweV?4Q-KP=g)&kEF01fctsamW|r%y7{;
zp?JJ8LRCFeX9F5dy{`40jOpV`>;7~}jYgk#t>mXk8Xe#7B(O>cInK|A<@8;7j`n^a
z(ZdZjk-y<fSUp&8O#aq^v?08St2_4Do<&nikJ|Q}<K0)Af?lW(unJp>A!@%m+3A;F
zrXZJW&Su~S;k1|IPZbfD{0{5u7v)wwA<sFw8A<><YP<In>bHJshmdoSHy1YQSy~sD
zx#JOWMXjKQVRhWUn1_o9URs`hcNT8Uau*ZMMb2^PRpGK;QW}^SIYKnKqmK97F80~w
z6OISl^YuNlx&pk4S*&OM02*qEPp?o-cX_!eV6$|o$;gHPD6wf+SeS5JXP${5<7H7{
zwwzH*@a6hgv7S>6Pz`A3=g*%N8igL@THW$s%a~vDqTfa-z$Ps0-Fwn2x`s@5AUZt}
z|N6K#9qMNQjsm38P_X{DO_IxFLPA1!SMLUeaYbNq*QuZU4Q5ULFo2+U&tqkYvC;k!
z$;!mYOnIS>2Rb|S6Q?o~wYdve*rCZvJB}=XoSRVPIObmrpp2Cun0m=x$Ex}*?}lF6
z%JqK=`#2fL2(#`=O!?I-@PU)gqJ6a_ED0r{2;KHi>F(wf;mx4NB_m&yD$S}^qzVxU
z-}ulT?WF~RL~@eUxti-^*VCX!E^~%>z(-;ETi4ap)ogU8AnCN^RYh8E*xfUrGM<nV
z0jS<k9s)Ks+AxmkOE|;kNd_FP4gBL<wph|*8a#V2D6#}#v-3ON;d6}d7i#+1ob>aF
zEKCCpeL%8oPCuWdPyz|Q1<EiNr0g5FLJ{oM5qX61znYVtyZj>x+TQt-T${uMBgp%n
zO(O5h;oIMP#1zJ&%r;H`b+VSvI@BkS?G3dxw}qu0c7%Dhx_zb-#C*xG5PQpIz~}nr
zd<yvQuJY}u&<v{Y`xVf#Dh~A8O@~`nb-qV3veO%G>+IuvVVX_OO6ipvMTcML!ArrY
zt^`&-AaG9vySNCDS&QGxuwmOH$=laGZx@d6d%k6|@V)8vmI>gkKUD+hmBt3&4J)d~
z94Vy{uU<#e=si#Y0r*hK$BzSo)YW@x#BKWw&wlgeATwU^7CEGQ1{~jZ!s4o_<Nq|G
zHrs)Odj<i0iDL7Xt7qN~O@##xln-VjEviS><+nWLd}P@YA$*$IGco{?Z4EePtd)5E
z*=6usVFIuh@b?E{L@JpFgWfS;d5Us+D8$&p0e|<oJyZ<jsO^?Gb-EBm3POm$%xvGJ
zlKr5iaFVTktKCv^yHhd&a8AL+S$KNm{Na7T;n<?teZqnJo7(@oyPI25l7XyI%t}JC
z*kJY2r;9E`I4rfs#|Ax>^X)TzyT)Mmd;xEbn9U!L#;ZLG0w77l*YCvMHb{avbmD7Q
zZE2A7${<3y29>2IpVJMjg4+E=7d8IVl?INJz+JdqHEw3#{AIO=ZQw<P4wTbW2>gTp
zPiybd+tysp#8RrC`#*v>v?jJD&L1!r_Yd-aL?u=e%<{Gty6-yUZChU}e<%L+`<(Do
z{E&cHY94WVSyhcyOyeGN5QhT4%axtz>^}DmeXGV5olc<#|HMepNTx2@kzV+Xh^tXB
z|Bkrbz;`-!b~6zfnOg5BPo8*HPMW?TrA0cVcTQkZK>6};1j0{VQBl!<^qV%xOs=IX
z;jqv^;AB>M`a^=oShJe>EJ-K*<-esS(p0x^7u!`B76a8KY2vM@`13v2>a&xwvhLyE
z%V=aO2A!$t>W;#Is$sING;yDp^hI9DhYuZG7Snw?Wuet(6)Xw0O739_>mbJJ@v6Mu
zU){;#aEbY(6AXA-$ak^auY341s3F(56shOAwdvoqs=O5}#gyEwG4-@SS$unU)$xIg
zl0(P@PHK2~_|Gd|LY%j4h#d57z|r3J#E|ZHdrzW(t)kVmouiC8y?LcF<7cDoB8W#2
zy5$~XP`=Fqwv3Fjv>mb_<YxMQr+33k(EK#cy1#x--^Qk3P_tNf`#f`H!yuwWS?%GS
z;KkQkgiA~*^BDRyGp>*;PG(N=>!86#)f$<b@0q^&eynVF7VeAN2o4_1R3DUVT(+vO
zse|;2u;17V;JAS@zJbVrb2-vk7=<4nz3t?=2{N++Crjhcc`B{XTuX@s@0SUn8PGuN
z#HJ)D$3d|c-6jBmg54zF<T~Q~(`}o3*=_LNPH`)c3~7i?zR}1p3CD?0IcZbqm?MvX
z0BgzE&YP^4YG2fsdqTB|gXiZ#?HhmFTq<Fy*#o&~)d^tZzco{(x;@i$zJ@=)GF=>A
zc7Kz)UKkPaf!|;!e<d#hiA`@4U!o*d=3&l$A{xQNoHyL0a@C^X7AJl@r(dGCJ=bFH
zy)pV5F7s?@>5)2res|p>ODaiBZ0yiX^)gfAtn<gXxUc%9#%2)0_t9hf0V?~`tJM{K
zjKe4B+ORO2f~Eu*)uOqf(GS)GI|NX-XIa$Y_-7S^XP@;Db4-UFbpf67`#a3=q_xaB
z{vRU@X{1Q43lCbbeVbmp%jheA@-S!T3S-w|y|2buj(R13zthm^$Ub=R$FXuU;>8HS
zwXl(66B81`z|t~3w?}`A^J!$s(l2_}=fT{|eq(2{X_$O|2ww3DfQo8)`PFsXJ3+R`
zZZ2-?33)`uZXJjkdCeL~Gm)#Ml;6~He|2E=rANpe75OQmCAIj`l$T^en@=Rb8l5Yd
zs!P~jcEu^hpl3H<_rPhemY-5Yj?c`x!7)OFvL-c?EmRydW=G_qcRq&PS0S&Lvvrgu
zcC#g`(BNOz^LbkM?>qCg*ng0>B=f`4?Gf2p`bMEJg#G6DZ0WX)ITUsxc@EWUIjz^f
zqw19FMy=z0Bc1abm3>r1sspJT0l!pkEJwvJ2&en&*z4M|th#DbKX{#Z{ThY8OicZg
zm=jivB=;I2?BN1)caYo}V<m;*Sl}A20zT*-33e0u8i52agxSTg2s64xF0fQLXA+i*
zwwbVHggR-;Fm|?vF*<S^=jLfFh4jQnyWnc8Y{8e*0VEk8L?Wn3MUeR^;9RW>h2&4V
z4}@>j)b>G>=|nPKzDD{@dFJDQN{_cYtD@<WZbt&WfJTefXULuZHiC13H~;V5$aax`
z<Nx#EB=Y}u2kZQ&2Y7)=9Z<Rd47~u8z4-5hk>(FTMgL|7+B@Vt+zkK!$+6cr^0}YK
UqCg!_2!QK_vc_{1*dqLY09x|HegFUf

literal 0
HcmV?d00001

diff --git a/docs/assets/images/readme-light.png b/docs/assets/images/readme-light.png
new file mode 100644
index 0000000000000000000000000000000000000000..f59fe8843787c25f5cd994e7170616561a481df2
GIT binary patch
literal 38943
zcmdpd1CwN3ux;Dsv~5jK+n%<&+L*R&+qP}nn)bA9+qPeQH{N|e;_Zm4I;Y}9RMgJg
znJaUx3|Ej7M})(L0|5a+l#~!v0s#R_0A2^cKmp$$`=}RyAFy^38jc_!NCW@9Ku2uL
zjDTO_IEkq{DchPjx#~L@gSfi7GMd|1IvVQR88g~Cm}Xw_;(~w>gGh=BskmibWV?C+
z4nIe>r*T}3uj2$)q){{q{J<3g&^gM~n-?uA7h7ul03<PRoNE@<l}oC=i>ggF!F~ax
zd0~FPjL=p8>2kEcXB|I}$O@y`{6K|)J^u7~y*tV}&N}u)e$IL4+J1zgp}_ioD>QO$
z2}=EcSCc$FiMjs&U9+_Rzuh#ou>bAf6BUIaDN1U%`i3C^Df8WQ)7o@dRc+DypFL;_
zs`JUEr;i~lkV9PZoen~oNMjbkzsZA<$>s0fja-v(($v$1hmSNIE9yMr-1c~*wbfdm
zos~FScqCs8TQzT0!(k)2tbJ+gEHLmi)>?O_Ru`V@--A)UHiq~&Icbd&eZEv}>87Kk
z<nYhn7X?M}t~B?bzkVUh(L)q}yyF>wKaHDsrFrD^c<P;L4l&a0;#ms6zLv9}9TY`q
zN;0WT_jDxSOuet%@!;kysyD)=<pM1g`edJ@eZ?N3YW>8A9(Sa@SZYf{x32BoWNpKc
z64zA62_}jb``BNv%i6JN1G;nnI#^lTld;n7cpOz$R)(RoydQ)jH_r;7DK4ju;P`={
z>dBikS5Qz8U_M`xjD?Mz3Bt_GjP?{h?xC;2M0?)9x0Jfp-@g}T*O#5Uru>H3F6lhH
zSzFd3<9k@si{bM!e(3Ml@Dj;%bZKGv9>bBq?M(0o?l5Kko!hsi_-}7v73~AD@Q)HB
z5o7(Yr(whWP8ZoSC5+mQp3T&i@^x#D?A)6TzZ#??`Wl0c9cr`!fJ<tKdj06RM#t8D
z!-EAf8+KBCw&zoPh|=y>XZyiQ&%F2l>~bkdUb&kUJ-+9|2}zFoQMzt@eSM*`jsRf3
zDKQwu+yCXw5IF#1LuqU>U%@QeJcDj*d^~m2_xtmFxk|&36)PpBtf{V!QwgE>k)r$i
z?|sdzb~ti*N2BTIQ-8ZF^asT`Uvo1Xj&v$xv+zdr$wKq{1MGkyA^yx{jm;MNpGPdB
z&3Wg7EI6Ck(IY<p#aFTX5$oTRR1}Je`EI>!RD8PL3UDs3;B&b@3bPjYHh&vLMW84E
zEYW_Ei+FG;K=6Jm4!zysa@CZWX`u=#m0qz-aSyR|=h3Tl+RK~C822q<THp3C01v2r
zTGN3yJq-=2O{<o+_v2!VmTk9l)%amX8TA+hiRe;%i;$EC23+vb`{x%j|L4Pemqw$B
zyMcj$k)eDs1vSvikg!P{k}Y%IR*hR$l5jF6+3IfjSri6xTuc6hd=O3N3Vau|G}Kp1
zNS<P&&wzbaCTdq6t~L7)zoW5q>N&29h1)(%%xH8C55(yr?G*wd=tKtFR1ypU2<TW<
zRr<5f%_fS)w*~<HFM@&4C!f=d`rF=<lp2oo%6{v*M$RxA{ZVdYMWkTV&#bN+MR(o0
zmZONhy}b;fKo8q2dvA+ef*~P7pX!&{i2e`g*i*>z0ki}xtk6(1?Vt1Nf<4b$o}csc
z^R`NWM+*{Bi%8Qr%9n_I7mgPxVg<+$<Tzs1nZmvtm?=4wMM&1;KC21@sKD#C&8yyA
z=Oqm~0;JH3*}_m*Sx5?1!s6gyF?bk~!Gl%HVV*s1fr`~ad9e$kpSM?EAqiqEKX-E=
z-Q7cz-j@}_CMRU3q@1epw?<ZtXV2DbU98sGHKoi!21{KpneWJ#v3t9S>+i1;h&!c4
zthV+`Ei2?wRS0|)8W^{ycXE&nwBeY1mFgWF6)zzc7v!%pO1^2=%5_({G}I_#qWbjF
zUjh>j@wx^(FgA1(Wjn#ERt}8&Z_GlTo$qcN&h{49y=)<bqb_GhU8FRe!^Sk-UBvoY
znAZ2!)GvKwe$F37jw67J1A~z!M&DaH`oXic^~mk-ZtPzi9BRNIRwyKxl9WU(okEX;
zj)&*jiekVU-@mx1m3Jde{e?TGCUgh|vxvVp2q$VR)IS=>i$C{U<V<2lJ6e3`yu(M?
zz+g|-a;Zu~Q%mc8p-3{3pOKM~U0O<t{0|r~GH%|<0mCP&5DW?Br~P8txgNE*sOxYv
zpTcl-tsUSc*nCRh_Yna{S%~$+i~g+66#hxUl4*|W?kO`XN%zR9dH9*wF;X<we!zTz
zj_0_D)g~Ts@6WLLnmV`Y)JQpHoP_~@CgN9z=ukn0J@~ZdjPhJ8Q9Z(+gAt%Qm-`(h
zr=v{XN<yg^(fZxlQM{w!6);j2!bz$)MHFUkpOop6xUJerX_`8gUss#Oi)oZuEaVZ`
z0*sb>UH_RBO3UFf=MpdGskxpvK|C#sR}rG=BAWZl-AlEltG7CubIR;rpw)$cmVocq
z>Fw=pEF8&{#V^eag*!H@TyP!%h-dbgG8P4A;o%>W%XE3?#$cX%J|Z&n+5q5I&HRfe
zyH`?Dns~#P8s}N%Zm4M|7m4pv4k~yMiwh=4C0U;aFT_w1hW+OiS4klExsF)db!}tW
zNWYdM>HFbQ$5QWEa~0JIlaFuYk2bm$>m!?IAbuMFc|Xtd23~%$xlX8u-qB)k{iFtJ
zFy~qPB%!RDy)>Q9{N0JmO>xSmmW7JG*`0{!_ebJxUrs=J*;6tGAStU5ZeQUfxj3q&
zDDqTimawk^QWVrqXW{lP{XCx>ynY=nb(-RGT6QAO#LZ`7XD=4^$n%Mzh6^E*aLw+i
z7_`&1tN8G*j@ef1a;%{u&Q*}({bbZ|4+aMffZFvTIkZ)r`Ds|;YP$&$H&eden2pw~
zhmi!yR8GH9zf-1LiS(txd897fus^c`uc7DPZN1KJJe9?TqoJX}0sv%-?l0zh71zBI
z)!ky2Z8=#*sSw0D!#?yP&P^A!a|lUswbs?$)I>7%xJi2>0fQNGOd6}?WyNAvHH<wM
zHF{CfCTMf@CAR?#uLKtCL9cdx+4EA*qw3BMw1yJ~qvy?i4l@pwY->AOtx#DgMmj5R
z@rC>I?;;!z%dd~u#MqW-Ygu8BR&lEihhCwdl}^hwe8MnnB;D)K9N9#^eSfcz46@*o
z*`VlJUtaVxV;^}wf@IiXB%{Vje(e{ic!7qG*gDL=-ja(}T5X}nbxH+pW2Qj0s=60d
zZ&>Hh>^_ZkuWv=jU$2p|>H4~g3LNKEsbs%UCs%ud1E1dsxxKZbf%#GEt9q+Yy#7MT
zr|8?WVf~cD`76Xn$7)c#a!`!MtD~#LVXO7(GziSMmhD2>!B)bjylCuaN>#P@Yv^UZ
zK+5H+gG$8(1BrW3eUQ;eUVB&-vZ3$urfYBA`eiJ7O$X9GjUQX1{{A+Jil-U68Qco%
z%;JqDNk6WF)KBst1ek?S*Rt)|BxQ=gD{Lbj;l>Zj*y&@>i?8?V@A(QqWk5=rRr-}c
zFFm2&jcU*bW>^irrm?rDX`^3gYpAyESnJlWwY49Sg(DPP_bdjTIUmrOEW4EX7Ec3Q
zEiUiEV<{8Etdp78^y)WbvkkXy#QDn4c=A^^zQz(vp?A}efa8&he5NiX-_wNn+2v?F
zw8;(IO{-1xl_SJ7I#xjaYvA1+&GF!E@6aOKy35P^Ace(rHV5xp!>)Ih+Xp2DfCK@Z
z|E;6MMBHDXvx?dmM^HPJF`S!I+Vye`oH$4U{g+1+#@vVJ)Z(ajAc>@E#Z>6{hNpnu
zWp|I$^b+Kz=d&jF`O!kW-uKC2t|Q%j(Sduu-J74y)Alwy^Q<bs@}~9`V0`ZSS}McO
z@?@RKSm5c>aQj-Z*1)vv!B|DR@iiIWqni@P+YfZ|&v|W8oAnl1_t(QzWhg}4hk}ZV
zcUsKA#hU>c^=E7tcTN6$QsQAGW~8tNe`)ATlx!`f049iL(6U>RX5^22aH2$q45Pus
zfF_oVY$?Xo9`E;QD%p%AB6M`JktiJHTpk*(z9FMj$;7ORqju@PI)A^~!Kk3%i%T|=
zzSp4Fk@*)2_?T~VViWc|kvpYbDK<z{_&V5*skNS1o3)OmX^d-M?~iSH=Ga@<!qfR0
zW#+x4i2*1gjY`f#`y4dYzt};i%u44d*mMgkmh!6BZos#t<G>!&Q-85i$Q3qmivptC
zaRp(ZOp8y=#(GR=X=8=Vi|fVVT!{W*>7a{rl$@~p?Ny*$X*xDyB!Qbn{Fq^ojeUZW
zfg2O%p}2rVnW5uTly+Ff6b>;mwXY*xDsD?K0C-9!3n^8N#T5K$5AxpqMUq=Z|Ihch
zRF_9uf>58HWc3OaB#}~93HMMq2aAa4I|QXbdZ6smaVK*-&{Y1Qks53-I}QV~KIUdy
z98M2qtXn}9&u4RP{p7K+F3+im)hbyjWqDy?<EP8TiaV<CCe^`Y-Z>j~E?FECj8JV$
zOK~--0^Hr2wXK&|Kc=#i=<pUlRfl8m{PjX%!j%%Mr&6Unxh&k)J7ix*|0FzC)~x*4
z=@BaHeH$@*qxklf(&`O7MMbqN7bSgyPtQ;9UYhO9?+tmrR6w(}em%0{Hg*GlQn$fS
zge5-G2)u$?_%%}?MQ^}ID_L$|jgSQbxp=V1k~nxM5)QY~V8A-Em`&LOYob3|OBdw|
z3-a$?^3e=xkRX(BASS<N)MzkTtJ5FI0iAF@Uy9Gn9FM|bkt>V1+HFNzmad%tE`zx$
zu&+6XjS4bCA#&+_>Zf_N_4-KU!^E;4D3WYvUq8{?*zi7i^;K{(9hT?Qo6!+0Yk~Qr
zFOC*3jIg{vH)b2A^l580ms3F|fl4)8O)Lc$XAp-XJXq?SStg;OP2?b;O;6$u;kM%#
zkHdOGMWKV~23rOx>_<tWDx4ALgbDXqMBW%Z*&G&SmV?~xa5&a!#lB8dKAlLX3k6Rl
zVjbjVoC$|xuMdio|GlP2Hca?6jUB`XP}rg$ykZD|Oofwsi#}MOf~-IkY+S#TuhML;
zkalr#k&~2^G~4WO{twxFl_4=vcFJ%wDbCMrV+GV(TaVj^=@5Ty?w^HUS%1G0*H%Z@
zb2<$lG0y$k<JLn@_FhI<JUX*4nmVV`M)cFaS&ipDzU^}jLD|fCknna8ZgM~iT}iGk
zoZ%Dc6^pRR_KPGpfcpbGYUnQ*)_MF(8^DD;st#Hz6^3*T4nCLjFJ*{<Ah%I;Py(4O
z_&^&L8~(&~BC6FaoN{vLT=W1TbCSJUe{i^WpkckwN_C?GMj|#?u3nto%e#%!=_DP&
z+-F4*G-udFO8vVk&nnqyg9pr%^ImxQk>~#IuE4haFlGJz{$AuUXY_QwbeUj*d9>pw
z0YO$-01~RXnPG(4r992nDcfN-jNtpLY~XOfnbpD3F*$m6S>_^xnOv<=oalDC+4tJQ
z&#&MgQ_MsAdO}nbg7|ce%Lol0!LiLMdY+Mpbh$7=;#Rr~n&CoZ=(GJ@e<&smv;gTZ
zsu1L7n7Bs4tQ;0SUf*r62te?m1R>P$Qsh4o<0Z4x96sHn$+48@$*Gjj;@?aVu?nd_
zE$t`F6VNrNNQtaa;+6Vcp9d1Op0ML*U4FzNrQ2K6k>a6ShbGSDJAQnzV|r7(xFr>N
zI1W(iEc@QhNi%(7&PYi~4R$!6i`v-KpzNu40AdNeYqQg;;0*a;q(up{$i?oS>)Q<(
zLI><84pz^Hk4!&Y>gn1?I>kjqrRM*teSj-9`1A!piiy8{Sci|y)DjBNukk-6QG?|p
zy<cpQjxTwpImq!2ymO99&D&xOgq}kSSoi|H7{n%*tl&D6$|p0CirugQVYD`moM@EG
zBsNjYG&&Wzz2Ou=8?uT&;keZ--J?D!GjqMId-{?+^AcH5U6ipCvBx)y5{X-A1$>o>
z?E;o}tE{!w7nd``-0!S9Kz=GgrOJAx==@Fg{%CTXfrdu$Y^W$HA%+s@y10iAhCT3E
z3rj6cWWz_cQ5kbfztu0Mjbj-v-3KBM9Si5*eM;60e-l(EW&9b!lIg{Z6K^hmtf#Dz
zP92=U_&iN)xWGsXUQu*1a3=#rKtJ+T3q-Ly8m%F*_muKo&$w5Y&p>Va)`>R~b+nbi
zZzcM-Pu+)72Q^9yi-&Tdu#;TwPCArCEm*CvLidM-k`BD?<rKDocdosJOt!%t&&|^!
zr%hcdM|9@cta}7Kdziroi$SxfZPVuVaam8#sC7LT1VZBFjiYWixCFWatb<CVUue14
z##6hc&ecIHy=boUT5VXFI?WR)>pPgdCl{n{)#{l+ugB{>43+E)G_Ae8U090dyhgdx
zR|Q)*tY%sJ1sl{9->XTCh4Ad_zW#gj>r7Zr3(ufF_9V!r^=zsQR@{?tdkbp&cPLr{
zX_>Ce?J2Xpf84=!u8Q4IdcD`vXn6C=JVl|wN(0qj4!o``=Z47F?L}Q^UiY?RASfbp
zzz#Q)_bV5*$3`<9)RUJi58|Tk!Y;l)d|+-Eon)hu=OrU)-rZGLdToF>A$B|p2mHmW
z7#8?;i=mHX9h$Iu-7VtUOSGx~_zgn{E*IqkF%mcIO}c5*>UQ1z{3b3g-YfNaMnOju
zK=i%zK&%tYPowX@q^p2=l1#s%K|)irN=4Z{^pi=;ZWb*J&XvufUE5cq$!s=+*>IR%
zr^yV}ZxvOPT(H%WDJu$t!+B?6x=<3?Yc>_++uBkj2IBsU)lE>T>xXsFOhCvCf^u1i
zc|}M$HCnj@4aPuH1gm0ETt6!&Nf8E!5`H?=<9W+Xt~pbc_P6p>{#gyPcz;UZE}eFC
z(o2`_^_(H|_N*`r$=-aY0Ln31>GF$F&p?S2&c=LS0H8UF0AEb?tgld7h<v?Bh$x?6
z3A!)RC>&?UYAD3#hxk<jlnX?8sD`92r7SJ;DGd_+Ee0ru(a8P6dU4j;$mn02r!UF$
zO+vZo`RSS`Tuj|sx4JI7^NherQ&>r24mIkW_zeVJIrzyi3a=V!-b>z8`BE~i^)Er<
zA8-h4GKABk=(g0f$O1kw8%*h+)s;n{9~7oNX@Lj1e_vFm%<BaT>>F>P3EhZk$;kE*
ziLv}-zx@_t+4%ZPgtzxule~Wh`PGDjk@QywoRx1v-c$7l>je;k{{hym9-oJ)di<X;
zJXh2;<Ed89Lmb^yI&lFdA2>zHWAmCpAMGf6)iF1JP=6Ltu}HeBs+D_^CM__~4fXV%
zqvA)$`ZTLmL25FPPk&B2O6r7eu^zx=V4$*`4C%F?#_9P!E~;H_=^QtpJ1!?nVvF)%
zZoL)Y>Y!JRKi_wbqtj4Zy61-ie*L3m5rm~Imc#2t+na%lP(+m=M}m+eIbmdN8AXAd
z69mBp>-Br_|NO^9YhZ3q!z`SSjEhYh6XsYQFVuquY6WGbAk#2CP3b#i@iA0P3{f%Q
z4-<TX%X&*flSGMD2U>wr25Udl%(7jpJ!zMZ4n|2G6K2rVF#J~^{b9PHvHGW=W`IfO
z@z?=$XYsn*;e3Q`@As^h2let}5O#vAFKvifp&O*6wBFcjgwWLf@i;)-$sinQ;}5QO
zY9f^`2Hf*BubL_w5xZp<JxIg#IJzs08=P6F)X_^me-YxYZajakytdnJFbk#_4$9Vx
za~pT(ZEWaPjF^#O16u#LEj{J9+JG8|<nWhTkSAK$6xr{n7cC2mvm)#D7SObdZ|U_K
zm~tt`xTPhRCda8Wuf-q;Vgm~~&%Hg*p6#!FJP|$l@cojOhTTfp`fvBgEJw!Ii^gfq
zWZ#TA#}5tR`PjE>*E}ciUngQxudlN?=Vr>!C{Eiq`rh)n++42{HRjxXapYqpH!tK&
zKH9SeC$_LY4Ix~d#(#@3uv2f==%yQ;*Ig<qDil?cADUFKm^6k7Tn}%2YkMTNx4IQD
zAP0ViMMsm6@aO1l+GCd@;f)>D<{Sl5*i6-iYZo;PBG|nqxuslQiKz<F)b@OE6eYRw
z>N@ERm^7loKyW@rIl9@c@kQY|VuAVV<E}X2F*DD4wcJSeeA-Y%9!={@q{KSszDZ!d
z*M_~`-r=Gd0^VF&8XKv@N`zBYvV*sm$80$uN>q{$h?G!oVBdpsc*Iv6?gp)7^~FR{
zX;w%urZSTm8j%5_M@MjwLa236>PTM57^Fxq;8kF)lA|eEy!PhOfhMNorkKI=<V|%+
z5kvViyi5+adqc-39beKQV3>6`7C&)~>8QaxN}GI9HRM&n$U8D@*7?P6_U{pM6#)RM
zp3kRsWiS3*Y8o1Iv$-w8Q?A3kEf}BKB^M`<tc;QvGfT1<j7-;vbRlqGtx2XJl!&=E
zRpbmlmyP(QQVOr&-CX_w6atgt!fMY?cb*9Fs_os<GE&hbk=<INB*S^luVcGm>J!|L
zuNXbDco>qC#YX6X5G0d-G*t66e9@9>+GBP3fj9y-2pT%v^IR-}H5xh+VX@G8YnLYd
zFZL>)GINCsSVL+_C~e)8iBvkWsf@cV^TIF%1==fC`cxg0CWbqHUs+rp7~!z6<m7{i
zSxKT+RtUl{0e!^X9#47SqhzN^Mf$cRSw0qeRJLBalQ|8<V#y>fSkS8ge4|a5?W>Bt
z>P#|141aVbi7g&AguCHeK07ATem0aYl&Yfd3CF`O@a$vcd(qPJ`#~mP9^nAm$YlBy
zLUy#hvEeXJWA&~{W7XlN%4(J7hb%^d{oh>wgfJ?2wZPs?7Wu9$9&u5be$Nl>_A!K?
zKSu8cyr;1Q3*%bWFR@xT93tuX`M)&IdZDFlLo*0uV^88Sbcs62X8B+XtYnF7db-6G
zn)CHDfqmh}kNat=_RTAoO|*Y;8_*Zu)HbuYP_50N;t}D>z3WOE<Gt%^F-rv&Esilm
zygmKbptKuT5!W$8UL#=h81O9^5Jtxs+1LU#JC)6Zcktla_9j+|_Gs#KfnMfsu<a0i
zDAW-F<}F#PCaujMTKSgT9X0xCXti<lzdjGfy>}l-&Fg+c(WuH|fVzYd4E)sO%^g1#
zT_)q<3?jXG|6I95vdZ|wLb5bUoMXaY&7W<JaHfCU8(FcciL%o<$v=nA)`A5gU~Rt#
zlVvCTi%Nc@oVqG59?#zsV>G#OcwA4wJNe-^_t%tsFrTUa%j*K~t+RxFEJjz*4sM>B
zs3IIMpOS~EY#h9)@%VP9l~GziF6b*NBF4r=xZ!EHJL#Usro1k>bT_3b$;(dMR^5=#
zLTHn1*X{5&Jc$Qb-97lqlgYgQX?@&Igk6YB3~8R4LmNtIP8@bm2dvRRG~MgNynL=M
z08pHiloSMyxTYbUr@JgKfhjC!W68sfpyXt0i5^3K?mC{{hvFX?%iGiTQp1&A!O(F%
zn=^6)K&@H{NpiT2z@+1rvCXRJztlVS)^dDK3GjO_xkqnv(Z-*Z3qHil_&V@4gi>E!
zC*K8wR?Ec}|4Cl{=r+@wQmD+#%+Q+LvTkjb5yC{=ZZFK6&E$BitLaB-t`I!RlMPme
zz4m(excduHjM~g%d=;LJt@(t(hy(ce_O&<wFo(uEm5m>1+couFF000`^${4VnLApR
zn6<&q9B*UJ&E%!cOr=0~W%-{0ehS}h@uIS^$$ucfz@y$FKwhxzHjPmI3p*b_VnmC{
zawi=rSMnPbAkPZn{ebH^H8eU2zrFjWO*{dgh(kGzH%tTDmd&K{5aS+UIhOu@l44A~
z8Qz`p1)Cqldwn~vb)9qmv;h%rzIn`+6&GL4bJWI5bC0Ss54Z}KC6bR!fn)zH)X%Ir
z=Qun(e53=y=>MSTLNf&SQAtaRf<!NxqzteLu81N`-rS5Ro<Mh{)7(D6u%4~<i{0_x
zkNh(6&(?LlE-3SmpR#Mt>O3{dv8b@8o4Rr*3;TT&6EfI;mnOBolQ1f4&QePDpVA(t
znv}0>G_4F#e3qWxD3Y6@lmcXz?Kv+#^xq1zJo}2hUIA@!-cR$>H+>_y2}yB(hbOh#
zo`&^)nU77UyB#*SS@xeZu35Ek$CenKZxKf>FYM89*unTG?L223U_-8G;EPK2i}A+T
z2~8pPc17zK4&U8t`3`vdrulCZfq-UI)O*T<l1=f3nj`aX{OPPQU$NR1l=&BTK?|#-
zVOTI$k}#x?ZbW{T)QPC=5^OJ0u+lX0?dJZvL`xEK4x8PvnH{(F=?lCNAQv5<9FvD3
zaktGmsgY2XYO4w+c@^Tk47X}$pPR~R-*b@nvxYI569#Z7Kw*T<<m=31hmP)4S*6pF
z7y9+Xh;V1t(VNzUy&SwQ0p`Lqho|GZaLjv$z80ALPk^N2Uw`mxEa4y`qGkR?)*=Q`
z&X0AUEZb_bJ_-pZHpP&;PG|`KiZ^S`R7M>_;HfX*AKZjeZNWvgxv|xsB^5Zhcl(!j
z=&-quGKuOK)xWm;6|aN~A3NkF9(ZuTm=4}xl`HUg{*Zay#k#u-@?;7Eq@r@{r*_I%
z;OT_Dnx~ZU_f=@$ooq5dSv2eIhkqkp-V=YP>p_y@uR--Ks_9HsTrbC^;OYWqE^A0w
z3ehV{W;;!h*=^SP!RN1b`^=vebag?4?(xsZvDB>Y9JW-m-$j@Q^HG>EuBsvo2H}Fm
zh&u69R~655uqkzXXv%45y59EmXz1w=e6M}OlJY8OvsqZm%Bl=LxD4^E)KoA#H$(K(
zfMuEbb9(56G39cG>^o8Eprr(jrC`;ookp{*0y4CS2PY9om^59s&r@C*)%SKr?$!Tl
z0i3knj72<o9QNnOuWV1bUEN6Bx4oD(|Bj_(cuKDnT4vp<v}`PTZ7)EQGMeVNK?l3~
z$)O_Y(eB{Z3T*(5#!IJV0vEl=<!&&);J!EgJc-?nY_eSi>(EY#6Ja#(!vi@=?Zxcu
z|1ZMn;^A~|Lw9l-#mk}2UCs^{?)QdG>&tTQS8~m{N1MUC8kl9qqn>Oms3|l}4c8A1
zSXtqXel-NkXB}kOc}A_eKYe~C!5kyqc^A%~hvgF7GWKCZv%721s@9CvDc|OnJV=-%
zK?+VIt0M-C+gpngG_F!K9E6dtF{}nVf|=!&f#MI2zlX%jgAl?8D1B0Z4x2FhS@^ZZ
zGC;|%ex}<k=IG?^y48{NoYL;Q_<^qzJf@JpFUb0W8HAF>dqY0;!<m~-HDsto9~Xhk
z7$rnMj~Wh$rUpgD=NrXcJ4}M#A3VBsCpxD5`|`+(HFyXyC9vRbof~9O;Y@TOsA<Q<
zWMtY(x}=Y%fCx=EUm?2=GIUeUYoiBW-aLakLii_!&+jt|);Cy0(?+dlHOsDV-3BQq
z%@(hLyx>Z{UK(on1U~w+yTPcd-iS!0TDtOIVa_yB{G*URWgf1qYoE{9OGMyDv)T;`
z;=CA%dxA;WD6`&AINS9o9*FJ+1?P<r1IEel)rfZjfKcE8SX}0sPNgnJKf&g;VtlBk
zS^XPDkBtayYaz@eey*T2l^BhI>70+7n0^%$MVRe&PG$|5)rJ*)DABwq+`{7m$_cdm
zZ&LLy=TWQ{ki^vn#^GDgX!7f{4bKU)HLqu1J<`#hl7fzJ-h2y6<^e*bD<!vQkJACY
z`Qwop@$vSip7&BA0shLUPJbbmTL#GQDzq=NiSC=l14-7zkamiZz2-#fxjG3h-G!Tn
zf;K!suo$%4>U7+#K=)WWbmEl&TK*V;SheL0lT2_87L~EkwbJ#*C1%&UytFOX&+A_F
z+bZvwD+ad9ZA~>CQZ%kDPW+j)3Yd_k^Zq5l%T;YwdtzPpSI`Tk??6nktd(j?vl~B>
zyVF1ZRBZO{<s&m73Q~daoE8<NyL==@T_8ja6(H1NwK<&o{RWoVVzs+As)hm=jGGiZ
ze_CY!)-iG}rik(3Puv$*RD4HF^TMCEC?`2_!d`6Vx@SqmP^qz1+L+xa=fdZ^AM{ca
zLbF;oKK+r*rqus7kypQExyEndlY_J=u2g}5kpP~&@pU$8)t>)4oaJsnAAl@&fN6{c
zWu^U9%G_|52Ne`)`(<ZKIBL|mJuhb{Ls0rNybB_oT&GO}gp@s~SuNg#7L~7ridPs0
zSm#HDZ2YUN=@`>*T_3po1Gf0i<e!GR-UM^_Q@v?y$nkoxv0G3atb>xvGc2`}QY|j!
zW&U~y?d7J@O!td8CSdMHZCO7t%hiLyRaZku+llV|pmYc(Vk03BwFe!t0s>Dj2_b_L
z#7QUjVK0*DYj=w0n%@6r?O?<+(IG!f*<AcJ>oQ0Jx#KdYlJLc4f6W*9Q{IDDz|v9+
zDTJkbd&MxG=yV;==6&AgUHt6sz7hJjr?$)ONvPiP#@D~^<RY^28Nltq07sDpNbOX$
zovVDSp6z^+<um~r(dX%1261rwbw1QCu9)m14Aj8$oit;QpE)w$M5-QxvEa9l1P$yW
zKm#wN??QlI#>t5cmZU#Nnnd(qER36X4npHNzKTch=MM-7rtUkYEuZ^b&g>~l%uRM3
z+Z%o`2r$qf6o?O@qC~9VA|0~lHT&Bsht@u90hKM?Bh{u!PcUr({v05KZncJ)c0V{Z
z7cG4A2y8_N#Ny#^?yo-6nXn4ThdxPUa5MFR=CD#^YpvXa#fuj07^X&Fh@i#GBJD3_
z@y;6DuX`Rx#49zRg(V@IXC#cWOh<n-Ioak&uwn8;bE}Yk#FuR0y>@PYUU`v-a)W5;
z1_~DXO6de2q*PuZ4e~@9v>n&dSXH9OgY?Dy>1$=QabxWkdiliAb>qm{@Fi#CN@XH;
zyB;^Q$w-$2Skv`RMZkpK<K6E&{$bxp%OVqx=~}yfj!Z~Mc-l=)x+`j7&o9iEK<ZG@
z^fcadxpnirky^uNHU#2z=ht|WUcILAkAGtXJhUmcH#J>GDpd|10>?=a-Rq_*7&hPI
zTMaudFyhXPIGE^(xTTxFZ0hJdB$%KLp*uAvOyQ(pC1mEwQ4p}-s0K%v9R}@c+K!;6
z_#Ii5#<=^S`xH>jGdM3?Hm)*4?tg@YqoBYweZ<j;a)!`A>=xkm$JgkKK;Pep%)5C#
z<$~aLC4=zy_6j)(x2^n!5u(NrgbB0gdrwa;;hE-L9U6>h=DajGcr6`9kPSof)P^4m
zkT|%<41PskBd|D&UWM&uzlQZoip<O{1qMy3k$*~{wXK|jc9unTjVN7<{ixeAL}Xg$
znjS~~(_$v~(ZKu`QE&tKs{tkHrgIhjU_;`)#G4*XrB66dr1_9DL4v0I`S;tW4=N@m
zpfw>VD4L0lAPWl}3lKV*gBDG-En_wds;X3e_j1m#Xhwi**}Cd}SeR1zuNC`*mxU_-
z5xwYlV9A9ibh9Uf=Mnpdb?bU9Vz6-kr3*ZaCxw>4Ds#T27T!!B47JD2c?<iV8S|_E
zwGX}H_`V0jZM~r*n~L{@Vy;BSOQesWR13W>Z|{)Z^=-^1r5A{Iqg9~+kE<M5n0>j6
z-$y^lw$+V;rT~@R$IgH9n0wWQ-%&79f8+F-)y{8s?k~>`eRp41AI{e^r_yos<4=(l
zbS_P7^-<k(84eo7^c+4X-vi=H%nNAq2zRbDnBHV)ZeIr+tv8y&5%JiWZ~O0Ia7g&G
zTS0V|nGpiBJ6#$}^Zcw|-@%|taBOG_&v#u!7`OymFA-Hft}89R*m@JX5A(~Xy%k(G
zyRI#0i-TUMF-elm{6KZAu*rwZR-FB@emz;HFXji6`_<jTFz_UF9#8BC9N`uBMT$z%
z#ch0`OS$Xxp&;6;kOZY=P!WQD#ipe4c^SC0c(KpLnCpc@%nxHL8My0e3c}190a8qJ
z+$mK7i#HjE#@ix88N-1j?ZuD{NE1}!1uziEAN4OD)rkdg&46QQH(&b)A+l!2-k5$O
zv%?W(#X%psEi3oxSsUYjp+r6_-90BzoLin!zb^(XW-U^zy!n{!4B!ZY*t8Z$0`d*j
z`WT9NtFGne(oG*1trXw6$YNOg#t43F!3nmkw+H|+$+}-nOdm)FDTe-8>9F&e`x%h_
zt!l6c9J1uoc9r857IdzzKT2)~s9n+`%7(cy3M%*aPhCHAxDkNbv|VX_eOj!Lpuz>;
zK!Z_7?3Aa+?mc>Fe176OTzwIsAOXNnGB_2C@JdM)lI`3!^r*<i5r~GbuU^<f^l$d`
zzC~44O}+Su`u>Co68~$IpDr+s_5LdQLk$6@EeAaZK_-lq03j2IL&R9i<$H@OUnA&(
zM3dj1TU*inkZ5Zm{Mm3ZI1)BWL-T3S<>LtR_pgmhK!u0Ny`!@m$?MCHbmWi1;!mX;
zg?sD9snpK!i{yz8TiwCEOp(Kaa`(Lt9s$g!_oLgCAZX1d_?+Bi1~PWYZVvgP48jy*
zu2-s+xn1*~lBbi|!V!3q{^<6*V^#wEW0=qut-hyj=-L~0iUXqlwv-iByBkAq(7vsP
zTohqj!lms`$+i)bAG3|7)9`rMoA}3k6k2-B#;D=Gqx&C3AEgeH514pTm9cuYMP%*A
zZHBW66KW59&ZpD41dpRJXt#%K*e1nh8YMxt9B&e_YTFk%C0~vYg4oyeHAex;h3?)$
z6arTdjg{mtwf`wBQhuAou1eTR?ug?UB_kB9LXePku@)q~d&y*!u&Wrui}hVUqZ1HJ
zCa20C9OddwZ}xU)&x<2Ko0$<Zg0M;5^qd}QvCI(DwvM5|%L;TBp3!!K3xPQoJWkA#
zweQVE+H+?Q4iEm!^A-5;2MIBssWf%1>AFCZ<8f9G)Oj>elS(|`OfLTQx3nad?;R`a
zFGNv5Z^fT??aeZ82T{Fa>r$t14XU^r<X-#l>L%_%NHT*OK(>Fea99rTU(B;jhSb<g
z-=A;p0*@9BmQ@axOKh8M4rlG_Z}S@vRNi#qo0lxEg2fsNSYO8qde9o)zj?h=Qz@-t
zYtH}Gc7HhpzrkhD<3{XB2J4IL)z?p*a@Ok%k3(gl*`J5mMlSschU%P|E@*=P_LM4x
zUd}K!-=mC(@h<t81+yY~HgbLR)pct<hg`afYuC}pB+34<B^M2B4hZn3QR=9PMGh7o
zX~dq}Gb3KTS`4slZN?c1Sma3>iCOMRsF-xNdK16WLHrXWoX-U8ABkrKMcJg149Uo1
z(raDsexS~7c#AoF<Xy!iVtw!QVJCInCud&l&PL&(MJohB<*gRg+9zD#;uz^4qx{l3
zj$pQzpt0k`H`6=_bp|NjdXO!TgCR~1i9l_534xXB5{P__1~AD8ET(m(UQpdDQE|v1
zl~S)-v8xmthhrf?*>hzxT!H&fA5-ut?muNFaJoCNUW;BgGKGms>_XnGU$*F`yV!4X
z)tjS>5BdB+NZ-r7rv=i2L&2qD$RD4Wl0X$+@C_|}DTUM7meFnsy08R)JVRtLdFv2p
zfPkz8C6Z|_Q!AVsW&g?6_OSi6{LfOnpBEtSrV|;&*I`X$ra3?I|F$|Gn}wr66bAQG
z|Iw+ewmN1-whR>}$v4+*`_3o@p<QMpM6i$VjT2P&!c+|!Dl`um0cte{Tp|9r^?uqC
zvg^$CYJp@DV!6!9CGu-j>qT3uZ+p9Qol25)6kUC5EpxGnY6=%#l8{-Iju5r*pak_h
zl$pPphzB&5qcNl{-?WyPlQFgJfK}VekM47HF~mqS&mhH@?Bw?^A1xAknY|yIB6~d>
zN&+Ie?yN&l+9L{9r@7beB?e-9jrf$tY_S=&-PFa3N8slaEQWLT&Jv$>5YO*Efvu5v
z{L%)KPGn`_;PpRuY3e<?*L}w5qN`$1eh-`*J-_9i^O7Kh6BMYsxTAEiSFBsGl%s+`
zIyVX3sT2S@(fU7tBH&R?<Xrk%{dpt-4dJ2S09EizSLl@{DsAHZ`}bL=?xE)H!omVg
zjOG$^(qi(l!@P>6ox)m6^|lVIW|!Ps9~z-F8xV#frw9jXgF`H$(r4ES>>Mn|_j2#-
zF8<zxt#tMvKV2Mllh4+u`M1u3UkS~WaxF-kK)|l(lJf-7M)+gKE7#9w)36=<#c}@<
z6O=G)zFj-4B>Y|Y=EnP8bn61v;SiNY0Zs5GQe!nf^<r!kT^T0y1bJb~Ob2Gl*TSo<
z-Gk8}jw&`Yo;QT}HEeY)rVQD#7uXmt7hb5w)NcBL8%!0}5S;i&v!KD`+d2g2Nr`gM
z?y8*d>()N*(s|c3lhS>6i&XCPu(aW~eqRK&(q5FvllJeuv3eUBSdsUa9~<kj=5P)<
z@+P}u-Z=|;y4s7ZcYsS+_<s`DRBNd21?p57>}S6KFUI@b%s^@EwI5T@ivi#>WS?_Q
z!#|nobw0kUbKWjnq0;oTE^{Qouw80db3#YpNc(FiO}}FD5xwFRDOzw9Yr+q2&0bv#
z5EXF18STGG;Puj+8AtMJUz+=+;W9;l8%#q1G&}jQ8RsQj2C3DpxdxwXXo))o=$VgU
zC@s5Ha6mz~PYU_sue{d4>h=54?fOJKUw1sYQusJdd>5d;_hhvjPd7J}>g$2NxY+mw
zTwkb+ns#W2!p(y3<FZ&y0i)0Ly!)e|Kb(LkrX%yo_CpcF+w>;p^ymWKV5P|&-ljD(
z_j6B;x<Z&3We_G9C4!idJ%)hCmHE89x6b6!gLtD!d!Ch7m+9~jQ&e1Bzc56n2c_i^
zsxT;8re6MJi4?U2P3mIdQdoqxSV?M%5MwBkDB1Cn%U;SyKW?3p@~*eG;&gshu6>=D
zPiViQ(ctxR`w5G!=ZGpQE{p$nNQpPCk4c4>bX&L=deBaNIsD)^E)Xx?lAeyHGu1$(
zkm7Bskav*2ocK=<mB(kKw+2{)G3KT-^xF22xyH}%g64o-@y>XVqHV>(*<C(F8i~O8
za?g=xrd~~=f2;n%wL_k{BU6SJmKp_0Y;kMUr*rs1m3m2~M>%AnNAM34elQgM4(%tM
zERoFVDGb{F;!4gI!)?{k9i{2+Cmeh+OQahSwfc5fA8eDc=_sq%wFM;eKHUFeziO&)
zmDxXvtZXxLV(D8?*krJV4yz9)sy>G_IP?(8!?oH?8aOyO$oJ{>b@qOIx{cI;iBH9D
z<}_Mb4|1sVI<E#_gmU66k<n`RnLHOcc9J?+1L~v*)ikH=M+|bj$4SRr=*UGvu<FhO
z`uGTdJuf%k^2~q+l-?fBssWBmI0a4Xn;B;stF_A!?1WSu)?ZgCK*6c*Gj0-uO1ilS
zv0q~F)F_zC4@#uWomN-*!|LyMXn#S!yNC$8MEiNsdDgA(8}y!UeTG8uoMGo>ng0F9
z0$R8PyGml6Wq0CdW+iR7Ib{1uuzuP&5|v-U5hZXY{I8}t@Av5BR1|0k!7vbD60qeo
zqZbVHdjqljMynj#4&UCc`P6WPvCuRWz&Qa<0wf>x#uRafYcg{;R`4LXWfL6tMTsKe
zkAFTj%N5-wLt~TtzjF=>eARe-y=$|83dF>eyjh<4sgdorSD(z>cS;HpP+34m(6ZbQ
z64;xUR|mip7&Aelc;<EDPXEE=hPM6D!|D4507zJ1a=GuN{uRiKzQ#ZR3b;%SHujJI
zSy$JW+UPv>1{<_LpT;}%9yFUfig76GBmMmFJN66BN&9QaKrtW?&tT~f?Qj&?OX7~3
zhHtJdE)3HLP}33f_9+_JT{#YC)Y^2idk_jL8bIoUp=p`&p4zQyBDS5N{^Q+~t?!MD
z!p$Er_&fb$|4VqR67cpLE|cd%Fet2OA4JYt43(2DytD>k6G7-uQ}T=&^iS(L^b&IQ
zIhA`>2l<fdP8vPMvd}6nTzRUM@A>Ck3wsfRdb;hGy>QkqN#)qsZ3!{QXug-S+?ZNB
zHYF)nW8*^(Xd-W~f|K7)$4PbD$6dR#6K>UMk#ey@mFMk=XtfFWrFWN<OqH!KZU-R?
zO+%KvLBdDpFvYgXN$E6`zg)U}5`nT#2$3P)hxbSr2#P)h&W-8{6okIHM}8k)chcO+
zZ;o;7Zb^@>2+@K{Fj-6pM9C@<eF<NxYo@BL=_`*9CA3Qwx04Qx7W$mg-y-ZAMu^IL
zx03vC!;|332^w7{Y_T%$INdDPK&3bd%LTZCsGCmo`Bb#&A{J2~>4Y+gEX~JWpsXjD
z0I!4LS>7RDch`eGq#?9;{Id?jCQoqfZkn>T%%>-FtE`M5pdS(a-L;vG#)Ox#hkPVK
z<M?hQi+2i?)SFp+G3J1|f2-4b#Aw|2@E}~}k4xfgKut(r#rsiYrfF>I<d`3F6bgO9
z&HL)AgLBcRGIp4rRA26{-*{ItY~O*#{-J3j^6s(MlA8vIwzqALn<VV>@&qS-<~q7F
z;b=%N0I=D*<Z&1)3r%FDb9tbC1KIBy>0ywCvdLOIXP!DDBf6A`W*djL9WX#W49g~0
zX8sUWU>WBW7hsZ#$QwO3Mdur|pa?=8F-M012b(*+JR|S$_Ha5<n08S&*%-?XwiF_B
zKc{2s-PBxdZBA$7X7|`}{mu4pIbl>|W2r^Cm2oabXg^(6$QO(AfI__vb2ABKv@~hu
z_`aA$RXqYi^rfU|5j~98eypv}DmZ(OmZ>{@>54Wf@20?`-$2O#=SEiPg61IWk`3bK
zCbU)tD(?<_rR+H|FevC3E|l`XGEg9b2KV9P<Es*L@*1ARiJFaR`Fkbl;n?QL|Af5r
zFn%_Ku*hgdV$yzMW<MXRpe+pakrc||lm1m9CZ_;^2C}Vxx_u;k5Pe>3>_96%tJDS&
z(2$VBM{8=6{8P<zIe-BLq<tNzJf=k8I_~eW<p=?15(?BpqOtjDj{{I=Pb&Um$8lN(
z#8YueMgUfI(2Arh<Seq5if>=#I7wiYpo-g+u&batShwl^oGrK;EKa~SNcKNgS-px*
z@$N$9RBFA|`FPPMv;rAm1O2-dp|Mm8)DL+@x2IdjgVuPc19}W^*t=hLN1!6B31S_?
zi?P{ck(z{yq}#;m!53}94ekS0Vv7ClHU6|WY!;!VFm(h)FnY=)bZlAZIKTP{yUVZ~
z5MnvFDTJ$;=!2XY=zWlvoHn;ET6VVPZj=<0s@|5JKgo+xgz;^w^<{=WRGE`I)@z4I
zNaMZo;17?0Ldlb$2T^ttYyQ}xVHU~dwHE5dS?j&>2%A{0mw&k-y>0|^FklA+m+di>
zXpQVB34_#dbnNbm$#vp(1B?Qn9V_Xb6(OUB1~~2sk#`n>04*<>AB3Cc5=<kN*9^|e
zO^Zdojh(}g5rMFT&?K?s5}e_4OTe7$PZTue_)SeVgmFO4YzAFX&HI<iqHYF<{^8Se
zSORy6yGlux|9eaxS)~SG@N^ma85?<XuS$!UfX+H%C?KD$v(NT=|2@WeZ-gzExwBQ7
z;BGBMLH5SIs)2t{RO15W)Og?bi-YVv^i#7huFArz2fe_elS^Jc?l7d!T`-6iE-qjh
zNX&oorZiIi!w&jGU&qm}Zz>hsZV|>pp-oT0C1FD1F4J%QBg$ztESF7Jmq=1{>pos0
zhg<$5nfToA6W8a8CB`qP6Bwb}mcssJ$q(8=`UpDg?h>x+merEiFg3vj3k$=b>PBvK
zV&gu}^_LK6#=46MYOAbdF6)XK@hQgCZPwR_XtkHN$5m}^-YCJu17~%dHnus071wC<
zx~ebD#FGL)oPG2!lkv7+c7m|uL6l2swe_@z^u;n_pq|p2>PPn6|Ec|`$!9?*+^nu;
z2tompn%6QX3)b~0V|s97{iw;-o`DNYuWH(;20&{eW%idPYJ0WLN_#a4>x+fm%SThL
zl{a+%yb<w_!rxT{<q0qoMMs6*_03ien(7vHEJSl&;B&I4)-K`X(&19Lv>wO!qo9JS
z`LL)d?;IsfLI)TjO1fY~)K5#G+23>bO~-@4PzyOpC}=dQP7V5jNwXX3wcQRBgWaQZ
zJ7=P&t~EZxp2Q>EAG|tzF7VptCryZ7sKHu`+}r~1wkE7ri>#bYT0VXqgU_mQy2<p{
znjbA9TBrbvF99%qZVVn^21Jm#Qoa1pltP*bgz_s^+8tiqKRjB6##zbgOo1)EqQp+_
z<N_rk4no;wgX^BeI-ZelleV{WY7VE8_mIYwzD%R5?V9@<qpQVVhWz2F`|)@Fb1jk|
zy2ip%E5_DHf%K~q=JjujuW#4>&b~bT(v<?CbR<Wu*|@f(SYx*5jiLa-|8QB6z_qTU
zDIqoHS9W0ZZTKuJ6@!G6Q0x0(=imxG@WzQEI1`5f_<^DNLx!6t1(-pABVIu$qC60k
zH)ExT=hBvkMLIw5*4{qxtzD$RX$CXDb<{t%liVI(l|8su+Qz5r_~o{=)rSrm4RKZ`
zr?whUD3VF0a52Ps>G34%X|<lA+P*ek27$9s;kiCqiH}=qY<f8}y#*|dhpEE9P{R&_
zT%<-nvUuzD_HQ)Mi#<(Rx|EksD0d~vLmwe>J0&<Bjv~e#?*>EDE`;r{q0!<4lxSXF
zFJQBnJ-ar%I2~bwfCQAX6Ni?+Ctm2GJO(4Ml#G7Up8o@w&pv~7<wA;CTd<}H07MTg
ztpPb;<hkz87eT38$`Na3g>_P5@*P95Ult>oM0&R2|NhCi+YM#xNo@f`O5wmux#OOS
z)<YaHXY4XisW*e~L^fM4xAg2<<0)!)wUo5+a_((XTqO*0*J=Q&g-jY^D&Sl&p*4Ev
zr}GG1o|2$mkHndPfoHXDULL4V`DUy#(ebg%==}431ZIu_$^BwYriTj?X!~>}<#u?T
ztNN9el2`B0h%ENkug2x8vn)f|gL9>8WI81P1neXU!zV?vr*C|86CWr;Ivy})>#e5V
zc)EZ+kFPR4a(B?3;<YCXh&9t)L3{eso6Mp&v7tK}XsffhaICZZcD%Kr9ly_&z57=k
zb1FN9ioHU?3irheiak@VA%8}M4JT>`frbmbcnaQQCoCA-t)nC%eOiaY5~(<^q|b>2
zm?9WEkiTFMcuznH(yFTHliwaEqD_W=hN7KWLq6gJidX8i(hLw~7drR=SSJ~*Tqe1l
z&kjJ$npF1e-Iv*-8rm~b>hK1A+GnVKhWNEp?$s06`aCozDLDu`hx2+ZT>!l|oTYZA
znd~Z=_oS7BqWoGy3@DKc%&a|uj-xE0S<_MP%$W=LAVO#9{Jd)FiSf;4o*7!#Q+3=b
zpM3#OxC|DOIumY+?bR4NV6K;d&8@DWU=T8%eP)wa#W?Z>0nU28CwQ4_Z;3m}xMPSm
zD2LF}M&2(=wNL&x96i?CqxRK4Ksncf4aHd;8px%ma)0@oNv^ZMW{)_81{J1f^Q2h0
zAcTbC`+Y21t;2M^(GDzxlZ@FiK%1JpP_$@6JY4WswI{yb_I)`g&D`yLskGw~M7!4#
z1eY}2mr;KN@dc|&gfm`e@t<bW*IW!<a((R}1rKxx73s9{r8#5O7O$Q5N~1yTDT#@6
zvW54f2+OO?5o9#}9SH3BM^^g3q{!CgeYs=TUkOrA*E}rukD@ME1noV;JaEp0zMtQ#
zJl8kW6`#?*Qxi`QThy?C2*kdK0K|(=1CiQGU0)A|W!?6OfJa&bv37P*1r<RBjQ9f9
z;%7OyxUI%fll6($PiFH@Nf?U!^k?@c|7<?Tsr1RI^k=W}_h^%Uy*xK3p7zyCwjf3`
z9mK|e!KbgEiQM$GG5=Q!fO$TAMI&7y^txj=y$|D$d?2GU3vK-HaT_rq$&yM~j^{^)
z$Puev-;<{;$A@p2DYA!btRcF2{fyuI77tR)RlhBxwKov~qj;KNwYv-eU;tiLbxz1g
zRgCvG9J+iYb5-yfQ@AuTuNc4Fi_fnTHYPeOoWZzkBY9om4UHubV-GlJ5C_!lZI((t
zfbn|5>l%6AG+ENLdFpQu66FL<Y2HHYGSevCZKB?p>1m0@6c<U*_EN_R;KH}CWx8<B
z-;VlbOrc%1-jOsxY?Ae~wxY^&5Bqkj*EB4MJw3>X&90RrYH&H^U#V_~Cf4<Nc`PGJ
zz5EAhjt1%!VgAo>7cdRz)^&#BD*ZSqlyb{L@zXPFRViAPo^+yN=V@{l126h?cW60k
zSCdTlr$Y5FcG7g)<50yj6^+L3=meht*boCaH$5HTv?#Sd;{hk}HE<h$7HIa12Jx~v
z-c`{xQch{o=kkblmK(yH|9bd805w6%z88J|M)|O93L%6LCi~zvA4uwW^49|F`t7&h
z_A7MiNQ0CseA}~qkIh|OCC@!E1e-;$d6$QbTrVABO>Et5luZ0`ZW&V-If)KeOB`k{
ziB9QSr$(V^+U!m8;FmZrlFzg%c1S<LH=G~vAnT4xjAUz_z(<k<iloj@x|_wi7Z(!O
zK|ZuLN*_{TcJ2G<=&lh0ePn#3>G=FoU)AVIg{TTx%&;TJM^dzjAO5JM#Hn$|MUfPg
z`T22K<@e<|8FOm;I5us!2Z)!>mW3lnib1Y3c?QXUX#D-Cz`a%*|9E1y_(}-Iy*h-Y
zvf1jJEkL46!Ez)s{4nLsp>jvdC)H!vH$v?a!4TjdDqIej+Qm)`9yD@feQEWTD_0u#
zEm$#y1U7D_j_JX|W|dUcc!V__W-Sfj=RZO?`y>~`&U12VFE>gH+c-^!Wt;3wnHR>Q
zwRZHNUzyF)MtmT6AX8|-uWI-1-A9eaT-KLAl&0Qv*^2_+<*y|LR0?u}s+!*2@x}MX
zy3%Cv)pqO)Zh3C)hP9uc{9Q_{7}j#@*1_ffoD`#&t!*A3>2v1bugXW)^`qJvS&OpV
z^l#!*x+M5D`T8{j!)=d`&&OgCbWgG3@k##JT)hzRoqfFHyuV)<*}cT6bsssgFa)>|
zpuAxBj_9PcO(jzA9WD0X);2+NQ5i1TQ|>42gub{2oWl6u7ooa6SZy}U;bE8@c6^#f
z-kxkicjr+S6-LLN-BA^Vs<Lu?&Pp^1(rTtl&mLTK!MQy2_%nR1Rdalr6<dbREy{sK
zJgn1wx~iJ$l7XdF8-3H@$lK04IfPluD*0|<Mb=lpRV=kz#NhKzawLX_sQk3dkW9+R
z^YYb#if^kcY_~V`XtWc_Ak<W~dGX2FVyYDU4>-MvW@ZGpNVw(kUnqPWF<a^^GApdX
zs5}%Kl+@xCP%L3Luw4k&3BfX(n`I448txl}c=8v{KyXu91UE>CInZl8Rw0Do`RAXn
zd%u4DdL5_$!m&mNHukp-8B2-vQ(PV>;H69WZMB2nRy&A~(inI`HT`<HNlW$-9c~yq
zN-9)#W}8{P-NvG|b_&ay*SJiMZ6o(f@ORSSJCjmYFJHd=(5kK<Bx1wz#+Oti-XsNQ
z9qzdt0_+sX4^_YXZ2Od1g)KL4$G&1+rs>7_NbOp)sVQZ?)<<a1{(}_N+~#9>^48V{
zb<b(3?a_jQ?DTiz;!?l+WkuDHxCqlnm!2MKIX>0Ks_oTUs9o81Tt~}o-BT<G0Wsk!
zabXP#Rsa(D-D>IUeZJWLioykvk{gCb;EK79o#`38GU~Zz+e#_9;F7DD_{OXB>3v+Y
z?Fx&Ex%|?LhzJXzn`s_j7nB(p6#I!iQX?Pz&$2x`>owh^=+QCW@Rs%HV4+jIxpP#g
zD%YgDGY9r{BwEZ0Zm-Vgzm&f(b?v?4+^kK({k9XyX7I{Txe+*{CTsg8pnEe-;{*&B
z3d3tm(+U5*h{F44VTrLLHWuOwPSV+5j9egWRc_7`3ReMF0ddE8QdwIdALxK1De$Pb
zZM;>zQoXF(+593@P^vQZw@{}37TRo^me4!3jgl+9K?K#F*zjHHci(;IcQ_n`g@wsJ
zefmftL?dtbk%Sc=uJEPZbj4kuF9D(s_k7v_I<QzE-}U)JX6~Lc<p3Tr?Pv+NJhyzq
z+6gCrmlAVmtdgoN?2y&huCXpHI27ZxC4>;Bi1@T~6%}`BQP!fuLiu@O=Ps`upK6O&
z6sU0N>O<GW+*wrNXWAks4y(fMLLZq0zE2zWqaeRpF1^p5k9EJj7>S`OE4Nqk?46f0
zV&v$S=8MDOAS7h()<u`gh0SWEZ=b$oFP}rGJTT|V3h6}{Y8Lf#Kz=rPUik$>`iF}7
zYpP^Kh$3zq8kzd#-15}Ia*dO^nyIRmXj1>RG&gHgL{irQ3VL5%^Ditf*u4YViDZj-
zb%^{H#CK2p!j(uKM(jl_yaqKlrRPA$963@b?z1>i)3^}9PyqvlviAg?cI+LxuJ2J*
z^%QLJU~+)J#t`h+3?FhU%K)9;r<$+lwrw$r)BCiMVbBvqB*W6`vbc3!->oRhk;@vJ
zLB^yRTe{qM`9F0bCyQfe-E#upXi|Q#`?ynA$>*MHk9}-oY{*haB}D4iw1%#vlzZ1)
zP!EBAq||+Q3N`?&nLBrGL0ga4mJmXSl+-RWE;}vs;%cvC*0R(Ogb@FVOiaxxEceI9
zN1C|e%rI7LuhxHFUiq5er+oU(^wM*rBKK6}WzKHcS8;CErs$-!WqYJ{S_jdxQ8-^}
ztd>5!NTglC>j$gVvQM%anznDo3|-f;+iivoDOur*oi4(q;4;O&PbP8=SSEcQO*ID)
z$&$RxNfC)@=^y=6dXg&jg;FQ-<;nh!@4GU(r?c8vQnHJEAjphLN_*~<Zno!pcd{~L
zsk1sFsq3Kf+?~shl?AnpA?2$>m<~*T>SvLv32qXCtAT#BLc-EBb47wC>5tCo63*fb
zPD97hR?;zlMfm;xbX8T2WSh`bD66r~OTg`i5DWM;D0dq~N>w1#0-^iw+5m&(MTY5=
zIol?mEjnCdK(B)!k`k!$NGd%7O&Zl~B5g)F$lzR>@a}cLL`~Du0jyT*QAZ}zC(ryg
z^`=WZfq0#xJJE29lzglR^^5dL-ycYc)OH*oZXU5>{kjSLrYFbHNME=$amfQyJ0$hH
zyjxq20pKgmM01xRg_wV|XYq$6r}iA!*KzTX{)U%lPm%AQyvS}tvWxw`!g7s-C=)T^
zDj)t>`gUnf##1$?ZELPSUCQx`bkAuws4d4Lexz4%)uj<6Mwv*8F)?|51*&SoW;322
zD^{&x?fMPG#l~PVnOMDMJ&!;90A{m^@`?%yi;76?T$k^ItGb%qyR)gNs9-?<epszm
z+-^5kiv?Xu?z}t-uU|50UOAgL`RFTJPUKuKeUwRkT1C#*M!MJKc^Q8If7HH@OzeEg
ztYwv3{&7i^8K9_KlhGmSxUx!p!0kgL39%`7{nGx;@8!1u?L=~a@$yA76T{X0(i1a8
z7&cCqk^u<%D9}}v#67%JI3v!$^E}As7ofT-uxp`23qOg9QQtca!&TlJBfC1{uwBq!
z1a7x`i`7~~3Zsv&Cz1eFUU0f0yS#CtiOmchA|Pu20-G&n$u;LV`Q*pF|7jIyl+h<>
zIhp+0!IhZQ16=n@T##KRar*<eDvE%V2yg`v$*6h(0KebA`M?G}GLSWS)|k{AFEarz
zgFJf3q{M^5&qz5g#2+$m!;~pUooMv2f^}Jpii_atwv3gG>T<#$p|q5ackrk!izqei
z*R~$3?O`d(FZFw*#Ol{2MdccQzc8|ES(R4ja9W??K??arbV}EO`2DgZKWi~SSYo$$
z6WUFsc^Plj_7<Z0CNI~;t5;x;akP-9myF6Eo2!WkQRv^pjv_!+6&!Y>&d68ad`I_g
zT^Tw0HBLOCFT3|-^UM<u<MDdQ$lS$`KmEdhQ~CjrmtVjgci%@!auUBSTEf<~%Wyc1
zeoQ74x(;72C}($u4_mn;RqWfdX}<13QS^5k9*am$8!DuTD9cTs0uY+eCBdX{Ygt~_
z>s@+uTLGK_F!)4=IQ2ODsUs&8vNOla30-XD6#GGnop9{9cn%9*y+S%M0+j25l%M!T
z<eQzuyMBSk0T+M`)MHnc>Zm|rVd1)nh#FEC(;w2muKzH*tofuHw-0s{K|&aGY7kym
z>4AbusPY)B&u)geP>8HAP;~Rq3cmX-jPlCH&w~(PxP+@Dd@JDNnsv$ne56BZ&9OC2
zOer!<=jI_5jXx-@hN4Qtq9}k6E5wJ^TU8~nyHt|z6f~8=tV`HkM!;vbSO`n2ds|*!
zz8=8i@f?-*aw(-0LcHGj#!D*&jDaIPWJG3ym;mNaGrpXbdAMiO4j4XS)!KCv`b}#~
ztE6fPKQv_#^VR^k6h&EbD8_3`RAg^0icU<s{rN8nNBRBIom=XC?zJh!=Uy`~?B4!8
zjCI-TlSrTHRIK`tQ;rK68J*Jkywa$!<w;igV_c*;9;hRd+1c5y$iyxmF4R3w4^{Rq
zwRWc(LV1=>q12gOUP)4{+0ZVoaVtLa_#;F`Me@U}*<5h$AZ{FH{B#yC`2(NN$Lu-t
zc>dWZ0hlv)KK)KQkx|b-!=KAnAcVkbv7#snilU(Fuw$!-huxw3T#zD5YV4H$KTC6W
z1^k`Et;$JBF{YOyl2V<~HZeab(lk3Y!4wI+?$cyltaaa18?Q_*T3+RmZ@&9|$>VX6
zrntNk-?XyW&<~)UNVek`L__%>kH;hYe!p0-U_nN=Zr!|qg@mBw-F?a2&UYF_taB9X
zDTCY~BIof#Sv5pEAT0(IL1d`T!`GMc+{b~5^?iU>b*N4I`G<u00@8u|B=iE1aE64h
z1dI^`2-rPxLkVGy26L8Dva?uF8YBSV_Cs+MBu5yQ#$6?ng3A3PfN`fogBT%?$5X27
zx_~3UvQ2ZSneY2uZn|{4M0^UQ9PT*<!7E^n6mmQk*X)c>=O0iE!^)3VIy@DgP?Nn{
zi`F|=cDujZ0mp5(BM_^ylo#TSSjmI28a8<^3Ncc-ZTv&$97MnS1gUxP6l_9DxpK~&
zIpr;%apa>UFJm$U)N1-RIw`HPL#+AnzG+rs!c`PiPL2sPeR=DU2+Ou?&&T$v>aQRE
zTXg?zd%TUjEq=+{^L5XIgB5FtwFRqW1Pb9oVUFfQRg6R$pVx34L`6jc;H<1fQIt9Y
z`1|6e48L(06TbY4g2E!=<6=3re?Ojiel*chk=%dpKLHSmz+y3CE_Eq!Aofek>}TDL
z)OjU%tq!r{er#B*^5GAq<5q2VeSP-@VJDVWYTKidQXlCMXC4Fw`5yN}$=1qp^ZDh4
zTb#gG!Tq-r$#xu55JCtkB^x(x6i%m80IXTFW=M2&^n27=pJGn|OE`F&u<Dr<hz~P}
zVTp?XiO>S4VeND7<+)zsGIpU|4Dxu{S(U8bZe{xKM%+$cV4O}sVRbX`vJMLb3~Ml)
z0B)2J35*a1kG<|(CzqU26*y*kF^x!ML*$ejb9M$<5fx#$?5(O{U2Td`CkGbR=-j#U
z-1PMH58is~t!KaZ;*09S!a^}}<Vf+{bI-w1R@XXx(vNex+;rtBQuxOr#V`)P&Sa*P
ze1Tc{H1ms@nfpV~SAMiILRPs3D++@^IR#`lBw;o)Cj3@eA@Y4I$L19y<z}IX<%%wU
z)f|pxr~LDjqefi*NWyJKtmM?Tuu5M2)PQcP$)4KU{RYERxEZN=kWwBqvOl#5Pf0t@
zuJE2;XQ5vapZvU{@~*yV*68$HF9UkoQdNbncB{h4-R(UcRxu(i(Za?|PnYlx9b(MY
zcAsX}t}n|=A77cDo*$LeW!!JN=aC^TSS4$r)Egr4`vVhcq%Nthu77@P+_V{o!$E9p
z3^RV1#a*}GO3xnM`Elx3{Oh5|m@(^TrhWAVAt4UTCKF(&R-5;uv~QNwSg-rmloe!t
z4-lEsWr#^=TG5`Yxn@;4OI6_d!Qq{@?(sa<ue+V)TdF&RJJb$6QmpEpLLZs=KJrWb
zuPY}1EnRway{fcA`?EAJb40*3@AGWh(T*b=wIp)n$dRJ3uu!<&Zn1jx>f2&sV!jhX
z7*;2#L)9LLI=<;nQX&oOU6&Z>lV~KNG{oLWWG#xT0*Kwyx0G=4=~aXQuLY6FT66>M
zl1($2izWO`8l?2`n@a*eEOALnTrICE&LCry)nr^>ml)%klbea<v?qcyy{@#R5JI(D
zt@l0p=%dr0e){Q<TB~GD>%&dN_8aMwzR%C>G3zF&D3=0j55p9Ez-&;i##+@YbMg=W
z$(%4FV;@PRSo!wifvetHI6+ibXDW)A4&ot@=K#q<5Qgz{b0B4efG$FaOOZUU3$akC
zc5hoXcJb&x-dd2_mKS)eVqF%ad|K;QyRj$s9?T($k2C^jbX`Ac>uWa$si`IXy4wf7
z_K$?;hMweDy*=AQW}c6jaFxEP)=0ZmxdXsv5ez;tM0kCY&Ix9#s+qpN@C@6kGkV+J
z6v9v=_oKc#SNBxxYekv5YP?dS2Nk@KNIV{6EiSL9ATPgwRjb$1AvuZffA|T2>gsBG
z_v*<nKTO4Lv+?=F$ta47qN+#<iUQhxY}se%)ga`vwQGu5$cz7db=;h^?%k#*MVqd3
zdnGa9DkpWd0g{wBlRY-VbkD`7hO%I-i*0+n^z3MTb;R)a_wE=Pbs>r<2_B=JNVemc
zK`n`_wMH&nxbVV|kdXHP6(nr^2)^lRyz&<}-HEk9{Jh-^J>m^3Wm7P#&@<j3WcHw}
zx$JsRk*Z2w_(u`rt}P?MbO1&>v*e*+r96K}abuODoFI>s!(z;*=RVBa)?m(-X7UCW
z+-Q7X-*$&>pHORbR8@8GuwlbKj*pL5YORuBkoo<;z{3mqoYXTX&iE-NEM<TY3=@*Y
zw7qpT8_BDR5PdQy|9El6q#4uq)rGzC)n6l4ja~SURbv;g6jJ_%5Vv3$uJ;E*B2z&=
zZ4su8E8kl9_PYNriQn(5_zPqBh*fKg%0;W!;5J)A&u#5Kn?g;ZO{tV}<)IkwP#`R^
zTYPj<>c5<MI|^3s^n7Yk1vj1>VZVNGMD(~HN;sj5jjl-+H7-K61C&>4M2D$t$aIq&
zW9Fo;_VfQXD7<T!L$#BdY5{QOZZE8o@|9V7%eSYFQT|X7#`7q6Kd7oIo*H-eZwnXm
zkK1qMv;n8^{@6EIvu*>GRaJaG@hd+3_%m!)E2j=Ph0xGYq|}YRq{O5OyunGK&0vG{
zu|;ZgOY^d3)gH5%mGEAjtm4j#Bd?5)GKng;PI|7Fv_vx*HDtSI2a7d2OeLq-N0nRW
z;p<{zIwzP}w6Ur{@t8i2Nk~02I=S=k+CJNfWGh2L2C)68eM$DYzW0%YS^~3i<3>?f
zSSUu19vv1H74@MIH35EqJP3>LhRV#ncjOMrin5uYOKj7&wv-46w;ANETWp|aShdf{
z2Cso4;J8c6VN@=hRR!jKSP0A#23EqT94r@CG!Cq;^5|6i>W?bIn1{n+%qyh{ZXqzN
zkinK#kALqIa;C$gJ7M{gb%#tQ)79U6^UV#lR>_egM~b5?TrvPFKK#&^{^bwfWE?l(
z9HEPoLA(NNq4h{*Ht-3^i?LR9%b5JblRLkhxqV+<*el*%(0A3?h3^QPeJ6;IK%PwN
ztl<tJ#QlD=zIN5yi-+xJ_|;z;Ycu6&zs3QkQZ`nV?tk}3W>m|dgPm7P&^;CSO45;1
zE}TFAz+!NZM3_}y_p|zjyqnmm%hw*S{^#l)#-*KiQV5ZuDxKoZB*mD>F7o5`Nvsxy
z(n<|ywa)Y<72JG&1UquPm{pXxs*t)xGQ!T7g}P^FP7Ad~6pT<MUDAj|ZrHe)&6~Hf
z^p9nfl$J4a^sDH)q{>x|#bU;0vvSux574t`ceZTZ&b2q*N^x-sH{Wm_KA#U=(~wf4
zDgu8C%Qb)x$#<IDFLWMmxW-s4SoRe8=n!kB!lmQaB<s>Wq$XI%DfTmWRTXFSHZsBE
zqfA^gFbq`)JU+?V9qvWC#s5gZ<Ltlo>|h^S+h==>WGmyfA+o8GitT7))LJ9KNCFu&
zXwW~YO`^#71~$F3Zyn`3jqK}&F(?r>NC=0#pqR=A8QR#)5M?)FBx_ONG!kzbBGCu`
zD1w_6*l2@vD-@cbs)i4Q=n&_Jt{&*)2K!zTf%1bAXQCW|1ToSMSwZI{g&QQZAvE2X
z>%L(9>yR)II$aKc!{HdwrAwD7g@uK_jT<)t$eIv?qn1(D2z##u)}DLrxg&NZho)(|
zJV(H(2#yC*fq2-f{7V^#T_86i#Gg8Hk)>4Muzl97{YrS*G;)*09Z`B2$a@rF8Vs1)
z&{&#?l#^GDU36ytJI*IAzW?I(<eU~Te8j4?>nHY`o)pzOv67vV`}Q<hg3E8UKHLQx
zkh%{Y=C}1&hZ34k&p)G=oy$)T8@6Vrdsu#{!SNTacOe8=OoAnwT%6g*&L^|W=on{W
zcY%+LTpt#bAiva)%cEmb1)Aor+oWY=WVoY~QXlwAum0g7HMCi`i^SL-xIH!Nc7uj}
z`}86^eG@_m6h&dh;`s<6m^tfbA|k@szJ3K#N`w%cd(Oase#<K=&~zPL>gc+}ybpW5
z*}8`U^tVg%)7LdTZZRoJQnZP0f3M(zej&_Q>ZIBuxu9PN-z};jK2l{*p^xZrm6&jq
z`D?1U@~m)n75KRDl(1_~=wiL5OOl0GzbeYE?X#Uow&R#Vtu?Z|yj-zbtv47)C9EA4
z*iaQ<GfUh)(J0u&<A=<Wx@4I;q(ne@b-h@(5RF~Ou_1v(ve*Sl5r$PW7!Dl<xQy3<
z!0m^Us=(DGHYBi(NipVG%jq1Jreh&9OLRP1W6PGR7;~@oHPF|Tu;#@&A}NIE`rdo*
z4Y>H?i<id5#mO2nf7Fu5V9cLC-%q{8{$nA8=<~O~nf+lNtJ^7K6_YPaQ$!emq6xn=
ziwdh>DfC*Mm6@pvyz)7DK-Ff}j#-+btNI->Qr-o0YVm~DgM#G21R0*T?!Uj>bi$)U
zs#-qfuZ^`CjEW888nfx(t;6=H<n#Ag)8!!Gsfmx2ns!Vj+Ef(o*||MCYUgpCtX(hY
z7s9u{8=L=r-EGWWZTys%Z>{FRt7FK_^HEu?Q&g_8WRr`hZj5E-(n{{RB8uGNN)#nf
zsJA3Hb5>+h+SkA9o*M=!frTonkgU+X_>h>)0--crHda;Z_nx&;Kqiw3m#doSs3@9?
z3=%@%_iOn58d8EU62=u&5rN=R)kb*GPcK_){HnV%&00vKLf>W;-J}ZqI;e`EuuQ}4
z(<!gg&@{=gK@t44tdgV{KW0_XG2Tp=T_H2i%aHydb!8>DWqazX0ksp!b{zE(LWt*|
zf4<Hd`N9h?gb5)G6>V28l&o*q7<wgw-7L_hkr<Ox5m;ZkxXQ3FCWqG_Q-on%+?-dl
z?Udlt17iy#?BFoh3wldwXjFEMZT*YRE|bUx+tONhc0>3z9a$r~Ed-=RH!j(gU9M5$
z3bd?QO@`$%Cd9aImk&1O)f3X<mEbM}TO0tPp`oX{-R?gM3kwBx>N>|JYA()O^V&x<
z?mk1T9{c;*pxh%Bu0%GOmvx9Bukf23lYe^Wr@ysZCijjSUCI5=;BJC5m@J`fuOBHT
z?!2`~Df8#gpTDW)6OJB8DP@<Q-9H(8Qpo61r$%A9pK6azh)p3e+Mqj{24$5R7Yzty
z(!6s1+*(a!s7hprN_MdivnmYhrNvq2$F2>L3Qg?x!1ucQ++M<*5HIS<+ZURPu4mfl
zoM7h6STi%1I_n-2Ex~TL1rW@pU{X!!x`yBHN0;#UO|g_$>C9VG#cwOTT;3p~+z(99
zT)J24ca;}ztP-NXS;`f4>I{k^`~Xp*3V&>NadI~svzApdX<j+^UlonjEXd6FGO(`$
zmq({xcf-wHP^Pgy!^3IE*_pS-wY4lS>$kc-TfWS8wBzVNZKn2#C!SdEc61jKU?8)b
zWL;v6A`Cb7@ks`8YPhgxml={c4H057q=9PDl;lpZ#jnH#m7dmRR!(>0Ozo5=ERChr
z_2VXl8Dz5|@|_wv<qfyU#{Byx1-kbg=0KOZ)WKa}WlK?%B#*}<f-<<<v9B13vU2RA
zzpWg*XcegQCEP&M$^C~Gg3A-Mus1jgv+)l#HtrEG`Zcj%8Qj5N>abzMRMliXyA`L^
zB>b*Id=+~D<{pc?z7n_hgDDFt%3>o_wr6|E%r`hxRH#CgTPG?^W!@UYI(fwzVeBsO
z(WkSObs28dnl(?;(Q+IP4InZx?fU+w^eE_ma*qN<`p@@Dao1;hRY0_4ID)KvA3L(W
zI9)n6i{ZX5Dk{R~+q=S+K(yFq^VV%hDe?RLC<2ymbu(>I1)1AD^c8{jz?phAd!+WZ
zP(?}aKHavTeMV(@baL8DwV!+a(g%i>b5oXwKB-n*9-RR_?bxjf*+o9K?Di5Fs!-w5
ziH$HJC2Zd1!KA?I9c~t_t9n(+dT&d+MY0`7JsJv-19-jO`n%Ge2tuiouc6v6aCs#T
zOJI^phe#vFFukZg;U=pDR3V88sc)MSVUR<gW{8r<gao>{qZ~#hnaT!6oQx7dpTxkp
z%L9cjC7C6XhR>leYhb)4fTru@Rv2w8YGa~VSP8pI>q(@|3|(UylT84ZSI6zEzeu5i
zJ)vO>pWp94ddbGfyB)Zz`XKc*z$jlNlrnUgL8WgK?@g)Pdj|)ORqrqDMx=k7K%BDz
z0&D;PAOJ~3K~#s4N!?aM9T0>N_pE+v(c+W;J#cc%t^c~X@#&T8)=%s=Z9iir1MV69
z_y_tc7Dt<x%k|{11F56y90O$$8>1{Mt28<}^}q3v>ZsutL^5x66}>uH>62<>d$yO9
zSQFD1J2|<VjkH7yeN(M;iZ@eOu5oHl8=H4|NYC{VS?=#8gjnDIlpgOrd3|g&z{``1
z-e0+TTdYj%{O<R9_5If<#=3ix>Yyw`r#wS4BfEl*9V|?r@go-vxq!dlbuU5)ygn~T
zu-dF3!Rz&6ci367^iRfrI)P1_w<3g*=~n9+H>oD1l!S+e5gs1S(m$7zXc8>_qmtW;
z?AWjd{E9Y6A9Hoj>M}b|C8JD#>c-fv4y(eKbIYHKO-}uy*eWs-brHRGryG+BX^9q+
zV@#M-7|_#>rb|AbQ%-!OiXsFSlRy!$Y)dt1iDn!&h4*KcZZEL~h`zTI$#xtYsFnQj
zcs$~hPd>>XJ9ex~2;sn<05K;+QH{cdl(4-}&^@j(X$rAG??fX;vNqN+qu97jV`}0L
zRRuam8Y1964R#bl_c#J{c{)YayTHLHv$(rdr*rhddjNKog4^GK)Dc*^5kS)=+Y1Fg
z&6s<~s6Y~1ROu(Pr2gKI34zYhM(k)XTwa}Rg@MQ7P6KOn-CMWYozd{)8iDd7ixCB5
z<lPS3mBWzS2n+=50e*ieK_Ce;yLs~FB}gWtt6%Z@gjS0gTs?NlwV?kG#QsY695P6x
zf?({*w-zoq<=<!L9`5-Z$=I+<ycicQuQI6z_FT0`rk&q%tAmF-cQt_2b^kGyXj6-*
zq_mNp6D&_y%z{#<Mtqcs@v};?m;^pea$&zvhW3~Iw5*bQuZX5&yg`r>qD(B<62hsy
zhx6#rj;5JE{=D*=snhS)b?V8zFf#4SLd}oOt@NChDom&L6NZN?QYb_M!UGnw;G3z_
zxoOySTz1J&>~<SMC@8{+pHzf`Pz34enar8<Gc$jf#j{U6j;bn{Os1NuI6nNE;m!u|
z`~5uf#IuaJISQ>pVhgfq7fGEjbXS?Csl#N(4tK}2E&(n4Qua9|xyy{Z@4Bs<NmW@r
z?OPH;(=n-nq-ZlCHiab{U7X&_PJXGM^_gz6i~PJdqlCB!6Ma%`lvQe+(aYX_?&_*d
zVTs*NcjoOVfOaC;j$;r51`MzPZj92^Mw!2&22yEW4L0W+1S>u?@MKjnVj4p#j8cda
z$S5&5Mzql#F(HQAyV7F_eQ(Y;^d@WLC2L*a`3*wo3Mv&G7M-}z11FP(m6F1sOznj5
zKtk;CYiui2aM#EJg<0$84N&6J$t*D@Duh9b6T)Z$l6<Gmo-zeJNTwvdX<-7l+bwG2
zB=`5y4jH}~rn1s>7l{7=$t|=N7%t}X5uA^zy~vZpM^hT|5g&edP!(ZTja_^n$T1up
zYjB)U#Ph)2hkGJNGHR@n2b{sJm~CyZA1QU*`RkBUZe6fo!LBwR@9<!LcA-Dn+fDv`
z$SI-b+U;R>v7Z@BoXlBSiNmJQG0seBrN&8JZTS3>UpGhb`RAYU`NXg2)2k<g&puQ6
zcG~nYHj6Ngo>ZtSiYQOd_Wi5IrR#}F>2}>j&9yvU4Nd6Kpp2}~CpqmnJJDf<WUu;w
zLbJeJW7+fhB_5wd*TLhHOkGsL^S8t^dqo*DCOk`rI8)7@zjhfl;x$bw*enX8Zi~ij
z5>$Eg086AB$%k5{UM00#$_vtW091PQr(T*|)IB;(O_Lx5%17hgej}=Pub#a4%76Lj
zgK>;~Zydj_n#OHIjlFDoo{wIgEbPjc{P2g9E6xa`Z>p_si@jx+hpF40^QGj&%2-=j
z?f6H-BG!&}>=&-R_S*jwLKr5qtZA_2{l<-SX}GRSU6M{wqLH9@Vz}Y%O)oOa2&5Mq
zBr~EWsJUaLK_J1%YECbb9G6gkYj=o*a<^fPtVLF-f=Tfc<=FrGuEZtDDh(ubc5`Ew
z_0A$uM<N9VS*}G{wMIs<f<)92=7=WBepUN4vP!_|X*vRY<6Dq)E_A#Mfbj6}$Im<O
zyorDQ`DdL!|A>?TdSZA7gn0sRpbzXMkna-hyq>3Syp@+HXC6o=E5<ImRe__0L>iFX
zy7HYRubp!LnL7{nRE}h9%4SqtI9Hq0y-Gs?AZ7gqZRoB44M`bvZ<|ZU`O0^rxrzYh
zH;i#RmQb3Tu>e>QncV5a%4&Vu4TB>(6jx~Em-zU<%c5AD?&gbM%eeJ|$hvAb3w9*&
z#e`3Y2oGn<chh;~!GDppYc~?%ijP!77p(FAyDUHBeJSfhBa5?l?2buDz2-wvHTO|9
z)D}^bauQMwW4<lMViEZKl4_5{?~{aC6|6cqJvzDsMFYQ->@M`-^5`tzSdGIXuqpzt
zl=yTII-pBT2uy%qgHS0rw}DGr2R_wYWTQV-mY4NI-S?24xmpO(Eg~`X@te<!cx7Xz
zC*zx`Ga_Gl;aQ<53Omv>c<F`zF#6RHHm>}F)C4n0(I$RgUdeGCtvq>QEPD$5tlVBr
z{~mVM>~w1%&nkT=XII9%!E5``vye4|{i&pJ#68&B(JH9bj|7N|i&N}&`>lWx7rFfL
zy{FDqxCI;YATi8HG->Dpj<myZ39z-$xPx~VL9YaZz=c{NHWZ3#!slE*LBq(!T5<5r
zdG+<Iq=cQtCVDD5VYYp@Ojfudqd4$7RK@7O;e4f(xEsVO#)TTzQGjZnPI{4HT{NqP
z1=SHmo;6*PQ=wDfY;1*Wh}!S9<E0ooW|8{ROD_!?IB;N{l*kcxt37`6NhtCMs410s
zFp%PG5KDOC#w&Pr@{0XAb=}xSrwJIx(YN;6kovvABZqq;M>4K|V%6FMh?OL=@5QGo
zW@~86Ee{U0Wf3))+ty<pZj|NhS`nL^I>c%g%dQz19=bhS^3ji_+&(mtC$5jBs9ZzS
zCBLP2X5xhZ5gB1@%AbAW5&m}b?I?=E4cA>AUsY8_?4A|33!(1>W;Gn6I6reqbYj{a
zAIPeS|2BmxHUzh-4%|*jwUpR}KnJV{gexe!Sv6#X7S43h8-<&daFm*@c<t)hXb~2T
z4^GisY}CEu%k#5d4{jcwm^Q#_F<rT<sxw0_jP0{-!v=1>^Byi5ay|pkH1;jeKKUqb
zzU^Sz{Bc~^FQl$q;=DCgWajy}^}+~t=6d<x&t)TWveKIrWIoy!vqnfsO#DA7L=z^b
zCfX(^>p%sNju3z7x;`a0H}|)OZBpanl6+?K?j}y3mz|R{q=o*c1D6FI-kzP4)2rt5
zpMd~*2MM!ta&j(h(1(Y;3mw<WkCXrz5>`klzt79d`@K1$oRXZp4d@bZP>6Ba+1dB)
z?Ro%GQc~6-W$$J_-jJP>b5cuvv?Xe!f9f)}Cr+H$Q3zpNYv~54*iqkD5k@SaMK!9<
z6jvG6L@ABLnPL}os!6)3g~I~9;$eG{VO`8BF~}uAY7~T8j6R|rjqiH9*{F)Mqo|$)
zN(noP1bq^9EX~z0YygknAg~RIoe)N1j4g-=3L)qbXT*2f%?8n}MMg1b(y+XCk851n
zro^SOt5o3A>VJE7v*E@Mu^98nDy_ecvdvJw6(WuUAUr&L5HQE-bRKz0H&5Kq3q_p`
zqBTnZCLxrccyjo8yfSfp%gtfI`wK$DF?}MAgpBM1fRNm<<DH+L?$#~`v{!7&P+o`+
z*RN64eOE=Xghn6AvWOnP$MgH481HbSC?|7sQl~Cs&N)70#2LLEm`#FrzAxdytD`9@
z*VtU%hl!t$BP67L&A;xNE4lXSE73GRpH7&>AIp|=)AiS#?DPA7OiJrHA<v}$Q*Phx
z0ps}-mE39fZYdx8P^)_5UNr<W0viI`KG?WI5L~1*_6qImB05udvru|xAefw-oNNc`
zq(q{VyFPX2tv8K2@%TPUS$R3#yLGh=IsaTtrpC1I;Ro*J!w?5QPJ4r4gTk4&ri%1j
zFLzuNg%B|7&&rJ4%ubDnYAyb1@;?ejP1=4)d9NbGf)2^a%Q___bggZWSy-40tivF4
z4PS%O($bpn0~+YxzrPt65`3P7Z-d4U_*^vp8w8IV{4!z0{&fJ(7VwZF#De7H<h3a&
zDJM5MIivV|a6c{nR+E#H`y=J?dzs6L9XfP4tVA-HJIBPtbfxy%g3BR>+YHg}aGN3b
z(>Dov#MN8vynci9Z7($9C~ILe8}+L@L>iXJT8J9?piWVN#~UKjZit%)L_O<*^hR~>
zn?_~{_#1ejgfRMxXoib~T8;U{g$9lpW>x4AX(SUhB#~af#<oI@okc3X8d364t6{P1
zn^a#3CoUAigRaq;S@&Oy#Zq?<9Z?cMPY#Pfs8ca2(X=*VK~ClAzsI!P6vF)Ghk)ib
zWk&;&OZ}nkmAhKN-$txhU+fg$?aw)iYHf4<NU8hr6l_3Bx&HUxe=ltLjAH{X>3i#w
z*`*t++&cYx*g30@gB9D{Y_IOe=O2#`Ad&TuJoxBSL_~!1{@DL=;rZvXWy?0+cxCjh
z(bf7ZwXLEPJM}Ut>Oyp)wn^WM<F%>-t(Upq_(gZ~v+kZJkiS^X%3POL{~L-(P8)jH
zZNo?B7ZfO7pN~OjpG9F|F~e`ao6@qfCi}bl&fB>B<`GPs=On_RaQo0m1TbT9g=bZ|
z`yMIhH+yM+H6qyv5(a3hx-2m<vEE`Q#FQqQTHJ1Tv%ALJ+}y!5VQElxy{W!hA5tEV
zlz(*S(4o~b|AdtHv^cD;-`iT_9SDLW8x%#U{~MTtl1SYC#%UlCc7t4XiK$PPDXWGJ
zxrW8Eu47@4styN`U>og#4ndjKPPal?_1<HYN?nFqyCFJ81}4w!D>_CQ^Ka-1FRs$q
zkS8dwHZI&^GU6tCB{Ytml(KOha&u^dkEQ*{lmN243LMvl19U_8H@j1>#p-t!#tA7O
zYQuRP5||X_yu&?*qXC<98KvnhatZ-(^+VGmRg0s|H{{;pZRp-Aq!7ncHJg@DzI%7c
z#$DQZFHI_1vUHORO_x+S&f$OK-p6Wf+5qs@yW=?d#1j~H?N#WyWX|0AAOxokIEAy%
zIO9$s4Dm)yQ9ImtZaBR<Tc1HHD>h00YvZ*ld~KX>_>1o5d)>WI(dDx#vF88&^U^3G
z6<L=&s7lVe?2@6PxVVJe+<cM}6Z!k_8~D%DkMq#u&!Fi|KYE7WIE>+Uk78Y>pKW`*
zyfwAB==-G=mzC#a%x?088vNynWDTNKRh?A(M$@z=$)BPq!EWn@xT=X`S-bP{_Q~zl
zO0Y;>pOzXIchI$36B83d5aPz5!^JeFVmGIxq#OXVX&p!@>nva%k4Ly%uKJyJ%*miO
zAVhAT;o{YsxXEKe;P~VkVRB*ApH44^?DE!G5fZ~1Ydz*x?7i>2pu#Cpj^;9~_aD2<
zbkd7eG+jZd5h6c6scDj`OCxjL)8M&iazJ$bvrpIcLciZH+-~=gLnfcOr61^b9)gi>
z;ql>T?X7{-mD{M*s62`xkpm9*9F7KD`{>Gb#T9buzO=O^b~{T^&21j;P#Yhq`~~78
zgW=5EQLtrm#NdyADx0FmUBdfg-^AR6ix*(l&vVGl&Eu}yZ=u@l=8n7XWAHg=bKn2n
z#k3hevNJt>dsJfT=%}Q$zoTh-W}c6GE{_)RQOY=}F?oaZO&zCI;b|i>`Gf9ehVI^^
z3tqB2lri^S5pByU_Jc0`5s7IxMkS_>*M&aiwKw0cyymLQx#Xe?xaa<d*|K#zy?gcK
zic2rzmDk5K+yAv!UBTTCz0R9cOV_Q*)OwfYWwod`dMp#kFA?H%g!o*-1OXEzEWn5_
zYz!$+OG!x?2#}YTmkw-hqRpj^j@5s5MH9zb#TqGLlu(qDYhGsxDgO>kqiLCyFrV4{
z!oiNID9TL)Sy%;p1S|<^>Huy!*zpb)wXu0KX3W?G7^Xx^By=4f*s!<?Hs(QTb>nZZ
zR+zkRV!i7)r(96wZJiaNV^pBKxzeNJ^*6mW^!Nis&1<E4S_73HNnV9<??>99Pm+=S
z9W0nGG2$(Qt+N2#Zws(Uy4~)TCX?yN%COP(CqW#FIB*lHe6rb3N1`D0TMofI4h6a`
z7`fo6kF7rx$jCLlsA;q)6J-gHKO|P(8vRIDNlD4lLowcw1}Rtgx_9e(+?%hCLQ$Hg
zJ{vb};mfbT;iVTwpzAshKJq00`oFvA+9j16ZoO;TsMlT_u%RUPtS-ss_y?|v{wzAo
z{ANtJiq9{(^}@&`O#G3WU_N6jzGug3m9nxG$)pZ^rMdV?ch5{sH0Oj`lo?l@6@Hr8
zB#4hx3AKyI9=Ixc(#hRy$z?fND}P_MDE0AYo_}&@dM5w-;BB6M?gg@T?dF;*FURBc
z^21Nf8yPRVcqmt1dPy&#w%|3Z!7)oDHC?|uJ3ITf?Ck8@a&mJ1z9%Q=mYkfNb2VM>
zh!B4^b&ino+WI%*n<kocPE1Tp3qF{foZJf|W-$O#h_4Ug5*lLno;`ck)V!AN$<3XR
zos)BoghAAjra;`>y?gg&g7)?e0vfrs{kr}?g!rUUV<``tdpp>*{mLt^xHL_x&!lZi
z;lk}Rgv2-GLV;5f%;jumBX+V!ykP}2wMI!Az-EH5AeV7f)BBnL5(!3}VsL`Yfx@f`
zq8-pP-blQ$1@-IKB?V5&#$3tHqCiVzEo5Ldxa{ogg{?EWR>BiECL<dqzO@}B7x2`L
zsZD*fENoFXgy?n1=F)}`5SQBd=ugr-6u9)!<r_*W*-NaXKnNielXY-g&qeoE;Vsz?
zuz1Ca6;0H3{c9j4IVq_*p_iiKVn&P{&AV^DhRI~&xlylh!Fhujc;@MZhK7>Xxl=zO
zl?NiLc_}@|*RykinXAqUJN~#%R+ewACN0sz;1e9y2d{}q>Yi-P&yw1mZ}=;H`EtK1
zXEq5BlJQy<3#IpyD7zSS%LS2x&N;zhKIeD`OE$Xb(ZS03Cx`UtlWJvsrZ*xwG3`c0
zGe2gx+0KoPi6%Naig(An#=jqXn&OfY{`1shd^Ke%+jcas=GCcV2Q$(fw=eTMW{G5T
z$ji$s@cDdKAw;8{qJ;jn|JR#|ZB$jY(Zfjyab=^1Qs(65=B_%(>FyOdIXMdv;@?et
zWUsEQJo7;NO-xMe3j`+a2=QlLUS7J-=bHiqXKkL`F(KiE10An5(YtqVMNyPK0BF^a
z5ya&q?S}O&(q5a1ELfXuxUqw4Vn^5viJjIWa8RGvQwxv3=`E`_Xz>YcqIBVb!depx
z86HGxoo-2bkz{SQAgfg1_KUjjD>%7l=OobUNpO68eBaiaU@JgTFXGT<N86Cf#Z6Cz
z(>VqdG5{n*X1b1zYdI9i%;iPDMsrL;TzvnOSjB8Rv90IfDcX$md(d_Ln5<^Auh_D6
z`xdXa@#fv@^>W|8ALr#4pCc+NlK*`=fr#)hhFxn!tVKsfvSh*R&bzj4xMjtXc@N!n
z`>kjE@Mjh4Gu+rMf)l&iShlsA4sm7_0au<G_P)&`oOwdK{7tR0e2YBr6r2*NGsa)3
zufsb+NNvnz1H)&AI}~$bl!+ZVUOL8`v6=;2cYDY!_7fi$f9Bg`UY)e;_g|k_zhd#w
z^UoawKyp$NFO3|*y$?Q$-|uJ4>o4)xQ~$wP*|-9%uInsVxH#9D6SGgcv41%t2~bc_
zkcX7>n)s+=?SFD}b2kE+jao?Al$E0~krd+FL!n!llauo~4O0Fapx=S^sj6yINjCoj
zC@3hXM2H!|A8JjXb4X#eTIC~;JQ6L0hy|cB3;YcXwic6NflH1s{yibNKIX8v3O45p
zHswhw+->)FYs!7x6oP*``!1Kd|EQ?e*_fxZr9e>ZG6qTrgA<bjzb|tGRVtqos`k{q
z4G9V9)^-y-Ae?i^M-<}hrVcBTK8JcL?Km=U@gu7?Rk&ocb;_f=^dBS?Q;Xu|2a6iF
zHoC6=($-@g$;izs`0uN4yxr*dV^2S){W^ca)M+z*kU#%2mkpb?@W_LK>UgTEl981~
zT54xrdub#Oj~Gw6mdv(2Ug9E5%v|bZPoWPXVAT${`MR^izf^U_TrKo7-q$L({;KbP
z)Z|v_=XI^pxl8zmNll+T@c6LVx=wtFQ={6e^TEtgIwzQ^a_eN|`Pkqbz;APY6vJ;E
zMvv}YiH(U4=&#ptJ-GMoyLj-?r+DSRWBiMjEdBg}ho5i<Q<c}pyu;dc8=jPM#XeQ=
z`b!eY8mw;OSYW0>)9&p-4c*3RX=#y2d0L}Zt>@O>E2NY@q+HwRICMUc{<?SXZbi!A
zcrLfk=liO`n2p@q0*3ER9;vM$rKC@vK6Q6NSXfxSOWzcfIO5lh`q;IwsD@Nd-y|d1
zMgUb_!PWvnVP)Hp$>7ZA=G^Aypl3@{?&a1%fm3H|p@Pe+AgD<?N;1BehI8<1MgmeD
z1eod#enN<Y3Z-auNO|Zjq`H25lgEWTpnLI%guoqU3Ogn%qO=t<i`3E0WpJz3kV6n3
zDe>g3MM_yDn|N0Il~9_OK4Z+g?>+U%lOy!aTeneET*Aw*k72^3uO2Jfv-8^jzV+^e
zFD8G@>o2{~C~OK~>7UEF;<}sIu``{4XP(Z&c|Xzh^t<`+XAh}~W~77=n?iPxpXe~v
za@iRnQ#2_ObdPe{G_Csk@mdvC`#>jk;3wV9yISQorxa)U{bF&y9*)_UoEDlqV~LZr
zBnzcZjZ=@avnk8NH_Jl#{NwR_JnlVWWB0B`HfZ3P3_ANPKAkZ6hvMw?+osQ)b<e|(
zKcnx-&Y`fdh^PNEO1|>O*b$|<=@a(#d;QB2NeOvP91i&9$flEJtS*;pQ0*J9*Lz9u
zc|cK4PR^kVXjYWsCXR(4NLN)=Rl|Vb)ZnQF1qF5X@7&znIT#t-4H4>dyKgv<F<S#+
zVPO*Bk3aq>p*Eo^q$7ki2x--ILnG1`<Y}^+4Edkli3X8U6#^>6v=*dnEawyK>aIm_
z$;R^Py;tZtP`LdHs$yJspQIWYq#)0wSA*?^G9cSMyc^i-#R0rtZ+`1f?x4_89qrbK
z5OLgnx4H3grOhjr95P5s?78WON-FP|z>xb^Y;ji0uY!+Di4Z1BtI4%B2d%mkZ+Rxb
zZ}aCjuLby5K}mMTE1!Hm@w_t!58c(h?<u_Y<~x3*yeTrd^C^A%95>{Hv2S2DH;&$@
zsBkjooe#L?%FF4JmI{E`Wa8iV-NU4>XLIHSx3FQiU~Re^yH)W0QYTlP6=v_(-8Nlu
z=wo_zw4St9df$CTb9%Pz#noLRb>7x0nJL|qQsT_H3DM&7xCr&fyDpCII&V!CiUNMW
zWb-Z$`BhOIclN`~{Qf5{8gf2ayLR*CSKlyw<}6&U=0jm`y8b#Ie&C)fqY_i^5Rx-K
z|8jEm2?GYw^Mq4<<HrB*{?eTEQTy|q{$+_I;BKiwZf@?%CJCpmH*{||lKqkJ-61W<
zsOybm!F2tAB7B?DjucY<ui^WedZS-9K3N=AO-X>a-+tSrY1%sDoq*mC1~#bj8c8-q
zm9o*s&|xu%RKE_bmm%E{MOA?$)~rhGmZtB)5KEv%(5o34)UAz9QO5OliD^t;rG&yt
z$;Mp6Qri%H9}ak1T3Y(2$z+l?o9#%++&&x@QiYJxV?BOwSMTSC?UTE!9rf5<tdG+4
zK*bQn7T(2VY5R~6PyTw)HFRCHDC6>1fDrx7LNNBy)520l-yW|$d~LM2eQ5N|PRSiV
znfS>^N#Ws5ORT;0`kQp>kiu=Z+*p_OJ!kF$rhfkeF)`7+_@Ad)wrC*{J@25z6NAgG
zlbPq^#4fgobB+(W{a;r`bygG+Uo5mU-_|NOexp|hNCZh>vF_z%-C4Q=-*AEa_Pmor
z2HkQ&M89sy=6+VQAg9=ml!C8UCh_!$SD80w9(Ub-3(m?)o_T&0FTXy9rGGBx>E~W#
zz`*lay=Gmr-^0_7KO(NU?BWkj>SmpL=S5N0*Z&c(-E~Q%5*==OJvzBdlVLi|;jc_2
zTYTV2DW#C|+eRM<ahdUtP?7SY;Exq44~bj50UBq7D#|{VPN_%7jvc!r#2G<{v$ON^
z76i8yit=gjF$n{bl9CQv<&Ihs<8ry`zyIF<f|Lsb8~Sy~C=mvstZG_UvbCY<kW(Ib
z%wdtZyb_OJ;?*QRP2%-Ss(q45FWAk26Wdc>U$?t0p-Y2kZ^)HomWW1K?Hw<Jp8u$C
zuIrmGzx?twweK5-8ngq1l*de#tPo=Lkyl9l$j7<&tle7eVPeB$_Jre)M>L~D(~77m
ziztQAo5~CL3u2`%rTleMmV3+MjV>iFQq^bv>HPhMVb`9}A*E?Y#<~p~`S{Z>xct(K
z1EhoI%v-=)?|i_9O@>-vbaWICKlm?JF8!VH<3DE?=doz3mGVlB%sd|^RSYf9$y!}%
zqyK#Ad1bWU>02v(<VcONS|yWo*G#wKJ)tBg^Ghkkg_<tOFY~iA$H!+2Be~@2e=_^0
zpP2UTciix|YcQKld^7cXZn*VM9(mwjjC<!z-W~HAD;Lk_^i%tD{VjJ;RNUBM37{x~
zM<2T1vSz1yK(Ef0*&DMw%CarhWmRr{mV#ciKi}xFzI#4sNFk;O2(C$;mXwr~nv|5(
zk;ZP|vh2LPLsrK$IXSt%5TenqUr4!6*;WmqY1;oaE)p;HBqt_5nv&Fn#<>vH1ZBo5
zhHTLN`#bhQKmsBzE>1d~PGK^c<Y%9K_W6DH-S?mnq65MN=S_tLH^A;61JA4#DBn|N
zxOQVh8e8RB3t6R&GqKC770Rpkc2?k@!|R7#C6KmH75Vlb)jl({abW-e7ko)XK~zam
zrIC2#)0=kfcfJ}1O$1SIEiElAeYvEhM30DwkhLKO?Wl)D?%DV1xkom1Sai7OaI~Vx
z=^yQ|nQl@9HULL*(-m!78}7U{NGUU-qoeoxnQ$zk#-sMA5aO|&`Kdz|ZE$^7RV_Ql
z$Hg|;Ue|RVeeyrt{`cX8hK6w8Lyz;sBM(wiQp&{1Q<(GPRF3acZ$DYJW*up%ofvfX
zSqwV+EOcFG<K`_a{c}0jDC;A3q-X5T$jsbP;wjxzjK1h2t?FC>=&%LIR)y{DlTSOR
z)8Mnu>>ZlclT%JUi6>q?mE@$vMsq7EEhR2ChV>gZlAe)CbhJ_Pz;3tk!8>nq!6jGm
z?g!&}{@JEwZhLg=ilUlRU!GhPT2$sA?NEI$=VoW`_r|xKNVb67+}y><Nl7I@K>CxS
zD3?ho8)-@+#PpgUYMUb^CB-3Nd{duEA(rh=UqT2|a#GT5jXRR3Aw=V3qUJlgwR`vO
zk8ZCiYJVv3+;h*#kt0V^SXe0CZnu2xwbv?#4I6fELPEkcK*byi7km#J--1>D-`<^v
zMOE%`0RO&c=FG5)h@)&O;SvgoVrni4u9=!k?t2KiW~sShg=l74wtKH-R`;sc)Vdmq
zWs>Fv4b8o9R}>evVP9vMbIyDJI1CKKfG~ohzMtpeQ4i;wc^P5m%e(yE2tY6^cVG~K
zlwA44P9^W|;%ry3zM?fG3f#)7SkT*A=>_r%?&i)FSAbhXXQ0R`l84O;iz~k&?+66E
z3$&XHkh0&%i;IhY>eHvsF5|GaMx!wX$%h;$Y4Jk{ZxWm;mU<x^iMLeF^pQiL<BQqw
zsI{IA{wpCVveUOJrCfK{#<k9?{y@l2fsl=TbyOAM)Bj!?=@5`k=@jA8p-4B<DJmk3
z)TQK-B7M0CNav+X!Hbl1H%NE4pme?a{r-OEy#Ky?_K!Vtc4nV_&a*qu%zS1h>j4nq
zeK>z9znIMUkYju}(CK$ss`y4xe-@`Ueokv5Z{xPh-LTjE&j30(=WA+OQP+-mYO7yg
zJ+oFvR%o!aQ*{KMFV?4p;_F)}X?PgyR0jtuO0Ufg{gWn9D^?CvfYj8~fY@g|T}1XO
zj0qPPH|xdFkWMM@PzE&?q~%_a>iSHX?d}G9051SL&x$cN+fqjOi71VWKmZ-?$H>r{
z*WHW{huNeZg%@LuU)skv`k&IW>!LFgO-R$<)DY(uX0GQPyw5?Z0X)v0N;{<Lr))cz
zG{2n}e%8*mx7$LwWQcx^F_!PBs;b)Y(^n!RO84+m!rt&c1+5es3xe}vv9(>vm>18S
z936}Fit`~52=2F<noSH2m;L9EizfW&M!fI%5VV>vBSF*#V_!?+dOb4I#J#6vr^G(8
zFXzZ#B)j%%Yh1*sxlu|x{9*fvFPt?GP#MU-kV;)WdL1k^lRrk<TO5Q!VX_HAnq80t
z3AtR1x7&Gfg?4Y>s-2ykA*#jQFHbg3vyBk!GNHJd-6t6jKTo(}NNDNW^S3Iie!tIU
zob}7<GI-1xSC&f|rY1#nfy3x|=hIaqmwU196lHAj!(~aJ5B*_lsG_HdDe&^tJ6_`~
zXYu2*cxX!l&PqR3JJefmd>Z@U)<0)_hpg^h;7&buSeTE)CC|I@ahoG+D_`lu3w-G%
zZ_+OyLj&u1u-*ejJbDKDk;YU`|4e7hr8?Pn3nd)$=4oHTdyvSQ)2JB<3FB{vPxuH4
zy1N~WY<AjUJ^`(I`Go}#fJ)E@Q=%Sr@qWCeFTkk@(@zogV;+^KsofpS;ew(Fa}Sse
z2vrF>RWT%Cky5ZPni*}*p18T}*VfXiIAt=%D0BDrwrYc9BWHu-3yXG>0mVBL@}SZf
z_jx{o(qR3_jf@X{6_{T!s>EeptY<janKtJ-r<|&(0n1xk=SScyys1Z{{gv<%jH>7S
zFl59b4DE1&Jvi?YhHw7|6`j}da${;|jfSo#BR^vFEN2FI0qg`nJnp0|OR6Y$n!zbE
zVpv^;e8W`GO`3eGg?us&<rYRRF_7f_Z|wk58|7aSLcKhK0c{}X5;37Ef6}8gQLph!
zfoQsv%RmhS8v<2i(ZxvUYbx)=2-5;S6JovEiT9jb@D!t!aQ)#&R8@KrjvC+bs<3GA
z8tRInBRJUKXSa2EeVfXpDhPD!nwI=#+2Usy?0ahN<)LSZtB{5}mpRL!x*CrA^V1*U
z39qr<^@*aO;R($wd|p;hOW84~+?bBJxd(cPX`d|FO?;)i;HXSx`s8+~AO!c!MZ#_V
z-M`%#hDsdm7eBUsA~TaS6W~duQ>!79yraK#iIWlRHoC^px6QQCR>U_rW@$BQKEcdE
zu2|r@vDweJS84Rb&5-i!sQKe9?7t0ShouNYN%ITstKCm=Q^Y*UB|ZOjBS&UF&DJvT
zJd6w1(((a|`(JEYrICHSW4$`XZ~H6rbER)`@7-)Z{{Ft>@yc!zUgj}vI5CWDP<ms^
zF~gE%$xF7tsduQiHI~VHZQknm-;X5{Ce_}aluW<#7bXhUyw%JmMBoS9($G-9k#pm(
zqCUJ}sb+6c_HKeyp7BYex6N)8eRp|r@8H71f*C1u&%V%0t|SPUJXum>*S)YH6AhI1
zGqC5Up-^^dQr?H=#i~+eO56pZq~v5|a6~=ApmIuIfwZWl<vz2gr{{v0LDIj%fX6S{
z^7Rt^_6>xz3Pa@vA6@rIMG{Zt<;{5{f}V8nhhUii3bn!Fg+e(5mYb21%24(^N!w&k
zLQ?ikrUu%kd`vWOiB8ss90tID>%%wO0j18GY^0u(q2X^4CZ?<d>0x!ABsIO#$%)a?
zgsZEoF~lNR1z(IS61z3uFUvRzuNJNOdgO$8<=>Wa9h`L)UWYOqrMQ$3y|VeS>XuEl
zjr(eiu@a~fI&M7EQ4N)S%7`3OSB|BJ(a8LZLn5pYVq}WVF)cm?<J5Fw%@TW)@xnGy
zn`EQ;;m*bv4ezFY;|h}S@-Ch%&c==AnHh<C*<(pb&AAh}Yh_48*<nh`=jdb`4Ie_?
zb_q9!Gr~3pr4x|pdXl*>@Zys)Vzd};b2Nv4^^V(BIdzlxyg9)FP8z#yp5fSZ$2E-J
zJ+k_K3ZpGItK|zwzY7Q%sq|*13U3NwFl)vxpxl!+iQ^zx@8eKji?uxr=13-|px9DZ
zf__QzWNWC)l%irUh3oP7pc{^`g>w2R@T<|~t09U(6v%i5@K0CajvXOEoT9)R9wT@7
zkQE}2$Uk@xYv!~<@eLTL^if(6IzyHjS7QMAp`lnn5HsHE=_(uip`oE?B#hD*h!=UV
zQV#+pKxZ0bacv_m*MQmW$4vsc{+PNk0hIAiD!zv)_8-Hfb1d3A%<v2>Edk65e<GDy
z17a;30yK7C)H_o`T*t4*(lL6bMdLeS&P&a=YgJWM8nveU+(~NhU0ek1MQORYxKv7*
zp3QRj>`>#Ghjm(es*vd36x3+*;NQo<X{ktm^|=H^e3B1+o{Ym!;=AfV`a3+lefYIi
z#_0yY;`!HwK1XwJA6|r|n~V)OS~?IRqc<}Ep^5~D{&Lh%`pi|F_LKz77rh3mV0sDK
zh`}X)6I#Ig22JPN#E36We(dm%1uBUYr1RVTXA=GBDMpYq=kaBxyA4tS^V)U+wWmIO
zeRSBAA|1#FEUy;urica-%3kd=ff#p#9oKX8sG#j`>9^660&%cl=kpd>xT}-t+1DoD
z4XN!De>sh;CnTWaN|CnVYs@-n8}l|)No$_FDbV9GPESDSsZ*TwS23Gdoc!^6si0?P
zXJ@oSr^-OC(1Q=89X{p=FdAoZQvB4mFv#C=1H~?RIVsD*u@N2-(Y=)YeZGQ`(x-EH
z*hLle$#%wQ?%}auA@$zVKK7KXM+tCpMqFKAA03~Z%u+vkB+=8^`6gDcc;Dm@r%rR@
z*VYsLMGT<>5@rsn5k^drA>fME(7^o<MkX!W((rd7NkEq}E;6_1B^M>FqRKO%E|io5
zfa{sEtLjy-tL^|IuuLWpH=6Sv&*L5?J@xm*=vPd}iCFy894`nhF7S0Ls89y}cmdLU
z;yZ&Jn`moma{)12Ta{Y_sd*bP%T-wB?DVvXmX;PGFOThMxXRjJt8X}y2``tHx)6~q
z!3Jf2O_hn)LJpEDK0C?%*vlBt>T<?nAz;$+dr#8283b#}+?M(3U^Wq+dK_@IpI&aj
zg9$;VB^%8>*PS=C6aP5yrqDd@{q`)_$FQZ9ZnH(!?^kqu@u%x02s=lCJZgO9B2xe2
zV8q7prW?f^y1ZqT=XPp3(;vk38T#SS#bak4U+!QAx7IW!zW7AGHz=^_xT{UbTpJDl
zvZ%fCZr)9UB=rG+_Sq3p4B=dB>r|h<$vW`yoTUWhrv!%EhK7i7T7yW8D&>31ri49L
zp?sEqaC|KBy~T|*GN8l^Qs28j<%Cq9Z{6ahN}RWW&N{m{92^rSPg9j;RotbS?1g9q
zQkb+gK|i*X>Kmc`N#3O7{)MPD9)>9oFRwceHa51RF)vs1#U?|Z4-YyO6-d)H<G8u`
zleM@;BJ@l*bl6#*2RF8O`MSHGSg4gQl4U05;|H4zN9*ONU;x_M+DTksTAOrNamk^{
z;T^rW9%bi)T6_?HVE`~1$BPoXrG=}O8>$2Hl~WV|I^bID^!$8@S5?oz!^5LN%xPAm
zuDbe><~j`Z#z*<r8usP0<_2k$@D_P|GV{GfB{tz@kTKu6WC?9gLld+XhgOaNS{I6K
zfEo{AASbU8m6V+Q<z`;cs*YH{jJ6xjT-e#%JY!&aUU5C1T_DZ31mlnQBT|vWUbn>l
zjDW97!+b1r5G^=qT|pU~%iBbv{~-JGx-kjPA+6TX^rWQPH#0SFuFv*o^atGNpPng#
zdSZ8dogL=qr!GIU`qh_$RXw2=U<m&CF@r)r-%pEkW%9MBoA3+r>{qoBzMp|Dd{5(`
zD<O*&;SV74z@qAVd}6UtjXpUA(73#RczW{t$h*P>dQ|&ckc2LHV1Tc~3jK+jqC4qK
zYjb_QZL!(WVUgFU621m)6!l?#p^ev7>Gv^Ik!3i>X@50%C_{2}d4*>Cv~zp=USnrV
zz@q+h=0Zd6+y{$4z0~&|`Ny`_4i1=#cwN!-v9{gX)>q$)KgG_5*+1EQbq^0okzvCM
z!6lZTUcE%S9*ew)&PZqQ`xp0eMu6CABtKL#fa&q+VQrSpp}uNL2Uv=@=ko7Xq+uxs
z3<i6S8YgSA)b@~M;8Z0iQ%V5jmzo1!?)@$&0GAMBv0<qdACY5iwBFbFuU~)W2V9?B
zOT^|q2n)stDiE?KOz0AGo~h}S5fXBZQ7fR0{aIYcj2G<6CntB0N=ZrS85tR&w50<%
z3L#^oqtzZZF}<!{xBe_9lE+d&<-~EGWBaUAVu^my^#Drw3eNmlHa)}py&A7tzW3Ip
z44(k&s}xS6<G=b6Z$QV8%beHFz|MH%i&&ZmwIyW+an*byaIVYqkdH36j1A}e?b=d5
zRtF1?eibA<`NA(S_t5b+yprs?TlnI#PHCGrab=U|+%7tc{4;0{jg};dw;46Z;oB;|
z8X|->9f7+yGN`rC&@Bv0Q#6L<GpkmXM8xlM)78G(y^Ghk;cb|sxZ*UwDtnv!>R*I)
zmIqoI$G>K@#gZ?8>ycuy;$aArB!T_TEvaK(jGL7kp6l==G1%1cG=&$jtD-=tTl0T<
zn#UFuaW&aPF@d$LLM(;Fy}4;wdEz8tVPOaup!){P{1{2<E_Z-nSN&;eZcYzI=BYni
z{8@SSVRK-V_E|YB$e?u6Aa7KqkoVi(f@>W|f8kGg@>obwY5BHuo8jogesR>x^WZ00
zOeXT&m!#RJ=o{KzW@O`po8>t=Vu3uo*aZ*v#oHMn+trlXy=SW&yDegW^G<qW5@TjY
zV!Da*0p`^yJW(GrbjTNzzp3euWyjpb+H@9P>Z#=2)e|ad!Q`xlOjk@eX1MiyQF68O
zU<0sjPt5YovfJCL_b;dS(M&{V|ChCOrikT<Y{&zGC?SstER9Sk6<Y)mWVn>{v_H_l
z0cQ4gzIx5-VbuD!n~Bf~H99NmVda2866=&}CABVkR#sL#z~CVlq@HtxhKECyRY5_a
z<j)^hCK8f3I|H7VW|>=N+(~_JLrIN*q0soTU?hh=p~!RHMh?}nPEda&f*m?80>D`l
z=RZ9xp4{a*FBBU(`cCN!(%;G1kL6o}+GGKo>hLx6m(cjaP*SNxGaWU1#sX0bM7~ro
z^Zdx(sSy~MTeG#b9Ue>*7sTY+BP%Zp=(>uJ(-6$BFG#$F>@Ww-9@-CKv0xtG{ib-@
z7@K~zzb%C&N-w`K6;xbr_MDYXYUs%A#6~>SX15{ykYAEJ(%tk&g+&+PqZ`GwR?O@6
z@peoY>=JrCwWOjs{>u8kx$4J{XZNjaLZE2@IslA)Z$H2>t9VxV|Lrm00q9g8g1wgc
zP|u@EEau^F`%heHKr^5t{ZUR;weB;p;O}rEs+!kbaN+zf+?KnzIBjC&K&3b&Omwm}
znEmf_A!2XLH!?`Tn%e;5p&v%fxiBLppvDcR&Bpi5&82~uON!jw+<&R5sReR6g)3x$
zqoX6c`pUrfjOOO%iRrq&S-Zuk_p^m}x7U9`3QtI@JXo@@R9iy>iN~@ri`A1e`fsUl
zr8>WS!UK<vD~-V9$d85JL^Owva=oRf3BhZ>jMg=7>U`Ck^`6;IR!%ZR&mT*t!7u9$
zAk|Uqb$OsAQGPpLac*H2Ym>>|F|0kw{4JY=#M#*IV+(wt!RvT;w467SXdBS%J&Tg4
zuZMa-sTO{E30OGYE@horcZ{$sLy3c#m%A7Ivd4FX-*@Ohk(6R6<m%g1{tgB?_;jAz
ztNjGDK4|YG<mO+na2{MuMw}@nTPq-7xhc46*=5P(Ky(ff*B-v8r30ZWeYfs-e6bE!
zk?=$+j&ioYd?`=5$*bZ3)ABD*1n5PGlDZr;=4>44m~mg&&cz&v+}U$WW4ISRby&)~
zeO2vat`0i?Nj(}WuSwnM5v0+zbApR5iCTf=b78LQysuypCAi@j#n~R%`}<;^Ul)r7
z-mi!ZP=t^w(!P^6b5GBuUO`t_4yNILw_oVn;;!5z9!&FV8=QO?V?o<GItr;169Zp2
zmn%ob2n29R=cDYFo!uWeKVDR9i%FZI7PN>2w**oO621oc{u-7ilfJd>lmv(Yihd8>
z*y9vf5mq0GFx~JIqLj^OmO+p0S(=f+ixpg2GBVxr%dp7<rE;w_=fy@<vswfS|0}L?
zcA4FT{~enSlacLG#KTsxk)5p|L?^&UJy@1b#k{9ueuFK$Xp(1I{xb5KeaKKLR0rea
zF`ntf?|@S7RGk|92Rl)*AF?WqT(k)PLRmt&%%$hhdL=FbkCwa0DiS(WiHfl3)S%co
z2o?Fy&=K=U1R=J92q%kfX;Z4<cugP)Edhcsq|c8hCQCeAJ)6(ku+!f%=NSEzz@X%P
z8f+UJZS(;~AFgtuKGQvQSXJauHVtmPDsTW~*Q{(P0w3Yaa?E{4)mKYN`zYAGmQ!>M
zTi6h$f>3xc4w-z2sb!8fH&)PN3S!~-Z2`I@kL-p0Y|}-;Oq&!=eDKM(U7~0R*OV<j
zk|ZTbm?VzlygX*HX{@`B(x-GhQ}X>(p2H~evCf!Rd<W{6TZ~%lTZLS4X6~W2_T9>!
z3Vye&j+qjl@60&WPghhZ@W1EOp05I!cs0x{sq*N@vl5GM+NTKHC#a~{A8j&5^R_??
zPzKpMo?Q;R>k$Lf6}yXz>-|18A#CkJg!&Gk#-Rzf0nakMID%CXLXg-lbovM!<$tfD
zU;gh6ZNcpSr4iab1V#UsiKeuQx%WRDgzd&**sA}J*2Tl8^8W=H+9)9YBb+9@`oaHM
a<PDSNm$7e|PI{66c&I9AJ+Dx-4*nnRE!XY<

literal 0
HcmV?d00001

diff --git a/docs/assets/images/use_philosophy.png b/docs/assets/images/use_philosophy.png
new file mode 100644
index 0000000000000000000000000000000000000000..23922d4e9476dcd0d3f457828807271ff58b1b01
GIT binary patch
literal 16463
zcmeHvXH=8V_U=m&P(VbQf{02JsY>q%0s;ckOXwX!@4bkEN>M;Sq<1MvAdt`rih_VN
zsR<<@ozQ#d#-B&e`G33j%UyS^d-yP{@aCP_v-h6;?3vjQubyfsQ(j`c1ONb~%Hv1c
z06-WH00dXbNbrAgiLu$k|04F1SJ5TI{{)lSyu^Pd@X}V62cSJSmH^-upz`RUuD{vp
zgrB#`Zsz_btd!jd&QrJlnyHo{mZ^v>x8|ujXGhd?8yYgSbb8W+>|RGl)&+Ta&s*vC
zCAq25R)HSl4Ib-F21anba%ey^mb)!hi8W{~)-G5%5bNcc@}kZ!bo;2g>F7I*<02^`
z*{yKkS0h(;^|6y)b_!L@KOcWJw}|DMiT^(5rnv<m;g=R>?)-iMAjCf=`tRfaQ?=Z`
z7~UF1sfMlfu%>JLqdtHzf&9xo9bq-#hWIr6p|(NLE6qfXn1i-gn)RxQvx;XP;xq)r
zSLA?Sjhvuff?a9r9To1xFEvjSb?jh9B4BX!1FZ`!O9RElH~EuU93pP^JyU;)6v%k<
z*PQMg1XwtmN}vosC@WpiUPHY=pPIC#Kfl<wR9*ehK)nd{dZ$~Q>hCbO2qV}I6;#z%
z!(hh3=Z3`g;Cw6MXv7I!GKcg$4Yrj!qlfF?@@rjkP@7`ehKHPARZ+8eMBx7AKrQs1
z40b>(C4GSUx8Qi`^Qa%I20E;6p?hnap5p*9=aU_1i42xB_RveM8~5jX5%csLyk#n?
z^`SFc;u2bHC#^ns#kd~53AuAla|YT_{i6yX!tkQQ`ylu3*WzJbULhm+VQZJ%7MaH9
z=hgjr|Lxr0i6>1Ota?^mp=<YmcxFivhy<5?T=S==qHdqTe?}!Bw&JKzA(|2pHSnVY
zT!fbg%X^YtC&NendOX<|eL13>vL_46U-i9L35Rco-}tu%%&2bxfrT7*|E{SkoCJgx
z894z=f23{U#;+OJAo1K^b^K(L17c}}+#>&2$zL%XSWNzq^S58Q3}Uiyk7x5P|1v&6
z03*8sG-=bkBl=^hmq`GDfJR><<)8UZ(Gd{l7&txi{MXtB2tBUe0@^NqsCoF01L2#-
zJ95B|s`}H*e`LF*K}g&p_HbL|&xFDa@B^EbT<nbaBik31aNruJ!z9U{fj3?Q1by7X
z55@n;79n==3NS<5J@$9t|1G6JM1UgpR}?DLOKCro_KtdeGd5nWz|<4F5qE};dKb1H
z4|q~8>9b33l@^BRn}!@}uy;k|3891UnwBCm7Iq%*s`$KdD(QwL*CTi5)cDj=-hHN3
zk$lUc?G@kUr{Ur6^?Q<yynV}rQyaPo*NyJ9CgKY2u<K!pidB?KIUY*ib8;#yFExJP
z1h%m@g*W=D+L)>bbG!TZs@BbXzIq)4k-rDlVdG|1c`$%$)eT72)X+O9P*;UGOT4=c
z<L^%JIUf6h2p<Y5EeQ6k({&Wr0qd}f3o0i5z?}x`F_LtF3pomvejrTJ$OYM5Gy_e#
z+S70Uj-xH;3&!fecg?*HSXGjZg;S#U_l_U>2p$!U%M+@41`7kjNFD8`o~p=^gV{~(
zY}(<kNJVax3q&f=Hi2T%ggw(!hn-(kF)``G-RxdYMNaVCXa^(OL&!AiKuGga*Rq0P
zd3=3_F;^2**QtK^v9VcH!JOS=ZBNf{HQPf&D!Uz*O883O4z`vgN(+J#)vQ3xh9R!d
z-zXrbM(aRF&}V!an!cJuK^YRV?P)SJvzB+Yq)*_W;YY!H1c`AO)(y(__Is?*(ug`a
z?&qk_N2N~g(z$&WpI9Ss4&sOo>#hAmP>{Vz8N^UXQzG${LvL>G$J&I6F_CG<rqyrL
zSaJ6KB+7Y#!rG3gES&5mCZ|y+%$96j5;yjj4+VV~U%B1)85=g~{-oa9nbid$DV&)9
zq`~wJt!`&nBen#sfPXcqCiCrgja4>bZur{#2c2Z?&~ightg|LrY{4^e(tlhw!j9gd
zpP;Wk<Q$_>01Ac9DqTA}ndB5EzIjiIOO`9e4tE*piY;E3;hB&uR6Abz_Bk6B%&7{V
zo2tAQus8?p9%go)XjYWvdH0=9JTZ~R=wLe)hJ2(jm2kBqNF;l0%C7lU-HqR?(Dj0`
zH*msF8OAzhtNO;oZ=zZK_DwOPJK)z{b66o_k@n5wbaERTti%$!!?eatXT9m%Fs>Y(
z$%6_gw)Udce7!Ekr9!05))dp=iY=OKTz;H(;5(?=VBnQr7;HlPuO)bg#-S2Fq0sYu
z(?p8pVXg;h96gx8#rG%nmHf+43*6D^s!zM^<Lqzsh3Q96)Ax11t=sv&tmhuB4^?yJ
zOF*7(rw_rL?Q>@vlN1cxt4clZ|F*~Y#p+SuK+<I=Jp-%1Rh3@6+u3VuXv{%Q-Zfjp
zzLpbdej4U_?`=mUu0Y{<-tlzX{QTe*qfsrp(b?(5qA`Kmni!X7XlcR1(g*Qwe_gsI
zj0BSt+k%fVL3%>6+HcRFQq55$kps;hW{yiNO+eI=)S7CwFZoyUQdjNwQP&;Xcz%gd
z9J|x&b~+orvzmsU)#j&8)l7YfZbF&3o`T<eV~816%p$)j&Vg={5ULwgE10Wdw^H3_
zPL0=e>f(t}RPOv1@H?74V(+^{lrwNTCSye6^9o8=&V`03Uz(NG98?^qbe=$+hlb=*
ztT9h~d0o_5W9wFUl1;V6x<-Or!@@zT>{aGwy5`U#4L+<Mn+vw5h)Im2-h}qIM^%wm
ze*^q|n6gjUWEBFJR<MRQj=e=T242u@IojI9D3}Z9>E3KC*FEx%>*l*2blIRo(A>7m
z9tUSVI-s-mE`C#1TA(hPsYnx<zxA}TAS{FLdf5YOcOSl2|BAZyG!6*_8}Ss`QloW{
z>fTp8KqlqtBb5><uj(h0<u7@F$xzEmDgS`FS;T;}m$f@Bok5qNxm_oy>khelanU=K
zZoT?B6W8td?47W|&LrE!BacKI-q`M4Yq?JL-^k~ng%AB?rAfdw0^=4g;+Dk|F{a<M
zbGTq^0~C<I=bLhHqv%oo2Rw4ve@*!t+ZB$8>e=*^-l%^Bm8&M?eE`5ID7so61zEex
z{9XjSM!!s4CP4oz-u2=Q`e&{kHPYFXzxvKWkmEh<o4$KL<UYQIoc6aY{L{Zf2O+Nr
zU|e{b3VrqfZd$bSZ75{=!?c(?{LwINY8Q%5wDs%GRTQdDPH7L4l$H6xUb6PI-N~kV
z&E~LMvLA7FC~LZz$;s|0_qxE(F>B7aH2&QyH)Tk5kgy0&P(%~rPWRizjhp3SE{7j>
z$Q+hMwYb*b@^-aT(nSQo8|5!A5%$UUZx|RdppU{nUHIk(;?~557}oo#rtYD+w362!
zIYMqo>9bQcjq6?N%4<kL-SQDfXz#SjcdR6`a*6jN?hdEusviCm70eUWJ`@LQMR+{;
ztrCb~jv}dCnE~$7nW=B4caIyt5p`6H4&zxoeb!rS3WsvzDM$x<e~f-j)=iAVR05fR
zOiQ?s58pBt*M7=%z8PBlvL*~%8azxFyoIDmM3-b1*eTKzf;AiuB=vdf+>rLiHTowN
z)A_8w;Kg4#=PRjP7}<#nQu6^EUf`D1ur@WJ$gb_N{O$xQRDy2JJ_lRe-Vhz3eQc-n
zVFx!4mGmFUW#oc=Ou>c?R^p$dg6Rok+$6Y`($W$qTTT7khaw)7+~ZI6U7x>t*64-R
zx%x|=5&C?6_Xu-_Dt$=LB_vfyia1o-Gk6?fmIC79-UhCwnVK37i`;j`*~h@aTmB_5
zbbUzrQL~ZXM!o@kM|h9k4=z!?El(Xb@sd|thr!z3aQ*nq@&sn(fw@<j+aDlfH6>d{
ze(p;?h1vrMr5{@l*my-15)+dYn#T<cIEOg73+g}0Yf}}g?tOQz4*?xUM!oqb@txw0
z(5C@3YmkmaW3fC5mK{FY35oYy1_szz5zRP6UXgxhX(@X?Qb}Z(y$f8-W(NsX51}j<
z-iRkxvvH`oR8p><P#(+H@dQToY)(%`TrW3hr)aRVNa6a{(T|Mxh$CS=889VJwh5n}
zanB$|S7q20SFjT(Ypw?wboBF9SB)!JcVoqazT@up@TrF5??lO+*w_@<8EN39mwx1>
z)=QgReO#0a8RdkdL{vmZUhCpyv%6T;yMzls$vfZ5Qhd{VG_+oKCC$lceQ=g}$2Vl$
z;}uwNR8X&s7@P+$c9rs*Z<W}na6L(=CUU$ZyigmwKpQRkd$qQV5#qC}zy{;H%zv`O
zE0XxMbj7*uUk*ujYf1&5E^?N=kpEXk0uWZx;M2vP6z+d~=vOY6MT^gE;l_7q|J%ub
zb@E?7`M<PwMDn;fiwnNJ$GN_Hm{j~YU{fXS_~cx5`s|e2w5%jaVsD9CXFWjff!COD
zV8`$aa+|sD`vp+s@NUrF3R>Z2z%2C=22s|y)yciJWS<=agZAwDh<ySPs;Jy?Tq@Bv
z)nk>(uIU{9pq7LwiW?jneB=!a?C8!4=qM@C)-ejuh)GUNnQshp@!1vUZLbU64m~~Y
zXhj7ZZf$MNq8GA~1H2ZNLVDxEpg4B=bxk1lg%j<_lR@D9hYxpZCOykc{7!w~UJHBy
zTOG2GAAZ`bU^}|Je0QYh8h6PO*Vf?}*k-1b=tr|c%D6w)e`%CmIi+{(_O{GKUf3d8
zl4XR*Kn%~p?h5uHhX%`C+rs3np<`TODV$@aOb2GEQx|C2z-MYRy3<u)TC4#-h#M&=
zn0jBf-Lz#zZW5vtpYl+rsACdQW;>7(mXX8~dtDs;z~P7PQotv7qtgy$Pxk>O*p<vA
z&(W%ry)m~`FtkqJ*BQ$fb~3=y8k&-wM7Ui<O@0MnEsE)WwMiEKiM@9nt-}{OrJRW7
zGq~eIAm!E<&;x7RTlJcoGcZf(l{vn%5-tAlwg|S+0fvz|ProuLtc)y|zMI|DlmcAi
zb$Yh|D#LyOt-qaQRaV(dV12Ov!k$J1@MV(H<-i8i6xiBz*7wRLQ=T8(Td`Mq7h9~U
zsj88ZJHVfuU|_~OL`Mfd*fp4KX`}4+Iegv}kD{y4DZaclpYGQ2J&i@hH1pUj_xS|}
z_6wXqV0j+ReW0jDCG~i%&VcW{XAU0Jdoy6`mFzdm(;3>ehX94ay=E9I&^98O;l~O@
z<<p*l$-;9XKM+29@z(`sOUt0|P9=z!Hrsd8LJyuER5U)gjXw;ySyuFjBAkeX;;s48
zDC*=gv&}py%Qui4id|TE`rMce2wYJ+Q1>7UpDbe_wAdcLNKoO}JI!En3w1EZ&!zx!
zFc02TOxqpeX(JR_A6d^Zj|2RB%5p3TuJQsHq8$tvnsD83oX@tnbFQQ~D-i)6sn=!)
z>VLi-i@6x?k$?F(jx5|#=f$o(p~b+HfWv^TZ>>~aQ=cMjzrKdeEfr^{z~S%;lK?F%
zvg!li+O?0)Ih&J1W!_P%x*fQFwM$fMsgUjY$Df*|du~ll<}0ZMrc<D6EHj-mc9Y-H
zM`eBK6M;2l`1R8+J=l<{BtVxexfE3A$l2r8l9{_wIulmMdJfdMRFp^jt_wZxNf$Gc
z*jeLE6JIUZIq(pNTQ@j^s^v&b{Iv|L!d>Xs!!I(Z81O2)ra}`Zkr1!UJEtbcLeE!6
zdZ~M2kq&JDvb!fG`+Vj0bwRsWXKb^=S<VEiJ>4C=obpl#_%QQDjkI%i{c<8ryJp96
znECXRo7V#@S(b(c4KOxE9h>>Ka4=5GsLmqwT}HXX7IY^%^Kzz_O@}P9GtX#@U65P7
zO=F2uLD?rYW1Oc13Qfo-)#!{$Cbbz@CX(PrdT`roxJJd+e}wJa?0YD&JX$wcO<U|7
zy`TNor)#gHIrBoC-4EZTeht_9uxAB}U{s)6ruW*gMpWF=QgAd9D$T|{zy9t-1O4Fa
z=h=@Y&1PC=?&P_nbtF`B+`{UKSv%5mrfG%^<%Q%qb4o7<RhlblF0-DJ`uU!jel2kl
zZ(M&~Zd!yL=vHP@FI~!^amA0*_+C*=8J{Q4hr(5whvCmZuw8R(irSvs3{@}KZ&dew
zzi|pG#MW+Vdp63=OP`l#NO{&i+~j?3;(q8i_iQ~_v$U(rcDjteE0;GtW2C^|w&R`T
zdNTA#S!~?8w-jc(B#F7iB2b20IAr*y=r_)rIGe;S;W0bdLG!?n)7*P`#2k)UP%+ZV
zrCph12ho2|ojKl9{dj3r$|C1U)3FS%4hSuF%x$#))jrPdRt;{@sA%Gx$Z@1DX2rD1
zIHhs-XvuT5_Uo&f?!E=vKtq<otwAoMolZ4R2V6>p8B0&h%KE35>!u|$XM9NoHkos@
zZ5}`S9;WU8^y;Y^im#5Pu)kpkqh`%aW}HXuJfmGfGjFNx#WG6mT3z?qML6AqFVCdG
zeIA~c8>95%Ke)W#EnG2a)+JfOCPNjX9z&1xmKOL&Yczd%&GUM<cI`7)Ge?@z&sO_T
zTp>q~p=N?y>_)HOFeiO(3bNTB;htUOYtU@X?gme^_)P{Bn|RF8dFCxX*H{VCMmkS0
zMqT4qp^ea|16HHpb0mZ;a&+v4p4#mq-Z#abuGsCekczk@vb#0kOD-(L4i}#7-&Bb1
zsAY0XK}smf3~H5iJ#g-ST4(0A__T|0l*pm4fJ<WWfKf#SX6}1y<F1UaE~yX8$o_TA
z&I1ACcqma%(#4u_4m4&JGiaG@dD!6AQIeM@7|N1=F<iv(U<4JYE8)9**K7W}PYUJ7
z(sIw;(>FqQiAh>?3iaT&-M6oMSn_jvUtqZqtu~uKqTf;3`t^1f{U`?E+3!ke5Irj?
zdO7tnV+P}pRzFw3{tG%vsI7TsXrx4!pKHAwX70u1rwCymxbxa&BZ;7($c;s+a2GGk
zV<VM9o0=|%jg%liS_j~Je;9Ut2UUZv72&eR)khYe>mL*4To6yMPRLB>l0FhT%fdJ>
zD^?*p;Np^BI%j3c!f!V-YE5pNk%xCkv<`$8?A;Lgelhfu@?6OSBj~ipx`fMDs!Qa=
zEpCj&JCgS;#)+@{4RrTllMvY<<q={#6!RjCQ~V!1QoUIod_N?pnEJX<_?W1WIl?jk
zixKXDn=l{mfOR>MY$HmUx2$jXpEVP+?1|b=eTms|>qIbWphA?0n|q*@K4(%!$8+|U
z{596SBaQ<_oo_;SSXKH34wt!M9qrknjQg)7pl9zltPBVt_ka(vh#uxXa~;zjrrGg#
zd)$4l@h)ogDhX#>tAchJ-b_M<_YfsPpSbxu6I{kLBIoy)A-IK!;%R=YcstZf7qseB
zTFO=f5?zwCE4Q&(Sj3nh;70>gwlS0LpKw#??=>n%at`?rpS_z62GF-Oq*SV<XQeAL
zHXpr*_K~fw%|*imD}4=bp41(i72ohZjZ7DH#7&Eo{J`1zt?)?iOlOTWMSEp3AC+z4
z_Oa44>MZI&#sgdU$w51n>W4GS00sM)5CN|u9O<KxtX13P<Vbd2UOoB{grct-ulwho
zuLe{DQv>?LV71}puR)Bbfs_>Cd%b%&WT`ew3im!(ZA8VifzRlOa*5}3H*62<=TtVr
z=>|#gH2uMLKt>VXfV`26Wc*yt-eX)B&wMXsMp>1OmD{wuTo`lp+6{pB%Y=Kc80U#m
z3kEq|u`;2x*od@daOr0nQ^45OBw&nPgpd96%S~#JcNR8D<;X*aRcq{)Zo0UXC--9#
z9di5PXOz`!(JRb9_+)k3o-A{xxePF;@?$||>sl;735jx&C5=#Hr2DiT^psWfZ+X9G
z$s&u9c6MS_0kQ+^emsjJLwB#ZFt``&E-zoCpJ7lA>Q!a}B4y1AOk2+A<;cl2xZj?w
zx3ukg8iGbvj-7hn$>g_-$;R`3<Ox3fs)T!^^y}MdcX|nX`q}-9CiaPn7rvg6@c19H
zm)zxb+;LeM3=4nKgfr`K7}4;B7(!*gVSUfci1Ei6G(k#6YCmNtvB`nFU`0fVH7N5F
zBTc!UU+;1<Wn^8}K-h{uXRPrA`pJ9~Zur|~6XtbU7!wAEEi`LY#@~1$i;VTdP7<Iy
z0<tUi)sVG~MOM^R+~xQk($H8pEk2ePa%Gpo=Qv|*&#>8kuJ9If^N234QKB$R2-Rby
zu^UE*(?M{be))x{D$>LEU|8yQ^=^;raQC)d7e-UoaU<qC;hbVk?^_WJZ=gJgZUME^
z_|Ac*@w{T~29B^Y+QUb&kSAzC9bIKNw8VrtJAR(-mORO-BwqadQuy8na<^tk^YKI3
zUikTb3KWFMeJBQZcbw4JV+^{>8$7qJvEj0e2W<|{;&1Ar6SqXM!Bt2Hy)MD9(=r~j
zMzPnh{05y!B8#}FBG@E|ABURslNc<hQAzHCqU89ljp#&bLxMM<zUyE)ip$awmhHs>
zr5Y?*YKRxkFQ5+%jW$<dt~mX*;fAIglR@(LI&loK>gwt#OCuBmYs1YN(mdG>I`qMn
zg%5GY*iV_W%VSDv`o8ss-~t~52D-sF#TO~U%c^%hdX$+ma~9Vpk#N3);Ftlc)Qm(n
z26N%Tr4KR%w*KeqU4_-PGJPZ@a>wV8xsCi012v<~8dGc&sqb5`U989FkK0&t)4B3Z
zhUJ$cKz*6|dQ-jlS^$is1tKwMv#TtqNkv`hy7)<_SX`QBIImi-;h90$mSyB^Hhhdj
z<#S=SL{-mk8oFsD54fBs-kAJRQ6^CV#zy_*3BkpY^d7gM9{SpnCk~&w^@f~V#!(MH
zbD$*e2^+Bw2<WJ3x>hGa?AR&2FZpm#9-Am?4i@^76H>GGKqYm|5sOHtbz+S^rQ{jl
zn{bcs$&~EMM}_JhIn``&tZD_~Hy@oHN!S$r4Dbfy`laepz_5VVW!t=zrOPrVOr!2N
zbg3NGx6<|jhVq9rh)SX&pA6mS@vSO9-|{j#*E<dL_03mrSdX?soj6mxiMx1_!cYHe
zM3}(J$nXeDBj;GE!>uC@mr<fY<Fyq#Bu5w%i#2ML@C}rRv}3km%|qViBVc&0cE(qk
z0x(5dn-b;vt}c$&WIq`oRY=dsAjxJOG`Rl*cn?L1tv8&@)SbY)3Tn6S?9&4P%j&OM
zfLg<>lNlD<nWzKaJjNS<pR2?@7F6Z3x4?FoK8~bmZOucYED!NL<Y<1GB)7+G<QCu*
z0x_*OINO|#8i0W9>tFEtqp!(;Av}F4pZCp=M8*$$$thk5V^$^981&3)f??Z38>B5f
zl(t^7!}op2eI<%1nJ09b{kF*#j<*jb!I>V>zOk9(nta18j@ATNa(`G0;J_Q4y|^$A
z%@!3{G2TC54_x~gb0h`{ALebg)>&i^G7bWq<AWw+2r3@6bMqe@dr9cmPQz~+)vXFY
z884oA8{)Gvs+9)otxY265~l{3WG>X>+;dcpcr?1cz3bk#@A{xA<GR6vtei;N4R8yy
zzO=ud6o9#PMeGKr0ANg>eJVX#7REs6W_WxS+jPcJ95cd5nbNrVdIf~);d*&tcborM
z<8)N(R1!WLXOLO>Ebg)kyVQDKM}6xcrN;W;&k^&>r}bB;>TaAKTjCnb1Wtm*-KWC&
zNVlKg7)0gY;8n=cyC8S`!sb)=Grf8Zm+XSbyaupe3Gay`c{qk&9PC<Cbb4ael%X$k
zu0Ek=@G9HCJu|T==}|ez=WE0I^9&1Wu|711cX`HPXb5*MCCRXPOg^mX;kgkh5f0{d
zZZXvYOOzHWnhZi2hLc|o_icUCxa@TO!Rpl#9xjj24(72Y$yw3+YVCK;g3qZkp%&=}
z@Yz+-4=W_`LV%qkcvds)h$c<I@)<2Xedo*@>D!XNYcIz#sjmSzP=^SZ+h``DM<i_q
z)6k}x9l@Ll9o#oF39aAlHXm)Nu%0LsG&j3_>`NE5YeYh^ng?+biMTtIwzxS&ipv~s
zo3ifPaP^#nux5@JsbQVLO9KM<lqk(*>>E!eF?=uGLTuxN&vSP*KeNlQVxoEaWk{UO
zmxD>qw0KP(Za%j@I_7Ht?@P==^~Jjp(V4O~HdJ*#Q>YX3cvHzy;~Z+<QWhUSO+%ys
zrI23kMh2oB^(Ze6dxsn{sg=VA%cWr3jp~?PBdwvkcdr0Ull>6{u*H_pw6ZV|AyJm#
zxH4$+hwv>7A17G$yx^f61x2_<{%3N`$19CxF_RS`%5QuELlS*nKC#GuE_~I{IzpJy
z5SDT(U165{G<u*S2aGT_8eV+=Rb9mHe(c5D5xxHD<MzD=nUc7$98b39Um$ZK$ae{b
zVaVX{PVcbgbsWw^hMgMQtFa%`%@J1|3xzLPm@u{0Eay{YthGeYaCV5DvfBxk^;A1G
zod#z4!hCPz-cubH+MzmE>U&!C8VpFxrA*tyb(dRrg%?61sle!lr<{H}*M*C;fa*Rc
z+qL^E!3J2VbS(!{a*>N1gVc_Iw;gbPV*kKy$HBfVr~PKo6JJs4$EI{*-x)|r<jA$N
z8|&KAPWry-f|{DW@_mT*J=;Dv!KT;bp4O66NQzjG7;p9nfk)48Mnrr+Lu2)rmZzOe
z+eMef9x|ENx|~`TzvwZZ)*zK*wmv8%Y1!poZm1xo7_bOvcUZ`l4Rk3KjOw3km5Ek=
zDz2x`zAl4$62E*+Q4*hh#O0rCAYhsIUwl%Ag&u0eeDqPy=Bhw010VbPC%5O<5WxG=
z7$oUgh0DRrtqydG%Z~8KP_N<XMhH4j)`}=dATRC0jT7l7v-iXIjLKvDM%hz4Uvfiz
z8Hvh|hp^8Wt`%{sb!G7VoU#;)AlTSILV9Y|(hqM0k2=xlc;Cpq(A?5S3X1p~I=kS#
zLOV~RQ>NWc;`TvMt#z^3K2y(`p}P;2L?v@zohIa%m?7rz*=Q~zt;NzS=<EeV5`Rlp
zTy>uB)6N%)6JB8OumE=4V(d&?=*N|CFzbd0eW$zQc0`T}C(qy|2Rewd(fUM}Cod-s
ze09d&Qx|ic=c)9ZyNNbT?0izTND6Z&&+!3iHZ2g?FOi70r`_c9&3s9e6EA#{#z!FN
zyJBd+l!~{TiSLq3%ccZgbLwXs(FICH60WXCLknH+xk57KcBXOMw)-hj$HgR?gojTX
zl(<pXXZ(tKts}l;*}>+~Xu*-LWp)wE!c8M2&v*5Dln0cr%8Eu^CeXB@3?Zd{q2)|L
z5_hx1I%7S<XJ~EEi;&3Tkc%>$pC^fxl#x^}eSEV-!~Y_}&uw9?ZtIk^O|mLZ6ehmt
z%^w|T3ceiPv&H;tcS0F1;{C<$9f{`CD*jaUyVt_Dd5<E6^@bO(*^`r>jr0?a9%LbH
z8;h2oGNo%Z>DU|ahrycdzXe(pO_WKi)OXM0H`})TV1CJ@Gr~75z_n2_>JvWZz`Y-Z
zKHg;-nct>)AltqnI7plU$}pPS2&PvX9d;7v?&qVkVDFUQsu{YK^CUw|>-D9W*;les
zx5tn!RQs#`ne7+I!ly~(R_QC(sE=(|H~_vrJmk1h66KuT88ou)CvKTSl;j;|DKY}9
zBh#{{1G2BWiDej2mnVqj(zTpvI6vdJjv5<DqRXI#9$&*-O!?MZMds4FC`J`1VUDa#
zRt(^PJ4HKN5LzrU`Vv~a+Yh!PY4KnL+}MGv)wc6_RPl28@~fJ`^=@bD&n_<$d=`$P
z3DJ-0S`g=VE{9W{^j!?Zo8T!}HtLYN=Kzj8Vm)7w9WcwPw4Qw*DubItS1H?H^QtDp
z-?UsK=vyLecF&7#nm*Pra2+*mA>fVa9=fo?QsWS#+dFcf^-Q=h?@&aE5zz5jO1&<)
z_O1-ptg7Mo?X=7M><+iVa{0U%GqoQ9?mL>Ofe9LSZ4@&n0+<PyHQ2aaTjSxPb!DjV
z(HoNpwO=NTFS_uSZ|W}U`S67sbJ8rXK(&4A#=7b4)*^Fgt0E&GG4lymo+(M&5@A-E
zb-JvJnB}L*_O;@QhVX?&`MzvXGcpMS&R(>9pE$FGh%uSOK^0pdL&bK$38V;%`g(oL
zMa^~9>=3cHH@?WT$kMoHu0tS@$Y+s$DA~P4xVFfX;u2#Qy(}L;uTe2)?tJ|k=&~Iw
zJ~!-SvwiMcLVmwVqL(zj;)#M2SpE<uvewgvMT<4Mb3_;#tkIipPNW#S+bxL<uw6O7
zwQ6jt*4f6pHXxjxC+(c~Qu?v82X0l3L6WM{-S<kR>^1E1n%`Vj(7H#xl)3+OK$<Xj
z<EO9vDd=<3vT1ED%O`}D9n`n&mhf7I+&~UQsS{SRTYXL=PE5(lYp`EQ49Ir%ObL3W
zNNzEjB-cDzDw*v}C7$uAz7*1hsfX4q#<L+^EyN&s-C8xdP|FD|Q*Wc!))FzoH$mEp
zA-%Y02Y~_w(#77OCG?RnhVU{7<<dy<G8A*_)N~<2G9p75Xg+YxA0m@SUw{%_TVi{w
z5ppb{@YFfDiuVn|X+9{9o@323E6$*y*Jwe=pY!cgg1$7i+}*$;mZ8ccnZxYc3jB3X
zl*~Mmb#YHj(kUF02q$jPw-haZc?c5!T<<Bm*h#^#v~dU2&hNj9Dk8KfNRo4Q&WWS0
z?qJ>aPH-F>%n^T+g?Vv`3p+%3<1l)|P$TZJz-3MCQfo$W<0yQ+*muAI%%wiGOR$;>
zscOrfbp9lWWaJ~#+3tQI5Tikrx7Yi{DIdfojG&0!28JNnA52K*pHcU;ex>)GX!0zq
z8c^QQe&CAw&|^ruYE^O0B<8`Mfs4H%tR1@9i+YQ6shXX6Ba<@X?Wq>WCw%1%E8Ny1
ze(`~TL^HpIi!3)-h@cPsP$4qWMZq3?(zLHVgRY6_Y8ib_2O?>c9TTM|5E!?$kQw%P
zIVKuqM$%UQ(yi97kuNlpyP<|TiOSpv?eHqof3V6Y-!!npyUx0eqNCrVVwi4sXQ206
zTU@27=ST6i!y`yLG%;7g=+@wO<LxT@*I`Q~Rd@)E5p^^iyzs2cJ32d{_uS2}35XyN
z$gs62m-K>0sexFB!ieMDCf#Vpy?;bPGNxo#LU*FM(r<YnfyF1T?#PrbCr>NYa+G-l
zBumEh%9y@d#hR{YH065bE5U^&`h2>t>~OzR#^T-4wl%(hffQdOW>2v6NLP!lzR-Iq
z2Q2TH{6_F;cHBTUvx-P}Ce7ZR-s#{rt$d}W?bC(C!U&nk-hKDZc5oj45;p2OkO@r%
zx%%yq`;z={679tFelR1$3^P{0$cec;44bfvhMe2u0F1@5Q&U=M)3$G9@`TG+&c&mf
zO;I?ETcA+gmFiW6k~!p@?J#awCVl?<W%2wjc^lYvv|7a41xIU=bcq$8;xHAAuVTHv
z)dr)bdL|gua(dLSTxgC8g}a^x1}tl8ih~K;X2=j%s=M<hHg;>srAPRQLz)G>hJB^p
zLmJ%6J$<Jg3_m*%<{%c)fguXYyY9_QNZRht{OM+7M~`yXK;^eLM{|!diL9uB)%u_s
zmXTK++?lI$C(2XueahbE)?i9r@y%71O}S(Wkc@qsVI}0qx!(z--<Hz<YJWOb0gj%p
z$5x-emkvTo?VoHzYnZ)t_U3CBAah|Y&tKJt%&JRV7K3Q5e_E3LS)Y~uqNwVEoITe^
zZXd3XG5j9-G=^JkILlo^q0x(vB44eNP}lZ(;YX%-kVlxHiN!&jeH^)H!foX73rOy=
zEYB^C`fVBeZrz1rhH3`)ijb_+kHc$L=>ql;SEQA^m!D;n4i@?I^NX6vwd!Ha>@|=R
z`yi&&zCHydIrh0c%;3#Y6d?#xcx$%g6Zyk2o@vE;CG#n5z6kP%(Z)_~c7&j^^0YBz
zQP*UE!XAj+?xm3fGRlR)^s*glP3WrANaB<{#_#754w^LP)-Z%7m7XL&jn?sN<dDGQ
z$$Nsn_YKQ?9>|SipNLT+E7;o4%CN|_&sreXRRq(J-2q17M`)Zl)m76+^`9aionJF_
zuX<~5*}FF}US!IpSpV2THIQlIwPU$Ai`486$DF#$g#{^EZ%T!9I%lf%$aE12QG{u4
z{1ekYgxC35J!1T@{<%7Y=hCQ&%;3mZS8vc$?*l)W`&pGKc)v9HNZcuN;c;@j8k}xc
z!H0Y-YT9qYr#N{1HFv>d5ZhS}Q<U2nW(UgJA7aXs#e4%;MPa_1roRANpFHp}q62M=
zrkGc@ruZFB%_VD>2OW~I^b^Z9=&utn`*fse^<3qr7ZQcYEb>%1k5&7uP4=MKzaV|8
zcro<2`XIIQ%I>8z1gkzlAJheW9STB@q@7qzyM1{+*(32h7g<0qqSB8Z(1G2!GgGwZ
z*A|UlsJtJK6gIo2_EH$6H)d=jI}our(-tQk(!Dz7C#I%7y0UF}QwzIoEM{Mx;!BNT
zu>igFTQ?4J9poPc_ePr4Z)@ruaxIYlL=)y*P(81AN2b=kh($<b^&%ssbR6&pEql#-
zu^0$0^5*DM)W?b!w}RrN$$1|01g@^fag<y@ilO&E1<E0L&F&U5y2E#?(9mncyXo)V
ztvIXsA_0e9SchOWxlWMuO8bg$QwqW&Q^I4usVlvqdD|ytRoAq>E2?GG4#3%m+?Gt`
z_Zq6YdN4D&ImGlu|8>Q#nZha-47+$ad~ZE|S)x=VeYEYFvBJaiiwqLrUJ*poN7>pp
zKSIJ?vhBz0Tx*aP82Tc_ke)_}F@ul6X~T~?Z1@0quqd4>k&8<jEL(a<$38iFB4}^%
zbdPxE{KIjVUp!11@WwCS+!4POvFpW251pAmJ&m3?Xh50WHpd;je~Mc}D|o-;<F~Zz
zGA4dTl|Jqv526gBV1lB4ytak<Zqlw{u~W9bltBWq%}4bdUDL9kLe=@;!XJC!a#gE@
z0-N#Id-C!=DR1_(_S`8%O3Hn-u7(mFY-fW-AKFDVp0zPO*lvCE0l}1462v9|=o<`m
zX02Z6@s|Qki5*nzo;y>F^>i!CN2NLt<#dzWl&Ft3fhIk~*c6ByPE-tcgay}De!41H
zNfu7(O=ZEYth}0a*=H1b2kDX(CfQIF%iCxM0aHiS$G<?=WiR%yZ@X-+j!JiwXF+z?
zm1U1|XBQSE-Kx#kU$uK{VOqPT=D7}Zghr%gSAKNa)~=0OguRWV0u>L?d)Kg7iXiu&
zSOcpj#|e&tKD-q#B;>ovcJWy`v6bvK$>mAZ@D@w3VV~jo&;s0Jul#MqkI&_+u9h<s
z&!~bX(FdYqoj+2P#X&%|z?l84gFOrA_~DIULvW6|bH&BWl~Uz%z4hAB8mPgvS%%=}
zRX?8%;rf;&QZ_ovGFa5?Di;w3LhnVTpWGGzFpLOf{>#HC3FrG^AS(`kuIoXNQ(pM}
z*x?H<>mfz0&Gs>7A@|E6QzTT3Y;UBtUd|%j_*rl_Kf_SauHG+>{+c|vbZ{8A+%L;B
zi!^;I$639dbQT%HvJmqgFpnX08uaPK(#x|QufCAxSxH5?Ok2;yu2h)M?p2Hm;J6p2
z#gCS82t}BX4X|U7jlbtL_nTgXP-!tK&&rnbwENf|FduObbG#Bl0NxvM7GQ(8!A;e@
zx|Q5T*&cep%O|AN{A8OybM@J9xfyHNFuH}kUPq-Iztu1_GMiaRNr2|se=i9Ak?Vyt
zWGGtwW(odjRI^gqs_v(bKl(-;I*HM6DB<7&3QFFx%)}6xJ+NBF>RF{i`xsi^Jwlm>
zaLUdl!lZyV4!NU_4>}nL(j=dhEssoMYnrDb<rtP>5kGwmCUXC2b|TyNBqr&LKV^T?
z0YngvC)fC?q<7xBs`&`;@rHN%^p1JnRwy#ziS)^}B{xIr*(A6^5MEzBT`G-odCwIx
zLq6lG*;kwi`Z{e!B;9`6EQs6i8bq|?0|{i^pdzT4#(yhl=f};sBbJNpJIX(mCMrJ<
zL%`Ka7A;JpR{O)0F1)(FRK}WLbGy?sRosRHMVESl3eu5cnZiJ**MN7QX)0gsNzf0y
zAd&K8bl4Ye0CQ}#QT5jOp)@S0cc4jN4~yRt?MZl5t{H-0z0Zc9p>Rb+1jk1tvQ$Jw
zIYvfr`lq;b3ajGds$D&*5ccbK6K^f{N4heHn~E+6xsZmO6G-g!$wbEq?z8OGd<(T%
z=}Y<A%2<@{lt9sljB99`j>63jgWQs>LQ>9`G{auwM6%UGpPngHUBchRkCq#o@Dzk!
zz;HtkpyMGU6Fr^+VQ1UsHmjJnh72K5b9bORli&F;X*iI0jo6A{qsnjiB_bC2uJ!D$
zsl9cgD@t#^I~`^XWDj&mQPv28xv~*;vdR|ZIlKhcprlN$S>gTTngg68P?ggAs%B#P
z#Q*X*98%JI^9GQ)dg#poXNZqymiX3+Il|v|64`G4{JK?xIyGNF_w>W@g$=xAGQH^H
zr5}Y=Qd=9WKVV_9X@d5yCTC(a0F%C-8dbP-(B9i<mQX|IpZ(_9hb%_*-j0UHX<1e2
zpB3Dm0I@E%VjzE~n#Q;pq{nk4L3muN<iPY0KAsPC@ipkKHJpb|xWb==^_?3H!~AN5
zG@KE2fnIo~M7kP0*}UDDfWKkqYTf?iae)b`!1-ZGvDH~0o@<o_jH)$UbfK7fdt1Tg
z)3X;_<6gaAZFyT#JV7<j0f&i6kw;(avofWrTcw->7{>Dbw#UP2vnH74L1MOWx|_}q
zUQ!akg^hbfcG}7aDw^UCkLUP9Z!1JoCikKa<jn%+lHd0rsYc$92%`XkibHkKDxC|s
z7Jo;mVqhxd9Vg>H0ySkAaOEJh=%Nbmp^8!COK4cH13BMbwCzgi+U}L+f1&sSvL_QW
z8~IBf8buCVeT0bM*s4d-*W(GZE;Bc45|=sB^tVY_$y|{>JVmSB;I1Iv5F90^dzOpc
zI|CvM15<4~%vT0Up6&Ub<W>UkLnW)P;*)gvTEiOGyJ(|drjZ3OoHy<V*P4XQP{RHD
z@1&tD2xiXi_&=zA4qW@x9*^cij>qv@xx^Te-jqz?^Ux0`v**di_x-)EwqgO_Lsr?<
zOSDk@41I`O+)-4GP)B?cwH+$?jK~6^p@eDH9l1Yvp|jP<2Cb?9I8RDUJFVeQAJ)3X
z{_<-cJK*1tH(+`3yQC6d1wD?h44O#yAo)`b`L+*VLH!&mbLlVJs+QpvP%k<|@w?Xe
zirlSSe5V&pLuhaQQ(28~-1+fA)vfLNAHI#8Jib$!`lGv2e|nuV!*hWqV+DT~jpA!;
zpVQw0IJ|eY{QvaY(u61AG+np<OSiemD+fGTT+ZM4(`$GTzP2++u;(x1=&Erz@V;+I
zdHOHwX`hf-Zr?8BFU^Oo8oxcvEqn9%Z>8gZ74lz&{MSSNKVC8s?@3xLTcZk#?R=o#
z<~;w7CRbi2%RM2Q8m+V*C{6Mlc9@QCTPAWh{S%)60t3QyA>gu?mr`On!x&gOHvKH>
zEPHG`tNkz66VJN5vr$CSGB#m?{x{2W=6~`xor#6MOV`?c%|O-F)&B==^o|P6cq27T
z@ei%>ii0bj<ocnY04K_q?KRq+Y)UHEaPV?7^+Y$gxp?A#e$_6bpdfmfFZ(ohtW1d(
zDAvr*GS>aD^IkKUsv$(d>~Ac;b)`uzk?r(8|JXi1z6z<tBM9-5FsG?wv0q1WsHo_q
z>|oZw75i@FZ@eUXbo(t^*yECt({;VFd6t;jd(@<IK~+a7a)%K}VwslrEsqcWMM?Vw
z1guk_`J{pcNt-UOgzA$+ELSP8i2C;JVk%`~Jw~lsE=e(XrdqF}j7=pg{Y~qC&{dxi
z2WuCksdD<WK^|Yl%ByIkHQAxdA^j3~9&s%4%71WwugKyd?4&$^^N(d(Xz)>AQ<vz)
zUtsvjiKn7w3$H)_L-+KT$HUJnr<^N)ute?z52u?|^O=7Di9i!iMU~c~q$T_VYA>_#
z*bzG1Nb?uNF8<;clt#$^g{V#Zu~u8}2S52gHj}!Ir=mU*eoFR7wp*ciG(9ld*7!4U
zIOi`M|44G_FGhXQ#8Xja!IKyN47~9ikAyTsa<~7;7IE;4ifZ32`Ipp?bp@YS<T2hQ
i{v+G}Q%X5Fzj6GX@NNZs9seHz02M`zM^Jf-Xa5V;XD-12

literal 0
HcmV?d00001

diff --git a/docs/blog/.authors.yml b/docs/blog/.authors.yml
deleted file mode 100644
index c3fe2ac10..000000000
--- a/docs/blog/.authors.yml
+++ /dev/null
@@ -1,5 +0,0 @@
-authors:
-  remilouf:
-    name: Remi Louf
-    description: author
-    avatar: https://avatars.githubusercontent.com/u/3885044?v=4
diff --git a/docs/blog/assets/4000_stars.png b/docs/blog/assets/4000_stars.png
deleted file mode 100644
index f3a596d7e13bd919064b87b463d2a4fc1fb5dc5f..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 57307
zcmXtgV{~Lq*LG~%=r|MGwkOHNwrwX9+qN;WZQFJ-vF$H&Ki~VKd#yg}be&y$S6y3I
zoiGJCad;RU7$6`Zcu5HnMIay$Cm<kT1Sp8_D?l9K58vO=b`t82KtKrn|31LOwxvej
z7qOj0)tr=UO`Kfy9gKloU0vzSZ7dxP_3e!5Y#mH9E_rZ(fQW!3MFf@IGS9PI-S8Gy
zo3EJH7d#t{#+$TySCoIt_S^#j%Yfts3J5_F6Xt=Vphytiqnt{^_X@ayk`muz8gdom
z*2vm}Kjm+oEu|g38M>%e(5=U-G);SJ$E(ycAMr5PKfd2L6o`OLh6)3R2`CaFQ98A6
z4hkSEn5nDq#TF!vK`<pMmI2~HNP!7@;5qsXp^!iV{0Pk;HT+F-e~O&?Y4kIc`@!VG
zt$-;1QdVxj#K7?AH#B3xi4rTo?xmzX-jicefRllEtf8?i*T+Kj10at=qU`A<V3vZ6
zB`R9_{pykG-u}uO-d_!;g=78qx7kRfhFn}7GzNl|?Yci{Qe_`CL0TA1rq+e~kf%wC
z6*S`-7Fa_^J(xY@4HWVPvt}&mEqUOZGkXw2z#*E+tm0J83<^*j`<6?&C?HMLMSN_s
zcitiWwMd#f;Cq(I-w!ZjX-CwUfu}34tfb}V<aB4V>j<3yma64-Y=_0DQ4ztw31%cz
z16#=Db^A{sIPCAc5y+B_KLn^)F;k(92G(<tML=st>i%HjBMgJ+S$4qB{hPCisxWNZ
zg)p#tvT>FH){r@X3ALU2*F5`051?qF+dsumqv1&IKZ)_hVB4%o;5l%f(?aoNy{Hvj
z(btj`ZGq%LuJoFfz9syAa6P`6;4(i<s)_m%r&QUz|LhH$8KzNz;;C+r(h9BOeTd?J
zLs<*Jwr$yi41;2ZkLa(+da?>f$WuMAT_atS>kX|E{L`qJ3rs`qnH+>&ai~_nCW(z}
zxdJ{FRE-vuGygT=c@ph^=9;|^z;MDx13C&D3F;|z!z~D_Q>S_$x(0K3L;Y_M7VVh;
z1P7Y&HW{i1X(IluAv2F6Evg3w7mg;r?;rmLVUAzP0@qG9R`32Pr44s5w2n3DSEekr
z3HuIj7T(PGKl6gQhzq8Gz2iM6gdWeFPK{#}Ej;uv`?3H3F;L}QC<IM`D1ev4Mx2~#
z`>XsYi;_W>L11St<o;7jMqN8pTN%Is(vD^)w1O--=nyfN$0kvKU@=iU6zMu#@ie9<
z{agNjTAZ}2ulNz}G+t+e5jWU_82UR^iBT8I%InpfkN;kQjqetzJtUa`BeN$)gmkY`
zI;@`ee-{7IUY^VBXDDhEr+8{{-qG3u@FN0iO3v%x(_m8#H0<KK<qj<j9eDuULfe78
z?pb^bX2Pzo3fKA<s5yLu(~o;=$Z62Zy6EPtT4LQ2qN;(#X>B=nS5;~f^b?NdTQC1<
zG&o*cat6D{9&k&ffMRR-DwY?tah;o}0r`&O2I)QfZ#&dS)Kvs(G1HE}5+b{YA=CdA
zI1Aqu7LgJ);<JN{#jy4K&vw`_F5rlpz<^s<y(>3dCWQuH8#|;XPRUEyqJ@{9Wnul8
z|CarYf$8H6r$4Odfwl{P5mIB~f-(*SwWjG|22{j|*pA@FM)!YOBdV+SIIg1|Z_}l6
z=HhOq#7FVC)3``<<$?fiQdW*nOcikPz@cm>j$CueDJe?yKxx6#X2|=u6}A1ce!|Iu
zZvB+X8l5u#LYOV9#@9BE6awf{?~R^5<5+<j9APCR5QETnJo2e-J`xEc$6jPFBE>n1
z9$4$m<*t?fGn*{lN*P$3c{)E@oU5pjSci2OWbK}4-}W!%a%jDL2H^2%v2dqP!0j3c
zV=q%LoFL`O%Q)PsZnF7R1UjEs?1)-&E{$N=2YPl*0_*P!RArdI1g;!8H0t9N(yDO5
z!5j2=0lYRL;{R=0^5}CKVpUALpwJB@pU(A}1e25N@URPV;q%-AHNA8R2ERi~iiI_y
z(n|FCvq2cJI0d9y8ci@91kV%3egRH61u>4mRcE=d=olsi1jw;N2zijILZeK{(b7PX
z#ChmoL2)P+m)zdFEp>pF;}ySJlLP!TzXNJjuR0$s5x|!LZXfvEc;G)Xu9jil1XShc
z1E~1<zkG@ordLOK!q=s2_bS&hfCu_+-Jx08k;^NIm38HFD+dvRxU^g1mFx<O_M~L{
zC8UNVBnEbG{raKBI84^6cTTTch;s)(m;=%84Z+`2w_JoSF4T1BQDAlzU=Ho~WhDFK
zvi+Gk(HZe5SzDT|FOnUi-&P6TkBi4SEFQallVQ>#yRR@rZUcAqV*d|mcJ&Kk+q95h
zAnm+5;Ia7tPb8Vggt7t5jJRj!+a2J#?$}-F7Fxl$rY_`F0YWG+Fm}a7`lRIfcnE+n
z8<Zt$K(6viK?&)e<C!X47!Y+P8KyndP-?k>5AwXoyAKGhY#<J9?V2WXkAy5SBRip~
z284kVE!{;;zhVFi7mrt`Uf3Y`@`m)UUnJ8X7hGBhhBYEtK=e%w4VgK~$twID!vD6R
z81sFkk)SF7Fb(=PxE`A~6n@Yr8PbA0E~KGprrnIO9SrX_K4wo!9<Qb-)R#I-ih*7b
zAK#a!xDHy}^lMS#UK5hH4Vq-D<CD94RXBJU4S0OO!g5emDQI;g(1FX&#CU)7hd?Wj
zpAgkx4_Ptq@UV1><C3xK?K6J&;dR<PW7EsQ+ksNiLYcse9zywlAf&*k2)?O|?2es^
zWh)HvyGl{GtqTVB2Xq=(`496i6VA#CRb-2&I2IqCFeUPMCLa(n3z+=KXyo?d@+cxe
zuZCbB0$&iNze1><Hm!df4h=Gox_XP_?N2#OYP_GAi>gQ~p9D-n1yfz|@%aFP?hnPx
z-wD!~<5EQnl09v`1Zw~JiCz;@n6{6;?A9zj=<2SM4UNJI*B^LtjWq8`=qxlEBI7C2
zy#a|Q%p)tl{3t__O|VE1%Wn%Mm9-)?1jiJGXI-;qmEj}psp=xJ3Vs6K`2K4aiMvF3
zrC6pGIC{B#APgvbtZhRcKR51H95<e(!8>kZa8uQoU|64*PugorY*zIU|NN&7J*GLZ
zO+mmdo&wspHWE_jKoTiyXeNR^PWfMAi(~l7cHm<<Ju%Yi5eAevV6^BII*sC;JGINF
zbv-vabBJA1rwB%}#ji@|8hyjJr8Px4RV-n%JG?epU-Ynk5}SHnoG)XzI8O}K!9;Vt
z$-mzp6tq6>%`-x@J_9DiiWaDP=JK$#{-+W>Qs}m;AYMoJ8#W`&6YBUj&+_wZ>%F`A
zItOg$JKYld2b?<7OYx;y@dPGgm_0j@n2UpvaP0cKB|amyY3&fmi6*Fy&?8yrG^-sY
z=|5GVG+vA7U>4=}#wPI!3UCOf!xl*8`==Voc2ifjN8{+4FXut+5|a0DK`98iLR8Lx
z8Y4=G<$tczL>!wcz$Q7g1L|OSJ=U5GKu;4$U`b0$tgI?Lsd=UN@wi?ZND33pAN=PE
zKa!dEcc|AX=`2L9Bw(<bTH}uZi>kGN6IJmAm#uG3c<L>@2(;8|m24Dm_eyPLDzEZ+
z1^^7#;zlKTcWi{H){X^B&XfP=Z1~C-om<Wi8~1?rs#ow50^bn@_Y<Hd$|9do4CjB2
z5ZgZzOt4wE1<0TTrru@p_!Lj2u|oMH>DNNXGsvPXJ*nZB_bBIy$Xu~pW5%_MU>``4
zWml%t3xT_8kZ(dws1kCJX;9A<#ln_c&8GCoQ9m3QgD~+Q1-A!CFk>t%ISParNL8V4
zyd=dNmHfs_yEcF#?J3i_(J!~rY6QkogEi=<FZWa1nCLCn$8wDCwg1kF?j{yBl$9No
zdyYXPAghfk{(k=?L%<$wO886?oY$toooWRK@tdM`(<V!)<ey7I7S#gRCPG@7zi{a$
z2N>>|47ki$Dl;4HZPvuq`SJ?yo8~0A93UsYA5gg6-sMMep6~bX3ZK};gvDQS+GK60
zW2g`be51dwF_|?hYhbztk`hCro;n64C<!qz2^qGy0YI)9H)K4h|D;YdlL<(W26mhU
z8vHldhmQ~+zltYrax*$Kggww$0|O%iiWbM5lH};*U{*>9Gb?_70@Bo7M4*siwG{ud
z8E&kEgbIz9CY!#8rX&tK@S$8CJz-yUbyr9;pY{-L%y!Q}?V!Fak5Rr&r{XD=(?i8S
zw=RZsgXTmAxJ6LFp$COPpYLn8p-uYo3V$8cP4J#Wq~v)BadU9!>x-&6s~WdQwpG2)
z2=^EsIP2H^DS7sH@*8w(WMGqG{eG~*%li;ewYqc6D-mV-Qt%$clo9AWh{WUs)O7CM
zp_*Ccc9tjiEHO<almGEUJapKn8X;+Q0to{L-?_anJbN`JN@_sq;!EBZe=Bz8m{+_$
zNbytT;s4U6`#9roy*1_EnbUnovUAdbGQ3tt#&B+v&-zb(FneUD%F0T2Dk`cwQKK*Y
z9IqKDU3Tf~U&;!cSI`!Guc;e^c+;izQ$0yp5#fL4khBd1`|^L_5k|40V2GfN5$4)|
z;H<G(4cx!jbMTU)9mHud;kcTxwQM#$NlMD7+3P*g*X^A}yG<I*t)Jhlt>38pnyuTi
z@jCAt7#SmXTx0phJA~>B9LC%V@J(K~b-+1@^RMUfF$uaXsf6i+TvWPTqYbsc{64%k
zC61-xBnB@M--*D?52=4Um;7=-c!n%u#Mgg6*mJM9uSx^FzBMkqB&X<h5Oaj<R3hBh
zg{>+ez-+Qas3x&5^klG@D@Y+oL5bJ-hbmI(AK;XP`x{gV9CiY6#r(|-k~!bXY4W~J
z;l_{gH2=zxnKzd(u?Uh@$7HHN3rn+i59}>k&E5-Q&+QXZ`OHv<_6u)MWDQXCbdZ%C
z^&qhKhRP2daTmU$&0A7JTJ34eyDH=A(fwfI8Q_l6#8+sK3)O$2=Rn%!>Wgp$tean!
z$ixFBlL%6GP8<7n@%ikqM7F}%LI@$Ld0*hvn>d&jX}3(@7S%XFK~uV)eH@yNWp8AH
z$g;ZAZki8F8PU1U`!qh9DXdD!jHn2@Gl{X3j=%j9^`Ws$Z9Bcs4(2id<n|U8NN~-2
zfmTOoQ(Ru3lp>D#4G2YgXqDJt@$UJ<^W^-j3=FP0q@nC>zQG-znF6iG?ndKp_ZyX`
zEWvs|HXJm61I?LDKLZ`aL|(Scf5^RsWrMi+3*dmtVAM$lkr2I8O6xOFXQtT|Q2E3g
z*qxW2A3;z47As6tx6=U=B7XuP1N-Ab_^|C2=4;70H6b%sauGzqzl>qOspYzP0fjh!
zu^64D?W@^_n$=2ph@OE-DNx#!-6oH*tKIEi7`8vkd9Dz7a6rfD>e4$TV60}vdfql^
zjOQ?fE<;WCx6;(6_hRD_oy`Na=0lIFhz|5}yLj}!kPcfUSZO8;0Ql;AoV!pCg)%Wj
zO}Zl}(eJf1dVgktndXBvm69?-<Z8q~4ueV8GbU@y@e1ZZ!)AUYH=i&|Bmn@u-HAK<
z(#Ll%dCNm{SHI%W@5L!)spN-BDWD@O@v2t1Yym(~cu{`Xfz!+vdOx@>bUzWt@0cu%
zQtVBPUS7l4m<E;8Bkip?Rv8>NYd-J1bThec9M?eF@Az`x6Fgq9r(XP$`oK!%w~JuK
znIuwx|4whAqD)8`w1}6zMGlZ%#27K*W}iRbtXhwJc=m^xZbm62uz1teFn^T{clv?J
zrl-)C7#~vOfrF#H@3<-FQ@rwtzz}mSs5h&SN%@yXf4utySYM<@O}bV5xLHy*Tv|l7
zvLc#UiCkSmxX>9m5`PyQAnbvGxKOC>@3CuR;Gv5ILnI|PC?wsLlkZQ;3W`e*7ZT)?
z3<~0ypSLR}&C{|0BhkpkwbYQ;X{hRU4OB23BSQOjCRCo6f@9g9NTpSI=dG#?a1rnO
z-7)Q@BYZ;6clM7F!-q!O-sE*w7b>4uMa|g0y(ca4@#nSYV~Kn*G3bsU5)2Z#wb-s6
zIAa%J<2G6Oc+uH3(r9t+Qe`B5zt)Xf#0plCH_ij6(W6W#u@DfoqeDAj&si5xFQByh
z)5ZeytXVa$-2IBHQ<R=FbmLsl87V_LHT<1;HPH4DXRZ$oIx@y2)*WHWl%7O5qv$+c
zLi21=bV#)*X9KIPn7ddlp=~xIZTrOnl74~V;6OJu5*z5Z@Ny_A67zy4rH7MnVyj9y
zv~boy9y;Y`A}F`aDa*h@gT=ZKnpkq@C96y%tC(5$TG;?*HUgPsQ4|&JjE#3GDSMPO
zg!Tq*I4I<Z-0#gI2aHJPJn2*xDFwr33tVfa<x`+Akjl&QZWmRNQDKg<Lz&qji@2nu
zAp%0L`)~>Edvgd6jz6T_k6ld@lo?=5EvR{=Y6wCloTT;cco7vp)<EXudEz8f-ow*_
z2o>Hoq!YKEF?rN^xxo?@Bm7DH{{qW?Cv+}}Vnn=KVrDEP{&|lx*nnFk60HeAaipKn
zq2`L+*TQ30ZIDk3qcZO2wr()6ZHY({;+m~`q3JPvq{^DYLtT)utd1CC`hmkqrK&$r
zf|FA96U|ZnROacgvNzCt#5W~XF#Mp7TT&6ia=V4GOF>Iuv1F4pl$M~S8&Zmm`PQ7N
zV!k%71r{Umf{p(|>0)hmT_`aMjQ$G<*!nu7rKS052YAIFysI^57J3b-7f1Sn0KrUi
zNzy3X2-BRCLc$ut6VLM?QnI{6{fJdMPCw9i1^wB-W0S`Btw$&;8gi??+5OrI0=>|~
zt9{DOHqLXHo>%JO^M?l!#%Ge0G=1&PB$A?rjEYKc_{2>}BET1#Yzao-H_1u<E9jGd
z*REm}YL*#AjC_k0I(z8Dv={$}al|GPyiUo++Qo@&rX%E7R$voD5Nz8|3=f1}60?Jt
zv95Dka3p&_6!8hzYWp3cq?6~GOlZ>}r^)T#Ik*j_2VqKUG9yG(X^9t>r0#BQ&7SWa
zHM*WbuKrGP{4Bkfi7&noyd_YQ%-{-Yph{p=RP3<}N(w$Tuvy7ixuQL<6V@FBK{0H{
z7776F$Y`EvFaX~&{sHCFiuWA{70(k{fBT=wY}qB^`s)MqHR)h@0ndj-POJ~C@R=qr
z^k3Ce6i3JXlAE@=>zXCXgyQQN?<TRUB|1Y(w2K|jGruObJYk$Im*W}r0l^gZ8R4`Y
zn&YKjPxy6+S;7{hSjr3nHMPj9ayAQRtFXzTT4q>7B%qn!wKFUJHQ|!kKx`>Nc1;l-
zNW~C~fZl!t=O(Y#Oxos+;4ulqiJU5c`NoQwCBhG_lkx!_whHG5DUEIo!LfVnLYWL4
z3_Sd;cC|_3WTZb8dHOdg3*OhjpVi%g!D5-}oTTy@LSqxZCAK_m<E4lE=F!o$HkEGL
zgFQS^*yD-;++L#j`=Un}J$2VfDf5pHOd6Q9eeDO*YnEqr@>vo?j3Eim$NHq`=ml`%
z4c_By)s@r%hlmOvV_qOp@MxJX)Km9PXaS15WOE)PVW_nnS7Y4nCm+=<ryB_^d#WY5
zB4%2R0S#313|ev!C?n_tQ)#eiJIa*CNVFns2Gx?)ER;pjAjvlCVU{fE)*`T6QREW}
z&}}rx>La0ujlqK&L?VmbpTdJ36r;~KF<6eD^me4=D|@tSGpaRULg+~gU0;0AMMXyE
z?bTT&OU69NZ@adQ%}QUZM+-D=Nsoilwdre;SlTkD#E=P$N1lapX#sKRI2d-?XVP!g
zDi>#&!SNC!YQQF%xU@f1=k7nnS$wXE^VT3xo>GPWjt<9p4#Z)FNj9zE`_-H&ry^w!
zwCj`nn(mVL8y6DS^u;&JMv*mqTH4lm6?W8e8Hbt1T#$G~*oM)++hb*0;=HdhS0Z=C
z(S-Pm0$}xg#9!v|A$uFcoo#QLBRa`_HB)oVJEUU|ei<1l_xR{WC*=){7c^BeZx)O~
zd(4u~ARH%bs+R0rp=fas-Wm<zm?}GI)p978-mN~mqFGI`zoU*dhEOS07!^WEqwCk^
ziiv`f@SY$)Bq0O$=3#Kk2hKnXVys!}_F{~q>y2+=Nz5KIfqE40IIvrI&Io>ZgLI}{
z-zX~HD)RODXV;u{-t(|rn}r&Rnk?-;4KmiJ2RO3p`D^xmaap<h6(Xs5Q~VJ0PJ;iF
z5=NdIUEaq(i+Oo`)Y%#v#auU@!v>G6yEAs|NGIR<g&Jm>3%>vG)JCgi_4>YS-=13S
z=E6i{&8x7aXh%YdoT@suY4S{x8k?}OyZP*oh}o+hW7Xo>O|<s!y6@bbm8GLQG(G~P
zDfQxrJZuCQ;E<({mGabC@?XV|drHCW_;2{gou6S1<kn9va7x;|k9Ah7=W$sH3z#DI
z_q4$Ld^u4#sdHA9b(QM;le6j(&U?xY(U)0^sg<FoXAl<O#=vmjkC3Jq%<o!GJkJ|Z
zov$5;dOadYN`)%$EHN~RO~L1~G!}vfsuO-SdTmlA$|w^W6zh`G!<<*St|z{)JE!_E
zf#?l(Nt~KSxT0800Hw#0w!E%Ywff!O;6mpU@!(PR?{B|T5W+ZSdB(XDlCop{#QZhe
zZ8y_3Z*@gI)8xYntsca|#B_dbox&XX`T~}oLR)azi8#u0&M;SU6;@piqbtgJ;wg8g
zuJ9-$LFLSIz2x_}WX7@{{y1aKq7R`ta~LsVDln>tz*QRR2jxuY)3=myMdbbn9QpEb
z*b<+>6~YZJWNoe7bw>~e9%OXcOdmXdOv46N*co*i1Y72Q%HL3_B`frkc$FeYVCrUg
zmx{WsQ_;Kix5tG4nYrik(;Kd--Ldq-gH*`e;O$H4$GK8@wpa2atK}%?I}(h~BdX?#
zK9SUM!;ZSG2HQL5Ly~IU!)OWB^KKZaY+B$cnh$ya#zdwkaR&8i>*d!+Xv8!xJl^i7
zU$SU_`Js_a8b+1SB{vjarP$@!<HNM8RM~I;wF^R_g8!=aII^j|XimY#>M>MBK74AR
z2z&vlLX&s&N}oObMmbN+Vp+}*2ql}Ebf-V+bDbBdaa4Sydw+`WPRU#>xvO~bQS11m
zjhTK%S*?l8>3GJy&F93skMv@VN)AW?F~4UNIL<f#Jw7&V_lRh38_+D_u6%5M(j4Q4
zAq!J)SaAyUALS@Lj2ODHTakSfUMYEPwYD(VTlKQc7`KTqtgo#)eUTX9IYVJXo)f|=
zi?2;(I-gi~eBP*LUUzyw4|l5O!;xsDSv9fE>im*)I3Wc@83*GayEAlbS-x*p;Y?Nq
zVL1J7h7>`<mHmyc18l@lZlG2kJIERsG<plp%dyTQ=VtNVq`xQc!Am{PytPhFjC^g$
z#Zp?bSC`}-0@x{i&ly)V#OAu@sd&Cr3#ZeA9PVjWm76O++;{t~s9_5|BO)(uW}RJV
z%A>FI>-rYk@xAYwoX@kLt~lfuYmuST7P>wVSnmkPW@cnQt9>Q6HWC;%E(yZ%J#e*Z
zgvbvO`Q^q4P$fBU8SwP!)4{bIkNDl6x4(u1`|vP6pKRh$P`FoT8$P*F9PaFkiV=Tj
zf)XJibapZG%$u{c!)yGvNJVE@Sy_#Xi;BD<WH#&*Tq3hQGoe)+e`wltqh2kKV>_Qk
z6<#(%Lb#w?77E(smS@+wVk$Dc5uprYIQy);fi84>@fpWV^=oxvOm8t`Ie)eqO=TND
z>7LD(U&2dx_|ThJ4wOoC!-d@xi^|uH*T#9SnrAJA21aHrZ&5Su?hshOz1BGB`EWyD
zEzLJFYHF6fKRGV&Kd@YmCQ!b9*h*5<UZ4nFSlab62`ls7(4!2a#l3!z*squlqE2`9
z5?*@2r9kC$-ZB)xI@Y{Jtqe&p2+XJ3O6D+G?_~Gln+7SK3glY1kT00IfHvgLoAtpZ
zhg`ZKw>mIDtRWkG=rWqyfLWfc+C4Z+IFgVxZ%3L=9Gm%aa+k=>)}3?Xu>C1`P9?D;
zC3AwafTkT!sXy$3y!L(s&mKjkYT5B(I+_q{)yf)Ba(#tKdAx!~MNTBs(@>kFZCF&^
zsV)<<D8E@=Fko#K(Bz(53V!xhD>i?Ua^zshlv5T0DJ&GQfT^00(Q0kQdI2xQz}Oob
zg$U`_F$SF0o~e5|tMMpo>z#ADVECeFMiK?M_Cv63x)XMxSB$(&R%1Q%w?5Vxf1J{2
zUTYLfuI_&kS-*oRMrZ!c+xZZc<~ZD2Cn9}+^Zo4k88UH>MNgC1iEd2U3@Pt&Y3F{T
zAoT&>qMuqkZ{`B|Kdl7kW)RQ?N*udR>*rQnQ7Y(+w*B}5lcO-aY6%M5I6o$k5)i`p
zb>$uTgDfl(^r?S0G!!0cak1_sPu$Gin2(wysYCZ8{=L<)-&*`U3*h*W8MFAYCN;3W
zB)YsR`4;fJ29=yLf4kY-P^}a6op7wNS&QIgFa{9A&1VlQ=4Q&Zdn}za^miikV`wLX
z=OKapmLLj+0k;p;OLf6oRwSpUN;|kgWsn-)PveTaH=nmpNz@}H8@?YrD>c42usE~N
zd;8=Qf5b`-XED26A<rZFHQgz)*e@U3@&z}f_2IdpuFk7c>yq^|j3^H&<7kq^I_E=z
z-;DehYopBDr@LPSzpfihN|q}*rKZ%H4;OH)nMP~oJ#Wddni<><k4xxxzfeZ+K|wBm
z_FT+N@H-e>{V8~a7WdM__}ggOWq$ICY-hV83BJ1>MqCGjta&5y2XB!fqBO(029@s{
zsDsJ5?B25hH&}$wyP(CESHdkM!H}5^;30(Kr`MJ{M}2*}MAH_J5^v`+@fQI4HB_&u
zX|~CRHcA?E{&f+i!?LS!Fe8rvC7z+KP6Ce^XZYA4K0MqG9b;!~f|QsSUsZ}}ECGm)
zg0deWk2+NOHrza5J7#q&y1F(@K$s-&6vyi@DOIy+N3$7%x|~5X9Xbp=EsazrL+Ew_
zbd=AGJgi_CKoCQFtZZbi$m$ns=8K*-tRDJ*T!1#-4tk;WwMOL)w;z1AevHdHvSSTp
zF5u@t#r5xeY}@vTt^Az$-!hrtIkGv^k~_w5`H9=MGtnS$ZxibN0_VVUbGU8GT#pTJ
z!%R}-AF4|Fu8?vE{|B)6=uRbJBg_M>M_Tesz1`-pXLOw?%g?is?Vt1<it*O$Y8=YT
zIzJ`%A*DxnJrJtS7fMHNt*H#>FI@NEo|Uxb$rYUrufE%z9*B91TQK--UE4PqLP)R^
z#Sl0c^u?F`G3Gk&;ELKI#_BD%x02~hafv6MaZSr`C{p3qMu&d=k_{dtOg5x1)rQx=
zX^T`JQw>n8tUI(nN{n9|G*W!RY*(d}MOh7?qV35Fr#Mq13!=%L$Iue{v8lA8u$#7r
z2tfxPwIDJwDBF*G(zm`61Ne0Ww-`g7P?r!0X@zQ<vcFADbTv25aX5GO;<ac}oE*9L
z@wFAvA6j`Fg=aL|;sJi5R2+1K<-iE(V%wJB|02vACsvP}3!Wv0Y9bD%T^H(waLyR|
zXvNGM*4i^f`=@<R3U7Z%(c_~LOA=b2%<yWl%Y=8bVkzGBJdT5h4v2<X<l=p~tmDjA
z;ghY-$DZzmpBL9u(O_Z(v|VL3v^>x_)6Z?a?8}Rh)pnxf;*MTHwo*a}VE#5i^pdM?
zG{NIHFe{3c4W>}8)Rhi1%^RC*QqA^Y+3+!W`)a0LpB#+HSNPER6f2u?u5f(-ZS-|a
z3KPSROu|WqmwaJs9dXI&)QL@ioCZ@906g~fLgVb~8}QDrCAPIe7?^^1R$`m!_cdN!
z=JSScQ+{*%-bZgw^5^3kJ05j|cWt%H8MSJ`>D!ljXAbAqJIxo%LdM{8FR^Q(TClWJ
z{^zq?L}s+<0Iu-GjO$W}&KY3rHizxwKJr2*SVP+mBx4uA#?`NS&@8mtVXKPLOZtv>
zl$!BiXQwN6!@LcXzW7gDo(z9aD<spHic;8c<|;(Oij#w2X=`LUGT4+}ZebU?KOq!<
zg)ox^6@r?^``jZROr=L5W1xUm$Ky*WQ>>N>VbE5nD!r|Nq;I(+{?dlWJ~7JlLUAw2
zj5JlQrL9vgjCS!9uFZO?8LScA3}9b%_~13tw!;`a$_#SA>4cF8d(q*(GQadWbNuY8
zf)l3Na@!Mo%wj7V-v5)E(-f5<k;Y;Ee1Mmm_{w@c!XQ^RdmjEj#L*E<Ap!jG>+pHP
z6Uy(CTE(GO(;02<?gVkQ0WNCZ^dljoI)A7Y&T7n|6NCGGuKtE>PPTFXux#~m7nRDI
zRG8zAGuZu37M*C62vrPHprrHbiqoOC)iW@Fg;rhW&^M>#VP3RUz%S5zeuU#u%^^DV
zH#d9ClyPQfkmJ5^!wOKoOFHf3cc~9~@yi>*@x|fK0GMLT7oR9UKcxK6TJNNMidw_I
zD&YlaXIRX-hiPB3xA%H1=P7vwv8z5tE(@ZbO4W^ttQ@c3{dV?WIa5cGt>OWvTl4Zr
z8p}T74w6*CmTH=mHov4Upji-)3UMas{)Z74-T~S}P)Lrmm>?#G%$8qn@Kp;eLz`$*
zX{esw%E+RJkXQSoQzVAfWiH)JN~LbQWT#i2NRkf>ub12-nXEsk^=OQjd4?ePNb;tm
zuNWVpzGH+BiX4T{Zuwp9CY1gQ^8L4A1-sK2iEI{dttl)<lrz|fYQNu8{y)J(^SIIu
zC1gpi2KGmJFU@fHzXHaFbMkLmRn0u^Wb39le$SfbM3JzPbu?cLHYDl|JWik4H}P5h
z3_tHj5Z?T=;`Kn=QeQvHySri*+Enrr*sh(5^RM3Xv$PDHO7(gqI94Z%iM^@GZgn*o
zakpQU=!mccqaCr``iy+hzdn+7)&;UtfyhChV6%~*6Ks9BUSm_;XUwu*Fi7H-C*;(9
zLrsWejZEJACZZ%Rck)6Q8J0jzW!7Z#R*-8iF4mij@%tqE9u=j?AkS4h%+QDzH93vB
zXI-Dr<-Dj>_7m2paRmd@{3!Pk9E?0KysN$-1v2IjqhRI%b|kK_w(R?4kVA(A_v>CG
z0w<yeT^f@!Z^YUAV}3I)XJFCp&eZdM#O4T!#DNC#RRm3CQt+Q~tfmNBm`42%&HH1M
z#2>k!9<-vw9lZ1;=!wvKMi+mXL;|<c*WH<z&2{eZp<iB{yr6<VFGEMjq{@c<(us5a
zt3tK2tkWF-$N5R(L>sb#&9`Et=pxFnZ4VI;#YQFUKy+2%B6{>Ju5sQKH<H1UUTa|l
zdj3(a<~!+}y6q)A7b9;lL6)LNO0f<i_%Or8u7QGcXBikPIlru1dlKjV+^rTs!XEMy
z^CP@ZNv=1|TsL38Jaa=I!bV?hf(y?G--#&g?Ob?pyjtPzn1}l7*)%=ojjw>d6HHYu
zy0iLJ*5$!bo0KLyiFTkanh(a-oh+leJu2+(lBCv5SuMwzfE%TJQewQP1`JNEx%N*=
zTHlcYheEKB45~c`L@X?a85&48gbiL5^nY9%%Q{(3C!OkM4r1b6X(H2OnbS16?AB0>
znm}1Z9+~k9SjyHXg!re_fOtFiK!F!e!UR5hnzBLBP{jsX&C$OVRpL11Bc1Px4szKM
z4qScuO%gFJpM-w_NR-uj{l5|CUYCRtheiLUl_+@)owU?(TW-ev4Z=n1+p0fq4cl4Q
zm@58i&6)AM$`?atKAl!gU~~LR#h~#1)%Pe@I3p4!6yUS8ZO3e4jau}y5goU4vMNl`
zclw~2dbj@-a$*-}PRXZx&5@Rdt)2XOdfo6!{o}j*vnMr;B_{}}ke8V5|2SvPvi^H2
z?>qK(nnt+-jTv?XwnuWTT|r<59?ySVd|~-~au2V1AsJPo5Ff<DQ1-c|Iw>M6jHH9`
zFGl1ynHNnF%ni!t{*;=c;;&QHAc|);&TZ3*9x>buXDLHP%E<};;u#Om$rN7Kyoz@?
zN0#w^l^pu`yZpWF@_XBMq1OhXPaJHd;uzXON1HcAE`O=&v$85ZW<->Eotpnq7@>O|
zH1Om}qTV|YR-j`iQ`Hj7&GY}xObtEtJ-NiFP*8cXc6kJ<m&e<lwyqXIz)4L^_p7T-
z1*}ydQiR6qrf{eJXUMZ!+5oz5B4BT<`W)mYP6!kW?Pq-D(}w&bcsJ37XMGqO9dfBo
zUEn}INW8RI0ZqiIDyD+V;IGmRII(SC`V&aP(7ZO7>Py6%5sd!23M0*K?a8qSy1etp
zcKc}j!9Z~Wqqzt6SPm1JYQuB6(0R!%DhlSuE5toYZ2A3*ul`_Mn_dgsh>MfCL(5!B
zFgp_?rbl^Ig{KAI>q_12vn`tx)RcLDAVIy0@w?iNu4_tUZiCD8`5>_0BJU0TVTz|H
zL`B>9|Ir~&f~=99_`w}Q#$vZZY2`xAKy=l?acw_-l!Z*?Ur#XSFO~unnytvh446)e
z_8Dms#!PccUV}5@Y74MDuXsu2)8L$FWsl#0ZC*aCKDK>tH(xM!wcqexQZlEr5x=4u
zyq`|g9+g4;C`HE(Mk9UieIon2#CKZ13@@&}WJ((wAzzwkG3CyMpoxsvbc{ZZ!Nv$Y
z(4d1n5`%0?zGtDc+ULYkCTBN)KZxmMgymGFi*fGbF%}*{Ne?y^Iy~@?MzG3?RhsJ<
zF)}D@AxWqPrDEi*m=gofz8y_M8#bGw>(A5&TMpY$r5S?`7uR`&h9<N)KdtCGtc|5K
z%yg0mDS^k!wWeCt&rTw;WUQcF<PX(V2cx2W19?i|f%P=ND3it>Gt1A9_}xGCp15vk
zS)mHG&;-ByLqoo451UPxC2-t{sO~tJ6sWOQ+qXE(woaK1SR9gKJl)7aS?dV<8-po`
z=$<QA0&2=p7`>ZYqz5;t>+Nr>&iLDzZ|a*=b?ezYSu7|dLI<p6pY|Vfgqs%>-H8Hp
zKX@)STVXb5SGBn2X19A^vpcwUS?l)?TCxpOgtdwa^RhC4>vpFCK$-7>=9lY6)I{po
zz#CytQik-Cg6@@~3{AuuBEQihXE)LHAqvy<2bI#-BfFxM^u?3w)$yqx0}3i-=c5IR
z96$Jr9dgF!H3dzzEhNZ1{vnK*1GW3nXXIGP=QsV)$r56p0WMPxvq^@1Wn(3>|6-ur
zGPgap!S|`Kc-PWq!{*E-mRENE-KT^6AKPSdYEl3!ix|*tl8XOR8xvoF@|6EIG#FDS
z|DZV~PMIru<oAk&WuDC<HH-;CpwQ2cjfjN}PxhfLrL_BHZhI4x*#vgVy|^>#I~rfV
z${60!zuUU-l5|5`&J>0|-o(M{)>kEMD&S^qD4oSNm_@#?_?N8|JDyi0Y@3e!fqy;`
zt_&CK#&L-5YJ`c83*2uoAJ@apKzaKWIPefW>#?lLEl3TwLCv4vnLW(pKw%;SNoj1o
zc%1z7v4*NwhJ?ueb+FC!^bljPdwPYfC2U1RRB}54WPLkB{_IDS-=!Ah6DvWdDbwl5
z?D|9MV?nUOZsngww<{DGrn((%wm3Kz2rf-USUZ1Wr;S@!AGV*_=Dp^L+g-B%=KW4b
z4VP$?-SpkiwGm|IG4Yqh73eVct6o4){ji^N+_TeDbm6<_D<`-882a9D!@>?Sy_n$X
zJ55wOhFF+X#(Ege_1Y-&>RQnY8ql_bAJmeuNp=h!PPF6_R~_ZH80^s)IDO_V|1o|{
zi*yA|)t6fRJh808P%1;<Tw{@ZW1U?$g185TXUQ;E&)lg!C9BivG2(c=A*upJ(lNco
z7U9_q&<_|svpwyPj$A7EF3OhXwuqjof2i5dUYU8`!OzL@r<J*GNB!XM_O!_|5fHgO
zp+TEEASK^&Au-WM_%2L~o~<?RdV|#u8&ka!Y9<2iuKUf_YU1=nMUOlyl^c#Sg??_Y
z$T<=PlJo{WgUpC-Ra`ULcP)l}lbTUgVgGft>y_qma>rTV_80VPNrUCGas5wIbctJx
zpvin+QKu4>0cZ7W-t921E(Fl(oDG%g{lyhkgEI2@_f_1I)O69<mCTzkN}^`E?a8+E
z9hp1LKfawNid1z6O3Cp2{!O~IfE8j|P@lG%#H?;rA5y}TO$)AV#`2~4@_4+~M;?&4
z1`KCCV(#ZAVomP8OAxP%ODJB4KN9PMY@tkcpzq|G9f}cVPG$IgZp1b(JKR3SuV24L
z?U@DsF$cKGykWTJSz=J*`gWO$e}FXDZb%Gn6Ocb0nI<$Wv#e{9!{{2be*VQ^B9{dy
zEA_wf>qJa%i;cpw$1}M*P|PkV73hY!&o+2*ZtJ{z1~eu`ww`!|PxGMm+a1F!2@2?k
zNLa_<7~y*{pi>Y`PfD}h-bY-mZSJ^;@KcGtW(u2sctNY-x4J(H0p@1G)Tx?OmSTC_
z5!X~IFF9inf(BHli2;TV4i3?gIpF_6C-#Z-3{^ja51Uo82s!*0n!q=Vzn*20M$-~s
zQSgU}4G01A!&|DxHgA8b@eyuo`@(iTSW>p#!Z0E5FyINC6$G%}Dg_SFcN=ULb&Hes
z476VX-a<(jl+Sh2dx<>Q;RYjNr%ITZV?uJgLyBE&r?={W0Rn3}o(4<GNET*w%|RI`
z;l|bQ`c&X!xpl<^Npb!bHiM@ZpW&Sdc07!5`$zVy|I;kKH*_0d=X#e3jH`nK2uiu*
zo_gzW9Q$LKUyszSZ#^izNO~l@mVb<4<#vW=jsHFNJ||mlca$Qe-?g({-C$6wKYv=?
z^(86xIRAdDU2@x2_aY8g2;fRi&!5pcJXxII71YfC%5<K8)}16{8@et#Hc-~!)xJg`
z*wYZGFhUwnXbPOv_&`~L8lr{-^{A)IagOtq%`y!1&F=dyl$$|Fb$lT$-4lO@vs_8~
zHcl2tldFW`dZOS-@;q|8@>8Vtdx6Iu*|fOqv8f5axmlxvpZIbg(|q^S8wvO6U*Usy
ziCmngxBCAG&F>Xiz0d0sar#5MZi=2RP-Q*M!r_ngVh>LaIi%94J^#Y|XyA1f*iGw#
z%v;<RTegHxBM!(@xc5d=US3+yrzHcgC@M<H6AR3h{yv>h=!W}`JN-etTN-R^ZOctP
zsW)4Zt0a}hmKnSfc$jKlYreJLqNFs+eKuiN$E9|mvsN)CL7yd}cc#L&?Ex8|HgaHD
zHJE006!&ztjhttFNX_=Xi<3CsJHUNK>O59xCi>Uq?~x^hgpp|UiKmR7l_1-dcNd)Q
zOrfxk?HPPc!a+Q_ENM$49y*4hzW|ai(SrGC%UUqGh(F1X-wMKr@WHN>X|{HZa3ly{
zpZ}n1y@n=2fwEz9&^o=hfL)LsdBwFQQeoEiG2(ln6`nM{py@pSh+J5>=BmV4)b$pe
z+Jf}ko*v5hl&YzUuk~$>4&2MXuH9Tum^sX9lFd+a99&dGG1}0zB{LoyFx%_+<Uqg-
z7m){nyx@()$jPZrslKq!@dpU>l?rt@N6q#@=@N-Bppl?~pW<rP(DaUsk}EII6lLs;
zAUo=t0F(XC-X4Ixr1s0UvzFuhh&}7*U>&e5V|9KYo_>9SBk^(&Uq9OKp#D@4GVZ2R
zm}J^ut;ve0)r<wf6$Jh3lYVO&GtO}-&B*skkE^b}Ee0AtZ$m&9{_M++R>hHI-?l64
z{$#%2U5Vt{zN?}j(7Pnr0LrV=L)&fL1>VS4UJ7hnKn4}r?CzC;7TQzGdfL`om^s;4
zztGfxm33G|h3s!gBpSOlj0@dA3S1;5K!2&Ji7ngfRx;b?)f@*0XU}@cMqWNZj;bh_
z-<RTDrqcZlf8wgkaKe^Oq!o`y@8!KgRW*Uh1Fx&mvh?L`f8@YEvULX}k?Gay9oK#M
z{#DKOViK(DBl>+?TWshL=*5vs3xFv*=DMm^Y;k<KDiDM`5lx>m;X%`a>yXc8UBE>!
z$f3=Syd_Qn3MpM<?Kp8ZFSo1_f4t7ljfa=!rYL_qiIJ(QDRjbkx&7gL2~dbr1%+OZ
ze&_B#fBH`!8TJN%->_LWmrw|@!>stWz$5oq*c6=-%E$9Qf$<a-JyIhJ>^F~(t{8HO
zF$@&1Uy&O%F@MIW?n}s?-I-%1K@|@Uy%7vOtpXgTcq95+DT2H`;to54pVC%;6*_A-
z3VEzRv~MNwax;U604K3V@rIrP!|>Er7c#-Yg^+BNgPoqHE6m0;^CmA7lA^)rb7ZQi
z*zU(rzIZXjqT7;uFWTZC!TQ&?<Mf8<<Wf{RZoA+z^9zH>_JtX_%(8b1M9v&hNXFb6
z<+>mILXC|FIgT#(X*d>thObF?q-2U;i&U@awPl7*_{pt@?z@H~yO7Ftyt9VB_+pN(
ze^Yhss8aU#D59j0b%M1A1#3sG{%UiUXxReW3XQ|zjs&oBVu#EN3M*M6Qn1x9M~%@N
ze?g*#6)Y{X3sA^s>>#E+gZK>S#5O}Vt;=&L8zBulCw>S2#S|ieOG_$SUU!qnSFo6L
zg|)Ser=9a~{8Lr9iHgGr%LlU)?7$VRY2_tfqEp{|FUpZ-!a2Z#t(x$q&KqiZRsMUe
zZ?4#(lEY`U0JcPm&sQx$HI#KlSHXF<XKnaxn%m-{3qLos_-U{bei$ugB9t2W-UEu6
z6<}o}^6-I}7!?lm(0*4!ZYUuwEG|17OOCtbhvEq?uW4gK1b6SDlL4hD5n5<vSQthb
z+(9>(=5cWK4+yMB3q;n3sDXh6U<T3IjWszKgv1KZ;vzMSF*~@iRd~QKNcBI4nOH$s
zZlb|xJek#M9U1jbdPau()21uuVWN`Lpk?MU^)ZiB8D~GM`#EU)-r{nNm(H!vyCFc`
zeydzc(E*kr(cljCD?(j;XocHmFfJM4zH%6>Mq_=|ZCn@G7B}Il71UJDMMm^O!Dpwi
zpd=egr6Og>Sa<$*5@xMiTYmiSq*}44feF`l`3f%^EV17o1@`7Gb_pY3k~S4Mp|FqV
zo9VWb#+LL4^szrFDT*m`msRQl>=Svx+7MSEP>5KBN~_AoSJi4?0%jZ~5hbx78VY^U
z`B68FdahZdhQZ9tikzPq`n#>q0s@7?^iz&Ce<K&*&~SR3T=@<4BmpV9K4}Uq%cg}A
zdqpWadx=VNf+7v25un;Rg$dYyRiHs;E{h&@r(^Ga?|=6eR&3SCFIdryE;vE2nzv`4
z`y$*-2u1!Piv!=|&_&ra23E%K$>|L;-O{(GCi)09{1#+UqZTrBguU`<OIO|g44}$b
zcAUF6m@APX+K&~Mz|{<pk_za`o=^7U_iO&gd}4J=p)-8g&KXL|6q)ONOL+Srxk*ga
z<FK-CX8znax3(N+p=Fvwaiz?p@e3591hwNpuBEIlhxN28n$MoByihZN5_fP&;Y3ev
z#urySMNnfRl>r(8KHIkr{RTLohUVtG*RyKSEY(oAZ!4QyS&=d`GwT#zEIFY&g}@IN
zW(gI16Uw`zSXdK$Ps?Tn3^adnjs^{4fuJMu2T(@TcDEuOp1R+JSH<ikS9cd*ruoJ(
zSdtgW@xu$lI}c^j9<ZgT-gVP=PalmhF<ZVG(eYk^J!W!1loU8F?;+e^*6MOW^Vm~L
z+{QB4<AtFP_tuIlp~1~{`H;w7(HnfQHt$Kuk{>lBjBwo}!Ie<d5edPOtk#mSur(7G
zkCvX_uS~A{gXCZ)r2+-}n+H=?{JCPv@Mc|G`%v?D$!u-i<YW14!a#9!Xdd><`^)gJ
zk;h_Bu6SMJbfUOaskwadk)Dpj_q+r8cPfi}quBfya4cf<+*i39Qg-`h?sn;}h>#Xk
z(K6z59X_L?DgN1j-QZQ5nEXwEyBOG)*m6wwmLc5YT;D%3yvushBWuI*e6jrrot>)q
zyZ5OpG<AWk$?l3@cfUSKljG??xb=J^K~mBVbWF~_pkC!?XO}`2yZB!Pm<^lwVi|z>
zEh&tz4-UL^SKe*fQJMSu_Cc06m%2DXFo*W+mL+c#vjNhIh7PO&V=)INx})kZqY!@t
z(jEHhfvgPpe5iB`%_bEWeph6TwT{3@pzJUg;gjPdB*Ax|#Bp9I3icMYWY%GZ$l-nK
zknhaDGwfiz^LFgVtequIvhmH>k7_ePObP?=EYMWm1ynJ@y4@q^{rIAWDvi;?ZC&6o
z0V*@9b*a-*fbvM8*A`DeHdy%d2N@OwFYhsl!KhqxNadGz@=keD<m-hoX>xSTCF}8?
z7?^I)v{H4)boaK}fO`qSBr9fi`0z=zitdm~`6X1=DJ%Mp9FgS`u+h|7&Qj8k)!SxY
z#g}_us@F++32r^##9c75&XNAO-tfi@1*m!QsBL0cbeOL-R#KTXFc|7`3v)|b$Yxls
zQIX!WCt@lXaWRjHOqb6P${?(v&~}Y)E=KK_clu3}Xu<4xS16QDb;`@#5vWsz*)OS^
zsc+)a#)gcB1`fc$m2d_QwpP^x5X($cFP5EJEcy^p9m`>jQ>Ik1qJskLtHdm+&iUOf
z34gj9G{W<R%Z`t~&hzH60r%3RF8$axwUfz`>7^<WdDA$T=xIEfGU}_lT5^0nbgii^
zjyApez&G{zB0|Zmy8fgthH$kb6Dln3=Og)@$Xm?#u%$`)J>3<jZIVCjmntCW#w6@(
zFlq+1POBDZ#3tvQWf8EBxa%RIJDTkX(+8uJdJpMQrN=qc`0*N4*Dt?~rZ~~W4Gf-z
zg1wu8`Gbd<>)K|V8DKi$QonJWS&Ax|R&!{r@zEhldp!XwSWr|HJOYC@JjA`7w>H4W
z*Tc+^X$49z?4#!~#k$CQmf|4GhnfQq3dUC$Ka6u#FMWaU0nSGl>+Vk1iHhjh?cBxf
z2uPaJ(}19N2x3Uf37P|bVo4>@c&@%J>;i1LG0N67gS%)cVz67+J7(O*H*@>r7R~1Q
z0o#a-f*5PJv`S=HNtp5!pv{>Ga%dOJc}G<I@d1ny0ju_LCSCT~y^J((aL@>ba*C6#
z-s%H8Po5H=oY?}przGf!`^|3qJ`t$USKgSxSakF*=t~(Gm%33FDgtEd%?q8TKKsZ-
z5-Fl5))6)H-4saC@d(pnSlRb1HaAEor|-B1Nj5(5aq5@(<Cnpyl@QE|$XeYUM#7sl
zu^NJe%(K;o!!#s2&~8DtC7Kia@eS^nm=D^>IDBpHhZgBN+5)$`Aqh=iKDmK<!X%%v
zxD2$Y;)*EHtDIP6<}MQP5{v<XcV%25>!g1@2iTAAJX$~Ve1Ga}_H<vh#cxDP{3<-P
zV^W`$b$M}F=<B6ZJl?Bj=}C*9IKK0h{7W2CbF##3_F}+j=;H}}`DeoB2n7+q8Bgea
z13#%lzs>Jti^r{ySK;@!x|`Ja%DZM<OIF3Jnb89lz)e+)iVfOJG+Q_{vJRp`<abNT
z5CAJ_0zyBu9ttH@R^Zp{;r_;k+-kvbv5NguDhlE+&>Kk6O7;p`54=&ZZ=1hLF8noU
z)NVC=kHe0kUKrebf)W)gItSPaZu=`2rUdMITJXm(q#vdk!uke?a)|XO%(H@s3SvlS
zvT?1^C<@=_vtcs5?o=>5u9<~JeqNrx0Fi*EChqsKk(SNx`WAayTACXVL3~026*YB!
zLqnqX(`E{%<6i|WEgn4b7_$x?dRPpa<9hd+(CZ~l+u5}>DT~vZs-HFM^T??vtF|rV
zzkoi0oU6MfAAgE}GtWVhENuI)N%Ir)Ut2L5WfS)5r(j?!v+HN`HDx2hyZK=dAPs%p
zt2z;AcBlxV^05Z`{+$uj4zkjU$)kx}klAQ;FFin0K<HC<8-O8H(9>vGR6%v%c=l!I
zn@EJ-cfT)|w&Ttc3DRO|<;$F-7jk0!32qQ5I*0@dA*c`xFg`^q=5yfBAn;0DJklOW
zt)Qpy{_<AX1qimcvHc$xAnKV|4TQeyh0#WsV>k3Sw|0EhWd0igbWPw;41dw_hAxm^
z-m}G_s}uLAT_zV61GjbEFjO8M#2A#Pnz+FXdho^e5zsf;d0o@LJ$ODgE6T2V8rT1d
zu?A7o<hg>KjjHJK_3Qi9172v(pI<pSX`4p#sB<UyXd-9~FyVlF1M!bF-t{uqWYqE1
zt3R0^C8n6`tI_@8etzIc;J8Fkib5d@ASb-5OgY<x%=Kiet5)%wNDTA@u;X4PTRBe?
z=Mx=&wf$9#5_@GSto?pM-U!`0@{vwV;swsAL)F5#sY$&^elrzeO6Er+=(|Q7r+LK%
z1qJKR+O_B`EG$59Cg?SrCf>Avq76mkk0(+oVexXE=KaOTI9skZ-E4E7tZFtKMo6a9
z9(%soUl**9OJ&fm8a;lfT7wQKXRWwKyCi;CQ0hQjO5YPgVXB`UlPddk;EVo$060O%
zzLBGF^dI1Lk6GRMBv8F-HMKwfK=LUQh#N4#H<Rg`R8*9tWa9e`AYu4O;to9mH8F|$
z?OUl{{xkLKenTtS2OgJB<D>wWo0{*IkpIx5Y+bdU>+XAj@nc5tj~g!{#%@Ol;d6U7
z(oYD1P=xPI#-kynhX9rrX<{dgY*^#F5*JUO+i#-V4Og@Mp{LpV-iJs{Ycy;7k{2Oi
zM`1DF{<4Z=jyQzg-E#d%RLhTtvtT>eV-V@-fMnO+ee5f(whDa2?e?J4ZsB#UIxEYG
zO-;i*ksXoFNht>b3Hw_Q03pP2=$55mcu7iG+@AUzNDx9i2rNXmaz*N=)(dzZm?wlt
zX>R3~KnM{Fd;m-%d@>gfh5;W7A>I%|M6^9iQBe`&#*O3l+iwqx%!LC@)A;t=Z#nL`
z<9OzoKi>O_={}t`YZe}lr&(93neE!?a&2GO0Gv)|ce~x*ZcV_2l$6Z-nuLM<i0{|W
zs7I$)>Q`f8*!SsIc&aMMKI=@dnXb{`kmzJ-_xt}vDhl2HbeY~lEBMU5?d&nwvU0I!
z<sdZ=o~lafb`{{<uog8Y4q045#m660_WM?1?zx*O?=9q_i%($Esbf)u+Q^a@YG*+R
zJv&!w5>3+#A}9>2o>^Wb0{3>=sC}sbgi6Lq<B0Fyo6UDVNW=CWbie&3l(=}qh^4io
z16GeMd~5nloWjkelO~`V8Bl@cf5VXkLX#BDn?vSF6A+3(YLeg9Zlb|x7TshEDeyFI
z94!Uz+8W|>(>o-Q^Ft20d#h%DD+2t(VtF<2bbBgtAVCOm3-CY`r#KP#KnQVW1YNK%
z0T)L|Rxn%x6a#lhLYUL(WYVNb{P4pMkr3Aenx=8b9e0qElf$K#UK*7-Iw~3(8p0-N
zLI}S4>MLq%Ye`H@YzOT+0fvQ6Rn<dihlB_)k|JtqsagCZN#n<Zsv>}f$_ng>DJUL|
z;#srNvNP#&ylw;touMkI-LQe`#ov;2(up{F_i4tZ=&R4BQ!?Fpr>3NlFmMP+2fH78
zhQhn<r*h>w;_kei=RWzGD=!$wIj0=2_xkiBV&Ov|6vas4mS`R~(&NUPI1#DaZ(})z
zg)+W(Uk1PaJnoWG*57&$?vhd?St2lLBf6wUu(!CBX|v`txK}O{#*ESxZsnv6yLO>f
zR1tf~AY;EJKP_A1moZ|@Y8(?G!0piqo!3?GcA}^@qa;p;Mhq}`e=8w`$O9%=40l@R
zEFgsF4LsfsQ=Ja{DH4K&5GSG&u4v=$LWoh35cK@>&$DF7l1PXOiJNY^iK3#SsL#_8
zV7Bqvym>RLR;>zqyUm+7vtq@HC~DS;#;^ehAr5WNNh?;YK&vPx{e<JaG@`z|l$7qh
zs9C=rEq^n~7oLaG46`?5sP@d5h1TFA_mYdlcrXX1J(%3gRa8X5Yp+r>Yc6(=M6<`^
zxcNq&{dy@APd$#2X?CiA{E39YgHXGqflyk<!Gu5v6(JPe{ps?6s(}({iPX84G~({s
z0IAO@0efm1eII{-oiDyY{w;UX|H+4Obj?MIM)O7i5>TYZE7LxwqOP9DuQ;Drv+G9-
z_ffabX%x+xLH4Bc5H=e~y%fqXE7t0B`MFe$EdX$N0!z6yl$R2d)^5h>HbK+0Lls4t
zx1ZIpvd-1e89NsW_oBy@g@Odu0r}_&1r>k`JtQU-=!PCk8QYAUM}!a`NGY3Eer*}1
zgt3krJ<(wsKAi=+&M81&boXi~x2l33Q+R3{bfvDYj>jKwo-@bA#W7;U2y%0CNli^9
zB_)N1h6XAtE2*fcVEy{_tY05GXQRBloF|@mf~TK8_zluR2s|DSpMU;29>e?5oRv;W
z-5d4UXP<HGvB!Fsb=IVdE^Gjbq6~_*uiJD`3Qbb=<B!Dk>w~RpSFg287$>^ytFMqA
zH<_oMZe%0+9>$F!0a;#7&DUQOcf=9I4;q9SR^L2DFmhNmDEoF1yPtjrb!Q<qn~K0j
z>@|Pl-7Q5NG4e>xy67|n0<El^s^5RZU0y}fu%Xy{^+X88P*I`@Kq@HKDhD1ARY52M
z&FukaJ*wS~R2A=VTVl%^@2aVe;)tWi<iDW4`#(1P_3!k4>OqnQ4)CrJ4GW!_=&+?=
z4<F3<n!^t5$C)RN!6aHO*+UJT<kqZRgN%tIre8nrEV~O!$lq4r?Ud2j`wZar5;_S|
zqPd)i*f{)2Uq^+aD1-Lr{6dH$fYU67Uy@S(F-#UMgb*SXxFp0X8_)^Oyyh&NLI?*i
z2DlCDbI}8s1bozTZN2jx2^<|_l_kJqz#=JSbu+S((XG1oV4d7^vJm1BDP{8!_ZFBw
zeL5v2C6?popMO5L+;R&?9d#5jF`+xc?A^PUS+iy_Wy%z`ZQB-foj2cnlZPI9h=ha=
z5!xD6nB*)RGUtyubLQ~qqmSZnI1b2u;|CfPK#@`oXm-su!fPvOXrN}<uQ<l!Arzeq
z)mD~~&?N)v8>m_ME%yGsNg6!JTglDpK`RiHeD@ucm5_PPIS8A**=MqDwb4L=yS$Rk
zk3GrGJMKp8D%2ASq>AT+u`G$tra`fD^;Hv5gn|$PwM!Z)$DKghNheaXdoN|5eop;{
z^+=BkDFuoqP>^97ob;7ELLlrmlo&g%h6dcVH6S&8znj!j6MmnT!>p$#Y<9X{c?p?U
zUC7qS_fq-$YU3i6A&+;fAwhF{c=)yVscmTBk-uGoU9}mh3E|IWmb;O)HSC=6Io&S0
z&=)i)B};!>OLc=2YkDZqq|YGetSl!cHO;VscKqO7Ddj-+FYCPDV*VOBW9LHQ46GAc
z<^o4ZDc=vrT`i@omr~A`Ql1Xn8DiD8#=Tt-V(j0*SSjTf&5=0(DP^UR&3ZKOw#8~@
zCT*+n;fEgv7c2y?zWOR3ee@B>9(!!FWDX!ZJDWfM`Op0J+ix6q+;KtIDK9T)?%cU;
zQTc$umMvRYwQ5z1$ef}mUN^1jw(Uf8p`h0d>Kgs2^nm!ff&v<L?j&u@aYzZ85V&e;
zQ5|vAuFuC+u#?mY6TohCTrgszno0val8UdthO|`D#~fpfgr*1r-6~hLW)1m&xe?dL
zA7hgqgd$J`)c5E?^*LuVW6?_Py7e+tp$5_@0ZM!VX~*P|cE)Kq%PT3KF_Zc=zoWTa
zNU0}%_-Ppus=1Fs!4?;X8WW4NvJy{yEmAZtajW^c5&dJ|!n;uX-Ks)S>3Yr?<lKBU
z+wOdX%9XzvF?A;Gi!8qU?kB!j_yZT5IgY$xg99S=LKSsx0!^~}<LPug<4lB~OI}L&
zX7N%`jZjdFL*l=SnzIgDd?I=w=mX+r(~EYBb!`kG#87mrPcXdPX`KrRD@$4-aIuuK
zwlzYflurSlTMSPyGE`fO(=C?&2;3&6)GWp=VYq<*0$BMo&uFPk6ScLqEL^xS=-Aa)
zU(NN`U*9UZ$;rt~pFZ6(uz&XK*{za$pke0BnU=+btzZ@zo;h=7i!FPdXdM*vH^;1K
zP9lvzNT^u09DA28IQsPVx)sgBvgJQ4MtY#j2`3ot*3dQ6w9;a#mMtgYs65mz>5c2f
zHkn_d*Q?w2#TRV7@fMW)^?)>{(rB<d$hhS%Jp0y{+<xQvq$b5Rk>w?KLLg#dNgg$d
z^fOM!Ra-;Rr=L*2b~RFS>%BpxKDTE4X-ry73zeDmDuh7TY{Vp|Al+{2OZMs3KO~yh
zUg=NrT24aDd;#5?tB^DC6mlnD!?wF0qH@*mUTWpDj<y}8<rUoh-0Sqn$>6T*E=LhQ
z658A)ZrTma-%t6^KcU9NlQ8VCfVFBH>iPcX-|-Q8i<bbQyR==kRoD}fz2~6gb#M)e
zuIR}T`vaZUxmLg^$#Js9@S9S~iZ)L1l*RHy;Dpu)6hia_hFT2&yESCagp|^WUScH}
zjub+)s(SDI`SYo-uMawQ>#eu8Mx+<Hxw%|^`Q<^!=FFLc$8&JESoU~4%$PBw)!WaW
zJsYRf89|LYVFiVDcu07k0-8q6vZcfwHUtqD4}hzomY9S@q}xU94@*#ccEjGMkFi0B
z$GwDoi@ztf!AZ*TV;j}$Xga?DsI6nizh9*2;m5I88nL@V(aY4K5_i^_e6g*ZVMh+(
z=pzO;vGp~X!7$m0SO-a?^2j`A63&V$_I>z2oZGkPra%dbG-*aSq(LVbwuB@cNnNNb
zF2Yq)sh4#zWQazz%NPF@<{VPjMiUZ*LiWig)9u#3viY8WQoVkIF;`?PbskOQ{{J_X
z;)-gf+<rYdUD8@borDDE_MMb2Sxok%bG*dN#ILK?vAeh=WJa_^5~J$HgFSWCsBs5e
zI(;mV-QJT5AqD~GTMWM<rQFw^`W#>w4WwI)z1LRZQpzR3T8rV<TD+{bs{~sAWcT5H
zAQ<}+z{J)FoHJ)m(B&gXj%3J?wwKSl_S$QME-xx7V(HSQZI<5=p&KP6+qZ9L<;s<<
zcA2hOvxW^DHbhvvP6UI39zYvq=#A=SZ5?%M){rpbNZlF*xSUQ@M?CK0A{z3y5SKR!
zB_@_o_8~L7wfeh7xZ>kT9x<|64O}H<<llQgweS5Go7;)*I~88Ht}C}YvFDz}{6#Cc
z;i^+nQNT-}8W;EQeHTJ{&0ZolmgHkblX3QW)NI&5(MKQRF5IoRq4MJyp)6klJ;qB4
zz-}WxGY75SMctlVNX=uWa#$UeupxzCB|Ra?IQb;HUi~LF-*F$cTel&k@FPd9wO&a+
zm@$`6=l{et=by^SV~z@NwT7*?1W!c;JEpx)udA;`#Ks1m4=I`X^$$o@@w&wu>uk>A
zh;!)Hxx~~1HKt>K^dGvH;8}__SX+oXp~I?FexZ~yKSFaXuvmUlE9F^j_mh;ep_MYd
zkWy9wzqD}fkoe}CZ-OpA-Y9+5Rtz6LoGx90XTw{~eK5c*CR|@%-zJx-+1s@f(S@RE
zTDSI?NFb=+z7?&inuK9~ea1Z+LRBeWz5>nTCh6EQIynj<OQHwH*RA>;b<jX!(yd;A
z^*0RXu0l3Uz8(49VuX-LA&g?bMq+}rlXm5$yfO1fF23{>l9Lkf`Ffj4&lY~9&v2Wn
zadD(gIEB;`Po!eW5-R7;!BbnUzwrQq&?b5__tTUY+a(=)Y8q8rH{q$PM#Cx|AzZ7P
zo+&D%pQBMx$UXNgvMxA>t#|wbXYoEi;@R>GSx8vECZ9)Me3v7K_Tld9|I|7%heX!Z
zvgPe}=yB1-ddV1kMVTewbb9z=;Sw)F4<=hkKuTh53R3f+I-0uE4+MBTo_2_L^+5>H
z7wZhoPU~Fru<DT=$zo|%vIYtvTGLF^O6N#c>s@TIyu~f<N=r-Gym@obu}HP)5<)O$
z%$T5K%a<>Y$Q&IFQc64?51)PZS)0h5l+v3a+llBxQ4}R7+LM@p=vAxM;K<G)u6s8<
zC|!bLxWHV?mqB6@$%h`&yeyz*<0h!6B;lyMMh|jxJ)O+eZrj4fn{Pv`Sz|2MWAn_~
zYpRsg_a8v*(8JiWb2n$4I4Z!^+FXi-kipjw2x2m_NI&;HZ2kJOciKnPE?;KE*M-ii
zRQzW(E&_X80!h94QoCy>4SV;5(8;Gk^DbB!xT>Pk?XrtVJpLFq-|-Nx%E~5MhU?fZ
zDlOy2e>_EOQVcIY^mpQ7Vw*XZ!L^eOF1Eh;E;-}I5!<V0BTK0ODOs>&8GFmB(2YpJ
zkE<Dn7aJQxY+`bY^yo-XRkgicu6GkW5u=vQ*tu|EMNoc=;2c(rLIH-iM34|72ff~1
zFsy8etY+}5#qz#JnU}D!eED)q!ZCb!i*`RGjy&?npv!;#)oNEF3%A>i+wJzcLfhk0
zP*4!i)3!AbLa=u2TJrPr+f{c>)9`pa-nM0b1lTl9%WfCuEG6};R}(v+KO#03DL`{M
zQEYaE$4$-36~y%JjV&W9pwMkF2nqFTe@D1Hqz*rvpleI0-mr<Sx7~%ka2KEi`aKP}
zghI}h7xT)eUvkAIrx0TlyA7mV`v1bNm8R%xD^dU|2?Gxyp>Kc67cHXl-S^0xI1x1~
zyP1v_e!F6<CsaF0{RUFIa|cx$*OSzLASkM#p%qCC&!y?rMN#Q>)Aek4_zAW?_!#{k
zdl1FX#pw^L8SComxaR)HD6OvK-De)6cWzFQ?ZeQ^>1O+X{zK+5c_bcwn0G6m-B5y*
zy!Y9cMnUP|=g5DVBQ{2ln>;X*Aeunaw5;|j&S?b4Utux)ij;D1yC`yiLI}|f>x8q7
z5u9o}dV$|y#QQMd`xeLw#ejmF!CD6^Spq#F!^(o!1S?kt=b3i9ot`~=wp=p{^zR?M
z-T96kJ18$NCpEQ2JTxqr<vuh`qq4FRx7*!rmM)LS!?bDBsH>}MlN^&wz4zXGOrAU$
zMNzyhFrx`6C90~DmX`L1DVt&wLS(edOqzykZ9a*U&cvi29+#6CM*^O`g}8U^A?bp#
zpoY<>SxV~iH{w<lQicSv=t)V<#w~1_{CDj83IRpWOwa?y@aAx8P~W`|3?0Iz89#IH
zNS!nUBPuQ7_d1a?S(x#cEr!(bC*dmENAZ{Qi0jddq%orrX14@m)?h@#EHx~J1QZgw
z_rh6PLh-5<bUE}el$aR5BnHul+=b~jHJFkFglc2JJ-3p7>%DA${uO%P{x`2xvpFi{
zY-r%NCtqU4y7fH!kK4&R%xps9uT}62R9|X`L^inC@y<J>A8`~(BM!$uqR(G}?K=xu
zv}_d!RnLp@76ELkr9coD8yhx}?x3)WFNg;2B6z}7J@Eh9QI7)^p<1;?jL=CbokEBm
z=v^5aqhF9^p<v}A(TFUe+eqO1nLql|0zzli)uC_SzSwM$tMX$-=s<6CHDW}Iy~WMK
zBq5cRm5dxYlFG_TFTraMLR4MdY?;BZplKRUKm9aQr%sKCyihPpqzEBMN=jnk!i5nf
z_8l2Epk4YN$*KzKcNdZ|Xi#8=fe=(}$j9b#lW>@sjbwG5qt|)UJk)R6L`-%rB0kZK
zjK_4r)ot6!roaE4n7w;Yga}&bS0eBT$e45*pD$R-DHD#t5o?E#d6I2_p+ih&7Ficx
zO4TnvQTX;dWSn_Awwzo9ni1D$y|I@{2|-MkbkbrSlq~&))ZxRi#U~nxCXLfJ{J%F(
z>dcZRiX)zWkKV`HYyU>kC$q>p|EvIN=!=K*$@F+M9((OQX3k&C{Wo64r18f$*4$DD
zU8Y1x@YL0_?d`Y8JZdyaBM<j(9hmE&6Z8*fFQB%;g-}$z=x~sP2vbcVAR#U`U_TEu
z2qD^q&>60`t1O0Jl2Yz&4?PZWgzD_EDZ+DYAUL73W+B00rE`IlvTddsKS(KSgb;fO
zUIVp7B-Dz~4HyvFdietf4zwJL7NIk(TZxH@Tyn`JJp1gk)Ya8_mmerrCn+Uvx0}+^
z(x}g69utSd!If8DNmf==tFm<{DCj)`qmG&_+t8$&gaP`ss^`5bNC|c8@{u+>@dF3x
z5ARUiLrAomT4YfX_I`c5EyK)>oP|Z?-*qSUox2c<F3Ud{U+5xGG(mk_A{i&2$owV0
zaoS0FP3TX=FBKv5gbbUF<YUH=dH#8neleG-#or@69&fjXaAnyN6393zkJ4o;aF!I)
zxY<?H(j#H)mGu1+lb*@o=bm8ao9|Hh%d&uN&H8XAG)?2*?@njxr?a`^$_u#ef^)o<
z#c*td!XRkwswy^5eUqGVCpERqH9}2oE&rXlK%c`j!214bBwwE+B_Xa^T{<}0tMTle
z=tajHqce6cd{~jGy%DHC*BDk7zhERU6s&ZPGyw&jga$*42%S|zM5GB_mo8mMw@AR-
zvSmv|=Q80<v{P07`Okmy@y8$2w{KtX@=i2`qA2v}(S!Hidygqori9fQ<Y0k<-b6XN
zkl@<B4W&yeN^&Y@_NlPxLF{tl2HdI1IJ$P#Ej3bvDD5F|S5~7{Rb%hb&1apH5@&f8
z`FG!oeajZVAG;r&@LKXDXc{rYhqJc0l74-=k=Z2~L2wJ143N`$cq^QsU#Rx9G_tPz
z6PnAzzIWckQ&#Fnl){MF69UC%Cwt5>RIXi5-KNcG(hUFi5P}tl7K@KlJ9_q{-#_nU
z^8-)Puzg2RiJGSI+VmMb^Pdm7`O<T_`N|88V$nV~t97{*9|v>oEn@2%Z`14CbBI4=
zpw~6(OAztjNaKrdeq`I;Vg#TG4Nd52@xD;WM*Arx$%*k%VCwCPlxR;NgvbJ{O6$KS
zr7VoL4hIP`EtW@v99a>%mL}0y=^V+Jzvg6yt+S?S6c-2Ym=<Y52i9)a!ouiY5<&=K
zVq%y$aUu&BE@aZANnUrds;Yl@a;GQ?ilSh%*_be40^fY|4QHQyHZd_Vf2iYQc8Z7!
z{12gXX94kjdKtm%0*~8;%^riIX}C9U!_l)37{T|t*UW;9NN|^zAzT`2SHEN|mz%Bs
zdIGU#wKs~)KUm8Dn~>lZDqT(*$J_-!GvS0hFPRJA3<0Dn3N{@fP*fYK<4z!T{5T5#
z`(NtU{BC6F2=uU(=4A-gM%I|o)D-Qba?KiJE0atGQmU5hi@9a$F-MVg!9+Iy;}Nvl
zx<E-D4^uyx&Eqe>$K=b;<Bn@C!=|X-`9sa=rL9s@o4<)2AAU-oD=sItM|Z<6=f~(o
z;Fd18Wa|5$f=wWWf~w)mIJUZWC8Q)D@Qt+FqZ;yUKzxwp^>kY2S^_Kg_^v3-WtIA#
zE`(@FAeoiU(QvU_$qd^^R#sLftU5Jx?Hc(CKUOmL?Q6g4JzcwY<%17C;Mr%NB{@0y
z50k+e3Yw-78ym~R4?oN&pL{~^-o5`YB+egZFW2ZoN?dyiv31M!ku0MQoTsW1SyF=9
zr#I3shRoVN2kw#*gpk-WjC+qJ+4k1kI2SBHK`~lO`Nv!tHa4Rc84{d{2_%mg&hl0H
zj2>|aKEe@<{6y!1(af572x5EnCij|as9g0s<uhj^T~2}q{{@2p6qT&JJY02klrR5P
zZ$jV1VcEKU(S)Ge<(Faak;9G`Uk1(NO$&eSy^neH#do>!igUU3>Pt}#!XD^m4L+wp
zB0Q4PAC^$E<R|)GdnL9mX~q@`vI&vHIDe7_KQ86h&0A5Via-z_ucE5@K1-?i_SVn5
zC#4`IIVp(H0}4@Pq(X=;zzr6|uXM)FwFp+^C>nu8Rx%Z!C5dlVF8z|0sS_5;ES85&
z=t@dTEH7p&_h?&SRb03|37x8{#Ky+*m%sdl1q&8%*kOlx$;%&+#0Vi6G-wd>=g;TP
zJMSbmHWoFU<jR8(3VJ)Ch$A4HOG;e33W>?>idi&RC<+Kc!@eRswRObz?TZNZ01pCB
zMI{~t_LLNmP%{5(D&Kww8`38us2sf)swTiTs4rWqYU$N8m$+EFG3rk}qW{(-qyQoH
zGA%-&I^xN^@Ivfane2J}P245LVZMDCfr5%4bL2>b+fCWhpV6e&XrEefu?ypv+U)eb
z_b#eeE~j|bESzo^PrvdWPrdyqcV2ffw_kG^HpT3W>JMr8Kds)$&W}FDQ&mUzD=$Mh
zVvRFo##s8TZ4f!o)9IgkbsE@oi>@l*-kYyt#3B7c+B9u;h1BHM)LHFNh|2Zqhg*PP
z2dR$Et`IGSRXB?kk!cG;k;Jg#E`<;=Sij&$LuV_QVaJJ<mzP`K)Y8(@TC7lWu##C)
z5)Hm<4KuE8*sx(NSg?R=ue}zB!|{i4FPbrgHk*x0FTIpSix!cWmj~$XS0M!LRo<y1
zK>_T|uGLn!H9V!|C^_AHRx=~b(6hG?&E>(~HJ7Gtx?qra>g&;LD)xj#>bGxa*VF&T
z?)Dhl`(n5B)%{6<?&9-EiQ|Y7EM2vMqmCTfXjU_PKKjrTH2jMQgCZ0Xk3O1=NoTQd
z+BE9cuJMbD3r8ZPjXDZdu~E8osZI#J+qZJBbjFDqpGfcf?<fD&*Lm=s$9Zkee4e}e
zdTzY*JXGT%YV5LZs-Ye~*-*HbZSTBK>d?c;I%NWiYIepjT)p1E@RB)AH#@T6hovlC
zzZFHZ0TPeUc;vaaS^E1HV_WH+%V#MRBqqcWYmZzRgZ+f4MfeIKQh>i%48JC&45y-X
zCw#CvSCOd&5x&fDUjjn?w%Su?n-C%_6HBP3?}>?#k4~_Xxj&oFCnqQK!V53(=9_Pl
znVEUeyI@Vinvs#g%P+so8*jWpMn;Br>|o#jP*4coqp{79ejly223c7}e2z}qz-YZC
zz`dssx64Ib*PO85EToiZNEAh|dCKEBib`}M;I+&JAuq;%g#=9~qz)U(udDMJewdl1
z-$CAL46+-W+m-CAuA=&vWmL_chvxBEW^e{>FF=9RBacMYBxS$+toM>7FxxlqTL^Lf
z0Ai7U001BWNkl<ZZeIbm-o5##>i~XU^fT{0a3>d>HVzfPkVlh)StUS7N%=3!D4aQ)
z9_OD+!jM7c1_AT<$HGYj)msY4f9Nq9+)kc*<6~Tk4Jo8i#92`1GF{Ai$wA+QhVc?f
zYEt6CV2Y3C4%|fWtoJ(L$xs^#AyR}8eS>`3D=W7n!s=z&6@|G%$qe@;Ak^DVd+Qt)
z33a(#mbbA;)}S+Cw+BC^`-{-oY&LH)zy%jvz&GD~!-*%J=(W_DR;q)+YG+#Igb?KA
z<uPyGJg&Ls8tis^z}UgCudRizp=p#L0C#mYo(3neIeM2diE(lyxQdH$A&AKeH?c|2
z?6gZk>ATZWR{rj7{U!Y4<$~^(KtsV9V<)~}AIi$A=$UJk)V2R$CUn>mlF0beWoU7B
zO5S=0t**uh0SuiiVP?^G$;(4?d8k;v0?k{FL^fGB6#RMGp2B_n-}QI1p}v+o#~(w^
z{BMv(JhBx2BwqS6NPS#NwCXx`fAAsB;xc+(b0zllG(U@Nu*r3coDdM#Et_r^oy+|1
zmhtoIt=<;`;e9#KS)hO1T!4F55oD$%88LnbG@^373Lz5E-L1h8j&<&8^un`^KX*n<
z$pa5oM5?0_x=`Ml_SQKp&ymYzEg~A_DmogWgFv!2XwV=&`|LCBzyE#`67+CeMNtkK
zt5s}lER!csX7=pa3>`WY;15sY{Gkw5DrL*ze|X(>wV+5uQi2~r(ARL6l_BC{5pi)r
zwhRu{;UGp;X;}CzHq-tjtlXqcBf9L_In+o6ha-mgxM+tYwE+SY#YXCh<B2<F6#L$M
z2iKlGNNJjtgJh1HrZQu<Ge;joLsbQptJmQ1(yhQ$h0vjk5~RlJ4O_VAhJTQm(S`S4
zdXkgwnM}#bm6R-8Vz{+^pTKL&=&8GY6WiZ?pY&r!lXcq3#+#$G&igQKK<56a4hJ!P
z`||YEkI+;FU)lqW%E1IVh{oP!rKh5ZaNPPG8d14k(Gwwpz1?(XwzeKtq{JPCxehi$
z7p{fQa_o;_p)(<+^jhN*5)!!YzWey>v(FefaNt4H_Nr&ko=lrIjmI8)jMUUrZ>(G=
zd{BVkIp1wWK+b4GJ!)(W#Kd@Gq)fOgDzGOcf*Rh%Cnc!KNl3dL6^}1ypKx7>5ZALO
zdrK?H%1XDWJlAm`jr~g;av167pUvKB)2YwTw~UC>`Hw+j1V|ys9eW&g1>33Lw%u#L
z^vd_&HkdqM&Y~Z=_~r*W>9{;ze)2xjQ&X{}Br^D(+u8E;i_{mF1uu#sCDQ4laOPYp
z*R7-Hm6sFSw+A56%sO?!cEay%(wmu#<ipwXS-EMOk^L$SBIi#go9gb5+{`X|j9ri^
z4<JOG3<x3OfZHvGr%EZKQD)<hjA%TCLdgvGHWx}phtxTg<w?^b|L$WoUo_@w2}~=M
ze?7MG<Hz&OH{Wo<1s8aiA4nl5CJ|K>g;P#Bg#`;1aORn3dfm(i(!R786d?Rp3#s90
zXh3>2T4KS~YHCs9;*CD7epF9_tg1pyi~}{idyYU-SFj6vVuIJ@27#6t-SaBFNf6SX
zT_T9-n#+zo`{>a<@^y+L4Iu<l(J$7RoNjcx{?8OGSV+xJOOcwU*R1od9fZ^fKvBs#
z<y4A)T#j>ZQImD7sF26w;iY#!<<7@n<e#@(!F{(}L7dH(5u7+|2$>VdQSkJONVg|Y
zpeCu?v6F&#KOp&#!Q@Oj3pL*1i*xW!C4Hn$>KAdz`t`VKY6D&hib~3v^71r{hOIs|
z{gn5&G&bHR0d&vF^4q%u7SS*TTu1O?uJtq%JJ*Sjup-~VYW%eop=5^pn+f%@&>?lU
zvX}vFZYm!PwY9Y^)?<G`N=a5$7H_}(HqSr*yw?SLAR#_Jo<|;egpWV|_zzoKt|d^=
zZ`#d50{g!Ik!5F{!=|U7!s8dO=0os<E$eC#@o@+>u(i9860NQtTYOwFUs+IiD#|H(
z?_F&14y0Xo+nKRbrUnL=3)D|~9fHK<WV5fhirmaBEKEdU>C!Hcek4VJl8{J`Yp<bZ
z%O=V{pM}Tm0jV{Lt81jc5TGjbJm)luzFCO7qB7_n`V+&-s%mb3?0G(%_bt=@{Rsaz
zaXhM`1{`Y?g<jWOhPz+~MPDs23~3UN)5YE|zoG2MU+8+-CBzTt8{+huy><;No0OC-
z{}pFxnO7&NNgjRv9f~TdjNP%e#+mfiw~FyagKjw)VgJrLFe3Vt5<<iRf43O!w9d5+
zR{p?f)cy?hv<}x@8fv|05V|(&9EwO;k*~JkKq7QzSr3Q9!FAVN$D&1x7(RS>hmAS&
zPvTG%g+mTGgt>F)a`)YLlaP?mne-3_6!hMV%|b{?^>3@Nmy}VobR~)=gIMW;KWJq`
z1JdEZXGJ$`DN;i^UC6k&@J^DH?0)}$#Oy0Z*i`IEiI#JEGdhjTP;Vg%0cv^%B^7ma
zNey;C=|E)8gy}{VdJR3r7DM($7a<&WN~Zr0&F$7ZOUNc0c_D;AIO6Df+Q}5on~&yl
z1-eOvZ@Yp$d%66!2dJ!b@!<;(F>FwOg9kN@wiEFQbi3n5DqeXV*PbG3H*8?f2k(>6
zy*phmI1ig6)~rxtv15kw*fhE+3jHs+g!mqM7Z4#O3w~O`v@gE}BN4*X#H&NV|7<vH
zf^kO;LNnwFsUgtl)h#Q?kvZT{6JdE*1Kolw@3hXf9af~Iy{%mxmC&_WXS>~Qc{%^l
z37tO_MPb;mVSM$~SKN5xjoy}69SLU1k(ihmF2DS8zWw%FjydL-P6F2gD1=#GNFwNV
z=>>GT;jbL}$b$$)H1>E|Z|ibHY)qhW#LPr=yAfg329rSDmTgph{4t<+DX`^qM;JBA
zM5A>Ze=~`LM8lq(NNs&B@$oVE{B|AvF|TVc3{omomDE#C!_mJVdtZGWX}09@`Y9V9
z5eY(~rlyjaH<G>c780<5VO{(2*Wa1+*Sk33xTAUH!CT2r?-C$K*8w924x|D}hYu&|
z#IfwW`EOLOUPsppFCt;+Apxwb=_M(B(@O%mqT+iCE3c~N-Y2Hwa^tgBnj4y`8-!7j
z0#px)3GtkC{ITe<diqT9F?O;u&C2ly^bMjaQi6mKcC52dU+=WewHcj+E;J%R=q$%N
z37rpmJ4s1ON(xUu{WKqZ@BukS)rpRTjEoFkdg&#mPMu0-X67HVIB;8_2q7xMSy2dx
z?UGLKtFIvS5WRMo3^!v?lL)^byw4aU5t>GGH6$ex;nvvm)?3(WYLG(UF0CYPa4+13
zw8o`#g9YhJ^b8PNTpXI)g-!1j<!9mP*rZO8ikJ6D^mCG&cQk3^kEh^`skqC^(9qK$
z&Fs*|dn5%odUYqEM-F8_F2(Hi>UO)C`oYKi{jq2HfA>x1mdnp0*4FeUWs*5Xw|}9y
zCAAwj;qKEDw`#}UB^?px2&f|$6cQSWiYQ;Zrjc}?-42h&{m)Hh<IX*Lc@o7FK<dn_
zR?`}rot{K;jE&-odhUDj4Uh_o)L%x@QWA}}Uk5zOBPi=~pm&hv4Rl)PA_1$H<@P2=
z;T8pMpkt8GdMRaC$$CjiNgcvM7mXsuk%XB<U?xmVnly<83l?zl$tMTYs%vj+n13S2
zs8OT%>Z`A~;)*K*YRh#Zd?-@NRy`m>m%3=WvhX+WHZmTC3hFnRLG`+|)P4IMieZgw
zC@aO0l8l<!g{JhasT_Y&Cv2)7Xov7O`5aIXLMLYef#Z-N<eq&td)|K+ci~>Iwaqkq
z1#D;hm3;V6JauK%Y}rgjRSh@a`xrm1TE|B(KgrmUhoktpH~k^eU6^PRt*o3Kvu06O
zT1K}s&ZPUjcd_&3*YQ+VH<~b*6sXM4r*ipHLtX#aN_~|X3l}qO#v+7J4KgQCZ3^9U
z(+ttl*jA_V6UL5Y$bjBxxB$gCM!HS0dsZ3=jt*R+CQL{vBhm-dsKkGd#qjG=%AJvz
z=MN66eo)bfP7g)6!fM+k^%PF4b=y<tuz%y}>FMoC=xS<$M?6^dc<qQVNmAdwefi{*
zPk88|hlr1l-!J6Jbg?$BE!T<GL6K5MzX_chqJFU=z)1P_U73y6^=Rz<?|-p1)B^&k
zfFdQ8J9d*W>L@&>N3qfN!D9RbfaXC}?Cj@E^ESiRw^k6-t2en9U&_8|AK)t3iRL{T
z!H>HT3K_?pz;8b+<*zr~$FLEH^Xj8_k&~GoG;zS^8I=f^hkf6F&&~x4$sT(gT}F>W
zsDk*z4k2aOPzv9l)+qb7$?PBmnPZNl$7NR{gigNv$^EA71>E(o*U%I@2u+WDM39l5
z#GQY=5(R-r_{H7<?|!y`AAZ}2kV;^pjDWtqdmJopd&Lowb1~3A$npl@i3rW{M~BtD
zHSFf|Er6B(OrvFeSld%)t7`IL|Hd;iGNjdYy?5{4*60)xRx-`jQwIY4OPn|y4(__^
zF2;`^zu)vZ>ZqfbGGz*BX=$D1NZJSmXj;c2!Vq>lQU+m9iU6Ap+^$9s|G@0j`fXdO
z`ThsYwpIc?cB^XnD!Pn63C-aM+{WzX+2lIcovWVu2JAK)PNOL0foGosq(n30>Ci7y
zdqyT*uf3kKnKNnFv_&sK9OTGK37Y2Si+SJj-?@wV%dw-m?wm8Q$CxpNe(M_+Oo0SV
zqh|9qwtx6BaXH=Sb?!OXyQKOY)`~*z^;c2y#eC{EZ^h(W5}G7ofZ;Ax1ZGKSspoxA
zT2;vn_dP>VWeq~n>+mU3K?+IHzA|om@L4n|^_aOv`=)pEl}KSyT7i&~!F{?0PiQ{y
z5Gg`;hsE*_I<0dNgO#PJdlcphme>MNCWGF5Ezu!#4p)H9W+OQ{Iq2A~U6Bs~wUU{Z
zmKO23q79SSx!rD-En9|Wl)-9mtX;bnr_<>z6nv1pc<&b|ilRjRHO3DT5RO=crZtUk
z69VjZM16xLtCAY*{p3?(8|v{-!w`U0T1;I@35jEm*NKa9+p_#k3cXOFv!;&3gg9#I
z+<wvyGzb*gsOpLUB{7lAtFNMB;R0$_tu{!VnSvk@c&MqV;;F}8VZ++3JT>(xjym;t
z_AS=q=}cFyuXwK>OzSGGVE61fRPWqHj|<KxX+VFU$J(#5vn?xw?ti_UUC+LPTWV+$
zw5n>BUwIww($Jn10n|I4+&$$5)@<7eip@(_%?xEJ6g--5S-_bSz5^o6pa20#jsAUl
z`v-5f{;=<7D2lQ#LZYlJI=dqzzY|bkvAmmz{8P({a9QbKg&i%F-xl;}p$@@?^aw6y
z=g!E!mv}rLcJB^;dGzerGoo{~2c*=K3l}e5%)Wj5_FDo%Nl6Jm{P4qmnfqXajdr+!
zs|g7REl6ye5U36Zo|0m{Ds*G(5E8ATlB%!2!D~D6ayo&Eq;%#?awbn^+Yd{y)#|ZZ
zrp?Feax>=AvtcFfsw&dclPNdqngtCuI{+b#j7}vkj_f~OPVs;Lhp4V0ZcH930!?ac
zT(_0CUVEQ&|L-i08gr!Xnv9Rbws{9N8+MQ|sHdJeY|3<N?ET><+%>giAAdYHGqFHG
zz!?)>hc;bKJ(c2F^Qip(CsL0)1`+R|-{c!mQ&NKKt#6Lo<L3TnUZ=9o$!RB!=F@pg
z@J4Hj#wV$1(ud-;NJ>2>&oB|j*cE#B$l;(k$<SO+)wbeNA;i(G*v{&ScA*ergvHp;
zQp)>VG1q~DO%}`J(8)}Th)XjVXt8{A3*}g?KTrrEWQ0sOLI{Q6E`P8-Zb8wnUq62T
z{r4uz^Yimtp_w(dZry6xYOG(sei5Cky`ZWpGiJ=d<B3YcsIc+rr=K!m!i3JO);7Wh
zM7=bDl9-4Zr1z*40wpO4XH^YC^BC&*yITcCUww_eqS9-f^9Iiw_Z8Qw-*N6OBIWFJ
zshajKNS_POSTW#VDJ5EYDcOU2v3^UT9=9b9jz_&PQePz?VjX1u=`zYbcn42)3GpYM
z%#7)uuzAyV?s@PwQqq#WB~Sn|jvK+w8FPs1nu#qj4m6GObsMSNu$hd!k;HY)vN%Q(
zAq_c3!iv^VPwwAtWY?4@Nj~Zb)K~{;qeeC%>|VXmTjxCU`ZWGC<7=-x-PJ1e_GkX?
zW1pecPYXy)h^4-v4yW5^%N0OMVmv*&<p$gAKt!|&T_58UDfA|ICN1dkdIvK?k*Og>
zh|n3<zC$gBH?~lmmCh;XF49Puy?SGvXwhOq7Yu9GtO>fjwFo7ze}5}(+ObhzU(dXG
z^Y+{PrlETF?Ag5V!VAR2gxhww6TzS~G&Ho-B;WE$;^I-zGb0=M$WYVMa97tN-JYf~
zR!%pib7!NP);r;~*BC@ZplTk9U!O|Gxs!1B>Vs%pV7IYnLx8)mklgGp?Al{`n-7A=
zJb*!Eh3PI;ZKPdt6(xnGym`}oq@?9=*8_KulIF9rfqqd*yGqW9$FlGHpJ~`r$nGz{
zLGx&IKW8HG7GzF%ZF17Eu<E5o*57m!NR7B*hf_Rbj`v&zg9J29^7hBGc;@ZT{G{q1
z%{bWsv4H;JlHO$+_up~_d50f@$E=YjBvQbT0ey+J+YXkq9QEt96A?o`*1fW=2v}>-
z4X~=ov#y1*)`l31q|Sp|sY4SC88Rg3^0Klr_Uwt=`_KCI>w_-0+wBY(Fd$;{v=^k5
zEL*mWf`a{bOEC?;g@uJIUc5M9StnW!B{nv;m<SdtZ43dZadDsu&^(Q}lfa&tiFCP;
z)m8XEwk1%rX(QSO<At$_%gvh&fVOcHrE})c{nk5hIqcv^2~78_|2D3|Lef)`DA`+d
zko|b{<L5C*k%otDTQ;+(xRP-v<#9}}95mEMYhzf}1R^dTXW3o~XU!w)gt4Rz8;sIi
zsE^c*rFtH-Kqw0R9-Tts;6dbGc@ZU_e2%BIJP4hW-i`k^`%C`u?E6LtqzGusEdst-
zS@?AQAAIf|zW;TD;WE|`QlJ1w96InI==cRd1(B^o*NG@XN?C(7p^9XEhQmUPEpMT$
z->|Ml*B(0811&PGjTm9&(I==YwzaU5HDt&T91cgs=4meoA(%OHCT@4A7QWGkS@d|u
zj2UPLdg+P%472rDR1+g?aq$qV1~za9Kw{6x#3sO5w9gyhuh%-0lzh7gn>S`5ka+oG
zw**2FRKD{TN}_|bE3QUURT>qz^NYBW)b7|pa)JY=N1?Xv;PLuMp_g;9Ne^glzMcOA
z>o;uYvYW4Gz@$mISN~4!$~9=xuSSRjG)eh}jTA0jN%m={64SpgLQydj5rWQz^!Chc
zWOT*Ry&H<%Mr?LAY2(MS_q`9%WZ>3jU6Ie`eaGJ)dz}VN1B_mM-u%>H2u2Nc|7>k-
zy*Cz9GwbL{jZwn}wVO*_dOty8M%xG>L;;~2sCS~{VVTA9(Gi^ESc~P=Kyb5?1*DV>
zz-o)(V_G7s8CdE3o0Kx#&K3X+9Xgcw_~0e&A{lJ|{rBGoT|RQ;$cW9+o@i)j;EON5
z*xzb+3Ek}3vn^v3I}sif^f#hN2In?~!-1NRh*lHWS6l)yx!KrbRO<E=`aS+6T7!$~
z@4iRDKd;pv0x5h>Kc&IR?)(2q`st_Qn0Th%no3GQ!CPWifTBVD-XfIRYLZiv*i%$`
z@OWIM9&1)yRL-nVzM^0MzDzuGJhm7cLRColzezaPuBB$>YH#n?`r-<TmaW2`oJ!ZR
zqlimMCj00-iho+FlOt(yc%)$ayYEr*-S@~w<}dI66hYR-=TW(MF%1Q}>%)|#X&Rr+
z`<Ayqn@fXR!^a%0{x#<fvHxb<2y+uDQH93H;X|zDMG(?(Jr{4>WP}hzb}lL9Rw75}
zD$vVAMCucGar;moVw-R~7_=BdhzudbYt4Kfh}Qgm_|9Va03k%AYP*fKSiU6Ojth-o
zrSs@WWak=1t*r=caMlz8s;V-2bZ~KzB6QogZ)f}V?Lo(mJMOrM%+X$uQtu43YSk*X
zY}vBk%Y>MvN_Ott$<n1u4>GS=`$7SbQf`RmBvMhZXJpb)QPw29P07kaiM8R{xyLZt
z=vfL4+qUA`z75lA<F}QcU?Z+hC?IUucI;r+Bmbi74cDWdI032X4oIZXYp+Q?uFB?i
zQl7t&LkDzY<?4+G4WH9A$&&AvuwwZNCZ0Wk9=*HzqTB?as-#_bE|nWMQ2Fa>O4qHW
zp{kVhk;6#HNk<X-@loRJ#P{w-<@!xXvj$xN4C_<96U}RN4enG`latB0<U;mNebZ}B
zcY8ei`^^uz{n3{>{iGATUUA{gW`)K@<3xbL%Q_y<)Xmg+=)m5jCM7gkD_W@Dxd+89
z_0A{}zA;7eI+aoml2Uew)CaiFVyb_K*zTf8?N1OS1J^b4`M)U6Z{iEQ1(gCIfD79y
zTnNz*YjQNZm13>7>xwmbYzfW;tXeC~YK_3tPd`2A@&yYPL?n*xop;^|x?ELNPCM<i
zHp_2MNGVYig_$#Fb|{OO$K&Di&p+SaY8=ooqm^13I2#uH-J-UPESzQKO=6ksS(y-@
zK<$<tNHb?bO3Ifn!{)a7u?bVGSH3_Z6$Q^vKT+`F3-r3{@2JO*$0JmIOc>TCsn>6-
zSewt`hxTRp+N~h9gM!o*7Zs76o5P7Gk3)^I`PF&R-IzjvP*vjk^kUbWui?nfA*FY3
z6ve;5i-IN;5_|L@CMCg-{{$Ai69{@=|7WsJ9M?FqG4Q*9l+#Y8VRJq;zpX_|ji;wh
z<MF8<Qtfhc|5LAmqIi=lkR}x~-L8I_txb4bXs4_q1bM@UG#>{TA;<<NrTH638{97{
z1u~_SQMpb#5j{vL_W?^ShOaZqceDjJ6A*P8fY$oczC*X*2E$)kDbpV}Su8IDT4ho@
z<BT(cE_b`#y!F;wtr6)(Lqh|vzy5mAvAn#z1FB1e38&M^tXVpDjzrv=|NX8l_q|!O
zW_2i%d7wc7P!uJy`Q1Sv(lZbhO}ok{i76=QS-7`v^b#v6p=#L*y>Kr00+Rv+`dnxd
z!vbX#n|05gL%|Er)9e0wNjhf|PMd<FVKl54C{j?fY9)vC@4<!*+whpdU<Wudv$N@y
zlg;Ya|Kmr5B>n|=kx==?9Fj-nvFq#kG?Z2n5GW_4p5UM;#OLN9rNmiQg4W<{oX-;Y
zZzFiUo|;4om8^>{X6GwYdH9*Pc=qkj@CZeBiz_zo+OkQ^p70yf3As=NjyOHOFK}uF
zyCPA%Hc5fdJdDjdvR&fmq$Vj?{4?G1M(KTUqX1P^qkf%sA}X-5cnt*}X{}Hp#0kI+
z7Q??AF8UV3h++H6V)#NKM5`^ULWsWrR(+=CNhw<uQqaGDe+CZ@o;drjfBlQ4ORYF{
z^SJfaTP=$eTM@~Qj@7GIlb@g8YWJquI!sX%1`HU$z<~o%Rkcl|PD;tPZQEGByrp?O
zoiJfOc{ZUZM??n#?Ah74t15l*Y+elnF?|M5yJH79ojS2{*5g{U2DR~nF?cXn;8Y5z
zfHq?`JDz!(oSXha){VE|N{H2oi`1<^b!*lU+u)>Ib~@`e@3Inpz`#7JYJ431&p8XE
z;w#}|d<6RO>HWY1^tkd*^ga75iWhy0v!XgskTk>k1;XiO<pmeA>%-|S)(4<u*$=FJ
z;xQO;6tC|n;nn}m!i`D56)$JAFa!pA*pXy)N#P&2T!U2foM?0P)Z{pxedrbx@b<Ay
zNsML0&_P7&+?q#8lNzOK^GWWWi<%g!M|zu31+?k`-HH8<PeY8|D}<OOgb2HknGhmT
z2yqWE7qDBb_DM?=Y{B*dcwY$dq!1$Pepf<>Y$3#pz{{<eGc25U-g!Zn*Vfi@!U-qv
z@WT&NS7#OK*%bNt`J8?B*}U?~D?!JF5S)ANxvkQ5f5Oaebvm7Gl4G~qx$?>@S-f~L
zKmGI*lP6CmA;GG&VDnHE1((a!>2_@cgu&2bfP5)s?}!S=$D!)CZKeAA1!P=ywecV~
z0^ubTzwsaTOnr@hA5AB&R}X5}t!Kv-f5t8~yj|P90)6tOmwXTq2rs3ACNz*>%Ns?v
z2me9ChOO*)@=4U4+wn+21A<=9Ji(GWl`XsWa@%$P*KV%4$iSq=O(Wb)>y<QXiumnU
zqeF;RQ%~_%3+Xc9SZs+&ph@a$YH%bb=qqcI@})l!*S!z1-MR*y&00l?v#Ol+OV@DM
zOCRt{!9M*o=`~K8>Tam2pHiRhLQ0>74v3AlQQP1~MS97bf*?IDg?*)^hH!<kNA%~T
zS0Bd~rRw9-*vE#FQcBjWqw8^_5kl`dYgI(K4Ol0HI2`Me68jZGh+BdG7i8!ODdpY!
zL7%=f-z8y<lroZ~I6~<&A2>0@DiXb1$M@*PfGdDXz=>YeHw8V0?O0$G5UNgw3+OGS
zY;_kPA%q>+X3S{?H+or=CBSw+o$=GTH*g#<90=8Qy#VOj3JYDiF%Ja=1@!IPH)I)(
z`1pA8^70rxdNjGYxum3|ker;1)9Iw5qJoNw3i9*w`R=>#EbGphIQit0`SQyzTd3~-
zM@>x)W5<qV<;s;|Tj->e0dCdo>}(!;>@hC6<PvN)n|HmhzWR#ElP9xb!v^m*Ewus;
z9z2-k%a`MDIK1(God^%+nQ%%e*Pvf~5rOw{W}M5I^mLq+)nJk|Ki`^q$N+3k4>eo1
z65q2sHEY&l$NzyZd<hf6=Z4i|<ATgAG!!Y(go690U)Xlt4Rrt4gA9J-Rd&AkIyLiV
z6YF$Q^8F8-aNA_wykQDATzwAlcDujk9UQ{zq83K5nrWr;T0u3*u1}|v(4!A2d3ixY
z&_sfRE)!2+-`oXs$s0w<-rZyk9^@s`LJ_3qjrPU3S)xe+6QtDmxdCfF001BWNkl<Z
zaqSlV_SEa_DXl<J6f~pjgMW-%0R1t~t53y%TnJ4<U4z>oe!lHABt@m=_*}6X<3^7_
zRYWjJ;Y0c|q#N%ofadW~xMV5aCyoc97<&_WNGVqfAvz?P>!1k7R1np2O^xsnp$VNp
zcfm%2V7}fKG6%ZbRS5Af;KdND*?>Ib(^gDrt-IAkj~+d^`s%BhI(4e$I<>X6EL^yd
zg$oz9Wh#Ua{PUmxY?F-rg^e3GvS!Vi785NgC90}&?6Jr4{PWK<WXO;vvBD>vbQ1ID
z&*!eY?qd4%=}0MC)K*MN$>z<QS+QaTd3kxAWUhtq9xW+l>w^q}pz%*dO-RHR6N9I+
zqRDt-|3PRmDz(3@0ts~+HhSv_2R<By--m!Wx+e1r#Ygyd6tMl;n<-l`pI-OeMvupy
zMD5jw@<j_tz{BB(4C4Er*LdsU1f8qHz{us)G`7Dom8vb91M*OP|LO;@?)Npg%1SLn
z>Ey#6>mXrpZ)$h$pzF{f*d4J(LW7x55keb}2hDrI8opgAaW*vY^lQ_(=(fk%U0#i*
zs37zrGK$a;NEGRpktzfCAdujh<X-4H2@&uE@+bWWkV0TrVBDBdUfS3an3lJy>S~_<
z&vX_qThVkJK$GnJW)V4KkH%)V8>&U^Hxtq8^ACVQVXR{nCl+x?DZc~$(+*P=0Fh!E
zuK_<tNLDEP0K5?iL61KADBZesi-f3<`17Ct%$PA_qCQWg;7{IW&z{}l-mO9i;^X7F
z^Ugc@{PWK_^w2|t#>ojG$jQlJ>eQ*c^wLYDr>BP<PiK-tm&?V>nKPr3Saz_(OXw6u
z`Gp7v8#KLRSmoXOK0H&CU6Y%3uE{ngd*aS^lkJ*p+pfvBZQHhu9sc|N{+{PWFFKB1
zeAen(7tiy&prB_E2v$}&rNbbtxFF7P&0`WJFVY^iXUvvOw6WyAy&QtNqQAOZc??GP
zs;EleV1N^MD<8@t7Y^tyuEhFyIsL8CL>cZqvm&2II_na3@ZR(6^^q@ehO`((2U31{
z)9-QN66YI3ib3*v-H_2SPNO1oV<&u%ozeY34Z)2f2t{#1$&u5AbN3gg49tbEfms#u
z*IOs$9zCw1$?i2fZx?ONMz^HA582Un(u^QySU6n5NLJaq#U_kSjKeuiznIQOgAcN)
zPnWslW1R4ZJH8xMRp`fy1tVSpiIc-MUGIbXva6$q6Nlrp^dHq7jE@aw2R396$C8$R
zq?T(n;bavOehV!zXUvYfB272m4<+qYbuTF-bYXKUNv*Jg$}t&fO^(nDsbQhLVJ0n%
zJK8CR$H(`<Nh>8-2n&7J&tSJd^uJz;LOcM;EUk#LnJaVCpm!pcTvS0Gh$*UYJXeg8
z?Rg6q&3g+AW*u&R3mIe{7YSWnR;z41r3OEVAu#_ba>B&IncruY%jHt*^z@Y5VzvSG
z<8fR!RM9g^lI&dD!44CX4L7;eZ*Q6yj2@3Du8nKeZu3MI@$uI(%+;T)@Ij*z%HJ8v
zUU@9Gpn5&Wc-`{E(1!joI4%6TU2d)?Eog^h52Zlqs_3UFjPV-1PUhl{n7VQd4~A68
z1No+e@6dkvOc)um;Tr=aa-0vU$ebnwVFeAAwE8!#noGSlwrQUfg=VjFjSvArMZTJI
zt}cUGlkJc6SQTpE?17~U7FDRJKa(JNT#XHRVi^UO=r^UEIO2u0Rz_Hm-wVdW*ad7!
zypE=t{cUr84b2DZBA&Oi5PW#}%bI}hcAjQYwf-71;qw~i{biCgScgRN;Bljw`c#su
zUU#n2F;<+7&NBF3JV>GWvD`!@ox0G{oh6K<AmO3=q`U7JDw#Bt;V<vc1Dn|If10g_
zzSy9*XX#kvepT`{EKCSQOBf45BYl23%H{$KhmihE0>>vp&uc-G$qY$LOG^=Bas1NK
zJafEscB|c{EpH2j^K~$&bOJoc24i%E!Yi!$f{+X+Gq~36$LRXO9kYrnaC=C-&Dp9r
zurjdv#cL~^Fap1Q&JJI1uye(}P2uwlzGJDV$(I}UrvwgvMp#v6uEB)+IU97UnL%L9
zuf1P1;S~jW7|<1!VYf%R`38%sjxCq)tr7S`382cHB7XPK&fMt~6bae<`kcS@n;l?s
zbrIIW&aax6SCi*3hvOlaDbST-OUGZqr>$U}WR$DsPu>QeCZ+p3hI`I4%LE)g6GpeU
z&`naT^i1g{7=SQ4Mk^80PKhvza6F!7Bev~l-KYM}6iv1H@9*T$6Z3sUd9aC7r=_;|
zfjM5UFCqV^^>=TE<eQs9(psqllw<Qw97LoXrqZ))fUuK%2hUjdHi?|tzxmVHz`U@J
zzj?Lr-~6N}d9x&oqYz}0V$ogP*hI@zW}fcBgre#1U7I}$e^CwS<O#@cVtWmU)NX}6
zn$Pj})QFU^gWR`IoM*|uh)Kr41fVkHU}F(}CGguU=Vy~)`hD;5;C?Um2$24%Dk`Y`
zr69q$J3A0HI`7Prv$fOzukBo`#b3|8I`kC!!gWF@r?3CLGl!$egU>pb&9BbDgH7+v
zhN)DRRQ?mbZ$gy5;MA@o#Ba8eiUc03GQ<@V(^BgzCOsCdTDM;wRoEtoI3ViliPRO@
zHz9XsNs198id5%GovWvz$|ISP03#E{#r$|K)9j8V_or3i!{nw4Y+cXTHS4y>C0(D)
z|5){4aiWKo*1FElEWW4pcyJ;cjLTwTyQpr;WVu}Z!Q7%7`fYu~WlL6ETs+!u5(fjD
zSy@%p;UH1f9*p1Vhj4cIrjqcfqNLRK)cM2*p6=uD-=$7fd%oPdMrEB-eTQFnJIVw*
z|GtA6D`>^qoTakUKuwQ0&Vi2n%mtUt$!ev^U-k$3;h{wRi;s&Co6p<o%Gur&A6|9C
z&+PQB$B(hWi7{!FyYpn9l1i#Yb9LYG+uTJk%co*cvPS~OIt9hWVTA2lwu9-~3$D1-
z(y#DTLhuR0rQ{i7)-|qIsmE+yNu9(Aejn&4Nnj)P?*emQ3bB9IJ1hDu6W<`Ki3%~b
zo1K)a)3$wZKYosFQQ>o!ukq0EnXIi5$M+PJrhf0HKGV>6zNn~sKG40=H`Juex5~O3
zLfiw5;h8^;hTRddS+eEbz=~!M>6-vpE!hb+Y@P%lXMRw~t%X!?SsSLjA7$@7^bAy;
z-p$k4*e*od?CMsvi!Q^(ZVWEocGE4f)E7}RDPnf<O|7}S`)~Z_CuAw$9cMdtN~04K
z@&4jj=wrHKhZ>vxMfdn1;SBhY=UZD`6(T{$81Rb>i0%kL_3NTYhb5uH|C>ZdN3@I=
zu=aNWZ8@EEIg?cQI@`cEXW=(bPou5us=!TM^JoLZcX%ib4GlAM^IUnVYU_>GW;~YU
zF$Dz$L1Sa`gM)+l#YYe7fxo$>`*&_p0w1nfuDhW+QxitaGMU_oXH{KU{>a=^mt0Nc
z4#KF9+O<Y-*xHV0b*58;Pw#}HnIYAmAD}yMMQYJ@7lcev=yypA>~OM>%hT^oP#bH7
zxy(2CJV}Uz>s6~+{4;e2HZ$VPoJgdeE%*+%<5mpEZ(V!RKEM2wtWF!CwUy*O^S@J5
zFTXE#U2kUuw$Drtrw&aS-7fD4qT_}Bn`;)T6ZdCF2(;}|9&q<d-kfEv01ElyeV+Xb
zheO};I8UYFqeO2D{LQ^sMaf|)!m;OQ)szcwqx%Xpx)q`DdCYTNkDBE5X*(kJerx+Q
zRNv$N?*Ff_*O|GlJtLSK{@Hcs#oX-&e|9qw$%t13Jl8x0M5)Ai;<}xS4;qd_r}s{9
z%QFVxQronowzMSG!rK|6X%hr%1r@Fa1=c}%dAq>g+PTIB1!oOxd(x?4Z9SmoN2Lh9
zRKI>jfC(7}xgBYLx=;(n)}W>=_x*I!d-}LX=7LU`KzHP!Sp+W{RO_BEQ8Kr=;#$>J
zQ5hT(_^_+&eEoBJ7SHalDLgH@OP4+6u;H|HvvK8}oRpMA=u$d-V9yH<6=2xm#gD0u
zI)&E+!h8XcR%ou}Ip^(Iewce?2O~vtuEF5febw@A>G!P58&wn9M|aNr<gEsHK|_tN
z`P<+XGp|!V2g{@;*W39HoCo1cb~{hjJ2%)z3RL?TfLI}hkG!}M50qX}gG$2<Zj~-J
zfaUBp$?dC%zHG!d(84P}jdg3{gB%~=$P-F;M=AswD(A-!+KxTsG;Debd-sxnoeN9c
zgFf+ub3;IdQ4SILV-A!q5JooNc~w~?hcx@BaV6UBjgtN#T%b9Y++uMrcw$dEz<GqJ
zZKL=vcG#T6c;S)^BNdnMvLU^C0=m~`@xt#a_eG44y@I_sQlSJj><>u6ljHO4q_wMq
ze|7v*vUP;HbX7R`mZnK)V};otj4UYY0;rbyzq1-fcY_H{m+w0un3lxT8zC~k?Z}94
zafW0K5Of1FQXI}LaSV5DEkqKIurQe;@HwKuXzbYR_j6M)t7~AWiFH3UYv_)5bZo3;
zKTbLYtV0F5Xwu~sKM{!G;o(UI|Gv07Q8~Jy{?kj5#<wC(89|Vf>Dm#t#ct1Bw*jvq
zNkT}ZHt)4Mm1xKR%>Rj$POSAMhZ9CFEP6;uq8j$>vNe2sRE++0IMmTm<MMOKA_G)`
z&voM<uszhYK7=2xx~-YcIicuR?CPi3*47q1%|kA)qdRo{v(4bk#d%IcZZ6rnbgf(m
z&S>yLTuEt2hw6<gP9rdI@?oO><HSTG%Zp;zAnJz)9@k)4XMx^~pWHrHJ$fj2PYB~}
zdGz*$50OvjyG|77m|W|6OTgr^*VqnrEnrj)$`f)ig-j|{CwSf-siI=rWl1OK=x0q|
zB`d%hBd@(Je`uUH<|be4d!SH0W7TU?v2JFcdG?15zt2OJdV@2cAggo@368>V-ap*g
z@5{kqw}<8MV_B|YqcwZ~(pZT<)Cq~QijKV4erR*$Cr^29gCLDV`f4h+LHyt}#{+xj
zduWAW6ju-oz%!<9ioeYaq3T0Og!8<*uNlk%ds!ah7h_MqU@k7TKgx8OU0FE^BYl-H
zAI98H^S(IOp_+c~?wAL7(H7NGGkrTYQycL^$&@Xs;2SrjL0)hY^nh(i(u$p2nWgfC
z`xfcu?^aHIV45{jP^Hz&rQ4Q;y{UsXS)oRI0R6aT#Y;)Z9%+cO8m>(lF(^?(bU`5?
zQ2XVJreYx8_ke<WU}t<1x%9u(c8l{LQbrayS@YM48njq)?PLi*^?G?5-ddfEO8*e0
z3x*j1?POGaU$#K}t(}QKm$Y#AF(GD>kizltUhcF}?9Vbk^;S-co(0P5<&jHyF&;AZ
zzc_;zfyv>G((eVa*xX1dx9@7TITq=ydM$0V79o2Wv>ddKaqU#&j)<V{6SG(Dtm<x!
zR=5ZViJKD62fSCjZ?gqy4=pjDH7zl_FHaxCd;=EVoM1CHM&t&cb8NST{yRMwOmS)7
zI+_d$87G!lzKRkkJ#B$BkIvt+BVN)8=}iT*+f7Fke%5e?eGpC^*n1w(=<du)GZ+uY
z5Q{E77KcckMG570gLh>2g4M)ER~szQ(8?PG+iGOMYd>6l!N~^3)R|R28jRz^^w;&J
zDi#yoQSrz;Lu2qFQm9%|G5N-!taMD1vv^7dudoD9)G!LPzSg$O0*4(E`j2CekrYNF
zvtJR!RlN>2^gC%=TO%ddpMP2CKfVgQ&uxI`)SjydP|8-%!f9%isP|DAJ@AgK?6YsL
z_|JLIlBA8@(o}vw#}w+$tf{8&Ps`-1b4PMH=)YG%A6sfn)g5^TCCZWG<4Q}7o16xd
ztjO4Ulq69M=nQmYs5XGUQ&U8SKPYTAtv@Z-dA?aJ**<gWZm(`_4c-o#^y9E<$@m-O
z5!rD>H5B|NheMJ!D}Znd<VY?TW9;!VFR+}0cZrv4JYLwBG)(0({8Re@pBc#8^?F7d
z1sWb$_XPPnI%D}AS+TC?(f9}F)~+96t7_wgw=h~ou`#qqSxqSU40}W3nnW@UuJgdo
zSto}Kciz(;(q4CbWgXhVQI<-FJGy(>2ke)Esn@AppYK;%&MU^s%F6vhHbi)@MMu}X
z(O_xA<dr&8hwB~~?3F7Jf1Sz1sj8+kS?k?2u<Vo|XKU5Ug}prIy~0iThn=I|H(wXH
zGTP6BD)aXZ03*48+y+mll$YIx$FbjDXcNPvN;m<pzOYuA)L}~g^4{u|!$`FC!DF$`
zkdkPmE-fqUB$BfPmAr1&P6*`VVF={YZqB>2>Dxd3{n6e`NJZxEpd{R!I(;{y*4{t5
zf^igb7Z<echj?N-uEd!6;zJhSR~Y_IV5@m;fuF4%XKE{1u{2bKEO1=I_2z7=pu8x_
zi=i!H_Eo$00i4tvv@C6HZ;EFhW{H1Hj5bW@`884~RO;7PV!rUQj3yl2O`Gt>NZ2Rw
z{sfl)t&!j93yCm}g~G(B^s3Go6f1+k0W2U6$klKd7wa>H9PnWKWK3l^Ktnt)-p)Hz
zT{i&UgH*S)us#lAUGJ4?{fIQg=;tP*89pHZyf4ipOiho49A?@Ladtz1K&Fw!Kaj&~
zB((s+#2iD6kYOK4J~x)nqEb_&0=}rPJ#NSD#MC1H^%7J0JlIR6DB^eS=j4u7Evk&c
z`SKT0Pz`du*U1$f9i8W0`qn;p!&CQ@mq-AZJDBnxr@)e(hl<@=&hhEHrt(8J$?eQX
zioEan^-j6ZiD)S|H&L>5;6-=Yzy3d{I_v8`WRA%}3mg!T--JPia<gw$#V_mPn|=8V
zlf@-<{t*+~(qAzL{=o?~K%_>%;An6sdSqqw@qK##=rKJH(RfT{j)HW_9<T49s5@JZ
zMXahCxci171)-!3W#Bg|6~76ooYS5msj7pUlEhH`agbcXwf!Z+bk~6_U4I|6g#z!q
zC&pf>CU{WUMl2>&gPuhxwKANW09@*af`PAha;()mvCndIBa9ZfgO}^L$17W6)kmF&
zB-NQ#gx)2A^M^>vefG@zgJ@HsEm=l3_49`!`6D%i0d6X0)W08tc@2?0kYNny?dAG|
zn{7AaG+><$;<$hTrJ38TEY1+d@}e-7Ef?M|;i~1kr`eyOYbFA|4FP=75`i_C!l))}
zI+WPn_m2G6n}+e560jD!S<)VTQiGYOAuaoz(8$2|;1EuxB^US6Y^Ci}U)KH|97__W
zrRC+}csc&iQblm#^Pe~qxJZ|yh{wGrw24iof7Q{{9Fz0D#Q+OdNbrNDv*y5m{-2~M
zSW#%jqADxnDk{X<xP|G^8ORsx{rrQ1%7{53`iJn)t?@p^mDnx5CZq7Ef<>%%bTYP!
zWB9&sY8CD03&Zl=pzdAh6QI|ri7zTFI!v!TfLIwH+dU*`sQxdSJR7a%!gwvFEbsNR
z#oxjr-m&UdDd%j)qz#HFgDJ^993JOydt!*i)XBL;jnv~@(&!^H|Ju3J7i%7qyK-K|
z1p-e?)%FYcd?KIT{0{_P(LkF{JS%a7$s|D?^qna|iIKc-SIH@Niv~I=JlWVs<N3LE
zR;UY!+x-1BrdN|HVv<2V@IAjV$y=<T`${8r!XuIDjac+y*|ns<)|4b-_Ode&hI<;1
zq&57zX?>HAB+!^prRPkq<98u<^R-lCT5B?e6-z|v_*4s8RJjZ8n%q6r-H<Yrh9MC&
zQzLd|7xDTKQm~?>1hDxtc)jMkQv6XMEqB{7Ud3BvI_xLSpJem*#WTFbsb6Vg5i!En
zb4LM8olXt5ArCc<8%Fbqf$KyVLRTV=bz8`}VwlfV%d$Xdl3p-2YZAOyqXpatb9>$i
zcH8!a$;df_+z8bAdJ24?(>q?wVnkPPes|Z<@r+%~j}nuu&F@ZYyV76q@d>YuP$i=T
zq1#8cb9o`O1l5wl_n6y84g5VU={mkc6nG-_+%H?YJDmmoTQy3`>*U&5)M8(lCSxay
zn#v3Cnyk(VoK{lpR>I-2CyNKecgI8ih@n&%n`yH(05Ru*D7mG7V2xV{QZ1v(k793l
znNtyU)DG#?*vmEhZdG;i4fkCaS(~a$Yf>?3ma6S9xRdB~N8%2U%5BbmjTm&)AHy$+
z^lEWnQn=JB(g(GyAhN$MDZOC^mRh0aA@|k94zre?N6GnL&5@)M)l=^<e}A}=t{~r5
z&;9ikFpPeIh-URtn7e%OarDb~G-D{f={;@`7@2WjVl!<$$75_(I!)TTswcwfcuN2F
za-^!jMka1_ylSeO=XD__1`xp5lcavJ=~ky80ezY9X9hV3g?Pd}>!Sd=q=6ec%{&zJ
z`>P10qkd2b|K-iuZ~K%brKMq{X#N+OXGmL5OoP*li-UlVYZzpKH`}G?N|bxD<Lbpt
zY7$MpBeEo5DXYMXP0)EAO6u$Cwk`s^Y&2ib?UYYw#!(<J<m60h@U+s`mSbpc$Q{G6
zCp@Q_SXaF#tLvT5Mj}W~>jq!rHgKB8#J|7Z6&}fk5?A073H=CW*$BNMbQ2@8)dVbQ
z4!~v)I0|df^GB`12^}ba-}$J(GE8j$OG%i5sA5T!+FQd51qe6o!A@T;^2Po=`|xkZ
zJ>WB3XvYo**n=8z1v8<7epA5z$yOf(TC6xgN+&OAfrANMux?-Op`?z$RUf)djKm>Z
zXQr+Gxd$hf!vcW%t2CwX&2T5)lyI!6%ke9=_g?V&&Gp<rCf6N%@m@AM*hsX;RNk;I
zttYy{C_H4mH>jr5(Bjy3#`9x;e{Blr8$fmGtES=ji@#~lV_*ocSoMMG&!^JU5lE|O
z!){x@$sO1eg}ky;U65h#vr7SveLTf<ixFC=)?Z8QAJi0ySIYL<Giq0F9F>ope~9>*
z!Gn@*m@U&!?ZKOpxoG0h4ycAq(KtJY^N6&q0oiLmhQ#eLJi$moJ``qLOZM=I#M~9y
z#W6h=b$A`Dt#4KDGEqX61f!!jb-13w)!yH7^+c#7h`c@H$MFND_}C;;0H(^ls3lOe
zdF&(p2pyKjxP5^~Sz;8(p_IUKtY}?Uc9toI@FkMYyEyo5Si>Hcq;UR6K!OLMiXU6l
z;oG$N)nR^dN4jH!<?b~`X|NI3XJ0YP)rIIfXc&QrV*Fw57nyhy;y+g$!)1qcTkF>k
z;|N?X-s#50+I>KD{73?mCqAmfx1Gf<q*ACgG!kZb&hp^sPg5c*J5GYAp1h%1tf{+8
z)djX+T^C{*_-tv9P1u6sJZUuI@}s9`#Cw4}u-`q5q)wPrJ5{ABrQ&8o(7XKBA^bnr
z9cVxADl&W~gP)(uY;LMrq|xXKzG>{bg@O7D3k;SJJ1VP4DD_>AinlN?w($a)LW>!2
z)%V^Sy5|xfttHPzX~*0$UOEk%ofZXd5Fp}){^>0k?k?jc(b_~LkrnABa$r{!Y!{Q;
zF*^(L5}7ftzeCREUHwV$tJ3=D!~YrJK@GhGlRec6%9WQ%Ml2F+ekq@eGVE=h(RyA0
z#0;Sfq!(m~ySn2nZaK;P@M*q;r-()=ZDXyX?F0$_DUvC0-r{RhRz-m(sYP@(s=io`
z!a;d!Oy5H)J#}8O<s{&N>VTTj1U`Je8qV5o!IGtub2xv$7)$ypcumjaNu{BH=336q
zr8s3^<52$79h;psi4n2&RJE(%dX={mq-)Al>keyYt-7yIgI5S6sX~HYurGgdCTkw%
znk0`VbsJY%mRqO_H2?wO-NSr!Vf>8c_hG9O!$Fy(6eSR*#g5GzOvWtR)9ZyI;BQmN
z>5eyvMk~F4a<=KbaDN10aP6OB|A)`7ZFO2Fr?7&>%MMivhy&D}5Iyq#d!&=m#}^G=
z_PH(5uT_1J+9RC8+8P92-x{Yq6b<KhRPlrMZHouju6uL-qRA?R>x;#Ou5XlV*728L
zz-*-jgw~f<gK8@Wuyy$6EtN}HTuhMRp1%uOue;vTsQd-EtB%$tMC0R=|8i2`*ceIV
zN65*NuyO~u6|mmh8n3Rofp}DzxFNUR*I|a;bYj{Hg*4(V@qWILOv*`T*Zd3{&>At-
zj&j<Ij9_|(pX#_bIhC-DBYAr}<S`ptW9+Irb0jILQR<~1A*Sv4o3J#Aa`!FYNluPP
z+sVK^;#!0>Y0B#jF-S$#?qTXP#D@7~uGbiuw?d4!=$tu2vkHXeAwz%wp9xm(yS{p2
zJnR@^GD^OY>=5CAaQ%A|$;5%OxQ`s%x3kQD0-QExPBlFV?$8cLq++1dRF(4{2qW^+
zMqK&X1_gM$=4qe=%rEAu>B9+`e5BJpU!ri?pMLMOU92HeFIN+hl}mmw&)Quixba54
zA>s>7+ifa=<f9kSz{N(8PBd!@o*&4QSk=UGADm|VPF;u_GO%0O1j=q)qW9cspSxCq
zpnqPx11jE%iKH;wAp-c``PRKZ4SDZY*GleY&q{G1JucW2G<`g|lqjO0Ff|V8b9Q!$
zQyd3DBS;m<WJb06)~BGHkk(?~C1^tf@QG|KyDUqsTa-ksPvsz$1J?IL>lxI)K>-AZ
zFFJ-OKrvRYs#X9aU~E?0e2pWg{7;<dDA1L`i>%Lfd@<nwuJ}1j9NYzicCP`7oGMKi
zrd;cj7W#m!UsU1aBUao6U4o{kyo4-1LDYlHcA}Dkh$U4moMs9v?v`*XnpJEa(^=vr
zUhJ@PJ{T}3_3=`b+zV=`S^M4Bi#Ve^BB^00MjA?NxLqpEkdvRE33z*5+%_d|2Ove(
z8K5!|uv@8GgI0htIDFLqjJ)2=SCcsyKgOTpN~mbyuaY)nMU^i8KtrRo#Oh(V{dZE%
zR~cQrnYXHt@)J4Ic4DB>>FEfA?pLZrrQ$e_+~5LfE_2=5@vzv-O{+g4uBAjTn;e7|
z-go$Egq|rj+O~K@gp9ZV3^DAz^6os}HVmVUj?0JkXD`5@+>Z+#^r8-*Rb%t~cietF
z2-(>4ChU>Qa0X5cQeQP15w!mw3-Ct$!)t29@c9|$w@7Fs)7kFtf||m}<RiL4;a^B0
z(KC8&wHphjO)`}KlGe=)$`lUzC;@#tFqeH!T_3th6*_=db_|>Rf5B)6_wgt@asr(h
z`c6*YHF(tA*B}#6Domg4Sp2aVMQL(GeW;~Cw_+Oj>8;ua-Ud+HX>dFQ$iaqnC*e63
z;7nmr7y6qHrv7k&WPKHQWlp_P!<tcjB?atFp?4}}m$!7EVTZ?geX@(FxRjyC42^;a
zZ$Crz?x{s})zraixIl0Hh#}E;Ck;nC$(kT%7S9pP!fdO-HF}r1r5KwgMOFReo4%Ls
zP;sXhf<PET4|-ZX3Cbm?-m3~xq;>lv;lDwkIjkuRGhN^6ZSp^S=&p2L6o?~=hNBSC
zn#Z>^NcmAe6u=D>&h+G?_7U07tIq1rY|#OOfiJisw{J`0a)+%|IwHm|sN6~H)w`7-
z9H;x-`SUgy3f8WBI5R3yU?O;PgzVng9aUA?>D(V{B;^N{9No{GeXxU$`Vr6H5ccTx
znS5`EBk_f$!0)`;ojZwM*xoonE>mNks2SwG*g4j$L&@MC-%m0p#EhB+y)oVYR^{VA
z9tLgw{K|FCWbInrCc>){hp@z<zfiGywm8U%x_;K3zS9vQC(8zdW&Dsvk>&GVkvHcD
zfxMH*bqBZMrakymt+1g2BVV#XTya<aJdQT5izEbx(E84;hJ`9A1Yb&;<jo$p`7{o!
ztBTi4zV(81!44ZCBwrAx5M4|Ce8R!7TZ=GNxSlX`{~TZ6{>r^K$(gtw@@lk6C6cb@
zOSSB0{E{!XYqLE3yl1j&3~ecV>})t53CWiFpY1!-zLFi+uAJX>`@jDQg%uu@%d2)~
zTU`_zAclboz0-D7N2w5OG&;T9{QiS=2fS)gF9uka@;Shuu(dd7L;uOy7X8C_AiRvd
zuwYrA#pLQELJ`|IyZ+A>#_Jvy;M=UVp*UqYb`{BL4arTB`HkxQ6j}Mt5HCA)tENKo
zyYD^>4U2gGm{sB1kUf3&o47-Ew+dIgtGb)L76{nExhvgOhxlL?h=%p&31)`Vt=%%)
zM0xwgrvkerS)UyDSBA9*alVy6vgT+{3!3n&EZK*CNkkr3ydt>@heo`oFI6b(j{oz|
zvB!5N)+D#_AFSVgjgAgS0mP5;no78?f8DlNk^mZ`oX^V0MJ}XC>|&qS{YbuQHzqW7
zdcu)<jYTGB4_O-q<?hDDv887iN}sSbm)-z2JQ>(Y={g_;Dbz54T}hKEw^UO09SlS`
zk1Jk6rNMePlIY=GiV9g8ERqi~2&UhC-=CHPZM5#;E`9dRt>4xx<q;rM{Mea`30eHx
zeB9Vzt{Si>ruc?dhq8_OB`K9UrY?I>+)%7#;mTY%=q>aH&WCV3?st^0nE8iMBYa?A
zvwfeh#=!qR0v`JGI-V>4Zijbc*e?(V`9tvx--lIDh&ha>$VgHhBQG*#Ta{apXWz+7
z-r61O8GqxmJlk~kv@>|c&g8w_@O+MR#t&`p=}U`BaRaI(D(|6kn<1KxKoXZl7PUSu
z;5f7Qy~?H}T0t4t{4V0oGw<5yt>qg!XL?PBeeaB*|E4mhsxTYnUnc&}6hb@{Y2Hey
z=HfD(AfT%4I}m%W*_tm|>rRvV%~KDOP?*}@^Jv!L<pG$?SljLw#B=%Or{R7HWqe8H
zxpzgr5DHK^aqvjD(0o%fLf~a1VL@f6cf(!YG5>uCV<-7PYbCe}nLu;QB(bMU#~AaV
z#1TG7=r}gh5?oMV*~<&D&aX-rY>Hi@SLm6AB$tZnHz!p~pmn~a8GpYIoNs8do5RfB
z4Jm2+s?*fr6dr9AdQ{B8v;<n-sLU-8xS}S6^eRG}zzOm&{vO&rUpDD?+i~_8U_(x>
zCA?LHHwoJIc<yZhNE{3&F01dOVSJ0ejkw(#;~#5Q=zBT$`Ah*$ocZs<+@v&Kxn6!O
zBzitVUgdjDYN>3x0Q#c&d{-fh^=0nc!|tAxS+6fk6wR)TMzm3OLx+|Pqx+5gt;QhY
z{oC8=Rv4P2)hsTaPL7~A|7W^2D+Gctg3^JiuYEDBn=$areM9mrRsi@G;uJ>DX7b6n
z`KtR(Iobsg?c1F=`HU0@ey+EHyHwCh_;7{w_G?vkA52tc6d1Jg3$8=Te^bQ2layFf
zOJp_}t#wXi#sUsr=<r2d=Tg}(Lr5|*OPRE|X^WarxaBij?rzEBLo&L@AoMT{FIKJd
zeuRbVX4wDvQO^4Cr3Hb>fU1gWFFcz!lgH;l`|$^V2GC`^7t2#D+YUN)wKOg}-ayV`
zKhZ+FB`mP+(saMANJ+ped!NZB?^9pkdK_HCUILCKe}K<eV@SN3NmJs&Us}pBw>2VK
zTF*wS_tNAO;&muGFM>Y~1FQeryIxeyQ(G0TV8^xR?0D{yOh-*0_i%)~s9;}#*R=k^
z%Moe#8cnA2bWM3OW%A(@#Mt$GW_j!VwH|d`MVTt+dzIJHlbJnj=TtQ+ai{n~-d(uW
z?=U>66K3phvW|VA=t1N<=)DIidg(<GWhOj!Vz7&t{X=#g=DVDunt^%UT0=wU9SLhn
z@I>$ZSP}lx_XEZDMeakQN44mZKj^b%U%+jN-=pRumU6gCSMJ?zh)k$5J$v*yG+4U+
zbtTzcw<F4KXsVxv;mpJd8j*&?6v77}LrM^JkwTH<Tpel4*$A~6nln~%%F;dWU0n<d
zMn#@vwW+;lUKw6P=*JxXfNQnxwJkIZ(r;3u4wRy&J0^SQ`pHvk9Pxa*gyXUba(#Z`
zcs_OVF?3R*5%cOs0g68Pm23JImn_9)REY{zK6@xPgla@UNAlH^iB}%K&NIPvS&Wjd
zin<m;(aQse>paBV3cqx^+UW}uJ(77_r?pbWM&f0~N7#sP*$-~6J6;&`e$0<$_zX;y
z5bWLz-2L8D0=Nn|pV;F8y!>)mOcnW!ldF%32{hwOF%Np-!b<hlR@Gj&4V7gxB;M;g
zV&d3v%|!OMx)pwTUzdZ=Ymt%Tz$pYiG$f}I@YBTtf9^+PMIspY5KWsBttT{vQ{#QI
zd$b}XNGY}LbS60+i_TxTWO1C!EHx24rN3*+-#7Y*-1*k0`SN6e;LD2(WR*z{TlEvk
z2d^sp+16J&S${0GVSDOyijbW@ny|D=S4JKmfT6TGx+@CXo^!Lj;r0}^NRr}F?rd)&
zOTNU>u&or7CNaJ8tDKlf&?8o48Y-);KQ3pu=b!yHA;$BblWMi6C@J7|-G%G*_`T)a
z4~3~QFAa$Dj#1T~u`~1|GI=5n`7J3(npB~bitY}3=6Ss=g6~7ZzVCGRr41oKn%Tf7
zkK~qrd-Ef6aA!H!?!VT8Td^N~S!pyxu(h#AYvBz5W%@dDY(#3rS!u)dnB)d%znTQs
zYF5d~x)Pp)5iW-{^=gEY(z8!P??03O{C*LAdj}Rtys*lYIA#flQscGeX6Q5(Om3Yq
zdhGXt4FJ`Y$|W|`hl&&5i-i`6C}S_BL&Z)DCuil4MGhYQ`>QrkhqVPO#zyZN!WW1h
zI$8aTq!mPE`WsUY%D)!$tp9d^y(NgipZs~uk>PR(;`}rC!O92X#mAyI*smFi!RI3x
zH3_({i#QC$OcnK#?(37wagE?VWt35;P)omr4{j?wE;fuewSh(nAHHcefDJa3{$E1{
zd}`!vPGARM*izJ^b->w?@hNYLn4!xLd%>>1RjjR9xz+Vpk)-O0fZJe4WfdrFC`efY
zdxlq!D1aV5K%^3z#y#kDfgY6U3gLjoc71_wy?4^0ptLTeC&{WK$)Xd<XT^um>gcUT
zcFg+SF{egrTDh6MMmMamB8afnY~T7@X3$Y6v@=MIk^MU0g+gPcAKI`GTw|#P!R@>W
zX?G$GFj_CwtiR6D;IRB5vuSTg(H@M>t2W}2m3s?M!<{SHbg_x}`A|}5nIItmT{gQp
zJzVR<-@Dvr%51dTygkWYd@ML80HfKk$;Ln;{`GUyU6`AIhy3?67Xk4mSi<z>ca=7x
z4CLeHpT+xtqgY(7u!lO+J@qy6C-Xaaj5eE=SnmqHh@XY=L9YeJ^KrZaYcrFrE78^Z
zCk90XT=6V&d}+CfnC-WRiW)o!IAuMHs;Cp|{u`@%;)TKC?|LVOaBUHQ0q-_Bqnj``
z(T3(3_;Fq{0QVZt^7LYAEgg}+{Ua#)U}8Y}_#V<GKkXL@USt?CqA+|C-|(D4E|Jbn
z9yF62qOIT7q#Y<?k7vm{{>D%oQhsGoYEOVrY`}0HwAB|gL&}71dWgA3BZ-_vW!7VQ
zus1c`?g|&b`J^)%v_RqJUTdyjL{)*XX>br9t;Fs~l^RBUdzKC!6-pQ7)f!R8Oo)qo
zAim|9Df~bl;@PqI^`eIlE;|t{t=HmMJ43YSNhUNW(y~x<Kellb0Q@ZFDTrPYwQx+y
z8LguCykbQ0_2Fotp3Hf!gQk04Pj{zWN40+zH$2}G`cj-Wk~xp~g28J#sEWugRd)Fz
zf@zRh#g4GtV3l-3z0$bF&W$Y1Ef^jT2l9WCo}MvfEe))}=`EOAjX~dfZ`X!9|78TP
z8OxKx<Oh`8{ksNJxt2AtVf)a4xSL*^KGz5MmBzcxHIA(1CMy>{#cr4Q5j+QEVn5tq
zai_kCQsojMATG>5bqiBrt>WRSv7YHr4@Dd9N>Xko8uzk&SzGGla7%fLq%nJ2bWy)=
z1FJ`4y}ybbdY(eOAys(c=bJ7<Yz+o#{4X?OuEfPy^gjU~W-Ih%s;KJeJ=2@`o%maj
zD*gu7gdw>Qh4lGaQtxY=VoCC<E8I7N@<Ns);soYOj3s|@Bw7busb6<oI|vU%_@~(3
zn3inX=<0Yla0$Hj@x9#zwx~S_!W)(683@%F?=jjkic)_|Z<NX>QN!<&!-kTthu#4V
zZ&EeUAI#I>@i%nC|7+;c2rTR*k(5rQqOBmo*@~lsy;#vh&d+s`&8g)b^YK33$nZHd
zkHBM3Xe=Bvb3N8)7Pi23yJs@8pqA9Ya6wA^W2vE+@PJV6dcGCH5W#(`?b=dHfAq9c
zbx2d?^J1`??H+VmCopQA4MIe6S)UB++8X)bc=MAn7S$C$^$XDwBswNQ2BUU`Dv`y9
z7f~G-N?)0)$CA?LcuK2x54YwV-;rF|{X^uqSLlADNlUYn|89?tziGqgdCS)nRJ{+@
zlSW_)v!uo1z;he%Q`qC>ilt?v>`aHtevz}spu9b|KxnzSLb;oEa2P(~@!N%f&(YUJ
z$!l-A77M?9W@RF~6o8RTG*<&^u6j&TU^bygQ<OAB+(Lz`0{1>#6=0R;W@&hNsPt;^
z&JhrNVQluN@%SBQXS5t2J#bVMB>R-6s?TeJaXUdeGQ0=h+rx<5(ai=53MJ3Aq|rZw
zP|A(>@(Nk`K_sLmgi-^`vAHaFnZcrTM43=#=7WpMzR6HEg8mjvn&|eKq$Dtq_**r0
z$ibwM7>@sU8D9YM3tbSKi#X*RmL5C~^zE=$4XR8U1WcDI(}J{q=uArsx)|E&SdKHP
z+Pt1=t`SnI3@6y)Bc3BIDjn0K5OHM%A6}F!m+jH=?A|1I$(6edF7M$ygkG@=XG_GV
z{;So@>_DzkybftG6^Ju7-ns%<k{g3T5cl7Qy)%FAxqJ5lYqEyQ(Bx{?ZD~~R#b3K6
zzC<mPwNnW<8J~cworFE)LVnf3*`d#Y41x84D-gl^h6N6@z7VZf{B97Hq=Hg3ZOhVI
zP-MC5y{-{W8Ln4<PsmZ><OYObag2?I^jex2^^jcmSfY}dc@LzFlT-HXb!!iHtdgyj
z6hTx)=>OdVx?^fJ!^VlGVYQCJGNTLq9m6Xv_`TrJh;G(9x-vUasaUe}-%fth-p;K9
zhY1n})tVS-(7Y#U2okT%BDGFs7#dg&`X7o<46(=g<Z6=hrPsaa#}z>Cwih4a1_%*t
zWM;XKOJk)QR~bQWEn`U<sO-s5D)UH}m~~yM3mcB0h+bi~6taIw5=j}m#L}UHQ{98}
zz-F9byDaFr-NC|T@aaW$=amE|SG2zzbxm`d(l=ajrwCWI1g%%U%1b3B@b+U;=QbfD
zS=Ls+t{j<b*Tyjo5r|o<X~+IOdv)YMc}Ps=e>d0A@s8Zde*5O3ZOeb>^TvW_x;1q#
z@E%*6D-3zZUZ$t=X&Lc?L+>)Vvz*2$cw`%3S&>8vh1MbhM1c{rEM#gEg`$31^;f|5
zPGVd7fMrU`tF|a-Y=vJq#`e1Towt#V&r=jOFM3}(&yYw#tu>u6Epr%So6b0eBAh3I
zsxxk5w6JAk2>vN{e{~egyd&nf@7>Apy3M84YK;RMO%zO)XCMa_{m_$O!(Z`~V}KP2
z0G^h{uQYNXsYb624y{Zr$W>4>h|SN8s7xxEK`l6;8si!<qO{87APwo;g`0ne>KU~I
zf&DgrhW)zC?+iDh1!HyYB)03rDxaY^9-HFnYV-)k`{q5){n9IW6(DY~g}(qPis?1~
zaO}M6TIK&-&>?QnB~RoI3?Squ-PXmca~+yOyB60Vj^~yiRS|9b$#cd8o;`S~T*k?C
z^j{yy+rYx@n1nySv`cTS)D#qe1^-{K)~xN$nU^L6HqTGAEdXMUc<I>-8%5iGJ}E$=
z4WX>4Ls2x_YL|!X#WbLCP=Ec|p}JF%-<?)i&YdHF^HV42oEvKEPOaXN4MhM1kHY^B
z@35uSn|2s!ma=q&Vflawn_~p3EKnx>VUHwgTn}U6D1iPDji5q&Yo@{ED?SMyLJryf
zv@Gm)o4S<3>5LdmP|Y*cZh^;AbRj9L9^|>Fk;H#PHb&T-JZPFtytV>|Q;sv3<qpqM
z*-TJX9_*z|eq$ttS^x4d0rtC+%xn!2{BNi-i{V$|EA9fec%w=_ebC%0Y;pKJc=0si
z2vFBNv9-j4$l6;_nG7V^Up|@>WgIyP;3kFtGa@U&<gb^?3s2S)f*!qF%cJX4KFDql
z57mCN2&55`c>>>qwMf6l#AE=K4@SYsYhH^;7?iOB75hpJapDTJ8twLrpSyaHOxE~x
zestdBwQ$LXjSXvheqlo<2R8j)%hKk4B1iJ|;BF$VdJXuBD)bNyzIjTc+RJh%|Hf7o
zx$-+zn}UZ*C<!#7Fuq$i4UTWZ(hWQvC6>Wqir}M+HaN|S=d$hMTk-0p3R+M$^fd14
z6$OvHq42UiGa4B95YE#3yujWLNEN{aqAl`y8B3|vs(w&GABkEb)qc|B;Zc8h?)qtI
zY1$tfte~P>H$d3jIk(`k<}c91VT}=6M|4hd0WM!C{+G>ZYEI8=Zt`;_kMThwu9ZSQ
zv2NWSw7c#2#H!bPpYK;!Q~BE9`_GqXPFvC3n5o1We+P!$OHAuHb>DR+Gbj8ErrvB^
z4>0E_jiRV&5lgTnF-le;;wdA4Q5yCjI6&ziJdcupxqUKE;P^$j7$9LiE5e#Cin+SR
zV{W<G-CIbb*81m%v}z@oWKIEQG*ml#vC!eO&#tci1|!yV5n#7*qb=EmMU9hIN8wLT
z&z9Q^U@^qzXSf{^F@_eZ*Oq4@a5SVv>7}7Xxo`LJRccQURMkM|4&`14ZmY=*;<qoZ
z+Ic}EBtrc0#!FM4L0$qck}`wgsv-YEghE6)s7^H(!N8M=yDgLbap4ji4sNZ(iQ6bE
zCiAbV_b36KHHsodkTi{zKyZxehvN<=H7Y~k{<8HX@F(Jh4^sZ26rRpza_@q{JGFM9
z%<r%<U{``|m|Pl2?u_XuugoV!bWDbBKM<I;gCps!b>&BYkSm>8c0$*vsOdW3OgW1}
zGb0j8ITe&x{%`iHs}q~@1$*G=NIWjH8*`uYR^;uB;m@*?fifz887y8DvTHpJr1Zpq
zy+Za5*e2q-ZBWeAUQ^@D%(~Fo>_L5d$KS@6JGIZpwR&1D5pg99+)(Qc)EACDCuZN2
z2~`z`ZjLIWTS>|QI?)hp)xOb%e}92q(s%fG>Y&rIdD(QScO9kPMsi7iqq*p?%6Z5q
zWL|qTD2t0=-YF1Bg%RAUbBz8kI!ScFa(CA)eyE5{o<;Yx<&3^2>+D`sd1l@^Sx6@T
zqn3wt7;7hLqrLwt+QC%NERZfn*lLTIdwi!BfC<;Cyv%N;0`@yb)^-vc=zaq`+c^%I
zL?isQKlM{DrIW31E!a>k({6?r=VK~L$TgLOiyG6kwAplIF^lrhM@LZ^FvgEHElnvK
z8zMc=pt;l1Pb~iwRad4)6+P^yr{_%j7XPHwNNAwgNs}ri1NIW|-<aWZg^_lB@`?T6
zq-(~WsrDp6kHrfE>i4bL%suap&0lf{zP>++wpXUwl~B%AS^CC6{sLBU48CDkqW10_
z%l)s$f0dpVM78%)20oos!G*~ezPw<&f-H24%S(Qq>nFMTdRpAn7)A8d`C-NdHE7VD
z{)awLj?FpKxF{8klQHap?11HtW3+rtiF|kB3mJX(Qq)6p`ylZ(V>34^M-LH+WLV-{
z#|<s`r@_9=0Yo>|prXxZ3tgf>_Dz#7SMP_fFjDJUGy2_iDUO)QO1#V^ly5;|Ocf%u
zc5BSa`@#y0ch3FzTjFYrOV;qQM$8a0Z1=6p5-f4@3FZ~{q0MIdA?pQ|@*^rg9FZ}V
zvUa_pBXm0B8b?~hX>lM0gfU1JKcJ!2;_xD^x`_x$3F(ef@Hu`*s8y`eBCt>FzDuyj
z5i@b_pdT%Zzjxah#|U<sa}XN}A3~ai7FGzSc#4}mncMxdmN%mWJyZyf?zW?tH93qC
z8l!3k?hrnYP}-fcsVcCOVeiidl5jqHz9cOUgqyox8Z(oMGZ`HANDKyuqN4|-_5HeZ
z9^^*j!gDpQMzZ5{hWNj$*PQ;U^{-UG@XPJV7x}`22QJ!=t1%{NY=6vh$<=?pv&2%H
z)ap|)K)Aq+e&XNhtDQ7xg674`Ca|sOS6X@oWVK;c;k`|1{COuA^>J=<nyKW5j&#?@
zwMZKG{Fr}!A7MJ9IhEK+hPE29F!#W4dm0Y}MOu!#61*fzbM3#$#h_t`n6Q!WmlwWN
zys&mKC$Fc(MsPk8wbxcGAlps<RI?B`aNC^lw*o_z<G262Ze~<Jdc%tM#1PNgF3K>4
zLc;<0&5uFj<s}`ICsC+8)I6Z4Z{*-^xT0WExop}JwATvR!NaNObYQOp*{GBB#X%an
z;|P$?^;l|Cc)pp;N}L3zEx@<UQ?6g*Ce{EYCt$W*V!n_!EG{lGFN}fWJ`HNAUd>q2
zgGQq|B8xXLztal}BZ`w%Mj=9j5#gC%wraA_ZIhQ8EaJ6X{~n{o?)-cK`|orrl1aDw
zV1kPPGVp&%=Py|9zkh%oFq{TN6#l=4pU;+VAD?_=c6wm0;d^Y<Y503S@_eI?S*K~S
z0J&Jv<+9Q=aL&{rfjYLdXjXVZg!z)ff|P=OI?9Ra^LT|K5rv{Es=nCdnThV>LN!2i
z?~C8_Kgsjxu;kEH*C?{ul+D<=rJ*b=t4(&iu=}eiwnkptMQ`~HM`>_9jvbMkU)K<E
zSt#sBp*ANUk@2$KlNOC>{%!eCH+WopO|9;TK*v^U&)-mJqB98}Hve^|hrerkF7m$|
z5iubLa#<7z_)391A3!y|G8rcDnK-z$A)#);qXIRcC;Jr$7fzDU4r`%s3w<gRWG}6S
zVL^JcNM1~v87(G-BS;oS!k`mpDi-ZWpG7jyZbz?6jm^l#LEf!s=tv1)xL;?0q8<7V
z5oP(htvJaK&GW`=&{Vr;=rHjQ^L$=Hax&8iDZnT?oFCjIAYt5q&$L*oIb1m!wy`wc
zZ$mEf>|Lf`Gal%51YtYgiR`=To&4|Hjou*s;aGy(saC$e-yOdM2WPdWeN9*Rs+n^z
zBd%`0!4N|(Ei{FyPb@^`CKQ{5Eu61PR24*h^G|r>nO$Z!`cX-uDmCH0Ogd0YvbVN>
z>ESRJ0V#eSbbJ>#^c|Dqd-OtP_Zju-Y_>G33O5;z;2Fl2kr`Iu&#So;8cdwi{_Rk?
zN$LZ`q1vqjUDN%YU{V9rvhwUtEZM3FVkM=i>k}Ax_WGn{$R=%zMnZK}s_(ytP3opl
z8j1hc;djF|6NIsiD*Zq{)+e2+koI(0jMJ_EYuwcmnzVwV|K@p`(M?_mfz?yztNMqo
zT8B2PcUT6Os{sf9un2C)yF@!2e>SV$KoFZyjsxqrc5;BDlHtIT2q~E;Mhz!zX#`!z
zr-+qUBN_4TaIgGwX%B}`xNG?}&xodR2z=i_{<uc@E3A|Xa|G9g(IaRUn+m`Uu{JYu
zTAW?zC5w^Rb}6;7*=9Ie+b<gc7jWIJJb*~`V{x@k`%dq_={X|1Kk;$D2FMt#B}ngT
z$(u5B&t$18_vA)g-Fpt8e}mpHr<x~LJNj2Gg5ep~YS1w7a0EYB3sHC=UkDQgKUgyV
z2a4Y(C9w(}y8cc}>>xGw;(~=~jU81j6&b!}`tv3_6He#se!PU*N&+3VFUK^AV_5D;
z_p6^AjL;+l<EZO?`9SGFsX=*MjXa+d;IVTb0WsT~v0^M~hS`NbZbZ>%+7aA@U_yMJ
zn$wHLjt%$4*E?!4{~yF59Wf{e{93(qfnA)_7&KvxO|vzw#7Hgz-LL8{@!DO4<o~w6
zTM*Nl#8mItD&A1D;^(4et4uNW3sq1m?<tsgiC9TByZzNYcS4{s4!=968tstI3L*nm
z-lUYYNtDf#*yzrxaVELx1B21`dmB80OaN)$HAHXb6-^Vl1Ik~&`R?f<*GLVjwlEgx
zefa_)k`xhCwg+odtrZJQRO7OkX|^M#Pbu*`(=K?(_P9b9NVs1f|1157W;%rL23Q(`
zC-;3jIJAN=F-T1?pygk3<zE_q9i;&s$34O^eY%1pqXop&O?gy!<)t;KpmroAjpCsr
zZ8RhGBdPJLzsuw(p3Ude@5MeK;568Qq8tTU$&~TkZ-s2sPx7l32518U(F%7It3zq%
z4{#CBfxdpO^=0}MVCW+>U_dFG8tsof_52WNLsMDM0D%yZHBDKn?P)4vHaq6eu^4=&
zu+k1n!_hJSM{;zmBnfUMn;|iv53cnY5XVyUe;@w02h;HjX6Y-;HYB3ZXnU!_8oQ#I
zniMFHoORuf!|KyJNTxGix^6?QvR~7&FpO+oPt=T~SA*8VQeIxKJg%pB^zPp+uYIYJ
z0`J+-t~{$YW9&Hg4}?cax}PQzfo^nuRr0@%<;`yBDpXFs`(Hs34rVkUvJnCoDKQRv
z26ENE_CQ*4MnhqG{xp29^kBR>uN%4K&U$@+6dQHWCXD+6<vad6|KJ`Aj1BxXq=yey
zp7>ApOb>zkuLkfP<NBi~g?Cu@uEs*eFnwvJ^Pg`J&y~U6+|+J^4dOyPw);$8rL_XT
z24|3%W{9K~>rQyC9->uN?t;cDJQVkLc(GxRo|inKLBHZK?8dxAWzllkK(h+yuUgX8
zH62JDf5b2iiD`sX0*D9Yt%=I8rkX-tb78|@J46i_@1Nn+@^LJv{`Ev;_dmyPP6Xsf
z=Xy_V%+y-WxL>3MmU;>Kej?x-+2H(tCL&Qs2aZqNz;_zk-CuG99)rq6TY}yf)CCp(
zkAn^6A48UD08dn#_t$xoM+!8N@7P+78zwXvg5)=)fx1b0W3;+{WaWM-Y6UA)kiY$Q
zp@mTw@4pDkJJFPe7SjP^V(U+k!gyFu^^;q-7|9uBKG3;&g*SSF8Bgw&aDDn?U4e-`
zlh`|KCwiy9X^*Y3WAFYR_OSQ=H_weyY{V3Vk*_NCf!^eaiKpOiBI}xPuHUWFj?@4&
z5P}VOq!k3nxr^j7f)CVtg0-Nc{u9!d$F5=}S=F|?L1_71id-sC8eA+xsm&-!Mgu|{
z8Ijiu>EP1#Hwr;Jp@W;9-R+3Rdz8$ld{-vBKd<ut%zo=nab*=pa|vk4qWdrL`0qP5
zYUtp&VR@apV165;I*DqLuP#dEmY_Anl#~}aEl}eD+*1z40R!|}At&Rc?DCqOrV4bA
zCIf^m@}Zbra!3147ksK|EZA)eRG0e2xqrjAg8ilgmX`lXEn%0diaF1IJ2fHfY2J%A
z=>G~5q=2+M0<qo$s7ZXF_#-hK{ucy-Ru;eg_|?$xxI}Wh)5A>kBM6hWvF?X<7xMqB
z>8it`+P?M~W*Cqdx{(+ZL_{g+lvD}nhM_wIM35RnngJA~yA+US2#Eo%q?B}rNOyw>
z-*A85J%66_Jo~J(;yv%)YwxuTyV<C;mGn{L_xtSpH#rS5{Xd}_+$*-m7YojwjeWL-
zmwr--sMo9dZl)LE=Q-?OVzBb&<yr2CaA|6I>!|<N2MTS)tg#4ellLh<=Fma$ka)#`
z{CB<1O8TKSB!j7oCMvCEo2OyL*8{Xi-iTj_8En6Btf#v;Mg_sL8PC5IXAZJ>ZI(XB
zjaYd2NyQ%|C}1%oi1bQk)qZ7yS{NVbyg1UonLVm$y`*PY9k}0WO{LFzy!akRIe1ot
z7d!DkoP5qGsp4}M3CVz(X+|wyuwRYB;d<|JLSYU(Cnuvj4IH(2+k+M0vm(z}&V;cs
z<U!EEy@O8qTpoNTF%2x<&C<Hx84{W7#EON@7IUJtu?ca*Ixek~5@}COjx(W+WjvY6
z3)tjj^ull`FB4xrIcK?aGqyMgrOwyfjea3J9|zD5A<gm=Q-zsgC;>M^h11<|g9^#X
z3S*@Qm&>aiuX7n8--)4A!PermA3C%qG$h@K=FbnbR5Inp%0AKxTIss6DQ2fQ77sr$
z_AGE+{;De~w=#oyduMs()_AUtM~;0#u-l1UQTSCd#-UCn&R&mv8#X)@=!s^_xEqMm
zZOuF7)}6p=0%cfjjVM3=%CK&<AZO#we|>39ur-05ikgdwR=yXv=MHbP<Wmxi>iONR
zqq96VzBo4eq2x<JDH__77@VY3{lJ#t{y~JM(_To-`1GK6=39Q%A0k-2ScsV$(s+;+
z6>qu(E(JR&aT*-I0LL0i#!`#6J%NP9gjrTCdWWwCdgUxXoHx!9BmT59lz2XVZ_N3A
zi)3*xsmANE+!uN>=}~1qI+5KC81R3gk=Az#Uk3``H*cQOJIV^mzwRSr2yc4OWFPGG
zT?;}6?nFZL9o03L)_qu-UX1GAo7Biy!fCS(ZpVMRL~9&eGT<=5_9UuXA2a)3%)f0v
zy|)f1lx(^n&nh#{Q)-U2w;#PMHn`LJc8a(bn|RfK{lRN3&XsU%g)z|s+ggE3k%sa-
z0Dh>CIQf4Z?h>=|q3{5Hk56;St@Js9zmxA#>ZbFodlYXz?M%LD)2<TzS0^?;SUR};
z=Q@Daq6|*~hUP`>pnk&-?G~9{r04$6bg_=*L66NRTx}xQTwzbT3e^{7$V8a+1pafk
z+R1q#JTF5zRgJM5&@^<0wo6c@<TgE3wD?c@EqHkHpx=@hhX7s0o8?H(V5ew{Gz(O?
z5g?8tto&CBJWHRmA^8XUap8=Nk)lZ}yGX=DSXN~BH)NNq`M@pwh&>RC40{{{Qq>w`
z2KkU+#gzGFMRoz-S4)q>Zk>g7!3I#vIWH{&|HxSu7Ml>(kf3EypaN~}a`Je#z_9;D
z0g2^eM*;9~CW<$Q>t9UXrF;3R?b2#7u}!e^_BE(LOU2`%0`*0a#W}beQw1V!=k+ld
zepndpt0$-09wBn;%iAH_>r9Orp@t4p9B@-$6WWJQd=}xl+s<96t{p%wm~0~S|El5D
z+jb0sAx}?FjeC20TR7osKto`vE@=4GK;rhu=H#Pk`4#p>d&Ja#MnhERg&BP-XlnA!
z;a$~)^BHTqm;r<w2D*=I#w=pZb_O`|>u|op-9CmZgC(Q>2aK+O*q7{=yj`?M!_RVf
z_$fXAHqC@o9cSWs{Y`Zk7S#3qM(3zCwtzjoCyP|Wae<*_{%6*g+X<`_Or!`GhaDRE
z6~}D1Ewvm<gPCrjx>u6#5(w*^Q>nF^MY0BO9wqhlchs_A_pigk@(i)(STZXyMcOA<
z`zJcXCiFKO1n_q<f%oBu4vPwi-z+o#(ts{O>}pQ4CUJUBF%5tRo{~Y^C>CnUT@zoC
zC*wLYTKe1q)zYN1YfzbsB@&E`E!o5ij1P7kM9Bpg&)Y{g(0S4A<IS+#>RT8sl%74~
zYmfYhbk{n8Ai7y*swJw_<(L3x&@P@JQI*91K4?N%rp_w*(C<de;ztT52Vy=rETIBF
zytf~K$&mi_pJ1?H3J=5B*O-IicLc8Yc<keLrl|GGietZQW@N5O;0qA_zkasEg=HqB
zMgcqn2Q`6C#`0o3@|C5X^%eOQ{Y9BY&?2sp@YugD<KK0sF)#-v%qaX!n}jCbPpcNT
z51lIGZo^AO>S~=l*Y=w|YU02as7L?19}SfO<5-7=(IZF}&Pevf_oNZU#j5Ps6PHn$
zP2w_A^|-~CRmGBf-t$oy@pL@rAv&KQ9Fn2KtUyc8e5Ko|i|6LNABC#0ORCH7SHSm{
z%H^CiW;(~TTidiwQGf3kkmlbnUGp9c-IJFqMlN9}9d?*7lw>K7;1pVK9`OWSpR7|M
z$1rYW`i@_X5z!YnI~m>{j3@D3ttVT<8-%~Wb_!Lmf85=rN&<gU_yD?tPx8$A&fuxi
z>ONR-8|Mz_(<9z{mii`weU+eA+&(^#C-Ks>>jixZ*S+KlMnZke`A;QOB?b>&wqd%%
zJU24;!KU&43_aa9$-NB;?Xx{y<2imc<Kj#0zmmbfYy6<uV4XL`S9ld9>eLzf3X}#A
zC<fbF=Q3gmZC5{>>df9r0!UXjLXpBAK`h&6;j(ou7yvyZ<&dE9Qeusgj8iXbj8<>B
zSsmzX+YA#@b&$~Ed`ckAi+uL=4O8hHx4<LCuY?#?U?xGsG^KFLjJA_qPj%W;54I}k
zQwq&a6dkI&nDdb<W_ZHZpXig@p>=tn9M7(j^RDYVd+F8ki6VNh=S|DTt4G!$4d>f6
zj^xb(>8JQXEGq=B*H8TfE=8@s<&Rv7naT_#oB4iB+VumQUGq|?&Qefe^W+~lgn7rl
zrS%R+`MAaW4Rr159swi<sBh{dhHE^9cgP?z-BjeEWFf4U<~5ZN%pIT4))st~i+`{2
zVq?A^g345}y)y8cWV;L|L~ss3H{uW(Vb?hP3>G46+3;kIFRlF^s0bTSJ%)M#UKr3X
zplUdiw^QZ$x^H#48d0zSTC5(o>d)FHPHXLntBoSW>XAH5l{5qleZ3Ab!>{C0&HCZ-
znB1v}H{&txQpC8NX6Ngvq14R`da6#)UQ<FkzfW%4<~>rN6D`{^5MWX7M1Cvq^3a`j
zo>!uN1CWAq%DDSnf_OV|WR(;#=3_#0qHj89dBxs|k_aC5$f+-_t{fe`-Hm7t8#VF9
zk7qKol??B`YNWX^<X~(%DIuYGvO7i5YzHfEEz2(P+49OCZA&lucBEWur;=;ehwun_
z_Zih=pwJ$2zeJ~hZ;k=ILxQz~@|D{UFI@L@K7H_9+Q*14t3MJlld*A<9zi@rA$bQW
zkawfhQY!4!OuWe#1;d@PR9-RMyWMy(dv`V>cs6Vh=_DonF?c45a<OOo1K$MiQ{$b&
z?!gIkpBi<SmfsuNU|~!*)1yI30bZqF3%q3tYh0PrN^Eo=>eT744srB+Ds*6b>Ru?b
zm4)*e=hdW3uRTuc3pCADpLBY)l6FTxAZhing=&?RnWlnR;b2PkvrChTqv%Gxt+Z*K
z$CtwE^giH71z9s?&tjLTD21$>ji|FXc>P?l1ER_uUZg;B1IM6pztP#Wh3Q3%P)~g*
z?Hd3E`FSHLoD9t4E)D;5!X<%9{HjV787n>E9#`4#>Y?hNz@1|J5-A1EjQ}tPrb#6>
zkMQx}<T(@}X2|f(ImfNV5>4V&QBTuBu12b?7o6{?F_VvU(|<%89yY8i0Sia9-1Nlq
zxz#j0M~`x^C8e1OKyn~CwnSnx;O!-`&UAQS{di-ziMGy|>r?zQ);|k5Ux==Je|2Y?
z12?pm?#b+f74p0NKg;qTO?&0?9ake#9b^7^r&BMSgo1LAe^IJg@&MRC*=JkjFz67I
zfLg~nG135F8M-@_E&|V0vg&YB?o^A2U(5RmyaFl6|M|Qtn(MvByHHj=-M?A@=w~sP
zybaps{*o^7{naL);jI5Yt8q#2Dt=JKAEm?+mKEB_??gQ5CU&VmWX@h#JZsJ)ITlIr
zf@1V|S|Go)p_%a2qBgJAJ@H1{7c~bgQgddK$M39De~Z-MMz7HcFlT?0@s$ES30%rI
zOgZ0_&XeFYhnus{lJup_5UmQec`-%bN7S6urr4b5)<*Rwdx#p@oMxy*6vq7RO3)Vi
zC;5UH<F-m`J8Ym`^Hq;_J)H=o+E^`SX&LE%%E}ls2#NF5lY0D^$xnB@Qv1Ls`5cxV
z%U?&0sU{MKJflEbRlq&$24dtUj6SGvkdY;eb6zhJn0L5#f%UM9Ek{8tmI`o>pg0S7
zTBNGgvqs%%6pkRY_G4Rut%WEb6hQ3Pveb|ZXXc+gafiomx;NWi?E#Go)SLy!_<Dc#
zciThwpD-t8bxFN}kuHpMZn^ya)FKMXE^S;Ax4xREu9kyxw<lJ|vKJHz3-UX&F3RfE
zS31oO{BH89{AHxKF@KH&5TE9lOU->sDMnROhLiSd#`o5~<y)eTOZ2?AxG-I@+jX^B
zc9zH{3NR7?`8^XkTqQBFY?D+&M8wM_RRy+e+A%ap*Sxi%F2HnbylOq>n)BfKQ({dO
zit->*C}2AZ=%o*brcQO)QCOkt3yjNs`1fJWFqW6WHZrvK_nvX!izrZx`TX-ws+{Ho
zV)g7nCJYg>HdJ(x{n3D;zjbf#2MDR*)yehtO&Hp0#5gXgk=B_jq3#1$s!MM^1t3bO
zS?)b{IaPY8-~e=f$ju$5>iJEW<dq=;AgsrKm~_Ei&j{H)Grt^ChDVoViHI82u8Od?
zP>s`9sY0#M?9?V%J1uTt5i(JY?Jc6}LZ3JjiZ47;DYpI+c?zjT(pN7BbF{5nB=6te
z8M*8{>7yxwHY^YqN-lhnwAE}_&~N{lZS&Ua?;Bj#n+=3~^DTaovXP0(bo0)e*wrxX
zC@AF(c#HABOFlUpFEvVNhL6TTaWK%<oi^bRNZMxrd6h8r$%zr5JfjVJM{U%+CG@Ew
zmLg5-eGc<E70xTuG^yheiH+^P4%alHfZ^2SOF<xEtTXn3fb1ZS@U&zdb?Osm!>Mft
z2WP}Y&aZ8<#;Rg3N;zV5@*@EP=N>^pa%1;9{4GvVGb&##7)<qRIwpwP+3U+p#&HOF
zCa`U;!-FS$cV`>?M1R8Izk<g<Euf+OWDv7KQtZ#ft@|Hai|0(PZU*xi%?c^%drd}x
z?|YEAe+vsoGJZlknXfw8lLX80D*UBAlkHPS6k`uOu@(RGD&xJ4d_QSgJolXlO=7<O
zsU}n@3ic#OjPP?3)VwsBq)L~&;yc<AHIn$lXnq2j)!OIdJO4Q@=iUW4J9V=TS$k$0
zuj0>PgMc-d=~<&_;`#L0%Qct&!jE(l-*A{_>)nP^9WCZWDA~L6@}dW=Zv&sSu%PG|
zc!Ia7ZM3$_`)ET)rW+S)jAyVOyn)}xgsBJWK7EU)^ch+sJX9ebQ55Wadv{W0mB&!O
z3V5=uzfnff-A{{()ySp@=S|J&;<3FKC{DO@(%{-5__BJVH!qhSxIj;;8wsz*BfuUp
zC24$xv>7=+PV5!}^_yuU1h2xgujzECIKb!HGa(s5XKFhzu4vTp&c+XJGi+5t9V7|>
zaQr74b1xbkH{txt9|6(+@wt(g$6JBPImo<p;O<6re4j}tQQT44xqD&I$Lq<j8jswe
zucTS=+cVi<87wQL>}7Aka9WR*_EDp?^Wf+@p&Iealm)MEYbl!QFC+c(xZ-Vpsk8XW
zCPzI1^WZ3T#djY(lb9Z9^))sU==_#hR!*F{ITbS?ad*`Ht7Yf!1wLuT>E-eL@t{(&
z#>mJ9Z}DFH-yq<}JL(1p%C~@rJm2tfmj46;)0JjfL9IJVJ=Nb_YMGmL<SHjikneTG
zQ7n$xhID#a$5uHPV*iU>ut#sV-uj|(RyY3j!1T@wRF(X~pc-jNa~P#O{xejztpLQ+
zKn*#%ct(G}-eH*1s!Zvays)RJO+~EibY}0V#@xO!Pb;{PD{z}|g<r3BcJR4{Ti(+}
zL|RP3Mj0>U)5X&!71VAz6rpTzFIQ8%$fxFcUvQa_c)dK?x-O-RT)y&O((Xumz7Z7%
zQ@_zVwYv^=bu7e-J|IA>u`UCe-95hl{7(sC%jEYLC1gttWnY3FblWi+Q<bu$Zoyav
z8prlxY{d@vDY1xV7T9%I&@vC`E&3igIj!E-zIX#mm7Ug%mYy>0EEpqRfd1l)8-ZIA
z2JTuR*m-sNbK3uay-L`sx46q4#bMG7+|Jx{Cc<&l+$LcdA$b0zYleLL5wNf+BB4Wf
zne=U7nr5g>cH+r1U4toCTx2zXl;)bzee64_HgXH;xNxqYokI#PE{kV4KtiWHO1cI<
zlKj*5$`ZJbo{UeQyeN|XUfQztIZ+}Mm(@n8SDVxg;{CG#_(Xypwtv|A`Gp7>mG3Hz
z)60%b>lOo>x2W&~U|XRE(v;CVvF74yHCz~NVV6muTPD)4zieGx`mVO^%`S@edRP#%
zTiF#}c=YWdi#$>NhP1gq@6_Q}nhtG;RUDmy$FwoO$Et5EIwIu*jwe1EJ_v!Kb&o5H
z9++b(?!DM^XT$BFoN|I`4enNqIC%QnP}OVDMT}?dQb6rJ7n!@HY`>Z{eUZS}x4(bS
z6SLoh7mTbazk*Io!q@EhE0E8yz7j0A4q?^xGI+ZW7)Tu}a;j~95O(GMI5+T1&TyOP
z!~<<fuWP~}l{wpOTCS6`XXCyKuxK=w_1Y;A`o<YGAY{>%+s2G)ZD+H@S`F&Y`~x};
zS1r+%R>LV)0IK@CDwB47(1oW;HZ#t^PpJZZ326bBtV=`1hbv_{N0jm_fdRx~@&o_8
zg>~Ca=m?Mi_{W>lWu_Vz7b2O1t`=_QHl5x?JNbrqhZmgRliG<Xuhd=@>tV(2k~~pV
zB|X*c&E@ZtMjci9#OVAH4r$xkYuXV{7O7n7|DKW3^`&0%%6R55Y~mu-Y%{A~<231N
zl+L}2bU))*fGr;<@996=PSV?dq6OBxyULU#r2;sB8urwA9NYLcBY8WIekz%Xf%~{Q
zWX@5fXvW5$#a?%jFLx~V!s@7?Iy_*Yt=c*6;s(m6NY>!<i(D#&^H4{UTL<z=-oTyU
zdW-C<b^8@t7o}1o)>k5Ky&by>%VzxGQpUmDHAbh;{+A+<Y(Kc557a>VSx@uX{2tmz
zoZ~3IvOOH3i|BNa0ndb?N4Cq}f6H3;*-8>$p7}h<*EzG>-WovXUHs%vImtGzGt?r@
zhracJ)LA4(Leu@dv7Gr+FZSpZ+L#zExf|y!!NfLQSW_~RqbOh`@|s?tv9$Q$kL%r?
z1Q`#P!q?W<;gKF@P~>3nLz}lAPR(Q?&+Vr6#5nNV=dsONWV%5s`h94z7@gV;l{%;k
zjgC66Q0x-=h^I<e(W>^~H2`xj8!4&U&3(m`@rG7f%?suf;mFn7Sj1tscZpu~$S7bA
zo5^+g7G+^k=?iSs8X$Uo+I;*qbM4s3cJ~?ST|A5Y$t-12&x?#>#2+`(U?{O97R>NT
z2#<z2-)zvNuC&ijA-74;Up!bsp89z5@;9Nb4Sflw>z$mb&0~$bt#Qe-YyIa{PFDkU
zzB|+qvvN2VwTXDqM@y$KW?~dd{sEWhsEib4;QjjO_{7QSkkaenQ>6p-zg@}^f=rZO
zP^Niir8qqEwQN2w2}<U0y~j;Aq&allZ81BK(H_y_6uX(#%$6@Gk{qAZlVD7Z++SPe
zTMzX7?INaG5%mCu8MkOmCW7zQ5o-+7?3UNJ2uyp~-mu<k#%k<<q2{?uAkTCoCE9M$
zW@^;F4$c6eILjjxFJn`%f5fL)@kvrMp^N?|A;zccR1h=U_=mbOwC0c2WQ#Ilxp&?f
zkAK!-e=(3E(a!Qzf;-LHRVL(T?;O()4*2H><w;Y1d*ojF_onw@$QXbM?2C`IXrsib
zNDM5(uZIAOnok+N*b4`?ZOEUQ)|&YtYC^-eX6JIR@{&tyZT>R0#@YQS);{uKdrBNY
z5~I5<3$5`-hPO~?iKMa#o=IiX>AKkKU!5Qlc=q`~0T#~%^m)rsat!m}AqAwl0z^)+
zLnQ!=yp!wY1OKCXpJ4q^JipRlq{wQxQ!4NcUU27(JXIJ7LfVYuN`Uc3<{RRg3I99K
z<}Iz|Xbde#SIUR-OO48^1Mku22hykO_ks+imng`T+1dTDvd8*e00V$48D<Sr{DBx`
zJE6>vPn)7Mv-Qiv?r@gh_W6*+*;XZuk#rmFt3V~Ui}w3K<_V)Hn-QJM{#XXvyO<^g
zU>C;lBet$ifMlBNApzAaxKnWe6olhSjy$6M3|{g&ezPh_vD9vvyP<!MAeR7ls_qrg
z8uiH01oPbWEY3tG)jY4pgd3cJGI~dTmdTO2MPuWaXOxEuq-T9|<ykTERDCi<&Fg{f
zRsCg3-ju;n4N<o4;hR$(>K}CuY=fWS%ZhE?qs1#vfVm73QERrZel*BC2)g97E9xGK
zl3`STbXSn$YP-53;MI{ivaKs^=0d$-9qX5qF@EXHZDsEVo?UAtS~Q6K9?g;#O3yaN
zPwZ-d^00Nf>eyY^Z}zwz$Cg^9`%zDoE%h!+M%))P<PK>RW%lseihNI(mufOiq2iF^
zDkwb&S~q2#!`-UR_~>XSsTQvfZBMIc$s@d)p2rX35g2NT_a=pynP&SMl!hyy=x1b`
zzqFn@p1V~H%b6jRGkD)B>$;pNZkV{rh1Zqo)jzPEmlhGE?pr`xjh@dx-km$2UV*L7
zoMdjMyRQDuJB-q>@X^dLb$&tgnEMfa`_L_m!V-}?iVV!~eIrBVnp%O0ZMLmTh|GB?
zoKshZG=!17%w9YSqYDCP#faZ?eNUl)n5mIrnJ|P^k<lc;7!=4HH1{wR|0dyUK{N|~
zJK~llW1nrW&w4W8cj=M0+MM-LKx8H@EQxR51$z)NAUTdWLCw*#;})pCd~CYEwuF93
z8uAw$F?1&giw<%}32g!ka}0_8_XpBh@O>z-t2Z*Fp~<elryXPj`l%;UFIF=;?BiJQ
c|J=PH3DqG=n%uYX1+dq1c~!ZxXNaKx1MMff)&Kwi

diff --git a/docs/blog/posts/roadmap-2024.md b/docs/blog/posts/roadmap-2024.md
deleted file mode 100644
index 2832660fc..000000000
--- a/docs/blog/posts/roadmap-2024.md
+++ /dev/null
@@ -1,91 +0,0 @@
----
-date: 2024-01-10
-categories:
-  - Roadmap
-authors:
-  - remilouf
----
-
-# Roadmap for 2024
-
-Outlines is not even one year old and it's already gone a long way! As we just reached 4000 stars, and before laying out the roadmap for the following year, we would like to pause and thank all of you for supporting us, using and contributing to the library!
-
-![4000 stars](../assets/4000_stars.png)
-
-## Thoughts
-
-Before delving into [the detailed roadmap](#detailed-roadmap), let me share a few thoughts and explain the general direction of the library. These thoughts are informed with my multiple interactions with users, either on [Twitter](https://twitter.com/remilouf) or in our [Discord server](https://discord.gg/ZxBxyWmW5n).
-
-*Outlines currently differentiates itself* from other libraries with its efficient JSON- and regex- constrained generation. A user-facing interface for grammar-structured generation (it had been hidden in the repository) was also recently added. But there is much more we can do along these lines. In 2024 will we will keep pushing in the direction of more accurate, faster constrained generation.
-
-Outlines also supports many models providers: `transformers`, `mamba`, `llama.cpp` and `exllama2`. Those *integrations represent a lot of maintenance*, and we will need to simplify them. For instance, `transformers` now supports quantized models, and we will soon deprecate the support for `autoawq` and `autogptq`.
-Thanks to a refactor of the library, it is now possible to use our constrained generation method by using logits processor with all other libraries, except `mamba`. We will look for libraries that provide state-space models and allow to pass a logits processor during inference. We will interface with `llama.cpp` and `exllama2` using logits processors.
-
-*We would like expand our work to the whole sampling layer*, and add new sampling methods that should make structured generation more accurate. This means we will keep the `transformers` integration as it is today and will expand our text generation logic around this library.
-
-Making workflows re-usable and easy to share is difficult today. That is why *we are big believers in [outlines functions](https://github.com/dottxt-ai/functions)*. We will keep improving the interface and adding examples.
-
-Finally, *we want to add a CLI tool*, `outlines serve`. This will allows you to either serve an API that does general constrained generation, or to serve Outlines function.
-
-## Detailed roadmap
-
-Here is a more detailed roadmap for the next 12 months. Outlines is a [community](https://discord.gg/ZxBxyWmW5n) effort, and we invite you to pick either topic and [contribute to the library](https://github.com/dottxt-ai/outlines). I will progressively add related [issues](https://github.com/dottxt-ai/outlines/issues) in the repository.
-
-### Many more examples and tutorials
-
-Let's be honest, Outlines is lacking clear and thorough examples. We want to change this!
-
-* How does Outlines work? What can you do with it?
-* What can you do with Outlines that is harder or impossible to do with other libraries?
-* How you can perform standard LLM workflows, for instance Chain of Thoughts, Tree of Thoughts, etc?
-* How does Oultines integrates with the larger ecosystem, for instance other libraries like LangChain and LlamaIndex?
-
-### Simplify the integrations
-
-We want to keep the current integrations but lower the maintenance cost so we can focus on what we bring to the table.
-
-* Deprecate every obsolete integration: `transformers` has recently integrated `autoawq` and `autogptq` for instance. ([PR](https://github.com/dottxt-ai/outlines/pull/527))
-* See if we can integrate to a library that provides state-space models via a logit processing function;
-* Integrate with llama.cpp via a logits processor;
-* Integrate with exllamav2 via a logits processor;
-
-### Push structured generation further
-
-We're just getting started!
-
-* Improve the performance of existing structured generation algorithms;
-* Improve the correctness of structured generation algorithms;
-* Add ready-to-use grammars in the [grammars](https://github.com/dottxt-ai/grammars) repository or in a submodule in Outlines.
-
-### Keep developing Outlines functions
-
-Functions are awesome, use them!
-
-* Implement a CLI `outlines serve` that allows to serve Outlines functions locally;
-* Add more functions to the [functions](https://github.com/dottxt-ai/functions) repository.
-
-### Serve structured generation
-
-We want to make it easier to serve structured generation and outlines functions.
-
-* Implement the outlines serve CLI `outlines serve`
-  - Serve local APIs that perform structured generation;
-  - Serve Outlines functions.
-
-### Improve the generation layer
-
-* Use `transformers`'s private API to prepare inputs for generation inside the `Transformers` class;
-* Support successions of model generation and text infilling for methods like Beam Search and SMC;
-* Differentiate by adding new caching methods: attention sink, trie-based caching, etc;
-* Differentiate by implementing SMC;
-* Implement Beam Search;
-* Add token healing.
-
-### A more seamless integration with OpenAI
-
-* Provide the same user interface for OpenAI and open source models so they are easily interchangeable;
-* Integrate the function calling API.
-
-## Last word
-
-This roadmap was influenced by the expressed interests of the community. If it doesn't reflect your needs please come and [share your experience with us](https://discord.gg/ZxBxyWmW5n).
diff --git a/docs/community/belonging.png b/docs/community/belonging.png
deleted file mode 100644
index 7346f9aeed55cb0b72aa6274849599fde0c153fb..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 142520
zcmdqJ1yod9_%|%JA|e7(5=uFwfPm6ml^8&9kS--fN{~(glNN@KL1Cmpy1^i%r5RG`
zMsgU2VZJ@ytJi@0{@44xYkhIoa^16-bM}7X_tbu#bDpc*kR>@zdwk!%eI)X7*I@hh
z9R~mQe?58_{AIplngaZf=mAXjuYK9AXJ_{9W7;QwP5PFz{`}C9C|#3E$*&{x)~?<Z
za!1eEpF4H%ps3uDnMSn=$DGQ<UcVJtDV6MA9}Tix{cz6nuR0FRR35tqz3`HJ|D%w<
zTnARXT3N{1^7G5hOwAk}O(pBi5b^OYZEnM>82XFj<x5yo5%Ax8>4*D>_8&ZYib?8^
zi%;@Q%yP@OcKhk?oBfE7!o+BHe;C~1&2#V|gn@*xx+ZT(=6ckZ-8OscN+j*~)O!~y
zyop#pQyn{Ut8|FVTOVxYs+4+Q4^pKPZyq`b3Dnj?U;l%Re;8ZSmI~>x_=|o|b^<w=
z<jlJo%Xs(Pqj$d#T%UY)uN_X^y+tGqO;YdiAlpTn-^i#ub1Hc)M&-yqGV|6rBy#@z
zfk(eFf0r7p^8wxwxF@aQV4Yjqx4ghQ7i8b&yopSjJi80oYE<C1_`XBCs+XW4SISe#
zT^17ovA<F9gZ|XAhfK)X8lHcI_$NVK8E^NIz$|4{a!506$$vce%Y}Cp1$Ck{)Zlda
znZ2;4k^^MxniaR#5=8rd;YEu6m?kbS$j>D8H}pvv96bu#`U@qvC(C6NfMpZZiI}}u
zo+bv;H@HxKcu!g*0j*_RvC@06OSIqP9FyD(u}SYQQuukfS(sD^EYd{a?k-CE#!(_K
zupN(=^R&P5_k(CDz2iq=vw4<>mHvQZ$Wi-i*A5^0wqNQipg+HoGPb=GII9CF8Ce%_
z-IEeOz%9q<?4I&UwE^<-=xv)m_9r^N{l)wI{_uWTg^<zZ^b#o~L#rRD;?v#Ovb^w_
zcU6q+pexYOnp`r`&jR~doflPrtsf{{S=<w~(>H;wGru^z+b#r>-bw-7pXh2VxA})_
zDQ05LsnQo2bHhJx^eF|fE-~TAx2J3pfR@&8aT~GjCEl%DfRdC!9+^M1^v@GrVL(9%
z=_Z*!(Rxeft%OLpZ5Qzm33wARG0D;FyYA@!(`~<qTgnB{x-uBIr~f;#k4U<4--SFK
zmc3|gBtMl5pPJ;_i`(1#fNdR?dNFkV2VK8is3ZVp#DvZM@PD(qiOlEu9Xv^&D1Q?~
zIxmx4T2(IRFD(4PHn?yA&N!{}J&n+BJN%$s#RH&jJy37A!13*wR8??8{F&A_|HAST
z{GrJfU3#4LlfSpRaQ!XgBWQa^gcLW>W1)(Z(q6155<;27O*Y>>g?A5_eA?qOi_?4A
zyu-Oq)jkwWL_M`vjY0@oRDlrT>op+qxIcuD&I)dY-y1@D{!9DsH<<ukG)g@<{<nNS
zz1`1tT=2f_5bdw>x}XQFyDVDZXH@?XA-M1+IuCHg^w=?WPh8CaTy?a>?H0PfN%6!f
z0<Ith1AB@F3Zf9)dh*0x!oB^FfGfF|Lu|hi&-CV{l<YO*#P6)To;sE6d~GjQi1zD&
zWyJP$m;nBMW_wv7&_>Kd8VJ(G{7-lKkGxJW5%}aN(%IA2p#=MxL<;^5KThwUXCJ$|
zpOHlTjBA3JKUD)T;W8LxuI@$78Z*GIB*&AN-&sG}75mV|U|pSW?C0I3(tvJYwFT@y
ztMq3X9u9cOWbzi?AGjwfksyjdDLWH>r}5j09z3_TP!jA;();HC#SDk`Qs6@=s{K9`
z!9NE~oWfgPosAq<YLXxHcoXpetbHbZIIt(ZWgx9_7Wp4)*xQrywnvhu4($QvpZWJD
zVrG(Si0C(forq!|-iveZ(|{w^+vd={RB(a|sQYbtarw^z|3ykr07v*kB;*H!gbT|H
zw_h47AN>9|sU(oM(pJ7-4F2WOqo-06A8x+!i7Gq)jdTZs^YG`VVfp^tZ+A1P5ONak
zF1@hdfB4RMD&Yn_cmva39>ZH3SjWAWi!pl|I>CHqUnw4z|5qeZACC3qlzoBDKKY>&
zT2&Cxq8uDF_FU%>kSwKC*WdZaF5d7{6hLiVW=8qnyzjCO2MBg)`(E+<*z~sx?>dkn
zTr$-gqTEwnL_i#ePC4x9wJyB|k&16_Hv6}we=O*I9EgMCVJh)npikr@R2*6wPJ2e&
zTT;OJ5?#~#VZ^^*1VTW?(Og!>x#u4HfvAo%&HDUMyOjBUda@Ug1&XX+6rIQhZj5<T
ze&GlGzg+;I<A?8vWkPzXHQe2kXHOOtF}d6hDM^kO|GJ*vsV+Qykr}DJlCAJBPJWR0
z?PARh2)~;51<{^fr|>k$!J)$rzwq$Wa^9XGKbS$Y@=N_DnZ%q!mW%_vquH|G?%v7)
zTEzFf=3t(imfN{7y$jlafq2^;;*VA=O%cgC2~_5wSzz~+>l#F()nymx9X-{jA{E7d
zKjA2F_Nj&YO!iTrxOd}2g`CJ6s?>*FdqqeQ7i0bP+-JgVSF@aIJ_I)hL{7H)Z)|K-
zmA=XNU3}j*z5VdmB|*X3M@seeH&mUT6)&t$I@jNPdh()oU9}!aTe?fiQLK@=8^N2k
zIbMvst<gGWP9`P)(o1`dNsD8k=n)A_vG_a2FDT#6e;!+xBDwtH^47;<CzQ7q*959(
zmP;C%HQrQVts&j$h5wpjyK{&t9F@_zI|vd(_v1b1MaSq2aPI9qhalnbY)=Cp_$-~q
z{PEhq;Nz7`@wO4CkcN#SzgVxg1m(rB4CPY200IKG)R~T{R_%n3z0$mYUU6nXdu{Tb
z4}5r9ZFz7p;%g&*zN0C}Pqhk%;|RLe`!7>JQ>2ppHjSO0QI_8J<5p)mvC1RP2@Vak
zm^9SEqiI&$>RG70XJ7kLAv;v7@y0Kde8;`i`k_#9)$EdSWjbBCD3}KtsxKV^lW#e`
zrUrdxC<Hz9c6aCC%|T4j3YF(!S4`t~JycGWeCt$xv7STW5vT^vbDeI)Z_c^|Ssdvg
z@qMGeA+JM3_iSyTgPQu=pD##%eu&!$V%$L&mx+B~;9$MolbFVYJUm^?qecD7URZ$j
ziX>+Ht5Zc=o<1(w^T~G={(@ic?1P~bl{=oxE<>$<$+vQ!xHTW$A(;RHnL+B+E9ap6
z^Wr0R_X~=P(Y~S=5B_8*jZ_!M65U13zQcC*YGPU%WU&-CJAd2lxa=bcF6XrmoVJ%m
z@*2d%3vS$(<JRdd$WYm|dp#>f>pkP7Kai%??*Vf?ZqT7?5W52GXkuV}t6BFx?B2sr
zi|YDV(%E%t`|r`{7e0vgcT<(1gDmo>AP$mnUAH^Xb&IZU%7YNWyd}sr`TXSyCZtU5
z1FfGp_!kNA3n9cXyOi+LsYg|zHQDzcNJAe>ciQ)M+UbigM2BL<P|UTvN{?ti&LA;E
z9M&=aJanY48D5oB$b<}TDJQ<^8uJ+Dyf`=FD#sW+Jn`>VKO1V|S?_Y2@_W7}m9EES
z%ZN1}K_+e`Za*M!9qw+b(c}_No5(_Sqt)euz74O-i5tH${12wRtEe`ovhbRW$@Y}5
zlC>6__zgS^M5UpO(J3T1%XE4;F<jLM;YE+TjDf=@fDjcoZ^hH+dp)UT{t)B^h0obK
z*@M6<TFXF{n>>!GHgZU#*pI;)Un_VS0)mi3#H^B1^4T$>$|0duhqb>e{}=ueBlgRO
zgr0p&1l4#c*N?2^T2;fPABUw|>fU$;bH4G#GGegJ#&zJ18bjsc?!Gzk^nRyd_u9kK
zP|t^<j%hgs3mkm<9u<UYZM0LR)lMHXwMLOz;ez;(aoJUG4?rOgE|$zJ#rwc7<*Nb`
z(vHJNf|-!+<<@Ong4bpsmmu>wm6hVdWg{=tXe+JvAS85u@z8xO;6)vdtA+}@#>Agj
zOfzPME-8S5M=RLOzA#S1&M>zPdhDhh7RfS#?#HJx)W6!}P@zlzbI%-a%V9Q*!Q*q#
z#VJDtXXH@aZ~`Qp0Gs!(kxwpQzCd?guhd3tLQP+e_p>9GJl?ib0y+I6;2$w8k#>Kv
zu2aL-Db^WDFGuK?C#$7zSS9byEIKG7)W$py!w?{_tBXorYdFM!8KTlz%+S0IqfQ-f
z`>VkQUvyyP=DYSezhA@ZcYLO^-d`=Ga?*aE(Kin_Y;s?}bQO6~4zlr}aM6NkjZ-d7
zCA<@mOQZmNgy+Y+HtD6zPty*2)EeCsB{MPpc|%k@OKSI`LLHy;-hob}L*&vPnt;c7
zR3Jjw<gfYfX&|KtwCMN<$HgrU`xo*;zG;6|mk|m+Z=!Rb@iK?6MGyUwyB>~vOQM0x
z8aRc#e<kOAG(_*v1#{t7jQ2lNR@?!J*~O+~FuSSoCAUyEDf#{>ANYZj9!`g3a(+#=
zQe}y%-Bp6s1WXC8f%@bAUOi0Yx->LT66z$NdK|6TFrv)LcZm~egJ1!RRwxYJuDUOu
zV<<@#`#6;G6$^sVxi6QJx^j;^GO>SuufFZO!(Zefcc8f~Q=LyBj|J4zJrIF?)lA6!
zYD(nE6xNqUI-jgOmWJMW7V7T#>B~~u+v$Jf#wYy2?F1Km#&fcN4?L#h-wJx+AEanq
zFn<qU6hiREm4?#YaXP24v}kxcfm<$3YN-4;%&*eGQa)!p1PQdhslPaNHOSrOkc<I_
zCx+SZo<hx;7J5DrjOfE>nuBC~<=o_?#WDqIJE5%WeS;Ky4jt)HVx~#~3_Vqj?wm1(
z8}Fr;KN4SkWwoFqC@K<jUYk89sL0l;UBZ3gc1(ou_0{dkyV2s8`2`saT5jt`{Rt{5
zdSYD%6e16B=?vpiY^aPJwBE6V7bv(NgK=OJ!NL0sDi(F&aXzf$$%RtKrwq6q{h+G6
z6lII&o5e5bbKMkV3xr7h$gXM|YAT%0xR!a>%HJeI+el>~CTck5IkAT6=)0aq{pytA
z*Gl@7^%d6g^Fk)q3mgh+p1q8w7BaT#$ahzh<<dcsaYy%**z~+9xxbYf#pfoctf#)V
zDyXVn-zc2@Oid>zU~)scT`cno|JB|thfGWVB1)_F2rcrf2{g1<muS!Unv!HsuRHvH
z)V3z`HlSm8`by+msZ%8@p_198utpNqGM&+ka-}KmrS5We#SqwQ17*5qGyF<<lF@>J
z&y>fKIZ)3<WS=5@8-3EP{z&6l4-;j&u;-iap9S@q+*egRD}G+)1S73d^y^nt=1X^j
zO`0_7;%b{l-*qq8POd0&Wv-m?=ld`)T5^D@jwei`-gHdcaQy4!-Fw=?YOim@7Z&Jq
z?Ok*#E>{aO#ED;SEpt@l(ybC|PVH$I$`g-ZEMw`hMA7=zkp@aLKe_(ef7zkHsV2sR
z)3Bu?oNpQxL2X-`dRJS0uuOZboA;Mg<R3SIB-XWEE~T>+8qtgD=nge#P+JvFYs1Ci
z#l4J&9&;i`-n|vS8Uiz2D35&}YOlq0t)bGO1TVkhnY$19weme8`10gclb|ZQ6xMu|
z8?-chch<gA3*0l7r{~qpyF`~7S5@Geulm}misOcup}ykT-qS|Hw6{7o&v0pzKbS18
zn~4k#rfJhKbA<8}n=JGB8+~HCYV@&=8A>~nt`Tv44sfwzlVtP;1z|;*EB<hsKQ2I`
z&r|&sFQnPSc}!vDK-l2idvvJn<WqKJGqfQ-wp>F%wK1a%95xoOOUv_|ZCHApQF(La
z0DRD;!A*~Qrdur}v;yHyzn&cozp~XB7xC!A2n~uMHfZRw%n3Se^^_)!%z#p2{pU73
zR_U8C1Nmom{i0KKk8H03B}L@b6<c1{X^t-DzfpC!)9)!n{#o9%II&URYr5-AUsrr$
zdL}XqYX`6MZM7Laf|#iF$?uZHZpnP^sNRvFiyJ?}_V}f|6BbEfEn>i>0y%NxAS663
ziwUV%51kK$T^3-aile|L;PrXrT<UWfxgV@mN^)zp&QZ}_SELaYc@jzUYFy^UBS;&M
zCIv!Ex74I6bfThl=W|@2UH&^hL(Qd#N$s)<LDLp3as=)}jUV@Fx6mu`Gah!*C!#Oq
z&7WeO>MgTW%XGChx7^Oijlj&}lCqBMl5DAwm2CQq0PFC33>+b>OnW#Z1z9;C4#TW~
zG~*JtZW(8<l1Bc`(6$G6<W@weUfCCuLU}GWv}iZr;;^30)m|0N?dY8XRT;+`Pa|1)
zMVy9)#*GL}*`0TJmmSQPD;1#h%kr`<{<xIFaKewh^592LnFsFJj%DOY<XtT5l0F4}
zxnLyWM=IqAOygkhnWUY(o7d}7$af+5AP49DFPLf+NQ&-(5F2VBe=(FXoh7NUGSV#`
zU-{KR!-~VrDR2l>z0F%26v_6*?d}=^03~7Xd#$XjP+wV(-3;w*Ks!|~g#oBLr7U!u
zwr!kNw5&$4RnDb;p077>zj8xwuzilTir~4}83iRB`|>I#>dMN}qT8|qI(g`+1WY4~
zPiSP17|Z9d*7H0k(+SQ>9mz|fs!1i5sb#_<Z*{4e{fv(3La7x|U;4_$>))DyDpUKt
z#FsQsQpu&IE*=*{D_S3)AL7QSSq8&l8_))WGnuV47X0qg(CGPEANZna62%kt+6|eR
zdxxPNLR1OkZBQ39R#I(p+RaIupKc;?Imxi=*_UtBvo=N43$8}2xvDod0IB;rS_QW=
zHw{oBGh($X3IwxDt06b$CNkm1LeTm4)4sB5JvQs16J+S3m~dEo3hZyHnUvoe__&<?
zp5&0QehzjbQEM$LiLTyWj*K@=rZ^1I19-`YxoQt{5;Yi%Dui^D%U&aRR8Xnpo<jC@
z=!<htgiF$~y5JH>JzqDsM|x@J4SXCCvJU2Ew<37Z6;Td5Q#V;RG3@)n&0>w1S6R&o
zv9afOR!~!hb1hG0RaH$Qjx~`WD=V23JUv}1u=CunvMMWx94}pO#y2YP*)_7gNc*}e
z(l2T_lNo6=NJD;Jb#k+@ob&HSgm>ZoovQ^??A|xqWVF9$QAa_y)>{()WophBK9~s~
z<k!cH5Syji16Wh08sq7>S7%3pkHcKiw*dIp=6K}f^D{Lx+>$7)IS8R!H*sZ{)GPZN
zGmct0fiXtd=VOga$*}^A2m#5e!c=b4f{0!Z@KX|1%}G_);%#*e4KK@{^vmm^xhA|z
z;aqctYkf#Q_c^|aq1&Wp++JJZ)GqIiHo2)?*PuSR^!)oQK+4bH`KQciq}uGfBB@iI
zsgcWC<+l(GEBWHCLHB9u2z%{c<kC=#LsP%wc#dp$dDY=&)DZWVa3-K7`L8eBs_(Nw
z@7xd7spIsLguh2t+BLaXE-}}ymttOC6&WQSds3Io*Ny7he*D;fv>C%W)~Gho!sdeI
zQ5ai2!W-BSdotUvW%YqZeY8e!fS1_iyv!Yq#j5k_jJYCD6wm7QV8r7K?zN6myJJ36
zZ{^$(7hJ!xTWIpQaDRBIt^al~TsJE8(OL5iC5Vs!D~)G|vRyKX%8FU~+Lwjib>u;a
z2|NeR1iT|lBt(c9nL#p19J$=cjSVNh8M?%{IZf3_?Xh&jrR`JDX7g)i!{*c#pGOJV
zRreb@HZg-H%TwVKhamM;cWm7!-%|29CP|CyrDwKLfdD00?Wn2ywQa>GWWq-8jMY<z
zn$Ll<a0gt%4Q*Uw-rcJ6<R5G@Zs5iBMy7U={!t35%jjaWY*Ix(?gh0?{amH#)yS1b
zQWZZdF+{+t^8450JHV-LhfbCaY5)h^P59!*b;#gt3CtBqIn1Z}YRx)et$s|`fuXQ}
zMD4P};n#2RxXF6DgOJDX<a6&MyKu0T4`JE*u`gg-tL$=+q0G(>J$Wj<78s<c-F*iX
zT>>z>m(VZsVT1GV?Nh8SOCJ!MFq)*LA;|ihf?leV`n0hc=H@vlpgLvlh3@jFzieyC
z6w+M3bo;NrgYTt^w^yon5F`;BP@k~vuPiNx*Tk*Fc;(VsjacL$9|XNhvj=Btg`$)t
zI0qS#8K+pil;PV_tde37cN`;P3;J!_PVMtm06UY)dR=EoHw^PJszLgva?(0XR|U+n
z{5oALQQLXcRqJ(?oY$z#c)eWj1XSEkZkfz@m7%(eqn{w&UW_y|_6^eo*|JD6%KeMX
z+T++yR8Jrc&&vAQk@q*`cTV4)7-@Jbsu}{5X;C179XCT#4BW2D0BP$N`JFaK+?qxc
z0WIc1&#J8k8FX&+Uy9+k5avtoY9DTAXsfy?yt5Gg`YvaF=j7{W)$SXmeXgrt+PiKI
zOK|GxmHlryo<yhpp2WpYo~{4%-e<49WptfQR(S#r*u7_uZEsLX&QrV^f69R&f_kQQ
zbm~J&+rnN@Ej~Xlm!>Fw2x4;i%wfm`*PL2Ovc}4TorDXIo17;$WE(0+`uk}a@$<_j
z0afUfIQ~55leCf?GJ({?*wz(S%{)sNT>%^UxUvE#qL2SNejN^xs#*n|u^4RZXn{gK
zI(>u?ZF6N6PB5rxRd*{y$PNzB5`t|`-IVcfI$#QXXwMZGDs=E@1;^Le{AxySZtl5A
zG=~r~@_t$l0NgPRI|1m5!IX{ENay5=X)+vr?KrIQeg+5wmpNG%06@JP*(K2H{4yh0
z)x0LRz}A|LSgQ^Kz0TXUkz-x8^d`J>;dA%;$JK0CZmt?V<$kJe?=FDQqe;;d;W2Gy
z=*aNwRn($a2+}*vT*NSt;dxwHe`C#u+Tq=WCRepvzsA^q+$0C{*ExiT4X#7K^b~6g
zzkxXS0xJ-$N7>x<5ZEi*FZvf6X7m<<JBC6D26P-PJ7X;b^$b~Q7^}KDI?dTD79E~9
z(nrQ;+V#=Tz6`jP5|GL0x5+e+OeQBy*6wun8_vu4a`0Cj^6#lx7l;`fsEi8?rBmeR
zL>d)OQs*r{^9!))=4?9uzyOp_l<i%|{`7|4jl|otnRp)BWc$JjjYV8`#0G~DCo7GE
zQ<EEIY%UBVYE{8-2*TP^06d}a9E+q%gICUd1NivVqEKWT>CI4Nr3CIVUV}idyCT~t
zNE$8q9PI<Bd&;932}!ON@tOJ*Dp{_~Ps0pGPl{_%jB(_uY$yF*d~hqS>CL0GTAugM
z7|ZvHq9%@HFRxj<pB#l60QsXnrpAZ54D<0W&2TvlPgavgZ8}v%U0pRTol+A=xs8%?
z@$7@&VDim>bR63r1wol-fe`T*CRZ`}u;aq9Z3c@9A)(o4kVU7H;Y{k5yzue0?0=Gn
z3m9w?#bQ_}V*-d$2uvHf0=36PESrM6hncaPB6Yd#30?t3`M=<iXDUv!b6d@>hw49e
zx@@(oW=lxfYnHDX)x4n~k|TSCY|<G+Vlqa@{e=Q)+R!{-FU(75jasx<^9J0BwY=pR
z)j`OEon3AQ?5svni&@;C@XwUpjR%hNUsn-v!zLl2jH%-`B8UovBLY5NQ2-9T%KZj)
zeVxwB$!mQExW7UON)Hf}JageXRh(!~r~I3C0;*-lZ-%17XdnpOsIY2Km1|4r#fjt>
zWaKiWye`o>##eH#_(#^xC<suJ-n4Uydjw%;;NxWrvVi>a<EqZsT2sA-*tLd0-eam~
z81pY5w&E`s|CaV^Ph|zX@S@|smNv|sZMF;Tlw_B&^&>NUa*#nw31M<pyjvPw>r})l
z$C-q2FZCC(mY+cN0U;jfwuEo1uzHbiE&9L>=~|qMd7&K=$8H?=4c|GH%z{R6t*$>N
zqt9*n{5tBg>$L!C3E(bkBcty9UFn)0fFIE)u8(dKpwwy2$bhjmxzjs+e&1)ZGTUdV
zs}OK@_^gsr!<J5b@Ww`vW?2Qde%uHxX)D87Gd)qZAv=8tBFV6&U#6&j<wHbG3?z&1
z?)E3@ONN@pWz+_;ZnBwLMdy2(b!dDOte%r;1mwkO@Xp8o-l;vo;!QD^OUiN*-9H6R
zHDM*08hKqB)`y{A_zf@us!-H@iPBu%zEHP#ul%7yoE||ib}EYGWUgDvO|Gm%$;V(T
zk?6+9?6naav)1xCgJuY9hde}fM7YJ#{P|_myR!4c9b{Q8zJ@g+5!5|-j`CTQSys80
zY8C|rS<k%+O-J8{cgH^N=pMQeTfq_JG8Ecp)#!J3;qG&?s{(xPAN!?DYSV)>i}ILb
z1hQ;LGVj=aS-Z(+<Y*GPGMje1VICfCqf3@iq)*eb;QPu&^MUb`?DaWK)EO(z{v_^v
zktbpTm!1p@avj!c%wUHhb~#6#7Tzz3McCw!!pizNk$rWmrQgyyN^1!OvC?snE$j!G
z;7sr%CRU^t(id*1;6}+a6CV2<cA1$~SahO<7ZGQ-<}tv9$-t*^lj&S?s+o2mmptQi
zt=YP%=0g^lm4|5D4GjxxzbFxNHARMeHW^Z#@^aiLr4papmqh3B$LHGBUy>{}Rkj4Z
ze%8~Fkz}Q-Q$T7U8ufZYsa>r4inb|fkYZrH#-hg+LhpNmcrz<^Wg+gT<}&yz?)v9U
z#`p*l*veJ-U;sSZ^tF9qfyNi$HSR#|sxdFYNo<@i6LO9HuaMA=%oHZ%{SO$HZMh&r
zx1{l5&J(J30!n55APBkS?9e7Bvvhy!K7FrMek4{0WSFRu$0NSZ-cyI}zzl{CK_0sy
zSwe*3a$3h&<m6|xtwTbq;Tm6tFO*6l?|T6<UXA-iu&BsaRekNYL7uHN{jt<97b*(g
z0$!V0^OkT!BIuW9_%`mQU6Thv%modkss|z68QDPkR)Rio-8UUWGOHLj<hW__coi`W
zc1BrX7vnv6v`dqT0L%>zGX&f)wsmU%88GB@oJ&sLTqlc()ypMk`%^OT{@bVv4Rb7t
zn6{kw7`wHYe@`EO?n6j%AM#xBRqbh$nXy)fTF>MRD#8nz+>XK0H9(6II$-AmACDW<
z_;T8yRw%X&l>L=FaW4p-O|0b)5fb=TG4ck|5@lwkdQCxm^LXT1<H+4%21qv~4{&J#
ziapPVFzRH2CnvsDHgH$MD-5%Jal*eI5%HfSZsWrITjd}^qR4WRsRs%!6JBkks&+YH
zZc6%7UQ3G(+*VuI!b1=U=dBhdBxvXm4tYRAK{%ZJFEL8WH6WrxV?BTirx>uT>f5=1
z$yYu8fkGNOaU2wL<|qwm=uY!DUH2dT&y|kBj$b7t!Y=WW<gCTA&@cKB)8sXebAYCn
z&V!IrAtWK#F(WfP2swnj=T1j&b^L#^yffv-?x5YD(U+(HIJPB{9Qa!;%jY?R*eg#5
zI}8f#)1F!Yh^VKjOG8gbv0zijk3dR$ec<rg&fW1Xct*|#J{YQBx{p;Dlu1&oUe9x2
z5gVtIZx!`{#}W!pz*f-L8%8*<hlJ)i?~azly8Rew=obbEXbe)NJZ%C5^rUD@L#?%P
z1~bFP-F8;=A7hip+Y~V?!c}3vb0@WQh@Y&z9c#Tt7{uOY?nn)HFgSShmfUAqJgx<l
z5*(mZ07vtKmf`QUAm$r#@=q+ybB<H;<j#LoKFZN+^I`o~AQLO-N1g$vG${9#XeQ(r
z1_9w~Kh6XHy^!)f4)TPd7O4FDux;>+B3bep?f}58x&Ct|)?*5xo}o*%Rd(2y6g(5*
z)o~gfY*cUlR|@u9co7PYK_#yuHA6xNm{~Ci!~lX1@|BO9y&M+`Ya6*$mC-9WInx_S
z1tN&D*HK`j>(KX<($L0H^WV>#znw66(-H%<J@y7g8Gu}%zuJ8laI-5Q<gdRU5GU3Y
zEi+^0{?*i^e2;-4oZ>{Fw!#d@kWiggPOp5-qEL)5Gpos$OTP>{eNQH(Bu`xslxyI|
zgjXkPEEXaww~7eQ4|~ugk8eOBx{Xa8`VfC4&G-9%X#i|ROXUK`0#q}(Qo7^=-%*tU
zbZXz?l#@4&C@5w(pnnk?9*VEL4_s&6Dp1^ic!z8s_<P_2iJ&Xu!UtIKg)NiVst`cQ
zWH$+{pV>>E7p(m1?(X6eAp;X3@-O{Pr~-^SQj)?%`jaZN#6^k0866>EdAU3Jvnq*T
zSHhs(IlJ&Dz)v32S!QD86(a*`^4Z;Or>c6F9f!?!m1uuBlbj4+095TfSv9!wWeuLx
zySsrMya%~o=cF89%#0aKxKP@=A(v*N)}f~p650v@#J${IFZ;Z2fT5f%j|<=y3wA)-
z9x4EHMzqs3rvSyTnOIrj|5813Ff@=ZjN~l+Pf_Q@Q6d2VkY)lvEasVNy=qh3Pwc+G
zlor@NuWzeX9^7nlcOERhhe()IBETFCSCOY}C=;s*&EG7km=LX|q@T!i`y)_!rvsAH
z-}ke}ABVyCfv5Nf(m#>;&na6x7zJC$ssPmY+z1Ih4gERj(cXBQ`S}TF8r?tR8}|r9
zUqr5tc_2VkgnxU9%bqYiv~=|xz#{0SLoe(et-$P6hg91`zz-jA)O)E23@RNEKLF%Y
z3sS>J9lNMss{7FVXEg^ys??H)$V!eI{buxlYuNuT*FTwis$NJ3{Q2~=kH-9{obity
zON)h}4%|#~@)-Mn-Nc_VEhL7mya&0R3BxtOz3hKWU<u(JWSo5?{U)H9@A`i&>jP&%
zc&|yz1na4ldIPAu`4x-5&;h2_63wK5p+*78sq7w59s)#pCjj|28iKw5K*+2jlNP<W
zmTt}8!yD;;8(2g{#9eFaR?oJ|R$QDr@AkcW_x#2=T_n*Wm;}r72<#OXz0L4J*~QAi
zxxvZ_p+sKy>KL^Ck_XU+y2|eW`$^Q(EC5?0PeB%$=lH)(^7r+8;D=og0~_)GFOcqK
z)BmUC_+Lb&8hI;Tv+r8oB#GbWy|#bHMcD?)`cD){OgOaFiNLmQE<SJX=s4Zd(t=tl
z?JY%wEj>0+Qc`;1&bl<E&psj$-Mobe>wRn>-(t5kp+7YnRbepiAl_TrZv-MC1#qCi
zXrzz_w>t4nYO@Jze@+=(ybWfB)^%I~5c)yJ_6HEYGw@%o;`=>7cwp}HlK`+0Eye!?
ze&{5PoDheHLsP^hv#w;2TCkRbe_ds!ZuIN#qTWn6w)gbVMsAN*zXDOF;A8<0$EQL5
z3;o)?pPoOPl=gjPr6l6t;uuAX^%zyXK9r50K=Dm2*F%Nqt0&ZLKiT4=hYIGry>!V0
zvMbBRCF1@V*I~MN{~*_bT6H0iP-gt!TVt!{jQFhcAt(fo1a<%=ReSTb6dSp!z5sX8
zzykO3T7cW=%+p2g{G__eR$*Di&(pB;bQbE_exj{-qx9HxnGk+kY;VmDV#{)4i`2kN
zZ_ugMB(3fh*&lKJ?OVbtPBx&<`8)0uW}_tP-7nSIRGDm~__s=M_YXA7L6p9B_k57G
zI{VafTmdEg+Q4h(At@|KVi``&XR0VBjvFo+tB_1KYBA0ekhQCSWL4ykldKdv2yv2W
ze6&y;d_ZEo)RK`@&r500t9GO_@cOSqtkXA{FuWzkg;YlXd9?Ci5VpD-xY+NqNU946
zKKDUZ#!bh9D@vfeAgpj0IPAPH2X3l|z`iP4l`S~hel`ngs?xCz=@{%dJ(glgLm#ck
z6)h^P&ZRZq@b^%5<>{|fGL@qb_<xAt`{3xT-9upKh2#W$W{!e5xI>eCtKyT$Z|^AJ
z^xx)qu3=Bp+^#0Pm^Zg#FM9b0P}Ob08v{(pQPS{}zl|VyAEr2Ezm9%&9Jb~5xQxMI
z<PfCceMFA+O%*3YA9#X-<0q^nRZ_imf7%`8*cLN>CEI4(6yvt?$zZeEqVmcsuPr6H
z`$hYWdXmuf2a0tmlj^NbmoGiIW+`?DmZ@FDWqPBUHNo}kB1v{Q7rW!vBNEj`h*i%s
z84=vi=||oeDE7P7Q}Y`dXX@v2>ANmAM7f}qa_!sSGuYgH54SSPh(|wLj#Ya8?&?!n
zi@90NyxaO>_?oNB<G%QL+mXntlG_v160S+kpCi0##?V7#SDyqwTX1qyQ0zCYr<S;v
zq?zyPz@<~cr0>4+BI*jRXnl+(e<ryPFOHbCb;pooZw}Gq8k(<1M9_*^YR+#Kc^Iw6
zr__GT9ZO0>LezP*PV<Umu8BF$wdCG*>}@Kh-zl`Iks)WRo2MbIYpKSMyh0cD9+5b!
z>Np@C!L3Uj8Rt2B*eVE@vn1Q}`CXd0wdB&Ev+0)jP!X?_c@89JNN{88lb8Lh=z{5L
z>1vp`Z!5k|Bgy3B=RWh114fP6<)&y4bG=G7eb3qFCHO{8!*+!St`$=ao%(b-OSxtN
zh&;6uG<@fKGW99_Co8D)7h@Te2OZ3U+wR)6o^)Fx4|9y8O{%krp;w+uSIf13w;hFU
z?nK2@Zf+Wp(W|p9=2FxC+cW&9z9zhh(zV3g;Uo>k(VEEe$jy{K3_l8s;Bvkq6F6k-
z>o4wtkLPp2>g7pa+p+C@T<!1DCYisJwaqlKXu%T@KT_bul7HoRa8Zw-E!JeJb`YgZ
zS)HcH<~TC-wa+xN*U6<*;Y_Q)4v$32z}NFs$zpbBrRBlqBZf;(?w6kgo*i-#v$C>X
zMx|tVRaYN^7_nI~U_Dkk`RvD2Uop@x(A(lAd4gFKFE}2B>8mx68uH+GoQ#%8y)FsG
zTzpAB5TnL9HjCI{q4T2<Jxt2T;N~?gY53l8zzGw;_N6DV8e@{FTb6t`Hc&X00_(Ms
z&u8ED(Hj4y>T^ww*#}C3MgQcHfk>p5K%Q5*nlXU_Mz`eyr%;JjWWg&0(Hm4d&9Bwr
z@^ciTU*x$?_?4i`6kcUHILsF<k@49j<)S3ixpYpW_1$a>3JdG#Ms^mdMcj?WRzIzk
zptlryNVz2CfND8nj*tYUm<^KWBit|a>kOQazOxb|zTr1H8KIefPfgcjF*?@{({V$%
z=tF{AeSt7zl!w>MD-7Rif{x*`Ph@^Rtf=1{-RC|KpNHU%5oXU8#YeWcx9d}|Qzf$)
zQH$F8_ubuo<cz_XjUvqu0o|ODCcZD9s}{Aj=E(<?iQFqK(BLPZc$O1?>Z)vT>yzio
z!!EMTuAi&!>Q#!=DWUishw+K)USwWPSE~J|7%Hket=5F(AoFit$F{JyIZt+{IpB9H
z1@a^#MVMJjQ&^^{1wF%@$6wDG?c}%BguF;H9O$j)W3#LlI_vVK{pncU$aZ>|nd*ZF
zXSXrDb$tyw#*(oPc#q}5fFnkj_ZU9Qq(Gz9G|_r>xA|!L-X>*9%H9NACyCEz-KS4I
zF`Lu^=CF<Y=;A>NUa^(?j%PGhKfmO2?NkC*`nq;J-Z7{;&fbTKwYGZa(z)N`)*8r@
zX5>@xp5TN;iw;ECREL(=Ri22x871TqNU#zZR;pNtvE<cJ9uUwk9(jT4#g{FwtSh*%
z618**&y(j=Y#Zu2?eKonocwT{E!Kf=bvRa-dec;A>6FctQ4bV4G@1uJ<UVm2LCPS)
zbrKt#bL~M>qGplv9bNpCfz0W}amMn+2bz}Xeqqy*7*8B>dF?9tOl`v<i2U77u~gO_
zm+q0gCLc<MSnFJ+W{+%t6dxH%OOO!7#PxFujw3B92kyw$-M})2Sh+8;Qq?u<2(6tJ
zw05t+DHsGwUwvvCIZ?J&ePeDda)X32nc__)`UqYXl-gWN4lT-FXIIT6i>rEBwl9C~
zMED!e<c^PA@e-Y=(X4dG9M*_;{B&-)|LVz>DyKQd*5_Jie1<At%6ysI{Q5gmvyo~H
zx;hXgz|fc2GOfg3LspLn@vx~A)Tlk6VYS&Noox#nK@wA4lXY8ub8$#D`Omjkq`=#F
zGx#=X4N>^_n(5Vxitg{T0QdCr?R?>G*`*f~JnR=TpC}dei41Q91sfZTtQC*cRZu0g
zR1MqJ>lwBA20d46imaW$V{z*Tnn+=x(QF#qi?Z$Q!#(Dozb7KOuC*&ATvsG*@J~Tg
zVhAmM-t9)7m}Z`=!k2b_XSuUG?K#(*5~ca@wOY1g2L4Of7sRm8RYu%Y;PNLuVfd4h
zYzJ;%xKo0i_6a&ck%;D4sz|S;wO0akq19c=J9d&+zU58qC{^aAQ^^&6FRBZARh{qx
zi`%O%@d?kf^oz|ZCOjBFRw0_%qOo0Ja|0~-H-}wT%uzU%P><W8xZ7nSCvlC-e5U#G
zd~;P^wOLaM`O^8@;)d239NyHd;PS6th;Xiv&IuCzY#Tv0kezl>g6OZ^n>oldvC-7#
zPlZT`UN9M3CLeo<_GH6~pu{|vw@XpIrCO{D3s$!6ZTdqxERu3A8!xb<2^PE@BGdHf
zK~YEjV*B3isRT-eGgTE;Lvy4OY_I3vSO{G#nb#O;u^!f&moqJ78Iv=m)ZO;)^gk?A
z%EXGX+G+B_`dZ#sZM~Vn#2OuVfBe`;^GQs5tWCg7q`t2GrV+bxP_fr?2C6MInxx^x
z)9tGvE9o+xA$9|$c&fHx;mWX?k_Mfca|YYc4V(`CD|Ws^uhgj1swnV)Bo;FtEPjZ|
z#<|zcuw1zEN<=e)D$)ZjN~dd96P0Vb_MF{pWRBw#WJ=57<e#3J@Ck!050URj5{21X
zrP7U7kqLAU5<J8Pnx=DOYowPXd9YjQl5sQ&2GIx)aodcAu`JnaTp+GGem!d)G<tpE
zG{t)K0x2Wv_MgIv;%RLp=11}oL(AyZvBxP4S|iqYBS{(~HcN)4NFF~<FLV-*jcdQo
zNk96HK8uCEou%e^eU8T<YRYfGN;i}xc}XtjIAr<2<wX}Mu%6pts277S?+~MDUc+i@
z3^6*|qvCknhk3I$MFc}J0_{atK%QLD_<}K67uU&HUtxxrF^;L8?}?wn^Cgr`Ev{QR
z_PHVY#-^Pj*0{2?N;!5`T5@chZ?KUc@z_|+ZRi;=9x=FxEsa`#dw_2#I>qqIYMX2x
zs?ztDb1?$667ztt8eKO`Y(QA+C)AB|F%Fc;ECcrJbIlWs(Tb{P9}GfJ#nzSMq{|(d
z?0Gq>zArL%zJ_UTHTukB$zNMixzXd^&Ce0Hi8$a)moN+E1LZ7U4R4GOV9gj@9D<);
zmDqetUAkPNd1(n*vY~7S;=X9bViw;*))6yF<&B$Km(h=JSb=vQ_&=~ko2xDYIGHpJ
zp`R63d1XDw^$=QFni3zuNnawR1<ro^{@~^SiLe}1as@{?kPaq^vF1Hh!qS?iC3SKf
zmGaPP+q#uLcH@I#zqbXVyg*jR!eSdkXJK|?g7HIr!1Kys{7dTU?jE{|k@jIS|H;;d
zCx*ep_8yKs;k(EN4P-M+*OPt(L<w-TNr)7U@h>`_NjJ{do7L#yHx<kux@65k<Xm>X
zJRN;Kvg7LOMhWzGvw>;il?~!7=NWDuY$^1HqEIoS1ZQEm3L=wI5lToX;WC6@Nq(?&
zL(a5W7rmN){+Q9(-|`KhsY>v!?OJP2gc&IPX9<TfSr+PZOi1O(jmB8+upBrTmC2EY
zUQuup|EH3%{QBnXMt4qq9l5YSaM1y=Uij+kjLF5g`g)_4)=opz`^UBoGbxVPH}oE(
z5I)1^s0)r+1<K|-rgOdcmHDXEO(S#Vp~N9KudkA*i=$uWa{RH`C7JaSu5`)sC1{D|
zG6w}W+*rM2z@dCIv~vrL_5_WDn9zv1wxZ~C2=x_f-Hk5nj;Fbp`-gR;eG>QIrm6Bo
zDmQw-#D0w8drsx;$pvCSeeKMhyWcAq?OraFBnwKG<q8jG!t1&PRcsMELc_>Bry#b)
zV6{FUBsmo=lcbp9sFj%BjS|$hPO!6q;-Kr(yA~Gy3a)*YRlU-U(}CxYnS>^lvo=99
zRNAs+0~r}*J(*ifS9NLzdaZ)=E%PzPV{66-)~Og{X@s}aH}(0xXA_M~6XSA13QI~M
zq4OkmpQfrpu9qweSNl;hc$AohoaY{19g%!C86mwq-6CAKrog+H8$5g%f+6}OPGvwH
zJmmTyxR)#CUC0aOv1en-&p_E{Y;~IXf1O1V%+0EDm0`II7u>jGzg{XC$VxorfRY9c
zweY6My6|dA=Kg_xQ9Y*-Cl9VVD<M|(9Br~l+?Ly%aE3Cib$QdAQL#tx$AkxMf>d?+
z!NVgn+tR*IYwj@Ov<ptqQmh)d^H^`s+4v3XAWf55yrMm7>Vw+!$97_K0@xR;A1th=
z<1GB670*6sG$&-%n*28Ux;vGf`9!K;UR2x2o`$n{%kSj2$x6P^6S;@sBpw@GvQMz$
zcTWmrjy4J9f`psOG=4D<JqMERa|yA_w+B0FPedjp^aQdg1{gQwj+e~wPSuAwWeF_I
z)cCZGxU8h{S+2=^4|Sbms7ewT!TrUvBhL6fB@M3St)$S{p2n&Td|){iBZM-IjlvH;
zu<Z_Rn_eEi^mFU)Z&x<0Pbj;UpMf*b1;UwbkKllFXUphAw7UqlGe9&uFd9ABVREGV
zC@j=p+`hhW`R+Rhc57UW^BmbyThPSwBa41mn=(URTM`<Oso|zIl1fWiBe?aWhU7fg
z8!71u<UyD8zI@P(u<jY<wvv?TN;M9-2IKQVJ8Q!E0*t2O8&ZQF-6K#=8D8xr&+(gl
z4hUTD@Ej{uZBp@IfwWChU0`sr*Ng?<-RTg`T<4p*+Y2@h$905`YJ1}Rooy$RayDKx
z7z>Lb@&*=B=#GOBa2nH@i#jQ4>Km<aLo(}bNoOawSxn4m-;z6T%n?!mNYBgG+g2_D
zJR6wl`h0~uS^8CCJX;%(_99cI;>bm-kWl*_Li)y28Se39sf>AXXYSB-lF&~kvBDLQ
z2yV?!=5<P^Sp)@zsfOkyW}K#{+ikmqt7y1=kJRVm%Y?15V^5k2*T?7noP6<a{T5g`
z2!XY?1T<}53!YuuX{!YRRYph0Ld+gpxV%E#mWf}$s5M2@q&pB@2Qcmtv^bCPA5Sp_
zwGE@&#r}Z-!}n(f>Q1^X#*?u0WNK4tAlyt@1T>BZ(;MK6`Y=jIj7Rjh8n>O-7Gv|8
zyhs>?#EN3J%AzW8t*Ev$EnbknOPgW!HAwm`0<%5LlBdXv`z5eclcXaqdKF*0l2Ex*
z2~C0x9p=|f{gd4*$}J=2qir{r*^SMR_TU`@a;M=lZNpv$1=;%av5KI$k@y-yEppFz
zV_OG*|3-5e-XMO9R=_;p85<eYw#eF1<O&Xe)+uxEZ_I-;iEEbFI^p`7g`!{$efM*9
zmJ(OgLiim<a7Jk6o>oU!)2%h<waK~sDTjibcn^3$R%Mp6(#jF^YcDSY%ls+2JR4S`
zoBSCpE<B+|Eq;`_?(Ll7D@rJb;=Vaw#)cxdndkLm!1D%9`1e}v8);|o*LOAdmEa&t
zW@%fOuE&`L#8+G|TEgTb{OfJ11<$lZhj}o_1kuxx7Y*^ep5G{wn{}GPOx3EQw;%Z0
zT9yvXjb<@!s&F}5t2cXYzKjx{i`FbOwk<e8b7rh4>a^%$z731O<Iu5w1cLGb!u^#j
zG0C4PXBqgqMb{EJp>bYbM*@TGhqqe}n(EXHFKylzD!CxWP%^=|yy@X&Jd!OF658y8
z0`-0U*B5jSWsW*?F(tUXRl~*k6u)6L{sYEnT#%o!OI~&m(tom^n<}2XDUxe8Pl?qR
z&KG;b&tqpH={k29k?wN2fMIQigO|5~Ojenw$^3Xj?nu2PeM1#ZZO}L(@W{?G+IhOn
zqI}-8&1jfD$E8U})FUBI6xZQzxLHSMW>%FP{i1{4UheExZBAY4ax6vV@WNC)U7I4W
zgu(^<qdQ45_2=*yBz4tt)D2v+UOML$+ofQ4uA-<Sg%pXNOkEl;%w+$>2Rm6-A9!+(
zv3U+i9lk`Bd^kCzF1lQ~j`w726EFF4+M$#QSsoU4d)6RQufzNd>@g$n&>y1;gill;
zJr%eK;wvM8r3-&y3D?c5>20+clq_d&NIZA77oobqG%;+!8$^Y0;c|2xtI9?7h}MPN
zVK(e;oFF7#hLc0WwH?=q-Ol14K?Z7t%I9f}*cDq_W#mmoxLi!Y@z?l=<*oxxvZp`V
zJKf@!lg}1#oo8DHX9z(`LL>Z7PB2n}1MoRsgBB*%QaS$oJ5P%jsA88Y9yr!rRehk4
zshh_#DX|XDw3=E_T4Y1?4VG~F4ObmY>}%orkG5k;4ik-0l5Xl^u493GQ_t)3&O!46
zG-JTLUPAna<V7~6A&5*6O{Cr{PN!rNSm;K>4qG{cqqAe$o0rToCTMf{Aeax__<3gh
zhA30f@Yp)vd5ioap6(McJG6-TY**hsbZ`A;R#5;8H8g@tr_f%jf`95UAt6~FJb<q6
zIJ7Dn&PFr#Xh9#GWexHXP9*4{z%_7*hMA$>A6o7RQ?3~+9*KyDjQ<%UQVHVc4=S7n
zvjMP5YhO4V-Fvh}CXj~lh}Y*;ln&?}nla2&G~Z$#OmQPT?oGs@GG-7&oj7a#%jcIU
zK|*#bAg@jIG#JDI)z_o+I|9;Bl4ZT!z8mFjO=t>WAAT=#)1S0}*MV!C%)e(rKNl_s
zAtYNTz@2B*L5FAWJH0)*Zx?><kE{=X_xC0al7S~02@QwtH(;3H$tKTkA9j%{{^qr%
zzL!I4A56jg!)W=XpZEW6@}TeILFoGsgWmXmG$+Bw5j+`>|2Re7wcSkww*7yyi6r#(
zAeh(S`)WNvzsqPD_$-p?D)R1nP7%mJ*Fj@?*TyIz#}FY*E1+M3TfQCa|J1wwC1d|T
z&@cQ+3IE?{zWrfp|AUTT7eb2dyvVFg-}<Fdfhsq{f`Wo;4<0<&a9ea(oEmrgqM@p)
z%3{!fEd?(vDYE77OpQBy(HISN<3)>@g@0?zX>F`)g0@fXf3&>+?@lHB|5USy(EZWR
zrc+@fWxIUvN5#tff2<EX-cmG}9YRME6jHcUoqz@2Ff-vl+UcLZwZ}*w6Y}(X49Nw>
z*XLq-?pj#|78MoEqjTc!yEe94?BPcW39UX@wNo!>G_Ga1gyivjn|+A_QzTuAZ}}|+
zlbr4VrwXpCu@*8*?|*2|fzT39GNe2K-jw{$rnnS4XyW4}D85&tyADa%98`2#2uyf@
z%?06v5e$&M1g`hbnUC*Q45;`&-;&=l2<adrYY=&cn+InD`99Qt9)A&h)~wvy4t<AP
znLaN!z-!(JZP3IszjMO>?~>^stKG_ah<k;#GC%QcAmsn0lPUYWV~($x2B#0l>2$_s
z`}*tjXPorwsy+3+?fXBNpDC9ig|W0i7!U3~L!<HpX!AsYr=1^w<~dq%Ir)af-RH25
z!oe4;AP2G#afCeh=zsL)6A9lt{C_Y&;wrRWr$L6XI7Sm$%;#?WM^Ck_zJ)i;MkzSb
z3;hd-AiSM@L9xbG=oY7HqRXemyE^%7-Kg%=xlQh3Ki?p18yMGFek+;%jHXvV{vH=?
zH{z{@6N<IUr9(LX0KO+JrS1c%xbAZ7W!ze)5q6<?u4}0|T3{?W>_+zH`c~P8cUKSB
z&#Gl$xAI8bDzl0P`G;MaMU7^=()e80rtjXn*El@4P&~pQxwEWHyHQ1lt2-&N`FV4*
zUh<5x<hs54*dy}u__iF}`tZ|0>a+FncK02fg>|=71|Q6I;v7=NgXRv@FHpuRp-NQC
zrx04yQ}G@;!BZOP@3L+B`A<riMp`;eZEsFSVc(BJ9rVyR#Tb6JVRUJ+F@D}*PEBAe
zXpSKVB`zpva-Ki>u1y94FVyGyRB2R?(aq>)KC!2VnHYAnBF3fcBCsSV+2xgw%wxWc
z+friS!nP4I!ZC5n>e!n0TLwraTc*|#+y%LrXU`N{m|V-=Fd^ArZf)v@z!)9UH!cXH
zTw|lk+NyF*C(=vCwk2Pe>4JLr`9L+>2ENNIzt?MXYi_iPlSe`DOyI0CILGj8aU8U*
zzdZWvgHf$t_;Nl1m$@B8`?+~sS-t@g;p%2tzEEU4Yo1c~S=v{`fmLoM%X*=e@rvEG
zIG0IfX3S1}r*iO-MK7lH+Y%1%iTGTb-<05D>PL$#7@88CHr<?p;~huD5<Cp`b-!ka
z3Qt>f$y^=lWM_yf!@qwbY=F5t=<tCYRg6^#v72Wfq0q4wuIdzL#gn|TyACE8$}c4(
zToGO@w@}a2EK8rRB!)q~SR_`E#kFj)fkE%z%t%({DF+}fatz$j(UIMrFUT<J_G_))
z4n3&x+-=c$ZHH<!Px8i+9g~i=q(l&|CN>W-;_xBNb~QX0E-|g2z-N3{f#pu`+;Ex0
zHYU=k=DkakE6Qjjewfh*PL}kf#!OixH96LEF%4t8!@Jn`IWEU@>8oX;x^qEcQN7=U
z#pnbTRc-STgMg<?@$nvqGdu!@o$t=)_6p^;bRB!KEh>a@YU3o1s1GA;dG2QIBBFlA
zWhX=Ywf08!n01(8q<BgVTbNAkr#rqzwRvjGqnR3`JLL(1DP~o=O)t9_3hM(H$rvR)
ztJlY>asBmUMHyyiDiLEh#95w@&{5<k#8G+eOg!;AJbjW0lP|rp%~mLa&eRbVT(Ht3
z6?ZI-Dw-Pqh&S6Rw23<Xv|_Fh<z?!*xn7-;t-Lbz8L_>_&SyPeve}^A8ZclzS8m`e
zCoV@qr%pK9LN0*uH@pl*^4ii?GQFO`K?q7RWD57vJ<|8&bMqtgq4ck4Am*HELyoct
zXJ`Pm09!(Wr_G@4Fe6Malhz^r4Q|8sZHh!><<8Y!19G&%iyczWw8L39tI4MQK^q6H
z!?r|q`i8_ravvzeQqeRJGyJ?6pJ&*zZBtfpopRf@d3Bn(I4MgseYjW{r+W6Y5&UAD
z^Y(K6tHMHVHw8}h%xBmA&vdUYko4tNFL^BaJ5`KLJFhgb11sJ!kU!}Ie+H&k|N2PD
z*;8XOHLSBHyc8zMIgYQK&qcUCu*_ZWyKzohz7~>XTiGnvfLFlyx0Z?~I8P(jonjf2
zS#0!|r?8_h-OQXi`p{qFN7u2I`8&7-A2?(n&>!u-(t_HMkBWTc>fxa{VB~*!O}yNc
z7aO}Yq5{4I){k#?5p?e}jI~91Iff?XfOq!DlbcJ-9CcmZdC<YjGMJEGU82w(p__B~
z9f637yjS;vp(}0gZ!-2VA0{=2xCWZ|Yt$Kdxj8j^yj;fPOG~ei%X-~y>vNoa#8bIc
zy^1Q?dVeHCJy>SJxy_^KIqu|$Ic}lF!}Dg`YM74Su=OI<0sZJ|F;86Ooazgf{Q0Ww
zp$~cZAZO>T`AsY)XGcIItxO%1>ejQ*n1i2@dVKZeGiy~?uI2TF^bhpzxb)q;;tj=}
z*PM{)l5{<tycR417XVckRFKe%@Ep$M1%ma_JL%~L66SMqGmNBiSEQkWv;Dsh6XmYS
z;GO^Cko2oM3Og=~%fhz}J3PGZVBbzIa#W7;=Cy;A*DW<~a`hR>U#c@OGL)Z>={s#P
z&q}K^Ul)4K$nerV#d*E?N=}XoXF}Mn6G33x)uMABipy=B7|V(8@uRUThn%Kbnnj%^
z7Z$}IEJgFvczSxq$?;W*j?6G9DYd-u#EC>jN8jnq9?~qf9uoDCbYD4sFf%JF_ud3l
zQ(MchAryJpQ&;5Tt7ADnry-`3Y^j>vdU>KEH}yJ_F1rQq*F*8zb@kVZ-;g%cO*dS8
z0^i^t&3VB0BG2dp>!&n({^LhcecG=qUCeHM<!#uB$DOB=Y!UbAGK<!Zyj|?=ohT*1
zyTD39PDXppQbp$Q7YxjP+JFhBZ#?nl^Gc<T?dDwM>OK1Q{z6Xcz`;S?sOZ?gs9qSF
zn(lW}mQ_FMofnXh0yU$eoXnrlV}sqYg^Y~_SRbIMxpk8aGXBA@I~gH4iV0tJ@HLQv
zL^MJKg}vooO4)wQ?q$`0P+KnDJ}`1`5rf>i@OgV~bPQ_IrD^df@1W~Ri<%xxli25e
zjBG%vpx_~n&GIQpi%uASuH(C#5!|abMe6YxFI~lz|1YM#Ixfnn>sCRer3Mg?8gc+Z
zQaT5uyHmQ64k@J@28M1Jx>FiSN$EzAZjo-dkMH|^_g?;n-wfxRz4zK{t$mJrq>NJ5
z#%T$vv~SXL1}(j5o84fOOU*+eCyR|v5}cmnqYG5JBaOv6xQ4a8?O=-``v#SjeTi<X
z1NU#@-(A9({@3o`${Un)G?4q5Yc-S(I|HXsjgl3)l8mAi5aU-xGi6#`-Q7sT$yK7#
z(qU_HXB50h9Js%s5m<36lk^$Vu)Wuch=CXTZ$I^i!@n(yNpiJ+3zmlDEp7F*p`bZ}
zCH)!xMu8q`5?Us^Ch*EM9|u-+>$VvVSH~Jsqay#N?^U>!Y3@#a!>O$`%rWZp8yK0U
z_^p*`E*1{5_gJXWB3CcY<CST1-S;ySoU3ffFA4fzFMz!nF3bel)CV2P3-=a#?xTTe
zr-CO|!Bm^PWyT47a9v=S-2r<&o^7@NFt@(Ym2^psC_-7Z84B?xGAq2Gy^K_zjg)QU
zO{|t=*OSYHM&Kn!tr$3gB|q(AxEfz|VXG_-!|<}0I|Yh~(1^m|^Tn3ho!Xf`g}v#l
z4~EJ2OZQVcf{a1&s+2xN&sJ>;=G%{cITOx}NizMi%;!l|?o1U&K2}J~XU6^elFQw_
zu4jiP?Y9sC6}vEY=Ucwq`W^eDig9mG8Y!glx{ou5J^Ui3SSdd{_f?r7*IV#es=T=K
zfwcN7Q~%@D&)8aHOd|e!{m6RiqRPxhAzi1|sDJe(WaQYT3x@e}?{Z!q7G#pKZp<eB
zIQ(RKJ2MZ(82!{6wuXj<)e8tG?4PH1{S#MjMj3J6tv8e9iw%W?^2KyE$$|fteiPV4
z^l817C<4f*=D-N6&l%DXaL&!Un}56a)b~C7zq>axLpYrD*8Ee2elvdPt3l?ukIOoL
zSE8c7XYj4|3=d;6C}nl$Nk(EU3o)p|<-D!Is*EWjb}isxZCMvFLQh46m0<+AFt%vH
zeu1{5!;D90+!+(HbdxxhDx+iODHrn0zL&`yWisw(KK@R%-152H&QCdhkr=?2`TO(z
zeZYGd3~V`T@)k`2QN#)k?mszk>G{xpP)*%Os_MK=RTLdvf$~zxsxJ&#!8}P`Ug6#K
z(sq<!&Sb!F;i082?vs@5!~`WbnP-ij+S@CJ<xfXB&S;N07<UYKNW_ZEPw&Nz&%N_@
z-U@jPGGYeH4&qG|Lzb$XqbIFa+v#mP{Z7z(rG#eq7!>ASPgMnrA*M^z$nvG4O|ITM
z9C|KCHWi&2dLhSr$pzM|Px-~_@&&M{;lN5DzTy)Qv~K?l_>^DBs7Lc<CdtR~M3D99
z8ub<3FXdJ>V%|hZ;K(%BvY-#HhmtMPX|2cGud>nU-4)`J^B}gEWG^qzM=p|yZ{kjO
zR$)I+Kd7^om?)ciqD%+0%b{;>n$bW`sU0Zz!x(V>Y$Whu&X4M^b|ya*4VE7stu0mK
z$!c5%RwQ7`kI2Wt;sdT*J)91-(X#sb2Xuwn>XVV^CGHm;rnNyH%;Cjkb53?gO7Tso
zzO$HF+&1f8&K!@ad>SRAPaWyVoS2w8k9kr>Joh1}$*flPRaHhpWf~up$Hr!ElbCso
zTBm*#&%Z;fSu^HsbLKnLH&7|?l8bX}*I*YgD=eTo-Ns&?|Awlwx5t0H(9=meG*gOA
zti+~G#2j)+tYc>{(&CKnF_tA}Sb51`vq`)vrJ&rsMAr8v<D-d11o7UUv=@GCllRr%
zze98j-jX|JlGrayz-ux{^h&7#&yi7c;;YzHvkE>Q80qOaHu*i=?#@;)>l?^XF)`uW
z&Q<nQ=%q=TmuZxR5WP+7`Ig6acz6_-ni~8gmU3fcZ))EEqeyMtj1oiS&f)LhUlhp&
zz&V<4dHML(ymXl=_1hyx)4zRhH<O8DMOlECsNsD+b3;d;xvSJGVkPCa4vN@f-W;Rh
zIK@-U6bzOKh0N1dxmQsav<nIfjyKxtsaNWwv$9scd(SDkc(xN36C-7{`<A}{HA#t&
zNiEgm)U5kMRcGxf)LB7~B)=;VHQ#!Qp4a~oY5Qj;ri%+F7p6nVZ;LKR504Lqh7`bI
za@y!uuv@4lv9=xx=-ixYaXZM<Z75}tR(PU+9t1BMn~yCgU{G8?T&N>0lC^c*9KahG
z9TmI!lif8mghA<dO}#%?g-WcY)i&_?>-c#3e3KJFUf#6rQ=~v+U7e|LiB8?i^wxp`
zx?%pX*w_!++OL0R3dA1FRh6M=F{zh-N#(Q{2qtLh-Ihya3cfsA%$3#f=M~^%$0EST
zcifw#J6vwYRLZt_%B`qU!Z006K4T3{daIWwo5*x%=zQjkM|rh5(|!oGDa|{F(-TN{
zxY^Db&;9wuXt+#qwAg_9K*i9_Bt(OJ;^YJ@g5cwgWvkiy+gpugk_rW|>VV(1F|}g0
zCXss5#{IW&>1NMUVQJo^$9xv#S_bwKUSelr4+$e%E2|c^Z<CkB6uT%&pDgJ;)0408
z$KpA(t9QFk(4vcg`hXZmzksbaprG(TilJR}_7-R`6KngJ#ppJbl`-sBKY8p%&Zp%C
zTav@%7>)yn@~t@xNFme2nE_SG1+>g`^_F$@k7yT4=B`Mjo?We~;rCu~hRCCVf91nw
zI+PIGR>xL>kEVe-sr3YR4jbpWtu~{*(4m`+tMP(QzCo*DeXE@vkAD3_pZT=cex#yK
zxOkdILopJq@^W%Z3C~?c%#~&^m_nU9=zeeNVr>%@dcqPTEoJt?{8EDck~w*LUWMy!
zOJw#7TP=3fW>gF;T20a7*6P%9xKw+_uX|vj+}U!opb~osXzQlcU`3@2Sn_&ovS#%%
zRS_@^@2EA?4rw?Cw}v7gl2?co)H1rToXC)Nf9hXfO#NWfn^ZX1Spk*ZxMI#VtOiGJ
z?A%aGN6D$RqVPc@u7e(KOs{Zj)y}S@7hVx@(@xl7U?l+k#Y~F}w-kN6y|!O`A|`00
zIp}T2bs^+#hYd3(;L<XXsDsg{4;#Ud@OY0x5hT);DzRo~zsKQw)mA-r>U>uH@!lat
zFD2JOjWWqz;($d!fLH}DoVVjj(a-wBOf%N&IjnP^oYV07j=h@-8+amzRY0B($h=%-
zfkz`>xR|Zh4KtQnm>;E*?nveRJSv5uA{H^&DPeum@n!^>&X<0squg#prQJ~HT3z+f
zTaFBK{LvdfGobolqRELpVpMBp_be~!8>YAWXHS8yu7cM~*{=4o&ioWXLdIjK_3%6h
z8zX!lJeqB^O&J_#oSKjsNtbX{a&!|<@oJZyE9@7MkjLJQoYdHns$tCG@$#V{T|j|h
zH4bTKmFFz%Y1eQtYD;Us>|Gd$lT)qSpfxG+YO&dr!OwtANkU`JaQ)cjMLqTRaJYwu
zFHms{k9lEWY<P;aOn78qdwMOS{(QLU-*Sp)%mRHl#$B%}wIi;*_eIi%W-0?e8?<x<
zMRdz*yKB;8B7OvSR(^vxSXi3|S|EphuTi&dC>snddb%dZZt?G5nUX?y&Xo6U8`~JO
z^5N_sbT?L@i4;7bav79=`fjj2>BEplvah9xXmhU<MF>Vmcip^AAFn^AN|Ae!ncdfC
za@>70Fq%}EZ)a>8C?zBks&|7yI;X6v4icX&hgzh$#`QAi@a%g$KsWpfViNMe3xZMh
z;>wABANzWt`L5PmLY;eKA}5`!2Jjau*&_ZiY@PQQ=l(SlH~}5Lw>jozNu)(GFS^gI
zIpb2`j0^osOJoELK9MGkLu33l8EI)DqyCTKY_~pLl`~Mpc*APcn{cMb2sL+Acae4B
zj&yFhq}NRyP+6cTzw`Vn8kZD89~l<*eZFQR>PU0e$QT#$?bB+%$LeTz%0xc*`rT0w
zIvS4m!-9DBCx5&7??V(;b!A$WCV$$zJWs^iePE)9@e3NMsNOlF*cUMMQmMRSf{x5$
zSjOr3d2eqxfliKL`o-abTm>S5JseK2Kx+E?`*YH{m%n~2^SO|)7Yx1t7vu=DZ)Y)N
zU+zv2*e`Wdok!u0rCuIgt@@8iGbzyHLiD~p_HoJ3E2cfOWUy@5hIaei?iJZok*#Yv
zSyK8=qyK7gGiAf;Ku4b(gV@{dR8XtB2j!{jgx+(!FLj}>vBHl|;=s`5K2i>z87Gc3
zj$Ddw{B9(YEpeVdCnmk-3&blgFR{DO%ct_rVmGCIIfOE#EcR~JwD$`f@5aez^{wXi
zhoO2rNd05DI#oNd0d0iQ=%m$y6L+Bym@hV&&FSxPdXw|*SIsx-t9FmZgfF;Qegwoq
z1I#Rn{k8B2AbS3Mp<M}jUCYa$VZ+2L{Gre2x150t4J_*5;DG9SDaJt2bQyC_!j~{{
zmlP3)_oNLclE3JM<JICZMlfwvmxKfrWPc=aTc)~4vwD+-gBG8VJMEwAdCT=m3HRO7
zO?J+0;zHIr?BF-9d+J@YEv)yw3>Q!OZZC0;>7|eNK4(!yqu2H<eOk|B#FUUn#5H;3
z&k~Z5<l2DVF=SFJm@0cqT(+fD-+LLN2_LFNCc3I}GYwwc)h5``mN$)TY~0DrSFgjR
zt^wHCCB0G}Wyg}$rc{x6t)tMS%3~ZlX7lyMm_~H%O0iu!91r%ZQBOfxhed&i;0B&Z
zrPdJgmN`#MP7)%mAf<hHyHQ&%fz&b^-*{!Mgq6CCG0U?Z^-QFe2EO+cmm^3bWJ=@E
z$LuC{CFq(4MpkRvmTg+fp9||hbOVp>rZSj;Fb-0=w38GCzArM~T!?YdAaO7J4u4BB
zSTI=SY5*MG>ErXgskH+=JQbB#cIQoLiQn6E&M(Y>v;oq2*Pa*+<cz7e0OoutD~}k5
zoY(n3e$60eQ*L!3p66?=j^t3yS{o^h9vB^z>KXgoz4(*j>StEh!YElHzuU#Bcw|@4
z->gCJfuAGdyD;QKr~r(q;Nku^*>uoDW<p|c=9vYJkj}#g$0@eke0VnH!TWsx&k4Ci
zWX5=0G9o`ERT$eiyyBGWi&a&CVwhub9ZsY_ppAJZ1uXp#k9%|Xn$6|x_wPS3veAJm
zBNMR9TWp;=40ezDZ28+N0!aVgvRr7y{=RR1LHmlg66E{T5#aD$MH!X$g2BELlk9n>
zUg3X>p%ByUCYWz&>Sk*qu`@zBtvp<e8K>Dd_qL*JxV7QU5FC1EMPbk$FhDj3%gR7I
zSbxQ6(2UV$Gh=b~*;xb$HJMFCV9UYY=nZ95`u9t9O<huPiB~jh6B3(dg*|yGZ{Pfq
z%eRY8%dvzS<3qUohp@Dt4n8re<h{7PJix?sL3#f9p&#3Bu?gKZAS<Ago}NA)JVGwu
z4h4`^wXKP%X|?^52uonFVgxEGYQACn=fieX9Ey*J^Cv7DigvWBO;&3-R7dFinVsHa
z6#BgWUyX@yalgDeT3VBI=NXowN&u_w(oRQ_aEb<Dkz_I!VFd;;DE;0%9J#e#Zo)f5
zN69IYPX&9)DhaarhJhG9Ft8ZbhGh=O%E)wG9tq%5P{f>{?_l9n)f%GNjXx_aqz{%t
z6#vnZko86v^B%fvayp}5tali#wfUJP6w08qZftHH9MM6j^Rf#us0I%M5fG@^EjK-v
zk&)r`xMcYg6kK;mcv==2iD@%e8XO{qz2K^p#TzaeNnGZu91#uzcgeOS_r{P5;1UtV
ztoKFIG+Dp9FJlU;8%}03oh@I8r)-2Ps5O>bKHj!C?{Q=m2nHr5a_g-&*vyC@=|10m
zxFP0mU_XC*w53-AE>ukK4=ad@!n%LBT4fO!oE`x+AR)bmh;5RY4@TYJ+A24u!1oRa
zoV)}juI`DAd}4U)L0!-H557MlhlPg+H#873X`V~AHrnfGg3J*<oHn5Ngz3&t;|<-5
zf}X#r0Ngs@10p4jVAiUne#?z=*>|LpFNN)Scm2Cgi7=G6*Mg-}@%mCvz043>DO*V5
z$?t*AY3{<LRmrCIci(8F%Ap7c|ENJ@0OmS{NY8}9sgf~S9!z`z`zPI@<HJ48ay#LV
zS+dMW;-Juy^3m8S9O<m2Pb(AI=8ADDjRT3CII#OoK9=$k<)JB<xrg%0=b3FrcReZ%
zYTup1;F^>gC@_$jY70dv?|}qDnO-)#D!rPDTIf?{rf-nIoM~KU;xLlYMY_fS&Ncx|
zA}FR5=&jL|n2uCyN?R52g9GxWlY~8Xd(VSS@jUVGu+H&z*Bv7HqlW{7f~{lp{DnjB
zI2vfm%SUzM(tVuvdXe8O`bju@0h@^7+_<Mou5keGidki2CkN%0Co3<0AcKk1rl;tW
z@s+*3^AO`u{TyqBq~#&t-U;fBX(Eq3uV*33p~JJw9&%ve3{rHL+dByRG>IDxI%z=t
zX7C|WQ6u&G`iMF9OzEZ-UT<kW^^?5)N4-z4R138$rH8}@9wF=tB4{USyteL}cu8#x
zn$>P!80S#!jkO^fLE5avCmNVI$+OJP;^D%I==*U=L`qs}&JV3<kl$FoT`X?llCk{3
zWEDjlQ&TOgmcPHYX@6L96?4nEiVtPG&n@-9RX}(+NY}gmTV+=6juh7!<%QAvCc$CB
zJgl$T+Z0`A>K80%5?5_0(z#>UOjT00@3z;}=I3>CZ0wahleFIOgf$VwR)}P(1=PCg
zd~l`AyEY{J48ee6Sl23ra+w_@Mj>5-nuya=r1@%@LD9NeNSTr7t$2|sp@S>}g-r?A
z-bBK3^2xNSR^-0Zt*p$gN$Gd^b7Ml#-b@D1taubTTh69rr;pjFL44+(#k$J2Muqlg
zyAEJ*_c{zWAz<YE&E<XJSo0Suws`^Vx(<H>SNW<#;bvQv3>`YFhjACD0r+fnBhQb-
zDI6!!`obW;&W!2!M0ozr6D%{r?zYrzIA`?7%vtE8{T_O+psE>-|G5|WB?}16;qz#U
zU%_J-Z{dR|iCA|=o3C7sJHe+$rT0}TW6JS?YR4<>3m!VP!oIirBeF?T44-=N#zUSl
zsIL6F=n4m=c5FZ+xRg!6TbcvL`YR=j1}ehU&-dLQDevx<XGd=e+DaSzvhhd<WAm$b
zLcL@&8!7maerB2LKys1gh`{W!&35vsr85)7$$7Ms4|2p#p97tT3evyfq~Sb9v_I_*
znP*(vXPy!>&Eb3c_ZHu?5J13v8AFv(*)RAjlr|`JJA&{eO!B`w4O2_#;PUhFlX?@4
zOteHqq)F%EYf`?t*4!S_ba3Z0V*poX&mHJ05CdR#|H?=bQT*&;aGCNPtXiC2-wQo7
z;L3&t2k0y=<~{75<Q%dLF)p+!du6K#EC$vS4=R=$g?`XtD{CIB`>gQdVz<vwr18=c
z`m3pMy3psN#Prc}aJ#$J7Du#>iTY&pkPpdHuh`5&oyEqad7ls#84ngV858ss1o?}s
z3rr!EAH^?`*>tDU;M(n1)y-5cOI)Hz<;PwzeSN|H7_!hE>miZ&J~1tbLO82|uv9WD
z28(8uPESCsNj_F=RmUyI6CDIxaP}i};=(_|a=SYi(`g_6**fbbHWItfwd1VC%JL1b
zhv5xsF;o97E(D%9RrSeecBbBXipzpX`EM3lc-}ixTnO!Vg{?vN<AunJ&WC<ksfg%5
zJCd42`1nTIn?+wtdqTu`olX(D-fN3ufiF2~qdRDn8>bQ&l;F371Q7KV+wN}(@O{=P
z%iYazf;FwaSQxGE?WLNYKivMJ2PYyPuJ_Fj#avHDN9O(eUJ)5(np^cH0Beq9@DUI)
z!7a^))Q<!Esk&ht4g)dSykF@elcj&EQK@3=6PO;|3<1ltKMk>SxVt?5F(&d*E(J<v
z*^4>u%tisrTpg|i8qHQFwh<fK7Zw%uOim(260;?;lu5dDx<9y$NAhz5L)*93*fjhN
zFek|xtzdR6YSm(#nW@=xQWOMIIx`339+W$A4_`0US#kGL_IFZ1ROn+NhsThIf>?W2
z4{6Q!Z|cvC<PmmHM<&8L5?zHReJm=?NvKwDY6rZ-S+pgdzq2pIfr*=|6boXIz_7Tq
zYE)9IfIqAwn1m>C*4QMqgf)U-;$!ufzhtI$--nC)UQsT+QJ)NvE{PV&(quM}pvw~`
znZ~6hRVtVZ%ZN7>=Og*-P?_c%l-LC0&;SP`nP=Wj_hs6$Q$3-Xw`>07kD4wb83#)D
zCEJVfEqutj=*3>3F>j3#zgo0ULVQN<Sx#V07ziWc#|M#~MmOSlbcY%k4!MEdj4L)H
zLHH20HQ4(eNl=vT%Rs&P7j)BhfLAaVSN@WrILwyqq-@G3Voa{Q9gQ1TaF^Gvy1ej`
z8@c`WZ^XdiIHC@0IcjS=<qb7$YGJNb;I`(R@P-%6+BShPMH_lEJ$)zdR8yrv%Y{_$
zNfHzI&tsjpi`fByElqvsZ{#j({eb7S(BSVoo|gO2`H!wI>@@sV&m(YIKDzDsX!Sze
z-I_I!`7uMM^>>Ocx&S*&{4Quf{If+_64-3(n@<MG0yPgf|7DbY5}Qq@^$bbUFN6)0
zf40~t$=s`0m*kUW;V~4rt!Bbf^%$n`y31`toc8kGYi4{4aq@Z1C79!bHc6FTU+z1m
z6X}9EtADgT!d(cj3Xu;M^eFl2nDZrz`rM0&p4mEjQkQTwmi<B3>e=(0@&js#kZNj+
zk&JL#twjGjqV4&%<0pEg3~4WqhoAp{+B@%Gb$SW_qaADJBBkLd`{0smcf3$CM0ZyS
zFZ%9~%~1jGEMz)I8aR||%6+hsOic-Mv2@xV6QM=5W4~sG|GhL`c&o>t^l<?aFmp6X
zxOIS1%Vx3EnI5w<C}ZP0{pT8QN{JdkflS=i5no=BFWd=OjEUL=oQE%RJZel7dr;0g
z>}2RjhtKKP-F3<%UiVZ5u`{R|vYF@B6Sq5CY`lc(E8`TZ>F{TNzJMhsL@I22fS`PZ
zkQGcJtmScnT|Y|WNF}sCU0Ct9B*)Xz_Z@s5?sdw+O3mYPQr6HP3K>8H(}l$yJ?<6V
zbVZ2;qSj75!S}>Sk$`d<SRj|YHeX|E^G&R4T7y}umSErG&vI99cmjNp#jp(*<Vgp%
zn6J)_e-2vP<TEIiMJV%N;BkS_3Qk24H@Sn`BbmrmwdQX7+SN{l@*&yT*_#>M7fM06
zcoJsW{*S@BB=vdcHSdzx!$AAY=d*9+GdX68kt#MZPtg}@&4(S&i%)NXev_K|Lygak
zGdEfvbZK)?ZQ9i*%q*3|i=)+HrOk^)Z4h@VR2mQUA@&Vl03ov<JOc?21XWjlsktib
zOIK(5--diPi;~lVYxp;{Jqehe+xitc13`ni7n*ItA4#%7I+UcEd)cufw*wGVNm?2$
ziOtA(;Pl;4LkKPe<d~<nP)dttovBQOyb%82;dq<l9ImwnViAxOxeAjC$X`I!Na;tJ
z={I^TJq&Z4G672x;JT7Nte>Fii%^zuYAgEgN+8e*6d0T2uD}Hl^664=*(W8b+h?HH
z#OX{h@h%B`?7}apmClCKN=N-GXbx2ae+@Zz!us)*)}N6io=&*wW+R0l_ka7yQ2WU!
zygvpznX8fy$RVx<A@X+UV8_Ns9|{B1Y0ILVJuWUCpo=l76)h*sISyeb0w9&H$S7YP
zjB(t<Pok*trMZ+rCvJE<e?aRe0Pht<zErU5ljRLwjp|uF;PSV?fJX3SPz2Fc3n`I+
zf{h>gd4?Hmx1$#z&8yb>liRzTOeqRt4WL8-Mad^nb_(=L(`pa|tiuz(*)2XVIF{(2
zKQjRb%f7?ME3)x>UNBLaF8<wTBSI~#?dkTaC8c*tN_(<_yahE!HCnZtl8J!mL7C4*
zT^^yFu27~}2BTZNmsyC0Z%Mm<T#80YW`nnChVka@F#5Vzq*WM7iaLn2EN+P?^<;lM
z+n+NZ>0O{Z@Z$}+er>K$aP_1G+_x~yC}iu6W#5bDIy90$rZ!~qt6gU8fUSs$LqF!Y
zJ(@F=aKTa5PY4u4aQDY5gVkJ2l_!(FbH7DNVx?%raOf|c-hVX8&ega-*9~u;Xa1AI
z7@(i*kpZpK75l%k4&;NsVc?J4h;tu|4prPFyScXGv%W&1((C+_9bcU|myMLFGl7B}
zHWdHCn1|Ys#I1M^_6f1AkaZ(Gu4zrQMv>_7{y#U9I$+v?Uv}R=&NhHxlRm$33rS#a
z2%?2ZZ*y#A!XEQwrnE&`f(9egq`i)X2;ddx{J%7}@hwO3O=)0hED98a2+G_F_Isww
ziVl;rQWt-qiB_jKm6jT6EJ=uim9p~aHOp{LAO3w#W_<uhFF#NE2!-NP%eea-DU~CV
z>C5h%wBS{gVbi@Y4pRzPG>;5gS@%5-ezML$cX#V7HVtWBm0s>IkRe7zr63ESXLP=S
z!oDCg2AwfOp;Xb!2HUxJru`%La}c#4q=oY+hV?IsNph*UKBI$!VzCtM^CdnUV3A!e
z-h~w*@4shklMHh*onKtB2GqkrC+J=VUr!$&?(!ugx5J{Nd&YAm8oCQ`V3YOMYFHG4
zVHa}GbudOctT}<>Yep3QlgtKV6htr^otQ{BTlo<D&~DuG66hF3WPe45!vUMg+NR9K
z=T+eq;S}yW0ryO}a#PCnIJ6(MnRQOBn#PUU|NSfS!aJAkk*;9O$rjwBrEd~rSwcDj
z5Y;e4Vgd-tNjOl>wB=o|{s_ryt0!>V&Bx*8n+3jnl{EbUf0o|2Xm8G6^y`r5>)|T6
zLWAQPj$M|3$Di)1&W@i=%mXd&Ec7J8;ANH+CQn9_*>XLWZDX>t#YqjhUm!D1MG0M?
z`P?WX1<htDYbnM!7!PW=GbVkgbZ#S*DCiK&ydcw62hhBStajf^QOB`N*CdnnRMZO%
z%K|ka2@nr%`+lf=v-Z+SW$f05s(RiwRp*eR%yE$<Qqoi))f$ko(VWMFG1U{ocO(n!
z)+x&TL-8s#=u~&zsEd=&H~BDG68^qN|B^sR4+Fm%vcbST(8XN8ik3v^^Ey$gFODdy
zz4g7^OBQ2M{`{mU*jNlVQYAmIr%`3BmJ5O`tRApZ-4}{e=?wvloGUqAqWYzDjD@_I
zFT0hvT;Z1Zi=DO_0d7OZRCZ`f9Nvv6ZO}&`{Vm-hIMkvS)lHI1@U#D$5u{bYavBOO
zA3Gkt$u+qgxux&B2~F&q2EW2G_r*)i8+@_m`q_VKh||+m>E^1;b1Z&b5dA0ck_6YA
z#LGN+qM-~t4qlAO(85M|szi~Mq{07s0lZ_%WtN7bnhND%l>Ns%Qdc&<VKulSoF%!F
z)WN>roNnWsA)>~kf=YvC3kK(20(Me*?$XC|l0Q!R-Ye+HbY<^Qn$uo|rP?oCo2-%o
z3NO-lnPN%XDWIawcQ)EBM2K*keIed{<GK3u_60BO^v`WN0ffW%M>xnjrx%K`7Ytq1
zMEoZ5aFJ+LiNH!AUmb()mpL;OFlk9m2_w%RkV!><aXc)PYQ;%uM85l-$(@<V63^kW
zI)wJ8sOfQeTcOGO((NYQp>C(AcLcsCl%dT={)XBhR578j_MjjP9$PdP0MHyG8o5^P
z)_1G&A0!h9xnZg~gT&6Jrq_vu1DvXc0<r_rgQJ68g*B_vpOy994~|pS63>Ht$hgHS
za0s9KHcO>RrZYsT*f^M5MqcLk*Kv3vro_EPZZF@`3i^BxQ^Q~(Dhv-S&Gwz8G1_g_
z0Bjjc-*B?%8S<Z=N&ZegtQ!;VL#6cjevVlvuOf&%&~f)i0M2$C7}Wa$ivlr_mr(oX
z7w~w{51a)(iO^S;x=dyWGIrm7*UKv<TR)S607SWpGHwTHDOEKP{9o@OXod)ti88HE
z70Ezdj7z!cru2K3C^$Uq9nPH0R#$6nytWqR14>n*A~dTEY2+q?<$Vgl4sUpm>Cpdz
z(ZE1o$08ms^RSCg04X)O6sKIb1709X-mnUD_Ih*MrI1RI$OH+X#KXhmb=iLNXSq2V
z^@?&cA<WrnHdLBA_Z&w=J+98k_kwV^Bc5XtnMB0!WTDe$i|LIZ!jN8E9s%lQ0FLZB
z7P#3{rFFuzf4%GB&L$ZE3g-z@v6_BT`}?cfeiuwyuUtB`tPYeH-Va52O#w0@SR-Dl
ziIQ5mQWr^*#YZpB+vwY0Ugk*J6#8h;X`%u<Cj8Xd?3Nbvu`p-na<$+>rq_j$5e9}=
z<q2OpbPRDAA)+`0q<o6dw1W*}`YT1?oC3W;D2dy7zv)O4!Frw7ubKM_MuVb)0)ES^
z_o;5#2MIGeN;B-2WMGE_kK@+!01&c`IazdPjBt_okdO1u(QhH68=xcpWMkPqn<Sx%
zjQ-qTZ=FX;OU=qJSwtVP?@Jmfz0mka@XyM024NMt5k~oJ@$ujQ2N=P^GMf4-sI&wy
zm)ru_D%V-mHB+Ak&+g=tll>eFjYo)pQO~lk{y}&6*m_br_hXH{lWpPHbdEU+sII|`
zpR1EiC|9a5?q`UcjGLHS37>_9Ss-+}&*Cr&p5OXuO+PF^U!B|OWvY!u;(r<m*bsj1
zO_&Vj%%oFrA}kMQjn_&XbAC~e-QSv^eXb$Ix+0;GV2@#fzoxO~W5EVOo{1*BUQ{|+
zIi^ogR_b$_vy5{KYmPA=*ORgst8Aj!-TDE&qkEm&5I<gjK8qC7lVr9RX2-A5OU>9^
zv9Ji~UP_Uy{x})hkDtEwC0uc+hJ+DvfGfkkgUeKhk?mWVhERxTr@|byez7?_j(@|p
zwr4dcWt3}m`Js5<V%Qwj*70z9z>AayxS@txTY3Ib@E-p|!N1XI=P>d#(4Umi2g}I^
z-Lro-^BbgI+Z2edHJ|jP8$~@C_?eYuC8Vwr_;4C?)a0fPLy~KuGq4;r_g@uqA>C=T
zi_fTT$^>Ac2)uPvCx=!FV(F#7&`yvWieqDUYecV#by4<=>0blD_EmcD=Sc51x+&dU
zW{UKU<u+FVhsVrfeXxI|rgM_JADIFI+x!!X`@t(R)ZqF2kJG{@6B0q0f<MDXyWbV$
z>=$L%sEV73m$2faQSvIobj0X!sY?r*iQej>d3bmT`dp&{qS_>{1_oH|U8>#k{K*YJ
zF}Lk(Z<7;-2#aq0Yry#fKcvY49%S>KQB0KH1FUlV9*@PM@60WCHpf0p1q0l~v=-qL
z(`yd1clsFy@RmaMp5qYH)fL}(eq5B6uSi@xfJQ_j2X9u#%^wRI6FJUgldU{htk8Sv
zE>sxG76Py&(h<l}i<=#m2f-ZZn?dObc@<n$RW-Q<br<q`6TI<5i387@T1|FI|AJaI
zNS<;3>WuBSc&`a<0}mpzNxN`+(OJV+k|v4(P5+B)jfV$zr%9+mrMkBf0Xlv8C_pT%
z-yJs`6v$xPV09)z)Lx$LFxOW%+TW4+-%}fqHhh0?{~ZPB0|FeIrvFxjgSO(?&4*jR
zNXG@F+2kn2%nN2V9dWmzV+yNzy&1B}P3TQ4&jAEDQ>i)YTRhHHOC4U2*~)FCbE*>1
z`MTs=qC9+fF_Mr1DmQW#&ky4jy6ecQz%^wHv!>}PrgXy9!WEdE?F6WW=MEUlDhm+&
zx<Z190HAt{o6Za9=07kzmuC~T1AUZ_Qj*-=`oZhZB3_G#|4%GWB+k$Tv!QUyY<!uG
zVND=_0tpZAz;`zdxxGq30P54ut0u?oqG0pJz~syo8|BeE+NxV|UGAu1QEUtMo(WlM
zt{wOJ2NDw#&PJ)B$G3Bi7t1t*C2`~4bB|vS+lA6A*ZU<=RBX$I*yvSzzq+270qViT
z8})3w{%L9I2Qqv)wlF*ug)zAr9%-@4qQ?54kh)+r5<Z?!BdpxrakXz{I;CPB8ZHI@
z1eX$&8Qji>YDp5K>Clz>PTUr_$&b};ab7aR*w@=-dmFx0G3&aP4h`^wkN7Vt15f=N
z=9^Br<Q3${3yg$j+W~xFh_^=sBkZk8n~i(Ja&En`rb3EI#K?vIv~u~LrwwJb=)4)n
zZ?rY=ACF0HRPq!4f6F$8B0^o<QKS#zBoPjm6L{)6ilZrK%g2S#(_fp#toI~v7%*=1
zz1Extj2V|qJ5|@Us><H@7tcxthfD@hjwF^1DZ!G5TDfe1VlFk40v&4HONcv>chSM-
z>v&&R549En?orn#VKv<Pc!2Qf{(E_n-*i64!)HKP9L*k3%9SGw9Rk#8ZOvz?HQ{i;
z#02J(Y5wW*^$Q;NY_fDNYk7HjN(@v~RDr0iNKMydsloRBY=vIY@ic~OCeSF`^Ao4h
z)3B<!?I^x({Pcvnw&nqgk56OUb+EAs5qZ4MbUT?dTB>&1kSh2SC5ccjat)YR?PRi<
z!w@YPN?hC1(cDuv_0;Aql+Q#0x7F(ac2(E-SjF@I7b7NM^ON;HUMKOiYj~?@aDx{>
z)UoI{N&hjIJpz(XT>_`(eew88?&lTtxjTiWbXrwL&Oi>Xb9O!`#swMfX7Dy8qs`92
z<S-)!5WbcmQ38N8&t-v=cNzp`2D&|#m|-wQ^fmO&O{U#{b{DwBM=^uEv&Fh|m6m`r
z9SnqJH0TZmqnpa$LCz^Mbs4}1CxNufooBZKw#EbyE*>X2ApW|4RJ{-!;_GQZw5!si
z?r|xLvTI>BOFg`fl!oQY(Oi<7)B9eU<EC?l%VqO{ehz6C2oEa{E$<2=sAZGgdN)u9
z=d1PaTu$x5s=BV{A4Xpq9wvpe!0}OuH>iPvcS-I&BNfa<r<3EgH|?7<g5QNjr^@*E
zP@^l+?Zd6I5<!97h^mEnvTy`d1CtCbJp<5aL^bSS)2UjGPAt-3wu6EV$hax{@v|7F
z-S_H|2c*3#(q*f{hjmn%rJ2Zda5>>X0}UGbo?mOPPIS0Dffi&CYeAyGw@@|c!PFqN
zRX4C5`>UX^-JB67es=Bb%$$-oIBbJQV8_M4R3u~0Jy@KfpU7{n+~A&iKi2WNYir{A
zL6*1WfQW;`@ShGagGT22Aj@xou+&tiXKi57sQgyE-Sy+<nChS3WNr>9S9=1A2x(*u
zNAQ-LH7k%`mC%dp(%-LT71wLcEu4L{v9M6Ku43i1=vf|$ux^jrj$|({4hI2#Kp$RF
zYPhCy>zi5vyg6(F4x(6VtWH&b`%V(%(D*;&g)s3&0YulCARo!oh>t;w3`U@Tpp@)a
zPSxFydvZCd`SfU>8jH5a-CoiW5lp)r6N@Xe3nYfm8^rkEZdb`t9qF|_3y(!8{n!&|
z`WG*aeg*9z)I6FhF6l$UqZzs5u@m3wwE)d;qRfMB_D4#MsgsC!dr+J(FNrv!IA&*C
z6G<$ZbS>}|x#;?=*~X99AHP~Bw`k5&s|sybK(Jb$K8~bBi5GpnG6`pc1Sf(epvC=}
zvIV(d1w`?>sjVIs(A$db{LB=4q`m`W$&)dWrH1Y*9AJU0s}h;D*2k&c)~4UfAUwUi
zjDKbvlL#Z|#M8hbDL`yWJjky0@Z5J$K=(<+cV7nt?l(v(Ai|VJ_FYc3Q#$p*1$HRJ
z_f^?oWGOB3C&1Kxn|S#-%D`V=(>V|bkt;^M0O>ER5Bq@}p1U5{;nN~5%m%VrXe+;3
z2}3*m*zdh_Kkza5a1=jJr$PAukuDwP!h@t$GGl-x-RzicUEo*D;op-8Pf>ar5;gHF
z{nelmIm{<uX4HSa_j=GCRe|z;QFR!|pkr>w7XyToHIA>zvrUCf-XdZ^pDc}Ta}@b~
zn^4=sznnLiG4WKtrGMg5IP*^xiFOiqu8>S`5?rHdCtNd=@Rk~ZEcsdapWwCt&<^LO
zQ8oY0e!2bIlPyHjVWVq>(4+-8y6}Ij71$g=roEpu<QRZL{KF-exy>WS-QEh^<V)oQ
z+yk%`OGnT3E#4RCiy>|%o$ImB=H!CUZ6~`*=jsKSRT_P6p9?CDiUUOO_|6;aU$p%N
zxnjc+VNue({gg(@Mp~z3F{;)97PvzBL`sXy>R*!ud;=F}O4I~8`M)+YA-}}#X4^&!
z`!7~|^MA40kTx_zsa<sXU@{RqPs6Um`W@0Dvnz&r-}aK95b3Kw1s~()?|)bQ-ah*0
z(zQ$HQE?YCv>y%|VpQRyt5w*LqQ25V2dBzaeIH$+2lu`9qkE={^F;cS!W@^t9rc4|
zMO^RZ11x_{yA&=DHph?P<6nC$h~#Y->%1-MKee4(a51*A`045{{pmw8Ng7PGyGT$M
z>oy+SW$m~V*FEp!dSuHsL)-sw&Q>w{3k~GCAEuE@o$umpELi~VPfofMnmm6$&u)sV
z%={(u@_6B_VO|ggFo4`x@<{sjZ9kM=8xL&`mgC@;75POOt9OSf{w7!uLxi%Ataf71
zE2Q<d#1quN?@s~T-6$wevH#I%2pb}|(mKSvfuC}nd@5<_xQw7dx;f(po0NBAV)}B4
z8XOi<5ry7`&5wV3aH{L-`nHCX*=>`=*(4J7BoJJmR<Mv?GM*3CM~r${#byb8u6743
zqpMJ<h~{RJ*-8V{+ouJWop=5#KFxA6wf_t*-J&Qb2W?(1YkPa)cUQ-zBd_=3V_*KV
zdd|jb1C)>Ap27xW28W(cy|oFow<-Dnxrk6M0o8Jb9<N`oERlv1+(w1_M64!djPt#n
z#Mu+zPidH<=TYK+D-^<Ga^xC58abS=tU0tplU}XL4u^@RQiH<OxxYTk`+Ns_mr$mZ
zyZ2J$w(XS<Q!Dj_4*eAO;P8z8D}~+scj2J5`evReH-5S4FG#_@9o87OWG>L<@E@Aw
zNJoziLVq8hKxxlyx8FeG!kSZpAXZWsd&J7)21~-Zr*1ukonssnlp^x|A?Ayms?zTZ
z@0jDG{-pnTQqTsB@Z)P!&v>|wWERs~<XK_j41e}Eg%-AFEKjJGJs0jw{B4+HKY$Nu
z7Zs#*#Gzbhg}_GN=gZUTzcW4bk_ZQ-4S#VO)gckW_hJZu?(2x@P=Fj6Xx78g7&U;x
zvbS8s8a6+xWoJk=z%f-<V-~?bK{KK?@R;X6M)^%|woYZPg{X72P|mGzLcD<d>2@na
zw?A^6r&>}H4%z8F&q`8UD1ztGmq$;9cI#Zq?@wWsXB1o;FbJ|uW|{w9YlkLPI3FF-
z94?h+4Dx=|=S-sOz2~>C+7#^di*8FEsJ)Ar<agcuLLnq<DxunJ$#gdoO{1IwP^ks)
zCf+W0@WH?YM*nJ$ISaUNWfc3!D12&Fz@NzSz0BOT1D_eA{;*dd*Xu^bM<PC*DkasK
zn_W|Ns~og2hbM=Yc-EkIce<JUJ>PQILyO?^ZANw5#S@Ll-cQpZMkl|4S3ZhHw(_RS
zO?s%5|EAmhC?}CtfD2nAlvY38argga{$-lLcAHyoM2HYk=wIcWu2Zm!5cpCnzryLI
zL<F03nC1lmWX@*v0dI$9CkG=Cszowt#(;Y$wo<=8eAP8smIEl=<rM%y!w$QKH}M`C
zzAI7=T2qi$m~fjF+$Nah04vZKY;PO|HXbdl-<<6nx{9Fp+j5xVK`Qhb#Irkm=idPl
z*E5m7-y0ttjjj23O+!FZgs()vz+}t9qWzANf`Z~7EBbe>ht~fM+E+!=4;nOm-OF#7
zwH^mvtmNlWm`)c@)Yh}QuJJi<qA62}utV?wvT}AtfA-FGA1GP=rq78cJE_RA3^Z{&
zWiRM{d;eMX8c8o24~Vn91*Ao1nMJhxT!~OHqqIs$KrI>Lu$Y~s7-HOWs!+k)<opW!
zhaH>ppZ*E1B=e=+MAKw>QgKEzrhxKZ?oFYlrl#5)onM64T;5-wY$yYMn$6A4kMq@Q
zo~O{KFWP|ofO5hN(}%imZEfdSR;>WGN&`F1KUwTjt$FmSRzE4PN``}SOQHx-3f1BZ
zhBvHQMVg_qfIrY`vXbLqe|+`5;~u=_687#VYW55g3URn7l@OtUlnzZGeWTg>tU6MN
z;!LMciC^M%P@DQktqfS6uB9Q%z4yV8>EUut#;ADy4CbZS#ued}m2@*#OIAP|G$QU*
z56mF9$x%H^Qw;cm-PUJITKP2_9zC50Q@;4Aq6l%jX|-~r@=Pu@4BjV4tP}Rr;c`-O
z#5vg%7GMh39i+gRRU6Ne*9eued5D+hHm{ER@5~Rx^?k(R3v#qu@+O;EYJ{N1)4fV&
z0+6L#S=*Q)93uWYjv(}!Ds(i7<GK`b7}DGMM=B?W#i~*Z6Pp0B)y6Nt^a=+@ZH?WK
zADy!rpo1iRA3A=;gwh@S*n;CQz~zUyeb=A${v{0jbA0|kT#fqNeKDc>t5L(fNvp{j
z;4tDPI|^1Gy{ycbz0XM2MOI&rs8tQ))#h@F-E_&21>rYCc2x#vkm6gn=XVB!l5)7E
z%u!{C<r8KcHlu&P@;kJj8H3LPLuI-`N{xRYWRe#y`8(B8j?4xbLO-OV44Ifqy*~KA
zXk%*qpc4Z>Dz-5hk#Jw>zA+8XzF^G?{M#Dd(R#W6;MI{-rQhZy5F;vv_ZuoJhUe|Q
z=mMCU*?@|R4F{C#eS-o19MSJU5sU+s_+Wgm`;7hKLOBdYq|F(f!C&o_b~F4!ysRq3
z)+ZZj$3Dcr=UjaN0SMl#{=9zp%*Tm|k8eV?SR|aITq};&>3*%LeargEf~3X9R)4R`
z&W1s+rhL<B7IK?djgN_-)OnyW`4gt--*H6|Abf;AM#@c6?kwvKQ;co>L?<;8_qtV>
z;cfKR9KF%GIVla><W!CQtyw0CTnIh{Yi#zor_<avgm~|H=IUt??MrBj=UheXj*LuL
z2YG3|9b{uuNB<3n8p8j6N@$`Tmbp|5jIdaVXh4gql38>zjCf?Vh=OYZO5Gb9wQEj4
z;HKhHQpU2I{wP7WCYTTp!dxhwDRcV!#ra*i=?}%-mG<=ht8^~=0O~*^0+LcFhQW2`
z(*q8r|8nYu)8BFGa-DYF__*AJ^RBK@?Z3&jIQ2S3isUDQ4B6H6ApIh$R@VHegyJ6)
zq%tg%3M8Bs_3MUO72REdq4n-M%AK-qmq*tp#fJI(`1k~Wx8m5ivkDVhG*@$k@T6@0
zZ&G2Fh)53Ob!Y>n`o;bXGQg_(Jp)F^UodMj5@7XHsfHmqIXM$RrA6NEj;_7&l9BWJ
z|6>QVy$BbEVz}|zUV2ceM!h&<kWUHnzJkxGef&rz6{YMmn<?mp5lJLt8?t)<I6l0b
zGzow8M5V5l9C<w#ofLKUU@Kv#8N>zmrT7t=ia2#cm(ek&8fGvr9^Q)~!X)7`#9w<y
z_Gqi&6A;r6b0)F{PV+8Joq30ZsfAyNGE)lVam56+wWOY9t`v}7oBeksY!^E?_PH*N
zM-c%*AhcudHumC6UtK_v4xJY6X~2U_SLNJ@`V+v!zp~e;cp)B5)9XeOqt5N_n@{We
zKZ>Bt?)IL+$cI)5f*ymum}pwOqDk28t$V{P50)H=%jK>$*?vwpid52!rz2{z0oqOC
zU}B&C<oa%kU>vF#UPV*!deW0>>Dv}*r8O>UGU+fbSaN)W%IIEDslA@q9O`>(jyqgV
z?VSdXnW<}da>xxN>?!<ayndl-a_H=1Z$&m#Px`KrZ^|#%*)3gjP+*WXRQlvx1%8^}
zAye+%Hxekt_I|p?(*EB#>&N-l8>Y%^E&BZ4-n|7H53y8c69NLrd5gMKOl&rp@V-H)
zfz&1&?0f&p8L*A|zmDe5qC52H`ipAhg-<vGv%1<t-WwzE+Z9P1NXBMxa9+zI#?Km_
znzrrO#(C2<vtT}cQ-Y8l`5DV~%cohRpEd2&JfAhkTk<TweiZTZ2s8J7JuI8#BpO>4
zpzQ%ya5h;bk2XkHQ9Fi36CP~*^BO*W_MNuh52Uvz?ByqE|6u=uz?NXb*1rrMoj><F
zSGX3k^Z`16TS9ZmEF|HL{i*jA9#BWvIpR3Lk&}P`|B9H{Qoqd;Z`EI<h%t2A^x*@L
zSB6>-fPun+bMk*mEC)SoegCLO_SFP&{NV`8vS1sflgDv@Vq<cpK1yO5m%0B=$3TDO
zO&u($*l?~{<ye?;!zfYVa-xB`$zz)fQ(3_KbDmAvm9dG#OBXrj%MO5btMb+?iwP0>
zT<5j9ZANoHEdyO`4lb|XiTdcor1p}+?~!rsh=$op>7kzT?(JcQfp*;e=}-T^n_YqT
zEMKDFmUb~unw%j<<Z{jVqhS*m*=+;;GY_q>H-sp|CViq?J+etv-(_3)k}l)65{4Ag
zb1m+0UT?zPT<_+M-G2=WN|uBJ?rm2KYd~r=33&I%fma-d651lF*fwHn7Cav%hQJk@
zJRLT6JZ0=@QH+X!UY8Jf5|IrDus13yI-Wsr#z>!6*VlBVMw8(>tz}FK{`2Xz9n4C|
zYri)6vn=j>Zfa&}_PsRa?J~2Lc;9!acMb0>gEoIIaKZ{qqo>fF_#v!3Ou2M!d6|o^
znTut8ea}@GTn`GV#7#_0{>~ZsQ*&@&xL>+k5xv2}P_4Bb%Kz}1(;`E+5e=o9nDzcS
zyiXMR7qBHeuB}VPv0&rjkPBPI5-%)NH>z4iO5;JGaq`*1<SQ#H<=8NayDM^0b#+>r
zSJH>=vV<aBR^zzpigq`rglOpK6Swn>d)xinl4A;B!`X@l!Wt(GO6EV@q<nnMvcBiX
zb2we2Y3ut92TjgfOccVN-#LOJmX`9RMoFz^O1W9yzKsJ0M~r(yz;BP5XZb+WRjQf3
zf4fkx59^c`^k`ponC`ry6_gQODSJpuO)X##`~YOcXu6EkqRy(-rmw$R-D%jVpF>~G
z;9x=PEV>R;;V{_qA>9`D3sS%_RN{&W#1tuy-51~!*w+JQH9S3evxIyxY;DQlS0`ev
zua4#yr|KKk*=Lfasqqvif8qV{T57a!@;alR+D1b|6AcA^V^v3o?eFM&BN|r4mbS=_
ze={Rb>ycv<g$iUZ6`WN$0OMYCb+x0Jg?V>w6EYKUqe2KavPvyNKwLZ~z)xQ;l^Pns
z{$OKn9t<dU0gq2R+5RTONfo+R=Ej{j^kJYDhLdouNxvpfa6o4Fs2?oQq*0S`YYUC(
zNuq0=j5x1)N6DPp^uOcsbX|^MqGefu9OnyF`M|GHt{>DJWYlZuCIeppbB*#k9HHMr
zIoS7-X?3Lzj%Cso2c0H=U7M(@5h;V64$3Ske_uAdHnZHygBKwezraMn8)QrmsAW=D
zAO<WZ#T~Cp(4<tYtBm}33yHqx=P)M9<_Rf%eajE?1tjOmsaj)e`<L%y*Nc@Guqjr4
zO3M_gvhDZsX9g9j3aPsv=?@6GtLjt&6!^Mc1xWZ>UghQfMEi>jAs-y;k_fDqw^0b)
zCzCSud=S4Y&d2pFk57Fx+bZ}|sMXaRp}?@4&HKv`1E;=YOr{SK=GOJ8+GBwM!5`e0
z&CKgx;GHFVv2o)o0{p%ow_N@&hKRrq@DtZ<N8YSn#P{iG5aL0tsOCLO#V857%t6L|
z$KR!2{NaKB^_mc_LfA5zVg<DF&vuj!A~|X?rldQA23~S3&eeBrr2vCOSy&4fX=sQC
z8&liCbIP#aIsOTR@bh?1R2vB~3$PWWNzwZenDqC2{rCcyE3orfj5LF0R2n$KVd?*}
zIsCcUR#Ih#SCD`f1r5Xd*zg6lkJ$sP;0ea5=-H}Ll9rul0JUcXIjf()-}|L6>_rI~
z<9Mz39J@XBi$E$6asZr}7D=QzoNm1=OA)1@5e&zbksCD`6TRYiRjfHj(A?~GZT_f2
z$4j1J_qVV3qtthNV^Qz7OyF?eV^*ET=Wc1Y)nj1es);^8(F@=GkXCoHRmqTRpj;Ml
zcoq;9M)i5tnp4bDHc?b#Mn^>mMCDhh{FGMPD7_I{?>7)eOOu&BZAs$DWE{CDR+XI#
zy~aIJgKz)&69bqwYN;UyX|+|&PA@RXo+~0iJl(-_A`Gf#Cd2=<^(YxcH5-fDey&4u
zWsYA#l&lxI(G};^P#g!Qtt}TNZxG6~MzoLT{vgTM%>VTQTwN4DO#EQsEf1H@JrFFL
zgFZaD1*R6PN2bxs#l-Yyp|mKmjPE`C7DhHUu_&FLX|70jx~jjt?cqq?(_0!*?wpLC
zc`U|c;!)06At1ZCn7p*Fc5?iPwByX7I#SMZ^7!}D$DMVi@P!iW&b2POU);0NTUzYT
z8METxIc)&@M;cgL1)$B~_>Kt5thA`VLMI2hP{YpB>7X$9=~s+3Z3U^#!SDtVYDFpf
z7rtK$aylJbp7y4Sc2|vZ2dT{au}UX#Hai|}b0yEVNq<BFgLr_U7ZyiF{^kUbn-l02
z&1geQ@%E}MM$_)6ZdC!KC{fTZg+&%(GebznU4LWN6+j8xMXOVBw5z}_gIM(J4TE{!
zYu|u4@E~t>8^;&fU<~Re%v|cX=QY>s0k;}9AMw9EJ>KcC6u&i>OJeCe+4w=XI`!Ua
z>|Uk(H85d~RaJ|Drwv)Ew<g%7_&?0ORajSF_wKDogLHRycXvrQ()k4`>5vo<32Eu>
z?v|7eK@gB`q@=syo&28XU)O&3zV_kv$dd(&xn_(x?t6R&A@BPUsMA-lD4I5wARbE7
z2>qVNVcY6{H2sa{oLaqjW3{jgk1d;&ayvgFbnL^~*2X;d=KSn><LzzltoP%jil}s}
z`@z{Oi`0(`b>66cx~<x$_Q0eJZrD#RALckCF1Zd4c&#16<#IS^G@U^UNL0a#(os#7
zMRk3UokkMHm|ki>K}6!(tNp&Ui9V$ln~-ah%AvO?zmTVp<we#lIXr81)*(pXM&=44
zJzAJ|yHrB>8S76j7uF3Q`O#G>L@WDijqB&WCTo^Ii1WwD9l+QlFnnvYnHG;+-%|I*
z1sdB%Slx@>4UOy+o9i!1Wa{;YGi^zHEu8V_2+k4Eg(VTO@oFai7grN0WMIeON=!bF
zA-EHN3)NjIy`6Lwm$xSIY0k<RrfXT;$wbPopDBxm+j#!USC&ye5j{*SRm8b26PiLg
zsZ*|KqxX?bl}FALKLLeuU47Ch&n}r+G)va!oU?&G6ja&Z+CV2JoJ%XavlN$4^t7hF
z`ulnR14toAaZNm84VSzz^}866+<)3@k@3I_k~19}Vxa!jsQ<cROiYR3t>%2$h3s!`
zKN9q==8dZ_*>ScOcTEO8zo9IOcJ;orf9945o9LU{#uA=%cJnc6I5NUau#LA|dK40+
z%2F*Xmz<cU$m^gp2PLbL`D&dtqqBGCbM=nhM&o)K+!P6*q>eD2xL~N5Yqg_7$1Sd@
z<*+^b2r1#^4x8%J<HH;|{d=N-Zkdmm^udviYZqg{OFQdKAx%JM4%8dOgw@iiW_tPf
zsuyYNed0y66JNig`!ZrRMUjg4a~`;T+O2PQ*f0V0stAHsI*D3wcsFRgD8#*AL%q?Y
zc0mZwg{q*huW5${?O|*@`z`TfaKme7<2ErHGYSk+VX5<qw9pP;1*@&Vk3<P1W%H%n
z2E&QwDa?TnxS?3UF9WsTAs)G=A&S=clVxydM9OT6auEeVe1C~eCKq_+0zhK*W~Lw3
zZ&WW^6Cye*AW*lD(DeuGqRV)G1=;z=jxO<KT5x^o>7Wrr@SA-2E4@-@2zc37xK*{)
zjdfO3G!SFq$VeGGPdth$LutJ@fe&K&kP>;nh)^eng8;LrpKNvYV#?D<`Bc4>TiSHp
z%EYpkrKcsz_R)*(_6zHIKQm%8GozlZwoEUSEJw0S*<c|m_3wTWVKZu)%+aObr+5+(
zM{-(?nLTwp2)Mx!Wf1);(V(3|g#VbAM>(Oc;k5Zz>;BsZP7Y3vQ|Vo9boy=~D4S_1
zkfteqTLVQzL}X%P1HiT=Oq8G5_+7S;86|0EQn{=nc#fLH@uNAat5ws@Xh2RrR4DtT
zOuKBj)|_tVy|r~%c{!8RfYiom0po0w18e54C(4&?kX^EAp`|DmaNi%~AO+Zk>T<j;
z#4mC=kIYQW<2k<Gi3yq$OBo3t^N`7hk8{30Nv)!FNPb2(ox~S#<qLJi9Kn+cPf4M_
ze<<THj7Hr2k(clN2CJ?{T=_jqn(WS~>i$A!Rs||QaMwsYd^JKuN7E|R4ES1yX|%Pq
z_C^A#(x4-PxYeCQ;!RrnYpHOyP7cU$HslA+@zqsGT^*<X-tI1bL`1XO9?nl=zUX7N
z(ISf1K&Ly<AtfapNxdL$ZTQFE<NKk@fZw(-+a9Px6eC{)MWm=aZ~5v>C^Alej*N(_
z`w$4C`|xrlLQ0EMX()pIUbPL5n5H7R>hYbcW91&wV#DK1P$RtCfx+b!bZ0PNvoGkP
z;|%5?rrVJUV7O(<*7&1#h3RzC`>F!c4QzB=a2P^GwOkgmq0oGVmBtbl%A?<F>2!W!
zM^!df*YcRZ+_QRvj~pGU?-bg6CZ$A`0@eL$fY%BiWrkJC>?ms(j>GlkTfljNx%kV&
zjC!i%Bx@Uk0+RwRFZRMeTG@z61cH@mhju&YlCPQggCN-k{10d<W}CJd<8QI(D(Ai<
zQ415jg$<3lQB9UYjNHRjO+`r3h|z*yw91F9&ELqSFZ#o2T840h4gP`kC|NYj<LN-A
zOfyTo>DuU~QKz&0)LT9D&f##rINiF?J_D4b4OZ2o#9?G|t}HjB4RLrPF34H5Yg0O0
zYch0kr0?9p9@Mq+nmevtKio%9|Namf7XNG|SkCZ-ZYY02+8M`GN3-z(=P*S0nGV22
zshqUix()n^@MmXbx`<jfN#Y1s>vo-L2vy007HDD4Pf_l^oM0#9!u7(L6k*mWflfED
z*B$C9FDIhv%_rOUV$vr~9c)$4j4CGFe-&LDRb8vLkL2YivlwNy`8%@+!j2J`d6JWq
zmqL@kuE?~=(zHmN@GjqutTwOp{o10MG=dH%Cnj>%jfmj><+V5m50j4Q3kHk2{1~(c
zkNF>u>aYq4^CTlx(>ypgV~E4Ucjh-k6trqPB)H5*&{uKPP9{$;%Z=*#@mN-U9id$*
zK%s61T6@FmXXS|-$gT)j4*lq>@Q%=YP(&}IUdVo=RQ&@#lFXiwK)ujvh3fYxE%2pq
zkhTx*evCqlO4ZR@)Lr5x31%P$&q^rZ@QJDTzMmfo1HKr8o|lEew%l=?56XcPidj$#
zqN{0<q`W+k<NqFK*Y)B4{_U<F13qx^$Q8U&lNLZ1ne|5jfmT~9%eGXw;>t4tNVN?g
z{QmF%^*Um)l&eR^Fkfku%mlcD;9g7Eq~1q-r^sR=JvNSf0nr|(w;hJ}(t!H&Wb^T6
z)CpP&B=v(?Lp<<bGK|SW4&++H_s_f7wB@Uc?IwEP99QWD@8-scAo50usXc=ZawX=k
zH-Mv>rttrFAdY3w@B}T9p=@AG?m*fKz#vx!iLSGWUNWd@UL;(5&PL+Jd<H-4g>1GX
z(DUTiR)@Wr1)z(4%(g`?R2$on2+s#E(`qjYky8pB`ofb%%mw5&9ImU;<>Xo3&M8aw
zKbKBFn>Z`MIQ)}Duz>1IbNa6@@jMc3=+N@<bOAsgBNM!X0b-gUV_9$HzX$2D9sI$Y
zD|-u16B)5a8Wq{jii`|Ln+gW}7ccO4=YkPm5(}N2Tms7=db8u2Sx!`+;5K|Qm;-?E
zp!6S%2TZ!<i2pw5&vW2Ezd?Y6tQcnODai}v0$hQ6pBMkkkZfE~m+}j#v1y&+*@ylL
zgvtNwi~si-wF7?~5ib#vjQX#i8xaL;FcB!$v2KIlG*$(9+5h#^|Btu$*Tj5&xcFay
z3bp^Mm;bNFz{|v*vg*Bh_8BOGcZLq<-T3cw{y+cT|M3V|5z6`ZU#7v^S_pgvj-7z@
z(f?c~|Jz5NhX|AYDVG~8wUnQKz~s*M<osvc{*RA5kGa%v=eyU>e;VhrDD{8(F8|{_
z|KI%6?vv0wQNRCqns|NYz4`CK_C-$?-O~q{y{@UL_y^-_{cGL_;|oztOw4ir^Oj@Y
zho|<d<sVPH4;8nhqiUW?*ZRr4?JO5f9ak5|%~9YBw2y#66~k}Ef0D4VlltKQ?>*~(
zKeFEm#lY8ac;QO<dg%F45c>UlvVncB?dmU<gwgxT@)~_q;+IxqwKn#cB#Q$;6tK9+
znWt-B5o^BFmUyqs!{aZA<Gh_aV{%Us^_@`aJ0Q`q8nt-_akBj!$4SKTjJ-Hqfy*)7
zX^#hvJ(q2t=>2p9A8uDV;94ex`_q)2q-~56^hJliS9-B0XHy|Ugz_KHHLqChN)KZa
zC67?~cXPde7=zS~bVfvzMM?t~guIB0aE8Z~w++&={!@0V!Q~@FOqumMRo<-SUq(ih
zV!meRyj|McZ;CIXY(cgDuB??g>j77U8obx`nHOvvw(zdI8mLAcm4ib=JpfG?n^d-v
z={<ROVc*pbjL|#4HaHhs+>k(p<p=%3zU?lsy6y-5Nouyf_zz#gzfT7&e5fsL?P9dq
z7uZ1r8{^>xa!PVB-0nawm#B`SskYwd1w@BfZrtTi%3~HXF$+i!Y{0vj4q_A%zEJS1
zI*}$5)AQ2w5$bIhaJ0&_dP31DSLW`r!*ohD5uVE*WMpUnoT>1tT`@%n-%snl`ryd^
z+Pp`m{q6%Hs}F7%R(Flk-t=|u7;b2|S436k4J_*!B8lA8Ox0t<@ajY5?-ar9sO6!@
zqaAf*Us<@iPF`Rc8Myj-x^(dOeaJFpYPEmXtUnl=VWdMrzJ=_SWXt`QI@!au*-ZWY
zqo!Cndc&UxZUo*N(v~Rh4lSqsiC%MvEZ4@Jw>HV`H3%mG{7}Ik#r}=rySVSUUW!nq
z4{hIlurVBg(t;9NH7tQPSUfPXw~yN27ozGdpxD<#u1xpKKFdVBC7S*m9sG?ShglaI
z@Kye8RAqusZ)5&vqou=we*6u*ecyZx*~?Xry}xEWPtn`|I+4KOFE{Mv`ZXQewGr1@
z4brH;C00~p9@%;olKrU(6?sU`8Wi?T!{O7ahua|>2ZY1F)X~vl(r=89Y;@in_DGe(
z7xH9F<*^S0{UDHu_=-c(n9@xXS@mIYb#RG9ouCwO_LA(MK?D(`qEQh1*92U6B_$=n
zoO8TW&lb0HU5d(z{VhY1QCJj9k1_`tL_>vus!hhkp5<i{#>A&kN^)7mPub*t%<o=|
zjVxmISk;^I8HnPBwd+?7r<N5(BekHY#z48gHdMIe!JvQD&mGd9$1TAs3S8wZKUzAw
z-~G{D#y7l!tF0U(`S!)!`#o+yTbl}4XVR-*0|JhZ-7Q_-vib2kBgy5r4L#^pRMG|t
zUUzj*C{liwDz-ED0E=&}OR#`-&R_n+^gys)c8o+AjgT;8tw)xORmBLWv(oS8u3!42
zjvFsaW?NGeVSqn(?5Vq?!|^_t=9m)ybsU5vzw`~m`i;OtdXK$`Or^eWsPKJwndAdM
zzntfF_8{BpD<|l1sORhu%rFdT51!O8s1VY%`~iN`@kX>Ljt_>z4R)|{e(|C|Le}X!
z_8o=GQgV<Nc>?NKwnkAD0GSPd;-c6jTI0Ssw9dbn-wChnk!G=*x`V2r?TlPxrb~<e
zaePmk*z;EM1rO@zvVemYIV&XvVNsZPj(#BjnoCIKH&JN9CiL@K!=ZQvbM@_4;00hj
zRIe7$tWd>H056ATsS%<zeF}Ul&X?rkqQM_OBoREkHOoK=Yf;S7H=Lk@PQ$6t4v{fa
z`^x&p#(v;86Wi2HVl@tkiHT_%Kl70X+v<0Jgnz=b<*ol2o~c^Ew<$@<$1ZV^f;WJD
z4t&%CjH>HGL3WRXq@c#63(DOEDEB>)guGw=%6zd)_ehaLzq9(MM*K0IX%TofMP3RG
zGvy$E?Y-s#YLwAOmg)~M&<&T6t9qY)@#o?ge-INSrhT9U$EMgTD2F(?ob8#fvzpHZ
zjD-hgfNEycL(nzcPlTnfE8r#yTqV>;5?R~Tl`!mPy^heZu<KntA$Y<z^_5%qzW*G(
zs>RQK57jm1sh`hBtI_}X=-`9mPx)j&@jaK@3gUjuk&>@6Y9FsO)Tg4QHXmOVhZ*+t
z6hv3c`rQhsHcKNZXToi5ZO@dsk<5p7rrFoee^J}NNvIZ`gMHly-WH=8|HRiYqz=8$
zVO+2jq85r@3H66+!zO}bq+=6KP>_MQo~IEWAy=YN#3K_)(AQ7M>n?OCy|`f;`cDMb
z#{L|(^k?&Wnt=EH6oX0M5`YodGMADvD`d*lnhk6t*E}{*a2_7X`ftgFH}OHv0f2LY
zJP9zF=SoUa@O}RBezf{+LuWKWyp^auIP4kK8-nU|wQnu8g7}+1MLpquRijEI{YQL2
zLWQu=I(6-Y`U&vB*O;1mdfs4gsQ>P4o=b9+y!*BB_NjHtH+m%A>eyipmgG^(6k-)p
zzHJ?f&>A!z1Q(N9Y&e@{1G$^3h_2;+wYqAANQImUB=R@0v7vnaD6uo#Gvk-#+Qj0*
zqOwNsS$LuzU@a4Ij&0|mGJlulCFfB(8(D0;=`}Q1sv{g*H)~io)h)+66EBfZFx^%U
zyhs4>pZa{M!Z38V4FR+6o_fDM?$0W&0tdv_Cdi!=6^FWPxI@39e;?|#!H(H2ehfvb
zQd!l0KY`3G=tD+Gua_riIXXUPv@&6IAQLI=)D&m)@Z+}}dSz#$2kv_!p^tIc47tve
zCVUAjuQ(YMgj5u}m(Hvlx8K1Fek&X;E>Cvok)hL?Lv2)MVPoeMmyt(YKU(gqFee9E
zSKfseriO;+q+f;6D}d581yHiKbc1|R@9_MASWhB?;Y+*sjpsxB8Du}~On^87P&E{6
z)gtMJ1~DMBqMws9+j7Aw<aG|eH(ddfKs&CLnEgJZ<=s?1IzwXB3qfx|3rkD#|AacC
zlGDnz@SF$2oUN^`1%MT{Ue(!{qxZWsyJZ5%Sk<+SGCl@ED5U%mhOO><SpGDTL6*#V
z*?vncbQ`*#M+KRgUju`-)6Td!Xe-_Rvm<-c383&sG6ZqAwk+S_EFyP}2Cf|*{*;%d
zqM@l}3S`#jp#+i@#xMwSim_FZxNWp$KV+SfaeTU!Q+m_4G4PAeV~q%fnC}xVhR6lD
zX_U{c44tRu&SeJ##U{rWpAh)^s%O8E<027sM}~revhc_&>rD{x<L2fD;`^jnGltBH
z4=2Ic<OJ0=m)(vk7qU%gdODGhpefX$?BbY1)8mY>sWpWD&M1*dHyoH}6&n4=dZBH#
zt3GFBFkc?5L}yLkQ<bgQ*w1%$N-JFQKdw%|K>LiNyH{3%j}!96-$A^)Cj?Adg3VWi
z-@7LWYOW5~7q=n@<v!5Nbp|(ZGjP8TpU}EL%Q8$UDG6=$z%`|fBBy7N-);jbd$Hi#
z*K2QExD-c5j+sPI){2zL))#dLsnsJc_#H+jj9fP|1pPl%8n)IRN&4Pi%rw7;jyb<5
z!?W-WjTeaY9k2I=OpnG5-NmHjT$h7nN{GuF!AlP>8=N>yBhey}Lk-+bABAc$3K%RH
z4Qh3{oesacQ#$=FY@WtM&&<@Y)*L6kZ%4|He5GGL1*SmS43lT;LYgp%48pr6CXjVg
z_~qUx4+K@yHoKmROsMB8n7__C5gB1tbgLB3>8gxOLic*}AQ-uiQqOf>u@!|XQdTg%
zz8~w&*SMXyPY&3$stwBcnx%&*I=q5!=igRio|TMopyOU3U%ek72Z;;kS|k-g0Ovx%
zepp)F-9X&N=rhq8!T84iZ3HzIJ*?UN65T8>bGU<6k}z&dG2tVJUF7V%F2_R~Ku3VW
z$z=n;#>oXXp7={=D1?Yk&)6RM%3#CD{!LC)>IPvtq%XBK?+rpFb~fpto34WL%9t@W
z)SY&Y2t)niBVd^sa=jjEtLr(;UFLv=ff1WJnf*4q<4;T?S2REsK_siv-I5PTYY=>e
zkX62Cm5qsIJ;*=!KDuExH`8Zgn|-C^aoF)I;J+d!9{DWna?KKT^t)2{yl+NH%$A>=
z2fLGrPcIun7yp?1MSFy=_>Yd0?5$9Un^Nz{2o^xKLN?po4`;)SXR2o05d5?L*^aRY
z!1+MW^$w}s{CPW0{%1SRidx~@veoqOBxc~Xr}?m!uq24?b)7x$9d%eRe*MKSf@$31
ziZy*8Mf7PLueWUS8xjLF&y>L00=cZC^|q*ueXscbr2l<Z``(K$>wj})U+&5u!3PFX
zm>tb}V??IDi=a<^YsNKQ9M(QVb!mj5U>3<-bG3|Q=70zNvOr$8_kP7UEnsd**FDit
z>VD-xMAY=Ef&XfB-+8R{>OLU5@xa8|wrHAV8^QyM#W2`Wj}(4Niq96d-*$KX(BBZC
zaQM7_e%rvrmu&+D>;^}f!mlE|d?Go1{<Jx$t>bWVT=R((H=}S&y*fbUv@P>FndR7m
zlc|(TA@|Aky~2vd`FzeutNtQD?wiLJmb#g_+F6gXd~RQNrKjsI6pP(*vC~&iovN@F
z0EjG(&poUy{dIPC8fDK^O$U*3<~ahUg&$s<t?0X%{$8!e@=vp~vnHp1!W^BPwDe`Q
zEpaSIGlPK)RCfiO<%!zbcp`Rq3~<`h_sFlkBX&70SNo7IzE_In@2_-{s{LZWF112^
zlNEE!i6Can;lRooLBt=25MWyiIJ9QF6A%IzQ5cF%P#l^p;tT7C$FIKOruO}z>j6$)
zLn)o_CHQx9dp=jUoy=oTU}hGNU@Zw8@wZYdMS0|5Sibl2ceh1zb$IB`hl~7R$5c6(
zAmA#Fe#Ump%v3!Q`#(K2sS|=n9v>f*IUpQ_{-A0l`+;r%nj#p5GFd<AbL^XCmu-lb
z03|B}j}~-3wD;o<ISQ*`K2KXknqcb10URge#VS7c<JCoHb4OMOD{-umUKmF4)lgCP
zk+Hh_(}?x#x7MDnR-(*c;qF*!B<cnHQ<vp&DS1^@wEigKo@piT&zH)6+Nz!5159!+
zs(c_;u7B0iUe~!Fz7dMnCYmn6H$WK`q{|$LH9eVaW^#CtSUop-hxK^-U~`Dk+qGn0
z#=5$(>_#1&_NMy7L!gFd!h_!HxX%^$<Jk$u9dFx5y!t+EgvG0I`@X7kKBWF;k1P&A
z<{b)0UXIMRUyte+W(>M15rqC!a(B9Cc3kdUDn0mw<Q_3%4T|`#W|9|9@xdYlbUs<T
z#iOb|El4a{FZ}%i-+E<WPek!o&%Hd*=7QVbK|WNeClPxaPmP)<0Qy6042puS24319
zba&0IFnNjJ@7MFDXxG(*c$mjpJ&@U9l}zL)g5`pP79@~dJ16wQL8y4sdgTD`;Q)d+
zh*@6*faJ)hFDY#N=!#77k2>lIQlQ<ma^q`>@5c=aG_^2@Lk~%l8k793q#d(eot|bn
z98V7t_L3HVGyq0PQ5kpn^S>59;`8Fidb={1{d_g!q*bqfMNOm5;;<5~m(f~{9C2(r
z6LYWVM9>ydM%Y%WJ#erg`#Thi(c8P8P#_TUe!S%9&Ps)xhY2~w+M4cQ_2cOULAA+?
zxr)3HN8clo2I&vbusKd+Rdah`ztr}r1T#jcUfLlM65`M?>(t>;R2t97=UZ4=*6c@y
zY>%zKc?a)YYuxpFjP&KG{`bLI==cwH-qWr_BZ-#29RX#X>lS7sL3i0Euo*H9jUFbn
zhxSX@^rKi}9ku3Z^KL-V6Nt1qSPd1VjiL~_Cr#GV?P+p1<kKnsek;Ay9=TDbEhkXC
zR2^IG=3-uSbCe;_Rr>+Nh9;WB$ORect0ZXf(I1Cr1>VO~#hSrqy<6e$D+(rm&_kn;
z#xBWaIgp;?clC-bcmARcZ!;P7P;M)O2q{7jdvhfS$!H%L{McRu$H425p%6v%mMsKQ
zS3<DF__}jvDDVNFHO}ekj}ukDP+up1l`f4zU;ii`yTa9rCE8zCQz}){_tl*Dg!GSe
z`N$OmNvssy?C7puzbB?iST-4}M{4T`r*R2lCHAtotSK<tY4S|gPBv&`QWJ-3<wbz4
z`A?f)<V1l&-Bm#-mVh;{+h5H@VTuT+lXdFezW(l?%b5^ruJ?<frqX|&$*RBQ@eu&C
zsI$4Xb?xVYH=px4+r-4g+J**7hJf4W5^H*%$DLn!C&xLrq~kw-DkpQ>;xI8W9iMJ0
z5b-&F98F5^fyHC}4=+GE6u4{G+x>#V+NWBmae?D`KTCCkh87w>I4HN+=7wayP%o~j
ziN#{j96CF@&Bn<cl+{UEtl4vofC(lMVx0Gf4{N8JjLp7xIJvobCmf7O?v0nlt2!?M
zs~S>-@n+C#XFO-{r(eN|spj~E(dXl^oU9nAN(2OiotY#3xgNsqs?k*Lo~D&ca~GC>
zN@cfVaoro;uQ!vj#CkP$A_<du+)}3%xqRil-S%e@mRdc`ZGkF=mno9qk!nPyUQ9~7
z@9B7mEW`juk^SL^0n_f;<w?Si{X1C5r?`W30fgN*A9B0m!%n+G)w1mbyhMZr{G)tl
zJTFeE?<&>5eJER$L@Szpq)F+M578?1&mL6I-=>4y754V^4F4o~IhtB`MBpmdJv*gI
z8EdvP_CXSpzMj<YPAitV{^|S6`<B}opDR1@2L_U<g)!g*Nwi|M*UdqTYGSYtziQF4
z5+!}u!x0Q)=JBRw^bu~O_qM)sNIKq$3QR{!s>!JZy}1v@mS%#A>LJiW-zZVzp|&L)
zlQ<~l6dx}3{GE0!@Z#1dP0Ss_T9ka%`hNyS`kNgQ8wgcC5*3PZLTZFQ;6+#8jiKXa
zM2wv4R97>=c+=WW!C%T)6T1BQbN|KdOr_K?nqg47<XiT2Y0h<}IFDVw_r8ZA2dL(Q
zx0q49T%88FHjDsV1bjMcO(-6B9J)Q*=<wf*;wjw0()pfRj$$(+7ifeTfgw$~=x66J
zXRMG%5KB>1H_|S)VZja2k@-Ia-g;#E@OF;y0&heYJn@xtc#v6rX3W`I=$AK^-W3rQ
zFbvQFJyqe1kB=J+m@qmGEu+JEPJ(?`*u%9gj{D@Hvw7w53=KlM#vH?q6t&U)*otK0
zJ6HNT6zI+@L{d@{$$^Ia4|q*gGPM}0QZjMOX?~v|KVtCBkO>B2q~x__cz&QU4nD}+
z<_I&}owbbqplohf_l`@Cz|wlc`r}ujK7eU2S~>igA}}Zvg+32CS1;0c*2QkOp7Ptg
z@_b`fMjrKoM`FQfGRyGM7iC7sPRS+fpv&6@f}Mv<Vej?!D=h!2^S{3d>FI~ZocYa<
zdH%`men$fr_dYZ_oKZZm?}iEfv+tt(#Tbe~yTB^))9(w+HKWY!o&SY^fIjR`R)1~z
zf)cLyp$o<(mZ;1Rs?>cuS}jMMjqzKZ?K~AX2Ich`37vk00G)>U%+wnj7bIg%`av?S
zRE2b)S4v(Ub{*t}#80dIk^EZaT{ly_-@lo$kg!Y(6%2J~fzB)8WV7TT52-8mk@;~1
z+q2yBHG^mAVbax|z}ODdhI446FsFxH4~8z%oI_frJ6~phggF?s$L#;8+#Rmu!_13H
zEVlD<H+4zsFrS7-&hi?sn|9tKY70mF^+L8pB&+~xY)a;5;WDeeMb=J~&zF0xV)>yb
zh<0-X;`S@%th@60?^<*WhVODdbUxaRw9-ax{9O_Jok>j^MKU(*`0GKusDd7zo=1Rr
z+2%Xh><3Q@b_T?|t>U9EKQW7{G1ta=Na&R#6+oY=?Zlcl^l#Ru&<HD(GX(HN?!9@Y
z0P?>^zF-6mBBz)pf=#WEXkniZfFB}{@L8@0!^>H`4kSM_wM4%$i=41hFEat*xIg#{
zF8g*8zkJs|LqGr?uOCsG`PHE?=$?}?zZp%+OA?^PKnZS~=KT?nKOv<Thf7Em`c0Ob
zX}4FTAM>d$)M+HXi!$y}D@H_Cd#w1CKRtLfSS6&T5mL&nuO~e1&?-7CHU}UljFFi#
zgR3rpX1_M+WHG9Rrj4>?QBhL`(?W#3_Youd`M;`u*u6$WtShLbQO&b_IAR2x?#PVP
zNpW$*Mk?|nl*C-7*{HD`cC)HeWjd(!ogMe5fSDax4GY@fj4g`}M47Gd4hlqPjmV)h
z5Vn|~LSGW)jNP$lSmor`pesK%BE~ko-!hVKxSFpE!ne?Iw;yQN7{+HmpIda_addB{
zVrED8SIR1{m!Xgqh|KC8&H_7|>C<AC%Z!BiL|xVBs%g0<c}igfmZbkrLV${FVAZQf
zQ&v`P_J0!TkJ1wI=h78_C_VC(p)C(l_C=+ZhYImn?ro@hswM|@@vw+w>X&e_G@7s>
za(KfrcwFC@ZR;Eq;2B-0lF=3mR=#3P=#uF%4cfhjgd0(&b}ueMk7agzZ@KxqxhLei
zADbTkQs9CWqB&f&T}{Kr!xEO3L|N_dWy`s4s*@~0*O&eGl>c@4l^c!vZL|Z70;7hM
zkn;B}h!wTU!dDeK!N9@DZnp09(DI@uInjlP2>I-RnErALj2rvDk-Xp<<SFaFUucf#
zQM{KrF0s6MgXKh?goft=cA|@-G2^GjCDvA=tcR-O=Ch`nr3b{O-er?j7BlJW^tE-~
zKx2H$%t7iWHMxS-eX3#BjmbDR7Pvq=LtoMXyY(y6I`h$j#eI6o2yqvJIJsZjep+r^
z2_kv6RNfj=h{9R05e|L#UK?i6mro^+ie$M(E6LgsZwnQvhu_AT*B6CJ<vgf7F1M`<
znNchoAPAliqycZq%U4({4}kUid3-O&nI)iU?H_Rp_F1)q4LOf|{U4&FAfSv=GYS*8
zS}93tr#g>%DdmAhm3<Jiwts-6&%5>urVl|(i?g9xZ2QgREU7APlYL*0i&cCoydG+N
zWbF~|(yLvw(||y7seZYHWel6B$qV}|KJ8spw1ANu;X2z}i+ly>pis8lRTzA0$rzbW
zvjE}q+f=mkLl6PZGPg~R_|@3lUPuy?g}6ygSP*{Ebh%9LDB*ktosB77)(%&R{08Vf
z!Cr679~-AL|K$*S!Z}BszjWGf>TTlC#s1nWBHhNfy9eP)skt(DQTq*@R?3hrDf%<y
z@<173@0zA!O{3GZHV&F|rd=Ly6z{@u`cuYM_nkTj(&AMJsZ1^~_9<3vq;;&YVMzS$
zc=E&hJ6-X^NE=k1pPAmAZdNxoTBf4qzs?!5eBvV1%-sc$qGVo&4t34|Ir1zZG=xHI
zi{}GjkXF~%BlLM4OeZX)t*N?(Z{mB(Q#j|UK-Jl+U3<Bdh6HTwO1+`boP;v%wQ+Gs
zcj_(tOaI%O%M8I}E~mJClO_Y%`^$x@>iYUmjp`5>a&_cQ%x^OkO@`v2#Q8Si_;Gi8
z&av9#JMrmW{+o9*jl^E(`&}#n=Z<o?vQeN46rN(MrltmuZB>wevW91U!U(L?xP>an
zQ-I^?Nnafts-wk+<+{?I9VaHqKNFXt3ByS4scb0;uzA}!zQ%DA0l7?~6Z+(WEj|c8
z9I;1QWnOdbJW}6X+!u<Xx!=Uib9-tdm=%Z*f$g&UI-M1n0s;sH$7g3fUvP(ZCW}-$
zIv9$LAr2pX2>dup`Q;b%{#eSZnmCXD;ZgkfxE%S;UETjFaj)5lA4+q*aGMQuLi*fi
zb<8Ei+RQ?uerewkm8NrwG66q1IpicDu8dE8*1xaygC*WYn7Fe%7%_mQHt@19MJ{Lc
z=!z5j4-!N`Rp!@%&APhdk0Me_NFuWJr>>?AJq~qxe)d;o{vwriG!bt(EyfDb68Up&
z>aXsiZTo8NL;2>bsg91gopf2)b<eFkZ;y{_taLTgbp8-S@7|faBnG&>8&LX`u7MW+
z884`xohz2Yty7&vi;(zeCYMYd9+f;Osy;qZQ6yBSD*gv7qOR7|Wz|ZRoF!w)A&eJ|
z-=m?Vc7G)NT16R4n;J3M?+a+H*WI)1;m)zH@4`@6o&Kf9O9ud{JsirdXXII{<pV<4
z^BUYdEWvCpl@yDEg27_TMvgqd9QYWDr!@T-!mc=*Fo;88*sFc?_3*gF5RSjfh@IWR
z>AU6~_QwQVjNUD7bUZHQO&n&mh9x+=!6ngAvD3P??{))Dn5Js|uwg2cH4cYkag!6I
z9Hc5KUCOEQFrlE4SIWu^xe#cd)5t~!sjtV+<?HhEdLR(`Mv}{gYu)O@CzDeYv(oUc
zyH8alHmJ3br&0Mmoq3l3gJnral@WyURD5~J*ms)joco`(DVwi)iJvg^wYJdsDp+6a
zMeD$h5hfu7!k<+2efWA)HhqHLNq|||-M!~Xkxb8yG%Dr$uz$PfCGy?&gTYSVsrn2(
z^!cs@QAW&H4kukIEs6s50{L<-yty&aCuD}*f}z{XgD*foC8OEGhkZ?uKcxtIS$(kp
zB|>qM@Tr9A7X>BIx-*rAx~f$wN6T`*$mo<`jb({OhxOZH@B-jm3-ruc>k4)OCzh_T
z#bgnqd`|GEPi(}15eLelRKXE@rXVEqueCNUF{_x&_EOVq>o}G&d!G!WhE^gnl=~Yf
z&a5@-bB<v_51V@eT9|Xu2=L(U(+IOdQ`RQRErgEmg)2@yl`oC2mG@3x__jLjg-Q@c
z4VqGf4HXlu{qBsfJNVRLmK#(nPv5Fpt(PJ4*|^8#a<B0ukGhjrO<R7(x=svgUtLPx
z)aNHIZ0zX<s%SQ-uGrUWb=ym1wN_Q_9A18L!?izA#3JW&r#z^!zwG!L;!Auy#Cst|
zG38My3W2!Lp((yJ#Rz!Xu?=n<%^eDfiA_y{A*WGKROF{hAt%Ra(mBXlX$buRr=|^+
zu#;m#pW!t&7S0(fF+*LR1wA^$%vZtr!6ft{#~Qgt<(;wX-;(k9oC(=qD#+e8D}6nm
ztu<pZL_A5iLCQjS%cEVQI^&VYh;Qu$-3|*RqqFq&iu_N+T8Dj+wx4Ryn=lws+-dS;
zmxn8%{goFw&LR8k&FJZiKiFQnPDHa$;c-Rfgkec8RY^<d$1^)btK2fU`p7kfou~Cq
zSK4wzPoKU>V9&GFqSnj56i+g8TRS)!Jp2}^FosPZCb5@y<ilyj&~fx|v&{pVG!?Kp
zIEe84BR7wSJ}<GOEzmjdAJf(I0+Vj|6x_Hh@j^mh@Z?UDB4`Vp&^noGudMh+*{!Qb
zQu27eOSrHiMgn3?C4U)u2PlUz!v1TQ?7Q_|C-f>#jc_o`A}J@Bgr_G73S?bgO4F9(
zs@iZFuEg$&YG@<?%Mc%gv`x-p0Ho$gJ!+M_M7jX&2e|H<bO}~+at<gJ2}$%T6+DbA
z5iN$76b&+xH&VD+9ToOSJyq#)LzL>dM@xM*2@~C$5;UrMW>D@*8bAHWIO1GkUD*-e
zs;{csnT@`hV1q~f*3Z#~K|z675ufDgN2sq~{<gCDQP}&^3|x}qlKvcRhtO+a)2?cT
zVU=Dfx@EbsGgM6Fg*cOf<RytS@noUGSAF0Hl?j4<P39Koe|@}GtRagI9cP<w%HiGU
z1)BqyR)5Q?sOwzoN5!G#erPkJz5Rv=S}~fwf8P~&RVsteib+csN=MHKnmEyAy=#>*
zTww+OmFq;@EBNRM5FXA+DtdgPk9^^~f8_{on5%w@w4&m$eh$ierJNxW8H|)4JjK0M
z=YBwtoR+q+6{(+mpS9CQ#1y;@<6A{`K+_f2xf52aRYrb4kDY#F?t8dr4~H6*VWqps
z*PQi0=~9OGv9SHFHfzjb4m3gpMG8T_92(-&_bwL<Ki>~B7Rau+?Fg!9+b}dLX~uA<
zPd^USRJ%LIP^lwsxO9N7Kny(+n{Nj@?rY&VIJvVXSc{D}uZ{H65;ID(y?j}<5v_YB
z>WU|Oag~;<dINuZNaQAemRGdEm}cdBy{0J&lY<+fvyasEY4!cjvYF((fK@ar7^W<z
zPY~Q<0tk#|$pOO33oE|nw>^uw@qf;uZy)Y^4t+W)a6Krb%S*(m#}196CH`zcYa10;
z%YS%tfIA#@QnvnC{`aC-?h?+lWqDg_yuFbhi3-UZt9hG+FR9C@r5ZT)A}D4KUmsPj
zC6@-Cd`#F3gzew%Hg8TwzcY~Ea8FQeN@4Z(|GjVi%?mG(HJ38(_^x~M+k>Ew5$2<K
z<u|GP<9^&%L2K*QL3JUA8zbZ}C+4B-?}(lgzspr-WM6>$|KSb3SiH1^fJ*)q6<)%9
z?u7BV!8bn<BDf~oDpxV;?;XSo;jf);&drIuk!rh&6{!-wP%S)My)W0#y82)^eiwfA
zud@wC9K@F};s*yiJ>YEf1MAQI8QGXIkw2O09La_J>0{@jNNQje6@gO20R1&H&xBFV
zezd+U3cI@aFaU4%RivdHH^hf2xMZx<zcfw4<~Hl2T`5`S|1CUYfhCn=f@m#ZhcB#N
zl9DX<i+1(BsqL55v5rqcO%lawY{T~-PJ}+gSRQ8bG3a)T#E8W1O*S<wEo|w$EXu`6
z!@W}=jl{z66#XVcPE&64K}w!tQW`_iAWk947qixb_dTgrIwEdFU_yhSM}P{k^c+)7
zERXn9hJLft@e2+OB-xmWo*rVCEjp;$v?jZ*M$x7>Msd&7IM87?Yc(%gU=IogQ|Aew
z8{==`f!9Sl7u+8)L$M|rcb<$htaFzB41HCt$(8u1U1?2W|LLQ@j3um>PYK~0DHn~^
zgy9#l*=;<n-d8OB(OKCwy3a?^jnu-tn{OwXR5<wqLwS?M8%lsGh_*J4QA%cuO}L%t
z0|X@PDx=r7wv3Cd9;6%nn^mpkMj9ls7*Vf~wzfK`g~~{b+XGz)PoX%wa}b>HRjU4S
z|2nQ?eY^72)=&D6(kS=*t|P2(R#=NK;efwkte|n4c#ny3Bh%#DJMTnS971lKP@<BE
ze@A@M<G*wEa;$KKTe)=o{IrVJUFYoQ7p4(eLM(|H>YNod6H9uwxmQX5^t(cntoTO-
zU@m-bj%VMBUI>9@en`e^6Dtpu<&|=r8LHs=;aS6AU+B9QAvgEcVYhc(IUh||TnZOm
zp;YD(u@w`{R1z|!czh}(K4OB{M!9X01{dk&1~+E9U$HCAQg&!v;+K1@v|{)IEqX_>
zqhemHRO+3Um6rDGH8@DIA@wsu&TnF9-!SdTR9|NO7V9g0uh;|`>$m+@<^O2NX7FQv
zbuku>59I5+`m+O&ovJ|oM~f5-8>Cg=&CZj)*3&k@mr^Jsk&#JVcYJm8*+=SfV4X7*
zBY|?NK4h`o{#U!d<Yi+iPJHFvPt}9Il>!eSqrE2NC9ruXg7U8u=-F4Y13jbv8z^WT
zOI=}D2V!ACEu{j`^Ik~)BSpP^>qjZ6ub}4p(U*By4PhZ=&;iih2kmVpHr6Q%37`Is
zy*H>*M^rf3c8f)$*VTPiqzV)`f#Ezz{rg-&AnD>OhMr#oxI8TH>Bt3QmF8Gv6IskQ
z$9pZ`Wer|rG-qUdfwLpRmiX54R{2ul)5@qZpH|WdqEtRst+DmP(~Ok-P~5d^W1tUN
z_t#Jr-_hw&Y>>&&{l?}!Rx5A8WKBb?TMMuGd1Sm$D&KB7!hRti;q5o6Vcv&Za(3}n
z)ut&oyl}a&u(1AYamn*QdH8GEcF%wgU#z05&D5%B)<UJMer4l<z5&y_#sfu95fKrR
zDx=LmD4dXCtn0hGwdHniJYHpfPF10#4ykYvky!6h=LK+!!RxN-7AVogOt8b)A4#c8
z7B$wblU?9`zTfDH#v^drCFd#@5-~DRZ0;WD^PTJsFH(aWGy9YWtDXTc>TL`}%9<n$
zD@y)78^fA+1XMu3gbi<ddmje=wY;jo)f46}G6&LGq^UXe75q3#LZKrOcJS2i>52_F
zX9-_&3r#7nOcLF~TJHApg)L56J2DixoJiq;;KdC8WmWAN7b-zR#o!=Ot+C(YSBIf`
zFl*sETwEJug2i+^Z|3Wm-~=G7c#uLnJ$D?0n$vUQh6A|YSHbE!=eHG%v5Fssd1zo*
zIqO{N0-hPWp_B`?r$UTKR&r$v<w68a8bt1%byCChy;2<}XU-Ixba9y6N!2tFK<yzH
zCh&6qwGu7szTEq}F1!@J`>|jf9)dT-u43#>!=ch>HOal$ajU7^Bw<g<(%J%8&7d`8
znd<gNkjp<#Yo%vn(DmVKAn88Wao4|kCO<b@v3f#li@RImG)CR2*&?tUl%Y4!sX<*H
zlF%7<k_$R18zYxs!6I~N@Py7Hs?RI<q-s_`PZf%4GRFQ<PST%eyn%67)D}Uomjjvz
zqXrUEGaNL>OMM^*T{dds*GXK2DZ+Iyo1MWh98j-;rn-QvKcKV!mm5N$#_U^KqC)f2
zJuYznDMgN4s6{bj$n|sPaz_|u=kSbHOx**WLf=pm6qH{5r@H7|F6pVI1-=f3BdO+e
zr$6=RNtg5OR;bezql_z*hgrtfRaYzCtUBds=Ol7hIb?;bu^eW-@KcqBNSZw5R%2D?
z7$Gwua`=45f?PIJp%vzX<2eLj)N5WyL2~TzDv0X?_qSM|<k<6#t?lc#{cmn)-|FiJ
zaikRs;ilLWCN$cDhtUJiS}@!`gI*e7Ah;(BbhG>%FRF@ltgHTY9hOFmi3SCat6evk
z^)MG@jqNnsz5)d*MHfTUKZV0cBnj)X@Be0Di4VX6rLj7J6y>P-<x}bV<MSFF1|6!A
z9M)I~iu3#diALX1YN3{lq7B-PDD4a=wTcd2@7<?hIdo%DNQr88(jd$!jP7jvHdD~x
zxTU#D*Y1|xtO9MfqtBTA5Pfkyc0w}+BQk|_Nxja?FoGa}IzxKS&m&7k!Y(UrOE>gp
zu4d~+`iz(RzPNL{L3YECTDjA*p~RvFTi@<<l}Ib?r&HT49GV@`N^E%oP)bz#%RNiA
zgpHC?dbcG#eP*7)&X({e+0mqj=B=_M4``j5&hOGsws9){k7xX2C75W^I=#!8m(MPA
zB+!T$PM(yaKz^?=+uy?5KP58%9mGlSpzepQc;KZJS&W*!OQ{n}V_@j$(Tz2Yk<gbR
z;=;-6H`x(*l3}BJ8uN^znu-<W$6bSJYz@`%$Y`DPS$~@Pn}X$XxCy$e-X}nf*C(x8
zJb>~|w7ck*OR)UW2(Io^8-6M0Psy|c*NdY6SbBJTb6|#n-Ex&E$Sc>}i5r|5FB=xT
z!ya#gVJwktgzTewCZy1&mszJ@$=-)@{A8h%vt}izlYR<__%o>ZrY~w};~a6`R*w?B
zipoZ%t1CTeZdS-L-!yFbi6)VnXQdv$1DJF!RA2MKX`+1ZML?9eoLQMRdFt@}n?#jY
z83{TiCb1Zf7yKBoI|Me$)v#es?IdBa1SJY4=z_M_`xLHkwuQtUPwop86-<{>x#_Ge
z@*CPPVQ%T)sWgHv3`G1C6<O3I#)&GfFMGx8@2;M3l+t*lB_yD3FErkTTXTX8(P0U|
zosbpFW9JtZuHCJz<ju^qCpV&48%Ge`P|}@AM==|<!N)UeqD976b^Z-&JeaT9hP6J<
zFWUB(c>6Y*90V#%<OdkF_kCN!zuZG9^ktOnI?S!EVHjN{5gB-l@v_b|3llg%4$G4U
zN{=+JymrO!E^>G7nHM4Ybg$6SgFRJsTduaFU`s6UAe@Ncae4eon`C=y6MlaocQEGS
z(nTz3`C`@<D?k713LvA&<A*7&d@s-b9hZ_B8H|bguC4y?6+!pXkHq$;AI0t8%vRNj
zJMas}l~kEqE#MbjiG$c^F#epDuo2zqcRonFXjCUbH->5EAd_Io%utjR@5Skc@iJV<
z$WSeu?N|)4_v%XM>ll3(MKyB+4N6ergC$cbEs9W=d%pFcx{^FSsFYD=w+MQ>Z!z+$
z)bUJilOM!<*tjeU8<p*Zk&>T8Wyo<iH~#ZS6=#m*4HJ;Kq#{jV)~*Xfv78VZZ#FJB
zAc<SfpM#w`VW-cfT2-(tm>^j|<AO%7dg_2p91{wRr*c!G$>3E-cYC+eA>45-5j!XM
zN#3*BnJGIov9IX4THq1&ncH*XCID(LZh`ZEK41dyvVXY^kCSm~rjV3iYg8u(_FD|-
zREbRyZ;E~gp>)FO9JgCszMtIQ5iI9YZ27sXT^5@R{YOUlpA(rkwzpU@a*d^7GvcYt
zGgtKfA6Q2suuk+oof@9RqVut-@bjG%0#A!oJ25lznvP&}TXr%{z?uN5f8gl_mi6&`
zC>91!p<C?19fSu1Nlpybe-0B8R7(CJjp45Ri>Wyyi`HjvT6T#AaB!Rz?sgUOLx%`4
zOld~5x~sDrj{>+%2Ia=&xW6j7DVv*9y6%j>G%_-3_Sxk?I<?_saFQyBDs~;^T{)N3
zvnkcXtx^8S5UW1i3k}-X1r;P<Q&47^f%56Wp<f564ymff6a^B&r^!4@YMCb>5%adD
zZ53Y#(2XTNWC8DzX@5jY+;J07n|(Xg(9+_&BvXV^%N0{5-hNY8)3yde@NBD2V6pnH
z9bHy&<xqb=Y=-DxWCaC<{e@OsR8o<N#`gH-qor0{y{7k(_4QmbWP|LIx1h<A;2)ba
z;Cw*^S6TBE`p%IX7#_&*s@g1N1Tm-4^dt7At>34V!aZ9oqKJSX*xIHd<&^Mo!uepA
zk_xbf!WZ2$+3+p3rz2jh2-~Jop)>hi66x1q{kp=wePOY_{q#-AfAwTjNX;ZCk`GB;
zp0a>VWIK$g$VzZ7xFYR%r~3oHzM?rHDk|!w0mWC1&js0jzinf4e+K$V87LxRdWtqP
zW6iMALAe!L_T%x@Fk_f$m~5E9Q08I0?L6kf6v9pG0D3-v@5x{OASxW7kVc&4*IVCP
zQ_@Z3Noh1Wm|5WsXs_cYS&ZoV>6Y{wMkk>E#@$Y#@IfN*#Q(`OdCa#4PZ*v4QwBM6
z_GiK6BkhdW^qVeT8JDjPNGZw(<#y7bJ6~JJs~sOCCjyN)_)jl#`Qr^AWwdL}-<hNR
z>HqKoT(l=IBhM|5l7mlD`pbwm*6FLdj3+^6=<6?M$`74N1?@K6C*pNi*mE?DRwUBu
zfgCx?O#?GMX#t10wY<{#p`0Op4>x@4S!MpTBV$(JqWXt-SS*JlU>jvcb`CKr!j;-A
z8eYUO;!tZIwZ$li#bsy09dyMh)~O;9Z2lF*QDnTGAUFjfT2B1O*|PHmWQ5M8t9Ndn
z3$eiA^a~{(pp+t)YoEmUZmw%hpkQK1jnz;QRRxFh;or>@d8iP}_9Ja#8fIz*qPbn6
zIM>!u;~M1o%4*t8#J9_?ms3qJ@Bq4WmGMdwVN(<xlE&9*xvy1s)$gx<rgtY04|k8R
zmj{KW@jBWZRLRrMmRP8k9l-_#{&lVJ_AiXU*%Xa`HWsl3DeKY7yqU?PUW!H$jS6-u
zQHSqzhp~Db#2mn8WB;U5f0Vxds~gR)0K?R<a75~jIc4v|8D0nut14pVT+6qhzsr+%
z7GT}Nu{zX~6w|plo9<Z(wK!IanP|WNg`u_J`~4OpN2WKm(sFw(pivTwdMF=ysi1!{
zNsb<38h|PrZME56J2C0#x@#6`p^lHMb3I-1rOf|8Q!hNM1V7!!_at6UE9@O6Oo|>V
z(Xdno=hv?_zO|ObuT57~T0JC-qB5MfbHqeo&d)e+^P(P`DFr=`$+Ro<Dy9}&4fsc>
zMo}sImSY$KrW$N@eC|(V_kAhYwJQwhb<dW2d-1?Bm9?`gpKk#)gLj8vUuh_^)u|hs
za9NE4LCKE1k(ad>tlPV2Vec#X#AUEI;w#{zii<P-Fxja#wb^KGX)*gdRibE@4pf<K
zJU&%)%_I&1(+>9of#lT03HrM(G!Zq!v8fV`)$Q#8qo0E#-+^WKWIZ_dvW;kOwuZ9B
z<0+n-nCl&znb|1^!`Yq#eCO@;waI)vJsSr{KX9y9JFev}(%WiQMUl$VLmjueKLi8Z
zv*g}p1iml<DDr^_@U6VQW-*S0``raRRYBN{iFIP9{gH$<<{_tb;*)=1$<<$II;^}*
zNTrj1Fj#7M<)K3Vg-3ruF#BK}9r&!JWloV~oWFE_nb>|)bn;bgc#$aIsn-9Hb@1o#
zh#^nA${f?V;m3oMYFWNoi=24~{;f-g)KJ`xfE!MPd<xm$OM9ZHhE)p&=Ef8{Ocmw1
zZh)BBv-~=@nrGy;x^Wc0klNe)CSJd~s>1;a4oL1R^yZ&9WP3q`EYU(6KPT3B&;;CR
zdF2Pv;uDp=f4dGk?XYZ`?7SX}MqqrC{^TQ<Cp-17;uCVLK}Mk~V7u*i>m}t|Sk*UF
z*^+COx(|iE8t@9g61l4+2T7@z#PlKP#@F^(p9+!5Y87Zn*RBr*Z8SnTNq0tt(<B`4
zN(J8J9L@I%(8rtBduW7G#VUAl<daLLP@3g&s66yHn(2zOVocP$@1nmDUMxWjbtD&T
zet@maLLDPs2h`}cVZm(M6)!t`+$iDRAzzP#N-af#UxWZ(;pEO5$qL8Ffnu#x&to`(
zc4gaIsfGUb2j*@GEB=Yh<<;l1W<rWi8))<vdjTEku(<EPIisB{W|$3G@>J(pbZene
zP%Z)12ocDEhIB5+;{4Br!XJr1X#xdYK~o**LS%6p{LcZMB&P=b`g2`2-Q3r4v`@yZ
z;h@hU9zJ1CeH~I6+#$MZFifh4Qd)W9!ATct-W<YMQPcbNxt=@ElY4$6E32OFX_QTy
zX){{Bp-fS)&>n>|D8BCBfaxpXrW_CoxBYYd&7c-d|KZ>5=wW8PnwdII&Klr(Q{4(L
z<=(qvwpGA!IU=Dn;E4Nb0yzr(Xvr)lZdD5&Tb*dEuz7eb4s14M0iUz{!#;)5hPL@4
zIjKB~Ys+&)D2=AX@Sw}%K<s80Kf##h+5q>|Wrl{2-p-y*%Y=+@2Q0duor6$Y^j(p1
zEixR1Gy}JN)SvA!T>xMNJ*)DGkn(lkR-bC?Jk19*ANwD3NIK;nuFO0!z)W+IEBO_A
zhd`Oci@QN)l=J%sHXlthFzyUEI2GK40`P`DBt-6cyKmfV$PV79%nEYe!2DRncs{2h
zrLW|BSiNdZ=tl2W_wl8cT7$}`xGtG$&dxu{ijh%T@9Ne9LrVcSFqX;`A%t7K>5Wts
zLPe7FQiqbyhAz)YUn5=Qn<3>zIfQ$dw2Y4xr@e;xS6Z@4Su=bf=Rze@2;uQDe7ipk
zlh)~aoIJm<fSvsuxMMk)><_RYZvpI(LQYO7Bk;F{Ywgo&fMjSygvm=wT3TAoN}`WI
zXX+jvR(D30M5AZ<`7@c{mWGK5tH|bASDGvuBV%Dfq3Q2;ymFKUopx2rw3+b;3C+KM
zJi;r<vjKt1y2kD=cJ|p0UqMDrOEbGb1}&(QE;!Qa2$bRb943-s=-c%()>CFwS#dwa
z%TB)sf;3)-e$f3bCi$Eq=_BQY%)Mbi<p7!p`i6R`CjI?m>#0#Hzb#IsQLI11!psaV
zr={Z95+$ME%E!<mmDq<Y%eU_CJVvEEA!_En4*(_pd9Kc?yRQ%1$>{`7#HU%0ze1$l
zJ5Os*8(DT3<D1i&Wk(Ys#LxHfTW%ZqK~1N}(<qnDpsYnoitdeNqbk1;LznyB7PDfi
zknTdo=rS+JJ$XTnqKW!C>c;DC*KRoBJ4w1lP>Ftc<v9P~VY`do5p4Hk&NR@WHGcqY
za7IzF^O+Mc3y^1~{y*~GGN`Vt+ZGK3g1bv_Te!PB1W#}a794^*1b26L4;owp0fG}8
zg1fu>n<P8?+vhv?ym!v4Ten`-tK!EDHOZQ5%s%=UqxIHW_j6X81#GU)5{QQANK}Iu
zrvMnLXUo0SYC&iTrVns5LbXMhRS&U^oW;SRy;CelAYg??@Wo?pM+EYRIypYvs%qoj
z=#(5je4#CGUNh0=_MAJqUshAd$TeUjNE<`dn$HleZ(!_QA;D^ZgIE)ACP7HVObZ4;
zPG7qhUe%`F`kZS81u=eM)HA;^7{w9glbUdLaY(t&>AQ=cxvQ@bTRH6ul$7fqBxeBm
zt~m^+a?R4w$6l-};?v`M_)VZg21|}aTpRZnUeT4w&hEIPUblI%1n2>jlL%CD>CWLk
zR5siM12WjzWRa_6tMKe_{x#6cAZENG-mwd~l?E?TTpUvd#hyj*$z3;Zzs+ubkmL*P
z;U*Bx#<=H&T`(&(3nsu$s~YWdPCsLy;Zo)Q3WHJFXz-l1uu62ysystap{;U>_9Wf!
z0o*dkD*>EkKnc3!pieqEdNgSdGfm|uo;6^xs0Mhaf+wgHShP+L`>(AGVaKu3d!NRq
zp}hMJ$0Q$Xd-Xk)M9G8>eCKv|(md^_8x%_9nmP<KGs*on&^Zv^rrkB+i%7jS-nyI+
zH}e|`LWnYn^3jZXCnbEfY!3nJdH!^kto32?!q=~Jk_Eg7%$w}{nDKeae^jjgwpaK|
zH@F*+{46B50c7EhS<fr`D?C+>sOw@PpK?<b*rZ1`gAfgISV-SDJe%ndal7)jG~ZR|
z-tN7zjPKgKIt<YD4XCVCgrM&h*MOn))b0acQ4$A3MfIsYdv{+&YXp##RtIm{!m<?E
zfD*=yOT<`m#Y5|x1E9}*WE=lzn_Y-J9N@#RYId;QTP?SZUR6Ra;MHT+?}W`4(V8DF
znaz%Y3z6$+>`&oelHc`x<p}4k5g7hVH3l5Abn?8CcEsYM$>Y!qan_QQiC}YJ>a)rN
z?xBhm<{a3m)Q`whhQW%8GaM2(=YWN+$x9ILI`y#Sy$EJD;H5_d0YxL15mTmB3mb??
zFvbpvwwwXbbhWh6DSNxcb9MkW2vqzPiDru-j2ylye}^2<k)6^0_&tQOa+E_#82|X@
zU~(uNR@@s_Y#&KIevoFuWQL>T@RN%S#495oGKwOoV*v6uLGZy0qJS5!2xN|tpgLkR
z`4Qz_%N3MGQXSe054x+<_3EPRDs9cLkbeiIRc8U(-`}4ZSpR<1{*)1KH{RW4wEFJu
zHJ>Xy@HDaba)j-$k6CQ;8O;9CSS8|3Yyv|W2(XT1<P@uSE2L2|F-|GEwR}4t22<F4
zqgj-I`d{2lPTw|v{#|@fkM*4wO(T->9WGEpve{7(`?C0;R!|m)qLO`gRb<q&Sx2V`
ze$iy_E1%vJ^Kg{Og>W{OkK<ASc!fm-Jf;<HnWc7!khcq<6x-fj)mGP)7AJ#%y<s<V
zN1gSOH-K>T?W!WT_v5OGlcJVWfSFex2inY33pduN$N$VQnG?(AL&nGm5)dVi^jo<A
z!VLdJ>SM7{VM}k<l9!gy)nmv=iLY0{>aEz223XwW$;PsH*r(`fzzJpLQh6amgn4;l
zCT}gVR)JXI;fs<qs}^8BH6%fKaAkoAVigXmhHtto&U~X&Tpk!yR94E6r9Ev%Woa-B
zUCM{I5xf4gT>hT_=8MM_?s0M4lks_-E(6Jaxve-_xCnWLh6lP|i396Xx(LL;GtYm8
zq7Gm&WjowJHRe1Rv)K;J2Tx0rp~lna(o$6eeGL>a(ToNZRH8~=-|b=2XL^1f-3VyV
z?fc#;!5$dix~ec3K7cawl(`r?-+FoD)bWw7ln8{;bqc0#*f4+MU`c5RV{Ftay42<&
zv{VMTy8&37H&AFo1Bv;!Jf4LBOz@SgqSyi1*FaIiWC;U6F-<V3wb$M@Bl3cmejOfq
znNLf1a@1ZC{kRH=N=iCwde)1V?F>5E_&zp;NZ{@L@sci*x!kI<lKxw@s!l;6DH8>(
ze0d<z{wid6hXQa46520xvL-EZ&}>je^h?fXs~&rc;&4Q7MaT~UtF&co-;`-j+EOz8
zDLfR82Sqv=`~)T1+5HAwGn9e_B3hV}JMqd_vpXb<dwU8og4X$on=&xH&FgZTn5R{~
zD~5giJLS(FdaI+oHPsaj(@e-(OQ@|=3^mi*^DZ(+)SJtbZ!&qUd*`$J4nQ}Q@Ew(6
zRBBcecsC^ZBn^$qK~!e#Aapa`aurJo9gDU79MJ$k+hzg}3;(Lt6A5vn9~3PH*h>9q
zK@U1NM}~jya?Zn@9oGnHs{`P6?!2p+ex$p@bSjsTm+d7eR#sr~LL_}gY3X`q+0bSm
z3Mfbqr8qn3wE=+Uf*C*MvfQ_o?-c?8gtfJ`R+Br<n?zbYO7D%N#$+yT$K9aYi@nap
z#pv{zc7WtwXSLAp(|Wc!Wb_RgMwsjqMebnmvUvFNmMU0Yv7X^O)IJ6!HI-NCr-f}J
znWaz?@M3vCnj~ogM27(UBit5HLeg7?r%gYR5c7jY5(W~Jt#hMAQj{TrfWuuj6CRtH
zAHbJk)K&PAPRqrGBN0K+F-50|<3b+P8=C<aFFG(HS+VjBNdRbjLk>uis>9Ga00C0x
z9ppvut}7rw>f!Kc*TZ4saXfrs#Jgj59%eFLfZ&72{z4l~kJN;GT>@zDRDpmy@jZ$d
z7WLy_oY)@nM24f#;<p8rN8q;7)(Vao@ie81xv%hFr>p(YX0#NAy|D(|MrlRN7k)a-
zv`T1LsI{m8X`GcF<@_-`wWHP6NNsjDJ1?4v`K{71?>_N>THIT+Xl4Z(R+C<66zbH3
zbKX}-BWXRkZ?F{h1Qc)YgU~esvjw(>-Xsc5UM(6Ycbz{Ji%NAbVK*Uz?4J<vuV1@a
zAz?Er#1zH-l!QDfEx(E?xLkaU$g1Wxw3mK5_vY2ca~)coh1K<{pkjJslv?B-TY}>Q
zjNn55&m1`xU1#Un&cV*O`04CH6xcArSr;_ZV)gAa13QIRp=nvI$&F0A>=iFEL*zb3
za(=xl0NP5{6_3~OQcs@?;%M3_d;oj4iK=LJo?#sO`fV=a=IgwY%`pibd2D=nlM9z{
z3?t)28*&fG(jcH|2wFs<;7oXr0(@h3+z@}!1&t|M%4DvVx@6*1)XbJ5L6%3W$Y`SF
z0bijp0BXMVK+q4(15x3oE=b}8y7gSN>~kVD!N&b-#o7$4d0xx+R_d?}Uu>h0Q$yT1
za#3vAwV3agfj*PlDMk2%FfBZ8GnIl~ooDa-blf{1xtiyeLxno_#{hY~@EOVt=6n+g
zQ0mk(%W#sE0*(M4LxZvk^)4>0hpJVMKN8}%P#%e%+C7x7prWbr5C8#h*f+0*v0~b!
z>?1y0>qnoP{Yuo=RyiVa^S>$sf{M?MJUODe?Ts$tAu{*Rt^`H;i!$+&FwO2&NxCSy
z{6OLKxSVMD7loy;av+&ekK}M$@UVpj`*?2|8XFf(Xgo-=2LRBqU7h(e2%w?Kn~Ek@
zR=YzTaD(rZDI1&(eyQNI0zPF)#HFP98;v7kP#3Gt9791sykG4C`mt_B5q~=!5~Sa+
zJwEw$!h7-Tw6ofb^tTnrJ0|O%;5hw5!8Z4ZuRLLb%~wBHOEYw<2IGJh@ga$1&Ve37
z3<pERg0obim~{Fn{J=RtP%>!KVw9tvn*VKg2!=X{uU{O8*^o*<gKr!L{ZKW1c{O}V
z2!aYqCcCI3oHUCEJdWvNXd9CXICat!c=@5ZZ5V&p<JVld!WOUs^6|S5O;mZ2oxkmm
z(Al09S!PJ+?w&yHuIt$y#+!O`h%;Jg@O5wE2ntmXv_f(*fR}u#34K=<7YqZ42_b<-
za+urspBSE9o4ztayZ?B6>3-9V<$QgNA|6R#nIv>W<x0%xIIjkwT-@1TrZ`*cgMDQ#
zq_F1k>s7t@(IR3r0Z;#3I=$?1hg0&bpdlq{G(9;vnAkUvgqSJ2K?hcs<N0kN{u;i$
zXS1Q_vZ5il@?feBNq!Pf6hkJdLKq5Bdf-%<#!6RTEHnawn7s^qaqzc+IXG>BLmyaA
z3^geqo3)-5z&5@6TKZ;;3)~tDJdfsdCGa4*truvO5EU19?nrn1WWn~~!v~Y4$Cr+i
zZkPK;8j6rihtob;J$KLUOHYx|*m#M4wc%Vc(l{L)Um+1RmEWguU0I77OyAOU<Hq2!
znSY)Bp15<S-KN!2k*~(%!;k?$f(p;eW4DtCP_8OC`YRa2fZxn}yWik@>xaES0WCCp
zm7Ks9bq!T-(0l~?F6g_*yYKwcYXDze9A%j;W07j!ojF3#M)YfI+Uq8-IYHAM0C1>s
zT$IYAFvk4;X4QQ27wy`b4IEfvvP5CxXcc`>R?j14E{wtOS3I(ZG!cGkR&xU5XAkke
z7(YNFDiqrjzM?uxNMLnH^|pW&D54Yb6NA-TEp`HWlH>E8aea6Hzn|)lLy$EdIKb*A
zWrK@6`o6mU=q4%E?>v{#Km83bY4`}S2m=8dD~TeB23SH2tE>PVIj#SR6Vt@A8^*~3
z`g4H~XlM05v6*TSaRWx3IsKPO(aDFu#`4co{}=xRe7sDMo?p}d)GrG00NC#I+EY9a
z?0>Qx`-gMaKR+MnT$j!F|Hjjbrv~adi6lSkCjXP$+&@3z7d_;+7ad^45qJsU(*I-M
zt)m|V0JTcS_+LIxUs0&^A;M&20KX*duF@Y~w?O;4T&E))XZ%ZSfe2j4ldY)7T&!Ve
z@)zg#Z7}$0Wha1IYhTN)_#*}Na#0V3xO05^+pmkaPQSSbrM<6yHk{Lv5K?XnP!W2Q
z9utCVyH5Jox0C9$x`sbjzZ{15!1W9)lu#?G;il>Z(c1FUf7V+^GPGUeKewy*gF#yw
zQ$98NH={Rn4hYX=rdK^ZJpjO}{{$xh*FOGK@+6%=3v(+`0Vgp2L@piI`;hLb@*pt-
zT>&Ei=RCTnB!I=iyR`259Hf^23^*~rPX=6s@65)~XOq=A>vtkaH2n1nYOAZe07Aad
zlYBZmLiV(X5z_o`E~%iX8lb4<`F;yI31F~<P)y95TZETX;mV{lcz~r{^)x#jUnq2P
z#{;zk@95=P$>*g^Kr4LGs&;v%YK;>utcysaY$rA(&fnnq2qVnEA03{bjrXelMB)Jx
zV}g<8oY%eONgkr<w=UQ5#7^SK$S;Kc1At4Ug!gxp`j7$@jRtycgt4jlYIN%9<4^Kp
zl$;G3-vA-Y>}&<f6)ohI2(U^Y5|&M)tBs=_@GJ5rdg3;wK3MO)4_+iqt;*z}<qIuj
z&a}ywkzw>&ai5u+H2|C~fS$WyLgleYoxf)H?-c%=ou9&i*b1q50><rEuhhTeG6dQi
z>ATFQ{4%=V^sp(cQEGS=%d5kM(W<pT=ksk8z^J8cJgW&61Z*ZwY$}>8B7U8kCJU=K
z&3_T3s3pDQzeJFL_|qTm>rg=XmUPEMMVl6uP>=9*8YKC(guC_U2IjL(yS$H$gp_dR
zi$%vMSzpwiFdtd{xxW=Y?0)<9J!6Q;_*`dwTQU!NBw}J+hBn=`L1DUtn=`s~=qFTG
zTya&8OR&Oh%{?!((ihe`CSYF@`)4f+3U&zqROzc1dpQFpZn=mnvHu=ve|!u)e%Z=A
zv#|ac+X<o&AD1J8Ha9obs@%C`hGeQfu8^d0J4jQCI*=<_?9=_&f`T0VXuG7X#d)bP
z<!;%mA{MWa@!9>yJk#CrZ7m7>P%!$?vARZb=kRQZ*nTc)Oi0aO5-sJKcsDE<kHj0b
z1a<=m>37aK!{!XqB>BLBPob&=FQ}`L5Z8CRm#hVV`sB*<QQnP&K3f&Rh4<*T>yA2G
z22dz%yR-lxbR4xZ>~C*qYyP(q#uBH9|CSno^#sTrr3<k~tAFoY{v;jX<vRIGnvOTI
zxn3OlfBu{Zs56k5Sy&u|->CwPymS##7ZJ(BrAgMcI1vvTh|Sx|J@z_ETVA5Bja8am
zR?YJR8SIyzkMkZf74Y;%-vzsR@|Jt@q1i04`l<H-0{0-TjG9ZQSS%t5i?#;Tzz<Pa
zF4OHF#6`O8{Esk5-+KY$Gw${i$bWw$3;>$0)N)U*LkFb1=W_SIzmYG-BR=)@9K&f`
zp+FC(y|99EQCL~QTd+J*WXCP8x(7$askD?^U@5?KZ$Xh*E)&V33PvAmyywXLX?EmP
z8t@610RBtuncS+&yJ|8|f1$i*1`w%u*!#2aTN_pae?Hee_wQKGPDOuZ|4;nTEpTEw
z7mc2asyN=AOGv<8tIh5C$)3B7y9mbHIC^{KQUD4X@BHxB(ztjlfU`l2LFU?Qt^`2E
z4#8p?y+Qn7XaT?Eil9&@jojbR(AhNsua09HgoO0{8{8JM!<OS?C`b{!t#<cop_Ru>
zLjySZqmkVrp9`vdxvt1D7N26B8Jnn|&i7)6c2KyHnQ3q&g{(n^_ODV=dQ@8JSr>@$
zOSSk6(8vzZO!ZtmtFzo60zO`ry$EGCQnNgmxP}m2X!LKUj3lL@ldZO%%_@O@Jf&$F
z2D+0I27a5VK?hH%6P+zlPJY!@nm8MT;1;xZce&dA1-oqcxdsSm^gCOX$m1OC!5<!(
zSzb!X=OWvXLg*zjtV|9h;(i~UqtsAU3&RBrI!_&TJL;`G;v5}2+bf>L^zG3P(>Vir
zH91DFOBaue@YR5+s8SfYGR=t1gmk#JeM(;GXfL{fS+=ayf%G9^s#iu<ysIp=!R0++
z^%zD9#!WK7*Xf`93>jFxnDvnPz#O{3ATM0Ly5fIkfn>&<so|hdq3HxT0_g9}5UR6W
zed^!GdB3&`8>{p|Qr^@1no|BF(8PmGOag=;mW(Xdwy~na^df@4t<v_RAc%#fj$<4Z
z6~*HT@i3FiKMZRO>`VE$@wbL|Aj`z0yT@6C_O;B?MDn?D4fnkXUsb@Wz{bHhI2i{h
zhejczbhzW#tTj!!K#)i{J0_`XX6xsEIEIGz%P)Hkc!osd5(f4H#ofCZ#lDjuL1Ger
z3!vW@r;{8pW<@B5pXAMUDfe;ly5G#U;pb#GW~U}*N6boUgC0-$Fj+BIvqd}o?HWf$
zcBW2%#ci`+q1osG^l7hvMJ!sFfucV&dawV%2McBgIMYF>%JF|F9LOJ>$s=>8Y0T0R
zlW1_A$pZL`Vc~B`NDs(Y4c?wCtH2A72t=c2IG$8?0FDn1-tYqjuf_Y}7MQ)(<pe1Y
zN;!^I%pbRQ;hfzWR1O)5h!0abHWY@JVBU51<;9`seAr;*&Fw@xUfyK!69^mPz@~F!
zrb7s}FI8EB)7f~f__HUhjGXR*JSUTE^|H3m*_no7$-!Cmqo~3{^~b9^gXs%ExKG9_
z-~F(s*x<;orKG^Sa{Px2Fh|-DF0fl)^a&7T+T3Z7rFbB}(3GI*5_r=Qoz8)BL56~$
zahq+%$B>kQv(}gETo5dhp98_th0U6<JpOr574XGtC)22j02*dhUL}lhhOCQjY*jIe
zU;A=esJmpKBwwT?8Ves`suW1R>FfG7H}Y}wE1%&9Nn_m~Go23wP0(EWXPq7lPP+3A
z*xnCbn|yHHB!^p({9VNlPG_%o_l{D(F3#fGFoCWuBhIgAWM4w(w{LrlS)<%3&lduc
z0f5T#S^Y~u{MFB2jCb7@wubfF)cSw(R;79_7AkQzv;`|ZZ&Zw)sc-$w;&c;t>u=*&
zna5^00Tl=!cFH;N&{XAM1zL@f!m-Y-4J3Z_91y?1yRFrTMJ0ceGn&XJ)S(B0<Eum}
ze~e$=KF}@#Y^RM%owlK$mV2@?^TT{<C{-#%wXOxstl>mqTdUkI;aPW2*!nqnN^5h!
z?sxc~oskz(73n^WkEp*_i&f0J`|86LW4*VBcED43U6b`729ao-MZ#HNrs`<n>vX_f
zd38J?nYZ1{0YO3zqQFgjoDVB#ZTAmOF8-+z^gh>(j#88Rg@mC-E<?MIor}L(LmauN
z`091Z1~8qY-F1nSwufW|+#^d47A5ivV9OAc;bxoh>LaN`u~bbLFwYqT+~VW?0ow|e
z80kp>!$QWCb-VEQO?0i)Okz3xKqXh9o61l{h796BBJ2`SrOyR~-<4DbxpjwWsuh(t
z-&bN17aHw=jyO}{t<&vZUR~YYd+xpfU(<9do7oI9Q3Md}W6E$w@fd$+FwYjwg1dW|
zU%BJ-IOe}g_E}nzH5u~cZ&`a)o(?y1KuV#M!5b!Wh6olj(t!u)gS|pog(k~%J)9pY
zQ_5gI<@wp)H|~Pd&{5PmMN8V)gAl0mKj_5^*l>Eyd$tU^=AUIm&w29?<JOP}=eG2W
zu;cEno=9+3T}Aw?oM<=tP95eqY6w$hI#a!xhFe7*MI?%^u93=&D$huLCQ6=d#&M#E
z%l_Ngg_IQGpT;gHHb+e{!h2+uZJr&VmVT9|VG(nXG>QM_(PBF>fWrW_reF*fM02tq
z|Jmo|$0McfL~jEbN#|;nY4Qi3Mj&W4&H$Iz<)fp(-rhoNc8jEW&47Q8*uRX1SDQh9
zZ6f*zsm{iXe&|i8VJ9aM(Xe6EM-qz7$9`<-myrOyp_yrF`c?|0WFif;XniB=NJkNg
z%1;82UPqaH{xg+~h6VZAHNG$Vr_EnL9obqC*GCX^UbMPR0Gs1WK8U(^5Lf7h%aKHM
zOkh)p_Gazm{hZAtTgvP_-8xA)AT}Nkh^K&X^u3_4uYkNb3Lg&+aGc|Gju+|fk)6S%
zA~iZnQ#YeL>;@*o7t;qPIeanqYXQdE>HzBlfbEh6{Ym1Fr%P{{5;^Nl34?STt@m_=
z?eX?1m*_&PZ&)0<<LOeHh5EaqCFEIvQGYfrQ_BRuf|gvbR%L=Yap?h3XVE<+owuC6
z_B|>E8OhuY>YddrtRI%qM+MCI0t<4Fs{p~gmciby17Y?FD0W+2OnE)so&6msfU^f+
z@R3o*WH_Q5lZ+(lSVzDK;Yg72c_DU>IH<wp<xcPkBZI-qE9v+Z?Jt;caXtKq-&O!`
zBNiw?^RjWEsaR@<mC%jid4L;}31S}UZ)oZ$HFLO>Q6N`>mTzLn<a7U2<mRN-(?4(y
zpQNsy6BbKzkH_ZYA5J&de7E`VnJu*}E59}6;wAv)vtith9pIv5NY?(|vuvUU$6^6v
z!|*Vj3zB4jHJR}V7+k%h4FpXN4qiAnzwKtH6J|P$xNPCG85dXmv&ax;rV@6?L2N*|
zDm)g7wLMp8?R^-^E$a?Q?}BEE4Dsj6dY8{moP^6L_^+?nDjg<WYHdoVuY?EWV0{$u
z!;Ux#$Ui>~NV>QmaXo64QQeUvhK!lIaw{_=A<vNe<c5`T0f)-f+wPCzV-3!6Lamz^
zw!X`ISQ+9nqA@>RsW|tx&B6ogGfA0g5n*ghdp(n1?u~u;cvdYI9<XQu4_i|4c++6`
z%9oL9hsK>nTIg6YKzO^G60);4D?Nx88QP)$D^X04Ok!AEhDa(S?{;%<{CFga;}zG~
z+b?R%3&W$ZZSiR<c*|k_Xo%?400&;1e8O^bU@9{@#h=FbNoYbKIBO7|goEmisz^?K
zz6Ia=8_#6l$S;?X=8(?<2a#~!34DUUNj3pU)=w8h^^zdolH_#xVy`8P+#4eM%33?v
z(3O9KhR3%DJ*=|PgDochQ_lYDYNPQ;B65)LgT-IcD|=mq9vJPt{8&B75P9G5R`Ol;
z3rHisXcUmxdR*M-BI>aDgryEaN7;iq3l|QkWDO&)Mm3&0!{p9JnH?PkU({X_#|U}S
zZ2#Nk@tCPZ6h%LJ|IuR3eyP>cM4*u=OF`)AUCy}KdT~R0edfzuR^|O*IvVs5?$WUW
z=L$`S%d^pTtEw<;?K0z$4CkX($IqI88)$S)&K-8saNpc-)xa+}^%=WpN&TfCxQ+k4
zGmg9W&N)?f;NR`UeXZVxJwm`6wJ&ItUo@B)SNqGFgi6r;V~19sV=a$Y7%YO1`g@1z
z8mV5%KAIZ?Gd=H!HOdaa%?T)@ayGn2++MGL-}9j$tSEb#N?i{HB*))oGV4W@y7oBB
z($KOdU)#AP8dmd~In4-lp2q<daHUZ75e1rdzJL7D=@J%&5(Z95cZ=}eDB8r@DwQ!H
zUI7gcg$*zZL3&F=1MD1EwnMtZw3OIL+Jj`Wl?+M>D&!3pb#PN5*(`iv*>JacWj?s#
zUVhFfvunTSPKu`=I?iJ@7PWe<!76X+70=8e(X2;m_oWyhiiqqZ06rSE1wG#lp6+$y
z+~pr1->;keco5v17dSGm9e}l%d?jRw4o2KxofYF$qEct(0O2+2+v;vRRFM@1uM9Mo
zdbdbJybF?hs0&&6rcY`+@!QblZe;d5mZ_9fSXc~mll&qcf6T%-u=`%Urx-FiFY*GE
zU?l88lEk46C?!-^l++wzk5)Ge?-Ptf^Ck~73rNXTn($*81@wfjFpS2KrSi?*1%yN6
zfizoPF+P`-*&O_28%)ZUv%<B18~Q5ug)`v6Gegt<XzLUIrn-h%?CGRT?R^ILCE%p5
z#;O=yCJeSO$7l8>GjW8|XT6IzGgQ7;paI8Mdyj!8U5V1$wZELG*`1j-aUVq&&!_{L
zD;L!41HO@$ELit!)p3>A7m^~0QZeug4=r^r=p$F!Pyc*;aa;gYM`$E3TrKzMk_fxj
zeTlC^&DHVT*|V~J-iOQeY{p|@C+C<=v2W6*20<fKMJrs_8)z_thgAm$Q7K};TaPj!
zvpv%I7D&6okU@KF(qU&xU|$f<#X!!ED^*w8?-JKdQ`zYq$2l@GQ+vNQr>{a)H~b~y
z2T+4GsTI9{Hw`JQX^NH39A%4X1(En?AiV6U*Y{MWdmyJx3d-;_&|nfK-Cwk`xQmRM
zbQF5`Dohf+zbnVMV-sVGJXqCo4SV&0LKYX``hZy!6^|DXpEuna=MZnM82lTo*mC!K
z;yl9-z>yM|IH-x+H+J(>wSpz=Nm6h8f*`AZyKG#bjh_=h2J*<K^nh~#CL2b{1U~T<
znuHEb?wy`|wU-=HvM+5?J4uvM+!yc;!MtfXTL4~Ic>X6d&zWFTDTUcQz%c+ML`I>+
zYQ(Mps(rN}VEjO3u-q7O4KnIdxO_RAf8e#+F+2WqKI^;<Fg6_bLk(Y|Aj#K{Ehrjo
zBh}ofiXTAnkZ-jJ$F>8t<LFxh25KajLGKl6pM4GMt@y+_tqAypi{S~3a_w;il6SJB
zbf9+vOsXB1A|c`(-2e_xp_gvyl#EPX6v4ou`Z(+EBf32U>+Wo5yi)GnPQ}jK)7J$=
zqSD7(d<u0M4&suBDf~D)!NF#B$Q^I;NM?t=1tK@;^5I1+qHP)<AijPPk2=vkI5boR
zXoaUXu20s~stt?j7cR`+;F2wcnRWCagWt!k{Ae_`<ZXHg+rd@H@&ue(XyNk2zuB6c
z=ouxzFgp$Ekmra3&JE)LSXFgdGPM=BY0*L%xmxMccz&ic2Z@j}@v2~xqPId6J6v%-
z7#)yPVaia<H6Ahtr2)1=$Dt{-27dRqb7`?*765Z1h1iYVzX*kn6%%sf*;0a4HRCZh
z2;t2s|5~?EqX{AG<!@7o@3Da=Gc6<)>u;9`Q@?~oi11jEtf%Gp6%hAi*}%)@?a#Tt
z=BQYIoq${(&df)(MoPJ*JBas1l}1VhT@<!V8QG_BFpoSIsl*g3IQVC~XS<JE_Ko=l
z%+<lx5}RSNoM?lkwgVVb^zN*A)4q_$Po)APBlwsyjapnNB!UTJ&c(!Er02hbarp@w
zfWpyrx1PHL1J#Zg0fp61B?Ze?o}|!}>pNXnm%#?77=2bx->@kvWewq?>3t@TE^+!N
zONQjatzTb}N@@yC4i3bL7uv+m4gkQ4LR8bhMDjlCZR=agk4Dpo;fV*ztrWPPmym8=
zTXDgXUeQ=Y$Y>yAPu;-ds8X}FA22uUB5lSMnL6#_2kUOJI!lZ;E~wFkVHmO+I4PJW
z(3!n0t*q={9Bf5!+#D>iE!Wyy<HLKQc8!c`E(VSsdidNrNGoB##rjR=v$_`5s0J#+
zDgm`arin%1V|XaSRr~u<r*w;@>;ST@_d}EAs+x+GJfn@uU@=>%x)g(~%GR2^LYt09
z-fzOsWJ>VSU2~z_C($Vcz0dGtUbSW=?Mq;y`ifHgvKy6tTyVm9_V^spJ4s~x;azQw
zyR$s3>}6YPAQrV*A!-pv!&&{l0rbtn)+&!N&U7hq4i1jNRF1^(R<|pH&zkka$RGQ0
zeU}Ni5BMBoGS6|#b)9_z${t3uj3CqaKPJXU0@7e&=%4Ek(pjkbCnIl9_D1?=EdXy)
z4XjcxZVIk;B3rsi-Ep68^6H`?<0Z&1N_wj?21*{GJkcaN0!E7puZC0Eo1k9FT73&b
zRPGc$)bM-H<B^$zYKU&1SUs#e`60g_+lA4QP-ly(MhF5hVc-$@ancCz(3+=o`(9Un
zUl#vp%kT~2hZOXJg24|?0C@!LA-b5&-jxBG7uE+sMaPe$>mo5Yf~)-hvS#RQ10pfR
zEn87Kkt7a2m-&NzS%r?&UTYAK!;W%dJe7NVy6+!vhSQUij=^M>8PunHJG?rJSrVtc
zGsk~JG4a93pO4J$9M7e%XFTxcrC9Bg05%gY*-shKS38bTm{2gq?LooAs#(e6m~!a@
z_E6^HtK*bjn(=e_Ls{9C414|n`&Q-hO{_yQd#tG%`m1Q+!mAzh_HuDBE!cHf3JN7e
z`ZP+Ar0&l?9KtaC?xwV*bEr;wU8R&dKxIaBfIHT}S>x<o{)Rb7-ag};?vM!CIapD?
z%6eFSj0PYtWe#2k$3H2NE5X-3pN<7OY;B5L)5j@5M(PTf=td$t`tNY90BmZa3CNaB
zSoshuwL=CLX0S`#7t1=0%obW&dS-fkHF0?s--5Euj83}_?8%jaj>){D`Pyt@w%8xD
zZ5lhz$k)V}9c6GBX62FumiS_jh!!{c^)6|}xGIJ;>)y|0rd+7itY}1C6}zLyM{_%^
zgUJy#1V*-%zXV7J`*H29xftUL#NH;TAy!Z72=6}c20+zk=xu3i7)J``-P<u>YO3l+
zX@&d{;9`0`V7zXG6@-W8tx%?;9(UpJVuvh_nN0{LxD4fl6D&0$W74^=J6)0(d}rJf
zLC~n9SD!@plcaBv7wR!Bb)y2CB7}nFYuZsaLzc_ZN&))?gG?m&fYL&Qm{{gY_gj?k
zg>G)z(hr0KlivW}$L%1p`!+^~Z`FV-e1&j5gojwZ%1uNkxs&15t8d-ouZ)ovz9)->
zD7pSwsMWRY#E;^;MAEu*HH76$L=;g#H4*@m@~~FIDMRnh+GPZS`bI3jJ;<;5w>bgD
zk-vEk>ieY#<?U!~{s-*uk4^b|jx5e@0D~83LJ<|-e0Q{<R`1!`v9<gX=tK0U{nPWN
z_TG!=x%j_l43w09i()sl+Q4_6_J+H%yT!qlt|<ZL=4M=;&Wl=0w)e>R12-m_q7x*y
zi`w{_jkH(sNS@dTWABMQ;ymnF_YCaudA!T2fP;79si(<eMZ4UzP`Gj*>N7XW4R)|3
zg*iG$XLTnw8T|)G2S6N8(<!9F$_vHmq-+|zs;fWX)uTp(qkJG70~<3WluhfmHwExO
zkpp1IN&FhIKP;LV=oxGN(=d;mVx1k%?2Z6jvu+B;N;WS%Ns^cjbIfqG+hSpUw8lwS
z{JO6yoHRf>WD(8-sAfoB6`VPX1*F8OL_5pY$km5{(UO`Dn3z_-OM2QlNaZ{+C+MTu
z`dJiSSV-x4BNmPE1?Z1%G2ftQrRrO=H<Qaw{m$`&@S+<>;QO!FQ;hYjdameXo(l$&
zq!eVbK4S||tbWajmh@F5I-dmdB%^7Cn|VpdeN8_XX>6nrm@G6(ms5FoVqKrxfU~!J
z1G+@;#OvHGkK=y4Qh9WLxMD+o;)SfQ;4>255WWl$V)rva9+SAaZ+?FX8{DhQQ%V(|
zABut^;wNKa4?8u;mtAF^UeIQ8nn40*@ZdAvm<-)W)*DMk!CBsv2<^*<cLu;73}u=!
z_t^UIDbrDa5NeYayImI*$p}zW2RRRpgaAB6HW(a>8mL4WzOb;GM2lk4hOq~^U+R8X
zRffzWSJ@iCQ`)*$XR5$EZ-8K;uq-f;2ViA>3z%s%pk!}Z+4{n7zUA?oJo0xbcaa`Q
zWT%cdB8ihgb%H_zD4{^1r_)n<AX9@jmbDR8951C`v&BE5_*{Z)s;=EOSzTg_Yx;+<
zgCaWHbVD*ExtjG}KQ@-=#Jm~qZk7F|h_@@dsjad*`4pa8I{a<juwC?eVUPQ2pu$%$
zuM6Pm=ME&8SWwptL`setTVn%;z=N>;l~hGHQzgDl3r$}Rn~g^VPwo7(zToE?i`_*V
z2yR>!?a!yLNuU7OeqWW1w9$4+of~1iCY2n8H9nV2bzTbN5+1Qqb#<lD;L{8b5*1fS
zspZlT*Vg1AwS*rsiIicq3@SDwHdM?@Y>QZ#S1`U`$-M3%G%36wyqMoJf*Xj6Lx8;^
z7KBERt{I~!W8Z3daxtH1=40VwSs3SQ{!&^$JSt>TQd*i>vRhtKf7rS>#Z~0%=%veK
z1a8yUKN!<~aHx6QjwEm|lDsSoUG9fmZ+K&;+b70>PdXVddphxSC)mi*Y;kfT-P>8f
z;Q2->l<?!rf{H?IV)FLhT2rJwG&KhI+hGa}=eEWw?MAn;7>Z5jp0Q&?_Axr8nyrrF
zr>a<6!__b?5N;+p*G;h%CihI1=jDq0yLsHgw<|O77$yX!CjN0q@sNlyXe2>sMp^Xe
zN-b{KtE2a~KW}!mPll^&t6l473L4EyD@-u}kWA{fLGa1t^39D^?Uf!X%LZq%@$LDJ
z?EszEJ286grgsT+ZnJ%>>X<N5FaG>|B@q22+}~?y0Sls-ix3WxQXrT`W&sx(MfpNZ
z!7ul4+XKKv2bTv{T(#F|{r*2X_wJjHPL$EQ;DqLpHOpI3(Kf;~pYp7{osp0|!^V>9
z7xFW{4|u#?!RA`&FV&uYa!dZ~wZ!N6Nj*dk8kAyQkz%R2G1NY9eL3~=Cd~wGfa9Wp
zD9vXG3EII^-;}64P<~nm@dBY)xWDFGL=;!Ewi~2@SdKMGt5}FKvC*oK-9{G6h)Q?s
zQ&+CiV|vwI!b`cQXc_()2h>GVVsff!9og{Hu`&)kN6ThmSgjPghjwE!YHAI@irAp>
zh<|TwYB*_bD+`tDFid?+VUFUcy-p5UOY0~E3T&jR1IFCvj9x&?GD&~U^ZQk;Redc@
zo=FvJ6Sv#Du=oUuqZyQj6^6%;euf+Q&XD9+hKw41=p$Pp@<Gv3ET=|`Rk7rJETe1*
zI+~Q3W!H;x9E>gOpS9fly~X-R1=2uvhy;bJE!Wi)vrm;U%w8Sv-Drn<yk$~p$OgQ4
z=A3j&3W;C5SmZ+}4TCp5>4=%c>D8S-i;9aib(g?Q2od9oi#6U~UtC^FJ!N^~ZjVJh
zczy-D8<<pW-gLTJdiwNPtC898O230l5d80-+|(ETanW;a=t?qmul><uv)l~_QhC)Q
z+P3R=fj&-WA?{Zh&GPrHkvuxbP&>6xjheXB3oqkS8t;e&1XO#5GuB|)y~b<%AtQ}p
z(JlD7*;R&bV+T?;_BJZ|>yKIvlwYlWy*H_!Sq0@0vG+tG1Gi{6x_=Spci)9!QNHS5
z^NY%H5;LTqarX;?KUCoyr4Zt4In4Qps)X3Stg2KqE%Vf|$?|1A4lm4ih-Z>x9~RVz
zM%l-N^(3dLQ@gKQeUgjhV@1Iu-2YgZ33`jjFWU6;=eoZb$MxrLRqf4Mwmso*+XdVX
z2=99|xi5OeV&bAfnyOyJnY3(M?K)`q>G14?pT%G^PDZ@iB&4OK4S|Lqwg<I6<#Rq(
zl`pW&3cu=&R_=Z)v;A=&Y-QzuD4>^u&*~N9e!eNrb>09fscB!htgB&lXo$B^l^oTt
zwb0NcAIQ)#Qrp{3l5Dg%fkDN=A>PQps6MY~<6v9C+{@XERM}`b>BLn(OKy%Lq|OlE
z*wr>{Jzw9dzTr7j(Q4aGj>4yGZ`qaXD7Z*Km^@uw%j9~LwCdOd*v?x7j}xbSa<_eL
z%mzzL{Uv?Yl{aTdQNb*yaYjQPaJpBp?7BFwghM{GLc$fhRXjx!hvv#B199ch+8iJ)
z*^wj`hLd-P)n>)zISgZbK+1B(pkjb)FlUz1s9TkGyyh`sRLPIS!zB^HXIbzSa^ag(
z8|e_jfl4TR>OCALZIs>cO=AfXL_FZQ?H1{WfUe>PA^GXg4-pkOx{^;PCMK4mB-1QB
z5K%s-(R9qz(w^?$Gs#Jqg{NAd%H3~{ROf9;Bzb+fNG%_aC{JevGQYTg0|gS$-#s1M
zJ1hjsKUC~r_srFf=lO2-c^f~fZ7_vI`*;cLT{c5l1-e6K5;Pxq+V+kND;hb&g0Aat
z+S?>y3*-{pGIVaPwCkI@3mNYefK2ihiAU>_XR>?k%&V0sm7m}oWL-cYR}-9k7Z%4m
z9+AdiC&#a>eAOyJn9z@~*HVJ0_X1pq>MlT=7}96hG}H<M+DDehso(r+@jx~0c$(kQ
z4_z(yE45fRzIQ3DFX{In7yB``X_^73Jg-IPbbF^J>z{jjPGghPZ!~l=_SNpf-LLbw
z>eT`>?Aqe?Su`yV+fEC+1J;xYmp12|7Go+B@vPH+)=GAK(Xt;EvppY`C}!unYB|dv
z5JX0tOp=ldPYku@7RS(iJ4rix12SVlr+0~>kQ+g3*zs<znD=?NvvU}KXQgpCY47MC
zr>pB0muF0VivfAbRVf;AH%hBqk-iardj3Fo+ZB_9U%J{#)Oy!nf7&gfDdNxgzR`66
z5n>^rI}&ZD5Ubf|R5hNhsDUI~Qh?XmE4f9r?I-%v($iJj#Gyr<eX7S+c)Wq!oTj!f
zHg8b(TJq{@u5#2#(hL=Y_aIEYmRy`S_mNi@M6y@Rz|Zwjo~8Hz?yDJcXQPMJlAD<1
z*t}Et_$cPRuc$X2hd_hdOjSushtPa>pKT_$Ma~_IqXZ}`&u?S3y<v{Nkg(tbUZ<Dh
zpyae%eGp*3{^s#!+lwwf@CxPDchb1h9t|$JqU97ena8Qz-rl~~WlMK?dMM9dKRNzS
z6awM+9*Inl*l3-d8^~vt62#pCRBCCghx4G<vOOo5$Q&C%y7$r;r?YS)eVg(EC)CGD
zZ#)?q(I`p~hU7D~OZ?z-{R9_TB6y{zyc#;`+Fm|B)TCNP_ij^Re0&^&I>dnJn^KZ-
zZ%O;uJqhk)Alc@JPNxIs<3g(QMq*(314=*V<Een8v||MIH%ZA+&83Nh8Tups+bC`a
z>p0{VZgNTqcIO6UJ7K_#C^Ye{l@^svuoQTrDRflCtuuv%Ql!AL;OuGg6-D3zI}=xn
zxG1|COS<dBQ2C>>4uv(pbEf7~NJL~TT1CqTB6@|Q+vj+#G=5lyNOsvrLU)2r%xphy
z(L;6$qDMnaqbYN6P^x<5sc@_&&D(c4XR027p)6%q<$Vl#_t}nta7s@?+rG*$^0mOn
zC8PQU5=(@(mlz{Y7uI?2cUj}NHs-7^@RD?BYChh=bDbt)H7&v*zke@QfadUZd1gtQ
zB3Z1vq-FZiz0h?Uhn&(c4(R9JhoP!f9-`|!E<9UlvlbPOWbvruIS2C+Y;kU`d4r*8
zXu>Z~?Pg(!UI@HT^~XnG6;E#Gm+X-pR<w%Exc*Xa&JctmaJP`%5B?G+b!KMpGX8vI
z11HM8clodlW%u+>GeQ-wYmR+C#*3f8*$(G$ASZ_;uc`C;E+r0^p}p-x+s`hYE9J(P
zCVR%^rr41{1e>mWn_7=p-mBiimtfr&SKb3vF;cVDC#ll)bGfaacH>p3uU~`09V}1J
zUlK}7Xx3{IKwCLOy!yihphMP*;`OJX6h)@&7hLF(>rePXpH{kivvkEi05ve(%onJ`
z+KD7Wp0^$@6>rDDqcsWXAiVIIOf9YB6kduUM*}4C#mv-v2Q?^3@=2#jW6H~Dam@MX
zW#|S};Zd&si!z-?<wcj-nTy<b*V&<F!#>o$;o%s7l_-4JG~k5K-+Z-l@%L&CT;yi{
zTCjWmTCkt85P8>VlRqvCDK*;Uau@6BD$*|;9H3^#(L#<F_QkmbO1-lPv#+b|FLhd^
zPk+@ZEUi+9@71jgUIlJ?mr{{;J}^nWDfP}iE^l+ARCX)}$E|e8!XmD2YPwNbhA;`&
zMy<``%F@)lC^jyQH`!&dDkdeB(qn<IgdPa|R--p>w97ubcr?~<ut`fh)`eW4;=HB<
zn!ccB##m70WeMM<I|j(`turcfK%6ZekHXF>Yf!gpr=w=3QO~aR^v%rPQ495$d|QAa
zhP(>+<Zuny{*LmNnA+oMNq#cG!ObpqHBDQI1PzH>?bg+8UEpedXR;Sa;(f&IB-J~2
zs<JyO>UWo^Kuhz|u1a8&BBW$zo2PpVNs^%0%stAHhOBhH`eOH;W}S#SEt`rmsEUY-
z<?ShjTujWY9T`27B222Lic2#bFgpZ)@ZRj=O{fL@xH@RI?k}fXD)mg^Do`sRbctJq
zMM0fnkwmUAldiwnr)l`G6f3XTZmro0yv^fsSFiJ{Mjz{Mw~z@08K#f^SZ_Mwigz<h
z*)5_FXawRe3tmYL_ebNLMJ5d9kX+0a$gA{kF+7fMJ)|Qmt*)(YL`x80-e28pWhP?j
zINL~m@3~RMC+30#?Ph3BsSbq7H`L!%lqWMFGe=b_KAr;dj>(boDDW%!k1zcp%#-Ao
z7V_8+b8~VG_t!1BmXBdj6OnV-SJu!B@}TImtlWpJ@%>pmZL(yN5=w8cB&;o>WS3N=
z%<Z<mvT46w5Wx%-V!(92kt+VIsZEidPCrU`VPf{NzA1&hSpoCjKH}YhNz~9@NxrRH
zdWoBbdl));{=j3Pd{i9kw9`1%hw#|fl6I#4%vFkQt{a%3IgJX@iuE{7Ab*p8L8L-~
zR)Lm2O5S|UCx!UnZJtYDdDq928fWyzUFq{6ca~bgF!j6k?AA=r<y^3OookNYlS~`=
zY>k71fKIBfe8$@3`;d7l88f#qh3rVII#fh_$>}Q)m7HojQhbz;o9P<T+UmFIslJG^
zT5v2w$uim3V~4~#J9AX+RNBLh6Bx0uQ?i6DbvkAxnsFC6sPko}URRS9A>WPrXR!!*
zk6CVeVl<cDtxZ^r%>x@++N?v(2Hs_lp@i2mZ9YEWjZcqv6Zu}6r%^5StF7-tR@k0r
zVrA;5W*#GbVlY%9Fw~hdXaz}}MoT&3Zo@f6$rB`4;r=`qUpV93d1QHh*HfF>#%?E~
z`Ews(p(wi{v#x1vH@BvZcpl7V^*ka5A2veALr5jo@eTeimNJ(I0p`a~z7XKv48+~F
zeeO43H;dcgEw0<HW<;u&kmOap4?HXrL>iJjPQjBMXWTc22X%iQuRxu44?Niptj0z8
z2#0<cj&U)5lF?t)#^V=<$r;B(nm4~reO^ksfThIi0q!VhM8{d@!^5((l%(w2&X*9-
zBZl?XoWyxqa>$g353?>Y0+ZxK*V1NdH#?LF+8eVdN*@IpF1e4x2=DMoEyEE`M_J_i
zY}%V;l#%G4-V3eKbWm^m4OHM(VxJ9Oos6{))J-ljaS&m4q>z7U=rW!)S}Ko^AK~5k
z;J=<c3ruq(9p_Ph+rd$ZH!d!XS~lJBBHZlA?>+k$A$K#7@heBm<TZ$#?^8#m-0z!#
zwav(+@TUjgeO)+PwT*1|C$|d0!mYE=v2{w|{k^wbQ$G}`Z>JjyZ(Hb@wSrj4fT3ma
z^xV;V3ATD+WPR&cNbA<Wfcq9r>3%Z1qC$7k>w-reL%-grvWRQFp_uuqNa`j1RoJfo
zrz>dpqlZT{uFEiW@M()SREUHv!kex)3bw)!HaYNRi*JoFdVMm7nAR)xWb<vcU#U7v
zih%OI+Con;k1{Fh1TDF3%zg4$kIOuwbY{;zA2Q}yzrbo)DA+yAiM-#%O?2bWvsO8X
zn$EoP3g2q=d~6nhJn@SH&W-EdV%AC3j5U069_SyZ)Q$ThJ?#VR9Uc4pv2>^o4oEqs
z-V|Y1X+6i<?;mVj@LZ9wf+%LzR&sQw>?r*hYJc>nEPXUo2&<AxP%2k}WDIjP88^L(
zD|d5Vae)T)v<D@gz@q|p%FDYwao%_sDEy+XsKHK>U|H#y$47osbBnm!YUbjZO1b_v
zbyDp%%xV-A=AEyN{lXM{?SkpdRz<Xl4o4MH)wSk^9Tyxo7)r-^&IAye42y0F)kHfc
zmI?jgc<aqhiwT=4+P3Qq^l_ayEbg!%+_IW`aB{{C`4rk<Q=AZ(-Gz7BWpYc29(Um@
zFm9Co=o655-LXjg?_djtq9*3?gTv*c$|Dd>i%Q=4j3tab(eS0G^W)7rrY7MVlak=w
zP7B#Vo;bcIhcpjt3b9qcTAWj=vPw^LJS&TdilN3+^&~6RaqZY5g`DF^LshB?!H~<O
zU%TWJto|OZd=a2FF*`<$pp2nZYnr=DSUJcn?dXplm(1swHkjlUw;q;CR23{G{my#>
zS}VBhb~?1bb`dnq@NSj*vkbq^ZOxLEPaFSQkpTpKa|}y$+(mMNI}GUh^$6cdmJ74v
z5?4${X2s`_rk`_pjuOch+84UBj~A+7&4q7;`+>Ga8xcH@d52CzwbOldWg=MPB5+iZ
zZURr+g-T7uJqL{_0$?`iQ94gs)2BmQJS`2+)A0w6T~86h)jX3a^(MHI+nd7AH|~L-
z-R*q8=Eq}-J`Q0`^-hEN^nMeY5|WUMTZKuEi9=;j9F2bIfYcbZDg=`NKQocZ$S1GP
z_SGLHCOQhdUgu*xo2OZM^~X<Q$h5Y%u6KD=ZQ73Rx-_~TO4Q9IVcK4h`wc>dz5TKY
zvv%L5`toR{h9ngZn|^9E1Npzzmf@8-xo@S!OD9|bHi^q*9u?ToaY*_#Ac*s=R6{M)
zG(ox^T4p95%YM+SqByQvXa$?F;4NDi`K95HO}m)tXS0qMl-7e;4&N1EK}CdG?yp_v
zS#enCMJ_Ptl2Me1E|1)fIg|OD6&OG0ujRbzB4H7RJ<PYcydZ#?*dWSDv%hRZB_yP?
zZa*y}uYd2|kGOetN7$eXdwRA}#<3J5LFh)CMB<F_2e|r$V8J72MKRm*y!`7ZKvskj
zH53-b3{K#y3@-vc^ND*eqB(xAp3RyzvDR80X$mhOvYQ&=rT3hbChgk-nds?6hD6H8
zt3JIVFl(u2*rWsH6Zvr0j?38C1PE#2aVn~D2(G;Du_A@ZAmIKalH-l8Zl;#!O}vuw
z2EiRA`juhw-@CVL3P$eFxWcl*8-WqPChkWSf<qs$jYBPn(np<u?iD9Cf?yqkF4Gs+
z^G^G?QCo6(Byds`=f!akW-CF}&)4>o#rTjM5b-IM8`3vk>zi7^SPa#F7a@ItO609a
z83q5Q3NIp|w-@?kW7Tqo#xxr|6q7zC+3hHQHHU~F7VSTb%M7_7aW{R(N^=H)=v_Xa
z)p5aQsq(@;iNlUllf04;_42cK!u_t4bhnmCU2hK^^*FYgnqS|{Ba`Vji3<Z!-h8Q%
z{ACH>Wg}j7{mAR7JzBwND*E~CMfq_8<ksgI7@=A|OwexZPqaA%EHs5O(cSfQ2@w&9
zjrY|wdF5?ZdHk?@vi}02$dJV$sPg<24b(ov0x#<;RbCPV8+bJ&Y2J@xI|$Z&<aa11
zyqizA`k)81+BmOzm0cQ|IJIXu=PeSKOwZ8~EN#o}WTMeS$W=g=cuThlMO11kjRd|8
z@I9RR1{m6zKX)WZuo&~>!9)ojLm(4UlJg1a>*oUasyo;EOdoXt?CN^U{UBBc7XB+$
z;ob~Ry&Q{$CNY&V+}RP-SDfmQ|4P9`-V4C9hRv4)WB9!4Gr^ca7{ZB5;(!6cg9ru&
z4f_7@Xxr}e^L|Nw{R}H(tKbT`2FhCK*SBSCf#?YviObY+uhb;}snN7ve8(S^Kw+>q
zv{J@&B=?UI5d{DC27q3VPwVZR0PuC^uMf%v!WS1LQwlw;6N{Ke$%RHC?JDF=R*)~{
zCn{}hUOHj1M3E}+phnH{8EO3I7mdH-Z<!XXHEL{e45Z0%7=xs<L<9gk!;3-r(SIYo
zz{OW8;$KH+vOn{M>I3PYh3D@T=${wIB+vId!u|~o4~qYXZ}4~U|G6MGLcah5?#7FK
zhLZm~cYE^*OYUGrL#Kv|6L(Vx>iowc`t$$(^{H-*U`>~}jp6MQ&>%ThB!6^Ds%rRu
zyt)7MWoB4lo;Pp@e{JBHU_xg91!?`kwl3d;A47q)Bg_2#X&Ju)fM4iO0Q|51>Yx8l
z%dc%>x{q*3UtKOR-_gJPhjIVsJNR`CJd>EmC*ba$y#Ej0{k4b`nV94u_pta{@b4k{
z=Y`0dPaT_q3H7l~z?Y4o0HP@?Df}N0&7Xn&*ARUT{dKMd0rTF12Z#?LCSaoer(yZW
zZ~WsFC!n71migax_a`y%umNKf>@Ts;Ga-mfa9W47diGsD=)d_$pRj<yCar&~0KlJ)
z&nZ=({r@iVh=99ui;$l0PWtEFsSw*!9sfm~>q&jW1nJGv7Ep^3ccW9F{D-d|1`Zzg
zMO=vV6y;yT`g@>>27&N{`SPkt;Wh9Vde4dE|6c+7UkB3vw<p^Fx#tFg2Zdxz05-!n
z@=tkaMP%9Qut6~=H<z56TKYltuKDumx?(F6=#JSzc;9^Kbk9LTqQCrf;xr=QY<93z
z({4_7d?fCLM5wf)iUtJQoZo*V$1jULmlqGvfhq920`>opD1f`Ec3-ZL=6%(7{*gnq
zkNg%eWT4RV|7XnKqhzTPTq6C4*#}Jn?iL^>on(CuHA~tlAqjH|7O=|<3lb4ax%z#U
z|L3~)uWRp7iT5MOO5(cBE5Xm{M)g0;xql?2f26BFnIMdwnA}fA1YJqoiqd#U>OW__
zzkl<-;dvQ4#Dso)%zYyXlOghz29yUO+6DhtR;`pkem@h;my>)G_0K%}e7(#GB$5cy
z*VW)aq7s!Q?k4*`yJ+hPezLmm%tBH7=MD*s1H1-UR$u?O7d&T_UmJDE|L(>ExI5(b
z1#ow`)PLg}|8Kn;8}-*zA#+-l{b$gn5q-r5$tmDxFau$-4(uoX=kGx;j*gjGIK%VS
zG&npw-g>G$zPp~DVCrW!fws1`&~$b~gW6%^9sSPXPW+OtYccD@G?X?b0~r{fHzXWp
zFW1(+7NDZ1<XF0w==`rj8X63U7UECbqGC$GDv<I#L4SR!bcTmxami?bZ0bf2-v7nh
zTSmpXWox4e?ry;$xJz&g?jGDdxCOUhA-KB-2*E8l6b?ZG!J%*q?i4QHOM34<ea`8=
z_Z#=e{ez4F1+~_ibInKQd>x6;z>*SDLlp31iHS<WvdjMa_FmiB9>JA8WR4&m%L6WP
zL*2`Sd$Zj-(bOJf+0Mkbb_!98Ppc0J$?4R8Y0U|7|Ft^gD2Tt(bqKhT$|U|S4~+wX
z-9xVKW?}?35mgLWu5WBZaF`p-R!{jh-<>z4ns>j3iW|0OWf9-(8$EqFx;yDXhi5Jv
z$q~|XXj=2O7<W1COO<hTJ&zMXcI0hA%xYUT=ex>sEiAPu1`rtT#O$m<3t5`7^xuE+
z@4WVp01{UKG6@E>$I0_m-Tho;y-Tym6!c2y?d_K{6E8?gO-1)Q1c(HOb_iZNw|z$2
z-I34PFL89#8`qOh1-Wm<#DWBGcU}+SE+<e2B;qn^Mm>i8#!mQ?vOTd^0|PFR{(Ay?
z^}O=pL}x+IajNj=e7L?|4T)2G_eScd{mM||sG@P1!FJYPeuxSK3rb7hxf$?iDmNsj
z3~^~0VJG-htNgbY{AZ3ru!71L)RUl~&eB^BSo*IDDsw|@2-r4S+UyTsre>#y`%?9p
z{QMpY6`&|6oc>`AFsc{=J+=^S-Z8I)Z3;6b{rej2QUZ{6r5wuC|54f82YYXhOH&Ck
zE9Q8v2W&<2tRy8W35xz6%VqWQ@|u{Rr_bcJPtY>*<I3ae1L$q~RDGA4@AdVAH`_*e
z4*XUfd@k$Zv~set;Z(h}G6sIxz8yC{It_N|b}OKz3R8M|`m^heSTBtkJK%xm5M+E7
zwHl2W;afVs+ns_UCxv_onUJKElpz4rF_mfht-0lfW77<gcsvf8SGicUUh|~N*c()=
zaeY2Nn=6O~ME%<J+iSB}JqDumjf4dAJrik}wePzL3E7O}&NUO0^YqMa^=ftYWl0qe
zQVcxyIJ;M@zi@tLQEyx=7}d)6mvFSU@)y;Vsx<`E>vBI`sqgMX$kbmuc1xst_xcXL
z5>WcGu~Asmpd{C!PrcH9RjI)LzVaa8Hhp4#hJlr(W&H4!a}}pUv%$l{UaC<7gbd{P
zWA#FWCFOb(Xa4NQ>TK6l-p?gMSuwdJjF={L(n`j~vOz*hTFys3@txx`)4s5Fe$F6W
z@(5TOP3Q^=k~L=Nk-W~k@x!dpJlBtk{<XMwON+&9%Fi>%*=(V+l}4w#*g{NhT`tZT
zSej!XiSBEA`_e?4ZR)#nBic-n8*MNl(S3?pqoN=(@BoMe+FW0?<|}W=O4ZQhm>~8E
zUs~Y5KLj$J#J~am%^PaQ0$hQ5p<)pMDj9KSXTw!8&Z_ZPB2g0)MPoy?7OXc+3fO9K
zYqTOa>GzwoA}Q)he0|b7Pb#Jn!hH0Fi$2FsiGWhbpd~G}TP2_J<BoBW2a=7a_Ejkx
z=)hblb$FWXc+^-9c!Ew8nn`AE&r^wMIl&57mc8^@I_EJg>%}d=$xNdQ%k~M}RLnT2
zi)yLm&^ev!nHZn2OS8JglpZ`t#FvqPiF&=HzSWPaW~52O8N<)NDK^yPca<1CDya9x
z)C^@m+!?f2gl{o9K<4#O2~c@!)2+vGJ)?J_g<D%*Tl+@`aXe&gE5+o&$bhD_b%N|q
zU(&ohYYTgMdGSNOktwP2(!I(D@T-6bT~F8hINbJd=ouMP8LQeR)B?;b|9zm6zg+o)
z8<+pWji=~x*lOT~1z6FtS>N;Mw$t1FiC@3S&E)%sURmW(=%nkE#t9rXsXsp#w+9L$
zhE&e#e^X{8=js|+e(zb7bWEo{+sU!D8oWI?Ty?)7FVl4#AFm2+vGQRfn#1+Dor)kb
zsY0(|XwG?Kv;As0aZJmPWn5L2!{-v@GQuuhcBZGRyTyZVKxMGBB)m2k9e>tCmO~@*
zusdIR&sf={HZ(k}c&V+`fL+-p#eswj`{2<Nb}Zl5a$Hzfuk*!pgPMy+tA8YoUXt@N
zA%LZASw$usc>2qRe?YT29&XuLL%>7wvlHU`^sJniZc9`;C^c%qF@QSm0Nf(EY<QN}
zk}3?a!<5nwzp)_7hd&k1ei98QJ7HxeF(q!3FiCP=;eAGDcRr=SRU;1}b{V}XtvrlY
ztbGi|RWs6Hd1&`bFjOI*M9ZyWl2>`cjTNFyCYosm1X-Ctr=*1J*Za!>S~l{HvidF&
z3WBTTgP8|dnG7=l#Y1dz;nuq@-N{;v$T)>($Dm|;qso+-7TuS}Q*FNK)lV-I09Lro
z^E`TD9=xX+g>AoHg`FMnc=MdKzA<&&Z&ejG&ubIcO<|^||A$z4`^*n1{lyRc+2*+Y
zvMNx5z;W9obBeUPd&;8gm{3Ae5|x1aL7d3_8pEIcUG%YEXiz+A-X5FSt4YsFOQVB@
zg*|(?dNku@(*Le<nZw6PM*)i)l36fosCT3}OZId<;L`u9xGbP`zp`~JlC{mi*BXBg
zSFQ4Imjs|BG;inZQ>$Iu;%W<I`Tm4341*h&<G37~M8e$d0X6vh)U0fPQLdIVXY51Z
z$Zepl`a^+LPQsy)53(`!-2o%uV96Oe?@CmbnFn(@Y!L^{CnxCWQwkXv)ZQ=q6gFF?
zE*$^ua(U4Ertj0&J1|_r!-S>nGL>=|=aC}#ZLdICm7U$?z>!tsi>jYwC{xFkL*oi?
zo)>-EA*+6+Ow6Kre78Iu$&2BM?*v3)c5Q6WO}#g(042eibE+fYa{YWLG*vaj_atBK
zb_?GSj~>u~JN#XHF~f!(1Mc*aw;C24Q@M;<8;d(^Z~hl?koFp5NM-c5O{f0z)z3x@
z#2^a1o7K^C<tU9r|0uc}np9TV0%mkatQ5XGsN1@jR91&IaD{NyR)is;@BQrs_&!yp
z(jEsD89U6<Ix-XPDIg~2=2oxoy<241wPd1m-WZnhtp#|r$oW|}GV5)KnU+lR-BGLg
zVv}>N?JUd9p<{qmZDUKViTG1mq`|E3_2$M<La$uVRu5n|(n!Hx(5y;`O?_P2-lDx<
zBOIioOFA_2qbGZKC<*kf25UQIGyWd@luR~+&+1h-v5g3rY7RfYwipc}EA_s=8(k=D
z{@Q<{6Jd449G3kg4<TCg;L<oI{!xst=G$lTipMH|@48q-YqpcW82nn>N7p|v>%(;Y
zo`?38@5M->W`=}L`J8>?Tb``?F{VIj{0XyIAr?o_jb)zvwZOkXn}0>59}$7FK;bRY
zBopw78|7JsdP7lUsTHkXd-tYw8}~2PSMv#_(EuUmDE&fgDB$rtLl82x`n7HFtIM#Z
zw2aqXMI#`AlX0~jaw>~3JD)nggzj9xRU`G<G?7(S#&3GndfBqiL?JWuBT3y+zpxS!
z=i}11)kQ??Ex@-^<&Q+wOMG2GHeA?zUfA3q_2@TX58c<bG%YIf7=9bj)DkqzSO8N4
z8zlD)@NY$s_sjcPgfcymdB|h?`C)v1ypXI+dxH@}PHrvBT3CL6H>q|zZBr#Z$iR?7
zrvC^JImVDBPuZEgcpaPIr|jN$?`1m}TstiO9&xHu>AP&65AfVo(WAC=v}9CDie53A
z|CILcfFn0&#j8&ZusyN><^Ftg2F=la4m3?lNKD)er_E6j=`z4uwDh6@4w#m4Zh&x8
zynSuSVlAM*rm#TJEUR!Ug^nX|8O;Y1I&R7{3@ElrigVtvTg~{NS1h&&?6g^<VPd)*
zP<NzeonlW>yIBaB-(_$CWliK%YHC>%)#JnzV3&PRa?`;`S6m{=J{!x_$HSMYQ`XN5
z`tiCQ@TN_uY1wK|VRl?6N6ca|P;5y@9>>jvz|_U&pH=u<q(>Bdp>B^m`yG5GtGLTu
zb#`$yCG2<kWIkLvwsbafx68-jKk7dOnLVF1Vn(2Iq~;r^2|P*#IQW2@;b4Y^vL}Cy
z9RJ_xZ&Ob^{?jU91FMww5$CJUa|uiE9GKSK-O^+gN${keooue=<{^SWbnZ3cGA^f5
z%j+N2TmYNa2?d=(;GRGNEC_9G!c*I2#@Zo#D9*<AFBU*(5PZA_YQ6*}B5u2+P-J{v
zsra;bim=B6{XtTm<wNiz$ltoWGQqw~YxwqR1-JM9M4>K5MAeeiHLJ)+jKK4#rA)FH
z@O^c1jwZ;SqiF^Nn2e)-Vi!MIXv@6^R0of@G9fs<EPMbo(1`P5+%20mOg>)r^I64e
zj|<Xfh2U-(_BA8Z%Ub}ew0Ji11d&u@b8A5&k0oHxR6nsdB~JXY^C&54wz=$ctRcee
zW1Y_QFd$N{nRxS`$O4>~S7|EG#du?0oN4VLUEwu6E)vC71Y>2=nalA{n?%L0wb36a
zV86Wc_s>a6P999wYSEX{QN+MFqg9S)2QKO?L~LrI4i!pDcu81o&9?K#4o@BF9iBrY
z-8LC8QD*tF?weGjrjyXnSsz%XStOK_IILInBFzibD0Y7vHt&g>IsscN5AbF(UW>_9
z+hgsl6)WemG%lV=C)@8%6xG_z4+pZW9sf_R;%^8>1MrdTFX<KPp4m?}9JTx4+C8xC
zjqUJO0CdykmpIDW_iL2t%~ysKzvk%hA8*AH+ouB>ntCO+q-=lJh{SD{GbP`YJ+)Cj
z*V_xBlV}2BzYp9GA-8@3DL^?me6`}yMbGi@xnRmowDnR>A&JxBA(6ohGR!)7i@w5B
zJ^<Lzso77RKb;znwjC;khsUO#9nTEiDGOTO@Hquyi$xgB8a@Yjn8@nKhpoLb6Adbt
zrw8;OL&=QL2Hh+FX?GL=xzQ9#EzJ_p<!eozH^+i&?jAc<grchB^pjS;H}I+m^gN4B
z?awe4n5WAyYe>Ya?IR}_A^6y2Td;&UYdFt0U%Ocqps^)%f6{X{DsW^fnRj#*i9DLR
zq0B4VH`TNnWF(`fr<W(M@I;!L#YAi5+TKe|Q$l`%09U2_w>$b_vee=!A)3t3xW$h7
z^`duIJHA?V*Y5%|v#oNGW~knP_InNBVeX<HdSE|OZ0s%R*1i<x+yI(f;l`<oN@=4r
zpTSHvRsIaU%_2Q~h8)dAY_XxJo9EhEzw7r3P`ka;L*+hxn+fB6^f&il5uB%{+amOf
z`ab(VR7IUI0D_<)54FYvJ`sKTD-Y^x8MxKktsJ%Acwf$24eu7W2BjLg9=g@qmwDV{
z7l$UT7~T5t9sCkk)ce58Tv%9CG*l2PT$*zOJjOH$sDm782byd1k38WD7w}eSY(Y2u
z6Ocd-vbzO&4YpfwEy-EG-_u5l5CgxilHP|hU<Nw4S>3Sp%ArS_;qfTmP=${6ag8w9
zqGt$wD>BvaVkjpe@ilU#TOTqz5_jw$<1yf-P-^f$ZT3BjeL80_I{@7cQiuphKeWcj
z1wC6wVlPf7iZ$#A$_5cBs0*s!Dm9(yy6qnt1yGX-ofj6_05BoY!4r^5e=Ih;78MkT
zj|x0N?H=zVe?SbN{a=#QKX5&pC?NkJ%9vyTKx^<*e%v})lA+AaZCMrIB{+@)fx6V6
zfT18RV?rMCYA&@ao@7eM^Z=Gi$nrUAE<NK}{*Y(~ZW=re23AZ$O6rHYvJkVgbIsYB
zG}&M<y5RfgH7hDA8hjlqSPoV6v_bY1mujeMd3*c4%@-)*IZz^E#Maf-x03c9u(W-X
z{BJ_b?>ns8hH@am*h$E2RK55&_o3UU6@1lQB5R$^6S`W(<$4#pA&iv?QS6cvlhf40
z$L)F<wl<6YA_t4eg`P9Qdn<rtB4c%OX`JE-bnd?U-kOvSnx!?S;~pIV;{}gDWn=;g
zKJ1}JCi$=~ZY*^?XaE{7iq6+9{2#s0=n2zRO?B>KGF_;#w~)RB5x3HoD=SuERshk?
zRg+TRjXc}9VaLjqpG!lf@BiIMX@S`LPvYtC9Q0RiG6oElICm&vWiY^*98?D^TmUBH
z7vQI65muTh*Lhvlewp#+&6}aQgc_IMI##7WbqfCrk^!)_mZNU7t3!}{e0*F@U7c}E
z@H`61xT=18w3K0Y?}+qU<fIr966ol5-+UBwCkMC#1g{od&j7ohyq|c%4~Hu?r`fHs
z>8ZbYHITtw#iGl81&vSnkGXl*&A8lLMmK>5`FW2Oxu2Kz6Z4By<Iac)8v2xzF}!M6
zlM@<Vj!%cIY7Y;VRkajx#CNU3RCV99AK-?Q70CjqqBS&1M+ei>6L_2xl16e~e{Mh9
zt^3DyG!*#$#p#QZf)d%8wi^Atp<yZ(dY!MtE}tW2Czj?Z*qvG8H~J)5_Hq0&Tz1xK
zp8x~e)0CPPqphJ4%b{X4uw@md)c{W+|L#LnZ!r}S5`7bz*bj>BtSX6;<s66ki&MF0
zPZ7VKZgM987yvkDDR&eSKA~&G%DUnosYe2EYY7|^Ti1-0nk>_CLm&gj=L$FltQ#H?
zlQ7i=^-9D_B+#}V-w~t=`*oI<7Tl{v%RomLcX!;O{BYk$jXvnZxXih<k_Pw={z;_&
z>ld9OK)QL|81{V004(2q@I*)wy72Ye3AIV{+<6vc=&|n~cI$w<N?sV*Rad4fT8aNu
zg75hL9Oc;SC)eyygZ%=WKXPashe<a~Gh{}Oiko|;8V&#HH&#hS8Dwx1AOqv&S=-;~
zT<iM|_Nf_ZIhM}ah(&aaI4xXkbr@Z~>Li)%&e-^so6Tb^xmG3^6(5`9ujcN1Uay7c
zXNNIHX5uzJGew7u&ikV{ZXFAqB^7XpR@Jr&@NcybkS09%d$yS4;@;t?IhhO!-wiF+
zjcPh{-xu9RAXx|%C@Ue-({VDhDx--!X+bqwg4B$ZtM2zxG7HPvmlz|l?M*+d)Q$4X
z`(G4r7S`3tv$UOvS@SeCmW&SkmyG>C?q@@w00Us_zp@jW<(@{Q-a96ITS>4k0N#>U
za;WT-vzYuEeFi*seF$8?a?4DIBZlPQdD3;hxd|{YK$dZ-D=NyzP;;sR+}V7PDRqnf
zsNIB|913>!acDOLszUN(V2nHE;Gpc;g}U-^+0#vIi6L^Mp;zKpfSI#S(BRN)21O{!
zs5mR>?(ds#Uu7n1NCl(gYvu6JX9*khANqVWk0oM6q%Xn3d{t7;M_9^GXfr7_U-R46
zf_a^BOEHIO5<QbuXQQ9N!?dE_?d)OuDH%dmuJUr1mh{bRaa<6G>eX{gQ^Zvc_32>A
z#5kBKnTY3y$8j6;Pp;O@A4d9gp`4c2xN0g&`YyE!eAg=UY5W>$E-uvE6sq2>EdbKc
zv}sJ0>uYf=GJW8$@?=v298K_GtNTZInN!2Mq_+S=MdjBnG@-9ouvlRKXkPSvhUePe
zw$*-J=lTJCkrnpl=4>X@aj+m($g(t-eKJRhTucIPiT7yt&Gr5N*p=sRS7ZOk<05~N
zyU&mnr&rt9C8bl@-@75}vS-r#?()!OT0=`f&yyg@{a3mbJdRp-UPeI+5~t-zoXF#;
z<yO!UK-MLdmJX>twE6P>`zoBk2EcTmGMZeko|Sv*P#hpDwnjkDw<&w}5N@#s+(c5*
z&<y21ehBl0OlzSn()vaK;>x{!&0hr@0}Il60o#8vH}C?6mWVpRu-Z4k%UY|;JeNsj
zAxsN{$Vw@$V|>!wj*o@_x&J9{$%+>V@Hi4r!5{Ov4dLkcE|~7_?8FVTwu<fT?SE<%
zYx@tNG)n9bRb2q68_zKhtk(ep#V{T&E^~2C%iEKc#$|7LLc;rjF$5h+F%`DZB#gK9
zDw|mkzkl>1Vq5IbFxT2l<KtDDYySJ?A%9u-3{t3Z{coPP0#G>t=<AMai5vv%4|(1x
z)_?f5K+I>E^Ro?0<IunUXCrve6_M!M4Qd~pm&ZS3eWo_f0geBE<ul`Ve|&72@&C8}
zk$;ubco;%(NUgc&!T-;0n^CO1xOJwne5PnX6ap8B$*|v)|Igo84SRu3SL5@qswWIk
zr^sRa>fXe9uIT{1@Bh`+^Z!lpeWn=xZ;T}W(1-v3`e~^bfu9lLzc^1)QBtb$gY}A9
zYTxYW7DPQhg7T`+!;oFi<*9_(bxv{?YV6x(0s?s7MY@p(vQc4BlvwK|BqcD6kJkW6
zj8A1?h-5|H!^4VT#_Qr8y%49e+GHIH+55W`H%bndWdm4AY-$>AC9rO0k>z%b>ZFX4
zzmYU5jxYjZ$j7%$(e>kOWN4zCum<K55+NC>obewMVb<5Ns2xXmHO$TY>fg-F!Wgt=
zv*%{yFz==}?=1^3Hp*K0=>3AMh89G={dP%xhBP3n@B8J$hwrj(eGSzg`;E^{#!*By
zo;TDESj`}CHZ!%5JO{^lVL`iMDI>hmJ4v%*8S{7F-)T9d#$4jrhU$)xkYF>0_9enq
zc1yg(hNHp42#-rZoW%WH83SmGb2hK!)LcrbiY<WlgW&IPru9I=SuDS3@V-sPRO~l7
zAs6?X7<5OxRWhuhqADVo2^W*w$&3I#yjti8emE|B;ID?kn!`p28PFzDJYRF#;hNi#
zR~$<ME&G=&X^^66Ahn>?)Bt5p2F2;AEEX=wYm6|UX{4yNRa{puqp+eV-Zmz07tX`P
zS1e%@FG>TShighsd`W2-mXH1SB!??Cvyk4B58Y)blL!?n0XI}WI5jVha*rfRL<Su2
z@ZdMc--Xb`{;W7lJd2e}^R?#@)a>+<g!>*PA~NjkHNXr}SrT_vMg!IdzkH#-s&GEt
zpRN1?g{8ic_3$P~s`7c}-6#;d>}>nmx<znF$j%M{JyUvlWmQ=T@r{^V7jL+{%)2ph
z3X3UsFX1#rDCbke1T9{}<cZ}OCPhd+pl5Qb*mx-OR5aUhpnc%`{rK=u3Hw6LTdJYZ
za`RiqBqcRQd;Z<h=#|@!&7eSh=MP%8MKk;FKf`+RrK9N6%O%2QQlv^MNSnpvq})MN
z#pXjtLIUF7Uq8I+$?wpv5cd`PvWeHkq?SQwQ<y&rrcf{L_w|+<?YysODSgKe)~ygm
zI*(O@@}hWN7(lQcq!0%dLJ<gi(Zf;AH}0iCD>8kmtW$PKQc?no4qqt8yC5mUE^Xj#
zC4iPmEntJ+8Mdf;`@r*n%7t)UV`-pjoDqkZm5r2REj$F36e)W`X(dK~>=4913hy4g
zM^Ck>A*I65gqh%QfeDBklH>WNo~+O#|8|82xS<-EFFYF0{ZMm@jTL3c?-5b1w2AD2
zQ}O<EoV?I{3lFgzFOE$s2;#4$6?OrE$bV4y?n0Xt7sq5@p{Q0@$3$Fx%<k4nf6>$g
zvm5(4E~ipy_>NL3`sUlW_u(<oiJu?+cMlE-P918m($L<(pc9t*AwK8&tb^xP>4`t4
zJSJt+4H!_zTAA-a11fY1ZF+NPLRJ>bPenI1BXtW<C%KDhaGU4iJAw*XzeA<LA3^A&
z2qg;|6wE(SMU#HW%NWOfR8N%Qpni!(zz4(OibTyvZQ$+WQ`GjM=SRskT_rtWaa(Ni
zeFNBgt7$uLs?QbQOoZYbu2TK@;MOo)S$b%*yqRvgv7VeZF^AzBg|@w0*ZUTR_ct`=
z`th3FM=`)hOMCa6GT#7#GYmE5ApNU^boD$MWsR-i8d{7h7-xBmceq^|;^|n4YSaQ%
zGD}dJbk_pDqunIHtB+$(aV4SXG&n+KNz{78b>>LFz)PMZ|ETdA#Hi4t45ZnU#+SB0
zX;Qt7wdy-}(?B)|=u;&9N#vw9N&GT8str2cK0AorH-;R()Gj0FWIoIDq`+l|qjmrx
zzbXU%rNTeX4a~tI=A^>S57sv2$)9w!UT`+eh~;@H&LPzMR2(d_C!}T3EE#00>0Gv-
zbs-vMy&tHx-w$J}HVC0ADnts&o|}W2?Nz%iDlU@`=Vw+#omf~n#AgYiPm><#?Qhjr
zg+NMQ>{c+N3d7PLimPfFP++R6CS>GL$79A%%#YL5w=%pV;ix;KCKXM*yGtfp{sd;g
z?dVU2gVqi*lG9D0qAn`kk81>37Wek|$w_Gjc2>MtEMAj{SN-LtP1_}hxc*AwO#w_k
za+J4EOOfv#O(T;roR+hU0AmBgtU51lPF6;~LQs)WVQZQAjhDTn?dhPJQH*e6y7Bq@
z)_Y1VJD9qXwhxJV3tvF#^RRno2Ed@8_9YJklOsXPD^J8^*wh)8oG~*lDLDm&EOJuJ
zkzR_VZv0xtIP+8>RhPU?APCEs)Ba9z?cpm{TN;cvQrl6bwbZB>U3W`CbWTuJ33^r5
zuc@U~{O0&5lkbK)Aug?4pV*k#o91D?u+}beZ<gSSpyTv3d`~FCgI3{GFJs>?CLm41
zY^1y11o#VtvjlpFhd(D4RkmlNR*pO+1L){JNq+g}!K%N#_7It-^G(Dw;(KF#=_z<}
z3AfZ?P1AwxSOq68)9A~ha++d3z@T`%Tt=+k?7btWW>zQB{u~pIxf?7W;tuS0e9UFT
zM~yyD(bd_)UYtfpq~)?5brS#OeBCu((Tt=F7H8)j;0d!N;vQP5F5}}9@3vjbL%%oz
z@|F`N-Ub(o|CKgod`_jq?t4c~*EE$D$8%y^gcp%Dq-7w8>cr$^1-P&|f5xhH-H(VU
zmnz)z$)s(dSH|4D@Zu68>&TN+APIR@5Y6P<dOpwuvUO`^m~Qti`tgGfqA2!c?NF|f
z0=hNdzKd}A-ou(!utpTptYqWeaO=^AKR&=Opj(vho=UHk<-XkW5?=KqiaWUXsNK&1
zJ@+~-wIFuYASlEU3N+3f4VjqUS}C$(Gc!b{iaI9w9sNqj3yTE17&Hv2Pu{E6u<?!|
z4A?Ppf<`HGez@%n=UBY8Hcz_crlFyh|L}py9cSu+pI^7As79%o^fLMTDOg*p;&*FT
zlol;`_DH)<pxE71_+YSW)Mu{5D-nC8-*~&HZ{ur;N6!2^b%V;^GizxV@UJq7_6tGW
zp6EE9tlDH%O65k`bkIIz!TVaioTrCx<iNyZ@9m8}+H9^S8zVtzmO(!S_<nQNB2!Uw
z^&kgN8=v>^8D<lBD2;}3t!eC=sj3q2%-ghw^O*p2#>K@4Ss&ZFymr4L;~6<{%go5j
zx{Go_S;b*BQd=Ntd;hG;2B*p>C~4HBo33%QP>Qv*2uusV=ug$5fZd;-`mTWA6WA=)
z_@SY(q&_h5j_}EKAljxP#=R$?zQgt@_R0IPOOPz!<QTkiY9v^7`J}7j^$;+AiNbx{
zi{PFNU{xx4BzlmjHTH`{x~1AfaI<*(T(*)~zhd&YAKzZGOLv&47}PIKS;jB%n{aXp
zRB4AJfUs_HmAJ);PZiPxw>Fazpos6}4((Z1YJ;IZGZtUwcUp4q<tuCRnyEt}oB7$4
zuTWpt-A$8IV}C8{AWRM%WDj9z>x5WaS(JP@iU}`xXkCidd7#1?ix?yqPLL{b*J}RG
zO~lG_z$BcVCL#7itfLnmg;LwT{z|IlQelqn=Z_(+Cq0b@u2mXoHoxg`MC98SO1m5B
zzC|hn5<|?y#S(A<h(rmk4DuRAXa*4~DUKK@QCQqv*y<BWb7oR9qP``cKhq745}EzL
zCtCJQ_4-<v^k#76a*4+8!A~vNBQI9bqjf{0gZC|&^6-c%QGO2Mv2c^mFVpv-Uknb~
z#8q>hu+^$xbl(wo=c|f!svx`rVd3FwU%a)uNIrS|5boQm;f8YfGSDE2ASznf2+^#(
z<;&E`$G3F1Qq^KehB~|HUN(vST6>54$r(c=6Z5aM^N_wi?F|reyEr#>(RGo~bEico
zJl^<Czl_`g?{V}5!-Y~OPG5&qqX({Nb^A~=T1DOJ2>i+$vwaS)i%dZbs#6xO41u+W
z`a0jTT*>l>D<?Mll8%0d%(?asy;cbcIZeZOFv>~l3CpD$2K5|!RXF)vXP_rGCVP`=
zd7nZj1&UwyJ?6O#GoR`wO*}N}O}r!&1lb%xwpfSqm0i~YS)wWon$sL)%QpV;2KRxC
zCBAy>bCc*`>|>MmIIM#<`frW}(q`av^jI1SNr4|b`+6qsC@><1@RD(>vsFq-Yi)6+
zR;Yp*AxUebRhD<v9lp=OKmqH+i=pj`JRtWj>9hqIUPCZ3yIw7?IsefRc*99_Yqk%2
zy`T_0sm{-SuIlYrTC4eRJ|@$+3n-a%^n8k|L-Q-kQU;32{Jg&<zQnD=^sUhc1NHV1
zj+$>U5)TC|s|*67Ovd}5!P~SN_309xZxe1@$uxSxO?KBGad8w?idEyt@z%bx;3ZXG
zcESbnlMv*ta=vHB^X0XO4f)ysyWhK>uPB>)0Fj02btJ4JU=M+~2^5tTC{9ec!YOGL
zh?tq-XD;r}d)d4uEbd`+w_ASb3{I3`^4zy!DD0>*;;21Oo_>j?L}pxoqsfrQYnE^K
zea+0nOtWK0+>gE4!M;+9NMvp!QdvP2k3j~bkX&JRpv(su&Ea|*FA6tn;*DAS`aE(F
zm+cH50h<Tw0?3)%KhEmr#*795j34=ND3BYmlJ&M@sP8qikbJdCp!p0X>|wMlCbzV4
z{M3{RZMK0ulsrZhCsPBv^Fa7m=^1nWw>AcBKN3Hv%(vNx%o9tK6?+)CxMlrXa=M>N
zU-y%8@AzJ!`e;mX`>G(_(kqo4E10ycn{)M}_9MY?0sUnC_THb$6ph>AB3}yP56OOn
z;Bl7=h2RJ}I&^tj;D!i)Qm^CfW5$fPdCg4CrLKrlQ-frSnI&_n;6701|F-Uy>TfTA
z^GE*GB<kg6`-<oYp=Ik;c9mF?OlJ6&Hw$X+%lT4Q1T3?LU>?nl@Z*;>CqhC(L*5k3
zoIb|mo3`RO$L055!uJZXLOw9jofvfb5c?s~j-MP#1Q9^1_H*VhO?Y(B{%RcWaAH6{
ze%}^6?HP8xJWY{o;j~|(Zj2$KXjQTKZUsS8rmGQU<0T@V&`yD;^s2#2v=Rcw>KBKH
z$wMwtxZ4Ja@6Aw|C>*RcMVqYj_1kJUC-7W6!w9oeW|swau&-%?{EQrP`8QuI4_c?U
zGV|X~`^=r=kbtodBN`Z18dS+iUr{T*lgKTbI~uMqI9btppozU59q29f|J>12iuZuM
zVmo_REg4=dH5h(XJ%1Hh<?FuCd-MQb6&u&qT?4(*QZ{KZIJdNvUrt?<A0L<Y1?{VG
zFq-SsYQHbglU45%?d7NUHm{FBUdE;UsXpX}5ftW>{w;bS;xIFb3oikfW<@+Aixf%A
zP$ok)%Tprz)0Ta<TYC}>?#4x=JX$M?U3&(G(=gjkuY#$}RmW<3>dP`KhIxE-14pgG
z6=YGVEevvnZn?bK+c>^^b0Xt@SQL$okZ7?m1IxADz+;%L-NCQ=>0B%#(bo8|y6{yb
z&e*A~)kqmcw!^j(BbF<&!ALzkSOX-)_1+C(;FNI0XKC=J`V!RH%EQf~3Y6VRDlfXA
z&SiZUj16MN`=rDovg@&a51tUN59{if72vNv3^1bk8&PsG>rH4VY>FyVFD)(EL_^CH
z`0m#{_VNEnjr)=tX}XNgSX+67=4Skg5{1dnm_5jRI$}smBtH&o1OX}5_L*dYf4<(D
znMn0ng_#nZhOmjCZ?&oBZ~AOZ7F8P6!Fuo_NkZ*&7YCg4NWW{>9-U&>TsG=O?FlI?
zw(&|}U*o;k{uat%{WT^K4{xv}eedF9u_O{V(W!}e5CZftiC;H7aw2+`!@zKRtzBJa
zCb6Q-_&~|ar=WG&QUWIO8Yiisvaq(Gn!RIz)k}r?fmJFj7BfN{gG?Lmt9Q4s>bg>u
zg%;*zis)mCuB}9MImO3suDHEPZ-2hI5U>-3m#gfvnur6m0kWgW&EYpjBF){Nr&gbc
zc3%B_jA;-jDiyAtUhI}$!7pWOy!r0*wgb6{8bN8>410Xm2PTMm5tZ_$PMc$Rh>ncg
ziY;D)rZ<~pueaebGf<KuMB^t`XH;C!AZboXOX-yDIW{t;+n4^HQmPK8w%-1Y>|d+`
z!;M}#99_C_nl3U<^=f(sM-W~hl0i>joL(~`+AVKww(gnCLV~N@>+aCF7(AMQres89
zMLWv<3JmrVk8mng=p!bf{aigZC3}ety#l2RB5f-R%WY(k?2gE_)Skh?s?fpS1_pKO
zw9oS7VY0va$a`^S9KY8D6p{gRP(f<#U0q0a9A+b?@dTy9r6PV-7umA`tiIEtBtbj7
z^L{ctt+S=ubUELa6`7e)p*(VLVK52DcD+S3zbV>}6&lg)Z$m#Wr)zdp-S(ALUE%+x
zPpcwNPIfri3s7ngJM=|%s#xN#@d-He9tmbAySxGoI67l|yOnJApQdd!Gh_`)vNvF@
zrV$IuH2SGnglCKNP=|07O-VU*r&t#QG`26Q>*22+J8^}a7AnuZ?|O-6S8Xh)anu<|
ze3jy4+QvWKaO+YY!&SUJ)#Xsi4<Vgd5lQn(TylI3?tYFbJhg}cWvz|5QQd`QPdC$t
zJ&9MN$n<*;^$UGvk0cl)giQeBq0y^?On>X$O5)2!Ul_9CiTb<!Xwz5Q$7uB-tR}2J
zd%O!_gkGrq1#gXJcMI0Oe1Mr@2agtLX`v#OR=n&E>)IXP)BzK-WQIpzxB+E^PlbDi
z#mS;fho1<MK5ajSKBpWlZ0#d16S2@0UMCN_?O~H=*IH8P*>SM7e<Kh;b5gu^uz&;<
z4)|{PR5Fg`v%G`H?5WxotKW*1&VlB)gH{$vptdLQ1ac}FoYql^Ok^}HR?6mJ^JP3`
z3(#b|ySj(kzo$_&!Lf@+DaDwnqhKRZWubgpv1GOuizp8L%<XLRZG~E~b;;8NyRy0$
zw*#(t%79iAo;LpWr`3ayho*x%h%)}o<mj@Yc8Hym&pRMYZY6VvOO5Mhq44qEn)SCR
z^;$ocL7z6XTu-IqA9Q0{yvNRoY0#{f1x!;!@M(4->r~%og@S=`CfFVbDN;wfQ1x9Z
z4~B0X8o4sa@cG>LzIgQg-VZ@iS}-kmH3-Cn?;{Mu*zF$e7HDXI%L{g6sy}_Wf*kzQ
z;75aiAq1i#>`I~Qb0o97h;GVBn)yYEw)I%0;5H;U?n~e3>DBVdJ;<axR#A+~D0Ax?
z^@wfFA7fA*VcZd?3k&lIQ794Y#G3x47v65@=z=X1U2+9MN+$3#Jp0(}m4pwAA_2$7
z9bejG*-9FOi-&<Jr%-9qt3|NZrK3yIAvGh|i-x%R<Lcp<LEi6~;Z%z#tLO9WP0=@d
zhg|7uJRKea+G;*Mp7NPPS};C`vq>}_o;E7hsWo}j&M#-LAknQA_ziJrzIaf&gIgni
z0dm%zSoGg9u;f*R8=BQN4lXJk@j(k0{9J%Aj5FKVIAz;o{jI-jx}lMOCM!oBUVzMf
zPpG`3qwLU^?BLXVzqb+7f?tS?v_uSvn=x{rb>@y`cxttL{4iRJcIF?I-})w@ClwO?
z8oyx$s4jlx&o%J2d2XSLG<JToR~NCex)$!YWu<mQj&)~rdDoh8L5nV~8iqVCW1|f2
z*8VRlG{fHq!V2%H3U{{S=4{X05Z(pQjfc+xZGR(|Gbh8qj4Ja?;Do{>(God#kK3OM
zvCF!g50nA`c9+w&NN21lSLO9Dp~e08^l3lc_qi~}Gx7HN;214Lao+kp^rbpAGiH7b
zVHi`yQ<OOs>?^gW+gJW>Jc<116%Xf^i>^co!6Rj}?H#l#6djjLS|Hb6dmm^D8MFCQ
z$enMeD4w$)?1$9Q29!lAKR~)zh*dxN*Bl`={66Ba><oMF6UjkD5@GJv9Ry6TDn|B=
zr(8?oR1yf&BK*J^tWKBqapO$@U7Ag%DQ|PF8Ii=x*n;x)qsQ*3I8!&wQv8UAo7P!n
zoM`UXhcQ}pK{3mUNgar`)qX=q0#iGn9_B)0@pp*oaEJ~|PI5?yczB-ZHuL}!+x#Oi
zM?<xxtN4y)bC&x^NYb-?<y>@F@;KssrrDlewS9Z$XS@iIiH1y<$E4et(8Mq1uaDsu
zkv(T>+B=@D-N+3A>n$&}ME&)7ZCL7qT+esB;VFOc^tX>+%|B#r)~-Wvv}#Uzp{AA|
zil1m3Z~dl`=|SaOkV*=h(&562z9rmb^<xtOd%|m7S9`1`;wQcf=a&$8tMVjUj}cqS
z&)b5UwzOlxQW^5<?YITl7H>cI`k*4(LY1RNCHN<f?z~mn87NDhc{O6|R?a6>jAki(
zu<4^65KwkbLko(2k666j^IAj3*@I}7MPtk9)S7wzFI?W^1u&~zCo+I~=hmjXzgD>j
zAdP238~M@W-gd9<phdl4A6Jz_xw>ZVZQ)bHZN`e%%=jKC#66B^E6I`5aJwnQkQgtL
z?ZaV_j*y3&n@lnpCIMc(ry*6{*hUp9T_Fr-Q1=EQK-(?^ipd$PRN~Wf(;0hl>}+T+
z@S@G|zdp%^ckLu=!h12Xvsu}IfkeadY2N#7@&Fu4f)0=DZ7qb>8w!?sR{y5c=H(2s
z?nDTVa-u<Rcwm)6dw^UTeK!t~HJf&uKNjDPrKqEQuqVTuwkYsJ_kMRN$BQcSOKH=}
zBM-3ZGfHf)npxRx1b|cL_o|DDxPE`>p3kQ#?U&;x{!?e}1*-}rZ55VVRnc)O3#dp`
za7=@xwWFhkGd@5dpjKdv%naBFv-R!HUtd!Jl&yv+_>}Za;W21fr@dIXqv3}5J!Vz2
zAREvq$ZFdji%M3|vS&cCv(k~{1+iSLCJce(3D1*MSsG04D#P8J$&%Rt!-dhfS{iXF
z`d&-<RL6d)oLshlG!@@r77^O8v?fZ`WLWcJwNK=XYel>A-*E^e{B!)q$>6HHp(7cX
zDzg;8+UFLCz*wSU*av6ng`QA~ejLzvr8YVo>)YUC&ZB%seC3sC2Zwgw-cjfEyHb3$
zsGxbDnaQ2reBi9ng69CZTM^t^M&Y=R#JROVat@M=qr_DnoNO=L?-UOrH)eSx2k@=G
z|K1p-vIxF|y>$ZECiPB_nUBlxXLV!3z@GlhE}t9Qxe4t`&+VOFHV@yay@Vd$pX?Ul
zj{=*4ez&)AIZ%eZuSGlLgs-+|;~^J$<W&54Q_q((4({ANEnV{XE7P@A)wcmBul_1@
zntgjyS2nYZkmTCMwb`hpD@i^8xR?200X8wN9IbdguC(fnaiBehouPo8U6GIwy1ibr
z=CC{kHG$2VGb%tpAHJoaW}~w|+ql7+#4&u`(HkgVp8}o2dKxY(Y*bj8nEp8zs>Uj~
z6E}HM78#6NK1|p?LGpvA6)&>S4a@(bv)~$N5BZjN;@!x68-AQ85ZonOxU3QX*i$7U
zS-Fz61hU9p09Xs)T9u{ZPN`Ahkb9*0B6U;2A`G#BsdJh#ioNDODLCTW9UNg%33{Of
z({rfy4XVKBteW3syfUoA+m0nTOgMj;;iIcrc7CoV?EvxOP#_UaNMJ&IYyR8Wv>O9#
zIw6B(Ps6gCPdu*+wf&4Nf8&&XT#=DFh5p{MqYU>-@dc3Ht`v0~^QFqdmKF0R@f4NX
z#G}DW5pkwpSJHlRo;YyMElv)7bkY`F-Z}@@1z4{@*IkJVkNxzsPvnnNHuls1R9ai%
z;Os3n&Qp#=f7|q0)AJg<*r@i&%COZ8b*l@Cz!lAR_9xX?FdCQ(nnhIzs(If`p$q>F
zctxT19|>%6wi!YO%_vj6++&$DW<**H?o|@H*kIG1#daUK|A;U#&Z2=EOgmYRiqA@6
z;St5!UZzVs^h+=0u#_S3lG{U`M#!p(x?8_nX7Ph*8`xUiXjYfRR%xsxI2~Q!hV;^d
zmmb3r$X{TgtSt1{cNDNp2mLZNB7cJ^z?WYcDE%c4t8#<umrwtW2kmwb_zvpS1acY<
zJv}^FpWqLu6i}L(@~aDxj@9YrZbqi7tI{5+t9eybg(P!q{Mseah@6ngvulA>+XiAJ
zdP~_*BjM*%1+)_L(6!s*B@)2bQq19UA7(4?zWf~eCgdgO;O<X>!<-Dpn0y?y>Qh50
z3Q2pKn@IHhV|-2g+*{jV9p1gF+18f?PVsn{Yy4gSuFY|98@aF&@G4i*zPRdc5^sBn
zP=fFwX0`v;91@b3y#28(wT*ov8G@)}*TciK%<R-vfIj>(Rl52Tu|016sr{?{1u(EC
zD#-253_B`L&8Sg5gJ>~Swlz~yi3P3v0=7}OgtL{KQOXA-9(^PPh&GIs>4Lc;-xR0M
zn1_UlQh*&kK%nP!j!E3^Nj`|-YV$F{CBP-AST}mZH6`n#SMb5ZtvUPzP-=s22ottN
z;Sc9AiJ)eBD(7U0;7tv!E5R57s4EP0gL_xgIuw1&K63}0gj=q)iU(6YOK$5|6R3=;
zk%LER9?#ROqKvsM7f|n;7D%e_PBTXx+{!~U<4muF?5u<<?MJ{j+@DN^1rtJX?6Zu~
z3>)mTx1dd)Q^fN;X<%r;1y&uQov{9ZZQk~q+kYV7>b)Ck0-#*eMTEzw4h!UqTMqr2
zsaKV4@fBT5cPX`+wGbu7%P(u&7(BDbg`-xEg`9g2h-M8yZ?B`u8ciV;)mJ$s5aX|;
zbaP0s*;j?lS9r6Aq^DL=8Ilo}I%eirdlz4nZKa>Dg}VivrhH=R+e5UW@@AvS%(}4k
zi@J_STet4@DyC{#1=-Z^-<fbf5DwaXG_5mJ7%skX3X{~7oVN^&6euWUJ`kfWL~-DC
z`(_16E^oT5WhjK%0+)SNTXogR9kIi^A*$+1D}(2!wyQUSv8Z&Q{50!?O#juu$LpHT
zJ(h(1ee?TJTQ&!OL9>;RFOJ%{YR1eTwH;(MD?sHZ9Er0d+UMIWUCb;Kqa;B;jS7;|
zvss_<HLa0Y#4>C9Z*KYH8Tn&}!I3(h2M|9+AuoiK``j0F#yd~rzQ?X#*?wWRjELDA
z&JpB_J4IJNy_M1t{n%fYeR<(gF<tZfB8xxmraJL>&G<-!Eo_7l4+ib>`<_Q>MNl?+
ztT}~b2OM`q5B}$hhts`3^OnA#^+&%7OG8O>yuSRF*LRJw3AR;klyNp?*{^zn6)=Q`
zSQo5hXQ9J07V@@hPV|G8aVy;8;P1Yjq<r+K_!EjJ7<DAGhLue{n#?{Ug0UTffL4b^
z4B7{@I&)=;KB9N#=2H$vi(ks$p7krdqwO)V4nlRNVp_)B#IZcPT~!CzXJH;8MecAd
zZu>;!z??Eh4+1#feo0<0#74Pq;x`Qt2in=-_Qps4cKJ|ans_>S*DX_?xi*ieq<oJ<
z*vLRYj0J9Z;K2w%#oMvzbT!V-r7NCr>h9j-#r4U*gMN~k89h)i@Jq2di`KTRRV4E@
z(Jxe(B3W#;+<y{q;~14Qnz_;m2rR5fD)CsF0?3hH_}NMsRg(=fmZDjx1)DAgJc@lN
zI|uLaKfhC=^}zQ51^pb5GYMy0wWniJyCAai5B^zfB`al_QyMmYH+bZU%JR|FiT3M5
zeZi8)H`RHdT7@C86MijC&P#sdl8N(Pp@{^e(_?bLR0dh$NWa?a$*;$+CiJ-ZNs@3G
zzS8&{Q_gd4Yoi_Wa~O69(A4sMH6yUxh$~WX%Ij{%O?X2~)VLe2`1cGQJD(!CW|@DT
zSsnRm`@A{iZZ9Nm9Y8>T!n#E<s6sxkbJudHUf_2&hG-VuYZ@9&Hm$cVlyhwzH~l9F
z^wdySEtcE0|4h^kuM2lVuZ=sRE&CIuZvgwC`4-3Q<U@A452^<#(8agnd9I;+w=w<a
z{7I0|1j__%4#URz3!)XpeIYqO`WhYd>RjqDI!)WDK4jBz>5BmokkZqh3jum<3RsA{
zO9hN7p(%dCZ?K0`_A(Whz|Uq#H$?M_nxpn?+W+;l)tsww-{}thYEL{FXt%<(-adYI
z3#dPLbx?F-LkW3-e7~nrOR<<MeBsGfyd(JF-Y__NW-nYNuQ;4=!89vH$$eSVb!;9)
z{=(!1y>?luon0wyaVdu{Y|P9Zfx{UB6p<1hJ-9!k%D7ool|4ujTA~TK0KO{*8Di6p
z3*Es($#*MatNAG=lQ-2(!?2^1*;9Rf@eN%mjCaj((#GH-^Q7xm{i=l)CdzqIqZN)*
zmFFBK*k4O;ifWOLy4YOCQYr9u@IWd!-Dm&2aHkj1Pl2LQu;f*cB6pz*0{v}YvB<0w
zxV3Am+bTF>pg{SVAa?#(qfz5mkgr`aT~Jk}HPpZ-Vp0zfOMB43$;nd!ar=BnmEJr>
ziK;XzhqJBC`Lf?36$o-RvgoQ_m>8A$jv~48YZlo*o<ag%4ooA`{{jZeH}li$d0e$O
z9{!I?>ia9mp8d5I`_mLy1%*1VY-!=yKRNNw*}PCsvj69qQ8ID@PD~}TerU?-szOqR
zmd3%gaNds{FV@z-63&S}`1XaVvBtb*kK;2~`J?^8Ev39<`OZH>jIS1yg)RUg1<nxf
zsr{SjO1$kJ%iE0fyKY51{bP3zk8eE+;+P(ol<C8L%1IeU>#b39C7KECf90V{+Ws!o
zEEh+6>p1Wp5-W<gmL)JL@qJYAe6f=$DhkH!O?bL^_%oKiuxGAHJU0<|reLXO=W_R4
zj;B=^?U|z!Iu@Xa!%}>86p2pjT3GXQ#cy>Bc0x%D2<W{}`zwcK8T|lT60YuAaw_15
zy98`&WJwJ#oS(=yHjDQ|bu~a5E&L0L6Ad{(4=)TJSOSTVjAR|vNPh%?AQ?@SzhJ2D
zmsU1ku=*qL<rMpDU?;*_M#xANuOz`@K)Lq!l?g&SbGvKtW`*E9tbcE@Y2`mGRR$vp
z-!A}T>v)?#Vx!1t8d!@G)2UI4hE!miflD|M;?mM6lT#hk*o8lFG{cftmEj?{24X4A
zxw*OB=#!hA!ad#Epd6H{I2&atVK`E(o7QMruY!?}Y>Z7_o_kXnndU5cQ%H%Lu=jQ%
zP|xIv>~_4GGw!!`WcF8}1JsFNOT{3c9&qn@HnKBbq9w%8+B%64eU|$AFPYCvs@qsh
zE#hspXb=J-`^2vK6^YHeTqg{*lgGt+*VY9oa_)pxZ)75Kkg3jl+KLl>?GZCo$D~zG
zJ}x1k4QZ-%sE2@oSU?hTnv=r<Q4FIF5w<QqolfI%BEpXV1<pxd<CTImTh+rG0uvrT
zu=OK=<a5TpB)_D1xxr0PBd8)YVwZ9wuvY0&WpiQQ=L4Ua?Ccy4t#Rtc2L0+JD`Q&U
zPqC=bYOCzhp}Y(a=0#&dxxW!G{Wi-RUC&EttWBvgT;1)|qn0ErK!UOzgzrAk_NV;#
zNP(ujhJl+9??XI-i`1{6Vp?6*!vzFdcEdmJ$nSu3=Q~$&RZT(xtAFTEr21*{+~lVs
z>hJPo0dCkWJ_=!W*$T<}&E0(RcMq6bdpx4-w99}a9!=Mqrd)_E?CKsgM^vTsnH&)K
zyun@K;(mLpaLcT^YPP}b*f7Urh0h7)+A8fXsSH~A;cKfzwb}AH9G%sOcIvM}oo={N
z3Er65w{F~j3L_GVzTJ8?%~1kbIa&36`*KcpTHtY$@K@y*>;PKweu)oX>XfGt@E)PV
zEE@+q@+Rgp&^_Y#dgg2=w%xD27I}Y%W#ig~4!%;q=BPRo;4P_neP@A*3xj|91;Js8
z0sKY+e8x1Ns|GDk(osd#)@qUeNYD+9x6En+;h3e-(>puw2N^SXs{@f&F4M?OppjTz
zKl91TmSCmE75YjM+ONOQ?A;4pnJMLVsqTvTiRPR31{0PYt=hUsc@1Evm_M+^5bl*t
zTXN#<%9iW`@Q%tLJA2RTlbVdPPPYQMoqg^#s+$Ib46A+*#hUoRaY0oxrl_GEUQhAY
z4D#cg$n`?Myrh&I<~h1w3G?CRc`m^>er#{eJG^s7V`c)kkN0;~6OsIe02G#p0OJux
z@L7G>uiZ}$qq~<0klSkd+>#e~ymW+N)TPI4FKC;cCcc-Lq=IE);$*^=6_S#^GtHJX
zEgRMkmM@}q)hi+-2<LH@uV?Jpz9WkOOk3YCh-by<#i%$aIjHdHJqw#cBnBCsbofNG
z?Cs{LNe=BxNyzO?dSC;tJg@*S%QP#R3Oo}|X6D{>2;r5YjDp%RtF{sS%Mmk^UGvE~
zuQ0-CGe8iCrtK#RLCZReijGd5iqh}mfgeoufw>T&Za9e13lT@3)9(|&_oL*eQPLsH
z%Uykn=3ayxGhWzVutxEGW*r>dh1N_O3d;8_OGPeGxj<#=#dIPWgKsp{p6Vq08gLa!
zaBN<1YzsK@yrn5b*=!t2lSfD0xg5TH`2u^6-)5vz$+|Y_U{v4A9*HW^q^M%%edQ@l
zOczxLIG?r%UqWV=18n@?ffP^t&yrUyp~osAT$^QcXl2=*DZ;lL6@LQFDS&pl#~sXt
zV~_R0Zz?de23kxP@Wn-hhn^x96z<Fs$SNqPZZue0HJe~)3;cCUbfO;%Md3!-AK*7|
z&&t}`*6nRO2}MGJ-P2%?q)+@W07-QJ$?BND6<apN?gf@6c}Uyw>G9#pg7O{oCF>E9
z$P)80stmzPuxmk*np)Ce=@3?43~%w+#Ty+R9Ui|$wwgQk&8}~*Ke^+Be)kEXFKCA)
zV$pM7N-b=!f!C3yQ?pyU#P@SDJW^w}3tNngp~C1ukT$Pxw9K(~9cv{-gC^+R!r((U
zC*Gn({hTqP&*iIInDKTXFuADR)EaqR^;}+m<T&Z&_`*Iih;el6XGc~jr_=TpWLqd4
zXQ>*A9~;Bo@8@T9t5))Gf$b%?JviP`gSATt=vt6iWhME}g>c(8tA6RX7cXLb_x@F+
zD&-@o#>(cSyU;*9zjQyUDuD)TAg9nJPW^Hh9gZuS8JsP{g#x*j6Z#dU+DVo_UY`zO
zOQTY&WY6PAj%MOnRThmK5DAV0-%uI`B{h=nq=x2aoSsJv%04nWp_vL&cZud7CaS_B
z52l|#w3#wsvV;|`w$XN_;CdfeHS$Qe;YEqSHV|sg4o&oy8Pc|m9&Z+X7!bvhrOKZ=
z!Nz!lfV}m%Y)z^5%l#p5^AT@zRJ2V)8yQ3!|9?n(%cwe+EekZby9al-1ef4JgKRXo
zgy6x0yKLNp1b4RtcXtVH!QI{MedOG8``qr=J>D4a5BtyFl+9Xe)m&><`F<51Yq`MO
z8eS}`aPFU#^LdNip=-GIF{?L8g4pw@pr+@OqF}{AkMsE*2`iGkHj#lNE*1VvLOz_k
zzO?LX3$2o_ORN&1x`M(JUE4U{&sL*R_O>OixNvg_)@XEgnx$F|S2m{#Md)pzctw{9
z*2B&<P#fJ-a|ZfSomA(J96oEjkr?g?S{0u~cF`zUn5BONx>lQwiz7Bk+0F6NFf>y!
zF_FQ$YmmEMf^|vnwr*A;D+NB2F|%=@vy3VTSEQu7lQceLxoa9hO}7;jYBcurHl<6_
zg|Ki<7iwy4o%o5w#BUd&bTjoG^bEuwbGO|<h8S#C+k4oG!XIzZx;it(NoTo-{%D=2
z0JW*n>VRX*Bx~-eq}_x+m6sjflo1qt5rX9EPpK90oh=aapd7KRxPUv|rxmd%2*3TF
z!|In7{$^&J478R8wkql58u=i74M5ba!pHaTS^$IEeBZ|7)ep$)ab#ni^I6&{=<@wN
z7J;uB2{W&ug{UnaZq5CmE0w^0Y5cxjeIG4&K0U%W35o3dS)GE7!y`{FcJFWigvY*0
zn|y8uk;_;QXG$Ly+{520Sf;&f3$<gJKCv6j`Qt-(+qRxOCF&ww$jjDESrRY8poaS2
zMC1&<q|v_lHY5^>YR1l6Ghgr_(n%h1E`Rn6AVtUEcLb}FiHDG1i00ft<(6bDRvvw;
z7EP%@4_xf%MjwHw%pk953B;t8vl;6)M5=G*%fD|lHDK)6^LiAFH7HhdPG9*I-Os0X
z>gO3L>|1qIE{G#<eQ9Ev2`jrS@l_H!6g4IDw>RjplJg5E*d#w(s@>Quvt~#}w}yxW
zwP4{;@%$(Rr;x#V*bR%`ywg8G;A9HN)<7IFty=_s)DTrzq)f<KG-t;H#$qpfwDrEH
zoSVcR0nspQI$@elkExOtjLAv?)7p8KqgbX$uJ=iSUi(SkXWVV?hpO!U)4)WkkO-C_
zY4<~cT7vFYqL^j7Ch2O|(U%eW<YZbR4swH#t{kgQd^i{HN(eqYwYK2}*|g&UEL1SB
z|A=iUO2DhgL3f|hLxV%OAE@!6k0R-OXdfCR)9aX;ejiG`(YqXk=*#N;8oDRT=x#Io
zdE5CB64#RcQ`9JhnU`?oYZH`9*W<3>xGiQKos^1vpR;&x$0_CfVK@A%RyL7kHQYTF
zU&)QZG=GzdE5nj3YxG{8zHHRs@ZwO%8i#E`E1sNu&xB6Jnxil7_Ut*lTsEtaN=dA7
zk}N_7wzcf)Y<i*Hh<Y29fvXFvsqn`USL=dqbvfiU*gto`^Pu%yp$ar!60;L+-vqW2
zz2z9ZWh_IX``{MX)7zc9FlN<tM3x+y`(i<gZZoqON1mhFz0tV?EW$&PL{fYree2$j
zF#8ltzYT)JJl-%}d`aQ8{)5v9Qf$hpJH&Jy8{A>Vz){515hzY;>w=l{^!=JG;5O%r
zqE3ghT#=lUddhYeo_;>d{oiq)tmW1P!YVa8=L%IEdy0B)lF{jB;p5e7ZVq5rIRyMX
zo|<8nqJ!RfH${z*B(J))E7x{zpax+Jt@k>gPr)J>W=C-6u)o74EIv35y>#L^P`*EH
zKh-|IPFxA_83KTV51LK4rRmZU31`Qmpx#ZH=_O}tzeTz6-cb`<j07Iinc&jP9$`%0
zUd<5?nfU7FME#8swVKbMS$~BHhdkdMyV>6r$^nvRX8U7#>}KO|=OaqGqYCJbi~{CT
zpNK+%Xhi;G7H$=6y(_d@Hko(TDOha)kAL0C!Q(F#%*~azN>2_#b^{)_7R$&<BqM`Q
zq<l3P-ki>RFnO;Pn#8RvLnA31NuR=1pVCx)YGI>?y5d0@gA5_jmWyM^@Ns`AJ-*T)
z;#-^?oR76Ju{@k$p(;8$G&F>WFz^vWK?Zai-ZN!8Kq_gm1)GAp;J7WV#oryB!uTs(
zeU&zPq8hXVqju}B*Aw$!&D1%dFMd<vnZQ|IK4VC*z+Jhi24UYP$^h>pLoN^SJy>0p
zxmN>SgopI?CS!Yq@6$26yWeTnq!S--Be!}M6wSdmWB}}eB1jHsJPh62_~35(`k=;~
z(I9CoG*^)Uo{UOyE6H8#XJLh+RD;XKCED_dWFf~9G2^#EACHOvsmjR0$s2ZzD90P4
zt77+Wf46m_gdWv6F1g#ku}m&5V~0$u>W5Ar*oPEt2p@-y1&me?8odL_mg`h+qe7TS
z0e)?k`?wsYp5VQA?3&fV1k_sIJ1)RmT%mONzt>YQV)gLwcs~ry3ztVsQBO2`Oim6u
zFk;~Jl!N|6{FL~}phx}J5CQjr4ZTrzq2&7J)(Jj3hNx3|>c!rj0Rx4k?gl2g4bP`L
z7e`ZDlkirMX*H>hmk?gqr?%x%N8d>618sj(z;6d-q!4Y?1E+9w`YZNtuZ3tJf1r8|
zd44O${Ckn!Awt@z-_Z#sJ@ETWNk^>am;=5W-335}PmCcj(daTc!qM-D1jfYnBaXhX
z%-V~5s=*MJN-}p0KI?w7c@8r$JbVsE9;aTukoWmKkT0@y?n%mTdT_JtRZfru05bSQ
z?B~xfbDDhUdTa9s7(mzcX<p`3&S9dF0~8eJ<iq<d0UnsoVqjv0rP=@f78wO30R<)>
zJB2Ao`@eAYFL;$ukX>UI+sGKdgNKiP(<mke1RgC|K<DY`Y5tzUKUnz}Wd8GmEzldC
z{wF`3zZClc|G!83-@c;?9~5w?;t1w=|6i@Q7(n|2a2~;<=&m>hk%Q8}96M;HJVa;d
z?fPt2`OC+xPlQipPo&R}$IrKVozt@a9?i?1_*dWm+lP-d1)2B&)tJK77oaPrvH?;q
z<O(l6gZ{sUQa9g^5Bx_T0tw-uID$hS3e1--0Y8J52|;+-B!7)Y5L#SXA5FGU`hPa>
zgFp$i>(5Wtz8?VmOyS>V;D4XDKmW50A5t`D=qLk#%71GxC&HVDJ_oh3>3<n)MFcSI
z)8rnth(Lq>n*aYB{6Bjdedf~xv=^94|1Z06lZW1?Xl0wkF!g$Pc#}SMxW)6azg_R*
zVzN5oy^&)mZjs|_<U5CpO{e|KWR=AJPo~$6UT(O}jm<fY`}VQs?<|u<Gv)Rd(r@ko
zSG;?)b=K)wybJy|z*)r1GUlEsHD`HrS-C%mWezB~)!v?OnqT4+aZlgykI27Mu6^X;
zT&`-l3`PWQR^y9w@O<E{&`Clb8_D-lqKVCwzxNqYXp!6CAOCLA4GNg%y4JhfMRHs~
z!@6zBNEqM~NMiYDSu$qm5uo>UdHkF#M`_qH)x#HNo4*`v$Ys1ms<kkiZRp}t$#e6-
zICg$M<Q2OMH%Yj!Qq1ax`+Rrp6|}R^-hutoIyWUO+&Vjdg3ql#X?p=*=<J&{e30b8
zK?E#0NH@YQh?|s3G)J6pW9se=Gv{>SglEoqEO)@r$kS-t@bK7ezxrO4))+%-(s)d>
z{^te;12Od(r$UuN*J%oI8BrUP8(SR&H!nAd+8L9#51jqUEiY%5vp^Gx<=4?!j0X8u
zq+y~t=*FhUZeY1X$9V_W2Zb{q%<RCbt?%gc?0oBlE}ztb<8t=dq^32SMLOHx>gW)8
zP&OYvCz&^k2m1k+Eq#cV&0>1q0=Kfhdc)NcUKu?-CmR%JZS>Hmju?H0@n^}K>0Xwc
zwOei4sZhIcd~q1&*=J~U+syu2l53vt;fzrb4Lrv1ZQQpFn_Q7D-qd5C|BKRnrSvto
zItS5}p4QEzmrS>)?P~#0N+eNkI`H&?04OLV_f4r49|`?+ZGyu+irY?`tBcz+-oWRd
z7nA#M3IP<k5eq(wJSWd1?a3CaXlZU%+Irpv@I_9n@1loGk<Z^hwchYIEK2h6BVOrC
zTkOhPlnp~hQ|#MlW;4_+eWU_emUJ$-2`PQ~l18HG<lx96K5(~1%MdpppwYrf0Yefz
zI%;jwb~iq--$Fw|ANw&OZ?p6bS6(jdxRs=)N%@J|;Nn{%N_uwLm-S@2oHo@0cG+L%
z&6U0es{12L@9I1BD=1Lmw3&-zsOutXD#;^psVF@&yc_{!sBE#nP*f*t>zECja{{hB
zTJ)Q{dYM87aYT$bwE)rJw6_F%*9}C0hg#nZE9ES-PzxPgz6wX84O`<NJ{@U)iP$=W
z84wUsu3gC7-ri-W3zEs%K%z!9;W@ZgMldpcqX+=|r&vxOpXD2Q%lIS-itZjhCh>tC
zI$v9juV2!3T%;D|7q-xo>l-f4E70lR&Uh4ZY4p4=I*~6qs~XW&5|TPfQ^JTnn|LwZ
z6VsGihZumhXM8c<zT46*!pZzXHovO;Mq8UT3Av|hObs&Shwm*K>IFE<6=e`1m4n&{
zGubn;SzDX6kN^wIU9c&Ei(bIQG=4VEHRE7+gXfkTBu5LW6dbqvfXH78SX0yp(#rV1
zmPItjy=%;JJ6ZhW7r5R?jnJ_0$l2nF^ADmqateA<vTNjI;xYoi{~1pOnPxVKhfjUj
zb6jw6y};%6u(+1|o;(_=`hzzPWEX<r?S#`*vc$jJh)%C8K$l}?1n(^X7d#?9(4I~-
zE&GWN1D#^CR1feqBkX)ArQiRaE{?F4T}wZ@!Tg_kd6}xf>SyA;`$Fh1bzA3gU+SI&
zIJJcdAVRoGeQ~82V`uZYs<?X&UQQCV>hdwhREd&VE`4JO%gRWOQ0mH5&n)q{JXNS`
zyz9$pV5k~@Pj^eu{?nF6jOgb#Rf?s`xgaA}Vt+m{&qR^uVClTxUfgQZ@0po8X8{z<
zlko+LCBJ_ECKAL_YP}Xu{Tj=Dg+yOn^+S4U9uur47Q`8)QJss!@QopEc*K^JthRUY
zk{3nnY-(r}uRv4pvz!u|;mTUGV#rs%$cxzqq-qXHAr*636?%2EcX|$@9msG*n3+>e
znIlGn?orfq9~qbxbJg-iGatrj#a#C;cTQK+G_Pk`jEt;|B11M|*122Nf82sxe2PLD
z<5;nItotXF6G)Ps4sLO1xbCU?4Y<Cf>&ugek!>8@kdnwI+9$HYr~CJxP43&I#niOl
zKq4FnByT@(HR~pPXB{p3a)40*ydge5)qRRzwJ8%07?Pf+MUJ_q;oCc|89ch=VP5;k
zOlc?XZ4T~){P5J{`qW;%=ZtrDw;FMp9szsdoa?VUV}^T)35y_PZk_BM-w!u34s6;9
z=>1BtKR<7}y1W<cf%EtlSkU4)WA)vVgd}ZNtU`^_fh6vs`tHr93-PcmsUeEh>#Ei!
z#q0oczcYB&_<qIX!a+BR9!LD>tWoYqExDLsO61v|npow%yEXrXeOmmV3-?0y_rFYb
z_;q4NI92gFN&v0Ekaz2P%E|1PvByi8V|$C_`;mFOvw#Lxa+#^0MGEYmz-=3QxAYpa
zuNjkD>{0GbTQoA?j4<XRMoMmAzD5RfICqdNuQ|H^;zmPBG$m$XtsozA&+Ky+uK2og
z;D0RnC5)pJdo49rCR}FoO*%#j*C$h+9c4U|ylEYMQ(;aP8(ql}#1!^ltvA>8^)Bx$
z9Qn~4+#E4php?oxDp{E&Md=f-W7s2`_}?cZjlfiD=oz@TJUx1uF%gtIQn;%)+RhP>
z@;o=EkS`q4a9Wpp+;{fyHQn#(g=H3Mr;*x{-b(-eSu*U&`H0(UXU_cf`O9}o>VC~H
z_9&`{3JtAAu2~F&7NLwPdS6EnfF;@xoK>(n>bWJv#3rdy=~=3gYU{7JM%gV~$n6F5
zFIlrWjblrw?YTi-#haI|{)W(-#_+H0Ny0W!W^MPVubH(Op|f}C3dwC>32r!)Y&lPI
zzNGMiSw;$>9{3!UUM7i#R_p%0fSE<v!zJ}NOH%I{-^{nAHjU`%aPIwYIBq8lN8g>h
zheXZN1!0Zs2ozO<B|q)rfN{vpbC-Af0uyhK!HXl5)kj|nW{^+|O^OvP!w4@-#vi9f
z+=mr&HP<~b?ukZmT=P83d-;5(ADJV$Su1U?evFm7F2tm`SY8xQlEYKIM1(o;@<%Um
zd?gU6lG4!$Oy(}2kIWbX2dt8|AJjqJks7~5e34GO`x8`{vu$s+muI)dETf=3gyNVk
zd-(LS9zU@W{o3q@puKkt1zoKC^6+qxvmm@rOE7$hNa4K0pV?=i<QUz<WvqUnVmd~N
z`>H7JH+vQf-dYUx<!fF*jnh>8a*Gg@dA`n?2JSPX0v%bin3)-Op4(^K@CgZ?Q14k@
zOaJb)jCnVP)0kiac4zm_nw2eMiGVj6(t0E3yE48-4KJZl#_}dQkt;gJp~djAQ~Mm`
zZQw43Q6Osdfj>ItlVyRYiueHF+!=4DYp&=M2=%~mXH(O<IWtC~Qa&JJk6DeYwDVGc
z#&C;rFJfkB%A4WlI<`*K_F2Q<xVb%bKD(h3M_j6}zQ$s#S~$ejNpZgnXaIGmt^4~a
zL+|#b#YI(#>nQFSDs*wz!`$yZUh-xx*P+m3&^P$@qao96mW`l=2kB2nPxFN}WcoMO
zlTcS%jwCD|KZS;K_(H?t;hBKb?&pIrPgYoUCo<a{22Wh1OeCLg7^uFfSF7{eOt&u~
zc4A6>7f0L@G3tIk2so}2c$7)Ky1BLf4lJR`N3BP5qY{ARTPxeFnO4*{F!EAOe6Je6
z&f?unVCUvB(&d|8;IU7TbD6N*bHt$0sXp^H(eeUoXd!f>yb!a!VP<B=>q+dghrx&K
z471hxJ^VZ8RJg_3GFFATok0a`rL{}`KK<W5N}s<6V_$6Eg_4W?YU-{>^?K-fpFXR>
z-Ldt(|LL`w<gZ6n52&EqZ<jiw9!lTZKjOodvpG1(f+)JxTK5h^+_oym%0XZ<eJBWW
zkYxr>5+863>Cy?p_R7c$mhzEcW+E`~ShA|!DSYly%d6K;L+r*>#Y>=z?f-F!B>?--
z2i0k?s%2^B;#U-k@<iPKjZ~|4(#zG?mRbwHJqsTiCa%#_920w^;*)RNYb>=+Q|@^*
zWA)N$n<hS3C_dghiy`yNVZ*O&a1<Iw?G3fhllD!@dxH_xSbCpE1WGGKqO1-d`;Prp
zcODNp9o-_s=u3+AhkgBUsz}L<nxI2|hNqIwj^pAQD$ZW=MY?B-@Ml!DT?^YJb-+-f
zjNu@)y^3E{)Lh4eWrcl0wG+UXN5zO?N&J=MV9Lt*)qvN*Y26(OyKBmI{hORW4oOMt
zV$A#WF5V%ULoA)j(9kzqPh#hRIMu&;aRyb7wBWa1vB2rt>!M9{O#8B)LnokCVXR~0
zYE>MZQ-X-XmaDwomY?s9<@8P;5U~i)apJ?Y5-lvlZ{>y?Eo#`rif7;0*(PnX4~PKt
z?eAp*?i{)C%FOx9AncO#ThS!JT48Kq^3ZB55{3%r0361WkO_1;%=GlsN+qUvcz+?P
zB1y#Qy!zXYVm#7$Jv!_;QC)i73RENHMv2;2YPy;V3JN2-6?H>hka1TU`Hw-^*ZAOG
zSSGy>0-zHnfp5RcHMAn%J<F3ZHHk-0KYFG|v_2h8P@itUz8XGUB-Pu7*m8qg$|T4F
z70Oh*`UQ5H^gzbEjvm)A8ijQZHyjpiMTQe2iLIE|*05}H1__)$nHvTZnGgq3U+)F3
zf0)^KEO2xiPLt6PTcL&;rS-40tUXeu0=@C)#M<`lofdRE<Xf^E9Z4!|5>|X8x<X*@
z=*hvw;y{gd3<sD`<&|TCrR@agCQ+OB{)SV`8gSFKE%r@cz)%g1Ei2nU>zRzaoeErh
zO(rNnHo7_s4)YK&fTKIf%an^}p0efYN|Drxazm6EBMWXs-FNIkIsA6PWRb0RShhbh
z|5HPcCWV+xN$_Pa+D`w9EpLrX{cLqujyO{Qu)duFojgt>zn8Az<h)ax({JV0@wLK7
z+=e1AOOIWieZA@0QW2e~qc}B2MWjSmx@|nqcsWP0Jt7vh)7UT~1yT1uK6l?ko2Z1D
z9PoZQlOo+dX=w8(y`EgX>ga<Yp*fN9yGbQpsjO(&bw1W-4p>E=c5t>=DVB9dRWWHZ
zqXDUy4(ry+ruMo$!KT3dm_r^O#&Krune}ONhJH__8&joVSwhHlr49h<x^p=7ST9_X
zO@YV|Wl5iaMjibFbe(U-Md;V)jWnZTy-Uwv`@c6d(#wV>zkRv{=$W8;bowXZ)gMJ%
z(;#D?X1a`m79K%o4^v~K1lDRrx*F0YQJzaUGQPmTW^`R-us9v{!(JCTL}o8pJMP}<
zF(N}|UoIK&J|VB-kZZuetIAg(c*r#fuwAWiy|);~o=?Zc)}BL7&=(k#0BCIXyY`@O
zWsPJ!b|rKt6rvZusOr3(KfYVb-O+0O4QG*@uHgca3$5kjx*=s+V);w7v^|87&dCal
z)=rHq^9n3Z*bBLwp?B5Ufs*-<6O3I2fk&Y&yy#3n%B0;UkNUiETOj+|1M`y2B3t1%
z?%4^EEtq41^+qFBm}$q?WhE>ec@#>cZGT^I3vMgvEDw<ev}CU4ERJ#*@Oo`TJ+&<W
zjLIV2;4Sr0Q`hS;-;5e$8HREhEprPOKzWCcsKL!b7m7r36;UOjw~&oVZ%WA0dh%&f
zVsv}88qXVjcK&$|;q-Je`)6k>!t{Ym1q;PN{EsCxppm5mbb1^k3K$uxA!?In!Ox&^
z=Gjs7GBgiIFDL0*_0=xd$@;u<1YHZkFe0^SqDXuxO(5yin?ZDjWYOvNaD)~H(~(aW
zCX*<x16E334_9Ajw|&?|Q;=1OgWBC%lI@r`y5q;r${jG&H>IoI&XO;$mQ@!wL=ob@
z6TrQLQ5zYsTa^8n<LQ+4oZQL3mt_}M=p`E}H8*<~A^82V1!^g_f+_F@L)NvCRw8GX
zXK;Q;eZm$$_xIX3A~Q=j<;&jSnpRA{FYaJOjm}D!d^pSZ;qV7DLwDu#i!+A70|(=W
z#+f9z=d#H3P-W-q@U+mZ3m3(Y9ucce5S!1NyX~{CA%dvlynTk$sP)fSjP(XQV4I;M
zH>N9yC<U@`VYU>KN6zV#@I%sB4gcmN^zVa?SH0ge1<Lc2q}rDfk_Ritf)eyw+iKT_
zd>^q^SDUD$(`R4fNnfjLovQ}W_Bg>T+S=lQuWJ${LkMFFOjxxZi)txHl6uvaWdwb_
zQ2=(Y(r4|{-U|)tL=GDA1`W6^Vv@+Dq$)hGA$Az9es8A!SeUBN%I1QO3rl^ZWz!}W
zga{wkpm_BD+f=v?!?$AAw6|V&nu_$Bt?C{2sZyt*M#XWE@*0F)E~J+ozgQ)~>0WR7
zG<KImpq_iaR;(d!+6M(c_xC#*h#HD_E7SUyjYRIYQ|7TUo?Ll;HbjiKFd@mby115y
ziuPq=)OEo?1ZW6rK5?Qm!!{!cY0ZD3*ny(^^bKJ;V@1dOl9Q-Z0TGs~P=`G-G<26j
zCCa`Yj_9FaP@MBQrx<^_=hvVy`DFRXRn^USWTQ+`c6r8VLnF%qR&{a2C(`MpS_lKO
ztTIO10a|L%!_}|SMbeLB;T2+v5Bs2x1!HI5*KJo!gkMqmPvLc20`OaEt-q(lIzEBt
zyFP*c)#-ZcGE-+8>6)aj;Zq^Nd(&y}rHV-3#K=t1{po?OTRrNBR5|S)UziEbDQfp}
zVY+dhw2?9kRI#OV?OnK$j=>(GuM%Z{MXyo0y00U~dp2~THN+)Rd4gU>auf=QQ4#d2
zW&Oxr-D7{l_S>OlIx~=XNcd07anBDH8Z^|PwUWO9$OQm!l)F_?eJ=u%vO>(6);w6=
zW@dirKA<0t&NDi@S0M@yHO-MN>(y#ZU>L(8oP+y`iMQ&rDbj^~0coK>tpns#Dp*Nq
z;hx~c+gaF|ua5(Zp5`$-eQ>oNF&ny;Y8NY(D=|*Gp>;m7+_iNf1xjgDm&Q5i!eabw
zq%?m{7}<#~Kuw?NN-cOr0;loB2n_mP-1c>-lBZ$M4bNpyD9_z6a9$U!>(6jd{yKJ^
zoy&_}Hf)N|TN*5Uz#S1!*Ut~-2K&q9_$40<*mD^zLN*b5DXu|Rt$Kq?9o}tcxN!1R
zb^&V~C`tqTw9oKNzlWQ@*zmqOg=So3qkzq1QEBnZW<A+(Mmo;GqK7xz*xbzQ9Tyar
zn{3|^Q4#Y0PKD!O6kI_K@8NA`_pr&mtjOXlPhUGuR;RCn%xKitzOCmW7(bbt9!yPz
zj@l>Biz1H5Vy=tZ;okoH84sLi3ucopR>)#FvAhVD&mnWblHy@ilD2PdYK+Umoc4lC
z_c}-34OBD@`he<!MnOH|KE<r+XBg2*iI(OP^_RuuL|qDsGqIcIIko|r9YE<i(vXAn
zbr7jT6AElJbdBF%`Hn8_vt1=G9t(dq5fyc7ZWaXMmo)!Nwvr%`dH#)uohA4;a5<>c
zoF&#Yn453P%L#Os`1d?q>stHK=|_@aNIoh4NZL9HZa`fpHfQbR7tRq740kn1yb#ee
zzNm0cvJ9+<7i;Q>-a~Y1?b?QyY^XdZp0RcD>@#qR3)3ir+c@^Ctw5Cg4jb)T!MTGK
zlz`J>US8(z7%O`8ln5PdxX+Yar8n_9Svn}L&Z0h?b8_qHu0fZj6tfSzZ>AN4VuEQV
zlLAVxd9AuaSN9sygB3+aG?I&6lxCm>)A9F?J+eXlm$DAAh9W0Ag*vu3gd~&vte?u=
z!Y2gXSHicy@nan;yL|neAZ2^i1FApUEA@A)Ia<xO;XK5eH01)aiy<5L#o@O_;ID%G
zw{kj>O`2n^2h6m|h1(t{{6qJu<+LM7v?H|xZ+(R|0Vu}c{@9&i1c=d|wP;M#III{b
z=(Yl1d3t$EO%dp7SQ!Nk8|+Y=?}OX)vHfjxs|tVD0=UqyaaK2=JfEFnHekZ4l}DmE
z9b_TPo>o+-HcT1EIf+5|iq&D}3Lt^ncH&ti;R`oIexmFT4ik69NRGf{jTptwAHt2;
z`r_{yv<EtJ+f7)XhU$E|Yf|TR?)DZC(-U?%poNzO|7MS3J1*?Uq90|&Q#&rVh3_1n
z<<6g?RYLp8btkSK0vip-FqxL{t&-vda@_%|Y3ON~{OdRk=*1zcpnw>5D&zCNNma6Q
zO_n%QRzs~-&;n07yi4|%Iz3+##!|U~FXu#MpTOWJ2D}TEkjo|RsWjKICktREv=$nT
zk|X^uL!ww25u<%?mf<5IMf&8TigmL_^EraCqZY}yL`d4rvE9l@{Iso;gU8<i2JG&#
z?vAX+Molb&myY{U^y_<~tDf#LA^DW6H{0O$#1l;Cj~xpjJe}1D^l)7RgNxJctC8`U
z9<pK^!QFD{bbPM|+4eE}K2p|t!yXIzcwEy%YEnqM%yz8b3N#`)vwVXzhcg8?$?*!J
zIbAw~V>LMz1;dZ31~|QxFr#V~vMI7%y!Me1jvnrMDejvBpoX3J{XCB?{Gqn{rIsCg
zrm$JtRvihMpz)%G;<qE!S>kB?_@-=Eb|i@~6uAJWBx4-Cg(Pv6qgKE2o5$sFj4n1u
zA?rdyaW3ZtQ;2@D-;9Ph(qsFtC3)S#52g?RI57f#a%16tS~kQNIE5JRt4-6%5^kJ$
z!<c^etstx0=xE@Qlk4CPm|DH&M|CrWV?tVlw59;WJ``=CDr4)5k}j=UwkPpTu2SH4
zcyNHLR4$@O<B@`*<8)|G5oTJRUaU45fry)8=Kj!1z2j@932saQ#ou@1sBMa52?s;)
z3O<-HwKV7+0})<Vd4NQ1I^8CHj1Gy}=J;N00r(@KM-~=+1@N{?Mgkp0cPRAS(sRJK
zguZ7#0bx%_Mqv%8ybnNnlfa~fqKw@RMZ}kB#1|XSc3b_Iy)CvgS5KgMECncvP;zwP
zTTkux$s~U%7k5)kY!2P5AZ*snzI_RSu0gTr^1S!3wn5lc#F-Mpuj{1k1;GSstwzmU
z(R2wef?My-HpyE36LvPoDL^X{uo88M)p%E>G&1)E`xUOVO@vW>^B!M2-a+k8HEPlB
zx_-~vrbluc1t^WxNS1~U;`&bEh(=>^1P{+4@}{^Q=)eT<kwyI`r*8mjSB^qQOZQ`A
z89Iv0T`~B~mQi)lvF9Ur(NX*Zm-QdZl)YxH`3@l((}`K#O7=(K{BJ!~w>3=q?A7KG
z`1nJaNrL1~*GdR)uN{MlcW*Bb-j*t`=8v*ZRK)GMRDY3lDPM4wN`=tB_<Az>e5+Rd
zqehI3E)NK^=17(b2hLfM@iyX42W%67ek>7eUQ@lmCd3{#NZN$uc9}5S9c$n{uUlqQ
zspr%J!SxW<rB-eeSrjg$T2Nd{tP{bVB);h?+SAdKpS9w1-cwnNEN#)F)P5CP;*{%4
z+JcuH)Z;T=dW~%0&KwAZhWd^RhMaJ^w$<nuOJkFQ<lT~=jk>EpjYf9I9{UFh2@kxO
zCm;`Z7156uN-2XFMYr9X-plmva&UX<mR!&@lQwI`I8<u2Q-N!cS1UtX6jFsMY@&NK
zT!;Q}!Xcl#ooZSHJNPpcw&YLIoI11-vrFm8&XVwVm9ja|<<cQGNeC3(IQL!~Kguy!
zCiebBOV(-wt=DHk$Xb#9Qsk(IFbb-)wVlW1^|dOd68>ERyhI8)`(UfM!G|iKXheDS
zFey11xHkE^c4pGEDLY^Hm{X3a1#bg(L=!zWidSjIfLNK1GqY;Tqeam$7mRs<BfxW^
zs9|6vIB@Km$lx5qam+FYT*_vthf|P*eCTm)U3PVJ6=Scv8>6!Nv?Ly)$CJbZ{B7X~
z7|jWeqcWMxd3<e5^q!tynu5CjwK))kId@ZoP9Elgj8go7WuoKP9J|KmP4{!8g`(E;
zo&95ulfGz-T%p)_;#4Cq3Ha#5%;w`%#}A{dud4+^01p)Mio)0fJaY>iHYY#6T0Z;L
zl&!WOnAWfuEs)xf4@QVy8==R-S<cy9KPu<iDn@+dGy6jic|>mv#a1b7x`?#F0*j9w
zE&$|h-j^p%k&A~A$W!8Jv4S?WhN9nCiHC$xBF-GHcM5r+O%25C3kG3(IQuLzV>1G|
zv`oXwAt9H5%5i&Qs?)Vf>72TJgeK*IDL!Tvme#<flOJubhlQde_?!E&Bi3INW6wh!
zr*K3=ZpLJ5oZ}kn97S}es-=qy4$bdi5sD8iQ0I!_cY2=bW;U<6=Imq!`L6AGP5#pL
zJO_$SvT9{7HpJWvJcoz3d1-OQ1JuVmU0qxVp-az|g0S5(o4MdKjeWO@)BJ`$Xhv+b
z!}s+<u)cLp<{hX!`3<Esvr49Ikn*FVP8^9ft@?!P9pB9acW+|9?t%LI?n5{iUw2Yn
zjQv#c<G%G+)aW-8nawN;@Q5ZtNq^!AM)3jXa+;4~Rw>QDnPo(C^gSS4;Fb12RW`NV
z<94xH#-?FW6~RfxR1a82NRWe4niA#ZJ8#l}TjbqT5uBid($)0<`RPkOB24;dsK7{!
zbYXab5mNeq`mTWuNj&79Q0^LF0UTe8y3HM^6U!!>q;X8c6e^>`ivgrDn}6OlaOy$6
ze=h*+zJbs7tfx)H+?AKWX^RjV?)5hdD8yG9uPL;#q#&p}w=*60$#Iqi`;18re?Ox3
z%e=Kb46@NUW^OaJGD=5(GdMU*WaSVuY*O&}D+eiEk;ULpK2}QUI1q=Rq@PT;U+)sC
zYB^KjfZy`rom~*J4A*naj<Z_K@I8GbRP3;Yw20M}G`2C#-6TLol7ZB?uVx|ndFI^H
zYOaX7%L^q#D0I;~UmK`*Q0=-;vDW(ZZ>iTJkfW3TwLR&YfRa6*Tp=lWDfHFcNIZ$B
zUKF^WRZZ>pY|sWz=trZ>r!m9)s}urQFks{xfDu6mQNtOiY^sm~N<;jkXOD`qhdAkq
zD_TWksFKTB%dvuDG8EbQj*jZD_?E>)t?1A+tFHHjo)Ly+aUF?KwGvJm2kgPE`I!}y
zSai$UX`u5Ze&~`fKJLfSVzh-F#4hIO@`2l1(6Ph&<m$liW)jSHsj+w6sFKw`;dQK%
zGwCM&<?$=N_B`sK!4Ss9c5G?B(^PMI2`Y=`-|nTJGrG+s$*?46A1Vl)H_HVSDCfKq
zxQ+xExTFXSCp!O#KyAHLyTny#8TUx+tm;w)OMN)BrtHMb1c$YEJaZ4D9NgU7yU}k%
z^=~(lMB@Uc!inLW)SbmyAAmfBKe5;i`;s;4bC&KXe^>rd>Fzf%xKY~O*z4D_27iLg
za#aI2iMZnkyQ?2_<*tMiJm=!=!UKqq(m<w?n%+0oO&2KVI0%qJfR)j)Sp*BC!*<&R
zFw5wFjLD-LJj?jBi8MuRUcwiQvQFm(h!bh?n^^Q2(&&4L<7D7faaCg&K8G*C8SY`l
z*kI8LIxvPg<{DroIA+&CV~?MU)?p15_0Ceb9-CZMu#?Zx97l_`oEIrFOx!WHJIX8a
z(E~@3uUB5Hifoak{0$t6KW1O{xs0mv`_y#z+z7@@dYg#_bD5liCZ;^{T+C5*^a{7W
z^_Xnk{15qQRJMV@Z`ZV0|2~#1Exy>()EtYj-ShKv7sy?-delk^bw}WQ=^&A_e5YR4
z^fBxEmXs2nJ6$7s7GntAzSn_m&>TuP0{n-K>R0WA5?Ldh0kop>K;CD!Fh^M&ZGP1l
zqqLmR^IVF+$$6MFmkwFRvm_E{()MPw9GWtY3^<}BuatlZ1%)05HWEJcZ*G~nJ=2zr
z50Z&BTew((8IU-i^jI;O04VLRagDqp@Sm;E-49)_9Cfhc%P7^59`BFNx^C3tKly@_
zH-Qj+P3R_h8;lZ(9qZWl=t=nE4$SRK_8fQp6U=7Gu#o-Ebasy;8~=(4-rJS0j+{+~
zLN}g$cQ`VBW~+qsr4BlPRRZ6JyhU@<r6kHB>E#6>#o)<jRn6cm+IFt&^2SD4z*-b)
zT~v27Rwd~6PIa>rqSi}XyR(ynvLa9x3pyg}dS5r!-%nT^$#1m!aYt&!Tu7f`7^sQO
zKA9V}Pz>GlT%r9Yo8S+Y-p@{Hw!A7{s^J(Bpjl=yhKA4rjw^;p{nzD^0H}ukfC>XD
zn~+0*(bK@mm?+F`v4(v9G^B82zD`b*cP%0TsTQ=M*uF{xX$OgjzTo>wA;2!6S^7b|
zevd*WFjm}F#X$jW1$Gn0FxeF$PNDDPTFshVNh0-HU3ar?<S+SSCv^mN1#(5cmGNU-
zW3vFSRz9ITIOh6dU0j(k&E|GI+J(x9v2Cq;#i&mO%g3kA4(o*{Yu4<FHUKRH3A44k
zy6peFmt}O6yCIvcpu0`Vv^VQ(!@*wQru_$rh<4bPhfiHbPzlb?g}YX8m9V)*14y3$
zKx(w%o6tDVCYImfw+QfWAjPsEg`$`<@gwY#PsrrV;uG!><_}X!T;nP);TyOqf;I^m
zQ?9&doTEkC131Oh`_`Qfq@`Y=u~UOCns5`j$ud5`oJ+g70KBrIhl}4f*_kfcvjlc9
z-^>NNJU|-?)WoY3AL9q%0mV2%O-+;zo8TGkiG>vjdHqEn%mzi94)5<HGX?u>djYiX
zw@=$=LfpH4(genERIwUFRX!sTvfo6v@r{A+nx+j15uJ=5`~+7H@Lk5=mu3=LXjyKS
zP=$XwFCa9T63y&#!(a+p+%{QknCWtEhmFpPTjb$SCwef*`B`HdUp-cyX~gy(LT8s~
z1Exql)q#0tKjdF{yod@OD5gTpadj@=4~<m^&fp%#wr1fgr>}e}aj!7TvuUs5WgLTH
zA4$L}@?tS`hqm$Eyx7;q&SARxBlpgR`swdYrC1eAI9n=OjnbX4G+jw>SSci~Db5kS
zP*_3ElSSfPG~zOom6-|G=R4EaR2R9+i*H_tLdw60!b7W@vJ8Hf>4mS2#@HQi2j&Jp
zx9)jhZ4Vh0Tj!Fz1TvyIv%<D-9A9X0pn{`an7+G&BwplW3tTrkxBqfMVjAVEY3b`)
zJpY_(f&oyaHci(=00lzG2O>>BQ9JzRU-?l+l{as(I2AvAU64wgr*s|prTuPnRoDi?
z54zh|Pt(@0LQI&6<G7rMfRah$6m=!k8dhW@SEV8|(W=Dq%%^$6bHxv*BmfN6BqV7A
zA$f0;(bN9Cu}_+Y8*NG&=TJu`?2RBQwt^mMvb}|5865WM>gKsBt}*lyN7J&zme8iY
z&W(VS%*JLm&=9f%cTCEeKonZ;yHxv40H|j9cw%vwH}dlLZ-FUu6jVaE)1O-atQ|Hf
zdk_VF=^~_Uo7CVt$d13O6v@f>;2Pxp<vJQj@4AEyZ6Z}^=xy@CLVb$L$rBGL+u5rj
zg~#=MY)>{JxRb>L(hujD2qjZ@umKL+r)d)?aSWVlJji!Y@-`^a7G=B3IO;wL4weO6
za)f}!W7l8({3(z`tG3)?``o6@UK5XSfV*UBR+)?pIr9OoFsxEXOM6uiVE`}8R+$1O
z4Vp}h$hSeEb|ycCNoTXRjR=uKMU{m1z)UfV-w2Z=ajYaMdxW#e!r~fE^nOfjX!uGj
zl<s`!9(@FH^kPUuSitDwC+CGi6W+O|o0nKBn5M>sA)&aWs#7=X%Jo`XEREf{<Q2cW
z&=n<JnKzzcey1pb2T+~;J0g43|B_y6t)$ZwZW>kp*RFnguZyu%eZ2q}!ru_a*AZDG
zhn&Zvt%KC}JkUsS%;=a`<44Pu83F1o6U5r~gF_WF7xqzu9T6-QPqrgJaZf&|_${t^
zT1KfEkZD#gd`+5qMHz`_62@o3s$9NVw_mX|qd~hjI6|<O>8LQwUqY@>zH6D7a>>I@
zE%-voyBU3*%vUNM-hqpvUqpqM?hX1hdmI^{R8<+jzl`aGQeE9zR8s+N8;NZCOkRQ)
z_dqsX86j$ed+`>w^SCZ~btz|BtYplu9gEllf$Qsf#kkgTbJ@>2N^<omKE2`LQTslo
z5^`s7@S4z(hICQYOfPQ_df563x!?gBUF_f~cf71>$NY+fZ^mNKD~N#l4kBIpW_j?B
zdZuFw`Iy^x=@5^5!p>rJ&r_Pe8(O!{5wKUMx^dnS7JCdskMbCax!4;r2y$58ECm50
zO5gT?RARem9*gXgo2Rwy^?gbSf1K)Zm*A<rOeNI~mlmO0d*^qKyM-b3LngQ9TGAN-
zZdm=&10Z8-?o@lZ0m|Zis|>NPH}ea%AFy#hLkTcoTYpp|S`|CIcyYx)W&&o=ull-w
zPx~Gr{=y~dQ*kcmkx8z2PftNzLZpR3;ou+HK>{8H3f5FD220q`Fj)iYQMyaYexJBK
z+wf!zPC%)(+#;ZwR%rX2(Mhq5MxZ@a97X#&L5k`7o_z82Me4Zi-)5J{Q@@qQqHop(
zlN?#=eak-L@*NUxy`g_&b&YVMqvIrcTi?xkp#bbEi5zcgTc*pD!E7!4W|(sRp+mL(
zc-D0q-<(ukN7v{<t}VuSf7xI2cSpsBn%1;&NOe5qtxy{~j9Z!)6+Se}WdmKRE7Zh<
zdyf-RIc*&_sm&=pGh*cebTx6-=EG-uLg*0o^}N0Ondeg5DJT5e7yhkB;|9Em7bhCd
zT+(&JboMC6qVk#5r5<W!;ic-G5L#TtJcpkB5suvnji_Asrc2GPmaft9Y#<=_c6)TJ
zNY(KZIYbnmsoKDin96qAKoB}fZ;>ouh1Y07OQjj@(fbJ)C;~EUMnNglWIgf8#YA^}
z2l1nC{Tj96WDV(2;Zsa96*C3Q@z;_IjF9R3S3w>ddk0bB+8tC!d?ho3OGyjM{DCke
zzENpBGY!<cAE(t?0}ZgZo#+x~xULa+^M2d?sjwQ0>ZzHwyDY1-C@6hYeg5oBTKeTx
z=;^8X+GZ)^iITOV2J2%p4_sNvk#m~hqK6$&z&L9(flesn7Ut+b0^CaDdFY@2dv{DP
zO=#bs<2MYOkcd?~{3t7fpONWh>h@C5TdEFIdTS6C1&mb?AD8f?H@fJuE5Z>cUcO#I
zWyA0`9eN{t@D?tiM~p+s?5v*Eu)6bGgS6Yy)f+Whm5pQn#^zvgUbnfunP?CCp2dlc
z#{WsC&C`JN4Y=CU+r60uPvE?qo_)UM-*G?aJPJMCFF&8R29_$+CjLn2%yYZBt({3I
zq#Cgmb)+(5XNoMEqOM6)QY)+YBr;2{-Sso5w|Tp3ol8ls%M6K0M)n&F1NI;hiv+zz
zhT{7mSh}|#76(rSm~2A25lQh+KeWA8%dXJ35?XQ#RH@8NlxSRK_Ej_vaF6aJTz$lE
zZO&)W3INz=7j(~+4iR)jbd4UoI_nguH`w)LFy3$-nMou|YZsyR8G0FehN)tC87MdL
zfC`9Jy|j^Vohm67NKvzpT_{u^X~eAuPI}qJZ7S9U1ccTs($)#j>=hIBk}R?Bi4w<)
zk$*{_r39snL3iV9%|E#NuqNXt0P-f1WCmO|>=DHu-{DUr;7P;Wh$8`Jwkot4IbN+m
zMSCTb?gVUzm&R2=1|@>XUd{LRR;~4oG>)@vZ=m@o$4d@YiZ{x>f(1O)JT90kowL$x
zxu~a=<=)aZAOwzTUVT&w7z`54u}S^*I@10?wE>ekOsd^85PDk1_sqyYyq(Ve!xT)h
zc8R)%*cj;vtgB`1&e5Ikb6L#zJqGYj(kMeDiF5Xan@&v-Y#I-o<pB>>Q2I6ARVE;4
zG8VGyfw-cg>s?(9t-p`oSkU^o(L)0;$eyo!V2x)U+)s^5DQcu7oC(0#h|3BccXS6%
z`HK-#90IDukw_Xpg(|Cs;=CI!3m4FP@;Zp`>3cXvY=|(?NP0^7uXKGYy=7$5X`&&U
z*z#9C1S;CiioKHZ`ekyQ@0QZ6y<}ll+F7Ja=j>uluj-OMw%te?os|~u7GFJAZeYYV
za*WS-naeqVCrdf7eN6_6Da!;RAZ}f<_=@sPq(pA(jTnUJtZhZ%W+gFEBW4!X$wEb@
z7H7oI&L~g_$kr?+MoP`Z*<o;6GHxkw=&v#Nk*7uQbytRFy*hOI$?X;n!NTP(C7UNC
zB3>1}JIaxo*L8T>_OmkjC?Nmy0cZG;w$N}@t|1>2nj<qb9x>nn?WZcBw9g%n1K_Gc
z3XGsguxHaT9$(%6JZp2?`?_;?7FVfwtNCz*Uvoz_z+;n8UEiviR1R-nThX^f8<Ex(
z{429#sU52S5Qq5U5%2+AcbvOPw9;M!KYhKeUOYhf&ixcE-~py<g0WKtk{p3NujpSg
zi#9n5@6ZmQ9KUyDmF)v*DQ$t^&GyU;o7}|Yd$Z8L<v@R_Ds-S#<byfHIOnY@C;!Mz
z+InA#?&c1)Uf2`)e+U~$k$v1Vg$WmgPSkZ$h%2@yO9u9OP`?dSiyOp!zFf#%{I^mA
zXu-W0{@~zwNVlTT$l2cGs<eY_nfXP`(xyTNXn%&k2n$3<xx?e?y$m4n-$Ec^^h9=O
z^enP-xfJnPSrvS4|6pNV3@YVRH4#WF|Bllbm#&tjTOUD-7=NYj)7A-J{~-?{J1M0=
zcYvJ~(~&?)y((R!|Eoda?=SfxnugtJW}vB-e>LS}jR{e)tluv283RHdzp-+ZJ!2vA
z!_{=#_sH3vAce@f2Xu|&_7Lmu##Z)`KjwH{O&3A6kA0^XA}M5_zj~OZM&Ut<5B9e&
z7j9F2|D2Lz^&exWf{tp1=2eNp+A3VgYqt@uKOp+Xs9MBadgg+zJ9Scqi}9C!CXoq<
zzU6SAsjPw0Ms&fK0`NatB{KLSd`L3TbZtt`FVAsvKIyIK2WOxwMv<}hcm4Wo8KD!b
za=5|&DH(A`hW^?Lyj^&UFye_vtCy&c@&Bb2_(RoF7Nr)AJogr1pr-W+(*F#o^PaR^
zQrxcO#L`oxiw9H(N-c_6x9({_=7<2X0iX`Z1`=UM$7QBGDNDaTy@mL}0s4m|$A?}l
z<A-9tVWDW>VM1~z0o!1Xbn8MUc$&Cc(~08}S426mkVEq22R>Ek0O7*?X`);mfLYj!
za_1j{pJhHs%)?_XQi!w1e@^bd2JQSJG&(><{_i0DfBEWvP1FA__xZ2y{H-?n$1eFF
z>YM*f{`8kz@I@096o<wE`yzaj`^TdGpN#-WaL-@|_J)B8;=cuXfPO3y6`E@G*5Sgh
zN)(v0HA=F54flf$>A#g_f8<F%h2p|?IdaCe_?$99*p!Ju|3f?VXEZe!=%8+{EKvGC
z#s=(2XkZnLAv3gGb6+a$|D*B$YoPcqH0vBO_P=ZNKPXko1)6S8v;x(Tp!`SSZCX8=
zdA_RkTNj=&F87=SlQ&5i3YB!rTvkSuG7GsSWG89xB~}MY03mFDxN=@@Ga7d3np9q?
z9}9-|NfHyjni*sHSsgg?TlmDvCeS=|P^EC%^b3nZlI2zP#$RHmN1pUMdA%6cI8&p0
ze$Ry$aZHkGbmO?z=zl7iz<(*3hRpF~8!5>YnV5ggZ>4Qx{jv~lWnYLPH&mE#b_$()
zx~iS-<>j9GOKK!~0*I4dO)5GA_Z4_bm$ryr_Pl;RpX)aP4uWC}r_6XS>Q6Vlzr}h5
z35NgF8I6{s9UnZ)&HW~Nq3z@waZ1Pga^F^6KepZ$zEE|&&*Ek6Jy72QUNbYXOo*MU
zp|;?@V#Xd+Ddslian36a%}(T$@0|^ryFt6vVBwZTs^<G9qxwBLJ>QZ=(r-2DPC;i+
zpsHZ1DYrf2nY`8GyO6oZK9<3V2<p?V!lyIP);`B|S=9%0DJY0JJx7(;vP?6e@@!vQ
zXE_&?yEg|-4qCgE%T&p{9QN?6b8(a4o-sLP_`afSr~Vqo%b-#ud=q7q1cSoZ7wR%q
z?@=ij8+`{5L#W~r0&1^-v^3Y_u!E{*mb?1?7i};qcT8wFr)Ej{iBM}~i}n278%;Mr
zF&Q|+@eLPH)^WeUaW8jv;Ev@R;}phV!R@Vp)zMSmdpNM3XYRJa-4X#%DT-HHu;7vN
zGn-o!csr$wRgjxoLCa0Z7RK%*JTm)f<7!DEE;%_LFe5stz0{x}8bPjuj8k^=j~WHa
z9>#t@z8e?a4+i?ZBlf&~G;%*d_UQhjI9kJ^f1_to@H4m2*P9yJz{gSf?^=LQULH_y
zc3+zp%I0N#7p)ZE*NWH{|64fKLLen@ncm_uxWPjV;|GKOHR`@mwxO9XQhB&54FDM3
z#mUMdkb@oVQfOvb)J?C<A_%;XddhfaaoPz+fp@7yf|xoN?*=7J>q<?0ypn2fF0ZWV
zUUsoN$w#Nh)?tNUjo;p`tyI#}%$y%*Ffs=caoMQX4yCN6Ej8;87f6mICW{sfi!eYr
zhQsNHpI-qhGKK^A2modLJg>Iwh~#DKQ`L$bHFX!WbZf=(CA}nz**=@qHFoY9%pNOH
zX?ts<8h+JoGXkXOu_T07(MiLv@roP))*5V9^79bvW?rksL4*vi%TQ6pq;AgPPBK2W
zz5Mx{(xh&RtnH`$x%jBq-q-o)R`D&1PkkGJ&UFhT>hZD^2>{YtA^a*d4Q);82OlEO
z+wg0b=nZ(`(+FP4BxJ@V1Lrrg6i&HgOANS!Rac^6I%&ke*2V1O8<gk2Q$RLc1G*l1
zNI{>X*tFz}$Ei3eTd%xuJ{za>K|H$i2f4*Bk{$ElFlurT4ci=3GYD`CWG`+ZBxvg#
zX8!lBkI`Ea_ou@tbJ%#e3SrYFy@}z?LYIKI+8{t?%|9A%IPE!@(^CKs2y;x0w%vR&
z^?8WcP7$-Fn0ew>ZQ#mTqUS_~xArYB6ROny<$=7%pu=ck{UC@vFgk;(Y|^S-H_JHY
zjlKm)XzDI(c?s}7g!Z#zh`a)#2#WPKGzdbfpm4q%aMp`XilZs75XgxJHUkMjaz*hf
zMV+jkJ(|k~D?UNjLD?VG_mn{5ZCC3@*)3ZFlsvS_JJ@k1`j9z}F^}a2b?&c)DqjtA
z^-?@@uZ-qhVwbM(H>||p=ow@G%TK(5aL(=IC%zqtpi;?44#9(dH>A@Kij<k9loj@D
zGY6DX-l^uhv6fL}TzeH4yvuT&fR^=(Iw0qsb$SVN0z$|CshR2!`=gnHXHhVt4u8+X
z?vG<1YWBnB@%HDs7f&2Y$u_RJ>_Hy<o7KXZOyG5Hg<R`*dRgJbv`;?^i9BxoVEt&9
z*KS3lm-Xn>Ouu}IxDE3>4(2<gF(Sf#Ez+8jLi99jl|G7BsHR6#7m=;^3-3`kL322n
z{o1x`WC)K@g@w5S2bmUP^fNEQ%?g5*X(f0p!y5PfEAiJYxf<FT^|!nOpMDUL67mSu
z90@GWI0-$5cK%En*u2G<*^JLI5-UghM2=|$DT_+XO@!VvE4I(|3x<@u2v9Te_~T$Y
zt<M2kB{B8P@-$xeai~RI9sj_r&TAD56dDS2zAm%gD1c&h<d2;6G_?&XF~jlrP=Rq&
zuA-SCQz{<-+2a^d>HcpaR+{v<s%R5<i+TGuP!yYAe+#Mgyr`YZMxb3>ewH}=*`6rx
z<ZjLR)8h{%xI0rb<5(=#EHpGgc{FEbDrsaK>&5?c2X~2~po`@U*+F@7beD4Ug|$$l
z=tO(;oM+h9Wyqc?8wFD}vN25h$E{gPx%9MR4_WN0M+T@pg|#V+H49Lv_4chBh0dc5
zJK5?*{tRpZ9{;Nr*j+MpBL+b)4o?wq9+%O9FKuiCWQ-4(&jljy9!ahTn|qzX30j4q
zY*78sKF?D+T_?G1!;W(Lqm&2ud$|VIiZs}>khxT`P$koHdQLqW3eq0K3L&P}1kX8j
z)r9oBgOk0yy}irT{CV<uk*@QlDj;EN_slqyoz-IB=BJOH1kO{*ZrS0L@eW=eegtMc
z-f5Cv3GOVMPn5TGSIkV~+WtShy=6d@Th~6W2+|=SAf3{w64E_@k`huwh^WMn5<^Ir
zGzdz|ASlRyNOzZnG!jF1ck|zPp7XrtJkNRG5C0FpFFK5P?|bjP_FC6kdtGbk=uK_=
z@8el^+nz@EM7`9nME*AKL<iW&T1=KCwTFw_#6OTLGXn9P56L`NoM>~LUDlUxg*lsU
zo_ae{P1C$tEF=fap)@L6o*p)6TYulq-JszHO5gKRXK(Yf2@TvDw~I*05I%K2y<HN|
z{4e|cxoZa(Jykq29!}2oYeq&ts`<k!HJ{+2hEyvLlZDso#M3AB?s?tkT6{hrQd@mV
zGVicEP*t0egq4bF8}xB}UV1KN$qIctMyEq`$h?LvTnkO5+4-gY6$wh2FeCh3S6;gr
zn!#{MsNhT+dWnY~ge~G{qav5ZV~<}=8lSG*pK`7(55uDA{^i3r+h*!R{cVY9+<PpN
zFi&fV>aKyBGJ;uxFzEWmE*y}xS2|Z))NVx1`bm0rwyu-R-aI00a&0YByiB9<Ug#ZY
zJI643tAEJ<+Fq!sI(0L4IJ*i3B1GlwMrruO%(gTukJfCp2A=IIoeU1qP@vvCz^q6;
z_N|~J>T%p-9THtTFL8P=gJ4dH|MvyvHl{L734R7O`)m8xbvQQ+Vw|-4(psnSlT6(*
z7DlUb!<gJ(L~yLx7!ydD0fM!M>uxOYRuWh-2i7BCy{-zZn(1rp&m6vgtLe&!8=rB$
zGc5F^R!~aG$=Qoy<#9pTcZDz&P9G!u0P+})6@ka0Pv1rk?u$#@KUO;ZDRMNMzD}0K
z)le&I`lb_&r^!}YBC06U+v7I1dfYTU=DcZ?6a(F-rlJRjABQp>?oNwj9YyK~aS%FO
z4`r2<j_C$;D|7!WM(K3BfQD;Lsl?Qw_8PfEii>Hd4KFeYs`^2kz<~rct~a^mj~J}7
zkZDtkVY{CFvgVhbth9PQY)wnTr2B^A82r61s>uj67wo$_K@={wr{dXhUtEO9AGzn*
zn2IOUapps;4hbt!=j40sKEC4HzQa5AjB;JNhF|h|BVXE>qhhyOH`}nY(-`&>8q!Uj
z&3jHqvuP(#hE;*C+l*=&x)RW-hYoEPb4<dy!X3R%hAOeC)z<6bf-UdE&mXi?hQbQu
z5%h9Iq~gYw5&?tz^Xybq2!@wL?DmUQ{YW@3V)<xz=ErLn#|7!Cnp^B6x*1D9v)o)B
z&j&OY?oyC;3iBcu`lS%eqf|DN@i8+=POwKOUShCVNXWfK-J#b<`KuAkw+?JL=s1Gk
zjIHh#%(T>UGGYW?>%BFQP_1<L?6tjDFnU<3^VkR#=b-$nydz{r8PX`#D=Q-Cqamv%
z$kBdTSlDssP&(yt4e1fDC5XiaWn^J?@DJZDu*Y;K{jP*yUXEo(y|KI<?aMIiy%I4o
z4TJGuwGTP_6wYrCOO7sSCLk8?9+oz-NbkAN44L{tlzt&s#^d)0s6T04Gv<R-*)QkL
zL~sMl$tZNZc;yz|V9-4C`EVB$zL`&Y-BkFFy{Ht_H#bP+uSNb8dZ5=nJG0+gW!mXA
zpwFc{Dsaaa|JLLKFC5}HrJ4Mi@89ZaF^Tlr8Y=K{plop5W&v=u(~4`#mmZs0D^@MD
zcW7RZ<P{0h(Ce6ab0S7vz{1Vth^a1EnMWB9Ix>oyT#I}+ZPG@rKuttWM#Y;-+LOn%
zQ)G?KfZplaV$;iND7O7reMi>Eh)Y-4TQ%(Ksqd}{Gwy9tZ|n5+GJkTF2xk_)my;6(
z4#rQ_>-vI(^wPwnD!v7`4tu0>M{gZI;%OZsq(G|D6(5<bvw(1kdd(COUjbV{1L3>J
z6jWO|ZxV30c!0eSES3xPT6!pXE|b#gii3HMhJ$-o58ag7jK@45%@M0*qn%;sO(|{y
zCF+PKRN+S<psMf|EB7YQKao1iInLp{Z!_TIGRmyE^_2p%gObKdE<^p9ig1>CqsM-S
z;;!Iia``ts=<`mF;1I=7c$KQd)yYVBCzMmMl-F3W=V?8mdAS@?dPmP?RuhATGT1pu
z;kqp~z;{Z+4rB1l#Rjrzy1}xB<Sz1#Dq6s7?$#Hk*rLTyan%bpACCylqzHI0YQCBp
zuNdJnT-2XEkKrp2Ivj8X&*{R8*KBGCj=0{aJP<Q4xLFva<T+Y+!D=>pjTcFwhp=5g
z+nSl)`*F&6@cc)WW>$v+=?Jd|!Hhh@BoZY>Gs;CGQNXt0OnW>w0{@Hy)3H=5{JKWD
znSXSe`?Z?CfoO;O73|6aJ!ebWw>;rf$T*wx8my|_kL=?Q80KrS+{6xn((K^V8W!XS
z$`w<FhdIXd8Y?Y`iD{I)<{wVpY_7O&b+|ZvZ+bkiTI0^Aw?TAQHE~Pc2sb;9K>1qf
z6JhFc4JsBEHc36jrZ;iz`S;+e?9r2*Ow9|EiF5gJq^$_JdEX&ioKEPw?zjjtZoI6A
z*6SjYj+ejh$v9Nd#e2@X$OiH;IBsz|YO|c4FG(n20KT&GA)r|0&Ud#vQyGszA-yVd
zn_W@C3Mm7-TB(6+9nWZZ?~FH!w9ZC}t4?L5hp8fRVD>|)qC+%O5+SE8Ry3I2f%@gG
z`;bC=+7gkW*@$g~XUrZIX|%6<dTUwb!`Y(AzLka5bDR^ON>Ea7mPwMYiC15NI3@nn
zU?zs%UEF?kN(9Y?f2o5HRw5z)&|qXqw;1zo`Hh5*2l^9s-8$elN56EqYQv<aVT|S$
zZ93tL(v1L!(!=)>b-MsgzW#W{nGzX#+C78K!b~cme%CO$ru-e!dTl2VEdWi1fPI-T
z)j0Mi)oXc#S;x9CL9{P9xVrzQISz}h7n0eYX0_+VRQJfMwU+V99~vnWip~*DvcUM!
zHMp#IZ|`t6J;;wbZpgOb1{M9}a!HO)v2`uV60KFG48KxxYdT}FB5QTmp3{Zxn_x+b
zo#n`Qw`$Edhc@mPoDB&KVmeXGT0d$AshzE->Tnq`8g$h2d$H@TVFrP_`-G_2M#)yg
z;(Zl@p}3;Vo`Kb3MF|&A$Fzx72H<~E;ZA&_`(gSOZ<V~K=6#PS4l0f-is`XjWJ7Ke
z+~3ZSNK$3Bz(uM&Z2KA-cBp#PjSY2|llgiY0%=xhg3IH0lczVODP(RRs5aXxUT)C;
zFuZ#&rl(nf#x?sfV>9gS3i-9#2aVw(Rc%w)*WJ7sSuQ+^ivu}iud^-pL`uR5l=A9F
z9|`WEQ46M?dzBUj-Zko+c;~2e*ixpKvp68}#uML>_!%Z?rfq$qi-DV2EvGh_xUAHd
z#rTF{m}g&z39>jyiI~l{NaQY8p<ubZu3SUd&j~Lgb}XTsCaV;~v9qQ*(KhM>vpp=r
zKE2WO0DV7R-<(DrVc&F#w}9_4LKPdqE|%7E@xck-*_*suc~umpwP9h+fcJGQ9n-4s
zU_<RupY-UB$Y;ryPW4q!Q6~{0{>J2~THd2?_zDO>Vp6gQlH{i_r}<?+hJRbj!Qmsw
z%n!K3H70iLnQ%yXZ6#uivvtS#kg~6bti!Qap{<XNW(OpBc2)Cl7_s9>cv-}5gbV(f
z%4dp*5ou2jA||oAKDW1(RtOIn7HSwVakEMC-l@oCu9pXQ6E{w82Q^c^7&_Cd6qMwD
zKU0bv9ib8`kf^C@T&)QtGU8mVJo{{{Kz`!oE3oy;N7|2v&qKBb65L)cMgS-dV#x?K
zo457@(cjf#yg;$MK1aB(RFKAPIYJHW>T5$3l75)g;u<}VH(((uzDxU{-XI<9lLDpS
ziS{A-L0^C3xX<9P8=iIB36d>U!K6CE_YwnVaj6^u1tG9^?}7(}?oc3$;%1yC554X3
ze!%S0H(}VMQAA~eC+zcGlSE&i3_MoSCCf-*#q^GNx<K!NHPbHBAVb_-O`XF$?q#FS
z#>P?zol!kMcU?G}E6P5#zyCy0WtJ$Uk@;8Xg>#q}Jwd3q&96YETkIz;)Y5n6zlq75
zX7n~pr?j@LvqTaaIax$Q$6r!6+0pyaUvTWCEPbT=*vifNSR>g(gpE56Q*E5+Sq0;b
zp>$(rYMum9xa7TQEH0nV#Nu2D60eVabsU1_f81dKL6>jy(X==~hI3GY(KDq(w(A#d
zI^RF>e~sn1&iVjqlb4}h#)MAV)YG{p{yb*=!2q&L+NylLnxEbFm0?CIy^P2P+@z3D
zjWAf3LM;X2bNxCV_PB49@>Zy0Lm0zmG_4nU(^?O~&zV*Jr0Jr-*jkHZyOxs>=Yf~A
z{P1o!Y|@RfwB&Jn)587PnRzdsmC??Y$rqXHFV~Pt!@bd)eWbzkq*A?ZMhnyOm>&sz
z9rMWR15r-$>EXLi_2@u?<;)MatUXME?4a32_kj>QuB}HtZ(QA(Oi&;Vm%FzIZpoSk
zHJ@S<Yu{NB#J{<vf60wUY*ky!8f^xLy@l(3yJ-xOH$9{tTMw2+Fu&hj>H&<o=b7j(
zIg(Mg%MK;PofY41MQx&`a+#KM&6s4Sx#c5OMZBX8oU(Gp3L0A`b7JYG-Fyc;ub+x7
zRMM(GMfz<nLoc`RWN$Vqv873q`NZMGfDGsGUAts2JomfZ6s12|m1<?_Tw0;tmNTAC
zmv#>69E`>(H?(Wg(hn8Ih00gIU6sFN;)piOg*r2`Y|!Z4f^I2@h*Jx_QGV!El}gUY
zg8u}StqQ_K_8{Vh$L0#5rygpv$TXZjn)d4OD*7-RA6mh_+`(9el3AW&;UJXKGpi9!
zxUndO+j(f6v!hyywWF+$tVLnlJag2ttBNw=54%wX<2p%rwr!i`R7J~jASK1+S{udr
zX<vbcisO!XR6N6pE&LMth}NC`i?c;^S`w00^knBOO15X|<KasaVs$P0yH8Xlx0<${
zq>7wz(?LC(6Zg{3D(I0K%nsfn(;UQ26WJ?$YZ6BsJBp<oMH-krOkwQ{8C<?QKUALl
zI7t^caa4+b6Pt(>99AFnVjafds`VoY%H~fQ@G`f;r-7Lt>D$6#k3##YqIVtvO-{Ef
z7p7H9`++{Ti^wj>HDIAHgw*w`o15Dbe|?Hzjhop_vj1+3)@51zmsgj+%gH3$eYA~?
zYd*mVnrL)*4VYsx-S;Im&{yCARH46><LqAf>HGT!!+@|dsjyfz@2^xXEN@$3&&zf8
zNz9-z?)p};FXr7Q$-Q9wXDJT><Mro)7#=3y1%tFseZ`LR0AwW`^0T_v`P*%>-Hv`L
z2Jfh%w#`S@gk3*B<dLw$CADn3B|#yUYW{8`c5Z6LReTgx{|iNG7RNpj!bsAnyBp#%
z`$eLWpENJB?FoN|ka|b%<E?uwx`xcA`k%%SQ^7M+TQv^F><&v+P0GVjk8U@)2hY&Y
zP;EwlggY_9G3r(Oma=dOr4#&z(H|8Bq!NWFN@u8F?;VHZY56lXWI~f&gI0BHLfvg0
zyu<K`n-pk3q}Sm5gz^_>J&j(9MuE2<<hoCC3AjNMY*-%T1!1ld8HE3h=69%5;eoft
zNAd9q*i>{E?_RKcv{kP7#s(?R$QwIrc=}nVrJQ;xp@<C{hC!F`+-UERN#dYu@CBnw
z?1?mvRJ32>I>c_dDf%E{;v9t8sXl8)W`v<<9K}Alo0-_P<3odn#B`NB3UN@Af3%Yc
z2<oK$@$y)$KGbBkca&6J?VHL)$3OD$9l1nZOElOj0dBf+qZUiE$ok@~4Sa?$dV6tA
z(n6o!hh^5sA+$rXrJCEU(yxU}LzD8fY=&CDg+J`IO#&(84Y{))d2&n>KU>BC)y5CC
z$<J_K5;O@M8A>eOVswopQW6~QW&W+(mQQv*HHE?GuVvjj5Hvg~(4E1XWpNQoaB{h~
zVwAIB74_}OVWK0Pe?@^vIyIoexB08Oraz37%g+HO|Cu$<dY~u)fY&^uyR~DJP=K<(
zbh)o>6!0+s600dJo6Q04+#)CY!ORa5@`@A@jPIcvwMa(#4Y;LAl9wPiN*5XWGzD_p
zalA_J-f%eUSE`<Wayo7~bHA`y6#*_<C+)mK9g^VY$(1pWxnkGC1;;X0LNlMKK3q(o
zTz&GD_Z|b584((xwX2!z)XVxwV%|VVM;#MBZ|g8M`&!p>xBWq2nZTqINsHh?%TIVC
z9k1I!3*3P4^eu!sxGh7jw&!?WS@!8TQWq-YA0ST&N@D$|e4(kCTNnmxea6bf@w-FO
z&GwlWsbOOu;COT#fT?4VnVK|?GP|l@s`%uJOz}mWubqa1q4k!#;a9gj=*v>Q;hDwJ
zB&SE%(J`d;?c!IWc5o$XSv-UrkL8}W1I`Bl^1;)PoTQk1F0|!qJM}80(ZtU~NERK^
zVSO4klPdg@G1ikzJWhJ&1s~*%iEisN=T)8d<+L6s61%x>F~=*hJ$KISDxfGSkZUrO
z5_354>Xp+as~5$>xn8=239F6b-)i#_3RGsyTd6z$bU9#w_U1MtOu8nv!S>wkfsls%
zI5m4m$3w0fQDVv2^qW6tC2M2o^Lrnz5DO_Fs+v@W8cRlVZmfS(LZGuGX`=#rov5GF
z%l%uh)bk(c2#S=a8%*_~+^SgFe)ThITYvq-+cF<1!#y<$=bU5f-z^Lf&-NO2WDO9X
zwLp(@s`;WKH2F0M!=W@w-(aS=-YR9__Jt`b-HWio<in7u=9PIIntah$QXo)r2ox3D
zuUuBujn}t6?x;*%)Erim#PEfp-ysAtn3IRk$g+|=tkg-*OZ$DS(5|O%Nz9E&bI*12
z<muy5#^Kw!(91w#J;G&6-1cqv$s|Z&xll_v=ctST2tDS~02d<ehtgsd&Tf!Z<1~3|
zqrsgUZw#;7=ZNt*MJmtfG~5~gRF$IDir=?)m{Xv?v6~lOtr767;ndfmZYkB;R=|Bu
zVv+Ib-ezwj`pLxUlj9r!19uC$BqHv$*z76w!6;ZXi5fbeq6Q7^6`>l;KBC{h>xzB^
zg5x}LRW<w;phONL@50l8;21v#w{dA0PSQLLAl3-ttBHDrEFhCrX6JnMs+gMNZ_X7r
z3;q@7YNRn(I=Mt{BF>sKY{rma?=dBreZku>PWwsZBnIx+t`xe_VmdPNe-oNu%!TZN
zRBthYf%LHQoec$(mP{&#K&}sMopF$n=ja(vjdMBf%3lbv6ns$hB6K1CRc79P;K*@Z
zU<D+-18+X44H+ztpguo$zxLa3-=Ir<!XNg~nuc*d0`dhN(0O8<@U`{ka4~%4T?_Rx
zJZ?FzufGWoBjM`mE718?82C%C(`GX~nzeHE`h)x#*NT-4;R01bz?<sC62Ui)PX)ZG
zYQ1due5LY3D99_1G}K#h+;^~Tc*-8u7p#%?-TPx|HK<opSY3P9NOm8w6`VoH^tOJA
zJw0ra^wPbi0pa;-)(wF70TH?&$eqVeIOTA}+|HU>uH4(0XNFg9K^1CqbV+$(&e2$i
zLTgSUE<*UQP6ds@Azs4=r`gZ{hay(4)qBqG@>hYWSvD6~eov0u>GL-tfc*4zw;-IK
z&hstF=u#}g24EBFdXSJa?$~ueLvgc7IH7@$?50iG4R#-640gQnjN!)UZ5~jN)^P6b
zWiq%;@dP*f!s;Ncv9gNuJJ>HLl{AF07}%^|62wJ2Yj5ld_lS)cHtuRFH1xSLLxqyP
zn40fqBnI@|cX+|eOg1nM-@f&dBg~G0E5YiCRo=IK4(7IaF@giZw4aTW^wtD<Cz;C0
zy~Y*uIlnF5k*x*8rhFZ{;a)PXv9Xv8LKre?C|ljaOqvv67Vt`i^9k3xa)>@dtaV<7
z;LuK_9Kw42yvd5WjYMz6{pnjK|M`QO*mQAxMvEuqr(`^B;**bL;BIb`3Pvh6327iJ
zCtNY+`~iD3TBPt=0?`B$V@^+$C815kO0_ARx@<KGJ(`&J`|wrRr277)`7QQx>xdFb
z5i5hJ#cG@h(gio<q5XI@f4dP&dtz~+&|SA$R1r9bf(um@(CLj)kaa3f^bCEcaeeKw
zp@lKnD}4y@{VgL?(7_60Pzp%|-bWcY%P5t309Az;!<?=*G3$5T#9_lHk#XO98tq%X
zllg5k)n-a`#aZQ)g?)cp{|_xRdY;@n!#rF(KO0n>(Bx1eSHv$ND`90A3%J(x9de2X
zcaa5>{*XJEINJ#+9(M{NdU{U|2(51ss*#k9%9uRHUR5h=pEdb}uEW*F5M4sgb$fix
zkJx|Jrr;}=OfVMpBZg1*&wR3CttA@gnwv+ZtrKGmRgJlAF>kQ)<VWV92OJ%9nvGUG
zihW&kN^Ws)vVC~gBT>_rMNN>-#yZE&wj`#D6?^@*KUH1_2)EE1zU$ZAXjv9IyjO5A
z^6T{dFkb)ha=tlm9@C@z(eEZfQS%f{HSG)8^xqSdArx_rrl3O8uvQ!V7Q4INclx*I
zEltb|xm^$LzsNIR%ePr`-08Vbg3D?oYm+x$a@c~ff*3}UPOKa(?8JRI98P`wbCwx}
zGIo%0HHvXZ%<anm;!_f=t#};&rKlyGB?+p};&1xqXq{EbKH}74ph))v;MLaFua*m+
zTE#qg!aDwO?!Cd-NfR?~RbB*IH%zdQ|FadbT&Su7>tn&oTQYD4a#9F!FjQE`vpP}+
zk;6~q414RSD{u#9YGJ(=Zn_z~3#Vo&3Kxlts-F*#cb9XW8ngN}a=bB!3xZZb9k>r3
zo}E$ouqi@vK_R+|;;7TUaE-ZVyM%{MZl73^N(+QmT8sh9Y_-$>On^wVec=-4k=<9$
z&wju51<BW!NoU?#)LPo*KQ*3s-OY>XP3h(|1`aYRiYJ=*G4J%nJB`fn(lc(2no$KB
zcZgXm|EpYRKeHxjAjjsti_0frcDFFhU7j*%#-t04T7ui#CTMV+NqSsgz2{56n9q9o
zQ3<%I40|D=?DDyy&$FOeCGqkV+E!70Efcg^{&i)I*Ka6A%tDA6<;#0-KHXa)fAW2E
zD-GHjlY+Dr&YzyrRVDNz48Dtc@lE5y0Ex48jtct9LFaM?T|reGY%_~XlL1;WA(%pQ
zPJ)BU#r<XfCTpz!x0~<CHW@h5@FoMqFbh({8Qm|g`4ExK{Fw_-HY#y{vxPRO&ji!|
zvEZqh&qZJmK8%>kIsGSO=QEGF*Q^z&x7v`rOXYXhv5{FKkKm(Mow!J_g!kiL&-xZM
z7IVKoxSwa^E;cPJr=@p(Bo6Wl%(+Xeg+6$=sJ}0hQ|1v;8#|4#myK$j-1$YIP4>#L
zo0C^ZlMw??V2EYvm*cH+7%C+r<vb0m1CA6yLUpov{QGg+ZQKvsk=6XmWsSQUhr?x7
zY<cg80dcq|M)fqBNA=d1{xw{uo#!7K${+bcuA6lb-6Y%34)S)N$wAJ86tK?|uVXq$
zJh>uCJwWje*B29ggU7BTJSlZyvWO`fjf0jl7nY*YDQE}K``_`6ds;z-q#tg#4~UJ~
zK&81wZ1rQI_`_~+18Yj#XwodFM2Sh40x1u4`7u#e@8^YbSiD<UD712DU}C>cGO4<b
zS%&oc-Xg9X0usinw~kgfQ>`Mu+z_FVdV&#R%qp{C>J``yPYG)Lm2sKBdEAS?=RWQB
z+ImkZPS8F6F}Tif@3Ak1F=x{{DXLYsrCDVXRHr?98*H<m+wOU9g=Pdlee!ZK8#<~k
z_8GV%f5C|8H?jzRWEO^*Vr=6=(Wc49S4;4Jmgqi;4A_W&J8^z@tb<X{V%_6Jh)y`u
zRedRB@eLK3Ra46Wu|IcWQYT!02-ND>i()iR!XWJzPQHz!`hqGr0ZlJ*J92IX+zpD=
zV$7mbkrG>Aw*P9OUBtLv?&Po)RA^?pR#!MeuXiN-sO9Elh>L7hjN7XH4~+$MMD(i3
z3)!dsviIZcj6+tojoBtjT)dyn)IIyLJiqN>xEUo}MApuED;0NS8tpnv5Q%QPWn53`
zcjG<JR>kPAAYGNf6HnH&H}F)p0M=ATx~1PY{8D2L5aDyXRiWWNVYvwAzA6b;B7co%
zX1m2tJ{5PVjSm3n5L?EX4~6N;4t#Cs^zjgL3fKwr)dFP@K(CDA@q@`k(xfsuoT2WC
zbEvE59uwb_KBGYeUU@S!M)}S#|HHZuWQ(juWMotWuA`Y}3sv7_S~k^#W%yTE^cJo5
zZ^p`OPUBPWuUVMG9zFcDRkeP88<6y63$sma>S;rYo7=<8Y^UBFR}`O;q}5_JUN@rl
zKG1$LO`{$obLV}@qfm$jZQ79p(oV;5*X>u+Q*_tx0E=^1xsIfn*KIBOCnIP#>(COV
zs)i)T;O66%ntj`^z(BulA{S?g*nzcEm73y;;GbeUNW)&VBZ@h4$$J@<&b+2%2i)Q{
zT~<Z!NJ>`*oR$UFhS6Z`w@B}|ciw>Dp!73xoNm!6o0%c=I$2DG*F>({Z=UtZvx;wR
z8q=C=y@MK|=3A#2a5Fni+U;^)uH90fnen2%TjGRcx*4)l>6QOLqa?Q?_TYbn)YhR?
zt##lr&j+v4@P`Gduuaiv`Nyp};a<h|m@OW1-8yqO4`QO*i1^z%T&iUdi4on)oFK{7
zGz}}77%%Q&!IClgiWC{?E{f=4q}jx?qWEw&>|^o2s6Nedm<Gs08_EvGi51!)Hob}6
ztoi0|L^#iSn%yn@UKQUYZ<PM#<}JU)_a+F7;Cmgpj=AvC$<wQZnJq4^DSN#}(M?<4
zoH4g&Ra5oXRzX#UP^)<V)wQL2&XtkoX_-*7>EeNk%9#p2!Kfd!8TLdjJHWSuDAO?)
zH{brki?qKLr})w2U_`^!Z1*1D*xQGKTQ!4S;=YHs(W$rzsv##hjJV_i&gD{X4NLpz
zdPdaA>ep#E4uyLqB}#IJJ?C)c?#9mEkQ7tdvLTAnnRsnS?BZW@gW*%>^deL-?~9q5
zlpp54A5+WFY$FZB#*W>VfMBRaOPA6s5!=tjdQgU_J48P$-)a%*sZ-AggoY8IH}zvz
zIesrRy>c8eh4^Bp`Q9XsMJ=4;JJWvHOJ)G1rYE0ywuN%|4*&tam&{pM{?Y`0^|f7_
z)7|j`l%;%1+9QXWFID$m7$iS<!cI_7Bh@}(!^Geg*~{*DEXH6Nz7ol1>Vo$vVDP~}
zi+*a$Bn2ks{v&I~fTagDqR83xAaaL{R;!ys!^0G9jnkd-TcEUg?c4jqH>1%w*MQ^T
zcG=`1!}G3tR(#U}l#L}aGwr3@qQ&0+ush<HVb>aP+t(fkjb>WA(50t~T=?np&xZ4t
zhr6#;`pAuY6egfHz)`QMKLD&~QfK@@-c|ZhJL8>LOZS8Osz3Rgj~d*v`fYy}NT!23
zMJwI%Cs{I|f6QWZw0*bGl*3`YjXIA$)U`Gu?6PvNyLd>wQ~tT+`w8V2*W`gSSAZt^
z8nEUMPJB06b{%&;-Zkq>zvnWaaju1!Uj;dp1)hV!l_ob2t*IF?oLN$L`R*U>;WF?t
zH_;gDe>FeCU$mxpo5kQBLF(#{&!+(6$+;Yu7kTT*GQ{0>zJ42?P|@5n8SL_=z$XXJ
zS9OD5vLJirqQT`^+S09u(Y}k)nlsjfeHcA)9q|dr@8(4lk_Pq~OF$0XAr%Dln>qBH
z$M&lv*~qe0a;Ghs=&7?gZRR9*_ZiF8ilE&3LgMB{8)oMt$h;14NA%rj;{9>ZHHm9X
zSj%k959``{v`6*|Z%o1u?{#Q91E1S@S>bJWVy!1lW>4eweWU4fo4s6o#%2-LIY8L|
zsG2>$gqM~P+2)vNlvhZ!q?2x5$nba8<Av1n_R_xaL`N3|j%6uv1JP^F<}YEL-@mUl
z@j8cz=DUIuX{cj*o<;C!Ll9&`HSYHMby*h{st*1VSd7LZM)oqXrTIhZcMCw}^SXes
z({<sz>Any#_Ndh`5}Dvzyc#%*>6=@kM5{ezU+2g7>lv|vo{)qH2&&WZ%&J;X#lZI4
zQ~-yEPp0>CgJ7Jz@=ua8&mJd7=Wfh3zIq?P^J^V8`;?*AuoAu_Dw9S(nyh)#U3}GY
zJF+n3#+b~YQp+j}y<WGPL?q4lZ)>qdNw=|Sg7I=ULPv<Yv-pr%(<H;+ezQ0bgC$p9
zlvU&^0;T+vRPfy>@hww2Jgo783STd^Z&%k({lNJ%iS~y`dzAv(luRnIWO=2?V&@Bs
zjBe)fpTj7{&|mbL>I`qOhK5|+Qp+$xIoSA3gRl8~G?-88?__+>HH5KtDDCUGwtant
z@6a03r>Fs{HNwqfN^0`!c>{6ko!@UFzoh0^<?PJeJN`r;rZyG2B`*!<oc!s?VkPJX
zOBMjdYI=2&n|*R|;ui<KHj^3XA|dv0^;JEo=`c|PKn7O&M9;`#jVA|O&&l|PUqlD+
zJ`h{G5UvH8p<DR~$5e4~V+f9ak&GTE3zlRU-}k682Gpai-o{NS>@Xs8;|95&<BZL#
zdj+h<P>`}0bhJxMM4@XLq|a7Ys+rlo(1YzlgY=|?-)lhB7c|P*2?k))>KbK6OHaO1
zai<l0sXHLhmN}+p&nR1laaaN}dc?g|`dBr_mUsYv#n(k%M99q(w<=Oc$PP+Hhy0W(
zDYzol6uh6Z>DlrkcAwc6BuJD5lirpYgXPub+)6bm>F(Q9Ks>u=ofA-)(Gnz~dUlcO
z)a@|`I?nK~8C@kyXmWiMEy@1IQ&A~lsnV&0L-4NPn+(<s&kR>WMhsMVBk3ezp53*+
zu_;9aTUGCK>KZ_Xf<e^Fu2~h#0jZRnENb>iU}n)&XO(dod`EePdPHX9!sSRLNl6P)
zuc~E<t6^=LJs_E?LuoqHV8quyL`xMcvzs<}<P@!rGjPRB`tt3YO6xsPRqFMJ2>f+n
zW!wj@54m<+WrA<)VUADS(wQ%OK$$Av+7S2+>xeuQ_sJKo0!t#nGi#pCrpF-DpcS2m
z)+?j7>f|P-&{T+5b94J1AIwp;kr=b=i@LM=_`}&M75Z%24?@Yp!)jAcwT$v=-}X~`
zcWTvknvUqZ2&LLq2ttUcoR?7kn?c1iaX4EF%ZC-{gsg_of1)54*lucobtem7QL&2S
zeTNnLf(5j&>UjLRhPbC6KqJ0ZLs?(7)J<MggfpSBsrp?LkD3JLr?Rnm^K>B)X{c$>
zn?lT?XRbEFI#~rHa5E(5nxebFz$~&&`+yAgi!9o=#DsA6Jg=;7lAXdpJ#ZXl_)XEs
z978#IijQN2T<WAQHcOpNQz5+nf@I&FhBj0E;jvrYrw3k-L6H)WGzWu#g2(HjpmY(V
zrt5iE*DV=YMkqteLsDw>i|MM~)#A_Rc4{_kTaO5!*wU%Y(H{>!zVd7qv7B0i2>Ts;
z?srRHSjV5`=earw1#Lgg%;ULN?RoDIoDO(-OgUp(rBM2L;q2Z&Xkm8AVwSzv4k9_#
zUOfQW#&o+;JTQg%m$k!<Lae{Uh5JQ{MqY2Q@vXbgk0wu_7M9dqF8t`@M!33UrIyC@
zn(o&JM;jj+CvG2d=TU)HTeY=2+Bqw79s9@p;w1)8T8*KfH3h>&gr=(RFgRu|6ir37
zc3M%hTz6)#E(e4xJ7ou2g2sp4n|fi=IdxL}34YQcve(dWu^)-srZv6Hym<5r)LFD}
zx7|4<qp+o?eqEzZ%l|Q0z|v+Qqrt&EvR-7)=B+ykyVB_(u(2Ti!P<s6K4$E3`A${A
zF}H31mNG*N@-SvH)mw;hKI;acK@DFeyPEyk80D{w1$urojzG!obj-`Mm5dpsGn~&?
zYzc{LXjn>UjUV{rS1^yB9QzYSNcH*m_M=56L0%ru@g==GP!<Igj;Zsjz6Iy}rTYrt
zRxO$o1Qq|YA5Rp852!rC5pT7&?drbgGIN4J1GJ2zM65Nh6LNUo>HN}ZB9t4tf9vW=
zOb1^)Vgd2O6(kOW7A3ggp?_mn{YhMrVnf4vxfD-9M%Yl{VZMk-AR73rqK)A$<L_2L
zR6fyBD%Z<#FhrzH$gX%BWbR78*uXytf*+c1hQ@xL1#mf7n>;2;6uSvz14^r<-u_#&
zp59w}W8<?&3vI|j8%c+8XCl_%9z&>pHfVo9!VPYwYvEob%j{Rg6A5&Oa_ZKkAo@9x
znKC9M#!%?uBEw>_l$gscX%ihVCkq|8>aQzSmj?KOg~P00>jDhUZq6a6+;L51|IZm+
zF={+yvHV$X)8XB8vOfush5tcadc)jsG|M8CzzQ-lZ$$ylhXP=fyx)*`S;dN0a1arj
zA-3VzhY#p(pch1+*6cdiga7E#^RG;4e@4*x<p1I+Q3nn&VO(wMm{i?k#FQ7#$Nh_q
z7?58JyMcb(K^P83psRq-&w^i(`h(U&&>j*OC;RF#?sve@Qc;6kiNtTN&HwsWqBtPs
zy|0$g;#O4!J%vgYUx6ya(Ni%Ee_e<Fg#twRyA6@uf43p}Uw;4hBj5g(Bm$iOO_KhB
z$po4yftkL*18E8)AXoHX?YI8#c@cqW9ZQ)@0isLyKb?X<XoP>G6oHZDIxs13T%`?t
zx*`?*PbaMZV=M+uiC&7UCAbHS-d~wu|HlIUzolkX-b)K%yEoZ*Mb*-vo7GaDryL|p
zcu}K{uogVjZ`_~9F5ZBd4X{NfGX9HWWVyW?%RBH=GEs+R{+}ht=gF-%%Gh66Zv7Uo
zi`r_@pr6bychP7E4?R&6j-Emk5DI#!)oXnXW+o<#$%8mw1FjQ8Uf6sm#^Ni`c@f~p
z$Iqvyhi#|xrcSECwDvsWJu(}9vax)ccSF8gPR+^b+4Cw`wRXJrzRG89ZJ7LXg$qVY
zmrs`{61B%uX_p(D{+_1?0oCxs*U<jM<;A1ovm(o$RHDnKp8mt_)xN&I%Cy<T`83zd
zy{9J<UF=@Qr;}gyaj`ThQ3TS;OarBj5or&pRMW~SjH^9vZA_lbOtcS69BYgexj@U{
ztM0D*JEPX*6CZ!Js-|LFb{l`0G>ZDR9mWV5bv1VZl|8OqM?=RV`uB$s*Xz<9EmW`p
z8g%yXZ@zH~Ik2FV$|DV7hdo!<v;FD~HL-(DNg^KaJ->nyvW9TzxxJo&QOf#91HOl!
zigYE9?GrX?Ja-?8xx2Dfd7ds9{p{Q6*eP+1;7&K#;;vO`C!~=e<}tiLRMv#XVce^v
z{`>IYJ?a9nu`~tWzHa2@g1lrVL)@3~6kXHDqdVB}+~lo@q?NqgomxvgS><)x7Iijn
zO+NndXAl7mHhI3suZi=E6yC<LMYYlwcWsl<31}rKd2lqVi2gdkKV~}iNGj@E7a3Pl
zK7E$`I+&_@;m%xLz1{Jz!%8FR+OwJLO%B0s!!qIs8qwRS5^fYVHQtryz6QcOONFu1
zYXP|vc(xNq^1AUeBl=*$hE6I)|9c`HXufgB<Dcx_fiZ4hjltk=-l`UKZ;>}h2nh`w
z6O#0HnTe0|B05-?EcnVHYPLmkBb^6wyt6n_VjS`HHW9-5Yi%XTzxU?P4?kGxA@6%^
za4fMkDRrfz`mk=x2*an9h27VlfHoA;u6t9eE@swzbW~lfvkS-z=Da+vQm9=-bPypO
zuA*8p4kjV(atVcjfdM`YRfWx)Dv8~bDu#FLzK|ekzhWrtydbknBlKnTtdJzT@x5b0
zW5<ee*>SU4<He1jhW7r&hI(1W##C_?)8qBAt(oTjt-aM#Q)ip1;UCjXtPO>;tc^Z?
zLAATC=MSq~FNDS5C(gsA&zFMFZ6eadsLX;3VW=A0vyPpyt(|%Be}6XC?^m6)M>b4W
zpH^@n^unFa=HByIkCl3R1HoEFD`kz`4ASL=-Y-WAK|(fXLSii&rkPqBYCJJz^z?B!
zIbn^f3A`09I)fVh0u*xpDbMtLnQ!PV6bi8M$r1N?0_n5pIL_tWnC9V)8xzL@Rfcw`
zXbPz&`j4Bz+AJ=$Aki#gah|`Jo~3(pV|{h{tl7%+<;xz-T_a;Js*UlpDxtPuhQ4Fm
z;fiaB&0uk>jft8`cz9;3THw9KeE+PhP0E_Z;ZW)B^RZ#LU(}wLnA;wGlhnf2&)LDx
z;6w<CxjA|7J<wN+A})O;>R5C6bf$}LvUVcChTw09@Q;lSBPBv`X*xP0Ss*X>$Pjz7
zkJ%yZeo^zKu!dR1!%7@K3N~s{FAi!E_mA+XQkZdofzV8ed1DrZL2<F}7f6Vsk%1Au
zl?sE{@rjq~`HI^}(RU$gk=FuEZ<uJ2&}uxFniE^x;_sULhh?tw@aejG*$Kr$O4dM~
z&Q4gvB<PJZ-Pl@WRfXjO^W2uU=%|+sY7FR1YyKE+OZ)R)Hv`xDGRMP+G~s>>uH4*Q
zBf|H3S(>!V=fR7->!Tz<7foW^?(FQGmdg7s;>fc$QCTa>ee)qNRE+8@+hLNPCI7rU
zY`DS-$;BI#9ZBae1cN&FcuavKcs5s_|LZQT3#{4Xg)43NWp*RqKP*lP*wxb-EZ?}6
zViK@4%^ZG2HJ^2aJ@)i~{F((bv{a8lYebjFvmz^A`707``X_+SxwtFy@uiy|ehTr(
zZx&FzGdw>EGMZP&8z}SKO)zz~n6KW>?lD@q*pZ~=9&>oi2zeUjU3lq6o;cUKT$E5@
z^KC{WcTL3p#6#Hm#N+Vv)Wz0GTG?%Gx}kbZsG*a5GqT)izQ}H>ciNWoyQokt7HqOg
zGKfhrGRTo!a)sygY;<;OYNLyABO23&KRMV`sU}_zkC(^j)N7>35=QRcEcJXTkrfhh
zy?ZdyYXsvJ5?YJr0y|&vJLI`3Lq5ApD@F>m9*r%EI<?gurMvhS#n5|~=+esZ8Q6c^
zJMzDeK#3kuaLu>{5o5TTl-ua+HtJw%1g&bx8z;76RW7-rFMCqOr+;NjF4tVHob*(N
z28Rppa719b)Z+A{iJwdfZu1TxJ<l@p1!neh$)~$@p?InmIMT{_j3LrWZcaTr?U5Bz
zOU=94(_Vsib)vvhwi?-O&5X|4)e_qwR*fnSs`CYyZ*ue7DYaglV>Zr5)}YRh&XzN`
zrY3c@bHfrO@i7VMmK)wB)t<c6O^Km%i*i_P;jMg?z9a0mAS3LyVtBZ{PW5{cH;fY8
zeB&sDo2S@mOKZ;Dg2yTcBg7~w_UCyU#!An3M}l2O9$^2E4f~g#e4T$KBl%zd(C{|u
z2}n<*)!U1y$*!%l<=G8{tYSnUZi2+JyIqsQE|qg~rCssZ4hJpNYrC~No@bHQlazo?
zWb(Y(n_6h#r6-PbM`I;4z%pK)O=9u5lP;(+C^8^EKAxuu(h3w}Lo;XEYsr>3EH}A5
z71!<%mx_mJ%RxPlcU>AI{xDg6=VIUIb!au4OgL^nIga*?D_bA#T9wRIWwK=;%axwl
z-aj{(UXDe=7e&>g9JY5@jnY8T4C8<?A+J&R`be-upjOJ*M3NEK{GWE<FEcK{0#=BT
zv~+<Im~0p64&ecz+r*N~Coesm&n+ho^tXe@RGMu*b%X|Ob9M{P^bNvJe40*MP7i79
zSAm-EeqBqO8}(iW=q0u={Anx2n(E<jUb6>nLI&@f*E?qfpBGUB5&=1vC%kHNZ^HfE
zu$a@4UYXtT?8dBK0BLUY4ig4@;h_@Vvsa47#d~}0ja&QYW7S$|tdR19<;cpfjn74B
z#mh)i#k}}B*;|PJdj?Ba)(({*8`W3#+BeQVXT;;|EGx2crNf~tT00f->J5@^oIs6E
zlDf7xnJM1`pbyxk&(8}oez_0^X)#;W(V1Tj-SK)h5dtqjoE;JB*xi^Lo81$1Wce;L
zxA*w$vRafy`%4p{Z4-B8sMo!hCU{7*;M<Y<M^V0Ui#@6SK=8*Na-|vIp(IY$)LiKB
zZJ5ZsCDo#KwhwE{54c)0gIpS86%Q>%n)Uuc8&v~KzkccfK-sWr)DFWyh}L4m=X>Oo
z7$~}fmHFNL=|jfIa^E<gSCPV{DMX0zstZ<A3pcmtD?qN1bd%U>bk)%2!hPuu&7&i@
z_ZRp@l}pbLdR2P9$T!_NyE!63e@AI+5U|?^Mmo~Uv%j!4KhYt66crDv&u)K4XYVqA
zTWB-2s-@y*_jnmN-y<g5KWE|_9x^_J1z5^^$G$H;a6jW~TFI(0PYsQ;O1ofgT5mM2
zai!dtxcIi6=HB=hEcoXxaj61bBHY<k6GNY+VF>2;WKE>dWmCwL_1(L74b8cz&6CU~
z<=7FuDqKd@WbDE_$GPM^lc$~AgT;g8Uzh{mtpdO_{X(VaIE*f!QPD-G=(ucMXWWaE
z#%*M4tA2gUI)ZyJB1M>{JGpY)$f02L>~qodUa3KrjdIcPA4e;qo6}pD$ME<*^+P9s
zXY%}$-fPWDUIgCy^BsS%q~CYs5oG<>7os}gl|2)5{tzeAoU6mSn;>~H#ruWV@PwKA
zp~L-w5|391813t`re(X6vDzbVyUpjW7kk^7i#))9<pbSER9oY<FED<*vRl8ZW@?2@
z=t&ENyGZ1HTuPmfr3=W)(tEk>#dJtvZFGL5XxcMbJ-t3AG(8j!^)DzHvc7}4fdYvU
z7fYBEB0Z@RIE0&?+TxOdl$eAx*xPB*6q<SY`87g3+K47X*u;^A>*>NE=CH`0y6+h1
zK0FwQd7skhAst?V<cSx^IfO%<#kE@^s(4k``iW1A^A3YErbA*E8vrn?P)=Cvk?=qb
z!3qCj7)JQf9Exd*;%NP^1ZaVr2=N0}I%tUlaHB~p20QLW!pOk*S@SOI^hL^mx(B=A
z>0ax^JLOWC7cDR~qdDTJf@=6dAg5750qMI0l+z&>6{0nr<tERztKMkfm#SS--?&8Y
z@#=GV*Yllbcz>8SeXCl^x<sY-V&~S(asQ#0_s(8`-OirSwA>p04$yFeTrBQ1;$9E)
zxN4Wt-k+7Z?ibl@O-=Vwd`SK?7l4O3Wbp1hC9TM7Mxc|wiq1C|wN775B++ph(;}KW
z5OA0LTA@hbG`V!GRQ4`*t-@r|>do0b_D0nzoFj%S;io4{xFkr=`AoXkn=@D7l{{^}
zYjwr#l#m)vF)|G&l6L$~mF*fK=_z`adCn_KC*PZYQKkQKiPo+hp#vkct24G<MxV7^
zZV;69H5T{*qGFyq1)}S9^^MbggFO{E0aVa4`_-(W2Ru3tnpOf`xoSO0XeBF#&*#rl
z0j6!)-sZiSy_L&oJ8|amUD=#ix?|;)o&nBud*kW!V6LU<%brPF+#=u^IQR7FKJB+D
zeSL)@rZQkH)n23fydkvPBLEDSP7Fqyv&xQtTA!XkHWXQI>bg*ic`zGRoplN7Tjl1a
zEan0jOf<U{Jv&iivweEhqqrtibhiG^_L%V#JxeNG?hE4@53*DVFA5#qca1&(%ZK!n
z0~mc;p{3s#e2+ZeY4uLz6%?bG%2G}oj2|rcw$nA_aBM%f6k@kABb1?*4D(QfR(p`?
zB<MEuO;)G$M>n?0dMuPZm;iQ+%7dI&eoQt974CMjB(3ZAPgY8u<VwjO%KULEy>gDo
zmkZy!oPB^z#Fc2d@6WiE!B2I=`&OyQN)1A_Sz9B#wVX9Y&FUNuhsaGq9{JvLEiBEM
z1MNaxr9JLQ3d8a8V>i90pCwiDnuJ|0^h>LMjcDY)3R!HbTljH2!a3h~+KOtuHd~)U
zWhA93su{I&Hkw)B(o<u%)K<7s22f!jYve}c<D%1Z;C#sBy$c!Sn&0-)9yqNHH|==8
zt6R|#moeM5Iy`;QHQkX|M4lgziMr{yt<Qa1^)aC((SLh<5dgAVqbzPC?&qgmxf<Hs
z;E?-MK|O}Iew(u^$7L2+lG&D8dYECvPtza*I$WTvf4gKH!e$Se#^#474s?5Hy=jJq
zCO2n)y416yvqOz8pX%~E%<LWK`!ee|H(g`7c~h;nhE9hnl_kG1+B_S$WI+`~jDd{y
zV?}!JNjM+P6wkbV;i)*jv2ql0Z@PCn(-elsJ2cd=)8XK{fA-RKXHHu04p!DA2D=(v
z0Op}>07XG_=m*bJLMKP>(-jNgMtTWFK8-k{X6-S?DwG6R@l>YceV}ig>$zSEX{k%t
zbYIOmrtCN|PoQ)s3>64~N+3{={t3Kkkvg2sM2kS<Z@MD@(sUdWyk8`iU-%C`xZ#zK
zCRhJM8Wh+X$r$a{nwIl%SZMhnpKab3M^V`{$WIb#uHdf<nNr$H|CetkWuk(C&Ti0A
zo1X+w?(vn29mp*Qo=T)Z&hTHM4SK#i;7C7vXSNUm40Zeequ*{JdMATrg<J8i^S^Zr
zkzNphs5Gty9$x@Rz5f5>js9)_{`T-X%-?uK^A9}Ihh;(Who}~qqP^xC=hG*@`<KP3
z3k1O=!DQD)TYAV#1Q1=BI|~23(pTTmVPR>OY^pFoUK)ym&{Lf0fA8tlH}sHE>CHD_
zLi`w48}iSu2<DAheFL~}&7}Xwh`w<&djwz><XRxg@#ieSyZg{E7m&*6jY;>+YWy3p
zl&DHY#(gb<Zl+igYR-7eH(h=hu?V8KNZ>AOW%<j5|Ba>U(17&dHjU4zbkQQLchIvP
zD%o$cUv9a034vp)8qg;8?`Qw*Lpoyth|q2IT(nCPoF1&>!VfprYd&rE_iS4ov!;^&
zXXH*e;3{(WnDZ)f*YHQ=E&?DmPEJms(?pt(mwV6)<1)`h<5?1le^_}Fa_o^zM{@3i
zJ-YqA=G|$!^K+CJ`Nz0vCE71@rlIr}lvbQzM}n?i{X+80^eFT4qLwnP?w`D7ustI<
zKG0JYZ;7+7wa2TzabN48b>d*3g0=88y-tdF8#d1`YHd8w{u2lKbdhgKhM+tgU6?_L
zZV%n-wDdytXq!;-HsTTN<%j3Yke6mSrlHcx<1?$!vs@Iad<=f<($!Mh|F`$BzH$9B
zycD0To%G6IGhsrXDKWmB+B_eEpKDmJ7vJima(OJRoVl(5+=-_iXC&~{CdutniYNrq
zO)YMKiska82x>@pd4f8tY{}+(`|qv%+Xq3=z42lJMId>0(ONC7Jo!fAKrTcw`{u4i
z?D4=o`J&@4!+Iic*d~+iAQ0*-7<btpea2wlE|NS+yJS6U1^Ypn*7Ud6=8}<et7jxN
z^7~_)kx?}7S>)y5mX-H(&+BOjZ6YExuBzSRqMF`9e=`7xzPR8+NIR#-b9_y+#2rb{
ze+%DxEA7&1w*Q}Y`08v?v2Rg%uOt0yj}|X}R1#SX(X(iOLU6>L<%Y(E8<rCyxj_&p
z)gdZ|Lnj|qMshJlcsbU+{=TnF=bsOlNiMpXS{&+Vzm{5}=Jj?p2dQ+2&eyh+O&}}x
zlVzd=zBlVms?0BZJ)O3!J!15@Xx_Iwob<~_h5l=fe*)`26$=?J5vM#)vLuzn>E7&C
z<EA%dzOe-P&hs7F>m9vA#vGbtwK}`T8c}_0rNQ(p#!P{f%T}*=(kQ#L@9ugUz~jRg
zt;Dr<_?I6Z4-bq;Ox(MqT#Ua!xP4kwO;gN2D{4wOokMu034NpbeFO}63dm%F;Zvf2
zJ9uyEoG@l|bc>lT#XQrzxoaacAZJ|XpKWsEDeGF1#gfHSh_w*Sh<3jOr*i>_Qd+i6
z^$zkl`81XHc;L~e%5VLN@4?>C2r@$0Wy1x)AJIpr8C%SkMl0npGI37GvqjR62oqL9
zwZ6mXY+gBb*_e!8x%9rHkN>C<rHZ7WZ-{x-FL(m*k2gDdcIa6G*xV#zs<1uo<RGS)
z2$l3@;sxY@Xe}4ruieR|&pM8OZ<S7;6>XF08z=VYvvw514w0wu*=9FBvJ0tw^FF(j
z(rtc=tVhQ01)hXx<~~xYS7FNiXDzFjBgHMn-WaQ&sHyWYE}qUUF`7feEIk_56R(@o
z)MFqMuOAfym8?b5TVx=WcbkE@KQXw2m>XoV&{lq$A+R}Q?3So$yBXgQ{XrU<7^A2C
z@qUfTKgtqcY2C<>@qCFf2tP_9ZC15)GJ&{%6Uf3-@9XiDR=zk^r-(%%^P&oihsTkf
zKUjf#(O$v<VL#29t5<>sgNd}49(0_3Y+CKnjP9H1g}nN0j~>ZL5ymNT`6>M=V!!vg
zP6p$|H`XcEK?UgmxC1+xc{uEx3MU*I9iM<4dG7BU+(K~e6(^pWfk;Qi(6x*bZXNC%
zw#K*mgs~Ym)vpZT;o6yVUF)3+hsB|Bdrz6a*lfp3^k7>qC-p&K|1yOCVGV9#oRsW#
z_{Hv=ialb_)#$LP)_ef5fg^_kcaO<TJi8W7-3#U(ZyJ2)N|4rs`W<J`-ZL)WL<egM
zYkZEu?v}2p^DeGl89IBcpDv&it!X8k79`{LLcj_}ItG7G^X#9OwWw~S#JKnxi`Tm#
zeDJDbhA+QD?%=5Gq_P*+upO7rlYtq&hlT~D0=qmn-bSB=Cj!>M9DmU&e90<&wX@~G
z&6U`DST~+HWM<RKRQckm61&t(k8(Vw|Crv@x|&R&ar-Dg{e0_2DCw8I2T48fScDYI
z<uCncP1{}V1C4fMtnQ~d24$-vx7+NiUMZ86z=HpDd`HiMLFgv0h{Msb^>KVAm+Z~n
zzS8p8B&TQUhk=Ca$`zJ1)w&W$(ujc)c1}MM1Gfh?QvW#Hf|qEr2A=rZ9}%ed@QF@i
z7K?P^T&0dr+r<jIo~9k@k#Pb*VOn%>x9^y+VaOd1XKV`D*q2O)gGCM#REyHQ$>BDr
zNSi)auYmJd6RS@*rB63YVCY57TDWF6J5rhbM*P`g)6k-(xpqe1(YJp*uyr|gdft&4
zHT1=NECG1namSvMoLR-wXL_)>DuJvarYv3vc#g2QbjOm_z2R%p9YkaL<7}Hr*QstJ
znpQvAF<t8CK4OP|+_PEM#?{iN>`KtS=f)eS4~?t3?#Zq9el^^Q_`jTjE0ZzAuAU3|
zjKS&Fcw7|%KdB)@ynT>NhpYN3e4!H@B}HY4GzU0xPgJ_&A@qY!Kd{FoRJRa_l}NBC
zrE^~b7`Kb`nNDi1WswcU-R3X8U~_}U)c}D|f0A;)=AR)Jz<L&uQLdSV^jRg3b?%r9
ze5gCL>C_fXP#bffmsWWi9#evqJCRMcfuLv8{+I(9P1IG6>RWmU14~URI|iR0bH~?4
z%!zBAlpJ+XjU>eL6x%dwIx88+$1Ko>==;URXchDw`Tp}lkE0dc|Ii~FU&X5Yb<~3!
z!cJA6@OZ}RHmhfbGPnPoSNBo6;W3hm>&Jm5B1u%9<SA?b!K@Vm1&TlaTF;{K*zE)n
zcdq$rkZoQN3RKMK3kjG2iu}_+DtF;aLDUi2KMiYwiPfZvRmjvZ5c%{&SN0SGa-*%b
zr}6)5?%D&HUgLg-4#r7HoJx|FxkO7!Q6yp%#@I%zG-^Vn+j7ZNx=3!nxmBb#w_z@A
zPSKKz%=L(oIZ1~}l3Y7Ty^qf6ZBFa``}G&|?0LSo=llJ9FWWErm<NmCm<Zgxo)fR&
zAtrGeKXV6KV%%p!GU%zrWu)e^#(WzbLEw7uMiVg{gL2G(%r+w*8Zz;ioP@ItObmG+
zT4MpNO{d_2v+EsWuIE1nP94&Xn@AKC#u8tO;NW-5sRSLPRy_v)2sa<CPB!rle28VP
z%bl>d^9Y>?kma|pL|kJ@n@!!Q^FuQQN_p>YY}z;Cs0E_(5FGP*AZ+)GIwLKXP3_}o
zLUa7ZRU~UPm+ihSFsC*YNK)>|C>k=Z8TVvwLr!f-Zrc6?i8~bf;$&^n(C9<24n$qg
z`hV)+)PPm5x^HowqEfa0h5m<NK5Z%nVHjS)gV()vUfLu5D*Jm^)!^dhLd~QNvg{Xo
znjuBn;a-8ezct$aL}Qjgr^$nO5HEY8I7M%n9dF<mo+6_wklB^QWO$?TG!%#J4rqGO
zgkYJ~O44?k405~xpigd?2c5n-cN=M7GsiOtSeutO!jd$QVXM$wb9Zl-DJ~isr&QRe
zchH+gKhWoHAe~EwRDO%uu?&E;4HZ&8f)E-NU2rr6KO0}&tI#t>WT@VzH`7@dM!s+E
z-jJk%2U^Y=c~H}frTM*gcR=c1##TRe17s8rDyhl2OzW;E3%qFp9~u@(iF01Kf-+(2
zZL4HOYlV|^Y;*T+&8TyVO90;6cE1NJ%l*)iIM6+OOt=K@@8x8Jb5P0axqwvmH%~GS
z<#btadO{j$)uuDDg?%LjpXW@_btY8T+XihG>jhpat95d86_&$YtMh;?5djlz7#b$9
z2g}4D&!te9&c3pk>O4w??=Z;7d_z}*KM5gpy5gPu<?3>nX<aNiBbmC}zybKlQOd6L
zlHV8zZD%U#uV|9k^9EH@vU@@c214;u@l}x7&RWT&;-UN1;oecGsV4T@lySWGohcd9
z*3~v~C;@YgZE6E^+09?z1O3NS%?={ZK^1RT2bp=pw~Xc`B&8z(Ho@w$7fue8ZjJ>}
zvY(a?z6_Qd6jIufIXSkxb<e7zty28gU;G3sV&5Wv5dSiSk=sp-;I*#w;-h<zX0;Cu
zmsMnEjZu{UE?FPiS?#ZHRvU5~>Tc&K)jdw+f7Thn^ZHe9*MD9phw<IyVH<}DWYMZ(
zN=8BpbQ$vYJYA@h@5K&y(&87^ro(p1k-r)RQ^q2LD?Imz$@?vc>!{oaV$LpG4La1e
zA`FWR1HR>vtwt7I14*%(Mx94Wrh!AJy8L_WU=10k*7fc^lb}hucEL~&GisFCHSM3{
zh}JkDk7bNcm>rUJ!YI1H!9TJ&3ivaHTn!^q4odaL)!FcCj3aX5G+*vY%rCjkVPiE=
zt>BS3y5^$*sI20F@7E;94EO_o{Bc0gskg|HFnx#aw=!73U_F231IGwY>wyGf)IzJf
zg(;639r~S2Lx#1@!3{wRQ#Gq~2ZS20?Kf{-mL=~h)I1^fXem`)qq}YH1%_x1W60b#
zstV%fB>>AA?k5CK9dAhbtaJ2bBy@qhWu>P>9O~jQE~n_yHY6!0;iH8*+?A)eW;iC{
z>~`0$m?n~j+N{nrR-c(&ZvRR}OdCW><~Yn}iK@Zw_PT%vJaTLW2C>ML`~WjrP+rX$
zdAs#tekk7cETn44YP$$36s9k)SW6=T^a+I*mL$*(9n=Reqg>-rtt%Pw=cgl~8%_sN
zHSj*MC#mW>!QHFV8rrL%5=ZolQo7r2lxjL;X)m{X)QRCHkti%nw;2QT7hW4IdCLPS
zI3txj2)m+>kDoH}?)%6N;PL{>M|n_Zl9!Kyur8pQ2Ry2(FT4M+tlN40;0mHRLn`Wn
z*{?d!HpqC|EJe_BCiht48FP5g?fm!zI(K3#a^|-D1;f&Do7%E5-t}i6rEO}D)d#dh
zbFCAV>GcxZ?c<aTS4u{!Y=M@{VUUH|xdcR5hC2<C;5m6ir$H=d!Lkf3ayO>3%tKaY
zzJ*M^)@A?eKxbjkzYK3??)qMusZ^D_DlJ8-H1bbL(6!`|<JBPgEe6XQ?QJ!6Jn-IL
zoP7{v{Dlj^LWC>aUA5eG{qg88e{eP|vgF+u+~RgQAu<tVF(5$eu{3HWV^OXHgJ&@_
z6@VSM3)vEV<CMmNz9yIO3>oRQ!CZ$uBA}9ZAGV`wpfD!&YiWnNn2P6oypo)UJq-yC
zDQvvi#I%b#lMQ)Lo`&@PRX6mQ7jD3Qo4WZ6!3erU4%=U0F$8pflKIi5u*?x{;AHn`
zm&QqGpKU>RZL=TSpV2hR8lYe<ILO&grw8j8FV3l-!kEU)OYr|m3aKaH^4P3i4<fC{
zO2Y)oM0>?x1`{!E^zjS2R<lmRp=cDRz}?^ge$(hWfM!XJ(isTbTC2+y(|7nA7<@fM
zcRKU1O-7J}NvksB-Nlx67U|$t7IzzRU`j|o{u{W+90#J`(wl+m%w;0I_PO#?tqkx5
zAi3!_@h16vEVf|0yNL;Wft?8}fFTxPd3h-%S`0%S^+6gc)w*+`@?v`ZJzJlK_8Hja
zP7?wp$C-(?vqY<>^-9<e&5Txn2-tapJjRk<Lz%dYhD^QJWQ!}B6&;}h>>4H@ws9Kr
z0t&DI5wj92wlIcgj)%YWX7JA;Zf-Yb5deO(<xCQ~2NAZZ&x7;<#}_jLN(p(GD9*I*
z<AhAEPJtylEyO<s1IGyuoS5Ttev&;WYSlH(g!qK2OW-*;@<!w(NOl2t^hyFsAIZdw
zj@1qOx9J=zG}tXvrfZVISdb=Z&%;dg)yxIAS<`n^k3oXa32EK1edul5{w$LRgT}u~
z>sSfS$YO+{mv4wxxTp#$IJs3%3LI8^C4smJK0aZ8=mlSIk^VG{##N-VUTZIMCie@M
z==5>+v-1G->1;>stgU$`JheaFZz;Y0j_McE@uBjwD-6o+=FJw$JG~C5sfsB!csTkt
z5oL@llc|$T%rsg%yRO0T3n4t?jqjf={``^Pb@9)?@@x?s@qV+Hf~W)z`%$?^=)<W|
zd6>~7`|^KdFt#u_=I(eZ>h1-11GC!g_G1nG3EQovW;=#CV<APU!GMfhbN;deNqNxb
z3*gXmv1V$y*sO8vUHOePmbC8Cb>OatrJo(TqYV8YwtQ9EfE`dct?N2{Cb2cGxh!yo
z7SvAj&NkbQ&88dCZ`f-j$OrsM3%U9tf7#iKUKVkv{;FTGnRI)86@;SCCBU~5+zYQ_
zb0O@ysY#*_O^1ZT2pXA&+Nm4C!zn)9sOvQrW%OUavQnnw^N|^@O6fugA5f;*w^QAA
zTpCFlPT(pYNP3${$Z1)cPHzI~CHH>cSkv3Gux{vSqi0;no~9%NZWN66>6F#1JqRm=
zjp%~gAc3-z;GHR(P`Z{sMmn}doNd9C)tFCmX1SaP?s9tMOtkmkf4JYoOj;u)Q$J}8
z{KlSXhe4q15>di!N5_?Hd-(nx*N8Dmnbwb5Ly3sz1}V5qR{_U`)N`$_f<95Agc%*@
z*TPRzJtM{W2!$nY;Obx@;gU0aYt*)7a#&0eCd!zth-gHFrwgI80kh#&@uxmK1>I0w
z*MuvB{%ZeWK>;3?^ohV8<ApFxdxL?_cajdGf-a$kDR7)_GA=t}vVKtN_>b&W-j@~@
z6sie?dHDdrU*;Qd2{2l*X4XlTadIc$XUa_ZJ#J&;{iXYo_ku1(gmjQBW|XMa-N@|P
z<FDg%NY-V#v8icyaBU4=n0z)VB`Tdw!^E}dl75_$1iUgTwIZRUfFiOWl$Jr;-9H0M
z?)*I|*u@sbS7!84D(uDpj1-oIoD$*~HiN3(?1($^HUwRV>O2o}&u(?7*XP)6uHw$k
zB@s(&5uh-<5Z1YPPf&b*?KuC*RZ*8<w1UcQ1}ZSfe|GLFs#T4NQJ<dGUxls+8cgy!
zbZttmewo>Bza?eiYs$@R@%1*=V^T=c!GteeN;!9<5MhGTH0=6;^Z{D^+3t^Ng=@|5
zP4_&$#Kx$9OASQNDp~>?l$DNZG$^a*NnS<~2bivS$D)Lc>+koLJEB&0-uFKYFD<~8
z4TU%D)u8JKP6>g~s;+kOz1bEA4wn~Z20R{pMORH5$I>BcU4A4C!au#bR!l+UmP*GZ
z3&bJ^o+d*zzZI5*a;UY6ugIet9v^@^lgEJ0x9xypru?kqN-Wy>MvNdPE0f6JH__c_
z36a-;%N_;8x=)&0BuVD`o}Fpj`wDYUMVz?{a70k;BWV8Nu8wj|P54F2st0Y4K&z*;
z=J|p8U_eEe$F>ueyvSst&ZRdqe>R(VfaksCTMWQ=Oh+bhN2dBsEen(wh))OAONr!U
zkti$z1a4W_zrrwIt4$x)!41IOG0pMr#n2L6iW;hwg)sl%G+|d49CD%)bUTLJ8dm!$
zxbQ~$+&n-vjkeF$z#%tgU`F5k6^_)Pc;13BwQ+r%*o&l;g0NK+{K}NyRSk<x(oG~d
zVagpTtS5+_FK#W(?gz8N3ZsP+E9@j}!t07ZYlBQ5e>1^5B)^Jb!Wk|#ZpTX|;A?`K
zQNQbp6H~b@F=+u}5%SpWPVKT6oCOQuS#~=uUHSMvKX%5&YI=7cR%dF~PZp@=z2Lq7
zQiIv3ZF!m`Ny;2IMt8t76%fXD?L%<V=!|f=&%S5P!!?b6TAWN4?R0%(i<m4YFOBqB
zWk2+B%Fhba0CZa1sTi-bI#2BEJKbYYlcD9~G~6?Qz*Va{&c+vWbq4A)kAm`5%mr=s
z;3W1@z0&Pg;e$ueI>+xy__jGf2lu21R#8a?rM?BYR7M(N(E;A@#qjp&(BK|ZBxtqu
zAX3&k*h$!_Lr_K&YTZhYaPRU!y_mO&ZpgnxufjK@?Kn)uTSajLO2}=<5&U!{t#|?A
zrr_AaPur}zgE&Kh_cAm|iEd}o1GwZzHBKAty4{HE!QR*h>B7;0EmhF&Gx1$f->Lgw
z;ZuFVwjgq-&D;y}FZ@)&9eOk!nOYn&xwl;XX=N%fmJNvwXLtCgC=DCAJw<WM+nbZT
z`X>Ch_d6oBQ_nWswEKbiqQ}xnuf2ssAN;y{VmH})(=yYMqy;FxH=_7gY2Ch!V;TLj
zmrwqU3GDbw7L0gv!h5(8bi?|C3a-4di+K*yu$#tzQO+e&ngqRwBU{A;E2)#z$_h)X
zZc3?1WX7iBxzhTiTd3jEp-579+%6URVVS`X))#@ZzlGvVxY;x6IyO%4Q0aN5TA>;E
zBekH9l_e*?F|}!JOUqTR{Eq5E^U07^yezI<+jUtC=b(ZxS56&n6;)_@GSEA&#M&&-
z{s)fdz@PyLXlUf)1VRT&HLmsj`jg%?9gi_I=#^GKC?CY7B*fNczfihTPnZBh_f<<D
z6@1T*^Ia<W??fWWEk!lBG5L|*r+mcnL+a0!>T!1)=4K|+PfYCbq1HH6ItxzvAJi+H
z=<<(F?MfG;RSDR2t>(^9?g`Jd@d^pFw?!#ZEwBYl)UWxVb=Nhe_qpfr-=py!l1Z{o
zsj&aG1<_wjJa6%JxWidKf`FKS5v<oK{>ZXxJqshD+zQEPxm|>`$A;v3*LU^e`$W-z
z1MAo3{0+YEJXi6w+>Va4JH?lretwU&dC+a|h)uM-HKB0h$H%7eVu=^n67|PzV_Q&x
zhwJkN;Q_0wSFFcc@}49Wvuo^3hbTMZD2@0hVQXyyw}?TlGVhL1i#6}`jr8BE&JT0-
zmGvVC57+Mph5G2Do^P;11(z0WzH(fCf7b=A-^I5AIJU&}<^xHJ2ht-*27!7UYmM?*
zP9$`hD>10xLXJgfr@q;LT~$<G->$4+5AkIYDP55y=y_Kp3&(K&h-`drMM|nszFU8r
zB-ZI|#S>rNqWj+dg*O@QB83zjp7oX8>Z@MfpyrS)Ry>Ko37<|Sy`yU!OWtEM_X&pZ
z^dkq-gE{Ka17e0MLD}D4N+@eBH1)IBXxzOOj6eJuy4Y7?kMD|ylC@^yYt?zM(kPA6
zYc<=#!1D3-gMV&Nf%306OV9C1Tq>h~L6Lv*QSh0a^R6c@KjPJa;&hjaG)>YKW{G96
zQqEz;{5qu$B<!;)B|D$q|KEP1iKRtEiR?4wdrvP~qCaKquYTwd4CWg5@veEgE&6AM
z`I}__(RAe+IT>e@#UZdVOgxHeZ6rO90Fmv9X&}EmR2YpHmqdMas>+IrbO+*M2dY>h
zgX2=NK)#qZxZh2jk<G!DN-1De$~Y+(k#pxehb$!ckcRMn$`Wz5A|+Fg1PxAH+L&kd
zVjbXWU&{WBn+TAjXg+0%8GiZqJVpBHU*=0nl&ezN98lP+J1DZfuVT@r33*(Uh`|`y
z0bOe+Kc)X{BPtTieq<HN3BLuUANY2_fQ+P;v>qvP(!AY>azHJ}{76%Au<z@(iC}ul
z;TYcGiYNo38b|>jL?g*DiNA`zC`w7CHjrhar)r|s&*;no<0+|Cx+#+4)uO2R|56mV
z*h+%TzC-3qkF5JI-LE_(EU&o#)RHxuS8LjikiXjOeko{ye)S*Nd20D;MQad1Dy_#-
z<o~<-H=8y>00V*C9((@y-hW=9+Reon_VP7W-`?_H4LH~V17WloE&Z>7RIXdx?0Ym>
z|8C9PXvweOZoZ4K9+v?+#^pqONyF=8M=$O$llbx>FrB@1-kq)tR04~$+9vbA8$_0t
z=FD{-I^`D;R#w9zRai12FMMA<NByl`7IClJMD1w40m@)1B<H_Fgh%QNDOY;EkYxO`
z?Tey;t0+HMY_9vW6-L*AS(_N%`Hl<bUlHnFZ7@@19X;gy(-kkk8Ko5U-NgMc|5fK1
zmm6x2i`7(7Fj%<N&%j*vXT%f{fu<uZ_g<SX1H}%qMB9wV1K4}V2#bdiU*LU1nnklx
apSPng?5)|4UTVAm{McFV*;%-Q82LY<tWGQd

diff --git a/docs/community/index.md b/docs/community/index.md
index 9519f5b56..517bf59c5 100644
--- a/docs/community/index.md
+++ b/docs/community/index.md
@@ -1,12 +1,9 @@
 # Community
 
-![Belonging](belonging.png)
-
 Outlines exists for a community of users who believe software doesn't need to be complicated. Who share the same passion for Large Language Models but don't want to compromise on robustness. Together, we are bringing these powerful models back to the world of software.
 
 ## Connect on Discord
 
 The Outlines community lives on our Discord server. There you can ask questions, share ideas or just chat with people like you. Don't be a stranger and [join us][discord].
 
-
 [discord]: https://discord.gg/UppQmhEpe8
diff --git a/docs/community/versioning.md b/docs/community/versioning.md
index 023b92537..a6e12f218 100644
--- a/docs/community/versioning.md
+++ b/docs/community/versioning.md
@@ -13,6 +13,10 @@ Each part of the version number (`major.minor.patch`) conveys information about
 - **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatability with existing use of core interfaces.
 - **Patch Releases** address bug fixes and incorporate low-risk changes to improve stability and performance.
 
+!!! note "Breaking Changes"
+
+    Outlines v1.0 introduced several breaking changes to the core interface. See [the migration guide](/user_guide/migration) for more details.
+
 ## Releases
 
 Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/dottxt-ai/outlines/releases).
diff --git a/docs/cookbook/classification.md b/docs/cookbook/classification.md
deleted file mode 100644
index 023e4543a..000000000
--- a/docs/cookbook/classification.md
+++ /dev/null
@@ -1,106 +0,0 @@
-# Classification
-
-Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
-
-
-This tutorial shows how one can implement multi-label classification using Outlines. We will use two functionalities of the library: `generate.choice` and `generate.json`.
-
-As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
-
-```python
-import outlines
-
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
-```
-
-We will use the following prompt template:
-
-```python
-from outlines import Template
-
-
-customer_support = Template.from_string(
-    """You are an experienced customer success manager.
-
-    Given a request from a client, you need to determine when the
-    request is urgent using the label "URGENT" or when it can wait
-    a little with the label "STANDARD".
-
-    # Examples
-
-    Request: "How are you?"
-    Label: STANDARD
-
-    Request: "I need this fixed immediately!"
-    Label: URGENT
-
-    # TASK
-
-    Request: {{ request }}
-    Label: """
-)
-```
-
-## Choosing between multiple choices
-
-Outlines provides a shortcut to do multi-label classification, using the `outlines.generate.choice` function to initialize a generator. Outlines uses multinomial sampling by default, here we will use the greedy sampler to get the label with the highest probability:
-
-```python
-from outlines.samplers import greedy
-
-generator = outlines.generate.choice(model, ["URGENT", "STANDARD"], sampler=greedy())
-```
-Outlines supports batched requests, so we will pass two requests to the model:
-
-```python
-requests = [
-    "My hair is one fire! Please help me!!!",
-    "Just wanted to say hi"
-]
-
-prompts = [customer_support(request) for request in requests]
-```
-
-We can now asks the model to classify the requests:
-
-```python
-labels = generator(prompts)
-print(labels)
-# ['URGENT', 'STANDARD']
-```
-
-Now, you might be in a hurry and don't want to wait until the model finishes completion. After all, you only need to see the first letter of the response to know whether the request is urgent or standard. You can instead stream the response:
-
-```python
-tokens = generator.stream(prompts)
-labels = ["URGENT" if "U" in token else "STANDARD" for token in next(tokens)]
-print(labels)
-# ['URGENT', 'STANDARD']
-```
-
-## Using JSON-structured generation
-
-Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
-
-```python
-from enum import Enum
-from pydantic import BaseModel
-
-
-class Label(str, Enum):
-    urgent = "URGENT"
-    standard = "STANDARD"
-
-
-class Classification(BaseModel):
-    label: Label
-```
-
-and we can use `generate.json` by passing this Pydantic model we just defined, and call the generator:
-
-```python
-generator = outlines.generate.json(model, Classification, sampler=greedy())
-labels = generator(prompts)
-print(labels)
-# [Classification(label=<Label.urgent: 'URGENT'>), Classification(label=<Label.standard: 'STANDARD'>)]
-```
diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
deleted file mode 100644
index b51f8d921..000000000
--- a/docs/cookbook/extract_event_details.py
+++ /dev/null
@@ -1,46 +0,0 @@
-from datetime import datetime
-
-from pydantic import BaseModel, Field
-
-from outlines import generate, models
-
-# Load the model
-model = models.mlxlm("mlx-community/Hermes-3-Llama-3.1-8B-8bit")
-
-
-# Define the event schema using Pydantic
-class Event(BaseModel):
-    title: str = Field(description="title of the event")
-    location: str
-    start: datetime = Field(
-        default=None, description="date of the event if available in iso format"
-    )
-
-
-# Get the current date and time
-now = datetime.now().strftime("%A %d %B %Y and it's %H:%M")
-
-# Define the prompt
-prompt = f"""
-Today's date and time are {now}
-Given a user message, extract information of the event like date and time in iso format, location and title.
-If the given date is relative, think step by step to find the right date.
-Here is the message:
-"""
-
-# Sample message
-message = """Hello Kitty, my grandmother will be here , I think it's better to postpone our
-appointment to review math lessons to next Friday at 2pm at the same place, 3 avenue des tanneurs, I think that one hour will be enough
-see you 😘 """
-
-# Create the generator
-generator = generate.json(model, Event)
-
-# Extract the event information
-event = generator(prompt + message)
-
-# Print the current date and time
-print(f"Today: {now}")
-
-# Print the extracted event information in JSON format
-print(event.json())
diff --git a/docs/core_concepts.md b/docs/core_concepts.md
new file mode 100644
index 000000000..7f53983ec
--- /dev/null
+++ b/docs/core_concepts.md
@@ -0,0 +1,13 @@
+---
+title: Core concepts
+---
+
+# Core concepts
+
+Coming soon. This will document various concepts at a high level, so users can understand Outlines before diving into specific implementations.
+
+1. Constrained decoding, tokens, and the basics of logit biasing
+2. Different ways to define output structure (regex, JSON schema, Pydantic models, context-free grammars)
+3. How finite state machines are used to guarantee output structure
+4. `Generator`, `Application`, `Template`,
+5. Prompt engineering vs. structured generation
diff --git a/docs/cookbook/atomic_caption.md b/docs/examples/atomic_caption.md
similarity index 75%
rename from docs/cookbook/atomic_caption.md
rename to docs/examples/atomic_caption.md
index a2b43e5e2..08782594f 100644
--- a/docs/cookbook/atomic_caption.md
+++ b/docs/examples/atomic_caption.md
@@ -1,5 +1,6 @@
 # Vision-Language Models with Outlines
-This guide demonstrates how to use Outlines with vision-language models, leveraging the new transformers_vision module. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
+
+This guide demonstrates how to use Outlines with vision-language models, leveraging the outlines `TransformersMultiModal` model. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
 
 We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
 
@@ -11,17 +12,21 @@ pip install outlines transformers torch
 ```
 
 ### Initializing the Model
-We'll use the transformers_vision function to initialize our vision-language model. This function is specifically designed to handle models that can process both text and image inputs. Today we'll be using the Pixtral model with the llama tokenizer. (Currently the mistral tokenizer is pending support).
+We'll use the `outlines.from_transformers` function to initialize our vision-language model. For this function to return a vision multi-modal model we need to pass in a transformers model and a transformers processor that can handle both text and image inputs. Today we'll be using the Pixtral model with the AutoProcessor.
 
 ```python
+import outlines
 import torch
 from transformers import (
-    LlavaForConditionalGeneration,
+    AutoProcessor,
+    LlavaForConditionalGeneration
 )
+
 model_name="mistral-community/pixtral-12b" # original magnet model is able to be loaded without issue
 model_class=LlavaForConditionalGeneration
+processor_class=AutoProcessor
 
-def get_vision_model(model_name: str, model_class: VisionModel):
+def get_vision_model(model_name: str, model_class, processor_class):
     model_kwargs = {
         "torch_dtype": torch.bfloat16,
         "attn_implementation": "flash_attention_2",
@@ -31,28 +36,25 @@ def get_vision_model(model_name: str, model_class: VisionModel):
         "device": "cuda",
     }
 
-    model = outlines.models.transformers_vision(
-        model.model_name,
-        model_class=model.model_class,
-        model_kwargs=model_kwargs,
-        processor_kwargs=processor_kwargs,
+    model = outlines.from_transformers(
+        model_class.from_pretrained(model_name, **model_kwargs),
+        processor_class.from_pretrained(model_name, **processor_kwargs),
     )
     return model
-model = get_vision_model(model_name, model_class)
+model = get_vision_model(model_name, model_class, processor_class)
 ```
 
 ### Defining the Schema
-Next, we'll define a schema for the output we expect from our vision-language model. This schema will help structure the model's responses.
-
+Next, we'll define a schema for the output we expect from our vision multi-modal model. This schema will help structure the model's responses. We use the `outlines.Generator` object to create a generator for our schema that will then be called with our prompt and images.
 
 ```python
+from enum import Enum
 from pydantic import BaseModel, Field, confloat, constr
 from pydantic.types import StringConstraints, PositiveFloat
 from typing import List
 from typing_extensions import Annotated
 
-from enum import StrEnum
-class TagType(StrEnum):
+class TagType(Enum):
     ENTITY = "Entity"
     RELATIONSHIP = "Relationship"
     STYLE = "Style"
@@ -86,7 +88,7 @@ class ImageData(BaseModel):
     short_caption: Annotated[str, StringConstraints(min_length=10, max_length=150)]
     dense_caption: Annotated[str, StringConstraints(min_length=100, max_length=2048)]
 
-image_data_generator = outlines.generate.json(model, ImageData)
+image_data_generator = outlines.Generator(model, ImageData)
 ```
 
 This schema defines the structure for image tags, including categories like Entity, Relationship, Style, etc., as well as short and dense captions.
@@ -95,7 +97,6 @@ This schema defines the structure for image tags, including categories like Enti
 
 We'll create a prompt that instructs the model on how to analyze the image and generate the structured output:
 
-
 ```python
 pixtral_instruction = """
 <s>[INST]
@@ -131,30 +132,34 @@ pixtral_instruction = """
 </Examples>
 </TagCategories>
 <ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
-<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG][/INST]
+<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG]<image>[/INST]
 """.strip()
 ```
 
-This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required.
+This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required. It's essential to include the <image> tag in the prompt at the location where the image will be inserted.
 
 ### Generating Structured Output
 Now we can use our model to generate structured output based on an input image:
 
 ```python
+from io import BytesIO
+from urllib.request import urlopen
+from PIL import Image
+
 def img_from_url(url):
     img_byte_stream = BytesIO(urlopen(url).read())
     return Image.open(img_byte_stream).convert("RGB")
 
 image_url="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
 image= img_from_url(image_url)
-result = image_data_generator(
-    pixtral_instruction,
-    [image]
-)
+result = image_data_generator({
+    "text": pixtral_instruction,
+    "images": image
+})
 print(result)
 ```
 
-This code loads an image from a URL, passes it to our vision-language model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
+This code loads an image from a URL, passes it to our vision multi-modal model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
 
 ```json
 {'tags_list': [{'tag': 'astronaut',
@@ -184,6 +189,6 @@ This code loads an image from a URL, passes it to our vision-language model alon
 ```
 
 ## Conclusion
-The transformers_vision module in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
+The `TransformersMultiModal` model in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
 
 By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
diff --git a/docs/cookbook/chain_of_density.md b/docs/examples/chain_of_density.md
similarity index 74%
rename from docs/cookbook/chain_of_density.md
rename to docs/examples/chain_of_density.md
index 10bda9771..8a822fc2e 100644
--- a/docs/cookbook/chain_of_density.md
+++ b/docs/examples/chain_of_density.md
@@ -26,42 +26,12 @@ The prompt also asks the model to return a list of JSON objects that contain the
 
 ![Figure 2 in the paper](./images/chain_of_density.png)
 
-We can now implement the prompt provided in the paper:
+We can now implement the prompt provided in the paper. We stored the prompt template in a text file, and we can load it using the `Template` class:
 
 ```python
 from outlines import Template
 
-
-chain_of_density = Template.from_string(
-    """Article: {{ article }}
-
-    You will generate increasingly concise, entity-dense summaries of the above Article.
-
-    Repeat the following 2 steps 5 times.
-
-    Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
-    Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
-
-    A Missing Entity is:
-    - Relevant: to the main story.
-    - Specific: descriptive yet concise (5 words or fewer).
-    - Novel: not in the previous summary.
-    - Faithful: present in the Article.
-    - Anywhere: located anywhere in the Article.
-
-    Guidelines:
-    - The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
-    - Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
-    - Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
-    - The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
-    - Missing entities can appear anywhere in the new summary.
-    - Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
-
-    Remember, use the exact same number of words for each summary.
-
-    Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
-    """
-)
+chain_of_density = Template.from_file("prompt_templates/chain_of_density.txt")
 ```
 
 ??? Note
@@ -84,19 +54,26 @@ class Summaries(BaseModel):
     summaries: conlist(Summary, max_length=5, min_length=5)
 ```
 
-We now generate the prompt by passing the article we want to summarize to the template. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use JSON-structured generation to generate the summaries:
+We now generate the prompt by passing the article we want to summarize to the prompt template previously loaded. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use the `Summaries` schema to generate the summaries with structured generation:
 
 ```python
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ")
+import outlines
+import transformers
 
-prompt = chain_of_density(article)
-result = outlines.generate.json(model, Summaries)(prompt)
+MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
+
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+)
+prompt = chain_of_density(article=article)
+result = model(prompt, Summaries, max_new_tokens=2000)
 ```
 
 We can now check the results:
 
 ```python
-print(result.model_dump())
+print(result)
 # {'summaries': [
 #     {
 #       'missing_entities': 'English mathematician, cryptanalyst, philosopher',
diff --git a/docs/cookbook/chain_of_thought.md b/docs/examples/chain_of_thought.md
similarity index 63%
rename from docs/cookbook/chain_of_thought.md
rename to docs/examples/chain_of_thought.md
index 5ac18af25..42d6c8550 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/examples/chain_of_thought.md
@@ -11,20 +11,23 @@ We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-
 pip install llama-cpp-python
 ```
 
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
+
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -38,20 +41,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Chain of thought
@@ -77,30 +69,27 @@ class Reasoning(BaseModel):
 json_schema = Reasoning.model_json_schema()
 ```
 
-We could generate a response using the json schema but for a change we will use the regex:
-
-```python
-from outlines.fsm.json_schema import convert_json_schema_to_str
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-schema_str = convert_json_schema_to_str(json_schema=json_schema)
-regex_str = build_regex_from_schema(schema_str)
-```
-
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(user_prompt):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON "
-        f"Here's the json schema you must adhere to:\n<schema>\n{json_schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + user_prompt
-        + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+generate_hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ json_schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    {{ user_prompt }}
+    <|im_end|>
+    <|im_start|>assistant
+    <schema>
+    """
+)
 ```
 
 For a given user prompt:
@@ -109,11 +98,11 @@ For a given user prompt:
 user_prompt = "9.11 and 9.9 -- which is bigger?"
 ```
 
-we can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+We can use `outlines.Generator` with the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-generator = generate.regex(model, regex_str)
-prompt = generate_hermes_prompt(user_prompt)
+generator = outlines.Generator(model, regex_str)
+prompt = generate_hermes_prompt(json_schema=json_schema, user_prompt=user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
diff --git a/docs/examples/classification.md b/docs/examples/classification.md
new file mode 100644
index 000000000..18a933b47
--- /dev/null
+++ b/docs/examples/classification.md
@@ -0,0 +1,85 @@
+# Classification
+
+Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
+
+
+This tutorial shows how one can implement multi-label classification using Outlines.
+
+As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
+
+```python
+import outlines
+import transformers
+
+MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
+
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+)
+```
+
+We will use a prompt template stored in a text file:
+
+```python
+from outlines import Template
+
+customer_support = Template.from_file("prompt_templates/classification.txt")
+```
+
+## Choosing between multiple choices
+
+Outlines provides a convenient way to do multi-label classification, passing a Literal type hint to the `outlines.Generator` object:
+
+```python
+from typing import Literal
+import outlines
+
+generator = outlines.Generator(model, Literal["URGENT", "STANDARD"])
+
+```
+Outlines supports batched requests, so we will pass two requests to the model:
+
+```python
+requests = [
+    "My hair is one fire! Please help me!!!",
+    "Just wanted to say hi"
+]
+
+prompts = [customer_support(request=request) for request in requests]
+```
+
+We can now ask the model to classify the requests:
+
+```python
+labels = generator(prompts)
+print(labels)
+# ['URGENT', 'STANDARD']
+```
+
+## Using JSON-structured generation
+
+Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
+
+```python
+from enum import Enum
+from pydantic import BaseModel
+
+
+class Label(str, Enum):
+    urgent = "URGENT"
+    standard = "STANDARD"
+
+
+class Classification(BaseModel):
+    label: Label
+```
+
+We can then create a generator with the Pydantic model we just defined and call it:
+
+```python
+generator = outlines.Generator(model, Classification)
+labels = generator(prompts)
+print(labels)
+# ['{"label":"URGENT"}', '{ "label": "STANDARD" }']
+```
diff --git a/docs/cookbook/dating_profiles.md b/docs/examples/dating_profiles.md
similarity index 92%
rename from docs/cookbook/dating_profiles.md
rename to docs/examples/dating_profiles.md
index 1cbea556c..5852e6df9 100644
--- a/docs/cookbook/dating_profiles.md
+++ b/docs/examples/dating_profiles.md
@@ -3,6 +3,7 @@
 In this example we will see how we can use Outlines to generate synthetic data for a dating application. This example was originally contributed by [Vibhor Kumar](https://github.com/veezbo).
 
 ```python
+import json
 from dataclasses import dataclass
 from enum import Enum
 
@@ -132,20 +133,21 @@ samples: list[Example] = [
 We will use Mosaic's MPT-7B model (requires 13GB of GPU memory) which can fit on a single GPU with a reasonable context window. We initialize it with Outlines:
 
 ```python
+MODEL_NAME = "mosaicml/mpt-7b-8k-instruct"
+
 config = transformers.AutoConfig.from_pretrained(
-    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
+    MODEL_NAME, trust_remote_code=True
 )
 config.init_device = "meta"
-model = outlines.models.transformers(
-    model_name="mosaicml/mpt-7b-8k-instruct",
-    device="cuda",
-    model_kwargs={
-        "config": config,
-        "trust_remote_code": True,
-        "torch_dtype": torch.bfloat16,
-        "device_map": {"": 0},
-    },
-)
+model_kwargs = {
+    "config": config,
+    "trust_remote_code": True,
+    "torch_dtype": torch.bfloat16,
+    "device_map": "cuda",
+}
+tf_model = transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME, **model_kwargs)
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+model = outlines.from_transformers(tf_model, tokenizer=tf_tokenizer)
 ```
 
 ## JSON-structured generation of profiles
@@ -162,9 +164,9 @@ every few  months. I also like weddings because I get to wear those suits, and
 it's  a good excuse for a date. I watch the latest series because I'm paying,
 with my hard-earned money, for every streaming service."""
 
-prompt = dating_profile_prompt(new_description, samples)
-profile = outlines.generate.json(model, DatingProfile)(prompt)
-parsed_profile = DatingProfile.model_validate_json(profile)
+prompt = dating_profile_prompt(description=new_description, examples=samples)
+profile = model(prompt, DatingProfile)
+parsed_profile = DatingProfile.model_validate_json(json.loads(profile))
 ```
 
 ## Results
diff --git a/docs/cookbook/deploy-using-bentoml.md b/docs/examples/deploy-using-bentoml.md
similarity index 91%
rename from docs/cookbook/deploy-using-bentoml.md
rename to docs/examples/deploy-using-bentoml.md
index 2926df0f6..5d000cb52 100644
--- a/docs/cookbook/deploy-using-bentoml.md
+++ b/docs/examples/deploy-using-bentoml.md
@@ -119,15 +119,22 @@ class Outlines:
     bento_model_ref = bentoml.models.get(BENTO_MODEL_TAG)
 
     def __init__(self) -> None:
-
         import outlines
         import torch
-        self.model = outlines.models.transformers(
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+
+        # Load tokenizer and model from the BentoML model reference path
+        hf_tokenizer = AutoTokenizer.from_pretrained(self.bento_model_ref.path)
+        hf_model = AutoModelForCausalLM.from_pretrained(
             self.bento_model_ref.path,
-            device="cuda",
-            model_kwargs={"torch_dtype": torch.float16},
+            torch_dtype=torch.float16,
+            low_cpu_mem_usage=True,
+            device_map="cuda"
         )
 
+        # Then use the loaded model with Outlines
+        self.model = outlines.from_transformers(hf_model, hf_tokenizer)
+
     ...
 ```
 
@@ -142,13 +149,14 @@ We then need to define an HTTP endpoint using `@bentoml.api` to decorate the met
         prompt: str = "Give me a character description.",
         json_schema: t.Optional[str] = DEFAULT_SCHEMA,
     ) -> t.Dict[str, t.Any]:
-
+        import json
         import outlines
+        from outlines.types import JsonSchema
 
-        generator = outlines.generate.json(self.model, json_schema)
+        generator = outlines.Generator(self.model, JsonSchema(json_schema))
         character = generator(prompt)
 
-        return character
+        return json.loads(character)
 ```
 
 Here `@bentoml.api` decorator defines `generate` as an HTTP endpoint that accepts a JSON request body with two fields: `prompt` and `json_schema` (optional, which allows HTTP clients to provide their own JSON schema). The type hints in the function signature will be used to validate incoming JSON requests. You can define as many HTTP endpoints as you want by using `@bentoml.api` to decorate other methods of `Outlines` class.
diff --git a/docs/cookbook/deploy-using-cerebrium.md b/docs/examples/deploy-using-cerebrium.md
similarity index 90%
rename from docs/cookbook/deploy-using-cerebrium.md
rename to docs/examples/deploy-using-cerebrium.md
index 6fbaab390..bf010d583 100644
--- a/docs/cookbook/deploy-using-cerebrium.md
+++ b/docs/examples/deploy-using-cerebrium.md
@@ -36,7 +36,7 @@ provider = "aws"
 region = "us-east-1"
 
 [cerebrium.dependencies.pip]
-outline = "==0.0.37"
+outline = "==1.0.0"
 transformers = "==4.38.2"
 datasets = "==2.18.0"
 accelerate = "==0.27.2"
@@ -48,9 +48,14 @@ Running code in Cerebrium is like writing normal python with no special syntax.
 
 ```python
 import outlines
+import transformers
+from outlines.types import JsonSchema
 
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
 schema = """{
     "title": "Character",
@@ -89,7 +94,7 @@ schema = """{
     }
 }"""
 
-generator = outlines.generate.json(model, schema)
+generator = outlines.Generator(model, JsonSchema(schema))
 ```
 
 On first deploy, it will download the model and store it on disk therefore for subsequent calls it will load the model from disk.
diff --git a/docs/cookbook/deploy-using-modal.md b/docs/examples/deploy-using-modal.md
similarity index 93%
rename from docs/cookbook/deploy-using-modal.md
rename to docs/examples/deploy-using-modal.md
index 15e200cb5..23a846a6d 100644
--- a/docs/cookbook/deploy-using-modal.md
+++ b/docs/examples/deploy-using-modal.md
@@ -68,7 +68,12 @@ When running longer Modal apps, it's recommended to download your language model
 # downloading models, setting up environment variables, etc.
 def import_model():
     import outlines
-    outlines.models.transformers(language_model)
+    import transformers
+
+    outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(language_model),
+        transformers.AutoTokenizer.from_pretrained(language_model)
+    )
 
 # This line tells the container to run the import_model function when it starts.
 outlines_image = outlines_image.run_function(import_model)
@@ -134,19 +139,22 @@ def generate(
     # so we need to import the necessary libraries here. You should
     # do this with any other libraries you might need.
     import outlines
+    import transformers
+    from outlines.types import JsonSchema
 
     # Load the model into memory. The import_model function above
     # should have already downloaded the model, so this call
     # only loads the model into GPU memory.
-    model = outlines.models.transformers(
-        language_model, device="cuda"
+    outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(language_model, device_map="cuda"),
+        transformers.AutoTokenizer.from_pretrained(language_model)
     )
 
     # Generate a character description based on the prompt.
     # We use the .json generation method -- we provide the
     # - model: the model we loaded above
     # - schema: the JSON schema we defined above
-    generator = outlines.generate.json(model, schema)
+    generator = outlines.Generator(model, JsonSchema(schema))
 
     # Make sure you wrap your prompt in instruction tags ([INST] and [/INST])
     # to indicate that the prompt is an instruction. Instruction tags can vary
diff --git a/docs/cookbook/earnings-reports.md b/docs/examples/earnings-reports.md
similarity index 94%
rename from docs/cookbook/earnings-reports.md
rename to docs/examples/earnings-reports.md
index 26c2a9ccf..cca1e9513 100644
--- a/docs/cookbook/earnings-reports.md
+++ b/docs/examples/earnings-reports.md
@@ -26,16 +26,14 @@ Choose your language model. We'll use Phi-3 mini, which is small enough to run o
 ```python
 import outlines
 import torch
+import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.models.transformers(
-    model_name,
-    device='auto',
-    model_kwargs={
-        # To reduce memory usage, we'll use bfloat16
-        "torch_dtype": torch.bfloat16,
-    },
+tf_model = transformers.AutoModelForCausalLM.from_pretrained(
+    model_name, device_map="cuda", torch_dtype=torch.bfloat16
 )
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
+model = outlines.from_transformers(tf_model, tf_tokenizer)
 ```
 
 ## Set up the data
@@ -156,7 +154,7 @@ year,revenue,operating_income,net_income,basic_earnings_per_share(
 \d{4},((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),(-?\d+(?:\.\d{1,2})?)){,3}
 ```
 
-Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.generate.regex` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
+Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.Generator` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
 
 ## Prompting the model
 
@@ -234,12 +232,12 @@ def extract_financial_data_prompt(columns_to_extract, income_statement):
 
 Now that we have our prompt and regular expression, we can run the model.
 
-Construct our regex extractor function. We'll use a greedy sampler, which samples the most likely next token at each step. It's a simple sampler that is more reproducible than multinomial sampling.
+Construct our regex extractor function.
 
 ```python
-csv_extractor = outlines.generate.regex(
-    model, csv_regex, sampler=outlines.samplers.greedy()
-)
+from outlines.types import Regex
+
+csv_extractor = outlines.Generator(model, Regex(csv_regex))
 ```
 
 Provide the prompt to the model and run it:
@@ -247,7 +245,7 @@ Provide the prompt to the model and run it:
 ```python
 csv_data = csv_extractor(
     extract_financial_data_prompt(columns_to_extract, income_statement),
-    max_tokens=1024,
+    max_new_tokens=1024,
 )
 
 print(csv_data)
diff --git a/docs/cookbook/extract_event_details.md b/docs/examples/extract_event_details.md
similarity index 94%
rename from docs/cookbook/extract_event_details.md
rename to docs/examples/extract_event_details.md
index 0f87d9586..063433ad2 100644
--- a/docs/cookbook/extract_event_details.md
+++ b/docs/examples/extract_event_details.md
@@ -8,7 +8,7 @@ place, 3 avenue des tanneurs, one hour will be enough see you 😘
 ```
 
 Let see how to extract the event details from the message with the MLX
-library dedicated to  Apple Silicon processor  (M series).
+library dedicated to Apple Silicon processor (M series).
 
 ```python
 --8<-- "docs/cookbook/extract_event_details.py"
diff --git a/docs/examples/extract_event_details.py b/docs/examples/extract_event_details.py
new file mode 100644
index 000000000..cd90bc439
--- /dev/null
+++ b/docs/examples/extract_event_details.py
@@ -0,0 +1,54 @@
+from datetime import datetime
+
+from mlx_lm import load
+from pydantic import BaseModel, Field
+
+import outlines
+from outlines import Generator, Template
+
+
+# Load the model
+model = outlines.from_mlxlm(*load("mlx-community/Hermes-3-Llama-3.1-8B-8bit"))
+
+
+# Define the event schema using Pydantic
+class Event(BaseModel):
+    title: str = Field(description="title of the event")
+    location: str
+    start: datetime = Field(
+        default=None, description="date of the event if available in iso format"
+    )
+
+# Load the prompt template from a string
+prompt_template = Template.from_string(
+    """
+    Today's date and time are {{ now }}
+    Given a user message, extract information of the event like date and time in iso format, location and title.
+    If the given date is relative, think step by step to find the right date.
+    Here is the message:
+    {{ message }}
+    """
+)
+
+# Get the current date and time
+now = datetime.now().strftime("%A %d %B %Y and it's %H:%M")
+
+# Sample message
+message = """Hello Kitty, my grandmother will be here, I think it's better to postpone our
+appointment to review math lessons to next Friday at 2pm at the same place, 3 avenue des tanneurs, I think that one hour will be enough
+see you 😘 """
+
+# Create the generator
+generator = Generator(model, Event)
+
+# Create the prompt
+prompt = prompt_template(now=now, message=message)
+
+# Extract the event information
+event = generator(prompt)
+
+# Print the current date and time
+print(f"Today: {now}")
+
+# Print the extracted event information
+print(event)
diff --git a/docs/cookbook/extraction.md b/docs/examples/extraction.md
similarity index 85%
rename from docs/cookbook/extraction.md
rename to docs/examples/extraction.md
index dfcca0cfd..cfb39ca5c 100644
--- a/docs/cookbook/extraction.md
+++ b/docs/examples/extraction.md
@@ -7,9 +7,14 @@ Getting LLMs to output the extracted entities in a structured format can be chal
 As always, we start with initializing the model. We will be using a quantized version of Mistal-7B-v0.1 (we're GPU poor):
 
 ```python
+import transformers
 import outlines
 
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name, device_map="cuda"),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
 ```
 
 And we will be using the following prompt template:
@@ -17,7 +22,6 @@ And we will be using the following prompt template:
 ```python
 from outlines import Template
 
-
 take_order = Template.from_string(
     """You are the owner of a pizza parlor. Customers \
     send you orders from which you need to extract:
@@ -70,14 +74,14 @@ orders = [
     "Hi! I would like to order two pepperonni pizzas and would like them in 30mins.",
     "Is it possible to get 12 margheritas?"
 ]
-prompts = [take_order(order) for order in orders]
+prompts = [take_order(order=order) for order in orders]
 
-generator = outlines.generate.json(model, Order)
+generator = outlines.Generator(model, Order)
 
 results = generator(prompts)
 print(results)
-# [Order(pizza=<Pizza.pepperonni: 'Pepperoni'>, number=2),
-#  Order(pizza=<Pizza.margherita: 'Margherita'>, number=12)]
+# ['{"pizza": "Pepperoni", "number": 2}',
+# '{"pizza": "Margherita", "number": 12}']
 ```
 
 There are several ways you could improve this example:
diff --git a/docs/cookbook/images/chain_of_density.png b/docs/examples/images/chain_of_density.png
similarity index 100%
rename from docs/cookbook/images/chain_of_density.png
rename to docs/examples/images/chain_of_density.png
diff --git a/docs/cookbook/images/coding_structure_diagram.png b/docs/examples/images/coding_structure_diagram.png
similarity index 100%
rename from docs/cookbook/images/coding_structure_diagram.png
rename to docs/examples/images/coding_structure_diagram.png
diff --git a/docs/cookbook/images/knowledge-graph-extraction.png b/docs/examples/images/knowledge-graph-extraction.png
similarity index 100%
rename from docs/cookbook/images/knowledge-graph-extraction.png
rename to docs/examples/images/knowledge-graph-extraction.png
diff --git a/docs/cookbook/images/nvidia-income.png b/docs/examples/images/nvidia-income.png
similarity index 100%
rename from docs/cookbook/images/nvidia-income.png
rename to docs/examples/images/nvidia-income.png
diff --git a/docs/cookbook/images/simtom.png b/docs/examples/images/simtom.png
similarity index 100%
rename from docs/cookbook/images/simtom.png
rename to docs/examples/images/simtom.png
diff --git a/docs/cookbook/images/trader-joes-receipt.jpg b/docs/examples/images/trader-joes-receipt.jpg
similarity index 100%
rename from docs/cookbook/images/trader-joes-receipt.jpg
rename to docs/examples/images/trader-joes-receipt.jpg
diff --git a/docs/cookbook/index.md b/docs/examples/index.md
similarity index 82%
rename from docs/cookbook/index.md
rename to docs/examples/index.md
index c36b98969..0181d137a 100644
--- a/docs/cookbook/index.md
+++ b/docs/examples/index.md
@@ -4,15 +4,23 @@ This part of the documentation provides a few cookbooks that you can browse to g
 
 - [Classification](classification.md): Classify customer requests.
 - [Named Entity Extraction](extraction.md): Extract information from pizza orders.
-- [Dating Profile](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
+- [Dating Profiles](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
 - [Chain Of Density](chain_of_density.md): Summarize documents using chain of density prompting and JSON-structured generation.
 - [Playing Chess](models_playing_chess.md): Make Phi-3 Mini play chess against itself using regex-structured generation.
 - [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
 - [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
+- [Structured Generation Workflow](structured_generation_workflow.md):
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
-- [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
 - [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
-- [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
 - [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
+- [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
+- [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
+- [Extract Events Details](extract_event_details.md):
+
+Run Outlines on the cloud:
+
+- [BentoML](deploy-using-bentoml.md)
+- [Cerebrium](deploy-using-cerebrium.md)
+- [Modal](deploy-using-modal.md)
diff --git a/docs/cookbook/knowledge_graph_extraction.md b/docs/examples/knowledge_graph_extraction.md
similarity index 60%
rename from docs/cookbook/knowledge_graph_extraction.md
rename to docs/examples/knowledge_graph_extraction.md
index 6a6877756..f4b21c84c 100644
--- a/docs/cookbook/knowledge_graph_extraction.md
+++ b/docs/examples/knowledge_graph_extraction.md
@@ -8,20 +8,23 @@ We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama
 pip install llama-cpp-python
 ```
 
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
+
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -35,20 +38,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Knowledge Graph Extraction
@@ -92,17 +84,24 @@ schema = KnowledgeGraph.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(user_prompt):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + user_prompt
-        + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+generate_hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    {{ user_prompt }}
+    <|im_end|>
+    <|im_start|>assistant
+    <schema>
+    """
+)
 ```
 
 For a given user prompt, for example:
@@ -111,27 +110,26 @@ For a given user prompt, for example:
 user_prompt = "Alice loves Bob and she hates Charlie."
 ```
 
-We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-from outlines import generate, models
+from outlines import Generator
 
-model = models.LlamaCpp(llm)
-generator = generate.json(model, KnowledgeGraph)
-prompt = generate_hermes_prompt(user_prompt)
+generator = Generator(model, KnowledgeGraph)
+prompt = generate_hermes_prompt(schema=schema, user_prompt=user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
 We obtain the nodes and edges of the knowledge graph:
 
 ```python
-print(response.nodes)
-print(response.edges)
-# [Node(id=1, label='Alice', property='Person'),
-# Node(id=2, label='Bob', property='Person'),
-# Node(id=3, label='Charlie', property='Person')]
-# [Edge(source=1, target=2, label='love', property='Relationship'),
-# Edge(source=1, target=3, label='hate', property='Relationship')]
+print(response)
+# {"nodes":[{"id":1,"label":"Alice","property":"loves,hates"},
+# {"id":2,"label":"Bob","property":"loved_by"},
+# {"id":3,"label":"Charlie","property":"hated_by"}],
+# "edges":[{"source":1,"target":2,"label":"loves","property":"love"},
+# {"source":1,"target":3,"label":"hates","property":"hate"}]}
+
 ```
 
 ## (Optional) Visualizing the Knowledge Graph
@@ -142,10 +140,10 @@ We can use the [Graphviz library](https://graphviz.readthedocs.io/en/stable/) to
 from graphviz import Digraph
 
 dot = Digraph()
-for node in response.nodes:
-    dot.node(str(node.id), node.label, shape='circle', width='1', height='1')
-for edge in response.edges:
-    dot.edge(str(edge.source), str(edge.target), label=edge.label)
+for node in response["nodes"]:
+    dot.node(str(node["id"]), node["label"], shape='circle', width='1', height='1')
+for edge in response["edges"]:
+    dot.edge(str(edge["source"]), str(edge["target"]), label=edge["label"])
 
 dot.render('knowledge-graph.gv', view=True)
 ```
diff --git a/docs/cookbook/models_playing_chess.md b/docs/examples/models_playing_chess.md
similarity index 77%
rename from docs/cookbook/models_playing_chess.md
rename to docs/examples/models_playing_chess.md
index a0df2eb2d..b8d89958f 100644
--- a/docs/cookbook/models_playing_chess.md
+++ b/docs/examples/models_playing_chess.md
@@ -1,6 +1,6 @@
 # Large language models playing chess
 
-In this example we will make a Phi-2 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
+In this example we will make a Phi-3 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
 
 ## The chessboard
 
@@ -18,29 +18,33 @@ board = chess.Board("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
 
 ## The opponents
 
-Phi-2 will be playing against itself:
+Phi-3 will be playing against itself:
 
 ```python
-from outlines import models
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
+import transformers
+import outlines
+
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
 ```
 
 ## A little help for the language model
 
-To make sure Phi-2 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
+To make sure Phi-3 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
 
 ```python
 import re
+from outlines.types.dsl import either, String
 
 def legal_moves_regex(board):
     """Build a regex that only matches valid moves."""
     legal_moves = list(board.legal_moves)
     legal_modes_str = [board.san(move) for move in legal_moves]
     legal_modes_str = [re.sub(r"[+#]", "", move) for move in legal_modes_str]
-    regex_pattern = "|".join(re.escape(move) for move in legal_modes_str)
-    regex_pattern = f"{regex_pattern}"
+    regex_pattern = either(*[String(move) for move in legal_modes_str])
     return regex_pattern
 ```
 
@@ -58,13 +62,11 @@ We update the prompt at each step so it reflects the state of the board after th
 ## Let's play
 
 ```python
-from outlines import generate
-
 board_state = " "
 turn_number = 0
 while not board.is_game_over():
     regex_pattern = legal_moves_regex(board)
-    structured = generate.regex(model, regex_pattern)(prompt + board_state)
+    structured = model(prompt + board_state, regex_pattern)
     move = board.parse_san(structured)
 
     if turn_number % 2 == 0 :  # It's White's turn
@@ -79,7 +81,7 @@ while not board.is_game_over():
     print(board_state)
 ```
 
-Interestingly enough, Phi-2 hates capturing.
+Interestingly enough, Phi-3 hates capturing.
 
 ```pgn
  e4 e5 1.Nf3 Ne7 3.b4 Nf5 5.Nc3 Ne7 7.Bb5 a6 9.Na4 b6 11.c3 Nec6 13.c4 a5 15.d4 Qg5 17.Nd2 Bb7 19.dxe5
diff --git a/docs/examples/prompt_templates/chain_of_density.txt b/docs/examples/prompt_templates/chain_of_density.txt
new file mode 100644
index 000000000..c4f9f55fa
--- /dev/null
+++ b/docs/examples/prompt_templates/chain_of_density.txt
@@ -0,0 +1,27 @@
+Article: {{ article }}
+
+You will generate increasingly concise, entity-dense summaries of the above Article.
+
+Repeat the following 2 steps 5 times.
+
+Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
+Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
+
+A Missing Entity is:
+- Relevant: to the main story.
+- Specific: descriptive yet concise (5 words or fewer).
+- Novel: not in the previous summary.
+- Faithful: present in the Article.
+- Anywhere: located anywhere in the Article.
+
+Guidelines:
+- The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
+- Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
+- Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
+- The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
+- Missing entities can appear anywhere in the new summary.
+- Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
+
+Remember, use the exact same number of words for each summary.
+
+Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
diff --git a/docs/examples/prompt_templates/classification.txt b/docs/examples/prompt_templates/classification.txt
new file mode 100644
index 000000000..e260a6771
--- /dev/null
+++ b/docs/examples/prompt_templates/classification.txt
@@ -0,0 +1,18 @@
+You are an experienced customer success manager.
+
+Given a request from a client, you need to determine when the
+request is urgent using the label "URGENT" or when it can wait
+a little with the label "STANDARD".
+
+# Examples
+
+Request: "How are you?"
+Label: STANDARD
+
+Request: "I need this fixed immediately!"
+Label: URGENT
+
+# TASK
+
+Request: {{ request }}
+Label:
diff --git a/docs/examples/prompt_templates/react_agent.txt b/docs/examples/prompt_templates/react_agent.txt
new file mode 100644
index 000000000..09583f42f
--- /dev/null
+++ b/docs/examples/prompt_templates/react_agent.txt
@@ -0,0 +1,27 @@
+<|im_start|>system
+You are a world class AI model who answers questions in JSON with correct Pydantic schema.
+Here's the json schema you must adhere to:
+<schema>
+{{ schema }}
+</schema>
+Today is {{ today }}
+You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation.
+At the end of the loop you output a Final Answer.
+Use Scratchpad to store the information from the Observation useful to answer the question
+Use Thought to describe your thoughts about the question you have been asked and reflect carefully about the Observation if it exists.
+Use Action to run one of the actions available to you.
+Use Action Input to input the arguments of the selected action - then return PAUSE.
+Observation will be the result of running those actions.
+Your available actions are:
+calculate:
+e.g. calulate: 4**2 / 3
+Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary
+wikipedia:
+e.g. wikipedia: Django
+Returns a summary from searching Wikipedia
+DO NOT TRY TO GUESS THE ANSWER. Begin!
+<|im_end|>
+<|im_start|>user
+{{ question }}
+<|im_end|>
+<|im_start|>assistant
diff --git a/docs/examples/prompt_templates/simtom_prospective_taking.txt b/docs/examples/prompt_templates/simtom_prospective_taking.txt
new file mode 100644
index 000000000..8e937ce60
--- /dev/null
+++ b/docs/examples/prompt_templates/simtom_prospective_taking.txt
@@ -0,0 +1,10 @@
+<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
+Your job is to output only the events that the specified character, {{character}}, knows about.
+
+Here are a few rules:
+1. A character knows about all events that they do.
+2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
+3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
+
+Story: {{story}}
+What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]
diff --git a/docs/examples/prompt_templates/simtom_simulation.txt b/docs/examples/prompt_templates/simtom_simulation.txt
new file mode 100644
index 000000000..ede934783
--- /dev/null
+++ b/docs/examples/prompt_templates/simtom_simulation.txt
@@ -0,0 +1,7 @@
+<s>[INST] {% for event in events %}
+{{event}}
+{% endfor %}
+You are {{name}}.
+Based on the above information, answer the following question:
+{{question}}
+You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]
diff --git a/docs/cookbook/qa-with-citations.md b/docs/examples/qa-with-citations.md
similarity index 68%
rename from docs/cookbook/qa-with-citations.md
rename to docs/examples/qa-with-citations.md
index 79a2214c3..213942d02 100644
--- a/docs/cookbook/qa-with-citations.md
+++ b/docs/examples/qa-with-citations.md
@@ -11,17 +11,19 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -35,20 +37,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Generate Synthetic Data
@@ -74,26 +65,28 @@ class Users(BaseModel):
     users: List[UserDetail]
 ```
 
-We can use a `generate.json` by passing this Pydantic class we just defined, and call the generator:
+We can use a `outlines.Generator` by passing this Pydantic class we just defined, and call the generator:
 
 ```python
-model = models.LlamaCpp(llm)
-generator = generate.json(model, Users)
+import json
+
+generator = outlines.Generator(model, Users)
 response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
-print(response.users)
-# [UserDetail(id=1, first_name='John', last_name='Doe', age=25),
-# UserDetail(id=2, first_name='Jane', last_name='Doe', age=30),
-# UserDetail(id=3, first_name='Bob', last_name='Smith', age=40),
-# UserDetail(id=4, first_name='Alice', last_name='Smith', age=35),
-# UserDetail(id=5, first_name='John', last_name='Smith', age=20)]
+response = json.loads(response)
+print(response['users'])
+# [{'id': 1, 'first_name': 'John', 'last_name': 'Doe', 'age': 25},
+# {'id': 2, 'first_name': 'Jane', 'last_name': 'Doe', 'age': 30},
+# {'id': 3, 'first_name': 'Bob', 'last_name': 'Smith', 'age': 40},
+# {'id': 4, 'first_name': 'Alice', 'last_name': 'Smith', 'age': 35},
+# {'id': 5, 'first_name': 'John', 'last_name': 'Smith', 'age': 20}]
 ```
 
 ```python
-for user in response.users:
-    print(user.first_name)
-    print(user.last_name)
-    print(user.age)
-    print(#####)
+for user in response['users']:
+    print(user['first_name'])
+    print(user['last_name'])
+    print(user['age'])
+    print("#####")
 # John
 # Doe
 # 25
@@ -135,23 +128,30 @@ schema = QuestionAnswer.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(question, context, schema=schema):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON with correct and exact citations "
-        "extracted from the `Context`. "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + "`Context`: "
-        + context
-        + "\n`Question`: "
-        + question + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON with correct and exact citations
+    extracted from the `Context`.
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    `Context`:
+    {{ context }}
+    `Question`:
+    {{ question }}
+    <|im_end|>
+    <|im_start|>assistant
+    """
+)
 ```
 
-We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
+We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
 
 ```python
 question = "What did the author do during college?"
@@ -161,11 +161,11 @@ I went to an arts high school but in university I studied Computational Mathemat
 As part of coop I worked at many companies including Stitchfix, Facebook.
 I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.
 """
-generator = generate.json(model, QuestionAnswer)
-prompt = generate_hermes_prompt(question, context)
+generator = outlines.Generator(model, QuestionAnswer)
+prompt = hermes_prompt(question=question, context=context, schema=schema)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 print(response)
-# QuestionAnswer(question='What did the author do during college?', answer='The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
+# {"question": "What did the author do during college?", "answer": "The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.", "citations": ["I went to an arts high school but in university I studied Computational Mathematics and physics.", "I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years."]}
 ```
 
 We can do the same for a list of question-context pairs:
@@ -219,12 +219,13 @@ for question, context in [
     (question4, context4),
     (question5, context5),
 ]:
-    final_prompt = my_final_prompt(question, context)
-    generator = generate.json(model, QuestionAnswer)
-    response = generator(final_prompt, max_tokens=1024, temperature=0, seed=42)
-    display(question)
-    display(response.answer)
-    display(response.citations)
+    prompt = hermes_prompt(question=question, context=context, schema=schema)
+    generator = outlines.Generator(model, QuestionAnswer)
+    response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+    response = json.loads(response)
+    print(question)
+    print(response['answer'])
+    print(response['citations'])
     print("\n\n")
 
 # 'Where was John born?'
diff --git a/docs/cookbook/react_agent.md b/docs/examples/react_agent.md
similarity index 71%
rename from docs/cookbook/react_agent.md
rename to docs/examples/react_agent.md
index 15b6ed369..bd3d51d6c 100644
--- a/docs/cookbook/react_agent.md
+++ b/docs/examples/react_agent.md
@@ -15,17 +15,19 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -39,20 +41,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Build a ReAct agent
@@ -115,51 +106,16 @@ from typing import Union
 
 class Decision(BaseModel):
     Decision: Union[Reason_and_Act, Final_Answer]
-```
-
-We could generate a response using the json schema but we will use the regex and check that everything is working as expected:
-
-```python
-from outlines.fsm.json_schema import convert_json_schema_to_str
-from outlines_core.fsm.json_schema import build_regex_from_schema
 
 json_schema = Decision.model_json_schema()
-schema_str = convert_json_schema_to_str(json_schema=json_schema)
-regex_str = build_regex_from_schema(schema_str)
-print(regex_str)
-# '\\{[ ]?"Decision"[ ]?:[ ]?(\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Thought"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Action"[ ]?:[ ]?("wikipedia"|"calculate")[ ]?,[ ]?"Action_Input"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\}|\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Final_Answer"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\})[ ]?\\}'
 ```
 
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic:
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic. We can load a template from a file for that:
 
 ```python
-import datetime
-
-def generate_hermes_prompt(question, schema=""):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON with correct Pydantic schema. "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema>\n"
-        "Today is " + datetime.datetime.today().strftime('%Y-%m-%d') + ".\n" +
-        "You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation. "
-        "At the end of the loop you output a Final Answer. "
-        "Use Scratchpad to store the information from the Observation useful to answer the question "
-        "Use Thought to describe your thoughts about the question you have been asked "
-        "and reflect carefully about the Observation if it exists. "
-        "Use Action to run one of the actions available to you. "
-        "Use Action Input to input the arguments of the selected action - then return PAUSE. "
-        "Observation will be the result of running those actions. "
-        "Your available actions are:\n"
-        "calculate:\n"
-        "e.g. calulate: 4**2 / 3\n"
-        "Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary\n"
-        "wikipedia:\n"
-        "e.g. wikipedia: Django\n"
-        "Returns a summary from searching Wikipedia\n"
-        "DO NOT TRY TO GUESS THE ANSWER. Begin! <|im_end|>"
-        "\n<|im_start|>user\n" + question + "<|im_end|>"
-        "\n<|im_start|>assistant\n"
-    )
+from outlines import Template
+
+hermes_prompt = Template.from_file("prompt_templates/react_agent.txt")
 ```
 
 We define a ChatBot class
@@ -175,7 +131,7 @@ class ChatBot:
         return result
 
     def execute(self):
-        generator = generate.regex(model, regex_str)
+        generator = outlines.Generator(model, Decision)
         result = generator(self.prompt, max_tokens=1024, temperature=0, seed=42)
         return result
 ```
@@ -194,7 +150,11 @@ def query(question, max_turns=5):
     previous_actions = []
     while i < max_turns:
         i += 1
-        prompt = generate_hermes_prompt(question=question, schema=Decision.model_json_schema())
+        prompt = generate_hermes_prompt(
+            question=question,
+            schema=Decision.model_json_schema(),
+            today=datetime.datetime.today().strftime('%Y-%m-%d')
+        )
         bot = ChatBot(prompt=prompt)
         result = bot(next_prompt)
         json_result = json.loads(result)['Decision']
diff --git a/docs/cookbook/read-pdfs.md b/docs/examples/read-pdfs.md
similarity index 79%
rename from docs/cookbook/read-pdfs.md
rename to docs/examples/read-pdfs.md
index dbe4ccb02..1854aba42 100644
--- a/docs/cookbook/read-pdfs.md
+++ b/docs/examples/read-pdfs.md
@@ -46,17 +46,19 @@ We've tested this example with [Pixtral 12b](https://huggingface.co/mistral-comm
 To use Pixtral:
 
 ```python
-from transformers import LlavaForConditionalGeneration
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
+processor_class = LlavaProcessor
 ```
 
 To use Qwen-2-VL:
 
 ```python
-from transformers import Qwen2VLForConditionalGeneration
+from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
+processor_class = AutoProcessor
 ```
 
 You can load your model into memory with:
@@ -64,17 +66,12 @@ You can load your model into memory with:
 ```python
 # This loads the model into memory. On your first run,
 # it will have to download the model, which might take a while.
-model = outlines.models.transformers_vision(
-    model_name,
-    model_class=model_class,
-    model_kwargs={
-        "device_map": "auto",
-        "torch_dtype": torch.bfloat16,
-    },
-    processor_kwargs={
-        "device": "auto",
-    },
-)
+model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
+processor_kwargs={"device_map": "cpu"}
+tf_model = model_class.from_pretrained(model_name, **model_kwargs)
+tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+model = outlines.from_transformers(tf_model, tf_processor)
 ```
 
 ## Convert the PDF to images
@@ -159,7 +156,7 @@ The structured output you can extract is exactly the same as everywhere else in
 
 Suppose you wished to go through each page of the PDF, and extract the page description, key takeaways, and page number.
 
-You can do this by defining a JSON schema, and then using `outlines.generate.json` to extract the data.
+You can do this by defining a JSON schema, and then using `outlines.Generator` to extract the data.
 
 First, define the structure you want to extract:
 
@@ -170,7 +167,7 @@ class PageSummary(BaseModel):
     page_number: int
 ```
 
-Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers `AutoProcessor` to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
+Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers processor to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
 
 Images are denoted with `type: "image"`, and text is denoted with `type: "text"`.
 
@@ -191,16 +188,15 @@ messages = [
                 """
             },
 
-            # Don't need to pass in an image, since we do this
-            # when we call the generator function down below.
+            # This a placeholder, the actual image is passed in when
+            # we call the generator function down below.
             {"type": "image", "image": ""},
         ],
     }
 ]
 
 # Convert the messages to the final prompt
-processor = AutoProcessor.from_pretrained(model_name)
-instruction = processor.apply_chat_template(
+prompt = tf_processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -209,10 +205,10 @@ Now we iterate through each image, and extract the structured information:
 
 ```python
 # Page summarizer function
-page_summary_generator = outlines.generate.json(model, PageSummary)
+page_summary_generator = outlines.Generator(model, PageSummary)
 
 for image in images:
-    result = page_summary_generator(instruction, [image])
+    result = page_summary_generator({"text": prompt, "images": image})
     print(result)
 ```
 
@@ -225,19 +221,21 @@ arXiv identifiers are optionally followed by a version number, i.e. `arXiv:YYMM.
 We can use a regular expression to define this patter:
 
 ```python
-paper_regex = r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?'
+from outlines.types import Regex
+
+paper_regex = Regex(r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?')
 ```
 
 We can build an extractor function from the regex:
 
 ```python
-id_extractor = outlines.generate.regex(model, paper_regex)
+id_extractor = outlines.Generator(model, paper_regex)
 ```
 
 Now, we can extract the arxiv paper identifier from the first image:
 
 ```python
-arxiv_instruction = processor.apply_chat_template(
+arxiv_instruction = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -256,7 +254,7 @@ arxiv_instruction = processor.apply_chat_template(
 )
 
 # Extract the arxiv paper identifier
-paper_id = id_extractor(arxiv_instruction, [images[0]])
+paper_id = id_extractor({"text": arxiv_instruction, "images": images[0]})
 ```
 
 As of the time of this writing, the arxiv paper identifier is
@@ -269,7 +267,9 @@ Your version number may be different, but the part before `vX` should match.
 
 ### Categorize the paper into one of several categories
 
-`outlines.generate.choice` allows the model to select one of several options. Suppose we wanted to categorize the paper into being about "language models", "economics", "structured generation", or "other".
+`outlines.Generator` also allows the model to select one of several options by providing a Literal type hint with the categories.
+
+Suppose we wanted to categorize the paper into being about "language models", "cell biology", or "other". We would then define the output type as `Literal["llms", "cell biology", "other"]`.
 
 Let's define a few categories we might be interested in:
 
@@ -284,7 +284,7 @@ categories = [
 Now we can construct the prompt:
 
 ```python
-categorization_instruction = processor.apply_chat_template(
+categorization_instruction = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -310,14 +310,13 @@ categorization_instruction = processor.apply_chat_template(
 Now we can show the model the first page and extract the category:
 
 ```python
+from typing import Literal
+
 # Build the choice extractor
-categorizer = outlines.generate.choice(
-    model,
-    categories
-)
+categorizer = outlines.Generator(model, Literal["llms", "cell biology", "other"])
 
 # Categorize the paper
-category = categorizer(categorization_instruction, [images[0]])
+category = categorizer({"text": categorization_instruction, "images": images[0]})
 print(category)
 ```
 
@@ -332,12 +331,12 @@ llms
 You can provide multiple images to the model by
 
 1. Adding additional image messages
-2. Providing a list of images to the `generate` function
+2. Providing a list of images to the generator
 
 For example, to have two images, you can do:
 
 ```python
-two_image_prompt = processor.apply_chat_template(
+two_image_prompt = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -355,17 +354,9 @@ two_image_prompt = processor.apply_chat_template(
 )
 
 # Pass two images to the model
-generator = outlines.generate.choice(
-    model,
-    ["hot dog", "not hot dog"]
-)
+generator = outlines.Generator(model, Literal["hot dog", "not hot dog"])
 
-result = generator(
-    two_image_prompt,
-
-    # Pass two images to the model
-    [images[0], images[1]]
-)
+result = generator({"text": two_image_prompt, "images": [images[0], images[1]]})
 print(result)
 ```
 
diff --git a/docs/cookbook/receipt-digitization.md b/docs/examples/receipt-digitization.md
similarity index 71%
rename from docs/cookbook/receipt-digitization.md
rename to docs/examples/receipt-digitization.md
index 0fa256db7..8ecb07969 100644
--- a/docs/cookbook/receipt-digitization.md
+++ b/docs/examples/receipt-digitization.md
@@ -38,18 +38,20 @@ If you want to use Qwen-2-VL, you can do the following:
 
 ```python
 # To use Qwen-2-VL:
-from transformers import Qwen2VLForConditionalGeneration
+from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
+processor_class = AutoProcessor
 ```
 
 If you want to use Pixtral, you can do the following:
 
 ```python
 # To use Pixtral:
-from transformers import LlavaForConditionalGeneration
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
+processor_class = LlavaProcessor
 ```
 
 ## Load the model
@@ -57,17 +59,12 @@ model_class=LlavaForConditionalGeneration
 Load the model into memory:
 
 ```python
-model = outlines.models.transformers_vision(
-    model_name,
-    model_class=model_class,
-    model_kwargs={
-        "device_map": "auto",
-        "torch_dtype": torch.bfloat16,
-    },
-    processor_kwargs={
-        "device": "cuda", # set to "cpu" if you don't have a GPU
-    },
-)
+model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
+processor_kwargs={"device_map": "cuda"}
+tf_model = model_class.from_pretrained(model_name, **model_kwargs)
+tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+model = outlines.from_transformers(tf_model, tf_processor)
 ```
 
 ## Image processing
@@ -166,7 +163,7 @@ class ReceiptSummary(BaseModel):
 
 ## Prepare the prompt
 
-We'll use the `AutoProcessor` to convert the image and the text prompt into a format that the model can understand. Practically,
+We'll use the `tf_processor` to convert the image and the text prompt into a format that the model can understand. Practically,
 this is the code that adds user, system, assistant, and image tokens to the prompt.
 
 ```python
@@ -194,8 +191,7 @@ messages = [
 ]
 
 # Convert the messages to the final prompt
-processor = AutoProcessor.from_pretrained(model_name)
-prompt = processor.apply_chat_template(
+prompt = tf_processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -222,17 +218,13 @@ Return the information in the following JSON schema:
 
 ```python
 # Prepare a function to process receipts
-receipt_summary_generator = outlines.generate.json(
-    model,
-    ReceiptSummary,
-
-    # Greedy sampling is a good idea for numeric
-    # data extraction -- no randomness.
-    sampler=outlines.samplers.greedy()
-)
+receipt_summary_generator = outlines.Generator(model, ReceiptSummary)
 
 # Generate the receipt summary
-result = receipt_summary_generator(prompt, [image])
+result = receipt_summary_generator(
+    {"text": prompt, "images": image},
+    max_new_tokens=1024
+)
 print(result)
 ```
 
@@ -241,32 +233,31 @@ print(result)
 The output should look like this:
 
 ```
-ReceiptSummary(
-    store_name="Trader Joe's",
-    store_address='401 Bay Street, San Francisco, CA 94133',
-    store_number=0,
-    items=[
-        Item(name='BANANA EACH', quantity=7, price_per_unit=0.23, total_price=1.61),
-        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CARAMEL CASHEW', quantity=2, price_per_unit=2.29, total_price=4.58),
-        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='SPINDRIFT ORANGE MANGO 8', quantity=1, price_per_unit=7.49, total_price=7.49),
-        Item(name='Bottle Deposit', quantity=8, price_per_unit=0.05, total_price=0.4),
-        Item(name='MILK ORGANIC GALLON WHOL', quantity=1, price_per_unit=6.79, total_price=6.79),
-        Item(name='CLASSIC GREEK SALAD', quantity=1, price_per_unit=3.49, total_price=3.49),
-        Item(name='COBB SALAD', quantity=1, price_per_unit=5.99, total_price=5.99),
-        Item(name='PEPPER BELL RED XL EACH', quantity=1, price_per_unit=1.29, total_price=1.29),
-        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25),
-        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25)
-    ],
-    tax=0.68,
-    total=41.98,
-    date='2023-11-04',
-    payment_method='debit',
-
-)
+{
+  "store_name": "Trader Joe's",
+  "store_address": "401 Bay Street, San Francisco, CA 94133",
+  "store_number": 0,
+  "items": [
+    {"name": "BANANA EACH", "quantity": 7, "price_per_unit": 0.23, "total_price": 1.61},
+    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CARAMEL CASHEW", "quantity": 2, "price_per_unit": 2.29, "total_price": 4.58},
+    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "SPINDRIFT ORANGE MANGO 8", "quantity": 1, "price_per_unit": 7.49, "total_price": 7.49},
+    {"name": "Bottle Deposit", "quantity": 8, "price_per_unit": 0.05, "total_price": 0.4},
+    {"name": "MILK ORGANIC GALLON WHOL", "quantity": 1,"price_per_unit": 6.79,"total_price": 6.79},
+    {"name": "CLASSIC GREEK SALAD", "quantity": 1, "price_per_unit": 3.49, "total_price": 3.49},
+    {"name": "COBB SALAD", "quantity": 1, "price_per_unit": 5.99, "total_price": 5.99},
+    {"name": "PEPPER BELL RED XL EACH", "quantity": 1, "price_per_unit": 1.29, "total_price": 1.29},
+    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
+    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
+  ],
+  "tax": 0.68,
+  "total": 41.98,
+  "date": "2023-11-04",
+  "payment_method": "debit"
+}
 ```
 
 Voila! You've successfully extracted information from a receipt using an LLM.
@@ -284,13 +275,13 @@ class ReceiptSummary(BaseModel):
 which gives you a result like
 
 ```
-ReceiptSummary(
+{
     ...
-    roast="You must be a fan of Trader Joe's because you bought enough
+    "roast": "You must be a fan of Trader Joe's because you bought enough
     items to fill a small grocery bag and still had to pay for a bag fee.
     Maybe you should start using reusable bags to save some money and the
     environment."
-)
+}
 ```
 
 Qwen is not particularly funny, but worth a shot.
diff --git a/docs/cookbook/simtom.md b/docs/examples/simtom.md
similarity index 68%
rename from docs/cookbook/simtom.md
rename to docs/examples/simtom.md
index a80ebc24f..53820258e 100644
--- a/docs/cookbook/simtom.md
+++ b/docs/examples/simtom.md
@@ -17,7 +17,7 @@ SimToM calls an LLM with two consecutive prompts:
 
 To implement SimToM with Outlines, we will need to:
 
-1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
+1. Write the prompts with [prompt templates](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
 2. Define the JSON object each prompt will return using Pydantic.
 3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/latest/reference/models/transformers/).
 
@@ -30,29 +30,8 @@ The authors have shared their code, prompts and data in [this GitHub repository]
 ```python
 from outlines import Template
 
-
-perspective_taking = Template.from_string(
-    """<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
-    Your job is to output only the events that the specified character, {{character}}, knows about.
-
-    Here are a few rules:
-    1. A character knows about all events that they do.
-    2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
-    3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
-
-    Story: {{story}}
-    What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]""" # noqa
-)
-
-simulation = Template.from_string(
-    """<s>[INST] {% for event in events %}
-    {{event}}
-    {% endfor %}
-    You are {{name}}.
-    Based on the above information, answer the following question:
-    {{question}}
-    You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]""" # noqa
-)
+perspective_taking = Template.from_file("prompt_templates/simtom_prospective_taking.txt")
+simulation = Template.from_file("prompt_templates/simtom_simulation.txt")
 ```
 
 ### JSON Structured Generation
@@ -65,13 +44,11 @@ We will need two Pydantic models for SimToM, one for each prompt:
 from pydantic import BaseModel, Field
 from typing import List
 
-
 class PerspectiveTaking(BaseModel):
     """This is for the first prompt."""
     character: str = Field(description="The character we extract the events for.")
     events: List[str] = Field(description="All events that the character knows about.")
 
-
 class Simulation(BaseModel):
     """This is for the second prompt."""
     answer: str
@@ -97,30 +74,37 @@ character = "Aria"
 We load `Mistral-7B-Instruct-v0.3`, create the prompt using the template we defined earlier, and generate a structured response. As a reminder, the goal of the first call is to get all the events a character, `Aria`, knows about.
 
 ```python
+import transformers
+import outlines
 # Load an LLM from Hugging Face
 MODEL_NAME = "mistral-community/Mistral-7B-Instruct-v0.3"
-model = outlines.models.transformers(MODEL_NAME, device="cuda")
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME),
+)
 
 perspective_prompt = perspective_taking(story=story, character=character)
 
 # Call Mistral 7B with the first prompt
-generator = outlines.generate.json(model, PerspectiveTaking)
-perspective = generator(perspective_prompt)
+generator = outlines.Generator(model, PerspectiveTaking)
+perspective = generator(perspective_prompt, max_new_tokens=1024)
 
-print(perspective.model_dump())
+print(perspective)
 # {'character': 'Aria', 'events': ['1 Aria entered the front_yard.', '3 The grapefruit is in the green_bucket.', '4 Aria moved the grapefruit to the blue_container.']}
 ```
 
 Not bad! We will now generate the second prompt with those events.
 
 ```python
-sim_prompt = simulation(events=perspective.events, name=character, question=question)
+import json
+
+sim_prompt = simulation(events=json.loads(perspective)["events"], name=character, question=question)
 
 # Call Mistral 7B with the second prompt
-generator = outlines.generate.json(model, Simulation)
-result = generator(sim_prompt)
+generator = outlines.Generator(model, Simulation)
+result = generator(sim_prompt, max_new_tokens=1024)
 
-print(result.model_dump())
+print(result)
 # {'answer': 'green_bucket'}
 ```
 
diff --git a/docs/cookbook/structured_generation_workflow.md b/docs/examples/structured_generation_workflow.md
similarity index 79%
rename from docs/cookbook/structured_generation_workflow.md
rename to docs/examples/structured_generation_workflow.md
index b9d7c28a4..524283f84 100644
--- a/docs/cookbook/structured_generation_workflow.md
+++ b/docs/examples/structured_generation_workflow.md
@@ -2,8 +2,7 @@
 
 This is a condensed version of [Coding for Structured Generation with LLMs](https://blog.dottxt.co/coding-for-structured-generation.html).
 
-For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this
-with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
+For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
 
 ## Unstructured approach
 
@@ -11,36 +10,30 @@ Before diving into how to use structure generation for this task let's start wit
 
 ```python
 import outlines
+import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.models.transformers(model_name)
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name)
+)
 ```
 
 Next we need a prompt for this model. Since we're focusing on structured generation, we won't be engaging in any form of "prompt hacking" and will be leaving this prompt untouched for the rest of this example.
 
 ```python
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-
-messages_phone = [
-            {"role": "user", "content": """
-            Please generate a realistic phone number for Washington State in the following format
-
-            (555) 555-5555
-
-            """}
-]
-
-# This allows us to properly format our prompt for
-# Phi-3 Mini's 'Instruct' interface.
-prompt_phone = tokenizer.apply_chat_template(messages_phone, tokenize=False)
+prompt_phone = """
+    Please generate a realistic phone number for Washington State in the following format
+    (555) 555-5555
+"""
 ```
 
 With our prompt ready we can now generate 10 example phone numbers
 
 ```python
-phone_generator_unstruct = outlines.generate.text(model)
-for _ in range(10):
-    print(phone_generator_unstruct(prompt_phone,max_tokens=12))
+phone_generator_unstruct = outlines.Generator(model)
+for _ in range(3):
+    print(phone_generator_unstruct(prompt_phone, max_new_tokens=12))
 ```
 
 > I'd be happy to help you generate a realistic phone\
@@ -81,7 +74,9 @@ For a simple example like this, we'll just be using a single phone number, for m
 The next step in the process is for use to define a simple regex that we feel correctly models our real data.
 
 ```python
-phone_regex_1 = r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}'
+from outlines.types import Regex
+
+phone_regex_1 = Regex(r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}')
 ```
 
 Next we need to validate this regex against our real data.
@@ -94,8 +89,8 @@ We'll start with a simple method of validation: just checking that our regex mat
 
 ```
 import re
-re.match(phone_regex_1, phone_number)
 
+re.match(phone_regex_1.pattern, phone_number)
 # <re.Match object; span=(0, 14), match='(206) 386-4636'>
 
 ```
@@ -107,8 +102,9 @@ Now that we have a match, we can move on to generating structured output!
 We're ready to see if structured generation can make an improvement over our initial unstructured approach:
 
 ```python
-phone_generator_v1 = outlines.generate.regex(model, phone_regex_1)
-for _ in range(10):
+phone_generator_v1 = outlines.Generator(model, phone_regex_1)
+
+for _ in range(3):
     print(phone_generator_v1(prompt_phone))
 ```
 > (206) 555-1234\
@@ -135,21 +131,21 @@ We've walked through the loop once, so we can go quickly now through each iterat
 We start by improving our structure:
 
 ```python
-phone_regex_2 = r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}'
+phone_regex_2 = Regex(r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}')
 ```
 
 Before rushing to another round of generation, let's validate this new regex. We'll add just a bit more sophistication over our last check:
 
 ```python
-re.match(phone_regex_2, phone_number)[0] == phone_number
+re.match(phone_regex_2.pattern, phone_number)[0] == phone_number
 # True
 ```
 Now that we've validated, let's generate with this new regex!
 
 ```python
-phone_generator_v2 = outlines.generate.regex(model,
-                                             phone_regex_2)
-for _ in range(10):
+phone_generator_v2 = outlines.Generator(model, phone_regex_2)
+
+for _ in range(3):
     print(phone_generator_v2(prompt_phone))
 ```
 
@@ -192,10 +188,10 @@ This prints out:
 Ah! We were missing the last digit, let's fix that and regenerate:
 
 ```python
-phone_regex_3_fixed = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]'
-phone_generator_v3 = outlines.generate.regex(model,
-                                             phone_regex_3_fixed)
-for _ in range(10):
+phone_regex_3_fixed = Regex(r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]')
+phone_generator_v3 = outlines.Generator(model, phone_regex_3_fixed)
+
+for _ in range(3):
     print(phone_generator_v3(prompt_phone))
 ```
 
diff --git a/docs/features/advanced/logits_processors.md b/docs/features/advanced/logits_processors.md
new file mode 100644
index 000000000..4e4d9e457
--- /dev/null
+++ b/docs/features/advanced/logits_processors.md
@@ -0,0 +1,97 @@
+---
+title: Logits Processors
+---
+
+# Logits Processors
+
+Logits processors are objects that control text generation by modifying the probability distribution of possible next tokens. They do this by adjusting the logits (raw model outputs) at each generation step, effectively biasing the model's token selection.
+Processors can be used to:
+1. Generate structured output (e.g., JSON that follows a specific schema)
+2. Prevent the model from generating specific words or tokens
+3. Implement custom token sampling strategies
+
+## Overview
+
+Outlines uses logits processors with steerable models — models that run locally and allow fine-grained control over the generation process. When using such models in Outlines, the output type provided is turned into a logits processor that is then passed to the inference engine.
+
+There are three models that support logits processors:
+
+- LlamaCpp
+- MLXLM
+- Transformers
+
+Instead of providing an output type that will be turned into a logits processor, it is possible to directly provide a logits processor. To do so, you must create a `Generator` instance using the `processor` keyword argument. You cannot directly call the model with a logits processor.
+
+For instance:
+
+```python
+import transformers
+from outlines import Generator, from_transformers
+from outlines.processors import RegexLogitsProcessor
+
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B"),
+    transformers.AutoTokenizer.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
+)
+
+# Create a regex logits processor that only returns hex unicode notations
+logits_processor = RegexLogitsProcessor(r"U\+[0-9A-Fa-f]{4,6}", model.tokenizer, model.tensor_library_name)
+
+# Create a generator with the logits processor and use it to generate text
+generator = Generator(model, processor=logits_processor)
+response = generator("What's the unicode for the hugging face emoji")
+
+print(response) # U+1F917
+```
+
+## Creating Custom Logits Processors
+
+You can create your own logits processor by subclassing the `OutlinesLogitsProcessor` class. This allows you to implement specific logic to modify logits as needed.
+Your logits processor needs to implement the `process_logits` method to modify the logits.
+`process_logits` accepts:
+- `input_ids`: the ids of the tokens of the existing sequences in a 2D tensor.
+- `logits`: the logits for the current generation step in a 2D tensor.
+
+In the example below, we create a custom logits processor to force the model to provide a response using only binary representation (so only the tokens for 0 and 1 are allowed):
+
+```python
+from outlines.processors.base_logits_processor import OutlinesLogitsProcessor, TensorType
+from outlines import Generator, from_transformers
+import transformers
+
+ALLOWED_TOKENS = [15, 16]  # token IDs corresponding to '0' and '1' in the model's vocabulary
+
+# Subclass OutlinesLogitsProcessor
+class BinaryLogitsProcessor(OutlinesLogitsProcessor):
+
+    def process_logits(self, input_ids: TensorType, logits: TensorType) -> TensorType:
+        # Create a mask for all tokens
+        mask = self.tensor_adapter.boolean_ones_like(logits)
+        # Set mask to False for the allowed tokens
+        for token_id in ALLOWED_TOKENS:
+            mask[:, token_id] = False
+        # Set non-allowed tokens to -inf so they are not selected
+        logits[mask] = float("-inf")
+        return logits
+
+# Create a regular model
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
+tf_model = transformers.AutoModelForCausalLM.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
+model = from_transformers(tf_model, tf_tokenizer)
+
+# Instantiate your custom logits processor
+logits_processor = BinaryLogitsProcessor(model.tensor_library_name)
+
+prompt = "Write the number 47 in binary. For example, 1010 is the binary representation of 10. Answer just with the binary number composed of 0s and 1s."
+formatted_prompt = tf_tokenizer.apply_chat_template(
+    [{"role": "user", "content": prompt}],
+    tokenize=False
+)
+
+# Create a generator with the custom logits processor instance and use it to generate text
+generator = Generator(model, processor=logits_processor)
+response = generator(formatted_prompt)
+
+print(response) # "101111"
+```
diff --git a/docs/features/core/generator.md b/docs/features/core/generator.md
new file mode 100644
index 000000000..508ba399c
--- /dev/null
+++ b/docs/features/core/generator.md
@@ -0,0 +1,122 @@
+---
+title: Generator API
+---
+
+# Generator
+
+
+The `Generator` class is the core component of Outlines v1. `Generator` accepts a [model](../reference/available_models) and an optional [output type](/api/types). If no output type is provided, the `Generator` will return unstructured text.
+
+!!! note
+
+    `Generator` is new as of Outlines v1, and replaces previous generator constructors:
+
+    - `generate.cfg`
+    - `generate.choice`
+    - `generate.format`
+    - `generate.fsm`
+    - `generate.json`
+    - `generate.regex`
+    - `generate.text`
+
+## Basic Usage
+
+```python
+from outlines import Generator, from_transformers
+import transformers
+
+# Initialize a model
+model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
+
+# Create a generator for plain text
+generator = Generator(model)
+result = generator("Write a short poem about AI.")
+
+# Print the result
+print(result)
+```
+
+## Structured Generation
+
+```python
+from pydantic import BaseModel
+from outlines import Generator, from_transformers
+import transformers
+
+# Define a Pydantic model for structured output
+class BookRecommendation(BaseModel):
+    title: str
+    author: str
+    year: int
+
+# Initialize a model
+model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
+
+# Create a generator for JSON output
+generator = Generator(model, BookRecommendation)
+
+# Generate a book recommendation
+result = generator("Recommend a science fiction book.")
+
+# Parse the JSON result into a Pydantic model
+book = BookRecommendation.model_validate_json(result)
+print(f"{book.title} by {book.author} ({book.year})")
+```
+
+## Parameters
+
+- `model`: The language model to use for generation
+- `output_type`: Optional. The type of output to generate
+
+## Generation Parameters
+
+When calling the generator, you can pass additional parameters to control the generation process. These parameters are passed through to the underlying model, so they depend on the specific model being used.
+
+Common parameters for most models include:
+- `max_new_tokens`: Maximum number of tokens to generate
+- `temperature`: Controls randomness (higher values = more random)
+- `top_p`: Controls diversity via nucleus sampling
+- `stop_strings`: String or list of strings at which to stop generation
+
+Example:
+```python
+result = generator(
+    "Write a short story.",
+    max_new_tokens=200,
+    temperature=0.7,
+    top_p=0.9,
+    stop_strings=["THE END", "###"]
+)
+```
+
+## Return Value
+
+The generator always returns a raw string containing the generated text. When generating structured outputs, you need to parse this string into the desired format.
+
+Unlike in Outlines v0, where the return type could be a parsed object, in v1 you are responsible for parsing the output when needed:
+
+```python
+# Outlines v1 approach
+from pydantic import BaseModel
+from outlines import Generator
+
+class Person(BaseModel):
+    name: str
+    age: int
+
+generator = Generator(model, Person)
+result = generator("Generate a person:")
+
+# Parse the result yourself
+person = Person.model_validate_json(result)
+```
+
+::: outlines.generator.Generator
diff --git a/docs/features/core/output_types.md b/docs/features/core/output_types.md
new file mode 100644
index 000000000..3a49c2623
--- /dev/null
+++ b/docs/features/core/output_types.md
@@ -0,0 +1,205 @@
+---
+title: Output Types
+---
+
+# Output Types
+
+Outlines provides a simple and intuitive way of defining the output structure of text generation. Possible output formats include basic Python types, multiple-choices, JSON schemas, regular expressions and context-free grammars.
+
+## Overview
+
+Outlines models accept a __prompt__ and an __output type__ when they are invoked, as well as additional inference keyword arguments that are forwarded on to the underlying model.
+
+Output types can be from the general Python ecosystem, including:
+- Most native Python types, such as `int` or `str`
+- Types from the `typing` module, such as `Literal`, `List`, `Dict`, `Enum`, etc
+- Types from popular theird party libraries such as Pydantic or GenSON.
+
+Outlines also provides special classes for certain output structures (more details below):
+- JSON schemas with `JsonSchema`
+- Regular expressions with `Regex`
+- Context-free grammars with `CFG`
+
+The general idea is that you should provide as an output type what you would give as the type hint of the return type of a function.
+
+Consider the following functions for instance:
+
+```python
+from datetime import date
+from typing import Dict, List, Literal, Union
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    birth_date: date
+    skills: Union[Dict, List[str]]
+
+def give_int() -> int:
+    ...
+
+def pizza_or_burger() -> Literal["pizza", "burger"]:
+    ...
+
+def create_character() -> Character:
+    ...
+```
+
+With an Outlines model, you can generate text that respects the type hints above by providing those as the output type:
+
+```python
+model("How many minutes are there in one hour", int) # "60"
+model("Pizza or burger", Literal["pizza", "burger"]) # "pizza"
+model("Create a character", Character, max_new_tokens=100) # '{"name": "James", "birth_date": "1980-05-10)", "skills": ["archery", "negociation"]}'
+```
+
+An important difference with function type hints though is that an Outlines generator always returns a string.
+You have to cast the response into the type you want yourself.
+
+For instance:
+
+```python
+result = model("Create a character", Character, max_new_tokens=100)
+casted_result = Character.model_validate_json(result)
+print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
+print(casted_result) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
+```
+
+## Output Type Categories
+
+We can group possible output types in several categories based on the use case they correspond to. While most of those types are native python or types coming from well-known third-party libraries, there are three Outlines-specific types: `JsonSchema`, `Regex` and `CFG`. Their use is explained below.
+
+### Basic Python Types
+
+The most straightfoward form of structed generation is to return an answer that conforms to a given basic type such as an int or a python list. You can use the basic Python types and the types from the `typing` library. For instance:
+
+```python
+from typing import Dict
+
+output_type = float # example of valid value: "0.05"
+output_type = bool # example of valid value: "True"
+output_type = Dict[int, str] # example of valid value: "{1: 'hello', 2: 'there'}"
+```
+
+You can combine types to create more complex response formats by relying on collection types and types such as `Union` and `Optional`. Let's consider for instance the output type below used to represent semi-structured data:
+
+```python
+from typing import Dict, List, Optional, Tuple, Union
+
+output_type = Dict[str, Union[int, str, List[Tuple[str, Optional[float]]]]]
+```
+
+Values created with this output type would be dictionnaries with string as keys and values made of either an integer, a string or a list of two elements tuples: a string and either a float or None. Example of a valid response for text generated with this output type (it would be contained in a string):
+
+```json
+{
+    "name": "Alice",
+    "age": 30,
+    "metrics": [("engagement", 0.85), ("satisfaction", None)]
+}
+```
+
+### Multiple Choices
+
+Outlines supports multiple choice classification by using the `Literal` or `Enum` output types. For instance:
+
+```python
+from enum import Enum
+from typing import Literal
+
+class PizzaOrBurger(Enum):
+    pizza = "pizza
+    burger = "burger
+
+# Equivalent multiple-choice output types
+output_type = Literal["pizza", "burger"]
+output_type = PizzaOrBurger
+```
+
+### JSON Schemas
+
+Multiple different common Python types are often used to store information equivalent to a JSON schema. The following can be used in Outlines to generate text that respects a JSON schema:
+
+- A Pydantic class
+- A Dataclass
+- A TypedDict
+- A [GenSON](https://github.com/wolverdude/GenSON) `SchemaBuilder`
+- A Callable (the parameters are turned into the keys and the type hinting is used to define the types of the values)
+
+For instance:
+
+```python
+from dataclasses import dataclass
+
+@dataclass
+class Character:
+    name: str
+    age: int
+
+ouput_type = Character
+
+def character(name: str, age: int):
+    return None
+
+output_type = character
+```
+
+There are two other JSON schema formats that require Outlines-specific classes: JSON schema strings and dictionaries.
+
+As those are contained in regular Python strings or dictionnaries, the associated output format would be ambiguous if they were to be provided directly. As a result, Outlines requires them to be wrapped in a `outlines.types.JsonSchema` object. For instance:
+
+```python
+from outlines.types import JsonSchema
+
+schema_string = '{"type": "object", "properties": {"answer": {"type": "number"}}}'
+output_type = JsonSchema(schema_string)
+
+schema_dict = {
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+output_type = JsonSchema(schema_dict)
+```
+
+### Regex Patterns
+
+Outlines provides support for text generation constrained by regular expressions. Since regular expressions are expressed as simple raw string literals, regex strings must wrapped in an `outlines.types.Regex` object to clarify the expected return type.
+
+```python
+from outlines.types import Regex
+
+regex = r"[0-9]{3}"
+output_type = Regex(regex)
+```
+
+The `outlines.types` module contains a few common regex patterns stored in variables you can import and directly use as output types. Common patterns include a sentence, an email address and an [ISBN reference](https://en.wikipedia.org/wiki/ISBN). For instance:
+
+```python
+from outlines.types import sentence
+
+print(type(sentence)) # outlines.types.dsl.Regex
+print(sentence.pattern) # [A-Z].*\s*[.!?]
+```
+
+To help you create complex regex patterns yourself, you can use the Outlines [regex DSL](../../utility/regex_dsl).
+
+### Context-Free Grammars
+
+Outlines allows you to generate text that respects the syntax of a context-free grammar. Context-free grammars are defined using [Lark](https://lark-parser.readthedocs.io/en/latest/index.html), a grammar language. Since grammars are expressed as a string, Large CFG strings should be be wrapped in an `outlines.types.CFG` object. For instance:
+
+```python
+from outlines.types import CFG
+
+grammar_string = """
+    start: expr
+    expr: "{" expr "}" | "[" expr "]" |
+"""
+output_type = CFG(grammar_string)
+```
+
+You can find a few Lark grammar examples in the [grammars module](../../api_reference/grammars.md).
+
+## Output type availability
+
+The output types presented above are not available for all models as some have only limited support for structured outputs. Please refer to the documentation of the specific model you wish to use to know what output types it supports.
diff --git a/docs/features/index.md b/docs/features/index.md
new file mode 100644
index 000000000..00df02ba3
--- /dev/null
+++ b/docs/features/index.md
@@ -0,0 +1,19 @@
+# Features
+
+This section presents in details the different features of Outlines.
+
+## Core Concepts
+
+- [Models](./models/index.md)
+- [Output Types](./core/output_types.md)
+- [Generators](./core/generator.md)
+
+## Utility
+
+- [Applications](./utility/application.md)
+- [Templates](./utility/templates.md)
+- [Regex DSL](./utility/regex_dsl.md)
+
+## Advanced
+
+- [Logits Processors](./advanced/logits_processors.md)
diff --git a/docs/features/models/anthropic.md b/docs/features/models/anthropic.md
new file mode 100644
index 000000000..dd0794386
--- /dev/null
+++ b/docs/features/models/anthropic.md
@@ -0,0 +1,98 @@
+---
+title: Anthropic
+---
+
+# Anthropic
+
+!!! Installation
+
+    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines. Install all optional dependencies of the `Anthropic` model with: `pip install outlines[anthropic]`.
+
+    You also need to have an Anthropic API key. This API key must either be set as an environment variable called `ANTHROPIC_API_KEY` or be provided to the `anthropic.Anthropic` class when instantiating it.
+
+## Model Initialization
+
+To create an Anthropic model instance, you can use the `from_anthropic` function. It takes 2 arguments:
+
+- `client`: an `anthropic.Anthropic` instance
+- `model_name`: the name of the model you want to use in subsequent model calls (optional)
+
+For instance:
+
+```python
+from anthropic import Anthropic
+import outlines
+
+# Create the Anthropic client
+client = Anthropic()
+
+# Create the model
+model = outlines.from_anthropic(
+    client,
+    "claude-3-5-sonnet-latest"
+)
+```
+
+Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-claude/models) for an up-to-date list of available models.
+
+## Text Generation
+
+Once you've created your Outlines `Anthropic` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+
+For instance:
+
+```python
+from anthropic import Anthropic
+import outlines
+
+# Create the model
+model = outlines.from_anthropic(
+    Anthropic(),
+    "claude-3-5-sonnet-latest"
+)
+
+# Call it to generate text
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
+```
+
+Anthropic does not support structured generation, so you cannot provide an output type. It does support streaming and vision inputs though.
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+from anthropic import Anthropic
+from outlines import from_anthropic
+from outlines.templates import Vision
+
+# Create the model
+model = from_anthropic(
+    Anthropic(),
+    "claude-3-5-sonnet-latest"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt
+prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+
+# Stream the response
+for chunk in model.stream(prompt, max_tokens=50):
+    print(chunk) # 'This...'
+```
+
+## Inference arguments
+
+When calling the model or streaming, you can provide keyword arguments that will be passed down to the Anthropic client. Make sure to include all the arguments you need to configure the client's behavior to your expected behavior. Some of the most common arguments include `max_tokens`, `temperature`, `stop_sequences` and `top_k`.
+
+See the [Anthropic API documentation](https://docs.anthropic.com/en/api/messages) for the full list of available arguments.
+
+!!! Warning
+
+    You must set a value for `max_tokens` with Anthropic models.
diff --git a/docs/features/models/dottxt.md b/docs/features/models/dottxt.md
new file mode 100644
index 000000000..306ef2c62
--- /dev/null
+++ b/docs/features/models/dottxt.md
@@ -0,0 +1,74 @@
+---
+title: Dottxt
+---
+
+# Dottxt
+
+!!! Installation
+
+    You need to install the `dottxt` python sdk to be able to use the Dottxt API in Outlines. Install all optional dependencies of the `Dottxt` model with: `pip install outlines[dottxt]`.
+
+    You also need to have a Dottxt API key. This API key must either be set as an environment variable called `DOTTXT_API_KEY` or be provided to the `dottxt.client.Dottxt` class when instantiating it.
+
+## Model Initialization
+
+To create an Dottxt model instance, you can use the `from_dottxt` function. It takes 3 arguments:
+
+- `client`: a `dottxt.client.Dottxt` instance
+- `model_name`: the name of the model you want to use in subsequent model calls (optional)
+- `model_revision`: the name of the revision to use for the model selected (optional)
+
+For instance:
+
+```python
+from dottxt.client import Dottxt
+import outlines
+
+# Create client
+client = Dottxt(api_key="...")
+
+# Create the model
+model = outlines.from_dottxt(
+    client,
+    "meta-llama/Llama-3.1-8B",
+    "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b"
+)
+```
+
+Use the `list_models` method of the Dottxt client to get a list of available model names and revisions for your account.
+
+## Text Generation
+
+Dottxt only supports constrained generation with JSON schema output types. You must always provide a value for the `output_type` parameter as unconstrained generation is not available.
+
+For instance:
+
+```python
+from typing import List
+from pydantic import BaseModel
+from dottxt.client import Dottxt
+import outlines
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_dottxt(
+    Dottxt(),
+    "meta-llama/Llama-3.1-8B",
+    "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b"
+)
+
+# Generate structured text
+result = model("Create a character", Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+## Inference arguments
+
+You can provide the same optional parameters you would pass to the `dottxt` sdk's client both during the initialization of the `Dottxt` class and when generating text. Some of the common inference arguments include `max_tokens`, `frequency_penalty`, `presence_penalty` and `temperature`.
+
+Consult the [dottxt python sdk GitHub repository](https://github.com/dottxt-ai/dottxt-python) for the full list of parameters.
diff --git a/docs/features/models/gemini.md b/docs/features/models/gemini.md
new file mode 100644
index 000000000..e6daccf26
--- /dev/null
+++ b/docs/features/models/gemini.md
@@ -0,0 +1,148 @@
+# Gemini
+
+!!! Installation
+
+    You need to install the `google.genai` libray to be able to use the Gemini API in Outlines. Install all optional dependencies of the `Gemini` model with: `pip install outlines[gemini]`.
+
+    You also need to have a Gemini API key. This API key must either be set as an environment variable called `GEMINI_API_KEY` or be provided to the `google.genai.Client` class when instantiating it.
+
+## Model Initialization
+
+To create a Gemini model instance, you can use the `from_gemini` function. It takes 2 arguments:
+
+- `client`: a `google.genai.Client` instance
+- `model_name`: the name of the model you want to use in subsequent model calls (optional)
+
+For instance:
+
+```python
+import outlines
+from google import genai
+
+# Create the client
+client = genai.Client()
+
+# Create the model
+model = outlines.from_gemini(
+    client,
+    "gemini-1.5-flash-latest"
+)
+```
+
+Check the [Gemini documentation](https://github.com/googleapis/python-genai) for an up-to-date list of available models.
+
+## Text Generation
+
+Once you've created your Outlines `Gemini` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+from google import genai
+
+# Create the model
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_output_tokens=20)
+print(result) # 'Riga'
+```
+
+Gemini also supports streaming. For instance:
+
+```python
+import outlines
+from google import genai
+
+# Create the model
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_output_tokens=20):
+    print(chunk) # 'In...'
+```
+
+## Structured Generation
+
+Gemini provides supports for some forms of structured output: multiple choice, JSON schema (with caveats) and lists of structured objects. To use it, call the model with an `output_type` on top of your prompt.
+
+#### Multiple Choice
+
+```python
+import outlines
+from google import genai
+from enum import Enum
+
+class PizzaOrBurger(Enum):
+    pizza = "pizza"
+    burger = "burger"
+
+# Create the model
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
+
+# Call it with the ouput type to generate structured text
+result = model("Pizza or burger?", PizzaOrBurger, max_output_tokens=20)
+print(result) # 'pizza'
+```
+
+#### JSON Schema
+
+Gemini supports only three types of objects used to define a JSON Schema:
+
+- Pydantic classes
+- Dataclasses
+- TypedDicts
+
+```python
+from typing import List
+from pydantic import BaseModel
+from google import genai
+import outlines
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
+
+# Call it with the ouput type to generate structured text
+result = model("Create a character", Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+#### Lists of Structured Objects
+
+A specificity of Gemini is that, despite not supporting regex, it does support a list of structured objects as an output type. To use it, put any of three available types described above in the typing `List` class
+
+```python
+from dataclasses import dataclass
+from google import genai
+import outlines
+
+@dataclass
+class Character:
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
+
+# Call it with the ouput type to generate structured text
+result = model("Create a character", list[Character])
+print(result) # '[{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}, {["name":...'
+```
+
+!!! Attention
+
+    The structured objects must be in a built-in `list`, not a `List` from the `typing` library
+
+## Inference arguments
+
+You can provide the same optional parameters you would pass to the `google.genai.Client` client both during the initialization of the Gemini model and when generating text. Some of the common inference arguments include `max_output_tokens`, `temperature`, and other generation parameters.
+
+Consult the [Google Generative AI documentation](https://github.com/googleapis/python-genai) for the full list of parameters.
diff --git a/docs/features/models/index.md b/docs/features/models/index.md
new file mode 100644
index 000000000..89c4fa124
--- /dev/null
+++ b/docs/features/models/index.md
@@ -0,0 +1,107 @@
+---
+title: Models
+---
+
+# Models
+
+## Overview
+
+Outlines models are objects that wrap an inference client or engine. Models provide a standardized interface to generate structured text.
+
+!!! Warning
+
+    The model loading functions have been modified in v1. While they used to be called `<name_inference_library>`, they are now called `from_<name_inference_library>`. The model classes' names and `__init__` methods are left unchanged.
+
+
+All Outlines model classes have an associated loader function to facilitate initializing a model instance. The name of this function is `from_` plus the name of the model in lower-case letters. For instance, Outlines has a `Transformers` model and an associated `from_transformers` loader function. The parameters to load a model are specific to each provider, please consult the documentation of the model you want to use for more information.
+
+After having created a model instance, you can either directly call it to generate text or first create a reusable generator that you would then call. In either case, you can provide an `output_type` to constrain the format of the generation output. See the [output types section](../core/output_types.md) for more information on constrained generation.
+
+For instance:
+
+```python
+from outlines import from_transformers, Generator
+import transformers
+
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+)
+
+# Call it directly
+response = model("How many countries are there in the world", max_new_tokens=20)
+print(result) # 'There are 200 countries in the world.'
+
+# Call it directly with an output_type
+response = model("How many countries are there in the world", int, max_new_tokens=20)
+print(result) # '200'
+
+# Create a generator first and then call it
+generator = Generator(model, int)
+response = generator("How many countries are there in the world")
+print(result) # '200'
+```
+
+## Features Matrix
+
+| | [Anthropic](../../models/anthropic) | [Dottxt](../../models/dottxt) | [Gemini](../../models/gemini) | [LlamaCpp](../../models/llamacpp) | [MLXLM](../../models/mlxlm) | [Ollama](../../models/ollama) | [OpenAI](../../models/openai) | [SGLang](../../models/sglang) | [TGI](../../models/tgi) | [Transformers](../../models/transformers) | [Transformers MultiModal](../../models/transformers_multimodal) | [VLLM](../../models/vllm) | [VLLMOffline](../../models/vllm_offline) |
+|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
+| **Output Types** | | | | | | | | | | | | | |
+| Simple Types | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| JSON Schema | ❌ | ✅ | 🟠 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Multiple Choice | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Regex | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Grammar | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | 🟠 | ❌ | ✅ | ✅ | ✅ | ✅ |
+| **Generation Features** | | | | | | | | | | | | | |
+| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
+| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ |
+
+## Model Types
+
+Models can be divided into two categories: local models and server-based models.
+
+In the case of local models, the text generation happens within the inference library object used to instantite the model. This gives Outlines direct access to the generation process (through a logits processor) and means all structured generation output types are available.
+
+The local models available are the following:
+
+- LlamaCpp
+- MLXLM
+- Transformers
+- TransformersMultiModal
+- VLLMOffline
+
+In the case of server-based models, the model is initialized with a client that sends a request to a server that is in charge of the actual text generation. As a result, we have limited control over text generation and some output types are not supported. The server on which the text generation happens can either be remote (with OpenAI or Anthopic for instance) or local (with SGLang for instance).
+
+The server-based models available are the following:
+
+- Anthropic
+- Dottxt
+- Gemini
+- Ollama
+- OpenAI
+- SgLang
+- TGI
+- VLLM
+
+Some models have an async version. To use them, just pass the async version of the provider object to their loading function. It will then return a `Async<ModelName>` instance with the same methods and features as the regular sync instance.
+
+For instance:
+
+```python
+from outlines import from_tgi
+from huggingface_hub import AsyncInferenceClient
+
+model = from_tgi(
+    AsyncInferenceClient("http://localhost:8000/v1")
+)
+print(type(model)) # outlines.models.tgi.AsyncTGI
+```
+
+The models that have an async version are the following:
+
+- SgLang
+- TGI
+- VLLM
diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
new file mode 100644
index 000000000..cf7dd3e0c
--- /dev/null
+++ b/docs/features/models/llamacpp.md
@@ -0,0 +1,168 @@
+---
+title: llama.cpp
+---
+
+# llama.cpp
+
+Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library](https://github.com/abetlen/llama-cpp-python). Llamacpp allows to run quantized models on machines with limited compute.
+
+!!! Installation
+
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration. Install all optional dependencies of the `LlamaCpp` model with: `pip install outlines[llamacpp]`.
+
+    See the [llama-cpp-python Github page](https://github.com/abetlen/llama-cpp-python?tab=readme-ov-file#supported-backends) for instructions on installing with CUDA, Metal, ROCm and other backends.
+
+
+## Model Initialization
+
+To load the model, you can use the `from_llamacpp` function. The single argument of the function is a `Llama` model instance from the `llama_cpp` library. Consult the [Llama class API reference](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama) for detailed information on how to create a model instance and on the various available parameters.
+
+For instance:
+
+```python
+import outlines
+from llama_cpp import Llama
+
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+from llama_cpp import Llama
+
+# Create the model
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_tokens=20)
+print(result) # 'Riga'
+```
+
+The `LlamaCpp` model also supports streaming. For instance:
+
+```python
+import outlines
+from llama_cpp import Llama
+
+# Create the model
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In...'
+```
+
+## Structured Generation
+
+The `LlamaCpp` model supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
+
+### Basic Type
+
+```python
+import outlines
+from llama_cpp import Llama
+
+output_type = int
+
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+from typing import List
+from pydantic import BaseModel
+import outlines
+from llama_cpp import Llama
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+result = model("Create a character.", output_type=Character, max_tokens=200)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import outlines
+from llama_cpp import Llama
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+result = model("What is the capital of France?", output_type)
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+from outlines.types import Regex
+import outlines
+from llama_cpp import Llama
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+result = model("Generate a fake social security number.", output_type)
+print(result) # '782-32-3789'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `__call__` method of the `llama_cpp.Llama` model. Some common inference arguments include `max_tokens`, `temperature`, `frequency_penalty` and `top_p`.
+
+See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__) for more information on inference parameters.
diff --git a/docs/features/models/mlxlm.md b/docs/features/models/mlxlm.md
new file mode 100644
index 000000000..d2cba95fc
--- /dev/null
+++ b/docs/features/models/mlxlm.md
@@ -0,0 +1,189 @@
+---
+title: mlx-lm
+---
+
+# mlx-lm
+
+Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
+
+!!! Note "Installation"
+
+    You need a device that [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration.
+
+    You need to install the `mlx` and `mlx-lm` libraries to be able to use mlx in Outlines. Install all optional dependencies of the `MLXLM` model with: `pip install outlines[mlxlm]`.
+
+## Model Initialization
+
+To create a MLXLM model instance, you can use the `from_mlxlm` function. It takes 2 arguments:
+
+- `model`: an `mlx.nn.Module` instance
+- `tokenizer`: a `transformers.PreTrainedTokenizer` instance
+
+However, we recommend you simply pass on the output of the `mlx_lm.load` function (it takes a model name as an argument).
+
+For instance:
+
+```python
+import outlines
+import mlx_lm
+
+# Create the model
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+import mlx_lm
+
+# Load the model
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_tokens=20)
+print(result) # 'Riga'
+```
+
+The `MLXLM` model also supports streaming. For instance:
+
+```python
+import outlines
+import mlx_lm
+
+# Load the model
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In...'
+```
+
+## Structured Generation
+
+As a local model, `MLXLM` supports all forms of structured generation available in Outlines.
+
+### Basic Type
+
+```python
+import outlines
+import mlx_lm
+
+output_type = int
+
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+from pydantic import BaseModel
+from typing import List
+import outlines
+import mlx_lm
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("Create a character.", output_type=Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import outlines
+import mlx_lm
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("What is the capital of France?", output_type)
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+from outlines.types import Regex
+import outlines
+import mlx_lm
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("Generate a fake social security number.", output_type)
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
+
+```python
+from outlines.types import CFG
+import outlines
+import mlx_lm
+
+arithmetic_grammar = """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
+
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("Write an addition.", output_type, max_tokens=20)
+print(result) # '23 + 48'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `mlx_lm.generate` function used to generate text.
+
+See the [MLXLM documentation](https://github.com/ml-explore/mlx-lm) for more information on inference parameters.
diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
new file mode 100644
index 000000000..eaad4c3e2
--- /dev/null
+++ b/docs/features/models/ollama.md
@@ -0,0 +1,103 @@
+---
+title: Ollama
+---
+
+# Ollama
+
+!!! Installation
+
+    To be able to use Ollama in Outlines, you must install both Ollama and the optional dependency libraries of the model.
+
+    - To download Ollama: https://ollama.com/download
+    - To install the ollama python sdk: `pip install outlines[ollama]`
+
+    Consult the [`ollama` documentation](https://github.com/ollama/ollama-python) for detailed information on installation and client initialization.
+
+## Model Initialization
+
+To create an Ollama model instance, you can use the `from_ollama` function. It takes 2 arguments:
+
+- `client`: an `ollama.Client` instance
+- `model_name`: the name of the model you want to use
+
+For instance:
+
+```python
+import ollama
+import outlines
+
+# Create the client
+client = ollama.Client()
+
+# Create the model
+model = outlines.from_ollama(
+    client,
+    "tinyllama",
+)
+```
+
+You can find the list of available models on the [Ollama library](https://ollama.com/library).
+
+## Text Generation
+
+Once you've created your Outlines `Ollama` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+
+For instance:
+
+```python
+import ollama
+import outlines
+
+# Create the model
+model = outlines.from_ollama(ollama.Client(), "tinyllama")
+
+# Call it to generate text
+response = model("What's the capital of Latvia?")
+print(response) # 'Riga'
+```
+
+Ollama also supports streaming. For instance:
+
+```python
+import ollama
+import outlines
+
+# Create the model
+model = outlines.from_ollama(ollama.Client(), "tinyllama")
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat"):
+    print(chunk) # 'In...'
+```
+
+## Structured Generation
+
+Ollama only provides support for structured generation based on a JSON schema. To use it, call the model with a JSON schema object as an `output_type` on top of your prompt.
+
+For instance:
+
+```python
+from typing import List
+from pydantic import BaseModel
+import ollama
+import outlines
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_ollama(ollama.Client(), "tinyllama")
+
+# Call it with the output type to generate structured text
+result = model("Create a character", Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+## Inference arguments
+
+When calling the model, you can provide keyword arguments that will be passed down to the `generate` method of the Ollama client.
+
+Consult the [Ollama REST API documentation](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-completion) for the full list of inference parameters.
diff --git a/docs/features/models/openai.md b/docs/features/models/openai.md
new file mode 100644
index 000000000..e4532d213
--- /dev/null
+++ b/docs/features/models/openai.md
@@ -0,0 +1,134 @@
+# OpenAI
+
+!!! Installation
+
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
+
+    You also need to have an OpenAI API key. This API key must either be set as an environment variable called `OPENAI_API_KEY` or be provided to the `openai.OpenAI` class when instantiating it.
+
+## Model Initialization
+
+To create an OpenAI model instance, you can use the `from_openai` function. It takes 2 arguments:
+
+- `client`: an `openai.OpenAI` or `openai.AzureOpenAI` instance
+- `model_name`: the name of the model you want to use
+
+For instance:
+
+```python
+import outlines
+import openai
+
+# Create the client
+client = openai.OpenAI()
+
+# Create the model
+model = outlines.from_openai(
+    client,
+    "gpt-4o"
+)
+```
+
+Check the [OpenAI documentation](https://platform.openai.com/docs/models) for an up-to-date list of available models. As shown above, you can use Azure OpenAI in Outlines the same way you would use OpenAI, just provide an `openai.AzureOpenAI` instance to the Outlines model class.
+
+## Text Generation
+
+Once you've created your Outlines `OpenAI` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+model = outlines.from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
+)
+
+# Call it to generate text
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
+```
+
+OpenAI supports streaming and vision inputs.
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+import openai
+import outlines
+from outlines.templates import Vision
+
+# Create the model
+model = outlines.from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt
+prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+
+# Stream the response
+for chunk in model.stream(prompt, max_tokens=50, stop="."):
+    print(chunk) # 'This...'
+```
+
+## Structured Generation
+
+OpenAI provides supports for some forms of structured output: JSON schemas and JSON syntax. To use it, call the model with an `output_type` on top of your prompt.
+
+#### JSON Schema
+
+```python
+from typing import List
+from pydantic import BaseModel
+import openai
+import outlines
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
+
+# Call it with the output type to generate structured text
+result = model("Create a character, use the json format.", Character, top_p=0.1)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+#### JSON Syntax
+
+What we mean by JSON syntax is what is sometimes called JSON mode, meaning that the model will return a valid JSON, but you do not get to specify its structure. To use this JSON mode, provide the `dict` type as an output type.
+
+```python
+import openai
+import outlines
+
+# Create the model
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
+
+# Call it with the output type to generate structured text
+result = model("Create a character, use the json format.", dict, temperature=0.5)
+print(result) # '{"first_name": "Henri", "last_name": "Smith", "height": "170"}'
+```
+
+## Inference arguments
+
+When calling the model, you can provide keyword arguments that will be passed down to the `chat.completions.create` method of the OpenAI client. Some of the most common arguments include `max_tokens`, `temperature`, `stop` and `top_p`.
+
+Another keyword argument of interest is `n`. If set with an integer value superior to 1, OpenAI will generate several sample responses and you will receive a list of strings as a response to your model call.
+
+See the [OpenAI API documentation](http://platform.openai.com/docs/api-reference/chat/create) for the full list of available arguments.
diff --git a/docs/features/models/openai_compatible.md b/docs/features/models/openai_compatible.md
new file mode 100644
index 000000000..4925d53d4
--- /dev/null
+++ b/docs/features/models/openai_compatible.md
@@ -0,0 +1,87 @@
+# OpenAI-Compatible APIs
+
+Many inference providers offer OpenAI-compatible APIs, allowing you to use the familiar OpenAI SDK while connecting to different backends. Outlines allows you can leverage various providers while maintaining consistent code.
+
+## What are OpenAI-Compatible APIs?
+
+OpenAI-compatible APIs implement the same REST endpoints and request/response formats as OpenAI's API, but serve different models or run on different infrastructure. This allows you to use the `openai` Python library with any compatible provider by simply changing the `base_url`.
+
+!!! Installation
+
+    You need to install the `openai` library to be able to use the OpenAI-compatible APIs in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
+
+## General Usage Pattern
+
+The standard approach is to use the OpenAI SDK with a custom base URL:
+
+```python
+import openai
+import outlines
+
+# Point to your OpenAI-compatible endpoint
+client = openai.OpenAI(
+    base_url="https://your-provider.com/v1",  # Custom endpoint
+    api_key="your-api-key"
+)
+
+# Use with Outlines
+model = outlines.from_openai(client, "model-name")
+```
+
+## Important: Provider-Specific Parameters
+
+!!! Warning "API-Specific Parameters"
+
+    Some providers require additional parameters in the API request for structured generation to work properly. These are typically passed as extra arguments when calling the model.
+
+    For example, some providers may need special parameters in the request body to enable guided generation or specify constraints. Always consult your provider's documentation for structured generation requirements.
+
+## Popular OpenAI-Compatible Providers
+
+Many providers offer OpenAI-compatible endpoints:
+
+- **Groq**
+- **Together AI**
+- **Anyscale**
+- **Fireworks AI**
+- **Perplexity**
+- **Local servers** (LocalAI, etc.)
+
+## Configuration Examples
+
+### Basic Setup
+```python
+import openai
+import outlines
+
+# Generic OpenAI-compatible setup
+client = openai.OpenAI(
+    base_url="https://api.your-provider.com/v1",
+    api_key="your-api-key"
+)
+
+model = outlines.from_openai(client, "provider-model-name")
+```
+
+### With Authentication Headers
+```python
+import openai
+import outlines
+
+# Some providers need custom headers
+client = openai.OpenAI(
+    base_url="https://api.your-provider.com/v1",
+    api_key="your-api-key",
+    default_headers={"Custom-Header": "value"}
+)
+
+model = outlines.from_openai(client, "provider-model-name")
+```
+
+## Related Documentation
+
+For specific implementations that use OpenAI-compatible APIs:
+
+- [SGLang](sglang.md): Local inference server with OpenAI-compatible endpoints
+- [vLLM](vllm.md): High-performance inference with OpenAI-compatible API
+- [OpenAI](openai.md): The original OpenAI API implementation
diff --git a/docs/features/models/sglang.md b/docs/features/models/sglang.md
new file mode 100644
index 000000000..95a6eb7b5
--- /dev/null
+++ b/docs/features/models/sglang.md
@@ -0,0 +1,171 @@
+---
+title: SGLang
+---
+
+# SGLang
+
+## Prerequisites
+
+The Outlines `SGLang` model is intended to be used along with an SGLang instance running on a separate server (can be local or remote). Make sure you have a SGLang server running and accessible before using the `SGLang` model. As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies of the `SGLang` model with: `pip install outlines[sglang]`.
+
+When launching your SGLang server, you can specify the backend engine to use for structured generation through the `grammar-backend` cli argument. Add `--grammar-backend outlines` to your command to use Outlines instead of the default engine.
+
+## Model Initialization
+
+To load the model, you can use the `from_sglang` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Make sure the value of the `base_url` argument of the `OpenAI` client points to your running SGLang server. Consult the [SGLang documentation](https://docs.sglang.ai/backend/send_request.html) on using an OpenAI client with an SGLang server for more information.
+
+Based on whether the `openai` client instance is synchronous or asynchronous, you will receive a `SGLang` or `AsyncSGLang` model instance.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the OpenAI client
+sync_openai_client = openai.OpenAI(base_url="http://localhost:11434")
+async_openai_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
+
+# Create a sync model
+sync_model = outlines.from_sglang(sync_openai_client)
+print(type(sync_model)) # <class 'outlines.models.sglang.SGLang'>
+
+# Create an async model
+async_model = outlines.from_sglang(async_openai_client)
+print(type(async_model)) # <class 'outlines.models.sglang.AsyncSGLang'>
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+# Call it to generate text
+result = model("Write a short story about a cat.", max_tokens=100)
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```
+
+The `SGLang` model also supports streaming. For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In ...'
+```
+
+## Structured Generation
+
+SGLang supports all output types available in Outlines (context-free grammars with caveats though, see the subsection below for more details). Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import openai
+import outlines
+
+output_type = int
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+import openai
+import outlines
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("Create a character.", Character, frequency_penalty=1.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import openai
+import outlines
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("What is the capital of France?", output_type, temperature=0)
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+import openai
+import outlines
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("Generate a fake social security number.", output_type, top_p=0.1)
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
+
+SGLang supports grammars, but expects an EBNF format instead of the Lark format Outlines uses. Thus, to use a context-free grammar with SGLang, provide a string using the EBNF syntax to the Outlines `CFG` object.
+
+```python
+import openai
+import outlines
+from outlines.types import CFG
+
+ebnf_grammar = """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+output_type = CFG(ebnf_grammar)
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("Is the weather good today?", output_type)
+print(result) # 'yes'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.
+
+An optional parameter of particular interest is `extra_body`, which is a dictionary containing arguments that are specific to SGLang and are not part of the standard `openai` interface.
+
+See the [SGLang documentation](https://docs.sglang.ai/backend/openai_api_completions.html) on parameters for the OpenAI-compatible server for more information on inference parameters.
diff --git a/docs/features/models/tgi.md b/docs/features/models/tgi.md
new file mode 100644
index 000000000..8fb807580
--- /dev/null
+++ b/docs/features/models/tgi.md
@@ -0,0 +1,137 @@
+---
+title: TGI
+---
+
+# TGI
+
+## Prerequisites
+
+The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed. Install all optional dependencoes of the `TGI` model with: `pip install outlines[tgi]`
+
+## Model Initialization
+
+To load the model, you can use the `from_tgi` function. The argument of the function is either an `InferenceClient` or `AsyncInferenceClient` instance from the `huggingface_hub` library. Consult the [HuggingFace documentation](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client) for more information on their inference client.
+
+Based on whether the inference client instance is synchronous or asynchronous, you will receive a `TGI` or an `AsyncTGI` model instance.
+
+For instance:
+
+```python
+import outlines
+import huggingface_hub
+
+# Create the inference client
+client = huggingface_hub.InferenceClient("http://localhost:11434")
+async_client = huggingface_hub.AsyncInferenceClient("http://localhost:11434")
+
+# Create a sync model
+sync_model = outlines.from_tgi(client)
+print(type(sync_model))  # <class 'outlines.models.tgi.TGI'>
+
+# Create an async model
+async_model = outlines.from_tgi(async_client)
+print(type(async_model))  # <class 'outlines.models.tgi.AsyncTGI'>
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+import huggingface_hub
+
+# Create the model
+client = huggingface_hub.InferenceClient("http://localhost:11434")
+model = outlines.from_tgi(client)
+
+# Call it to generate text
+result = model("Write a short story about a cat.", stop_sequences=["."])
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```The `TGI` model supports streaming. For instance:
+
+```python
+import outlines
+import huggingface_hub
+
+# Create the model
+client = huggingface_hub.InferenceClient("http://localhost:11434")
+model = outlines.from_tgi(client)
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", stop_sequences=["."]):
+    print(chunk) # 'In ...'
+```
+
+## Structured Generation
+
+TGI supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import outlines
+import huggingface_hub
+
+output_type = int
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```### JSON Schema
+
+```python
+import outlines
+import huggingface_hub
+from typing import List
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```### Multiple Choice
+
+```python
+import outlines
+import huggingface_hub
+from typing import Literal
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("What is the capital of France?", output_type, temperature=0)
+print(result) # 'Paris'
+```### Regex
+
+```python
+import outlines
+import huggingface_hub
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("Generate a fake social security number.", output_type, top_p=0.1)
+print(result) # '782-32-3789'
+```
+## Inference parameters
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `text_generation` method of the TGI client.
+
+Common parameters include `max_new_tokens`, `stop_sequences`, `temperature`, `top_k`, `top_p`, and others as specified in the [TGI inference client documentation](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client#huggingface_hub.InferenceClient).
diff --git a/docs/features/models/transformers.md b/docs/features/models/transformers.md
new file mode 100644
index 000000000..9dc1cab79
--- /dev/null
+++ b/docs/features/models/transformers.md
@@ -0,0 +1,207 @@
+---
+title: Transformers
+---
+
+# Transformers
+
+!!! Installation
+
+    You need to install the `transformers` library to be able to use the Transformers in Outlines. Install all optional dependencies of the `Transformers` model with: `pip install outlines[transformers]`.
+
+    See the [HuggingFace documentation](https://huggingface.co/docs/transformers/en/installation) for more information on installing `transformers` with CPU, GPU...
+
+## Model Initialization
+
+To load the model, you can use the `from_transformers` function. It takes 2 arguments:
+
+- `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
+- `tokenizer_or_processor`: a `transformers` tokenizer (created with `AutoTokenizer` for instance, it must be an instance of either `PreTrainedTokenizer` or `PreTrainedTokenizerFast`)
+
+For instance:
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Create the transformers model and tokenizer
+hf_model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+hf_tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+
+# Create the Outlines model
+model = outlines.from_transformers(hf_model, hf_tokenizer)
+```
+
+If you provide a processor instead of a tokenizer for the second argument of the `from_transformers` function, you would get a `TransformersMultiModal` instance. See the [TransformersMultiModal model documentation](./transformers_multimodal.md) for more information on using multimodal models in Outlines.
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Create model
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_new_tokens=20)
+print(result) # 'Riga'
+```
+
+The `Transformers` model also supports batch generation. To use it, provide a list of prompts instead of a single prompt. You will receive as result a list of completions. For instance:
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Create model
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Create a list of prompts that will be used in a single batch
+prompts = [
+    "What's the capital of Lithuania?",
+    "What's the capital of Latvia?",
+    "What's the capital of Estonia?"
+]
+
+# Call it to generate text
+result = model(prompts, max_new_tokens=20)
+print(result) # ['Vilnius', 'Riga', 'Tallinn']
+```
+
+## Structured Generation
+
+As a local model, `Transformers` supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+output_type = int
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("How many countries are there in the world?", output_type, max_new_tokens=5)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from pydantic import BaseModel
+from typing import List
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Create a character.", output_type=Character, max_new_tokens=200, repetition_penalty=0.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("What is the capital of France?", output_type, max_new_tokens=10, temperature=0)
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Generate a fake social security number.", output_type, max_new_tokens=20, top_p=0.5)
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from outlines.types import CFG
+
+arithmetic_grammar = """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Write an addition.", output_type, max_new_tokens=100)
+print(result) # '23 + 48'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `generate` method of the `transformers` model. Some common inference arguments include `max_new_tokens`, `temperature`, `repetition_penalty` and `top_p`.
+
+See the [transformers documentation](https://huggingface.co/docs/transformers/en/main_classes/text_generation) for more information on inference parameters.
+
+!!! Warning
+
+    The `max_new_tokens` inference parameter has a default value of 20. This is insufficient for most tasks and will result in the generation output not respecting the output type (because the response is truncated). We recommend you always provide a value for this argument.
diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
new file mode 100644
index 000000000..b2b846f6c
--- /dev/null
+++ b/docs/features/models/transformers_multimodal.md
@@ -0,0 +1,89 @@
+---
+title: Transformers MultiModal
+---
+
+# Transformers MultiModal
+
+The Outlines `TransformersMultiModal` model inherits from `Transformers` and shares most of its interface. Please start by reading the [Transformers documentation](./transformers.md) as this document only focuses on the specificities of `TransformersMultiModal` compared to `Transformers`.
+
+On top of the difference of arguments to provide to the `from_transformers` function as described below, the only other specificity of the `TransformersMultiModal` model is the format of the prompt to use when calling the model. For all other elements such as the output types and inference arguments, it works exactly the same way as `Transformers`.
+
+## Model Initialization
+
+To load the model, you can use the `from_transformers` function. It takes 2 arguments:
+
+- `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
+- `tokenizer_or_processor`: a `transformers` processor (created with `AutoProcessor` for instance, it must be an instance of `ProcessorMixin`)
+
+For instance:
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoProcessor
+
+# Create the transformers model and processor
+hf_model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+hf_processor = AutoProcessor.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+
+# Create the Outlines model
+model = outlines.from_transformers(hf_model, hf_processor)
+```
+
+## Model Input
+
+Instead of a string or a list of strings (for batch generation), you should provide a dictionnary or a list of dictionnaries as a prompt when calling the `TransformersMultiModal` model. The dictionnary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
+
+```python
+{
+    "text": Union[str, List[str]]
+    "<other_keys_depending_on_your_processor>": Union[Any, List[Any]]
+}
+```
+
+Some common keys to include for processors are `images` for vision models or `audios` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets or if you are using batch generation).
+
+Here's an example of using a vision multimodal model:
+
+```python
+from io import BytesIO
+from urllib.request import urlopen
+
+from PIL import Image
+from pydantic import BaseModel
+from transformers import (
+    LlavaForConditionalGeneration,
+    AutoProcessor,
+)
+
+import outlines
+
+TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+
+class Animal(BaseModel):
+    specie: str
+    color: str
+    weight: int
+
+def get_image_from_url():
+    img_byte_stream = BytesIO(urlopen(IMAGE_URL).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+# Create a model
+model = outlines.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+    AutoProcessor.from_pretrained(TEST_MODEL),
+)
+
+# Call it with a model input dict containing a text prompt and an image + an output type
+result = model(
+    {"text": "<image>Describe this animal.", "images": get_image_from_url(IMAGE_URL)},
+    Animal,
+)
+print(result) # '{"specie": "cat", "color": "white", "weight": 4}'
+print(Animal.model_validate_json(result)) # specie=cat, color=white, weight=4
+```
+
+!!! Warning
+
+    Make sure your prompt (the value of the `text` key) contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are images present in the value of the `images` key.
diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
new file mode 100644
index 000000000..21a048f9f
--- /dev/null
+++ b/docs/features/models/vllm.md
@@ -0,0 +1,186 @@
+---
+title: vLLM
+---
+
+# vLLM
+
+## Prerequisites
+
+The Outlines `VLLM` model is intended to be used along with a vLLM instance running on a separate server (can be local or remote). Make sure you have a vLLM server running and accessible before using the `VLLM` model. As the vLLM client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies for the `VLLM` model with: `pip install openai`.
+
+If you want to use the vllm offline inference mode instead of the server mode, please refer to the [VLLMOffline](./vllm_offline.md) model documentation.
+
+## Model Initialization
+
+To load the model, you can use the `from_vllm` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Make sure the value of the `base_url` argument of the `OpenAI` client points to your running vLLM server. Consult the [vLLM documentation](https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html) on using an OpenAI client with a vLLM server for more information.
+
+Based on whether the `openai` client instance is synchronous or asynchronous, you will receive a `VLLM` or `AsyncVLLM` model instance.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the OpenAI client
+sync_openai_client = openai.OpenAI(base_url="http://localhost:8080")
+async_openai_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+
+# Create a sync model
+sync_model = outlines.from_vllm(sync_openai_client)
+print(type(sync_model)) # <class 'outlines.models.vllm.VLLM'>
+
+# Create an async model
+async_model = outlines.from_vllm(async_openai_client)
+print(type(async_model)) # <class 'outlines.models.vllm.AsyncVLLM'>
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+# Call it to generate text
+result = model("Write a short story about a cat.", max_tokens=100)
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```
+
+The `VLLM` model also supports streaming. For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In ...'
+```
+
+## Structured Generation
+
+vLLM supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import openai
+import outlines
+
+output_type = int
+
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+import openai
+import outlines
+from typing import List
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import openai
+import outlines
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+result = model("What is the capital of France?", output_type, temperature=0)
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+import openai
+import outlines
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+result = model("Generate a fake social security number.", output_type, top_p=0.1)
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
+
+```python
+import openai
+import outlines
+from outlines.types import CFG
+
+arithmetic_grammar = """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
+
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+result = model("Write an addition.", output_type, extra_body={"guided_decoding_backend": "outlines"})
+print(result) # '23 + 48'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.
+
+An optional parameter of particular interest is `extra_body`, which is a dictionary containing arguments that are specific to vLLM and are not part of the standard `openai` interface. Among those, `guided_decoding_backend` allows you to select the library used by the vLLM server to control structured generation. You can use the value `outlines` to generated structured text with Outlines.
+
+See the [vLLM documentation](https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html#extra-parameters) on extra parameters for the OpenAI-compatible server for more information on inference parameters.
diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
new file mode 100644
index 000000000..38c476e59
--- /dev/null
+++ b/docs/features/models/vllm_offline.md
@@ -0,0 +1,176 @@
+---
+title: vLLM Offline
+---
+
+# vLLM Offline
+
+Outlines provides an integration with [vLLM](https://docs.vllm.ai/en/latest/) using the [vllm library](https://github.com/vllm-project/vllm). This model allows you to use vLLM in the "Offline Inference" mode, meaning that text generation happens within the model, there is no separate server. If you want to use vLLM with a server, see the [VLLM model documentation](./vllm.md).
+
+!!! Note "Installation"
+
+    You need to install the `vllm` library to be able to use the `VLLMOffline` model: `pip install vllm`. Due to a library version conflict between outlines and vllm, you MUST install `vllm` before installing `outlines`.
+
+    When installing `outlines` (after having first installed `vllm`), you may encounter the following error: `ERROR: pip's dependency resolver does not currently take into account all the packages that are installed`. You can safely ignore it.
+
+    See the [vLLM documentation](https://docs.vllm.ai/en/latest/getting_started/installation/index.html) for instructions on how to install vLLM for CPU, ROCm...
+
+## Model Initialization
+
+To load the model, you can use the `from_vllm_offline` function. The single argument of the function is a `LLM` model instance from the `vllm` library. You will then receive a `VLLMOffline` model instance you can use to generate text.
+
+Consult the [LLM class API reference](https://docs.vllm.ai/en/latest/api/vllm/index.html#vllm.LLM) for detailed information on how to create an `LLM` instance and on the various available parameters.
+
+For instance:
+
+```python
+import outlines
+from vllm import LLM
+
+# Create the model
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+```
+
+!!! Note
+
+    When initializing the `vllm.LLM` object, you can specify a `guided_decoding_backend` to choose what library will be used by vLLM to constrain the generation. Consult the [vLLM documentation](https://docs.vllm.ai/en/v0.8.2/features/structured_outputs.html) on structured output for the list of possible values.
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+from vllm import LLM, SamplingParams
+
+# Create the model
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", sampling_params=SamplingParams(max_tokens=20))
+print(result) # 'Riga'
+```
+
+## Structured Generation
+
+The `VLLMOffline` model supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import outlines
+from vllm import LLM
+
+output_type = int
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+import outlines
+from vllm import LLM, SamplingParams
+from typing import List
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Create a character.", output_type=Character, sampling_params=SamplingParams(frequency_penalty=1.5, max_tokens=200))
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import outlines
+from vllm import LLM, SamplingParams
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("What is the capital of France?", output_type, sampling_params=SamplingParams(temperature=0))
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+import outlines
+from vllm import LLM, SamplingParams
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Generate a fake social security number.", output_type, sampling_params=SamplingParams(top_p=0.1))
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
+
+```python
+import outlines
+from vllm import LLM, SamplingParams
+from outlines.types import CFG
+
+arithmetic_grammar = """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Write an addition.", output_type)
+print(result) # '23 + 48'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `generate` method of the `LLM` model instance. An argument of particular interest is `sampling_params`. It takes as a value a `vllm.SamplingParams` instance containing parameters such as max_tokens or temperature.
+
+See the [vLLM documentation](https://docs.vllm.ai/en/latest/api/vllm/sampling_params.html#vllm.sampling_params.SamplingParams) on sampling parameters for more information on inference parameters.
diff --git a/docs/features/utility/application.md b/docs/features/utility/application.md
new file mode 100644
index 000000000..4faccb10d
--- /dev/null
+++ b/docs/features/utility/application.md
@@ -0,0 +1,65 @@
+---
+title: Application
+---
+
+# Application
+
+The `Application` class enables you to encapsulate a prompt template and an output type into a reusable component.
+
+## Overview
+
+An `Application` combines a prompt template with an output type, creating a reusable component that can be applied to different models.
+
+Applications are useful for simplifying repeated tasks where you have a well-defined `Template` and a fixed output type, such as classification tasks or data extraction.
+
+To create an `Application` instance, initialize the class with a prompt template and an output type. You can then call the application with a model and the variables defined in your template in a dictionary.
+
+For instance:
+
+```python
+from typing import Literal
+import transformers
+from outlines import Application, Template, from_transformers
+
+# Create a template
+template_str = "Is {{ name }} a boy or a girl name?"""
+template = Template.from_string(template_str)
+
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Create the application and call it to generate text
+application = Application(template, Literal["boy", "girl"])
+response = application(model, {"name": "Alice"}, max_new_tokens=10)
+
+print(response) # "girl"
+```
+
+Instead of providing an Outlines `Template` instance, you can provide a `Callable` that returns a string. The parameters of the callable are used as the variables of the template such that you must provide values for them in the dictionary when calling the application.
+
+For instance, we can create the same example as above using a a function instead of a template:
+
+```python
+from typing import Literal
+import transformers
+from outlines import Application, from_transformers
+
+# Create a function that will be used as a template
+def template(name: str) -> str:
+    return f"Is {name} a boy or a girl name?"
+
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Create the application with the function template and call it to generate text
+application = Application(template, Literal["boy", "girl"])
+response = application(model, {"name": "Alice"}, max_new_tokens=10)
+
+print(response) # "girl"
+```
diff --git a/docs/reference/regex_dsl.md b/docs/features/utility/regex_dsl.md
similarity index 99%
rename from docs/reference/regex_dsl.md
rename to docs/features/utility/regex_dsl.md
index f42b6dd34..740a0772b 100644
--- a/docs/reference/regex_dsl.md
+++ b/docs/features/utility/regex_dsl.md
@@ -1,4 +1,8 @@
-# DSL to express constraints
+---
+title: Regex DSL
+---
+
+# Regex DSL
 
 This library provides a Domain-Specific Language (DSL) to construct regular expressions in a more intuitive and modular way. It allows you to create complex regexes using simple building blocks that represent literal strings, patterns, and various quantifiers. Additionally, these custom regex types can be used directly as types in [Pydantic](https://pydantic-docs.helpmanual.io/) schemas to enforce pattern constraints during text generation.
 
diff --git a/docs/features/utility/template.md b/docs/features/utility/template.md
new file mode 100644
index 000000000..631114f95
--- /dev/null
+++ b/docs/features/utility/template.md
@@ -0,0 +1,85 @@
+---
+title: Template
+---
+
+# Template
+
+Outlines templates provide a way of creating reusable prompt structures with placeholders for dynamic content.
+
+## Overview
+
+To create a `Template` instance, you can use two class methods:
+- `from_string`: Creates a template from a string containing a Jinja2 template
+- `from_file`: Creates a template from a file containing a Jinja2 template
+
+After creating a template, you can call it with the variables required by the template as keyword arguments.
+
+For instance:
+
+```python
+from outlines import Template
+
+# Create a template from a string
+template_str = """
+Hello, {{ name }}!
+The weather today is {{ weather }}.
+"""
+template = Template.from_string(template_str)
+
+# Create a template from a file, assuming the content of template_str is put into a file
+template = Template.from_file("path_to/my_file.txt")
+
+# Call the template to render the prompt
+prompt: str = template(name="Alice", weather="sunny")
+print(prompt)  # "Hello, Alice!\nThe weather today is sunny."
+```
+
+## Composite Templates
+
+Templates can be nested and composed to create complex prompt structures:
+
+```python
+from outlines import Template
+
+# Create component templates
+user_template = Template.from_string("User: {{ query }}")
+system_template = Template.from_string("System: {{ instruction }}")
+
+# Create a composite template
+chat_template = Template.from_string("""
+{{ system }}
+{{ user }}
+""")
+
+# Fill in nested templates
+prompt = chat_template(
+    system=system_template(instruction="You are a helpful assistant."),
+    user=user_template(query="What is machine learning?")
+)
+
+print(prompt)
+# System: You are a helpful assistant.
+#
+# User: What is machine learning?
+```
+
+## Custom Filters
+
+You can add custom filters to your Outlines template to extend the templating functionality. To do so, provide as second argument a dictionnary with filter names as keys and filter functions as values. The filter can then be used in your jinja2 template following the regular syntax. When rendering a prompt, the function will be applied to the associated variable.
+
+For instance:
+
+```python
+from outlines import Template
+
+def uppercase(text: str) -> str:
+    return text.upper()
+
+# Add custom filter when creating template
+template = Template.from_string(
+    "Hello {{ name | uppercase }}!",
+    filters={"uppercase": uppercase}
+)
+prompt = template(name="alice")
+print(prompt)  # "Hello ALICE!"
+```
diff --git a/docs/reference/chat_templating.md b/docs/guide/chat_templating.md
similarity index 100%
rename from docs/reference/chat_templating.md
rename to docs/guide/chat_templating.md
diff --git a/docs/guide/core_concepts.md b/docs/guide/core_concepts.md
new file mode 100644
index 000000000..7f53983ec
--- /dev/null
+++ b/docs/guide/core_concepts.md
@@ -0,0 +1,13 @@
+---
+title: Core concepts
+---
+
+# Core concepts
+
+Coming soon. This will document various concepts at a high level, so users can understand Outlines before diving into specific implementations.
+
+1. Constrained decoding, tokens, and the basics of logit biasing
+2. Different ways to define output structure (regex, JSON schema, Pydantic models, context-free grammars)
+3. How finite state machines are used to guarantee output structure
+4. `Generator`, `Application`, `Template`,
+5. Prompt engineering vs. structured generation
diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
new file mode 100644
index 000000000..80a6c2456
--- /dev/null
+++ b/docs/guide/getting_started.md
@@ -0,0 +1,363 @@
+---
+title: Getting Started
+---
+
+# Getting Started
+
+## Installation
+
+We recommend using `uv` to install Outlines. You can find `uv` installation instructions [here](https://github.com/astral-sh/uv).
+
+```bash
+uv pip install 'outlines[transformers]'
+```
+
+or the classic `pip`:
+
+```bash
+pip install 'outlines[transformers]'
+```
+
+For more information, see the [installation guide](./installation).
+
+## Creating a Model
+
+Outlines contains a variety of models that wrap LLM inference engines/clients. For each of them, you need to install the model's associated library as described in the [installation guide](../installation).
+
+The full list of available models along with detailed explanation on how to use them can be found in the [models page](../features/models/index.md) of the Features section of the documentation.
+
+For a quick start, you can find below an example of how to initialize all supported models in Outlines:
+
+=== "Anthropic"
+
+    ```python
+    import outlines
+    from anthropic import Anthropic
+
+    # Create an Anthropic client
+    anthropic_client = Anthropic()
+
+    # Create an Outlines model
+    model = outlines.from_anthropic(anthropic_client, "claude-3-haiku-20240307")
+    ```
+
+=== "Dottxt"
+
+    ```python
+    import outlines
+    from dottxt.client import Dottxt
+
+    # Create an Dottxt client
+    dottxt_client = Dottxt()
+
+    # Create an Outlines model
+    model = outlines.from_dottxt(dottxt_client)
+    ```
+
+=== "Gemini"
+
+    ```python
+    import outlines
+    from google.generativeai import GenerativeModel
+
+    # Create a Gemini client
+    gemini_client = GenerativeModel()
+
+    # Create an Outlines model
+    model = outlines.from_gemini(gemini_client, "gemini-1-5-flash")
+    ```
+
+=== "llama.cpp"
+
+    ```python
+    import outlines
+    from llama_cpp import Llama
+
+    # Model to use, it will be downloaded from the HuggingFace hub
+    repo_id = "TheBloke/Llama-2-13B-chat-GGUF"
+    file_name = "llama-2-13b-chat.Q4_K_M.gguf"
+
+    # Create a Llama.cpp model
+    llama_cpp_model = Llama.from_pretrained(repo_id, file_name)
+
+    # Create an Outlines model
+    model = outlines.from_llamacpp(llama_cpp_model)
+    ```
+
+=== "mlx-lm"
+
+    ```python
+    import outlines
+    import mlx_lm
+
+    # Create an MLXLM model with the output of mlx_lm.load
+    # The model will be downloaded from the HuggingFace hub
+    model = outlines.from_mlxlm(
+        **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    )
+    ```
+
+=== "Ollama"
+
+    ```python
+    import outlines
+    from ollama import Client
+
+    # Create an Ollama client
+    ollama_client = Client()
+
+    # Create an Outlines model, the model must be available on your system
+    model = outlines.from_ollama(ollama_client, "tinyllama")
+    ```
+
+=== "OpenAI"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+
+    # Create an OpenAI client instance
+    openai_client = OpenAI()
+
+    # Create an Outlines model
+    model = outlines.from_openai(openai_client, "gpt-4o")
+    ```
+
+=== "SgLang"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+
+    # You must have a separete SgLang server running
+    # Create an OpenAI client with the base URL of the SgLang server
+    openai_client = OpenAI(base_url="http://localhost:11434/v1")
+
+    # Create an Outlines model
+    model = outlines.from_sglang(openai_client)
+    ```
+
+=== "TGI"
+
+    ```python
+    # SgLang
+
+    import outlines
+    from huggingface_hub import InferenceClient
+
+    # You must have a separete TGI server running
+    # Create an InferenceClient client with the base URL of the TGI server
+    tgi_client = InferenceClient("http://localhost:8080")
+
+    # Create an Outlines model
+    model = outlines.from_tgi(tgi_client)
+    ```
+
+=== "Transformers"
+
+    ```python
+    import outlines
+    from transformers
+
+    # Define the model you want to use
+    model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+
+    # Create a HuggingFace model and tokenizer
+    hf_model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
+    hf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
+
+    # Create an Outlines model
+    model = outlines.from_transformers(hf_model, hf_tokenizer)
+    ```
+
+=== "vLLM (offline)"
+
+    ```python
+    import outlines
+    from vllm import LLM
+
+    # Create a vLLM model
+    vllm_model = LLM("microsoft/Phi-3-mini-4k-instruct")
+
+    # Create an Outlines model
+    model = outlines.from_vllm_offline(vllm_model)
+    ```
+
+=== "vLLM (online)"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+
+    # You must have a separete vLLM server running
+    # Create an OpenAI client with the base URL of the VLLM server
+    openai_client = OpenAI(base_url="http://localhost:11434/v1")
+
+    # Create an Outlines model
+    model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
+    ```
+
+## Generating Text
+
+Once you have created the Outlines model for your inference engine/client, you are already all set to generate text! Models are callable such that you can simply call them with a text prompt. For instance:
+
+```python
+model = <your_model_as_defined_above>
+
+# Call the model to generate text
+result = model("Write a short story about a cat.")
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```
+
+Most models also support streaming through the use of a `streaming` method. You can directly use with a prompt just like regular text generation. For instance:
+
+```python
+model = <your_model_as_defined_above>
+
+# Stream text
+for chunk in model.streaming("Write a short story about a cat.")
+    print(chunk) # 'In ...'
+```
+
+## Structured Generation
+
+Outlines follows a simple pattern that mirrors Python's own type system for structured output. Simply specify the desired output type as you would when using type hinting with a function, and Outlines will ensure your data matches that structure exactly.
+
+Supported output types can be organized in 5 categories:
+
+- [Basic Types](../../features/core/output_types#basic-python-types): `int`, `float`, `bool`...
+- [Multiple Choices](../../features/core/output_types#multiple-choices): using `Literal` or `Enum`
+- [JSON Schemas](../../features/core/output_types#json-schemas): using a wide range of possible objects including Pydantic models and dataclasses
+- [Regex](../../features/core/output_types#regex-patterns): through the Outlines's `Regex` object
+- [Context-free Grammarq](../../features/core/output_types#context-free-grammars): through the Outlines's `CFG` object
+
+Consult the section on [Output Types](../../features/core/output_types.md) in the features documention for more detailed information on all supported types for each output type category.
+
+In the meantime, you can find below examples of using each of the five output type categories:
+
+=== "Basic Yypes"
+
+    ```python
+    model = <your_model_as_defined_above>
+
+    # Generate an integer
+    result = model("How many countries are there in the world?", int)
+    print(result) # '200'
+    ```
+
+=== "Multiple Choice"
+
+    ```python
+    from enum import Enum
+
+    # Define our multiple choice output type
+    class PizzaOrBurger(Enum):
+        pizza = "pizza
+        burger = "burger
+
+    model = <your_model_as_defined_above>
+
+    # Generate text corresponding to either of the choices defined above
+    result = model("What do you want to eat, a pizza or a burger?", PizzaOrBurger)
+    print(result) # 'pizza'
+    ```
+
+=== "JSON Schemas"
+
+    ```python
+    from datetime import date
+    from typing import Dict, List, Union
+    from pydantic import BaseModel
+
+    model = <your_model_as_defined_above>
+
+    # Define the class we will use as an output type
+    class Character(BaseModel):
+        name: str
+        birth_date: date
+        skills: Union[Dict, List[str]]
+
+    # Generate a character
+    result = model("Create a character", Character)
+    print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
+    print(Characte.model_validate_json(result);) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
+    ```
+
+=== "Regex"
+
+    ```python
+    from outlines.types import Regex
+
+    model = <your_model_as_defined_above>
+
+    # Define our regex for a 3 digit number
+    output_type = Regex(r"[0-9]{3}")
+
+    # Generate the number
+    result = model("Write a 3 digit number", output_type)
+    print(result) # '236'
+    ```
+
+=== "Context-free Grammars"
+
+    ```python
+    from outlines.types import CFG
+
+    model = <your_model_as_defined_above>
+
+    # Define your Lark grammar as string
+    arithmetic_grammar = """
+        ?start: sum
+
+        ?sum: product
+            | sum "+" product   -> add
+            | sum "-" product   -> sub
+
+        ?product: atom
+            | product "*" atom  -> mul
+            | product "/" atom  -> div
+
+        ?atom: NUMBER           -> number
+            | "-" atom         -> neg
+            | "(" sum ")"
+
+        %import common.NUMBER
+        %import common.WS_INLINE
+
+        %ignore WS_INLINE
+    """
+
+    # Generate an arithmetic operation
+    result = model("Write an arithmetic operation", CFG(grammar_string))
+    print(result) # '2 + 3'
+    ```
+
+It's important to note that not all output types are available for all models due to limitations in the underline inference libraries. The [Models](../features/models/index.md) section of the features documentation includes a features matrix to easily visualize output type availabilities.
+
+## Generators
+
+Generators are an important type of objects in Outlines that are used to encapsulate a model and an output type. After having created a generator, you can call it using a similar interface to a model and it will generate text conforming to the output type you initially provided.
+
+This feature is useful if you want to generate text several times for given model and output type. Not only does it prevent having to include the same output type at each call, but it also allows us to compile the output type only once instead of doing it at each generation (which is important for local models as this operation can be expensive).
+
+For instance:
+
+```python
+from typing import Literal
+from outlines import Generator
+
+model = <your_model_as_defined_above>
+
+# Create a generator
+generator = Generator(model, Literal["pizza", "burger"])
+
+# Call it as you would call a model
+result = generator("What do you want to eat, a pizza or a burger?")
+print(result) # pizza
+```
+
+You can find more information on generators in the dedicated page on [Generators](../features/core/generator.md) in the features documentation.
+
+## Other features
+
+On top of more detailed explanation on the concepts already discussed here, the [Features](../features/index.md) section of the documentation contains information on additional Outlines features such as applications, prompt templates, the regex DSL...
diff --git a/docs/guide/installation.md b/docs/guide/installation.md
new file mode 100644
index 000000000..c626841fd
--- /dev/null
+++ b/docs/guide/installation.md
@@ -0,0 +1,73 @@
+---
+title: Installation
+---
+
+# Installation
+
+## Dependency Management
+
+We recommend using modern Python packaging tools such as `uv` for managing python dependencies.
+
+### uv (Recommended)
+
+```bash
+# Install uv
+curl -LsSf https://astral.sh/uv/install.sh | sh
+
+# Create a virtual environment and install Outlines
+uv venv
+source .venv/bin/activate
+uv pip install outlines
+```
+
+or with basic pip:
+
+```bash
+pip install outlines
+```
+
+
+
+
+
+## Optional Dependencies
+
+To use Outlines models, you need to install the Python libraries for the associated inference engines/clients. Such libraries are not part of the general installation as you should only install the libraries needed for the specific models you want to use.
+
+Outlines models with the installation of their associated additional depencies:
+
+- [Anthropic](features/models/anthropic.md): `pip install anthropic`
+- [Dottxt](features/models/dottxt.md): `pip install dottxt`
+- [Gemini](features/models/gemini.md): `pip install google-generativeai`
+- [Llamacpp](features/models/llamacpp.md): `pip install llama-cpp-python`
+- [Mlx-lm](features/models/mlxlm.md): `pip install mlx mlx-lm`
+- [Ollama](features/models/ollama.md): `pip install ollama` (after having downloaded Ollama in your system)
+- [OpenAI](features/models/openai.md): `pip install openai`
+- [SGLang](features/models/sglang.md): `pip install openai`
+- [TGI](features/models/tgi.md): `pip install huggingface_hub`
+- [Transformers](features/models/transformers.md): `pip install transformers`
+- [TransformersMultiModal](features/models/transformers_multimodal.md): `pip install transformers`
+- [vLLM (online server)](features/models/vllm.md): `pip install openai`
+- [vLLM (offline)](features/models/vllm_offline.md): `pip install vllm`
+
+If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
+
+!!! warning "Hardware Requirements"
+
+    If you are using a local model, your model may require specific hardware. Please check the documentation for these libraries.
+
+    Some libraries like `vllm` and `llama-cpp-python` require specific hardware, such as a compatible GPU. `mlx-lm` on its side is designed for Apple Silicon, so it may not be appropriate for your use case if you are on a different platform.
+
+## Bleeding Edge
+
+You can install the latest version of Outlines from the repository's `main` branch:
+
+```sh
+pip install git+https://github.com/dottxt-ai/outlines.git@main
+```
+
+This can be useful, for instance, when a fix has been merged but not yet released.
+
+## Installing for Development
+
+See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs/guide/migration.md b/docs/guide/migration.md
new file mode 100644
index 000000000..df766d8c8
--- /dev/null
+++ b/docs/guide/migration.md
@@ -0,0 +1,246 @@
+# Outlines 1.0 migration guide
+
+Outlines 1.0 introduces some breaking changes that affect the way you use the library. You are likely concerned by all of the following sections, so please read this document carefully until the end.
+
+This guide will help you migrate your code to the new version.
+
+All previous functionality will be supported until Outlines version 1.1.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
+
+## Removed or modified features
+- [Generate functions](#generate-functions)
+- [Models](#models)
+- [Samplers](#samplers)
+- [Functions](#functions)
+- [Text generation return types](#text-generation-return-types)
+- [Inference arguments](#inference-arguments)
+
+### Generate functions
+
+The whole `generate` module has been removed. That includes the functions `generate.cfg`, `generate.choice`, `generate.format`,`generate.fsm`, `generate.json`, `generate.regex` and `generate.text`.
+
+You should replace these functions by the [`Generator`](../features/core/generator.md) object along with the right output type as an argument (on top of the model). The output type can either be a python type or be an object from the `outlines.types` module. You can find more information about the output types in the [Output Types](../features/core/output_types.md) section of the features documentation.
+
+Associated v1 output types for each deprecated function:
+- `generate.cfg` -> `outlines.types.CFG`
+- `generate.choice` -> `typing.Literal` or `typing.Union`
+- `generate.format` -> native python types (`str`, `int` etc.)
+- `generate.fsm` -> `outlines.types.FSM`
+- `generate.json` -> `pydantic.BaseModel`, `typing.TypedDict`, `dataclasses.dataclass`, `genson.schema.SchemaBuilder` or `outlines.types.JsonSchema`
+- `generate.regex` -> `outlines.types.Regex`
+- `generate.text` -> no output type (`None`)
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = ...
+generator = generate.choice(model, ["foo", "bar"])
+```
+
+You should now use:
+
+```python
+from typing import Literal
+from outlines import Generator
+
+model = ...
+generator = Generator(model, Literal["foo", "bar"])
+```
+
+### Models
+
+The model classes found in the `outlines.models` module are maintained but there are a few important changes to be aware of.
+
+The functions used to created a model have been replaced by equivalent functions named with a `from_` prefix. The function `outlines.models.transformers` has been replaced by `outlines.from_transformers` for instance. On top of this change of name, the arguments have been modified. You should refer to the [models documentation](../features/models/index.md) for more details, but the overall idea is that you now need to provide a model/client instance from the inference library the Outlines model is wrapping.
+
+For instance, instead of:
+
+```python
+from outlines import models
+
+model = models.llamacpp(
+    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+)
+```
+
+You should now do:
+
+```python
+from llama_cpp import Llama
+from outlines import from_llamacpp
+
+llamacpp_model = Llama.from_pretrained(
+    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+)
+model = from_llamacpp(llamacpp_model)
+```
+
+The `load_lora` methods that are present on the `VLLM` and `LlamaCpp` models have been removed. You should now handle lora loading through the `Llama` instance in the case of the `LlamaCpp` model or provide it as a keyword argument when calling the model in the case of the `VLLM` model.
+
+For instance, instead of:
+
+```python
+from outlines import from_vllm
+from vllm import LLM
+
+model = from_vllm(
+    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
+)
+model.load_lora("path/to/lora/file")
+
+response = model("foo")
+```
+
+You should now do:
+
+```python
+from outlines import from_vllm
+from vllm import LLM
+from vllm.lora.request import LoRARequest
+
+model = from_vllm(
+    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
+)
+lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
+
+response = model("foo", lora_request=lora_request)
+```
+
+The `ExLlamav2` model has been removed without replacement. This inference library is not fully compatible with Outlines, so we decided to remove it. You can still use it until final deprecation, but we recommend you to migrate to a different inference library right now.
+
+### Samplers
+
+The `outlines.samplers` module has been removed without replacement. You should now use the arguments of the inference library model to control the sampling. Depending on the model you use, this could be done at initialization or when calling the model to generate text (so when calling the outlines model or a generator).
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = <transformers_model>
+
+generator = generate.text(model, samplers.beam_search(2))
+response = generator("foo")
+```
+
+You should now do:
+
+```python
+from outlines import Generator
+
+model = <transformers_model>
+
+generator = Generator(model)
+response = generator("foo", num_beams=2)
+```
+
+### Functions
+
+The `outlines.function` module has been removed. It is replaced by the `outlines.applications` module. An [`Application`](../features/utility/application.md) serves a similar purpose as a `Function`: it encapsulates a prompt template and an output type. A difference is that can `Application` is not instantiated with a model name. Instead, you should provide a model instance along with the prompt when calling it.
+
+For instance, instead of:
+
+```python
+from outlines import Function
+
+prompt_template = ...
+output_type = ...
+
+fn = Function(
+    prompt_template,
+    output_type,
+    "hf-internal-testing/tiny-random-GPTJForCausalLM",
+)
+
+result = fn("foo")
+```
+
+You should now do:
+
+```python
+from outlines import Application
+
+prompt_template = ...
+output_type = ...
+
+application = Application(
+    prompt_template,
+    output_type,
+)
+
+model = ...
+
+result = application(model, "foo")
+```
+
+### Text generation return types
+
+In the previous version of Outlines, the return type of the generators depended on the output type provided. For instance, if you passed a Pydantic model to the `generate.json` function, the return type was a Pydantic model instance. In the v1, the return type of a generator is always a `str`, the raw text generated by the model. You are responsible for parsing the text into the desired format.
+
+For instance, instead of:
+
+```python
+from pydantic import BaseModel
+from outlines import generate
+
+class Foo(BaseModel):
+    bar: str
+
+model = ...
+
+generator = generate.json(model, Foo)
+
+result = generator("foo")
+print(result.bar)
+```
+
+You should now do:
+
+```python
+from pydantic import BaseModel
+from outlines import Generator
+
+class Foo(BaseModel):
+    bar: str
+
+model = ...
+
+generator = Generator(model, Foo)
+
+result = generator("foo")
+result = Foo.model_validate_json(result) # parse the text into the Pydantic model instance
+print(result.bar)
+```
+
+The [Output Types](../features/core/output_types.md) section of the features documentation includes extensive details on available output types.
+
+### Inference arguments
+
+In the previous version of Outlines, some of the inference arguments were standardized across the models and were provided as positional arguments to the generator or through the sampling params dictionary. Additionally, various default values were added by outlines to the inference library models. This is no longer the case. You should refer to the documentation of the inference library you use to find the right arguments for your use case and pass them as keyword arguments to the outlines generator when calling it.
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = <transformers_model>
+
+generator = generate.text(model)
+
+result = generator("foo", 256, ".", 10) # 256 tokens, stop at "." and seed 10
+```
+
+You should now do:
+
+```python
+from outlines import Generator
+
+model = <transformers_model>
+
+generator = Generator(model)
+
+result = generator("foo", max_new_tokens=256, stop_strings=".", seed=10)
+```
diff --git a/docs/guide/selecting_an_inference_backend.md b/docs/guide/selecting_an_inference_backend.md
new file mode 100644
index 000000000..2d52941fd
--- /dev/null
+++ b/docs/guide/selecting_an_inference_backend.md
@@ -0,0 +1,5 @@
+This guide should provide a general overview of the available models in the [API reference](/api/models/).
+
+## Models
+
+-  [Anthropic](/api/models/anthropic)
diff --git a/docs/index.md b/docs/index.md
index 3692d32c9..1ae1999bb 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,8 +1,162 @@
 ---
-title: Outlines
-template: home.html # Note that this is managed in overrides/home.html
+title: Welcome to Outlines!
 hide:
   - navigation
-  - toc
-  - feedback
 ---
+
+#
+
+<figure markdown>
+![](/assets/images/logo-light-mode.svg#only-light){ width="500" }
+![](/assets/images/logo-dark-mode.svg#only-dark){ width="500" }
+</figure>
+
+LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
+
+Outlines guarantees structured outputs during generation — directly from any LLM.
+
+- **Works with any model** - Same code runs across OpenAI, Ollama, vLLM, and more
+- **Simple integration** - Just pass your desired output type: `model(prompt, output_type)`
+- **Guaranteed valid structure** - No more parsing headaches or broken JSON
+- **Provider independence** - Switch models without changing code
+- **Rich structure definition** - Use Json Schema, regular expressions or context-free grammars
+
+<figure markdown>
+[Get Started](guide/getting_started){ .md-button .md-button--primary }
+[View Examples](examples/){ .md-button }
+[API Reference](api_reference/){ .md-button }
+[GitHub](https://github.com/dottxt-ai/outlines){ .md-button }
+</figure>
+
+## See it in action
+
+```python
+from pydantic import BaseModel
+from typing import Literal
+import outlines
+import openai
+
+class Customer(BaseModel):
+    name: str
+    urgency: Literal["high", "medium", "low"]
+    issue: str
+
+client = openai.OpenAI()
+model = outlines.from_openai(client, "gpt-4o")
+
+customer = model(
+    "Alice needs help with login issues ASAP",
+    Customer
+)
+# ✓ Always returns valid Customer object
+# ✓ No parsing, no errors, no retries
+```
+
+## Quick install
+
+```shell
+pip install outlines
+```
+
+## Features
+
+<div class="grid cards" markdown>
+
+- :material-shield-check: **Reliable** - Guaranteed schema compliance -- always valid JSON.
+- :material-puzzle: **Feature-rich** - Supports a large proportion of the JSON Schema spec, along with regex and context-free grammars.
+- :material-lightning-bolt: **Fast** - Microseconds of overhead vs seconds of retries. Compilation happens once, not every request.
+- :material-lightbulb: **Simple** - Outlines is a low-abstraction library. Write code the way you normally do with LLMs. No agent frameworks needed.
+
+</div>
+
+## Supported inference APIs, libraries & servers
+
+- [vLLM](reference/models/vllm.md)
+- [Transformers](reference/models/transformers.md)
+- [llama.cpp](reference/models/llamacpp.md)
+- [Ollama](reference/models/ollama.md)
+- [MLX-LM](reference/models/mlxlm.md)
+- [SgLang](reference/models/sglang.md)
+- [TGI](reference/models/tgi.md)
+- [OpenAI](reference/models/openai.md)
+- [Anthropic](reference/models/anthropic.md)
+- [Gemini](reference/models/gemini.md)
+
+## Who is using Outlines?
+
+Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) use Outlines. Prominent companies and organizations that use Outlines include:
+
+<div class="grid cards" markdown>
+  <div class="row"><img src="../logos/amazon.png" width="200"></div>
+  <div class="row"><img src="../logos/apple.png" width="200"></div>
+  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
+  <div class="row"><img src="../logos/canoe.png" width="200"></div>
+  <div class="row"><img src="../logos/cisco.png" width="200"></div>
+  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
+  <div class="row"><img src="../logos/databricks.png" width="200"></div>
+  <div class="row"><img src="../logos/datadog.png" width="200"></div>
+  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
+  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
+  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
+  <div class="row"><img src="../logos/harvard.png" width="200"></div>
+  <div class="row"><img src="../logos/hf.png" width="200"></div>
+  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
+  <div class="row"><img src="../logos/meta.png" width="200"></div>
+  <div class="row"><img src="../logos/mit.png" width="200"></div>
+  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
+  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
+  <div class="row"><img src="../logos/nyu.png" width="200"></div>
+  <div class="row"><img src="../logos/safran.png" width="200"></div>
+  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
+  <div class="row"><img src="../logos/shopify.png" width="200"></div>
+  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
+  <div class="row"><img src="../logos/tinder.png" width="200"></div>
+  <div class="row"><img src="../logos/upenn.png" width="200"></div>
+</div>
+
+Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
+
+Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
+
+
+## Outlines people
+
+Outlines would not be what it is today without a community of dedicated developers:
+
+<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
+</a>
+
+## About .txt
+
+Outlines is built with ❤️ by [.txt](https://dottxt.co).
+
+.txt solves the critical problem of reliable structured output generation for large language models. Our [commercially-licensed libraries][dottxt-doc] ensure 100% compliance with JSON Schema, regular expressions and context-free grammars while adding only microseconds of latency. Unlike open-source alternatives, we offer superior reliability, performance, and enterprise support.
+
+
+## Acknowledgements
+
+<div class="grid" markdown>
+
+<figure markdown>
+  <a href="https://www.normalcomputing.ai">
+  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
+  </a>
+</figure>
+
+</div>
+
+Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
+
+[discord]: https://discord.gg/R9DSu34mGd
+[aesara]: https://github.com/aesara-devs
+[blackjax]: https://github.com/blackjax-devs/blackjax
+[pythological]: https://github.com/pythological
+[hy]: https://hylang.org/
+[.txt]: https://dottxt.co
+[vllm]: https://github.com/vllm-project/vllm
+[tgi]: https://github.com/huggingface/text-generation-inference
+[lorax]: https://github.com/predibase/lorax
+[xinference]: https://github.com/xorbitsai/inference
+[sglang]: https://github.com/sgl-project/sglang/
+[dottxt-doc]: https://docs.dottxt.co
diff --git a/docs/installation.md b/docs/installation.md
deleted file mode 100644
index f88bc7125..000000000
--- a/docs/installation.md
+++ /dev/null
@@ -1,49 +0,0 @@
----
-title: Installation
----
-
-# Installation
-
-You can install Outlines with `pip`:
-
-```sh
-pip install outlines
-```
-
-Outlines supports OpenAI, Transformers, Mamba, llama.cpp, and ExLlamaV2, but **you will need to install them manually**:
-
-```sh
-pip install openai
-pip install transformers datasets accelerate torch
-pip install llama-cpp-python
-pip install exllamav2 transformers torch
-pip install mamba_ssm transformers torch
-pip install vllm
-```
-
-If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
-
-## Optional Dependencies
-
-Outlines provides multiple optional dependency sets to support different backends and use cases. You can install them as needed using:
-
-- `pip install "outlines[vllm]"` for [vLLM](https://github.com/vllm-project/vllm), optimized for high-throughput inference.
-- `pip install "outlines[transformers]"` for [Hugging Face Transformers](https://huggingface.co/docs/transformers/index).
-- `pip install "outlines[mlx]"` for [MLX-LM](https://github.com/ml-explore/mlx-lm), optimized for Apple silicon.
-- `pip install "outlines[openai]"` to use OpenAI’s API.
-- `pip install "outlines[llamacpp]"` for [llama.cpp](https://github.com/ggerganov/llama.cpp), a lightweight LLM inference engine.
-- `pip install "outlines[exllamav2]"` for [ExLlamaV2](https://github.com/turboderp/exllamav2), optimized for NVIDIA GPUs.
-
-## Bleeding Edge
-
-You can install the latest version of Outlines from the repository's `main` branch:
-
-```sh
-pip install git+https://github.com/dottxt-ai/outlines.git@main
-```
-
-This can be useful, for instance, when a fix has been merged but not yet released.
-
-## Installing for Development
-
-See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs/logos/cohere.png b/docs/logos/cohere.png
new file mode 100644
index 0000000000000000000000000000000000000000..85f6e6d9904b0d49302a2d933181279e974a7c86
GIT binary patch
literal 113536
zcmeEvcRbbmAAg-FBXP_^=qR#M*UCO+Mksqz%FLEMI@hfZWfdvAtn7q{aJo?<dy|p9
zPxj{b{;2M~<#YP|9>2f8e_Z!Gbl#uW`~4cv*L;5jTvAn_AY&xkvu6**1w~o)J$v>i
zLI08L2fumH;CUAOZ?Chu!udVfb<E>?_OR`_AbU>J-FT{p{Be(t9{-lx^b<qRM217O
zWOCukwvF0Bw&onfiEZbjI-604o;-O)LgwyoMkkt(KqQGpO<cZAOH?VGmm2)pTrGNE
zuey@C+O5Dx<-~P+_imqLneEtfo1^wx&fvnI$w{?SPd2;rmbZN7Bs@s2Wx18&EwV<w
zd@S_hxXnsLLdAx@wP!B^<^vh6B5|8`6#PH#tV5F|Ng&{setRnBJTZ~_?Ky5$0tCRf
z*h1&_|85I(REl)f8E0Ylx3lgRx|)gL2}r0;YNFY+PUZ~29PbBN1eZPG{h<>s?Zs$5
zeG^DSm<UPI8^ZhB^FJm4{rVph{>OxG_Vzzb_@5^HPZR#}7ymPd|Cz)8%;A6L0IcAD
zO!)t10_HG_-`bHhr@zM)4g|+MLaC~qaqyIt@>y+1j{j9vOXH-(glc+kIYJ?N)hxZ@
zeO2x;ryHNRgp;b6*dMi5y!Z0!U7s9Yv@#1U+Q!G;kNuA*v`B_ll#<TUI+n-RY)g*H
zAjdBSpAC0n4y=~kHVAfkPCXuHlIpEc=jfHJF8bJ+Ia6@{t@(*V!t6c13Bt>M+KoYc
zJE@0hB?6H~r=h2~BWF*A{=Wli>U$5beo9uF58EeN_X%kspCrIu$<98iAfI&iL(dfr
z<3lm3)3G02l>Su>GzL*~uKlf-?}%SWco~)Gp^+F$ZB!2n*H&8ReYIpI@AS(VewjF_
za1IDnrD&-X7r!Jk20-Uo(=nYK>UiQWT4yIeKNZttrct4XoTWfIY6owOxu|_TZaUG&
zY5K>%$z0Fzudl0SDY=H#L^OsqxaLwdg4Osa-ls}A*lud{v^+P4V9Rt9WO$Q)2gJT*
zM<#Sd?$)E@%2xevEQRY;!s4j7SQtcOkBUkQt9!F^|NC$vl7&!Qs&j?YrNy+ECQkOz
z5KZ;@w%g-O(#<jgmJ!f~yN<JkTGJ%`vBSOuVvL}mW2K0!l&4+N$_VXjZO!%(<fwcF
zyZ&=Cu74~CeDJ*7Zz{W{m{cT$>eClxEM!?z`s}G(x)ou}EME^j>%|{?F)h_GnW!CY
z%b@NM6Ls~iu65Q>8cm`d9ge9BsH*wWJfcOHzr(k0;Jg>H$@7(xYi^q%ZSISCMOGSh
zp>THey6kqa4aBl=WsqfNuM6`*JE3KJvH4mj#&xA3Z2p<LsK^XyOkK%xvud9v&QXc_
zKWSvbMTDDtMPG_y5cbXK7x@ci(H6_TWf%9DpFrP2my?hrtqpV3o`a-cNj}wUbYyMp
zKz-W5gNC8bWo+t(RHAhu*blURMKnLQxWlP_I}CGt6%rp3zMQIvw{MwXWtV{-(ZkOc
zI<Y0*bPU>Jra!KY$4w%$iFsowe4f}t)IHGUoUBA;WS7C+8i(Wm<tK}N|8`2%dZbJ`
z?d+DnGUsQf?x`uG28d=t2@#Qc&rR0iW01Dpr0x*b^HHkvIG&o%hq{9rh`DK9nY**6
z9}4`!9OL3#=kb(!d_-KBxyx;`0KYjGrOdq}2^(rE#jJpOb4W?X<S<9YU2m)nq($6|
zs`;X!+KsAokd=7DV9_Y>-_4OQ>-1-1H_EUs?;1xMnWP;ygSID&#CahmV_uPyBz38#
zvq1}7qH%Kd9wYA#%!tk9V+E<Xq=itn)2~<-cX;#9CXZ4p&W1F8a;{aH%3>D%lEajj
z=jJ+g?E`e!7GS{YCgop9sn{Cg5e(DCQy=$PeH~D&wG|y0pdJ6XDVmn{lO#-ofOrFa
z8b66J_k`MQ;0hMivU205A!1+I5C|?ce@M+e^-XKY3vYC6+|GM7q}26@-Gmug)3<K|
zDJdB~0%+~<W@m$@r9@IJTy~!t-^^Bh+;(c_vo9X5@)Q~{1m|0bzJ*Tjvulw6-#m%I
z$?52nJN7G<@uDQz)xF(hV>JKF4f+N>KeiZ4w{lh1Q%AN%*k|URU#@*$S#li<UJ!a#
zBzRW&OtT<()|fJeuXQ58KXqt&pQt3ef&@D%t|ii+F%a7Nk4F&|tFtL=7$}X`tIOJ6
z@thgKRvC3ZPXYhj0HNNgGUhBKVBjZ&Oo?k8Q`+2prNE=216Ih~v`;MNoc|sW$jlC@
z7+)KfQ=g_Me%j|9Ur$qn6!e&B@EE)DW1G-pjw41*e8w+uvi7)KsLN$npTaVS@O&J5
zfBdHh8PiqG24D?qMBC|e%D*I*irp!L1V;*rIh^6KJ4W?^c1L07%l2>&3}8F3Kc9(Y
z;W;j==8L9MZP(umLGj~_NAgv`2LA>sf4h<UBBaukv>j4XzH?tN;^q8+(WxC_sJax{
zAIko1J8PFrO#&AlX;88}=_zr;f02r*R|Jf+)I+ni{_%tY;$cG|&WYq^JMO8)2-ZR&
zP}zqpoler`C($Q-AU-6o<b_I>ww&@<R`B$up9b|PQx(7#q1n!z&3vHNJ@oXDZPEOk
zQn1CSqFM-6CFvgl?R-FgB)0n~y8Q*G!|k<HU+a@LME{->o$wcU2(@u>7$A7*FWjn*
ziD%OV<>>cUS07$N)?=}?axsY<{}nb+3M&k=NdRqLZIz#Ug7EfMT%6oIbf$CQOo&Ix
z(3!p#APmx5X+&Zk#6NM8;6X_~r2P3A;MC}?Yr)mg!Lq)Sd<ARe1uvWRA?1q#%GYjp
zu^c>Q41o#zIFaWRh!0`bFlq><|K&W=B%elN{OldgeNChfn(YUxB`%@ahE*V=!8sDY
zdgD-Yhr-1BQ?Ta85|n}vN>y}*gZ$@b?7_5>c%M|*=o%0;*y44?+hwQRl7vXh14R+(
z4U7X{Zy-ht4i4_y_8V?UPNObC&CwypTTWRjlS2pnSLqUX5%G1%a>t=y_Zq&%HJ$1|
z@<mCh#!SG2wtKTag73cI#(c?mA9q-JRRL8=34E8fRt3j@%m{j1f)S!{6x-Jg+<}2e
z{miC!Nnkp_D2yrqGcmpy3Gj87r?}vw;=)HId`xwQs^{6~G)i8`@d4BLkLF=sk~6C!
z+ea`989|dHQkPv1&eO0pcmYxOmIxdLx|3jwc&B5$mG!!Fqmg!1$>c$fZ6?_-&p;77
z$$rvtnr~mK|Lou#b|4Kh09P`As%dHT5OkqU*H>0P-rbLnQ4}pbt!~({kIyFTCuAU@
zI{OND*GVbIC??M902{RX5+LAidqqFMJ<Er<S4OQ*6C>QO1iHJ*#<Zloto|=cp#x3w
zJJ@Ir&;>~@t}fck9RRD&z5+Gh8gN`REBTEh<wGqL{+tq=a=zze*V?35S}lJT%Pn-p
z-Ka3X3AF9j=4nWpwh)$-AU7dqQy>gD75KAdhl!CM5-3$mnF}X=k|{VZ3sMp37w7D$
zXJ$V0PO|BxFew07a0H^r8?p$Z>SN!0{a8ceXfiE77VBJdrhw}w+Xc@!z)*$HJgDjG
zx_#4`H5DKPW|1B^yJd>A@!*6uw1>;Arvh@rO&Gc2XpqtEx^=F<YDGCw6mq<{T77<W
zxxQ}O%rD5C5%B*4FgWGHWF7GB&uXUEtf#EPqWM3Dr6a%8JLl|pm;ao`{UnB6bf7lk
z?OPpw!}U5pIXSQoRq_?`PpbFCL%!`JDpN{)c;$(r;}=Hd&R?tc;cEStf>(NKqxWyE
zmr47!znu%RzltDg1F=tm@mw7gw2*(Z`nU%sRH8;S$PoK089Q+BlWOMBR3Wz?p^Q7v
zTGT|FEzfYE!2W6Z*h0DWb$!6c%ya!|gW@cXE~a0jVg}@pvoD0gU+#p2|2Ykt123lA
zsZKigtUx&>ge`#}*riL$5C$K$xlVhxcF>q{l}o4_(jlX@_1+8r!q8i|Kq9A#Zl3Wm
zW%?BRjZcBXx{z4tK{U(vrye*t21WczHP~2!YPN|vjS;zZ{}i=>?Jp1M#~>ipmYz%A
z=rWjC%;^Vl5L%pg<GO{oP%8*8aT7#M4J{o4iBC+T)DLE5|7G-xYJwPrbVEaR_4>W$
zd9N<*%f}EzN+3+*-lW+BR(q1hh)9*qw!D?8zlV*Kt$5g$K{@*0_;JVHgZV-!1vRz;
z_BD%qK97aO2wEmtS`EHnv-u}aOGx>iXO!3&6rHP=tw_}WY$pmB^j|)r6CNT&N8Jss
zzw57|&-NKDcbzk}K(aNM05jPtR%-?pQ~8g_U3eCQ-;!-%FD}x8T>5_i2RyY*o9)Bg
za=53KS<03t6g=n#-9lRnSua7BMvSpFGfNdz3${Ux+MovR)4$UHPhJDI4Zc722WFx0
z@`ZXU$j23Do|y);hru>Zm9tHQ9hD%u<Gm`a?o#q?x+E<OV9#c`aW?KcPQMM&Bh<iR
z>QTe4Qa60>yBpxJFQ`GtuN0mM$%h>AZct5{oziaOb@<>1RAm6)je_uR^#46jgA8Kz
zVT7i~+pY1}<ugxH3Y2c)#7Ie!+DLmrM7t1=^B5dtVik5@iK1P#@Ij?-&$P?_%+*`C
zTw<pRR*IAn<z8u@chmR5HolsIsLvj<OB)77cAC}<rxROVL++ouYSMmTCE$f9#;;a%
zyS?(s>ZDh;GX+3zB=95(kB^=QxdGWhHSKa9R4)~>uPH|$>fW#We{mH<i?>ZlW|ewj
zfw$0k^Wi+utYbj4DutLH01PKTINjx8m#*mx!nwwou8Cc;^fc@b4xi_tHy`oiL^*}j
zKLdvOZ$cI>vy)45&Dm&F%vYbu{T1-`A~NWz4mh#I=*D=or|KBIl>+&v0st#%H!Erc
z5gF~-?FnTyvDvYa$W-&KaYe_g7zHliTz*!~F-{EsaJb(MM^9fhX^QLg2qJw3D#Zdj
z^8<Sk*Qq;7Oa9m>xAX8Fa=$4nk$y2TdKW}A$YZ2?^7B{kF4xxeHolKi0!kzU^ogaI
zH;bCBK@aogW3GklP`g5fhM3pO=AWzr+Kdm^*Ku>}EgWW?$!@%6{Gi+|oDA@O5>!B0
zCD5jp?RCx?$#d_qE2qx>B7CsemQJ^t%eY`=h_9Tw#7(4IU+)Zps8}B`p|w3x6hS1$
z_2Q<})F-97oB&#2-%6bIu8>dtX%iqp<QmxAw~r~#XZ|LB#qi09Fc0Pr5D8R!uY<t3
zKLf!|lbG{~s}zM2HiMF|pC2%W#Z4n;WrFlKjgE@e9s|gl0SwY;^{6TU7xIH@_kHb?
z7E-V)DT_0et3Oi)0Qi7x<86wHNwIk!D_&f)X*EbCffg`N-7xPMDV2KyuBzpFU+6pz
zFJGZ5_UmHcwoiSXRjSBaYM<~cpNX{u&pQuA&M&x?ZlMdwZureC)!sc6KUYmVD**BU
zgY_1MpZ5+R<1fTrjbH}ePc1ooU#8?a8kPa|vhu6@sM#FiabH_H-0N1`H7Ytae`47V
z4~jKFKGHi<?=#N1Jl$Q`5_@za5Q_9jlC%@efwPc2aXQ~-^mUs0YALp(S0l*g(J!21
zyJU=c-;cXv(c|4CE~R6Pct|tG56sW6C)J9Cswf`!$a=AYg7X7^3U-?j$<+UA&;l?b
z95>2Sd$`-r$0s+nxQ%x&;sbC{9*4odB;y(?#<Ihkauh1e)($M6xc*}nzzHq0%Mh!N
zIUJWW%)6Aq@3+V#4RGNK*aoFZTpKCXiYLZaCn&aAA}Vs5vU>R|i~>+%!Uxtg>`m6*
zmU=C-%!CA7JeS}N*f`hoI)D^xq{k!<pR{s2d>|Ed)y(9O)yZEJ@D@4%pje+)Q)3MS
zaFZ7Uj@(yGumDa{((5K8P(Ue6SU@AIWy%5**T4<#-CE$o=Wtq1!se;=c1BrTlLOHj
zh|gQ^Mp_?7Tt{{YpK)z#4DX9Uo#LtaH1bpWwxbKW^i|T`&uh2I$U&$Bp@+CzvIc@^
zhV!(b=-LMD>>AdE`|z(D-h)vm`y_0=Xc6h;BttWH4&s(^)FaYxJ=0Z<%e5`u2C~IT
zpAP+;cJ7cXGDtuwy9EiG_s!!a0Hf`|Qn$X4nH_mo4UNq^RoJg-hFk39jz1msYdr84
zX`(9&w<r4+F90^U<bVLt3oma0;xe9ojmvR}UzHo=Cq!G4ax6yAYef_REX4w$%)DVR
zLb5sJ;2LBXUvu&KadWHspa0dS+0gyRhz_OFme8k6;!|XAp_hSMD#ok2fRC~_VwuGS
z<a})ciw=_PqPPbWK{lkwZ;)AAYL`5j#g0x807UD~ir)rt578aYbd9h9w;j){x63mF
zs&<%|c_xPPRdEwq0B2#qH6MlcTy(&Qij4}n<5MZ@RuY-5U`Y#+CUtnVM-CjYOKUHG
znIa|({24u9u)$%vV`$SXP#H)US3l89j||bZ;amSDeuj?nnLTvN)ym51!vPEX4F(jz
zGQeYNpUqg+gat(R`}e9=akTc>s(LQT{d;!n1cIo^f@h!O$1RWx8KWIfkW%SB1<P+=
zdIWgyi+LK*IL(zT32LG185uv9Np=Jcfc8Lin#V`T5nMTc3%3T4r0qjb8i>|TO9(C9
zsaoY!qTIV%R0$dgx1?~_Qy1$?b2JW!-Q9!PhCl#h{VtlOmIHww6vWoDX&uigP`Vz5
zkr(z|*nHjlPa7tG12oKogykt7gia%t=Xmcr5^jMS7*dxx9QP}9p-m-;gOg|n$(tA5
zax7@-I4%hTMt9nMenZrD?pv+N{ezirbL5!80dIv}@gxbea>5ZV6~UBI^I0I|BVesX
zOzioDd_<Y7iB(Q-|4*t9d=AErp(?C-o=L#jetJoMQ6Yg90R5)JJ#)5(WN1yK@%>l3
zA-aq^RAX(na{RRj>KHmf7f{J9OQ$)Rg{!)@Sm<1FPnq1*VYeLPkX2m9xo3^GhvTet
z6Nx+_IPs56xsK@JePQdp;8^>1Asf50WH+N>|LZmZe_7bRN;4L$sa*yX4+Dy4veT(g
z0yNH^02P7)tcx|*-h;cM5VnS95Fd-2s{6DpD?1d&B0Eo143WJZ`RoLu^yKM3MAtGz
zoBSp#dlz@=FhA1;nj_lV@;Ggq1PKBt9#GNFNwp?Y;&k}3ay^8RIrEJqDk$8}ec$fs
zqJ=c{R+hZ&R+QAyI$feRIgn(XnmTl%ffe!YVJkim+lh6T^}BeDY!{REB9=*#!i`H$
zOBP0hNNE}gVx|w5=@8dx6lFA8<5fg`&!F~K3%~K$ZDmTR&^e)?@zp-=kNNM9UwRHu
z?2>UMDTsH!SBqnUv9X&Xu>rf~N#Y%g3jw$*-4E=e3uu7Nd|os<jY!_rX!b8dJx?`R
z&+}>fMSvgtnL)SEPe7Hyu=9DV&2{<H;+O&uV6|~uN@E_798{b3_8h8?m#=uI^vBCH
zySGd7nbSc~Zc6`oYhUnRz`p||)E+fwz^Ienzj(jwkJN<;?DKGIhj+VmoN7f4liYGY
zmb|^aj~w7%8*ph^(VgcK1VMH%$XS7VDjh_dn0%xw;g{V)#>QgUf2q#n!@#EvDNxGG
z=Kw;jt+7TVBM(0*o?>!hDLRIVQ;&K|0f(aQ2+^_!PB%=7%C_DP@-40tk!+-bYK9E}
z*h!*SDpTIp0$L-e0YRva`)@k(J+cEVeydbbxo)#qZXt6ER}Hw#<NPuZZAEk^Ro?Fp
zr?o}w;ghuu`#~oDD=0J+<H`kjlo?n$@|N?m><j$St6=qTi8Bq&*%aK;boT*D$+d8i
zWbEj%Z+(%U>6^+SSZ%c*-8nSXxn4<^^$H8(0wVPjW!{N2MQZ)eMkVII>{AH-B$tGQ
z2?79*jyp7)MGE`jr>3SFAGzvXOt=Qn?152{BdNqC(`)T@_Ey^5A}QFs@!f|87*vc7
zUP-Lf-@L2eXLSHG2f(-5xt-(0*P94-jn*1xffV-20CyOEQR$YVRc8Y$$Jh9C%KnTc
zS~CE4+qY~ZhM?qzaC<&%K7;-5_<g;kyR>o-MwlcC%c=`htM`t&64jXaUO!}yn~pUb
z;se9sC+T%V%q79)O>ovnq)%N{Ya8^0iC(|U><oq)Y*)$Qqx^9Yv_6_uR_!_K{&z3-
za0-lPzlA=AC=}7?^)&U|@?%gpx{Q$I22GpBd(FZaZQ@hR<x`ja{r9bd&WiT3zS=Gs
zgXC5j5znGni+A5%peQWNueplw;RO#EtPx_wtbaE}tC@RMA<3}R6^Ix?+PpTi6Dw9(
z9Xj@J_N|+Ndlit2zmc*LhdpbSRWS8a+5S8GOotJo1Jm~k*4$->BJeLCp#6X`ugx2)
z$bsk~OCvE-E6Pe(JqG*Nui6QDkvj74pj9%@DISJ)gTk6gDHj<~sFcx2)Jcf}!NERT
zo^-0+vf7KtA{%PCZ^3J5;%A@1l*QSg2Gq;9SS63DNXPVA`r6BwL7iK$szrx5pdq$q
zOWrKvo5EAvNQAjb1*#|eirGnNPln#%l=l-?_q>d;IJw&`0B%nr3gcWI#-Bm_vh&F}
zO5|w*6l~qi5&<B1r(w5pcPUAcMiEXo$D}aKjy+eid~Yt_OkoEdK9ZzB5myA}Z(`S}
zj@<G7G=D#OE_rh4PkVnh#cyKI)Mp#rFuNx(UtT_e5T+ix4BT0Wam*`H0Et_zk#R0f
zVJi3l+V1DDr(Q%TM5T#(o37n4C@#~Wptei|NFVfA2|1Sky-gO(yXhh8?Uh>Z$^ysl
z4EmMSM5g`xE4cVK^V`|Qe+ChGUIb@1V@NSVER!0ZbzJTAZ`{aGoRx(IgoPMf*?Uog
zG7(U5UosqJ<aj{_Dm2^;H;%pN1$F5l$uq$t%W@To?IjGOyImtF=!ju}NEs|9_)U83
zN=OzSf&CZGC>Ii4y*BaEzQ&`UpMBsUGaTa<UZLy8=PIAPe%%{)ewip3l-tD0!sv*K
z<@<Y2Po!Yob=w7#fv?^zvNn{1!oehadqa`sfCAdwNOU0(S@Gtb1%-ZNOqfwCx$}vw
z^M{8oz;Jv_9AhBjGox$(fVQy3?cotDt`uqjV$hVp%S|x{XnLs0#(ds4ZT)NY9P$C6
zDycry&et+x(l#JmL%MJu9M&RRDe*d8k;r}ht!pi$*M&}t>=qM%o}sGRdrSf%yy2$W
z?2kayqV>13h!dXVpvrwaHLjkM_a|b-n}6z_e>*lti#%!NgSq}%+di))U&mIV8^=K`
z2MTeehi#FdU%uoMQMHKeko~p9ytCT}BuS6$F1F|Jf*67~4D3}l(zc}+dARJQ7-OyO
zp~Q_{ehqL8I0Vtr0I1ldEd0uG-4N)~D8r36y)=iTx;!paM=*i>6n`_v3{EgzY+ozi
zVkr8&cWNX|)?DWjcz|;=a-@i2a|i)-ijHQdI(z`O_Tp}#f^KO6u1r2_OA;BToMpOF
z2BM|KS}n$;cH&pl>4$_gHm)a~&26`6l-|u>5L;1IZD(DM-Y#&}RQCrZ6%e|t?~hSq
zL%mX+{;d8n!UyU22UINDtrDMk9CQSlZXd9AE~GxzFbQ~TTz+nI`3<h^W9~dpqqm8y
zSi$JsGcZ62sM4==Fgy}Aaex}})2A|7FscA&#nPsE5tLMiy9(w%Yo@)CM%CT=xwrj&
z8yGY4p+4)Dk({S836a2wmwJ17;;kTfR!bjRD8ssIs!=}kfZMd#u@|M6Gx}Nr{A11~
zkO06zrdv~=4F=T=x)8N-OPRyNX}b>!Z1p_G&$4AgP4|m_4T(!EP>9KFS((jKUt4C6
zhUhXLmZ0J)%k8reVuj@}jbfNd;iYIly>;$%*#u|6s5(9I0Q64ZC&-_Z*YA%BLs&zn
zvb@}HIplH$YTNM5<i+5MA@OXI30|V#X=yX51rOto6eRNsL_dUKGngLO*aR_)$!T$M
zul(r<y5pA;z}C@*$pB(O&!=02Rm!{C3pHQ`Q-1;ie?bpQFT)@QGci)8@FI1A3V&9P
zqY<j6>r2!AcUgVrda*7S#_x!nhC8+#$>)f9oaj9P{F)20vi10-zW{9#Eg^`um)rG~
z!9tKYWZ#1G;vjM%dy@mkfOME9x#WJ_UWha@bx&X83>%zyDgGF0I{m7Ym?iR|Dt_I}
z+U+arJQ<1!n}E|x(dml7bAS0bVZ%N!d=^HHLCZK4_ZAr6=(j@IC7<i0`6}S3!b-P6
zWQ+T%xp<fTuAji$L|CKO#W3S7KefbbO&v?8(FKh`>*+h(R4a@iA<xYXpA%<)_U9d#
zz3xTaCf9t{)u9p3zrlQfO4lDOq;P*l4HYGa#vFndbdb4E4(&c8_>I`{uYTMns%l-*
zBFo8#Vn+`n_<&mIbL3$_z5bPvkgMvvrSW}i^Wj~EIOz1CRl?O|&)YI$-1Jc(RxW`K
z7y*xd8?`$XChMV@g*IwiMK*@`r?H`b_&mwxaD0E02h@$tJ!B#N05qtH)IXef32MsZ
zF@B55oU!@&R;v#$_x=6L9Kv^-nml<44D*;xx|b!AN`TC)dXVkKehly@lq|jrTmati
zdNM~0V3H&dK%c@)gCcN>E1kw&G&Hj_{XjD{93k_t71Zywl9_uO{@LY!KDZ>}`b4a6
zn$)H-*{2Lik|rr_?YUp>3Dlu6QX#258;YpmRKl_xtvddJA$7F3W2>5*TDKHPXVKO5
zk*?`V^@OFM*(ub9faZXksmm@M+>0>ccJRu3xv&Dcom}PhW^Gc?P4gt>@2Rq(*_qQi
zF?Urq=EW-|SGcGEoxp~nE$@7sz3p<YHR43WOJFqD|6r=-{hn<zuYeh%blBn_knJRK
za0OZ{$Fb)#ClnMtIuG1sppEb~aiVVlnUy~)7CYa4S&af_<Dh}b9cor~Mk9yb^UeT~
z_DQx-{bS+RiSBS@_&q|!$^8kd#*)XsA*#%#b%7dC2OrPe_trVMuh$k^k+bB}S?P?h
zN=(L3G@Yv)T^}DCE2+P7;Sk$bDiBT7`-7UIJV!=-HPUZCybhI~{84mSx{CLlVFl{D
z{W<vp85NYdb~~}?-p8sChwD6m-E1t?iXpG9?|C!Vr8-sSk%Nd2U|`FGfrv$Rr(K&X
zcy`4NIco_`x9m3dvFjjUO9cU2d1sBlSZN{Tczpbd`4ztvq*f~PmJ|ZX$;Hm=!7|t7
zA~ay6-?ecz{=vlI*ZwTfy|LpY%rd5yFFkDoi^$-ueNCS2gEEuAV*7Hf#Qt$Ea9lN`
zNOSRHQ1bVzYSm55@sR-BZrK{R_5woT`Od*70HIS*cYUln*Ea2meIR!4Y4{V0V3^0R
z-z#|CzSwOzGVTkAE%${T5<l+nrDE8EGG(Msu44SCSkoE!)z5KrsY$~8wp;Igz|5J;
zg6Z7|W1xU*oO?|5`ZqH8CC8JJRaoz1odaOF3$(YKc&=X9dcM4=ParD08`yY{EBl%r
z1LjM>L^pfoa!$>cLoj6lki=LJAF|W$@_wau;U<0QR4Qpw^%mk2>pKlI{=gW&1H`Zj
zSd9v0H-4>~QJ%s(ALaaG1XNc6^VBG@FO1vj3`#iDj42ozLfy3<ol5Mjl-@jWLRyVY
zYwp0Mk=ergCQcLY-_qFK!{${_nf^Nau=5Nh?$9QFP?PjcOKN5+c?X=+A%F2~8=|JK
zLt|d-O(%?+!{p&5sRv0i(=APnMloV-twwwUSQi6opmmt;^se6V4k?s3KUWT@+#=Fc
zFNmkw5Xq;!k9*|YyUdCSU+xnsd<LH!T<A9`f04gB)opHp*H4{4YCyf$@0iJhd%`1<
zh;EMeb=D~xVc4_&FqH;~L!UQuX0{d0inloEe+)9Xwv&P;F9Ygo(IjItY|;xRFW5eC
zG72nmX0^}O@!bVz$Gm)_8hbd3aXlF{Z?C{=$}L<wsYKr8jPiTrm}fvw<UXc%YkePs
zDds|P8?`<*hvy^(LR6~bahgdBMb_9;X<)&d9yuw+@wlZ?W5=#9?-ZmNks%1!m=iR=
z=&K@JW(VS4Z+er^LXGJ2TD95KhodrkLs6$D(^CXt*J12vs%B^2U)&6|vr(4b1nVY3
zyaX%#LDHn{HzX}}WO4l#>|W9RMTpfgQ;9G8+S{aWevY6A(IALQP>+X8*<akWUbqku
zV;6+2Ep0APh4q*(d=a8^(F`{8{D$R+Vm{{pW$-QSd{BS0cV*?h#sSJr6<LYoAzCm#
z2dCjY2)E?7Z}S+SX^K}bKk6es&`>EJRja};C*i(UjNxUK_$NM&2*K49z!)uU6*5OJ
zv(@BvFyGl?O2{j?r(I{Yx~8+(WEij<-t%??7D05Aanrf^t7n2bJH>Tg_X#6RI01B>
zTvPf{o_n+<g{qs%czY0@)v9D*rsDm=XC^U^q`dEyGK1prJDeHL$jf{^{rlmq*^5}J
z?Rjgxt^I}8YEWz9o}XUsoO(xNCQZv#o?QR_4|rg1lS^p2|7EkiouPf$;qDTY6yUS-
zWA+*C6b{l>8?k58Vh}0tB%_I)-Xx#V@bit%;MP$GID3J=Ea!bMm<L$MHCRfemGKdZ
zyaKBhFUk-mTXX$i1nf(fN==OWgkyRsA<-69KBzza$C<V2m?r*+n71pi42I$w>Cca!
zc-gFzXXt+Gsl2HlIoc((b~Nq5A=wJk_#ks;c=v@t8(X%F6y7|?u0Q4ob{t}2`lOqc
zg;n|(W$$VsSJT590=P#)%XR69V&9wX;@2Sz9(TGnp(*Zd@tJPYyQ%5rNKxN(`vZb7
zy(0iB#pl>QOakW@=_<qHIc5Z?;(LwK#92GFN>A%tT~%K6c$e%;dIwI06ZRqE%O}JA
zd_O@m#)=2(oMhw4R%$k+N;`a6Cg04?2BjYA^)LV>ed&6uU87YynC}vVuIuZ2{KU_q
zeIYR{4j91V=Ljoz_m^v3Sa(rUST<3OJ;qT4>HFlHrHVes18nkfOWu@hkSmerKm0tY
zJOHcYr)Dh%yVNBDo%f)*bKPQO>vcwP)5UsFb@d!c7cBN!UwZp4L~6JVdp2_aE@$*@
ztJ-X-?v(>Bn~~yP>CJ1P)6&W>qW7mL0Djle7!$2OFaI(+{armUGXR~^06Ik%=bSmZ
z_$N?e2D=L(XTFUV=xw>QupNTkER`-R#_$3I|ICx_a;a4Dg70bgi6#AhwI8zp6Fz`d
zPq7{9&%%OW8VR5(luc8`<?H2Gy!Fti{qdb*5G}MW>Lk5|!gKkg_ORVC81x&=O@@_7
z`7B>;*`b?w*?p2AyqQRw<TlBZq0y0$N=_`4s~{bY?yZ2e08p}wnRnx_`-Gn?t8;0o
z1*T(HcC@q-cvMh0V}vb+gGM?vCfv9DjO4PJQB#nxr!3efJA>AHVFiv4+0|h=e1lWw
z@OI8XzOfE6aA_x3*Ck7a<>v*bfEL~Cylpwkg5evCj45{f`j;AL4n2YQbD)GQ$Zq5b
zZsCmfv_{w!Llf7vg&G6B0Mh)Jk<z+X5sC0wgx3#%Sr&etX9p4O4jSWx;(#UAK3e-#
zFNM|JLjXITpogo!X6fc=sf(=+A}zZ#%1PMuyT|a6K(Uyp?$v;&C_)s)g_k%j>1Sr|
zl2P+xICL8X9~|j9AGv%FkrPneQd~FaXgZgaDO0%K1`5cNeU7H)M+LPTb#)(Xv;u_U
ziQ)ITVm0sg+SYX`I~%X*KY0qIEXzC%I3K9N_!|41diOlZjt2EoDRO1&KF%&L5h-pR
z_JsO%8hxE6bl>ZOcsbar1T=JW$);sOL$Wp3y#0GO(<cGsIyV6n#VW(+`9z*d{lP$8
zmDFx#cHRWFxh(P$89e`KK|LSGz!h8A+A_#@y$})MWft2DpG}sHN5y860An6@)vbwP
z?xniM`3mYt9YLlJ5<1>L6({`7?7{Dqwagq)h|Juw3tVx81{+~)t050j2-=_Ip0B)u
z+#nrCn%`wsHl2)!jMuRewhzMYJx^d^3Bq4lIpmFpD}L0=4H3)mZ4h?z0NvrZ%wAfg
zn#?aZ3Vx86JQdnZDPlg-+cl^Gd=tQTOp%gw5w1#eaV}=i2pTwqiB1yuiMH}DXW%F9
zT{w6d3m8MCEyt?#V=O=~9$cni2ZO<I>bA)iTHf%yHl79fnl<H&%sL27rbV=7&m!0#
znt`cf6H54`lW0wmO{7rB+)j5aG2ng^G+57cCPJ(XSE(VcCLP5SAAr>eF7G0Q9!D@S
zeraa}w#g-%bL7RLe%Q&vcXv|X_Z}AAK1s-r2w<35Z+vRupU6wQ-^>9hMzckTmVqud
zzjRL1jb29Nz7Hn0@VO1i!eexq&3Ik2{gcgm{0q8$UDLM!oAy-&7b%#!zx33Ql?VcD
zbvR)zq@q<<(<{eUB{xN5)7ntLpzg($^3Z=PeLv4}_rDFd3@EHKXq**;4(CLijp4i7
zpdv-a>55xfS;1KxIOM~tE)hn+buAR&MwtrM*w;PJzsphrBLdvQ(fUZBs%2;8He7kU
zGGTK*Fw{=>T(i!8=PF6Qmn{RhX~#{X-RVVCSH9NtP*uNcAq45Tp9c1jRAYX?3xx22
zc?jQaHS52%fuGK@)a!l;O|@mHE0q}ZTJ9sjpCokP8;hsHep+h6Ipg1N(nv^bEeKq7
z7@SK@RDDJOR+Et{7ZbJpysc7%#sUF7o~th{g&>&fyN|We&TD5Ou+uA;<c*H^QybEY
zUy1bX^MDv@6RLaIUo2K14apKbuzd_Ae3}3ZIFSy;aMD|h<nXt*r1+fN3zEOfSZFdU
zL_KhnUfu9A0g_-$s;>crJ&tZ7X(iv#!{|ZY_sU`!ABy|Lk<Q2PS^)0Mu-dgu^zFS3
zFc=I9Kb?k-2yO_{+45(y5nyQ`I!{keY!eq(YBCjZXHhQ`98$NhNvluz%u;M^xU_EZ
zTMC%D6v)C&*h8ac!OYquzVf^&@Q%Yr^R-OfofhgrK5fgF4j)yLEIh;U1EAAcygKM<
zR}QFcrX*(-+3MVUt`{J0wfEq!w+6nW&@#hH257UCDw6O5^?u|33S_szoAua@g;#W%
z(AYW*PdxSd>p+)KHMplFx>Vw?-fng<Y7t^Ko&iQzeY}!3M3V3Bfq}T4v(Y^ta{03L
z<Td3nkTuTj??M5eQF=_yu<g;+Kv-bBh4v#!@>VhP+$vl%JFJljxhQ%`FFgI%@spGo
z_zi;D`>mwQ(d*~LfKvA2qKLs1-J;@|;*s-Y1Zb3#dKZD(yh1s}^|<dtCyvs4<Adjz
z;o%w#`^{L#<PHA?#+@&ueFhPiQ{Q$}D(e$wAGct^SA+?fkPVs7zp0$GU$HbfWbvF1
zFkSoN^)LH~Pjv}-zD1wd<U0b#a<Gf92?#F9k~5Mb%Fln$OJR-6+_I9rpOc2Ig^?3j
z)X)wPObE38dDcA@-yPLKv)wdAkq4hR{m0Etks*n%l*r4+2>3A~mWyn1aR$$8O;mpn
zOa}+dEE83b@M|T$v3oFdCk%N>;#~|lX{L8aXfzkNB4ha)gj(suI%}!RLRBjJRkNCC
zTQ~;@V&EoHZ;!8EKNcFV_wq;DIqdV(0?L0TpBD-%d;;As)G!fRsZN$u#ZF$ZD$jLS
zWT6KbDdp1>nCbe`f&6!OL$;h{V;Y*|@4;3SpzDv)UC&#$6F%*^9&f~^wZzacumQ?!
z(yXG}SMPQ6!sTILuYZsyEtCX}zh;Pdk}#D=4pry$AN#p`Z~S#wB7bB2d9!|pIZOw4
z$DZ6x61BHa9UKN1LL!!D4C1`qtrs}iQ@0!E<eA{rEnx~_0w9Pp^61k~^MIK5I$xcX
zpx)*OT1Qucd)$j?%;b&rF|F^de|8c~Go{BY_s+G*&?J_qD+%z~=Myj<00cwc7tGAJ
z_^*U(fKq3jm%YaX=rx^}Jy`gr5xUI>XG1S|LkCua#@7r?e=M?^2urIe;~T|L7Uq8h
z_mN<iIY?ueXQP0Jjkr|J1-#v8X`$dEh&{zn90pz2&cMPCH2C}7VbO&<fjGC?1J4uZ
z3EiTPsH*9^X6H~D%?q!c(B06g-10LT<%chd0_7ZjvrMNZ`Nw4+71dd1o?g?POV}{S
z1DEBbg{&7p{LS@^k7du&+;*lxto*l`krtVj{&Mi=L-)TWo=d6su^KG%S%1V9p#~UQ
zCT=oYA@eY?UM;Nm%>V^G%oyB`CEca6OxYOP*F3+?r2cIxY@SX{LPD>U4^?M0yPq6h
zpF!V>SUv2N<fZ6!XT@QVBt_usbOdL?_@<Xrg-Nz$E}@;85`)?rIlaxiU(uh*j~1Q)
zg{eQ4Z!~yNO|y@MtE}8;v1PYUUh$ji@p2#5G2-8<?|R~@kI|sDztvBP)T9b-*jFnO
zA%i3`c)2gx!qSxt%f@DWh#RGDd1TmH?%btwvS?6~J<Jt5b@8t9TclX$lfQ9axQ98;
zTb(1z&rgn3HP7i2kV}ZP;`}&uysxIlZ{5hIv@fTCg(coMu4hL^80<!%uiK0ZbuG>5
zQ0v}U)|a!6u>tXxa^QZo{@slZ=u#pa)R>MVdVC8-3Kr^p^ceQ;ODq07MCEl~Mde1F
zXt_Jqpo>7fOu|exwp>zrhW0zTKHtIGXoJcVUWYcsCO8{lNdcrdMsC=PUz!mflC;&?
z3#gG9HtEYVPv)zAL~g%)WFfZelF&ET$;cimV(hyZChj(4c;lj(oIUO7&^anQj!PE{
z`0q1^R1nDKh_q!)L+5<l)ard)g_d4fcn}{<s`e7GiMuh?@}W<>aLLcEmeQ3#)FP<B
zMj4E+zY%Y*dPRkj<glX0fj)+lVuLvWiuLCR5>#dzT`0Q!q_eZNtFLSrY!g^<rbjxB
znuOzEgj?@d6XdLP+lIpdcs)m(5vdQQ*1C=MHa-43u=k`X+wif^Cr`H>*$!eJuZ3@J
zgG+eUWJ!vCf_<@@HLViwdWV4maQDS#DnGmW(&uv0Szn_=ZYSZvL-0a>#HJmdScN+?
z!$i?z-vQsFLN}Y#MkVItz$Fk^A(|Ett5-yil$2UzE{QJjN8U0~jF^;aJE78ORgOY!
zUHc8Oi)`hIeX(BW-zkZdWn(m(<8dRkSpyFwBjqcMqE(;5POpF&`T*SedQb1W_Mzi<
z!{u@4bU~e5sdMsbw(*bk9pA3S%xJ>HsbvB#b*-m&lc}SmF@hqA(PtAWm%8~1+tK2%
ze-R!oOs9#;+VD%hPSe58>PLuxvN(`WhJWc$2@xBqqEPUYL<Q@nk^~dTS4Kz=0(y$`
z)*>aj7)^hC`QxaSo~qTG6dafS^Rkxz3Zm^?|DDRonKu@bw{^1KH3l1U0+YD;C;z)#
zu+uMC_e)1!RuPCDYmtX9_<NZMmTr>QT4knTG-dj59cdw(m7lGB-aK7wg{@XWg?r24
zg-|1bJXZY=B*DgJfS%BP{u;;1YwP@|O4f)IA@Fl5Yz@aTE?3LN*XuZh>^yb$rQIpJ
z^{lx}5H-`=ux1TgF9J6Z$Sy9Hl+LWlJFMxG9ahXrcD+z1=4(BG^pEM&d2seBtbW}>
zzaw%Qvfy<Y?QMC{6_jnm){2|t=jgT%q*b#wy%_;7xrBoVu5DjG&zYPWD!Dra$kU<v
zOruRPPE!IQk^@y4_@!EGk(t$bQUZ1xL=8svi%$1=c_b={yvf99zGK0SwN%ik*V)_a
z+TJ8UQx=4i*U5W&?<vt<2-l(qG_Uu+v^BZ-3b=^4ORo&)2plkx$haB5w3To<h{?DO
zoqmq8|1Xv3nIgA$iwCxxeG~+rze<NUzDs{#m{(cpi=FFv^>`h%FXPbGbdEKzZk}oN
z^&E27=yduB+cw#YsHIj6tB=LkNv7##MfX57m!-R`%K7Cao}G)(BB0MsWHEjCCDl_F
zf@#_^fRvD?a*;*u_!((xrr<Xy)MpP;yuxu<+Y`LNCiljotlP`wp_oW%#ur%`V|1gH
zG`I`e5L#?a0Q$g){nSMLg7523$m>kmw=gXZlFB9Bx~ZEp%(EHE6ekI!dV-2<o8`W4
zOXn!KKd+?kSEq1Cd{Ddi*~;PWMi%z4721f7fb=#H@zZWz%Y1$pN$;L8WfS84NR1ch
zda|-eyP9<(_$2&P1Bsw=KMz36)F`djowfi~9Jb|BC7F=Z`!%)T#v*~Wryp~;Uf#X_
z#acGP=`doy`nyZKvn-V<DheO&Oy=CjvMV!dKrb|b(G|-_xH8|u(LMu_u{w$*Yv0bq
zhl8mp;pW!H8j|oUo<Bm*xOrt+%RO3h0PuB4AqQb|Pj%fE++IXnmV*xiF(V`^w>6z-
z*50S=R~IV#x^CO$GIXYIvwrcY(O^_M?2OSZ^deFAq??ztW@!g?jwVo!WG6aZ5i!v-
zVd}lwnCB=0GR`9^>%lJtTWJrQ7Yc=@&u%h2oEDJ_Ol>cEW(~^%wo57)djKM=Y343z
zIw(`N0_uXmOfIo$9@M}ox&WSEUi$GPz!6mFp9E6+5Uru3u`eyE43n$TP5Cjas3>Kz
z1@7tqh)k@>y$kxLdkVVfl1zeu_BrKqFW5F(ZClC4=sm4^M&LkZL`}(?GgFd`8aj=q
zL(|9EPl`=bDpdGe>#Ok*5Ya0f18`@h=|<<)hCx)pr;vf_lbX7qRx~-V<rArOJexq}
zr9lw9GikDZVWv61OJ&A}P3v7Zzu2^eTm_T0uF44l#O~r!0WtoT#6Wsw2owX=E6cbz
zx=R($80U5j5Ioe}hxMM<2ZU3ypKaaI?CHgA3fqd!J*H?kIf_(h%ZipFAS(vUo07g*
zrYvP9$wPeo4C;4<(_0u4V`p6h1r5>~{w6@KnW(oNzr@m+-?VeAyl&R%{B2=S8mzTW
zBY<a%)MG#?0z9>vMwWn{TQ5lks;&zl9pj@~)$hWWL9CC5veyHzA|{lcRDBFuv>+QS
zK3x%SO(E?4T0TN^cRat7c!-e=t$f(Jh!>gERY~<HKdjg~cKyEVdp`lWwY!X3>bD6{
z14B%2T(1-_$1rMZj0^9O6AV=KQkIp-B}_}7Y)EIJ#5uM7Wt92>I-pPW?-b=MHVlyC
zkJmJt;BF)Ue3^hqIRH{~Vy?~O{QzNvJdPx$21}nUXo`z!xePb|3@zggox?Auj&2p#
z(cLV0eG8+hVqDRoxxOm;fyYK&0e*gr>Ul0Yc;o)<na|PIhL;Nrfc~})U7*#NCTIV^
z{+m^WIL-3z$+nVbt|HK}-g=##t)LMx_86tmk*j#m*m9Xb`fYvcwBl-RN$TjS=N>AH
zcEJ9%{6`%OEe3Q9!upK-xw&Cb6G#ockEn@_MXWq|`7OZZ*>Vu}TYseutXrvJZz8w8
zBu5NyEta=%ij1GvpN&tOSfaJ)D3T`T_XmDlk_#JHSra74eqM)n12^4z^A`W!=NI^X
zTSykipTzCk$SJY-SK`u4X>%(w_vwIE1%c9uImwW3VY{`w7GF6hINfZa!$wO5`+m{#
zoB<(1Yeq=Cv-+ktuQ^Xxd!j)=Jb*X7VSUxPZncsw=a|%g1I&ND3P8k%01+9!AtK`&
zOrz?$57z)v==rnV_kMSDT2Q9H55%*iwa?Rsmx(scr~hsL<u4l5ju*?>B3ql!6S)m#
z;HD!1G*~`B2hXnwlxjq2>gjBMEK@4SG&XYttND9hPF5y?yAuQm)>2o^z6W-uSIeE1
zWumCHLUCWUKut~5So;gvGxz`#M2hOck@fWp#vLG1%mI-2h4YdNNAHm^R*l~-s;hqz
z|HRAoo%K#0ayuRbC#);$m)$ZQh#Lt+3s(@L$r|)27WFo|)9>4X?;4{y94$5;>BO?C
zC^vWyc1&h3f`YE<wNUZtJX<%JNs(8tFq(2XzAjE1h3ZN<IhNkJ1ZX`)__qy6y2h6b
zHr;8nzhS#CEzsDf@ySXmb|XCSJwr4+<V10W0;pG%9WrmTx3d9jdN(aCHXhK9m(d0B
zLvNh<R~i1X8bOWzIp0FJLUV0<SFnVvcYBtG#*OjwF~5>O8|4V(5FF$~E2e%a%MBd8
z%8ju<W&vmf3raAW#P(Z7X|oUYMQw9TU7)vh!K5z<orqCTI(QhZ((q|%zdGHU<;Jvt
z$fOt(&9`GH0+N16tiIO0wJ?Pjev**ZLy^SoJ`aFV5;=aE?*6Gi2pmqIExs3DSNg1x
zh=R|aO{?Hu-<#K3hA&6EuxHb)8;1#mz2@Xe%xX{K$76eL+%|tD5}Mvo%8)8Kv-iwl
z@64;YmGG<fBEr-?S2h-=)-MaH0h7GyW_+<Ao<+1*lc1a3gV8no?AK0dSK6O(G72*O
zTyg#sjx7$St=Pay`bc;wlT3l2b&FX=rVoSBtUOgVb{`+8XhwIB?-V=}s8)1A@ELzN
ziVGs(1tJE#Y(*pC?+Y5Seel?P(#xe{VQK9LRHDZl1VP@i8(z0j_h4r{09`v;kX}TA
z^nz7E0AMY*3tld@CjdK6CP%;PeJ9@)!rP54VqRbrjKb{wXN|=`CU)b^Zx($%*`PG8
z<Em1>#S?4}GD@wn-g|W`UZ|!-ZdyG8WdJsZMBG)oo_T+P?y~!dPk?)*D_(vk;Co$G
zEaiNd93gQYAX-`~DNVd2r~{;lax=NXSnhR0F;bggc!uszLBx-_c~PR93_(^M_<3_|
zs6tlB!$58z@B)0Vn#hj(1&BE%kr$jHt~o7Nwp1Jmn~x_P1ETv6Ah=>X<KuitAB;c@
z7@D~U1r?83bcV_*2*mfSBtzk<dXM8ILw?W!)rA0Tk>782R>uFr@5T1$sa~hI7NI*l
zAag7CaLy>YR$qp-)H;@Z9}Z4{x%YD*)`8aula1m73Wr%|m0Zj%57p07(_%}<5?V8Y
z;F9_rKXde6awjC8qV3#|v)d0fy7UQV5_>QPBuRcVZWew=mGdCxIc$UEh+mb<HXE&p
zgvXs*xLRT-i5%%>FS<wOAdu7Q-}=AhMtc#3REnHSC3*OYrW;VX^0Qx~mERwz8J*wk
zESmJ7{=MZ6pP69gTtNe%K8R%#m#)`{GXxSaK!kS1wY*z=KRF=rpd2$OKD|zflDa(i
zcmO{80;eZU!Y+Su^9qp%nR)t<y{U`!04djKVWdQgJ0UJ>0~ZzvUh@!#xdWoTkA3LI
zBij$L3xN+l&_ZukgIo4li7nZEW5dOsi4h3qgx@{@x~%L?%}Oj9&`bW{;N5`-!-Vp^
zp;qWd{9{e(raf}DT=_R^$T2Qf<>FqHyhMjUNExwlls0xr2Q^@KB7_U}Y9}CF^!j^{
z@q!AJJ6D<hi74BD#{)LRD)HvbXP(CIi?Rh~aoR&6kABNGu7R-mp5FC_Z)*;Jt5&Q3
zPCDeg;H9mbSq3>W-^FtyeRCt`UHuVdUo`?U#bpsmf!x>^Lyp0d9;Y3q19Y~JDQuW~
z&?w0hz>h<y&_UL90m*IO?!lEZAk+X>nMN!8eeB)Hv(J8g#o!MS_iKauK@Ueu@<tPm
zeU$_nM_qpVYWik;*kBytvIkXB1ezzc1RwRJ^dJRTGgpq$bm>}W|GVMEhT91Oog!}&
zHovmI*mXTNfOMuEmz$QqhJO6eV<yAsot89P0iTHPMI2!Gx}5mtzCWZ13wrnCgdM#Z
zhwdpcs__z1xl4Qfq^sYZeGE8op|7%|rQ%xMYl9~DtAwn_pN#j^O!?%2LgTMm642h*
zw3?PrgWAs^?P|U9ZwOIAg!0Z7L4-`?&xmdnD~`MUZh;N7AfpIk*#oJd+`5%l%vR5S
zbM``NaCV8nZfc;@<J~7yy1Y;y{C`J)2O^LNa;gp{P<q&@iV%zi1a<n^H|^>v`J%kG
z3%ft_Cn>F*4pfN_4yI`LOZ}@+mY_iBs<=vBu6QmTlps390hsl-+iZh|S}|VHX6OO6
zJPrI^G9nUY%>*Ir)%Hid2o?y|J@%Pt&}@s3ZrJ~(aAVA7(u%L!bdVi-ud(OsbEv!U
z^$M*9O*Cq>FyeF+{KXyMb^y2y#N4*Sf3zg|{0NALv`_ww&}>QA{G`%h75FM86&t{+
zaCHk`PuKmBfXGTQKLSz01s`YPIf{Gjn_uIy7>rvD5w(;jpgIS&{#`$*t|NYCfeZI<
zp6v>Qheoui9(YySvsF9%l^Xq_!S|#<!M|1m2QG~2I&q(a8XkkGif*HQ&y|ZWEAbN9
zL9yju_{6dM{IW@dzdM-#=3K_j?VeG)by|9e`Cj{g)33y?j1vwmqd32If@Gj?q_xfc
zHNy8;fnl!DO4(SxJr>K})_+m~4(&lShoK`M3O6o3E$Q_>P~^pnGmxVwnHW?RpY_Y%
z&y>v#D%G%#0`#r2h?Xel_>JCNpGmYNVOGMSCoY}0*{HwAsj_insy^Om(CnxJ0qA?g
z89L|;;#8kUg;nh;ps2vzSsS7%HB%J6pZvtV6YvxLz-uJqctDm!{(?$35ES{;!`r?_
zKNFa>L3(^(WEtP}X0h*>`QirV9OFh#PjmM&uh&!JAgHa^S-=qM%85!*rdNiarw2nS
z`!KX76Kc!8jHH-k5UufPSqGIdP3HTFD{M?N9tHP0;q~N&H-<R}l)~c*H*RYSY5|d7
z6Ok5cm69BG*ZX)ifIv-*sEj|<*~aCHdAHqhqDKJJ3N+J%*0RSg#?XbSM6_3R!(ZxR
z1Bc@Wp{^I7R|IhoXku`8E`G1dM)E#73%^5jzdR<qDZWmv-tPT5Q&AxDnnrzpdj;`n
zzYECJ%(s_=;SxVclr{7n<;Q$S2J*bNbEj_83+h9sYX|WWJ_u{V^&4H;i&+Lq<K0BI
z7f^UT9GEq?{H;mWz|CsrMTnC(Ftn>y^F6N%cjU$ZosI_J<UVQVjcp~ul5^NBU|QN5
z`%ve*TyT<v7vuoeudilo_%vFXAdzl*!)7-KR56HvUJgucx!}2K42EnFeLIS&e3GNP
zT|SVNuqh;rBSfXsr3IFIQv}A!{{RP!c%9Xi{;at!<JIdRLg8#R(J4@acQff!&wxt!
z<Gld)#<OLi7Z~rhRY7C6)P<(o<DCMGlzYJ+dI0u)wYXK!lC3~2+Z8?w!bFhsj<iWi
zpUs=D;pux@odSa9h$&*|$N37_aw+;9cnd`5>dfas#=3;9D?sOua`eB5WKyIsXz_*H
z6~N%L2!F+p-kw;W`Y%&g_#j7F3jHv`!t0)8K6=X=pkZDR$0gsBK=Wj-F!yqRwn|<;
z4LpD{$DylNMdt;ab0pswWpYqmVbgl41H9NvawL4MSOvV6J@#$YfRP(0!E~#fXz+3e
zbl&ALSF1wg@(J9JV@^TWtuB%aIG6i8?$%Z9mxqoaN%Xt3JjW%p7H`iu7y6`wo-NBu
z07;?eDBZ;OrM;ue-~H0G9W1=V_971?jaNz5JO0E4L1^nm0zobnJj^zrF)>YC<jr}b
zLNAJ|GmZmXP<p`wdVrv1(*_?P;X;YZI>3~MP*Uj)ng5q#JO&5v1P8ab1kF+%Jzy;n
zn8=b+&_t>9CSWKsv@6=6FcEfa#)1bS0u&`;>e%lT>miZ4Se~@x`U4KFwwtz~p5FG}
z<6|no)F;(<;T7w&A!bs(segt+#eTcZp|F5yZ+R$p4-<fZo(K}yF+Yqvc%4#vts97r
zkk=y&oxuWq=ecg_1^`hED0r8`hzgyrDy28C+rBC@k^cHUkcZ7d-{*s%`F3I`9=@7|
zdqM0p^yK!8p+(1fR|swyoJMJh9{g}UHn7`p!EX}wBCJlk<2D{m$3unb_6vZ+vHkXw
z9nYFs{wkl=dTBsFlZEINa}%X|#M9nWXz^g6u;|p66eQ?Fn8@BZNf2xkQV{D0RwGaK
z{;}K`)uxeg98$8^SbQZw&Qhe~HNrj!+zZUj0tvSF(Ts7;Q$YWndK5q^2j18?_Mnm)
z-r9&JxW1Bu(sc7;W{TDmz>8`ZCgg$r>N*M)M+rOE=iqg3zJ<#bYpqSs??d}5iTGaB
zxt0r$O?xn&qz6$0TfRrNa+LX;nSfpPsDf;s9p#%ucnu?IQq^dk+i09-qEdReA>^$X
z1iP&KB;__N3G-nQdH8<PTc1-=y<4|I=O$@tXNOOpHE&tVZpac|Cxom*+Ut!i#c4kj
zv(}W|p{C{11qNh#e#44@e+pcN|1J{(o8*%FoJja+`Xt}GP0s?UBcOz%^CsQHX@1#2
zuLy#U|B??Tr^i8M<5Qe5Uc5Eu*|}Cod0owbD^A0%+$JpK^i!b&BJ0hiewm&Ws~J!+
z@}(Eg_w<IKAAbI7XM~4eUN+m6Vmt%N(_TjNaC9@s1rGj(A1}~K<6q{R^YA&Kp3ie?
z2kLpn2Jr!kA-{3K#owyZ8B8W~Vo>TiMgKiACm2m?YM7eXoY|hr&T=+MH}_7Pg!)r2
zqj|-q`H-^}gn7YzG8DDh>g~4l8WgXcYIivJ6mYN&yRt}lRXXQa6A@lxw?0V&{x|8K
zjVH3BfkA(?%YPi@?lZJuKsf5vyvR1}UcGS<KLvvS!jv6d>;NghPS_wN;UMUFI64mR
zHX59xdF_e~RLpxhc{KV)IdAIOQo=elK{90%lyqQ9%wp-mgzTgInH^$ZKs}FPINVtM
zr7k?c-9pQO!MD7T66g9HMZ+^pJB++{N^H6>1gk?h3<ZDnmU(tK-@T=NvCytG8{sH-
zBNXDvDX&a{Ehf>X=-(*GpXfTZvHDM?iDZ|r-+(g4k258YnWaV5#c~KA?<ILsM`Qd<
ztme|=IUx<mJeQ@~thSYc-R%h*^)P)$$zy}%)&5Vf_#?$rTz+ILabjEjElOU5C2%qq
ztu)Iy-&`uq?<m{IxTDAiPP<MF;$scgrF951Q;*om`^EPBhv4@kwf-;V!|y@I>&4~9
zE`_ihnf;B9RdYGWFy8-c?_0AP&u<}-M&b=Z-}D>&SLR2w9Y4lxUcYzZ*FTZ+y`CR_
znQhw<gk=oZFLKaeq9KR=Rq{LZ6D|={qa6CBTSc11PobRjP#@3rzk-ZmMHqDRUt)`F
zND~8K9ra$sFjbKhg}iU0UwRI6<gv=9#K7O$=8u>FiscFw^*O6=4-u$_D}nco)f<h~
zZEBtPo9qIv4-9C6%E@2}s!3P0`2vAkWs_Xs4FFMKNa0o7kr)-zf3joWOo1{!g(zMB
zL}miFf?BD+#!qCl`CM4<gG|Aj3{2Xogbj=n&i6ADe?(KdU__7O1q!%Py8px1b;ncr
z{qee8GAf(CN^VxRmSp7?GDG&JGO}fF-G)TjBV;5aWQ6Q-E6P^(CNq0wuitsDk?QvQ
z=lkmWa-Zj%^V#G5InTS)so2;Hn{%E&A2bV=PhpRE<j$ccH_R=HZikgKTD}H6$EW@H
zLvM{pp1}QC+9P{<>;;AviM|wiBg<0ngj{!%$U1v=)F<E!_}ELD)dX6?9OQjm7RM&c
zq0e_<iA~;w@7Y>)b4Pp^Ont)Q)<o0dYXBiv`goV_A0Z|+A%LuM_eQ!-QPYlJh6#g2
ziJEo9@)8@<nN#E}?hcm=+y4QrU}c|qnnLNzx2T|z+ave=SS;BLPh8b9?k_OMvvxh~
zr7RRTS)SXL?$%b3Lp4scS{50Nh;A@wZ5s#c5*c=G1D{T$^*Ud6R+YkDcW9=p;1zPd
zWiWNL+5EI1Y8sOBpN!;@vuyMY=NO-#`RjK4?bp1GMS2PF1MF)T>#18NFcT;{Pw0{o
zdix`rXdOO(MM~$p)ISLP80oW>cWcX*r~@DjQjEWe;e*sosfS0-rh}ip|NGDl=Krv#
zJ#QbNd+xqJ)gWZw`9<ZHg$AiA4;0Yytab`HyuIKi<71;PG4S*s0-ti)Zy<Y9+QNxD
zC+7@m0TevcWj8u`oozBVs)G7g{(aITYT(Yt)yHA^y{G1Ka_Hp<fb`01eo7U7X4g>~
zN~itrvMl;6j=amV2fy~Ef|2LWi$|?BoJp7=%cW-CPs!_5IU7hHKmYG4@bM@#j|Q)x
zZfnhi{-@3c8F&={^?}dbExU#9g1`P*cSU{2J0dUX`vUaUwk&h92vJrLU+1;^Z^=4+
z)S_CNzHpRlbi)%Xtx&h{2F2@je$+*AKP#TN3MOcPnwLa^)Sc1i%s)VcC!^4vph=eW
zB=>PMzz2eT_B&HgW|s!hP2KGGjQxkoRKc&<jIdF;JAUz@aaPIcVudwBQQNF5MdYH3
z2>HK@_v{)7Ehpi@!;_<nX{w9zgm{Vov6oy%rDD^eLD&EE|2!z$*XelnxK>-{MsH|t
zO)ESZiUa!B*!ai{MF&TM_cs}WS8?igu>ilBiVPdn2FYv%2-Yp6bo!-W3s4(nR?tW0
zh5-K%Xde6%YRP^RA?nuQVn4+pwcce2H-`AxKhB<TYVAcus(r}*#eZKA4x)lkfXk)j
z_^uF5{3agA(9`q<fJ<1*wpx{bO7HcLe##TligT^eqc=oEt~4cX@}deLDOfk~mg(F=
z2Cw^+(PIHK+(%F1O*}1seX6CmV1Z8(9?lMSIoHXR(6RL&@08j<n?z_1fO4)s;&b`-
zM8CV+7oI-iy9JP!)&W=)Oatei9VtsPDCqHjdELIYMjO}T016LeGNlgyo~M?jpXcA5
zxn&fZ2js)ApnMoc4!(rL2JJNT>?PjK*q04b|L4IpbTh~YZ&6u<tucd}BHa+sgROGX
zIf-=T`TuUwiD=R~&QxsHEJ7cY=z>zGL9W|xFK7>nZ+=M`Tw=i5*?Uga7L9waVLha$
z@C4Vv`{=7sK(NVvYBeZu`h2Fv?7zE}Hz+Ge?x_Qq*`@!k+QV`oKm}^)rcK8d1&Po<
zA*nI<z?GhcTs1YtukMvoFl8J8=1N~TwKY-Huz!PO<<_`ui8c1}Ti+%4PBWPn03dY0
z>0n%OFHE;w;~WfdV!g0j&<{%GxcJ*C7tfoFWVkXA6;OkFuc2zwQA$gau3bnL`HyfX
z_lzI?)~9Rsbwv0AWH2en9Md&9xhdN)xpB}<tJWW>`v+EMi&FfaoqNI3IEg>Ffg<A~
zI9@z&>bS}sg)<vAb@V!#t;EOvflNCD@*(%BHig^s!w10?At50Ad~cItyOH@LJKfjw
z`8f5*@4u<tu5y3^U5VwuQ0vP)>yf(h>lc`1!7vrBJR4HOm$6*YS$5E2#_8pxG<J9A
zxn$AhuyDt&#C8NlOSCKm)&;rZd+CgYRhMn2)!`I5t%>#Fi>iIXC#Za#%IZMl)hGis
z20tbRS_&D<E#|XrgUjAll3Z&4CNatAp;q&}%`)G}jVTi&Kr1UitKiPm(W=-+Hy4qj
zJ2!}Iu|v?q+5K*Ri7b8ZFbBSPw|FTGNZ%)5<ETCQsV8Qu6?`|sHu4u%xGz_cV5Jv=
z_?_**kc#YNSH3j~7@>c{rHim#_Ato-tz6pW>A;>_wa09*->pT>;@8^+x@86;gV$SB
z!qKcCZ%u`b8Jnrue6|;jdk4)R{bSAaQF~{1H0l8AJ1}xbTdkGsPeh*FDhz7KQduC5
zz*@J8c?EQ(7L@tiJD`~lRV6?mr6HQBVSv0S4VHJ>u(St7Hms}}403b6<<wH@zCe!m
zHULyPSwEzG&bbpgOD7ht^SpQxD?kR700voQBLIRcK51itDFf<bnY<E+yKriGP^|a+
zG4B)U|25d<U@t^B;-u9-bolr&q{k;g_=hZ*S~?P^tl?0xMin_*FB<1e<fze!)z8M<
zf~_Pgn+qC)K;j9tvZfXcO6$RA4cLUTe>{t(@|V$GIi`#i$_%RbqYW+M0>-VpS)muT
zuDQcZfX{lb+#t+n2vC3JEb+C%9d(zma^!9}`V1A-lF-;g4X$P5Mz3)x@P!Oe<y3IT
zG4!?j+{w}~K@g&jb$kVtK=@OmHQ~!dcw}+GDDf(pstMrCdrG02`Oi@<bG{?x<>@rT
zmqM^JRSm|Q(+w_!P=IDQbJfb<Gs+zUk6U$p%9IDd-mPaRmo!`GXt6&GNkl8Ko*ObE
zgC3-BZs>MHd)^=jp^f>GohwpC7W$9t=kYE|BfdZwQ5EjD?+d`UCV;grVB@uk0ZC4g
zr%lVN(x2FGzMy8atVIkdfRR3}uc`;j@q@&H<rN8i*Tf78C^MdI`AyTVSbK&jCj3E9
zu0&5U&_$BvNy<%wS3$0NxHJE`>FzYfQ%_}_0dK4~5ulXFE9TdHFW>IRxNGHEVhu7x
zhBTm*{)6D5c~IGN5Lv&iHdlc)4+Ni51PD37xjB=l0>~;Z)b+anuK{KB->7cdeBL@o
zCzu*LTaML=gTQ!kq{6-EYvfwWEpuZo82Anjm{$>~=I&1VksVDzVxih7OoKJ1LCxX2
z@F#u%<^N7DcAbmhlNifUAj%N!{n*wdQ5rN~A;wCrV~SN}PnLkh!;}0Bz}d^vZW`Eu
zFS&9<I;3FFz0|XZqcrCBeTU^%Z(s$gSp0gp?@l`RPg&c$6uw-Eqk|A~pcCyzbSHAL
z(7G&LIq$&_th%6@hOA>LHU)r#8qbPqtqIWGYY_t_Z`ZCTR>qJ~r0o)-sWP~PUGrCP
zN@!9qr@EV4q^;j%WZ1uI`k?31zgJC0Ga8+pYg^$Mad%51bP>CcbOgail{HnhjMYa4
z*~`Y$J^WfUBqu_O+hABprYMT|()C_!B67X0eYzxv)qe=AJ`A1N9gH3p^V6MDpQY}{
zs!4siu!kq_4PVP$IxhgnQ-n}s9ZOVqBA=Gl_-|$W3kr_rz)6!PNd!2sX<=b7lLPp}
z>h|Q0!sM5slOrlc9?MwCo0kMWb_&KV&RTC$y#tpc69FOi>Kzb{wwlwrzR00{^Pii%
zkCgG(b{!n*b=$0q<fj7PSCXRz4^wyPgE~4Nom(2Nk|MUoFXTD;y;=}55xou0_QLp0
zOmdKkqSl-S2<7K24Hy1}$q#4AuqHOLvbSF}sePNyjar4~FZz7ED;E{xF6aJhH8t3p
zbsd-5^2*jiKwFXH*SBLGq<5hb-_fbH(-AVOl}Arp=S;s!L4mEJan}bulzLU*7J#i>
zvt~7sjL8QK<j!9M8m=eOmCj__g!P0yuuidtzDg*#RxUO%7Q2sW>ln{be&(BO9ro-^
zk)Ywmw%nT}WZmo&LSP76v!m2OoD5;}ZbSiWk(*VVjrNVPH`ojcauPoGWgZb-kfFa;
z+L{)A4S;RrL9F8^(!9JkOsFC%l0yA79$OtO<}{2~)M=rI-*rfKvFKiXiHD7HAcU!e
z##<yn5$y?Gf7?H%TcAlP?s%Iq2<)7qCF}%)9|J|gF-sh&>=F#vAX+6>JQho>2=P@B
zb#lZ^z<6+!zt?Fq7&KNR4xnf;YjTrEKGpHA;5Sy3punbE$!9EWeSSI<cDK7HYuh{p
zpT>eZ#mxa)Ph4dk#_+`^yPXqL=x=}GJ2@?m?fGy{zk!EI@7){Mf8uPl+YoHv0{|hQ
zbgj0?cD~j6HU!HD0^xW;nusBn&+9c^Qrq~&mhIk~%6Qwmq{Vq_AC<J0%|)#3bAUE&
z*s6ZGp+KmSv&9cxWP$-E)%IM!*D-L4_~C~y#>!GO*fv{r#?n>to#ST7mrYUA9sy<@
z0bsb91HpE~%<qV+dDWGL($C2rHRA8YMrCAQ2#l6m;-LZ;=sw|KK5@DYbzxnEX0*X~
zf0X`#Ntbg&nJCmH)`O}zsK>4hNQ&P_szD~uicGlX-1czar7{jPY;_Ri08YbUd3BdL
z41pa84#4r)ffX}u*BjEAL7x#?J!Y@kwfx+(O?5My6kGHpXO41F0-w-<W{avK6~GOO
zJN}vXnJGxdB1u-xJ$d)|uMhQN;?1_6s?o4Q0|PE$pe(3QV9<^4WHD(vN5bc5GT!2Z
zG7?d2se4Qj?C@rmODPy~t>dx~sl+3G142{9U~`lI)V|2Lt`}RIY>kAxD@(*~MNiXh
z4Sjyzy53%wJ5_WFx^2s2XN+{}LdSol459@V|EOOhE5T=dOON5MxsUgAP`M=NeFGiK
z7+lEC<dt<Mp9Pb-tfB(BpDC2EZM$y<hf|70&>q0nhG%j!p=+#;L4m|T=sBm3vRNZP
zuH<{7&Dd;>YCih5$a?MP^4E23`L^TSY`vc)f&85W7@V=M*OfNuKq0A>zMoTx^E5w(
zu76m%Pm~q&4_hiOfXS3eg-<>05TXMEx~%Oj-!<g_7)=dfnt5I^>brI!%R;+jqy&^_
z(a?aZ$AD^wmQMt=P*QN?VgoMuvt5wtzUnLLJ#kwoT`{NDi3t1JI_+Z0!+g+48)lGf
z#W}AROwyjp`o79W-~S_AMF|^{8TAl9&S4rcmjzt}K?$W&Eg&&8E*@;1>56UaAtjO3
z{6tTKO{&577zXP{x>i3yH#*q*(t&lK0}zoh8Uhi?h)DCqBzs|#4iU`=g(8<;8m={O
zL5r>|BIrYhRZ!W)Lqm1w$d^dF{;j)U?C=8Jb>$W2PHYiDMih)CfhQX%1kp?Do}uyo
zKDC2|@}OFlaF58hFb6{sH)()4z>+z|+yvhCwZBHa`iu6Q!kgv6jL4YtK)8bQlRpvi
zfpX8YmklR~ET~ej6*ic|&^&bIqZ^}9?n^3}-T+sCiU3ZRdr)+JuIY}ugS)wFw|h?d
zX{xlpM{kkcIOd)w)6HDZY_zAfx2KG2D}w|+8RWA{8~mD955LJcJ9}L=sw8y5RviH!
z`?4}UVRUqJ1{%O)*0~IQz|GPUn0egdXYcCD0<8wKPAIA=VX>IDq|16D>oNdOlFJv|
z06vvw0GtZ<K$L~4<9b<HB(8mhj<@IQ6N(zx9u5X|BNY)zL(Z<H5AL)PJVdNo*BxMR
zYET|)&Pd_nBh#Rt+cLs_3xMKHT^*72Nq`77g9ZRol2wrgV1Jl4$SsRoJD_}?w>bq^
zq$HUs!_{w&b|NZj&9wPJHRw11K?VDs&^ozRbI14c+*qM2J>67O#@HAZgZbtGh~nz%
zHC}{usy)s)4}PJW1eh!poczrm*vn|TIfZ{XQzfcp5s{bQfYnQPB3S`KV0jiIa*6jg
z5K`<=N(z<~D8>|~BhGsXOxx8x_0-I<gf2JGRRt@43V7RgMKsY>RHQ2&`+Srf6)OFD
z9oJ-i&4!s4!6;;-JE2QzeykmjG~TkUnsq;+u49IUuj7m};NFwBv?W%(I@4|M?o2y9
zwuI)KT7bnePA!+R>iLm0KqQlzoyc&n><?oF5nRq5l|nz#YTD)gQ;i>Vh!dvt7l26x
zVe^qV>UAd2Weg;_SBgva2FrbLpkk2bOP=oQ%2SkGx}epoIr(V<V3|;vP*Hfnpar4w
z?LT*I5k8ZnK34F*ttBb}pFP1d*XJ3C>JQiPPR}I-w?ZRbxuAAuJQH!;d9G3@ii(QM
z+tVNS0@n3kz<9-a&bRr=lrB&7xF4~;;|%H*F;#%#N+rdo^}vvHZlY(PM@BFQ7F&N&
z9b8vdo9a)sP~W{cdtI0r+B$@&=U!I>m-|Q$zbrgzD2L@6D`+nqE1$>y;-$GJ28+UM
zJ!Rmd+;~k;@phQX<&D$cR4eckK2KI#gs>G4`wU>Tl@A@hgW}6EPKXbdYhc&f;$UpL
z_XVBcA*`9b6zQ^Q-O4l2Im#6k(QY81g--h&Ja^Rnz~M2fKwlc~rjEzhK2d?{q*#{p
za`_B(3SL6A&~*flRnI3+wu0GnDfzgymYkT|SQJp1bLKhuD8(1e{Amm5^BML-ovOf|
zqM*qt?xiV<mZBHb?lOZvsj=RP;-b?x=IO8iyy@ock(Sg}ODG)$kV#O2>q=8*^X>lm
z=8jL7<<C~|Rj5=i7A&Y;-4)njH=Ko4)uQi=8>!Q)!*ih;m}nmok~yXluCsjYZc>u7
ziVRpbPtBSkM7O2D{yA&|z)y_cpgO>hR4ITaE8R*RPIw*Y%g-7A;Tm_N6qec4?T_F+
z{IJ=`1x<J<mv{$b_@n^{*Ymj+4o0QF&@~*&z!G2Hy^r*%7!flWEZ+dI#iE0*N+7{N
z7XZVRdcu{L=u~95gI(3ibXAxPD;A9wCarkKab4Z;xX97UMyL{Y`O$q;&>5=>sKZxQ
zzS-w}>i?WBT1JEI{-`&vkX%+y_uLH@>A5EopkB5l)WPUTQ7hF2P4#K|@(Zn8s6Z@E
z?TD7H^Pr8#)uyjnX2A2H7{O*ao+hc$QUsC_c?Ah|y>03E+rl=<Bg%bR5Rv(iBD|kS
zBnK<9LckP%Z}ssPqJ}OiBGek=){$7MO}dw1qgtsaxz2l1`y^;V3Xlf~`-Dxkabfog
z4{Y~Cpu!2jR9D>9m-!}A7OxU0nn9R^WQ3A>N3#^#)t%vM`+|#fZnqa?t+C8)qdt*~
z&-S!zI{060y2Yms%o<7w^a#}%+xSfRL0EZKv?%+Smd)R$@=YT6d}Zs&m74DEeDAS#
zIW|b^q1mNVOL?puUN?MK{M#%C^-12{Q#zJ`<tjh$+2&=@Cdr|<ka;Ztb|#h3+4dMs
zq-iN*yo0eVsnp-2qjH9*6I<66WfH+q(S}^4^BAaLp&&VGp1l5~EAhok2f4WQ@24$H
z__4@QHM#h-@ep*(^X)FsF$2Jw!Gm;FC-1y%+yrMCI&T>C{I6&dARAfi$`HjFLKoKj
z!!Id9*Nr^~wR+<+Zd&ci+s3yQ2G?^wb(R}rTavWx6mR4$;`i(qplT@i)-bMoJh(j@
zhmCJCU8JnLDp->Gug6dw_=7LJ#SN)Nnx_mX==Lm{RH_ibR+#&&C&#-@V?7{oh5wyP
zKvRA~23-H`Jcj5PNFW{6`<`>&J99f6bW_<qYA^pr*~*<j9eVi2{s?%*#1;F_MB_$D
z%r)D2!9d}}-sz#<RW4Rrc{PR>p|JsIk8%3ZH*_9O{dxN#vn+J;L@h+7&w@npL!T~1
zF5mG!acR9(?o!s@SZJgIPQgRv@p#;uH;Ubg@rmyygP2vp!tHUfchb{VvAWXtABRHz
zF1=q=&~=Y{;iJ5owaJ?mZNh|V0oLJe4tO?e+P!^}c6D&E_&VhPdaLoP`r>WEzkOaf
zQwEpSy0wY>W|o%5NJ&%n+$yBhXd;kj6nW?qf~rP{r$c`xmHKX{>X#aPB}Wy5?<bG=
zZ*cWObDI(J>aq|o^UMw&VoAT#4*O!ubgIp3{hK~l_P*3)$L+^V!CFNJdCZUV&eaQ9
z);Wf@7Qe-y-7{=}IgZy@nPY<OOsmWNy>C2`viJT23nv!2q83o8{3SBm_k47LdQMe@
zsy`A;nN^67ex%^Y`TGXi7*1Gg+{yIXHH+oy%@%vtEYA3!(5X=0fg3v7n|+R-Qejfi
zf8fJ&mbZUj0S8J{CBfF4<F0Gz?zPZB)8lQ(&CVpw=874?v#5?M=ic|jL?HL@JKbuD
zp}5&g#aA;>JJ-Z+am#R82A%bG+GWeV{X_YoT`WM!5?MeDqT@d^tplm*f_!w88Ub%E
zjK<P1d}z58AYx|b+>z8KYQ%3QJ0g7{angq5UXtg1LQV!<n@8}ncqQZ02Pr7nTaHqF
zHBPY=W~`p?>UIB=JyPRt{FtHMipwG`CA4_Vb;~Wfe%xLK#Z|vyDkWy3PIvzR3~x{v
z$j%jVg(BSO0Lj_!3N#1d#K;5t|I@%Q!+$~R$3}3dO>VA*tirv5?FYrcgKj4k3MG{W
zvW*N%*xw%@7yIo&%zFaZ`<$+!>ZVUDbf&b{>-0tsCff)@KVyD-;84=X)r8v%V+pG|
zQhetr^jV&Q4I!Xi{rXSbRkTfMp<8B7YL-?&;;ZvCB+%{%*nrW^C`nQ+PDt7e<m`Uo
zl^G<0WpR!SNy^Q<Lfdk)$}|an=wztdgaBBP<8Xc{3#|KOIY{w5QK<611-F;>b3mbF
zksN`YTcg@tv&jBVd18nrF$^%!X#W^hbi*V(k-NZIV=nv~-hN_j(@y*?8f|~=J?kBz
zA0;INPs?)RjzeccGQf|Ei}oKUC60|}eB)-bK_77J2J!^!uhZ`RtCGU9m~kElby!F%
zD{mmTl0)(WX7|kaXZ^eyjLHP0Z11>cNqgVcez^SC=I<lrakN9%%Jb()sCum@Yu&|-
zC7ez}bQMtmOuJam7?rd+uONS@m(f;4fY6T`%L!ud3yu?TO}KyJ)ah+sQ6@dnfDio&
zu?h$f>wf*6q-vw$mPDxeGo`y9G4imtb<B32lF!rzMD}*K@N)StFh7P^O3Zc;SN<Je
zn(CwXfbt6gK2Jr}uVH(XEaR}&=Ii%QZWP?);JDR4b`wKkL<m^<wxj8{V_L3Gu94Sl
zuk$mG{voOmf8+2SRPL^RFK8LcWk~(c74wUJE;N#L)9`pDuF~^`Zi8emz021y^!R<#
z5~OFlIa^EZb8R=zOT7Q_0NS-5JJ@w4UwuV(Sl7kLGNhZc9*hKzg<v7~rAib}xs*EJ
z`*Nr<`T7YAyBG-0`bqXDU5`_+RoKu9Fs3JcavTf0Na301-{YV*ENd5Ne3@Pm(_nVp
za2o8IW<)a}<jalpbb>>+p?cPrM6pOY<5#w<l!u!$VkRA-781(b9hhUqT?M?RmdSra
zV-h}QVLw_%dR^B9+STuyN7TVA@-EivkEEN7FKfN@_QO~JL^q&zgX<=ag--Y7$>U+^
zp=DXhmS&g!eEI+ij$d@!AT%iSFjcC0-p^+ELM#DB)+KR{lg9n*^HwV?c(+Lqc15Q~
zu$}PtrJ<d8CZP$ZR!24HoklE*rWSebbfDU{DYX9hz#*1XUJnRgem$!XK36?q6lTY^
zY+(AkeZU;GOrh^)XuXpl|6IqD&d#-dHB{5iJ8S{>1ZYG_-WaEC=?t&lx_Rji85SB`
zfprJCb#%AnO9y;8b4dbYV~CHx1QzOp<Glr9oCN7G`Ps23eP2mxp)y7=sQlME#J$ky
zJ|jowt<i#<p_D**uT$GQC<+9pi1LlaSLAMvHfRWlHLJLeMFO(ntD5S1|E(=ejq^u&
zDqn2amP|QSu>Z#k`U_i3a?Pr}W(lmie=+a?iX6X)v%Sdl68~DyqK3BkO^f7hwVH7U
zQ)shi$(3RX?sc3yUt0PmBJ)3pIloz0gd8<$q}JwMU8Z#_hx&?>5nTq`_K&_1>mn$g
zTZ=1sjU^bhif@i81}Cup^?aN}A4!<O!t>S1_v0h!(|ZmAG1Cgn-K;6OR^OuwGTl58
z<Gt)gdf#{qh5sTRGR6`%&bSd%m}+hGJx5o6`v?6jK>njfDA$^rZ`k(wPxRi=dGy!n
zi2L`mKOS29)*(t`g5Hr%R4$b+qCD^>DT+lQL9wVnrTvc8b2f<s*p$bEzR#(8s8v<%
zY~k9f&l!&_@}+I!`QZZ68(kjqw0mnDWtW`aOa9HF?Ajm71OB&1rq7W_OFhAIYOIAK
z9r<X7wuMZf_yW*9$w*l9jfK*npG!FEQTp40wO@%)0{CqYGiN@Q+^o+{jV>(_lJ(9!
zvVF=V5?w@a_-h}7Z<UGBRwBe6r6FZm$7|{7DE_2^5v8R>?u(0sDXH^y7lprW`*46+
zISW6YhFBufh)I4ob^Zdsa~t8i-&v251V38TWI&qIXM@=dxs`+Z6x-AXtUD5<*|$8L
z<UAzM#%`1z@Z|L0567wX;aEqL!LK}-ej+)42WbyIjJk#geC#7ulVj>J$w`}ROj_rE
zoX9`;x6&NXxi2(mo~2W-lyjx4r-x@^NOhOMo7e86$caSt;K>nVhfAc!Dk|V~cz>Q7
zk3kb{P8d(v^4_+~e<rpG6IdW)1|cWy28S<aA#DemsMRuXNwA1q3@mNsYFMqdwd<GH
zINeXmcX-pc1#EWzEzcmN?R<8=)s}qRHGntz^Z5Y8j931!fmOS=TXV@=IRkm@q3zAx
zBLzB>*eH3mrNhJHX0g7YuauK4!I(VBU#}qGn#r!=8tXJ5Y_Yd1QCgmC+oe#C3^;qe
z=pfuUd|dWkYeNMNJ@H>hFB1cSk!IGh3gw=!Jv~w0vX%imf-4PqEP>Xx++_2&r4k7W
zSk*?(6s0kACB=DmPKVt2&f+%Z4xmao(MYdE`T!X1-ODS2`ubrgm8IuiddR=sB{B%U
z%4kA(Xsxn9wEC2f{MpPMyMlIy2q@wC772Ww(8g?G9H*}-^1)yJ7MTQs6ce{gp@n^r
z*<!XmzPJ-V?(U)iQXQ>;WO|wa%fGFb-a%*ODJqpgRxX|9p3Uq2JcgJ6577De17NsM
z$J~>eh5PW|>9}>Pj+UjzXJS`1f4z!?a`}C}TgUsn`2svc2I?Ie7d?j@k9P&3B2Tlc
zMkcN373PC~DQ*lBwrP!HcW)`SZV0F2{;p=npn>cW9I2W6M0NM>-3kGX7U$?evL~g~
zY=2ta`$%cH6Qd3}ZWv(;hx($W^m#I}9U0m5f;RA66;JxgMiBjjvt5WJc1jff-_fx!
z#d2|2ZSBvc`W2<MDyk<1cuPBEMrK2y=D2|P{4to)jfGzB(Zh(pIY5Ww#*s^?Q>~=A
zW$Em!Lw7=+0J8-YIS7(LHt;2tFMZ0p^Qk}8AJU78B*>G_^)RPOGa1jjlc-{SaEEqf
z${^@4_2rEG;ApGES0CB(i`bZdJtb@7Y_;80yQd~%r*|}KnGo{S<nPGdm?v+kNNiCS
z_Q9|38iQS*2mI7VCSry+MMDL2C3lvLgAx&^I=*u^OiWCwotC->v@6drC6J5mKDE1d
z@$CD8%!FJMCl*{fS69Z@$h3m3cD6T@1K8$p_BPARhf!NGLzwf~kni}<|1ft*C%E#C
zk^QNdz_KeXk&uJ;ZX*B=-n})Bwk+qshd-r7Jof%L2a(O=YA&tTg7sMj<v6mPr;lg@
zm|tIQ9z83JgX8ML>+xy-i!7}E6{tVg_QE;Evm`Z{+Be2cL(?fgfwbKroWlu-aFp*T
z--K<>J`}9jP?y(#>rH%W_td*S(SaaOFZT(hc4nIGiloG;dg{0x@}jOm8BzAxaVpK2
zgd{f)DQkifk`sT4D>@pd$FH6H)kkkf4q76do&8S$QBBgi*&mM4mR`oBOLJH8J^$SN
zTO5}xy--|>j^LsYu9m|R4%>V80>&o!$DpYy9*V5)vTd(v+7^j~Gx=@TmOpbjL@kV|
z`#XwrRJ$qvLF-)nibM?m_6~%BtLL`ipL+HJm>bu?xE9ZHx??+=+dIQO3Wvt-^aUi<
zoO(&#snhw}{PFYwd;jqgM`8{k*R5Krn4mD1K`yoT(2i*M$sBA|Fn3EjoX-aIbj68}
z<nFFutWW|6ZE>@OEeScPM%$Z}%d>PsVw%N>_|l(=Fv0-F+0B^i-NHHb!&A+se<$rh
z9sycf-sk6e`7)~VhV7TbnT3qAE<`>)DSt>%15=-qRKoa!XIgo~t3G0l%F!h|+mjPG
z==Q0^p=Cl*1eH8&)XTHv&e9)ZipjY4#d*k=@F;&58Bx*p2TFvYCDe(1dOMi}$^n>b
zuaSIzC&L{PEM#=GGz^_v`eP5o;{IsXe4Pf@<#*Lh4WtQgc$9X25mN>^T$cl;I^Y<I
zvb+eq>Qe#PhTrXO?L>0FG?9)gqdQJCACYI-9ji|yc7<ctvLgJ!J4hYh2CZhH2Zsv+
z59LyL{>2KugN>AdoNQe6H3RQmlK5Q-=rKnm<^?L?&rb>)0+CMs=51?N?%wgtQ-4YU
z2}NFKmwS%8`C#qj<Yd050R!dEPvWG338^|J@Z8q-?p<bDg_c5@xu-@GvWCCAeUu{J
z&%CDLt7pA|2=*84lq--E5EG{=@C3+F$)9~`?^c-w80Eg|su$d0(H<K_6wpz&^Z6~k
z_Kx;zW?W(y=#%y+^Atrcd({}8Og)|u(ca%gp1}P3_K^WN1zH7n%``@iN^nZ>>}*UN
z65+v@j5qAA#uPYq#9Ij&QgIHE{=V&l@K~m=XBtV>EZvp>(M35<I_@ruvIpTx`Mf_X
z%EQB>rbW8m_<~Y=^5tD`wrhX;*=HUU>hxG#GIZ4#k-Eou&2D%7bg=%<C|Bc%)|V?!
zT?Ha&mIL+H;wVx5f1Ldrm1VbyLFi8(H}gu!_P(Zl@o4uaXdNhAZ}gW$4J5RxJVq%?
zRmG9sZ`l3wjxo7O@xy-Q2FgZ4rE_lTjRynOE`82IInqOlfz>iRV1jEew&Ofy5&0wF
zLtQ2;`q8kidC7jXcrv$$c!W-GPepg4tz?|!6XO;EqfLtRa(@`lm`AY17?vWvnw0#O
zU^bP-kh6Fi+mZ)Zi$ejL9dg=~z6I#G)z$7QF{9$RPp_kbUH;hrQ<94O?2Bg$<5J!a
zOqFHF?gg_m#~|M$AkTF+KY!{g+Dw<3n{VOxfj@+wT)vN@GQHMm_<=JOSEqW#bJY{O
z7Tz-)FyT!ko}(n4T;pZ4<OqAo{w*>_-h+SSJEZKo1v;;4+vneJiUK?QVMLmHhm8(m
z2i@i)55rMH_=-HTYrR%mLihhzAK`ZxbtN!YD!j2dr-+|u{2hFk{=_DL{_n0_`8c8x
z15$maFqzt=g11k8_W^#C2c264UbdRpt*qwMg*G&VNA1!FT?h1MQrfdLniVCnuuvH6
zJNG_-COEl2=#N^*aZqBh$7y)Ekrgb}+1Y9RN&Q*Gu8l&$p-M~ND|sVuu|0RwbHeJ0
zgA&u!J(69!#(HNtKT8x(*G7S~n&oWxIRB{|4k}}NanUEB4BYHj%w>|1H|DcT?U(o&
z^#eNmwRd;Ab0lF(hqCzU^A;T+Ay@2CM8LCS&JVy?z#D>NKP8OyfxKnBgelh>*Pcpc
zPgb^H#Gr^c(>nLnw5}k>ho7{=Uw!xe$Pa)L6wg8)2LgwrSt_13%wY3sC$HH34Z06U
z)2T9!E%TJ7-E_ac#^F+!{V!G>0f|>XFH-jmh@5Rr7$c65?{aC|2!RkrvqRTB)P`#+
zk5w?$$ZMBN<RIlx{BEkodm8qIN4A@d3U{f(M*JG<%|p9VwjD?GP2#Ld_n5!ZDlm{3
zQTN-mBM0GgXF#&<9`ZIw7ZpAn_@&UWMZ(Gcfr<R@r4sQJwDtLj){!grkRdS=jp5nd
zpPV?5qQ_18lOp(v(QZ{Y2|k%i6rM>wuK#NZ@Gn#oL0h_}o6*pG=jNT}WTp0|^KvJ5
zcgPHd{d9;dQ7d?Ic%oaAK%SC36L0EWOdW5}2*-(|7ed0+t*=^shY}Z=o7Owk67UN~
zNSEc^F8RJjDm7-;zi_#AGAr=E-S=ZCApQ;=XjPl~rbhFf@%IK5g<pZ#ZgIFS;U3i+
z<K(S02_a8sqPgFhk56S8Hf6aa)tbYh!&ynz6Y8|fSfm~()`%|N#iO)EjeTBqJnwpL
zjJNW4HHwq#vkgqcXD_;IJMMKxp)I;<@99AxI-*nJc;shL;Ryx$oDr0XHM=f;=K!Sa
zGbeb~+BL$WOn9{I*COzzBz6q|5@837&~1A?xK9AmR``>6{m(9=gk<I+I>KEBVF>6)
zhjI(;4Z2rqTH8-iD!Er_0)Ko5CB7*Gbf)AuJ{D00Dx62A{73zmhv|1i-EArL6FC@~
zBN<K=6wL9XU~0Jc%scwXT^6GlKy_QP!Kfi#55`@4#@D|7ZQ#$TfXBLbesgGu6&PnD
z4U<;OLWxAAu@QctPMF6Y9nJ$fFb<qFA~}c}Yn0`cyw%2Gw1ii78ut4+7*L@%a?WOS
zy=+cpi;|W<SFoo;Mo`ryBrw77=B^ZpHh9LGRxs@@fl(P`CeGA_75yO!i)>K#Og_`D
zWVFXB1`VJ~nf0M|jKpt@-$s}lcuQq)YwXzw%zGq=+osYOulg0{r7VOT4wUme_UrYe
zAfJH234jeyT%xJmEIuHW#qj&?r4P1}8Q3{GidqgOKbF$DVIZNtXT(N<3<vjQB|4&T
zk|PqeSeL8`ijG$QE`|4~egsWqPUZM$3}`=E|FyEa7AQQ(9)>+8%NQG8H5uf~ved{p
z|J&NcLI*>&1IDXfKHy#A)Xw2{J1YKc*N7l?U9l4!m;P%5X_G)sNwZetVe1pEzn>nT
zB2R4CdeMb3S8u2)xEOy@oMVr}s9eB+@;fLb3Wu7S`h0!JLKpMs08WqB(5|h=h%<_r
z;Afu6J>o&mDdXay56rj>zvK-09jFpT&kGu2IWsdewds=Ao&QPBp50?HI55%}t|?pT
zl9sulVwc;yXl3lAK)$EveP&SIR2Lz~fI_`etS@;T8fIhVpQDGV>F@n2rVWT;DrLvZ
zv~rqU@wwZFb%lN%;w}_a59j-_zJiHeycp!m(phpX)0Zim?)|Q?L+?RE!qQ(%jG?Wp
zm2=Y^&PT$tf2F(*XMjNzpv5Csi&=czQPtD)EmvZ7jn1Oh?iatHgOuQ$@-vbVUC9wm
zp-daAT<YKFdi_p<cU1C9g_nuz%DZlu551mHu9Wq>X&U=GKio+(Oa#BKzxU>ALGH7b
z=rrI?Iw%;^uErdI9THCRx>-;gKR}ovAocs>j+8UCR)w1RB`F1+7S6o&-dQL2WCr;W
z(2{ZeeiD`0K@a(FS7;mdBo*xbzEuDpM^jfCW2C*fmGnN+9EbACS8whS!5qnTnS;fo
zL-v<mQ7-rbTt-*>E?Ly}`}7W*+-$kJ8>$kEZ-=CUwe$>VIo4}Wu9EL@<siI)t?zud
zQCj{`y@f`OadWb%6U$wnJ@NIk0OVSs)E9(%FrPge#7Y}4BSCwrpR~(k?<bHIJR;gh
z5k~1Tmp`r^Ds&|XC-3l{OdLQ(0u$B|M(j?Dq9bUEQqpMe<afdmPNIv=+F!LDrO5lD
zWzEKx%c1?=Qc2e7zkR!qI|6o5E@wal$32afGbUEm-1E|q1V1FREY?p@Rx_7T0X2nV
zQwC|jWCp4G1DJ6S^DmKj<kB$(uWrR%|H&Ra9{0fP<H!9{13G@8LX;JpIXsgEm~w)`
z%h+Fk_Xi3{5%wB^2itN-b#!!O_PC1WN*beouYw+bfVQjb7|M6-_Exbu`rXa^7YmRG
z3$Ts3>y_8=$VJegfScwwVU42}DY{>S5l>;I{wqZ#n^BL)JKZ<cNeQ<*j2K2D+`tR^
z?F|@YUO`pF#{qRWZrQT#H)7~C{GZ|L*R7*-4Q9P`CObGZ_ockZStyXI504~19)eU-
z*s#B``XPhzyTLw5AQle-Jz8Oja22(icFh<os>oaQ%S13p0wW&l6L?+%IP)mGocwzK
zvMu{Sx3-@<rBaCJed&Zh!$u7PYc(1>?orA8A|YPc)%(3oKA%|pNl;sKa?P6d1^HTR
zEC+rgVFN>7gw_X_>P78u-RNt$1iM%Ox6W!KV;Pmb&7zh;fyQg=zXrr(;1(AW^Gu8g
zQWW-uB`9)&w&$-dx@&2Cp<!W{{grg19svcGthg#q?+W9-VcX1^8e}TH+R=D?pTNcu
zwq*ryRTx;$Pn(qvJ&wzbr~Re8bCB|GJu`ZtiyEkVVW;|HzLVj(tmTd5eM$^M`_Kuy
zN-iOUMteJ`L`zBCbNXc;7<2#~=jva0;8R_Gb*rkUJ(J{6>D|Z|Gs!#E`fYyu!#Ioe
zGE**u6uLm64VhYi{4Z)If&4IdRqVih20ZEH^xjmLAg;iP*AhR35AV?ksP^eTtpwev
z7b<I=DruVS5k7U<zp6;$DZrd$M+p>@o%#CptDj?<=^8gDzu>(g=6ywBq!m0Te{CVE
zQnSGPTjrBM1@HYnzbFf;^AW>Ip-$YA-~|&2RgSS=@`50?M?IgqvXn~V+G++hG6%Fh
z_IXykBcL$aq+6bM3UwT$sjZRb=B?sYWe84^eL5kYj7*oB9RA)lRnArrpdPS4@z^c2
zQNwRDTWZc{6Fr`y3AkQ<_KN?$7kH-Lucjp=ui;9^OVfOBIhyj0<C^6@8qG0-Al>%`
z`(tDioC!qPBv+63a2jpB(WfWc<M{y?KG}~8jkQb>xe4zF(jM8q-d|4p1WI1qrq?Ku
zhz?j_L_#sYU%KB#mJSJtz32zHq?4fB7sfHOS%T9&nOl7F3$ETi6BTnDsJ7Runp1c|
zs@t9|>f7Y`=HSZ&o%{6L4~}}PNi!VY`aW*{m_*bgd7FJrvlu_{3WB8S4m=Slv~hSs
zt<_4+-Ez94`_Ax=#Nl&1&Q-Pc2WjKxo35tvWWU@yxC;WPg%|oX@j=vWf~Jctt{-YS
zL&?dlbr$zI8AJk%H!!^-Y&g|Kt0X{Hxcn#i(_guyoE0$Aj6yINLe2#>Hk?Eoc$i0$
zT4}wzeP5#obs4WM%b*~$RLFS+Q+9Y#ruR!Gk%%gAj93>wJ>i)9ux8NW3lh-eY|Vq2
zf7)OBD3e6TYrq=U(^~5E-x0{D?qeq%sGqX-9~KdS6R^lRw|rTem^{kQAoR*_E#uNY
zTHK_xG&0e(3toCcSkM_NL^bqQZ=VfAq(P?E$!hRALKb%P=+UrLdMAU3@f1h(fc?3b
z=R>s5v`4_<`NCuFTaxdyAKk;daed!j`dPtV($0vP!{^w6g79?Lqy;NRjl`=R+h=Q-
z@dfC!*s?a-aa?#5gM1#d|I%>@NXLJYCD6hi62IsYu$Aw-ni%JL_}+=oU$z86Y>Rq3
z7<sDW!Fwq_i{QbVMABFGZ6N@(Tv?6INYcVRSS+u73M$Uh*JoqAK|L(r^~=ZKM;hU1
z$~wldJ&<mv&}!{oUu*EVh3Q)E?6ZLbyrp|Y)*0TT+~?`xF?WvMDa~(O+H+rY;u(f4
zm7=-&W<^ViZ@RBh?i|11Zb=u!2MkaT<hF0fZM-8oI~8@0QquhmAUR!7$jq-lN}*Yt
zmO5aM)ARhd;8qusZ(RJ(LwIpFFs*w6mn6-BhL*_PqAv&o{HZD=3d)6RhYDZOxhune
ziIsRhFm*vf2b#omRp3nAE;iCDp8jT5H`;^8^Y^(Q6pT=SYl$thiapPMHZv8c$NRS}
zAc(*g=I&RJ;>C%^B;;>uOs&3(lUC<z-0e|;;PpI+e<O#(H8$6C^iz~mtFANscXc0;
zkh+QB9d?psICzK|9Z+3MY^it<!BBkozkNLOLd+Zvr7Tz}vQ4d~<@)Gm`di}z2LKS1
z4k(6*AVy(lE=sraYbdvHj?xD71iSr`oOse1PU}LK8k6=D%E0hzB$H2g?~m73KsXaj
zE2exE;R+SFo9`@bMdZ7ykZ!R{^pb&3$?j}9h8Wh_wc%8)H*CkHz9@J4lJ1S&KS3^v
z`T2p;r<&CEcA|)Lu;%BswEpc2MIX7&iAnC28e_;|r6D$>5&Ffg7$}Pr$$i0n#u*ut
z0q8+6UOk2vcWtMyo+s1Lq)49dE4@R{;AmP@y1bK~FVG0>y|9j>w%ZiOY%`V}2xnEI
z@Fn<d;D!VyO|46j4HYj$&v5SPJMtTR{j}OyAqU~sx1&GWl(Ws+e?zEA0!2-ZTQ~JV
z&h2(bEdp%x<~sHP*>pW&f&tx#`r2{rqF5786U!5OuLqPZQw8$<jL7!~o`{leRS*kB
z`jpz@9SNPlG9klQC%M5rBMLg)o^g(@?UO3$7F#POk9&Q#flWIUM<m16uisOY_0Ys)
zK*kt2ZAhH@p)N5kgQVGqh?@vUM#rIKYwE|%$Jd9_x<4ub&LfN9Q){!s=O}`e+oP5A
zb%!Xzo~aC9Q^I|v5?(}iMA0a%ipE_@?t7Z#h2GiHeCKwVPW71t=jDR1;+z4YTT}15
z&^p!U)0Irq9VTZiCTrPb?Y%01m;$}JNd|)7_hhGufE#?#q_~w!A7>)X$9t>!&BHyR
z0@5t4(PieN@0|X_ymV6E5*!HihCsmPTj2CDrFV%>z8|k#I@0w;yrc2ZiTN+nqNix8
zdh|1C+-1|wou6bms(E3i^yvCh#kvEyAc*0_GJMRZwBQLpt}5rzb?arN{4E)SQX;`+
zt>b#`?h}rOuh545gkz#>anNf_YNFwNOv3yesWhLR#qm?^{G@U?97G4Z9FjYLa+Bbx
zc6D3oAF`L6gH3!000%>J5_Z{tZ(Q@itNvX0an|5cUH9QlKJB?#&gVQvzFOMs`Ba-9
z;(GHZ94n)P*BUc*O^7`-K4xZeh(#A}y{=4^*@hp=gS3`4{)#ib-+5GJt?d(Bb14Rz
zz|66xy;3bH6BN|0yR9gVck<~-s`7#u3Hb=8rM|~WJZ$K5*x>=;cGWOiJ@Y$5ZZXn_
z#V$K_#TGM<vIIt1w;Ej7xOCB(X`qC$X2FbiHR8$o74ydx7-g1o0{%8$=nyIR;5#NK
ziP~6Pj>>lzlV?&IV%gL8G7Vi|j-|Rku9#@3tUmx=d+n@*FR}^e(#x%%S3~fb)@S)g
zeF(GyYiu6HEr|-+d(AC~chHSW1PUH^%3}${zjVrWs_i+;`50D+o@fM!e~*$rt0qAr
zgkao_<3fF?4#wy8^<Hm(rO9%~5B8cG5ISk8%o)A;h?m6*Gooy>+$1b~qF+q*@GDbA
z_MHO@U%oGiNE9ZvGxl_ET(vhNc%Xad$7YOP1XJC~HO(S^e@w>#K?@n(R~6KFh_ehR
z2tOjIFBtmbXZvXm^bL%3V*c}(6rAyKfW8)7p0$tTlxf#|?h|!R{;Qd<*M*+Q9@^rS
zBP|sQULWaatmLUOeVu=a_2+Z?>NPjlA1XwXH{5KgF&=eY?Y{f>d3L~^N}S~Iv9TL=
zsd0i*r#ueqE`&_RS-z;4>>E9YH+WNbz&EO$O=XmO_-b&aZ{S<Wh6(}m5AsimXT<bB
zTzAa-Tr(YJ*@NnS`bHEVGMlJeP=+dh2&}n3@f+TGvNh&Tp-%loF*@SYJK7Jy*7985
zvE><K(w5{KN^s(e!28u^11lQ7+2|DEEaSjd@k=wVvg;)ZQO^{zrAtlw$P^32hR0#^
z&YM$3v%W8C6!ehbF(l$0Y`qNePE6_?yq`txLh&X8(Z%TL5U-IH|6OeBZH7LlkF}IP
ztJC`nU){Db7ZF9kU?yCSF1}T!qQPM*Y}v+PxK}nVGQqxi`bncy7&SQNbqYGWeQ7Mw
ztBo>Qn&(p*AaI0u4p2e8Xs$_O&v;Cwlx=(iRb9%&LRL-sXfMo-W%+8D5HS?q+uY)4
zc#%E;JQ>TB@qIm!C{tQ)d~Qsr*lGIB8+``ZWOz=sZElz{TT?e#$+@X1#rKI2!jAY3
z5chLe1Y+s^Y@mAjnSxL)j+L@F`(MeKoGwc2Mh-fwSj8SHV4;?7I|4LiKH62NN5G)W
zbIw=2?~KzM-zpo+8a5K4c33ClhWu164yxC#$?@<eQ@bsuU|FUDY`IIl&mdl$a0~z?
zJO139-&~fNa}UgR=96qB-3{I<ypidZ7B=k<rgG|ZhV?}$uO2{&5OncOmyiruloWpM
zmJJ#ekqC6q5T*0w<A~3ZEa>jC7$2|iahu2`a^v7AABE5>BA~vg*d*%-dSe;FF$pV!
zvuQOh`7#fF;c5G5a889}$HdZQSvG%+PrnT%^+l0rkNMp7NZ8h!8eF&MN^C|9@^}o2
zx|2S#vXvG>W8;VI7_;f2!Wv2y3hc|zk$`<~vdeV~q+Xp06?Y<H-IeJBuq|2iptn`S
z(0E5k`ez()yya=|;i<Xxo~!P?U)VFq7NXO?cM>te<<ImviJ~?%(>kj%+bP60=dW_b
z)!qLHUWCL4Mj@CuI+X;edl&Zp=+W0s7dqV;vi=Kr@GAAA*6vbBQuz8Sl*-(KVXjVL
z$;L=g05iUr<YD+6^UPOKjWZ6%-5T_1Loc58mC7QANsu_)?OSkr)joHp?nEIgOZ*Qg
z4iZrU<ihwzM1}$?dDM0~7o9dtrOY4S58H`N#6!>#RvdOXge{;L%c<6U7kH^p6b^kK
z4bR}-T4o8OW+=AE8^0$6I#bIJBlr%rHt^`>w$_|16cp`Ff}94qC|Kj>i*~>Q0wz$i
zRTQ%8RT*4#m3U2<Lqxcjk;PM~mJSrY$_(~aexF-<bM-O<<nr)L^JYeuA`Gdott3kG
zF|g)WppegSa6NA3Mwle6y0kx<e&){jSqlIyz+tX~h4g$LAb@c@i6XlAay?2~ZCF0=
z3hztXyghu8=L_`sd6VFF{ci7&qeaS|w1AxGW2nQN9die(R}yO`6asn(K~I8Q8e!PL
zHE29OK3%%fW<guSW<_|t7O=<92563i(5=<ez(qlcnKcpRi;I`^o0E2(1U{){t9l@Z
zTv}{uVR57iF5?f=qz7Qm{*va5X)ybG&|q|c>AGc0zcv7p1t8N4m8*&-IdCG2S}epp
zx<I*r1wwuFe@f(oNl?5IOv%YixA3(nTwB`LDfAOZnN7L5em1QO&#Ald)Kr`Wyek_m
zOeLU2rapYJQkKGf#{8>W*L~zcoSuND?^E7PZm#QduqjODhrA8-Fou~~fREYI(w!Ss
zIEMEzlaL|OV=ogYzJ*#iU#sM<&lc_BHN!%36jU1pE<2BJ4u4P7w+o3|qPS<0bL{}?
z75=)+oC!scmgCR-H6gxi#ykjuEnWg_*kJUzqa@<R$mY>TpRBJV5`5=>HJy=)a3=<i
zdql%&T1BQ>ir4%>7f|v{Mv}gQ<JOOF(Pu}q&POt{Fl$lUT|WSB_I%5|lv0&d$Z=&|
z!`b4*XTB~-uo$3@V!uz&H4y#hU;=OYCD(5?ZirC}xxENcl3^yVayVE8u960U$4T{d
zHGmQGLqE*Qb(3otxp76S;RT*3JwTKnk&XG%ZIOvi=UXj0G<~v19@E%mfTmZ@*ZETd
zgW?Sfq`tzJ5NkTPq{XKp-_F_dS$|g<0ekmVn&-Z=e=Y__)*W?pMChIGd5Y+BD{7oH
zBwN=n0;#qqw=@f}u~p-=AM)RF`AT>fs&yGbU6q!=L>?Fm)eDzZ7BTp(<qHzDGQ2no
zJ0U)RFxMBh2e;5__jH%6=YKT&1e$Zihf!<8-8{F;PE1EJHL3+{(Yqr#k0oBe4xi)n
z3tYeCXxUud8Kt`TDuhUXzT^-g<e5o;d{KP=bQ-`B$+fq`;$`+qAJ%pA3Gi3UiHwLX
zN7K+7nQ7(K7Sh5xe02BG{Dh`27i!|pW@Oda7DzjPhYaVsM4xDA8qvqDb^PATKS&fI
z3waT$2!3K|1Z4_4Cf(**Tf>4*B5CTvEbo=IJ!<I?ESs3y-C)Y7Zqe+WWd<ztGe;$=
zM<sV%;%oeoV9h`iVScoO{yw@6XM)PWZS-dP^T`ZKLVGts6)1+DB?l*{x2ER?CqQCq
zJW)M@CB}2p;vHEBG%|Lx9R?Znsf$Eh^Ia9=g`J(bXVOkRCPwPY!ItWstXw`<m9RIC
zFch1}@mQY&r$(6FTj?Sp7^da5o~Xuv6C^;S9-K`U4@y@@G%)$p<ERy?P<J)?l?*Dq
zUB@LJj!c*4f%9tRfQEHPT<~k?=Q8M_obL}f6lzU9;!x8b{Y&b&VinCmP(xwG0hCWD
zHEn7f4-4Qypj$ukgKhyqL4k1qZaApbcj$TIGY-O!wAIaJT&(kT)ep95R(7*bBD6Nk
zbeaNGyk-+a1{@yfT}SxBA`C2CRZ5@eoXz5r;5&;k^<b4PxOB?UW+VQEFC08!!cbmO
zrUvBrrL6A+9zu?M+u<9!{Y(^^R1Fe@Ofx2CqGRY&XUeKB?Bp|lF2@=RJke8Nw6bk8
zKY}yNH8Lm#G2?d9-A9hWCrBi+s<hoV3g|U_CZx0=oD`A{GAjMYS%70HCw^o8cFnWK
zjm}8^+~OUg9bdwng!T?dt8ubyFZK>K<1iuxFg8>Gp+NJT;*iOm^R>f=9x@k~4x;b?
zi<2lUM;$A<Mp?!S-;Sj`DTz|Y^8$;o1J_tS*M<iQW%1;1jNYy(-d%lF*`rc*a$>i3
z=17I7UvT3U*JT7x=Ii2O!O$fdRWMt1n=xbdqhZg90@wDqdI@T>l$l4OV3>!nh-F2;
zYN4JFlMRgOkAga5C8!DnKeLVl7(~z<%gM=kYhU^;I&j=MV3BQZ(R8;G=}3`b$XF_=
z7TpW!V-aWVZJ>|*2Q6lG2Y#g58t{FICtJDF-bQm8MIs)7whJ3xU|S!3H$t=Q(oxk2
z98y0c$hEacMAd;s0^mZ>6c0Ar3fJoe$hNY6lHH!qW&yHpkZth8dfKPnQh5%J)2bL{
zSu@yJv=CF>qS9Mrlo)78z3L3!@(SszH^0iUM24$yPC$^3-j5XcDi+?m#(-(Jw!7vn
z52wTJiWEvRGYa4F0r7)0{)lj&k#*^L1!cFP^t3IaO>)r6%%?!b;R)D$XyoQ?KXSpu
zPcbS;`zKKECW4>hS4w-=zZ^-BUppdq5>){;A}78P9{~qqKAV)ZX(5(Zaw{Q=agyZA
zF}>}86%ht=%Zkv+$+0c34{Y~U1|Ir2Dy>~Z%x7-q+1H7afj*~diZvnsStg2(m&&5t
z<bRly_wqO|$uCcWjDf?x+=#fu*Ci+*mO0Nbsqg^LbQiJQ;LTZEuWudN)9F@Ch*4f5
z&7y^`2%Z_`Lt97EyM`UX)`ReVR@mfcx(aRcnlUm7JWc}aF2%QprvXxrL3qcMeA7AG
z$TL{<<jQ}El7x`#f|B}M`}TV6dgW|EFh)T2ut-GY018IXW%<zXn7g@2@tRMoiMt4C
zDV03ZA_Tas3IbD}M$P)gjYl68_3Ax!@ffu4M?YfR^kf0h)tkg`Qd2p`JMEjoIM+hr
zKE?VyJ+jRR2I&}`hUarL4N!fawB<P;AQXJthdh|$s`x1z@cGy+d`u%OfO#fl(0cU~
ziICvpiM8UOvIjQ69shusG1cw|m_f=a*QmP0lG|4pAS-=W)Ofq0A}5VPsm#{a8(3RA
za_F3EZzlxL2wWZy8vv$CIyy-4RCgQ$RJ@gBo$;tKWYIV`{==eVq&9%*mNoF}0Jxsi
zbdnqN*$~DMc76*nIvz0QU?$!q^F#o1hHq3ew&ts9TW>}zRE69((21{?fk9x|PFS74
zI)C=L8g%h=Ik_SPkHck~;SW{Uc66Pmpfz5Rv&-PI76dYR_BasRhu`{vVjyD<z?u!E
z`OB=`qVnqi>=9;)DK`L7VGss5n;NRar)r=H#uB}tSfH;*#B@K2*3%f|Pw>8KH<a$U
zbJ^q`7kH(g73z!T;^fl{UwU3z9UcRg<RGvl6vUzNz`LS8;GOIhuuXs28OpHc1&YGf
zTswFI0Osm_u{M-1Z4R(JY&x<SFc(?|Zl3qy!V@#s#eqyq4?S93KyHvDro+a`+gg3D
z*iRG`rBwlAaQ{HMXa9dtUm^C)*-`Jn>xxAsI<F&Y2b~jQ?XNIkf)r*@eyZO3q!$Ve
z1!mqY%p?Jktg`N>71@b_FfXpKb?<;hUsE~EbOS!IWov`9Oz-8f=r3m5RV<_keEE^0
z$gu~|)E~MNzv*JLyXstezA5#!ZIOz*hRzn9vR>0ud22E5!a@<r0)&2=IpbbSNGZ{w
z$U6;Z7@pt9jLpT|g{j%sro6!`SrRI7LDPEWu$?%4A9)+liE;5R3usYGkYXU<-I*g*
z6IlUvvhqATsFt$`-8s>;zEueDK>`($KsDs!6v65h4kHg~_a_)6B61Y2>C_z;Ju&NK
zzIc`6@InwLK0n|?7i7^l;FZ3>=97ZRCrFT8WzjLPgs$;w6058lM*ZUO^WqBi3({gs
zttyij7kL7d{jLBpaX=lV610CYVx@C#URLUo(+ry>2D1-%8x_2+!&Gz9KD83pjq?lS
zfPmizHShBBazGbAHFfm|V`5YMq6Q<9t81QdbhtoBfGsSfo5{=9=}fQThQKGbD&~FE
zP)d&0;;;|isQt0zcFo{|X7UX?jSfhko)YKP>qSSLwVn`udRG`96mcP>WbDl>z(j=H
zaR9S2`ch$&HuyZCiZF{X1ydG5oMlIacDqXG7Nn(^IWA==_k*sn_%n1u^p%AuQ>NvB
z2-k0Mu4D?&=%9`A!86V6K4_%5r-bp^IrFhX39|>7O*YPZ&;=-lCnBP2)tGI6m_Udv
zmurr%)bk|4IgBq?ezEouiryZ<fNU;jp$}1+rykWD5#PKgp<jWq`ekb9=zH}pcKxkH
zd2DqY6(ENy2bFB4lZJK>3%yt^H3yo=tO%|cCg4gGQrTKx!ZdZn^Y07k5u}X<>tvOr
z&c_Gc0e4TrHyLIsWVKIrUx*#KfLb>V_1SJjBh+A1CCuaWb&!pkoiEj+;V>hvfLf(R
zR+-SmGBL+a6{_>Z(4yl7B~dmSBh2;#==vI@P)#9OGvT(WY%OQvsRRKFm!Z&Y*b{B)
z5q5>mDEE7M1p$U@B$N)#U0aSHTfcC5T3(L@Qu`}~1_RjyiZ`NHg~;L2JfqSQYvk4z
z+if^7L``5HQEGQjOzidvd7#6%9G1OBYuAH{-_<7>97-Q*#DA!L{{&_~{hTvw7lcec
zIu1lXb0z_pJER+s+o9ILX2?CL;z4@Xpk-L8Nac-wgz1%dKD#fG!uTGCGAGqsYgSV4
z0W{Pr$!WPIn=_%JoWPYoN*K_iFo%R=`i{);XuAt0j(S^%I4@6LtPVbp=M2`c0U?#>
z2)p?I*!t?QD7Wry89@h7Iuwr5oq~YU(%p#CA|a)Klz_lkAW9A0iZn<!0wV?uN=U;<
zibzN|-+pFL;r)I8oXhJl^URL5@3q#w?zJC?3#2f4HQlE9>y5*LS`6_!nW?7K)8?2)
zM?@uj*N|q@)W0niQtr~ke3=B!X&j`^=(x}(5P<43TLqPH)nGSW4*mK|lG(ETA1%Nm
zO}hJ|IK-cs`7V)4gEAo~AOyBT<v}3m53x^lcj(wHVHHvy<kA>u#&Ez|a3Hj$D4(X_
zHthRWR74_H$O_!Hf%!a3=;Ec~`d#juPUVvUjTQLKn7k#3J}hFcL8N`Z%zC<iUcNt+
z2OwoC_!&ko{*Mp>s&HVcdEA?MlXP5GMC+J{fxQ|<XKaYPRO75}-)JH~mUI?)oMjlT
z%{>Xu$U2ARZkG9rTPR%YLWlWFF%85icK)}K3xCQ8QaK835bTl`4M22SU;>D&M)oBe
zByZ<y9{k>sA$notyNo%BzmO;{<d{RTyoqCy$3oWyV~~4illAu$XN!wjud+>vDt0;l
zg&-fo%2q19bpziCz~|TUjJ+rU@_#u<h*#Pf;0mo5=p>}gyukFJq>Y9}ieBJ79K+!9
zQy`>bUGHnSf2ef$lmyxWe7>$~QB<8qJRuxDh~U|=YPl>sZ5HN`@}w^v(@-Ey3iP+8
zT{p>XLTYLQhl+u!ClacMJSX;_hkVr)0?;*f@VG{^;=%`xx_g`<HcAkHq@zL@B3tkN
zygOGZ6a|X}F^CD7=bdkYY}0MlvDv0N)-iX-86hwQ_`!&_o7*TZxVRI%XG+sDKMWFt
z-;lVnV1<zfI+Pq30(^caSd?SJ&V7IbsUXiue+%>7<Z#!LxRK$tx{ZSCs;Y2&@bM-0
zLz;6x=zybpfc3&7GVp%^I?7Xc>fb$B`B4yBn;qzlnTRx)2>zD=i1kxS#e}?kbJ@g`
znKcfEWMUY3PT+GCYfLof>cNc~%dNdj&1z}@Rl`xA2aG~pi0e5^td(n`Fj)vV67=+f
z?X-J_3&W2MiGCsgg@A+_dgEqPxWGhTrNP9W>Bfzc^L(MGO5uFI_#gZ1^TMkjZZJia
zeTKPl##9yAH+Qu^UsSyY$dhPPMzs3W-W$av>x}bAePa7dcMoIN0@!}mqhej)*AEm|
zs<~ePbUD0c0S<9!%tq!eD?Q4H(<f8nX*MrDx{Vx(h7@$t+PJ1}TW<w;_#(!NtKB|v
z$qz)S1Ka-wBQ{wYzER=ytW}E7cJW3GgE+KYp#7q}^;fQ9ipY8pb4(g7%ogOhQJ-})
z1;JmnWU&29`wn9q0=XiyACR1?=K=sf2+RRroUv~i!O(*<(9#Qaom_d!#sy490m?vA
z-?H7zy3ed0MG?u0UvubwuHbm^T+qJxU^;x}*UD$~!lWHIB{=|-vy#tLL1;#S6~S1P
zYOHFpwXQmw&c{5G&13|@)phfz8qt<VP6MD5S(9kpIv?b3s|4as-*vrDnXcnCcSgHv
zd9h%nW+eA;{8-QAeUH80I6}L4F%5ha2ZVLvr!lq>PI}I>EamBSo8oN0U0hVrt~aNE
zqAUsILH+BeyHq0Fo?LVP1jSr@gWnv$!F*1F;T)|SmbmWPAukJ%rn5t<Pao!v@U1#n
z`D#Yzr#;B<BlKuI0XSUrlV=>>HiAq)t!^b*vZ>HSeaY#8P$xO^-F<VoKBz3$7`!!%
zdJ3h$=U$>FV#G~<U-#Q`jcX6P{D6PM1v(GFlrNFp7#QmtbH#AY3USW?5*`3vlOhx&
zo|opvwae&BnVeoLzVIRYg9T}P82A=y2(*MOo5c>8R2kpxl~BF=;0!wyhp=P$2lKe|
z*6UZ5{lp@L4)uMASbPWPn;$;L-CleSUbX|at3U=DK9U5!K{=mwn`j_i4_!ulkMxu4
zGIjoS4LbM1Af>)uHDEIBYv5rt!<22Sp$*C$HU6#huL+Zr$PBN1u+(M7docC{4+F;}
zf#uTIGZz1QI9hr0YlyNahMxA339^C01Znzpn0|1kS~*G+X>Ts=Ci9C>8?4nhFaJ`l
zW_{!qpA7K=1p1d2o;HEtBb~bcviDp~an}sXL2;UMx~AMAX{^FqSghv$Ib5)@&Ul#_
zI5>bs-~xVCH3(q=zEC6ZZIVlO7H)kC7<Nmo(m=X;8W%@^!oLMqp~<oka>2E8_^Pmh
zubH>EH*dKk97HDJWS;0$9y+*85|eFpT8q+Cro$fgbZXBItITDB`Yt+Le3E`jU}80?
zAB8ZxRz!)Z3_YnD{!Qgorq-yO6XJOO5%8-J`}mf2lS0qD+zxUu+V~1YU@f1aba^!O
z@6smOx(;0m6cs-I_729z{6XKuHp{{ESHVUz<H8F!fVyp&5g-lUOKuFtg9$Ow(6nq2
zc#WnIzEdcsiLhgyr;XW#{4f%}`+m~i*zm%W^<2{WA@{Ky{oG;}<%SGkx_1yeHWF7Q
zZ(Er~&)x@`gCE6_m=H^vehiio5>yv%^WXm4+KMS}+i$Wb5W9d_YYAn_#rp9Z=*KqS
z=qM)NeU{PZD5R+xrJE3y183|z;jTv4$bzrU=U)c)gQ6m5{gkgmpg(GQgp%@${8-Ta
z;06ZKNzPTHbXmo$8rq78&pyOo0ArTHUm_ooX)eap{G}a$vVuk-CvWC4(jY-^z8KNB
zJAYwF1{eI82bGfq!%B_G`;ldIj6F5qLmB}HdA}UeQVo3ZzZ{e&2x{r(usnYlE6)9g
z?jTI#QPZ`lW+XWccFXw+ivR|hB8^{dzqAdw@hw+YH>%l=IpxN(7GygN@J)N9W0-Ai
z^MW?(eVaZvkpo{ip)=s<p(5s4YR9BJCv?s0-OI|9vW8C*u#<wcA=CRMil0No7E8Oi
zE4p7)fkg-cyOzV==78C?u*;ZkApMZ?=GOrcEVQ1V#&DGbSl+TdgsNO9wxo-ZtNv-h
z3vs8nR(0DnBRAn6_0ZYTnVe>{5w^l~=j49dp<^#H6=BZRILDIe+?uA^X2+(%`1zC`
zII}8nb;TVK!Wc_rp@<c*Qu)N?Z5a|dfJ^9pCX2PnV>haAFKWrSUsMP-WLe#4HIiTG
z{p6ju+Z=)K*SPqJOm^5;ihj@Uf{RhV{v%$y8Mhu}t)hcX<H7AgnNhlXU&sF2nENP_
zP7fmm>L$wHCmFFJ#4eu3m5H8*J1ztDgcKUC1Go_(k*;dH315Ld5;#Quy0%d$=IS<9
zV!8DUnE@sC(r8WC==l53UybJ4TtigZRL1*8#4lim4v#cN^F+k6wf$F-x2mn=FYSf5
z`P|v7GkN-OW>6ct1IMBm_k`rN*K|`SQ!)fDu-N+G-!2jYVuya-S<(op|KcSa@&g{f
z+_SD94PqM{8hSwN)1&c19G+&3-d9Z?6N{KXc$DUcyv#fGUk|o%jvfh{-rj#w5W3T+
z^o)rD#sQIXk7P4|9|GRe2iFc5<yz#%#ka<uk!>LaY52T*Nof)j-H+1XXAEB{7AvZK
zXN?iAI-LL`7Ix#gxt+#Lw`w;GupESkXn>9R&CS-aHy@`hGu~|ngcS4#dVUI=lJyY*
z37d=Cy|Uf`E@B%1#=n3-|LjPJ7#2U-<580twu>F?J_W->_*08zD5B_{_s9#}vm^3^
z4GQoxswk0Ghmx9PxTU5SI&`|Ys&V0_OY7|qhS5+j8G@0}wS837B{IOK()O7p`HOcF
zL)b0#2d^W3S5wa?HU+9pyMu)>766g^tsp#8`}7n$KsjAVb8UI!n|D?f(HXUXXBRIz
zP`xu)J~>6h)qVG?nnn&YLFMkH2npUTau<n<hD$eC&m{jcEOxz5wgW|1T`WtePe#ON
zT|^a^JC|msT$3amG1Ee_gP}GIH;x_zZ%9Ubba$1Z4+1fIswbx<hNX}ORCDiHI`5d^
zrk8~G;~uu})(Xt?4Io&9-E6tXjG`2Om2yvcsaS_9@6*O4yACihaaTM5I$MOnuisFE
z7tlhV74e*2nkoNQLwqK6l~j3GR8*Ty4WraTL=wI`r^H4lbbt6EhsP^Pm^>+(txNuv
zH90U$aB5R|v_`Zl{7CxyjbAw<9P+sS<ElX1>P4~To!Z|{z9T|!Vv6T}Cb6`4(nxlo
zTfkf6s^4JiBOA=62c{bqmmtTM3Zw1tv!KJ9G7Fncao^nugX(^m_(2bGF@grNZtiTn
z*{8rYfRt;Bx#<x2KjKMxTJK%6x=KsS`Ja7$(E^Rl;H6TUO`_cTsM&kW^Q6CNY(q9q
zD<`n{X%HlbHBhU2aYULDq{X;o#3(eG8IDwsM_p$63SF4e&+pxe-vuP$*`#N&6C+Uy
z^-p-{HvZtYUv%>4%kt=h+l!{CiM-a;(6LTt2?P&k183NC3X)P`auO0PExAwRsTQyC
zA8}<hX$#m!P-&+!wx_k15>|6@S8ppKrTb0>H5BJ%u_tp{8s@MUvP3+h4@w-=iEmpO
zEBwhw5zl&mvvzT3-p%<f@8XX4!p;h9!Hea~%4>Kxl=R0EpJNx~qanJCXL0{@x`R!W
zf>|O`qAbg$B1wIpvDqVPs6(JutLw#nF^a~>qV~!+b}K%cIjjBC%EV1;r<)h=mo%!^
zy*Hs%D}S#>JLT1ub~;1N`_}@U&(vwnNQqOYA(X#F%FbQur>N<;?;jV;lNxWK^8$!b
zFgsDW2dj;9@Or;q|CtsX#fvypIuqLC7z+WbmsN>+y)xa--g_^3Dm^?Y<HlZ|c9f~P
zHDBZwH-}K`hzD}OHOfb&#X+U{-Tu9U0sRz#l<{H?m)_RP@-#cGb8D*yDcPnAy3DyT
z_CcTbcXT@?*8C1Oqz)E$#MDxkWRVy5TFr=6mornQbefH6Exp>NLNg=1@+Q+28>5vp
z_dHZ3cvl)fM=P^vIx1~7JwvKL@Ocp|Dd}dp{VXNlxXPAnz>^Y;QyOgxdrG;nVE=~6
zCb~`Xj!zU{%Af5%N6v3lWfgE*Lwu?9WW@>Kg{COY8T3sxHLz|U*gDpG!Ec~{=GNXF
z`@Xgq9@P={OC3j%<<I3IJsRkM?(()P*7Y4G7jsQ;S(!k`6qU~J8eN)0Ac4#pT2XP?
zSB0zjCbf458`bu;m1|C7qA%z#4zlKNf9l?AO)P2jN&^mC9LAm2GSS3wxD%M|0RO=u
z$TPhz-k|FnV%dmyE>!98&m@KZs`W9uJnaLY%WxXOttOzKgMl(Gm+^l(%2A%U1geW;
z+SzQH(XP+b3NChviT~BOi3gxQTwfLu;kmO*DFe3qDwspW%7o+L3SdYP0GHC>CccG{
zK1Ib&;gtxMg!4+=7FcWXmtMsaR`T*@O{lxNt$)WKyo%$7Qg26c^Z(2tf}%2eE@L~%
zro*?Ot;1<HI6^$64-4hs@iS!o4%Ul~cJ`G-AUTeWJwcfLyz2L#uKhp1bQbU1dY<>J
zlF858ch|KeOs{`ne|f0HK}^D5#;)#XIBdsHLRWT7k%JADRMo(u=HCs?DnvQ928_kJ
zByntcGxr*J{@3_d`Orzh9yide?s(Qf2CU8y8%oA*=bh&tS<zuQ11_BQn7o*{^)Zr7
z^3HMzOKob)#Q8skD7d&0%1t3dVvp|&@K>AQJO_^jvHh9kp;6_qJIxipHt0=vaKK-~
zQRd4UzoYQiV1}s7Im#TFk|FEmPTRb4IP*W2gv(U;vrAw1C*9cEL+jgP(V79L$zKAU
z1=dhd*Kvnk?d}_Gx1zz%mxG^o913JR9QdCl;N*6)&#yFkSSvZ55vt^6`?S#+llMYD
zp~Z{i(H(G8CF(`Q{c&!T3};7$>Yob;u!cbIgpgVMqAegM+f{`82@6#f|6?cS_Au`B
zO=Eg~A_y;}$N!3Du|EvnytMmkT+F7uUto>SD!ZFytiXru{vkmDepwJlq$DDKkS0Ix
zChQawNhL4<6knj{pQ&LMK-w7h`b@w~wa>5I*ZbOCtCg#%7@8US<B|c6hg5b|?Qi=)
zF!vHL_utuvw)^KWXT_oh^UHJ<-Qyo58r!INnS46s%zt=9=}DeRH=Dd|pP$Wb2$Utj
zZGcWvhe_C<iY#^!1A|d-tH$rTt%=G`(}g?q^;5k0+uUF1h|s?8?V;=aVWWJHU6USk
z<WcD~SJ}s7V(P#9NJ)L8#Nh|e8*7``*>lQnlypre{;VLq8iK8TH%Z+wSmfFHP(8Sw
zng3J0wf4B<m#lnLLEBJ|UA6xlot@eA%T>7`Q2$6b6FIeKgNkUi?X_)1a;Zlmq=BJe
zAvcFv0O9e+0KmTxIRtqf#GP(@NlYczmBsYju6+3Ocf=jILu6)1OX=e8k~<$+SYT1m
zLAq+Bad^%S7t?}6tBLrbR&l7ZD04I($KQ^uNCR_O6|ae52QU2xp;MP}Q(1r&8E!zO
z)guFV#R{|cT6A>cO8UO6%QCMML2~A=tTdHtDYGZgRIGyOp@%D1*nfy-{KByKM7n!_
z&nd<xiYT3`D6gNX?^O}i?Gj^2w=e=K{69D`g%_0F)Aas+XVRYZl;R~E#1g=yrygwB
zO3&Yc*l{`yExYlR>Wpv*B;D;A6*~LjZx0Q<1}uLlGq~-F0|f6cba|@(k0-=FhaTKv
z&to=bY`~c`pFyQ8NTwAR^mia@cxqDlev?ISrd`_~6mNfs3;h4XqjWpo!n@55TiOaW
ziEGw2N?wlbCrSTqn9?Kv*&+4N%kTQC^CF*AyULlX`9GydIeBbN=@HO4HkKnZfuE8^
znj)oP|Jvw;DEYh7)g5O;doSaDAk9hzM!{(zh`j>p@9{_nl1%l*?TVN@*JsV6yd3(i
z;%)zS{R^AGz^eQedT=DDbd)QtfE48qDuFa074O@0fSEXB66aQb3Mwh-&Y70K`-Q--
zwD-O5y`6)yAU%yJCM*o&6yv}IA%E8C1nd@kxIEQUD<@5mC5hhF>s+=HHtxe5?K23T
z>s<*6RB3)s&cK|hz{#8$N<Su4XVySx<R|<zd)IBcGPzxu*2hYOQSf150~Zjvmsz|Y
z>XzS9$DJTQk8lPiMMF(a{1<jH1Gz=`@}W_k_SX3YQIm|31`Sm}${gAX;w@YpUEFus
zhIT>zG+=%U!-6K>9_eBPTSdl%f%SBqa#R{s9gmqsep2e;{78&&YHsW9krzlzCO*6v
z>?j|^avdCPi-P3we=7j!7=4_@WxZS9wr{ITt~+jxlJs7j{I~x7@&0>#_cr<`eoTZ`
zX#0=^Mu1JcGQw)|XN=H30#Y-VxAtBD2d!eqW@~4_Ak#N1j=$cC*$~CoVM*J#zeHM3
z5*Q4=-*3z!f9(7Du==gt?IaPYo1%TjHZ7~EmFsDL?_cppb>r>Q2@ZTNuy_Qp_=eJ_
zDc0G4`vmNvh@9Q{oiFj2j-h3SE^~XJ$u)w1Yb>6EqT<_1h^M~1e~uM*oE{(pQH9dc
z&OVaBDCt|DYInAOG_ke+!c|@?;#B+-cLLO!C-r2Xa%;ao<Uu!@9OFraAOSG{)RXHl
zH9B-Oas(91pFXgdF%O85;Y-{8+d~_|%DZFc?n}1@o~bX<lUM;e)~=H%{%faTF93mf
zosa9Hw#oom1Rd22naICbLBd`D4q;sRYSENLWH(7xK3Kh<UV_Ijz1U1Z{oGwLj@eN1
zqCL{_1Q$>7Z{@wyA?$OC1vixw%sWQ}v`-ybcyNW@tyu~09A3BZw4xN!Pm8UL-VuP<
zIXs&=oM;xV3nhL-t@TE|*=J;5slj$XXR3w$_cw~eAfJOOCeD28`_pFFEhlZ;QKEC~
z?=wI^d>$5RWV`w*1J8RDaR%ytV4y6cWrC{me~+@92D*G|&h$=tkA<L^QP;~RO8UPp
z(qH-tEWc(hQ}bT({ptMB0P-MaxPy`Z)1LBRPYi4MXSgJ>g(!31JK@SR%EI>>0+ujL
z!68^^k~?jAB$Na%!tfC6;1~vVIfi4x^*>)`WYYR^K~%*$MmhdM+!x%xet`tk@v0<;
zdXo3`EMerAe}nLd?11#dpRZvSvxP&e(sS=(UApV@xKZ7X`&W-#LyNE>>>m;wcs3V)
zjh+B}In_zk@?S3%urBl=1S@9pB8}Lj*f=-m80r744PX`nOxAx+JA9~h=8y~Bgb-kH
zqzj`r_%nnYX00DQbw(MK$o0G~rO*Esm3}GGzsh$jvvEQ4S5rQdBK<XbOJIBqV?+Z-
zY(Sd+4x+a;z`7=rH)1#I)nl@M{dC}1FN}5%k6u}i+TN9+zX87d-t_GIV_&vJ$lUua
z6DLXPqEP(r%rPsX+D5<jd~-5!q3OPC2hH34_T;}s1@n$4pokt@VSVsZ6=`bPO@HR7
z#Vq1&KG5-Dd<0?c>Ke(v5rH|&ke+KJUc9@^rz9#IsW$oVq(Fuy)oj^aposVXKcVKh
z`E`SHrbWJ+!v|aKx-n_gxJRN`U?>jisCQi5-!^TXNZ{%KDes8LTs(SQ;0osaQa(;9
z`}cTjIb*f3Ei>2wpttZ;t}j1Mpt`(Iz>X_J1_s{c7zW9|{|@jGxW;~~n)+MRPp^?x
z2Ibv4jzA*@+&??*kE+Bo_Q<^~6mfI2>+FPYz-iK~!V~pHIsE+pQUwSqN&!^R`PS6<
z>38vdRna&a;p}UotYU<0Xf;0IBDA2;!-f3l`ao+w9PaHjJ)NB($=BB`bhhDd+yNH0
zg&ECJ`*v%1)h}}PO2{>H9k6SxpBh^MQLTp~wKLy+*(%dSNS{x;*)&z}=O-7xB0aBg
zOEKj^<@}q-&d7lfq<f_O={Rd8Nt5Di<mY`FZ+G$fi{YzjlbR~)PmB{R2ia{O4~r<R
z$Et;@c*!j*ruC#)r3dI02ShGQ<>+m^5|&tfW#Te2{rk+sA^IX&L8SCuBtykFU9E=Y
zt=PkVIgJI+68g?w<ki!OE5)a8k04#bNx+&vw`T<6Pd!}}b?1=$8@FJU>}xDSO637M
zgX;S|^{>3*()_|!Q|+udC-S6JtIfI=r?zX=HdGAxR@QWKvSuYMP4@h_C&WVzNv}Hy
zH<P&xDPh6h32U3v7b1^d5~_9sRUgwVTz9_#lucn)V8``Yl|n$*!X5n0)GfZgta_Q~
zL8EDJ)$9ReXLM<18ns-m`apNc!?ND$gDrRY1D@bLE01Q^%<>-^bMv&EJC}*DxbR8=
zmK$@nHY2haCt^Kc8~5{(h)0VOM<i)*IC|ej|Jfy6#0;48Cq+Q8S@56<jW)DRW_yvf
zDt8NzyHUt_t}UV>UehWUTILakSrO!cg34Qc+f{cXuilX;+UCPNInw;a=lMTc00Z;h
z^xO=!fkOM3G!139DT$=+f96X9*D?R_gQ|Aizhm{`wG1XbMqptq<MABNrFZe(*^lZu
z<`0>@88pk*Q()<A>RIjSygP0ngnKb!$1XjOe&bs3=BuD|Bgd(huXbl_4%IdjlsCF{
zhjaDraj}f~y<1%RcdG}eNRuor+7t&XetJX5lgbX9S5SolD0qeOIXGU=aSgM)?N_U?
ze9|0W4K-qA$fCKYlHJ88G+tk5WiWE_hRQew2AP%2-Rd?OVtBt3=Q022%Eky7FivMl
z4IwBvuqyT8ku(pON3Tp$98h?hz(%j$W8-j(Z6k^rdW{STyX;m&^!>XQu5k03Y@9D=
ziP7qp+ZFv*V*}?@uDD=+;-fBDw2gVK`dgD@YNC?KXKnu`B_JWJ#7jQM6a32q^=M!>
zamhL2w|N!#wJ25OeZaan!0I+@#ac5jFlb)j-TICs;vV|`A4Rf%C{Lzq$)ptW-fsAs
z&rZ>j3=(@keGaT*J-@Rjd({n;0_;w^9El*btPv@ft!Rlv;$pUpLuAnpphOf07-Kx4
z_}WTJa+Y6&idRJ2DErnmA!2#|LHx=Cx#((Tx9vKU9uM95NBo$}9_W9<a#ubH$1H0v
zUh_SITrIrl(P-Oy#@g%muaY2UK#uyU5g45vBYLLnvTe^8=Rm@gSBaXN5PriK6rRtU
zw!3=G`?_v#MEa4rzSN%P;~2+)OJC^K{N=^v_EW}Q_9vVJ|2lsFYJso>72)liy`lAY
zY#*I0gEB#w<2lOpmISVWnrYAN{<ZDVv?^RR%4+NG#DGtMhcHvEHnr}%=`j|p4Odfu
z#)|k!aE6!Gt2WMrrTcDgQrqPt6JTUTZ)*d7<XK=k>)j_{ZR9tX14E>L!0J3ld1>DI
z#k7sF42UtS6H5qgF|sV#D33*LmC?kxyydB}kWK3%r-E?&)i2oNt?wmjb;ddm2Z(2+
zk34^z9WBb#oxC4&<@y_4Pk@^ZHRNOrB(^w<SDcBP$Cy?;RJ~NwT1g~DVGFz^_LJM?
zI_o@h3$#Ype`7)gdIp$?+lb!5_)TxWo68YoI&twwf}==Pv~$dSU!r4Wh4E<@B3LI#
zZi70(5mGZmiA^uErjN3%Y$%Jmbe(oN0gIsC3~8%>-9=<pk<2G0yftwJ)89+4hZytm
z-Rsblyji{OSC>A=@Y?#{4ToSW7zit8r4oj^c6WQj_&8VxK$Jbnn2nGw!+SRo8Q1lB
zg2nqlSBg5w4E}@zjode}=6YLm&nlKw8DsAJ+6W`>KAWGvvaNb9l1n_^ID#9}=4z;e
zOA(H{oK$c<9pF}3+^*6h&JlRp%oQO!V_e^*>7C)y;i!WVt5&N_pZq4V^my}I+Iip@
z!_ENKXUSQhUZ(4%<>d^PzrOMWjFQsxN)*k&y@je0x~v}*M4F|*V&u%Erby4F`16T>
zKXEfhKjXSGAMKct_*3aG;3k-Nnz|Xrw$!FYyg$;OVEEQeF`Ss8_}%Rn#&vL-BfCk$
zo-1v}njl$}$L4YlrU~NJPmN#^%Q?{|bR*XiE{Z8Hur>d3<7WT~R*C(T@6)O6f)Gc=
zAjrxukybKopxfGYg-uvy=PzZ;{o#J0FyIjx{{U@5@mZ-eoD{b#m;j~;8z__FwlJYz
z7!m0!(?uq4+j~9Gn4-XCWN=DRGmDC<vrx_7Hen_s$0%mPQ1PSQv?|xDuPyJ9=_TYG
zu@SwSsD=LGiIeK5liktLoW$0~FbJN}$m~E=Ij*sQTbq=;HC_1XASO5mb^eTdt@EwX
zb2Y5%CXgl>E+>sHr~LF-hsA`=b_V{D<!JC6PUW;UKJ9SZBTbb=QVB$X*GvgtlVpW}
zg}zmg8LeeaH*L10*O<r=PF^XS(JCadR`AVj>P@(+ra4$v`ctr6?||v5fj(yS;A=-)
zujxtf5WyS;L-+3PEPT+0tA73g2k~u(NjL?f7=N=NTBOM}UX8~}Eb=i)U<Il_CfiTD
zD?NJ7jCL`n6xzWC(i=&1+oi3Nlp2i4BXKW~KYKjPt&-RqwJc4yDgz6luNr?9j!Taw
zoJpZC>9+Ech44pO53i25gwGHH5gutMF`{?HI(*UGyV6}nuWeZV-Bfb|=1j?7;UQcm
z;=`A}_e1|7ScYB^8%CHu^-@T*pe3M#a&sgT5%7=tD7TW~dUMs&`9b#bn75*ok$|Fy
zy+N;fjp_29{iPJ(YW{)Wk_civ`ocls1|svqmGz|Cn@>?`0HM_5!%mXGK_HV@`bLLo
zrpYxrZKOd>nG@H7p-^F2m0M(~pa>&5m45XM;L_`>tTm=x7lmHFzjM@*H(&U-Qj6Ho
z*5F)w^&h~cme)<^;keA`7n-9t_a_^t)}570U_lBLK;y9lF6CsKJ70NaK_2RXfa}RX
z!ERx;>l;f!E@w%bBT<My>Its!-J-71)8)0h8{}RIF(jVxCkQztiQpgQpb_W(R2XCq
z-V@?sEz3ekQ{a75HOx0YaETBP!fS#qbJH_2FZ9?8w@zPFiPxSVXggZ|K+I4(ZEz(F
z(MtJ~!y-U^MCj(VQN>rEVWYjvWR9zN0<9|XIulo30{OuKd8MXK!PNH3E6jR|SD?%n
zztpt-qTRiJCgtcIwJ<8B0fxByw7wUeWK!j&0S4jnJ`aTA{ThwLboDwdPM*k@K&yTG
zk#r%<G%_>VP;+eNaj+6USW6I&J@s*}OXFfmx1&H59o;&9UK~1ymJDM4^J8bNjZMW%
zg@94`@Pcj*|IP%1%mH0dmc`AKdxYh>AIHh55zK_U*YguWA<76z`UC1%+|`GcIJ&vf
zcj00^+xOc?I1OnE+%`$1t$n39fp+t`j1%=+ge#^>Pf^$L-o07lj^jg?ipgQ9o1OH$
z(vQ`r>(2L{!mu{1T>P<ozTeVT*u04rk3<Om(n5H<#5765VkJxIv^iI!ZvKY2=9HQ=
z<(GIq+Nv+oRwljNUV{5H5tjI|`F&Ytma3=kYFu|7WQ4S7#&<zvpvtnNguH+m`k?`X
zBh<Dwkg>j+nHrxmTn1IMjQlc~mfpd9P3OXC){3Mgc3HE)!ODQ;7qqCRl%*StT<Cs=
zkznn!YBf<e*3~v<haEW%Ku3uuoUX&<-67$zelpxqdMO^H>pABo9+P_B#WCxz4%Lxd
zXFDqij04UrkK*llxg+UlB6Ut#<+c|XsSEDsIsrTN#igmcqAXXB+&>C}3<tcr*yaq)
zP)D&KeZXYueY$h?9Ach&DR-uK)<Cy9EwhdjxHtN+D>z&&wXI&;2RzQb9Kc+QD9z7P
z4&|m}Ua<tW4Lr&PPeI6Vkga5CY7VI?z49F3R(v4MBE1$4&f=Y|S81B9JvR^&R{?4>
zvX~#FgBL86ul`Jt;}pIHG8lvnyzu2X)6v~F#@n5XbC}qf7z{Cu0{fzW-atd%d>A(R
z)vK^gifQ(TbN$E1kVC}T=xkN!tQc-lL~6%I0}EioiPm<mR{wXsB&bO^AR=AA04|(T
z82-`#5G9`Jo<s-%!p#92&F+^{z8V?p0(9-p;LCgnsisx@xrKUh)3Q5&)N!T;xwV<y
za_Hw;w!I@RT&WP%9_O$fh`W)w?G8AIK@I+kiU3f+@KW<!!!2iB5*yEh;hn(gTg$2V
z<gPtt4jY8tNWsXBGEl$-TbOmj{0HSu5W8grYA*MNJNi7uUjtYo^ZMPBAMDTPFrt{P
zQ(?K7s}9Qmn|TIT@^;lep~aWS1Yoebh~C;yKsFs2`|ZuK3f*oksuPgJuU{RZ%3J8s
zNh{P!-*~d~&onxNEE~u)wepGPW)C_o1;2pH5M;>NCj$(W_J4nI3IEe-g_$uY8_Gag
zzeD&gTS-{ZP;eL96|kieH#io<=gz)=3cV*o&IFyZIul1YU))+QR}0?)(&TL@a9%Ol
zc34gL0vyx7r7PF<FF+1z6sAdb$Z%$hXfEc!JsWokEx+kU-8uu}^E85CrrG~m*ZkPB
zPpksb?jYfr=@)c@vu1WFPjHW7jTRwJok|YX&#wv1GMru>3gdA2=6JNM{_;D^06EMp
zU-U$<>6AX%l474Sz5jS4h``-dufB-+B(@o-jBikVtrqP%0^wR3(Bl&0FG;)@yZm|8
zB3oY!XVGD+=U|sGPP2~}Cjw0BJq&0ZjDdAXzCQ)pbr75cU~Kn<W^0f?bFV~e9~~XZ
zs}p7{l<0I^s^?Gj%&BD(e-Lawc-v01=7)5Zb2S1*Dd4xzhISHr7m9w!;Nz0OubZnc
zN(4l4uSJ-jBs<P$aiWWj;!DmJ<o?upCa(exYFhq_!q)_zimg=wJj6nDK%@#TbFbHG
z=`3P7DWz%YtpX%{@_46KWbdnOX9I5~8u;NKQ2;)y17r2Rty}&*Ae=zGKL>&n8llG^
zW_Mh1@9WuCWv&$o+4}*C<IW^YPorsXPzA8Mf>X*55iRla`#l17ef(tM$EZdbQvTUz
zpy>NUYdMUG8=(HJk<+%Iee#9jO!FQg(*3!XLn<&}0ld2G>T1FH)Pg<9xHVhQp*Tf$
zqwH5*jGD6JO_rOtt2!qCJ7O#(fico5qId2J*K(S^y`u|NG^eyCUw*{*w6Ql`F8@fG
z2}n`7n`qy-lv^Vw$6mVN0~Q%ncD}j4O%fHbpVnl#pjAVm_K&;=v-pCiwr#B|yjS<-
zKC)mNoVA>L9UtL@n~$PdduPp<FHn7_kpp|nOc)z|!{gwAGAs1n1BZI3CgZy~*HOVS
z>MK6_m%{>!`2e<%JXlcmFm#NiKr}c-6DkQ2<9_BqU*Mrb&gv-Ze%>Mog+k%i1iP6E
ziW((WTzyYz!i>QQjt2f&tD{Y{K}G&pybU)%<u_T<Y7mLgHQO<uWPq8sy-f7yi}!mz
z+!18#1cUroC(jaF)x#dI?K&K&e7ocWd%FAOG|LWw%27k~N4dizRfMbi_8q-POD-%f
z-Dv~C)TiA?*&<hQ%!Z3oRJIM23sQ!a!8=%Ba9pmX^Q~)UBo3mWze-vi@00;WXgM;!
zz!v`OnCoQ58(w=ly?)uRNeWqE5ch>UjdP+U`x42L{M!}gdgJ_?m$GE>z(%6CaO<+q
zcWV^o7l@yMu8UnV0Nm=g`I+o$eWbhGF*=YW)*ls7K43OjTKT(^+%v|&0wOt{#i*n=
z6K<w{<E>huJQ<NWi3`NpA*e;>gRrSncb{umB7s?@<9^HAT=Zj)`sv4FyP?vKHa=!V
zVDm7nKkL4_?Uh3rP25Mg=g_<J@rF|Cf@C3@YPLZT7-&-SOZ9@_;S6Y+U^YNU?=HQ(
zpkFOtJUz+z<?pT6(B^{_#SiSz!E13FXO$i(f!;gSY$lAeD9*r?UdPFyTP}MQA}In>
z{cm<u?@-BK5{(15S1iMxPJObU^xBxz90wCQVn)&eE8V~a@O^+mA=4?6u*nyOArU(V
z5e}!NF`r8&+pM-0_bqmNXv1_7xR(DA4a=_yC%;7{0-@9qh(&P1^XyxMe_06>G;k~6
zLh1o@D}$jMu8(%BOqExTU@)X=k$!?}{8->2%4N@<DKb6_=uNIcy!D8HiQxOb2gj=^
zAk641id$c;U{EOsNu}kO-}sRC%&uSbuhoyTC!oMxJjQ?SR`0;`^0Y;7@X{M;AH2Ip
zb#+&#ZcB-%2#yaSt&duymMkIxFioUFiJd=AfK&v}312R&&fg>HifV$&L*(eqy;(Jg
z_+w;8`&O@=@wxP<`8fzWhDa*c#>W<sGy^*PvsUK+sC6h-iXwB`$_3R7AL?io97VJ-
zl@4hoGh?&?N=(Buox7KhLw7XdDm1IW8h?{JRdt%5YgzV4l#r+#RRi_7?ah2C=k9CD
zKh^(T0SZ2w^&Axg+-QKkK-}wVAfO=_r-5tp;4H@aMYes8)~?aBX9cX)34|y9W(>Co
zN=p3moQ`dTIih##HJ9-^0+(D1`&zI=_m{&~_&|4W+JyM3O^RZ!2vV5=058F;aFdQU
z&SE0qjG`znG<=8lh77kYzju{7(cC1)QBB4~l##>$QhQVh7+vi1T|KI#0R?rn7GbOI
zT}kG5x!Ed=R66>%td-v4y(?LpKo(@V5lutK_7d6VQ-xoifK(0AX3L=uJBJv$PC={^
z<@zFR`(xzs_g;Wks;Aco3_LT4ZL<eg6X{cf*vrl3)O912n2{9M)>dzFD<m}m)pn0o
zpJX@U;b{MSJvn+|j2frvK}*J%(x8NJIy**D4A};>J>jHIvp=RCwx2cOvfdeh2)|(D
zV;PF2j&Gf+zSe;QO^^`lqa3<ZpmwZ!r=BiCvImmhKlS--gbWFOs7-LhQTt2Bf?VBZ
zuxmG&xos%>4kwik<<!)YA)rrMDUq8DBW3S!t$`QN!Bug&U?Z`2`v92$f>;u$6t>A7
zVMgX82m0X-2L`S7E}qI-_O`@KVvW(IeSO<aSLgs<v|vN8O!wLrY~O2ITj-B^1Hv47
zl4Adz=5K=h>j5~MfaSp<7#^&V)U`=BQ;8=t7iT>7)JM3+)rD6R_AMWA>gtn$RPa<0
z*;0tEbxZPQ#-!pW>^1v{A8?;Gm%SJTHeO{?ptjFZT^zdA@+`azX`1}<^O4(Bft7{f
z8E-$F{`yQJuT_O%hS>;WG^6?=y6xD}gc^hT+2Qw^UGJQ|&5zMZp?a+)_PG4P*PL)q
zt$fM_L~NoQGv-FKL(Bn<{*^nI+UJA&SVtI?_iie&%$DCC?k!oNz5FRZywnd$e4z6Z
z;=jW<cfLuUjbE8GumJUQ^mS<EfYonDCBdWc6_|w*0`af}X}iWRd?@xmK~wOA1wE5e
znpytRN3koh5E%|#g{V~R@7LvWlA6^7u@CRjfs$UXHWfVOZu_{$14lu^AN39Qr=o(H
zaOUmKov+HzAlw<g?_ji@O8ko_*jF)s&%y|40c~F--f&Hx4Z2%9U)miMb<rlthPB;G
ze3M@4dSYRJ-`D}o+JYDDYnoB(tlVa#v&aWL|0cWKbrf~^!kb7xgD%N%o@-F`2T+gE
zb2BMd$z1w+K=U-e)GkLv2?^1RiqbKY?j&1!o#Ei68FBT>Z)&s|(Wiru+XGA41uvJs
zE*8BEnp%Ti>^08egqYfvBJZ>-m7y0$2ys12P+>WUO50xQ3O{y5AfOBUqk!0_c6TTC
zBP2K4UA*FcHf1|bn~7;`Vrcc#f*l7Gk5Tg{-5c3UdEu?T&7ftEC^NCo3~7tk4@<mf
zYdqI#E{Q#E*oAy4iwz<!Ha-?7a|T9DT8~QmwDD{AlzPz|bGx#k$2N4VE|LByHTKg~
ztJXS+pvSw;$G2Y1Mf-o=*4jzcUwW;41uEI2KZ07b-n&JS>+?SEm1}bDPI@WtG#(l>
zXiOf=%k-1)vg@Ws*unE=YiA0P>Au|0{W=OxT@gI$@m?9jc!0e1m2rTxMl86OP^itD
zp;r(W!>(o=gw1L}m;6Ct^3o*`lF+<hjg+P*K{Ryjc<7Bp)JI1WF_U3Cd;dHX;Y`q9
z7U%Hc(u7ZJBO_GaPo<7Y>zBM|nq=KKMn9{52=D;A*ly+@*yI0DmdDLu1W7kvjH?|~
zKRQHe>;u^gP+5-aCh=4v`&I86T@00g#mH2jEq2gHQI{Dr_0@dGoTtFo{&Th8x)0u-
z6MjPm)2BSwMm=WN<)`Py{#VO|c`U+(pt7T>H1i-mQPJi44e+prYT_<&S%9ngVXHXU
zw~ShFX(5La-exXl^b3PG7*kSJvPg;FR@_QLx)&_5ha4#hgO%ZA{gW5;pCHPw0!Jj?
z3<oD!6+&vhtvEOak3O2B8RF^(s!!j2@lcn5Q40sSdFO03LQbd4ibqlP1Q@!~0~0TZ
zdDOqqQurq;C|5iL{-G73PRtkN4AFOk`?bFfiBQZ^Jy8TgOh|W?xJ=eM%vmFNtTfHZ
zgkJsxRIv*!i5=Nxpm?o6M{B-cf{S%e{+P>A>78NqUtY#38unSylJ_Mxz}2L&THm!f
zKaQ<^a?d1#KO2l)#}jxy{1J|m8&_~f)vhuYg>eQj+C-|O#AYf~+^q(PCg~wy0!7DI
z6Z?^ejdEO<ZAr6&rC)peFplFXXm<s9^gr+ikppR#X0J%cRgVe_D~*{RLy*kpq)V0@
zXn;xr9ZL!AEzBlExYg`(l1`%<wnyqIiFm)ew7^=g+jS=G_}F6ceHbIu%_vGVdcr@-
z1chEBfsHyha9L&3sSL6NN<YDXZ&ndD2KjaccY+k;eNGR4pSoYm((jLZl=j2HW#vXy
z%qQ@hp!tf!2zDPKNjfp)Tsoa{3-fC}>}$C}nt=(DoD(%y&cU8TQh*EhigU?yV^fd9
zDWGub9L>ZF5})dcFAC|reOEFC?)`kJclai0J!RCGRTX0Ky6w5!Bydx7$nax95zT>B
zF=f89lH7k@7zRuZTG{6Wl_@cx_$BbQ?*X40z?EC0sunsA4g<o&*JuGwqyIw=t;8^I
z_0RQrpYTS~z;tN_sdqBVp^e4=r5|6ldi*8Fo$<|EfPQN<3RYq>>nYaCPY(RZ0vf3S
z!q4NL5;8pSGIIa)Y*5}@09HoL&dpP2=w_(9!_0i>&JIs5C!hJWw@{hTzQ+{XrPqc~
zcE{45SWmtr&(qkvwpfVe)%YcGcAiabhtQ6dfLe$+NR&TFUD~?!+d&cPlf_OGdLVic
z&)D!wqWZ=-86BAC@2b|O%fCCw1&||*14QBY>ROof7$mF&^r)ZW{NpfCb0Y+T&Wh^Y
zmY<r|M=J)9vi30SP3S=X(cVd>_*b+iyv|;eii*63RRJRAsrVm$tlr@pTiEc#LBI)z
zk#xyLZ+w0Y>Akz$AQTAEq}W+NX6#Vle~+Vg(F^@(WWTbOZ`mgn!j*S1{Ta(0(YeeQ
z=!PCqV2mp38R~oV+<~*usY@pjB)9~%pV9?*;0%Q4kIwf5!j2$zPF8(>wm|PbYY8BJ
zWO7`}6L-(xnDq>@Y!Lcm0|;{$m~V{r%ersX*Cj7k-S?zLEa5?-bc6J-WyqfmnyNC#
zRuPm5`z(uRHzN1+i1MC8C{@UIFN@DEiESe{qA&3?W;#)%OZHF<9d`XxjP6M?V8#a@
zCj;;(p0gjBEwBY4JmGIr+x;&vI?j06IQ{crW1^g=dNT0=`?EK>U1eNyILuc*F!$F}
z+mbiyR<7V93ULWq14j8OqHd)hyS5x`j3bVp2kX#glAWfN=@t}KjYX2DFrbroM5s$5
z_5Nr9RJ|nlv15reK|M@fJG1LO&^aZI3nO8ld?AbM!jb!*H{NK8;cT=D&Ii+Fo19>c
zQ3XD;iutNdF0Ag(K8uSO;h~U*lX<ByZ-IVpf6#<dH03sF)^l)7R-K!#_gG$$Jxbb~
zK)k};B?g?R6fWg+esUnM5MmHEr3v--DD_dw;0v9(nS8iCWU#Rc;{)vo&lCev_#c&W
zQ>h!$^gliRhJVHb{i}@ZT<>sO!|Hq)P3Q13h}<|snKFS>qAtCht&z*Dj)SEF5kZGW
z{%M9E@Lq6=Xlg#By)&i4rDH4*N#GOzn!$vj?k%M!ER**EmX3eWCWUyk=V>1}U9JRR
zvA0kN$2XZG<;6MhoYAt}pBW*0zp`czDT9B<?-e(A1?87Wbvl#i4_DHxfo0n%(2czx
z9f*&4$5VIK7W0k}k}MRR{^IHWnN!bBwkvYumU9$VS8TF~Bo()xe(>*E(7(XVjnfx?
ziY%6PUfprfxIXV$X2n2Ru#@+I)~rWVbHSib9FBuI4oCp1Z`aQJ7tyv{8d427nr(yF
zC|@QH$~E4}<1V;`2i^6xePH$Mlw{0G4s>3;z7?fRVtKS7=6SN6VhW6i1%A`KXllDg
zmPkn$w;b^N#D0p)9<Ptf1_4ivBRG{X*frlT2Uu!7*G1mS$<cx3v&xHSdCwuRLt}TU
zoZ9wF{5b9!k^xmeiMVds{jzadKBBOLpbK;X<PQnX&sL%TJSuB)^s7t0TI*v1`EwZD
zVkR86uG&9|JR`kv2cKeDwnEAMPN_}=rnLWe;j;3j_nLmRPP_hk;OCYp@YWdYLJLbA
z`u^^_Z`Q%v{S`ka4rQqf_<4oKenv`&rbi%l2T4ao{3_j#<JJau(0#=cp-$|u!f_D}
zRldXe$lhv!DatA<%5e$VV(gua0qvUy@_sQ=6OQvY5i)g~y2=4^|ND+t1Sfv%Y*$A4
zYGcmjkV{rjY+?N+?ZQkFPYQB%Hf=2fBc}xS+6?z)U$)AJbPIB1<bt<gPe)(Bv2?%P
zV5j@{3QP>-$G-AE5RPyhzW0c4#V{DaK<_3jReY4X;=df^-RCZ+vHDbIY1FR3+{G(A
zIO^HwuHn>)az2Vtusn))S=lF~6mv3mBWOBg>~wsZeFwy54)g@-YSoIRPL)FDlgeQ3
zRqwXpCqZ*dIgSGHesoPSKB_JyvTpUL4ieIzp*q7k$t}x5Z!zdQ{*G;13E84<%Hv{n
z%fggONIY=*!c^AoZ+?5|T>s<F8rS3tsj$^&chXr8oAeK%Q_Bd)H$=~Ik4-S?X^-V9
z*>=81=|%396t+^yGR%sDtZ7|-ZdObp=Xdi3K0q~W%UY(Ve9onG{2%WCU%H=`^V<0N
zdBFF*rdbhusNG6DxobNbYbLk`J_=rJ8mZFQ-<6vmY;)C6UinoRU7hywo7Bj-yyGa&
z4g+lgsx!%AI|#rR-}V7U^?Bx_t{>@A#}*;85y4?IM>qxlPMO<A`*TZonMMWwf{>tg
zeVt=Lx*apxx>2Y?$mfm1#Cfr6z!#7|Q60ChD|Dkv-xWdrME0k2V1BxY^hdez!FYe^
zYo$qGoU(oKzmmcUHK20*Ut5=^r@-keR`d3{#qQj6NfJD4;{G^?^61rrIb<#I8SF<H
z`NEcE9hy%_5no!SK3|JzZOiTY(NDMZJ>|n2kB7|1$Yw2EHK~r$bYfx_qN|_TQ1_lR
z&O~lzf{LYFoFC6`k`tKY@lhVl+PYTH!E5z{^a2!Grn4U0aV3j5^8O5Pq=DK@xBJam
zzCvCuytoi8BT!oWbozDz%y)gDyY$23`UvjY_JBjgLo~rRZ${H@%2<w*U0w$2xihs*
znFD*s428Ngct>Y^0zpG~qnX&j@XKZ{O{4j)JMaL5M0shFKQHjKTH0ya`FUY@-;dYP
zq`kIgHCYD;keJ-if`mVsal75uo-!U@J8-zcB-%JJNh67ln?7<AYy!eSf?~R}sp_pb
z_g-Me#^GhkOKcZbDl@uz*lSygEWODxI0M@bWL;~1quAulfPc?%W4Qra@U^<{J#iYa
z<nc3*_+)0{0bsbMyp$C`U_1CgC3ASpLE_{A<&ZB0p$xD)N&V&4$8cHr@%<$Nx@iE4
zwG&z_92bBZWGaL0OShPzBzw03>k?-H7OY5E5Jud62i5F&toEz81Y6?;EVTk&&C35?
zx&}pFPq?)Bh8}2?e}C;nj{G$XEO=8KWQGOF2ATS69<4oa=*BDX!6Is>BwLBSga3At
zDe@rSzAM~ZklXC^?06b-iRzFcJ7@4xUx-SlsJdf0>NnuWRD6B6G$xs`^LRw%sYR>n
zJ;Y)lN#m;0Ag}W$cR>E{yR&h*4>K=+CbD^K3!Atqmg_hmI;=I2v4&L``*}Br>Q*#>
zKA<4)sGn{IPaB>7wxL1uqz_ZWDT#_Je>M?uQ_}10O(pro4$kC$Owx)7BDOdTN8jV8
z#dCX1#iK1q*^8k<gp{(oh9x1%G18m_eiP;9ks(p*(L~z3?gy>e<=rIzls>OCg}t4_
z;5<wTr(0!iVMg;h%kpxo*>=Z^#l-SkmV<O^@%sDSSC72x1>@(`%v(d&#4dUc-lhEt
zhnSO1553+ez@?ZWNA-GNpTVY7^vpQjLvL#iw!;|1S>nO?ma>s%7u#3Uj-FNwG8c;o
zThaQS3odQe0CmNIjRTMQhd>Nq{P3_Q4=%u?Gs-_$dP1vJe+DXYmhsl+oKb`-sz-et
z>OMR+(y=8Y<|IWO2zi$lv+VEzTGCXRnv@~V9I9z=;OxAb+TJ%S{0*mi2ho$}5eMR%
zuduf`{lz6lUi0MwAuC$VAkSdAk2VZZf%c>B%Af?RIvwM%qnwQWAU!F}<3|2L)bl$k
z%|UQbmoyWem8HCzcszy#_q^~HvPP(jTq8Ri);Rvd(RZTr`clk&3zT%9PaL8gllhEc
z{$dIvdCaT#aBRcixw!Sy%H5Cf<w=eosyf)4`jI!c27fmW-SOf2{N_ARDh^Mi83H-m
z1Ek+woq(mo_Zb$EFYT<&+_bIvRm@YAtMEDqG`~>~8Mhd?!UK+0mw-v_s+*Ti-Y2^8
zWz$}%0TR8h@9BvrVM_$^%3}IH=T2f_EHDvEC@Sy@BkVbW?the}+f?Z6dx*(90_@jd
z)4o2s-V-gj`XQG3R661B%9o$6|L^g#U^e&ClSGySF@Z%-5GN(YjoLiVCz-1s2tv@u
zR7Akl*}!9W1=R;W6Z+#rbXfm}?pIexJjvO6R`qkINLJQm?40C`P`staV{C6m5a70D
zkSfHeik2C!#Qsm{$Ips3>$IdD3zhE&$D*MwB`5=U`w@Y;ydNY)*q9U4)HUzOo&7d+
z(D=jP^)(i}Z*FfB_J8D67j-s=L?wPW&Yyz%AjQ&{b#d^{cQUAIQ@EIEF^lEvENjy^
zC}|tsV-5DBJPPP#b~<0EUhvXWFfXj|d=z(z^S)^2ae#u0hoXKEttEI2&m=w+enSEE
z-8d~~q&B}Rv3i~+e^|NzTlZ1XyhNe&G^l>35Wba*kn#OdgRFf+W^*5t%pbx9_-WJ%
z;Y+WaE&p{1f-6tC09iL&lYK+L%xAfS)3M5cOiC=rqUApf=7K9^yJzxh4tD$3Qnha4
z-mTr-sObrIs4~;Z;Xdv1xA7B$R-oL8U%C5a+F_!1XpM~6Kp*Ule$|I_(=Vp|6z=Xi
zt@GzUroc@6D``J{@UOd7QK52d0^njgsN2QBx;0lQ^y;yb3>K1XQx0@h!(*I4RQS|e
zgi?7LFz(uJOdHDa`r$bLVMWx^@9=mFjSp<bO~T#DNr9H)UX;(l^LMRc|MRp6YNy$;
z_>x&&RzET}J*b+WZh2CoPt1_2^<(svFFVl1t02)`g86m53?Ip}pNe?T#$w(JYVSrB
zDVaZgb^qeWU;FNl$EZ=@cN#!YK)EoY#~kNT{u1O=BmZ(P&@R{wn_E~^m(dpu(FDXh
zyN&gQ3GevKqvr%Xk&^-4wU|(1yFcQ=;?(NON|-HhP2u~w#XDDJGx+2fxXxp9QqYip
zv2|_nypv?A{>@imeEBh@-cgML$Jjqir<0L;-q>$cxBC!W?U0)YN(0;ksBq|qf4GVb
zall(>EB*StPSr)GwYZLegbRa&|Ir;g!;lmtO(#0e`|gLIgKP}%zd>@&$@|}dg7H=4
zS>`leA?(Pq3jIgys_OdmJmHOs(&kQEP8C%i3iO<$*gqel0-~R=a=^LW@$6XO!>@r<
zmJF0%Ukr`*kA7V~88=#3yZHN$6BG}SZV4@8IXU*3Z2;1}Dd~s~5MaXie+$<@!e7#n
zQT{BMzbkA!4xC@IrMRW1JomIzq?DxfIk5*xO9E(t08hOB^Q@)L9pE`W+&Af{cm$uf
zXgY=k%2b;cKx-?}e1gHlhJkW<K9VDW&u%s%H(jmK`A@L5tB=w-ZL|?pkINhfO3Jd-
zO>ep4<{>@B5o>$Gjvi|Tm~!4iKtP#0&0?Qp=_%W<biJSXv3+TndvQ*{gvgfoZr(+I
z5Tfbmry&Fldqo8y@kSHpfSM3;uYdound?8duI3#-mWj9h92nV;Pmx^F9&L-!=cVq{
z7{5xyP5~u|_j6STM2R@&NR~qPPY-9qyPEntL&yKME@DOq8rVVsB}Dy<Jb+mfR2V5I
zIv<BV#51;OeT;MuasKnNLy)-iZGk1L8hmE%RR(Gkbl^qK#$_`C#K*=4R0eqEzrfoP
z5SjDkq?w03##XA67gjW*2BxgU!X6*a#GDIuz0}x@>2foK#8MHfgO+6($CD-ShC1!`
zc5bz?i-@%6FPdaHKN~D0@qkNb&yDdxS6kxX!pPVMewUZtvyy%usB;ZV1)G_@C;jjE
z>Vt|pj-a=quF(fKa-XC=P9SE;c~n^@K8VdIv#m~nDkZcUKnqmg;c{|ONn^}X#H7~e
z$D0Lr+hWQ)&~2vwLJA-<@0tZn1N8^Bm|uEeH@yx<2L#roD!VNVLk+PXe8R?$jK5uL
zurK<~oS74dKkxlcu=did>YRc~wtx2$c&}o(ab$Cbly-E)m9#uyuq`d;Ki?W%`+A{t
zwn$1j{-Oe-4)&+R;0SQeJy=t1AE}x?EbE)E@!ix`7GsV@{5&!^fLBd0vv$kwgL5vZ
zH0#?8lphi{I--jTUXH7h-S(`|!vs8VL$oi*bGUyzd6G~neQQhN9x`*^zU#E}v5A6b
zrQ_zeT6X&w?WxRajDuQ!1d4D?e=+vpc3drQkI&>`^i&RGNx27!iXEkbvcDJ^b9Ten
zkIfQn6(#>3XLqzD{}<QiA)jP8t@{=UHLpl#_Y_!Me<6yUfSfF9&v5Si6;_aqUnK=}
zgJ(_JRq_Wh9{=;K9L7RP5IE^sk;#?=GOx$o`RPHSM=5tiw#054Gd6Pvs?KR|!wr<g
zCC8W_h85F6HNEOzYo9BC`$Uee7D*i{0#v>XH!nrRggw=3TNoBIT8Q}ASr9l-tAioj
z=v83r_^oRu6zLx=FQlM;PRY`_Rj4*@T-H2VWCrWjqw>n~4c}nBV`qT9P)d~%beP^~
zKJzHQ$hB2rq4%8gaFUuR*BZBt&R|m~vU{q)$#1F)*4h|#{@yr-ilBvs3!OYz0g0J<
zh4H#mIDlVPNWoupn2otCr4W^cEr9}&D-mEAL*LLO4|h^kZrc?MdF(B{LAs-k2?s$f
z)f2%46cp|6Gu3WLQ|Vl)a$3A%mEq9cA48vEq4S*PaBv2w-IZ?=PVTORVY>j%e11DA
zOUWF0a~^cs>3wiS#PX}&ey9D;rhZ)pde}Xn#K>vm4OPBihc#C4vcl6mef9P|s$mBQ
z;EIYDGc~I>kG|s=Is*+rO&uy!A+jv!f3gOM2M{JrauYyG=O$3=n7X)fh>P$@Bh={a
zsyh3YhnzFL3WrRFonTV-(TxP>cZdaBRBm>!{qTVjQcF0Gw-7IdY}GXX8g0?EfF16<
zLjbCeyU(x7UA*%Rf?&7@ykqx;&HP$y+AKnW`IveWght)C!?l@JwMFESg}}=_b)zF!
znYi%Qrg;~MU)!d7I07vW1RGFYmiz*$OFg?2GuIW(*OtAB+%!jYl5EwmT=3sBAN$V#
zz#fp?Uz<}t;7?J2VQRF~BuyppjK#QG27ea(!}=xbq71qFL{4ax;+^si5N*#B8{|<s
zjzr8#C<m*Ix8FB$g+k<@Sh_M}hG_Nz+X4gES;cp!eq${vFc;MuJtxJGKZWed4de6G
z>>7&bib+1UdO**5x6QKq-t%jrH}rww{+^p0T9e;}J;(eYi3ftM>^DovFQlq)W9=Q|
zod+?Q*E?x^<+=#R2g6PO=Sg_it%_$KTl8~vMFBC^_BQjPL)A61gwen;ChUkTxV4tI
z%3aYp=R{t*mG!nhGo9ctqMSoU9xu9Cu%f$60M1HY%B9>A<xXStd5R6|4;UK$A6r))
z)`a)<bq*;NCLkz1RJueRB}}DTLXb8PK?Ugsrvejc=~hY_MY^U^qJ*T<lG4)6-a9sR
z{J!shK9A$>oO4fp&N+8<gzgIx0?lr*_24|4bdQfWcdv%|Z&PcX0t^nQR-dj~dHuuR
zTl@U{Y(@u2FW952n0``#3Bx_5J_uQj8r!^k26xbj#ae4c@rw~-Tto|Vz{G3x*j}m@
zV991YOM@{2cie`ZC${sHSF2`Se9Z~gn{*hG+9q7IoO%DgP;|&kS`gf*l!nziRog=T
zKJBv?GEtV4lFPsPwE}0vZBu&eqSyL>^M`Lv{&rcKI$sW+!fxl#M}5bO7OX7KzUXRq
za<!$XPoB*AvRnREK*vj6vZr-F<_}=?w9O_7X4__|ptD&H-yhyk@b-{yiEfrK2n<=C
zMb~j&#M=NlP5$e#%~Lej#TOkY@y(bJ!7KYiIe0zeFEj|fPE~Y{GQ%`)eB<fH>j=RH
z2RWG$8vIBHYLq->o7&*q@1;??AghrN1Fzzu9b}3R;@?>Q<GN8$O8<zv<|nG$jh>J^
z$pRC{CA=q1*nnynw9PDBYSQ{+N_Gd=1g1}}+c()!gCTQJ)yEpTj+B@TUH+fAX$qB7
zmYR4>5>S2YEcNA@D?CqFA<}uNipe}OLgFF#+fc}VjQwhR$oDJ*n{6lM#;}~C2%J6+
zvrk_SK$H|oo0`&eFT(<Hh1A?FFo0>d3;7%oxJAN3fPulltrCY|K`QS&6~__o1a;G<
zMK|Lci?m*$fkC);hVcatE_hNK;r;Bl0M_bJaG}oW13zplQ<6Ct;u{E;X~1U2*&ABL
z8yY|zunI8nmu9P-_fo*SGQ{UR1__*GYVn1{*j=hSgr|4>d7nEq#bySqFq&mDc2i2?
z5w`=7pJY)cVCa_9mj->h<qX)dFuv7x13z={k1`o8jM@Bs@Zf?-%;ZRPjJI~pTqk{D
zg=+3J{6uUC;j4dFL)LEOdN~Pwq{n4tk@k<)6&3eCw!;QmM)*D7bK*a7qj(bnb>TG<
z`mT5MMRR>!{>;*Rq~QuTzO9^Bb^#<nu4PDe6MLCIx<Jsg*Ste*PsSgVA<B*pM}l^E
z&0L3sffxJ6W!MYVKU1S1*|k79tSwR19Q1-yD8I<N9^VCJ3c0sWRKWEg7vlf?KDDK{
z>-P4}6mE(-VI_rgV|Z;YyKPxoZIkU7A?HHrQ@D5A?#ySJ9k?r94XBC}%wXmGAJzIr
zQ8M!mylY_b?IHEvCEBNhPNCb_Y5<w+SsubQNg9|w9osZHpi%<*7{lJ@Jm7uLPVaYM
z)ZdF9@3@`q6|(MaS&+f+dX*iqt0=f2=wfg(aQF`-`X&|0v*K!CJEm@V67{2nAB@3N
zpPAJ2|6h6W4GjWdBFNA7X^9!AM_w_mG)I^yi3?o`!9DuBClUjiNAY0NP8mb4mDFlb
zp`2FvQAM8TfB0J<UpGD5TY_Cr7|XB(SD5Kzd-AUHL`8hcrtg<WwDJU{LFSOC3>M7o
zI(3O3UP|Q8FS<Y55pqKWaeH5p1XSJy1<E8VfA<#*py*w}{wQB^E^o;+-kugyq~I?=
z)<zVlQ=a%Xa5ufskUIR@7rv(v2{eaE!>%b6yiACHq99{Y(0(^PLXinc^6qNZ<D`b9
z95#w876t0E7|5(xI&ErXhK5q?9qDSu$~7lcTGPqZ-#eab64*t_`yO3eKPrq>yG|-C
zMwPnM(ZJ6R=!TZuI?jfCd;p9^@;$n3=`eRN!x^GmNUt_4%e0+q??pz9|6eHXA4z?X
z!CZu;6Ym;Xr;;_hPUzZKgLD&SvC4O+ZE8>NodP|^8Qhj(u{Ed8JQs!aD1r2^ZVSq%
zNHbgM_U}8(^Z=xE;En2$q})(m+WcUB_P+eQvGK3I@ZkxceqCwE-WQshLd?^yPZS=?
z%KhyJEt8;c4x%-<ZvW#9{=ac#gU@k=Vyaw|9c4lw0d6NKGeX@Vwszw~^XGnhC4^Q)
zM(v$6ln)UTw~#j<x!EEnq>Zw2`)SQFym0kl=E=Py{xEsu%24@=+-$>T+Us^^kQ17`
zc8pUw?yGf2g+ua~EVY{OETAq^8pi6!kC67u?2M7y_fk^Agi9+7)M)-3gA*F~?TYPV
z8BZ?xd24@Q8P%|{wcA9%+gKymGSPb!J;vBVyK>9DT$>WTEW~u<nBy?D^b;4P9s7rO
zKR9Sa<9Q>T=)?8939#70zS=$|1n4@mjvgKV`oXNjI*j-{O;5R=stKbnrNc8r&@y_@
z$@<q$Cy@Ei|I7jm6+E3zI=7$WYeQLpW2vh%>EYhm8`6hLOBb|Y8)1vM6c8`EEvRi=
zq@%sD8d#jND?gDKZ0Bhr+@^4uAbvcFS+wgdfWm5>u1A$YkgBVhK~Yb+!i(cQW%?>s
zBQAGbe~D*^p%mk*nPu10phfoE1Zc;{7aYSYqWe@M2?7zz&9lyBHb5!zh&%!E15ZGT
z?6m$$4WgG!b0$cy5n2_mA77p`lWDMX;#U#902<x2lbgk|k2`-E(^o*t0#XXne1Vqn
ztdnxIs8tJTy59b9+fN?PK(;npju`txuign=*Z#^~@~?&T1JN7_1J!;!Fyafvyn))*
zOd!r%a{6ahR~6IYj!TCl%y-$~#u~q)h6x|$rBj%U1_q#feeblHhk`Szu$5gM?dSV$
zRE|qX*Scey+4>pP@25HZDHO{LzJ8(R=DT8=+uB6L>9Hlo>bn`g9iA3gRZZ$OUbm9z
z%djS)gX7JOppe%n+$#p&Y!t6ttHMw%7X|aQ-e$68?~L1Ul<7+y60YUIfBL$>Z$w6?
z%VJ|*`41}87r~NUsp+$;gRMdOmc(}c7zu+ermoa=+<p-(<AXf`IQ`AZo>;l?EUE!(
zJcNEtU^VN60%b;Fuiy=%!jUNoLzkZ)uby>Xc;2aF6T6RBKX@<)w#lb1Oz;4I<5N8a
zx?g9G!OmF6d%BJ;H{VJR3&ej~uMh(O-IS%>?X0ZaWQ-rlxMOo-f>xPxEr}t)yAasD
zLOzOq0dB&amZu}icY#@xWcR@6__DQZ${_zEnQw$od6KErB)7T5Q>`V3o#LhWRWxC8
zB1FLO$HU_N>h>7ob=|tdGxc$4-NjIEZB(KZ0#S7dZp}x0)Tpy!&o`H=(n=X`&!I|J
zhh4-LyTbb&U+iO)f-`s|V5O;$o?f7vf%F{^+sMTDS2~!DEsDcW4zF&@vTT)7g^t*}
zCQ97z;Hb*-69b{P>y4#!lch357KOS(F_Vq3G@yunpmOzw0^R!0rLF*kY_mL_=<9oU
z6F_=F1ypmIZ*OTSg@=;?H#GkO&6z!#D}n!S0%aOVR9JB*LUl*&;LCuArzta@NO=s1
zdl-(|t6m;?E=wRvcWAR?im7)nG?QN*XjuguzSX=umL#T3OcI=Obo<@rc*YAqCY!!@
zfT-o#r61;TCunF`x$a56{UdhDKHGG+R7&hJ<BmYqKpS+mh(1>xCjIVxyV^d3LVhE$
zZEWe!x0uS2Z4p!3;7g&{>~V1%uMHVqAqn6-fw$yVYg8OdjN{C%O!b#cCubCs7?3ZO
z%ropoDalA0Vph4lk<A<2j2G2+9JoIQDk^wQ3jDe{F>Wsbf5<~j-stko``c^1TSnI_
zKT+DZV7I@$9Y<p&w-q_}>jfXAGR$jyR8<SB9Z@k7+<ql9DK9RTagaGAQNPvs7J-+Y
z6Adl~J#Xs;4Q7j{k53sn{2bmi7Icby)B)QcrJiF@#!?Gm{94%Sp(VFrxZ_&lVnys=
zHkw2iVZfGJoHkpY*gOAjirIK5lXBdv3K1gZy#7*czlq*}9-Db_Tiv27S5SGtron<;
zr&YJPDcV}L#)$HiIv@VKlLjH&SS=6}%s2PAo#_|D&V+@U*t<FN?BiDB6v{REF7{e}
z_aYmKTkunxMX~fjgzNokP0T$wfzzinB~pdr^~Z868jCqw`^CKOrYvtg+b=V55S(h=
zlqoZYD-}n7OlonYs-F;wQNUjl_>t_uiuMqT-b7EQVE&G(9L#_bWuA(dag)zJGveXM
z*R$@h43AEO<GjUh_NA`Lba}OTKGX9d1SPvnF?xYYO=4iR^m38LRuL|ckigzXAKmBu
zwdIBuU6WUX8&$MIE(7ga4|Z|v2@2pIDo5i@?XW=b$ou!nORrFC--fn8f-!;R(%f|5
zBi1T54La0v_e$NUR^~ZXopIm`hm^;EL^FpB&p_m82u__L1=(Dk;5f7Kl0c5p3wrtH
z^aA#MUb+$d&SvGcC2wC6H#J~~ml{5Zx=erVh1VBQFsPo!-<WAIM9f-QVs*Qj&ESoP
z0dO@RNo{K^9_$zMImkp8W1+70-59?kx7c+szF(W7l{tpLk@%lIju(ND9!gp%r+JCA
z_sR+@Er^d&EHJ7wSBcE|Af%JJfjhqRX8Y*<EMZlLG#qON<Kth=;lm&H9aOED)BF;5
zP_IDB32B;;o#M!#!^D#20S-R#?JHpi8`-*yuM8(NN!%Oz;x_bU`m!E>lOcqv!#u&c
z3OW7KVy!3q2IWkx-F|M;E^yr2&)6vlcUy87;;gh`fJLT=H;DZM=c<><o&1vLHu|5`
z9{0s73PT`fC8JZ!Q8dSXqZ)Vl#T>%|GXHy&nELj2pQbhJnx%?XJU38))reBzODCt>
zu6(%loFIxL@=1~U@fSW>q<&fBOF4IRpUu7r#8j1MSANM56djl7jDB<KB~Zs1W}uEQ
zInsG4PAD@GE=><YvlXN^VPyL$zh{!>+e1M;g$*`hdrcCDV6z6^^cRbdXqz7gmtSd?
z5WlgZu<(J1sR&F>&GX5MVr(xs<u?16$Ug1yBR9MTG6v6ah*Tynr#xfr(RDJ!&5ekR
zdyvVHjKVFBtM}7G=7Sec5~nZ7Lg;Shz?(}s`W!m9PvI!IZDr+tO~&&okua3m#>`V@
z-3BgXJ8rHY<4z>LWHN0<wz2J=mDu#=>^2ph=-Mlprh?J2?SerHdE9z@HU!JONuIJE
z@V;`X{*z^Tkat+ie#<@+JUp|R)<Y5=lL7yP6DGTzT^b#%faJxQV+6(JL(vvkfIYpN
zh3)lM8qN1<tj_TUu{6Yk+{$Dn5gjv->_4fzW7WKUS*9Y%^lc5n+I-FiH>>9K=P~F@
z1Z@9@2?e&)fluAm;&Zp@PGy`+#?ipNH1l;%L0BtaZ<ugrpN=;y_yETciX*~FVdKcY
zyHtW~Y0gJ!A(+?3c$&1sZ{Rd5)WvbLGNkP#<v*oxhFB}-G~4-ftEF4Hv1<`u2AEhT
z>tMX#7W70a)tCGGCy>mi!WMoJ=|T*0GpnO#=l!d~_J*LDrTtgw0L{%rSWah&s1=?C
ziv-`@fYnyZ)L{AkUBgdu!x;Ve`eS+VYE@VNM~cAi#Re*G+RaXOMy0uG;Y6ScM4%Hg
z>_y8LAxd!8zoENl3dN}yjV0pCX1#Y{Z+)Qq)FI5+)V%UqujVE<k2vTCe_*+YFqdM5
zQ4tYwlYDf&`+c%q8_p_7z3vP;!G{%2?Dwu#D5iziM+16T!`%F)(ux5$z2oD{4zS7Z
zMAAsW&}(KbC4ah~u2o$5I{;lmO1{rfxr#9|f{Bi85gAnBM8|y9g%4K|sg@*aS}SYh
zg_h5R$8U+0?X_;|%0l8q8iRozTV~Oa-YU6&_`1nY9dbQ|9KSqlq@7)z?yuEE9ofAv
z)qtaxb2@rF){c`1cyxeLvJ9OM<=as{MWm`IVR}&w>yEC#(#x;vaZD|Zj;jdho&IAb
zKPmd+^y8C#dTlGp9eOEE&}dTs;wO{jM0%#-ifZX#EKKCup`~L1O*qq_`|LXTq%gb-
zJYv9BUj;;4liwR?S8e6CAOhR?F!I3EkytKaCTu&hmBKEr8xwElpujD=8Q|g$8h^#$
zSNJMBrm|0Z_AbnRe<*0racgItA}Ws9VCkq8Tns$gJLa`bLmLAD`ktR&2$*?_#(PgW
zwqSi`Y`{gCH6VqTFq<iA14}|iNnm8>gu=%1W?Dw|FQ*$To18<e5}W2fB9%^Pdfe20
zw)ah8`rw`Pl9uA)%5`<_#ZSJyItXGGsDGtnWTtj%6k`LoRHW}`cA6Zd%-D6!EsOE)
z$Rwx38&xVjc#n6E#<6!PQZ7ZFsyNGPpN_T4qXsq^IlEIIW^~ubD6QN8(rih};x}h6
z)W5v8Cv)Tm315oOmQQ~?<$Mrr5aa*C6-hx5-A={})r=)UNl&@V*e{TE>U6vh)+b5G
z+WV>`I7oaxc{BhyT!Jlx{dDXMAeJL^lPf=c;7!x=+XT%+)PUFnQ<h5l3ORBuMtGGv
z3b^`Nb-TFISy5&;_5b<v-+rLFwql*;zHk71^k7nMT8ejv<#OMdy%^DnJ9AFL(Mez?
zkhUt+s$`~&IMq(FhH$5I<_t=1R8`Sw-(v%`9=d7da*Z-0yth<wPi-^dU@*TgGOoAu
zH>X%_-qHWaP!LtPMYY}Vyr3Q{Up<u}!T`wlP?B4c>D}QsNvO}lVK_3h9D6D0WopL%
z%LPYpwWhVo&Lc0&jB^U$?=Mf~CKY*iP_!E2T{j&K?(d&EIyEQKaBtig`DE6;PmDXq
z*Ly)iSK~>iW&Q{-@wC9`ni*vV-CIKG#B&D6#6(>01?+ZQd#ulYbm=9g6o4Ic9+|0M
z4}c3lj4lst11QfmJG@b8E8}!={zrf5`iqc7lOz8T0zE3uGo#97QE*FBIRr(ar*JgP
zX`_}uCBk|fpAfx?0WH=?Cd-XRFM?tkHs1YL2JFF_o<p0eS|&5{{1*=M0)q7h&2)={
z^)E(K;#J9ZPnFt0k*O%Q*!QS>icmQcOv;101R!V3oxb8qM=k?;$r`>`++;6g;y1@@
z@`XK3&eeT?BJf6mK-Kxa+<m(67&ArEHrUPhQo49koae|@Z~{Z)mtWw7BYg7s6J+;J
z$nSnoKHiKN@MvoKG@Hq|&-Z8Rv(?|9ioIW5I3-STL>8RE^;OyW7JH#RFDL#C9o!k%
zcevCHKAih$KAU6z0fubYE0w}41L1>Cc2PKnJB+UgQf8cC6T7mfk30fEGTWqcMH6`e
z3R?gDdu>w_>sb%=e%r5D{Yj5CADkc!=RY|X1dH>99Ks875qGZFFtAB2m`&qviD?BQ
z4}B2AOIwl#5+YM?CnH7_xHhuZ`8TP!KqiAv>x3XUqfDOfFZ}%>Y-GZB@P_<8&!iWE
z3{>;5dpP1ZS9bn5_-si2pr;nGgN5P<ke@H27l#R7+HMvnK&^P^glU8>-T)xWeBMm1
zK{C#Oy*HVc-|us;fGPLoN(p-7NX?2n>+ufzVGufUl@#C3mYD5Gx_$?X)8NZEZdUK4
z;ru)`KG0yupx_#U2Kp~xGaD>=|0@F68-(R5BQJ{aCo|Q>E8ABK0d_>>-uJ#A1Q!?B
zM$muxM`5Hv6(XNKl)%dY(6v+SRX}>bYw@;Bg7XEtSM8mer7vo3mj86>3fbooBlv?y
zE&!}#P=ECjy&MntCe7X6i)#SO`5)pvI|IE*SG(%|p9{!ECZOxBMYC2>V*_@c?|}f4
zSE^N={(A-04|>h+d;OZDs<;5(UR(IRu}_i!iFLg{Z$IG{w;q~YT^Xm`9B7O&DUrt>
z9*o2VUMe^~H)X(U3CHVRV?8wb^EdnM@ck@tJ#wkM=GJh$Mf^G&&<Dt+x({6cdmD0(
zcLymmf@Er10@KQK@v1;LC7%|>^!EPM<SxC&8|Hq%m+54$fuiiY+n;dD4x~h}@{}^8
zRywGm<)o^U(VnsWA;Od!!d=_mZJk-^#%&(Y9Raekw*#L51M}<z2E2iqb9HcxI|@{Y
zr#p|`E1Oo@XA;UVJ;uoa&WmTOi^2+;X@NV;)SCxe9QkY1cn|4+1=u?66wGec+oTC*
z&GH`-$RTzsayeDc>;h>u_L7!iCy-uVCgkQEI5j`$Dl^mX(x0p~Wh5RmSGM0MZj#s<
zY781mY~@cdai!q$$^=}Y@t-Sj+byiGk9jWmA0a*xMLjo`noZ%?VioTb!@3CrM?|qI
z@ZW1-fki$ne=tkczk@24dxn)`X`=Z>EDe;z@shDOg~fd%&qUNqNVLRAw#~{-=kF3F
zVg5)U4d^dyx>@lhuZJ=*r;H%+`)*`N1HwK0begfS{u_55j+uex|1yKcdUt%3-k!IP
za2@@1{FlnU8}X%8uyORznuUmu;SlHVy#RLmA8~GCElQobKN@HTnw<Q92%w?K@NB#B
z($2mS_(+}n;)4IRaL6*Y{avrIqy&7P;Xi**4MBh;H2E`_E5Rv}IT80qOon1JI!h;h
z_f9!$n9h4s1}RrMOYg6sxxz4sAa9axdaa@UZd>St<`)FxucMS1s6`&NJw6-R9$mF`
zaPRC!8&COleH6amRU0YSVQjF^(~5>68`HZ1hDj1<ONlF2c0h-R5oQxaR3gM6cQq|%
zOR=pz7WRy|22U1j;p+dO3)B!&D?j^UU&L8!0hv?m*ZEA@3vZQuFa+pHo=lAgsg{ur
z{(F<45EF%*=AZ~Rb2E<SLVAHOu%OJD$A<|x`d+Dv%z)zKsBUJ%M;1+-cv$`W;%@kp
zna~;KJN?PiRoUvR)0pF5`gEx$_mX{`jjQHLU)DUo&+s)4pyCjn2z4X9$EV@h;3&)R
z$2(OCvF*C@r%mYkt@T#$-{=CB3ek`Ejo`Fe33YVh-*p|`mhX)HDnoo2S0Yl7S#)<)
zwcf;_YV<fZ38!{aG^Ju9^sp+PGSm4#&g?bD4D<rNQg2zT9D?*y(zBkb+_n)SK8j7q
z6183GdVA_avf3*zFAz#LN4^*Hl#1E?W|el=u^*Hnv?0#WG;HLx2VPR=zA>d#JDY%I
zb$WiG*t1;o>dAe`-Wcuo8`d2~VIpN6(p)%YY>@%gyNAqr=C3D}F9=o1Jwm6hb<@IL
z0aZEcou&1RHF;)<xZi|=QKT44+U`X3$jdQYo|Xcc*b1zGHPzN`q?}_>YwMaaK@F3_
zAl3i|qg0NrxbkYEhjRWaOg*nvL5>)hLzY5q-92{3Kf9$Xlp20Wt-J($_Y;qcK*w@m
z3o}GImVgQDm^$wZ3i_+$Ah21#2s4%?AX0?tUhXuZ0}%A4yy$PHKb|9!7sRxyUUW*v
z37;NikSy|Q-ft;QWqAepSuR2EC9Jvog2Mk6lvwY>G}$Je*FX5~j{Tx^()ia^z$M+x
z(~#okI7YE%BFN<xc?x}c*Bs(lPQ08iZOP&qy?T^rFqWBWU&;lXRnRo!!ObB9eTAa~
z&NrBC?e1QL>whU56~cc`gBf*RIKHTQhq?3Qn~Im9p)eSTE)R2pd?xm38kz^CbMsQD
zoa*0#uGMOGsn`x9Y?6uy_t1lgsVPf{v!1#x;bCQx)SH0aLN7DO?bob3s1COW9_==f
z$4avt4}libDWTuCUnX)ysu{$#tDcN~6jR^x51H`bMQ)3&1m3K3YS9x>pgft18xA*1
zxV*p6RRg^7_P^x6kni20nd)ON0J=-sqoeeNM?V`d?#Wa`6y;GG(E~cfk@t@rnyq~F
z(<Xy{p7qnW0Ah!W`bNHP<B?pz`p6tqS%(CR-L0DA7+~14KDTB<dzTKj-i(-wGu|pL
z*1i*?8MLtQBDZQQWNUtT1s%n;!uFB44GbO&2?8k=Rxh`qO#qiCb1Pkyd|50x@k_4&
zEA9JTlb7IdL?I^XIn7jboPno7&TJ?XC-dY_Sl^R_5B!l)<+WHQB~DczTxx8;_z3~r
zma&X;>?bWTylrHj{r#h3YTLQfhi6F=oWjU4?EN1}$+F4zR8<MPsNGm>ER{T{95m)~
zbj;qW*<B1ihX}hHL#Q%jkC*TjiU)CPCfT&SbDG#L&{DEyb|PIOX3mrWLh8CvqgAA`
z(<|gUDeR8Rmt$FrZYRW-p3{6FI&+zn018@?r#QRy47fNJ`tYosw<{lK4Yu37Qkou|
zK+I$+byOC@az*nrovkd$<#=x~&ZJg2j9bI&X-Ioz9SD+o1QBaimk=-T{&Ha)1&}F5
z_^{ONW>a@!%Q>RWmfB%hZDCuFjOCKj#O3CTtYt0$a!poPm?i?Z6;N>|5sovZ_dq|w
zxC&W(2}=)~q>uD!-yg>GULykwX$m)^*Q%|aPcZYo0#5ZHD?jKZ^Jw;5Cu)(<Dr}qt
z+Uj)lop;X{`l$PY4;o>!V}tfjiTQ*^oDCMtLZof0jiufUUkw9iB?`1yxf`c!*YBr|
zH5Ab1j;C$yW38{8pqKjwqJKY>m`Xlsl7UFuePCf8&y$6t%tXNY(#ZP;(3T_an5fLm
zJ&j!cGYcRs9`E$|!=Ub{ItnV_)jl@e)o;Iy;23Uz)E@UHrd9`5GKourws2`sD$k+l
zkydayB-?jJf0$$SQVu2Z|ND{;<n9uM;!cc)F5$Wyt^K9He=01XJ3Agl5+4Y);tjT~
zaM>B}S^BPR0^;Ul>$u{b#`uwO11kaIvkb%a<5!;Y<}F&{S~JD6N_qOKNAV8y&v`>d
zkkUvm$4?O%LGLoE0PmLvMG@vzw6J>G^9g!*mxCN6X){UfnoD1)%zokl`|Xyp`SA@I
zv2XN5vbA(GYSrI7NriL$RdygM-Im!_A?nK+vCS*xTnea))X5iwh}p*67xcUYpZga1
z#E%HFLq5h_M(Kil$?lrJ?;bH?U_O8ko?x)AU$_topk*J=NS!<{(7U+82~XS7y|!<r
z#lH|a5b0W<P&Qd27*zO?;V784X{~QAU^F?&U0s&6P(wg1H~CwqFA2nB1F2Tavt4&+
ze$B;sS*6(je&MrC9N`Y2Iw>ovL9Kgp#Y~&)1&%Tkp2SNcG&FC^i4O~2TmZ^@AJ-Y(
zB{#JB7CB#j7h9FLocCIZ*n2aWpqM64GM;)lf64^45Zebjnbrx#R@<K&HWr`M(GYs#
zztWLPgNnUQgQ<yUI#2ln+wwA#=@E*W$|TPCO&>saZ|?-WtWSHGfw)M+cphZyo9qn*
z)sKo5oJ!vkQOEnp%+f(^(ps_eFuPE<v815<fH=Cxa^GQwU_@j7;k^!HF@dF+rD|O2
zzpdENG`WG+=Ka^ruOdt72W^rY9^9!7NRn{@J!hE&G!@nR=flKYz9~W)mt!%)oXsp)
z)=rjo`@NK92$!-{NW)ol|II;oz#8&YaxPVOM{4>V(H<Ipu@u8vZnY2#`zg<NOk3ZE
z$TMZ~1`q05t6a(`TZ*1p1sN5^_K$`Fk#czL4U#~vQ+TCpr9y=zh7`>&yfEClDfa9X
zD7Ku&(dksL!uMh4ZxWD!>0OBK0j<gpD)yp4b3c&>{Pt=wF!#J<yZw!f<DM$MlsIxV
zDUv0udC+4;NPqIJE2`9Jsij4%(qbiCo7j_JzLRg)1DrHSwO7-6<}4zAkX>iA1)1Q{
z^$ETrb$!(T9RX9!I#I$6tlY?fj>P9|KENy=___DDp0eCGzu-qUN?TodQ)||V_gtWG
z1ox0cN(A}EuOl0^>4O*MidGFi&1BHy`4zf@Vqygu0~M>S?(B8$)`x=I1~S|0JYP!i
zH3A`s*c>mp&>uQqyc?I17u4Q*cgG9ohp^{C^JD@(e3>MOn=t&7Z;L{tUZ3#ht)3g_
zg09ZRS*M2GmH%d-PIM7uVb|E?E6lDYuiiKNDqx3CZbkEP$*qk{oK*f{3PPkyrfb-c
zwWv>!W_U=}AIbY@%$t%b`d2Mygj<UX@p=9^n3@>U?iTUYA%y9=NE(hdul2q@@B<aw
zF;D7ggYYg(F9RdvsV8pTQk}hA=9UbmQs~DyUSBMOiW5gx@Px?5JCJw_n<?^6m;yC9
zJknE~xiJ%<?*Dun;lcZcnvk;bp*YqCxgVB~cHQz0y~_>pd@!gP7DsIRz$0x_0+`9|
zhJP_I{11cDi5{Cfw{FcwQnF}C;yEytg-rT^ZqrNNvMKPy^|(M3_OoP&0ILi2LA>k<
z-}V4?AM|Efqu^t=>cgyXVc_YBxvM*UaiwSBfzwyH&k%0v7WwuyONRuLY=t&Hc}&o(
zL107yk!krSBLkP)js|@04j7rayXy%w_qaP8q}!qB_0;1sbOfMCF9PZ5X*s)*Y-VIr
zq5-j+tzEJm&GDu<a|BNgzB}s2@w-s;kxY)~>W&qU+G#~O90ts@VYb}V6#WBzkAK83
znGgm}LR}|E?RV!Y&Com}uSm+ub1$7)tVv(r@4z3D*rw<{yr)tXa}@OyaCl~&Y4k@x
zP+*uK|7C3$J`+ZzQucxFngst&$xB(f-?z8?HqY4-0q#P#P}Xa0sLwAEpgRJaXsyB_
z?j$Z~3BQoyOJi?JHgc;{-$Jd=!1ox@^T7-Pb)AT`8Tj^uV*RZLi3h`5BDphUAe&w4
zeVVkILVQ{)G{8~bu7SN&FJna#{n1*=$8}Zkw>b0&40~9CVUJ=er>|IJcT}ZB-X*cr
z=e&N2+E<8G^@!?Fmq%qG2rpw%*bDCiPth5JDEwP}dGMf}-S}&?vxVZYAHFR>#&kVt
zt)|1ZevoA~>QE{*;8kfs<qDD06*`TI8(Cqz*zIYjF#DeSInXgh79IpbO@363_!v{{
zC9UaDYy2_4=JZ7owsBg1`NT&dP4gjJYpKSw9jf&nF8L3&1!Hli8S^33^JKH*1r-g2
z3G$aXQ6dR?lPGOw6t9y}@lE(zZg#WG*?oH72x5W3#q*ZcMelm8bZ|W%nY`Fo5iARg
zGmPuq8@y6r*h$)QJ9_#bYw-ijV6TbR8PmA0I@e41TTqvSA<v+vfTrihxWIlI{vbkm
z`R?mIX`rsoCohV$tzMdc4bB2(YZLeI6H8UIQ^b<zGh$0$Z$BDqSzWfws22Eg?<k3u
z=55=Q_4ebkCkS|=Xc(r(Y5rxiTet2{ZRH`q7gg?q7q-?sxZzzo_s_KxE3By^g!JrU
z%!qoA`^<2tn#6fWoX)x~Ryb^yG)%_)JF_GJxqAm#Q=(}m@USgdS((jROEq&IV)FW`
z9uG2)wcFrh3EhlWOH*`<%(FpNjlXyX-wn3<4KPUS_<4evFC5dCTgy2bH>jK0`jJQ{
zkftk;DhbOKpYS|T<-g->Zkj~Mq|k@+6_pUgj-^vrcm}2fJlQ&F8$#Oqnro-wwZffo
zt=dF>g1RIJ*|;#i5_`Yp_SK<4;0BdeHyR4st7*B7G~az6CVV3)i7gX4d6hN1$jFvA
zc#7idtf8}m+Ct-$+P!nc+_@({o)26^>4gFiG&IW3j?22=Ij>>key!qZa=b^$KZ2r4
zxjwP*MRfiiIZ1BMM(Sr{@qCSszu?5q+4KiQtc}$QxCql9^hOL4IQJ#xLP#0OlV9NL
z$vQ)}#CUx-0S3EgZyBn=;HHMmcAwsn`wB!P)!i^1tf|@k{Ib~<;}t0K<VWOLfJmKB
zzY^cr6b*{+L2hlXn*^WwiINHJT9zc@wEk2r%m~crMFpbOeoO=>{z-*#vEJ!)u;zOJ
z06$EV0buY6P33Dm-=ihn>OR3$-LwBqNX+RCy#Uw`Tgn1;$+l`_o`F;QH<RaRx4`<E
zG@=%R=@AkueZ~yOTesWtqE3FQH~HLvYnJh1;=yo_rVNwTV=g#B7nNZOU|^zhDKwSI
zA?*~%YPaX*Q$(_a!K2MY+WPwtL=3CuvD9%<02%=x|3dAIC-GmJqNSgsO?dy!F7(o`
zdv{D#(6_FWeKmU2^mM5vrp3Rd<}AL~xu+ne{?Muuq=9cbTdD?CZ=Wx?rppbEQ^H&`
zYQgKrxy;>7f_{8CzHa4igEeYD6Un64#qX;uC{%3slSZues**H%Ux52v3E^(ewazp;
z;0<~Ld?^vhI!OgCy;^~O^L;D5As-n}{4$ckI~m^-4fD91Pr-Z}WEe4CYK@aE_13mD
zM2e!J14bLh-1?k{4%S;s?3U}qe=~dv%m2#YKdq5D{$p}YPD!0bNko-!!Qqsn>xG~&
zhzO^DRFDn%3r=^-)=)({94;jWVF00GwM!QtWL8E~pvC|$#IKu;oB+<71hzm*;&++?
zug{OTP|tVd)0YHF4>mhffZ@ME74C<uEt9H*#1MBi$Z*_P^K`&x(o}^~CA(-dQJ^=U
zY*-W`w-?Q^3QVGpc94F&1YmqwdFU#byUH=C?ch$tm>VGaW+NOBSa}&HovP*IM7$%1
zu;eLm3GY3ZMz^A<^*~hUQ0kzt=R}wb1fb>Vxk9-Jj$n;qbh!<TV@@ae7b8I`c~i#o
zhcj;!5iWoVrNj+4?`VHEr}laVh~uc4m!AfS)jPENO=-&v0xoz2$)>K7#%wLUF7a&p
z<ZaZPY<Iu(>e@KRty}&l{fQDu<Z}|+x%hWCmuz$+oT1<30SJ6ym{jlKR)x!mc6L*D
zZc*SCfjA^o08H@QKCF4;TN?oU_-CFonTH9KV2FJb`n!>JM*_(cj3rQ*>&d#o<sgN;
zAtRZ|W>Z7Ru6OeuvvyAu{JsRyoh!5YBr~^Rc_op8$PYB0&7a3qiOz-Ho!xv#5?QPV
zxU6*zqD%eU6dfs7Bg<Ph^JlD26crbwDzUmwwfccPGS?PN!%)E%8rn$cqyG5;)LS}*
zkfJ<ml?0KZ1Bd5vL3EHyUYDyyjg^Mm!@{22{wzLsvf>CH*|CtHAiqQAo`(ZJ!qCes
zJjC2_J;yj8WL;Z7F}_4Zp3aLR*&7BQ;p5Z?!(%2gt3k$JF55-yef1Gdwy}v+FgzyN
zaKbWDolRa1WNa<+HHR7_jc*Y#Tc^q<EBVkPc~y7hVzCP;2M|9)_bc6|0+H%d+}p|B
ztz#GPKFF8$m#oxcf<=97b*(Rk3i$OHFVE&l)jPy4MdKc*9o|^yUaD$6>CGDsSj|Y7
zUi~!A-%_fa{R6v#*3%6M@yF{(BVRsep}2?{3la#=(BoUD(`NIo1oL+PE$Q<m0zAa(
zYpMe&uIYKsnv>{)oxZM9YX%tvgzCHsx(cel3-AP^S0VoRMzs=8a6*KB`HsTV(5eBU
zTTZ={g4jV5h2@x=%WeBmpD!mGHpW((v?lns5-6c_pmmgag*B)z&QnunbDARH4(*<d
zG~`f%>DMQs&K-_)afT#%?f{j+M$8My)m-xJ*5M?QBN?A8o7dcZgy({nq-$^Tn&8-b
zkw0EFH13Omd2Rrc)|uzLXZL&ou2_r*aa(!iu*kJ)5?oP~7FM)Yp6g0{-Fg8?zvVY8
z0iUVz(tOIH$8bH!w=J>>@nQ(gIIXvD{qb`@jU27hJ9osf%(`Km`6<cA*C$qoL!6#h
zP{Y<v>Kyj?A5X=-*xQ9jOStNdVZ&82uf#vbY{Z%G$Sk`ivfPB`?0p)uPBFTz4$7>=
zlO_jV3hE}6-@mCmD(Y3b7RO061&KKTt+J_s(MWHORJ&mFjwKa^LO+7{30R->#qE}(
zm%md4-o+2nYS^Rb!RnBzEETtj4b(kS+h*X(@qg00Ef_pf3%bvUz{*39QDDyNsK2AV
zx1HUH9mir~fB>p_Pb#K-Bvp*^1G57by{w+*eCdc~KN5gMwfKcO=9D~su}^Y2NI&(Q
z!SfrzwA^_`VOvlub#TG269f9<#MU2%OKX2y@W+ov`EmVD1JNxX{oxKz5hlIZV-xM+
zRNY(1JuL4RuPy$MMko(Y@CY(HFVSeL+ppO?`w1Dq|3}$3nK=;Vj&d@|gASd9ew3BG
zDV!ytdD`8d3KdQ^(_gsW7PchITRKgcsl-e0I%lw8CL+zasO*##jc5gM$v*>>T;PaA
z2Td94+_0bCc&{)<vmV(oC7JF1_KZL%aLJF1WS9(6RNfKAaSa!gU`}OwJgfV#a^SAx
z80Uqd27tB5?LwbY2FHrb2chea=00u2%Ln5{KDs@PtDs+9HoX7=6IOzLwAOn6gXHD&
z{b>jYOb32~UI*==!Yd``48Is45?I7c41KcC;nVEnl<pVw<G0ets@Dr^g7lvKBK3;w
z*exHNokG8%V-Gq)bSc_~Y_*Bqh)&PaVXaenNZ?dST>UT00H-P_Q*-F>H3L<=u!pLv
zBr?MtXMQw3K}tix)(KWlZIbF=6A6!^^i?5u8=LN1U;25|JmC??8OW#OplcDajNZhp
zVtKJiSj)^762`IUf30|)2e8~zlLNz4s4CoY2CnW{PK|;S(MK)%4B0rTIE_%RqB@4U
zi5n-<a)VYbinUZaEwA=d!v<WTyZSh`k+me6_HC=pb$;p=73`F_G5*o8HaFkC>k>rz
zQYXtMpI%iuKL2iHaeySIJ@hp)NcasG=+O9}mtSA*)WyTF{y9iNmhk3zcjpHah;<Bf
zyoZvt0hWF=8}*ThIR7k6cbA;_&zic??m9Y5UovhsD}JX{d}93M3i^5F)%$N?d!B`X
z6nJ1Y3Gtv)zSo*W-BG2Kxh{_q^p>Lck9?qbC{t+FO{z0i=XhG~6%FH>S!)kBWB3r}
z19UL;>HF7e_xw6WlyZ5|UfioiqAnUU^OP>CSIq00+GOg}+1KLutY%M}N5IJWz`Ix1
zwhC>cyW?sBaXbUDM^0ap+VXF|b!+(0A?{o*JbBV6FA&6^k06Sk`*I684Qy8~!X_z0
zhK)bXWDx+Uy%WP}OU@o+veDd`+3MH+Jb6Ti0wngDl3HF@z;vy&TusYzdJH}`n);L2
zR-B&wRdLgw#9!((DE4XxRYFW8QXG4TjYJh?++3-Jcpx>`Xc=gp`3{(G_*m*@un4|5
zW>w<Gho!n3d&@Tx%?IwXChRQ%QdL>m2*_{O{4Sd!!qjk?8XlS`Qxb_zjRX2T=3Y#f
z4z@l@(V`ZAx1>hER>K&G>J>KTLfGHndS3*64Z+@YW_BQ&bxpbg!M@wA`|O+kFhloM
zw7({<NfIj6r*wM)EK<p&NE?EwD*S~+QT8#AcCow^aL!6By)@4H>}Js5L%|A8c*41v
zs!bx;L&Fjk3u;kTX^W+|FH#e(tTzIdb;h|Pr4&GaNea$V0*YQjQjR?nyf~jW>*g7I
z7<H2p7uu1{la!z4vRH{Od5o@sTiWSUT1IgM=t5?J;z*~E-sq!bsn}8*y5t_YJTsx<
z-%Z5u3%{NFZa<LdMMr_UhBG+a2;#NnF2dd+E)_%Bl8-ic6}%E&8MRG~D4GHS)baY+
z8!d=HG|P|nL<L$u*t4em<EUAX<(i+N?`H>CY)g?<NiYAjmDIYC=Gv_Mr3*y)$D1@{
z;JU3Espl|gxO*kX(8OeBgJaD+N_EDY3q;Ii{fx5sM(~QuK#%gUF`eCMC$ZjXEIIO^
z`4$BrvmN%0`>uEf(V1M2k!qFB9h>}xL9}cy$e>jql34mnQ@c3cW!F>*6b3y?oSI?U
zpte@2%x*;jv-w)k!@K}CtwO-JzA-*EyB_~H8ARS7aRftJ{ANc5oeF`E&fKYT5;aPQ
zmlDz3+z~!2geOh!39{<U*<&hcpj5a??#s+B^azzDusuw*=6YY673Re3EOL~X!XRXz
zm(f;^_0H(?UbTus$yaBXIf4GzxXxbv8WSr*>P7Y8P+i=I^*w(wbPU=y5Be{<aJU9A
z7KsUyV9m{^aI;XNHWMm@qAzBtMv?L2Aa+nb=87G}X<3lZzRrLRVQx$f+ZDcDA$>Rv
zmkOXjb$FLa?3!}#fG5Xkh%D%N7_L7`ahCaJ`<ZCH1@oK4wpp+*-(^Y_H#`UsyY*uk
zsRlZ$cR+KJM{_%<@z^~f{B7StEd%h=W6m$HT*34)E=M0Pd0sIN7PD#&^l}lGN(&^8
z-He6ibkb?hMsQ8oxx~ll1_Oi?udOE|A)g0pL{+pVebV6qrK1PvA+o@K6^h2%S<zic
zG$it_M+}EDle-2MgK3&2e@393+8$$g!0lWo*9)Hpn*_-rZR)yaEOvYD1f@essFAK4
z$sdFqqAgh;t{AV+EuVV{+Za7qbS>uZZ6840qHwpHt?60eOmO&mMR-I}c4!Nn>ki*l
zW<#VW(uVAwBE=gRHH>@r!o&HvnkUGUU`202T_vOp_BQQ^iy09DkOX>NsqQ^sT-B>P
z($bHb;R(nZsDr<>!4c_T4=zDJqQVv*mQmy>-=m70#1CcV38ADYfnw)9M@O!eX4?{R
zO{>i&@7n28H=4D#qJ~?^UHq_)C0XneVL9$HOr!9};q2`$1eT%}l(~Rfy9>ZswD*HB
zAYIt50sy(<m$G3Y6a<$U4Z-|q*+CcecFOk0zshw)$5ho-Mcir4?CNaBNt6^Az$sfz
zw@p}|kP6cz>ae4RDGWW9dJNP&zDINR>Gh#QVWRIJdxxplt&xL0w%hPpdXBdER{VVk
z(>fafcw5BBj<FPuOK2wK{rKrl)-w>X>Ryd0zJ04HW2GSQ(t0~=n_WX37)aZtA47-0
zki&#bpdIO;BI%w2Y6zi>T{XnL);dX)&m2RYp>%icO%M9+wN~!LvG_F7{qyC$kL?91
zXd_#&mXf-DwZ)?K7wYSB5MPjlES8F<Ac2QH3*Wx#)3SHJ{vpWCN`UA$3ZMG^hK{ae
z>SOInrSYz|Om@yMbltx=Amj)J)WX4l+8|^3hogqvfJUWD=2#z)Bbm$9NNoP;OKO;+
zn3im&c1hFBjct=K+_D_`cW|a0%bhx_Ts6c-<g~r%{%n_cALU{|ox!0KFiWPOu^-XZ
zezes>fd#R9%k6b4Z1s65%<<BU1sMSbH^E}3LvOdod09_q@A~tEP8iPKrvNU}9jm#S
zNM)@uV7PNrRnGxXnOXgtXZ%Z($P|pmP50VI{M?qtVLX0kM#jT|8Y45b`2gl(RI+!9
z(V~`OYjH&QQIZg35PhgpYs#F}N%iJ&@++ovz-?z?7=5$AN?vKmNSsYnr$6pKh0-8m
zm(I_A-+3+^Kk%K79yDwc=|iw3!?#ZnS~|G=jE-bAwZm<669CuQexIT(B!SV@E%irA
zIt&y&gjX$<dIix^87`h7-E-Z@8&KOB6$x>J#^btb5MCH?Va{GD&V5RAWyCJ*9#k{M
zJjYsPZg0%;h#Uw5Lb>6Yz`?foqn)%Z!7sJ!g|!CCh4?KguFBQIVho#x8H!h6bJ3%F
zrgyDW!6uKV#p2>W{po}$RGR=`KqtP{!1n8(SpXN6D?_+usr&JW(;$B*rPm~-*V@G&
zZb?Sim(3FP=O*iWi!pTjnISgZA4ePyDlop?gQ6Qz@A+3K619<#FSsWi&VDapX#w7m
z&30EshDkX0@Hno(01u##I;C}i=1`l=U3M_737~RBI@;tC7}k^^QX>wUlFjkL;<(Lx
zFm_;bhy46mAtbo>gTvQz><7dvxos_V9||ntXas#rCbaS90%+}I)mzyf3M#~;n_$Ds
z=c>ob69dsbdRH0#!AVGA5&#^!rgybS&BffRd+7(p;ep?^7QIdfUTAoSUViiTR8Kzc
z;x&X(?v7xhFzMFSlfIZxTpq<a)&N>A68w`-t`eF0k*h;pOq1q_^Jk@n{qMR`GXNB#
z<ISjY528ga@f0`nw)_$NI~vFhsHVH<!};}(ev6fHH6FBJKE&#4$j<J2sTiJrdow^4
zD;pM;H^+P2u3!7414#u#vplsTbi5#RdFrIR&(X_B!RI~7)D4rbEb6UHY>J>f@|pa&
zi3=Zvh1Ztq7DRQV8MQ*Sv9-#s;>plP1=B*Dp3^b4Z-Iv%Zo@Wi-WIQVi&M%Zf5PIa
z+R5ZEDqwLqvU+mSVcJT3lLPj%dSWwx=xGID)T3egj2Bg{jD~IM$-;2$qh)o`n1eus
zKVg%Ja_vUC_}FTQ+Ft1BS<?DaIVK4z0Q|P~-F_X|<gfjT=}oIJ8M>MF+5E)0SPkgm
z(;&LCeYfF?Ew<`AAy#e4boavE0zsG|q0-GB)k3Q$%>(b-eTh$VJ{g3r`C2-n&^6Qw
z;g;GgwD$b@Eofl%*AAM@AGBLKL@QSh+-)joXAKDRbUlz7j{l3D<IXyk4#xd?x2CAy
z#Nd#ycWC>Yo#Wue#y_?L7V=ZpJ9;?gw%^d0*e(}?pI7Aqu^1FZY;&Sv%?0i0a!QFz
zzBiiomJCv9*A*-sSZ*|qUZG)l4XWthk+21S_M#5Izp)x4O5uKYFlWnisw0PLdtCIG
z^<8i^QirlmG&BVmz!x{N1XA!H{#A^Lv9|?5YyixR1R*<d^ExZ?Ipr4EdXC|g%%;06
ze~3H?A%T@_VtO2;JsP8>ZLeNlL|-=j1&+PSnRoY-q{PmE!sO@9xaPesi-r=4DYh*z
zkoD^I#(El8T1aBm@6Ddwf!$JcNj=x*G@%2C(~Xai7hvV_Rl1e<+!wlpZ}*_bz$~$Y
z9-K4+9$c)(N-yTY4nG@VJ8*K35Ft~PQ|k5lAXY}9cZ;1gQD7a%PAr{D?pZab_PmSg
z-;zcMf&<9ylh)c-3WCb-LT3w4Yj7R(`o`B`7bc-%)eI(*2lVH-Q})t63bH=x;yG!t
z*|Je(x~N(*sE<Hve_{R^+jX0-S0lT@PZr~akn{v%GM%GHC9z@lQdjrUMGlx=5xDnv
zpAAbmP+*#DC1j*F44-E-Ks8<O_Ec?!hOBf)AxUr#%nVR-8*8Jk`Qw8v?qjCf5K!EC
zI(pn*%)Ca@8j1V{KrFviI9<cbSVn2O5cLcUm)xEkZX-`I$|x+qR`wl2!xe=bh~ONb
z^XxZ~y;IHc-qf1oFBXI3hE_3WtTk+=jpTC4>a&v{Q3Re{p>?F=taY^Ky=+BR*7WG6
zJ;dFcdVK;+^x3nAl3fPRqG0=sB^L!kq>sbWubVP);G7?xZH5Sp(=GU=G|#Tl>LXdN
zq@aypol7SolIKP<02mczT#Y{8da?JG!x-IndeY$z`^)CJ6&wm`NzRZ+lSexqU8*zH
z=iqZD752yf6?9y27$;jnX)D}x)d_)UB$R>;ocdV)jpZ|Ns<O;xcxJP?08Z>&!N@=p
zFu`HXtpUGA2KMeVjk<T&QIZP_v^HyvN6e`IDd-OUy4BoRVzK{}bhck7SQ5!>HaKAw
zy1hM-_rk4z{WDry&`=bGlRVg&=B^faIJnd{b>z-*NG%O;ms02CcQs_Pd5y)N@7>`Z
z`~HSQT?&<)9SCNp$%dC?$RrHVcoNSz9+@u&EzSy~&(*bWo)E!F3&~C*1seB>{CCd1
zBGXne&8;jEmiKh4tvmPkcEGF@Mc;8TX|U!sHLq*X-~rJA+l85%l=P-mlI067V33fW
z&5vv^LYzU-BINvh)0V^&bK3OovTNMI4W5r-x5`tDwf@b^fX_0##;C#;o`L1HpsO<(
zbf`9v6@{-(HquD~EKj!P#+VX6KTd&Z_1RL*!TS$_-1c%m$+R_aYx1CxtjjAJ6Y`H?
z(1exm6aDrj=ti~Kl$LVUC>z)}1)gY$3L{hxdbU&mt)MDMH{K)i8Y;|_^Wk7+&g=G*
zFOWY;z{q5E0J@%D>dj-^|KnBYFqon+j&Yv(96ML-Bs8XTMne?@Tg$5Z?CFhpqC4M3
z^H*<&+<q5)&>NRK;@)8e>#HJ-$FHkFbZ9{jv75DUHSG|qQV6SY?Z(%?7LEk9UzKZ8
zB&$thM21(g%wx<VuB|pPw#);P{^)v{W7R{};eojTdIz9Nna%WWqEfj)sf+(nf3+lP
zNTfCP<E1X<w{sk@w^h`rXMgpbO$fmL-Sg^`A<{phwzrcEGSrj&z@_WBcRpzeGfy{R
z;I4d2c=rbNk$iMsX4E&mOQOvC#aU{UFx~ps>D=<QoZ{78F*wWlG-w?j`NZ85E9ySy
zVvKy;`;G6l3nzc1KsSM>>oigCjNl56Wo!v*Fo<yO6W&v2>X6|`fXSJ>2CAwDA?({p
zvHYw*M?Y;wh|u83Ux~=g2TUbnAu9!5%e*^T8OKozI4rFiA^5chU~`eM$&=pXr}qXD
zX?59ZAM^}}RhJw*!yz;Po?F6A%DVHxnJ*0--z$Ijyc_`Y%icw<N`uEq0O;Ua5RJw>
z0$GgDbWZOY*{Etd>&fXZ_w$<KB=||05)agN9f-oWt^fry04lXxLpzxavuU}-8BTb6
z2S^qZO0vs+jA@}{^T%4b#b*T5Y2=V2E-4h1;|Xqwc6qcsJS6z3YF^>#y}fRlEZur6
zz==!W{2YZNL6w<X@#rTzP35{TTGa2~I~?HW>Qm5EETyXvCZa#T(pYgXF}<VV0xfv$
z5aZh_1*18PUIMnet$(l=K&lm!*aG_f>@4C690?wvb#pRa`xWLI8NvBB<)6M*nXpZ4
zuO5v}PiXQ!fZ7Gtw9>6rNJlaHBiba8)~oE}bS}=1Ay=SQ=sveQJqv*?+Scc3U~j50
z1My4hYTT6)@GfmHeHsEB1qI$<s1KN@pI_4*CWtpK<*E@Ov(W(fLy=sK%lSP^L5^<Q
z0U*1Y^YUY)3m70eaTi|S%kgngtZ~af^iTV!PBAG`YrD=zwGe@uQadB0a?neRDXqxZ
zmktE>S#-r@J}EO!lTj3uqvoky3H@{|t@ogHa=;t?&Jz1*@~eKl_UNAb2mEs7V48q}
zfcHnXlqFC@P$irTej<OW7!5$MF7$5^{rcITO~AFIu*Vg^X>)_VfH(&zu<}W?b1dti
zZam@wVY~sEbPIx5ZmY52v@XLA_%)dUMsdJt-%B!kz$l_h5H5|{+hg^ANhlv}V!mfX
z>tun>EKG7Q*23O88s5Fg`6nqtl7r2OpU461cda5x(<-Qi>}d^G;5XW&Z(bm6Fi5bA
z)GL^uhqryo6nnLYZ~--Go7TF&X06GNOe(`cEvt$&F2hE!x4C6nG5&vJ_<4}K#Z<+5
z-P{q=G7(h0dzn1U5agJnRC;6U03L}_n4E_X*Jnwl|G?b~+dzbV{KO+*hMN#wBO&Zk
z<@4nMbcxnV`UUpCq@qYxXUf0mCb&&es`cv*`M-v}i33_VmhfKx(3ss_7N$xtq0v{Z
z|1x@SWd-0RwQj(=wJbmhC<Q3hckhsCCXLi|VouCZ{c}5P8nh_djm~LCtvntj7O_7m
zrlX=cNHIY|$m)-F3%7^vCD<S#jjW?zON_uFLu?Reo$^p4B|S^1>w**a@Z89SQj5LX
z5GMmUu1CMiUnLk=cdYsgJ?(<l;dbP+6RRaZO^nyc%Z#qc*My8I@^j;u>&rz32FWsQ
zl)N8l8hi~rtTTmS>2%pqn#zIH5chB2oIN2be=sbDObo(KP}&6-(2o^n-G!2YnHcXd
zDj5qXfMC>KNG_n-peMp`Bv`2rH3Fx<F!`j|lLP?$Bm9p#y<bebRk~Zq^tn}>QNgcg
z0PzbUQG#_fF`Qjs^wI9n%&YSUA<Dpf^dX@fwXg<r>FLkH-z3&0NcSd^!LNzBi?5Dx
zrmQB^+SWzJ96)(e+6!*C1<B51d^v`Fw~P!f?KxMReF)PxETCe$`xn3Z#=CqBxGiuK
z{8l`{GT$<xBqf&KD=g%W_1Ww-pefjbhYQ|1&eF8cSFXfBQedb9C<r+LSq88DT_*86
zLzs&Ij^R0pX3Fd|<I9OKFatoYtp>EqR$ZLs(AbZF7IYZaRx<6v86XB8_D2pW7t~dH
z2Uj<gRBn?&pqm2t9zuM(8@#?{xw7QRU#~QB`jMU5<PT66RoTV)T;xTa!$JN(!|zC@
zJWdUx!daFQ$0LX3wwG-i<#__Cw=R=YGd#zPS(mx@m&FF|E}^4X6c4EWRRbk~;0e2$
zmlw~cNf|9(uHn>B1h%ClHNAc$yjV^8lI>#Wv5}=wFcy>a07B}eVv_CRH}B8W`{}c?
z>3c9P={?v8tCr}YE&Jn{$ih^HK21h!FRx6fcvRjgtvw@j5d_WKOC6oe=Mm-!$6@ms
zbI;C#rXtdUM2dO3OR5%}H@iiyW|{*0FuxSXSl@FrhukR{yT5nznsOOFF=&L%>B(v@
z>fm+N+aU8Vz|~pRp5Av`%v&U-&VsP<admG^%8&h#*~}Mp)!9ogYSA1d2O(y(zl=-P
zTYrsJ2coyagsZ>)mE|G&5_VZ}50G_#Iy9eh@E{}$OdjZ8cm<J#H5jC`(un~FK6fol
z#TM8<Xw%xd^SRy!rsX`XTW9Nem4>ZV7o~3E?D5dc8tV&&gFX1O8F1Eu%Xg4JNTezp
zv8YKJ^B`={J1XGxY;eoUk&!Gf#g@d{sRE2$b!E02qWmA=CM#)GhuRvjCT?picg=4k
zP;Ps69|?X;ayj!l`9VI*qo)t9w9R&D_GZ>P&O2r)9KZh}N0XOM@$u=_>+)I1X0-nO
zOBz(CKfEs*d)=jv(7XITC;XZ7cRJZCqFiqtK7ojRbuFfOm~;Ki=daltUB#ZIxrV3S
zRjpOd4~lr5Ub@qzf-QWvZM1fgKS~lxww#Aa7uaaKNy=o8>CA9(#TEtseXYav3Di<`
zK*)T{zuWmm75Q<8BPbTo)?Oj&xll>-W+LA6ijO#}g01_!l>-e<5GkXX?iOoHjXp<{
zaj{M4Bm|LDHXC0!T6}iof%)I4xa^JTR?1j17mV8osZ!uQeGoFpY05r!=yf<`rH@2J
z#1Jd_`KA614Q_DcdG#R=vma&j2*C)jk@#NKw0){qKxNUVXr=A%lH5=|W!PwPYJuCp
z^9iOko$G+9bl`=(npLMNsoyB8N${OyF;LnWHm>MX#$A?aHA$mE<GaG@;=v#{T#dKI
zmA}x|3tk1hkvRHlYyC~+U*%p-#N1QbN&e9rxXWSvt=sF<%P5Q&*qL2iXmv%E-PsLS
zH2%K%ngXz9sswGE{_?KSOZ1FxdDsCI>cskTqru%{4nEkkpR-Xu8h_xT)?28UzHlp#
zpM6VrjR%KgGGPq3XD?7ES4`=1UQGO7d)FD(bhh+0h7~Lzc(H&2t^_tB1VNNetQfjM
zK@!?tL=X^^D&6R+YehpSQbJT(L_m;834!dYumR~sP$0N~fly5h5L({z53tX3@BRLM
zk}vQie@@PsGxIyYnKLt~Mj#4z_(_ZdgVk<<4W$+PjVr3po#6oH;T|C2e~u9{w}(1u
z$*W{U*Nlo)4?@<tV-!{8#)pOh(-OL(d^0&q%r)i>uvI3>kJteZ&@MWn4vg#Pl%r5#
ziMxda_g-DGz&GdM^?L9zQ!a!Dd8PX`k=xh(lk4hz8*SZ~N*^9RJMR4)*pvuVi4S@F
zN`Z4*CA_jbOE;gn=n2iR2sV@1DfAT1iHwSB4WYgo16#XFzIJBVjxXGt)X+0Ew_TNG
z_PKsOV(5GKy$gQizjqa=quu2IZC6{{@wj1I?)J)I$JTt`__?hOHv4vi1u3Y%Leh+?
z_gC}y4=mN7%yi;x71ngq>WZT^nir=zT!+m8qOZyy6rRvt2jC@bt})}F%~`AS2qySb
zBB(9jG4rd`Iw=Y`KH$?X{Q)|R*OdFtMbT_!0n~RWLcM?9LeTihdIcc}cVu(%hBPhx
zZ#^FV$;te*qu0>(fUTw&?krbxp*&0rJw9R9AB`~yX+;GA_<+yhC{?umc+uUe0u^I$
zi=zJON=G6hi*%0|{8K^YpR2>h$mR-mzirD#rG<39t?ouW9#TcSi@?Cs?8bOa*<4Go
z>w2awpNQcvj$C8HH<I>iCB4KS_#?*<V$vqEU6q0LuPf36PY^T4SU0u)*`+Q^LPQuf
zJ?VpKSk3ww-dfaNG(`e{ujY<t8wR%L-s5DX5)%f@SW;jj-uKW9RMx`D^7FlJC0UP^
z;F6oNnH+Z1{M*opu__b`wK@A*=jw!Oq+U)Ywy;euekl_5m7Ryls%TTexbbzPQ!#R2
z+{H-}&kG|xnbT7YslFWi5Tcj8zrLX7pp=&p5zN{CXGjukWY$ZYz$p^^vY1ssILSPW
z<r57!O40J*ZtEv|DEiDP5X5cCy%)Q9g*7~c6e>T0po0RNY-;TFBIzkkC^s+p0POuY
zL6Kc1m~Hefu;OAg3#hBne99eItdU_9Gk_d5ZQI-S?y`=oO0y!cs6KsqRuh~{l$bol
z4VZEAy8<b!DFN$>*K#t_sXAs17!2XUMrWhI@hREcCSOCSPQ=W%v69x+V8xtkKiik;
z)#RejXGF7)gZ-}21R+HKqw7n~Qnb=bwL_XdSImO@c0%YzlAmpzWj(_8a0xv(6o_$_
z%&bYklU*&!br5lXFK2Z@>q5eO!xARKZE;sWxWorm9Fy8I4CVS(DL!E`{sl?)s0C*n
zWAY|xrUEK?GWl1TX?o>L=gJ9@V9Lf3%#0!n%cvatdMsz$a4OCciSjwFvS-yO#z7Ys
zJlE0&r_Uc4&xP!ED^$2}n%H<TA8iu0^mu1uhMLeUUy};Xts6)?ZfZSZ)XgAyhz5LB
z<_!E$LHAr`dqg-!dgz}Szp91_`bfNr0KfUxtfpPzUndq#-Uk6Fb_DE(@w&tBx&*gX
zSe=pb(RQlnAvorlXC%ig)T`~aT*HKCoF{A`5I^nCe_?bXxPDk3LvDE&1$GEfF(6L`
zO}oIMse+cE3!0|W8V?$gyMXFIZ9B3Sv?dP&fc+zgSlNM}6?`5XM4<btC3_yVEciQ>
zPpFBLCZ^aDhw?uKhFUl!o>x!?9|URG8iF#)j@Z;OI25#Pcl2(k7S=c!kcS&m!f2S8
zik*4mRqqM*w@y`lKX{x{4X;Z982fCgZa)+(i9lT*NI@y<p1G}6$Z8TSg&tl#w&srZ
zD`pLLsQagW>(s$kL_a<UvaYj%2jNp^u7a-K(2n5OR!!D<E~dEo1J$<ZCzApl%FmDA
zBKYL*hja(|>WbhJ3dOpsDzlSwT4^-}t>gp+Z_#IZrvB#ZZ4I+0@fIV%*9rj87%?!G
zn7E;xZ);|d-k3@>&d-wX0N$%eVoAp+th)GR_hoaDV+$ll#`YcaeOTK4s#-@&L0{j4
zWkH|^;7di>OVd3kbj?1CjktL`FyiO~YI{<(?2<lI>$6(3Ii$EXgtc39oz5i|(u0|}
zJjm<A$d+AfNYtg%UjD**8XrgiTDrVpo8@y^E(oWCUM(%7iVCMeL(sY7Kn-kvLR{cW
z084AaaJw8Wce58t<m6PE0|{}Sxs@-ROF}*BBVT~d$OQ6+sEF=MaSDbilQz72CX6o6
z)jgxiUR7anU2box{(`CI=gw!VUB^pp@rH^SQnRV`q>b&mO~xuRjG0za<-hCRN&-Ej
zY^coMkDsYHqFHzt+POWK0XJ~P#_M6EN8d+Jtf-iz6@j__B%QuJJ*w{tKD*!F5zGwA
zOG%V=$fwE!lM~@AoYLDpat_`345;%i!W=8O?pffs3foOKUJ!5A4uP>3eW+oo1|eK*
z*=i4M?LT0886%-}8n!t7f#+MJiU)x9ZP|!ahQrk`6fNUR5z-2-Pq=wmYInMd;F1fs
z(fND9Ir$L=p*Nv#V^2v0=({Dz$HZjMl+stKlT@^S1tr@&Zue)e1<X!luar^85eQ6B
zVBbu4Nm><ylYBBMzW7gdq}t9;-{3!;rVJ{#YTm2o(L@J-7lZ&aM%%-=r^fwE$KE7$
z;%fSI1hUpF#LD$N+8qU-D9m!6W;U<eZXy^PD4l#i%79v?%>!Vgv4fRVUj@KISPPz*
zZIhqvT_J!}-lv_?p5@s+mk<S+7sUXU>JT;bXw!J5HtU&cTS~muY6c`TcI%G$$RWC6
zdW)Iyx3qz(I*6e$fN#aUS!^U3<w((bGE+hd#Q2ccLRlK~>UL(2{JkzZ6-|F#jO??A
zR6)BN$a)Fu;kL)j?AxJ{`0??T5^VPj2~#^?&h>JRlHg`^Oe@rlX$L5WZP%YNws4P?
z;QlS-WDbG}&CG0S0>_m%Oz@_?@usKnlg3Z^<yGM6VMl`E#~#yCdwMK!5V`p22(VJ|
ziq686CsjFVWy8a^v=<3FfIU6{TF1U`(fmfL{RIEQdHsM>NR&VRaJ1x>WOqW!*>cfe
zY*au*+e0bS?kmt3&Fcu(zHcg3X!jY>{3|A9jq>7=A?f|AtbjCE(Y?_t1`frqvY|d4
z`vkq#Dj4xj^2{+dFo3lL<t?-Hb{@AlYTWFxSVNY!ca=oc=V9x8v4K}brD8jWR<VvS
zimliiv`s0<TKPPvY9p)#0#>-TC$CEx?A&#J)uPEyuBq)e=Jytn;F4A;=!V3A$vijv
z#KOp#%-%2V(4v#&2)3@F{HJYU$|b6-JmM}L#2)9B@I#3+#WmG1_)A()+}!#4d%$eb
z84&01X1=>rie^lGjNL~Mi_0YGe(pU3@&sc_(BaYKGT+R~%Wr=Xmm7N*GPC8~qORVO
zj=DrlDlN?d5djE3&;b%$P-^LzQF8ffZDeHOXYlH)Fre6xoWZm3h<yN&B5_#P^g)AV
zT(HVw&bMk7YG`!!87J)N5r-QX^0Q-EVM>g=iI_{&o%2=CjxiX$msZB#2oKG<a11&0
zH$*Z0k0{Fn(o+O>^4{${oM>Zv7T?CUbm54F&9LZ@0${y+;RP(a@aS^yaoTUt1l=!7
z9CC@M9cb-cyp?Gjds+VA3PMjG=}*Okd)=z9Gw@OlnKw1&WI;020T_aO9(I7h08DFp
zS~?bG#rvHRTc-hFs|3yRB08VZaw;uoF}MjO)B@=9a}F8Bq_>u5sx?iFyw_Suvz<H9
zuR#?QyU;^L-{KQZFW{?xON7$PJRp;omi&GPz1zS8H^IJggqX!%Qk^D5%i)0s8DmEo
zd>CAh01AHi?J^-Wv*@qU!^alHSLo>3X0C?IPw{Tziv1@3e9k@rWsWa`)3u{8K0eZ2
zf-B51(MkV;z%=eW2Y)a1J%{-wbAOX2JuzXk;>r2f*V`evHv1B+^WK=6rAH9npD7&b
zUG8t6^0QqbRnPY;*@(~0RwY6M&x4RQtvZmrSNlCIWaRn)(JsyB#+o$>WX$m+XVNlo
zZl>=~JKI78J=ciXc-IQ0QTLbp=U>H4fUqdrRTfy^R&QrurYEX5av6)Gcr7c9o7C7o
zs-s%^(Ch6DT|JJhn;_qU83#OT-j^ekCJC+tUpb>CEtQxw9WMV(3iyYPJ}&7`Th0#s
zy?Lf2Kg|?cgSH+lRf{oj#WOhgil#81is^S)Hhp(Fi8on;yq0vQD?H+JkhT=L3xYf(
z7(PWVG)XS`DxqTbdj@7?zZI|ej#=*=xa8Ok=0jCq9mC4(mr;|w^HXU_Vtzni0C&$b
zZ1KF;Ou=hQ-d$DmF~JSjmDX?vttqL4Z}kMcb#G$%g%=)t_|#o;jR&dRGspw7Xf=Q<
z{A}0cb~(y0{4BP~=alLJ$xU~e(<ELE>C557du+bo2r^$BAhpV8Zq5U78hzn7(QsNO
zzNW9WTV@5vWZ*u$_gKCcjbxTe_paF;y~XJ`c<VlBvgFhAo3u)^6<Ey)^Q>UECnUBd
z_m2iEq&&evvgb+4;T)qVCnsTF0PWdhIq|M1!4w?>!oioonX%&Mimx?8FQO@nuo?BC
zCd<+)kLhsp*q4hTP*DYQ26Rsn{tg}YhP-FTg4||al1u_PfoJMRUw}Tyunfxvd>t<&
zwx`~*cR3$H$*MoQ18+-3fY5TA(ikmfrQB?ofMyBw$``J!%VETJI?0GG$oK~X5(`%V
zbyyvL6-l0O4+IXV5M|v^A8&jm(Esv2fPT5c+wF6m#s6MRSl8?zZl*fyL1-%Ii+JbF
z4NwceXiaXeJ$roFmKT4yjJNJ<Lkf5Rk^5TXB+f&`_c&O9bxAH?@;d>|aQ`EyZJhjs
z<Tiez>}NUeTWD6r{jZli-&e$XpHxALm6F-F(GM3gZ|^%ob|QvjvL_egU03e4$DT9i
zuv{Z?Xd^nR5pu7aq2P6A+_@t#GaNVp&3wn)KE%v$j-8WLnPS7M$T;?69NX2wS%?rT
zE1JTVFr81ND;G!BK5UCOoc3boOh+mU#S{#^7j98tTSvbwW9-c+9rS&799&;fo&u=Y
zy!-w+cmxu#+vLRg&;<tz*TnPwsL9angj*XH2D;?-)qkwcQ>8uK`F=T1Ye23mcw~2+
zM#%RMIKCq0;jWLrDlK1Fty8xvCE@yB(<fp$pdPc-v!AO0st*Z^;GFp+b1-g;_iG@f
zCFDq_E>UOt&K;Xn&`h#_JzVJ&Zz7mC0THL>Wk9;2`s!jwCr4f2i41R45p*sWqK>7a
z8KtVRE|!d_#y>p@x=Xzn;#@;;DFiwqPaTus8h9BWTt9yG3f3)+r0(%^Gmd~WBV+qg
zN6fzU>F{!9Xm4`<EmlZCXH03l#=W<V=x%mCUvmaWmI$q`lyLFYM2WNen8wq<vxeo|
z(a-+EVO?`_Ieq9RRNlbM1SgucLjDb@5_5{H_9T*;E<p_;lm=aFa$aA_a-lk<esaNs
zEg4x`YPYPM3w8us6n*^kvA2SoNKaQ{j{CA4*X_0Ap=gPD6|@Bbi=#xl=pF&pwGl8N
zvY2w-y>_d2D%!<(QUW>jT@~dGI0O#WY|UHBe2J>>TnZWeyl)vJ)a(FuyP0O&->zA@
z(%O}=&davH<>B&2GyoN+<LWv*%N9D)UdL51c0SF7&Qm%7k578Gmn6Si0lqo@`!4+`
zoR5iS*22*ytx#D)w0y464l6P?m|gS~(kFU9)-qN?P_r>0;Nu_b#~Qv)`e0c%uw1N8
zFqNOhS(qmKm+$LcXL^vcU>=ia8~Hl~h*ba(xv8h6EGXrGV_Y=Pkv2oI_*mZkuWAW?
z*Wd-#n#!jtlz*xJ$TmPT&Ms%fnaFpQaXaHJMk-aPP1bciWY>x1!r51GFoe*#v@_2V
z>Dc5H{5@oTwoDvwW9oy5NR8AmXl+sD2K?(VD<6)d{b*tK;FBoH?{@u@_Vs7kOXs`X
z5{4YFM&4BVlfInya$m}?>U*Mjd&?JlIwB`A#Vc#@GGK8KzTTCXz%-a!@V-24X_0Bb
zMU6RT>&J~koF3{ku+AiANXn`JZ6Yc$zu-OVh#f6d9V*92C-*3(o4UhfR0TJvL)=Z{
zRTa*aU&w1m3$*|i0aEc2y_=feaQjlYMMb*3FN=yHC%<ExDH<L41AJ(+0%1MT+08;Y
z-$IgUa+w{rMr6IoPX8JF<8aX0*geDiwRdePZgB2j7uSES`ohT`Od=Yqvkoj0T^dhH
z^$>|No@4VzdyD69c)q>Wax!b{O0%=sr?3oMIOj2FUyDk>)+E|D*#xRC7g}q_5F!G6
z%J;BBvc6V~te=>z9cVI*shw>Kfk6|tx&>(R`CP;JHH7u@Al8&l!jTisk8x15+h5CA
z97;yz?zCo^6XzapN{~sq*V%h8c?(V)v?GK{`Bemu+WX`;Y0*q|k+#-4442I`F+Ih|
z?TGZcKZ+eCOSF&iFD%uo_70qbq>~H>aEC#MfT-@gni!6D8^xMF7jW!(LF{PDq-dje
zd3aU3B?B5LW^s<cNH0Y0RSbQTo3>IS$dr*3M`Yeg8%?u%aW+V#89rX*L_Bt6nJTnL
zArT+dhbn>%KQ<X~T0|D&gzq^h8WHQIZb0bV(EeCx{*{nMO!3O>^vWc~<M9qf2MD&~
z!uk^<u1UoM9z%auFBoovCM#xrHC3W^>)V%};C(nG|AwYSqL*rTSN~n;LYhF39ai(S
zr}mE05g(IUT$Iy;mTcK3+T-#N4Os(jVm31$omFj~gbC7gl~P8GG7?+nN=QYe!6RfM
z=}XL{9z|GN1^VtI1zV|tt(cix#eQ$!tfkqle3wnvMnVeCA6FFOa7$;p!#;;J=oN1q
zh*A=WeJY@645JZ08)~X=ae1O&RW;s}UvffsBEauFq-kjI?R?v?18Tv5ecA7M#fh}b
z;tvLo#@qAWsj6F_kPb6VMhk<nA_P^G=FiEfze`h89+RqGy<@wo4TXJ84bvlzp48Pq
zs0aJl<RWZO1^#uzWK`V0nvw&XK?r-gTB-44PUr-k3gNq;hx0zR_GRKbRRs_d?)d$H
zB#2mEmEdaFn*7ohr;qfU>3Zg4gXO3W9nT=`C>(P&6Xmz}kZih1{$cjF0zIS^7%elg
zpM1?wD`|B#NbTu_Gr%)5d>w40KIVi)y!Ht^ij7fY9T_n08-O-xYCs}1X>eojpf20V
zcu3dN@leLAt2LU4tUs4I^a@qp;!Q&{`oGs^CpuA|B;d%8h_&jGF*X4q-!K}E6PDa+
zwIM!I5QJHDK53MJgxrdYj)tohuG%()REHShH`=DA7wL1qVe7wPOte@Yl>)Z@uO=X_
zyk<kYCZ^m#?{E?^l}_t(zI7{r9Mt$+6td>P8R*<3M1!ooPv+Vvbu?u+Ai$L##JR3a
zej9Zt$9~iU8|YWe;T-$a5?GCo(q$#Z2U@l=5VWWRwE6d&5>O<|PId)5Ct@|<tXZ3&
z=m46{{pB5PHQw$)<aHI9$s~%L9TIBw_7~_wP;!%*WdE8*iooZ4S`v*ca(Zw&5u<lD
z#tEZPurhIp=qD)0q_5lO<TYl@hW3p%^Q`;Q4OZ&9tqu1xcO}>;C8U@mgjChIQ-QNF
zN@g8#i+Ezu47W-H5tf~56dw_3*q!44=7M*yW%<>&Cic5q7sd7hR;niH=NeLbA7>tC
ztsY0EHNMO;-cM6<ts|RlR3ZmU+j6@^%JclE%HVnL7feTD&mtpZh}=Hm*f!D=@6Vtc
zk?7Jr1{EQle|yxra+X6A$p<0Scpjp?7ItY{7vCgLT7T?|fwp-z{-IS)&~YD=pGTC)
zq_@Ofi7CgxB>~cNEY(+;m~z9urf=UX8_PNZrY7bgBQzk!&$3vTY81~y`%Kufx(?r7
zbnsbpz)o1Rrd2-^D>O$I@D{?86}<tO_SK2$LR{!2I=ylvAZ)C`)AgS$X%GX+i1?ho
zUMAM}`;-n&rpZp)w&4p>U?fU4E>UZ<WR;{emgr1Rw;<ZEqa`?Ay4%MhH%?@=P4<wH
zS?$vP5wNCpNV`evx-K?<dcmbo_46p%?c=48rdSP8ABGVIe)nFYC5~r>g5m~x>4VOv
z<KX(O`Hhp8Nz~YG?S<vg)dN7deBSK<3~wbN?|=SSjluqt<M2NZHcJAzp1gkq+%ZFl
z@jvea8W9Oj-A_jNm9L=_pK$;6;E##?x9EOI<cCClSj~Uvzz-bxZ;1UNkslKIar*wJ
z4*cLFz*+kbD)vJnKP0l8N%4clulgPTZ+^|Jh@8IWRq`#5xi#PibHe6$!BO|?{|Dl<
Bsw4ma

literal 0
HcmV?d00001

diff --git a/docs/logos/vllm.png b/docs/logos/vllm.png
new file mode 100644
index 0000000000000000000000000000000000000000..1ead9972879c29e17160f1d3ead150482467b1e8
GIT binary patch
literal 89976
zcmeFaWn5Hi_dg6HD4_y^q9QP&gpx}4C?e9`Eh^I8ISL|@2GU_rA|c(3prTSrcS%YO
z-SAv{XwUhb`~UKJ;kiHOb8q(SeZ}haUF+J%S58Lk7{OTrEG(>J;<s)nU||vFV__ZA
z!aEG^)UozI!NS6(Hxdz%GZ7IJu`so;R<zVNxO?r6#qGNaV%M+m^73F|F+4TU(^I&`
z#PC)Bgr45l8pcZm)(#4fANwoldDNo5z(2uj;a}m=VOm<fr-^$juyEMUo^&nxg2H>v
zI`*Kt{6!^UzJh|^bL1Q~yjF&V;j+G-UNu%`NR=c3Vw?)DdGQiOwGP&8a%@VxI~Ld<
z8qSZC>Twb;P-a8~!wuo9hk|QDaqh7r$NMFHGH?;5NhNUjNZ7O37w@oUN*=w6kN2kY
zb%*0gC@@MOvTfY2-Ic%ReAD;A1-7>eCk1Ao&D)asANBV?a~jciTz;LG_*{(tZA0SQ
zc%H<O7e1nT%U&RngGAS>--aUfk;e}m+Enrii#UJip{LNHic;Qh@uGL-3{Ho8+Mnn@
zgmvCkLN!)GxCc<~#$7e>d(zTam%%k277Y6Y)**0(4gQDaV}yl+xyHhx1OJbO1&f32
z+#$@5JG66sNDKPWSYt>HJc4Vaq-LuoEyZVGVaB2lQJ=-p%o2J4OTdv2T$<gr)rUKp
znVQ@1ISSHXp5O!5&}~*4IOY*s6G0j^X*sxvh4o!H7YiE;8;uYF91a(-zH^UH;l|DF
z*TG+cH1}<7E%{hk9UL539N1YbtPNQKt9e=3uCiXe$_$=hwsA7I)pukzx4E!2$j&%7
z?%Ehw8(G>KS(wA2arJLo*x3rw&_EOI{<F1CTcdk>Gnw0LZwqXY75ats3JV+SzrA+X
z(dhqp4f<v4HO#)YrW1e$<CC>Ex(gNzElcQ%0A_*%_x6@20FA;YXXJR-RPBb5*<EuR
z=pA-8HZ}p)fBorSUsd0Gm51xVoBMv*ep7%IBGf*@ZE*#27qCT$K!A0Z#X<x<eZLH`
zun<_{H?Ao;VvlzbW`{Jn!O=FyUz2=*NxyG@b>(BD|1mCcvFv-I?*ghc*p=kHE=F@j
zzioIsWtAh%ZkCwN;dbY$Swn(&Km?a$)nKV3&IPg?@`p)~^?EOED6w6W>xKv6Eq1mI
zKQfV@8gm`)oE2P{djGj_(Y42YZL`d6DsOqR%`jHs+e0jD7!DpW9O3!*;x7^YO2S_)
z_^SnfwcxK7{MCZLTJTp3{%XNrE%>Vif3@JR7W~zMzgqBD3;t@sUoH5n1%I{RuNM5(
zg1=huR}21X!Cx)-|CtsHNX12EIeJSTtY^L^S1WOz5C8QgAd=%=b=cI@#6<LI1l^0O
z!K{jw>AoKIzSi)*u`_+G6y=MCULFP&vmg4F!_f2D==6m}PYE6!eBD*5KCKC-`4_`E
zQ>(E)iX}~Q31>?e8cUmNz0(EhgIY=#sD<5UR%48F+FdfGCbVQzCoIQrOsq=8E10CD
z*Tlrcl;+C398Fgo`xP=J($+1Pm|mt)q8QT>xR_)2eR^ZK<RgQvda91)N8BFAhRAN)
zq*D)%K*iYs5tOT8X;r1{I|h7W_>X*4q21V*7^k6d!&n}x{wPZi#Z~8p)(-}iJ}yyK
zujaGLnQJJ@r%P9h(Q9*x8}nUb-qr@?Wew$mO=0b2Xt$nqouyLyx)z5`^3nxC<KdOE
zsm-(B=X+M?Jtk743Gc2>wi)-deD@NYph}%`XNz(f)OoLxZNGMBxzyz&@suyR*nU%$
zU#BRP(C2cctkIkFAj?vfQv0$c2^acRhNj7rBGDW6tCJ)m{BHBmGAf=An^YPnu1ub<
zSakcaNKNRo+CJ(*HA%Yqy*aR@)Tw5=)L<P~hJPZvTcbZGJygs3+auzRX}NBwG+l7Q
zaiPtyg^JI1BvbK0ns(Jn+Hl{PFFM_xa<XVBwwP;SeSLC(vpBu;+muPm!bZWX%@wVb
z9%1e|XeZz2^PIltl%<bfnrLRpTc4QMZZR2_iJ9_CR_T3F5SK1>KhL8(IV5?oezCY&
zcBz!?d(fTI(V~r(6~4QrqrxgCc^0>Md3&Oqi%%)^yb(@&F1q@?QD&}XG}dCJ&PUPB
zraHOO#6~Jr#w=#kLn=mK`J4E2I?ANaz1Oy-@3DH^!s}cN7x$IegGrIhnz_Zpzo<8t
z-bHyuhO<>%_hP)x7Icpej)a~=1;d7l&O03zG%BTd@dY2qz#SEMwUg>2m*UOPHZ0tQ
zC#Xc>_0b@2`pu>9*gfDC_z{>`5LF5K5cLx3t`8Bl5NCTk#an@^mn}QY^0X<hyRg0K
z!}r@ZRzI#HTS`RA?h1}Q{=<AFX~M^cX5H*m6_A);2Bc7~p;_3ZERY;tJcVb|?e!@{
z*)$oUs@l@i#MNeIl*o4h;fZ(%^CM^7NNHg*x5yA}0#dr&PLO>Q1zT+98S_3k4~Hn+
z)i{;%<%wI5f;O1xlM)D^@!ww5E0sKKQ7So<a0>igPljg`gWF(q96{-pu3P3l{#;hb
zNs>~Tj22Az4%R2lYcW1K853em%$KeimzY@Db7f3LackYg46i+LCq>UG^OEc=1zHnA
zANO*|_pow0`YT{QVz{LO%5_)_^U7@!w&6jOw^IefF&z~T<QE5*&vN}d@mNJ~v_8J9
z%w=)%{%j}@v`Wl4rMOI%b&E^BF$eZVLyA|l>U3T(^GYa0u-jy8J)e1gThoDwqhR)y
zbuIbn+C>ZFi80Q<S6oGzHK0fEEWVuSX=_`ixH80IxIH`;@;07Yua3Xf3*4O<G4tG@
zkz)yEwhX%SK3?1t3mJx6?QkWiv=fk<I1~{!TdWhAbv~BcqB~l@FUhK^EgwG=VG0Wp
zS{bR#HS0K~***g(x3yq~U{8uE*s%nq1Dzjm0;DyWIv4M^3`Uwfx=7<d@bDX^-oj|C
zR@YE0>5~wTn0gf@6{tLMy3#dres@l}4)d$hCcT6Br$Iy9?PcSvso9~M`<6X=R|ber
z(_#~Lis7qd9gG(AIB=NUPajh@I{|T~8K>Gooq}W<dJV-u;HcQb&>wiSp>Ok2e1_#*
zT83w1Yrx?QE_~wqYI+R`(pja`0TyaRjTv|NjwxL}IV;IG=e)hUt8~4>*=Hz-`}S4<
z$pG>dj;rgs9A7OY5eRuOo#+{HC(Lw~u%Y5UuS_TR&reQTv}GCAd3?KP`<ey;zlTH*
zMM{W?Rg&N43g+M81>y=0lO^jHp6?G1a-Rf`|8qg;zjDWumo@g6TmL8D`pyEIscb(N
zp-S0CoF`1^@vORyfe22MS{Ca-YoJM6BR&>E`Sn}*XEE(*5?E*)9A@yv--<O>kE!A{
z3H3xIYKtYbKyB#usSVq3d~%Ok&*n<{CZAnTshe=U8ly2Oc;5+@!`)@;;NX$$#BA)l
zyLf~%OovRiPN(i5f8xxM?UhGcw8e2TVj0YiJ*!=NSSlZSDp(B_6&l7e|D-aIlF`o2
zt}DbD4ud!+j_fLOC>z^Yo$_EO-VtYPm?$y&wIj(*`QrzsAb#!{{ng;CSpkp$xDFV;
z*nrgry!44queNIXagWPXNvOB?q@s1ygCF?B48f?i<=HYvKt<SHGQ-P*=*R3f>R?|O
zqUkW}Ay-ob1$Jq;bzhwv4HIh?yX6W^{NpH?gPkk~W1XFdfx_#*IP5!(?yHCo3=CvF
zYTd36X3v(=+uT@JG`1)-QupDb+2IMC?K|vLe$ds7p#x0uM6lz&YkKFhj1w0k@%i;r
z9P@X8Rd8l7(^V(%FLw?^t^^<e((6NWCHH^6HVHb%I};nx5TOG7BBl538SN&$mftgF
zl^y#G<nu6YF+<G9Ya}Ojgl>ylr=?J0jmPPJVJs1uILcM^?avp4&@)GA^=c_3=&6Ux
z)4s)Ln9S%7T+om|0hhah4ApTSJu*}m>tR4*^PF^NJVY_rO(p-rRr=ixf(zt1g4S?O
zpNDjJAl@Zuf(f|82{0U2#J16qIWozq@8ca$4czgMf^^>TOJFz+8uSklqV?|X5*N+-
zP#LhP?GdvjfbmxmAM@FjG|UCZwYa>zmJ(OebDS7Pw4Gw>x$EM3Lk!|$4Jq3&XWiAB
zm-Z9w${GdMgVYIU9+u)0KURcSUogze%iGk5Y9nVK1JRWA753*Rx3?1~gw6V-WjLN=
zXM0;y-qJ)7p2NQhFK(@5Mn+x+`?w{Kvg5}tU1q_K!2GET<apyW2kiqO0J_I*ro9q4
z5=g+_{pEW5e5f~?COYzhGO}+&6xiC;%mqE9i~ARew|lps-{R=>rXR*0f~fZR8p@Tt
zf}!duv<72fS{|&O6WCmvjbEKA+jJm&7RZ4|{1u6CRIC~8tKp3Oel6+x#a$lA<pIBC
z_dI#`#@0nWDITj<L8xqyE=Eaz&;arBi?b7fNR~_(HmgAWAmQ+j?(Pj=wuE01?MYw2
zn%EwrTFN%&iA=kT+j<C&Ok0I*5d}zR4{YO2<kU2&-~n>opp#2h8lNO@3zG~5k6s}N
zo4S3@ju&VcTm~@z!<Ti?G02<>?0(CWkPuU_i`NK8_L?xK*&#O}zYo6qEDAfYkQR?#
zbM2B1bAj-~*;19}QvnwdicnB+2~6^=FLLTxL9!?iKMve#C!^m%GClg_i)DC(vUR7U
zn{y<<4^#L8>vIh)LJO@1hSLl&2ws6Fp&|Y|!_N~s$``w{bME|okT=^#75*c`Q`Gbr
z>Af^$O(b_{bGq&(VK|`_0tb%_NQiGjeOsH8q5g?yc)LVFVuNi^V|_jkv2{nPfzB#p
z`%!1|UG0Z#1M(i&K}2#g^VTzQO|aRanpG_|Prlry_1{5QNFE&f$<Ad1A^|iC*Xj<V
z<jz<~H4sAt*$oUz@2n5QYNwNEk{l;WUMtA3Ap}7j=$!om@@z-}z~>nO=ZBw4t$uy^
z5Xog0!THt52N<JZHF&X8f`G@yn#X&U6yFXxjO1-moC@rqu7bgL?-oV~qR8<0+Y3Ww
z(GA2pb6NK<ZYzf*i1wtqT|=v-q3yiAq*p7c=wZ7W!|yOTp@05&*v|>kF~~eyqI*C3
zO%_d+I(&Ya=elbT`$>V7cQs%!*xBPQB@bSl7Dd{PHAm;%t$G<t^_hi3+7sag^E<`7
z5@}EwC>rJ@aLI?etS#w6Oy5h$z<`~wc>%Q$>E!`3vVW+n486F4M~^=VhBg~F3W$~P
z+&CC|<mzSMLPk&=7v6g;hk6t+j`vS;F5lBLeu#yv@2l{s)6KVh#a+9kzRT6e*HJaL
zWED)}2iDH;1d*d*$Ey3{;|oj>fAq56PDD=35a!97u6Jl9!d4|+XZ?#=N3OI-jlL|z
zkXw4oGlL$ky(r`jJLNqO63h*()Ues7Pk$t*c#X`E<~>o%i`cTY+(5m%Z%P$ITbHv%
zq7Oe(aB*1;tzem)6!#Nqow)#fQ6HbRjG^H$YoIlE#GVZgJ_BMcc6Kt@`7OhNkpe)F
z3c?o^21^PO7%@h=5{6C92)e(9-a|eGa&^-@q~#f-%6rGz+Gu{(DwDN-Is}{oMkKV<
zYw^fLoa+I}UL=Y<1%lIV*d6VCDsgZ!jJD`uVW+no_f&W*A}A;r4{Su&UGhbmfOlGW
z*(z`?;cjA~WBoxI9Vy*AO9TSP*h@q=x-u4hfXR4`jJ-OgI#&iEIfDlQCJO{cgl!2O
z5*zm=v*yAD)?#!y|6zeqIAHxF616h-^*Zym%)4O<GUTGD#3Hi07YDafGLFl=q7QE3
zHtmrNv-Z@J40*Xlx99F97T@-N!vw6j1keS&{2B-RA4-6hh!tSS=PLCYFE}m_hDs8i
zVp^(Ca(D|2J^?-PMv!pAn>W8~yd|Zs?^%jRBpE2T)9}Py_1N3!0m>mRz!tk4dIpai
zKbN2Ml&g^z0T+TS)aKlyV-&UitTtzMVjR3rEWr7U3uDh5SUpk;-%C5PvufD*6AnhQ
z#6fT?-q;9<08)SeSPdTIZ?}c42X2`Gt17Fa36%RbSyBy&wzjszr?8@^y%jpjch(&w
z{(+@#{dD>+Qq=F1cAjBf%(pk40U#b?c#K+@=wS6&>hq~%zQtF+t5tBhP{^NIU6MMe
zxb-KKlr9h2UifM_)npu^DdZ#5h=X?nD3>G`dZ_4*xiJXR;nKkAM69Gvt@eI=7!PcP
zTf)5H*i$TIF}_z<M^#r>moC9(Wu4|oznNUzIjUn^ulTlYLmV-PZOCt*rrk-y_ttYz
zP}RxRGW0%<XIM4`R;yDpOq1MFhr6~{bcC{KuUxp~^U7t<7$oDX^~Rhn3EZO<L>a_7
ztWVT(ab`X%88a6zDO=B=Das4O24CRJd`GP{uP|vaGi&9egKQ!_Q6PcwMfHF7+gun7
z)ioE_<=a!#U{PQm46T$t-8e|Bf5?qD*~stTE3F&rtMsR?@`{rH&xU$>LCoFq=WEdx
zuFl}Ydv6S=ff(!@Q4ENm_YnaufwX?1&K+hOuh7>V_v$wg+WQ|o$LwL2v)V8YW$=1%
zDTr9tuiSUS0k)q>aAo98S=UUU-7QZ93(iX>^%;7f<CNb;-#^%kT-Y?A@c89>`vWum
z%T7>OFiJ|w*zt)8N@3I`AE0V#aSNlzqoboY1#iQWcZHpJgfs)yMFiV#@IlNG>glrV
z8W4IPng$;3xy%wCsR1cnurLspjW6#nkyv!(rp}=&h&G)S&o%&W?oFarWItgs(VmmS
zwtVZYJJLAkcZ2qCL&fd04-9dh!eL~Va`FF*?ro>SKz1GLt=-uZJi#C;=7~tbF>ZN?
z9zDL9qP#x37v~~gLioVU$m4*6%2a$42x>a-z$hNhxGM+jqA0ZP^T60=eq`&RwNZo;
z03)EV^^LM+Qx<Mxn(<=PZgUq##b{tKoJ=^n)gWlSyg5}fi}=auJs!V>tio>*IM$xD
z?pdL|$A}-x$CZ?<r&pxa7#LJ**W%!<`5^L-M>RJ$FSKl0eR9})ssB_4N^o*I^k96r
zUxFaMUc&d1?Lj-6Ni3+3+Jl_jfZAhix=r0z>>M!V7tjOGWj9(Z1$iz+ch7?mZkvs8
z7ubg0vyv+$BnSC=S_#$fbIm9Ul2RlLKWvJCBTQjiHy)kLU!5!}J8f@4#cp~H7mt|X
zEn;as-AmRcza;pZ_&al?wPWK>o(hx_8&IFMH%?+NAl;hXF$WcP#h{pN&iF(fM3*>E
z;7nl0-@SnT0N}?l3a+U68VmQ?b@_{oAP~@@M|axPX%{%p|K={8^g6UhAR353{U1Yj
zEY8;U24vx;y}xvXJSymJL%c{*1o;zvSU>b$uqF^sbDy+PWL-4>Vp=;(t+a*OLjde)
zz}LD%iwX-T@pRcK6MOggm92p~W}WhR9+MaCO}+QP4nJf_90#6?(Fx90v#Z8hf6W7y
z6rSNQ)h+pwRoC4LCs|@I_(Ny|C2jc}&vuab`%J?&qwc36%oc|h78Y1r>}|}=t28bk
z;8ieSb7O_w*BoDpkWuW8fK&o8rmVMi;=wh-KX{TClKxOmQ8vf##JJEWu(S?*^dBA-
z&Q9`PtH62&@*j0l2oSgEDd7g5<Gt!M-T%2kEQd5;&meqF2F`?+l1G2<)<0Jc?jWpi
zoSmH|fR)+2L%!JWpDYE^W6P<M#SJI50U;%lVHyCtY306TEVl~1y0s6qGxp{s=Mvuw
z2RE@RwmW3;_G~fHd=>RZ2!tK>)1_{!2}2BbUY&pkBjCkP?<xRT)aQhf@*doYAOqrl
z;xn7%vmFJ%j6`q|gfxhUmmo-u7>K?rH$wpF*3^inxPjEdI>sg@MIhc2rMg={!Vm(P
z+;4&|3vv7Qt4EB@dg3VT*PpzUJBI`%2NoHwt6S=G2n2hF?X6y57BC<UXL2nW2OgDt
zklB3t1f1<E2rpk+0a}%DX4XsW!KIHCf#=|<)NcMyio9i#J0A2x<cHf~+(0Tpjo-j8
zxgHiF1+axQv0B!>8nM##2@CF5w9;hqsLf+fc-~oWhr`c65E*krn)0B{3V!BE&ZF?4
zg&E>XHlRz3f_e1zn$D1^6+2Gn=)YPey359K9(X-ZL~DtgDF7S|Dehla-Wy3Cj7{|q
zn*URAW@s!&e!QxGfJ?~qpr*>x_c;m=m&?4<@fWGAz!iJ1&nNf5B;r&cdOwesd^+ge
z5CX6!<D7G{xmPhMni@c4KLi#RLl4Cz!Ss=<%=&r5)C|V%Gc?4Jh$Xi)fWGxmX?E`+
zvGCv<fX)|<fA47X)-sVVaLUU$7iTWDjlF~zr$Mh*Y7(6j1N{ou2BxHNLw(9^s=FjU
z<McvZCRO4RkfU%E@D1r2fb3pRLsXvZN*tmS0DM6%Yj<V;l;?fo1pRd8WSFqWhFE!8
z9Vz?Sm<%xFE5wo@i-Wy=_m%SgC;Nm$2P}HT0|)Eis&{R|&IRHWu8w1$cReVM<;dQ<
z!gYlN&Ss4y*LAh}y}sBZe0PBJg}*ZJi64eoEVlWFN(Lq>_Fdn;*b9=tU@Ihg<q?3{
z43v_UirHTL2lg2oTMF!{JOB8GtGzqTDUd>@pTvv(g8mFZqZX=#s9m~XA+O;SGYH}N
zSNx}pLnJptT6nB>#GhnTS-aV8H4=-3yaQWZm?&a%Jy8)vpOOj{EVenv1LgHLop%rB
z_5YlKmt<soe61nFZ<WIOL>e1DF^?#m$$T^AjaX2O^(SD(cWHZ+9$=f6*K7aV?CII!
ze1FsMUd4|t0|d+U7lH(4e!d>0rG|-;<5@^K{Q_cs@1&sY@*WrO=K`UwIn_k9yLimS
zX>EXSoLG89PFg@~xsB4~o71DXduX>n8&HzG6$cU?iwai{=fLZyE+a>9PqFH(eOq5$
z6q~}*8vshtw2uB-@M1eWBwzv&3i61m@xD#%?rz_O+)-pDwMX%c#pZ%VmfzC=COk4-
zc=7qWt{~Ah`&_5=Xz%?9RxsqObk42+6lvsBJjW9IixS_9X^8EuHY+pkFJ!A8UwVXv
ztR@hc`yo;`w>Dd6$R&i_Bg)rI5DngOvi)@PKSu_Xk|B7nnKxNYAL}}^jsTHqF2eD`
z_tA-o36Bs=bhsmD@FF15h7P7T{uuG_2}@St;O}gemqR3+!f0e5_9AjY?uP%RVyE5$
z1YZPMU^DE_%M>%e8GkM?I`|h!YrfSl=AM=2ytSAw>;o9ngJInLOwUaz?lVd@dt}-Z
zqJ3mW=w;(HzgK7P3pb1@xR>4D0;*L3_+H#`{`f_C9v?oaGzR{^DdoXz3YrzDX>M+=
zN3NgCmm}hiKM4fp8x>X40hz|{V_%fnrDz;JAPG0N{I<{Ap;VkQOsCj9B-<TYHpu^G
zwIfzmK=}*{sRE*6jb!cjrBu!%uLyH3dfc5pWIj8FTqE=vb1ulrnhjhtq`|<iTPAIk
z0JxX5vhIWC8Vg_-b8~Y`6t@VY>@W%2C_wM(=UcXFJNM%O+QXa!&}n-s*1SwzH(;%@
zQGt~F-8t5S-?@pjzpL+PIiU9=Qgr|1|I|vbc7J>(-O@EP*3zZ^_%no(JUFjUCYAKP
zEh<`FXs3`9*vF##QoyW!S!;y0jN#}X_Ia1D%;Y>E3`++4mMO~5i_-DIISfJ?9;dNq
zw36<osOQI_n%edd|3?bGZ`$o=GtbLuO1S`!n0NQWS$X2eno#Z2W43F`o!cCzjT`dx
zk0^fbEG!NSu#jAjIDU5`%*8YwD95o{UOa7XzKxtfVc$1&!ESn!Z%>r&(dSM6@<9U{
z_X4xkc<ToU-6Fprt=o=s^#XDCee20`Z^E9~aL=p6Uq6%UMj;uzb|(o|#rDAeI`-8!
za%Uks``mp5bAe0F12A_<Rb4l`k6ETnC7_nH_#2qk4LO6<Bmg*GfC0E`>EXj8L|pYZ
zUi}6&lK6Os;$M{B-^H))k~#Hc2C5{d+j%bzI5^4ngGfG{do>X`se%1JPr8l9he`lo
z#e-bC{{&qy39`GpJBy!>&z<u^-yIr%O^cHG`*bi8oE`Qe{7sHz?|ciu97-D}e&FwO
zQ9IA=-|c5|$1%;Gl`2&&tIw1Dh7kf4-h||}c!XVblHCs?K%xGmV=wZE`@A=~&Ml1+
z%lYnw+&tw(>Bt4f0|*r}w7d9DsDQtJbYKZ6Uv*Z|!vUB)h|>gx?pPoksdo6*TjTJp
zvk?SvhzNj&k&Kl&Rx#(T<&lI2X7mCuLQ?W2)AB5yV(3+*5^lNOnm<S%WaH1?_`*j7
z%nFlN_9X*-{M{rOZU};sx;9UIsMfGc@qb=C#h-(iQ<W$a?aYx`JSLF`6sJP}!^GRd
z%K%nFh`>x*(xPYQ8e}zHY!s|!2EN6QHYD~?#w>8*P&nMBu=^yqk-3eR$q+2b=2CCf
zGY@Wh|GZzb5d~U8S6mjxgYNNteetj($3)c#WFuvv)~5^Ui31RhBHzMlYisMoo}=+j
z9E;7JF6wihsJV3^L&0ljFM9U@4(qdEg$dsNscO4-uEN>O+OumfDJCef6HD9^0l-cn
zGgF};e-<`@KLUwUx?E$SQv?A2EP@vbBFV+0(iWGsJ=W(>t#Q9i+Rj}7Kgxh3Uq4a0
zPwO$;^M+tHq4vCtH;kBuhY@nNo_Wb7Og1ih60ZnWCSX__)x%`$Uh~xigfsDjRiQo2
z{<+!LJW;){zJQB^fQ9@=SV?-q!`Tft-1240$aIIU?wY65$Dqo-h;rn9^6gok9TK0`
zL5<~9P?Ws;G&kBX1=YfI?W&)v6586^Z^)~;x(uZRIp2f|srKWo(*`LjsfBUjW<v$3
zHnqHowzD<dn^OYAaQhu0L7L((JrvLVBEoTSRy`i<#hD6xUQCcHO3c=CQlD&aYY8{6
z!N-rg-dPxLi#_X|K-e(AUMHYPKxnZk2Gx6jfo`<7fEp(#`%J#jNU}zqVZT>R*fQ)k
zDk>KhQ&AVP+Sh+HQEH26$ip~miAmT3%73_)KZQ6h2d1jyG-yaK9nsh>W7DkJ`gwix
zk?6C;VJ6eQ6JNbiY1&0B8o8!zK|?gHr*L@9JM-HoCpNQiwMTdR3g9$=z8B7#CLd%h
zC><H|n|zKa!Xz4SD&Qvt444%Gfj+UYS@nNfXE&@}lNdiPGPwDPR91=0v=#mF)ps6%
zS6xM>yU(Y@@&o6Q`5pJY-THh>k4Lk*;RJ)8vBGYl0u)JMi)7Pw|0{y6Aebx7{dA1G
z6V0!GZP0|s#r+8avg8#My)yiR!4?4Pl<SQT)#0-M+9TYg$L^gh1Xx@j#cShk<>{s4
z&_xyJFS5tQ!@wZ9$(^O&)sBC^fcwFIRR($JfAK$~TXox1-S^UXultvwnn=Stv{Y8a
z3=dGi3vOIwvSz)0S^8U_+muVso8c&%+UD)r?sg=U2(m<Xf=6JL|CHE$ci_}8&@|DI
z9_79=x{Sva&oEWGI>`@Ga|4CnI(aw*v=G!{Q&?1Vq&>A*&MA;19Aw{xgD*9OG+06>
znuQdyNnQ}f43dy3?idF-N5EA=;ZG;_z|$=;EoK`q#F*mSKzcv3eYd`dh`SF<>k0@M
z@xmv~)pqEz8>29ZZ_Q~*)8^N&ddV!TKF)MkERGX4)||RJuU%xHlaO_8&bmoiRZ!4r
zX24<qu5^A^bOz)ASI-@f2tKf3(0AcS-B0Nf4x&#u9grlcH_WOSRb!(;Vmll;zG*m8
zGW~@@SgS*{)xZ)H-5oE=7dadC^$e_nz^msKcQ&ZXf6KtVa2frP$Q><`Glqs=BC>Z4
z?mxblusG&;Mz*TDIQ|g=|AWE2BuNugt~H!30+G)x74dDF-YE|YE_T`o$4CtG(aI4c
z#i*PHL^^WK7;kX}BwhiwlN_r3RGzgEC)_7U0G@NxBNscc(*sTp^DO)qET`)suz<_2
z{C&cW9YHEjW*?I_(XS)JNgPf3H5swg#hVHeomt*v9!ljgf-d%EiwnP#IMU>Ij1e~M
z8Vcqhe0Xne%$5*+xW$ikm(IV0LuLU7dp4SSt2aQo_q893^V}20v{nj}jSJsgK+BP0
zjK{{ZL*U!@An+*G=<V5D@A3FOT}dtct4Zva+d{2F=`6Z2S&{m&s-7)7k%h#}y|+JU
zo429IU{WAVHdp1!{}iT!q(!`im52;#cZfRHM)BHPQNu<*`I5#@PENAAh;$O)k|8;b
zfUngmL2V+#Sc$I%A+zhjqvhc82>bL*p`euZzA{$p$FBcbBpT4{eJ07GPGwIf6fwng
zlY1pzGIV1EP>_YHOD9_<P7S7+XC9pkkX0MM5ak(D_az^Xr6o`jDP?maG~Sh2fN6MA
zc8U$XCHvF7{=v;VO7L+y*W19tQ`g*g496QQ>RC4j<*p8P5`Q?}03vR|_eh_^P}(`K
z_DtdmyM@q-`31)+zQ$b?8j%gcSuy^(69-Cb*f4o~kS!=xFhi054Mx3EKl>3{8BatY
zu2xYbk0U_6$D*#@mB{gxK9!aX#h4`!Dv)96EB+Y8fCoa`U;V(RZVt=e=L?mVi4iE6
z>s~BfoYXqZdVfcRke!e}n|vgNIv_;E1)$LJ)dTA-@Mp+dApClIG@An>&;X@!7E@pu
zzq~dBSyG#ecMX_r2EIP6e=X{_!PUGrC>=+IL_Z2H6SnHVA*f8Cz0j;_v3?`JmWt8#
z*|Xhf@2kHjIsVmumWmIo0vIObO`HVTSoaoOa;+b?NWIfno6_CwzCcBER@c`jF8=UD
zO?DM|SVAVg4MFPt&Uvz6_+|I{q{F5NrFg(dfVf>Yx%5vo-2-cmZxUN5<o{XFKPa>y
zm|7;m8-o$mMR1NaEc@oqKc(Waj3YZWxgk2{(tH8Ah6iO<iaD|Z=%pA`bvXMb3zM~n
z?_R$xQU#Rlx-T28|7qis@#hv6+LswP-6%0>cM2eI>a%7dNX+hN_X<uq)fv)BN4%J4
zQw*^9_yAYf0kk|!$#UWWvyg;>DsQ#1!kZOmj-?OkONX(&B0H78sb9t4e11nx5si>M
z39dc+L!SPiE@BARdCb=m^t7#3+^$Y<RHV<%EOMA)N!2QtA-Y|qS&DO$2L1IyP-iai
zGEd0@URV2{VIO)VTiL2=m}#e-H)g77wJS(?0EiFL3y%LEm`E6od4Auc9^SF{S+^w$
zCbjta`8_VU!m1aWwB>POxTo;SNDXs=O?O8szGut<v1f>R;b=m~?{{zapk^c;)C7QY
zL;6p`9^7A?4l*OSek^DHbwI}2()2G~yrSc*l={;Qeu7X(6i^M#{Ii22lFn>MA8Pld
z;j-E2siXFMg}?f0u1i+T+Zx*XCWcwIGM7c&UM$=RJMY{?QA_=c;8qZ-t+5sVM9^Li
zc|!i^@B=G<AV6nMg321j>77nIL|i^hzojoI8hG>1(i!;S-qG(i+17f-s_8i5@nHqz
zcx8l%cDA;zEQ`JzmyqrAHUKE0Hc_yN*sIh2zb4eor$(gfdv-nd^TvSqS^&^W>R0~O
zeL}12bmFuG)XF;5Q|vU``s({9+sSs5bZy<DK)Pz)pLYEEy6<*TDKf}+?^LS(Fn#+t
zKwJbemr0DfJ|{GI(f-N#St5QF53X@7vy|2IAyG_)h$YIT?3|n~^*I5t2;GT7UXbO)
zrfZ9dX^8Re$PwtmBeSIMk&-ssoqyjPTcOI9m%E2UCmXXCV3Jla)K-o2;v=FwuqmXH
z%4J;S%u(rEI2v^5YW6y<GwwumIqxg{9XQ-p<`V1pr;6#XQYD~wb^ZTB2T_ZlUCcOq
ztMi1R&~rI$<N}j0w8~pFavg3<Ly;C6Z5Qnf-jM=m?|SRUmFzI<myAevBsxp?c-bL`
zdEvzzr3erUADQTmsg=E+pPu!Sk<z(j+q;}0MaRDOKla#icOn0N0orC26QtB_KA%(_
zH`DR^Y+?jXkr=wJkLi|ua18(U7=?y`8hP=+$0O9P9cbEL3?ylxtfhm@7V6|;xtY6e
z`RhB7qp4G#vHD*`t$^v1_Zzncw6aWA+yd=5i)k6p=fm`$@Jh#7!e)o-Hm}Rd%J$7j
zMEv;H)sZ4Tma<nMj5`CtY9jI!2et4Y0u!u`lw>Km(*~l7pnF`Xqd2+l>nmA#rfoB&
zN4;opV9caE)&pM$K+$!ogn5kZ7iGz6*Obt?v{0gZ0#bdu8@K@n7%a|1Ceb~~fhGo0
z7_V@t(j4vq-DfBTj9W5b62UNz1;DVvB=)o$p9h#f$<M~POfE~be(vIX!1GA+<+V7F
z+Oq~x@_*@ni%QMwUa{$OHsW^?CfVIiGGYD|X*WEaeL}dCaUa;E2;SUrEAvl@EJfp)
zflzWSrggsjL5ga+&?0E?FS=UZ7=P{+3VG-V9n_d{h$KNs?>hwW2l-HTk{@%-+;}Km
zVBXS`dUkhaQP9m(!f@dd(ZRj`V<D$OAR@6gGv1bEL6PlA8lU0flAM~Fn(J32dwq6R
z&JhWu1nsam)uX!vG_aRBuXAR)JyF{~yV&~81>!HLc?aA3P{$+WN5JU|*kaVNFw6fH
z$A+V1*FMR1T7ASNznF4PJoN~bV=ptm{kWs*#Zd`%1tgj-&<bEuWt2uM>5{9U!fiA9
zE+wsS*pB|(UZw+TZ@80}K5o^U_Nl`CXn=~&a|btDjbRGv3Mbfl;&IY7!2&txdUfKj
zloguz1Xyi3X**T_BuGAux{IBl;_KtXg5^r8X<Exq(KiF}*r|C2Gh9~b%jc(CQHGRI
z&1}nG{&2?jU^6NypgOoyfYMD4Qv6iJKtwY%dZkrWw|t*t-pRfhmlxjEX<2bBHk`vq
zfjyE*OS3AJHMS6RCFPN`T)`*7gZ0%=7<6=W2owHtfelAH=!_PQrLFqVxp2%Eb`7+0
z?CMSMAy7#j&<pnZZ^sGjA_-m**;FjA&5&wDgT@t2C#V+;G_L4gzGt8LL|botWu99I
zAl?QB4kLA&36bV2Qu!>UjhmfY85L%`>cpURfbb3c@c$5``{4+H7?6TY*FxRHX?z4^
zoi24h--xkR8;swkV+#Yd&yxC|Kdn~vh7I_IK-RCI*1IwMVK92owIcG^clAJ7({17X
zk08iVpMlePD*_?}wM*tt&6tkr0Uc0@>fr^a4dM%A?~p;(;1-bLiP<-;-U;~GT6&<z
zb*d{vzryRKL)6dX9z%xiLot@rA6WMtGD5&_L*NCK>%NYjJz)b!1vU+*gir(TY9CJb
zbSa0bFkU>515@GXI8BcIg`-2SknJWs^&;v+Ky;l|P3Tl?{^vl+6P5P2`F#y2D(xAp
z7_Mu>Z;(EwalSz+N_PI;*`&D4ckHvm51%r7Wji=<^4nC&)>XXea_O<DUEQe7aIl*S
zXWyX*8QzORn#v55X>ymjk)J2E4HN22(n}Vq=f_&6qQf7MJu^W9we14dk+p`i-2U_8
zL<ctkVP{!k*DDG4Hd^64#_+h(T@85#P9C^q4vX3E+KPX__<;+PmX@vq-T&4)<hIe_
zEFg=g@D`mE{w8&e;9CwV^KMnQO(mtnxnL8oi6@(_l|uF+%MDDq`~$l^!`LuiV1@**
zD4vIU)c@TnZ)sH0O9W8(OXE6g;VaHTH62%Dxy;(>%WGR6wm-27M4ZI|!R=hp?GIs(
zl!^o@L0)!~nVDI4d<J=RS0@p1&tV(!-Qqkof6RoO!+Rn8fKRaP`&Kk>D}$O(voY6z
z`~spBcL6jrR?6^=z7<_oxnHyr!l-()Ma!mUu_Ms>B_b17B~_!zwc;4}x#)4Gl+Z4t
z#OMW-C-xzor0&Nzjnfg1vmvWt6Q56PE9V^zU;?FN)BFEe!%mpJZ9jsYkqWBHa^0WZ
zs>7~$G%K$g#`k=CPVYfm9G>Q(@>^bP1E)Na!@Ur+p{|`Qeg1^v?Ah1gz@(1a#&X{C
zh7i8S5I5D!VtZM_MW}5KM>eqWzo$jSncyDh8;NrX5?p8v(dL&ZbkQxeHL`7x=5}N5
zQvNEG4l`+fe?|Z((#zuJGJ`Jq7{`8g5uf93)|J0X_cDXvi$GPLb5L0C1(|zn*s+n|
z4)Ps=m$ttjunBp7F8(VZh2LZu@9(Ckt2To9l7@i>k-X+K?a2IM<EL^<PUPC)C~53S
ze^-Zze8iPq%X*a#$m<t6CEh<I2-SgGE>cvOM{TzSRNaw8&9i9ECZB~wEh!BJGV005
z9+177B~2!co}B=d8UM-rUxBO%j3#YE+EB~$dV*B=%;sR1zUF%6Ih*SG^@R?L?%b*M
zJs=SdRrl%b?hX9W$^8#HE0f?^kae5|k)4+EVaaJYDCEqSd1-gLB;?~H4@>p_6r*1E
z%lIm2s618JIg_aD73=U>TZ~811c_#E)=1NkeK5VXW8@9IrFaIC1~_z#@;_B39wjeE
z$+;Z&@OUOVD?r>49I*WjGWu@WT#|k=<R?5Y?{YypnW&}bpDrr>x^dJ0^an0ijn8JB
zG8O2(k=tnjb&JAqc7vpUpAvKkmZyD*ET8=1@qF+sqkP9(@#icAgg!;8uUXxQM<kE{
z1lpto$AxIdZg<}hnkt?ZqI9o4Sy->tvp3ge+FtTy3Gv^%HI;2qx%D35=r~Mu>$aiT
zZRlyaOV}+%JI~E9S@!$VZ<xorEqr6YW;y{P%Bc;IVMb9<FnA)4;g(0Tou!isL+f%%
zotGD(2IW@GK%V}`;l2_QdxbHac4D+;-m!Cu|4C8M1TzGSH@-{U&6exlnQtZhuG!Iu
za`-!Adhy$0PoC7}D*jcNtibdqUvNnAe$TjSdno}pP`0{^o;8zPea(16IVZN`<Tn-i
z-7WiqmQX<}MVj;fi<coNP}^Q+J#^&w`zCbq>BUz*RKHh2jVSizlRLEG{NP-lUm=JK
zwIYhj9pusA1lSuQw-FK5iT><;y&xwp0bbH1iYn~R|L+%agvgAl>EFOP>+1M@Xy6z+
zsj4--ni1R@@sLAWL&n+-R6B20NAo+BMZKWVytkzEB;7mKgu>tu?#1hIYD~?vG$i6X
zVq(nHOC8A>gH^GIodRGs05+%{oFJ%lR_NEqa`O7HvVA?<m^YD1_%Gf?PgDvmd`Jgw
zEWN!V!MGUUQ4VMT3JJY<_m(!&2|43dm8_Vkb8e&7HFe?=)cf@FQexE1*utkLdxqy<
zZ4sgz_dC&JbEn(0Bp!lRV<PO-%uL4JS*0%JiZ7{<+&JIUwaGBFfset`9Mog==P}NJ
z6Sus7A>I&(+I!=`@Z1mI2_Mr-g#k0&RaVfz<#eXj0T_)V)iHvr=%b;nuj&D!KU=52
zxG+y?tQ4*{#g^p%MrTLD3P5o{Fi$bvazE>{b>r6JtS8yy4N8ULm#RaE8tXK(K1{9;
zHXk|eYi{nrO0#SBcE%Yc;Q@GZSv@%BId@U(p_E10`oa*c<jZsN4__S<*Gq&!&Z-XN
zLYWOVPW(KUy{HjTL|GdyZc5eJnHp&db*}|DDeaw#01m0}MqF`*TleewD?Q=zTDGHI
z-YzO>td!YiaSs>$Uw@(cY12`<$o+@sLck?4VL`0bmou}ON{aC&mrnT&YIkmSuuO)$
zY|=Hv0T!Kq07S0*DePuht@j#}<U^01X46wXWumxunr{KB9B-UU!5n_s*(&8J2H_X3
zrsAVX)LN=qjz1~AU0xRJ7<YeoF5=T%Lr85=<OI2);2TJ^#=O<2c+_dkacP7ouGZA$
z@5$_#?Gk&Nm?X!?53<kCQ>En1TSHz(pHVeUAUpN+FrJc#9so#sG|SxGI3G!@odyxG
zU#-yUxCO0{JpQB<%B?y&nP(gHACN(neX`*mB-jCRAjw6g;eIFVh*j6&TchVTcbn-k
zzudXG_M{%W=!uZ+R6VP`UF|NwwL>}tF0RDM@e~p5iJTUlplI8AaYFt^2E&h<xu(8p
zr7OQBtyOZZN$H3uO}Jztxh5p>B~$K=M%gv1vsch8-sO};?L%7ap*F`ZoW=tscXeV0
zR#Ojs(a{Xu?!x(})VjKk=;XV0Q>H0t=6$-)SBdWDmq5GUQQDmgBXWkRej)IE3E9Mw
zz_i-Um9cbT0C6`Rv7hHzq0eAMQdksEOM=FO)}~P{%BiLlRSxI5k@Uv=k%*Ej$zSX!
zcF&W?T?c+r6;FYCzs0$gf|_ZO^jR}E?~Cz>Qg<Bh3mxn3h~Ycl#s8$-=2>V5DdVH<
zfM_>-2)>7W2tt$?0}!I*aJ2krDzNCxzY@V`_ghq3FmMBZTncX5{nG{jkiy+<1p-vy
zGYk!r?ciyhuLcvwyPWzO2$I9LaLJJUTaFWfQE<IjC|fJ-A1{lPcU&nQe*K0&GrqQH
zA&?4>Kg_JF%@Z8!`aKU$ImcvK_LV0@M>{~E-2P3$e_TTbcIJ-gfzNRCQQev@o~_OD
zb2-`s&|FyyGc(Flq(dCYCxo32J#N3CMt~=akzbA%*^Qlw233<-NrOq~eEs&A86{wq
z9VSZ=`{&>#R#UxvELtF)XCgh3OeZWW&e8f6Y0zULN)3n4flB&`Z{79fR~JEioUIS)
zXbqr{m}g#jx`#9=1o1)5mgud@(icp3=R5rTZsmY>a9|MAl_f$XIG3<kGUY~Lt>#=L
zsZ(Tc#sS8N%_%}6U?++37B5mjel(K6HpG`jXys>+5a<0Lak$|OH}~+2f>UTq?v{ph
z$9*K3xtNilbk?@(g7@aekg_*zi?E7-q0K9sro~Ar(bc2x%RDs4lv7lWIRc!0V=y@}
zMxzMy`%RTOtwjK=*3R0?OmOud%`w48ku&sd0o1}9tk0fGTW5hj_%BNty^faF>@U=t
zaX?Y+sUj6<-|tpWNnxM&idB?f%Jvl|^l-Q>`MlP@^<Lz~fgkrChzp)^f22_r()jHu
z`4$)Dl(EamltVi7_Ompp6p??vdvl0BZ=-UN?vzgCish&xMeexH@hM6bay!4*-TJC&
z4#EHdCJhB2!Y~Heq;AKn7D@Z1?)`0#+m4O!oL!?lDRdO=_B+3k{(*{{c#wVk;OFyG
z5OO_)$V91d>obAEc;~<dSSRv&+-Dv4#PsU}2zLt;<=ndkN9fVR`mkkVi%y*?-3++o
z-9q0<y*?Nbz>oj?)2oL^EPNqB(*)J~kV<azP9A$h?YZ%`+0tlA2Y*B=PI;}eUC!AK
z%t3fWoCa`sP0zTFL-ZI>hHIH{#{$9hc;2$op}7cB$TH$dNzNQ@opZ=q0xy>rVx7P%
zi|}^{nM{-?N@<;Tmuce_G$Bo)+Li!>3}neiKQZoOA#$9+>&Rol;S#wO58ztVJic|-
zFRHvxH%)cREk2P08t$5mtd`JyRUHJD&^H_Q^8y}x6G`N;R8iIi+;DxEbCdP)*adQ^
zg+w>K-ShDGctrjY4Q~B%oLHLWp>cr0_oehK*E)2+tRjE-de?bgg(4xkL}mGXLl4Nq
z1sX>e2A<V>Z*rvMFTCy`*vuW6|IT=Hq=2{mP0P#t!1Sa1xXbyIJha2RBW;;VWFx+p
z3sourWx9hDLLEa$V+5RL^1yk8?)2{!XIaEx>A;MC%<Fa@bFopfj*NdK8@lB0l4*8=
z7kv0)YlFd8LC&PerC+~?A^~p_2-DYhsmMwXsa+qH_U7y>lO!)u`CV!t6<=y$-cH=1
zEFw~|`ROqJek1}j;9I3$=yc6D017HbxY^aagFdd;%d^9q(sno5C8%%{WTIF6uD=Sd
ztnX2o)%T&yHTv|JguTbU9H~U$#eKOk{;HMuPRMo{YIpU%X15D4L8+TDzL)pJe0s48
z-(aW>L6wKCw*K!2@$yq?R7C~Joc52w@%cKxV<!-C-(iD6scAF8kG3jx%H)<MDSQRk
z5C!3hutSHBsx{Tu&#Xqh?3v62%_Vc|=vDD_dv4-eanVp800kdNIU738#4Klk4A=eq
zd*AMe<?APRp{-y>KsOuCt|$BG1}B6KdP_>`vQpYich=AL6s1|tR<MdHcU(2?`4V8|
z6$wrv=3Zs_>U2=FK)QbfM}qQks>I0Grx(h8g0Hl^^ABmhc8W#obeBz?V63dk#*9oF
z3Y4teGxSB)=%wlM9(=!R51P|%0-&Q)Ib=*;c2A<7od;yQI&|&S-YsxA>?Cl-x&`J9
zF6sH{OG^%hWWekd^t(5?78a@Y1jzSHnt6;{e4WnI&x<ooI5|eQBOkj?LCy%7k?%Ij
z3tB)7;#X@lJV@zf)><sflEj*SE67gZsHADNL6r&NbnV06&N=}ALHF3A0XHpU3U{Ve
z5D)S>4egF@MnETB8ce^1>?s+tlfbl7jqm>Vhu$uwf+c-Fd1XV6GdK25qtF}EMKtpx
zKMii1(~I}*-xGwBq~Y8QI6EH?ke8u%LxHDOmwpAc2(O_l$ZfNu9brW$pc7A<Q@@*1
z9PId1D;L4Z`;kUP*>$_LCn`)k8jdgr>#3_zV%({!LUP7Ike&PTQH|@Jv3cES4abF*
z%GYnW^GcA;p)d126irMwfdn~naNWW}w&Ud=QJ`>m1Jy$ib|B_^H)h5GZS`vbLQ%2d
zH8@gOTg5A4x(wysqTocn`3EEn59Glu^6B6^N9!U|YUVO|+1YCh<K0vq4P4wbZ44aM
zM%rAL<#*y@y+xeP@oquk)cyMZgBI4OUlZ3xND%6=+FoAOKowZqnh`Jg$jBklPX1DR
zV$YZSr4W$1X;F|52<&Wjhvab&VVp}$MN3C3&VH%~ZcM1*#}8uas9%o(wll8-j&rv0
z(#bh{<3P}aER#-g{`HdhqDV_>E<RB2+@J-h%x;hc1)ofoJvjk8$RK8}q(X#RYqw)d
zmgeR!3Ppl+Z9CmjQj^;)FI4G2udi4<cIL=lXW|uTTo+ybXl{4)D(2fo`<+YN0?Y+O
zrUvvNGJV}{_-y=WT7Lf%zX~_jG16y((!RAU#qNdDQ9RSCzA~9lzD2uC7KD1cd`L;*
zxg+Ld9h?8=cK<H9x@mf;gDv{d$M%!K_yRH2k2!bra=k+-`<slRXGO2ev_6z_IgJ}P
z4kawf7EZKf4f|{7I2;ISAqfl?Mb1R7gD*v`e`9J^`4$14;l~Cc8ejY~x8IDqFmPtz
z>ls<$b=x{u<&s%r3JQuU(7I_eydSF%hMNAet^~c?XG7ufklhWc)Vbuke5B>1hKHR?
zAN9-PnCk0~Kt<)(1_w7c>tjEw7~nmBkByBPbLLuMwus2V0b)mMb`GOqXqBy4^;aOj
z`JiUdin$mPoQ{a^(+T4w;4fbiQi)qD1Rd(SXDPy-djg%l2hMrF9DJ*Fceh_;KnW#G
zWy3f9eJ>_N#O1>JUP<M7OXlgIvm*V*Hq7itL+jF$%!JGWx=Tt=9~&)7P7R(Z5Y?~D
zTvA3`mGJ!In9QwkL_WR3Ibru=?M{Osv&qhaRUlbQtzm45_{Gw_Ksg0hkVmLaujP{R
z3j6?NrV((gVAJ}zr0QOaW-X+vZC;G~+Fg<D1TP#YG-jMft97d{y<h7K_<2&6<&s^x
z+0E%UL>9&Y?H?BB>wdYM_{TZsuL&rCXgD{l_H1snQo=IT-B#aSKIQjQSykK4C|`I;
zvAJE0?lRsBUDoIHvbip+k?D^b09(_5lPMed{8Sq`!|pY%l6ZB;u&v^*DKxq>KHYDn
zMzRsh*Q#eY<B7Cg4W~n<q*%2Uze>s*B=V<NR}6t<d{WlH;MXy$EB`!~l6e`F`Fb$m
zoyu5>k~ZYX&qFT#vBZ30>^LEldgg$3UYC6bf0nM1t9MtSo!`wMs)=DKfe^gyC1D|t
zLOC{4TxtIS;BOdjkdwyI|77)v?0VN+dGsRF<zvjW++ET4xQdM`WP&U_EJ4QQm%G{B
zg7b_{Zo}h>5%0GxgiJ{T0M+%cSGF80CFv6Ch!ER!??WOM`}fMFuR+d#0#Zu{cZ+W=
z#Zxjlpg6y3=d|eI@p|O#a~WrA00!!ycAGOj5nCVEU`p9LUx|f`z=Pbuo!<{XA1UQ5
zLH~$Kl<*;|&&?D}njg*Ogu)`8GAAEO`_`d+_@7Y`HY}R#)@SVyAoK~~-xfY~1M`S3
z$6o<rX=)$J13L@Cr)$aE+uQlZp~Ujltl*u*yKZi7bH1{!{m=V`r7pWu-4oNH-DP6V
z2~e|CnZaZ)+x7#V1xrnhcQX{yC>~8w1Rq#&es4Q4blq7}&=O%{I@9;+JL~O-IQy*`
z#-CclEz9xg*X}`};kEc^R*+}}`~=OstiN4}G=@a=zQ4F6+?{QtsPo+{s}@pUtJaeB
zk6aI*+0|cPyHn!>J{GfOEs`Ms&rXp}e}~OrUC%BP3u!o^$8o*!drEWB_1Qj863riX
zEeRxttX)b61`T&IARxeIqp@6dcWc`h$Tb450IX&okPq;d!6s5%hpNM=wR4P{&<mn|
z-`}8s|6Ihg@LI|eXfDl)0G+(TSuU%!cdk0UXL+uCY3GF8k88l9@#E1QFqixUAYNOa
za?eHkSx8+fc5ZTupElu|JpHtLYSGM}m2UjpoP;yu3*P;P8QNuBCQLtZ&;v9o>D}e8
zQL_MGTu?d9VyxkNra>jS-rNrJ)!*QQUzb}AoQqa~0A{R2gRiJfjHV)n_IAey9qF#D
z$=+vTJy8^pjh%90B`aNwKI7fOTADH!a&jySmpy+mA<3gI)k?I6fd9zJ{s&R%Nr(@x
zO9m6!7ZEe;EtkO;B`83a5*;blQdC=D1bt{ngCKt>N4Bi7G~>>VV-T7o3H~dB-@ein
zr~~wi)6mCz(uJN|B)mx?&b7x5aJT1O(~>yFA~0)ZY@HA7N^Z4(`nn4{$4TOLj!!D+
zMDGN^Tb>5(8DIH4{YB}_!!d$X@-CH?jb|-cYIn&vxy>FkgQHXCWl+5PXT0_`90diF
z;PX6XQOaE(tX0!>D4}#kctnlyrxS5L)LC0(J6bRB;;iN`m!8#b>AteD+lQW$2fyQ{
zqYaY1xl6@U`7X!yOEQR$0Ri_MfjeIj+HY7jJ4;`p;t_YwmwVl<t&Xy$!T&f%kEq3g
zaLhM10}9U@-gU_D1SqPzeN6=CL%vSAht(jh%jlv9jW8DB%oz{811R|yW&w!V8q_wf
zmsh^)%JD8jhbBMCKE>I)MVyp_a7<Dt8<?;Ede8;f$!H6^Z^c6TPedLab>f7=Z*gM0
zP3jif^F+LujUm&>EK9~DwB0a-M+Opt0tp7$4CJns_XA-YXCWxTl!d%`7IO9i8W_5m
z9Du@{>+}df;w2J+!qrHaL0~vB^(7ifSl_W@lLq&Ut~XuZ;)pF&d6Q`{+D#llkk|~c
zI<J@Mm)u7B`Tf#e4W0FyFR+RRsHZz=x>W!Y>2y8mDD7(M0pZNVhigT~<GWu>+>PF%
zVe$4~XfHFF9z|<`!`Q6y;<Rkx0Eu6-%O`3PMkb3VgHPMo$qDP1dr7_%Fum<#Vq~-&
zX}M{qc$?Ie|AW#-r+Lvu!(9r#gA!BDoPd4hB*bNa33RBYGL~!}a1u2gT}wj=&8-E%
zD72jXa1FL&lXeA$_aq*nqNG0Zw;`wxT3`NAnyCvQfDx$A$awX&=w0>RZ}hQiKPrfX
z1>jYjxp}$z6^}%r<9{BxAiv)=1jzT$w?cBxC}X+_w<WNM7QOPN{-Yu_Q|DspsJG9s
zfHL3P;JlCti{Fqw*c~M@qJHvk>hJ-y0l005(?@4BRN_or798*2h|U<jF=dZS`28)y
zso^P_542Q7X7K|gk^0eU`gBq7otdUMhCKr-e+OBlioLr$aPa_PVq(-<z5JP?R4@l!
z{T`$UT;wxV^>V8Ij~2RG=@5>++3uU`i@II+f}i7@xJXQe*G$VmnY<-*EF?haPN+m<
zqAq0D3A~nJOdy`489fj;4e~zn8ql@m$IGDXTSN?fE~kf7=g^Ufe%rL;L{tJiD`+3#
zwdYzVQ`PpC$Pavd>7mE_d-1=9BGi%uN1AcHtIFKW`UZ`NPRaIU(CW@ly>FkHoanl}
zk#m}=R!7MlQIxe1_|J0Nns_hj5K-|eNCPyN6EBpuhoeA^&Hcn=(|l_iZ)GMcvAxBn
z^ar(S_g521Y~$VT-$37*zRgp6HT>OT$k#MP^)v7BQ)xSa5+uem%*0eLIlJdg`+CZ*
z#G~6!E`|twNSHWQW#iq{DZKIAVs&nm{gNuf{senlr4-<%C*(|lK-<zs4kxsM51Oib
z05W@A<;G>43xWYuy%Xvh;adngV!b8`zC!qhy_V;fg^g73YJEi(p-F-D=4GsHi}?bC
zYy)=t%qK`G%WL{{h)UE}njP+uHEUU`j;5=;uTx&0xSU)3(4{k$O3+z=g#6b(QYudr
zHBR?z%wi{cW3kMGG$<FhoTa;rU~#H2hjP%g8EXIC$YR!X&A#WPBfK#kq}$NinlE5t
z^L_`k0%W0t4wLQATtJszJi?K!dHvhx_`3Cm>hcxc+DmH<&sr4KZ-&x$rMx0OHd(@J
z9`zZN!{?hl0Y*ehOKe|dLv&G|o#qTA8Y2XtY?&t4M2`{OB!5J{j73TB141Fy@b1ix
zDuBN>+H9;X58unX4i4NBPj(zHTxd7Zv=@k}oB|(grNp3}vuZ^VyVe44oT0aGGLU9p
z=^X)A1W=%O1$$Lq?%G^Jd3$WP*3!i55TS4HQt^fK5GG2hEh?WVhTW+s&j0DE;Bvf3
zg~7si+Okmd>L-!oyl>Ttj@Dn9X;Rf)@0bxRvAx4s)dkL3EqIuK7`LcHnywwH^EiO6
z57v9H!x+AQs+;hy_P)AfiIT}cbQID+XSAmoIQ=-~A{i^HDxosVO;P<O!lgnH*?LBe
z>}dhB@{OI1XRCfRHgd3oe15-zV+KTN;*_&YMt%vPSj@>gH&@evGg-E|flWxxA~0`8
zf9`7(BHyRO;df)QM|43spS;}SWY-&T<Z5`+_H{Ddu2IDK7wm6_k(la9W66T^=!oC&
zDczNB|MD855~Z`jx&eWzr53*)wcZ3=)jmye@gF2Z^qRrwqb$oU0Hm~#`|EhIaxYX5
z&MroN1{oiV2@KTbHab(%OCxeJ`IPG*gS*u2GPSOhc-G{_B_#S@(o&sTjne1cisiT#
zN*K>!vY%n!{x6YYK%?Bb7{*CPp^a1%g2nDxt^bd%_l~FffB(l{i83<EDp3>-L@9)D
zN~L6HCM!us_MV+el8lV3>`fUVdn6LF$=-W!$M(CP=O{kkcfbC6-P~S|=XpFIkLz(=
z_v?PYU%Ds5(Cuk=JEK~LpL(jP4A50`&hXFx#DI<_@^!k+eBtaWF2ISA0MP$H&I!8)
z;q!QXg_UcNj|(dnju<Rix2Rf?A}UXhS_S#UHZ{f#XqO=Ela4JVwW8#d6we0_9ze5R
zC_5&kg%+Lrck^`v?_#~<&K|}c*b3ks7mMjbE`_D3W6rB)p{xuK&c%Ly@p!PKQ`ix+
z{QC9l?K`)>e%YG`A}`vM*(>2z*lAIZjgjjw4-7qB<5$y5Q+fye2IB<YS&L4N3m1^m
zzP?s?JD;0PvqWeyg1jJOduv@d;6skSPH5m-PFGEk^WUf`CtWN|!}=zHZO_I_WJ@(v
z)g86)G$>fsSEQz?wq@tHG>ga`(0P`D+@s!WOB1{T|J-;S5J0Mvpqp#9%CY~WwTK3Y
zP7YpxqK}ZijS{R^o)YF|Z7VLbPCUGYK)`X6-Q9cwpZ}Z!{4Z>9sauXHXz$sCu;ZRj
znpF8X-zzdx)vRc8iAv`&s4z&7zW@;B3PUIOQLUNE`*5{(;Fo;o2$PxRp+1OwAfrCg
zkPenvLSAigGXBE_OBSEa0Uhv9Zvl8CU47J0E*snkpMr3dwxi5}mx8i=5y&NiCJ<U2
zD1+4G8~MlMqZEL(`OQksLi*+3HKVwLMhKGu%;*2Y4-|~E2hl)k$K<WIl}HuxWv0$o
zuHU0gZo3mJizl*f_-@VIQ<c|2%X+pjQ&wx-1?5N3xj%s`yAOHN@H6nF+?u?&jXau-
z0_$G?cXZt2KRmL2DY^v%34ar@=u<3Atg*}-;$;_y-e!Y=V%ygE56_d^x<;0IMT(NA
zCQN1Hxc+RqXT;fUU1vOfb{CS6`8wmZ#nGiYx0sZz{%N#6Qbshl{AOWFG8n3Zc%mYm
zqE_H|{9i{mToq5}qsb3{;EcB=x8@&!Zr6FFlbEvbz=>wU5n(xWTC}LHYo~qn_`O(&
zDKNUfVx%D2;kt+EbP!6dg^N{se|gc2Ub5)a)PnUFEHSkkf3Atga}(z;FcawVVtOco
zwZnuQrak_V8!^^#pqWHoU*iR`vP(Pl$AR#P<h<!u&{jR^rS;;MuUdD31Jo|GI<eth
z?2kc=6Sm+QnCBY*w`M~<A;{JK%C&nA03myTo(+%E?0Y*pAY71OJcK~>O*G!#r4tik
z_1&91KeScA6H3s(oB$^@-LtY_Ff0ULB{KuK10=vFY!@YVzOha>bTKhAlRF3+ZbdE)
zqC=@0;(qz|tMmJx)P_1($G#nZe9490^v>L9lxJbCPc`>neGa3S%|3x_9mlu3*gc9-
z5Gi3ctCH@W-PDuRT=h(56Hm`B;5W-sA9x)vDw0tBt)%nei4(ar9fwT4Q7;H81zxdS
z6->ZxTJ&2&h-<H8aCDmRMBtfYR+02IGLbB(<XcJbe>G}=s0p1HSFP+ILh3-_9vC%X
zDLTKd4duL1$>#6sRch(KtM?!JE5N5l9p#8N3L7~gG9@xftG!tmUu!tX0UN6mqeEqA
zGaB<$gZedhX)s>3#OE{`z}i>CS{e84{u1uD(aDFOu7|9uUVb8Pdjb0BkkCrWYo}6?
z&jZ{cifCG8-roD^Fb;Zeh9kqXpE^bGpF3c>c=_i_DV>5%+<t9!PLJD=)ZY6vG_Ln}
zbEg<hpqRZU!y0?nd-u<MIWMi&T4;HHtd4DlO=x~9!TI2{vXX&iOWvAUS7!#3WnFA+
ztj)hF1blvWW9VLVJ8aCAr%W1u|HvHxed9R9OyjDKwoYtkYWC)gipl_Q`%m$2&!`%p
zQS)+knC9rSqkOv+3J3;M_FPu@dZ(Z{l3>&R(wU1+cKeSPKu3{K!ffehDZg%&bHAD0
zJIULzSY>*Ggv;#hS4-e6G|uz#<L)u~lL?OcqLfA-93oNUoRk}&=YEj(=$84!=PXBF
zQ5K!pCS#{GleNV$^`GL`^JIPk{r*@843ZG;a_6|R+Yj<0L0UwYs^9K@x)(iIwu;%Q
zVxF3`x5QG17EPy(eh((8sJP%6NPjV$J26E+Z+_Nmsb$Es%21%Pjg{D1;Lh39@A_AV
z{kmNxxsj9_SFOHEo=`nocLjccnj(0(peKCL^=EJDcd|T3XsA7#x#9IjTtB!J8mHS_
z>>mshpV8c&i7)=7oL>0FG=}{AxjY0eFCr<*D(aRIncID*^9ZZ4`^m$C-5=>Or|QY;
z7Ehl2y47Np%*GzFCO!9&R7KDCjb_iOIwVy$2C<&N*adfctda8WZ^C|*mO-cXT1<du
zNm#vJ@^Wh%Xo@3|7@TyB{X#m6GcWVvN9b9SUslmLNK6vueJlhLJEs=(nHzs%FSLUd
zpoH#}EZBnq$cZvuD;_n&l$4M=+3RaI%ah-9Dtw}4@q=4z9X}|Tr}O~C)&<vpoQ{2X
zc`U}@^Co$9&`ZQJ$@AJ3GPjs*j!v`55SIt((gTP(xVwt8=%cf&W!~%bS|K}IfM@5s
zrtciw{bHtwVuS3Avg_`T^q8c(14q7%kT`929dllyt-Y{0r9Ywf-uY6;k<ktZ{S48R
z*3XNy<j1&5Co>7gcFWwIEgX)A#1RQiTB|&tawQ>X>eoYSZ!QL~aY{yt>BGDa9hRDl
zBvGu~QV~*NLiNfzAxnzfQaABjX$f0i<d&X2e|;x>pr`>NPr>`b^f#Cg4&G(FmNB~A
zTBRbq`CYjzvnaP<F}*BhBw@|G*|zuj@77n)O>jB^mlcX1zjOcH@@o}XqcuyZ;w3}O
zMh318z)idIQ&ea^?E7odB%2d=r8F9%?`|KN7FoVy8){Q)1Q&hW1*Ozt!}UJri*S_)
zfD`M=qi%J|@!fl?PHi7=UI!b?p7uJLhoEGC%$klGoBLW@wR7{f<vsH@vlPGFthVBc
zAd)M*Wt_(3Cj$l2hh2B=4|nq6pLwYBod0PKbT5J9@#bB%@hO84DZwSVO>o89PN@0X
zk~~3BiE;ZrJ>_q)d8guJ1k^M2NVCg-LtJ|okzwRR*t9wvk9RN|B;WDF8#+kM`I$9j
zHi^$!TevX!;${VlOphFm_xEVbvrY<?rMd_a+8_UDfm6f>U|Ew<kW$KjD*9we1?*ex
zY~HpDNlP&W$>90wz?e^@d8d6M{^nNPN7KrUZ!ZrDtGX7-Hi;T`2lz(vy=UFo6pZrl
zU~zi6%)3s<@T`4(GF?eYyeCa6RLgX>w}MKZkrz;rN=jOCxl*>K*T%-Di865*IJR;d
z+<)k`u)9V7><2$eLX4Jzb5|&5%0mFx#@>sb7(2FgJfwQOK~;XW+r)Z*3Cmb=Bv)HY
zON%IYy%b*DaC-L5O8d^z-}Tv%OV?NKhBoQDXzt_pX7cB|`}2*|RfK1zon!WQ2Bt(C
zTCC|!_s|JQ181OC3sCVpuBg>-dSbGEXRu#$G~@Y%YTVcr+()^>@Q-Y#g0!?tCN?Y=
zB8Vu@Y$$@*j!D;MNHDBL&2GvIWFzz+tXxbtPp<wHRns#xbbLdAUM7AOru*A8llI=u
z$@0<#Mvb<uE%h#Y03c-m?Kfk8v30KjTZJG?NR>Zd`jf%7ZBi)8tnqd0x03SV?j!$5
z+3J??ayrx_jCu<1LA%*PtEDm4?iDmjxrM+@JuW8i?WDFa7zMZyJGV-C{!;4`NO%f7
z-LPL{(^EO?IR!prR(~hNXz;@6>><4-u_wOW(s-7TG4e^&&8P5E)NKE!^0z0u^0_(v
zbkb{b<Ecm|vMzniMoQzMf5tP_Qxr7WPB{SZsy0}C;4c_YOmZ|T9~k^K3Y7&Wlw1FI
zMMXutWgx=;VMF*o0{Z^w_q%^SS`otQVok`e-5;6f0OZiL{Q*M$S{@}yRopc>1KqC9
zArZQ8V!ZEkytf7O$_$H~21JF|WQ?@qOYD#CwsUq0=+w(-xcx@}cX{VxX<3sdOw<WU
z#!K1@4;CYY8Rq0K5Fl{KclYB1ye;wC9?f)iAZ(?5-PNIGRI0#sa^O!Q1o(#!)Camj
z7m)E5JLf*$8|f{NOzV(r&sHZmx1`uJu9eW$o=#*|577(?exP-bIPc{q57M2dB6{WG
zI*pEgS{RKDA0d>0x_t~*z#a>r(f75h=>d`?E+Qf#dtdyfTA$jJE4mHQRV6(Zu6NFr
z-~IF;Qx63W7n&YKTS{wh4xi@92gAU;Ezg7Yd)kR87yN6TQxyIhyD3~74Iw$&_S(vg
zG0q}^sMM|gJh2{|XR2Q(s`V|ZH+f4xbjJa63eH2*O~wtU2Z<HM2XLu_Jb=+jr@lrY
z$2B|_FWH8BNUE$C!LAp<3s%kd?PL9bK~SudDSdfdahO~DTitaHXr(&AU@kqH>l%=s
zJ00$h?x*A`p!;*;@d?dk5g58ech@)q2(x{fdRgVP{z7x#C)_3+EmZo1X^KK)#+$wr
zpt5FNvqjc_7*?ob({!VrnwI|!B+Y+$nu<<?=I%=8s3HMzSZgp2Yvh>b4$9T&5txkL
z4xu9ri`VR#^RS}wgc{HlVZbLGCKSeO(%-8#xVuBT7!`wLDNj~$4=?`+2gM2gCgGwl
z=EJpyte-Yof=T>^F?tCITe{nj_zK{?bdR=Jj!uVd3vDQNR!3afy=eaKl;_3!4$>R9
zfDc5`7d%z%TsIg@Q)D$%>F4q!_a{EB70h*VV4aK%>phg@{{Xr->MrX^!AZQ4?|0Yk
zu@vJIWszlKg}*Kbu*YTOx+|W}RW`UdTIg$52&<`NPglgqoc6aTyaGKCns(@3S7C!)
zAao|4jzsxZ2bvR6{499R+hNhq(rep7bVUOGgtXnhmi+eG>;0DrTUH(^U9{8er~GX4
zBAD06qlB!#=(7o34!uyP(KOAZ)BCD^t8=zMvv!jNu7A&zaqC`|pE)9JCVN4&y97X4
zoiAA)nh-n9k(0O5WR-HEbG5Fj-s?)sX^aQS6^y5P?+IIn)B!>LJsTN!dPlj?vi?gS
zy*<bUCQ~OwL85V<W$ZV5wQ~9sh9Y3%?$`f-k~|0`Hy7b$kS4E~*InH1t3J1VwA;5K
zIXIQiA?MsK?+%p*5zmO$hjI68jL}+eE;TSIb}QQV&csrm$X#!1V9Ie~4cIzAWlT_z
zM;&>+AS<JdBcJHeCd|Mo%B^NRyBnSUgoZ>`3YuSr4B(_YlEnX%_+SU*|3DqRBH*fM
zGF`G+y5>rGR!z2yiBh1+i%BF>%Y$!N$5kRV*VP7JM%5_l4+6mjwvcwiv2xdUH+VN|
z-2+zA;CYR?Sfd*DYYr3lV@{I@t*m`U@vTiikNq8=oILdcnj4PCP6XJl)baSF@BS`s
z-=Gso=z$DMNyazmmFOA#bstj5URF_2af!JwlQE2FC{ALEw%3dHHwHPeEC8guXmkMb
zC0myDCPfZ{N7MZy4L9p?pNb+wxQu<#Y=Jwm6$NHG`eHHPS_B_t&mX5G=By1;zh<%U
z+rH^@r7Fo47lLUMC%WUu<4Vh}N`dB~>`}d-*A>Z+Sg`|;jgzey`j6n8<YmuN@UXNZ
z35wqMkrdMss0rPHcRlCBN*-L*S28_i0mO$wcK@;&=|xbNMK;2GIROFsWvsF6&U*-<
z0u76ganPUjgovF8d%n(k!Uw27c9bM3k`D3ajuXStPgqm*NG3b7nJj<KhA$%7c_z&)
z-=WOVeFJ-1o>0+b64<wQ2KNIsW7G)I!Rf8X8Rml>S=h;GA2LbR{`2>&!BcOh5L)D9
zV#@8R-qZ0iQL(S9MbxIeYZ}`U!aBx2^8TG^bmTO=;774HPVK%VODV={sbdb9bmtj+
zGd;~}YK*8;&Xa-Zr8tYP_y?BT=}AYMmwC1{P1f%r2yEErO8Nh(ZGL(b!?ybEew?>v
zdhpWyT<|tL$8Nf<CLYk3cOKgnS)q#Pnq=9mhE4{h`w79d?U<q?)PHs_EE7TouLvXw
z-2D+!b+&d-eWSgnzA^po#zdKw-&S9yMacN=sCSSGXZ}!Vt*e_|_-7u#O$~SLWydSP
zjt(03di@QY-q!q#XEM%{=EPnq2g}t-D?~3C_M?+6WC3J7Hg-ybTe30h%_n!VO(-p|
zkwD9Bq)l;mNY@$sP-TUq7)XNM;Bs&^rK-P$hg@IF^e0MwA)PkhEcH<mkIM$-W?+vf
zDi0`d?Gh|F2}Se+#)k`vnHd=mL&r}gdTlhs+>1*6Z~eIZQC5EuEH=`SZ)3)x-%~gW
zXTaCWXxnYdb2X^gkXWfIGx&MX&c0|v&<CDj8S1~UkdH<wHDjxU-alp0lE^pG;rIY;
zHli{k>1Ynw6KQt-e{>M>I~W%=0R#x{Y9`W`QZQuD2+XG(8ocy_Odo>PZ#YO-Ee(Q9
zrxPDa@eTQ|f}^2hY|>f_=kDU<BW9Q_fr@{%C#W+3k_Jvoaa;%ua(zTquZ?Gi>7rVi
zRx?Q7o_%0@ZgNYzDu&1Jf6)bIk?_B{B-fX_hBRPm`REEbD5Ws-em8%@1IVBe_?q#A
zR487YSX*1eW|L$X`Fqrp3VX>L_;6<y9NN99qOn8W);Vo2lo8x1V8!A@1B>I$6T-ko
zVZOn5?V|pA7qRP4*(2C>0b1e18OvHJeIzp9@qDhYW;@L48u*%ETAd7I?j7(T(v&za
zd;6Th^nZyr(vPsOO9pZcB~g!aG<QMkfru0I2twp>xr7zFr}s`g>}jHqqq@7%;HhVE
zV~T~a_Pq$f_mAe;>i`-0x&egE8dqfZbm2jAX1DbiOBS1g<V-)z2iJ7<BEn{nw>{`r
zrO&(_D>;SC>RbcZVx&>^>doZb!lmCT!dU<TK5WiUdW864NVOF>DDr#kK2572l$%yN
zUl5l)a#LID=`Q93oza*rC24YzD(}kc`O$g}1I-bC<i7r!WA6Ni-f8K?>ePx8nfa-f
zz-bNBefso{WhBv05R7iQr7l=w@T0xQsa|M$BAJg)0!QSHij!u)F0qkc?R}mDOXb$J
zlGC-S{CUFu{)InRSR0}~7BEXV^5;fGe}pWRNRXEO0z+W>7r&0{=*@MbH@Pk4LiYlw
zPrO-Ue@@m;ynD0nBsWbDG=6$9d4}50SGehgh&Ggm2|GDrVAO%Rc*BCYts^okDF;}I
zUuu}eo51T<h+#T`#z#nJeh)UPuy1?z&cv^Zzt=S*1}~Ub^k6|Y7v|7^kMM)mvwqyX
zDFe{)zp-pa6Fj5oen-vzBlk3J_^UgvIyp;-7+CdG<ecwt?&@%AsTdB_vxG#!$hy|9
z+e{?(t?Z%Q6ZHz|@TU^J$FlyhOAm$IwKR7A^rLx~tnBP4{pHbE;u46Z@_Cb%lH7_d
z@r2r~1P-Yq#Q@;W4`;XpGKq0tD31o#eE%?_zxZ0x%3Xh4=LDoEg)n61=rPLSMGkyW
z4ZBj7aICoJT@6Y44oYoElplWKR=mZw>n{nt1VB%AdI`94Lr$#Mm8FbRE|GKd@%?a7
z5vKO^+TrZ}uN!MEMUGfL$=|5-C)_*+*3;KAggAaCunj2C(3+cXNTtlNIKuGpgN6dH
zcBO9>e=ECB-;8Zb-dj-yi>6X>nz%h=1EM3NAHp9-?-bk_OtbBiKAvyC(qIyj*p^tE
zn=$98-2^Ab$Mb0Km7UPE`)b&Cz|u1}H(o_9QJ;{PF3*2Fa$4rRUcB2lUF2FkQ>j%_
z3W>UC-vmX`)J)x1p+q;j8+mKO^qkEtE;BvZ-;N(3bTDVTH56Q6(Pz9OV1LqiprfY9
zg#9C1m}mF9b&+E+6x+9ZhIouR*~vCtOu9F6IW{(jTbO3u7nnSSOQ}U3uGW4X7;@ZN
z9C1$(`P^~RZooiSCGT09fNPs&WsfRb9j0qry~(a9{rkP*u;e3z+vLlu)9*^It2P%o
zrZX9%8ehk6e9!(Y+_z@Z26@ruBb7@vEce2d7TZWeJ{}IUsmJN(rAeTdEt!<pZ4o4i
zIH5G%F6K_@noc&kBsb%T$<uK^dg;;q)|h-9@u$gBJ)m`ZiKs}(55DR!Bd!0|uMQ^a
zhs{Q6x>8<Na~(U;Ek3W<7pgazd@7J#O{V#Xadpy%43nJ%?j~{-j*h*@;R!1~D2D+7
zTWYletF>jTE0yX-Dp^Wfw(Djc4-B0QbdqSZgo{2`*YsD8SkmOBYjv#YPIP_^nD5D>
zGg=Q|C^9ta5%>xNVe!SB-J6e{E04<W&K;^NODURiIKf9FvsMG3<pV3rl?&y7Dw_?d
zfVVcvlHluRyKfh<Jg#QjG7o-^<AeVUdhXO%_UoyVah?VEgQyddbaJ{ARWFUDn%qHB
zxv*I)vx^Jy5ezA>HRLff+gme5Y#l?JQf6jmi+vzs>CYF4FRB92IQL`gf4ChBP6nUe
z-2Kj8f#YD}qXmVsTU7f}VlE;z3XQd?@^?OF!crn^)zA%`GM9lhpdsG|u(0CiO;+p#
zn0BaJ^?k~5lfS&4p-+R}IT4!q%pJEB!gB?tv$cv4eW)@z*|ffFeyEB(dv!S2#<`6Q
znwSpOTBWZaiszrwKSR+i-rr-v-N?J+k_AHAA}SVLo<ZDMXs3uzT6aua4;6Kabf*B}
z^w0OjFIejRP34kh1(@nm)_*@mrF?$&^$@w10{PL4y-WW>_nXv;f7ee1yfjYHX#_vE
z*-Xn<!~D@o?6-mU+dNYx;4?xO8B$jJk#AuH#E`l==9b@q5|J{R(I~mzGN};1hsRO@
z<WNE71*YTJSY8*4YCP!Vph9T<8C$aa%tu+zHf#qD#h?Q@<<I~<_dZ`roBA$NtNwcX
z8l#8b{}T7^xD*R?&v|SkO1;MS2BhivK-D8t1&32jkdjl2!0cPgpCH(@Q(*O;>Sr#8
z?yyX@e{QUsONA_ET0lpa=MVLaA|C0ia-(>eaX8y2RJMu2q|k~|UbYpDv;f|*!`2qh
z77MivqjVng<R!mDIPaBHEh#;V$yKQTXNoBj37ct-ay|Z;3#8S}jP`-pXZiCUgJ9{j
znvyZ3I%RJJW@ts~MM93J-StM~r{+rHG;Skyp_oL$<hk7xWl$OF)N8ec?Z{;aD653%
zHrx-te)%Pct2N{tA<TQnJDAwZO0z!rrGJH$Oefx#c(UVYQL&)!@?)_oA09Lr$tzys
zwbnVZBwg%1s0%s6y^ePkI6}C&d+{-yk8@D~sai>l92@6m?-s&JN44Xl-+nkUJ;(c+
zcb|PkWJqjAI#AzbS5!OBvd5WkMgR3y@REe2s*%s=3j%th=}BH4e_A#zGT9$G;E!Sz
zQ_p+yECX^5PnYApjy<1d-w`zC61YXx%4Mr`X^U&z36qd1I^mq1NHYbw(PxwEHD{&5
zH}X*PU%YYPLHs8GJGBLuuxGl0+PM1ObZ0bQ`fTn4KB&8@8Xt;qoAzXzXkD=wZ&H`n
z`gFzPaQA<+evrLKUs`%Q+?i$6b1q+Z(1^7?p@uMniNf)9M|>3N$sGd$aPm_!_6^HX
zz|Q-7d?$Q+;cPekw85tu>APYT0jP=Bzn_8?Xd_HcPAGvv24WrCm=Y3BUoHj1%&KE6
zzvg9o^vLz{&4~bYPf>xf80)o-@_q(?@|N9euAL1*_>Hdc@O<UHmSuwAvSVP+v_4Fu
zLF??|u3!rujjJx$YgQ-3;`c{@!V7$pBB}S_*!vL7UE@eJE}7wHmF6{k7_r_vAXHy(
z3Brv#MaABJ_3xAVTR7s1q^T!Apv<;O)28f}#w|qt!7cp{(T}Od(S`x@hBuCDmGw-4
zVOQ?c_6*(a2W3q6FEv3G>9FO-Ss3v4VIEfnq@TBumVv#ZN5r3r8Ax$$354PG8dZ86
z66Z=g>a<A2+g>I(*T{SwkDAu{k<HwF>KQqgz$5W)V)dy%fKv@XkD6~n$%4H2CG{bN
z*!D+hqbKfYpTlA0L^QBDX9w@#23R$rG(dK)P8>|%q%mktiWmc>AzfJS<WG4V9pIId
z1$rRWCX3V2aU=JkkMSN15`}|9-)ZkQJ{Se?PE0@1;gH8hLq5v*q~nTDkDC9AGXt&m
z<x>=fUM?7&l|~tKufldITg-w$yQF;~{%dZ-ceYo2+u8Z^KTHFp^PMN_*dQ({0LJ(g
zue(lO#VHEOn^8+Z?P%BOTCVn3UV5JwpL)@%4|v2UD^>d#*sWpj)JyPoEs`@Bd}x~m
zSzp*12(B<S2IwjFPZ1GJbYS@G40DVfw=x>VkM8DqVo*7$-Iahr0LYnQB{_6;IC3H~
zKI1E>6g@yeMRdq!IrHbtZ03Sg+0XBQ-!4ox#?^BED<3C}eh{92F$<LR=|j%BTNxKF
z!bz!hkzhKI!Qi)<J|v+4sD;0FQNs$cFo=aY=)ku5^xXSXmQ20w`|_T<3OwA}zR2<8
zj|30I2OF}!F3Jk0k*9_6@SPty%#Yh|oV~0PUP4G=?N(euU7t3=TL3=Hg!vNTIZhjR
z7;t&(+WPvM?0<>{!pD*twSpM?K?JA`$SGlcSO*tf)I>_rRxtH6k3|<^*C^VBNLN9w
z=}bvh+Y<@lR8<4BO#FR(w<!jCKz8Gq|2{}#FBLEr*eAMe`sAm+!x8N0VLxi?alaCE
zxv!!H+7jV*4wu@Kj5hKZ0YA#Vs-KzUg8GQEW>lvALs7hw2QjBzM|ul;eu0QJM34);
zM&X6pe(R1@vsK=bnhJ-3)9J?!>!k!kGd8B}Q?iyk15`0pF61X2oys4$?$rT#Ba(XA
z`|7w)JAl_0YaY=NTHaPxaA5g0v$O<y);OZCLc9K0Gkv4d80Aw#J$Ha<{b-Yq->+)%
zSd>4jh5c5zbH9^p)e!p9j-RmVG^rA3<dJ%&=D0dEiny(ch3a(|{zk$JL&TOymhgv`
zlyc9L>fHqe0WE9Ie`7WZTxl-dJL&BeNx$=fSJC@UZVYiV{ot53k*=Y<lU1BLq9$Rr
z@5_uWwzgu3hGT1~`Bwf2Xt`#k8w9=BB|ilrzubEP0J2@F*laAyCc;oDb^Sc!9MYjZ
z=~8{$woLR1=ObYJ6?NxY=r0pbXxy4{+Foxd0(-y0&o8TA?N*?S(wh6ot=JAF3Shr9
zSezJ}<<>dOrqTX>WT16~QJMiA=P!EJVBR2tzxDhHZR^%=g44A?p*1mj`0gLQyS5|N
zEs*_IK_kMzm-6ywU?IMn23HFbtr9LP{>g5kNyMD(=`iuc&FdqO`<|_;?c%SOqr{^=
zOqf!X`_?>~a%c5Xl~CwNt>TM?VNpc`%BTFclZ&$!8@~%W=vIOupb}l%66-Td5L5ah
zPdnQ6nSyjOlf|%80x%Rd5JyPmqt4GA6Z=0-8L%bp0D{Gt!_D5gDeNl78c`t*7lAPF
zC%b?H<{xbVmQI`2-w${OkO?l<T@a-=DgDK9{rApzTI`l!K-4xO4`1NGBvXR%aiQGI
z^<pidkj_KeU~jJj7Ap5?*xiLX+m_phM0$(&O+;losFu9^<@Vi)Ms#F=B|;2$Y%KQ`
zrAIp)7{WmdaM$XGQN=CuWgfGDdQ{4xT%cRV;PTIkQRofK2pM(fr`bd(K@;UfW1Q4>
zdVW)fLI?!t)koP*ba`%fR2;5iME412ZQto5`Lln~HHdrbG45yBO^3Qc5ahwYdm;LX
zoBOL8&kVxJ*Tu6VQ#l34?N^_+15cvd3FtqG_D|!z!6WKzS)?7Kk&X%0B#P@yudRO?
zkvy&rj!#7?W#{n$z@P|0!>7v`WI1dG*BZPk`Yq4tL90g{k>-$zE?}k*#~n}m!PAO^
zDW=rByY5cxbC<k|kF?6TfGbIqpXRn6t|i=<@PqC!3_3=;sI!JDq^8I|8aaF3b*jUI
zU>YiH`zz;HD|d!D?w}vY7Ykp*gK0t=ETE({S~@7%_D1s{rd#tY=JNdymM)E?ZYy?D
zL08697z_ix9YEaANR05>0=_l)?v8<kG!NR?rtQqQ18aGq&iq&F`96aUp2EE85+Erb
zIWybzWz>!6-d%no%A>9NW=&0I`VSRSNa?Q{Kl^77yA3o3)+!JSwmsWEr<(df-Y)O&
zOwb#scHac?D9pB=>#vN6$%u;(ak(8AV%csa8&sHa`Ha_6n82}Mj-j2<t$NOD%;}v)
zL~t^0yFDVP%*)0cweQrS3=S^^?z9&w`+b}KH}3{b20iAhR(Qzeq;L52>58`XbLF?B
zRC}Ws_P)Cl`rM9=`lAn`6Vw3^F<Apz?BmeHIGGKa5y~&hs#MN)wOwby0BhmX@0Rzu
zh*^)1@l4Y&Bg1t!&m}W~?mN3vO^(CDWuXUfw%vD<P}@;Xe0+n2=fU#{PN*=EK{q|a
z{CuLY!wegq0EF7N+r1u;G)VO*tCSNC-eSBK{i*8S=W(R94jQlQ?jfsbdEEFQ{!<i=
zI@Senk75{feOg+9nCs-?+#8r5hxjDhZ-61jcxX#bmx9$wQx(>_r~dyqkiifHxTB<g
zT>FV@{mRL~)!E&CarS6lwxXQ84!+*$*`NLW%3t{s+#PIhU%2(14iH+PzYHcUHZeDT
z1v+-~BbAX59CV{hVO&sE!k&mmVp03)-k_EfESFT7kYQ6mztK`O4@nRjsR_yJ0cMW9
zT<lxovR7e(OFm~La<A%JSj2|PMBP>u(kT9~<CpmXDO@RreKV@(0Q#rxAICYFG|6gs
zYH<b~{Q<Ub7tVfb#b;$q)llxDWvg@N;9^geDsq>ZF-)5X2_rIcY!A>p5VH$n{y7ik
zsBrZn$10$Ae>~1#$Lt?b_UKa3Ma>dz&Fj{&mL^VMb|z#FS_JZ+L(ztc<}2Hb{OgnH
zMPIv^u!t2l)B-z_0%{COuEPzukM*MgzngBYx-=)XF-I6w2}xLqh>xQb2gkHx!C`mb
z+i+!-&vbVKr`zeC*0ViMEq}J(C_mcMX=6Zr5=fgZq+ahsVLr}p5pV@r4t?i)AFfM7
z$b|j1V?w(UPHzYsYbgyDnp>^a?ZRHbDg#imK>y*+jL{=Zbw%qsQ~jZD0=@9jHAJ7<
zC<^RW+iNnLS%tBOz-_7{Yo;0@_dQlG7WA5|O(VuubZ;Nd<P+~)!j>f9u+-}eQ$ozt
z@{SK)F$Q0SdT`t0tf}Of9x7YYtl!$l7g8hqfq{p>wFc%BaBj>!=Okpl8D3Z@bU*Cq
zdkO{Jor8Lw$~axX+2@QlHul!3p>xUYHST$SD|bjPm;+bCJ0FU*Qm91FvYl>YmI#kX
zrkd(9x^M6R85TwIPX?kX3xyETS95rg_PLO`Dq2{@wgwv&uEqmjg1#y6>s|fJ0MhkP
zyOl-9$S>JT3U4mjRT|C5uN5D_5mj*M4>AZe%k3@&?(C<)0TsT?^!Q}a+NA=P%8#fs
z5Ks0A=3H5mJ>zy?t_**9riacViL-3UYUE(a@lfO%^+dIPE2sRqDDE$gEFic=g2eDu
zBi7$X_H1vEfA0CJhb7R3a_INHFR6^nDnX1;m6xq9*Jqg2-IXYx^tuNeTtyWfdS8*Y
zepboC>ZTXwgRdR=5M+*f8Q2H32A3u)P6FKZXoAkoB{vx384ak_$rm3w{4)~O*jOdz
zuFNhj;Rmq<Y(A6`+?V<vgcvV_eQwfuPaP{s>{WoKe{5?y^6;0&DPS#IjlZR4TfS#@
zE8=ySAMM1*X07Yl(3M}|*I&{Mc15^NV9fH|&=!BB;XiW$q4?u-_8*$@VsC(O5m#sM
zSGyF&{lMK|S7mTlN)x(8bT9J^-Vh>zU!3k*ugN|-LNo{j?08JH5IN=N+wg@3AA-%|
z2<Lx8*jo@S3G@Ko9xLFN7X7Xm&e27jnMZ;Dy4{jNLodW_KAWN2csq`JTWm<R@D>$F
zm~O6g;7y7}>@fjE(IU4yL3}Rmecz=-4(Y|&)}lKP)aGHnY73qe2U}>?M0yRyo5)a7
z=cjkVs?(?-<g3yBW5rO6S$LWeMkc^=3Hx;cb&xphlGg32_Dp>j^YFWmZir0xsHQ)W
zcr>3Ne+u#Rz<NHaeg47AfxZfA4J!Sdey@)^9RO3NUjN94;_0Vwca{w;rdADjgDLoU
zV-$qEoD-oBWpghwv{imlKu&5@_GbL|?-fav<mdGLqHulzJ7EYHJsOy1VK#=@a-Yk|
zLMDc`-yBsCL_~SP6PR4j|N6*nGR((y326Lq3bgR6e6E2$3+TqC1E1(W*zJjWA=2N=
zR};i=!*2wzYo%3pDXa%HVsJV?7KiKjiwaWu`@K5Caxwq>mW>vm3!L}l8oU$UK(R75
z;KOz*0h^lM`2y97w~nYGa=@hT0zE=ZZ(QFBzvM;FXOK&;3+FY8s<k7AWM(kK^n`Ar
z6{sTAALKS26nxEV@`==FI9pJ{?Cc(^U>n5pF6DBHA+qm(#-OwY6amhu3RE<2$5&IM
zb}+8~k)eH|2Wd?M&{U9+Fy@pL;y*{oEYHYJXF|@L;B%T2`PKLUc(Sr{7P!R5^gNbE
z>IiA1rp`n_o6}OWViYXSf{|9q5i|Y`k5;UlMNIq#xA-99V<2aJSv2DH`J;PWwa}Kt
z*4Z7B=QOdw1ngrhz!4{><3zdRC6df|G`L%yP<fbrv1%#d8HV7%V&@Uk$pV(-C{h>@
zsZRY5QxCZ1zo3*>@>XAvaI838=_f#Z<ug-_^VRJiPl1Q<>^BoA7=CTGdwnMpFhGik
zC%+wUKmK!zz-@%G!#Cab3|MA##i{p{k37{|o8N9~zLl1ADqSx90b%sr@ce=P7iIMU
zSBBVI#z`i7n&<?c0=LiUwTpL5E9!sv=n#3jsjh;L2uC-V1X`C8g6G3kWU|kek+@Q|
zl*@WGDdidQm1otIcTsp)P$pYFmwR8r9W*K!G^`I=o;?2}yg+|M%nW>76>!YK)-$tE
z-|hEYZ@{J|sDAv{9QGIH?QHvFS(J%LXNs73BEYIa>K*A3FNJ+4sY8DU`e2*4<~PB?
z;Pit>*c+=FC{|O!f=2yDVU32O&K7>3!5DV?o1vBOOESc#R7!K7O$xJruEw_)_mPHD
z_uZN_x*P_W4JTv0@I$`&Xw^|Cs#&vjPpZN{SpvoAyXVfALu49pK623y!xOgs9gS!)
zzEY3Y8nf0*IfiH36r;$W%^oAkt240<E-svcSOZ+k@~>s!ItG4Ro%`o*ARmXm25N&L
z%@55ux@xXc@5_@Ly|o9(Pjr?=MHj6Do%9ye&E;WdGri~IoDtmZOZ2c}bv}|;HqQ>b
zyCY>dc7NfzS(Bm9gS+xZ;<#~FR!s3pu##V{Lf}_?E{@sK{t_K7wa9tP`v7JlzCg}A
zc0}YUAIOezJys3AF1bs>`}<{EU5lY$P0K}cNddm0Ix;-XeXB*zb|zg`S3kZn^5;;s
z9qi08b2)P=knT)ct<d_-H~=7x^a5T)U?UhV8jZe!s~`l`xZ|38J$x*HbK6Spt^LL`
znltxF0}m6D-^jh7L>KZNtfd=RAhhw6E_VR>=H&S=O$$nc7B4Q$aSc%?S(*Iqdxi&n
zz|k$xvcBtCc97j7!vsxWO%B~|d7_cr!wLEj9fz3Db4uFm<Q%YQPvhXBrd}PvwP3!S
zkyf;79iD#axp@OjkI2!!B$IignZAmQn;f*QMNayYM?p{U>!0ild`oCfs26D?kT_}H
z)u~xaJV;;Hp4lvTPmwimR3T}HJxmKasjzTU?_)9eXGK9$yP5q%_h=}v3rlJ#^>g@{
z&jjEW3#K5bzx~3WK8}mGQxvIgb^>0nDB>fFwK!=dyJh41pT8ChL8-2cx@)>0*IGEi
zZ$6q1iBJQ(w3yj^W~Ja{le}V5nmZPd4LJgwNc`acS5Uy!>RjBKrd4#5N4B+0*(HnU
z8W<}TjMRp0r#<UA+cyL9{Vl|F?(UtKSiV=hx(`n{imItLG9TI<Q2HDdhag#0!8khs
z@J|5Gx#I^I)?*ChqKNUvXq8ewO;?Sj+#parpK7#CD&zUOb)<XYyGXEvYM@Kup6#)-
z+dywjlU0O(&j-J#k4-CHMFj5NUnb@eu500NkiKNsW@1B^2ow#AbC41zjV0e$-rfE4
zh$PN&L~RKBqh3<9pvUHXjHXEMDKb2KJrKt^E&{E@-1=GiMcW2~*H|W1$;<Mva$r*J
za8LhkduU%sG8#lm(*bu&2XRuQ4O(A7dhW%@zKgp192z3W9kT(3W{#F^qz6k`&o1m1
z;Fc)q-F+m&|LYI{(?h$<3f6nICh0=l$K-CkB`Aq2JF(JVNkz-Jl!zr4dNejQncw>d
z2#(6de<a+%p*j?Owgc#96LX-9wvn;rN>+^p<f^B|mKJky>p9}rt0dt3_#HjZKG%}q
z%Y|O=3?A>H0HSn|SN1N52m6eST6hvKg$-#j@2V3_8_v#_HjKWDEH@tu&Srhm2~wJ=
zW~=*?ZxAyqzJF;L|3h06`dXP70mb&_N*+V~Dp?NT2pF5K)e&K(d(PWo1?&I$Onf4D
zc6wrWGE-Igiy~VeU69GQTVMCRJpMK9f=addjY}F0y|29n&+yuKliwgB^5UfJr^sIY
zS$65bI}xM48&6N}yJ3+3h4kq0Nx#R+9Nw1AmcyYeR23hVVWdD{qoeH!;r5UP;hFXA
z?e4kI=0K~hh1FW8EelMKlau%P=tt8eF#2o)?!|lo>Iy1eq7-W?tErtT86v=&=cLo5
znpX(1;Sz=ukF~mC%sEsygA5dz?C{!XvY(j^HMb?BmaAz3#=5&{92w>s=iJ=agG1cr
ze5l_Dn8<rqGX|$CvQlGEUvkW$trQ*(DgM%h{0UVD!y04^s5jg|cgKVC6wqmDX-1*o
zu-c*L&Z+lsBTva3C|mi$Tig8hv?)a3>RB+iv0XqUqNDfI!^0zRdWy3eQYaSV=O}wm
z;rC%o2cvucxZHz(Lw<JfTWv2Kns7%gJbyN8n<?=&&o7oDDz0<k#lJ|%2Pm7hNR)ED
zM;I^iT%mq9JuzV?b@3k>=>O1sgr#;D*pBlUSbw9JJZ~`SU$_`EkY(7l7R~%$;nHWL
zf~(VvVW2k(S_rJQA6@|$ex4ZC^A!{UFHsv$0zLR%k>GkNSUJxDOee=@fY;$H6?fu`
zm7mCqb69!T^RYuY5{4bVf|-u>bCl$~kC?Rvl{K_GzX>kt)JNV@Coj7iN`juvV=$G8
zJ%#*scrUCZffL?21j8hN{ykgWBb}V40Z-vAGq4Tw9?{o03^PJbBE~1%b>ke<bM_%~
z`7?yk{_EeLJ$ojz@#G!t52MQX;X);XVP%brwPgvoP7P$9N$}n?#awat1ObIo_DZHx
ziksa&ng^~4l-b(%hUS}CSR&+z7A+<{X_lr*hUCU(W{QNV2CbI~`f0~4b&B5JWHlyd
zK5Xd&cV@T{5Es@u`_muJjK`Xtg3aW87@B`Sx;~lFzTrJ21y{qCwwZ9cwT}IAdYIr4
zXfpU>{rq}tCqCEaK(55E!ly$w@&Mq;;rjaendnQrSdZbIME)W#@SH}5S#YW4-~$vb
z<{@WFzbt#I?kQXCw~amzi1ANkF?JTfzs=3dCuu`on8)nN`gT4BW@O)&<Qw{t?UT1I
zh0g4P30mT`7z}<1o#oZ(?(SuTJWxP-{NnfhWnCR%6DM?E(@Kx;Mlv#r)Nh;r;ix(A
zi82|p=Bk~7<b)ZrjE-}fvmT$DPe0&`%lUFPHIlx&x4WZb9X%<3W`Li-zSK0tSQ1%j
zv^bc<YU}0T1TPB|#?`?GL+`7wQ5ey8+d+c8ZK<zQJ!}xRF?hDNtI%Qnt-|B&fs@-S
zA5$Er^hq9EQVbe-)AKy-wB!CC{*51_I``DR@v6lKFzugjViOqj4MJbB^64=;zJ)Ae
znNO#PqEQ+X9;2>vPFoZK@*3vGm`~+8WXl+g6YtS~i0_{lU&B{OQQPc>cVY8R`rZAr
zXU|4Ct@=ByCOuI+&MST1Ves@x8}$vV^f<B26%))dZK+X7fTRP=d843ss;|#RaD)C&
zONHtNJl75OooWt_!&lEw>+A7!s{!>nX|89a{fnnmSSQB$>+d=rR_e?<OE=zqGSR9W
zy0<6D2>2w!f8x!Nd{jHiI!}q3f_HK54HPNYX&waPqW5+53dX$xtjvopeWT*Kyj0EG
zV?PAbko<Wr_aP&6?y9ep&?<17d9c6REIlh^(sMs_-dE}|lOo@qL{B;xISYKQxMaim
zh*IWwXP4?}ot1$48=rF@`A`ImC7X3P)K_^2bbM?(eDrt_@tKbL`nlwlvRluBUdZC@
zc?~<)=P;Y`+E`qXg|)SH;rGF<4VXvCyao?Y-^9D$e&zegt#ya><)Lmaxn3xd)erZa
zB()?Wav#`hgfn(H^_!reJ`=8?;l+xcIz5rnSSBcc?^K{YF?+VzJ&TKBF$sw>cx+WZ
z-y(oUDRkt}|MD&1Z0^H@=x_fNyd`FgiluW-_D7xE98D>Am!#o|s!K>n@Pi#z9@hRY
zE&a+cXxbS*s}-!hRIXG`-rf6zuLul`$&b6@zMeBxU0H!wRq`keN#{Dsd|sq)xbs3y
zrTN&A_}ijn(`2!Tws}s&>v<<(Dd$@S)BEpk;DVth5HiX34`^a5)es<LR^J=<nb3>X
zljvO?3EO`C?RtU0jdn;fP{?qMO^ozI!EykthONXUUk7V7UAFvC<C>RM#4+7nc=jG2
zEwiF<7Ri^ozbdUO_h7F*si4DgB?eI%Ph;Kkgkb=Yghu10-w5wB<NYHyG**kJ<Ks?e
z`@_5`0{WE4LyyX+>&4M4+msko5Rria@@#jUVVALfxQ^qIhEw3}-@ji4<hu)FH>d<^
zPYrNPr?~h0YN)f?4{{VOI1H;lwHL%MEW~lD7hbcS_X*pu3go=G51)+Ujgwlor_FO0
zPF${s?k8S}iO<$gF-86iBL&sD&-_+a<=&T_1dg7A_`pX}%6sl!mc-54;00I7bH2q&
ze4YJ8Fz<@N`G(A2N6(#j23IR9et%6{%J|sW-CG>Gl8+ueIt8=Sgf~^(f6Y~?Z_RJF
z7CP9ScKJ1^G-oy|ar*RWfy;sU(R<R)0C`w3WgGe(%#Vg$6?1|ZKZsFUuOvYExar_Q
zmBbL`dk+dfQ!DXN8U8w9Rn>4lDYNPk5BBl*hFG)`V|S+`hQ<Hkyabd9h?%ex4X@xB
zsp#v|PjbJN4;1jz2hO|!iquWS>0f1DHD-MG+)ydPbMz+{RAQeEC4qNYwKMO-r>&5n
z7X*hHrPGVaze<D^>Kj$};gOuHQ%k%ldT5eNp%5Gy@?QMIq6uEPQn(zed<r;f`dgWs
zU$GwCAhp7DS&6EzRw}Bszo{%-YfdQ~J}bO>_w?(^#iIzdbDfss;fi#uo7Z0F8y9K_
z`l(m`^umiMj`D976|B>$BW3Il!cM;W-(Iwr>3rN}8qTFYddfOZxZ!qT`uAe>*=v17
z-K_6_5EvLrHV>19)z-85iS=&B+yDN|3XAtx2}*R2GP8oD{nh7@9j8{z&AH0$_3aw|
z2Ph~fY(_%j_KA?b-rlI)-tbs)R8u64a3OWAohgK@yW6_;`3w4^^z`pI3Kko9srGtX
zQk;u^PHRPyGx-)5fz2q-`6-O`3-NN*<3FK(qyFk?5oB~<lwZ=b=NBv4us*G8X~FP*
z`(K?s+~GmY7aPkqVFsA3<A#;blb^MzP>PFoAi55t0I6Hxs)h*_5E2p!V=^_F*7jGm
zTqrIs{_v10>aRT`d<fZjjhE|8%GY-EsMnC&ThwhO6t#(W|Ni|dCu^t85rOohcv?dh
z8WD5l!rNgQ|Ld&=J6?k7=<(z0Kfpz^;Ir%f!6%~#?Lg3EzpsZaaS-3kkK;xgVM`x_
z<IHSrZ6isK^K9G>)srPb?<eTrZkZb5rzqV~c%d}G5=pK#MiCcm1A>bo3GgV>$u~6A
z0U|)JGz{u(vZ(3abm8y6gtyI7*C@n5ikT)gMTUqL{^QwD%LIvSoJ>X3rCl9tbFT+z
zhqFq4$S;R&uZ9(5b(&ViEWNWvdUlI2Q7t_$)NH}L<`%8$;!<~ax81zleM$W}Gl4jg
zB=(&ng-XIVFgEhbu)&@V6d69v$GNql>g&hE7`?KKSB6LZT+cG0<w!m~ym-u6C?F*E
z{U0xY<{TKl^<$;%`<NW<DZc+bp*INW;r)dPtPc9b5z9<obK8h>T8qP!F$8staQzY&
z9r1KrjG76q;<I@-H97eXiFXq=pClyqTm$s%@G`%hD8!|W_)iYkl{wJRyAU>=qbQ=N
z&(SqAmvrwdA!bcAeafY@+S{wqr(Q>Viii<1_>I?I4f}f5u)=G?L0N#a{i_oHQsGxH
zUYp>7=wx%oTjFX3SRD$G^czl&)Udb%-XA)bvJE5YhD{|;XaWDEa}J#ByF$y`n|6D!
z&x^m|@noNz?86ZeP>hV>Zg&6u)Wq9|7L>{8FDJ);LC=w2EF;qQL~(O-bK9vef|!`1
zNajcRI*s|1A9QZfryw_@*3E1NHWK;~N%k<c-R6RVL+7_m5$F%wvlv-$M}=m)t#8wA
zZ>u*tx?fFRd=I_p&Bn&Et&8O=loz5#U>+>zn;!ALZl%kizCU*RC|+FXpsjDdUp~g{
zsBD<e&8Blq->ca$B8t_}!waNoHJ2L7*X5jhY)cv(vhiZNpuQw@GeUQ{l%cag{#5kf
z-oya+I#5;k{qE&<yfC%B{ewh6pKIaPQp!mah;d<R_lRKBB#?ZBS1YviU8~K0ftubz
zy|Z95&n8*s8U7P|doKTxKUQzOMX=v6;(rns_SD{=BimCSrA)z5RhY6hgFBao0NsZF
zqq1*$3jVc>$z2JK)PiH(d*bJxW6cL7#j3U!tT?a7#}d6DohUSfxT~+5b*(3x&CVP~
z$k&>ia5|qU0#<BFM}eKyId)7rV)i;xhPj#PS+Oa`w#&6Bb|T7dn+n{yUw$cvU&*yi
z6rsiN<?*3W8;@>bR&FlcWhVS}=AuIA$En#`kENw0)sz3ue#-FX4h&*b#I&-;C#Jca
z4rBL3dw`=P6?$P6vY#}>uHjePqqoMB2~rN@BFIQ8#kbN)bbGGJ70!LfI|V6m_Z2!9
z-`pg{t@<tjZEFrM|8Vy~5|8kW=O}JEhC^1yGvniHwZ*7CL(O+q7GnI(USK2IPt495
zpEA?YZ5SIH8-0D;(7MM!wNrR|wRU?6%*uRYz65zsKm$yjzR2cL?)jI;_u6*L5x4e+
zR%*ZB>6jNGv1^(|XUlNyvL6#nT<a7%ESMJ*YW~>!20CWbF0!zuH*V5U1(dWkRI!{f
zG(<4fl^~4yf>_xV>^2J5g$h$5iW^Htjur>D`7nWr%I7DEgZl^Ef0BPz&09X{UCX?A
ztn_%RRQuOw9ceC==H(s?r%$fcxg=k0N;@UE9DP<$S8Ai@>|21(_pI5QQ?O(tf0m5k
zfT#V|whE>~l35ys?%+{$N`J1+W<!RjCBkiC<RrRrDuUKxq?gsVxE9Qg=H!q?L!Sa{
zm|D|s2OH*Mgh6Gr8+~<k(g(OCc*-DyzQx;ZJ<!9z>V~l6orEr3GKm*)wl#oEx1|&#
zf}u?uwWpdmVJY<cud)hV3RdHi@48E|U!QRuNi+m#!&dFuQf?C92XRt3{W9oTaqz1j
zu6_KPo#dV0*_UIOr6mXUf7VH;7f5Cp+yap@5dZ$qUrk|aO>v<@+}xS{#jgg|E8LQc
zw$}Vlt|%QY5*iNBibQ<+)=vFS9_}YDH5yIzPs*;^$plfo{{au2;d$Q<zM(OVHxw1R
zotzUkbO=u-;pqYL88^pg6CRQBpSyDiGs+~|{K+TDLcf5;sldyBW<bpB#$Z2Im>3s#
zwvW9**daehaIN>CuyvjO1*q7zM`{GR4m<=SpPABN2TZgR^&Z2B=cp)1_l+{*al~fS
zH`wr;tuNTJ4AC<^_j4kreL|E7(NPEDB9oVU<Gy;7bB?~U`bK+0{Qo#Z<MzES%{l)n
zVdzu46@?7bH+pxJgEwH6<#eY_skjh4h%?fOV{7zQk+ne-{PmYZ=iW_t2leFdxpkP?
zG+x?$*?NL&5av*v3!7U%XhBnPOXe+O9SXfd;M%Zqb8-9tnQlBi@jFrpf~4{3X-CnE
zcTm#=tbSs%Arv{H@lnMqI%#%{7TJ8xkWJi$VBdUp##S*bI34z=S2PqO-fP2#Us0X$
z5xGyQgZ#}QeAy-b@+^X$4Px%;xFF7i?L@!;*lt?mW_iVeOO9kyl^gx6d!SPY4*A(C
zw81_*8-G86SWJCFO<%Lo)!9qxl#Iq+#l2H1iEd=JrM&njSRj7cH=Xv}{F;(7W9I45
zbb8OX{sYrDoKJ<ftb!=O@)`nFbHA*?(8uKx;K{4n4!HBu6e?FjL@}>m@y};bWn<n$
z55a!`xtASVW;M=L$dp}LXo_^^mx+zo>5H6b*Mm`%SreBpbAHbc>ugqY{jfYk^ns6y
z%k=*TBJ(BuWp3grU?MEMCBSklhSwb^5~2t(aC-=94O3em1DPj^Fbzexv^LO1<w}Sy
zVYFEBS+f{DKIXsI9z4#Al~`P6YICHnOi(H*;lpi`PDdwghp6%^YK8beUZ08jTC8tB
zN5ZfEX1gOS<sZ`2_`Bf`?)!$z5x;YHcNl=Xru#$XpzMbx;O<h%v-<gJH@BOECtf6i
zd}34tlG=h8dq*rOrAhyc1v%#p75}XV&oXgn$r%z~QT6%cGg6w?c!88}1fKaC=B2DP
z{9@tg`#_9@fQ6;S#e4xI1faje$)zC5cOOiR<Hy}kT&IEE78v7=b7VW`i`ozn(*_e6
zPIkHax1Uvdzf^JL*%%ue>v!$Hy4M5zAtA(aH*L8c0MiAfyPDt=@;#}?%L4#wMs7T}
z=}x+(*fYmnqh&R;ZCC?Z`pew5FZ??Y5fgZT1f@0v#BzPX8s~qhihZUWotknFUHCY%
zaeLy)6-r3OaeHzfo-PzMf}h!NZC^$&nSH9AmvfC941<cLzKX_8w{~Upk$&8d-m2Mz
zG{hl7j1@AbTpDLzVpIFr3FYfz^4R_)6;i?!yQP@J2U8(ehX#pSuj#<Z+7@Vr3-gQd
z#Zs>K_PcRzjl*!c=>#S@9L9_b3w*Tj2qPq)u3?|5P7fVsVWMvO<79@d+QMO&V6`$_
zNzxD$oZrBdnt1J)cr*Ze`_*R2xc+w}3&owZVg%J8X8>VbcXn7mw-{GuO-;?ysrF3H
zdt3+2DDS+4(ZrelY@{Si2XTwQXo84t3h{S9pjFCqDb`~G^NN0N34f_-44)Z%hEYff
zT@S1|=UuEsdc1SMb9l4-R3MCc@`28!?vYQL$VCVCHOW>956CKZPU=D8Qd0WmAF7?|
zR=rw<XJ$#zI)W3*18%gOg`2cJd#5cS`DL5x<dUKw_N_d9`<!E1LX-O`55aB6)C=<h
zRy-}q+RC0}5(Fbbp`lxn|HiC+Nswt}PQ3j5LG$b}EqZ7e@q=LIZ+h@$o-K6_?<DES
zTW1Wx<LQVtQ+<q!1YDeu9S{3U?<1wiHq+0r#S|Vm^!QCDG5c5-C8?!^%XmlHqq4Uy
zH?Hpyo9?DDd8mA}4cB!4R&2(|`J&&%A03w4Fl87N_v+INd2N3-89q?ip{SSR^c=+g
z2n1h_zsnq~mrg#(!VBb84sRNqyMJk8rF60C{`)t~vTPR>$gGTJCMP#kLfmBYUfpGh
z^^upIA4YI>v;vU(GG#j+B3t-Rtd_jl3%moTDr`Cxd?f^LF8D;~#EGZ;Ew^#@C)lv;
zeFQezd7A~?Nyj5ZPI?Q$x%;k=!AgHt=-_?k!NEUxo&3m3PH}79<GRSKEX)sVm8W;u
zm^_+><r4O9S^&QNDJ6UZo8SNmpx^x$E;}NTPCTMdaTVsFo#2NnzED;qo&wJqaG%>W
zk$2^oXzuUaiN*Plc!bi1`CvccKgg)fO~NF~)9?uCSEJTRq+^au(44!~JR1M4>)Qe_
z+bmW7@zNQ4@c6q%3>-MKdtdZ`K|`god}V?tGEkcHEoMPsoUes1fo9fBpt<VwpZX8A
z23NV%HeCfSyMBz6-{PEMgCk6{n|5)vT{9mxuHi2@jnAXIWUNGI@0wa|^h;(}+im5@
z?Nsc4?jdC9Pov>qHH;2g755O66=T3dM%9KQg+|z?0lB!n|En3K%w~nfmAGAamt?d*
zbdfkf_&j>fbu=h?n8%0*sVN*Lpo=y0>o)h7-EwIs?WbV8$cw(mS+LkFC@5$qaMwH;
zAH)u`^V#3Zw;oc*9Nmp)_;jb>Vi7q?idBu^LtOwD9r?EH*8z{iKZD`$Pqk;T@Lx8v
zNQg@VoFSq2IW!7#X$ky@kM#~XOeig{l*wXH<Hlylkd`BeKh6H0rDU=z$oFBi0gBFG
z3+ez7k#E&d9tUJHMrW@S=u*;LU2fH4Lt6c$L~)w7d!9y?wYRLxm*ZUqZr_LEi^UB2
zFu2vOL|to+rol@mnskj!v-?_0)2CEVz~9bV|7EM|`fI3ZH_!b$xE~<s=N*wpvP^Gi
ziyeBxIkZX)&iK=rb7qW%A=6+WevsIDU>;3ZxP!XCg)-gly`nmTm;(Upj7L(l5|x!@
z%zyvz9l73hD^Xg7#cnZk(gvZ`w-E{Pf*UDmMrFKTFtG(RtN%j^eZbgjs6yi96YvO3
zDY8PXx^0BT9~qjRIeU?xpH6V#SZy{(&Y@qOc)%W4d2T?wvk<6If)W+h`G*DV*j4xs
z{eft-zne>sG6tp5_@Po?*6THuEIYL2j{P#!(CAhDHv+}Q4hWP363O3g*BDdFh9#k(
zjIABuChuVmnJDhZ9VToL{pk*OdXBP~Gj<cA7fINamMV+HoIY?dUlc~`Wwu`yk`Z)t
zjJhf?gX{qfSJ(sU#E8LJ_!IQXNV;OB{_i8$Ew4-s)$DlXj%b`pMWgI3wYo4c8@|%2
zL|knqCnpz0THqaT@283Fd4PCP2zRzkA9s?SCKhS_{!&#{sPYGLB`uGj-w{r>A7>Dl
zGrsNZ_v92_0YhWGcXz?J9Po@XSO3Rhhp_jKq%<U-4Z%gO|EThEbjL&TE4u-^u<6K_
zQI_2qi6|L3<ecM`NjRN5GxNoa`Jk+`S2A@QS-d(pa<e*BY$S8zwLxF#H&k(M(Y0#M
z&6c-pTO#g))oQee#_+8k^T(%f#Q(Z$5bB28=&)`q4J^RWOkVPmFvUY?dpd}B#BBMA
zGyai*$AMh@?Zqfmv>PuE9Mawk`I&OS_N$>)vcYaYurJQdj(c`%u<<Ecg0eo-{Y0bn
z>XD;vyxA^Wix^B1FV+9)dtiSDzI#4lsnK$45$J~$2V+iH&_BQ?Qo4{xO+<Lw`u>ML
z1}{~Q(?9Od<QGPo*?aHfc3_P<Q%Gu5Bn?Ywh)Q`G4o<hqs&+??iTlW)>%SSFPK=0%
zFnXMEO_%ac`g1!*T{VUrWZR>npg%#!qKlg|gbE+RhBs0$qJANjte<1Ae`a!G!Yw&>
z{Qu+ZzvH?7zyEQ(9uX2rRw`7sj7TWs(Lj+AB4qT;mQ6N2(v}e+WMn3@kR27-p+ebv
zZ$jqx+#f@p)${#%{eI8C9+#eVzu#}Sv(LGnb6$2{uf1-BgOK6QIEu2oVeC$fYsMgg
z^0T+=c6x;>VKVWlb90NSC(U-yswrdcNVvulODvyhoP1EgAVIam_Fub1a3@Dc67}#+
zU=us}pf7^;$i$bw*R9DFR<3dsyOeI8k3l6`2t0o;2iIi*#rzn=k0=$k{|A+0`Mvn~
zt4f7uWSCdatrzCuzlD$d1DAv@`m=80S)5T%-EEy7C;aI_oo$n*u}%({{?b{0^fluA
z{CuPQoZ`Y)v5dU^C=bg-H%uI1-W+sYQB0WX9mMGi%(ES!-@n}9wbpDj%g!>584n_c
z{aO^*tgf1;=Q;@lsXKU;lhg`D)rO33B7&>D6BzBEluT~qdsrS`nH47|-G78i#N!*S
zqmAd6cQEIB6ooYTEQB1<w3Tfd^_SY@!=~=Zcvs}mWP~f)K`0gNy&Q{n3wOZG9MFF0
zrOAJKt(FGLQ51Ji=JIR58FQ_zdH;=Q$B{0ka~OKPVYq8fe<<=`=(Lk?6ii1UeH7aF
z?%kW!((N8GGdK5!?IP=^qqH=0<29^TQENIH#Sucd%fsv^gLUDu-bplAb>}aP3FVsr
z6~7Xg($A}hx<|b@e5OH_iE>g$@@o<k&;63!*kp{{ixvx;l26M1y*OHL6mpOsGu((k
zku<02Ec-d9rkglOjGHPgpKtrxTOggRy=h`8X^j$Snx8+#@KY_GCYIc~)Ldv0c}6q>
zDniE<)FQ(lo@ITkg?Tk&Adr)mmgYEOKx6Qi+MvH3hMX_AYXO2KTV2DWRRv2U$x)M$
z#z&j9f_{A=BN9a(7A9=ZsL=!$LM%}1;2jju`q*1B7n>cbfhrFnbbHUcw!FM76&&*<
z9&d`V${0^JmfcwILF0-|S>m41wm6DCqA*`DCtWLhq?7S@GzMoVuz#8J5pdTXO|v^V
zw`~sLu8wTee~jYUeIrWi<vlXuTS7}q2Wsx3wUZcmncEIrxM5(>=L<E*_vEEkFO%q)
zo=a{4n{H=E2M?r%T@=aT?LX2iPjJ8!S@wcWH)yL3qreP`4~jL<&&?Se*3@KD(FX)B
zzi&Z~2BmYo*RKm8rFtq|Qahn(y&unQT8F_b6`3#ANLe6ypxz0?EWse*(A@FS&%AVn
z^bw_^bcDq_iOSw-Z;5W5sh?LblHu=QbE@FOjt!53`FIJW^q;|wV%Ny2+Z=A(jpyZj
z(I<#!npn9bs$Sz?tHKY27zsId3;Su&3@BIiR9Eez+?NcQh(CI2xu0hmRZ;B)w<_@6
zyZnNJ&fR~{h_aw;gzh8Tg>*kRG+gw5ID2UTsY27Gcdxl!if7#9$!M{`3zOq&@g}W?
zTR_luLQ$v9n9Wuo9yMKlF2|zfie~udxws%|U^X>@z_)MT){*>W&I}@MB`_VyN&+Cn
z6Wq#`d7M9>w>Dk;N0L&}LHaVzO%oO|XCWi4bh|wcM}d;(;3`W!d076LPe7QKce#7j
zv`)q2{0{qUCE3I7uD7@zJ$lr3068kU9c5E4090?nLQqHzq@l{d!^5V8J5sbdBBeGQ
zQ}m^t%`OjfX-IKSRM@z}DQ=aG_y+hR7Ny%(JM{iMPQdyu0>g`^nXQ@Uk9H5G6iFHg
zuaiso?a=tx-r9;vFD~PscRRRNCuGX>Ns)j8`o;GZi6zi7@YVxbF|OAcJN!)M)=eKW
zRdQ{CpBJPzF<F)hNVU3IxHnEKkOI#$u~I|rpq1lg`Z-Ox&*wvt2t^<^GBVN;x#Pt=
zG)!D0-i~Fh3Zx6RL*ueL^28~aBT}e$j%Yn?`(dI!br#E=7d0W(M{cXF|7nLe45ZL3
zH0lnI5lhw<^V<#B-lMEc8^?d@4Weh)K#lKZKc)kRPk?qY3?l8<z6?}lGK$7Mx8)R|
zbF|*IIy^eMY{vQVkLBBN25DLZ6)ovn#g%y5DvOqulg)u&C;wRf)O+%ELAGIt%GtrM
zz5lH`s4^(<7*0OwO&~BA*81v%Cw(FDU3H!1H_p5MxtT@&<a4Ax@6U1Hm<DDkW%;<L
zF*|nJPx#TzkC6xIX+nxS7^_!nl%iFVa!CZGP8(x8zPE!-7tA=Pf{!mosPBuoO~4F-
zUSWrbvAAHeE3mw$xy<~Bbv663%@YvwXs3ECEko4t;PnDj{9}%>vCFmtQHBt|(o1*w
zMkz|!<*Pg~Yt-~Bc}d)@)MH^~wT@ni@iSJDDEYL>JHQJ)=c5v67IFGO#__C6QcF(5
z?cFWn1_bwTk4r&GYZE6JTmVjZ#QPK`Dk5IVMIt)!iIyCicy$LO4ShGe;g&HY{#kX>
zM?oL<wYj>wuD_zqd-`W%<4wTV_ptAF1#hA_yq|IA*>XRWxqG|1x>m7epkL}XOA+4J
zq#6DI$9Ye^z>Tmgp$@x`gief6FCw2*^s(1IDjDih6KDn<K1l@JCwe*BWhw_>3Cx3_
zsk;9)H+EMV<-x8nM^c(yyG5TNtJOL)!J+tYVBYC83f9myWJaxyHM2tmg|nnM=6w!4
zCw4`R4I6xb>j&ySB&f#t0MxN0mO4mgHN#&r9u`30K|^7b6mMd$@JZ1i-V^37Y$yFp
zCF6y?s4AG3r1BR18K3wV;GlA}>CA@pOCohvG{z!YU7zLoSEV2yvvxkVT$l-S2og5I
zD+Ouy$<vK+Dx&Ks%7e>g=K~Ua(2|9GNpO&V`E#P<7NhyfuBm4#oyO=9w?@y+`CIYH
zlPA}YdtCgaI-GHufDzb4NVudf7^>nrJ2f@+Zr2juQ4APkPLD6|c&Oz|a3u&FZ$LeC
z`@$8z<QVLG(H@}j>gu!xQUj)IgF0)p6-G=`kYc@PSABiG&)-8O(iIhOABcCsV6}N>
z&!0E>sc9igKyJBr|NcUu2G$FkQOgscVC&igxN&29`{Y5y-BOKpUp|mxdhO^Ilh4in
z<3;Q=D*grI0bz!YuD1r20j-Qu1-mzqjG!<2w6R64Gi~CBI;D-G^TE;N^vHbY?Ivfq
zolgGYugef*2V3%y?m|BHkD~8GP(W|f_(f;-Oke|<wqx8$S(qf1;b!glwNaOX&Bk%%
zeVqYkg$?Q+6{%z%rHx`Y^TTyd9$KcY2$Q~s5u+OAE_oe|m;uBEW95f9WreYV#R~<w
z$~{P!WGwj9B~8U+=}VNy0!%4bKdd3mh3i+q>Sl}@l78L5kFp4a1Jh>?iw55=NnN<a
zC;yMGosq15zp_)Oi?J~P6jUE$``0$vZ-jH><hugo`SRZ6<cR2;r{xR})rDR6*PTwS
z<cN&Wn6wAAP;Yh3(|Ug$KMIXy9N0^4c}JxmPSO@b;iZwJrn_ILC&w#<$2P~YF+0bp
z>Zgd)UO3`0Js$X1QVXGu*Sj!@%S2z{Tj%WPTN^LmKZTYjLVEbdlhsp~%;edH&eIXE
zw0-J%S&0qffUEouK|V$`i$Mjimnb0xxZG_a=o0s&q0sm<Q|>i+hx2!5LtFxhPqcD3
z>|QPK$~*#ov+Q3#frxa!ihhFKBCW=i>CpoR4%`!;csH;k1yG*S(~N3iC%qdsu?pu)
zkZx%mX9ryG+D^oodna9ewQ9k_FEYzrgOXfT<pYuWP6g@lZckHp9*=2bs}JnL&~%th
zdV9St#gGP@gF}Ytr8y4kGJnUXC@AO{_U_H=ha_Ne@;lfE&SKs3fJ0GY`hn7lJxk41
zoJ8XY_LaBG$g$mC5pM*vc(MLl<%pEkkHUe&*3l*7t=#6jR$v(@6E9MenLFe6^DWe5
z#{Hb@xBcSbUn;ZayncfB@+A$OAMY0@#UePIJW$B~<LYa(tF!yczHj~~zXfUMo|n}3
z>aGsZB(TNbB2`On{JiMis6hu9o+d85O@$ojX~d@9+jCz}e6FGH-iK){0H99>{23AX
zBr4=+q@=k6#5eb|D|=!6iyMf4k$zC^+%_Ub(eB$-$V#N=zs2RG`_)0seE2Mi&#=WI
zj?J+&x#zZ8*5$!3a+GO#ZYZMM4=l%)6}fs1!t{dOyAV}Rko8SJsA!%#$5tS2Bh5~l
zdMXU0{MT+=XL?K4;{Sc;>1cn{ye+NmcUt@i3vsOyerZ3tX<WLS#oq;jbJgOrc?wgJ
zBr&QtCMqqQcbMNSshUxvSvuASX>7;QPfc5{pP_mhi(XleaAmPOwg;=fz9T?`j<8C6
zpNPpLblatsFEC~PVL^^hxA+(*EJu5Ug|2?sPB<k_+_#fC+2ow^zl&#oEe?p#M^kF;
zGmuilETu=OBM~JX1CJ8ZO3A^uqr<~LJMTGtJF}UNgj@K^MOEBjwGJH_db)$dipdvQ
zCZ4P(zH(S(H}UBu+mv*_S<(%9R_fWKa-1-LhYy@T(qCz43$B#K%G(igJI!Y?jO+p@
zj?wN<E-qeo{SRTQIo}1VWtRqX16>UjImps9yY3(My$_Mwt(<1JTj<T4UULRW?A#@W
zUp{_L5lGxOw(po-(y898>rZtGV!D~Hd#|ju2~LkGWMvpzsZDf9nfFZZW4pntm1R<%
za*C3glw9q@K`a9OCVBvwL0Eiy2XcB0BeIGE+J#tHsT}D#3m`YsNl_-I8&!e`!o;`5
z$4j4XAo13QoD6`{|I}7^7AyH%QCvvLPulz9O*M~vp%d2@-lAw;mgW<y2Y-mFK@hb~
zi_$FoSDI7D&Z2&ht064~BZGF>Vv_;1wDBD#XH&pE9B>QJ8s$+3o^=@5UlQG1>#O##
zLva1##Ze6r3>|r{-}is4zu*<cFb#}XJ<%+UFAJb4!1grXWR1Iu%287DEV}Bomoo%_
z2&(V$U0hwQusK1T+StVK=x9DnZeD2`Aj=B#(gGvwFJqvmsg0r4^5pfK|56bq8KCYc
zwhQAY_19}5nW#~nuisxbnk^^r230@QD$Y0G4Y`R%z!t(Zg2JNUhSJTiBD`EVsswQl
z2Od9uyrg)e=^D}Y?b{(Cv;OHCr4!-rU`1Jh29hBUa_{2PzhOvuXV#~dN*ar>vzH=6
z-J`$URLn6l{$jr*izKz9=+gO>&Itek{w3><*4IT%OXOk`5g2ERo<6-DS8OZU)>{B5
z_(e~$YhykLLg*|sup1pqI6izcNh|4yP4|$uGZt`ZAf9k}v0%Dv+*xqNKIVOMag6xh
zV?!e&E~=31#4r*iBqk;c-zIa0f@#kiR~A3PX=Lp3=ObX*J4c{8KRAFu;#j%Bvd6v6
zn>~V{KMLPo;Ty~=4gT8H-hKd(VH07&7mV=66?Y^1Ss?koO~4Ib>NQZ+cUl}f5aC_t
zbnSR%hyxwbThGq9<k$X*`vdn}7CT1!Onkz2?)zV0eTan~;H~fMlzIhlfVvvj=v%w2
zVR@B@m5Ip^39+%VBqn%=TJj~Z@7$JxIJRU-MeO@%RRunVdm32s_FCiuz1@jZ0w6p+
z`yNDuAD1<5WZ+A=i5ibL%K~>NV+P1u@K+26C0o~*8p`;dSDM6pd3^!q;OiUv*i22K
z0Eg88oAGq_d&>E*uT$3bRdf@JTTxo|Zj@Mo$B!dCj%zoBabw_U4C}?p8eVq!x{$Pb
zqb|~<OrljaJhD`UMSv0NZYJa%$?I|}r|Z{f;&^Z^^c!^;?2kKC2$18UNS<c_RoU_9
z6|mlFA+5W+ds6tS$r)^#1~2$5oZ#sB6S{2tX@UJSU#ontnthWRG@JYmfMPGtzwx03
z82wI42FHg>;n>R$<Auyz<{ckwADSu5%aH%TU4fmKfxxj#E(}idq~-QBgTt-*SSdK*
z{&)<4e~Zp7%pXY&yYkD~=#z&;jazD>8P1V@P7KJExfzkvMW{LnE9`44$)jT6xhIl~
zV0zf-%eXhC9q^xj6MdIRMn=X`3`tZ}DOV+(`vEiH+uoQpzD!RzwD<fbK~+8nKGiR%
z?N3_JZpG%|XAMEU8@t6U-DVv)ZhV|5zcn%T@!O(z29U>mEes_dv=0i}bAlMp;8u!Z
zYM{Dlcw?9qYvSx^pey9?Q)`qR&C_AD2ZA%PPB@$=l}#?bkXIKm+%dtVHg0g}*O94y
zQ#k%9K<mn-L5q1E*VtZmrcc_JMoi<d#?{5*h{yA#s2+N+gneS^uQt^5rXMd9u;q@E
zNV%h#j(3Dh$+_yQP5FU^y#R)1Jxo=c&~-qpvOb((UqV6IacNTx<F93w9uT2xcX=5O
z?hzTZ%*%P5MogDbkpaPr;8S-;$AxuV#f28eB4?fcz8|p^{>Xxq&pi4bco}j)2}M*O
zS<s%ZY{8#zBQ-xkijyk&qMz|;F!(phEYJ}|p7l)*mv4~<9BL6KPy5P=a*h~{W1e(E
z2AcZUISbzvfs5=KZ2n|p;5W|i#u~E}a%9SL)QSEls`Ca#zD9%41tIm8`eggJG3Q5{
z+aWRjaK?iN30C|9;=fiJ%3u22Xy!2FTif_pjg2i-jcn(Qac{c0v0eRH{2g6u%1G?;
z%Mfnda2N!otewn`HpL0fLVefzfQwhvuV;%eH@?`!i<l$i5S{cl2I4>5`Z7q-F8N;I
zoQ28zk#6gs7`^V$0xU!fnx0}~dk;*L!%AipSAHleQnmK0@NtBh<B;?hFVDeU$ak2$
z0fZAGUtNPJ8{?VLUd0JomV0!Uv&}BqL)O1&p<Y(v*Bm~u^-TaZ?9j5x#M@9_r03(#
zH?e1CZt{j_D63Uewt?MFycPQcx`?oKON=@h#Nynhvj{67O@EHi^oLJTpUXzK-T$W!
zfi*^ru<p449UN;MuP!bo7~c{L3q$WF>QQ2tpUgjWBnuC_#6&{}@U64>)*4I-a_5af
z@{*H_16NH~-VU)<8XxjHM*fKMj^5}4J*p*-FmwNX1PR_m<U?dB(9O~g{j$>Bc?nuj
zSf_(klFpu&R5SiHM*TNChIgBUBxhqzzEN8i()F4P=M?yy!!a`XjIuVv0xF?_&sp#s
zvH7*PP>uSxx1fYcODVW+SbWRqP8GgQ$K40ZOIKRI4Lm+goBGX&97bkezq+v#;NR7<
z=piI9Ufrv#x4Ne_7D@;rgK1XQ#{H-DQ}%b{bTl^p7+91(C^BvLs^=lWBmJe4D!kEc
zNh?UFhclOPhr<!AEKZjJp{Soy0}}pQH+T?!)T2Rl?YnP^pUDjP5UA1F>c}P4j0~P{
zf0I4){@5T#P9L%Q&O4o3j~+WV_n=QG=oAz2%jqcbrg(NNXIn?#9Bs!Jonf?o?m)k@
zLu0TyCYP(s3+6tqqP1qt9dwa&|Ggg9&{i#p^q-qQ0U~Y+B;v;JZ0S1<o~jz(HCOJ;
zhk+0E)UFp;9MXzvTImZaL~FM_S$*+tYWqT&H~qT6&+CTJSh*Ec$bLLtvLY2!$VRU2
zCi*aOk3xGoe@L~@e!;+N6qJ-QNQ1|ufuQsGJr(IUJ)8R-E%0-PNo!6)Yy#OXJo6i9
zM7z~ktxDg>3~}}*mYHZ%8NWXLETP)J1VKn>xMpDH$89)X6z%1ffB6^j-f@_!vmpF3
zOx!o#p*nSSM^RsZ{99<TaGbnK>57k9aAbQx1`^It^Z=*G0parGuF|)LB7+~{*e*h@
zsShMELL3(Ip9XreIY*&v+jXJ6y}d(ikJ?`fIcOF!jFa6$Hq^8$+5D$!05Q>E2iPwS
zwr6*TH<=0+Ys6&js_Z$1(@iXXC@<OdyA%_{n%`7hRK!FB6&4hWd%9%ZWV`gCatK$x
z*y~g5G^I&^{BNcpnneLT#X@Esm*MzMMODSrnto(5#{^`%b|Ht@{CEG|a7rY|k3`^h
z(|fII+z#$A34=_dMQ0XU{tMGR45G8B2#cVh`xuFLC+PUDIN^%psdNCxY3#wGnlP-8
z)D2J;!R>rz@js7INm46XnW@%m`Tjk$a%C{kURG8%3x?u^)G*W-?;qUkpG0dytyV<7
zDHlgRG&gZo`AU;soZ!*(@zPxt%$5rB+%xB5OCILBx@vdLlC-q6So-QQv_ySe>3_0s
zE5G=F5f?4Cyp;*~<l0-k*y%`pa?rq|QJwTOg$(KuE5>b=D`_eH5}j^QKWmITgjJW_
z%55Y!Yl<+WwvuUp8&LqtS_RAg@@4_~*UabUioPVBmp#Nq0}&~4CiHhXTm2Owuod-y
zyvAOjc&3U+k47Z*oOp;j8&$LTm!j1>`8hdqZQ?`&>{RQ;n$MWe*~RY0N8<q2<CiAZ
zw(iH@%EdVE0^@hx;2BTO$?tNk>4gpf^<WM!Djgl2PEssFq92m~jS~LB$>CTSkrpp(
z_%XAY#oGpFDux3GR!3lHSf<3T?5`>xAnTixdihn4d3Nc>24<{PP=UgVf?Nk4+>8DP
z+vgRM;-UHv*;COV)U=HMr|H^;FRG!>;#3aSH@}k7YAfJ>B5w%#w)O95G!}nKOs=+V
zKhmXSbK<qB2OW}!xv(%VH)|}oZS#eX0OYX9<D@bE)O+a|#Fs=M4?49ZeE5fijon0n
zg}mi<3(JpJvZ0k&VvV!6c(zB`GWg2DKjoZwH{@9MNIHxpC4S6#FuOVk6Hz-HU`~nq
z@y<&xe@!|?=9_?+-@BxS{}cgZFyw<5f|vJv^i;{y&ktJ`q!rzTp@rZx)78aAk^$;i
z`jM#VeO&h6!V;f<L1dR@70G+)9o~z4Ir<t*ux0Eup+U{k-VawF^Mg5ALY+3BJ=W}^
zmfJ$gvK130l>P<cDcSB=hrfsD3{dkN`E&7SjipNaa;zTMuJ@E3teUUiPh|GJtE01X
zo!O2-1%Rxnv><HR9Y#nt17m;)n~ew$?jio$uwB$kddt~*)fU$<9Fo%yf)E54^QU(z
zTo`m3@yo`SpkZC^5O&jj<%WSyTJfcLZpqsNZnC4imgA}E+V_#u4@Lq<m`Wbx(O>?W
zDT3|(wOa<j=#AUA_W!XG0=W!#a9?I6kSweFlA!qo85p{@T$}sR-Dhtt=Vn5gq=9*<
z{ULk;|Jev2MD;D2SI(-rW9XYJ^YsRUgi6FMHV~MEq_yKmeyU4e7`(%=nQsQTyCSQv
zl$AEKWx%LyM?V-j<6s%4bYf{JHWp^T%`$(6PHbBaKp0O1Z){om-F@Oh4sA#tpnM{y
zT_f*nMAPjlA;ED-=|45KvYITv^;ZZ#)wxu3Tgg8fZDp$#NeUUCN}IqdF1NAh4u9t0
zJ9!MJ!f|2mF$8NV<IRXqELS{(7$8*p^<9oaQ~BXb6wF_5iXqLp6nFKOlhl)Q249%O
z$7z?i+Kya~T9m%>{{+7gQBsRSJk%iy@bLO;nrlWsXZjm5BY&h6f3l^p-9)EoCy;R7
zj<96Bt#f(;AL55Ht<8yVy09mbt>cNffQO#pvHe^zFHqIn=g&$U2-2+f6QN!;0K9!6
zxkuB*zvd)ZOfiePonAKOv%6R$WMc>X%zuXGCLHi#tP_Q1PYo82_A1@90qT7bINz`0
zKDdv9ua!UJ^BZA2kP)$S|02G0Shb-yCqFxCfK5(3=hgE{eM-=B|CRG$*@Hsdm(l8e
zTXx$B-b!#kWJ7dahNEiedKotJ$Y&(xWz>*X#M_i=h&!#soXyw^e{@s@C7qW!ltgn`
zlhVq1a&mJ03U^=#u7T$_2hNV}{>exs*kU^%WG8vlkfU_pk9D!Me$112wga1qO@uUK
zQxl-j4-diFE~+nSIDBNDy_|1hXJ4DnG);@~B0eD(Wb(aa!i!nF)Nuh8-L>OVnsM?l
z(a=8i-gU%2M~i`SskITt@8pE$9H?jXwLo)Xjk=_$D1*+8?}q8s*SAt{CUQOMz@0SR
zX}q{$wh<ZS=St1^n0ICww>EpM-@NA@YK6t&kk9DEj0_w`5}XdVb4cAszp5&=i$pm;
z`!y;3S0W4MyHxeh_%)?!FSPS}OnY<0^JKQX&KrvHmV{NEj#d^8xbhf|#4P@N;L3c-
zW_pcJ=~Bahk{(Mh*Z1D_?}tUCXe11m)94gUUe=JM+()oxr8>_#=pNcUL%c(!s29g=
z#jjyC66P3$-|;q=`mo)EwJ?;y{O-%6WXEKngSfWMODykbY_`Xfg`!FK)TpnI{-t7I
zd@B7_FLpt~&KDSQhBPiC293_P*2y>iSfA9}_}51%oGT6%VX;uQHuQ0x<bPYMO5g<|
zq8^dvZ`HH{ws;ZZybHYi@YSc6cwQc0K`@b&-KT}PsT|9>eU#Mno;A$Hx1%tT^TQ@_
zR#w&t&}Ds{40#C)1+stWDs3M&g;NXT+Rfi&OmB7we<gXMhY`Oh;u{g3hntWhWXr?G
zh{>LlZ_c(mM2(lWfwN)7Pb_I34V|V@J(k>G^2#9;-^`ncna308*N<pv$VTOmw&0r1
zZld2Hk8W~l3>;Ttex%r%ml!D&p-KbT#3cPBaB|mq{)&VB^RY(20p<xf)Fw0`AT=4+
z!5*QG%W+}e>Z7R(3dOl5!pV-7sMs@s@Al9gxjZ$FB%6HbUmW*1cNpskV7@{pP`u@N
zRxI@uV91I&9^Sv{7??4n@2)*N5Wm&D-T>oXE<VX0CNBN>9zUC=8+fm+_Imc00>qx4
zY)AD^xMTY_Sr>LvP)IOf3)NqtP#q8!?~l#U9lHpv6TkAcSU(v}P6VZB;)(?P1%BoS
zt2vkccqcwUxJ`8TUR_g}#KC2{3$CRR3$h;cmRs%B6Cs^dkOMvLg~@8%!>@xIrGqyX
zc>Qr<gd_ROZ4Gx($vYB&&!~JxREU;(%q!=WuC<{tiFX5SJ`rC`zEdL^(hIpznK<+A
zZyn-3RP_eYyB~gpPezXtt+ZK2WQP)*{yfM@OKi0GqN6W@>medHbF}vf)9g8N=G^hO
z!y#FKNH$cAX7r)Fb22w4C#N$w?=vfIbsym|o8@hTwa(=sStmV~VZ2LGi4;~=7>@~h
z0nYaobRl0nTz*XbEs@O9yo`H9kn+XWyZ7!n-T{xIo^@-x--wsPt~f~%WBAuBTmV8`
z>eHFTBgFzB`zEi_+4kFOAxMC>!T#6humA_hNCtvgGu86B+h_aLZ^EiVbbDnIGOMAe
zC^>nwHeRKYyk5+mk`tc>-uUi6f1E?07ky*3_(L0<gD7ApuR2Skw&#h}lvN-E=G6XA
z*(KkUoxGNz_|IYM6D|5*^2S02)#goM(Fp+4Y_K+j82ECZyO>1_by4A4p@>UFm^Zmx
z%a6SZyy9VVu9wt*)IJD&Ly+jFnQ;Gh>~?tQt+E>UR{Qd2;_v#Z_v%CP)7;I@-UVp{
z=d|wYQbP**mn1#LUX@Rf!AO^z=n-TIcYDA&Ec^g^vs{{q)c%!1{`EezIb6JDY7V}w
z1hW&Nba&aZ^T^408NUK!*ZEQ9Q?o?s&?-YR2~B_U{}CqrPEQ>qJzh4zio43_7r>mx
zuKk|5V)-u^uYZadXEc<O!q7s{y`|swr#Hg<VVw0%P3opSQ}`-N5|X1X6Fi5_+5tvB
ztQlpX`n&cE1(d3Uojm(h+<REg!$k#tjaqNZu}CX4T8m{kn2ypJhgf$_;@!-zSogQD
zCCD)}h0fd7w%6>%{eM`T3CkTmNMEAHX<&)d0ak=r0QFbkP2r=hDGkYuW|Pn^J!Ir#
z)iy?+!VvbTMJ}`XB6pA^(|V5QJf|!f)wv`9L38Qgw%Vc3K*f)-%2eR*Tz2j>B-mNM
z))ML~T8PeV4rug@96l7FB$?rpJ545;#dE@=ryCE3gnmr#;qmbzzW-6FTY^#_#SM`H
zC7^78d0t+$R4bh(<(+NSqLL@>?OdJbfE{{M2>|7{e+N1K;voFxDdzyoI(!ksci?fp
zMw|ALopb;*B&6>KQ>5Mq23znDFP((h6fO&Q5>k%n<lAc_Ib2?<7jN__ag@cx!QBMt
zZk@#<%aS1y(_*t=`t71v0H2X-Y?Q=`&nU$yfJ|qlR&>vT=k1kamBeJTNGkh~Pg#Cp
z9!lM#a`7thL+T?O{+D0~jUG%<%XgYV3VE0?c$wLTb;Gh2tbr*_H_h`4tqQ{TfV0$_
zfs|)!wYlM$!V<BqpHwW{cT3o`ej4lJ`O<`KV-5tA>&oPR6uN%XwistfPD%1ha^UN>
z(va1c*jIP6HAk(<@UW^TI~y|VMAQL*-Jbse5Fn@ug37BwX7e~ti{bGoocW+Y9i+C&
zIYfz-@9FikJ<@(dZ|=2NBzqwm<?Xs~)=THh^M{U*z%-Hlk4*aS-~1E#W+^Ew>JK6M
zA&B?ZSua~J4~*!KWIoaV2X#%@9%(Rhb4^lv$sQc{fNVbtRk2OE-SieeF}Go+iv1b4
z!6#rb1(T4k0T%k-XyMJ-aYW!CeyW|YPxl+ceF$bJf)$&P&R@U`^~v5y#NwF%OR|%_
zzP;TYsd?h0^!JMay6Q8bY%&kH<_&ZYs^Jxd6ciGx>a<=G>UYY(LqyZtRIT!hXI9Yn
zKM*6$VkY(iE0p(h!&}CQja2Y)?ytwgIiie$q2~&j9%r{_%FfkN7v|653Bm;FCj>B0
zaKvB8B7zK)h}Kgg?IOO2W1LiNTP^l9%!Y}0J<jCIxUTQ`wf!Pl(DVlw3YPSD1p99k
z*@~KllnYdYWC%>KazX!HZ<;&#!!TkwVcAcItu<&UI$!0ri1{6S7G{TeWeaF44UJ~K
z{%5D0iFO<mm`zALP4n^`awnWfmFRhh9wF%X_hmx!551#Z=93-xSRjB77<QP}X>4Ur
zD=cDfS@>2iak`5h#dD9{bGrZPIuirKV$IcE_zvDbzWZPLw_X_((!LFah)vKZj_$Qy
zKHDn)CFt?>&#d(g>JNe`#wZhxiHo~A0;M7PSJOTjfiqg@Ll9v-$Oy;IaiE6qLVip8
zlYhMdyr+DmOK)MR0-rjncnabn%zrs)pyI?f(BQ2JwL#S*;_arykZODOu;s)0p(XVs
zwHdK3#7>f&o4ShsbuFyC0<%l%pT)2ofV|OvJK_-#!U337C<kn1e`N6M*RPMI+QNLV
z7(7<Q;<#S{j=Oj#HiFm#1K7@thn`alz<x9EYMSG0i5)&({KalKjT<KeV*d5WowInM
zC`6jZCmAJYm1b3x)JJodpQqW)D;f*CI_E$|LwxVE1n9TqLG0`QO$6+FD3G3yt53EQ
z>VPY9`gJcItd<)c-^!iVcav|032n{KzEcg2KlQ!qy0AS^2SEk@k1FpjsE-Z-xovot
zlE{M%!0u^)o$WFCsR2HXom-5pWi`j~t47!!brsWjd-?d!B)#iylqF*qeA!v3)*mIa
z3#4*kMGP$u%3G_-0t*Z{rwqB_Q*NqwZMt#PB%}{JTmJA-ywHz=7gCph3zJ%w8M#k9
zm*b&&TM;h<dvpmlbo}Av<u$nrOA~pB>I_V?`>-nM2ZXPUJcGDSYCJvS&@7gPy`2tR
zmr1zr5?@IU9V;+l3f~~*uZp-L+TRow)INnbb&BQ@SCrK7+)cQp31b1*UH>n2^~<(`
zuK;r*I1}Fyh(`kMnMa0f^4r!dHkV7u1A<EPM0EC(K1f}ZO;Q-@Nc#VyHvfo}A1^Cy
ztTCsghUmWI6|`K44V~(&A$<~72JO2A<suqN*P9kU*1pBL-_R%S`H^|sh~1Xg#ckXK
zy+crXgwNcd0MWlusW;K^xheSr7T>E+6zrXB+jn4A)jcP(;3jW{+jSE`(P{usQy|P5
zuxDz(`@~qUNNO&L%Dr$^c(6ZI!*mWTra3x9V<${hLP3nJc)Q^5-=M;dbcgcZ$1g6x
ztdKqN6}1WBn@H7eZsmwE&apf$WT-@liL$fZrChTKc(>o51@qo>Zx&tItA`%O^OnzU
zfI3~<?}M#Rz%id*0twf2W^|c)^Pnnu--+mVwrm+$o=RVvHOU&Y91IK$jQ>}HKt;O4
zNfj40gPLK&^}tkydagX(#$@sW!HVyTdXZ)}6urGV2-@b`TMG&b)&*hx>?YXlzx2~4
zviKhaB`cJ5Sp`?nCwk3`ZDBMnxOLY^&uqv<3Pw9s4mH1-F!@;VvBn$06I_;4QmQM*
z0%s~>z=74gA}og7Ldg{ny#a3a$iEw06%K%cjs%}Qjbj3|E|RKStMc%Sv6Y@RK6*x-
zxJABkr~=deZq9L{?dOO8QGFypp1*@e-3yykfKyWP0>1XS4&$^*wia*u@!duFyRYp%
z9EFladV|Irm2G;<VXAl}Vn_pM?t69+M^&63FviaCXQ?Lz4@8EO5sT@64WUKIol61O
z4Q`um4E(-*ZQ8t+k~rlNU5_1QLl-qIdq>7ZcN!ecG(NMADVb0C&^4l#aeHE7B4_*G
z!JIgy5q4OgU&HQ-58ls#CIo)zz1jl9_!E7nzij=HxkzI9Qe65iWOfMivkQ8<f^vQ!
zcq>Gj(7Q(eCGgb+KpJEKF$nm#IAqXQQ1u}$+*kfm9XR4Zvq^aSuVxb}lbh(IEH@SU
zg%@A9RC(Rab(c3>J+*7`Zpl^%lfE>3UXvgZw;kf|rI=V?TDlLR%bybQ`=4kXgw!n0
zlUl+EXnaOGbT9{$x_#@wi2R_~qc3(yD>7|1afdoSGLkCaDYqHRiTz%KOVMGCrTxgX
zuv<G8^K|O(Xkd{R@EMTbI1c%Z&ybGBeiTd;Ml9cc-PL<`qS8ZSFwNr|^^Z?XJtmKd
zkp`_9q;W$M0R!zIYH4ld;W~F_2lk&vCxL-g=oiK!e&f!8<&gG1Nc-FM!E})0xrs(x
ztav%6QHT~_3Ml5&swMflZz$d!Z9t*7X5eZ+@<~FNXa`qH$pw;y`lvhP^d1t}Zre~O
zrm}vi=D_X=e1T-6Wbj)W@DzfO(6c$-{%Ra<U<7HFI&eekkK+W4MOO$azCUrKXHDMc
zr~A>jc{8_u`gDznlVdw_U|ev}<-H6H-{}6@;rfSMM&}`S*W5frxI0ba9UW!v>fu_u
z)RP<Ev`(bz>X;Yi_K>9<QQGc3F*KB$2E!e1{jFH@r-l#X2KuWyQxN(ep6YZN>$HVD
zn<mMW6=S4g$eUfOn9SZ9cZiuO`Kp&IV3bI_ugBf^?ZSU&T;MXa7SVMgxoWuzeCrKo
z3fj3$zVqU@o_)%T5!$hhuJa=LENA-E^~*s1-mI;yts})$e<qBXi%O*$`*>)T`fKFG
z2;k8UNPy2sAv-}>f7IfOwHq7`k_6K<_`Kur{B?+ICB=JrYD9!4X&|>rC3fUM)Kd@V
z*pGsWYed0w6A!lMC!D!6ei_zs9sVm)g>X{}bTEgivg9s4zyBK9zH&V!@CZ(}Fwe-J
zD7k((X{Y;-YE`m&@_6r_25E)8ZP8d99V8xa+<03AiO1kzZ(0r9B%9v_knkepsQwP1
zv@+k`n!&%#_MkGikD>#g<m`g&#S7-%qBCh0sS)+pzf&XdK=!sTUuIwZj|w;#QNYwj
zfrJZtj~k$6t!8;h&gVtn@`7SiLGkDNj~_jj#i*$H$lQB>pe+REwQ~9WNPe&dQx5sr
z83yWI$Uld~%9Tmm+!t~LN0abO>`~ASFv)|j<zX^z%q&u*Gqul#j;SjV$Y=|($U_jq
zR+HS2)d4kM<`^OuKgG9L&-u4eDt}znv9mzry8?`T>u($<kJ{=iT)jEj{(XVZG1o1H
zQ~I-t_0D+VmV_h=UarHu<V5EW^kLq=iV%IZ@1%pdyh(T#W&J$!bK-xZPmL`?W47q=
z&+Hf<8q3`p3M$_?1M<zfSR6)3*vcM8(s9WhVtd%uZ<}Z5%2%mb%jY=oX{>OfZJ9!U
zvPPnKcG2B9*Z7#-0Ab>6v<QmrWLXhU7VCL-B}w`Nqc3*{k@IF;sqM&-6chWId`HIr
z%0c4ylf^YwmNbGk=Mf(ookW!47$0U?3HBrUE2^nbMX9*&H-(L1M+6^ZLugVF*om2=
zSx7olE-CQnElHbkn}z4-#jKUD-z~K-7FVFE8yjop1_lS$IfeeOXj%ZWWE3AY`El9=
z{fW4cv+5_&UM8IH0k~cF&e_a`5dP6mLTgQd4ZBW$$hFIP{q-qBvb0E@PSu44SUad>
zfH>OGoT!d%TvZovYXh~&E>CI=bN*o!u-t)pOf_Y+aPC6jDZITjq&}x)!z^V#rWeFD
zSk&h-V?Vy?nv|$0Mr5gd*_(UsdiI_*qQ$6>HiZ`{cWV;dhRwqATWlqBhUHpyxe4o4
zLhy)*AsZ_m-KJ<`bu{VfwVAmOiGjsUht6KCBo3xm7a|@Q7?6Cn7iL!dm(TU<`TEg7
zhh0y!jd54D1v5|7v*;_|OQNx6=-SZlo$n;(GM*Gt9g_2?mHu*Cn&I}kPoLJ(Yrqqp
z;qvdZyDWn<6E58PMWbAaACXhAlmpH8EBQVr1XP5heSpE%?xGVecu$YiV8@u=&wdg?
zFBuV*38SITMa%K>{G@ctSHwo;>1Xd5@fou_Pj$Wf5{lHwVRF|3`r)x~TO`oO7@?X9
zj0Dp*$5r}JKNFqrbT}-abQeFsW=^lfJ7#U)C>~ABlEHy+S>>5>PaCBma^2y+q@l=L
ziJV1zM4!L3_C88ihOuI+mQI0AnckLags=RU3i$0xhrA%S=#Fk=#ou(a28tY3>IN2>
zC3B9mo)=uN814&>={xM#L4HI-Lug+_RMh$vqrXnM#S>tIWEc#%-ICU}2M7$(LVtf!
z)W&oLw&^u^_&k14BdS;6%)}q*x3)x19~Q!M(0cj37W@0QwzeokgoXd#<Y&l*uiwVe
ze30<<LYNTCr6eWk^J=3h>Nc^(Kih(GXhcqiI`Cw8j@YV2<Oej?EWLmKJ}-r(AEp*i
ztp!L2G{5S1yV9|(5-(?r#3_zGj0f6tIvh&bkUHiE`iubzPs&HXe@;czO_73G5e^w_
z`3+A0jon&Sw9MVsH2!{Ku2pqr1WEU{;>B*&YwKANmyh<-h-%eJ78R)^wK12!qIlo$
z_ON2p!l{jG`}dCuq+BEx#qxn5x^XCStCe`AlVz(TcF&h8h7F>OiGi6wzZgktH!qC0
zF5F2anfPy#+a98#bbTGd!(n%o0Y|YCYtS-`X~i7`4QD}CI&I#}5#qZ#`{ow;{<#|Y
zhJ~nNwJxIgrg$`U`x}4vTg%_Reap!oJLmBK_BFO(B$4z4$K_fAAGq*lOnuZ_@j*rO
zFin;XcB^JbeV3PTs5)qEU?~x{H{FoU)yULz__VseEebq9WiA7Ce6_C?Xwxbo{#V<!
z1+yE<cJ4$Rwf=bfZ@VKdF;}V0ex<|Kh*`Q@N$?|w65Hev^R<~zbf5{0wks$otZMy_
z%={cl(H?U-KoHmy^e_0<hOFp@XciQ=Qc5gcIo4>(Hr>DbvhL;T3pZR`mgEtk@6%(c
zT&MEE4Y0E31r-yfjP1A)I0PzYcPQtF-@910b3k&lE9D1)sfM&zJjTPDIVe-xzBruu
zV8t?FPwMHw=klRNCNaA-L^_(fIcD;^BbQM7VOd}C;Qvc$3J_sLeaHJ5Zx!(2Z)k<g
z>Ee@Ov7?9ZWJJqjMyiV9#miRvgevQDZeP4?hSB~oFhBjOi6oNk{7mZ9$0?Vj=d4Q6
zZHRSQSgA%LRm0sG9YKZ({rI}dzj+UXB6rFxC<){d#=Rr@DavLgzdFdPyhEzzyMz5L
z*Kbe7Z_KET^B#`O>|VI2pO+u@eF=0x`v0l}kQU7|{muCLHd@~b<y}7U^ZI_-eXhw8
zF4;{@G2zUv9&;ZpO1<_V2H?gUtIoISSR?s|NXMc_rP@Qvcke@H4p#AYE!@Z()FW&u
z!lXgUXcA9AM2Ej42L|(F7L^=H%*~-#SgdYn-xqFikNikziJGBMzUd7Kp;1@}=P3d<
zAG4c^)U)yZy)zryy&Y_F{;Vn%W-cRJ5+-uO&JaI0Vna!nbU_;)^Rw(of(4SGeUVnU
zo+GnJvyi*zk*p$HasBziZ?<Gd_*102J5N_@?=RnVcU{MHZyb*2{!I*=ZV@<(nd;A}
zqAF;{Z5jiECjv%v(1?2{K4;HTA6n9=+D7kJl%A;-OMjrFVxN?A{`{F(5!=H|PfBbZ
zgfo*vLqnB}|3@{QkB}$|wT+<D$8@5#iUy^8r?@68*gUu%TDtb!d2^aLNDZhT9Qw1T
zr{@gTWD@@V-%f!u4|ic+({g=c$ri!|xzgX)TLAvrR^CytKsTc|+hBZr-GjGsUDU@~
z6t&DQ5>qzplk&-Yc6RnH`{Veyzl<^3xPAN&fDKOXfAo%u@Cu@Qs&B<J6P+H@@9s%P
zur{2fukCv?Qu$?C;-bMlS4$3)if!hDbMtcXGohls=4Y5*PLR#1l<=2Hm_(j%;9jcI
zc>B7R?|eWD=uo4gyt(Ng%O^b(LPncMrhk<Gh`LzhvFfO}YNM6cp|sj!c{H`^#>Yvw
zXY;RSwcrn%QpP_%U0RKJ<@4i9!sqht8^V(=*Ik}pT6NJ(y)kZX4o|r8d8u37YSxUI
z;;2a0fq!y*dd<e?NAyxPm>sV%|2RI{eELF>pM}Ts!nOV9(yjKHOBCCSpKMK%N?Qor
z9Z1()o}cPhTU(oIGk!f&GV<e_@a({kw_DTPukhrn(?u4vE?JGmMlAhUjF9>wdAe0&
zu&_f`f}Phqx7gwlMRV&*J-_tzo$2InRz{j{tX)`6FCRaWEM-X>7~I%<<BeTU=>w~K
z&aP}eHq|uW!jfywZPf-p$lTm{b>Z4#j?<5Y3#&FeE)^GRYp!aMCQGf4n15-E()=-M
z(cJ8{)T?Zj?6K;xx^VVP4aK!qG2g6p>tw&-5r*)u#ailhiAUMLcB-dK`et>1TvRT1
zflZtRjS@NE$JNyrQr0hAOL1(DUol~`{*l+njo$ljGYkImbr69Ni;w;IsuGPNi?3d>
z&BXJC)m|R{wWvBDWm}t**C?1tl{yxd!#xXdCOtnREOc>hZq5oT3ei(2<GM&6$;EfO
zCTfIMt8c93w=2^6kx>nWmyC>zv_GF-Z_lATcEx4usLpT{Y71%K=@OR^Z&LyM2r_+p
z;zB*Kd^_VloM}ey0@FNlFd>1!dLO8P^CCOR+|zd@uY3G(?S9|iEN+*kmuj9viptSw
z7X0!wJbc;ZmrX0#LiGN@M2Fvjo0)vbL-cduzPp!3@nFMW0;Wu~mH7qDRl0t>+x0}3
zMWW1oroM`whN7pNp4~IoM65E=5p~bFDK_sxZOV2%%_~>Lqi8n-NPrZhl{&Ec?G$<x
zn^m|A4)s3n;!gyxV#7VWLp6nwJD6}0x1j{oD5n;KXWgLJhZoO!m~w=SIg}bFEI1S|
zh!F=TCx<@p^YdGG*<XI%D0&OtO2L^vzPH%YK7{K!+IgwOMMTXbf?$~mcLi0ru5Cf5
zFT}DE{^TM}@!F4lLECsH`5sCKunKV~ainv}NzM!vflA}eYyg$UZ}dy0?L5a)`0<Cw
z*v%9N>=dC7donsG>U8TiPJ;h|i6LD?$8!pCVY`*!O3_r8#pg@t-}{Z}6d#WMv0^so
zZmhc(gJJAZhA)3UmNQZ9@v_slns?VO!<#4A>SZw*nx%SdL$T=za9j0CxQjfFNp-Y1
zjdW#0P@m!$q3}>RU5LNox3{Ckh}@!8e4T^Ya2%~n4Mslw!hZ4j*~9*p&xAk9My9Sh
zsXv;T+0iEM$rYAjFfr{ez7k&1L9C}M6N<mIZFCIeIa?i*a4QG|thbJ~msC37gAf||
z5?7M4Fs1K!HS9_=6=ULO{gMzJ&2TI0{7hQk|MB=txC^u7f%b5XYrTwbOB(Jp)||Sa
zsOYgN=GDK6I=(_drH!2hCvXhjQ$OVJ2W$#ALn~D70YUMnflK;;XAg;bN06OgFh}k^
zsoKKH$6qR+KO59{Izw|g&B*iKH4l%qUH3%K>JS*glIlgH8Lu8mr}8^oSE#@(VE*0D
zj-A0b8x!~fs)@+hnr904y9qCJIjF4TIZA)#B)w}xlR%5p{Pj7JPln9(5|5q_3z4Aq
zT+1FB_$eQHSj=s4YR@KHwT*^UuP<e1#z`IQ?O>jhUY@-BK1~(Rx*Mhi{TvKC|5tlC
zLV>|rfko1$r)uu=D@Wg#<%f;UtoAm(+UK}VjD*}DU)M>xbNcrBtoh08BD#CXe`mZ#
z1maGkn^D)gn;Z`Jo0hm}MCJu^!Ll{rQU~$pO@+Rd*irPG72QKVkcgPNj1}C;F`swT
z6nC$!GfG$T#BQfj{Q0oh{e<)63!hde5)SLEF!#dsv9U2(5|Kj^1X{Q3Ax(EEVs2#$
z!nR*ATm=e}Dn&lW-?=l@RL3bMqjwDW7d$6-x_E4kOMoD{XiuW|!;Rs+_a@fz*0MU9
z)fc3|b8c>CIwtuURozXrkfQyqYmGzmC~lh$T$dLeX?jBu^GOr?Rpc8=N<b7gMSL3`
zT(mVxQkkpmi5wQ01+SlaS+Z6jxfXvBLZu}7*kU`I)yDm8hwC{7L~Phj9COW^{+zb3
zi%Hzcr5Qy}P*NXgU+m0rEyA7#g8|xYf1iTeVue?W2!Eh$zz^h~XsaSjv80Mm%+;V|
z@|30I>M^=@!Z|;<eERslw9##JCOALf=Ukpt*YjaQbA>;bKkI{qYeGTT`4lBD$S+UI
zxNrACzz+YCZUHyY$vZcNp7k1-vY_Gx{cj3uDtae}FZK?TL~}$bbbOB;IHn4_2|!<J
zg6(!cj&(TcNcDQo$v7?Eh5Ru_h5UANfnt`$>|XsM^F4A11l*Pr;*skrDb5kjusvqu
zCb2)NkkW(LfMzi6gU}C%dlpl}_mF6Kup1K<M0R;h|6{+%ALUmlQ~TYZ-&Az`Sr^ms
zTu)x5YscyEhd}GFtOSMX=#ey6ui~N{X{roZim+i~c3vzQG&oaA@Iw8WxAH%n5u%qN
z6ac|!zW>6p#yJ)z)|x}<@^wmxJ|^e~kA&F;p*wpnKYP2M_kyA1sU~_lx=*xx&SZEN
zy+Z}6>@@2kGVEv6gMn>E$!?yGg+Gxt@QbN8em^`I#w_BVoV%)RyE-FsRNJi3?tN-A
z6Z`n`(Ovl^4I^d=Wu>L1+q&a&S%s7dWYQ-^g@=a+9kfrzZ9WiC24O(wIpDhj391A#
zKXKsy-g%~uO)et;+LGq7Q*Zg2Z0^!~AGR}j)Ra#;J>&!8Y$4HBo}>5Uen~Q_n%q2H
z(!|!v%xr{Q;o%hmeRk@wxYwSXJH>$eh@s4zKho+-!Zsn?addow3^$;@@S29e*-2>8
z#{A<p*8PdIFJHZRbE8_X-*DMo_pzKr8f$*455+`aU|{DF;Xr(z6LW?9(H-qH5)*6e
zbw!ha=44_q+tlZa<5gu_V9_TlWYF7!`-@jljX@-Dso5Gr#f?@M<%OhjHLH6o9)WF?
z9cYRjp044j2#t@ACwFE@KT0)}iT*q|FO2P2cBi(Ev5?JZD)7xE0{LT+f;ajOz7K0Y
z8j7nf{bEpb10sM+JFD?6BP{hS{*gVB9zzp3!dDoZ+MSHcJ6&$fx}QvUz7WlqaYZM!
z(_+aEZt{!*IW2)KGY6F`%LN|cPBaHTM6-sj4_=KC(?emQYtIo0{GlX(Gh6Ja@dYI+
zn1;3%bjEYPqK)TqC9*q#MeIyq^&%3F?v;N_<ja?0NNGYy0N~$t2F!vexDr&W4%p&|
z7lh2e^1K9w0x>;A4z5H+CXWbSh)VcSHg0;Uud`<6bslbaHf<$?O`r|qN)ZPP7M2*<
zFw?no?XUS=y_O#zmVGNE=e!r+>IA9csWQdM@?fhLLolzH&k$_NiS$zw_=s=|q;pqF
zNVp|98d>nVhh-h*y}p`|e)mgO<frO$#qJ67XU+$NG%(0^5LxQv()bv!hg6Haw{wxO
zx+X%v50+43s(j0tun8U*O?G+mp=h%0A6c56$}Fzo6<bJRdGLpW2%5BXp<Tn9{+`Ff
zW%bt_Q9cvxE?tzE1E?JSl!;v1-Vd;7#;?t(krq`(Lz;MSNly7_qB_h%d4gtfJ6N~;
z%HmVzk2D2r{jcIVh(hN1R|5`&ZJQfsq~g4HbC+7x(YlRE-EAz^UEfxB-SPJ8R2Fzj
zZc`LKb+BWu@=N9Dh~Fw}hH96IgT-L{J|-mv1^1tKz4-`_n&Id=+qzQwU^k*{(34<`
zibNU@;YEQOGnE(Gtil#*Zi-KJr6L%U@mRtd9OcUx>o9de=dQGzT_@jV(#JGEb8-F_
zTkok$oWK*_@OjK-V9IyD*vy4Z{`(ZM+c}99F#!~EXG6ts#`TXb@bX_ggS?#v>7hEU
zl-@^hX;O*leBFD@5BBhD%l^ojYZx?>YUyOklB@zNJ2&q%TDaaWM5fo|B(O4BH4&di
zNX^-@>3`5mOZcFVJSSQSL{RdbS@6H#1;AE1D1}uEewiU8uf`4`@@HexpXkWN_$@xL
zu55uaMd*OfGbB}DS0f}ORQ&i?W(wlzsioGt#rtc#4k8ze_Jn&}{n7Xb3@~~p5|dpG
z?k9|MH#w+WG>VbhRmP!xdoGppd_dkkrAEt9Z4yQ%XG|;oYmr<an8dM}2XZ&j#;Ehg
z-B{2di!3}QM`~|V!8~@!D2o-G>yuGwSaWEI34s;y>>o{pyiz|Pe&dyFSk<zq;3Zim
zp;$im9-Mi8-Rr_93=QcnKW)2DmzZv_`)hD;aAe&M6+)r5!VA54D{<l4tibM%SW>yu
zW_0`r(&)wN$eZYQ#P0`hbe!Z|+%Jz;VJ1qD=%0Ufmf&aY;LTX~VP5s@e(|E1!9IW^
zDrxn@sfDEChTY)nMs?^WsY&kX7)$k0RF}bn9-dq&dGLnK6Lk)8BiN9XoIy<dA1S$^
z5HLln0gZh4^XlIw`mEZ<BVXQmwBq2sFNHmUiLnd2UMso#1cTYr8n~I5nCQvzt7gT*
zf$DZxFi^bv4=#Hgjojti1~Yua&o9PQW&|FQ7iy#?R3WnHM5pfhd7hMuNze8j(KOw!
zrRB6}|GL<i*X)Hp>b(8(>3OmVJ3G7e9Q_CKxNTfioEhlFfI%nBmq14B_vS)OIeYfk
zO9)~xgbW9>ByJ>_L7KBj@H$daRY$=ckyt0~IlB^n>PJG<#XXn9iuTtke$}EpDQ>K%
z1rcng;5kC|0QltU2R$~~4@YsMk?ZLpOxL?C-)XYcVK}^d3qc9seOqeC2<ygF1{2_f
zEb{uDqf*l*_kWtHU*WG}7SnNc<!@<_IZE6jhNS)7YVL)XJ@eGerXvlPc6-qob`HFC
zB3mAT<HwD>T*EW3{g+>cX#lT!)ublyB0)@NkRkbz#^v7#;=YaUelKUJ=uCnmMRe;b
zlIGrvmda<ii0KC!Ru~%_ugm|^DQI6J=rwWjy?YJUbsEi&&@jUcuckQt0GHJtY95Mz
z8!u5Irx_-e;8!clU{=%Dzg;^x$-f}9P+Qy5%H+-$He$@&!lZ+FHASCj)gDsxZWY!J
zNs8rbGUC&+$}(FyT7qqw`VnI^1IDOIWKi%=V}z(NU2ijjU02?MnJ(*fm(W?Qw3|NJ
ze0;|iwmWZ0+XRs~mQ~Y2ifOIXbe?Z9Mg0lha1(9k1O>hIpHXbH%UyOzK`<8i5_}+>
z^nrNXfjVB1fGXE+y!4|diRdQZu2*IhExzl>&Mk@$ObmE1t|5L1f{2EZk>p)rZQ-!Y
zL^<$K2;o_Iw4{3V-gYNNx8W)?64W5!q$)xi@4)}SP(u|=x2H0MpxvT_N>(rh9jz}5
z!(YQ!gpZ8SUFn*?Gv@UjAyY=z&d-PI)mX@K67K9ksLL{m>UCz;*ZpW*R9$_6fdp7)
zo)!G|o3&xhTZ5H<S6{~v^=0&ujez(bx}ZP5$Wgq{H)N&glbNBcUXpafwL0VS2b_r|
z2Jv+5kP;C3`QhF!sm=OX3M036*WfJn1kr=Q9Fc)$f25O@ksiis;m+3t0k>xra33O>
z9Di|o-#evZ@x2z;-|Xg$dxFXt*Y^4NUW7-`a@QRx$G4An#}br<zbr7tfQ|Yi?DSR_
zq&$-SVjeGEf@I&SFC}~r#yd3iYM>w!j$!WvYy1W@BxkNr%N94vKj6>}8!@zN;eLvw
zafD*OfB){cDUH*oT5`A8?UVOKtj)GUZ1#_@IDN87lSJDiH)mPJMo31$%`MeW>y4zy
zj1HC8L5sQ8i?(=Kh_JR?^;(YFR6giaW;sv2Z_QSe|1P#@Q-tXHCROu4I6)9_p}HS6
zLxk|UH+mc9PQirdQMnkU8_MyfL;ba)OA|bS3e~hvwDm1}<m|1D4)5vidgqlW2;eu)
zH)xfyIRhAC51V*PZLV-x-jJxPacQ<wp#VYE7XUK(Mx;6)2uvbqMZ)Q5kp0@fqW-NY
z2`YxDb?x<hlG?LMQBe;9SSO@aTR5XjZ1*MRXmAJB-P=`VPJBU64tD}H8PWSu1`e=_
zk%(2qyoPYKYHcTBq<y^)+KJbYUqj^P2wq{_**z8WVzYFu6VWNTO~hvit!fmEcUR74
zwjY1)fq_zglVB!9P|Mdh=_W<u0M==b>xzRPGZ03Ic7$_vXqw^Y`x1@>K8QLJcsL9o
zh&Jqf4k@#QO07+vNU=S$sM<~BD1xdbY7u+q4q2hPutrBh!V?@RMQdivdBL-Y=nQFr
z-!J@pJ3)_b(5#;hWV3CS-<6;tT2b;GOu1CDmBkhX?|n-eeO?XyY^*$>A;D^`&{Z<m
zZ_?RkL|5Jdo6kQP5pxj?QQu|YJpz&dT?eodiP*2$p&OfxN%S*>_v|_h31|5hSki*o
z!`z`-raU}6o7uhITcoa$-i1QgZ?EEo3vPnGy_*72mPFL~fNxVFU-FJ(SFsL7fSc$T
zto4Tn{sh^X{Q6U2Awi3|Qss^>-fMnO_rvqey_XGvgU0hJ3K0RMZErJzLx-UXM~`Ja
zOv&P6F2BnYEC;|ATJUP^%r>%<k|0tCYP^Nhw;g%Dae;W=F-6i`wQBMN7r1^O#>;V~
zG_gmtWhg6~L+{PIBGt^-rR5i?voqhVCWKoq8EIbV!@Dgssw^wQjQiFeie<k>&Vtzv
z26@}v0{lxbl*DG)TJaya-xH2=2N@>l|Am_X*+j1s^)$CTie;@uD!PakOEMK@Nq#?1
z`-+WPRIr#y!KYcvX#ZN_&*7=9Kko#e+Cg~CPh`W_x}Cvo<VE#GaG9=WcsCOa%W`tO
zV1$zr^=1g4Q$)7^+3JH%3Ss^?+RK%*idI(5FCW=dlP8CoDk&+gZ_3J}*NCq#R<G~h
zi*-Hh`9bk@D3jx#gVzplTouH0j0}emwmL&;bNaz@TB*&9?+tS)Loe3?wfILTwJglN
zmP6zo95%)AYP2${oBy+u)Ei|X?!Cw#eN!lL$q9e>heFzlia-+Adw5@MyjdGX(zy~?
z%*<4n;qbgHB2hQe;^NB}HK<<5cLqbglStNg|4&6kT-kP?d!NdhBo_$MCrtyPdw8Z9
z!GW3Kf&=O6rZ^waCnHz~!u81=hFddD(2614TE~vkY4_|GbVY2*jhW_q8tza;nHz}(
zWv?{3n_pyGE|z<{ThUXSKlw49X)!&dw<)ggDXWxW$38;e5wb25d>bK5%kHUW>z0eu
z_`&~APO-T1oZm~+M{wv749yv-`9aV9+>%W%XMp9kXnA<@N@m_`C6quVM;76Dyq=Lc
zQz9ZIdoA)2cJ*&Ql35*0B*11jLa^vLM10q{wh;a#!x3Zs@N7Wo`D)d+C>Q4&lLuaC
zI)N)@X5<S~E%iSA+HJr<v31{?%>U0XsE;V=7V%u=kKz3JrCsZM0H;e8*+fA@E*+84
zN0He0@71AnKqzPSUw*s(F(P)EYfoixl9dn6ibGkV;0CJ^en4;#`D-Gm4q2(qlSdjc
zIg?Elp^cS8pN$N24mFdWh2UJK`)8oTve#4yD$M}Yow6Wi>c>HB(t_b481*+&HvfM`
zTSGY|JyMHHd#*;!OpcEYR)!cm>drhZ2?oP%G3t?dGBYzX%1nKL;7jR$B?@4*i5Y*P
z5H^h%A2a~B3jUJ!1{r{S<}xMKE`o>?&9e&}m0VRC5iT!JA8&OYS-G$p5JhB3b(xYQ
z*SI|)A%Ust;YC6`$3Tmj3NbBYKFJ+u_zdB?3}axr$YC`?08C&QQmR}?vC}GTOrgX?
zlB#+3J<mK(vo6>2t#r{TOVc~XSohVtB92_*>v1QtcaKNC4G7Y-0Km&n1eElLCHnJE
z$`J2W)Q*dgm@cU&=KI9EvMlO*{QY^Ih9vJV4KlUWwa=C06iEQ+h{YVa9TyLIuff*#
zKRlkO>X3kL@Y!E*j(ShbNyU{qdE>R#T`SjzhG&(E#^&bNy!|Y9$$|+dk&=SV=U6-O
z9KDz`JwDa%SFEihaX!E)XSw_TYwz4+n!3UOZV)GP3uGA+){1UIhS>(JQwl=?2kK;O
z3_(F4B817K;IQDMl+ub*nM!$thXxAFDKrDy@+h=CYS9S<gKcbw1zNdlZ~_G?6|Z2s
z+wQnoYQUIff1mr;&3ALp`}dviob#P?f;V=?ewR==_&y%5Yz;}EMT?`9Omgc;1Cl|?
zf3(vS&x%2QCQY~K=n3wZrJOFA5#M*uQL$g#6f!%i@!%ki3W~UjT#KII3q0F;+i(du
z9veN8pOu=%B!fk8cR3Y}*$Uq9V{+onF^*rLDniLrR5A%1<XTAdQ4wt;-?M6zhiDGx
zb>P(_dok)l`ygYJ5*O)veZAtEH1Svf#yKu0R`htrrmqVYiapEu^f3x<il6Dr*`V1{
zz;2oW$MlSl1{p0)r+s2(>d@afIx>4+Fqlr1tW6k>VD#<(I3=E%dQaObo7Uy&V-9;3
z)ZW*B_(KC-h$J5solEtvV+(v+1!Y-{2fC>a)w|UZk#9=xa*&nGc{~F-JH)xfuST`|
zTAWN=0Sm|T>Zqfub>)nol=X6o=&o2t1*KNSpP;s*+#3{j6s7A=gd)D%PBzN)cW&dx
zK$$4Noyr4?^le!=F<37hXGrft97`f>1&LTXhrm3BQ+7m%Qb0ZfF;3glF#jhl`xt!`
zo}GHcZt6a<U|{uRg@dtjasfDA?!R=!8^dahwmj1}<MnjkTYF_wXsSA?c0Jyw{wf(w
zos>=~xV?FODV=3{G_>64*MSQcuhp=n7aIs_x;~pUMHW`mxvcU7hh1}6_ue8*O1Ef9
z&(NAltU)1v+?$$GsV0K+&2FL1R7^kiB#AnsaofgTcQUkZ_IYpO?DB`_jnR)TC`VPl
zzp!A=yEdYAX}YkI#d-Yh&j9(jt#eE%Z9Ml;7g_+A>d$e^EHe3(OjK{6Y^$hB#)cor
zWHRH*)wW$M6D@z)2xh6B3?K&Xk1+`C(Y_x?UhKvMllOw{-U@Ep9=xeI$-}#<StULF
zaM<HctVE?+A7Z4YkjXmQTtulw16XX-r4iWg0p{06Nj=L<Vc=RJYCzCrH=JR%DDSd?
zwdT2#<yS-kL8oi%_w9e%erXV|iY9bZt}ft^^FMw4^T=~mHmc*wA&){4Uti}^Fd7Or
zBJ{qOmwntVom~1@GVdNcFaQqh(qDLK-Ez5>u}Idq7HK^)%npdaBJ#?02}+c*zL-0g
zIW7@xp7}l_<MR&MvkXKpKSHq9cFjqKkDM|@6Pk-xnLCz$ZS+?b&Gt^Kgqb(aFNXh1
z&-lcO75Ma^9bfMKpwS_>F(~U8(2g6<)Yxfjulv;6QwAPB4U=?UUDV~~;<O~$EGY*4
znf0PL$HAtutfZ%>M}Uf)o$LeR5)&UfpV_^`?1}K&#5w_EZ1m-|i0pT4*usO3c)X!I
zR#PADo(qn|w!2-{l=dIeWv#IyV+A#fDYMqQ6}j|Jj}`jUb>PMaO#8LL`O_ogRVPBu
zhMT;0d9MmIk0@T%0L+bK=*r1=Rr5CN#lh|NajSUwW0H3d7U{$Lbyt-2&CCZ4M8U;F
zZ5Hir{&{)b2=~PO)f>#U#5hsY4SZ)ALU5<Ec=Di~Q|ol;Y-Br0eH#0O=eJHG0{#<)
zw(KO)i*cYOcjoGVlGktmKP_*9Py&&GTtEsS1&{(r0i*y@04abJKnfrQkOD{nqySO?
pDS#9}3LpiL0!YFCrl8h-!6cShZH@{^lCJ<y(9Vz@+<>@Ke*;LjI~f1~

literal 0
HcmV?d00001

diff --git a/docs/quickstart.md b/docs/quickstart.md
deleted file mode 100644
index ec445dcba..000000000
--- a/docs/quickstart.md
+++ /dev/null
@@ -1,228 +0,0 @@
----
-title: Quickstart
----
-
-# Quickstart
-
-After [installing Outlines](installation.md), the fastest way to get to up to speed with the library is to get acquainted with its few core elements. We advise you to take a quick look at this page to see everything Outlines has to offer before diving in the [documentation](reference/index.md).
-
-
-## Core elements
-
-### Models
-
-The first step when writing a program with Outlines is to initialize a model. Weights will be loaded on the device at this step:
-
-```python
-import outlines
-
-model = outlines.models.transformers(
-    "microsoft/Phi-3-mini-4k-instruct",
-    device="cuda"  # optional device argument, default is cpu
-)
-```
-
-Outlines supports a wide variety of inference engines and model weight types. More details on different models can be found in the [Outlines Models](./reference/models/models.md) documentation page.
-
-### Generation
-
-Once the model is initialized you can build an `outlines.generate` generator. This generator can be called with a prompt directly.
-
-([Outlines Structured Generation Full Documentation](./reference/generation/generation.md))
-
-=== "Text"
-
-    ```python
-    generator = outlines.generate.text(model)
-
-    result = generator("Question: What's 2+2? Answer:", max_tokens=100)
-    print(result)
-    # The answer is 4
-
-	# Outlines also supports streaming output
-    stream = generator.stream("What's 2+2?", max_tokens=4)
-    for i in range(5):
-        token = next(stream)
-        print(repr(token))
-	# '2'
-	# '+'
-	# '2'
-	# ' equals'
-	# '4'
-    ```
-
-=== "Structured"
-
-    Along with typical language model generation behavior via, `outlines.generate.text`, Outlines supports structured generation, which guarantees the tokens generated by the model will follow a predefined structure. Structures can be defined by a regex pattern, JSON schema, python object type, or a Lark grammar defining a parsable language such as SQL or Python.
-
-	Example: using pydantic to enforce a JSON schema
-
-    ```python
-    from enum import Enum
-    from pydantic import BaseModel, constr, conint
-
-    class Character(BaseModel):
-        name: constr(max_length=10)
-        age: conint(gt=18, lt=99)
-        armor: (Enum('Armor', {'leather': 'leather', 'chainmail': 'chainmail', 'plate': 'plate'}))
-        strength: conint(gt=1, lt=100)
-
-    generator = outlines.generate.json(model, Character)
-
-    character = generator(
-        "Generate a new character for my awesome game: "
-        + "name, age (between 1 and 99), armor and strength. "
-        )
-    print(character)
-    # Character(name='Zara', age=25, armor=<Armor.leather: 'leather'>, strength=85)
-	```
-
-## [Deploy using vLLM and FastAPI](./reference/serve/vllm.md)
-
-Outlines can be deployed as a LLM service using [vLLM][vllm]{:target="_blank"} and [FastAPI][fastapi]{:target="_blank"}. The server supports asynchronous processing of incoming requests, and benefits from the performance of vLLM.
-
-First start the server:
-
-```python
-python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-Or you can start the server with Outlines' official Docker image:
-
-```bash
-docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used.
-
-
-You can then query the model in shell by passing a prompt and a [JSON Schema][jsonschema]{:target="_blank"} specification for the structure of the output:
-
-```bash
-curl http://127.0.0.1:8000/generate \
-    -d '{
-        "prompt": "Question: What is a language model? Answer:",
-        "schema": {"type": "string"}
-        }'
-```
-
-Or use the [requests][requests]{:target="_blank"} library from another python program. You can read the [vLLM documentation][vllm]{:target="_blank"} for more details.
-
-## Utilities
-
-### [Prompt templates](./reference/prompting.md)
-
-Prompting can lead to messy code. Outlines' prompt functions are python functions that contain a template for the prompt in their docstring. We use a powerful templating language to allow you to loop over lists, dictionaries, add conditionals, etc. directly from the prompt. When called, a prompt function returns the rendered template:
-
-```python
-from outlines import Template
-
-few_shots = Template.from_string(
-    """{{ instructions }}
-
-    Examples
-    --------
-
-    {% for example in examples %}
-    Q: {{ example.question }}
-    A: {{ example.answer }}
-
-    {% endfor %}
-    Question
-    --------
-
-    Q: {{ question }}
-    A:
-    """
-)
-
-instructions = "Please answer the following question following the examples"
-examples = [
-    {"question": "2+2=?", "answer":4},
-    {"question": "3+3=?", "answer":6}
-]
-question = "4+4 = ?"
-
-prompt = few_shots(instructions, examples, question)
-print(prompt)
-# Please answer the following question following the examples
-
-# Examples
-# --------
-
-# Q: 2+2=?
-# A: 4
-
-# Q: 3+3=?
-# A: 6
-
-# Question
-# --------
-
-# Q: 4+4 = ?
-# A:
-```
-
-### Outlines functions
-
-Once you are done experimenting with a prompt and an output structure, it is useful to be able to encapsulate all of these in a single function that can be called from other parts of the program. This is what `outlines.Function` allows you to do:
-
-=== "function.py"
-
-    ```python
-    from pydantic import BaseModel
-
-    import outlines
-
-
-    tell_a_joke = outlines.Template.from_string(
-        """Tell me a joke about {{ topic }}."""
-    )
-
-    class Joke(BaseModel):
-        setup: str
-        punchline: str
-
-    generate_joke = outlines.Function(
-        tell_a_joke,
-        Joke,
-        "microsoft/Phi-3-mini-4k-instruct"
-    )
-    ```
-
-=== "Call a function"
-
-    ```python
-    from .function import generate_joke
-
-    response = generate_joke("baseball")
-
-    # haha
-    # Joke(setup='Why was the baseball in a bad mood?', punchline='Because it got hit around a lot.')
-    ```
-
-=== "Call a function stored on GitHub"
-
-    You can load a function that is stored on a repository on GitHub directly from Outlines. Say `Someone` stores a function in `joke.py` at the root of the `TheirRepo` repository:
-
-    ```python
-    import outlines
-
-    joke = outlines.Function.from_github("Someone/TheirRepo/joke")
-    response = joke("baseball")
-    ```
-    It make it easier for the community to collaborate on the infinite number of use cases enabled by these models!
-
-
-## Going further
-
-If you need more inspiration you can take a look at the [cookbook](cookbook/index.md) or watch [Remi Louf's AI Engineer World’s Fair Presentation on Outlines](https://www.youtube.com/live/R0X7mPagRiE?t=775s). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/dottxt-ai/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
-
-
-[pydantic]: https://docs.pydantic.dev/latest
-[jsonschema]: https://json-schema.org/
-[fastapi]: https://fastapi.tiangolo.com/
-[cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
-[ebnf]: https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form
-[requests]: https://requests.readthedocs.io/en/latest/
-[vllm]: https://docs.vllm.ai/en/latest/index.html
diff --git a/docs/reference/generation/cfg.md b/docs/reference/generation/cfg.md
deleted file mode 100644
index f0104ec63..000000000
--- a/docs/reference/generation/cfg.md
+++ /dev/null
@@ -1,154 +0,0 @@
-# Grammar-structured generation
-
-You can pass any context-free grammar in the EBNF format and Outlines will generate an output that is valid to this grammar:
-
-```python
-from outlines import models, generate
-
-arithmetic_grammar = """
-    ?start: expression
-
-    ?expression: term (("+" | "-") term)*
-
-    ?term: factor (("*" | "/") factor)*
-
-    ?factor: NUMBER
-           | "-" factor
-           | "(" expression ")"
-
-    %import common.NUMBER
-"""
-
-model = models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = generate.cfg(model, arithmetic_grammar)
-sequence = generator(
-  "Alice had 4 apples and Bob ate 2. "
-  + "Write an expression for Alice's apples:"
-)
-
-print(sequence)
-# (8-2)
-```
-
-###### Disclaimer
-
-!!! Note "Experimental"
-
-    Outlines current **community-contributed** implementation of CFG-structured generation is experimental. This does not reflect the performance of [.txt](https://dottxt.co)'s product, where we have optimized grammar-structured generation to be as fast as regex-structured generation. Additionally, it does not fully align with the approach described in our [technical report](https://arxiv.org/pdf/2307.09702), aside from its use of incremental/partial parsing. This feature is still a work in progress, requiring performance enhancements and bug fixes for an ideal implementation. For more details, please see our [grammar-related open issues on GitHub](https://github.com/dottxt-ai/outlines/issues?q=is%3Aissue+is%3Aopen+label%3Agrammar).
-
-!!! Note "Greedy"
-
-    To mitigate performance issues, CFG-structured generation will use rejection sampling and iterate over the candidate tokens highest logit first,, completing once a single valid token ID is selected. This is effectively greedy generation.
-
-## Ready-to-use grammars
-
-Outlines contains a (small) library of grammars that can be imported and use directly. We can rewrite the previous example as:
-
-```python
-from outlines import models, generate
-
-arithmetic_grammar = outlines.grammars.arithmetic
-
-model = models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = generate.cfg(model, arithmetic_grammar)
-sequence = generator(
-  "Alice had 4 apples and Bob ate 2. "
-  + "Write an expression for Alice's apples:"
-)
-
-print(sequence)
-# (8-2)
-```
-
-The following grammars are currently available:
-
-- Arithmetic grammar via `outlines.grammars.arithmetic`
-- JSON grammar via `outlines.grammars.json`
-
-If you would like more grammars to be added to the repository, please open an [issue](https://github.com/dottxt-ai/outlines/issues) or a [pull request](https://github.com/dottxt-ai/outlines/pulls).
-
-
-## Grammar guide
-
-A grammar is a list of rules and terminals that define a *language*:
-
-- Terminals define the vocabulary of the language; they may be a string, regular expression or combination of these and other terminals.
-- Rules define the structure of that language; they are a list of terminals and rules.
-
-Outlines uses the [Lark library](https://github.com/lark-parser/lark) to make Large Language Models generate text in a language of a grammar, it thus uses grammars defined in a format that Lark understands, based on the [EBNF syntax](https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form). Read the [Lark documentation](https://lark-parser.readthedocs.io/en/stable/grammar.html) for more details on grammar, the following is a small primer that should help get your started.
-
-In the following we will define a [LOGO-like toy language](https://github.com/lark-parser/lark/blob/master/examples/turtle_dsl.py) for python's [turtle](https://docs.python.org/3/library/turtle.html) library.
-
-### Terminals
-
-A turtle can take 4 different `MOVEMENT` move instructions: forward (`f`), backward (`b`), turn right (`r`) and turn left (`l`). It can take `NUMBER` number of steps in each direction, and draw lines in a specified `COLOR`. These define the vocabulary of our language:
-
-```ebnf
-MOVEMENT: "f"|"b"|"r"|"l"
-COLOR: LETTER+
-
-%import common.LETTER
-%import common.INT -> NUMBER
-%import common.WS
-%ignore WS
-```
-
-The lines that start with `%` are called "directive". They allow to import pre-defined terminals and rules, such as `LETTER` and `NUMBER`. `LETTER+` is a regular expressions, and indicates that a `COLOR` is made of at least one `LETTER`. The last two lines specify that we will ignore white spaces (`WS`) in the grammar.
-
-### Rules
-
-We now need to define our rules, by decomposing instructions we can send to the turtle via our python program. At each line of the program, we can either choose a direction and execute a given number of steps, change the color used to draw the pattern. We can also choose to start filling, make a series of moves, and stop filling. We can also choose to repeat a series of move.
-
-We can easily write the first two rules:
-
-```ebnf
-instruction: MOVEMENT NUMBER   -> movement
-           | "c" COLOR [COLOR] -> change_color
-```
-
-where `movement` and `change_color` represent aliases for the rules. A whitespace implied concatenating the elements, and `|` choosing either of the elements. The `fill` and `repeat` rules are slightly more complex, since they apply to a code block, which is made of instructions. We thus define a new `code_block`  rule that refers to `instruction` and finish implementing our rules:
-
-```ebnf
-instruction: MOVEMENT NUMBER            -> movement
-           | "c" COLOR [COLOR]          -> change_color
-           | "fill" code_block          -> fill
-           | "repeat" NUMBER code_block -> repeat
-
-code_block: "{" instruction "}"
-```
-
-We can now write the full grammar:
-
-```ebnf
-start: instruction+
-
-instruction: MOVEMENT NUMBER            -> movement
-            | "c" COLOR [COLOR]          -> change_color
-            | "fill" code_block          -> fill
-            | "repeat" NUMBER code_block -> repeat
-
-code_block: "{" instruction+ "}"
-
-MOVEMENT: "f"|"b"|"l"|"r"
-COLOR: LETTER+
-
-%import common.LETTER
-%import common.INT -> NUMBER
-%import common.WS
-%ignore WS
-```
-
-Notice the `start` rule, which defines the starting point of the grammar, i.e. the rule with which a program must start. This full grammars allows us to parse programs such as:
-
-```python
-c red yellow
-    fill { repeat 36 {
-        f200 l170
-    }}
-```
-
-The result of the parse, the parse tree, can then easily be translated into a Python program that uses the `turtle` library to draw a pattern.
-
-### Next steps
-
-This section provides a very brief overview of grammars and their possibilities. Check out the [Lark documentation](https://lark-parser.readthedocs.io/en/stable/index.html) for more thorough explanations and more examples.
diff --git a/docs/reference/generation/choices.md b/docs/reference/generation/choices.md
deleted file mode 100644
index aed5af5ab..000000000
--- a/docs/reference/generation/choices.md
+++ /dev/null
@@ -1,16 +0,0 @@
-# Multiple choices
-
-Oultines allows you to make sure the generated text is chosen between different options:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.choice(model, ["skirt", "dress", "pen", "jacket"])
-answer = generator("Pick the odd word out: skirt, dress, pen, jacket")
-
-```
-
-!!! Note "Performance"
-
-    `generation.choice` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate from the same list of choices several times make sure that you only call `generate.choice` once.
diff --git a/docs/reference/generation/creating_grammars.md b/docs/reference/generation/creating_grammars.md
deleted file mode 100644
index 78e41282a..000000000
--- a/docs/reference/generation/creating_grammars.md
+++ /dev/null
@@ -1,99 +0,0 @@
-# Overview
-
-Outlines allows the use of [Lark](https://github.com/lark-parser/lark) grammars to guide generation. These grammars are used to construct parsers that filter out incompatible tokens during the generation process The result is a generation that adheres to the grammar's production rules.
-
-# Primer on Creating Grammars
-
-To create grammars for Outlines, a solid understanding of Lark grammars is necessary. Here's how you can get started:
-
-- Read Lark's grammars documentations [here](https://lark-parser.readthedocs.io/en/latest/grammar.html).
-- Review Outlines' existing grammars [here](/outlines/grammars).
-
-
-# Compatibility With Outlines
-
-It's important to note that not all Lark grammars work with Outlines. Changes may be necessary to ensure compatability.
-
-### LALR(1) Parser
-
-Outlines utilizes Larks LALR(1) parser, meaning the grammar must be unambiguous at least up to the next token (one token lookahead). Read Lark's official LALR(1) parser documentation [here](https://lark-parser.readthedocs.io/en/stable/parsers.html#lalr-1).
-
-If your grammar is ambiguous, you will recieve the following error at runtime:
-
-```
-GrammarError: Reduce/Reduce collision in Terminal('B') between the following rules:
-```
-
-### Regex Terminal Restrictions
-
-Outlines converts terminals to finite state machines using the [Interegular](https://github.com/MegaIng/interegular/) library. Not all regular expressions work with Interegular, mitigation is described in the subsections which follow.
-
-
-#### Avoid Lookarounds
-
-Examples of removing lookaround while maintaining the same functionality
-
-##### Example: Escaped String
-
-From Outlines' modified `ESCAPED_STRING` in [common.lark](/outlines/grammars/common.lark).
-
-Before:
-```
-_STRING_INNER: /.*?/
-_STRING_ESC_INNER: _STRING_INNER /(?<!\\)(\\\\)*?/
-
-ESCAPED_STRING : "\"" _STRING_ESC_INNER "\""
-```
-
-After:
-```
-_NON_CONTROL_CHAR: /([^"\\\x00-\x1F\x7F-\x9F])/
-_ESCAPED_CHAR: /\\/ (_NON_CONTROL_CHAR | /\\/ | /"/)
-ESCAPED_STRING_INNER: _NON_CONTROL_CHAR | _ESCAPED_CHAR
-ESCAPED_STRING: /"/ ESCAPED_STRING_INNER* /"/
-```
-
-#### Avoid Backreferences
-
-Backreferences, for example `([ab]^*)\1`, cannot be simulated by a finite state machine, and will result in an error if used.
-
-# Creating a Valid Grammar
-
-You can use Outlines' test suite to verify your grammar.
-
-### 1) Create Your Grammar
-
-Create your grammar file named `your_new_grammar.lark`, adhering to the guidelines provided above. Add it to `outlines/grammars/` (ensure attribution is included and license is compatible).
-
-Update `outlines/grammars.py` with a line including your grammar.
-
-### 2) Test Your Grammar
-
-Test grammar for false negatives, ensure sample grammars can be generated:
-- Add valid example outputs which are compliant with the grammar to `tests/benchmark/cfg_samples/your_new_grammar/`
-- Run the tests for your grammar via `pytest -s tests/fsm/test_cfg_guide.py::test_cfg_grammar_sample -k "your_new_grammar"`
-
-Test grammar for false positives, ensure invalid outputs aren't generated.
-
-Currently there isn't a builtin false positive testing utility. It is recommended you smoke test via
-```
-from outlines import models, generate, grammars
-model = models.transformers("mistralai/Mistral-7B-v0.1")
-generator = generate.cfg(model, grammars.your_new_grammar)
-result = generator(<your prompt to generate output for your grammar>)
-print(result)
-```
-
-# Converting
-There are a few tools available for converting from other grammars to lark. These tools serve as a starting point. However, you will typically need to make additional adjustments to ensure full compatibility and proper functioning within Outlines.
-
-Tools:
-- Larks built in "Nearley-to-Lark" converter https://lark-parser.readthedocs.io/en/latest/tools.html
-- Convert ANTLR4 to Lark (Note, most antlr4 grammars are not LALR(1) compatible, so will require additional tweaking) https://github.com/kaby76/Domemtech.Trash/blob/main/src/trconvert/readme.md
-- Extract EBNF from Yacc files https://www.bottlecaps.de/rr/ui
-
-Reference Grammars:
-- Github Lark Grammars https://github.com/search?q=path%3A*.lark&type=code
-- Github Nearley Grammars https://github.com/search?q=path%3A*.ne+%22-%3E%22&type=code
-- Antlr4 grammars https://github.com/antlr/grammars-v4/
-- Grammar zoo https://slebok.github.io/zoo/index.html#html
diff --git a/docs/reference/generation/custom_fsm_ops.md b/docs/reference/generation/custom_fsm_ops.md
deleted file mode 100644
index 5c4be96fa..000000000
--- a/docs/reference/generation/custom_fsm_ops.md
+++ /dev/null
@@ -1,37 +0,0 @@
-# Custom FSM Operations
-
-Outlines is fast because it compiles regular expressions into an index ahead of inference. To do so we use the equivalence between regular expressions and Finite State Machines (FSMs), and the library [interegular](https://github.com/MegaIng/interegular) to perform the translation.
-
-Alternatively, one can pass a FSM built using `integular` directly to structure the generation.
-
-## Example
-
-### Using the `difference` operation
-
-In the following example we build a fsm which recognizes only the strings valid to the first regular expression but not the second. In particular, it will prevent the words "pink" and "elephant" from being generated:
-
-```python
-import interegular
-from outlines import models, generate
-
-
-list_of_strings_pattern = """\["[^"\s]*"(?:,"[^"\s]*")*\]"""
-pink_elephant_pattern = """.*(pink|elephant).*"""
-
-list_of_strings_fsm = interegular.parse_pattern(list_of_strings_pattern).to_fsm()
-pink_elephant_fsm = interegular.parse_pattern(pink_elephant_pattern).to_fsm()
-
-difference_fsm = list_of_strings_fsm - pink_elephant_fsm
-
-difference_fsm_fsm.accepts('["a","pink","elephant"]')
-# False
-difference_fsm_fsm.accepts('["a","blue","donkey"]')
-# True
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.fsm(model, difference_fsm)
-response = generator("Don't talk about pink elephants")
-```
-
-To see the other operations available, consult [interegular's documentation](https://github.com/MegaIng/interegular/blob/master/interegular/fsm.py).
diff --git a/docs/reference/generation/format.md b/docs/reference/generation/format.md
deleted file mode 100644
index 749baa8b6..000000000
--- a/docs/reference/generation/format.md
+++ /dev/null
@@ -1,23 +0,0 @@
-# Type constraints
-
-We can ask completions to be restricted to valid python types:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.format(model, int)
-answer = generator("When I was 6 my sister was half my age. Now I’m 70 how old is my sister?")
-print(answer)
-# 67
-```
-
-The following types are currently available:
-
-- int
-- float
-- bool
-- datetime.date
-- datetime.time
-- datetime.datetime
-- We also provide [custom types](types.md)
diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
deleted file mode 100644
index 930ad9d22..000000000
--- a/docs/reference/generation/generation.md
+++ /dev/null
@@ -1,216 +0,0 @@
----
-title: Generation
----
-
-# Generation
-
-Once an [Outlines model](../models/models.md) is constructed you can use `outlines.generate` to generate text. Standard LLM generation is possible via `outlines.generate.text`, along with a variety of structured generation methods described below. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
-
-Before generating text, you must construct an `outlines.model`. Example:
-
-```python
-import outlines
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
-```
-### Text generator
-
-```python
-generator = outlines.generate.text(model)
-
-result = generator("Question: What's 2+2? Answer:", max_tokens=100)
-print(result)
-# The answer is 4
-
-# Outlines also supports streaming output
-stream = generator.stream("What's 2+2?", max_tokens=4)
-for i in range(5):
-	token = next(stream)
-	print(repr(token))
-# '2'
-# '+'
-# '2'
-# ' equals'
-# '4'
-```
-
-### [Multi-label classification](./choices.md)
-
-Outlines allows you to do multi-label classification by guiding the model so it can only output either of the specified choices:
-
-```python
-import outlines
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = outlines.generate.choice(model, ["Blue", "Red", "Yellow"])
-
-color = generator("What is the closest color to Indigo? ")
-print(color)
-# Blue
-```
-
-### [JSON-structured generation](./json.md)
-
-Outlines can guide models so that they output valid JSON **100%** of the time. You can either specify the structure using [Pydantic][pydantic]{:target="_blank"} or a string that contains a [JSON Schema][jsonschema]{:target="_blank"}:
-
-=== "Pydantic"
-
-    ```python
-    from enum import Enum
-    from pydantic import BaseModel, constr, conint
-
-    import outlines
-
-    class Armor(str, Enum):
-        leather = "leather"
-        chainmail = "chainmail"
-        plate = "plate"
-
-
-    class Character(BaseModel):
-        name: constr(max_length=10)
-        age: conint(gt=18, lt=99)
-        armor: Armor
-        strength: conint(gt=1, lt=100)
-
-    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-    generator = outlines.generate.json(model, Character)
-
-    character = generator(
-        "Generate a new character for my awesome game: "
-        + "name, age (between 1 and 99), armor and strength. "
-        )
-    print(character)
-    # name='Orla' age=21 armor=<Armor.plate: 'plate'> strength=8
-    ```
-
-=== "JSON Schema"
-
-    ```python
-    import outlines
-
-    schema = """{
-        "$defs": {
-            "Armor": {
-                "enum": ["leather", "chainmail", "plate"],
-                "title": "Armor",
-                "type": "string"
-            }
-        },
-        "properties": {
-            "name": {"maxLength": 10, "title": "Name", "type": "string"},
-            "age": {"title": "Age", "type": "integer"},
-            "armor": {"$ref": "#/$defs/Armor"},
-            "strength": {"title": "Strength", "type": "integer"}\
-        },
-        "required": ["name", "age", "armor", "strength"],
-        "title": "Character",
-        "type": "object"
-    }"""
-
-    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-    generator = outlines.generate.json(model, schema)
-    character = generator(
-        "Generate a new character for my awesome game: "
-        + "name, age (between 1 and 99), armor and strength. "
-        )
-    print(character)
-    # {'name': 'Yuki', 'age': 24, 'armor': 'plate', 'strength': 3}
-    ```
-
-!!! Note
-
-    We advise you to constrain the length of the strings fields when first testing your schema, especially with small models.
-
-### [Grammar-structured generation](./cfg.md)
-
-Outlines also allows to generate text that is valid to any [context-free grammar][cfg]{:target="_blank"} (CFG) in the [EBNF format][ebnf]{:target="_blank"}. Grammars can be intimidating, but they are a very powerful tool! Indeed, they determine the syntax of every programming language, valid chess moves, molecule structure, can help with procedural graphics generation, etc.
-
-Here we show a simple example of a grammar that defines arithmetic operations:
-
-```python
-from outlines import models, generate
-
-arithmetic_grammar = """
-    ?start: sum
-
-    ?sum: product
-        | sum "+" product   -> add
-        | sum "-" product   -> sub
-
-    ?product: atom
-        | product "*" atom  -> mul
-        | product "/" atom  -> div
-
-    ?atom: NUMBER           -> number
-         | "-" atom         -> neg
-         | "(" sum ")"
-
-    %import common.NUMBER
-    %import common.WS_INLINE
-
-    %ignore WS_INLINE
-"""
-
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = generate.cfg(model, arithmetic_grammar, max_tokens=100)
-
-result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
-print(result)
-# 5+5+5+5+5
-```
-
-
-EBNF grammars can be cumbersome to write. This is why Outlines provides grammar definitions in the `outlines.grammars.` module
-
-```python
-from outlines import models, generate, grammars
-
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = generate.cfg(model, grammars.arithmetic, max_tokens=100)
-
-result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
-print(result)
-# 5+5+5+5+5
-```
-
-The available grammars are listed [here](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
-
-
-### [Regex-structured generation](./regex.md)
-
-Slightly simpler, but no less useful, Outlines can generate text that is in the language of a [regular expression](https://www.regular-expressions.info/tutorial.html). For instance to force the model to generate IP addresses:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
-
-regex_str = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-generator = generate.regex(model, regex_str)
-
-result = generator("What is the IP address of localhost?\nIP: ")
-print(result)
-# 127.0.0.100
-```
-
-### [Generate a given Python type](./types.md)
-
-We provide a shortcut to regex-structured generation for simple use cases. Pass a Python type to the `outlines.generate.format` function and the LLM will output text that matches this type:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = generate.format(model, int)
-
-result = generator("What is 2+2?")
-print(result)
-# 4
-```
-
-
-[jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
-[pydantic]: https://docs.pydantic.dev/latest
-[cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
-[ebnf]: https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form
diff --git a/docs/reference/generation/json.md b/docs/reference/generation/json.md
deleted file mode 100644
index 05dc69654..000000000
--- a/docs/reference/generation/json.md
+++ /dev/null
@@ -1,178 +0,0 @@
-# JSON structured generation
-
-Outlines can make any open source model return a JSON object that follows a structure that is specified by the user. This is useful whenever we want the output of the model to be processed by code downstream: code does not understand natural language but rather the structured language it has been programmed to understand.
-
-There are mostly two reasons why someone would want to get an output formatted as JSON from a LLM:
-
-1. Parse the answer (e.g. with Pydantic), store it somewhere, return it to a user, etc.
-2. Call a function with the result
-
-Outlines has you covered in both cases! Indeed, to define the structure of the JSON you want the model to follow you can either provide a Pydantic model, or a function. No need to duplicate code!
-
-## Using Pydantic
-
-Outlines can infer the structure of the output from a Pydantic model. The result is an instance of the model that contains the values returned by the LLM:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models, generate
-
-
-class User(BaseModel):
-    name: str
-    last_name: str
-    id: int
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.json(model, User)
-result = generator(
-    "Create a user profile with the fields name, last_name and id"
-)
-print(result)
-# User(name="John", last_name="Doe", id=11)
-```
-
-!!! Note "JSON and whitespaces"
-
-    By default Outlines prevents the model from generating json with syntactic newlines, tabs, or multiple spaces. The default `whitespace_pattern` is `r"[ ]?"`. Small models tend to enter an infinite repetition loop if the `whitespace_pattern` allows infinite spacing. If you would like to allow the model to generate multiple tabs, newlines, and spaces, you can set the whitespace pattern as follows:
-
-    ```python
-    generator = generate.json(model, User, whitespace_pattern=r"[\n\t ]*")
-    ```
-
-!!! Note "Performance"
-
-    `generation.json` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate several times with the same schema make sure that you only call `generate.json` once.
-
-!!! Tip "Custom types"
-
-    Outlines provides [custom Pydantic types](types.md) so you do not have to write regular expressions for common types, such as phone numbers or zip codes.
-
-## Using a JSON Schema
-
-Instead of a Pydantic model you can pass a string that represents a [JSON Schema](https://json-schema.org/) specification to `generate.json`:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models
-from outlines import generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
-schema = """
-{
-  "title": "User",
-  "type": "object",
-  "properties": {
-    "name": {"type": "string"},
-    "last_name": {"type": "string"},
-    "id": {"type": "integer"}
-  },
-  "required": ["name", "last_name", "id"]
-}
-"""
-
-generator = generate.json(model, schema)
-result = generator(
-    "Create a user profile with the fields name, last_name and id"
-)
-print(result)
-# User(name="John", last_name="Doe", id=11)
-```
-
-## From a function's signature
-
-Outlines can infer the structure of the output from the signature of a function. The result is a dictionary, and can be passed directly to the function using the usual dictionary expansion syntax `**`:
-
-```python
-from outlines import models
-from outlines import generate
-
-def add(a: int, b: int):
-    return a + b
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.json(model, add)
-result = generator("Return two integers named a and b respectively. a is odd and b even.")
-
-print(add(**result))
-# 3
-```
-
-A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
-
-
-## From a dynamic JSON schema builder - GenSON
-
-Outlines integrated [GenSON](https://github.com/wolverdude/GenSON) builders to be able to dynamicly declare JSON schemas. It can be used as follow:
-
-```python
-from genson import SchemaBuilder
-
-from outlines import models
-from outlines import generate
-
-builder = SchemaBuilder()
-builder.add_schema({"type": "object", "properties": {}})
-builder.add_object({"name": "Toto", "age": 5})
-
-model = models.transformers(
-    "HuggingFaceTB/SmolLM2-135M",
-    device="auto",
-)
-generator = generate.json(model, builder)
-
-res = generator("Return a json of a young boy")
-print(res)
-# {"name": "Ben", "age": 10}
-```
-
-Anytime you are updating the schema through the builder, you need to redifine the outline generator to include these changes. From the the previous example:
-
-```python
-from genson import SchemaBuilder
-
-from outlines import models
-from outlines import generate
-
-builder = SchemaBuilder()
-builder.add_schema({"type": "object", "properties": {}})
-builder.add_object({"name": "Toto", "age": 5})
-
-model = models.transformers(
-    "HuggingFaceTB/SmolLM2-135M",
-    device="auto",
-)
-generator = generate.json(model, builder)
-
-res = generator("Return a json of a young boy")
-print(res)
-# {"name": "Ben", "age": 10}
-
-builder.add_object({"hobby": "sports"})
-generator = generate.json(model, builder)
-
-res = generator("Return a json of a youg boy whose hobby is coding")
-print(res)
-# {"name": "Ben", "age": 10, "hobby": "coding"}
-```
-
-!!! Note
-
-    Beware of [GenSON](https://github.com/wolverdude/GenSON)'s behavior regarding dynamic amending of schemas through their builder. Here is an example of how you could lose `required` informations and generate json with missing fields:
-
-    ```python
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"name": "Toto", "age": 5})
-
-    print(builder.to_schema())
-    # {'$schema': 'http://json-schema.org/schema#', 'type': 'object', 'properties': {'name': {'type': 'string'}, 'age': {'type': 'integer'}}, 'required': ['age', 'name']}
-
-    builder.add_object({"hobby": "sport"})
-    print(builder.to_schema())
-    # {'name': {'type': 'string'}, 'age': {'type': 'integer'}, 'hobby': {'type': 'string'}}}
-    ```
diff --git a/docs/reference/generation/regex.md b/docs/reference/generation/regex.md
deleted file mode 100644
index 23df35567..000000000
--- a/docs/reference/generation/regex.md
+++ /dev/null
@@ -1,27 +0,0 @@
-# Regular expressions
-
-Outlines can guarantee that the text generated by the LLM will be valid to a regular expression:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
-generator = generate.regex(
-    model,
-    r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)",
-)
-
-prompt = "What is the IP address of the Google DNS servers? "
-answer = generator(prompt, max_tokens=30)
-
-print(answer)
-# What is the IP address of the Google DNS servers?
-# 2.2.6.1
-```
-
-If you find yourself using `generate.regex` to restrict the answers' type you can take a look at [type-structured generation](types.md) instead.
-
-!!! Note "Performance"
-
-    `generate.regex` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate several times using the same regular expression make sure that you only call `generate.regex` once.
diff --git a/docs/reference/generation/structured_generation_explanation.md b/docs/reference/generation/structured_generation_explanation.md
deleted file mode 100644
index aa27a7a85..000000000
--- a/docs/reference/generation/structured_generation_explanation.md
+++ /dev/null
@@ -1,54 +0,0 @@
-# How does Outlines work?
-
-
-Language models generate text token by token, using the previous token sequence as input and sampled logits as output. This document explains the structured generation process, where only legal tokens are considered for the next step based on a predefined automata, e.g. a regex-defined [finite-state machine](https://en.wikipedia.org/wiki/Finite-state_machine) (FSM) or [Lark](https://lark-parser.readthedocs.io/en/stable/) grammar.`
-
-
-## Worked Example
-
-Let's consider a worked example with a pattern for whole and decimal numbers:
-
-`^\d*(\.\d+)?$`.
-
-### Creating Automata
-
-The pattern is first converted into an automata. Below is a brief explanation of the automata conversion and its representation.
-
-**Automata Diagram:**
-
-```mermaid
-graph LR
-    node0("1-9") --> node1("1-9")
-    node1 --> node1
-    node1 --> nodeEND{{END}}
-    node1 --> nodePeriod(".")
-    nodePeriod --> node2("1-9")
-    node2 --> node2
-    node2 --> nodeEND{{END}}
-```
-
-### Generating a Token
-
-Let's assume that we're in the middle of generation, and so far "748" has been generated. Here is the automata with the current state highlighted in green, with the legal next characters being another number (1-9), a dot (.), or end of sequence.
-
-```mermaid
-graph LR
-    node0("1-9") --> node1("1-9")
-    node1 --> node1
-    node1 --> nodeEND{{END}}
-    node1 --> nodePeriod(".")
-    nodePeriod --> node2("1-9")
-    node2 --> node2
-    node2 --> nodeEND{{END}}
-
-    style node1 fill:#090
-```
-
-Generating a token requires the following steps:
-
-- Feed the previous input sequence ("748") into the language model.
-- Language model runs a forward pass and produces token logits.
-- Outlines logits processor sets the probability of illegal tokens to 0%.
-- A token is sampled from the set of legal tokens.
-
-![Generation and Logits Processing Flow Chart](../../assets/images/logits_processing_diagram.svg)
diff --git a/docs/reference/generation/types.md b/docs/reference/generation/types.md
deleted file mode 100644
index eb6d7382b..000000000
--- a/docs/reference/generation/types.md
+++ /dev/null
@@ -1,83 +0,0 @@
-# Custom types
-
-Outlines provides custom Pydantic types so you can focus on your use case rather than on writing regular expressions:
-
-| Category | Type | Import | Description |
-|:--------:|:----:|:-------|:------------|
-| ISBN | 10 & 13 | `outlines.types.ISBN` | There is no guarantee that the [check digit][wiki-isbn] will be correct |
-| Airport | IATA | `outlines.types.airports.IATA` | Valid [airport IATA codes][wiki-airport-iata] |
-| Country | alpha-2 code | `outlines.types.airports.Alpha2` | Valid [country alpha-2 codes][wiki-country-alpha-2] |
-|  | alpha-3 code | `outlines.types.countries.Alpha3` | Valid [country alpha-3 codes][wiki-country-alpha-3] |
-|  | numeric code | `outlines.types.countries.Numeric` | Valid [country numeric codes][wiki-country-numeric] |
-|  | name | `outlines.types.countries.Name` | Valid country names |
-|  | flag | `outlines.types.countries.Flag` | Valid flag emojis |
-| | email | `outlines.types.Email` | Valid email address |
-
-Some types require localization. We currently only support US types, but please don't hesitate to create localized versions of the different types and open a Pull Request. Localized types are specified using `types.locale` in the following way:
-
-```python
-from outlines import types
-
-types.locale("us").ZipCode
-types.locale("us").PhoneNumber
-```
-
-Here are the localized types that are currently available:
-
-| Category | Locale | Import | Description |
-|:--------:|:----:|:-------|:------------|
-| Zip code | US | `ZipCode` | Generate US Zip(+4) codes |
-| Phone number  | US | `PhoneNumber` | Generate valid US phone numbers |
-
-
-You can use these types in Pydantic schemas for JSON-structured generation:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models, generate, types
-
-# Specify the locale for types
-locale = types.locale("us")
-
-class Client(BaseModel):
-    name: str
-    phone_number: locale.PhoneNumber
-    zip_code: locale.ZipCode
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.json(model, Client)
-result = generator(
-    "Create a client profile with the fields name, phone_number and zip_code"
-)
-print(result)
-# name='Tommy' phone_number='129-896-5501' zip_code='50766'
-```
-
-Or simply with `outlines.generate.format`:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models, generate, types
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.format(model, types.locale("us").PhoneNumber)
-result = generator(
-    "Return a US Phone number: "
-)
-print(result)
-# 334-253-2630
-```
-
-
-We plan on adding many more custom types. If you have found yourself writing regular expressions to generate fields of a given type, or if you could benefit from more specific types don't hesite to [submit a PR](https://github.com/dottxt-ai/outlines/pulls) or [open an issue](https://github.com/dottxt-ai/outlines/issues/new/choose).
-
-
-[wiki-isbn]: https://en.wikipedia.org/wiki/ISBN#Check_digits
-[wiki-airport-iata]: https://en.wikipedia.org/wiki/IATA_airport_code
-[wiki-country-alpha-2]: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2
-[wiki-country-alpha-3]: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3
-[wiki-country-numeric]: https://en.wikipedia.org/wiki/ISO_3166-1_numeric
diff --git a/docs/reference/index.md b/docs/reference/index.md
deleted file mode 100644
index 0cd42ba88..000000000
--- a/docs/reference/index.md
+++ /dev/null
@@ -1,15 +0,0 @@
-# Reference
-
-## Structured generation
-
-While LLM capabilities are increasingly impressive, we can make their output more reliable by steering the generation. Outlines thus offers mechanisms to specify high level constraints on text completions by generative language models.
-
-Stopping sequence
-By default, language models stop generating tokens after and <EOS> token was generated, or after a set maximum number of tokens. Their output can be verbose, and for practical purposes it is often necessary to stop the generation after a given sequence has been found instead. You can use the stop_at keyword argument when calling the model with a prompt:
-
-```python
-import outlines.models as models
-
-complete = models.openai("gpt-4o-mini")
-expert = complete("Name an expert in quantum gravity.", stop_at=["\n", "."])
-```
diff --git a/docs/reference/models/exllamav2.md b/docs/reference/models/exllamav2.md
deleted file mode 100644
index a4e727840..000000000
--- a/docs/reference/models/exllamav2.md
+++ /dev/null
@@ -1,15 +0,0 @@
-# ExllamaV2
-
-The `outlines.models.exllamav2` model requires a Logits Processor component for compatibility with Outlines structured generation. While ExLlamaV2 doesn't natively support this feature, a third-party fork provides the necessary functionality. You can install it with the following command:
-
-```bash
-pip install git+https://github.com/lapp0/exllamav2@sampler-logits-processor
-```
-
-Install other requirements:
-
-```bash
-pip install transformers torch
-```
-
-*Coming soon*
diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
deleted file mode 100644
index 51b62eca8..000000000
--- a/docs/reference/models/llamacpp.md
+++ /dev/null
@@ -1,230 +0,0 @@
-# Llama.cpp
-
-Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library][llamacpp]. Llamacpp allows to run quantized models on machines with limited compute.
-
-!!! Note "Installation"
-
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration. See the [installation section](#installation) for instructions to install `llama-cpp-python` with CUDA, Metal, ROCm and other backends. To get started quickly you can also run:
-
-    ```bash
-    pip install "outlines[llamacpp]"
-    ```
-
-## Load the model
-
-You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
-
-```python
-from outlines import models
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-```
-
-This will download the model files to the hub cache folder and load the weights in memory.
-
-You can also initialize the model by passing the path to the weights on your machine. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
-
-```python
-from outlines import models
-from llama_cpp import Llama
-
-llm = Llama("./phi-2.Q4_K_M.gguf")
-model = models.LlamaCpp(llm)
-```
-
-If you need more control, you can pass the same keyword arguments to the model as you would pass in the [llama-ccp-library][llamacpp]:
-
-```python
-from outlines import models
-
-model = models.llamacpp(
-    "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf"
-    n_ctx=512,  # to set the context length value
-)
-```
-
-**Main parameters:**
-
-| Parameters | Type | Description | Default |
-|------------|------|-------------|---------|
-| `n_gpu_layers`| `int` | Number of layers to offload to GPU. If -1, all layers are offloaded | `0` |
-| `split_mode` | `int` | How to split the model across GPUs. `1` for layer-wise split, `2` for row-wise split | `1` |
-| `main_gpu` | `int` | Main GPU | `0` |
-| `tensor_split` | `Optional[List[float]]` | How split tensors should be distributed across GPUs. If `None` the model is not split. | `None` |
-| `n_ctx` | `int` | Text context. Inference from the model if set to `0` | `0` |
-| `n_threads` | `Optional[int]` | Number of threads to use for generation. All available threads if set to `None`.| `None` |
-| `verbose` | `bool` | Print verbose outputs to `stderr` | `False` |
-
-See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters.
-
-### Load the model on GPU
-
-!!! Note
-
-    [Make sure](#cuda) that you installed `llama-cpp-python` with GPU support.
-
- To load the model on GPU, pass `n_gpu_layers=-1`:
-
-```python
-from outlines import models
-
-model = models.llamacpp(
-    "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf",
-    n_gpu_layers=-1,  # to use GPU acceleration
-)
-```
-
-This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
-
-### Load LoRA adapters
-
-You can load LoRA adapters dynamically:
-
-```python
-from outlines import models, generate
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-generator = generate.text(model)
-answer_1 = generator("prompt")
-
-model.load_lora("./path/to/adapter.gguf")
-answer_2 = generator("prompt")
-```
-
-To load another adapter you need to re-initialize the model. Otherwise the adapter will be added on top of the previous one:
-
-```python
-from outlines import models
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-model.load_lora("./path/to/adapter1.gguf")  # Load first adapter
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-model.load_lora("./path/to/adapter2.gguf")  # Load second adapter
-```
-
-## Generate text
-
-In addition to the parameters described in the [text generation section](../text.md) you can pass extra keyword arguments, for instance to set sampling parameters not exposed in Outlines' public API:
-
-```python
-from outlines import models, generate
-
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-generator = generate.text(model)
-
-answer = generator("A prompt", presence_penalty=0.8)
-```
-
-**Extra keyword arguments:**
-
-The value of the keyword arguments you pass to the generator suspersede the values set when initializing the sampler or generator. All extra sampling methods and repetition penalties are disabled by default.
-
-| Parameters | Type | Description | Default |
-|------------|------|-------------|---------|
-| `suffix` | `Optional[str]` | A suffix to append to the generated text. If `None` no suffix is added. | `None` |
-| `echo` | `bool` | Whether to preprend the prompt to the completion. | `False` |
-| `seed` | `int` | The random seed to use for sampling. | `None` |
-| `max_tokens` | `Optional[int]` | The maximum number of tokens to generate. If `None` the maximum number of tokens depends on `n_ctx`. | `16` |
-| `frequence_penalty` | `float` | The penalty to apply to tokens based on their frequency in the past 64 tokens. | `0.0` |
-| `presence_penalty` | `float` | The penalty to apply to tokens based on their presence in the past 64 tokens. | `0.0` |
-| `repeat_penalty` | `float` | The penalty to apply to repeated tokens in the past 64 tokens. | `1.` |
-| `stopping_criteria` | `Optional[StoppingCriteriaList]` | A list of stopping criteria to use. | `None`
-| `logits_processor` | `Optional[LogitsProcessorList]` | A list of logits processors to use. The logits processor used for structured generation will be added to this list. | `None`
-| `temperature` | `float` | The temperature to use for sampling | `1.0` |
-| `top_p` | `float` | The top-p value to use for [nucleus sampling][degeneration]. | `1.` |
-| `min_p` | `float` | The min-p value to use for [minimum-p sampling][minimum-p]. | `0.` |
-| `typical_p` | `float` | The p value to use for [locally typical sampling][locally-typical]. | `1.0` |
-| `stop` | `Optional[Union[str, List[str]]]` | A list of strings that stop generation when encountered. | `[]` |
-| `top_k` |  `int` | The top-k value used for [top-k sampling][top-k]. Negative value to consider all logit values. | `-1.` |
-| `tfs_z` | `float` | The [tail-free sampling][tail-free] parameter. | `1.0` |
-| `mirostat_mode` | `int` | The [mirostat sampling][mirostat] mode. | `0` |
-| `mirostat_tau` | `float` | The target cross-entropy for [mirostat sampling][mirostat].| `5.0` |
-| `mirostat_eta` | `float` | The learning rate used to update `mu` in [mirostat sampling][mirostat]. | `0.1` |
-
-See the [llama-cpp-python documentation][llama-cpp-python-call] for the full and up-to-date list of parameters and the [llama.cpp code][llama-cpp-sampling-params] for the default values of other
-sampling parameters.
-
-### Streaming
-
-
-## Installation
-
-You need to install the `llama-cpp-python` library to use the llama.cpp integration.
-
-### CPU
-
-For a *CPU-only* installation run:
-
-```bash
-pip install llama-cpp-python
-```
-
-!!! Warning
-
-    Do not run this command if you want support for BLAS, Metal or CUDA. Follow the instructions below instead.
-
-### CUDA
-
-```bash
-CMAKE_ARGS="-DLLAMA_CUDA=on" pip install llama-cpp-python
-```
-
-It is also possible to install pre-built wheels with CUDA support (Python 3.10 and above):
-
-```bash
-pip install llama-cpp-python \
-  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/<cuda-version>
-```
-
-Where `<cuda-version>` is one of the following, depending on the version of CUDA installed on your system:
-
-- `cu121` for CUDA 12.1
-- `cu122` for CUDA 12.2
-- `cu123` CUDA 12.3
-
-### Metal
-
-```bash
-CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
-```
-
-It is also possible to install pre-build wheels with Metal support (Python 3.10 or above, MacOS 11.0 and above):
-
-```bash
-pip install llama-cpp-python \
-  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/metal
-```
-
-### OpenBLAS
-
-```bash
-CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
-```
-
-### Other backend
-
-`llama.cpp` supports many other backends. Refer to the [llama.cpp documentation][llama-cpp-python-install] to use the following backends:
-
-- CLBast (OpenCL)
-- hipBLAS (ROCm)
-- Vulkan
-- Kompute
-- SYCL
-
-
-
-
-[llamacpp]: https://github.com/abetlen/llama-cpp-python
-[llama-cpp-python-call]: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
-[llama-cpp-python-install]: https://github.com/abetlen/llama-cpp-python/tree/08b16afe11e7b42adec2fed0a781123383476045?tab=readme-ov-file#supported-backends
-[llama-cpp-sampling-params]: https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
-[mirostat]: https://arxiv.org/abs/2007.14966
-[degeneration]: https://arxiv.org/abs/1904.09751
-[top-k]: https://arxiv.org/abs/1805.04833
-[minimum-p]: https://github.com/ggerganov/llama.cpp/pull/3841
-[locally-typical]: https://arxiv.org/abs/2202.00666
-[tail-free]: https://www.trentonbricken.com/Tail-Free-Sampling
diff --git a/docs/reference/models/mlxlm.md b/docs/reference/models/mlxlm.md
deleted file mode 100644
index d435b9c1f..000000000
--- a/docs/reference/models/mlxlm.md
+++ /dev/null
@@ -1,93 +0,0 @@
-# mlx-lm
-
-Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
-
-!!! Note "Installation"
-
-    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To get started quickly you can also run:
-
-    ```bash
-    pip install "outlines[mlxlm]"
-    ```
-
-
-## Load the model
-
-You can initialize the model by passing the name of the repository on the HuggingFace Hub. The official repository for mlx-lm supported models is [mlx-community](https://huggingface.co/mlx-community).
-
-```python
-from outlines import models
-
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-```
-
-This will download the model files to the hub cache folder and load the weights in memory.
-
-The arguments `model_config` and `tokenizer_config` are available to modify loading behavior. For example, per the `mlx-lm` [documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms#supported-models), you must set an eos_token for `qwen/Qwen-7B`. In outlines you may do so via
-
-```
-model = models.mlxlm(
-    "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit",
-    tokenizer_config={"eos_token": "<|endoftext|>", "trust_remote_code": True},
-)
-```
-
-**Main parameters:**
-
-(Subject to change. Table based on [mlx-lm.load docstring](https://github.com/ml-explore/mlx-examples/blob/main/llms/mlx_lm/utils.py#L429))
-
-| Parameters         | Type   | Description                                                                                      | Default |
-|--------------------|--------|--------------------------------------------------------------------------------------------------|---------|
-| `tokenizer_config` | `dict` | Configuration parameters specifically for the tokenizer. Defaults to an empty dictionary.        | `{}`    |
-| `model_config`     | `dict` | Configuration parameters specifically for the model. Defaults to an empty dictionary.            | `{}`    |
-| `adapter_path`     | `str`  | Path to the LoRA adapters. If provided, applies LoRA layers to the model.                        | `None`  |
-| `lazy`             | `bool` | If False, evaluate the model parameters to make sure they are loaded in memory before returning. | `False` |
-
-
-## Generate text
-
-You may generate text using the parameters described in the [text generation documentation](../text.md).
-
-With the loaded model, you can generate text or perform structured generation, e.g.
-
-```python
-from outlines import models, generate
-
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-generator = generate.text(model)
-
-answer = generator("A prompt", temperature=2.0)
-```
-
-## Streaming
-
-You may creating a streaming iterable with minimal changes
-
-```python
-from outlines import models, generate
-
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-generator = generate.text(model)
-
-for token_str in generator.text("A prompt", temperature=2.0):
-    print(token_str)
-```
-
-## Structured
-
-You may perform structured generation with mlxlm to guarantee your output will match a regex pattern, json schema, or lark grammar.
-
-Example: Phone number generation with pattern `"\\+?[1-9][0-9]{7,14}"`:
-
-```python
-from outlines import models, generate
-
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-
-phone_number_pattern = "\\+?[1-9][0-9]{7,14}"
-generator = generate.regex(model, phone_number_pattern)
-
-model_output = generator("What's Jennys Number?\n")
-print(model_output)
-# '8675309'
-```
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
deleted file mode 100644
index eab70f2a1..000000000
--- a/docs/reference/models/models.md
+++ /dev/null
@@ -1,56 +0,0 @@
----
-title: Models
----
-
-# Models
-
-Outlines supports generation using a number of inference engines (`outlines.models`). Loading a model using outlines follows a similar interface between inference engines:
-
-```python
-import outlines
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-model = outlines.models.transformers_vision("llava-hf/llava-v1.6-mistral-7b-hf")
-model = outlines.models.vllm("microsoft/Phi-3-mini-128k-instruct")
-model = outlines.models.llamacpp(
-    "microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf"
-)
-model = outlines.models.exllamav2("bartowski/Phi-3-mini-128k-instruct-exl2")
-model = outlines.models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
-
-model = outlines.models.openai(
-    "gpt-4o-mini",
-    api_key=os.environ["OPENAI_API_KEY"]
-)
-```
-
-
-# Feature Matrix
-|                   | [Transformers](transformers.md) | [Transformers Vision](transformers_vision.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md)* |
-|-------------------|--------------|---------------------|------|-----------|-----------|-------|---------|
-| **Device**        |              |                     |      |           |           |       |         |
-| Cuda              | ✅           | ✅                  | ✅   | ✅        | ✅        | ❌    | N/A     |
-| Apple Silicon     | ✅           | ✅                  | ❌   | ✅        | ✅        | ✅    | N/A     |
-| x86 / AMD64       | ✅           | ✅                  | ❌   | ✅        | ✅        | ❌    | N/A     |
-| **Sampling**      |              |                     |      |           |           |       |         |
-| Greedy            | ✅           | ✅                  | ✅   | ✅*       | ✅        | ✅    | ❌      |
-| Multinomial       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Multiple Samples  | ✅           | ✅                  |      | ❌        |           | ❌    | ✅      |
-| Beam Search       | ✅           | ✅                  | ✅   | ❌        | ✅        | ❌    | ❌      |
-| **Generation**    |              |                     |      |           |           |       |         |
-| Batch             | ✅           | ✅                  | ✅   | ❌        | ?         | ❌    | ❌      |
-| Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
-| Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| **Structured**    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| JSON Schema       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Choice            | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Regex             | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
-| Grammar           | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
-
-
-## Caveats
-
-- OpenAI doesn't support structured generation due to limitations in their API and server implementation.
-- `outlines.generate` ["Structured"](../generation/generation.md) includes methods such as `outlines.generate.regex`, `outlines.generate.json`, `outlines.generate.cfg`, etc.
-- MLXLM only supports Apple Silicon.
-- llama.cpp greedy sampling available via multinomial with `temperature = 0.0`.
diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
deleted file mode 100644
index 0777ad8b7..000000000
--- a/docs/reference/models/openai.md
+++ /dev/null
@@ -1,204 +0,0 @@
-# OpenAI and compatible APIs
-
-!!! Installation
-
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively:
-
-    ```bash
-    pip install "outlines[openai]"
-    ```
-
-## OpenAI models
-
-Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.openai`:
-
-```python
-from outlines import models
-
-
-model = models.openai("gpt-4o-mini")
-model = models.openai("gpt-4o")
-```
-
-Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.AsyncOpenAI` as keyword arguments:
-
-```python
-import os
-from outlines import models
-
-
-model = models.openai(
-    "gpt-4o-mini",
-    api_key=os.environ["OPENAI_API_KEY"]
-)
-```
-
-The following table enumerates the possible parameters. Refer to the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list.
-
-**Parameters:**
-
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
-| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
-| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
-| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if no specified. | `None` |
-| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
-| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
-| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
-| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
-| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
-
-## Azure OpenAI models
-
-Outlines also supports Azure OpenAI models:
-
-```python
-from outlines import models
-
-
-model = models.azure_openai(
-    "azure-deployment-name",
-    "gpt-4o-mini",
-    api_version="2024-07-18",
-    azure_endpoint="https://example-endpoint.openai.azure.com",
-)
-```
-
-!!! Question "Why do I need to specify model and deployment name?"
-
-    The model name is needed to load the correct tokenizer for the model. The tokenizer is necessary for structured generation.
-
-
-You can pass any parameter you would pass to `openai.AsyncAzureOpenAI`. You can consult the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/lib/azure.py) for an up-to-date list.
-
-**Parameters:**
-
-
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `azure_endpoint` | `str` | Azure endpoint, including the resource. Infered from `AZURE_OPENAI_ENDPOINT` if not specified | `None` |
-| `api_version` | `str` | API version. Infered from `AZURE_OPENAI_API_KEY` if not specified | `None` |
-| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
-| `azure_ad_token` | `str` | Azure active directory token. Inference from `AZURE_OPENAI_AD_TOKEN` if not specified | `None` |
-| `azure_ad_token_provider` | `AzureADTokenProvider` | A function that returns an Azure Active Directory token | `None` |
-| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
-| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
-| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if not specified. | `None` |
-| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
-| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
-| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
-| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
-| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
-
-## Models that follow the OpenAI standard
-
-Outlines supports models that follow the OpenAI standard. You will need to initialize the OpenAI client properly configured and pass it to `outlines.models.openai`
-
-```python
-import os
-from openai import AsyncOpenAI
-from outlines import models
-from outlines.models.openai import OpenAIConfig
-
-
-client = AsyncOpenAI(
-    api_key=os.environ.get("PROVIDER_KEY"),
-    base_url="http://other.provider.server.com"
-)
-config = OpenAIConfig("model_name")
-model = models.openai(client, config)
-```
-
-!!! Warning
-
-    You need to pass the async client to be able to do batch inference.
-
-## Structured Generation Support
-
-Outlines provides support for [OpenAI Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode) via `outlines.generate.json`, `outlines.generate.choice`
-
-```python
-from pydantic import BaseModel, ConfigDict
-import outlines.models as models
-from outlines import generate
-
-model = models.openai("gpt-4o-mini")
-
-class Person(BaseModel):
-    model_config = ConfigDict(extra='forbid')  # required for openai
-    first_name: str
-    last_name: str
-    age: int
-
-generator = generate.json(model, Person)
-generator("current indian prime minister on january 1st 2023")
-# Person(first_name='Narendra', last_name='Modi', age=72)
-
-generator = generate.choice(model, ["Chicken", "Egg"])
-print(generator("Which came first?"))
-# Chicken
-```
-
-!!! Warning
-
-    Structured generation support only provided to OpenAI-compatible endpoints which conform to OpenAI's standard. Additionally, `generate.regex` and `generate.cfg` are not supported.
-
-
-## Advanced configuration
-
-For more advanced configuration option, such as support proxy, please consult the [OpenAI SDK's documentation](https://github.com/openai/openai-python):
-
-
-```python
-from openai import AsyncOpenAI, DefaultHttpxClient
-from outlines import models
-from outlines.models.openai import OpenAIConfig
-
-
-client = AsyncOpenAI(
-    base_url="http://my.test.server.example.com:8083",
-    http_client=DefaultHttpxClient(
-        proxies="http://my.test.proxy.example.com",
-        transport=httpx.HTTPTransport(local_address="0.0.0.0"),
-    ),
-)
-config = OpenAIConfig("model_name")
-model = models.openai(client, config)
-```
-
-It is possible to specify the values for `seed`, `presence_penalty`, `frequence_penalty`, `top_p` by passing an instance of `OpenAIConfig` when initializing the model:
-
-```python
-from outlines.models.openai import OpenAIConfig
-from outlines import models
-
-
-config = OpenAIConfig(
-    presence_penalty=1.,
-    frequency_penalty=1.,
-    top_p=.95,
-    seed=0,
-)
-model = models.openai("gpt-4o-mini", config)
-```
-
-## Monitoring API use
-
-It is important to be able to track your API usage when working with OpenAI's API. The number of prompt tokens and completion tokens is directly accessible via the model instance:
-
-```python
-from openai import AsyncOpenAI
-import outlines.models
-
-
-model = models.openai("gpt-4o")
-
-print(model.prompt_tokens)
-# 0
-
-print(model.completion_tokens)
-# 0
-```
-
-These numbers are updated every time you call the model.
diff --git a/docs/reference/models/tgi.md b/docs/reference/models/tgi.md
deleted file mode 100644
index 3f0568103..000000000
--- a/docs/reference/models/tgi.md
+++ /dev/null
@@ -1,3 +0,0 @@
-# Text-generation-inference (TGI)
-
-TGI uses Outlines to provide structured generation, see [their documentation](https://huggingface.co/docs/text-generation-inference/en/basic_tutorials/using_guidance).
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
deleted file mode 100644
index 3649b9b13..000000000
--- a/docs/reference/models/transformers.md
+++ /dev/null
@@ -1,148 +0,0 @@
-# transformers
-
-
-!!! Installation
-
-    You need to install the `transformer`, `datasets` and `torch` libraries to be able to use these models in Outlines, or alternatively:
-
-    ```bash
-    pip install "outlines[transformers]"
-    ```
-
-
-Outlines provides an integration with the `torch` implementation of causal models in the [transformers][transformers] library. You can initialize the model by passing its name:
-
-```python
-from outlines import models
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
-```
-
-If you need more fine-grained control you can also initialize the model and tokenizer separately:
-
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from outlines import models
-
-llm = AutoModelForCausalLM.from_pretrained("gpt2", output_attentions=True)
-tokenizer = AutoTokenizer.from_pretrained("gpt2")
-model = models.Transformers(llm, tokenizer)
-```
-
-## Using Logits Processors
-
-There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
-
-1. Use Outlines generation wrapper, `outlines.models.transformers`
-2. Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
-
-Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
-
-### Using `outlines.models.transformers`
-
-```python
-import outlines
-
-time_regex_pattern = r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?"
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
-generator = outlines.generate.regex(model, time_regex_pattern)
-
-output = generator("The the best time to visit a dentist is at ")
-print(output)
-# 2:30 pm
-```
-
-### Using models initialized via the `transformers`  library
-
-```python
-import outlines
-import transformers
-
-
-model_uri = "microsoft/Phi-3-mini-4k-instruct"
-
-outlines_tokenizer = outlines.models.TransformerTokenizer(
-    transformers.AutoTokenizer.from_pretrained(model_uri)
-)
-phone_number_logits_processor = outlines.processors.RegexLogitsProcessor(
-    "\\+?[1-9][0-9]{7,14}",  # phone number pattern
-    outlines_tokenizer,
-)
-
-generator = transformers.pipeline('text-generation', model=model_uri)
-
-output = generator(
-    "Jenny gave me her number it's ",
-	logits_processor=transformers.LogitsProcessorList([phone_number_logits_processor])
-)
-print(output)
-# [{'generated_text': "Jenny gave me her number it's 2125550182"}]
-# not quite 8675309 what we expected, but it is a valid phone number
-```
-
-[transformers]: https://github.com/huggingface/transformers
-
-
-## Alternative Model Classes
-
-`outlines.models.transformers` defaults to `transformers.AutoModelForCausalLM`, which is the appropriate class for most standard large language models, including Llama 3, Mistral, Phi-3, etc.
-
-However other variants with unique behavior can be used as well by passing the appropriate class.
-
-### Mamba
-
-[Mamba](https://github.com/state-spaces/mamba) is a transformers alternative which employs memory efficient, linear-time decoding.
-
-To use Mamba with outlines you must first install the necessary requirements:
-```
-pip install causal-conv1d>=1.2.0 mamba-ssm torch transformers
-```
-
-Then you can either create an Mamba-2 Outlines model via
-```python
-import outlines
-
-model = outlines.models.mamba("state-spaces/mamba-2.8b-hf")
-```
-
-or explicitly with
-```python
-import outlines
-from transformers import MambaForCausalLM
-
-model = outlines.models.transformers(
-    "state-spaces/mamba-2.8b-hf",
-    model_class=MambaForCausalLM
-)
-```
-
-
-
-Read [`transformers`'s documentation](https://huggingface.co/docs/transformers/en/model_doc/mamba) for more information.
-
-### Encoder-Decoder Models
-
-You can use encoder-decoder (seq2seq) models like T5 and BART with Outlines.
-
-Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
-
-T5 Example:
-```python
-import outlines
-from transformers import AutoModelForSeq2SeqLM
-
-model_pile_t5 = outlines.models.transformers(
-    model_name="EleutherAI/pile-t5-large",
-    model_class=AutoModelForSeq2SeqLM,
-)
-```
-
-Bart Example:
-```python
-model_bart = outlines.models.transformers(
-    model_name="facebook/bart-large",
-    model_class=AutoModelForSeq2SeqLM,
-)
-```
diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
deleted file mode 100644
index 451e54a8f..000000000
--- a/docs/reference/models/transformers_vision.md
+++ /dev/null
@@ -1,115 +0,0 @@
-# Transformers Vision
-
-Outlines allows seamless use of [vision models](https://huggingface.co/learn/computer-vision-course/en/unit4/multimodal-models/tasks-models-part1).
-
-`outlines.models.transformers_vision` shares interfaces with, and is based on [outlines.models.transformers](./transformers.md).
-
-Tasks supported include
-
-- image + text -> text
-- video + text -> text
-
-
-
-## Example: Using [Llava-Next](https://huggingface.co/docs/transformers/en/model_doc/llava_next) Vision Models
-
-Install dependencies
-`pip install torchvision pillow flash-attn`
-
-Create the model
-```python
-import outlines
-from transformers import LlavaNextForConditionalGeneration
-
-model = outlines.models.transformers_vision(
-    "llava-hf/llava-v1.6-mistral-7b-hf",
-    model_class=LlavaNextForConditionalGeneration,
-	device="cuda",
-)
-```
-
-Create convenience function to load a `PIL.Image` from URL
-```python
-from PIL import Image
-from io import BytesIO
-from urllib.request import urlopen
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-```
-
-### Describing an image
-
-```python
-description_generator = outlines.generate.text(model)
-description_generator(
-    "<image> detailed description:",
-    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")]
-)
-```
-
-> This is a color photograph featuring a Siamese cat with striking blue eyes. The cat has a creamy coat and a light eye color, which is typical for the Siamese breed. Its features include elongated ears, a long, thin tail, and a striking coat pattern. The cat is sitting in an indoor setting, possibly on a cat tower or a similar raised platform, which is covered with a beige fabric, providing a comfortable and soft surface for the cat to rest or perch. The surface of the wall behind the cat appears to be a light-colored stucco or plaster.
-
-#### Multiple Images
-
-To include multiple images in your prompt you simply add more `<image>` tokens to the prompt
-
-```python
-image_urls = [
-	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-3.png",  # triangle
-	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-11.png",  # hexagon
-]
-description_generator = outlines.generate.text(model)
-description_generator(
-    "<image><image>What shapes are present?",
-    list(map(img_from_url, image_urls)),
-)
-```
-
-> There are two shapes present. One shape is a hexagon and the other shape is an triangle. '
-
-
-### Classifying an Image
-
-```python
-pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
-planet_generator = outlines.generate.regex(model, pattern)
-
-planet_generator(
-    "What planet is this: <image>",
-    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg")]
-)
-```
-
-> Saturn
-
-
-### Extracting Structured Image data
-
-```python
-from pydantic import BaseModel
-from typing import List, Optional
-
-class ImageData(BaseModel):
-    caption: str
-    tags_list: List[str]
-    object_list: List[str]
-    is_photo: bool
-
-image_data_generator = outlines.generate.json(model, ImageData)
-
-image_data_generator(
-    "<image> detailed JSON metadata:",
-    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg")]
-)
-```
-
-> `ImageData(caption='An astronaut on the moon', tags_list=['moon', 'space', 'nasa', 'americanflag'], object_list=['moon', 'moon_surface', 'space_suit', 'americanflag'], is_photo=True)`
-
-
-## Resources
-
-### Choosing a model
-- https://mmbench.opencompass.org.cn/leaderboard
-- https://huggingface.co/spaces/WildVision/vision-arena
diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
deleted file mode 100644
index 8789b588e..000000000
--- a/docs/reference/models/vllm.md
+++ /dev/null
@@ -1,235 +0,0 @@
-# vLLM
-
-
-!!! Note "Installation"
-
-    You need to install the `vllm` library to use the vLLM integration. See the [installation section](#installation) for instructions to install vLLM for CPU or ROCm. To get started you can also run:
-
-    ```bash
-    pip install "outlines[vllm]"
-    ```
-
-## Load the model
-
-Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
-
-
-```python
-from outlines import models
-
-model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
-```
-
-Or alternatively:
-
-```python
-import vllm
-from outlines import models
-
-llm = vllm.LLM("microsoft/Phi-3-mini-4k-instruct")
-model = models.VLLM(llm)
-```
-
-
-Models are loaded from the [HuggingFace hub](https://huggingface.co/).
-
-
-!!! Warning "Device"
-
-    The default installation of vLLM only allows to load models on GPU. See the [installation instructions](#installation) to run models on CPU.
-
-
-You can pass any parameter that you would normally pass to `vllm.LLM`, as keyword arguments:
-
-```python
-from outlines import models
-
-model = models.vllm(
-    "microsoft/Phi-3-mini-4k-instruct",
-    trust_remote_code=True,
-    gpu_memory_utilization=0.7
-)
-```
-
-**Main parameters:**
-
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `tokenizer_mode`| `str`  | "auto" will use the fast tokenizer if available and "slow" will always use the slow tokenizer. | `auto`
-| `trust_remote_code`| `bool` | Trust remote code when downloading the model and tokenizer. | `False` |
-| `tensor_parallel_size`| `int` | The number of GPUs to use for distributed execution with tensor parallelism.| `1` |
-| `dtype`| `str` | The data type for the model weights and activations. Currently, we support `float32`, `float16`, and `bfloat16`. If `auto`, we use the `torch_dtype` attribute specified in the model config file. However, if the `torch_dtype` in the config is `float32`, we will use `float16` instead.| `auto` |
-| `quantization`| `Optional[str]` | The method used to quantize the model weights. Currently, we support "awq", "gptq" and "squeezellm". If None, we first check the `quantization_config` attribute in the model config file. If that is None, we assume the model weights are not quantized and use `dtype` to determine the data type of the weights.| `None` |
-| `revision`| `Optional[str]` | The specific model version to use. It can be a branch name, a tag name, or a commit id.| `None` |
-| `tokenizer_revision`| `Optional[str]`| The specific tokenizer version to use. It can be a branch name, a tag name, or a commit id.| `None` |
-| `gpu_memory_utilization`| `float` | The ratio (between 0 and 1) of GPU memory to reserve for the model weights, activations, and KV cache. Higher values will increase the KV cache size and thus improve the model's throughput. However, if the value is too high, it may cause out-of-memory (OOM) errors.| `0.9` |
-| `swap_space`| `int` | The size (GiB) of CPU memory per GPU to use as swap space. This can be used for temporarily storing the states of the requests when their `best_of` sampling parameters are larger than 1. If all requests will have `best_of=1`, you can safely set this to 0. Otherwise, too small values may cause out-of-memory (OOM) errors.| 4 |
-| `enforce_eager`| `bool` | Whether to enforce eager execution. If True, we will disable CUDA graph and always execute the model in eager mode. If False, we will use CUDA graph and eager execution in hybrid.| `False` |
-| `enable_lora` | `bool` | Whether to enable loading LoRA adapters | `False` |
-
-See the [vLLM code](https://github.com/vllm-project/vllm/blob/8f44facdddcf3c704f7d6a2719b6e85efc393449/vllm/entrypoints/llm.py#L72) for a list of all the available parameters.
-
-### Use quantized models
-
-vLLM supports AWQ, GPTQ and SqueezeLLM quantized models:
-
-
-```python
-from outlines import models
-
-model = models.vllm("TheBloke/Llama-2-7B-Chat-AWQ", quantization="awq")
-model = models.vllm("TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", quantization="gptq")
-model = models.vllm("https://huggingface.co/squeeze-ai-lab/sq-llama-30b-w4-s5", quantization="squeezellm")
-```
-
-!!! Warning "Dependencies"
-
-    To use AWQ model you need to install the autoawq library `pip install autoawq`.
-
-    To use GPTQ models you need to install the autoGTPQ and optimum libraries `pip install auto-gptq optimum`.
-
-
-### Multi-GPU usage
-
-To run multi-GPU inference with vLLM you need to set the `tensor_parallel_size` argument to the number of GPUs available when initializing the model. For instance to run inference on 2 GPUs:
-
-
-```python
-from outlines import models
-
-model = models.vllm(
-    "microsoft/Phi-3-mini-4k-instruct"
-    tensor_parallel_size=2
-)
-```
-
-### Load LoRA adapters
-
-You can load LoRA adapters and alternate between them dynamically:
-
-```python
-from outlines import models
-
-model = models.vllm("facebook/opt-350m", enable_lora=True)
-model.load_lora("ybelkaa/opt-350m-lora")  # Load LoRA adapter
-model.load_lora(None)  # Unload LoRA adapter
-```
-
-## Generate text
-
-In addition to the parameters described in the [text generation section](../text.md) you can pass an instance of `SamplingParams` directly to any generator via the `sampling_params` keyword argument:
-
-```python
-from vllm.sampling_params import SamplingParams
-from outlines import models, generate
-
-
-model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
-params = SamplingParams(n=2, frequency_penalty=1., min_tokens=2)
-answer = generator("A prompt", sampling_params=params)
-```
-
-This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
-
-!!! Note
-
-    The values passed via the `SamplingParams` instance supersede the other arguments to the generator or the samplers.
-
-**`SamplingParams` attributes:**
-
-| Parameters | Type             | Description            | Default |
-|:-----------|------------------|:-----------------------|---------|
-| `n` | `int` | Number of output sequences to return for the given prompt. | `1` |
-| `best_of` | `Optional[int]` | Number of output sequences that are generated from the prompt. From these `best_of` sequences, the top `n` sequences are returned. `best_of` must be greater than or equal to `n`. This is treated as the beam width when `use_beam_search` is True. By default, `best_of` is set to `n`. | `None` |
-| `presence_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens.| `0.0` |
-| `frequency_penalty` | `float` | Float that penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens. | `0.0`
-| `repetition_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the prompt and the generated text so far. Values > 1 encourage the model to use new tokens, while values < 1 encourage the model to repeat tokens. | `1.0` |
-| `temperature` | `float` | Float that controls the randomness of the sampling. Lower values make the model more deterministic, while higher values make the model more random. Zero means greedy sampling. | `1.0` |
-| `top_p` | `float` |  Float that controls the cumulative probability of the top tokens to consider. Must be in (0, 1]. Set to 1 to consider all tokens. | `1.0` |
-| `top_k` | `int` |  Integer that controls the number of top tokens to consider. Set to -1 to consider all tokens. | `-1` |
-| `min_p` |`float` |  Float that represents the minimum probability for a token to be considered, relative to the probability of the most likely token. Must be in [0, 1]. Set to 0 to disable this. | `0.0` |
-| `seed` | `Optional[int]` | Random seed to use for the generation. | `None` |
-| `use_beam_search` | `bool` |  Whether to use beam search instead of sampling. | `False` |
-| `length_penalty` | `float` | Float that penalizes sequences based on their length. Used in beam search.  | `1.0` |
-| `early_stopping` | `Union[bool, str]` |  Controls the stopping condition for beam search. It accepts the following values: `True`, where the generation stops as soon as there are `best_of` complete candidates; `False`, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; `"never"`, where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm). | `False` |
-| `stop` | `Optional[Union[str, List[str]]]` |  List of strings that stop the generation when they are generated. The returned output will not contain the stop strings. | `None` |
-| `stop_token_ids` | `Optional[List[int]]` |  List of tokens that stop the generation when they are generated. The returned output will contain the stop tokens unless the stop tokens are special tokens. | `None` |
-| `include_stop_str_in_output` | `bool` |  Whether to include the stop strings in output text. Defaults to False. | `False` |
-| `ignore_eos` | `bool` |  Whether to ignore the EOS token and continue generating tokens after the EOS token is generated. | `False` |
-| `max_tokens` | `int` |  Maximum number of tokens to generate per output sequence. | `16` |
-| `min_tokens` | `int` | Minimum number of tokens to generate per output sequence before EOS or stop_token_ids can be generated | `0` |
-| `skip_special_tokens` | `bool` | Whether to skip special tokens in the output. | `True` |
-| `spaces_between_special_tokens` | `bool` |  Whether to add spaces between special tokens in the output.  Defaults to True. | `True` |
-
-### Streaming
-
-!!! Warning
-
-    Streaming is not available for the offline vLLM integration.
-
-
-## Installation
-
-By default the vLLM library is installed with pre-commpiled C++ and CUDA binaries and will only run on GPU:
-
-```python
-pip install vllm
-```
-
-### CPU
-
-You need to have the `gcc` compiler installed on your system. Then you will need to install vLLM from source. First clone the repository:
-
-```bash
-git clone https://github.com/vllm-project/vllm.git
-cd vllm
-```
-
-Install the Python packages needed for the installation:
-
-```bash
-pip install --upgrade pip
-pip install wheel packaging ninja setuptools>=49.4.0 numpy
-pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/cpu
-```
-
-and finally run:
-
-```bash
-VLLM_TARGET_DEVICE=cpu python setup.py install
-```
-
-See the [vLLM documentation][vllm-install-cpu] for more details, alternative installation methods (Docker) and performance tips.
-
-### ROCm
-
-
-You will need to install vLLM from source. First install Pytorch on ROCm:
-
-```bash
-pip install torch==2.2.0.dev20231206+rocm5.7 --index-url https://download.pytorch.org/whl/nightly/rocm5.7 # tested version
-```
-
-You will then need to install flash attention for ROCm following [these instructions][rocm-flash-attention]. You can then install `xformers=0.0.23` and apply the patches needed to adapt Flash Attention for ROCm:
-
-```bash
-pip install xformers==0.0.23 --no-deps
-bash patch_xformers.rocm.sh
-```
-
-And finally build vLLM:
-
-```bash
-cd vllm
-pip install -U -r requirements-rocm.txt
-python setup.py install # This may take 5-10 minutes.
-```
-
-See the [vLLM documentation][vllm-install-rocm] for alternative installation methods (Docker).
-
-
-[vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
-[vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
-[rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
diff --git a/docs/reference/prompting.md b/docs/reference/prompting.md
deleted file mode 100644
index 986ca82fc..000000000
--- a/docs/reference/prompting.md
+++ /dev/null
@@ -1,395 +0,0 @@
-# Prompt templating
-
-Outlines provides a powerful domain-specific language to write and manage
-prompts, via what we call *prompt functions*.  Prompt functions are Python
-functions that contain a template for the prompt in their docstring, and their
-arguments correspond to the variables used in the prompt. When called, a prompt
-function returns the template rendered with the values of the arguments.
-
-The aim of prompt functions is to solve several recurrent problems with prompting:
-
-1. **Building complex prompts quickly leads to messy code.** This problem has
-   already been solved in the web development community by using templating, so
-   why not use it here?
-2. **Composing prompts is difficult.** Why not just compose functions?
-3. **Separating prompts from code.** Encapsulation in functions allows a clean
-   separation between prompts and code. Moreover, like any function, prompt
-   functions can be imported from other modules.
-
-Outlines uses the [Jinja templating
-engine](https://jinja.palletsprojects.com/en/3.1.x/) to render prompts, which
-allows to easily compose complex prompts.
-
-!!! warning "Prompt rendering"
-
-    Prompt functions are opinionated when it comes to prompt rendering. These opinions are meant to avoid common prompting errors, but can have unintended consequences if you are doing something unusual. We advise to always print the prompt before using it. You can also [read the
-    reference](#formatting-conventions) section if you want to know more.
-
-## Your first prompt
-
-The following snippet showcases a very simple prompt. The variables between
-curly brackets `{{  }}` are placeholders for the values of the arguments you
-will pass to the prompt function.
-
-=== "Code"
-
-    ```python title="greetings.py"
-    from outlines import Template
-
-    prompt = """Hello, {{ name }}!
-    {{ question }}"""
-
-    greetings = Template.from_string(prompt)
-    prompt = greetings("user", "How are you?")
-    print(prompt)
-    ```
-
-=== "Output"
-
-    ```text
-    Hello, user!
-    How are you?
-    ```
-
-If a variable is missing in the function's arguments, Jinja2 will throw an `UndefinedError` exception:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    prompt = """Hello, {{ surname }}!"""
-    greetings = Template.from_string(prompt)
-    prompt = greetings("user")
-    ```
-
-=== "Output"
-
-    ```text
-    Traceback (most recent call last):
-      File "<stdin>", line 9, in <module>
-      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 38, in __call__
-          return render(self.template, **bound_arguments.arguments)
-      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 213, in render
-          return jinja_template.render(**values)
-      File "/home/remi/micromamba/envs/outlines/lib/python3.9/site-packages/jinja2/environment.py", line 1301, in render
-          self.environment.handle_exception()
-      File "/home/remi/micromamba/envs/outlines/lib/python3.9/site-packages/jinja2/environment.py", line 936, in handle_exception
-          raise rewrite_traceback_stack(source=source)
-      File "<template>", line 1, in top-level template code
-      jinja2.exceptions.UndefinedError: 'surname' is undefined
-    ```
-
-## Importing prompts from files
-
-Outlines allows you to read a prompt template from a text file. This way you can build "white space perfect" prompts, and version them independently from your code. We have found ourselves gravitating around this pattern a lot since Outlines came out:
-
-=== "prompt.txt"
-    ```text
-    """Hello, {{ name }}!
-    {{ question }}
-    """
-    ```
-
-=== "generate.py"
-
-    ```python
-    from outlines import Template
-
-    greetings = Template.from_file("prompt.txt")
-    prompt = greetings("John Doe", "How are you today?")
-    ```
-
-=== "Output"
-
-    ```text
-    Hello, John Doe!
-    How are you today?
-    ```
-
-## Few-shot prompting
-
-Few-shot prompting can lead to messy code. Prompt functions allow you to loop
-over lists or dictionaries from the template. In the following example we
-demonstrate how we can generate a prompt by passing a list of dictionaries with
-keys `question` and `answer` to the prompt function:
-
-=== "Code"
-
-    ```text title="prompt.txt"
-    {{ instructions }}
-
-    Examples
-    --------
-
-    {% for example in examples %}
-    Q: {{ example.question }}
-    A: {{ example.answer }}
-
-    {% endfor %}
-    Question
-    --------
-
-    Q: {{ question }}
-    A:
-    ```
-
-    ```python title="render.py"
-    from outlines import Template
-
-    instructions = "Please answer the following question following the examples"
-    examples = [
-        {"question": "2+2=?", "answer":4},
-        {"question": "3+3=?", "answer":6}
-    ]
-    question = "4+4 = ?"
-
-    few_shots = Template.from_file("prompt.txt")
-    prompt = few_shots(instructions, examples, question)
-    print(prompt)
-    ```
-
-=== "Output"
-
-    ```text
-    Please answer the following question following the examples
-
-    Examples
-    --------
-
-    Q: 2+2=?
-    A: 4
-
-    Q: 3+3=?
-    A: 6
-
-    Question
-    --------
-
-    Q: 4+4 = ?
-    A:
-    ```
-
-## Conditionals, filters, etc.
-
-Jinja2 has many features beyond looping that are not described here:
-conditionals, filtering, formatting, etc. Please refer to the [Jinja
-documentation](https://jinja.palletsprojects.com/en/3.1.x/>) for more
-information about the syntax of the templating language. The Jinja syntax is
-powerful, and we recommend you take some time to read their documentation if you
-are building complex prompts.
-
-
-## Tools
-
-Several projects (e.g.[Toolformer](https://arxiv.org/abs/2302.04761), [ViperGPT](https://viper.cs.columbia.edu/), [AutoGPT](https://github.com/Significant-Gravitas/Auto-GPT), etc.) have shown that we can "teach" language models to use external functions by describing what these functions do in the prompt. In these projects the same information is often repeated twice: the function implementation, name, docstring, or arguments are copy-pasted in the prompt. This is cumbersome and error prone; you can directly pull this information from within an Outlines prompt function:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    def my_tool(arg1: str, arg2: int):
-        """Tool description.
-
-        The rest of the docstring
-        """
-        pass
-
-    prompt = """{{ question }}
-
-    COMMANDS
-    1. {{ tool | name }}: {{ tool | description }}, args: {{ tool | args }}
-
-    {{ tool | source }}
-    """
-
-    tool_prompt = Template.from_string(prompt)
-    prompt = tool_prompt("Can you do something?", my_tool)
-    print(prompt)
-    ```
-
-=== "Output"
-
-    ```text
-    Can you do something?
-
-    COMMANDS
-    1. my_tool: Tool description., args: arg1: str, arg2: int
-
-    def my_tool(arg1: str, arg2: int):
-        """Tool description.
-
-        The rest of the docstring
-        """
-        pass
-    ```
-
-## JSON response format
-
-To build reliable chains with language models we often need to instruct them the
-format in which we would like them to return their response.
-
-Without prompt templating, the information is repeated twice between creating
-the parsing function (e.g. a Pydantic model), and writing the desired schema in
-the prompt. This can lead to errors that are hard to debug.
-
-Outlines allows you to directly pull the JSON schema of a pydantic model, or
-pretty print a dictionary from within an Outlines prompt function
-
-=== "Code"
-
-    ```python
-    from pydantic import BaseModel, Field
-
-   from outlines import Template
-
-    class MyResponse(BaseModel):
-        field1: int = Field(description="an int")
-        field2: str
-
-
-    my_prompt = Template.from_string("""{{ response_model | schema }}""")
-    prompt = my_prompt(MyResponse)
-    print(prompt)
-    # {
-    #   "field1": "an int",
-    #   "field2": "<field2>"
-    # }
-    ```
-
-=== "Output"
-
-    ```python
-    response = {
-        "field1": "<field1>",
-        "field2": "a string"
-    }
-
-    my_prompt(MyResponse)
-    # {
-    #   "field1": "<field1>",
-    #   "field2": "a string"
-    # }
-    ```
-
-## Formatting conventions
-
-Prompt templates are opinionated when it comes to rendering a template read from
-a string, and these opinions are meant to avoid prompting mistakes and help with
-formatting.
-
-### Whitespaces
-
-If you have experience working with strings between triple quotes you know that
-indenting has an influence on the string's formatting. Prompt functions adopt
-a few conventions so you don't have to think about indents when writing prompt.
-
-First, whether you start the prompt right after the triple quotes or on the line
-below does not matter for formatting:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-
-    prompt1 = Template.from_string(
-    """My prompt
-    """
-    )
-
-    prompt2 = Template.from_string(
-    """
-    My prompt
-    """
-    )
-
-    print(prompt1())
-    print(prompt2())
-    ```
-
-=== "Output"
-
-    ```text
-    My prompt
-    My prompt
-    ```
-
-Indentation is relative to the second line of the docstring, and leading spaces are removed:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    example1 = Template.from_string(
-    """First line
-    Second line
-    """
-    )
-
-    example2 = Template.from_string(
-    """
-        Second line
-        Third line
-    """
-    )
-
-    example3 = Template.from_string(
-    """
-        Second line
-          Third line
-    """
-    )
-
-    print(example1())
-    print(example2())
-    print(example3())
-    ```
-
-=== "Output"
-
-    ```text
-    First line
-    Second line
-
-    Second line
-    Third line
-
-    Second line
-      Third line
-    ```
-
-Trailing whitespaces are not removed, unless they follow a linebreak symbol `\` (see [linebreaks](#linebreaks)).
-
-### Linebreaks
-
-You can use the backslash `\` to break a long line of text. It will render as a single line:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    example = Template.from_string(
-    """
-    Break in \
-    several lines \
-    But respect the indentation
-        on line breaks.
-    And after everything \
-    Goes back to normal
-    """
-    )
-
-    print(example())
-    ```
-
-=== "Output"
-
-    ```text
-    Break in several lines But respect the indentation
-        on line breaks.
-    And after everything Goes back to normal
-    ```
diff --git a/docs/reference/samplers.md b/docs/reference/samplers.md
deleted file mode 100644
index 9cba5cefa..000000000
--- a/docs/reference/samplers.md
+++ /dev/null
@@ -1,178 +0,0 @@
-# Samplers
-
-Outlines offers different sequence sampling algorithms, and we will integrate more in the future. You can read [this blog post](https://huggingface.co/blog/how-to-generate) for an overview of the different sampling algorithm.
-
-Samplers provide control over the sampling process, allowing you to influence the output of the model. This can include controlling randomness (temperature), biasing towards certain tokens (top-k, top-p), or sequence generation (beam search).
-
-## Multinomial sampling
-
-[Multinomial sampling](https://en.wikipedia.org/wiki/Multinomial_distribution) is the default sampling algorithm in Outlines.
-
-As an example, suppose we have only two possible tokens: "H" and "T". For a fixed prompt such as "Flip a coin, did you get heads or tails?" The language model calculates probability for each token:
-
-| Token | Probability |
-|-------|-------------|
-| "H"   | 0.5         |
-| "T"   | 0.5         |
-
-You'd expect to receive "H" 50% of the time and "T" 50% of the time.
-
-### Parameters
-
-- `samples`: Number of samples to generate (default: 1)
-- `top_k`: Only consider the top k tokens (optional)
-- `top_p`: Only consider the top tokens with cumulative probability >= p (optional)
-- `temperature`: Controls randomness of sampling (optional)
-
-### Default behavior
-
-Outlines defaults to the multinomial sampler without top-p or top-k sampling, and temperature equal to 1.
-
-Not specifying a sampler is equivalent to:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial()
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# 4
-```
-
-### Batching
-
-You can ask the generator to take multiple samples by passing the number of samples when initializing the sampler:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial(3)
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# [4, 4, 4]
-```
-
-If you ask multiple samples for a batch of prompts the returned array will be of shape `(num_samples, num_batches)`:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial(3)
-
-generator = generate.text(model, sampler)
-answer = generator(["What is 2+2?", "What is 3+3?"])
-
-print(answer)
-# [[4, 4, 4], [6, 6, 6]]
-```
-
-### Temperature
-
-You can control the temperature with
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial(3, temperature=0.5)
-
-generator = generate.text(model, sampler)
-answer = generator(["What is 2+2?", "What is 3+3?"])
-
-print(answer)
-```
-
-If you would like to use `temperature=0.0`, please use `sampler=samplers.greedy()` instead.
-
-### Top-k sampling
-
-You can ask Outlines to only consider the top-k logits at each step by specifying the value of the `top-k` keyword argument when initializing the sampler.
-
-```python
-sampler = samplers.multinomial(3, top_k=10)
-```
-
-### Top-p sampling
-
-You can ask Outlines to only consider the highest probability tokens such that their cumulative probability is greater than a threshold `p`. Specify the `top_p` keyword argument when initializing the sampler:
-
-
-```python
-sampler = samplers.multinomial(3, top_p=0.95)
-```
-
-## Greedy sampler
-
-Greedy sampling selects the token with the highest probability at each step. It's deterministic and always produces the same output for a given input.
-
-To use the greedy sampler, initialize the generator with the sampler:
-
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.greedy()
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# 4
-```
-
-You cannot ask for multiple samples with the greedy sampler since it does not clear what the result should be. Only the most likely token can be returned.
-
-
-## Beam Search
-
-Beam search maintains multiple candidate sequences at each step, potentially finding better overall sequences than greedy or multinomial sampling.
-
-To use Beam Search, initialize the generator with the sampler:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.beam_search(beams=5)
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# 4
-```
-!!! Warning "Compatibility"
-
-    Only models from the `transformers`  and `exllamav2 ` libraries are compatible with Beam Search.
-
-### Parameters
-
-- `beams`: Number of beams to use (default: 1)
-
-## Sampler Comparison
-
-Here's a table comparing the different samplers:
-
-| Sampler | Pros | Cons | Use Cases |
-|---------|------|------|-----------|
-| Greedy | Deterministic, fast | May produce repetitive text | When you need consistent, predictable output |
-| Multinomial | Balances exploration and exploitation | Results may vary between runs | General-purpose text generation, creative tasks |
-| Beam Search | Can find globally better sequences | More computationally expensive | When sequence quality is critical, e.g., translation |
-
-For most use cases, we recommend using the default [multinomial sampler](#multinomial-sampling).
diff --git a/docs/reference/serve/lmstudio.md b/docs/reference/serve/lmstudio.md
deleted file mode 100644
index db383186a..000000000
--- a/docs/reference/serve/lmstudio.md
+++ /dev/null
@@ -1,89 +0,0 @@
-# Serve with LM Studio
-
-!!! tip "Would rather not self-host?"
-
-    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
-
-[LM Studio](https://lmstudio.ai/) is an application that runs local LLMs. It flexibly mixes GPU and CPU compute in hardware-constrained environments.
-
-As of [LM Studio 0.3.4](https://lmstudio.ai/blog/lmstudio-v0.3.4), it natively supports Outlines for structured text generation, using an OpenAI-compatible endpoint.
-
-## Setup
-
-1. Install LM Studio by visiting their [downloads page](https://lmstudio.ai/download).
-2. Enable the LM Studio [server functionality](https://lmstudio.ai/docs/basics/server).
-3. Download [a model](https://lmstudio.ai/docs/basics#1-download-an-llm-to-your-computer).
-4. Install Python dependencies.
-```bash
-pip install pydantic openai
-```
-
-## Calling the server
-
-By default, LM Studio will serve from `http://localhost:1234`. If you are serving on a different port or host, make sure to change the `base_url` argument in `OpenAI` to the relevant location.
-
-```python
-class Testing(BaseModel):
-    """
-    A class representing a testing schema.
-    """
-    name: str
-    age: int
-
-openai_client = openai.OpenAI(
-    base_url="http://0.0.0.0:1234/v1",
-    api_key="dopeness"
-)
-
-# Make a request to the local LM Studio server
-response = openai_client.beta.chat.completions.parse(
-    model="hugging-quants/Llama-3.2-1B-Instruct-Q8_0-GGUF",
-    messages=[
-        {"role": "system", "content": "You are like so good at whatever you do."},
-        {"role": "user", "content": "My name is Cameron and I am 28 years old. What's my name and age?"}
-    ],
-    response_format=Testing
-)
-```
-
-You should receive a `ParsedChatCompletion[Testing]` object back:
-
-```python
-ParsedChatCompletion[Testing](
-    id='chatcmpl-3hykyf0fxus7jc90k6gwlw',
-    choices=[
-        ParsedChoice[Testing](
-            finish_reason='stop',
-            index=0,
-            logprobs=None,
-            message=ParsedChatCompletionMessage[Testing](
-                content='{ "age": 28, "name": "Cameron" }',
-                refusal=None,
-                role='assistant',
-                function_call=None,
-                tool_calls=[],
-                parsed=Testing(name='Cameron', age=28)
-            )
-        )
-    ],
-    created=1728595622,
-    model='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-Q4_K_M.gguf',
-    object='chat.completion',
-    service_tier=None,
-    system_fingerprint='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-
-Q4_K_M.gguf',
-    usage=CompletionUsage(
-        completion_tokens=17,
-        prompt_tokens=47,
-        total_tokens=64,
-        completion_tokens_details=None,
-        prompt_tokens_details=None
-    )
-)
-```
-
-You can retrieve your `Testing` object with
-
-```python
-response.choices[0].message.parsed
-```
diff --git a/docs/reference/serve/vllm.md b/docs/reference/serve/vllm.md
deleted file mode 100644
index 14277a526..000000000
--- a/docs/reference/serve/vllm.md
+++ /dev/null
@@ -1,73 +0,0 @@
-# Serve with vLLM
-
-!!! tip "Would rather not self-host?"
-
-    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
-
-Outlines can be deployed as an LLM service using the vLLM inference engine and a FastAPI server. vLLM is not installed by default so will need to install Outlines with:
-
-```bash
-pip install outlines[serve]
-```
-
-You can then start the server with:
-
-```bash
-python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used. The `--model` argument allows you to specify any model of your choosing.
-
-To run inference on multiple GPUs you must pass the `--tensor-parallel-size` argument when initializing the server. For instance, to run inference on 2 GPUs:
-
-
-```bash
-python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct" --tensor-parallel-size 2
-```
-
-
-### Alternative Method: Via Docker
-
-You can install and run the server with Outlines' official Docker image using the command
-
-```bash
-docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-## Querying Endpoint
-
-You can then query the model in shell by passing a prompt and either
-
-1. a [JSON Schema][jsonschema]{:target="_blank"} specification or
-2. a [Regex][regex]{:target="_blank"} pattern
-
-with the `schema` or `regex` parameters, respectively, to the `/generate` endpoint. If both are specified, the schema will be used. If neither is specified, the generated text will be unconstrained.
-
-For example, to generate a string that matches the schema `{"type": "string"}` (any string):
-
-```bash
-curl http://127.0.0.1:8000/generate \
-    -d '{
-        "prompt": "What is the capital of France?",
-        "schema": {"type": "string", "maxLength": 5}
-        }'
-```
-
-To generate a string that matches the regex `(-)?(0|[1-9][0-9]*)(\.[0-9]+)?([eE][+-][0-9]+)?` (a number):
-
-```bash
-curl http://127.0.0.1:8000/generate \
-    -d '{
-        "prompt": "What is Pi? Give me the first 15 digits: ",
-        "regex": "(-)?(0|[1-9][0-9]*)(\\.[0-9]+)?([eE][+-][0-9]+)?"
-        }'
-```
-
-Instead of `curl`, you can also use the [requests][requests]{:target="_blank"} library from another python program.
-
-Please consult the [vLLM documentation][vllm]{:target="_blank"} for details on additional request parameters. You can also [read the code](https://github.com/dottxt-ai/outlines/blob/main/outlines/serve/serve.py) in case you need to customize the solution to your needs.
-
-[requests]: https://requests.readthedocs.io/en/latest/
-[vllm]: https://docs.vllm.ai/en/latest/index.html
-[jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
-[regex]: https://www.regular-expressions.info/tutorial.html
diff --git a/docs/reference/text.md b/docs/reference/text.md
deleted file mode 100644
index f36a67da0..000000000
--- a/docs/reference/text.md
+++ /dev/null
@@ -1,86 +0,0 @@
-# Text generation
-
-Outlines provides a unified interface to generate text with many language models, API-based and local. The same pattern is used throughout the library:
-
-1. Instantiate a generator by calling `outlines.generate.text` with the model to be used.
-2. Call the generator with the prompt and (optionally) some generation parameters.
-
-
-```python
-from outlines import models, generate
-
-model = models.openai("gpt-4o-mini")
-generator = generate.text(model)
-answer = generator("What is 2+2?")
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-answer = generator("What is 2+2?")
-```
-
-By default Outlines uses the multinomial sampler with `temperature=1`. See [this section](samplers.md) to learn how to use different samplers.
-
-## Streaming
-
-Outlines allows you to stream the model's response by calling the `.stream` method of the generator with the prompt:
-
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
-tokens = generator.stream("What is 2+2?")
-for token in tokens:
-    print(token)
-```
-
-## Parameters
-
-### Limit the number of tokens generated
-
-To limit the number of tokens generated you can pass the `max_tokens` positional argument to the generator:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
-answer = generator("What is 2+2?", 5)
-answer = generator("What is 2+2?", max_tokens=5)
-```
-
-### Stop after a given string is generated
-
-You can also ask the model to stop generating text after a given string has been generated, for instance a period or a line break. You can pass a string or a line of string for the `stop_at` argument:
-
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
-answer = generator("What is 2+2?", stop_at=".")
-answer = generator("What is 2+2?", stop_at=[".", "\n"])
-```
-
-*The stopping string will be included in the response.*
-
-
-### Seed the generation
-
-It can be useful to seed the generation in order to get reproducible results:
-
-```python
-import torch
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
-seed = 789001
-
-answer = generator("What is 2+2?", seed=seed)
-```
diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index e71a6124f..381f5aa11 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -1,149 +1,22 @@
 @font-face {
-  font-family: 'Source Code Pro Custom', monospace;
-  src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
+    font-family: "Source Code Pro Custom", monospace;
+    src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
 }
 
 :root > * {
-  --md-default-bg-color: #FFFFFF;
-  --md-code-bg-color: #2E3440;
-  --md-code-fg-color: #FFFFFF;
-  --md-text-font-family: "Inter";
-  --md-code-font: "Source Code Pro Custom";
-  --md-typeset-a-color: #d16626; /*this is the brand color*/
-
-  /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
-  --md-mermaid-label-fg-color: #000000;
-  --md-mermaid-edge-color: #000000;
-}
-
-.index-pre-code {
-  max-width: 700px;
-  left: 50%;
-}
-
-.index-pre-code pre>code {
-  text-align: left;
-}
-
-.md-clipboard::after {
-  color: #FFFFFF;
-  transition: color 0.3s ease-in-out;
-}
-
-.md-clipboard:hover::after {
-  color: #D8DEE9;
-}
-
-.md-source-file {
-  text-align: center;
-  padding: 24px 0;
-}
-
-.md-typeset pre>code {
-  border-radius: .2rem;
-  box-shadow: 10px 5px 5px #D8DEE9;
-}
-
-.md-typeset p > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset strong > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-content p > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset td > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset li > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset code {
-  font-weight: 500;
-}
-
-.md-typeset pre {
-  margin-left: .5rem;
-  margin-right: .5rem;
-  margin-top: 2rem;
-  margin-bottom: 2rem;
-}
-
-.language-python {
-  background: #FFFFFF ! important
-}
-
-.language-bash {
-  background: #FFFFFF ! important
-}
-
-.language-toml {
-  background: #FFFFFF ! important
-}
-
-.language-text {
-  background: #FFFFFF ! important
-}
-
-.language-json {
-  background: #FFFFFF ! important
-}
-
-h1.title {
-  color: #FFFFFF;
-  margin: 0px 0px 5px;
-}
-
-h2.subtitle {
-  margin: 5px 0px 25px;
-  font-size: 1rem;
-  max-width: 540px;
-  margin: 0 auto;
-}
-
-.md-typeset {
-  line-height: 24px;
-  font-weight: 400;
-}
-
-.md-typeset h1 {
-  font-weight: bold;
-  color: #000000;
-}
-
-.md-typeset h2 {
-  font-weight: bold;
-  color: #000000;
-}
-
-span.md-ellipsis {
-  color: black;
-}
-
-.md-nav__link--active {
-  background-color: #ECEFF4;
-}
-
-.md-typeset ol li {
-  margin-bottom: .1rem;
-}
-
-.md-typeset ul li {
-  margin-bottom: .1rem;
+    /* Accent color shades */
+    /* --md-accent-fg-color: #d36237;
+    --md-accent-fg-color--transparent: #d36237;
+    --md-accent-bg-color: hsla(0, 0%, 100%, 1);
+    --md-accent-bg-color--light: hsla(0, 0%, 100%, 0.7); */
+
+    /* Primary color shades */
+    /* --md-accent-fg-color: hsla(#d36237, 1);
+    --md-accent-fg-color--transparent: hsla(#d36237, 0.1);
+    --md-accent-bg-color: #d36237
+    --md-accent-bg-color--light: hsla(#d36237, 0.7); */
+
+    --md-text-font-family: "Inter";
+    --md-code-font: "Source Code Pro Custom";
+    /* --md-typeset-a-color: #d36237; /*this is the brand color */
 }
diff --git a/docs/welcome.md b/docs/welcome.md
deleted file mode 100644
index 728a30a97..000000000
--- a/docs/welcome.md
+++ /dev/null
@@ -1,183 +0,0 @@
----
-title: Welcome to Outlines!
----
-
-Outlines is a Python library that allows you to use Large Language Model in a simple and robust way (with structured generation). It is built by [.txt][.txt]{:target="_blank"}, and is already used in production by many companies.
-
-## What models do you support?
-
-We support [Openai](reference/models/openai.md), but the true power of Outlines is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [exllama2](reference/models/exllamav2.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
-
-## What are the main features?
-
-<div class="grid cards" markdown>
--   :material-code-json:{ .lg .middle } __Make LLMs generate valid JSON__
-
-    ---
-
-    No more invalid JSON outputs, 100% guaranteed
-
-    [:octicons-arrow-right-24: Generate JSON](reference/generation/json.md)
-
--   :material-keyboard-outline:{ .lg .middle } __JSON mode for vLLM__
-
-    ---
-
-    Deploy a LLM service using Outlines' JSON structured generation and vLLM
-
-    [:octicons-arrow-right-24: Deploy outlines](reference/serve/vllm.md)
-
-
--   :material-regex:{ .lg .middle } __Make LLMs follow a Regex__
-
-    ---
-
-    Generate text that parses correctly 100% of the time
-
-    [:octicons-arrow-right-24: Guide LLMs](reference/generation/regex.md)
-
--    :material-chat-processing-outline:{ .lg .middle } __Powerful Prompt Templating__
-
-     ---
-
-     Better manage your prompts' complexity with prompt templating
-
-    [:octicons-arrow-right-24: Learn more](reference/prompting.md)
-</div>
-
-## Why use Outlines?
-
-
-Outlines is built at [.txt][.txt] by engineers with decades of experience in software engineering, machine learning (Bayesian Statistics and NLP), and compilers. [.txt][.txt] is a VC-backed company fully focused on the topic of structured generation and is committed to make the community benefit from its experience.
-
-We are also open source veterans and have authored/maintained many libraries over the years: the [Aesara][aesara]{:target="_blank"} and [Pythological][pythological]{:target="_blank"} ecosystems, [Blackjax][blackjax]{:target="_blank"} and [Hy][hy]{:target="_blank"} among many others.
-.
-
-Outlines does not use unnecessary abstractions that tend to get in your way. We have a laser focus on reliable text generation with LLMs, a clear roadmap to push the state of the art in this area and a commitment to clean and robust code.
-
-And last but not least, unlike alternatives, Outlines' structured generation introduces **no overhead** during inference.
-
-
-## Who is using Outlines?
-
-Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) are using Outlines. Some of the prominent companies and organizations that are using Outlines include:
-
-<head>
-  <style>
-  .row {
-      display: inline-block;
-      width: 100%;
-      margin-bottom: 50px;
-      margin-top: 0px !important;
-      break-inside: avoid;
-  }
-
-  /* Create two equal columns that sits next to each other */
-  .column {
-      column-count: 3;
-      column-gap: 20px;
-      padding: 20px;
-  }
-
-  </style>
-</head>
-<body>
-
-<div class="column">
-  <div class="row"><img src="../logos/amazon.png" width="200"></div>
-  <div class="row"><img src="../logos/apple.png" width="200"></div>
-  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
-  <div class="row"><img src="../logos/canoe.png" width="200"></div>
-  <div class="row"><img src="../logos/cisco.png" width="200"></div>
-  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
-  <div class="row"><img src="../logos/databricks.png" width="200"></div>
-  <div class="row"><img src="../logos/datadog.png" width="200"></div>
-  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
-  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
-  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
-  <div class="row"><img src="../logos/harvard.png" width="200"></div>
-  <div class="row"><img src="../logos/hf.png" width="200"></div>
-  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
-  <div class="row"><img src="../logos/meta.png" width="200"></div>
-  <div class="row"><img src="../logos/mit.png" width="200"></div>
-  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
-  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
-  <div class="row"><img src="../logos/nyu.png" width="200"></div>
-  <div class="row"><img src="../logos/safran.png" width="200"></div>
-  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
-  <div class="row"><img src="../logos/shopify.png" width="200"></div>
-  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
-  <div class="row"><img src="../logos/tinder.png" width="200"></div>
-  <div class="row"><img src="../logos/upenn.png" width="200"></div>
-</div>
-
-</body>
-
-Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
-
-Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
-
-## Philosophy
-
-**Outlines**  is a library for neural text generation. You can think of it as a
-more flexible replacement for the `generate` method in the
-[transformers](https://github.com/huggingface/transformers) library.
-
-**Outlines**  helps developers *structure text generation* to build robust
-interfaces with external systems. It provides generation methods that
-guarantee that the output will match a regular expressions, or follow
-a JSON schema.
-
-**Outlines**  provides *robust prompting primitives* that separate the prompting
-from the execution logic and lead to simple implementations of few-shot
-generations, ReAct, meta-prompting, agents, etc.
-
-**Outlines**  is designed as a *library* that is meant to be compatible the
-broader ecosystem, not to replace it. We use as few abstractions as possible,
-and generation can be interleaved with control flow, conditionals, custom Python
-functions and calls to other libraries.
-
-**Outlines**  is *compatible with every auto-regressive model*. It only interfaces with models
-via the next-token logits distribution.
-
-## Outlines people
-
-Outlines would not be what it is today without a community of dedicated developers:
-
-<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
-  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
-</a>
-
-## Acknowledgements
-
-<div class="grid" markdown>
-
-
-<figure markdown>
-  <a href="http://www.dottxt.co">
-  ![Normal Computing logo](assets/images/dottxt.png){ width="150" }
-  </a>
-</figure>
-
-<figure markdown>
-  <a href="https://www.normalcomputing.ai">
-  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
-  </a>
-</figure>
-
-</div>
-
-Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
-
-
-[discord]: https://discord.gg/R9DSu34mGd
-[aesara]: https://github.com/aesara-devs
-[blackjax]: https://github.com/blackjax-devs/blackjax
-[pythological]: https://github.com/pythological
-[hy]: https://hylang.org/
-[.txt]: https://dottxt.co
-[vllm]: https://github.com/vllm-project/vllm
-[tgi]: https://github.com/huggingface/text-generation-inference
-[lorax]: https://github.com/predibase/lorax
-[xinference]: https://github.com/xorbitsai/inference
-[sglang]: https://github.com/sgl-project/sglang/
diff --git a/examples/babyagi.py b/examples/babyagi.py
index a8f1de8ba..4a817906d 100644
--- a/examples/babyagi.py
+++ b/examples/babyagi.py
@@ -8,13 +8,14 @@
 from collections import deque
 from typing import Deque, List
 
+from openai import OpenAI
+
 import outlines
-import outlines.models as models
 from outlines import Template
 
 
-model = models.openai("gpt-4o-mini")
-complete = outlines.generate.text(model)
+model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
+complete = outlines.Generator(model)
 
 ## Load the prompts
 perform_task_ppt = Template.from_file("prompts/babyagi_perform_task.txt")
@@ -76,11 +77,14 @@ def one_cycle(objective: str, task_list, next_task_id: int):
 
     task = task_list.popleft()
 
-    prompt = perform_task_ppt(objective, task)
+    prompt = perform_task_ppt(objective=objective, task=task)
     result = complete(prompt)
 
     prompt = create_tasks_ppt(
-        objective, first_task["task_name"], result, [first_task["task_name"]]
+        objective=objective,
+        task=first_task["task_name"],
+        result=result,
+        previous_tasks=[first_task["task_name"]],
     )
     new_tasks = complete(prompt)
 
@@ -91,7 +95,9 @@ def one_cycle(objective: str, task_list, next_task_id: int):
         task_list.append({"task_id": next_task_id, "task_name": task})
 
     prompt = prioritize_tasks_ppt(
-        objective, [task["task_name"] for task in task_list], next_task_id
+        objective=objective,
+        tasks=[task["task_name"] for task in task_list],
+        next_task_id=next_task_id,
     )
     prioritized_tasks = complete(prompt)
 
diff --git a/examples/beam-cloud/app.py b/examples/beam-cloud/app.py
index fb6c2cb2b..781bc1d3d 100644
--- a/examples/beam-cloud/app.py
+++ b/examples/beam-cloud/app.py
@@ -1,14 +1,21 @@
+from typing import Literal
+
 from beam import Image, endpoint, env
 
+
 if env.is_remote():
     import outlines
 
 
 # Pre-load models when the container first starts
 def load_models():
+    from transformers import AutoModelForCausalLM, AutoTokenizer
     import outlines
 
-    model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+    model = outlines.models.from_transformers(
+        AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+        AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    )
     return model
 
 
@@ -34,6 +41,6 @@ def predict(context, **inputs):
     # Unpack cached model from context
     model = context.on_start_value
     # Inference
-    generator = outlines.generate.choice(model, ["Positive", "Negative"])
+    generator = outlines.Generator(model, Literal["Positive", "Negative"])
     answer = generator(prompt)
     return {"answer": answer}
diff --git a/examples/bentoml/service.py b/examples/bentoml/service.py
index 98d46894a..370bac901 100644
--- a/examples/bentoml/service.py
+++ b/examples/bentoml/service.py
@@ -1,7 +1,7 @@
 import typing as t
 
 import bentoml
-from import_model import BENTO_MODEL_TAG
+from import_model import BENTO_MODEL_TAG, MODEL_ID
 
 DEFAULT_SCHEMA = """{
     "title": "Character",
@@ -55,13 +55,17 @@ class Outlines:
 
     def __init__(self) -> None:
         import torch
+        from transformers import AutoModelForCausalLM, AutoTokenizer
 
         import outlines
 
-        self.model = outlines.models.transformers(
-            self.bento_model_ref.path,
-            device="cuda",
-            model_kwargs={"torch_dtype": torch.float16},
+        self.model = outlines.from_transformers(
+            AutoTokenizer.from_pretrained(MODEL_ID),
+            AutoModelForCausalLM.from_pretrained(
+                MODEL_ID,
+                torch_dtype=torch.float16,
+                low_cpu_mem_usage=True,
+            )
         )
 
     @bentoml.api
@@ -72,7 +76,7 @@ async def generate(
     ) -> t.Dict[str, t.Any]:
         import outlines
 
-        generator = outlines.generate.json(self.model, json_schema)
+        generator = outlines.Generator(self.model, outlines.json_schema(json_schema))
         character = generator(prompt)
 
         return character
diff --git a/examples/cerebrium/main.py b/examples/cerebrium/main.py
index b61cdb6e5..4e1987197 100644
--- a/examples/cerebrium/main.py
+++ b/examples/cerebrium/main.py
@@ -1,6 +1,13 @@
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
+    AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
+)
+
 
 schema = {
     "title": "Character",
@@ -29,14 +36,13 @@
     },
 }
 
-generator = outlines.generate.json(model, schema)
-
 
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
-    character = generator(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
+    character = model(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
+        outlines.json_schema(schema),
     )
 
     print(character)
diff --git a/examples/cfg.py b/examples/cfg.py
index 99edf35f1..bd3b58a2e 100644
--- a/examples/cfg.py
+++ b/examples/cfg.py
@@ -1,5 +1,8 @@
-import outlines.generate as generate
-import outlines.models as models
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+import outlines
+from outlines.types import CFG
+
 
 nlamb_grammar = r"""
     start: sentence
@@ -75,15 +78,20 @@
     %ignore WS
 """
 
-model = models.transformers("hf-internal-testing/tiny-random-gpt2")
+model_name = "hf-internal-testing/tiny-random-gpt2"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name),
+)
+
 batch_size = 10
 for grammar in [nlamb_grammar, calc_grammar, dyck_grammar, json_grammar]:
-    generator = generate.cfg(model, grammar, max_tokens=model.model.config.n_positions)
-    sequences = generator([" "] * batch_size)
+    generator = outlines.Generator(model, CFG(grammar))
+    sequences = generator([" "] * batch_size, max_tokens=model.model.config.n_positions)
     for seq in sequences:
         try:
-            parse = generator.fsm.parser.parse(seq)
+            parse = generator.logits_processor.guide.parser.parse(seq)
             assert parse is not None
             print("SUCCESS", seq)
-        except Exception:  # will also fail if goes over max_tokens / context window
+        except Exception:  # will also fail if goes over max_new_tokens / context window
             print("FAILURE", seq)
diff --git a/examples/dating_profile.py b/examples/dating_profile.py
index d26984223..ff488698e 100644
--- a/examples/dating_profile.py
+++ b/examples/dating_profile.py
@@ -6,7 +6,7 @@
 from pydantic import BaseModel, conlist
 
 import outlines
-from outlines import models, Template
+from outlines import Template
 
 
 class QuestionChoice(str, Enum):
@@ -86,26 +86,17 @@ class Example:
 # Below requires ~13GB of GPU memory
 # https://huggingface.co/mosaicml/mpt-7b-8k-instruct
 # Motivation: Reasonably large model that fits on a single GPU and has been fine-tuned for a larger context window
-config = transformers.AutoConfig.from_pretrained(
-    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
-)
-config.init_device = "meta"
-model = models.transformers(
-    model_name="mosaicml/mpt-7b-8k-instruct",
-    device="cuda",
-    model_kwargs={
-        "config": config,
-        "trust_remote_code": True,
-        "torch_dtype": torch.bfloat16,
-        "device_map": {"": 0},
-    },
+model_name = "mosaicml/mpt-7b-8k-instruct"
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
 )
 
 new_description = "I'm a laid-back lawyer who spends a lot of his free-time gaming. I work in a corporate office, but ended up here after the start-up I cofounded got acquired, so still play ping pong with my cool coworkers every day. I have a bar at home where I make cocktails, which is great for entertaining friends. I secretly like to wear suits and get a new one tailored every few months. I also like weddings because I get to wear those suits, and it's a good excuse for a date. I watch the latest series because I'm paying, with my hard-earned money, for every streaming service."
 
 dating_profile_prompt = Template.from_file("prompts/dating_profile.txt")
 prompt = dating_profile_prompt(description=new_description, examples=samples)
-profile = outlines.generate.json(model, DatingProfile)(prompt)  # type: ignore
+profile = model(prompt, outlines.json_schema(DatingProfile), max_tokens=500)  # type: ignore
 print(profile)
 
 # Sample generated profiles
diff --git a/examples/llamacpp_example.py b/examples/llamacpp_example.py
index 22d0da3ba..f6afc40ed 100644
--- a/examples/llamacpp_example.py
+++ b/examples/llamacpp_example.py
@@ -1,6 +1,7 @@
 from enum import Enum
 
 from pydantic import BaseModel, constr
+from llama_cpp import Llama
 
 import outlines
 
@@ -30,10 +31,10 @@ class Character(BaseModel):
 
 if __name__ == "__main__":
     # curl -L -o mistral-7b-instruct-v0.2.Q5_K_M.gguf https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q5_K_M.gguf
-    model = outlines.models.llamacpp("./mistral-7b-instruct-v0.2.Q5_K_M.gguf")
+    model = outlines.from_llamacpp(Llama("./mistral-7b-instruct-v0.2.Q5_K_M.gguf"))
 
     # Construct structured sequence generator
-    generator = outlines.generate.json(model, Character)
+    generator = outlines.Generator(model, Character)
 
     # Draw a sample
     seed = 789005
diff --git a/examples/llamacpp_processor.py b/examples/llamacpp_processor.py
index 909df38fc..a0bf4aa3f 100644
--- a/examples/llamacpp_processor.py
+++ b/examples/llamacpp_processor.py
@@ -3,7 +3,7 @@
 from llama_cpp import Llama, LogitsProcessorList
 from pydantic import BaseModel, constr
 
-from outlines.generate.processors import JSONLogitsProcessor
+from outlines.processors import JSONLogitsProcessor
 from outlines.models.llamacpp import LlamaCppTokenizer
 
 
@@ -36,7 +36,7 @@ class Character(BaseModel):
 
     prompt = "Instruct: You are a leading role play gamer. You have seen thousands of different characters and their attributes.\nPlease return a JSON object with common attributes of an RPG character. Give me a character description\nOutput:"
 
-    logits_processor = JSONLogitsProcessor(Character, tokenizer)
+    logits_processor = JSONLogitsProcessor(Character, tokenizer, tensor_library_name="numpy")
 
     json_str = llama.create_completion(
         prompt,
diff --git a/examples/math_generate_code.py b/examples/math_generate_code.py
index c65537ddb..a5a53f084 100644
--- a/examples/math_generate_code.py
+++ b/examples/math_generate_code.py
@@ -1,9 +1,11 @@
 """Example from https://dust.tt/spolu/a/d12ac33169"""
 
+import openai
+
 import outlines
-import outlines.models as models
 from outlines import Template
 
+
 examples = [
     {"question": "What is 37593 * 67?", "code": "37593 * 67"},
     {
@@ -35,8 +37,8 @@ def execute_code(code):
     return result
 
 
-prompt = answer_with_code_prompt(question, examples)
-model = models.openai("gpt-4o-mini")
-answer = outlines.generate.text(model)(prompt)
+prompt = answer_with_code_prompt(question=question, examples=examples)
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
+answer = model(prompt)
 result = execute_code(answer)
 print(f"It takes Carla {result:.0f} minutes to download the file.")
diff --git a/examples/meta_prompting.py b/examples/meta_prompting.py
index 18f1ad41a..ed4b7455a 100644
--- a/examples/meta_prompting.py
+++ b/examples/meta_prompting.py
@@ -12,11 +12,15 @@
 
 import argparse
 
+import openai
+
 import outlines
-import outlines.models as models
 from outlines import Template
 
 
+client = openai.OpenAI()
+
+
 def split_into_steps(question, model_name: str):
     solve = Template.from_string(
         """{{question}}
@@ -24,16 +28,15 @@ def split_into_steps(question, model_name: str):
         """
     )
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
-    prompt = solve(question)
-    answer = generator(prompt, 500)
+    prompt = solve(question=question)
+    answer = model(prompt, max_tokens=500)
     prompt += (
         answer
         + "\n what is the only option that displays the same type of relationship as : :?"
     )
-    answer = generator(prompt, 500)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -49,13 +52,12 @@ def fill_in_the_blanks(question, model_name: str):
 
     solve = Template.from_string("""{{memory}}. Let's begin.""")
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
-    prompt = determine_goal(question)
-    answer = generator(prompt, stop_at=["."])
-    prompt = solve(prompt + answer)
-    answer = generator(prompt, max_tokens=500)
+    prompt = determine_goal(question=question)
+    answer = model(prompt, stop=["."])
+    prompt = solve(memory=prompt + answer)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -89,13 +91,12 @@ def ask_an_expert(question, model_name: str):
         """
     )
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
-    prompt = find_expert(question)
-    expert = generator(prompt, stop_at=['"'])
-    prompt = get_answer(question, expert, prompt + expert)
-    answer = generator(prompt, max_tokens=500)
+    prompt = find_expert(question=question)
+    expert = model(prompt, stop=['"'])
+    prompt = get_answer(question=question, expert=expert, memory=prompt+expert)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -117,13 +118,12 @@ def ask_an_expert_simple(question, model_name: str):
         """
     )
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
-    prompt = find_expert(question)
-    expert = generator(prompt, stop_at=["\n", "."])
-    prompt = get_answer(expert, prompt + expert)
-    answer = generator(prompt, max_tokens=500)
+    prompt = find_expert(question=question)
+    expert = model(prompt, stop=["\n", "."])
+    prompt = get_answer(expert=expert, memory=prompt+expert)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
diff --git a/examples/modal_example.py b/examples/modal_example.py
index 34d4ee606..fe0e6bc67 100644
--- a/examples/modal_example.py
+++ b/examples/modal_example.py
@@ -4,7 +4,7 @@
 
 
 outlines_image = modal.Image.debian_slim(python_version="3.11").pip_install(
-    "outlines==0.0.37",
+    "outlines==1.0.0",
     "transformers==4.38.2",
     "datasets==2.18.0",
     "accelerate==0.27.2",
@@ -12,9 +12,11 @@
 
 
 def import_model():
-    import outlines
+    from transformers import AutoModelForCausalLM, AutoTokenizer
 
-    outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+    model_id = "mistralai/Mistral-7B-Instruct-v0.2"
+    _ = AutoTokenizer.from_pretrained(model_id)
+    _ = AutoModelForCausalLM.from_pretrained(model_id)
 
 
 outlines_image = outlines_image.run_function(import_model)
@@ -58,17 +60,22 @@ def import_model():
 }"""
 
 
-@app.function(image=outlines_image, gpu=modal.gpu.A100(memory=80))
+@app.function(image=outlines_image, gpu="A100-40GB")
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
     import outlines
+    from transformers import AutoModelForCausalLM, AutoTokenizer
 
-    model = outlines.models.transformers("mistralai/Mistral-7B-v0.1", device="cuda")
+    model_id = "mistralai/Mistral-7B-Instruct-v0.2"
+    model = outlines.from_transformers(
+        tokenizer=AutoTokenizer.from_pretrained(model_id),
+        model=AutoModelForCausalLM.from_pretrained(model_id, device="cuda"),
+    )
 
-    generator = outlines.generate.json(model, schema)
-    character = generator(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
+    character = model(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
+        outlines.json_schema(schema),
     )
 
     print(character)
diff --git a/examples/parsing.py b/examples/parsing.py
index 9e08eae14..c8aac5a2e 100644
--- a/examples/parsing.py
+++ b/examples/parsing.py
@@ -36,7 +36,6 @@
     start="file_input",
 )
 
-
 class ParserLogitsProcessor(LogitsProcessor):
     """Bias invalid token scores according to a running parse state."""
 
diff --git a/examples/pick_odd_one_out.py b/examples/pick_odd_one_out.py
index f1a992411..36ec9dfe7 100644
--- a/examples/pick_odd_one_out.py
+++ b/examples/pick_odd_one_out.py
@@ -10,34 +10,37 @@
 
 """
 
-import outlines
-import outlines.models as models
-
+import json
 
-build_ooo_prompt = outlines.Template.from_string(
-    """
-    Pick the odd word out: skirt, dress, pen, jacket.
-    skirt is clothing, dress is clothing, pen is an object, jacket is clothing.
-    So the odd one is pen.
+import openai
 
-    Pick the odd word out: Spain, France, German, England, Singapore.
-    Spain is a country, France is a country, German is a language, ...
-    So the odd one is German.
+import outlines
+from outlines import Generator
+from outlines.types import JsonSchema
 
-    Pick the odd word out: {{ options | join(", ") }}.
 
-    """
-)
+build_ooo_prompt = outlines.Template.from_file("prompts/pick_odd_one_out.txt")
 
 options = ["sea", "mountains", "plains", "sock"]
-
-model = models.openai("gpt-4o-mini")
-gen_text = outlines.generate.text(model)
-gen_choice = outlines.generate.choice(model, options)
-
-prompt = build_ooo_prompt(options)
-reasoning = gen_text(prompt, stop_at=["Pick the odd word", "So the odd one"])
+options_schema = JsonSchema({
+    "type": "object",
+    "properties": {
+        "result": {
+            "type": "string",
+            "enum": options
+        }
+    },
+    "required": ["result"]
+})
+
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
+gen_text = Generator(model)
+gen_choice = Generator(model, options_schema)
+
+prompt = build_ooo_prompt(options=options)
+reasoning = gen_text(prompt, stop=["Pick the odd word", "So the odd one"])
 prompt += reasoning
-result = gen_choice(prompt)
+raw_result = gen_choice(prompt)
+result = json.loads(raw_result)["result"]
 prompt += result
 print(result)
diff --git a/examples/prompts/babyagi_create_task.txt b/examples/prompts/babyagi_create_task.txt
new file mode 100644
index 000000000..434063aad
--- /dev/null
+++ b/examples/prompts/babyagi_create_task.txt
@@ -0,0 +1,9 @@
+Objective: {{ objective }}
+Current Task: {{ task }}
+Result: {{ result }}
+Previous Tasks: {{ previous_tasks }}
+
+Based on the result, create a list of new tasks that will help achieve the objective.
+Please provide the tasks in the following format:
+1. [Task description]
+2. [Task description]
diff --git a/examples/prompts/babyagi_perform_task.txt b/examples/prompts/babyagi_perform_task.txt
new file mode 100644
index 000000000..bd8098924
--- /dev/null
+++ b/examples/prompts/babyagi_perform_task.txt
@@ -0,0 +1,5 @@
+Objective: {{ objective }}
+Task: {{ task }}
+
+Please perform the task and provide a concise result in the following format:
+Result: [Your concise result here]
diff --git a/examples/prompts/babyagi_prioritize_task.txt b/examples/prompts/babyagi_prioritize_task.txt
new file mode 100644
index 000000000..f42c8b05f
--- /dev/null
+++ b/examples/prompts/babyagi_prioritize_task.txt
@@ -0,0 +1,7 @@
+Tasks: {{ tasks }}
+Next Task ID: {{ next_task_id }}
+
+Please prioritize the tasks based on their importance and urgency to achieve the objective.
+Provide the prioritized tasks in the following format:
+1. [Task ID]. [Task description]
+2. [Task ID]. [Task description]
diff --git a/examples/prompts/dating_profile.txt b/examples/prompts/dating_profile.txt
new file mode 100644
index 000000000..129be4582
--- /dev/null
+++ b/examples/prompts/dating_profile.txt
@@ -0,0 +1,11 @@
+You are a world-renowned matchmaker who understands the modern dating market. Your job is to generate dating app profiles for male clients interested in women based on a provided description. The profiles should be authentic, show off their strengths, and maximize their likelihood of getting matches on dating apps.
+Here are some examples of past clients that you have successfully created profiles for:
+{% for example in examples %}
+Description:
+{{ example.description }}
+Profile:
+{{ example.profile }}
+{% endfor %}
+Here is the new client who you need to create a profile for:
+Description: {{ description }}
+Profile:
diff --git a/examples/prompts/pick_odd_one_out.txt b/examples/prompts/pick_odd_one_out.txt
new file mode 100644
index 000000000..4a00ceb00
--- /dev/null
+++ b/examples/prompts/pick_odd_one_out.txt
@@ -0,0 +1,9 @@
+Pick the odd word out: skirt, dress, pen, jacket.
+skirt is clothing, dress is clothing, pen is an object, jacket is clothing.
+So the odd one is pen.
+
+Pick the odd word out: Spain, France, German, England, Singapore.
+Spain is a country, France is a country, German is a language, ...
+So the odd one is German.
+
+Pick the odd word out: {{ options | join(", ") }}.
diff --git a/examples/prompts/self_consistency.txt b/examples/prompts/self_consistency.txt
new file mode 100644
index 000000000..29f1f8a74
--- /dev/null
+++ b/examples/prompts/self_consistency.txt
@@ -0,0 +1,6 @@
+{% for example in examples %}
+Q: {{ example.question }}
+A: {{ example.answer }}
+{% endfor %}
+Q: {{ question }}
+A:
diff --git a/examples/react.py b/examples/react.py
index 9c1954e88..b7c807726 100644
--- a/examples/react.py
+++ b/examples/react.py
@@ -11,12 +11,14 @@
 
 """
 
+import json
+
 import requests  # type: ignore
+from openai import OpenAI
 
 import outlines
-from outlines import Template
-import outlines.generate as generate
-import outlines.models as models
+from outlines import Generator, Template
+from outlines.types import JsonSchema
 
 
 build_reAct_prompt = Template.from_string(
@@ -47,32 +49,56 @@ def search_wikipedia(query: str):
     return ".".join(list(page.values())[0]["extract"].split(".")[:2])
 
 
-prompt = build_reAct_prompt("Where is Apple Computers headquarted? ")
-model = models.openai("gpt-4o-mini")
-
-mode_generator = generate.choice(model, choices=["Tho", "Act"])
-action_generator = generate.choice(model, choices=["Search", "Finish"])
-text_generator = generate.text(model)
+prompt = build_reAct_prompt(question="Where is Apple Computers headquarted? ")
+model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
+
+# Define JSON schemas for mode and action
+mode_schema = JsonSchema({
+    "type": "object",
+    "properties": {
+        "result": {
+            "type": "string",
+            "enum": ["Tho", "Act"]
+        }
+    },
+    "required": ["result"]
+})
+action_schema = JsonSchema({
+    "type": "object",
+    "properties": {
+        "result": {
+            "type": "string",
+            "enum": ["Search", "Finish"]
+        }
+    },
+    "required": ["result"]
+})
+
+mode_generator = Generator(model, mode_schema)
+action_generator = Generator(model, action_schema)
+text_generator = Generator(model)
 
 for i in range(1, 10):
-    mode = mode_generator(prompt, max_tokens=128)
-    prompt = add_mode(i, mode, "", prompt)
+    mode_output = mode_generator(prompt, max_tokens=128)
+    mode = json.loads(mode_output)["result"]  # Extract the result from the JSON output
+    prompt = add_mode(i=i, mode=mode, result="", prompt=prompt)
 
     if mode == "Tho":
-        thought = text_generator(prompt, stop_at="\n", max_tokens=128)
+        thought = text_generator(prompt, stop="\n", max_tokens=128)
         prompt += f"{thought}"
     elif mode == "Act":
-        action = action_generator(prompt, max_tokens=128)
+        action_output = action_generator(prompt, max_tokens=128)
+        action = json.loads(action_output)["result"]  # Extract the result from the JSON output
         prompt += f"{action} '"
 
-        subject = text_generator(prompt, stop_at=["'"], max_tokens=128)
+        subject = text_generator(prompt, stop=["'"], max_tokens=128)
         # Apple Computers headquartered
         subject = " ".join(subject.split()[:2])
         prompt += f"{subject}'"
 
         if action == "Search":
             result = search_wikipedia(subject)
-            prompt = add_mode(i, "Obs", result, prompt)
+            prompt = add_mode(i=i, mode="Obs", result=result, prompt=prompt)
         else:
             break
 
diff --git a/examples/sampling.ipynb b/examples/sampling.ipynb
index d6b50e33d..75561451c 100644
--- a/examples/sampling.ipynb
+++ b/examples/sampling.ipynb
@@ -7,14 +7,15 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "#OK\n",
     "import functools as ft\n",
     "import re\n",
     "\n",
     "import numpy as np\n",
     "import matplotlib.pylab as plt\n",
+    "import openai\n",
     "\n",
-    "import outlines.models as models\n",
-    "import outlines.text as text"
+    "import outlines"
    ]
   },
   {
@@ -40,7 +41,7 @@
    "id": "2a3514d6-d5d7-46e9-9b69-1251d337e094",
    "metadata": {},
    "source": [
-    "In this example we will look at completion results for questions from the GSM8K dataset, using few-shots prompts with 5 examples. We first use `outlines.text.prompt` to build the few-shot prompt. `outlines.text.prompt` is a decorator around \"prompt functions\" which contain the prompt template in its docstring. Outlines uses the Jinja2 templating engine to render the prompt when the function is called with the variables' values; it thus allows you to build complex prompts very easily."
+    "In this example we will look at completion results for questions similar to those in the GSM8K dataset, using few-shots prompts with 5 examples. We first use `outlines.Template` to build the few-shot prompt. Outlines uses the Jinja2 templating engine to render the object when the function is called with the variables' values; it thus allows you to build complex prompts very easily."
    ]
   },
   {
@@ -86,8 +87,7 @@
     "]\n",
     "\n",
     "\n",
-    "@text.prompt\n",
-    "def few_shot_prompt(question, examples):\n",
+    "few_shot_prompt = outlines.Template.from_string(\n",
     "    \"\"\"\n",
     "    {% for example in examples %}\n",
     "    Q: {{ example.question }}\n",
@@ -96,9 +96,9 @@
     "    Q: {{ question }}\n",
     "    A:\n",
     "    \"\"\"\n",
+    ")\n",
     "\n",
-    "\n",
-    "# Prompt functions can be partially evaluated like any other function\n",
+    "# Template instances can be partially evaluated because they are callable objects\n",
     "gsm8k_prompt = ft.partial(few_shot_prompt, examples=examples)"
    ]
   },
@@ -107,7 +107,7 @@
    "id": "1eae0ec8-89f0-43fc-b055-6fcd64cbc03b",
    "metadata": {},
    "source": [
-    "## When `text-davinci-003` is uncertain"
+    "## When `gpt-4o-mini` is uncertain"
    ]
   },
   {
@@ -115,7 +115,7 @@
    "id": "a273ed78-e813-467e-85f3-16d7f283ba87",
    "metadata": {},
    "source": [
-    "Let us now sample 20 completions with the `text-davinci-003` model. Outlines is sampling first, and allows to draw several samples with both OpenAI and `transformers` models easily:"
+    "Let us now sample 20 completions with the `gpt-4o-mini` model. Outlines is sampling first, and allows to draw several samples with both OpenAI and `transformers` models easily:"
    ]
   },
   {
@@ -125,11 +125,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model = models.text_completion.openai(\"text-davinci-003\", max_tokens=128)\n",
+    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4o\")\n",
     "\n",
-    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
-    "prompt = gsm8k_prompt(question)\n",
-    "answers = model(prompt, samples=20)"
+    "question = \"When I was 6, my sister was half the age of my brother. When I was 14, my sister was 3 years younger than my brother. Now I'm 70, how old is my sister now?\"\n",
+    "prompt = gsm8k_prompt(question=question)\n",
+    "answers = model(prompt, n=20, max_tokens=512)"
    ]
   },
   {
@@ -201,7 +201,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
@@ -222,7 +222,7 @@
    "id": "661a1135-ac2d-4a49-a786-d04a7ba68b48",
    "metadata": {},
    "source": [
-    "We see that there is an important variabilty in the answers given by `text-davinci-003`. Depending on the number of samples taken, even self-consistency sampling may lead to the wrong result here."
+    "We see that there is an important variabilty in the answers given by `gpt-4o-mini`. Depending on the number of samples taken, even self-consistency sampling may lead to the wrong result here."
    ]
   },
   {
@@ -248,7 +248,7 @@
    "id": "0b15b230-b667-4c9c-8a5d-366dd61de9b7",
    "metadata": {},
    "source": [
-    "## `text-davinci-003` on an easier question"
+    "## `gpt-4o-mini` on an easier question"
    ]
   },
   {
@@ -266,9 +266,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "question = \"John buys 2 pairs of shoes for each of his 3 children. They cost $60 each. How much did he pay?\"\n",
+    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
     "prompt = gsm8k_prompt(question)\n",
-    "answers = model(question, samples=20)"
+    "answers = model(question, samples=20, max_tokens=512)"
    ]
   },
   {
@@ -286,7 +286,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
@@ -316,7 +316,7 @@
    "id": "22f31872-aab7-4a68-b9f2-d335a4f1a875",
    "metadata": {},
    "source": [
-    "## How `gpt-4` compares to `text-davinci-003`\n",
+    "## How `gpt-4` compares to `gpt-4o-mini`\n",
     "\n",
     "Let us now look at how GPT4 fares on the original question:"
    ]
@@ -328,11 +328,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model = models.text_completion.openai(\"gpt-4\", max_tokens=128)\n",
+    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4\")\n",
     "\n",
-    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
+    "question = \"When I was 6, my sister was half the age of my brother. When I was 14, my sister was 3 years younger than my brother. Now I'm 70, how old is my sister now?\"\n",
     "prompt = gsm8k_prompt(question)\n",
-    "answers = model(prompt, samples=20)"
+    "answers = model(prompt, samples=20, max_tokens=512)"
    ]
   },
   {
@@ -350,7 +350,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqwAAAHgCAYAAABgsD+6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAAsTAAALEwEAmpwYAAAQI0lEQVR4nO3dYahf9X3H8c+vua6J3Ui4hKKDOpVMxTmn1Tlwatt0rKKUziZgfJa5B2tg0TkY5EEZLepWYYITwTJ0o5Wi08ypXYdjMxG76OpkLc51jQUdU4ZiHaagREg8e5C/0WhiNWlyP7l9vSDk/s/v/Lnffx6Ed05+95wxTVMAAKDVhxZ6AAAAeC+CFQCAaoIVAIBqghUAgGqCFQCAaoIVAIBqcz9h3T2vAAA4EsaBFlxhBQCgmmAFAKCaYAV4H15++eXcdtttueyyy7Jq1aosW7Ysy5cvzwUXXJDbb789b7zxxn7f9+ijj+aSSy7J/Px8li1bljPPPDM33XRTdu/efYQ/AcDRa/yER7PawwqQ5Ktf/Wo2bNiQ448/Pp/61Kdywgkn5MUXX8y9996bHTt2ZM2aNbnnnnsyxltbsO6///6sWbMmS5cuzeWXX575+fl885vfzPbt27N27drcc889C/iJAOoccA+rYAV4H7Zs2ZJXX301l156aT70obf+c+qFF17Ieeedl+eeey6bN2/OmjVrkiQ//vGPs2rVquzYsSPbtm3LueeemyTZuXNnVq9encceeyx33nln1q1btyCfB6CQH7oCOBSrV6/OZz/72X1iNUmOO+64fOELX0iSPPzww3uPb968OS+99FLWrVu3N1aTZOnSpbnuuuuSJLfeeuvhHxxgERCsAIfomGOOSZLMzb11p8AtW7YkSS6++OJ3nX/RRRfl2GOPzaOPPprXX3/9yAwJcBQTrACHYNeuXfn617+eZN843b59e5LklFNOedd75ubmctJJJ2XXrl155plnjsygAEcxwQpwCDZt2pSnnnoql1xyST7zmc/sPb5jx44kyfLly/f7vjePv/LKK4d9RoCjnWAFOEg333xzbrzxxpx22mm54447FnocgEVLsAIchFtuuSVXX311Tj/99GzdujXz8/P7rL95BfXNK63v9ObxFStWHNY5ARYDwQrwAd10003ZuHFjzjjjjGzdujXHHXfcu8459dRTkyRPP/30u9Z27dqVZ599NnNzczn55JMP+7wARzvBCvAB3HDDDbnmmmty1llnZevWrfnoRz+63/NWr16dJHnwwQfftfbII4/ktddey/nnn58Pf/jDh3VegMVAsAK8T9dee202bdqUc845Jw899FBWrlx5wHPXrl2blStX5q677soTTzyx9/jOnTvzxS9+MUmyYcOGwz4zwGLgSVcA78PXvva1rF+/PkuWLMnGjRv3+9P/J554YtavX7/39X333Ze1a9dm6dKlWbduXebn5/PAAw/sfTTr3Xffvc+jXAF+xnk0K8Ch+NKXvpQvf/nL73nOJz7xiX2edpUk27Zty/XXX5/HHnssO3fuzKpVq3LllVfmqquuypIlSw7jxABHHcEKAEC1AwarPawAAFQTrAAAVBOsAABUm1voAQ7kxE3fWugRAAB+pvz3Vy5d6BH2yxVWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKqNaZoOvDjGg0lWHrlxABaFlUl+tNBDABxlfjRN08X7W3jPYAXggxtjPDFN07kLPQfAYmFLAAAA1QQrAADVBCvAT99fLvQAAIuJPawAAFRzhRUAgGpzCz0AwNFsjLEiyW1JzkgyJbkyyR8mOXV2yookr0zTdNaRnw5gcRCsAIfmL5I8OE3T2jHGzyU5dpqmy99cHGPcmGTHgk0HsAjYwwpwkMYYy5N8L8nJ037+Mh1jjCT/k2T1NE0/PMLjASwa9rACHLyTkryU5K/HGN8dY9w2xvjI29YvTPKiWAU4NIIV4ODNJfl4klunaTo7yatJNr1t/Yokdy7EYACLiWAFOHjPJ3l+mqbvzF5vzp6AzRhjLsnnk/zNAs0GsGgIVoCDNE3TC0meG2O8eUeATyf5/uzr30ryg2manl+Q4QAWEXcJADg0G5N8Y3aHgGeS/O7s+LrYDgDwU+EuAQAAVLMlAACAaoIVAIBqghUAgGqCFQCAaoIVAIBqghXgHcYYvzPGmMYYpy30LAAIVoD9uSLJv8x+XxCzJ2UBEMEKsI8xxs8nuSDJ72XPzf8zxvjkGOPhMcbmMcYPxhjfGGOM2dpXxhjfH2M8Ocb48zHGkjHGs2OPFWOM3WOMi2bnPjLG+OUxxkfGGH81xnh8jPHdMcbnZuvrxxgPjDG2JHloYf4EAPr4FzzAvj6X5MFpmp4eY7w8xjhndvzsJL+S5H+TbEvym2OM/0pyWZLTpmmaxhgrpmnaPcbYnuT0JCcl+fckF44xvpPkY9M0/XCM8adJtkzTdOUYY0WSx8cY/zz7Ph9PcuY0Tf93pD4wQDtXWAH2dUWSu2Zf35W3tgU8Pk3T89M0vZHke0lOTLIjyc4kt48xPp/ktdm5305y0ezXn2XPFdtfT/Jvs/XfTrJpjPG9JA8nWZrkhNnaP4lVgH25wgowM8aYT7I6ya+OMaYkS5JMSb6V5PW3nbo7ydw0TbvGGOcl+XSStUn+YPb+R5JsSPKLSf4kyR8n+WT2hGySjCRrpmna/o7v/xtJXj0sHw7gKOYKK8Bb1ia5Y5qmX5qm6cRpmj6W5NkkF+7v5Nl+1+XTNP1DkmuS/Nps6fEk5yd5Y5qmndlzRfb3sydkk+Qfk2x82z7Ysw/T5wFYFAQrwFuuSPJ37zj2tznw3QJ+IcnfjzGezJ67CvxRkkzT9HqS55L86+y8b8/O/Y/Z62uTHJPkyTHGf85eA3AAY5qmhZ4BAAAOyBVWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKr9P8bb7HZA9fu3AAAAAElFTkSuQmCC",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
diff --git a/examples/self_consistency.py b/examples/self_consistency.py
index 4acba9d58..061ff594c 100644
--- a/examples/self_consistency.py
+++ b/examples/self_consistency.py
@@ -1,9 +1,9 @@
 import re
 
 import numpy as np
+import openai
 
 import outlines
-import outlines.models as models
 from outlines import Template
 
 examples = [
@@ -44,21 +44,12 @@
 question = "When I was 6 my sister was half my age. Now I’m 70 how old is my sister?"
 
 
-few_shots = Template.from_string(
-    """
-    {% for example in examples %}
-    Q: {{ example.question }}
-    A: {{ example.answer }}
-    {% endfor %}
-    Q: {{ question }}
-    A:
-    """
-)
+few_shots = Template.from_file("prompts/self_consistency.txt")
 
-model = models.openai("gpt-4o-mini")
-generator = outlines.generate.text(model)
-prompt = few_shots(question, examples)
-answers = generator(prompt, samples=10)
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
+generator = outlines.Generator(model)
+prompt = few_shots(question=question, examples=examples)
+answers = generator(prompt, n=10)
 
 digits = []
 for answer in answers:
@@ -71,7 +62,7 @@
         print(f"Could not parse the completion: '{answer}'")
 
 unique_digits, counts = np.unique(digits, return_counts=True)
-results = {d: c for d, c in zip(unique_digits, counts)}
+results = {int(d): int(c) for d, c in zip(unique_digits, counts)}
 print(results)
 
 max_count = max(results.values())
diff --git a/examples/simulation_based_inference.ipynb b/examples/simulation_based_inference.ipynb
index 6ad52e58e..e1f888324 100644
--- a/examples/simulation_based_inference.ipynb
+++ b/examples/simulation_based_inference.ipynb
@@ -20,8 +20,9 @@
     "import requests\n",
     "import re\n",
     "\n",
-    "import outlines.models as models\n",
-    "import outlines.text as text\n",
+    "import openai\n",
+    "\n",
+    "import outlines\n",
     "\n",
     "random.seed(0)"
    ]
@@ -49,7 +50,7 @@
    "id": "aabb4db6-fd94-4c42-ab7f-97c3de45b2cc",
    "metadata": {},
    "source": [
-    "In this example we will use GPT 3.5 to solve problems from the GSM-8K dataset. The state-of-the-art performance on this dataset is obtained using few-shot prompting with 5 examples. However, it is not clear how one should select these examples. Here, we will use **simulation-based inference** to try to infer which examples we should be using to get the best out of the model's abilities to solve the problem.\n",
+    "In this example we will use GPT 4 mini to solve problems from the GSM-8K dataset. The state-of-the-art performance on this dataset is obtained using few-shot prompting with 5 examples. However, it is not clear how one should select these examples. Here, we will use **simulation-based inference** to try to infer which examples we should be using to get the best out of the model's abilities to solve the problem.\n",
     "\n",
     "Let's start with downloading the dataset:"
    ]
@@ -112,8 +113,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "@text.prompt\n",
-    "def few_shots(question, examples):\n",
+    "few_shots = outlines.Template.from_string(\n",
     "    \"\"\"\n",
     "    {% for example in examples %}\n",
     "    Q: {{ example.question }}\n",
@@ -122,16 +122,16 @@
     "    Q: {{ question }}\n",
     "    A:\n",
     "    \"\"\"\n",
+    ")\n",
     "\n",
-    "\n",
-    "model = models.text_completion.openai(\"text-davinci-003\", max_tokens=128)\n",
+    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4o-mini\")\n",
     "\n",
     "\n",
     "# TODO: This could largely benefit from vectorization in #52\n",
     "def one_train_example(problem, example_set):\n",
     "    example_ids = random.choices(range(0, len(example_set)), k=5)\n",
     "    examples = [example_set[i] for i in example_ids]\n",
-    "    prompt = few_shots(problem[\"question\"], examples)\n",
+    "    prompt = few_shots(question=problem[\"question\"], examples=examples)\n",
     "    answers_raw = model(prompt, samples=20)\n",
     "\n",
     "    samples = []\n",
@@ -203,7 +203,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
diff --git a/examples/transformers_integration.py b/examples/transformers_integration.py
deleted file mode 100644
index 16e2b8ec3..000000000
--- a/examples/transformers_integration.py
+++ /dev/null
@@ -1,25 +0,0 @@
-"""Example of integrating `outlines` with `transformers`."""
-
-from pydantic import BaseModel
-from transformers import pipeline
-
-from outlines.integrations.transformers import JSONPrefixAllowedTokens
-
-
-class Person(BaseModel):
-    first_name: str
-    surname: str
-
-
-pipe = pipeline("text-generation", model="mistralai/Mistral-7B-v0.1")
-prefix_allowed_tokens_fn = JSONPrefixAllowedTokens(
-    schema=Person, tokenizer_or_pipe=pipe, whitespace_pattern=r" ?"
-)
-results = pipe(
-    ["He is Tom Jones", "She saw Linda Smith"],
-    return_full_text=False,
-    do_sample=False,
-    max_new_tokens=50,
-    prefix_allowed_tokens_fn=prefix_allowed_tokens_fn,
-)
-print(results)
diff --git a/examples/vllm_integration.py b/examples/vllm_offline_integration.py
similarity index 80%
rename from examples/vllm_integration.py
rename to examples/vllm_offline_integration.py
index ee1b24a33..e320c6184 100644
--- a/examples/vllm_integration.py
+++ b/examples/vllm_offline_integration.py
@@ -4,7 +4,7 @@
 from pydantic import BaseModel
 from transformers import AutoTokenizer
 
-from outlines.models.vllm import adapt_tokenizer
+from outlines.models.vllm_offline import adapt_tokenizer
 from outlines.processors import JSONLogitsProcessor
 
 
@@ -17,7 +17,10 @@ class Person(BaseModel):
 llm = vllm.LLM(model=MODEL_ID, max_model_len=512)
 tokenizer = adapt_tokenizer(AutoTokenizer.from_pretrained(MODEL_ID))
 logits_processor = JSONLogitsProcessor(
-    schema=Person, tokenizer=tokenizer, whitespace_pattern=r" ?"
+    schema=Person,
+    tokenizer=tokenizer,
+    tensor_library_name="torch",
+    whitespace_pattern=r" ?"
 )
 result = llm.generate(
     ["He is Tom Jones", "She saw Linda Smith"],
diff --git a/mkdocs.yml b/mkdocs.yml
index 2e77a0440..ce075062b 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -2,7 +2,7 @@
 site_name: Outlines
 site_author: The Outlines developers
 site_description: >-
-  Structured text generation with LLMs
+    Structured text generation with LLMs
 
 # Repository
 repo_name: dottxt-ai/outlines
@@ -11,173 +11,225 @@ repo_url: https://github.com/dottxt-ai/outlines
 # Copyright
 copyright: Copyright &copy; 2023- The Outlines Developers
 
+# Documentation directory
+docs_dir: docs
+
 # Configuration
 theme:
-  name: material
-  custom_dir: docs/overrides
-  palette:
-    - scheme: default
-      primary: white
-  logo: assets/images/logo-square.svg
-  favicon: assets/images/logo-simple.png
-  icon:
-    repo: fontawesome/brands/github
-  features:
-    - content.code.copy
-    - navigation.expand
-    - navigation.tabs
-    - navigation.sections
-    - header.autohide
-    - announce.dismiss
-  font:
-    text: Inter
-    code: Source Code Pro
+    name: material
+    palette:
+        # Palette toggle for light mode
+        - media: "(prefers-color-scheme: light)"
+          scheme: default
+          primary: white
+    logo: assets/images/logo-square.svg
+    favicon: assets/images/logo-square.png
+    icon:
+        repo: fontawesome/brands/github
+    features:
+        - content.code.copy
+        - navigation.expand
+        - navigation.tabs
+        - navigation.sections
+        - header.autohide
+        - announce.dismiss
+    font:
+        text: Inter
+        code: Source Code Pro
 
 # Additional configuration
 extra:
-  social:
-    - icon: fontawesome/brands/github
-      link: https://github.com/dottxt-ai
-    - icon: fontawesome/brands/twitter
-      link: https://twitter.com/remilouf
-  generator: false
-  analytics:
-    provider: google
-    property: !ENV GOOGLE_ANALYTICS_KEY
-  version:
-    provider: mike
-    default: latest
-    alias: true
+    social:
+        - icon: fontawesome/brands/github
+          link: https://github.com/dottxt-ai
+        - icon: fontawesome/brands/twitter
+          link: https://twitter.com/remilouf
+    generator: false
+    analytics:
+        provider: google
+        property: !ENV GOOGLE_ANALYTICS_KEY
+    version:
+        provider: mike
+        default: latest
+        alias: true
 
 # Extensions
 markdown_extensions:
-  - admonition
-  - def_list
-  - attr_list
-  - md_in_html
-  - pymdownx.highlight:
-      anchor_linenums: true
-      line_spans: __span
-      pygments_lang_class: true
-      noclasses: True
-      pygments_style: nord
-  - pymdownx.superfences:
-      custom_fences:
-        - name: mermaid
-          class: mermaid
-          format: !!python/name:pymdownx.superfences.fence_code_format
-  - pymdownx.tabbed:
-      alternate_style: true
-  - pymdownx.inlinehilite
-  - pymdownx.details
-  - pymdownx.emoji:
-      emoji_index: !!python/name:material.extensions.emoji.twemoji
-      emoji_generator: !!python/name:material.extensions.emoji.to_svg
-  - pymdownx.snippets:
+    - admonition
+    - def_list
+    - attr_list
+    - md_in_html
+    - pymdownx.highlight:
+          anchor_linenums: true
+          line_spans: __span
+          pygments_lang_class: true
+          noclasses: True
+    - pymdownx.superfences:
+          custom_fences:
+              - name: mermaid
+                class: mermaid
+                format: !!python/name:pymdownx.superfences.fence_code_format
+    - pymdownx.tabbed:
+          alternate_style: true
+    - pymdownx.inlinehilite
+    - pymdownx.details
+    - pymdownx.emoji:
+          emoji_index: !!python/name:material.extensions.emoji.twemoji
+          emoji_generator: !!python/name:material.extensions.emoji.to_svg
+    - pymdownx.snippets:
 
 extra_css:
-  - stylesheets/extra.css
+    - stylesheets/extra.css
 
 plugins:
-  - blog
-  - mkdocstrings:
-      default_handler: python
-      handlers:
-        python:
-          options:
-            docstring_style: numpy
-            show_submodules: true
-  - search
-  - section-index
-  - social:
-      cards_layout_options:
-        color: #173a58
-  - git-committers:
-      repository: dottxt-ai/outlines
-      branch: main
-  - git-revision-date-localized:
-      enable_creation_date: true
-      type: timeago
+    - blog
+    - mkdocstrings:
+          update_features_models_documentation default_handler: python
+          handlers:
+              python:
+                  options:
+                      docstring_style: numpy
+                      show_submodules: true
+    - search
+    - section-index
+    - social:
+          cards_layout_options:
+              color: #173a58
+    - redirects:
+          redirect_maps:
+              "welcome.md": "index.md"
+              "reference/chat_templating.md": "guide/chat_templating.md"
+              "blog.md": "https://blog.dottxt.co"
+              "cookbook/index.md": "examples/index.md"
+              "cookbook/classification.md": "examples/classification.md"
+              "cookbook/extraction.md": "examples/extraction.md"
+              "cookbook/dating_profiles.md": "examples/dating_profiles.md"
+              "cookbook/chain_of_density.md": "examples/chain_of_density.md"
+              "cookbook/models_playing_chess.md": "examples/models_playing_chess.md"
+              "cookbook/simtom.md": "examples/simtom.md"
+              "cookbook/qa-with-citations.md": "examples/qa-with-citations.md"
+              "cookbook/knowledge_graph_extraction.md": "examples/knowledge_graph_extraction.md"
+              "cookbook/structured_generation_workflow.md": "examples/structured_generation_workflow.md"
+              "cookbook/chain_of_thought.md": "examples/chain_of_thought.md"
+              "cookbook/react_agent.md": "examples/react_agent.md"
+              "cookbook/atomic_caption.md": "examples/atomic_caption.md"
+              "cookbook/read-pdfs.md": "examples/read-pdfs.md"
+              "cookbook/earnings-reports.md": "examples/earnings-reports.md"
+              "cookbook/receipt-digitization.md": "examples/receipt-digitization.md"
+              "cookbook/extract_event_details.md": "examples/extract_event_details.md"
+              "cookbook/deploy-using-bentoml.md": "examples/deploy-using-bentoml.md"
+              "cookbook/deploy-using-cerebrium.md": "examples/deploy-using-cerebrium.md"
+              "cookbook/deploy-using-modal.md": "examples/deploy-using-modal.md"
+
+              # Root files
+              "quickstart.md": "getting_started/index.md"
+
+              # API to features redirects
+              "api/index.md": "features/index.md"
+              "api/json_schema.md": "features/types/json_schema.md"
+              "api/regex.md": "features/types/regex.md"
+              "api/applications.md": "features/utility/application.md"
+              "api/templates.md": "features/utility/template.md"
+              "api/guide.md": "features/advanced/guide.md"
+              "api/parsing.md": "features/advanced/parsing.md"
+              "api/models.md": "features/available_models.md"
+
+              # API subdirectories to features
+              "api/types/": "features/types/index.md"
+              "api/core/": "features/index.md"
+              "api/advanced/": "features/index.md"
+
+    - git-committers:
+        repository: dottxt-ai/outlines
+        branch: main
+    - git-revision-date-localized:
+        enable_creation_date: true
+        type: timeago
+
+    - gen-files:
+        scripts:
+        - scripts/gen_ref_pages.py
+    - literate-nav:
+        nav_file: SUMMARY.md
 
 nav:
-  - Home: index.md
-  - Getting Started:
-      - Welcome to Outlines: welcome.md
-      - Installation: installation.md
-      - Quickstart: quickstart.md
-      - Licence: licence.md
-  - Cookbook:
-      - cookbook/index.md
-      - Classification: cookbook/classification.md
-      - Named Entity Extraction: cookbook/extraction.md
-      - Generate synthetic data: cookbook/dating_profiles.md
-      - Summarize a document: cookbook/chain_of_density.md
-      - Playing chess: cookbook/models_playing_chess.md
-      - Perspective-taking prompting: cookbook/simtom.md
-      - Question-answering with citations: cookbook/qa-with-citations.md
-      - Knowledge Graph Extraction: cookbook/knowledge_graph_extraction.md
-      - Structured Generation Workflow: cookbook/structured_generation_workflow.md
-      - Chain of Thought (CoT): cookbook/chain_of_thought.md
-      - ReAct Agent: cookbook/react_agent.md
-      - Vision-Language Models: cookbook/atomic_caption.md
-      - Structured Generation from PDFs: cookbook/read-pdfs.md
-      - Earnings reports to CSV: cookbook/earnings-reports.md
-      - Digitizing receipts with vision models: cookbook/receipt-digitization.md
-      - Extract events details from text: cookbook/extract_event_details.md
-      - Run on the cloud:
-          - BentoML: cookbook/deploy-using-bentoml.md
-          - Cerebrium: cookbook/deploy-using-cerebrium.md
-          - Modal: cookbook/deploy-using-modal.md
-  - Docs:
-      - reference/index.md
-      - Generation:
-          - Overview: reference/generation/generation.md
-          - Chat templating: reference/chat_templating.md
-          - Regex DSL: reference/regex_dsl.md
-          - Text: reference/text.md
-          - Samplers: reference/samplers.md
-          - Structured generation:
-              - How does it work?: reference/generation/structured_generation_explanation.md
-              - Classification: reference/generation/choices.md
-              - Regex: reference/generation/regex.md
-              - Type constraints: reference/generation/format.md
-              - JSON (function calling): reference/generation/json.md
-              - Grammar: reference/generation/cfg.md
-              - Creating Grammars: reference/generation/creating_grammars.md
-              - Custom FSM operations: reference/generation/custom_fsm_ops.md
-      - Utilities:
-          - Serve with vLLM: reference/serve/vllm.md
-          - Serve with LM Studio: reference/serve/lmstudio.md
-          - Custom types: reference/generation/types.md
-          - Prompt templating: reference/prompting.md
-          - Outlines functions: reference/functions.md
-      - Models:
-          - Overview: reference/models/models.md
-          - Open source:
-              - Transformers: reference/models/transformers.md
-              - Transformers Vision: reference/models/transformers_vision.md
-              - Llama.cpp: reference/models/llamacpp.md
-              - vLLM: reference/models/vllm.md
-              - TGI: reference/models/tgi.md
-              - ExllamaV2: reference/models/exllamav2.md
-              - MLX: reference/models/mlxlm.md
-              - Mamba: reference/models/transformers/#mamba
-          - API:
-              - OpenAI: reference/models/openai.md
-  - API Reference:
-      - api/index.md
-      - api/models.md
-      - api/templates.md
-      - api/json_schema.md
-      - api/guide.md
-      - api/parsing.md
-      - api/regex.md
-      - api/samplers.md
-  - Community:
-      - community/index.md
-      - Feedback 🫶: community/feedback.md
-      - Chat with us ☕: https://discord.com/invite/R9DSu34mGd
-      - How to contribute 🏗️: community/contribute.md
-      - Your projects 👏: community/examples.md
-      - Versioning Guide 📌: community/versioning.md
-  - Blog: https://blog.dottxt.co
+    - Home: index.md
+
+    - Guide:
+          - Getting Started: guide/getting_started.md
+          - Installation: guide/installation.md
+          - Migrating to v1: guide/migration.md
+          - Licence: licence.md
+          - Utility:
+                - Chat Templating for Instruct Models: guide/chat_templating.md
+
+    - Features:
+          - Overview: features/index.md
+
+          - Core:
+                - Models:
+                    - Overview: features/models/index.md
+                    - Anthropic: features/models/anthropic.md
+                    - Dottxt: features/models/dottxt.md
+                    - Gemini: features/models/gemini.md
+                    - Llamacpp: features/models/llamacpp.md
+                    - Mlx-lm: features/models/mlxlm.md
+                    - Ollama: features/models/ollama.md
+                    - OpenAI: features/models/openai.md
+                    - OpenAI compatible API: features/models/openai_compatible.md
+                    - SGLang: features/models/sglang.md
+                    - TGI: features/models/tgi.md
+                    - Transformers: features/models/transformers.md
+                    - TransformersMultiModal: features/models/transformers_multimodal.md
+                    - vLLM (online server): features/models/vllm.md
+                    - vLLM (offline): features/models/vllm_offline.md
+                - Output Types:
+                    - Overview: features/core/output_types.md
+                    - Basic Types: features/core/output_types#basic-python-types
+                    - Multiple-Choices: features/core/output_types#multiple-choices
+                    - JSON: features/core/output_types#json-schemas
+                    - Regex: features/core/output_types#regex-patterns
+                    - Context-free Grammars: features/core/output_types#context-free-grammars
+                - Generator: features/core/generator.md
+
+          - Utilities:
+                - Application: features/utility/application.md
+                - Regex DSL: features/utility/regex_dsl.md
+                - Template: features/utility/template.md
+
+          - Advanced:
+                - Logits Processors: features/advanced/logits_processors.md
+
+    - API Reference: api_reference/
+
+    - Examples:
+          - examples/index.md
+          - Classification: examples/classification.md
+          - Named Entity Extraction: examples/extraction.md
+          - Dating Profiles: examples/dating_profiles.md
+          - Chain of Density: examples/chain_of_density.md
+          - Playing chess: examples/models_playing_chess.md
+          - SimTom: examples/simtom.md
+          - Q&A with Citations: examples/qa-with-citations.md
+          - Knowledge Graph Extraction: examples/knowledge_graph_extraction.md
+          - Structured Generation Workflow: examples/structured_generation_workflow.md
+          - Chain of Thought (CoT): examples/chain_of_thought.md
+          - ReAct Agent: examples/react_agent.md
+          - Vision-Language Models: examples/atomic_caption.md
+          - Structured Generation from PDFs: examples/read-pdfs.md
+          - Earnings Reports to CSV: examples/earnings-reports.md
+          - Receipt Digitization: examples/receipt-digitization.md
+          - Extract Events Details: examples/extract_event_details.md
+          - Run on the cloud:
+                - BentoML: examples/deploy-using-bentoml.md
+                - Cerebrium: examples/deploy-using-cerebrium.md
+                - Modal: examples/deploy-using-modal.md
+    - Community:
+        - community/index.md
+        - Feedback 🫶: community/feedback.md
+        - Our Discord Server ☕: https://discord.com/invite/R9DSu34mGd
+        - How to Contribute 🏗️: community/contribute.md
+        - Community Projects 👏: community/examples.md
+        - Versioning Guide 📌: community/versioning.md
+
+    - Blog: https://blog.dottxt.co
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 628c88f0e..1eebb8bf0 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -1,22 +1,83 @@
 """Outlines is a Generative Model Programming Framework."""
 
-import outlines.generate
+import sys
+from types import ModuleType
+
 import outlines.grammars
-import outlines.models
 import outlines.processors
 import outlines.types
-from outlines.base import vectorize
+from outlines.applications import Application
 from outlines.caching import clear_cache, disable_cache, get_cache
-from outlines.function import Function
+from outlines.generator import Generator
+from outlines.models import (
+    from_dottxt,
+    from_openai,
+    from_transformers,
+    from_gemini,
+    from_anthropic,
+    from_ollama,
+    from_llamacpp,
+    from_mlxlm,
+    from_sglang,
+    from_tgi,
+    from_vllm,
+    from_vllm_offline,
+)
 from outlines.templates import Template, prompt
+from outlines.types import regex, json_schema, cfg
+from outlines.templates import Vision
+
+from .v0_legacy import (
+    generate,
+    samplers,
+    models as legacy_models,
+    function,
+)
+from .v0_legacy.function import Function
+
+model_list = [
+    "from_anthropic",
+    "from_dottxt",
+    "from_gemini",
+    "from_llamacpp",
+    "from_mlxlm",
+    "from_ollama",
+    "from_openai",
+    "from_tgi",
+    "from_transformers",
+    "from_vllm",
+]
 
 __all__ = [
     "clear_cache",
     "disable_cache",
     "get_cache",
-    "Function",
+    "Application",
+    "Generator",
+    "regex",
+    "json_schema",
+    "cfg",
     "prompt",
-    "Prompt",
-    "vectorize",
+    "Template",
     "grammars",
-]
+] + model_list
+
+
+# v0 legacy
+
+generate_module = ModuleType("generate")
+generate_module.__dict__.update(generate.__dict__)
+
+function_module = ModuleType("function")
+function_module.__dict__.update(function.__dict__)
+
+samplers_module = ModuleType("samplers")
+samplers_module.__dict__.update(samplers.__dict__)
+
+outlines.models.__dict__.update(legacy_models.__dict__)
+
+sys.modules["outlines.generate"] = generate_module
+sys.modules["outlines.samplers"] = samplers_module
+sys.modules["outlines.function"] = function_module
+
+__all__ += ["Function"]
diff --git a/outlines/applications.py b/outlines/applications.py
new file mode 100644
index 000000000..4b4b34187
--- /dev/null
+++ b/outlines/applications.py
@@ -0,0 +1,107 @@
+"""Encapsulate a prompt template and an output type into a reusable object."""
+
+from typing import Any, Callable, Dict, Optional, Union
+
+from outlines.generator import (
+    BlackBoxGenerator,
+    Generator,
+    SteerableGenerator,
+)
+from outlines.models.base import Model
+from outlines.templates import Template
+
+
+class Application:
+    """
+    Application is a class that encapsulates a prompt template and an
+    output type. It can be called to generate a response by providing a
+    model, the values to be substituted in the template in a dictionary
+    and optional inference parameters.
+
+    Parameters
+    ----------
+    template : Union[Template, Callable]
+        A callable that takes arguments and returns a prompt string.
+    output_type : Any
+        The expected output type of the generated response.
+
+    Examples
+    --------
+    ```python
+    from pydantic import BaseModel
+    from transformers import AutoModelForCausalLM, AutoTokenizer
+    from outlines import models, Application
+    from outlines.types import JsonType
+    from outlines.templates import Template
+
+    class OutputModel(BaseModel):
+        result: int
+
+    model = models.from_transformers(
+        AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+        AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+    )
+
+    template_string = "What is 2 times {{ num }}?"
+    template = Template.from_string(template_string)
+
+    application = Application(template, JsonType(OutputModel))
+
+    result = application(model, {"num": 3}, max_new_tokens=20)
+    print(result)  # Expected output: { "result" : 6 }
+    ```
+
+    """
+    def __init__(
+        self,
+        template: Union[Template, Callable],
+        output_type: Optional[Any] = None,
+    ):
+        """
+        Parameters
+        ----------
+        template
+            The template to use to build the prompt.
+        output_type
+            The output type provided to the generator.
+
+        """
+        self.template = template
+        self.output_type = output_type
+        self.generator: Optional[Union[
+            BlackBoxGenerator, SteerableGenerator
+        ]] = None
+        self.model: Optional[Model] = None
+
+    def __call__(
+        self,
+        model: Model,
+        template_vars: Dict[str, Any],
+        **inference_kwargs
+    ) -> Any:
+        """
+        Parameters
+        ----------
+        model
+            The model to use to generate the response.
+        template_vars
+            The variables to be substituted in the template.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+        Returns
+        -------
+        Any
+            The generated response.
+        """
+        if model is None:
+            raise ValueError("you must provide a model")
+        # We save the generator to avoid creating a new one for each call.
+        # If the model has changed since the last call, we create a new
+        # generator.
+        if model != self.model:
+            self.model = model
+            self.generator = Generator(model, self.output_type)  # type: ignore
+
+        prompt = self.template(**template_vars)
+        assert self.generator is not None
+        return self.generator(prompt, **inference_kwargs)
diff --git a/outlines/caching.py b/outlines/caching.py
index 6882bac6b..7cb2ed169 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -1,3 +1,5 @@
+"""Caching and memoization of function calls."""
+
 import asyncio
 import contextlib
 import functools
@@ -12,7 +14,7 @@
 _caching_enabled = True
 
 
-class CloudpickleDisk(Disk):
+class CloudpickleDisk(Disk): # pragma: no cover
     def __init__(self, directory, compress_level=1, **kwargs):
         self.compress_level = compress_level
         super().__init__(directory, **kwargs)
@@ -57,11 +59,11 @@ def get_cache():
     if outlines_cache_dir:
         # OUTLINES_CACHE_DIR takes precendence
         cache_dir = outlines_cache_dir
-    elif xdg_cache_home:
+    elif xdg_cache_home:  # pragma: no cover
         cache_dir = os.path.join(xdg_cache_home, ".cache", "outlines")
     elif home_dir != "/":
         cache_dir = os.path.join(home_dir, ".cache", "outlines")
-    else:
+    else:  # pragma: no cover
         # home_dir may be / inside a docker container without existing user
         tempdir = tempfile.gettempdir()
         cache_dir = os.path.join(tempdir, ".cache", "outlines")
@@ -108,7 +110,7 @@ def decorator(cached_function: Callable):
 
         base = (full_name(cached_function),)
 
-        if asyncio.iscoroutinefunction(cached_function):
+        if asyncio.iscoroutinefunction(cached_function):  # pragma: no cover
 
             async def wrapper(*args, **kwargs):
                 if not _caching_enabled:
diff --git a/outlines/fsm/__init__.py b/outlines/fsm/__init__.py
index e69de29bb..4e38718ef 100644
--- a/outlines/fsm/__init__.py
+++ b/outlines/fsm/__init__.py
@@ -0,0 +1 @@
+"""Finite state machines builder for CFG generation."""
diff --git a/outlines/fsm/json_schema.py b/outlines/fsm/json_schema.py
deleted file mode 100644
index 578ee7626..000000000
--- a/outlines/fsm/json_schema.py
+++ /dev/null
@@ -1,83 +0,0 @@
-import inspect
-import json
-import warnings
-from enum import Enum
-from typing import Callable, Type, Union
-
-from pydantic import BaseModel, create_model
-
-
-def convert_json_schema_to_str(json_schema: Union[dict, str, Type[BaseModel]]) -> str:
-    """Convert a JSON schema to a string.
-
-    Parameters
-    ----------
-    json_schema
-        The JSON schema.
-
-    Returns
-    -------
-    str
-        The JSON schema converted to a string.
-
-    Raises
-    ------
-    ValueError
-        If the schema is not a dictionary, a string or a Pydantic class.
-    """
-    if isinstance(json_schema, dict):
-        schema_str = json.dumps(json_schema)
-    elif isinstance(json_schema, str):
-        schema_str = json_schema
-    elif issubclass(json_schema, BaseModel):
-        schema_str = json.dumps(json_schema.model_json_schema())
-    else:
-        raise ValueError(
-            f"Cannot parse schema {json_schema}. The schema must be either "
-            + "a Pydantic class, a dictionary or a string that contains the JSON "
-            + "schema specification"
-        )
-    return schema_str
-
-
-def get_schema_from_signature(fn: Callable) -> dict:
-    """Turn a function signature into a JSON schema.
-
-    Every JSON object valid to the output JSON Schema can be passed
-    to `fn` using the ** unpacking syntax.
-
-    """
-    signature = inspect.signature(fn)
-    arguments = {}
-    for name, arg in signature.parameters.items():
-        if arg.annotation == inspect._empty:
-            raise ValueError("Each argument must have a type annotation")
-        else:
-            arguments[name] = (arg.annotation, ...)
-
-    try:
-        fn_name = fn.__name__
-    except Exception as e:
-        fn_name = "Arguments"
-        warnings.warn(
-            f"The function name could not be determined. Using default name 'Arguments' instead. For debugging, here is exact error:\n{e}",
-            category=UserWarning,
-        )
-    model = create_model(fn_name, **arguments)
-
-    return model.model_json_schema()
-
-
-def get_schema_from_enum(myenum: type[Enum]) -> dict:
-    if len(myenum) == 0:
-        raise ValueError(
-            f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
-        )
-    choices = [
-        get_schema_from_signature(elt.value.func)
-        if callable(elt.value)
-        else {"const": elt.value}
-        for elt in myenum
-    ]
-    schema = {"title": myenum.__name__, "oneOf": choices}
-    return schema
diff --git a/outlines/fsm/parsing.py b/outlines/fsm/parsing.py
index e48fb69e4..3831c57c1 100644
--- a/outlines/fsm/parsing.py
+++ b/outlines/fsm/parsing.py
@@ -1,3 +1,5 @@
+"""Build a parser from a grammar to create a finite state machine."""
+
 from copy import copy, deepcopy
 from dataclasses import dataclass
 from functools import lru_cache
diff --git a/outlines/fsm/types.py b/outlines/fsm/types.py
deleted file mode 100644
index f6409aa66..000000000
--- a/outlines/fsm/types.py
+++ /dev/null
@@ -1,76 +0,0 @@
-import datetime
-from enum import EnumMeta
-from typing import Any, Protocol, Tuple, Type
-
-from outlines.types import Regex, boolean as boolean_regex, date as date_regex
-from outlines.types import datetime as datetime_regex
-from outlines.types import (
-    integer as integer_regex,
-    number as number_regex,
-    time as time_regex,
-)
-
-
-class FormatFunction(Protocol):
-    def __call__(self, sequence: str) -> Any: ...
-
-
-def python_types_to_regex(python_type: Type) -> Tuple[Regex, FormatFunction]:
-    # If it is a custom type
-    if isinstance(python_type, Regex):
-        custom_regex_str = python_type.pattern
-
-        def custom_format_fn(sequence: str) -> str:
-            return str(sequence)
-
-        return Regex(custom_regex_str), custom_format_fn
-
-    if isinstance(python_type, EnumMeta):
-        values = python_type.__members__.keys()
-        enum_regex_str: str = "(" + "|".join(values) + ")"
-
-        def enum_format_fn(sequence: str) -> str:
-            return str(sequence)
-
-        return Regex(enum_regex_str), enum_format_fn
-
-    if python_type is float:
-
-        def float_format_fn(sequence: str) -> float:
-            return float(sequence)
-
-        return number_regex, float_format_fn
-    elif python_type is int:
-
-        def int_format_fn(sequence: str) -> int:
-            return int(sequence)
-
-        return integer_regex, int_format_fn
-    elif python_type is bool:
-
-        def bool_format_fn(sequence: str) -> bool:
-            return bool(sequence)
-
-        return boolean_regex, bool_format_fn
-    elif python_type == datetime.date:
-
-        def date_format_fn(sequence: str) -> datetime.date:
-            return datetime.datetime.strptime(sequence, "%Y-%m-%d").date()
-
-        return date_regex, date_format_fn
-    elif python_type == datetime.time:
-
-        def time_format_fn(sequence: str) -> datetime.time:
-            return datetime.datetime.strptime(sequence, "%H:%M:%S").time()
-
-        return time_regex, time_format_fn
-    elif python_type == datetime.datetime:
-
-        def datetime_format_fn(sequence: str) -> datetime.datetime:
-            return datetime.datetime.strptime(sequence, "%Y-%m-%d %H:%M:%S")
-
-        return datetime_regex, datetime_format_fn
-    else:
-        raise NotImplementedError(
-            f"The Python type {python_type} is not supported. Please open an issue."
-        )
diff --git a/outlines/generate/api.py b/outlines/generate/api.py
deleted file mode 100644
index 5d3c52a8a..000000000
--- a/outlines/generate/api.py
+++ /dev/null
@@ -1,609 +0,0 @@
-import datetime
-from copy import copy
-from dataclasses import dataclass
-from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
-
-from outlines.generate.generator import sequence_generator
-
-if TYPE_CHECKING:
-    import torch
-
-FormattedOutput = Union[
-    str, int, float, bool, datetime.date, datetime.time, datetime.datetime
-]
-
-
-class SequenceGenerator:
-    def __init__(
-        self,
-        fsm,
-        model,
-        sampler,
-        device,
-    ):
-        self.fsm = fsm
-        self.model = model
-        self.sampler = sampler
-        self.tokenizer = model.tokenizer
-        self.device = device
-        self.num_samples = sampler.samples
-
-    def get_generated_token_ids(
-        self,
-        prompt_token_ids: "torch.Tensor",
-        token_ids: "torch.Tensor",
-    ) -> List["torch.Tensor"]:
-        """Get the tokens generated so far.
-
-        Parameters
-        ----------
-        prompt_token_ids
-            Tensor that contains the token ids of the sequences' prompts.
-        token_ids
-            The generated token ids.
-
-        Returns
-        -------
-        A tensor that contains the token ids that have been generated so far.
-
-        """
-        prompt_lengths = [len(prompt) for prompt in prompt_token_ids]
-        token_ids = [
-            cur_token_ids[length:]
-            for cur_token_ids, length in zip(token_ids, prompt_lengths)
-        ]
-
-        return token_ids
-
-    def is_stop_sequence_found(
-        self, generated_sequences: List[str], stop_sequences: List[str]
-    ) -> bool:
-        """Determine whether one of the stop sequences has been generated.
-
-        Parameters
-        ----------
-        generated_sequences
-            The list of sequences generated so far.
-        stop_sequences
-            The list that contains the sequence which stop the generation when
-            found.
-
-        Returns
-        -------
-        True if at least one of the stop sequences has been found in each generated
-        sequence.
-
-        """
-        return all(
-            [
-                any([seq in generated for seq in stop_sequences])
-                for generated in generated_sequences
-            ]
-        )
-
-    def strip_stop_sequences(
-        self, sequence: str, stop_sequences: Optional[List[str]]
-    ) -> str:
-        """Remove the stop sequences from the generated sequences.
-
-        Parameters
-        ----------
-        sequence
-            One of the generated sequences.
-        stop_sequences
-            The list that contains the sequence which stop the generation when
-            found.
-
-        """
-        if stop_sequences:
-            match_indexes = [sequence.find(seq) for seq in stop_sequences]
-            if any([index != -1 for index in match_indexes]):
-                # select the stop_sequence that is found first in the sequence
-                min_match_index_value = min([i for i in match_indexes if i != -1])
-                min_match_index_pos = match_indexes.index(min_match_index_value)
-                sequence = sequence[
-                    : match_indexes[min_match_index_pos]
-                    + len(stop_sequences[min_match_index_pos])
-                ]
-
-        return sequence
-
-    def format_sequence(self, sequence: str) -> FormattedOutput:
-        """Translate the generated sequence to another type.
-
-        This method is for instance overridden when generating JSON to either
-        return a dictionnary or a Pydantic model.
-
-        Parameters
-        ----------
-        sequence
-            A generated sequences.
-
-        Returns
-        -------
-        The formatted sequence.
-
-        """
-        return sequence
-
-    def __call__(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        rng: Optional["torch.Generator"] = None,
-    ) -> Union[FormattedOutput, List[FormattedOutput], List[List[FormattedOutput]]]:
-        """Generate the full text sequence.
-
-        Since `SequenceGenerator.stream` calls the tokenizer at every step this
-        method loops over the generator returned by `sequence_generator` itself
-        so the tokenizer is called only once after all token ids have been
-        generated.
-
-        Parameters
-        ----------
-        prompts
-            A string or list of strings that are passed to the model before
-            generating the first token.
-        max_tokens
-            An integer representing maximum number of tokens that will be generated
-            (per prompt)
-        stop_at
-            A string or list of strings at which the text generated will stop
-        rng
-            The random number generator. Defaults to a non-seeded `torch.Generator`
-            instance.
-
-        Returns
-        -------
-        The generation(s), potentially cast to another type.
-        """
-        import torch
-
-        if isinstance(prompts, str):
-            prompts = [prompts]
-
-        if isinstance(stop_at, str):
-            stop_at = [stop_at]
-
-        stop_sequences = stop_at
-        num_samples = self.num_samples
-
-        if rng is None:
-            rng = torch.Generator(device=self.device)
-            rng.seed()
-
-        prompt_token_ids, attention_masks = self.tokenizer.encode(prompts)
-        prompt_token_ids = prompt_token_ids.to(self.device)
-        attention_masks = attention_masks.to(self.device)
-
-        # To draw multiple samples we repeat the prompt as many times
-        # as there are samples. We copy the FSMs and initialize the
-        # FSM states.
-        num_samples = self.num_samples
-        batch_size = len(prompts)
-
-        prompt_token_ids = torch.repeat_interleave(prompt_token_ids, num_samples, dim=0)
-        attention_masks = torch.repeat_interleave(attention_masks, num_samples, dim=0)
-        fsm_states = [0 for _ in range(batch_size * num_samples)]
-        fsms = [self.fsm.copy() for _ in range(batch_size * num_samples)]
-        weights = torch.zeros(
-            (batch_size * num_samples), dtype=torch.float, device=self.device
-        )
-
-        states = sequence_generator(
-            self.model,
-            self.sampler,
-            fsms,
-            prompt_token_ids,
-            weights,
-            attention_masks,
-            fsm_states,
-            rng=rng,
-        )
-
-        while True:
-            try:
-                last_state = next(states)
-                if max_tokens or stop_sequences:
-                    token_ids = last_state.token_ids
-                    generated_token_ids = self.get_generated_token_ids(
-                        prompt_token_ids, token_ids
-                    )
-                    if max_tokens and len(generated_token_ids[0]) >= max_tokens:
-                        break
-                    if stop_sequences and self.is_stop_sequence_found(
-                        self.tokenizer.decode(generated_token_ids), stop_sequences
-                    ):
-                        break
-            except StopIteration:
-                break
-
-        token_ids = last_state.token_ids
-        generated_token_ids = self.get_generated_token_ids(prompt_token_ids, token_ids)
-
-        generated = self.tokenizer.decode(generated_token_ids)
-        stripped = [
-            self.strip_stop_sequences(sequence, stop_sequences)
-            for sequence in generated
-        ]
-        formatted = [self.format_sequence(sequence) for sequence in stripped]
-
-        # We reshape the output to (batch_size, sample_size)
-        output: List[List[FormattedOutput]] = list()
-        for i in range(0, batch_size * num_samples, num_samples):
-            output.append(formatted[i : i + num_samples])
-
-        # We remove leading dimensions for the output
-        if batch_size == 1 and num_samples == 1:
-            return output[0][0]
-        elif batch_size == 1:
-            return output[0]
-        elif num_samples == 1:
-            return [samples[0] for samples in output]
-        else:
-            return output
-
-    def stream(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        rng: Optional["torch.Generator"] = None,
-    ) -> Iterator[Union[List[str], str, List[List[str]]]]:
-        """Generate the text sequence one token at a time.
-
-        Since `Tokenizer.decode` strips the whitespaces from the tokens we have no
-        choice but to decode the generated token ids at each step and compare the
-        current decoded strings to the previously decoded strings.
-
-        Parameters
-        ----------
-        prompts
-            A string or list of strings that are passed to the model before
-            generating the first token.
-        max_tokens
-            An integer representing maximum number of tokens that will be generated
-            (per prompt)
-        stop_at
-            A string or list of strings at which the text generated will stop
-        rng
-            The random number generator. Defaults to a non-seeded `torch.Generator`
-            instance.
-
-        Returns
-        -------
-        A string or list of strings that contain the generated text.
-
-        """
-        import torch
-
-        if isinstance(prompts, str):
-            prompts = [prompts]
-
-        if isinstance(stop_at, str):
-            stop_at = [stop_at]
-
-        stop_sequences = stop_at
-        num_samples = self.num_samples
-
-        prompt_token_ids, attention_masks = self.tokenizer.encode(prompts)
-        prompt_token_ids = prompt_token_ids.to(self.device)
-        attention_masks = attention_masks.to(prompt_token_ids.device)
-
-        # To draw multiple samples we repeat the prompt as many times
-        # as there are samples. We copy the FSMs and initialize the
-        # FSM states.
-        num_samples = self.num_samples
-        batch_size = len(prompts)
-
-        prompt_token_ids = torch.repeat_interleave(prompt_token_ids, num_samples, dim=0)
-        attention_masks = torch.repeat_interleave(attention_masks, num_samples, dim=0)
-        fsm_states = [0 for _ in range(batch_size * num_samples)]
-        fsms = [self.fsm.copy() for _ in range(batch_size * num_samples)]
-        weights = torch.zeros(
-            (batch_size * num_samples),
-            dtype=torch.float,
-            device=prompt_token_ids.device,
-        )
-
-        if rng is None:
-            rng = torch.Generator(device=prompt_token_ids.device)
-            rng.seed()
-
-        states = sequence_generator(
-            self.model,
-            self.sampler,
-            fsms,
-            prompt_token_ids,
-            weights,
-            attention_masks,
-            fsm_states,
-            rng=rng,
-        )
-
-        def token_generator() -> Iterator[Union[List[str], str, List[List[str]]]]:
-            previously_generated_sequences = [
-                "" for _ in range(batch_size)
-            ] * num_samples
-            num_generated = 0
-            is_stop_at_reached = [False for _ in range(batch_size)] * num_samples
-            while True:
-                if (max_tokens and num_generated >= max_tokens) or all(
-                    is_stop_at_reached
-                ):
-                    return
-                try:
-                    sequence = next(states)
-                    num_generated += 1
-                except StopIteration:
-                    return
-                generated_token_ids = sequence.token_ids[:, -num_generated:]
-                generated_sequences = self.tokenizer.decode(generated_token_ids)
-                if stop_sequences:
-                    is_stop_at_reached = [
-                        stop
-                        or self.is_stop_sequence_found(
-                            [generated_sequence], stop_sequences
-                        )
-                        for generated_sequence, stop in zip(
-                            generated_sequences, is_stop_at_reached
-                        )
-                    ]
-
-                    generated_sequences = [
-                        (
-                            self.format_sequence(
-                                self.strip_stop_sequences(sequence, stop_sequences)
-                            )
-                            if stop
-                            else sequence
-                        )
-                        for sequence, stop in zip(
-                            generated_sequences, is_stop_at_reached
-                        )
-                    ]
-                next_tokens = [
-                    token[len(sequence) :]
-                    for token, sequence, stop in zip(
-                        generated_sequences,
-                        previously_generated_sequences,
-                        is_stop_at_reached,
-                    )
-                ]
-                previously_generated_sequences = generated_sequences
-                # We reshape the output to (batch_size, sample_size)
-                output: List[List[str]] = list()
-                for i in range(0, batch_size * num_samples, num_samples):
-                    output.append(next_tokens[i : i + num_samples])
-
-                # We remove leading dimensions for the output
-                if batch_size == 1 and num_samples == 1:
-                    yield output[0][0]
-                elif batch_size == 1:
-                    yield output[0]
-                elif num_samples == 1:
-                    yield [samples[0] for samples in output]
-                else:
-                    yield output
-
-        return token_generator()
-
-
-@dataclass(frozen=True)
-class GenerationParameters:
-    """Generation parameters used in Outlines' public API."""
-
-    max_tokens: Optional[int]
-    stop_at: Optional[Union[str, List[str]]]
-    seed: Optional[int]
-
-
-@dataclass(frozen=True)
-class SamplingParameters:
-    """Sampling parameters available in Outlines."""
-
-    sampler: str
-    num_samples: int = 1
-    top_p: Optional[float] = None
-    top_k: Optional[int] = None
-    temperature: Optional[float] = None
-
-
-class SequenceGeneratorAdapter:
-    """Class used to unify the interface to the model providers'
-    generation functions.
-
-    Attributes
-    ----------
-    model
-        The wrapped model.
-    logits_processor
-        The logits processor to use to generate text.
-    sampler
-        The sampler to use to generate text.
-
-    """
-
-    def __init__(self, model, logits_processor, sampler):
-        self.model = model
-        self.logits_processor = logits_processor
-
-        self.sampling_params = sampler.sampling_params
-
-    def prepare_generation_parameters(
-        self,
-        max_tokens: Optional[int],
-        stop_at: Optional[Union[str, List[str]]],
-        seed: Optional[int],
-    ):
-        if isinstance(stop_at, str):
-            stop_at = [stop_at]
-
-        generation_params = GenerationParameters(
-            max_tokens,
-            stop_at,
-            seed,
-        )
-
-        return generation_params
-
-    def format_sequence(self, sequence: str) -> FormattedOutput:
-        """Translate the generated sequence to another type.
-
-        This method is for instance overridden when generating JSON to either
-        return a dictionnary or a Pydantic model.
-
-        Parameters
-        ----------
-        sequence
-            A generated sequences.
-
-        Returns
-        -------
-        The formatted sequence.
-
-        """
-        return sequence
-
-    def _format(self, sequences):
-        """Apply formatting to every string in a completion."""
-        if isinstance(sequences, list):
-            return [self._format(sequence) for sequence in sequences]
-        else:
-            return self.format_sequence(sequences)
-
-    def __call__(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """Generate text from a prompt of list of prompts."""
-
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-
-        completions = self.model.generate(
-            prompts,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-        return self._format(completions)
-
-    def stream(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """Return a text generator from a prompt or a list of prompts."""
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-        return self.model.stream(
-            prompts,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-
-class VisionSequenceGeneratorAdapter(SequenceGeneratorAdapter):
-    def __call__(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[str, Any],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """
-        Generate text from a prompt of list of prompts.
-
-        Media: A URI to construct media or media object itself. Used as AutoProcessor argument.
-        """
-        prompts, media = self._validate_prompt_media_types(prompts, media)
-
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-
-        completions = self.model.generate(
-            prompts,
-            media,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-        return self._format(completions)
-
-    def stream(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: List[Union[str, Any, List[Union[str, Any]]]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """Return a text generator from a prompt or a list of prompts."""
-        prompts, media = self._validate_prompt_media_types(prompts, media)
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-        return self.model.stream(
-            prompts,
-            media,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-    @classmethod
-    def _validate_prompt_media_types(
-        cls,
-        prompts: Union[str, List[str]],
-        media: Union[str, Any, List[Union[str, Any]]],
-    ) -> Union[Any, List[Any]]:
-        """
-        Prepare media as PIL.Image and ensure for every prompt str there is one List[PIL.Image]
-        """
-
-        def valid_types(prompts, media):
-            from PIL import Image  # type: ignore
-
-            if isinstance(prompts, list):
-                if not isinstance(media, list) or len(prompts) != len(media):
-                    return False
-                for subprompt, submedia in zip(prompts, media):
-                    if not isinstance(subprompt, str) or not all(
-                        isinstance(m, Image.Image) for m in submedia
-                    ):
-                        return False
-            elif isinstance(prompts, str):
-                if not all(isinstance(m, Image.Image) for m in media):
-                    return False
-            return True
-
-        if not valid_types(prompts, media):
-            raise TypeError(
-                "Expected (prompts, media) to be of type "
-                "(str, List[Image])), or (List[str], List[List[Image]]) "
-                f"instead got prompts={prompts}, media={media}"
-            )
-
-        return prompts, media
diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
deleted file mode 100644
index b677040d5..000000000
--- a/outlines/generate/cfg.py
+++ /dev/null
@@ -1,54 +0,0 @@
-from functools import singledispatch
-
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import LlamaCpp, OpenAI, TransformersVision
-from outlines.samplers import Sampler, multinomial
-
-
-@singledispatch
-def cfg(
-    model, cfg_str: str, sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    """Generate text in the language of a Context-Free Grammar
-
-    Arguments
-    ---------
-    model:
-        An `outlines.model` instance.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGeneratorAdapter` instance that generates text.
-
-    """
-    from outlines.processors import CFGLogitsProcessor
-
-    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@cfg.register(TransformersVision)
-def cfg_vision(model, cfg_str: str, sampler: Sampler = multinomial()):
-    from outlines.processors import CFGLogitsProcessor
-
-    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
-    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@cfg.register(LlamaCpp)
-def cfg_llamacpp(model, cfg_str: str, sampler: Sampler = multinomial()):
-    raise NotImplementedError("Not yet available due to bug in llama_cpp tokenizer")
-
-
-@cfg.register(OpenAI)
-def cfg_openai(model, cfg_str: str, sampler: Sampler = multinomial()):
-    raise NotImplementedError(
-        "Cannot use grammar-structured generation with an OpenAI model"
-        + "due to the limitations of the OpenAI API."
-    )
diff --git a/outlines/generate/choice.py b/outlines/generate/choice.py
deleted file mode 100644
index afb998f52..000000000
--- a/outlines/generate/choice.py
+++ /dev/null
@@ -1,59 +0,0 @@
-import json as pyjson
-import re
-from enum import Enum
-from functools import singledispatch
-from typing import Callable, List, Union
-
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-from outlines.fsm.json_schema import get_schema_from_enum
-from outlines.generate.api import SequenceGeneratorAdapter
-from outlines.models import OpenAI
-from outlines.samplers import Sampler, multinomial
-
-from .json import json
-from .regex import regex
-
-
-@singledispatch
-def choice(
-    model, choices: Union[List[str], type[Enum]], sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    if isinstance(choices, type(Enum)):
-        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
-    else:
-        choices = [re.escape(choice) for choice in choices]  # type: ignore
-        regex_str = r"(" + r"|".join(choices) + r")"
-
-    generator = regex(model, regex_str, sampler)
-    if isinstance(choices, type(Enum)):
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    else:
-        generator.format_sequence = lambda x: x
-
-    return generator
-
-
-@choice.register(OpenAI)
-def choice_openai(
-    model: OpenAI, choices: List[str], sampler: Sampler = multinomial()
-) -> Callable:
-    """
-    Call OpenAI API with response_format of a dict:
-    {"result": <one of choices>}
-    """
-
-    choices_schema = pyjson.dumps(
-        {
-            "type": "object",
-            "properties": {"result": {"type": "string", "enum": choices}},
-            "additionalProperties": False,
-            "required": ["result"],
-        }
-    )
-    generator = json(model, choices_schema, sampler)
-
-    def generate_choice(*args, **kwargs):
-        return generator(*args, **kwargs)["result"]
-
-    return generate_choice
diff --git a/outlines/generate/format.py b/outlines/generate/format.py
deleted file mode 100644
index 56ed10ac1..000000000
--- a/outlines/generate/format.py
+++ /dev/null
@@ -1,48 +0,0 @@
-from functools import singledispatch
-
-from outlines.fsm.types import python_types_to_regex
-from outlines.generate.api import SequenceGeneratorAdapter
-from outlines.models import OpenAI
-from outlines.samplers import Sampler, multinomial
-
-from .regex import regex
-
-
-@singledispatch
-def format(
-    model, python_type, sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    """Generate structured data that can be parsed as a Python type.
-
-    Parameters
-    ----------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    python_type:
-        A Python type. The output of the generator must be parseable into
-        this type.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGenerator` instance that generates text constrained by the Python type
-    and translates this text into the corresponding type.
-
-    """
-    regex_str, format_fn = python_types_to_regex(python_type)
-    regex_str = regex_str.pattern
-    generator = regex(model, regex_str, sampler)
-    generator.format_sequence = format_fn
-
-    return generator
-
-
-@format.register(OpenAI)
-def format_openai(model, python_type, sampler: Sampler = multinomial()):
-    raise NotImplementedError(
-        "Cannot use Python type-structured generation with an OpenAI model"
-        + " due to the limitations of the OpenAI API."
-    )
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
deleted file mode 100644
index 1950812d2..000000000
--- a/outlines/generate/fsm.py
+++ /dev/null
@@ -1,31 +0,0 @@
-from functools import singledispatch
-
-import interegular
-
-from outlines.fsm.guide import RegexGuide
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import TransformersVision
-from outlines.samplers import Sampler, multinomial
-
-
-@singledispatch
-def fsm(
-    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    from outlines.processors import GuideLogitsProcessor
-
-    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@fsm.register(TransformersVision)
-def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()):
-    from outlines.processors import GuideLogitsProcessor
-
-    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
-    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
diff --git a/outlines/generate/generator.py b/outlines/generate/generator.py
deleted file mode 100644
index e506aa035..000000000
--- a/outlines/generate/generator.py
+++ /dev/null
@@ -1,312 +0,0 @@
-import dataclasses
-import math
-from typing import TYPE_CHECKING, Callable, Iterable, Iterator, List, Optional, Tuple
-
-if TYPE_CHECKING:
-    import torch
-
-    from outlines.fsm.guide import Guide
-
-
-class ContextLengthExceededError(Exception):
-    pass
-
-
-@dataclasses.dataclass(frozen=True)
-class GenerationState:
-    token_ids: "torch.Tensor"
-    kv_cache: "torch.Tensor"
-    logits: "torch.Tensor"
-    weights: "torch.Tensor"
-    fsm_states: List[int]
-
-
-def sequence_generator(
-    model: Callable,
-    sampler: Callable,
-    fsms: List["Guide"],
-    token_ids: "torch.Tensor",
-    sequence_weights: "torch.Tensor",
-    attention_masks: "torch.Tensor",
-    fsm_states: List[int],
-    rng: "torch.Generator",
-) -> Iterator[GenerationState]:
-    """Generates sequences of tokens.
-
-    Parameters
-    ----------
-    model
-        A callable that generates a probability distribution over the
-        vocabulary when passed a tensor of token ids.
-    sampler
-        A callable that returns the next token ids, their ancestor sequence and
-        the updated sequence weights when passed a distribution over the
-        vocabulary.
-    token_ids
-        A tensor of token ids on which the sequence distribution is conditioned, of
-        shape ``(n_seqs, n_prompt_tokens)``
-    sequence_weights
-        A tensor that contains the initial weights of the sequences, of shape
-        ``(n_seqs,)``
-    attention_masks
-        A tensor of tensors that represent the tokens considered at the attention
-        layer, of shape ``(n_seqs, n_prompt_tokens)``.
-    fsms
-        List of finite-state machines that drive the text generation,
-        one for each sequence in the batch.
-    fsm_states
-        The initial states of the finite-state machine for each sequence in the batch.
-
-    Yields
-    ------
-    A new sequence.
-
-    """
-    import torch
-
-    if rng is None:
-        rng = torch.Generator()
-
-    kv_cache = None
-
-    while True:
-        try:
-            logits, kv_cache = model(token_ids, attention_masks, kv_cache)
-        except IndexError:  # Exceeding the context length
-            raise ContextLengthExceededError(
-                "The input length exceeds the context length of the model."
-            )
-
-        allowed_tokens = get_allowed_tokens(fsms, fsm_states)
-        biased_logits = bias_logits(logits, allowed_tokens)
-        next_token_ids, ancestors, sequence_weights = sampler(
-            biased_logits, sequence_weights, rng
-        )
-
-        token_ids = update_token_ids(token_ids, next_token_ids, ancestors)
-        attention_masks = update_attention_masks(attention_masks, ancestors)
-        kv_cache = reorder_kv_cache(kv_cache, ancestors)
-        if len(ancestors) > 1:
-            fsms = reorder_fsms(fsms, ancestors)
-            fsm_states = reorder_fsm_states(fsm_states, ancestors)
-
-        fsm_states = get_next_fsm_states(fsms, fsm_states, next_token_ids)
-        is_finished = is_generation_finished(fsms, fsm_states)
-
-        if is_finished:
-            yield GenerationState(
-                token_ids,
-                kv_cache,
-                logits,
-                sequence_weights,
-                fsm_states,
-            )
-            return
-
-        yield GenerationState(
-            token_ids,
-            kv_cache,
-            logits,
-            sequence_weights,
-            fsm_states,
-        )
-
-
-def get_next_fsm_states(
-    fsms: List["Guide"], fsm_states: List[int], next_token_ids: "torch.Tensor"
-) -> List[int]:
-    """
-
-    Parameters
-    ----------
-    fsm
-        The finite-state machine used to monitor this batch.
-    next_token_ids
-        The tokens that were just generated.
-
-    Returns
-    -------
-    A `torch.Tensor` object that represents the next logit mask.
-
-    """
-    return [
-        fsm.get_next_state(fsm_state, int(token_id[0]))
-        for fsm, fsm_state, token_id in zip(fsms, fsm_states, next_token_ids)
-    ]
-
-
-def get_allowed_tokens(
-    fsms: List["Guide"], fsm_states: List[int]
-) -> List[Optional[Iterable[int]]]:
-    """Get the new instructions for each sequence from the finite-state machine.
-
-    Parameters
-    ----------
-    fsm
-        The finite-state machine used to monitor this batch.
-    fsm_states
-        The FSM states corresponding to each sequence in the batch.
-
-    Returns
-    -------
-    A nested list that contains the ids of the logits to keep.
-
-    """
-    return [
-        fsm.get_next_instruction(state).tokens for fsm, state in zip(fsms, fsm_states)
-    ]
-
-
-def is_generation_finished(fsms: List["Guide"], fsm_states: List[int]) -> bool:
-    """Determine if the generation is finished.
-
-    A generation is considered finished if the FSM of every sequence in the
-    batch is in a final state.
-
-    A better solution is to return finished sequences as soon as their FSM
-    is in a final state.
-
-    Parameters
-    ----------
-    fsm
-        The finite-state machine used to monitor this batch.
-    fsm_states
-        The FSM states corresponding to each sequence in the batch.
-
-    Returns
-    -------
-    Whether all sequences are finished sampling.
-
-    """
-    return all([fsm.is_final_state(state) for fsm, state in zip(fsms, fsm_states)])
-
-
-def update_token_ids(
-    token_ids: "torch.Tensor", next_token_ids: "torch.Tensor", ancestors: "torch.Tensor"
-) -> "torch.Tensor":
-    """Append the sampled tokens to the running sequence of tokens.
-
-    Parameters
-    ----------
-    token_ids
-        The current token sequences
-    next_token_ids
-        The tokens that were just generated and that we need to append
-        to the existing sequences.
-    ancestors
-        The sequences to which the token ids need to be added.
-
-    Returns
-    -------
-    A new sequence of token ids that contains the tokens that were
-    just generated.
-
-    """
-    import torch
-
-    token_ids = torch.index_select(token_ids, 0, ancestors)
-    return torch.concatenate([token_ids, next_token_ids], dim=-1)
-
-
-def update_attention_masks(
-    attention_masks: "torch.Tensor", ancestors: "torch.Tensor"
-) -> "torch.Tensor":
-    """Expand the attention masks.
-
-    Parameters
-    ----------
-    attention_masks
-        The attention masks for each sequence in the batch.
-    ancestors
-        The sequences to which the token ids need to be added.
-
-    Returns
-    -------
-    The attention masks padded with 1s.
-
-    """
-    import torch
-
-    attention_masks = torch.index_select(attention_masks, 0, ancestors)
-    return torch.concatenate(
-        [
-            attention_masks,
-            torch.ones(
-                attention_masks.shape[:-1] + (1,), device=attention_masks.device
-            ),
-        ],
-        axis=-1,
-    )
-
-
-def reorder_fsms(fsms: List["Guide"], ancestors: "torch.Tensor") -> List["Guide"]:
-    reordered_fsms = []
-    for ancestor in ancestors:
-        reordered_fsms.append(fsms[ancestor].copy())
-
-    return reordered_fsms
-
-
-def reorder_fsm_states(fsm_states: List[int], ancestors: "torch.Tensor") -> List[int]:
-    reordered_states = []
-    for ancestor in ancestors:
-        reordered_states.append(fsm_states[ancestor])
-
-    return reordered_states
-
-
-def reorder_kv_cache(
-    kv_cache: Optional[Tuple], ancestors: "torch.Tensor"
-) -> Optional[Tuple]:
-    """Re-order the KV-cache based on the ancestors.
-
-    In transformers, the object that stores the KV-cache is a tuple who elements
-    are the key cache and the value cache. Each of these caches are tuples where
-    each element correpond to a layer. To each layer corresponds a tensor whose
-    first dimension is the batch size.
-
-    """
-    import torch
-
-    if kv_cache is None:
-        return None
-
-    new_kv_cache: Tuple = tuple()
-    for cache_item in kv_cache:
-        new_cache_item: Tuple = tuple()
-        for layer in cache_item:
-            layer = torch.index_select(layer, 0, ancestors.to(layer.device))
-            new_cache_item += (layer,)
-        new_kv_cache += (new_cache_item,)
-
-    return new_kv_cache
-
-
-def bias_logits(logits: "torch.Tensor", allowed_token_ids: List) -> "torch.Tensor":
-    """Mask the logits.
-
-    The function iterates over a nested list where each list corresponds to the
-    indices that need to be masked for each row in the array.
-
-    Parameters
-    ----------
-    logits
-        Two dimensional tensor that contains the next-token probability
-        distribution.
-    allowed_token_ids
-        A list that contains the tokens that can be generated by the model.
-
-    Returns
-    -------
-    A view of the original logits tensor where some values are masked.
-
-    """
-    import torch
-
-    biased_logits = torch.full_like(logits, -math.inf, device=logits.device)
-    for i, ids in enumerate(allowed_token_ids):
-        if ids is not None:
-            biased_logits[i, ids] = logits[i, ids]
-        else:
-            biased_logits[i] = logits[i]
-    return biased_logits
diff --git a/outlines/generate/json.py b/outlines/generate/json.py
deleted file mode 100644
index ab05df1f4..000000000
--- a/outlines/generate/json.py
+++ /dev/null
@@ -1,123 +0,0 @@
-import json as pyjson
-from enum import Enum
-from functools import singledispatch
-from typing import Callable, Optional, Union
-
-from genson import SchemaBuilder
-from outlines_core.fsm.json_schema import build_regex_from_schema
-from pydantic import BaseModel
-
-from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
-from outlines.generate.api import SequenceGeneratorAdapter
-from outlines.models import OpenAI
-from outlines.samplers import Sampler, multinomial
-
-from .regex import regex
-
-
-@singledispatch
-def json(
-    model,
-    schema_object: Union[str, object, Callable],
-    sampler: Sampler = multinomial(),
-    whitespace_pattern: Optional[str] = None,
-) -> SequenceGeneratorAdapter:
-    """
-    Generate structured JSON data with a `Transformer` model based on a specified JSON Schema.
-
-    Parameters
-    ----------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    schema_object:
-        The JSON Schema to generate data for. Can be a JSON string, a Pydantic model, or a callable
-        that returns a JSON schema.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-    whitespace_pattern
-        Pattern to use for JSON syntactic whitespace (doesn't impact string literals)
-        Example: allow only a single space or newline with `whitespace_pattern=r"[\n ]?"`
-
-    Returns
-    -------
-    A `SequenceGenerator` instance that generates text constrained by the schema_object and
-    transforms the result if BaseModel is used.
-
-    """
-    if isinstance(schema_object, type(BaseModel)):
-        schema = pyjson.dumps(schema_object.model_json_schema())
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: schema_object.parse_raw(x)
-    elif isinstance(schema_object, type(Enum)):
-        schema = pyjson.dumps(get_schema_from_enum(schema_object))
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    elif isinstance(schema_object, SchemaBuilder):
-        schema = schema_object.to_json()
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    elif callable(schema_object):
-        schema = pyjson.dumps(get_schema_from_signature(schema_object))
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    elif isinstance(schema_object, str):
-        schema = schema_object
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    else:
-        raise ValueError(
-            f"Cannot parse schema {schema_object}. The schema must be either "
-            + "a Pydantic object, a function or a string that contains the JSON "
-            + "Schema specification"
-        )
-
-    return generator
-
-
-@json.register(OpenAI)
-def json_openai(
-    model, schema_object: Union[str, object], sampler: Sampler = multinomial()
-):
-    if not isinstance(sampler, multinomial):
-        raise NotImplementedError(
-            r"The OpenAI API does not support any other sampling algorithm "
-            + "than the multinomial sampler."
-        )
-
-    if isinstance(schema_object, type(BaseModel)):
-        schema = schema_object.model_json_schema()
-        schema["additionalProperties"] = False
-        schema = pyjson.dumps(schema)
-        format_sequence = lambda x: schema_object.parse_raw(x)
-    elif isinstance(schema_object, str):
-        schema = schema_object
-        format_sequence = lambda x: pyjson.loads(x)
-    else:
-        raise ValueError(
-            f"Cannot parse schema {schema_object}. The schema must be either "
-            + "a Pydantic object, a function or a string that contains the JSON "
-            + "Schema specification"
-        )
-
-    # create copied, patched model with normalized json schema set
-    generator = model.new_with_replacements(
-        response_format={
-            "type": "json_schema",
-            "json_schema": {
-                "name": "default",
-                "strict": True,
-                "schema": pyjson.loads(schema),
-            },
-        }
-    )
-
-    generator.format_sequence = format_sequence
-
-    return generator
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
deleted file mode 100644
index 326701f4b..000000000
--- a/outlines/generate/regex.py
+++ /dev/null
@@ -1,66 +0,0 @@
-from functools import singledispatch
-
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import OpenAI, TransformersVision
-from outlines.samplers import Sampler, multinomial
-from outlines.types import Regex
-
-
-@singledispatch
-def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
-    """Generate structured text in the language of a regular expression.
-
-    Parameters
-    ----------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    regex_str:
-        The regular expression that the output must follow.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGeneratorAdapter` instance that generates text constrained by the
-    regular expression.
-
-    """
-    from outlines.processors import RegexLogitsProcessor
-
-    if isinstance(regex_str, Regex):
-        regex_str = regex_str.pattern
-
-    logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@regex.register(TransformersVision)
-def regex_vision(
-    model,
-    regex_str: str | Regex,
-    sampler: Sampler = multinomial(),
-):
-    from outlines.processors import RegexLogitsProcessor
-
-    if isinstance(regex_str, Regex):
-        regex_str = regex_str.pattern
-
-    logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
-    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@regex.register(OpenAI)
-def regex_openai(
-    model: OpenAI,
-    regex_str: str,
-    sampler: Sampler = multinomial(),
-):
-    raise NotImplementedError(
-        "Cannot use regex-structured generation with an OpenAI model"
-        + "due to the limitations of the OpenAI API."
-    )
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
deleted file mode 100644
index 32530d0c4..000000000
--- a/outlines/generate/text.py
+++ /dev/null
@@ -1,50 +0,0 @@
-from functools import singledispatch
-
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import OpenAI, TransformersVision
-from outlines.samplers import Sampler, multinomial
-
-
-@singledispatch
-def text(model, sampler: Sampler = multinomial()) -> SequenceGeneratorAdapter:
-    """Generate text with a `Transformer` model.
-
-    Note
-    ----
-    Python 3.11 allows dispatching on Union types and
-    this should greatly simplify the code.
-
-    Arguments
-    ---------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGeneratorAdapter` instance that generates text.
-
-    """
-    return SequenceGeneratorAdapter(model, None, sampler)
-
-
-@text.register(TransformersVision)
-def text_vision(model, sampler: Sampler = multinomial()):
-    return VisionSequenceGeneratorAdapter(model, None, sampler)
-
-
-@text.register(OpenAI)
-def text_openai(model: OpenAI, sampler: Sampler = multinomial()) -> OpenAI:
-    if not isinstance(sampler, multinomial):
-        raise NotImplementedError(
-            r"The OpenAI API does not support any other sampling algorithm "
-            + "than the multinomial sampler."
-        )
-
-    return model
diff --git a/outlines/generator.py b/outlines/generator.py
new file mode 100644
index 000000000..1623fe85e
--- /dev/null
+++ b/outlines/generator.py
@@ -0,0 +1,347 @@
+"""Encapsulate a model and an output type into a reusable object."""
+
+from typing import (
+    Any,
+    AsyncIterator,
+    Iterator,
+    Optional,
+    Union,
+)
+
+from outlines.models import (
+    AsyncBlackBoxModel,
+    BlackBoxModel,
+    SteerableModel,
+)
+from outlines.models.base import AsyncModel, Model
+from outlines.processors import (
+    CFGLogitsProcessor,
+    GuideLogitsProcessor,
+    OutlinesLogitsProcessor,
+    RegexLogitsProcessor,
+)
+from outlines.processors.guide import RegexGuide
+from outlines.types import CFG, FSM
+from outlines.types.dsl import python_types_to_terms, to_regex
+
+
+class BlackBoxGenerator:
+    """Synchronous generator for which we don't control constrained
+    generation.
+
+    The output type provided is not compiled into a logits processor, but is
+    instead directly passed on to the model.
+
+    """
+    output_type: Optional[Any]
+
+    def __init__(self, model: BlackBoxModel, output_type: Optional[Any]):
+        """
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        output_type
+            The output type that will be used to constrain the generation.
+
+        """
+        self.model = model
+        self.output_type = output_type
+
+        if isinstance(self.output_type, FSM):
+            raise NotImplementedError(
+                "FSM generation is not supported for API-based models"
+            )
+
+    def __call__(self, prompt: Any, **inference_kwargs) -> Any:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate(
+            prompt, self.output_type, **inference_kwargs
+        )
+
+    def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate_stream(
+            prompt, self.output_type, **inference_kwargs
+        )
+
+
+class AsyncBlackBoxGenerator:
+    """Asynchronous generator for which we don't control constrained
+    generation.
+
+    The output type provided is not compiled into a logits processor, but is
+    instead directly passed on to the model.
+
+    """
+    output_type: Optional[Any]
+
+    def __init__(self, model: AsyncBlackBoxModel, output_type: Optional[Any]):
+        """
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        output_type
+            The output type that will be used to constrain the generation.
+
+        """
+        self.model = model
+        self.output_type = output_type
+
+        if isinstance(self.output_type, FSM):
+            raise NotImplementedError(
+                "FSM generation is not supported for API-based models"
+            )
+
+    async def __call__(self, prompt: Any, **inference_kwargs) -> Any:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return await self.model.generate(
+            prompt, self.output_type, **inference_kwargs
+        )
+
+    async def stream(
+        self, prompt: Any, **inference_kwargs
+    ) -> AsyncIterator[Any]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        async for chunk in self.model.generate_stream(  # pragma: no cover
+            prompt, self.output_type, **inference_kwargs
+        ):
+            yield chunk
+
+
+class SteerableGenerator:
+    """Represents a generator for which we control constrained generation.
+
+    The generator is responsible for building and storing the logits processor
+    (which can be quite expensive to build), and then passing it to the model
+    when the generator is called.
+
+    The argument defining constrained generation can be of 2 types associated
+    to different methods to create an instance of the generator:
+    - `output_type` (through `__init__`): an output type as defined in the
+      `outlines.types` module
+    - `processor` (through `from_processor`): an already built logits processor
+       as defined in the `outlines.processors` module
+
+    The 2 parameters are mutually exclusive.
+
+    """
+    logits_processor: Optional[OutlinesLogitsProcessor]
+
+    def __init__(self, model: SteerableModel, output_type: Optional[Any]):
+        """
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        output_type
+            The output type expressed as a Python type
+
+        """
+        self.model = model
+        if output_type is None:
+            self.logits_processor = None
+        else:
+            term = python_types_to_terms(output_type)
+            if isinstance(term, CFG):
+                cfg_string = term.definition
+                self.logits_processor = CFGLogitsProcessor(
+                    cfg_string,
+                    self.model.tokenizer,
+                    self.model.tensor_library_name,
+                )
+            elif isinstance(term, FSM):
+                guide = RegexGuide.from_interegular_fsm(
+                    term.fsm,
+                    self.model.tokenizer,
+                )
+                self.logits_processor = GuideLogitsProcessor(
+                    self.model.tokenizer, guide, self.model.tensor_library_name
+                )
+            else:
+                regex_string = to_regex(term)
+                self.logits_processor = RegexLogitsProcessor(
+                    regex_string,
+                    self.model.tokenizer,
+                    self.model.tensor_library_name,
+                )
+
+    @classmethod
+    def from_processor(
+        cls, model: SteerableModel, processor: OutlinesLogitsProcessor
+    ):
+        """Create a generator from a logits processor.
+
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        processor
+            An instance of an OutlinesLogitsProcessor.
+
+        """
+        if not isinstance(processor, OutlinesLogitsProcessor):
+            raise TypeError(
+                "The processor argument must be an instance of "
+                "OutlinesLogitsProcessor"
+            )
+        instance = cls.__new__(cls)
+        instance.model = model
+        instance.logits_processor = processor
+
+        return instance
+
+    def __call__(self, prompt: Any, **inference_kwargs) -> Any:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate(
+            prompt, self.logits_processor, **inference_kwargs
+        )
+
+    def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate_stream(
+            prompt, self.logits_processor, **inference_kwargs
+        )
+
+
+def Generator(
+    model: Union[Model, AsyncModel],
+    output_type: Optional[Any] = None,
+    *,
+    processor: Optional[OutlinesLogitsProcessor] = None,
+) -> Union[SteerableGenerator, BlackBoxGenerator, AsyncBlackBoxGenerator]:
+    """Create a generator for the given model and output parameters.
+
+    The 2 parameters output_type and processor are mutually exclusive. The
+    parameters processor is only supported for SteerableModel instances
+    (typically local models) and is intended to be only used by advanced users.
+
+    Parameters
+    ----------
+    model
+        An instance of an Outlines model.
+    output_type
+        The output type expressed as a Python type or a type defined in the
+        outlines.types.dsl module.
+    processor
+        An instance of an OutlinesLogitsProcessor.
+
+    Returns
+    -------
+    Union[SteerableGenerator, BlackBoxGenerator, AsyncBlackBoxGenerator]
+        A generator instance.
+
+    """
+    provided_output_params = sum(
+        param is not None
+        for param in [output_type, processor]
+    )
+    if provided_output_params > 1:
+        raise ValueError(
+            "At most one of output_type or processor can be provided"
+        )
+
+    if isinstance(model, SteerableModel): # type: ignore
+        if processor is not None:
+            return SteerableGenerator.from_processor(model, processor) # type: ignore
+        else:
+            return SteerableGenerator(model, output_type) # type: ignore
+    else:
+        if processor is not None:
+            raise NotImplementedError(
+                "This model does not support logits processors"
+            )
+        if isinstance(model, AsyncBlackBoxModel): # type: ignore
+            return AsyncBlackBoxGenerator(model, output_type) # type: ignore
+        elif isinstance(model, BlackBoxModel): # type: ignore
+            return BlackBoxGenerator(model, output_type) # type: ignore
+        else:
+            raise ValueError(
+                "The model argument must be an instance of "
+                "SteerableModel, BlackBoxModel or AsyncBlackBoxModel"
+            )
diff --git a/outlines/grammars.py b/outlines/grammars.py
index f0c122964..e2050148a 100644
--- a/outlines/grammars.py
+++ b/outlines/grammars.py
@@ -1,10 +1,29 @@
+"""A few common Lark grammars."""
+
 from pathlib import Path
 
 GRAMMAR_PATH = Path(__file__).parent / "grammars"
 
 
-def read_grammar(grammar_file_name, base_grammar_path=GRAMMAR_PATH):
-    """Read grammar file from default grammar path"""
+def read_grammar(
+    grammar_file_name: str,
+    base_grammar_path: Path = GRAMMAR_PATH,
+) -> str:
+    """Read grammar file from default grammar path.
+
+    Parameters
+    ----------
+    grammar_file_name
+        The name of the grammar file to read.
+    base_grammar_path
+        The path to the directory containing the grammar file.
+
+    Returns
+    -------
+    str
+        The contents of the grammar file.
+
+    """
     full_path = base_grammar_path / grammar_file_name
     with open(full_path) as file:
         return file.read()
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index fe6f861ac..d818c5db5 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -8,12 +8,39 @@
 
 from typing import Union
 
-from .exllamav2 import ExLlamaV2Model, exl2
-from .llamacpp import LlamaCpp, llamacpp
-from .mlxlm import MLXLM, mlxlm
-from .openai import OpenAI, azure_openai, openai
-from .transformers import Transformers, TransformerTokenizer, mamba, transformers
-from .transformers_vision import TransformersVision, transformers_vision
-from .vllm import VLLM, vllm
+from .anthropic import Anthropic, from_anthropic
+from .base import Model, ModelTypeAdapter
+from .dottxt import Dottxt, from_dottxt
+from .gemini import Gemini, from_gemini
+from .llamacpp import LlamaCpp, from_llamacpp
+from .mlxlm import MLXLM, from_mlxlm
+from .ollama import Ollama, from_ollama
+from .openai import OpenAI, from_openai
+from .sglang import AsyncSGLang, SGLang, from_sglang
+from .tgi import AsyncTGI, TGI, from_tgi
+from .transformers import (
+    Transformers,
+    TransformerTokenizer,
+    TransformersMultiModal,
+    from_transformers,
+)
+from .vllm_offline import VLLMOffline, from_vllm_offline
+from .vllm import AsyncVLLM, VLLM, from_vllm
 
-LogitsGenerator = Union[Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM]
+SteerableModel = Union[LlamaCpp, MLXLM, Transformers]
+BlackBoxModel = Union[
+    Anthropic,
+    Dottxt,
+    Gemini,
+    Ollama,
+    OpenAI,
+    SGLang,
+    TGI,
+    VLLM,
+    VLLMOffline,
+]
+AsyncBlackBoxModel = Union[
+    AsyncTGI,
+    AsyncSGLang,
+    AsyncVLLM,
+]
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
new file mode 100644
index 000000000..4ed3828d9
--- /dev/null
+++ b/outlines/models/anthropic.py
@@ -0,0 +1,227 @@
+"""Integration with Anthropic's API."""
+
+from typing import TYPE_CHECKING, Any, Iterator, Optional, Union
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.templates import Vision
+
+if TYPE_CHECKING:
+    from anthropic import Anthropic as AnthropicClient
+
+__all__ = ["Anthropic", "from_anthropic"]
+
+
+class AnthropicTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `Anthropic` model.
+
+    `AnthropicTypeAdapter` is responsible for preparing the arguments to
+    Anthropic's `messages.create` method: the input (prompt and possibly
+    image).
+    Anthropic does not support defining the output type, so
+    `format_output_type` is not implemented.
+
+    """
+
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
+        """Generate the `messages` argument to pass to the client.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        dict
+            The `messages` argument to pass to the client.
+
+        """
+        if isinstance(model_input, str):
+            return self.format_str_model_input(model_input)
+        elif isinstance(model_input, Vision):
+            return self.format_vision_model_input(model_input)
+        raise TypeError(
+            f"The input type {input} is not available with Anthropic. "
+            "The only available types are `str` and `Vision`."
+        )
+
+    def format_str_model_input(self, model_input: str) -> dict:
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": model_input,
+                }
+            ]
+        }
+
+    def format_vision_model_input(self, model_input: Vision) -> dict:
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": [
+                        {
+                            "type": "image",
+                            "source": {
+                                "type": "base64",
+                                "media_type": model_input.image_format,
+                                "data": model_input.image_str,
+                            },
+                        },
+                        {"type": "text", "text": model_input.prompt},
+                    ],
+                }
+            ]
+        }
+
+    def format_output_type(self, output_type):
+        """Not implemented for Anthropic."""
+        if output_type is None:
+            return {}
+        else:
+            raise NotImplementedError(
+                f"The output type {output_type} is not available with "
+                "Anthropic."
+            )
+
+
+class Anthropic(Model):
+    """Thin wrapper around the `anthropic.Anthropic` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `anthropic.Anthropic` client.
+
+    """
+    def __init__(
+        self, client: "AnthropicClient", model_name: Optional[str] = None
+    ):
+        """
+        Parameters
+        ----------
+        client
+            An `anthropic.Anthropic` client.
+        model_name
+            The name of the model to use.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = AnthropicTypeAdapter()
+
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
+        """Generate text using Anthropic.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            As structured generation is not supported by Anthropic, the value
+            of this argument must be `None`. Otherwise, an error will be
+            raised at runtime.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The response generated by the model.
+
+        """
+        messages = self.type_adapter.format_input(model_input)
+
+        if output_type is not None:
+            raise NotImplementedError(
+                f"The type {output_type} is not available with Anthropic."
+            )
+
+        if (
+            "model" not in inference_kwargs
+            and self.model_name is not None
+        ):
+            inference_kwargs["model"] = self.model_name
+
+        completion = self.client.messages.create(
+            **messages,
+            **inference_kwargs,
+        )
+        return completion.content[0].text
+
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using Anthropic.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            As structured generation is not supported by Anthropic, the value
+            of this argument must be `None`. Otherwise, an error will be
+            raised at runtime.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        messages = self.type_adapter.format_input(model_input)
+
+        if output_type is not None:
+            raise NotImplementedError(
+                f"The type {output_type} is not available with Anthropic."
+            )
+
+        if (
+            "model" not in inference_kwargs
+            and self.model_name is not None
+        ):
+            inference_kwargs["model"] = self.model_name
+
+        stream = self.client.messages.create(
+            **messages,
+            stream=True,
+            **inference_kwargs,
+        )
+
+        for chunk in stream:
+            if (
+                chunk.type == "content_block_delta"
+                and chunk.delta.type == "text_delta"
+            ):
+                yield chunk.delta.text
+
+
+def from_anthropic(
+    client: "AnthropicClient", model_name: Optional[str] = None
+) -> Anthropic:
+    """Create an Outlines `Anthropic` model instance from an
+    `anthropic.Anthropic` client instance.
+
+    Parameters
+    ----------
+    client
+        An `anthropic.Anthropic` client instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Anthropic
+        An Outlines `Anthropic` model instance.
+
+    """
+    return Anthropic(client, model_name)
diff --git a/outlines/models/base.py b/outlines/models/base.py
new file mode 100644
index 000000000..08beece18
--- /dev/null
+++ b/outlines/models/base.py
@@ -0,0 +1,377 @@
+"""Base classes for all models and model type adapters."""
+
+from abc import ABC, abstractmethod
+from typing import Any, AsyncIterator, Iterator, Optional
+
+
+class ModelTypeAdapter(ABC):
+    """Base class for all model type adapters.
+
+    A type adapter instance must be given as a value to the `type_adapter`
+    attribute when instantiating a model.
+    The type adapter is responsible for formatting the input and output types
+    passed to the model to match the specific format expected by the
+    associated model.
+
+    """
+
+    @abstractmethod
+    def format_input(self, model_input: Any) -> Any:
+        """Format the user input to the expected format of the model.
+
+        For API-based models, it typically means creating the `messages`
+        argument passed to the client. For local models, it can mean casting
+        the input from str to list for instance.
+        This method is also used to validate that the input type provided by
+        the user is supported by the model.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        Any
+            The formatted input to be passed to the model.
+
+        """
+        ...
+
+    @abstractmethod
+    def format_output_type(self, output_type: Optional[Any] = None) -> Any:
+        """Format the output type to the expected format of the model.
+
+        For API-based models, this typically means creating a `response_format`
+        argument. For local models, it means formatting the logits processor to
+        create the object type expected by the model.
+
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        Any
+            The formatted output type to be passed to the model.
+
+        """
+        ...
+
+class Model(ABC):
+    """Base class for all synchronous models.
+
+    This class defines a shared `__call__` method that can be used to call the
+    model directly.
+    All models inheriting from this class must define a `type_adapter`
+    attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
+    attribute are used in the `generate` method to format the input and output
+    types received by the model.
+    Additionally, local models must define a `tensor_library_name` attribute.
+
+    """
+    type_adapter: ModelTypeAdapter
+    tensor_library_name: str
+
+    def __call__(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
+        """Call the model.
+
+        Users can call the model directly, in which case we will create a
+        generator instance with the output type provided and call it.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        generator("prompt")
+        ```
+        and
+        ```python
+        model("prompt", Foo)
+        ```
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        from outlines import Generator
+
+        return Generator(self, output_type)(model_input, **inference_kwargs)
+
+    def stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Iterator[Any]:
+        """Stream a response from the model.
+
+        Users can use the `stream` method from the model directly, in which
+        case we will create a generator instance with the output type provided
+        and then invoke its `stream` method.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        for chunk in generator("prompt"):
+            print(chunk)
+        ```
+        and
+        ```python
+        for chunk in model.stream("prompt", Foo):
+            print(chunk)
+        ```
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Iterator[Any]
+            A stream of responses from the model.
+
+        """
+        from outlines import Generator
+
+        generator = Generator(self, output_type)
+        return generator.stream(model_input, **inference_kwargs) # type: ignore
+
+    @abstractmethod
+    def generate(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
+        """Generate a response from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        ...
+
+    @abstractmethod
+    def generate_stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Iterator[Any]:
+        """Generate a stream of responses from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Iterator[Any]
+            A stream of responses from the model.
+
+        """
+        ...
+
+class AsyncModel(ABC):
+    """Base class for all asynchronous models.
+
+    This class defines a shared `__call__` method that can be used to call the
+    model directly.
+    All models inheriting from this class must define a `type_adapter`
+    attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
+    attribute are used in the `generate` method to format the input and output
+    types received by the model.
+    Additionally, local models must define a `tensor_library_name` attribute.
+
+    """
+    type_adapter: ModelTypeAdapter
+    tensor_library_name: str
+
+    async def __call__(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
+        """Call the model.
+
+        Users can call the model directly, in which case we will create a
+        generator instance with the output type provided and call it.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        await generator("prompt")
+        ```
+        and
+        ```python
+        await model("prompt", Foo)
+        ```
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        from outlines import Generator
+
+        generator = Generator(self, output_type)
+        return await generator(model_input, **inference_kwargs)
+
+    async def stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> AsyncIterator[Any]:
+        """Stream a response from the model.
+
+        Users can use the `stream` method from the model directly, in which
+        case we will create a generator instance with the output type provided
+        and then invoke its `stream` method.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        async for chunk in generator("prompt"):
+            print(chunk)
+        ```
+        and
+        ```python
+        async for chunk in model.stream("prompt", Foo):
+            print(chunk)
+        ```
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        AsyncIterator[Any]
+            A stream of responses from the model.
+
+        """
+        from outlines import Generator
+
+        generator = Generator(self, output_type)
+
+        async for chunk in generator.stream(model_input, **inference_kwargs):  # type: ignore
+            yield chunk
+
+    @abstractmethod
+    async def generate(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
+        """Generate a response from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        ...
+
+    @abstractmethod
+    async def generate_stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> AsyncIterator[Any]:
+        """Generate a stream of responses from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        AsyncIterator[Any]
+            A coroutine that will produce an async iterator of responses from the model.
+
+        """
+        ...
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
new file mode 100644
index 000000000..62c40a7c5
--- /dev/null
+++ b/outlines/models/dottxt.py
@@ -0,0 +1,215 @@
+"""Integration with Dottxt's API."""
+
+import json
+from typing import TYPE_CHECKING, Any, Optional
+
+from pydantic import TypeAdapter
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.types import CFG, JsonSchema, Regex
+from outlines.types.utils import (
+    is_dataclass,
+    is_genson_schema_builder,
+    is_pydantic_model,
+    is_typed_dict,
+)
+
+if TYPE_CHECKING:
+    from dottxt import Dottxt as DottxtClient
+
+__all__ = ["Dottxt", "from_dottxt"]
+
+
+class DottxtTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `Dottxt` model."""
+
+    def format_input(self, model_input: str) -> str:
+        """Format the prompt to pass to the client.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        str
+            The input to pass to the client.
+
+        """
+        if isinstance(model_input, str):
+            return model_input
+        raise TypeError(
+            f"The input type {model_input} is not available with Dottxt. "
+            "The only available type is `str`."
+        )
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> str:
+        """Format the output type to pass to the client.
+
+        TODO: `int`, `float` and other Python types could be supported via
+        JSON Schema.
+
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        str
+            The output type to pass to the client.
+
+        """
+        # Unsupported languages
+        if output_type is None:
+            raise TypeError(
+                "You must provide an output type. Dottxt only supports "
+                "constrained generation."
+            )
+        elif isinstance(output_type, Regex):
+            raise TypeError(
+                "Regex-based structured outputs will soon be available with "
+                "Dottxt. Use an open source model in the meantime."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs will soon be available with "
+                "Dottxt. Use an open source model in the meantime."
+            )
+
+        elif isinstance(output_type, JsonSchema):
+            return output_type.schema
+        elif is_dataclass(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return json.dumps(schema)
+        elif is_typed_dict(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return json.dumps(schema)
+        elif is_pydantic_model(output_type):
+            schema = output_type.model_json_schema()
+            return json.dumps(schema)
+        elif is_genson_schema_builder(output_type):
+            return output_type.to_json()
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not supported by Dottxt. "
+                "Consider using a local mode instead."
+            )
+
+
+class Dottxt(Model):
+    """Thin wrapper around the `dottxt.client.Dottxt` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `dottxt.client.Dottxt` client.
+
+    """
+
+    def __init__(
+        self,
+        client: "DottxtClient",
+        model_name: Optional[str] = None,
+        model_revision: Optional[str] = None,
+    ):
+        """
+        Parameters
+        ----------
+        client
+            A `dottxt.Dottxt` client.
+        model_name
+            The name of the model to use.
+        model_revision
+            The revision of the model to use.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.model_revision = model_revision
+        self.type_adapter = DottxtTypeAdapter()
+
+    def generate(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
+        """Generate text using Dottxt.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
+
+        """
+        prompt = self.type_adapter.format_input(model_input)
+        json_schema = self.type_adapter.format_output_type(output_type)
+
+        if (
+            "model_name" not in inference_kwargs
+            and self.model_name is not None
+        ):
+            inference_kwargs["model_name"] = self.model_name
+
+        if (
+            "model_revision" not in inference_kwargs
+            and self.model_revision is not None
+        ):
+            inference_kwargs["model_revision"] = self.model_revision
+
+        completion = self.client.json(
+            prompt,
+            json_schema,
+            **inference_kwargs,
+        )
+        return completion.data
+
+    def generate_stream(
+        self,
+        model_input,
+        output_type=None,
+        **inference_kwargs,
+    ):
+        """Not available for Dottxt."""
+        raise NotImplementedError(
+            "Dottxt does not support streaming. Call the model/generator for "
+            + "regular generation instead."
+        )
+
+
+def from_dottxt(
+    client: "DottxtClient",
+    model_name: Optional[str] = None,
+    model_revision: Optional[str] = None,
+) -> Dottxt:
+    """Create an Outlines `Dottxt` model instance from a `dottxt.Dottxt`
+    client instance.
+
+    Parameters
+    ----------
+    client
+        A `dottxt.Dottxt` client instance.
+    model_name
+        The name of the model to use.
+    model_revision
+        The revision of the model to use.
+
+    Returns
+    -------
+    Dottxt
+        An Outlines `Dottxt` model instance.
+
+    """
+    return Dottxt(client, model_name, model_revision)
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
new file mode 100644
index 000000000..c66c984af
--- /dev/null
+++ b/outlines/models/gemini.py
@@ -0,0 +1,296 @@
+"""Integration with Gemini's API."""
+
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Iterator,
+    Optional,
+    Union,
+    get_args,
+)
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.templates import Vision
+from outlines.types import CFG, JsonSchema, Regex
+from outlines.types.utils import (
+    is_dataclass,
+    is_enum,
+    get_enum_from_literal,
+    is_genson_schema_builder,
+    is_literal,
+    is_pydantic_model,
+    is_typed_dict,
+    is_typing_list,
+)
+
+if TYPE_CHECKING:
+    from google.genai import Client
+
+__all__ = ["Gemini", "from_gemini"]
+
+
+class GeminiTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `Gemini` model.
+
+    `GeminiTypeAdapter` is responsible for preparing the arguments to Gemini's
+    client `models.generate_content` method: the input (prompt and possibly
+    image), as well as the output type (either JSON or multiple choice).
+
+    """
+
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
+        """Generate the `contents` argument to pass to the client.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        dict
+            The `contents` argument to pass to the client.
+
+        """
+        if isinstance(model_input, str):
+            return {"contents": [model_input]}
+        elif isinstance(model_input, Vision):
+            from google.genai import types
+
+            image_part = types.Part.from_bytes(
+                data=model_input.image_str,
+                mime_type=model_input.image_format
+            ),
+            return {"contents": [model_input.prompt, image_part]}
+        else:
+            raise TypeError(
+                f"The input type {input} is not available with Gemini. "
+                "The only available types are `str` and `Vision`."
+            )
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the `generation_config` argument to pass to the client.
+
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        dict
+            The `generation_config` argument to pass to the client.
+
+        """
+
+        # Unsupported output pytes
+        if isinstance(output_type, Regex):
+            raise TypeError(
+                "Neither regex-based structured outputs nor the `pattern` "
+                "keyword in Json Schema are available with Gemini. Use an "
+                "open source model or dottxt instead."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs are not available with Gemini. "
+                "Use an open source model or dottxt instead."
+            )
+        elif is_genson_schema_builder(output_type):
+            raise TypeError(
+                "The Gemini SDK does not accept Genson schema builders as an "
+                "input. Pass a Pydantic model, typed dict or dataclass "
+                "instead."
+            )
+        elif isinstance(output_type, JsonSchema):
+            raise TypeError(
+                "The Gemini SDK does not accept Json Schemas as an input. "
+                "Pass a Pydantic model, typed dict or dataclass instead."
+            )
+
+        if output_type is None:
+            return {}
+
+        # Structured types
+        elif is_dataclass(output_type):
+            return self.format_json_output_type(output_type)
+        elif is_typed_dict(output_type):
+            return self.format_json_output_type(output_type)
+        elif is_pydantic_model(output_type):
+            return self.format_json_output_type(output_type)
+
+        # List of structured types
+        elif is_typing_list(output_type):
+            return self.format_list_output_type(output_type)
+
+        # Multiple choice types
+        elif is_enum(output_type):
+            return self.format_enum_output_type(output_type)
+        elif is_literal(output_type):
+            enum = get_enum_from_literal(output_type)
+            return self.format_enum_output_type(enum)
+
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not supported by Gemini. "
+                "Consider using a local model or dottxt instead."
+            )
+
+    def format_enum_output_type(self, output_type: Optional[Any]) -> dict:
+        return {
+            "response_mime_type": "text/x.enum",
+            "response_schema": output_type,
+        }
+
+    def format_json_output_type(self, output_type: Optional[Any]) -> dict:
+        return {
+            "response_mime_type": "application/json",
+            "response_schema": output_type,
+        }
+
+    def format_list_output_type(self, output_type: Optional[Any]) -> dict:
+        args = get_args(output_type)
+
+        if len(args) == 1:
+            item_type = args[0]
+
+            # Check if list item type is supported
+            if (
+                is_pydantic_model(item_type)
+                or is_typed_dict(item_type)
+                or is_dataclass(item_type)
+            ):
+                return {
+                    "response_mime_type": "application/json",
+                    "response_schema": output_type,
+                }
+
+            else:
+                raise TypeError(
+                    "The only supported types for list items are Pydantic "
+                    + "models, typed dicts and dataclasses."
+                )
+
+        raise TypeError(
+            f"Gemini only supports homogenous lists: "
+            "list[BaseModel], list[TypedDict] or list[dataclass]. "
+            f"Got {output_type} instead."
+        )
+
+
+class Gemini(Model):
+    """Thin wrapper around the `google.genai.Client` client.
+
+    This wrapper is used to convert the input and output types specified by
+    the users at a higher level to arguments to the `google.genai.Client`
+    client.
+
+    """
+
+    def __init__(self, client: "Client", model_name: Optional[str] = None):
+        """
+        Parameters
+        ----------
+        client
+            A `google.genai.Client` instance.
+        model_name
+            The name of the model to use.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = GeminiTypeAdapter()
+
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs,
+    ) -> str:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema, a list of such types, or a multiple choice type.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The response generated by the model.
+
+        """
+        contents = self.type_adapter.format_input(model_input)
+        generation_config = self.type_adapter.format_output_type(output_type)
+
+        completion = self.client.models.generate_content(
+            **contents,
+            model=inference_kwargs.pop("model", self.model_name),
+            config={**generation_config, **inference_kwargs}
+        )
+
+        return completion.text
+
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs,
+    ) -> Iterator[str]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema, a list of such types, or a multiple choice type.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        contents = self.type_adapter.format_input(model_input)
+        generation_config = self.type_adapter.format_output_type(output_type)
+
+        stream = self.client.models.generate_content_stream(
+            **contents,
+            model=inference_kwargs.pop("model", self.model_name),
+            config={**generation_config, **inference_kwargs},
+        )
+
+        for chunk in stream:
+            if hasattr(chunk, "text") and chunk.text:
+                yield chunk.text
+
+
+def from_gemini(client: "Client", model_name: Optional[str] = None) -> Gemini:
+    """Create an Outlines `Gemini` model instance from a
+    `google.genai.Client` instance.
+
+    Parameters
+    ----------
+    client
+        A `google.genai.Client` instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Gemini
+        An Outlines `Gemini` model instance.
+
+    """
+    return Gemini(client, model_name)
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 7c7ad64df..92d5fe15f 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,26 +1,28 @@
-import dataclasses
-import pickle
+"""Integration with the `llama-cpp-python` library."""
+
 import warnings
+from functools import singledispatchmethod
 from typing import (
     TYPE_CHECKING,
+    Any,
     Dict,
     Iterator,
     List,
     Optional,
     Set,
     Tuple,
-    TypedDict,
     Union,
 )
 
-from typing_extensions import Unpack
-
-from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
+from outlines.processors import CFGLogitsProcessor, OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     from llama_cpp import Llama, LogitsProcessorList
 
+__all__ = ["LlamaCpp", "from_llamacpp"]
+
 
 class LlamaCppTokenizer(Tokenizer):
     def __init__(self, model: "Llama"):
@@ -33,7 +35,8 @@ def __init__(self, model: "Llama"):
 
         self.tokenizer = model.tokenizer()
 
-        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
+        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613
+        # is resolved
         self._hf_tokenizer = None
         try:
             self.vocabulary = model.tokenizer_.hf_tokenizer.get_vocab()
@@ -47,7 +50,8 @@ def __init__(self, model: "Llama"):
         # ensure stable ordering of vocabulary
         self.vocabulary = {
             tok: tok_id
-            for tok, tok_id in sorted(self.vocabulary.items(), key=lambda x: x[1])
+            for tok, tok_id
+            in sorted(self.vocabulary.items(), key=lambda x: x[1])
         }
 
         self._hash = None
@@ -57,14 +61,19 @@ def decode(self, token_ids: List[int]) -> List[str]:
         return [decoded_bytes.decode("utf-8", errors="ignore")]
 
     def encode(
-        self, prompt: Union[str, List[str]], add_bos: bool = True, special: bool = True
+        self,
+        prompt: Union[str, List[str]],
+        add_bos: bool = True,
+        special: bool = True,
     ) -> Tuple[List[int], List[int]]:
         if isinstance(prompt, list):
             raise NotImplementedError(
                 "llama-cpp-python tokenizer doesn't support batch tokenization"
             )
         token_ids = self.tokenizer.tokenize(
-            prompt.encode("utf-8", errors="ignore"), add_bos=add_bos, special=special
+            prompt.encode("utf-8", errors="ignore"),
+            add_bos=add_bos,
+            special=special,
         )
         # generate attention mask, missing from llama-cpp-python
         attention_mask = [
@@ -77,7 +86,10 @@ def convert_token_to_string(self, token: str) -> str:
             from transformers.file_utils import SPIECE_UNDERLINE
 
             token_str = self._hf_tokenizer.convert_tokens_to_string([token])
-            if token.startswith(SPIECE_UNDERLINE) or token == "<0x20>":
+            if (
+                token.startswith(SPIECE_UNDERLINE)
+                or token == "<0x20>"
+            ):  # pragma: no cover
                 token_str = " " + token_str
             return token_str
         else:
@@ -89,8 +101,15 @@ def __eq__(self, other):
         return self.__getstate__() == other.__getstate__()
 
     def __hash__(self):
+        # We create a custom hash as pickle.dumps(self) is not stable
         if self._hash is None:
-            self._hash = hash(pickle.dumps(self))
+            self._hash = hash((
+                tuple(sorted(self.vocabulary.items())),
+                self.eos_token_id,
+                self.eos_token,
+                self.pad_token_id,
+                tuple(sorted(self.special_tokens)),
+            ))
         return self._hash
 
     def __getstate__(self):
@@ -107,241 +126,160 @@ def __setstate__(self, state):
         raise NotImplementedError("Cannot load a pickled llamacpp tokenizer")
 
 
-class LlamaCppParams(TypedDict, total=False):
-    suffix: Optional[str]
-    temperature: float
-    top_p: float
-    min_p: float
-    typical_p: float
-    seed: int
-    max_tokens: int
-    logits_processor: "LogitsProcessorList"
-    stop: Optional[Union[str, List[str]]]
-    frequence_penalty: float
-    presence_penalty: float
-    repeat_penalty: float
-    top_k: int
-    tfs_z: float
-    mirostat_mode: int
-    mirostat_tau: float
-    mirostat_eta: float
-    stream: bool
-
-
-class LlamaCpp:
-    """Represents a model provided by the `llama-cpp-python` library.
-
-    We wrap models from model providing libraries in order to give all of
-    them the same interface in Outlines and allow users to easily switch
-    between providers. This class wraps the `llama_cpp.Llama` class from the
-    `llama-cpp-python` library.
+class LlamaCppTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `LlamaCpp` model.
+
+    `LlamaCppTypeAdapter` is responsible for preparing the arguments to
+    `llama-cpp-python`'s `Llama.__call__` method: the input (a string prompt),
+    as well as the logits processor (an instance of `LogitsProcessorList`).
 
     """
 
-    def __init__(self, model: "Llama"):
-        self.model = model
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
 
-    @property
-    def tokenizer(self):
-        return LlamaCppTokenizer(self.model)
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
 
-    def prepare_generation_parameters(
-        self,
-        generation_parameters: GenerationParameters,
-        sampling_parameters: SamplingParameters,
-        structure_logits_processor,
-        **llama_cpp_params: Unpack[LlamaCppParams],
-    ):
-        """Prepare the generation parameters.
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
 
-        `llama-cpp-python` uses different default values
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. "
+            "The `llama-cpp-python` library does not support batch inference. "
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str) -> str:
+        return model_input
+
+    def format_output_type(
+        self, output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> "LogitsProcessorList":
+        """Generate the logits processor argument to pass to the model.
+
+        Parameters
+        ----------
+        output_type
+            The logits processor provided.
+
+        Returns
+        -------
+        LogitsProcessorList
+            The logits processor to pass to the model.
 
         """
         from llama_cpp import LogitsProcessorList
 
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-
-        # We update `llama_cpp_params` with the values the user passed to the
-        # generator.
-        if "stop" not in llama_cpp_params:
-            llama_cpp_params["stop"] = stop_at
-        if "seed" not in llama_cpp_params:
-            llama_cpp_params["seed"] = seed
-
-        # Somehow `llama-cpp-python` generates `max_tokens + 1`  tokens
-        if "max_tokens" not in llama_cpp_params:
-            if max_tokens is None:
-                llama_cpp_params["max_tokens"] = -1  # indicates unlimited tokens
-            else:
-                llama_cpp_params["max_tokens"] = max_tokens - 1
-        else:
-            llama_cpp_params["max_tokens"] = llama_cpp_params["max_tokens"] - 1
+        return LogitsProcessorList([output_type])
 
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
-        )
 
-        # We update the `llama_cpp_params` with the sampling values that
-        # were specified by the user via the `Sampler` class, unless they
-        # are also specified in `llama_cpp_params`. We also disable other
-        # sampling methods that are enabled by default and reset the temperature
-        # value.
-        #
-        # See https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
-        # for the default values in `llama.cpp` and indications to disable the sampling modes.
-        # Mirostat sampling, tail-free sampling and all penalties are disabled by default.
-        #
-        # See https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
-        # for default values in `llama-cpp-python`
-        if sampler == "beam_search":
-            raise NotImplementedError(
-                "The `llama_cpp_python` library does not support Beam Search."
-            )
-        if num_samples != 1:
-            raise NotImplementedError(
-                "The `llama_cpp_python` library does not allow to take several samples."
-            )
-        if "top_p" not in llama_cpp_params:
-            if top_p is not None:
-                llama_cpp_params["top_p"] = top_p
-            else:
-                llama_cpp_params["top_p"] = 1.0
-
-        if "min_p" not in llama_cpp_params:
-            llama_cpp_params["min_p"] = 0.0
-
-        if "top_k" not in llama_cpp_params:
-            if top_k is not None:
-                llama_cpp_params["top_k"] = top_k
-            else:
-                llama_cpp_params["top_k"] = -1
-
-        if "temperature" not in llama_cpp_params:
-            if temperature is not None:
-                llama_cpp_params["temperature"] = temperature
-            else:
-                llama_cpp_params["temperature"] = 1.0
-
-        if "repeat_penalty" not in llama_cpp_params:
-            llama_cpp_params["repeat_penalty"] = 1.0
-
-        # The choice to stream or not should happen via the high-level API
-        llama_cpp_params["stream"] = False
-
-        if structure_logits_processor is not None:
-            if "logits_processor" in llama_cpp_params:
-                llama_cpp_params["logits_processor"].append(structure_logits_processor)
-            else:
-                llama_cpp_params["logits_processor"] = LogitsProcessorList(
-                    [structure_logits_processor]
-                )
-
-        return llama_cpp_params
+class LlamaCpp(Model):
+    """Thin wrapper around the `llama_cpp.Llama` model.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `llama_cpp.Llama` model.
+    """
+
+    tensor_library_name = "numpy"
+
+    def __init__(self, model: "Llama"):
+        """
+        Parameters
+        ----------
+        model
+            A `llama_cpp.Llama` model instance.
+
+        """
+        self.model = model
+        self.tokenizer = LlamaCppTokenizer(self.model)
+        self.type_adapter = LlamaCppTypeAdapter()
 
     def generate(
         self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **llama_cpp_params: Unpack[LlamaCppParams],
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
     ) -> str:
         """Generate text using `llama-cpp-python`.
 
         Parameters
         ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-        llama_cpp_params
-            Keyword arguments that can be passed to
-            `llama_cpp_python.Llama.__call__`.  The values in `llama_cpp_params`
-            supersede the values of the parameters in `generation_parameters` and
-            `sampling_parameters`.  See the `llama_cpp_python` documentation for
-            a list of possible values: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        **inference_kwargs
+            Additional keyword arguments to pass to the `Llama.__call__`
+            method of the `llama-cpp-python` library.
 
         Returns
         -------
-        The generated text.
+        str
+            The text generated by the model.
 
         """
-        if not isinstance(prompts, str):
+        if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
-                "The `llama-cpp-python` library does not support batch inference."
+                "CFG generation is not supported for LlamaCpp due to bug in "
+                "the llama_cpp tokenizer"
             )
 
-        llama_cpp_params = self.prepare_generation_parameters(
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
-            **llama_cpp_params,
+        completion = self.model(
+            self.type_adapter.format_input(model_input),
+            logits_processor=self.type_adapter.format_output_type(output_type),
+            **inference_kwargs,
         )
-        completion = self.model(prompts, **llama_cpp_params)
         result = completion["choices"][0]["text"]
 
         self.model.reset()
 
         return result
 
-    def stream(
+    def generate_stream(
         self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **llama_cpp_params: Unpack[LlamaCppParams],
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
     ) -> Iterator[str]:
         """Stream text using `llama-cpp-python`.
 
         Parameters
         ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-        llama_cpp_params
-            Keyword arguments that can be passed to
-            `llama_cpp_python.Llama.__call__`.  The values in `llama_cpp_params`
-            supersede the values of the parameters in `generation_parameters` and
-            `sampling_parameters`.  See the `llama_cpp_python` documentation for
-            a list of possible values: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        **inference_kwargs
+            Additional keyword arguments to pass to the `Llama.__call__`
+            method of the `llama-cpp-python` library.
 
         Returns
         -------
-        A generator that return strings.
+        Iterator[str]
+            An iterator that yields the text generated by the model.
 
         """
-
-        if not isinstance(prompts, str):
+        if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
-                "The `llama-cpp-python` library does not support batch inference."
+                "CFG generation is not supported for LlamaCpp due to bug in "
+                "the llama_cpp tokenizer"
             )
 
-        llama_cpp_params = self.prepare_generation_parameters(
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
-            **llama_cpp_params,
+        generator = self.model(
+            self.type_adapter.format_input(model_input),
+            logits_processor=self.type_adapter.format_output_type(output_type),
+            stream=True,
+            **inference_kwargs,
         )
-        llama_cpp_params["stream"] = True
-        generator = self.model(prompts, **llama_cpp_params)
 
         def token_generator() -> Iterator[str]:
             while True:
@@ -354,54 +292,37 @@ def token_generator() -> Iterator[str]:
 
         return token_generator()
 
-    def load_lora(self, adapter_path: str):
+    def load_lora(self, adapter_path: str) -> None:  # pragma: no cover
+        """Load a LoRA adapter. Deprecated since v1.0.0."""
+        warnings.warn("""
+            The `load_lora` method is deprecated starting from v1.0.0.
+            Support for it will be removed in v1.1.0.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
         if self.model._model.apply_lora_from_file(
             adapter_path,
             1.0,
         ):
-            raise RuntimeError(f"Failed to apply LoRA from lora path: {adapter_path}")
-
+            raise RuntimeError(
+                f"Failed to apply LoRA from lora path: {adapter_path}"
+            )
 
-def llamacpp(
-    repo_id: str, filename: Optional[str] = None, **llamacpp_model_params
-) -> LlamaCpp:
-    """Load a model from the `llama-cpp-python` library.
 
-    We use the `Llama.from_pretrained` classmethod that downloads models
-    directly from the HuggingFace hub, instead of asking users to specify
-    a path to the downloaded model. One can still load a local model
-    by initializing `llama_cpp.Llama` directly.
+def from_llamacpp(model: "Llama"):
+    """Create an Outlines `LlamaCpp` model instance from a
+    `llama_cpp.Llama` instance.
 
     Parameters
     ----------
-    repo_id
-        The name of the model repository.
-    filename:
-        A filename of glob pattern to match the model file in the repo.
-    llama_cpp_model_params
-        Llama-specific model parameters. See the `llama-cpp-python` documentation
-        for the full list: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__
+    model
+        A `llama_cpp.Llama` instance.
 
-    """
-    from llama_cpp import Llama
-
-    # Default to using the model's full context length
-    if "n_ctx" not in llamacpp_model_params:
-        llamacpp_model_params["n_ctx"] = 0
-
-    if "verbose" not in llamacpp_model_params:
-        llamacpp_model_params["verbose"] = False
-
-    # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
-    if "tokenizer" not in llamacpp_model_params:
-        warnings.warn(
-            "The pre-tokenizer in `llama.cpp` handles unicode improperly "
-            + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
-            + "Outlines may raise a `RuntimeError` when building the regex index.\n"
-            + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
-            + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
-        )
-
-    model = Llama.from_pretrained(repo_id, filename, **llamacpp_model_params)
+    Returns
+    -------
+    LlamaCpp
+        An Outlines `LlamaCpp` model instance.
 
+    """
     return LlamaCpp(model)
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 843107d66..bea4ad293 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -1,247 +1,185 @@
-import dataclasses
-from typing import TYPE_CHECKING, Generator, Iterator, List, Optional, Tuple, Union
+"""Integration with the `mlx_lm` library."""
 
-from .transformers import TransformerTokenizer
+from functools import singledispatchmethod
+from typing import TYPE_CHECKING, Iterator, List, Optional
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.models.transformers import TransformerTokenizer
+from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
-    import mlx.core as mx
     import mlx.nn as nn
     from transformers import PreTrainedTokenizer
 
-    from outlines.generate.api import GenerationParameters, SamplingParameters
-    from outlines.processors import OutlinesLogitsProcessor
+__all__ = ["MLXLM", "from_mlxlm"]
 
 
-class MLXLM:
-    """
-    Represents an `mlx_lm` model
-    """
+class MLXLMTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `MLXLM` model."""
 
-    def __init__(
-        self,
-        model: "nn.Module",
-        tokenizer: "PreTrainedTokenizer",
-    ):
-        self.model = model
-        self.mlx_tokenizer = tokenizer  # returns mlx tensors, used for encode()
-        self.tokenizer = TransformerTokenizer(
-            tokenizer._tokenizer
-        )  # _tokenizer is HF Tokenizer
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
 
-    def generate(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: "GenerationParameters",
-        logits_processor,
-        sampling_parameters: "SamplingParameters",
-    ) -> str:
-        streamer = self.stream(
-            prompts, generation_parameters, logits_processor, sampling_parameters
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. "
+            "The `mlx_lm` library does not support batch inference."
         )
-        return "".join(list(streamer))
 
-    def stream(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: "GenerationParameters",
-        logits_processor,
-        sampling_parameters: "SamplingParameters",
-    ) -> Iterator[str]:
-        """Generate text using `mlx_lm`.
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        return model_input
+
+    def format_output_type(
+        self, output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Optional[List[OutlinesLogitsProcessor]]:
+        """Generate the logits processor argument to pass to the model.
 
         Parameters
         ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
+        output_type
+            The logits processor provided.
 
         Returns
         -------
-        The generated text.
+        Optional[list[OutlinesLogitsProcessor]]
+            The logits processor argument to be passed to the model.
+
         """
-        import mlx.core as mx
+        if not output_type:
+            return None
+        return [output_type]
 
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
-        )
-        if max_tokens is None:
-            max_tokens = int(1e9)
-
-        if not isinstance(prompts, str):
-            raise NotImplementedError(
-                "The `mlx-lm` library does not support batch inference."
-            )
-        if sampler == "beam_search":
-            raise NotImplementedError(
-                "The `mlx-lm` library does not support Beam Search."
-            )
-        if num_samples != 1:
-            raise NotImplementedError(
-                "The `mlx-lm` library does not allow to take several samples."
-            )
-        if top_k is not None:
-            raise NotImplementedError("The `mlx-lm` library does not support top_k.")
-        if seed is not None:
-            raise NotImplementedError("The `mlx-lm` library does not support seed.")
-        if stop_at is not None:
-            raise NotImplementedError("The `mlx-lm` library does not support stop_at.")
-
-        generate_kwargs = {
-            "temp": temperature,
-            "top_p": top_p,
-            "sampler": sampler,
-            "logits_processor": logits_processor,
-        }
-
-        # Adapted from
-        # https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L267
-        prompt_tokens = mx.array(self.mlx_tokenizer.encode(prompts))
-
-        detokenizer = self.mlx_tokenizer.detokenizer
-        detokenizer.reset()
-
-        for (token, prob), n in zip(
-            self.generate_step(prompt_tokens, **generate_kwargs),
-            range(max_tokens),
-        ):
-            if token == self.tokenizer.eos_token_id:
-                break
-            detokenizer.add_token(token)
-            yield detokenizer.last_segment
 
-        detokenizer.finalize()
-        yield detokenizer.last_segment
+class MLXLM(Model):
+    """Thin wrapper around an `mlx_lm` model.
 
-    def generate_step(
-        self,
-        prompt: "mx.array",
-        temp: Optional[float],
-        top_p: Optional[float],
-        sampler: str,
-        logits_processor: "OutlinesLogitsProcessor",
-    ) -> Generator[Tuple[int, float], None, None]:
-        """
-        Adapted from
-        https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L129
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `mlx_lm` library.
+
+    """
 
-        A generator producing token ids based on the given prompt from the model.
+    tensor_library_name = "mlx"
 
+    def __init__(
+        self,
+        model: "nn.Module",
+        tokenizer: "PreTrainedTokenizer",
+    ):
+        """
         Parameters
         ----------
-        prompt
-            The input prompt.
-        temp
-            The temperature for sampling, if 0 the argmax is used.
-        top_p
-            Nulceus sampling, higher means model considers more less likely words.
-        sampler
-            The sampler string defined by SequenceGeneratorAdapter
-        logits_processor
-            Augment logits before sampling.
-        """
-        import mlx.core as mx
-        import mlx_lm
-
-        temperature: float = temp or 1.0
+        model
+            An instance of an `mlx_lm` model.
+        tokenizer
+            An instance of an `mlx_lm` tokenizer or of a compatible
+            `transformers` tokenizer.
 
-        def sample(logits: "mx.array") -> Tuple["mx.array", float]:
-            softmax_logits = mx.softmax(logits)
-
-            if temperature == 0.0 or sampler == "greedy":
-                token = mx.argmax(logits, axis=-1)
-            elif sampler == "multinomial":
-                if top_p is not None and top_p > 0 and top_p < 1.0:
-                    token = mlx_lm.sample_utils.top_p_sampling(
-                        logits, top_p, temperature
-                    )
-                else:
-                    token = mx.random.categorical(logits * (1 / temperature))
-            else:
-                raise ValueError(f"Invalid mlx-lm sampler: `{sampler}`")
+        """
+        self.model = model
+        # self.mlx_tokenizer is used by the mlx-lm in its generate function
+        self.mlx_tokenizer = tokenizer
+        # self.tokenizer is used by the logits processor
+        self.tokenizer = TransformerTokenizer(tokenizer._tokenizer)
+        self.type_adapter = MLXLMTypeAdapter()
 
-            prob = softmax_logits[0, token]
-            return token, prob
+    def generate(
+        self,
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **kwargs,
+    ) -> str:
+        """Generate text using `mlx-lm`.
 
-        cache = mlx_lm.models.cache.make_prompt_cache(self.model)
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        kwargs
+            Additional keyword arguments to pass to the `mlx-lm` library.
 
-        # kv cache contains processed input IDs, we pass the unprocessed inputs and cache to model()
-        unprocessed_input_ids = prompt
-        generated_ids: List[int] = []
+        Returns
+        -------
+        str
+            The text generated by the model.
 
-        while True:
-            logits = self.model(unprocessed_input_ids[None], cache=cache)
-            logits = logits[:, -1, :]
+        """
+        from mlx_lm import generate
+
+        return generate(
+            self.model,
+            self.mlx_tokenizer,
+            self.type_adapter.format_input(model_input),
+            logits_processors=self.type_adapter.format_output_type(output_type),
+            **kwargs,
+        )
 
-            if logits_processor is not None:
-                # convert to logits_processor 1d expectation, apply, then convert back
-                logits_1d = logits.reshape(-1)
-                logits_1d = logits_processor(generated_ids, logits_1d)
-                logits = logits_1d.reshape(1, -1)
+    def generate_stream(
+        self,
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **kwargs,
+    ) -> Iterator[str]:
+        """Stream text using `mlx-lm`.
 
-            new_token_single, prob = sample(logits)
-            new_token = new_token_single.item()
-            yield new_token, prob
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        kwargs
+            Additional keyword arguments to pass to the `mlx-lm` library.
 
-            generated_ids.append(new_token)
-            unprocessed_input_ids = new_token_single
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
 
+        """
+        from mlx_lm import stream_generate
+
+        for gen_response in stream_generate(
+            self.model,
+            self.mlx_tokenizer,
+            self.type_adapter.format_input(model_input),
+            logits_processors=self.type_adapter.format_output_type(output_type),
+            **kwargs,
+        ):
+            yield gen_response.text
 
-def mlxlm(
-    model_name: str,
-    tokenizer_config: dict = {},
-    model_config: dict = {},
-    adapter_path: Optional[str] = None,
-    lazy: bool = False,
-):
-    """Instantiate a model from the `mlx_lm` library and its tokenizer.
 
-    Signature adapted from
-    https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L422
+def from_mlxlm(model: "nn.Module", tokenizer: "PreTrainedTokenizer") -> MLXLM:
+    """Create an Outlines `MLXLM` model instance from an `mlx_lm` model and a
+    tokenizer.
 
     Parameters
     ----------
-    Args:
-        path_or_hf_repo (Path): The path or the huggingface repository to load the model from.
-        tokenizer_config (dict, optional): Configuration parameters specifically for the tokenizer.
-            Defaults to an empty dictionary.
-        model_config(dict, optional): Configuration parameters specifically for the model.
-            Defaults to an empty dictionary.
-        adapter_path (str, optional): Path to the LoRA adapters. If provided, applies LoRA layers
-            to the model. Default: ``None``.
-        lazy (bool): If False eval the model parameters to make sure they are
-            loaded in memory before returning, otherwise they will be loaded
-            when needed. Default: ``False``
+    model
+        An instance of an `mlx_lm` model.
+    tokenizer
+        An instance of an `mlx_lm` tokenizer or of a compatible
+        transformers tokenizer.
 
     Returns
     -------
-    A `MLXLM` model instance.
+    MLXLM
+        An Outlines `MLXLM` model instance.
 
     """
-    try:
-        import mlx.core as mx
-        import mlx_lm
-    except ImportError:
-        raise ImportError(
-            "The `mlx_lm` library needs to be installed in order to use `mlx_lm` models."
-        )
-    if not mx.metal.is_available():
-        raise RuntimeError("You cannot use `mlx_lm` without Apple Silicon (Metal)")
-
-    model, tokenizer = mlx_lm.load(
-        model_name,
-        tokenizer_config=tokenizer_config,
-        model_config=model_config,
-        adapter_path=adapter_path,
-        lazy=lazy,
-    )
     return MLXLM(model, tokenizer)
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
new file mode 100644
index 000000000..862c71f69
--- /dev/null
+++ b/outlines/models/ollama.py
@@ -0,0 +1,213 @@
+"""Integration with the `ollama` library."""
+
+import json
+from typing import TYPE_CHECKING, Any, Iterator, Optional
+
+from pydantic import TypeAdapter
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.types import CFG, JsonSchema, Regex
+from outlines.types.utils import (
+    is_dataclass,
+    is_genson_schema_builder,
+    is_pydantic_model,
+    is_typed_dict,
+)
+
+if TYPE_CHECKING:
+    from ollama import Client as OllamaClient
+
+__all__ = ["Ollama", "from_ollama"]
+
+
+class OllamaTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `Ollama` model."""
+
+    def format_input(self, model_input: str) -> str:
+        """Generate the prompt argument to pass to the model.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
+
+        """
+        if isinstance(model_input, str):
+            return model_input
+        raise TypeError(
+            f"The input type {model_input} is not available. "
+            "Ollama does not support batch inference."
+        )
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> Optional[str]:
+        """Format the output type to pass to the client.
+
+        TODO: `int`, `float` and other Python types could be supported via
+        JSON Schema.
+
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        Optional[str]
+            The formatted output type to be passed to the model.
+
+        """
+        if isinstance(output_type, Regex):
+            raise TypeError(
+                "Regex-based structured outputs are not supported by Ollama. "
+                "Use an open source model in the meantime."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs are not supported by Ollama. "
+                "Use an open source model in the meantime."
+            )
+
+        if output_type is None:
+            return None
+        elif isinstance(output_type, JsonSchema):
+            return json.loads(output_type.schema)
+        elif is_dataclass(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return schema
+        elif is_typed_dict(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return schema
+        elif is_pydantic_model(output_type):
+            schema = output_type.model_json_schema()
+            return schema
+        elif is_genson_schema_builder(output_type):
+            return output_type.to_json()
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not supported by Ollama. "
+                "Consider using a local model instead."
+            )
+
+
+class Ollama(Model):
+    """Thin wrapper around the `ollama.Client` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `ollama.Client` client.
+
+    """
+
+    def __init__(
+        self,client: "OllamaClient", model_name: Optional[str] = None,
+    ):
+        """
+        Parameters
+        ----------
+        client
+            The `ollama.Client` client.
+        model_name
+            The name of the model to use.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = OllamaTypeAdapter()
+
+    def generate(self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **kwargs: Any,
+    ) -> str:
+        """Generate text using Ollama.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
+
+        """
+        if "model" not in kwargs and self.model_name is not None:
+            kwargs["model"] = self.model_name
+
+        response = self.client.generate(
+            prompt=self.type_adapter.format_input(model_input),
+            format=self.type_adapter.format_output_type(output_type),
+            **kwargs,
+        )
+        return response.response
+
+    def generate_stream(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using Ollama.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        if "model" not in kwargs and self.model_name is not None:
+            kwargs["model"] = self.model_name
+
+        response = self.client.generate(
+            prompt=self.type_adapter.format_input(model_input),
+            format=self.type_adapter.format_output_type(output_type),
+            stream=True,
+            **kwargs,
+        )
+        for chunk in response:
+            yield chunk.response
+
+
+def from_ollama(
+    client: "OllamaClient", model_name: Optional[str] = None
+) -> Ollama:
+    """Create an Outlines `Ollama` model instance from an `ollama.Client`
+    client.
+
+    Parameters
+    ----------
+    client
+        A `ollama.Client` client instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Ollama
+        An Outlines `Ollama` model instance.
+
+    """
+    return Ollama(client, model_name)
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index e46107bd4..2112e1974 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -2,15 +2,419 @@
 
 import copy
 import functools
+import json
+import warnings
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Callable,
+    Dict,
+    Iterator,
+    List,
+    Optional,
+    Union,
+)
 from dataclasses import asdict, dataclass, field, replace
-from typing import Callable, Dict, List, Optional, Tuple, Union
 
-import numpy as np
+from pydantic import BaseModel, TypeAdapter
 
-from outlines.base import vectorize
 from outlines.caching import cache
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.templates import Vision
+from outlines.types import JsonSchema, Regex, CFG
+from outlines.types.utils import (
+    is_dataclass,
+    is_typed_dict,
+    is_pydantic_model,
+    is_genson_schema_builder,
+    is_native_dict
+)
+from outlines.v0_legacy.base import vectorize
 
-__all__ = ["OpenAI", "openai", "azure_openai"]
+if TYPE_CHECKING:
+    from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
+
+__all__ = ["OpenAI", "from_openai"]
+
+
+class OpenAITypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `OpenAI` model.
+
+    `OpenAITypeAdapter` is responsible for preparing the arguments to OpenAI's
+    `completions.create` methods: the input (prompt and possibly image), as
+    well as the output type (only JSON).
+
+    """
+
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
+        """Generate the `messages` argument to pass to the client.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the client.
+
+        """
+        if isinstance(model_input, str):
+            return self.format_str_model_input(model_input)
+        elif isinstance(model_input, Vision):
+            return self.format_vision_model_input(model_input)
+        raise TypeError(
+            f"The input type {input} is not available with OpenAI. "
+            "The only available types are `str` and `Vision`."
+        )
+
+    def format_str_model_input(self, model_input: str) -> dict:
+        """Generate the `messages` argument to pass to the client when the user
+        only passes a prompt.
+
+        """
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": model_input,
+                }
+            ]
+        }
+
+    def format_vision_model_input(self, model_input: Vision) -> dict:
+        """Generate the `messages` argument to pass to the client when the user
+        passes a prompt and an image.
+
+        """
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": model_input.prompt},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:{model_input.image_format};base64,{model_input.image_str}"  # noqa: E702
+                            },
+                        },
+                    ],
+                }
+            ]
+        }
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the `response_format` argument to the client based on the
+        output type specified by the user.
+
+        TODO: `int`, `float` and other Python types could be supported via
+        JSON Schema.
+
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        dict
+            The formatted output type to be passed to the client.
+
+        """
+        # Unsupported languages
+        if isinstance(output_type, Regex):
+            raise TypeError(
+                "Neither regex-based structured outputs nor the `pattern` keyword "
+                "in Json Schema are available with OpenAI. Use an open source "
+                "model or dottxt instead."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs are not available with OpenAI. "
+                "Use an open source model or dottxt instead."
+            )
+
+        if output_type is None:
+            return {}
+        elif is_native_dict(output_type):
+            return self.format_json_mode_type()
+        elif is_dataclass(output_type):
+            output_type = TypeAdapter(output_type).json_schema()
+            return self.format_json_output_type(output_type)
+        elif is_typed_dict(output_type):
+            output_type = TypeAdapter(output_type).json_schema()
+            return self.format_json_output_type(output_type)
+        elif is_pydantic_model(output_type):
+            output_type = output_type.model_json_schema()
+            return self.format_json_output_type(output_type)
+        elif is_genson_schema_builder(output_type):
+            schema = json.loads(output_type.to_json())
+            return self.format_json_output_type(schema)
+        elif isinstance(output_type, JsonSchema):
+            return self.format_json_output_type(json.loads(output_type.schema))
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not available with OpenAI. "
+                "Use an open source model or dottxt instead."
+            )
+
+    def format_json_output_type(self, schema: dict) -> dict:
+        """Generate the `response_format` argument to the client when the user
+        specified a `Json` output type.
+
+        """
+        # OpenAI requires `additionalProperties` to be set
+        if "additionalProperties" not in schema:
+            schema["additionalProperties"] = False
+
+        return {
+            "response_format": {
+                "type": "json_schema",
+                "json_schema": {
+                    "name": "default",
+                    "strict": True,
+                    "schema": schema,
+                },
+            }
+        }
+
+    def format_json_mode_type(self) -> dict:
+        """Generate the `response_format` argument to the client when the user
+        specified the output type should be a JSON but without specifying the
+        schema (also called "JSON mode").
+
+        """
+        return {"response_format": {"type": "json_object"}}
+
+
+class OpenAI(Model):
+    """Thin wrapper around the `openai.OpenAI` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client.
+
+    """
+
+    def __init__(
+        self,
+        client: Union["OpenAIClient", "AzureOpenAIClient"],
+        model_name: Optional[Union[str, "OpenAIConfig"]] = None,
+        **kwargs
+    ):
+        """Initialize the OpenAI model.
+
+        To provide temporary backwards compatibility with Outlines v0,
+        the class can be instantiated with a `OpenAIConfig` instance as
+        a value for the `model_name` argument. This is deprecated and will
+        be removed in v1.1.0. Please provide a model name instead.
+
+        Parameters
+        ----------
+        client
+            The `openai.OpenAI` client.
+        model_name
+            The name of the model to use.
+
+        """
+
+        # legacy mode
+        if isinstance(model_name, OpenAIConfig) or kwargs.get("config"):
+            warnings.warn("""
+                The `openai` function is deprecated starting from v1.0.0.
+                Do not use it. Support for it will be removed in v1.1.0.
+                Instead, you should instantiate a `OpenAI` model with the
+                `outlines.from_openai` function that takes an openai library
+                client and a model name as arguments. Similarly, you cannot
+                instantiate a `OpenAI` model directly with a `OpenAIConfig`
+                instance anymore, but must provide a client and a model name
+                instead.
+                For example:
+                ```python
+                from openai import OpenAI as OpenAIClient
+                from outlines import from_openai
+                client = OpenAIClient()
+                model = from_openai(client, "gpt-4o")
+                ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+            )
+            config = (
+                model_name
+                if isinstance(model_name, OpenAIConfig)
+                else kwargs.pop("config")
+            )
+            self.legacy_instance = OpenAILegacy(
+                client, config, kwargs.get("system_prompt")
+            )
+        # regular mode
+        else:
+            self.client = client
+            self.model_name = model_name
+            self.type_adapter = OpenAITypeAdapter()
+
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Union[type[BaseModel], str]] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using OpenAI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema or an empty dictionary.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The text generated by the model.
+
+        """
+        import openai
+
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        try:
+            result = self.client.chat.completions.create(
+                **messages,
+                **response_format,
+                **inference_kwargs,
+            )
+        except openai.BadRequestError as e:
+            if e.body["message"].startswith("Invalid schema"):
+                raise TypeError(
+                    f"OpenAI does not support your schema: {e.body['message']}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise e
+
+        messages = [choice.message for choice in result.choices]
+        for message in messages:
+            if message.refusal is not None:
+                raise ValueError(
+                    f"OpenAI refused to answer the request: {message.refusal}"
+                )
+
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
+
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Union[type[BaseModel], str]] = None,
+        **inference_kwargs,
+    ) -> Iterator[str]:
+        """Stream text using OpenAI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema or an empty dictionary.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        import openai
+
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        stream = self.client.chat.completions.create(
+            stream=True,
+            **messages,
+            **response_format,
+            **inference_kwargs
+        )
+
+        for chunk in stream:
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+    ### Legacy !!!
+
+    def __call__(self, *args, **kwargs):
+        if hasattr(self, "legacy_instance"):
+            return self.legacy_instance(*args, **kwargs)
+        else:
+            return super().__call__(*args, **kwargs)
+
+    def stream(self, *args, **kwargs):
+        if hasattr(self, "legacy_instance"):
+            return self.legacy_instance.stream(*args, **kwargs)
+        else:
+            return super().stream(*args, **kwargs)
+
+    def new_with_replacements(self, **kwargs):
+        if hasattr(self, "legacy_instance"):
+            return self.legacy_instance.new_with_replacements(self, **kwargs)
+        raise NotImplementedError("This method is only available in legacy mode")
+
+    def __str__(self):
+        if hasattr(self, "legacy_instance"):
+            return str(self.legacy_instance)
+        else:
+            return super().__str__()
+
+    def __repr__(self):
+        if hasattr(self, "legacy_instance"):
+            return repr(self.legacy_instance)
+        else:
+            return super().__repr__()
+
+def from_openai(
+    client: Union["OpenAIClient", "AzureOpenAIClient"],
+    model_name: Optional[str] = None,
+) -> OpenAI:
+    """Create an Outlines `OpenAI` model instance from an `openai.OpenAI`
+    client.
+
+    Parameters
+    ----------
+    client
+        An `openai.OpenAI` client instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    OpenAI
+        An Outlines `OpenAI` model instance.
+
+    """
+    return OpenAI(client, model_name)
+
+
+### Legacy !!!
 
 
 @dataclass(frozen=True)
@@ -66,7 +470,7 @@ class OpenAIConfig:
     user: str = field(default_factory=str)
 
 
-class OpenAI:
+class OpenAILegacy():
     """An object that represents the OpenAI API."""
 
     def __init__(
@@ -91,6 +495,7 @@ def __init__(
 
         self.client = client
         self.config = config
+        self.system_prompt = system_prompt
 
         # We count the total number of prompt and generated tokens as returned
         # by the OpenAI API, summed over all the requests performed with this
@@ -98,7 +503,7 @@ def __init__(
         self.prompt_tokens = 0
         self.completion_tokens = 0
 
-        self.format_sequence = lambda seq: seq
+        self.format_sequence = lambda x: x
 
     def __call__(
         self,
@@ -109,7 +514,7 @@ def __call__(
         system_prompt: Optional[str] = None,
         temperature: Optional[float] = None,
         samples: Optional[int] = None,
-    ) -> np.ndarray:
+    ):
         """Call the OpenAI API to generate text.
 
         Parameters
@@ -149,7 +554,7 @@ def __call__(
         )  # type: ignore
 
         response, prompt_tokens, completion_tokens = generate_chat(
-            prompt, system_prompt, self.client, config
+            prompt, system_prompt or self.system_prompt, self.client, config
         )
         self.prompt_tokens += prompt_tokens
         self.completion_tokens += completion_tokens
@@ -161,9 +566,11 @@ def stream(self, *args, **kwargs):
             "Streaming is currently not supported for the OpenAI API"
         )
 
-    def new_with_replacements(self, **kwargs):
-        new_instance = copy.copy(self)
-        new_instance.config = replace(new_instance.config, **kwargs)
+    def new_with_replacements(self, model, **kwargs):
+        new_instance = copy.copy(model)
+        new_instance.legacy_instance.config = replace(
+            new_instance.legacy_instance.config, **kwargs
+        )
         return new_instance
 
     def __str__(self):
@@ -179,7 +586,7 @@ async def generate_chat(
     system_prompt: Union[str, None],
     client,
     config: OpenAIConfig,
-) -> Tuple[np.ndarray, int, int]:
+):
     """Call OpenAI's Chat Completion API.
 
     Parameters
@@ -200,6 +607,7 @@ async def generate_chat(
     A tuple that contains the model's response(s) and usage statistics.
 
     """
+    import numpy as np
 
     @error_handler
     @cache()
@@ -250,54 +658,3 @@ def call(*args, **kwargs):
             raise e
 
     return call
-
-
-@functools.singledispatch
-def openai(model_or_client, *args, **kwargs):
-    return OpenAI(model_or_client, *args, **kwargs)
-
-
-@openai.register(str)
-def openai_model(
-    model_name: str,
-    config: Optional[OpenAIConfig] = None,
-    **openai_client_params,
-):
-    try:
-        from openai import AsyncOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
-        )
-
-    if config is not None:
-        config = replace(config, model=model_name)  # type: ignore
-    else:
-        config = OpenAIConfig(model=model_name)
-
-    client = AsyncOpenAI(**openai_client_params)
-
-    return OpenAI(client, config)
-
-
-def azure_openai(
-    deployment_name: str,
-    model_name: Optional[str] = None,
-    config: Optional[OpenAIConfig] = None,
-    **azure_openai_client_params,
-):
-    try:
-        from openai import AsyncAzureOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
-        )
-
-    if config is not None:
-        config = replace(config, model=deployment_name)  # type: ignore
-    if config is None:
-        config = OpenAIConfig(model=deployment_name)
-
-    client = AsyncAzureOpenAI(**azure_openai_client_params)
-
-    return OpenAI(client, config)
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
new file mode 100644
index 000000000..0c2a51da1
--- /dev/null
+++ b/outlines/models/sglang.py
@@ -0,0 +1,372 @@
+"""Integration with an SGLang server."""
+
+import json
+import warnings
+from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
+
+from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
+from outlines.models.openai import OpenAITypeAdapter
+from outlines.types.dsl import (
+    CFG,
+    JsonSchema,
+    python_types_to_terms,
+    to_regex,
+)
+from outlines.templates import Vision
+
+if TYPE_CHECKING:
+    from openai import AsyncOpenAI, OpenAI
+
+__all__ = ["AsyncSGLang", "SGLang", "from_sglang"]
+
+
+class SGLangTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `SGLang` and `AsyncSGLang` models."""
+
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
+        """Generate the prompt argument to pass to the client.
+
+        We rely on the OpenAITypeAdapter to format the input as the sglang
+        server expects input in the same format as OpenAI.
+
+        Parameters
+        ----------
+        model_input
+            The input passed by the user.
+
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the client.
+
+        """
+        return OpenAITypeAdapter().format_input(model_input)
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the structured output argument to pass to the client.
+
+        Parameters
+        ----------
+        output_type
+            The structured output type provided.
+
+        Returns
+        -------
+        dict
+            The formatted output type to be passed to the client.
+
+        """
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            warnings.warn(
+                "SGLang grammar-based structured outputs expects an EBNF "
+                "grammar instead of a Lark grammar as is generally used in "
+                "Outlines. The grammar cannot be used as a structured output "
+                "type with an outlines backend, it is only compatible with "
+                "the sglang and llguidance backends."
+            )
+            return {"extra_body": {"ebnf": term.definition}}
+        elif isinstance(term, JsonSchema):
+            return OpenAITypeAdapter().format_json_output_type(
+                json.loads(term.schema)
+            )
+        else:
+            return {"extra_body": {"regex": to_regex(term)}}
+
+
+class SGLang(Model):
+    """Thin wrapper around the `openai.OpenAI` client used to communicate with
+    an SGLang server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    SGLang server.
+
+    """
+
+    def __init__(self, client, model_name: Optional[str] = None):
+        """
+        Parameters
+        ----------
+        client
+            An `openai.OpenAI` client instance.
+        model_name
+            The name of the model to use.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = SGLangTypeAdapter()
+
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using SGLang.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input,
+            output_type,
+            **inference_kwargs,
+        )
+
+        response = self.client.chat.completions.create(**client_args)
+
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The SGLang server refused to answer the request: "
+                    f"{message.refusal}"
+                )
+
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
+
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using SGLang.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = self.client.chat.completions.create(
+            **client_args, stream=True,
+        )
+
+        for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
+        """Build the arguments to pass to the SGLang client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        client_args = {
+            **messages,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+class AsyncSGLang(AsyncModel):
+    """Thin async wrapper around the `openai.OpenAI` client used to communicate
+    with an SGLang server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    SGLang server.
+
+    """
+
+    def __init__(self, client, model_name: Optional[str] = None):
+        """
+        Parameters
+        ----------
+        client
+            An `openai.AsyncOpenAI` client instance.
+        model_name
+            The name of the model to use.
+
+        Parameters
+        ----------
+        client
+            An `openai.AsyncOpenAI` client instance.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = SGLangTypeAdapter()
+
+    async def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using `sglang`.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        response = await self.client.chat.completions.create(**client_args)
+
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The sglang server refused to answer the request: "
+                    f"{message.refusal}"
+                )
+
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
+
+    async def generate_stream( # type: ignore
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> AsyncIterator[str]:
+        """Return a text generator.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        AsyncIterator[str]
+            An async iterator that yields the text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = await self.client.chat.completions.create(
+            **client_args,
+            stream=True,
+        )
+
+        async for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
+        """Build the arguments to pass to the SGLang client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        client_args = {
+            **messages,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+def from_sglang(
+    client: Union["OpenAI", "AsyncOpenAI"],
+    model_name: Optional[str] = None,
+) -> Union[SGLang, AsyncSGLang]:
+    """Create a `SGLang` or `AsyncSGLang` instance from an `openai.OpenAI` or
+    `openai.AsyncOpenAI` instance.
+
+    Parameters
+    ----------
+    client
+        An `openai.OpenAI` or `openai.AsyncOpenAI` instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Union[SGLang, AsyncSGLang]
+        An Outlines `SGLang` or `AsyncSGLang` model instance.
+
+    """
+    from openai import AsyncOpenAI, OpenAI
+
+    if isinstance(client, OpenAI):
+        return SGLang(client, model_name)
+    elif isinstance(client, AsyncOpenAI):
+        return AsyncSGLang(client, model_name)
+    else:
+        raise ValueError(
+            f"Unsupported client type: {type(client)}.\n"
+            "Please provide an OpenAI or AsyncOpenAI instance."
+        )
diff --git a/outlines/models/tgi.py b/outlines/models/tgi.py
new file mode 100644
index 000000000..d9f3ffd75
--- /dev/null
+++ b/outlines/models/tgi.py
@@ -0,0 +1,340 @@
+"""Integration with a TGI server."""
+
+import json
+from functools import singledispatchmethod
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    AsyncIterator,
+    Iterator,
+    Optional,
+    Union,
+)
+
+from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
+from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
+
+if TYPE_CHECKING:
+    from huggingface_hub import AsyncInferenceClient, InferenceClient
+
+__all__ = ["AsyncTGI", "TGI", "from_tgi"]
+
+
+class TGITypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `TGI` and `AsyncTGI` models."""
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the client.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available with TGI. "
+            + "Please provide a string."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str) -> str:
+        return model_input
+
+    @format_input.register(list)
+    def format_list_input(self, model_input: list):
+        raise NotImplementedError("TGI does not support batch inference.")
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the structured output argument to pass to the client.
+
+        Argument
+        --------
+        output_type
+            The structured output type provided.
+
+        Returns
+        -------
+        dict
+            The structured output argument to pass to the client.
+
+        """
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            raise NotImplementedError(
+                "TGI does not support CFG-based structured outputs."
+            )
+        elif isinstance(term, JsonSchema):
+            return {
+                "grammar": {
+                    "type": "json",
+                    "value": json.loads(term.schema),
+                }
+            }
+        else:
+            return {
+                "grammar": {
+                    "type": "regex",
+                    "value": to_regex(term),
+                }
+            }
+
+
+class TGI(Model):
+    """Thin wrapper around a `huggingface_hub.InferenceClient` client used to
+    communicate with a `TGI` server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the
+    `huggingface_hub.InferenceClient` client.
+
+    """
+
+    def __init__(self, client):
+        """
+        Parameters
+        ----------
+        client
+            A huggingface `InferenceClient` client instance.
+
+        """
+        self.client = client
+        self.type_adapter = TGITypeAdapter()
+
+    def generate(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
+        """Generate text using TGI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input,
+            output_type,
+            **inference_kwargs,
+        )
+
+        return self.client.text_generation(**client_args)
+
+    def generate_stream(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using TGI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = self.client.text_generation(
+            **client_args, stream=True,
+        )
+
+        for chunk in stream:  # pragma: no cover
+            yield chunk
+
+    def _build_client_args(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
+        """Build the arguments to pass to the TGI client."""
+        prompt = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+
+        client_args = {
+            "prompt": prompt,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+class AsyncTGI(AsyncModel):
+    """Thin async wrapper around a `huggingface_hub.AsyncInferenceClient`
+    client used to communicate with a `TGI` server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the
+    `huggingface_hub.AsyncInferenceClient` client.
+
+    """
+
+    def __init__(self, client):
+        """
+        Parameters
+        ----------
+        client
+            A huggingface `AsyncInferenceClient` client instance.
+
+        """
+        self.client = client
+        self.type_adapter = TGITypeAdapter()
+
+    async def generate(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
+        """Generate text using TGI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        response = await self.client.text_generation(**client_args)
+
+        return response
+
+    async def generate_stream( # type: ignore
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> AsyncIterator[str]:
+        """Stream text using TGI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        AsyncIterator[str]
+            An async iterator that yields the text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = await self.client.text_generation(
+            **client_args, stream=True
+        )
+
+        async for chunk in stream:  # pragma: no cover
+            yield chunk
+
+    def _build_client_args(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
+        """Build the arguments to pass to the TGI client."""
+        prompt = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+
+        client_args = {
+            "prompt": prompt,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+def from_tgi(
+    client: Union["InferenceClient", "AsyncInferenceClient"],
+) -> Union[TGI, AsyncTGI]:
+    """Create an Outlines `TGI` or `AsyncTGI` model instance from an
+    `huggingface_hub.InferenceClient` or `huggingface_hub.AsyncInferenceClient`
+    instance.
+
+    Parameters
+    ----------
+    client
+        An `huggingface_hub.InferenceClient` or
+        `huggingface_hub.AsyncInferenceClient` instance.
+
+    Returns
+    -------
+    Union[TGI, AsyncTGI]
+        An Outlines `TGI` or `AsyncTGI` model instance.
+
+    """
+    from huggingface_hub import AsyncInferenceClient, InferenceClient
+
+    if isinstance(client, InferenceClient):
+        return TGI(client)
+    elif isinstance(client, AsyncInferenceClient):
+        return AsyncTGI(client)
+    else:
+        raise ValueError(
+            f"Unsupported client type: {type(client)}.\n"
+            + "Please provide an HuggingFace InferenceClient "
+            + "or AsyncInferenceClient instance."
+        )
diff --git a/outlines/models/tokenizer.py b/outlines/models/tokenizer.py
index 1a5708d85..5463c5eaa 100644
--- a/outlines/models/tokenizer.py
+++ b/outlines/models/tokenizer.py
@@ -1,7 +1,9 @@
-from typing import Dict, Hashable, List, Protocol, Set, Tuple, Union
+from typing import Dict, Hashable, List, Protocol, Set, Tuple, Union, TYPE_CHECKING
 
-import numpy as np
-from numpy.typing import NDArray
+
+if TYPE_CHECKING:
+    import numpy as np
+    from numpy.typing import NDArray
 
 
 class Tokenizer(Hashable, Protocol):
@@ -13,11 +15,11 @@ class Tokenizer(Hashable, Protocol):
 
     def encode(
         self, prompt: Union[str, List[str]]
-    ) -> Tuple[NDArray[np.int64], NDArray[np.int64]]:
+    ) -> "Tuple['NDArray[np.int64]', 'NDArray[np.int64]']":
         """Translate the input prompts into arrays of token ids and attention mask."""
         ...
 
-    def decode(self, token_ids: NDArray[np.int64]) -> List[str]:
+    def decode(self, token_ids: "NDArray[np.int64]") -> List[str]:
         """Translate an array of token ids to a string or list of strings."""
         ...
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 6b204ec6e..01c7b34a5 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,20 +1,22 @@
-import dataclasses
-import inspect
-from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, Union
+"""Integration with the `transformers` library."""
 
-from outlines.generate.api import GenerationParameters, SamplingParameters
+from functools import singledispatchmethod
+from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union
+
+from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
+from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     import torch
-    from transformers import PreTrainedModel, PreTrainedTokenizer
-
-    from outlines.processors import OutlinesLogitsProcessor
-
-__all__ = ["transformers"]
-
+    from transformers import (
+        PreTrainedTokenizer,
+        PreTrainedModel,
+        ProcessorMixin,
+        LogitsProcessorList,
+    )
 
-KVCacheType = Tuple[Tuple["torch.DoubleTensor", "torch.DoubleTensor"], ...]
+__all__ = ["Transformers", "TransformersMultiModal", "from_transformers"]
 
 
 def get_llama_tokenizer_types():
@@ -25,28 +27,28 @@ def get_llama_tokenizer_types():
     """
     try:
         from transformers.models.llama import LlamaTokenizer
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class LlamaTokenizer:  # type: ignore
             pass
 
     try:
         from transformers.models.llama import LlamaTokenizerFast
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class LlamaTokenizerFast:  # type: ignore
             pass
 
     try:
         from transformers.models.code_llama import CodeLlamaTokenizer
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class CodeLlamaTokenizer:  # type: ignore
             pass
 
     try:
         from transformers.models.code_llama import CodeLlamaTokenizerFast
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class CodeLlamaTokenizerFast:  # type: ignore
             pass
@@ -126,229 +128,190 @@ def __setstate__(self, state):
         self.__init__(state["tokenizer"])
 
 
-class Transformers:
-    """Represents a `transformers` model."""
-
-    def __init__(
-        self,
-        model: "PreTrainedModel",
-        tokenizer: "PreTrainedTokenizer",
-    ):
-        self.model = model
-        self.tokenizer = TransformerTokenizer(tokenizer)
+class TransformersTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `Transformers` model."""
 
-    def forward(
-        self,
-        input_ids: "torch.LongTensor",
-        attention_mask: "torch.LongTensor",
-        past_key_values: Optional[Tuple] = None,
-    ) -> Tuple["torch.FloatTensor", Optional[KVCacheType]]:
-        """Compute a forward pass through the transformer model.
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
 
         Parameters
         ----------
-        input_ids
-            The input token ids.  Must be one or two dimensional.
-        attention_mask
-            The attention mask.  Must be one or two dimensional.
-        past_key_values
-            A tuple of tuples containing the cached key and value tensors for each
-            attention head.
+        model_input
+            The input passed by the user.
 
         Returns
         -------
-        The computed logits and the new cached key and value tensors.
+        str
+            The formatted input to be passed to the model.
 
         """
-        try:
-            import torch
-        except ImportError:
-            ImportError(
-                "The `torch` library needs to be installed to use `transformers` models."
-            )
-        assert 0 < input_ids.ndim < 3
-
-        if past_key_values:
-            input_ids = input_ids[..., -1].unsqueeze(-1)
-
-        with torch.inference_mode():
-            output = self.model(
-                input_ids,
-                attention_mask=attention_mask,
-                return_dict=True,
-                output_attentions=False,
-                output_hidden_states=False,
-                past_key_values=past_key_values,
-            )
-
-        return output.logits, output.past_key_values
+        raise NotImplementedError(
+            f"The input type {input} is not available."
+            "Please use a string or a list of strings."
+        )
 
-    def __call__(
-        self,
-        input_ids: "torch.LongTensor",
-        attention_mask: "torch.LongTensor",
-        past_key_values: Optional[Tuple] = None,
-    ) -> "torch.FloatTensor":
-        logits, kv_cache = self.forward(input_ids, attention_mask, past_key_values)
-        next_token_logits = logits[..., -1, :]
+    @format_input.register(str)
+    def format_str_input(self, model_input: str) -> str:
+        return model_input
 
-        return next_token_logits, kv_cache
+    @format_input.register(list)
+    def format_list_input(self, model_input: List[str]) -> List[str]:
+        return model_input
 
-    def generate(
+    def format_output_type(
         self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Union[str, List[str], List[List[str]]]:
-        """Generate text using `transformers`.
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Optional["LogitsProcessorList"]:
+        """Generate the logits processor argument to pass to the model.
 
         Parameters
         ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
+        output_type
+            The logits processor provided.
 
         Returns
         -------
-        The generated text
+        Optional[LogitsProcessorList]
+            The logits processor to pass to the model.
+
         """
-        if isinstance(prompts, str):
-            # convert to 2d
-            input_ids, attention_mask = self.tokenizer.encode([prompts])
-        else:
-            input_ids, attention_mask = self.tokenizer.encode(prompts)
+        from transformers import LogitsProcessorList
 
-        inputs = {
-            "input_ids": input_ids.to(self.model.device),
-            "attention_mask": attention_mask.to(self.model.device),
-        }
-        if (
-            "attention_mask"
-            not in inspect.signature(self.model.forward).parameters.keys()
-        ):
-            del inputs["attention_mask"]
+        if output_type is not None:
+            return LogitsProcessorList([output_type])
+        return None
 
-        generation_kwargs = self._get_generation_kwargs(
-            prompts,
-            generation_parameters,
-            logits_processor,
-            sampling_parameters,
-        )
-        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
 
-        # if single str input and single sample per input, convert to a 1D output
-        if isinstance(prompts, str):
-            generated_ids = generated_ids.squeeze(0)
+class Transformers(Model):
+    """Thin wrapper around a `transformers` model and a `transformers`
+    tokenizer.
 
-        return self._decode_generation(generated_ids)
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `transformers` model and
+    tokenizer.
+
+    """
 
-    def stream(
+    def __init__(
         self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Iterator[Union[str, List[str]]]:
+        model: "PreTrainedModel",
+        tokenizer: "PreTrainedTokenizer",
+    ):
         """
-        Temporary stream stand-in which implements stream() signature
-        and equivalent behaviour but isn't yielded until generation completes.
+        Parameters:
+        ----------
+        model
+            A `PreTrainedModel`, or any model that is compatible with the
+            `transformers` API for models.
+        tokenizer
+            A `PreTrainedTokenizer`, or any tokenizer that is compatible with
+            the `transformers` API for tokenizers.
 
-        TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
         """
-        if isinstance(prompts, str):
-            # convert to 2d
-            input_ids, attention_mask = self.tokenizer.encode([prompts])
+        # We need to handle the cases in which jax/flax or tensorflow
+        # is not available in the environment.
+        try:
+            from transformers import FlaxPreTrainedModel
+        except ImportError:  # pragma: no cover
+            FlaxPreTrainedModel = None
+
+        try:
+            from transformers import TFPreTrainedModel
+        except ImportError:  # pragma: no cover
+            TFPreTrainedModel = None
+
+        tokenizer.padding_side = "left"
+        self.model = model
+        self.transformer_tokenizer = tokenizer
+        self.tokenizer = TransformerTokenizer(tokenizer)
+        self.type_adapter = TransformersTypeAdapter()
+
+        if (
+            FlaxPreTrainedModel is not None
+            and isinstance(model, FlaxPreTrainedModel)
+        ):
+            self.tensor_library_name = "jax"
+        elif (
+            TFPreTrainedModel is not None
+            and isinstance(model, TFPreTrainedModel)
+        ):
+            self.tensor_library_name = "tensorflow"
         else:
-            input_ids, attention_mask = self.tokenizer.encode(prompts)
+            self.tensor_library_name = "torch"
+
+    def _prepare_model_inputs(
+        self,
+        model_input: Union[str, List[str], dict],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Tuple[Union[str, List[str]], dict]:
+        """Turn the user input into arguements to pass to the model"""
+        prompts = self.type_adapter.format_input(model_input)
+        input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
             "input_ids": input_ids.to(self.model.device),
             "attention_mask": attention_mask.to(self.model.device),
         }
-        if (
-            "attention_mask"
-            not in inspect.signature(self.model.forward).parameters.keys()
-        ):
-            del inputs["attention_mask"]
 
-        generation_kwargs = self._get_generation_kwargs(
-            prompts,
-            generation_parameters,
-            logits_processor,
-            sampling_parameters,
-        )
-        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
+        return prompts, inputs
 
-        # if single str input and single sample per input, convert to a 1D output
-        if isinstance(prompts, str):
-            generated_ids = generated_ids.squeeze(0)
+    def generate(
+        self,
+        model_input: Union[str, List[str], dict],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, List[str]]:
+        """Generate text using `transformers`.
 
-        for i in range(generated_ids.size(-1)):
-            output_group_ids = generated_ids.select(-1, i).unsqueeze(-1)
-            yield self._decode_generation(output_group_ids)
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response. For
+            multi-modal models, the input should be a dictionary containing the
+            `text` key with a value of type `Union[str, List[str]]` and the
+            other keys required by the model.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        inference_kwargs
+            Additional keyword arguments to pass to the `generate` method
+            of the `transformers` model.
+
+        Returns
+        -------
+        Union[str, List[str]]
+            The text generated by the model.
 
-    def _get_generation_kwargs(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> dict:
-        """
-        Convert outlines generation parameters into model.generate kwargs
         """
-        from transformers import GenerationConfig, LogitsProcessorList, set_seed
+        prompts, inputs = self._prepare_model_inputs(model_input, output_type)
+        logits_processor = self.type_adapter.format_output_type(output_type)
 
-        max_new_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
+        generated_ids = self._generate_output_seq(
+            prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )
-        if max_new_tokens is None:
-            max_new_tokens = int(2**30)
 
-        # global seed, not desirable
-        if seed is not None:
-            set_seed(seed)
+        # if single str input, convert to a 1D outputt
+        if isinstance(prompts, str):
+            generated_ids = generated_ids.squeeze(0)
+
+        return self._decode_generation(generated_ids)
 
-        if logits_processor is not None:
-            logits_processor_list = LogitsProcessorList([logits_processor])
-        else:
-            logits_processor_list = None
-
-        generation_config = GenerationConfig(
-            max_new_tokens=max_new_tokens,
-            stop_strings=stop_at,
-            num_return_sequences=(num_samples or 1),
-            top_p=top_p,
-            top_k=top_k,
-            temperature=temperature,
-            do_sample=(sampler == "multinomial"),
-            num_beams=(num_samples if sampler == "beam_search" else 1),
-            eos_token_id=self.tokenizer.eos_token_id,
-            pad_token_id=self.tokenizer.pad_token_id,
-        )
+    def generate_stream(self, model_input, output_type, **inference_kwargs):
+        """Not available for `transformers` models.
+
+        TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
 
-        return dict(
-            logits_processor=logits_processor_list,
-            generation_config=generation_config,
-            tokenizer=self.tokenizer.tokenizer,
+        """
+        raise NotImplementedError(
+            "Streaming is not implemented for Transformers models."
         )
 
-    def _generate_output_seq(
-        self, prompts, inputs, generation_config, **generation_kwargs
-    ):
+    def _generate_output_seq(self, prompts, inputs, **inference_kwargs):
         input_ids = inputs["input_ids"]
         output_ids = self.model.generate(
-            **inputs, generation_config=generation_config, **generation_kwargs
+            **inputs,
+            tokenizer=self.transformer_tokenizer,
+            **inference_kwargs,
         )
 
         # encoder-decoder returns output_ids only, decoder-only returns full seq ids
@@ -358,12 +321,10 @@ def _generate_output_seq(
             generated_ids = output_ids[:, input_ids.shape[1] :]
 
         # if batch list inputs AND multiple samples per input, convert generated_id to 3D view
-        num_samples = generation_config.num_return_sequences or 1
-
+        num_samples = inference_kwargs.get("num_return_sequences", 1)
         if num_samples > 1 and isinstance(prompts, list):
             batch_size = input_ids.size(0)
-            num_return_sequences = generation_config.num_return_sequences or 1
-            generated_ids = generated_ids.view(batch_size, num_return_sequences, -1)
+            generated_ids = generated_ids.view(batch_size, num_samples, -1)
 
         return generated_ids
 
@@ -377,82 +338,164 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
                 self.tokenizer.decode(generated_ids[i])
                 for i in range(len(generated_ids))
             ]
-        else:
+        else:  # pragma: no cover
             raise TypeError(
                 f"Generated outputs aren't 1D, 2D or 3D, but instead are {generated_ids.shape}"
             )
 
 
-def transformers(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-    model_class=None,
-    tokenizer_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
+class TransformersMultiModalTypeAdapter(ModelTypeAdapter):
+    """Type adapter for `TransformersMultiModal` model."""
 
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    tokenizer_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the tokenizer.
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt arguments to pass to the model.
 
-    Returns
-    -------
-    A `TransformersModel` model instance.
+        Argument
+        --------
+        model_input
+            The input passed by the user.
 
-    """
-    if model_class is None or tokenizer_class is None:
-        try:
-            from transformers import AutoModelForCausalLM, AutoTokenizer
-        except ImportError:
-            raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the model.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. Please provide a "
+            + "dictionary containing at least the 'text' key with a value "
+            + "of type Union[str, List[str]]. You should also include the "
+            + "other keys required by your processor (for instance, 'images' "
+            + "or 'audios')."
+            + "Make sure that the text is correctly formatted for the model "
+            + "(e.g. include <image> or <|AUDIO|> tags) and that the number "
+            + "of text tags match the number of additional assets provided."
+        )
+
+    @format_input.register(dict)
+    def format_list_input(self, model_input: dict) -> dict:
+        if "text" not in model_input:
+            raise ValueError(
+                "The input must contain the 'text' key along with the other "
+                + "keys required by your processor."
             )
-    if model_class is None:
-        model_class = AutoModelForCausalLM
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
+        return model_input
 
-    if device is not None:
-        model_kwargs["device_map"] = device
+    def format_output_type(
+        self,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Optional["LogitsProcessorList"]:
+        """Generate the logits processor argument to pass to the model.
 
-    model = model_class.from_pretrained(model_name, **model_kwargs)
+        Argument
+        --------
+        output_type
+            The logits processor provided.
 
-    tokenizer_kwargs.setdefault("padding_side", "left")
-    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
+        Returns
+        -------
+        Optional[LogitsProcessorList]
+            The logits processor to pass to the model.
 
-    return Transformers(model, tokenizer)
+        """
+        from transformers import LogitsProcessorList
 
+        if output_type is not None:
+            return LogitsProcessorList([output_type])
+        return None
 
-def mamba(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-):
-    try:
-        from transformers import MambaForCausalLM
 
-    except ImportError:
-        raise ImportError(
-            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
-        )
+class TransformersMultiModal(Transformers):
+    """Thin wrapper around a `transformers` model and a `transformers`
+    processor.
 
-    return transformers(
-        model_name=model_name,
-        device=device,
-        model_kwargs=model_kwargs,
-        tokenizer_kwargs=tokenizer_kwargs,
-        model_class=MambaForCausalLM,
-    )
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `transformers` model and
+    processor.
+
+    """
+
+    def __init__(self, model: "PreTrainedModel", processor):
+        """Create a TransformersMultiModal model instance
+
+        We rely on the `__init__` method of the `Transformers` class to handle
+        most of the initialization and then add elements specific to vision
+        models.
+
+        Parameters
+        ----------
+        model
+            A `PreTrainedModel`, or any model that is compatible with the
+            `transformers` API for models.
+        processor
+            A `ProcessorMixin` instance.
+
+        """
+        self.processor = processor
+        self.processor.padding_side = "left"
+        self.processor.pad_token = "[PAD]"
+
+        tokenizer: "PreTrainedTokenizer" = self.processor.tokenizer
+
+        super().__init__(model, tokenizer)
+
+        self.type_adapter = TransformersMultiModalTypeAdapter()
+
+    def _prepare_model_inputs(
+        self,
+        model_input: Union[str, List[str], dict],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Tuple[Union[str, List[str]], dict]:
+        """Turn the user input into arguements to pass to the model"""
+        model_input = self.type_adapter.format_input(model_input)
+        inputs = self.processor(
+            **model_input, padding=True, return_tensors="pt"
+        ).to(self.model.device)
+
+        return model_input["text"], inputs
+
+
+def from_transformers(
+    model: "PreTrainedModel",
+    tokenizer_or_processor: Union["PreTrainedTokenizer", "ProcessorMixin"],
+) -> Union[Transformers, TransformersMultiModal]:
+    """Create an Outlines `Transformers` or `TransformersMultiModal` model
+    instance from a `PreTrainedModel` instance and a `PreTrainedTokenizer` or
+    `ProcessorMixin` instance.
+
+    `outlines` supports `PreTrainedModelForCausalLM`,
+    `PreTrainedMambaForCausalLM`, `PreTrainedModelForSeq2Seq` and any model
+    that implements the `transformers` model API.
+
+    Parameters
+    ----------
+    model
+        A `transformers.PreTrainedModel` instance.
+    tokenizer_or_processor
+        A `transformers.PreTrainedTokenizer` or
+        `transformers.ProcessorMixin` instance.
+
+    Returns
+    -------
+    Union[Transformers, TransformersMultiModal]
+        An Outlines `Transformers` or `TransformersMultiModal` model instance.
+
+    """
+    from transformers import (
+        PreTrainedTokenizer, PreTrainedTokenizerFast, ProcessorMixin)
+
+    if isinstance(
+        tokenizer_or_processor, (PreTrainedTokenizer, PreTrainedTokenizerFast)
+    ):
+        tokenizer = tokenizer_or_processor
+        return Transformers(model, tokenizer)
+    elif isinstance(tokenizer_or_processor, ProcessorMixin):
+        processor = tokenizer_or_processor
+        return TransformersMultiModal(model, processor)
+    else:
+        raise ValueError(
+            "We could determine whether the model passed to `from_transformers`"
+            + " is a text-2-text or a multi-modal model. Please provide a "
+            + "a transformers tokenizer or processor."
+        )
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
deleted file mode 100644
index c8a86536e..000000000
--- a/outlines/models/transformers_vision.py
+++ /dev/null
@@ -1,138 +0,0 @@
-from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
-
-from outlines.generate.api import GenerationParameters, SamplingParameters
-from outlines.models import Transformers
-
-if TYPE_CHECKING:
-    from outlines.processors import OutlinesLogitsProcessor
-
-
-class TransformersVision(Transformers):
-    def __init__(self, model, tokenizer, processor):
-        super().__init__(model, tokenizer)
-        self.processor = processor
-
-    def generate(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[List[Any], List[List[Any]]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Union[str, List[str], List[List[str]]]:
-        """Generate text using `transformers`.
-
-        Parameters
-        ----------
-        prompts
-            A prompt or list of prompts.
-        media
-            A List[PIL.Image] or List[List[PIL.Image]]
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-
-        Returns
-        -------
-        The generated text
-        """
-        inputs = self.processor(
-            text=prompts, images=media, padding=True, return_tensors="pt"
-        ).to(self.model.device)
-
-        generation_kwargs = self._get_generation_kwargs(
-            prompts,
-            generation_parameters,
-            logits_processor,
-            sampling_parameters,
-        )
-        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
-
-        # if single str input and single sample per input, convert to a 1D output
-        if isinstance(prompts, str):
-            # Should always be true until NotImplementedError above is fixed
-            generated_ids = generated_ids.squeeze(0)
-
-        return self._decode_generation(generated_ids)
-
-    def stream(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[Any, List[Any]],  # TODO: docstring
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Iterator[Union[str, List[str]]]:
-        raise NotImplementedError
-
-
-def transformers_vision(
-    model_name: str,
-    model_class,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    processor_kwargs: dict = {},
-    tokenizer_class=None,
-    processor_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
-
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    model_class
-        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
-        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    processor_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the processor.
-
-    Returns
-    -------
-    A `TransformersModel` model instance.
-
-    """
-    if processor_class is None or tokenizer_class is None:
-        try:
-            from transformers import AutoProcessor, AutoTokenizer
-        except ImportError:
-            raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
-            )
-    if processor_class is None:
-        processor_class = AutoProcessor
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
-
-    if device is not None:
-        model_kwargs["device_map"] = device
-
-    model = model_class.from_pretrained(model_name, **model_kwargs)
-
-    processor_kwargs.setdefault("padding_side", "left")
-    processor_kwargs.setdefault("pad_token", "[PAD]")
-    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-    if tokenizer_class is None:
-        if getattr(processor, "tokenizer", None):
-            tokenizer = processor.tokenizer
-        else:
-            tokenizer = AutoTokenizer.from_pretrained(model_name, **processor_kwargs)
-    else:
-        tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
-
-    return TransformersVision(model, tokenizer, processor)
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index b9b035d1f..f787e7072 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,227 +1,362 @@
-import dataclasses
-from typing import TYPE_CHECKING, List, Optional, Union
+"""Integration with a vLLM server."""
 
-from outlines.generate.api import GenerationParameters, SamplingParameters
+import json
+from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
+
+from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
+from outlines.models.openai import OpenAITypeAdapter
+from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
+from outlines.templates import Vision
 
 if TYPE_CHECKING:
-    from transformers import PreTrainedTokenizerBase
-    from vllm import LLM
-    from vllm.sampling_params import SamplingParams
+    from openai import AsyncOpenAI, OpenAI
 
+__all__ = ["VLLM", "AsyncVLLM", "from_vllm"]
 
-class VLLM:
-    """Represents a vLLM model.
 
-    We wrap models from model providing libraries in order to give all of
-    them the same interface in Outlines and allow users to easily switch
-    between providers. This class wraps the `vllm.LLM` class from the
-    `vllm` library.
+class VLLMTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `VLLM` and `AsyncVLLM` models."""
 
-    """
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
+        """Generate the prompt argument to pass to the client.
 
-    def __init__(self, model: "LLM"):
-        self.model = model
-        self.lora_request = None
+        We rely on the OpenAITypeAdapter to format the input as the vLLM server
+        expects input in the same format as OpenAI.
 
-        self.tokenizer = self._get_tokenizer()
+        Parameters
+        ----------
+        model_input
+            The input passed by the user.
 
-    def _get_tokenizer(self):
-        if hasattr(self.model, "get_tokenizer"):
-            tokenizer = self.model.get_tokenizer()
-        elif hasattr(self.model, "tokenizer"):
-            if hasattr(self.model.tokenizer, "tokenizer"):
-                tokenizer = self.model.tokenizer.tokenizer
-            else:
-                tokenizer = self.model.tokenizer
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the model.
+
+        """
+        return OpenAITypeAdapter().format_input(model_input)
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the structured output argument to pass to the client.
+
+        Parameters
+        ----------
+        output_type
+            The structured output type provided.
+
+        Returns
+        -------
+        dict
+            The structured output argument to pass to the model.
+
+        """
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            return {"guided_grammar": term.definition}
+        elif isinstance(term, JsonSchema):
+            extra_body = {"guided_json": json.loads(term.schema)}
+            if term.whitespace_pattern:
+                extra_body["whitespace_pattern"] = term.whitespace_pattern
+            return extra_body
         else:
-            raise ValueError(
-                "The provided LLM instance neither has a "
-                "`tokenizer` attribute or a `get_tokenizer` method."
-            )
-        return adapt_tokenizer(tokenizer=tokenizer)
+            return {"guided_regex": to_regex(term)}
 
-    def generate(
+
+class VLLM(Model):
+    """Thin wrapper around the `openai.OpenAI` client used to communicate with
+    a `vllm` server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    `vllm` server.
+    """
+
+    def __init__(
         self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor,
-        sampling_parameters: SamplingParameters,
-        *,
-        sampling_params: Optional["SamplingParams"] = None,
-        use_tqdm: bool = True,
+        client: "OpenAI",
+        model_name: Optional[str] = None,
     ):
+        """
+        Parameters
+        ----------
+        client
+            An `openai.OpenAI` client instance.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = VLLMTypeAdapter()
+
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
         """Generate text using vLLM.
 
         Parameters
         ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-        sampling_params
-            An instance of `vllm.sampling_params.SamplingParams`. The values
-            passed via this dataclass supersede the values of the parameters
-            in `generation_parameters` and `sampling_parameters`. See the
-            vLLM documentation for more details: https://docs.vllm.ai/en/latest/dev/sampling_params.html.
-        use_tqdm
-            A boolean in order to display progress bar while inferencing
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        The generated text, of shape `(n_batch, n_samples)`. If there are only
-        one batch and several samples, the list is of shape `(n_samples)`. If
-        this is a batch with several sequences but only one sample the list is
-        of shape `(n_batch)`. If there is only one sequence and one sample, a
-        string is returned.
+        Union[str, list[str]]
+            The text generated by the model.
 
         """
-        from vllm.sampling_params import SamplingParams
-
-        if sampling_params is None:
-            sampling_params = SamplingParams()
-
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-
-        # We only update the values in `sampling_params` if they
-        # are specified by the user when calling the generator.
-        if max_tokens is not None:
-            sampling_params.max_tokens = max_tokens
-        if stop_at is not None:
-            if isinstance(stop_at, str):
-                stop_at = [stop_at]
-            sampling_params.stop = stop_at
-        if seed is not None:
-            sampling_params.seed = seed
-
-        sampling_params.logits_processors = (
-            [logits_processor] if logits_processor is not None else []
+        client_args = self._build_client_args(
+            model_input,
+            output_type,
+            **inference_kwargs,
         )
 
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
+        response = self.client.chat.completions.create(**client_args)
+
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The vLLM server refused to answer the request: "
+                    f"{message.refusal}"
+                )
+
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
+
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using vLLM.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
         )
 
-        # We only update the values in `sampling_params` that
-        # were not specified by the user.
-        if sampling_params.n == 1:
-            sampling_params.n = num_samples
-            sampling_params.best_of = num_samples
-        if top_p is not None and sampling_params.top_p == 1.0:
-            sampling_params.top_p = top_p
-        if top_k is not None and sampling_params.top_k == -1:
-            sampling_params.top_k = top_k
-            # TODO: remove this if statement once fixed
-            # https://github.com/vllm-project/vllm/issues/5404#issuecomment-2175972897
-            if top_k == 1:
-                sampling_params.repetition_penalty = 0
-        if temperature is not None and sampling_params.temperature == 1.0:
-            sampling_params.temperature = temperature
-        if sampler == "beam_search":
-            sampling_params.use_beam_search = True
-
-        results = self.model.generate(
-            prompts,
-            sampling_params=sampling_params,
-            lora_request=self.lora_request,
-            use_tqdm=use_tqdm,
+        stream = self.client.chat.completions.create(
+            **client_args, stream=True,
         )
-        results = [[sample.text for sample in batch.outputs] for batch in results]
 
-        batch_size = len(results)
-        sample_size = len(results[0])
+        for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
 
-        if batch_size == 1 and sample_size == 1:
-            return results[0][0]
-        elif batch_size == 1:
-            return results[0]
-        elif sample_size == 1:
-            return [batch[0] for batch in results]
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
+        """Build the arguments to pass to the OpenAI client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        extra_body = inference_kwargs.pop("extra_body", {})
+        extra_body.update(output_type_args)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        client_args = {
+            **messages,
+            **inference_kwargs,
+        }
+        if extra_body:
+            client_args["extra_body"] = extra_body
+
+        return client_args
+
+
+class AsyncVLLM(AsyncModel):
+    """Thin async wrapper around the `openai.OpenAI` client used to communicate
+    with a `vllm` server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    `vllm` server.
+    """
 
-        return results
+    def __init__(
+        self,
+        client: "AsyncOpenAI",
+        model_name: Optional[str] = None,
+    ):
+        """
+        Parameters
+        ----------
+        client
+            An `openai.AsyncOpenAI` client instance.
 
-    def stream(self, *args, **kwargs):
-        """Return a text generator.
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = VLLMTypeAdapter()
 
-        Streaming is not yet available for `vllm.LLM`.
+    async def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using vLLM.
 
-        TODO: Implement the streaming functionality ourselves.
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The text generated by the model.
 
         """
-        raise NotImplementedError(
-            "Streaming is not available for the vLLM integration."
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
         )
 
-    def load_lora(self, adapter_path: Optional[str]):
-        from vllm.lora.request import LoRARequest
+        response = await self.client.chat.completions.create(**client_args)
 
-        if adapter_path is None:
-            self.lora_request = None
-        else:
-            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
-
-
-def vllm(model_name: str, **vllm_model_params):
-    """Load a vLLM model.
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The vLLM server refused to answer the request: "
+                    f"{message.refusal}"
+                )
 
-    Parameters
-    ---------
-    model_name
-        The name of the model to load from the HuggingFace hub.
-    vllm_model_params
-        vLLM-specific model parameters. See the vLLM code for the full list:
-        https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/llm.py
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
 
-    """
-    from vllm import LLM
+    async def generate_stream( # type: ignore
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> AsyncIterator[str]:
+        """Stream text using vLLM.
 
-    model = LLM(model_name, **vllm_model_params)
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
+        inference_kwargs
+            Additional keyword arguments to pass to the client.
 
-    return VLLM(model)
+        Returns
+        -------
+        AsyncIterator[str]
+            An async iterator that yields the text generated by the model.
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
 
+        stream = await self.client.chat.completions.create(
+            **client_args,
+            stream=True,
+        )
 
-def adapt_tokenizer(tokenizer: "PreTrainedTokenizerBase") -> "PreTrainedTokenizerBase":
-    """Adapt a tokenizer to use to compile the FSM.
+        async for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
 
-    The API of Outlines tokenizers is slightly different to that of `transformers`. In
-    addition we need to handle the missing spaces to Llama's tokenizer to be able to
-    compile FSMs for this model.
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
+        """Build the arguments to pass to the OpenAI client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        extra_body = inference_kwargs.pop("extra_body", {})
+        extra_body.update(output_type_args)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        client_args = {
+            **messages,
+            **inference_kwargs,
+        }
+        if extra_body:
+            client_args["extra_body"] = extra_body
+
+        return client_args
+
+
+def from_vllm(
+    client: Union["OpenAI", "AsyncOpenAI"],
+    model_name: Optional[str] = None,
+) -> Union[VLLM, AsyncVLLM]:
+    """Create an Outlines `VLLM` or `AsyncVLLM` model instance from an
+    `openai.OpenAI` or `openai.AsyncOpenAI` instance.
 
     Parameters
     ----------
-    tokenizer
-        The tokenizer of the model.
+    client
+        An `openai.OpenAI` or `openai.AsyncOpenAI` instance.
+    model_name
+        The name of the model to use.
 
     Returns
     -------
-    PreTrainedTokenizerBase
-        The adapted tokenizer.
-    """
-    from transformers import SPIECE_UNDERLINE
-
-    tokenizer.vocabulary = tokenizer.get_vocab()
-    tokenizer.special_tokens = set(tokenizer.all_special_tokens)
-
-    def convert_token_to_string(token: Union[str, bytes]) -> str:
-        string = tokenizer.convert_tokens_to_string([token])
+    Union[VLLM, AsyncVLLM]
+        An Outlines `VLLM` or `AsyncVLLM` model instance.
 
-        # A hack to handle missing spaces to HF's Llama tokenizers
-        if (
-            type(token) is str
-            and token.startswith(SPIECE_UNDERLINE)
-            or token == "<0x20>"
-        ):
-            return " " + string
-
-        return string
-
-    tokenizer.convert_token_to_string = convert_token_to_string
-
-    return tokenizer
+    """
+    from openai import AsyncOpenAI, OpenAI
+
+    if isinstance(client, OpenAI):
+        return VLLM(client, model_name)
+    elif isinstance(client, AsyncOpenAI):
+        return AsyncVLLM(client, model_name)
+    else:
+        raise ValueError(
+            f"Unsupported client type: {type(client)}.\n"
+            "Please provide an OpenAI or AsyncOpenAI instance."
+        )
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
new file mode 100644
index 000000000..f970b2f9a
--- /dev/null
+++ b/outlines/models/vllm_offline.py
@@ -0,0 +1,206 @@
+"""Integration with the `vllm` library (offline mode)."""
+
+import json
+import warnings
+from functools import singledispatchmethod
+from typing import TYPE_CHECKING, Any, List, Optional, Union
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
+
+if TYPE_CHECKING:
+    from transformers import PreTrainedTokenizerBase
+    from vllm import LLM
+
+__all__ = ["VLLMOffline", "from_vllm_offline"]
+
+
+class VLLMOfflineTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `VLLMOffline` model."""
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. "
+            "Please use a string or a list of strings."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str) -> str:
+        return model_input
+
+    @format_input.register(list)
+    def format_list_input(self, model_input: List[str]) -> List[str]:
+        return model_input
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the structured output argument to pass to the model.
+
+        For vLLM, the structured output definition is set in the
+        `GuidedDecodingParams` constructor that is provided as a value to the
+        `guided_decoding` parameter of the `SamplingParams` constructor, itself
+        provided as a value to the `sampling_params` parameter of the `generate`
+        method.
+
+        Parameters
+        ----------
+        output_type
+            The structured output type provided.
+
+        Returns
+        -------
+        dict
+            The arguments to provide to the `GuidedDecodingParams` constructor.
+
+        """
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            return {"grammar": term.definition}
+        elif isinstance(term, JsonSchema):
+            guided_decoding_params = {"json": json.loads(term.schema)}
+            if term.whitespace_pattern:
+                guided_decoding_params["whitespace_pattern"] = term.whitespace_pattern
+            return guided_decoding_params
+        else:
+            return {"regex": to_regex(term)}
+
+
+class VLLMOffline(Model):
+    """Thin wrapper around a `vllm.LLM` model.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `vllm.LLM` model.
+
+    """
+
+    def __init__(self, model: "LLM"):
+        """Create a VLLM model instance.
+
+        Parameters
+        ----------
+        model
+            A `vllm.LLM` model instance.
+
+        """
+        self.model = model
+        self.type_adapter = VLLMOfflineTypeAdapter()
+        self.lora_request = None # v0 legacy, to be removed
+
+    def generate(
+        self,
+        model_input: Union[str, List[str]],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, List[str], List[List[str]]]:
+        """Generate text using vLLM.
+
+        Parameters
+        ----------
+        prompt
+            The prompt based on which the model will generate a response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        inference_kwargs
+            Additional keyword arguments to pass to the `generate` method
+            in the `vllm.LLM` model.
+
+        Returns
+        -------
+        Union[str, List[str], List[List[str]]]
+            The text generated by the model.
+
+        """
+        from vllm.sampling_params import GuidedDecodingParams, SamplingParams
+
+        sampling_params = inference_kwargs.pop("sampling_params", None)
+
+        if sampling_params is None:
+            sampling_params = SamplingParams()
+
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        if output_type_args:
+            sampling_params.guided_decoding = GuidedDecodingParams(**output_type_args)
+
+        results = self.model.generate(
+            self.type_adapter.format_input(model_input),
+            sampling_params=sampling_params,
+            lora_request=self.lora_request, # v0 legacy, to be removed
+            **inference_kwargs,
+        )
+        results = [[sample.text for sample in batch.outputs] for batch in results]
+
+        batch_size = len(results)
+        sample_size = len(results[0])
+
+        if batch_size == 1 and sample_size == 1:
+            return results[0][0]
+        elif batch_size == 1:
+            return results[0]
+        elif sample_size == 1:
+            return [batch[0] for batch in results]
+
+        return results
+
+    def generate_stream(self, model_input, output_type, **inference_kwargs):
+        """Not available for `vllm.LLM`.
+
+        TODO: Implement the streaming functionality ourselves.
+
+        """
+        raise NotImplementedError(
+            "Streaming is not available for the vLLM integration."
+        )
+
+    def load_lora(self, adapter_path: Optional[str]) -> None:
+        """Load a LoRA adapter. Deprecated since v1.0.0.
+
+        Use the `lora_request` argument when calling the model or generator
+        instead.
+
+        """
+        warnings.warn("""
+            The `load_lora` method is deprecated starting from v1.0.0.
+            Support for it will be removed in v1.1.0.
+            Please use the v1 of the `outlines` library by using the
+            `outlines.from_vllm` function to create a `VLLM` model
+            instance.
+            In the v1, you must pass the `lora_request` argument as
+            a keyword argument when calling the model or generator.
+            """)
+
+        from vllm.lora.request import LoRARequest
+
+        if adapter_path is None:
+            self.lora_request = None
+        else:
+            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
+
+
+def from_vllm_offline(model: "LLM") -> VLLMOffline:
+    """Create an Outlines `VLLMOffline` model instance from a `vllm.LLM`
+    instance.
+
+    Parameters
+    ----------
+    model
+        A `vllm.LLM` instance.
+
+    Returns
+    -------
+    VLLMOffline
+        An Outlines `VLLMOffline` model instance.
+
+    """
+    return VLLMOffline(model)
diff --git a/outlines/processors/__init__.py b/outlines/processors/__init__.py
index f0f0f829b..f07ce0cbb 100644
--- a/outlines/processors/__init__.py
+++ b/outlines/processors/__init__.py
@@ -1,3 +1,5 @@
+"""Processors and guides to control generation in steerable models."""
+
 from .structured import (
     CFGLogitsProcessor,
     GuideLogitsProcessor,
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index d6fe346e0..bde129b2c 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -1,159 +1,121 @@
-from abc import abstractmethod
-from typing import TYPE_CHECKING, List, Protocol, Type, Union
-
-import numpy as np
-import torch
-from numpy.typing import NDArray
-
-if TYPE_CHECKING:
-    import mlx.core as mx
-
-
-Array = Union[NDArray, torch.Tensor, List, "mx.array"]
+"""Base class for logits processors."""
 
+from abc import abstractmethod
+from typing import TypeVar
 
-def is_mlx_array_type(array_type):
-    try:
-        import mlx.core as mx
-    except ImportError:
-        return False
-    return issubclass(array_type, mx.array)
-
+from outlines.processors.tensor_adapters import (
+    TensorAdapterImplementation,
+    tensor_adapters,
+)
 
-def is_jax_array_type(array_type):
-    try:
-        import jaxlib
-    except ImportError:
-        return False
-    return issubclass(array_type, jaxlib.xla_extension.ArrayImpl) or isinstance(
-        array_type, jaxlib.xla_extension.ArrayImpl
-    )
+TensorType = TypeVar('TensorType')
 
 
-class OutlinesLogitsProcessor(Protocol):
+class OutlinesLogitsProcessor:
+    """Base class for logits processors.
+    This class implements a shared `__call__` method is called by the models
+    and returns the processed logits. It relies on the `process_logits` method
+    that must be implemented by the subclasses to do the actual processing. The
+    `tensor_adapter` attribute, created at initialization based on the
+    tensor library name specified in the constructor, is used to manipulate the
+    tensors using the appropriate library for the model (numpy, torch...).
     """
-    Base class for logits processors which normalizes types of logits:
-    - ndarray (used by llama-cpp-python), converted to torch.Tensor
-    - mlx.core.array (used by mlx-lm), converted to torch.Tensor
-    - torch.Tensor (used by everything else)
-
-    Normalization of types and conversion to torch.Tensor
-    doesn't move memory, it just casts the type.
+    tensor_adapter: TensorAdapterImplementation
 
-    Normalizing the types allows all logits processors inheriting from this class
-    to implement a single method for all the business logit: `process_logits()`
-    """
+    def __init__(self, tensor_library_name: str):
+        """
+        Parameters
+        ----------
+        tensor_library_name
+            The name of the library to use to manipulate tensors. Possible
+            values are "jax", "mlx", "numpy", "tensorflow" and "torch". You
+            must choose the library that your model is using.
+        """
+        tensor_adapter_class = tensor_adapters.get(tensor_library_name)
+        if tensor_adapter_class is None:
+            raise NotImplementedError(
+                f"Library {tensor_library_name} is not available"
+            )
+        self.tensor_adapter = tensor_adapter_class()  # type: ignore
 
     @abstractmethod
     def process_logits(
-        self, input_ids: List[List[int]], logits: torch.Tensor
-    ) -> torch.Tensor:
-        """
-        input_ids and logits are always 2D tensors for handling a batch of sequences.
-
-        - input_ids -> List[List[tokens]]
-        - logits -> 2D_Tensor[logit floats]
-
-        Important to keep in mind when designing universal logits processors
-        - logits processors are only used once and never re-applied for a new sequence generator
-        - Some models only pass output_ids, some models such as llamacpp and transformers prefix with input_ids
-        - Some sampling methods, such as beam search, result in unstable sequence ordering in models like vLLM
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Main method to implement for logits processors subclasses.
+        This method applies a mask on the logits to bias the generation.
+        It is called by the `__call__` method that standardizes the shape of
+        `input_ids` and `logits` to ensure they are 2D tensors.
+        Elements to keep in mind when designing universal logits processors:
+        - logits processors are only used once and never re-applied for a new
+        sequence generator
+        - Some models only pass output_ids, some models such as llamacpp and
+        transformers prefix with input_ids
+        - Some sampling methods, such as beam search, result in unstable
+        sequence ordering in models like vLLM
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences in a 2D tensor.
+        logits
+            The logits for the current generation step in a 2D tensor.
+        Returns
+        -------
+        TensorType
+            The processed logits as a 2D tensor.
         """
-        pass
+        ...
 
-    @torch.no_grad()
     def __call__(
-        self,
-        input_ids: Array,
-        logits: Array,
-    ) -> Array:
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Entrypoint for logits processors, this is the method that is
+        called by the model.
+        Because different models use different structures to store the
+        input_ids and logits, we standardize their format to 2D tensors
+        before calling the `process_logits` method. After processing, the
+        logits are cast back to the original array library type before being
+        returned.
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences in a tensor.
+        logits
+            The logits for the current generation step in a tensor.
+        Returns
+        -------
+        TensorType
+            The processed logits as a tensor.
         """
-        Apply logits processor
-
-        1) Unify type
-        - convert input_ids: either ndarray, mlx array, List[int], or Tensor -> List[List[int]]
-        - convert logits: either ndarray, mlx array, or Tensor -> 2D float Tensor
-        2) Unify shape, ensure logits and input_ids are 2D
-        3) Call self.process_logits() to perform business logic
-        4) Cast logits back to original array library type
-        """
-        # ensure logits are torch Tensors
-        torch_logits = self._to_torch(logits)
-        input_ids = self._to_torch(input_ids)
-
-        assert torch_logits.shape[:-1] == input_ids.shape[:-1]
-
-        # Guarantee passed as 2D Tensors, then covert back to original (1D or 2D) shape
-        if len(torch_logits.shape) == 2:
-            processed_logits = self.process_logits(input_ids, torch_logits)
-        elif len(torch_logits.shape) == 1:
-            processed_logits = self.process_logits(
-                input_ids.unsqueeze(0), torch_logits.unsqueeze(0)
-            ).squeeze(0)
-
-        # return logits as passed array type
-        return self._from_torch(processed_logits, type(logits))
-
-    @staticmethod
-    def _to_torch(tensor_like: Array) -> torch.Tensor:
-        """Convert various types to torch.Tensor."""
-        if isinstance(tensor_like, torch.Tensor):
-            return tensor_like
-
-        elif isinstance(tensor_like, np.ndarray):
-            return torch.from_numpy(tensor_like)
-
-        elif isinstance(tensor_like, (list, tuple)):
-            return torch.tensor(tensor_like)
-
-        elif is_mlx_array_type(type(tensor_like)):
-            import mlx.core as mx
-
-            # https://ml-explore.github.io/mlx/build/html/usage/numpy.html#pytorch
-            if tensor_like.dtype == mx.bfloat16:
-                tensor_like = tensor_like.astype(mx.float32)
-            return torch.from_dlpack(np.array(tensor_like, copy=False))
-
-        elif is_jax_array_type(type(tensor_like)):
-            import jax
-
-            torch_tensor = torch.from_dlpack(jax.dlpack.to_dlpack(tensor_like))
-            return torch_tensor
-
-        else:
-            raise TypeError(
-                "LogitsProcessor must be called with either np.NDArray, "
-                "torch.Tensor, list, or mlx.core.array typed logits. "
-                f"Logits type: `{type(tensor_like)}`"
+        # if input_ids is 1D and logits is 2D with a single sequence,
+        # reshape input_ids to 2D (needed for mlx-lm)
+        if (
+            len(self.tensor_adapter.shape(input_ids)) == 1
+            and len(self.tensor_adapter.shape(logits)) == 2
+            and self.tensor_adapter.shape(logits)[0] == 1
+        ):
+            input_ids = self.tensor_adapter.unsqueeze(input_ids)
+
+        assert (
+            self.tensor_adapter.shape(logits)[:-1]
+            == self.tensor_adapter.shape(input_ids)[:-1]
+        )
+
+        # Guarantee passed as 2D Tensors, then covert back to original
+        # (1D or 2D) shape
+        if len(self.tensor_adapter.shape(logits)) == 2:
+            processed_logits = self.process_logits(input_ids, logits)
+        elif len(self.tensor_adapter.shape(logits)) == 1:
+            processed_logits = self.tensor_adapter.squeeze(
+                self.process_logits(
+                    self.tensor_adapter.unsqueeze(input_ids),
+                    self.tensor_adapter.unsqueeze(logits),
+                ),
             )
-
-    @staticmethod
-    def _from_torch(tensor: torch.Tensor, target_type: Type) -> Array:
-        """Convert torch.Tensor to the specified target type."""
-        if target_type == torch.Tensor:
-            return tensor
-
-        elif target_type == np.ndarray:
-            return tensor.detach().numpy()
-
-        elif target_type is list:
-            return tensor.detach().tolist()
-
-        elif target_type is tuple:
-            return tuple(tensor.detach().tolist())
-
-        elif is_mlx_array_type(target_type):
-            import mlx.core as mx
-
-            # numpy doesn't support bfloat16, mlx doesn't support direct conversion from torch
-            return mx.array(tensor.float().numpy())
-
-        elif is_jax_array_type(target_type):
-            import jax
-
-            return jax.dlpack.from_dlpack(tensor)
-
         else:
-            raise TypeError(
-                f"Failed to convert torch tensors to target_type `{target_type}`"
+            raise ValueError(
+                f"Logits shape {self.tensor_adapter.shape(logits)} is not "
+                + "supported"
             )
+
+        return processed_logits
diff --git a/outlines/fsm/guide.py b/outlines/processors/guide.py
similarity index 58%
rename from outlines/fsm/guide.py
rename to outlines/processors/guide.py
index 6b97d7729..bebddd9cb 100644
--- a/outlines/fsm/guide.py
+++ b/outlines/processors/guide.py
@@ -1,9 +1,13 @@
+"""Guides to control generation in steerable models.
+
+Logits processors rely on guides to control the generation process.
+"""
+
 import collections
 import copy
 import warnings
-from typing import TYPE_CHECKING, Any, Generator, Union
+from typing import TYPE_CHECKING, Any, Generator, ValuesView, Union
 
-import torch
 from lark.indenter import DedentError
 from lark.lexer import UnexpectedCharacters, UnexpectedToken
 from outlines_core.fsm.guide import Generate
@@ -15,6 +19,7 @@
 )
 
 from outlines import grammars
+from outlines.caching import cache
 from outlines.fsm.parsing import PartialLark, PartialParserState
 
 if TYPE_CHECKING:
@@ -27,59 +32,106 @@
 class Guide(CoreGuide):
     """Base definition of a generation guide.
 
-    A generation guide defines the behavior of a finite-state machine that guides
-    a text generation procedure. Unlike the DFAs built from regular expressions
-    guides can also emit a `Write` instructions which tells the model that it can
-    append a sequence of tokens (or token word) instead of generating it.
+    A generation guide defines the behavior of a finite-state machine that
+    guides a text generation procedure. Unlike the DFAs built from regular
+    expressions guides, it can also emit a `Write` instructions which tells
+    the model that it can append a sequence of tokens (or token word) instead
+    of generating it.
 
     """
-
     initial_state: Any
 
 
 class StopAtEOSGuide(Guide):
     """Guide to generate tokens until the EOS token has been generated."""
-
     final_state = 1
-    start_state = 0  # TODO: remove start_state, use only initial_state
     initial_state = 0
 
     def __init__(self, tokenizer: "Tokenizer"):
-        """Initialize the generation guide.
-
-        model
-            The logit generator used to generate the next token.
+        """
+        Parameters
+        ----------
+        tokenizer
+            The tokenizer used to convert tokens to ids.
 
         """
         self.eos_token_id = tokenizer.eos_token_id
         self.vocabulary = tokenizer.vocabulary.values()
 
     def get_next_instruction(self, state: int) -> Instruction:
+        """Return the next instruction.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        Instruction
+            An `Instruction` instance.
+
+        """
         if self.is_final_state(state):
             return Write([self.eos_token_id])
         return Generate(None)
 
     def get_next_state(self, state: int, token_id: int) -> int:
+        """Return the next state.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+        token_id
+            The id of the token that was just generated.
+
+        Returns
+        -------
+        int
+            The next state.
+
+        """
         if token_id == self.eos_token_id or state == self.final_state:
             return self.final_state
 
         return self.initial_state
 
-    def is_final_state(self, state: int):
+    def is_final_state(self, state: int) -> bool:
+        """Return whether the given state is a final state.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether the given state is a final state.
+
+        """
         return state == self.final_state
 
-    def copy(self):
+    def copy(self) -> "StopAtEOSGuide":
+        """Return itself as there is no need to copy."""
         return self
 
 
+@cache()
 def cached_create_states_mapping(regex_string, tokenizer, *args, **kwargs):
-    return uncached_create_states_mapping(regex_string, tokenizer, *args, **kwargs)
+    """Wrap the uncached create_states_mapping function in a cache."""
+    return uncached_create_states_mapping(
+        regex_string, tokenizer, *args, **kwargs
+    )
 
 
 class RegexGuide(CoreRegexGuide):
-    """
-    Guide to generate text in the language of a regular expression.
-    CoreRegexGuide with outlines cache
+    """Guide to generate text in the language of a regular expression.
+
+    This class is a wrapper around the CoreRegexGuide class that adds a cache
+    to the create_states_mapping function.
+
     """
 
     @classmethod
@@ -89,6 +141,23 @@ def from_regex(
         tokenizer,
         **kwargs,
     ):
+        """Create a RegexGuide from a regular expression.
+
+        Parameters
+        ----------
+        regex_string
+            The regular expression to generate text from.
+        tokenizer
+            The tokenizer to use to convert tokens to ids.
+        kwargs
+            Additional keyword arguments to pass to the CoreRegexGuide constructor.
+
+        Returns
+        -------
+        RegexGuide
+            A RegexGuide instance.
+
+        """
         return super().from_regex(
             regex_string,
             tokenizer,
@@ -101,15 +170,25 @@ def from_regex(
 
 
 class CFGGuide(Guide):
-    """Guide to generate text that is in the language of a context-free Lark grammar."""
+    """Guide to generate text that is in the language of a context-free Lark
+    grammar.
+
+    """
 
-    def __init__(self, cfg_string: str, tokenizer):
+    def __init__(self, cfg_string: str, tokenizer: "Tokenizer"):
         """
-        Construct the PartialLark parser and set the empty initial_state (PartialParserState)
+        Parameters
+        ----------
+        cfg_string
+            The context-free grammar to generate text from.
+        tokenizer
+            The tokenizer to use to convert tokens to ids.
+
         """
         warnings.warn(
-            "Outlines' public *community-contributed* CFG structured generation is experimental. "
-            "Please review https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
+            "Outlines' public *community-contributed* CFG structured generation "
+            "is experimental. Please review "
+            "https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
         )
 
         self.cfg_string = cfg_string
@@ -142,25 +221,32 @@ def get_next_instruction(self, state: CFGState) -> Instruction:
 
         Returns
         -------
-        A `Generate` instance that contains the model and the allowed token ids.
+        Instruction
+            A `Generate` instance that contains the model and the allowed token
+            ids.
 
         """
+        import torch
 
         if state.parser_state is None:
             return Write(torch.tensor([self.eos_token_id]))
 
         valid_tokens = list(
-            self.iter_valid_token_ids(state, self.tokenizer.vocabulary.values())
+            self.iter_valid_token_ids(
+                state, self.tokenizer.vocabulary.values()
+            )
         )
+
         if len(valid_tokens) == 1:
             return Write(torch.tensor(valid_tokens))
+
         return Generate(torch.tensor(valid_tokens))
 
     def iter_valid_token_ids(
-        self, state: CFGState, candidate_token_ids: list
+        self, state: CFGState, candidate_token_ids: ValuesView[int]
     ) -> Generator[int, None, None]:
-        """
-        Iterate over the given token_ids and yield those that are valid for the current parser state.
+        """Iterate over the given token_ids and yield those that are valid for
+        the current parser state.
 
         Parameters
         ----------
@@ -173,11 +259,8 @@ def iter_valid_token_ids(
         ------
         int
             Valid token ids.
-        """
-        if state.parser_state is None:
-            yield self.eos_token_id
-            return
 
+        """
         for token_id in candidate_token_ids:
             if token_id == self.eos_token_id:
                 if self.can_terminate_state(state):
@@ -196,9 +279,10 @@ def iter_valid_token_ids(
                     pass
 
     def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
-        """
-        Update the state of the guide.
-        Decode the token_id, and calculate the new parser_state with the token applied.
+        """Update the state of the guide.
+
+        Decode the token_id, and calculate the new parser_state with the token
+        applied.
 
         Parameters
         ----------
@@ -209,7 +293,8 @@ def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
 
         Returns
         -------
-        The guides new PartialParserState
+        CFGState
+            The guides new PartialParserState
 
         """
         if state.parser_state is None or token_id == self.eos_token_id:
@@ -221,7 +306,8 @@ def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
     def _get_parser_state_token_applied(
         self, state: CFGState, token_id: int
     ) -> PartialParserState:
-        """
+        """Apply the given token_id to the parser state.
+
         Don't mutate `parser_state`, copy to protect
 
         Get the token string
@@ -229,6 +315,19 @@ def _get_parser_state_token_applied(
           - else: normalized (with possibly leading whitespace)
 
         Don't allow empty ("") tokens, raise ValueError
+
+        Parameters
+        ----------
+        state
+            The guide's current PartialParserState, or None if complete
+        token_id
+            The id of the token that was just generated.
+
+        Returns
+        -------
+        PartialParserState
+            The parser state with the token applied.
+
         """
         parser_state = copy.copy(state.parser_state)  # prevent side effects
 
@@ -252,12 +351,38 @@ def _get_parser_state_token_applied(
         return parser_state
 
     def is_final_state(self, state: CFGState) -> bool:
-        # TODO: remove this method, use can_terminate_state and must_terminate_state
-        # here and in RegexGuide per https://github.com/dottxt-ai/outlines/issues/885
+        """Return whether the given state is a final state.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether the given state is a final state.
+
+        """
+        # TODO: remove this method, use can_terminate_state and
+        # must_terminate_state here and in RegexGuide per
+        # https://github.com/dottxt-ai/outlines/issues/885
         return self.can_terminate_state(state)
 
     def can_terminate_state(self, state: CFGState) -> bool:
-        """Generation is allowed to terminate"""
+        """Return whether generation is allowed to terminate.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether generation is allowed to terminate.
+
+        """
         if state.parser_state is not None:
             try:
                 copy.copy(state.parser_state).feed_eof()
@@ -266,11 +391,32 @@ def can_terminate_state(self, state: CFGState) -> bool:
         return True
 
     def must_terminate_state(self, state: CFGState) -> bool:
-        """Generation must terminate, no legal continuations"""
-        return state.parser_state is None or set(state.parser_state.accepts()).issubset(
-            {"$END"}
+        """Indicate whether generation must terminate as there are no legal
+        continuations.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether generation must terminate.
+
+        """
+        return (
+            state.parser_state is None or
+            set(state.parser_state.accepts()).issubset({"$END"})
         )
 
     def copy(self) -> "CFGGuide":
-        """Create a copy of the Guide."""
+        """Create a copy of the Guide.
+
+        Returns
+        -------
+        CFGGuide
+            A copy of the Guide.
+
+        """
         return CFGGuide(self.cfg_string, self.tokenizer)
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 8ce69c32a..f1a55b902 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -1,4 +1,4 @@
-"""
+"""Logits processors for structured generation.
  _______________________________
 / Don't want to self-host?       \
 \\ Try .json at http://dottxt.co /
@@ -27,79 +27,90 @@
 import math
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
 
-import torch
-from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel
+from outlines_core.fsm.json_schema import build_regex_from_schema
 
-from outlines.fsm.guide import CFGGuide, Guide, RegexGuide
-from outlines.fsm.json_schema import convert_json_schema_to_str
-
-from .base_logits_processor import OutlinesLogitsProcessor
+from outlines.processors.base_logits_processor import (
+    OutlinesLogitsProcessor,
+    TensorType
+)
+from outlines.processors.guide import (
+    CFGGuide,
+    Guide,
+    RegexGuide
+)
+from outlines.types import JsonSchema
 
 if TYPE_CHECKING:
     from outlines.models.tokenizer import Tokenizer
 
 
 class GuideLogitsProcessor(OutlinesLogitsProcessor):
-    """Bias generation using a finite
+    """Bias generation using a guide.
 
     Attributes
     ----------
     tokenizer
-        The tokenizer used to convert tokens to ids.
+        The outlines tokenizer used to convert tokens to ids.
     guide
-        The `outlines.fsm.Guide` which is used to bias the logits.
+        The outlines guide used to bias the logits.
     """
-
     tokenizer: "Tokenizer"
     guide: Guide
     _guide_states: Dict[int, Any]
     _seq_start_idx: Optional[int]
 
-    def __init__(self, tokenizer: "Tokenizer", guide: Guide):
-        """A Guide-based logits processor.
-
+    def __init__(
+        self, tokenizer: "Tokenizer", guide: Guide, tensor_library_name: str
+    ):
+        """
         Parameters
         ----------
         tokenizer
             The tokenizer used to convert tokens to ids.
         guide
-            The `outlines.fsm.Guide. which is used to bias the logits.
+            The `outlines.processors.guide.Guide` that is used to bias the
+            logits.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
+
         """
+        super().__init__(tensor_library_name=tensor_library_name)
         self.tokenizer = tokenizer
         self.guide = guide
         self._guide_states = {hash(tuple([])): self.guide.initial_state}
         self._seq_start_idx = None
 
     def process_logits(
-        self, input_ids: torch.LongTensor, logits: torch.FloatTensor
-    ) -> torch.Tensor:
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
         """Use the Guide to bias the logits before sampling the next token.
 
         Parameters
         ----------
         input_ids
-            The input token ids.
+            The ids of the tokens of the existing sequences.
         logits
-            The logits.
+            The logits for the current generation step.
 
         Returns
         -------
-        torch.Tensor
+        TensorType
             The biased logits.
+
         """
         if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0])
+            self._seq_start_idx = len(input_ids[0]) # type: ignore
 
         sequence_states: List[int] = []  # vector of states corresponding to `input_ids`
 
-        for seq_ids in input_ids:
+        for seq_ids in input_ids: # type: ignore
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(gen_ids.tolist()))
+            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
 
             if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
-                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
+                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
+                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])
@@ -110,136 +121,165 @@ def process_logits(
             allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
             allowed_tokens_batch.append(allowed_tokens)
             batch_indices.append(
-                torch.full_like(allowed_tokens, i)
+                self.tensor_adapter.full_like(allowed_tokens, i)
             )  # Store batch index for each allowed token
 
-        allowed_tokens_concat = torch.cat(allowed_tokens_batch).to(logits.device)
-        batch_indices_concat = torch.cat(batch_indices).to(logits.device)
-
-        mask = torch.ones_like(logits, dtype=torch.bool)
+        device = self.tensor_adapter.get_device(logits)
+        allowed_tokens_concat = self.tensor_adapter.to_device(
+            self.tensor_adapter.concatenate(allowed_tokens_batch),
+            device
+        )
+        batch_indices_concat = self.tensor_adapter.to_device(
+            self.tensor_adapter.concatenate(batch_indices),
+            device
+        )
+
+        mask = self.tensor_adapter.boolean_ones_like(logits)
         mask[batch_indices_concat, allowed_tokens_concat] = False
-        logits.masked_fill_(mask, float("-inf"))
+        logits = self.tensor_adapter.apply_mask(logits, mask, float("-inf"))
 
         return logits
 
     def copy(self) -> "GuideLogitsProcessor":
         """Return a copy of the logits processor."""
-        return GuideLogitsProcessor(tokenizer=self.tokenizer, guide=self.guide.copy())
+        return GuideLogitsProcessor(
+            tokenizer=self.tokenizer,
+            guide=self.guide.copy(),
+            tensor_library_name=self.tensor_adapter.library_name
+        )
 
 
 class RegexLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a regular expression.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    guide
-        The `outlines.fsm.RegexGuide. which is used to bias the logits.
-    """
-
-    def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
-        """Compile the RegexGuide that drives the regex-guided generation.
+    """Bias generation based on a regular expression."""
+    guide: RegexGuide
 
+    def __init__(
+        self,
+        regex_string: str,
+        tokenizer: "Tokenizer",
+        tensor_library_name: str,
+    ):
+        """
         Parameters
         ----------
         regex_string
-            A string that represents a regular expression
+            A string that represents a regular expression.
         tokenizer
-            An Outlines tokenizer
+            An Outlines tokenizer.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
+
         """
+        # Build a guide from the regex string and then pass it to the
+        # GuideLogitsProcessor superclass.
         guide = RegexGuide.from_regex(regex_string, tokenizer)
-        super().__init__(tokenizer=tokenizer, guide=guide)
+        super().__init__(tokenizer=tokenizer, guide=guide, tensor_library_name=tensor_library_name)
 
 
 class JSONLogitsProcessor(RegexLogitsProcessor):
-    """Bias generation based on a JSON schema.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    guide
-        The `outlines.fsm.RegexGuide. which is used to bias the logits.
-    """
-
+    """Bias generation based on a JSON schema."""
     def __init__(
         self,
         schema: Union[dict, Type[BaseModel], str],
         tokenizer: "Tokenizer",
+        tensor_library_name: str,
         whitespace_pattern: Optional[str] = None,
     ):
-        """Compile the Guide that drives the JSON-guided generation.
-
+        """
         Parameters
         ----------
         schema
             A JSON schema that encodes the structure we want the model to generate.
         tokenizer
             The tokenizer used to convert tokens to ids.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
         whitespace_pattern
             Pattern to use for JSON syntactic whitespace (doesn't impact string
             literals). For example, to allow only a single space or newline with
-            `whitespace_pattern=r"[\n ]?"`
+            `whitespace_pattern=r"[\n ]?"`.
+
         """
-        schema_str = convert_json_schema_to_str(json_schema=schema)
+        # Convert the JSON schema into a regex string and then pass it to the
+        # RegexLogitsProcessor superclass.
+        schema_str = JsonSchema(schema).schema
         regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(regex_string=regex_string, tokenizer=tokenizer)
+        super().__init__(
+            regex_string=regex_string,
+            tokenizer=tokenizer,
+            tensor_library_name=tensor_library_name
+        )
 
 
 class CFGLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a context-free grammar.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    guide
-        The `outlines.fsm.CFGGuide. which is used to bias the logits.
-    """
-
+    """Bias generation based on a context-free grammar."""
     guide: CFGGuide
 
-    def __init__(self, cfg_str: str, tokenizer: "Tokenizer"):
-        """Compile the CFGGuide that drives the CFG-guided generation.
-
+    def __init__(
+        self, cfg_str: str, tokenizer: "Tokenizer", tensor_library_name: str
+    ):
+        """
         Parameters
         ----------
         cfg_str
-            A string that represents a grammar
+            A string that represents a grammar.
         tokenizer
             The tokenizer used to convert tokens to ids.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
+
         """
+        # Build a guide from the CFG string and then pass it to the
+        # GuideLogitsProcessor superclass.
         cfg_guide = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
-        super().__init__(tokenizer=tokenizer, guide=cfg_guide)
+        super().__init__(
+            tokenizer=tokenizer,
+            guide=cfg_guide,
+            tensor_library_name=tensor_library_name
+        )
 
     def process_logits(
-        self, input_ids: torch.LongTensor, logits: torch.Tensor
-    ) -> torch.Tensor:
-        """Same behavior as GuideLogitsProcessor, but uses rejection sampling"""
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Same behavior as GuideLogitsProcessor, but uses rejection
+        sampling.
+
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences.
+        logits
+            The logits for the current generation step.
+
+        Returns
+        -------
+        TensorType
+            The biased logits.
+
+        """
         if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0])
+            self._seq_start_idx = len(input_ids[0]) # type: ignore
 
         sequence_states: List = []  # vector of states corresponding to `input_ids`
 
-        for seq_ids in input_ids:
+        for seq_ids in input_ids: # type: ignore
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(gen_ids.tolist()))
+            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
 
-            if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
-                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
+            if curr_state_key not in self._guide_states: # pragma: no cover
+                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
+                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])
 
-        mask = torch.full_like(logits, -math.inf)
+        mask = self.tensor_adapter.full_like(logits, -math.inf)
         for i, guide_state in enumerate(sequence_states):
             first_legal_token = next(
                 self.guide.iter_valid_token_ids(
-                    guide_state, torch.argsort(logits[i], descending=True)
+                    guide_state, self.tensor_adapter.argsort_descending(logits[i]) # type: ignore
                 )
             )
-            mask[i, [first_legal_token]] = logits[i, [first_legal_token]]
+            mask[i, [first_legal_token]] = logits[i, [first_legal_token]] # type: ignore
 
         return mask
diff --git a/outlines/processors/tensor_adapters/__init__.py b/outlines/processors/tensor_adapters/__init__.py
new file mode 100644
index 000000000..8e3afe7f6
--- /dev/null
+++ b/outlines/processors/tensor_adapters/__init__.py
@@ -0,0 +1,26 @@
+"""Library specific objects to manipulate tensors."""
+
+from typing import Union
+
+from .jax import JAXTensorAdapter
+from .mlx import MLXTensorAdapter
+from .numpy import NumpyTensorAdapter
+from .tensorflow import TensorFlowTensorAdapter
+from .torch import TorchTensorAdapter
+
+
+tensor_adapters = {
+    "jax": JAXTensorAdapter,
+    "mlx": MLXTensorAdapter,
+    "numpy": NumpyTensorAdapter,
+    "tensorflow": TensorFlowTensorAdapter,
+    "torch": TorchTensorAdapter,
+}
+
+TensorAdapterImplementation = Union[
+    JAXTensorAdapter,
+    MLXTensorAdapter,
+    NumpyTensorAdapter,
+    TensorFlowTensorAdapter,
+    TorchTensorAdapter,
+]
diff --git a/outlines/processors/tensor_adapters/base.py b/outlines/processors/tensor_adapters/base.py
new file mode 100644
index 000000000..ffaac58df
--- /dev/null
+++ b/outlines/processors/tensor_adapters/base.py
@@ -0,0 +1,257 @@
+"""Base class for tensor adapters."""
+
+from abc import ABC, abstractmethod
+from typing import TYPE_CHECKING, TypeVar, Any, Union
+
+if TYPE_CHECKING:
+    import torch
+
+TensorType = TypeVar('TensorType')
+
+
+class TensorAdapter(ABC):
+    """Abstract base class for tensor adapters.
+
+    This class defines the interface for tensor adapters that are used to
+    manipulate tensors in different libraries. Concrete implementations of
+    this class should provide specific implementations for each method as
+    well as providing a `library_name` attribute.
+
+    TODO: Update the version of outlines-core used to receive plain arrays
+    instead of torch tensors. In the meantime, implementations of this class
+    must make sure that their `full_like` and `concatenate` methods can
+    handle torch tensors.
+
+    """
+    library_name: str
+
+    @abstractmethod
+    def shape(self, tensor: TensorType) -> list[int]:
+        """Get the shape of the tensor.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to get the shape of.
+
+        Returns
+        -------
+        list[int]
+            The shape of the tensor. The list contains as many elements as
+            there are dimensions in the tensor.
+
+        """
+        ...
+
+    @abstractmethod
+    def unsqueeze(self, tensor: TensorType) -> TensorType:
+        """Add a dimension to the tensor at axis 0.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to add a dimension to.
+
+        Returns
+        -------
+        TensorType
+            The tensor with an additional dimension.
+
+        """
+        ...
+
+    @abstractmethod
+    def squeeze(self, tensor: TensorType) -> TensorType:
+        """Remove a dimension from the tensor at axis 0.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to remove a dimension from.
+
+        Returns
+        -------
+        TensorType
+            The tensor with one less dimension.
+
+        """
+        ...
+
+    @abstractmethod
+    def to_list(self, tensor: TensorType) -> list:
+        """Convert the tensor to a list.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to convert to a list.
+
+        Returns
+        -------
+        list
+            The tensor as a list.
+
+        """
+        ...
+
+    @abstractmethod
+    def to_scalar(self, tensor: TensorType) -> Any:
+        """Return the only element of the tensor.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to return the only element of.
+
+        Returns
+        -------
+        Any
+            The only element of the tensor.
+
+        """
+        ...
+
+    @abstractmethod
+    def full_like(self, tensor: "torch.Tensor", fill_value: Any) -> TensorType: # type: ignore
+        """Create a tensor with the same shape as the input tensor filled
+        with a scalar value.
+
+        ATTENTION: This method receives a torch tensor regardless of the
+        library used.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to create a new tensor with the same shape.
+        fill_value
+            The value to fill the new tensor with.
+
+        Returns
+        -------
+        TensorType
+            A tensor with the same shape as the input tensor filled with the
+            specified value.
+
+        """
+        ...
+
+    @abstractmethod
+    def concatenate(
+        self, tensors: list[Union["torch.Tensor", TensorType]]
+    ) -> TensorType:
+        """Concatenate a list of tensors along axis 0.
+
+        ATTENTION: This method can either receive a list of torch tensors or
+        a list of tensors from the library used.
+
+        Parameters
+        ----------
+        tensors
+            The list of tensors to concatenate.
+
+        Returns
+        -------
+        TensorType
+            The concatenated tensor.
+
+        """
+        ...
+
+    @abstractmethod
+    def get_device(self, tensor: TensorType) -> str:
+        """Get the name of the tensor's device.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to get the device of.
+
+        Returns
+        -------
+        str
+            The name of the tensor's device.
+
+        """
+        ...
+
+    @abstractmethod
+    def to_device(self, tensor: TensorType, device: str) -> TensorType:
+        """Move the tensor to a specified device.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to move to a specified device.
+        device
+            The name of the device to move the tensor to.
+
+        Returns
+        -------
+        TensorType
+            The tensor moved to the specified device.
+
+        """
+        ...
+
+    @abstractmethod
+    def boolean_ones_like(self, tensor: TensorType) -> TensorType:
+        """Create a boolean ones tensor with the same shape as the input
+        tensor.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to create a boolean ones tensor with the same shape.
+
+        Returns
+        -------
+        TensorType
+            A boolean ones tensor with the same shape as the input tensor.
+
+        """
+        ...
+
+    @abstractmethod
+    def apply_mask(
+        self, tensor: TensorType, mask: TensorType, value: Any
+    ) -> TensorType:
+        """Fill the elements of the tensor where the mask is True with the
+        specified value.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to fill.
+        mask
+            The mask to apply to the tensor.
+        value
+            The value to fill the tensor with.
+
+        Returns
+        -------
+        TensorType
+            The tensor with the mask applied.
+
+        """
+        ...
+
+    @abstractmethod
+    def argsort_descending(
+        self, tensor: TensorType
+    ) -> TensorType:
+        """Return the indices that would sort the tensor in descending order
+        along axis -1.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to sort.
+
+        Returns
+        -------
+        TensorType
+            The indices that would sort the tensor in descending order along
+            axis -1.
+
+        """
+        ...
diff --git a/outlines/processors/tensor_adapters/jax.py b/outlines/processors/tensor_adapters/jax.py
new file mode 100644
index 000000000..30cd2ecc9
--- /dev/null
+++ b/outlines/processors/tensor_adapters/jax.py
@@ -0,0 +1,50 @@
+"""Tensor adapter for the `jax` library."""
+
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class JAXTensorAdapter(TensorAdapter):
+    library_name = "jax"
+
+    def __init__(self):
+        import jax
+
+        self.jax = jax
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.jax.numpy.expand_dims(tensor, axis=0)
+
+    def squeeze(self, tensor):
+        return self.jax.numpy.squeeze(tensor, axis=0)
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        return self.jax.numpy.full_like(tensor, fill_value)
+
+    def concatenate(self, tensors):
+        return self.jax.numpy.concatenate(tensors, axis=0)
+
+    def to_device(self, tensor, device):
+        return tensor
+
+    def get_device(self, tensor):
+        return None
+
+    def boolean_ones_like(self, tensor):
+        return self.jax.numpy.ones_like(tensor, dtype=bool)
+
+    def apply_mask(self, tensor, mask, value):
+        result = tensor.copy()
+        result = result.at[mask].set(value)
+        return result
+
+    def argsort_descending(self, tensor):
+        return self.jax.numpy.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/mlx.py b/outlines/processors/tensor_adapters/mlx.py
new file mode 100644
index 000000000..5714808d4
--- /dev/null
+++ b/outlines/processors/tensor_adapters/mlx.py
@@ -0,0 +1,60 @@
+"""Tensor adapter for the `mlx` library."""
+
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class MLXTensorAdapter(TensorAdapter):
+    library_name = "mlx"
+
+    def __init__(self):
+        import mlx.core
+
+        self.mlx = mlx.core
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.mlx.expand_dims(tensor, 0)
+
+    def squeeze(self, tensor):
+        if tensor.shape[0] == 1:
+            return tensor[0]
+        return tensor
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        # Compatible with receiving a torch tensor
+        return self.mlx.full(tensor.shape, fill_value)
+
+    def concatenate(self, tensors):
+        # Can handle both torch and mlx tensors
+        return self.mlx.concatenate(
+            [
+                self.mlx.array(t) if not isinstance(t, self.mlx.array) else t
+                for t in tensors
+            ],
+            axis=0
+        )
+
+    def get_device(self, tensor):
+        return None
+
+    def to_device(self, tensor, device):
+        return tensor
+
+    def boolean_ones_like(self, tensor):
+        return self.mlx.ones(tensor.shape, dtype=self.mlx.bool_)
+
+    def apply_mask(self, tensor, mask, value):
+        result = tensor.astype(tensor.dtype)
+        result = self.mlx.where(mask, self.mlx.array(value), result)
+        return result
+
+    def argsort_descending(self, tensor):
+        return self.mlx.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/numpy.py b/outlines/processors/tensor_adapters/numpy.py
new file mode 100644
index 000000000..831220444
--- /dev/null
+++ b/outlines/processors/tensor_adapters/numpy.py
@@ -0,0 +1,50 @@
+"""Tensor adapter for the `numpy` library."""
+
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class NumpyTensorAdapter(TensorAdapter):
+    library_name = "numpy"
+
+    def __init__(self):
+        import numpy
+
+        self.numpy = numpy
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.numpy.expand_dims(tensor, axis=0)
+
+    def squeeze(self, tensor):
+        return self.numpy.squeeze(tensor, axis=0)
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        return self.numpy.full_like(tensor, fill_value)
+
+    def concatenate(self, tensors):
+        return self.numpy.concatenate(tensors, axis=0)
+
+    def get_device(self, tensor):
+        return None
+
+    def to_device(self, tensor, device):
+        return tensor
+
+    def boolean_ones_like(self, tensor):
+        return self.numpy.ones_like(tensor, dtype=bool)
+
+    def apply_mask(self, tensor, mask, value):
+        result = tensor.copy()
+        result[mask] = value
+        return result
+
+    def argsort_descending(self, tensor):
+        return self.numpy.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/tensorflow.py b/outlines/processors/tensor_adapters/tensorflow.py
new file mode 100644
index 000000000..382f847b6
--- /dev/null
+++ b/outlines/processors/tensor_adapters/tensorflow.py
@@ -0,0 +1,50 @@
+"""Tensor adapter for the `tensorflow` library."""
+
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class TensorFlowTensorAdapter(TensorAdapter):
+    library_name = "tensorflow"
+
+    def __init__(self):
+        import tensorflow as tf
+
+        self.tf = tf
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.tf.expand_dims(tensor, axis=0)
+
+    def squeeze(self, tensor):
+        return self.tf.squeeze(tensor, axis=0)
+
+    def to_list(self, tensor):
+        return tensor.numpy().tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.numpy().item()
+
+    def full_like(self, tensor, fill_value):
+        return self.tf.fill(self.tf.shape(tensor), fill_value)
+
+    def concatenate(self, tensors):
+        return self.tf.concat(tensors, axis=0)
+
+    def get_device(self, tensor):
+        return tensor.device
+
+    def to_device(self, tensor, device):
+        # TensorFlow handles device placement differently
+        with self.tf.device(device):
+            return self.tf.identity(tensor)
+
+    def boolean_ones_like(self, tensor):
+        return self.tf.ones_like(tensor, dtype=self.tf.bool)
+
+    def apply_mask(self, tensor, mask, value):
+        return self.tf.where(mask, self.tf.constant(value, dtype=tensor.dtype), tensor)
+
+    def argsort_descending(self, tensor):
+        return self.tf.argsort(tensor, direction='DESCENDING')
diff --git a/outlines/processors/tensor_adapters/torch.py b/outlines/processors/tensor_adapters/torch.py
new file mode 100644
index 000000000..4b624e960
--- /dev/null
+++ b/outlines/processors/tensor_adapters/torch.py
@@ -0,0 +1,48 @@
+"""Tensor adapter for the `torch` library."""
+
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class TorchTensorAdapter(TensorAdapter):
+    library_name = "torch"
+
+    def __init__(self):
+        import torch
+
+        self.torch = torch
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return tensor.unsqueeze(0)
+
+    def squeeze(self, tensor):
+        return tensor.squeeze(0)
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        return self.torch.full_like(tensor, fill_value)
+
+    def concatenate(self, tensors):
+        return self.torch.cat(tensors, dim=0)
+
+    def get_device(self, tensor):
+        return tensor.device
+
+    def to_device(self, tensor, device):
+        return tensor.to(device)
+
+    def boolean_ones_like(self, tensor):
+        return self.torch.ones_like(tensor, dtype=self.torch.bool)
+
+    def apply_mask(self, tensor, mask, value):
+        return self.torch.masked_fill(tensor, mask, value)
+
+    def argsort_descending(self, tensor):
+        return self.torch.argsort(tensor, descending=True)
diff --git a/outlines/release_note.md b/outlines/release_note.md
new file mode 100644
index 000000000..35812623b
--- /dev/null
+++ b/outlines/release_note.md
@@ -0,0 +1,392 @@
+# Release Note
+
+### Why a new major version?
+
+The v1 intends on making Outlines more closely focused on constrained generation. To do so, we delegate a wider range of tasks to the users and inference libraries. On top of making Outlines leaner, this design provides more flexibility to the users and let them use interfaces they are already familiar with.
+
+Our approach is inspired by the unix best practices — each element does one thing well, and we compose those functional elements.
+
+As this new version deprecates some previously available features of Outlines, we have written a migration guide that gives detailed information on how to upgrade your v0 code to v1.
+
+### Deprecated
+
+All deprecated features listed below will be removed in version 1.1.0. Until then, a warning will be displayed with information on how to migrate your code to v1.
+
+- The model loader functions from the `models` module (`transformers`, `openai`, etc.) have been deprecated. They are replaced by equivalent functions prefixed with `from_` such as `from_transformers`, `from_openai`, etc. The new loader functions accept different arguments compared to the old ones. They now typically require an instance of an engine/client from the associated inference library. This change was made to avoid duplicating inference library logic and to give users more control over inference engine/client initialization.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+```python
+# v0
+from outlines import models
+from transformers import BertForSequenceClassification, BertTokenizer
+
+model = models.transformers(
+    model_name="prajjwal1/bert-tiny",
+    model_class=BertForSequenceClassification,
+    tokenizer_class=BertTokenizer,
+    model_kwargs={"use_cache": False},
+    tokenizer_kwargs={"model_max_length": 512},
+)
+
+# v1
+import outlines
+from transformers import BertForSequenceClassification, BertTokenizer
+
+hf_model = BertForSequenceClassification.from_pretrained("prajjwal1/bert-tiny", use_cache=False)
+hf_tokenizer = BertTokenizer.from_pretrained("prajjwal1/bert-tiny", model_max_length=512)
+model = outlines.from_transformers(hf_model, hf_tokenizer)
+```
+
+- The `generate` module and the associated functions (`json`, `choice`…) have been deprecated. They are replaced by the `Generator` constructor. While you had to select the right generate function for your output type, you can now provide any output type supported by Outlines to the unique `Generator` object.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/generator)
+
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import generate, models
+
+class Character(BaseModel):
+	name: str
+
+model = models.openai("gpt-4o")
+generator = generate.json(model, Character)
+
+# v1
+from openai import OpenAI
+from pydantic import BaseModel
+from outlines import Generator, from_openai
+
+class Character(BaseModel):
+	name: str
+
+model = from_openai(OpenAI())
+generator = Generator(model, Character)
+```
+
+- The `TransformersVision` model has been deprecated. It's replaced by `TransformersMultiModal`, which is more general as it supports additional input types beyond images, such as audio. When calling it, instead of providing the prompt and image assets separately, both should now be included in a single dictionary. The model is loaded with `from_transformers` just like the `Transformers` model, but the second argument must be a processor instead of a tokenizer.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models/transformers_multimodal)
+
+
+```python
+# v0
+from io import BytesIO
+from urllib.request import urlopen
+from PIL import Image
+from transformers import LlavaForConditionalGeneration
+from outlines import models, generate
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+model = models.transformers_vision(
+    model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+    model_class=LlavaForConditionalGeneration,
+)
+generator = generate.text(model)
+result = generator(
+    "Describe the image <image>",
+    img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
+)
+
+# v1
+from io import BytesIO
+from urllib.request import urlopen
+from PIL import Image
+from transformers import LlavaForConditionalGeneration, AutoProcessor
+import outlines
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+model = outlines.from_transformers(
+	LlavaForConditionalGeneration.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration"),
+	AutoProcessor.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration")
+)
+image = img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
+result = model({"text": "Describe the image <image>", "images": image})
+```
+
+- The `Exllamav2` model has been deprecated without replacement because its interface is not fully compatible with Outlines. We had to implement cumbersome patching to make it work, so we decided to remove it entirely.
+
+- The `function` module and the associated `Function` class have been deprecated. They are replaced by the `Application` class, which serves a similar purpose to `Function`. There are two notable differences: an `Application` is not initialized with a model (a model must be provided when calling the object), and template variables must be provided in a dictionary instead of as keyword arguments when calling the `Application`.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/utility/application)
+
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import Function, Template
+
+class Character(BaseModel):
+	name: str
+
+template = Template.from_string("Create a {{ gender }} character.")
+fn = Function(template, Character, "hf-internal-testing/tiny-random-GPTJForCausalLM")
+response = fn(gender="female")
+
+# v1
+from pydantic import BaseModel
+from outlines import Application, Template, from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+class Character(BaseModel):
+	name: str
+
+model = from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+template = Template.from_string("Create a {{ gender }} character.")
+app = Application(template, Character)
+response = app(model, {"gender": "female"})
+```
+
+- The `samplers` module and the associated objects (`multinomial`, `greedy`…) have been deprecated. You should now use the inference arguments specific to the inference library your  model is based on to control the sampling.
+
+```python
+# v0
+from outlines import generate, models, samplers
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model, samplers.beam_search(2))
+response = generator("Write a short story about a cat", max_tokens=10)
+
+# v1
+from outlines import Generator, from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+response = model("Write a short story about a cat", num_beams=2)
+```
+
+- The `load_lora` methods on the `VLLM` and `LlamaCpp` models have been deprecated. You should now load through the `Llama` instance provided when initializing the model in the case of the `LlamaCpp` model, and provide it as a keyword argument when calling the model in the case of the `VLLM` model.
+
+```python
+# v0
+from outlines import models
+from vllm import LLM
+
+model = models.vllm("erwanf/gpt2-mini")
+model.load_lora("path/to/lora/file")
+response = model("Write a short story about a cat.")
+
+#v1
+from outlines import from_vllm
+from vllm import LLM
+from vllm.lora.request import LoRARequest
+
+model = from_vllm(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
+response = model("Write a short story about a cat.", lora_request=lora_request)
+```
+
+### Modified
+
+Some objects are maintained, but their interface or behavior has been modified.
+
+- The interface of `Model` classes (`Transformers`, `OpenAI`, etc.) has been significantly modified. Models can now be called directly with a prompt and an output type without having to create a generator first. Additionally, all models have a `stream` method that can be invoked directly by the user.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import generate, models
+
+class Character(BaseModel):
+		name: str
+
+model = models.openai("gpt-4o")
+generator = generate.json(model, Character)
+result = generator("Create a character")
+
+# v1
+from openai import OpenAI
+from pydantic import BaseModel
+from outlines import from_openai
+
+class Character(BaseModel):
+	name: str
+
+model = from_openai(OpenAI(), "gpt-4o")
+result = model("Create a character", Character)
+```
+
+- The interface of the `__init__` method of the `OpenAI` model class has been modified. While it previously accepted a client and an `OpenAIConfig` object instance, it now accepts a client and a model name. The inference arguments from the config object should now be specified when calling the model to more closely align with the OpenAI Python library's functionality. If you provide an `OpenAIConfig` instance when initializing the model, a deprecation warning will appear and your model will behave like a v0 model.
+We recommend using the `from_openai` function instead of initializing models directly.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models/openai)
+
+
+```python
+# v0
+from outlines.models.openai import OpenAI, OpenAIConfig
+from openai import OpenAI as OpenAIClient
+
+model = OpenAI(
+	OpenAIClient(),
+	OpenAIConfig(model="gpt-4o", stop=["."])
+)
+
+# v1
+import outlines
+from openai import OpenAI
+
+model = outlines.from_openai(OpenAIClient(), "gpt-4o")
+```
+
+- The return type of text generation is now consistently a string (or list/lists of strings for multiple samples or batching). In v0, Outlines automatically cast the inference result into the type provided by the user for constrained generation, but we have removed this behavior. This change was made to create more consistent behavior and to give users more freedom in deciding how to handle the generation result.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import generate, models
+
+class Character(BaseModel):
+	name: str
+
+model = models.openai("gpt-4o")
+generator = generate.json(model, Character)
+result = generator("Create a character")
+print(result) # name='James'
+
+# v1
+import openai
+from pydantic import BaseModel
+from outlines import from_openai
+
+class Character(BaseModel):
+		name: str
+
+model = from_openai(OpenAI())
+result = model("Create a character", Character)
+print(result) # {"name": "James"}
+print(Character.model_validate_json(result)) # name='James'
+```
+
+- While Outlines was trying to standardize inference argument names across models in v0, we decided to stop doing so and to directly pass on the inference arguments provided by the user to the inference engine/client. Our objective is to let the user use all arguments they are accustomed to with their inference library instead of having to learn Outlines-defined arguments. The deprecation of the `samplers` mentioned above is a part of this change of approach.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+```python
+# v0
+from outlines import generate, models
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+result = generator("Create a character", max_tokens=256, stop_at=".")
+
+# v1
+from outlines import from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+	AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+	AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+result = model("Create a character", max_new_tokens=256, stop_strings=".")
+```
+
+### Added features
+
+- There are 8 additional models available. All of them are loaded with an associated `from_` function that accepts an inference engine/client instance.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+    - `Dottxt`
+    - `Anthopic`
+    - `Gemini`
+    - `Ollama`
+    - `SGLang`
+    - `TGI`
+    - `TransformersMultiModel`
+    - `VLLM`
+- Some server-based models now have an async version. To create an async model, just provide an async client instance when using the loader function. The async models are the following.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+    - `AsyncSGLang`
+    - `AsyncTGI`
+    - `AsyncVLLM`
+
+```python
+import outlines
+from huggingface_hub import AsyncInferenceClient
+
+async_model = outlines.from_tgi(AsyncInferenceClient("http://localhost:11434"))
+```
+
+- As explained previously, the `Generator` constructor has been added. It accepts a model and an output type as arguments and returns a generator object that can be used to generate text by providing a prompt and inference arguments. The interest of a generator is that it's reusable such that the user does not have to specify the output type they want each time and the output type compilation (when applicable) happens only once.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/generator)
+
+```python
+# direct model calling
+from typing import Literal
+from outlines import from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+		AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
+		AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
+)
+result = model("Pizza or burger", Literal["pizza", "burger"])
+
+# using a generator
+from outlines import Generator, from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+		AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+		AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+generator = Generator(model, Literal["pizza", "burger"])
+result = generator("Pizza or burger")
+```
+
+- As explained previously, the `Application` class has been added. An `Application` is initialized with a prompt template and an output type. The application object returned can then be called with a model, a dictionary containing values for the template variables and inference arguments. The objective of this object is to let users easily switch from a model to another for a given set of prompt and output type.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/utility/application)
+
+```python
+from pydantic import BaseModel
+from outlines import Application, Template
+
+class Character(BaseModel):
+	name: str
+
+template = Template.from_string("Create a {{ gender }} character.")
+app = Application(template, Character)
+response = app(model, {"gender": "female"})
+```
+
+- The regex DSL and the associated `Term` classes and functions have been added. Terms (`Regex`, `String`…) can be used as output types to generate text with models or generators (they are turned into a regex). The term functions (`either`, `optional`, `at_least`…) are useful to build more complex regex patterns by combining terms. On top of the objects related to regex patterns, there are also 3 terms that are intended to be used by themselves as output types: `JsonSchema`, `CFG` and `FSM`.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/ouput_types)
+
+```python
+# term used directly as an output type
+from outlines import from_transformers
+from outlines.types import JsonSchema
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+		AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
+		AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
+)
+json_schema = '{"type": "object", "properties": {"answer": {"type": "number"}}}'
+result = model("What's 2 + 2? Respond in a json", JsonSchema(json_schema))
+
+# creating a complex regex pattern
+from outlines import from_transformers
+from outlines.types import at_least, either, integer, optional
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+	AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
+	AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
+)
+regex_term = "I have " + integer + either("dog", "cat") + optional("s")
+result = model("How many pets do you have", regex_term)
+```
diff --git a/outlines/samplers.py b/outlines/samplers.py
deleted file mode 100644
index 3fef673b1..000000000
--- a/outlines/samplers.py
+++ /dev/null
@@ -1,353 +0,0 @@
-import math
-from dataclasses import dataclass
-from typing import TYPE_CHECKING, Callable, Optional, Protocol, Tuple
-
-if TYPE_CHECKING:
-    import torch
-
-
-class Sampler(Protocol):
-    samples: int
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        rng: "torch.Generator",
-    ) -> "torch.DoubleTensor": ...
-
-
-@dataclass(frozen=True)
-class SamplingParameters:
-    """Sampling parameters available in Outlines."""
-
-    sampler: str
-    num_samples: int = 1
-    top_p: Optional[float] = None
-    top_k: Optional[int] = None
-    temperature: Optional[float] = None
-
-
-class GreedySampler:
-    """Greedy Sampling algorithm.
-
-    Greedy sampling consists in choosing the token with the largest
-    likelihood at every step.
-
-    We don't allow more than one sample. We could attribute this a meaning, for
-    instance the k-th sample represents the k-th most likely token. In which
-    case it would be equivalent to beam search without the sequence weights.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(self):
-        self.samples = 1
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        _,
-    ) -> "torch.DoubleTensor":
-        """Call the greedy sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
-        next_token_ids = torch.argmax(logprobs, dim=-1, keepdim=True)
-
-        ancestors = torch.arange(
-            next_token_logits.shape[0], device=next_token_logits.device
-        )
-        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
-
-        return next_token_ids, ancestors, weights
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("greedy", self.samples, None, None, 0.0)
-
-
-greedy = GreedySampler
-
-
-class MultinomialSampler:
-    """Multinomial sampling algorithm.
-
-    Multinomial sampling consists in randomly sampling the next token assuming
-    its distribution is a Categorical distribution parametrized by the
-    next-token logits.
-
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(
-        self,
-        samples: int = 1,
-        *,
-        top_k: Optional[int] = None,
-        top_p: Optional[float] = None,
-        temperature: Optional[float] = None,
-    ):
-        self.samples = samples
-        self.top_k = top_k
-        self.top_p = top_p
-        self.temperature = temperature
-
-        self.logits_processors = []
-        if top_k is not None:
-            self.logits_processors.append(keep_top_k_logits(top_k))
-        elif top_p is not None:
-            self.logits_processors.append(keep_top_p_logits(top_p))
-
-        if temperature is not None:
-            self.logits_processors.append(rescale_logits(temperature))
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        rng: "torch.Generator",
-    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
-        """Call the multinomial sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        altered_next_token_logits = next_token_logits
-        for logit_processor in self.logits_processors:
-            altered_next_token_logits = logit_processor(next_token_logits)
-
-        probs = torch.nn.functional.softmax(altered_next_token_logits, dim=-1)
-        next_token_ids = torch.multinomial(probs, num_samples=1, generator=rng)
-
-        logprobs = torch.nn.functional.log_softmax(altered_next_token_logits, dim=-1)
-        ancestors = torch.arange(
-            altered_next_token_logits.shape[0], device=next_token_logits.device
-        )
-        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
-
-        return next_token_ids, ancestors, weights
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters(
-            "multinomial",
-            self.samples,
-            self.top_p,
-            self.top_k,
-            self.temperature,
-        )
-
-
-multinomial = MultinomialSampler
-
-
-def keep_top_k_logits(k: int) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that masks logits values smaller than the top `k` ones.
-
-    Parameters
-    ----------
-    k
-        The ranking below which logit values are replaced by `-math.inf`.
-
-    """
-    import torch
-
-    if not isinstance(k, int) or k < 1:
-        raise ValueError(f"`k` must be a strictly positive integers, got {k} instead.")
-
-    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
-        num_to_keep = min(k, logits.size(-1))
-        mask_idx = logits < torch.topk(logits, num_to_keep)[0][..., -1, None]
-        return logits.masked_fill(mask_idx, -math.inf)
-
-    return logits_processor
-
-
-def keep_top_p_logits(p: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that masks the lowest probability tokens whose
-    cumulative probability is below a certain threshold.
-
-    Parameters
-    ----------
-    p
-        The value of the threshold. We keep the highest probability tokens whose
-        cumulative distribution is greater than or equal to `p` and mask the
-        others. Its value must be between 0 (excluded) and 1 (included).
-
-    """
-    import torch
-
-    if p <= 0.0 or p > 1.0:
-        raise ValueError(
-            f"`p` must be a floating point number between 0 (excluded) and 1 (included), got {p} instead."
-        )
-
-    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
-        sorted_logits, sorted_idx = torch.sort(logits, descending=False)
-        cumulative_probabilties = torch.nn.functional.softmax(
-            sorted_logits, dim=-1
-        ).cumsum(dim=-1)
-
-        sorted_masked_idx = cumulative_probabilties <= (1 - p)
-        mask_idx = torch.scatter(sorted_masked_idx, 1, sorted_idx, sorted_masked_idx)
-        return logits.masked_fill(mask_idx, -math.inf)
-
-    return logits_processor
-
-
-def rescale_logits(temperature: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that rescales the token probabilities exponentially.
-
-    Parameters
-    ----------
-    temperature
-        The value by which we rescale the logits.
-
-    """
-
-    if not isinstance(temperature, float) or temperature < 0.0:
-        raise ValueError(
-            f"`temperature` must be a strictly positive floating point number, got {temperature} instead."
-        )
-    elif temperature == 0.0:
-        raise ValueError(
-            "Please use the greedy sampler instead of setting the temperature to 0."
-        )
-
-    def logits_processor(logits: "torch.Tensor") -> "torch.Tensor":
-        return logits / temperature
-
-    return logits_processor
-
-
-class BeamSearchSampler:
-    """Beam Search sampling algorithm.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence. Equivalent to the
-        number of beams.
-    """
-
-    def __init__(self, beams: int = 1):
-        self.samples = beams
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        _,
-    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
-        """Call the beam search sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
-        weights = logprobs + sequence_weights.unsqueeze(1).expand_as(next_token_logits)
-
-        # Flatten scores to (n_batch, n_samples * vocab_size)
-        # and find the top-k weights for each batch.
-        batch_size = next_token_logits.shape[0] // self.samples
-        vocab_size = next_token_logits.shape[-1]
-        weights = weights.view(batch_size, self.samples * vocab_size)
-
-        # If the weights are all equal to 0 we are at the beginning of the search
-        # and thus only need to sample from one set of token logits for each
-        # batch.
-        if torch.all(sequence_weights == 0):
-            weights = weights[:, :vocab_size]
-
-        weights, indices = torch.topk(
-            weights, self.samples, dim=1, largest=True, sorted=True
-        )
-
-        ancestors = torch.div(indices, vocab_size, rounding_mode="floor")
-        next_token_ids = indices % vocab_size
-
-        # Re-shape the weights, next_token_ids and ancestors to (n_batch * n_samples, 1)
-        first_batch_idx = torch.arange(
-            0, batch_size * self.samples, self.samples, device=next_token_logits.device
-        ).unsqueeze(1)
-        ancestors = ancestors + first_batch_idx
-
-        ancestors = ancestors.view(self.samples * batch_size)
-        weights = weights.view(self.samples * batch_size)
-        next_token_ids = next_token_ids.view(self.samples * batch_size, 1)
-
-        return next_token_ids, ancestors, weights
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
-
-
-beam_search = BeamSearchSampler
diff --git a/outlines/serve/__init__.py b/outlines/serve/__init__.py
deleted file mode 100644
index e69de29bb..000000000
diff --git a/outlines/serve/serve.py b/outlines/serve/serve.py
deleted file mode 100644
index 61d3ed7af..000000000
--- a/outlines/serve/serve.py
+++ /dev/null
@@ -1,140 +0,0 @@
-#  _______________________________
-# / Don't want to self-host?      \
-# \ Try .json at http://dottxt.co /
-#  -------------------------------
-#        \   ^__^
-#         \  (oo)\_______
-#            (__)\       )\/\
-#                ||----w |
-#                ||     ||
-#
-#
-# Copyright 2024- the Outlines developers
-# Copyright 2023 the vLLM developers
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-
-#     http://www.apache.org/licenses/LICENSE-2.0
-
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-import argparse
-import json
-from typing import AsyncGenerator
-
-import uvicorn
-from fastapi import FastAPI, Request
-from fastapi.responses import JSONResponse, Response, StreamingResponse
-from vllm.engine.arg_utils import AsyncEngineArgs
-from vllm.engine.async_llm_engine import AsyncLLMEngine
-from vllm.sampling_params import SamplingParams
-from vllm.utils import random_uuid
-
-from outlines.models.vllm import adapt_tokenizer
-from outlines.processors import JSONLogitsProcessor, RegexLogitsProcessor
-
-TIMEOUT_KEEP_ALIVE = 5  # seconds.
-TIMEOUT_TO_PREVENT_DEADLOCK = 1  # seconds.
-app = FastAPI()
-engine = None
-tokenizer = None
-
-
-@app.get("/health")
-async def health() -> Response:
-    """Health check."""
-    return Response(status_code=200)
-
-
-@app.post("/generate")
-async def generate(request: Request) -> Response:
-    """Generate completion for the request.
-
-    The request should be a JSON object with the following fields:
-    - prompt: the prompt to use for the generation.
-    - schema: the JSON schema to use for the generation (if regex is not provided).
-    - regex: the regex to use for the generation (if schema is not provided).
-    - stream: whether to stream the results or not.
-    - other fields: the sampling parameters (See `SamplingParams` for details).
-    """
-    assert engine is not None
-
-    request_dict = await request.json()
-    prompt = request_dict.pop("prompt")
-    stream = request_dict.pop("stream", False)
-
-    json_schema = request_dict.pop("schema", None)
-    regex_string = request_dict.pop("regex", None)
-    if json_schema is not None:
-        logits_processors = [JSONLogitsProcessor(json_schema, tokenizer)]
-    elif regex_string is not None:
-        logits_processors = [RegexLogitsProcessor(regex_string, tokenizer)]
-    else:
-        logits_processors = []
-
-    sampling_params = SamplingParams(
-        **request_dict,
-        logits_processors=logits_processors,  # type: ignore
-    )
-    request_id = random_uuid()
-
-    results_generator = engine.generate(prompt, sampling_params, request_id)  # type: ignore
-
-    # Streaming case
-    async def stream_results() -> AsyncGenerator[bytes, None]:
-        async for request_output in results_generator:
-            prompt = request_output.prompt
-            text_outputs = [prompt + output.text for output in request_output.outputs]
-            ret = {"text": text_outputs}
-            yield (json.dumps(ret) + "\0").encode("utf-8")
-
-    if stream:
-        return StreamingResponse(stream_results())
-
-    # Non-streaming case
-    final_output = None
-    async for request_output in results_generator:
-        if await request.is_disconnected():
-            # Abort the request if the client disconnects.
-            await engine.abort(request_id)  # type: ignore
-            return Response(status_code=499)
-        final_output = request_output
-
-    assert final_output is not None
-    prompt = final_output.prompt
-    text_outputs = [prompt + output.text for output in final_output.outputs]
-    ret = {"text": text_outputs}
-    return JSONResponse(ret)
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--host", type=str, default=None)
-    parser.add_argument("--port", type=int, default=8000)
-    parser.add_argument("--ssl-keyfile", type=str, default=None)
-    parser.add_argument("--ssl-certfile", type=str, default=None)
-    parser = AsyncEngineArgs.add_cli_args(parser)
-    args = parser.parse_args()
-
-    # Adds the `engine_use_ray`,  `disable_log_requests` and `max_log_len`
-    # arguments
-    engine_args: AsyncEngineArgs = AsyncEngineArgs.from_cli_args(args)  # type: ignore
-
-    # Sets default for the model (`facebook/opt-125m`)
-    engine = AsyncLLMEngine.from_engine_args(engine_args)
-    tokenizer = adapt_tokenizer(tokenizer=engine.engine.tokenizer.tokenizer)
-
-    uvicorn.run(
-        app,
-        host=args.host,
-        port=args.port,
-        log_level="debug",
-        timeout_keep_alive=TIMEOUT_KEEP_ALIVE,
-        ssl_keyfile=args.ssl_keyfile,
-        ssl_certfile=args.ssl_certfile,
-    )
diff --git a/outlines/templates.py b/outlines/templates.py
index 16588e813..cd6a4fbf5 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -1,3 +1,6 @@
+"""Create templates to easily build prompts."""
+
+import base64
 import functools
 import inspect
 import json
@@ -5,12 +8,46 @@
 import re
 import textwrap
 from dataclasses import dataclass
+from io import BytesIO
 from pathlib import Path
 from typing import Any, Callable, Dict, Optional, Type, cast
 import warnings
 
 import jinja2
-import pydantic
+from PIL import Image
+from pydantic import BaseModel
+
+
+@dataclass
+class Vision:
+    """Contains the input for a vision model.
+
+    Provide an instance of this class as the `model_input` argument to a model
+    that supports vision.
+
+    Parameters
+    ----------
+    prompt
+        The prompt to use to generate the response.
+    image
+        The image to use to generate the response.
+
+    """
+    prompt: str
+    image: Image.Image
+
+    def __post_init__(self):
+        image = self.image
+
+        if not image.format:
+            raise TypeError(
+                "Could not read the format of the image passed to the model."
+            )
+
+        buffer = BytesIO()
+        image.save(buffer, format=image.format)
+        self.image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
+        self.image_format = f"image/{image.format.lower()}"
 
 
 @dataclass
@@ -21,7 +58,6 @@ class Template:
     template can be accessed by callers.
 
     """
-
     template: jinja2.Template
     signature: Optional[inspect.Signature]
 
@@ -30,7 +66,8 @@ def __call__(self, *args, **kwargs) -> str:
 
         Returns
         -------
-        The rendered template as a Python ``str``.
+        str
+            The rendered template as a Python string.
 
         """
         if self.signature is not None:
@@ -42,7 +79,8 @@ def __call__(self, *args, **kwargs) -> str:
 
     @classmethod
     def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
-        """Create a `Template` instance from a string containing a Jinja template.
+        """Create a `Template` instance from a string containing a Jinja
+        template.
 
         Parameters
         ----------
@@ -51,16 +89,20 @@ def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
 
         Returns
         -------
-        An instance of the class with the provided content as a template.
+        Template
+            An instance of the class with the provided content as a template.
+
         """
         return cls(build_template_from_string(content, filters), None)
 
     @classmethod
     def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
-        """Create a `Template` instance from a file containing a Jinja template.
+        """Create a `Template` instance from a file containing a Jinja
+        template.
 
-        Note: This method does not allow to include and inheritance to reference files
-        that are outside the folder or subfolders of the file given to `from_file`.
+        Note: This method does not allow to include and inheritance to
+        reference files that are outside the folder or subfolders of the file
+        given to `from_file`.
 
         Parameters
         ----------
@@ -70,10 +112,14 @@ def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
         Returns
         -------
         Template
-            An instance of the Template class with the template loaded from the file.
+            An instance of the Template class with the template loaded from the
+            file.
+
         """
-        # We don't use a `Signature` here because it seems not feasible to infer one from a Jinja2 environment that is
-        # split across multiple files (since e.g. we support features like Jinja2 includes and template inheritance)
+        # We don't use a `Signature` here because it seems not feasible to
+        # infer one from a Jinja2 environment that is
+        # split across multiple files (since e.g. we support features like
+        # Jinja2 includes and template inheritance)
         return cls(build_template_from_file(path, filters), None)
 
 
@@ -203,12 +249,12 @@ def create_jinja_env(
     - `source`: get a function's source code
     - `signature`: get a function's signature
     - `args`: get a function's arguments
-    - `schema`: isplay a JSON Schema
+    - `schema`: display a JSON Schema
 
     Users may pass additional filters, and/or override existing ones.
 
-    Arguments
-    ---------
+    Parameters
+    ----------
     loader
        An optional `BaseLoader` instance
     filters
@@ -287,7 +333,7 @@ def get_fn_source(fn: Callable):
     re_search = re.search(re.compile(r"(\bdef\b.*)", re.DOTALL), source)
     if re_search is not None:
         source = re_search.group(0)
-    else:
+    else:  # pragma: no cover
         raise TypeError("Could not read the function's source code")
 
     return source
@@ -300,7 +346,7 @@ def get_fn_signature(fn: Callable):
 
     source = textwrap.dedent(inspect.getsource(fn))
     re_search = re.search(re.compile(r"\(([^)]+)\)"), source)
-    if re_search is None:
+    if re_search is None:  # pragma: no cover
         signature = ""
     else:
         signature = re_search.group(1)
@@ -321,12 +367,9 @@ def get_schema_dict(model: Dict):
     return json.dumps(model, indent=2)
 
 
-@get_schema.register(type(pydantic.BaseModel))
-def get_schema_pydantic(model: Type[pydantic.BaseModel]):
+@get_schema.register(type(BaseModel))
+def get_schema_pydantic(model: Type[BaseModel]):
     """Return the schema of a Pydantic model."""
-    if not isinstance(model, type(pydantic.BaseModel)):
-        raise TypeError("The `schema` filter only applies to Pydantic models.")
-
     if hasattr(model, "model_json_schema"):
         def_key = "$defs"
         raw_schema = model.model_json_schema()
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index d7f40309a..85af9adc0 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,21 +1,40 @@
-from enum import Enum
+"""Output types for structured generation and regex DSL."""
+
+from dataclasses import dataclass, is_dataclass
+from enum import Enum, EnumMeta
+from typing import Union
+
+from jsonschema import Draft202012Validator as Validator
+from jsonschema.exceptions import SchemaError
+from pydantic import BaseModel, TypeAdapter
+from typing_extensions import _TypedDictMeta  # type: ignore
 
 from . import airports, countries, locale
 from outlines.types.dsl import (
     Regex,
-    json_schema,
+    CFG,
+    FSM,
+    JsonSchema,
     regex,
-    either,
+    cfg,
+    fsm,
+    json_schema,
     optional,
+    either,
     exactly,
     at_least,
     at_most,
     between,
-    one_or_more,
     zero_or_more,
+    one_or_more,
+    # deprecated
+    repeat,
+    times,
 )
 
+
 # Python types
+string = Regex(r'"[^"]*"')
 integer = Regex(r"[+-]?(0|[1-9][0-9]*)")
 boolean = Regex("(True|False)")
 number = Regex(rf"{integer.pattern}(\.[0-9]+)?([eE][+-][0-9]+)?")
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 73de3876c..1a0a0ca1c 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,12 +1,77 @@
-import json as json
+"""Regular expression DSL and output types for structured generation.
+
+This module contains elements related to three logical steps in the use of
+output types for structured generation:
+
+1. Definition of `Term` classes that contain output type definitions. That
+   includes both terms intended to be used by themselves such as `JsonSchema`
+   or `CFG` and terms that are part of the regular expression DSL such as
+   `Alternatives` or `KleeneStar` (and the related functions).
+2. Conversion of Python types into `Term` instances (`python_types_to_terms`).
+3. Conversion of a `Term` instance into a regular expression (`to_regex`).
+
+"""
+
+import json
 import re
+import sys
+import warnings
 from dataclasses import dataclass
-from typing import Any, List, Union
-
-from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler
+from enum import EnumMeta
+from types import FunctionType
+from typing import (
+    Any,
+    List,
+    Optional as OptionalType,
+    Union,
+    get_args,
+)
+import interegular
+import jsonschema
+from genson import SchemaBuilder
+from outlines_core.fsm.json_schema import build_regex_from_schema
+from pydantic import (
+    BaseModel,
+    GetCoreSchemaHandler,
+    GetJsonSchemaHandler,
+    TypeAdapter,
+)
 from pydantic.json_schema import JsonSchemaValue
 from pydantic_core import core_schema as cs
-from outlines_core.fsm.json_schema import build_regex_from_schema
+
+import outlines.types as types
+from outlines import grammars
+from outlines.types.utils import (
+    get_schema_from_signature,
+    is_int,
+    is_int_instance,
+    is_float,
+    is_float_instance,
+    is_str,
+    is_str_instance,
+    is_bool,
+    is_datetime,
+    is_date,
+    is_time,
+    is_native_dict,
+    is_dict_instance,
+    is_dataclass,
+    is_typed_dict,
+    is_pydantic_model,
+    is_genson_schema_builder,
+    is_literal,
+    is_union,
+    is_enum,
+    is_callable,
+    is_typing_list,
+    is_typing_tuple,
+    is_typing_dict,
+)
+
+if sys.version_info >= (3, 12):  # pragma: no cover
+    from typing import _TypedDictMeta  # type: ignore
+else:  # pragma: no cover
+    from typing_extensions import _TypedDictMeta  # type: ignore
 
 
 class Term:
@@ -31,18 +96,30 @@ class also handles validation.
 
     """
 
-    def __add__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
-        if isinstance(other, str):
-            other = String(other)
+    def __add__(self: "Term", other: "Term") -> "Sequence":
+        if is_str_instance(other):
+            other = String(str(other))
 
         return Sequence([self, other])
 
-    def __radd__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
-        if isinstance(other, str):
-            other = String(other)
+    def __radd__(self: "Term", other: "Term") -> "Sequence":
+        if is_str_instance(other):
+            other = String(str(other))
 
         return Sequence([other, self])
 
+    def __or__(self: "Term", other: "Term") -> "Alternatives":
+        if is_str_instance(other):
+            other = String(str(other))
+
+        return Alternatives([self, other])
+
+    def __ror__(self: "Term", other: "Term") -> "Alternatives":
+        if is_str_instance(other):
+            other = String(str(other))
+
+        return Alternatives([other, self])
+
     def __get_validator__(self, _core_schema):
         def validate(input_value):
             return self.validate(input_value)
@@ -124,6 +201,16 @@ def one_or_more(self) -> "KleenePlus":
     def zero_or_more(self) -> "KleeneStar":
         return zero_or_more(self)
 
+    # deprecated
+    def times(self, count: int) -> "QuantifyExact":
+        return times(self, count)
+
+    # deprecated
+    def repeat(self, min_count: int, max_count: int) -> Union[
+        "QuantifyMinimum", "QuantifyMaximum", "QuantifyBetween"
+    ]:
+        return repeat(self, min_count, max_count)
+
 
 @dataclass
 class String(Term):
@@ -138,6 +225,14 @@ def __repr__(self):
 
 @dataclass
 class Regex(Term):
+    """Class representing a regular expression.
+
+    Parameters
+    ----------
+    pattern
+        The regular expression as a string.
+
+    """
     pattern: str
 
     def _display_node(self) -> str:
@@ -147,22 +242,119 @@ def __repr__(self):
         return f"Regex(pattern='{self.pattern}')"
 
 
+@dataclass
+class CFG(Term):
+    """Class representing a context-free grammar.
+
+    Parameters
+    ----------
+    definition
+        The definition of the context-free grammar as a string.
+
+    """
+    definition: str
+
+    def _display_node(self) -> str:
+        return f"CFG('{self.definition}')"
+
+    def __repr__(self):
+        return f"CFG(definition='{self.definition}')"
+
+    def __eq__(self, other):
+        if not isinstance(other, CFG):
+            return False
+        return self.definition == other.definition
+
+    @classmethod
+    def from_file(cls, path: str) -> "CFG":
+        """Create a CFG instance from a file containing a CFG definition.
+
+        Parameters
+        ----------
+        path : str
+            The path to the file containing the CFG definition.
+        Returns
+        -------
+        CFG
+            A CFG instance.
+
+        """
+        with open(path, "r") as f:
+            definition = f.read()
+        return cls(definition)
+
+
+@dataclass
+class FSM(Term):
+    """Class representing a finite state machine.
+
+    Parameters
+    ----------
+    fsm
+        The finite state machine to store. This object must be an instance of
+        `interegular.fsm.FSM`.
+
+    """
+    fsm: interegular.fsm.FSM
+
+    def _display_node(self) -> str:
+        return f"FSM({self.fsm.__repr__()})"
+
+    def __repr__(self):
+        return f"FSM(fsm={self.fsm.__repr__()})"
+
+
 class JsonSchema(Term):
-    def __init__(self, schema: Union[dict, str, type[BaseModel]]):
-        if isinstance(schema, dict):
+    """Class representing a JSON schema.
+
+    The JSON schema object from which to instantiate the class can be a
+    dictionary, a string, a Pydantic model, a typed dict, a dataclass, or a
+    genSON schema builder.
+
+    """
+    def __init__(
+        self,
+        schema: Union[
+            dict, str, type[BaseModel], _TypedDictMeta, type, SchemaBuilder
+        ],
+        whitespace_pattern: OptionalType[str] = None,
+    ):
+        """
+        Parameters
+        ----------
+        schema
+            The object containing the JSON schema.
+        whitespace_pattern
+            The pattern to use to match whitespace characters.
+
+        """
+        schema_str: str
+
+        if is_dict_instance(schema):
             schema_str = json.dumps(schema)
-        elif isinstance(schema, str):
-            schema_str = schema
-        elif issubclass(schema, BaseModel):
-            schema_str = json.dumps(schema.model_json_schema())
+        elif is_str_instance(schema):
+            schema_str = str(schema)
+        elif is_pydantic_model(schema):
+            schema_str = json.dumps(schema.model_json_schema())  # type: ignore
+        elif is_typed_dict(schema):
+            schema_str = json.dumps(TypeAdapter(schema).json_schema())
+        elif is_dataclass(schema):
+            schema_str = json.dumps(TypeAdapter(schema).json_schema())
+        elif is_genson_schema_builder(schema):
+            schema_str = schema.to_json()  # type: ignore
         else:
             raise ValueError(
-                f"Cannot parse schema {json_schema}. The schema must be either "
-                + "a Pydantic class, a dictionary or a string that contains the JSON "
-                + "schema specification"
+                f"Cannot parse schema {schema}. The schema must be either "
+                + "a Pydantic class, typed dict, a dataclass, a genSON schema "
+                + "builder or a string or dict that contains the JSON schema "
+                + "specification"
             )
 
         self.schema = schema_str
+        self.whitespace_pattern = whitespace_pattern
+
+    def __post_init__(self):
+        jsonschema.Draft7Validator.check_schema(json.loads(self.schema))
 
     def _display_node(self) -> str:
         return f"JsonSchema('{self.schema}')"
@@ -170,6 +362,35 @@ def _display_node(self) -> str:
     def __repr__(self):
         return f"JsonSchema(schema='{self.schema}')"
 
+    def __eq__(self, other):
+        if not isinstance(other, JsonSchema):
+            return False
+        try:
+            self_dict = json.loads(self.schema)
+            other_dict = json.loads(other.schema)
+            return self_dict == other_dict
+        except json.JSONDecodeError:  # pragma: no cover
+            return self.schema == other.schema
+
+    @classmethod
+    def from_file(cls, path: str) -> "JsonSchema":
+        """Create a JsonSchema instance from a .json file containing a JSON
+        schema.
+
+        Parameters
+        ----------
+        path:
+            The path to the file containing the JSON schema.
+        Returns
+        -------
+        JsonSchema
+            A JsonSchema instance.
+
+        """
+        with open(path, "r") as f:
+            schema = json.load(f)
+        return cls(schema)
+
 
 @dataclass
 class KleeneStar(Term):
@@ -322,6 +543,14 @@ def regex(pattern: str):
     return Regex(pattern)
 
 
+def cfg(definition: str):
+    return CFG(definition)
+
+
+def fsm(fsm: interegular.fsm.FSM):
+    return FSM(fsm)
+
+
 def json_schema(schema: Union[str, dict, type[BaseModel]]):
     return JsonSchema(schema)
 
@@ -331,6 +560,7 @@ def either(*terms: Union[str, Term]):
 
     This factory function automatically translates string arguments
     into `String` objects.
+
     """
     terms = [String(arg) if isinstance(arg, str) else arg for arg in terms]
     return Alternatives(terms)
@@ -374,11 +604,306 @@ def one_or_more(term: Union[Term, str]) -> KleenePlus:
     return KleenePlus(term)
 
 
+def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
+    """Convert Python types to Outlines DSL terms that constrain LLM output.
+
+    Parameters
+    ----------
+    ptype
+        The Python type to convert
+    recursion_depth
+        Current recursion depth to prevent infinite recursion
+
+    Returns
+    -------
+    Term
+        The corresponding DSL `Term` instance.
+
+    """
+    if recursion_depth > 10:
+        raise RecursionError(
+            f"Maximum recursion depth exceeded when converting {ptype}. "
+            "This might be due to a recursive type definition."
+        )
+
+    # First handle Term instances
+    if isinstance(ptype, Term):
+        return ptype
+
+    # Basic types
+    if is_int(ptype):
+        return types.integer
+    elif is_float(ptype):
+        return types.number
+    elif is_bool(ptype):
+        return types.boolean
+    elif is_str(ptype):
+        return types.string
+    elif is_native_dict(ptype):
+        return CFG(grammars.json)
+    elif is_time(ptype):
+        return types.time
+    elif is_date(ptype):
+        return types.date
+    elif is_datetime(ptype):
+        return types.datetime
+
+    # Basic type instances
+    if is_str_instance(ptype):
+        return String(ptype)
+    elif is_int_instance(ptype) or is_float_instance(ptype):
+        return Regex(str(ptype))
+
+    # Structured types
+    structured_type_checks = [
+        lambda x: is_dataclass(x),
+        lambda x: is_typed_dict(x),
+        lambda x: is_pydantic_model(x),
+    ]
+    if any(check(ptype) for check in structured_type_checks):
+        schema = TypeAdapter(ptype).json_schema()
+        return JsonSchema(schema)
+
+    elif is_genson_schema_builder(ptype):
+        schema = ptype.to_json()
+        return JsonSchema(schema)
+
+    if is_enum(ptype):
+        return Alternatives(
+            [
+                python_types_to_terms(member, recursion_depth + 1)
+                for member in _get_enum_members(ptype)
+            ]
+        )
+
+    args = get_args(ptype)
+    if is_literal(ptype):
+        return _handle_literal(args)
+    elif is_union(ptype):
+        return _handle_union(args, recursion_depth)
+    elif is_typing_list(ptype):
+        return _handle_list(args, recursion_depth)
+    elif is_typing_tuple(ptype):
+        return _handle_tuple(args, recursion_depth)
+    elif is_typing_dict(ptype):
+        return _handle_dict(args, recursion_depth)
+
+    if is_callable(ptype):
+        return JsonSchema(get_schema_from_signature(ptype))
+
+    type_name = getattr(ptype, "__name__", ptype)
+    raise TypeError(
+        f"Type {type_name} is currently not supported. Please open an issue: "
+        "https://github.com/dottxt-ai/outlines/issues"
+    )
+
+
+def _get_enum_members(ptype: EnumMeta) -> List[Any]:
+    regular_members = [member.value for member in ptype]  # type: ignore
+    function_members = []
+    for key, value in ptype.__dict__.items():
+        if (
+            isinstance(value, FunctionType)
+            and not (key.startswith('__') and key.endswith('__'))
+            and key != '_generate_next_value_'  # Skip this specific method that causes issues
+        ):
+            function_members.append(value)
+    return regular_members + function_members
+
+
+def _handle_literal(args: tuple) -> Alternatives:
+    return Alternatives([python_types_to_terms(arg) for arg in args])
+
+
+def _handle_union(args: tuple, recursion_depth: int) -> Alternatives:
+    # Handle the Optional[T] type
+    if len(args) == 2 and (type(None) in args or None in args):
+        other_ptype = next(arg for arg in args if arg not in (type(None), None))
+        return Alternatives(
+            [
+                python_types_to_terms(other_ptype, recursion_depth + 1),
+                String("None"),
+            ]
+        )
+    return Alternatives(
+        [python_types_to_terms(arg, recursion_depth + 1) for arg in args]
+    )
+
+
+def _handle_list(args: tuple, recursion_depth: int) -> Sequence:
+    if args is None or len(args) > 1:
+        raise TypeError(
+            f"Only homogeneous lists are supported. Got multiple type arguments {args}."
+        )
+    item_type = python_types_to_terms(args[0], recursion_depth + 1)
+    return Sequence(
+        [
+            String("["),
+            item_type,
+            KleeneStar(Sequence([String(", "), item_type])),
+            String("]"),
+        ]
+    )
+
+
+def _handle_tuple(args: tuple, recursion_depth: int) -> Union[Sequence, String]:
+    if len(args) == 0 or args == ((),):
+        return String("()")
+    elif len(args) == 2 and args[1] is Ellipsis:
+        item_term = python_types_to_terms(args[0], recursion_depth + 1)
+        return Sequence(
+            [
+                String("("),
+                item_term,
+                KleeneStar(Sequence([String(", "), item_term])),
+                String(")"),
+            ]
+        )
+    else:
+        items = [python_types_to_terms(arg, recursion_depth + 1) for arg in args]
+        separator = String(", ")
+        elements = []
+        for i, item in enumerate(items):
+            elements.append(item)
+            if i < len(items) - 1:
+                elements.append(separator)
+        return Sequence([String("("), *elements, String(")")])
+
+
+def _handle_dict(args: tuple, recursion_depth: int) -> Sequence:
+    if args is None or len(args) != 2:
+        raise TypeError(f"Dict must have exactly two type arguments. Got {args}.")
+    # Add dict support with key:value pairs
+    key_type = python_types_to_terms(args[0], recursion_depth + 1)
+    value_type = python_types_to_terms(args[1], recursion_depth + 1)
+    return Sequence(
+        [
+            String("{"),
+            Optional(
+                Sequence(
+                    [
+                        key_type,
+                        String(":"),
+                        value_type,
+                        KleeneStar(
+                            Sequence([String(", "), key_type, String(":"), value_type])
+                        ),
+                    ]
+                )
+            ),
+            String("}"),
+        ]
+    )
+
+
+# deprecated
+def repeat(term: Term, min_count: int, max_count: int) -> Union[
+    QuantifyMinimum, QuantifyMaximum, QuantifyBetween
+]:
+    if min_count is None and max_count is None:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `between`, `at_least` or `at_most` instead.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        raise ValueError(
+            "You must provide a value for at least `min_count` or `max_count`"
+        )
+    if max_count is None:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `at_least` instead.
+            For instance:
+            ```python
+            from outlines.types import Regex, at_least
+            digit = Regex(r'[0-9]')
+            # 2 ways of doing the same thing
+            at_least_5_digits = digit.at_least(5)
+            at_least_5_digits = at_least(digit, 5)
+            ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return at_least(min_count, term)
+    if min_count is None:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `at_most` instead.
+            For instance:
+            ```python
+            from outlines.types import Regex, at_most
+            digit = Regex(r'[0-9]')
+            # 2 ways of doing the same thing
+            at_most_5_digits = digit.at_most(5)
+            at_most_5_digits = at_most(digit, 5)
+            ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return at_most(max_count, term)
+    else:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `between` instead.
+            For instance:
+            ```python
+            from outlines.types import Regex, between
+            digit = Regex(r'[0-9]')
+            # 2 ways of doing the same thing
+            five_to_eight_digits = digit.between(5, 8)
+            five_to_eight_digits = between(digit, 5, 8)
+            ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return between(min_count, max_count, term)
+
+
+# deprecated
+def times(term: Term, count: int) -> QuantifyExact:
+    warnings.warn("""
+        The `times` function/method is deprecated starting from v0.2.2.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Use `exactly` instead.
+        For instance:
+        ```python
+        from outlines.types import Regex, exactly
+        digit = Regex(r'[0-9]')
+        # 2 ways of doing the same thing
+        five_digits = digit.exactly(5)
+        five_digits = exactly(digit, 5)
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    return exactly(count, term)
+
+
 def to_regex(term: Term) -> str:
     """Convert a term to a regular expression.
 
     We only consider self-contained terms that do not refer to another rule.
 
+    Parameters
+    ----------
+    term
+        The term to convert to a regular expression.
+
+    Returns
+    -------
+    str
+        The regular expression as a string.
+
     """
     match term:
         case String():
@@ -386,7 +911,7 @@ def to_regex(term: Term) -> str:
         case Regex():
             return f"({term.pattern})"
         case JsonSchema():
-            regex_str = build_regex_from_schema(term.schema)
+            regex_str = build_regex_from_schema(term.schema, term.whitespace_pattern)
             return f"({regex_str})"
         case KleeneStar():
             return f"({to_regex(term.term)})*"
diff --git a/outlines/types/locale/__init__.py b/outlines/types/locale/__init__.py
index 511631d84..97c9abff0 100644
--- a/outlines/types/locale/__init__.py
+++ b/outlines/types/locale/__init__.py
@@ -1 +1,3 @@
+"""Locale-specific regex patterns."""
+
 from . import us
diff --git a/outlines/types/locale/us.py b/outlines/types/locale/us.py
index 0bda82d44..275aa492e 100644
--- a/outlines/types/locale/us.py
+++ b/outlines/types/locale/us.py
@@ -1,3 +1,5 @@
+"""Locale-specific regex patterns for the United States."""
+
 from outlines.types.dsl import Regex
 
 zip_code = Regex(r"\d{5}(?:-\d{4})?")
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
new file mode 100644
index 000000000..64537643e
--- /dev/null
+++ b/outlines/types/utils.py
@@ -0,0 +1,206 @@
+"""Utility functions for the types module."""
+
+import dataclasses
+import datetime
+import inspect
+import sys
+import warnings
+from enum import Enum, EnumMeta
+from typing import (
+    Annotated,
+    Any,
+    Callable,
+    Dict,
+    Literal,
+    List,
+    NewType,
+    Tuple,
+    Union,
+    get_args,
+    get_origin,
+)
+
+import interegular
+from genson import SchemaBuilder
+from pydantic import BaseModel, create_model
+
+if sys.version_info >= (3, 12): # pragma: no cover
+    from typing import _TypedDictMeta  # type: ignore
+else: # pragma: no cover
+    from typing_extensions import _TypedDictMeta  # type: ignore
+
+
+# Type identification
+
+
+def is_int(value: Any) -> bool:
+    return (
+        value is int
+        or get_origin(value) is int
+        or (get_origin(value) is Annotated and get_args(value)[0] is int)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is int)
+    )
+
+
+def is_int_instance(value: Any) -> bool:
+    return isinstance(value, int)
+
+
+def is_float(value: Any) -> bool:
+    return (
+        value is float
+        or get_origin(value) is float
+        or (get_origin(value) is Annotated and get_args(value)[0] is float)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is float)
+    )
+
+
+def is_float_instance(value: Any) -> bool:
+    return isinstance(value, float)
+
+
+def is_str(value: Any) -> bool:
+    return (
+        value is str
+        or get_origin(value) is str
+        or (get_origin(value) is Annotated and get_args(value)[0] is str)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is str)
+    )
+
+
+def is_str_instance(value: Any) -> bool:
+    return isinstance(value, str)
+
+
+def is_bool(value: Any) -> bool:
+    return (
+        value is bool
+        or get_origin(value) is bool
+        or (get_origin(value) is Annotated and get_args(value)[0] is bool)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is bool)
+    )
+
+
+def is_dict_instance(value: Any) -> bool:
+    return isinstance(value, dict)
+
+
+def is_datetime(value: Any) -> bool:
+    return value is datetime.datetime or get_origin(value) is datetime.datetime
+
+
+def is_date(value: Any) -> bool:
+    return value is datetime.date or get_origin(value) is datetime.date
+
+
+def is_time(value: Any) -> bool:
+    return value is datetime.time or get_origin(value) is datetime.time
+
+
+def is_native_dict(value: Any) -> bool:
+    return value is dict
+
+
+def is_typing_dict(value: Any) -> bool:
+    return get_origin(value) is dict
+
+
+def is_typing_list(value: Any) -> bool:
+    return get_origin(value) is list
+
+
+def is_typing_tuple(value: Any) -> bool:
+    return get_origin(value) is tuple
+
+
+def is_union(value: Any) -> bool:
+    return get_origin(value) is Union
+
+
+def is_literal(value: Any) -> bool:
+    return get_origin(value) is Literal
+
+
+def is_dataclass(value: Any) -> bool:
+    return isinstance(value, type) and dataclasses.is_dataclass(value)
+
+
+def is_typed_dict(value: Any) -> bool:
+    return isinstance(value, _TypedDictMeta)
+
+
+def is_pydantic_model(value):
+    # needed because generic type cannot be used with `issubclass`    # for Python versions < 3.11
+    if get_origin(value) is not None:
+        return False
+
+    return isinstance(value, type) and issubclass(value, BaseModel)
+
+
+def is_genson_schema_builder(value: Any) -> bool:
+    return isinstance(value, SchemaBuilder)
+
+
+def is_enum(value: Any) -> bool:
+    return isinstance(value, EnumMeta)
+
+
+def is_callable(value: Any) -> bool:
+    return callable(value) and not isinstance(value, type)
+
+
+def is_interegular_fsm(value: Any) -> bool:
+    return isinstance(value, interegular.fsm.FSM)
+
+
+# Type conversion
+
+
+def get_enum_from_literal(value) -> Enum:
+    return Enum(
+        value.__name__,
+        {str(arg): arg for arg in get_args(value)}
+    )
+
+
+def get_schema_from_signature(fn: Callable) -> dict:
+    """Turn a function signature into a JSON schema.
+
+    Every JSON object valid to the output JSON Schema can be passed
+    to `fn` using the ** unpacking syntax.
+
+    """
+    signature = inspect.signature(fn)
+    arguments = {}
+    for name, arg in signature.parameters.items():
+        if arg.annotation == inspect._empty:
+            raise ValueError("Each argument must have a type annotation")
+        else:
+            arguments[name] = (arg.annotation, ...)
+
+    try:
+        fn_name = fn.__name__
+    except Exception as e:
+        fn_name = "Arguments"
+        warnings.warn(
+            f"The function name could not be determined. Using default name 'Arguments' instead. For debugging, here is exact error:\n{e}",
+            category=UserWarning,
+        )
+    model = create_model(fn_name, **arguments)
+
+    return model.model_json_schema()
+
+
+def get_schema_from_enum(myenum: type[Enum]) -> dict:
+    if len(myenum) == 0:
+        raise ValueError(
+            f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
+        )
+    choices = [
+        get_schema_from_signature(elt.value.func)
+        if callable(elt.value)
+        else {"const": elt.value}
+        for elt in myenum
+    ]
+    schema = {"title": myenum.__name__, "oneOf": choices}
+    return schema
diff --git a/outlines/base.py b/outlines/v0_legacy/base.py
similarity index 67%
rename from outlines/base.py
rename to outlines/v0_legacy/base.py
index 29d42c54c..e545b5bcf 100644
--- a/outlines/base.py
+++ b/outlines/v0_legacy/base.py
@@ -4,28 +4,10 @@
 import inspect
 from typing import Callable, Optional
 
-import numpy as np
-
-# Import required functions based on NumPy version
-np_major_version = int(np.__version__.split(".")[0])
-if np_major_version >= 2:
-    from numpy.lib._function_base_impl import (
-        _calculate_shapes,
-        _parse_gufunc_signature,
-        _parse_input_dimensions,
-        _update_dim_sizes,
-    )
-else:
-    from numpy.lib.function_base import (
-        _calculate_shapes,
-        _parse_gufunc_signature,
-        _parse_input_dimensions,
-        _update_dim_sizes,
-    )
 
 # Allow nested loops for running in notebook. We don't enable it globally as it
 # may interfere with other libraries that use asyncio.
-if hasattr(builtins, "__IPYTHON__"):
+if hasattr(builtins, "__IPYTHON__"): # pragma: no cover
     try:
         import nest_asyncio
 
@@ -51,21 +33,65 @@ class vectorize:
     """
 
     def __init__(self, func: Callable, signature: Optional[str] = None):
+        self.setup_numpy(signature)
         self.func = func
         self.signature = signature
         self.is_coroutine_fn = inspect.iscoroutinefunction(func)
 
         functools.update_wrapper(self, func)
 
-        if signature is not None:
-            # Parse the signature string into a Python data structure.
-            # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
-            self._in_and_out_core_dimensions = _parse_gufunc_signature(signature)
-        else:
+    def setup_numpy(self, signature: Optional[str] = None):
+        """Setup NumPy and import required functions based on NumPy version.
+
+        As numpy is an optional dependency only needed for the legacy OpenAI
+        model, we put the import in a try block to make sure no error is raised
+        if it's not installed.
+        """
+        try:
+            import numpy as np
+            self.np = np
+            # Import required functions based on NumPy version
+            np_major_version = int(np.__version__.split(".")[0])
+            if np_major_version >= 2:
+                from numpy.lib._function_base_impl import (
+                    _calculate_shapes,
+                    _parse_gufunc_signature,
+                    _parse_input_dimensions,
+                    _update_dim_sizes,
+                )
+            else: # pragma: no cover
+                from numpy.lib.function_base import (
+                    _calculate_shapes,
+                    _parse_gufunc_signature,
+                    _parse_input_dimensions,
+                    _update_dim_sizes,
+                )
+            self._calculate_shapes = _calculate_shapes
+            self._parse_gufunc_signature = _parse_gufunc_signature
+            self._parse_input_dimensions = _parse_input_dimensions
+            self._update_dim_sizes = _update_dim_sizes
+            if signature is not None:
+                # Parse the signature string into a Python data structure.
+                # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
+                self._in_and_out_core_dimensions = self._parse_gufunc_signature(signature)
+            else:
+                self._in_and_out_core_dimensions = None
+        except (ModuleNotFoundError, ImportError): # pragma: no cover
+            self.np = None
+            self._calculate_shapes = None
+            self._parse_gufunc_signature = None
+            self._parse_input_dimensions = None
+            self._update_dim_sizes = None
             self._in_and_out_core_dimensions = None
 
     def __call__(self, *args, **kwargs):
         """Call the vectorized function."""
+        # Raise an error ourselves if NumPy is not installed.
+        if not self.np: # pragma: no cover
+            raise ImportError(
+                "NumPy is required to use the legacy version of the OpenAI "
+                "model."
+            )
         if not args and not kwargs:
             return self.call_thunk()
         elif self.signature is not None:
@@ -104,14 +130,14 @@ def call_no_signature(self, *args, **kwargs):
 
         """
         # Convert args and kwargs to arrays
-        args = [np.array(arg) for arg in args]
-        kwargs = {key: np.array(value) for key, value in kwargs.items()}
+        args = [self.np.array(arg) for arg in args]
+        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
 
         # Broadcast args and kwargs
-        broadcast_shape = np.broadcast(*args, *list(kwargs.values())).shape
-        args = [np.broadcast_to(arg, broadcast_shape) for arg in args]
+        broadcast_shape = self.np.broadcast(*args, *list(kwargs.values())).shape
+        args = [self.np.broadcast_to(arg, broadcast_shape) for arg in args]
         kwargs = {
-            key: np.broadcast_to(value, broadcast_shape)
+            key: self.np.broadcast_to(value, broadcast_shape)
             for key, value in kwargs.items()
         }
 
@@ -127,9 +153,9 @@ def call_no_signature(self, *args, **kwargs):
             results if isinstance(results, tuple) else (results,) for results in outputs
         ]
         outputs = tuple(
-            [np.asarray(x).reshape(broadcast_shape).squeeze() for x in zip(*outputs)]
+            [self.np.asarray(x).reshape(broadcast_shape).squeeze() for x in zip(*outputs)]
         )
-        outputs = tuple([x.item() if np.ndim(x) == 0 else x for x in outputs])
+        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
 
         n_results = len(list(outputs))
 
@@ -149,25 +175,25 @@ def call_with_signature(self, *args, **kwargs):
             )
 
         # Convert args and kwargs to arrays
-        args = [np.asarray(arg) for arg in args]
-        kwargs = {key: np.array(value) for key, value in kwargs.items()}
+        args = [self.np.asarray(arg) for arg in args]
+        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
 
         # Find the arguments' broadcast shape, and map placeholder
         # variables in the signature to the number of dimensions
         # they correspond to given the arguments.
-        broadcast_shape, dim_sizes = _parse_input_dimensions(
+        broadcast_shape, dim_sizes = self._parse_input_dimensions(
             args + list(kwargs.values()), input_core_dims
         )
 
         # Calculate the shape to which each of the arguments should be broadcasted
         # and reshape them accordingly.
-        input_shapes = _calculate_shapes(broadcast_shape, dim_sizes, input_core_dims)
+        input_shapes = self._calculate_shapes(broadcast_shape, dim_sizes, input_core_dims)
         args = [
-            np.broadcast_to(arg, shape, subok=True)
+            self.np.broadcast_to(arg, shape, subok=True)
             for arg, shape in zip(args, input_shapes)
         ]
         kwargs = {
-            key: np.broadcast_to(value, broadcast_shape)
+            key: self.np.broadcast_to(value, broadcast_shape)
             for key, value in kwargs.items()
         }
 
@@ -195,18 +221,18 @@ def call_with_signature(self, *args, **kwargs):
         # number of dimensions of the outputs accordingly.
         for results, core_dims in zip(flat_outputs, output_core_dims):
             for result in results:
-                _update_dim_sizes(dim_sizes, result, core_dims)
+                self._update_dim_sizes(dim_sizes, result, core_dims)
 
         # Calculate the shape to which each of the outputs should be broadcasted
         # and reshape them.
-        shapes = _calculate_shapes(broadcast_shape, dim_sizes, output_core_dims)
+        shapes = self._calculate_shapes(broadcast_shape, dim_sizes, output_core_dims)
         outputs = tuple(
             [
-                np.hstack(results).reshape(shape).squeeze()
+                self.np.hstack(results).reshape(shape).squeeze()
                 for shape, results in zip(shapes, zip(*outputs))
             ]
         )
-        outputs = tuple([x.item() if np.ndim(x) == 0 else x for x in outputs])
+        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
 
         return outputs[0] if n_results == 1 else outputs
 
@@ -226,7 +252,7 @@ def vectorize_call(self, broadcast_shape, args, kwargs):
 
         """
         outputs = []
-        for index in np.ndindex(*broadcast_shape):
+        for index in self.np.ndindex(*broadcast_shape):
             current_args = tuple(arg[index] for arg in args)
             current_kwargs = {key: value[index] for key, value in kwargs.items()}
             outputs.append(self.func(*current_args, **current_kwargs))
@@ -252,7 +278,7 @@ def vectorize_call_coroutine(self, broadcast_shape, args, kwargs):
 
         async def create_and_gather_tasks():
             tasks = []
-            for index in np.ndindex(*broadcast_shape):
+            for index in self.np.ndindex(*broadcast_shape):
                 current_args = tuple(arg[index] for arg in args)
                 current_kwargs = {key: value[index] for key, value in kwargs.items()}
                 tasks.append(self.func(*current_args, **current_kwargs))
@@ -268,32 +294,3 @@ async def create_and_gather_tasks():
             loop.close()
 
         return outputs
-
-
-def _update_arrays_type(arrays, results):
-    """Update the dtype of arrays.
-
-    String arrays contain strings of fixed length. Here they are initialized with
-    the type of the first results, so that if the next results contain longer
-    strings they will be truncated when added to the output arrays. Here we
-    update the type if the current results contain longer strings than in the
-    current output array.
-
-    Parameters
-    ----------
-    arrays
-        Arrays that contain the vectorized function's results.
-    results
-        The current output of the function being vectorized.
-
-    """
-
-    updated_arrays = []
-    for array, result in zip(arrays, results):
-        if array.dtype.type == np.str_:
-            if array.dtype < np.array(result).dtype:
-                array = array.astype(np.array(result).dtype)
-
-        updated_arrays.append(array)
-
-    return tuple(updated_arrays)
diff --git a/outlines/function.py b/outlines/v0_legacy/function.py
similarity index 58%
rename from outlines/function.py
rename to outlines/v0_legacy/function.py
index aff21d68f..465df53f9 100644
--- a/outlines/function.py
+++ b/outlines/v0_legacy/function.py
@@ -1,16 +1,58 @@
 import importlib.util
+import warnings
 from dataclasses import dataclass
 from typing import TYPE_CHECKING, Callable, Optional, Tuple, Union
 
 import requests
 
-from outlines import generate, models
+from outlines.v0_legacy import generate
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.models import transformers
 
-if TYPE_CHECKING:
-    from outlines.generate.api import SequenceGenerator
+if TYPE_CHECKING: # pragma: no cover
     from outlines.templates import Template
 
 
+def function_warning():
+    warnings.warn("""
+        The `Function` class is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Instead, you should use the `outlines.Application` class that
+        implements a similar functionality.
+        An `Application` is initialized with a prompt template and an
+        output type. It can then be called with a model and a prompt
+        to generate a response that follows the format specified by the
+        output type.
+        For example:
+        ```python
+        from pydantic import BaseModel
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+        from outlines import models, Application
+        from outlines.types import JsonType
+        from outlines.templates import Template
+
+        class OutputModel(BaseModel):
+            result: int
+
+        model = models.from_transformers(
+            AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+            AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+        )
+
+        template_string = "What is 2 times {{ num }}?"
+        template = Template.from_string(template_string)
+
+        application = Application(template, JsonType(OutputModel))
+
+        result = application(model, num=3)
+        print(result)  # Expected output: { "result" : 6 }
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2
+    )
+
+
 @dataclass
 class Function:
     """Represents an Outlines function.
@@ -21,15 +63,18 @@ class Function:
     prompt template.
 
     """
-
     prompt_template: "Template"
     schema: Union[str, Callable, object]
     model_name: str
-    generator: Optional["SequenceGenerator"] = None
+    generator: Optional["GeneratorV0Adapter"] = None
+
+    def __post_init__(self):
+        function_warning()
 
     @classmethod
     def from_github(cls, program_path: str, function_name: str = "fn"):
         """Load a function stored on GitHub"""
+        function_warning()
         program_content = download_from_github(program_path)
         function = extract_function_from_file(program_content, function_name)
 
@@ -37,7 +82,7 @@ def from_github(cls, program_path: str, function_name: str = "fn"):
 
     def init_generator(self):
         """Load the model and initialize the generator."""
-        model = models.transformers(self.model_name)
+        model = transformers(self.model_name)
         self.generator = generate.json(model, self.schema)
 
     def __call__(self, *args, **kwargs):
@@ -55,11 +100,18 @@ def __call__(self, *args, **kwargs):
             Values to pass to the prompt template as keyword arguments.
 
         """
-        if self.generator is None:
-            self.init_generator()
+        # We do not want to raise the deprecation warning about the `json`
+        # function or the transformers function here.
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore",
+                category=DeprecationWarning
+            )
+            if self.generator is None:
+                self.init_generator()
 
-        prompt = self.prompt_template(*args, **kwargs)
-        return self.generator(prompt)
+            prompt = self.prompt_template(*args, **kwargs)
+            return self.generator(prompt)
 
 
 def download_from_github(short_path: str):
@@ -94,11 +146,10 @@ def download_from_github(short_path: str):
 
 def extract_function_from_file(content: str, function_name: str) -> Tuple[Callable]:
     """Extract a function object from a downloaded file."""
-
     spec = importlib.util.spec_from_loader(
         "outlines_function", loader=None, origin="github"
     )
-    if spec is not None:
+    if spec is not None:  # pragma: no cover
         module = importlib.util.module_from_spec(spec)
         exec(content, module.__dict__)
 
diff --git a/outlines/generate/__init__.py b/outlines/v0_legacy/generate/__init__.py
similarity index 62%
rename from outlines/generate/__init__.py
rename to outlines/v0_legacy/generate/__init__.py
index f28cbd80d..617f7b2e3 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/v0_legacy/generate/__init__.py
@@ -1,4 +1,3 @@
-from .api import SequenceGenerator
 from .cfg import cfg
 from .choice import choice
 from .format import format
@@ -6,3 +5,13 @@
 from .json import json
 from .regex import regex
 from .text import text
+
+__all__ = [
+    "cfg",
+    "choice",
+    "format",
+    "fsm",
+    "json",
+    "regex",
+    "text",
+]
diff --git a/outlines/v0_legacy/generate/api.py b/outlines/v0_legacy/generate/api.py
new file mode 100644
index 000000000..625bac105
--- /dev/null
+++ b/outlines/v0_legacy/generate/api.py
@@ -0,0 +1,175 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+from dataclasses import asdict
+from typing import Any, Dict, List, Optional, Union
+
+from outlines.generator import Generator
+from outlines.v0_legacy.models.exllamav2 import ExLlamaV2Model, exllamav2_params_adapter
+from outlines.v0_legacy.models.transformers import transformers_params_adapter
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
+from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
+from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
+from outlines.models import (
+    LlamaCpp,
+    MLXLM,
+    Transformers,
+    VLLMOffline,
+)
+
+V0GeneratorModel = Union[
+    ExLlamaV2Model,
+    LlamaCpp,
+    MLXLM,
+    Transformers,
+    TransformersVision,
+    VLLMOffline,
+]
+
+
+class GeneratorV0Adapter:
+    """Object used to bridge the gap between the legacy `generate` functions
+    and the current `Generator` object.
+
+    The legacy `generate` functions return an instance of this class. When
+    initialized, it stores an instance of the current `Generator` object
+    and calls it when it's called.
+    """
+
+    def __init__(self, model, output_type, sampler):
+        if not isinstance(model, V0GeneratorModel):  # type: ignore
+            raise ValueError(
+                "You can only use the v0 API with models that were already"
+                + f"available in v0. Got {type(model)}."
+            )
+        self.model = model
+        self.generator = Generator(model, output_type)
+        self.sampling_params = asdict(sampler.sampling_params)
+
+    def format_sequence(self, sequence: str):
+        return sequence
+
+    def create_inference_params(
+        self, max_tokens, stop_at, seed, **model_specific_params
+    ):
+        """Turn the arguments received by the generator into arguments
+        for the model associated to it.
+        """
+        generation_params = {
+            "max_tokens": max_tokens,
+            "stop_at": stop_at,
+            "seed": seed,
+        }
+
+        if isinstance(self.model, LlamaCpp):
+            adapter_function = llamacpp_params_adapter
+        elif isinstance(self.model, VLLMOffline): # pragma: no cover
+            adapter_function = vllm_offline_params_adapter
+        elif isinstance(self.model, Transformers):
+            adapter_function = transformers_params_adapter
+        elif isinstance(self.model, ExLlamaV2Model): # pragma: no cover
+            adapter_function = exllamav2_params_adapter
+        elif isinstance(self.model, MLXLM): # pragma: no cover
+            adapter_function = mlxlm_params_adapter
+
+        inference_params = adapter_function(
+            self.sampling_params,
+            generation_params,
+            model_specific_params
+        )
+        return inference_params
+
+    def __call__(
+        self,
+        prompts: Union[str, List[str], Dict[str, Any]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        result = self.generator(
+            prompts,
+            **self.create_inference_params(
+                max_tokens, stop_at, seed, **model_specific_params
+            )
+        )
+        if isinstance(result, list):
+            return [self.format_sequence(r) for r in result]
+        else:
+            return self.format_sequence(result)
+
+    def stream(
+        self,
+        prompts: Union[str, List[str], Dict[str, Any]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        return self.generator.stream(
+            prompts,
+            **self.create_inference_params(
+                max_tokens, stop_at, seed, **model_specific_params
+            )
+        )
+
+
+class GeneratorVisionV0Adapter(GeneratorV0Adapter):
+    """This object does the same as `GeneratorV0Adapter` but for the
+    `TransformersVision` model. The only difference between the two is that
+    it receives a media argument in addition to the prompts.
+    """
+
+    @staticmethod
+    def merge_prompts_and_media(
+        prompts: Union[str, List[str]],
+        media: Union[Any, List[Any]],
+    ) -> Dict[str, Any]:
+        """Merge prompts and media into a single dictionary
+
+        This function is needed as the `TransformersMultiModal` model requires
+        the prompt and assets to be contained in a single argument while the
+        legacy `TransformersVision` model requires separate arguments for the
+        prompt and assets.
+        """
+        return {
+            "text": prompts,
+            "images": media,
+        }
+
+    def __call__(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[str, List[str]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        merged_input = self.merge_prompts_and_media(prompts, media)
+        return super().__call__(
+            merged_input,
+            max_tokens,
+            stop_at,
+            seed,
+            **model_specific_params
+        )
+
+    def stream(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[str, List[str]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        merged_input = self.merge_prompts_and_media(prompts, media)
+        return super().stream(
+            merged_input,
+            max_tokens,
+            stop_at,
+            seed,
+            **model_specific_params
+        )
diff --git a/outlines/v0_legacy/generate/cfg.py b/outlines/v0_legacy/generate/cfg.py
new file mode 100644
index 000000000..0f33b6efb
--- /dev/null
+++ b/outlines/v0_legacy/generate/cfg.py
@@ -0,0 +1,80 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models.llamacpp import LlamaCpp
+from outlines.models.openai import OpenAI
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.types.dsl import CFG
+
+
+def cfg(
+    model, cfg_str: str, sampler: Sampler = multinomial()
+) -> GeneratorV0Adapter:
+    """Generate text in the language of a Context-Free Grammar
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Support for it will be removed in v1.1.0.
+    Use the `Generator` object instead:
+
+    ```python
+    from outlines import Generator
+    from outlines.types import CFG
+    cfg_str = "..."  # Context-Free Grammar as a string
+    generator = Generator(model, CFG(cfg_str))
+    ```
+
+    You can then call the generator created with a prompt to generate text
+    that matches the Context-Free Grammar.
+
+    Arguments
+    ---------
+    model:
+        The Outlines v0 model to use to generate text.
+    cfg_str:
+        The Context-Free Grammar to use to generate text.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+
+    warnings.warn("""
+        The `cfg` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Use the `Generator` object instead:
+        ```python
+        from outlines import Generator
+        from outlines.types import CFG
+        cfg_str = '...'  # Context-Free Grammar as a string
+        generator = Generator(model, CFG(cfg_str))
+        ```
+        You can then call the generator created with a prompt to generate
+        text in the language of a Context-Free Grammar.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use CFG-based structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API."
+        )
+    elif isinstance(model, LlamaCpp):
+        raise NotImplementedError(
+            "Not yet available due to bug in llama_cpp tokenizer"
+        )
+    elif isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, CFG(cfg_str), sampler)
+    return GeneratorV0Adapter(model, CFG(cfg_str), sampler)
diff --git a/outlines/v0_legacy/generate/choice.py b/outlines/v0_legacy/generate/choice.py
new file mode 100644
index 000000000..eff2fcffd
--- /dev/null
+++ b/outlines/v0_legacy/generate/choice.py
@@ -0,0 +1,126 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import re
+import json as pyjson
+import warnings
+from enum import Enum
+from typing import Callable, List, Union
+
+from outlines.models.openai import OpenAI
+from outlines.types.utils import get_schema_from_enum
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
+from outlines.v0_legacy.generate.json import json
+from outlines.v0_legacy.generate.regex import regex
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+
+def choice(
+    model,
+    choices: Union[List[str], type[Enum]],
+    sampler: Sampler = multinomial(),
+) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter, Callable]:
+    """Generate a choice from a list of options.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Support for it will be removed in v1.1.0.
+    Use the `Generator` object instead:
+
+    ```python
+    from typing import Literal
+    from outlines import Generator
+    generator = Generator(model, Literal["foo", "bar"])
+    ```
+
+    You can then call the generator created with a prompt to generate a
+    choice from the list of options.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    choices:
+        A list of options to choose from.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate a choice from the list of options.
+
+    """
+    warnings.warn("""
+        The `choice` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Use the `Generator` object instead:
+        ```python
+        from outlines import Generator
+        from typing import Literal
+        generator = Generator(model, Literal['foo', 'bar'])
+        ```
+        You can then call the generator created with a prompt to generate
+        a choice from the list of options.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        if isinstance(choices, list):
+            return choice_openai(model, choices, sampler)
+        raise ValueError(
+            "The `choice` function with OpenAI only supports a list of "
+            + "strings as choices."
+        )
+
+    if isinstance(choices, type(Enum)):
+        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
+    else:
+        choices = [re.escape(choice) for choice in choices]  # type: ignore
+        regex_str = r"(" + r"|".join(choices) + r")"
+
+    # We do not want to raise the deprecation warning through the regex function
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            category=DeprecationWarning,
+        )
+        generator = regex(model, regex_str, sampler)
+
+    if isinstance(choices, type(Enum)):
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    else:
+        generator.format_sequence = lambda x: x
+
+    return generator
+
+
+def choice_openai(
+    model,
+    choices: List[str],
+    sampler: Sampler = multinomial(),
+) -> Callable:
+
+    choices_schema = pyjson.dumps(
+        {
+            "type": "object",
+            "properties": {"result": {"type": "string", "enum": choices}},
+            "additionalProperties": False,
+            "required": ["result"],
+        }
+    )
+
+    # We do not want to raise the deprecation warning through the json function
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            category=DeprecationWarning
+        )
+        generator = json(model, choices_schema, sampler)
+
+    def generate_choice(*args, **kwargs): # pragma: no cover
+        return generator(*args, **kwargs)["result"]
+
+    return generate_choice
diff --git a/outlines/v0_legacy/generate/format.py b/outlines/v0_legacy/generate/format.py
new file mode 100644
index 000000000..cd574ac4e
--- /dev/null
+++ b/outlines/v0_legacy/generate/format.py
@@ -0,0 +1,70 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models.openai import OpenAI
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.samplers import Sampler, multinomial
+
+
+def format(
+    model, python_type, sampler: Sampler = multinomial()
+) -> GeneratorV0Adapter:
+    """Generate structured data that can be parsed as a Python type.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Support for it will be removed in v1.1.0.
+    Use the `Generator` object instead::
+
+    ```python
+    from outlines import Generator
+    generator = Generator(model, int)
+    ```
+
+    You can then call the generator created with a prompt to generate a
+    integer.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    python_type:
+        A Python type. The output of the generator must be parseable into
+        this type.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `SequenceGenerator` instance that generates text constrained by the Python type
+    and translates this text into the corresponding type.
+
+    """
+    warnings.warn("""
+        The `format` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Use the `Generator` object instead:
+        ```python
+        from outlines import Generator
+        generator = Generator(model, int)
+        ```
+        You can then call the generator created with a prompt to generate
+        an integer.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use Python type-structured generation with an OpenAI model"
+            + " due to the limitations of the OpenAI API."
+        )
+    elif isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, python_type, sampler)
+    return GeneratorV0Adapter(model, python_type, sampler)
diff --git a/outlines/v0_legacy/generate/fsm.py b/outlines/v0_legacy/generate/fsm.py
new file mode 100644
index 000000000..23a5477fe
--- /dev/null
+++ b/outlines/v0_legacy/generate/fsm.py
@@ -0,0 +1,77 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+import interegular
+
+from outlines.models.openai import OpenAI
+from outlines.types import FSM
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+
+
+def fsm(
+    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
+) -> GeneratorV0Adapter:
+    """Generate text that follows a Finite State Machine.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Support for it will be removed in v1.1.0.
+    Use the `Generator` object instead:
+
+    ```python
+    import interegular
+    from outlines import Generator
+    from outlines.types import FSM
+    fsm = FSM(interegular.parse_pattern(r"[a-z]+").to_fsm())
+    generator = Generator(model, fsm)
+    ```
+
+    You can then call the generator created with a prompt to generate text
+    that follows the Finite State Machine.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    fsm:
+        The interregular `FSM` to use to generate text.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text that follows the Finite State Machine.
+
+    """
+    warnings.warn("""
+        The `fsm` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Use the `Generator` object instead:
+        ```python
+        from outlines import Generator
+        from outlines.types import FSM
+        fsm_str = '...'  # Finite State Machine as a string
+        generator = Generator(model, FSM(fsm_str))
+        ```
+        You can then call the generator created with a prompt to generate
+        text that follows the Finite State Machine.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use FSM-based generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API."
+        )
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, FSM(fsm), sampler)
+    return GeneratorV0Adapter(model, FSM(fsm), sampler)
diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
new file mode 100644
index 000000000..6260b6471
--- /dev/null
+++ b/outlines/v0_legacy/generate/json.py
@@ -0,0 +1,145 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import json as pyjson
+import warnings
+from typing import Callable, Optional, Union
+
+from pydantic import BaseModel
+
+from outlines.models.openai import OpenAI
+from outlines.types.dsl import JsonSchema
+from outlines.types.utils import get_schema_from_signature, is_callable
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+
+
+def json(
+    model,
+    schema_object: Union[str, object, Callable],
+    sampler: Sampler = multinomial(),
+    whitespace_pattern: Optional[str] = None,
+) -> GeneratorV0Adapter:
+    """Generate structured text that follows a JSON Schema.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Support for it will be removed in v1.1.0.
+    Use the `Generator` object instead:
+
+    ```python
+    from outlines import Generator
+    from outlines.types import JsonSchema
+    generator = Generator(model, JsonSchema(schema_object))
+    ```
+
+    You can then call the generator created with a prompt to generate text that
+    matches the JSON Schema.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    schema_object:
+        The JSON Schema to generate data for. Can be a JSON string, a Pydantic
+        model, or a callable that returns a JSON schema.
+    sampler:
+        The sampler defining the sampling parameters.
+    whitespace_pattern
+        Pattern to use for JSON syntactic whitespace (doesn't impact string
+        literals).
+        Example: allow only a single space or newline with
+        `whitespace_pattern=r"[\n ]?"`.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+    warnings.warn("""
+        The `json` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Use the `Generator` object instead:
+        ```python
+        from outlines import Generator
+        from outlines.types import JsonSchema
+        schema_str = '...'  # JSON Schema as a string
+        generator = Generator(model, JsonSchema(schema_str))
+        ```
+        You can then call the generator created with a prompt to generate
+        JSON data that matches the schema.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        return json_openai(model, schema_object, sampler)
+
+    if is_callable(schema_object):
+        json_schema = JsonSchema(get_schema_from_signature(schema_object))  # type: ignore
+    else:
+        json_schema = JsonSchema(schema_object, whitespace_pattern)
+
+    if isinstance(model, TransformersVision):
+        generator = GeneratorVisionV0Adapter(model, json_schema, sampler)
+    else:
+        generator = GeneratorV0Adapter(model, json_schema, sampler)  # type: ignore
+
+    if isinstance(schema_object, type(BaseModel)):
+        setattr(
+            generator,
+            'format_sequence',
+            lambda x: schema_object.model_validate_json(x)
+        )
+    else:
+        setattr(generator, 'format_sequence', lambda x: pyjson.loads(x))
+
+    return generator
+
+
+def json_openai(
+    model,
+    schema_object: Union[str, object, Callable],
+    sampler: Sampler = multinomial(),
+):
+    if not isinstance(sampler, multinomial):
+        raise NotImplementedError(
+            "The OpenAI API does not support any other sampling algorithm "
+            + "than the multinomial sampler."
+        )
+
+    if isinstance(schema_object, type(BaseModel)):
+        schema = schema_object.model_json_schema()
+        schema["additionalProperties"] = False
+        schema = pyjson.dumps(schema)
+        format_sequence = lambda x: schema_object.model_validate_json(x)
+    elif isinstance(schema_object, str):
+        schema = schema_object
+        format_sequence = lambda x: pyjson.loads(x)
+    else:
+        raise ValueError(
+            f"Cannot parse schema {schema_object}. The schema must be either "
+            + "a Pydantic object or a string that contains the JSON Schema "
+            + "specification"
+        )
+
+    # create copied, patched model with normalized json schema set
+    generator = model.new_with_replacements(
+        response_format={
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": pyjson.loads(schema),
+            },
+        }
+    )
+
+    generator.format_sequence = format_sequence
+
+    return generator
diff --git a/outlines/v0_legacy/generate/regex.py b/outlines/v0_legacy/generate/regex.py
new file mode 100644
index 000000000..261f72037
--- /dev/null
+++ b/outlines/v0_legacy/generate/regex.py
@@ -0,0 +1,67 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models.openai import OpenAI
+from outlines.types import Regex
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.samplers import Sampler, multinomial
+
+
+def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
+    """Generate structured text in the language of a regular expression.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Support for it will be removed in v1.1.0.
+    Use the `Generator` object instead:
+
+    ```python
+    from outlines import Generator
+    from outlines.types import Regex
+    generator = Generator(model, Regex(r"^[a-z]+$"))
+    ```
+
+    You can then call the generator created with a prompt to generate text
+    that matches the regular expression.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    regex_str:
+        The regular expression that the output must follow.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+    warnings.warn("""
+        The `regex` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Use the `Generator` object instead:
+        ```python
+        from outlines import Generator
+        from outlines.types import Regex
+        generator = Generator(model, Regex(r'regex_str'))
+        ```
+        You can then call the generator created with a prompt to generate
+        text that matches the regular expression.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use regex-structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API."
+        )
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, regex_str, sampler)
+    return GeneratorV0Adapter(model, regex_str, sampler)
diff --git a/outlines/v0_legacy/generate/text.py b/outlines/v0_legacy/generate/text.py
new file mode 100644
index 000000000..148cf131d
--- /dev/null
+++ b/outlines/v0_legacy/generate/text.py
@@ -0,0 +1,64 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import Union
+
+from outlines.models.openai import OpenAI
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.samplers import Sampler, multinomial
+
+
+def text(model, sampler: Sampler = multinomial()) -> Union[
+    GeneratorV0Adapter, GeneratorVisionV0Adapter, OpenAI
+]:
+    """Generate unconstrained text.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Support for it will be removed in v1.1.0.
+    Use the `Generator` object instead:
+
+    ```python
+    from outlines import Generator
+    generator = Generator(model)
+    ```
+
+    You can then call the generator created with a prompt to generate text.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+    warnings.warn("""
+        The `text` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Use the `Generator` object instead:
+        ```python
+        from outlines import Generator
+        generator = Generator(model)
+        ```
+        You can then call the generator created with a prompt to generate
+        text.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        return model
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, None, sampler)
+    return GeneratorV0Adapter(model, None, sampler)
diff --git a/outlines/v0_legacy/models/__init__.py b/outlines/v0_legacy/models/__init__.py
new file mode 100644
index 000000000..1dc012c37
--- /dev/null
+++ b/outlines/v0_legacy/models/__init__.py
@@ -0,0 +1,21 @@
+from .exllamav2 import ExLlamaV2Model, exl2
+from .llamacpp import llamacpp
+from .mlxlm import mlxlm
+from .openai import azure_openai, openai
+from .transformers import mamba, transformers
+from .transformers_vision import TransformersVision, transformers_vision
+from .vllm_offline import vllm
+
+__all__ = [
+    "ExLlamaV2Model",
+    "TransformersVision",
+    "azure_openai",
+    "exl2",
+    "llamacpp",
+    "mamba",
+    "mlxlm",
+    "openai",
+    "transformers",
+    "transformers_vision",
+    "vllm"
+]
diff --git a/outlines/models/exllamav2.py b/outlines/v0_legacy/models/exllamav2.py
similarity index 79%
rename from outlines/models/exllamav2.py
rename to outlines/v0_legacy/models/exllamav2.py
index efd70f96a..683512002 100644
--- a/outlines/models/exllamav2.py
+++ b/outlines/v0_legacy/models/exllamav2.py
@@ -1,18 +1,30 @@
-import dataclasses
-from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, TypedDict, Union
-
-import torch
-from typing_extensions import Unpack
-
-from outlines.generate.api import GenerationParameters, SamplingParameters
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import (
+    TYPE_CHECKING,
+    Dict,
+    Iterator,
+    List,
+    Optional,
+    Tuple,
+    TypedDict,
+    Union
+)
 
 if TYPE_CHECKING:
     import torch.LongTensor
-    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler
+    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler  # type: ignore
+
+
+def exllamav2_params_adapter(sampling_params: dict, generation_params: dict, kwargs: dict) -> dict:
+    """Argument names mapping is done in the model class for the `exllamav2` model."""
+    kwargs.update(**sampling_params, **generation_params)
+    return kwargs
 
 
 class ExllamaV2Params(TypedDict, total=False):
-    max_tokens: int
     stop_conditions: Optional[List[Union[int, str]]]
     seed: Optional[int]
     gen_settings: "ExLlamaV2Sampler.Settings"
@@ -30,6 +42,8 @@ def convert_token_to_string(self, token):
         return token
 
     def decode(self, token_ids: "torch.LongTensor") -> List[str]:
+        import torch
+
         decoded = self.exl2_tokenizer.decode(
             torch.tensor(token_ids),
             decode_special_tokens=False,
@@ -39,7 +53,7 @@ def decode(self, token_ids: "torch.LongTensor") -> List[str]:
         return decoded
 
 
-class ExLlamaV2Model:
+class ExLlamaV2Model():
     """Represents a `exl2` model."""
 
     def __init__(
@@ -48,18 +62,37 @@ def __init__(
         tokenizer: "OutlinesExLlamaV2Tokenizer",
         max_seq_len: int,
     ):
+        warnings.warn("""
+            The `ExLlamaV2Model` model is deprecated starting from v1.0.0.
+            Support for it will be removed in v1.1.0.
+            As the `exllamav2` library is not compatible with some key features
+            of `outlines`, we decided to remove support for it.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
         self.generator = generator
         self.tokenizer = tokenizer
         self.max_seq_len = max_seq_len
 
+    def __call__(self, *args, **kwargs):
+        raise NotImplementedError(
+            "As a v0 model, the `exllamav2` model does not support the "
+            + "`__call__` method."
+        )
+
+    def stream(self, *args, **kwargs):
+        raise NotImplementedError(
+            "As a v0 model, the `exllamav2` model does not support the "
+            + "`stream` method."
+        )
+
     def prepare_generation_parameters(
         self,
         prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        sampling_parameters: SamplingParameters,
-        structure_logits_processor,
-        **exllamav2_params: Unpack[ExllamaV2Params],
-    ) -> Tuple[ExllamaV2Params, Union[str, List[str]]]:
+        logits_processor,
+        kwargs
+    ) -> Tuple[Dict, Union[str, List[str]]]:
         """Prepare the generation parameters.
 
         `exllamav2` uses different default values
@@ -67,10 +100,12 @@ def prepare_generation_parameters(
         """
         from exllamav2.generator import ExLlamaV2Sampler
 
+        exllamav2_params = {}
+
         if isinstance(prompts, str):
             prompts = [prompts]
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
 
+        max_tokens = kwargs.get("max_tokens")
         if max_tokens is None:
             max_tokens = []
             for prompt in prompts:
@@ -85,28 +120,40 @@ def prepare_generation_parameters(
                 max_tokens for _ in range(len(prompts))
             ]
 
+        stop_at = kwargs.get("stop_at")
         stop_conditions = [self.generator.tokenizer.eos_token_id]
-        if isinstance(generation_parameters.stop_at, str):
-            stop_conditions.append(generation_parameters.stop_at)
-        elif isinstance(generation_parameters.stop_at, list):
-            for stop_at in generation_parameters.stop_at:
+        if isinstance(stop_at, str):
+            stop_conditions.append(stop_at)
+        elif isinstance(stop_at, list):
+            for stop_at in stop_at:
                 stop_conditions.append(stop_at)
         exllamav2_params["stop_conditions"] = stop_conditions
-        exllamav2_params["seed"] = seed
+
+        exllamav2_params["seed"] = kwargs.get("seed")
 
         gen_settings = ExLlamaV2Sampler.Settings()
-        if sampling_parameters.temperature is not None:
-            gen_settings.temperature = sampling_parameters.temperature
-        if sampling_parameters.top_p is not None:
-            gen_settings.top_p = sampling_parameters.top_p
-        if sampling_parameters.top_k is not None:
-            gen_settings.top_k = sampling_parameters.top_k
-        gen_settings.logits_processor = structure_logits_processor
+
+        temperature = kwargs.get("temperature")
+        if temperature is not None:
+            gen_settings.temperature = temperature
+
+        top_p = kwargs.get("top_p")
+        if top_p is not None:
+            gen_settings.top_p = top_p
+
+        top_k = kwargs.get("top_k")
+        if top_k is not None:
+            gen_settings.top_k = top_k
+
+        gen_settings.logits_processor = logits_processor
+
         exllamav2_params["gen_settings"] = gen_settings
-        if sampling_parameters.num_samples > 1:
-            prompts = prompts * sampling_parameters.num_samples
+
+        num_samples = kwargs.get("num_samples")
+        if num_samples is not None and num_samples > 1:
+            prompts = prompts * num_samples
             exllamav2_params["max_new_tokens"] = (
-                exllamav2_params["max_new_tokens"] * sampling_parameters.num_samples
+                exllamav2_params["max_new_tokens"] * num_samples
             )
 
         if len(prompts) == 1:
@@ -115,7 +162,7 @@ def prepare_generation_parameters(
         return exllamav2_params, prompts
 
     def reformat_output(
-        self, output: Union[str, List[str]], sampling_parameters: SamplingParameters
+        self, output: Union[str, List[str]], sampling_parameters
     ):
         """
         The purpose of this function is to reformat the output from exllamav2's output format to outline's output format.
@@ -148,16 +195,13 @@ def reformat_output(
     def generate(
         self,
         prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **exllamav2_params: Unpack[ExllamaV2Params],
+        logits_processor,
+        **kwargs
     ) -> Union[str, List[str]]:
         exllamav2_params, prompts = self.prepare_generation_parameters(
             prompts,
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
+            logits_processor,
+            kwargs
         )
         """
         In exllamav2, it needs the max amount of new tokens generated.
@@ -177,23 +221,20 @@ def generate(
             seed=exllamav2_params["seed"],
         )
 
-        return self.reformat_output(output, sampling_parameters)
+        return self.reformat_output(output, kwargs)
 
-    def stream(
+    def generate_stream(
         self,
         prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **exllamav2_params: Unpack[ExllamaV2Params],
+        logits_processor,
+        **kwargs
     ) -> Iterator[Union[str, List[str]]]:
         from exllamav2.generator import ExLlamaV2DynamicJob
 
         exllamav2_params, prompts = self.prepare_generation_parameters(
             prompts,
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
+            logits_processor,
+            kwargs
         )
 
         order = {}
@@ -237,7 +278,7 @@ def token_generator() -> Iterator[str]:
                         next_text[idx] = text
                     if r["eos"]:
                         next_text[idx] = ""
-                yield self.reformat_output(next_text, sampling_parameters)
+                yield self.reformat_output(next_text, kwargs)
             return
 
         return token_generator()
@@ -291,7 +332,7 @@ def exl2(
 
     """
     try:
-        from exllamav2 import (
+        from exllamav2 import (  # type: ignore
             ExLlamaV2,
             ExLlamaV2Cache,
             ExLlamaV2Cache_Q4,
diff --git a/outlines/v0_legacy/models/llamacpp.py b/outlines/v0_legacy/models/llamacpp.py
new file mode 100644
index 000000000..6c29630d5
--- /dev/null
+++ b/outlines/v0_legacy/models/llamacpp.py
@@ -0,0 +1,147 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import Optional
+
+from outlines.models import LlamaCpp
+
+
+def llamacpp_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `llamacpp` model
+    to the new `Generator` API.
+    """
+    # generation_params
+    if (
+        generation_params.get("stop_at") is not None
+        and kwargs.get("stop") is None
+    ):
+        kwargs["stop"] = generation_params["stop_at"]
+
+    if (
+        generation_params.get("seed") is not None
+        and kwargs.get("seed") is None
+    ):
+        kwargs["seed"] = generation_params["seed"]
+
+    if (
+        generation_params.get("max_tokens") is not None
+        and kwargs.get("max_tokens") is None
+    ):
+        kwargs["max_tokens"] = generation_params["max_tokens"]
+    else:
+        kwargs["max_tokens"] = -1
+
+    # sampling_params
+    if sampling_params.get("sampler") == "beam_search":
+        raise NotImplementedError(
+            "The `llama_cpp_python` library does not support Beam Search."
+        )
+
+    if sampling_params.get("num_samples") != 1:
+        raise NotImplementedError(
+            "The `llama_cpp_python` library does not allow to take several "
+            + "samples."
+        )
+
+    if (
+        sampling_params.get("top_p") is not None
+        and kwargs.get("top_p") is None
+    ):
+        kwargs["top_p"] = sampling_params["top_p"]
+    else:
+        kwargs["top_p"] = 1.0
+
+    if (
+        sampling_params.get("top_k") is not None
+        and kwargs.get("top_k") is None
+    ):
+        kwargs["top_k"] = sampling_params["top_k"]
+    else:
+        kwargs["top_k"] = -1
+
+    if (
+        sampling_params.get("temperature") is not None
+        and kwargs.get("temperature") is None
+    ):
+        kwargs["temperature"] = sampling_params["temperature"]
+    else:
+        kwargs["temperature"] = 1.0
+
+    # model specific params
+    if kwargs.get("min_p") is None:
+        kwargs["min_p"] = 0.0
+
+    if kwargs.get("repeat_penalty") is None:
+        kwargs["repeat_penalty"] = 1.0
+
+    # The choice to stream or not should happen via the high-level API
+    kwargs.pop("stream", None)
+
+    return kwargs
+
+
+def llamacpp(
+    repo_id: str, filename: Optional[str] = None, **llamacpp_model_params
+) -> LlamaCpp:
+    """Load a model from the `llama-cpp-python` library.
+
+    We use the `Llama.from_pretrained` classmethod that downloads models
+    directly from the HuggingFace hub, instead of asking users to specify
+    a path to the downloaded model. One can still load a local model
+    by initializing `llama_cpp.Llama` directly.
+
+    Parameters
+    ----------
+    repo_id
+        The name of the model repository.
+    filename:
+        A filename of glob pattern to match the model file in the repo.
+    llama_cpp_model_params
+        Llama-specific model parameters. See the `llama-cpp-python` documentation
+        for the full list: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__
+
+    """
+    from llama_cpp import Llama
+
+    warnings.warn("""
+        The `llamacpp` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Instead, you should instantiate a `LlamaCpp` model with the
+        `outlines.from_llamacpp` function that takes a `llama_cpp.Llama`
+        instance as argument. For example:
+        ```python
+        from llama_cpp import Llama
+        from outlines import from_llamacpp
+        model = from_llamacpp(Llama.from_pretrained(
+            repo_id="<repo_id>",
+            filename="<filename>",
+        ))
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    # Default to using the model's full context length
+    if "n_ctx" not in llamacpp_model_params:
+        llamacpp_model_params["n_ctx"] = 0
+
+    if "verbose" not in llamacpp_model_params:
+        llamacpp_model_params["verbose"] = False
+
+    # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
+    if "tokenizer" not in llamacpp_model_params:
+        warnings.warn(
+            "The pre-tokenizer in `llama.cpp` handles unicode improperly "
+            + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
+            + "Outlines may raise a `RuntimeError` when building the regex index.\n"
+            + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
+            + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
+        )
+
+    model = Llama.from_pretrained(repo_id, filename, **llamacpp_model_params)
+
+    return LlamaCpp(model)
diff --git a/outlines/v0_legacy/models/mlxlm.py b/outlines/v0_legacy/models/mlxlm.py
new file mode 100644
index 000000000..a69594173
--- /dev/null
+++ b/outlines/v0_legacy/models/mlxlm.py
@@ -0,0 +1,153 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+from typing import Optional
+import warnings
+
+from outlines.models import MLXLM
+
+
+def mlxlm_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `mlxlm` model
+    to the new `Generator` API.
+    """
+    # generation_params
+    if generation_params.get("stop_at") is not None:
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support `stop_at`."
+        )
+
+    if generation_params.get("seed") is not None:
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support `seed`."
+        )
+
+    if (
+        generation_params.get("max_tokens") is not None
+        and kwargs.get("max_tokens") is None
+    ):
+        kwargs["max_tokens"] = generation_params["max_tokens"]
+    else:
+        kwargs["max_tokens"] = int(1e9)
+
+    # sampling_params
+    if sampling_params.get("sampler") == "beam_search":
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support Beam Search."
+        )
+
+    if sampling_params.get("num_samples") != 1:
+        raise NotImplementedError(
+            "The `mlx-lm` library does not allow to take several samples."
+        )
+
+    if (
+        sampling_params.get("top_k") is not None
+        or kwargs.get("top_k") is not None
+    ):
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support top_k."
+        )
+
+    if (
+        sampling_params.get("top_p") is not None
+        or kwargs.get("top_p") is not None
+    ):
+        kwargs.pop("top_p", None)
+        warnings.warn(
+            """
+            The `top_p` parameter is not available anymore.
+            The argument will be ignored. Sorry for the inconvenience.
+            Please migrate to the v1 of Outlines.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+
+    if (
+        sampling_params.get("temperature") is not None
+        or kwargs.get("temperature") is not None
+    ):
+        kwargs.pop("temperature", None)
+        warnings.warn(
+            """
+            The `temperature` parameter is not available anymore.
+            The argument will be ignored. Sorry for the inconvenience.
+            Please migrate to the v1 of Outlines.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+
+    return kwargs
+
+
+def mlxlm(
+    model_name: str,
+    tokenizer_config: dict = {},
+    model_config: dict = {},
+    adapter_path: Optional[str] = None,
+    lazy: bool = False,
+):
+    """Instantiate a model from the `mlx_lm` library and its tokenizer.
+
+    Signature adapted from
+    https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L422
+
+    Parameters
+    ----------
+    Args:
+        path_or_hf_repo (Path): The path or the huggingface repository to load the model from.
+        tokenizer_config (dict, optional): Configuration parameters specifically for the tokenizer.
+            Defaults to an empty dictionary.
+        model_config(dict, optional): Configuration parameters specifically for the model.
+            Defaults to an empty dictionary.
+        adapter_path (str, optional): Path to the LoRA adapters. If provided, applies LoRA layers
+            to the model. Default: ``None``.
+        lazy (bool): If False eval the model parameters to make sure they are
+            loaded in memory before returning, otherwise they will be loaded
+            when needed. Default: ``False``
+
+    Returns
+    -------
+    A `MLXLM` model instance.
+
+    """
+    warnings.warn("""
+        The `mlxlm` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Instead, you should instantiate a `MLXLM` model with the
+        `outlines.from_mlxlm` function that takes a mlxlm model
+        and a tokenizer as arguments. For example:
+        ```python
+        from mlx_lm import load
+        from outlines import from_mlxlm
+        model = from_mlxlm(
+            load("<model_name>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    try:
+        import mlx.core as mx
+        import mlx_lm
+    except ImportError:
+        raise ImportError(
+            "The `mlx_lm` library needs to be installed in order to use `mlx_lm` models."
+        )
+    if not mx.metal.is_available():
+        raise RuntimeError("You cannot use `mlx_lm` without Apple Silicon (Metal)")
+
+    model, tokenizer = mlx_lm.load(
+        model_name,
+        tokenizer_config=tokenizer_config,
+        model_config=model_config,
+        adapter_path=adapter_path,
+        lazy=lazy,
+    )
+    return MLXLM(model, tokenizer)
diff --git a/outlines/v0_legacy/models/openai.py b/outlines/v0_legacy/models/openai.py
new file mode 100644
index 000000000..46546b003
--- /dev/null
+++ b/outlines/v0_legacy/models/openai.py
@@ -0,0 +1,64 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import functools
+from dataclasses import replace
+from typing import Optional
+
+from outlines.models.openai import OpenAI, OpenAIConfig
+
+
+@functools.singledispatch
+def openai(model_or_client, *args, **kwargs):
+    if len(args) == 1 and isinstance(args[0], OpenAIConfig):
+        return OpenAI(model_or_client, args[0], **kwargs)
+    elif kwargs.get("config"):
+        return OpenAI(model_or_client, kwargs.pop("config"), **kwargs)
+    else:
+        return OpenAI(model_or_client, OpenAIConfig(), **kwargs)
+
+
+@openai.register(str)
+def openai_model(
+    model_name: str,
+    config: Optional[OpenAIConfig] = None,
+    **openai_client_params,
+):
+    try:
+        from openai import AsyncOpenAI
+    except ImportError:
+        raise ImportError(
+            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
+        )
+
+    if config is not None:
+        config = replace(config, model=model_name)  # type: ignore
+    else:
+        config = OpenAIConfig(model=model_name)
+
+    client = AsyncOpenAI(**openai_client_params)
+
+    return OpenAI(client, config)
+
+
+def azure_openai(
+    deployment_name: str,
+    model_name: Optional[str] = None,
+    config: Optional[OpenAIConfig] = None,
+    **azure_openai_client_params,
+):
+    try:
+        from openai import AsyncAzureOpenAI
+    except ImportError:
+        raise ImportError(
+            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
+        )
+
+    if config is not None:
+        config = replace(config, model=deployment_name)  # type: ignore
+    if config is None:
+        config = OpenAIConfig(model=deployment_name)
+
+    client = AsyncAzureOpenAI(**azure_openai_client_params)
+
+    return OpenAI(client, config)
diff --git a/outlines/v0_legacy/models/transformers.py b/outlines/v0_legacy/models/transformers.py
new file mode 100644
index 000000000..eb936c491
--- /dev/null
+++ b/outlines/v0_legacy/models/transformers.py
@@ -0,0 +1,196 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import Optional
+
+from outlines.models import Transformers
+
+
+def transformers_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `transformers` model
+    to the new `Generator` API.
+    """
+    from transformers import set_seed
+
+    # generation_params
+    if (
+        generation_params.get("stop_at") is not None
+        and kwargs.get("stop_strings") is None
+    ):
+        kwargs["stop_strings"] = generation_params["stop_at"]
+
+    if (
+        generation_params.get("max_tokens") is not None
+        and kwargs.get("max_new_tokens") is None
+    ):
+        kwargs["max_new_tokens"] = generation_params["max_tokens"]
+    else:
+        kwargs["max_new_tokens"] = int(2**30)
+
+    if (
+        generation_params.get("seed") is not None
+        or kwargs.get("seed") is not None
+    ):
+        seed = (
+            generation_params.get("seed")
+            if generation_params.get("seed") is not None
+            else kwargs.get("seed")
+        )
+        set_seed(seed) # global seed
+
+    # sampling_params
+    if sampling_params.get("sampler") == "multinomial":
+        kwargs["do_sample"] = True
+    elif sampling_params.get("sampler") == "beam_search":
+        kwargs["num_beams"] = sampling_params.get("num_samples", 1)
+
+    if (
+        sampling_params.get("num_samples") is not None
+        and kwargs.get("num_return_sequences") is None
+    ):
+        kwargs["num_return_sequences"] = sampling_params["num_samples"]
+    else:
+        kwargs["num_return_sequences"] = 1
+
+    if (
+        sampling_params.get("top_p") is not None
+        and kwargs.get("top_p") is None
+    ):
+        kwargs["top_p"] = sampling_params["top_p"]
+
+    if (
+        sampling_params.get("top_k") is not None
+        and kwargs.get("top_k") is None
+    ):
+        kwargs["top_k"] = sampling_params["top_k"]
+
+    if (
+        sampling_params.get("temperature") is not None
+        and kwargs.get("temperature") is None
+    ):
+        kwargs["temperature"] = sampling_params["temperature"]
+
+    return kwargs
+
+
+def transformers(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    tokenizer_kwargs: dict = {},
+    model_class=None,
+    tokenizer_class=None,
+):
+    """Instantiate a model from the `transformers` library and its tokenizer.
+
+    Parameters
+    ----------
+    model_name
+        The name of the model as listed on Hugging Face's model page.
+    device
+        The device(s) on which the model should be loaded. This overrides
+        the `device_map` entry in `model_kwargs` when provided.
+    model_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the model.
+    tokenizer_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the tokenizer.
+
+    Returns
+    -------
+    A `TransformersModel` model instance.
+
+    """
+    warnings.warn("""
+        The `transformers` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Instead, you should instantiate a `Transformers` model with the
+        `outlines.from_transformers` function that takes a transformers model
+        and a tokenizer as arguments. For example:
+        ```python
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+        from outlines import from_transformers
+        model = from_transformers(
+            AutoModelForCausalLM.from_pretrained("<model_name>"),
+            AutoTokenizer.from_pretrained("<model_name>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if model_class is None or tokenizer_class is None:
+        try:
+            from transformers import AutoModelForCausalLM, AutoTokenizer
+        except ImportError:  # pragma: no cover
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if model_class is None:
+        model_class = AutoModelForCausalLM
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
+
+    if device is not None:
+        model_kwargs["device_map"] = device
+
+    model = model_class.from_pretrained(model_name, **model_kwargs)
+
+    tokenizer_kwargs.setdefault("padding_side", "left")
+    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
+
+    return Transformers(model, tokenizer)
+
+
+def mamba(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    tokenizer_kwargs: dict = {},
+):
+    warnings.warn("""
+        The `mamba` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Instead, you should instantiate a `Transformers` model with the
+        `outlines.from_transformers` function that takes a transformers model
+        and a tokenizer as arguments. For Mamba, you would use a mamba model
+        and a mamba tokenizer. For example:
+        ```python
+        from transformers import MambaForCausalLM, MambaTokenizer
+        from outlines import from_transformers
+        model = from_transformers(
+            MambaForCausalLM.from_pretrained("<repo_id>"),
+            MambaTokenizer.from_pretrained("<repo_id>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    try:
+        from transformers import MambaForCausalLM
+
+    except ImportError:  # pragma: no cover
+        raise ImportError(
+            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
+        )
+
+    # We do not want to raise the deprecation warning through the transformers function
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            category=DeprecationWarning,
+        )
+        return transformers(
+            model_name=model_name,
+            device=device,
+            model_kwargs=model_kwargs,
+            tokenizer_kwargs=tokenizer_kwargs,
+            model_class=MambaForCausalLM,
+    )
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
new file mode 100644
index 000000000..1c7657a19
--- /dev/null
+++ b/outlines/v0_legacy/models/transformers_vision.py
@@ -0,0 +1,107 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+from typing import TYPE_CHECKING, Optional
+import warnings
+
+from outlines.models.transformers import (
+    Transformers,
+    TransformersMultiModalTypeAdapter,
+)
+
+
+class TransformersVision(Transformers):
+    def __init__(self, model, tokenizer, processor):
+        self.processor = processor
+        self.processor.padding_side = "left"
+        self.processor.pad_token = "[PAD]"
+
+        super().__init__(model, tokenizer)
+
+        self.type_adapter = TransformersMultiModalTypeAdapter()
+
+
+    def _prepare_model_inputs(self, model_input, output_type):
+        model_input = self.type_adapter.format_input(model_input)
+        inputs = self.processor(
+            **model_input, padding=True, return_tensors="pt"
+        ).to(self.model.device)
+
+        return model_input["text"], inputs
+
+
+def transformers_vision(
+    model_name: str,
+    model_class,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    processor_kwargs: dict = {},
+    tokenizer_class=None,
+    processor_class=None,
+):
+    """Instantiate a model from the `transformers` library and its tokenizer.
+
+    Parameters
+    ----------
+    model_name
+        The name of the model as listed on Hugging Face's model page.
+    model_class
+        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
+        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
+    device
+        The device(s) on which the model should be loaded. This overrides
+        the `device_map` entry in `model_kwargs` when provided.
+    model_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the model.
+    processor_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the processor.
+
+    Returns
+    -------
+    A `TransformersModel` model instance.
+
+    """
+    warnings.warn("""
+        The `transformers_vision` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Instead, you should instantiate a `TransformersMultiModal` model with
+        the `outlines.from_transformers` function that takes a transformers
+        model and a processor as arguments. For example:
+        ```python
+        from transformers import AutoModelForCausalLM, AutoProcessor
+        from outlines import from_transformers
+        model = from_transformers(
+            AutoModelForCausalLM.from_pretrained("<model_name>"),
+            AutoProcessor.from_pretrained("<model_name>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if processor_class is None or tokenizer_class is None:
+        try:
+            from transformers import AutoProcessor, AutoTokenizer
+        except ImportError:  # pragma: no cover
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if processor_class is None:
+        processor_class = AutoProcessor
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
+
+    if device is not None:
+        model_kwargs["device_map"] = device
+
+    model = model_class.from_pretrained(model_name, **model_kwargs)
+
+    processor_kwargs.setdefault("padding_side", "left")
+    processor_kwargs.setdefault("pad_token", "[PAD]")
+    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+    tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
+
+    return TransformersVision(model, tokenizer, processor)
diff --git a/outlines/v0_legacy/models/vllm_offline.py b/outlines/v0_legacy/models/vllm_offline.py
new file mode 100644
index 000000000..8efb7cfc0
--- /dev/null
+++ b/outlines/v0_legacy/models/vllm_offline.py
@@ -0,0 +1,102 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models import VLLMOffline
+
+
+def vllm_offline_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `vllm` model
+    to the new `Generator` API.
+    """
+    from vllm.sampling_params import SamplingParams
+
+    # vllm has a `sampling_params` argument that is a `SamplingParams` object
+    # this is unrelated to the `sampling_params` argument in the legacy
+    # `Generator` API
+    sampling_params_kwarg = kwargs.get("sampling_params", SamplingParams())
+
+    # generation_params
+    if generation_params.get("stop_at") is not None:
+        if isinstance(generation_params["stop_at"], str):
+            sampling_params_kwarg.stop_strings = [generation_params["stop_at"]]
+        else:
+            sampling_params_kwarg.stop_strings = generation_params["stop_at"]
+
+    if generation_params.get("seed") is not None:
+        sampling_params_kwarg.seed = generation_params["seed"]
+
+    if generation_params.get("max_tokens") is not None:
+        sampling_params_kwarg.max_tokens = generation_params["max_tokens"]
+
+    # sampling_params
+    if sampling_params.get("sampler") == "beam_search":
+        sampling_params_kwarg.use_beam_search = True
+
+    if (
+        sampling_params.get("num_samples") is not None
+        and sampling_params_kwarg.n == 1
+    ):
+        sampling_params_kwarg.n = sampling_params["num_samples"]
+        sampling_params_kwarg.best_of = sampling_params["num_samples"]
+
+    if (
+        sampling_params.get("top_p") is not None
+        and sampling_params_kwarg.top_p == 1.0
+    ):
+        sampling_params_kwarg.top_p = sampling_params["top_p"]
+
+    if (
+        sampling_params.get("top_k") is not None
+        and sampling_params_kwarg.top_k == -1
+    ):
+        sampling_params_kwarg.top_k = sampling_params["top_k"]
+        if sampling_params.get("top_k") == 1:
+            sampling_params_kwarg.repetition_penalty = 0
+
+    if (
+        sampling_params.get("temperature") is not None
+        and sampling_params_kwarg.temperature == 1.0
+    ):
+        sampling_params_kwarg.temperature = sampling_params["temperature"]
+
+    return {"sampling_params": sampling_params_kwarg, **kwargs}
+
+
+def vllm(model_name: str, **vllm_model_params):
+    """Load an offline vLLM model.
+
+    Parameters
+    ---------
+    model_name
+        The name of the model to load from the HuggingFace hub.
+    vllm_model_params
+        vLLM-specific model parameters. See the vLLM code for the full list:
+        https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/llm.py
+
+    """
+    warnings.warn("""
+        The `vllm` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.1.0.
+        Instead, you should instantiate a `VLLMOffline` model with the
+        `outlines.from_vllm_offline` function that takes a vLLM model name as
+        argument.
+        For example:
+        ```python
+        from vllm import LLM
+        from outlines import from_vllm_offline
+        model = from_vllm_offline(LLM("<model_name>"))
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    from vllm import LLM
+
+    model = LLM(model_name, **vllm_model_params)
+
+    return VLLMOffline(model)
diff --git a/outlines/v0_legacy/samplers.py b/outlines/v0_legacy/samplers.py
new file mode 100644
index 000000000..9f1ff5cc6
--- /dev/null
+++ b/outlines/v0_legacy/samplers.py
@@ -0,0 +1,118 @@
+from dataclasses import dataclass
+from typing import Optional, Protocol
+
+
+@dataclass(frozen=True)
+class SamplingParameters:
+    """Sampling parameters available in Outlines."""
+    sampler: str
+    num_samples: int = 1
+    top_p: Optional[float] = None
+    top_k: Optional[int] = None
+    temperature: Optional[float] = None
+
+
+class Sampler(Protocol):
+    """
+    Protocol for all samplers.
+
+    As outlines now delegates the handling of the generation loop
+    to the models, the samplers are only used to define the sampling
+    parameters that are passed to the model.
+    """
+
+    samples: int
+
+    def sampling_params(self) -> SamplingParameters:
+        ...
+
+
+class GreedySampler:
+    """Greedy Sampling algorithm.
+
+    Greedy sampling consists in choosing the token with the largest
+    likelihood at every step.
+
+    We don't allow more than one sample. We could attribute this a meaning, for
+    instance the k-th sample represents the k-th most likely token. In which
+    case it would be equivalent to beam search without the sequence weights.
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence.
+
+    """
+
+    def __init__(self):
+        self.samples = 1
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters("greedy", self.samples, None, None, 0.0)
+
+
+greedy = GreedySampler
+
+
+class MultinomialSampler:
+    """Multinomial sampling algorithm.
+
+    Multinomial sampling consists in randomly sampling the next token assuming
+    its distribution is a Categorical distribution parametrized by the
+    next-token logits.
+
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence.
+
+    """
+
+    def __init__(
+        self,
+        samples: int = 1,
+        *,
+        top_k: Optional[int] = None,
+        top_p: Optional[float] = None,
+        temperature: Optional[float] = None,
+    ):
+        self.samples = samples
+        self.top_k = top_k
+        self.top_p = top_p
+        self.temperature = temperature
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters(
+            "multinomial",
+            self.samples,
+            self.top_p,
+            self.top_k,
+            self.temperature,
+        )
+
+
+multinomial = MultinomialSampler
+
+
+class BeamSearchSampler:
+    """Beam Search sampling algorithm.
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence. Equivalent to the
+        number of beams.
+    """
+
+    def __init__(self, beams: int = 1):
+        self.samples = beams
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
+
+
+beam_search = BeamSearchSampler
diff --git a/pyproject.toml b/pyproject.toml
index dfed39973..9216216d2 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -28,36 +28,42 @@ dependencies = [
    "jinja2",
    "lark",
    "nest_asyncio",
-   "numpy",
    "cloudpickle",
    "diskcache",
    "pydantic>=2.0",
    "referencing",
    "jsonschema",
    "requests",
+   "pillow",
    "tqdm",
    "typing_extensions",
    "iso3166",
    "airportsdata",
-   "torch",
    "outlines_core==0.1.26",
    "genson",
+   "torch",
 ]
 dynamic = ["version"]
 
 [project.optional-dependencies]
-vllm = ["vllm", "transformers", "numpy<2"]
-transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
-mlxlm = ["mlx-lm", "datasets"]
+anthropic = ["anthropic"]
+dottxt = ["dottxt"]
+gemini = ["google-genai"]
+llamacpp = ["huggingface-hub", "llama-cpp-python"]
+mlxlm = ["datasets", "mlx", "mlx-lm"]
+ollama = ["ollama"]
 openai = ["openai"]
-llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
-exllamav2 = ["exllamav2"]
+sglang = ["openai"]
+tgi = ["huggingface_hub"]
+transformers = ["accelerate", "datasets", "transformers"]
+vllm = ["openai"]
 test = [
     "pre-commit",
     "pytest",
     "pytest-benchmark",
     "pytest-cov",
     "pytest-mock",
+    "pytest-asyncio",
     "coverage[toml]>=5.1",
     "diff-cover",
     "accelerate",
@@ -68,18 +74,22 @@ test = [
     "huggingface_hub",
     "openai>=1.0.0",
     "datasets",
-    "transformers",
+    "anthropic",
+    "google-genai",
+    "transformers<=4.52.1",
     "pillow",
-    "exllamav2",
     "jax",
+    "flax",
+    "numpy>=2.0.0,<2.2.0",
+    "torch",
+    "tensorflow",
+    "tf-keras",
+    "ollama",
+    "dottxt",
+    "sentencepiece",
+    "mkdocs_gen_files",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
-serve = [
-    "vllm>=0.3.0",
-    "uvicorn",
-    "fastapi",
-    "pydantic>=2.0",
-]
 
 [project.urls]
 homepage = "https://github.com/dottxt-ai/outlines"
@@ -108,6 +118,8 @@ filterwarnings = [
     "ignore::FutureWarning:huggingface_hub.*",
     "ignore::UserWarning",
     "ignore::DeprecationWarning:pyairports.*",
+    "ignore::DeprecationWarning:jax.*",
+    "ignore::DeprecationWarning:flax.*",
 ]
 
 [tool.mypy]
@@ -116,13 +128,13 @@ enable_incomplete_feature = ["Unpack"]
 
 [[tool.mypy.overrides]]
 module = [
-    "exllamav2.*",
     "jax",
     "jaxlib",
     "jax.numpy",
     "jinja2",
     "jsonschema.*",
-    "openai.*",
+    "anthropic.*",
+    "google.*",
     "mamba_ssm.*",
     "mlx_lm.*",
     "mlx.*",
@@ -141,36 +153,55 @@ module = [
     "lark.*",
     "interegular.*",
     "datasets.*",
+    "openai.*",
     "requests.*",
     "responses.*",
     "vllm.*",
-    "uvicorn.*",
-    "fastapi.*",
     "iso3166.*",
     "airportsdata.*",
     "outlines_core.*",
     "genson",
+    "ollama.*",
+    "dottxt.*",
+    "tensorflow",
+    "tensorflow.*",
+    "tf-keras",
+    "tf-keras.*",
+    "mkdocs_gen_files.*",
 ]
 ignore_missing_imports = true
 
 [tool.coverage.run]
+# we omit the files that require a GPU or Apple Silicon
+# as well as the models that make API calls
 omit = [
     "outlines/_version.py",
+    "outlines/models/anthropic.py",
+    "outlines/models/dottxt.py",
+    "outlines/models/gemini.py",
+    "outlines/models/mlxlm.py",
+    "outlines/models/openai.py",
+    "outlines/models/vllm_offline.py",
+    "outlines/processors/tensor_adapters/mlx.py",
+    "outlines/v0_legacy/models/anthropic.py",
+    "outlines/v0_legacy/models/dottxt.py",
+    "outlines/v0_legacy/models/exllamav2.py",
+    "outlines/v0_legacy/models/gemini.py",
+    "outlines/v0_legacy/models/mlxlm.py",
+    "outlines/v0_legacy/models/openai.py",
+    "outlines/v0_legacy/models/vllm_offline.py",
     "tests/*",
 ]
 branch = true
 relative_files = true
 
 [tool.coverage.report]
-omit = [
-    "tests/*",
-]
+show_missing = true
 exclude_lines = [
     "pragma: no cover",
     "if TYPE_CHECKING:",
     "\\.\\.\\.",
 ]
-show_missing = true
 
 [tool.diff_cover]
 compare_branch = "origin/main"
diff --git a/requirements-doc.txt b/requirements-doc.txt
index 05f9f36ec..91f621d6c 100644
--- a/requirements-doc.txt
+++ b/requirements-doc.txt
@@ -6,4 +6,7 @@ mkdocs-section-index
 mkdocstrings[python]
 mkdocs-git-committers-plugin-2
 mkdocs-git-revision-date-localized-plugin
+mkdocs-redirects
+mkdocs-gen-files
+mkdocs-literate-nav
 mike
diff --git a/scripts/gen_ref_pages.py b/scripts/gen_ref_pages.py
new file mode 100644
index 000000000..48c226442
--- /dev/null
+++ b/scripts/gen_ref_pages.py
@@ -0,0 +1,53 @@
+"""Generate the API reference pages and navigation automatically.
+
+This script is based on the `gen_ref_pages.py` script in the
+[mkdocstrings](https://mkdocstrings.github.io/recipes/#automatic-code-reference-pages) project.
+
+To exclude a file or module from being included in the generated API reference,
+add a part of its path to the `EXCLUDED_FILES` list.
+"""
+
+from pathlib import Path
+
+import mkdocs_gen_files
+
+
+CODEBASE_DIR_NAME = "outlines"
+OUTPUT_DIR_NAME = "api_reference"
+EXCLUDED_FILES = ["_version", "v0_legacy"]
+
+
+nav = mkdocs_gen_files.Nav()
+root = Path(__file__).parent.parent
+src = root / CODEBASE_DIR_NAME
+
+for path in sorted(src.rglob("*.py")):
+    module_path = path.relative_to(src).with_suffix("")
+    doc_path = path.relative_to(src).with_suffix(".md")
+    full_doc_path = Path(OUTPUT_DIR_NAME, doc_path)
+
+    parts = tuple(module_path.parts)
+
+    if any(part in EXCLUDED_FILES for part in parts):
+        continue
+
+    if parts[-1] == "__init__":
+        if len(parts) == 1:
+            doc_path = Path("index.md")
+            full_doc_path = Path(OUTPUT_DIR_NAME, doc_path)
+            parts = (CODEBASE_DIR_NAME,)
+        else:
+            parts = parts[:-1]
+            doc_path = doc_path.with_name("index.md")
+            full_doc_path = full_doc_path.with_name("index.md")
+
+    nav[parts] = doc_path.as_posix()
+
+    with mkdocs_gen_files.open(full_doc_path, "w") as fd:
+        ident = ".".join(parts)
+        fd.write(f"::: {CODEBASE_DIR_NAME}.{ident}")
+
+    mkdocs_gen_files.set_edit_path(full_doc_path, path.relative_to(root))
+
+with mkdocs_gen_files.open(f"{OUTPUT_DIR_NAME}/SUMMARY.md", "w") as nav_file:
+    nav_file.writelines(nav.build_literate_nav())
diff --git a/tests/fsm/test_json_schema.py b/tests/fsm/test_json_schema.py
deleted file mode 100644
index 23864e029..000000000
--- a/tests/fsm/test_json_schema.py
+++ /dev/null
@@ -1,78 +0,0 @@
-import json
-from contextlib import nullcontext
-from enum import Enum
-from functools import partial
-from typing import List
-
-import pytest
-from outlines_core.fsm.json_schema import build_regex_from_schema
-from pydantic import BaseModel, constr
-
-from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
-
-
-def test_function_basic():
-    def test_function(foo: str, bar: List[int]):
-        pass
-
-    result = get_schema_from_signature(test_function)
-    assert result["type"] == "object"
-    assert list(result["properties"].keys()) == ["foo", "bar"]
-    assert result["properties"]["foo"]["type"] == "string"
-    assert result["properties"]["bar"]["type"] == "array"
-    assert result["properties"]["bar"]["items"]["type"] == "integer"
-
-
-def test_function_no_type():
-    def test_function(foo, bar: List[int]):
-        pass
-
-    with pytest.raises(ValueError):
-        get_schema_from_signature(test_function)
-
-
-def test_from_pydantic():
-    class User(BaseModel):
-        user_id: int
-        name: str
-        maxlength_name: constr(max_length=10)
-        minlength_name: constr(min_length=10)
-        value: float
-        is_true: bool
-
-    schema = json.dumps(User.model_json_schema())
-    regex_str = build_regex_from_schema(schema)
-    assert isinstance(regex_str, str)
-
-
-def add(a: float, b: float) -> float:
-    return a + b
-
-
-class MyEnum(Enum):
-    add = partial(add)
-    a = "a"
-    b = 2
-
-
-# if you don't register your function as callable, you will get an empty enum
-class EmptyEnum(Enum):
-    add = add
-
-
-@pytest.mark.parametrize(
-    "enum,expectation",
-    [
-        (MyEnum, nullcontext()),
-        (EmptyEnum, pytest.raises(ValueError)),
-    ],
-)
-def test_enum_schema(enum, expectation):
-    with expectation:
-        schema = get_schema_from_enum(enum)
-        regex_str = build_regex_from_schema(json.dumps(schema))
-        assert isinstance(regex_str, str)
-        assert schema["title"] == enum.__name__
-        assert len(schema["oneOf"]) == len(enum)
-        for elt in schema["oneOf"]:
-            assert type(elt) in [int, float, bool, type(None), str, dict]
diff --git a/tests/fsm/test_types.py b/tests/fsm/test_types.py
deleted file mode 100644
index 4017a817c..000000000
--- a/tests/fsm/test_types.py
+++ /dev/null
@@ -1,22 +0,0 @@
-import datetime as pydatetime
-
-import pytest
-
-from outlines.fsm.types import python_types_to_regex
-from outlines import types
-
-
-@pytest.mark.parametrize(
-    "python_type,custom_type",
-    [
-        (int, types.integer),
-        (float, types.number),
-        (bool, types.boolean),
-        (pydatetime.date, types.date),
-        (pydatetime.time, types.time),
-        (pydatetime.datetime, types.datetime),
-    ],
-)
-def test_python_types(python_type, custom_type):
-    test_regex, _ = python_types_to_regex(python_type)
-    assert custom_type.pattern == test_regex.pattern
diff --git a/tests/generate/__init__.py b/tests/generate/__init__.py
deleted file mode 100644
index e69de29bb..000000000
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
deleted file mode 100644
index abd9c72a4..000000000
--- a/tests/generate/conftest.py
+++ /dev/null
@@ -1,66 +0,0 @@
-from importlib import reload
-
-import pytest
-import torch
-
-
-def is_metal_available():
-    try:
-        import mlx.core as mx
-        import mlx_lm  # noqa: F401
-
-        assert mx.metal.is_available()
-    except (ImportError, AssertionError):
-        return False
-    return True
-
-
-def pytest_collection_modifyitems(config, items):
-    """
-    If mlxlm and Metal aren't available, skip mlxlm tests
-    If CUDA isn't available, skip vllm and transformers_vision
-    """
-    if not torch.cuda.is_available():
-        skip_marker = pytest.mark.skip(
-            reason="Skipping test because CUDA is not available"
-        )
-        for item in items:
-            if "model_fixture" in item.fixturenames:
-                model_param = item.callspec.params.get("model_fixture", None)
-                if (
-                    model_param.startswith("model_transformers_vision")
-                    or model_param.startswith("model_vllm")
-                    or model_param.startswith("model_exllamav2")
-                ):
-                    item.add_marker(skip_marker)
-
-    if not is_metal_available():
-        skip_marker = pytest.mark.skip(
-            reason="Skipping test because mlx-lm or Metal are not available"
-        )
-        for item in items:
-            if "model_fixture" in item.fixturenames:
-                model_param = item.callspec.params.get("model_fixture", None)
-                if model_param.startswith("model_mlxlm"):
-                    item.add_marker(skip_marker)
-
-
-@pytest.fixture
-def temp_cache_dir():
-    import os
-    import tempfile
-
-    import outlines.caching
-    import outlines.fsm.guide
-
-    with tempfile.TemporaryDirectory() as tempdir:
-        os.environ["OUTLINES_CACHE_DIR"] = tempdir
-        outlines.caching.get_cache.cache_clear()
-        reload(outlines)
-        reload(outlines.fsm.guide)
-        cache_status = outlines.caching._caching_enabled
-        try:
-            outlines.caching._caching_enabled = True
-            yield
-        finally:
-            outlines.caching._caching_enabled = cache_status
diff --git a/tests/generate/test_api.py b/tests/generate/test_api.py
deleted file mode 100644
index 7188022f5..000000000
--- a/tests/generate/test_api.py
+++ /dev/null
@@ -1,33 +0,0 @@
-from io import BytesIO
-from urllib.request import urlopen
-
-import pytest
-from PIL import Image  # type: ignore
-
-from outlines.generate.api import VisionSequenceGeneratorAdapter
-
-IMG_URI = "https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg"
-PIL_IMG = Image.open(BytesIO(urlopen(IMG_URI).read())).convert("RGB")
-
-
-@pytest.mark.parametrize(
-    "prompts,media,type_error",
-    [
-        ("single prompt", [PIL_IMG], False),
-        (["prompt0", "prompt1"], [[PIL_IMG], [PIL_IMG]], False),
-        ("single prompt", [PIL_IMG, PIL_IMG], False),
-        (["prompt0", "prompt1"], [[PIL_IMG, PIL_IMG], [PIL_IMG]], False),
-        ("single prompt", "this isn't an image, it's a string", True),
-        ("single prompt", PIL_IMG, True),
-        (["prompt0", "prompt1"], [PIL_IMG], True),
-        (["prompt0", "prompt1"], [[PIL_IMG]], True),
-        (["prompt0", "prompt1"], [[[PIL_IMG]], [[PIL_IMG]]], True),
-    ],
-)
-def test_vision_sequence_generator_validate_types(prompts, media, type_error):
-    """Ensure inputs are validated correctly"""
-    if type_error:
-        with pytest.raises(TypeError):
-            VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
-    else:
-        VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
deleted file mode 100644
index 3d36faa5d..000000000
--- a/tests/generate/test_generate.py
+++ /dev/null
@@ -1,439 +0,0 @@
-import contextlib
-import re
-from enum import Enum
-
-import pytest
-
-import outlines.generate as generate
-import outlines.models as models
-import outlines.samplers as samplers
-
-##########################################
-# Model Fixtures
-##########################################
-
-
-@pytest.fixture(scope="session")
-def model_llamacpp(tmp_path_factory):
-    return models.llamacpp(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-    )
-
-
-@pytest.fixture(scope="session")
-def model_exllamav2(tmp_path_factory):
-    from huggingface_hub import snapshot_download
-
-    tmp_dir = tmp_path_factory.mktemp("model_download")
-    model_path = snapshot_download(
-        repo_id="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4.6-exl2",
-        cache_dir=tmp_dir,
-    )
-
-    return models.exl2(
-        model_path=model_path,
-        cache_q4=True,
-        paged=False,
-    )
-
-
-@pytest.fixture(scope="session")
-def model_mlxlm(tmp_path_factory):
-    return models.mlxlm("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
-
-
-@pytest.fixture(scope="session")
-def model_mlxlm_phi3(tmp_path_factory):
-    return models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
-
-
-@pytest.fixture(scope="session")
-def model_transformers_random(tmp_path_factory):
-    return models.transformers("hf-internal-testing/tiny-random-gpt2", device="cpu")
-
-
-@pytest.fixture(scope="session")
-def model_transformers_opt125m(tmp_path_factory):
-    return models.transformers("facebook/opt-125m", device="cpu")
-
-
-@pytest.fixture(scope="session")
-def model_mamba(tmp_path_factory):
-    return models.mamba(model_name="state-spaces/mamba-130m-hf", device="cpu")
-
-
-@pytest.fixture(scope="session")
-def model_bart(tmp_path_factory):
-    from transformers import AutoModelForSeq2SeqLM
-
-    return models.transformers(
-        "facebook/bart-base", device="cpu", model_class=AutoModelForSeq2SeqLM
-    )
-
-
-@pytest.fixture(scope="session")
-def model_transformers_vision(tmp_path_factory):
-    import torch
-    from transformers import LlavaNextForConditionalGeneration
-
-    return models.transformers_vision(
-        "llava-hf/llava-v1.6-mistral-7b-hf",
-        model_class=LlavaNextForConditionalGeneration,
-        device="cuda",
-        model_kwargs=dict(
-            torch_dtype=torch.bfloat16,
-            load_in_4bit=True,
-            low_cpu_mem_usage=True,
-        ),
-    )
-
-
-@pytest.fixture(scope="session")
-def model_vllm(tmp_path_factory):
-    return models.vllm("facebook/opt-125m", gpu_memory_utilization=0.1)
-
-
-# TODO: exllamav2 failing in main, address in https://github.com/dottxt-ai/outlines/issues/808
-# TODO: t5 tokenizer doesn't work with streaming
-"""
-@pytest.fixture(scope="session")
-def model_exllamav2(tmp_path_factory):
-    return models.exllamav2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        device="cpu"
-    )
-
-@pytest.fixture(scope="session")
-def model_t5(tmp_path_factory):
-    from transformers import AutoModelForSeq2SeqLM
-
-    return models.transformers(
-        "EleutherAI/pile-t5-base", device="cpu", model_class=AutoModelForSeq2SeqLM
-    )
-"""
-
-
-ALL_MODEL_FIXTURES = (
-    "model_llamacpp",
-    "model_exllamav2",
-    "model_mlxlm",
-    "model_mlxlm_phi3",
-    "model_transformers_random",
-    "model_transformers_opt125m",
-    "model_mamba",
-    "model_bart",
-    "model_transformers_vision",
-    "model_vllm",
-)
-
-
-class MyEnum(Enum):
-    foo = "foo"
-    bar = "bar"
-    baz = "baz"
-
-
-ALL_SAMPLE_CHOICES_FIXTURES = (
-    ["foo", "bar", "baz"],
-    MyEnum,
-)
-
-
-##########################################
-# Stuctured Generation Inputs
-##########################################
-
-
-@pytest.fixture()
-def sample_schema():
-    from pydantic import BaseModel, conint, conlist, constr
-
-    class SampleSchema(BaseModel):
-        title: constr(max_length=10)
-        numbers: conlist(conint(strict=True), min_length=3, max_length=3)
-        labels: conlist(constr(min_length=1, max_length=5), min_length=3, max_length=3)
-
-    return SampleSchema
-
-
-@pytest.fixture()
-def sample_choices():
-    return ["foo", "bar", "baz"]
-
-
-@pytest.fixture()
-def sample_lark_grammar():
-    # from https://github.com/lark-parser/lark/blob/master/docs/grammar.md
-    return """
-    ?start: hello_world "!" number
-    hello_world: ("hello" | "world") ~ 3
-    number: ("0".."9") ~ 5
-    thanks: "Thank"i " for testing!"
-    """
-
-
-REGEX_PATTERNS = [
-    "a b c d e",  # ensure proper tokenizer whitespace prefix handling
-    "(123456789)|(abcdefghijklmnop)",  # ensure consistent correct sequence handling during batch
-    r"([a-z]{10})@([a-z]{5})\.([a-z]{3})",  # email example
-]
-
-
-###########################################
-# Model/Generator Pair Behavior Definitions
-###########################################
-
-
-def enforce_not_implemented(model_fixture, *task_names):
-    """
-    Per `NOT_IMPLEMENTED`, mapping, if a model hasn't implemented a task,
-    assert an NotImplementedError is raised. Otherwise, run normally
-    """
-    NOT_IMPLEMENTED = {
-        "stream": ["model_transformers_vision", "model_vllm"],
-        "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-        "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-        "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-        "cfg": ["model_llamacpp"],  # TODO: fix llama_cpp tokenizer
-    }
-    for task_name in task_names:
-        if model_fixture in NOT_IMPLEMENTED.get(task_name, []):
-            return pytest.raises(NotImplementedError)
-    else:
-        return contextlib.nullcontext()
-
-
-def get_inputs(fixture_name, batch_size=None):
-    """Get generator kwargs, just the prompt by default, but include images for transformers_visian"""
-    from io import BytesIO
-    from urllib.request import urlopen
-
-    from PIL import Image  # type: ignore
-
-    prompts = ["abcd", "efgh", "1234", "5678", "foo", "bar", "baz", "bif"]
-    prompts = prompts[0] if batch_size is None else prompts[:batch_size]
-
-    if fixture_name.endswith("_vision"):
-        img_url = "https://python-pillow.org/pillow-perf/static/space_pil_lanczos.png"
-        img = Image.open(BytesIO(urlopen(img_url).read())).convert("RGB")
-
-        if batch_size is None:
-            return {"prompts": f"<image> {prompts}", "media": [img]}
-        else:
-            return {
-                "prompts": [f"<image> {p}" for p in prompts],
-                "media": [[img] for _ in range(batch_size)],
-            }
-
-    else:
-        return {"prompts": prompts}
-
-
-###########################################
-# Tests
-###########################################
-
-
-@pytest.mark.parametrize("sampler_name", ("greedy", "multinomial", "beam_search"))
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_text(request, model_fixture, sampler_name):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model, getattr(samplers, sampler_name)())
-    with enforce_not_implemented(model_fixture, sampler_name):
-        res = generator(**get_inputs(model_fixture), max_tokens=10)
-        assert isinstance(res, str)
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex(request, model_fixture, pattern):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    res = generator(**get_inputs(model_fixture), max_tokens=20)
-    assert re.fullmatch(pattern, res) is not None, res
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_fsm(request, model_fixture, pattern):
-    import interegular
-
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.fsm(model, interegular.parse_pattern(pattern).to_fsm())
-    res = generator(**get_inputs(model_fixture))
-    assert re.fullmatch(pattern, res) is not None, res
-
-
-@pytest.mark.skip(
-    "Fix issues with JSON, some models fail this test https://github.com/dottxt-ai/outlines/issues/985"
-)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_json(request, model_fixture, sample_schema):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.json(model, sample_schema)
-    # asserts valid within call
-    generator(**get_inputs(model_fixture), max_tokens=100)
-
-
-def test_integrate_genson_generate_json(request):
-    from genson import SchemaBuilder
-
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"name": "Toto", "age": 5})
-
-    model = request.getfixturevalue("model_transformers_opt125m")
-
-    generator = generate.json(model, builder)
-    res = generator("Return a json of a young boy")
-
-    assert "name" in res
-    assert "age" in res
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
-def test_generate_choice(request, model_fixture, sample_choices):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.choice(model, sample_choices)
-    res = generator(**get_inputs(model_fixture))
-    if isinstance(sample_choices, type(Enum)):
-        assert res in [elt.value for elt in sample_choices]
-    else:
-        assert res in sample_choices
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
-def test_generate_choice_twice(request, model_fixture, sample_choices):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.choice(model, sample_choices)
-    res = generator(**get_inputs(model_fixture))
-    if isinstance(sample_choices, type(Enum)):
-        assert res in [elt.value for elt in sample_choices]
-    else:
-        assert res in sample_choices
-
-    res = generator(**get_inputs(model_fixture))
-    if isinstance(sample_choices, type(Enum)):
-        assert res in [elt.value for elt in sample_choices]
-    else:
-        assert res in sample_choices
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_format_bool(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.format(model, bool)
-    res = generator(**get_inputs(model_fixture))
-    assert isinstance(res, bool)
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_cfg(request, model_fixture, sample_lark_grammar):
-    from lark import Lark
-
-    from outlines import grammars
-
-    model = request.getfixturevalue(model_fixture)
-    with enforce_not_implemented(model_fixture, "cfg"):
-        generator = generate.cfg(model, sample_lark_grammar)
-        res = generator(**get_inputs(model_fixture))
-        # validate legal with the grammar via lark
-        # TODO: cleanup PartialLark so doesn't modify Lark globally
-        import importlib
-
-        import lark.lark
-
-        importlib.reload(lark.lark)
-        Lark(
-            sample_lark_grammar, parser="lalr", import_paths=[grammars.GRAMMAR_PATH]
-        ).parse(res)
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_text_stream(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model)
-    with enforce_not_implemented(model_fixture, "stream"):
-        for token in generator.stream(**get_inputs(model_fixture), max_tokens=10):
-            assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_stream(request, model_fixture, pattern):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    with enforce_not_implemented(model_fixture, "stream"):
-        output = ""
-        for token in generator.stream(**get_inputs(model_fixture), max_tokens=20):
-            output += token
-        assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_batch_text(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model)
-    with enforce_not_implemented(model_fixture, "batch"):
-        res = generator(**get_inputs(model_fixture, 2), max_tokens=10)
-        assert isinstance(res, list)
-        assert isinstance(res[0], str)
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_batch(request, model_fixture, pattern):
-    """Ensure batch requests work and fsm order is maintained"""
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    with enforce_not_implemented(model_fixture, "batch"):
-        outputs = generator(**get_inputs(model_fixture, 4), max_tokens=20)
-        for output in outputs:
-            assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_batch_stream(request, model_fixture, pattern):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    with enforce_not_implemented(model_fixture, "batch", "stream"):
-        outputs = ["", ""]
-        for tokens in generator.stream(**get_inputs(model_fixture, 2), max_tokens=20):
-            outputs[0] += tokens[0]
-            outputs[1] += tokens[1]
-        for output in outputs:
-            assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_single_multinomial(request, model_fixture, pattern):
-    """Ensure batch requests work and fsm order is maintained"""
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern, sampler=samplers.multinomial(4))
-    with enforce_not_implemented(model_fixture, "multiple_samples"):
-        output_sample_groups = generator(**get_inputs(model_fixture), max_tokens=40)
-        for output in output_sample_groups:
-            assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-@pytest.mark.parametrize("sampler_name", ("multinomial", "beam_search"))
-def test_generate_regex_batch_multi_sample(
-    request, model_fixture, pattern, sampler_name
-):
-    """Ensure batch requests work and fsm order is maintained"""
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(
-        model, pattern, sampler=getattr(samplers, sampler_name)(4)
-    )
-    with enforce_not_implemented(model_fixture, "batch", "multiple_samples"):
-        output_batch_groups = generator(**get_inputs(model_fixture, 4), max_tokens=40)
-        for output_sample_groups in output_batch_groups:
-            for output in output_sample_groups:
-                assert re.fullmatch(pattern, output) is not None, output
diff --git a/tests/generate/test_generator.py b/tests/generate/test_generator.py
deleted file mode 100644
index 5a2edf8dc..000000000
--- a/tests/generate/test_generator.py
+++ /dev/null
@@ -1,497 +0,0 @@
-import math
-from typing import Generator
-
-import pytest
-import torch
-
-from outlines.fsm.guide import Generate
-from outlines.generate.api import SequenceGenerator
-from outlines.generate.generator import (
-    bias_logits,
-    get_allowed_tokens,
-    get_next_fsm_states,
-    is_generation_finished,
-    sequence_generator,
-    update_attention_masks,
-    update_token_ids,
-)
-
-
-def test_sequence_generator_class():
-    class MockFSM:
-        first_state = 0
-
-        def get_next_state(self, state, next_token_ids):
-            return 4
-
-        def get_next_instruction(self, *_):
-            return Generate([4])
-
-        def is_final_state(self, _):
-            return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            # Input: "test"
-            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
-
-        def decode(self, tokens):
-            return ["testx"[i] for i in tokens]
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3, 4]], dtype=torch.float), None
-
-    class sampler:
-        def __init__(self):
-            self.samples = 1
-
-        def __call__(self, biased_logits, *_):
-            return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
-
-    # Stream
-    generator = SequenceGenerator(MockFSM(), MockModel(), sampler(), "cpu")
-    assert generator.device == "cpu"
-    assert isinstance(generator.tokenizer, MockTokenizer)
-    assert isinstance(generator.fsm, MockFSM)
-
-    sequence = generator.stream("test")
-    assert isinstance(sequence, Generator)
-
-    next(sequence)
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-    # Call
-    generator = SequenceGenerator(MockFSM(), MockModel(), sampler(), "cpu")
-    result = generator("test")
-    assert result == "x"
-
-
-def test_sequence_generator_1d_single_iteration():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return 0
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, _):
-            return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3]]),
-        torch.tensor([[1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        [MockFSM()],
-        token_ids,
-        torch.tensor([0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-    result = next(sequence)
-
-    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3]]))
-    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_sequence_generator_1d_several_iterations():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return state + 1
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, state):
-            if state < 2:
-                return False
-            else:
-                return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3]]),
-        torch.tensor([[1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        [MockFSM()],
-        token_ids,
-        torch.tensor([0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-
-    result = next(sequence)
-    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3]]))
-    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
-
-    result = next(sequence)
-    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3, 3]]))
-    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_sequence_generator_2d_single_iteration():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return 0
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, _):
-            return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]), torch.tensor(
-                [[1, 1, 1, 1], [1, 1, 1, 1]]
-            )
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return (
-            torch.argmax(biased_logits, keepdims=True, dim=-1),
-            torch.tensor([0, 1]),
-            None,
-        )
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]),
-        torch.tensor([[1, 1, 1, 1], [1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0, 0]
-    fsms = [MockFSM(), MockFSM()]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        fsms,
-        token_ids,
-        torch.tensor([0, 0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-
-    result = next(sequence)
-    assert torch.equal(
-        result.token_ids, torch.tensor([[0, 1, 2, 3, 3], [4, 5, 6, 7, 2]])
-    )
-    assert torch.equal(
-        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
-    )
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_sequence_generator_2d_several_iterations():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return state + 1
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, state):
-            if state < 2:
-                return False
-            else:
-                return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]), torch.tensor(
-                [[1, 1, 1, 1], [1, 1, 1, 1]]
-            )
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return (
-            torch.argmax(biased_logits, keepdims=True, dim=-1),
-            torch.tensor([0, 1]),
-            None,
-        )
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]),
-        torch.tensor([[1, 1, 1, 1], [1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0, 0]
-    fsms = [MockFSM(), MockFSM()]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        fsms,
-        token_ids,
-        torch.tensor([0, 0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-
-    result = next(sequence)
-    assert torch.equal(
-        result.token_ids, torch.tensor([[0, 1, 2, 3, 3], [4, 5, 6, 7, 2]])
-    )
-    assert torch.equal(
-        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
-    )
-
-    result = next(sequence)
-    assert torch.equal(
-        result.token_ids, torch.tensor([[0, 1, 2, 3, 3, 3], [4, 5, 6, 7, 2, 2]])
-    )
-    assert torch.equal(
-        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
-    )
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_get_next_fsm_states():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return 0
-
-        def copy(self):
-            return self
-
-    result = get_next_fsm_states([MockFSM()], [0], torch.tensor([[0]]))
-    assert result == [0]
-
-    result = get_next_fsm_states(
-        [MockFSM(), MockFSM()], [0, 0], torch.tensor([[0], [0]])
-    )
-    assert result == [0, 0]
-
-
-def test_get_get_next_instructions():
-    class MockFSM:
-        def get_next_instruction(self, _):
-            return Generate([1, 2, 3, 4])
-
-    result = get_allowed_tokens([MockFSM()], [0])
-    assert result == [[1, 2, 3, 4]]
-
-    result = get_allowed_tokens([MockFSM(), MockFSM()], [0, 1])
-    assert result == [[1, 2, 3, 4], [1, 2, 3, 4]]
-
-
-def test_is_generation_finished():
-    class MockFSMFinished:
-        def is_final_state(self, _):
-            return True
-
-    result = is_generation_finished([MockFSMFinished(), MockFSMFinished()], [1, 1])
-    assert result is True
-
-    class MockFSMNotFinished:
-        def is_final_state(self, state):
-            if state == 0:
-                return False
-            else:
-                return True
-
-    result = is_generation_finished(
-        [MockFSMNotFinished(), MockFSMNotFinished()], [0, 1]
-    )
-    assert result is False
-
-
-@pytest.mark.parametrize(
-    "token_ids,next_token_ids,ancestors,expected_result",
-    [
-        (
-            torch.tensor([[1]]),
-            torch.tensor([[2]]),
-            torch.tensor([0]),
-            torch.tensor([[1, 2]]),
-        ),
-        (
-            torch.tensor([[1], [2]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([0, 1]),
-            torch.tensor([[1, 3], [2, 4]]),
-        ),
-        (
-            torch.tensor([[1], [2]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([0, 0]),
-            torch.tensor([[1, 3], [1, 4]]),
-        ),
-        (
-            torch.tensor([[1], [2]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([1, 0]),
-            torch.tensor([[2, 3], [1, 4]]),
-        ),
-        (
-            torch.tensor([[1, 2], [3, 5]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([1, 0]),
-            torch.tensor([[3, 5, 3], [1, 2, 4]]),
-        ),
-    ],
-)
-def test_update_token_ids(token_ids, next_token_ids, ancestors, expected_result):
-    result = update_token_ids(token_ids, next_token_ids, ancestors)
-    assert torch.equal(result, expected_result)
-
-
-@pytest.mark.parametrize(
-    "attention_masks,ancestors,expected_result",
-    [
-        (
-            torch.tensor([[1, 1]], dtype=torch.float),
-            torch.tensor([0]),
-            torch.tensor([[1, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([0, 0]),
-            torch.tensor([[1, 1, 1], [1, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([0, 1]),
-            torch.tensor([[0, 1, 1], [1, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([1, 0]),
-            torch.tensor([[1, 1, 1], [0, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([1, 1]),
-            torch.tensor([[1, 1, 1], [1, 1, 1]], dtype=torch.float),
-        ),
-    ],
-)
-def test_expand_attention_masks(attention_masks, ancestors, expected_result):
-    result = update_attention_masks(attention_masks, ancestors)
-    assert torch.equal(result, expected_result)
-
-
-@pytest.mark.parametrize(
-    "logits,indices_to_mask,expected",
-    [
-        (
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-            [[0, 1, 2, 3]],
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-            [[0, 2, 3]],
-            torch.tensor([[1, -math.inf, 3, 4]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-            [[0, 2]],
-            torch.tensor([[1, -math.inf, 3, -math.inf]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3], [4, 5, 6]], dtype=torch.float),
-            [[1, 2], [0, 1]],
-            torch.tensor([[-math.inf, 2, 3], [4, 5, -math.inf]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3], [4, 5, 6]], dtype=torch.float),
-            [[0, 2], [1]],
-            torch.tensor(
-                [[1, -math.inf, 3], [-math.inf, 5, -math.inf]], dtype=torch.float
-            ),
-        ),
-    ],
-)
-def test_bias_logits(logits, indices_to_mask, expected):
-    masked_logits = bias_logits(logits, indices_to_mask)
-    assert torch.equal(masked_logits, expected)
diff --git a/tests/generate/test_integration_exllamav2.py b/tests/generate/test_integration_exllamav2.py
deleted file mode 100644
index 12c4143b3..000000000
--- a/tests/generate/test_integration_exllamav2.py
+++ /dev/null
@@ -1,363 +0,0 @@
-import importlib
-from unittest.mock import patch
-
-import pytest
-
-import outlines.models as models
-from outlines.generate.api import GenerationParameters, SamplingParameters
-from outlines.models.exllamav2 import ExLlamaV2Model
-
-
-@pytest.fixture(scope="session")
-def model_exllamav2(tmp_path_factory):
-    return models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=False,
-    )
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_import_error(request, model_fixture):
-    with patch.dict("sys.modules", {"exllamav2": None}):
-        with pytest.raises(ImportError):
-            models.exl2(
-                model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-                cache_q4=True,
-                paged=False,
-            )
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_attributes(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    assert hasattr(model, "generator")
-    assert hasattr(model, "tokenizer")
-    assert model.tokenizer.convert_token_to_string(1) == 1
-    assert hasattr(model, "max_seq_len")
-    assert isinstance(model.max_seq_len, int)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_prompt_types(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-    prompt = ["test"]
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_no_max_tokens(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_test_stop_at(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_multisampling(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, list)
-    assert isinstance(output[0], str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_prepare_generation_parameters(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    exllamav2_params, prompts = model.prepare_generation_parameters(
-        prompt, generation_params, sampling_params, structure_logits_processor
-    )
-    assert isinstance(exllamav2_params, dict)
-    assert isinstance(prompts, list)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_prompt_types(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-    prompt = ["test"]
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_no_max_tokens(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_test_stop_at(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_multisampling(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, list)
-        assert isinstance(token[0], str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_seed(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, seed=1, stop_at=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_reformat_output(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-    )
-    output = "test"
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert reformatted_output == output
-    output = ["test"]
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert reformatted_output == output[0]
-    output = ["test", "test"]
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-    )
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert len(reformatted_output) == 2
-    assert reformatted_output[0] == "test"
-    assert reformatted_output[1] == "test"
-    output = ["test", "test"]
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert len(reformatted_output) == 2
-    assert reformatted_output[0] == "test"
-    assert reformatted_output[1] == "test"
-    output = ["test", "test", "test", "test"]
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert len(reformatted_output) == 2
-    assert reformatted_output[0] == ["test", "test"]
-    assert reformatted_output[1] == ["test", "test"]
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_max_chunk_size(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=False,
-        max_chunk_size=128,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_cache_default(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        paged=False,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def is_flash_attn_available():
-    try:
-        importlib.import_module("flash_attn")
-    except (ImportError, AssertionError):
-        return False
-    return True
-
-
-@pytest.mark.skipif(not is_flash_attn_available(), reason="flash-attn is not installed")
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_paged(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=True,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_draft_model(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=False,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_draft_model_cache_default(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        paged=False,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_set_max_seq_len(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        max_seq_len=2048,
-        paged=False,
-        cache_q4=True,
-    )
-    assert isinstance(model, ExLlamaV2Model)
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
deleted file mode 100644
index fd5be2171..000000000
--- a/tests/generate/test_integration_llamacpp.py
+++ /dev/null
@@ -1,374 +0,0 @@
-import datetime
-import re
-
-import pytest
-from pydantic import BaseModel, constr
-
-import outlines.generate as generate
-import outlines.models as models
-import outlines.samplers as samplers
-
-TEST_MODEL = "./llama-test-model/TinyMistral-248M-v2-Instruct.Q4_K_M.gguf"
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    return models.llamacpp(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-    )
-
-
-@pytest.mark.parametrize(
-    "generator_type,params",
-    (
-        (generate.text, []),
-        (generate.regex, ("[0-9]",)),
-        # (generate.cfg, (grammars.arithmetic,)),  # Awaiting CFG fix
-    ),
-)
-def test_llamacpp_generation_api(model, generator_type, params):
-    generator = generator_type(model, *params)
-
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=".")
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res1 = generator("test", seed=1, max_tokens=10)
-    res2 = generator("test", seed=1, max_tokens=10)
-    assert isinstance(res1, str)
-    assert isinstance(res2, str)
-    assert res1 == res2
-
-
-def test_llama_cpp_streaming_api(model):
-    generator = generate.text(model)
-    token_generator = generator.stream("test", max_tokens=10)
-    tokens = [token for token in token_generator]
-    assert len(tokens) <= 10
-    assert isinstance(tokens[0], str)
-
-
-@pytest.mark.xfail(reason="Batch inference is not available in `llama-cpp-python`.")
-def test_llamacpp_batch_inference(model):
-    generator = generate.text(model)
-    res = generator(["test", "test1"])
-    assert len(res) == 2
-
-
-def test_llamacpp_sampling_params(model):
-    generator = generate.text(model)
-
-    params = {
-        "frequency_penalty": 1.0,
-        "presence_penalty": 1.0,
-    }
-    res = generator("test", seed=1, max_tokens=10, **params)
-    assert isinstance(res, str)
-
-
-def test_llamacpp_greedy_sampling(model):
-    sampler = samplers.greedy()
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=20)
-    assert isinstance(res, str)
-
-
-def test_llamacpp_multinomial_sampling(model):
-    sampler = samplers.multinomial()
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(1, temperature=1.0)
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(1, top_k=1)
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(1, top_p=0.5)
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-
-def test_llamacpp_several_samples(model):
-    sampler = samplers.multinomial(3)
-    generator = generate.text(model, sampler)
-    with pytest.raises(NotImplementedError, match="allow to take several samples"):
-        generator("test")
-
-
-def test_llamacpp_beam_search(model):
-    sampler = samplers.beam_search(1)
-    generator = generate.text(model, sampler)
-
-    with pytest.raises(NotImplementedError, match="does not support Beam Search"):
-        generator("test")
-
-
-def test_llamacpp_text_stop(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.text(model)(prompt, stop_at="a", max_tokens=100)
-    assert isinstance(sequence, str)
-    assert sequence.find("a") == -1
-
-
-def test_llamacpp_regex(model):
-    prompt = (
-        "<|im_start|>user\nWrite an email address<|im_end|>\n<|im_start|>assistant\n"
-    )
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # One prompt
-    sequence = generator(prompts=prompt)
-    assert isinstance(sequence, str)
-    assert re.fullmatch(pattern=regex_str, string=sequence) is not None
-
-
-def test_llamacpp_integer(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, int)(prompt, max_tokens=10)
-    assert isinstance(sequence, int)
-    assert sequence != ""
-    int(sequence)
-
-
-def test_llamacpp_float(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, float)(prompt, max_tokens=10)
-    assert isinstance(sequence, float)
-
-    assert sequence != ""
-    float(sequence)
-
-
-def test_llamacpp_bool(model):
-    prompt = (
-        "<|im_start|>user\nIs this True or False?<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, bool)(prompt, max_tokens=10)
-    assert isinstance(sequence, bool)
-
-    assert sequence != ""
-    bool(sequence)
-
-
-def test_llamacpp_date(model):
-    prompt = (
-        "<|im_start|>user\nWhat day is it today?<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, datetime.date)(prompt, max_tokens=20, seed=10)
-    assert isinstance(sequence, datetime.date)
-
-
-def test_llamacpp_time(model):
-    prompt = "<|im_start|>user\nWhat time is it?<|im_end|>\n<|im_start|>assistant\n"
-    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10)
-    assert isinstance(sequence, datetime.time)
-
-
-def test_llamacpp_datetime(model):
-    prompt = "<|im_start|>user\nWhat time is it?<|im_end|>\n<|im_start|>assistant\n"
-    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20)
-    assert isinstance(sequence, datetime.datetime)
-
-
-def test_llamacpp_choice(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.choice(model, ["test", "choice"])(prompt)
-    assert sequence == "test" or sequence == "choice"
-
-
-def test_llamacpp_json_basic(model):
-    prompt = "<|im_start|>user\nOutput some JSON<|im_end|>\n<|im_start|>assistant\n"
-
-    class Spam(BaseModel):
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam, whitespace_pattern="")(
-        prompt, max_tokens=100, temperature=0.0, seed=1
-    )
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.spam, str)
-    assert isinstance(result.fuzz, bool)
-    assert len(result.spam) <= 10
-
-
-def test_llamacpp_json_schema(model):
-    prompt = "<|im_start|>user\nOutput some JSON<|im_end|>\n<|im_start|>assistant\n"
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "boolean"},
-           "bar": {"type": "string", "maxLength": 4}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    result = generate.json(model, schema, whitespace_pattern="")(
-        prompt, max_tokens=100, temperature=0, seed=10
-    )
-    assert isinstance(result, dict)
-    assert isinstance(result["foo"], bool)
-    assert isinstance(result["bar"], str)
-
-
-@pytest.mark.parametrize(
-    "repo,model_path,hf_tokenizer_uri",
-    [
-        ("Qwen/Qwen1.5-0.5B-Chat-GGUF", "*q2*.gguf", "Qwen/Qwen1.5-0.5B-Chat"),
-        ("TheBloke/phi-2-GGUF", "*Q2*.gguf", "microsoft/phi-2"),
-    ],
-)
-def test_byte_tokenizer_regression(repo, model_path, hf_tokenizer_uri):
-    """Reproduce https://github.com/dottxt-ai/outlines/issues/820"""
-    import llama_cpp
-
-    model = models.llamacpp(
-        repo,
-        model_path,
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            hf_tokenizer_uri
-        ),
-    )
-    generator = generate.choice(model, ["skirt", "dress", "pen", "jacket"])
-    generator("Pick the odd word out: skirt, dress, pen, jacket")
-
-
-def test_llama_cpp_pre_tokenizer_remains_broken():
-    """If fails, llama.cpp pre-tokenizer is fixed -> revert #892, remove `with pytest.raises`"""
-    repo = "Qwen/Qwen1.5-0.5B-Chat-GGUF"
-    model_path = "*q2*.gguf"
-
-    model = models.llamacpp(repo, model_path)
-    with pytest.raises(RuntimeError):
-        generate.choice(model, ["skirt", "dress", "pen", "jacket"])
-
-
-@pytest.mark.skip("Caching for guide was temporarily turned off")
-def test_RegexGuide_caching(model, temp_cache_dir):
-    import llama_cpp
-
-    import outlines.caching
-    from outlines.fsm.guide import cached_create_states_mapping
-
-    assert outlines.caching._caching_enabled
-
-    regex = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-    prompt = "What is the IP address of the Google DNS servers? "
-
-    cache = outlines.caching.get_cache()
-
-    # Returns (hits, misses)
-    _ = cache.stats(enable=True)
-    assert cache.statistics
-
-    assert cached_create_states_mapping.__memory__ is cache
-
-    generator = generate.regex(model, regex, sampler=samplers.greedy())
-    assert cache.stats() == (0, 1)
-
-    model_2 = models.llamacpp(
-        "Qwen/Qwen1.5-0.5B-Chat-GGUF",
-        "*q2*.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "Qwen/Qwen1.5-0.5B-Chat"
-        ),
-    )
-    generator_2 = generate.regex(model_2, regex, sampler=samplers.greedy())
-    assert cache.stats() == (0, 2)
-
-    # These two different models and tokenizers should not have the same state
-    # mapping results
-    assert (
-        generator.logits_processor.guide.states_to_token_maps
-        != generator_2.logits_processor.guide.states_to_token_maps
-    )
-
-    generator_3 = generate.regex(model_2, regex, sampler=samplers.greedy())
-    assert cache.stats() == (1, 2)
-    assert (
-        generator_2.logits_processor.guide.states_to_token_maps
-        == generator_3.logits_processor.guide.states_to_token_maps
-    )
-
-    # Just for fun...
-    structured = generator(prompt, max_tokens=30)
-    structured_2 = generator_2(prompt, max_tokens=30)
-
-    assert re.fullmatch(regex, structured)
-    assert re.fullmatch(regex, structured_2)
-    assert structured != structured_2
-
-
-@pytest.mark.xfail(
-    reason="Some versions of the Hermes-2-Pro-Llama-3 model have a broken config"
-)
-def test_tokenizer_vocabulary_decode_sanity():
-    """Assert the decoded newline token (198) is the same as the normalized vocab token"""
-    import llama_cpp
-
-    model = models.llamacpp(
-        "bartowski/Meta-Llama-3-8B-Instruct-GGUF",
-        "Meta-Llama-3-8B-Instruct-IQ1_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B",
-        ),
-    )
-    tokenizer = generate.regex(model, "a").logits_processor.tokenizer
-
-    decoded_nl_token = tokenizer.decode([198])[0]
-    vocab_nl_token = tokenizer.convert_token_to_string(
-        [token for token, token_id in tokenizer.vocabulary.items() if token_id == 198][
-            0
-        ]
-    )
-    assert decoded_nl_token == vocab_nl_token
-
-
-def test_no_length_constraint_when_unset():
-    """Assert that models.llamacpp doesn't have an implicit max_tokens preventing full sequence generation"""
-    import llama_cpp
-
-    model = models.llamacpp(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "Locutusque/TinyMistral-248M-Instruct"
-        ),
-    )
-
-    long_pattern = "abcdefg" * 10
-    generator = generate.regex(model, long_pattern)
-
-    output = generator("a")
-    assert re.match(long_pattern, output)
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
deleted file mode 100644
index 92c5d789c..000000000
--- a/tests/generate/test_integration_transformers.py
+++ /dev/null
@@ -1,569 +0,0 @@
-import datetime
-import re
-from enum import Enum
-from functools import partial
-from typing import List, Union
-
-import pytest
-import torch
-from outlines_core.fsm.regex import reduced_vocabulary
-from pydantic import BaseModel, constr
-
-import outlines.generate as generate
-import outlines.models as models
-from outlines.models.transformers import Transformers, TransformerTokenizer
-from outlines.samplers import beam_search, greedy, multinomial
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    return models.transformers(
-        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
-    )
-
-
-def test_transformers_integration_text(model):
-    sequence = generate.text(model)(
-        "Write a short sentence ", seed=10000, max_tokens=10
-    )
-    assert isinstance(sequence, str)
-    assert model.tokenizer.eos_token not in sequence
-
-    sequence = generate.text(model)(
-        "Write a short sentence ",
-        max_tokens=20,
-        stop_at="a",
-        seed=10000,
-    )
-    assert isinstance(sequence, str)
-
-    prompts = ["Write a short sentence ", "And another one "]
-    sequence = generate.text(model)(
-        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0], str)
-
-
-def test_transformers_integration_text_multiple_samples(model):
-    sampler = multinomial(2)
-
-    sequence = generate.text(model, sampler=sampler)(
-        "Write a short sentence ", seed=10000
-    )
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert model.tokenizer.eos_token not in sequence
-
-    prompts = ["Write a short sentence ", "And another one "]
-    sequence = generate.text(model, sampler=sampler)(
-        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0], list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0][0], str)
-
-
-def test_transformers_integration_streaming(model):
-    sequence = generate.text(model).stream(
-        "Write a short sentence ", max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-
-    token = next(sequence)
-    assert isinstance(token, str)
-
-    remaining = "".join([token for token in sequence])
-    assert isinstance(remaining, str)
-
-    sequence = generate.text(model).stream(
-        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    tokens = next(sequence)
-    assert isinstance(tokens, list)
-    assert isinstance(tokens[0], str)
-    assert isinstance(tokens[1], str)
-
-
-def test_transformers_integration_streaming_batch_samples(model):
-    sampler = multinomial(samples=2)
-
-    sequence = generate.text(model, sampler=sampler).stream(
-        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    tokens = next(sequence)
-    assert isinstance(tokens, list)
-    assert len(tokens) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[0]) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[1]) == 2
-
-
-def test_transformers_integration_streaming_batch_beam_search(model):
-    sampler = beam_search(beams=2)
-
-    sequence = generate.regex(model, r"ab[cd]e", sampler=sampler).stream(
-        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=["c", "d"], seed=10000
-    )
-    tokens = next(sequence)
-    assert isinstance(tokens, list)
-    assert len(tokens) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[0]) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[1]) == 2
-
-
-def test_transformers_integration_text_stop(model):
-    prompt = "Write a short sentence "
-    sequence = generate.text(model)(prompt, stop_at="a", seed=10000)
-    assert sequence[len(prompt) :].find("a") == -1
-
-
-def test_transformers_various_regexes(model):
-    prompt = "Write an email address"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # One prompt
-    sequence = generator(prompt, seed=0)
-    assert re.fullmatch(regex_str, sequence) is not None
-
-
-def test_transformers_various_regexes_prompt_list(model):
-    prompt = "Write an email address"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # Two prompts
-    sequence = generator([prompt, prompt], seed=0)
-    assert re.fullmatch(regex_str, sequence[0]) is not None
-    assert re.fullmatch(regex_str, sequence[1]) is not None
-
-
-def test_transformers_various_regexes_prompt_list_multiple_samples(model):
-    sampler = multinomial(samples=2)
-    prompt = "Write an email address"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str, sampler=sampler)
-
-    # Two prompts
-    sequence = generator([prompt, prompt], seed=0, max_tokens=500)
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert re.fullmatch(regex_str, sequence[0][0]) is not None
-    assert re.fullmatch(regex_str, sequence[0][1]) is not None
-    assert re.fullmatch(regex_str, sequence[1][0]) is not None
-    assert re.fullmatch(regex_str, sequence[1][1]) is not None
-
-
-def test_transformers_various_regexes_prompt_list_beam_search(model):
-    sampler = beam_search(5)
-    prompt_1 = "Write an email address"
-    prompt_2 = "Random"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str, sampler=sampler)
-
-    # Two prompts
-    sequence = generator([prompt_1, prompt_2], seed=0)
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert len(sequence[0]) == 5
-    assert re.fullmatch(regex_str, sequence[0][0]) is not None
-    assert re.fullmatch(regex_str, sequence[0][1]) is not None
-    assert re.fullmatch(regex_str, sequence[1][0]) is not None
-    assert re.fullmatch(regex_str, sequence[1][1]) is not None
-
-
-def test_transformers_integration_integer(model):
-    prompt = "Write a short sentence"
-    sequence = generate.format(model, int)(prompt, max_tokens=10, seed=0)
-
-    assert isinstance(sequence, int)
-
-
-def test_transformers_integration_integer_array(model):
-    prompts = ["Give me a number", "And another one"]
-    sequence = generate.format(model, int)(prompts, max_tokens=10, seed=0)
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0], int)
-    assert isinstance(sequence[1], int)
-
-
-def test_transformers_integration_float(model):
-    prompt = "Write a short sentence"
-    sequence = generate.format(model, float)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, float)
-
-
-def test_transformers_integration_bool(model):
-    prompt = "Is this True or False?"
-    sequence = generate.format(model, bool)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, bool)
-
-
-def test_transformers_integration_date(model):
-    prompt = "What day is it today?"
-    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, datetime.date)
-
-
-def test_transformers_integration_time(model):
-    prompt = "What time is it?"
-    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, datetime.time)
-
-
-def test_transformers_integration_datetime(model):
-    prompt = "What time is it?"
-    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20, seed=0)
-
-    assert sequence != 0
-    assert isinstance(sequence, datetime.datetime)
-
-
-def test_transformers_integration_choice(model):
-    prompt = "Write a short sentence "
-    sequence = generate.choice(model, ["test", "choice"])(prompt, seed=0)
-
-    assert sequence == "test" or sequence == "choice"
-
-
-def test_transformers_integration_with_pad_token():
-    model_name = "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM"
-    model = models.transformers(model_name, device="meta")
-    assert model.tokenizer.pad_token_id == 1
-    assert model.tokenizer.pad_token == "<pad>"
-
-
-def test_transformers_json_basic(model):
-    prompt = "Output some JSON "
-
-    class Spam(BaseModel):
-        foo: int
-        bar: float
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam)(prompt, max_tokens=500, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.foo, int)
-    assert isinstance(result.bar, float)
-    assert isinstance(result.spam, str)
-    assert isinstance(result.fuzz, bool)
-    assert len(result.spam) <= 10
-
-
-def test_transformers_json_schema(model):
-    prompt = "Output some JSON "
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "integer"},
-           "bar": {"type": "string", "maxLength": 4}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    result = generate.json(model, schema)(prompt, max_tokens=500, seed=0)
-    assert isinstance(result, dict)
-    assert isinstance(result["foo"], int)
-    assert isinstance(result["bar"], str)
-
-
-def test_transformers_json_batch(model):
-    prompts = ["Output some JSON ", "Output more JSON"]
-
-    class Spam(BaseModel):
-        foo: int
-        bar: float
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam)(prompts, max_tokens=500, seed=0)
-    assert isinstance(result[0], BaseModel)
-    assert isinstance(result[1], BaseModel)
-
-
-def test_transformers_json_batch_multiple_samples(model):
-    sampler = multinomial(samples=2)
-    prompts = ["Output some JSON ", "Output more JSON"]
-
-    class Spam(BaseModel):
-        foo: int
-        bar: float
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam, sampler=sampler)(
-        prompts, max_tokens=500, seed=0
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert isinstance(result[0][0], BaseModel)
-    assert isinstance(result[0][1], BaseModel)
-    assert isinstance(result[1][0], BaseModel)
-    assert isinstance(result[1][1], BaseModel)
-
-
-def test_transformers_json_str_enum(model):
-    prompt = "Output some JSON "
-
-    class Name(str, Enum):
-        john = "John"
-        marc = "Marc"
-        michel = "Michel"
-
-    class User(BaseModel):
-        user_id: int
-        name: Name
-
-    result = generate.json(model, User)(prompt, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.user_id, int)
-    assert result.name in ["John", "Marc", "Michel"]
-
-
-def test_transformers_json_int_enum(model):
-    prompt = "Output some JSON "
-
-    class Id(int, Enum):
-        one = 1
-        two = 2
-
-    class User(BaseModel):
-        user_id: Id
-
-    result = generate.json(model, User)(prompt, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.user_id, int)
-    assert result.user_id in [1, 2]
-
-
-def add(a: int, b: int) -> int:
-    return a + b
-
-
-def mul(c: float, d: float) -> float:
-    return c * d
-
-
-def test_transformers_json_function_enum(model):
-    prompt = "Output some JSON "
-
-    class Operation(Enum):
-        add = partial(add)
-        mul = partial(mul)
-
-    result = generate.json(model, Operation)(prompt, seed=0)
-    assert isinstance(result, dict)
-    assert len(result) == 2
-    for k, v in result.items():
-        assert k in ["a", "b", "c", "d"]
-        assert isinstance(v, (int, float))
-
-
-def test_transformers_json_array(model):
-    prompt = "Output some JSON "
-
-    class User(BaseModel):
-        user_id: int
-        value: List[float]
-
-    result = generate.json(model, User)(prompt, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.user_id, int)
-    assert isinstance(result.value, list)
-    for value in result.value:
-        assert isinstance(value, float) or isinstance(value, int)
-
-
-@pytest.mark.xfail(reason="The implementation of `anyOf` is incorrect")
-def test_transformers_json_union(model):
-    prompt = "Output some JSON "
-
-    class Spam(BaseModel):
-        foo: int
-        bar: Union[constr(max_length=10), float]
-
-    result = generate.json(model, Spam)(prompt, max_tokens=100, seed=4)
-    assert isinstance(result, BaseModel)
-    assert (
-        isinstance(result.bar, int)
-        or isinstance(result.bar, float)
-        or isinstance(result.bar, str)
-    )
-
-
-def test_transformers_json_function(model):
-    prompt = "Output arguments for the function"
-
-    def function(foo: int, bar: List[int]):
-        return foo + sum(bar)
-
-    sequence = generate.json(model, function)(prompt, max_tokens=100, seed=4)
-    assert isinstance(sequence, dict)
-    assert isinstance(function(**sequence), int)
-
-
-def test_transformers_logits_vocab_size(model):
-    # Artificially increase the weights/logits size relative
-    # to the vocabulary
-    model.model.resize_token_embeddings(pad_to_multiple_of=3)
-
-    assert len(model.tokenizer.vocabulary) == 1024
-    assert model.model.base_model.wte.weight.shape[0] == 1026
-
-    generator = generate.choice(model, ["True", "False"])
-
-    sequence = generator("blah", seed=101)
-    assert sequence == "False"
-
-
-def test_transformers_json_custom_ws(model):
-    prompt = "Output some JSON with newlines"  # try to force model to use newlines
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "integer"},
-           "bar": {"type": "integer"}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    generator = generate.json(model, schema, whitespace_pattern=r"[ ]?")
-    generator.format_sequence = lambda x: x  # patch to return raw text
-    assert "\n" not in generator(prompt, max_tokens=500, seed=0)
-
-
-def test_transformers_reduced_vocabulary_caching():
-    from transformers import AutoTokenizer
-
-    hf_tokenizer = AutoTokenizer.from_pretrained("gpt2")
-    tokenizer = TransformerTokenizer(hf_tokenizer)
-    tokenizer2 = TransformerTokenizer(hf_tokenizer)
-
-    # TODO: We might actually want only one copy of a given tokenizer.
-    assert tokenizer is not tokenizer2
-
-    vocab = reduced_vocabulary(tokenizer)
-    vocab2 = reduced_vocabulary(tokenizer2)
-
-    assert vocab2 is vocab
-
-
-@pytest.mark.skip(reason="Custom Sampler Disabled in Transformers Integration")
-def test_custom_sampler(model):
-    seen = False
-    target_token_ids = model.tokenizer.encode(["c"])[0]
-
-    class biased_sampler:
-        def __init__(self, samples: int = 1):
-            self.samples = samples
-
-        def __call__(
-            logits: torch.DoubleTensor, samples: int, *_
-        ) -> torch.DoubleTensor:
-            nonlocal seen
-
-            if not seen:
-                seen = True
-                return target_token_ids, torch.tensor([0]), None
-            else:
-                return (
-                    torch.tensor([[model.tokenizer.eos_token_id]]),
-                    torch.tensor([0]),
-                    None,
-                )
-
-    generator = generate.choice(model, ["a", "b", "c"], sampler=biased_sampler(1))
-    sequence = generator(
-        """What is 1+1?
-    a. 3
-    b. 4
-    c. 2"""
-    )
-
-    assert sequence == "c"
-
-
-def test_transformers_use_existing_model_and_tokenizer():
-    from transformers import AutoModelForCausalLM, AutoTokenizer
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    hf_tokenizer = AutoTokenizer.from_pretrained(model_name)
-    hf_model = AutoModelForCausalLM.from_pretrained(model_name)
-    model = Transformers(hf_model, hf_tokenizer)
-    sequence = generate.text(model)("Write a short sentence ", seed=10000)
-    assert isinstance(sequence, str)
-
-
-@pytest.mark.skip("Caching for guide was temporarily turned off")
-def test_RegexGuide_caching(temp_cache_dir):
-    import outlines.caching
-    from outlines.fsm.guide import cached_create_states_mapping
-
-    assert outlines.caching._caching_enabled
-
-    regex = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-    prompt = "What is the IP address of the Google DNS servers? "
-
-    cache = outlines.caching.get_cache()
-
-    # Returns (hits, misses)
-    _ = cache.stats(enable=True)
-    assert cache.statistics
-
-    assert cached_create_states_mapping.__memory__ is cache
-
-    model = models.transformers(
-        "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM", device="cpu"
-    )
-    generator = generate.regex(model, regex, sampler=greedy())
-    assert cache.stats() == (0, 1)
-
-    model_2 = models.transformers(
-        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
-    )
-    generator_2 = generate.regex(model_2, regex, sampler=greedy())
-    assert cache.stats() == (0, 2)
-
-    # These two different models and tokenizers should not have the same state
-    # mapping results
-    assert (
-        generator.logits_processor.guide.states_to_token_maps
-        != generator_2.logits_processor.guide.states_to_token_maps
-    )
-
-    generator_3 = generate.regex(model_2, regex, sampler=greedy())
-    assert cache.stats() == (1, 2)
-    assert (
-        generator_2.logits_processor.guide.states_to_token_maps
-        == generator_3.logits_processor.guide.states_to_token_maps
-    )
-
-    # Just for fun...
-    structured = generator(prompt, max_tokens=30)
-    structured_2 = generator_2(prompt, max_tokens=30)
-
-    assert re.fullmatch(regex, structured)
-    assert re.fullmatch(regex, structured_2)
-    assert structured != structured_2
diff --git a/tests/generate/test_integration_transformers_vision.py b/tests/generate/test_integration_transformers_vision.py
deleted file mode 100644
index ee4f84c06..000000000
--- a/tests/generate/test_integration_transformers_vision.py
+++ /dev/null
@@ -1,116 +0,0 @@
-from io import BytesIO
-from urllib.request import urlopen
-
-import pytest
-from PIL import Image
-from transformers import AutoProcessor, LlavaForConditionalGeneration
-
-import outlines
-from outlines.models.transformers_vision import transformers_vision
-
-IMAGE_URLS = [
-    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
-    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
-    "https://upload.wikimedia.org/wikipedia/commons/b/be/Tamias-rufus-001.jpg",
-]
-
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    return transformers_vision(
-        "trl-internal-testing/tiny-LlavaForConditionalGeneration",
-        model_class=LlavaForConditionalGeneration,
-        device="cpu",
-    )
-
-
-@pytest.fixture(scope="session")
-def processor(tmp_path_factory):
-    return AutoProcessor.from_pretrained("llava-hf/llava-interleave-qwen-0.5b-hf")
-
-
-def test_single_image_text_gen(model, processor):
-    conversation = [
-        {
-            "role": "user",
-            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
-        },
-    ]
-    generator = outlines.generate.text(model)
-    sequence = generator(
-        processor.apply_chat_template(conversation),
-        [img_from_url(IMAGE_URLS[0])],
-        seed=10000,
-        max_tokens=10,
-    )
-    assert isinstance(sequence, str)
-
-
-def test_multi_image_text_gen(model, processor):
-    """If the length of image tags and number of images we pass are > 1 and equal,
-    we should yield a successful generation.
-    """
-    conversation = [
-        {
-            "role": "user",
-            "content": [
-                {"type": "text", "text": "What do all these have in common?"},
-            ]
-            + [{"type": "image"} for _ in range(len(IMAGE_URLS))],
-        },
-    ]
-    generator = outlines.generate.text(model)
-    sequence = generator(
-        processor.apply_chat_template(conversation),
-        [img_from_url(i) for i in IMAGE_URLS],
-        seed=10000,
-        max_tokens=10,
-    )
-    assert isinstance(sequence, str)
-
-
-def test_mismatched_image_text_gen(model, processor):
-    """If the length of image tags and number of images we pass are unequal,
-    we should raise an error.
-    """
-    conversation = [
-        {
-            "role": "user",
-            "content": [
-                {"type": "text", "text": "I'm passing 3 images, but only 1 image tag"},
-                {"type": "image"},
-            ],
-        },
-    ]
-    generator = outlines.generate.text(model)
-    with pytest.raises(ValueError):
-        _ = generator(
-            processor.apply_chat_template(conversation),
-            [img_from_url(i) for i in IMAGE_URLS],
-            seed=10000,
-            max_tokens=10,
-        )
-
-
-def test_single_image_choice(model, processor):
-    conversation = [
-        {
-            "role": "user",
-            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
-        },
-    ]
-    choices = ["cat", "dog"]
-    generator = outlines.generate.choice(model, choices)
-    sequence = generator(
-        processor.apply_chat_template(conversation),
-        [img_from_url(IMAGE_URLS[0])],
-        seed=10000,
-        max_tokens=10,
-    )
-    assert isinstance(sequence, str)
-    assert sequence in choices
diff --git a/tests/generate/test_integration_vllm.py b/tests/generate/test_integration_vllm.py
deleted file mode 100644
index d812d3cf2..000000000
--- a/tests/generate/test_integration_vllm.py
+++ /dev/null
@@ -1,243 +0,0 @@
-import datetime
-import re
-
-import pytest
-import torch
-from pydantic import BaseModel, constr
-
-try:
-    from vllm.sampling_params import SamplingParams
-except ImportError:
-    pass
-
-import outlines.generate as generate
-import outlines.grammars as grammars
-import outlines.models as models
-import outlines.samplers as samplers
-
-pytestmark = pytest.mark.skipif(
-    not torch.cuda.is_available(), reason="vLLM models can only be run on GPU."
-)
-
-
-@pytest.fixture(scope="module")
-def model():
-    return models.vllm("gpt2", gpu_memory_utilization=0.5)
-
-
-@pytest.mark.parametrize(
-    "generator_type,params", ((generate.text, []), (generate.regex, ("[0-9]",)))
-)
-def test_vllm_generation_api(model, generator_type, params):
-    generator = generator_type(model, *params)
-
-    res = generator("test")
-    assert isinstance(res, str)
-
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=".")
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res1 = generator("test", seed=1)
-    res2 = generator("test", seed=1)
-    assert isinstance(res1, str)
-    assert isinstance(res2, str)
-    assert res1 == res2
-
-    res = generator(["test", "test1"])
-    assert len(res) == 2
-
-
-def test_vllm_sampling_params(model):
-    generator = generate.text(model)
-
-    sampling_params = SamplingParams(n=2)
-    res = generator("test", sampling_params=sampling_params)
-    assert len(res) == 2
-    assert isinstance(res[0], str)
-    assert isinstance(res[1], str)
-
-    sampling_params = SamplingParams(seed=2)
-    res1 = generator("test", sampling_params=sampling_params)
-    res2 = generator("test", sampling_params=sampling_params)
-    assert res1 == res2
-
-
-def test_vllm_greedy_sampling(model):
-    sampler = samplers.greedy()
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert isinstance(res, str)
-
-
-def test_vllm_multinomial_sampling(model):
-    sampler = samplers.multinomial()
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(3)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert len(res) == 3
-    assert isinstance(res[0], str)
-    assert isinstance(res[1], str)
-
-    sampler = samplers.multinomial(2, top_k=1)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert res[0] == res[1]
-
-    sampler = samplers.multinomial(1, top_p=0.5)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(2, temperature=0.00001)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert res[0] == res[1]
-
-
-def test_vllm_beam_search(model):
-    sampler = samplers.beam_search(1)
-    generator = generate.text(model, sampler)
-    res1 = generator("test")
-    sampler = samplers.greedy()
-    generator = generate.text(model, sampler)
-    res2 = generator("test")
-    assert res1 == res2
-
-    sampler = samplers.beam_search(2)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert len(res) == 2
-    assert res[0] != res[1]
-
-
-def test_vllm_text_stop(model):
-    prompt = "Write a short sentence containing 'You': "
-    sequence = generate.text(model)(prompt, max_tokens=100, seed=10)
-    assert sequence.find("news") != -1
-
-    sequence = generate.text(model)(prompt, stop_at="news", max_tokens=100, seed=10)
-    assert isinstance(sequence, str)
-    assert sequence.find("news") == -1
-
-
-def test_vllm_regex(model):
-    prompt = "Write an email address: "
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # One prompt
-    sequence = generator(prompts=prompt)
-    assert isinstance(sequence, str)
-    assert re.fullmatch(pattern=regex_str, string=sequence) is not None
-
-
-def test_vllm_integer(model):
-    prompt = "Give me an integer: "
-    sequence = generate.format(model, int)(prompt, max_tokens=10)
-    assert isinstance(sequence, int)
-    assert sequence != ""
-    int(sequence)
-
-
-def test_vllm_float(model):
-    prompt = "Give me a floating-point number: "
-    sequence = generate.format(model, float)(prompt, max_tokens=10)
-    assert isinstance(sequence, float)
-
-    assert sequence != ""
-    float(sequence)
-
-
-def test_vllm_bool(model):
-    prompt = "Is this True or False? "
-    sequence = generate.format(model, bool)(prompt, max_tokens=10)
-    assert isinstance(sequence, bool)
-
-    assert sequence != ""
-    bool(sequence)
-
-
-def test_vllm_date(model):
-    prompt = "What day is it today? "
-    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10)
-    assert isinstance(sequence, datetime.date)
-
-
-def test_vllm_time(model):
-    prompt = "What time is it? "
-    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10)
-    assert isinstance(sequence, datetime.time)
-
-
-def test_vllm_datetime(model):
-    prompt = "What time is it? "
-    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20)
-    assert isinstance(sequence, datetime.datetime)
-
-
-def test_vllm_choice(model):
-    prompt = "Which one between 'test' and 'choice'? "
-    sequence = generate.choice(model, ["test", "choice"])(prompt)
-    assert sequence == "test" or sequence == "choice"
-
-
-def test_vllm_json_basic(model):
-    prompt = "Output some JSON. "
-
-    class Spam(BaseModel):
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    sampling_params = SamplingParams(temperature=0)
-    result = generate.json(model, Spam, whitespace_pattern="")(
-        prompt, max_tokens=100, seed=1, sampling_params=sampling_params
-    )
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.spam, str)
-    assert isinstance(result.fuzz, bool)
-    assert len(result.spam) <= 10
-
-
-def test_vllm_json_schema(model):
-    prompt = "Output some JSON. "
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "boolean"},
-           "bar": {"type": "string", "maxLength": 4}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    sampling_params = SamplingParams(temperature=0)
-    result = generate.json(model, schema, whitespace_pattern="")(
-        prompt, max_tokens=100, seed=10, sampling_params=sampling_params
-    )
-    assert isinstance(result, dict)
-    assert isinstance(result["foo"], bool)
-    assert isinstance(result["bar"], str)
-
-
-@pytest.mark.xfail(
-    reason="The CFG logits processor for vLLM has not been implemented yet."
-)
-def test_vllm_cfg(model):
-    prompt = "<|im_start|>user\nOutput a short and valid JSON object with two keys.<|im_end|>\n><|im_start|>assistant\n"
-    result = generate.cfg(model, grammars.arithmetic)(prompt, seed=11)
-    assert isinstance(result, str)
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
new file mode 100644
index 000000000..66b3c74db
--- /dev/null
+++ b/tests/models/test_anthropic.py
@@ -0,0 +1,106 @@
+import io
+from typing import Generator
+
+from anthropic import Anthropic as AnthropicClient
+from PIL import Image
+import pytest
+
+import outlines
+from outlines.models.anthropic import Anthropic
+from outlines.templates import Vision
+
+MODEL_NAME = "claude-3-haiku-20240307"
+
+
+@pytest.fixture(scope="session")
+def model():
+    return Anthropic(AnthropicClient(), MODEL_NAME)
+
+
+@pytest.fixture(scope="session")
+def model_no_model_name():
+    return Anthropic(AnthropicClient())
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+    image.format = "PNG"
+
+    return image
+
+
+def test_init_from_client():
+    client = AnthropicClient()
+
+    # With model name
+    model = outlines.from_anthropic(client, MODEL_NAME)
+    assert isinstance(model, Anthropic)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+    # Without model name
+    model = outlines.from_anthropic(client)
+    assert isinstance(model, Anthropic)
+    assert model.client == client
+    assert model.model_name is None
+
+
+def test_anthropic_wrong_inference_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model = Anthropic(AnthropicClient(), MODEL_NAME)
+        model.generate("prompt", foo=10, max_tokens=1024)
+
+
+def test_anthropic_wrong_input_type():
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(TypeError, match="is not available"):
+        model = Anthropic(AnthropicClient(), MODEL_NAME)
+        model.generate(Foo("prompt"))
+
+
+def test_anthropic_wrong_output_type():
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model = Anthropic(AnthropicClient(), MODEL_NAME)
+        model.generate("prompt", Foo(1))
+
+
+@pytest.mark.api_call
+def test_anthropic_simple_call(model):
+    result = model.generate("Respond with one word. Not more.", max_tokens=1024)
+    assert isinstance(result, str)
+
+
+@pytest.mark.xfail(reason="Anthropic requires the `max_tokens` parameter to be set")
+@pytest.mark.api_call
+def test_anthropic_direct_call(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        model_name=MODEL_NAME,
+        max_tokens=1024,
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_anthropic_simple_vision(model, image):
+    result = model.generate(
+        Vision("What does this logo represent?", image), max_tokens=1024
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_anthopic_streaming(model):
+    result = model.stream("Respond with one word. Not more.", max_tokens=1024)
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
new file mode 100644
index 000000000..0bd1277af
--- /dev/null
+++ b/tests/models/test_dottxt.py
@@ -0,0 +1,131 @@
+import json
+import os
+
+import pytest
+from dottxt.client import Dottxt as DottxtClient
+from pydantic import BaseModel
+
+import outlines
+from outlines import Generator
+from outlines.models.dottxt import Dottxt
+
+
+MODEL_NAME = "dottxt/dottxt-v1-alpha"
+MODEL_REVISION = "d06c86726aadd8dadb92c5b9b9e3ce8ef246c471"
+
+
+class User(BaseModel):
+    first_name: str
+    last_name: str
+    user_id: int
+
+
+@pytest.fixture(scope="session")
+def api_key():
+    """Get the Dottxt API key from the environment, providing a default value
+    if not found.
+
+    This fixture should be used for tests that do not make actual api calls,
+    but still require to initialize the Dottxt client.
+
+    """
+    api_key = os.getenv("DOTTXT_API_KEY")
+    if not api_key:
+        return "MOCK_API_KEY"
+    return api_key
+
+
+@pytest.fixture(scope="session")
+def model_name_and_revision(api_key):
+    client = DottxtClient(api_key=api_key)
+    model_list = client.list_models()
+    return (model_list[0].name, model_list[0].revision)
+
+
+@pytest.fixture(scope="session")
+def model(api_key, model_name_and_revision):
+    client = DottxtClient(api_key=api_key)
+    return Dottxt(
+        client,
+        model_name_and_revision[0],
+        model_name_and_revision[1],
+    )
+
+
+@pytest.fixture(scope="session")
+def model_no_model_name(api_key):
+    client = DottxtClient(api_key=api_key)
+    return Dottxt(client)
+
+
+@pytest.mark.api_call
+def test_dottxt_init_from_client(api_key, model_name_and_revision):
+    client = DottxtClient(api_key=api_key)
+
+    # Without model name
+    model = outlines.from_dottxt(client)
+    assert isinstance(model, Dottxt)
+    assert model.client == client
+    assert model.model_name is None
+
+    # With model name
+    model = outlines.from_dottxt(
+        client,
+        model_name_and_revision[0],
+        model_name_and_revision[1],
+    )
+    assert isinstance(model, Dottxt)
+    assert model.client == client
+    assert model.model_name == model_name_and_revision[0]
+    assert model.model_revision == model_name_and_revision[1]
+
+
+def test_dottxt_wrong_output_type(model_no_model_name):
+    with pytest.raises(TypeError, match="You must provide an output type"):
+        model_no_model_name("prompt")
+
+
+def test_dottxt_wrong_input_type(model_no_model_name):
+    with pytest.raises(TypeError, match="is not available"):
+        model_no_model_name(["prompt"], User)
+
+
+@pytest.mark.api_call
+def test_dottxt_wrong_inference_parameters(model_no_model_name):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model_no_model_name("prompt", User, foo=10)
+
+
+@pytest.mark.api_call
+def test_dottxt_direct_pydantic_call(model_no_model_name):
+    result = model_no_model_name("Create a user", User)
+    assert "first_name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_dottxt_direct_jsonschema_call(
+    model_no_model_name, model_name_and_revision
+):
+    result = model_no_model_name(
+        "Create a user",
+        User,
+        model_name=model_name_and_revision[0],
+        model_revision=model_name_and_revision[1],
+    )
+    assert "first_name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_dottxt_generator_pydantic_call(model):
+    generator = Generator(model, User)
+    result = generator("Create a user")
+    assert "first_name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_dottxt_streaming(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="Dottxt does not support streaming"
+    ):
+        model.stream("Create a user", User)
diff --git a/tests/models/test_dottxt_type_adapter.py b/tests/models/test_dottxt_type_adapter.py
new file mode 100644
index 000000000..0253af1ea
--- /dev/null
+++ b/tests/models/test_dottxt_type_adapter.py
@@ -0,0 +1,137 @@
+import io
+import json
+import pytest
+import sys
+from dataclasses import dataclass
+
+from PIL import Image
+from genson import SchemaBuilder
+from pydantic import BaseModel
+
+from outlines.models.dottxt import DottxtTypeAdapter
+from outlines.templates import Vision
+from outlines.types import cfg, json_schema, regex
+
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return DottxtTypeAdapter()
+
+
+def test_dottxt_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert result == message
+
+
+def test_dottxt_type_adapter_input_invalid(adapter, image):
+    prompt = Vision("prompt", image)
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_dottxt_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="You must provide an output type"):
+        adapter.format_output_type(None)
+
+    with pytest.raises(TypeError, match="The type `str` is not supported"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not supported"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="Regex-based structured outputs will soon be"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs will soon be"):
+        adapter.format_output_type(cfg(""))
+
+
+def test_dottxt_type_adapter_output_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_output_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_output_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_output_genson_schema_builder(adapter, schema):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    result_dict = json.loads(result)
+    assert isinstance(result_dict, dict)
+    expected_schema = {
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"],
+    }
+    assert result_dict == expected_schema
+
+
+def test_dottxt_type_adapter_json_schema_str(adapter, schema):
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(json_schema(schema_str))
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_json_schema_dict(adapter, schema):
+    result = adapter.format_output_type(json_schema(schema))
+    assert result == json.dumps(schema)
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
new file mode 100644
index 000000000..5df7b2669
--- /dev/null
+++ b/tests/models/test_gemini.py
@@ -0,0 +1,221 @@
+import io
+import json
+import sys
+from dataclasses import dataclass
+from enum import Enum
+from typing import Generator, Literal
+
+import pytest
+from PIL import Image
+from google.genai import Client
+from pydantic import BaseModel, ValidationError
+
+import outlines
+from outlines.models.gemini import Gemini
+from outlines.templates import Vision
+
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
+MODEL_NAME = "gemini-1.5-flash-latest"
+
+
+@pytest.fixture(scope="session")
+def model():
+    return Gemini(Client(), MODEL_NAME)
+
+
+@pytest.fixture(scope="session")
+def model_no_model_name():
+    return Gemini(Client())
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.mark.api_call
+def test_gemini_init_from_client():
+    client = Client()
+
+    # Without model name
+    model = outlines.from_gemini(client)
+    assert isinstance(model, Gemini)
+    assert model.client == client
+    assert model.model_name is None
+
+    # With model name
+    model = outlines.from_gemini(client, MODEL_NAME)
+    assert isinstance(model, Gemini)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+
+@pytest.mark.api_call
+def test_gemini_wrong_inference_parameters(model):
+    with pytest.raises(ValidationError):
+        model.generate("prompt", foo=10)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_call(model):
+    result = model.generate("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_gemini_direct_call(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        model=MODEL_NAME
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_vision(model, image):
+    result = model.generate(Vision("What does this logo represent?", image))
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_pydantic(model):
+    class Foo(BaseModel):
+        bar: int
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_vision_pydantic(model, image):
+    class Logo(BaseModel):
+        name: int
+
+    result = model.generate(Vision("What does this logo represent?", image), Logo)
+    assert isinstance(result, str)
+    assert "name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_gemini_nested_pydantic(model):
+    class Bar(BaseModel):
+        fu: str
+
+    class Foo(BaseModel):
+        sna: int
+        bar: Bar
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "sna" in json.loads(result)
+    assert "bar" in json.loads(result)
+    assert "fu" in json.loads(result)["bar"]
+
+
+@pytest.mark.xfail(
+    reason="The Gemini SDK's serialization method does not support Json Schema strings."
+)
+@pytest.mark.api_call
+def test_gemini_simple_json_schema_string(model):
+    schema = "{'properties': {'bar': {'title': 'Bar', 'type': 'integer'}}, 'required': ['bar'], 'title': 'Foo', 'type': 'object'}"
+    result = model.generate("foo?", schema)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.xfail(
+    reason="The Gemini SDK's serialization method does not support Json Schema dictionaries."
+)
+@pytest.mark.api_call
+def test_gemini_simple_json_schema_dict(model):
+    schema = {
+        "properties": {"bar": {"type": "integer"}},
+        "required": ["bar"],
+        "type": "object",
+    }
+    result = model.generate("foo?", schema)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_typed_dict(model):
+    class Foo(TypedDict):
+        bar: int
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_dataclass(model):
+    @dataclass
+    class Foo:
+        bar: int
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_choice_enum(model):
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert result == "Foo" or result == "Bar"
+
+
+@pytest.mark.api_call
+def test_gemini_sample_choice_literal(model):
+    result = model.generate("foo?", Literal["Foo", "Bar"])
+    assert isinstance(result, str)
+    assert result == "Foo" or result == "Bar"
+
+
+@pytest.mark.xfail(
+    reason="Gemini supports lists for choices but we do not as it is semantically incorrect."
+)
+@pytest.mark.api_call
+def test_gemini_simple_choice_list(model):
+    choices = ["Foo", "Bar"]
+    result = model.generate("foo?", choices)
+    assert isinstance(result, str)
+    assert result == "Foo" or result == "Bar"
+
+
+@pytest.mark.api_call
+def test_gemini_simple_list_pydantic(model):
+    class Foo(BaseModel):
+        bar: int
+
+    result = model.generate("foo?", list[Foo])
+    assert isinstance(json.loads(result), list)
+    assert isinstance(json.loads(result)[0], dict)
+    assert "bar" in json.loads(result)[0]
+
+
+@pytest.mark.api_call
+def test_gemini_streaming(model):
+    result = model.stream("Respond with one word. Not more.")
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
new file mode 100644
index 000000000..46834eae9
--- /dev/null
+++ b/tests/models/test_gemini_type_adapter.py
@@ -0,0 +1,204 @@
+import io
+import pytest
+import sys
+from dataclasses import dataclass
+from enum import Enum, EnumMeta
+from typing import Literal
+
+from PIL import Image
+from genson import SchemaBuilder
+from google.genai import types
+from pydantic import BaseModel
+
+from outlines import cfg, json_schema, regex
+from outlines.models.gemini import GeminiTypeAdapter
+from outlines.templates import Vision
+
+if sys.version_info >= (3, 12):
+    from typing import TypedDict, is_typeddict
+else:
+    from typing_extensions import TypedDict, is_typeddict
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return GeminiTypeAdapter()
+
+
+def test_gemini_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert result == {"contents": [message]}
+
+
+def test_gemini_type_adapter_input_vision(adapter, image):
+    input_message = Vision("hello", image)
+    result = adapter.format_input(input_message)
+    image_part = types.Part.from_bytes(
+        data=input_message.image_str,
+        mime_type=input_message.image_format
+    ),
+    assert result == {"contents": [input_message.prompt, image_part]}
+
+
+def test_dottxt_type_adapter_input_invalid(adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_gemini_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="The type `str` is not supported"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not supported"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="Neither regex-based"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs"):
+        adapter.format_output_type(cfg(""))
+
+    with pytest.raises(TypeError, match="The Gemini SDK does not accept"):
+        adapter.format_output_type(SchemaBuilder())
+
+    with pytest.raises(TypeError, match="The Gemini SDK does not"):
+        adapter.format_output_type(json_schema(""))
+
+
+def test_gemini_type_adapter_output_none(adapter):
+    result = adapter.format_output_type(None)
+    assert result == {}
+
+
+def test_gemini_type_adapter_output_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": User,
+    }
+
+
+def test_gemini_type_adapter_output_list_dataclass(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(list[User])
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": list[User],
+    }
+
+
+def test_gemini_type_adapter_output_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": User,
+    }
+
+
+def test_gemini_type_adapter_output_list_typed_dict(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(list[User])
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": list[User],
+    }
+
+
+def test_gemini_type_adapter_output_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": User,
+    }
+
+
+def test_gemini_type_adapter_output_list_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(list[User])
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": list[User],
+    }
+
+
+def test_gemini_type_adapter_output_enum(adapter):
+    class Foo(Enum):
+        Bar = "bar"
+        Fuzz = "fuzz"
+
+    result = adapter.format_output_type(Foo)
+    assert result == {
+        "response_mime_type": "text/x.enum",
+        "response_schema": Foo,
+    }
+
+
+def test_gemini_type_adapter_output_literal(adapter):
+    Foo = Literal["bar", "fuzz"]
+    result = adapter.format_output_type(Foo)
+
+    assert isinstance(result, dict)
+    assert len(result) == 2
+    assert result["response_mime_type"] == "text/x.enum"
+    assert isinstance(result["response_schema"], EnumMeta)
+    assert len(result["response_schema"].__members__) == 2
+    assert result["response_schema"].bar.value == "bar"
+    assert result["response_schema"].fuzz.value == "fuzz"
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
new file mode 100644
index 000000000..6a77528f6
--- /dev/null
+++ b/tests/models/test_llamacpp.py
@@ -0,0 +1,134 @@
+import json
+from enum import Enum
+
+import pytest
+from llama_cpp import Llama
+from pydantic import BaseModel
+
+from outlines.models.llamacpp import (
+    LlamaCpp,
+    LlamaCppTokenizer,
+    LlamaCppTypeAdapter,
+    from_llamacpp
+)
+from outlines.processors import RegexLogitsProcessor
+from outlines.types.dsl import Regex, CFG
+
+
+def test_load_model():
+    model = from_llamacpp(
+        Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+    )
+
+    assert isinstance(model, LlamaCpp)
+    assert isinstance(model.model, Llama)
+    assert isinstance(model.tokenizer, LlamaCppTokenizer)
+    assert isinstance(model.type_adapter, LlamaCppTypeAdapter)
+    assert model.tensor_library_name == "numpy"
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    return LlamaCpp(
+        Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+    )
+
+
+def test_llamacpp_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_llamacpp_regex(model):
+    regex_str = Regex(r"[0-9]").pattern
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer, model.tensor_library_name)
+    result = model.generate("Respond with one word. Not more.", logits_processor)
+    assert isinstance(result, str)
+
+
+def test_llamacpp_json(model):
+    class Foo(BaseModel):
+        bar: str
+
+    result = model("foo? Respond with one word.", Foo, max_tokens=1000)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+def test_llamacpp_choice(model):
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    result = model("foo?", Foo)
+    assert result == "Foo" or result == "Bar"
+
+
+def test_llamacpp_cfg(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="CFG generation is not supported for LlamaCpp"
+    ):
+        model("Respond with one word. Not more.", CFG('start: "a"'))
+
+
+def test_llamacpp_text_stop(model):
+    result = model.generate("Write the letter a.", None, stop="a", max_tokens=100)
+    assert "a" not in result
+
+
+def test_llamacpp_stream_simple(model):
+    generator = model.stream("Respond with one word. Not more.", None)
+
+    for x in generator:
+        assert isinstance(x, str)
+
+
+def test_llamacpp_stream_regex(model):
+    generator = model.stream("Respond with one word. Not more.", Regex(r"[0-9]"))
+
+    x = next(generator)
+    assert isinstance(x, str)
+
+
+def test_llamacpp_stream_json(model):
+    class Foo(BaseModel):
+        bar: int
+
+    generator = model.stream("foo?", Foo)
+
+    x = next(generator)
+    assert x == "{"
+
+
+def test_llamacpp_stream_cfg(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="CFG generation is not supported for LlamaCpp"
+    ):
+        model.stream("Respond with one word. Not more.", CFG('start: "a"'))
+
+
+def test_llamacpp_stream_choice(model):
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    generator = model.stream("foo?", Foo)
+
+    x = next(generator)
+    assert x[0] in ("B", "F")
+
+
+def test_llamacpp_stream_text_stop(model):
+    generator = model.stream("Write the letter a.", None, stop="a", max_tokens=100)
+
+    result = next(generator)
+    assert isinstance(result, str)
+    assert result != "a"
diff --git a/tests/models/test_llamacpp_tokenizer.py b/tests/models/test_llamacpp_tokenizer.py
new file mode 100644
index 000000000..dbf2507f3
--- /dev/null
+++ b/tests/models/test_llamacpp_tokenizer.py
@@ -0,0 +1,133 @@
+import pytest
+import sys
+from unittest.mock import patch
+
+import llama_cpp
+import transformers
+
+from outlines.models.llamacpp import LlamaCppTokenizer
+
+
+@pytest.fixture
+def model():
+    model = llama_cpp.Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+    setattr(
+        model.tokenizer_,
+        "hf_tokenizer",
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return model
+
+
+@pytest.fixture
+def model_no_hf_tokenizer():
+    model = llama_cpp.Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+    del model.tokenizer_
+    return model
+
+
+@pytest.fixture
+def different_model():
+    model = llama_cpp.Llama.from_pretrained(
+        "TheBloke/phi-2-GGUF",
+        "phi-2.Q4_K_M.gguf",
+    )
+    return model
+
+
+@pytest.fixture
+def tokenizer(model):
+    return LlamaCppTokenizer(model)
+
+
+@pytest.fixture
+def another_tokenizer(model):
+    return LlamaCppTokenizer(model)
+
+
+@pytest.fixture
+def tokenizer_no_hf_tokenizer(model_no_hf_tokenizer):
+    return LlamaCppTokenizer(model_no_hf_tokenizer)
+
+
+@pytest.fixture
+def different_tokenizer(different_model):
+    return LlamaCppTokenizer(different_model)
+
+
+def test_llama_cpp_tokenizer_init(tokenizer, tokenizer_no_hf_tokenizer):
+    # regular case
+    assert tokenizer.eos_token_id is not None
+    assert tokenizer.pad_token_id is not None
+    assert isinstance(tokenizer.vocabulary, dict)
+
+    # tokenizer with no hf_tokenizer
+    assert tokenizer_no_hf_tokenizer.eos_token_id is not None
+    assert tokenizer_no_hf_tokenizer.pad_token_id is not None
+    assert isinstance(tokenizer_no_hf_tokenizer.vocabulary, dict)
+
+
+def test_llama_cpp_tokenizer_encode(tokenizer):
+    # batch case
+    with pytest.raises(NotImplementedError):
+        token_ids, attention_mask = tokenizer.encode(["foo", "bar"])
+
+    # regular case
+    token_ids, attention_mask = tokenizer.encode("Hello, world!")
+    assert token_ids is not None
+    assert attention_mask is not None
+    assert len(token_ids) == len(attention_mask)
+
+
+def test_llama_cpp_tokenizer_decode(tokenizer):
+    token_ids, _ = tokenizer.encode("Hello, world!")
+    decoded_text = tokenizer.decode(token_ids)
+    assert isinstance(decoded_text, list)
+    assert "".join(decoded_text).strip() == "Hello, world!"
+
+
+def test_llama_cpp_tokenizer_convert_token_to_string(
+    tokenizer,
+    tokenizer_no_hf_tokenizer
+):
+    # with self._hf_tokenizer
+    token_str = tokenizer.convert_token_to_string("<0x20>")
+    assert isinstance(token_str, str)
+
+    # without self._hf_tokenizer
+    token_str = tokenizer_no_hf_tokenizer.convert_token_to_string("<0x20>")
+    assert isinstance(token_str, str)
+
+
+def test_llama_cpp_tokenizer_eq(tokenizer, another_tokenizer, different_tokenizer):
+    assert not tokenizer == 1
+    assert tokenizer == another_tokenizer
+    assert tokenizer != different_tokenizer
+
+
+def test_llama_cpp_tokenizer_hash(tokenizer, another_tokenizer, different_tokenizer):
+    assert isinstance(hash(tokenizer), int)
+    assert hash(tokenizer) == hash(another_tokenizer)
+    assert hash(tokenizer) != hash(different_tokenizer)
+
+
+def test_llama_cpp_tokenizer_getstate(tokenizer):
+    state = tokenizer.__getstate__()
+    assert isinstance(state, tuple)
+    assert len(state) == 5
+    assert isinstance(state[0], dict)
+    assert isinstance(state[1], int)
+    assert isinstance(state[2], str)
+    assert isinstance(state[3], int)
+    assert isinstance(state[4], list)
+
+
+def test_llama_cpp_tokenizer_setstate(tokenizer):
+    with pytest.raises(NotImplementedError):
+        tokenizer.__setstate__(None)
diff --git a/tests/models/test_llamacpp_type_adapter.py b/tests/models/test_llamacpp_type_adapter.py
new file mode 100644
index 000000000..e49586d5d
--- /dev/null
+++ b/tests/models/test_llamacpp_type_adapter.py
@@ -0,0 +1,41 @@
+import pytest
+
+import llama_cpp
+from llama_cpp import LogitsProcessorList
+
+from outlines.models.llamacpp import LlamaCppTypeAdapter, LlamaCppTokenizer
+from outlines.processors.structured import RegexLogitsProcessor
+
+
+@pytest.fixture
+def adapter():
+    return LlamaCppTypeAdapter()
+
+
+@pytest.fixture
+def logits_processor():
+    model = llama_cpp.Llama.from_pretrained(
+        "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+    return RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=LlamaCppTokenizer(model),
+        tensor_library_name="numpy",
+    )
+
+
+def test_llamacpp_type_adapter_format_input(adapter):
+    # anything else than a string
+    with pytest.raises(NotImplementedError):
+        adapter.format_input(["Hello, world!"])
+
+    # string case
+    assert adapter.format_input("Hello, world!") == "Hello, world!"
+
+
+def test_llamacpp_type_adapter_format_output_type(adapter, logits_processor):
+    formatted = adapter.format_output_type(logits_processor)
+    assert isinstance(formatted, LogitsProcessorList)
+    assert len(formatted) == 1
+    assert isinstance(formatted[0], RegexLogitsProcessor)
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index 20e59da81..8ec889230 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -1,9 +1,20 @@
 import pytest
-
-from outlines.models.mlxlm import mlxlm
+import re
+from enum import Enum
+from typing import Generator
+
+import outlines
+from outlines.types import Regex
+from outlines.models.mlxlm import (
+    MLXLM,
+    MLXLMTypeAdapter,
+    from_mlxlm
+)
 from outlines.models.transformers import TransformerTokenizer
+from pydantic import BaseModel
 
 try:
+    import mlx_lm
     import mlx.core as mx
 
     HAS_MLX = mx.metal.is_available()
@@ -15,86 +26,94 @@
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_model():
-    model = mlxlm(TEST_MODEL)
-    assert hasattr(model, "model")
-    assert hasattr(model, "tokenizer")
+def test_mlxlm_model_initialization():
+    model = from_mlxlm(*mlx_lm.load(TEST_MODEL))
+    assert isinstance(model, MLXLM)
+    assert isinstance(model.model, mlx_lm.models.llama.Model)
+    assert isinstance(
+        model.mlx_tokenizer, mlx_lm.tokenizer_utils.TokenizerWrapper
+    )
     assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, MLXLMTypeAdapter)
+    assert model.tensor_library_name == "mlx"
 
 
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_tokenizer():
-    model = mlxlm(TEST_MODEL)
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    model, tokenizer = mlx_lm.load(TEST_MODEL)
+    return outlines.from_mlxlm(model, tokenizer)
+
 
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_tokenizer(model):
     # Test single string encoding/decoding
     test_text = "Hello, world!"
-    token_ids = mx.array(model.mlx_tokenizer.encode(test_text))
+    token_ids, _ = model.tokenizer.encode(test_text)
+    token_ids = mx.array(token_ids)
     assert isinstance(token_ids, mx.array)
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_generate():
-    from outlines.generate.api import GenerationParameters, SamplingParameters
+def test_mlxlm_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
 
-    model = mlxlm(TEST_MODEL)
-    prompt = "Write a haiku about programming:"
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
 
-    # Test with basic generation parameters
-    gen_params = GenerationParameters(max_tokens=50, stop_at=None, seed=None)
 
-    # Test with different sampling parameters
-    sampling_params = SamplingParameters(
-        sampler="multinomial", num_samples=1, top_p=0.9, top_k=None, temperature=0.7
-    )
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_invalid_input_type(model):
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model(["Respond with one word. Not more."])
+
 
-    # Test generation
-    output = model.generate(prompt, gen_params, None, sampling_params)
-    assert isinstance(output, str)
-    assert len(output) > 0
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_invalid_inference_kwargs(model):
+    with pytest.raises(TypeError):
+        model("Respond with one word. Not more.", foo="bar")
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_stream():
-    from outlines.generate.api import GenerationParameters, SamplingParameters
+def test_mlxlm_inference_kwargs(model):
+    result = model("Write a short story about a cat.", max_tokens=2)
+    assert isinstance(result, str)
+    assert len(result) < 20
 
-    model = mlxlm(TEST_MODEL)
-    prompt = "Count from 1 to 5:"
 
-    gen_params = GenerationParameters(max_tokens=20, stop_at=None, seed=None)
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_regex(model):
+    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
 
-    sampling_params = SamplingParameters(
-        sampler="greedy",  # Use greedy sampling for deterministic output
-        num_samples=1,
-        top_p=None,
-        top_k=None,
-        temperature=0.0,
-    )
 
-    # Test streaming
-    stream = model.stream(prompt, gen_params, None, sampling_params)
-    tokens = list(stream)
-    assert len(tokens) > 0
-    assert all(isinstance(token, str) for token in tokens)
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_JsonType(model):
+    class Character(BaseModel):
+        name: str
 
-    # Test that concatenated streaming output matches generate output
-    streamed_text = "".join(tokens)
-    generated_text = model.generate(prompt, gen_params, None, sampling_params)
-    assert streamed_text == generated_text
+    result = model("Create a character with a name.", Character)
+    assert "name" in result
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_errors():
-    model = mlxlm(TEST_MODEL)
+def test_mlxlm_choice(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
 
-    # Test batch inference (should raise NotImplementedError)
-    with pytest.raises(NotImplementedError):
-        from outlines.generate.api import GenerationParameters, SamplingParameters
+    result = model("Cat or dog?", Foo)
+    assert result in ["cat", "dog"]
 
-        gen_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
-        sampling_params = SamplingParameters("multinomial", 1, None, None, 1.0)
-        model.generate(["prompt1", "prompt2"], gen_params, None, sampling_params)
 
-    # Test beam search (should raise NotImplementedError)
-    with pytest.raises(NotImplementedError):
-        sampling_params = SamplingParameters("beam_search", 1, None, None, 1.0)
-        model.generate("test prompt", gen_params, None, sampling_params)
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_stream_text_stop(model):
+    generator = model.stream(
+        "Respond with one word. Not more.", None, max_tokens=100
+    )
+    assert isinstance(generator, Generator)
+    assert isinstance(next(generator), str)
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
new file mode 100644
index 000000000..408fa2ceb
--- /dev/null
+++ b/tests/models/test_ollama.py
@@ -0,0 +1,101 @@
+import json
+from enum import Enum
+from typing import Annotated
+
+import pytest
+from ollama import Client as OllamaClient
+from pydantic import BaseModel, Field
+
+import outlines
+from outlines.models import Ollama
+
+
+MODEL_NAME = "tinyllama"
+
+
+@pytest.fixture
+def model():
+    return Ollama(OllamaClient(), MODEL_NAME)
+
+
+@pytest.fixture
+def model_no_model_name():
+    return Ollama(OllamaClient())
+
+
+def test_init_from_client():
+    client = OllamaClient()
+
+    # With model name
+    model = outlines.from_ollama(client, MODEL_NAME)
+    assert isinstance(model, Ollama)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+    # Without model name
+    model = outlines.from_ollama(client)
+    assert isinstance(model, Ollama)
+    assert model.client == client
+    assert model.model_name is None
+
+
+def test_wrong_inference_parameters(model):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model.generate(
+            "Respond with one word. Not more.", None, foo=10
+        )
+
+
+def test_ollama_simple(model):
+    result = model.generate(
+        "Respond with one word. Not more.", None
+    )
+    assert isinstance(result, str)
+
+
+def test_ollama_direct(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        None,
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
+def test_ollama_json(model):
+    class Foo(BaseModel):
+        foo: Annotated[str, Field(max_length=1)]
+
+    result = model("Respond with one word. Not more.", Foo)
+    assert isinstance(result, str)
+    assert "foo" in json.loads(result)
+
+
+def test_ollama_wrong_output_type(model):
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    with pytest.raises(TypeError, match="is not supported"):
+        model.generate("foo?", Foo)
+
+
+def test_ollama_wrong_input_type(model):
+    with pytest.raises(TypeError, match="is not available"):
+        model.generate(["foo?", "bar?"], None)
+
+
+def test_ollama_stream(model):
+    generator = model.stream("Write a sentence about a cat.")
+    assert isinstance(next(generator), str)
+
+
+def test_ollama_stream_json(model_no_model_name):
+    class Foo(BaseModel):
+        foo: Annotated[str, Field(max_length=2)]
+
+    generator = model_no_model_name.stream("Create a character.", Foo, model=MODEL_NAME)
+    generated_text = []
+    for text in generator:
+        generated_text.append(text)
+    assert "foo" in json.loads("".join(generated_text))
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
new file mode 100644
index 000000000..58b6c6aa9
--- /dev/null
+++ b/tests/models/test_ollama_type_adapter.py
@@ -0,0 +1,131 @@
+import io
+import json
+import pytest
+import sys
+from dataclasses import dataclass
+
+from genson import SchemaBuilder
+from PIL import Image
+from pydantic import BaseModel
+
+from outlines.models.ollama import OllamaTypeAdapter
+from outlines.templates import Vision
+from outlines.types import cfg, json_schema, regex
+
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return OllamaTypeAdapter()
+
+
+def test_ollama_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert result == message
+
+
+def test_ollama_type_adapter_input_invalid(adapter, image):
+    prompt = Vision("prompt", image)
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_ollama_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="The type `str` is not supported"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not supported"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="Regex-based structured outputs are not"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs are not"):
+        adapter.format_output_type(cfg(""))
+
+
+def test_ollama_type_adapter_output_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == schema
+
+
+def test_ollama_type_adapter_output_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == schema
+
+
+def test_ollama_type_adapter_output_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == schema
+
+
+def test_ollama_type_adapter_output_genson_schema_builder(adapter, schema):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    assert result == json.dumps({
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"]
+    })
+
+
+def test_ollama_type_adapter_json_schema_str(adapter, schema):
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(json_schema(schema_str))
+    assert result == schema
+
+
+def test_ollama_type_adapter_json_schema_dict(adapter, schema):
+    result = adapter.format_output_type(json_schema(schema))
+    assert result == schema
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index f4b97f36a..455087707 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,125 +1,173 @@
-import importlib
+import io
 import json
-from contextlib import contextmanager
-from unittest import mock
-from unittest.mock import MagicMock
+import os
+from typing import Annotated, Generator
 
 import pytest
-from openai import AsyncOpenAI
-
-from outlines import generate
-from outlines.models.openai import OpenAI, OpenAIConfig
-
-
-def module_patch(path):
-    """Patch functions that have the same name as the module in which they're implemented."""
-    target = path
-    components = target.split(".")
-    for i in range(len(components), 0, -1):
-        try:
-            # attempt to import the module
-            imported = importlib.import_module(".".join(components[:i]))
-
-            # module was imported, let's use it in the patch
-            patch = mock.patch(path)
-            patch.getter = lambda: imported
-            patch.attribute = ".".join(components[i:])
-            return patch
-        except Exception:
-            continue
-
-    # did not find a module, just return the default mock
-    return mock.patch(path)
-
-
-def test_openai_call():
-    with module_patch("outlines.models.openai.generate_chat") as mocked_generate_chat:
-        mocked_generate_chat.return_value = ["foo"], 1, 2
-        async_client = MagicMock(spec=AsyncOpenAI, api_key="key")
-
-        model = OpenAI(
-            async_client,
-            OpenAIConfig(max_tokens=10, temperature=0.5, n=2, stop=["."]),
-        )
-
-        assert model("bar")[0] == "foo"
-        assert model.prompt_tokens == 1
-        assert model.completion_tokens == 2
-        mocked_generate_chat_args = mocked_generate_chat.call_args
-        mocked_generate_chat_arg_config = mocked_generate_chat_args[0][3]
-        assert isinstance(mocked_generate_chat_arg_config, OpenAIConfig)
-        assert mocked_generate_chat_arg_config.max_tokens == 10
-        assert mocked_generate_chat_arg_config.temperature == 0.5
-        assert mocked_generate_chat_arg_config.n == 2
-        assert mocked_generate_chat_arg_config.stop == ["."]
-
-        model("bar", samples=3)
-        mocked_generate_chat_args = mocked_generate_chat.call_args
-        mocked_generate_chat_arg_config = mocked_generate_chat_args[0][3]
-        assert mocked_generate_chat_arg_config.n == 3
-
-
-@contextmanager
-def patched_openai(completion, **oai_config):
-    """Create a patched openai whose chat completions always returns `completion`"""
-    with module_patch("outlines.models.openai.generate_chat") as mocked_generate_chat:
-        mocked_generate_chat.return_value = completion, 1, 2
-        async_client = MagicMock(spec=AsyncOpenAI, api_key="key")
-        model = OpenAI(
-            async_client,
-            OpenAIConfig(max_tokens=10, temperature=0.5, n=2, stop=["."]),
-        )
-        yield model
-
-
-def test_openai_choice_call():
-    with patched_openai(completion='{"result": "foo"}') as model:
-        generator = generate.choice(model, ["foo", "bar"])
-        assert generator("hi") == "foo"
-
-
-def test_openai_choice_call_invalid_server_response():
-    with patched_openai(completion="not actual json") as model:
-        generator = generate.choice(model, ["foo", "bar"])
-        with pytest.raises(json.decoder.JSONDecodeError):
-            generator("hi")
-
-
-def test_openai_json_call_pydantic():
-    from pydantic import BaseModel, ConfigDict, ValidationError
-
-    class Person(BaseModel):
-        model_config = ConfigDict(extra="forbid")  # required for openai
-        first_name: str
-        last_name: str
-        age: int
-
-    completion = '{"first_name": "Usain", "last_name": "Bolt", "age": 38}'
-
-    # assert success for valid response
-    with patched_openai(completion=completion) as model:
-        generator = generate.json(model, Person)
-        assert generator("fastest person") == Person.parse_raw(completion)
-
-    # assert fail for non-json response
-    with patched_openai(completion="usain bolt") as model:
-        generator = generate.json(model, Person)
-        with pytest.raises(ValidationError):
-            assert generator("fastest person")
-
-
-def test_openai_json_call_str():
-    person_schema = '{"additionalProperties": false, "properties": {"first_name": {"title": "First Name", "type": "string"}, "last_name": {"title": "Last Name", "type": "string"}, "age": {"title": "Age", "type": "integer"}}, "required": ["first_name", "last_name", "age"], "title": "Person", "type": "object"}'
-
-    output = {"first_name": "Usain", "last_name": "Bolt", "age": 38}
-
-    # assert success for valid response
-    with patched_openai(completion=json.dumps(output)) as model:
-        generator = generate.json(model, person_schema)
-        assert generator("fastest person") == output
-
-    # assert fail for non-json response
-    with patched_openai(completion="usain bolt") as model:
-        generator = generate.json(model, person_schema)
-        with pytest.raises(json.decoder.JSONDecodeError):
-            assert generator("fastest person")
+import requests
+from PIL import Image
+from openai import OpenAI as OpenAIClient
+from pydantic import BaseModel, Field
+
+import outlines
+from outlines.models.openai import OpenAI
+from outlines.templates import Vision
+from outlines.types import json_schema
+
+MODEL_NAME = "gpt-4o-mini-2024-07-18"
+
+
+@pytest.fixture(scope="session")
+def api_key():
+    """Get the OpenAI API key from the environment, providing a default value if not found.
+
+    This fixture should be used for tests that do not make actual api calls,
+    but still require to initialize the OpenAI client.
+
+    """
+    api_key = os.getenv("OPENAI_API_KEY")
+    if not api_key:
+        return "MOCK_VALUE"
+    return api_key
+
+
+@pytest.fixture(scope="session")
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture(scope="session")
+def model(api_key):
+    return OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
+
+
+@pytest.fixture(scope="session")
+def model_no_model_name(api_key):
+    return OpenAI(OpenAIClient(api_key=api_key))
+
+
+def test_init_from_client(api_key):
+    client = OpenAIClient(api_key=api_key)
+
+    # With model name
+    model = outlines.from_openai(client, "gpt-4o")
+    assert isinstance(model, OpenAI)
+    assert model.client == client
+    assert model.model_name == "gpt-4o"
+
+    # Without model name
+    model = outlines.from_openai(client)
+    assert isinstance(model, OpenAI)
+    assert model.client == client
+    assert model.model_name is None
+
+
+def test_openai_wrong_inference_parameters(model):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model.generate("prompt", foo=10)
+
+
+def test_openai_wrong_input_type(model):
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(TypeError, match="is not available"):
+        model.generate(Foo("prompt"))
+
+
+def test_openai_wrong_output_type(model):
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(TypeError, match="is not available"):
+        model.generate("prompt", Foo(1))
+
+
+@pytest.mark.api_call
+def test_openai_simple_call(model):
+    result = model.generate("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_openai_simple_call_multiple_samples(model):
+    result = model.generate("Respond with one word. Not more.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+@pytest.mark.api_call
+def test_openai_direct_call(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_openai_simple_vision(image, model):
+    result = model.generate(Vision("What does this logo represent?", image))
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_openai_simple_pydantic(model):
+    class Foo(BaseModel):
+        bar: int
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_openai_simple_pydantic_refusal(model):
+    class Foo(BaseModel):
+        bar: Annotated[str, Field(int, pattern=r"^\d+$")]
+
+    with pytest.raises(TypeError, match="OpenAI does not support your schema"):
+        _ = model.generate("foo?", Foo)
+
+
+@pytest.mark.api_call
+def test_openai_simple_vision_pydantic(image, model):
+    class Logo(BaseModel):
+        name: int
+
+    result = model.generate(Vision("What does this logo represent?", image), Logo)
+    assert isinstance(result, str)
+    assert "name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_openai_simple_json_schema(model):
+    class Foo(BaseModel):
+        bar: int
+
+    schema = json.dumps(Foo.model_json_schema())
+
+    result = model.generate("foo?", json_schema(schema))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_openai_streaming(model):
+    result = model.stream("Respond with one word. Not more.")
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
new file mode 100644
index 000000000..fcafbb8de
--- /dev/null
+++ b/tests/models/test_openai_type_adapter.py
@@ -0,0 +1,195 @@
+import io
+import json
+import pytest
+import sys
+from dataclasses import dataclass
+from typing import Literal
+
+from genson import SchemaBuilder
+from PIL import Image
+from pydantic import BaseModel
+
+from outlines import cfg, json_schema, regex, Vision
+from outlines.models.openai import OpenAITypeAdapter
+
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+        "additionalProperties": False,
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return OpenAITypeAdapter()
+
+
+def test_openai_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert isinstance(result, dict)
+    assert len(result["messages"]) == 1
+    assert result["messages"][0]["role"] == "user"
+    assert result["messages"][0]["content"] == message
+
+
+def test_openai_type_adapter_input_vision(adapter, image):
+    input_message = Vision("hello", image)
+    result = adapter.format_input(input_message)
+    assert isinstance(result, dict)
+
+    messages = result["messages"]
+    assert len(messages) == 1
+
+    message = messages[0]
+    assert message["role"] == "user"
+    assert len(message["content"]) == 2
+    assert message["content"][0]["type"] == "text"
+    assert message["content"][0]["text"] == "hello"
+
+    assert message["content"][1]["type"] == "image_url"
+    assert (
+        message["content"][1]["image_url"]["url"]
+        == f"data:image/png;base64,{input_message.image_str}"
+    )
+
+
+def test_dottxt_type_adapter_input_invalid(adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_openai_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="The type `str` is not available"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not available"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="The type `Literal` is not available"):
+        adapter.format_output_type(Literal[1, 2])
+
+    with pytest.raises(TypeError, match="Neither regex-based"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs"):
+        adapter.format_output_type(cfg(""))
+
+    class Foo(BaseModel):
+        bar: str
+
+    with pytest.raises(TypeError, match="The type `list` is not available"):
+        adapter.format_output_type(list[Foo])
+
+
+def test_openai_type_adapter_output_none(adapter):
+    result = adapter.format_output_type(None)
+    assert result == {}
+
+
+def test_openai_type_adapter_json_mode(adapter):
+    result = adapter.format_output_type(dict)
+    assert result == {"response_format": {"type": "json_object"}}
+
+
+def test_openai_type_adapter_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_genson_schema_builder(adapter, schema):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    expected_schema = {
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"],
+        "additionalProperties": False  # OpenAI adds this
+    }
+    assert result["response_format"]["json_schema"]["schema"] == expected_schema
+
+
+def test_openai_type_adapter_json_schema_str(adapter, schema):
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(json_schema(schema_str))
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_json_schema_dict(adapter, schema):
+    result = adapter.format_output_type(json_schema(schema))
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
new file mode 100644
index 000000000..60960f463
--- /dev/null
+++ b/tests/models/test_sglang.py
@@ -0,0 +1,272 @@
+# ATTENTION: When running this test with an actual SGLang server, use the
+# llguidance backend (--grammar-backend llguidance)
+# The outlines backend does not support the EBNF grammar. The xgrammar
+# backend is slow and buggy.
+
+import os
+import re
+import warnings
+from typing import AsyncGenerator, Generator
+
+import pytest
+from openai import AsyncOpenAI, OpenAI
+
+from outlines.models.sglang import SGLang, AsyncSGLang, from_sglang
+from outlines.types.dsl import CFG, Regex, JsonSchema
+from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
+
+
+EBNF_YES_NO_GRAMMAR = """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+
+
+# If the SGLANG_SERVER_URL environment variable is set, use the real SGLang server
+# Otherwise, use the mock server
+sglang_server_url = os.environ.get("SGLANG_SERVER_URL")
+sglang_model_name = os.environ.get(
+    "SGLANG_MODEL_NAME", "qwen/qwen2.5-0.5b-instruct"
+)
+if sglang_server_url:
+    openai_client = OpenAI(base_url=sglang_server_url)
+    async_openai_client = AsyncOpenAI(base_url=sglang_server_url)
+else:
+    warnings.warn("No SGLang server URL provided, using mock server")
+    openai_client = MockOpenAIClient()
+    async_openai_client = MockAsyncOpenAIClient()
+
+mock_responses = [
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': sglang_model_name,
+        },
+        "foo"
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': sglang_model_name,
+            'stream': True
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'n': 2,
+            'model': sglang_model_name,
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+            'response_format': {
+                'type': 'json_schema',
+                'json_schema': {
+                    'name': 'default',
+                    'strict': True,
+                    'schema': {
+                        'type': 'object',
+                        'properties': {'bar': {'type': 'string'}},
+                        'additionalProperties': False
+                    }
+                }
+            }
+        },
+        '{"foo": "bar"}'
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'regex': '([0-9]{3})',
+            },
+        },
+        "123"
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'ebnf': EBNF_YES_NO_GRAMMAR,
+            },
+        },
+        "yes"
+    ),
+]
+
+
+# If the SGLANG_SERVER_URL environment variable is not set, add the mock
+# responses to the mock clients
+if not sglang_server_url:
+    async_openai_client.add_mock_responses(mock_responses)
+    openai_client.add_mock_responses(mock_responses)
+
+
+@pytest.fixture
+def sync_model():
+    return SGLang(openai_client, model_name=sglang_model_name)
+
+
+@pytest.fixture
+def sync_model_no_model_name():
+    return SGLang(openai_client)
+
+
+@pytest.fixture
+def async_model():
+    return AsyncSGLang(async_openai_client, model_name=sglang_model_name)
+
+
+@pytest.fixture
+def async_model_no_model_name():
+    return AsyncSGLang(async_openai_client)
+
+
+def test_sglang_init():
+    # We do not rely on the mock server here because we need an object
+    # of type OpenAI and AsyncOpenAI to test the init function.
+    openai_client = OpenAI(base_url="http://localhost:11434")
+    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
+
+    # Sync with model name
+    model = from_sglang(openai_client, sglang_model_name)
+    assert isinstance(model, SGLang)
+    assert model.client == openai_client
+    assert model.model_name == sglang_model_name
+
+    # Sync without model name
+    model = from_sglang(openai_client)
+    assert isinstance(model, SGLang)
+    assert model.client == openai_client
+    assert model.model_name is None
+
+    # Async with model name
+    model = from_sglang(async_openai_client, sglang_model_name)
+    assert isinstance(model, AsyncSGLang)
+    assert model.client == async_openai_client
+    assert model.model_name == sglang_model_name
+
+    # Async without model name
+    model = from_sglang(async_openai_client)
+    assert isinstance(model, AsyncSGLang)
+    assert model.client == async_openai_client
+    assert model.model_name is None
+
+    with pytest.raises(ValueError, match="Unsupported client type"):
+        from_sglang("foo")
+
+
+def test_sglang_sync_simple_call(sync_model):
+    result = sync_model("Respond with a single word.",)
+    assert isinstance(result, str)
+
+
+def test_sglang_sync_streaming(sync_model_no_model_name):
+    result = sync_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=sglang_model_name,
+    )
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
+
+
+def test_sglang_sync_multiple_samples(sync_model):
+    result = sync_model("Respond with a single word.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+def test_sglang_sync_json(sync_model):
+    json_string = (
+        '{"type": "object", "properties":'
+        + ' {"bar": {"type": "string"}}}'
+    )
+    result = sync_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+def test_sglang_sync_regex(sync_model):
+    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+def test_sglang_sync_cfg(sync_model):
+    with pytest.warns(
+        UserWarning,
+        match="SGLang grammar-based structured outputs expects an EBNF"
+    ):
+        result = sync_model("foo?", CFG(EBNF_YES_NO_GRAMMAR), max_tokens=10)
+        assert isinstance(result, str)
+        assert result in ["yes", "no"]
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_simple_call(async_model):
+    result = await async_model("Respond with a single word.",)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_streaming(async_model_no_model_name):
+    result = async_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=sglang_model_name,
+    )
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_multiple_samples(async_model):
+    result = await async_model("Respond with a single word.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_json(async_model):
+    json_string = (
+        '{"type": "object", "properties":'
+        + ' {"bar": {"type": "string"}}}'
+    )
+    result = await async_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_regex(async_model):
+    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_cfg(async_model):
+    result = await async_model("foo?", CFG(EBNF_YES_NO_GRAMMAR), max_tokens=10)
+    assert isinstance(result, str)
+    assert result in ["yes", "no"]
diff --git a/tests/models/test_sglang_type_adapter.py b/tests/models/test_sglang_type_adapter.py
new file mode 100644
index 000000000..7530174f9
--- /dev/null
+++ b/tests/models/test_sglang_type_adapter.py
@@ -0,0 +1,145 @@
+import io
+import json
+import pytest
+from dataclasses import dataclass
+
+from PIL import Image
+from outlines.models.sglang import SGLangTypeAdapter
+from outlines.types import CFG, JsonSchema
+
+from outlines import Vision
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
+
+
+@pytest.fixture
+def type_adapter():
+    return SGLangTypeAdapter()
+
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
+
+@pytest.fixture
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+def test_sglang_type_adapter_input_text(type_adapter):
+    message = "prompt"
+    result = type_adapter.format_input(message)
+    assert isinstance(result, dict)
+    assert len(result["messages"]) == 1
+    assert result["messages"][0]["role"] == "user"
+    assert result["messages"][0]["content"] == message
+
+
+def test_sglang_type_adapter_input_vision(type_adapter, image):
+    input_message = Vision("hello", image)
+    result = type_adapter.format_input(input_message)
+    assert isinstance(result, dict)
+
+    messages = result["messages"]
+    assert len(messages) == 1
+
+    message = messages[0]
+    assert message["role"] == "user"
+    assert len(message["content"]) == 2
+    assert message["content"][0]["type"] == "text"
+    assert message["content"][0]["text"] == "hello"
+
+    assert message["content"][1]["type"] == "image_url"
+    assert (
+        message["content"][1]["image_url"]["url"]
+        == f"data:image/png;base64,{input_message.image_str}"
+    )
+
+
+def test_sglang_type_adapter_input_invalid(type_adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = type_adapter.format_input(prompt)
+
+
+def test_sglang_type_adapter_output_type(
+    type_adapter,
+    cfg_instance,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    with pytest.warns(
+        UserWarning,
+        match="SGLang grammar-based structured outputs expects an EBNF"
+    ):
+        assert type_adapter.format_output_type(cfg_instance) == {
+            "extra_body": {"ebnf": CFG_STRING}
+        }
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "response_format": {
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": {
+                    **json.loads(JSON_SCHEMA_STRING),
+                    "additionalProperties": False,
+                },
+            },
+        }
+    }
+    # whitespace pattern is ignored
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "response_format": {
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": {
+                    **json.loads(JSON_SCHEMA_STRING),
+                    "additionalProperties": False,
+                },
+            },
+        }
+    }
+    assert type_adapter.format_output_type(int) == {
+        "extra_body": {"regex": "([+-]?(0|[1-9][0-9]*))"}
+    }
diff --git a/tests/models/test_tgi.py b/tests/models/test_tgi.py
new file mode 100644
index 000000000..52704f6c0
--- /dev/null
+++ b/tests/models/test_tgi.py
@@ -0,0 +1,180 @@
+import os
+import re
+import warnings
+from typing import AsyncGenerator, Generator
+
+import pytest
+from huggingface_hub import InferenceClient, AsyncInferenceClient
+
+from outlines.models.tgi import TGI, AsyncTGI, from_tgi
+from outlines.types.dsl import CFG, Regex, JsonSchema
+from tests.test_utils.mock_tgi_client import MockTGIInferenceClient, MockAsyncTGIInferenceClient
+
+
+YES_NO_GRAMMAR = """
+?start: answer
+
+answer: "yes" | "no"
+"""
+
+# If the TGI_SERVER_URL environment variable is set, use the real TGI server
+# Otherwise, use the mock server
+tgi_server_url = os.environ.get("TGI_SERVER_URL")
+if tgi_server_url:
+    tgi_client = InferenceClient(tgi_server_url)
+    async_tgi_client = AsyncInferenceClient(tgi_server_url)
+else:
+    warnings.warn("No TGI server URL provided, using mock server")
+    tgi_client = MockTGIInferenceClient()
+    async_tgi_client = MockAsyncTGIInferenceClient()
+
+mock_responses = [
+    (
+        {
+            'prompt': 'Respond with a single word.',
+            'max_new_tokens': 10,
+        },
+        "foo"
+    ),
+    (
+        {
+            'prompt': 'Respond with a single word.',
+            'max_new_tokens': 10,
+            'stream': True
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'prompt': 'foo?',
+            'max_new_tokens': 10,
+            'grammar': {
+                'type': 'json',
+                'value': {
+                    'type': 'object',
+                    'properties': {
+                        'bar': {'type': 'string'}
+                    },
+                    'required': ['bar']
+                }
+            }
+        },
+        '{"foo": "bar"}'
+    ),
+    (
+        {
+            'prompt': 'foo?',
+            'max_new_tokens': 10,
+            'grammar': {
+                'type': 'regex',
+                'value': '([0-9]{3})',
+            },
+        },
+        "123"
+    ),
+]
+
+# If the TGI_SERVER_URL environment variable is not set, add the mock
+# responses to the mock clients
+if not tgi_server_url:
+    async_tgi_client.add_mock_responses(mock_responses)
+    tgi_client.add_mock_responses(mock_responses)
+
+
+@pytest.fixture
+def sync_model():
+    return TGI(tgi_client)
+
+
+@pytest.fixture
+def async_model():
+    return AsyncTGI(async_tgi_client)
+
+
+def test_tgi_init():
+    model = from_tgi(
+        InferenceClient("http://localhost:11434"),
+    )
+    assert isinstance(model, TGI)
+
+    model = from_tgi(
+        AsyncInferenceClient("http://localhost:11434"),
+    )
+    assert isinstance(model, AsyncTGI)
+
+    with pytest.raises(ValueError, match="Unsupported client type"):
+        from_tgi("foo")
+
+
+def test_tgi_sync_simple_call(sync_model):
+    result = sync_model("Respond with a single word.", max_new_tokens=10)
+    assert isinstance(result, str)
+
+
+def test_tgi_sync_streaming(sync_model):
+    result = sync_model.stream(
+        "Respond with a single word.",
+        max_new_tokens=10,
+    )
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
+
+
+def test_tgi_sync_json(sync_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
+    result = sync_model("foo?", JsonSchema(json_string), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+def test_tgi_sync_regex(sync_model):
+    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+def test_tgi_sync_cfg(sync_model):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support CFG-based structured outputs",
+    ):
+        sync_model("foo?", CFG(YES_NO_GRAMMAR), max_new_tokens=10)
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_simple_call(async_model):
+    result = await async_model("Respond with a single word.", max_new_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_streaming(async_model):
+    result = async_model.stream("Respond with a single word.", max_new_tokens=10)
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_json(async_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
+    result = await async_model("foo?", JsonSchema(json_string), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_regex(async_model):
+    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_cfg(async_model):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support CFG-based structured outputs",
+    ):
+        await async_model("foo?", CFG(YES_NO_GRAMMAR), max_new_tokens=10)
diff --git a/tests/models/test_tgi_model_adapter.py b/tests/models/test_tgi_model_adapter.py
new file mode 100644
index 000000000..543ab75ff
--- /dev/null
+++ b/tests/models/test_tgi_model_adapter.py
@@ -0,0 +1,93 @@
+import json
+import pytest
+
+from outlines.models.tgi import TGITypeAdapter
+from outlines.types import CFG, JsonSchema
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
+
+
+@pytest.fixture
+def type_adapter():
+    return TGITypeAdapter()
+
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
+
+@pytest.fixture
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+
+def test_tgi_type_adapter_input_text(type_adapter):
+    message = "prompt"
+    assert message == type_adapter.format_input(message)
+
+
+def test_tgi_type_adapter_input_invalid(type_adapter):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support batch inference.",
+    ):
+        type_adapter.format_input(["foo", "bar"])
+    with pytest.raises(
+        NotImplementedError,
+        match="is not available with TGI",
+    ):
+        type_adapter.format_input({"foo": "bar"})
+
+
+def test_tgi_type_adapter_output_type(
+    type_adapter,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "grammar": {
+            "type": "json",
+            "value": json.loads(JSON_SCHEMA_STRING),
+        }
+    }
+    # whitespace_pattern is ignored
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "grammar": {
+            "type": "json",
+            "value": json.loads(JSON_SCHEMA_STRING),
+        }
+    }
+    assert type_adapter.format_output_type(int) == {
+        "grammar": {
+            "type": "regex",
+            "value": "([+-]?(0|[1-9][0-9]*))",
+        }
+    }
+
+
+def test_tgi_type_adapter_output_type_invalid(
+    type_adapter,
+    cfg_instance,
+):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support CFG-based structured outputs.",
+    ):
+        type_adapter.format_output_type(cfg_instance)
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 1404d287a..cd6f60811 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -1,117 +1,206 @@
+import re
+from enum import Enum
+
+from pydantic import BaseModel
 import pytest
-import torch
-from transformers import AutoTokenizer
-from transformers.models.gpt2 import GPT2TokenizerFast
+import transformers
 
-from outlines.models.transformers import TransformerTokenizer, transformers
+import outlines
+from outlines.models.transformers import (
+    Transformers,
+    TransformerTokenizer,
+    TransformersTypeAdapter,
+)
+from outlines.types import Regex
 
-TEST_MODEL = "hf-internal-testing/tiny-random-GPTJForCausalLM"
 
+TEST_MODEL = "erwanf/gpt2-mini"
+TEST_MODEL_MAMBA = "hf-internal-testing/tiny-random-MambaForCausalLM"
+TEST_MODEL_BART = "trl-internal-testing/tiny-BartModel"
 
-def test_tokenizer():
-    tokenizer = AutoTokenizer.from_pretrained(TEST_MODEL, padding_side="left")
-    tokenizer = TransformerTokenizer(tokenizer)
-    assert tokenizer.eos_token_id == 0
-    assert tokenizer.pad_token_id == 0
-    assert isinstance(tokenizer.tokenizer, GPT2TokenizerFast)
 
-    token_ids, attention_mask = tokenizer.encode("Test")
-    assert token_ids.ndim == 2
-    assert token_ids.shape[0] == 1
-    assert isinstance(token_ids, torch.LongTensor)
-    assert token_ids.shape == attention_mask.shape
+def test_transformers_instantiate_invalid():
+    with pytest.raises(ValueError):
+        outlines.from_transformers(
+            transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+            int,
+        )
 
-    token_ids, attention_mask = tokenizer.encode(["Test", "Test"])
-    assert token_ids.ndim == 2
-    assert token_ids.shape[0] == 2
-    assert isinstance(token_ids, torch.LongTensor)
-    assert token_ids.shape == attention_mask.shape
 
-    token_ids, attention_mask = tokenizer.encode(["Test", "A long sentence"])
-    assert token_ids.shape == attention_mask.shape
-    assert attention_mask[0][0] == tokenizer.pad_token_id
+def test_transformers_instantiate_simple():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+    )
+    assert isinstance(model, Transformers)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersTypeAdapter)
+    assert model.tensor_library_name == "torch"
 
-    text = tokenizer.decode(torch.tensor([[0, 1, 2]]))
-    isinstance(text, str)
 
-    text = tokenizer.decode(torch.tensor([[0, 1, 2], [3, 4, 5]]))
-    isinstance(text, list)
-    isinstance(text[0], str)
-    isinstance(text[1], str)
+def test_transformers_instantiate_flax_model():
+    model = outlines.from_transformers(
+        transformers.FlaxAutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+    )
+    assert isinstance(model, Transformers)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersTypeAdapter)
+    assert model.tensor_library_name == "jax"
+
 
-    tokenizer = AutoTokenizer.from_pretrained(
-        TEST_MODEL, additional_special_tokens=["<t1>", "<t2>"]
+def test_transformers_instantiate_tensorflow_model():
+    model = outlines.from_transformers(
+        transformers.TFAutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
     )
-    tokenizer = TransformerTokenizer(tokenizer)
-    assert "<t1>" in tokenizer.special_tokens
-    assert "<t2>" in tokenizer.special_tokens
+    assert isinstance(model, Transformers)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersTypeAdapter)
+    assert model.tensor_library_name == "tensorflow"
 
 
-def test_llama_tokenizer():
-    tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/llama-tokenizer")
-    tokenizer = TransformerTokenizer(tokenizer)
+def test_transformers_instantiate_mamba():
+    model = outlines.from_transformers(
+        transformers.MambaForCausalLM.from_pretrained(TEST_MODEL_MAMBA),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+    )
+    assert isinstance(model, Transformers)
 
-    # Broken
-    assert tokenizer.tokenizer.convert_tokens_to_string(["▁baz"]) == "baz"
-    assert tokenizer.tokenizer.convert_tokens_to_string(["<0x20>"]) == ""
-    assert tokenizer.tokenizer.convert_tokens_to_string(["▁▁▁"]) == "  "
 
-    # Not broken
-    assert tokenizer.convert_token_to_string("▁baz") == " baz"
-    assert tokenizer.convert_token_to_string("<0x20>") == " "
-    assert tokenizer.convert_token_to_string("▁▁▁") == "   "
+def test_transformers_instantiate_tokenizer_kwargs():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(
+            TEST_MODEL, additional_special_tokens=["<t1>", "<t2>"]
+        ),
+    )
+    assert "<t1>" in model.tokenizer.special_tokens
+    assert "<t2>" in model.tokenizer.special_tokens
 
 
-def test_model():
-    model = transformers(TEST_MODEL, device="cpu")
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.model.device.type == "cpu"
+@pytest.fixture
+def model():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+    )
+    return model
 
-    model = transformers(TEST_MODEL, model_kwargs={"device_map": "cpu"})
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.model.device.type == "cpu"
 
-    model = transformers(TEST_MODEL, device="cpu", model_kwargs={"device_map": "cuda"})
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.model.device.type == "cpu"
-
-    input_ids = torch.tensor([[0, 1, 2]])
-    logits, kv_cache = model(input_ids, torch.ones_like(input_ids))
-    assert logits.type() == "torch.FloatTensor"
-    assert logits.ndim == 2
-    assert logits.shape[0] == 1
-    assert len(kv_cache) == model.model.config.n_layer
-    assert len(kv_cache[0]) == 2
-    assert kv_cache[0][0].shape[1] == model.model.config.n_head
-    assert kv_cache[0][0].shape[2] == 3  # number of tokens
-
-    input_ids = torch.tensor([[0, 1, 2], [3, 4, 5], [6, 7, 8]])
-    logits, kv_cache = model(input_ids, torch.ones_like(input_ids))
-    assert logits.type() == "torch.FloatTensor"
-    assert logits.ndim == 2
-    assert logits.shape[0] == 3
-    assert len(kv_cache) == model.model.config.n_layer
-    assert len(kv_cache[0]) == 2
-    assert kv_cache[0][0].shape[1] == model.model.config.n_head
-    assert kv_cache[0][0].shape[2] == 3  # number of tokens
-
-    with pytest.raises(AssertionError):
-        input_ids = torch.tensor([[[0, 1, 2], [3, 4, 5]], [[6, 7, 8], [0, 1, 2]]])
-        logits = model(input_ids, torch.ones_like(input_ids))
-
-
-def test_tokenizer_eq_hash():
-    tokenizer_hf = AutoTokenizer.from_pretrained("gpt2")
-
-    tokenizer = TransformerTokenizer(tokenizer_hf)
-    tokenizer_2 = TransformerTokenizer(tokenizer_hf)
-
-    assert tokenizer == tokenizer_2
-    assert hash(tokenizer) == hash(tokenizer_2)
-
-    tokenizer_hf_2 = AutoTokenizer.from_pretrained("gpt2")
-    tokenizer_hf_2.add_tokens(["test_token"])
-
-    tokenizer_3 = TransformerTokenizer(tokenizer_hf_2)
-    assert tokenizer != tokenizer_3
-    assert hash(tokenizer) != hash(tokenizer_3)
+@pytest.fixture
+def model_bart():
+    model = outlines.from_transformers(
+        transformers.BartForConditionalGeneration.from_pretrained(TEST_MODEL_BART),
+        transformers.BartTokenizer.from_pretrained(TEST_MODEL_BART),
+    )
+    return model
+
+
+def test_transformers_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_transformers_call(model, model_bart):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+    result = model_bart("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+def test_transformers_inference_kwargs(model):
+    result = model("Respond with one word. Not more.", max_new_tokens=100)
+    assert isinstance(result, str)
+
+
+def test_transformers_invalid_inference_kwargs(model):
+    with pytest.raises(ValueError):
+        model("Respond with one word. Not more.", foo="bar")
+
+
+def test_transformers_regex(model):
+    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
+
+
+def test_transformers_json(model):
+    class Character(BaseModel):
+        name: str
+
+    result = model("Create a character with a name.", Character)
+    assert "name" in result
+
+
+def test_transformers_choice(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Foo)
+    assert result in ["cat", "dog"]
+
+
+def test_transformers_batch_samples(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+    result = model(
+        "Respond with one word. Not more.", num_return_sequences=2, num_beams=2
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."]
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."],
+        num_return_sequences=2,
+        num_beams=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+
+
+def test_transformers_batch_samples_constrained(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Foo, num_return_sequences=2, num_beams=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Foo,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Foo,
+        num_return_sequences=2,
+        num_beams=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+        assert item[0] in ["cat", "dog"]
+        assert item[1] in ["cat", "dog"]
+
+
+def test_transformers_streaming(model):
+    with pytest.raises(NotImplementedError, match="Streaming is not implemented"):
+        model.stream("Respond with one word. Not more.")
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
new file mode 100644
index 000000000..448bec221
--- /dev/null
+++ b/tests/models/test_transformers_multimodal.py
@@ -0,0 +1,194 @@
+# we only test vision models here as audio models are too heavy to run on CI
+
+import re
+from enum import Enum
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image
+from pydantic import BaseModel
+from transformers import (
+    LlavaForConditionalGeneration,
+    AutoProcessor,
+)
+
+import outlines
+from outlines.models.transformers import (
+    TransformersMultiModal,
+    TransformerTokenizer,
+    TransformersMultiModalTypeAdapter,
+)
+from outlines.types import Regex
+
+TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+IMAGE_URLS = [
+    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
+    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
+]
+
+
+@pytest.fixture
+def images():
+    def img_from_url(url):
+        img_byte_stream = BytesIO(urlopen(url).read())
+        return Image.open(img_byte_stream).convert("RGB")
+
+    return [img_from_url(url) for url in IMAGE_URLS]
+
+
+@pytest.fixture
+def model():
+    return outlines.from_transformers(
+        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+        AutoProcessor.from_pretrained(TEST_MODEL),
+    )
+
+
+def test_transformers_vision_instantiate_simple():
+    model = outlines.from_transformers(
+        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+        AutoProcessor.from_pretrained(TEST_MODEL),
+    )
+    assert isinstance(model, TransformersMultiModal)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersMultiModalTypeAdapter)
+    assert model.tensor_library_name == "torch"
+
+
+def test_transformers_vision_simple(model, images):
+    result = model.generate(
+        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
+        None,
+        max_new_tokens=2,
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_vision_call(model, images):
+    result = model(
+        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
+        max_new_tokens=2,
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_vision_wrong_number_images(model, images):
+    with pytest.raises(ValueError):
+        model(
+            {
+                "text": "<image>Describe this image in one sentence:",
+                "images": [images[0], images[1]],
+            }
+        )
+
+
+def test_transformers_vision_wrong_input_type(model):
+    with pytest.raises(NotImplementedError):
+        model.generate("invalid input", None)
+
+
+def test_transformers_inference_kwargs(model, images):
+    result = model(
+        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
+        max_new_tokens=2,
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_invalid_inference_kwargs(model, images):
+    with pytest.raises(ValueError):
+        model(
+            {
+                "text": "<image>Describe this image in one sentence:",
+                "images": images[0],
+            },
+            foo="bar",
+        )
+
+
+def test_transformers_several_images(model, images):
+    result = model(
+        {
+            "text": "<image><image>Describe this image in one sentence:",
+            "images": [images[0], images[1]],
+        },
+        max_new_tokens=2,
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_vision_json(model, images):
+    class Foo(BaseModel):
+        name: str
+
+    result = model(
+        {"text": "<image>Give a name to this animal.", "images": images[0]},
+        Foo,
+        max_new_tokens=10,
+    )
+    assert "name" in result
+
+
+def test_transformers_vision_regex(model, images):
+    result = model(
+        {"text": "<image>How old is it?", "images": images[0]},
+        Regex(r"[0-9]")
+    )
+
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
+
+
+def test_transformers_vision_choice(model, images):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model(
+        {"text": "<image>Is it a cat or a dog?", "images": images[0]},
+        Foo,
+    )
+
+    assert isinstance(result, str)
+    assert result in ["cat", "dog"]
+
+
+def test_transformers_vision_batch_samples(model, images):
+    result = model(
+        {"text": "<image>Describe this image in one sentence.", "images": images[0]},
+        num_return_sequences=2,
+        num_beams=2,
+        max_new_tokens=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        {
+            "text": [
+                "<image>Describe this image in one sentence.",
+                "<image>Describe this image in one sentence.",
+            ],
+            "images": [images[0], images[1]],
+        },
+        max_new_tokens=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        {
+            "text": [
+                "<image>Describe this image in one sentence.",
+                "<image>Describe this image in one sentence.",
+            ],
+            "images": [images[0], images[1]],
+        },
+        num_return_sequences=2,
+        num_beams=2,
+        max_new_tokens=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
new file mode 100644
index 000000000..28569d25f
--- /dev/null
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -0,0 +1,48 @@
+import pytest
+
+import outlines
+import transformers
+from transformers import LogitsProcessorList
+
+from outlines.models.transformers import TransformersMultiModalTypeAdapter
+from outlines.processors.structured import RegexLogitsProcessor
+
+
+@pytest.fixture
+def adapter():
+    return TransformersMultiModalTypeAdapter()
+
+
+@pytest.fixture
+def logits_processor():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=model.tokenizer,
+        tensor_library_name="torch",
+    )
+
+
+def test_transformers_multimodal_type_adapter_format_input(adapter):
+    with pytest.raises(NotImplementedError):
+        adapter.format_input("hello")
+
+    with pytest.raises(ValueError):
+        adapter.format_input({"foo": "bar"})
+
+    assert adapter.format_input({"text": "foo"}) == {"text": "foo"}
+
+
+def test_transformers_multimodal_type_adapter_format_output_type(
+    adapter, logits_processor
+):
+    formatted = adapter.format_output_type(logits_processor)
+    assert isinstance(formatted, LogitsProcessorList)
+    assert len(formatted) == 1
+    assert isinstance(formatted[0], RegexLogitsProcessor)
+
+    formatted = adapter.format_output_type(None)
+    assert formatted is None
diff --git a/tests/models/test_transformers_tokenizer.py b/tests/models/test_transformers_tokenizer.py
new file mode 100644
index 000000000..bf3452302
--- /dev/null
+++ b/tests/models/test_transformers_tokenizer.py
@@ -0,0 +1,142 @@
+import pytest
+
+import transformers
+
+from outlines.models.transformers import (
+    get_llama_tokenizer_types,
+    TransformerTokenizer,
+)
+
+
+TEST_MODEL = "erwanf/gpt2-mini"
+TEST_MODEL_SEQ2SEQ = "hf-internal-testing/tiny-random-t5"
+
+
+@pytest.fixture
+def tokenizer():
+    return transformers.AutoTokenizer.from_pretrained(TEST_MODEL)
+
+
+@pytest.fixture
+def tokenizer_no_pad_token_id(tokenizer):
+    tokenizer.pad_token_id = None
+    return tokenizer
+
+
+@pytest.fixture
+def tokenizer_seq2seq():
+    return transformers.AutoTokenizer.from_pretrained(TEST_MODEL_SEQ2SEQ)
+
+
+@pytest.fixture
+def transformer_tokenizer(tokenizer):
+    return TransformerTokenizer(tokenizer)
+
+
+@pytest.fixture
+def another_transformer_tokenizer(tokenizer):
+    return TransformerTokenizer(tokenizer)
+
+
+@pytest.fixture
+def transformer_tokenizer_seq2seq(tokenizer_seq2seq):
+    return TransformerTokenizer(tokenizer_seq2seq)
+
+
+def test_get_llama_tokenizer_types():
+    tokenizer_types = get_llama_tokenizer_types()
+    assert tokenizer_types[0] is transformers.models.llama.LlamaTokenizer
+    assert tokenizer_types[1] is transformers.models.llama.LlamaTokenizerFast
+    assert tokenizer_types[2] is transformers.models.code_llama.CodeLlamaTokenizer
+    assert tokenizer_types[3] is transformers.models.code_llama.CodeLlamaTokenizerFast
+
+
+def test_transformer_tokenizer_init(
+    tokenizer,
+    tokenizer_no_pad_token_id
+):
+    # tokenizer with a pad_token_id
+    transformer_tokenizer = TransformerTokenizer(tokenizer)
+    assert transformer_tokenizer.tokenizer == tokenizer
+    assert transformer_tokenizer.eos_token_id == tokenizer.eos_token_id
+    assert transformer_tokenizer.pad_token_id == tokenizer.pad_token_id
+    assert transformer_tokenizer.special_tokens == set(tokenizer.all_special_tokens)
+    assert transformer_tokenizer.vocabulary == tokenizer.get_vocab()
+
+    # tokenizer with no pad_token_id
+    transformer_tokenizer_no_pad_token_id = TransformerTokenizer(tokenizer_no_pad_token_id)
+    assert transformer_tokenizer_no_pad_token_id.tokenizer == tokenizer_no_pad_token_id
+    assert transformer_tokenizer_no_pad_token_id.eos_token_id == tokenizer_no_pad_token_id.eos_token_id
+    assert transformer_tokenizer_no_pad_token_id.pad_token_id == tokenizer_no_pad_token_id.eos_token_id
+    assert transformer_tokenizer_no_pad_token_id.special_tokens == set(tokenizer_no_pad_token_id.all_special_tokens)
+    assert transformer_tokenizer_no_pad_token_id.vocabulary == tokenizer_no_pad_token_id.get_vocab()
+
+
+def test_transformer_tokenizer_encode(transformer_tokenizer):
+    input_ids, attention_mask = transformer_tokenizer.encode("Hello, world!")
+    assert input_ids is not None
+    assert attention_mask is not None
+    assert input_ids.shape == attention_mask.shape
+
+
+def test_transformer_tokenizer_decode(transformer_tokenizer):
+    input_ids, _ = transformer_tokenizer.encode("Hello, world!")
+    decoded_text = transformer_tokenizer.decode(input_ids)
+    assert isinstance(decoded_text, list)
+    assert "Hello, world!" in decoded_text[0]
+
+
+def test_transformer_tokenizer_convert_token_to_string(transformer_tokenizer):
+    # regular
+    transformer_tokenizer.is_llama = False
+    token = transformer_tokenizer.tokenizer.tokenize("Hello")[0]
+    string = transformer_tokenizer.convert_token_to_string(token)
+    assert isinstance(string, str)
+    assert "Hello" in string
+
+    # is_llama + <0x20>
+    transformer_tokenizer.is_llama = True
+    string = transformer_tokenizer.convert_token_to_string("<0x20>")
+    assert isinstance(string, str)
+    assert " " in string
+
+
+def test_transformer_tokenizer_eq(
+    transformer_tokenizer,
+    another_transformer_tokenizer,
+    transformer_tokenizer_seq2seq,
+):
+    # different types of object
+    assert transformer_tokenizer.__eq__(1) == NotImplemented
+
+    # regular case
+    assert transformer_tokenizer == another_transformer_tokenizer
+    assert transformer_tokenizer != transformer_tokenizer_seq2seq
+
+    # with model name and kwargs attributes
+    transformer_tokenizer.model_name = "foo"
+    transformer_tokenizer.kwargs = {"foo": "bar"}
+    another_transformer_tokenizer.model_name = "foo"
+    another_transformer_tokenizer.kwargs = {"foo": "bar"}
+    assert transformer_tokenizer == another_transformer_tokenizer
+
+
+def test_transformer_tokenizer_hash(
+    transformer_tokenizer,
+    another_transformer_tokenizer,
+    transformer_tokenizer_seq2seq,
+):
+    assert isinstance(hash(transformer_tokenizer), int)
+    assert hash(transformer_tokenizer) == hash(another_transformer_tokenizer)
+    assert hash(transformer_tokenizer) != hash(transformer_tokenizer_seq2seq)
+
+
+def test_transformer_tokenizer_getstate_setstate(
+    transformer_tokenizer,
+    another_transformer_tokenizer,
+):
+    state = transformer_tokenizer.__getstate__()
+    assert "tokenizer" in state
+
+    another_transformer_tokenizer.__setstate__(state)
+    assert another_transformer_tokenizer == transformer_tokenizer
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
new file mode 100644
index 000000000..44f6ec4e6
--- /dev/null
+++ b/tests/models/test_transformers_type_adapter.py
@@ -0,0 +1,50 @@
+import pytest
+
+import transformers
+from transformers import LogitsProcessorList
+
+import outlines
+from outlines.models.transformers import TransformersTypeAdapter
+from outlines.processors.structured import RegexLogitsProcessor
+
+
+@pytest.fixture
+def adapter():
+    return TransformersTypeAdapter()
+
+
+@pytest.fixture
+def logits_processor():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=model.tokenizer,
+        tensor_library_name="torch",
+    )
+
+
+def test_transformers_type_adapter_format_input(adapter):
+    with pytest.raises(NotImplementedError):
+        adapter.format_input(int)
+
+    assert adapter.format_input("Hello, world!") == "Hello, world!"
+
+    assert adapter.format_input(["Hello, world!", "Hello, world!"]) == [
+        "Hello, world!",
+        "Hello, world!",
+    ]
+
+
+def test_transformers_type_adapter_format_output_type(
+    adapter, logits_processor
+):
+    formatted = adapter.format_output_type(logits_processor)
+    assert isinstance(formatted, LogitsProcessorList)
+    assert len(formatted) == 1
+    assert isinstance(formatted[0], RegexLogitsProcessor)
+
+    formatted = adapter.format_output_type(None)
+    assert formatted is None
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
new file mode 100644
index 000000000..76a9da814
--- /dev/null
+++ b/tests/models/test_vllm.py
@@ -0,0 +1,254 @@
+import os
+import re
+import warnings
+from typing import AsyncGenerator, Generator
+
+import pytest
+from openai import AsyncOpenAI, OpenAI
+
+from outlines.models.vllm import VLLM, AsyncVLLM, from_vllm
+from outlines.types.dsl import CFG, Regex, JsonSchema
+from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
+
+
+YES_NO_GRAMMAR = """
+?start: answer
+
+answer: "yes" | "no"
+"""
+
+
+# If the VLLM_SERVER_URL environment variable is set, use the real vLLM server
+# Otherwise, use the mock server
+vllm_server_url = os.environ.get("VLLM_SERVER_URL")
+vllm_model_name = os.environ.get(
+    "VLLM_MODEL_NAME", "microsoft/Phi-3-mini-4k-instruct"
+)
+if vllm_server_url:
+    openai_client = OpenAI(base_url=vllm_server_url)
+    async_openai_client = AsyncOpenAI(base_url=vllm_server_url)
+else:
+    warnings.warn("No VLLM server URL provided, using mock server")
+    openai_client = MockOpenAIClient()
+    async_openai_client = MockAsyncOpenAIClient()
+
+mock_responses = [
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': vllm_model_name,
+        },
+        "foo"
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': vllm_model_name,
+            'stream': True
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'n': 2,
+            'model': vllm_model_name,
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+            'guided_json': {
+                'type': 'object',
+                'properties': {
+                    'bar': {'type': 'string'}
+                }
+            },
+            }
+        },
+        '{"foo": "bar"}'
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'guided_regex': '([0-9]{3})',
+            },
+        },
+        "123"
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'guided_grammar': YES_NO_GRAMMAR,
+            },
+        },
+        "yes"
+    ),
+]
+
+
+# If the VLLM_SERVER_URL environment variable is not set, add the mock
+# responses to the mock clients
+if not vllm_server_url:
+    async_openai_client.add_mock_responses(mock_responses)
+    openai_client.add_mock_responses(mock_responses)
+
+
+@pytest.fixture
+def sync_model():
+    return VLLM(openai_client, vllm_model_name)
+
+
+@pytest.fixture
+def sync_model_no_model_name():
+    return VLLM(openai_client)
+
+
+@pytest.fixture
+def async_model():
+    return AsyncVLLM(async_openai_client, vllm_model_name)
+
+
+@pytest.fixture
+def async_model_no_model_name():
+    return AsyncVLLM(async_openai_client)
+
+
+def test_vllm_init():
+    # We do not rely on the mock server here because we need an object
+    # of type OpenAI and AsyncOpenAI to test the init function.
+    openai_client = OpenAI(base_url="http://localhost:11434")
+    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
+
+    # Sync with model name
+    model = from_vllm(openai_client, vllm_model_name)
+    assert isinstance(model, VLLM)
+    assert model.client == openai_client
+    assert model.model_name == vllm_model_name
+
+    # Sync without model name
+    model = from_vllm(openai_client)
+    assert isinstance(model, VLLM)
+    assert model.client == openai_client
+    assert model.model_name is None
+
+    # Async with model name
+    model = from_vllm(async_openai_client, vllm_model_name)
+    assert isinstance(model, AsyncVLLM)
+    assert model.client == async_openai_client
+    assert model.model_name == vllm_model_name
+
+    # Async without model name
+    model = from_vllm(async_openai_client)
+    assert isinstance(model, AsyncVLLM)
+    assert model.client == async_openai_client
+    assert model.model_name is None
+
+    with pytest.raises(ValueError, match="Unsupported client type"):
+        from_vllm("foo")
+
+
+def test_vllm_sync_simple_call(sync_model):
+    result = sync_model("Respond with a single word.",)
+    assert isinstance(result, str)
+
+
+def test_vllm_sync_streaming(sync_model_no_model_name):
+    result = sync_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=vllm_model_name,
+    )
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
+
+
+def test_vllm_sync_multiple_samples(sync_model):
+    result = sync_model("Respond with a single word.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+def test_vllm_sync_json(sync_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}}'
+    result = sync_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+def test_vllm_sync_regex(sync_model):
+    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+def test_vllm_sync_cfg(sync_model):
+    result = sync_model("foo?", CFG(YES_NO_GRAMMAR), max_tokens=10)
+    assert isinstance(result, str)
+    assert result in ["yes", "no"]
+
+
+@pytest.mark.asyncio
+async def test_vllm_async_simple_call(async_model):
+    result = await async_model("Respond with a single word.",)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_vllm_async_streaming(async_model_no_model_name):
+    result = async_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=vllm_model_name,
+    )
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+@pytest.mark.asyncio
+async def test_vllm_async_multiple_samples(async_model):
+    result = await async_model("Respond with a single word.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+@pytest.mark.asyncio
+async def test_vllm_async_json(async_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}}'
+    result = await async_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+@pytest.mark.asyncio
+async def test_vllm_async_regex(async_model):
+    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+@pytest.mark.asyncio
+async def test_vllm_async_cfg(async_model):
+    result = await async_model("foo?", CFG(YES_NO_GRAMMAR), max_tokens=10)
+    assert isinstance(result, str)
+    assert result in ["yes", "no"]
diff --git a/tests/models/test_vllm_offline.py b/tests/models/test_vllm_offline.py
new file mode 100644
index 000000000..01d5d6a59
--- /dev/null
+++ b/tests/models/test_vllm_offline.py
@@ -0,0 +1,117 @@
+import re
+from enum import Enum
+
+import pytest
+from pydantic import BaseModel
+
+try:
+    from vllm import LLM, SamplingParams
+    HAS_VLLM = True
+except ImportError:
+    HAS_VLLM = False
+
+import outlines
+from outlines.models.vllm_offline import (
+    VLLMOffline,
+    VLLMOfflineTypeAdapter,
+    from_vllm_offline
+)
+from outlines.types import Regex
+
+
+TEST_MODEL = "erwanf/gpt2-mini"
+
+pytestmark = pytest.mark.skipif(
+    not HAS_VLLM,
+    reason="vLLM models can only be run on GPU."
+)
+
+
+def test_vllm_model_initialization():
+    model = from_vllm_offline(LLM(TEST_MODEL))
+    assert isinstance(model, VLLMOffline)
+    assert isinstance(model.model, LLM)
+    assert isinstance(model.type_adapter, VLLMOfflineTypeAdapter)
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    model = outlines.from_vllm_offline(LLM(TEST_MODEL))
+    return model
+
+
+def test_vllm_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_vllm_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+def test_vllm_inference_kwargs(model):
+    result = model("Write a short story about a cat.", sampling_params=SamplingParams(max_tokens=2), use_tqdm=True)
+    assert isinstance(result, str)
+    assert len(result) <= 20
+
+
+def test_vllm_invalid_inference_kwargs(model):
+    with pytest.raises(TypeError):
+        model("Respond with one word. Not more.", foo="bar")
+
+
+def test_vllm_regex(model):
+    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
+
+
+def test_vllm_json(model):
+    class Character(BaseModel):
+        name: str
+
+    result = model("Create a character with a name.", Character)
+    assert "name" in result
+
+
+def test_vllm_choice(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Foo)
+    assert result in ["cat", "dog"]
+
+
+def test_vllm_batch_samples(model):
+    result = model(
+        "Respond with one word. Not more.",
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."]
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."],
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+
+
+def test_vllm_streaming(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="Streaming is not available for the vLLM integration."
+    ):
+        model.stream("Respond with one word. Not more.")
diff --git a/tests/models/test_vllm_offline_type_adapter.py b/tests/models/test_vllm_offline_type_adapter.py
new file mode 100644
index 000000000..b8a18b6d1
--- /dev/null
+++ b/tests/models/test_vllm_offline_type_adapter.py
@@ -0,0 +1,84 @@
+import io
+import json
+from dataclasses import dataclass
+
+import pytest
+
+from outlines.models.vllm_offline import VLLMOfflineTypeAdapter
+from outlines.types import CFG, JsonSchema, Regex
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
+
+
+@pytest.fixture
+def type_adapter():
+    return VLLMOfflineTypeAdapter()
+
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
+
+@pytest.fixture
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+@pytest.fixture
+def regex_instance():
+    return Regex(r"[0-9]+")
+
+
+def test_vllm_type_adapter_input_text(type_adapter):
+    # string
+    message = "prompt"
+    result = type_adapter.format_input(message)
+    assert message == result
+
+    # list of strings
+    messages = ["foo", "bar"]
+    result = type_adapter.format_input(messages)
+    assert messages == result
+
+
+def test_vllm_type_adapter_input_invalid(type_adapter):
+    with pytest.raises(NotImplementedError):
+        _ = type_adapter.format_input({"foo": "bar"})
+
+
+def test_vllm_type_adapter_output_type(
+    type_adapter,
+    cfg_instance,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+    regex_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    assert type_adapter.format_output_type(cfg_instance) == {
+        "grammar": CFG_STRING
+    }
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "json": json.loads(JSON_SCHEMA_STRING)
+    }
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "json": json.loads(JSON_SCHEMA_STRING),
+        "whitespace_pattern": "\n"
+    }
+    assert type_adapter.format_output_type(regex_instance) == {
+        "regex": "([0-9]+)"
+    }
diff --git a/tests/models/test_vllm_type_adapter.py b/tests/models/test_vllm_type_adapter.py
new file mode 100644
index 000000000..46d021a8c
--- /dev/null
+++ b/tests/models/test_vllm_type_adapter.py
@@ -0,0 +1,121 @@
+import io
+import json
+import pytest
+from dataclasses import dataclass
+
+from PIL import Image
+from outlines.models.vllm import VLLMTypeAdapter
+from outlines.types import CFG, JsonSchema
+
+from outlines import Vision
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
+
+
+@pytest.fixture
+def type_adapter():
+    return VLLMTypeAdapter()
+
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
+
+@pytest.fixture
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+def test_vllm_type_adapter_input_text(type_adapter):
+    message = "prompt"
+    result = type_adapter.format_input(message)
+    assert isinstance(result, dict)
+    assert len(result["messages"]) == 1
+    assert result["messages"][0]["role"] == "user"
+    assert result["messages"][0]["content"] == message
+
+
+def test_vllm_type_adapter_input_vision(type_adapter, image):
+    input_message = Vision("hello", image)
+    result = type_adapter.format_input(input_message)
+    assert isinstance(result, dict)
+
+    messages = result["messages"]
+    assert len(messages) == 1
+
+    message = messages[0]
+    assert message["role"] == "user"
+    assert len(message["content"]) == 2
+    assert message["content"][0]["type"] == "text"
+    assert message["content"][0]["text"] == "hello"
+
+    assert message["content"][1]["type"] == "image_url"
+    assert (
+        message["content"][1]["image_url"]["url"]
+        == f"data:image/png;base64,{input_message.image_str}"
+    )
+
+
+def test_vllm_type_adapter_input_invalid(type_adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = type_adapter.format_input(prompt)
+
+
+def test_vllm_type_adapter_output_type(
+    type_adapter,
+    cfg_instance,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    assert type_adapter.format_output_type(cfg_instance) == {
+        "guided_grammar": CFG_STRING
+    }
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "guided_json": json.loads(JSON_SCHEMA_STRING)
+    }
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "guided_json": json.loads(JSON_SCHEMA_STRING),
+        "whitespace_pattern": "\n"
+    }
+    assert type_adapter.format_output_type(int) == {
+        "guided_regex": "([+-]?(0|[1-9][0-9]*))"
+    }
diff --git a/tests/processors/test_base_processor.py b/tests/processors/test_base_processor.py
index d2a1e1af2..3b729410a 100644
--- a/tests/processors/test_base_processor.py
+++ b/tests/processors/test_base_processor.py
@@ -7,74 +7,96 @@
 
 from outlines.processors.base_logits_processor import OutlinesLogitsProcessor
 
-arrays = {
-    "list": [[1.0, 2.0], [3.0, 4.0]],
-    "np": np.array([[1, 2], [3, 4]], dtype=np.float32),
-    "jax": jnp.array([[1, 2], [3, 4]], dtype=jnp.float32),
-    "torch": torch.tensor([[1, 2], [3, 4]], dtype=torch.float32),
-}
-
 try:
     import mlx.core as mx
-
-    arrays["mlx"] = mx.array([[1, 2], [3, 4]], dtype=mx.float32)
-    arrays["mlx_bfloat16"] = mx.array([[1, 2], [3, 4]], dtype=mx.bfloat16)
+    HAS_MLX = True
 except ImportError:
-    pass
+    HAS_MLX = False
 
-try:
-    import jax.numpy as jnp
 
-    arrays["jax"] = jnp.array([[1, 2], [3, 4]], dtype=jnp.float32)
-except ImportError:
-    pass
+libraries = ["numpy", "torch", "jax"]
+if HAS_MLX:
+    libraries.append("mlx")
 
+# we check the accepted shapes:
+# - both 1D
+# - both 2D
+# - input_ids 1D and logits 2D with a single sequence
+# we raise an error if the shapes are not accepted:
+# - input_ids 2D and logits 1D
+# - input_ids 1D and logits 2D, but with multiple sequences
+# - both 3D
+arrays = {
+    "numpy": [
+        (np.array([1, 2], dtype=np.float32), np.array([1, 2], dtype=np.int32), None),
+        (np.array([[1, 2], [3, 4]], dtype=np.float32), np.array([[1, 2], [3, 4]], dtype=np.int32), None),
+        (np.array([1, 2], dtype=np.float32), np.array([[1, 2]], dtype=np.int32), None),
+        (np.array([[1, 2]], dtype=np.float32), np.array([1, 2], dtype=np.int32), AssertionError),
+        (np.array([1, 2], dtype=np.float32), np.array([[1, 2], [3, 4]], dtype=np.int32), AssertionError),
+        (np.array([[[1, 2]]], dtype=np.float32), np.array([[[1, 2]]], dtype=np.int32), ValueError),
+    ],
+    "torch": [
+        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([1, 2], dtype=torch.int32), None),
+        (torch.tensor([[1, 2], [3, 4]], dtype=torch.float32), torch.tensor([[1, 2], [3, 4]], dtype=torch.int32), None),
+        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([[1, 2]], dtype=torch.int32), None),
+        (torch.tensor([[1, 2]], dtype=torch.float32), torch.tensor([1, 2], dtype=torch.int32), AssertionError),
+        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([[1, 2], [3, 4]], dtype=torch.int32), AssertionError),
+        (torch.tensor([[[1, 2]]], dtype=torch.float32), torch.tensor([[[1, 2]]], dtype=torch.int32), ValueError),
+    ],
+    "jax": [
+        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), None),
+        (jnp.array([[1, 2], [3, 4]], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), None),
+        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2]], dtype=jnp.int32), None),
+        (jnp.array([[1, 2]], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), AssertionError),
+        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), AssertionError),
+        (jnp.array([[[1, 2]]], dtype=jnp.float32), jnp.array([[[1, 2]]], dtype=jnp.int32), ValueError),
+    ],
+}
+if HAS_MLX:
+    arrays["mlx"] = [
+        (mx.array([1, 2], dtype=mx.float32), mx.array([1, 2], dtype=mx.int32), None),
+        (mx.array([[1, 2], [3, 4]], dtype=mx.float32), mx.array([[1, 2], [3, 4]], dtype=mx.int32), None),
+        (mx.array([1, 2], dtype=mx.float32), mx.array([[1, 2]], dtype=mx.int32), None),
+        (mx.array([[1, 2]], dtype=mx.float32), mx.array([1, 2], dtype=mx.int32), AssertionError),
+        (mx.array([1, 2], dtype=mx.float32), mx.array([[1, 2], [3, 4]], dtype=mx.int32), AssertionError),
+        (mx.array([[[1, 2]]], dtype=mx.float32), mx.array([[[1, 2]]], dtype=mx.int32), ValueError),
+    ]
 
-# Mock implementation of the abstract class for testing
 class MockLogitsProcessor(OutlinesLogitsProcessor):
-    def process_logits(
-        self, input_ids: List[List[int]], logits: torch.Tensor
-    ) -> torch.Tensor:
-        # For testing purposes, let's just return logits multiplied by 2
-        return logits * 2
-
-
-@pytest.fixture
-def processor():
-    """Fixture for creating an instance of the MockLogitsProcessor."""
-    return MockLogitsProcessor()
-
-
-@pytest.mark.parametrize("array_type", arrays.keys())
-def test_to_torch(array_type, processor):
-    data = arrays[array_type]
-    torch_tensor = processor._to_torch(data)
-    assert isinstance(torch_tensor, torch.Tensor)
-    assert torch.allclose(
-        torch_tensor.cpu(), torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
-    )
-
-
-@pytest.mark.parametrize("array_type", arrays.keys())
-def test_from_torch(array_type, processor):
-    torch_tensor = torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
-    data = processor._from_torch(torch_tensor, type(arrays[array_type]))
-    assert isinstance(data, type(arrays[array_type]))
-    if array_type == "mlx_bfloat16":
-        # For bfloat16, we expect the output to be float32 due to the conversion
-        assert data.dtype == mx.float32
-        assert np.allclose(np.array(data), np.array([[1, 2], [3, 4]], dtype=np.float32))
-    else:
-        assert np.allclose(data, arrays[array_type])
-
-
-@pytest.mark.parametrize("array_type", arrays.keys())
-def test_call(array_type, processor):
-    input_ids = arrays[array_type]
-    logits = arrays[array_type]
-    processed_logits = processor(input_ids, logits)
-
-    assert isinstance(processed_logits, type(arrays[array_type]))
-    assert np.allclose(
-        np.array(processed_logits), np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
-    )
+    def process_logits(self, input_ids, logits):
+        # check that input_ids and logits received are 2D tensors
+        assert len(self.tensor_adapter.shape(input_ids)) == 2
+        assert len(self.tensor_adapter.shape(logits)) == 2
+        return logits
+
+
+@pytest.mark.parametrize("library", libraries)
+def test_base_logits_processor_init(library):
+    processor = MockLogitsProcessor(library)
+    assert processor.tensor_adapter is not None
+    with pytest.raises(NotImplementedError):
+        processor = MockLogitsProcessor("foo")
+
+
+@pytest.mark.parametrize("library", libraries)
+def test_base_logits_processor_call(library):
+    processor = MockLogitsProcessor(library)
+    input_values = arrays[library]
+    for input_value in input_values:
+        input_ids, logits, expected_error = input_value
+        if expected_error is not None:
+            with pytest.raises(expected_error):
+                processor(input_ids, logits)
+        else:
+            original_shape = processor.tensor_adapter.shape(logits)
+            processed_logits = processor(input_ids, logits)
+            # we check that the shape of logits is preserved
+            assert processor.tensor_adapter.shape(processed_logits) == original_shape
+
+
+@pytest.mark.parametrize("library", libraries)
+def test_base_logits_processor_init_library_name(library):
+    processor = MockLogitsProcessor(library)
+    assert processor.tensor_adapter is not None
+    with pytest.raises(NotImplementedError):
+        processor = MockLogitsProcessor("foo")
diff --git a/tests/fsm/test_cfg_guide.py b/tests/processors/test_cfg_guide.py
similarity index 97%
rename from tests/fsm/test_cfg_guide.py
rename to tests/processors/test_cfg_guide.py
index d92afa625..6091fd4c2 100644
--- a/tests/fsm/test_cfg_guide.py
+++ b/tests/processors/test_cfg_guide.py
@@ -4,8 +4,9 @@
 import pytest
 from transformers import AutoTokenizer
 
-from outlines import grammars, models
-from outlines.fsm.guide import CFGGuide
+from outlines import grammars
+from outlines.models.transformers import TransformerTokenizer
+from outlines.processors.guide import CFGGuide
 
 
 @pytest.fixture
@@ -341,12 +342,12 @@ def decode(self, token_ids):
 
 @pytest.fixture(scope="session")
 def tokenizer_sentencepiece_gpt2():
-    return models.TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
+    return TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
 
 
 @pytest.fixture(scope="session")
 def tokenizer_sentencepiece_llama1():
-    return models.TransformerTokenizer(
+    return TransformerTokenizer(
         AutoTokenizer.from_pretrained(
             "trl-internal-testing/tiny-random-LlamaForCausalLM"
         )
@@ -355,16 +356,14 @@ def tokenizer_sentencepiece_llama1():
 
 @pytest.fixture(scope="session")
 def tokenizer_tiktoken_llama3():
-    return models.TransformerTokenizer(
+    return TransformerTokenizer(
         AutoTokenizer.from_pretrained("yujiepan/llama-3-tiny-random")
     )
 
 
 @pytest.fixture(scope="session")
 def tokenizer_character_level_byt5():
-    return models.TransformerTokenizer(
-        AutoTokenizer.from_pretrained("google/byt5-small")
-    )
+    return TransformerTokenizer(AutoTokenizer.from_pretrained("google/byt5-small"))
 
 
 # Collects all samples within cfg_samples/ and makes adding
diff --git a/tests/fsm/test_guide.py b/tests/processors/test_guide.py
similarity index 85%
rename from tests/fsm/test_guide.py
rename to tests/processors/test_guide.py
index bf25c43c4..0238e37d3 100644
--- a/tests/fsm/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -1,6 +1,23 @@
 import pytest
 
-from outlines.fsm.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
+import llama_cpp
+import transformers
+
+import outlines
+from outlines.processors.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
+from outlines import caching
+
+try:
+    import mlx_lm
+    HAS_MLX = True
+except ImportError:
+    HAS_MLX = False
+
+try:
+    import vllm
+    HAS_VLLM = True
+except ImportError:
+    HAS_VLLM = False
 
 
 def assert_expected_tensor_ids(tensor, ids):
@@ -17,7 +34,7 @@ class MockTokenizer:
 
     fsm = StopAtEOSGuide(MockTokenizer())
 
-    instruction = fsm.get_next_instruction(fsm.start_state)
+    instruction = fsm.get_next_instruction(fsm.initial_state)
     assert isinstance(instruction, Generate)
     assert instruction.tokens is None
 
@@ -25,11 +42,13 @@ class MockTokenizer:
     assert isinstance(instruction, Write)
     assert instruction.tokens == [2]
 
-    assert fsm.get_next_state(fsm.start_state, 2) == fsm.final_state
-    assert fsm.get_next_state(fsm.start_state, 1) == fsm.start_state
-    assert fsm.is_final_state(fsm.start_state) is False
+    assert fsm.get_next_state(fsm.initial_state, 2) == fsm.final_state
+    assert fsm.get_next_state(fsm.initial_state, 1) == fsm.initial_state
+    assert fsm.is_final_state(fsm.initial_state) is False
     assert fsm.is_final_state(fsm.final_state) is True
 
+    assert isinstance(fsm.copy(), StopAtEOSGuide)
+
 
 def test_regex_vocabulary_error():
     class MockTokenizer:
@@ -181,6 +200,46 @@ def convert_token_to_string(self, token):
     assert fsm.is_final_state(state)
 
 
+def test_regex_guide_caching():
+    assert caching._caching_enabled
+
+    cache = caching.get_cache()
+    cache.clear()
+    _, _ = cache.stats(enable=True, reset=True) # (hits, misses)
+
+    regex = r"[0-9]{3}"
+
+    models = [
+        outlines.from_transformers(
+            transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+            transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini")
+        ),
+        outlines.from_llamacpp(llama_cpp.Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        ))
+    ]
+    if HAS_MLX:
+        models.append(outlines.from_mlxlm(
+            *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+        ))
+    if HAS_VLLM:
+        models.append(outlines.from_vllm_offline(vllm.LLM("TinyLlama/TinyLlama-1.1B-Chat-v1.0")))
+
+    for i, model in enumerate(models):
+        # First call for each model should be a miss
+        RegexGuide.from_regex(regex, model.tokenizer)
+        expected_misses = i + 1
+        expected_hits = i
+        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
+
+        # Second call for each model
+        RegexGuide.from_regex(regex, model.tokenizer)
+        expected_misses = i + 1
+        expected_hits = i + 1
+        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
+
+
 def test_cfg():
     class MockTokenizer:
         vocabulary = {"{": 1, "}": 2, "[": 3, "]": 4, "eos": 5}
@@ -238,6 +297,8 @@ def decode(self, token_ids):
     assert guide.must_terminate_state(state)
     assert guide.can_terminate_state(state)
 
+    assert isinstance(guide.copy(), CFGGuide)
+
 
 def test_cfg_early_termination():
     class MockTokenizer:
@@ -465,3 +526,14 @@ def decode(self, token_ids):
     state = guide.get_next_state(state, token_id=4)
     assert guide.must_terminate_state(state)
     assert guide.can_terminate_state(state)
+    assert guide.is_final_state(state)
+
+    cfg_str = """
+        start: CHAR
+        CHAR: /./
+    """
+
+    guide = CFGGuide(cfg_str, tokenizer)
+
+    state = guide.get_next_state(guide.initial_state, token_id=1)
+    assert guide.get_next_state(state, token_id=4).parser_state is None
diff --git a/tests/processors/test_structured.py b/tests/processors/test_structured.py
new file mode 100644
index 000000000..54afae2b2
--- /dev/null
+++ b/tests/processors/test_structured.py
@@ -0,0 +1,135 @@
+import pytest
+from typing import Dict
+
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+import outlines
+from outlines.processors.guide import RegexGuide
+from outlines.processors.structured import (
+    GuideLogitsProcessor,
+    RegexLogitsProcessor,
+    JSONLogitsProcessor,
+    CFGLogitsProcessor,
+)
+
+
+arithmetic_grammar = """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+
+@pytest.fixture
+def tokenizer():
+    model = outlines.from_transformers(
+        AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return model.tokenizer
+
+
+def test_structured_init_guide_logits_processor(tokenizer):
+    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
+    processor = GuideLogitsProcessor(
+        tokenizer=tokenizer,
+        guide=guide,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, GuideLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is guide
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_init_guide_logits_processor_copy(tokenizer):
+    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
+    processor = GuideLogitsProcessor(
+        tokenizer=tokenizer,
+        guide=guide,
+        tensor_library_name="torch",
+    )
+    processor_copy = processor.copy()
+    assert processor_copy is not processor
+    assert processor_copy.tokenizer is tokenizer
+    assert processor_copy.guide is guide
+    assert processor_copy.tensor_adapter.library_name == "torch"
+
+
+def test_structured_guide_logits_processor_call(tokenizer):
+    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
+    processor = GuideLogitsProcessor(
+        tokenizer=tokenizer,
+        guide=guide,
+        tensor_library_name="torch",
+    )
+    vocab_token_ids = list(tokenizer.vocabulary.values())
+    logits = torch.randn(1, len(vocab_token_ids))
+    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
+    output = processor(input_ids, logits)
+    assert output.shape == (1, len(vocab_token_ids))
+
+
+def test_structured_init_json_logits_processor(tokenizer):
+    processor = JSONLogitsProcessor(
+        schema={"type": "object", "properties": {"name": {"type": "string"}}},
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, JSONLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is not None
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_init_regex_logits_processor(tokenizer):
+    processor = RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, RegexLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is not None
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_init_cfg_logits_processor(tokenizer):
+    processor = CFGLogitsProcessor(
+        cfg_str=arithmetic_grammar,
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, CFGLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is not None
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_cfg_logits_processor_call(tokenizer):
+    processor = CFGLogitsProcessor(
+        cfg_str=arithmetic_grammar,
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    vocab_token_ids = list(tokenizer.vocabulary.values())
+    logits = torch.randn(1, len(vocab_token_ids))
+    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
+    output = processor(input_ids, logits)
+    assert output.shape == (1, len(vocab_token_ids))
+    processor(input_ids, logits)
diff --git a/tests/processors/test_tensor_adapters.py b/tests/processors/test_tensor_adapters.py
new file mode 100644
index 000000000..4a5bcc1e1
--- /dev/null
+++ b/tests/processors/test_tensor_adapters.py
@@ -0,0 +1,276 @@
+import pytest
+from pytest import mark
+
+import jax
+import jax.numpy as jnp
+import numpy as np
+import tensorflow as tf
+import torch
+
+from outlines.processors.tensor_adapters import (
+    NumpyTensorAdapter,
+    TorchTensorAdapter,
+    MLXTensorAdapter,
+    JAXTensorAdapter,
+    TensorFlowTensorAdapter,
+)
+
+try:
+    import mlx_lm
+    import mlx.core as mx
+
+    HAS_MLX = mx.metal.is_available()
+except ImportError:
+    HAS_MLX = False
+
+
+adapters = {
+    "jax": JAXTensorAdapter(),
+    "numpy": NumpyTensorAdapter(),
+    "tensorflow": TensorFlowTensorAdapter(),
+    "torch": TorchTensorAdapter(),
+}
+if HAS_MLX:
+    adapters["mlx"] = MLXTensorAdapter()
+
+frameworks = ["jax", "numpy", "tensorflow", "torch", "mlx"]
+
+def create_tensor(framework, shape, dtype=None):
+    if framework == "torch":
+        return torch.randn(*shape)
+    elif framework == "numpy":
+        return np.random.randn(*shape)
+    elif framework == "jax":
+        key = jax.random.PRNGKey(0)
+        return jax.random.normal(key, shape=shape)
+    elif framework == "tensorflow":
+        return tf.random.normal(shape)
+    elif framework == "mlx":
+        if not HAS_MLX:
+            pytest.skip("MLX not available")
+        return mx.random.normal(shape)
+
+def compare_tensors(framework, tensor1, tensor2):
+    if framework == "torch":
+        return torch.allclose(tensor1, tensor2)
+    elif framework == "numpy":
+        return np.array_equal(tensor1, tensor2)
+    elif framework == "jax":
+        return jax.numpy.array_equal(tensor1, tensor2)
+    elif framework == "tensorflow":
+        return tf.reduce_all(tf.equal(tensor1, tensor2))
+    elif framework == "mlx":
+        if not HAS_MLX:
+            pytest.skip("MLX not available")
+        return mx.array_equal(tensor1, tensor2)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_shape(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (2,))
+    result_1d = adapters[framework].shape(tensor_1d)
+    assert len(result_1d) == 1
+    assert result_1d[0] == 2
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (2, 3))
+    result_2d = adapters[framework].shape(tensor_2d)
+    assert len(result_2d) == 2
+    assert result_2d[0] == 2
+    assert result_2d[1] == 3
+
+    # 3d tensor
+    tensor_3d = create_tensor(framework, (2, 2, 3))
+    result_3d = adapters[framework].shape(tensor_3d)
+    assert len(result_3d) == 3
+    assert result_3d[0] == 2
+    assert result_3d[1] == 2
+    assert result_3d[2] == 3
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_unsqueeze(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (2,))
+    result_1d = adapters[framework].unsqueeze(tensor_1d)
+    assert result_1d.shape == (1, 2)
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (2, 3))
+    result_2d = adapters[framework].unsqueeze(tensor_2d)
+    assert result_2d.shape == (1, 2, 3)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_squeeze(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (1,))
+    result_1d = adapters[framework].squeeze(tensor_1d)
+    with pytest.raises(TypeError):
+        len(result_1d)
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (1, 2))
+    result_2d = adapters[framework].squeeze(tensor_2d)
+    assert result_2d.shape == (2,)
+
+    # 3d tensor
+    tensor_3d = create_tensor(framework, (1, 2, 3))
+    result_3d = adapters[framework].squeeze(tensor_3d)
+    assert result_3d.shape == (2, 3)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_to_list(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (2,))
+    result_1d = adapters[framework].to_list(tensor_1d)
+    assert isinstance(result_1d, list)
+    assert len(result_1d) == 2
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (2, 3))
+    result_2d = adapters[framework].to_list(tensor_2d)
+    assert isinstance(result_2d, list)
+    assert len(result_2d) == 2
+    assert len(result_2d[0]) == 3
+    assert len(result_2d[1]) == 3
+
+    # 3d tensor
+    tensor_3d = create_tensor(framework, (2, 2, 3))
+    result_3d = adapters[framework].to_list(tensor_3d)
+    assert isinstance(result_3d, list)
+    assert len(result_3d) == 2
+    assert len(result_3d[0]) == 2
+    assert len(result_3d[1]) == 2
+    assert len(result_3d[0][0]) == 3
+    assert len(result_3d[0][1]) == 3
+    assert len(result_3d[1][0]) == 3
+    assert len(result_3d[1][1]) == 3
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_to_scalar(framework):
+    # multi-elements tensor, should raise an error
+    tensor_multi = create_tensor(framework, (2, 3))
+    if framework == "torch":
+        with pytest.raises(RuntimeError):
+            adapters[framework].to_scalar(tensor_multi)
+    else:
+        with pytest.raises(ValueError):
+            adapters[framework].to_scalar(tensor_multi)
+
+    # single-element tensor
+    tensor_single = create_tensor(framework, (1, 1))
+    scalar = adapters[framework].to_scalar(tensor_single)
+    assert isinstance(scalar, float)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_full_like(framework):
+    tensor = create_tensor(framework, (2, 3))
+    result = adapters[framework].full_like(tensor, 0)
+    assert result.shape == (2, 3)
+    for i in range(2):
+        for j in range(3):
+            assert result[i, j] == 0
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_concatenate(framework):
+    # 1d tensors
+    tensor1 = create_tensor(framework, (2,))
+    tensor2 = create_tensor(framework, (2,))
+    result = adapters[framework].concatenate([tensor1, tensor2])
+    assert result.shape == (4,)
+    assert result[0] == tensor1[0]
+    assert result[1] == tensor1[1]
+    assert result[2] == tensor2[0]
+    assert result[3] == tensor2[1]
+
+    # 2d tensors
+    tensor1 = create_tensor(framework, (2, 3))
+    tensor2 = create_tensor(framework, (2, 3))
+    result = adapters[framework].concatenate([tensor1, tensor2])
+    assert result.shape == (4, 3)
+    for i in range(2):
+        for j in range(3):
+            assert result[i, j] == tensor1[i, j]
+            assert result[i + 2, j] == tensor2[i, j]
+
+    # 3d tensors
+    tensor1 = create_tensor(framework, (2, 2, 3))
+    tensor2 = create_tensor(framework, (2, 2, 3))
+    result = adapters[framework].concatenate([tensor1, tensor2])
+    assert result.shape == (4, 2, 3)
+    for i in range(2):
+        for j in range(2):
+            for k in range(3):
+                assert result[i, j, k] == tensor1[i, j, k]
+                assert result[i + 2, j, k] == tensor2[i, j, k]
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_get_to_device(framework):
+    tensor = create_tensor(framework, (2, 3))
+    device = adapters[framework].get_device(tensor)
+    device_tensor = adapters[framework].to_device(tensor, device)
+
+    if framework == "torch":
+        assert isinstance(device_tensor.device.type, str)
+        assert compare_tensors(framework, device_tensor, tensor)
+    else:
+        assert compare_tensors(framework, device_tensor, tensor)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_boolean_ones_like(framework):
+    tensor = create_tensor(framework, (2, 3))
+    ones = adapters[framework].boolean_ones_like(tensor)
+
+    assert ones.shape == (2, 3)
+    for i in range(2):
+        for j in range(3):
+            assert ones[i, j]
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_apply_mask(framework):
+    tensor = create_tensor(framework, (2, 3))
+
+    if framework == "torch":
+        mask = torch.randn(2, 3) > 0
+    elif framework == "numpy":
+        mask = np.random.randn(2, 3) > 0
+    elif framework == "jax":
+        key = jax.random.PRNGKey(0)
+        mask = jax.random.normal(key, shape=(2, 3)) > 0
+    elif framework == "tensorflow":
+        mask = tf.random.normal((2, 3)) > 0
+    elif framework == "mlx":
+        if not HAS_MLX:
+            pytest.skip("MLX not available")
+        mask = mx.random.normal((2, 3)) > 0
+
+    masked = adapters[framework].apply_mask(tensor, mask, float("-inf"))
+
+    assert masked.shape == (2, 3)
+    for i in range(2):
+        for j in range(3):
+            if mask[i, j]:
+                assert masked[i, j] == float("-inf")
+            else:
+                assert masked[i, j] == tensor[i, j]
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_argsort_descending(framework):
+    tensor = create_tensor(framework, (2, 3))
+    indices = adapters[framework].argsort_descending(tensor)
+
+    assert indices.shape == (2, 3)
+    for i in range(2):
+        sorted_values = [tensor[i][idx] for idx in indices[i]]
+        for j in range(len(sorted_values) - 1):
+            assert sorted_values[j] >= sorted_values[j + 1]
diff --git a/tests/test_applications.py b/tests/test_applications.py
new file mode 100644
index 000000000..5fed7bf23
--- /dev/null
+++ b/tests/test_applications.py
@@ -0,0 +1,93 @@
+from typing import Any
+
+import jinja2
+import pytest
+import transformers
+
+from outlines import from_transformers
+from outlines.applications import Application
+from outlines.templates import Template
+
+
+@pytest.fixture(scope="session")
+def model():
+    return from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("gpt2"),
+        transformers.AutoTokenizer.from_pretrained("gpt2"),
+    )
+
+
+@pytest.fixture(scope="session")
+def another_model():
+    return from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("gpt2"),
+        transformers.AutoTokenizer.from_pretrained("gpt2"),
+    )
+
+
+def test_application_initialization():
+    template = Template.from_string("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    assert application.template == template
+    assert application.output_type == output_type
+    assert application.model is None
+    assert application.generator is None
+
+
+def test_application_generator_no_model():
+    template = Template.from_string("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    with pytest.raises(ValueError):
+        application(None, {"value": "example"})
+
+
+def test_application_template_call(model):
+    template = Template.from_string("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+    result = application(model, {"value": "example"}, max_new_tokens=10)
+
+    assert isinstance(result, str)
+
+
+def test_application_callable_call(model):
+    def template(value):
+        return f"Test {value}"
+
+    output_type = None
+    application = Application(template, output_type)
+    result = application(model, {"value": "example"}, max_new_tokens=10)
+
+    assert isinstance(result, str)
+
+
+def test_application_template_error(model):
+    template = Template.from_string("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    with pytest.raises(jinja2.exceptions.UndefinedError):
+        application(model, {"foo": "bar"})
+
+
+def test_application_generator_reuse(model, another_model):
+    template = Template.from_string("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    application(model, {"value": "example"}, max_new_tokens=10)
+    first_generator = application.generator
+    first_model = application.model
+
+    application(model, {"value": "example"}, max_new_tokens=10)
+    assert application.model == first_model
+    assert application.generator == first_generator
+
+    application(another_model, {"value": "example"}, max_new_tokens=10)
+    assert application.model == another_model
+    assert application.model != first_model
+    assert application.generator != first_generator
diff --git a/tests/test_cache.py b/tests/test_cache.py
index eb4ec406e..149d856c0 100644
--- a/tests/test_cache.py
+++ b/tests/test_cache.py
@@ -1,9 +1,19 @@
 import os
 import tempfile
 import unittest
+from importlib import reload
 
 import diskcache
 import pytest
+from diskcache import Cache, UNKNOWN
+from outlines.caching import CloudpickleDisk
+
+
+@pytest.fixture
+def temp_dir():
+    """Create a temporary directory for testing."""
+    directory = tempfile.mkdtemp()
+    yield directory
 
 
 @pytest.fixture
@@ -188,3 +198,24 @@ def fn():
     # scope has exited, cache is enabled again
     fn()
     assert mock.call_count == 2
+
+
+@pytest.fixture
+def temp_cache_dir():
+    import os
+    import tempfile
+
+    import outlines.caching
+    import outlines.processors.guide
+
+    with tempfile.TemporaryDirectory() as tempdir:
+        os.environ["OUTLINES_CACHE_DIR"] = tempdir
+        outlines.caching.get_cache.cache_clear()
+        reload(outlines)
+        reload(outlines.processors.guide)
+        cache_status = outlines.caching._caching_enabled
+        try:
+            outlines.caching._caching_enabled = True
+            yield
+        finally:
+            outlines.caching._caching_enabled = cache_status
diff --git a/tests/test_generator.py b/tests/test_generator.py
new file mode 100644
index 000000000..5ce54dc4b
--- /dev/null
+++ b/tests/test_generator.py
@@ -0,0 +1,255 @@
+import pytest
+from typing import AsyncGenerator, Generator as TypingGenerator, Literal
+
+import interegular
+import transformers
+
+import outlines
+from outlines.generator import (
+    BlackBoxGenerator,
+    SteerableGenerator,
+    Generator,
+    AsyncBlackBoxGenerator,
+)
+from outlines.models import AsyncVLLM, VLLM
+from outlines.processors import (
+    OutlinesLogitsProcessor,
+    RegexLogitsProcessor,
+)
+from outlines.types import CFG, FSM
+from tests.test_utils.mock_openai_client import (
+    MockAsyncOpenAIClient,
+    MockOpenAIClient,
+)
+
+
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+
+
+# We used the mocked vllm model to test the black box generator
+async_openai_client = MockAsyncOpenAIClient()
+openai_client = MockOpenAIClient()
+mock_responses = [
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Write a very short sentence'}
+            ],
+            'model': MODEL_NAME,
+            'max_tokens': 10,
+            'extra_body': {'guided_regex': '("[^"]*")'},
+        },
+        "Mock response"
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Write a very short sentence'}
+            ],
+            'model': MODEL_NAME,
+            'max_tokens': 10,
+            'extra_body': {'guided_regex': '("[^"]*")'},
+            'stream': True,
+        },
+        ["Mock", "response"]
+    ),
+]
+async_openai_client.add_mock_responses(mock_responses)
+openai_client.add_mock_responses(mock_responses)
+
+
+@pytest.fixture(scope="session")
+def steerable_model():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return model
+
+
+@pytest.fixture(scope="session")
+def sample_processor():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    processor = RegexLogitsProcessor(
+        regex_string="[0-9]{3}",
+        tokenizer=model.tokenizer,
+        tensor_library_name=model.tensor_library_name,
+    )
+    return processor
+
+
+@pytest.fixture(scope="module")
+def black_box_sync_model():
+    return VLLM(openai_client, MODEL_NAME)
+
+
+@pytest.fixture(scope="module")
+def black_box_async_model():
+    return AsyncVLLM(async_openai_client, MODEL_NAME)
+
+
+# SteerableGenerator
+
+
+def test_steerable_generator_init_valid_processor(steerable_model, sample_processor):
+    generator = SteerableGenerator.from_processor(steerable_model, sample_processor)
+    assert generator.logits_processor == sample_processor
+    assert generator.model == steerable_model
+
+
+def test_steerable_generator_init_invalid_processor(steerable_model):
+    with pytest.raises(TypeError):
+        SteerableGenerator.from_processor(steerable_model, Literal["foo", "bar"])
+
+
+def test_steerable_generator_init_cfg_output_type(steerable_model):
+    generator = SteerableGenerator(steerable_model, CFG('start: "a"'))
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_steerable_generator_init_fsm_output_type(steerable_model):
+    generator = SteerableGenerator(
+        steerable_model,
+        FSM(interegular.parse_pattern(r"abc").to_fsm())
+    )
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_steerable_generator_init_other_output_type(steerable_model):
+    generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_steerable_generator_init_invalid_output_type(steerable_model, sample_processor):
+    with pytest.raises(ValueError):
+        SteerableGenerator(steerable_model, sample_processor)
+
+
+def test_steerable_generator_call(steerable_model):
+    generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
+    result = generator("foo", max_new_tokens=10)
+    assert isinstance(result, str)
+
+
+def test_steerable_generator_stream(steerable_model):
+    with pytest.raises(NotImplementedError):
+        generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
+        result = generator.stream("foo", max_tokens=10)
+        assert isinstance(result, TypingGenerator)
+        assert isinstance(next(result), str)
+
+
+# BlackBoxGenerator
+
+
+def test_black_box_generator_init_valid(black_box_sync_model):
+    generator = BlackBoxGenerator(black_box_sync_model, Literal["foo", "bar"])
+    assert generator.model == black_box_sync_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_black_box_generator_init_invalid(black_box_sync_model):
+    with pytest.raises(NotImplementedError):
+        BlackBoxGenerator(black_box_sync_model, FSM("foo"))
+
+
+def test_black_box_generator_call(black_box_sync_model):
+    generator = BlackBoxGenerator(black_box_sync_model, str)
+    result = generator("Write a very short sentence", max_tokens=10)
+    assert isinstance(result, str)
+
+
+def test_black_box_generator_stream(black_box_sync_model):
+    generator = BlackBoxGenerator(black_box_sync_model, str)
+    result = generator.stream("Write a very short sentence", max_tokens=10)
+    assert isinstance(result, TypingGenerator)
+    assert isinstance(next(result), str)
+
+
+
+# AsyncBlackBoxGenerator
+
+
+def test_async_black_box_generator_init_valid(black_box_async_model):
+    generator = AsyncBlackBoxGenerator(black_box_async_model, Literal["foo", "bar"])
+    assert generator.model == black_box_async_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_async_black_box_generator_init_invalid(black_box_async_model):
+    with pytest.raises(NotImplementedError):
+        AsyncBlackBoxGenerator(black_box_async_model, FSM("foo"))
+
+
+@pytest.mark.asyncio
+async def test_async_black_box_generator_call(black_box_async_model):
+    generator = AsyncBlackBoxGenerator(black_box_async_model, str)
+    result = await generator("Write a very short sentence", max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_async_black_box_generator_stream(black_box_async_model):
+    generator = AsyncBlackBoxGenerator(black_box_async_model, str)
+    result = generator.stream("Write a very short sentence", max_tokens=10)
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+# Generator
+
+
+def test_generator_init_no_model():
+    with pytest.raises(ValueError):
+        Generator(None, Literal["foo", "bar"])
+
+
+def test_generator_init_multiple_output_type(steerable_model, sample_processor):
+    with pytest.raises(ValueError):
+        Generator(steerable_model, Literal["foo", "bar"], processor=sample_processor)
+
+
+def test_generator_steerable_output_type(steerable_model):
+    generator = Generator(steerable_model, Literal["foo", "bar"])
+    assert isinstance(generator, SteerableGenerator)
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_generator_steerable_processor(steerable_model, sample_processor):
+    generator = Generator(steerable_model, processor=sample_processor)
+    assert isinstance(generator, SteerableGenerator)
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_generator_black_box_sync_output_type(black_box_sync_model):
+    generator = Generator(black_box_sync_model, Literal["foo", "bar"])
+    assert isinstance(generator, BlackBoxGenerator)
+    assert generator.model == black_box_sync_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_generator_black_box_sync_processor(black_box_sync_model, sample_processor):
+    with pytest.raises(NotImplementedError):
+        Generator(black_box_sync_model, processor=sample_processor)
+
+
+def test_generator_black_box_async_output_type(black_box_async_model):
+    generator = Generator(black_box_async_model, Literal["foo", "bar"])
+    assert isinstance(generator, AsyncBlackBoxGenerator)
+    assert generator.model == black_box_async_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_generator_black_box_async_processor(black_box_async_model, sample_processor):
+    with pytest.raises(NotImplementedError):
+        Generator(black_box_async_model, processor=sample_processor)
diff --git a/tests/test_grammars.py b/tests/test_grammars.py
index 3d704f094..69486684b 100644
--- a/tests/test_grammars.py
+++ b/tests/test_grammars.py
@@ -1,7 +1,7 @@
 import pytest
 
 import outlines.grammars as grammars
-from outlines.fsm.guide import CFGGuide
+from outlines.processors.guide import CFGGuide
 
 
 @pytest.mark.parametrize("grammar", [grammars.json, grammars.arithmetic])
diff --git a/tests/test_samplers.py b/tests/test_samplers.py
deleted file mode 100644
index 10a7be26f..000000000
--- a/tests/test_samplers.py
+++ /dev/null
@@ -1,277 +0,0 @@
-import math
-
-import pytest
-import torch
-
-from outlines.samplers import (
-    BeamSearchSampler,
-    GreedySampler,
-    MultinomialSampler,
-    beam_search,
-    greedy,
-    keep_top_k_logits,
-    keep_top_p_logits,
-    multinomial,
-    rescale_logits,
-)
-
-
-def compute_logprobs(logits):
-    return torch.nn.functional.log_softmax(logits, dim=-1)
-
-
-def test_aliases():
-    assert greedy == GreedySampler
-    assert multinomial == MultinomialSampler
-    assert beam_search == BeamSearchSampler
-
-
-def test_greedy():
-    sampler = GreedySampler()
-    logits = torch.tensor([[1.0, 2.0, 5.0]])
-    weights = torch.tensor([0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[2]]))
-    assert ancestors.equal(torch.tensor([0]))
-    assert weights.equal(logprobs[..., 2])
-
-    sampler = GreedySampler()
-    logits = torch.tensor([[10.0, 0.0, 3.0], [-math.inf, 2.0, 5.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [2]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
-
-    params = sampler.sampling_params
-    assert params.sampler == "greedy"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 0.0
-
-
-def test_multinomial():
-    rng = torch.Generator()
-    rng.manual_seed(239)
-
-    sampler = MultinomialSampler()
-    logits = torch.tensor([[1.0, 4.0, 5.0]])
-    weights = torch.tensor([0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[2]]))
-    assert ancestors.equal(torch.tensor([0]))
-    assert weights.equal(torch.tensor([logprobs[:, 2]]))
-
-    sampler = MultinomialSampler()
-    logits = torch.tensor([[10.0, 0.0, 9.0], [-math.inf, 4.0, 5.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [2]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
-
-    sampler = MultinomialSampler(samples=5, top_k=10, top_p=0.9, temperature=0.8)
-    params = sampler.sampling_params
-    assert params.sampler == "multinomial"
-    assert params.num_samples == 5
-    assert params.top_p == 0.9
-    assert params.top_k == 10
-    assert params.temperature == 0.8
-
-
-def test_multinomial_init():
-    sampler = MultinomialSampler()
-    assert sampler.logits_processors == []
-
-    sampler = MultinomialSampler(3)
-    assert sampler.logits_processors == []
-
-    sampler = MultinomialSampler(top_k=1)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_p=0.95)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_k=1, top_p=0.95)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(temperature=1.0)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_k=1, temperature=1.0)
-    assert len(sampler.logits_processors) == 2
-
-    sampler = MultinomialSampler(top_p=0.95, temperature=1.0)
-    assert len(sampler.logits_processors) == 2
-
-
-def test_top_k():
-    with pytest.raises(ValueError, match="`k` must be a strictly"):
-        keep_top_k_logits(-1)
-
-    with pytest.raises(ValueError, match="`k` must be a strictly"):
-        keep_top_k_logits(0.1)
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0]])
-
-    logits_processor = keep_top_k_logits(1)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
-
-    logits_processor = keep_top_k_logits(10)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[1.0, 2.0, 3.0, 4.0]]))
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
-    logits_processor = keep_top_k_logits(2)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [[-math.inf, -math.inf, 3.0, 4.0], [-math.inf, -math.inf, 7.0, 8.0]]
-        )
-    )
-
-
-def test_top_p():
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(-0.1)
-
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(0.0)
-
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(1.1)
-
-    logits = torch.tensor([[1.0, 1.01, 1.02, 4.0]])
-
-    logits_processor = keep_top_p_logits(0.1)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
-
-    logits_processor = keep_top_p_logits(0.95)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, 1.01, 1.02, 4.0]]))
-
-    logits_processor = keep_top_p_logits(1.0)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[1.0, 1.01, 1.02, 4.0]]))
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
-
-    logits_processor = keep_top_p_logits(0.1)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [-math.inf, -math.inf, -math.inf, 4.0],
-                [-math.inf, -math.inf, -math.inf, 8.0],
-            ]
-        )
-    )
-
-    logits_processor = keep_top_p_logits(0.95)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [-math.inf, 2.0, 3.0, 4.0],
-                [-math.inf, 6.0, 7.0, 8.0],
-            ]
-        )
-    )
-
-    logits_processor = keep_top_p_logits(1.0)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [1.0, 2.0, 3.0, 4.0],
-                [5.0, 6.0, 7.0, 8.0],
-            ]
-        )
-    )
-
-
-def test_rescale():
-    with pytest.raises(ValueError, match="`temperature` must"):
-        rescale_logits(1)
-
-    with pytest.raises(ValueError, match="`temperature` must"):
-        rescale_logits(-0.1)
-
-    with pytest.raises(ValueError, match="Please use the greedy sampler"):
-        rescale_logits(0.0)
-
-
-def test_beam_search():
-    # Two beams, single sequence
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
-    init_weights = torch.tensor([0, 1.0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [1]]))
-    assert ancestors.equal(torch.tensor([1, 0]))
-    assert weights.equal(
-        torch.tensor([init_weights[1] + logprobs[1][0], logprobs[0][1]])
-    )
-
-    # Make sure that initial samples are different
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [0.0, 1.0]])
-    init_weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[1], [0]]))
-    assert ancestors.equal(torch.tensor([0, 0]))
-    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[0][0]]))
-
-    # One beam, batch of two sequences. Reduces to Greedy Search.
-    sampler = BeamSearchSampler(1)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[1], [0]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[1][0]]))
-
-    # Two beams, batch of two sequences
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0], [3.0, 2.0], [0.0, 1.0]])
-    init_weights = torch.tensor([0, 0, 2.0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [1], [0], [1]]))
-    assert ancestors.equal(torch.tensor([1, 0, 2, 2]))
-    assert weights.equal(
-        torch.tensor(
-            [
-                logprobs[1][0],
-                logprobs[0][1],
-                init_weights[2] + logprobs[2][0],
-                init_weights[2] + logprobs[2][1],
-            ]
-        )
-    )
-
-    sampler = BeamSearchSampler(beams=3)
-    params = sampler.sampling_params
-    assert params.sampler == "beam_search"
-    assert params.num_samples == 3
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 1.0
diff --git a/tests/test_templates.py b/tests/test_templates.py
index 2fda3caa3..cf9d0ed5f 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -1,12 +1,75 @@
+import base64
+import json
 import os
 import tempfile
 from typing import Dict, List
 
 import pytest
+from PIL import Image
+from io import BytesIO
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.templates import Template, build_template_from_string
+from outlines.templates import (
+    Template,
+    build_template_from_string,
+    Vision,
+    get_fn_name,
+    get_fn_args,
+    get_fn_description,
+    get_fn_source,
+    get_fn_signature,
+    get_schema,
+)
+
+
+def sample_function(x, y=2):
+    """This is a sample function."""
+    return x + y
+
+def function_with_annotations(x: int, y: str) -> str:
+    """Function with annotations."""
+    return f"{x} {y}"
+
+def function_with_no_docstring(x, y):
+    return x * y
+
+class CallableClass:
+    def __call__(self):
+        pass
+
+class PydanticClass(BaseModel):
+    foo: str
+
+
+def test_vision_initialization():
+    # Create a simple image for testing
+    image = Image.new("RGB", (10, 10), color="red")
+    image.format = "PNG"
+
+    # Initialize the Vision object
+    vision = Vision(prompt="Test prompt", image=image)
+
+    # Check that the prompt is set correctly
+    assert vision.prompt == "Test prompt"
+
+    # Check that the image is encoded correctly
+    buffer = BytesIO()
+    image.save(buffer, format=image.format)
+    expected_image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
+    assert vision.image_str == expected_image_str
+
+    # Check that the image format is set correctly
+    assert vision.image_format == "image/png"
+
+
+def test_vision_invalid_image_format():
+    # Create an image without a format
+    image = Image.new("RGB", (10, 10), color="blue")
+
+    # Expect a TypeError when the image format is not set
+    with pytest.raises(TypeError, match="Could not read the format"):
+        Vision(prompt="Test prompt", image=image)
 
 
 def render(content: str, **kwargs):
@@ -441,3 +504,60 @@ def test_template_from_str_with_extra_linebreaks():
     """
     template = build_template_from_string(content)
     assert template.render(name="World") == "Hello, World!\n"
+
+
+def test_get_fn_name():
+    with pytest.raises(TypeError):
+        get_fn_name(1)
+    assert get_fn_name(sample_function) == "sample_function"
+    assert get_fn_name(function_with_annotations) == "function_with_annotations"
+    no_name_func = lambda x: x
+    assert get_fn_name(no_name_func) == "<lambda>"
+    assert get_fn_name(CallableClass()) == "CallableClass"
+
+
+def test_get_fn_args():
+    with pytest.raises(TypeError):
+        get_fn_args(1)
+    assert get_fn_args(sample_function) == "x, y=2"
+    assert get_fn_args(function_with_annotations) == "x: int, y: str"
+
+
+def test_get_fn_description():
+    with pytest.raises(TypeError):
+        get_fn_description(1)
+    assert get_fn_description(sample_function) == "This is a sample function."
+    assert get_fn_description(function_with_annotations) == "Function with annotations."
+    assert get_fn_description(function_with_no_docstring) == ""
+
+
+def test_get_fn_source():
+    with pytest.raises(TypeError, match="The `source` filter only applies to callables."):
+        get_fn_source(1)
+    source = (
+        'def sample_function(x, y=2):\n'
+        '    """This is a sample function."""\n'
+        '    return x + y'
+    )
+    assert get_fn_source(sample_function).strip() == source
+
+
+def test_get_fn_signature():
+    with pytest.raises(TypeError, match="The `source` filter only applies to callables."):
+        get_fn_signature(1)
+    sample_function_signature = "x, y=2"
+    assert get_fn_signature(sample_function) == sample_function_signature
+    function_with_annotations_signature = "x: int, y: str"
+    assert get_fn_signature(function_with_annotations) == function_with_annotations_signature
+
+
+def test_get_schema():
+    with pytest.raises(NotImplementedError):
+        get_schema(1)
+
+    dict_schema = {"foo": "bar"}
+    dict_schema_output = get_schema(dict_schema)
+    assert dict_schema_output == '{\n  "foo": "bar"\n}'
+
+    pydantic_schema_output = get_schema(PydanticClass)
+    assert pydantic_schema_output == '{\n  "foo": "<foo>"\n}'
diff --git a/tests/test_utils/mock_openai_client.py b/tests/test_utils/mock_openai_client.py
new file mode 100644
index 000000000..5f466e30f
--- /dev/null
+++ b/tests/test_utils/mock_openai_client.py
@@ -0,0 +1,117 @@
+from typing import List, Dict, Any, Optional
+from unittest.mock import MagicMock
+
+from tests.test_utils.utils import hash_dict
+
+
+class MockChoice:
+    def __init__(
+        self,
+        content: str,
+        finish_reason: str = "stop",
+        refusal: Optional[str] = None
+    ):
+        self.message = MagicMock()
+        self.message.content = content
+        self.message.refusal = refusal
+        self.finish_reason = finish_reason
+        self.delta = MagicMock()
+        self.delta.content = content
+
+
+class MockCompletionResponse:
+    def __init__(self, choices: List[MockChoice]):
+        self.choices = choices
+
+
+class MockStreamingChunk:
+    def __init__(self, content: Optional[str] = None):
+        self.choices = []
+        if content is not None:
+            choice = MagicMock()
+            delta = MagicMock()
+            delta.content = content
+            choice.delta = delta
+            self.choices = [choice]
+
+
+class MockOpenAIClient:
+    """Mock for OpenAI client that can be used to test vLLM integration"""
+
+    def __init__(self):
+        self.chat = MagicMock()
+        self.chat.completions = MagicMock()
+        self.chat.completions.create = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        def _create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return self._create_streaming_response(response)
+            else:
+                return self._create_standard_response(response)
+
+        self.chat.completions.create.side_effect = _create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+    def _create_standard_response(self, response):
+        if isinstance(response, str):
+            response = [response]
+        choices = [MockChoice(content=chunk) for chunk in response]
+        return MockCompletionResponse(choices=choices)
+
+    def _create_streaming_response(self, response):
+        chunks = [MockStreamingChunk(content=chunk) for chunk in response]
+        return iter(chunks)
+
+
+class MockAsyncOpenAIClient:
+    """Mock for AsyncOpenAI client that can be used to test AsyncVLLM integration"""
+
+    def __init__(self):
+        self.chat = MagicMock()
+        self.chat.completions = MagicMock()
+        self.chat.completions.create = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        async def _async_create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return self._create_async_streaming_response(response)
+            else:
+                return await self._create_async_standard_response(response)
+
+        self.chat.completions.create.side_effect = _async_create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+    async def _create_async_standard_response(self, response):
+        """Create an async standard (non-streaming) response"""
+        if isinstance(response, str):
+            response = [response]
+        choices = [MockChoice(content=chunk) for chunk in response]
+        return MockCompletionResponse(choices=choices)
+
+    async def _create_async_streaming_response(self, response):
+        """Create an async streaming response generator"""
+        chunks = [MockStreamingChunk(content=chunk) for chunk in response]
+
+        for chunk in chunks:
+            yield chunk
diff --git a/tests/test_utils/mock_tgi_client.py b/tests/test_utils/mock_tgi_client.py
new file mode 100644
index 000000000..fef054682
--- /dev/null
+++ b/tests/test_utils/mock_tgi_client.py
@@ -0,0 +1,63 @@
+from typing import Any, Dict
+from unittest.mock import MagicMock
+
+from tests.test_utils.utils import hash_dict
+
+
+class MockTGIInferenceClient:
+    """Mock for TGI `InferenceClient` that can be used to test the TGI model"""
+
+    def __init__(self):
+        self.text_generation = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        def _create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return iter(response)
+            else:
+                return response
+
+        self.text_generation.side_effect = _create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+
+class MockAsyncTGIInferenceClient:
+    """Mock for TGI `InferenceClient` that can be used to test the TGI model"""
+
+    def __init__(self):
+        self.text_generation = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        async def _async_create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return self._create_async_streaming_response(response)
+            else:
+                return response
+
+        self.text_generation.side_effect = _async_create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+    async def _create_async_streaming_response(self, response):
+        """Create an async streaming response generator"""
+        for chunk in response:
+            yield chunk
diff --git a/tests/test_utils/utils.py b/tests/test_utils/utils.py
new file mode 100644
index 000000000..d3a3fa969
--- /dev/null
+++ b/tests/test_utils/utils.py
@@ -0,0 +1,23 @@
+import hashlib
+import pickle
+import sys
+
+
+def hash_dict(d) -> str:
+    def make_hashable(obj):
+        if isinstance(obj, (bool, int, float, str, type(None))):
+            if isinstance(obj, str):
+                return sys.intern(obj)
+            return obj
+        if isinstance(obj, dict):
+            return tuple(sorted(
+                (sys.intern(k) if isinstance(k, str) else k, make_hashable(v))
+                for k, v in obj.items()
+            ))
+        if isinstance(obj, (list, tuple)):
+            return tuple(make_hashable(e) for e in obj)
+        return str(obj)
+
+    hashable_obj = make_hashable(d)
+    pickled_obj = pickle.dumps(hashable_obj, protocol=4)
+    return hashlib.sha256(pickled_obj).hexdigest()
diff --git a/tests/types/test_custom_types.py b/tests/types/test_custom_types.py
index 46df9ba4a..691b681b8 100644
--- a/tests/types/test_custom_types.py
+++ b/tests/types/test_custom_types.py
@@ -4,7 +4,6 @@
 from pydantic import BaseModel
 
 from outlines import types
-from outlines.fsm.types import python_types_to_regex
 from outlines.types.dsl import to_regex
 
 
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index 8351ef319..4e13b1d38 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -1,6 +1,25 @@
+import datetime
+import json
+import sys
+import tempfile
+from dataclasses import dataclass
+from enum import Enum
+from typing import (
+    Dict,
+    List,
+    Literal,
+    Tuple,
+    Union,
+    get_args,
+    Optional as PyOptional
+)
+
+import interegular
 import pytest
+from genson import SchemaBuilder
 from pydantic import BaseModel
 
+from outlines import grammars, types
 from outlines.types.dsl import (
     Alternatives,
     JsonSchema,
@@ -16,6 +35,14 @@
     String,
     Term,
     either,
+    CFG,
+    FSM,
+    _handle_dict,
+    _handle_list,
+    _handle_literal,
+    _handle_tuple,
+    _handle_union,
+    json_schema,
     one_or_more,
     zero_or_more,
     optional,
@@ -24,34 +51,48 @@
     at_least,
     exactly,
     regex,
-    json_schema,
+    python_types_to_terms,
+    # deprecated
+    repeat,
+    times,
 )
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
 
 def test_dsl_init():
     string = String("test")
     assert string.value == "test"
     assert repr(string) == "String(value='test')"
+    assert string.display_ascii_tree() == "└── String('test')\n"
 
     regex = Regex("[0-9]")
     assert regex.pattern == "[0-9]"
     assert repr(regex) == "Regex(pattern='[0-9]')"
+    assert regex.display_ascii_tree() == "└── Regex('[0-9]')\n"
 
     schema = JsonSchema('{ "type": "string" }')
     assert schema.schema == '{ "type": "string" }'
     assert repr(schema) == 'JsonSchema(schema=\'{ "type": "string" }\')'
+    assert schema.display_ascii_tree() == "└── JsonSchema('{ \"type\": \"string\" }')\n"
 
     kleene_star = KleeneStar(string)
     assert kleene_star.term == string
     assert repr(kleene_star) == "KleeneStar(term=String(value='test'))"
+    assert kleene_star.display_ascii_tree() == "└── KleeneStar(*)\n    └── String('test')\n"
 
     kleene_plus = KleenePlus(string)
     assert kleene_plus.term == string
     assert repr(kleene_plus) == "KleenePlus(term=String(value='test'))"
+    assert kleene_plus.display_ascii_tree() == "└── KleenePlus(+)\n    └── String('test')\n"
 
     optional = Optional(string)
     assert optional.term == string
     assert repr(optional) == "Optional(term=String(value='test'))"
+    assert optional.display_ascii_tree() == "└── Optional(?)\n    └── String('test')\n"
 
     alternatives = Alternatives([string, regex])
     assert alternatives.terms[0] == string
@@ -60,6 +101,7 @@ def test_dsl_init():
         repr(alternatives)
         == "Alternatives(terms=[String(value='test'), Regex(pattern='[0-9]')])"
     )
+    assert alternatives.display_ascii_tree() == "└── Alternatives(|)\n    ├── String('test')\n    └── Regex('[0-9]')\n"
 
     sequence = Sequence([string, regex])
     assert sequence.terms[0] == string
@@ -68,21 +110,25 @@ def test_dsl_init():
         repr(sequence)
         == "Sequence(terms=[String(value='test'), Regex(pattern='[0-9]')])"
     )
+    assert sequence.display_ascii_tree() == "└── Sequence\n    ├── String('test')\n    └── Regex('[0-9]')\n"
 
     exact = QuantifyExact(string, 3)
     assert exact.term == string
     assert exact.count == 3
     assert repr(exact) == "QuantifyExact(term=String(value='test'), count=3)"
+    assert exact.display_ascii_tree() == "└── Quantify({3})\n    └── String('test')\n"
 
     minimum = QuantifyMinimum(string, 3)
     assert minimum.term == string
     assert minimum.min_count == 3
     assert repr(minimum) == "QuantifyMinimum(term=String(value='test'), min_count=3)"
+    assert minimum.display_ascii_tree() == "└── Quantify({3,})\n    └── String('test')\n"
 
     maximum = QuantifyMaximum(string, 3)
     assert maximum.term == string
     assert maximum.max_count == 3
     assert repr(maximum) == "QuantifyMaximum(term=String(value='test'), max_count=3)"
+    assert maximum.display_ascii_tree() == "└── Quantify({,3})\n    └── String('test')\n"
 
     between = QuantifyBetween(string, 1, 3)
     assert between.term == string
@@ -92,6 +138,7 @@ def test_dsl_init():
         repr(between)
         == "QuantifyBetween(term=String(value='test'), min_count=1, max_count=3)"
     )
+    assert between.display_ascii_tree() == "└── Quantify({1,3})\n    └── String('test')\n"
 
     with pytest.raises(
         ValueError, match="`max_count` must be greater than `min_count`"
@@ -99,6 +146,44 @@ def test_dsl_init():
         QuantifyBetween(string, 3, 1)
 
 
+def test_dsl_term_methods():
+    a = String("a")
+    b = Regex("[0-9]")
+    c = "c"
+
+    assert a + b == Sequence([a, b])
+    assert a + c == Sequence([a, String(c)])
+    assert a.__radd__(b) == Sequence([b, a])
+    assert a.__radd__(c) == Sequence([String(c), a])
+
+    assert a | b == Alternatives([a, b])
+    assert a | c == Alternatives([a, String(c)])
+    assert a.__ror__(b) == Alternatives([b, a])
+    assert a.__ror__(c) == Alternatives([String(c), a])
+
+    core_schema = a.__get_pydantic_core_schema__("", "")
+    validator = a.__get_validator__(core_schema)
+    assert validator("a") == "a"
+    with pytest.raises(
+        ValueError,
+        match="Input should be in the language of the regular expression",
+    ):
+        validator("b")
+
+    assert a.__get_pydantic_json_schema__("", "") == {"type": "string", "pattern": "a"}
+
+    assert a.matches("a")
+    assert not a.matches("b")
+
+    assert a.display_ascii_tree() == "└── String('a')\n"
+
+    with pytest.raises(NotImplementedError):
+        Term()._display_node()
+
+    assert a._display_children("") == ""
+
+    assert a.__str__() == "└── String('a')\n"
+
 def test_dsl_sequence():
     a = String("a")
     b = String("b")
@@ -144,7 +229,7 @@ def test_dsl_alternatives():
 def test_dsl_optional():
     a = String("a")
 
-    opt = optional(a)
+    opt = a.optional()
     assert isinstance(opt, Optional)
 
     opt = optional("a")
@@ -158,7 +243,7 @@ def test_dsl_optional():
 def test_dsl_exactly():
     a = String("a")
 
-    rep = exactly(2, a)
+    rep = a.exactly(2)
     assert isinstance(rep, QuantifyExact)
     assert rep.count == 2
 
@@ -173,7 +258,7 @@ def test_dsl_exactly():
 def test_dsl_at_least():
     a = String("a")
 
-    rep = at_least(2, a)
+    rep = a.at_least(2)
     assert isinstance(rep, QuantifyMinimum)
     assert rep.min_count == 2
 
@@ -188,7 +273,7 @@ def test_dsl_at_least():
 def test_dsl_at_most():
     a = String("a")
 
-    rep = at_most(2, a)
+    rep = a.at_most(2)
     assert isinstance(rep, QuantifyMaximum)
     assert rep.max_count == 2
 
@@ -203,7 +288,7 @@ def test_dsl_at_most():
 def test_between():
     a = String("a")
 
-    rep = between(1, 2, a)
+    rep = a.between(1, 2)
     assert isinstance(rep, QuantifyBetween)
     assert rep.min_count == 1
     assert rep.max_count == 2
@@ -219,7 +304,7 @@ def test_between():
 def test_dsl_zero_or_more():
     a = String("a")
 
-    rep = zero_or_more(a)
+    rep = a.zero_or_more()
     assert isinstance(rep, KleeneStar)
 
     rep = zero_or_more("a")
@@ -233,7 +318,7 @@ def test_dsl_zero_or_more():
 def test_dsl_one_or_more():
     a = String("a")
 
-    rep = one_or_more(a)
+    rep = a.one_or_more()
     assert isinstance(rep, KleenePlus)
 
     rep = one_or_more("a")
@@ -252,6 +337,49 @@ def test_dsl_aliases():
     assert isinstance(test, JsonSchema)
 
 
+def test_dsl_repeat():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `repeat` function/method is deprecated",
+    ):
+        a = String("a")
+
+        # error missing min_count or max_count
+        with pytest.raises(
+            ValueError,
+            match="You must provide a value for at least `min_count` or "
+            "`max_count`",
+        ):
+            repeat(a, None, None)
+
+        # at_least
+        assert repeat(a, 2, None) == at_least(2, a)
+        assert repeat("a", 2, None) == at_least(2, "a")
+        assert a.repeat(2, None) == a.at_least(2)
+
+        # at_most
+        assert repeat(a, None, 2) == at_most(2, a)
+        assert repeat("a", None, 2) == at_most(2, "a")
+        assert a.repeat(None, 2) == a.at_most(2)
+
+        # between
+        assert repeat(a, 1, 2) == between(1, 2, a)
+        assert repeat("a", 1, 2) == between(1, 2, "a")
+        assert a.repeat(1, 2) == a.between(1, 2)
+
+
+def test_dsl_times():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `times` function/method is deprecated",
+    ):
+        a = String("a")
+
+        assert times(a, 2) == exactly(2, a)
+        assert times("a", 2) == exactly(2, "a")
+        assert a.times(2) == a.exactly(2)
+
+
 def test_dsl_term_pydantic_simple():
     a = String("a")
 
@@ -297,3 +425,367 @@ def test_dsl_display():
         tree
         == "└── Sequence\n    ├── KleeneStar(*)\n    │   └── Alternatives(|)\n    │       ├── String('a')\n    │       └── String('b')\n    └── Regex('[0-9]')\n"
     )
+
+
+def test_cfg():
+    cfg_string = """
+?start: expr
+?expr: NUMBER
+"""
+    cfg = types.cfg(cfg_string)
+    assert isinstance(cfg, CFG)
+    assert cfg.definition.strip() == "?start: expr\n?expr: NUMBER"
+    assert cfg._display_node() == "CFG('\n?start: expr\n?expr: NUMBER\n')"
+    assert cfg.__repr__() == "CFG(definition='\n?start: expr\n?expr: NUMBER\n')"
+    assert cfg == types.cfg(cfg_string)
+    assert not cfg == "a"
+
+
+def test_fsm():
+    interegular_fsm = interegular.parse_pattern(r"a").to_fsm()
+    fsm = types.fsm(interegular_fsm)
+    assert isinstance(fsm, FSM)
+    assert fsm.fsm == interegular_fsm
+    assert fsm._display_node() == f"FSM({fsm.fsm.__repr__()})"
+    assert fsm.__repr__() == f"FSM(fsm={fsm.fsm.__repr__()})"
+
+
+def test_json_schema():
+    # init dict
+    schema = types.json_schema({"type": "string"})
+    assert schema.schema == '{"type": "string"}'
+
+    # init str
+    schema = types.json_schema('{"type": "string"}')
+    assert schema.schema == '{"type": "string"}'
+
+    # init Pydantic model
+    class PydanticModel(BaseModel):
+        foo: str
+    schema = types.json_schema(PydanticModel)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "PydanticModel", "type": "object"}'
+
+    # init TypedDict
+    class SomeTypedDict(TypedDict):
+        foo: str
+    schema = types.json_schema(SomeTypedDict)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "SomeTypedDict", "type": "object"}'
+
+    # init dataclass
+    @dataclass
+    class DataClass:
+        foo: str
+    schema = types.json_schema(DataClass)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "DataClass", "type": "object"}'
+
+    # init SchemaBuilder
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    schema = types.json_schema(builder)
+    assert schema.schema == '{"$schema": "http://json-schema.org/schema#", "type": "object"}'
+
+    # init unsupported type
+    with pytest.raises(ValueError, match="Cannot parse schema"):
+        types.json_schema(1)
+
+    # other methods
+    schema = types.json_schema('{"type": "string"}')
+    schema.__post_init__()
+    assert schema._display_node() == "JsonSchema('{\"type\": \"string\"}')"
+    assert schema.__repr__() == "JsonSchema(schema='{\"type\": \"string\"}')"
+    assert schema == types.json_schema('{"type": "string"}')
+    assert not schema == "a"
+
+
+def test_dsl_cfg_from_file():
+    grammar_content = """
+    ?start: expression
+    ?expression: term (("+" | "-") term)*
+    ?term: factor (("*" | "/") factor)*
+    ?factor: NUMBER
+    """
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.txt', delete=True) as temp_file:
+        temp_file.write(grammar_content)
+        temp_file.flush()
+        temp_file_path = temp_file.name
+        cfg = CFG.from_file(temp_file_path)
+        assert cfg == CFG(grammar_content)
+
+
+def test_dsl_json_schema_from_file():
+    schema_content = """
+    {
+        "type": "object",
+        "properties": {
+            "name": {
+                "type": "string"
+            }
+        }
+    }
+    """
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=True) as temp_file:
+        temp_file.write(schema_content)
+        temp_file.flush()
+        temp_file_path = temp_file.name
+        schema = JsonSchema.from_file(temp_file_path)
+        assert schema == JsonSchema(schema_content)
+
+
+def test_dsl_python_types_to_terms():
+    with pytest.raises(RecursionError):
+        python_types_to_terms(None, 11)
+
+    term = Term()
+    assert python_types_to_terms(term) == term
+
+    assert python_types_to_terms(int) == types.integer
+    assert python_types_to_terms(float) == types.number
+    assert python_types_to_terms(bool) == types.boolean
+    assert python_types_to_terms(str) == types.string
+    assert python_types_to_terms(datetime.time) == types.time
+    assert python_types_to_terms(datetime.date) == types.date
+    assert python_types_to_terms(datetime.datetime) == types.datetime
+    assert python_types_to_terms(dict) == types.CFG(grammars.json)
+
+    string_instance = "a"
+    assert python_types_to_terms(string_instance) == String(string_instance)
+    int_instance = 1
+    assert python_types_to_terms(int_instance) == Regex(r"1")
+    float_instance = 1.0
+    assert python_types_to_terms(float_instance) == Regex(r"1.0")
+
+    @dataclass
+    class DataClass:
+        a: int
+        b: str
+
+    assert python_types_to_terms(DataClass) == JsonSchema(
+        {
+            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
+            "required": ["a", "b"],
+            "title": "DataClass",
+            "type": "object",
+        }
+    )
+
+    class SomeTypedDict(TypedDict):
+        a: int
+        b: str
+
+    assert python_types_to_terms(SomeTypedDict) == JsonSchema(
+        {
+            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
+            "required": ["a", "b"],
+            "title": "SomeTypedDict",
+            "type": "object",
+        }
+    )
+
+    class PydanticModel(BaseModel):
+        a: int
+        b: str
+
+    assert python_types_to_terms(PydanticModel) == JsonSchema(
+        {
+            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
+            "required": ["a", "b"],
+            "title": "PydanticModel",
+            "type": "object",
+        }
+    )
+
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+    assert python_types_to_terms(builder) == JsonSchema(
+        {
+            "$schema": "http://json-schema.org/schema#",
+            "type": "object",
+            "properties": {"hi": {"type": ["integer", "string"]}},
+            "required": ["hi"]
+        }
+    )
+
+    interegular_fsm = interegular.parse_pattern(r"abc").to_fsm()
+    assert python_types_to_terms(types.fsm(interegular_fsm)).fsm is interegular_fsm
+
+    def func(a: int, b: str):
+        return (a, b)
+
+    assert python_types_to_terms(func) == JsonSchema(
+        {
+            "type": "object",
+            "properties": {
+                "a": {"title": "A", "type": "integer"},
+                "b": {"title": "B", "type": "string"},
+            },
+            "required": ["a", "b"],
+            "title": "func",
+        }
+    )
+
+    class SomeEnum(Enum):
+        a = "a"
+        b = int
+        c = func
+
+    result = python_types_to_terms(SomeEnum)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 3
+    assert result.terms[0] == String("a")
+    assert result.terms[1] == types.integer
+    assert isinstance(result.terms[2], JsonSchema)
+    schema_dict = json.loads(result.terms[2].schema)
+    assert schema_dict == {
+        "properties": {
+            "a": {"title": "A", "type": "integer"},
+            "b": {"title": "B", "type": "string"},
+        },
+        "required": ["a", "b"],
+        "title": "func",
+        "type": "object",
+    }
+
+    # for generic types we only test the dispatch as the functions that
+    # convert to terms are tested in distinct tests below
+    assert python_types_to_terms(Literal["a", "b"]) == _handle_literal(("a", "b"))
+    assert python_types_to_terms(Union[int, str]) == _handle_union((int, str), recursion_depth=0)
+    assert python_types_to_terms(list[int]) == _handle_list((int,), recursion_depth=0)
+    assert python_types_to_terms(tuple[int, str]) == _handle_tuple((int, str), recursion_depth=0)
+    assert python_types_to_terms(dict[int, str]) == _handle_dict((int, str), recursion_depth=0)
+
+
+def test_dsl_handle_literal():
+    literal = Literal["a", 1]
+    result = _handle_literal(get_args(literal))
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert result.terms[0] == String("a")
+    assert result.terms[1] == Regex(r"1")
+
+
+def test_dsl_handle_union():
+    # test simple Union
+    simple_union = Union[int, str]
+    result = _handle_union(get_args(simple_union), recursion_depth=0)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert result.terms[0] == types.integer
+    assert result.terms[1] == types.string
+
+    # test with Optional[T]
+    optional_type = PyOptional[int]
+    result = _handle_union(get_args(optional_type), recursion_depth=0)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert result.terms[0] == types.integer
+    assert result.terms[1] == String("None")
+
+    # test with more complex types
+    class TestModel(BaseModel):
+        field: str
+
+    class TestEnum(Enum):
+        a = "a"
+        b = "b"
+
+    complex_union = Union[TestModel, TestEnum]
+    result = _handle_union(get_args(complex_union), recursion_depth=0)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert isinstance(result.terms[0], JsonSchema)
+    assert isinstance(result.terms[1], Alternatives)
+    assert len(result.terms[1].terms) == 2
+    assert result.terms[1].terms[0] == String("a")
+    assert result.terms[1].terms[1] == String("b")
+
+
+def test_dsl_handle_list():
+    with pytest.raises(TypeError):
+        _handle_list((int, str), recursion_depth=0)
+
+    # simple type
+    list_type = list[int]
+    result = _handle_list(get_args(list_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 4
+    assert result.terms[0] == String("[")
+    assert result.terms[1] == types.integer
+    assert isinstance(result.terms[2], KleeneStar)
+    assert result.terms[2].term == Sequence([String(", "), types.integer])
+    assert result.terms[3] == String("]")
+
+    # more complex type
+    list_type = list[Union[int, str]]
+    result = _handle_list(get_args(list_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 4
+    assert result.terms[0] == String("[")
+    assert result.terms[1] == _handle_union(get_args(Union[int, str]), recursion_depth=0)
+    assert isinstance(result.terms[2], KleeneStar)
+    assert result.terms[2].term == Sequence([String(", "), _handle_union(get_args(Union[int, str]), recursion_depth=0)])
+    assert result.terms[3] == String("]")
+
+
+def test_dsl_handle_tuple():
+    # empty tuple
+    tuple_type = Tuple[()]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, String)
+    assert result.value == "()"
+
+    # tuple with ellipsis
+    tuple_type = tuple[int, ...]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 4
+    assert result.terms[0] == String("(")
+    assert result.terms[1] == types.integer
+    assert isinstance(result.terms[2], KleeneStar)
+    assert result.terms[2].term == Sequence([String(", "), types.integer])
+    assert result.terms[3] == String(")")
+
+    # tuple with fixed length
+    tuple_type = tuple[int, str]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 5
+    assert result.terms[0] == String("(")
+    assert result.terms[1] == types.integer
+    assert result.terms[2] == String(", ")
+    assert result.terms[3] == types.string
+    assert result.terms[4] == String(")")
+
+    # tuple with fixed length and complex types
+    tuple_type = tuple[int, Union[str, int]]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 5
+    assert result.terms[0] == String("(")
+    assert result.terms[1] == types.integer
+    assert result.terms[2] == String(", ")
+    assert result.terms[3] == _handle_union(get_args(Union[str, int]), recursion_depth=0)
+    assert result.terms[4] == String(")")
+
+
+def test_dsl_handle_dict():
+    # args of incorrect length
+    with pytest.raises(TypeError):
+        incorrect_dict_type = dict[int, str, int]
+        _handle_dict(get_args(incorrect_dict_type), recursion_depth=0)
+
+    # correct type
+    dict_type = dict[int, str]
+    result = _handle_dict(get_args(dict_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 3
+    assert result.terms[0] == String("{")
+    assert isinstance(result.terms[1], Optional)
+    assert isinstance(result.terms[1].term, Sequence)
+    assert len(result.terms[1].term.terms) == 4
+    assert result.terms[1].term.terms[0] == types.integer
+    assert result.terms[1].term.terms[1] == String(":")
+    assert result.terms[1].term.terms[2] == types.string
+    assert result.terms[1].term.terms[3] == KleeneStar(Sequence([String(", "), types.integer, String(":"), types.string]))
+    assert result.terms[2] == String("}")
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
new file mode 100644
index 000000000..4d73312d6
--- /dev/null
+++ b/tests/types/test_types_utils.py
@@ -0,0 +1,409 @@
+import datetime
+import pytest
+import sys
+from dataclasses import dataclass
+from enum import Enum
+from functools import partial
+from typing import (
+    Annotated,
+    Any,
+    Dict,
+    List,
+    Literal,
+    NewType,
+    Optional,
+    Tuple,
+    Union
+)
+
+import interegular
+from genson import SchemaBuilder
+from pydantic import BaseModel, constr
+
+from outlines.types.dsl import JsonSchema
+from outlines.types.utils import (
+    get_enum_from_literal,
+    get_schema_from_enum,
+    get_schema_from_signature,
+    is_bool,
+    is_callable,
+    is_date,
+    is_dataclass,
+    is_datetime,
+    is_enum,
+    is_float,
+    is_float_instance,
+    is_genson_schema_builder,
+    is_int,
+    is_int_instance,
+    is_interegular_fsm,
+    is_literal,
+    is_native_dict,
+    is_pydantic_model,
+    is_str,
+    is_str_instance,
+    is_time,
+    is_typed_dict,
+    is_typing_dict,
+    is_typing_list,
+    is_typing_tuple,
+    is_union
+)
+
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
+
+# Type identification
+
+
+@pytest.fixture
+def sample_enum():
+    class SampleEnum(Enum):
+        A = 1
+        B = 2
+
+    return SampleEnum
+
+@pytest.fixture
+def sample_complex_enum():
+    def add_func(a: float, b: float) -> float:
+        return a + b
+
+    class SampleComplexEnum(Enum):
+        add = partial(add_func)
+        a = "a"
+        b = 2
+
+    return SampleComplexEnum
+
+@pytest.fixture
+def sample_empty_enum():
+    def add_func(a: float, b: float) -> float:
+        return a + b
+
+    # the enum is empty because the function is not registered as callable
+    class SampleEmptyEnum(Enum):
+        add = add_func
+
+    return SampleEmptyEnum
+
+@pytest.fixture
+def sample_class():
+    class SampleClass:
+        pass
+
+    return SampleClass
+
+@pytest.fixture
+def sample_dataclass():
+    @dataclass
+    class SampleDataclass:
+        field1: str
+        field2: int
+
+    return SampleDataclass
+
+@pytest.fixture
+def sample_typed_dict():
+    class SampleTypedDict(TypedDict):
+        name: str
+        age: int
+
+    return SampleTypedDict
+
+@pytest.fixture
+def sample_pydantic_model():
+    class SamplePydanticModel(BaseModel):
+        name: str
+        age: int
+
+    return SamplePydanticModel
+
+@pytest.fixture
+def sample_schema_builder():
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+    return builder
+
+@pytest.fixture
+def sample_function():
+    def sample_function(foo: str, bar: List[int]):
+        pass
+
+    return sample_function
+
+@pytest.fixture
+def sample_function_missing_type():
+    def sample_function(foo, bar: List[int]):
+        pass
+
+    return sample_function
+
+
+def test_is_int():
+    assert is_int(int)
+    assert not is_int(float)
+    assert not is_int(1)
+    assert not is_int(List[int])
+    assert not is_int(Dict[int, int])
+    assert is_int(Annotated[int, "some metadata"])
+    assert not is_int(Annotated[str, "some metadata"])
+    assert is_int(NewType("UserId", int))
+    assert not is_int(NewType("UserId", str))
+
+
+def test_is_int_instance():
+    assert is_int_instance(1)
+    assert not is_int_instance(1.0)
+    assert not is_int_instance("1")
+    assert not is_int_instance(int)
+
+
+def test_is_float():
+    assert is_float(float)
+    assert not is_float(int)
+    assert not is_float(1.0)
+    assert not is_float(List[float])
+    assert not is_float(Dict[float, float])
+    assert is_float(Annotated[float, "some metadata"])
+    assert not is_float(Annotated[int, "some metadata"])
+    assert is_float(NewType("UserId", float))
+    assert not is_float(NewType("UserId", int))
+
+
+def test_is_float_instance():
+    assert is_float_instance(1.0)
+    assert not is_float_instance(1)
+    assert not is_float_instance("1.0")
+    assert not is_float_instance(float)
+
+
+def test_is_str():
+    assert is_str(str)
+    assert not is_str(int)
+    assert not is_str("hello")
+    assert not is_str(List[str])
+    assert not is_str(Dict[str, str])
+    assert is_str(Annotated[str, "some metadata"])
+    assert not is_str(Annotated[int, "some metadata"])
+    assert is_str(NewType("UserId", str))
+    assert not is_str(NewType("UserId", int))
+
+
+def test_is_str_instance():
+    assert is_str_instance("hello")
+    assert is_str_instance("")
+    assert is_str_instance("123")
+    assert not is_str_instance(123)
+    assert not is_str_instance(str)
+
+
+def test_is_bool():
+    assert is_bool(bool)
+    assert not is_bool(int)
+    assert not is_bool(True)
+    assert is_bool(Annotated[bool, "some metadata"])
+    assert not is_bool(Annotated[int, "some metadata"])
+    assert is_bool(NewType("UserId", bool))
+    assert not is_bool(NewType("UserId", int))
+
+
+def test_is_datetime():
+    assert is_datetime(datetime.datetime)
+    assert not is_datetime(datetime.date)
+    assert not is_datetime(datetime.time)
+    assert not is_datetime(datetime.datetime.now())
+
+
+def test_is_date():
+    assert is_date(datetime.date)
+    assert not is_date(datetime.datetime)
+    assert not is_date(datetime.time)
+    assert not is_date(datetime.date.today())
+
+
+def test_is_time():
+    assert is_time(datetime.time)
+    assert not is_time(datetime.datetime)
+    assert not is_time(datetime.date)
+    assert not is_time(datetime.time(12, 30))
+
+
+def test_is_native_dict():
+    assert is_native_dict(dict)
+    assert not is_native_dict({})
+    assert not is_native_dict({"key": "value"})
+    assert not is_native_dict(list)
+    assert not is_native_dict(dict[str, int])
+
+
+def test_is_typing_dict():
+    assert is_typing_dict(dict[str, int])
+    assert is_typing_dict(Dict[int, str])
+    assert not is_typing_dict(dict)
+    assert not is_typing_dict({})
+
+
+def test_is_typing_list():
+    assert is_typing_list(list[int])
+    assert is_typing_list(List[int])
+    assert not is_typing_list(list)
+    assert not is_typing_list([])
+    assert not is_typing_list(dict)
+
+
+def test_is_typing_tuple():
+    assert is_typing_tuple(tuple[int, str])
+    assert is_typing_tuple(Tuple[int, str])
+    assert not is_typing_tuple(tuple)
+    assert not is_typing_tuple(())
+    assert not is_typing_tuple(list)
+
+
+def test_is_union():
+    assert is_union(Union[int, str])
+    assert is_union(Optional[int])
+    assert not is_union(list)
+    assert not is_union(["a", "b"])
+    assert not is_union(Literal[int, str])
+
+
+def test_is_literal():
+    assert is_literal(Literal["a", "b"])
+    assert not is_literal(str)
+    assert not is_literal("a")
+    assert not is_literal(["a", "b"])
+    assert not is_literal(Union[str, int])
+
+
+def test_is_dataclass(
+    sample_dataclass,
+    sample_class,
+    sample_typed_dict,
+    sample_pydantic_model
+):
+    assert is_dataclass(sample_dataclass)
+    assert not is_dataclass(sample_dataclass(field1="test", field2=123))
+    assert not is_dataclass(dict)
+    assert not is_dataclass(sample_class)
+    assert not is_dataclass(sample_typed_dict)
+    assert not is_dataclass(sample_pydantic_model)
+
+
+def test_is_typed_dict(
+    sample_typed_dict,
+    sample_class,
+    sample_dataclass,
+    sample_pydantic_model
+):
+    assert is_typed_dict(sample_typed_dict)
+    assert not is_typed_dict(sample_typed_dict(name="test", age=30))
+    assert not is_typed_dict(dict)
+    assert not is_typed_dict(sample_class)
+    assert not is_typed_dict(sample_dataclass)
+    assert not is_typed_dict(sample_pydantic_model)
+
+
+def test_is_pydantic_model(
+    sample_pydantic_model,
+    sample_class,
+    sample_dataclass,
+    sample_typed_dict
+):
+    assert is_pydantic_model(sample_pydantic_model)
+    assert not is_pydantic_model(sample_pydantic_model(name="test", age=30))  # Instance
+    assert not is_pydantic_model(dict)
+    assert not is_pydantic_model(sample_class)
+    assert not is_pydantic_model(sample_dataclass)
+    assert not is_pydantic_model(sample_typed_dict)
+
+
+def test_is_genson_schema_builder(
+    sample_schema_builder,
+    sample_class,
+    sample_dataclass,
+    sample_typed_dict,
+    sample_pydantic_model
+):
+    assert is_genson_schema_builder(sample_schema_builder)
+    assert not is_genson_schema_builder(dict)
+    assert not is_genson_schema_builder(str)
+    assert not is_genson_schema_builder({"type": 'object', "properties": {}})
+    assert not is_genson_schema_builder('{"type": "object", "properties": {}}')
+    assert not is_genson_schema_builder(sample_class)
+    assert not is_genson_schema_builder(sample_dataclass)
+    assert not is_genson_schema_builder(sample_typed_dict)
+    assert not is_genson_schema_builder(sample_pydantic_model)
+
+
+def test_is_enum(sample_enum):
+    assert is_enum(sample_enum)
+    assert not is_enum(sample_enum.A)
+    assert not is_enum(dict)
+    assert not is_enum(Literal["a", "b"])
+    assert not is_enum(["a", "b"])
+
+
+def test_is_callable(sample_function, sample_class, sample_dataclass, sample_typed_dict, sample_pydantic_model):
+    assert is_callable(sample_function)
+    assert is_callable(lambda x: x)
+    assert not is_callable(dict)
+    assert not is_callable(sample_class)
+    assert not is_callable(sample_dataclass)
+    assert not is_callable(sample_typed_dict)
+    assert not is_callable(sample_pydantic_model)
+
+
+def test_is_interegular_fsm():
+    fsm = interegular.parse_pattern(r"[0-9]+").to_fsm()
+    assert is_interegular_fsm(fsm)
+    assert not is_interegular_fsm(dict)
+
+
+# Type conversion
+
+
+def test_get_enum_from_literal(sample_enum):
+    basic_enum = get_enum_from_literal(Literal["a", "b"])
+    assert(is_enum(basic_enum))
+    assert basic_enum.a.value == "a"
+    assert basic_enum.b.value == "b"
+
+    complex_enum = get_enum_from_literal(Literal["a", 1, True, None, sample_enum.A])
+    assert is_enum(complex_enum)
+    assert complex_enum.a.value == "a"
+    assert getattr(complex_enum, "1").value == 1
+    assert getattr(complex_enum, "True").value
+    assert getattr(complex_enum, "None").value is None
+    assert getattr(complex_enum, "SampleEnum.A").value == sample_enum.A
+
+
+def test_get_schema_from_signature(sample_function, sample_function_missing_type):
+    result = get_schema_from_signature(sample_function)
+    assert result["type"] == "object"
+    assert list(result["properties"].keys()) == ["foo", "bar"]
+    assert result["properties"]["foo"]["type"] == "string"
+    assert result["properties"]["bar"]["type"] == "array"
+    assert result["properties"]["bar"]["items"]["type"] == "integer"
+
+    # in case of a function missing type annotations
+    with pytest.raises(ValueError):
+        get_schema_from_signature(sample_function_missing_type)
+
+
+def test_get_schema_from_enum(sample_complex_enum, sample_empty_enum):
+    schema = get_schema_from_enum(sample_complex_enum)
+    assert JsonSchema(schema)
+    assert schema["title"] == sample_complex_enum.__name__
+    assert len(schema["oneOf"]) == len(sample_complex_enum)
+    for elt in schema["oneOf"]:
+        assert type(elt) in [int, float, bool, type(None), str, dict]
+
+    # in case of an empty enum because the function member is not registered as callable
+    with pytest.raises(ValueError):
+        get_schema_from_enum(sample_empty_enum)
diff --git a/tests/v0_legacy/generate/test_api_legacy.py b/tests/v0_legacy/generate/test_api_legacy.py
new file mode 100644
index 000000000..75cf72ff0
--- /dev/null
+++ b/tests/v0_legacy/generate/test_api_legacy.py
@@ -0,0 +1,186 @@
+from io import BytesIO
+from dataclasses import asdict
+from typing import Generator
+from urllib.request import urlopen
+
+import pytest
+import transformers
+from PIL import Image
+from anthropic import Anthropic as AnthropicClient
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers(
+            "erwanf/gpt2-mini"
+        )
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=transformers.LlavaForConditionalGeneration,
+            processor_class=transformers.AutoProcessor,
+        )
+
+
+@pytest.fixture(scope="session")
+def llama_cpp_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        return models.llamacpp(
+            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+
+
+def create_generator_v0_adapter(model):
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(
+            model,
+            int,
+            samplers.greedy(),
+        )
+    else:
+        return GeneratorV0Adapter(
+            model,
+            int,
+            samplers.greedy(),
+        )
+
+
+def test_generator_v0_adapter_init(transformers_model):
+    # wrong model type
+    model = models.from_anthropic(AnthropicClient(), "claude-3-haiku-20240307")
+    with pytest.raises(
+        ValueError,
+        match="You can only use the v0 API with models that were already"
+        + "available in v0. Got <class 'outlines.models.anthropic.Anthropic'>.",
+    ):
+        GeneratorV0Adapter(model, None, samplers.greedy())
+
+    # unknown output type
+    with pytest.raises(
+        TypeError,
+        match="is currently not supported",
+    ):
+        GeneratorV0Adapter(
+            transformers_model, type, samplers.greedy()
+        )
+
+    # valid initialization
+    generator_v0_adapter = GeneratorV0Adapter(
+        transformers_model,
+        int,
+        samplers.greedy(),
+    )
+    assert isinstance(generator_v0_adapter, GeneratorV0Adapter)
+    assert generator_v0_adapter.model == transformers_model
+    assert generator_v0_adapter.sampling_params == asdict(samplers.greedy().sampling_params)
+    assert isinstance(generator_v0_adapter.generator, SteerableGenerator)
+    assert generator_v0_adapter.generator.model == transformers_model
+    assert isinstance(generator_v0_adapter.generator.logits_processor, RegexLogitsProcessor)
+
+
+def test_generator_v0_adapter_create_inference_params(
+    llama_cpp_model,
+    transformers_vision_model,
+    transformers_model,
+):
+    args = {
+        "max_tokens": 10,
+        "stop_at": "\n",
+        "seed": 42
+    }
+    kwargs = {
+        "foo": "bar",
+    }
+
+    result = (
+        create_generator_v0_adapter(llama_cpp_model)
+        .create_inference_params(**args, **kwargs)
+    )
+    assert isinstance(result, dict)
+
+    result = (
+        create_generator_v0_adapter(transformers_vision_model)
+        .create_inference_params(**args, **kwargs)
+    )
+    assert isinstance(result, dict)
+
+    result = (
+        create_generator_v0_adapter(transformers_model)
+        .create_inference_params(**args, **kwargs)
+    )
+    assert isinstance(result, dict)
+
+
+def test_generator_vision_v0_adapter_merge_prompts_and_media(transformers_vision_model):
+    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
+    prompts = ["Hello, world!", "Goodbye, world!"]
+    media = [
+        img_from_url(
+            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+        ),
+        img_from_url(
+            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+        )
+    ]
+    result = generator_v0_adapter.merge_prompts_and_media(prompts, media)
+    assert isinstance(result, dict)
+    assert result["text"] == prompts
+    assert result["images"] == media
+
+
+def test_generator_v0_adapter_call_stream(llama_cpp_model, transformers_model):
+    generator_v0_adapter = create_generator_v0_adapter(transformers_model)
+    result = generator_v0_adapter("Hello, world!", max_tokens=10)
+    assert isinstance(result, str)
+    result = generator_v0_adapter(["Hello, world!", "Goodbye, world!"], max_tokens=10)
+    assert isinstance(result, list)
+    for r in result:
+        assert isinstance(r, str)
+
+    generator_v0_adapter = create_generator_v0_adapter(llama_cpp_model)
+    gen = generator_v0_adapter.stream("Hello, world!", max_tokens=10)
+    assert isinstance(gen, Generator)
+
+
+def test_generator_v0_adapter_vision_call_stream(transformers_vision_model):
+    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
+    media = img_from_url(
+        "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+    )
+    result = generator_v0_adapter("Hello, world!<image>", media, max_tokens=10)
+    assert isinstance(result, str)
+    with pytest.raises(
+        NotImplementedError,
+        match="Streaming is not implemented for Transformers models",
+    ):
+        gen = generator_v0_adapter.stream("Hello, world!<image>", media, max_tokens=10)
+        assert isinstance(gen, Generator)
diff --git a/tests/v0_legacy/generate/test_cfg_legacy.py b/tests/v0_legacy/generate/test_cfg_legacy.py
new file mode 100644
index 000000000..2b8a2734e
--- /dev/null
+++ b/tests/v0_legacy/generate/test_cfg_legacy.py
@@ -0,0 +1,111 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import grammars, models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import CFGLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.cfg import cfg
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def llama_cpp_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        return models.llamacpp(
+            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf"
+        )
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_cfg_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use CFG-based structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API.",
+        ):
+            cfg(openai_model, grammars.arithmetic)
+
+
+def test_cfg_legacy_llama_cpp(llama_cpp_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Not yet available due to bug in llama_cpp tokenizer",
+        ):
+            cfg(llama_cpp_model, grammars.arithmetic)
+
+
+def test_cfg_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        generator = cfg(transformers_vision_model, grammars.arithmetic)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
+
+
+def test_cfg_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        generator = cfg(transformers_model, grammars.arithmetic, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_choice_legacy.py b/tests/v0_legacy/generate/test_choice_legacy.py
new file mode 100644
index 000000000..fbc4650e2
--- /dev/null
+++ b/tests/v0_legacy/generate/test_choice_legacy.py
@@ -0,0 +1,97 @@
+import json as pyjson
+from dataclasses import asdict
+from enum import Enum
+from typing import Callable
+
+import pytest
+from pydantic import BaseModel
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.choice import choice
+
+
+class EnumChoice(Enum):
+    foo = "foo"
+    bar = "bar"
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+def test_choice_legacy_standard_enum(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `choice` function is deprecated",
+    ):
+        generator = choice(transformers_model, EnumChoice, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_choice_legacy_standard_list_strings(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `choice` function is deprecated",
+    ):
+        generator = choice(transformers_model, ["foo", "bar"])
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert generator.format_sequence("foo") == "foo"
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_choice_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `choice` function is deprecated",
+    ):
+        with pytest.raises(
+            ValueError,
+            match="The `choice` function with OpenAI only supports a list of "
+            + "strings as choices.",
+        ):
+            choice(openai_model, EnumChoice)
+
+        generator = choice(openai_model, ["foo", "bar"])
+        assert isinstance(generator, Callable)
diff --git a/tests/v0_legacy/generate/test_format_legacy.py b/tests/v0_legacy/generate/test_format_legacy.py
new file mode 100644
index 000000000..81503fb8f
--- /dev/null
+++ b/tests/v0_legacy/generate/test_format_legacy.py
@@ -0,0 +1,93 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.format import format
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_format_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `format` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use Python type-structured generation with an OpenAI model"
+            + " due to the limitations of the OpenAI API.",
+        ):
+            format(openai_model, int)
+
+
+def test_format_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `format` function is deprecated",
+    ):
+        generator = format(transformers_vision_model, int)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        RegexLogitsProcessor
+    )
+
+
+def test_format_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `format` function is deprecated",
+    ):
+        generator = format(transformers_model, int, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        RegexLogitsProcessor
+    )
diff --git a/tests/v0_legacy/generate/test_fsm_legacy.py b/tests/v0_legacy/generate/test_fsm_legacy.py
new file mode 100644
index 000000000..fe6e5ba2e
--- /dev/null
+++ b/tests/v0_legacy/generate/test_fsm_legacy.py
@@ -0,0 +1,104 @@
+from dataclasses import asdict
+
+import interegular
+import pytest
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import GuideLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.fsm import fsm
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_fsm_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `fsm` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use FSM-based generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API.",
+        ):
+            fsm(
+                openai_model,
+                interegular.parse_pattern(r"[a-z]+").to_fsm()
+            )
+
+
+def test_fsm_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `fsm` function is deprecated",
+    ):
+        generator = fsm(
+            transformers_vision_model,
+            interegular.parse_pattern(r"[a-z]+").to_fsm()
+        )
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        GuideLogitsProcessor
+    )
+
+
+def test_fsm_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `fsm` function is deprecated",
+    ):
+        generator = fsm(
+            transformers_model,
+            interegular.parse_pattern(r"[a-z]+").to_fsm(),
+            samplers.greedy()
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        GuideLogitsProcessor
+    )
diff --git a/tests/v0_legacy/generate/test_json_legacy.py b/tests/v0_legacy/generate/test_json_legacy.py
new file mode 100644
index 000000000..d09d0347b
--- /dev/null
+++ b/tests/v0_legacy/generate/test_json_legacy.py
@@ -0,0 +1,200 @@
+import json as pyjson
+import pytest
+from dataclasses import asdict
+
+from pydantic import BaseModel
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.json import json
+
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "foo": {
+            "type": "string"
+        }
+    },
+    "required": ["foo"]
+}
+"""
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_json_legacy_standard_callable(transformers_model):
+    def foo(bar: str) -> str:
+        return bar
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_model, foo, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_standard_json_string(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_model, JSON_SCHEMA_STRING)
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_standard_pydantic(transformers_model):
+    class Foo(BaseModel):
+        foo: str
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_model, Foo)
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"foo": "value"}')
+        == Foo.model_validate_json('{"foo": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_vision_model, JSON_SCHEMA_STRING)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_openai_wrong_sampler(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="The OpenAI API does not support any other sampling "
+            + "algorithm than the multinomial sampler.",
+        ):
+            json(openai_model, JSON_SCHEMA_STRING, samplers.greedy())
+
+
+def test_json_legacy_openai_pydantic(openai_model):
+    class Foo(BaseModel):
+        foo: str
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(openai_model, Foo)
+    assert isinstance(generator, models.OpenAI)
+    assert (
+        generator.format_sequence('{"foo": "value"}')
+        == Foo.model_validate_json('{"foo": "value"}')
+    )
+
+
+def test_json_legacy_openai_json_string(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(openai_model, JSON_SCHEMA_STRING)
+    assert isinstance(generator, models.OpenAI)
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+
+
+def test_json_legacy_openai_not_supported(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        with pytest.raises(ValueError, match="Cannot parse schema"):
+            json(openai_model, int)
diff --git a/tests/v0_legacy/generate/test_regex_legacy.py b/tests/v0_legacy/generate/test_regex_legacy.py
new file mode 100644
index 000000000..86da00c35
--- /dev/null
+++ b/tests/v0_legacy/generate/test_regex_legacy.py
@@ -0,0 +1,90 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.regex import regex
+
+
+REGEX_STRING = r"^[a-z]+$"
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_regex_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `regex` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use regex-structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API.",
+        ):
+            regex(openai_model, REGEX_STRING)
+
+
+def test_regex_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `regex` function is deprecated",
+    ):
+        generator = regex(transformers_vision_model, REGEX_STRING)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
+
+
+def test_regex_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `regex` function is deprecated",
+    ):
+        generator = regex(transformers_model, REGEX_STRING, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_text_legacy.py b/tests/v0_legacy/generate/test_text_legacy.py
new file mode 100644
index 000000000..e7412b295
--- /dev/null
+++ b/tests/v0_legacy/generate/test_text_legacy.py
@@ -0,0 +1,83 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.text import text
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_text_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = text(openai_model)
+    assert generator == openai_model
+    assert isinstance(generator, models.OpenAI)
+
+
+def test_text_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = text(transformers_vision_model)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert generator.generator.logits_processor is None
+
+
+def test_text_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = text(transformers_model, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert generator.generator.logits_processor is None
diff --git a/tests/v0_legacy/models/test_exllamav2_legacy.py b/tests/v0_legacy/models/test_exllamav2_legacy.py
new file mode 100644
index 000000000..06fa9e17e
--- /dev/null
+++ b/tests/v0_legacy/models/test_exllamav2_legacy.py
@@ -0,0 +1,76 @@
+import warnings
+
+import pytest
+
+try:
+    import exllamav2  # type: ignore
+    HAS_EXV2 = True
+except ImportError:
+    HAS_EXV2 = False
+    warnings.warn(
+        "Skipping exllamav2 legacy tests because exllamav2 is not available."
+    )
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+pytestmark = pytest.mark.skipif(
+    not HAS_EXV2,
+    reason="Exllamav2 is not available."
+)
+
+
+@pytest.fixture(scope="session")
+def model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `exllamav2` function is deprecated",
+    ):
+        model = models.exl2(
+            model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+            max_seq_len=1024,
+            cache_q4=False,
+        )
+    return model
+
+
+def test_exllamav2_legacy_init(model):
+    assert isinstance(model, models.ExllamaV2)
+    assert hasattr(model, "generator")
+    assert hasattr(model, "tokenizer")
+    assert model.tokenizer.convert_token_to_string(1) == 1
+    assert hasattr(model, "max_seq_len")
+    assert isinstance(model.max_seq_len, int)
+
+
+def test_exllamav2_legacy_call(model):
+    # minimal
+    generator = generate.text(model)
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator("Respond with a single word.")
+    assert isinstance(result, str)
+
+    # with args
+    generator = generate.text(
+        model,
+        samplers.multinomial(samples=2, temperature=10, top_p=0.5, top_k=2)
+    )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(["Write a short story.", "Write a poem."], 2, "e", 2)
+    assert isinstance(result, list)
+    for sequence in result:
+        assert isinstance(sequence, list)
+        for s in sequence:
+            assert isinstance(s, str)
+            assert len(s) < 20
+
+    # stream
+    generator = generate.text(model)
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator.stream("Respond with a single word.", 2)
+    assert isinstance(result, GeneratorV0Adapter)
+    for r in result:
+        assert isinstance(r, str)
diff --git a/tests/v0_legacy/models/test_llamacpp_legacy.py b/tests/v0_legacy/models/test_llamacpp_legacy.py
new file mode 100644
index 000000000..aead5bd86
--- /dev/null
+++ b/tests/v0_legacy/models/test_llamacpp_legacy.py
@@ -0,0 +1,169 @@
+import warnings
+from dataclasses import asdict
+
+import pytest
+from llama_cpp import Llama
+from llama_cpp.llama_tokenizer import LlamaHFTokenizer, LlamaTokenizer # type: ignore
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+
+def test_llamacpp_legacy_init():
+    # without tokenizer
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        with pytest.warns(
+            match="The pre-tokenizer in `llama.cpp` handles unicode improperly"
+        ):
+            model = models.llamacpp(
+                repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+                filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            )
+    assert isinstance(model, models.LlamaCpp)
+    assert isinstance(model.model, Llama)
+    assert not model.model.verbose
+
+    # with tokenizer
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        model = models.llamacpp(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            tokenizer=LlamaHFTokenizer.from_pretrained("distilbert-base-uncased"),
+            n_ctx=0,
+            verbose=True,
+        )
+    assert isinstance(model, models.LlamaCpp)
+    assert isinstance(model.model, Llama)
+    assert model.model.verbose
+    assert isinstance(model.model.tokenizer(), LlamaTokenizer)
+
+
+def test_llamacpp_legacy_params_adapter_generation_params():
+    result = llamacpp_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert not result.get("stop")
+    assert not result.get("seed")
+    assert result.get("max_tokens") == -1
+
+    result = llamacpp_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
+        {}
+    )
+    assert result.get("stop") == "foo"
+    assert result.get("seed") == 2
+    assert result.get("max_tokens") == 10
+
+
+def test_llamacpp_legacy_params_adapter_sampling_params():
+    with pytest.raises(
+        NotImplementedError,
+        match="does not support Beam Search"
+    ):
+        result = llamacpp_params_adapter(
+            asdict(samplers.beam_search().sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    with pytest.raises(
+        NotImplementedError,
+        match="does not allow to take several samples"
+    ):
+        result = llamacpp_params_adapter(
+            asdict(samplers.multinomial(samples=2).sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    result = llamacpp_params_adapter(
+        {"sampler": "multinomial", "num_samples": 1, "top_p": None, "top_k": None, "temperature": None},
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("top_p") == 1.0
+    assert result.get("top_k") == -1
+    assert result.get("temperature") == 1.0
+
+    result = llamacpp_params_adapter(
+        {"sampler": "greedy", "num_samples": 1, "top_p": 2, "top_k": 0, "temperature": 0.5},
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("top_p") == 2
+    assert result.get("top_k") == 0
+    assert result.get("temperature") == 0.5
+
+
+def test_llamacpp_legacy_params_adapter_model_specific_params():
+    result = llamacpp_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("min_p") == 0.0
+    assert result.get("repeat_penalty") == 1.0
+
+    result = llamacpp_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
+        {"min_p": 0.5, "repeat_penalty": 1.5, "stream": True}
+    )
+    assert result.get("min_p") == 0.5
+    assert result.get("repeat_penalty") == 1.5
+    assert not hasattr(result, "stream")
+
+
+def test_llamacpp_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        model = models.llamacpp(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                top_p=0.9,
+                top_k=10,
+                temperature=0.7
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!",
+        10,
+        "foo",
+        2,
+        min_p=0.5,
+        repeat_penalty=1.5,
+    )
+    assert isinstance(result, str)
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!",
+        10,
+        "foo",
+        2,
+        min_p=0.5,
+        repeat_penalty=1.5,
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_mlxlm_legacy.py b/tests/v0_legacy/models/test_mlxlm_legacy.py
new file mode 100644
index 000000000..e70943ef2
--- /dev/null
+++ b/tests/v0_legacy/models/test_mlxlm_legacy.py
@@ -0,0 +1,193 @@
+import warnings
+from dataclasses import asdict
+
+import pytest
+
+try:
+    import mlx.core as mx
+    import mlx_lm  # noqa: F401
+
+    assert mx.metal.is_available()
+    HAS_MLX = True
+except ImportError:
+    HAS_MLX = False
+    warnings.warn("Skipping MLXLM legacy tests because MLX is not available.")
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
+
+pytestmark = pytest.mark.skipif(
+    not HAS_MLX,
+    reason="MLX tests require Apple Silicon."
+)
+
+
+def test_mlxlm_legacy_init():
+    # minimal
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mlxlm` function is deprecated",
+    ):
+        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
+    assert isinstance(model, models.MLXLM)
+    assert isinstance(model.model, mlx_lm.models.llama.Model)
+
+    # with optional arguments
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mlxlm` function is deprecated",
+    ):
+        model = models.mlxlm(
+            "mlx-community/SmolLM-135M-Instruct-4bit",
+            tokenizer_config={"max_length": 100},
+            model_config={"max_length": 100},
+            adapter_path=None,
+            lazy=False,
+        )
+    assert isinstance(model, models.MLXLM)
+    assert isinstance(model.model, mlx_lm.models.llama.Model)
+    assert isinstance(model.mlx_tokenizer, mlx_lm.tokenizer_utils.TokenizerWrapper)
+
+
+def test_mlxlm_legacy_params_adapter_generation_params():
+    # default
+    result = mlxlm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("stop") is None
+    assert result.get("seed") is None
+    assert result.get("max_tokens") == int(1e9)
+
+    # max_tokens
+    result = mlxlm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": 2},
+        {}
+    )
+    assert result.get("max_tokens") == 2
+
+    # stop_at
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support `stop_at`."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial().sampling_params),
+            {"stop_at": "foo", "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # seed
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support `seed`."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial().sampling_params),
+            {"stop_at": None, "seed": 2, "max_tokens": None},
+            {}
+        )
+
+
+def test_mlxlm_legacy_params_adapter_sampling_params():
+    # beam search
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support Beam Search."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.beam_search().sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # num_samples
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not allow to take several samples."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial(samples=2).sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # top_k
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support top_k."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial(top_k=2).sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # temperature
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `temperature` parameter is not available",
+    ):
+        result = mlxlm_params_adapter(
+            {
+                "sampler": "multinomial",
+                "num_samples": 1,
+                "top_p": None,
+                "top_k": None,
+                "temperature": 0.5
+            },
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+        assert result.get("temperature") is None
+
+    # top_p
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `top_p` parameter is not available",
+    ):
+        result = mlxlm_params_adapter(
+            {
+                "sampler": "multinomial",
+                "num_samples": 1,
+                "top_p": 2,
+                "top_k": None,
+                "temperature": None
+            },
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+        assert result.get("top_p") is None
+
+
+def test_mlxlm_legacy_params_adapter_model_specific_params():
+    result = mlxlm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {"length_penalty": 0.5}
+    )
+    assert result.get("length_penalty") == 0.5
+
+
+def test_mlxlm_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mlxlm` function is deprecated",
+    ):
+        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
+
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial()
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator("Hello, world!", 10)
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_openai_legacy.py b/tests/v0_legacy/models/test_openai_legacy.py
new file mode 100644
index 000000000..c07443dab
--- /dev/null
+++ b/tests/v0_legacy/models/test_openai_legacy.py
@@ -0,0 +1,159 @@
+import os
+
+import pytest
+from openai import AsyncAzureOpenAI, AsyncOpenAI
+
+from outlines import models, generate
+from outlines.models.openai import OpenAI, OpenAIConfig, OpenAILegacy
+
+
+# Set the OPENAI_API_KEY environment variable to "MOCK_VALUE" if it is not set
+# not to raise an error when the OpenAI client is initialized
+api_key = os.getenv("OPENAI_API_KEY")
+if not api_key:
+    os.environ["OPENAI_API_KEY"] = "MOCK_VALUE"
+azure_api_key = os.getenv("AZURE_OPENAI_API_KEY")
+if not azure_api_key:
+    os.environ["AZURE_OPENAI_API_KEY"] = "MOCK_VALUE"
+
+
+def test_openai_legacy_init():
+    # through the openai function with a client and no config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        client = AsyncOpenAI()
+        model = models.openai(client, system_prompt="You are a helpful assistant.")
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.config == OpenAIConfig()
+    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # through the openai function with a client and config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            client,
+            OpenAIConfig(stop=["."]),
+            system_prompt="You are a helpful assistant.",
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.config == OpenAIConfig(stop=["."])
+    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # through the openai function with model name and no config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            "gpt-4o",
+            max_retries=1,
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.client.max_retries == 1
+    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o")
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # through the openai function with model name and config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            "gpt-4o",
+            OpenAIConfig(stop=["."]),
+            max_retries=1,
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.client.max_retries == 1
+    assert model.legacy_instance.config == OpenAIConfig(
+        model="gpt-4o", stop=["."]
+    )
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # directly through the OpenAI class with client, config and system prompt
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.OpenAI(
+            AsyncOpenAI(),
+            OpenAIConfig(model="gpt-4o", n=1),
+            system_prompt="You are a helpful assistant.",
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o", n=1)
+    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+
+def test_azure_openai_legacy_init():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.azure_openai(
+            "foo",
+            max_retries=1,
+            base_url="https://foo.openai.azure.com/",
+            api_version="2023-03-15-preview",
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncAzureOpenAI)
+    assert model.legacy_instance.client.max_retries == 1
+    assert model.legacy_instance.config == OpenAIConfig(model="foo")
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+
+@pytest.mark.api_call
+def test_openai_legacy_call_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            "gpt-4o",
+            max_retries=1,
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(model)
+    result = generator(
+        "Hello, world!",
+        10,
+        ["."],
+        system_prompt="You are a helpful assistant.",
+        temperature=0.5,
+        samples=1
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_legacy.py b/tests/v0_legacy/models/test_transformers_legacy.py
new file mode 100644
index 000000000..2c34c2517
--- /dev/null
+++ b/tests/v0_legacy/models/test_transformers_legacy.py
@@ -0,0 +1,172 @@
+from dataclasses import asdict
+
+import pytest
+from transformers import (
+    GPT2LMHeadModel,
+    GPT2TokenizerFast,
+    BertForSequenceClassification,
+    BertTokenizer,
+    MambaForCausalLM,
+)
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.models.transformers import transformers_params_adapter
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+
+def test_transformers_legacy_init():
+    # Test with default model and tokenizer classes
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="erwanf/gpt2-mini",
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, GPT2LMHeadModel)
+    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
+
+    # Test with non default model and tokenizer classes
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="bert-base-uncased",
+            model_class=BertForSequenceClassification,
+            tokenizer_class=BertTokenizer,
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, BertForSequenceClassification)
+    assert isinstance(model.tokenizer.tokenizer, BertTokenizer)
+
+    # Test with model and tokenizer kwargs and device
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="erwanf/gpt2-mini",
+            model_kwargs={"use_cache": False},
+            tokenizer_kwargs={"model_max_length": 512},
+            device="cpu",
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, GPT2LMHeadModel)
+    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
+    assert model.tokenizer.tokenizer.model_max_length == 512
+    assert not model.model.config.use_cache
+    assert model.model.device.type == "cpu"
+
+
+def test_mamba_legacy_init():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mamba` function is deprecated",
+    ):
+        model = models.mamba(
+            model_name="erwanf/gpt2-mini",
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, MambaForCausalLM)
+    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
+
+
+def test_transformers_legacy_params_adapter_generation_params():
+    result = transformers_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert not result.get("stop")
+    assert not result.get("seed")
+    assert result.get("max_new_tokens") == 2**30
+
+    result = transformers_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
+        {}
+    )
+    assert result.get("stop_strings") == "foo"
+    assert not result.get("seed")
+    assert result.get("max_new_tokens") == 10
+
+
+def test_transformers_legacy_params_adapter_sampling_params():
+    result = transformers_params_adapter(
+        {
+            "sampler": "multinomial",
+            "num_samples": None,
+            "top_p": None,
+            "top_k": None,
+            "temperature": None,
+        },
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {},
+    )
+    assert result.get("do_sample")
+    assert result.get("num_return_sequences") == 1
+    assert result.get("top_p") is None
+    assert result.get("top_k") is None
+    assert result.get("temperature") is None
+
+    result = transformers_params_adapter(
+        {
+            "sampler": "beam_search",
+            "num_samples": 2,
+            "top_p": 2,
+            "top_k": 0,
+            "temperature": 0.5,
+        },
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {},
+    )
+    assert result.get("num_beams") == 2
+    assert result.get("num_return_sequences") == 2
+    assert result.get("top_p") == 2
+    assert result.get("top_k") == 0
+    assert result.get("temperature") == 0.5
+
+
+def test_transformers_legacy_params_adapter_model_specific_params():
+    result = transformers_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {"length_penalty": 1.0}
+    )
+    assert result.get("length_penalty") == 1
+    assert not result.get("foo")
+
+
+def test_transformers_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="erwanf/gpt2-mini",
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                samples=1,
+                top_p=0.9,
+                top_k=10,
+                temperature=0.7
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!",
+        10,
+        "foo",
+        2,
+        length_penalty=0.5,
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_vision_legacy.py b/tests/v0_legacy/models/test_transformers_vision_legacy.py
new file mode 100644
index 000000000..bcaf4ea3e
--- /dev/null
+++ b/tests/v0_legacy/models/test_transformers_vision_legacy.py
@@ -0,0 +1,110 @@
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image
+from transformers import (
+    AutoProcessor,
+    LlavaForConditionalGeneration,
+    LlamaTokenizer,
+    LlamaTokenizerFast,
+    LlavaProcessor,
+)
+
+from outlines import models, generate, samplers
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+
+def test_transformers_vision_legacy_init():
+    # Test with only model name and model class
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+        )
+    assert isinstance(model, models.TransformersVision)
+    assert isinstance(model.model, LlavaForConditionalGeneration)
+    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
+    assert isinstance(model.processor, LlavaProcessor)
+
+    # Test with model, tokenizer, and processor classes
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+            tokenizer_class=LlamaTokenizer,
+            processor_class=AutoProcessor,
+        )
+    assert isinstance(model, models.TransformersVision)
+    assert isinstance(model.model, LlavaForConditionalGeneration)
+    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizer)
+    assert isinstance(model.processor, LlavaProcessor)
+
+    # Test with model, tokenizer, processor kwargs, and device
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+            model_kwargs={"output_attentions": True},
+            processor_kwargs={"num_additional_image_tokens": 1},
+            device="cpu",
+        )
+    assert isinstance(model, models.TransformersVision)
+    assert isinstance(model.model, LlavaForConditionalGeneration)
+    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
+    assert isinstance(model.processor, LlavaProcessor)
+    assert model.model.config.output_attentions
+    assert model.processor.num_additional_image_tokens == 1
+    assert model.model.device.type == "cpu"
+
+
+def test_transformers_vision_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                samples=1,
+                top_p=0.9,
+                top_k=10,
+                temperature=0.7
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!<image>",
+        img_from_url(
+            "https://upload.wikimedia.org/wikipedia/commons"
+            + "/2/25/Siam_lilacpoint.jpg"
+        ),
+        10,
+        "foo",
+        2,
+        length_penalty=0.5,
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_vllm_offline_legacy.py b/tests/v0_legacy/models/test_vllm_offline_legacy.py
new file mode 100644
index 000000000..318eeddb7
--- /dev/null
+++ b/tests/v0_legacy/models/test_vllm_offline_legacy.py
@@ -0,0 +1,181 @@
+import warnings
+from dataclasses import asdict
+
+import pytest
+
+try:
+    from vllm import LLM
+    HAS_VLLM = True
+except ImportError:
+    HAS_VLLM = False
+    warnings.warn("Skipping vLLM legacy tests because vLLM is not available.")
+
+from outlines import generate, models, samplers
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
+
+pytestmark = pytest.mark.skipif(
+    not HAS_VLLM,
+    reason="vLLM models can only be run on GPU."
+)
+
+
+@pytest.fixture
+def model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `vllm` function is deprecated",
+    ):
+        model_instance = models.vllm(
+            "erwanf/gpt2-mini",
+            seed=1,
+        )
+    return model_instance
+
+
+def test_vllm_legacy_init(model):
+    assert isinstance(model, models.VLLM)
+    assert isinstance(model.model, LLM)
+
+
+def test_vllm_legacy_params_adapter_generation_params():
+    # default
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].stop_strings is None
+    assert result["sampling_params"].seed is None
+    assert result["sampling_params"].max_tokens is None
+
+    # max_tokens
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": 2},
+        {}
+    )
+    assert result["sampling_params"].max_tokens == 2
+
+    # seed
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": 2, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].seed == 2
+
+    # stop_at str
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].stop_strings == ["foo"]
+
+    # stop_at list
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": ["foo", "bar"], "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].stop_strings == ["foo", "bar"]
+
+
+def test_vllm_legacy_params_adapter_sampling_params():
+    # default
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].n == 1
+    assert result["sampling_params"].top_p == 1.0
+    assert result["sampling_params"].top_k == -1
+    assert result["sampling_params"].temperature == 1.0
+
+    # beam search
+    result = vllm_offline_params_adapter(
+        asdict(samplers.beam_search().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].use_beam_search is True
+
+    # num_samples
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial(samples=2).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].n == 2
+    assert result["sampling_params"].best_of == 2
+
+    # top_p
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial(top_p=0.5).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].top_p == 0.5
+
+    # top_k
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial(top_k=2).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].top_k == 2
+
+    # temperature
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial(temperature=0.5).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].temperature == 0.5
+
+
+def test_vllm_legacy_params_adapter_model_specific_params():
+    result = vllm_offline_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {"use_tqdm": False}
+    )
+    assert not result.get("use_tqdm")
+
+
+def test_vllm_legacy_call_generation(model):
+    # default
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial()
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator("Hello, world!")
+    assert isinstance(result, str)
+
+    # with args
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                samples=2, top_p=0.5, top_k=2, temperature=0.5
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!", 10, stop_at="foo", seed=1, use_tqdm=False
+    )
+    assert isinstance(result, list)
+    for completion in result:
+        assert isinstance(completion, str)
diff --git a/tests/test_base.py b/tests/v0_legacy/test_base_legacy.py
similarity index 99%
rename from tests/test_base.py
rename to tests/v0_legacy/test_base_legacy.py
index 24c84f7d9..f445b4f5d 100644
--- a/tests/test_base.py
+++ b/tests/v0_legacy/test_base_legacy.py
@@ -2,7 +2,7 @@
 import pytest
 from numpy.testing import assert_array_equal
 
-from outlines.base import vectorize
+from outlines.v0_legacy.base import vectorize
 
 
 def test_vectorize_docstring():
diff --git a/tests/test_function.py b/tests/v0_legacy/test_function_legacy.py
similarity index 59%
rename from tests/test_function.py
rename to tests/v0_legacy/test_function_legacy.py
index 8bb12976e..1fd03ebb9 100644
--- a/tests/test_function.py
+++ b/tests/v0_legacy/test_function_legacy.py
@@ -4,25 +4,76 @@
 from requests.exceptions import HTTPError
 
 import outlines
-from outlines.function import Function, download_from_github, extract_function_from_file
+from outlines.v0_legacy.function import (
+    Function,
+    download_from_github,
+    extract_function_from_file,
+)
 
 
 def test_function_basic():
-    with pytest.deprecated_call(match="The @prompt decorator"):
+    with pytest.warns(DeprecationWarning, match="The @prompt decorator"):
+        with pytest.warns(
+            DeprecationWarning, match="The `Function` class is deprecated"
+        ):
+            @outlines.prompt
+            def test_template(text: str):
+                """{{ text }}"""
+
+            class Foo(BaseModel):
+                id: int
 
-        @outlines.prompt
-        def test_template(text: str):
-            """{{ text }}"""
+            fn = Function(
+                test_template,
+                Foo,
+                "erwanf/gpt2-mini",
+            )
 
-    class Foo(BaseModel):
-        id: int
+            assert fn.generator is None
 
-    fn = Function(test_template, Foo, "hf-internal-testing/tiny-random-GPTJForCausalLM")
+            result = fn("test")
+            assert isinstance(result, BaseModel)
+            assert fn.generator is not None
+            fn("test")
 
-    assert fn.generator is None
 
-    result = fn("test")
-    assert isinstance(result, BaseModel)
+@responses.activate
+def test_function_from_github():
+    content = """
+import outlines
+from pydantic import BaseModel
+
+model = "gpt2"
+
+
+prompt = outlines.Template.from_string("{{ text }}")
+
+
+class User(BaseModel):
+    id: int
+    name: str
+
+
+fn = outlines.Function(
+    prompt,
+    User,
+    "gpt2",
+)
+"""
+    responses.add(
+        responses.GET,
+        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
+        body=content,
+        status=200,
+    )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `Function` class is deprecated",
+    ):
+        fn = Function.from_github("dottxt-ai/outlines/program", "fn")
+        assert (
+            str(type(fn)) == "<class 'outlines.v0_legacy.function.Function'>"
+        )
 
 
 def test_download_from_github_invalid():
@@ -103,8 +154,10 @@ class User(BaseModel):
     "gpt2",
 )
     """
-
-    with pytest.deprecated_call(match="The @prompt decorator"):
+    with pytest.raises(
+        DeprecationWarning,
+        match="The @prompt decorator is deprecated",
+    ):
         fn = extract_function_from_file(content, "function")
         assert (
             str(type(fn)) == "<class 'outlines.function.Function'>"
@@ -131,7 +184,9 @@ class User(BaseModel):
     "gpt2",
 )
     """
-
-    with pytest.deprecated_call(match="The @prompt decorator"):
+    with pytest.raises(
+        DeprecationWarning,
+        match="The @prompt decorator is deprecated",
+    ):
         with pytest.raises(AttributeError, match="Could not find"):
             extract_function_from_file(content, "function")
diff --git a/tests/v0_legacy/test_samplers_legacy.py b/tests/v0_legacy/test_samplers_legacy.py
new file mode 100644
index 000000000..a4d750116
--- /dev/null
+++ b/tests/v0_legacy/test_samplers_legacy.py
@@ -0,0 +1,60 @@
+from outlines.v0_legacy.samplers import (
+    BeamSearchSampler,
+    GreedySampler,
+    MultinomialSampler,
+    beam_search,
+    greedy,
+    multinomial,
+)
+
+
+def test_aliases():
+    assert greedy == GreedySampler
+    assert multinomial == MultinomialSampler
+    assert beam_search == BeamSearchSampler
+
+
+def test_greedy():
+    sampler = GreedySampler()
+    params = sampler.sampling_params
+    assert params.sampler == "greedy"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 0.0
+
+
+def test_multinomial():
+    sampler = MultinomialSampler()
+    params = sampler.sampling_params
+    assert params.sampler == "multinomial"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature is None
+
+    sampler = multinomial(samples=5, top_k=10, top_p=0.9, temperature=0.8)
+    params = sampler.sampling_params
+    assert params.sampler == "multinomial"
+    assert params.num_samples == 5
+    assert params.top_p == 0.9
+    assert params.top_k == 10
+    assert params.temperature == 0.8
+
+
+def test_beam_search():
+    sampler = BeamSearchSampler()
+    params = sampler.sampling_params
+    assert params.sampler == "beam_search"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 1.0
+
+    sampler = beam_search(beams=3)
+    params = sampler.sampling_params
+    assert params.sampler == "beam_search"
+    assert params.num_samples == 3
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 1.0
diff --git a/uv.lock b/uv.lock
index f1deefbe6..2d927f136 100644
--- a/uv.lock
+++ b/uv.lock
@@ -625,28 +625,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/02/cc/b7e31358aac6ed1ef2bb790a9746ac2c69bcb3c8588b41616914eb106eaf/exceptiongroup-1.2.2-py3-none-any.whl", hash = "sha256:3111b9d131c238bec2f8f516e123e14ba243563fb135d3fe885990585aa7795b", size = 16453 },
 ]
 
-[[package]]
-name = "exllamav2"
-version = "0.2.7"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "fastparquet" },
-    { name = "ninja" },
-    { name = "numpy" },
-    { name = "pandas" },
-    { name = "pillow" },
-    { name = "pygments" },
-    { name = "regex" },
-    { name = "rich" },
-    { name = "safetensors" },
-    { name = "sentencepiece" },
-    { name = "torch" },
-    { name = "websockets" },
-]
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/b5/04/300012908311c9a904d44c74a037821e7cde530c648afff4ad6bed4e22c3/exllamav2-0.2.7-py3-none-any.whl", hash = "sha256:410d1a69fed329f6f1c10de6988a3b6b33e1fa715aa4fe32c545edd1abd1254b", size = 1512625 },
-]
-
 [[package]]
 name = "fastapi"
 version = "0.115.6"
@@ -1917,9 +1895,6 @@ dependencies = [
 ]
 
 [package.optional-dependencies]
-exllamav2 = [
-    { name = "exllamav2" },
-]
 llamacpp = [
     { name = "datasets" },
     { name = "llama-cpp-python" },
@@ -1946,7 +1921,6 @@ test = [
     { name = "datasets" },
     { name = "diff-cover", version = "9.2.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.9.17'" },
     { name = "diff-cover", version = "9.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.9.17'" },
-    { name = "exllamav2" },
     { name = "huggingface-hub" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
@@ -1989,8 +1963,6 @@ requires-dist = [
     { name = "datasets", marker = "extra == 'transformers'" },
     { name = "diff-cover", marker = "extra == 'test'" },
     { name = "diskcache" },
-    { name = "exllamav2", marker = "extra == 'exllamav2'" },
-    { name = "exllamav2", marker = "extra == 'test'" },
     { name = "fastapi", marker = "extra == 'serve'" },
     { name = "genson" },
     { name = "huggingface-hub", marker = "extra == 'test'" },

From 73f2ba999c9e9b54f1185b46de22771ce73ee300 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Wed, 18 Jun 2025 15:41:44 +0200
Subject: [PATCH 193/505] Revert "Merge v1.0 into main (#1619)"

This reverts commit 7fc5f1cf031c1e31bbc69cf89dfc572625e107cf.
---
 .github/workflows/tests.yml                   |  29 +-
 .github/workflows/tests_nightly.yml           |  60 --
 Dockerfile                                    |  23 +
 README.md                                     | 802 +++++++-----------
 benchmarks/bench_cfg_guide.py                 |   2 +-
 benchmarks/bench_json_schema.py               |   2 +-
 benchmarks/bench_regex_guide.py               |   2 +-
 docs/api/guide.md                             |   1 +
 docs/{api_reference => api}/index.md          |   0
 docs/api/json_schema.md                       |   1 +
 docs/api/models.md                            |   1 +
 docs/api/parsing.md                           |   1 +
 docs/api/regex.md                             |   1 +
 docs/api/samplers.md                          |   1 +
 docs/api/templates.md                         |   1 +
 docs/assets/images/install.png                | Bin 12669 -> 0 bytes
 docs/assets/images/logo-square.png            | Bin 13864 -> 3323 bytes
 docs/assets/images/logo-square.svg            | 192 ++---
 docs/assets/images/readme-dark.png            | Bin 38912 -> 0 bytes
 docs/assets/images/readme-light.png           | Bin 38943 -> 0 bytes
 docs/assets/images/use_philosophy.png         | Bin 16463 -> 0 bytes
 docs/blog/.authors.yml                        |   5 +
 docs/blog/assets/4000_stars.png               | Bin 0 -> 57307 bytes
 docs/blog/posts/roadmap-2024.md               |  91 ++
 docs/community/belonging.png                  | Bin 0 -> 142520 bytes
 docs/community/index.md                       |   3 +
 docs/community/versioning.md                  |   4 -
 docs/{examples => cookbook}/atomic_caption.md |  53 +-
 .../chain_of_density.md                       |  51 +-
 .../chain_of_thought.md                       |  87 +-
 docs/cookbook/classification.md               | 106 +++
 .../{examples => cookbook}/dating_profiles.md |  30 +-
 .../deploy-using-bentoml.md                   |  22 +-
 .../deploy-using-cerebrium.md                 |  11 +-
 .../deploy-using-modal.md                     |  16 +-
 .../earnings-reports.md                       |  24 +-
 .../extract_event_details.md                  |   2 +-
 docs/cookbook/extract_event_details.py        |  46 +
 docs/{examples => cookbook}/extraction.md     |  16 +-
 .../images/chain_of_density.png               | Bin
 .../images/coding_structure_diagram.png       | Bin
 .../images/knowledge-graph-extraction.png     | Bin
 .../images/nvidia-income.png                  | Bin
 docs/{examples => cookbook}/images/simtom.png | Bin
 .../images/trader-joes-receipt.jpg            | Bin
 docs/{examples => cookbook}/index.md          |  14 +-
 .../knowledge_graph_extraction.md             | 102 +--
 .../models_playing_chess.md                   |  28 +-
 .../qa-with-citations.md                      | 125 ++-
 docs/{examples => cookbook}/react_agent.md    |  90 +-
 docs/{examples => cookbook}/read-pdfs.md      |  79 +-
 .../receipt-digitization.md                   |  99 ++-
 docs/{examples => cookbook}/simtom.md         |  52 +-
 .../structured_generation_workflow.md         |  62 +-
 docs/core_concepts.md                         |  13 -
 docs/examples/classification.md               |  85 --
 docs/examples/extract_event_details.py        |  54 --
 .../prompt_templates/chain_of_density.txt     |  27 -
 .../prompt_templates/classification.txt       |  18 -
 .../examples/prompt_templates/react_agent.txt |  27 -
 .../simtom_prospective_taking.txt             |  10 -
 .../prompt_templates/simtom_simulation.txt    |   7 -
 docs/features/advanced/logits_processors.md   |  97 ---
 docs/features/core/generator.md               | 122 ---
 docs/features/core/output_types.md            | 205 -----
 docs/features/index.md                        |  19 -
 docs/features/models/anthropic.md             |  98 ---
 docs/features/models/dottxt.md                |  74 --
 docs/features/models/gemini.md                | 148 ----
 docs/features/models/index.md                 | 107 ---
 docs/features/models/llamacpp.md              | 168 ----
 docs/features/models/mlxlm.md                 | 189 -----
 docs/features/models/ollama.md                | 103 ---
 docs/features/models/openai.md                | 134 ---
 docs/features/models/openai_compatible.md     |  87 --
 docs/features/models/sglang.md                | 171 ----
 docs/features/models/tgi.md                   | 137 ---
 docs/features/models/transformers.md          | 207 -----
 .../models/transformers_multimodal.md         |  89 --
 docs/features/models/vllm.md                  | 186 ----
 docs/features/models/vllm_offline.md          | 176 ----
 docs/features/utility/application.md          |  65 --
 docs/features/utility/template.md             |  85 --
 docs/guide/core_concepts.md                   |  13 -
 docs/guide/getting_started.md                 | 363 --------
 docs/guide/installation.md                    |  73 --
 docs/guide/migration.md                       | 246 ------
 docs/guide/selecting_an_inference_backend.md  |   5 -
 docs/index.md                                 | 162 +---
 docs/installation.md                          |  49 ++
 docs/logos/cohere.png                         | Bin 113536 -> 0 bytes
 docs/logos/vllm.png                           | Bin 89976 -> 0 bytes
 docs/quickstart.md                            | 228 +++++
 docs/{guide => reference}/chat_templating.md  |   0
 docs/reference/generation/cfg.md              | 154 ++++
 docs/reference/generation/choices.md          |  16 +
 .../reference/generation/creating_grammars.md |  99 +++
 docs/reference/generation/custom_fsm_ops.md   |  37 +
 docs/reference/generation/format.md           |  23 +
 docs/reference/generation/generation.md       | 216 +++++
 docs/reference/generation/json.md             | 178 ++++
 docs/reference/generation/regex.md            |  27 +
 .../structured_generation_explanation.md      |  54 ++
 docs/reference/generation/types.md            |  83 ++
 docs/reference/index.md                       |  15 +
 docs/reference/models/exllamav2.md            |  15 +
 docs/reference/models/llamacpp.md             | 230 +++++
 docs/reference/models/mlxlm.md                |  93 ++
 docs/reference/models/models.md               |  56 ++
 docs/reference/models/openai.md               | 204 +++++
 docs/reference/models/tgi.md                  |   3 +
 docs/reference/models/transformers.md         | 148 ++++
 docs/reference/models/transformers_vision.md  | 115 +++
 docs/reference/models/vllm.md                 | 235 +++++
 docs/reference/prompting.md                   | 395 +++++++++
 .../utility => reference}/regex_dsl.md        |   6 +-
 docs/reference/samplers.md                    | 178 ++++
 docs/reference/serve/lmstudio.md              |  89 ++
 docs/reference/serve/vllm.md                  |  73 ++
 docs/reference/text.md                        |  86 ++
 docs/stylesheets/extra.css                    | 161 +++-
 docs/welcome.md                               | 183 ++++
 examples/babyagi.py                           |  18 +-
 examples/beam-cloud/app.py                    |  11 +-
 examples/bentoml/service.py                   |  16 +-
 examples/cerebrium/main.py                    |  16 +-
 examples/cfg.py                               |  22 +-
 examples/dating_profile.py                    |  21 +-
 examples/llamacpp_example.py                  |   5 +-
 examples/llamacpp_processor.py                |   4 +-
 examples/math_generate_code.py                |  10 +-
 examples/meta_prompting.py                    |  48 +-
 examples/modal_example.py                     |  23 +-
 examples/parsing.py                           |   1 +
 examples/pick_odd_one_out.py                  |  47 +-
 examples/prompts/babyagi_create_task.txt      |   9 -
 examples/prompts/babyagi_perform_task.txt     |   5 -
 examples/prompts/babyagi_prioritize_task.txt  |   7 -
 examples/prompts/dating_profile.txt           |  11 -
 examples/prompts/pick_odd_one_out.txt         |   9 -
 examples/prompts/self_consistency.txt         |   6 -
 examples/react.py                             |  56 +-
 examples/sampling.ipynb                       |  48 +-
 examples/self_consistency.py                  |  23 +-
 examples/simulation_based_inference.ipynb     |  18 +-
 examples/transformers_integration.py          |  25 +
 ...ine_integration.py => vllm_integration.py} |   7 +-
 mkdocs.yml                                    | 366 ++++----
 outlines/__init__.py                          |  77 +-
 outlines/applications.py                      | 107 ---
 outlines/{v0_legacy => }/base.py              | 141 +--
 outlines/caching.py                           |  10 +-
 outlines/fsm/__init__.py                      |   1 -
 outlines/{processors => fsm}/guide.py         | 236 +-----
 outlines/fsm/json_schema.py                   |  83 ++
 outlines/fsm/parsing.py                       |   2 -
 outlines/fsm/types.py                         |  76 ++
 outlines/{v0_legacy => }/function.py          |  75 +-
 outlines/{v0_legacy => }/generate/__init__.py |  11 +-
 outlines/generate/api.py                      | 609 +++++++++++++
 outlines/generate/cfg.py                      |  54 ++
 outlines/generate/choice.py                   |  59 ++
 outlines/generate/format.py                   |  48 ++
 outlines/generate/fsm.py                      |  31 +
 outlines/generate/generator.py                | 312 +++++++
 outlines/generate/json.py                     | 123 +++
 outlines/generate/regex.py                    |  66 ++
 outlines/generate/text.py                     |  50 ++
 outlines/generator.py                         | 347 --------
 outlines/grammars.py                          |  23 +-
 outlines/models/__init__.py                   |  43 +-
 outlines/models/anthropic.py                  | 227 -----
 outlines/models/base.py                       | 377 --------
 outlines/models/dottxt.py                     | 215 -----
 outlines/{v0_legacy => }/models/exllamav2.py  | 141 ++-
 outlines/models/gemini.py                     | 296 -------
 outlines/models/llamacpp.py                   | 393 +++++----
 outlines/models/mlxlm.py                      | 340 +++++---
 outlines/models/ollama.py                     | 213 -----
 outlines/models/openai.py                     | 483 ++---------
 outlines/models/sglang.py                     | 372 --------
 outlines/models/tgi.py                        | 340 --------
 outlines/models/tokenizer.py                  |  12 +-
 outlines/models/transformers.py               | 555 ++++++------
 outlines/models/transformers_vision.py        | 138 +++
 outlines/models/vllm.py                       | 483 ++++-------
 outlines/models/vllm_offline.py               | 206 -----
 outlines/processors/__init__.py               |   2 -
 outlines/processors/base_logits_processor.py  | 246 +++---
 outlines/processors/structured.py             | 218 ++---
 .../processors/tensor_adapters/__init__.py    |  26 -
 outlines/processors/tensor_adapters/base.py   | 257 ------
 outlines/processors/tensor_adapters/jax.py    |  50 --
 outlines/processors/tensor_adapters/mlx.py    |  60 --
 outlines/processors/tensor_adapters/numpy.py  |  50 --
 .../processors/tensor_adapters/tensorflow.py  |  50 --
 outlines/processors/tensor_adapters/torch.py  |  48 --
 outlines/release_note.md                      | 392 ---------
 outlines/samplers.py                          | 353 ++++++++
 outlines/serve/__init__.py                    |   0
 outlines/serve/serve.py                       | 140 +++
 outlines/templates.py                         |  85 +-
 outlines/types/__init__.py                    |  27 +-
 outlines/types/dsl.py                         | 567 +------------
 outlines/types/locale/__init__.py             |   2 -
 outlines/types/locale/us.py                   |   2 -
 outlines/types/utils.py                       | 206 -----
 outlines/v0_legacy/generate/api.py            | 175 ----
 outlines/v0_legacy/generate/cfg.py            |  80 --
 outlines/v0_legacy/generate/choice.py         | 126 ---
 outlines/v0_legacy/generate/format.py         |  70 --
 outlines/v0_legacy/generate/fsm.py            |  77 --
 outlines/v0_legacy/generate/json.py           | 145 ----
 outlines/v0_legacy/generate/regex.py          |  67 --
 outlines/v0_legacy/generate/text.py           |  64 --
 outlines/v0_legacy/models/__init__.py         |  21 -
 outlines/v0_legacy/models/llamacpp.py         | 147 ----
 outlines/v0_legacy/models/mlxlm.py            | 153 ----
 outlines/v0_legacy/models/openai.py           |  64 --
 outlines/v0_legacy/models/transformers.py     | 196 -----
 .../v0_legacy/models/transformers_vision.py   | 107 ---
 outlines/v0_legacy/models/vllm_offline.py     | 102 ---
 outlines/v0_legacy/samplers.py                | 118 ---
 pyproject.toml                                |  77 +-
 requirements-doc.txt                          |   3 -
 scripts/gen_ref_pages.py                      |  53 --
 tests/{processors => fsm}/test_cfg_guide.py   |  15 +-
 tests/{processors => fsm}/test_guide.py       |  82 +-
 tests/fsm/test_json_schema.py                 |  78 ++
 tests/fsm/test_types.py                       |  22 +
 tests/generate/__init__.py                    |   0
 tests/generate/conftest.py                    |  66 ++
 tests/generate/test_api.py                    |  33 +
 tests/generate/test_generate.py               | 439 ++++++++++
 tests/generate/test_generator.py              | 497 +++++++++++
 tests/generate/test_integration_exllamav2.py  | 363 ++++++++
 tests/generate/test_integration_llamacpp.py   | 374 ++++++++
 .../generate/test_integration_transformers.py | 569 +++++++++++++
 .../test_integration_transformers_vision.py   | 116 +++
 tests/generate/test_integration_vllm.py       | 243 ++++++
 tests/models/test_anthropic.py                | 106 ---
 tests/models/test_dottxt.py                   | 131 ---
 tests/models/test_dottxt_type_adapter.py      | 137 ---
 tests/models/test_gemini.py                   | 221 -----
 tests/models/test_gemini_type_adapter.py      | 204 -----
 tests/models/test_llamacpp.py                 | 134 ---
 tests/models/test_llamacpp_tokenizer.py       | 133 ---
 tests/models/test_llamacpp_type_adapter.py    |  41 -
 tests/models/test_mlxlm.py                    | 133 ++-
 tests/models/test_ollama.py                   | 101 ---
 tests/models/test_ollama_type_adapter.py      | 131 ---
 tests/models/test_openai.py                   | 292 +++----
 tests/models/test_openai_type_adapter.py      | 195 -----
 tests/models/test_sglang.py                   | 272 ------
 tests/models/test_sglang_type_adapter.py      | 145 ----
 tests/models/test_tgi.py                      | 180 ----
 tests/models/test_tgi_model_adapter.py        |  93 --
 tests/models/test_transformers.py             | 283 +++---
 tests/models/test_transformers_multimodal.py  | 194 -----
 ...st_transformers_multimodal_type_adapter.py |  48 --
 tests/models/test_transformers_tokenizer.py   | 142 ----
 .../models/test_transformers_type_adapter.py  |  50 --
 tests/models/test_vllm.py                     | 254 ------
 tests/models/test_vllm_offline.py             | 117 ---
 .../models/test_vllm_offline_type_adapter.py  |  84 --
 tests/models/test_vllm_type_adapter.py        | 121 ---
 tests/processors/test_base_processor.py       | 148 ++--
 tests/processors/test_structured.py           | 135 ---
 tests/processors/test_tensor_adapters.py      | 276 ------
 tests/test_applications.py                    |  93 --
 .../test_base_legacy.py => test_base.py}      |   2 +-
 tests/test_cache.py                           |  31 -
 ...st_function_legacy.py => test_function.py} |  85 +-
 tests/test_generator.py                       | 255 ------
 tests/test_grammars.py                        |   2 +-
 tests/test_samplers.py                        | 277 ++++++
 tests/test_templates.py                       | 122 +--
 tests/test_utils/mock_openai_client.py        | 117 ---
 tests/test_utils/mock_tgi_client.py           |  63 --
 tests/test_utils/utils.py                     |  23 -
 tests/types/test_custom_types.py              |   1 +
 tests/types/test_dsl.py                       | 508 +----------
 tests/types/test_types_utils.py               | 409 ---------
 tests/v0_legacy/generate/test_api_legacy.py   | 186 ----
 tests/v0_legacy/generate/test_cfg_legacy.py   | 111 ---
 .../v0_legacy/generate/test_choice_legacy.py  |  97 ---
 .../v0_legacy/generate/test_format_legacy.py  |  93 --
 tests/v0_legacy/generate/test_fsm_legacy.py   | 104 ---
 tests/v0_legacy/generate/test_json_legacy.py  | 200 -----
 tests/v0_legacy/generate/test_regex_legacy.py |  90 --
 tests/v0_legacy/generate/test_text_legacy.py  |  83 --
 .../v0_legacy/models/test_exllamav2_legacy.py |  76 --
 .../v0_legacy/models/test_llamacpp_legacy.py  | 169 ----
 tests/v0_legacy/models/test_mlxlm_legacy.py   | 193 -----
 tests/v0_legacy/models/test_openai_legacy.py  | 159 ----
 .../models/test_transformers_legacy.py        | 172 ----
 .../models/test_transformers_vision_legacy.py | 110 ---
 .../models/test_vllm_offline_legacy.py        | 181 ----
 tests/v0_legacy/test_samplers_legacy.py       |  60 --
 uv.lock                                       |  28 +
 300 files changed, 12037 insertions(+), 21850 deletions(-)
 delete mode 100644 .github/workflows/tests_nightly.yml
 create mode 100644 Dockerfile
 create mode 100644 docs/api/guide.md
 rename docs/{api_reference => api}/index.md (100%)
 create mode 100644 docs/api/json_schema.md
 create mode 100644 docs/api/models.md
 create mode 100644 docs/api/parsing.md
 create mode 100644 docs/api/regex.md
 create mode 100644 docs/api/samplers.md
 create mode 100644 docs/api/templates.md
 delete mode 100644 docs/assets/images/install.png
 delete mode 100644 docs/assets/images/readme-dark.png
 delete mode 100644 docs/assets/images/readme-light.png
 delete mode 100644 docs/assets/images/use_philosophy.png
 create mode 100644 docs/blog/.authors.yml
 create mode 100644 docs/blog/assets/4000_stars.png
 create mode 100644 docs/blog/posts/roadmap-2024.md
 create mode 100644 docs/community/belonging.png
 rename docs/{examples => cookbook}/atomic_caption.md (75%)
 rename docs/{examples => cookbook}/chain_of_density.md (74%)
 rename docs/{examples => cookbook}/chain_of_thought.md (63%)
 create mode 100644 docs/cookbook/classification.md
 rename docs/{examples => cookbook}/dating_profiles.md (92%)
 rename docs/{examples => cookbook}/deploy-using-bentoml.md (91%)
 rename docs/{examples => cookbook}/deploy-using-cerebrium.md (90%)
 rename docs/{examples => cookbook}/deploy-using-modal.md (93%)
 rename docs/{examples => cookbook}/earnings-reports.md (94%)
 rename docs/{examples => cookbook}/extract_event_details.md (94%)
 create mode 100644 docs/cookbook/extract_event_details.py
 rename docs/{examples => cookbook}/extraction.md (85%)
 rename docs/{examples => cookbook}/images/chain_of_density.png (100%)
 rename docs/{examples => cookbook}/images/coding_structure_diagram.png (100%)
 rename docs/{examples => cookbook}/images/knowledge-graph-extraction.png (100%)
 rename docs/{examples => cookbook}/images/nvidia-income.png (100%)
 rename docs/{examples => cookbook}/images/simtom.png (100%)
 rename docs/{examples => cookbook}/images/trader-joes-receipt.jpg (100%)
 rename docs/{examples => cookbook}/index.md (82%)
 rename docs/{examples => cookbook}/knowledge_graph_extraction.md (60%)
 rename docs/{examples => cookbook}/models_playing_chess.md (77%)
 rename docs/{examples => cookbook}/qa-with-citations.md (68%)
 rename docs/{examples => cookbook}/react_agent.md (71%)
 rename docs/{examples => cookbook}/read-pdfs.md (79%)
 rename docs/{examples => cookbook}/receipt-digitization.md (71%)
 rename docs/{examples => cookbook}/simtom.md (68%)
 rename docs/{examples => cookbook}/structured_generation_workflow.md (79%)
 delete mode 100644 docs/core_concepts.md
 delete mode 100644 docs/examples/classification.md
 delete mode 100644 docs/examples/extract_event_details.py
 delete mode 100644 docs/examples/prompt_templates/chain_of_density.txt
 delete mode 100644 docs/examples/prompt_templates/classification.txt
 delete mode 100644 docs/examples/prompt_templates/react_agent.txt
 delete mode 100644 docs/examples/prompt_templates/simtom_prospective_taking.txt
 delete mode 100644 docs/examples/prompt_templates/simtom_simulation.txt
 delete mode 100644 docs/features/advanced/logits_processors.md
 delete mode 100644 docs/features/core/generator.md
 delete mode 100644 docs/features/core/output_types.md
 delete mode 100644 docs/features/index.md
 delete mode 100644 docs/features/models/anthropic.md
 delete mode 100644 docs/features/models/dottxt.md
 delete mode 100644 docs/features/models/gemini.md
 delete mode 100644 docs/features/models/index.md
 delete mode 100644 docs/features/models/llamacpp.md
 delete mode 100644 docs/features/models/mlxlm.md
 delete mode 100644 docs/features/models/ollama.md
 delete mode 100644 docs/features/models/openai.md
 delete mode 100644 docs/features/models/openai_compatible.md
 delete mode 100644 docs/features/models/sglang.md
 delete mode 100644 docs/features/models/tgi.md
 delete mode 100644 docs/features/models/transformers.md
 delete mode 100644 docs/features/models/transformers_multimodal.md
 delete mode 100644 docs/features/models/vllm.md
 delete mode 100644 docs/features/models/vllm_offline.md
 delete mode 100644 docs/features/utility/application.md
 delete mode 100644 docs/features/utility/template.md
 delete mode 100644 docs/guide/core_concepts.md
 delete mode 100644 docs/guide/getting_started.md
 delete mode 100644 docs/guide/installation.md
 delete mode 100644 docs/guide/migration.md
 delete mode 100644 docs/guide/selecting_an_inference_backend.md
 create mode 100644 docs/installation.md
 delete mode 100644 docs/logos/cohere.png
 delete mode 100644 docs/logos/vllm.png
 create mode 100644 docs/quickstart.md
 rename docs/{guide => reference}/chat_templating.md (100%)
 create mode 100644 docs/reference/generation/cfg.md
 create mode 100644 docs/reference/generation/choices.md
 create mode 100644 docs/reference/generation/creating_grammars.md
 create mode 100644 docs/reference/generation/custom_fsm_ops.md
 create mode 100644 docs/reference/generation/format.md
 create mode 100644 docs/reference/generation/generation.md
 create mode 100644 docs/reference/generation/json.md
 create mode 100644 docs/reference/generation/regex.md
 create mode 100644 docs/reference/generation/structured_generation_explanation.md
 create mode 100644 docs/reference/generation/types.md
 create mode 100644 docs/reference/index.md
 create mode 100644 docs/reference/models/exllamav2.md
 create mode 100644 docs/reference/models/llamacpp.md
 create mode 100644 docs/reference/models/mlxlm.md
 create mode 100644 docs/reference/models/models.md
 create mode 100644 docs/reference/models/openai.md
 create mode 100644 docs/reference/models/tgi.md
 create mode 100644 docs/reference/models/transformers.md
 create mode 100644 docs/reference/models/transformers_vision.md
 create mode 100644 docs/reference/models/vllm.md
 create mode 100644 docs/reference/prompting.md
 rename docs/{features/utility => reference}/regex_dsl.md (99%)
 create mode 100644 docs/reference/samplers.md
 create mode 100644 docs/reference/serve/lmstudio.md
 create mode 100644 docs/reference/serve/vllm.md
 create mode 100644 docs/reference/text.md
 create mode 100644 docs/welcome.md
 delete mode 100644 examples/prompts/babyagi_create_task.txt
 delete mode 100644 examples/prompts/babyagi_perform_task.txt
 delete mode 100644 examples/prompts/babyagi_prioritize_task.txt
 delete mode 100644 examples/prompts/dating_profile.txt
 delete mode 100644 examples/prompts/pick_odd_one_out.txt
 delete mode 100644 examples/prompts/self_consistency.txt
 create mode 100644 examples/transformers_integration.py
 rename examples/{vllm_offline_integration.py => vllm_integration.py} (80%)
 delete mode 100644 outlines/applications.py
 rename outlines/{v0_legacy => }/base.py (67%)
 rename outlines/{processors => fsm}/guide.py (58%)
 create mode 100644 outlines/fsm/json_schema.py
 create mode 100644 outlines/fsm/types.py
 rename outlines/{v0_legacy => }/function.py (58%)
 rename outlines/{v0_legacy => }/generate/__init__.py (62%)
 create mode 100644 outlines/generate/api.py
 create mode 100644 outlines/generate/cfg.py
 create mode 100644 outlines/generate/choice.py
 create mode 100644 outlines/generate/format.py
 create mode 100644 outlines/generate/fsm.py
 create mode 100644 outlines/generate/generator.py
 create mode 100644 outlines/generate/json.py
 create mode 100644 outlines/generate/regex.py
 create mode 100644 outlines/generate/text.py
 delete mode 100644 outlines/generator.py
 delete mode 100644 outlines/models/anthropic.py
 delete mode 100644 outlines/models/base.py
 delete mode 100644 outlines/models/dottxt.py
 rename outlines/{v0_legacy => }/models/exllamav2.py (79%)
 delete mode 100644 outlines/models/gemini.py
 delete mode 100644 outlines/models/ollama.py
 delete mode 100644 outlines/models/sglang.py
 delete mode 100644 outlines/models/tgi.py
 create mode 100644 outlines/models/transformers_vision.py
 delete mode 100644 outlines/models/vllm_offline.py
 delete mode 100644 outlines/processors/tensor_adapters/__init__.py
 delete mode 100644 outlines/processors/tensor_adapters/base.py
 delete mode 100644 outlines/processors/tensor_adapters/jax.py
 delete mode 100644 outlines/processors/tensor_adapters/mlx.py
 delete mode 100644 outlines/processors/tensor_adapters/numpy.py
 delete mode 100644 outlines/processors/tensor_adapters/tensorflow.py
 delete mode 100644 outlines/processors/tensor_adapters/torch.py
 delete mode 100644 outlines/release_note.md
 create mode 100644 outlines/samplers.py
 create mode 100644 outlines/serve/__init__.py
 create mode 100644 outlines/serve/serve.py
 delete mode 100644 outlines/types/utils.py
 delete mode 100644 outlines/v0_legacy/generate/api.py
 delete mode 100644 outlines/v0_legacy/generate/cfg.py
 delete mode 100644 outlines/v0_legacy/generate/choice.py
 delete mode 100644 outlines/v0_legacy/generate/format.py
 delete mode 100644 outlines/v0_legacy/generate/fsm.py
 delete mode 100644 outlines/v0_legacy/generate/json.py
 delete mode 100644 outlines/v0_legacy/generate/regex.py
 delete mode 100644 outlines/v0_legacy/generate/text.py
 delete mode 100644 outlines/v0_legacy/models/__init__.py
 delete mode 100644 outlines/v0_legacy/models/llamacpp.py
 delete mode 100644 outlines/v0_legacy/models/mlxlm.py
 delete mode 100644 outlines/v0_legacy/models/openai.py
 delete mode 100644 outlines/v0_legacy/models/transformers.py
 delete mode 100644 outlines/v0_legacy/models/transformers_vision.py
 delete mode 100644 outlines/v0_legacy/models/vllm_offline.py
 delete mode 100644 outlines/v0_legacy/samplers.py
 delete mode 100644 scripts/gen_ref_pages.py
 rename tests/{processors => fsm}/test_cfg_guide.py (97%)
 rename tests/{processors => fsm}/test_guide.py (85%)
 create mode 100644 tests/fsm/test_json_schema.py
 create mode 100644 tests/fsm/test_types.py
 create mode 100644 tests/generate/__init__.py
 create mode 100644 tests/generate/conftest.py
 create mode 100644 tests/generate/test_api.py
 create mode 100644 tests/generate/test_generate.py
 create mode 100644 tests/generate/test_generator.py
 create mode 100644 tests/generate/test_integration_exllamav2.py
 create mode 100644 tests/generate/test_integration_llamacpp.py
 create mode 100644 tests/generate/test_integration_transformers.py
 create mode 100644 tests/generate/test_integration_transformers_vision.py
 create mode 100644 tests/generate/test_integration_vllm.py
 delete mode 100644 tests/models/test_anthropic.py
 delete mode 100644 tests/models/test_dottxt.py
 delete mode 100644 tests/models/test_dottxt_type_adapter.py
 delete mode 100644 tests/models/test_gemini.py
 delete mode 100644 tests/models/test_gemini_type_adapter.py
 delete mode 100644 tests/models/test_llamacpp.py
 delete mode 100644 tests/models/test_llamacpp_tokenizer.py
 delete mode 100644 tests/models/test_llamacpp_type_adapter.py
 delete mode 100644 tests/models/test_ollama.py
 delete mode 100644 tests/models/test_ollama_type_adapter.py
 delete mode 100644 tests/models/test_openai_type_adapter.py
 delete mode 100644 tests/models/test_sglang.py
 delete mode 100644 tests/models/test_sglang_type_adapter.py
 delete mode 100644 tests/models/test_tgi.py
 delete mode 100644 tests/models/test_tgi_model_adapter.py
 delete mode 100644 tests/models/test_transformers_multimodal.py
 delete mode 100644 tests/models/test_transformers_multimodal_type_adapter.py
 delete mode 100644 tests/models/test_transformers_tokenizer.py
 delete mode 100644 tests/models/test_transformers_type_adapter.py
 delete mode 100644 tests/models/test_vllm.py
 delete mode 100644 tests/models/test_vllm_offline.py
 delete mode 100644 tests/models/test_vllm_offline_type_adapter.py
 delete mode 100644 tests/models/test_vllm_type_adapter.py
 delete mode 100644 tests/processors/test_structured.py
 delete mode 100644 tests/processors/test_tensor_adapters.py
 delete mode 100644 tests/test_applications.py
 rename tests/{v0_legacy/test_base_legacy.py => test_base.py} (99%)
 rename tests/{v0_legacy/test_function_legacy.py => test_function.py} (59%)
 delete mode 100644 tests/test_generator.py
 create mode 100644 tests/test_samplers.py
 delete mode 100644 tests/test_utils/mock_openai_client.py
 delete mode 100644 tests/test_utils/mock_tgi_client.py
 delete mode 100644 tests/test_utils/utils.py
 delete mode 100644 tests/types/test_types_utils.py
 delete mode 100644 tests/v0_legacy/generate/test_api_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_cfg_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_choice_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_format_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_fsm_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_json_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_regex_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_text_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_exllamav2_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_llamacpp_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_mlxlm_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_openai_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_transformers_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_transformers_vision_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_vllm_offline_legacy.py
 delete mode 100644 tests/v0_legacy/test_samplers_legacy.py

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index c6e37d684..46d3d5ff6 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -2,7 +2,7 @@ name: Tests
 
 on:
   pull_request:
-    branches: [main,v1.0]
+    branches: [main]
   push:
     branches: [main]
 
@@ -29,11 +29,6 @@ jobs:
       uses: actions/setup-python@v4
       with:
         python-version: ${{ matrix.python-version }}
-    - name: Install Ollama
-      run: |
-        curl -fsSL https://ollama.com/install.sh | sh
-        ollama --version
-        ollama pull tinyllama
     - name: Set up test environment
       run: |
         python -m pip install --upgrade pip
@@ -51,7 +46,7 @@ jobs:
         echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
-        uv run pytest -x --cov=outlines -m 'not api_call'
+        uv run pytest -x --cov=outlines
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
     - name: Upload coverage data
@@ -87,30 +82,16 @@ jobs:
         with:
           name: coverage-data
 
-      - name: Determine base branch for comparison
-        id: base-branch
-        run: |
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            # For PRs, use the remote tracking branch
-            COMPARE_BRANCH="origin/${{ github.base_ref }}"
-            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
-          else
-            # For push events, compare against the parent commit
-            COMPARE_BRANCH="${{ github.event.before }}"
-            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
-          fi
-          echo "Using $COMPARE_BRANCH for coverage comparison"
-
-      - name: Fetch base branch for coverage diff
+      - name: Fetch main for coverage diff
         run: |
-          git fetch --no-tags --prune origin ${COMPARE_BRANCH#origin/}
+          git fetch --no-tags --prune origin main
 
       - name: Combine coverage & fail if it's <100%.
         run: |
           python -m coverage combine
           python -m coverage html --skip-covered --skip-empty
           python -m coverage xml
-          diff-cover coverage.xml --compare-branch=$COMPARE_BRANCH --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
+          diff-cover coverage.xml --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
 
       - name: Upload HTML report if check failed.
         uses: actions/upload-artifact@v4
diff --git a/.github/workflows/tests_nightly.yml b/.github/workflows/tests_nightly.yml
deleted file mode 100644
index c34a707fa..000000000
--- a/.github/workflows/tests_nightly.yml
+++ /dev/null
@@ -1,60 +0,0 @@
-name: Nightly Tests
-
-# Run every day at 6:00 AM UTC
-on:
-  schedule:
-    - cron: "0 6 * * *"
-
-jobs:
-  tests:
-    name: Run the tests
-    runs-on: ubuntu-latest
-    env:
-      OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
-      ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
-      GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
-      DOTTXT_API_KEY: ${{ secrets.DOTTXT_API_KEY }}
-    strategy:
-      fail-fast: false
-      matrix:
-        python-version: ["3.10"]
-
-    steps:
-    - uses: actions/checkout@v3
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v4
-      with:
-        python-version: ${{ matrix.python-version }}
-    - name: Install Ollama
-      run: |
-        curl -fsSL https://ollama.com/install.sh | sh
-        ollama --version
-        ollama pull tinyllama
-    - name: Set up test environment
-      run: |
-        python -m pip install --upgrade pip
-        pip install uv
-        uv venv
-        uv pip install -e .[test]
-    - name: Create matrix id
-      id: matrix-id
-      env:
-        MATRIX_CONTEXT: ${{ toJson(matrix) }}
-      run: |
-        echo $MATRIX_CONTEXT
-        export MATRIX_ID=`echo $MATRIX_CONTEXT | md5sum | cut -c 1-32`
-        echo $MATRIX_ID
-        echo "::set-output name=id::$MATRIX_ID"
-    - name: Run tests
-      run: |
-        uv run pytest -x --cov=outlines
-      env:
-        COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
-    - name: Slack notification on failure
-      if: failure()
-      uses: 8398a7/action-slack@v3
-      with:
-        status: ${{ job.status }}
-        fields: commit,author,ref
-      env:
-        SLACK_WEBHOOK_URL: ${{ secrets.SLACK_TESTS_ERRORS_WEBHOOK_URL }}
diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 000000000..7e6f2e68a
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,23 @@
+### Build stage
+FROM python:3.10 AS builder
+
+WORKDIR /outlines
+
+RUN pip install --upgrade pip
+
+# Copy necessary build components
+COPY pyproject.toml .
+COPY outlines ./outlines
+
+# Install outlines and outlines[serve]
+# .git required by setuptools-scm
+RUN --mount=source=.git,target=.git,type=bind \
+    pip install --no-cache-dir .[serve]
+
+### Runtime stage
+FROM python:3.10
+WORKDIR /outlines
+COPY --from=builder /outlines /outlines
+
+# https://dottxt-ai.github.io/outlines/reference/vllm/
+ENTRYPOINT ["python3", "-m", "outlines.serve.serve"]
diff --git a/README.md b/README.md
index 0d85aebf1..b20fc821c 100644
--- a/README.md
+++ b/README.md
@@ -1,669 +1,449 @@
 <div align="center" style="margin-bottom: 1em;">
 
-<img src="./docs/assets/images/logo-light-mode.svg#gh-light-mode-only" alt="Outlines Logo" width=300></img>
-<img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=300></img>
+<img src="./docs/assets/images/logo-light-mode.svg#gh-light-mode-only" alt="Outlines Logo" width=500></img>
+<img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=500></img>
 
 
- 🗒️ *Structured outputs for LLMs* 🗒️
+ 🗒️ *Make LLMs speak the language of every application.* 🗒️
 
-Made with ❤👷️ by the team at [.txt](https://dottxt.co)
-<br>Trusted by NVIDIA, Cohere, HuggingFace, vLLM, etc.
+Made with ❤👷️ by the team at [.txt](https://dottxt.co).
 
-<!-- Project Badges -->
-[![PyPI Version][pypi-version-badge]][pypi]
+[![Documentation][documentation-badge]][documentation]
+[![Contributors][contributors-badge]][contributors]
 [![Downloads][downloads-badge]][pypistats]
-[![Stars][stars-badge]][stars]
-
-<!-- Community Badges -->
 [![Discord][discord-badge]][discord]
-[![Blog][dottxt-blog-badge]][dottxt-blog]
-[![Twitter][twitter-badge]][twitter]
+
+[Youtube channel][youtube-dottxt] | [.txt blog][blog-dottxt] | [Twitter][dottxt-twitter]
+
 
 </div>
 
-Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales).
 
-## Table of Contents
+``` bash
+pip install outlines
+```
 
-- [Why Outlines?](#why-outlines)
-- [Quickstart](#quickstart)
-- [Real-World Examples](#real-world-examples)
-  - [🙋‍♂️ Customer Support Triage](#customer-support-triage)
-  - [📦 E-commerce Product Categorization](#e-commerce-product-categorization)
-  - [💊 Pull Medication Data from Unstructured Notes](#pull-medication-data-from-unstructured-notes)
-  - [📊 Parse Event Details with Incomplete Data](#parse-event-details-with-incomplete-data)
-  - [🗂️ Categorize Documents into Predefined Types](#categorize-documents-into-predefined-types)
-  - [📅 Schedule a Meeting with Function Calling](#schedule-a-meeting-with-function-calling)
-  - [📝 Dynamically Generate Prompts with Re-usable Templates](#dynamically-generate-prompts-with-re-usable-templates)
-- [They Use Outlines](#they-use-outlines)
-- [Model Integrations](#model-integrations)
-- [Core Features](#core-features)
-- [Other Features](#other-features)
-- [About .txt](#about-txt)
-- [Community](#community)
+First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/latest/welcome/)/
 
-<div align="center"><img src="./docs/assets/images/install.png" width=300></img></div>
+Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales) with us.
 
-## Why Outlines?
+## Features
 
-LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
+- 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
+- 🔥 Fast [JSON generation](#efficient-json-generation-following-a-pydantic-model) following a JSON schema or a Pydantic model
+- 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
+- 📝 Generate text that follows a [regex](#efficient-regex-structured-generation) or a [context-free grammar](#using-context-free-grammars-to-guide-generation)
+- 🖍️ Simple and powerful prompting primitives based on the [Jinja templating engine](https://jinja.palletsprojects.com/)
+- 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/latest/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
 
-Outlines guarantees structured outputs during generation — directly from any LLM.
 
-- **Works with any model** - Same code runs across OpenAI, Ollama, vLLM, and more
-- **Simple integration** - Just pass your desired output type: `model(prompt, output_type)`
-- **Guaranteed valid structure** - No more parsing headaches or broken JSON
-- **Provider independence** - Switch models without changing code
+Outlines  has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!
 
+## Why should I use structured generation?
 
-### The Outlines Philosophy
+* It doesn't add any overhead during inference (cost-free)
+* It allows Open Source models to beat closed source models ([Mistral](https://x.com/dottxtai/status/1797692104023363765), [GPT-4](https://x.com/dottxtai/status/1798443290913853770))
+* [It speeds up inference](http://blog.dottxt.co/coalescence.html)
+* [It improves the performance of base models (GSM8K)](http://blog.dottxt.co/performance-gsm8k.html)
+* [It improves the performance of finetuned models (CoNNL)](https://predibase.com/blog/lorax-outlines-better-json-extraction-with-structured-generation-and-lora)
+* [It improves model efficiency (less examples needed)](https://huggingface.co/blog/evaluation-structured-outputs)
 
-<div align="center"><img src="./docs/assets/images/use_philosophy.png" width=300></img></div>
+## .txt company
 
-Outlines follows a simple pattern that mirrors Python's own type system. Simply specify the desired output type, and Outlines will ensure your data matches that structure exactly:
+<div align="center">
+<img src="./docs/assets/images/dottxt-light.svg#gh-light-mode-only" alt="dottxt logo" width=100></img>
+<img src="./docs/assets/images/dottxt-dark.svg#gh-dark-mode-only" alt="dottxt logo" width=100></img>
+</div>
 
-- For a yes/no response, use `Literal["Yes", "No"]`
-- For numerical values, use `int`
-- For complex objects, define a structure with a [Pydantic model](https://docs.pydantic.dev/latest/)
+We started a company to keep pushing the boundaries of structured generation. Learn more about [.txt](https://dottxt.ai), and  [license our commercial product](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF) if you need SOTA performance and quality ✨
 
-## Quickstart
+## Structured generation
 
-Getting started with outlines is simple:
+The first step towards reliability of systems that include large language models
+is to ensure that there is a well-defined interface between their output and
+user-defined code. **Outlines** provides ways to control the generation of
+language models to make their output more predictable.
 
-### 1. Install outlines
+The following methods of structured generation are supported:
 
-``` shell
-pip install outlines
-```
+- [Multiple choices](#multiple-choices)
+- [Type constraints](#type-constraint)
+- [Efficient regex-structured generation](#efficient-regex-structured-generation)
+- [Efficient JSON generation following a Pydantic model](#efficient-json-generation-following-a-pydantic-model)
+- [Using context-free grammars to guide generation](#using-context-free-grammars-to-guide-generation)
+- [Open functions](#open-functions)
 
-### 2. Connect to your preferred model
+### Chat template tokens
 
-``` python
-import outlines
-from transformers import AutoTokenizer, AutoModelForCausalLM
+Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself. Chat template tokens are not needed for base models.
 
+Please see [the documentation](https://dottxt-ai.github.io/outlines/latest/reference/chat_templating) on chat templating for more.
 
-MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
-    AutoTokenizer.from_pretrained(MODEL_NAME)
-)
-```
+### Multiple choices
 
-### 3. Start with simple structured outputs
+You can reduce the completion to a choice between multiple possibilities:
 
 ``` python
-from typing import Literal
-from pydantic import BaseModel
+import outlines
 
+model_name = "HuggingFaceTB/SmolLM2-360M-Instruct"
+model = outlines.models.transformers(model_name)
 
-# Simple classification
-sentiment = model(
-    "Analyze: 'This product completely changed my life!'",
-    Literal["Positive", "Negative", "Neutral"]
-)
-print(sentiment)  # "Positive"
+# You must apply the chat template tokens to the prompt!
+# See below for an example.
+prompt = """
+<|im_start|>system
+You extract information from text.
+<|im_end|>
+
+<|im_start|>user
+What food does the following text describe?
+
+Text: I really really really want pizza.
+<|im_end|>
+<|im_start|>assistant
+"""
 
-# Extract specific types
-temperature = model("What's the boiling point of water in Celsius?", int)
-print(temperature)  # 100
+generator = outlines.generate.choice(model, ["Pizza", "Pasta", "Salad", "Dessert"])
+answer = generator(prompt)
+
+# Likely answer: Pizza
 ```
 
-### 4. Create complex structures
+You can also pass in choices with an `Enum`:
 
-``` python
-from pydantic import BaseModel
+````python
 from enum import Enum
 
-class Rating(Enum):
-    poor = 1
-    fair = 2
-    good = 3
-    excellent = 4
-
-class ProductReview(BaseModel):
-    rating: Rating
-    pros: list[str]
-    cons: list[str]
-    summary: str
-
-review = model(
-    "Review: The XPS 13 has great battery life and a stunning display, but it runs hot and the webcam is poor quality.",
-    ProductReview,
-    max_new_tokens=200,
-)
+class Food(str, Enum):
+    pizza = "Pizza"
+    pasta = "Pasta"
+    salad = "Salad"
+    dessert = "Dessert"
 
-review = ProductReview.model_validate_json(review)
-print(f"Rating: {review.rating.name}")  # "Rating: good"
-print(f"Pros: {review.pros}")           # "Pros: ['great battery life', 'stunning display']"
-print(f"Summary: {review.summary}")     # "Summary: Good laptop with great display but thermal issues"t(result)
-```
+generator = outlines.generate.choice(model, Food)
+answer = generator(prompt)
+# Likely answer: Pizza
+````
 
-## Real-world examples
+### Type constraints
 
-Here are production-ready examples showing how Outlines solves common problems:
+You can instruct the model to only return integers or floats:
 
-<details id="customer-support-triage"><summary><b>🙋‍♂️ Customer Support Triage</b>
-<br>This example shows how to convert a free-form customer email into a structured service ticket. By parsing attributes like priority, category, and escalation flags, the code enables automated routing and handling of support issues.
-</summary>
 
 ``` python
 import outlines
-from enum import Enum
-from pydantic import BaseModel
-from transformers import AutoTokenizer, AutoModelForCausalLM
-from typing import List
-
-
-MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
-    AutoTokenizer.from_pretrained(MODEL_NAME)
-)
 
+model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
 
-def alert_manager(ticket):
-    print("Alert!", ticket)
+prompt = "<s>result of 9 + 9 = 18</s><s>result of 1 + 2 = "
+answer = outlines.generate.format(model, int)(prompt)
+print(answer)
+# 3
 
+prompt = "sqrt(2)="
+generator = outlines.generate.format(model, float)
+answer = generator(prompt, max_tokens=10)
+print(answer)
+# 1.41421356
+```
 
-class TicketPriority(str, Enum):
-    low = "low"
-    medium = "medium"
-    high = "high"
-    urgent = "urgent"
+### Efficient regex-structured generation
 
-class ServiceTicket(BaseModel):
-    priority: TicketPriority
-    category: str
-    requires_manager: bool
-    summary: str
-    action_items: List[str]
+Outlines also comes with fast regex-structured generation. In fact, the `choice` and
+`format` functions above all use regex-structured generation under the
+hood:
 
+``` python
+import outlines
 
-customer_email = """
-Subject: URGENT - Cannot access my account after payment
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
-I paid for the premium plan 3 hours ago and still can't access any features.
-I've tried logging out and back in multiple times. This is unacceptable as I
-have a client presentation in an hour and need the analytics dashboard.
-Please fix this immediately or refund my payment.
-"""
+prompt = """
+<|im_start|>system You are a helpful assistant.
+<|im_end|>
 
-prompt = f"""
 <|im_start|>user
-Analyze this customer email:
-
-{customer_email}
+What is an IP address of the Google DNS servers?
 <|im_end|>
 <|im_start|>assistant
-"""
-
-ticket = model(
-    prompt,
-    ServiceTicket,
-    max_new_tokens=500
-)
-
-# Use structured data to route the ticket
-ticket = ServiceTicket.model_validate_json(ticket)
-if ticket.priority == "urgent" or ticket.requires_manager:
-    alert_manager(ticket)
-```
-</details>
-
-<details id="e-commerce-product-categorization"><summary><b>📦 E-commerce product categorization</b>
-<br>This use case demonstrates how outlines can transform product descriptions into structured categorization data (e.g., main category, sub-category, and attributes) to streamline tasks such as inventory management. Each product description is processed automatically, reducing manual categorization overhead.
-</summary>
+The IP address of a Google DNS server is
 
-```python
-import outlines
-from pydantic import BaseModel
-from transformers import AutoTokenizer, AutoModelForCausalLM
-from typing import List, Optional
+"""
 
+generator = outlines.generate.text(model)
+unstructured = generator(prompt, max_tokens=30)
 
-MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
-    AutoTokenizer.from_pretrained(MODEL_NAME)
+generator = outlines.generate.regex(
+    model,
+    r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)",
+    sampler=outlines.samplers.greedy(),
 )
+structured = generator(prompt, max_tokens=30)
 
+print(unstructured)
+# 8.8.8.8
+#
+# <|im_end|>
 
-def update_inventory(product, category, sub_category):
-    print(f"Updated {product.split(',')[0]} in category {category}/{sub_category}")
-
-
-class ProductCategory(BaseModel):
-    main_category: str
-    sub_category: str
-    attributes: List[str]
-    brand_match: Optional[str]
-
-# Process product descriptions in batches
-product_descriptions = [
-    "Apple iPhone 15 Pro Max 256GB Titanium, 6.7-inch Super Retina XDR display with ProMotion",
-    "Organic Cotton T-Shirt, Men's Medium, Navy Blue, 100% Sustainable Materials",
-    "KitchenAid Stand Mixer, 5 Quart, Red, 10-Speed Settings with Dough Hook Attachment"
-]
-
-template = outlines.Template.from_string("""
-<|im_start|>user
-Categorize this product:
-
-{{ description }}
-<|im_end|>
-<|im_start|>assistant
-""")
+print(structured)
+# 8.8.8.8
+```
 
-# Get structured categorization for all products
-categories = model(
-    [template(description=desc) for desc in product_descriptions],
-    ProductCategory,
-    max_new_tokens=200
-)
+Unlike other libraries, regex-structured generation in Outlines is almost as fast
+as non-structured generation.
 
-# Use categorization for inventory management
-categories = [
-    ProductCategory.model_validate_json(category) for category in categories
-]
-for product, category in zip(product_descriptions, categories):
-    update_inventory(product, category.main_category, category.sub_category)
-```
-</details>
+### Efficient JSON generation following a Pydantic model
 
-<details id="parse-event-details-with-incomplete-data"><summary><b>📊 Parse event details with incomplete data</b>
-<br>This example uses outlines to parse event descriptions into structured information (like event name, date, location, type, and topics), even handling cases where the data is incomplete. It leverages union types to return either structured event data or a fallback “I don’t know” answer, ensuring robust extraction in varying scenarios.
-</summary>
+Outlines users can guide the generation process so the output is *guaranteed* to follow a [JSON schema](https://json-schema.org/) or [Pydantic model](https://docs.pydantic.dev/latest/):
 
 ```python
-import outlines
-from typing import Union, List, Literal
-from pydantic import BaseModel
 from enum import Enum
-from transformers import AutoTokenizer, AutoModelForCausalLM
-
-
-MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
-    AutoTokenizer.from_pretrained(MODEL_NAME)
-)
+from pydantic import BaseModel, constr
 
-class EventType(str, Enum):
-    conference = "conference"
-    webinar = "webinar"
-    workshop = "workshop"
-    meetup = "meetup"
-    other = "other"
-
-
-class EventInfo(BaseModel):
-    """Structured information about a tech event"""
-    name: str
-    date: str
-    location: str
-    event_type: EventType
-    topics: List[str]
-    registration_required: bool
-
-# Create a union type that can either be a structured EventInfo or "I don't know"
-EventResponse = Union[EventInfo, Literal["I don't know"]]
-
-# Sample event descriptions
-event_descriptions = [
-    # Complete information
-    """
-    Join us for DevCon 2023, the premier developer conference happening on November 15-17, 2023
-    at the San Francisco Convention Center. Topics include AI/ML, cloud infrastructure, and web3.
-    Registration is required.
-    """,
-
-    # Insufficient information
-    """
-    Tech event next week. More details coming soon!
-    """
-]
-
-# Process events
-results = []
-for description in event_descriptions:
-    prompt = f"""
-<|im_start>system
-You are a helpful assistant
-<|im_end|>
-<|im_start>user
-Extract structured information about this tech event:
-
-{description}
-
-If there is enough information, return a JSON object with the following fields:
+import outlines
 
-- name: The name of the event
-- date: The date where the event is taking place
-- location: Where the event is taking place
-- event_type: either 'conference', 'webinar', 'workshop', 'meetup' or 'other'
-- topics: a list of topics of the conference
-- registration_required: a boolean that indicates whether registration is required
+class Weapon(str, Enum):
+    sword = "sword"
+    axe = "axe"
+    mace = "mace"
+    spear = "spear"
+    bow = "bow"
+    crossbow = "crossbow"
 
-If the information available does not allow you to fill this JSON, and only then, answer 'I don't know'.
-<|im_end|>
-<|im_start|>assistant
-"""
-    # Union type allows the model to return structured data or "I don't know"
-    result = model(prompt, EventResponse, max_new_tokens=200)
-    results.append(result)
-
-# Display results
-for i, result in enumerate(results):
-    print(f"Event {i+1}:")
-    if isinstance(result, str):
-        print(f"  {result}")
-    else:
-        # It's an EventInfo object
-        print(f"  Name: {result.name}")
-        print(f"  Type: {result.event_type}")
-        print(f"  Date: {result.date}")
-        print(f"  Topics: {', '.join(result.topics)}")
-    print()
-
-# Use structured data in downstream processing
-structured_count = sum(1 for r in results if isinstance(r, EventInfo))
-print(f"Successfully extracted data for {structured_count} of {len(results)} events")
-```
-</details>
 
-<details id="categorize-documents-into-predefined-types"><summary><b>🗂️ Categorize documents into predefined types</b>
-<br>In this case, outlines classifies documents into predefined categories (e.g., “Financial Report,” “Legal Contract”) using a literal type specification. The resulting classifications are displayed in both a table format and through a category distribution summary, illustrating how structured outputs can simplify content management.
-</summary>
+class Armor(str, Enum):
+    leather = "leather"
+    chainmail = "chainmail"
+    plate = "plate"
 
-```python
-import outlines
-from typing import Literal, List
-import pandas as pd
-from transformers import AutoTokenizer, AutoModelForCausalLM
 
+class Character(BaseModel):
+    name: constr(max_length=10)
+    age: int
+    armor: Armor
+    weapon: Weapon
+    strength: int
 
-MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
-    AutoTokenizer.from_pretrained(MODEL_NAME)
-)
 
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
-# Define classification categories using Literal
-DocumentCategory = Literal[
-    "Financial Report",
-    "Legal Contract",
-    "Technical Documentation",
-    "Marketing Material",
-    "Personal Correspondence"
-]
+# Construct structured sequence generator
+generator = outlines.generate.json(model, Character)
 
-# Sample documents to classify
-documents = [
-    "Q3 Financial Summary: Revenue increased by 15% year-over-year to $12.4M. EBITDA margin improved to 23% compared to 19% in Q3 last year. Operating expenses...",
+# Draw a sample
+seed = 789001
 
-    "This agreement is made between Party A and Party B, hereinafter referred to as 'the Parties', on this day of...",
+character = generator("Give me a character description", seed=seed)
 
-    "The API accepts POST requests with JSON payloads. Required parameters include 'user_id' and 'transaction_type'. The endpoint returns a 200 status code on success."
-]
+print(repr(character))
+# Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
 
-template = outlines.Template.from_string("""
-<|im_start|>user
-Classify the following document into exactly one category among the following categories:
-- Financial Report
-- Legal Contract
-- Technical Documentation
-- Marketing Material
-- Personal Correspondence
-
-Document:
-{{ document }}
-<|im_end|>
-<|im_start|>assistant
-""")
+character = generator("Give me an interesting character description")
 
-# Classify documents
-def classify_documents(texts: List[str]) -> List[DocumentCategory]:
-    results = []
+print(repr(character))
+# Character(name='Vivian Thr', age=44, armor=<Armor.plate: 'plate'>, weapon=<Weapon.crossbow: 'crossbow'>, strength=125)
+```
 
-    for text in texts:
-        prompt = template(document=text)
-        # The model must return one of the predefined categories
-        category = model(prompt, DocumentCategory, max_new_tokens=200)
-        results.append(category)
+The method works with union types, optional types, arrays, nested schemas, etc. Some field constraints are [not supported yet](https://github.com/dottxt-ai/outlines/issues/215), but everything else should work.
 
-    return results
+### Efficient JSON generation following a JSON Schema
 
-# Perform classification
-classifications = classify_documents(documents)
+Sometimes you just want to be able to pass a JSON Schema instead of a Pydantic model. We've got you covered:
 
-# Create a simple results table
-results_df = pd.DataFrame({
-    "Document": [doc[:50] + "..." for doc in documents],
-    "Classification": classifications
-})
+``` python
+import outlines
 
-print(results_df)
+schema = '''{
+    "title": "Character",
+    "type": "object",
+    "properties": {
+        "name": {
+            "title": "Name",
+            "maxLength": 10,
+            "type": "string"
+        },
+        "age": {
+            "title": "Age",
+            "type": "integer"
+        },
+        "armor": {"$ref": "#/definitions/Armor"},
+        "weapon": {"$ref": "#/definitions/Weapon"},
+        "strength": {
+            "title": "Strength",
+            "type": "integer"
+        }
+    },
+    "required": ["name", "age", "armor", "weapon", "strength"],
+    "definitions": {
+        "Armor": {
+            "title": "Armor",
+            "description": "An enumeration.",
+            "enum": ["leather", "chainmail", "plate"],
+            "type": "string"
+        },
+        "Weapon": {
+            "title": "Weapon",
+            "description": "An enumeration.",
+            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
+            "type": "string"
+        }
+    }
+}'''
 
-# Count documents by category
-category_counts = pd.Series(classifications).value_counts()
-print("\nCategory Distribution:")
-print(category_counts)
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = outlines.generate.json(model, schema)
+character = generator("Give me a character description")
 ```
-</details>
 
-<details>
-<summary id="schedule-a-meeting-with-function-calling"><b>📅 Schedule a meeting from requests with Function Calling</b>
-<br>This example demonstrates how outlines can interpret a natural language meeting request and translate it into a structured format matching a predefined function’s parameters. Once the meeting details are extracted (e.g., title, date, duration, attendees), they are used to automatically schedule the meeting.
-</summary>
+### Using context-free grammars to guide generation
 
-```python
+Formal grammars rule the world, and Outlines makes them rule LLMs too. You can pass any context-free grammar in the EBNF format and Outlines will generate an output that is valid to this grammar:
+
+``` python
 import outlines
-import json
-from typing import List, Optional
-from datetime import date
-from transformers import AutoTokenizer, AutoModelForCausalLM
 
+arithmetic_grammar = """
+    ?start: expression
 
-MODEL_NAME = "microsoft/phi-4"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
-    AutoTokenizer.from_pretrained(MODEL_NAME)
-)
+    ?expression: term (("+" | "-") term)*
 
+    ?term: factor (("*" | "/") factor)*
 
-# Define a function with typed parameters
-def schedule_meeting(
-    title: str,
-    date: date,
-    duration_minutes: int,
-    attendees: List[str],
-    location: Optional[str] = None,
-    agenda_items: Optional[List[str]] = None
-):
-    """Schedule a meeting with the specified details"""
-    # In a real app, this would create the meeting
-    meeting = {
-        "title": title,
-        "date": date,
-        "duration_minutes": duration_minutes,
-        "attendees": attendees,
-        "location": location,
-        "agenda_items": agenda_items
-    }
-    return f"Meeting '{title}' scheduled for {date} with {len(attendees)} attendees"
+    ?factor: NUMBER
+           | "-" factor
+           | "(" expression ")"
 
-# Natural language request
-user_request = """
-I need to set up a product roadmap review with the engineering team for next
-Tuesday at 2pm. It should last 90 minutes. Please invite john@example.com,
-sarah@example.com, and the product team at product@example.com.
+    %import common.NUMBER
 """
 
-# Outlines automatically infers the required structure from the function signature
-prompt = f"""
-<|im_start|>user
-Extract the meeting details from this request:
+model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
+generator = outlines.generate.cfg(model, arithmetic_grammar)
+sequence = generator("Alice had 4 apples and Bob ate 2. Write an expression for Alice's apples:")
 
-{user_request}
-<|im_end|>
-<|im_start|>assistant
-"""
-meeting_params = model(prompt, schedule_meeting, max_new_tokens=200)
+print(sequence)
+# (8-2)
+```
 
-# The result is a dictionary matching the function parameters
-meeting_params = json.loads(meeting_params)
-print(meeting_params)
+This was a very simple grammar, and you can use `outlines.generate.cfg` to generate syntactically valid Python, SQL, and much more than this. Any kind of structured text, really. All you have to do is search for "X EBNF grammar" on the web, and take a look at the [Outlines `grammars` module](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
 
-# Call the function with the extracted parameters
-result = schedule_meeting(**meeting_params)
-print(result)
-# "Meeting 'Product Roadmap Review' scheduled for 2023-10-17 with 3 attendees"
-```
-</details>
+### Open functions
 
-<details>
-<summary id="dynamically-generate-prompts-with-re-usable-templates"><b>📝 Dynamically generate prompts with re-usable templates</b>
-<br>Using Jinja-based templates, this example shows how to generate dynamic prompts for tasks like sentiment analysis. It illustrates how to easily re-use and customize prompts—including few-shot learning strategies—for different content types while ensuring the outputs remain structured.
-</summary>
+Outlines can infer the structure of the output from the signature of a function. The result is a dictionary, and can be passed directly to the function using the usual dictionary expansion syntax `**`:
 
 ```python
 import outlines
-from typing import List, Literal
-from transformers import AutoTokenizer, AutoModelForCausalLM
 
 
-MODEL_NAME = "microsoft/phi-4"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
-    AutoTokenizer.from_pretrained(MODEL_NAME)
-)
-
+def add(a: int, b: int):
+    return a + b
 
-# 1. Create a reusable template with Jinja syntax
-sentiment_template = outlines.Template.from_string("""
-<|im_start>user
-Analyze the sentiment of the following {{ content_type }}:
+model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
+generator = outlines.generate.json(model, add)
+result = generator("Return json with two integers named a and b respectively. a is odd and b even.")
 
-{{ text }}
+print(add(**result))
+# 3
+```
 
-Provide your analysis as either "Positive", "Negative", or "Neutral".
-<|im_end>
-<|im_start>assistant
-"""
+A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
 
-# 2. Generate prompts with different parameters
-review = "This restaurant exceeded all my expectations. Fantastic service!"
-prompt = sentiment_template(content_type="review", text=review)
+You can also embed various functions into an enum to generate params:
 
-# 3. Use the templated prompt with structured generation
-result = model(prompt, Literal["Positive", "Negative", "Neutral"])
-print(result)  # "Positive"
+```python
+from enum import Enum
+from functools import partial
 
-# Templates can also be loaded from files
-example_template = outlines.Template.from_file("templates/few_shot.txt")
+import outlines
 
-# Use with examples for few-shot learning
-examples = [
-    ("The food was cold", "Negative"),
-    ("The staff was friendly", "Positive")
-]
-few_shot_prompt = example_template(examples=examples, query="Service was slow")
-print(few_shot_prompt)
-```
-</details>
 
-## They use outlines
+def add(a: int, b: int) -> int:
+    return a + b
 
-<div align="center">
-<img src="./docs/assets/images/readme-light.png#gh-light-mode-only" alt="Users Logo"></img>
-<img src="./docs/assets/images/readme-dark.png#gh-dark-mode-only" alt="Users Logo"></img>
-</div>
+def mul(c: float, d: float) -> float:
+    return c * d
 
-## Model Integrations
+class Operation(Enum):
+    add = partial(add)
+    mul = partial(mul)
 
-| Model type | Description | Documentation |
-|---------|-------------|:-------------:|
-| **Server Support** | vLLM and Ollama | [Server Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
-| **Local Model Support** | transformers and llama.cpp | [Model Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
-| **API Support** | OpenAI and Gemini | [API Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
+model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
+generator = outlines.generate.json(model, Operation)
+result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
 
-## Core Features
+print(result)
+# {'c': -3.14, 'd': 1.5}
+```
 
-| Feature | Description | Documentation |
-|---------|-------------|:-------------:|
-| **Multiple Choices** | Constrain outputs to predefined options | [Multiple Choices →](https://dottxt-ai.github.io/outlines/latest/guides/choice_generation) |
-| **️Function Calls** | Infer structure from function signatures | [Function Guide →](https://dottxt-ai.github.io/outlines/latest/guides/function_generation) |
-| **JSON/Pydantic** | Generate outputs matching JSON schemas | [JSON Guide →](https://dottxt-ai.github.io/outlines/latest/guides/json_generation) |
-| **Regular Expressions** | Generate text following a regex pattern | [Regex Guide →](https://dottxt-ai.github.io/outlines/latest/guides/regex_generation) |
-| **Grammars** | Enforce complex output structures | [Grammar Guide →](https://dottxt-ai.github.io/outlines/latest/guides/cfg_generation) |
+## Prompting
 
-## Other Features
+Building prompts can get messy. **Outlines** makes it easier to write and manage
+prompts by encapsulating templates inside "template functions". Template
+functions use the Jinja2 templating engine to help build complex prompts in a
+concise manner.
 
-| Feature | Description | Documentation |
-|---------|-------------|:-------------:|
-| **Prompt templates** | Separate complex prompts from code | |
-| **Custome types** | Intuitive interface to build complex types | |
-| **Applications** | Encapsulate templates and types into functions | |
+Template functions are created by loading a Jinja2 template from a text file.
+Assume you have the following prompt template defined in `prompt.txt`:
 
-## About .txt
+``` text
+You are a sentiment-labelling assistant.
 
-<div align="center">
-<img src="./docs/assets/images/dottxt-light.svg#gh-light-mode-only" alt="dottxt logo" width=100></img>
-<img src="./docs/assets/images/dottxt-dark.svg#gh-dark-mode-only" alt="dottxt logo" width=100></img>
-</div>
+{% for example in examples %}
+{{ example[0] }} // {{ example[1] }}
+{% endfor %}
+{{ to_label }} //
+```
 
-Outlines is developed and maintained by [.txt](https://dottxt.co), a company dedicated to making LLMs more reliable for production applications.
+You can then load it and call it with:
 
-Our focus is on advancing structured generation technology through:
+``` python
+import outlines
 
-- 🧪 **Cutting-edge Research**: We publish our findings on [structured generation](http://blog.dottxt.co/performance-gsm8k.html)
-- 🚀 **Enterprise-grade solutions**: You can license [our enterprise-grade libraries](https://docs.dottxt.co).
-- 🧩 **Open Source Collaboration**: We believe in building in public and contributing to the community
+examples = [
+    ("The food was disgusting", "Negative"),
+    ("We had a fantastic night", "Positive"),
+    ("Recommended", "Positive"),
+    ("The waiter was rude", "Negative")
+]
 
-Follow us on [Twitter](https://twitter.com/dottxtai) or check out our [blog](https://blog.dottxt.co/) to stay updated on our latest work in making LLMs more reliable.
+labelling = outlines.Template.from_file("prompt.txt")
+prompt = labelling("Just awesome", examples)
+```
 
-## Community
+This helps:
 
-<div align="center" style="margin-bottom: 1em;">
+- Keep content separate from the code
+- Design "white space perfect" prompts
 
-[![Contributors][contributors-badge]][contributors]
-[![Stars][stars-badge]][stars]
-[![Downloads][downloads-badge]][pypistats]
-[![Discord badge][discord-badge]][discord]
+It is more maintainable and means prompts can be versioned separately from the code.
 
-</div>
+## Join us
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
+- 🔨 **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/latest/community/contribute/).
 - 🐞 **Found a bug?** Open an [issue](https://github.com/dottxt-ai/outlines/issues)
-- 🧩  **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/latest/community/contribute/).
 
 
 ## Cite Outlines
 
 ```
 @article{willard2023efficient,
-  title={Efficient Guided Generation for Large Language Models},
+  title={Efficient Guided Generation for LLMs},
   author={Willard, Brandon T and Louf, R{\'e}mi},
   journal={arXiv preprint arXiv:2307.09702},
   year={2023}
 }
 ```
 
+[documentation]: https://dottxt-ai.github.io/outlines/latest/welcome/
+[documentation-badge]: https://img.shields.io/readthedocs/outlines
 [contributors]: https://github.com/dottxt-ai/outlines/graphs/contributors
 [contributors-badge]: https://img.shields.io/github/contributors/dottxt-ai/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
-[dottxt-blog]: https://blog.dottxt.co/
-[dottxt-blog-badge]: https://img.shields.io/badge/dottxt%20blog-a6b4a3
 [dottxt-twitter]: https://twitter.com/dottxtai
-[dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
 [discord]: https://discord.gg/R9DSu34mGd
-[discord-badge]: https://img.shields.io/discord/1182316225284554793?color=ddb8ca&logo=discord&logoColor=white&style=flat-square
-[downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=A6B4A3&logo=python&logoColor=white&style=flat-square
+[discord-badge]: https://img.shields.io/discord/1182316225284554793?color=81A1C1&logo=discord&logoColor=white&style=flat-square
+[downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=89AC6B&logo=python&logoColor=white&style=flat-square
 [pypistats]: https://pypistats.org/packages/outlines
-[pypi-version-badge]: https://img.shields.io/pypi/v/outlines?style=flat-square&logoColor=white&color=ddb8ca
-[pypi]: https://pypi.org/project/outlines/
-[stars]: https://github.com/dottxt-ai/outlines/stargazers
-[stars-badge]: https://img.shields.io/github/stars/dottxt-ai/outlines?style=flat-square&logo=github&color=BD932F&logoColor=white
-[twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=flat-square&logo=x&logoColor=white&color=bd932f
-[twitter]: https://x.com/dottxtai
+[dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
+[youtube-dottxt]: https://www.youtube.com/@dottxt-ai
+[blog-dottxt]: https://blog.dottxt.co/
diff --git a/benchmarks/bench_cfg_guide.py b/benchmarks/bench_cfg_guide.py
index a4d288700..477104435 100644
--- a/benchmarks/bench_cfg_guide.py
+++ b/benchmarks/bench_cfg_guide.py
@@ -4,7 +4,7 @@
 
 import outlines.grammars
 from outlines.caching import cache_disabled
-from outlines.processors.guide import CFGGuide
+from outlines.fsm.guide import CFGGuide
 from outlines.models.transformers import TransformerTokenizer
 
 random.seed(42)
diff --git a/benchmarks/bench_json_schema.py b/benchmarks/bench_json_schema.py
index 6ed223ef4..3a1f72cb6 100644
--- a/benchmarks/bench_json_schema.py
+++ b/benchmarks/bench_json_schema.py
@@ -1,7 +1,7 @@
 from outlines_core.fsm.json_schema import build_regex_from_schema
 
 from outlines.caching import cache_disabled
-from outlines.processors.guide import RegexGuide
+from outlines.fsm.guide import RegexGuide
 
 from .common import setup_tokenizer  # noqa: E402
 
diff --git a/benchmarks/bench_regex_guide.py b/benchmarks/bench_regex_guide.py
index 14b15da89..fa23a724f 100644
--- a/benchmarks/bench_regex_guide.py
+++ b/benchmarks/bench_regex_guide.py
@@ -1,5 +1,5 @@
 from outlines.caching import cache_disabled
-from outlines.processors.guide import RegexGuide
+from outlines.fsm.guide import RegexGuide
 
 from .common import setup_tokenizer
 
diff --git a/docs/api/guide.md b/docs/api/guide.md
new file mode 100644
index 000000000..1c3160c8c
--- /dev/null
+++ b/docs/api/guide.md
@@ -0,0 +1 @@
+::: outlines.fsm.guide
diff --git a/docs/api_reference/index.md b/docs/api/index.md
similarity index 100%
rename from docs/api_reference/index.md
rename to docs/api/index.md
diff --git a/docs/api/json_schema.md b/docs/api/json_schema.md
new file mode 100644
index 000000000..471cb3a80
--- /dev/null
+++ b/docs/api/json_schema.md
@@ -0,0 +1 @@
+::: outlines.fsm.json_schema
diff --git a/docs/api/models.md b/docs/api/models.md
new file mode 100644
index 000000000..cb9497be1
--- /dev/null
+++ b/docs/api/models.md
@@ -0,0 +1 @@
+::: outlines.models
diff --git a/docs/api/parsing.md b/docs/api/parsing.md
new file mode 100644
index 000000000..e9662999c
--- /dev/null
+++ b/docs/api/parsing.md
@@ -0,0 +1 @@
+::: outlines.fsm.parsing
diff --git a/docs/api/regex.md b/docs/api/regex.md
new file mode 100644
index 000000000..5ef91db46
--- /dev/null
+++ b/docs/api/regex.md
@@ -0,0 +1 @@
+::: outlines.generate.regex
diff --git a/docs/api/samplers.md b/docs/api/samplers.md
new file mode 100644
index 000000000..2b9b34234
--- /dev/null
+++ b/docs/api/samplers.md
@@ -0,0 +1 @@
+::: outlines.samplers
diff --git a/docs/api/templates.md b/docs/api/templates.md
new file mode 100644
index 000000000..6150512d4
--- /dev/null
+++ b/docs/api/templates.md
@@ -0,0 +1 @@
+::: outlines.templates
diff --git a/docs/assets/images/install.png b/docs/assets/images/install.png
deleted file mode 100644
index 2b89b72680c6dff599fc96dc5a2cbfe014a48903..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 12669
zcmeHu^;?u(*Y-8EAYmenq=<mh4GKs%(y4&bokI*N;H}an4bl=r4-JZdbTc$acg!%r
zFvK_NgZFnF??3P!-}msthHLL@*V=2HYwhb?{G*{JPfAQp3;+PBqJpdz0N_Bde;*JL
zU`vvp2{YJVcy3P=b%?M(engh<u+JbjE%_%vNgwSx05AfIvX6DVO}A$V{dBCFF81eC
zsBPZ99l%?^_nPr_K7{it@rEHFrvVSCgS;o*ZN$u7%wmiBqo!<qpsein_mC3ODMNr&
zP~#?Enbys_l@vn?iHR3TG2yLjm1vI8o+ZD&&1GdX>y^aShIBr@IiE?FD}*>ij1b^g
z0m;HM4i`02yA8slkOqDg(nJtryuZz@ctj8~J7+b@%XUV<mGnP-0UWX$jDTp-R}quT
z&Jf^-z@LG!qYJA*fVEPBRrhOuMg~AQjPT#H{wn4lf^afgsZ)mKeou=nuoL6{KR&S$
z&V0sWABU#n=V-}L1J}CC?Mmz5Z*}QtUV-Gg&PXUW@u^z)EO?MoChQVj({t6qzjyUg
z{LU#5sFq)e+4pVC0DHN3W*4e@-ExYSLq--<YS!HF@w?94tp!&T>-)S+LtT4*m*)Zm
z5#S@*R*j5Jb_^>mS`pFC#ox#v@)38U5{;vdUnbNzk{q@9(En8n=`|e5{oVqVf^(!0
z8r_yDHT7{g1j(|ER#rfsQU2til}LMYd6Kf+8FvS?S5Eq!eQ{ceA-JI*Q=W(T??yCi
z#Qs$UMv7`kzmUiFOU~PGPfjCoGq{iU*t%j?f<GZCzX(RiU#h@o&dA-8_-+}KJJ($S
z+K28-9^C1hnQLRIL;ulE?2F&7?bZ6hp+;2;KQ+>5u&)WbFFmiI_0hV=^X<i7Rl~Et
z?u<?|thWu``7yM_d;kcuIK>!<4mEZ{TX6q<YwN~d=So#N*Y340lD9Vj4<TEp&rpbJ
zn#|pmvdhDPq~bJnRTb)GzeNz=#h2#tp1<N_&1_5duW<FR>DT-EC?s-c6W#z`<RljS
zHLyJ%tkUb@UJCaAOQr81XkkfEn8S<Wz9QM_!C;S?F^x)>;GS5!R)7%Ljf=G07=>x)
zxc8MrPhZu(TWqHI+MZJ-I&3N?r)#9bAVM}>$D`!BdboT;G5J`3qtxT;hinQ%)htmP
z)+J^Y3wsNH&gpwKWD#c`A{+IO4b$0-X@P=5Y)>qUO`Lgjh6tIrxR$XdD@U5G!LCR5
zuocth7jh?DvNf%{328bH&P3@Vm7HE%X4kO@ESrD258YE#tW}Smp`R#ij1xyH<`!W@
zmP>U@zZm7jM%+tK((ada`dp(OX43Uos^FnT=`{1`j;*2~N~u>pE}N?}v?`$C&VLS^
zpL_lj@M-a|GnI;aJ#H;pkjf@HIgQQH>#R}76sqo-5Z~Dp&5O>QWcP|YwuHiG>5Eg`
zBYN&=-;<cqQ}Wzov4OX9IqF58p#942()Eas^g1KNVysqZCdy$4aEdYS<WLlmQ?9&H
zKTp(`@X_d4tJfx;?`x);JyXZ-53Zt=o>>Iz+=^2k3YS$3WR;n2I~?OFidj(nBQjx~
zmTMpl@Y(sNpE3M5PU@;VuND-*b~xjCZxCD<E|eDw{c3ccbyQSWj&`I|v62z4zNw>Z
zV-R=hU{m7I7&N&F3$%}7$;(r6%{!p5W!I{0)VZ(Ror97pUn9M$7*5Fi8v)GIr9Y5E
zsL#)ne#Uu<IIu`mJ=2fQSthiE&-x=eu6vEhWP~&~8`mgL!c?O^R_^(BDs_pMyQv4C
zuIG4ak_S<rzh~Qu8JT{ds6y@}467TtO0&xyUG#E_0;jbO$LIPVVGrJVH3-rO^{1@7
zEyByrxwW^tI!19kJj;|=ZuVJ_wy3oU2_3&{BE}L*cYcZ*rl_gP(8}f7xKs7hobHe-
zr9sco!<NqT7ztOGnGf2=uwtTUL*8Uz=yUo182Xdj*zxxQRgX8*6R8oUP7mei$V|<Q
zxi1`_nDd8UP&FH!c%eeXw`jJiJcPZ^bYZ6*$lAgDw3}wSEJMt9mn`PYc}iNpmUl8I
zeyI4YB}M%B59$T&;ThTkBbjqjRi2%isguF%H<k?M9OqaACw4!0-_s8bxbQxwU!US?
zuvj-|uX5=Ip*J}lnCXefGUpcd+oEUr*uMniNbSV+tnyJ2hkv;reD}B1DUzf=kU|3a
z8UtmfvXwDC1Gl;O+hG+oI=b7OaZ<A5B*$}$p4|hIlIxLlvklGWs0a5Betg+AMg*F^
zE24D1$YydusYWkV7lcK<Jvi!;SYDXYXDv?uO!<&SnEvB$^q0PgZ)gdOsHUa&Bb4B3
zbgvyz*Gn=TuT^+}gFf=OhBg{hIwDS|V;zPXm2C%#O}~|-PcT^{`<GoiU-$iVy6qt7
zGwVOa-6>q|@T;{r-SL_>Vwk=&^@XqX!~;>|r)i;o=8?tW;srMLb}xEey9eo*H}Of>
zFumb3h>Q0P7UZ}LwC6K+A%WWy-+dP2iTSRk+&Fw#N_G)%YRpt1QO;af^ID?TPdhW4
zr_=n-_-zy7KM_ZUOVDj9pd<TuX{EY&>76||ges8ZK{V!&Hd5&Rw<%iR1IYE-^`h(R
zLy`>+&#3Lj(!F%MrUR{iByaLkgI_fUN;x;w4tMy(MGp;r%$jo6*&65RRF26RO<E)4
znbLUnUiSC-X2=m&{s6q?zNnXUswSk2{lw2`H}u^-zwTqMS0D~%YyiB`rA)QfBOse(
ztDQu<*0Q{OC3<i6XK5Jz20%}+T=q*pzn6D+_|B=~MS<?kc%=mzZpCg`o*#_@$MxTF
zLC|ehY2a#>$Q6i12EKSOa3_(W|LO0^8MS`xkqXC}NO(wyp#P7b&0r<LuIL97#qFTV
zB_XcWkUuic{w3pb(yLP~vM25H=syrqY58k_s;}cjqUBu9<3x3=&D$bEzYQXfq;eei
zQ9REy^&-;ftH@ba5c2o<uR3wK1OeNuGw1|<PN4M{cZV3NIKk;UsQwsbLU*&gH;r(h
z)%nX)$Mv~U&Kc29sfAh2_8j@~D5Dt!L%;kOW)-TmZzabhah;`B+r8SZVA}NlZ=>0s
zNPk59_GZxU<wBV84Bdeq7mQ`ll_2>QOGRJX!gth_m8X!?N|-|Kr#%>)X7)U=iZe21
zCkm~%6<I<T&C1MG4#r7{^|7>joTFkiqInqqR5!jkZB`}XrrIoLv1nEN_k$mT#U<Ru
zvA9!5%7@G%3*6yqXH2#lX@3+R@&+F}>xk!Q=6tiYcaDi#a_KMZG65fTGAK`5%$;+M
z3*^FrG2OxS+5S_pzZ+A*_m|Xpsz@-0PG2|jWWLJXI8?NC)E&R8(Q1a?Ve0g)CWQsM
z?S+!&h5y<ag%}Q(Byb`#&%oyw9iMg^;ap0FiI0o#G*dR#Bo93Amc};qO>wQ;DBdL}
z+$?qa1ILadt?A7H)Hp{GT+vGhBaxwRJjP+=mBxk_%pdkB#dg8AuYZj`?+lX1lXiGP
zN7lsG%G!9{WQdJf)m8P5d;KOu7s}VaeClnJKdTi&rb>P1{=S^UxX?H%xAiMu`%R_(
z!?@@fQN_tBmk(37vI8X3w<cHPzMVZoEcRI@qMv$>g!sQ1RS$imeCdng?ki4kEV$Nh
z`q%Vt#(nulTyXz>LN?t$(dSA8txaSX89xtx2V&2aTONV-G@O#DuBUynz8zZAs1+4?
zFOfrW=>E_7luBIsh?{ykitjrf<nGliuk)4_!+FN7G8%*ar|+hC@IQRFLt_9|>Wx9m
z%w0J{BYcC&_jYt%Bp^0ZnusI3yCt-?OX8Q(wOIt_@Y=Yddg(qE!$_+B0CR>FkuW!K
zlB_E9VaLu@IRP)d^1Ld`p;>+DWrVVzuk4qeK%v#Be}k8?0@K3#nC_YK;)jZYj_UFE
zwfn6luKR0^)+&jMupQ*k6{Xkra9si4V6icZR`B41Lk(QKUH<r%=SD4Mpoi7T!IOWm
z8w=ovS15q(iX6+!D3g&5p9u13;(-qF<z1ju9Rj>c=VJZiWU*ZyaYzFnJFPRd{zk6;
zIC#<+dk(70F7FgUIOZyXBymsv_#P167WN7^`BJ&_vWv6|F$M6V+#vtve?k~W^K<|-
zAZeuZ2T`#7o@+sXJ2Aaaa{k8T80_8ca6*gsvI|8F4qnSA``h<}f7>dy7-yFTD*NN3
zFC*`uPd6EXcFPJ4j?4YENMi4U{?)O63=P}mdo}=i$d(+e^k0!&O7XuU`QN1_AvVd8
zk@sG7n6fOkhsC|`A~tZoz(wqLFoYTyA|<&bBusKuliAo)9?E)djdSz8?2AtiI4SVq
zy!ExvZi<DKwJQjhkm+8|y1K(~ZrHR-$F}Q;ru?YSP<t4S)!~+imX6Mdd60kj2D8_+
zXT9aBCERB^*{P9LUHrU8M-QB2%0W*xj8kk_75(Ll?Y{RHn&Lb<_|pXL)LsE_D6Px)
zYY6nq^zH^UGCMoFa<Fa>z@vC2g0;+tQ@oEJJr3|%j^jPqswgH$$X}d$`5ZMKeP2Ze
z8$3r!aB*@fJ2syQu1~?BH+Q&V4cx4|z2*Yoez%B`aWUG_R>!43d94p#p?|#H*E98b
z)?kW}xdGZUBaM%qRC;P6Xex@@VY9+0FU`m)TOO%5kJt=m>vFhmuu>|(1N@hy6%!j^
z$M_uMl>$BS5$Sn0d!iSoB-z?|;k<$8IGcoeFGA@+rFu|P*VT_yW8~pZ&1bsSy-8e{
z!)?2w`=@juJi|yVHuSVt;>_;`uxF``4*)gJKbgzzClJD<kYZJ!;mi!=eE0ypQ@`69
zYtxiRO3km3JLeaZotx7c&8ocAMrkJb091NN_0rn9YdcrIbna1yf;JeiOBl?0Nsfkz
z?Z&30(EgK0FO6fqw`m4G%T+-)UZoLK2{`vjp7wzEu;ELKm4ul}V5dTw1E(uAq>u{g
z?HtD@FF`04!90m50A`I*^8ON&`uBNth5@H8F|l!RavFw6iP2@(MEZZ!hRR+La&U9U
zt0oJ!Eju%VO3@{EFF0=ii_E|V|35o3^@_T>b`!O3dG=F>%Ess34(FU{)Ql9(*z-Gf
z5VJK^h5+NI33#A~k~ro_Gma4~Fee#d>!!I{Hsw{Z2rbZ6O2PzO2U?rFy_s-`Xmx*i
zewf(TqOs9i^@}~u@<CPiD`Y~$T~F5Y6nAGES-H4Isc{T2%Z5H@VQ{OX9SM}+Yi`~e
zM}!Bj>r#Qh`Rs@sqm8v=Qb{$P=m8#jtD@iXCo{?NQjB_hOlazZnL5vkuIUD=3O__A
zI|3_tMIWD92TB=7N25fSj;n*U)<N>xltBlZbLx)uC|yNG8&-13V@Q$dSz?N77g1+-
zH=m61OwQ-e-+BTu?I|5%h&D<)nv@R-&u`<UxJ<6|+8;kGG;t?Oo!XA>>QX|tt7w?P
zJd2BZ`V(gAJ>mI<eK|>@R`1u4$*ybnJHR`&@?%K|TLwDO%7nSo6(L^lJqqnmvgoKN
z?q}Ao0y_@35v>tYKj1WihJ_8u<>jA5AZ1lVECK@05KqHmGiB6bxlQU7bT`M!Va`zd
ziSlQX=W8#}d#&WfcEyL+jsxb(vRZm3>brYpFtzQiI8sJky*DGlyUBJXM?6Ix*)J`B
z6c6Whw=5C}E+Z=GW14x*=>66*37{3utDz@z6~(=1LV8sbyz9M34Rxfn!qRFID4k8@
zlf3j-G;2G!W=_Y3@lgG{&8J#k`x@qrZsl%E4amMY5H8{SfV6c-0^ggI2|rFyW}Qj>
z&5`PQ{`YfoiX0sg_&;#zSm$ein!t;I%@2Mf`NJySiY3G2Cm?>f-#e$-@|>?cZ(F27
zZU91wB5zJWLiWeboLeK3lIR0%pQWFnz2CRas*NKDi6QIjHi&2g`@t-Eaw>64=hV6{
zy+~|ir)TD0HR}^B`Y2%NyYljbiloVHPeKW}*Q~!DJ@}xE+H7Qy2qJp%$jLIz7(7!I
zof3QV0OjFvM9ARB7$I?}pEYHLjW5^%dY`2z*X<>{Vtf%4-DF9jZgKf!)gieQj=Aci
zK;Ywn1oZK?7BWo=`H}B+)2gzvlI38GdKY4O7miuflxogZO=UZa*k(A8FmecJk+7Xa
zrNZVc0<yT~m~K+<;)IhxsCgSqp(qmYa#x)7etSf;y>A&oqFL0Na0W|%#$tntr#sCT
z=TBSVPHG}@j>l)D3$C@vm82@><QZ_VbNwd?uxB@BH;_?`*i*qXnU=~Ep{NT@8PkM4
z^4Y0<>0lO_dE*;Ty?r#wPIQD`dRuiMGASy`db)|fG2iSx+^^tJNaC$bC|TV1@3Iv`
zWzs7>X(sGC^&e3p*0)gE-P{$2+no2tIh@XrT!nB|axP5GVtCBg4)S<%gx=T{NY$L*
z&#NFpf))wzp?>2Ve0+SZN~nTiBA04iqMW2m&-ptCw@E_>#KH%jP7)sRIrC(!P_e*q
zC@z%v>?dxx#8{n<d)4N~XkpQ|+9b@$3DsmLJ+gRqG$+J~zA(*#OAG<}n3In=&su!L
zY{{Dl@-@H!vrj)eJ9~GwVocG@je}6=;YQ%WzB~e{J|&w+ey~2%eKM_4=pP)+>$M|p
za5SnXa1Z4KH+Bo7a=2r27HDGAyw$j@Wn4aYa_X2cr52JVtmBK?;8K6!`x>~j%MpAs
z#Ww!~(AU(o3D}pqzuO4g8sL$kV;dsBMK?Rz%wt@mq5H|=UC1Ho0}EjL!8W3G=4jgF
zMSo+s^|8xI@vt1sru01Eh;|BQ`42x}Xn|VRTS8Ns3XjO-nWgEO_{62Sw6C*0c-DPG
z{Q>XKo%9912oysQjmBo^nZ(fyJI$C=zFXNQ^sH#i!E$(|mz0b(^ag-xbH}eFL!a-p
z=K6NSOCO%CB+DJW7`A9NsIWG)5Ey?9*q;OWU(XMQ)nW=uY9}`)M%U?(EB(+icxp9<
z3#J0|g;sUgl%VS!-)~6~K?I<CEq9lk=a8ajR$Z|L)-{5?CqL_74KfXCQe9+j=YsY^
zXdEAEKju{-pLaEQb*Nt~tl_fuC@5Kf)AQL0l|lWHZlE#!M9zna&xK0W7bkCqs+}~6
zpE9c9(^3IaLK1$+)m~V&_!aT+N3F83Pa<M<jon|K6&9Pegseo!UI$RP2g`b<@vQw^
z1%t`i43L*;KDov=f{AJh&dIa6f%+HYwV5v2AdOYI84q}oC^(5O@O0l5sVZdXmO1OU
zNQ$)c(+qPP<+^u@ArH7XBj>eSd0^+LTO(fsDlLN#qm?L4y_<|(J6Wv@(_L#<P5B}Q
z@Q88DO@j0d@U9;(ek1lmu5_-ZLkl#2PTtiFC1&9BGT!eeWB~}sAwL^dt5kD8iPW<v
z&5HP+r---}ol)2^Oe%Y$M)i!`Dm{`u?j+n1W_ajKE|bc6ww4t!CKg&zb|ere-?IM8
zq9^D3B{4+}r>-;gvTFnF#jxq;q#xD@!yUs7=LlmjKL3NA2BS4p?m^_;zU`&pSZ{3S
z(&jhJXR#`~qJTU9{?=|cvo-Tpi`lv_y=W~)ts=4PaDMdY4&C*1QI9XE;EYD#g!OEh
z(k$dGt&$V4)xf&Yo*NCG<=ztX274_y70&a_IOY;KW!HhKgjXT8!@-A$b*3}}uikhG
z@Fw@pkiR%8{#6~(&t#{jvfJ;_yYC(YSLuOPpZ?9&uOF$r6zV;<mr}iYQcU@zpiON;
zlVGnMkrzca9LvkMtSx#Q1tcjC4m}LjR{K)rR<-6FZhLNzmoabklcv3cXyemT0v?**
za92pgr7l|Qn#FN-(Hjlb^}9_*g7*1hEL34YjUAjj(I#MG3^vZ<9264Z(<^wiGUKq4
zH{pR*4UZbP^`RWZh<5)-Ewms&V{?Y!V2;86AnJkso^;q~*vaF-LxY!{xAcrHdO;)f
z6MhwwdaDOpd`EqwKXpmFR3(_MKt%D)-x{=u4;3kmJZmZTV0C@+BsHOgUq*>s;y6;^
z<%98S{W2~8<kC%g?yx8py9epav?&Iy7BvoGJh#dn(JNBVSpj<ggTb+FQV;qJ?tDY%
zi`TwFw(!}2M#2iRU8kw)-f2ScSmrDnj`<nr;q=kzKkRAugl2_%ljh(fGJ!g?&E`QU
z^iQculM;bpkEdbOR%bkP3!~_1BSELCCr#W-g*^I#C;=Z@9JOoC#=uA$gJ_vP%%Z>_
zZ+2GoO^5%AOC2N2_%%iH46>lL1XX{eHcseMRI3WZ%T=Cf1swCIPjlB*{U;vi8fZBI
z2NzmU9^gwM0d#+{qw~|F7e)BY0=OaYo^Fh$DcGT4g8JpNRD>N&8EYIb`_ruZ22o?i
zrR`*-5E2H{SJeeX0aMN1WDtYx$u25AwaBsBi7?Givl>vp)1hR<_684GZ`>&G$qOwk
z@>%c5x{-3P1%$hFyW+>=RnHCaa`~Q=StULg>@YL1x$WSDoc}|MyJ=qE?g#_DF9opq
z$DSztmwrU?qU$<IORAwWlA;XtO$=M2JVzUa(CE7|AtN^0{(W^Gn@ep8rm?%dv|qg&
zq`+BDlJpSio7gKm=$$UW->8|TddFr5@Sbvj{nMvtzYfwsSFx8u*UP9RpVw49*ACr2
zAZZPJwdOlJJ*5YxE-aLI%!6^p$q1>k$<=#anh4G3tY{XLXksWjZ87xLHV=tAsKO0}
z_%AR6=aV>^q!4vU9{l9M_uu3vzcTRyYC>Q8nj<p`K+Y>YiRB)gpwgz0Jsun^rn4m-
zVYJnyn4w8WZu6f@HE7IMiEV+AG%Qw0gTc7ubX|e-UkQ5yAa+DSnF|PS`Cs2>LU{3t
zVq%-mPc}0spRh<kROxZfA5+SYNCk2j>b-J11jf9`3=9lhhZNZjXp|X7^CARp7MhAm
zDZ{p}c)rDY01=_&TTO)E_qzHwy6l`<ia{PdYZbj3sO5!6cgt9hfTzBytajw{-$`<i
znvL1h_CGK?pIOW8>}E*)V|rF*nAwdP<>`m>zIFr~G0+XAx?r}4*u?5PWm-h%^eAb!
z#$B9gKIa|)8{pOGD~rpokvy@8@*!%P$M6!c@&o}}3Xp&MBrV_|(;EP;(frB<NE3AU
zG7%`yHc5G~o|o=Wsn=Yvz1y5#8^D%sSv|DzTo>5fAFV^>s_xjhD?@Xe$RcDiSg&A3
zVY?_`bxmO#EsYX5AsT*x!py0c>26ByAKxH^i97<&<BQ=Of$TS!C``hNK1g@<m=J7<
z9w^%{e7}O2nmgBh-}C%q{p$3L`QE6n>=4bs2y$Rsj5}3WvrlA|b{MJ0N}e26=CS$C
zy#4%Po|mV=bKk3G{4_d3mA-~=X4VE@8Z<VYw3??GP@NYf^5|j)e2ra}cZ<;PPilg#
zzjl{fk&=TjNBYBQ-wi5W8elj>T=BRw3Y>0iX>^IEM5ywb?c{xw(yJpEeo7V_+X?<)
z7WrD05Ilkt5nHLt>$)H}RI9Ge!7<IgE8Zka?1eXH$Ufq>HeMDZaG^u_=5(MXBELnB
zakJUbiR<2;lk#T?6+IIVXHI5^=0t06eZ4}sF!>EjXtMZ3T~=^Y9_}ulR()R;ZpNWt
zNQImte+jAG+SJu3JN1IJ;Vpx(iyznkl%b|Z6*e{K;PoVNacaa)I#7D!Ji^zpzE`sF
z1h+1eAzYO!q2unr5Z*6j&Kk(I#~qFxh54?2NOo)(OzJp{2yA$@<GugnO;Fa#(*e~f
zf)zO~xIOOTzRr~A_PQWDrHOlxvEP21mZpJ)Llkm7uaMfWQK9*2k}t0>STHsHr9-vD
zp`#zXi2mwrzynoH0KHnVtx_7y5d<v+?ZH-4U(WoK{rL;;nvd7+n5v@y39+ch)~jus
z<Kq_*-BNP_Vy`4BQ=RzXZC9*u*K<!J^ZS{cL1ERwj*6D65s=(!`4<t&4~Orm-Epfs
zpn0StZhb6%(I&LiAz(IIrX*wpqy{Xv4IRD<FZaTz`5gVTszUL=w&<Ac!URS?#P!ad
zlc^DZzQM<hW{x0ttV`$2g5TmBtLs<%Rih6HD~ZoGPToA&8}Ju<`6;7-MoH*4fQk=}
z=~Z%_vC|Qf9y975g%K5%#e8h9%1k%V&QmM%UFLsld&l8b2X3);_O6dZjE(hZa}$sI
z*Mz>d!#P1{nH$;i?%2EbW$PChxBe=@n6W6_#vB$b(LQq18OuQ_ehU<;KPOaMBA+HQ
zY#m!*PmhN8%zBXaY7Uw^hfah`v$JeAIUn_=8okgSo}@G~2S%FL6Iy1h)oXW~&UOZ5
zio-H`S1KEQQRQ8AM`_^At+!4ca`q_e**&l#ZzH{OLeu?zaiq!{<Tle8?||C?&v&wI
zu}{wure4po?QGbUL~HtFkvWbS^_w~d-45E$l(>qG6Ych@9YpQ3SN-3l!Pb|-o4rC-
z`@`y^`VTU|3G4T*W(b}#K&M(B1Z>zCoJ`t1hLWZ#i`ooXx$e}i>RYW>KVnD(_l59M
zNjNJrHO6$=4}VxQOEH(n$QP;zY>4DHUu6K7ti6A}eP*;j&8bCAvBS?o_JfA|jha|n
z(nalhH>w;F49~7Bzi&8N+SNFJI@gtRgH2zXh<b|1P%nAI?@=R9X>;m<v6`4etg(Y`
ziY?Dx`Eqwreef}5%qy`7Ix)3&Q};|1br^n2OWGmsEyAU9G)WOAAk<fOagwO$Lt6hE
za*0>l9|r~msKbsL9MRce&<4hKI9FAdsK`Ii3bofy(%0Kt^$|x8WlNqO5Wn%Q=47KV
zy`!!)4^~^KP<lozBe1$pR4EIRCy;hfn9<vHsMy~;nhSi1x|8bCo;XGrLN98ZVbsF!
z4kPqWAzNPyM^Q6XzEqHsh8LlyPfCZP=AR%p-H<C>p}zQbgwkBCb!UlJ=<&r!ing`-
z1l@XFr&bJ61C7-O;VhFF^P|kK+O0)o|E%|p%=ybZ1U!U!AoXF=nB|0R0bM5f#ct^%
zk%r-|(Xr8kv?77R!vp%>^PjMrZs4YvUgNY(_7rpf+Kuguo%$n5z65q}o~V+vv;ZWR
z>lWmEOk;a0H^H$W(QahmwrJhFER19jImQ#yui`ucKG#j8|7z@}%+WX#TXOx7;p9%E
z+#Fv&zj&Y`d){U|<%wJ#eAo`!=(uv5p0*vz|M+D^ZK2{k9fd6nI?5?__@vhR)SJ_L
zW8k1G4y~(RYHBn|lW&AK3M(JovCAn#X!p|5BDfo+MU&!MeAkRyifNkSB&IwSl5DcL
z<&P?>lLnDr?g_yx`9b@6VS{GH1@MDj>la=Z7(<x5p~VhYA?)h#(fIm`Y2{3vN7Y8Y
z1eiPt(63i@=T0oKgm22uE!HKO$P*q24^EdYELpH8B9wB%uCRvTOH0gY_s~{tLYrsn
z$ZNUxe6b&sicL#HZVOF#IM52M2_pj%9eV>u{nh6l?fNv5el}gKJGS{y`xJLqjvlKr
z=L1cLlXgXS-6Y&!oi$o@7d^}$7H}VLG=lB8p3X1I1rChDz!2`6!#_8_XXBxl7S+S%
zl4L-dl&{zPXE6us4qO|)XDIkG)r+FlRxC3p@H<X^r_D~V>Gx;`TymIc29Qe}M52NR
zV(DpS;sQ>%gc8Fn#YQ=lLz7l#g^mW;`hq~t*pR!t!aU)Y?uMBAD~0qzv+JRYrA;YM
zKStWIo7tV!k%=F?sXWx>89e6QBa}`Z9vR3am^Atvl$R;fZ*jb5F&5H$l7sbmnJwTR
zTAs!^=*7V5RT|H7l{7z-#{R($8|YkUE+N!wcXlk~E9o}pDvusmh}~3lu;-C@eIMzr
zM^vZru+bLz^{;)3yHet)_JYtIE1KgP=g@faj;d_vV3QrCh{pncbfSrymG%54xaa*&
z|C}K=jpXUbq*i;l4CR|?%h%riyXqI4>*REFv!o4EZ!erCwR4A=+aqZ7cAFxpXAey4
zC{6gs-tOa)_JKxt1u|X%BNhWE;eIy|4E*U*VP|U}uyAZ)*MJUkX<FwPEqD|ZdiZXm
z7>NVEm$jb;hYy&SOnqG&6C<TNK^J<RcyGB>Sh%;`mc(9>IU8t1-8a57M~G`J9tL_E
zzli-ux`7xQl*A{e#rPn16ZnzJ2SpYA@!Y8qrH;>X8$D0j&cD>=IpiZ<58%1nx@&V#
zvoP|Z5NupYeFmWu_b$E4CzP}MY9(Qvu$*Avs3&xm#;K7C@hwP!=gGE*D0=BeZ^E~>
zu_d1i3q4?DFUKv9-<#}?aLWd0lpaUopw%qYJ3et1w4FGxHx?_?y){~xhyF(I)J3b$
zyacBN&a7-B83Ol(k*)K>75S@XQdjx&?IO|-)olCQN-bKG+!O9_>GjKn26;rt%s93O
z@(TbPivld_HI}b$X`xQ`7|Q5@sx-$6;f;5*64`Sc?1mVejnXvwy0sUrp@qwsQ8Pmc
zh$jPeyqXY+2m*m9svkNgr=uTf6R%S@-=2CxFs3-QPwV!Q*g>8ypvt#H&>?<#7aAGI
zP*fS9q9)LqRHsppDaUxuTP5B&v-k>$&f2E&#xCON=Hx)#M=HL_Q@8L}<aHONe87g?
zT2)cHzmK(3%L_tUqs7Uw>f?cfN^T(sWY#)Wkp!uH=xoF~_+ttpSzATjjG@8>+#$2;
zc2;(ndgZ{W2^~_z3yMsWs-H87W$4EVzY0-BM^<NP;(`J4j`IM|6kIE&-w7oU=~EI4
zfcyo`b%2~%kmr)ydyM5JL-!XL0s5!*rkCU-5oEF5gk`Zu)-e_L2gSo(qW{o*&gFmv
zB(w~*4if)3-f+n=4&W+~Gw$MCQi1as5N2P%#{Tjd2nxiKp|#Y1`NmM}Bi-113il=9
zC=D!8UnZI2(XvSc?)q=s{$^JKph&w*zLUcF6PC`jAnjZK&!zw@MOt#nk4m>#U>`#T
zg!4XJa&Fj(3I03re?|Qt6`^G_w*zST??J<a2K5*hlsrFi{;MnB^C3{z?TuR;nL2&9
zy56A5a6^|It|z#YqkP2a|J5-9P`bS-W&%N<8B6O6(2Ba29UI3y@wrC-SLqndV<6o}
zBa_cKxk+*kB5~&-naof(v@v-wf=T?}T<sQmP<pr5S=}>Eda_@yWh4@g&m`YdPokv4
z=CkhW|7$?`Bun*$%Ltw({U(vvXc;ny;c?4d=Dc^^9)b6GzW-lov!5T#BUMc>6Yude
z<EeFlEjSMk^WEMfGj5le+Z!YZ?|+B<f$z7FyvX5pwi-g(t^9<Gs|d*FaXol<EwS^~
zKY?s`AxeFVf9d`oc>Jf&a&<1y$)~3a&Pv2$z{Up~b@S(6c8U3Di#XLk-MFJhvPXM4
zdd4~qdI8B*!tKRHoVVJ4!Fvr~I`D4Y!|yRmV#7D%*-hcjF6eOcinUxAOT?c%E3AQU
zQ0_fT+CB=n2HTZM|ABaWA3l}IeVh-S&c~BZ`kO*6%@~0F_OU;#J96{#NZ!52(uwy!
z-u#^+^9$>~M_?-gCf_Ih1qsb-jKI_2s3(_D5Oj>KXjeDV)V^FZ<9#eOm;X?U=n|AD
zq<(RAHk`KaFR?^c4$I5kdU~1p2f|yB$5u!=$f^DnR6dqD%=)Hq_isMNDYjxUuQ;FM
z@36P9IUG87IPm_io&j5NT6s(0Zzla`CTt=IF>~<WME!kcY(;wY=iI-;>R@G{X^~Cw
hclH0fg>o90?Hy*n8bj!a{r?F-QBF;^<cay~{{vwg+7|!-

diff --git a/docs/assets/images/logo-square.png b/docs/assets/images/logo-square.png
index ebf8cb8cb9e1d7d59f427a2ddda99de5d12e5ba7..9bbb39243415b826a71ba65121b20ada5a7b9503 100644
GIT binary patch
literal 3323
zcmc&%X;4#H8V#<vfo)fVB9L~likU${76E068zM47+wj;43J5eIgguf#5NSc1B~AAz
zvg<H(h8MyzAwh%$Xl1bt1SJ6_K%xyqLPFS*K*EEGT~^OHwk=a-)%>{is_w1(z5AW-
zob%m#;|n+E9b5Ib!eFo+E@w`C34>`IP>;WAZc@(>$KYu7XG_c(-*^}dfq6YNt}w}f
zdU8*KlTU&NCM*GaG42u!i^ZCL6^)7yxfpZF6cZO-JZ-NBgKfX=a_YF}<r2wA(DiX)
zVArfPej>`+eAUmyrKmLb{CB12HfHM?{V9U{^z0`Gzi9{!uN0hf+E4jogvp<;o;har
zjV9&Jt#3=6Dv(xJzd$v3?(?Mh&?eoi>9-4aoIvl|I|$jt8=8GLfDA3NUvJq8MIp^@
zH!sblC}Mi?a_lUtVH6+eVx3oOKZ+Zn4GI@T7{?a*Lu2PKYj1Bu3J8Bzagzypk6<sO
zA>!$<UVZ;1$*4%Eh$PiU%G%~`$P~n!k|TsM5*`B|BHvsIQc=fg9td-b`dD#yDuwnv
zp7z9h0c<_mLSGrkFU<Yfgg=6VAi;Te(^?Auh%FTp=3xe#IlDz?=gGP&yRUA1xbwj0
znHMz7H*CAU&*#|o_I1IUzb-|n5|@(W6UX^vk(}X3GMm6O)g{bY8d-WPo_CQ&W$hI@
z8T=s1!O_GTk(qN?jc-F14uv>+*)wf+>a${62jmYG-F0oDm@c0nE8F_Q03lg4EMPry
zGFZNupY9vDJPWR<GK4u(l3F&Ko)#54CUU}~XmV1sE&7$ipda+0$*%<+I-%@nZ#JX3
z!r}g8$7Xe6k_^jc3qJ8AvcXd|i@0A&na7scLgV@QJ=_XjM!6IWT=v&NPrfwJF4Y@G
zUqUUEfRSbdMlfSYjz+D{iMv^yq9^msmq%&{>Vr;k3p%m>N9-2XiM^&$7ont-@wBcW
zI;fYzmLXlS)wvS+!fb^usNLFFu&Fg|(0<pt-$d&zAaeo(38MiHN;m*@w%6~CoFVX@
zM$HE5PWYF@3k!o<L?Hmp!m2-9YMVB*+;?qq#5QIY)RuvbMAiX61~yjk()o4GZwK1X
zOi?4HhDC)vgb-A$Ksz&R=$Znk4Km=>Fav<q!?y8(z=)Ma9~nxnC3hk-2-8{Q9s)Lk
zsyjV{w6l7mGNx-G_3no<5@J^-yr`SAdXrxu1&JA!*&vP(dZ!VmoVg*h*<MEFJ!V6l
zpx%ywI!Q;m5VP7nDLfsfCTaj*`7AmcSrEP2t()2VJU5qHF-ULkGn0uLC0)^7M8j%F
zIxiS+-YOY;Ryo!aVj}D$L;>7w2yyi8GCdAxYG}v^O%-=1@v_(_Flf|*18#z8<ey4c
zG;_DPN1hZmAj7fvoyRu3MZLGce8b8RX4CySj_E|EG?6Can-^mcmKO6$QgkYAp~T5w
z*-pM?s=vE1Sw4JtHF@gQ@#aHPh$Lzou*qSul!-w@<W#1Y)SHp!h)@+9S%x=;KfvlY
zY9W_r#4iDDVy`4iLWo@?@RpTSIAz{DZ5}C}rwuhde--q&#??Crw;%|n3!7!GR2$PF
zaejM!F}A;qt`;40nW9{(>^RqGpQ6lG#p=|hsg*ok1-eVRd=0mxe?VftmFI_r`O$aX
zc4xGW3s|BjhyYJUtjOW=S5z-;gQKyc418*DrzjA?QNAp24#RQz298ZD>G_2{s=O3;
zn%v*Wk|Im9Ymh%==dc29Y#@U;?h`M_jx?HKu%LWM(6=Oa^f0)JymK^RrnX-s-zQaE
zg0w8>&YjWR@X7m~=S_mc?jTtCT+TMtakX({Y&JvJGawG3q+ckD;7ubetf*WqjG^6}
zO@=dsD(3bbLF45t!5;7Z?I1XaNKlslFq%Gj5OI(21K69xiqf}>3!=$y-OcbTMI;16
zD|h#RKN;=YdFq2c7*kLOS`%xR-DFC7l|oQS8(1CdyJ76uGAJzZbx=atVENv{NuZqU
zi|^`YMb*oS)FL6~uo{?)!*h>}39+FQXZQO60AnPOD=o$CNb2Jur2WUU{Bt3lYmwBW
z%Y9XMzl|~sO2$uxw0`!tA#rggb9L)@{?2zT`JKw&bM*ljI;|W({0Ij>^Hhgg*{_Q2
z9r#kktE4C&<rjMK_ErcNB=W+6&fZB%YV1OZ$$6e_(%+?B$LC$Mi=&el8*~k74i%G4
z6rNm?{O1G2um7tHyz|R{H;9kE^oEDC71AjN^ODHnf-!}wt5s?%Yp+t=F1i2!QduuZ
zkaF;CZa54P;F2AE)~2!MuK}*#D#J;dW-c?V``(wFTZO%+ujvIC9IamwaO_}aY&8}B
z56XK3;e>}-%#mld%()uhT7S~$`lNqG$QE9FI+bQGGxYeh8jNHx8ogVwzaY}s1&RRp
z+6wg`@JA3Fg;#kI-5Hz2|4<l{kT2ps_W@`raZq{{7KSWyKJniE|Iy3)PCyn3DTu49
z$Bn7{P_@N4?Po4h$Te%QY;lZ(|I4{L5af!}MbK!K;!0^lk%Je2DZ&w}lJ{@;k4wrs
zG2hU0dY2bKTOo3%hB>IVJik>j$0R^YZBVz|tN93-R@>wGGn&75`FUU5-@h`gbyv<3
zXIp@lwKqQ|oJ>KzF8EUvAt_lh+wUPYKdkmIup@&p#PGmq*!kQ4l6Ut1kT<}u2HEb6
zjpAyI8>4;vSRELdjD|`*q_7tIkFIa?S};7xTrmVj8m;G1mr(2$RZT}X>n_GY76tji
zi!_p?{n92?+d(0F^pJ%m`Tj*+_){dn&eNies%Bi&&cg<{xeqcP+{!yp>p;Ls*%}~d
zoyT9I4}`;cSrsdI_1O<soLrO8J5xj(nPn0QWr@U(SEQ?n<DFfMqFc$dg4mY~sw*qy
zi#@CJ%N!J}V9T122x-J_X(UG?R{lV}^WzGS$d=2aGm~Tm<%-7>qMs(g`ynC+0WkJx
ooF=m)A%l%qHm_eBt7g+qwHY~_RyAv@A1*K#C%02ICoWw57oI%f82|tP

literal 13864
zcmeHtbySqy+CE5#AT3D8&?22fcS$z_!Y~ZoGc+PCDGky}!w^!^B?uxRjdX)FGSUdY
z(Kn9oI_Ep*_pbGQ|DCDz?7i>%+WWrtb?^N=Yfprxx*{$%1vUx_3a+w}oHhyyYTS=6
z7ACSLpj~hP1%>jXkFFt18{|pn;^u5+>j0sHdAmU9AYQgsC@5Yt6B$MxRLuzvH$0IP
zGOuoj1$&gIcT)O0%F1ao%5R6C&gHPE4CpE_E*<vF%>{Dr+<ZR}yy?nGFyW=&NU&46
z|4asVLSP|j=Xz|&**6q%c~w&Bd9u2>u@|&Df!IF4VVbWzIX*{(PQe{pq7E=0tTqly
zJ}F-IcV8cq`Xu*+2)HniD)&ihVgCLupHy?Id%@$QOQvt`?GlUa6CILyX0}aUm^+_R
zCBI8r)(SBoyf#znuUG0%V^N;m<bI72iddXvxOaea@clewF_w{aja|I;yYz_Fqw&La
zHQF+_r{MA_QN;A)PVxFP4Hb3gN;7{ahpU+fipK|5c<X~|5`Ni1&9|E*Y|U;?vLQJF
zHIbx5rRtw@pF9E{G~h$g)lBc^7O*3@c;zG{lXHR$@J&EZR=2P28kcvO%wmc<*34EQ
zM0TaRmr`8Qgobc8KRthMZ7|>AGkN>_=e_Dme+#wEULN1WeKk_w!tyC@>$mR)#@ygF
zcdgDp#y%U+Xn#{L(>Vig*}`tjOspo9O&N1`>H>I3s>CRER_1(qW@!+Mkyl$rP{R6l
zQTCh7Gx|Wuu#-75?^f{e#`)cV$d@93=wm|^xDl;mwSfnxaTEXt(=L49&xIQ2GugbH
z(*1C92c68q4fg45V^f%)!x<HK7g7A#$IpcqVO28WXBv1Iw00xC1YlL3ULt#kzM?n_
z6*iVYi++jQiC}f$W`e!b`&3N}Wu{bZJKG68e4CzhyNZIz849SNf|FmftE~OJ`=V@Y
zVNtg4R$YDIWi<DS5<@KSw#}HDWM^!eRW@XDI>oeM6q~)G_UHk;;;n1l-XNanel3Et
zdH=+!uD7iH`&EbCNmaUN%4p2_pq*bj!XArVIHH;y6UuD(FdK4<YNpT1po(7DNL{eG
zl;J_H><88iHTEb_=bN&q*)|93IY%v{eY5GowZl%QeaX7a_$_~%#>Ksp<cY%3VPVqd
zJWZ8Wv60?my>xc9zJ34s?|$P=S<DSu!?v{Iv4eYjy~PHz7W-2$TNcHD0&2Uoni*<Y
z&th!%W|~mDDJKsmo$pqr^gI1t(qd$k6huQNJ$6yb7CY?no^>-gOMM2FaikY!Pj$V}
z^3=!lzlCqYqa<eE?3BeYiAY*+mwo=U;PcXe*8VV*cQmpu_a(fH26Zfq6I{kz>%TPy
zntkDD=2|hW(5>7Qno(vct8jD=+{M(Nt&{*?<7y4n<jY&DD+ftXJ;XcMX7$hON>)S(
zdbr^T<*L`pIJkayzq)t4%d_mh+iaRwUCnxK5_OUD7(TbH+>$v9cS}Q@beg7pl3MU1
zYQ3PXNX8}m=q&AU2H^so@wTUY6ZF0kdE*Us^HU!7gP}5~hFQ*`N@cLjM%JN}RN5ik
zE;}+lV2AbNQ7tp3^YG^h@AV&+?@APA_Q#T^euAu2#DpHiIoOJtG_Ng0X7WvByLw$Q
zIzr=%FkK@rjJsGHrv!<}_?QJv_Sd?#vt|R(c?MhIRpWB^yxNUpyJuC=Lhe2?GfgSB
zINN_Jir4tAHX{6SAS+j%?n*sNAtiD96d)oVJ|k?`%X3$YGhYd`%l*9Gbx6!_yP%;2
z)mKWd?z|dWQ8ge?5z@HcM$BQ{oYGaEvi#26ozW>XUtP|uZ3aVERjjD)OF&>PINN`H
zt<ifU8u5BepR}!3ks+1~Y|u^E`SOm+@e2y-AfTswCF?hTU59BbL(jp84N5Wr4PF_a
z@h@njv!JVV;TxrC^{GcTa8Iuuf*l)^BaleA+M}9#E!kqGn_u9!cT=!TwVqSIb5aN+
zH+-uwrG$mS{w*bf&IM|gXY5*{-B$@ep>Vm{?hPFb3~iAU-ngr3U-rJD3o!wa{WyeX
zy;s~wa5J}`ddq}(6_Y+%>Wb@LiFt2mO|1LFcC~K$8HY0i_?g-P(>l*#A}&@v%EG#r
zyO4Z}$V3kcIde*^V6&ho=_-_Nj(~+*K(ma<H4Z-RYr00Du+<``>ycb~MmIp52a~#t
z^oG48Cy0Q0w3w$i0yy6~1kN-*d6fC|t~r}R7)HVpmedj%r(5VST4yE3p?q^Oo=jU)
z%sFeb`%zDuA;#FI_LNddMfhXtbIe3e*S^6f5OwP(N5s1r-fcEl+pML-bY=*^|C=&z
zHxw)GZktsx2Ymt?l{S;oo@#Hz4!IyEK+>i$)c~edv8=fD82v1;@|6rA@iW=Sm?b_5
z6w7?843>1jzRuU^2td($pK0T`@>J9$6HG3=xXDLAv=N?hx)vN;E`5BIgc1fqHL-US
zmOG6m@ro<z-JY(TLoT?C017G;e^+UHBd_GaHC!B<_>RLHEozk(vKI&ZfJjv>rQueY
z)Iet`Jox*_JE_h#+@2ql7){YuikBm!CM^Bn)aGn_3FZX&1-vEJbP_S7=<n4Ay<;An
zNwjS=DqC?Li1jI}aJA+vp9lj(mRmVHh$w(wOVfCXsAOl+^RL*PQdmj?Y^ykxckL(@
zTRx=a5ugFV_iqPT0GCBzU~{oVLoVI~$kA>_=3W0q0iyLgG-mHyOy&c~HDiu@J`B5l
z9;9f#Rc)lX{C1f(D1jqFU{Ijt`qSfpfqq&xC0|P7ZG;qF+N|umD-M(n1H2sqgyPZU
z+Csk0?kGuOTJ0^elZY3c5~Gp?ckzQ}3nUEHBpgtqHxTp90*bEdGFs$AA-OxnnmzL)
zy(Mv0_%sidO0yGYLvCZ6#4#lWmU7;!mSL<+MqA4mXVjKmvGf}Ct10@71~s>Mh;l;~
zQi3soQB8)mLE!Evef;LJa8KENjF-BMsq~Pc@EgMoGT8%lX`qK3Ost&X`!ITb`!I&n
z=QlN-{MjbY*%@>d>+Mm(Jyc<nJ{~W<E2D*T>6ZG_-f~~~O=_Cb&e@VG>z4(8d6O|f
zje^-=uq8$G@i_?^@i0qE>Fem)ueXD!;MzM6A9isM2*eaKJP|>O>cykGW8d=qWk@5%
zmk=SVjC%W|a@=d;_q3JF6iQ`;{R?H?FIlevF+sFsY)#k|c|GECXywu0ZAKF|&db2H
zt9s`Aa~NTs7_x5+4e9&SITNo=siI#E$kp!VUo}bH7xOS97Y5u!(s7SStR<%@4esy`
zXgc0@mk%IeO25Nfhhl2x1ib1yCOg105N1$eG7ZelRit!yU0A0`*PPT)zVeZ_vJYw-
zDRW(WcLX#WLxn@K=R)d-9oAqS{y?oj7QssBCSR2pX~1ZJogo^xbH~!g7<Ttopph+c
zjLkho45?HLf7zZ?;9~~9Z-=jNb1mdgLpWy`H3OLEKnqDUY*#_DbfgcIvgG&_F|WSM
zm}8NL%U#Gie>Kmr)L&f@UFw4Q(8fPc=q4{E$Ic`6>L|+B+MZ5~Am&UvxqqJ|anr4E
z?rZbsxp4oL0%s+Fjo&9yc6yhEHiC2KCxZwDom4*!YA4lfHmbf2hUw=1PwyGXQ-ba2
z2)DQ_%!NgiT6fz>(0OD=8b5on15%<}Z(eaDq``}<yJ{?E@LdyrTK*Fw5aF#iC>l)i
znrJ@N!R;t$cA*3_ngAG3{-JYjmau9Fam2YHi41OK7oTZB)mgN&wz|A|x!|kYHm};E
z*7X*(BD5Bd9a#?_(#hLmc$t_FMZAqt$xLFR+8<w+<4H=SFaD53@|{ZRzL(nJLk?$>
z)qt<7V;>*kyi*H=>o%fot)Rn;H%&7y>b1T*gVQ~O7r-f^W39A0UM>o)y;p8>h~$?!
z`T4pA3z?|j!qVS`f4ntFG#z_K;URi0Us9L>=IYZxYSf1q1)<mnfez*)BPMuBI!QQz
z=t3ownm{pvw>i)gwuJg(pY${a7@0|Bk+b~Q%_VTsu=>Ntav9YUU2npd_jF$KM{LVW
z@!b!ls4(vP*bt?x1L+DE9JI$PCLBY*<s|xLf-Jb3?0ZOO`Eto>_2~nPL}5nfV#ytn
z4i#rzf4Iy0%Z@9;#~95_rmq{)*ON$#Pt&7D(ao|v7W^YUX(Y>!#$>csQ>jRffFBI*
z83GG-lw9K|t<+u$@Jf&;m|$_^H*UYj_-v8Ghu7f349PKcyLYzdg|ir1SZ;jWBrDXy
z`K$r!D<YCC6BHuztt<O#UOZ4Zo_h(Tegj=%k9N3!T2gm3Y0-yh#OH)y&1dOlMCo+q
zjlCvX-Ae9oRJS-=IfA;%hVkl9h6I;)$hm}JWCV<L7*V78Jj9CHD9WX!#ZJ$(eB{Pt
z8(Fp?ay+HkYYWE}Pi#^ToDnLS?KU0%U`k1G52v)hgX7Ft;q(PjY4NMqSnJePx1u8G
z5j_ng(8r7j3=GNqbpU;YR)@|;mhBX7o8<(t@Nmj9YJ|(Yw3<bK2iCo(w&8I7?JZ(E
zA)Jr-^J)~r!Ot34#bb$Q2nxmunl6^(Je}H$kCzB^WiOpV0+R<wU#_WP@esW3Ae0$?
z*L31?cncbZ@2&IVgU}t$xcFvTRQNUn6>L<9t>d^g_jL|V5Sc0O46aYb+Y(P4EThrH
z2LlWly(gq<BoW`F1M4qd91Y}j`zM{=#+??LPv=UaY+k^H@$V+<E#zQyHN0Jds5qDo
z*CG<`4&gIYVD~X;1O~Onw>q>H2e2)^&Z}tVXEjbV&=4>b7O^}fTE5I@fpV>m=%m*B
z8oK47guW*7=-`^8wM3_={=(u;zqBh}C&H+^cK7Vb$Qc&5@P}0K-Cb=s7QTKCGu;v_
zj<2U2{<Xj`@lCo3TSr+IFKAg-_*&3hATN<D9+l;GwjarW?*n0Y7z5c><p}dik`hxA
zn<Z->A3?5FqUD1w4>ZF-hPCA7ajNm>#9>njpD<peH*crtF?Bm(YY0quJlHPbZ@b$N
z8g^@NgeHyVz_2)uN{oICSf+fel7`bT0XKFhwV;@MjS8S`bCFI|W=E|HKC>jO(j|KP
zpy(ww@VT?5gwBv5H(m1+*5KDfc0_=B1KLn*{GBHsl=l?aFwnK6?cgJ&2UzCXWLmc$
zTC$826vv~|b1^dHyaJ5R^-WkB?>v1FTl^XoSY%aDtjwaxpp2E?PC7Ksz=X-Iri?mB
z&UaZYR#t$gF9Z#hV=8&{Il=%u)hA^;MS{IwsIO_?vG__(Wi7Lb;SQ^SfON1=Ud&<O
z)==D%e7;FzJS?clk#FDO$g#r9^on^b-+-4#@6GdE(#k3S>RWeF-bcgd`{eUwKlqRK
z9-ouU7mN<{$vj!W;OJyUq?AoEr0IdMQ=6F(2qvTQSj!VmD3wEquS+}j>1#oj2(7c?
zg0bSAu$du)p3(9iY9IX~>)Ir{fR7=X9a)bB9nh#lzrbheL&Q0WU7Z%g3EHyhL_K*M
zSUvC<`99qFIzKAi_b4Na45!rgWRCa&6E=>yPGJ>)cz(JLI#>^++A4~7h&nVJfxxT2
zrM9rWf}P1(>JVs;Ykll}d(ODUYt<qjrXN%EwvJmH(vytOyPq0YAFe0`mA&F%c=>?s
z*_{%*B{hW`qwdrIMKNgq?LIfd_it;*8wx2@UyBIQDaWay_sNG+U6Bvj^tGdYxFz>;
zpc%CWK64)g%4*SgA+w$+a>Hn`#~O%I(h(4p&T<EiLNlZEZV}6=G_6$IF8@p9*AU%<
z1+yhKd$Z&O(uMmIn*-eOTAE4@(v-vtJsTM_qr+_yY^7*=(nphp{m~sJN&QQbRD22q
ziGJu6rT3fjq9p8$RYkGtP||iDk9kj`6wEjYat3{wB=3eFhA4B0mmIurIAl*=$m+y#
zsKyPaCF8a}mz}q2E8USudb)q7|M|+(-opV6wgu0-pRSTto?3aOzcNV-RC>rO*NwmA
z)oB6XIyLw6GAc2b6w7_JL9>1zb!k8^-gxwohoHqh-!4T=FIh^zl=?VH;5^CM9<Fy3
zh4Iu?tM&H!r8usDvLTquhLI{RC@ypRCOB}gIbQEor$Pm7^u@)3%LYF5+C~>)@c1lB
zYw+~lwW|K=P1gGb1%dWE>~B6|oG6K>tZGh{<2Zz4R4hNl`@(e{6!?VNdJ0vE05`WF
zWUKB{f#O)DEc9L6JCAEt_w-MVHFYT<cLc=_!D;(o)&_nbB5Lm3)MvM|rC_)cvE$hY
zjlUCkA09K0!BuNFIf(ftf$Ubkb7-xCg*+b`>Aj^gf;q<erb2N?m)g#0b5`}w0@md_
zFv@7sh{k=({O|qB^n=gZF2r;kW?2_f5BQa*gK51+5rU?n2~7p>sWB^SP@2*lUNa{~
z^@2aFUT1vLqMw;7Om_c-tqT$|WV;jOZSEb6uiH+`)J|NrKv^04<r4{FYcyWsin)cN
z&Q)`{n7#Dmyv@D^AmMjnA3gK+*d_YdYp61cwgMz<BD05C5YZs;unnivl(m3nplyZi
z;)G4{1*4`i6UnQz6;?k4u(pHtowR^j>jq2^MPIGpbW~Kkv3d5|=yW@+YJ3s)%yh$j
zS?jv|(pf;V8-ujb5?5GS2Y#{Zc?#N?R4_d?`Qom*+!02{ttEJeXF2hV#*)*9lHo0f
zb@#L0=o8s(0ZHeQ3zp-iZ`t@6H<<*_qHPg5#4E8pjuM!4>b{u`Z;RurWn4Lub7-Er
zOSg8GefRnRa95vD2o$U`6;KZ;V9iIA%ZCc5S8NAlefi=U!|}FQ^T-ad5PEURzA?pQ
z_2SVbOoB^mpqfJ8>Bu}am4UnGYrMax0TIglJ^j|r)#>jd*T-9<JJ`&LsKgZ^&r1lU
zh?cTgA`ZXCaP*5+aj;FP^~ADSZW|PBG~3gz3E2fTR1Ckm5RD-@(C}(|?<w9C5&lm6
z^vU2>jbDqEgh_sJ>e*8`4OX?jZa_%+o`_UkCd?<ucULr=i9=5P<AMnJxAz5`9|y4g
zScgqN7aPEg15OVd%oivF>mE%t$=^J+4(6)xd(#QT{mvUqwcJX-uVmmTj);KK$Ebs(
z(3iKD2FJZgQZqhc=&V1*cAHBK$;112=g7~R$}w^3sF$t{Jm=S6DT#hAS>%LR<-TE#
z8?02i{-k}n4w2##9G%(TG$hW<R_P044uq<wp1jyI)OUO-n;Y!$*l}#%Bh+rziOoBl
zWt%opRBtu$)@F^X_5@fv!rPNku_5v7>3L%@Q6ts!B{!${ABJ@1<1?JVn#|AJr)lw=
z+i1+DQw9X&fs(P`>nu0?Ca%ZkZny;(1#T3$4tfQjv7AZvJ6@|FK1-Rs%35mgp$VWN
zaRcPv0`{m;SC!gRK&~F!<!bjDC)f1T6&9m?VjEfrK!LZ<ZduM?;b3QF;*@?sxH|)B
zi}yr~J+_B@QVWjR9_^{sWmCeizRl%Q&cI{W7av$5%%<<5atsJ%0G^elE`HqdB&j(s
zw``MZPH>*l%d<i~>3JnsGPw1jx{hR{>!qU)wP>>vW}Dea2C4XRWw2^T=B)!;pzF?<
zTV&BX{~oG~yTiC^LfN)32lbm-U3E-IOxs(H@Dd%FWJZ2<q3sFSXrl?!yEOGNwtQ>P
zHK*`J)1v*D8TD~n??TZn?|RV-XD#ufS6$>qjvhkUPCV`RM#Z#M6xa!A?^JSOh*93D
z4^Z`%OQL+Fv?pUTCEMD(FBy4yd)a>~_dJmTxkZ?X@ZaQjw~vE(?6kc_vtrfN9=IpT
z;(|q}8|onU;JDs)OD#UE(%syt3-%+*{KpbSERr)e=FinBszvC672BzEtFIkI=Tbkp
zzJvx3J;*X{pl{hWOE$WR8LHan`|<0<vsj=xVPmMh8DX^W)}z!3h4r#n#8M&r3EN{S
zo-f5tLOQIueDC+a5jG)g8`e^?>GOoE@XYT~%~-dqnWF90n<94WzxrmXS9_a|6n^$Q
zi65GfVDEo^QZ7v{VPurQ(aJtuJH#%0tF+_uOZn{Y^P20v<oiZ)E!u;d!`WI~r*LwA
zk-&4OupD)B|E7=w=Cd19#V5vk=N{VZEKGcd!HHJT7=D>kL3y)7?IVu73(>v`qDq#&
zF!z{-wIY!+Tw%Enz>$kd1h3>Gm)*_;!bc->upRG`WqVLksP+1VaK_@hY@_rd%dhB$
zw;!N-vf>>RJjk(o-EBUBAE;t^-^ZhGmZIrSfxEaWl)F77K~{UiSw~Ter1%ps2a=Vg
zuTj>Jz0z6<y}Mve-02zY!F<+ia4)W&Y$dBaY%FcV@1TU0nkG)XC|TWOY<iP2q3%;d
zozB>&Duy`mP4m>k&q>hM4nfQL1Bf>j;z<ahymz^53}Y6jNOi*ZgrznGBJzSHJ&HLl
zEgI=0YiH%?$l(LhD(31q!RFA{T8A8MhgpJyX2mtTAyoNeTuJB5q29!+KoQ3qx@)0G
z2OqlLX!#_cN<~_>*V>-4ijNuYaW-|h=>X^zjyQ3~R{=hmk${D#cnpYOVhJCz9dx5h
zi__jKQO?V#kLk~PA<wjo^3Wt0P6^$=B~#q$k#s*&`MNr(m-~XLl$uY>BWS?fWgoA~
z)4^nJI1O)jP|9!vU7AzCctyBZ%t<|^D%32`>V_%<NGfL-cOdp9r1viMyR7Eoz&D<c
z{c0pO^*<&@HL*Ko$#Sq`A9A$-LZ9fNY!MvK$lsoWU*y(oXYy8gs4sjJorey~!S<Q5
z$yNvtTfORC+U1DH;awH#33zlXes+gn3<2)C{&EB{Pwxc2m3WjcJ*o48Jd0U_E|Ug~
zT+D$2-OrxSh?;#V@?&CeA0NId;OqPP2xL~gBgS@J_~vvpT|?y36k?~mf+1ekWb)iJ
zeo7{K$~=thCEM-e(xGLYZSq+ZCvj@>$&VyM4{pLiVbic-#Odb?s>d4|MVl*d&9a*o
zd+;@HFroON%&-k8lWc0jU3*?7!^e`Tq&R|dXd1?QiS1sI_4U5^-Di>?CPv(M%MzJR
z;~qUhm8$l6@!o1cNMrl}@~vBbc(qFpAIEd=)l&qwLXJ9r9aOfmta*F-vAAK4@ofd7
zXu4CpFEN)eZl<Wx=aru@pWWE)3;FcSi;C9;>7=aO!U{a|y9lRcydI_}J{FAK6AAZm
z5YkOZd+X@n$F^<bGDt)*ZrE{?G+eCCTeN?CrX?1O-{)U+8I{6N5y>8;*q)6+_obrZ
zxw!t`<{B6Ov#qNG@9GgX@s9x(i*J&=T|!f12Ccz4es134UYg@yW-6ashqH)_6H@iP
zU_G>xv_s2l;uc*xm<aD8a}jCF&BC2~_yrWne}sWBla4jFa0Nsp6UR*LWxwey0`gVd
z+vlLYH!(4cQ_x$`djH6{mUL$ub=l^)ww(qqcsr)FcO#s4XEamzKKV|NNqPbm<du#}
z%`!8H-Dhq&S$S@;mG4<YbPx5n&gDpsn#Wm7g>NW*U`}4v?vuIOpWB1R{90FhJEJa-
zhhgJGp}Ss82OKjnw9qVQyR7`)WS;ld!u}|^ynN?mq8S3v*9(hL*NgIf5=HQiUEzs<
zl|WboiLr*gFc>AME|(z(_UayMBszi7i8B1nhp(7-8o1lcukm%G5TX(f2_}$7d?B{7
zvYN`WvVWZFA<yzMpT>(TbxBdZH`A#Da8U2kjnbRDSJORMDb{0^3Dsn;-8OgIdOn;_
zuk7l2&jbh0|Dis$Nv6BI3kiBBK7nVa@`$|3m6^5;;ZCQk<ifn~Tq|0LPnxa%6xpk-
zw`uCk-40IjcUy9A(}(bc6c0QE>nCV`5pLUGUfnUxw04&`-NY*KjzX1)_X;B~RBKTe
ztFxudJdI!|V}y5acIPph#+}9QMD{voMsebZ@mW}nLc0kqsy_FqP}U@r5NZ(h>Z%p_
z(eG4!^p2|0uc+P-KgS97e(D{u!=I;Qff^M@CnHTl$=aomsh@QZ$96^oc8K5PA7O&s
zDA@6VCl;IkLj~;>zQ3nE33E?tzWJ>${;rv^Ai6g={B|!8Pj`i5h&yiXnrv}AE5K8I
zmhi;hwu*tS1=qx+o*#nWFo%kneRn#^v`<5a^<Ka!-$fzS@H3om<vy(fr>++$mc{UK
zTo)It=zDCw<z3Cbv%Z(7sx|GjwzO_}I<In|$kSH}TjVLLp_-})*x8X2Wa(@H;q-EJ
zL7u{*pomF&xq!g-5Ez{W#M;(L9B|Ow4xqEO6bBdxsBx>g$U<yvm3-VFIzH;UU>|!h
z&=Mdifi31Gf&_4cz(90fjt)*x5ifDTPh1h?^^a;U0Nqa$n7uf_P)(Cg*4Yh0$Ir>n
z$;~0}W$VEUkie!BbF;J((Uw#A9RgVr2iU-1E+Sl9o}QkZo_w6nZq{5pKp>Edo0p51
zmjh|R0rhr*fxI}JpbS4Ce#4N1K*4UdE-+hXC%PY)APZ-Am^c7{Y^VEUe2y+^YJb8z
zL4Rif$p@Dg$c2lClbg%Yk?U6vC`{f13G(}Z{-p<07kOCCr44~PySsrQ@*WT;7{jj+
zmf%1AUEJLqeuiTS=7Kmt9Fe9_WUoAb=~7WyP4iEW9~4;II=cMyLX!O#OPH<IKgjya
z*nYJ94CmK@Al?7O{fqS<vHvtiTB)gt$T@@Ee+*AqP8{%~e-TS(u&t%Y&r2{bFSmuI
zkQE1nSBQ^;U(m{e1IW)O#9<*QY-J&AX(4FE5BUX3*$D~*Ie{TRppf94wn!X)D<M8^
zK??y6K^`kh4t^dWghQAg4Cde!6ymcIwtxr<S_=LGq2XqWR3*sa*HQg|vP43GEV%_O
zL6$rmAR%rG4t^mEK@JO`pfCqFzXcz+6>>nlV7{MFmS7PDXE#R>ayo4tLDmp17bokV
z4L=AMk=9fe2k>%o|D#6J0R*!`I*0>QZJpe`{?VXo>j=?-fqt;bBg79B<`o74h4_Sp
z1$cS?VWbangCdpq1CxiFlkYd&AJZa&3<gOo=!Z^`06#sDv53gJK|nBPH(h6E2XVlU
zA<_Nt{3&lbvEL>|$rg&V@cyCr-!-oTas93ITM9VX{%oS7`zc!y5cs!Fpdb&3<<E|g
ze!n$=Z9q=e5M+M;E~r21ZU4n&frWX2Ag~oL2d^bikb_@X0LWnh0`hTytUwR}9&Qkb
z-|Dx(|3rs6Tfsa*ZV+i}WTePwkOKM{4IS%GD%t)z7Ec?<51zPr1vt0`Ie7SVxrIge
z1VxaF;uaF&<_2*69x&IBto|cnF|L2&MC_-*FUtVZ@3%VS@`7Bgxc*$Oe&_55jsJ(Q
z-{<20um>dce>?eG{Qj4&|I+oh82DSp|5ew2>H1p?{4L}Es_TD^F6@6i@Iai9cR`-W
z$C**0(-+7`Axw)$igGA7Ki;oeixQDFcU+W=kk3z&&Og4W0c~(SWaVv`vYPzuElgZ|
zMkXGN_D3ivIBd#t(z;$Vdvj*t1rX|vuZz&3w0Qw{13rGoPsPibBSHE6;LuTQ+=v9g
zc-reAPDa2uD{D+HolG%Kgj6v0U{=k^flP4601-}#3V>(+WS~%>hBI)>$!Tn0C=(%J
zS}QaX^ND}+&}{<|*y`uCa6ZU)a5{gwq9)D3G31`D4CEv~kpLd)CC-QTKAL<u=s^<`
z?umOdV_{W2>-)-ba5-&g3L2?_|NVZ_%Fv15!AY)tM#hEL(ANHPPWHD|kJF{+kwdcs
z<EknpqCtyWpIDk(?Xho27t-CX-nzhL_ne$RW?42+D4+xcpwQ8wU}2%$x`q7wTDoB(
zmrg5~9?D@@2?8A?!Y3zK((N@;%BJHfW97O@LXGv_KWEkxs*t-Xvm9j8YmzMbc0J0I
zd3E|RCtTb8kz~#&$0dXC*R*@CipzE73f(Hp=}siG?V2Yc^(UR-S9`B%cQgH~X2|?r
zeQD9#N>|!ar2Jg|K-Bqbb8>M7hY{4L?)%Yk?X}XfyZHS6$-)}v>BqY0SB6l=8>syW
z@;Cy&)r7QuOM3achwKebxi-|%?x6FX{mwU`vjClMafuALLkAM>wU&b`yWh4kH{PzU
zwFlYv4<6a?Jd9mq<61ZITi6h}c|dD?p4Jac*O{-j_w{$WkXu`+nTPpO9u1<QprfOp
zqDo1j5D@&h{A=lYD&_j+%INMlGlrvzcM#D>o0AT@saZD1IUGh~k7rL@<W%pM_r>fO
z)K^LXcN^8p9GgWIO$@3FRb1@_s!o=MbHXQ?+vdW*jpPc%^cY-BbbCfl<`dV;zu+Q$
zvd~ZX_4J)Uk$m?w6OLPc&e~^v!Qhso1*I5g(kI630CkSX@d)tuy&ZX4M~8~K3YC$D
z%EOs-7Hb2TD#MD`NNnQDhc&+gT;pS!ruUMCFFobED?kUl#mq0vybfN2ne0h7MT&|{
zJRJ3293%hB++PK{5sRuBD=K_<C<t9cD7rdA$7UH)O^pSLA8#J8Uw&8~Rkeno#hQs~
zTvMLtEYDaX=gPVe(-!?14zR#~i(K4qOjlAU?3v@&93fu$%Y3SS<8!=i64GB4^;2Sh
z>I+%QFn4GRt>d3Hn(L@uIUp+d;Otb3$(39N^!(bwKzKXyq(mDrKby|NtTIVu3S4()
z0P+K$J!HP9eBh(qSJWyTY2WWXyH`GWWPnT%e2VvrU$&jinv1I5?%j63GA3KATRGsX
zQ#bSH*7wYm+_iti9+NyOup>H2JO4x=%-_+w)zvnc0-uCa)SBTm+oN6xh<7?)bxv_G
z>cA4aw$b<`>*+o|e(TaST2KF|!)9do*&yqQou@C@_23vA=YLklUv=`oQ2I@Ne<<;v
zN`i|0SUtVV(_07ngsJIrYB`hIbM3y@OQu~5)~+Wjr}-vN+bq>yoE@*VEl@Mrr#Pki
z80UYhOAugNm)zgy^m}JZKw!tLXp#wNf6r)liR0tI_&RbsGjnAR0w1kfiAmp@i0?n?
z8*%wWUi2KmfbwTTprZat!q3vu&1N${IDeYVuevn7Sv5DiwA0-6-sPIUSUTqUQ+fM~
zl2(4c<Nj(|R>@=vty&EV{wHok=phs$&Ihj-Gj8o29B0RdZz892NOe$7#o)eUl7;)J
z-vKG{VpGezTmP<SbK*17Niw7K9X|%QmJ!Bnwt}m!c?8bsUEiFpnz4g2fP`K(^X9e`
z|LC#2MvmT1^!w?p=aE)%aJ0RrI63YV=pjvgTSwt7y>xb$m*>d;PT3W)UaC0dE2!>^
z8^XY5={qK=Zk+R*0(kGv4*&i12dFI>8-f`Xj~kOljhMAG^zs_HBcEOjeuG;{N|;nL
zzc(*AYr!2&WJ}&l)bZ(5M$&In-F(=<=F<87c1~3IW|;p1jj$IZ;3WNQa{WIf(JxE(
zKPA}TX_Jbni#GS8%Fbxx*lZYPEQTkoG|nS#AMoBmeQo9Tf)a2<llZF5ntv}A3zeMs
z2?+I%O#_+He<t<M4w0odcl>bdqw|y6_13R+5?{U5C}T@em6j@gJ&$O;R7EA5MhhhT
zbJzJ-0RImBZ?%mZ{78-!lg%E57YNm$rc!OHw9JF2i;U<Y>B_+QbcO?-J9d5wr-~r~
z(RV7}{Jzxx%(Y+ChP*_6KBKpV3aGpJvucZh3C7Q&X?W35Ux&?H`go?_;(qWSDEV8X
z{$nQpiC84cxZ%%;cE?4vRjkv{-IJmIu0PbHjD;nTnCoi2*3!#Z{FQv3me3%*+w1>F
zs;i9aEP&ofY0Xu>i&}al%y{;}yajegu~#U@FC+MME&r!`g<quh_uDkV91LW%k7`m?
z3%=yo5smA;m~?CQHSa`D(oD&IfBeZyGy#dcONJE{C>j2L;Q6QY%GSCFzm@?t5qyyW
znnHU{2bel23h&t;{<??%(>nUgO8q~!eYSfH{#Mgc8z^9{AoswC+opC%>?%GAo*+=^
z`#tAHA(#AmL#K_HD*{4XX==J)RLy)0&!RkI7$kIs!H0m3>#}e8XMv=I&0+02HDmiU
zhKJU;c~!!mBXw*-G&G(1H#VNW)(#Uu-+#}12eS=%=OHfU_eo+QGeHm$H-+uF*VsLD
z%ZL0KTWQd1?>V}Fs+Iq+a%7mppW)aO5GHGO86r2&6LWTaVIa7x9=RchhexU2QrOUJ
sp3UIJ)ad!`?(na9`kNZKRc<hIAT+s3$>CAR`)Cwpd3Cu88S|k32Lf%CzW@LL

diff --git a/docs/assets/images/logo-square.svg b/docs/assets/images/logo-square.svg
index 2c1b9321e..efc68a642 100644
--- a/docs/assets/images/logo-square.svg
+++ b/docs/assets/images/logo-square.svg
@@ -1,16 +1,16 @@
 <?xml version="1.0" encoding="UTF-8" standalone="no"?>
 <svg
-   width="100"
-   height="100"
-   viewBox="0 0 100 100"
+   width="69.82"
+   height="94.089996"
+   viewBox="0 0 69.82 94.089996"
    fill="none"
    version="1.1"
    id="svg39"
-   sodipodi:docname="logo-square.svg"
+   sodipodi:docname="Outlines_white dottxt.ai.svg"
    inkscape:export-filename="../projects/outlines/docs/assets/images/logo-dark-mode.png"
    inkscape:export-xdpi="137.93103"
    inkscape:export-ydpi="137.93103"
-   inkscape:version="1.1.2 (0a00cf5339, 2022-02-04)"
+   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
    xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
    xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
    xmlns="http://www.w3.org/2000/svg"
@@ -24,104 +24,98 @@
      inkscape:pageopacity="0.0"
      inkscape:pagecheckerboard="0"
      inkscape:deskcolor="#d1d1d1"
-     inkscape:zoom="3.3793722"
-     inkscape:cx="30.479034"
-     inkscape:cy="41.279857"
-     inkscape:window-width="1424"
-     inkscape:window-height="875"
-     inkscape:window-x="8"
-     inkscape:window-y="40"
+     inkscape:zoom="0.27368421"
+     inkscape:cx="171.73077"
+     inkscape:cy="45.673077"
+     inkscape:window-width="1416"
+     inkscape:window-height="820"
+     inkscape:window-x="30"
+     inkscape:window-y="38"
      inkscape:window-maximized="0"
-     inkscape:current-layer="g39"
-     inkscape:pageshadow="2"
-     showgrid="false" />
+     inkscape:current-layer="g39" />
   <g
      clip-path="url(#clip0_56_2)"
      id="g39">
-    <g
-       id="g863"
-       transform="translate(15.09,2.95)">
-      <path
-         d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
-         fill="#e35a26"
-         id="path12" />
-      <path
-         d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
-         fill="#e35a26"
-         id="path13" />
-      <path
-         d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path14" />
-      <path
-         d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path15" />
-      <path
-         d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path16" />
-      <path
-         d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path17" />
-      <path
-         d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path18" />
-      <path
-         d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
-         fill="#e35a26"
-         id="path19" />
-      <path
-         d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
-         fill="#e35a26"
-         id="path20" />
-      <path
-         d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
-         fill="#e35a26"
-         id="path21" />
-      <path
-         d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
-         fill="#e35a26"
-         id="path22" />
-      <path
-         d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-         fill="#e35a26"
-         id="path23" />
-      <path
-         d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-         fill="#e35a26"
-         id="path24" />
-      <path
-         d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-         fill="#e35a26"
-         id="path25" />
-      <path
-         d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-         fill="#e35a26"
-         id="path26" />
-      <path
-         d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path27" />
-      <path
-         d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path28" />
-      <path
-         d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path29" />
-      <path
-         d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
-         fill="#e35a26"
-         id="path30" />
-      <path
-         d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
-         fill="#e35a26"
-         id="path31" />
-    </g>
+    <path
+       d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
+       fill="#e35a26"
+       id="path12" />
+    <path
+       d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
+       fill="#e35a26"
+       id="path13" />
+    <path
+       d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path14" />
+    <path
+       d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path15" />
+    <path
+       d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path16" />
+    <path
+       d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path17" />
+    <path
+       d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path18" />
+    <path
+       d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
+       fill="#e35a26"
+       id="path19" />
+    <path
+       d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
+       fill="#e35a26"
+       id="path20" />
+    <path
+       d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
+       fill="#e35a26"
+       id="path21" />
+    <path
+       d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
+       fill="#e35a26"
+       id="path22" />
+    <path
+       d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path23" />
+    <path
+       d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path24" />
+    <path
+       d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path25" />
+    <path
+       d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path26" />
+    <path
+       d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path27" />
+    <path
+       d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path28" />
+    <path
+       d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path29" />
+    <path
+       d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path30" />
+    <path
+       d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
+       fill="#e35a26"
+       id="path31" />
   </g>
   <defs
      id="defs39">
diff --git a/docs/assets/images/readme-dark.png b/docs/assets/images/readme-dark.png
deleted file mode 100644
index d4526256d462dde768cbcbde886afe07bea2769a..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 38912
zcmdpdV|Qg;&~4aJ$4)x7ZT9KdIk7vo?T&359d~Towr$&d&vVCp?~k}^jLosX>{?Z|
zs^*+4OhHZ@5grd73=9lWQbGg>1_luiy7!0u20B0VP|t%daCQ<Jj$mL&eg7TcL$)PG
zpqIE#qUugcwkA%l`VPimuCA_(<~Ej&hWd8KjJ6J@8JGNcU|_^xk|IBq-7?R!T-`8L
zUWbHC4;R<(rT&1y5fh7wKw*j~r{_|a8(dX-+Z$t68)zf^wmn;|uKFm>uMebj;RO~@
z-jf#`Q8rCwZV{yfl9Pj@1K{u6v-y!*91k<OQanvN@3Na<X(+z^zu}8nSc=H}zr!pI
z2Qb_He@CwF|0fT*ai}ohrw@+T+K`|Tfc~esxw*Qiai@y7AW_yj!0+e3Jnj1=-zuPz
z(!bE|D10u$2I}*bI@2Vkqk8kX{T>eP=HQzH4rXR<Mp%-54yG$YZOv-isinEe<(2tI
zc2|)q-Q?xW+0^y*EDdos!B(AHjJhz=oRsrayN3jbV!;+>%6-`u{eY?(YisJ@SaxPJ
zVn8VE)?bGkbw)f#M@PE;zjG)279_Y>SgI`M+S-|cWxKW7>JWn|%iZ0%GtI9-<KW5F
zsmgijQ)^#$zUUo%CT+>|8oTBeZRvC@EINHPuQ>~-H$iV~Og!n125S^u_c7A>REHQA
zCGhnzY;_L*_&oKpeMb&AI-S7>p5#V4z1^d%J-wL&jUdx98O0!@iyHUT^I)ZlkP{Hn
zSEG}KjlnfE_>&Qa9A=(J>8b*quQof1H*Bsdsinfp$}ax65d#%5_;Mk5w`ZI?poZt1
zy?u#MmQSAEnoLYhtB8bryecP55>PNwlY)#Xeukm+UMu6_;QT<R99Q4sBd0yFPvE#G
z4OG}@(W-Sgl+`8=R8Pvz{xncpTUv*eKF7m0PmynkQL4gaBOZVB@v!&Nwckb-g_TWt
zHK94+8sKTePSN}Ocg?9)3R{Yg<o*JPz0u}c&7Tx+rvJc+JYqfWB*Q#MH4rRB{9+_&
zKa<nu?TTe;>I3Z5HOi11fj0<O-rvkGU&aMi-g6CXS;LNeW2KNa0vnOF9;yLIaZdc8
z@#JhEnD_zS5HW<U>qb)R`}KLGNuPX&A&N^Lc#M>aSDg*z^7iq$<)ot*pOBCs==1(8
zU?-uR>`7ZJn?V^>VGDAzVCKs(J>HO^Z1xGGh-LtFc+2}kK1Ut($}@C&5u$dz+01QG
zeN78$dTji3al*k;S4_0bBrR;U#lH|6NvziThV5z^?hW2EHUY1<y1OTXl+1u>u2bNO
zr-kWR?`BrHzHS#eC!o;KKdnRolO%V#w9ce0Ba6p}k)JHGDmm%+pQ{@a)9C0kTCU;I
zq!7QI@p-I}tLF=I$8n{9dDfSCcvlAr7Z)fT!X1!}oWW0=cn1>X1#w*_d5*NLI5U{k
zRGg`Q3nWWJ5=@vlif`W{5%3<<sg|qnJ-u}e4-Y?|r(0I8q-12!?d|UF2g0HE8^#Dw
zBEw7Rhf0F}{_nx=e9oF!zpFQS*@rn7P521#qnXMQ5je{KZ=to4=G@YPbgf`Lzsw<U
z{=gGjgYQUHa8Xdqdp9~FKmKX|Tb4g$+uF$RZB6TVzyiJ9GG_3-%k&rdTes^WHllh{
zpQDn!>fr5ge`3Zt*3Am$FS5ky*x#RZMbd{=)D%@BeFrl%jC1a<3H+BITJ;7j5l#7P
zltY?MPa)B;Ar<|FdpD8=-7fE2am2ymwSG8+Ba8An5Vsk&ohR9;sctD{Wolrk9CF)q
z!T!z)^>CTE3XL%9Fx`6J!I5(VnEhxltQ3FSLgo>WypxBXBF1g_4_OXYnXk6b&d#dA
zAmZHzfZU&khKBdY>LX`k8{+rxrv_H1v&+NHmrohzajrCFErSXS8(QO=Rh42EeqYk^
z@?UE=jWwlVW(oXuZbC!0doUq)uddM9&Yn_hd`}m2kx~*TG|KK9Nk_q7XNedY%Qag=
z@0f-4Q5-BKrm)tAG9ui{jh7EMbHS|iCPBt$W4fbj881^H`@m(P%i2br6XAW1iE(wr
zSA4(3^e+R0RfaAI9JeB3@>LQJ;bu!<jM9JeAri86a`#y|eiRR))!#u%8vn@FebH#+
z<Q(ucbR$RnZjq-goE$Mc3gPhi{!%O=Dk@M=cB)>MG6SF3^Eps$3vTp<f)NX>>LNL0
z4#fwvo=U0j0?U_2g>hX1A1^D5jtULMX|vgAO=3(rD`lJFcbx2TlM>+INE#yaEPuaZ
zfii|fL8mOaIQNsNn<6NpvxuO{GP$J#cTA;Tf1hlcENUFcb!YL|9onhU36Y637y2D<
z(I~dSv6xUt=!B?aI&xE6Rf-Cr27Y7O4DTkiP^;F2;*6*+O+WGLF?GV<s!QN#kq?jJ
zRHXC*m-I>u>*@HloN;#aH)vp;l{X%uc(9XS=;+ZoSK18SxV6|bhOr}PyC_n|cgS7y
z#IPwkEWMd;iD@%SFK~er^hIw0d@TY*`O0OHv=#z(HmaCda?CDk)J;I2_0f~_vO_q8
zQ)E-@tNx`<ch*6pc*6xdVLy+k1|O9qO@o+sPUM?ZA2WY^MZ&4>aG`Vxtz5B^*Za-T
zMVr&fZ8OOK&}8D_NFl442pWCCJUd~0&G&GzH#OoN3UkLEP&N;$`+%c4ptYkol2E|I
z{QyU*kI(%cmP91v-yk_1Ff$<5l$Zxv0vrIwmXzp)APmz^T!&H)phc)Ty<^?VO4yhW
z#$`wRE8b-gh3kBCK`DN(ykKSP$bN)D@0&DR;aISAcJb1OEcCvz%*n~Y5=J1uSa{o&
zpDFiIsEvjK2p#U1({r&j#U;Wv-<@aVl(R4;DW?9jMz@l*XWA!+obmL0>>D4|nPp8K
z{@b)M#x7(Ox;TWC!U+4@@F;RM)7Df|7xz<%4-Ye!tLfrMhn$CzvhUCyq~|iKu%D8l
zZDr2WY&h<6-{f{S3B86|^4*@t62^9>LYu>$64$_9HFS>olBtKA>(3juHmk;vgJ8tY
zl72ZRd2Xs*BD$tlQ`;5n<_2~uvf`Rex?d8UD8D2crD)^LM>}GRvpvdQPKt9n*9>a-
zX#-&^a88)~CTx2Tzy7e@ach+V*7T<>;|n(I6lK@Ue7bxW!!hXyv$^dJ^j}<_j&gi;
zFY!1!pE;&aRu#j{ivD3V14!k;^#yM=7*PBWMt+DcWv|?vv}4LzNB7+1eOcS4w>@xw
zSxB3$8<q{s+?2>C#(!*mWC@RS3Y?sr5NN&inaRj3*>Q-xm3qb(J`wSK6k}g()v&h8
z9&N_$(0jkPKAR;-R=@qpc{hkU?*;qU)Ka5?;X-g-b#Li~Z-b3Y)l{SY%L>9*I35La
z{r$r_VNFBfv`b0cMwMfL6q9kLKwT#@>!lP0Akui^)X_(2BOl4~#QnET;9Xepk01>v
zTE2`|YyEQ*$Jz%v)a!>YAKy91f*#a&$9@T`8ZIjpq$iPtB*_YGy-!_yO<&LAD+s`y
zr8dS$T}%Td5&crt*A=v*ae*|xZ2G5aLh0`$E5e}T6lI|!cFXtcL(<;fzD`+%K!Jr;
z;TyhQFyfSd(>5#{1SeaddbbJ0Ke%z)5_oEuha4z&0tnt`_Umum7ceQv@8UIds9a#L
z0`>|DgaE0QsS~G7mqiWJ!l8kIlVI0cllZD1OK|TAsM?VAC6#E1KcyhTx9k2WQ(b-T
zuYJf`$NbzQ1}Y`d^;P1T!bM1PX!L#FJ_snJPkh?iik+^)fNa0S6&xii3k1e@$TVeA
zI_p5c^bddP2guTTeQBLQ0%9>y%lLg)R(596k@VgbcAY)$Z3AP+;==zb;m_@*<e>Mw
zCl!{D*~O$#f4n)>@^hcwE6AMYN#FW|UT+?H%Te!Uv|81q?u)zXU~DL==<DhJS?S_U
z;nC47%P;GM<*U2I!QXJ!%g#Zbjo3mDzm2XuxfrL^Z9ecX=^s`|!qi%)*Hn7EOru&G
za*?X;$Kd6#Qteq1;?q;Z_}II3R?ZpbUiCk5)`uPG>h^XN6<66|FP%k@R0Ed;-b!@K
z1W!aT5i6=l!kXlh*Dp}ez<Lm$P;FyCg&6m6aLFl?Xt>k0IneLd8jn>X!oj)2aUTaI
zR___AvtVk!S>#Mx_)GAKEECVeBf^#5ear2K9I&B&{G&VM+W0|@^Gqt@^v_LFiU!iJ
zTbilkezwo?aQp_(;1*@Wwv&7OXr@SNM4&$3qxykZTydX)PcZ$?TM6I5&Z@=WH5m2r
z*d^v)W3}Hy)G+l%=O6zqU6G{+OKe8%=NT5x8V!_9#zL*;$wS5lD$YjQ+8J)*hBky+
zm(IoM7P&~*^6d4us8IH`X0RYdh)ES$a_~^8CLNVp(?j=KO3|9KTc)RCxR5e7oFs$-
zqaV|GEY@7y9u{bEhJ%qcn~qZgQ|S|jYjodu*|v|g^uO}csen-Z+{BBh7%@Ze|KN53
z0Yd3WK{mq-LBOjW!>7rR@=m+aU(=@W)Fy!_y;huGcY0s1YF#^hA$rbt6I?{wze%GL
zZvz?76CEB>Nr2%^ewDJ;(rm;(Pj}Du{|(7(zI3%PWrya3EDi>yoqb#GTN64lTl?_x
zGgFYMqVi;b%8}G^&X>i{bNJ_Bz%M_!ATXYE8NP+@Km@AIIVdO57g2&DoO?KuQN-*b
zPr(J*0LkD8f1pcC4dUNqzMa|Cp=by9`_L~kKc6XM$Km4Qjz;3L^@bo3yq1=g_0fL2
ze1`SuK=7gnkv9MmlYHOp^Z&AuQKq0)HY89u3^+%lc2bFQo9C~7MO|it3HG6flSYO4
zS|ss)z1MhrSklwuw^?syXlZGQ|4ljwM?ooCrod~1>sTesKT=ciw2~k98xU@qQ1!za
zBHP@;^ZV%#rhLV@anxxlUb5XSH%a0*svt?7zKl9TnuDAOY$h9(9prpVHA^G&-${uM
z`TjU!5v2omRww|OaGN>zG>lhER5$`5u+7RPeyZV)z0h?oH78}+^`LDr@)~i^?EHlH
zd+L_D&ND%7`F?X+R|s^!<JU>Ya_+dsMoY9f+l~iS?X6C4Qv?z@Na4F)TXg-gj1c);
zSTl4a>t4J@=M4pI!DS_k_0P)SdJ&Sz7vXp({R4a{8#%d2QILOV5MW_Zf=pMIL+#x*
zCGey@H`y4y6#Jf)oaFA!KO?N0@zTYU^TWO5r{m7-d=Bu(%?2UI3RxvuxW{;W|9eoH
z(L@V6&A~rVZu#O?z2k%lM9G&0I_Q@Cs-YN2zo6iQFo?(V*hkj_cXUQxHK+7>{yUCM
zq+e@v_*_LlIr&~M=$Lg4a2N&5{5s{QhDsqTy&I#-gBr1bI`gvznpWl8q5Hc?$N+ay
z{k?uSC0YTY=*|bqGhNc|2}Ytz5kv#$pRxTV+4aj8!$`?N_PA=llk>_E^gcu&@c`eR
z2ngCEIU4Gt3a$P%RI=UZ=$O|H-G+cHKZy!|pF?uutruh$55ywmw)=gb%3`Rhf|XFg
zV^aHN-IR8h^(F{G$=<-g@CPl=0+LlQvk8P5u#{v(5NRf&%v3G~n0W4205}NlE-5F4
zx{B-;GLBmXZFYKtoGHWV*-3ZXoN$H6jn1Y8k#2IX;HMu?G9>`4e>)~)H<4Oht63i^
zGE?fFjMkNMTGD59uAgtaN#jU4fPQq)B@NH|<kgzJTmWWYQ&daAu5t_gDH`)#9wTPU
zHV2Gwl%9kQ)I=FcbaQGiG~n=4L)hp^pApOTabn@FtQUX8Z;09JRCcTdLsC7rouFYs
z@7M84l(@n+5^D?$6}prn*kcsTNHr@Pkr3f#L>EjbQDYup2<8)lwkHK-Ha09YiHNe7
z*ZWhQ=@g-$%lR@QK9>y!$SOcs5?)>R9p69_(wJns($X)?($b?b^SNmE7L<Ts#~nz=
z{QR}242(7RX`!Hzbu(;o`A0mkm=Pn*MRiMdOx1|bdr)0NN-sh(U#T!2_Mo`cI5ibT
z#t;}neScxl#zR@NhyP#Q{juOK$4$cHoEI7!FCM!tdvT)}Bw<qO-;{#~>fDS3;C8wT
zkF`&ui7hSCpJO(9|IG!D=BE!%;m|%osLALM3nu!1L0sV1cK(bC9VK-*p3ZBmo8&!H
zA`uY+8+GYxzw2)Kd#b_7<`f3)$;QZF2F!J&4@~;jk&yJTJ(vQOjy>rW6=1+h$-;88
zrcq~-<v1tHbx{a}kV{zEl5lD@7&nS-Vs9W|h|D8sq9=m!T9^;UQ8u10V9H1LjC@sU
z2=3=P<OL#ckF5Mjbhvnbxr<RaDmO-rOJ)5n!Sq(Bs874~aaz!P-;Wjt<_#=wzm)}l
z(!uH*{>+O`(%L`<SNj4PAaxo@5}2CKSQ*Eg5%(UEE>b{@oEjMt%_cVbhDS^KkBmgI
zw@XKmwWuKMO^GW^a|~RTh~|iz<R0~Upr*d(5A8=Ss7y|RA@aHwRZ*FO?9M{Gn-tO(
z30~XY4?JWbr9&rz-`L&P9}&v<LKyo}A;Abjk?in%`h34^yK*ryGNOhNNbpz15L!Mt
zMVP|$r}_^d2^DsiC@9KiehNzlhya>?Mlq0rd%0)Cfpc~D4(gt{5x2TD3^rcOlkv7X
z2KxWt02K!=2hS^wTSE_Iq<GF3htEzv4lr~n-{JS}h@_5Fe;0|1pH;hDtOkjYy9f%9
zc8={>&sppQUQ2A|P{tS2Z+8J%^OH$@pqSo9-FAbdEgmrl4fFq`8GE*v-xAuNjmOmn
z6Z5-yRMSII%E+R-O1OcO_0|7Wj*g(mjvGs1>YB0Q-1-i+hF28j7k{B5ABp6bh=Zs8
zbMG151pQtj)=H4LtY6l~OA%$Zk`(0MGheHgRo*u%ra2uPTwH2a>&>YCOC7hs6NcnV
zF1;U=<X5BU`<kxjam7Ve8qQePN=jM|^b#EvzQe{x$I|0vkqd>)l3`_MCF#-d)WG~>
zw4mQfT+m^$Fux)TxH*y4-fo{=&b|LtSq=Y4u5P}I31Sf6{k=l;N7Ev^;YT2fl-Jw7
z&|oH}4^U3AyU9I<ONeE)f6Wo0#lTD?E5((7I*#{kM!n3FN2z8O4pRk+%ex%#`)4kg
zAo&YS#!xD%TdY+wIMxXvqtb;)-n3ZBvgHf6xp4Bl9dT;0r^27j@H7dG`+z$wRC0Js
zM9IAQIE5M?miVMu@zO-?yj3``XkNiqZ@xA4*2?e2(o}^}6moODE>9O$5gHvE3}kDC
zk+iERaV`Z<jVgNDK_~{d5TBLZb+wH)nEo7qqrnrh{no!}j;8?BpMTg2r1VPXbo@M3
z(9=A?Qke5{Q8E9>3G0#Igirnn`0C=Imvx1bE-DH$g*SA&kD4p!DzcvwA6_ah3Ug*{
z9&l3jRzyT9cF-=EJHaqFHy8K?0AwQ*&g-u@$9A=zO0pnHM4PvK@?>aC3xIPTDa-b1
z@Q9+=#~1XegXfl=K6HJ)p4Ean?K?7mt@V5$rEA2%*$mL{uPwBx9S+)ulxA3L=HsB$
zQZVuxLwGm}GJ6u@8-k2?Fe3hAm1qQ(fL;zW8cAUDPE>DR)bCY&x7tPGLMJRNEJh*1
zGY(3ilt81-?EqlsDMMT|XzFDnjhZ5{*?Dy^RlUBbe9>`hX>=HNp$9___$F7N&6#a1
z^a-$+gbUA)U8eMVI9o&?UjHQ$Q&}&65=aV_UWAT{I+;omF#Pq*=0iu_y}LHjnZ-y*
z(LT5~ik*(|VT69h`cXp3r{{Ak*u|hHX>uq>uZn=Wv%{^!xaoG}&Fg$3^Ry{|3dl9b
z!-@!!BjB^E9O?HY^6oWUX$y-|;HDaq@jbXkoYv!N!JXca>#B5VJbyqtQRvSy{A+-O
zVKJ_!!FymPLSDaHe({zZuD&qECo#C_r0tL4fAlNd?OH-(A)yodVqCS7SY5)4I0@=G
zZuCn=lYTV~avf|PntZV}bernrs5b4du}g*ToWYFZ4VxOQT;<qi@N{@xqv*l+iwMTn
z;n(B8F`v|d#|Yg-bNa>&p3(&bt^Ez!s2aFULe!LyNxOgO<CtA|0@ddC7$=8^ahUac
z1IH5+F*@4c5E<s_R$jtcs7}Jwkte$TDzljmP1-oh#m_~1$7~jDClq^7WqQ}T|C^9;
zQz{qot%!O`;Z<+8%hSy195STzouSMd67|GfDCZ3JJ~-|KK6*QIcy#j#2?%`n4{D#e
z#V5w!4jL{EY_-}s;~IVoyj?-2xIg}VumK(9G%w3xV%??BxVO)CuKoI?=JwWs{IdVi
zgoLjI+}$sC6t8`JGES8Vc{{f0G$cj8YEF8}?n>VJw6k}$k%($aZmMTxY^Q}-`Q%N0
zzeacL{45{rGMrzQcE4-DK8ba5!yE9wAHYaStp0t8eIK?lSQI10BVO8?$D{Ui9k3v7
zP#*?DR#rN<LNtPGBvq3?cO1Oi3hBj6ObB(AD!*wOpO4MI7*wAc%3716g-OS0vluqF
zS23dU@aTL`&!qgEveKuDH}(7ZF3NOycQoS7CN@m<q3?us4W^(OAsY}Bwsj@Xm&&0H
zIh9T2%Cl_+*Y$;oFXt4<U2h%-!Z`T}W8Aj=U_MnyNRi*a{UfUIi0?y_(KJUSMf0!8
z(%CJunR*M4m9F~M(ThqrQ)4)k5)l!Rbn7Nr^7`38?i3WlEV5sd>v)@7V=rsnU$Z+3
zvvh=!ILrlFOkUA84-*~nv_v4>9Jxq1h$<>FqUQR_&i%nOpusy|O9RW_!LAB1?LSK!
zs5<|-{FF?SyV4b?PPI%1+u6}~I$JFiLxtOR0l9co&&2`@9Yn>T;%{__Vy$A&rUW}~
zfJH;8&RS#OvgM_5{uys;B|2_KU?ogbhoyn?mjV(7^s|8m3kWlLG5?!UjS|GrCXf-W
zDgvOxh*z5J7b&eyU9HW6c5mHoKOs9fYM5Z*(Yeoxr<@Jkrvl&O{zd{el|!=u%2+4i
zL7T!+EOEjx&_1cP3Ym&){BvTX5IeuQYn`n)2$0Kc&8O1Xit7dp(jbJp5yvX4=!?3`
zzlo~#WdPWjddsIt%EDb0_bdK9%+a-zg1|!UHkS*xm?1I;M@LZ$3yX;LhGVafWh7!e
zj1Ut$SegeyS?j(UBb}np@+GEBS!;3c>SfX?0m3s-jsnXyA62xrZNw-z{kk1`s1%U2
zrv#M4^N505ne_azQ4!V1o_7~bzRmB5<ji7thCpjt3z$+fgu(*@l<ELi$~oz=5Wr<;
zD=X4m$#mMObe?06yRlm{Vt#FuY$%UgTyOq0Zy1rCc0#OcIoy|5QMqyBS0AU`B~#%f
zw`jL?t^rh?a@|?R*LeiT>g2NIW)R3GSgjr`)WJXCHZW>KR)|Ry*`Ottn(q>9Q<M}|
zHxrAeWWJ5VYWx*vaxR6<N~hEnt3|`n9s6OkEl%SuFSWU-EZ1dQ!+nZVUK}!-MS?Y%
z5T9a!i^kxDHBUil`aNC_fXncGY9E#6sVK?R+>@Vhrqlbac>A~3sI$EA$YL24g?K>W
z4AFB&@Uy~CCkYkf=10jwhmP^s*w|t!xlA)x-mLsW6MIX+4=jr|B%F8S0zeyxA5(X@
zsCS0Bm43EWT2mwnv><fZS40%u9np<QWVJn+z5hE(WHu>6K_wpk`7r@D;Q5GAzN+9c
zSSzGWCyxBx_%;xSUvOk>YhR7WwBg=<vV^&46K|;SSMQ9|^@~>^H(+?ZB|^_v?`ImF
zdKH&Y7xA%uWEf?x`T6kjNEyeOdW(nbfS2k&Xcd*~SMroh%t4jCd^;*QB#A7mQWDS7
zbuY*7Tv5>gDTP4)3A}QqF1Hib&JQ2PEibN+_V?T@n<o5(k-)VJ{hzP9w3cb(lukP0
zu+VKPJ49;zc54+i?>1Gm{=O^w20f#|Z;#b%pc>7?xpvg`5HphgKiWhXHczbaB$YHi
z;f&sCi#Bi{(osZtg%C@kxinf|6kxi~Mz$ZEjwGi+{H?jYJn0%4JfswsWEp<lKO+#A
zY(F-Q>>!Fwe6$N;<U2?AB#pVjT4K;{9*s~$dV&Wni*?EO@gFo_G*C(?AwxNmu;QkZ
z^U(Q=>I1j|1xkvz0sY7PiZDra^(l-xv>>j}A-Jpz8&aVbk&|ETA@I5S_fq<z%#fL)
zD7B^T^GEh5g&(i3=S4$Pk|#w*Z+|dxAPT@vMQLrOHFl`oP)e!JE8>c$Drc3AKApty
z>qQi10c*9TW@$$I1xo4>jNW|2**FM=wA1^x+%M}K+PfkCLosaR2yEG^*3Rf_nmj{e
zCqplD-CL5~@SAFf^Annhj}+5Ei-vo9jw%!y003Z4lWD(s^o;B@g6Mk3yFZXgzeOSP
zzJ<<1f=Nqj<Mv!O+x$94&rbdPk<nxY*=@zDNJw$N(`z8;(e}rt4HGR$%3s0WYTEsy
zkU#s499cF+f~H0@AdV*@nSL*BzE~l!x1=WJ_UN#JIy`)&VAr9gdFRTt*61WJO0I51
zL_z`<C<QOY4&37DuKSueP<CCISh$o_C2VKKYEm-=5gTOk$e3{?7UF+Op7%Q?X=wPS
zPEVHOawBvUZ$`W=J#)0$V>9c8Mb1ooQ2g^6kkFo;M~vsh7c<iYC}h_of8+Z1b>*Vv
zB=Rfc$X0>omXB?4bhYtUYwjJCu~jn&^ZGEe$`RC_6p7ay+cBeg{_JtjUtBIIo=kc1
zX-2VI;5-y5s^ywJaSCVDX{DZ-nL$>%)Y0PMpw1tyNzL`%TST+ON=@u*Y+gxWT;}Vl
zsq&lpUoF5cN85(rNe>K>lk%ffW1{+qoLZD1w1l|@-}iK`CjJ&ds;!mU_IUlQ*DNZL
zy{$$8p0l{t;L3!};|LB?MQgtAcx=q)gIYr(VPPpELz*(XDl`9522f4B?Idzy19P~V
zA^tVB5xMXDdmqB8r1=MYs;E$CQF66QVGG=7^CCvF=8*R^gJ)rx5fOvB!XC-ubr@M@
z0du;D2P`qt$fUbb>3QD@>hMxQqD-?FBocR@%r+|`?(h$UOHDg+h}GDWk%*YUAXveY
zOyB>WpSivZnXUPSX2{h3XXdY1_5D!JGzX%nVt|W?80-G|&!lqflCp52+@0IgOjQGu
z{H>l`3plO}7xyfA@A$8ETi;gO=hLBhJyL*j@FRk3l!R2c7=^Q&kocPxW2ixI?6q4d
zYhmM4%K*X%tPmlbq;pgp?-V~9a8X^sc@*M?+rO!#Ma&S1_6QVuIb{{>@`z5%H1uA-
z^SB=*AeB<!@uwRW!O=3bO8ErmNSnb*1W2Et{+qbHn@|u%^Qff4(aez8h3Dd<^AK1d
zp`i7ywtE~3k1DeIFMLD=c}yzH4s9?B-&M)D%1GK?2DYDBb6M5PQGDjC(rN~%(c*X1
z(W!|>oIz4Xan5aJsa7wVn9C$cct&8EkCwnE+925}E?yD3r_tURnO7WmAI?af#dcd?
zT?%%m#-m5Q-gOt&TRuMjPU<@`(jpux`cnnUqz*Ts%%Ewlc|U(<Tpl@j>E6S;e_(L?
z0lJGa9OeD}?CSk!!j(4OR%&c?8X5kV6Z3*Y7!Wr22thQLFtTOdnIMELP?)&;6*1#@
z?jO~sFxbKGA6*kB0|q=8Bj~WtiWtycLVxrU8ccp1Vp}2~U5zzYG1J|BFnIDWeb)CJ
z5W1B8)hXX#_5JPob{xfr%%chAM!HI_hntk@<t@wEWLBg@tpne#wS+@rQAx(dqvIrc
zEg>sV5%fC;BU8XI_5Gr1m7GL->E(2@khHdu<*7K#EhR>DelvQQjy$D<8ECS=IkeAi
z|I!0E_$J<c_7ML3{M<%^jboXUE2Wa5U|D_bCJ>9J`fk1#G?njmo`tiSN$~UV(-1$q
z_~g1zlW*#sDSgOf;QBHXw)Pf36>?FqwrqV=udv!xcc6+h<6`n>z_&EbbTy#Vq=f~(
zCO!bQK{rH`m1{!1LuhIw1EdI;Ar>ji*R=YD=S@<&H}MHcv%+FD96K(iSA`l2yqc{`
zco;ZFIVf%E;uG?=w8`uA$QWAr8<2DsERM+6CLe0v6Am<W$@AhU(kV9ecsy;D6zA7o
z^9lzOMOCi|d6)HVGosJ&e&HxIHVzDdX&UI*%)Op<mP{K9TJk)8kzR6QmlX#CROxt0
z3eQnUQB2iM9U1QI5*=^&qNGIQkL9w#pus}XMz2{n{vySsW?=ZG2&J1tBn!jTz{*}|
znZj_9prnc*YW?+bH8~ZCvd&BSdXFH7%yBVsXUXK7#nh#)T02+(Yi*Vj`&A0~kt{Sz
z3m=b=l~UNdF;A#1B<SoHn#~DRJ;CI4Y+mPd_ghw0(h;N4U~>#yUj(a*XYk1`4sM=!
zn-gJzHh~=B$fEv9+TuJ&tzl{MI3HvX`3Jd#Bzumyx#DBlFh<45qU&kXlf|U&#a6i5
zapH1CMcz!Ui@`cYe5Z%Y;Kg2Np?S2)RNSHO*x(XJZGJnL`(=RksRL_>Xzd^G#8xIj
zj}R4x!^_Kqj9^0sv?2*FbmcF#@~A1vc(Ig}56LtZ!@=sF0GKJE<0LDZ9bhqg19%0Z
z&Qc?hyYiTk3gy&5pku~F<lj>3(ughryF<zF6<tS;c*D@GmK<RXB;pjA49*JCMI?sN
z&4%m9xweWdFZXftDWABm?WcxKvyi?ct$MKy*Q+gqd?z@$CPLo&7;iPJ=FyV5Uuhrz
z>V|^#HaI4asGi3eL=@)1zw>{vzxxP%HM_!df$&726a-9DC#2>dIwpL1b@jq#Y4Fxc
z=%#Y-20d{BU_Pv$0+f&M_O=&Z-=qLc!~hVLOXEe8KCpz89#iG#F+|Q(2y=>QGND|C
z*<ABbWlZ16touQRKT+_LW3_Qi3W#WZ?_7x*tI%t*>2kH{W<swLVsa2bgW%RH7nF~9
zl7RR@5Bu508@#M@{gjRb%iUgU7K{4Ch|++KA_QNs9}I&Po|mMT(JmYb^w%M^z=(sf
z_Aok>c=>5P1cD+E3MNx<IBqQp-=$SMD>kb1y2fab^_Fibhoy`$e<c5Ay70K3&IpN+
zfb~GxdJ**hD&#6ZtSRd2U>YK#1R-NSJ#1eQl-BN$m%i3yWuxomh5ow}m|QW_)6-||
zx>eLqQj5NMd%tD_;an`}K8zLgYF%A~3|*D*XVll(tsg$y4HVbBb60-L@}Fpv;B`@I
zxNC{f7)PuxGMdU~`>J^rgpqgyY#8assW<9OW7evmU7w*)h|JO6Zr@8_bXXchIV8KQ
zq{-2`gcNp+l`(>93>k^NZ_vIW9miCjVUo`6-NdI5-D`QjJe-GzfdWp^pKxTgTJ@IS
ztv$pmjD*A`+v6QI+xv!NRGV?!#Bsp8vbqFTmXEy@1;P)k^zYbe(+p&5!FpVp4$koV
zgZ$HE|8cJG%u$Mj=!#Zqn#MfOqQ}Ry(hmz8`#-|M8(QAkswCv#1%q9Ft9x!x=*WZQ
zs@@t0I=Mxkt5^}+mBGv8!pFQ-n~hSb;6?N(t_VG@!WEZHa8r{wBst|5&xuYfs=Z6X
z|Hkt97ZW|nI&`b5SFF&%roF2>(#7Pt4f4IC%2zCmS5{lu+PB_0AyS$?mp+5&epro{
zf4KjYayuB4IK7~*&w$=({ADUl&^S@Xyllj(!we$vLPwn`GVUH54O}P;+)iR>Ah1TT
zPSbcblPfXVIzIQZ>YtbO@T?mH2~u9;Hp?5p1ZC%v31c4cB9Ly9?fyK&Dgo5n5+@}k
zB_A7S4w1n0KGXqH`9W_?5ylIuF`rj4w=Ujyb$4GDSu#*r=*Gszm}^U5YI^RtTsppp
zoc}Oo=Rl3Y`P69dg$vtD-q7<}($d{WJ9H0Cps-Y!B&M^N9UlrTYGeexp5+L#(9KY=
zY+z83!<EqGc0h~}Gz}aSI~J2paAJIp4-ZCAu{j=W^;)@`7w@NLdcpywxt_c?pXnai
z6#@75QoW_`$jzhM=f8JbhX@FeUW}|@6M<MzjGzQoB9g^W<>C1Bt{#ts=MJI;QqlCA
zu2xEn*1s-p?^OD4NRA7H`<HK>*fogk6{bdn2}K1FYAEdhf*mLbT{jc*1WP_a3R&%_
zeQ681@2uQ`C7DKdATF&58}J>B*b=VU5h;qD21TW7d03C@M~jR?sn$-+#lzBjSGmwF
z-rJqb`KXPsjEQzbG!kwOG_-MS?A?@*Z}XC;1R~ebg>@cHTz&qT=ACd=i?|luZVprV
z_$UL`5Qpk*%}}(@c_3UVR-PuFfFSS`FB{UAm?8tzJOC}*HV@_g)4pva&<B9kB0y57
zw+~fhP-buNk>97tGOiA#h1pTJ-#ZJgJ-u#f*11R*3k#=d&Lzo$sjB3jpnz7G#wzgF
z$CHV@7$~hXi}c{7v%l5`q#29vgQi3iyL(TY`x{)k(82=8x1gX<=uwMkMmO=xhgzb=
zj-QhK;-d4_l@W2{PSN9y=KJ(PC#=HG*&1^H6>@#Fx|POnZZPn}$CsIF*`nb$3!IdN
zx-<kJw$pMbT30Z~rBXxXNW`P1gIG=bYw|sx;KE?yUZzntT;|-({Lts>M7OKJ))XJj
z{?|i`<^r^uirbIjD#PJlrFZ>1o>7F^Fpo>jf&h-Yt8e&}Zg?Xb&WLf>{gH9+{~C$w
zK^?fDKTDiA@cR`c(UNk`tw^C#u^buASNT=><1{X&!#)xxBNIl7UvXJqTl4vGr=U1r
zCCSku_v8`ss?K4aT#9i)$7!eBiOtW?^wGjUj9_MPYij^AlgP*mLrF0w_1X6#&Rw_b
zHO>o-n5wj7;2Gbm(Snt{*}3t15Z`o%T<#$)6@%aO^FAkpO*oPg!wUr-1!EQh4DaQS
zS*rcgW{ISQHU1L#f-1(nbbVA}|As<Ll_$$2n~Z_u_Og`Q;D*RlZgjst{RTDjcy~6O
z8UNhqFiEO4Dli+!7?QUq<12q{UAmZ1xVa5z59*fYLDt=~>PFa+c$W2ex$9JsYW~~f
zRgZ0_^xtrWvS|XvunCe=@Y3GKcyfq=Fr<VNZuy|XZza}2M0n&ZH_3Yh#hTl?3eOJt
zuj<N3xL>_|u5$H0_7l|-;;&`E-g9`$1Xj*gouQtjB})~+ZRwqHCx%Rgp+vj%&de#(
zBo18`U}qC=3T55HH=Vo6uV}&9@D8$R;s*&)sieq%$*#B{Vu^zP*T$C~FE27E*V(4B
z0Cay<vrQ2pvb7xmt+n2^zj6v3R9O{Viy9gAZ3)lE0w6`Jsjwa3>5!UY5axYsD571o
zqmBlNUbMmM|8<J_z@77=)O-5+OuJMhZ%B&5l3J?>d3qkHzRwx4X5nLF7nf@`NP)3G
zU04Sgh9-XLz6|1vYrU3>I=-P>J{=I4R_)IiAX+=>2Gl)9``<oG<PN|~){PD`hrHU!
z>Hi7(Cygt_9T_nNUp&u{q;VH9O!GZf3Fo(24$3=tn4}6CI0VpH<q^$h8q$gI_g?ne
zUbDRsIi*UENnL(f!x44cItSGm7$)u(xGZZ)aTcApSnT@Rs0nU2jk6a4#$^X%GTkJi
zSZA6tdS3ME-qr*qI;Z7CMnZZ~9yNptI0(sD-HBUg=Zv@B3AqFp6#m%^M%qEzV<G13
z?sVBkZ=$Rnc+!)@*_zswpQ!lp>m<vnZ!f`W<!qo}mSNl^UDv%sm4Y1nM;Eg+eDWf%
zF!RmlOTPR=N-=Um4kU;54mG?N`-$fa4nr>zHM7Matj6BJSr3t@wvix0?7B~&yOd(H
z)gbkY(+dJy@R;aF#0eF76GTO=lejD_$*9&VI10$TC1rT@FiN$EqmMuE%4vH(=Nc5Z
zp<Ek!;V2mGzA<6-$+_xtsx0Vzw7R>!KJr)oI3%Z)`RfP*q<I$<y>?)=?Zj})AU613
zo1bn0gq?yp*Rf^()xOBcHK<A1+8dp@7fCm-Bu&kaL1nIYE565qq=9B{pkBG){(L3j
zaPve=(cW(-N(f8h*}VAhxw`a%JG=ay?K-m<*QHI7{&7!P$bkWk-=^s;Wv@opkT|H;
z{ya_JYVGK6i40YL*)}p1U%R%W50n6}I!|#-0L9s;8NMMUt8x@XXoDrirPO35MqaH&
zONO`!!FIqsX?b#PirrP`$N6Sm%KJOS?d>fM1L9(cvPzO10WVljYkXm0VPbO9jzimv
z<Cv<?hc~pw3XcR_m5)~USBF-pXJO9I@F3VcyIY$eHN6SrLveQr|9q#d&JQmkSMZD`
z8^f{23#;{eDtyilNmUml&Wxx(MR6^T*lB#O_QYH%{=EQ~tCg`}S>IaEDP6!h7Hge1
z|GNiLRae~9%0LJw`v%ZIwkN}1%qUfdqBpNrzc_`DF<cH;fH*N!0W(JEIr%@RU#Jct
z%}w%qYhsAU>8#Y~m$Jz!d#QXjmCa9ML<J?RAPvP-=QUb;lLyj0zz!#QvDAp<!lWd-
zRw|phANSRZ1x4U#MBo4x`jUWs_PW~K4JgOb69{68Q#T?R!H+bb^dViXI<6Q4+rDo~
z>Cg(=yQf9|Mq|H1!yXF1eIAYxe5HniHQ73P*LGGYN#f19)yZDfL#|7}LEnCq4tGBo
z``t2K@C?SZnGn%e#AYCj`=qP*r!s=<GIL3E%7|&~vkpJwAeK`qbSCu~zz+m8|1!N@
zSP+bFSk(#1t;lZ3ZX!)hEr4n=L_<T9d-CX58*6IHup}Uc)}TgBw+dQn+>`0Vf?cN;
z`f7jdXiyzyU|YP<`PekKbq!5;Vt>>W$ERe~2b57b?RVjIaF$)V?v1TM06!n+<b@XM
ziqz-+BAnd6QlK-1nH(OyoKHO5&52>BAR|(-Di4^Tw&&>fKlki3^YcPa3TN~q(NuZ7
zv(EY3ouX8Ng#ixqQG*Wyn;u(21y%<6u6)O%mUIV{*m9_ty1-~C%F6Po;pK`GRQ@fo
zdX2m78W>)`e5hAoopa(#N?xRW_OlcET2Nk8Kxk05C{}pT5$3Dol2&;6bVrT-y?f25
zYd#n}qh_SpiKyz#D<4;>SxmNGdp>A7_};dSCU>z0BdTcEqAfua79LrkHbwYeqfgfL
z^~!dXS?*Yf)=QIbp{z}V7AA2V8(%8||8WFTBL0O5taY=-HBmvaP{R@8&Rptb$W^1l
z-}*S*kB-PUBeKk&S8C|f`*$bN+tZIPse|o5M|y{BEBjv!m5gT?x4&RFNn_G8`*s+P
z7m-24;ZM5FI8^n$N>aq<CP<;UC<2CpUH~4IM^A;Pqj53jKn9;y+B7mT=36dI{1lHL
zKWA82Sh4bX?_ou`2lGAbLPDZuCt;Fqm%bAVvQCbCjGSQ67Ai*EK?!__JN?-&V^lah
zRVFUiuZqym4S{wJij2g@1s6igI<1PmPGeQ*MybJX>;An_m|Y;Df2W^tu7)QWIOz{b
z9^kxUN6+>!tTsTca`o=zT~&*(mND6ODf{NR#5?Aeo6Ix%xGMD`YO`%2r3-rF%Hqb@
z7c9R|`CK&!#EP<7{zUcrQ{cF^{UAs-7K0LtQh56?B{zAB%X(QUoqxreCX!0jvZZqI
zSOQUoWRFT~JLQqXu+}(NsiOs|w+2Qw+Tzwr$u{Ip0ajMut^ZQM*~eurUSY@y6msdb
zIZ<*_2G7m{u?iUa&u9uA^P!8IqDf&PcFAO$Z@KQ)<nPp5?Ksoc7UMfy+0xd&i1{qJ
z;!sgDXEwt7c7L7XVR8JKuo$$c*>-4A?{vTnf`(S7&g(8k1^#l<Vc`XuoxM3X<hV~g
zr-pnEUsqEtO<nsS2}&^K>y*QZb`O{a-JD~)XU=Z|Vvtlpzk57gHW05_7z>rk0_>D#
zz9}#vS1MveF+}3CO$3=1T39xFXtOL#MQN@d^C1-V4%SUH#VLaY@M*fr?!HWAKI%1N
zZ45P5urGNkQ5Kgg*;5~j8tH!OrJ?4ah?$K~@-&V^!<52|HT)6r*n~qnB(7Ppc!>!E
z&!tRo%<Uf=S<>T92NkFm_3wwT91W2`y698%kbKuVb89LW<<WGZH{y<^{$u)p(KE;Z
zBnz7{y$_gb$_XR$9k+R-JCGA$Nw)1f;<Kb}_Us7cj3fh?xH*$@mKI9s;cGeSZCApm
zS0Js+!Lk8wrZ4~EjZp<S@wERQ8Mv~JZN~awWRc{mNJeOc0#Dag-Y5py8y~D${8AU)
ze&Qsj4G!u^`#~XNjNS)w{6<j7R{9OEGr3*giwOX}Yp7w*7sL7m$9gM8{myi6>w`xc
zo9Xv#Aq({Y-HZL@8HVjg>ia~__QoBe$D*s-qnCJ0Y6obEBHON9>$%?Zz=mvH3qxD0
z_FUS#wj~E_$@rV&BRyTw>)Z=~)Fo6rtSRFUj?`;)U<_1<AHsO7T!<7*QOr+Mnd>v7
zAUqZJ-%klFsST~n1^0`Xvkjubz|wU##`Jp?WBRS#!+RT-U9x}-UV_~Tk)YDZbE5FI
zVlt@vQ=<yB)1|T|-*^3WeV17T?u+UG2vK;!^j6{T*@8fuaxb$%yB3F2Z5HM97bJIe
z&jB2`92wniAwmi4DbfIEAxm;~P78!<7Vyn%7LX=NPJb4=*CzTmE4&hqgZHoiY-JXY
z_NQH-eM7#)qr^`d6tyj$bUL6ErA(^AO<P%67M%~xCeXS10~YRD-({LCAdD^@$ppP$
z^v}a`qY(dLs<Ih2x=j7eYd}6LE&H<)_QF<-)KyA8^$1v!hb@;@Rb!$-14^wah#!F%
z!3uBZ-`l%N;b<U{otVmKz`cn(o#2$Nko_5rAHL$EeJbuzyF^k*KRPz4>&(As$zFhF
zQb%1<_Ittw87Tt*;-%lwqJ|ZM8d*%jrJOH2xzk}o049ww<?nyK#xIjawVpiqe%M)(
zhm{qY4N3CR9kyN*-;Dj9xn_o<6c*hUL!*R7P))8iE{?H*!MNpeLlK7ZXC=q$=^sXc
zLKnSQ+C{*`PxJjoy|({Xdlp`pOr2E%l2<w@rE_#>2m3oUTVK)dLWwXDs?;8Q;$s?q
zd3A}B$EQRR#b1Fhk8}u+hgd<!cWTKFAJ-B#KQKqyx1{a00_aRJp(G+XXTHe!a5V(S
zDStkZ)0rdy2Qp^Mg7)uBrJnjC1_nOCx)ha~5F`sQ#<N-#G6lR-%V%5kU(H%Rdx!>3
z(leg?5G+zZ>2+>>AzmS@{jzz<sdG=`NiiBr;k$&qJI6~VI+X)~$&&|%u-OICa$HI0
zA0qa4OLGCnJE%j8p(sjNcG&)D^P3!{qDmV|MbuK)Ujo|@PkvG`A__Q!VX)%Y;wj>c
z=<}J?8o9r`rfn2m!Pnu-L2)~qmE>!w=WVG4J40_s5j5@l&E6U`mHx5B$;6Ocrth|!
zlSQ0GJl8R}XOERWU0Zwv24{?Gi_4Mimm1=0d9K+r%V)=%XrLAh%xAS)|5wC!MAZB)
z9)6Op99kowWSW>;Xz!`DO)tSu)CRgCarMwBCSUp;y_CUi1PQ58C!ekJ!_F}SQc%2;
z>SpK&KSuWn#9stT@hLtrIC85lGCr}Xs!J%IGF1Rj?lBI@bn7C*ok9Afsv&P0X$vU!
zdhm^NI8>yhV!;te@U_QR90BIUdX>WR4OEQY8dO@HH4V3qHV7L+OlV;_{cg0hR0kTM
zmQG>^-Cq?1$v;$+32&G3=-AZN7@?*55-_E8W~TsqA7_@A`4v_*;z1d?J>G5e(F9If
z67m6g7(<LB6(LpprVg5MMS>O?Y<}Y0CB@8IqGroMiI<6)O2sUiJ5eqSU2!|@Y?}=K
zfR6?`G;1&n(J+|qELlvdbd3+og>PE%q2=RxHEX~MlE^oO+Afq{LB~UxA<%9y4Tng2
zOLhow>$nl4S!Uk#ABZo&aPyH3=`D3uYsrr@<pzMynn^k?5o+^@Ns$i$LNs`s_h4pe
zdU@36^wo$)%siw7%^bdG#UL>a%}92l)#fx61^EDWtdYr^Ig#n00a||`3=<?z5S|8f
zVI)3Aa9Q`O4D!*veJ=dZ3^Vkxre;XXMo+?vsF%{$<=;P<g?W5*BFIEP$k3l8qgZ1;
z&xazyr`nxm=BIshtc;FeLhY!_*H2CQ&GWn(>r*#JI2$h>?17#L1ID~zER^fI1SI>_
zam75r7lPY!WT5S7n-+%Tba&&hEkEa4{-kS>Ksd$m3d!jZBdsANNbnd2dg+B!LRrc#
z@_~K$LFbPpvfrp1QLz?m@BKxI1K&0V)$)<359o19*x{ix2>nhNi$BjW<2dZ61F*aT
zZ&eZ8IkHfQKDvS>D^7LFU$es!6BA2rk+>#WnBAErMKY^7lw;yq1Kl*r{5JW7UM@L)
zi4aa^$<uE+=EiEsng=Wpu1w<%ii^5`NXX7<8s*D^nAmO;)#RKtuD~^>NYXCyXM4hB
zGN=&Qyb@(Y-%g@BNfd1T-MxW<d!do&xe9a(wdJ&nqk#H+jb2L+>y&xT4_Ez^c?FcN
z%9X>`x4A+J%_%w6MPoOKP}aD#3)3fXp}%s-#uF}FT2HUcrT!pCR+sTV&|25#;*YN}
z4brcp<-_Q=t=J^fn5_G|A5L>(z|DJyGC%)V9f0x9PobmDW!Rjl^RdlFW`ui98k%yM
zGld)8i?eOj5$zI`a1_o-egA963jA-ucH-Z4BdFAWeTk-G!-(R99^9qUK@$fDE>OtC
zp;c2qD!@rZrZcvgDm;Wa0{8Vx7SM*@RmM(&f`U{=7R`g(RzpzQWA^8ZIRCzotmpcc
z9m{OpoIGh|tfn7Mjxvn0-6(n_xKm&zJUo-H8dZTWLt6bG8cwxR`ve=HEnw+`{nn>8
zq?3tZY6430Gdr}cg!c-SW7o_!cHDQ#DQK8wyVW$~I=WIrZfv#_bmKQ|5T!3KxD<-k
z@m%}T9kw!%{5}&|-*KlusJr;;Sdm01irN!#tkdhAb7KBb@_Z@7O!=lw1*i=G599W+
zV8{q$Xq(%4J&~UO>occ(scAG*cJNhWI}D%T$jC`jMI80gxwSCIf5rp*Vqs}sOdQhP
zRC*y{1`Ih0yi72<Ai5xAB(eEG5?E$papAP@@B_@b70l?-{Nz7kN{FF>5?CkkL#EM*
zLW@VXGCH(?F|T)>Zg1GtMbqfTz{FYlK-fi*av8FhIZh+>mCmX0?M)j4vKVW)&Un@!
zqNSCvh<O}42Zjbxe82+$YGIQtVZ<o-_rHG7Hfob{Qj+Z_Uo+M;pr061#w@@NG6$A}
znQgDnT`QV<73_~zan*XqYlqU-1q#(A@1|u;*V|cY$($I`BLAK`I`tTZLM^h+lCOGg
z!GT9nVWNkK*U{Tim7Om8<G9^H!wU5Gv!m{l#Sxl$<_xHo0dSMA)BVznJsLX;`nSK|
z^Fjq+ZW|)t9bHqJBtP{tzTZco4%ueBx&f)3Z>PVUOiT&Es7EI=-@66~CiGn%E?dT=
z2bS6TKyCbPK)nIr7sliMtG&CkOxbo{(15m+7|z=rB)zWEqh?QM<i>RP*w_4E9ofDK
zM>j5Tas5LK%C`6mrTX*JiAC;5DU*r<^&of$`FE&qU;=+Y0_$5?H*zni@2P?Sq^g)u
zFK$_KSy^=Xl$T|SZFi9T>!-i}5x0QO=UWHP=ZDRAY*`!e({db_BX<5iz9xx-uNA(l
zXX>&JWFgbWVlIh1ioi|L9te15AG3=8)dIMn4(Kv2qYP|TlIkjwJb7F-bUx~$h|(`J
z5@7rc|Dzm7NDNo-)M;%p`8HHw!$~d-A>uBWBkXcGKEASXe*W}+I;I!scw$K4e%MD%
z==+94vG(?Uk|9&{3qy#nhWq&;A3*c6?KZUsmHT~!FPsska6FxBry!Ky%V&l5>+E(K
zp04Sa;n})lDsx&!*{=P(vIbckkkR_WvVoJL<t^TK6ZoQIUO4zb26(n`r&$T2J$cb3
z^_TUk&P}}}sDRyWxI!3Vm{~N&E?dGpIIk)lpr7_#rO*<Sf1{&tBCp?%a!pZ=t8MI0
zNCYa$oFWv;eQdN?j*cZ;mu`;>f?I7K?KbJiL0G}JA-SnHRK)9MmG_ILzid?%7S2|5
zsdz_sC5qOje?`W-OO6c$B@uqY@ki(6dV#|MZa4O%)!zb1v)A<c+pDS)7|ac3{Tk3Z
z=l=&nLA|~`eDTs{(q9quLvsCWVd@&nMG__gmjUJzHgN6uVYG8G`P`57)-kA68aY0A
zRKc88K`dAk$VEddxn)!--8&v$GGuzLnVCyMn6oknx3~74b-D$TRxqj>zUtc3`n+Zy
zkLSpHJst-vn*M@nR&EC(j`v)SJt9{iKkz8}yXi9*q#y6uv}GRozbzS?J~`#1mbOYZ
z)o#=h6nRYC+HNC0QUjJ9i|e&Et}^Yx4;B=kcHb2d@wZ<R#!Isc5CVqwbl^?*Q0~-u
zW=5_wn}oP=WXS31dEQ^cV!Ga4oW1`804wfgJEh5~f#XQB^>U?7E*hm$@-;4cOB1~@
z6CxC&^`Vqhprc}BX!jQu_DYpfBnn3TKZg+L5}Zgfq(*vFD3$pIoKj^a9Vjex7Dtv#
zxJ`PIQp~S@92S#!L6_)&xx@Pf*ni$!!OW#4T9;^xDI`F_Y8G^EZ(+ud#f%sbz=#2X
z%vn>WuHWOnvp6rcCL8lFDI@`=j+83!<P$%M1ybW9M1!`7)tn)9euxj{;N-<(7AAfO
zFtefR_+}$P3P_RgD}tq7oyD(QERU!M_M|1Ep9#h#gmRPgSJvnxbtwbdVF{_fRcjsN
zT11FOW`6C*O4S4l)&#O}Z6K%ibaC0R3Wg7G5+3Y-O;YYuShLN+{9glE{f7hH&osk>
z_15vst9t#@T0-kRPD@Mk2{b*iN&b34Ue~dgO_T8Y@t(^uLITMM`IZi!rF}l{2+M?>
z#Fq`bOfR$#)o(vmbtQ!m;<opTN3{Bdq>d|RD^kk6OP4N9YxV0L1M>1xvx3?uocVv>
z<p2BZegT(71e@B21}dD<+d)C8#+LoAzbisqdeb?<7mOGX2%x0p<W2^|gOQz`T@fCa
zuw<`PujnZbUn0G2s^FZ{16jGt%9vpnFm=k~{IYU2PNxfp!%j$WFoA&qXqraafpngI
z?j`<q{q>AGXE=&Up}4r1(z0@p5}VCNubw^V)2jy;j=q{*mE~M}hMVu^l<=_DL|tG%
zc#!Or-lCFrfwgt0h5pFc^CIlsJ6P!uW#*M{3$qU7dOriseCVo(kxq{!qrgj1nO^ka
z!lE^6cUAnSI4iAQzu~{EkgVpp3*@S#<zirko5vM}NmAnX-4W0U!DMtMS9&OUVu@HU
zVVe-l(Is&Lf(4i*s7NYNC=`-xBnNbbJ!XY%htq||mnWBsKAOrk2(DA0{|OzSb^@X6
zdV$Gg>WbaR9qer#A(;^CW!wIS`BzfHhFx|x?6PB0AR*36Vh0bAAsRwJQJIO9EDL+n
zEqI&gM3~^)60xQ5`iNV5R9y?D{{8pgw>e>wl2!3ed~D#gz&XczF0F+GRUlVOcthAK
z=ci0xcFf(de|C&m$x}b>+)f^oSjjQ>e-W!#?9C|+cT{+b)6gmlky5tlZZ^J@=cZ%>
z6NC_=N3U)rPyQ_;xGF|c8`{J1)R}!95y|O>hp5u6Q|{91c?5+nzuti7ik_y!mq@V@
zW^5M0dvCtVTkn2Ar;Z)C>2KFJI)34jpO`q|R=Os{H##OGGm8QJ`_ivZZ%TH)N2o<H
zY{-YnAG#Me{2=Sp3KfulFwaZ(4p#C@G|uef=r}0R{*U+P7nRtp3Vpj+$t_a(VO8n<
zh1sdI{GInNFC+n`j+8FopN}jSA6SgqjuYUJS$d%oE&wjjBwb&;NVa%Owg(pqn?o=h
z6Qh93k@RS-1d%^8QdU`g9qmD})+UgsTB{TpAJpB=+|`F(DQY_GO*cg2kzS)!@c7)B
z0y@^bI8w@jdGqFZ0|EjF4i1)m`}S=UYZ<w0nc8*S=v$-|>-qC_aw-*q**d(FHfzDA
z6R|k_pNS2-#0%}i<oIKe!5x@5bc9gMtzHM!mHj(aVdQ@bAsG?Z^=yZYDVm1M>(!=6
zDYr+&CFFZG8EiHSqQlLmtA+<he6g&w{Ez*vpZ8^Y&M7F<(u*s#d+MIYtXIz2B9zkW
zkRjH?_~O+8hgFbWs1hCB4v(kla3;6gjmdQAx7zFVVm6xz4Gp142P}dzmyQj=;mmKH
z^dhC4S$CUa@=lrcb7|LA+bRcU<a<{I*p$#N(U!CByfi$tuuP+S2RN9gqDv(&JT@U-
zQFz{NQTkQ7^u@(Fk&~s|;5{iM>rmxte(=eK;$NDHi5P;i>5elkK+<wyuJkfUN=6De
zH^9s}QWA9HHlk}AXquL0Hk(11AgBwJCL22>;8G!=W!oJNALv9;@avLpDMk#cY@HLk
zd@P+Ii8U5UUDwkTMUgg}t>!OM2ytSU0IFjD=kwFz$6a=VFiL+lTCS@Nk&NIYsi>c(
ze749~D1CmMXY?0f;&U4gZ2b7tPr5|Qdt2gsWo2cJDgcW;=oqy{+zg;=nzsB{T(32W
zjO-rLJ;5^f;cFs8JzmLkU*-)AkBk4i&_!}ik(v-6ZRMKbA-wi|;r{GG_57l&l+?z$
zbLHmdmWRj1f4xQb-qD;Q<MVVEJ3}=>gB6DM3m`eegV}7xY%#uHG|lKyttdv)r`zqu
zW;1R@UauEbJ#?g;*=)vDb&Qm7)`&oUOmmZ*;$}iw08!0ICM}XWJEeDjNlx1Gy4wrV
z)6;>og%FCAQWrvqnH63BdFSYer}}g;=9pdRl_**$D0A)~6VbnKyp6Z!6!qCkK`QY2
zNg?Tn|BRA(z`v7{#D6Q!V3-hGjbs#po-K0)Ato&$UeP%Ph0`RQE?@|TdmJ!ap(i;y
z%x1H6yWPpvA6UYA*9u7hrw8n2$SH%uN^q-&g)-0zk%15%aM&3wQK1^c`a4<qd+UT!
z_2}fGSC^ycU{>Oi<d;LaTcGM-HAA=qVnU$8TR+(*w0F(#nx-ZDo~kW~$ZAqQpSLW2
z?C5a<F_WhI*`J8G1^lE--m$dPe@cFRL7Pf^`m?Zlk9a;hoZIbI*{Xj-La|)Xn&$_Q
znj3fCPNbCUmn~b?ShJ3{gVocid&iJan?+!=2)aaD<If))^xe`;<yvvMj);NeOb=NF
z>XXIUDX9Q;MWnB&l*o5iNcSB>X<A6GlO7(rJd(&zGj@wGgrvztfW!D6c>Dbi`QN;S
zm=pz@&Bo4Md)Tpg4Hk=qtn3^%ZP`j<*IJv^^iMx$)8^k%HH~*(e;J3vj;5+uEfzE>
z>D1nW5JoTTf-;?`W)$FAB|T{HzLd48<U$BDpi3$BI@^`voV28+n{BtB(c2LR;Py&!
z)v(a77yec*x^=K(HVOK6v9Wc(vm2arZ6A`UQ>6~9tonIm@{b}8jWZAo5`tb563hx=
z4=xn8$lvf9<Aq3Cj>%;vSaTAELX4u)359rFr@Ju3XdeYatKa*R9_!XYRaH?|R<=DP
zq$<%Sau6J-Q)|B13E8DkS;rEmRc$YJ8iKNGB-nkdmR=1iJcinLSQ_lVWmGAvwmKSJ
z4A_8kB<vATH@f{4jADO>stA1RrHynbbgRbm5Eygv?VIh<Af*6GUG@LmTwciXLO~^H
zKFKzbgK7`=c(wyncea7HW09;Tb=LfE<HwFhNbxD+__e(|OfrznRHXcP-<R`G77#zq
zJowy(0~@Cgp5CQB_cT5F$Qb?DcU@xpk7;#Jn5SSbx~J?<QT7L*F#EvnwL20wca624
z;_^tgrnq(sL6=9bi>eX*nFT6RS2u>mcAey~iNV=Hj$7rnZJzMBt~ZCq#7_ZwRgW#s
zOxqY9({*)<q7F-FBI#zV;$Ys)67r=+!XP6KUQrZmHbaC41_tuOS2Gzmej?XheFb*A
z4U5G>MrIZnnOV$Tu!!rfz5;+1tA1t6)@@7}e>2mbeI6+#4!Z*om`o;g9p3n+fYXXC
zTp}zaHX|X(gJhNNuFw_FM|HP{#U%U#sb3ru6LWEPc6LR0Y{G3F61qN_zJFhjjtPm|
zsVXBl>G~+v?Q#lDCvC=(;x1?O31I2wa+fZ}=WtR;9v07EC^LcYs(t{NyhQAGn24`D
z6ZhmLLPMaSW5ZJweWw}e+Q^3Wo=?*HEL^y7`-BM-yg~>IQK#46S=fY&xWWTFbD&3i
zLsY6d94Laqih6Ck3A93&2;+ZXa1R&fpH|6F>l<p7izGZEp*=<l|5yb)KsVqm9jw5w
z0-hI;;dAW3Zf-6b;!`8vr9xUEl=|dXggPKT+^{&hRoImao<q1#jD59#p#gCa*s11W
zadGixT{nbgSy`FbzJ0r2m(Yn$>a6+SCXBr_ACy^~xV2V25>_b44=T;-{OnH``)Q1D
zlH<k3-O6*(VfuKxwRJ4pp&bUDcPzdvs@o#+Cy9@gQmPRViRZrgL%}#HaTPnnEcece
zu9*eu!pS#8bq26o1d7Rgw!<#e2^WXm_}2dl(;{QLd^Wgy!28i*W<R+O^2H^(d)Qq~
zB;E8D7J3T{I`D3ulMXQ!gb;YET-LYWG7bQz%Y_hvYp*gy<Dwsz(zj19R<2ruuIng@
zLRNMTJ9q6SF(H8;zBa-c%@#8jiy5H^bX^kQ7M!m%KT2nf?x7rgdTB;Ry`-YJD5dzy
zGsA{`vZUn2@b(Gs59#B0KQ7Xg8X4QAPk5*$zIxoUEfsRXZ)FopauPn-q^4ZIagVFL
zW>yxIW~c0dlR~nk5LCO%TBX3>Nt%O{lBA?0z_6UDsw%F!>Z*${w3>zJvp_L}ei)UT
zR2`D@pj$h`qURe&2!`O?m1B$vHU11Hk1OQ&WE*+KMi8~Y5Rl{QNUsE7yo65#Osmst
zbC`sg5*|^Y`c@z4(3yIB0ZLsDMx{rxJJ)bg`%>bB9UkZw1t|p*PZO!&zQok8Jkq{>
z`<uV|>Z`ps-+c1{pOIC|q_#uKtOdU$UO(nc-J*Sf6c-)u`P3m(B0fi#)6>41zw5As
zdJ@XU(3Q~=(NjT*M`-ON1c9g&$U>>hG)$&lWkuf1kyA&u+3qDL0u!IxaA4yngForg
zj=Njq{ANecF=~sr1u12j$z=NNSX}Q|P*}LH1bAC??`{jL^!6!?4BjsD%f`)LSGMAs
zvqOZUfWs<Y>fFx!(kb2Sbnj?=M3?fSyGDoG7yMRUQzNcAUa&YPb?zTA@mtfSHlSlO
z>NX0DE-$94yRGW2|50<h>PMvP*twfCh7MuIo-f$2X$wP78_ZQ>Ml)yLLWZ9|ikF}J
z7Z;p25~)ipRq>WOU^|Q?Q7@7orQ0JE(~AJ2Uq5Sg*0#$fBTnt%pu#05e)?nah^vPO
zTXv?o1EesG3$!WSRZWspq_TKp*@q=LX&;BRSNbaooAWbMcGkVdU#_#wN&d8mq@*OV
zefxG%R#qmaOqt?HO--E{85#Mh5W?sYZ4HB{)BGPDVuw>YLcdsu@UbL%HQ1HyUqn`;
zOW2=ZlldGGqVdXuxda6mBJXw|5n0Wh66WYeE2MhU0bbJ~9<Z7v&)t(xmzcUPcpdiU
z<5xu1o%hZhaQTSHP&@RAfx#UivJnq&hx+=;YPH@tY}l~%n>TO1)a7!Cyu3V-l$0dE
zh%IV-5+^46zMgX+GAQRVkaq()$8joNAy|qOH#*JU9;vhDKA!sJyq(Pvk+ZL#ZQAh4
z%Ap%yTlLb$uvJ?TV!t9-hR~;jctIeZ5%3&{_l2S?7E<mF4m0oC{QBy*H@?35+>P&V
zY}rh+zYyzp%jb`_+l|ZRYVf%cB1uuqwpOnnse7Gx3U&jmTfF$sXYmfL@bpV)Z{2iR
ztjw~R&6oYQ%el|vm1GsDj5#yJ92so744`*soAr{@f=@A<1i=oaLqu%+sDz%~ZtdNt
z+eR?jE=q-|bM@xG)rz3Tn51d-D`7|^PN$QqbMw>B%UQO31wny<%>M2N0IXIkGpE1D
zV-MfY%@gin#j0O1nG`H$v*Fo<`UZt2SgCs`lJbMXjC}_H!U`&$7&)xh{E!Y^dq#$u
zZw46NFTj-8!P=*Nn3=F(Q(Ro6Y1kEK1+#Xilizl`&^1vpa#+tj7Yzy6qU++}2G6<q
z(>cjWPE4wW%=YcuMP6Q>xbemttvBC%^Oq)*>1qI^1_%BJYaawxBc(he1Hr5q@r8*|
z5a-jP+4pPRVX3h`)}ZQ;UK_LCJKn>)kL7SyxW>drWPt#p8hKtpz%+->D-Y#zRzGL`
zomr(4w~s&!wn0!6?v8drwXS-T0#S{WDoXhpe!U0we+ingiWWkIgoTBDv1ZMh(Rq1!
z%J%KsMN(3dXoDVZ05-h0!J9H`{(DxGejuI$3XkFCvjx1Yqnwg5YyO2PU(Efgxi0K2
zuPzVU@Y?DL-G;@jQ53O8z@q~Cil&o@ECLdxl=leaQsH!N+4TCViECb2b4-eC|CzY^
zznfF@ikaTh=e_3Xb9$RB0UcWUyuP^ewj(t!z>?O!*0Dy2bIm1L2ev2*dvCiqEaItg
zQIpMP(QWa@a{6?!(LUT15*2Db3!tb>qkn>ptU{IEoo(2yrXL=-JmQnyU2J`XIL!i3
zoSFK+?b5quPcwRuNI(}=ZW<}keOe?cih{eUB+%~Uy<B?ng>>%RiN_wfpJl%oosrl6
zwt>{N1Kd3J2JRStGppCEtA1igB#I*NXwAFlRw7xfyUQ$^e71U<-9qd=V<IjcnrPe6
zy@NF|uUN&aNxH{bDXGv%h_Q<HVdfG2;%#h7c9NXoX5#4Zhab2yGBG09gcN)sgb*<?
zF@ZJDt68%-$w^LNs$AM4FE3Bz<>e_4KKS4>CX?x60Ll-*+%w>(D`CeQ^|Q9?pQQfd
z4xtbg=znam9s0%?f~<3R{e6X%^$L6Uj`#5Tv`oyyDh}_{CV)Xs`X^;^Mqi)Ru#}QK
zpC&W0A<!!tdPg_3s8$ndH)1(`PUd%P`a;<CE-e2W{4fxTcU14O1_uXEpEGAp{Qmv>
zYs6$*v`SX9d)A!%lv#8CVOI5C5*`Eopk>?JR^yWJlMwP2v!eIhKWpA&X|v{RZ>|e_
z-ODTcY<zvy%epY{R*)Y`<PeVZ9xf3;A0+Qt6>a^-S5{m^8-5Ru2kZ987u;T)O^<9o
z_NZ<ng=pd8wtjGB8U3|YRsB=6MXV;gL;NWbv8g+PJ9JG<&hS2+lI0;b!px)BM)K39
z3VL_8(W{dUt*S+?UX{DBFVjO-q003mLabe)Em%xg;^ML_U}Sa+T|WA)=4^K7Y$2+u
zTGaVumz9^3pI<;$b`DZXW_<Pqy?gZ_EHsocqeqdFdVqNg7W2op9o+NKV=P?!6I*`&
zgL6lmjizbnx{j0*i>YDnh{jl;yQmP_^SMTWb^xqalS2_O{^BsDuuLO3z{JKqPP%ol
zvM<X^QJKcTu6E2Ofs}A&?*LBkZLBrd?{O{;P&~gy#wL8KTLNCIKIcgxd6G7b5Tc$-
zd%=PQ-K|#ZBLJx7u;dcR-_Z1lO8mGkx<nY3M_=MXWKA%9ba4GKQW_c1b>|qY!=3r?
zY$jY(3hfWEvJKbZ+;Vs(3m(peC_iJ1ow8m#LyM`t1=ZJ@3Sc!s*Qkc^yT#3L*_7;t
zMHhmz#seJMrAwFRs$JaBwn_mtM)KDS@=|BbdpR|$VzAJ~U|=e+0w_7cQ|cfN0<)!L
zf~tycsk7!?xc~F{pC!*+&`A1yeU>LJw{LlU<rSM=U%l93RyGUC-!XEvS_}bwg`&*g
z^xDc-PFg3M&)xssoSI$CjFvp7*&1++>PK!tN|~FNm$&s;T<=(+>x$*?*G6_9*2DI@
z!=_C9eZO&kt90qaM_Y&pHX(!&8I+viA=sgCFi*whku3kcf+0Qa{IsbYi&^0D1lI&b
z7UZO^Nk^NpLO*oXErF4E;#H+_PposlEhs4D+N;J8-=#ApC8Y!f1=6=qFG|bGP!xr!
zPd-k1MkZH|8O>`i{+EobES{Y<g)@ephR5qg)3lmpzupXsWQNpOp?mfg*~H6rx7*CZ
zoRP1x<A9sQI4fKBxk%3R5Ep4CyHI7`+A{ja8=kGWNDE;>3MCa9xy33O`D%|Y9fDFP
z-4J=x(4Mspkdw(aM~T##A{2s;_dHq*g*Md&uR6gdFE3BHTrLqC8+#96D7&pMK=IB)
zpP-teN+Urfw9faaB8>7sJ96q9?gFg!<ygBQBb&RLoKk2X)Uaz#oEOFx89!wv$TdTS
zVys#NbZDnSSTmNIS*7(CfRO?9ujQ~9kM%lpj*m24_jRfC=;W3+Z!uY!2^*h;;YRfc
zlgTt@&YU@2uD|~J)GDha)22<MEoO52B9|>wf#0is{+TfTsxaN-O;jZPq$JU3mntJc
z1VeT2W$6M+B_bO^DpKw?>HMKPDh?b;%fiiXu8xzMep^VG`F35BcntEH%t#)xDcXR|
zZ>+rGln2hvYwZjE8CbX1{GaGBb-dkj_=WE4meK;a^wKwM(O2Kr>a9FAXA$1wG=N{W
z)oGRYXTanZyGzP7&z+Zs+cjODJ7-N907c~*B^5?d@XvlKX6$*P{NFbPoY6ag>_U|l
zTPrwsP#{Y;S1`1fgLONd=(<+#{ZXZ9@;u#j*+5~53U8zUxDM!wVKMZ*6UKMx%(SPT
zsChhP@}od?%%j4C_uW;quU%9mJ$rPkJ+G#tX<Ch#lqSg4bOMDS&`)-+4$RUkrB}%N
zWlBoD`;W~6o!eXZd2<EVjR@i8Zwj!Q1z~|E7OgMmj!|K(+E&Twy&MEN6gKQ}(Lce?
zuiGnm=B9QD(cxx1s^q=7p5Lp_c~VFoB`;hezc}7=_$yP_8tHbsh1qN#W$cl#=fnDg
z!yFJ9AgS~U3Mv}51l(T$y<?0b<<$gOAua?m>KLVb?+}rJ#(u5J@!`=RzZ}{hVto;;
z89#?hu~Smvsl9Uupq<ZkWX-?ExOPDe<2k*WWN*G`@LbG_@q7ZUMoFRkI$?YJzJN1c
z2Xi2RDK<9t5|_*MUX@j{%?rrJNSXe9F|e`fr`4D}d$uVtJMJtbcOm7KAR39SXjRUY
zy7+11t1B-#_2IL#j>Gla6!$*++riB<27lZs>gelA2Aw&ypULWI^){aFoUI@=bY1-1
z+SfYP6c+9)x#avlOU~#MaG5~h@k(?F{Sxdf*=XF$N-K2i7KOVm3+L<QrF3j(A-hQB
z)E*9A{wANx&Im?Pz+x$@cY`j=-kTc{7ys}qy>jNgrl5NAZKL$CNO2PqqR_XagMv~O
zkj=U)+HAHaJKLDdCR9yBRaJBeXPy<vf^-jQnI5jFv@=N5_niH#yGWJVjFPOB#lGiK
z6r)RbNT9;518(9X%_L`d_;FJ?H;xQpKvz54Qr)PU@!Fl$*H|m3=c}C3&G;VKu*c~w
zEq2$GK|JZ&K1rL!H!kn#r=JcILUaZUH*95{??JO7NsNMsKtaba=pSdqH=62v1XLZ8
za~lSr#D>&&Usw7p;?0U7*6Na3dgzSmBSA`8H<<@8y6@P1)`KpujxLSY#%h92VfFXx
zx+FQTezJ}tjJVJNRj*5Apkc+Vdv47MCAH<!6-DXecDqHLx{@vOH2-Ssczb!!=GRtF
z=$aYx8@gPMl;eQoAtI|8AcXjS^$V;2Jf4<TWK%Lnoxu%Y3yV4LSk#WIYKtgU)%~X^
zMDEMlU9LB_?su^-%fsfqE(%Ij1P~gipi8h?1iRDSv<o$JR^I>?tuLoroR#cC6}P9h
zv5mvw&;UAh>f{KEiN7N(rt2LJ;r)JxRKN3eaWB)|jK40N$8K%MrKbngL_X=wxF=d|
zR`e!%>IxySn9YXZlyJ^~0LGpd${GCv$d&cKuQH^@eBFIOTJ^t&$HiYA9+U7;`}PSj
zHBU^U0ywM+>-V@A*ws#9na069m9zQ=pzE-8zl)GS6D1WI9ixozzKnb?^VgN*@<^6!
zsQ4y7|6on2<C7N2leBrN-P)ZycXm+oH|_HQmOgiAMHumo5dn}|Xmqjxqrms>TqAL%
z+7&DW#77vVbO3V8>$|`KVuK;Cyw)n|)gYtTPjbsqkX<HmdTW)hs1kp^<d)Sw`A$A!
zS(mUUPp86DTNEwW1__Z3yWtqruZ&vfWmOdAc#562&+OTzO&J|-MwpX8b`~dA0ov0Y
zcHT)!O1fq0)Tyo5Ilk>P;n_`Tn`aE3e)v|&?)4rimSa#qQtG&}w<4uHxMIbMy{&$|
z<3Mr7{w3XebzA+!4N=1qVytZ4@8<kd%ye#NA*V>CeVCbVSC;aR8>1MMXeT1XM6XUZ
zb{{mvC4fgoOHGZ`0CcnEt9vd9yIcaF{aA8^=5>w#UKc~V33JB|RkbC@C=Ql<Q9`8&
z;!m|<w<(mCm9cI64yJ$lIo-P@q9_WA5Lj(Cq^{#~yNQg9prD`-o6W|mU)Q3lsxhwW
zM+iY-Q4w?JFJfOxDyI)Q6-9v$<`rSvs}m<ooLQ$Zxf{vHT7~98-lBNLJ=Y8m{dbpW
z3$uPMd89*!SX4(o001BWNkl<Z4t+8*GAf-O+yJU3sdVe~>|{gN;ljbj?fl10<y5*Q
zL3RbNDhaeJlvV0v6{_5JaTqgxELJn~J^!lvnx7PsCuy_PEkF*iZr!@F5hF%;gb-F7
zG2n=<O%Ty^NH3Oj4zK@hZ81Ujc7`}SSY!w-r?*N>wyzaoK{g|cJHNsZo%t2@v%1YD
z!v%h@j>Ryy9Kr)6A@-wJ2u`mgv!woc+6Ogk-&p3-$*YQc47JzK;MS#NZyuDnYpbNk
zg+gqoe?v%iiHb|Dy+@er5L(-AUf1<PRaHfGJ-no(B++K#B<~s%Kv>W)6wXB#13`8M
z90;K!$VJNCpsYn`Kl0x%Qd|1+>)u%2(W=LOEEIWx^j+kfFvzi^!_Hp<e0scRa-y>-
zndjSuaKzniE`9I~Tg;%_&ur;=`J$F(p;qJo{M05sr$<0!$L<|<O-p|D`+{Ku66`Eo
zUruC*iJqNoY)f&G5Mv=Z!_DH26<j(rh<@=l!h%hBygHrQn|W_R1?5_}sa>b64}xR5
zJwCL%?UErq4a322JDry<Tiv<X6q9<>$6Do2lgvS8JAxj<OpmJWHy@=s>6B>Y)i>T{
z?TQ~+y=EO-w{6Gea--`yLI{KqC_->BBa;=Yer41Z*D>zK>#$j^M#&%vs-|I56vD&9
zxckoA@p?VnanD2Cc$UDOuXCHJVReqLG$+~U|0>E#`K5c$#OFs1H9Tc!_6g|n!wT=9
zu$Y8S;b9>cy}QsufLUWef{h@%LXbltA=XM>iORNAH(g^blvV141S)J#b+dGH1)bVi
z_+fP!8}>L4YPuKz)FdFB6p|<TQ{Z;H#hY)w>6tWX(taUCFA&iERwLo2nw&DD3rv@Y
z+SUWrv=1`k0{7)t^{w{6?mXz(u2xKT41*%4AxcvVjabRLF74=GLtr}V6b?@*fIcx2
zi|+*w16`Nw%csU&4Tb>rwMG^@b@mr1h7D2FFHIz+WM7^}iOcxRn@x}yX%u4jMM}xR
zVx63_`uB(UYrs<bd|fVA&5fkZ#7R!Nr9CF?A;N>l12IUd41smt9!Yl)!$97SR9rkU
zZW&U%OhET#w19PP)9b5FwMg^#5;`^7ko`nJA`~fK-u%YOg(s754$XvTH>LePWAG;(
zqMG0BMpd0Pe{_NEP9Lh6?IDL6-x9cTeg~<e>*7z*7SSKN>U}Og);jjVD<cRAFmd7F
zAa<v_nY*TptRj^OmxM8}tDUqQ51I~9AtquY%<RqZ&~xZG?wW8LQBje~`i+}!9eve}
zv3oP6X-~SxxP@kUR!g}|1%3&O?fT@WTICCOn(C*!=vC)juyzHbMqffq`v^jV9GFZB
zss^eCQYe5REu)Z;0|Q9O$>XP=mJsAHVYQl3guw0bqH7wOt}*uqaM%=L0}JUG6+(GM
z!`sj=x|_Aqv#=yP^;v*|Qf<ZR?Ukp;M_bsi$BAy@_WLK^G2!VaCy}3Dz}j`|dFPFP
zai&LSP**#OfM2#&GHOT=sW~2IuP7lsUnM-q#7!535)o`dN{9$HCCpu0_CjRG?yC#Z
zlhfg3jO0n$IMs<Z0E-td?x`qBA^@rra(-=iKtTm;&uZA48Vn)XC)#lJ0#vwRPhM^I
zv!%-F2!@DEErguXS|w2J0<U{l@o0kNJZcJ;9WGMBz5;N1>kH8?5%pcx1r-{}c}k6I
zy{@%VO3A(gNpY2k46qo_)1Qc}^l0qJk>r#$><3n`8Qir6-2(#yFICmUY`u8ANJ>)p
z$4&R5s5^ltfmoUskY)!)A><OAd%okbab1pP8e3lbb%=l;flf^hJ`oTiO4r5Y<2{YG
zz=mX=bE~5Hv6At%zuF@^p4aj-YDiVtAL;eFmLH4j9S@4KQ*Sz$<LUeIHw8<w3sph_
z6^8V5aQ~GN+<r+IU;k3d_B1y`dOAqU@o>T5ApS^k(`(p7rc8c}>u<iDoZLJH^zVxh
zoPT<6hw}Va`O9ATx?pON+kI8_bxN{RUfQC2XU+5JdQ)Di5nxxi={y^k4D`?=B8zab
zk7&zYVyt_KRQ3=qcVTyJ=V01it{xsjzb*=qru`%Yr_#x>mkzdMx&)@sIp_eLgEEM<
zrxUImAU?*-!r#iV`sC*9k}BWo&OKQ5aZ(yVlSSE4Q>HH}y8oF^@_rrC(*aW4e8uIX
z&^3)m9)FUUm}p*o<5OlV6I^qS@x7)9*pu#I{caaMI@);Vrgog&KahxE6JAyF@uH#w
zbAK(lyeKO*zFI`WNg;WXKM8yH?iJm;cYlED(mz|0AZL~T!S1R!$drPHi8W>u^lWb=
zn*fx%jaG!!GO&Gc{aC_-C8A!LogiDoI*t_{L2{nftjO!SNML_~q}W+s@U#npaGw~>
zgT=LD%u4+^q!sBDRaUvX9Y*}8Z#<-?OAZui?8s55^!UFo6z_nwkJMhqY`*-*AAjst
zw~WaN=`EW)A(+ZN-y-Fk2yvuAs|s)hX0evZV=p|C6Kr^ORaYJTe^L_O^ym`@0e5YD
zd3j?Em`+l~gl9LU<rm9Onmaze&hLuZ(dzXhrLN-6`2!)u)|D$)W;cCA+a*6Ub?5eN
zdq=%BxA2|pLY4dyjZYRAV>1gLy*85eVP+~^I$l-s<?>R_x%}@u`p|t?EM@`&1KG0m
z58nUq6F?U2&+;OKTw0Ws^3T%r^qQ5llu`y;)H@dIuC;6R+62r|s-4aMD`)BKQZ{^F
z&b9>=9Q?^i%6umU3tg1`?51+L8}k~KA#MvF&o3e^*@Jhzm&`>@vKKfxu+T~RVi$Qo
zx+q@i!uhiY_fijHg_lm7Bopd{r)NrykF<(Hk0Qntr==Cu&ToUaFe~*fp~wX}MJgb-
zj-PnHzHa?SI>g43*foKEeS7ovyPq+0g_)u<jUFAWY)^A>&*kBq*3-evrNwMYcB1K$
z>5EHt{ql$N^up}4dA?y0Co?Ti(q>6YN>a9O-!As=->+D$R`crBs}BkxBGKINMGScA
z#4B1%#*fQezrLbbfevAYIP<kn9VmhvqYfEyp|!5$YH{h+4B^--y8f{e*q3Lx!Rrzp
zpwlg?`HQ!@*RLkXX2fp#HX+%UFDa}Pm=z;&Cag+m=9cTE6&lx!41|P;hQ}6EXdEmS
zc(wY8JPwQDhpAg8Nio9>=|=o1;BvX1>fgWrOQE5mde^R9^{>DFTIWP^aVL!rqtdeg
zK}*J;IT7+^{yXd2<{q$P_71Bnv+x(-j24{7F+~1-@Iz<6a=d48qVs;zX`M$8lDlnY
z*=VffqAhl2EDxljq9WkeUw?HDpYTF(k6{nwDrV~u#aA{LPvKs&7d(KJ@|7Qd{PFLt
zd94!#A%ywVQ+lla$KRrcR=RcGo?FDFLxSiVZ==$sGkawPH{AXVV{f>&=9uE*5@vq+
zH4)+A^zGA&J<0pn`o}iDnEl<)dAY@Rm*noN*B&i6rdxYE`l>t40p0tF!*5^jLSjZ7
zQJi6+)cHTH+*vH-#f6#sSNhH?G|2wi)mMzZZp7KcOzongC@U}LV0s2uT|S!V=-R#p
zsc8q9IPq4l8B#$+h?z>4&RcT|xq5gAJvv%hw7y)My}azyqMZG|H<$z`Es`f`qg0pv
zsm|D*K7D#;Aw(o#SkCH-NUI6@#~N)7Pw8kB84t7?`&1p$3Snn9RQQzj=^Spjs!fU^
z=&I#WkPQaL!N3kiOF~}(Dlodp1pCw|E~-?>DQjNw$fC-IBGO?o5`+A8vP+B-4C-La
zDXfmQGN(+S2qV6;Yh?XnU0$8txf=TmOf@3XZZX{Er*tr0uK{sJEzN3l74p~DJT#ll
zr=GCx?WF6iI6Ysru!sb3AjJ%x7}xCL(Ulnm57TBvBnTmHJl=C?3rw84;Xq!gJggqv
z9ea&Av$aJe)V4*`H2#!jfBYe(RQG4OlU8l3B+#yK@8#iqzq*vtN}XBDE1B@%^Yug|
zKv-BP_uqXdeR}oeU;llHO`Csb=k8?gzGM7(XAd9pTWD<Jkm|7!5s9H@wD(jYCw`<?
zWF$AEAGa^|BNUOP=)9>_c#F{QmAd?UNQZ=8b?4PBzQgior#^XOVPUap-h#z^_{nED
z>~`+HV|+ak2@v0<6Q6uCi|;pukz1_t?W$6)JS&(U9jz3WY5csU;+w`qB($Amo3X&?
zW{#DB3n&0mfDOPV{P)hi3b@?Q_H2wes;0OXIL*)Y_keXk0Pv=t{aL`2I*)GvE^gvl
zI^f1=*OrNqG4&fzcBq36aqbU*HBB4~5DvU}i2WMy5FC9SwZp+TV>>i7w5CoLUY|bH
zR&#AUp&|^+Tx6hOF-$KutdtcVLrg}87?#KCPBHC*jLc`RUSBW@VF=X5P=q1+cIVa?
zsOiND;SL?EZ=0DqXgZ|%G#Ij(4VU{N4#VPSK0VqAOiKOZB#=|;C8NZQE{z>llVO>S
z3~cy3?dJNmGreVXc8ln`mrF`Y5>uy6JuxhjF6%QOFKVT;2O^otq+3qo#ZPk&HE`pr
zD`QQHc%m&kf20(rZF_B9biapA&uMz}Uz;5Vc)ooww^=PTtKFm}AncgMM*@{ns;iI1
z^-dHDvr-qtc1f75>GHV|0|Lc;V<MQf+{NQhzs?2cj_`X-dPXMy`tOU(`0#Clf`TY0
zDCBRq-pNx>JQ|krz@u*gLjka;{RVdrxN>WXtEWrlns;Tz|L!pbMt2jfE3L|qysnmO
zIYLfQq_}lxAN%=hce=g@b^+)%{jNtJy07<)nP2ebrymg+8G+mF;fbfG;qiF5<f022
zoHH&qhR;5o#rO%gaZW#-ZgEypvORpVr1+PTLi3%6nonC1l7leB<dKpNyaBw1A;zjH
z2FCjtuUCyd)J;<Xo(5ukwrkXiP-`RS-$i{PIi!(sP2mRS1OLVl%l>HW+&|MiAvq41
z(8QSKz%lLcBZOeuv}rYubzQGbl@9jFhp6xv?&u&N3t_ke!t9V!4w+S*X|hX=M4awX
z606yeR3@c<VoP%|DbTGQ?8vT-#nh!Bvs4m)h)})U^7^rx)uPm&P;3g{SV&cq?9G>y
zxh*y4jtMb}uKTuh1f#B|Zy)lYgqmNaG?w7iC4f#C9v?Rdg~wa<8iG!k)c?}d=R=5l
z+ge;?L!{!7?qR@pO%DBwGj__Rv_EDJ{wOw_`<pY~WU)2;R&QaX=E0r29Vz9eWy_Wo
zH+{sPi>$Pi7s6r^(pLT9d^IE_DCR#eyvJE*4D~y<veL=Jlm5XgFaC$1pdfx)_6zTQ
zIGqW%-NflbPoqbV?gPSNx^ykcIglC>8z0=cy@i2^_TJN%6y4$0l?&g|Dt^4n6d2v3
zmBeHTI&Y}uS{Cwl(A$kZBjo((a1-lxIs=6eA~>pRcf~AZaBwiAM_tUl4?o86v(Dgw
zdnWSYfBwm>cic<&ZiyszO=xiLc2SXhHe)8^C*010b=zHkq<G#gw8<$lFZqy$qQ8o3
zyCd*C@GJF&-%7xrH9-NIvSge}LlFtE8pEP_EU*DrV^}7qG%@x_xuY3l4!K%sEw*pp
zUL#&MY}k-WwevZ!6GWa`J5Wj^`?_wsziNblm|(;8U7cW4>4BX&qCq0gk>IbhOhJV>
zWGjO_AB$>-(E6Qqjuc%=cI85u+xXmvIiO!`eG6<|ik*^OIkiRB!FX+>YF}Gj*Y}-}
zu5Cp;OLgK^t0Knn#7+Gh+q*nzxm|(VT7MqL0%gGQo<Un-%Rc2rk4ojC#&|p)G3K#(
zQ6@`3zt*0QJ8vgab0ekrv9+&tqEeil@>O<;^7AWiOm9>~>bm6NN&n!{hwi6yr;e;y
zx1L#F&Su63@9@Li`KA3%J!|>0l`9_#3o^ekXi)dfCMmn+mud{}7r^koffJ;nJ?KKN
zep{<Z-zvR+uW>ZwOPyD=a!(f8LV>)jR|k92n6rW@F4xhe>@a9Rx8GtSmCu6B-Z#hJ
zdEc8~eErS-|9k%5D2l?rpMM#%*~}|1{)fp=B~f1C=lc){4+~|+^y!pn(fbRtQYQKK
z>h{N9MMzdN7?_7)DFje~IgO0Hq=|D~*2wnbv0^s?&(b`1{NXSZ7}$()BY_?*eC@;H
zzWeSg(sjMcDl4G-ZS@Cex{;;5BO3}U8}`e#n2Z?0-qD8T(xX9zyJeh_k%9GF4eC-*
z=Jr3<={2&-t1&Bw?%91fIK2X=SKzQfk9LNY)7N4tffAP?IFs`lis{60pw#NKva+&e
zEi<_mWb%!@K{oaIZ7FQPeU0r54L1#wf)2-MF0D=1<2`}4!p&1R94IQ|<EF+1#U>6H
zCbO-naYsv)TN@$dpRT|5;V}4=0V7Ts*!X(rh5vhvbB3SAnWqn7@4o%K^7`An`O1r=
zrlqlH$<lAK^NMdMu`$DBB5U&XQ3Foz9q@X;cpIDdy10Hsh=>X?-*Lmp&?{6$yYWM<
zV)ts@)9AGhmjhDcWv#3tUzgX8IV-qV*Es8Se>*=UXj8J2zVSBtC)l6-$JlnKIcy^L
z;K75=qU@B1ckkW%_;)|d<-WV_z-qPf&ifzJzFibgJo*riJ@Is-RW?FHLKr>jqCO$*
zx+XSfZhti)35F%HZY!(Cgx^IN0<AIp2}%7B=Q|QAFx>COKqXCQ!~PThF`Gw<KWpTA
z8pcD4M#dgfq?ENTo78U_4iFe_?tS^NHB*vNQeU)zAtHN6L!TH3akQMJEzBoes?zKK
zz4Yzl9b)%yEj$tetj2Te8{4qOqgRt;l}ff{!k%0~nah8b#nf*>)><f4RXqXSWTwAy
z=+ABoBd=uAE&kg@N^wDh_MB}Mg?7h#3T=svY5H?s73ZPGSR4_p-xd+0zqXXJSXI>x
z$KrY?DyphAnho>A-1(H3S8&tct|z~sfJskI<;_=LARxd&?_NE5{@H1_-+t3|Ib%m&
zSW@Y+T$fp>5+7p`F%f2TT~bn^bMC-EE;=>PG-^o5Gg2{K_=Q&a>;H7;QJ1m!U3z&{
zE6*vG^8CR)>}QWUEx6mzo(>{IOca%Cv=1{A6K)o&2I+a;Q~IAeZ2gV5+);7s*c)e$
zz0pV}n)Jv6>`zVQhq((l?bK5^eaNZ2`R)geUbCd6RJWlwm4gAaofyf+_yQw7ueu+w
zz<5qwfop-zjU0C$@Oss8KVt|rAO9m!81aQ`{T$x`IK|KQ^}xZyc>+x^4)|Zy?*fPf
z&H)|<&h|6jgb}xSMGIUz81R>=`9{_GukqP_EASN1a!ar#m_2(oBS(&`d8{ak;jd6D
zAg4||xxk2347P*QW4O2rDg>D&kW~U9c1d(Fgw_#`fps#KTY`qMivGu`J|1uX{jZM<
zLBlxFa<@))nV{Gykh1EpEYN8G7-+3+7g=30sWchNQy|z_s0IfI4*^cBE*YXpZ$e8G
zjYS`X5!O-<DdkYVS@vxay?MzaCo#BX%9f0svrc`#U8sD>-+q_Nr6?A=|7?#IM0c0s
zE!>X~V&$@B%Z_PM<cY?{En61t+LL@&uO9XLoNw8>jVGV^7fa@ShtuWazK169+`p#M
zzMZjl5<)O>!mZWAt;u`$Ss#C5GJpKBg`OR4^o_Uk@uDKSb+D3Cq|&d8?V{v#_Xo)Z
z+UO-t&zo6NA9t%M0LNjjDJ8H-cQIdgZIy6EWQggG^G^wwv^Ud()gt(AWf{Z!2Oy>7
z_x&z3sqoPK_jGvTu}3;stqmi4p8MBS?zrb6Vq>BidCqVyzjiDG`t@bR*~99MOFM9o
zx$_q;%FRt_R+wH}w@9uA-p9y5ejg)S`Yqs5pcjo>2fG8^s(!CdSn<cGMz)W}zrCh4
z-P#h!ue!!aps|{6mtqLcqvb{nF<F<=s_TE|XMB5FM!+_kY15|FEcE{U?|%mfA&mZg
zo>Hh=hQVYqEP^Is2&wo8=pR?x-mt_a*pnyNl_MKiC|e56XH6;^Q5U|}NA0LP@-;Fa
zU6Pz9*_oqIRB4FFhydspXSn!7>_)OskgZ`qZ+C4*xyfXzm$iM|Gq(SJy#r|Vn6m~*
z={M*~isKvd(UJ&p0@Vj?BWz32Ui51C4MyvD%7hrjZ13CJb8zKuL+W0nlqX&o$}z_2
z)~8Oq_o0JX*|qv8CnuNj6YtxZlbL<i4L9AM_P`^L^T53m>Di-OgX6!SGmkav*3-RP
zBHw)VHItushItz#S%qGTD|Aw`y+ntbnX|T>znvR)aeJF-dT~y~_^nc%^rBj>X2|Bp
z-B%)cU#nog?)uuSRSdIQ#I?6v5ISZ4x^g<Tw@_53vGbst=rA+ub~%ZPO=R*vKjxVy
zAIG1F1TdLQy!pxty!guNjK2EDt(oZu```1xqrascsI`A(XXkM9gu61!$}5}Q?z626
z$wT1;ZU>4Q**~bx?~Mt-#w;8D+}aXoc7fufL<c4ToBixF11+)2_2;Udjp1%zh#_iy
zIW9>z02?-J@KWviw?~1cjs{jWBNMtVR+Awp`^MI~oy**UgRSJLE%%9uY|PT=yS==X
z47?d7)O5@Vu^TS$#HjkQoM4nls_SkSCJ3r$=y#jV<{E*Kl$3OQ&2MH)M`~lBj~$W*
zu|%n#EVqMb_5R7n3MnZ(-g9VcTsLJ)Mp>nN-`~FA-or);)1m5lwG7qqk*ccxWfN_x
zDbLxTvvu2!5ocd8dg=WSPtvD7`#htryk1K=aPZZ_+_ZJO_av{r=z{YYewP1|uZxx}
z<>{pVuxIZ+OeTdp$KS$Nvlo+RAC0P;sdVe8n#5)ntlQ<}fvX~~Oz2+xfk1wfjrPGy
zT3Oag-QDDhY@bwlPAl{LF1=p~!E~=F@S5=#g}tEZkecHq*r8Bdp%LHS!pBP`{my-a
zUw)j==|h^1Yzqtw;LTTFz-G1DHH*y5%*uWGyi3NE|KsU@GWB22alz;-H*Wo7``HB<
zdz<ZD{WA+mFlsK&Yh+*D0Qg1VH^c5!Xb>0V0Q}Yc`W@JJxDz}IGz=@OKL$0jy}1ma
zx=vPoKB@ZO1AOgg|M}4H3$-=y4HEwT`|rI{%2dEe2I=f$98D`UYDW6Dysj?tU7fvL
zZKZ5UyjsKF-XV>|*9Thb7Y(oU81ChkCL%x(7iz?U_H5tK<t;HvA?@>7FT{)hdu=gW
zuh+W=z;3sn7}hGY9>Fn<kE}w{poL&*euRnU_495MNMy$Go<mz?*8%a8s)+`%X10K+
z)~_EaC2pg?cJ}h+%eS}s_1XZ%*{Lb{1x4?D{q46I%N9=862kKt5O2FXET+qp$&Woe
z_WEnC^n2W|Yky<T{6)O`#{bcyoAEWbYj-kX;bA=Y>@?=iUBVe7Z{WklP72C2wxzg;
z3N<mjU%;(#k>;(pUmSLi8(#ZDcP@BOtDw|RX=@djtGjqjuShP?^c&6_6fkP!z`(hQ
zvDOcU_Hu}IyPP<zf}giku)iRLQP)1q4|9HE;_bJRo14evr>1enh>N)3(kuCF)>n-t
z)5Jta^XW(L^oR(I{3b5aH0G3euYJqvFSXA;`_P@9nX|6Mp=2J;Yx-vwlF)d4#k#@q
zl^9~u7ZVN30G;b|Y&FNq8d;ad>OgjBWP5Y*XN^pzehipD_qIaQH78q?cJ10p0JGUF
zJswZ(ihFQUeXAruRw-=Hg2GDKU`gn(7;O+kTcN>3R%v|!>#*q9EdD=xCZ*vVGD}*1
zPC+(9IQxqQEph7X$cBT(^|RX*3!L`fx+C>Hd-l}pl5yN!+s6yP?`HOi)GYF-Q84X1
z-g9V+jCp)hW`#?Cr@;=Bxw%a?ju^Ev4{vD(z*67-$bVHtLV{fpe0<aSp>cO#7Gdfa
zZ^tCW3zuAY!BdYt?7#G<S5=wx<TE^Q&z*EjtgXMeJ9#go#$3yeUAyVju>&vsCyB4W
zTh6@?JVCn-J(%@#32v`sSl@u)Gy6EOn#H)1?9|J);=SitwXAHd?y0#&q)L_N)lz+-
z?s`X2%LhwAuU?&OAs3z+XbyA;%Wu1!q~}Xc9eNHI-uNFjty#eX_fEv&aPa;|GnoGA
z=ZwAKI%a(E7MEUpA&*a)#y_6^XQQt}|Gs^gI%V>?xuxp%$PiQJ_)Ef+2d|8<#Dpt%
zhq=Xdhx3~L6@+9~Binr~1<iGDH}>H_9%X+TE0`X76z6mRH~HB<gZf{K>wrCe#@)s-
zsNYvN#IV{LmY$ye0fwa(g1f@VFTvzvX>+NtFJG`N6LQOBtHmHzQ|W2wR@ZdFzI;h)
zq2xf3<UkP|D3YWUNRsow>66i2SP8|4O#VE2bScR%mu%0L?8#FaWRD9I40|8i)f%c*
zRb6z&6<4I1&1M-K94!0x?R)ZzstyAlyQl(j=vsnp2~~Li(iy);LuQV6Nte{D2u+S~
z>>+Wj9`@kQ*$Pre*Yy)CA?`RLQ&L<~%Kp<A6-}J{e$J+>0<XR&-JL(;oMG0+Lf7B>
za5^O=rKF^$v3t*60K8r=&%N|2=L|oK9^Gr>DMLd;xciRrEL%LEm)@SkS(n^Tsbv5k
zEOMdivQKz)m%o9wp-jq)KGQ0Hd{rxFrdG+zTKS$Vsa}E9U-Fs)h6ur>E4Ehh{dz0i
z2VTs;i>9!0#jkw%*~eUQ*`-)47V`58c;cyPy!GCPO#1tS4D8>Ro;|wr_lNG|%TGV#
zqZu>#dHG7e&k5l2(HA47GBh&8^xLyD^9!ean)~0ZLTy-aR?6px^P0BxtnKFDzv^Z~
z=3^w@`1%|##t_^(hWPX+BzIK(9BUd|2po+&?HXv%g+l{A_1PwYnHVL9>Jp35cCuB4
zb@l4{`s=S3U|1wajvTpT@7}#1*=)9n0Cc((F8K+TkA(`Ka1WOXX+=g%WO#s)jVtMQ
z000+VNkl<ZsT^YS-^cn$@o2C&zhTWvk0vN=pf5xtzZ;|!K(FYQDNJ6C1j^l#!U`y;
z6pe~%dlU@+2s)3cyVaSTocy%eY?d~g?WBNg1Yv}`L^nC`ghv((i5t<FcjqKcaP|Y6
z^3uK={6?USrvU6R16%f9(+E}Zky6+7<*k0PHbhZYN-FTN=+|%N>=mW|K*(UH)Bj>U
zBQuNVUw(}b-+7%gPand|ufB=RX2WbY<954w?BV+{n`)zJo|^V=?!9XQ9pYl?(Jhf4
z-4c1|zPs?MDtq?qHSO57`}364G%~ZYDJm}3D=I3{b)5^Xmc*dIpgHZLB1OlJadhk2
zmA-v?(Xm5Zz4@f4XK)}rgHs0&A}T79e?2pW<ttY6#@p{x*P=S&>|xwF{uVy?cn0T<
z^jG8stQIQ*tnbV%`a?>6a&dOr{v&xk{|Z8~xhAJ#U@1n{b6whD)Moq*=#PK5abu#X
zwZC9HHL|_Dk?qaFU&74K_%CZT6G)<=@VS#?WTEZbx3A30%abmbORB0W-+c4UDGxvV
z@F0`PWSGvPhx7mKop*dw<^IRt&q>lGEv2l=6rqS9vMb_cPf-*F#j2oQC#WdH1u`r+
zKtz@ba(`UB0!0KBMN|-FiR_J_MW7TYvy|>Ow8?pXf1IW*Y0}b2OOfaOdgb+MPR@Dq
zO`7C<&UeyhFYJtj=vm;|AXb-<kZ#<}no)RDArOMCh0nDvilj`4J<nyCjsr$ppg+$*
z?TSWg9)EG^hImZMG;GXr`97No4eC7#t=<EB_56>>#Kgp*-MV#)4h#&GHk(blQw)R<
zKJ#))2a$~NoiE*Jixl3gv;@15w56SASQA0FfDu7JI!F@)LlvpgL5fBxqSB;;bm;-<
zy-Jl5gFpyXFo1vr5b2$OAXS>wP(m*u^j_|IzwgKWfA`<)&dff~o;lB%GkeZ?`SncH
z2|=lQ_u=d4Ku8kyy+#E|ni;{v$>K|l#L>$q8!{UBr|K+CfN`<rxsEX4aGz@yw8*5X
zF=h-pg!GAP%Cj@FWy1MH@|UV$uEV(yR)0@%9pkP9zYh#bWms5Oj-}gZ;e@kak($R>
z#=YNLnton}*}$R4qAhSfVgDETs7AuETPWe~iZtjNpFB9T{qw-H`FfX6|GQ}?aQnv|
zT+)zGF!1^i>?|elc`hI$nj<9Y9N!RKUhm}&Wj&Y?_iSl^4vx3%^p*|fn`ZlNsOf6s
z^(;-vb8*=Dmlw6u1S$CrQWi^2{^&X6Nt7>M^ArKSgj&QS4v>$vx-?mZSwKWE7)(q;
zLU+np%0Vh)`KQCkFq5F+`<9XQzFrQ>Zv-`((dWl)uQfmXR%J)Nk@kPZ{xKTkR(|~x
zbZfAHMQ=PZ796X6$Afn)E!#u1dHyOd&0IQ8bpaY6H;6BcZ|ptz9zHt|cMM0Gf9fb?
z$=z4dK%?k)3>xR1(|7neCB-8Yh>md)R2~Y-NO69ChqZp`xyJkV??0(AF0He~b$6@z
zb?%p&9?>()Zsy0cZ!U)EDOQ{7sdZeVeo`7detm{}J?aJFBz4w%89Zg<t<Hvu_aJb}
zPv*%)dw6jzA8p-?zS;=z0diEAl`dcDr!KrdS!uu<wOEQ<t5UbvUq{9@7Z?maUZ0i3
z-)Rg+^X6U4#k?3Hy-m@FajBX1Jv(jdJHH^`#!W*%L|aePLr4TdjL-!_)9xaT^aDAo
zYFpX^zq)Rg&&P>)i~~V0617E&CF55;U&1u^aoE!iw%B)r_;Z4_KPXBd23=6Jw7@oD
z!_)`UdDsqs8+f=y55FRwC$e4}VT-tOt}(}W;~h1|XuPNfI2+%a#hfke3f>bfZ-~I{
zsmRU~b3ji`vW_KY(O8dNA>O99$_V$rcNYuq!JjAE)5_NQ`M-)^Na2ul96jyNke*X`
z*OR;}l>s5xCNJ1K%*BK({M~(TLB*DNlV9;Rjz=Z<s~zeKZ@k+_-hSUg?x@yYcLw2&
zcJ@c@XcoTv3RG-%AORLrDr;IxLJt~4fv~2#$qhHQ#@qgmTh5U%l`VHmJ^a%~3N?Ba
zb*y_I#thnUhCllF2%5KO|FmlUSIDyub_uJ;PXJ5W1?AxQQ}4Cu#Tch7ozeeqv57pM
zRkOfm>Z%+lh0-cKP(W7qZ`x)ARdx3rNbSr!TD?4(uO?0}75E4(m$kDoDhp{dMNQ2K
z#R$^*Omo0X;wdQ}eC5uL!7$vND%ZZ|%MSYPzT86_gvVnR#&ShIQkK9!XJw7cS}%(@
z8CH4U%NSoC^Zo_M!etaZ%XTUUs3RTCCxedK<6|ZJu^+@}GykyBQ3+nn^E7HcO{7<}
z%f<PS>WRUoJ^>L4iN|;lt>CE@`{d+gJhOP>v`I~#8AdLx-_}^d32caw3v3DGp`|`3
zu!FkPn)(fVteItdu}OtF+1qrDU32mU9bYs<AG>#;l9L8kL~MS2pKR&sV8?KK1*^$<
z*i0_ho*01ltgw&dowKu>^_geoT!q|Hwa%&#Am%qgfsxfA`_}yg^KO9b_htR)o(00c
z`8-f<?2d^crDbEwWW}0QJ0q-rC-IcI%r^Qt;HIkh9Me-*3CrJ-^v{MCpcl4o$utrq
z29qP29pQVqz2Vjj<j*RfRBHbTj9-8WhJ1Bv&~*vm2;Xe29&cOEcX_7X?Owl1#-OzR
z@C)bYHu2%v(?e90^pE>SRMFN007w36pc}^ZS%@*trE>D%(<d4Q2?>dDR->qg)gc8E
zkx1@X!Cop?>PTu*%~{#Y8s*kjoyyf8TwJHBV>#xs#j_QBL4#L0Q8t+qv^>a4BL9SC
zn470UD{<j|INp)GQ@Nu{wUAIe3;`NLyO<m;?p^nty;t|}G5`JAr25g;A&i3Mz|>{i
z%3b~TAa!u0{({itaaT?<dHu;;z4f?DyUE=mitU-6Py+HM|09dVl~RhJp|$fS>n>sc
z!>y;K>5?)#<Deem6zkzpXxibz(o7Y*e>rtr^I233i9Z*O1SE5OuvvXsJ^r3#Cd_&+
z7m_h8>n)$i^Af(Rv1C7xnXp<V{YxhGa%Ut4DPF;0w%rY*OCmmv3I`7py=rSXF6Bl~
zjdgGK0`X_<_EJ}#1J<(>2_ct)_|7y9kGgNMFuL+BF_aNOje+|vw$uOl^N3q(fR*Cq
z<8vZ_Ojlo40k~S+>Oj_XA+Vy$07*(8KDXDZYI1A1<dSKiO$?$g)PIV_f4Qcl6ztPF
zXSY9P)V+9|uF~57;mhB*<xYZZMieM=W%lI&SxWF-aDwprO9d5kHS^ry+D|DdRrbf*
zm^a&XtJ!Ek8Ed9-^}rSi1OoXR8m4VC!rryBuB-eYZPQX0@=_JF+qe$&{vnM(VBPnK
zh{;b;w^Dzy4F@l$NMUn24lET?0kH6?a>>}9(F-Mb^42+k3Rrx3=64m#o%3<h|MGCW
z#;<&iUfeA3YLd^4T}(~Fbt)~G`mq0SPc}>AR*P78(dS?EI?eJNug=rlmxTu`=bAm3
zn($4mZ^w;lg1PADVu|Ka>==Z&&kd^Rn?NWk{AH>bx2o<HOHgRV*XaEzR@c4pTy$NK
z!1+-#+zsD6_$9Z@ZAXhjlgbwcEzm&fD1TL~zD84v*}2>~KE!niJTtdxH-P~;KMbxd
z(tdBi7YI8#O%5Sgo&^s#`5vWcLO37Db&(EgeP<#0PUVG+ZBowQa%nd&tkGyLQu{;r
zrPPgETJ2yl63s_svda1zg>SYdE4wC<FFuz&)csa&bGR|~wph1FtktSw1{BeKe&NLU
zOjoa$<#uuk>F)G>OTNrG^4~+B{E2=9t`pw~8_RVkM^_rGUJ))>@-Ih5L`3BB^YfE@
z2oFb*lai9sf{D^!@rutU>DbD#C5{3oARa|UMZ1)&Qh|fcQa;4--YWN2o|e3L?@5+<
z!ZLshjhER;xw0XPa`(EAyrJuOJH_NKF2Ffk;YeZj6~CsfjZNQlt=nSZ<@vEvzC-#?
z<I;c9fBN)keZTEzv*$J}+OXCRViT4jSMiwP2kvelQx2x;XO&NJS5vY<m(Db^8KWuM
zM44^eW?(-qk;m!32tgf=1kqoYewmvHriQ#aFA`|mUF;GMD||RgagLpSrxYx+S7iy*
zfxi-CK_J*3vv*C?^A0sR+1-_NCV0|y1evH17a0+fm5oeGGYmFuPWnBfXuGvWE8#Fh
z8anL)d13J-3($JEfK7d5exb)R2v_%s>cI{6&0`wqbNSYDS%yV|++4l2iDn^6m*T<E
z7mF;4Vo67D1H>T1zV$x1_yPxtkawhS<-9F`A2tUY=kMDKN~cx%az17Wy?eAG&PWf4
zoi2KaYu6lJj;9lDde|sonC2lM>=ou>Rvi+Cm+!mKZC@K-pTwMLP>48Ya!4`2FHV!(
zudgmSd3kR|qe1_QZOE`0J31CSPySdIsYjKiD}@ACNkOKb1KCK*OA$>XROBSO5$Qh6
zH5Zz&tXqk`-5SIOskFDNiCFy-Qoj)PE&+`BHgq&Kwf#szzxV4Acg;TD&-}-(aDlQD
z<#Df}YRGi6wzf8q_YfX6e}$SV0i#zFSeX~L`$j%x5u%($jap3yXyQ`$;4@aw)e)&#
zSxxdF&_W!OnXly`z{AZvB}Pt@|MP?QqbqweV?4Q-KP=g)&kEF01fctsamW|r%y7{;
zp?JJ8LRCFeX9F5dy{`40jOpV`>;7~}jYgk#t>mXk8Xe#7B(O>cInK|A<@8;7j`n^a
z(ZdZjk-y<fSUp&8O#aq^v?08St2_4Do<&nikJ|Q}<K0)Af?lW(unJp>A!@%m+3A;F
zrXZJW&Su~S;k1|IPZbfD{0{5u7v)wwA<sFw8A<><YP<In>bHJshmdoSHy1YQSy~sD
zx#JOWMXjKQVRhWUn1_o9URs`hcNT8Uau*ZMMb2^PRpGK;QW}^SIYKnKqmK97F80~w
z6OISl^YuNlx&pk4S*&OM02*qEPp?o-cX_!eV6$|o$;gHPD6wf+SeS5JXP${5<7H7{
zwwzH*@a6hgv7S>6Pz`A3=g*%N8igL@THW$s%a~vDqTfa-z$Ps0-Fwn2x`s@5AUZt}
z|N6K#9qMNQjsm38P_X{DO_IxFLPA1!SMLUeaYbNq*QuZU4Q5ULFo2+U&tqkYvC;k!
z$;!mYOnIS>2Rb|S6Q?o~wYdve*rCZvJB}=XoSRVPIObmrpp2Cun0m=x$Ex}*?}lF6
z%JqK=`#2fL2(#`=O!?I-@PU)gqJ6a_ED0r{2;KHi>F(wf;mx4NB_m&yD$S}^qzVxU
z-}ulT?WF~RL~@eUxti-^*VCX!E^~%>z(-;ETi4ap)ogU8AnCN^RYh8E*xfUrGM<nV
z0jS<k9s)Ks+AxmkOE|;kNd_FP4gBL<wph|*8a#V2D6#}#v-3ON;d6}d7i#+1ob>aF
zEKCCpeL%8oPCuWdPyz|Q1<EiNr0g5FLJ{oM5qX61znYVtyZj>x+TQt-T${uMBgp%n
zO(O5h;oIMP#1zJ&%r;H`b+VSvI@BkS?G3dxw}qu0c7%Dhx_zb-#C*xG5PQpIz~}nr
zd<yvQuJY}u&<v{Y`xVf#Dh~A8O@~`nb-qV3veO%G>+IuvVVX_OO6ipvMTcML!ArrY
zt^`&-AaG9vySNCDS&QGxuwmOH$=laGZx@d6d%k6|@V)8vmI>gkKUD+hmBt3&4J)d~
z94Vy{uU<#e=si#Y0r*hK$BzSo)YW@x#BKWw&wlgeATwU^7CEGQ1{~jZ!s4o_<Nq|G
zHrs)Odj<i0iDL7Xt7qN~O@##xln-VjEviS><+nWLd}P@YA$*$IGco{?Z4EePtd)5E
z*=6usVFIuh@b?E{L@JpFgWfS;d5Us+D8$&p0e|<oJyZ<jsO^?Gb-EBm3POm$%xvGJ
zlKr5iaFVTktKCv^yHhd&a8AL+S$KNm{Na7T;n<?teZqnJo7(@oyPI25l7XyI%t}JC
z*kJY2r;9E`I4rfs#|Ax>^X)TzyT)Mmd;xEbn9U!L#;ZLG0w77l*YCvMHb{avbmD7Q
zZE2A7${<3y29>2IpVJMjg4+E=7d8IVl?INJz+JdqHEw3#{AIO=ZQw<P4wTbW2>gTp
zPiybd+tysp#8RrC`#*v>v?jJD&L1!r_Yd-aL?u=e%<{Gty6-yUZChU}e<%L+`<(Do
z{E&cHY94WVSyhcyOyeGN5QhT4%axtz>^}DmeXGV5olc<#|HMepNTx2@kzV+Xh^tXB
z|Bkrbz;`-!b~6zfnOg5BPo8*HPMW?TrA0cVcTQkZK>6};1j0{VQBl!<^qV%xOs=IX
z;jqv^;AB>M`a^=oShJe>EJ-K*<-esS(p0x^7u!`B76a8KY2vM@`13v2>a&xwvhLyE
z%V=aO2A!$t>W;#Is$sING;yDp^hI9DhYuZG7Snw?Wuet(6)Xw0O739_>mbJJ@v6Mu
zU){;#aEbY(6AXA-$ak^auY341s3F(56shOAwdvoqs=O5}#gyEwG4-@SS$unU)$xIg
zl0(P@PHK2~_|Gd|LY%j4h#d57z|r3J#E|ZHdrzW(t)kVmouiC8y?LcF<7cDoB8W#2
zy5$~XP`=Fqwv3Fjv>mb_<YxMQr+33k(EK#cy1#x--^Qk3P_tNf`#f`H!yuwWS?%GS
z;KkQkgiA~*^BDRyGp>*;PG(N=>!86#)f$<b@0q^&eynVF7VeAN2o4_1R3DUVT(+vO
zse|;2u;17V;JAS@zJbVrb2-vk7=<4nz3t?=2{N++Crjhcc`B{XTuX@s@0SUn8PGuN
z#HJ)D$3d|c-6jBmg54zF<T~Q~(`}o3*=_LNPH`)c3~7i?zR}1p3CD?0IcZbqm?MvX
z0BgzE&YP^4YG2fsdqTB|gXiZ#?HhmFTq<Fy*#o&~)d^tZzco{(x;@i$zJ@=)GF=>A
zc7Kz)UKkPaf!|;!e<d#hiA`@4U!o*d=3&l$A{xQNoHyL0a@C^X7AJl@r(dGCJ=bFH
zy)pV5F7s?@>5)2res|p>ODaiBZ0yiX^)gfAtn<gXxUc%9#%2)0_t9hf0V?~`tJM{K
zjKe4B+ORO2f~Eu*)uOqf(GS)GI|NX-XIa$Y_-7S^XP@;Db4-UFbpf67`#a3=q_xaB
z{vRU@X{1Q43lCbbeVbmp%jheA@-S!T3S-w|y|2buj(R13zthm^$Ub=R$FXuU;>8HS
zwXl(66B81`z|t~3w?}`A^J!$s(l2_}=fT{|eq(2{X_$O|2ww3DfQo8)`PFsXJ3+R`
zZZ2-?33)`uZXJjkdCeL~Gm)#Ml;6~He|2E=rANpe75OQmCAIj`l$T^en@=Rb8l5Yd
zs!P~jcEu^hpl3H<_rPhemY-5Yj?c`x!7)OFvL-c?EmRydW=G_qcRq&PS0S&Lvvrgu
zcC#g`(BNOz^LbkM?>qCg*ng0>B=f`4?Gf2p`bMEJg#G6DZ0WX)ITUsxc@EWUIjz^f
zqw19FMy=z0Bc1abm3>r1sspJT0l!pkEJwvJ2&en&*z4M|th#DbKX{#Z{ThY8OicZg
zm=jivB=;I2?BN1)caYo}V<m;*Sl}A20zT*-33e0u8i52agxSTg2s64xF0fQLXA+i*
zwwbVHggR-;Fm|?vF*<S^=jLfFh4jQnyWnc8Y{8e*0VEk8L?Wn3MUeR^;9RW>h2&4V
z4}@>j)b>G>=|nPKzDD{@dFJDQN{_cYtD@<WZbt&WfJTefXULuZHiC13H~;V5$aax`
z<Nx#EB=Y}u2kZQ&2Y7)=9Z<Rd47~u8z4-5hk>(FTMgL|7+B@Vt+zkK!$+6cr^0}YK
UqCg!_2!QK_vc_{1*dqLY09x|HegFUf

diff --git a/docs/assets/images/readme-light.png b/docs/assets/images/readme-light.png
deleted file mode 100644
index f59fe8843787c25f5cd994e7170616561a481df2..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 38943
zcmdpd1CwN3ux;Dsv~5jK+n%<&+L*R&+qP}nn)bA9+qPeQH{N|e;_Zm4I;Y}9RMgJg
znJaUx3|Ej7M})(L0|5a+l#~!v0s#R_0A2^cKmp$$`=}RyAFy^38jc_!NCW@9Ku2uL
zjDTO_IEkq{DchPjx#~L@gSfi7GMd|1IvVQR88g~Cm}Xw_;(~w>gGh=BskmibWV?C+
z4nIe>r*T}3uj2$)q){{q{J<3g&^gM~n-?uA7h7ul03<PRoNE@<l}oC=i>ggF!F~ax
zd0~FPjL=p8>2kEcXB|I}$O@y`{6K|)J^u7~y*tV}&N}u)e$IL4+J1zgp}_ioD>QO$
z2}=EcSCc$FiMjs&U9+_Rzuh#ou>bAf6BUIaDN1U%`i3C^Df8WQ)7o@dRc+DypFL;_
zs`JUEr;i~lkV9PZoen~oNMjbkzsZA<$>s0fja-v(($v$1hmSNIE9yMr-1c~*wbfdm
zos~FScqCs8TQzT0!(k)2tbJ+gEHLmi)>?O_Ru`V@--A)UHiq~&Icbd&eZEv}>87Kk
z<nYhn7X?M}t~B?bzkVUh(L)q}yyF>wKaHDsrFrD^c<P;L4l&a0;#ms6zLv9}9TY`q
zN;0WT_jDxSOuet%@!;kysyD)=<pM1g`edJ@eZ?N3YW>8A9(Sa@SZYf{x32BoWNpKc
z64zA62_}jb``BNv%i6JN1G;nnI#^lTld;n7cpOz$R)(RoydQ)jH_r;7DK4ju;P`={
z>dBikS5Qz8U_M`xjD?Mz3Bt_GjP?{h?xC;2M0?)9x0Jfp-@g}T*O#5Uru>H3F6lhH
zSzFd3<9k@si{bM!e(3Ml@Dj;%bZKGv9>bBq?M(0o?l5Kko!hsi_-}7v73~AD@Q)HB
z5o7(Yr(whWP8ZoSC5+mQp3T&i@^x#D?A)6TzZ#??`Wl0c9cr`!fJ<tKdj06RM#t8D
z!-EAf8+KBCw&zoPh|=y>XZyiQ&%F2l>~bkdUb&kUJ-+9|2}zFoQMzt@eSM*`jsRf3
zDKQwu+yCXw5IF#1LuqU>U%@QeJcDj*d^~m2_xtmFxk|&36)PpBtf{V!QwgE>k)r$i
z?|sdzb~ti*N2BTIQ-8ZF^asT`Uvo1Xj&v$xv+zdr$wKq{1MGkyA^yx{jm;MNpGPdB
z&3Wg7EI6Ck(IY<p#aFTX5$oTRR1}Je`EI>!RD8PL3UDs3;B&b@3bPjYHh&vLMW84E
zEYW_Ei+FG;K=6Jm4!zysa@CZWX`u=#m0qz-aSyR|=h3Tl+RK~C822q<THp3C01v2r
zTGN3yJq-=2O{<o+_v2!VmTk9l)%amX8TA+hiRe;%i;$EC23+vb`{x%j|L4Pemqw$B
zyMcj$k)eDs1vSvikg!P{k}Y%IR*hR$l5jF6+3IfjSri6xTuc6hd=O3N3Vau|G}Kp1
zNS<P&&wzbaCTdq6t~L7)zoW5q>N&29h1)(%%xH8C55(yr?G*wd=tKtFR1ypU2<TW<
zRr<5f%_fS)w*~<HFM@&4C!f=d`rF=<lp2oo%6{v*M$RxA{ZVdYMWkTV&#bN+MR(o0
zmZONhy}b;fKo8q2dvA+ef*~P7pX!&{i2e`g*i*>z0ki}xtk6(1?Vt1Nf<4b$o}csc
z^R`NWM+*{Bi%8Qr%9n_I7mgPxVg<+$<Tzs1nZmvtm?=4wMM&1;KC21@sKD#C&8yyA
z=Oqm~0;JH3*}_m*Sx5?1!s6gyF?bk~!Gl%HVV*s1fr`~ad9e$kpSM?EAqiqEKX-E=
z-Q7cz-j@}_CMRU3q@1epw?<ZtXV2DbU98sGHKoi!21{KpneWJ#v3t9S>+i1;h&!c4
zthV+`Ei2?wRS0|)8W^{ycXE&nwBeY1mFgWF6)zzc7v!%pO1^2=%5_({G}I_#qWbjF
zUjh>j@wx^(FgA1(Wjn#ERt}8&Z_GlTo$qcN&h{49y=)<bqb_GhU8FRe!^Sk-UBvoY
znAZ2!)GvKwe$F37jw67J1A~z!M&DaH`oXic^~mk-ZtPzi9BRNIRwyKxl9WU(okEX;
zj)&*jiekVU-@mx1m3Jde{e?TGCUgh|vxvVp2q$VR)IS=>i$C{U<V<2lJ6e3`yu(M?
zz+g|-a;Zu~Q%mc8p-3{3pOKM~U0O<t{0|r~GH%|<0mCP&5DW?Br~P8txgNE*sOxYv
zpTcl-tsUSc*nCRh_Yna{S%~$+i~g+66#hxUl4*|W?kO`XN%zR9dH9*wF;X<we!zTz
zj_0_D)g~Ts@6WLLnmV`Y)JQpHoP_~@CgN9z=ukn0J@~ZdjPhJ8Q9Z(+gAt%Qm-`(h
zr=v{XN<yg^(fZxlQM{w!6);j2!bz$)MHFUkpOop6xUJerX_`8gUss#Oi)oZuEaVZ`
z0*sb>UH_RBO3UFf=MpdGskxpvK|C#sR}rG=BAWZl-AlEltG7CubIR;rpw)$cmVocq
z>Fw=pEF8&{#V^eag*!H@TyP!%h-dbgG8P4A;o%>W%XE3?#$cX%J|Z&n+5q5I&HRfe
zyH`?Dns~#P8s}N%Zm4M|7m4pv4k~yMiwh=4C0U;aFT_w1hW+OiS4klExsF)db!}tW
zNWYdM>HFbQ$5QWEa~0JIlaFuYk2bm$>m!?IAbuMFc|Xtd23~%$xlX8u-qB)k{iFtJ
zFy~qPB%!RDy)>Q9{N0JmO>xSmmW7JG*`0{!_ebJxUrs=J*;6tGAStU5ZeQUfxj3q&
zDDqTimawk^QWVrqXW{lP{XCx>ynY=nb(-RGT6QAO#LZ`7XD=4^$n%Mzh6^E*aLw+i
z7_`&1tN8G*j@ef1a;%{u&Q*}({bbZ|4+aMffZFvTIkZ)r`Ds|;YP$&$H&eden2pw~
zhmi!yR8GH9zf-1LiS(txd897fus^c`uc7DPZN1KJJe9?TqoJX}0sv%-?l0zh71zBI
z)!ky2Z8=#*sSw0D!#?yP&P^A!a|lUswbs?$)I>7%xJi2>0fQNGOd6}?WyNAvHH<wM
zHF{CfCTMf@CAR?#uLKtCL9cdx+4EA*qw3BMw1yJ~qvy?i4l@pwY->AOtx#DgMmj5R
z@rC>I?;;!z%dd~u#MqW-Ygu8BR&lEihhCwdl}^hwe8MnnB;D)K9N9#^eSfcz46@*o
z*`VlJUtaVxV;^}wf@IiXB%{Vje(e{ic!7qG*gDL=-ja(}T5X}nbxH+pW2Qj0s=60d
zZ&>Hh>^_ZkuWv=jU$2p|>H4~g3LNKEsbs%UCs%ud1E1dsxxKZbf%#GEt9q+Yy#7MT
zr|8?WVf~cD`76Xn$7)c#a!`!MtD~#LVXO7(GziSMmhD2>!B)bjylCuaN>#P@Yv^UZ
zK+5H+gG$8(1BrW3eUQ;eUVB&-vZ3$urfYBA`eiJ7O$X9GjUQX1{{A+Jil-U68Qco%
z%;JqDNk6WF)KBst1ek?S*Rt)|BxQ=gD{Lbj;l>Zj*y&@>i?8?V@A(QqWk5=rRr-}c
zFFm2&jcU*bW>^irrm?rDX`^3gYpAyESnJlWwY49Sg(DPP_bdjTIUmrOEW4EX7Ec3Q
zEiUiEV<{8Etdp78^y)WbvkkXy#QDn4c=A^^zQz(vp?A}efa8&he5NiX-_wNn+2v?F
zw8;(IO{-1xl_SJ7I#xjaYvA1+&GF!E@6aOKy35P^Ace(rHV5xp!>)Ih+Xp2DfCK@Z
z|E;6MMBHDXvx?dmM^HPJF`S!I+Vye`oH$4U{g+1+#@vVJ)Z(ajAc>@E#Z>6{hNpnu
zWp|I$^b+Kz=d&jF`O!kW-uKC2t|Q%j(Sduu-J74y)Alwy^Q<bs@}~9`V0`ZSS}McO
z@?@RKSm5c>aQj-Z*1)vv!B|DR@iiIWqni@P+YfZ|&v|W8oAnl1_t(QzWhg}4hk}ZV
zcUsKA#hU>c^=E7tcTN6$QsQAGW~8tNe`)ATlx!`f049iL(6U>RX5^22aH2$q45Pus
zfF_oVY$?Xo9`E;QD%p%AB6M`JktiJHTpk*(z9FMj$;7ORqju@PI)A^~!Kk3%i%T|=
zzSp4Fk@*)2_?T~VViWc|kvpYbDK<z{_&V5*skNS1o3)OmX^d-M?~iSH=Ga@<!qfR0
zW#+x4i2*1gjY`f#`y4dYzt};i%u44d*mMgkmh!6BZos#t<G>!&Q-85i$Q3qmivptC
zaRp(ZOp8y=#(GR=X=8=Vi|fVVT!{W*>7a{rl$@~p?Ny*$X*xDyB!Qbn{Fq^ojeUZW
zfg2O%p}2rVnW5uTly+Ff6b>;mwXY*xDsD?K0C-9!3n^8N#T5K$5AxpqMUq=Z|Ihch
zRF_9uf>58HWc3OaB#}~93HMMq2aAa4I|QXbdZ6smaVK*-&{Y1Qks53-I}QV~KIUdy
z98M2qtXn}9&u4RP{p7K+F3+im)hbyjWqDy?<EP8TiaV<CCe^`Y-Z>j~E?FECj8JV$
zOK~--0^Hr2wXK&|Kc=#i=<pUlRfl8m{PjX%!j%%Mr&6Unxh&k)J7ix*|0FzC)~x*4
z=@BaHeH$@*qxklf(&`O7MMbqN7bSgyPtQ;9UYhO9?+tmrR6w(}em%0{Hg*GlQn$fS
zge5-G2)u$?_%%}?MQ^}ID_L$|jgSQbxp=V1k~nxM5)QY~V8A-Em`&LOYob3|OBdw|
z3-a$?^3e=xkRX(BASS<N)MzkTtJ5FI0iAF@Uy9Gn9FM|bkt>V1+HFNzmad%tE`zx$
zu&+6XjS4bCA#&+_>Zf_N_4-KU!^E;4D3WYvUq8{?*zi7i^;K{(9hT?Qo6!+0Yk~Qr
zFOC*3jIg{vH)b2A^l580ms3F|fl4)8O)Lc$XAp-XJXq?SStg;OP2?b;O;6$u;kM%#
zkHdOGMWKV~23rOx>_<tWDx4ALgbDXqMBW%Z*&G&SmV?~xa5&a!#lB8dKAlLX3k6Rl
zVjbjVoC$|xuMdio|GlP2Hca?6jUB`XP}rg$ykZD|Oofwsi#}MOf~-IkY+S#TuhML;
zkalr#k&~2^G~4WO{twxFl_4=vcFJ%wDbCMrV+GV(TaVj^=@5Ty?w^HUS%1G0*H%Z@
zb2<$lG0y$k<JLn@_FhI<JUX*4nmVV`M)cFaS&ipDzU^}jLD|fCknna8ZgM~iT}iGk
zoZ%Dc6^pRR_KPGpfcpbGYUnQ*)_MF(8^DD;st#Hz6^3*T4nCLjFJ*{<Ah%I;Py(4O
z_&^&L8~(&~BC6FaoN{vLT=W1TbCSJUe{i^WpkckwN_C?GMj|#?u3nto%e#%!=_DP&
z+-F4*G-udFO8vVk&nnqyg9pr%^ImxQk>~#IuE4haFlGJz{$AuUXY_QwbeUj*d9>pw
z0YO$-01~RXnPG(4r992nDcfN-jNtpLY~XOfnbpD3F*$m6S>_^xnOv<=oalDC+4tJQ
z&#&MgQ_MsAdO}nbg7|ce%Lol0!LiLMdY+Mpbh$7=;#Rr~n&CoZ=(GJ@e<&smv;gTZ
zsu1L7n7Bs4tQ;0SUf*r62te?m1R>P$Qsh4o<0Z4x96sHn$+48@$*Gjj;@?aVu?nd_
zE$t`F6VNrNNQtaa;+6Vcp9d1Op0ML*U4FzNrQ2K6k>a6ShbGSDJAQnzV|r7(xFr>N
zI1W(iEc@QhNi%(7&PYi~4R$!6i`v-KpzNu40AdNeYqQg;;0*a;q(up{$i?oS>)Q<(
zLI><84pz^Hk4!&Y>gn1?I>kjqrRM*teSj-9`1A!piiy8{Sci|y)DjBNukk-6QG?|p
zy<cpQjxTwpImq!2ymO99&D&xOgq}kSSoi|H7{n%*tl&D6$|p0CirugQVYD`moM@EG
zBsNjYG&&Wzz2Ou=8?uT&;keZ--J?D!GjqMId-{?+^AcH5U6ipCvBx)y5{X-A1$>o>
z?E;o}tE{!w7nd``-0!S9Kz=GgrOJAx==@Fg{%CTXfrdu$Y^W$HA%+s@y10iAhCT3E
z3rj6cWWz_cQ5kbfztu0Mjbj-v-3KBM9Si5*eM;60e-l(EW&9b!lIg{Z6K^hmtf#Dz
zP92=U_&iN)xWGsXUQu*1a3=#rKtJ+T3q-Ly8m%F*_muKo&$w5Y&p>Va)`>R~b+nbi
zZzcM-Pu+)72Q^9yi-&Tdu#;TwPCArCEm*CvLidM-k`BD?<rKDocdosJOt!%t&&|^!
zr%hcdM|9@cta}7Kdziroi$SxfZPVuVaam8#sC7LT1VZBFjiYWixCFWatb<CVUue14
z##6hc&ecIHy=boUT5VXFI?WR)>pPgdCl{n{)#{l+ugB{>43+E)G_Ae8U090dyhgdx
zR|Q)*tY%sJ1sl{9->XTCh4Ad_zW#gj>r7Zr3(ufF_9V!r^=zsQR@{?tdkbp&cPLr{
zX_>Ce?J2Xpf84=!u8Q4IdcD`vXn6C=JVl|wN(0qj4!o``=Z47F?L}Q^UiY?RASfbp
zzz#Q)_bV5*$3`<9)RUJi58|Tk!Y;l)d|+-Eon)hu=OrU)-rZGLdToF>A$B|p2mHmW
z7#8?;i=mHX9h$Iu-7VtUOSGx~_zgn{E*IqkF%mcIO}c5*>UQ1z{3b3g-YfNaMnOju
zK=i%zK&%tYPowX@q^p2=l1#s%K|)irN=4Z{^pi=;ZWb*J&XvufUE5cq$!s=+*>IR%
zr^yV}ZxvOPT(H%WDJu$t!+B?6x=<3?Yc>_++uBkj2IBsU)lE>T>xXsFOhCvCf^u1i
zc|}M$HCnj@4aPuH1gm0ETt6!&Nf8E!5`H?=<9W+Xt~pbc_P6p>{#gyPcz;UZE}eFC
z(o2`_^_(H|_N*`r$=-aY0Ln31>GF$F&p?S2&c=LS0H8UF0AEb?tgld7h<v?Bh$x?6
z3A!)RC>&?UYAD3#hxk<jlnX?8sD`92r7SJ;DGd_+Ee0ru(a8P6dU4j;$mn02r!UF$
zO+vZo`RSS`Tuj|sx4JI7^NherQ&>r24mIkW_zeVJIrzyi3a=V!-b>z8`BE~i^)Er<
zA8-h4GKABk=(g0f$O1kw8%*h+)s;n{9~7oNX@Lj1e_vFm%<BaT>>F>P3EhZk$;kE*
ziLv}-zx@_t+4%ZPgtzxule~Wh`PGDjk@QywoRx1v-c$7l>je;k{{hym9-oJ)di<X;
zJXh2;<Ed89Lmb^yI&lFdA2>zHWAmCpAMGf6)iF1JP=6Ltu}HeBs+D_^CM__~4fXV%
zqvA)$`ZTLmL25FPPk&B2O6r7eu^zx=V4$*`4C%F?#_9P!E~;H_=^QtpJ1!?nVvF)%
zZoL)Y>Y!JRKi_wbqtj4Zy61-ie*L3m5rm~Imc#2t+na%lP(+m=M}m+eIbmdN8AXAd
z69mBp>-Br_|NO^9YhZ3q!z`SSjEhYh6XsYQFVuquY6WGbAk#2CP3b#i@iA0P3{f%Q
z4-<TX%X&*flSGMD2U>wr25Udl%(7jpJ!zMZ4n|2G6K2rVF#J~^{b9PHvHGW=W`IfO
z@z?=$XYsn*;e3Q`@As^h2let}5O#vAFKvifp&O*6wBFcjgwWLf@i;)-$sinQ;}5QO
zY9f^`2Hf*BubL_w5xZp<JxIg#IJzs08=P6F)X_^me-YxYZajakytdnJFbk#_4$9Vx
za~pT(ZEWaPjF^#O16u#LEj{J9+JG8|<nWhTkSAK$6xr{n7cC2mvm)#D7SObdZ|U_K
zm~tt`xTPhRCda8Wuf-q;Vgm~~&%Hg*p6#!FJP|$l@cojOhTTfp`fvBgEJw!Ii^gfq
zWZ#TA#}5tR`PjE>*E}ciUngQxudlN?=Vr>!C{Eiq`rh)n++42{HRjxXapYqpH!tK&
zKH9SeC$_LY4Ix~d#(#@3uv2f==%yQ;*Ig<qDil?cADUFKm^6k7Tn}%2YkMTNx4IQD
zAP0ViMMsm6@aO1l+GCd@;f)>D<{Sl5*i6-iYZo;PBG|nqxuslQiKz<F)b@OE6eYRw
z>N@ERm^7loKyW@rIl9@c@kQY|VuAVV<E}X2F*DD4wcJSeeA-Y%9!={@q{KSszDZ!d
z*M_~`-r=Gd0^VF&8XKv@N`zBYvV*sm$80$uN>q{$h?G!oVBdpsc*Iv6?gp)7^~FR{
zX;w%urZSTm8j%5_M@MjwLa236>PTM57^Fxq;8kF)lA|eEy!PhOfhMNorkKI=<V|%+
z5kvViyi5+adqc-39beKQV3>6`7C&)~>8QaxN}GI9HRM&n$U8D@*7?P6_U{pM6#)RM
zp3kRsWiS3*Y8o1Iv$-w8Q?A3kEf}BKB^M`<tc;QvGfT1<j7-;vbRlqGtx2XJl!&=E
zRpbmlmyP(QQVOr&-CX_w6atgt!fMY?cb*9Fs_os<GE&hbk=<INB*S^luVcGm>J!|L
zuNXbDco>qC#YX6X5G0d-G*t66e9@9>+GBP3fj9y-2pT%v^IR-}H5xh+VX@G8YnLYd
zFZL>)GINCsSVL+_C~e)8iBvkWsf@cV^TIF%1==fC`cxg0CWbqHUs+rp7~!z6<m7{i
zSxKT+RtUl{0e!^X9#47SqhzN^Mf$cRSw0qeRJLBalQ|8<V#y>fSkS8ge4|a5?W>Bt
z>P#|141aVbi7g&AguCHeK07ATem0aYl&Yfd3CF`O@a$vcd(qPJ`#~mP9^nAm$YlBy
zLUy#hvEeXJWA&~{W7XlN%4(J7hb%^d{oh>wgfJ?2wZPs?7Wu9$9&u5be$Nl>_A!K?
zKSu8cyr;1Q3*%bWFR@xT93tuX`M)&IdZDFlLo*0uV^88Sbcs62X8B+XtYnF7db-6G
zn)CHDfqmh}kNat=_RTAoO|*Y;8_*Zu)HbuYP_50N;t}D>z3WOE<Gt%^F-rv&Esilm
zygmKbptKuT5!W$8UL#=h81O9^5Jtxs+1LU#JC)6Zcktla_9j+|_Gs#KfnMfsu<a0i
zDAW-F<}F#PCaujMTKSgT9X0xCXti<lzdjGfy>}l-&Fg+c(WuH|fVzYd4E)sO%^g1#
zT_)q<3?jXG|6I95vdZ|wLb5bUoMXaY&7W<JaHfCU8(FcciL%o<$v=nA)`A5gU~Rt#
zlVvCTi%Nc@oVqG59?#zsV>G#OcwA4wJNe-^_t%tsFrTUa%j*K~t+RxFEJjz*4sM>B
zs3IIMpOS~EY#h9)@%VP9l~GziF6b*NBF4r=xZ!EHJL#Usro1k>bT_3b$;(dMR^5=#
zLTHn1*X{5&Jc$Qb-97lqlgYgQX?@&Igk6YB3~8R4LmNtIP8@bm2dvRRG~MgNynL=M
z08pHiloSMyxTYbUr@JgKfhjC!W68sfpyXt0i5^3K?mC{{hvFX?%iGiTQp1&A!O(F%
zn=^6)K&@H{NpiT2z@+1rvCXRJztlVS)^dDK3GjO_xkqnv(Z-*Z3qHil_&V@4gi>E!
zC*K8wR?Ec}|4Cl{=r+@wQmD+#%+Q+LvTkjb5yC{=ZZFK6&E$BitLaB-t`I!RlMPme
zz4m(excduHjM~g%d=;LJt@(t(hy(ce_O&<wFo(uEm5m>1+couFF000`^${4VnLApR
zn6<&q9B*UJ&E%!cOr=0~W%-{0ehS}h@uIS^$$ucfz@y$FKwhxzHjPmI3p*b_VnmC{
zawi=rSMnPbAkPZn{ebH^H8eU2zrFjWO*{dgh(kGzH%tTDmd&K{5aS+UIhOu@l44A~
z8Qz`p1)Cqldwn~vb)9qmv;h%rzIn`+6&GL4bJWI5bC0Ss54Z}KC6bR!fn)zH)X%Ir
z=Qun(e53=y=>MSTLNf&SQAtaRf<!NxqzteLu81N`-rS5Ro<Mh{)7(D6u%4~<i{0_x
zkNh(6&(?LlE-3SmpR#Mt>O3{dv8b@8o4Rr*3;TT&6EfI;mnOBolQ1f4&QePDpVA(t
znv}0>G_4F#e3qWxD3Y6@lmcXz?Kv+#^xq1zJo}2hUIA@!-cR$>H+>_y2}yB(hbOh#
zo`&^)nU77UyB#*SS@xeZu35Ek$CenKZxKf>FYM89*unTG?L223U_-8G;EPK2i}A+T
z2~8pPc17zK4&U8t`3`vdrulCZfq-UI)O*T<l1=f3nj`aX{OPPQU$NR1l=&BTK?|#-
zVOTI$k}#x?ZbW{T)QPC=5^OJ0u+lX0?dJZvL`xEK4x8PvnH{(F=?lCNAQv5<9FvD3
zaktGmsgY2XYO4w+c@^Tk47X}$pPR~R-*b@nvxYI569#Z7Kw*T<<m=31hmP)4S*6pF
z7y9+Xh;V1t(VNzUy&SwQ0p`Lqho|GZaLjv$z80ALPk^N2Uw`mxEa4y`qGkR?)*=Q`
z&X0AUEZb_bJ_-pZHpP&;PG|`KiZ^S`R7M>_;HfX*AKZjeZNWvgxv|xsB^5Zhcl(!j
z=&-quGKuOK)xWm;6|aN~A3NkF9(ZuTm=4}xl`HUg{*Zay#k#u-@?;7Eq@r@{r*_I%
z;OT_Dnx~ZU_f=@$ooq5dSv2eIhkqkp-V=YP>p_y@uR--Ks_9HsTrbC^;OYWqE^A0w
z3ehV{W;;!h*=^SP!RN1b`^=vebag?4?(xsZvDB>Y9JW-m-$j@Q^HG>EuBsvo2H}Fm
zh&u69R~655uqkzXXv%45y59EmXz1w=e6M}OlJY8OvsqZm%Bl=LxD4^E)KoA#H$(K(
zfMuEbb9(56G39cG>^o8Eprr(jrC`;ookp{*0y4CS2PY9om^59s&r@C*)%SKr?$!Tl
z0i3knj72<o9QNnOuWV1bUEN6Bx4oD(|Bj_(cuKDnT4vp<v}`PTZ7)EQGMeVNK?l3~
z$)O_Y(eB{Z3T*(5#!IJV0vEl=<!&&);J!EgJc-?nY_eSi>(EY#6Ja#(!vi@=?Zxcu
z|1ZMn;^A~|Lw9l-#mk}2UCs^{?)QdG>&tTQS8~m{N1MUC8kl9qqn>Oms3|l}4c8A1
zSXtqXel-NkXB}kOc}A_eKYe~C!5kyqc^A%~hvgF7GWKCZv%721s@9CvDc|OnJV=-%
zK?+VIt0M-C+gpngG_F!K9E6dtF{}nVf|=!&f#MI2zlX%jgAl?8D1B0Z4x2FhS@^ZZ
zGC;|%ex}<k=IG?^y48{NoYL;Q_<^qzJf@JpFUb0W8HAF>dqY0;!<m~-HDsto9~Xhk
z7$rnMj~Wh$rUpgD=NrXcJ4}M#A3VBsCpxD5`|`+(HFyXyC9vRbof~9O;Y@TOsA<Q<
zWMtY(x}=Y%fCx=EUm?2=GIUeUYoiBW-aLakLii_!&+jt|);Cy0(?+dlHOsDV-3BQq
z%@(hLyx>Z{UK(on1U~w+yTPcd-iS!0TDtOIVa_yB{G*URWgf1qYoE{9OGMyDv)T;`
z;=CA%dxA;WD6`&AINS9o9*FJ+1?P<r1IEel)rfZjfKcE8SX}0sPNgnJKf&g;VtlBk
zS^XPDkBtayYaz@eey*T2l^BhI>70+7n0^%$MVRe&PG$|5)rJ*)DABwq+`{7m$_cdm
zZ&LLy=TWQ{ki^vn#^GDgX!7f{4bKU)HLqu1J<`#hl7fzJ-h2y6<^e*bD<!vQkJACY
z`Qwop@$vSip7&BA0shLUPJbbmTL#GQDzq=NiSC=l14-7zkamiZz2-#fxjG3h-G!Tn
zf;K!suo$%4>U7+#K=)WWbmEl&TK*V;SheL0lT2_87L~EkwbJ#*C1%&UytFOX&+A_F
z+bZvwD+ad9ZA~>CQZ%kDPW+j)3Yd_k^Zq5l%T;YwdtzPpSI`Tk??6nktd(j?vl~B>
zyVF1ZRBZO{<s&m73Q~daoE8<NyL==@T_8ja6(H1NwK<&o{RWoVVzs+As)hm=jGGiZ
ze_CY!)-iG}rik(3Puv$*RD4HF^TMCEC?`2_!d`6Vx@SqmP^qz1+L+xa=fdZ^AM{ca
zLbF;oKK+r*rqus7kypQExyEndlY_J=u2g}5kpP~&@pU$8)t>)4oaJsnAAl@&fN6{c
zWu^U9%G_|52Ne`)`(<ZKIBL|mJuhb{Ls0rNybB_oT&GO}gp@s~SuNg#7L~7ridPs0
zSm#HDZ2YUN=@`>*T_3po1Gf0i<e!GR-UM^_Q@v?y$nkoxv0G3atb>xvGc2`}QY|j!
zW&U~y?d7J@O!td8CSdMHZCO7t%hiLyRaZku+llV|pmYc(Vk03BwFe!t0s>Dj2_b_L
z#7QUjVK0*DYj=w0n%@6r?O?<+(IG!f*<AcJ>oQ0Jx#KdYlJLc4f6W*9Q{IDDz|v9+
zDTJkbd&MxG=yV;==6&AgUHt6sz7hJjr?$)ONvPiP#@D~^<RY^28Nltq07sDpNbOX$
zovVDSp6z^+<um~r(dX%1261rwbw1QCu9)m14Aj8$oit;QpE)w$M5-QxvEa9l1P$yW
zKm#wN??QlI#>t5cmZU#Nnnd(qER36X4npHNzKTch=MM-7rtUkYEuZ^b&g>~l%uRM3
z+Z%o`2r$qf6o?O@qC~9VA|0~lHT&Bsht@u90hKM?Bh{u!PcUr({v05KZncJ)c0V{Z
z7cG4A2y8_N#Ny#^?yo-6nXn4ThdxPUa5MFR=CD#^YpvXa#fuj07^X&Fh@i#GBJD3_
z@y;6DuX`Rx#49zRg(V@IXC#cWOh<n-Ioak&uwn8;bE}Yk#FuR0y>@PYUU`v-a)W5;
z1_~DXO6de2q*PuZ4e~@9v>n&dSXH9OgY?Dy>1$=QabxWkdiliAb>qm{@Fi#CN@XH;
zyB;^Q$w-$2Skv`RMZkpK<K6E&{$bxp%OVqx=~}yfj!Z~Mc-l=)x+`j7&o9iEK<ZG@
z^fcadxpnirky^uNHU#2z=ht|WUcILAkAGtXJhUmcH#J>GDpd|10>?=a-Rq_*7&hPI
zTMaudFyhXPIGE^(xTTxFZ0hJdB$%KLp*uAvOyQ(pC1mEwQ4p}-s0K%v9R}@c+K!;6
z_#Ii5#<=^S`xH>jGdM3?Hm)*4?tg@YqoBYweZ<j;a)!`A>=xkm$JgkKK;Pep%)5C#
z<$~aLC4=zy_6j)(x2^n!5u(NrgbB0gdrwa;;hE-L9U6>h=DajGcr6`9kPSof)P^4m
zkT|%<41PskBd|D&UWM&uzlQZoip<O{1qMy3k$*~{wXK|jc9unTjVN7<{ixeAL}Xg$
znjS~~(_$v~(ZKu`QE&tKs{tkHrgIhjU_;`)#G4*XrB66dr1_9DL4v0I`S;tW4=N@m
zpfw>VD4L0lAPWl}3lKV*gBDG-En_wds;X3e_j1m#Xhwi**}Cd}SeR1zuNC`*mxU_-
z5xwYlV9A9ibh9Uf=Mnpdb?bU9Vz6-kr3*ZaCxw>4Ds#T27T!!B47JD2c?<iV8S|_E
zwGX}H_`V0jZM~r*n~L{@Vy;BSOQesWR13W>Z|{)Z^=-^1r5A{Iqg9~+kE<M5n0>j6
z-$y^lw$+V;rT~@R$IgH9n0wWQ-%&79f8+F-)y{8s?k~>`eRp41AI{e^r_yos<4=(l
zbS_P7^-<k(84eo7^c+4X-vi=H%nNAq2zRbDnBHV)ZeIr+tv8y&5%JiWZ~O0Ia7g&G
zTS0V|nGpiBJ6#$}^Zcw|-@%|taBOG_&v#u!7`OymFA-Hft}89R*m@JX5A(~Xy%k(G
zyRI#0i-TUMF-elm{6KZAu*rwZR-FB@emz;HFXji6`_<jTFz_UF9#8BC9N`uBMT$z%
z#ch0`OS$Xxp&;6;kOZY=P!WQD#ipe4c^SC0c(KpLnCpc@%nxHL8My0e3c}190a8qJ
z+$mK7i#HjE#@ix88N-1j?ZuD{NE1}!1uziEAN4OD)rkdg&46QQH(&b)A+l!2-k5$O
zv%?W(#X%psEi3oxSsUYjp+r6_-90BzoLin!zb^(XW-U^zy!n{!4B!ZY*t8Z$0`d*j
z`WT9NtFGne(oG*1trXw6$YNOg#t43F!3nmkw+H|+$+}-nOdm)FDTe-8>9F&e`x%h_
zt!l6c9J1uoc9r857IdzzKT2)~s9n+`%7(cy3M%*aPhCHAxDkNbv|VX_eOj!Lpuz>;
zK!Z_7?3Aa+?mc>Fe176OTzwIsAOXNnGB_2C@JdM)lI`3!^r*<i5r~GbuU^<f^l$d`
zzC~44O}+Su`u>Co68~$IpDr+s_5LdQLk$6@EeAaZK_-lq03j2IL&R9i<$H@OUnA&(
zM3dj1TU*inkZ5Zm{Mm3ZI1)BWL-T3S<>LtR_pgmhK!u0Ny`!@m$?MCHbmWi1;!mX;
zg?sD9snpK!i{yz8TiwCEOp(Kaa`(Lt9s$g!_oLgCAZX1d_?+Bi1~PWYZVvgP48jy*
zu2-s+xn1*~lBbi|!V!3q{^<6*V^#wEW0=qut-hyj=-L~0iUXqlwv-iByBkAq(7vsP
zTohqj!lms`$+i)bAG3|7)9`rMoA}3k6k2-B#;D=Gqx&C3AEgeH514pTm9cuYMP%*A
zZHBW66KW59&ZpD41dpRJXt#%K*e1nh8YMxt9B&e_YTFk%C0~vYg4oyeHAex;h3?)$
z6arTdjg{mtwf`wBQhuAou1eTR?ug?UB_kB9LXePku@)q~d&y*!u&Wrui}hVUqZ1HJ
zCa20C9OddwZ}xU)&x<2Ko0$<Zg0M;5^qd}QvCI(DwvM5|%L;TBp3!!K3xPQoJWkA#
zweQVE+H+?Q4iEm!^A-5;2MIBssWf%1>AFCZ<8f9G)Oj>elS(|`OfLTQx3nad?;R`a
zFGNv5Z^fT??aeZ82T{Fa>r$t14XU^r<X-#l>L%_%NHT*OK(>Fea99rTU(B;jhSb<g
z-=A;p0*@9BmQ@axOKh8M4rlG_Z}S@vRNi#qo0lxEg2fsNSYO8qde9o)zj?h=Qz@-t
zYtH}Gc7HhpzrkhD<3{XB2J4IL)z?p*a@Ok%k3(gl*`J5mMlSschU%P|E@*=P_LM4x
zUd}K!-=mC(@h<t81+yY~HgbLR)pct<hg`afYuC}pB+34<B^M2B4hZn3QR=9PMGh7o
zX~dq}Gb3KTS`4slZN?c1Sma3>iCOMRsF-xNdK16WLHrXWoX-U8ABkrKMcJg149Uo1
z(raDsexS~7c#AoF<Xy!iVtw!QVJCInCud&l&PL&(MJohB<*gRg+9zD#;uz^4qx{l3
zj$pQzpt0k`H`6=_bp|NjdXO!TgCR~1i9l_534xXB5{P__1~AD8ET(m(UQpdDQE|v1
zl~S)-v8xmthhrf?*>hzxT!H&fA5-ut?muNFaJoCNUW;BgGKGms>_XnGU$*F`yV!4X
z)tjS>5BdB+NZ-r7rv=i2L&2qD$RD4Wl0X$+@C_|}DTUM7meFnsy08R)JVRtLdFv2p
zfPkz8C6Z|_Q!AVsW&g?6_OSi6{LfOnpBEtSrV|;&*I`X$ra3?I|F$|Gn}wr66bAQG
z|Iw+ewmN1-whR>}$v4+*`_3o@p<QMpM6i$VjT2P&!c+|!Dl`um0cte{Tp|9r^?uqC
zvg^$CYJp@DV!6!9CGu-j>qT3uZ+p9Qol25)6kUC5EpxGnY6=%#l8{-Iju5r*pak_h
zl$pPphzB&5qcNl{-?WyPlQFgJfK}VekM47HF~mqS&mhH@?Bw?^A1xAknY|yIB6~d>
zN&+Ie?yN&l+9L{9r@7beB?e-9jrf$tY_S=&-PFa3N8slaEQWLT&Jv$>5YO*Efvu5v
z{L%)KPGn`_;PpRuY3e<?*L}w5qN`$1eh-`*J-_9i^O7Kh6BMYsxTAEiSFBsGl%s+`
zIyVX3sT2S@(fU7tBH&R?<Xrk%{dpt-4dJ2S09EizSLl@{DsAHZ`}bL=?xE)H!omVg
zjOG$^(qi(l!@P>6ox)m6^|lVIW|!Ps9~z-F8xV#frw9jXgF`H$(r4ES>>Mn|_j2#-
zF8<zxt#tMvKV2Mllh4+u`M1u3UkS~WaxF-kK)|l(lJf-7M)+gKE7#9w)36=<#c}@<
z6O=G)zFj-4B>Y|Y=EnP8bn61v;SiNY0Zs5GQe!nf^<r!kT^T0y1bJb~Ob2Gl*TSo<
z-Gk8}jw&`Yo;QT}HEeY)rVQD#7uXmt7hb5w)NcBL8%!0}5S;i&v!KD`+d2g2Nr`gM
z?y8*d>()N*(s|c3lhS>6i&XCPu(aW~eqRK&(q5FvllJeuv3eUBSdsUa9~<kj=5P)<
z@+P}u-Z=|;y4s7ZcYsS+_<s`DRBNd21?p57>}S6KFUI@b%s^@EwI5T@ivi#>WS?_Q
z!#|nobw0kUbKWjnq0;oTE^{Qouw80db3#YpNc(FiO}}FD5xwFRDOzw9Yr+q2&0bv#
z5EXF18STGG;Puj+8AtMJUz+=+;W9;l8%#q1G&}jQ8RsQj2C3DpxdxwXXo))o=$VgU
zC@s5Ha6mz~PYU_sue{d4>h=54?fOJKUw1sYQusJdd>5d;_hhvjPd7J}>g$2NxY+mw
zTwkb+ns#W2!p(y3<FZ&y0i)0Ly!)e|Kb(LkrX%yo_CpcF+w>;p^ymWKV5P|&-ljD(
z_j6B;x<Z&3We_G9C4!idJ%)hCmHE89x6b6!gLtD!d!Ch7m+9~jQ&e1Bzc56n2c_i^
zsxT;8re6MJi4?U2P3mIdQdoqxSV?M%5MwBkDB1Cn%U;SyKW?3p@~*eG;&gshu6>=D
zPiViQ(ctxR`w5G!=ZGpQE{p$nNQpPCk4c4>bX&L=deBaNIsD)^E)Xx?lAeyHGu1$(
zkm7Bskav*2ocK=<mB(kKw+2{)G3KT-^xF22xyH}%g64o-@y>XVqHV>(*<C(F8i~O8
za?g=xrd~~=f2;n%wL_k{BU6SJmKp_0Y;kMUr*rs1m3m2~M>%AnNAM34elQgM4(%tM
zERoFVDGb{F;!4gI!)?{k9i{2+Cmeh+OQahSwfc5fA8eDc=_sq%wFM;eKHUFeziO&)
zmDxXvtZXxLV(D8?*krJV4yz9)sy>G_IP?(8!?oH?8aOyO$oJ{>b@qOIx{cI;iBH9D
z<}_Mb4|1sVI<E#_gmU66k<n`RnLHOcc9J?+1L~v*)ikH=M+|bj$4SRr=*UGvu<FhO
z`uGTdJuf%k^2~q+l-?fBssWBmI0a4Xn;B;stF_A!?1WSu)?ZgCK*6c*Gj0-uO1ilS
zv0q~F)F_zC4@#uWomN-*!|LyMXn#S!yNC$8MEiNsdDgA(8}y!UeTG8uoMGo>ng0F9
z0$R8PyGml6Wq0CdW+iR7Ib{1uuzuP&5|v-U5hZXY{I8}t@Av5BR1|0k!7vbD60qeo
zqZbVHdjqljMynj#4&UCc`P6WPvCuRWz&Qa<0wf>x#uRafYcg{;R`4LXWfL6tMTsKe
zkAFTj%N5-wLt~TtzjF=>eARe-y=$|83dF>eyjh<4sgdorSD(z>cS;HpP+34m(6ZbQ
z64;xUR|mip7&Aelc;<EDPXEE=hPM6D!|D4507zJ1a=GuN{uRiKzQ#ZR3b;%SHujJI
zSy$JW+UPv>1{<_LpT;}%9yFUfig76GBmMmFJN66BN&9QaKrtW?&tT~f?Qj&?OX7~3
zhHtJdE)3HLP}33f_9+_JT{#YC)Y^2idk_jL8bIoUp=p`&p4zQyBDS5N{^Q+~t?!MD
z!p$Er_&fb$|4VqR67cpLE|cd%Fet2OA4JYt43(2DytD>k6G7-uQ}T=&^iS(L^b&IQ
zIhA`>2l<fdP8vPMvd}6nTzRUM@A>Ck3wsfRdb;hGy>QkqN#)qsZ3!{QXug-S+?ZNB
zHYF)nW8*^(Xd-W~f|K7)$4PbD$6dR#6K>UMk#ey@mFMk=XtfFWrFWN<OqH!KZU-R?
zO+%KvLBdDpFvYgXN$E6`zg)U}5`nT#2$3P)hxbSr2#P)h&W-8{6okIHM}8k)chcO+
zZ;o;7Zb^@>2+@K{Fj-6pM9C@<eF<NxYo@BL=_`*9CA3Qwx04Qx7W$mg-y-ZAMu^IL
zx03vC!;|332^w7{Y_T%$INdDPK&3bd%LTZCsGCmo`Bb#&A{J2~>4Y+gEX~JWpsXjD
z0I!4LS>7RDch`eGq#?9;{Id?jCQoqfZkn>T%%>-FtE`M5pdS(a-L;vG#)Ox#hkPVK
z<M?hQi+2i?)SFp+G3J1|f2-4b#Aw|2@E}~}k4xfgKut(r#rsiYrfF>I<d`3F6bgO9
z&HL)AgLBcRGIp4rRA26{-*{ItY~O*#{-J3j^6s(MlA8vIwzqALn<VV>@&qS-<~q7F
z;b=%N0I=D*<Z&1)3r%FDb9tbC1KIBy>0ywCvdLOIXP!DDBf6A`W*djL9WX#W49g~0
zX8sUWU>WBW7hsZ#$QwO3Mdur|pa?=8F-M012b(*+JR|S$_Ha5<n08S&*%-?XwiF_B
zKc{2s-PBxdZBA$7X7|`}{mu4pIbl>|W2r^Cm2oabXg^(6$QO(AfI__vb2ABKv@~hu
z_`aA$RXqYi^rfU|5j~98eypv}DmZ(OmZ>{@>54Wf@20?`-$2O#=SEiPg61IWk`3bK
zCbU)tD(?<_rR+H|FevC3E|l`XGEg9b2KV9P<Es*L@*1ARiJFaR`Fkbl;n?QL|Af5r
zFn%_Ku*hgdV$yzMW<MXRpe+pakrc||lm1m9CZ_;^2C}Vxx_u;k5Pe>3>_96%tJDS&
z(2$VBM{8=6{8P<zIe-BLq<tNzJf=k8I_~eW<p=?15(?BpqOtjDj{{I=Pb&Um$8lN(
z#8YueMgUfI(2Arh<Seq5if>=#I7wiYpo-g+u&batShwl^oGrK;EKa~SNcKNgS-px*
z@$N$9RBFA|`FPPMv;rAm1O2-dp|Mm8)DL+@x2IdjgVuPc19}W^*t=hLN1!6B31S_?
zi?P{ck(z{yq}#;m!53}94ekS0Vv7ClHU6|WY!;!VFm(h)FnY=)bZlAZIKTP{yUVZ~
z5MnvFDTJ$;=!2XY=zWlvoHn;ET6VVPZj=<0s@|5JKgo+xgz;^w^<{=WRGE`I)@z4I
zNaMZo;17?0Ldlb$2T^ttYyQ}xVHU~dwHE5dS?j&>2%A{0mw&k-y>0|^FklA+m+di>
zXpQVB34_#dbnNbm$#vp(1B?Qn9V_Xb6(OUB1~~2sk#`n>04*<>AB3Cc5=<kN*9^|e
zO^Zdojh(}g5rMFT&?K?s5}e_4OTe7$PZTue_)SeVgmFO4YzAFX&HI<iqHYF<{^8Se
zSORy6yGlux|9eaxS)~SG@N^ma85?<XuS$!UfX+H%C?KD$v(NT=|2@WeZ-gzExwBQ7
z;BGBMLH5SIs)2t{RO15W)Og?bi-YVv^i#7huFArz2fe_elS^Jc?l7d!T`-6iE-qjh
zNX&oorZiIi!w&jGU&qm}Zz>hsZV|>pp-oT0C1FD1F4J%QBg$ztESF7Jmq=1{>pos0
zhg<$5nfToA6W8a8CB`qP6Bwb}mcssJ$q(8=`UpDg?h>x+merEiFg3vj3k$=b>PBvK
zV&gu}^_LK6#=46MYOAbdF6)XK@hQgCZPwR_XtkHN$5m}^-YCJu17~%dHnus071wC<
zx~ebD#FGL)oPG2!lkv7+c7m|uL6l2swe_@z^u;n_pq|p2>PPn6|Ec|`$!9?*+^nu;
z2tompn%6QX3)b~0V|s97{iw;-o`DNYuWH(;20&{eW%idPYJ0WLN_#a4>x+fm%SThL
zl{a+%yb<w_!rxT{<q0qoMMs6*_03ien(7vHEJSl&;B&I4)-K`X(&19Lv>wO!qo9JS
z`LL)d?;IsfLI)TjO1fY~)K5#G+23>bO~-@4PzyOpC}=dQP7V5jNwXX3wcQRBgWaQZ
zJ7=P&t~EZxp2Q>EAG|tzF7VptCryZ7sKHu`+}r~1wkE7ri>#bYT0VXqgU_mQy2<p{
znjbA9TBrbvF99%qZVVn^21Jm#Qoa1pltP*bgz_s^+8tiqKRjB6##zbgOo1)EqQp+_
z<N_rk4no;wgX^BeI-ZelleV{WY7VE8_mIYwzD%R5?V9@<qpQVVhWz2F`|)@Fb1jk|
zy2ip%E5_DHf%K~q=JjujuW#4>&b~bT(v<?CbR<Wu*|@f(SYx*5jiLa-|8QB6z_qTU
zDIqoHS9W0ZZTKuJ6@!G6Q0x0(=imxG@WzQEI1`5f_<^DNLx!6t1(-pABVIu$qC60k
zH)ExT=hBvkMLIw5*4{qxtzD$RX$CXDb<{t%liVI(l|8su+Qz5r_~o{=)rSrm4RKZ`
zr?whUD3VF0a52Ps>G34%X|<lA+P*ek27$9s;kiCqiH}=qY<f8}y#*|dhpEE9P{R&_
zT%<-nvUuzD_HQ)Mi#<(Rx|EksD0d~vLmwe>J0&<Bjv~e#?*>EDE`;r{q0!<4lxSXF
zFJQBnJ-ar%I2~bwfCQAX6Ni?+Ctm2GJO(4Ml#G7Up8o@w&pv~7<wA;CTd<}H07MTg
ztpPb;<hkz87eT38$`Na3g>_P5@*P95Ult>oM0&R2|NhCi+YM#xNo@f`O5wmux#OOS
z)<YaHXY4XisW*e~L^fM4xAg2<<0)!)wUo5+a_((XTqO*0*J=Q&g-jY^D&Sl&p*4Ev
zr}GG1o|2$mkHndPfoHXDULL4V`DUy#(ebg%==}431ZIu_$^BwYriTj?X!~>}<#u?T
ztNN9el2`B0h%ENkug2x8vn)f|gL9>8WI81P1neXU!zV?vr*C|86CWr;Ivy})>#e5V
zc)EZ+kFPR4a(B?3;<YCXh&9t)L3{eso6Mp&v7tK}XsffhaICZZcD%Kr9ly_&z57=k
zb1FN9ioHU?3irheiak@VA%8}M4JT>`frbmbcnaQQCoCA-t)nC%eOiaY5~(<^q|b>2
zm?9WEkiTFMcuznH(yFTHliwaEqD_W=hN7KWLq6gJidX8i(hLw~7drR=SSJ~*Tqe1l
z&kjJ$npF1e-Iv*-8rm~b>hK1A+GnVKhWNEp?$s06`aCozDLDu`hx2+ZT>!l|oTYZA
znd~Z=_oS7BqWoGy3@DKc%&a|uj-xE0S<_MP%$W=LAVO#9{Jd)FiSf;4o*7!#Q+3=b
zpM3#OxC|DOIumY+?bR4NV6K;d&8@DWU=T8%eP)wa#W?Z>0nU28CwQ4_Z;3m}xMPSm
zD2LF}M&2(=wNL&x96i?CqxRK4Ksncf4aHd;8px%ma)0@oNv^ZMW{)_81{J1f^Q2h0
zAcTbC`+Y21t;2M^(GDzxlZ@FiK%1JpP_$@6JY4WswI{yb_I)`g&D`yLskGw~M7!4#
z1eY}2mr;KN@dc|&gfm`e@t<bW*IW!<a((R}1rKxx73s9{r8#5O7O$Q5N~1yTDT#@6
zvW54f2+OO?5o9#}9SH3BM^^g3q{!CgeYs=TUkOrA*E}rukD@ME1noV;JaEp0zMtQ#
zJl8kW6`#?*Qxi`QThy?C2*kdK0K|(=1CiQGU0)A|W!?6OfJa&bv37P*1r<RBjQ9f9
z;%7OyxUI%fll6($PiFH@Nf?U!^k?@c|7<?Tsr1RI^k=W}_h^%Uy*xK3p7zyCwjf3`
z9mK|e!KbgEiQM$GG5=Q!fO$TAMI&7y^txj=y$|D$d?2GU3vK-HaT_rq$&yM~j^{^)
z$Puev-;<{;$A@p2DYA!btRcF2{fyuI77tR)RlhBxwKov~qj;KNwYv-eU;tiLbxz1g
zRgCvG9J+iYb5-yfQ@AuTuNc4Fi_fnTHYPeOoWZzkBY9om4UHubV-GlJ5C_!lZI((t
zfbn|5>l%6AG+ENLdFpQu66FL<Y2HHYGSevCZKB?p>1m0@6c<U*_EN_R;KH}CWx8<B
z-;VlbOrc%1-jOsxY?Ae~wxY^&5Bqkj*EB4MJw3>X&90RrYH&H^U#V_~Cf4<Nc`PGJ
zz5EAhjt1%!VgAo>7cdRz)^&#BD*ZSqlyb{L@zXPFRViAPo^+yN=V@{l126h?cW60k
zSCdTlr$Y5FcG7g)<50yj6^+L3=meht*boCaH$5HTv?#Sd;{hk}HE<h$7HIa12Jx~v
z-c`{xQch{o=kkblmK(yH|9bd805w6%z88J|M)|O93L%6LCi~zvA4uwW^49|F`t7&h
z_A7MiNQ0CseA}~qkIh|OCC@!E1e-;$d6$QbTrVABO>Et5luZ0`ZW&V-If)KeOB`k{
ziB9QSr$(V^+U!m8;FmZrlFzg%c1S<LH=G~vAnT4xjAUz_z(<k<iloj@x|_wi7Z(!O
zK|ZuLN*_{TcJ2G<=&lh0ePn#3>G=FoU)AVIg{TTx%&;TJM^dzjAO5JM#Hn$|MUfPg
z`T22K<@e<|8FOm;I5us!2Z)!>mW3lnib1Y3c?QXUX#D-Cz`a%*|9E1y_(}-Iy*h-Y
zvf1jJEkL46!Ez)s{4nLsp>jvdC)H!vH$v?a!4TjdDqIej+Qm)`9yD@feQEWTD_0u#
zEm$#y1U7D_j_JX|W|dUcc!V__W-Sfj=RZO?`y>~`&U12VFE>gH+c-^!Wt;3wnHR>Q
zwRZHNUzyF)MtmT6AX8|-uWI-1-A9eaT-KLAl&0Qv*^2_+<*y|LR0?u}s+!*2@x}MX
zy3%Cv)pqO)Zh3C)hP9uc{9Q_{7}j#@*1_ffoD`#&t!*A3>2v1bugXW)^`qJvS&OpV
z^l#!*x+M5D`T8{j!)=d`&&OgCbWgG3@k##JT)hzRoqfFHyuV)<*}cT6bsssgFa)>|
zpuAxBj_9PcO(jzA9WD0X);2+NQ5i1TQ|>42gub{2oWl6u7ooa6SZy}U;bE8@c6^#f
z-kxkicjr+S6-LLN-BA^Vs<Lu?&Pp^1(rTtl&mLTK!MQy2_%nR1Rdalr6<dbREy{sK
zJgn1wx~iJ$l7XdF8-3H@$lK04IfPluD*0|<Mb=lpRV=kz#NhKzawLX_sQk3dkW9+R
z^YYb#if^kcY_~V`XtWc_Ak<W~dGX2FVyYDU4>-MvW@ZGpNVw(kUnqPWF<a^^GApdX
zs5}%Kl+@xCP%L3Luw4k&3BfX(n`I448txl}c=8v{KyXu91UE>CInZl8Rw0Do`RAXn
zd%u4DdL5_$!m&mNHukp-8B2-vQ(PV>;H69WZMB2nRy&A~(inI`HT`<HNlW$-9c~yq
zN-9)#W}8{P-NvG|b_&ay*SJiMZ6o(f@ORSSJCjmYFJHd=(5kK<Bx1wz#+Oti-XsNQ
z9qzdt0_+sX4^_YXZ2Od1g)KL4$G&1+rs>7_NbOp)sVQZ?)<<a1{(}_N+~#9>^48V{
zb<b(3?a_jQ?DTiz;!?l+WkuDHxCqlnm!2MKIX>0Ks_oTUs9o81Tt~}o-BT<G0Wsk!
zabXP#Rsa(D-D>IUeZJWLioykvk{gCb;EK79o#`38GU~Zz+e#_9;F7DD_{OXB>3v+Y
z?Fx&Ex%|?LhzJXzn`s_j7nB(p6#I!iQX?Pz&$2x`>owh^=+QCW@Rs%HV4+jIxpP#g
zD%YgDGY9r{BwEZ0Zm-Vgzm&f(b?v?4+^kK({k9XyX7I{Txe+*{CTsg8pnEe-;{*&B
z3d3tm(+U5*h{F44VTrLLHWuOwPSV+5j9egWRc_7`3ReMF0ddE8QdwIdALxK1De$Pb
zZM;>zQoXF(+593@P^vQZw@{}37TRo^me4!3jgl+9K?K#F*zjHHci(;IcQ_n`g@wsJ
zefmftL?dtbk%Sc=uJEPZbj4kuF9D(s_k7v_I<QzE-}U)JX6~Lc<p3Tr?Pv+NJhyzq
z+6gCrmlAVmtdgoN?2y&huCXpHI27ZxC4>;Bi1@T~6%}`BQP!fuLiu@O=Ps`upK6O&
z6sU0N>O<GW+*wrNXWAks4y(fMLLZq0zE2zWqaeRpF1^p5k9EJj7>S`OE4Nqk?46f0
zV&v$S=8MDOAS7h()<u`gh0SWEZ=b$oFP}rGJTT|V3h6}{Y8Lf#Kz=rPUik$>`iF}7
zYpP^Kh$3zq8kzd#-15}Ia*dO^nyIRmXj1>RG&gHgL{irQ3VL5%^Ditf*u4YViDZj-
zb%^{H#CK2p!j(uKM(jl_yaqKlrRPA$963@b?z1>i)3^}9PyqvlviAg?cI+LxuJ2J*
z^%QLJU~+)J#t`h+3?FhU%K)9;r<$+lwrw$r)BCiMVbBvqB*W6`vbc3!->oRhk;@vJ
zLB^yRTe{qM`9F0bCyQfe-E#upXi|Q#`?ynA$>*MHk9}-oY{*haB}D4iw1%#vlzZ1)
zP!EBAq||+Q3N`?&nLBrGL0ga4mJmXSl+-RWE;}vs;%cvC*0R(Ogb@FVOiaxxEceI9
zN1C|e%rI7LuhxHFUiq5er+oU(^wM*rBKK6}WzKHcS8;CErs$-!WqYJ{S_jdxQ8-^}
ztd>5!NTglC>j$gVvQM%anznDo3|-f;+iivoDOur*oi4(q;4;O&PbP8=SSEcQO*ID)
z$&$RxNfC)@=^y=6dXg&jg;FQ-<;nh!@4GU(r?c8vQnHJEAjphLN_*~<Zno!pcd{~L
zsk1sFsq3Kf+?~shl?AnpA?2$>m<~*T>SvLv32qXCtAT#BLc-EBb47wC>5tCo63*fb
zPD97hR?;zlMfm;xbX8T2WSh`bD66r~OTg`i5DWM;D0dq~N>w1#0-^iw+5m&(MTY5=
zIol?mEjnCdK(B)!k`k!$NGd%7O&Zl~B5g)F$lzR>@a}cLL`~Du0jyT*QAZ}zC(ryg
z^`=WZfq0#xJJE29lzglR^^5dL-ycYc)OH*oZXU5>{kjSLrYFbHNME=$amfQyJ0$hH
zyjxq20pKgmM01xRg_wV|XYq$6r}iA!*KzTX{)U%lPm%AQyvS}tvWxw`!g7s-C=)T^
zDj)t>`gUnf##1$?ZELPSUCQx`bkAuws4d4Lexz4%)uj<6Mwv*8F)?|51*&SoW;322
zD^{&x?fMPG#l~PVnOMDMJ&!;90A{m^@`?%yi;76?T$k^ItGb%qyR)gNs9-?<epszm
z+-^5kiv?Xu?z}t-uU|50UOAgL`RFTJPUKuKeUwRkT1C#*M!MJKc^Q8If7HH@OzeEg
ztYwv3{&7i^8K9_KlhGmSxUx!p!0kgL39%`7{nGx;@8!1u?L=~a@$yA76T{X0(i1a8
z7&cCqk^u<%D9}}v#67%JI3v!$^E}As7ofT-uxp`23qOg9QQtca!&TlJBfC1{uwBq!
z1a7x`i`7~~3Zsv&Cz1eFUU0f0yS#CtiOmchA|Pu20-G&n$u;LV`Q*pF|7jIyl+h<>
zIhp+0!IhZQ16=n@T##KRar*<eDvE%V2yg`v$*6h(0KebA`M?G}GLSWS)|k{AFEarz
zgFJf3q{M^5&qz5g#2+$m!;~pUooMv2f^}Jpii_atwv3gG>T<#$p|q5ackrk!izqei
z*R~$3?O`d(FZFw*#Ol{2MdccQzc8|ES(R4ja9W??K??arbV}EO`2DgZKWi~SSYo$$
z6WUFsc^Plj_7<Z0CNI~;t5;x;akP-9myF6Eo2!WkQRv^pjv_!+6&!Y>&d68ad`I_g
zT^Tw0HBLOCFT3|-^UM<u<MDdQ$lS$`KmEdhQ~CjrmtVjgci%@!auUBSTEf<~%Wyc1
zeoQ74x(;72C}($u4_mn;RqWfdX}<13QS^5k9*am$8!DuTD9cTs0uY+eCBdX{Ygt~_
z>s@+uTLGK_F!)4=IQ2ODsUs&8vNOla30-XD6#GGnop9{9cn%9*y+S%M0+j25l%M!T
z<eQzuyMBSk0T+M`)MHnc>Zm|rVd1)nh#FEC(;w2muKzH*tofuHw-0s{K|&aGY7kym
z>4AbusPY)B&u)geP>8HAP;~Rq3cmX-jPlCH&w~(PxP+@Dd@JDNnsv$ne56BZ&9OC2
zOer!<=jI_5jXx-@hN4Qtq9}k6E5wJ^TU8~nyHt|z6f~8=tV`HkM!;vbSO`n2ds|*!
zz8=8i@f?-*aw(-0LcHGj#!D*&jDaIPWJG3ym;mNaGrpXbdAMiO4j4XS)!KCv`b}#~
ztE6fPKQv_#^VR^k6h&EbD8_3`RAg^0icU<s{rN8nNBRBIom=XC?zJh!=Uy`~?B4!8
zjCI-TlSrTHRIK`tQ;rK68J*Jkywa$!<w;igV_c*;9;hRd+1c5y$iyxmF4R3w4^{Rq
zwRWc(LV1=>q12gOUP)4{+0ZVoaVtLa_#;F`Me@U}*<5h$AZ{FH{B#yC`2(NN$Lu-t
zc>dWZ0hlv)KK)KQkx|b-!=KAnAcVkbv7#snilU(Fuw$!-huxw3T#zD5YV4H$KTC6W
z1^k`Et;$JBF{YOyl2V<~HZeab(lk3Y!4wI+?$cyltaaa18?Q_*T3+RmZ@&9|$>VX6
zrntNk-?XyW&<~)UNVek`L__%>kH;hYe!p0-U_nN=Zr!|qg@mBw-F?a2&UYF_taB9X
zDTCY~BIof#Sv5pEAT0(IL1d`T!`GMc+{b~5^?iU>b*N4I`G<u00@8u|B=iE1aE64h
z1dI^`2-rPxLkVGy26L8Dva?uF8YBSV_Cs+MBu5yQ#$6?ng3A3PfN`fogBT%?$5X27
zx_~3UvQ2ZSneY2uZn|{4M0^UQ9PT*<!7E^n6mmQk*X)c>=O0iE!^)3VIy@DgP?Nn{
zi`F|=cDujZ0mp5(BM_^ylo#TSSjmI28a8<^3Ncc-ZTv&$97MnS1gUxP6l_9DxpK~&
zIpr;%apa>UFJm$U)N1-RIw`HPL#+AnzG+rs!c`PiPL2sPeR=DU2+Ou?&&T$v>aQRE
zTXg?zd%TUjEq=+{^L5XIgB5FtwFRqW1Pb9oVUFfQRg6R$pVx34L`6jc;H<1fQIt9Y
z`1|6e48L(06TbY4g2E!=<6=3re?Ojiel*chk=%dpKLHSmz+y3CE_Eq!Aofek>}TDL
z)OjU%tq!r{er#B*^5GAq<5q2VeSP-@VJDVWYTKidQXlCMXC4Fw`5yN}$=1qp^ZDh4
zTb#gG!Tq-r$#xu55JCtkB^x(x6i%m80IXTFW=M2&^n27=pJGn|OE`F&u<Dr<hz~P}
zVTp?XiO>S4VeND7<+)zsGIpU|4Dxu{S(U8bZe{xKM%+$cV4O}sVRbX`vJMLb3~Ml)
z0B)2J35*a1kG<|(CzqU26*y*kF^x!ML*$ejb9M$<5fx#$?5(O{U2Td`CkGbR=-j#U
z-1PMH58is~t!KaZ;*09S!a^}}<Vf+{bI-w1R@XXx(vNex+;rtBQuxOr#V`)P&Sa*P
ze1Tc{H1ms@nfpV~SAMiILRPs3D++@^IR#`lBw;o)Cj3@eA@Y4I$L19y<z}IX<%%wU
z)f|pxr~LDjqefi*NWyJKtmM?Tuu5M2)PQcP$)4KU{RYERxEZN=kWwBqvOl#5Pf0t@
zuJE2;XQ5vapZvU{@~*yV*68$HF9UkoQdNbncB{h4-R(UcRxu(i(Za?|PnYlx9b(MY
zcAsX}t}n|=A77cDo*$LeW!!JN=aC^TSS4$r)Egr4`vVhcq%Nthu77@P+_V{o!$E9p
z3^RV1#a*}GO3xnM`Elx3{Oh5|m@(^TrhWAVAt4UTCKF(&R-5;uv~QNwSg-rmloe!t
z4-lEsWr#^=TG5`Yxn@;4OI6_d!Qq{@?(sa<ue+V)TdF&RJJb$6QmpEpLLZs=KJrWb
zuPY}1EnRway{fcA`?EAJb40*3@AGWh(T*b=wIp)n$dRJ3uu!<&Zn1jx>f2&sV!jhX
z7*;2#L)9LLI=<;nQX&oOU6&Z>lV~KNG{oLWWG#xT0*Kwyx0G=4=~aXQuLY6FT66>M
zl1($2izWO`8l?2`n@a*eEOALnTrICE&LCry)nr^>ml)%klbea<v?qcyy{@#R5JI(D
zt@l0p=%dr0e){Q<TB~GD>%&dN_8aMwzR%C>G3zF&D3=0j55p9Ez-&;i##+@YbMg=W
z$(%4FV;@PRSo!wifvetHI6+ibXDW)A4&ot@=K#q<5Qgz{b0B4efG$FaOOZUU3$akC
zc5hoXcJb&x-dd2_mKS)eVqF%ad|K;QyRj$s9?T($k2C^jbX`Ac>uWa$si`IXy4wf7
z_K$?;hMweDy*=AQW}c6jaFxEP)=0ZmxdXsv5ez;tM0kCY&Ix9#s+qpN@C@6kGkV+J
z6v9v=_oKc#SNBxxYekv5YP?dS2Nk@KNIV{6EiSL9ATPgwRjb$1AvuZffA|T2>gsBG
z_v*<nKTO4Lv+?=F$ta47qN+#<iUQhxY}se%)ga`vwQGu5$cz7db=;h^?%k#*MVqd3
zdnGa9DkpWd0g{wBlRY-VbkD`7hO%I-i*0+n^z3MTb;R)a_wE=Pbs>r<2_B=JNVemc
zK`n`_wMH&nxbVV|kdXHP6(nr^2)^lRyz&<}-HEk9{Jh-^J>m^3Wm7P#&@<j3WcHw}
zx$JsRk*Z2w_(u`rt}P?MbO1&>v*e*+r96K}abuODoFI>s!(z;*=RVBa)?m(-X7UCW
z+-Q7X-*$&>pHORbR8@8GuwlbKj*pL5YORuBkoo<;z{3mqoYXTX&iE-NEM<TY3=@*Y
zw7qpT8_BDR5PdQy|9El6q#4uq)rGzC)n6l4ja~SURbv;g6jJ_%5Vv3$uJ;E*B2z&=
zZ4su8E8kl9_PYNriQn(5_zPqBh*fKg%0;W!;5J)A&u#5Kn?g;ZO{tV}<)IkwP#`R^
zTYPj<>c5<MI|^3s^n7Yk1vj1>VZVNGMD(~HN;sj5jjl-+H7-K61C&>4M2D$t$aIq&
zW9Fo;_VfQXD7<T!L$#BdY5{QOZZE8o@|9V7%eSYFQT|X7#`7q6Kd7oIo*H-eZwnXm
zkK1qMv;n8^{@6EIvu*>GRaJaG@hd+3_%m!)E2j=Ph0xGYq|}YRq{O5OyunGK&0vG{
zu|;ZgOY^d3)gH5%mGEAjtm4j#Bd?5)GKng;PI|7Fv_vx*HDtSI2a7d2OeLq-N0nRW
z;p<{zIwzP}w6Ur{@t8i2Nk~02I=S=k+CJNfWGh2L2C)68eM$DYzW0%YS^~3i<3>?f
zSSUu19vv1H74@MIH35EqJP3>LhRV#ncjOMrin5uYOKj7&wv-46w;ANETWp|aShdf{
z2Cso4;J8c6VN@=hRR!jKSP0A#23EqT94r@CG!Cq;^5|6i>W?bIn1{n+%qyh{ZXqzN
zkinK#kALqIa;C$gJ7M{gb%#tQ)79U6^UV#lR>_egM~b5?TrvPFKK#&^{^bwfWE?l(
z9HEPoLA(NNq4h{*Ht-3^i?LR9%b5JblRLkhxqV+<*el*%(0A3?h3^QPeJ6;IK%PwN
ztl<tJ#QlD=zIN5yi-+xJ_|;z;Ycu6&zs3QkQZ`nV?tk}3W>m|dgPm7P&^;CSO45;1
zE}TFAz+!NZM3_}y_p|zjyqnmm%hw*S{^#l)#-*KiQV5ZuDxKoZB*mD>F7o5`Nvsxy
z(n<|ywa)Y<72JG&1UquPm{pXxs*t)xGQ!T7g}P^FP7Ad~6pT<MUDAj|ZrHe)&6~Hf
z^p9nfl$J4a^sDH)q{>x|#bU;0vvSux574t`ceZTZ&b2q*N^x-sH{Wm_KA#U=(~wf4
zDgu8C%Qb)x$#<IDFLWMmxW-s4SoRe8=n!kB!lmQaB<s>Wq$XI%DfTmWRTXFSHZsBE
zqfA^gFbq`)JU+?V9qvWC#s5gZ<Ltlo>|h^S+h==>WGmyfA+o8GitT7))LJ9KNCFu&
zXwW~YO`^#71~$F3Zyn`3jqK}&F(?r>NC=0#pqR=A8QR#)5M?)FBx_ONG!kzbBGCu`
zD1w_6*l2@vD-@cbs)i4Q=n&_Jt{&*)2K!zTf%1bAXQCW|1ToSMSwZI{g&QQZAvE2X
z>%L(9>yR)II$aKc!{HdwrAwD7g@uK_jT<)t$eIv?qn1(D2z##u)}DLrxg&NZho)(|
zJV(H(2#yC*fq2-f{7V^#T_86i#Gg8Hk)>4Muzl97{YrS*G;)*09Z`B2$a@rF8Vs1)
z&{&#?l#^GDU36ytJI*IAzW?I(<eU~Te8j4?>nHY`o)pzOv67vV`}Q<hg3E8UKHLQx
zkh%{Y=C}1&hZ34k&p)G=oy$)T8@6Vrdsu#{!SNTacOe8=OoAnwT%6g*&L^|W=on{W
zcY%+LTpt#bAiva)%cEmb1)Aor+oWY=WVoY~QXlwAum0g7HMCi`i^SL-xIH!Nc7uj}
z`}86^eG@_m6h&dh;`s<6m^tfbA|k@szJ3K#N`w%cd(Oase#<K=&~zPL>gc+}ybpW5
z*}8`U^tVg%)7LdTZZRoJQnZP0f3M(zej&_Q>ZIBuxu9PN-z};jK2l{*p^xZrm6&jq
z`D?1U@~m)n75KRDl(1_~=wiL5OOl0GzbeYE?X#Uow&R#Vtu?Z|yj-zbtv47)C9EA4
z*iaQ<GfUh)(J0u&<A=<Wx@4I;q(ne@b-h@(5RF~Ou_1v(ve*Sl5r$PW7!Dl<xQy3<
z!0m^Us=(DGHYBi(NipVG%jq1Jreh&9OLRP1W6PGR7;~@oHPF|Tu;#@&A}NIE`rdo*
z4Y>H?i<id5#mO2nf7Fu5V9cLC-%q{8{$nA8=<~O~nf+lNtJ^7K6_YPaQ$!emq6xn=
ziwdh>DfC*Mm6@pvyz)7DK-Ff}j#-+btNI->Qr-o0YVm~DgM#G21R0*T?!Uj>bi$)U
zs#-qfuZ^`CjEW888nfx(t;6=H<n#Ag)8!!Gsfmx2ns!Vj+Ef(o*||MCYUgpCtX(hY
z7s9u{8=L=r-EGWWZTys%Z>{FRt7FK_^HEu?Q&g_8WRr`hZj5E-(n{{RB8uGNN)#nf
zsJA3Hb5>+h+SkA9o*M=!frTonkgU+X_>h>)0--crHda;Z_nx&;Kqiw3m#doSs3@9?
z3=%@%_iOn58d8EU62=u&5rN=R)kb*GPcK_){HnV%&00vKLf>W;-J}ZqI;e`EuuQ}4
z(<!gg&@{=gK@t44tdgV{KW0_XG2Tp=T_H2i%aHydb!8>DWqazX0ksp!b{zE(LWt*|
zf4<Hd`N9h?gb5)G6>V28l&o*q7<wgw-7L_hkr<Ox5m;ZkxXQ3FCWqG_Q-on%+?-dl
z?Udlt17iy#?BFoh3wldwXjFEMZT*YRE|bUx+tONhc0>3z9a$r~Ed-=RH!j(gU9M5$
z3bd?QO@`$%Cd9aImk&1O)f3X<mEbM}TO0tPp`oX{-R?gM3kwBx>N>|JYA()O^V&x<
z?mk1T9{c;*pxh%Bu0%GOmvx9Bukf23lYe^Wr@ysZCijjSUCI5=;BJC5m@J`fuOBHT
z?!2`~Df8#gpTDW)6OJB8DP@<Q-9H(8Qpo61r$%A9pK6azh)p3e+Mqj{24$5R7Yzty
z(!6s1+*(a!s7hprN_MdivnmYhrNvq2$F2>L3Qg?x!1ucQ++M<*5HIS<+ZURPu4mfl
zoM7h6STi%1I_n-2Ex~TL1rW@pU{X!!x`yBHN0;#UO|g_$>C9VG#cwOTT;3p~+z(99
zT)J24ca;}ztP-NXS;`f4>I{k^`~Xp*3V&>NadI~svzApdX<j+^UlonjEXd6FGO(`$
zmq({xcf-wHP^Pgy!^3IE*_pS-wY4lS>$kc-TfWS8wBzVNZKn2#C!SdEc61jKU?8)b
zWL;v6A`Cb7@ks`8YPhgxml={c4H057q=9PDl;lpZ#jnH#m7dmRR!(>0Ozo5=ERChr
z_2VXl8Dz5|@|_wv<qfyU#{Byx1-kbg=0KOZ)WKa}WlK?%B#*}<f-<<<v9B13vU2RA
zzpWg*XcegQCEP&M$^C~Gg3A-Mus1jgv+)l#HtrEG`Zcj%8Qj5N>abzMRMliXyA`L^
zB>b*Id=+~D<{pc?z7n_hgDDFt%3>o_wr6|E%r`hxRH#CgTPG?^W!@UYI(fwzVeBsO
z(WkSObs28dnl(?;(Q+IP4InZx?fU+w^eE_ma*qN<`p@@Dao1;hRY0_4ID)KvA3L(W
zI9)n6i{ZX5Dk{R~+q=S+K(yFq^VV%hDe?RLC<2ymbu(>I1)1AD^c8{jz?phAd!+WZ
zP(?}aKHavTeMV(@baL8DwV!+a(g%i>b5oXwKB-n*9-RR_?bxjf*+o9K?Di5Fs!-w5
ziH$HJC2Zd1!KA?I9c~t_t9n(+dT&d+MY0`7JsJv-19-jO`n%Ge2tuiouc6v6aCs#T
zOJI^phe#vFFukZg;U=pDR3V88sc)MSVUR<gW{8r<gao>{qZ~#hnaT!6oQx7dpTxkp
z%L9cjC7C6XhR>leYhb)4fTru@Rv2w8YGa~VSP8pI>q(@|3|(UylT84ZSI6zEzeu5i
zJ)vO>pWp94ddbGfyB)Zz`XKc*z$jlNlrnUgL8WgK?@g)Pdj|)ORqrqDMx=k7K%BDz
z0&D;PAOJ~3K~#s4N!?aM9T0>N_pE+v(c+W;J#cc%t^c~X@#&T8)=%s=Z9iir1MV69
z_y_tc7Dt<x%k|{11F56y90O$$8>1{Mt28<}^}q3v>ZsutL^5x66}>uH>62<>d$yO9
zSQFD1J2|<VjkH7yeN(M;iZ@eOu5oHl8=H4|NYC{VS?=#8gjnDIlpgOrd3|g&z{``1
z-e0+TTdYj%{O<R9_5If<#=3ix>Yyw`r#wS4BfEl*9V|?r@go-vxq!dlbuU5)ygn~T
zu-dF3!Rz&6ci367^iRfrI)P1_w<3g*=~n9+H>oD1l!S+e5gs1S(m$7zXc8>_qmtW;
z?AWjd{E9Y6A9Hoj>M}b|C8JD#>c-fv4y(eKbIYHKO-}uy*eWs-brHRGryG+BX^9q+
zV@#M-7|_#>rb|AbQ%-!OiXsFSlRy!$Y)dt1iDn!&h4*KcZZEL~h`zTI$#xtYsFnQj
zcs$~hPd>>XJ9ex~2;sn<05K;+QH{cdl(4-}&^@j(X$rAG??fX;vNqN+qu97jV`}0L
zRRuam8Y1964R#bl_c#J{c{)YayTHLHv$(rdr*rhddjNKog4^GK)Dc*^5kS)=+Y1Fg
z&6s<~s6Y~1ROu(Pr2gKI34zYhM(k)XTwa}Rg@MQ7P6KOn-CMWYozd{)8iDd7ixCB5
z<lPS3mBWzS2n+=50e*ieK_Ce;yLs~FB}gWtt6%Z@gjS0gTs?NlwV?kG#QsY695P6x
zf?({*w-zoq<=<!L9`5-Z$=I+<ycicQuQI6z_FT0`rk&q%tAmF-cQt_2b^kGyXj6-*
zq_mNp6D&_y%z{#<Mtqcs@v};?m;^pea$&zvhW3~Iw5*bQuZX5&yg`r>qD(B<62hsy
zhx6#rj;5JE{=D*=snhS)b?V8zFf#4SLd}oOt@NChDom&L6NZN?QYb_M!UGnw;G3z_
zxoOySTz1J&>~<SMC@8{+pHzf`Pz34enar8<Gc$jf#j{U6j;bn{Os1NuI6nNE;m!u|
z`~5uf#IuaJISQ>pVhgfq7fGEjbXS?Csl#N(4tK}2E&(n4Qua9|xyy{Z@4Bs<NmW@r
z?OPH;(=n-nq-ZlCHiab{U7X&_PJXGM^_gz6i~PJdqlCB!6Ma%`lvQe+(aYX_?&_*d
zVTs*NcjoOVfOaC;j$;r51`MzPZj92^Mw!2&22yEW4L0W+1S>u?@MKjnVj4p#j8cda
z$S5&5Mzql#F(HQAyV7F_eQ(Y;^d@WLC2L*a`3*wo3Mv&G7M-}z11FP(m6F1sOznj5
zKtk;CYiui2aM#EJg<0$84N&6J$t*D@Duh9b6T)Z$l6<Gmo-zeJNTwvdX<-7l+bwG2
zB=`5y4jH}~rn1s>7l{7=$t|=N7%t}X5uA^zy~vZpM^hT|5g&edP!(ZTja_^n$T1up
zYjB)U#Ph)2hkGJNGHR@n2b{sJm~CyZA1QU*`RkBUZe6fo!LBwR@9<!LcA-Dn+fDv`
z$SI-b+U;R>v7Z@BoXlBSiNmJQG0seBrN&8JZTS3>UpGhb`RAYU`NXg2)2k<g&puQ6
zcG~nYHj6Ngo>ZtSiYQOd_Wi5IrR#}F>2}>j&9yvU4Nd6Kpp2}~CpqmnJJDf<WUu;w
zLbJeJW7+fhB_5wd*TLhHOkGsL^S8t^dqo*DCOk`rI8)7@zjhfl;x$bw*enX8Zi~ij
z5>$Eg086AB$%k5{UM00#$_vtW091PQr(T*|)IB;(O_Lx5%17hgej}=Pub#a4%76Lj
zgK>;~Zydj_n#OHIjlFDoo{wIgEbPjc{P2g9E6xa`Z>p_si@jx+hpF40^QGj&%2-=j
z?f6H-BG!&}>=&-R_S*jwLKr5qtZA_2{l<-SX}GRSU6M{wqLH9@Vz}Y%O)oOa2&5Mq
zBr~EWsJUaLK_J1%YECbb9G6gkYj=o*a<^fPtVLF-f=Tfc<=FrGuEZtDDh(ubc5`Ew
z_0A$uM<N9VS*}G{wMIs<f<)92=7=WBepUN4vP!_|X*vRY<6Dq)E_A#Mfbj6}$Im<O
zyorDQ`DdL!|A>?TdSZA7gn0sRpbzXMkna-hyq>3Syp@+HXC6o=E5<ImRe__0L>iFX
zy7HYRubp!LnL7{nRE}h9%4SqtI9Hq0y-Gs?AZ7gqZRoB44M`bvZ<|ZU`O0^rxrzYh
zH;i#RmQb3Tu>e>QncV5a%4&Vu4TB>(6jx~Em-zU<%c5AD?&gbM%eeJ|$hvAb3w9*&
z#e`3Y2oGn<chh;~!GDppYc~?%ijP!77p(FAyDUHBeJSfhBa5?l?2buDz2-wvHTO|9
z)D}^bauQMwW4<lMViEZKl4_5{?~{aC6|6cqJvzDsMFYQ->@M`-^5`tzSdGIXuqpzt
zl=yTII-pBT2uy%qgHS0rw}DGr2R_wYWTQV-mY4NI-S?24xmpO(Eg~`X@te<!cx7Xz
zC*zx`Ga_Gl;aQ<53Omv>c<F`zF#6RHHm>}F)C4n0(I$RgUdeGCtvq>QEPD$5tlVBr
z{~mVM>~w1%&nkT=XII9%!E5``vye4|{i&pJ#68&B(JH9bj|7N|i&N}&`>lWx7rFfL
zy{FDqxCI;YATi8HG->Dpj<myZ39z-$xPx~VL9YaZz=c{NHWZ3#!slE*LBq(!T5<5r
zdG+<Iq=cQtCVDD5VYYp@Ojfudqd4$7RK@7O;e4f(xEsVO#)TTzQGjZnPI{4HT{NqP
z1=SHmo;6*PQ=wDfY;1*Wh}!S9<E0ooW|8{ROD_!?IB;N{l*kcxt37`6NhtCMs410s
zFp%PG5KDOC#w&Pr@{0XAb=}xSrwJIx(YN;6kovvABZqq;M>4K|V%6FMh?OL=@5QGo
zW@~86Ee{U0Wf3))+ty<pZj|NhS`nL^I>c%g%dQz19=bhS^3ji_+&(mtC$5jBs9ZzS
zCBLP2X5xhZ5gB1@%AbAW5&m}b?I?=E4cA>AUsY8_?4A|33!(1>W;Gn6I6reqbYj{a
zAIPeS|2BmxHUzh-4%|*jwUpR}KnJV{gexe!Sv6#X7S43h8-<&daFm*@c<t)hXb~2T
z4^GisY}CEu%k#5d4{jcwm^Q#_F<rT<sxw0_jP0{-!v=1>^Byi5ay|pkH1;jeKKUqb
zzU^Sz{Bc~^FQl$q;=DCgWajy}^}+~t=6d<x&t)TWveKIrWIoy!vqnfsO#DA7L=z^b
zCfX(^>p%sNju3z7x;`a0H}|)OZBpanl6+?K?j}y3mz|R{q=o*c1D6FI-kzP4)2rt5
zpMd~*2MM!ta&j(h(1(Y;3mw<WkCXrz5>`klzt79d`@K1$oRXZp4d@bZP>6Ba+1dB)
z?Ro%GQc~6-W$$J_-jJP>b5cuvv?Xe!f9f)}Cr+H$Q3zpNYv~54*iqkD5k@SaMK!9<
z6jvG6L@ABLnPL}os!6)3g~I~9;$eG{VO`8BF~}uAY7~T8j6R|rjqiH9*{F)Mqo|$)
zN(noP1bq^9EX~z0YygknAg~RIoe)N1j4g-=3L)qbXT*2f%?8n}MMg1b(y+XCk851n
zro^SOt5o3A>VJE7v*E@Mu^98nDy_ecvdvJw6(WuUAUr&L5HQE-bRKz0H&5Kq3q_p`
zqBTnZCLxrccyjo8yfSfp%gtfI`wK$DF?}MAgpBM1fRNm<<DH+L?$#~`v{!7&P+o`+
z*RN64eOE=Xghn6AvWOnP$MgH481HbSC?|7sQl~Cs&N)70#2LLEm`#FrzAxdytD`9@
z*VtU%hl!t$BP67L&A;xNE4lXSE73GRpH7&>AIp|=)AiS#?DPA7OiJrHA<v}$Q*Phx
z0ps}-mE39fZYdx8P^)_5UNr<W0viI`KG?WI5L~1*_6qImB05udvru|xAefw-oNNc`
zq(q{VyFPX2tv8K2@%TPUS$R3#yLGh=IsaTtrpC1I;Ro*J!w?5QPJ4r4gTk4&ri%1j
zFLzuNg%B|7&&rJ4%ubDnYAyb1@;?ejP1=4)d9NbGf)2^a%Q___bggZWSy-40tivF4
z4PS%O($bpn0~+YxzrPt65`3P7Z-d4U_*^vp8w8IV{4!z0{&fJ(7VwZF#De7H<h3a&
zDJM5MIivV|a6c{nR+E#H`y=J?dzs6L9XfP4tVA-HJIBPtbfxy%g3BR>+YHg}aGN3b
z(>Dov#MN8vynci9Z7($9C~ILe8}+L@L>iXJT8J9?piWVN#~UKjZit%)L_O<*^hR~>
zn?_~{_#1ejgfRMxXoib~T8;U{g$9lpW>x4AX(SUhB#~af#<oI@okc3X8d364t6{P1
zn^a#3CoUAigRaq;S@&Oy#Zq?<9Z?cMPY#Pfs8ca2(X=*VK~ClAzsI!P6vF)Ghk)ib
zWk&;&OZ}nkmAhKN-$txhU+fg$?aw)iYHf4<NU8hr6l_3Bx&HUxe=ltLjAH{X>3i#w
z*`*t++&cYx*g30@gB9D{Y_IOe=O2#`Ad&TuJoxBSL_~!1{@DL=;rZvXWy?0+cxCjh
z(bf7ZwXLEPJM}Ut>Oyp)wn^WM<F%>-t(Upq_(gZ~v+kZJkiS^X%3POL{~L-(P8)jH
zZNo?B7ZfO7pN~OjpG9F|F~e`ao6@qfCi}bl&fB>B<`GPs=On_RaQo0m1TbT9g=bZ|
z`yMIhH+yM+H6qyv5(a3hx-2m<vEE`Q#FQqQTHJ1Tv%ALJ+}y!5VQElxy{W!hA5tEV
zlz(*S(4o~b|AdtHv^cD;-`iT_9SDLW8x%#U{~MTtl1SYC#%UlCc7t4XiK$PPDXWGJ
zxrW8Eu47@4styN`U>og#4ndjKPPal?_1<HYN?nFqyCFJ81}4w!D>_CQ^Ka-1FRs$q
zkS8dwHZI&^GU6tCB{Ytml(KOha&u^dkEQ*{lmN243LMvl19U_8H@j1>#p-t!#tA7O
zYQuRP5||X_yu&?*qXC<98KvnhatZ-(^+VGmRg0s|H{{;pZRp-Aq!7ncHJg@DzI%7c
z#$DQZFHI_1vUHORO_x+S&f$OK-p6Wf+5qs@yW=?d#1j~H?N#WyWX|0AAOxokIEAy%
zIO9$s4Dm)yQ9ImtZaBR<Tc1HHD>h00YvZ*ld~KX>_>1o5d)>WI(dDx#vF88&^U^3G
z6<L=&s7lVe?2@6PxVVJe+<cM}6Z!k_8~D%DkMq#u&!Fi|KYE7WIE>+Uk78Y>pKW`*
zyfwAB==-G=mzC#a%x?088vNynWDTNKRh?A(M$@z=$)BPq!EWn@xT=X`S-bP{_Q~zl
zO0Y;>pOzXIchI$36B83d5aPz5!^JeFVmGIxq#OXVX&p!@>nva%k4Ly%uKJyJ%*miO
zAVhAT;o{YsxXEKe;P~VkVRB*ApH44^?DE!G5fZ~1Ydz*x?7i>2pu#Cpj^;9~_aD2<
zbkd7eG+jZd5h6c6scDj`OCxjL)8M&iazJ$bvrpIcLciZH+-~=gLnfcOr61^b9)gi>
z;ql>T?X7{-mD{M*s62`xkpm9*9F7KD`{>Gb#T9buzO=O^b~{T^&21j;P#Yhq`~~78
zgW=5EQLtrm#NdyADx0FmUBdfg-^AR6ix*(l&vVGl&Eu}yZ=u@l=8n7XWAHg=bKn2n
z#k3hevNJt>dsJfT=%}Q$zoTh-W}c6GE{_)RQOY=}F?oaZO&zCI;b|i>`Gf9ehVI^^
z3tqB2lri^S5pByU_Jc0`5s7IxMkS_>*M&aiwKw0cyymLQx#Xe?xaa<d*|K#zy?gcK
zic2rzmDk5K+yAv!UBTTCz0R9cOV_Q*)OwfYWwod`dMp#kFA?H%g!o*-1OXEzEWn5_
zYz!$+OG!x?2#}YTmkw-hqRpj^j@5s5MH9zb#TqGLlu(qDYhGsxDgO>kqiLCyFrV4{
z!oiNID9TL)Sy%;p1S|<^>Huy!*zpb)wXu0KX3W?G7^Xx^By=4f*s!<?Hs(QTb>nZZ
zR+zkRV!i7)r(96wZJiaNV^pBKxzeNJ^*6mW^!Nis&1<E4S_73HNnV9<??>99Pm+=S
z9W0nGG2$(Qt+N2#Zws(Uy4~)TCX?yN%COP(CqW#FIB*lHe6rb3N1`D0TMofI4h6a`
z7`fo6kF7rx$jCLlsA;q)6J-gHKO|P(8vRIDNlD4lLowcw1}Rtgx_9e(+?%hCLQ$Hg
zJ{vb};mfbT;iVTwpzAshKJq00`oFvA+9j16ZoO;TsMlT_u%RUPtS-ss_y?|v{wzAo
z{ANtJiq9{(^}@&`O#G3WU_N6jzGug3m9nxG$)pZ^rMdV?ch5{sH0Oj`lo?l@6@Hr8
zB#4hx3AKyI9=Ixc(#hRy$z?fND}P_MDE0AYo_}&@dM5w-;BB6M?gg@T?dF;*FURBc
z^21Nf8yPRVcqmt1dPy&#w%|3Z!7)oDHC?|uJ3ITf?Ck8@a&mJ1z9%Q=mYkfNb2VM>
zh!B4^b&ino+WI%*n<kocPE1Tp3qF{foZJf|W-$O#h_4Ug5*lLno;`ck)V!AN$<3XR
zos)BoghAAjra;`>y?gg&g7)?e0vfrs{kr}?g!rUUV<``tdpp>*{mLt^xHL_x&!lZi
z;lk}Rgv2-GLV;5f%;jumBX+V!ykP}2wMI!Az-EH5AeV7f)BBnL5(!3}VsL`Yfx@f`
zq8-pP-blQ$1@-IKB?V5&#$3tHqCiVzEo5Ldxa{ogg{?EWR>BiECL<dqzO@}B7x2`L
zsZD*fENoFXgy?n1=F)}`5SQBd=ugr-6u9)!<r_*W*-NaXKnNielXY-g&qeoE;Vsz?
zuz1Ca6;0H3{c9j4IVq_*p_iiKVn&P{&AV^DhRI~&xlylh!Fhujc;@MZhK7>Xxl=zO
zl?NiLc_}@|*RykinXAqUJN~#%R+ewACN0sz;1e9y2d{}q>Yi-P&yw1mZ}=;H`EtK1
zXEq5BlJQy<3#IpyD7zSS%LS2x&N;zhKIeD`OE$Xb(ZS03Cx`UtlWJvsrZ*xwG3`c0
zGe2gx+0KoPi6%Naig(An#=jqXn&OfY{`1shd^Ke%+jcas=GCcV2Q$(fw=eTMW{G5T
z$ji$s@cDdKAw;8{qJ;jn|JR#|ZB$jY(Zfjyab=^1Qs(65=B_%(>FyOdIXMdv;@?et
zWUsEQJo7;NO-xMe3j`+a2=QlLUS7J-=bHiqXKkL`F(KiE10An5(YtqVMNyPK0BF^a
z5ya&q?S}O&(q5a1ELfXuxUqw4Vn^5viJjIWa8RGvQwxv3=`E`_Xz>YcqIBVb!depx
z86HGxoo-2bkz{SQAgfg1_KUjjD>%7l=OobUNpO68eBaiaU@JgTFXGT<N86Cf#Z6Cz
z(>VqdG5{n*X1b1zYdI9i%;iPDMsrL;TzvnOSjB8Rv90IfDcX$md(d_Ln5<^Auh_D6
z`xdXa@#fv@^>W|8ALr#4pCc+NlK*`=fr#)hhFxn!tVKsfvSh*R&bzj4xMjtXc@N!n
z`>kjE@Mjh4Gu+rMf)l&iShlsA4sm7_0au<G_P)&`oOwdK{7tR0e2YBr6r2*NGsa)3
zufsb+NNvnz1H)&AI}~$bl!+ZVUOL8`v6=;2cYDY!_7fi$f9Bg`UY)e;_g|k_zhd#w
z^UoawKyp$NFO3|*y$?Q$-|uJ4>o4)xQ~$wP*|-9%uInsVxH#9D6SGgcv41%t2~bc_
zkcX7>n)s+=?SFD}b2kE+jao?Al$E0~krd+FL!n!llauo~4O0Fapx=S^sj6yINjCoj
zC@3hXM2H!|A8JjXb4X#eTIC~;JQ6L0hy|cB3;YcXwic6NflH1s{yibNKIX8v3O45p
zHswhw+->)FYs!7x6oP*``!1Kd|EQ?e*_fxZr9e>ZG6qTrgA<bjzb|tGRVtqos`k{q
z4G9V9)^-y-Ae?i^M-<}hrVcBTK8JcL?Km=U@gu7?Rk&ocb;_f=^dBS?Q;Xu|2a6iF
zHoC6=($-@g$;izs`0uN4yxr*dV^2S){W^ca)M+z*kU#%2mkpb?@W_LK>UgTEl981~
zT54xrdub#Oj~Gw6mdv(2Ug9E5%v|bZPoWPXVAT${`MR^izf^U_TrKo7-q$L({;KbP
z)Z|v_=XI^pxl8zmNll+T@c6LVx=wtFQ={6e^TEtgIwzQ^a_eN|`Pkqbz;APY6vJ;E
zMvv}YiH(U4=&#ptJ-GMoyLj-?r+DSRWBiMjEdBg}ho5i<Q<c}pyu;dc8=jPM#XeQ=
z`b!eY8mw;OSYW0>)9&p-4c*3RX=#y2d0L}Zt>@O>E2NY@q+HwRICMUc{<?SXZbi!A
zcrLfk=liO`n2p@q0*3ER9;vM$rKC@vK6Q6NSXfxSOWzcfIO5lh`q;IwsD@Nd-y|d1
zMgUb_!PWvnVP)Hp$>7ZA=G^Aypl3@{?&a1%fm3H|p@Pe+AgD<?N;1BehI8<1MgmeD
z1eod#enN<Y3Z-auNO|Zjq`H25lgEWTpnLI%guoqU3Ogn%qO=t<i`3E0WpJz3kV6n3
zDe>g3MM_yDn|N0Il~9_OK4Z+g?>+U%lOy!aTeneET*Aw*k72^3uO2Jfv-8^jzV+^e
zFD8G@>o2{~C~OK~>7UEF;<}sIu``{4XP(Z&c|Xzh^t<`+XAh}~W~77=n?iPxpXe~v
za@iRnQ#2_ObdPe{G_Csk@mdvC`#>jk;3wV9yISQorxa)U{bF&y9*)_UoEDlqV~LZr
zBnzcZjZ=@avnk8NH_Jl#{NwR_JnlVWWB0B`HfZ3P3_ANPKAkZ6hvMw?+osQ)b<e|(
zKcnx-&Y`fdh^PNEO1|>O*b$|<=@a(#d;QB2NeOvP91i&9$flEJtS*;pQ0*J9*Lz9u
zc|cK4PR^kVXjYWsCXR(4NLN)=Rl|Vb)ZnQF1qF5X@7&znIT#t-4H4>dyKgv<F<S#+
zVPO*Bk3aq>p*Eo^q$7ki2x--ILnG1`<Y}^+4Edkli3X8U6#^>6v=*dnEawyK>aIm_
z$;R^Py;tZtP`LdHs$yJspQIWYq#)0wSA*?^G9cSMyc^i-#R0rtZ+`1f?x4_89qrbK
z5OLgnx4H3grOhjr95P5s?78WON-FP|z>xb^Y;ji0uY!+Di4Z1BtI4%B2d%mkZ+Rxb
zZ}aCjuLby5K}mMTE1!Hm@w_t!58c(h?<u_Y<~x3*yeTrd^C^A%95>{Hv2S2DH;&$@
zsBkjooe#L?%FF4JmI{E`Wa8iV-NU4>XLIHSx3FQiU~Re^yH)W0QYTlP6=v_(-8Nlu
z=wo_zw4St9df$CTb9%Pz#noLRb>7x0nJL|qQsT_H3DM&7xCr&fyDpCII&V!CiUNMW
zWb-Z$`BhOIclN`~{Qf5{8gf2ayLR*CSKlyw<}6&U=0jm`y8b#Ie&C)fqY_i^5Rx-K
z|8jEm2?GYw^Mq4<<HrB*{?eTEQTy|q{$+_I;BKiwZf@?%CJCpmH*{||lKqkJ-61W<
zsOybm!F2tAB7B?DjucY<ui^WedZS-9K3N=AO-X>a-+tSrY1%sDoq*mC1~#bj8c8-q
zm9o*s&|xu%RKE_bmm%E{MOA?$)~rhGmZtB)5KEv%(5o34)UAz9QO5OliD^t;rG&yt
z$;Mp6Qri%H9}ak1T3Y(2$z+l?o9#%++&&x@QiYJxV?BOwSMTSC?UTE!9rf5<tdG+4
zK*bQn7T(2VY5R~6PyTw)HFRCHDC6>1fDrx7LNNBy)520l-yW|$d~LM2eQ5N|PRSiV
znfS>^N#Ws5ORT;0`kQp>kiu=Z+*p_OJ!kF$rhfkeF)`7+_@Ad)wrC*{J@25z6NAgG
zlbPq^#4fgobB+(W{a;r`bygG+Uo5mU-_|NOexp|hNCZh>vF_z%-C4Q=-*AEa_Pmor
z2HkQ&M89sy=6+VQAg9=ml!C8UCh_!$SD80w9(Ub-3(m?)o_T&0FTXy9rGGBx>E~W#
zz`*lay=Gmr-^0_7KO(NU?BWkj>SmpL=S5N0*Z&c(-E~Q%5*==OJvzBdlVLi|;jc_2
zTYTV2DW#C|+eRM<ahdUtP?7SY;Exq44~bj50UBq7D#|{VPN_%7jvc!r#2G<{v$ON^
z76i8yit=gjF$n{bl9CQv<&Ihs<8ry`zyIF<f|Lsb8~Sy~C=mvstZG_UvbCY<kW(Ib
z%wdtZyb_OJ;?*QRP2%-Ss(q45FWAk26Wdc>U$?t0p-Y2kZ^)HomWW1K?Hw<Jp8u$C
zuIrmGzx?twweK5-8ngq1l*de#tPo=Lkyl9l$j7<&tle7eVPeB$_Jre)M>L~D(~77m
ziztQAo5~CL3u2`%rTleMmV3+MjV>iFQq^bv>HPhMVb`9}A*E?Y#<~p~`S{Z>xct(K
z1EhoI%v-=)?|i_9O@>-vbaWICKlm?JF8!VH<3DE?=doz3mGVlB%sd|^RSYf9$y!}%
zqyK#Ad1bWU>02v(<VcONS|yWo*G#wKJ)tBg^Ghkkg_<tOFY~iA$H!+2Be~@2e=_^0
zpP2UTciix|YcQKld^7cXZn*VM9(mwjjC<!z-W~HAD;Lk_^i%tD{VjJ;RNUBM37{x~
zM<2T1vSz1yK(Ef0*&DMw%CarhWmRr{mV#ciKi}xFzI#4sNFk;O2(C$;mXwr~nv|5(
zk;ZP|vh2LPLsrK$IXSt%5TenqUr4!6*;WmqY1;oaE)p;HBqt_5nv&Fn#<>vH1ZBo5
zhHTLN`#bhQKmsBzE>1d~PGK^c<Y%9K_W6DH-S?mnq65MN=S_tLH^A;61JA4#DBn|N
zxOQVh8e8RB3t6R&GqKC770Rpkc2?k@!|R7#C6KmH75Vlb)jl({abW-e7ko)XK~zam
zrIC2#)0=kfcfJ}1O$1SIEiElAeYvEhM30DwkhLKO?Wl)D?%DV1xkom1Sai7OaI~Vx
z=^yQ|nQl@9HULL*(-m!78}7U{NGUU-qoeoxnQ$zk#-sMA5aO|&`Kdz|ZE$^7RV_Ql
z$Hg|;Ue|RVeeyrt{`cX8hK6w8Lyz;sBM(wiQp&{1Q<(GPRF3acZ$DYJW*up%ofvfX
zSqwV+EOcFG<K`_a{c}0jDC;A3q-X5T$jsbP;wjxzjK1h2t?FC>=&%LIR)y{DlTSOR
z)8Mnu>>ZlclT%JUi6>q?mE@$vMsq7EEhR2ChV>gZlAe)CbhJ_Pz;3tk!8>nq!6jGm
z?g!&}{@JEwZhLg=ilUlRU!GhPT2$sA?NEI$=VoW`_r|xKNVb67+}y><Nl7I@K>CxS
zD3?ho8)-@+#PpgUYMUb^CB-3Nd{duEA(rh=UqT2|a#GT5jXRR3Aw=V3qUJlgwR`vO
zk8ZCiYJVv3+;h*#kt0V^SXe0CZnu2xwbv?#4I6fELPEkcK*byi7km#J--1>D-`<^v
zMOE%`0RO&c=FG5)h@)&O;SvgoVrni4u9=!k?t2KiW~sShg=l74wtKH-R`;sc)Vdmq
zWs>Fv4b8o9R}>evVP9vMbIyDJI1CKKfG~ohzMtpeQ4i;wc^P5m%e(yE2tY6^cVG~K
zlwA44P9^W|;%ry3zM?fG3f#)7SkT*A=>_r%?&i)FSAbhXXQ0R`l84O;iz~k&?+66E
z3$&XHkh0&%i;IhY>eHvsF5|GaMx!wX$%h;$Y4Jk{ZxWm;mU<x^iMLeF^pQiL<BQqw
zsI{IA{wpCVveUOJrCfK{#<k9?{y@l2fsl=TbyOAM)Bj!?=@5`k=@jA8p-4B<DJmk3
z)TQK-B7M0CNav+X!Hbl1H%NE4pme?a{r-OEy#Ky?_K!Vtc4nV_&a*qu%zS1h>j4nq
zeK>z9znIMUkYju}(CK$ss`y4xe-@`Ueokv5Z{xPh-LTjE&j30(=WA+OQP+-mYO7yg
zJ+oFvR%o!aQ*{KMFV?4p;_F)}X?PgyR0jtuO0Ufg{gWn9D^?CvfYj8~fY@g|T}1XO
zj0qPPH|xdFkWMM@PzE&?q~%_a>iSHX?d}G9051SL&x$cN+fqjOi71VWKmZ-?$H>r{
z*WHW{huNeZg%@LuU)skv`k&IW>!LFgO-R$<)DY(uX0GQPyw5?Z0X)v0N;{<Lr))cz
zG{2n}e%8*mx7$LwWQcx^F_!PBs;b)Y(^n!RO84+m!rt&c1+5es3xe}vv9(>vm>18S
z936}Fit`~52=2F<noSH2m;L9EizfW&M!fI%5VV>vBSF*#V_!?+dOb4I#J#6vr^G(8
zFXzZ#B)j%%Yh1*sxlu|x{9*fvFPt?GP#MU-kV;)WdL1k^lRrk<TO5Q!VX_HAnq80t
z3AtR1x7&Gfg?4Y>s-2ykA*#jQFHbg3vyBk!GNHJd-6t6jKTo(}NNDNW^S3Iie!tIU
zob}7<GI-1xSC&f|rY1#nfy3x|=hIaqmwU196lHAj!(~aJ5B*_lsG_HdDe&^tJ6_`~
zXYu2*cxX!l&PqR3JJefmd>Z@U)<0)_hpg^h;7&buSeTE)CC|I@ahoG+D_`lu3w-G%
zZ_+OyLj&u1u-*ejJbDKDk;YU`|4e7hr8?Pn3nd)$=4oHTdyvSQ)2JB<3FB{vPxuH4
zy1N~WY<AjUJ^`(I`Go}#fJ)E@Q=%Sr@qWCeFTkk@(@zogV;+^KsofpS;ew(Fa}Sse
z2vrF>RWT%Cky5ZPni*}*p18T}*VfXiIAt=%D0BDrwrYc9BWHu-3yXG>0mVBL@}SZf
z_jx{o(qR3_jf@X{6_{T!s>EeptY<janKtJ-r<|&(0n1xk=SScyys1Z{{gv<%jH>7S
zFl59b4DE1&Jvi?YhHw7|6`j}da${;|jfSo#BR^vFEN2FI0qg`nJnp0|OR6Y$n!zbE
zVpv^;e8W`GO`3eGg?us&<rYRRF_7f_Z|wk58|7aSLcKhK0c{}X5;37Ef6}8gQLph!
zfoQsv%RmhS8v<2i(ZxvUYbx)=2-5;S6JovEiT9jb@D!t!aQ)#&R8@KrjvC+bs<3GA
z8tRInBRJUKXSa2EeVfXpDhPD!nwI=#+2Usy?0ahN<)LSZtB{5}mpRL!x*CrA^V1*U
z39qr<^@*aO;R($wd|p;hOW84~+?bBJxd(cPX`d|FO?;)i;HXSx`s8+~AO!c!MZ#_V
z-M`%#hDsdm7eBUsA~TaS6W~duQ>!79yraK#iIWlRHoC^px6QQCR>U_rW@$BQKEcdE
zu2|r@vDweJS84Rb&5-i!sQKe9?7t0ShouNYN%ITstKCm=Q^Y*UB|ZOjBS&UF&DJvT
zJd6w1(((a|`(JEYrICHSW4$`XZ~H6rbER)`@7-)Z{{Ft>@yc!zUgj}vI5CWDP<ms^
zF~gE%$xF7tsduQiHI~VHZQknm-;X5{Ce_}aluW<#7bXhUyw%JmMBoS9($G-9k#pm(
zqCUJ}sb+6c_HKeyp7BYex6N)8eRp|r@8H71f*C1u&%V%0t|SPUJXum>*S)YH6AhI1
zGqC5Up-^^dQr?H=#i~+eO56pZq~v5|a6~=ApmIuIfwZWl<vz2gr{{v0LDIj%fX6S{
z^7Rt^_6>xz3Pa@vA6@rIMG{Zt<;{5{f}V8nhhUii3bn!Fg+e(5mYb21%24(^N!w&k
zLQ?ikrUu%kd`vWOiB8ss90tID>%%wO0j18GY^0u(q2X^4CZ?<d>0x!ABsIO#$%)a?
zgsZEoF~lNR1z(IS61z3uFUvRzuNJNOdgO$8<=>Wa9h`L)UWYOqrMQ$3y|VeS>XuEl
zjr(eiu@a~fI&M7EQ4N)S%7`3OSB|BJ(a8LZLn5pYVq}WVF)cm?<J5Fw%@TW)@xnGy
zn`EQ;;m*bv4ezFY;|h}S@-Ch%&c==AnHh<C*<(pb&AAh}Yh_48*<nh`=jdb`4Ie_?
zb_q9!Gr~3pr4x|pdXl*>@Zys)Vzd};b2Nv4^^V(BIdzlxyg9)FP8z#yp5fSZ$2E-J
zJ+k_K3ZpGItK|zwzY7Q%sq|*13U3NwFl)vxpxl!+iQ^zx@8eKji?uxr=13-|px9DZ
zf__QzWNWC)l%irUh3oP7pc{^`g>w2R@T<|~t09U(6v%i5@K0CajvXOEoT9)R9wT@7
zkQE}2$Uk@xYv!~<@eLTL^if(6IzyHjS7QMAp`lnn5HsHE=_(uip`oE?B#hD*h!=UV
zQV#+pKxZ0bacv_m*MQmW$4vsc{+PNk0hIAiD!zv)_8-Hfb1d3A%<v2>Edk65e<GDy
z17a;30yK7C)H_o`T*t4*(lL6bMdLeS&P&a=YgJWM8nveU+(~NhU0ek1MQORYxKv7*
zp3QRj>`>#Ghjm(es*vd36x3+*;NQo<X{ktm^|=H^e3B1+o{Ym!;=AfV`a3+lefYIi
z#_0yY;`!HwK1XwJA6|r|n~V)OS~?IRqc<}Ep^5~D{&Lh%`pi|F_LKz77rh3mV0sDK
zh`}X)6I#Ig22JPN#E36We(dm%1uBUYr1RVTXA=GBDMpYq=kaBxyA4tS^V)U+wWmIO
zeRSBAA|1#FEUy;urica-%3kd=ff#p#9oKX8sG#j`>9^660&%cl=kpd>xT}-t+1DoD
z4XN!De>sh;CnTWaN|CnVYs@-n8}l|)No$_FDbV9GPESDSsZ*TwS23Gdoc!^6si0?P
zXJ@oSr^-OC(1Q=89X{p=FdAoZQvB4mFv#C=1H~?RIVsD*u@N2-(Y=)YeZGQ`(x-EH
z*hLle$#%wQ?%}auA@$zVKK7KXM+tCpMqFKAA03~Z%u+vkB+=8^`6gDcc;Dm@r%rR@
z*VYsLMGT<>5@rsn5k^drA>fME(7^o<MkX!W((rd7NkEq}E;6_1B^M>FqRKO%E|io5
zfa{sEtLjy-tL^|IuuLWpH=6Sv&*L5?J@xm*=vPd}iCFy894`nhF7S0Ls89y}cmdLU
z;yZ&Jn`moma{)12Ta{Y_sd*bP%T-wB?DVvXmX;PGFOThMxXRjJt8X}y2``tHx)6~q
z!3Jf2O_hn)LJpEDK0C?%*vlBt>T<?nAz;$+dr#8283b#}+?M(3U^Wq+dK_@IpI&aj
zg9$;VB^%8>*PS=C6aP5yrqDd@{q`)_$FQZ9ZnH(!?^kqu@u%x02s=lCJZgO9B2xe2
zV8q7prW?f^y1ZqT=XPp3(;vk38T#SS#bak4U+!QAx7IW!zW7AGHz=^_xT{UbTpJDl
zvZ%fCZr)9UB=rG+_Sq3p4B=dB>r|h<$vW`yoTUWhrv!%EhK7i7T7yW8D&>31ri49L
zp?sEqaC|KBy~T|*GN8l^Qs28j<%Cq9Z{6ahN}RWW&N{m{92^rSPg9j;RotbS?1g9q
zQkb+gK|i*X>Kmc`N#3O7{)MPD9)>9oFRwceHa51RF)vs1#U?|Z4-YyO6-d)H<G8u`
zleM@;BJ@l*bl6#*2RF8O`MSHGSg4gQl4U05;|H4zN9*ONU;x_M+DTksTAOrNamk^{
z;T^rW9%bi)T6_?HVE`~1$BPoXrG=}O8>$2Hl~WV|I^bID^!$8@S5?oz!^5LN%xPAm
zuDbe><~j`Z#z*<r8usP0<_2k$@D_P|GV{GfB{tz@kTKu6WC?9gLld+XhgOaNS{I6K
zfEo{AASbU8m6V+Q<z`;cs*YH{jJ6xjT-e#%JY!&aUU5C1T_DZ31mlnQBT|vWUbn>l
zjDW97!+b1r5G^=qT|pU~%iBbv{~-JGx-kjPA+6TX^rWQPH#0SFuFv*o^atGNpPng#
zdSZ8dogL=qr!GIU`qh_$RXw2=U<m&CF@r)r-%pEkW%9MBoA3+r>{qoBzMp|Dd{5(`
zD<O*&;SV74z@qAVd}6UtjXpUA(73#RczW{t$h*P>dQ|&ckc2LHV1Tc~3jK+jqC4qK
zYjb_QZL!(WVUgFU621m)6!l?#p^ev7>Gv^Ik!3i>X@50%C_{2}d4*>Cv~zp=USnrV
zz@q+h=0Zd6+y{$4z0~&|`Ny`_4i1=#cwN!-v9{gX)>q$)KgG_5*+1EQbq^0okzvCM
z!6lZTUcE%S9*ew)&PZqQ`xp0eMu6CABtKL#fa&q+VQrSpp}uNL2Uv=@=ko7Xq+uxs
z3<i6S8YgSA)b@~M;8Z0iQ%V5jmzo1!?)@$&0GAMBv0<qdACY5iwBFbFuU~)W2V9?B
zOT^|q2n)stDiE?KOz0AGo~h}S5fXBZQ7fR0{aIYcj2G<6CntB0N=ZrS85tR&w50<%
z3L#^oqtzZZF}<!{xBe_9lE+d&<-~EGWBaUAVu^my^#Drw3eNmlHa)}py&A7tzW3Ip
z44(k&s}xS6<G=b6Z$QV8%beHFz|MH%i&&ZmwIyW+an*byaIVYqkdH36j1A}e?b=d5
zRtF1?eibA<`NA(S_t5b+yprs?TlnI#PHCGrab=U|+%7tc{4;0{jg};dw;46Z;oB;|
z8X|->9f7+yGN`rC&@Bv0Q#6L<GpkmXM8xlM)78G(y^Ghk;cb|sxZ*UwDtnv!>R*I)
zmIqoI$G>K@#gZ?8>ycuy;$aArB!T_TEvaK(jGL7kp6l==G1%1cG=&$jtD-=tTl0T<
zn#UFuaW&aPF@d$LLM(;Fy}4;wdEz8tVPOaup!){P{1{2<E_Z-nSN&;eZcYzI=BYni
z{8@SSVRK-V_E|YB$e?u6Aa7KqkoVi(f@>W|f8kGg@>obwY5BHuo8jogesR>x^WZ00
zOeXT&m!#RJ=o{KzW@O`po8>t=Vu3uo*aZ*v#oHMn+trlXy=SW&yDegW^G<qW5@TjY
zV!Da*0p`^yJW(GrbjTNzzp3euWyjpb+H@9P>Z#=2)e|ad!Q`xlOjk@eX1MiyQF68O
zU<0sjPt5YovfJCL_b;dS(M&{V|ChCOrikT<Y{&zGC?SstER9Sk6<Y)mWVn>{v_H_l
z0cQ4gzIx5-VbuD!n~Bf~H99NmVda2866=&}CABVkR#sL#z~CVlq@HtxhKECyRY5_a
z<j)^hCK8f3I|H7VW|>=N+(~_JLrIN*q0soTU?hh=p~!RHMh?}nPEda&f*m?80>D`l
z=RZ9xp4{a*FBBU(`cCN!(%;G1kL6o}+GGKo>hLx6m(cjaP*SNxGaWU1#sX0bM7~ro
z^Zdx(sSy~MTeG#b9Ue>*7sTY+BP%Zp=(>uJ(-6$BFG#$F>@Ww-9@-CKv0xtG{ib-@
z7@K~zzb%C&N-w`K6;xbr_MDYXYUs%A#6~>SX15{ykYAEJ(%tk&g+&+PqZ`GwR?O@6
z@peoY>=JrCwWOjs{>u8kx$4J{XZNjaLZE2@IslA)Z$H2>t9VxV|Lrm00q9g8g1wgc
zP|u@EEau^F`%heHKr^5t{ZUR;weB;p;O}rEs+!kbaN+zf+?KnzIBjC&K&3b&Omwm}
znEmf_A!2XLH!?`Tn%e;5p&v%fxiBLppvDcR&Bpi5&82~uON!jw+<&R5sReR6g)3x$
zqoX6c`pUrfjOOO%iRrq&S-Zuk_p^m}x7U9`3QtI@JXo@@R9iy>iN~@ri`A1e`fsUl
zr8>WS!UK<vD~-V9$d85JL^Owva=oRf3BhZ>jMg=7>U`Ck^`6;IR!%ZR&mT*t!7u9$
zAk|Uqb$OsAQGPpLac*H2Ym>>|F|0kw{4JY=#M#*IV+(wt!RvT;w467SXdBS%J&Tg4
zuZMa-sTO{E30OGYE@horcZ{$sLy3c#m%A7Ivd4FX-*@Ohk(6R6<m%g1{tgB?_;jAz
ztNjGDK4|YG<mO+na2{MuMw}@nTPq-7xhc46*=5P(Ky(ff*B-v8r30ZWeYfs-e6bE!
zk?=$+j&ioYd?`=5$*bZ3)ABD*1n5PGlDZr;=4>44m~mg&&cz&v+}U$WW4ISRby&)~
zeO2vat`0i?Nj(}WuSwnM5v0+zbApR5iCTf=b78LQysuypCAi@j#n~R%`}<;^Ul)r7
z-mi!ZP=t^w(!P^6b5GBuUO`t_4yNILw_oVn;;!5z9!&FV8=QO?V?o<GItr;169Zp2
zmn%ob2n29R=cDYFo!uWeKVDR9i%FZI7PN>2w**oO621oc{u-7ilfJd>lmv(Yihd8>
z*y9vf5mq0GFx~JIqLj^OmO+p0S(=f+ixpg2GBVxr%dp7<rE;w_=fy@<vswfS|0}L?
zcA4FT{~enSlacLG#KTsxk)5p|L?^&UJy@1b#k{9ueuFK$Xp(1I{xb5KeaKKLR0rea
zF`ntf?|@S7RGk|92Rl)*AF?WqT(k)PLRmt&%%$hhdL=FbkCwa0DiS(WiHfl3)S%co
z2o?Fy&=K=U1R=J92q%kfX;Z4<cugP)Edhcsq|c8hCQCeAJ)6(ku+!f%=NSEzz@X%P
z8f+UJZS(;~AFgtuKGQvQSXJauHVtmPDsTW~*Q{(P0w3Yaa?E{4)mKYN`zYAGmQ!>M
zTi6h$f>3xc4w-z2sb!8fH&)PN3S!~-Z2`I@kL-p0Y|}-;Oq&!=eDKM(U7~0R*OV<j
zk|ZTbm?VzlygX*HX{@`B(x-GhQ}X>(p2H~evCf!Rd<W{6TZ~%lTZLS4X6~W2_T9>!
z3Vye&j+qjl@60&WPghhZ@W1EOp05I!cs0x{sq*N@vl5GM+NTKHC#a~{A8j&5^R_??
zPzKpMo?Q;R>k$Lf6}yXz>-|18A#CkJg!&Gk#-Rzf0nakMID%CXLXg-lbovM!<$tfD
zU;gh6ZNcpSr4iab1V#UsiKeuQx%WRDgzd&**sA}J*2Tl8^8W=H+9)9YBb+9@`oaHM
a<PDSNm$7e|PI{66c&I9AJ+Dx-4*nnRE!XY<

diff --git a/docs/assets/images/use_philosophy.png b/docs/assets/images/use_philosophy.png
deleted file mode 100644
index 23922d4e9476dcd0d3f457828807271ff58b1b01..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 16463
zcmeHvXH=8V_U=m&P(VbQf{02JsY>q%0s;ckOXwX!@4bkEN>M;Sq<1MvAdt`rih_VN
zsR<<@ozQ#d#-B&e`G33j%UyS^d-yP{@aCP_v-h6;?3vjQubyfsQ(j`c1ONb~%Hv1c
z06-WH00dXbNbrAgiLu$k|04F1SJ5TI{{)lSyu^Pd@X}V62cSJSmH^-upz`RUuD{vp
zgrB#`Zsz_btd!jd&QrJlnyHo{mZ^v>x8|ujXGhd?8yYgSbb8W+>|RGl)&+Ta&s*vC
zCAq25R)HSl4Ib-F21anba%ey^mb)!hi8W{~)-G5%5bNcc@}kZ!bo;2g>F7I*<02^`
z*{yKkS0h(;^|6y)b_!L@KOcWJw}|DMiT^(5rnv<m;g=R>?)-iMAjCf=`tRfaQ?=Z`
z7~UF1sfMlfu%>JLqdtHzf&9xo9bq-#hWIr6p|(NLE6qfXn1i-gn)RxQvx;XP;xq)r
zSLA?Sjhvuff?a9r9To1xFEvjSb?jh9B4BX!1FZ`!O9RElH~EuU93pP^JyU;)6v%k<
z*PQMg1XwtmN}vosC@WpiUPHY=pPIC#Kfl<wR9*ehK)nd{dZ$~Q>hCbO2qV}I6;#z%
z!(hh3=Z3`g;Cw6MXv7I!GKcg$4Yrj!qlfF?@@rjkP@7`ehKHPARZ+8eMBx7AKrQs1
z40b>(C4GSUx8Qi`^Qa%I20E;6p?hnap5p*9=aU_1i42xB_RveM8~5jX5%csLyk#n?
z^`SFc;u2bHC#^ns#kd~53AuAla|YT_{i6yX!tkQQ`ylu3*WzJbULhm+VQZJ%7MaH9
z=hgjr|Lxr0i6>1Ota?^mp=<YmcxFivhy<5?T=S==qHdqTe?}!Bw&JKzA(|2pHSnVY
zT!fbg%X^YtC&NendOX<|eL13>vL_46U-i9L35Rco-}tu%%&2bxfrT7*|E{SkoCJgx
z894z=f23{U#;+OJAo1K^b^K(L17c}}+#>&2$zL%XSWNzq^S58Q3}Uiyk7x5P|1v&6
z03*8sG-=bkBl=^hmq`GDfJR><<)8UZ(Gd{l7&txi{MXtB2tBUe0@^NqsCoF01L2#-
zJ95B|s`}H*e`LF*K}g&p_HbL|&xFDa@B^EbT<nbaBik31aNruJ!z9U{fj3?Q1by7X
z55@n;79n==3NS<5J@$9t|1G6JM1UgpR}?DLOKCro_KtdeGd5nWz|<4F5qE};dKb1H
z4|q~8>9b33l@^BRn}!@}uy;k|3891UnwBCm7Iq%*s`$KdD(QwL*CTi5)cDj=-hHN3
zk$lUc?G@kUr{Ur6^?Q<yynV}rQyaPo*NyJ9CgKY2u<K!pidB?KIUY*ib8;#yFExJP
z1h%m@g*W=D+L)>bbG!TZs@BbXzIq)4k-rDlVdG|1c`$%$)eT72)X+O9P*;UGOT4=c
z<L^%JIUf6h2p<Y5EeQ6k({&Wr0qd}f3o0i5z?}x`F_LtF3pomvejrTJ$OYM5Gy_e#
z+S70Uj-xH;3&!fecg?*HSXGjZg;S#U_l_U>2p$!U%M+@41`7kjNFD8`o~p=^gV{~(
zY}(<kNJVax3q&f=Hi2T%ggw(!hn-(kF)``G-RxdYMNaVCXa^(OL&!AiKuGga*Rq0P
zd3=3_F;^2**QtK^v9VcH!JOS=ZBNf{HQPf&D!Uz*O883O4z`vgN(+J#)vQ3xh9R!d
z-zXrbM(aRF&}V!an!cJuK^YRV?P)SJvzB+Yq)*_W;YY!H1c`AO)(y(__Is?*(ug`a
z?&qk_N2N~g(z$&WpI9Ss4&sOo>#hAmP>{Vz8N^UXQzG${LvL>G$J&I6F_CG<rqyrL
zSaJ6KB+7Y#!rG3gES&5mCZ|y+%$96j5;yjj4+VV~U%B1)85=g~{-oa9nbid$DV&)9
zq`~wJt!`&nBen#sfPXcqCiCrgja4>bZur{#2c2Z?&~ightg|LrY{4^e(tlhw!j9gd
zpP;Wk<Q$_>01Ac9DqTA}ndB5EzIjiIOO`9e4tE*piY;E3;hB&uR6Abz_Bk6B%&7{V
zo2tAQus8?p9%go)XjYWvdH0=9JTZ~R=wLe)hJ2(jm2kBqNF;l0%C7lU-HqR?(Dj0`
zH*msF8OAzhtNO;oZ=zZK_DwOPJK)z{b66o_k@n5wbaERTti%$!!?eatXT9m%Fs>Y(
z$%6_gw)Udce7!Ekr9!05))dp=iY=OKTz;H(;5(?=VBnQr7;HlPuO)bg#-S2Fq0sYu
z(?p8pVXg;h96gx8#rG%nmHf+43*6D^s!zM^<Lqzsh3Q96)Ax11t=sv&tmhuB4^?yJ
zOF*7(rw_rL?Q>@vlN1cxt4clZ|F*~Y#p+SuK+<I=Jp-%1Rh3@6+u3VuXv{%Q-Zfjp
zzLpbdej4U_?`=mUu0Y{<-tlzX{QTe*qfsrp(b?(5qA`Kmni!X7XlcR1(g*Qwe_gsI
zj0BSt+k%fVL3%>6+HcRFQq55$kps;hW{yiNO+eI=)S7CwFZoyUQdjNwQP&;Xcz%gd
z9J|x&b~+orvzmsU)#j&8)l7YfZbF&3o`T<eV~816%p$)j&Vg={5ULwgE10Wdw^H3_
zPL0=e>f(t}RPOv1@H?74V(+^{lrwNTCSye6^9o8=&V`03Uz(NG98?^qbe=$+hlb=*
ztT9h~d0o_5W9wFUl1;V6x<-Or!@@zT>{aGwy5`U#4L+<Mn+vw5h)Im2-h}qIM^%wm
ze*^q|n6gjUWEBFJR<MRQj=e=T242u@IojI9D3}Z9>E3KC*FEx%>*l*2blIRo(A>7m
z9tUSVI-s-mE`C#1TA(hPsYnx<zxA}TAS{FLdf5YOcOSl2|BAZyG!6*_8}Ss`QloW{
z>fTp8KqlqtBb5><uj(h0<u7@F$xzEmDgS`FS;T;}m$f@Bok5qNxm_oy>khelanU=K
zZoT?B6W8td?47W|&LrE!BacKI-q`M4Yq?JL-^k~ng%AB?rAfdw0^=4g;+Dk|F{a<M
zbGTq^0~C<I=bLhHqv%oo2Rw4ve@*!t+ZB$8>e=*^-l%^Bm8&M?eE`5ID7so61zEex
z{9XjSM!!s4CP4oz-u2=Q`e&{kHPYFXzxvKWkmEh<o4$KL<UYQIoc6aY{L{Zf2O+Nr
zU|e{b3VrqfZd$bSZ75{=!?c(?{LwINY8Q%5wDs%GRTQdDPH7L4l$H6xUb6PI-N~kV
z&E~LMvLA7FC~LZz$;s|0_qxE(F>B7aH2&QyH)Tk5kgy0&P(%~rPWRizjhp3SE{7j>
z$Q+hMwYb*b@^-aT(nSQo8|5!A5%$UUZx|RdppU{nUHIk(;?~557}oo#rtYD+w362!
zIYMqo>9bQcjq6?N%4<kL-SQDfXz#SjcdR6`a*6jN?hdEusviCm70eUWJ`@LQMR+{;
ztrCb~jv}dCnE~$7nW=B4caIyt5p`6H4&zxoeb!rS3WsvzDM$x<e~f-j)=iAVR05fR
zOiQ?s58pBt*M7=%z8PBlvL*~%8azxFyoIDmM3-b1*eTKzf;AiuB=vdf+>rLiHTowN
z)A_8w;Kg4#=PRjP7}<#nQu6^EUf`D1ur@WJ$gb_N{O$xQRDy2JJ_lRe-Vhz3eQc-n
zVFx!4mGmFUW#oc=Ou>c?R^p$dg6Rok+$6Y`($W$qTTT7khaw)7+~ZI6U7x>t*64-R
zx%x|=5&C?6_Xu-_Dt$=LB_vfyia1o-Gk6?fmIC79-UhCwnVK37i`;j`*~h@aTmB_5
zbbUzrQL~ZXM!o@kM|h9k4=z!?El(Xb@sd|thr!z3aQ*nq@&sn(fw@<j+aDlfH6>d{
ze(p;?h1vrMr5{@l*my-15)+dYn#T<cIEOg73+g}0Yf}}g?tOQz4*?xUM!oqb@txw0
z(5C@3YmkmaW3fC5mK{FY35oYy1_szz5zRP6UXgxhX(@X?Qb}Z(y$f8-W(NsX51}j<
z-iRkxvvH`oR8p><P#(+H@dQToY)(%`TrW3hr)aRVNa6a{(T|Mxh$CS=889VJwh5n}
zanB$|S7q20SFjT(Ypw?wboBF9SB)!JcVoqazT@up@TrF5??lO+*w_@<8EN39mwx1>
z)=QgReO#0a8RdkdL{vmZUhCpyv%6T;yMzls$vfZ5Qhd{VG_+oKCC$lceQ=g}$2Vl$
z;}uwNR8X&s7@P+$c9rs*Z<W}na6L(=CUU$ZyigmwKpQRkd$qQV5#qC}zy{;H%zv`O
zE0XxMbj7*uUk*ujYf1&5E^?N=kpEXk0uWZx;M2vP6z+d~=vOY6MT^gE;l_7q|J%ub
zb@E?7`M<PwMDn;fiwnNJ$GN_Hm{j~YU{fXS_~cx5`s|e2w5%jaVsD9CXFWjff!COD
zV8`$aa+|sD`vp+s@NUrF3R>Z2z%2C=22s|y)yciJWS<=agZAwDh<ySPs;Jy?Tq@Bv
z)nk>(uIU{9pq7LwiW?jneB=!a?C8!4=qM@C)-ejuh)GUNnQshp@!1vUZLbU64m~~Y
zXhj7ZZf$MNq8GA~1H2ZNLVDxEpg4B=bxk1lg%j<_lR@D9hYxpZCOykc{7!w~UJHBy
zTOG2GAAZ`bU^}|Je0QYh8h6PO*Vf?}*k-1b=tr|c%D6w)e`%CmIi+{(_O{GKUf3d8
zl4XR*Kn%~p?h5uHhX%`C+rs3np<`TODV$@aOb2GEQx|C2z-MYRy3<u)TC4#-h#M&=
zn0jBf-Lz#zZW5vtpYl+rsACdQW;>7(mXX8~dtDs;z~P7PQotv7qtgy$Pxk>O*p<vA
z&(W%ry)m~`FtkqJ*BQ$fb~3=y8k&-wM7Ui<O@0MnEsE)WwMiEKiM@9nt-}{OrJRW7
zGq~eIAm!E<&;x7RTlJcoGcZf(l{vn%5-tAlwg|S+0fvz|ProuLtc)y|zMI|DlmcAi
zb$Yh|D#LyOt-qaQRaV(dV12Ov!k$J1@MV(H<-i8i6xiBz*7wRLQ=T8(Td`Mq7h9~U
zsj88ZJHVfuU|_~OL`Mfd*fp4KX`}4+Iegv}kD{y4DZaclpYGQ2J&i@hH1pUj_xS|}
z_6wXqV0j+ReW0jDCG~i%&VcW{XAU0Jdoy6`mFzdm(;3>ehX94ay=E9I&^98O;l~O@
z<<p*l$-;9XKM+29@z(`sOUt0|P9=z!Hrsd8LJyuER5U)gjXw;ySyuFjBAkeX;;s48
zDC*=gv&}py%Qui4id|TE`rMce2wYJ+Q1>7UpDbe_wAdcLNKoO}JI!En3w1EZ&!zx!
zFc02TOxqpeX(JR_A6d^Zj|2RB%5p3TuJQsHq8$tvnsD83oX@tnbFQQ~D-i)6sn=!)
z>VLi-i@6x?k$?F(jx5|#=f$o(p~b+HfWv^TZ>>~aQ=cMjzrKdeEfr^{z~S%;lK?F%
zvg!li+O?0)Ih&J1W!_P%x*fQFwM$fMsgUjY$Df*|du~ll<}0ZMrc<D6EHj-mc9Y-H
zM`eBK6M;2l`1R8+J=l<{BtVxexfE3A$l2r8l9{_wIulmMdJfdMRFp^jt_wZxNf$Gc
z*jeLE6JIUZIq(pNTQ@j^s^v&b{Iv|L!d>Xs!!I(Z81O2)ra}`Zkr1!UJEtbcLeE!6
zdZ~M2kq&JDvb!fG`+Vj0bwRsWXKb^=S<VEiJ>4C=obpl#_%QQDjkI%i{c<8ryJp96
znECXRo7V#@S(b(c4KOxE9h>>Ka4=5GsLmqwT}HXX7IY^%^Kzz_O@}P9GtX#@U65P7
zO=F2uLD?rYW1Oc13Qfo-)#!{$Cbbz@CX(PrdT`roxJJd+e}wJa?0YD&JX$wcO<U|7
zy`TNor)#gHIrBoC-4EZTeht_9uxAB}U{s)6ruW*gMpWF=QgAd9D$T|{zy9t-1O4Fa
z=h=@Y&1PC=?&P_nbtF`B+`{UKSv%5mrfG%^<%Q%qb4o7<RhlblF0-DJ`uU!jel2kl
zZ(M&~Zd!yL=vHP@FI~!^amA0*_+C*=8J{Q4hr(5whvCmZuw8R(irSvs3{@}KZ&dew
zzi|pG#MW+Vdp63=OP`l#NO{&i+~j?3;(q8i_iQ~_v$U(rcDjteE0;GtW2C^|w&R`T
zdNTA#S!~?8w-jc(B#F7iB2b20IAr*y=r_)rIGe;S;W0bdLG!?n)7*P`#2k)UP%+ZV
zrCph12ho2|ojKl9{dj3r$|C1U)3FS%4hSuF%x$#))jrPdRt;{@sA%Gx$Z@1DX2rD1
zIHhs-XvuT5_Uo&f?!E=vKtq<otwAoMolZ4R2V6>p8B0&h%KE35>!u|$XM9NoHkos@
zZ5}`S9;WU8^y;Y^im#5Pu)kpkqh`%aW}HXuJfmGfGjFNx#WG6mT3z?qML6AqFVCdG
zeIA~c8>95%Ke)W#EnG2a)+JfOCPNjX9z&1xmKOL&Yczd%&GUM<cI`7)Ge?@z&sO_T
zTp>q~p=N?y>_)HOFeiO(3bNTB;htUOYtU@X?gme^_)P{Bn|RF8dFCxX*H{VCMmkS0
zMqT4qp^ea|16HHpb0mZ;a&+v4p4#mq-Z#abuGsCekczk@vb#0kOD-(L4i}#7-&Bb1
zsAY0XK}smf3~H5iJ#g-ST4(0A__T|0l*pm4fJ<WWfKf#SX6}1y<F1UaE~yX8$o_TA
z&I1ACcqma%(#4u_4m4&JGiaG@dD!6AQIeM@7|N1=F<iv(U<4JYE8)9**K7W}PYUJ7
z(sIw;(>FqQiAh>?3iaT&-M6oMSn_jvUtqZqtu~uKqTf;3`t^1f{U`?E+3!ke5Irj?
zdO7tnV+P}pRzFw3{tG%vsI7TsXrx4!pKHAwX70u1rwCymxbxa&BZ;7($c;s+a2GGk
zV<VM9o0=|%jg%liS_j~Je;9Ut2UUZv72&eR)khYe>mL*4To6yMPRLB>l0FhT%fdJ>
zD^?*p;Np^BI%j3c!f!V-YE5pNk%xCkv<`$8?A;Lgelhfu@?6OSBj~ipx`fMDs!Qa=
zEpCj&JCgS;#)+@{4RrTllMvY<<q={#6!RjCQ~V!1QoUIod_N?pnEJX<_?W1WIl?jk
zixKXDn=l{mfOR>MY$HmUx2$jXpEVP+?1|b=eTms|>qIbWphA?0n|q*@K4(%!$8+|U
z{596SBaQ<_oo_;SSXKH34wt!M9qrknjQg)7pl9zltPBVt_ka(vh#uxXa~;zjrrGg#
zd)$4l@h)ogDhX#>tAchJ-b_M<_YfsPpSbxu6I{kLBIoy)A-IK!;%R=YcstZf7qseB
zTFO=f5?zwCE4Q&(Sj3nh;70>gwlS0LpKw#??=>n%at`?rpS_z62GF-Oq*SV<XQeAL
zHXpr*_K~fw%|*imD}4=bp41(i72ohZjZ7DH#7&Eo{J`1zt?)?iOlOTWMSEp3AC+z4
z_Oa44>MZI&#sgdU$w51n>W4GS00sM)5CN|u9O<KxtX13P<Vbd2UOoB{grct-ulwho
zuLe{DQv>?LV71}puR)Bbfs_>Cd%b%&WT`ew3im!(ZA8VifzRlOa*5}3H*62<=TtVr
z=>|#gH2uMLKt>VXfV`26Wc*yt-eX)B&wMXsMp>1OmD{wuTo`lp+6{pB%Y=Kc80U#m
z3kEq|u`;2x*od@daOr0nQ^45OBw&nPgpd96%S~#JcNR8D<;X*aRcq{)Zo0UXC--9#
z9di5PXOz`!(JRb9_+)k3o-A{xxePF;@?$||>sl;735jx&C5=#Hr2DiT^psWfZ+X9G
z$s&u9c6MS_0kQ+^emsjJLwB#ZFt``&E-zoCpJ7lA>Q!a}B4y1AOk2+A<;cl2xZj?w
zx3ukg8iGbvj-7hn$>g_-$;R`3<Ox3fs)T!^^y}MdcX|nX`q}-9CiaPn7rvg6@c19H
zm)zxb+;LeM3=4nKgfr`K7}4;B7(!*gVSUfci1Ei6G(k#6YCmNtvB`nFU`0fVH7N5F
zBTc!UU+;1<Wn^8}K-h{uXRPrA`pJ9~Zur|~6XtbU7!wAEEi`LY#@~1$i;VTdP7<Iy
z0<tUi)sVG~MOM^R+~xQk($H8pEk2ePa%Gpo=Qv|*&#>8kuJ9If^N234QKB$R2-Rby
zu^UE*(?M{be))x{D$>LEU|8yQ^=^;raQC)d7e-UoaU<qC;hbVk?^_WJZ=gJgZUME^
z_|Ac*@w{T~29B^Y+QUb&kSAzC9bIKNw8VrtJAR(-mORO-BwqadQuy8na<^tk^YKI3
zUikTb3KWFMeJBQZcbw4JV+^{>8$7qJvEj0e2W<|{;&1Ar6SqXM!Bt2Hy)MD9(=r~j
zMzPnh{05y!B8#}FBG@E|ABURslNc<hQAzHCqU89ljp#&bLxMM<zUyE)ip$awmhHs>
zr5Y?*YKRxkFQ5+%jW$<dt~mX*;fAIglR@(LI&loK>gwt#OCuBmYs1YN(mdG>I`qMn
zg%5GY*iV_W%VSDv`o8ss-~t~52D-sF#TO~U%c^%hdX$+ma~9Vpk#N3);Ftlc)Qm(n
z26N%Tr4KR%w*KeqU4_-PGJPZ@a>wV8xsCi012v<~8dGc&sqb5`U989FkK0&t)4B3Z
zhUJ$cKz*6|dQ-jlS^$is1tKwMv#TtqNkv`hy7)<_SX`QBIImi-;h90$mSyB^Hhhdj
z<#S=SL{-mk8oFsD54fBs-kAJRQ6^CV#zy_*3BkpY^d7gM9{SpnCk~&w^@f~V#!(MH
zbD$*e2^+Bw2<WJ3x>hGa?AR&2FZpm#9-Am?4i@^76H>GGKqYm|5sOHtbz+S^rQ{jl
zn{bcs$&~EMM}_JhIn``&tZD_~Hy@oHN!S$r4Dbfy`laepz_5VVW!t=zrOPrVOr!2N
zbg3NGx6<|jhVq9rh)SX&pA6mS@vSO9-|{j#*E<dL_03mrSdX?soj6mxiMx1_!cYHe
zM3}(J$nXeDBj;GE!>uC@mr<fY<Fyq#Bu5w%i#2ML@C}rRv}3km%|qViBVc&0cE(qk
z0x(5dn-b;vt}c$&WIq`oRY=dsAjxJOG`Rl*cn?L1tv8&@)SbY)3Tn6S?9&4P%j&OM
zfLg<>lNlD<nWzKaJjNS<pR2?@7F6Z3x4?FoK8~bmZOucYED!NL<Y<1GB)7+G<QCu*
z0x_*OINO|#8i0W9>tFEtqp!(;Av}F4pZCp=M8*$$$thk5V^$^981&3)f??Z38>B5f
zl(t^7!}op2eI<%1nJ09b{kF*#j<*jb!I>V>zOk9(nta18j@ATNa(`G0;J_Q4y|^$A
z%@!3{G2TC54_x~gb0h`{ALebg)>&i^G7bWq<AWw+2r3@6bMqe@dr9cmPQz~+)vXFY
z884oA8{)Gvs+9)otxY265~l{3WG>X>+;dcpcr?1cz3bk#@A{xA<GR6vtei;N4R8yy
zzO=ud6o9#PMeGKr0ANg>eJVX#7REs6W_WxS+jPcJ95cd5nbNrVdIf~);d*&tcborM
z<8)N(R1!WLXOLO>Ebg)kyVQDKM}6xcrN;W;&k^&>r}bB;>TaAKTjCnb1Wtm*-KWC&
zNVlKg7)0gY;8n=cyC8S`!sb)=Grf8Zm+XSbyaupe3Gay`c{qk&9PC<Cbb4ael%X$k
zu0Ek=@G9HCJu|T==}|ez=WE0I^9&1Wu|711cX`HPXb5*MCCRXPOg^mX;kgkh5f0{d
zZZXvYOOzHWnhZi2hLc|o_icUCxa@TO!Rpl#9xjj24(72Y$yw3+YVCK;g3qZkp%&=}
z@Yz+-4=W_`LV%qkcvds)h$c<I@)<2Xedo*@>D!XNYcIz#sjmSzP=^SZ+h``DM<i_q
z)6k}x9l@Ll9o#oF39aAlHXm)Nu%0LsG&j3_>`NE5YeYh^ng?+biMTtIwzxS&ipv~s
zo3ifPaP^#nux5@JsbQVLO9KM<lqk(*>>E!eF?=uGLTuxN&vSP*KeNlQVxoEaWk{UO
zmxD>qw0KP(Za%j@I_7Ht?@P==^~Jjp(V4O~HdJ*#Q>YX3cvHzy;~Z+<QWhUSO+%ys
zrI23kMh2oB^(Ze6dxsn{sg=VA%cWr3jp~?PBdwvkcdr0Ull>6{u*H_pw6ZV|AyJm#
zxH4$+hwv>7A17G$yx^f61x2_<{%3N`$19CxF_RS`%5QuELlS*nKC#GuE_~I{IzpJy
z5SDT(U165{G<u*S2aGT_8eV+=Rb9mHe(c5D5xxHD<MzD=nUc7$98b39Um$ZK$ae{b
zVaVX{PVcbgbsWw^hMgMQtFa%`%@J1|3xzLPm@u{0Eay{YthGeYaCV5DvfBxk^;A1G
zod#z4!hCPz-cubH+MzmE>U&!C8VpFxrA*tyb(dRrg%?61sle!lr<{H}*M*C;fa*Rc
z+qL^E!3J2VbS(!{a*>N1gVc_Iw;gbPV*kKy$HBfVr~PKo6JJs4$EI{*-x)|r<jA$N
z8|&KAPWry-f|{DW@_mT*J=;Dv!KT;bp4O66NQzjG7;p9nfk)48Mnrr+Lu2)rmZzOe
z+eMef9x|ENx|~`TzvwZZ)*zK*wmv8%Y1!poZm1xo7_bOvcUZ`l4Rk3KjOw3km5Ek=
zDz2x`zAl4$62E*+Q4*hh#O0rCAYhsIUwl%Ag&u0eeDqPy=Bhw010VbPC%5O<5WxG=
z7$oUgh0DRrtqydG%Z~8KP_N<XMhH4j)`}=dATRC0jT7l7v-iXIjLKvDM%hz4Uvfiz
z8Hvh|hp^8Wt`%{sb!G7VoU#;)AlTSILV9Y|(hqM0k2=xlc;Cpq(A?5S3X1p~I=kS#
zLOV~RQ>NWc;`TvMt#z^3K2y(`p}P;2L?v@zohIa%m?7rz*=Q~zt;NzS=<EeV5`Rlp
zTy>uB)6N%)6JB8OumE=4V(d&?=*N|CFzbd0eW$zQc0`T}C(qy|2Rewd(fUM}Cod-s
ze09d&Qx|ic=c)9ZyNNbT?0izTND6Z&&+!3iHZ2g?FOi70r`_c9&3s9e6EA#{#z!FN
zyJBd+l!~{TiSLq3%ccZgbLwXs(FICH60WXCLknH+xk57KcBXOMw)-hj$HgR?gojTX
zl(<pXXZ(tKts}l;*}>+~Xu*-LWp)wE!c8M2&v*5Dln0cr%8Eu^CeXB@3?Zd{q2)|L
z5_hx1I%7S<XJ~EEi;&3Tkc%>$pC^fxl#x^}eSEV-!~Y_}&uw9?ZtIk^O|mLZ6ehmt
z%^w|T3ceiPv&H;tcS0F1;{C<$9f{`CD*jaUyVt_Dd5<E6^@bO(*^`r>jr0?a9%LbH
z8;h2oGNo%Z>DU|ahrycdzXe(pO_WKi)OXM0H`})TV1CJ@Gr~75z_n2_>JvWZz`Y-Z
zKHg;-nct>)AltqnI7plU$}pPS2&PvX9d;7v?&qVkVDFUQsu{YK^CUw|>-D9W*;les
zx5tn!RQs#`ne7+I!ly~(R_QC(sE=(|H~_vrJmk1h66KuT88ou)CvKTSl;j;|DKY}9
zBh#{{1G2BWiDej2mnVqj(zTpvI6vdJjv5<DqRXI#9$&*-O!?MZMds4FC`J`1VUDa#
zRt(^PJ4HKN5LzrU`Vv~a+Yh!PY4KnL+}MGv)wc6_RPl28@~fJ`^=@bD&n_<$d=`$P
z3DJ-0S`g=VE{9W{^j!?Zo8T!}HtLYN=Kzj8Vm)7w9WcwPw4Qw*DubItS1H?H^QtDp
z-?UsK=vyLecF&7#nm*Pra2+*mA>fVa9=fo?QsWS#+dFcf^-Q=h?@&aE5zz5jO1&<)
z_O1-ptg7Mo?X=7M><+iVa{0U%GqoQ9?mL>Ofe9LSZ4@&n0+<PyHQ2aaTjSxPb!DjV
z(HoNpwO=NTFS_uSZ|W}U`S67sbJ8rXK(&4A#=7b4)*^Fgt0E&GG4lymo+(M&5@A-E
zb-JvJnB}L*_O;@QhVX?&`MzvXGcpMS&R(>9pE$FGh%uSOK^0pdL&bK$38V;%`g(oL
zMa^~9>=3cHH@?WT$kMoHu0tS@$Y+s$DA~P4xVFfX;u2#Qy(}L;uTe2)?tJ|k=&~Iw
zJ~!-SvwiMcLVmwVqL(zj;)#M2SpE<uvewgvMT<4Mb3_;#tkIipPNW#S+bxL<uw6O7
zwQ6jt*4f6pHXxjxC+(c~Qu?v82X0l3L6WM{-S<kR>^1E1n%`Vj(7H#xl)3+OK$<Xj
z<EO9vDd=<3vT1ED%O`}D9n`n&mhf7I+&~UQsS{SRTYXL=PE5(lYp`EQ49Ir%ObL3W
zNNzEjB-cDzDw*v}C7$uAz7*1hsfX4q#<L+^EyN&s-C8xdP|FD|Q*Wc!))FzoH$mEp
zA-%Y02Y~_w(#77OCG?RnhVU{7<<dy<G8A*_)N~<2G9p75Xg+YxA0m@SUw{%_TVi{w
z5ppb{@YFfDiuVn|X+9{9o@323E6$*y*Jwe=pY!cgg1$7i+}*$;mZ8ccnZxYc3jB3X
zl*~Mmb#YHj(kUF02q$jPw-haZc?c5!T<<Bm*h#^#v~dU2&hNj9Dk8KfNRo4Q&WWS0
z?qJ>aPH-F>%n^T+g?Vv`3p+%3<1l)|P$TZJz-3MCQfo$W<0yQ+*muAI%%wiGOR$;>
zscOrfbp9lWWaJ~#+3tQI5Tikrx7Yi{DIdfojG&0!28JNnA52K*pHcU;ex>)GX!0zq
z8c^QQe&CAw&|^ruYE^O0B<8`Mfs4H%tR1@9i+YQ6shXX6Ba<@X?Wq>WCw%1%E8Ny1
ze(`~TL^HpIi!3)-h@cPsP$4qWMZq3?(zLHVgRY6_Y8ib_2O?>c9TTM|5E!?$kQw%P
zIVKuqM$%UQ(yi97kuNlpyP<|TiOSpv?eHqof3V6Y-!!npyUx0eqNCrVVwi4sXQ206
zTU@27=ST6i!y`yLG%;7g=+@wO<LxT@*I`Q~Rd@)E5p^^iyzs2cJ32d{_uS2}35XyN
z$gs62m-K>0sexFB!ieMDCf#Vpy?;bPGNxo#LU*FM(r<YnfyF1T?#PrbCr>NYa+G-l
zBumEh%9y@d#hR{YH065bE5U^&`h2>t>~OzR#^T-4wl%(hffQdOW>2v6NLP!lzR-Iq
z2Q2TH{6_F;cHBTUvx-P}Ce7ZR-s#{rt$d}W?bC(C!U&nk-hKDZc5oj45;p2OkO@r%
zx%%yq`;z={679tFelR1$3^P{0$cec;44bfvhMe2u0F1@5Q&U=M)3$G9@`TG+&c&mf
zO;I?ETcA+gmFiW6k~!p@?J#awCVl?<W%2wjc^lYvv|7a41xIU=bcq$8;xHAAuVTHv
z)dr)bdL|gua(dLSTxgC8g}a^x1}tl8ih~K;X2=j%s=M<hHg;>srAPRQLz)G>hJB^p
zLmJ%6J$<Jg3_m*%<{%c)fguXYyY9_QNZRht{OM+7M~`yXK;^eLM{|!diL9uB)%u_s
zmXTK++?lI$C(2XueahbE)?i9r@y%71O}S(Wkc@qsVI}0qx!(z--<Hz<YJWOb0gj%p
z$5x-emkvTo?VoHzYnZ)t_U3CBAah|Y&tKJt%&JRV7K3Q5e_E3LS)Y~uqNwVEoITe^
zZXd3XG5j9-G=^JkILlo^q0x(vB44eNP}lZ(;YX%-kVlxHiN!&jeH^)H!foX73rOy=
zEYB^C`fVBeZrz1rhH3`)ijb_+kHc$L=>ql;SEQA^m!D;n4i@?I^NX6vwd!Ha>@|=R
z`yi&&zCHydIrh0c%;3#Y6d?#xcx$%g6Zyk2o@vE;CG#n5z6kP%(Z)_~c7&j^^0YBz
zQP*UE!XAj+?xm3fGRlR)^s*glP3WrANaB<{#_#754w^LP)-Z%7m7XL&jn?sN<dDGQ
z$$Nsn_YKQ?9>|SipNLT+E7;o4%CN|_&sreXRRq(J-2q17M`)Zl)m76+^`9aionJF_
zuX<~5*}FF}US!IpSpV2THIQlIwPU$Ai`486$DF#$g#{^EZ%T!9I%lf%$aE12QG{u4
z{1ekYgxC35J!1T@{<%7Y=hCQ&%;3mZS8vc$?*l)W`&pGKc)v9HNZcuN;c;@j8k}xc
z!H0Y-YT9qYr#N{1HFv>d5ZhS}Q<U2nW(UgJA7aXs#e4%;MPa_1roRANpFHp}q62M=
zrkGc@ruZFB%_VD>2OW~I^b^Z9=&utn`*fse^<3qr7ZQcYEb>%1k5&7uP4=MKzaV|8
zcro<2`XIIQ%I>8z1gkzlAJheW9STB@q@7qzyM1{+*(32h7g<0qqSB8Z(1G2!GgGwZ
z*A|UlsJtJK6gIo2_EH$6H)d=jI}our(-tQk(!Dz7C#I%7y0UF}QwzIoEM{Mx;!BNT
zu>igFTQ?4J9poPc_ePr4Z)@ruaxIYlL=)y*P(81AN2b=kh($<b^&%ssbR6&pEql#-
zu^0$0^5*DM)W?b!w}RrN$$1|01g@^fag<y@ilO&E1<E0L&F&U5y2E#?(9mncyXo)V
ztvIXsA_0e9SchOWxlWMuO8bg$QwqW&Q^I4usVlvqdD|ytRoAq>E2?GG4#3%m+?Gt`
z_Zq6YdN4D&ImGlu|8>Q#nZha-47+$ad~ZE|S)x=VeYEYFvBJaiiwqLrUJ*poN7>pp
zKSIJ?vhBz0Tx*aP82Tc_ke)_}F@ul6X~T~?Z1@0quqd4>k&8<jEL(a<$38iFB4}^%
zbdPxE{KIjVUp!11@WwCS+!4POvFpW251pAmJ&m3?Xh50WHpd;je~Mc}D|o-;<F~Zz
zGA4dTl|Jqv526gBV1lB4ytak<Zqlw{u~W9bltBWq%}4bdUDL9kLe=@;!XJC!a#gE@
z0-N#Id-C!=DR1_(_S`8%O3Hn-u7(mFY-fW-AKFDVp0zPO*lvCE0l}1462v9|=o<`m
zX02Z6@s|Qki5*nzo;y>F^>i!CN2NLt<#dzWl&Ft3fhIk~*c6ByPE-tcgay}De!41H
zNfu7(O=ZEYth}0a*=H1b2kDX(CfQIF%iCxM0aHiS$G<?=WiR%yZ@X-+j!JiwXF+z?
zm1U1|XBQSE-Kx#kU$uK{VOqPT=D7}Zghr%gSAKNa)~=0OguRWV0u>L?d)Kg7iXiu&
zSOcpj#|e&tKD-q#B;>ovcJWy`v6bvK$>mAZ@D@w3VV~jo&;s0Jul#MqkI&_+u9h<s
z&!~bX(FdYqoj+2P#X&%|z?l84gFOrA_~DIULvW6|bH&BWl~Uz%z4hAB8mPgvS%%=}
zRX?8%;rf;&QZ_ovGFa5?Di;w3LhnVTpWGGzFpLOf{>#HC3FrG^AS(`kuIoXNQ(pM}
z*x?H<>mfz0&Gs>7A@|E6QzTT3Y;UBtUd|%j_*rl_Kf_SauHG+>{+c|vbZ{8A+%L;B
zi!^;I$639dbQT%HvJmqgFpnX08uaPK(#x|QufCAxSxH5?Ok2;yu2h)M?p2Hm;J6p2
z#gCS82t}BX4X|U7jlbtL_nTgXP-!tK&&rnbwENf|FduObbG#Bl0NxvM7GQ(8!A;e@
zx|Q5T*&cep%O|AN{A8OybM@J9xfyHNFuH}kUPq-Iztu1_GMiaRNr2|se=i9Ak?Vyt
zWGGtwW(odjRI^gqs_v(bKl(-;I*HM6DB<7&3QFFx%)}6xJ+NBF>RF{i`xsi^Jwlm>
zaLUdl!lZyV4!NU_4>}nL(j=dhEssoMYnrDb<rtP>5kGwmCUXC2b|TyNBqr&LKV^T?
z0YngvC)fC?q<7xBs`&`;@rHN%^p1JnRwy#ziS)^}B{xIr*(A6^5MEzBT`G-odCwIx
zLq6lG*;kwi`Z{e!B;9`6EQs6i8bq|?0|{i^pdzT4#(yhl=f};sBbJNpJIX(mCMrJ<
zL%`Ka7A;JpR{O)0F1)(FRK}WLbGy?sRosRHMVESl3eu5cnZiJ**MN7QX)0gsNzf0y
zAd&K8bl4Ye0CQ}#QT5jOp)@S0cc4jN4~yRt?MZl5t{H-0z0Zc9p>Rb+1jk1tvQ$Jw
zIYvfr`lq;b3ajGds$D&*5ccbK6K^f{N4heHn~E+6xsZmO6G-g!$wbEq?z8OGd<(T%
z=}Y<A%2<@{lt9sljB99`j>63jgWQs>LQ>9`G{auwM6%UGpPngHUBchRkCq#o@Dzk!
zz;HtkpyMGU6Fr^+VQ1UsHmjJnh72K5b9bORli&F;X*iI0jo6A{qsnjiB_bC2uJ!D$
zsl9cgD@t#^I~`^XWDj&mQPv28xv~*;vdR|ZIlKhcprlN$S>gTTngg68P?ggAs%B#P
z#Q*X*98%JI^9GQ)dg#poXNZqymiX3+Il|v|64`G4{JK?xIyGNF_w>W@g$=xAGQH^H
zr5}Y=Qd=9WKVV_9X@d5yCTC(a0F%C-8dbP-(B9i<mQX|IpZ(_9hb%_*-j0UHX<1e2
zpB3Dm0I@E%VjzE~n#Q;pq{nk4L3muN<iPY0KAsPC@ipkKHJpb|xWb==^_?3H!~AN5
zG@KE2fnIo~M7kP0*}UDDfWKkqYTf?iae)b`!1-ZGvDH~0o@<o_jH)$UbfK7fdt1Tg
z)3X;_<6gaAZFyT#JV7<j0f&i6kw;(avofWrTcw->7{>Dbw#UP2vnH74L1MOWx|_}q
zUQ!akg^hbfcG}7aDw^UCkLUP9Z!1JoCikKa<jn%+lHd0rsYc$92%`XkibHkKDxC|s
z7Jo;mVqhxd9Vg>H0ySkAaOEJh=%Nbmp^8!COK4cH13BMbwCzgi+U}L+f1&sSvL_QW
z8~IBf8buCVeT0bM*s4d-*W(GZE;Bc45|=sB^tVY_$y|{>JVmSB;I1Iv5F90^dzOpc
zI|CvM15<4~%vT0Up6&Ub<W>UkLnW)P;*)gvTEiOGyJ(|drjZ3OoHy<V*P4XQP{RHD
z@1&tD2xiXi_&=zA4qW@x9*^cij>qv@xx^Te-jqz?^Ux0`v**di_x-)EwqgO_Lsr?<
zOSDk@41I`O+)-4GP)B?cwH+$?jK~6^p@eDH9l1Yvp|jP<2Cb?9I8RDUJFVeQAJ)3X
z{_<-cJK*1tH(+`3yQC6d1wD?h44O#yAo)`b`L+*VLH!&mbLlVJs+QpvP%k<|@w?Xe
zirlSSe5V&pLuhaQQ(28~-1+fA)vfLNAHI#8Jib$!`lGv2e|nuV!*hWqV+DT~jpA!;
zpVQw0IJ|eY{QvaY(u61AG+np<OSiemD+fGTT+ZM4(`$GTzP2++u;(x1=&Erz@V;+I
zdHOHwX`hf-Zr?8BFU^Oo8oxcvEqn9%Z>8gZ74lz&{MSSNKVC8s?@3xLTcZk#?R=o#
z<~;w7CRbi2%RM2Q8m+V*C{6Mlc9@QCTPAWh{S%)60t3QyA>gu?mr`On!x&gOHvKH>
zEPHG`tNkz66VJN5vr$CSGB#m?{x{2W=6~`xor#6MOV`?c%|O-F)&B==^o|P6cq27T
z@ei%>ii0bj<ocnY04K_q?KRq+Y)UHEaPV?7^+Y$gxp?A#e$_6bpdfmfFZ(ohtW1d(
zDAvr*GS>aD^IkKUsv$(d>~Ac;b)`uzk?r(8|JXi1z6z<tBM9-5FsG?wv0q1WsHo_q
z>|oZw75i@FZ@eUXbo(t^*yECt({;VFd6t;jd(@<IK~+a7a)%K}VwslrEsqcWMM?Vw
z1guk_`J{pcNt-UOgzA$+ELSP8i2C;JVk%`~Jw~lsE=e(XrdqF}j7=pg{Y~qC&{dxi
z2WuCksdD<WK^|Yl%ByIkHQAxdA^j3~9&s%4%71WwugKyd?4&$^^N(d(Xz)>AQ<vz)
zUtsvjiKn7w3$H)_L-+KT$HUJnr<^N)ute?z52u?|^O=7Di9i!iMU~c~q$T_VYA>_#
z*bzG1Nb?uNF8<;clt#$^g{V#Zu~u8}2S52gHj}!Ir=mU*eoFR7wp*ciG(9ld*7!4U
zIOi`M|44G_FGhXQ#8Xja!IKyN47~9ikAyTsa<~7;7IE;4ifZ32`Ipp?bp@YS<T2hQ
i{v+G}Q%X5Fzj6GX@NNZs9seHz02M`zM^Jf-Xa5V;XD-12

diff --git a/docs/blog/.authors.yml b/docs/blog/.authors.yml
new file mode 100644
index 000000000..c3fe2ac10
--- /dev/null
+++ b/docs/blog/.authors.yml
@@ -0,0 +1,5 @@
+authors:
+  remilouf:
+    name: Remi Louf
+    description: author
+    avatar: https://avatars.githubusercontent.com/u/3885044?v=4
diff --git a/docs/blog/assets/4000_stars.png b/docs/blog/assets/4000_stars.png
new file mode 100644
index 0000000000000000000000000000000000000000..f3a596d7e13bd919064b87b463d2a4fc1fb5dc5f
GIT binary patch
literal 57307
zcmXtgV{~Lq*LG~%=r|MGwkOHNwrwX9+qN;WZQFJ-vF$H&Ki~VKd#yg}be&y$S6y3I
zoiGJCad;RU7$6`Zcu5HnMIay$Cm<kT1Sp8_D?l9K58vO=b`t82KtKrn|31LOwxvej
z7qOj0)tr=UO`Kfy9gKloU0vzSZ7dxP_3e!5Y#mH9E_rZ(fQW!3MFf@IGS9PI-S8Gy
zo3EJH7d#t{#+$TySCoIt_S^#j%Yfts3J5_F6Xt=Vphytiqnt{^_X@ayk`muz8gdom
z*2vm}Kjm+oEu|g38M>%e(5=U-G);SJ$E(ycAMr5PKfd2L6o`OLh6)3R2`CaFQ98A6
z4hkSEn5nDq#TF!vK`<pMmI2~HNP!7@;5qsXp^!iV{0Pk;HT+F-e~O&?Y4kIc`@!VG
zt$-;1QdVxj#K7?AH#B3xi4rTo?xmzX-jicefRllEtf8?i*T+Kj10at=qU`A<V3vZ6
zB`R9_{pykG-u}uO-d_!;g=78qx7kRfhFn}7GzNl|?Yci{Qe_`CL0TA1rq+e~kf%wC
z6*S`-7Fa_^J(xY@4HWVPvt}&mEqUOZGkXw2z#*E+tm0J83<^*j`<6?&C?HMLMSN_s
zcitiWwMd#f;Cq(I-w!ZjX-CwUfu}34tfb}V<aB4V>j<3yma64-Y=_0DQ4ztw31%cz
z16#=Db^A{sIPCAc5y+B_KLn^)F;k(92G(<tML=st>i%HjBMgJ+S$4qB{hPCisxWNZ
zg)p#tvT>FH){r@X3ALU2*F5`051?qF+dsumqv1&IKZ)_hVB4%o;5l%f(?aoNy{Hvj
z(btj`ZGq%LuJoFfz9syAa6P`6;4(i<s)_m%r&QUz|LhH$8KzNz;;C+r(h9BOeTd?J
zLs<*Jwr$yi41;2ZkLa(+da?>f$WuMAT_atS>kX|E{L`qJ3rs`qnH+>&ai~_nCW(z}
zxdJ{FRE-vuGygT=c@ph^=9;|^z;MDx13C&D3F;|z!z~D_Q>S_$x(0K3L;Y_M7VVh;
z1P7Y&HW{i1X(IluAv2F6Evg3w7mg;r?;rmLVUAzP0@qG9R`32Pr44s5w2n3DSEekr
z3HuIj7T(PGKl6gQhzq8Gz2iM6gdWeFPK{#}Ej;uv`?3H3F;L}QC<IM`D1ev4Mx2~#
z`>XsYi;_W>L11St<o;7jMqN8pTN%Is(vD^)w1O--=nyfN$0kvKU@=iU6zMu#@ie9<
z{agNjTAZ}2ulNz}G+t+e5jWU_82UR^iBT8I%InpfkN;kQjqetzJtUa`BeN$)gmkY`
zI;@`ee-{7IUY^VBXDDhEr+8{{-qG3u@FN0iO3v%x(_m8#H0<KK<qj<j9eDuULfe78
z?pb^bX2Pzo3fKA<s5yLu(~o;=$Z62Zy6EPtT4LQ2qN;(#X>B=nS5;~f^b?NdTQC1<
zG&o*cat6D{9&k&ffMRR-DwY?tah;o}0r`&O2I)QfZ#&dS)Kvs(G1HE}5+b{YA=CdA
zI1Aqu7LgJ);<JN{#jy4K&vw`_F5rlpz<^s<y(>3dCWQuH8#|;XPRUEyqJ@{9Wnul8
z|CarYf$8H6r$4Odfwl{P5mIB~f-(*SwWjG|22{j|*pA@FM)!YOBdV+SIIg1|Z_}l6
z=HhOq#7FVC)3``<<$?fiQdW*nOcikPz@cm>j$CueDJe?yKxx6#X2|=u6}A1ce!|Iu
zZvB+X8l5u#LYOV9#@9BE6awf{?~R^5<5+<j9APCR5QETnJo2e-J`xEc$6jPFBE>n1
z9$4$m<*t?fGn*{lN*P$3c{)E@oU5pjSci2OWbK}4-}W!%a%jDL2H^2%v2dqP!0j3c
zV=q%LoFL`O%Q)PsZnF7R1UjEs?1)-&E{$N=2YPl*0_*P!RArdI1g;!8H0t9N(yDO5
z!5j2=0lYRL;{R=0^5}CKVpUALpwJB@pU(A}1e25N@URPV;q%-AHNA8R2ERi~iiI_y
z(n|FCvq2cJI0d9y8ci@91kV%3egRH61u>4mRcE=d=olsi1jw;N2zijILZeK{(b7PX
z#ChmoL2)P+m)zdFEp>pF;}ySJlLP!TzXNJjuR0$s5x|!LZXfvEc;G)Xu9jil1XShc
z1E~1<zkG@ordLOK!q=s2_bS&hfCu_+-Jx08k;^NIm38HFD+dvRxU^g1mFx<O_M~L{
zC8UNVBnEbG{raKBI84^6cTTTch;s)(m;=%84Z+`2w_JoSF4T1BQDAlzU=Ho~WhDFK
zvi+Gk(HZe5SzDT|FOnUi-&P6TkBi4SEFQallVQ>#yRR@rZUcAqV*d|mcJ&Kk+q95h
zAnm+5;Ia7tPb8Vggt7t5jJRj!+a2J#?$}-F7Fxl$rY_`F0YWG+Fm}a7`lRIfcnE+n
z8<Zt$K(6viK?&)e<C!X47!Y+P8KyndP-?k>5AwXoyAKGhY#<J9?V2WXkAy5SBRip~
z284kVE!{;;zhVFi7mrt`Uf3Y`@`m)UUnJ8X7hGBhhBYEtK=e%w4VgK~$twID!vD6R
z81sFkk)SF7Fb(=PxE`A~6n@Yr8PbA0E~KGprrnIO9SrX_K4wo!9<Qb-)R#I-ih*7b
zAK#a!xDHy}^lMS#UK5hH4Vq-D<CD94RXBJU4S0OO!g5emDQI;g(1FX&#CU)7hd?Wj
zpAgkx4_Ptq@UV1><C3xK?K6J&;dR<PW7EsQ+ksNiLYcse9zywlAf&*k2)?O|?2es^
zWh)HvyGl{GtqTVB2Xq=(`496i6VA#CRb-2&I2IqCFeUPMCLa(n3z+=KXyo?d@+cxe
zuZCbB0$&iNze1><Hm!df4h=Gox_XP_?N2#OYP_GAi>gQ~p9D-n1yfz|@%aFP?hnPx
z-wD!~<5EQnl09v`1Zw~JiCz;@n6{6;?A9zj=<2SM4UNJI*B^LtjWq8`=qxlEBI7C2
zy#a|Q%p)tl{3t__O|VE1%Wn%Mm9-)?1jiJGXI-;qmEj}psp=xJ3Vs6K`2K4aiMvF3
zrC6pGIC{B#APgvbtZhRcKR51H95<e(!8>kZa8uQoU|64*PugorY*zIU|NN&7J*GLZ
zO+mmdo&wspHWE_jKoTiyXeNR^PWfMAi(~l7cHm<<Ju%Yi5eAevV6^BII*sC;JGINF
zbv-vabBJA1rwB%}#ji@|8hyjJr8Px4RV-n%JG?epU-Ynk5}SHnoG)XzI8O}K!9;Vt
z$-mzp6tq6>%`-x@J_9DiiWaDP=JK$#{-+W>Qs}m;AYMoJ8#W`&6YBUj&+_wZ>%F`A
zItOg$JKYld2b?<7OYx;y@dPGgm_0j@n2UpvaP0cKB|amyY3&fmi6*Fy&?8yrG^-sY
z=|5GVG+vA7U>4=}#wPI!3UCOf!xl*8`==Voc2ifjN8{+4FXut+5|a0DK`98iLR8Lx
z8Y4=G<$tczL>!wcz$Q7g1L|OSJ=U5GKu;4$U`b0$tgI?Lsd=UN@wi?ZND33pAN=PE
zKa!dEcc|AX=`2L9Bw(<bTH}uZi>kGN6IJmAm#uG3c<L>@2(;8|m24Dm_eyPLDzEZ+
z1^^7#;zlKTcWi{H){X^B&XfP=Z1~C-om<Wi8~1?rs#ow50^bn@_Y<Hd$|9do4CjB2
z5ZgZzOt4wE1<0TTrru@p_!Lj2u|oMH>DNNXGsvPXJ*nZB_bBIy$Xu~pW5%_MU>``4
zWml%t3xT_8kZ(dws1kCJX;9A<#ln_c&8GCoQ9m3QgD~+Q1-A!CFk>t%ISParNL8V4
zyd=dNmHfs_yEcF#?J3i_(J!~rY6QkogEi=<FZWa1nCLCn$8wDCwg1kF?j{yBl$9No
zdyYXPAghfk{(k=?L%<$wO886?oY$toooWRK@tdM`(<V!)<ey7I7S#gRCPG@7zi{a$
z2N>>|47ki$Dl;4HZPvuq`SJ?yo8~0A93UsYA5gg6-sMMep6~bX3ZK};gvDQS+GK60
zW2g`be51dwF_|?hYhbztk`hCro;n64C<!qz2^qGy0YI)9H)K4h|D;YdlL<(W26mhU
z8vHldhmQ~+zltYrax*$Kggww$0|O%iiWbM5lH};*U{*>9Gb?_70@Bo7M4*siwG{ud
z8E&kEgbIz9CY!#8rX&tK@S$8CJz-yUbyr9;pY{-L%y!Q}?V!Fak5Rr&r{XD=(?i8S
zw=RZsgXTmAxJ6LFp$COPpYLn8p-uYo3V$8cP4J#Wq~v)BadU9!>x-&6s~WdQwpG2)
z2=^EsIP2H^DS7sH@*8w(WMGqG{eG~*%li;ewYqc6D-mV-Qt%$clo9AWh{WUs)O7CM
zp_*Ccc9tjiEHO<almGEUJapKn8X;+Q0to{L-?_anJbN`JN@_sq;!EBZe=Bz8m{+_$
zNbytT;s4U6`#9roy*1_EnbUnovUAdbGQ3tt#&B+v&-zb(FneUD%F0T2Dk`cwQKK*Y
z9IqKDU3Tf~U&;!cSI`!Guc;e^c+;izQ$0yp5#fL4khBd1`|^L_5k|40V2GfN5$4)|
z;H<G(4cx!jbMTU)9mHud;kcTxwQM#$NlMD7+3P*g*X^A}yG<I*t)Jhlt>38pnyuTi
z@jCAt7#SmXTx0phJA~>B9LC%V@J(K~b-+1@^RMUfF$uaXsf6i+TvWPTqYbsc{64%k
zC61-xBnB@M--*D?52=4Um;7=-c!n%u#Mgg6*mJM9uSx^FzBMkqB&X<h5Oaj<R3hBh
zg{>+ez-+Qas3x&5^klG@D@Y+oL5bJ-hbmI(AK;XP`x{gV9CiY6#r(|-k~!bXY4W~J
z;l_{gH2=zxnKzd(u?Uh@$7HHN3rn+i59}>k&E5-Q&+QXZ`OHv<_6u)MWDQXCbdZ%C
z^&qhKhRP2daTmU$&0A7JTJ34eyDH=A(fwfI8Q_l6#8+sK3)O$2=Rn%!>Wgp$tean!
z$ixFBlL%6GP8<7n@%ikqM7F}%LI@$Ld0*hvn>d&jX}3(@7S%XFK~uV)eH@yNWp8AH
z$g;ZAZki8F8PU1U`!qh9DXdD!jHn2@Gl{X3j=%j9^`Ws$Z9Bcs4(2id<n|U8NN~-2
zfmTOoQ(Ru3lp>D#4G2YgXqDJt@$UJ<^W^-j3=FP0q@nC>zQG-znF6iG?ndKp_ZyX`
zEWvs|HXJm61I?LDKLZ`aL|(Scf5^RsWrMi+3*dmtVAM$lkr2I8O6xOFXQtT|Q2E3g
z*qxW2A3;z47As6tx6=U=B7XuP1N-Ab_^|C2=4;70H6b%sauGzqzl>qOspYzP0fjh!
zu^64D?W@^_n$=2ph@OE-DNx#!-6oH*tKIEi7`8vkd9Dz7a6rfD>e4$TV60}vdfql^
zjOQ?fE<;WCx6;(6_hRD_oy`Na=0lIFhz|5}yLj}!kPcfUSZO8;0Ql;AoV!pCg)%Wj
zO}Zl}(eJf1dVgktndXBvm69?-<Z8q~4ueV8GbU@y@e1ZZ!)AUYH=i&|Bmn@u-HAK<
z(#Ll%dCNm{SHI%W@5L!)spN-BDWD@O@v2t1Yym(~cu{`Xfz!+vdOx@>bUzWt@0cu%
zQtVBPUS7l4m<E;8Bkip?Rv8>NYd-J1bThec9M?eF@Az`x6Fgq9r(XP$`oK!%w~JuK
znIuwx|4whAqD)8`w1}6zMGlZ%#27K*W}iRbtXhwJc=m^xZbm62uz1teFn^T{clv?J
zrl-)C7#~vOfrF#H@3<-FQ@rwtzz}mSs5h&SN%@yXf4utySYM<@O}bV5xLHy*Tv|l7
zvLc#UiCkSmxX>9m5`PyQAnbvGxKOC>@3CuR;Gv5ILnI|PC?wsLlkZQ;3W`e*7ZT)?
z3<~0ypSLR}&C{|0BhkpkwbYQ;X{hRU4OB23BSQOjCRCo6f@9g9NTpSI=dG#?a1rnO
z-7)Q@BYZ;6clM7F!-q!O-sE*w7b>4uMa|g0y(ca4@#nSYV~Kn*G3bsU5)2Z#wb-s6
zIAa%J<2G6Oc+uH3(r9t+Qe`B5zt)Xf#0plCH_ij6(W6W#u@DfoqeDAj&si5xFQByh
z)5ZeytXVa$-2IBHQ<R=FbmLsl87V_LHT<1;HPH4DXRZ$oIx@y2)*WHWl%7O5qv$+c
zLi21=bV#)*X9KIPn7ddlp=~xIZTrOnl74~V;6OJu5*z5Z@Ny_A67zy4rH7MnVyj9y
zv~boy9y;Y`A}F`aDa*h@gT=ZKnpkq@C96y%tC(5$TG;?*HUgPsQ4|&JjE#3GDSMPO
zg!Tq*I4I<Z-0#gI2aHJPJn2*xDFwr33tVfa<x`+Akjl&QZWmRNQDKg<Lz&qji@2nu
zAp%0L`)~>Edvgd6jz6T_k6ld@lo?=5EvR{=Y6wCloTT;cco7vp)<EXudEz8f-ow*_
z2o>Hoq!YKEF?rN^xxo?@Bm7DH{{qW?Cv+}}Vnn=KVrDEP{&|lx*nnFk60HeAaipKn
zq2`L+*TQ30ZIDk3qcZO2wr()6ZHY({;+m~`q3JPvq{^DYLtT)utd1CC`hmkqrK&$r
zf|FA96U|ZnROacgvNzCt#5W~XF#Mp7TT&6ia=V4GOF>Iuv1F4pl$M~S8&Zmm`PQ7N
zV!k%71r{Umf{p(|>0)hmT_`aMjQ$G<*!nu7rKS052YAIFysI^57J3b-7f1Sn0KrUi
zNzy3X2-BRCLc$ut6VLM?QnI{6{fJdMPCw9i1^wB-W0S`Btw$&;8gi??+5OrI0=>|~
zt9{DOHqLXHo>%JO^M?l!#%Ge0G=1&PB$A?rjEYKc_{2>}BET1#Yzao-H_1u<E9jGd
z*REm}YL*#AjC_k0I(z8Dv={$}al|GPyiUo++Qo@&rX%E7R$voD5Nz8|3=f1}60?Jt
zv95Dka3p&_6!8hzYWp3cq?6~GOlZ>}r^)T#Ik*j_2VqKUG9yG(X^9t>r0#BQ&7SWa
zHM*WbuKrGP{4Bkfi7&noyd_YQ%-{-Yph{p=RP3<}N(w$Tuvy7ixuQL<6V@FBK{0H{
z7776F$Y`EvFaX~&{sHCFiuWA{70(k{fBT=wY}qB^`s)MqHR)h@0ndj-POJ~C@R=qr
z^k3Ce6i3JXlAE@=>zXCXgyQQN?<TRUB|1Y(w2K|jGruObJYk$Im*W}r0l^gZ8R4`Y
zn&YKjPxy6+S;7{hSjr3nHMPj9ayAQRtFXzTT4q>7B%qn!wKFUJHQ|!kKx`>Nc1;l-
zNW~C~fZl!t=O(Y#Oxos+;4ulqiJU5c`NoQwCBhG_lkx!_whHG5DUEIo!LfVnLYWL4
z3_Sd;cC|_3WTZb8dHOdg3*OhjpVi%g!D5-}oTTy@LSqxZCAK_m<E4lE=F!o$HkEGL
zgFQS^*yD-;++L#j`=Un}J$2VfDf5pHOd6Q9eeDO*YnEqr@>vo?j3Eim$NHq`=ml`%
z4c_By)s@r%hlmOvV_qOp@MxJX)Km9PXaS15WOE)PVW_nnS7Y4nCm+=<ryB_^d#WY5
zB4%2R0S#313|ev!C?n_tQ)#eiJIa*CNVFns2Gx?)ER;pjAjvlCVU{fE)*`T6QREW}
z&}}rx>La0ujlqK&L?VmbpTdJ36r;~KF<6eD^me4=D|@tSGpaRULg+~gU0;0AMMXyE
z?bTT&OU69NZ@adQ%}QUZM+-D=Nsoilwdre;SlTkD#E=P$N1lapX#sKRI2d-?XVP!g
zDi>#&!SNC!YQQF%xU@f1=k7nnS$wXE^VT3xo>GPWjt<9p4#Z)FNj9zE`_-H&ry^w!
zwCj`nn(mVL8y6DS^u;&JMv*mqTH4lm6?W8e8Hbt1T#$G~*oM)++hb*0;=HdhS0Z=C
z(S-Pm0$}xg#9!v|A$uFcoo#QLBRa`_HB)oVJEUU|ei<1l_xR{WC*=){7c^BeZx)O~
zd(4u~ARH%bs+R0rp=fas-Wm<zm?}GI)p978-mN~mqFGI`zoU*dhEOS07!^WEqwCk^
ziiv`f@SY$)Bq0O$=3#Kk2hKnXVys!}_F{~q>y2+=Nz5KIfqE40IIvrI&Io>ZgLI}{
z-zX~HD)RODXV;u{-t(|rn}r&Rnk?-;4KmiJ2RO3p`D^xmaap<h6(Xs5Q~VJ0PJ;iF
z5=NdIUEaq(i+Oo`)Y%#v#auU@!v>G6yEAs|NGIR<g&Jm>3%>vG)JCgi_4>YS-=13S
z=E6i{&8x7aXh%YdoT@suY4S{x8k?}OyZP*oh}o+hW7Xo>O|<s!y6@bbm8GLQG(G~P
zDfQxrJZuCQ;E<({mGabC@?XV|drHCW_;2{gou6S1<kn9va7x;|k9Ah7=W$sH3z#DI
z_q4$Ld^u4#sdHA9b(QM;le6j(&U?xY(U)0^sg<FoXAl<O#=vmjkC3Jq%<o!GJkJ|Z
zov$5;dOadYN`)%$EHN~RO~L1~G!}vfsuO-SdTmlA$|w^W6zh`G!<<*St|z{)JE!_E
zf#?l(Nt~KSxT0800Hw#0w!E%Ywff!O;6mpU@!(PR?{B|T5W+ZSdB(XDlCop{#QZhe
zZ8y_3Z*@gI)8xYntsca|#B_dbox&XX`T~}oLR)azi8#u0&M;SU6;@piqbtgJ;wg8g
zuJ9-$LFLSIz2x_}WX7@{{y1aKq7R`ta~LsVDln>tz*QRR2jxuY)3=myMdbbn9QpEb
z*b<+>6~YZJWNoe7bw>~e9%OXcOdmXdOv46N*co*i1Y72Q%HL3_B`frkc$FeYVCrUg
zmx{WsQ_;Kix5tG4nYrik(;Kd--Ldq-gH*`e;O$H4$GK8@wpa2atK}%?I}(h~BdX?#
zK9SUM!;ZSG2HQL5Ly~IU!)OWB^KKZaY+B$cnh$ya#zdwkaR&8i>*d!+Xv8!xJl^i7
zU$SU_`Js_a8b+1SB{vjarP$@!<HNM8RM~I;wF^R_g8!=aII^j|XimY#>M>MBK74AR
z2z&vlLX&s&N}oObMmbN+Vp+}*2ql}Ebf-V+bDbBdaa4Sydw+`WPRU#>xvO~bQS11m
zjhTK%S*?l8>3GJy&F93skMv@VN)AW?F~4UNIL<f#Jw7&V_lRh38_+D_u6%5M(j4Q4
zAq!J)SaAyUALS@Lj2ODHTakSfUMYEPwYD(VTlKQc7`KTqtgo#)eUTX9IYVJXo)f|=
zi?2;(I-gi~eBP*LUUzyw4|l5O!;xsDSv9fE>im*)I3Wc@83*GayEAlbS-x*p;Y?Nq
zVL1J7h7>`<mHmyc18l@lZlG2kJIERsG<plp%dyTQ=VtNVq`xQc!Am{PytPhFjC^g$
z#Zp?bSC`}-0@x{i&ly)V#OAu@sd&Cr3#ZeA9PVjWm76O++;{t~s9_5|BO)(uW}RJV
z%A>FI>-rYk@xAYwoX@kLt~lfuYmuST7P>wVSnmkPW@cnQt9>Q6HWC;%E(yZ%J#e*Z
zgvbvO`Q^q4P$fBU8SwP!)4{bIkNDl6x4(u1`|vP6pKRh$P`FoT8$P*F9PaFkiV=Tj
zf)XJibapZG%$u{c!)yGvNJVE@Sy_#Xi;BD<WH#&*Tq3hQGoe)+e`wltqh2kKV>_Qk
z6<#(%Lb#w?77E(smS@+wVk$Dc5uprYIQy);fi84>@fpWV^=oxvOm8t`Ie)eqO=TND
z>7LD(U&2dx_|ThJ4wOoC!-d@xi^|uH*T#9SnrAJA21aHrZ&5Su?hshOz1BGB`EWyD
zEzLJFYHF6fKRGV&Kd@YmCQ!b9*h*5<UZ4nFSlab62`ls7(4!2a#l3!z*squlqE2`9
z5?*@2r9kC$-ZB)xI@Y{Jtqe&p2+XJ3O6D+G?_~Gln+7SK3glY1kT00IfHvgLoAtpZ
zhg`ZKw>mIDtRWkG=rWqyfLWfc+C4Z+IFgVxZ%3L=9Gm%aa+k=>)}3?Xu>C1`P9?D;
zC3AwafTkT!sXy$3y!L(s&mKjkYT5B(I+_q{)yf)Ba(#tKdAx!~MNTBs(@>kFZCF&^
zsV)<<D8E@=Fko#K(Bz(53V!xhD>i?Ua^zshlv5T0DJ&GQfT^00(Q0kQdI2xQz}Oob
zg$U`_F$SF0o~e5|tMMpo>z#ADVECeFMiK?M_Cv63x)XMxSB$(&R%1Q%w?5Vxf1J{2
zUTYLfuI_&kS-*oRMrZ!c+xZZc<~ZD2Cn9}+^Zo4k88UH>MNgC1iEd2U3@Pt&Y3F{T
zAoT&>qMuqkZ{`B|Kdl7kW)RQ?N*udR>*rQnQ7Y(+w*B}5lcO-aY6%M5I6o$k5)i`p
zb>$uTgDfl(^r?S0G!!0cak1_sPu$Gin2(wysYCZ8{=L<)-&*`U3*h*W8MFAYCN;3W
zB)YsR`4;fJ29=yLf4kY-P^}a6op7wNS&QIgFa{9A&1VlQ=4Q&Zdn}za^miikV`wLX
z=OKapmLLj+0k;p;OLf6oRwSpUN;|kgWsn-)PveTaH=nmpNz@}H8@?YrD>c42usE~N
zd;8=Qf5b`-XED26A<rZFHQgz)*e@U3@&z}f_2IdpuFk7c>yq^|j3^H&<7kq^I_E=z
z-;DehYopBDr@LPSzpfihN|q}*rKZ%H4;OH)nMP~oJ#Wddni<><k4xxxzfeZ+K|wBm
z_FT+N@H-e>{V8~a7WdM__}ggOWq$ICY-hV83BJ1>MqCGjta&5y2XB!fqBO(029@s{
zsDsJ5?B25hH&}$wyP(CESHdkM!H}5^;30(Kr`MJ{M}2*}MAH_J5^v`+@fQI4HB_&u
zX|~CRHcA?E{&f+i!?LS!Fe8rvC7z+KP6Ce^XZYA4K0MqG9b;!~f|QsSUsZ}}ECGm)
zg0deWk2+NOHrza5J7#q&y1F(@K$s-&6vyi@DOIy+N3$7%x|~5X9Xbp=EsazrL+Ew_
zbd=AGJgi_CKoCQFtZZbi$m$ns=8K*-tRDJ*T!1#-4tk;WwMOL)w;z1AevHdHvSSTp
zF5u@t#r5xeY}@vTt^Az$-!hrtIkGv^k~_w5`H9=MGtnS$ZxibN0_VVUbGU8GT#pTJ
z!%R}-AF4|Fu8?vE{|B)6=uRbJBg_M>M_Tesz1`-pXLOw?%g?is?Vt1<it*O$Y8=YT
zIzJ`%A*DxnJrJtS7fMHNt*H#>FI@NEo|Uxb$rYUrufE%z9*B91TQK--UE4PqLP)R^
z#Sl0c^u?F`G3Gk&;ELKI#_BD%x02~hafv6MaZSr`C{p3qMu&d=k_{dtOg5x1)rQx=
zX^T`JQw>n8tUI(nN{n9|G*W!RY*(d}MOh7?qV35Fr#Mq13!=%L$Iue{v8lA8u$#7r
z2tfxPwIDJwDBF*G(zm`61Ne0Ww-`g7P?r!0X@zQ<vcFADbTv25aX5GO;<ac}oE*9L
z@wFAvA6j`Fg=aL|;sJi5R2+1K<-iE(V%wJB|02vACsvP}3!Wv0Y9bD%T^H(waLyR|
zXvNGM*4i^f`=@<R3U7Z%(c_~LOA=b2%<yWl%Y=8bVkzGBJdT5h4v2<X<l=p~tmDjA
z;ghY-$DZzmpBL9u(O_Z(v|VL3v^>x_)6Z?a?8}Rh)pnxf;*MTHwo*a}VE#5i^pdM?
zG{NIHFe{3c4W>}8)Rhi1%^RC*QqA^Y+3+!W`)a0LpB#+HSNPER6f2u?u5f(-ZS-|a
z3KPSROu|WqmwaJs9dXI&)QL@ioCZ@906g~fLgVb~8}QDrCAPIe7?^^1R$`m!_cdN!
z=JSScQ+{*%-bZgw^5^3kJ05j|cWt%H8MSJ`>D!ljXAbAqJIxo%LdM{8FR^Q(TClWJ
z{^zq?L}s+<0Iu-GjO$W}&KY3rHizxwKJr2*SVP+mBx4uA#?`NS&@8mtVXKPLOZtv>
zl$!BiXQwN6!@LcXzW7gDo(z9aD<spHic;8c<|;(Oij#w2X=`LUGT4+}ZebU?KOq!<
zg)ox^6@r?^``jZROr=L5W1xUm$Ky*WQ>>N>VbE5nD!r|Nq;I(+{?dlWJ~7JlLUAw2
zj5JlQrL9vgjCS!9uFZO?8LScA3}9b%_~13tw!;`a$_#SA>4cF8d(q*(GQadWbNuY8
zf)l3Na@!Mo%wj7V-v5)E(-f5<k;Y;Ee1Mmm_{w@c!XQ^RdmjEj#L*E<Ap!jG>+pHP
z6Uy(CTE(GO(;02<?gVkQ0WNCZ^dljoI)A7Y&T7n|6NCGGuKtE>PPTFXux#~m7nRDI
zRG8zAGuZu37M*C62vrPHprrHbiqoOC)iW@Fg;rhW&^M>#VP3RUz%S5zeuU#u%^^DV
zH#d9ClyPQfkmJ5^!wOKoOFHf3cc~9~@yi>*@x|fK0GMLT7oR9UKcxK6TJNNMidw_I
zD&YlaXIRX-hiPB3xA%H1=P7vwv8z5tE(@ZbO4W^ttQ@c3{dV?WIa5cGt>OWvTl4Zr
z8p}T74w6*CmTH=mHov4Upji-)3UMas{)Z74-T~S}P)Lrmm>?#G%$8qn@Kp;eLz`$*
zX{esw%E+RJkXQSoQzVAfWiH)JN~LbQWT#i2NRkf>ub12-nXEsk^=OQjd4?ePNb;tm
zuNWVpzGH+BiX4T{Zuwp9CY1gQ^8L4A1-sK2iEI{dttl)<lrz|fYQNu8{y)J(^SIIu
zC1gpi2KGmJFU@fHzXHaFbMkLmRn0u^Wb39le$SfbM3JzPbu?cLHYDl|JWik4H}P5h
z3_tHj5Z?T=;`Kn=QeQvHySri*+Enrr*sh(5^RM3Xv$PDHO7(gqI94Z%iM^@GZgn*o
zakpQU=!mccqaCr``iy+hzdn+7)&;UtfyhChV6%~*6Ks9BUSm_;XUwu*Fi7H-C*;(9
zLrsWejZEJACZZ%Rck)6Q8J0jzW!7Z#R*-8iF4mij@%tqE9u=j?AkS4h%+QDzH93vB
zXI-Dr<-Dj>_7m2paRmd@{3!Pk9E?0KysN$-1v2IjqhRI%b|kK_w(R?4kVA(A_v>CG
z0w<yeT^f@!Z^YUAV}3I)XJFCp&eZdM#O4T!#DNC#RRm3CQt+Q~tfmNBm`42%&HH1M
z#2>k!9<-vw9lZ1;=!wvKMi+mXL;|<c*WH<z&2{eZp<iB{yr6<VFGEMjq{@c<(us5a
zt3tK2tkWF-$N5R(L>sb#&9`Et=pxFnZ4VI;#YQFUKy+2%B6{>Ju5sQKH<H1UUTa|l
zdj3(a<~!+}y6q)A7b9;lL6)LNO0f<i_%Or8u7QGcXBikPIlru1dlKjV+^rTs!XEMy
z^CP@ZNv=1|TsL38Jaa=I!bV?hf(y?G--#&g?Ob?pyjtPzn1}l7*)%=ojjw>d6HHYu
zy0iLJ*5$!bo0KLyiFTkanh(a-oh+leJu2+(lBCv5SuMwzfE%TJQewQP1`JNEx%N*=
zTHlcYheEKB45~c`L@X?a85&48gbiL5^nY9%%Q{(3C!OkM4r1b6X(H2OnbS16?AB0>
znm}1Z9+~k9SjyHXg!re_fOtFiK!F!e!UR5hnzBLBP{jsX&C$OVRpL11Bc1Px4szKM
z4qScuO%gFJpM-w_NR-uj{l5|CUYCRtheiLUl_+@)owU?(TW-ev4Z=n1+p0fq4cl4Q
zm@58i&6)AM$`?atKAl!gU~~LR#h~#1)%Pe@I3p4!6yUS8ZO3e4jau}y5goU4vMNl`
zclw~2dbj@-a$*-}PRXZx&5@Rdt)2XOdfo6!{o}j*vnMr;B_{}}ke8V5|2SvPvi^H2
z?>qK(nnt+-jTv?XwnuWTT|r<59?ySVd|~-~au2V1AsJPo5Ff<DQ1-c|Iw>M6jHH9`
zFGl1ynHNnF%ni!t{*;=c;;&QHAc|);&TZ3*9x>buXDLHP%E<};;u#Om$rN7Kyoz@?
zN0#w^l^pu`yZpWF@_XBMq1OhXPaJHd;uzXON1HcAE`O=&v$85ZW<->Eotpnq7@>O|
zH1Om}qTV|YR-j`iQ`Hj7&GY}xObtEtJ-NiFP*8cXc6kJ<m&e<lwyqXIz)4L^_p7T-
z1*}ydQiR6qrf{eJXUMZ!+5oz5B4BT<`W)mYP6!kW?Pq-D(}w&bcsJ37XMGqO9dfBo
zUEn}INW8RI0ZqiIDyD+V;IGmRII(SC`V&aP(7ZO7>Py6%5sd!23M0*K?a8qSy1etp
zcKc}j!9Z~Wqqzt6SPm1JYQuB6(0R!%DhlSuE5toYZ2A3*ul`_Mn_dgsh>MfCL(5!B
zFgp_?rbl^Ig{KAI>q_12vn`tx)RcLDAVIy0@w?iNu4_tUZiCD8`5>_0BJU0TVTz|H
zL`B>9|Ir~&f~=99_`w}Q#$vZZY2`xAKy=l?acw_-l!Z*?Ur#XSFO~unnytvh446)e
z_8Dms#!PccUV}5@Y74MDuXsu2)8L$FWsl#0ZC*aCKDK>tH(xM!wcqexQZlEr5x=4u
zyq`|g9+g4;C`HE(Mk9UieIon2#CKZ13@@&}WJ((wAzzwkG3CyMpoxsvbc{ZZ!Nv$Y
z(4d1n5`%0?zGtDc+ULYkCTBN)KZxmMgymGFi*fGbF%}*{Ne?y^Iy~@?MzG3?RhsJ<
zF)}D@AxWqPrDEi*m=gofz8y_M8#bGw>(A5&TMpY$r5S?`7uR`&h9<N)KdtCGtc|5K
z%yg0mDS^k!wWeCt&rTw;WUQcF<PX(V2cx2W19?i|f%P=ND3it>Gt1A9_}xGCp15vk
zS)mHG&;-ByLqoo451UPxC2-t{sO~tJ6sWOQ+qXE(woaK1SR9gKJl)7aS?dV<8-po`
z=$<QA0&2=p7`>ZYqz5;t>+Nr>&iLDzZ|a*=b?ezYSu7|dLI<p6pY|Vfgqs%>-H8Hp
zKX@)STVXb5SGBn2X19A^vpcwUS?l)?TCxpOgtdwa^RhC4>vpFCK$-7>=9lY6)I{po
zz#CytQik-Cg6@@~3{AuuBEQihXE)LHAqvy<2bI#-BfFxM^u?3w)$yqx0}3i-=c5IR
z96$Jr9dgF!H3dzzEhNZ1{vnK*1GW3nXXIGP=QsV)$r56p0WMPxvq^@1Wn(3>|6-ur
zGPgap!S|`Kc-PWq!{*E-mRENE-KT^6AKPSdYEl3!ix|*tl8XOR8xvoF@|6EIG#FDS
z|DZV~PMIru<oAk&WuDC<HH-;CpwQ2cjfjN}PxhfLrL_BHZhI4x*#vgVy|^>#I~rfV
z${60!zuUU-l5|5`&J>0|-o(M{)>kEMD&S^qD4oSNm_@#?_?N8|JDyi0Y@3e!fqy;`
zt_&CK#&L-5YJ`c83*2uoAJ@apKzaKWIPefW>#?lLEl3TwLCv4vnLW(pKw%;SNoj1o
zc%1z7v4*NwhJ?ueb+FC!^bljPdwPYfC2U1RRB}54WPLkB{_IDS-=!Ah6DvWdDbwl5
z?D|9MV?nUOZsngww<{DGrn((%wm3Kz2rf-USUZ1Wr;S@!AGV*_=Dp^L+g-B%=KW4b
z4VP$?-SpkiwGm|IG4Yqh73eVct6o4){ji^N+_TeDbm6<_D<`-882a9D!@>?Sy_n$X
zJ55wOhFF+X#(Ege_1Y-&>RQnY8ql_bAJmeuNp=h!PPF6_R~_ZH80^s)IDO_V|1o|{
zi*yA|)t6fRJh808P%1;<Tw{@ZW1U?$g185TXUQ;E&)lg!C9BivG2(c=A*upJ(lNco
z7U9_q&<_|svpwyPj$A7EF3OhXwuqjof2i5dUYU8`!OzL@r<J*GNB!XM_O!_|5fHgO
zp+TEEASK^&Au-WM_%2L~o~<?RdV|#u8&ka!Y9<2iuKUf_YU1=nMUOlyl^c#Sg??_Y
z$T<=PlJo{WgUpC-Ra`ULcP)l}lbTUgVgGft>y_qma>rTV_80VPNrUCGas5wIbctJx
zpvin+QKu4>0cZ7W-t921E(Fl(oDG%g{lyhkgEI2@_f_1I)O69<mCTzkN}^`E?a8+E
z9hp1LKfawNid1z6O3Cp2{!O~IfE8j|P@lG%#H?;rA5y}TO$)AV#`2~4@_4+~M;?&4
z1`KCCV(#ZAVomP8OAxP%ODJB4KN9PMY@tkcpzq|G9f}cVPG$IgZp1b(JKR3SuV24L
z?U@DsF$cKGykWTJSz=J*`gWO$e}FXDZb%Gn6Ocb0nI<$Wv#e{9!{{2be*VQ^B9{dy
zEA_wf>qJa%i;cpw$1}M*P|PkV73hY!&o+2*ZtJ{z1~eu`ww`!|PxGMm+a1F!2@2?k
zNLa_<7~y*{pi>Y`PfD}h-bY-mZSJ^;@KcGtW(u2sctNY-x4J(H0p@1G)Tx?OmSTC_
z5!X~IFF9inf(BHli2;TV4i3?gIpF_6C-#Z-3{^ja51Uo82s!*0n!q=Vzn*20M$-~s
zQSgU}4G01A!&|DxHgA8b@eyuo`@(iTSW>p#!Z0E5FyINC6$G%}Dg_SFcN=ULb&Hes
z476VX-a<(jl+Sh2dx<>Q;RYjNr%ITZV?uJgLyBE&r?={W0Rn3}o(4<GNET*w%|RI`
z;l|bQ`c&X!xpl<^Npb!bHiM@ZpW&Sdc07!5`$zVy|I;kKH*_0d=X#e3jH`nK2uiu*
zo_gzW9Q$LKUyszSZ#^izNO~l@mVb<4<#vW=jsHFNJ||mlca$Qe-?g({-C$6wKYv=?
z^(86xIRAdDU2@x2_aY8g2;fRi&!5pcJXxII71YfC%5<K8)}16{8@et#Hc-~!)xJg`
z*wYZGFhUwnXbPOv_&`~L8lr{-^{A)IagOtq%`y!1&F=dyl$$|Fb$lT$-4lO@vs_8~
zHcl2tldFW`dZOS-@;q|8@>8Vtdx6Iu*|fOqv8f5axmlxvpZIbg(|q^S8wvO6U*Usy
ziCmngxBCAG&F>Xiz0d0sar#5MZi=2RP-Q*M!r_ngVh>LaIi%94J^#Y|XyA1f*iGw#
z%v;<RTegHxBM!(@xc5d=US3+yrzHcgC@M<H6AR3h{yv>h=!W}`JN-etTN-R^ZOctP
zsW)4Zt0a}hmKnSfc$jKlYreJLqNFs+eKuiN$E9|mvsN)CL7yd}cc#L&?Ex8|HgaHD
zHJE006!&ztjhttFNX_=Xi<3CsJHUNK>O59xCi>Uq?~x^hgpp|UiKmR7l_1-dcNd)Q
zOrfxk?HPPc!a+Q_ENM$49y*4hzW|ai(SrGC%UUqGh(F1X-wMKr@WHN>X|{HZa3ly{
zpZ}n1y@n=2fwEz9&^o=hfL)LsdBwFQQeoEiG2(ln6`nM{py@pSh+J5>=BmV4)b$pe
z+Jf}ko*v5hl&YzUuk~$>4&2MXuH9Tum^sX9lFd+a99&dGG1}0zB{LoyFx%_+<Uqg-
z7m){nyx@()$jPZrslKq!@dpU>l?rt@N6q#@=@N-Bppl?~pW<rP(DaUsk}EII6lLs;
zAUo=t0F(XC-X4Ixr1s0UvzFuhh&}7*U>&e5V|9KYo_>9SBk^(&Uq9OKp#D@4GVZ2R
zm}J^ut;ve0)r<wf6$Jh3lYVO&GtO}-&B*skkE^b}Ee0AtZ$m&9{_M++R>hHI-?l64
z{$#%2U5Vt{zN?}j(7Pnr0LrV=L)&fL1>VS4UJ7hnKn4}r?CzC;7TQzGdfL`om^s;4
zztGfxm33G|h3s!gBpSOlj0@dA3S1;5K!2&Ji7ngfRx;b?)f@*0XU}@cMqWNZj;bh_
z-<RTDrqcZlf8wgkaKe^Oq!o`y@8!KgRW*Uh1Fx&mvh?L`f8@YEvULX}k?Gay9oK#M
z{#DKOViK(DBl>+?TWshL=*5vs3xFv*=DMm^Y;k<KDiDM`5lx>m;X%`a>yXc8UBE>!
z$f3=Syd_Qn3MpM<?Kp8ZFSo1_f4t7ljfa=!rYL_qiIJ(QDRjbkx&7gL2~dbr1%+OZ
ze&_B#fBH`!8TJN%->_LWmrw|@!>stWz$5oq*c6=-%E$9Qf$<a-JyIhJ>^F~(t{8HO
zF$@&1Uy&O%F@MIW?n}s?-I-%1K@|@Uy%7vOtpXgTcq95+DT2H`;to54pVC%;6*_A-
z3VEzRv~MNwax;U604K3V@rIrP!|>Er7c#-Yg^+BNgPoqHE6m0;^CmA7lA^)rb7ZQi
z*zU(rzIZXjqT7;uFWTZC!TQ&?<Mf8<<Wf{RZoA+z^9zH>_JtX_%(8b1M9v&hNXFb6
z<+>mILXC|FIgT#(X*d>thObF?q-2U;i&U@awPl7*_{pt@?z@H~yO7Ftyt9VB_+pN(
ze^Yhss8aU#D59j0b%M1A1#3sG{%UiUXxReW3XQ|zjs&oBVu#EN3M*M6Qn1x9M~%@N
ze?g*#6)Y{X3sA^s>>#E+gZK>S#5O}Vt;=&L8zBulCw>S2#S|ieOG_$SUU!qnSFo6L
zg|)Ser=9a~{8Lr9iHgGr%LlU)?7$VRY2_tfqEp{|FUpZ-!a2Z#t(x$q&KqiZRsMUe
zZ?4#(lEY`U0JcPm&sQx$HI#KlSHXF<XKnaxn%m-{3qLos_-U{bei$ugB9t2W-UEu6
z6<}o}^6-I}7!?lm(0*4!ZYUuwEG|17OOCtbhvEq?uW4gK1b6SDlL4hD5n5<vSQthb
z+(9>(=5cWK4+yMB3q;n3sDXh6U<T3IjWszKgv1KZ;vzMSF*~@iRd~QKNcBI4nOH$s
zZlb|xJek#M9U1jbdPau()21uuVWN`Lpk?MU^)ZiB8D~GM`#EU)-r{nNm(H!vyCFc`
zeydzc(E*kr(cljCD?(j;XocHmFfJM4zH%6>Mq_=|ZCn@G7B}Il71UJDMMm^O!Dpwi
zpd=egr6Og>Sa<$*5@xMiTYmiSq*}44feF`l`3f%^EV17o1@`7Gb_pY3k~S4Mp|FqV
zo9VWb#+LL4^szrFDT*m`msRQl>=Svx+7MSEP>5KBN~_AoSJi4?0%jZ~5hbx78VY^U
z`B68FdahZdhQZ9tikzPq`n#>q0s@7?^iz&Ce<K&*&~SR3T=@<4BmpV9K4}Uq%cg}A
zdqpWadx=VNf+7v25un;Rg$dYyRiHs;E{h&@r(^Ga?|=6eR&3SCFIdryE;vE2nzv`4
z`y$*-2u1!Piv!=|&_&ra23E%K$>|L;-O{(GCi)09{1#+UqZTrBguU`<OIO|g44}$b
zcAUF6m@APX+K&~Mz|{<pk_za`o=^7U_iO&gd}4J=p)-8g&KXL|6q)ONOL+Srxk*ga
z<FK-CX8znax3(N+p=Fvwaiz?p@e3591hwNpuBEIlhxN28n$MoByihZN5_fP&;Y3ev
z#urySMNnfRl>r(8KHIkr{RTLohUVtG*RyKSEY(oAZ!4QyS&=d`GwT#zEIFY&g}@IN
zW(gI16Uw`zSXdK$Ps?Tn3^adnjs^{4fuJMu2T(@TcDEuOp1R+JSH<ikS9cd*ruoJ(
zSdtgW@xu$lI}c^j9<ZgT-gVP=PalmhF<ZVG(eYk^J!W!1loU8F?;+e^*6MOW^Vm~L
z+{QB4<AtFP_tuIlp~1~{`H;w7(HnfQHt$Kuk{>lBjBwo}!Ie<d5edPOtk#mSur(7G
zkCvX_uS~A{gXCZ)r2+-}n+H=?{JCPv@Mc|G`%v?D$!u-i<YW14!a#9!Xdd><`^)gJ
zk;h_Bu6SMJbfUOaskwadk)Dpj_q+r8cPfi}quBfya4cf<+*i39Qg-`h?sn;}h>#Xk
z(K6z59X_L?DgN1j-QZQ5nEXwEyBOG)*m6wwmLc5YT;D%3yvushBWuI*e6jrrot>)q
zyZ5OpG<AWk$?l3@cfUSKljG??xb=J^K~mBVbWF~_pkC!?XO}`2yZB!Pm<^lwVi|z>
zEh&tz4-UL^SKe*fQJMSu_Cc06m%2DXFo*W+mL+c#vjNhIh7PO&V=)INx})kZqY!@t
z(jEHhfvgPpe5iB`%_bEWeph6TwT{3@pzJUg;gjPdB*Ax|#Bp9I3icMYWY%GZ$l-nK
zknhaDGwfiz^LFgVtequIvhmH>k7_ePObP?=EYMWm1ynJ@y4@q^{rIAWDvi;?ZC&6o
z0V*@9b*a-*fbvM8*A`DeHdy%d2N@OwFYhsl!KhqxNadGz@=keD<m-hoX>xSTCF}8?
z7?^I)v{H4)boaK}fO`qSBr9fi`0z=zitdm~`6X1=DJ%Mp9FgS`u+h|7&Qj8k)!SxY
z#g}_us@F++32r^##9c75&XNAO-tfi@1*m!QsBL0cbeOL-R#KTXFc|7`3v)|b$Yxls
zQIX!WCt@lXaWRjHOqb6P${?(v&~}Y)E=KK_clu3}Xu<4xS16QDb;`@#5vWsz*)OS^
zsc+)a#)gcB1`fc$m2d_QwpP^x5X($cFP5EJEcy^p9m`>jQ>Ik1qJskLtHdm+&iUOf
z34gj9G{W<R%Z`t~&hzH60r%3RF8$axwUfz`>7^<WdDA$T=xIEfGU}_lT5^0nbgii^
zjyApez&G{zB0|Zmy8fgthH$kb6Dln3=Og)@$Xm?#u%$`)J>3<jZIVCjmntCW#w6@(
zFlq+1POBDZ#3tvQWf8EBxa%RIJDTkX(+8uJdJpMQrN=qc`0*N4*Dt?~rZ~~W4Gf-z
zg1wu8`Gbd<>)K|V8DKi$QonJWS&Ax|R&!{r@zEhldp!XwSWr|HJOYC@JjA`7w>H4W
z*Tc+^X$49z?4#!~#k$CQmf|4GhnfQq3dUC$Ka6u#FMWaU0nSGl>+Vk1iHhjh?cBxf
z2uPaJ(}19N2x3Uf37P|bVo4>@c&@%J>;i1LG0N67gS%)cVz67+J7(O*H*@>r7R~1Q
z0o#a-f*5PJv`S=HNtp5!pv{>Ga%dOJc}G<I@d1ny0ju_LCSCT~y^J((aL@>ba*C6#
z-s%H8Po5H=oY?}przGf!`^|3qJ`t$USKgSxSakF*=t~(Gm%33FDgtEd%?q8TKKsZ-
z5-Fl5))6)H-4saC@d(pnSlRb1HaAEor|-B1Nj5(5aq5@(<Cnpyl@QE|$XeYUM#7sl
zu^NJe%(K;o!!#s2&~8DtC7Kia@eS^nm=D^>IDBpHhZgBN+5)$`Aqh=iKDmK<!X%%v
zxD2$Y;)*EHtDIP6<}MQP5{v<XcV%25>!g1@2iTAAJX$~Ve1Ga}_H<vh#cxDP{3<-P
zV^W`$b$M}F=<B6ZJl?Bj=}C*9IKK0h{7W2CbF##3_F}+j=;H}}`DeoB2n7+q8Bgea
z13#%lzs>Jti^r{ySK;@!x|`Ja%DZM<OIF3Jnb89lz)e+)iVfOJG+Q_{vJRp`<abNT
z5CAJ_0zyBu9ttH@R^Zp{;r_;k+-kvbv5NguDhlE+&>Kk6O7;p`54=&ZZ=1hLF8noU
z)NVC=kHe0kUKrebf)W)gItSPaZu=`2rUdMITJXm(q#vdk!uke?a)|XO%(H@s3SvlS
zvT?1^C<@=_vtcs5?o=>5u9<~JeqNrx0Fi*EChqsKk(SNx`WAayTACXVL3~026*YB!
zLqnqX(`E{%<6i|WEgn4b7_$x?dRPpa<9hd+(CZ~l+u5}>DT~vZs-HFM^T??vtF|rV
zzkoi0oU6MfAAgE}GtWVhENuI)N%Ir)Ut2L5WfS)5r(j?!v+HN`HDx2hyZK=dAPs%p
zt2z;AcBlxV^05Z`{+$uj4zkjU$)kx}klAQ;FFin0K<HC<8-O8H(9>vGR6%v%c=l!I
zn@EJ-cfT)|w&Ttc3DRO|<;$F-7jk0!32qQ5I*0@dA*c`xFg`^q=5yfBAn;0DJklOW
zt)Qpy{_<AX1qimcvHc$xAnKV|4TQeyh0#WsV>k3Sw|0EhWd0igbWPw;41dw_hAxm^
z-m}G_s}uLAT_zV61GjbEFjO8M#2A#Pnz+FXdho^e5zsf;d0o@LJ$ODgE6T2V8rT1d
zu?A7o<hg>KjjHJK_3Qi9172v(pI<pSX`4p#sB<UyXd-9~FyVlF1M!bF-t{uqWYqE1
zt3R0^C8n6`tI_@8etzIc;J8Fkib5d@ASb-5OgY<x%=Kiet5)%wNDTA@u;X4PTRBe?
z=Mx=&wf$9#5_@GSto?pM-U!`0@{vwV;swsAL)F5#sY$&^elrzeO6Er+=(|Q7r+LK%
z1qJKR+O_B`EG$59Cg?SrCf>Avq76mkk0(+oVexXE=KaOTI9skZ-E4E7tZFtKMo6a9
z9(%soUl**9OJ&fm8a;lfT7wQKXRWwKyCi;CQ0hQjO5YPgVXB`UlPddk;EVo$060O%
zzLBGF^dI1Lk6GRMBv8F-HMKwfK=LUQh#N4#H<Rg`R8*9tWa9e`AYu4O;to9mH8F|$
z?OUl{{xkLKenTtS2OgJB<D>wWo0{*IkpIx5Y+bdU>+XAj@nc5tj~g!{#%@Ol;d6U7
z(oYD1P=xPI#-kynhX9rrX<{dgY*^#F5*JUO+i#-V4Og@Mp{LpV-iJs{Ycy;7k{2Oi
zM`1DF{<4Z=jyQzg-E#d%RLhTtvtT>eV-V@-fMnO+ee5f(whDa2?e?J4ZsB#UIxEYG
zO-;i*ksXoFNht>b3Hw_Q03pP2=$55mcu7iG+@AUzNDx9i2rNXmaz*N=)(dzZm?wlt
zX>R3~KnM{Fd;m-%d@>gfh5;W7A>I%|M6^9iQBe`&#*O3l+iwqx%!LC@)A;t=Z#nL`
z<9OzoKi>O_={}t`YZe}lr&(93neE!?a&2GO0Gv)|ce~x*ZcV_2l$6Z-nuLM<i0{|W
zs7I$)>Q`f8*!SsIc&aMMKI=@dnXb{`kmzJ-_xt}vDhl2HbeY~lEBMU5?d&nwvU0I!
z<sdZ=o~lafb`{{<uog8Y4q045#m660_WM?1?zx*O?=9q_i%($Esbf)u+Q^a@YG*+R
zJv&!w5>3+#A}9>2o>^Wb0{3>=sC}sbgi6Lq<B0Fyo6UDVNW=CWbie&3l(=}qh^4io
z16GeMd~5nloWjkelO~`V8Bl@cf5VXkLX#BDn?vSF6A+3(YLeg9Zlb|x7TshEDeyFI
z94!Uz+8W|>(>o-Q^Ft20d#h%DD+2t(VtF<2bbBgtAVCOm3-CY`r#KP#KnQVW1YNK%
z0T)L|Rxn%x6a#lhLYUL(WYVNb{P4pMkr3Aenx=8b9e0qElf$K#UK*7-Iw~3(8p0-N
zLI}S4>MLq%Ye`H@YzOT+0fvQ6Rn<dihlB_)k|JtqsagCZN#n<Zsv>}f$_ng>DJUL|
z;#srNvNP#&ylw;touMkI-LQe`#ov;2(up{F_i4tZ=&R4BQ!?Fpr>3NlFmMP+2fH78
zhQhn<r*h>w;_kei=RWzGD=!$wIj0=2_xkiBV&Ov|6vas4mS`R~(&NUPI1#DaZ(})z
zg)+W(Uk1PaJnoWG*57&$?vhd?St2lLBf6wUu(!CBX|v`txK}O{#*ESxZsnv6yLO>f
zR1tf~AY;EJKP_A1moZ|@Y8(?G!0piqo!3?GcA}^@qa;p;Mhq}`e=8w`$O9%=40l@R
zEFgsF4LsfsQ=Ja{DH4K&5GSG&u4v=$LWoh35cK@>&$DF7l1PXOiJNY^iK3#SsL#_8
zV7Bqvym>RLR;>zqyUm+7vtq@HC~DS;#;^ehAr5WNNh?;YK&vPx{e<JaG@`z|l$7qh
zs9C=rEq^n~7oLaG46`?5sP@d5h1TFA_mYdlcrXX1J(%3gRa8X5Yp+r>Yc6(=M6<`^
zxcNq&{dy@APd$#2X?CiA{E39YgHXGqflyk<!Gu5v6(JPe{ps?6s(}({iPX84G~({s
z0IAO@0efm1eII{-oiDyY{w;UX|H+4Obj?MIM)O7i5>TYZE7LxwqOP9DuQ;Drv+G9-
z_ffabX%x+xLH4Bc5H=e~y%fqXE7t0B`MFe$EdX$N0!z6yl$R2d)^5h>HbK+0Lls4t
zx1ZIpvd-1e89NsW_oBy@g@Odu0r}_&1r>k`JtQU-=!PCk8QYAUM}!a`NGY3Eer*}1
zgt3krJ<(wsKAi=+&M81&boXi~x2l33Q+R3{bfvDYj>jKwo-@bA#W7;U2y%0CNli^9
zB_)N1h6XAtE2*fcVEy{_tY05GXQRBloF|@mf~TK8_zluR2s|DSpMU;29>e?5oRv;W
z-5d4UXP<HGvB!Fsb=IVdE^Gjbq6~_*uiJD`3Qbb=<B!Dk>w~RpSFg287$>^ytFMqA
zH<_oMZe%0+9>$F!0a;#7&DUQOcf=9I4;q9SR^L2DFmhNmDEoF1yPtjrb!Q<qn~K0j
z>@|Pl-7Q5NG4e>xy67|n0<El^s^5RZU0y}fu%Xy{^+X88P*I`@Kq@HKDhD1ARY52M
z&FukaJ*wS~R2A=VTVl%^@2aVe;)tWi<iDW4`#(1P_3!k4>OqnQ4)CrJ4GW!_=&+?=
z4<F3<n!^t5$C)RN!6aHO*+UJT<kqZRgN%tIre8nrEV~O!$lq4r?Ud2j`wZar5;_S|
zqPd)i*f{)2Uq^+aD1-Lr{6dH$fYU67Uy@S(F-#UMgb*SXxFp0X8_)^Oyyh&NLI?*i
z2DlCDbI}8s1bozTZN2jx2^<|_l_kJqz#=JSbu+S((XG1oV4d7^vJm1BDP{8!_ZFBw
zeL5v2C6?popMO5L+;R&?9d#5jF`+xc?A^PUS+iy_Wy%z`ZQB-foj2cnlZPI9h=ha=
z5!xD6nB*)RGUtyubLQ~qqmSZnI1b2u;|CfPK#@`oXm-su!fPvOXrN}<uQ<l!Arzeq
z)mD~~&?N)v8>m_ME%yGsNg6!JTglDpK`RiHeD@ucm5_PPIS8A**=MqDwb4L=yS$Rk
zk3GrGJMKp8D%2ASq>AT+u`G$tra`fD^;Hv5gn|$PwM!Z)$DKghNheaXdoN|5eop;{
z^+=BkDFuoqP>^97ob;7ELLlrmlo&g%h6dcVH6S&8znj!j6MmnT!>p$#Y<9X{c?p?U
zUC7qS_fq-$YU3i6A&+;fAwhF{c=)yVscmTBk-uGoU9}mh3E|IWmb;O)HSC=6Io&S0
z&=)i)B};!>OLc=2YkDZqq|YGetSl!cHO;VscKqO7Ddj-+FYCPDV*VOBW9LHQ46GAc
z<^o4ZDc=vrT`i@omr~A`Ql1Xn8DiD8#=Tt-V(j0*SSjTf&5=0(DP^UR&3ZKOw#8~@
zCT*+n;fEgv7c2y?zWOR3ee@B>9(!!FWDX!ZJDWfM`Op0J+ix6q+;KtIDK9T)?%cU;
zQTc$umMvRYwQ5z1$ef}mUN^1jw(Uf8p`h0d>Kgs2^nm!ff&v<L?j&u@aYzZ85V&e;
zQ5|vAuFuC+u#?mY6TohCTrgszno0val8UdthO|`D#~fpfgr*1r-6~hLW)1m&xe?dL
zA7hgqgd$J`)c5E?^*LuVW6?_Py7e+tp$5_@0ZM!VX~*P|cE)Kq%PT3KF_Zc=zoWTa
zNU0}%_-Ppus=1Fs!4?;X8WW4NvJy{yEmAZtajW^c5&dJ|!n;uX-Ks)S>3Yr?<lKBU
z+wOdX%9XzvF?A;Gi!8qU?kB!j_yZT5IgY$xg99S=LKSsx0!^~}<LPug<4lB~OI}L&
zX7N%`jZjdFL*l=SnzIgDd?I=w=mX+r(~EYBb!`kG#87mrPcXdPX`KrRD@$4-aIuuK
zwlzYflurSlTMSPyGE`fO(=C?&2;3&6)GWp=VYq<*0$BMo&uFPk6ScLqEL^xS=-Aa)
zU(NN`U*9UZ$;rt~pFZ6(uz&XK*{za$pke0BnU=+btzZ@zo;h=7i!FPdXdM*vH^;1K
zP9lvzNT^u09DA28IQsPVx)sgBvgJQ4MtY#j2`3ot*3dQ6w9;a#mMtgYs65mz>5c2f
zHkn_d*Q?w2#TRV7@fMW)^?)>{(rB<d$hhS%Jp0y{+<xQvq$b5Rk>w?KLLg#dNgg$d
z^fOM!Ra-;Rr=L*2b~RFS>%BpxKDTE4X-ry73zeDmDuh7TY{Vp|Al+{2OZMs3KO~yh
zUg=NrT24aDd;#5?tB^DC6mlnD!?wF0qH@*mUTWpDj<y}8<rUoh-0Sqn$>6T*E=LhQ
z658A)ZrTma-%t6^KcU9NlQ8VCfVFBH>iPcX-|-Q8i<bbQyR==kRoD}fz2~6gb#M)e
zuIR}T`vaZUxmLg^$#Js9@S9S~iZ)L1l*RHy;Dpu)6hia_hFT2&yESCagp|^WUScH}
zjub+)s(SDI`SYo-uMawQ>#eu8Mx+<Hxw%|^`Q<^!=FFLc$8&JESoU~4%$PBw)!WaW
zJsYRf89|LYVFiVDcu07k0-8q6vZcfwHUtqD4}hzomY9S@q}xU94@*#ccEjGMkFi0B
z$GwDoi@ztf!AZ*TV;j}$Xga?DsI6nizh9*2;m5I88nL@V(aY4K5_i^_e6g*ZVMh+(
z=pzO;vGp~X!7$m0SO-a?^2j`A63&V$_I>z2oZGkPra%dbG-*aSq(LVbwuB@cNnNNb
zF2Yq)sh4#zWQazz%NPF@<{VPjMiUZ*LiWig)9u#3viY8WQoVkIF;`?PbskOQ{{J_X
z;)-gf+<rYdUD8@borDDE_MMb2Sxok%bG*dN#ILK?vAeh=WJa_^5~J$HgFSWCsBs5e
zI(;mV-QJT5AqD~GTMWM<rQFw^`W#>w4WwI)z1LRZQpzR3T8rV<TD+{bs{~sAWcT5H
zAQ<}+z{J)FoHJ)m(B&gXj%3J?wwKSl_S$QME-xx7V(HSQZI<5=p&KP6+qZ9L<;s<<
zcA2hOvxW^DHbhvvP6UI39zYvq=#A=SZ5?%M){rpbNZlF*xSUQ@M?CK0A{z3y5SKR!
zB_@_o_8~L7wfeh7xZ>kT9x<|64O}H<<llQgweS5Go7;)*I~88Ht}C}YvFDz}{6#Cc
z;i^+nQNT-}8W;EQeHTJ{&0ZolmgHkblX3QW)NI&5(MKQRF5IoRq4MJyp)6klJ;qB4
zz-}WxGY75SMctlVNX=uWa#$UeupxzCB|Ra?IQb;HUi~LF-*F$cTel&k@FPd9wO&a+
zm@$`6=l{et=by^SV~z@NwT7*?1W!c;JEpx)udA;`#Ks1m4=I`X^$$o@@w&wu>uk>A
zh;!)Hxx~~1HKt>K^dGvH;8}__SX+oXp~I?FexZ~yKSFaXuvmUlE9F^j_mh;ep_MYd
zkWy9wzqD}fkoe}CZ-OpA-Y9+5Rtz6LoGx90XTw{~eK5c*CR|@%-zJx-+1s@f(S@RE
zTDSI?NFb=+z7?&inuK9~ea1Z+LRBeWz5>nTCh6EQIynj<OQHwH*RA>;b<jX!(yd;A
z^*0RXu0l3Uz8(49VuX-LA&g?bMq+}rlXm5$yfO1fF23{>l9Lkf`Ffj4&lY~9&v2Wn
zadD(gIEB;`Po!eW5-R7;!BbnUzwrQq&?b5__tTUY+a(=)Y8q8rH{q$PM#Cx|AzZ7P
zo+&D%pQBMx$UXNgvMxA>t#|wbXYoEi;@R>GSx8vECZ9)Me3v7K_Tld9|I|7%heX!Z
zvgPe}=yB1-ddV1kMVTewbb9z=;Sw)F4<=hkKuTh53R3f+I-0uE4+MBTo_2_L^+5>H
z7wZhoPU~Fru<DT=$zo|%vIYtvTGLF^O6N#c>s@TIyu~f<N=r-Gym@obu}HP)5<)O$
z%$T5K%a<>Y$Q&IFQc64?51)PZS)0h5l+v3a+llBxQ4}R7+LM@p=vAxM;K<G)u6s8<
zC|!bLxWHV?mqB6@$%h`&yeyz*<0h!6B;lyMMh|jxJ)O+eZrj4fn{Pv`Sz|2MWAn_~
zYpRsg_a8v*(8JiWb2n$4I4Z!^+FXi-kipjw2x2m_NI&;HZ2kJOciKnPE?;KE*M-ii
zRQzW(E&_X80!h94QoCy>4SV;5(8;Gk^DbB!xT>Pk?XrtVJpLFq-|-Nx%E~5MhU?fZ
zDlOy2e>_EOQVcIY^mpQ7Vw*XZ!L^eOF1Eh;E;-}I5!<V0BTK0ODOs>&8GFmB(2YpJ
zkE<Dn7aJQxY+`bY^yo-XRkgicu6GkW5u=vQ*tu|EMNoc=;2c(rLIH-iM34|72ff~1
zFsy8etY+}5#qz#JnU}D!eED)q!ZCb!i*`RGjy&?npv!;#)oNEF3%A>i+wJzcLfhk0
zP*4!i)3!AbLa=u2TJrPr+f{c>)9`pa-nM0b1lTl9%WfCuEG6};R}(v+KO#03DL`{M
zQEYaE$4$-36~y%JjV&W9pwMkF2nqFTe@D1Hqz*rvpleI0-mr<Sx7~%ka2KEi`aKP}
zghI}h7xT)eUvkAIrx0TlyA7mV`v1bNm8R%xD^dU|2?Gxyp>Kc67cHXl-S^0xI1x1~
zyP1v_e!F6<CsaF0{RUFIa|cx$*OSzLASkM#p%qCC&!y?rMN#Q>)Aek4_zAW?_!#{k
zdl1FX#pw^L8SComxaR)HD6OvK-De)6cWzFQ?ZeQ^>1O+X{zK+5c_bcwn0G6m-B5y*
zy!Y9cMnUP|=g5DVBQ{2ln>;X*Aeunaw5;|j&S?b4Utux)ij;D1yC`yiLI}|f>x8q7
z5u9o}dV$|y#QQMd`xeLw#ejmF!CD6^Spq#F!^(o!1S?kt=b3i9ot`~=wp=p{^zR?M
z-T96kJ18$NCpEQ2JTxqr<vuh`qq4FRx7*!rmM)LS!?bDBsH>}MlN^&wz4zXGOrAU$
zMNzyhFrx`6C90~DmX`L1DVt&wLS(edOqzykZ9a*U&cvi29+#6CM*^O`g}8U^A?bp#
zpoY<>SxV~iH{w<lQicSv=t)V<#w~1_{CDj83IRpWOwa?y@aAx8P~W`|3?0Iz89#IH
zNS!nUBPuQ7_d1a?S(x#cEr!(bC*dmENAZ{Qi0jddq%orrX14@m)?h@#EHx~J1QZgw
z_rh6PLh-5<bUE}el$aR5BnHul+=b~jHJFkFglc2JJ-3p7>%DA${uO%P{x`2xvpFi{
zY-r%NCtqU4y7fH!kK4&R%xps9uT}62R9|X`L^inC@y<J>A8`~(BM!$uqR(G}?K=xu
zv}_d!RnLp@76ELkr9coD8yhx}?x3)WFNg;2B6z}7J@Eh9QI7)^p<1;?jL=CbokEBm
z=v^5aqhF9^p<v}A(TFUe+eqO1nLql|0zzli)uC_SzSwM$tMX$-=s<6CHDW}Iy~WMK
zBq5cRm5dxYlFG_TFTraMLR4MdY?;BZplKRUKm9aQr%sKCyihPpqzEBMN=jnk!i5nf
z_8l2Epk4YN$*KzKcNdZ|Xi#8=fe=(}$j9b#lW>@sjbwG5qt|)UJk)R6L`-%rB0kZK
zjK_4r)ot6!roaE4n7w;Yga}&bS0eBT$e45*pD$R-DHD#t5o?E#d6I2_p+ih&7Ficx
zO4TnvQTX;dWSn_Awwzo9ni1D$y|I@{2|-MkbkbrSlq~&))ZxRi#U~nxCXLfJ{J%F(
z>dcZRiX)zWkKV`HYyU>kC$q>p|EvIN=!=K*$@F+M9((OQX3k&C{Wo64r18f$*4$DD
zU8Y1x@YL0_?d`Y8JZdyaBM<j(9hmE&6Z8*fFQB%;g-}$z=x~sP2vbcVAR#U`U_TEu
z2qD^q&>60`t1O0Jl2Yz&4?PZWgzD_EDZ+DYAUL73W+B00rE`IlvTddsKS(KSgb;fO
zUIVp7B-Dz~4HyvFdietf4zwJL7NIk(TZxH@Tyn`JJp1gk)Ya8_mmerrCn+Uvx0}+^
z(x}g69utSd!If8DNmf==tFm<{DCj)`qmG&_+t8$&gaP`ss^`5bNC|c8@{u+>@dF3x
z5ARUiLrAomT4YfX_I`c5EyK)>oP|Z?-*qSUox2c<F3Ud{U+5xGG(mk_A{i&2$owV0
zaoS0FP3TX=FBKv5gbbUF<YUH=dH#8neleG-#or@69&fjXaAnyN6393zkJ4o;aF!I)
zxY<?H(j#H)mGu1+lb*@o=bm8ao9|Hh%d&uN&H8XAG)?2*?@njxr?a`^$_u#ef^)o<
z#c*td!XRkwswy^5eUqGVCpERqH9}2oE&rXlK%c`j!214bBwwE+B_Xa^T{<}0tMTle
z=tajHqce6cd{~jGy%DHC*BDk7zhERU6s&ZPGyw&jga$*42%S|zM5GB_mo8mMw@AR-
zvSmv|=Q80<v{P07`Okmy@y8$2w{KtX@=i2`qA2v}(S!Hidygqori9fQ<Y0k<-b6XN
zkl@<B4W&yeN^&Y@_NlPxLF{tl2HdI1IJ$P#Ej3bvDD5F|S5~7{Rb%hb&1apH5@&f8
z`FG!oeajZVAG;r&@LKXDXc{rYhqJc0l74-=k=Z2~L2wJ143N`$cq^QsU#Rx9G_tPz
z6PnAzzIWckQ&#Fnl){MF69UC%Cwt5>RIXi5-KNcG(hUFi5P}tl7K@KlJ9_q{-#_nU
z^8-)Puzg2RiJGSI+VmMb^Pdm7`O<T_`N|88V$nV~t97{*9|v>oEn@2%Z`14CbBI4=
zpw~6(OAztjNaKrdeq`I;Vg#TG4Nd52@xD;WM*Arx$%*k%VCwCPlxR;NgvbJ{O6$KS
zr7VoL4hIP`EtW@v99a>%mL}0y=^V+Jzvg6yt+S?S6c-2Ym=<Y52i9)a!ouiY5<&=K
zVq%y$aUu&BE@aZANnUrds;Yl@a;GQ?ilSh%*_be40^fY|4QHQyHZd_Vf2iYQc8Z7!
z{12gXX94kjdKtm%0*~8;%^riIX}C9U!_l)37{T|t*UW;9NN|^zAzT`2SHEN|mz%Bs
zdIGU#wKs~)KUm8Dn~>lZDqT(*$J_-!GvS0hFPRJA3<0Dn3N{@fP*fYK<4z!T{5T5#
z`(NtU{BC6F2=uU(=4A-gM%I|o)D-Qba?KiJE0atGQmU5hi@9a$F-MVg!9+Iy;}Nvl
zx<E-D4^uyx&Eqe>$K=b;<Bn@C!=|X-`9sa=rL9s@o4<)2AAU-oD=sItM|Z<6=f~(o
z;Fd18Wa|5$f=wWWf~w)mIJUZWC8Q)D@Qt+FqZ;yUKzxwp^>kY2S^_Kg_^v3-WtIA#
zE`(@FAeoiU(QvU_$qd^^R#sLftU5Jx?Hc(CKUOmL?Q6g4JzcwY<%17C;Mr%NB{@0y
z50k+e3Yw-78ym~R4?oN&pL{~^-o5`YB+egZFW2ZoN?dyiv31M!ku0MQoTsW1SyF=9
zr#I3shRoVN2kw#*gpk-WjC+qJ+4k1kI2SBHK`~lO`Nv!tHa4Rc84{d{2_%mg&hl0H
zj2>|aKEe@<{6y!1(af572x5EnCij|as9g0s<uhj^T~2}q{{@2p6qT&JJY02klrR5P
zZ$jV1VcEKU(S)Ge<(Faak;9G`Uk1(NO$&eSy^neH#do>!igUU3>Pt}#!XD^m4L+wp
zB0Q4PAC^$E<R|)GdnL9mX~q@`vI&vHIDe7_KQ86h&0A5Via-z_ucE5@K1-?i_SVn5
zC#4`IIVp(H0}4@Pq(X=;zzr6|uXM)FwFp+^C>nu8Rx%Z!C5dlVF8z|0sS_5;ES85&
z=t@dTEH7p&_h?&SRb03|37x8{#Ky+*m%sdl1q&8%*kOlx$;%&+#0Vi6G-wd>=g;TP
zJMSbmHWoFU<jR8(3VJ)Ch$A4HOG;e33W>?>idi&RC<+Kc!@eRswRObz?TZNZ01pCB
zMI{~t_LLNmP%{5(D&Kww8`38us2sf)swTiTs4rWqYU$N8m$+EFG3rk}qW{(-qyQoH
zGA%-&I^xN^@Ivfane2J}P245LVZMDCfr5%4bL2>b+fCWhpV6e&XrEefu?ypv+U)eb
z_b#eeE~j|bESzo^PrvdWPrdyqcV2ffw_kG^HpT3W>JMr8Kds)$&W}FDQ&mUzD=$Mh
zVvRFo##s8TZ4f!o)9IgkbsE@oi>@l*-kYyt#3B7c+B9u;h1BHM)LHFNh|2Zqhg*PP
z2dR$Et`IGSRXB?kk!cG;k;Jg#E`<;=Sij&$LuV_QVaJJ<mzP`K)Y8(@TC7lWu##C)
z5)Hm<4KuE8*sx(NSg?R=ue}zB!|{i4FPbrgHk*x0FTIpSix!cWmj~$XS0M!LRo<y1
zK>_T|uGLn!H9V!|C^_AHRx=~b(6hG?&E>(~HJ7Gtx?qra>g&;LD)xj#>bGxa*VF&T
z?)Dhl`(n5B)%{6<?&9-EiQ|Y7EM2vMqmCTfXjU_PKKjrTH2jMQgCZ0Xk3O1=NoTQd
z+BE9cuJMbD3r8ZPjXDZdu~E8osZI#J+qZJBbjFDqpGfcf?<fD&*Lm=s$9Zkee4e}e
zdTzY*JXGT%YV5LZs-Ye~*-*HbZSTBK>d?c;I%NWiYIepjT)p1E@RB)AH#@T6hovlC
zzZFHZ0TPeUc;vaaS^E1HV_WH+%V#MRBqqcWYmZzRgZ+f4MfeIKQh>i%48JC&45y-X
zCw#CvSCOd&5x&fDUjjn?w%Su?n-C%_6HBP3?}>?#k4~_Xxj&oFCnqQK!V53(=9_Pl
znVEUeyI@Vinvs#g%P+so8*jWpMn;Br>|o#jP*4coqp{79ejly223c7}e2z}qz-YZC
zz`dssx64Ib*PO85EToiZNEAh|dCKEBib`}M;I+&JAuq;%g#=9~qz)U(udDMJewdl1
z-$CAL46+-W+m-CAuA=&vWmL_chvxBEW^e{>FF=9RBacMYBxS$+toM>7FxxlqTL^Lf
z0Ai7U001BWNkl<ZZeIbm-o5##>i~XU^fT{0a3>d>HVzfPkVlh)StUS7N%=3!D4aQ)
z9_OD+!jM7c1_AT<$HGYj)msY4f9Nq9+)kc*<6~Tk4Jo8i#92`1GF{Ai$wA+QhVc?f
zYEt6CV2Y3C4%|fWtoJ(L$xs^#AyR}8eS>`3D=W7n!s=z&6@|G%$qe@;Ak^DVd+Qt)
z33a(#mbbA;)}S+Cw+BC^`-{-oY&LH)zy%jvz&GD~!-*%J=(W_DR;q)+YG+#Igb?KA
z<uPyGJg&Ls8tis^z}UgCudRizp=p#L0C#mYo(3neIeM2diE(lyxQdH$A&AKeH?c|2
z?6gZk>ATZWR{rj7{U!Y4<$~^(KtsV9V<)~}AIi$A=$UJk)V2R$CUn>mlF0beWoU7B
zO5S=0t**uh0SuiiVP?^G$;(4?d8k;v0?k{FL^fGB6#RMGp2B_n-}QI1p}v+o#~(w^
z{BMv(JhBx2BwqS6NPS#NwCXx`fAAsB;xc+(b0zllG(U@Nu*r3coDdM#Et_r^oy+|1
zmhtoIt=<;`;e9#KS)hO1T!4F55oD$%88LnbG@^373Lz5E-L1h8j&<&8^un`^KX*n<
z$pa5oM5?0_x=`Ml_SQKp&ymYzEg~A_DmogWgFv!2XwV=&`|LCBzyE#`67+CeMNtkK
zt5s}lER!csX7=pa3>`WY;15sY{Gkw5DrL*ze|X(>wV+5uQi2~r(ARL6l_BC{5pi)r
zwhRu{;UGp;X;}CzHq-tjtlXqcBf9L_In+o6ha-mgxM+tYwE+SY#YXCh<B2<F6#L$M
z2iKlGNNJjtgJh1HrZQu<Ge;joLsbQptJmQ1(yhQ$h0vjk5~RlJ4O_VAhJTQm(S`S4
zdXkgwnM}#bm6R-8Vz{+^pTKL&=&8GY6WiZ?pY&r!lXcq3#+#$G&igQKK<56a4hJ!P
z`||YEkI+;FU)lqW%E1IVh{oP!rKh5ZaNPPG8d14k(Gwwpz1?(XwzeKtq{JPCxehi$
z7p{fQa_o;_p)(<+^jhN*5)!!YzWey>v(FefaNt4H_Nr&ko=lrIjmI8)jMUUrZ>(G=
zd{BVkIp1wWK+b4GJ!)(W#Kd@Gq)fOgDzGOcf*Rh%Cnc!KNl3dL6^}1ypKx7>5ZALO
zdrK?H%1XDWJlAm`jr~g;av167pUvKB)2YwTw~UC>`Hw+j1V|ys9eW&g1>33Lw%u#L
z^vd_&HkdqM&Y~Z=_~r*W>9{;ze)2xjQ&X{}Br^D(+u8E;i_{mF1uu#sCDQ4laOPYp
z*R7-Hm6sFSw+A56%sO?!cEay%(wmu#<ipwXS-EMOk^L$SBIi#go9gb5+{`X|j9ri^
z4<JOG3<x3OfZHvGr%EZKQD)<hjA%TCLdgvGHWx}phtxTg<w?^b|L$WoUo_@w2}~=M
ze?7MG<Hz&OH{Wo<1s8aiA4nl5CJ|K>g;P#Bg#`;1aORn3dfm(i(!R786d?Rp3#s90
zXh3>2T4KS~YHCs9;*CD7epF9_tg1pyi~}{idyYU-SFj6vVuIJ@27#6t-SaBFNf6SX
zT_T9-n#+zo`{>a<@^y+L4Iu<l(J$7RoNjcx{?8OGSV+xJOOcwU*R1od9fZ^fKvBs#
z<y4A)T#j>ZQImD7sF26w;iY#!<<7@n<e#@(!F{(}L7dH(5u7+|2$>VdQSkJONVg|Y
zpeCu?v6F&#KOp&#!Q@Oj3pL*1i*xW!C4Hn$>KAdz`t`VKY6D&hib~3v^71r{hOIs|
z{gn5&G&bHR0d&vF^4q%u7SS*TTu1O?uJtq%JJ*Sjup-~VYW%eop=5^pn+f%@&>?lU
zvX}vFZYm!PwY9Y^)?<G`N=a5$7H_}(HqSr*yw?SLAR#_Jo<|;egpWV|_zzoKt|d^=
zZ`#d50{g!Ik!5F{!=|U7!s8dO=0os<E$eC#@o@+>u(i9860NQtTYOwFUs+IiD#|H(
z?_F&14y0Xo+nKRbrUnL=3)D|~9fHK<WV5fhirmaBEKEdU>C!Hcek4VJl8{J`Yp<bZ
z%O=V{pM}Tm0jV{Lt81jc5TGjbJm)luzFCO7qB7_n`V+&-s%mb3?0G(%_bt=@{Rsaz
zaXhM`1{`Y?g<jWOhPz+~MPDs23~3UN)5YE|zoG2MU+8+-CBzTt8{+huy><;No0OC-
z{}pFxnO7&NNgjRv9f~TdjNP%e#+mfiw~FyagKjw)VgJrLFe3Vt5<<iRf43O!w9d5+
zR{p?f)cy?hv<}x@8fv|05V|(&9EwO;k*~JkKq7QzSr3Q9!FAVN$D&1x7(RS>hmAS&
zPvTG%g+mTGgt>F)a`)YLlaP?mne-3_6!hMV%|b{?^>3@Nmy}VobR~)=gIMW;KWJq`
z1JdEZXGJ$`DN;i^UC6k&@J^DH?0)}$#Oy0Z*i`IEiI#JEGdhjTP;Vg%0cv^%B^7ma
zNey;C=|E)8gy}{VdJR3r7DM($7a<&WN~Zr0&F$7ZOUNc0c_D;AIO6Df+Q}5on~&yl
z1-eOvZ@Yp$d%66!2dJ!b@!<;(F>FwOg9kN@wiEFQbi3n5DqeXV*PbG3H*8?f2k(>6
zy*phmI1ig6)~rxtv15kw*fhE+3jHs+g!mqM7Z4#O3w~O`v@gE}BN4*X#H&NV|7<vH
zf^kO;LNnwFsUgtl)h#Q?kvZT{6JdE*1Kolw@3hXf9af~Iy{%mxmC&_WXS>~Qc{%^l
z37tO_MPb;mVSM$~SKN5xjoy}69SLU1k(ihmF2DS8zWw%FjydL-P6F2gD1=#GNFwNV
z=>>GT;jbL}$b$$)H1>E|Z|ibHY)qhW#LPr=yAfg329rSDmTgph{4t<+DX`^qM;JBA
zM5A>Ze=~`LM8lq(NNs&B@$oVE{B|AvF|TVc3{omomDE#C!_mJVdtZGWX}09@`Y9V9
z5eY(~rlyjaH<G>c780<5VO{(2*Wa1+*Sk33xTAUH!CT2r?-C$K*8w924x|D}hYu&|
z#IfwW`EOLOUPsppFCt;+Apxwb=_M(B(@O%mqT+iCE3c~N-Y2Hwa^tgBnj4y`8-!7j
z0#px)3GtkC{ITe<diqT9F?O;u&C2ly^bMjaQi6mKcC52dU+=WewHcj+E;J%R=q$%N
z37rpmJ4s1ON(xUu{WKqZ@BukS)rpRTjEoFkdg&#mPMu0-X67HVIB;8_2q7xMSy2dx
z?UGLKtFIvS5WRMo3^!v?lL)^byw4aU5t>GGH6$ex;nvvm)?3(WYLG(UF0CYPa4+13
zw8o`#g9YhJ^b8PNTpXI)g-!1j<!9mP*rZO8ikJ6D^mCG&cQk3^kEh^`skqC^(9qK$
z&Fs*|dn5%odUYqEM-F8_F2(Hi>UO)C`oYKi{jq2HfA>x1mdnp0*4FeUWs*5Xw|}9y
zCAAwj;qKEDw`#}UB^?px2&f|$6cQSWiYQ;Zrjc}?-42h&{m)Hh<IX*Lc@o7FK<dn_
zR?`}rot{K;jE&-odhUDj4Uh_o)L%x@QWA}}Uk5zOBPi=~pm&hv4Rl)PA_1$H<@P2=
z;T8pMpkt8GdMRaC$$CjiNgcvM7mXsuk%XB<U?xmVnly<83l?zl$tMTYs%vj+n13S2
zs8OT%>Z`A~;)*K*YRh#Zd?-@NRy`m>m%3=WvhX+WHZmTC3hFnRLG`+|)P4IMieZgw
zC@aO0l8l<!g{JhasT_Y&Cv2)7Xov7O`5aIXLMLYef#Z-N<eq&td)|K+ci~>Iwaqkq
z1#D;hm3;V6JauK%Y}rgjRSh@a`xrm1TE|B(KgrmUhoktpH~k^eU6^PRt*o3Kvu06O
zT1K}s&ZPUjcd_&3*YQ+VH<~b*6sXM4r*ipHLtX#aN_~|X3l}qO#v+7J4KgQCZ3^9U
z(+ttl*jA_V6UL5Y$bjBxxB$gCM!HS0dsZ3=jt*R+CQL{vBhm-dsKkGd#qjG=%AJvz
z=MN66eo)bfP7g)6!fM+k^%PF4b=y<tuz%y}>FMoC=xS<$M?6^dc<qQVNmAdwefi{*
zPk88|hlr1l-!J6Jbg?$BE!T<GL6K5MzX_chqJFU=z)1P_U73y6^=Rz<?|-p1)B^&k
zfFdQ8J9d*W>L@&>N3qfN!D9RbfaXC}?Cj@E^ESiRw^k6-t2en9U&_8|AK)t3iRL{T
z!H>HT3K_?pz;8b+<*zr~$FLEH^Xj8_k&~GoG;zS^8I=f^hkf6F&&~x4$sT(gT}F>W
zsDk*z4k2aOPzv9l)+qb7$?PBmnPZNl$7NR{gigNv$^EA71>E(o*U%I@2u+WDM39l5
z#GQY=5(R-r_{H7<?|!y`AAZ}2kV;^pjDWtqdmJopd&Lowb1~3A$npl@i3rW{M~BtD
zHSFf|Er6B(OrvFeSld%)t7`IL|Hd;iGNjdYy?5{4*60)xRx-`jQwIY4OPn|y4(__^
zF2;`^zu)vZ>ZqfbGGz*BX=$D1NZJSmXj;c2!Vq>lQU+m9iU6Ap+^$9s|G@0j`fXdO
z`ThsYwpIc?cB^XnD!Pn63C-aM+{WzX+2lIcovWVu2JAK)PNOL0foGosq(n30>Ci7y
zdqyT*uf3kKnKNnFv_&sK9OTGK37Y2Si+SJj-?@wV%dw-m?wm8Q$CxpNe(M_+Oo0SV
zqh|9qwtx6BaXH=Sb?!OXyQKOY)`~*z^;c2y#eC{EZ^h(W5}G7ofZ;Ax1ZGKSspoxA
zT2;vn_dP>VWeq~n>+mU3K?+IHzA|om@L4n|^_aOv`=)pEl}KSyT7i&~!F{?0PiQ{y
z5Gg`;hsE*_I<0dNgO#PJdlcphme>MNCWGF5Ezu!#4p)H9W+OQ{Iq2A~U6Bs~wUU{Z
zmKO23q79SSx!rD-En9|Wl)-9mtX;bnr_<>z6nv1pc<&b|ilRjRHO3DT5RO=crZtUk
z69VjZM16xLtCAY*{p3?(8|v{-!w`U0T1;I@35jEm*NKa9+p_#k3cXOFv!;&3gg9#I
z+<wvyGzb*gsOpLUB{7lAtFNMB;R0$_tu{!VnSvk@c&MqV;;F}8VZ++3JT>(xjym;t
z_AS=q=}cFyuXwK>OzSGGVE61fRPWqHj|<KxX+VFU$J(#5vn?xw?ti_UUC+LPTWV+$
zw5n>BUwIww($Jn10n|I4+&$$5)@<7eip@(_%?xEJ6g--5S-_bSz5^o6pa20#jsAUl
z`v-5f{;=<7D2lQ#LZYlJI=dqzzY|bkvAmmz{8P({a9QbKg&i%F-xl;}p$@@?^aw6y
z=g!E!mv}rLcJB^;dGzerGoo{~2c*=K3l}e5%)Wj5_FDo%Nl6Jm{P4qmnfqXajdr+!
zs|g7REl6ye5U36Zo|0m{Ds*G(5E8ATlB%!2!D~D6ayo&Eq;%#?awbn^+Yd{y)#|ZZ
zrp?Feax>=AvtcFfsw&dclPNdqngtCuI{+b#j7}vkj_f~OPVs;Lhp4V0ZcH930!?ac
zT(_0CUVEQ&|L-i08gr!Xnv9Rbws{9N8+MQ|sHdJeY|3<N?ET><+%>giAAdYHGqFHG
zz!?)>hc;bKJ(c2F^Qip(CsL0)1`+R|-{c!mQ&NKKt#6Lo<L3TnUZ=9o$!RB!=F@pg
z@J4Hj#wV$1(ud-;NJ>2>&oB|j*cE#B$l;(k$<SO+)wbeNA;i(G*v{&ScA*ergvHp;
zQp)>VG1q~DO%}`J(8)}Th)XjVXt8{A3*}g?KTrrEWQ0sOLI{Q6E`P8-Zb8wnUq62T
z{r4uz^Yimtp_w(dZry6xYOG(sei5Cky`ZWpGiJ=d<B3YcsIc+rr=K!m!i3JO);7Wh
zM7=bDl9-4Zr1z*40wpO4XH^YC^BC&*yITcCUww_eqS9-f^9Iiw_Z8Qw-*N6OBIWFJ
zshajKNS_POSTW#VDJ5EYDcOU2v3^UT9=9b9jz_&PQePz?VjX1u=`zYbcn42)3GpYM
z%#7)uuzAyV?s@PwQqq#WB~Sn|jvK+w8FPs1nu#qj4m6GObsMSNu$hd!k;HY)vN%Q(
zAq_c3!iv^VPwwAtWY?4@Nj~Zb)K~{;qeeC%>|VXmTjxCU`ZWGC<7=-x-PJ1e_GkX?
zW1pecPYXy)h^4-v4yW5^%N0OMVmv*&<p$gAKt!|&T_58UDfA|ICN1dkdIvK?k*Og>
zh|n3<zC$gBH?~lmmCh;XF49Puy?SGvXwhOq7Yu9GtO>fjwFo7ze}5}(+ObhzU(dXG
z^Y+{PrlETF?Ag5V!VAR2gxhww6TzS~G&Ho-B;WE$;^I-zGb0=M$WYVMa97tN-JYf~
zR!%pib7!NP);r;~*BC@ZplTk9U!O|Gxs!1B>Vs%pV7IYnLx8)mklgGp?Al{`n-7A=
zJb*!Eh3PI;ZKPdt6(xnGym`}oq@?9=*8_KulIF9rfqqd*yGqW9$FlGHpJ~`r$nGz{
zLGx&IKW8HG7GzF%ZF17Eu<E5o*57m!NR7B*hf_Rbj`v&zg9J29^7hBGc;@ZT{G{q1
z%{bWsv4H;JlHO$+_up~_d50f@$E=YjBvQbT0ey+J+YXkq9QEt96A?o`*1fW=2v}>-
z4X~=ov#y1*)`l31q|Sp|sY4SC88Rg3^0Klr_Uwt=`_KCI>w_-0+wBY(Fd$;{v=^k5
zEL*mWf`a{bOEC?;g@uJIUc5M9StnW!B{nv;m<SdtZ43dZadDsu&^(Q}lfa&tiFCP;
z)m8XEwk1%rX(QSO<At$_%gvh&fVOcHrE})c{nk5hIqcv^2~78_|2D3|Lef)`DA`+d
zko|b{<L5C*k%otDTQ;+(xRP-v<#9}}95mEMYhzf}1R^dTXW3o~XU!w)gt4Rz8;sIi
zsE^c*rFtH-Kqw0R9-Tts;6dbGc@ZU_e2%BIJP4hW-i`k^`%C`u?E6LtqzGusEdst-
zS@?AQAAIf|zW;TD;WE|`QlJ1w96InI==cRd1(B^o*NG@XN?C(7p^9XEhQmUPEpMT$
z->|Ml*B(0811&PGjTm9&(I==YwzaU5HDt&T91cgs=4meoA(%OHCT@4A7QWGkS@d|u
zj2UPLdg+P%472rDR1+g?aq$qV1~za9Kw{6x#3sO5w9gyhuh%-0lzh7gn>S`5ka+oG
zw**2FRKD{TN}_|bE3QUURT>qz^NYBW)b7|pa)JY=N1?Xv;PLuMp_g;9Ne^glzMcOA
z>o;uYvYW4Gz@$mISN~4!$~9=xuSSRjG)eh}jTA0jN%m={64SpgLQydj5rWQz^!Chc
zWOT*Ry&H<%Mr?LAY2(MS_q`9%WZ>3jU6Ie`eaGJ)dz}VN1B_mM-u%>H2u2Nc|7>k-
zy*Cz9GwbL{jZwn}wVO*_dOty8M%xG>L;;~2sCS~{VVTA9(Gi^ESc~P=Kyb5?1*DV>
zz-o)(V_G7s8CdE3o0Kx#&K3X+9Xgcw_~0e&A{lJ|{rBGoT|RQ;$cW9+o@i)j;EON5
z*xzb+3Ek}3vn^v3I}sif^f#hN2In?~!-1NRh*lHWS6l)yx!KrbRO<E=`aS+6T7!$~
z@4iRDKd;pv0x5h>Kc&IR?)(2q`st_Qn0Th%no3GQ!CPWifTBVD-XfIRYLZiv*i%$`
z@OWIM9&1)yRL-nVzM^0MzDzuGJhm7cLRColzezaPuBB$>YH#n?`r-<TmaW2`oJ!ZR
zqlimMCj00-iho+FlOt(yc%)$ayYEr*-S@~w<}dI66hYR-=TW(MF%1Q}>%)|#X&Rr+
z`<Ayqn@fXR!^a%0{x#<fvHxb<2y+uDQH93H;X|zDMG(?(Jr{4>WP}hzb}lL9Rw75}
zD$vVAMCucGar;moVw-R~7_=BdhzudbYt4Kfh}Qgm_|9Va03k%AYP*fKSiU6Ojth-o
zrSs@WWak=1t*r=caMlz8s;V-2bZ~KzB6QogZ)f}V?Lo(mJMOrM%+X$uQtu43YSk*X
zY}vBk%Y>MvN_Ott$<n1u4>GS=`$7SbQf`RmBvMhZXJpb)QPw29P07kaiM8R{xyLZt
z=vfL4+qUA`z75lA<F}QcU?Z+hC?IUucI;r+Bmbi74cDWdI032X4oIZXYp+Q?uFB?i
zQl7t&LkDzY<?4+G4WH9A$&&AvuwwZNCZ0Wk9=*HzqTB?as-#_bE|nWMQ2Fa>O4qHW
zp{kVhk;6#HNk<X-@loRJ#P{w-<@!xXvj$xN4C_<96U}RN4enG`latB0<U;mNebZ}B
zcY8ei`^^uz{n3{>{iGATUUA{gW`)K@<3xbL%Q_y<)Xmg+=)m5jCM7gkD_W@Dxd+89
z_0A{}zA;7eI+aoml2Uew)CaiFVyb_K*zTf8?N1OS1J^b4`M)U6Z{iEQ1(gCIfD79y
zTnNz*YjQNZm13>7>xwmbYzfW;tXeC~YK_3tPd`2A@&yYPL?n*xop;^|x?ELNPCM<i
zHp_2MNGVYig_$#Fb|{OO$K&Di&p+SaY8=ooqm^13I2#uH-J-UPESzQKO=6ksS(y-@
zK<$<tNHb?bO3Ifn!{)a7u?bVGSH3_Z6$Q^vKT+`F3-r3{@2JO*$0JmIOc>TCsn>6-
zSewt`hxTRp+N~h9gM!o*7Zs76o5P7Gk3)^I`PF&R-IzjvP*vjk^kUbWui?nfA*FY3
z6ve;5i-IN;5_|L@CMCg-{{$Ai69{@=|7WsJ9M?FqG4Q*9l+#Y8VRJq;zpX_|ji;wh
z<MF8<Qtfhc|5LAmqIi=lkR}x~-L8I_txb4bXs4_q1bM@UG#>{TA;<<NrTH638{97{
z1u~_SQMpb#5j{vL_W?^ShOaZqceDjJ6A*P8fY$oczC*X*2E$)kDbpV}Su8IDT4ho@
z<BT(cE_b`#y!F;wtr6)(Lqh|vzy5mAvAn#z1FB1e38&M^tXVpDjzrv=|NX8l_q|!O
zW_2i%d7wc7P!uJy`Q1Sv(lZbhO}ok{i76=QS-7`v^b#v6p=#L*y>Kr00+Rv+`dnxd
z!vbX#n|05gL%|Er)9e0wNjhf|PMd<FVKl54C{j?fY9)vC@4<!*+whpdU<Wudv$N@y
zlg;Ya|Kmr5B>n|=kx==?9Fj-nvFq#kG?Z2n5GW_4p5UM;#OLN9rNmiQg4W<{oX-;Y
zZzFiUo|;4om8^>{X6GwYdH9*Pc=qkj@CZeBiz_zo+OkQ^p70yf3As=NjyOHOFK}uF
zyCPA%Hc5fdJdDjdvR&fmq$Vj?{4?G1M(KTUqX1P^qkf%sA}X-5cnt*}X{}Hp#0kI+
z7Q??AF8UV3h++H6V)#NKM5`^ULWsWrR(+=CNhw<uQqaGDe+CZ@o;drjfBlQ4ORYF{
z^SJfaTP=$eTM@~Qj@7GIlb@g8YWJquI!sX%1`HU$z<~o%Rkcl|PD;tPZQEGByrp?O
zoiJfOc{ZUZM??n#?Ah74t15l*Y+elnF?|M5yJH79ojS2{*5g{U2DR~nF?cXn;8Y5z
zfHq?`JDz!(oSXha){VE|N{H2oi`1<^b!*lU+u)>Ib~@`e@3Inpz`#7JYJ431&p8XE
z;w#}|d<6RO>HWY1^tkd*^ga75iWhy0v!XgskTk>k1;XiO<pmeA>%-|S)(4<u*$=FJ
z;xQO;6tC|n;nn}m!i`D56)$JAFa!pA*pXy)N#P&2T!U2foM?0P)Z{pxedrbx@b<Ay
zNsML0&_P7&+?q#8lNzOK^GWWWi<%g!M|zu31+?k`-HH8<PeY8|D}<OOgb2HknGhmT
z2yqWE7qDBb_DM?=Y{B*dcwY$dq!1$Pepf<>Y$3#pz{{<eGc25U-g!Zn*Vfi@!U-qv
z@WT&NS7#OK*%bNt`J8?B*}U?~D?!JF5S)ANxvkQ5f5Oaebvm7Gl4G~qx$?>@S-f~L
zKmGI*lP6CmA;GG&VDnHE1((a!>2_@cgu&2bfP5)s?}!S=$D!)CZKeAA1!P=ywecV~
z0^ubTzwsaTOnr@hA5AB&R}X5}t!Kv-f5t8~yj|P90)6tOmwXTq2rs3ACNz*>%Ns?v
z2me9ChOO*)@=4U4+wn+21A<=9Ji(GWl`XsWa@%$P*KV%4$iSq=O(Wb)>y<QXiumnU
zqeF;RQ%~_%3+Xc9SZs+&ph@a$YH%bb=qqcI@})l!*S!z1-MR*y&00l?v#Ol+OV@DM
zOCRt{!9M*o=`~K8>Tam2pHiRhLQ0>74v3AlQQP1~MS97bf*?IDg?*)^hH!<kNA%~T
zS0Bd~rRw9-*vE#FQcBjWqw8^_5kl`dYgI(K4Ol0HI2`Me68jZGh+BdG7i8!ODdpY!
zL7%=f-z8y<lroZ~I6~<&A2>0@DiXb1$M@*PfGdDXz=>YeHw8V0?O0$G5UNgw3+OGS
zY;_kPA%q>+X3S{?H+or=CBSw+o$=GTH*g#<90=8Qy#VOj3JYDiF%Ja=1@!IPH)I)(
z`1pA8^70rxdNjGYxum3|ker;1)9Iw5qJoNw3i9*w`R=>#EbGphIQit0`SQyzTd3~-
zM@>x)W5<qV<;s;|Tj->e0dCdo>}(!;>@hC6<PvN)n|HmhzWR#ElP9xb!v^m*Ewus;
z9z2-k%a`MDIK1(God^%+nQ%%e*Pvf~5rOw{W}M5I^mLq+)nJk|Ki`^q$N+3k4>eo1
z65q2sHEY&l$NzyZd<hf6=Z4i|<ATgAG!!Y(go690U)Xlt4Rrt4gA9J-Rd&AkIyLiV
z6YF$Q^8F8-aNA_wykQDATzwAlcDujk9UQ{zq83K5nrWr;T0u3*u1}|v(4!A2d3ixY
z&_sfRE)!2+-`oXs$s0w<-rZyk9^@s`LJ_3qjrPU3S)xe+6QtDmxdCfF001BWNkl<Z
zaqSlV_SEa_DXl<J6f~pjgMW-%0R1t~t53y%TnJ4<U4z>oe!lHABt@m=_*}6X<3^7_
zRYWjJ;Y0c|q#N%ofadW~xMV5aCyoc97<&_WNGVqfAvz?P>!1k7R1np2O^xsnp$VNp
zcfm%2V7}fKG6%ZbRS5Af;KdND*?>Ib(^gDrt-IAkj~+d^`s%BhI(4e$I<>X6EL^yd
zg$oz9Wh#Ua{PUmxY?F-rg^e3GvS!Vi785NgC90}&?6Jr4{PWK<WXO;vvBD>vbQ1ID
z&*!eY?qd4%=}0MC)K*MN$>z<QS+QaTd3kxAWUhtq9xW+l>w^q}pz%*dO-RHR6N9I+
zqRDt-|3PRmDz(3@0ts~+HhSv_2R<By--m!Wx+e1r#Ygyd6tMl;n<-l`pI-OeMvupy
zMD5jw@<j_tz{BB(4C4Er*LdsU1f8qHz{us)G`7Dom8vb91M*OP|LO;@?)Npg%1SLn
z>Ey#6>mXrpZ)$h$pzF{f*d4J(LW7x55keb}2hDrI8opgAaW*vY^lQ_(=(fk%U0#i*
zs37zrGK$a;NEGRpktzfCAdujh<X-4H2@&uE@+bWWkV0TrVBDBdUfS3an3lJy>S~_<
z&vX_qThVkJK$GnJW)V4KkH%)V8>&U^Hxtq8^ACVQVXR{nCl+x?DZc~$(+*P=0Fh!E
zuK_<tNLDEP0K5?iL61KADBZesi-f3<`17Ct%$PA_qCQWg;7{IW&z{}l-mO9i;^X7F
z^Ugc@{PWK_^w2|t#>ojG$jQlJ>eQ*c^wLYDr>BP<PiK-tm&?V>nKPr3Saz_(OXw6u
z`Gp7v8#KLRSmoXOK0H&CU6Y%3uE{ngd*aS^lkJ*p+pfvBZQHhu9sc|N{+{PWFFKB1
zeAen(7tiy&prB_E2v$}&rNbbtxFF7P&0`WJFVY^iXUvvOw6WyAy&QtNqQAOZc??GP
zs;EleV1N^MD<8@t7Y^tyuEhFyIsL8CL>cZqvm&2II_na3@ZR(6^^q@ehO`((2U31{
z)9-QN66YI3ib3*v-H_2SPNO1oV<&u%ozeY34Z)2f2t{#1$&u5AbN3gg49tbEfms#u
z*IOs$9zCw1$?i2fZx?ONMz^HA582Un(u^QySU6n5NLJaq#U_kSjKeuiznIQOgAcN)
zPnWslW1R4ZJH8xMRp`fy1tVSpiIc-MUGIbXva6$q6Nlrp^dHq7jE@aw2R396$C8$R
zq?T(n;bavOehV!zXUvYfB272m4<+qYbuTF-bYXKUNv*Jg$}t&fO^(nDsbQhLVJ0n%
zJK8CR$H(`<Nh>8-2n&7J&tSJd^uJz;LOcM;EUk#LnJaVCpm!pcTvS0Gh$*UYJXeg8
z?Rg6q&3g+AW*u&R3mIe{7YSWnR;z41r3OEVAu#_ba>B&IncruY%jHt*^z@Y5VzvSG
z<8fR!RM9g^lI&dD!44CX4L7;eZ*Q6yj2@3Du8nKeZu3MI@$uI(%+;T)@Ij*z%HJ8v
zUU@9Gpn5&Wc-`{E(1!joI4%6TU2d)?Eog^h52Zlqs_3UFjPV-1PUhl{n7VQd4~A68
z1No+e@6dkvOc)um;Tr=aa-0vU$ebnwVFeAAwE8!#noGSlwrQUfg=VjFjSvArMZTJI
zt}cUGlkJc6SQTpE?17~U7FDRJKa(JNT#XHRVi^UO=r^UEIO2u0Rz_Hm-wVdW*ad7!
zypE=t{cUr84b2DZBA&Oi5PW#}%bI}hcAjQYwf-71;qw~i{biCgScgRN;Bljw`c#su
zUU#n2F;<+7&NBF3JV>GWvD`!@ox0G{oh6K<AmO3=q`U7JDw#Bt;V<vc1Dn|If10g_
zzSy9*XX#kvepT`{EKCSQOBf45BYl23%H{$KhmihE0>>vp&uc-G$qY$LOG^=Bas1NK
zJafEscB|c{EpH2j^K~$&bOJoc24i%E!Yi!$f{+X+Gq~36$LRXO9kYrnaC=C-&Dp9r
zurjdv#cL~^Fap1Q&JJI1uye(}P2uwlzGJDV$(I}UrvwgvMp#v6uEB)+IU97UnL%L9
zuf1P1;S~jW7|<1!VYf%R`38%sjxCq)tr7S`382cHB7XPK&fMt~6bae<`kcS@n;l?s
zbrIIW&aax6SCi*3hvOlaDbST-OUGZqr>$U}WR$DsPu>QeCZ+p3hI`I4%LE)g6GpeU
z&`naT^i1g{7=SQ4Mk^80PKhvza6F!7Bev~l-KYM}6iv1H@9*T$6Z3sUd9aC7r=_;|
zfjM5UFCqV^^>=TE<eQs9(psqllw<Qw97LoXrqZ))fUuK%2hUjdHi?|tzxmVHz`U@J
zzj?Lr-~6N}d9x&oqYz}0V$ogP*hI@zW}fcBgre#1U7I}$e^CwS<O#@cVtWmU)NX}6
zn$Pj})QFU^gWR`IoM*|uh)Kr41fVkHU}F(}CGguU=Vy~)`hD;5;C?Um2$24%Dk`Y`
zr69q$J3A0HI`7Prv$fOzukBo`#b3|8I`kC!!gWF@r?3CLGl!$egU>pb&9BbDgH7+v
zhN)DRRQ?mbZ$gy5;MA@o#Ba8eiUc03GQ<@V(^BgzCOsCdTDM;wRoEtoI3ViliPRO@
zHz9XsNs198id5%GovWvz$|ISP03#E{#r$|K)9j8V_or3i!{nw4Y+cXTHS4y>C0(D)
z|5){4aiWKo*1FElEWW4pcyJ;cjLTwTyQpr;WVu}Z!Q7%7`fYu~WlL6ETs+!u5(fjD
zSy@%p;UH1f9*p1Vhj4cIrjqcfqNLRK)cM2*p6=uD-=$7fd%oPdMrEB-eTQFnJIVw*
z|GtA6D`>^qoTakUKuwQ0&Vi2n%mtUt$!ev^U-k$3;h{wRi;s&Co6p<o%Gur&A6|9C
z&+PQB$B(hWi7{!FyYpn9l1i#Yb9LYG+uTJk%co*cvPS~OIt9hWVTA2lwu9-~3$D1-
z(y#DTLhuR0rQ{i7)-|qIsmE+yNu9(Aejn&4Nnj)P?*emQ3bB9IJ1hDu6W<`Ki3%~b
zo1K)a)3$wZKYosFQQ>o!ukq0EnXIi5$M+PJrhf0HKGV>6zNn~sKG40=H`Juex5~O3
zLfiw5;h8^;hTRddS+eEbz=~!M>6-vpE!hb+Y@P%lXMRw~t%X!?SsSLjA7$@7^bAy;
z-p$k4*e*od?CMsvi!Q^(ZVWEocGE4f)E7}RDPnf<O|7}S`)~Z_CuAw$9cMdtN~04K
z@&4jj=wrHKhZ>vxMfdn1;SBhY=UZD`6(T{$81Rb>i0%kL_3NTYhb5uH|C>ZdN3@I=
zu=aNWZ8@EEIg?cQI@`cEXW=(bPou5us=!TM^JoLZcX%ib4GlAM^IUnVYU_>GW;~YU
zF$Dz$L1Sa`gM)+l#YYe7fxo$>`*&_p0w1nfuDhW+QxitaGMU_oXH{KU{>a=^mt0Nc
z4#KF9+O<Y-*xHV0b*58;Pw#}HnIYAmAD}yMMQYJ@7lcev=yypA>~OM>%hT^oP#bH7
zxy(2CJV}Uz>s6~+{4;e2HZ$VPoJgdeE%*+%<5mpEZ(V!RKEM2wtWF!CwUy*O^S@J5
zFTXE#U2kUuw$Drtrw&aS-7fD4qT_}Bn`;)T6ZdCF2(;}|9&q<d-kfEv01ElyeV+Xb
zheO};I8UYFqeO2D{LQ^sMaf|)!m;OQ)szcwqx%Xpx)q`DdCYTNkDBE5X*(kJerx+Q
zRNv$N?*Ff_*O|GlJtLSK{@Hcs#oX-&e|9qw$%t13Jl8x0M5)Ai;<}xS4;qd_r}s{9
z%QFVxQronowzMSG!rK|6X%hr%1r@Fa1=c}%dAq>g+PTIB1!oOxd(x?4Z9SmoN2Lh9
zRKI>jfC(7}xgBYLx=;(n)}W>=_x*I!d-}LX=7LU`KzHP!Sp+W{RO_BEQ8Kr=;#$>J
zQ5hT(_^_+&eEoBJ7SHalDLgH@OP4+6u;H|HvvK8}oRpMA=u$d-V9yH<6=2xm#gD0u
zI)&E+!h8XcR%ou}Ip^(Iewce?2O~vtuEF5febw@A>G!P58&wn9M|aNr<gEsHK|_tN
z`P<+XGp|!V2g{@;*W39HoCo1cb~{hjJ2%)z3RL?TfLI}hkG!}M50qX}gG$2<Zj~-J
zfaUBp$?dC%zHG!d(84P}jdg3{gB%~=$P-F;M=AswD(A-!+KxTsG;Debd-sxnoeN9c
zgFf+ub3;IdQ4SILV-A!q5JooNc~w~?hcx@BaV6UBjgtN#T%b9Y++uMrcw$dEz<GqJ
zZKL=vcG#T6c;S)^BNdnMvLU^C0=m~`@xt#a_eG44y@I_sQlSJj><>u6ljHO4q_wMq
ze|7v*vUP;HbX7R`mZnK)V};otj4UYY0;rbyzq1-fcY_H{m+w0un3lxT8zC~k?Z}94
zafW0K5Of1FQXI}LaSV5DEkqKIurQe;@HwKuXzbYR_j6M)t7~AWiFH3UYv_)5bZo3;
zKTbLYtV0F5Xwu~sKM{!G;o(UI|Gv07Q8~Jy{?kj5#<wC(89|Vf>Dm#t#ct1Bw*jvq
zNkT}ZHt)4Mm1xKR%>Rj$POSAMhZ9CFEP6;uq8j$>vNe2sRE++0IMmTm<MMOKA_G)`
z&voM<uszhYK7=2xx~-YcIicuR?CPi3*47q1%|kA)qdRo{v(4bk#d%IcZZ6rnbgf(m
z&S>yLTuEt2hw6<gP9rdI@?oO><HSTG%Zp;zAnJz)9@k)4XMx^~pWHrHJ$fj2PYB~}
zdGz*$50OvjyG|77m|W|6OTgr^*VqnrEnrj)$`f)ig-j|{CwSf-siI=rWl1OK=x0q|
zB`d%hBd@(Je`uUH<|be4d!SH0W7TU?v2JFcdG?15zt2OJdV@2cAggo@368>V-ap*g
z@5{kqw}<8MV_B|YqcwZ~(pZT<)Cq~QijKV4erR*$Cr^29gCLDV`f4h+LHyt}#{+xj
zduWAW6ju-oz%!<9ioeYaq3T0Og!8<*uNlk%ds!ah7h_MqU@k7TKgx8OU0FE^BYl-H
zAI98H^S(IOp_+c~?wAL7(H7NGGkrTYQycL^$&@Xs;2SrjL0)hY^nh(i(u$p2nWgfC
z`xfcu?^aHIV45{jP^Hz&rQ4Q;y{UsXS)oRI0R6aT#Y;)Z9%+cO8m>(lF(^?(bU`5?
zQ2XVJreYx8_ke<WU}t<1x%9u(c8l{LQbrayS@YM48njq)?PLi*^?G?5-ddfEO8*e0
z3x*j1?POGaU$#K}t(}QKm$Y#AF(GD>kizltUhcF}?9Vbk^;S-co(0P5<&jHyF&;AZ
zzc_;zfyv>G((eVa*xX1dx9@7TITq=ydM$0V79o2Wv>ddKaqU#&j)<V{6SG(Dtm<x!
zR=5ZViJKD62fSCjZ?gqy4=pjDH7zl_FHaxCd;=EVoM1CHM&t&cb8NST{yRMwOmS)7
zI+_d$87G!lzKRkkJ#B$BkIvt+BVN)8=}iT*+f7Fke%5e?eGpC^*n1w(=<du)GZ+uY
z5Q{E77KcckMG570gLh>2g4M)ER~szQ(8?PG+iGOMYd>6l!N~^3)R|R28jRz^^w;&J
zDi#yoQSrz;Lu2qFQm9%|G5N-!taMD1vv^7dudoD9)G!LPzSg$O0*4(E`j2CekrYNF
zvtJR!RlN>2^gC%=TO%ddpMP2CKfVgQ&uxI`)SjydP|8-%!f9%isP|DAJ@AgK?6YsL
z_|JLIlBA8@(o}vw#}w+$tf{8&Ps`-1b4PMH=)YG%A6sfn)g5^TCCZWG<4Q}7o16xd
ztjO4Ulq69M=nQmYs5XGUQ&U8SKPYTAtv@Z-dA?aJ**<gWZm(`_4c-o#^y9E<$@m-O
z5!rD>H5B|NheMJ!D}Znd<VY?TW9;!VFR+}0cZrv4JYLwBG)(0({8Re@pBc#8^?F7d
z1sWb$_XPPnI%D}AS+TC?(f9}F)~+96t7_wgw=h~ou`#qqSxqSU40}W3nnW@UuJgdo
zSto}Kciz(;(q4CbWgXhVQI<-FJGy(>2ke)Esn@AppYK;%&MU^s%F6vhHbi)@MMu}X
z(O_xA<dr&8hwB~~?3F7Jf1Sz1sj8+kS?k?2u<Vo|XKU5Ug}prIy~0iThn=I|H(wXH
zGTP6BD)aXZ03*48+y+mll$YIx$FbjDXcNPvN;m<pzOYuA)L}~g^4{u|!$`FC!DF$`
zkdkPmE-fqUB$BfPmAr1&P6*`VVF={YZqB>2>Dxd3{n6e`NJZxEpd{R!I(;{y*4{t5
zf^igb7Z<echj?N-uEd!6;zJhSR~Y_IV5@m;fuF4%XKE{1u{2bKEO1=I_2z7=pu8x_
zi=i!H_Eo$00i4tvv@C6HZ;EFhW{H1Hj5bW@`884~RO;7PV!rUQj3yl2O`Gt>NZ2Rw
z{sfl)t&!j93yCm}g~G(B^s3Go6f1+k0W2U6$klKd7wa>H9PnWKWK3l^Ktnt)-p)Hz
zT{i&UgH*S)us#lAUGJ4?{fIQg=;tP*89pHZyf4ipOiho49A?@Ladtz1K&Fw!Kaj&~
zB((s+#2iD6kYOK4J~x)nqEb_&0=}rPJ#NSD#MC1H^%7J0JlIR6DB^eS=j4u7Evk&c
z`SKT0Pz`du*U1$f9i8W0`qn;p!&CQ@mq-AZJDBnxr@)e(hl<@=&hhEHrt(8J$?eQX
zioEan^-j6ZiD)S|H&L>5;6-=Yzy3d{I_v8`WRA%}3mg!T--JPia<gw$#V_mPn|=8V
zlf@-<{t*+~(qAzL{=o?~K%_>%;An6sdSqqw@qK##=rKJH(RfT{j)HW_9<T49s5@JZ
zMXahCxci171)-!3W#Bg|6~76ooYS5msj7pUlEhH`agbcXwf!Z+bk~6_U4I|6g#z!q
zC&pf>CU{WUMl2>&gPuhxwKANW09@*af`PAha;()mvCndIBa9ZfgO}^L$17W6)kmF&
zB-NQ#gx)2A^M^>vefG@zgJ@HsEm=l3_49`!`6D%i0d6X0)W08tc@2?0kYNny?dAG|
zn{7AaG+><$;<$hTrJ38TEY1+d@}e-7Ef?M|;i~1kr`eyOYbFA|4FP=75`i_C!l))}
zI+WPn_m2G6n}+e560jD!S<)VTQiGYOAuaoz(8$2|;1EuxB^US6Y^Ci}U)KH|97__W
zrRC+}csc&iQblm#^Pe~qxJZ|yh{wGrw24iof7Q{{9Fz0D#Q+OdNbrNDv*y5m{-2~M
zSW#%jqADxnDk{X<xP|G^8ORsx{rrQ1%7{53`iJn)t?@p^mDnx5CZq7Ef<>%%bTYP!
zWB9&sY8CD03&Zl=pzdAh6QI|ri7zTFI!v!TfLIwH+dU*`sQxdSJR7a%!gwvFEbsNR
z#oxjr-m&UdDd%j)qz#HFgDJ^993JOydt!*i)XBL;jnv~@(&!^H|Ju3J7i%7qyK-K|
z1p-e?)%FYcd?KIT{0{_P(LkF{JS%a7$s|D?^qna|iIKc-SIH@Niv~I=JlWVs<N3LE
zR;UY!+x-1BrdN|HVv<2V@IAjV$y=<T`${8r!XuIDjac+y*|ns<)|4b-_Ode&hI<;1
zq&57zX?>HAB+!^prRPkq<98u<^R-lCT5B?e6-z|v_*4s8RJjZ8n%q6r-H<Yrh9MC&
zQzLd|7xDTKQm~?>1hDxtc)jMkQv6XMEqB{7Ud3BvI_xLSpJem*#WTFbsb6Vg5i!En
zb4LM8olXt5ArCc<8%Fbqf$KyVLRTV=bz8`}VwlfV%d$Xdl3p-2YZAOyqXpatb9>$i
zcH8!a$;df_+z8bAdJ24?(>q?wVnkPPes|Z<@r+%~j}nuu&F@ZYyV76q@d>YuP$i=T
zq1#8cb9o`O1l5wl_n6y84g5VU={mkc6nG-_+%H?YJDmmoTQy3`>*U&5)M8(lCSxay
zn#v3Cnyk(VoK{lpR>I-2CyNKecgI8ih@n&%n`yH(05Ru*D7mG7V2xV{QZ1v(k793l
znNtyU)DG#?*vmEhZdG;i4fkCaS(~a$Yf>?3ma6S9xRdB~N8%2U%5BbmjTm&)AHy$+
z^lEWnQn=JB(g(GyAhN$MDZOC^mRh0aA@|k94zre?N6GnL&5@)M)l=^<e}A}=t{~r5
z&;9ikFpPeIh-URtn7e%OarDb~G-D{f={;@`7@2WjVl!<$$75_(I!)TTswcwfcuN2F
za-^!jMka1_ylSeO=XD__1`xp5lcavJ=~ky80ezY9X9hV3g?Pd}>!Sd=q=6ec%{&zJ
z`>P10qkd2b|K-iuZ~K%brKMq{X#N+OXGmL5OoP*li-UlVYZzpKH`}G?N|bxD<Lbpt
zY7$MpBeEo5DXYMXP0)EAO6u$Cwk`s^Y&2ib?UYYw#!(<J<m60h@U+s`mSbpc$Q{G6
zCp@Q_SXaF#tLvT5Mj}W~>jq!rHgKB8#J|7Z6&}fk5?A073H=CW*$BNMbQ2@8)dVbQ
z4!~v)I0|df^GB`12^}ba-}$J(GE8j$OG%i5sA5T!+FQd51qe6o!A@T;^2Po=`|xkZ
zJ>WB3XvYo**n=8z1v8<7epA5z$yOf(TC6xgN+&OAfrANMux?-Op`?z$RUf)djKm>Z
zXQr+Gxd$hf!vcW%t2CwX&2T5)lyI!6%ke9=_g?V&&Gp<rCf6N%@m@AM*hsX;RNk;I
zttYy{C_H4mH>jr5(Bjy3#`9x;e{Blr8$fmGtES=ji@#~lV_*ocSoMMG&!^JU5lE|O
z!){x@$sO1eg}ky;U65h#vr7SveLTf<ixFC=)?Z8QAJi0ySIYL<Giq0F9F>ope~9>*
z!Gn@*m@U&!?ZKOpxoG0h4ycAq(KtJY^N6&q0oiLmhQ#eLJi$moJ``qLOZM=I#M~9y
z#W6h=b$A`Dt#4KDGEqX61f!!jb-13w)!yH7^+c#7h`c@H$MFND_}C;;0H(^ls3lOe
zdF&(p2pyKjxP5^~Sz;8(p_IUKtY}?Uc9toI@FkMYyEyo5Si>Hcq;UR6K!OLMiXU6l
z;oG$N)nR^dN4jH!<?b~`X|NI3XJ0YP)rIIfXc&QrV*Fw57nyhy;y+g$!)1qcTkF>k
z;|N?X-s#50+I>KD{73?mCqAmfx1Gf<q*ACgG!kZb&hp^sPg5c*J5GYAp1h%1tf{+8
z)djX+T^C{*_-tv9P1u6sJZUuI@}s9`#Cw4}u-`q5q)wPrJ5{ABrQ&8o(7XKBA^bnr
z9cVxADl&W~gP)(uY;LMrq|xXKzG>{bg@O7D3k;SJJ1VP4DD_>AinlN?w($a)LW>!2
z)%V^Sy5|xfttHPzX~*0$UOEk%ofZXd5Fp}){^>0k?k?jc(b_~LkrnABa$r{!Y!{Q;
zF*^(L5}7ftzeCREUHwV$tJ3=D!~YrJK@GhGlRec6%9WQ%Ml2F+ekq@eGVE=h(RyA0
z#0;Sfq!(m~ySn2nZaK;P@M*q;r-()=ZDXyX?F0$_DUvC0-r{RhRz-m(sYP@(s=io`
z!a;d!Oy5H)J#}8O<s{&N>VTTj1U`Je8qV5o!IGtub2xv$7)$ypcumjaNu{BH=336q
zr8s3^<52$79h;psi4n2&RJE(%dX={mq-)Al>keyYt-7yIgI5S6sX~HYurGgdCTkw%
znk0`VbsJY%mRqO_H2?wO-NSr!Vf>8c_hG9O!$Fy(6eSR*#g5GzOvWtR)9ZyI;BQmN
z>5eyvMk~F4a<=KbaDN10aP6OB|A)`7ZFO2Fr?7&>%MMivhy&D}5Iyq#d!&=m#}^G=
z_PH(5uT_1J+9RC8+8P92-x{Yq6b<KhRPlrMZHouju6uL-qRA?R>x;#Ou5XlV*728L
zz-*-jgw~f<gK8@Wuyy$6EtN}HTuhMRp1%uOue;vTsQd-EtB%$tMC0R=|8i2`*ceIV
zN65*NuyO~u6|mmh8n3Rofp}DzxFNUR*I|a;bYj{Hg*4(V@qWILOv*`T*Zd3{&>At-
zj&j<Ij9_|(pX#_bIhC-DBYAr}<S`ptW9+Irb0jILQR<~1A*Sv4o3J#Aa`!FYNluPP
z+sVK^;#!0>Y0B#jF-S$#?qTXP#D@7~uGbiuw?d4!=$tu2vkHXeAwz%wp9xm(yS{p2
zJnR@^GD^OY>=5CAaQ%A|$;5%OxQ`s%x3kQD0-QExPBlFV?$8cLq++1dRF(4{2qW^+
zMqK&X1_gM$=4qe=%rEAu>B9+`e5BJpU!ri?pMLMOU92HeFIN+hl}mmw&)Quixba54
zA>s>7+ifa=<f9kSz{N(8PBd!@o*&4QSk=UGADm|VPF;u_GO%0O1j=q)qW9cspSxCq
zpnqPx11jE%iKH;wAp-c``PRKZ4SDZY*GleY&q{G1JucW2G<`g|lqjO0Ff|V8b9Q!$
zQyd3DBS;m<WJb06)~BGHkk(?~C1^tf@QG|KyDUqsTa-ksPvsz$1J?IL>lxI)K>-AZ
zFFJ-OKrvRYs#X9aU~E?0e2pWg{7;<dDA1L`i>%Lfd@<nwuJ}1j9NYzicCP`7oGMKi
zrd;cj7W#m!UsU1aBUao6U4o{kyo4-1LDYlHcA}Dkh$U4moMs9v?v`*XnpJEa(^=vr
zUhJ@PJ{T}3_3=`b+zV=`S^M4Bi#Ve^BB^00MjA?NxLqpEkdvRE33z*5+%_d|2Ove(
z8K5!|uv@8GgI0htIDFLqjJ)2=SCcsyKgOTpN~mbyuaY)nMU^i8KtrRo#Oh(V{dZE%
zR~cQrnYXHt@)J4Ic4DB>>FEfA?pLZrrQ$e_+~5LfE_2=5@vzv-O{+g4uBAjTn;e7|
z-go$Egq|rj+O~K@gp9ZV3^DAz^6os}HVmVUj?0JkXD`5@+>Z+#^r8-*Rb%t~cietF
z2-(>4ChU>Qa0X5cQeQP15w!mw3-Ct$!)t29@c9|$w@7Fs)7kFtf||m}<RiL4;a^B0
z(KC8&wHphjO)`}KlGe=)$`lUzC;@#tFqeH!T_3th6*_=db_|>Rf5B)6_wgt@asr(h
z`c6*YHF(tA*B}#6Domg4Sp2aVMQL(GeW;~Cw_+Oj>8;ua-Ud+HX>dFQ$iaqnC*e63
z;7nmr7y6qHrv7k&WPKHQWlp_P!<tcjB?atFp?4}}m$!7EVTZ?geX@(FxRjyC42^;a
zZ$Crz?x{s})zraixIl0Hh#}E;Ck;nC$(kT%7S9pP!fdO-HF}r1r5KwgMOFReo4%Ls
zP;sXhf<PET4|-ZX3Cbm?-m3~xq;>lv;lDwkIjkuRGhN^6ZSp^S=&p2L6o?~=hNBSC
zn#Z>^NcmAe6u=D>&h+G?_7U07tIq1rY|#OOfiJisw{J`0a)+%|IwHm|sN6~H)w`7-
z9H;x-`SUgy3f8WBI5R3yU?O;PgzVng9aUA?>D(V{B;^N{9No{GeXxU$`Vr6H5ccTx
znS5`EBk_f$!0)`;ojZwM*xoonE>mNks2SwG*g4j$L&@MC-%m0p#EhB+y)oVYR^{VA
z9tLgw{K|FCWbInrCc>){hp@z<zfiGywm8U%x_;K3zS9vQC(8zdW&Dsvk>&GVkvHcD
zfxMH*bqBZMrakymt+1g2BVV#XTya<aJdQT5izEbx(E84;hJ`9A1Yb&;<jo$p`7{o!
ztBTi4zV(81!44ZCBwrAx5M4|Ce8R!7TZ=GNxSlX`{~TZ6{>r^K$(gtw@@lk6C6cb@
zOSSB0{E{!XYqLE3yl1j&3~ecV>})t53CWiFpY1!-zLFi+uAJX>`@jDQg%uu@%d2)~
zTU`_zAclboz0-D7N2w5OG&;T9{QiS=2fS)gF9uka@;Shuu(dd7L;uOy7X8C_AiRvd
zuwYrA#pLQELJ`|IyZ+A>#_Jvy;M=UVp*UqYb`{BL4arTB`HkxQ6j}Mt5HCA)tENKo
zyYD^>4U2gGm{sB1kUf3&o47-Ew+dIgtGb)L76{nExhvgOhxlL?h=%p&31)`Vt=%%)
zM0xwgrvkerS)UyDSBA9*alVy6vgT+{3!3n&EZK*CNkkr3ydt>@heo`oFI6b(j{oz|
zvB!5N)+D#_AFSVgjgAgS0mP5;no78?f8DlNk^mZ`oX^V0MJ}XC>|&qS{YbuQHzqW7
zdcu)<jYTGB4_O-q<?hDDv887iN}sSbm)-z2JQ>(Y={g_;Dbz54T}hKEw^UO09SlS`
zk1Jk6rNMePlIY=GiV9g8ERqi~2&UhC-=CHPZM5#;E`9dRt>4xx<q;rM{Mea`30eHx
zeB9Vzt{Si>ruc?dhq8_OB`K9UrY?I>+)%7#;mTY%=q>aH&WCV3?st^0nE8iMBYa?A
zvwfeh#=!qR0v`JGI-V>4Zijbc*e?(V`9tvx--lIDh&ha>$VgHhBQG*#Ta{apXWz+7
z-r61O8GqxmJlk~kv@>|c&g8w_@O+MR#t&`p=}U`BaRaI(D(|6kn<1KxKoXZl7PUSu
z;5f7Qy~?H}T0t4t{4V0oGw<5yt>qg!XL?PBeeaB*|E4mhsxTYnUnc&}6hb@{Y2Hey
z=HfD(AfT%4I}m%W*_tm|>rRvV%~KDOP?*}@^Jv!L<pG$?SljLw#B=%Or{R7HWqe8H
zxpzgr5DHK^aqvjD(0o%fLf~a1VL@f6cf(!YG5>uCV<-7PYbCe}nLu;QB(bMU#~AaV
z#1TG7=r}gh5?oMV*~<&D&aX-rY>Hi@SLm6AB$tZnHz!p~pmn~a8GpYIoNs8do5RfB
z4Jm2+s?*fr6dr9AdQ{B8v;<n-sLU-8xS}S6^eRG}zzOm&{vO&rUpDD?+i~_8U_(x>
zCA?LHHwoJIc<yZhNE{3&F01dOVSJ0ejkw(#;~#5Q=zBT$`Ah*$ocZs<+@v&Kxn6!O
zBzitVUgdjDYN>3x0Q#c&d{-fh^=0nc!|tAxS+6fk6wR)TMzm3OLx+|Pqx+5gt;QhY
z{oC8=Rv4P2)hsTaPL7~A|7W^2D+Gctg3^JiuYEDBn=$areM9mrRsi@G;uJ>DX7b6n
z`KtR(Iobsg?c1F=`HU0@ey+EHyHwCh_;7{w_G?vkA52tc6d1Jg3$8=Te^bQ2layFf
zOJp_}t#wXi#sUsr=<r2d=Tg}(Lr5|*OPRE|X^WarxaBij?rzEBLo&L@AoMT{FIKJd
zeuRbVX4wDvQO^4Cr3Hb>fU1gWFFcz!lgH;l`|$^V2GC`^7t2#D+YUN)wKOg}-ayV`
zKhZ+FB`mP+(saMANJ+ped!NZB?^9pkdK_HCUILCKe}K<eV@SN3NmJs&Us}pBw>2VK
zTF*wS_tNAO;&muGFM>Y~1FQeryIxeyQ(G0TV8^xR?0D{yOh-*0_i%)~s9;}#*R=k^
z%Moe#8cnA2bWM3OW%A(@#Mt$GW_j!VwH|d`MVTt+dzIJHlbJnj=TtQ+ai{n~-d(uW
z?=U>66K3phvW|VA=t1N<=)DIidg(<GWhOj!Vz7&t{X=#g=DVDunt^%UT0=wU9SLhn
z@I>$ZSP}lx_XEZDMeakQN44mZKj^b%U%+jN-=pRumU6gCSMJ?zh)k$5J$v*yG+4U+
zbtTzcw<F4KXsVxv;mpJd8j*&?6v77}LrM^JkwTH<Tpel4*$A~6nln~%%F;dWU0n<d
zMn#@vwW+;lUKw6P=*JxXfNQnxwJkIZ(r;3u4wRy&J0^SQ`pHvk9Pxa*gyXUba(#Z`
zcs_OVF?3R*5%cOs0g68Pm23JImn_9)REY{zK6@xPgla@UNAlH^iB}%K&NIPvS&Wjd
zin<m;(aQse>paBV3cqx^+UW}uJ(77_r?pbWM&f0~N7#sP*$-~6J6;&`e$0<$_zX;y
z5bWLz-2L8D0=Nn|pV;F8y!>)mOcnW!ldF%32{hwOF%Np-!b<hlR@Gj&4V7gxB;M;g
zV&d3v%|!OMx)pwTUzdZ=Ymt%Tz$pYiG$f}I@YBTtf9^+PMIspY5KWsBttT{vQ{#QI
zd$b}XNGY}LbS60+i_TxTWO1C!EHx24rN3*+-#7Y*-1*k0`SN6e;LD2(WR*z{TlEvk
z2d^sp+16J&S${0GVSDOyijbW@ny|D=S4JKmfT6TGx+@CXo^!Lj;r0}^NRr}F?rd)&
zOTNU>u&or7CNaJ8tDKlf&?8o48Y-);KQ3pu=b!yHA;$BblWMi6C@J7|-G%G*_`T)a
z4~3~QFAa$Dj#1T~u`~1|GI=5n`7J3(npB~bitY}3=6Ss=g6~7ZzVCGRr41oKn%Tf7
zkK~qrd-Ef6aA!H!?!VT8Td^N~S!pyxu(h#AYvBz5W%@dDY(#3rS!u)dnB)d%znTQs
zYF5d~x)Pp)5iW-{^=gEY(z8!P??03O{C*LAdj}Rtys*lYIA#flQscGeX6Q5(Om3Yq
zdhGXt4FJ`Y$|W|`hl&&5i-i`6C}S_BL&Z)DCuil4MGhYQ`>QrkhqVPO#zyZN!WW1h
zI$8aTq!mPE`WsUY%D)!$tp9d^y(NgipZs~uk>PR(;`}rC!O92X#mAyI*smFi!RI3x
zH3_({i#QC$OcnK#?(37wagE?VWt35;P)omr4{j?wE;fuewSh(nAHHcefDJa3{$E1{
zd}`!vPGARM*izJ^b->w?@hNYLn4!xLd%>>1RjjR9xz+Vpk)-O0fZJe4WfdrFC`efY
zdxlq!D1aV5K%^3z#y#kDfgY6U3gLjoc71_wy?4^0ptLTeC&{WK$)Xd<XT^um>gcUT
zcFg+SF{egrTDh6MMmMamB8afnY~T7@X3$Y6v@=MIk^MU0g+gPcAKI`GTw|#P!R@>W
zX?G$GFj_CwtiR6D;IRB5vuSTg(H@M>t2W}2m3s?M!<{SHbg_x}`A|}5nIItmT{gQp
zJzVR<-@Dvr%51dTygkWYd@ML80HfKk$;Ln;{`GUyU6`AIhy3?67Xk4mSi<z>ca=7x
z4CLeHpT+xtqgY(7u!lO+J@qy6C-Xaaj5eE=SnmqHh@XY=L9YeJ^KrZaYcrFrE78^Z
zCk90XT=6V&d}+CfnC-WRiW)o!IAuMHs;Cp|{u`@%;)TKC?|LVOaBUHQ0q-_Bqnj``
z(T3(3_;Fq{0QVZt^7LYAEgg}+{Ua#)U}8Y}_#V<GKkXL@USt?CqA+|C-|(D4E|Jbn
z9yF62qOIT7q#Y<?k7vm{{>D%oQhsGoYEOVrY`}0HwAB|gL&}71dWgA3BZ-_vW!7VQ
zus1c`?g|&b`J^)%v_RqJUTdyjL{)*XX>br9t;Fs~l^RBUdzKC!6-pQ7)f!R8Oo)qo
zAim|9Df~bl;@PqI^`eIlE;|t{t=HmMJ43YSNhUNW(y~x<Kellb0Q@ZFDTrPYwQx+y
z8LguCykbQ0_2Fotp3Hf!gQk04Pj{zWN40+zH$2}G`cj-Wk~xp~g28J#sEWugRd)Fz
zf@zRh#g4GtV3l-3z0$bF&W$Y1Ef^jT2l9WCo}MvfEe))}=`EOAjX~dfZ`X!9|78TP
z8OxKx<Oh`8{ksNJxt2AtVf)a4xSL*^KGz5MmBzcxHIA(1CMy>{#cr4Q5j+QEVn5tq
zai_kCQsojMATG>5bqiBrt>WRSv7YHr4@Dd9N>Xko8uzk&SzGGla7%fLq%nJ2bWy)=
z1FJ`4y}ybbdY(eOAys(c=bJ7<Yz+o#{4X?OuEfPy^gjU~W-Ih%s;KJeJ=2@`o%maj
zD*gu7gdw>Qh4lGaQtxY=VoCC<E8I7N@<Ns);soYOj3s|@Bw7busb6<oI|vU%_@~(3
zn3inX=<0Yla0$Hj@x9#zwx~S_!W)(683@%F?=jjkic)_|Z<NX>QN!<&!-kTthu#4V
zZ&EeUAI#I>@i%nC|7+;c2rTR*k(5rQqOBmo*@~lsy;#vh&d+s`&8g)b^YK33$nZHd
zkHBM3Xe=Bvb3N8)7Pi23yJs@8pqA9Ya6wA^W2vE+@PJV6dcGCH5W#(`?b=dHfAq9c
zbx2d?^J1`??H+VmCopQA4MIe6S)UB++8X)bc=MAn7S$C$^$XDwBswNQ2BUU`Dv`y9
z7f~G-N?)0)$CA?LcuK2x54YwV-;rF|{X^uqSLlADNlUYn|89?tziGqgdCS)nRJ{+@
zlSW_)v!uo1z;he%Q`qC>ilt?v>`aHtevz}spu9b|KxnzSLb;oEa2P(~@!N%f&(YUJ
z$!l-A77M?9W@RF~6o8RTG*<&^u6j&TU^bygQ<OAB+(Lz`0{1>#6=0R;W@&hNsPt;^
z&JhrNVQluN@%SBQXS5t2J#bVMB>R-6s?TeJaXUdeGQ0=h+rx<5(ai=53MJ3Aq|rZw
zP|A(>@(Nk`K_sLmgi-^`vAHaFnZcrTM43=#=7WpMzR6HEg8mjvn&|eKq$Dtq_**r0
z$ibwM7>@sU8D9YM3tbSKi#X*RmL5C~^zE=$4XR8U1WcDI(}J{q=uArsx)|E&SdKHP
z+Pt1=t`SnI3@6y)Bc3BIDjn0K5OHM%A6}F!m+jH=?A|1I$(6edF7M$ygkG@=XG_GV
z{;So@>_DzkybftG6^Ju7-ns%<k{g3T5cl7Qy)%FAxqJ5lYqEyQ(Bx{?ZD~~R#b3K6
zzC<mPwNnW<8J~cworFE)LVnf3*`d#Y41x84D-gl^h6N6@z7VZf{B97Hq=Hg3ZOhVI
zP-MC5y{-{W8Ln4<PsmZ><OYObag2?I^jex2^^jcmSfY}dc@LzFlT-HXb!!iHtdgyj
z6hTx)=>OdVx?^fJ!^VlGVYQCJGNTLq9m6Xv_`TrJh;G(9x-vUasaUe}-%fth-p;K9
zhY1n})tVS-(7Y#U2okT%BDGFs7#dg&`X7o<46(=g<Z6=hrPsaa#}z>Cwih4a1_%*t
zWM;XKOJk)QR~bQWEn`U<sO-s5D)UH}m~~yM3mcB0h+bi~6taIw5=j}m#L}UHQ{98}
zz-F9byDaFr-NC|T@aaW$=amE|SG2zzbxm`d(l=ajrwCWI1g%%U%1b3B@b+U;=QbfD
zS=Ls+t{j<b*Tyjo5r|o<X~+IOdv)YMc}Ps=e>d0A@s8Zde*5O3ZOeb>^TvW_x;1q#
z@E%*6D-3zZUZ$t=X&Lc?L+>)Vvz*2$cw`%3S&>8vh1MbhM1c{rEM#gEg`$31^;f|5
zPGVd7fMrU`tF|a-Y=vJq#`e1Towt#V&r=jOFM3}(&yYw#tu>u6Epr%So6b0eBAh3I
zsxxk5w6JAk2>vN{e{~egyd&nf@7>Apy3M84YK;RMO%zO)XCMa_{m_$O!(Z`~V}KP2
z0G^h{uQYNXsYb624y{Zr$W>4>h|SN8s7xxEK`l6;8si!<qO{87APwo;g`0ne>KU~I
zf&DgrhW)zC?+iDh1!HyYB)03rDxaY^9-HFnYV-)k`{q5){n9IW6(DY~g}(qPis?1~
zaO}M6TIK&-&>?QnB~RoI3?Squ-PXmca~+yOyB60Vj^~yiRS|9b$#cd8o;`S~T*k?C
z^j{yy+rYx@n1nySv`cTS)D#qe1^-{K)~xN$nU^L6HqTGAEdXMUc<I>-8%5iGJ}E$=
z4WX>4Ls2x_YL|!X#WbLCP=Ec|p}JF%-<?)i&YdHF^HV42oEvKEPOaXN4MhM1kHY^B
z@35uSn|2s!ma=q&Vflawn_~p3EKnx>VUHwgTn}U6D1iPDji5q&Yo@{ED?SMyLJryf
zv@Gm)o4S<3>5LdmP|Y*cZh^;AbRj9L9^|>Fk;H#PHb&T-JZPFtytV>|Q;sv3<qpqM
z*-TJX9_*z|eq$ttS^x4d0rtC+%xn!2{BNi-i{V$|EA9fec%w=_ebC%0Y;pKJc=0si
z2vFBNv9-j4$l6;_nG7V^Up|@>WgIyP;3kFtGa@U&<gb^?3s2S)f*!qF%cJX4KFDql
z57mCN2&55`c>>>qwMf6l#AE=K4@SYsYhH^;7?iOB75hpJapDTJ8twLrpSyaHOxE~x
zestdBwQ$LXjSXvheqlo<2R8j)%hKk4B1iJ|;BF$VdJXuBD)bNyzIjTc+RJh%|Hf7o
zx$-+zn}UZ*C<!#7Fuq$i4UTWZ(hWQvC6>Wqir}M+HaN|S=d$hMTk-0p3R+M$^fd14
z6$OvHq42UiGa4B95YE#3yujWLNEN{aqAl`y8B3|vs(w&GABkEb)qc|B;Zc8h?)qtI
zY1$tfte~P>H$d3jIk(`k<}c91VT}=6M|4hd0WM!C{+G>ZYEI8=Zt`;_kMThwu9ZSQ
zv2NWSw7c#2#H!bPpYK;!Q~BE9`_GqXPFvC3n5o1We+P!$OHAuHb>DR+Gbj8ErrvB^
z4>0E_jiRV&5lgTnF-le;;wdA4Q5yCjI6&ziJdcupxqUKE;P^$j7$9LiE5e#Cin+SR
zV{W<G-CIbb*81m%v}z@oWKIEQG*ml#vC!eO&#tci1|!yV5n#7*qb=EmMU9hIN8wLT
z&z9Q^U@^qzXSf{^F@_eZ*Oq4@a5SVv>7}7Xxo`LJRccQURMkM|4&`14ZmY=*;<qoZ
z+Ic}EBtrc0#!FM4L0$qck}`wgsv-YEghE6)s7^H(!N8M=yDgLbap4ji4sNZ(iQ6bE
zCiAbV_b36KHHsodkTi{zKyZxehvN<=H7Y~k{<8HX@F(Jh4^sZ26rRpza_@q{JGFM9
z%<r%<U{``|m|Pl2?u_XuugoV!bWDbBKM<I;gCps!b>&BYkSm>8c0$*vsOdW3OgW1}
zGb0j8ITe&x{%`iHs}q~@1$*G=NIWjH8*`uYR^;uB;m@*?fifz887y8DvTHpJr1Zpq
zy+Za5*e2q-ZBWeAUQ^@D%(~Fo>_L5d$KS@6JGIZpwR&1D5pg99+)(Qc)EACDCuZN2
z2~`z`ZjLIWTS>|QI?)hp)xOb%e}92q(s%fG>Y&rIdD(QScO9kPMsi7iqq*p?%6Z5q
zWL|qTD2t0=-YF1Bg%RAUbBz8kI!ScFa(CA)eyE5{o<;Yx<&3^2>+D`sd1l@^Sx6@T
zqn3wt7;7hLqrLwt+QC%NERZfn*lLTIdwi!BfC<;Cyv%N;0`@yb)^-vc=zaq`+c^%I
zL?isQKlM{DrIW31E!a>k({6?r=VK~L$TgLOiyG6kwAplIF^lrhM@LZ^FvgEHElnvK
z8zMc=pt;l1Pb~iwRad4)6+P^yr{_%j7XPHwNNAwgNs}ri1NIW|-<aWZg^_lB@`?T6
zq-(~WsrDp6kHrfE>i4bL%suap&0lf{zP>++wpXUwl~B%AS^CC6{sLBU48CDkqW10_
z%l)s$f0dpVM78%)20oos!G*~ezPw<&f-H24%S(Qq>nFMTdRpAn7)A8d`C-NdHE7VD
z{)awLj?FpKxF{8klQHap?11HtW3+rtiF|kB3mJX(Qq)6p`ylZ(V>34^M-LH+WLV-{
z#|<s`r@_9=0Yo>|prXxZ3tgf>_Dz#7SMP_fFjDJUGy2_iDUO)QO1#V^ly5;|Ocf%u
zc5BSa`@#y0ch3FzTjFYrOV;qQM$8a0Z1=6p5-f4@3FZ~{q0MIdA?pQ|@*^rg9FZ}V
zvUa_pBXm0B8b?~hX>lM0gfU1JKcJ!2;_xD^x`_x$3F(ef@Hu`*s8y`eBCt>FzDuyj
z5i@b_pdT%Zzjxah#|U<sa}XN}A3~ai7FGzSc#4}mncMxdmN%mWJyZyf?zW?tH93qC
z8l!3k?hrnYP}-fcsVcCOVeiidl5jqHz9cOUgqyox8Z(oMGZ`HANDKyuqN4|-_5HeZ
z9^^*j!gDpQMzZ5{hWNj$*PQ;U^{-UG@XPJV7x}`22QJ!=t1%{NY=6vh$<=?pv&2%H
z)ap|)K)Aq+e&XNhtDQ7xg674`Ca|sOS6X@oWVK;c;k`|1{COuA^>J=<nyKW5j&#?@
zwMZKG{Fr}!A7MJ9IhEK+hPE29F!#W4dm0Y}MOu!#61*fzbM3#$#h_t`n6Q!WmlwWN
zys&mKC$Fc(MsPk8wbxcGAlps<RI?B`aNC^lw*o_z<G262Ze~<Jdc%tM#1PNgF3K>4
zLc;<0&5uFj<s}`ICsC+8)I6Z4Z{*-^xT0WExop}JwATvR!NaNObYQOp*{GBB#X%an
z;|P$?^;l|Cc)pp;N}L3zEx@<UQ?6g*Ce{EYCt$W*V!n_!EG{lGFN}fWJ`HNAUd>q2
zgGQq|B8xXLztal}BZ`w%Mj=9j5#gC%wraA_ZIhQ8EaJ6X{~n{o?)-cK`|orrl1aDw
zV1kPPGVp&%=Py|9zkh%oFq{TN6#l=4pU;+VAD?_=c6wm0;d^Y<Y503S@_eI?S*K~S
z0J&Jv<+9Q=aL&{rfjYLdXjXVZg!z)ff|P=OI?9Ra^LT|K5rv{Es=nCdnThV>LN!2i
z?~C8_Kgsjxu;kEH*C?{ul+D<=rJ*b=t4(&iu=}eiwnkptMQ`~HM`>_9jvbMkU)K<E
zSt#sBp*ANUk@2$KlNOC>{%!eCH+WopO|9;TK*v^U&)-mJqB98}Hve^|hrerkF7m$|
z5iubLa#<7z_)391A3!y|G8rcDnK-z$A)#);qXIRcC;Jr$7fzDU4r`%s3w<gRWG}6S
zVL^JcNM1~v87(G-BS;oS!k`mpDi-ZWpG7jyZbz?6jm^l#LEf!s=tv1)xL;?0q8<7V
z5oP(htvJaK&GW`=&{Vr;=rHjQ^L$=Hax&8iDZnT?oFCjIAYt5q&$L*oIb1m!wy`wc
zZ$mEf>|Lf`Gal%51YtYgiR`=To&4|Hjou*s;aGy(saC$e-yOdM2WPdWeN9*Rs+n^z
zBd%`0!4N|(Ei{FyPb@^`CKQ{5Eu61PR24*h^G|r>nO$Z!`cX-uDmCH0Ogd0YvbVN>
z>ESRJ0V#eSbbJ>#^c|Dqd-OtP_Zju-Y_>G33O5;z;2Fl2kr`Iu&#So;8cdwi{_Rk?
zN$LZ`q1vqjUDN%YU{V9rvhwUtEZM3FVkM=i>k}Ax_WGn{$R=%zMnZK}s_(ytP3opl
z8j1hc;djF|6NIsiD*Zq{)+e2+koI(0jMJ_EYuwcmnzVwV|K@p`(M?_mfz?yztNMqo
zT8B2PcUT6Os{sf9un2C)yF@!2e>SV$KoFZyjsxqrc5;BDlHtIT2q~E;Mhz!zX#`!z
zr-+qUBN_4TaIgGwX%B}`xNG?}&xodR2z=i_{<uc@E3A|Xa|G9g(IaRUn+m`Uu{JYu
zTAW?zC5w^Rb}6;7*=9Ie+b<gc7jWIJJb*~`V{x@k`%dq_={X|1Kk;$D2FMt#B}ngT
z$(u5B&t$18_vA)g-Fpt8e}mpHr<x~LJNj2Gg5ep~YS1w7a0EYB3sHC=UkDQgKUgyV
z2a4Y(C9w(}y8cc}>>xGw;(~=~jU81j6&b!}`tv3_6He#se!PU*N&+3VFUK^AV_5D;
z_p6^AjL;+l<EZO?`9SGFsX=*MjXa+d;IVTb0WsT~v0^M~hS`NbZbZ>%+7aA@U_yMJ
zn$wHLjt%$4*E?!4{~yF59Wf{e{93(qfnA)_7&KvxO|vzw#7Hgz-LL8{@!DO4<o~w6
zTM*Nl#8mItD&A1D;^(4et4uNW3sq1m?<tsgiC9TByZzNYcS4{s4!=968tstI3L*nm
z-lUYYNtDf#*yzrxaVELx1B21`dmB80OaN)$HAHXb6-^Vl1Ik~&`R?f<*GLVjwlEgx
zefa_)k`xhCwg+odtrZJQRO7OkX|^M#Pbu*`(=K?(_P9b9NVs1f|1157W;%rL23Q(`
zC-;3jIJAN=F-T1?pygk3<zE_q9i;&s$34O^eY%1pqXop&O?gy!<)t;KpmroAjpCsr
zZ8RhGBdPJLzsuw(p3Ude@5MeK;568Qq8tTU$&~TkZ-s2sPx7l32518U(F%7It3zq%
z4{#CBfxdpO^=0}MVCW+>U_dFG8tsof_52WNLsMDM0D%yZHBDKn?P)4vHaq6eu^4=&
zu+k1n!_hJSM{;zmBnfUMn;|iv53cnY5XVyUe;@w02h;HjX6Y-;HYB3ZXnU!_8oQ#I
zniMFHoORuf!|KyJNTxGix^6?QvR~7&FpO+oPt=T~SA*8VQeIxKJg%pB^zPp+uYIYJ
z0`J+-t~{$YW9&Hg4}?cax}PQzfo^nuRr0@%<;`yBDpXFs`(Hs34rVkUvJnCoDKQRv
z26ENE_CQ*4MnhqG{xp29^kBR>uN%4K&U$@+6dQHWCXD+6<vad6|KJ`Aj1BxXq=yey
zp7>ApOb>zkuLkfP<NBi~g?Cu@uEs*eFnwvJ^Pg`J&y~U6+|+J^4dOyPw);$8rL_XT
z24|3%W{9K~>rQyC9->uN?t;cDJQVkLc(GxRo|inKLBHZK?8dxAWzllkK(h+yuUgX8
zH62JDf5b2iiD`sX0*D9Yt%=I8rkX-tb78|@J46i_@1Nn+@^LJv{`Ev;_dmyPP6Xsf
z=Xy_V%+y-WxL>3MmU;>Kej?x-+2H(tCL&Qs2aZqNz;_zk-CuG99)rq6TY}yf)CCp(
zkAn^6A48UD08dn#_t$xoM+!8N@7P+78zwXvg5)=)fx1b0W3;+{WaWM-Y6UA)kiY$Q
zp@mTw@4pDkJJFPe7SjP^V(U+k!gyFu^^;q-7|9uBKG3;&g*SSF8Bgw&aDDn?U4e-`
zlh`|KCwiy9X^*Y3WAFYR_OSQ=H_weyY{V3Vk*_NCf!^eaiKpOiBI}xPuHUWFj?@4&
z5P}VOq!k3nxr^j7f)CVtg0-Nc{u9!d$F5=}S=F|?L1_71id-sC8eA+xsm&-!Mgu|{
z8Ijiu>EP1#Hwr;Jp@W;9-R+3Rdz8$ld{-vBKd<ut%zo=nab*=pa|vk4qWdrL`0qP5
zYUtp&VR@apV165;I*DqLuP#dEmY_Anl#~}aEl}eD+*1z40R!|}At&Rc?DCqOrV4bA
zCIf^m@}Zbra!3147ksK|EZA)eRG0e2xqrjAg8ilgmX`lXEn%0diaF1IJ2fHfY2J%A
z=>G~5q=2+M0<qo$s7ZXF_#-hK{ucy-Ru;eg_|?$xxI}Wh)5A>kBM6hWvF?X<7xMqB
z>8it`+P?M~W*Cqdx{(+ZL_{g+lvD}nhM_wIM35RnngJA~yA+US2#Eo%q?B}rNOyw>
z-*A85J%66_Jo~J(;yv%)YwxuTyV<C;mGn{L_xtSpH#rS5{Xd}_+$*-m7YojwjeWL-
zmwr--sMo9dZl)LE=Q-?OVzBb&<yr2CaA|6I>!|<N2MTS)tg#4ellLh<=Fma$ka)#`
z{CB<1O8TKSB!j7oCMvCEo2OyL*8{Xi-iTj_8En6Btf#v;Mg_sL8PC5IXAZJ>ZI(XB
zjaYd2NyQ%|C}1%oi1bQk)qZ7yS{NVbyg1UonLVm$y`*PY9k}0WO{LFzy!akRIe1ot
z7d!DkoP5qGsp4}M3CVz(X+|wyuwRYB;d<|JLSYU(Cnuvj4IH(2+k+M0vm(z}&V;cs
z<U!EEy@O8qTpoNTF%2x<&C<Hx84{W7#EON@7IUJtu?ca*Ixek~5@}COjx(W+WjvY6
z3)tjj^ull`FB4xrIcK?aGqyMgrOwyfjea3J9|zD5A<gm=Q-zsgC;>M^h11<|g9^#X
z3S*@Qm&>aiuX7n8--)4A!PermA3C%qG$h@K=FbnbR5Inp%0AKxTIss6DQ2fQ77sr$
z_AGE+{;De~w=#oyduMs()_AUtM~;0#u-l1UQTSCd#-UCn&R&mv8#X)@=!s^_xEqMm
zZOuF7)}6p=0%cfjjVM3=%CK&<AZO#we|>39ur-05ikgdwR=yXv=MHbP<Wmxi>iONR
zqq96VzBo4eq2x<JDH__77@VY3{lJ#t{y~JM(_To-`1GK6=39Q%A0k-2ScsV$(s+;+
z6>qu(E(JR&aT*-I0LL0i#!`#6J%NP9gjrTCdWWwCdgUxXoHx!9BmT59lz2XVZ_N3A
zi)3*xsmANE+!uN>=}~1qI+5KC81R3gk=Az#Uk3``H*cQOJIV^mzwRSr2yc4OWFPGG
zT?;}6?nFZL9o03L)_qu-UX1GAo7Biy!fCS(ZpVMRL~9&eGT<=5_9UuXA2a)3%)f0v
zy|)f1lx(^n&nh#{Q)-U2w;#PMHn`LJc8a(bn|RfK{lRN3&XsU%g)z|s+ggE3k%sa-
z0Dh>CIQf4Z?h>=|q3{5Hk56;St@Js9zmxA#>ZbFodlYXz?M%LD)2<TzS0^?;SUR};
z=Q@Daq6|*~hUP`>pnk&-?G~9{r04$6bg_=*L66NRTx}xQTwzbT3e^{7$V8a+1pafk
z+R1q#JTF5zRgJM5&@^<0wo6c@<TgE3wD?c@EqHkHpx=@hhX7s0o8?H(V5ew{Gz(O?
z5g?8tto&CBJWHRmA^8XUap8=Nk)lZ}yGX=DSXN~BH)NNq`M@pwh&>RC40{{{Qq>w`
z2KkU+#gzGFMRoz-S4)q>Zk>g7!3I#vIWH{&|HxSu7Ml>(kf3EypaN~}a`Je#z_9;D
z0g2^eM*;9~CW<$Q>t9UXrF;3R?b2#7u}!e^_BE(LOU2`%0`*0a#W}beQw1V!=k+ld
zepndpt0$-09wBn;%iAH_>r9Orp@t4p9B@-$6WWJQd=}xl+s<96t{p%wm~0~S|El5D
z+jb0sAx}?FjeC20TR7osKto`vE@=4GK;rhu=H#Pk`4#p>d&Ja#MnhERg&BP-XlnA!
z;a$~)^BHTqm;r<w2D*=I#w=pZb_O`|>u|op-9CmZgC(Q>2aK+O*q7{=yj`?M!_RVf
z_$fXAHqC@o9cSWs{Y`Zk7S#3qM(3zCwtzjoCyP|Wae<*_{%6*g+X<`_Or!`GhaDRE
z6~}D1Ewvm<gPCrjx>u6#5(w*^Q>nF^MY0BO9wqhlchs_A_pigk@(i)(STZXyMcOA<
z`zJcXCiFKO1n_q<f%oBu4vPwi-z+o#(ts{O>}pQ4CUJUBF%5tRo{~Y^C>CnUT@zoC
zC*wLYTKe1q)zYN1YfzbsB@&E`E!o5ij1P7kM9Bpg&)Y{g(0S4A<IS+#>RT8sl%74~
zYmfYhbk{n8Ai7y*swJw_<(L3x&@P@JQI*91K4?N%rp_w*(C<de;ztT52Vy=rETIBF
zytf~K$&mi_pJ1?H3J=5B*O-IicLc8Yc<keLrl|GGietZQW@N5O;0qA_zkasEg=HqB
zMgcqn2Q`6C#`0o3@|C5X^%eOQ{Y9BY&?2sp@YugD<KK0sF)#-v%qaX!n}jCbPpcNT
z51lIGZo^AO>S~=l*Y=w|YU02as7L?19}SfO<5-7=(IZF}&Pevf_oNZU#j5Ps6PHn$
zP2w_A^|-~CRmGBf-t$oy@pL@rAv&KQ9Fn2KtUyc8e5Ko|i|6LNABC#0ORCH7SHSm{
z%H^CiW;(~TTidiwQGf3kkmlbnUGp9c-IJFqMlN9}9d?*7lw>K7;1pVK9`OWSpR7|M
z$1rYW`i@_X5z!YnI~m>{j3@D3ttVT<8-%~Wb_!Lmf85=rN&<gU_yD?tPx8$A&fuxi
z>ONR-8|Mz_(<9z{mii`weU+eA+&(^#C-Ks>>jixZ*S+KlMnZke`A;QOB?b>&wqd%%
zJU24;!KU&43_aa9$-NB;?Xx{y<2imc<Kj#0zmmbfYy6<uV4XL`S9ld9>eLzf3X}#A
zC<fbF=Q3gmZC5{>>df9r0!UXjLXpBAK`h&6;j(ou7yvyZ<&dE9Qeusgj8iXbj8<>B
zSsmzX+YA#@b&$~Ed`ckAi+uL=4O8hHx4<LCuY?#?U?xGsG^KFLjJA_qPj%W;54I}k
zQwq&a6dkI&nDdb<W_ZHZpXig@p>=tn9M7(j^RDYVd+F8ki6VNh=S|DTt4G!$4d>f6
zj^xb(>8JQXEGq=B*H8TfE=8@s<&Rv7naT_#oB4iB+VumQUGq|?&Qefe^W+~lgn7rl
zrS%R+`MAaW4Rr159swi<sBh{dhHE^9cgP?z-BjeEWFf4U<~5ZN%pIT4))st~i+`{2
zVq?A^g345}y)y8cWV;L|L~ss3H{uW(Vb?hP3>G46+3;kIFRlF^s0bTSJ%)M#UKr3X
zplUdiw^QZ$x^H#48d0zSTC5(o>d)FHPHXLntBoSW>XAH5l{5qleZ3Ab!>{C0&HCZ-
znB1v}H{&txQpC8NX6Ngvq14R`da6#)UQ<FkzfW%4<~>rN6D`{^5MWX7M1Cvq^3a`j
zo>!uN1CWAq%DDSnf_OV|WR(;#=3_#0qHj89dBxs|k_aC5$f+-_t{fe`-Hm7t8#VF9
zk7qKol??B`YNWX^<X~(%DIuYGvO7i5YzHfEEz2(P+49OCZA&lucBEWur;=;ehwun_
z_Zih=pwJ$2zeJ~hZ;k=ILxQz~@|D{UFI@L@K7H_9+Q*14t3MJlld*A<9zi@rA$bQW
zkawfhQY!4!OuWe#1;d@PR9-RMyWMy(dv`V>cs6Vh=_DonF?c45a<OOo1K$MiQ{$b&
z?!gIkpBi<SmfsuNU|~!*)1yI30bZqF3%q3tYh0PrN^Eo=>eT744srB+Ds*6b>Ru?b
zm4)*e=hdW3uRTuc3pCADpLBY)l6FTxAZhing=&?RnWlnR;b2PkvrChTqv%Gxt+Z*K
z$CtwE^giH71z9s?&tjLTD21$>ji|FXc>P?l1ER_uUZg;B1IM6pztP#Wh3Q3%P)~g*
z?Hd3E`FSHLoD9t4E)D;5!X<%9{HjV787n>E9#`4#>Y?hNz@1|J5-A1EjQ}tPrb#6>
zkMQx}<T(@}X2|f(ImfNV5>4V&QBTuBu12b?7o6{?F_VvU(|<%89yY8i0Sia9-1Nlq
zxz#j0M~`x^C8e1OKyn~CwnSnx;O!-`&UAQS{di-ziMGy|>r?zQ);|k5Ux==Je|2Y?
z12?pm?#b+f74p0NKg;qTO?&0?9ake#9b^7^r&BMSgo1LAe^IJg@&MRC*=JkjFz67I
zfLg~nG135F8M-@_E&|V0vg&YB?o^A2U(5RmyaFl6|M|Qtn(MvByHHj=-M?A@=w~sP
zybaps{*o^7{naL);jI5Yt8q#2Dt=JKAEm?+mKEB_??gQ5CU&VmWX@h#JZsJ)ITlIr
zf@1V|S|Go)p_%a2qBgJAJ@H1{7c~bgQgddK$M39De~Z-MMz7HcFlT?0@s$ES30%rI
zOgZ0_&XeFYhnus{lJup_5UmQec`-%bN7S6urr4b5)<*Rwdx#p@oMxy*6vq7RO3)Vi
zC;5UH<F-m`J8Ym`^Hq;_J)H=o+E^`SX&LE%%E}ls2#NF5lY0D^$xnB@Qv1Ls`5cxV
z%U?&0sU{MKJflEbRlq&$24dtUj6SGvkdY;eb6zhJn0L5#f%UM9Ek{8tmI`o>pg0S7
zTBNGgvqs%%6pkRY_G4Rut%WEb6hQ3Pveb|ZXXc+gafiomx;NWi?E#Go)SLy!_<Dc#
zciThwpD-t8bxFN}kuHpMZn^ya)FKMXE^S;Ax4xREu9kyxw<lJ|vKJHz3-UX&F3RfE
zS31oO{BH89{AHxKF@KH&5TE9lOU->sDMnROhLiSd#`o5~<y)eTOZ2?AxG-I@+jX^B
zc9zH{3NR7?`8^XkTqQBFY?D+&M8wM_RRy+e+A%ap*Sxi%F2HnbylOq>n)BfKQ({dO
zit->*C}2AZ=%o*brcQO)QCOkt3yjNs`1fJWFqW6WHZrvK_nvX!izrZx`TX-ws+{Ho
zV)g7nCJYg>HdJ(x{n3D;zjbf#2MDR*)yehtO&Hp0#5gXgk=B_jq3#1$s!MM^1t3bO
zS?)b{IaPY8-~e=f$ju$5>iJEW<dq=;AgsrKm~_Ei&j{H)Grt^ChDVoViHI82u8Od?
zP>s`9sY0#M?9?V%J1uTt5i(JY?Jc6}LZ3JjiZ47;DYpI+c?zjT(pN7BbF{5nB=6te
z8M*8{>7yxwHY^YqN-lhnwAE}_&~N{lZS&Ua?;Bj#n+=3~^DTaovXP0(bo0)e*wrxX
zC@AF(c#HABOFlUpFEvVNhL6TTaWK%<oi^bRNZMxrd6h8r$%zr5JfjVJM{U%+CG@Ew
zmLg5-eGc<E70xTuG^yheiH+^P4%alHfZ^2SOF<xEtTXn3fb1ZS@U&zdb?Osm!>Mft
z2WP}Y&aZ8<#;Rg3N;zV5@*@EP=N>^pa%1;9{4GvVGb&##7)<qRIwpwP+3U+p#&HOF
zCa`U;!-FS$cV`>?M1R8Izk<g<Euf+OWDv7KQtZ#ft@|Hai|0(PZU*xi%?c^%drd}x
z?|YEAe+vsoGJZlknXfw8lLX80D*UBAlkHPS6k`uOu@(RGD&xJ4d_QSgJolXlO=7<O
zsU}n@3ic#OjPP?3)VwsBq)L~&;yc<AHIn$lXnq2j)!OIdJO4Q@=iUW4J9V=TS$k$0
zuj0>PgMc-d=~<&_;`#L0%Qct&!jE(l-*A{_>)nP^9WCZWDA~L6@}dW=Zv&sSu%PG|
zc!Ia7ZM3$_`)ET)rW+S)jAyVOyn)}xgsBJWK7EU)^ch+sJX9ebQ55Wadv{W0mB&!O
z3V5=uzfnff-A{{()ySp@=S|J&;<3FKC{DO@(%{-5__BJVH!qhSxIj;;8wsz*BfuUp
zC24$xv>7=+PV5!}^_yuU1h2xgujzECIKb!HGa(s5XKFhzu4vTp&c+XJGi+5t9V7|>
zaQr74b1xbkH{txt9|6(+@wt(g$6JBPImo<p;O<6re4j}tQQT44xqD&I$Lq<j8jswe
zucTS=+cVi<87wQL>}7Aka9WR*_EDp?^Wf+@p&Iealm)MEYbl!QFC+c(xZ-Vpsk8XW
zCPzI1^WZ3T#djY(lb9Z9^))sU==_#hR!*F{ITbS?ad*`Ht7Yf!1wLuT>E-eL@t{(&
z#>mJ9Z}DFH-yq<}JL(1p%C~@rJm2tfmj46;)0JjfL9IJVJ=Nb_YMGmL<SHjikneTG
zQ7n$xhID#a$5uHPV*iU>ut#sV-uj|(RyY3j!1T@wRF(X~pc-jNa~P#O{xejztpLQ+
zKn*#%ct(G}-eH*1s!Zvays)RJO+~EibY}0V#@xO!Pb;{PD{z}|g<r3BcJR4{Ti(+}
zL|RP3Mj0>U)5X&!71VAz6rpTzFIQ8%$fxFcUvQa_c)dK?x-O-RT)y&O((Xumz7Z7%
zQ@_zVwYv^=bu7e-J|IA>u`UCe-95hl{7(sC%jEYLC1gttWnY3FblWi+Q<bu$Zoyav
z8prlxY{d@vDY1xV7T9%I&@vC`E&3igIj!E-zIX#mm7Ug%mYy>0EEpqRfd1l)8-ZIA
z2JTuR*m-sNbK3uay-L`sx46q4#bMG7+|Jx{Cc<&l+$LcdA$b0zYleLL5wNf+BB4Wf
zne=U7nr5g>cH+r1U4toCTx2zXl;)bzee64_HgXH;xNxqYokI#PE{kV4KtiWHO1cI<
zlKj*5$`ZJbo{UeQyeN|XUfQztIZ+}Mm(@n8SDVxg;{CG#_(Xypwtv|A`Gp7>mG3Hz
z)60%b>lOo>x2W&~U|XRE(v;CVvF74yHCz~NVV6muTPD)4zieGx`mVO^%`S@edRP#%
zTiF#}c=YWdi#$>NhP1gq@6_Q}nhtG;RUDmy$FwoO$Et5EIwIu*jwe1EJ_v!Kb&o5H
z9++b(?!DM^XT$BFoN|I`4enNqIC%QnP}OVDMT}?dQb6rJ7n!@HY`>Z{eUZS}x4(bS
z6SLoh7mTbazk*Io!q@EhE0E8yz7j0A4q?^xGI+ZW7)Tu}a;j~95O(GMI5+T1&TyOP
z!~<<fuWP~}l{wpOTCS6`XXCyKuxK=w_1Y;A`o<YGAY{>%+s2G)ZD+H@S`F&Y`~x};
zS1r+%R>LV)0IK@CDwB47(1oW;HZ#t^PpJZZ326bBtV=`1hbv_{N0jm_fdRx~@&o_8
zg>~Ca=m?Mi_{W>lWu_Vz7b2O1t`=_QHl5x?JNbrqhZmgRliG<Xuhd=@>tV(2k~~pV
zB|X*c&E@ZtMjci9#OVAH4r$xkYuXV{7O7n7|DKW3^`&0%%6R55Y~mu-Y%{A~<231N
zl+L}2bU))*fGr;<@996=PSV?dq6OBxyULU#r2;sB8urwA9NYLcBY8WIekz%Xf%~{Q
zWX@5fXvW5$#a?%jFLx~V!s@7?Iy_*Yt=c*6;s(m6NY>!<i(D#&^H4{UTL<z=-oTyU
zdW-C<b^8@t7o}1o)>k5Ky&by>%VzxGQpUmDHAbh;{+A+<Y(Kc557a>VSx@uX{2tmz
zoZ~3IvOOH3i|BNa0ndb?N4Cq}f6H3;*-8>$p7}h<*EzG>-WovXUHs%vImtGzGt?r@
zhracJ)LA4(Leu@dv7Gr+FZSpZ+L#zExf|y!!NfLQSW_~RqbOh`@|s?tv9$Q$kL%r?
z1Q`#P!q?W<;gKF@P~>3nLz}lAPR(Q?&+Vr6#5nNV=dsONWV%5s`h94z7@gV;l{%;k
zjgC66Q0x-=h^I<e(W>^~H2`xj8!4&U&3(m`@rG7f%?suf;mFn7Sj1tscZpu~$S7bA
zo5^+g7G+^k=?iSs8X$Uo+I;*qbM4s3cJ~?ST|A5Y$t-12&x?#>#2+`(U?{O97R>NT
z2#<z2-)zvNuC&ijA-74;Up!bsp89z5@;9Nb4Sflw>z$mb&0~$bt#Qe-YyIa{PFDkU
zzB|+qvvN2VwTXDqM@y$KW?~dd{sEWhsEib4;QjjO_{7QSkkaenQ>6p-zg@}^f=rZO
zP^Niir8qqEwQN2w2}<U0y~j;Aq&allZ81BK(H_y_6uX(#%$6@Gk{qAZlVD7Z++SPe
zTMzX7?INaG5%mCu8MkOmCW7zQ5o-+7?3UNJ2uyp~-mu<k#%k<<q2{?uAkTCoCE9M$
zW@^;F4$c6eILjjxFJn`%f5fL)@kvrMp^N?|A;zccR1h=U_=mbOwC0c2WQ#Ilxp&?f
zkAK!-e=(3E(a!Qzf;-LHRVL(T?;O()4*2H><w;Y1d*ojF_onw@$QXbM?2C`IXrsib
zNDM5(uZIAOnok+N*b4`?ZOEUQ)|&YtYC^-eX6JIR@{&tyZT>R0#@YQS);{uKdrBNY
z5~I5<3$5`-hPO~?iKMa#o=IiX>AKkKU!5Qlc=q`~0T#~%^m)rsat!m}AqAwl0z^)+
zLnQ!=yp!wY1OKCXpJ4q^JipRlq{wQxQ!4NcUU27(JXIJ7LfVYuN`Uc3<{RRg3I99K
z<}Iz|Xbde#SIUR-OO48^1Mku22hykO_ks+imng`T+1dTDvd8*e00V$48D<Sr{DBx`
zJE6>vPn)7Mv-Qiv?r@gh_W6*+*;XZuk#rmFt3V~Ui}w3K<_V)Hn-QJM{#XXvyO<^g
zU>C;lBet$ifMlBNApzAaxKnWe6olhSjy$6M3|{g&ezPh_vD9vvyP<!MAeR7ls_qrg
z8uiH01oPbWEY3tG)jY4pgd3cJGI~dTmdTO2MPuWaXOxEuq-T9|<ykTERDCi<&Fg{f
zRsCg3-ju;n4N<o4;hR$(>K}CuY=fWS%ZhE?qs1#vfVm73QERrZel*BC2)g97E9xGK
zl3`STbXSn$YP-53;MI{ivaKs^=0d$-9qX5qF@EXHZDsEVo?UAtS~Q6K9?g;#O3yaN
zPwZ-d^00Nf>eyY^Z}zwz$Cg^9`%zDoE%h!+M%))P<PK>RW%lseihNI(mufOiq2iF^
zDkwb&S~q2#!`-UR_~>XSsTQvfZBMIc$s@d)p2rX35g2NT_a=pynP&SMl!hyy=x1b`
zzqFn@p1V~H%b6jRGkD)B>$;pNZkV{rh1Zqo)jzPEmlhGE?pr`xjh@dx-km$2UV*L7
zoMdjMyRQDuJB-q>@X^dLb$&tgnEMfa`_L_m!V-}?iVV!~eIrBVnp%O0ZMLmTh|GB?
zoKshZG=!17%w9YSqYDCP#faZ?eNUl)n5mIrnJ|P^k<lc;7!=4HH1{wR|0dyUK{N|~
zJK~llW1nrW&w4W8cj=M0+MM-LKx8H@EQxR51$z)NAUTdWLCw*#;})pCd~CYEwuF93
z8uAw$F?1&giw<%}32g!ka}0_8_XpBh@O>z-t2Z*Fp~<elryXPj`l%;UFIF=;?BiJQ
c|J=PH3DqG=n%uYX1+dq1c~!ZxXNaKx1MMff)&Kwi

literal 0
HcmV?d00001

diff --git a/docs/blog/posts/roadmap-2024.md b/docs/blog/posts/roadmap-2024.md
new file mode 100644
index 000000000..2832660fc
--- /dev/null
+++ b/docs/blog/posts/roadmap-2024.md
@@ -0,0 +1,91 @@
+---
+date: 2024-01-10
+categories:
+  - Roadmap
+authors:
+  - remilouf
+---
+
+# Roadmap for 2024
+
+Outlines is not even one year old and it's already gone a long way! As we just reached 4000 stars, and before laying out the roadmap for the following year, we would like to pause and thank all of you for supporting us, using and contributing to the library!
+
+![4000 stars](../assets/4000_stars.png)
+
+## Thoughts
+
+Before delving into [the detailed roadmap](#detailed-roadmap), let me share a few thoughts and explain the general direction of the library. These thoughts are informed with my multiple interactions with users, either on [Twitter](https://twitter.com/remilouf) or in our [Discord server](https://discord.gg/ZxBxyWmW5n).
+
+*Outlines currently differentiates itself* from other libraries with its efficient JSON- and regex- constrained generation. A user-facing interface for grammar-structured generation (it had been hidden in the repository) was also recently added. But there is much more we can do along these lines. In 2024 will we will keep pushing in the direction of more accurate, faster constrained generation.
+
+Outlines also supports many models providers: `transformers`, `mamba`, `llama.cpp` and `exllama2`. Those *integrations represent a lot of maintenance*, and we will need to simplify them. For instance, `transformers` now supports quantized models, and we will soon deprecate the support for `autoawq` and `autogptq`.
+Thanks to a refactor of the library, it is now possible to use our constrained generation method by using logits processor with all other libraries, except `mamba`. We will look for libraries that provide state-space models and allow to pass a logits processor during inference. We will interface with `llama.cpp` and `exllama2` using logits processors.
+
+*We would like expand our work to the whole sampling layer*, and add new sampling methods that should make structured generation more accurate. This means we will keep the `transformers` integration as it is today and will expand our text generation logic around this library.
+
+Making workflows re-usable and easy to share is difficult today. That is why *we are big believers in [outlines functions](https://github.com/dottxt-ai/functions)*. We will keep improving the interface and adding examples.
+
+Finally, *we want to add a CLI tool*, `outlines serve`. This will allows you to either serve an API that does general constrained generation, or to serve Outlines function.
+
+## Detailed roadmap
+
+Here is a more detailed roadmap for the next 12 months. Outlines is a [community](https://discord.gg/ZxBxyWmW5n) effort, and we invite you to pick either topic and [contribute to the library](https://github.com/dottxt-ai/outlines). I will progressively add related [issues](https://github.com/dottxt-ai/outlines/issues) in the repository.
+
+### Many more examples and tutorials
+
+Let's be honest, Outlines is lacking clear and thorough examples. We want to change this!
+
+* How does Outlines work? What can you do with it?
+* What can you do with Outlines that is harder or impossible to do with other libraries?
+* How you can perform standard LLM workflows, for instance Chain of Thoughts, Tree of Thoughts, etc?
+* How does Oultines integrates with the larger ecosystem, for instance other libraries like LangChain and LlamaIndex?
+
+### Simplify the integrations
+
+We want to keep the current integrations but lower the maintenance cost so we can focus on what we bring to the table.
+
+* Deprecate every obsolete integration: `transformers` has recently integrated `autoawq` and `autogptq` for instance. ([PR](https://github.com/dottxt-ai/outlines/pull/527))
+* See if we can integrate to a library that provides state-space models via a logit processing function;
+* Integrate with llama.cpp via a logits processor;
+* Integrate with exllamav2 via a logits processor;
+
+### Push structured generation further
+
+We're just getting started!
+
+* Improve the performance of existing structured generation algorithms;
+* Improve the correctness of structured generation algorithms;
+* Add ready-to-use grammars in the [grammars](https://github.com/dottxt-ai/grammars) repository or in a submodule in Outlines.
+
+### Keep developing Outlines functions
+
+Functions are awesome, use them!
+
+* Implement a CLI `outlines serve` that allows to serve Outlines functions locally;
+* Add more functions to the [functions](https://github.com/dottxt-ai/functions) repository.
+
+### Serve structured generation
+
+We want to make it easier to serve structured generation and outlines functions.
+
+* Implement the outlines serve CLI `outlines serve`
+  - Serve local APIs that perform structured generation;
+  - Serve Outlines functions.
+
+### Improve the generation layer
+
+* Use `transformers`'s private API to prepare inputs for generation inside the `Transformers` class;
+* Support successions of model generation and text infilling for methods like Beam Search and SMC;
+* Differentiate by adding new caching methods: attention sink, trie-based caching, etc;
+* Differentiate by implementing SMC;
+* Implement Beam Search;
+* Add token healing.
+
+### A more seamless integration with OpenAI
+
+* Provide the same user interface for OpenAI and open source models so they are easily interchangeable;
+* Integrate the function calling API.
+
+## Last word
+
+This roadmap was influenced by the expressed interests of the community. If it doesn't reflect your needs please come and [share your experience with us](https://discord.gg/ZxBxyWmW5n).
diff --git a/docs/community/belonging.png b/docs/community/belonging.png
new file mode 100644
index 0000000000000000000000000000000000000000..7346f9aeed55cb0b72aa6274849599fde0c153fb
GIT binary patch
literal 142520
zcmdqJ1yod9_%|%JA|e7(5=uFwfPm6ml^8&9kS--fN{~(glNN@KL1Cmpy1^i%r5RG`
zMsgU2VZJ@ytJi@0{@44xYkhIoa^16-bM}7X_tbu#bDpc*kR>@zdwk!%eI)X7*I@hh
z9R~mQe?58_{AIplngaZf=mAXjuYK9AXJ_{9W7;QwP5PFz{`}C9C|#3E$*&{x)~?<Z
za!1eEpF4H%ps3uDnMSn=$DGQ<UcVJtDV6MA9}Tix{cz6nuR0FRR35tqz3`HJ|D%w<
zTnARXT3N{1^7G5hOwAk}O(pBi5b^OYZEnM>82XFj<x5yo5%Ax8>4*D>_8&ZYib?8^
zi%;@Q%yP@OcKhk?oBfE7!o+BHe;C~1&2#V|gn@*xx+ZT(=6ckZ-8OscN+j*~)O!~y
zyop#pQyn{Ut8|FVTOVxYs+4+Q4^pKPZyq`b3Dnj?U;l%Re;8ZSmI~>x_=|o|b^<w=
z<jlJo%Xs(Pqj$d#T%UY)uN_X^y+tGqO;YdiAlpTn-^i#ub1Hc)M&-yqGV|6rBy#@z
zfk(eFf0r7p^8wxwxF@aQV4Yjqx4ghQ7i8b&yopSjJi80oYE<C1_`XBCs+XW4SISe#
zT^17ovA<F9gZ|XAhfK)X8lHcI_$NVK8E^NIz$|4{a!506$$vce%Y}Cp1$Ck{)Zlda
znZ2;4k^^MxniaR#5=8rd;YEu6m?kbS$j>D8H}pvv96bu#`U@qvC(C6NfMpZZiI}}u
zo+bv;H@HxKcu!g*0j*_RvC@06OSIqP9FyD(u}SYQQuukfS(sD^EYd{a?k-CE#!(_K
zupN(=^R&P5_k(CDz2iq=vw4<>mHvQZ$Wi-i*A5^0wqNQipg+HoGPb=GII9CF8Ce%_
z-IEeOz%9q<?4I&UwE^<-=xv)m_9r^N{l)wI{_uWTg^<zZ^b#o~L#rRD;?v#Ovb^w_
zcU6q+pexYOnp`r`&jR~doflPrtsf{{S=<w~(>H;wGru^z+b#r>-bw-7pXh2VxA})_
zDQ05LsnQo2bHhJx^eF|fE-~TAx2J3pfR@&8aT~GjCEl%DfRdC!9+^M1^v@GrVL(9%
z=_Z*!(Rxeft%OLpZ5Qzm33wARG0D;FyYA@!(`~<qTgnB{x-uBIr~f;#k4U<4--SFK
zmc3|gBtMl5pPJ;_i`(1#fNdR?dNFkV2VK8is3ZVp#DvZM@PD(qiOlEu9Xv^&D1Q?~
zIxmx4T2(IRFD(4PHn?yA&N!{}J&n+BJN%$s#RH&jJy37A!13*wR8??8{F&A_|HAST
z{GrJfU3#4LlfSpRaQ!XgBWQa^gcLW>W1)(Z(q6155<;27O*Y>>g?A5_eA?qOi_?4A
zyu-Oq)jkwWL_M`vjY0@oRDlrT>op+qxIcuD&I)dY-y1@D{!9DsH<<ukG)g@<{<nNS
zz1`1tT=2f_5bdw>x}XQFyDVDZXH@?XA-M1+IuCHg^w=?WPh8CaTy?a>?H0PfN%6!f
z0<Ith1AB@F3Zf9)dh*0x!oB^FfGfF|Lu|hi&-CV{l<YO*#P6)To;sE6d~GjQi1zD&
zWyJP$m;nBMW_wv7&_>Kd8VJ(G{7-lKkGxJW5%}aN(%IA2p#=MxL<;^5KThwUXCJ$|
zpOHlTjBA3JKUD)T;W8LxuI@$78Z*GIB*&AN-&sG}75mV|U|pSW?C0I3(tvJYwFT@y
ztMq3X9u9cOWbzi?AGjwfksyjdDLWH>r}5j09z3_TP!jA;();HC#SDk`Qs6@=s{K9`
z!9NE~oWfgPosAq<YLXxHcoXpetbHbZIIt(ZWgx9_7Wp4)*xQrywnvhu4($QvpZWJD
zVrG(Si0C(forq!|-iveZ(|{w^+vd={RB(a|sQYbtarw^z|3ykr07v*kB;*H!gbT|H
zw_h47AN>9|sU(oM(pJ7-4F2WOqo-06A8x+!i7Gq)jdTZs^YG`VVfp^tZ+A1P5ONak
zF1@hdfB4RMD&Yn_cmva39>ZH3SjWAWi!pl|I>CHqUnw4z|5qeZACC3qlzoBDKKY>&
zT2&Cxq8uDF_FU%>kSwKC*WdZaF5d7{6hLiVW=8qnyzjCO2MBg)`(E+<*z~sx?>dkn
zTr$-gqTEwnL_i#ePC4x9wJyB|k&16_Hv6}we=O*I9EgMCVJh)npikr@R2*6wPJ2e&
zTT;OJ5?#~#VZ^^*1VTW?(Og!>x#u4HfvAo%&HDUMyOjBUda@Ug1&XX+6rIQhZj5<T
ze&GlGzg+;I<A?8vWkPzXHQe2kXHOOtF}d6hDM^kO|GJ*vsV+Qykr}DJlCAJBPJWR0
z?PARh2)~;51<{^fr|>k$!J)$rzwq$Wa^9XGKbS$Y@=N_DnZ%q!mW%_vquH|G?%v7)
zTEzFf=3t(imfN{7y$jlafq2^;;*VA=O%cgC2~_5wSzz~+>l#F()nymx9X-{jA{E7d
zKjA2F_Nj&YO!iTrxOd}2g`CJ6s?>*FdqqeQ7i0bP+-JgVSF@aIJ_I)hL{7H)Z)|K-
zmA=XNU3}j*z5VdmB|*X3M@seeH&mUT6)&t$I@jNPdh()oU9}!aTe?fiQLK@=8^N2k
zIbMvst<gGWP9`P)(o1`dNsD8k=n)A_vG_a2FDT#6e;!+xBDwtH^47;<CzQ7q*959(
zmP;C%HQrQVts&j$h5wpjyK{&t9F@_zI|vd(_v1b1MaSq2aPI9qhalnbY)=Cp_$-~q
z{PEhq;Nz7`@wO4CkcN#SzgVxg1m(rB4CPY200IKG)R~T{R_%n3z0$mYUU6nXdu{Tb
z4}5r9ZFz7p;%g&*zN0C}Pqhk%;|RLe`!7>JQ>2ppHjSO0QI_8J<5p)mvC1RP2@Vak
zm^9SEqiI&$>RG70XJ7kLAv;v7@y0Kde8;`i`k_#9)$EdSWjbBCD3}KtsxKV^lW#e`
zrUrdxC<Hz9c6aCC%|T4j3YF(!S4`t~JycGWeCt$xv7STW5vT^vbDeI)Z_c^|Ssdvg
z@qMGeA+JM3_iSyTgPQu=pD##%eu&!$V%$L&mx+B~;9$MolbFVYJUm^?qecD7URZ$j
ziX>+Ht5Zc=o<1(w^T~G={(@ic?1P~bl{=oxE<>$<$+vQ!xHTW$A(;RHnL+B+E9ap6
z^Wr0R_X~=P(Y~S=5B_8*jZ_!M65U13zQcC*YGPU%WU&-CJAd2lxa=bcF6XrmoVJ%m
z@*2d%3vS$(<JRdd$WYm|dp#>f>pkP7Kai%??*Vf?ZqT7?5W52GXkuV}t6BFx?B2sr
zi|YDV(%E%t`|r`{7e0vgcT<(1gDmo>AP$mnUAH^Xb&IZU%7YNWyd}sr`TXSyCZtU5
z1FfGp_!kNA3n9cXyOi+LsYg|zHQDzcNJAe>ciQ)M+UbigM2BL<P|UTvN{?ti&LA;E
z9M&=aJanY48D5oB$b<}TDJQ<^8uJ+Dyf`=FD#sW+Jn`>VKO1V|S?_Y2@_W7}m9EES
z%ZN1}K_+e`Za*M!9qw+b(c}_No5(_Sqt)euz74O-i5tH${12wRtEe`ovhbRW$@Y}5
zlC>6__zgS^M5UpO(J3T1%XE4;F<jLM;YE+TjDf=@fDjcoZ^hH+dp)UT{t)B^h0obK
z*@M6<TFXF{n>>!GHgZU#*pI;)Un_VS0)mi3#H^B1^4T$>$|0duhqb>e{}=ueBlgRO
zgr0p&1l4#c*N?2^T2;fPABUw|>fU$;bH4G#GGegJ#&zJ18bjsc?!Gzk^nRyd_u9kK
zP|t^<j%hgs3mkm<9u<UYZM0LR)lMHXwMLOz;ez;(aoJUG4?rOgE|$zJ#rwc7<*Nb`
z(vHJNf|-!+<<@Ong4bpsmmu>wm6hVdWg{=tXe+JvAS85u@z8xO;6)vdtA+}@#>Agj
zOfzPME-8S5M=RLOzA#S1&M>zPdhDhh7RfS#?#HJx)W6!}P@zlzbI%-a%V9Q*!Q*q#
z#VJDtXXH@aZ~`Qp0Gs!(kxwpQzCd?guhd3tLQP+e_p>9GJl?ib0y+I6;2$w8k#>Kv
zu2aL-Db^WDFGuK?C#$7zSS9byEIKG7)W$py!w?{_tBXorYdFM!8KTlz%+S0IqfQ-f
z`>VkQUvyyP=DYSezhA@ZcYLO^-d`=Ga?*aE(Kin_Y;s?}bQO6~4zlr}aM6NkjZ-d7
zCA<@mOQZmNgy+Y+HtD6zPty*2)EeCsB{MPpc|%k@OKSI`LLHy;-hob}L*&vPnt;c7
zR3Jjw<gfYfX&|KtwCMN<$HgrU`xo*;zG;6|mk|m+Z=!Rb@iK?6MGyUwyB>~vOQM0x
z8aRc#e<kOAG(_*v1#{t7jQ2lNR@?!J*~O+~FuSSoCAUyEDf#{>ANYZj9!`g3a(+#=
zQe}y%-Bp6s1WXC8f%@bAUOi0Yx->LT66z$NdK|6TFrv)LcZm~egJ1!RRwxYJuDUOu
zV<<@#`#6;G6$^sVxi6QJx^j;^GO>SuufFZO!(Zefcc8f~Q=LyBj|J4zJrIF?)lA6!
zYD(nE6xNqUI-jgOmWJMW7V7T#>B~~u+v$Jf#wYy2?F1Km#&fcN4?L#h-wJx+AEanq
zFn<qU6hiREm4?#YaXP24v}kxcfm<$3YN-4;%&*eGQa)!p1PQdhslPaNHOSrOkc<I_
zCx+SZo<hx;7J5DrjOfE>nuBC~<=o_?#WDqIJE5%WeS;Ky4jt)HVx~#~3_Vqj?wm1(
z8}Fr;KN4SkWwoFqC@K<jUYk89sL0l;UBZ3gc1(ou_0{dkyV2s8`2`saT5jt`{Rt{5
zdSYD%6e16B=?vpiY^aPJwBE6V7bv(NgK=OJ!NL0sDi(F&aXzf$$%RtKrwq6q{h+G6
z6lII&o5e5bbKMkV3xr7h$gXM|YAT%0xR!a>%HJeI+el>~CTck5IkAT6=)0aq{pytA
z*Gl@7^%d6g^Fk)q3mgh+p1q8w7BaT#$ahzh<<dcsaYy%**z~+9xxbYf#pfoctf#)V
zDyXVn-zc2@Oid>zU~)scT`cno|JB|thfGWVB1)_F2rcrf2{g1<muS!Unv!HsuRHvH
z)V3z`HlSm8`by+msZ%8@p_198utpNqGM&+ka-}KmrS5We#SqwQ17*5qGyF<<lF@>J
z&y>fKIZ)3<WS=5@8-3EP{z&6l4-;j&u;-iap9S@q+*egRD}G+)1S73d^y^nt=1X^j
zO`0_7;%b{l-*qq8POd0&Wv-m?=ld`)T5^D@jwei`-gHdcaQy4!-Fw=?YOim@7Z&Jq
z?Ok*#E>{aO#ED;SEpt@l(ybC|PVH$I$`g-ZEMw`hMA7=zkp@aLKe_(ef7zkHsV2sR
z)3Bu?oNpQxL2X-`dRJS0uuOZboA;Mg<R3SIB-XWEE~T>+8qtgD=nge#P+JvFYs1Ci
z#l4J&9&;i`-n|vS8Uiz2D35&}YOlq0t)bGO1TVkhnY$19weme8`10gclb|ZQ6xMu|
z8?-chch<gA3*0l7r{~qpyF`~7S5@Geulm}misOcup}ykT-qS|Hw6{7o&v0pzKbS18
zn~4k#rfJhKbA<8}n=JGB8+~HCYV@&=8A>~nt`Tv44sfwzlVtP;1z|;*EB<hsKQ2I`
z&r|&sFQnPSc}!vDK-l2idvvJn<WqKJGqfQ-wp>F%wK1a%95xoOOUv_|ZCHApQF(La
z0DRD;!A*~Qrdur}v;yHyzn&cozp~XB7xC!A2n~uMHfZRw%n3Se^^_)!%z#p2{pU73
zR_U8C1Nmom{i0KKk8H03B}L@b6<c1{X^t-DzfpC!)9)!n{#o9%II&URYr5-AUsrr$
zdL}XqYX`6MZM7Laf|#iF$?uZHZpnP^sNRvFiyJ?}_V}f|6BbEfEn>i>0y%NxAS663
ziwUV%51kK$T^3-aile|L;PrXrT<UWfxgV@mN^)zp&QZ}_SELaYc@jzUYFy^UBS;&M
zCIv!Ex74I6bfThl=W|@2UH&^hL(Qd#N$s)<LDLp3as=)}jUV@Fx6mu`Gah!*C!#Oq
z&7WeO>MgTW%XGChx7^Oijlj&}lCqBMl5DAwm2CQq0PFC33>+b>OnW#Z1z9;C4#TW~
zG~*JtZW(8<l1Bc`(6$G6<W@weUfCCuLU}GWv}iZr;;^30)m|0N?dY8XRT;+`Pa|1)
zMVy9)#*GL}*`0TJmmSQPD;1#h%kr`<{<xIFaKewh^592LnFsFJj%DOY<XtT5l0F4}
zxnLyWM=IqAOygkhnWUY(o7d}7$af+5AP49DFPLf+NQ&-(5F2VBe=(FXoh7NUGSV#`
zU-{KR!-~VrDR2l>z0F%26v_6*?d}=^03~7Xd#$XjP+wV(-3;w*Ks!|~g#oBLr7U!u
zwr!kNw5&$4RnDb;p077>zj8xwuzilTir~4}83iRB`|>I#>dMN}qT8|qI(g`+1WY4~
zPiSP17|Z9d*7H0k(+SQ>9mz|fs!1i5sb#_<Z*{4e{fv(3La7x|U;4_$>))DyDpUKt
z#FsQsQpu&IE*=*{D_S3)AL7QSSq8&l8_))WGnuV47X0qg(CGPEANZna62%kt+6|eR
zdxxPNLR1OkZBQ39R#I(p+RaIupKc;?Imxi=*_UtBvo=N43$8}2xvDod0IB;rS_QW=
zHw{oBGh($X3IwxDt06b$CNkm1LeTm4)4sB5JvQs16J+S3m~dEo3hZyHnUvoe__&<?
zp5&0QehzjbQEM$LiLTyWj*K@=rZ^1I19-`YxoQt{5;Yi%Dui^D%U&aRR8Xnpo<jC@
z=!<htgiF$~y5JH>JzqDsM|x@J4SXCCvJU2Ew<37Z6;Td5Q#V;RG3@)n&0>w1S6R&o
zv9afOR!~!hb1hG0RaH$Qjx~`WD=V23JUv}1u=CunvMMWx94}pO#y2YP*)_7gNc*}e
z(l2T_lNo6=NJD;Jb#k+@ob&HSgm>ZoovQ^??A|xqWVF9$QAa_y)>{()WophBK9~s~
z<k!cH5Syji16Wh08sq7>S7%3pkHcKiw*dIp=6K}f^D{Lx+>$7)IS8R!H*sZ{)GPZN
zGmct0fiXtd=VOga$*}^A2m#5e!c=b4f{0!Z@KX|1%}G_);%#*e4KK@{^vmm^xhA|z
z;aqctYkf#Q_c^|aq1&Wp++JJZ)GqIiHo2)?*PuSR^!)oQK+4bH`KQciq}uGfBB@iI
zsgcWC<+l(GEBWHCLHB9u2z%{c<kC=#LsP%wc#dp$dDY=&)DZWVa3-K7`L8eBs_(Nw
z@7xd7spIsLguh2t+BLaXE-}}ymttOC6&WQSds3Io*Ny7he*D;fv>C%W)~Gho!sdeI
zQ5ai2!W-BSdotUvW%YqZeY8e!fS1_iyv!Yq#j5k_jJYCD6wm7QV8r7K?zN6myJJ36
zZ{^$(7hJ!xTWIpQaDRBIt^al~TsJE8(OL5iC5Vs!D~)G|vRyKX%8FU~+Lwjib>u;a
z2|NeR1iT|lBt(c9nL#p19J$=cjSVNh8M?%{IZf3_?Xh&jrR`JDX7g)i!{*c#pGOJV
zRreb@HZg-H%TwVKhamM;cWm7!-%|29CP|CyrDwKLfdD00?Wn2ywQa>GWWq-8jMY<z
zn$Ll<a0gt%4Q*Uw-rcJ6<R5G@Zs5iBMy7U={!t35%jjaWY*Ix(?gh0?{amH#)yS1b
zQWZZdF+{+t^8450JHV-LhfbCaY5)h^P59!*b;#gt3CtBqIn1Z}YRx)et$s|`fuXQ}
zMD4P};n#2RxXF6DgOJDX<a6&MyKu0T4`JE*u`gg-tL$=+q0G(>J$Wj<78s<c-F*iX
zT>>z>m(VZsVT1GV?Nh8SOCJ!MFq)*LA;|ihf?leV`n0hc=H@vlpgLvlh3@jFzieyC
z6w+M3bo;NrgYTt^w^yon5F`;BP@k~vuPiNx*Tk*Fc;(VsjacL$9|XNhvj=Btg`$)t
zI0qS#8K+pil;PV_tde37cN`;P3;J!_PVMtm06UY)dR=EoHw^PJszLgva?(0XR|U+n
z{5oALQQLXcRqJ(?oY$z#c)eWj1XSEkZkfz@m7%(eqn{w&UW_y|_6^eo*|JD6%KeMX
z+T++yR8Jrc&&vAQk@q*`cTV4)7-@Jbsu}{5X;C179XCT#4BW2D0BP$N`JFaK+?qxc
z0WIc1&#J8k8FX&+Uy9+k5avtoY9DTAXsfy?yt5Gg`YvaF=j7{W)$SXmeXgrt+PiKI
zOK|GxmHlryo<yhpp2WpYo~{4%-e<49WptfQR(S#r*u7_uZEsLX&QrV^f69R&f_kQQ
zbm~J&+rnN@Ej~Xlm!>Fw2x4;i%wfm`*PL2Ovc}4TorDXIo17;$WE(0+`uk}a@$<_j
z0afUfIQ~55leCf?GJ({?*wz(S%{)sNT>%^UxUvE#qL2SNejN^xs#*n|u^4RZXn{gK
zI(>u?ZF6N6PB5rxRd*{y$PNzB5`t|`-IVcfI$#QXXwMZGDs=E@1;^Le{AxySZtl5A
zG=~r~@_t$l0NgPRI|1m5!IX{ENay5=X)+vr?KrIQeg+5wmpNG%06@JP*(K2H{4yh0
z)x0LRz}A|LSgQ^Kz0TXUkz-x8^d`J>;dA%;$JK0CZmt?V<$kJe?=FDQqe;;d;W2Gy
z=*aNwRn($a2+}*vT*NSt;dxwHe`C#u+Tq=WCRepvzsA^q+$0C{*ExiT4X#7K^b~6g
zzkxXS0xJ-$N7>x<5ZEi*FZvf6X7m<<JBC6D26P-PJ7X;b^$b~Q7^}KDI?dTD79E~9
z(nrQ;+V#=Tz6`jP5|GL0x5+e+OeQBy*6wun8_vu4a`0Cj^6#lx7l;`fsEi8?rBmeR
zL>d)OQs*r{^9!))=4?9uzyOp_l<i%|{`7|4jl|otnRp)BWc$JjjYV8`#0G~DCo7GE
zQ<EEIY%UBVYE{8-2*TP^06d}a9E+q%gICUd1NivVqEKWT>CI4Nr3CIVUV}idyCT~t
zNE$8q9PI<Bd&;932}!ON@tOJ*Dp{_~Ps0pGPl{_%jB(_uY$yF*d~hqS>CL0GTAugM
z7|ZvHq9%@HFRxj<pB#l60QsXnrpAZ54D<0W&2TvlPgavgZ8}v%U0pRTol+A=xs8%?
z@$7@&VDim>bR63r1wol-fe`T*CRZ`}u;aq9Z3c@9A)(o4kVU7H;Y{k5yzue0?0=Gn
z3m9w?#bQ_}V*-d$2uvHf0=36PESrM6hncaPB6Yd#30?t3`M=<iXDUv!b6d@>hw49e
zx@@(oW=lxfYnHDX)x4n~k|TSCY|<G+Vlqa@{e=Q)+R!{-FU(75jasx<^9J0BwY=pR
z)j`OEon3AQ?5svni&@;C@XwUpjR%hNUsn-v!zLl2jH%-`B8UovBLY5NQ2-9T%KZj)
zeVxwB$!mQExW7UON)Hf}JageXRh(!~r~I3C0;*-lZ-%17XdnpOsIY2Km1|4r#fjt>
zWaKiWye`o>##eH#_(#^xC<suJ-n4Uydjw%;;NxWrvVi>a<EqZsT2sA-*tLd0-eam~
z81pY5w&E`s|CaV^Ph|zX@S@|smNv|sZMF;Tlw_B&^&>NUa*#nw31M<pyjvPw>r})l
z$C-q2FZCC(mY+cN0U;jfwuEo1uzHbiE&9L>=~|qMd7&K=$8H?=4c|GH%z{R6t*$>N
zqt9*n{5tBg>$L!C3E(bkBcty9UFn)0fFIE)u8(dKpwwy2$bhjmxzjs+e&1)ZGTUdV
zs}OK@_^gsr!<J5b@Ww`vW?2Qde%uHxX)D87Gd)qZAv=8tBFV6&U#6&j<wHbG3?z&1
z?)E3@ONN@pWz+_;ZnBwLMdy2(b!dDOte%r;1mwkO@Xp8o-l;vo;!QD^OUiN*-9H6R
zHDM*08hKqB)`y{A_zf@us!-H@iPBu%zEHP#ul%7yoE||ib}EYGWUgDvO|Gm%$;V(T
zk?6+9?6naav)1xCgJuY9hde}fM7YJ#{P|_myR!4c9b{Q8zJ@g+5!5|-j`CTQSys80
zY8C|rS<k%+O-J8{cgH^N=pMQeTfq_JG8Ecp)#!J3;qG&?s{(xPAN!?DYSV)>i}ILb
z1hQ;LGVj=aS-Z(+<Y*GPGMje1VICfCqf3@iq)*eb;QPu&^MUb`?DaWK)EO(z{v_^v
zktbpTm!1p@avj!c%wUHhb~#6#7Tzz3McCw!!pizNk$rWmrQgyyN^1!OvC?snE$j!G
z;7sr%CRU^t(id*1;6}+a6CV2<cA1$~SahO<7ZGQ-<}tv9$-t*^lj&S?s+o2mmptQi
zt=YP%=0g^lm4|5D4GjxxzbFxNHARMeHW^Z#@^aiLr4papmqh3B$LHGBUy>{}Rkj4Z
ze%8~Fkz}Q-Q$T7U8ufZYsa>r4inb|fkYZrH#-hg+LhpNmcrz<^Wg+gT<}&yz?)v9U
z#`p*l*veJ-U;sSZ^tF9qfyNi$HSR#|sxdFYNo<@i6LO9HuaMA=%oHZ%{SO$HZMh&r
zx1{l5&J(J30!n55APBkS?9e7Bvvhy!K7FrMek4{0WSFRu$0NSZ-cyI}zzl{CK_0sy
zSwe*3a$3h&<m6|xtwTbq;Tm6tFO*6l?|T6<UXA-iu&BsaRekNYL7uHN{jt<97b*(g
z0$!V0^OkT!BIuW9_%`mQU6Thv%modkss|z68QDPkR)Rio-8UUWGOHLj<hW__coi`W
zc1BrX7vnv6v`dqT0L%>zGX&f)wsmU%88GB@oJ&sLTqlc()ypMk`%^OT{@bVv4Rb7t
zn6{kw7`wHYe@`EO?n6j%AM#xBRqbh$nXy)fTF>MRD#8nz+>XK0H9(6II$-AmACDW<
z_;T8yRw%X&l>L=FaW4p-O|0b)5fb=TG4ck|5@lwkdQCxm^LXT1<H+4%21qv~4{&J#
ziapPVFzRH2CnvsDHgH$MD-5%Jal*eI5%HfSZsWrITjd}^qR4WRsRs%!6JBkks&+YH
zZc6%7UQ3G(+*VuI!b1=U=dBhdBxvXm4tYRAK{%ZJFEL8WH6WrxV?BTirx>uT>f5=1
z$yYu8fkGNOaU2wL<|qwm=uY!DUH2dT&y|kBj$b7t!Y=WW<gCTA&@cKB)8sXebAYCn
z&V!IrAtWK#F(WfP2swnj=T1j&b^L#^yffv-?x5YD(U+(HIJPB{9Qa!;%jY?R*eg#5
zI}8f#)1F!Yh^VKjOG8gbv0zijk3dR$ec<rg&fW1Xct*|#J{YQBx{p;Dlu1&oUe9x2
z5gVtIZx!`{#}W!pz*f-L8%8*<hlJ)i?~azly8Rew=obbEXbe)NJZ%C5^rUD@L#?%P
z1~bFP-F8;=A7hip+Y~V?!c}3vb0@WQh@Y&z9c#Tt7{uOY?nn)HFgSShmfUAqJgx<l
z5*(mZ07vtKmf`QUAm$r#@=q+ybB<H;<j#LoKFZN+^I`o~AQLO-N1g$vG${9#XeQ(r
z1_9w~Kh6XHy^!)f4)TPd7O4FDux;>+B3bep?f}58x&Ct|)?*5xo}o*%Rd(2y6g(5*
z)o~gfY*cUlR|@u9co7PYK_#yuHA6xNm{~Ci!~lX1@|BO9y&M+`Ya6*$mC-9WInx_S
z1tN&D*HK`j>(KX<($L0H^WV>#znw66(-H%<J@y7g8Gu}%zuJ8laI-5Q<gdRU5GU3Y
zEi+^0{?*i^e2;-4oZ>{Fw!#d@kWiggPOp5-qEL)5Gpos$OTP>{eNQH(Bu`xslxyI|
zgjXkPEEXaww~7eQ4|~ugk8eOBx{Xa8`VfC4&G-9%X#i|ROXUK`0#q}(Qo7^=-%*tU
zbZXz?l#@4&C@5w(pnnk?9*VEL4_s&6Dp1^ic!z8s_<P_2iJ&Xu!UtIKg)NiVst`cQ
zWH$+{pV>>E7p(m1?(X6eAp;X3@-O{Pr~-^SQj)?%`jaZN#6^k0866>EdAU3Jvnq*T
zSHhs(IlJ&Dz)v32S!QD86(a*`^4Z;Or>c6F9f!?!m1uuBlbj4+095TfSv9!wWeuLx
zySsrMya%~o=cF89%#0aKxKP@=A(v*N)}f~p650v@#J${IFZ;Z2fT5f%j|<=y3wA)-
z9x4EHMzqs3rvSyTnOIrj|5813Ff@=ZjN~l+Pf_Q@Q6d2VkY)lvEasVNy=qh3Pwc+G
zlor@NuWzeX9^7nlcOERhhe()IBETFCSCOY}C=;s*&EG7km=LX|q@T!i`y)_!rvsAH
z-}ke}ABVyCfv5Nf(m#>;&na6x7zJC$ssPmY+z1Ih4gERj(cXBQ`S}TF8r?tR8}|r9
zUqr5tc_2VkgnxU9%bqYiv~=|xz#{0SLoe(et-$P6hg91`zz-jA)O)E23@RNEKLF%Y
z3sS>J9lNMss{7FVXEg^ys??H)$V!eI{buxlYuNuT*FTwis$NJ3{Q2~=kH-9{obity
zON)h}4%|#~@)-Mn-Nc_VEhL7mya&0R3BxtOz3hKWU<u(JWSo5?{U)H9@A`i&>jP&%
zc&|yz1na4ldIPAu`4x-5&;h2_63wK5p+*78sq7w59s)#pCjj|28iKw5K*+2jlNP<W
zmTt}8!yD;;8(2g{#9eFaR?oJ|R$QDr@AkcW_x#2=T_n*Wm;}r72<#OXz0L4J*~QAi
zxxvZ_p+sKy>KL^Ck_XU+y2|eW`$^Q(EC5?0PeB%$=lH)(^7r+8;D=og0~_)GFOcqK
z)BmUC_+Lb&8hI;Tv+r8oB#GbWy|#bHMcD?)`cD){OgOaFiNLmQE<SJX=s4Zd(t=tl
z?JY%wEj>0+Qc`;1&bl<E&psj$-Mobe>wRn>-(t5kp+7YnRbepiAl_TrZv-MC1#qCi
zXrzz_w>t4nYO@Jze@+=(ybWfB)^%I~5c)yJ_6HEYGw@%o;`=>7cwp}HlK`+0Eye!?
ze&{5PoDheHLsP^hv#w;2TCkRbe_ds!ZuIN#qTWn6w)gbVMsAN*zXDOF;A8<0$EQL5
z3;o)?pPoOPl=gjPr6l6t;uuAX^%zyXK9r50K=Dm2*F%Nqt0&ZLKiT4=hYIGry>!V0
zvMbBRCF1@V*I~MN{~*_bT6H0iP-gt!TVt!{jQFhcAt(fo1a<%=ReSTb6dSp!z5sX8
zzykO3T7cW=%+p2g{G__eR$*Di&(pB;bQbE_exj{-qx9HxnGk+kY;VmDV#{)4i`2kN
zZ_ugMB(3fh*&lKJ?OVbtPBx&<`8)0uW}_tP-7nSIRGDm~__s=M_YXA7L6p9B_k57G
zI{VafTmdEg+Q4h(At@|KVi``&XR0VBjvFo+tB_1KYBA0ekhQCSWL4ykldKdv2yv2W
ze6&y;d_ZEo)RK`@&r500t9GO_@cOSqtkXA{FuWzkg;YlXd9?Ci5VpD-xY+NqNU946
zKKDUZ#!bh9D@vfeAgpj0IPAPH2X3l|z`iP4l`S~hel`ngs?xCz=@{%dJ(glgLm#ck
z6)h^P&ZRZq@b^%5<>{|fGL@qb_<xAt`{3xT-9upKh2#W$W{!e5xI>eCtKyT$Z|^AJ
z^xx)qu3=Bp+^#0Pm^Zg#FM9b0P}Ob08v{(pQPS{}zl|VyAEr2Ezm9%&9Jb~5xQxMI
z<PfCceMFA+O%*3YA9#X-<0q^nRZ_imf7%`8*cLN>CEI4(6yvt?$zZeEqVmcsuPr6H
z`$hYWdXmuf2a0tmlj^NbmoGiIW+`?DmZ@FDWqPBUHNo}kB1v{Q7rW!vBNEj`h*i%s
z84=vi=||oeDE7P7Q}Y`dXX@v2>ANmAM7f}qa_!sSGuYgH54SSPh(|wLj#Ya8?&?!n
zi@90NyxaO>_?oNB<G%QL+mXntlG_v160S+kpCi0##?V7#SDyqwTX1qyQ0zCYr<S;v
zq?zyPz@<~cr0>4+BI*jRXnl+(e<ryPFOHbCb;pooZw}Gq8k(<1M9_*^YR+#Kc^Iw6
zr__GT9ZO0>LezP*PV<Umu8BF$wdCG*>}@Kh-zl`Iks)WRo2MbIYpKSMyh0cD9+5b!
z>Np@C!L3Uj8Rt2B*eVE@vn1Q}`CXd0wdB&Ev+0)jP!X?_c@89JNN{88lb8Lh=z{5L
z>1vp`Z!5k|Bgy3B=RWh114fP6<)&y4bG=G7eb3qFCHO{8!*+!St`$=ao%(b-OSxtN
zh&;6uG<@fKGW99_Co8D)7h@Te2OZ3U+wR)6o^)Fx4|9y8O{%krp;w+uSIf13w;hFU
z?nK2@Zf+Wp(W|p9=2FxC+cW&9z9zhh(zV3g;Uo>k(VEEe$jy{K3_l8s;Bvkq6F6k-
z>o4wtkLPp2>g7pa+p+C@T<!1DCYisJwaqlKXu%T@KT_bul7HoRa8Zw-E!JeJb`YgZ
zS)HcH<~TC-wa+xN*U6<*;Y_Q)4v$32z}NFs$zpbBrRBlqBZf;(?w6kgo*i-#v$C>X
zMx|tVRaYN^7_nI~U_Dkk`RvD2Uop@x(A(lAd4gFKFE}2B>8mx68uH+GoQ#%8y)FsG
zTzpAB5TnL9HjCI{q4T2<Jxt2T;N~?gY53l8zzGw;_N6DV8e@{FTb6t`Hc&X00_(Ms
z&u8ED(Hj4y>T^ww*#}C3MgQcHfk>p5K%Q5*nlXU_Mz`eyr%;JjWWg&0(Hm4d&9Bwr
z@^ciTU*x$?_?4i`6kcUHILsF<k@49j<)S3ixpYpW_1$a>3JdG#Ms^mdMcj?WRzIzk
zptlryNVz2CfND8nj*tYUm<^KWBit|a>kOQazOxb|zTr1H8KIefPfgcjF*?@{({V$%
z=tF{AeSt7zl!w>MD-7Rif{x*`Ph@^Rtf=1{-RC|KpNHU%5oXU8#YeWcx9d}|Qzf$)
zQH$F8_ubuo<cz_XjUvqu0o|ODCcZD9s}{Aj=E(<?iQFqK(BLPZc$O1?>Z)vT>yzio
z!!EMTuAi&!>Q#!=DWUishw+K)USwWPSE~J|7%Hket=5F(AoFit$F{JyIZt+{IpB9H
z1@a^#MVMJjQ&^^{1wF%@$6wDG?c}%BguF;H9O$j)W3#LlI_vVK{pncU$aZ>|nd*ZF
zXSXrDb$tyw#*(oPc#q}5fFnkj_ZU9Qq(Gz9G|_r>xA|!L-X>*9%H9NACyCEz-KS4I
zF`Lu^=CF<Y=;A>NUa^(?j%PGhKfmO2?NkC*`nq;J-Z7{;&fbTKwYGZa(z)N`)*8r@
zX5>@xp5TN;iw;ECREL(=Ri22x871TqNU#zZR;pNtvE<cJ9uUwk9(jT4#g{FwtSh*%
z618**&y(j=Y#Zu2?eKonocwT{E!Kf=bvRa-dec;A>6FctQ4bV4G@1uJ<UVm2LCPS)
zbrKt#bL~M>qGplv9bNpCfz0W}amMn+2bz}Xeqqy*7*8B>dF?9tOl`v<i2U77u~gO_
zm+q0gCLc<MSnFJ+W{+%t6dxH%OOO!7#PxFujw3B92kyw$-M})2Sh+8;Qq?u<2(6tJ
zw05t+DHsGwUwvvCIZ?J&ePeDda)X32nc__)`UqYXl-gWN4lT-FXIIT6i>rEBwl9C~
zMED!e<c^PA@e-Y=(X4dG9M*_;{B&-)|LVz>DyKQd*5_Jie1<At%6ysI{Q5gmvyo~H
zx;hXgz|fc2GOfg3LspLn@vx~A)Tlk6VYS&Noox#nK@wA4lXY8ub8$#D`Omjkq`=#F
zGx#=X4N>^_n(5Vxitg{T0QdCr?R?>G*`*f~JnR=TpC}dei41Q91sfZTtQC*cRZu0g
zR1MqJ>lwBA20d46imaW$V{z*Tnn+=x(QF#qi?Z$Q!#(Dozb7KOuC*&ATvsG*@J~Tg
zVhAmM-t9)7m}Z`=!k2b_XSuUG?K#(*5~ca@wOY1g2L4Of7sRm8RYu%Y;PNLuVfd4h
zYzJ;%xKo0i_6a&ck%;D4sz|S;wO0akq19c=J9d&+zU58qC{^aAQ^^&6FRBZARh{qx
zi`%O%@d?kf^oz|ZCOjBFRw0_%qOo0Ja|0~-H-}wT%uzU%P><W8xZ7nSCvlC-e5U#G
zd~;P^wOLaM`O^8@;)d239NyHd;PS6th;Xiv&IuCzY#Tv0kezl>g6OZ^n>oldvC-7#
zPlZT`UN9M3CLeo<_GH6~pu{|vw@XpIrCO{D3s$!6ZTdqxERu3A8!xb<2^PE@BGdHf
zK~YEjV*B3isRT-eGgTE;Lvy4OY_I3vSO{G#nb#O;u^!f&moqJ78Iv=m)ZO;)^gk?A
z%EXGX+G+B_`dZ#sZM~Vn#2OuVfBe`;^GQs5tWCg7q`t2GrV+bxP_fr?2C6MInxx^x
z)9tGvE9o+xA$9|$c&fHx;mWX?k_Mfca|YYc4V(`CD|Ws^uhgj1swnV)Bo;FtEPjZ|
z#<|zcuw1zEN<=e)D$)ZjN~dd96P0Vb_MF{pWRBw#WJ=57<e#3J@Ck!050URj5{21X
zrP7U7kqLAU5<J8Pnx=DOYowPXd9YjQl5sQ&2GIx)aodcAu`JnaTp+GGem!d)G<tpE
zG{t)K0x2Wv_MgIv;%RLp=11}oL(AyZvBxP4S|iqYBS{(~HcN)4NFF~<FLV-*jcdQo
zNk96HK8uCEou%e^eU8T<YRYfGN;i}xc}XtjIAr<2<wX}Mu%6pts277S?+~MDUc+i@
z3^6*|qvCknhk3I$MFc}J0_{atK%QLD_<}K67uU&HUtxxrF^;L8?}?wn^Cgr`Ev{QR
z_PHVY#-^Pj*0{2?N;!5`T5@chZ?KUc@z_|+ZRi;=9x=FxEsa`#dw_2#I>qqIYMX2x
zs?ztDb1?$667ztt8eKO`Y(QA+C)AB|F%Fc;ECcrJbIlWs(Tb{P9}GfJ#nzSMq{|(d
z?0Gq>zArL%zJ_UTHTukB$zNMixzXd^&Ce0Hi8$a)moN+E1LZ7U4R4GOV9gj@9D<);
zmDqetUAkPNd1(n*vY~7S;=X9bViw;*))6yF<&B$Km(h=JSb=vQ_&=~ko2xDYIGHpJ
zp`R63d1XDw^$=QFni3zuNnawR1<ro^{@~^SiLe}1as@{?kPaq^vF1Hh!qS?iC3SKf
zmGaPP+q#uLcH@I#zqbXVyg*jR!eSdkXJK|?g7HIr!1Kys{7dTU?jE{|k@jIS|H;;d
zCx*ep_8yKs;k(EN4P-M+*OPt(L<w-TNr)7U@h>`_NjJ{do7L#yHx<kux@65k<Xm>X
zJRN;Kvg7LOMhWzGvw>;il?~!7=NWDuY$^1HqEIoS1ZQEm3L=wI5lToX;WC6@Nq(?&
zL(a5W7rmN){+Q9(-|`KhsY>v!?OJP2gc&IPX9<TfSr+PZOi1O(jmB8+upBrTmC2EY
zUQuup|EH3%{QBnXMt4qq9l5YSaM1y=Uij+kjLF5g`g)_4)=opz`^UBoGbxVPH}oE(
z5I)1^s0)r+1<K|-rgOdcmHDXEO(S#Vp~N9KudkA*i=$uWa{RH`C7JaSu5`)sC1{D|
zG6w}W+*rM2z@dCIv~vrL_5_WDn9zv1wxZ~C2=x_f-Hk5nj;Fbp`-gR;eG>QIrm6Bo
zDmQw-#D0w8drsx;$pvCSeeKMhyWcAq?OraFBnwKG<q8jG!t1&PRcsMELc_>Bry#b)
zV6{FUBsmo=lcbp9sFj%BjS|$hPO!6q;-Kr(yA~Gy3a)*YRlU-U(}CxYnS>^lvo=99
zRNAs+0~r}*J(*ifS9NLzdaZ)=E%PzPV{66-)~Og{X@s}aH}(0xXA_M~6XSA13QI~M
zq4OkmpQfrpu9qweSNl;hc$AohoaY{19g%!C86mwq-6CAKrog+H8$5g%f+6}OPGvwH
zJmmTyxR)#CUC0aOv1en-&p_E{Y;~IXf1O1V%+0EDm0`II7u>jGzg{XC$VxorfRY9c
zweY6My6|dA=Kg_xQ9Y*-Cl9VVD<M|(9Br~l+?Ly%aE3Cib$QdAQL#tx$AkxMf>d?+
z!NVgn+tR*IYwj@Ov<ptqQmh)d^H^`s+4v3XAWf55yrMm7>Vw+!$97_K0@xR;A1th=
z<1GB670*6sG$&-%n*28Ux;vGf`9!K;UR2x2o`$n{%kSj2$x6P^6S;@sBpw@GvQMz$
zcTWmrjy4J9f`psOG=4D<JqMERa|yA_w+B0FPedjp^aQdg1{gQwj+e~wPSuAwWeF_I
z)cCZGxU8h{S+2=^4|Sbms7ewT!TrUvBhL6fB@M3St)$S{p2n&Td|){iBZM-IjlvH;
zu<Z_Rn_eEi^mFU)Z&x<0Pbj;UpMf*b1;UwbkKllFXUphAw7UqlGe9&uFd9ABVREGV
zC@j=p+`hhW`R+Rhc57UW^BmbyThPSwBa41mn=(URTM`<Oso|zIl1fWiBe?aWhU7fg
z8!71u<UyD8zI@P(u<jY<wvv?TN;M9-2IKQVJ8Q!E0*t2O8&ZQF-6K#=8D8xr&+(gl
z4hUTD@Ej{uZBp@IfwWChU0`sr*Ng?<-RTg`T<4p*+Y2@h$905`YJ1}Rooy$RayDKx
z7z>Lb@&*=B=#GOBa2nH@i#jQ4>Km<aLo(}bNoOawSxn4m-;z6T%n?!mNYBgG+g2_D
zJR6wl`h0~uS^8CCJX;%(_99cI;>bm-kWl*_Li)y28Se39sf>AXXYSB-lF&~kvBDLQ
z2yV?!=5<P^Sp)@zsfOkyW}K#{+ikmqt7y1=kJRVm%Y?15V^5k2*T?7noP6<a{T5g`
z2!XY?1T<}53!YuuX{!YRRYph0Ld+gpxV%E#mWf}$s5M2@q&pB@2Qcmtv^bCPA5Sp_
zwGE@&#r}Z-!}n(f>Q1^X#*?u0WNK4tAlyt@1T>BZ(;MK6`Y=jIj7Rjh8n>O-7Gv|8
zyhs>?#EN3J%AzW8t*Ev$EnbknOPgW!HAwm`0<%5LlBdXv`z5eclcXaqdKF*0l2Ex*
z2~C0x9p=|f{gd4*$}J=2qir{r*^SMR_TU`@a;M=lZNpv$1=;%av5KI$k@y-yEppFz
zV_OG*|3-5e-XMO9R=_;p85<eYw#eF1<O&Xe)+uxEZ_I-;iEEbFI^p`7g`!{$efM*9
zmJ(OgLiim<a7Jk6o>oU!)2%h<waK~sDTjibcn^3$R%Mp6(#jF^YcDSY%ls+2JR4S`
zoBSCpE<B+|Eq;`_?(Ll7D@rJb;=Vaw#)cxdndkLm!1D%9`1e}v8);|o*LOAdmEa&t
zW@%fOuE&`L#8+G|TEgTb{OfJ11<$lZhj}o_1kuxx7Y*^ep5G{wn{}GPOx3EQw;%Z0
zT9yvXjb<@!s&F}5t2cXYzKjx{i`FbOwk<e8b7rh4>a^%$z731O<Iu5w1cLGb!u^#j
zG0C4PXBqgqMb{EJp>bYbM*@TGhqqe}n(EXHFKylzD!CxWP%^=|yy@X&Jd!OF658y8
z0`-0U*B5jSWsW*?F(tUXRl~*k6u)6L{sYEnT#%o!OI~&m(tom^n<}2XDUxe8Pl?qR
z&KG;b&tqpH={k29k?wN2fMIQigO|5~Ojenw$^3Xj?nu2PeM1#ZZO}L(@W{?G+IhOn
zqI}-8&1jfD$E8U})FUBI6xZQzxLHSMW>%FP{i1{4UheExZBAY4ax6vV@WNC)U7I4W
zgu(^<qdQ45_2=*yBz4tt)D2v+UOML$+ofQ4uA-<Sg%pXNOkEl;%w+$>2Rm6-A9!+(
zv3U+i9lk`Bd^kCzF1lQ~j`w726EFF4+M$#QSsoU4d)6RQufzNd>@g$n&>y1;gill;
zJr%eK;wvM8r3-&y3D?c5>20+clq_d&NIZA77oobqG%;+!8$^Y0;c|2xtI9?7h}MPN
zVK(e;oFF7#hLc0WwH?=q-Ol14K?Z7t%I9f}*cDq_W#mmoxLi!Y@z?l=<*oxxvZp`V
zJKf@!lg}1#oo8DHX9z(`LL>Z7PB2n}1MoRsgBB*%QaS$oJ5P%jsA88Y9yr!rRehk4
zshh_#DX|XDw3=E_T4Y1?4VG~F4ObmY>}%orkG5k;4ik-0l5Xl^u493GQ_t)3&O!46
zG-JTLUPAna<V7~6A&5*6O{Cr{PN!rNSm;K>4qG{cqqAe$o0rToCTMf{Aeax__<3gh
zhA30f@Yp)vd5ioap6(McJG6-TY**hsbZ`A;R#5;8H8g@tr_f%jf`95UAt6~FJb<q6
zIJ7Dn&PFr#Xh9#GWexHXP9*4{z%_7*hMA$>A6o7RQ?3~+9*KyDjQ<%UQVHVc4=S7n
zvjMP5YhO4V-Fvh}CXj~lh}Y*;ln&?}nla2&G~Z$#OmQPT?oGs@GG-7&oj7a#%jcIU
zK|*#bAg@jIG#JDI)z_o+I|9;Bl4ZT!z8mFjO=t>WAAT=#)1S0}*MV!C%)e(rKNl_s
zAtYNTz@2B*L5FAWJH0)*Zx?><kE{=X_xC0al7S~02@QwtH(;3H$tKTkA9j%{{^qr%
zzL!I4A56jg!)W=XpZEW6@}TeILFoGsgWmXmG$+Bw5j+`>|2Re7wcSkww*7yyi6r#(
zAeh(S`)WNvzsqPD_$-p?D)R1nP7%mJ*Fj@?*TyIz#}FY*E1+M3TfQCa|J1wwC1d|T
z&@cQ+3IE?{zWrfp|AUTT7eb2dyvVFg-}<Fdfhsq{f`Wo;4<0<&a9ea(oEmrgqM@p)
z%3{!fEd?(vDYE77OpQBy(HISN<3)>@g@0?zX>F`)g0@fXf3&>+?@lHB|5USy(EZWR
zrc+@fWxIUvN5#tff2<EX-cmG}9YRME6jHcUoqz@2Ff-vl+UcLZwZ}*w6Y}(X49Nw>
z*XLq-?pj#|78MoEqjTc!yEe94?BPcW39UX@wNo!>G_Ga1gyivjn|+A_QzTuAZ}}|+
zlbr4VrwXpCu@*8*?|*2|fzT39GNe2K-jw{$rnnS4XyW4}D85&tyADa%98`2#2uyf@
z%?06v5e$&M1g`hbnUC*Q45;`&-;&=l2<adrYY=&cn+InD`99Qt9)A&h)~wvy4t<AP
znLaN!z-!(JZP3IszjMO>?~>^stKG_ah<k;#GC%QcAmsn0lPUYWV~($x2B#0l>2$_s
z`}*tjXPorwsy+3+?fXBNpDC9ig|W0i7!U3~L!<HpX!AsYr=1^w<~dq%Ir)af-RH25
z!oe4;AP2G#afCeh=zsL)6A9lt{C_Y&;wrRWr$L6XI7Sm$%;#?WM^Ck_zJ)i;MkzSb
z3;hd-AiSM@L9xbG=oY7HqRXemyE^%7-Kg%=xlQh3Ki?p18yMGFek+;%jHXvV{vH=?
zH{z{@6N<IUr9(LX0KO+JrS1c%xbAZ7W!ze)5q6<?u4}0|T3{?W>_+zH`c~P8cUKSB
z&#Gl$xAI8bDzl0P`G;MaMU7^=()e80rtjXn*El@4P&~pQxwEWHyHQ1lt2-&N`FV4*
zUh<5x<hs54*dy}u__iF}`tZ|0>a+FncK02fg>|=71|Q6I;v7=NgXRv@FHpuRp-NQC
zrx04yQ}G@;!BZOP@3L+B`A<riMp`;eZEsFSVc(BJ9rVyR#Tb6JVRUJ+F@D}*PEBAe
zXpSKVB`zpva-Ki>u1y94FVyGyRB2R?(aq>)KC!2VnHYAnBF3fcBCsSV+2xgw%wxWc
z+friS!nP4I!ZC5n>e!n0TLwraTc*|#+y%LrXU`N{m|V-=Fd^ArZf)v@z!)9UH!cXH
zTw|lk+NyF*C(=vCwk2Pe>4JLr`9L+>2ENNIzt?MXYi_iPlSe`DOyI0CILGj8aU8U*
zzdZWvgHf$t_;Nl1m$@B8`?+~sS-t@g;p%2tzEEU4Yo1c~S=v{`fmLoM%X*=e@rvEG
zIG0IfX3S1}r*iO-MK7lH+Y%1%iTGTb-<05D>PL$#7@88CHr<?p;~huD5<Cp`b-!ka
z3Qt>f$y^=lWM_yf!@qwbY=F5t=<tCYRg6^#v72Wfq0q4wuIdzL#gn|TyACE8$}c4(
zToGO@w@}a2EK8rRB!)q~SR_`E#kFj)fkE%z%t%({DF+}fatz$j(UIMrFUT<J_G_))
z4n3&x+-=c$ZHH<!Px8i+9g~i=q(l&|CN>W-;_xBNb~QX0E-|g2z-N3{f#pu`+;Ex0
zHYU=k=DkakE6Qjjewfh*PL}kf#!OixH96LEF%4t8!@Jn`IWEU@>8oX;x^qEcQN7=U
z#pnbTRc-STgMg<?@$nvqGdu!@o$t=)_6p^;bRB!KEh>a@YU3o1s1GA;dG2QIBBFlA
zWhX=Ywf08!n01(8q<BgVTbNAkr#rqzwRvjGqnR3`JLL(1DP~o=O)t9_3hM(H$rvR)
ztJlY>asBmUMHyyiDiLEh#95w@&{5<k#8G+eOg!;AJbjW0lP|rp%~mLa&eRbVT(Ht3
z6?ZI-Dw-Pqh&S6Rw23<Xv|_Fh<z?!*xn7-;t-Lbz8L_>_&SyPeve}^A8ZclzS8m`e
zCoV@qr%pK9LN0*uH@pl*^4ii?GQFO`K?q7RWD57vJ<|8&bMqtgq4ck4Am*HELyoct
zXJ`Pm09!(Wr_G@4Fe6Malhz^r4Q|8sZHh!><<8Y!19G&%iyczWw8L39tI4MQK^q6H
z!?r|q`i8_ravvzeQqeRJGyJ?6pJ&*zZBtfpopRf@d3Bn(I4MgseYjW{r+W6Y5&UAD
z^Y(K6tHMHVHw8}h%xBmA&vdUYko4tNFL^BaJ5`KLJFhgb11sJ!kU!}Ie+H&k|N2PD
z*;8XOHLSBHyc8zMIgYQK&qcUCu*_ZWyKzohz7~>XTiGnvfLFlyx0Z?~I8P(jonjf2
zS#0!|r?8_h-OQXi`p{qFN7u2I`8&7-A2?(n&>!u-(t_HMkBWTc>fxa{VB~*!O}yNc
z7aO}Yq5{4I){k#?5p?e}jI~91Iff?XfOq!DlbcJ-9CcmZdC<YjGMJEGU82w(p__B~
z9f637yjS;vp(}0gZ!-2VA0{=2xCWZ|Yt$Kdxj8j^yj;fPOG~ei%X-~y>vNoa#8bIc
zy^1Q?dVeHCJy>SJxy_^KIqu|$Ic}lF!}Dg`YM74Su=OI<0sZJ|F;86Ooazgf{Q0Ww
zp$~cZAZO>T`AsY)XGcIItxO%1>ejQ*n1i2@dVKZeGiy~?uI2TF^bhpzxb)q;;tj=}
z*PM{)l5{<tycR417XVckRFKe%@Ep$M1%ma_JL%~L66SMqGmNBiSEQkWv;Dsh6XmYS
z;GO^Cko2oM3Og=~%fhz}J3PGZVBbzIa#W7;=Cy;A*DW<~a`hR>U#c@OGL)Z>={s#P
z&q}K^Ul)4K$nerV#d*E?N=}XoXF}Mn6G33x)uMABipy=B7|V(8@uRUThn%Kbnnj%^
z7Z$}IEJgFvczSxq$?;W*j?6G9DYd-u#EC>jN8jnq9?~qf9uoDCbYD4sFf%JF_ud3l
zQ(MchAryJpQ&;5Tt7ADnry-`3Y^j>vdU>KEH}yJ_F1rQq*F*8zb@kVZ-;g%cO*dS8
z0^i^t&3VB0BG2dp>!&n({^LhcecG=qUCeHM<!#uB$DOB=Y!UbAGK<!Zyj|?=ohT*1
zyTD39PDXppQbp$Q7YxjP+JFhBZ#?nl^Gc<T?dDwM>OK1Q{z6Xcz`;S?sOZ?gs9qSF
zn(lW}mQ_FMofnXh0yU$eoXnrlV}sqYg^Y~_SRbIMxpk8aGXBA@I~gH4iV0tJ@HLQv
zL^MJKg}vooO4)wQ?q$`0P+KnDJ}`1`5rf>i@OgV~bPQ_IrD^df@1W~Ri<%xxli25e
zjBG%vpx_~n&GIQpi%uASuH(C#5!|abMe6YxFI~lz|1YM#Ixfnn>sCRer3Mg?8gc+Z
zQaT5uyHmQ64k@J@28M1Jx>FiSN$EzAZjo-dkMH|^_g?;n-wfxRz4zK{t$mJrq>NJ5
z#%T$vv~SXL1}(j5o84fOOU*+eCyR|v5}cmnqYG5JBaOv6xQ4a8?O=-``v#SjeTi<X
z1NU#@-(A9({@3o`${Un)G?4q5Yc-S(I|HXsjgl3)l8mAi5aU-xGi6#`-Q7sT$yK7#
z(qU_HXB50h9Js%s5m<36lk^$Vu)Wuch=CXTZ$I^i!@n(yNpiJ+3zmlDEp7F*p`bZ}
zCH)!xMu8q`5?Us^Ch*EM9|u-+>$VvVSH~Jsqay#N?^U>!Y3@#a!>O$`%rWZp8yK0U
z_^p*`E*1{5_gJXWB3CcY<CST1-S;ySoU3ffFA4fzFMz!nF3bel)CV2P3-=a#?xTTe
zr-CO|!Bm^PWyT47a9v=S-2r<&o^7@NFt@(Ym2^psC_-7Z84B?xGAq2Gy^K_zjg)QU
zO{|t=*OSYHM&Kn!tr$3gB|q(AxEfz|VXG_-!|<}0I|Yh~(1^m|^Tn3ho!Xf`g}v#l
z4~EJ2OZQVcf{a1&s+2xN&sJ>;=G%{cITOx}NizMi%;!l|?o1U&K2}J~XU6^elFQw_
zu4jiP?Y9sC6}vEY=Ucwq`W^eDig9mG8Y!glx{ou5J^Ui3SSdd{_f?r7*IV#es=T=K
zfwcN7Q~%@D&)8aHOd|e!{m6RiqRPxhAzi1|sDJe(WaQYT3x@e}?{Z!q7G#pKZp<eB
zIQ(RKJ2MZ(82!{6wuXj<)e8tG?4PH1{S#MjMj3J6tv8e9iw%W?^2KyE$$|fteiPV4
z^l817C<4f*=D-N6&l%DXaL&!Un}56a)b~C7zq>axLpYrD*8Ee2elvdPt3l?ukIOoL
zSE8c7XYj4|3=d;6C}nl$Nk(EU3o)p|<-D!Is*EWjb}isxZCMvFLQh46m0<+AFt%vH
zeu1{5!;D90+!+(HbdxxhDx+iODHrn0zL&`yWisw(KK@R%-152H&QCdhkr=?2`TO(z
zeZYGd3~V`T@)k`2QN#)k?mszk>G{xpP)*%Os_MK=RTLdvf$~zxsxJ&#!8}P`Ug6#K
z(sq<!&Sb!F;i082?vs@5!~`WbnP-ij+S@CJ<xfXB&S;N07<UYKNW_ZEPw&Nz&%N_@
z-U@jPGGYeH4&qG|Lzb$XqbIFa+v#mP{Z7z(rG#eq7!>ASPgMnrA*M^z$nvG4O|ITM
z9C|KCHWi&2dLhSr$pzM|Px-~_@&&M{;lN5DzTy)Qv~K?l_>^DBs7Lc<CdtR~M3D99
z8ub<3FXdJ>V%|hZ;K(%BvY-#HhmtMPX|2cGud>nU-4)`J^B}gEWG^qzM=p|yZ{kjO
zR$)I+Kd7^om?)ciqD%+0%b{;>n$bW`sU0Zz!x(V>Y$Whu&X4M^b|ya*4VE7stu0mK
z$!c5%RwQ7`kI2Wt;sdT*J)91-(X#sb2Xuwn>XVV^CGHm;rnNyH%;Cjkb53?gO7Tso
zzO$HF+&1f8&K!@ad>SRAPaWyVoS2w8k9kr>Joh1}$*flPRaHhpWf~up$Hr!ElbCso
zTBm*#&%Z;fSu^HsbLKnLH&7|?l8bX}*I*YgD=eTo-Ns&?|Awlwx5t0H(9=meG*gOA
zti+~G#2j)+tYc>{(&CKnF_tA}Sb51`vq`)vrJ&rsMAr8v<D-d11o7UUv=@GCllRr%
zze98j-jX|JlGrayz-ux{^h&7#&yi7c;;YzHvkE>Q80qOaHu*i=?#@;)>l?^XF)`uW
z&Q<nQ=%q=TmuZxR5WP+7`Ig6acz6_-ni~8gmU3fcZ))EEqeyMtj1oiS&f)LhUlhp&
zz&V<4dHML(ymXl=_1hyx)4zRhH<O8DMOlECsNsD+b3;d;xvSJGVkPCa4vN@f-W;Rh
zIK@-U6bzOKh0N1dxmQsav<nIfjyKxtsaNWwv$9scd(SDkc(xN36C-7{`<A}{HA#t&
zNiEgm)U5kMRcGxf)LB7~B)=;VHQ#!Qp4a~oY5Qj;ri%+F7p6nVZ;LKR504Lqh7`bI
za@y!uuv@4lv9=xx=-ixYaXZM<Z75}tR(PU+9t1BMn~yCgU{G8?T&N>0lC^c*9KahG
z9TmI!lif8mghA<dO}#%?g-WcY)i&_?>-c#3e3KJFUf#6rQ=~v+U7e|LiB8?i^wxp`
zx?%pX*w_!++OL0R3dA1FRh6M=F{zh-N#(Q{2qtLh-Ihya3cfsA%$3#f=M~^%$0EST
zcifw#J6vwYRLZt_%B`qU!Z006K4T3{daIWwo5*x%=zQjkM|rh5(|!oGDa|{F(-TN{
zxY^Db&;9wuXt+#qwAg_9K*i9_Bt(OJ;^YJ@g5cwgWvkiy+gpugk_rW|>VV(1F|}g0
zCXss5#{IW&>1NMUVQJo^$9xv#S_bwKUSelr4+$e%E2|c^Z<CkB6uT%&pDgJ;)0408
z$KpA(t9QFk(4vcg`hXZmzksbaprG(TilJR}_7-R`6KngJ#ppJbl`-sBKY8p%&Zp%C
zTav@%7>)yn@~t@xNFme2nE_SG1+>g`^_F$@k7yT4=B`Mjo?We~;rCu~hRCCVf91nw
zI+PIGR>xL>kEVe-sr3YR4jbpWtu~{*(4m`+tMP(QzCo*DeXE@vkAD3_pZT=cex#yK
zxOkdILopJq@^W%Z3C~?c%#~&^m_nU9=zeeNVr>%@dcqPTEoJt?{8EDck~w*LUWMy!
zOJw#7TP=3fW>gF;T20a7*6P%9xKw+_uX|vj+}U!opb~osXzQlcU`3@2Sn_&ovS#%%
zRS_@^@2EA?4rw?Cw}v7gl2?co)H1rToXC)Nf9hXfO#NWfn^ZX1Spk*ZxMI#VtOiGJ
z?A%aGN6D$RqVPc@u7e(KOs{Zj)y}S@7hVx@(@xl7U?l+k#Y~F}w-kN6y|!O`A|`00
zIp}T2bs^+#hYd3(;L<XXsDsg{4;#Ud@OY0x5hT);DzRo~zsKQw)mA-r>U>uH@!lat
zFD2JOjWWqz;($d!fLH}DoVVjj(a-wBOf%N&IjnP^oYV07j=h@-8+amzRY0B($h=%-
zfkz`>xR|Zh4KtQnm>;E*?nveRJSv5uA{H^&DPeum@n!^>&X<0squg#prQJ~HT3z+f
zTaFBK{LvdfGobolqRELpVpMBp_be~!8>YAWXHS8yu7cM~*{=4o&ioWXLdIjK_3%6h
z8zX!lJeqB^O&J_#oSKjsNtbX{a&!|<@oJZyE9@7MkjLJQoYdHns$tCG@$#V{T|j|h
zH4bTKmFFz%Y1eQtYD;Us>|Gd$lT)qSpfxG+YO&dr!OwtANkU`JaQ)cjMLqTRaJYwu
zFHms{k9lEWY<P;aOn78qdwMOS{(QLU-*Sp)%mRHl#$B%}wIi;*_eIi%W-0?e8?<x<
zMRdz*yKB;8B7OvSR(^vxSXi3|S|EphuTi&dC>snddb%dZZt?G5nUX?y&Xo6U8`~JO
z^5N_sbT?L@i4;7bav79=`fjj2>BEplvah9xXmhU<MF>Vmcip^AAFn^AN|Ae!ncdfC
za@>70Fq%}EZ)a>8C?zBks&|7yI;X6v4icX&hgzh$#`QAi@a%g$KsWpfViNMe3xZMh
z;>wABANzWt`L5PmLY;eKA}5`!2Jjau*&_ZiY@PQQ=l(SlH~}5Lw>jozNu)(GFS^gI
zIpb2`j0^osOJoELK9MGkLu33l8EI)DqyCTKY_~pLl`~Mpc*APcn{cMb2sL+Acae4B
zj&yFhq}NRyP+6cTzw`Vn8kZD89~l<*eZFQR>PU0e$QT#$?bB+%$LeTz%0xc*`rT0w
zIvS4m!-9DBCx5&7??V(;b!A$WCV$$zJWs^iePE)9@e3NMsNOlF*cUMMQmMRSf{x5$
zSjOr3d2eqxfliKL`o-abTm>S5JseK2Kx+E?`*YH{m%n~2^SO|)7Yx1t7vu=DZ)Y)N
zU+zv2*e`Wdok!u0rCuIgt@@8iGbzyHLiD~p_HoJ3E2cfOWUy@5hIaei?iJZok*#Yv
zSyK8=qyK7gGiAf;Ku4b(gV@{dR8XtB2j!{jgx+(!FLj}>vBHl|;=s`5K2i>z87Gc3
zj$Ddw{B9(YEpeVdCnmk-3&blgFR{DO%ct_rVmGCIIfOE#EcR~JwD$`f@5aez^{wXi
zhoO2rNd05DI#oNd0d0iQ=%m$y6L+Bym@hV&&FSxPdXw|*SIsx-t9FmZgfF;Qegwoq
z1I#Rn{k8B2AbS3Mp<M}jUCYa$VZ+2L{Gre2x150t4J_*5;DG9SDaJt2bQyC_!j~{{
zmlP3)_oNLclE3JM<JICZMlfwvmxKfrWPc=aTc)~4vwD+-gBG8VJMEwAdCT=m3HRO7
zO?J+0;zHIr?BF-9d+J@YEv)yw3>Q!OZZC0;>7|eNK4(!yqu2H<eOk|B#FUUn#5H;3
z&k~Z5<l2DVF=SFJm@0cqT(+fD-+LLN2_LFNCc3I}GYwwc)h5``mN$)TY~0DrSFgjR
zt^wHCCB0G}Wyg}$rc{x6t)tMS%3~ZlX7lyMm_~H%O0iu!91r%ZQBOfxhed&i;0B&Z
zrPdJgmN`#MP7)%mAf<hHyHQ&%fz&b^-*{!Mgq6CCG0U?Z^-QFe2EO+cmm^3bWJ=@E
z$LuC{CFq(4MpkRvmTg+fp9||hbOVp>rZSj;Fb-0=w38GCzArM~T!?YdAaO7J4u4BB
zSTI=SY5*MG>ErXgskH+=JQbB#cIQoLiQn6E&M(Y>v;oq2*Pa*+<cz7e0OoutD~}k5
zoY(n3e$60eQ*L!3p66?=j^t3yS{o^h9vB^z>KXgoz4(*j>StEh!YElHzuU#Bcw|@4
z->gCJfuAGdyD;QKr~r(q;Nku^*>uoDW<p|c=9vYJkj}#g$0@eke0VnH!TWsx&k4Ci
zWX5=0G9o`ERT$eiyyBGWi&a&CVwhub9ZsY_ppAJZ1uXp#k9%|Xn$6|x_wPS3veAJm
zBNMR9TWp;=40ezDZ28+N0!aVgvRr7y{=RR1LHmlg66E{T5#aD$MH!X$g2BELlk9n>
zUg3X>p%ByUCYWz&>Sk*qu`@zBtvp<e8K>Dd_qL*JxV7QU5FC1EMPbk$FhDj3%gR7I
zSbxQ6(2UV$Gh=b~*;xb$HJMFCV9UYY=nZ95`u9t9O<huPiB~jh6B3(dg*|yGZ{Pfq
z%eRY8%dvzS<3qUohp@Dt4n8re<h{7PJix?sL3#f9p&#3Bu?gKZAS<Ago}NA)JVGwu
z4h4`^wXKP%X|?^52uonFVgxEGYQACn=fieX9Ey*J^Cv7DigvWBO;&3-R7dFinVsHa
z6#BgWUyX@yalgDeT3VBI=NXowN&u_w(oRQ_aEb<Dkz_I!VFd;;DE;0%9J#e#Zo)f5
zN69IYPX&9)DhaarhJhG9Ft8ZbhGh=O%E)wG9tq%5P{f>{?_l9n)f%GNjXx_aqz{%t
z6#vnZko86v^B%fvayp}5tali#wfUJP6w08qZftHH9MM6j^Rf#us0I%M5fG@^EjK-v
zk&)r`xMcYg6kK;mcv==2iD@%e8XO{qz2K^p#TzaeNnGZu91#uzcgeOS_r{P5;1UtV
ztoKFIG+Dp9FJlU;8%}03oh@I8r)-2Ps5O>bKHj!C?{Q=m2nHr5a_g-&*vyC@=|10m
zxFP0mU_XC*w53-AE>ukK4=ad@!n%LBT4fO!oE`x+AR)bmh;5RY4@TYJ+A24u!1oRa
zoV)}juI`DAd}4U)L0!-H557MlhlPg+H#873X`V~AHrnfGg3J*<oHn5Ngz3&t;|<-5
zf}X#r0Ngs@10p4jVAiUne#?z=*>|LpFNN)Scm2Cgi7=G6*Mg-}@%mCvz043>DO*V5
z$?t*AY3{<LRmrCIci(8F%Ap7c|ENJ@0OmS{NY8}9sgf~S9!z`z`zPI@<HJ48ay#LV
zS+dMW;-Juy^3m8S9O<m2Pb(AI=8ADDjRT3CII#OoK9=$k<)JB<xrg%0=b3FrcReZ%
zYTup1;F^>gC@_$jY70dv?|}qDnO-)#D!rPDTIf?{rf-nIoM~KU;xLlYMY_fS&Ncx|
zA}FR5=&jL|n2uCyN?R52g9GxWlY~8Xd(VSS@jUVGu+H&z*Bv7HqlW{7f~{lp{DnjB
zI2vfm%SUzM(tVuvdXe8O`bju@0h@^7+_<Mou5keGidki2CkN%0Co3<0AcKk1rl;tW
z@s+*3^AO`u{TyqBq~#&t-U;fBX(Eq3uV*33p~JJw9&%ve3{rHL+dByRG>IDxI%z=t
zX7C|WQ6u&G`iMF9OzEZ-UT<kW^^?5)N4-z4R138$rH8}@9wF=tB4{USyteL}cu8#x
zn$>P!80S#!jkO^fLE5avCmNVI$+OJP;^D%I==*U=L`qs}&JV3<kl$FoT`X?llCk{3
zWEDjlQ&TOgmcPHYX@6L96?4nEiVtPG&n@-9RX}(+NY}gmTV+=6juh7!<%QAvCc$CB
zJgl$T+Z0`A>K80%5?5_0(z#>UOjT00@3z;}=I3>CZ0wahleFIOgf$VwR)}P(1=PCg
zd~l`AyEY{J48ee6Sl23ra+w_@Mj>5-nuya=r1@%@LD9NeNSTr7t$2|sp@S>}g-r?A
z-bBK3^2xNSR^-0Zt*p$gN$Gd^b7Ml#-b@D1taubTTh69rr;pjFL44+(#k$J2Muqlg
zyAEJ*_c{zWAz<YE&E<XJSo0Suws`^Vx(<H>SNW<#;bvQv3>`YFhjACD0r+fnBhQb-
zDI6!!`obW;&W!2!M0ozr6D%{r?zYrzIA`?7%vtE8{T_O+psE>-|G5|WB?}16;qz#U
zU%_J-Z{dR|iCA|=o3C7sJHe+$rT0}TW6JS?YR4<>3m!VP!oIirBeF?T44-=N#zUSl
zsIL6F=n4m=c5FZ+xRg!6TbcvL`YR=j1}ehU&-dLQDevx<XGd=e+DaSzvhhd<WAm$b
zLcL@&8!7maerB2LKys1gh`{W!&35vsr85)7$$7Ms4|2p#p97tT3evyfq~Sb9v_I_*
znP*(vXPy!>&Eb3c_ZHu?5J13v8AFv(*)RAjlr|`JJA&{eO!B`w4O2_#;PUhFlX?@4
zOteHqq)F%EYf`?t*4!S_ba3Z0V*poX&mHJ05CdR#|H?=bQT*&;aGCNPtXiC2-wQo7
z;L3&t2k0y=<~{75<Q%dLF)p+!du6K#EC$vS4=R=$g?`XtD{CIB`>gQdVz<vwr18=c
z`m3pMy3psN#Prc}aJ#$J7Du#>iTY&pkPpdHuh`5&oyEqad7ls#84ngV858ss1o?}s
z3rr!EAH^?`*>tDU;M(n1)y-5cOI)Hz<;PwzeSN|H7_!hE>miZ&J~1tbLO82|uv9WD
z28(8uPESCsNj_F=RmUyI6CDIxaP}i};=(_|a=SYi(`g_6**fbbHWItfwd1VC%JL1b
zhv5xsF;o97E(D%9RrSeecBbBXipzpX`EM3lc-}ixTnO!Vg{?vN<AunJ&WC<ksfg%5
zJCd42`1nTIn?+wtdqTu`olX(D-fN3ufiF2~qdRDn8>bQ&l;F371Q7KV+wN}(@O{=P
z%iYazf;FwaSQxGE?WLNYKivMJ2PYyPuJ_Fj#avHDN9O(eUJ)5(np^cH0Beq9@DUI)
z!7a^))Q<!Esk&ht4g)dSykF@elcj&EQK@3=6PO;|3<1ltKMk>SxVt?5F(&d*E(J<v
z*^4>u%tisrTpg|i8qHQFwh<fK7Zw%uOim(260;?;lu5dDx<9y$NAhz5L)*93*fjhN
zFek|xtzdR6YSm(#nW@=xQWOMIIx`339+W$A4_`0US#kGL_IFZ1ROn+NhsThIf>?W2
z4{6Q!Z|cvC<PmmHM<&8L5?zHReJm=?NvKwDY6rZ-S+pgdzq2pIfr*=|6boXIz_7Tq
zYE)9IfIqAwn1m>C*4QMqgf)U-;$!ufzhtI$--nC)UQsT+QJ)NvE{PV&(quM}pvw~`
znZ~6hRVtVZ%ZN7>=Og*-P?_c%l-LC0&;SP`nP=Wj_hs6$Q$3-Xw`>07kD4wb83#)D
zCEJVfEqutj=*3>3F>j3#zgo0ULVQN<Sx#V07ziWc#|M#~MmOSlbcY%k4!MEdj4L)H
zLHH20HQ4(eNl=vT%Rs&P7j)BhfLAaVSN@WrILwyqq-@G3Voa{Q9gQ1TaF^Gvy1ej`
z8@c`WZ^XdiIHC@0IcjS=<qb7$YGJNb;I`(R@P-%6+BShPMH_lEJ$)zdR8yrv%Y{_$
zNfHzI&tsjpi`fByElqvsZ{#j({eb7S(BSVoo|gO2`H!wI>@@sV&m(YIKDzDsX!Sze
z-I_I!`7uMM^>>Ocx&S*&{4Quf{If+_64-3(n@<MG0yPgf|7DbY5}Qq@^$bbUFN6)0
zf40~t$=s`0m*kUW;V~4rt!Bbf^%$n`y31`toc8kGYi4{4aq@Z1C79!bHc6FTU+z1m
z6X}9EtADgT!d(cj3Xu;M^eFl2nDZrz`rM0&p4mEjQkQTwmi<B3>e=(0@&js#kZNj+
zk&JL#twjGjqV4&%<0pEg3~4WqhoAp{+B@%Gb$SW_qaADJBBkLd`{0smcf3$CM0ZyS
zFZ%9~%~1jGEMz)I8aR||%6+hsOic-Mv2@xV6QM=5W4~sG|GhL`c&o>t^l<?aFmp6X
zxOIS1%Vx3EnI5w<C}ZP0{pT8QN{JdkflS=i5no=BFWd=OjEUL=oQE%RJZel7dr;0g
z>}2RjhtKKP-F3<%UiVZ5u`{R|vYF@B6Sq5CY`lc(E8`TZ>F{TNzJMhsL@I22fS`PZ
zkQGcJtmScnT|Y|WNF}sCU0Ct9B*)Xz_Z@s5?sdw+O3mYPQr6HP3K>8H(}l$yJ?<6V
zbVZ2;qSj75!S}>Sk$`d<SRj|YHeX|E^G&R4T7y}umSErG&vI99cmjNp#jp(*<Vgp%
zn6J)_e-2vP<TEIiMJV%N;BkS_3Qk24H@Sn`BbmrmwdQX7+SN{l@*&yT*_#>M7fM06
zcoJsW{*S@BB=vdcHSdzx!$AAY=d*9+GdX68kt#MZPtg}@&4(S&i%)NXev_K|Lygak
zGdEfvbZK)?ZQ9i*%q*3|i=)+HrOk^)Z4h@VR2mQUA@&Vl03ov<JOc?21XWjlsktib
zOIK(5--diPi;~lVYxp;{Jqehe+xitc13`ni7n*ItA4#%7I+UcEd)cufw*wGVNm?2$
ziOtA(;Pl;4LkKPe<d~<nP)dttovBQOyb%82;dq<l9ImwnViAxOxeAjC$X`I!Na;tJ
z={I^TJq&Z4G672x;JT7Nte>Fii%^zuYAgEgN+8e*6d0T2uD}Hl^664=*(W8b+h?HH
z#OX{h@h%B`?7}apmClCKN=N-GXbx2ae+@Zz!us)*)}N6io=&*wW+R0l_ka7yQ2WU!
zygvpznX8fy$RVx<A@X+UV8_Ns9|{B1Y0ILVJuWUCpo=l76)h*sISyeb0w9&H$S7YP
zjB(t<Pok*trMZ+rCvJE<e?aRe0Pht<zErU5ljRLwjp|uF;PSV?fJX3SPz2Fc3n`I+
zf{h>gd4?Hmx1$#z&8yb>liRzTOeqRt4WL8-Mad^nb_(=L(`pa|tiuz(*)2XVIF{(2
zKQjRb%f7?ME3)x>UNBLaF8<wTBSI~#?dkTaC8c*tN_(<_yahE!HCnZtl8J!mL7C4*
zT^^yFu27~}2BTZNmsyC0Z%Mm<T#80YW`nnChVka@F#5Vzq*WM7iaLn2EN+P?^<;lM
z+n+NZ>0O{Z@Z$}+er>K$aP_1G+_x~yC}iu6W#5bDIy90$rZ!~qt6gU8fUSs$LqF!Y
zJ(@F=aKTa5PY4u4aQDY5gVkJ2l_!(FbH7DNVx?%raOf|c-hVX8&ega-*9~u;Xa1AI
z7@(i*kpZpK75l%k4&;NsVc?J4h;tu|4prPFyScXGv%W&1((C+_9bcU|myMLFGl7B}
zHWdHCn1|Ys#I1M^_6f1AkaZ(Gu4zrQMv>_7{y#U9I$+v?Uv}R=&NhHxlRm$33rS#a
z2%?2ZZ*y#A!XEQwrnE&`f(9egq`i)X2;ddx{J%7}@hwO3O=)0hED98a2+G_F_Isww
ziVl;rQWt-qiB_jKm6jT6EJ=uim9p~aHOp{LAO3w#W_<uhFF#NE2!-NP%eea-DU~CV
z>C5h%wBS{gVbi@Y4pRzPG>;5gS@%5-ezML$cX#V7HVtWBm0s>IkRe7zr63ESXLP=S
z!oDCg2AwfOp;Xb!2HUxJru`%La}c#4q=oY+hV?IsNph*UKBI$!VzCtM^CdnUV3A!e
z-h~w*@4shklMHh*onKtB2GqkrC+J=VUr!$&?(!ugx5J{Nd&YAm8oCQ`V3YOMYFHG4
zVHa}GbudOctT}<>Yep3QlgtKV6htr^otQ{BTlo<D&~DuG66hF3WPe45!vUMg+NR9K
z=T+eq;S}yW0ryO}a#PCnIJ6(MnRQOBn#PUU|NSfS!aJAkk*;9O$rjwBrEd~rSwcDj
z5Y;e4Vgd-tNjOl>wB=o|{s_ryt0!>V&Bx*8n+3jnl{EbUf0o|2Xm8G6^y`r5>)|T6
zLWAQPj$M|3$Di)1&W@i=%mXd&Ec7J8;ANH+CQn9_*>XLWZDX>t#YqjhUm!D1MG0M?
z`P?WX1<htDYbnM!7!PW=GbVkgbZ#S*DCiK&ydcw62hhBStajf^QOB`N*CdnnRMZO%
z%K|ka2@nr%`+lf=v-Z+SW$f05s(RiwRp*eR%yE$<Qqoi))f$ko(VWMFG1U{ocO(n!
z)+x&TL-8s#=u~&zsEd=&H~BDG68^qN|B^sR4+Fm%vcbST(8XN8ik3v^^Ey$gFODdy
zz4g7^OBQ2M{`{mU*jNlVQYAmIr%`3BmJ5O`tRApZ-4}{e=?wvloGUqAqWYzDjD@_I
zFT0hvT;Z1Zi=DO_0d7OZRCZ`f9Nvv6ZO}&`{Vm-hIMkvS)lHI1@U#D$5u{bYavBOO
zA3Gkt$u+qgxux&B2~F&q2EW2G_r*)i8+@_m`q_VKh||+m>E^1;b1Z&b5dA0ck_6YA
z#LGN+qM-~t4qlAO(85M|szi~Mq{07s0lZ_%WtN7bnhND%l>Ns%Qdc&<VKulSoF%!F
z)WN>roNnWsA)>~kf=YvC3kK(20(Me*?$XC|l0Q!R-Ye+HbY<^Qn$uo|rP?oCo2-%o
z3NO-lnPN%XDWIawcQ)EBM2K*keIed{<GK3u_60BO^v`WN0ffW%M>xnjrx%K`7Ytq1
zMEoZ5aFJ+LiNH!AUmb()mpL;OFlk9m2_w%RkV!><aXc)PYQ;%uM85l-$(@<V63^kW
zI)wJ8sOfQeTcOGO((NYQp>C(AcLcsCl%dT={)XBhR578j_MjjP9$PdP0MHyG8o5^P
z)_1G&A0!h9xnZg~gT&6Jrq_vu1DvXc0<r_rgQJ68g*B_vpOy994~|pS63>Ht$hgHS
za0s9KHcO>RrZYsT*f^M5MqcLk*Kv3vro_EPZZF@`3i^BxQ^Q~(Dhv-S&Gwz8G1_g_
z0Bjjc-*B?%8S<Z=N&ZegtQ!;VL#6cjevVlvuOf&%&~f)i0M2$C7}Wa$ivlr_mr(oX
z7w~w{51a)(iO^S;x=dyWGIrm7*UKv<TR)S607SWpGHwTHDOEKP{9o@OXod)ti88HE
z70Ezdj7z!cru2K3C^$Uq9nPH0R#$6nytWqR14>n*A~dTEY2+q?<$Vgl4sUpm>Cpdz
z(ZE1o$08ms^RSCg04X)O6sKIb1709X-mnUD_Ih*MrI1RI$OH+X#KXhmb=iLNXSq2V
z^@?&cA<WrnHdLBA_Z&w=J+98k_kwV^Bc5XtnMB0!WTDe$i|LIZ!jN8E9s%lQ0FLZB
z7P#3{rFFuzf4%GB&L$ZE3g-z@v6_BT`}?cfeiuwyuUtB`tPYeH-Va52O#w0@SR-Dl
ziIQ5mQWr^*#YZpB+vwY0Ugk*J6#8h;X`%u<Cj8Xd?3Nbvu`p-na<$+>rq_j$5e9}=
z<q2OpbPRDAA)+`0q<o6dw1W*}`YT1?oC3W;D2dy7zv)O4!Frw7ubKM_MuVb)0)ES^
z_o;5#2MIGeN;B-2WMGE_kK@+!01&c`IazdPjBt_okdO1u(QhH68=xcpWMkPqn<Sx%
zjQ-qTZ=FX;OU=qJSwtVP?@Jmfz0mka@XyM024NMt5k~oJ@$ujQ2N=P^GMf4-sI&wy
zm)ru_D%V-mHB+Ak&+g=tll>eFjYo)pQO~lk{y}&6*m_br_hXH{lWpPHbdEU+sII|`
zpR1EiC|9a5?q`UcjGLHS37>_9Ss-+}&*Cr&p5OXuO+PF^U!B|OWvY!u;(r<m*bsj1
zO_&Vj%%oFrA}kMQjn_&XbAC~e-QSv^eXb$Ix+0;GV2@#fzoxO~W5EVOo{1*BUQ{|+
zIi^ogR_b$_vy5{KYmPA=*ORgst8Aj!-TDE&qkEm&5I<gjK8qC7lVr9RX2-A5OU>9^
zv9Ji~UP_Uy{x})hkDtEwC0uc+hJ+DvfGfkkgUeKhk?mWVhERxTr@|byez7?_j(@|p
zwr4dcWt3}m`Js5<V%Qwj*70z9z>AayxS@txTY3Ib@E-p|!N1XI=P>d#(4Umi2g}I^
z-Lro-^BbgI+Z2edHJ|jP8$~@C_?eYuC8Vwr_;4C?)a0fPLy~KuGq4;r_g@uqA>C=T
zi_fTT$^>Ac2)uPvCx=!FV(F#7&`yvWieqDUYecV#by4<=>0blD_EmcD=Sc51x+&dU
zW{UKU<u+FVhsVrfeXxI|rgM_JADIFI+x!!X`@t(R)ZqF2kJG{@6B0q0f<MDXyWbV$
z>=$L%sEV73m$2faQSvIobj0X!sY?r*iQej>d3bmT`dp&{qS_>{1_oH|U8>#k{K*YJ
zF}Lk(Z<7;-2#aq0Yry#fKcvY49%S>KQB0KH1FUlV9*@PM@60WCHpf0p1q0l~v=-qL
z(`yd1clsFy@RmaMp5qYH)fL}(eq5B6uSi@xfJQ_j2X9u#%^wRI6FJUgldU{htk8Sv
zE>sxG76Py&(h<l}i<=#m2f-ZZn?dObc@<n$RW-Q<br<q`6TI<5i387@T1|FI|AJaI
zNS<;3>WuBSc&`a<0}mpzNxN`+(OJV+k|v4(P5+B)jfV$zr%9+mrMkBf0Xlv8C_pT%
z-yJs`6v$xPV09)z)Lx$LFxOW%+TW4+-%}fqHhh0?{~ZPB0|FeIrvFxjgSO(?&4*jR
zNXG@F+2kn2%nN2V9dWmzV+yNzy&1B}P3TQ4&jAEDQ>i)YTRhHHOC4U2*~)FCbE*>1
z`MTs=qC9+fF_Mr1DmQW#&ky4jy6ecQz%^wHv!>}PrgXy9!WEdE?F6WW=MEUlDhm+&
zx<Z190HAt{o6Za9=07kzmuC~T1AUZ_Qj*-=`oZhZB3_G#|4%GWB+k$Tv!QUyY<!uG
zVND=_0tpZAz;`zdxxGq30P54ut0u?oqG0pJz~syo8|BeE+NxV|UGAu1QEUtMo(WlM
zt{wOJ2NDw#&PJ)B$G3Bi7t1t*C2`~4bB|vS+lA6A*ZU<=RBX$I*yvSzzq+270qViT
z8})3w{%L9I2Qqv)wlF*ug)zAr9%-@4qQ?54kh)+r5<Z?!BdpxrakXz{I;CPB8ZHI@
z1eX$&8Qji>YDp5K>Clz>PTUr_$&b};ab7aR*w@=-dmFx0G3&aP4h`^wkN7Vt15f=N
z=9^Br<Q3${3yg$j+W~xFh_^=sBkZk8n~i(Ja&En`rb3EI#K?vIv~u~LrwwJb=)4)n
zZ?rY=ACF0HRPq!4f6F$8B0^o<QKS#zBoPjm6L{)6ilZrK%g2S#(_fp#toI~v7%*=1
zz1Extj2V|qJ5|@Us><H@7tcxthfD@hjwF^1DZ!G5TDfe1VlFk40v&4HONcv>chSM-
z>v&&R549En?orn#VKv<Pc!2Qf{(E_n-*i64!)HKP9L*k3%9SGw9Rk#8ZOvz?HQ{i;
z#02J(Y5wW*^$Q;NY_fDNYk7HjN(@v~RDr0iNKMydsloRBY=vIY@ic~OCeSF`^Ao4h
z)3B<!?I^x({Pcvnw&nqgk56OUb+EAs5qZ4MbUT?dTB>&1kSh2SC5ccjat)YR?PRi<
z!w@YPN?hC1(cDuv_0;Aql+Q#0x7F(ac2(E-SjF@I7b7NM^ON;HUMKOiYj~?@aDx{>
z)UoI{N&hjIJpz(XT>_`(eew88?&lTtxjTiWbXrwL&Oi>Xb9O!`#swMfX7Dy8qs`92
z<S-)!5WbcmQ38N8&t-v=cNzp`2D&|#m|-wQ^fmO&O{U#{b{DwBM=^uEv&Fh|m6m`r
z9SnqJH0TZmqnpa$LCz^Mbs4}1CxNufooBZKw#EbyE*>X2ApW|4RJ{-!;_GQZw5!si
z?r|xLvTI>BOFg`fl!oQY(Oi<7)B9eU<EC?l%VqO{ehz6C2oEa{E$<2=sAZGgdN)u9
z=d1PaTu$x5s=BV{A4Xpq9wvpe!0}OuH>iPvcS-I&BNfa<r<3EgH|?7<g5QNjr^@*E
zP@^l+?Zd6I5<!97h^mEnvTy`d1CtCbJp<5aL^bSS)2UjGPAt-3wu6EV$hax{@v|7F
z-S_H|2c*3#(q*f{hjmn%rJ2Zda5>>X0}UGbo?mOPPIS0Dffi&CYeAyGw@@|c!PFqN
zRX4C5`>UX^-JB67es=Bb%$$-oIBbJQV8_M4R3u~0Jy@KfpU7{n+~A&iKi2WNYir{A
zL6*1WfQW;`@ShGagGT22Aj@xou+&tiXKi57sQgyE-Sy+<nChS3WNr>9S9=1A2x(*u
zNAQ-LH7k%`mC%dp(%-LT71wLcEu4L{v9M6Ku43i1=vf|$ux^jrj$|({4hI2#Kp$RF
zYPhCy>zi5vyg6(F4x(6VtWH&b`%V(%(D*;&g)s3&0YulCARo!oh>t;w3`U@Tpp@)a
zPSxFydvZCd`SfU>8jH5a-CoiW5lp)r6N@Xe3nYfm8^rkEZdb`t9qF|_3y(!8{n!&|
z`WG*aeg*9z)I6FhF6l$UqZzs5u@m3wwE)d;qRfMB_D4#MsgsC!dr+J(FNrv!IA&*C
z6G<$ZbS>}|x#;?=*~X99AHP~Bw`k5&s|sybK(Jb$K8~bBi5GpnG6`pc1Sf(epvC=}
zvIV(d1w`?>sjVIs(A$db{LB=4q`m`W$&)dWrH1Y*9AJU0s}h;D*2k&c)~4UfAUwUi
zjDKbvlL#Z|#M8hbDL`yWJjky0@Z5J$K=(<+cV7nt?l(v(Ai|VJ_FYc3Q#$p*1$HRJ
z_f^?oWGOB3C&1Kxn|S#-%D`V=(>V|bkt;^M0O>ER5Bq@}p1U5{;nN~5%m%VrXe+;3
z2}3*m*zdh_Kkza5a1=jJr$PAukuDwP!h@t$GGl-x-RzicUEo*D;op-8Pf>ar5;gHF
z{nelmIm{<uX4HSa_j=GCRe|z;QFR!|pkr>w7XyToHIA>zvrUCf-XdZ^pDc}Ta}@b~
zn^4=sznnLiG4WKtrGMg5IP*^xiFOiqu8>S`5?rHdCtNd=@Rk~ZEcsdapWwCt&<^LO
zQ8oY0e!2bIlPyHjVWVq>(4+-8y6}Ij71$g=roEpu<QRZL{KF-exy>WS-QEh^<V)oQ
z+yk%`OGnT3E#4RCiy>|%o$ImB=H!CUZ6~`*=jsKSRT_P6p9?CDiUUOO_|6;aU$p%N
zxnjc+VNue({gg(@Mp~z3F{;)97PvzBL`sXy>R*!ud;=F}O4I~8`M)+YA-}}#X4^&!
z`!7~|^MA40kTx_zsa<sXU@{RqPs6Um`W@0Dvnz&r-}aK95b3Kw1s~()?|)bQ-ah*0
z(zQ$HQE?YCv>y%|VpQRyt5w*LqQ25V2dBzaeIH$+2lu`9qkE={^F;cS!W@^t9rc4|
zMO^RZ11x_{yA&=DHph?P<6nC$h~#Y->%1-MKee4(a51*A`045{{pmw8Ng7PGyGT$M
z>oy+SW$m~V*FEp!dSuHsL)-sw&Q>w{3k~GCAEuE@o$umpELi~VPfofMnmm6$&u)sV
z%={(u@_6B_VO|ggFo4`x@<{sjZ9kM=8xL&`mgC@;75POOt9OSf{w7!uLxi%Ataf71
zE2Q<d#1quN?@s~T-6$wevH#I%2pb}|(mKSvfuC}nd@5<_xQw7dx;f(po0NBAV)}B4
z8XOi<5ry7`&5wV3aH{L-`nHCX*=>`=*(4J7BoJJmR<Mv?GM*3CM~r${#byb8u6743
zqpMJ<h~{RJ*-8V{+ouJWop=5#KFxA6wf_t*-J&Qb2W?(1YkPa)cUQ-zBd_=3V_*KV
zdd|jb1C)>Ap27xW28W(cy|oFow<-Dnxrk6M0o8Jb9<N`oERlv1+(w1_M64!djPt#n
z#Mu+zPidH<=TYK+D-^<Ga^xC58abS=tU0tplU}XL4u^@RQiH<OxxYTk`+Ns_mr$mZ
zyZ2J$w(XS<Q!Dj_4*eAO;P8z8D}~+scj2J5`evReH-5S4FG#_@9o87OWG>L<@E@Aw
zNJoziLVq8hKxxlyx8FeG!kSZpAXZWsd&J7)21~-Zr*1ukonssnlp^x|A?Ayms?zTZ
z@0jDG{-pnTQqTsB@Z)P!&v>|wWERs~<XK_j41e}Eg%-AFEKjJGJs0jw{B4+HKY$Nu
z7Zs#*#Gzbhg}_GN=gZUTzcW4bk_ZQ-4S#VO)gckW_hJZu?(2x@P=Fj6Xx78g7&U;x
zvbS8s8a6+xWoJk=z%f-<V-~?bK{KK?@R;X6M)^%|woYZPg{X72P|mGzLcD<d>2@na
zw?A^6r&>}H4%z8F&q`8UD1ztGmq$;9cI#Zq?@wWsXB1o;FbJ|uW|{w9YlkLPI3FF-
z94?h+4Dx=|=S-sOz2~>C+7#^di*8FEsJ)Ar<agcuLLnq<DxunJ$#gdoO{1IwP^ks)
zCf+W0@WH?YM*nJ$ISaUNWfc3!D12&Fz@NzSz0BOT1D_eA{;*dd*Xu^bM<PC*DkasK
zn_W|Ns~og2hbM=Yc-EkIce<JUJ>PQILyO?^ZANw5#S@Ll-cQpZMkl|4S3ZhHw(_RS
zO?s%5|EAmhC?}CtfD2nAlvY38argga{$-lLcAHyoM2HYk=wIcWu2Zm!5cpCnzryLI
zL<F03nC1lmWX@*v0dI$9CkG=Cszowt#(;Y$wo<=8eAP8smIEl=<rM%y!w$QKH}M`C
zzAI7=T2qi$m~fjF+$Nah04vZKY;PO|HXbdl-<<6nx{9Fp+j5xVK`Qhb#Irkm=idPl
z*E5m7-y0ttjjj23O+!FZgs()vz+}t9qWzANf`Z~7EBbe>ht~fM+E+!=4;nOm-OF#7
zwH^mvtmNlWm`)c@)Yh}QuJJi<qA62}utV?wvT}AtfA-FGA1GP=rq78cJE_RA3^Z{&
zWiRM{d;eMX8c8o24~Vn91*Ao1nMJhxT!~OHqqIs$KrI>Lu$Y~s7-HOWs!+k)<opW!
zhaH>ppZ*E1B=e=+MAKw>QgKEzrhxKZ?oFYlrl#5)onM64T;5-wY$yYMn$6A4kMq@Q
zo~O{KFWP|ofO5hN(}%imZEfdSR;>WGN&`F1KUwTjt$FmSRzE4PN``}SOQHx-3f1BZ
zhBvHQMVg_qfIrY`vXbLqe|+`5;~u=_687#VYW55g3URn7l@OtUlnzZGeWTg>tU6MN
z;!LMciC^M%P@DQktqfS6uB9Q%z4yV8>EUut#;ADy4CbZS#ued}m2@*#OIAP|G$QU*
z56mF9$x%H^Qw;cm-PUJITKP2_9zC50Q@;4Aq6l%jX|-~r@=Pu@4BjV4tP}Rr;c`-O
z#5vg%7GMh39i+gRRU6Ne*9eued5D+hHm{ER@5~Rx^?k(R3v#qu@+O;EYJ{N1)4fV&
z0+6L#S=*Q)93uWYjv(}!Ds(i7<GK`b7}DGMM=B?W#i~*Z6Pp0B)y6Nt^a=+@ZH?WK
zADy!rpo1iRA3A=;gwh@S*n;CQz~zUyeb=A${v{0jbA0|kT#fqNeKDc>t5L(fNvp{j
z;4tDPI|^1Gy{ycbz0XM2MOI&rs8tQ))#h@F-E_&21>rYCc2x#vkm6gn=XVB!l5)7E
z%u!{C<r8KcHlu&P@;kJj8H3LPLuI-`N{xRYWRe#y`8(B8j?4xbLO-OV44Ifqy*~KA
zXk%*qpc4Z>Dz-5hk#Jw>zA+8XzF^G?{M#Dd(R#W6;MI{-rQhZy5F;vv_ZuoJhUe|Q
z=mMCU*?@|R4F{C#eS-o19MSJU5sU+s_+Wgm`;7hKLOBdYq|F(f!C&o_b~F4!ysRq3
z)+ZZj$3Dcr=UjaN0SMl#{=9zp%*Tm|k8eV?SR|aITq};&>3*%LeargEf~3X9R)4R`
z&W1s+rhL<B7IK?djgN_-)OnyW`4gt--*H6|Abf;AM#@c6?kwvKQ;co>L?<;8_qtV>
z;cfKR9KF%GIVla><W!CQtyw0CTnIh{Yi#zor_<avgm~|H=IUt??MrBj=UheXj*LuL
z2YG3|9b{uuNB<3n8p8j6N@$`Tmbp|5jIdaVXh4gql38>zjCf?Vh=OYZO5Gb9wQEj4
z;HKhHQpU2I{wP7WCYTTp!dxhwDRcV!#ra*i=?}%-mG<=ht8^~=0O~*^0+LcFhQW2`
z(*q8r|8nYu)8BFGa-DYF__*AJ^RBK@?Z3&jIQ2S3isUDQ4B6H6ApIh$R@VHegyJ6)
zq%tg%3M8Bs_3MUO72REdq4n-M%AK-qmq*tp#fJI(`1k~Wx8m5ivkDVhG*@$k@T6@0
zZ&G2Fh)53Ob!Y>n`o;bXGQg_(Jp)F^UodMj5@7XHsfHmqIXM$RrA6NEj;_7&l9BWJ
z|6>QVy$BbEVz}|zUV2ceM!h&<kWUHnzJkxGef&rz6{YMmn<?mp5lJLt8?t)<I6l0b
zGzow8M5V5l9C<w#ofLKUU@Kv#8N>zmrT7t=ia2#cm(ek&8fGvr9^Q)~!X)7`#9w<y
z_Gqi&6A;r6b0)F{PV+8Joq30ZsfAyNGE)lVam56+wWOY9t`v}7oBeksY!^E?_PH*N
zM-c%*AhcudHumC6UtK_v4xJY6X~2U_SLNJ@`V+v!zp~e;cp)B5)9XeOqt5N_n@{We
zKZ>Bt?)IL+$cI)5f*ymum}pwOqDk28t$V{P50)H=%jK>$*?vwpid52!rz2{z0oqOC
zU}B&C<oa%kU>vF#UPV*!deW0>>Dv}*r8O>UGU+fbSaN)W%IIEDslA@q9O`>(jyqgV
z?VSdXnW<}da>xxN>?!<ayndl-a_H=1Z$&m#Px`KrZ^|#%*)3gjP+*WXRQlvx1%8^}
zAye+%Hxekt_I|p?(*EB#>&N-l8>Y%^E&BZ4-n|7H53y8c69NLrd5gMKOl&rp@V-H)
zfz&1&?0f&p8L*A|zmDe5qC52H`ipAhg-<vGv%1<t-WwzE+Z9P1NXBMxa9+zI#?Km_
znzrrO#(C2<vtT}cQ-Y8l`5DV~%cohRpEd2&JfAhkTk<TweiZTZ2s8J7JuI8#BpO>4
zpzQ%ya5h;bk2XkHQ9Fi36CP~*^BO*W_MNuh52Uvz?ByqE|6u=uz?NXb*1rrMoj><F
zSGX3k^Z`16TS9ZmEF|HL{i*jA9#BWvIpR3Lk&}P`|B9H{Qoqd;Z`EI<h%t2A^x*@L
zSB6>-fPun+bMk*mEC)SoegCLO_SFP&{NV`8vS1sflgDv@Vq<cpK1yO5m%0B=$3TDO
zO&u($*l?~{<ye?;!zfYVa-xB`$zz)fQ(3_KbDmAvm9dG#OBXrj%MO5btMb+?iwP0>
zT<5j9ZANoHEdyO`4lb|XiTdcor1p}+?~!rsh=$op>7kzT?(JcQfp*;e=}-T^n_YqT
zEMKDFmUb~unw%j<<Z{jVqhS*m*=+;;GY_q>H-sp|CViq?J+etv-(_3)k}l)65{4Ag
zb1m+0UT?zPT<_+M-G2=WN|uBJ?rm2KYd~r=33&I%fma-d651lF*fwHn7Cav%hQJk@
zJRLT6JZ0=@QH+X!UY8Jf5|IrDus13yI-Wsr#z>!6*VlBVMw8(>tz}FK{`2Xz9n4C|
zYri)6vn=j>Zfa&}_PsRa?J~2Lc;9!acMb0>gEoIIaKZ{qqo>fF_#v!3Ou2M!d6|o^
znTut8ea}@GTn`GV#7#_0{>~ZsQ*&@&xL>+k5xv2}P_4Bb%Kz}1(;`E+5e=o9nDzcS
zyiXMR7qBHeuB}VPv0&rjkPBPI5-%)NH>z4iO5;JGaq`*1<SQ#H<=8NayDM^0b#+>r
zSJH>=vV<aBR^zzpigq`rglOpK6Swn>d)xinl4A;B!`X@l!Wt(GO6EV@q<nnMvcBiX
zb2we2Y3ut92TjgfOccVN-#LOJmX`9RMoFz^O1W9yzKsJ0M~r(yz;BP5XZb+WRjQf3
zf4fkx59^c`^k`ponC`ry6_gQODSJpuO)X##`~YOcXu6EkqRy(-rmw$R-D%jVpF>~G
z;9x=PEV>R;;V{_qA>9`D3sS%_RN{&W#1tuy-51~!*w+JQH9S3evxIyxY;DQlS0`ev
zua4#yr|KKk*=Lfasqqvif8qV{T57a!@;alR+D1b|6AcA^V^v3o?eFM&BN|r4mbS=_
ze={Rb>ycv<g$iUZ6`WN$0OMYCb+x0Jg?V>w6EYKUqe2KavPvyNKwLZ~z)xQ;l^Pns
z{$OKn9t<dU0gq2R+5RTONfo+R=Ej{j^kJYDhLdouNxvpfa6o4Fs2?oQq*0S`YYUC(
zNuq0=j5x1)N6DPp^uOcsbX|^MqGefu9OnyF`M|GHt{>DJWYlZuCIeppbB*#k9HHMr
zIoS7-X?3Lzj%Cso2c0H=U7M(@5h;V64$3Ske_uAdHnZHygBKwezraMn8)QrmsAW=D
zAO<WZ#T~Cp(4<tYtBm}33yHqx=P)M9<_Rf%eajE?1tjOmsaj)e`<L%y*Nc@Guqjr4
zO3M_gvhDZsX9g9j3aPsv=?@6GtLjt&6!^Mc1xWZ>UghQfMEi>jAs-y;k_fDqw^0b)
zCzCSud=S4Y&d2pFk57Fx+bZ}|sMXaRp}?@4&HKv`1E;=YOr{SK=GOJ8+GBwM!5`e0
z&CKgx;GHFVv2o)o0{p%ow_N@&hKRrq@DtZ<N8YSn#P{iG5aL0tsOCLO#V857%t6L|
z$KR!2{NaKB^_mc_LfA5zVg<DF&vuj!A~|X?rldQA23~S3&eeBrr2vCOSy&4fX=sQC
z8&liCbIP#aIsOTR@bh?1R2vB~3$PWWNzwZenDqC2{rCcyE3orfj5LF0R2n$KVd?*}
zIsCcUR#Ih#SCD`f1r5Xd*zg6lkJ$sP;0ea5=-H}Ll9rul0JUcXIjf()-}|L6>_rI~
z<9Mz39J@XBi$E$6asZr}7D=QzoNm1=OA)1@5e&zbksCD`6TRYiRjfHj(A?~GZT_f2
z$4j1J_qVV3qtthNV^Qz7OyF?eV^*ET=Wc1Y)nj1es);^8(F@=GkXCoHRmqTRpj;Ml
zcoq;9M)i5tnp4bDHc?b#Mn^>mMCDhh{FGMPD7_I{?>7)eOOu&BZAs$DWE{CDR+XI#
zy~aIJgKz)&69bqwYN;UyX|+|&PA@RXo+~0iJl(-_A`Gf#Cd2=<^(YxcH5-fDey&4u
zWsYA#l&lxI(G};^P#g!Qtt}TNZxG6~MzoLT{vgTM%>VTQTwN4DO#EQsEf1H@JrFFL
zgFZaD1*R6PN2bxs#l-Yyp|mKmjPE`C7DhHUu_&FLX|70jx~jjt?cqq?(_0!*?wpLC
zc`U|c;!)06At1ZCn7p*Fc5?iPwByX7I#SMZ^7!}D$DMVi@P!iW&b2POU);0NTUzYT
z8METxIc)&@M;cgL1)$B~_>Kt5thA`VLMI2hP{YpB>7X$9=~s+3Z3U^#!SDtVYDFpf
z7rtK$aylJbp7y4Sc2|vZ2dT{au}UX#Hai|}b0yEVNq<BFgLr_U7ZyiF{^kUbn-l02
z&1geQ@%E}MM$_)6ZdC!KC{fTZg+&%(GebznU4LWN6+j8xMXOVBw5z}_gIM(J4TE{!
zYu|u4@E~t>8^;&fU<~Re%v|cX=QY>s0k;}9AMw9EJ>KcC6u&i>OJeCe+4w=XI`!Ua
z>|Uk(H85d~RaJ|Drwv)Ew<g%7_&?0ORajSF_wKDogLHRycXvrQ()k4`>5vo<32Eu>
z?v|7eK@gB`q@=syo&28XU)O&3zV_kv$dd(&xn_(x?t6R&A@BPUsMA-lD4I5wARbE7
z2>qVNVcY6{H2sa{oLaqjW3{jgk1d;&ayvgFbnL^~*2X;d=KSn><LzzltoP%jil}s}
z`@z{Oi`0(`b>66cx~<x$_Q0eJZrD#RALckCF1Zd4c&#16<#IS^G@U^UNL0a#(os#7
zMRk3UokkMHm|ki>K}6!(tNp&Ui9V$ln~-ah%AvO?zmTVp<we#lIXr81)*(pXM&=44
zJzAJ|yHrB>8S76j7uF3Q`O#G>L@WDijqB&WCTo^Ii1WwD9l+QlFnnvYnHG;+-%|I*
z1sdB%Slx@>4UOy+o9i!1Wa{;YGi^zHEu8V_2+k4Eg(VTO@oFai7grN0WMIeON=!bF
zA-EHN3)NjIy`6Lwm$xSIY0k<RrfXT;$wbPopDBxm+j#!USC&ye5j{*SRm8b26PiLg
zsZ*|KqxX?bl}FALKLLeuU47Ch&n}r+G)va!oU?&G6ja&Z+CV2JoJ%XavlN$4^t7hF
z`ulnR14toAaZNm84VSzz^}866+<)3@k@3I_k~19}Vxa!jsQ<cROiYR3t>%2$h3s!`
zKN9q==8dZ_*>ScOcTEO8zo9IOcJ;orf9945o9LU{#uA=%cJnc6I5NUau#LA|dK40+
z%2F*Xmz<cU$m^gp2PLbL`D&dtqqBGCbM=nhM&o)K+!P6*q>eD2xL~N5Yqg_7$1Sd@
z<*+^b2r1#^4x8%J<HH;|{d=N-Zkdmm^udviYZqg{OFQdKAx%JM4%8dOgw@iiW_tPf
zsuyYNed0y66JNig`!ZrRMUjg4a~`;T+O2PQ*f0V0stAHsI*D3wcsFRgD8#*AL%q?Y
zc0mZwg{q*huW5${?O|*@`z`TfaKme7<2ErHGYSk+VX5<qw9pP;1*@&Vk3<P1W%H%n
z2E&QwDa?TnxS?3UF9WsTAs)G=A&S=clVxydM9OT6auEeVe1C~eCKq_+0zhK*W~Lw3
zZ&WW^6Cye*AW*lD(DeuGqRV)G1=;z=jxO<KT5x^o>7Wrr@SA-2E4@-@2zc37xK*{)
zjdfO3G!SFq$VeGGPdth$LutJ@fe&K&kP>;nh)^eng8;LrpKNvYV#?D<`Bc4>TiSHp
z%EYpkrKcsz_R)*(_6zHIKQm%8GozlZwoEUSEJw0S*<c|m_3wTWVKZu)%+aObr+5+(
zM{-(?nLTwp2)Mx!Wf1);(V(3|g#VbAM>(Oc;k5Zz>;BsZP7Y3vQ|Vo9boy=~D4S_1
zkfteqTLVQzL}X%P1HiT=Oq8G5_+7S;86|0EQn{=nc#fLH@uNAat5ws@Xh2RrR4DtT
zOuKBj)|_tVy|r~%c{!8RfYiom0po0w18e54C(4&?kX^EAp`|DmaNi%~AO+Zk>T<j;
z#4mC=kIYQW<2k<Gi3yq$OBo3t^N`7hk8{30Nv)!FNPb2(ox~S#<qLJi9Kn+cPf4M_
ze<<THj7Hr2k(clN2CJ?{T=_jqn(WS~>i$A!Rs||QaMwsYd^JKuN7E|R4ES1yX|%Pq
z_C^A#(x4-PxYeCQ;!RrnYpHOyP7cU$HslA+@zqsGT^*<X-tI1bL`1XO9?nl=zUX7N
z(ISf1K&Ly<AtfapNxdL$ZTQFE<NKk@fZw(-+a9Px6eC{)MWm=aZ~5v>C^Alej*N(_
z`w$4C`|xrlLQ0EMX()pIUbPL5n5H7R>hYbcW91&wV#DK1P$RtCfx+b!bZ0PNvoGkP
z;|%5?rrVJUV7O(<*7&1#h3RzC`>F!c4QzB=a2P^GwOkgmq0oGVmBtbl%A?<F>2!W!
zM^!df*YcRZ+_QRvj~pGU?-bg6CZ$A`0@eL$fY%BiWrkJC>?ms(j>GlkTfljNx%kV&
zjC!i%Bx@Uk0+RwRFZRMeTG@z61cH@mhju&YlCPQggCN-k{10d<W}CJd<8QI(D(Ai<
zQ415jg$<3lQB9UYjNHRjO+`r3h|z*yw91F9&ELqSFZ#o2T840h4gP`kC|NYj<LN-A
zOfyTo>DuU~QKz&0)LT9D&f##rINiF?J_D4b4OZ2o#9?G|t}HjB4RLrPF34H5Yg0O0
zYch0kr0?9p9@Mq+nmevtKio%9|Namf7XNG|SkCZ-ZYY02+8M`GN3-z(=P*S0nGV22
zshqUix()n^@MmXbx`<jfN#Y1s>vo-L2vy007HDD4Pf_l^oM0#9!u7(L6k*mWflfED
z*B$C9FDIhv%_rOUV$vr~9c)$4j4CGFe-&LDRb8vLkL2YivlwNy`8%@+!j2J`d6JWq
zmqL@kuE?~=(zHmN@GjqutTwOp{o10MG=dH%Cnj>%jfmj><+V5m50j4Q3kHk2{1~(c
zkNF>u>aYq4^CTlx(>ypgV~E4Ucjh-k6trqPB)H5*&{uKPP9{$;%Z=*#@mN-U9id$*
zK%s61T6@FmXXS|-$gT)j4*lq>@Q%=YP(&}IUdVo=RQ&@#lFXiwK)ujvh3fYxE%2pq
zkhTx*evCqlO4ZR@)Lr5x31%P$&q^rZ@QJDTzMmfo1HKr8o|lEew%l=?56XcPidj$#
zqN{0<q`W+k<NqFK*Y)B4{_U<F13qx^$Q8U&lNLZ1ne|5jfmT~9%eGXw;>t4tNVN?g
z{QmF%^*Um)l&eR^Fkfku%mlcD;9g7Eq~1q-r^sR=JvNSf0nr|(w;hJ}(t!H&Wb^T6
z)CpP&B=v(?Lp<<bGK|SW4&++H_s_f7wB@Uc?IwEP99QWD@8-scAo50usXc=ZawX=k
zH-Mv>rttrFAdY3w@B}T9p=@AG?m*fKz#vx!iLSGWUNWd@UL;(5&PL+Jd<H-4g>1GX
z(DUTiR)@Wr1)z(4%(g`?R2$on2+s#E(`qjYky8pB`ofb%%mw5&9ImU;<>Xo3&M8aw
zKbKBFn>Z`MIQ)}Duz>1IbNa6@@jMc3=+N@<bOAsgBNM!X0b-gUV_9$HzX$2D9sI$Y
zD|-u16B)5a8Wq{jii`|Ln+gW}7ccO4=YkPm5(}N2Tms7=db8u2Sx!`+;5K|Qm;-?E
zp!6S%2TZ!<i2pw5&vW2Ezd?Y6tQcnODai}v0$hQ6pBMkkkZfE~m+}j#v1y&+*@ylL
zgvtNwi~si-wF7?~5ib#vjQX#i8xaL;FcB!$v2KIlG*$(9+5h#^|Btu$*Tj5&xcFay
z3bp^Mm;bNFz{|v*vg*Bh_8BOGcZLq<-T3cw{y+cT|M3V|5z6`ZU#7v^S_pgvj-7z@
z(f?c~|Jz5NhX|AYDVG~8wUnQKz~s*M<osvc{*RA5kGa%v=eyU>e;VhrDD{8(F8|{_
z|KI%6?vv0wQNRCqns|NYz4`CK_C-$?-O~q{y{@UL_y^-_{cGL_;|oztOw4ir^Oj@Y
zho|<d<sVPH4;8nhqiUW?*ZRr4?JO5f9ak5|%~9YBw2y#66~k}Ef0D4VlltKQ?>*~(
zKeFEm#lY8ac;QO<dg%F45c>UlvVncB?dmU<gwgxT@)~_q;+IxqwKn#cB#Q$;6tK9+
znWt-B5o^BFmUyqs!{aZA<Gh_aV{%Us^_@`aJ0Q`q8nt-_akBj!$4SKTjJ-Hqfy*)7
zX^#hvJ(q2t=>2p9A8uDV;94ex`_q)2q-~56^hJliS9-B0XHy|Ugz_KHHLqChN)KZa
zC67?~cXPde7=zS~bVfvzMM?t~guIB0aE8Z~w++&={!@0V!Q~@FOqumMRo<-SUq(ih
zV!meRyj|McZ;CIXY(cgDuB??g>j77U8obx`nHOvvw(zdI8mLAcm4ib=JpfG?n^d-v
z={<ROVc*pbjL|#4HaHhs+>k(p<p=%3zU?lsy6y-5Nouyf_zz#gzfT7&e5fsL?P9dq
z7uZ1r8{^>xa!PVB-0nawm#B`SskYwd1w@BfZrtTi%3~HXF$+i!Y{0vj4q_A%zEJS1
zI*}$5)AQ2w5$bIhaJ0&_dP31DSLW`r!*ohD5uVE*WMpUnoT>1tT`@%n-%snl`ryd^
z+Pp`m{q6%Hs}F7%R(Flk-t=|u7;b2|S436k4J_*!B8lA8Ox0t<@ajY5?-ar9sO6!@
zqaAf*Us<@iPF`Rc8Myj-x^(dOeaJFpYPEmXtUnl=VWdMrzJ=_SWXt`QI@!au*-ZWY
zqo!Cndc&UxZUo*N(v~Rh4lSqsiC%MvEZ4@Jw>HV`H3%mG{7}Ik#r}=rySVSUUW!nq
z4{hIlurVBg(t;9NH7tQPSUfPXw~yN27ozGdpxD<#u1xpKKFdVBC7S*m9sG?ShglaI
z@Kye8RAqusZ)5&vqou=we*6u*ecyZx*~?Xry}xEWPtn`|I+4KOFE{Mv`ZXQewGr1@
z4brH;C00~p9@%;olKrU(6?sU`8Wi?T!{O7ahua|>2ZY1F)X~vl(r=89Y;@in_DGe(
z7xH9F<*^S0{UDHu_=-c(n9@xXS@mIYb#RG9ouCwO_LA(MK?D(`qEQh1*92U6B_$=n
zoO8TW&lb0HU5d(z{VhY1QCJj9k1_`tL_>vus!hhkp5<i{#>A&kN^)7mPub*t%<o=|
zjVxmISk;^I8HnPBwd+?7r<N5(BekHY#z48gHdMIe!JvQD&mGd9$1TAs3S8wZKUzAw
z-~G{D#y7l!tF0U(`S!)!`#o+yTbl}4XVR-*0|JhZ-7Q_-vib2kBgy5r4L#^pRMG|t
zUUzj*C{liwDz-ED0E=&}OR#`-&R_n+^gys)c8o+AjgT;8tw)xORmBLWv(oS8u3!42
zjvFsaW?NGeVSqn(?5Vq?!|^_t=9m)ybsU5vzw`~m`i;OtdXK$`Or^eWsPKJwndAdM
zzntfF_8{BpD<|l1sORhu%rFdT51!O8s1VY%`~iN`@kX>Ljt_>z4R)|{e(|C|Le}X!
z_8o=GQgV<Nc>?NKwnkAD0GSPd;-c6jTI0Ssw9dbn-wChnk!G=*x`V2r?TlPxrb~<e
zaePmk*z;EM1rO@zvVemYIV&XvVNsZPj(#BjnoCIKH&JN9CiL@K!=ZQvbM@_4;00hj
zRIe7$tWd>H056ATsS%<zeF}Ul&X?rkqQM_OBoREkHOoK=Yf;S7H=Lk@PQ$6t4v{fa
z`^x&p#(v;86Wi2HVl@tkiHT_%Kl70X+v<0Jgnz=b<*ol2o~c^Ew<$@<$1ZV^f;WJD
z4t&%CjH>HGL3WRXq@c#63(DOEDEB>)guGw=%6zd)_ehaLzq9(MM*K0IX%TofMP3RG
zGvy$E?Y-s#YLwAOmg)~M&<&T6t9qY)@#o?ge-INSrhT9U$EMgTD2F(?ob8#fvzpHZ
zjD-hgfNEycL(nzcPlTnfE8r#yTqV>;5?R~Tl`!mPy^heZu<KntA$Y<z^_5%qzW*G(
zs>RQK57jm1sh`hBtI_}X=-`9mPx)j&@jaK@3gUjuk&>@6Y9FsO)Tg4QHXmOVhZ*+t
z6hv3c`rQhsHcKNZXToi5ZO@dsk<5p7rrFoee^J}NNvIZ`gMHly-WH=8|HRiYqz=8$
zVO+2jq85r@3H66+!zO}bq+=6KP>_MQo~IEWAy=YN#3K_)(AQ7M>n?OCy|`f;`cDMb
z#{L|(^k?&Wnt=EH6oX0M5`YodGMADvD`d*lnhk6t*E}{*a2_7X`ftgFH}OHv0f2LY
zJP9zF=SoUa@O}RBezf{+LuWKWyp^auIP4kK8-nU|wQnu8g7}+1MLpquRijEI{YQL2
zLWQu=I(6-Y`U&vB*O;1mdfs4gsQ>P4o=b9+y!*BB_NjHtH+m%A>eyipmgG^(6k-)p
zzHJ?f&>A!z1Q(N9Y&e@{1G$^3h_2;+wYqAANQImUB=R@0v7vnaD6uo#Gvk-#+Qj0*
zqOwNsS$LuzU@a4Ij&0|mGJlulCFfB(8(D0;=`}Q1sv{g*H)~io)h)+66EBfZFx^%U
zyhs4>pZa{M!Z38V4FR+6o_fDM?$0W&0tdv_Cdi!=6^FWPxI@39e;?|#!H(H2ehfvb
zQd!l0KY`3G=tD+Gua_riIXXUPv@&6IAQLI=)D&m)@Z+}}dSz#$2kv_!p^tIc47tve
zCVUAjuQ(YMgj5u}m(Hvlx8K1Fek&X;E>Cvok)hL?Lv2)MVPoeMmyt(YKU(gqFee9E
zSKfseriO;+q+f;6D}d581yHiKbc1|R@9_MASWhB?;Y+*sjpsxB8Du}~On^87P&E{6
z)gtMJ1~DMBqMws9+j7Aw<aG|eH(ddfKs&CLnEgJZ<=s?1IzwXB3qfx|3rkD#|AacC
zlGDnz@SF$2oUN^`1%MT{Ue(!{qxZWsyJZ5%Sk<+SGCl@ED5U%mhOO><SpGDTL6*#V
z*?vncbQ`*#M+KRgUju`-)6Td!Xe-_Rvm<-c383&sG6ZqAwk+S_EFyP}2Cf|*{*;%d
zqM@l}3S`#jp#+i@#xMwSim_FZxNWp$KV+SfaeTU!Q+m_4G4PAeV~q%fnC}xVhR6lD
zX_U{c44tRu&SeJ##U{rWpAh)^s%O8E<027sM}~revhc_&>rD{x<L2fD;`^jnGltBH
z4=2Ic<OJ0=m)(vk7qU%gdODGhpefX$?BbY1)8mY>sWpWD&M1*dHyoH}6&n4=dZBH#
zt3GFBFkc?5L}yLkQ<bgQ*w1%$N-JFQKdw%|K>LiNyH{3%j}!96-$A^)Cj?Adg3VWi
z-@7LWYOW5~7q=n@<v!5Nbp|(ZGjP8TpU}EL%Q8$UDG6=$z%`|fBBy7N-);jbd$Hi#
z*K2QExD-c5j+sPI){2zL))#dLsnsJc_#H+jj9fP|1pPl%8n)IRN&4Pi%rw7;jyb<5
z!?W-WjTeaY9k2I=OpnG5-NmHjT$h7nN{GuF!AlP>8=N>yBhey}Lk-+bABAc$3K%RH
z4Qh3{oesacQ#$=FY@WtM&&<@Y)*L6kZ%4|He5GGL1*SmS43lT;LYgp%48pr6CXjVg
z_~qUx4+K@yHoKmROsMB8n7__C5gB1tbgLB3>8gxOLic*}AQ-uiQqOf>u@!|XQdTg%
zz8~w&*SMXyPY&3$stwBcnx%&*I=q5!=igRio|TMopyOU3U%ek72Z;;kS|k-g0Ovx%
zepp)F-9X&N=rhq8!T84iZ3HzIJ*?UN65T8>bGU<6k}z&dG2tVJUF7V%F2_R~Ku3VW
z$z=n;#>oXXp7={=D1?Yk&)6RM%3#CD{!LC)>IPvtq%XBK?+rpFb~fpto34WL%9t@W
z)SY&Y2t)niBVd^sa=jjEtLr(;UFLv=ff1WJnf*4q<4;T?S2REsK_siv-I5PTYY=>e
zkX62Cm5qsIJ;*=!KDuExH`8Zgn|-C^aoF)I;J+d!9{DWna?KKT^t)2{yl+NH%$A>=
z2fLGrPcIun7yp?1MSFy=_>Yd0?5$9Un^Nz{2o^xKLN?po4`;)SXR2o05d5?L*^aRY
z!1+MW^$w}s{CPW0{%1SRidx~@veoqOBxc~Xr}?m!uq24?b)7x$9d%eRe*MKSf@$31
ziZy*8Mf7PLueWUS8xjLF&y>L00=cZC^|q*ueXscbr2l<Z``(K$>wj})U+&5u!3PFX
zm>tb}V??IDi=a<^YsNKQ9M(QVb!mj5U>3<-bG3|Q=70zNvOr$8_kP7UEnsd**FDit
z>VD-xMAY=Ef&XfB-+8R{>OLU5@xa8|wrHAV8^QyM#W2`Wj}(4Niq96d-*$KX(BBZC
zaQM7_e%rvrmu&+D>;^}f!mlE|d?Go1{<Jx$t>bWVT=R((H=}S&y*fbUv@P>FndR7m
zlc|(TA@|Aky~2vd`FzeutNtQD?wiLJmb#g_+F6gXd~RQNrKjsI6pP(*vC~&iovN@F
z0EjG(&poUy{dIPC8fDK^O$U*3<~ahUg&$s<t?0X%{$8!e@=vp~vnHp1!W^BPwDe`Q
zEpaSIGlPK)RCfiO<%!zbcp`Rq3~<`h_sFlkBX&70SNo7IzE_In@2_-{s{LZWF112^
zlNEE!i6Can;lRooLBt=25MWyiIJ9QF6A%IzQ5cF%P#l^p;tT7C$FIKOruO}z>j6$)
zLn)o_CHQx9dp=jUoy=oTU}hGNU@Zw8@wZYdMS0|5Sibl2ceh1zb$IB`hl~7R$5c6(
zAmA#Fe#Ump%v3!Q`#(K2sS|=n9v>f*IUpQ_{-A0l`+;r%nj#p5GFd<AbL^XCmu-lb
z03|B}j}~-3wD;o<ISQ*`K2KXknqcb10URge#VS7c<JCoHb4OMOD{-umUKmF4)lgCP
zk+Hh_(}?x#x7MDnR-(*c;qF*!B<cnHQ<vp&DS1^@wEigKo@piT&zH)6+Nz!5159!+
zs(c_;u7B0iUe~!Fz7dMnCYmn6H$WK`q{|$LH9eVaW^#CtSUop-hxK^-U~`Dk+qGn0
z#=5$(>_#1&_NMy7L!gFd!h_!HxX%^$<Jk$u9dFx5y!t+EgvG0I`@X7kKBWF;k1P&A
z<{b)0UXIMRUyte+W(>M15rqC!a(B9Cc3kdUDn0mw<Q_3%4T|`#W|9|9@xdYlbUs<T
z#iOb|El4a{FZ}%i-+E<WPek!o&%Hd*=7QVbK|WNeClPxaPmP)<0Qy6042puS24319
zba&0IFnNjJ@7MFDXxG(*c$mjpJ&@U9l}zL)g5`pP79@~dJ16wQL8y4sdgTD`;Q)d+
zh*@6*faJ)hFDY#N=!#77k2>lIQlQ<ma^q`>@5c=aG_^2@Lk~%l8k793q#d(eot|bn
z98V7t_L3HVGyq0PQ5kpn^S>59;`8Fidb={1{d_g!q*bqfMNOm5;;<5~m(f~{9C2(r
z6LYWVM9>ydM%Y%WJ#erg`#Thi(c8P8P#_TUe!S%9&Ps)xhY2~w+M4cQ_2cOULAA+?
zxr)3HN8clo2I&vbusKd+Rdah`ztr}r1T#jcUfLlM65`M?>(t>;R2t97=UZ4=*6c@y
zY>%zKc?a)YYuxpFjP&KG{`bLI==cwH-qWr_BZ-#29RX#X>lS7sL3i0Euo*H9jUFbn
zhxSX@^rKi}9ku3Z^KL-V6Nt1qSPd1VjiL~_Cr#GV?P+p1<kKnsek;Ay9=TDbEhkXC
zR2^IG=3-uSbCe;_Rr>+Nh9;WB$ORect0ZXf(I1Cr1>VO~#hSrqy<6e$D+(rm&_kn;
z#xBWaIgp;?clC-bcmARcZ!;P7P;M)O2q{7jdvhfS$!H%L{McRu$H425p%6v%mMsKQ
zS3<DF__}jvDDVNFHO}ekj}ukDP+up1l`f4zU;ii`yTa9rCE8zCQz}){_tl*Dg!GSe
z`N$OmNvssy?C7puzbB?iST-4}M{4T`r*R2lCHAtotSK<tY4S|gPBv&`QWJ-3<wbz4
z`A?f)<V1l&-Bm#-mVh;{+h5H@VTuT+lXdFezW(l?%b5^ruJ?<frqX|&$*RBQ@eu&C
zsI$4Xb?xVYH=px4+r-4g+J**7hJf4W5^H*%$DLn!C&xLrq~kw-DkpQ>;xI8W9iMJ0
z5b-&F98F5^fyHC}4=+GE6u4{G+x>#V+NWBmae?D`KTCCkh87w>I4HN+=7wayP%o~j
ziN#{j96CF@&Bn<cl+{UEtl4vofC(lMVx0Gf4{N8JjLp7xIJvobCmf7O?v0nlt2!?M
zs~S>-@n+C#XFO-{r(eN|spj~E(dXl^oU9nAN(2OiotY#3xgNsqs?k*Lo~D&ca~GC>
zN@cfVaoro;uQ!vj#CkP$A_<du+)}3%xqRil-S%e@mRdc`ZGkF=mno9qk!nPyUQ9~7
z@9B7mEW`juk^SL^0n_f;<w?Si{X1C5r?`W30fgN*A9B0m!%n+G)w1mbyhMZr{G)tl
zJTFeE?<&>5eJER$L@Szpq)F+M578?1&mL6I-=>4y754V^4F4o~IhtB`MBpmdJv*gI
z8EdvP_CXSpzMj<YPAitV{^|S6`<B}opDR1@2L_U<g)!g*Nwi|M*UdqTYGSYtziQF4
z5+!}u!x0Q)=JBRw^bu~O_qM)sNIKq$3QR{!s>!JZy}1v@mS%#A>LJiW-zZVzp|&L)
zlQ<~l6dx}3{GE0!@Z#1dP0Ss_T9ka%`hNyS`kNgQ8wgcC5*3PZLTZFQ;6+#8jiKXa
zM2wv4R97>=c+=WW!C%T)6T1BQbN|KdOr_K?nqg47<XiT2Y0h<}IFDVw_r8ZA2dL(Q
zx0q49T%88FHjDsV1bjMcO(-6B9J)Q*=<wf*;wjw0()pfRj$$(+7ifeTfgw$~=x66J
zXRMG%5KB>1H_|S)VZja2k@-Ia-g;#E@OF;y0&heYJn@xtc#v6rX3W`I=$AK^-W3rQ
zFbvQFJyqe1kB=J+m@qmGEu+JEPJ(?`*u%9gj{D@Hvw7w53=KlM#vH?q6t&U)*otK0
zJ6HNT6zI+@L{d@{$$^Ia4|q*gGPM}0QZjMOX?~v|KVtCBkO>B2q~x__cz&QU4nD}+
z<_I&}owbbqplohf_l`@Cz|wlc`r}ujK7eU2S~>igA}}Zvg+32CS1;0c*2QkOp7Ptg
z@_b`fMjrKoM`FQfGRyGM7iC7sPRS+fpv&6@f}Mv<Vej?!D=h!2^S{3d>FI~ZocYa<
zdH%`men$fr_dYZ_oKZZm?}iEfv+tt(#Tbe~yTB^))9(w+HKWY!o&SY^fIjR`R)1~z
zf)cLyp$o<(mZ;1Rs?>cuS}jMMjqzKZ?K~AX2Ich`37vk00G)>U%+wnj7bIg%`av?S
zRE2b)S4v(Ub{*t}#80dIk^EZaT{ly_-@lo$kg!Y(6%2J~fzB)8WV7TT52-8mk@;~1
z+q2yBHG^mAVbax|z}ODdhI446FsFxH4~8z%oI_frJ6~phggF?s$L#;8+#Rmu!_13H
zEVlD<H+4zsFrS7-&hi?sn|9tKY70mF^+L8pB&+~xY)a;5;WDeeMb=J~&zF0xV)>yb
zh<0-X;`S@%th@60?^<*WhVODdbUxaRw9-ax{9O_Jok>j^MKU(*`0GKusDd7zo=1Rr
z+2%Xh><3Q@b_T?|t>U9EKQW7{G1ta=Na&R#6+oY=?Zlcl^l#Ru&<HD(GX(HN?!9@Y
z0P?>^zF-6mBBz)pf=#WEXkniZfFB}{@L8@0!^>H`4kSM_wM4%$i=41hFEat*xIg#{
zF8g*8zkJs|LqGr?uOCsG`PHE?=$?}?zZp%+OA?^PKnZS~=KT?nKOv<Thf7Em`c0Ob
zX}4FTAM>d$)M+HXi!$y}D@H_Cd#w1CKRtLfSS6&T5mL&nuO~e1&?-7CHU}UljFFi#
zgR3rpX1_M+WHG9Rrj4>?QBhL`(?W#3_Youd`M;`u*u6$WtShLbQO&b_IAR2x?#PVP
zNpW$*Mk?|nl*C-7*{HD`cC)HeWjd(!ogMe5fSDax4GY@fj4g`}M47Gd4hlqPjmV)h
z5Vn|~LSGW)jNP$lSmor`pesK%BE~ko-!hVKxSFpE!ne?Iw;yQN7{+HmpIda_addB{
zVrED8SIR1{m!Xgqh|KC8&H_7|>C<AC%Z!BiL|xVBs%g0<c}igfmZbkrLV${FVAZQf
zQ&v`P_J0!TkJ1wI=h78_C_VC(p)C(l_C=+ZhYImn?ro@hswM|@@vw+w>X&e_G@7s>
za(KfrcwFC@ZR;Eq;2B-0lF=3mR=#3P=#uF%4cfhjgd0(&b}ueMk7agzZ@KxqxhLei
zADbTkQs9CWqB&f&T}{Kr!xEO3L|N_dWy`s4s*@~0*O&eGl>c@4l^c!vZL|Z70;7hM
zkn;B}h!wTU!dDeK!N9@DZnp09(DI@uInjlP2>I-RnErALj2rvDk-Xp<<SFaFUucf#
zQM{KrF0s6MgXKh?goft=cA|@-G2^GjCDvA=tcR-O=Ch`nr3b{O-er?j7BlJW^tE-~
zKx2H$%t7iWHMxS-eX3#BjmbDR7Pvq=LtoMXyY(y6I`h$j#eI6o2yqvJIJsZjep+r^
z2_kv6RNfj=h{9R05e|L#UK?i6mro^+ie$M(E6LgsZwnQvhu_AT*B6CJ<vgf7F1M`<
znNchoAPAliqycZq%U4({4}kUid3-O&nI)iU?H_Rp_F1)q4LOf|{U4&FAfSv=GYS*8
zS}93tr#g>%DdmAhm3<Jiwts-6&%5>urVl|(i?g9xZ2QgREU7APlYL*0i&cCoydG+N
zWbF~|(yLvw(||y7seZYHWel6B$qV}|KJ8spw1ANu;X2z}i+ly>pis8lRTzA0$rzbW
zvjE}q+f=mkLl6PZGPg~R_|@3lUPuy?g}6ygSP*{Ebh%9LDB*ktosB77)(%&R{08Vf
z!Cr679~-AL|K$*S!Z}BszjWGf>TTlC#s1nWBHhNfy9eP)skt(DQTq*@R?3hrDf%<y
z@<173@0zA!O{3GZHV&F|rd=Ly6z{@u`cuYM_nkTj(&AMJsZ1^~_9<3vq;;&YVMzS$
zc=E&hJ6-X^NE=k1pPAmAZdNxoTBf4qzs?!5eBvV1%-sc$qGVo&4t34|Ir1zZG=xHI
zi{}GjkXF~%BlLM4OeZX)t*N?(Z{mB(Q#j|UK-Jl+U3<Bdh6HTwO1+`boP;v%wQ+Gs
zcj_(tOaI%O%M8I}E~mJClO_Y%`^$x@>iYUmjp`5>a&_cQ%x^OkO@`v2#Q8Si_;Gi8
z&av9#JMrmW{+o9*jl^E(`&}#n=Z<o?vQeN46rN(MrltmuZB>wevW91U!U(L?xP>an
zQ-I^?Nnafts-wk+<+{?I9VaHqKNFXt3ByS4scb0;uzA}!zQ%DA0l7?~6Z+(WEj|c8
z9I;1QWnOdbJW}6X+!u<Xx!=Uib9-tdm=%Z*f$g&UI-M1n0s;sH$7g3fUvP(ZCW}-$
zIv9$LAr2pX2>dup`Q;b%{#eSZnmCXD;ZgkfxE%S;UETjFaj)5lA4+q*aGMQuLi*fi
zb<8Ei+RQ?uerewkm8NrwG66q1IpicDu8dE8*1xaygC*WYn7Fe%7%_mQHt@19MJ{Lc
z=!z5j4-!N`Rp!@%&APhdk0Me_NFuWJr>>?AJq~qxe)d;o{vwriG!bt(EyfDb68Up&
z>aXsiZTo8NL;2>bsg91gopf2)b<eFkZ;y{_taLTgbp8-S@7|faBnG&>8&LX`u7MW+
z884`xohz2Yty7&vi;(zeCYMYd9+f;Osy;qZQ6yBSD*gv7qOR7|Wz|ZRoF!w)A&eJ|
z-=m?Vc7G)NT16R4n;J3M?+a+H*WI)1;m)zH@4`@6o&Kf9O9ud{JsirdXXII{<pV<4
z^BUYdEWvCpl@yDEg27_TMvgqd9QYWDr!@T-!mc=*Fo;88*sFc?_3*gF5RSjfh@IWR
z>AU6~_QwQVjNUD7bUZHQO&n&mh9x+=!6ngAvD3P??{))Dn5Js|uwg2cH4cYkag!6I
z9Hc5KUCOEQFrlE4SIWu^xe#cd)5t~!sjtV+<?HhEdLR(`Mv}{gYu)O@CzDeYv(oUc
zyH8alHmJ3br&0Mmoq3l3gJnral@WyURD5~J*ms)joco`(DVwi)iJvg^wYJdsDp+6a
zMeD$h5hfu7!k<+2efWA)HhqHLNq|||-M!~Xkxb8yG%Dr$uz$PfCGy?&gTYSVsrn2(
z^!cs@QAW&H4kukIEs6s50{L<-yty&aCuD}*f}z{XgD*foC8OEGhkZ?uKcxtIS$(kp
zB|>qM@Tr9A7X>BIx-*rAx~f$wN6T`*$mo<`jb({OhxOZH@B-jm3-ruc>k4)OCzh_T
z#bgnqd`|GEPi(}15eLelRKXE@rXVEqueCNUF{_x&_EOVq>o}G&d!G!WhE^gnl=~Yf
z&a5@-bB<v_51V@eT9|Xu2=L(U(+IOdQ`RQRErgEmg)2@yl`oC2mG@3x__jLjg-Q@c
z4VqGf4HXlu{qBsfJNVRLmK#(nPv5Fpt(PJ4*|^8#a<B0ukGhjrO<R7(x=svgUtLPx
z)aNHIZ0zX<s%SQ-uGrUWb=ym1wN_Q_9A18L!?izA#3JW&r#z^!zwG!L;!Auy#Cst|
zG38My3W2!Lp((yJ#Rz!Xu?=n<%^eDfiA_y{A*WGKROF{hAt%Ra(mBXlX$buRr=|^+
zu#;m#pW!t&7S0(fF+*LR1wA^$%vZtr!6ft{#~Qgt<(;wX-;(k9oC(=qD#+e8D}6nm
ztu<pZL_A5iLCQjS%cEVQI^&VYh;Qu$-3|*RqqFq&iu_N+T8Dj+wx4Ryn=lws+-dS;
zmxn8%{goFw&LR8k&FJZiKiFQnPDHa$;c-Rfgkec8RY^<d$1^)btK2fU`p7kfou~Cq
zSK4wzPoKU>V9&GFqSnj56i+g8TRS)!Jp2}^FosPZCb5@y<ilyj&~fx|v&{pVG!?Kp
zIEe84BR7wSJ}<GOEzmjdAJf(I0+Vj|6x_Hh@j^mh@Z?UDB4`Vp&^noGudMh+*{!Qb
zQu27eOSrHiMgn3?C4U)u2PlUz!v1TQ?7Q_|C-f>#jc_o`A}J@Bgr_G73S?bgO4F9(
zs@iZFuEg$&YG@<?%Mc%gv`x-p0Ho$gJ!+M_M7jX&2e|H<bO}~+at<gJ2}$%T6+DbA
z5iN$76b&+xH&VD+9ToOSJyq#)LzL>dM@xM*2@~C$5;UrMW>D@*8bAHWIO1GkUD*-e
zs;{csnT@`hV1q~f*3Z#~K|z675ufDgN2sq~{<gCDQP}&^3|x}qlKvcRhtO+a)2?cT
zVU=Dfx@EbsGgM6Fg*cOf<RytS@noUGSAF0Hl?j4<P39Koe|@}GtRagI9cP<w%HiGU
z1)BqyR)5Q?sOwzoN5!G#erPkJz5Rv=S}~fwf8P~&RVsteib+csN=MHKnmEyAy=#>*
zTww+OmFq;@EBNRM5FXA+DtdgPk9^^~f8_{on5%w@w4&m$eh$ierJNxW8H|)4JjK0M
z=YBwtoR+q+6{(+mpS9CQ#1y;@<6A{`K+_f2xf52aRYrb4kDY#F?t8dr4~H6*VWqps
z*PQi0=~9OGv9SHFHfzjb4m3gpMG8T_92(-&_bwL<Ki>~B7Rau+?Fg!9+b}dLX~uA<
zPd^USRJ%LIP^lwsxO9N7Kny(+n{Nj@?rY&VIJvVXSc{D}uZ{H65;ID(y?j}<5v_YB
z>WU|Oag~;<dINuZNaQAemRGdEm}cdBy{0J&lY<+fvyasEY4!cjvYF((fK@ar7^W<z
zPY~Q<0tk#|$pOO33oE|nw>^uw@qf;uZy)Y^4t+W)a6Krb%S*(m#}196CH`zcYa10;
z%YS%tfIA#@QnvnC{`aC-?h?+lWqDg_yuFbhi3-UZt9hG+FR9C@r5ZT)A}D4KUmsPj
zC6@-Cd`#F3gzew%Hg8TwzcY~Ea8FQeN@4Z(|GjVi%?mG(HJ38(_^x~M+k>Ew5$2<K
z<u|GP<9^&%L2K*QL3JUA8zbZ}C+4B-?}(lgzspr-WM6>$|KSb3SiH1^fJ*)q6<)%9
z?u7BV!8bn<BDf~oDpxV;?;XSo;jf);&drIuk!rh&6{!-wP%S)My)W0#y82)^eiwfA
zud@wC9K@F};s*yiJ>YEf1MAQI8QGXIkw2O09La_J>0{@jNNQje6@gO20R1&H&xBFV
zezd+U3cI@aFaU4%RivdHH^hf2xMZx<zcfw4<~Hl2T`5`S|1CUYfhCn=f@m#ZhcB#N
zl9DX<i+1(BsqL55v5rqcO%lawY{T~-PJ}+gSRQ8bG3a)T#E8W1O*S<wEo|w$EXu`6
z!@W}=jl{z66#XVcPE&64K}w!tQW`_iAWk947qixb_dTgrIwEdFU_yhSM}P{k^c+)7
zERXn9hJLft@e2+OB-xmWo*rVCEjp;$v?jZ*M$x7>Msd&7IM87?Yc(%gU=IogQ|Aew
z8{==`f!9Sl7u+8)L$M|rcb<$htaFzB41HCt$(8u1U1?2W|LLQ@j3um>PYK~0DHn~^
zgy9#l*=;<n-d8OB(OKCwy3a?^jnu-tn{OwXR5<wqLwS?M8%lsGh_*J4QA%cuO}L%t
z0|X@PDx=r7wv3Cd9;6%nn^mpkMj9ls7*Vf~wzfK`g~~{b+XGz)PoX%wa}b>HRjU4S
z|2nQ?eY^72)=&D6(kS=*t|P2(R#=NK;efwkte|n4c#ny3Bh%#DJMTnS971lKP@<BE
ze@A@M<G*wEa;$KKTe)=o{IrVJUFYoQ7p4(eLM(|H>YNod6H9uwxmQX5^t(cntoTO-
zU@m-bj%VMBUI>9@en`e^6Dtpu<&|=r8LHs=;aS6AU+B9QAvgEcVYhc(IUh||TnZOm
zp;YD(u@w`{R1z|!czh}(K4OB{M!9X01{dk&1~+E9U$HCAQg&!v;+K1@v|{)IEqX_>
zqhemHRO+3Um6rDGH8@DIA@wsu&TnF9-!SdTR9|NO7V9g0uh;|`>$m+@<^O2NX7FQv
zbuku>59I5+`m+O&ovJ|oM~f5-8>Cg=&CZj)*3&k@mr^Jsk&#JVcYJm8*+=SfV4X7*
zBY|?NK4h`o{#U!d<Yi+iPJHFvPt}9Il>!eSqrE2NC9ruXg7U8u=-F4Y13jbv8z^WT
zOI=}D2V!ACEu{j`^Ik~)BSpP^>qjZ6ub}4p(U*By4PhZ=&;iih2kmVpHr6Q%37`Is
zy*H>*M^rf3c8f)$*VTPiqzV)`f#Ezz{rg-&AnD>OhMr#oxI8TH>Bt3QmF8Gv6IskQ
z$9pZ`Wer|rG-qUdfwLpRmiX54R{2ul)5@qZpH|WdqEtRst+DmP(~Ok-P~5d^W1tUN
z_t#Jr-_hw&Y>>&&{l?}!Rx5A8WKBb?TMMuGd1Sm$D&KB7!hRti;q5o6Vcv&Za(3}n
z)ut&oyl}a&u(1AYamn*QdH8GEcF%wgU#z05&D5%B)<UJMer4l<z5&y_#sfu95fKrR
zDx=LmD4dXCtn0hGwdHniJYHpfPF10#4ykYvky!6h=LK+!!RxN-7AVogOt8b)A4#c8
z7B$wblU?9`zTfDH#v^drCFd#@5-~DRZ0;WD^PTJsFH(aWGy9YWtDXTc>TL`}%9<n$
zD@y)78^fA+1XMu3gbi<ddmje=wY;jo)f46}G6&LGq^UXe75q3#LZKrOcJS2i>52_F
zX9-_&3r#7nOcLF~TJHApg)L56J2DixoJiq;;KdC8WmWAN7b-zR#o!=Ot+C(YSBIf`
zFl*sETwEJug2i+^Z|3Wm-~=G7c#uLnJ$D?0n$vUQh6A|YSHbE!=eHG%v5Fssd1zo*
zIqO{N0-hPWp_B`?r$UTKR&r$v<w68a8bt1%byCChy;2<}XU-Ixba9y6N!2tFK<yzH
zCh&6qwGu7szTEq}F1!@J`>|jf9)dT-u43#>!=ch>HOal$ajU7^Bw<g<(%J%8&7d`8
znd<gNkjp<#Yo%vn(DmVKAn88Wao4|kCO<b@v3f#li@RImG)CR2*&?tUl%Y4!sX<*H
zlF%7<k_$R18zYxs!6I~N@Py7Hs?RI<q-s_`PZf%4GRFQ<PST%eyn%67)D}Uomjjvz
zqXrUEGaNL>OMM^*T{dds*GXK2DZ+Iyo1MWh98j-;rn-QvKcKV!mm5N$#_U^KqC)f2
zJuYznDMgN4s6{bj$n|sPaz_|u=kSbHOx**WLf=pm6qH{5r@H7|F6pVI1-=f3BdO+e
zr$6=RNtg5OR;bezql_z*hgrtfRaYzCtUBds=Ol7hIb?;bu^eW-@KcqBNSZw5R%2D?
z7$Gwua`=45f?PIJp%vzX<2eLj)N5WyL2~TzDv0X?_qSM|<k<6#t?lc#{cmn)-|FiJ
zaikRs;ilLWCN$cDhtUJiS}@!`gI*e7Ah;(BbhG>%FRF@ltgHTY9hOFmi3SCat6evk
z^)MG@jqNnsz5)d*MHfTUKZV0cBnj)X@Be0Di4VX6rLj7J6y>P-<x}bV<MSFF1|6!A
z9M)I~iu3#diALX1YN3{lq7B-PDD4a=wTcd2@7<?hIdo%DNQr88(jd$!jP7jvHdD~x
zxTU#D*Y1|xtO9MfqtBTA5Pfkyc0w}+BQk|_Nxja?FoGa}IzxKS&m&7k!Y(UrOE>gp
zu4d~+`iz(RzPNL{L3YECTDjA*p~RvFTi@<<l}Ib?r&HT49GV@`N^E%oP)bz#%RNiA
zgpHC?dbcG#eP*7)&X({e+0mqj=B=_M4``j5&hOGsws9){k7xX2C75W^I=#!8m(MPA
zB+!T$PM(yaKz^?=+uy?5KP58%9mGlSpzepQc;KZJS&W*!OQ{n}V_@j$(Tz2Yk<gbR
z;=;-6H`x(*l3}BJ8uN^znu-<W$6bSJYz@`%$Y`DPS$~@Pn}X$XxCy$e-X}nf*C(x8
zJb>~|w7ck*OR)UW2(Io^8-6M0Psy|c*NdY6SbBJTb6|#n-Ex&E$Sc>}i5r|5FB=xT
z!ya#gVJwktgzTewCZy1&mszJ@$=-)@{A8h%vt}izlYR<__%o>ZrY~w};~a6`R*w?B
zipoZ%t1CTeZdS-L-!yFbi6)VnXQdv$1DJF!RA2MKX`+1ZML?9eoLQMRdFt@}n?#jY
z83{TiCb1Zf7yKBoI|Me$)v#es?IdBa1SJY4=z_M_`xLHkwuQtUPwop86-<{>x#_Ge
z@*CPPVQ%T)sWgHv3`G1C6<O3I#)&GfFMGx8@2;M3l+t*lB_yD3FErkTTXTX8(P0U|
zosbpFW9JtZuHCJz<ju^qCpV&48%Ge`P|}@AM==|<!N)UeqD976b^Z-&JeaT9hP6J<
zFWUB(c>6Y*90V#%<OdkF_kCN!zuZG9^ktOnI?S!EVHjN{5gB-l@v_b|3llg%4$G4U
zN{=+JymrO!E^>G7nHM4Ybg$6SgFRJsTduaFU`s6UAe@Ncae4eon`C=y6MlaocQEGS
z(nTz3`C`@<D?k713LvA&<A*7&d@s-b9hZ_B8H|bguC4y?6+!pXkHq$;AI0t8%vRNj
zJMas}l~kEqE#MbjiG$c^F#epDuo2zqcRonFXjCUbH->5EAd_Io%utjR@5Skc@iJV<
z$WSeu?N|)4_v%XM>ll3(MKyB+4N6ergC$cbEs9W=d%pFcx{^FSsFYD=w+MQ>Z!z+$
z)bUJilOM!<*tjeU8<p*Zk&>T8Wyo<iH~#ZS6=#m*4HJ;Kq#{jV)~*Xfv78VZZ#FJB
zAc<SfpM#w`VW-cfT2-(tm>^j|<AO%7dg_2p91{wRr*c!G$>3E-cYC+eA>45-5j!XM
zN#3*BnJGIov9IX4THq1&ncH*XCID(LZh`ZEK41dyvVXY^kCSm~rjV3iYg8u(_FD|-
zREbRyZ;E~gp>)FO9JgCszMtIQ5iI9YZ27sXT^5@R{YOUlpA(rkwzpU@a*d^7GvcYt
zGgtKfA6Q2suuk+oof@9RqVut-@bjG%0#A!oJ25lznvP&}TXr%{z?uN5f8gl_mi6&`
zC>91!p<C?19fSu1Nlpybe-0B8R7(CJjp45Ri>Wyyi`HjvT6T#AaB!Rz?sgUOLx%`4
zOld~5x~sDrj{>+%2Ia=&xW6j7DVv*9y6%j>G%_-3_Sxk?I<?_saFQyBDs~;^T{)N3
zvnkcXtx^8S5UW1i3k}-X1r;P<Q&47^f%56Wp<f564ymff6a^B&r^!4@YMCb>5%adD
zZ53Y#(2XTNWC8DzX@5jY+;J07n|(Xg(9+_&BvXV^%N0{5-hNY8)3yde@NBD2V6pnH
z9bHy&<xqb=Y=-DxWCaC<{e@OsR8o<N#`gH-qor0{y{7k(_4QmbWP|LIx1h<A;2)ba
z;Cw*^S6TBE`p%IX7#_&*s@g1N1Tm-4^dt7At>34V!aZ9oqKJSX*xIHd<&^Mo!uepA
zk_xbf!WZ2$+3+p3rz2jh2-~Jop)>hi66x1q{kp=wePOY_{q#-AfAwTjNX;ZCk`GB;
zp0a>VWIK$g$VzZ7xFYR%r~3oHzM?rHDk|!w0mWC1&js0jzinf4e+K$V87LxRdWtqP
zW6iMALAe!L_T%x@Fk_f$m~5E9Q08I0?L6kf6v9pG0D3-v@5x{OASxW7kVc&4*IVCP
zQ_@Z3Noh1Wm|5WsXs_cYS&ZoV>6Y{wMkk>E#@$Y#@IfN*#Q(`OdCa#4PZ*v4QwBM6
z_GiK6BkhdW^qVeT8JDjPNGZw(<#y7bJ6~JJs~sOCCjyN)_)jl#`Qr^AWwdL}-<hNR
z>HqKoT(l=IBhM|5l7mlD`pbwm*6FLdj3+^6=<6?M$`74N1?@K6C*pNi*mE?DRwUBu
zfgCx?O#?GMX#t10wY<{#p`0Op4>x@4S!MpTBV$(JqWXt-SS*JlU>jvcb`CKr!j;-A
z8eYUO;!tZIwZ$li#bsy09dyMh)~O;9Z2lF*QDnTGAUFjfT2B1O*|PHmWQ5M8t9Ndn
z3$eiA^a~{(pp+t)YoEmUZmw%hpkQK1jnz;QRRxFh;or>@d8iP}_9Ja#8fIz*qPbn6
zIM>!u;~M1o%4*t8#J9_?ms3qJ@Bq4WmGMdwVN(<xlE&9*xvy1s)$gx<rgtY04|k8R
zmj{KW@jBWZRLRrMmRP8k9l-_#{&lVJ_AiXU*%Xa`HWsl3DeKY7yqU?PUW!H$jS6-u
zQHSqzhp~Db#2mn8WB;U5f0Vxds~gR)0K?R<a75~jIc4v|8D0nut14pVT+6qhzsr+%
z7GT}Nu{zX~6w|plo9<Z(wK!IanP|WNg`u_J`~4OpN2WKm(sFw(pivTwdMF=ysi1!{
zNsb<38h|PrZME56J2C0#x@#6`p^lHMb3I-1rOf|8Q!hNM1V7!!_at6UE9@O6Oo|>V
z(Xdno=hv?_zO|ObuT57~T0JC-qB5MfbHqeo&d)e+^P(P`DFr=`$+Ro<Dy9}&4fsc>
zMo}sImSY$KrW$N@eC|(V_kAhYwJQwhb<dW2d-1?Bm9?`gpKk#)gLj8vUuh_^)u|hs
za9NE4LCKE1k(ad>tlPV2Vec#X#AUEI;w#{zii<P-Fxja#wb^KGX)*gdRibE@4pf<K
zJU&%)%_I&1(+>9of#lT03HrM(G!Zq!v8fV`)$Q#8qo0E#-+^WKWIZ_dvW;kOwuZ9B
z<0+n-nCl&znb|1^!`Yq#eCO@;waI)vJsSr{KX9y9JFev}(%WiQMUl$VLmjueKLi8Z
zv*g}p1iml<DDr^_@U6VQW-*S0``raRRYBN{iFIP9{gH$<<{_tb;*)=1$<<$II;^}*
zNTrj1Fj#7M<)K3Vg-3ruF#BK}9r&!JWloV~oWFE_nb>|)bn;bgc#$aIsn-9Hb@1o#
zh#^nA${f?V;m3oMYFWNoi=24~{;f-g)KJ`xfE!MPd<xm$OM9ZHhE)p&=Ef8{Ocmw1
zZh)BBv-~=@nrGy;x^Wc0klNe)CSJd~s>1;a4oL1R^yZ&9WP3q`EYU(6KPT3B&;;CR
zdF2Pv;uDp=f4dGk?XYZ`?7SX}MqqrC{^TQ<Cp-17;uCVLK}Mk~V7u*i>m}t|Sk*UF
z*^+COx(|iE8t@9g61l4+2T7@z#PlKP#@F^(p9+!5Y87Zn*RBr*Z8SnTNq0tt(<B`4
zN(J8J9L@I%(8rtBduW7G#VUAl<daLLP@3g&s66yHn(2zOVocP$@1nmDUMxWjbtD&T
zet@maLLDPs2h`}cVZm(M6)!t`+$iDRAzzP#N-af#UxWZ(;pEO5$qL8Ffnu#x&to`(
zc4gaIsfGUb2j*@GEB=Yh<<;l1W<rWi8))<vdjTEku(<EPIisB{W|$3G@>J(pbZene
zP%Z)12ocDEhIB5+;{4Br!XJr1X#xdYK~o**LS%6p{LcZMB&P=b`g2`2-Q3r4v`@yZ
z;h@hU9zJ1CeH~I6+#$MZFifh4Qd)W9!ATct-W<YMQPcbNxt=@ElY4$6E32OFX_QTy
zX){{Bp-fS)&>n>|D8BCBfaxpXrW_CoxBYYd&7c-d|KZ>5=wW8PnwdII&Klr(Q{4(L
z<=(qvwpGA!IU=Dn;E4Nb0yzr(Xvr)lZdD5&Tb*dEuz7eb4s14M0iUz{!#;)5hPL@4
zIjKB~Ys+&)D2=AX@Sw}%K<s80Kf##h+5q>|Wrl{2-p-y*%Y=+@2Q0duor6$Y^j(p1
zEixR1Gy}JN)SvA!T>xMNJ*)DGkn(lkR-bC?Jk19*ANwD3NIK;nuFO0!z)W+IEBO_A
zhd`Oci@QN)l=J%sHXlthFzyUEI2GK40`P`DBt-6cyKmfV$PV79%nEYe!2DRncs{2h
zrLW|BSiNdZ=tl2W_wl8cT7$}`xGtG$&dxu{ijh%T@9Ne9LrVcSFqX;`A%t7K>5Wts
zLPe7FQiqbyhAz)YUn5=Qn<3>zIfQ$dw2Y4xr@e;xS6Z@4Su=bf=Rze@2;uQDe7ipk
zlh)~aoIJm<fSvsuxMMk)><_RYZvpI(LQYO7Bk;F{Ywgo&fMjSygvm=wT3TAoN}`WI
zXX+jvR(D30M5AZ<`7@c{mWGK5tH|bASDGvuBV%Dfq3Q2;ymFKUopx2rw3+b;3C+KM
zJi;r<vjKt1y2kD=cJ|p0UqMDrOEbGb1}&(QE;!Qa2$bRb943-s=-c%()>CFwS#dwa
z%TB)sf;3)-e$f3bCi$Eq=_BQY%)Mbi<p7!p`i6R`CjI?m>#0#Hzb#IsQLI11!psaV
zr={Z95+$ME%E!<mmDq<Y%eU_CJVvEEA!_En4*(_pd9Kc?yRQ%1$>{`7#HU%0ze1$l
zJ5Os*8(DT3<D1i&Wk(Ys#LxHfTW%ZqK~1N}(<qnDpsYnoitdeNqbk1;LznyB7PDfi
zknTdo=rS+JJ$XTnqKW!C>c;DC*KRoBJ4w1lP>Ftc<v9P~VY`do5p4Hk&NR@WHGcqY
za7IzF^O+Mc3y^1~{y*~GGN`Vt+ZGK3g1bv_Te!PB1W#}a794^*1b26L4;owp0fG}8
zg1fu>n<P8?+vhv?ym!v4Ten`-tK!EDHOZQ5%s%=UqxIHW_j6X81#GU)5{QQANK}Iu
zrvMnLXUo0SYC&iTrVns5LbXMhRS&U^oW;SRy;CelAYg??@Wo?pM+EYRIypYvs%qoj
z=#(5je4#CGUNh0=_MAJqUshAd$TeUjNE<`dn$HleZ(!_QA;D^ZgIE)ACP7HVObZ4;
zPG7qhUe%`F`kZS81u=eM)HA;^7{w9glbUdLaY(t&>AQ=cxvQ@bTRH6ul$7fqBxeBm
zt~m^+a?R4w$6l-};?v`M_)VZg21|}aTpRZnUeT4w&hEIPUblI%1n2>jlL%CD>CWLk
zR5siM12WjzWRa_6tMKe_{x#6cAZENG-mwd~l?E?TTpUvd#hyj*$z3;Zzs+ubkmL*P
z;U*Bx#<=H&T`(&(3nsu$s~YWdPCsLy;Zo)Q3WHJFXz-l1uu62ysystap{;U>_9Wf!
z0o*dkD*>EkKnc3!pieqEdNgSdGfm|uo;6^xs0Mhaf+wgHShP+L`>(AGVaKu3d!NRq
zp}hMJ$0Q$Xd-Xk)M9G8>eCKv|(md^_8x%_9nmP<KGs*on&^Zv^rrkB+i%7jS-nyI+
zH}e|`LWnYn^3jZXCnbEfY!3nJdH!^kto32?!q=~Jk_Eg7%$w}{nDKeae^jjgwpaK|
zH@F*+{46B50c7EhS<fr`D?C+>sOw@PpK?<b*rZ1`gAfgISV-SDJe%ndal7)jG~ZR|
z-tN7zjPKgKIt<YD4XCVCgrM&h*MOn))b0acQ4$A3MfIsYdv{+&YXp##RtIm{!m<?E
zfD*=yOT<`m#Y5|x1E9}*WE=lzn_Y-J9N@#RYId;QTP?SZUR6Ra;MHT+?}W`4(V8DF
znaz%Y3z6$+>`&oelHc`x<p}4k5g7hVH3l5Abn?8CcEsYM$>Y!qan_QQiC}YJ>a)rN
z?xBhm<{a3m)Q`whhQW%8GaM2(=YWN+$x9ILI`y#Sy$EJD;H5_d0YxL15mTmB3mb??
zFvbpvwwwXbbhWh6DSNxcb9MkW2vqzPiDru-j2ylye}^2<k)6^0_&tQOa+E_#82|X@
zU~(uNR@@s_Y#&KIevoFuWQL>T@RN%S#495oGKwOoV*v6uLGZy0qJS5!2xN|tpgLkR
z`4Qz_%N3MGQXSe054x+<_3EPRDs9cLkbeiIRc8U(-`}4ZSpR<1{*)1KH{RW4wEFJu
zHJ>Xy@HDaba)j-$k6CQ;8O;9CSS8|3Yyv|W2(XT1<P@uSE2L2|F-|GEwR}4t22<F4
zqgj-I`d{2lPTw|v{#|@fkM*4wO(T->9WGEpve{7(`?C0;R!|m)qLO`gRb<q&Sx2V`
ze$iy_E1%vJ^Kg{Og>W{OkK<ASc!fm-Jf;<HnWc7!khcq<6x-fj)mGP)7AJ#%y<s<V
zN1gSOH-K>T?W!WT_v5OGlcJVWfSFex2inY33pduN$N$VQnG?(AL&nGm5)dVi^jo<A
z!VLdJ>SM7{VM}k<l9!gy)nmv=iLY0{>aEz223XwW$;PsH*r(`fzzJpLQh6amgn4;l
zCT}gVR)JXI;fs<qs}^8BH6%fKaAkoAVigXmhHtto&U~X&Tpk!yR94E6r9Ev%Woa-B
zUCM{I5xf4gT>hT_=8MM_?s0M4lks_-E(6Jaxve-_xCnWLh6lP|i396Xx(LL;GtYm8
zq7Gm&WjowJHRe1Rv)K;J2Tx0rp~lna(o$6eeGL>a(ToNZRH8~=-|b=2XL^1f-3VyV
z?fc#;!5$dix~ec3K7cawl(`r?-+FoD)bWw7ln8{;bqc0#*f4+MU`c5RV{Ftay42<&
zv{VMTy8&37H&AFo1Bv;!Jf4LBOz@SgqSyi1*FaIiWC;U6F-<V3wb$M@Bl3cmejOfq
znNLf1a@1ZC{kRH=N=iCwde)1V?F>5E_&zp;NZ{@L@sci*x!kI<lKxw@s!l;6DH8>(
ze0d<z{wid6hXQa46520xvL-EZ&}>je^h?fXs~&rc;&4Q7MaT~UtF&co-;`-j+EOz8
zDLfR82Sqv=`~)T1+5HAwGn9e_B3hV}JMqd_vpXb<dwU8og4X$on=&xH&FgZTn5R{~
zD~5giJLS(FdaI+oHPsaj(@e-(OQ@|=3^mi*^DZ(+)SJtbZ!&qUd*`$J4nQ}Q@Ew(6
zRBBcecsC^ZBn^$qK~!e#Aapa`aurJo9gDU79MJ$k+hzg}3;(Lt6A5vn9~3PH*h>9q
zK@U1NM}~jya?Zn@9oGnHs{`P6?!2p+ex$p@bSjsTm+d7eR#sr~LL_}gY3X`q+0bSm
z3Mfbqr8qn3wE=+Uf*C*MvfQ_o?-c?8gtfJ`R+Br<n?zbYO7D%N#$+yT$K9aYi@nap
z#pv{zc7WtwXSLAp(|Wc!Wb_RgMwsjqMebnmvUvFNmMU0Yv7X^O)IJ6!HI-NCr-f}J
znWaz?@M3vCnj~ogM27(UBit5HLeg7?r%gYR5c7jY5(W~Jt#hMAQj{TrfWuuj6CRtH
zAHbJk)K&PAPRqrGBN0K+F-50|<3b+P8=C<aFFG(HS+VjBNdRbjLk>uis>9Ga00C0x
z9ppvut}7rw>f!Kc*TZ4saXfrs#Jgj59%eFLfZ&72{z4l~kJN;GT>@zDRDpmy@jZ$d
z7WLy_oY)@nM24f#;<p8rN8q;7)(Vao@ie81xv%hFr>p(YX0#NAy|D(|MrlRN7k)a-
zv`T1LsI{m8X`GcF<@_-`wWHP6NNsjDJ1?4v`K{71?>_N>THIT+Xl4Z(R+C<66zbH3
zbKX}-BWXRkZ?F{h1Qc)YgU~esvjw(>-Xsc5UM(6Ycbz{Ji%NAbVK*Uz?4J<vuV1@a
zAz?Er#1zH-l!QDfEx(E?xLkaU$g1Wxw3mK5_vY2ca~)coh1K<{pkjJslv?B-TY}>Q
zjNn55&m1`xU1#Un&cV*O`04CH6xcArSr;_ZV)gAa13QIRp=nvI$&F0A>=iFEL*zb3
za(=xl0NP5{6_3~OQcs@?;%M3_d;oj4iK=LJo?#sO`fV=a=IgwY%`pibd2D=nlM9z{
z3?t)28*&fG(jcH|2wFs<;7oXr0(@h3+z@}!1&t|M%4DvVx@6*1)XbJ5L6%3W$Y`SF
z0bijp0BXMVK+q4(15x3oE=b}8y7gSN>~kVD!N&b-#o7$4d0xx+R_d?}Uu>h0Q$yT1
za#3vAwV3agfj*PlDMk2%FfBZ8GnIl~ooDa-blf{1xtiyeLxno_#{hY~@EOVt=6n+g
zQ0mk(%W#sE0*(M4LxZvk^)4>0hpJVMKN8}%P#%e%+C7x7prWbr5C8#h*f+0*v0~b!
z>?1y0>qnoP{Yuo=RyiVa^S>$sf{M?MJUODe?Ts$tAu{*Rt^`H;i!$+&FwO2&NxCSy
z{6OLKxSVMD7loy;av+&ekK}M$@UVpj`*?2|8XFf(Xgo-=2LRBqU7h(e2%w?Kn~Ek@
zR=YzTaD(rZDI1&(eyQNI0zPF)#HFP98;v7kP#3Gt9791sykG4C`mt_B5q~=!5~Sa+
zJwEw$!h7-Tw6ofb^tTnrJ0|O%;5hw5!8Z4ZuRLLb%~wBHOEYw<2IGJh@ga$1&Ve37
z3<pERg0obim~{Fn{J=RtP%>!KVw9tvn*VKg2!=X{uU{O8*^o*<gKr!L{ZKW1c{O}V
z2!aYqCcCI3oHUCEJdWvNXd9CXICat!c=@5ZZ5V&p<JVld!WOUs^6|S5O;mZ2oxkmm
z(Al09S!PJ+?w&yHuIt$y#+!O`h%;Jg@O5wE2ntmXv_f(*fR}u#34K=<7YqZ42_b<-
za+urspBSE9o4ztayZ?B6>3-9V<$QgNA|6R#nIv>W<x0%xIIjkwT-@1TrZ`*cgMDQ#
zq_F1k>s7t@(IR3r0Z;#3I=$?1hg0&bpdlq{G(9;vnAkUvgqSJ2K?hcs<N0kN{u;i$
zXS1Q_vZ5il@?feBNq!Pf6hkJdLKq5Bdf-%<#!6RTEHnawn7s^qaqzc+IXG>BLmyaA
z3^geqo3)-5z&5@6TKZ;;3)~tDJdfsdCGa4*truvO5EU19?nrn1WWn~~!v~Y4$Cr+i
zZkPK;8j6rihtob;J$KLUOHYx|*m#M4wc%Vc(l{L)Um+1RmEWguU0I77OyAOU<Hq2!
znSY)Bp15<S-KN!2k*~(%!;k?$f(p;eW4DtCP_8OC`YRa2fZxn}yWik@>xaES0WCCp
zm7Ks9bq!T-(0l~?F6g_*yYKwcYXDze9A%j;W07j!ojF3#M)YfI+Uq8-IYHAM0C1>s
zT$IYAFvk4;X4QQ27wy`b4IEfvvP5CxXcc`>R?j14E{wtOS3I(ZG!cGkR&xU5XAkke
z7(YNFDiqrjzM?uxNMLnH^|pW&D54Yb6NA-TEp`HWlH>E8aea6Hzn|)lLy$EdIKb*A
zWrK@6`o6mU=q4%E?>v{#Km83bY4`}S2m=8dD~TeB23SH2tE>PVIj#SR6Vt@A8^*~3
z`g4H~XlM05v6*TSaRWx3IsKPO(aDFu#`4co{}=xRe7sDMo?p}d)GrG00NC#I+EY9a
z?0>Qx`-gMaKR+MnT$j!F|Hjjbrv~adi6lSkCjXP$+&@3z7d_;+7ad^45qJsU(*I-M
zt)m|V0JTcS_+LIxUs0&^A;M&20KX*duF@Y~w?O;4T&E))XZ%ZSfe2j4ldY)7T&!Ve
z@)zg#Z7}$0Wha1IYhTN)_#*}Na#0V3xO05^+pmkaPQSSbrM<6yHk{Lv5K?XnP!W2Q
z9utCVyH5Jox0C9$x`sbjzZ{15!1W9)lu#?G;il>Z(c1FUf7V+^GPGUeKewy*gF#yw
zQ$98NH={Rn4hYX=rdK^ZJpjO}{{$xh*FOGK@+6%=3v(+`0Vgp2L@piI`;hLb@*pt-
zT>&Ei=RCTnB!I=iyR`259Hf^23^*~rPX=6s@65)~XOq=A>vtkaH2n1nYOAZe07Aad
zlYBZmLiV(X5z_o`E~%iX8lb4<`F;yI31F~<P)y95TZETX;mV{lcz~r{^)x#jUnq2P
z#{;zk@95=P$>*g^Kr4LGs&;v%YK;>utcysaY$rA(&fnnq2qVnEA03{bjrXelMB)Jx
zV}g<8oY%eONgkr<w=UQ5#7^SK$S;Kc1At4Ug!gxp`j7$@jRtycgt4jlYIN%9<4^Kp
zl$;G3-vA-Y>}&<f6)ohI2(U^Y5|&M)tBs=_@GJ5rdg3;wK3MO)4_+iqt;*z}<qIuj
z&a}ywkzw>&ai5u+H2|C~fS$WyLgleYoxf)H?-c%=ou9&i*b1q50><rEuhhTeG6dQi
z>ATFQ{4%=V^sp(cQEGS=%d5kM(W<pT=ksk8z^J8cJgW&61Z*ZwY$}>8B7U8kCJU=K
z&3_T3s3pDQzeJFL_|qTm>rg=XmUPEMMVl6uP>=9*8YKC(guC_U2IjL(yS$H$gp_dR
zi$%vMSzpwiFdtd{xxW=Y?0)<9J!6Q;_*`dwTQU!NBw}J+hBn=`L1DUtn=`s~=qFTG
zTya&8OR&Oh%{?!((ihe`CSYF@`)4f+3U&zqROzc1dpQFpZn=mnvHu=ve|!u)e%Z=A
zv#|ac+X<o&AD1J8Ha9obs@%C`hGeQfu8^d0J4jQCI*=<_?9=_&f`T0VXuG7X#d)bP
z<!;%mA{MWa@!9>yJk#CrZ7m7>P%!$?vARZb=kRQZ*nTc)Oi0aO5-sJKcsDE<kHj0b
z1a<=m>37aK!{!XqB>BLBPob&=FQ}`L5Z8CRm#hVV`sB*<QQnP&K3f&Rh4<*T>yA2G
z22dz%yR-lxbR4xZ>~C*qYyP(q#uBH9|CSno^#sTrr3<k~tAFoY{v;jX<vRIGnvOTI
zxn3OlfBu{Zs56k5Sy&u|->CwPymS##7ZJ(BrAgMcI1vvTh|Sx|J@z_ETVA5Bja8am
zR?YJR8SIyzkMkZf74Y;%-vzsR@|Jt@q1i04`l<H-0{0-TjG9ZQSS%t5i?#;Tzz<Pa
zF4OHF#6`O8{Esk5-+KY$Gw${i$bWw$3;>$0)N)U*LkFb1=W_SIzmYG-BR=)@9K&f`
zp+FC(y|99EQCL~QTd+J*WXCP8x(7$askD?^U@5?KZ$Xh*E)&V33PvAmyywXLX?EmP
z8t@610RBtuncS+&yJ|8|f1$i*1`w%u*!#2aTN_pae?Hee_wQKGPDOuZ|4;nTEpTEw
z7mc2asyN=AOGv<8tIh5C$)3B7y9mbHIC^{KQUD4X@BHxB(ztjlfU`l2LFU?Qt^`2E
z4#8p?y+Qn7XaT?Eil9&@jojbR(AhNsua09HgoO0{8{8JM!<OS?C`b{!t#<cop_Ru>
zLjySZqmkVrp9`vdxvt1D7N26B8Jnn|&i7)6c2KyHnQ3q&g{(n^_ODV=dQ@8JSr>@$
zOSSk6(8vzZO!ZtmtFzo60zO`ry$EGCQnNgmxP}m2X!LKUj3lL@ldZO%%_@O@Jf&$F
z2D+0I27a5VK?hH%6P+zlPJY!@nm8MT;1;xZce&dA1-oqcxdsSm^gCOX$m1OC!5<!(
zSzb!X=OWvXLg*zjtV|9h;(i~UqtsAU3&RBrI!_&TJL;`G;v5}2+bf>L^zG3P(>Vir
zH91DFOBaue@YR5+s8SfYGR=t1gmk#JeM(;GXfL{fS+=ayf%G9^s#iu<ysIp=!R0++
z^%zD9#!WK7*Xf`93>jFxnDvnPz#O{3ATM0Ly5fIkfn>&<so|hdq3HxT0_g9}5UR6W
zed^!GdB3&`8>{p|Qr^@1no|BF(8PmGOag=;mW(Xdwy~na^df@4t<v_RAc%#fj$<4Z
z6~*HT@i3FiKMZRO>`VE$@wbL|Aj`z0yT@6C_O;B?MDn?D4fnkXUsb@Wz{bHhI2i{h
zhejczbhzW#tTj!!K#)i{J0_`XX6xsEIEIGz%P)Hkc!osd5(f4H#ofCZ#lDjuL1Ger
z3!vW@r;{8pW<@B5pXAMUDfe;ly5G#U;pb#GW~U}*N6boUgC0-$Fj+BIvqd}o?HWf$
zcBW2%#ci`+q1osG^l7hvMJ!sFfucV&dawV%2McBgIMYF>%JF|F9LOJ>$s=>8Y0T0R
zlW1_A$pZL`Vc~B`NDs(Y4c?wCtH2A72t=c2IG$8?0FDn1-tYqjuf_Y}7MQ)(<pe1Y
zN;!^I%pbRQ;hfzWR1O)5h!0abHWY@JVBU51<;9`seAr;*&Fw@xUfyK!69^mPz@~F!
zrb7s}FI8EB)7f~f__HUhjGXR*JSUTE^|H3m*_no7$-!Cmqo~3{^~b9^gXs%ExKG9_
z-~F(s*x<;orKG^Sa{Px2Fh|-DF0fl)^a&7T+T3Z7rFbB}(3GI*5_r=Qoz8)BL56~$
zahq+%$B>kQv(}gETo5dhp98_th0U6<JpOr574XGtC)22j02*dhUL}lhhOCQjY*jIe
zU;A=esJmpKBwwT?8Ves`suW1R>FfG7H}Y}wE1%&9Nn_m~Go23wP0(EWXPq7lPP+3A
z*xnCbn|yHHB!^p({9VNlPG_%o_l{D(F3#fGFoCWuBhIgAWM4w(w{LrlS)<%3&lduc
z0f5T#S^Y~u{MFB2jCb7@wubfF)cSw(R;79_7AkQzv;`|ZZ&Zw)sc-$w;&c;t>u=*&
zna5^00Tl=!cFH;N&{XAM1zL@f!m-Y-4J3Z_91y?1yRFrTMJ0ceGn&XJ)S(B0<Eum}
ze~e$=KF}@#Y^RM%owlK$mV2@?^TT{<C{-#%wXOxstl>mqTdUkI;aPW2*!nqnN^5h!
z?sxc~oskz(73n^WkEp*_i&f0J`|86LW4*VBcED43U6b`729ao-MZ#HNrs`<n>vX_f
zd38J?nYZ1{0YO3zqQFgjoDVB#ZTAmOF8-+z^gh>(j#88Rg@mC-E<?MIor}L(LmauN
z`091Z1~8qY-F1nSwufW|+#^d47A5ivV9OAc;bxoh>LaN`u~bbLFwYqT+~VW?0ow|e
z80kp>!$QWCb-VEQO?0i)Okz3xKqXh9o61l{h796BBJ2`SrOyR~-<4DbxpjwWsuh(t
z-&bN17aHw=jyO}{t<&vZUR~YYd+xpfU(<9do7oI9Q3Md}W6E$w@fd$+FwYjwg1dW|
zU%BJ-IOe}g_E}nzH5u~cZ&`a)o(?y1KuV#M!5b!Wh6olj(t!u)gS|pog(k~%J)9pY
zQ_5gI<@wp)H|~Pd&{5PmMN8V)gAl0mKj_5^*l>Eyd$tU^=AUIm&w29?<JOP}=eG2W
zu;cEno=9+3T}Aw?oM<=tP95eqY6w$hI#a!xhFe7*MI?%^u93=&D$huLCQ6=d#&M#E
z%l_Ngg_IQGpT;gHHb+e{!h2+uZJr&VmVT9|VG(nXG>QM_(PBF>fWrW_reF*fM02tq
z|Jmo|$0McfL~jEbN#|;nY4Qi3Mj&W4&H$Iz<)fp(-rhoNc8jEW&47Q8*uRX1SDQh9
zZ6f*zsm{iXe&|i8VJ9aM(Xe6EM-qz7$9`<-myrOyp_yrF`c?|0WFif;XniB=NJkNg
z%1;82UPqaH{xg+~h6VZAHNG$Vr_EnL9obqC*GCX^UbMPR0Gs1WK8U(^5Lf7h%aKHM
zOkh)p_Gazm{hZAtTgvP_-8xA)AT}Nkh^K&X^u3_4uYkNb3Lg&+aGc|Gju+|fk)6S%
zA~iZnQ#YeL>;@*o7t;qPIeanqYXQdE>HzBlfbEh6{Ym1Fr%P{{5;^Nl34?STt@m_=
z?eX?1m*_&PZ&)0<<LOeHh5EaqCFEIvQGYfrQ_BRuf|gvbR%L=Yap?h3XVE<+owuC6
z_B|>E8OhuY>YddrtRI%qM+MCI0t<4Fs{p~gmciby17Y?FD0W+2OnE)so&6msfU^f+
z@R3o*WH_Q5lZ+(lSVzDK;Yg72c_DU>IH<wp<xcPkBZI-qE9v+Z?Jt;caXtKq-&O!`
zBNiw?^RjWEsaR@<mC%jid4L;}31S}UZ)oZ$HFLO>Q6N`>mTzLn<a7U2<mRN-(?4(y
zpQNsy6BbKzkH_ZYA5J&de7E`VnJu*}E59}6;wAv)vtith9pIv5NY?(|vuvUU$6^6v
z!|*Vj3zB4jHJR}V7+k%h4FpXN4qiAnzwKtH6J|P$xNPCG85dXmv&ax;rV@6?L2N*|
zDm)g7wLMp8?R^-^E$a?Q?}BEE4Dsj6dY8{moP^6L_^+?nDjg<WYHdoVuY?EWV0{$u
z!;Ux#$Ui>~NV>QmaXo64QQeUvhK!lIaw{_=A<vNe<c5`T0f)-f+wPCzV-3!6Lamz^
zw!X`ISQ+9nqA@>RsW|tx&B6ogGfA0g5n*ghdp(n1?u~u;cvdYI9<XQu4_i|4c++6`
z%9oL9hsK>nTIg6YKzO^G60);4D?Nx88QP)$D^X04Ok!AEhDa(S?{;%<{CFga;}zG~
z+b?R%3&W$ZZSiR<c*|k_Xo%?400&;1e8O^bU@9{@#h=FbNoYbKIBO7|goEmisz^?K
zz6Ia=8_#6l$S;?X=8(?<2a#~!34DUUNj3pU)=w8h^^zdolH_#xVy`8P+#4eM%33?v
z(3O9KhR3%DJ*=|PgDochQ_lYDYNPQ;B65)LgT-IcD|=mq9vJPt{8&B75P9G5R`Ol;
z3rHisXcUmxdR*M-BI>aDgryEaN7;iq3l|QkWDO&)Mm3&0!{p9JnH?PkU({X_#|U}S
zZ2#Nk@tCPZ6h%LJ|IuR3eyP>cM4*u=OF`)AUCy}KdT~R0edfzuR^|O*IvVs5?$WUW
z=L$`S%d^pTtEw<;?K0z$4CkX($IqI88)$S)&K-8saNpc-)xa+}^%=WpN&TfCxQ+k4
zGmg9W&N)?f;NR`UeXZVxJwm`6wJ&ItUo@B)SNqGFgi6r;V~19sV=a$Y7%YO1`g@1z
z8mV5%KAIZ?Gd=H!HOdaa%?T)@ayGn2++MGL-}9j$tSEb#N?i{HB*))oGV4W@y7oBB
z($KOdU)#AP8dmd~In4-lp2q<daHUZ75e1rdzJL7D=@J%&5(Z95cZ=}eDB8r@DwQ!H
zUI7gcg$*zZL3&F=1MD1EwnMtZw3OIL+Jj`Wl?+M>D&!3pb#PN5*(`iv*>JacWj?s#
zUVhFfvunTSPKu`=I?iJ@7PWe<!76X+70=8e(X2;m_oWyhiiqqZ06rSE1wG#lp6+$y
z+~pr1->;keco5v17dSGm9e}l%d?jRw4o2KxofYF$qEct(0O2+2+v;vRRFM@1uM9Mo
zdbdbJybF?hs0&&6rcY`+@!QblZe;d5mZ_9fSXc~mll&qcf6T%-u=`%Urx-FiFY*GE
zU?l88lEk46C?!-^l++wzk5)Ge?-Ptf^Ck~73rNXTn($*81@wfjFpS2KrSi?*1%yN6
zfizoPF+P`-*&O_28%)ZUv%<B18~Q5ug)`v6Gegt<XzLUIrn-h%?CGRT?R^ILCE%p5
z#;O=yCJeSO$7l8>GjW8|XT6IzGgQ7;paI8Mdyj!8U5V1$wZELG*`1j-aUVq&&!_{L
zD;L!41HO@$ELit!)p3>A7m^~0QZeug4=r^r=p$F!Pyc*;aa;gYM`$E3TrKzMk_fxj
zeTlC^&DHVT*|V~J-iOQeY{p|@C+C<=v2W6*20<fKMJrs_8)z_thgAm$Q7K};TaPj!
zvpv%I7D&6okU@KF(qU&xU|$f<#X!!ED^*w8?-JKdQ`zYq$2l@GQ+vNQr>{a)H~b~y
z2T+4GsTI9{Hw`JQX^NH39A%4X1(En?AiV6U*Y{MWdmyJx3d-;_&|nfK-Cwk`xQmRM
zbQF5`Dohf+zbnVMV-sVGJXqCo4SV&0LKYX``hZy!6^|DXpEuna=MZnM82lTo*mC!K
z;yl9-z>yM|IH-x+H+J(>wSpz=Nm6h8f*`AZyKG#bjh_=h2J*<K^nh~#CL2b{1U~T<
znuHEb?wy`|wU-=HvM+5?J4uvM+!yc;!MtfXTL4~Ic>X6d&zWFTDTUcQz%c+ML`I>+
zYQ(Mps(rN}VEjO3u-q7O4KnIdxO_RAf8e#+F+2WqKI^;<Fg6_bLk(Y|Aj#K{Ehrjo
zBh}ofiXTAnkZ-jJ$F>8t<LFxh25KajLGKl6pM4GMt@y+_tqAypi{S~3a_w;il6SJB
zbf9+vOsXB1A|c`(-2e_xp_gvyl#EPX6v4ou`Z(+EBf32U>+Wo5yi)GnPQ}jK)7J$=
zqSD7(d<u0M4&suBDf~D)!NF#B$Q^I;NM?t=1tK@;^5I1+qHP)<AijPPk2=vkI5boR
zXoaUXu20s~stt?j7cR`+;F2wcnRWCagWt!k{Ae_`<ZXHg+rd@H@&ue(XyNk2zuB6c
z=ouxzFgp$Ekmra3&JE)LSXFgdGPM=BY0*L%xmxMccz&ic2Z@j}@v2~xqPId6J6v%-
z7#)yPVaia<H6Ahtr2)1=$Dt{-27dRqb7`?*765Z1h1iYVzX*kn6%%sf*;0a4HRCZh
z2;t2s|5~?EqX{AG<!@7o@3Da=Gc6<)>u;9`Q@?~oi11jEtf%Gp6%hAi*}%)@?a#Tt
z=BQYIoq${(&df)(MoPJ*JBas1l}1VhT@<!V8QG_BFpoSIsl*g3IQVC~XS<JE_Ko=l
z%+<lx5}RSNoM?lkwgVVb^zN*A)4q_$Po)APBlwsyjapnNB!UTJ&c(!Er02hbarp@w
zfWpyrx1PHL1J#Zg0fp61B?Ze?o}|!}>pNXnm%#?77=2bx->@kvWewq?>3t@TE^+!N
zONQjatzTb}N@@yC4i3bL7uv+m4gkQ4LR8bhMDjlCZR=agk4Dpo;fV*ztrWPPmym8=
zTXDgXUeQ=Y$Y>yAPu;-ds8X}FA22uUB5lSMnL6#_2kUOJI!lZ;E~wFkVHmO+I4PJW
z(3!n0t*q={9Bf5!+#D>iE!Wyy<HLKQc8!c`E(VSsdidNrNGoB##rjR=v$_`5s0J#+
zDgm`arin%1V|XaSRr~u<r*w;@>;ST@_d}EAs+x+GJfn@uU@=>%x)g(~%GR2^LYt09
z-fzOsWJ>VSU2~z_C($Vcz0dGtUbSW=?Mq;y`ifHgvKy6tTyVm9_V^spJ4s~x;azQw
zyR$s3>}6YPAQrV*A!-pv!&&{l0rbtn)+&!N&U7hq4i1jNRF1^(R<|pH&zkka$RGQ0
zeU}Ni5BMBoGS6|#b)9_z${t3uj3CqaKPJXU0@7e&=%4Ek(pjkbCnIl9_D1?=EdXy)
z4XjcxZVIk;B3rsi-Ep68^6H`?<0Z&1N_wj?21*{GJkcaN0!E7puZC0Eo1k9FT73&b
zRPGc$)bM-H<B^$zYKU&1SUs#e`60g_+lA4QP-ly(MhF5hVc-$@ancCz(3+=o`(9Un
zUl#vp%kT~2hZOXJg24|?0C@!LA-b5&-jxBG7uE+sMaPe$>mo5Yf~)-hvS#RQ10pfR
zEn87Kkt7a2m-&NzS%r?&UTYAK!;W%dJe7NVy6+!vhSQUij=^M>8PunHJG?rJSrVtc
zGsk~JG4a93pO4J$9M7e%XFTxcrC9Bg05%gY*-shKS38bTm{2gq?LooAs#(e6m~!a@
z_E6^HtK*bjn(=e_Ls{9C414|n`&Q-hO{_yQd#tG%`m1Q+!mAzh_HuDBE!cHf3JN7e
z`ZP+Ar0&l?9KtaC?xwV*bEr;wU8R&dKxIaBfIHT}S>x<o{)Rb7-ag};?vM!CIapD?
z%6eFSj0PYtWe#2k$3H2NE5X-3pN<7OY;B5L)5j@5M(PTf=td$t`tNY90BmZa3CNaB
zSoshuwL=CLX0S`#7t1=0%obW&dS-fkHF0?s--5Euj83}_?8%jaj>){D`Pyt@w%8xD
zZ5lhz$k)V}9c6GBX62FumiS_jh!!{c^)6|}xGIJ;>)y|0rd+7itY}1C6}zLyM{_%^
zgUJy#1V*-%zXV7J`*H29xftUL#NH;TAy!Z72=6}c20+zk=xu3i7)J``-P<u>YO3l+
zX@&d{;9`0`V7zXG6@-W8tx%?;9(UpJVuvh_nN0{LxD4fl6D&0$W74^=J6)0(d}rJf
zLC~n9SD!@plcaBv7wR!Bb)y2CB7}nFYuZsaLzc_ZN&))?gG?m&fYL&Qm{{gY_gj?k
zg>G)z(hr0KlivW}$L%1p`!+^~Z`FV-e1&j5gojwZ%1uNkxs&15t8d-ouZ)ovz9)->
zD7pSwsMWRY#E;^;MAEu*HH76$L=;g#H4*@m@~~FIDMRnh+GPZS`bI3jJ;<;5w>bgD
zk-vEk>ieY#<?U!~{s-*uk4^b|jx5e@0D~83LJ<|-e0Q{<R`1!`v9<gX=tK0U{nPWN
z_TG!=x%j_l43w09i()sl+Q4_6_J+H%yT!qlt|<ZL=4M=;&Wl=0w)e>R12-m_q7x*y
zi`w{_jkH(sNS@dTWABMQ;ymnF_YCaudA!T2fP;79si(<eMZ4UzP`Gj*>N7XW4R)|3
zg*iG$XLTnw8T|)G2S6N8(<!9F$_vHmq-+|zs;fWX)uTp(qkJG70~<3WluhfmHwExO
zkpp1IN&FhIKP;LV=oxGN(=d;mVx1k%?2Z6jvu+B;N;WS%Ns^cjbIfqG+hSpUw8lwS
z{JO6yoHRf>WD(8-sAfoB6`VPX1*F8OL_5pY$km5{(UO`Dn3z_-OM2QlNaZ{+C+MTu
z`dJiSSV-x4BNmPE1?Z1%G2ftQrRrO=H<Qaw{m$`&@S+<>;QO!FQ;hYjdameXo(l$&
zq!eVbK4S||tbWajmh@F5I-dmdB%^7Cn|VpdeN8_XX>6nrm@G6(ms5FoVqKrxfU~!J
z1G+@;#OvHGkK=y4Qh9WLxMD+o;)SfQ;4>255WWl$V)rva9+SAaZ+?FX8{DhQQ%V(|
zABut^;wNKa4?8u;mtAF^UeIQ8nn40*@ZdAvm<-)W)*DMk!CBsv2<^*<cLu;73}u=!
z_t^UIDbrDa5NeYayImI*$p}zW2RRRpgaAB6HW(a>8mL4WzOb;GM2lk4hOq~^U+R8X
zRffzWSJ@iCQ`)*$XR5$EZ-8K;uq-f;2ViA>3z%s%pk!}Z+4{n7zUA?oJo0xbcaa`Q
zWT%cdB8ihgb%H_zD4{^1r_)n<AX9@jmbDR8951C`v&BE5_*{Z)s;=EOSzTg_Yx;+<
zgCaWHbVD*ExtjG}KQ@-=#Jm~qZk7F|h_@@dsjad*`4pa8I{a<juwC?eVUPQ2pu$%$
zuM6Pm=ME&8SWwptL`setTVn%;z=N>;l~hGHQzgDl3r$}Rn~g^VPwo7(zToE?i`_*V
z2yR>!?a!yLNuU7OeqWW1w9$4+of~1iCY2n8H9nV2bzTbN5+1Qqb#<lD;L{8b5*1fS
zspZlT*Vg1AwS*rsiIicq3@SDwHdM?@Y>QZ#S1`U`$-M3%G%36wyqMoJf*Xj6Lx8;^
z7KBERt{I~!W8Z3daxtH1=40VwSs3SQ{!&^$JSt>TQd*i>vRhtKf7rS>#Z~0%=%veK
z1a8yUKN!<~aHx6QjwEm|lDsSoUG9fmZ+K&;+b70>PdXVddphxSC)mi*Y;kfT-P>8f
z;Q2->l<?!rf{H?IV)FLhT2rJwG&KhI+hGa}=eEWw?MAn;7>Z5jp0Q&?_Axr8nyrrF
zr>a<6!__b?5N;+p*G;h%CihI1=jDq0yLsHgw<|O77$yX!CjN0q@sNlyXe2>sMp^Xe
zN-b{KtE2a~KW}!mPll^&t6l473L4EyD@-u}kWA{fLGa1t^39D^?Uf!X%LZq%@$LDJ
z?EszEJ286grgsT+ZnJ%>>X<N5FaG>|B@q22+}~?y0Sls-ix3WxQXrT`W&sx(MfpNZ
z!7ul4+XKKv2bTv{T(#F|{r*2X_wJjHPL$EQ;DqLpHOpI3(Kf;~pYp7{osp0|!^V>9
z7xFW{4|u#?!RA`&FV&uYa!dZ~wZ!N6Nj*dk8kAyQkz%R2G1NY9eL3~=Cd~wGfa9Wp
zD9vXG3EII^-;}64P<~nm@dBY)xWDFGL=;!Ewi~2@SdKMGt5}FKvC*oK-9{G6h)Q?s
zQ&+CiV|vwI!b`cQXc_()2h>GVVsff!9og{Hu`&)kN6ThmSgjPghjwE!YHAI@irAp>
zh<|TwYB*_bD+`tDFid?+VUFUcy-p5UOY0~E3T&jR1IFCvj9x&?GD&~U^ZQk;Redc@
zo=FvJ6Sv#Du=oUuqZyQj6^6%;euf+Q&XD9+hKw41=p$Pp@<Gv3ET=|`Rk7rJETe1*
zI+~Q3W!H;x9E>gOpS9fly~X-R1=2uvhy;bJE!Wi)vrm;U%w8Sv-Drn<yk$~p$OgQ4
z=A3j&3W;C5SmZ+}4TCp5>4=%c>D8S-i;9aib(g?Q2od9oi#6U~UtC^FJ!N^~ZjVJh
zczy-D8<<pW-gLTJdiwNPtC898O230l5d80-+|(ETanW;a=t?qmul><uv)l~_QhC)Q
z+P3R=fj&-WA?{Zh&GPrHkvuxbP&>6xjheXB3oqkS8t;e&1XO#5GuB|)y~b<%AtQ}p
z(JlD7*;R&bV+T?;_BJZ|>yKIvlwYlWy*H_!Sq0@0vG+tG1Gi{6x_=Spci)9!QNHS5
z^NY%H5;LTqarX;?KUCoyr4Zt4In4Qps)X3Stg2KqE%Vf|$?|1A4lm4ih-Z>x9~RVz
zM%l-N^(3dLQ@gKQeUgjhV@1Iu-2YgZ33`jjFWU6;=eoZb$MxrLRqf4Mwmso*+XdVX
z2=99|xi5OeV&bAfnyOyJnY3(M?K)`q>G14?pT%G^PDZ@iB&4OK4S|Lqwg<I6<#Rq(
zl`pW&3cu=&R_=Z)v;A=&Y-QzuD4>^u&*~N9e!eNrb>09fscB!htgB&lXo$B^l^oTt
zwb0NcAIQ)#Qrp{3l5Dg%fkDN=A>PQps6MY~<6v9C+{@XERM}`b>BLn(OKy%Lq|OlE
z*wr>{Jzw9dzTr7j(Q4aGj>4yGZ`qaXD7Z*Km^@uw%j9~LwCdOd*v?x7j}xbSa<_eL
z%mzzL{Uv?Yl{aTdQNb*yaYjQPaJpBp?7BFwghM{GLc$fhRXjx!hvv#B199ch+8iJ)
z*^wj`hLd-P)n>)zISgZbK+1B(pkjb)FlUz1s9TkGyyh`sRLPIS!zB^HXIbzSa^ag(
z8|e_jfl4TR>OCALZIs>cO=AfXL_FZQ?H1{WfUe>PA^GXg4-pkOx{^;PCMK4mB-1QB
z5K%s-(R9qz(w^?$Gs#Jqg{NAd%H3~{ROf9;Bzb+fNG%_aC{JevGQYTg0|gS$-#s1M
zJ1hjsKUC~r_srFf=lO2-c^f~fZ7_vI`*;cLT{c5l1-e6K5;Pxq+V+kND;hb&g0Aat
z+S?>y3*-{pGIVaPwCkI@3mNYefK2ihiAU>_XR>?k%&V0sm7m}oWL-cYR}-9k7Z%4m
z9+AdiC&#a>eAOyJn9z@~*HVJ0_X1pq>MlT=7}96hG}H<M+DDehso(r+@jx~0c$(kQ
z4_z(yE45fRzIQ3DFX{In7yB``X_^73Jg-IPbbF^J>z{jjPGghPZ!~l=_SNpf-LLbw
z>eT`>?Aqe?Su`yV+fEC+1J;xYmp12|7Go+B@vPH+)=GAK(Xt;EvppY`C}!unYB|dv
z5JX0tOp=ldPYku@7RS(iJ4rix12SVlr+0~>kQ+g3*zs<znD=?NvvU}KXQgpCY47MC
zr>pB0muF0VivfAbRVf;AH%hBqk-iardj3Fo+ZB_9U%J{#)Oy!nf7&gfDdNxgzR`66
z5n>^rI}&ZD5Ubf|R5hNhsDUI~Qh?XmE4f9r?I-%v($iJj#Gyr<eX7S+c)Wq!oTj!f
zHg8b(TJq{@u5#2#(hL=Y_aIEYmRy`S_mNi@M6y@Rz|Zwjo~8Hz?yDJcXQPMJlAD<1
z*t}Et_$cPRuc$X2hd_hdOjSushtPa>pKT_$Ma~_IqXZ}`&u?S3y<v{Nkg(tbUZ<Dh
zpyae%eGp*3{^s#!+lwwf@CxPDchb1h9t|$JqU97ena8Qz-rl~~WlMK?dMM9dKRNzS
z6awM+9*Inl*l3-d8^~vt62#pCRBCCghx4G<vOOo5$Q&C%y7$r;r?YS)eVg(EC)CGD
zZ#)?q(I`p~hU7D~OZ?z-{R9_TB6y{zyc#;`+Fm|B)TCNP_ij^Re0&^&I>dnJn^KZ-
zZ%O;uJqhk)Alc@JPNxIs<3g(QMq*(314=*V<Een8v||MIH%ZA+&83Nh8Tups+bC`a
z>p0{VZgNTqcIO6UJ7K_#C^Ye{l@^svuoQTrDRflCtuuv%Ql!AL;OuGg6-D3zI}=xn
zxG1|COS<dBQ2C>>4uv(pbEf7~NJL~TT1CqTB6@|Q+vj+#G=5lyNOsvrLU)2r%xphy
z(L;6$qDMnaqbYN6P^x<5sc@_&&D(c4XR027p)6%q<$Vl#_t}nta7s@?+rG*$^0mOn
zC8PQU5=(@(mlz{Y7uI?2cUj}NHs-7^@RD?BYChh=bDbt)H7&v*zke@QfadUZd1gtQ
zB3Z1vq-FZiz0h?Uhn&(c4(R9JhoP!f9-`|!E<9UlvlbPOWbvruIS2C+Y;kU`d4r*8
zXu>Z~?Pg(!UI@HT^~XnG6;E#Gm+X-pR<w%Exc*Xa&JctmaJP`%5B?G+b!KMpGX8vI
z11HM8clodlW%u+>GeQ-wYmR+C#*3f8*$(G$ASZ_;uc`C;E+r0^p}p-x+s`hYE9J(P
zCVR%^rr41{1e>mWn_7=p-mBiimtfr&SKb3vF;cVDC#ll)bGfaacH>p3uU~`09V}1J
zUlK}7Xx3{IKwCLOy!yihphMP*;`OJX6h)@&7hLF(>rePXpH{kivvkEi05ve(%onJ`
z+KD7Wp0^$@6>rDDqcsWXAiVIIOf9YB6kduUM*}4C#mv-v2Q?^3@=2#jW6H~Dam@MX
zW#|S};Zd&si!z-?<wcj-nTy<b*V&<F!#>o$;o%s7l_-4JG~k5K-+Z-l@%L&CT;yi{
zTCjWmTCkt85P8>VlRqvCDK*;Uau@6BD$*|;9H3^#(L#<F_QkmbO1-lPv#+b|FLhd^
zPk+@ZEUi+9@71jgUIlJ?mr{{;J}^nWDfP}iE^l+ARCX)}$E|e8!XmD2YPwNbhA;`&
zMy<``%F@)lC^jyQH`!&dDkdeB(qn<IgdPa|R--p>w97ubcr?~<ut`fh)`eW4;=HB<
zn!ccB##m70WeMM<I|j(`turcfK%6ZekHXF>Yf!gpr=w=3QO~aR^v%rPQ495$d|QAa
zhP(>+<Zuny{*LmNnA+oMNq#cG!ObpqHBDQI1PzH>?bg+8UEpedXR;Sa;(f&IB-J~2
zs<JyO>UWo^Kuhz|u1a8&BBW$zo2PpVNs^%0%stAHhOBhH`eOH;W}S#SEt`rmsEUY-
z<?ShjTujWY9T`27B222Lic2#bFgpZ)@ZRj=O{fL@xH@RI?k}fXD)mg^Do`sRbctJq
zMM0fnkwmUAldiwnr)l`G6f3XTZmro0yv^fsSFiJ{Mjz{Mw~z@08K#f^SZ_Mwigz<h
z*)5_FXawRe3tmYL_ebNLMJ5d9kX+0a$gA{kF+7fMJ)|Qmt*)(YL`x80-e28pWhP?j
zINL~m@3~RMC+30#?Ph3BsSbq7H`L!%lqWMFGe=b_KAr;dj>(boDDW%!k1zcp%#-Ao
z7V_8+b8~VG_t!1BmXBdj6OnV-SJu!B@}TImtlWpJ@%>pmZL(yN5=w8cB&;o>WS3N=
z%<Z<mvT46w5Wx%-V!(92kt+VIsZEidPCrU`VPf{NzA1&hSpoCjKH}YhNz~9@NxrRH
zdWoBbdl));{=j3Pd{i9kw9`1%hw#|fl6I#4%vFkQt{a%3IgJX@iuE{7Ab*p8L8L-~
zR)Lm2O5S|UCx!UnZJtYDdDq928fWyzUFq{6ca~bgF!j6k?AA=r<y^3OookNYlS~`=
zY>k71fKIBfe8$@3`;d7l88f#qh3rVII#fh_$>}Q)m7HojQhbz;o9P<T+UmFIslJG^
zT5v2w$uim3V~4~#J9AX+RNBLh6Bx0uQ?i6DbvkAxnsFC6sPko}URRS9A>WPrXR!!*
zk6CVeVl<cDtxZ^r%>x@++N?v(2Hs_lp@i2mZ9YEWjZcqv6Zu}6r%^5StF7-tR@k0r
zVrA;5W*#GbVlY%9Fw~hdXaz}}MoT&3Zo@f6$rB`4;r=`qUpV93d1QHh*HfF>#%?E~
z`Ews(p(wi{v#x1vH@BvZcpl7V^*ka5A2veALr5jo@eTeimNJ(I0p`a~z7XKv48+~F
zeeO43H;dcgEw0<HW<;u&kmOap4?HXrL>iJjPQjBMXWTc22X%iQuRxu44?Niptj0z8
z2#0<cj&U)5lF?t)#^V=<$r;B(nm4~reO^ksfThIi0q!VhM8{d@!^5((l%(w2&X*9-
zBZl?XoWyxqa>$g353?>Y0+ZxK*V1NdH#?LF+8eVdN*@IpF1e4x2=DMoEyEE`M_J_i
zY}%V;l#%G4-V3eKbWm^m4OHM(VxJ9Oos6{))J-ljaS&m4q>z7U=rW!)S}Ko^AK~5k
z;J=<c3ruq(9p_Ph+rd$ZH!d!XS~lJBBHZlA?>+k$A$K#7@heBm<TZ$#?^8#m-0z!#
zwav(+@TUjgeO)+PwT*1|C$|d0!mYE=v2{w|{k^wbQ$G}`Z>JjyZ(Hb@wSrj4fT3ma
z^xV;V3ATD+WPR&cNbA<Wfcq9r>3%Z1qC$7k>w-reL%-grvWRQFp_uuqNa`j1RoJfo
zrz>dpqlZT{uFEiW@M()SREUHv!kex)3bw)!HaYNRi*JoFdVMm7nAR)xWb<vcU#U7v
zih%OI+Con;k1{Fh1TDF3%zg4$kIOuwbY{;zA2Q}yzrbo)DA+yAiM-#%O?2bWvsO8X
zn$EoP3g2q=d~6nhJn@SH&W-EdV%AC3j5U069_SyZ)Q$ThJ?#VR9Uc4pv2>^o4oEqs
z-V|Y1X+6i<?;mVj@LZ9wf+%LzR&sQw>?r*hYJc>nEPXUo2&<AxP%2k}WDIjP88^L(
zD|d5Vae)T)v<D@gz@q|p%FDYwao%_sDEy+XsKHK>U|H#y$47osbBnm!YUbjZO1b_v
zbyDp%%xV-A=AEyN{lXM{?SkpdRz<Xl4o4MH)wSk^9Tyxo7)r-^&IAye42y0F)kHfc
zmI?jgc<aqhiwT=4+P3Qq^l_ayEbg!%+_IW`aB{{C`4rk<Q=AZ(-Gz7BWpYc29(Um@
zFm9Co=o655-LXjg?_djtq9*3?gTv*c$|Dd>i%Q=4j3tab(eS0G^W)7rrY7MVlak=w
zP7B#Vo;bcIhcpjt3b9qcTAWj=vPw^LJS&TdilN3+^&~6RaqZY5g`DF^LshB?!H~<O
zU%TWJto|OZd=a2FF*`<$pp2nZYnr=DSUJcn?dXplm(1swHkjlUw;q;CR23{G{my#>
zS}VBhb~?1bb`dnq@NSj*vkbq^ZOxLEPaFSQkpTpKa|}y$+(mMNI}GUh^$6cdmJ74v
z5?4${X2s`_rk`_pjuOch+84UBj~A+7&4q7;`+>Ga8xcH@d52CzwbOldWg=MPB5+iZ
zZURr+g-T7uJqL{_0$?`iQ94gs)2BmQJS`2+)A0w6T~86h)jX3a^(MHI+nd7AH|~L-
z-R*q8=Eq}-J`Q0`^-hEN^nMeY5|WUMTZKuEi9=;j9F2bIfYcbZDg=`NKQocZ$S1GP
z_SGLHCOQhdUgu*xo2OZM^~X<Q$h5Y%u6KD=ZQ73Rx-_~TO4Q9IVcK4h`wc>dz5TKY
zvv%L5`toR{h9ngZn|^9E1Npzzmf@8-xo@S!OD9|bHi^q*9u?ToaY*_#Ac*s=R6{M)
zG(ox^T4p95%YM+SqByQvXa$?F;4NDi`K95HO}m)tXS0qMl-7e;4&N1EK}CdG?yp_v
zS#enCMJ_Ptl2Me1E|1)fIg|OD6&OG0ujRbzB4H7RJ<PYcydZ#?*dWSDv%hRZB_yP?
zZa*y}uYd2|kGOetN7$eXdwRA}#<3J5LFh)CMB<F_2e|r$V8J72MKRm*y!`7ZKvskj
zH53-b3{K#y3@-vc^ND*eqB(xAp3RyzvDR80X$mhOvYQ&=rT3hbChgk-nds?6hD6H8
zt3JIVFl(u2*rWsH6Zvr0j?38C1PE#2aVn~D2(G;Du_A@ZAmIKalH-l8Zl;#!O}vuw
z2EiRA`juhw-@CVL3P$eFxWcl*8-WqPChkWSf<qs$jYBPn(np<u?iD9Cf?yqkF4Gs+
z^G^G?QCo6(Byds`=f!akW-CF}&)4>o#rTjM5b-IM8`3vk>zi7^SPa#F7a@ItO609a
z83q5Q3NIp|w-@?kW7Tqo#xxr|6q7zC+3hHQHHU~F7VSTb%M7_7aW{R(N^=H)=v_Xa
z)p5aQsq(@;iNlUllf04;_42cK!u_t4bhnmCU2hK^^*FYgnqS|{Ba`Vji3<Z!-h8Q%
z{ACH>Wg}j7{mAR7JzBwND*E~CMfq_8<ksgI7@=A|OwexZPqaA%EHs5O(cSfQ2@w&9
zjrY|wdF5?ZdHk?@vi}02$dJV$sPg<24b(ov0x#<;RbCPV8+bJ&Y2J@xI|$Z&<aa11
zyqizA`k)81+BmOzm0cQ|IJIXu=PeSKOwZ8~EN#o}WTMeS$W=g=cuThlMO11kjRd|8
z@I9RR1{m6zKX)WZuo&~>!9)ojLm(4UlJg1a>*oUasyo;EOdoXt?CN^U{UBBc7XB+$
z;ob~Ry&Q{$CNY&V+}RP-SDfmQ|4P9`-V4C9hRv4)WB9!4Gr^ca7{ZB5;(!6cg9ru&
z4f_7@Xxr}e^L|Nw{R}H(tKbT`2FhCK*SBSCf#?YviObY+uhb;}snN7ve8(S^Kw+>q
zv{J@&B=?UI5d{DC27q3VPwVZR0PuC^uMf%v!WS1LQwlw;6N{Ke$%RHC?JDF=R*)~{
zCn{}hUOHj1M3E}+phnH{8EO3I7mdH-Z<!XXHEL{e45Z0%7=xs<L<9gk!;3-r(SIYo
zz{OW8;$KH+vOn{M>I3PYh3D@T=${wIB+vId!u|~o4~qYXZ}4~U|G6MGLcah5?#7FK
zhLZm~cYE^*OYUGrL#Kv|6L(Vx>iowc`t$$(^{H-*U`>~}jp6MQ&>%ThB!6^Ds%rRu
zyt)7MWoB4lo;Pp@e{JBHU_xg91!?`kwl3d;A47q)Bg_2#X&Ju)fM4iO0Q|51>Yx8l
z%dc%>x{q*3UtKOR-_gJPhjIVsJNR`CJd>EmC*ba$y#Ej0{k4b`nV94u_pta{@b4k{
z=Y`0dPaT_q3H7l~z?Y4o0HP@?Df}N0&7Xn&*ARUT{dKMd0rTF12Z#?LCSaoer(yZW
zZ~WsFC!n71migax_a`y%umNKf>@Ts;Ga-mfa9W47diGsD=)d_$pRj<yCar&~0KlJ)
z&nZ=({r@iVh=99ui;$l0PWtEFsSw*!9sfm~>q&jW1nJGv7Ep^3ccW9F{D-d|1`Zzg
zMO=vV6y;yT`g@>>27&N{`SPkt;Wh9Vde4dE|6c+7UkB3vw<p^Fx#tFg2Zdxz05-!n
z@=tkaMP%9Qut6~=H<z56TKYltuKDumx?(F6=#JSzc;9^Kbk9LTqQCrf;xr=QY<93z
z({4_7d?fCLM5wf)iUtJQoZo*V$1jULmlqGvfhq920`>opD1f`Ec3-ZL=6%(7{*gnq
zkNg%eWT4RV|7XnKqhzTPTq6C4*#}Jn?iL^>on(CuHA~tlAqjH|7O=|<3lb4ax%z#U
z|L3~)uWRp7iT5MOO5(cBE5Xm{M)g0;xql?2f26BFnIMdwnA}fA1YJqoiqd#U>OW__
zzkl<-;dvQ4#Dso)%zYyXlOghz29yUO+6DhtR;`pkem@h;my>)G_0K%}e7(#GB$5cy
z*VW)aq7s!Q?k4*`yJ+hPezLmm%tBH7=MD*s1H1-UR$u?O7d&T_UmJDE|L(>ExI5(b
z1#ow`)PLg}|8Kn;8}-*zA#+-l{b$gn5q-r5$tmDxFau$-4(uoX=kGx;j*gjGIK%VS
zG&npw-g>G$zPp~DVCrW!fws1`&~$b~gW6%^9sSPXPW+OtYccD@G?X?b0~r{fHzXWp
zFW1(+7NDZ1<XF0w==`rj8X63U7UECbqGC$GDv<I#L4SR!bcTmxami?bZ0bf2-v7nh
zTSmpXWox4e?ry;$xJz&g?jGDdxCOUhA-KB-2*E8l6b?ZG!J%*q?i4QHOM34<ea`8=
z_Z#=e{ez4F1+~_ibInKQd>x6;z>*SDLlp31iHS<WvdjMa_FmiB9>JA8WR4&m%L6WP
zL*2`Sd$Zj-(bOJf+0Mkbb_!98Ppc0J$?4R8Y0U|7|Ft^gD2Tt(bqKhT$|U|S4~+wX
z-9xVKW?}?35mgLWu5WBZaF`p-R!{jh-<>z4ns>j3iW|0OWf9-(8$EqFx;yDXhi5Jv
z$q~|XXj=2O7<W1COO<hTJ&zMXcI0hA%xYUT=ex>sEiAPu1`rtT#O$m<3t5`7^xuE+
z@4WVp01{UKG6@E>$I0_m-Tho;y-Tym6!c2y?d_K{6E8?gO-1)Q1c(HOb_iZNw|z$2
z-I34PFL89#8`qOh1-Wm<#DWBGcU}+SE+<e2B;qn^Mm>i8#!mQ?vOTd^0|PFR{(Ay?
z^}O=pL}x+IajNj=e7L?|4T)2G_eScd{mM||sG@P1!FJYPeuxSK3rb7hxf$?iDmNsj
z3~^~0VJG-htNgbY{AZ3ru!71L)RUl~&eB^BSo*IDDsw|@2-r4S+UyTsre>#y`%?9p
z{QMpY6`&|6oc>`AFsc{=J+=^S-Z8I)Z3;6b{rej2QUZ{6r5wuC|54f82YYXhOH&Ck
zE9Q8v2W&<2tRy8W35xz6%VqWQ@|u{Rr_bcJPtY>*<I3ae1L$q~RDGA4@AdVAH`_*e
z4*XUfd@k$Zv~set;Z(h}G6sIxz8yC{It_N|b}OKz3R8M|`m^heSTBtkJK%xm5M+E7
zwHl2W;afVs+ns_UCxv_onUJKElpz4rF_mfht-0lfW77<gcsvf8SGicUUh|~N*c()=
zaeY2Nn=6O~ME%<J+iSB}JqDumjf4dAJrik}wePzL3E7O}&NUO0^YqMa^=ftYWl0qe
zQVcxyIJ;M@zi@tLQEyx=7}d)6mvFSU@)y;Vsx<`E>vBI`sqgMX$kbmuc1xst_xcXL
z5>WcGu~Asmpd{C!PrcH9RjI)LzVaa8Hhp4#hJlr(W&H4!a}}pUv%$l{UaC<7gbd{P
zWA#FWCFOb(Xa4NQ>TK6l-p?gMSuwdJjF={L(n`j~vOz*hTFys3@txx`)4s5Fe$F6W
z@(5TOP3Q^=k~L=Nk-W~k@x!dpJlBtk{<XMwON+&9%Fi>%*=(V+l}4w#*g{NhT`tZT
zSej!XiSBEA`_e?4ZR)#nBic-n8*MNl(S3?pqoN=(@BoMe+FW0?<|}W=O4ZQhm>~8E
zUs~Y5KLj$J#J~am%^PaQ0$hQ5p<)pMDj9KSXTw!8&Z_ZPB2g0)MPoy?7OXc+3fO9K
zYqTOa>GzwoA}Q)he0|b7Pb#Jn!hH0Fi$2FsiGWhbpd~G}TP2_J<BoBW2a=7a_Ejkx
z=)hblb$FWXc+^-9c!Ew8nn`AE&r^wMIl&57mc8^@I_EJg>%}d=$xNdQ%k~M}RLnT2
zi)yLm&^ev!nHZn2OS8JglpZ`t#FvqPiF&=HzSWPaW~52O8N<)NDK^yPca<1CDya9x
z)C^@m+!?f2gl{o9K<4#O2~c@!)2+vGJ)?J_g<D%*Tl+@`aXe&gE5+o&$bhD_b%N|q
zU(&ohYYTgMdGSNOktwP2(!I(D@T-6bT~F8hINbJd=ouMP8LQeR)B?;b|9zm6zg+o)
z8<+pWji=~x*lOT~1z6FtS>N;Mw$t1FiC@3S&E)%sURmW(=%nkE#t9rXsXsp#w+9L$
zhE&e#e^X{8=js|+e(zb7bWEo{+sU!D8oWI?Ty?)7FVl4#AFm2+vGQRfn#1+Dor)kb
zsY0(|XwG?Kv;As0aZJmPWn5L2!{-v@GQuuhcBZGRyTyZVKxMGBB)m2k9e>tCmO~@*
zusdIR&sf={HZ(k}c&V+`fL+-p#eswj`{2<Nb}Zl5a$Hzfuk*!pgPMy+tA8YoUXt@N
zA%LZASw$usc>2qRe?YT29&XuLL%>7wvlHU`^sJniZc9`;C^c%qF@QSm0Nf(EY<QN}
zk}3?a!<5nwzp)_7hd&k1ei98QJ7HxeF(q!3FiCP=;eAGDcRr=SRU;1}b{V}XtvrlY
ztbGi|RWs6Hd1&`bFjOI*M9ZyWl2>`cjTNFyCYosm1X-Ctr=*1J*Za!>S~l{HvidF&
z3WBTTgP8|dnG7=l#Y1dz;nuq@-N{;v$T)>($Dm|;qso+-7TuS}Q*FNK)lV-I09Lro
z^E`TD9=xX+g>AoHg`FMnc=MdKzA<&&Z&ejG&ubIcO<|^||A$z4`^*n1{lyRc+2*+Y
zvMNx5z;W9obBeUPd&;8gm{3Ae5|x1aL7d3_8pEIcUG%YEXiz+A-X5FSt4YsFOQVB@
zg*|(?dNku@(*Le<nZw6PM*)i)l36fosCT3}OZId<;L`u9xGbP`zp`~JlC{mi*BXBg
zSFQ4Imjs|BG;inZQ>$Iu;%W<I`Tm4341*h&<G37~M8e$d0X6vh)U0fPQLdIVXY51Z
z$Zepl`a^+LPQsy)53(`!-2o%uV96Oe?@CmbnFn(@Y!L^{CnxCWQwkXv)ZQ=q6gFF?
zE*$^ua(U4Ertj0&J1|_r!-S>nGL>=|=aC}#ZLdICm7U$?z>!tsi>jYwC{xFkL*oi?
zo)>-EA*+6+Ow6Kre78Iu$&2BM?*v3)c5Q6WO}#g(042eibE+fYa{YWLG*vaj_atBK
zb_?GSj~>u~JN#XHF~f!(1Mc*aw;C24Q@M;<8;d(^Z~hl?koFp5NM-c5O{f0z)z3x@
z#2^a1o7K^C<tU9r|0uc}np9TV0%mkatQ5XGsN1@jR91&IaD{NyR)is;@BQrs_&!yp
z(jEsD89U6<Ix-XPDIg~2=2oxoy<241wPd1m-WZnhtp#|r$oW|}GV5)KnU+lR-BGLg
zVv}>N?JUd9p<{qmZDUKViTG1mq`|E3_2$M<La$uVRu5n|(n!Hx(5y;`O?_P2-lDx<
zBOIioOFA_2qbGZKC<*kf25UQIGyWd@luR~+&+1h-v5g3rY7RfYwipc}EA_s=8(k=D
z{@Q<{6Jd449G3kg4<TCg;L<oI{!xst=G$lTipMH|@48q-YqpcW82nn>N7p|v>%(;Y
zo`?38@5M->W`=}L`J8>?Tb``?F{VIj{0XyIAr?o_jb)zvwZOkXn}0>59}$7FK;bRY
zBopw78|7JsdP7lUsTHkXd-tYw8}~2PSMv#_(EuUmDE&fgDB$rtLl82x`n7HFtIM#Z
zw2aqXMI#`AlX0~jaw>~3JD)nggzj9xRU`G<G?7(S#&3GndfBqiL?JWuBT3y+zpxS!
z=i}11)kQ??Ex@-^<&Q+wOMG2GHeA?zUfA3q_2@TX58c<bG%YIf7=9bj)DkqzSO8N4
z8zlD)@NY$s_sjcPgfcymdB|h?`C)v1ypXI+dxH@}PHrvBT3CL6H>q|zZBr#Z$iR?7
zrvC^JImVDBPuZEgcpaPIr|jN$?`1m}TstiO9&xHu>AP&65AfVo(WAC=v}9CDie53A
z|CILcfFn0&#j8&ZusyN><^Ftg2F=la4m3?lNKD)er_E6j=`z4uwDh6@4w#m4Zh&x8
zynSuSVlAM*rm#TJEUR!Ug^nX|8O;Y1I&R7{3@ElrigVtvTg~{NS1h&&?6g^<VPd)*
zP<NzeonlW>yIBaB-(_$CWliK%YHC>%)#JnzV3&PRa?`;`S6m{=J{!x_$HSMYQ`XN5
z`tiCQ@TN_uY1wK|VRl?6N6ca|P;5y@9>>jvz|_U&pH=u<q(>Bdp>B^m`yG5GtGLTu
zb#`$yCG2<kWIkLvwsbafx68-jKk7dOnLVF1Vn(2Iq~;r^2|P*#IQW2@;b4Y^vL}Cy
z9RJ_xZ&Ob^{?jU91FMww5$CJUa|uiE9GKSK-O^+gN${keooue=<{^SWbnZ3cGA^f5
z%j+N2TmYNa2?d=(;GRGNEC_9G!c*I2#@Zo#D9*<AFBU*(5PZA_YQ6*}B5u2+P-J{v
zsra;bim=B6{XtTm<wNiz$ltoWGQqw~YxwqR1-JM9M4>K5MAeeiHLJ)+jKK4#rA)FH
z@O^c1jwZ;SqiF^Nn2e)-Vi!MIXv@6^R0of@G9fs<EPMbo(1`P5+%20mOg>)r^I64e
zj|<Xfh2U-(_BA8Z%Ub}ew0Ji11d&u@b8A5&k0oHxR6nsdB~JXY^C&54wz=$ctRcee
zW1Y_QFd$N{nRxS`$O4>~S7|EG#du?0oN4VLUEwu6E)vC71Y>2=nalA{n?%L0wb36a
zV86Wc_s>a6P999wYSEX{QN+MFqg9S)2QKO?L~LrI4i!pDcu81o&9?K#4o@BF9iBrY
z-8LC8QD*tF?weGjrjyXnSsz%XStOK_IILInBFzibD0Y7vHt&g>IsscN5AbF(UW>_9
z+hgsl6)WemG%lV=C)@8%6xG_z4+pZW9sf_R;%^8>1MrdTFX<KPp4m?}9JTx4+C8xC
zjqUJO0CdykmpIDW_iL2t%~ysKzvk%hA8*AH+ouB>ntCO+q-=lJh{SD{GbP`YJ+)Cj
z*V_xBlV}2BzYp9GA-8@3DL^?me6`}yMbGi@xnRmowDnR>A&JxBA(6ohGR!)7i@w5B
zJ^<Lzso77RKb;znwjC;khsUO#9nTEiDGOTO@Hquyi$xgB8a@Yjn8@nKhpoLb6Adbt
zrw8;OL&=QL2Hh+FX?GL=xzQ9#EzJ_p<!eozH^+i&?jAc<grchB^pjS;H}I+m^gN4B
z?awe4n5WAyYe>Ya?IR}_A^6y2Td;&UYdFt0U%Ocqps^)%f6{X{DsW^fnRj#*i9DLR
zq0B4VH`TNnWF(`fr<W(M@I;!L#YAi5+TKe|Q$l`%09U2_w>$b_vee=!A)3t3xW$h7
z^`duIJHA?V*Y5%|v#oNGW~knP_InNBVeX<HdSE|OZ0s%R*1i<x+yI(f;l`<oN@=4r
zpTSHvRsIaU%_2Q~h8)dAY_XxJo9EhEzw7r3P`ka;L*+hxn+fB6^f&il5uB%{+amOf
z`ab(VR7IUI0D_<)54FYvJ`sKTD-Y^x8MxKktsJ%Acwf$24eu7W2BjLg9=g@qmwDV{
z7l$UT7~T5t9sCkk)ce58Tv%9CG*l2PT$*zOJjOH$sDm782byd1k38WD7w}eSY(Y2u
z6Ocd-vbzO&4YpfwEy-EG-_u5l5CgxilHP|hU<Nw4S>3Sp%ArS_;qfTmP=${6ag8w9
zqGt$wD>BvaVkjpe@ilU#TOTqz5_jw$<1yf-P-^f$ZT3BjeL80_I{@7cQiuphKeWcj
z1wC6wVlPf7iZ$#A$_5cBs0*s!Dm9(yy6qnt1yGX-ofj6_05BoY!4r^5e=Ih;78MkT
zj|x0N?H=zVe?SbN{a=#QKX5&pC?NkJ%9vyTKx^<*e%v})lA+AaZCMrIB{+@)fx6V6
zfT18RV?rMCYA&@ao@7eM^Z=Gi$nrUAE<NK}{*Y(~ZW=re23AZ$O6rHYvJkVgbIsYB
zG}&M<y5RfgH7hDA8hjlqSPoV6v_bY1mujeMd3*c4%@-)*IZz^E#Maf-x03c9u(W-X
z{BJ_b?>ns8hH@am*h$E2RK55&_o3UU6@1lQB5R$^6S`W(<$4#pA&iv?QS6cvlhf40
z$L)F<wl<6YA_t4eg`P9Qdn<rtB4c%OX`JE-bnd?U-kOvSnx!?S;~pIV;{}gDWn=;g
zKJ1}JCi$=~ZY*^?XaE{7iq6+9{2#s0=n2zRO?B>KGF_;#w~)RB5x3HoD=SuERshk?
zRg+TRjXc}9VaLjqpG!lf@BiIMX@S`LPvYtC9Q0RiG6oElICm&vWiY^*98?D^TmUBH
z7vQI65muTh*Lhvlewp#+&6}aQgc_IMI##7WbqfCrk^!)_mZNU7t3!}{e0*F@U7c}E
z@H`61xT=18w3K0Y?}+qU<fIr966ol5-+UBwCkMC#1g{od&j7ohyq|c%4~Hu?r`fHs
z>8ZbYHITtw#iGl81&vSnkGXl*&A8lLMmK>5`FW2Oxu2Kz6Z4By<Iac)8v2xzF}!M6
zlM@<Vj!%cIY7Y;VRkajx#CNU3RCV99AK-?Q70CjqqBS&1M+ei>6L_2xl16e~e{Mh9
zt^3DyG!*#$#p#QZf)d%8wi^Atp<yZ(dY!MtE}tW2Czj?Z*qvG8H~J)5_Hq0&Tz1xK
zp8x~e)0CPPqphJ4%b{X4uw@md)c{W+|L#LnZ!r}S5`7bz*bj>BtSX6;<s66ki&MF0
zPZ7VKZgM987yvkDDR&eSKA~&G%DUnosYe2EYY7|^Ti1-0nk>_CLm&gj=L$FltQ#H?
zlQ7i=^-9D_B+#}V-w~t=`*oI<7Tl{v%RomLcX!;O{BYk$jXvnZxXih<k_Pw={z;_&
z>ld9OK)QL|81{V004(2q@I*)wy72Ye3AIV{+<6vc=&|n~cI$w<N?sV*Rad4fT8aNu
zg75hL9Oc;SC)eyygZ%=WKXPashe<a~Gh{}Oiko|;8V&#HH&#hS8Dwx1AOqv&S=-;~
zT<iM|_Nf_ZIhM}ah(&aaI4xXkbr@Z~>Li)%&e-^so6Tb^xmG3^6(5`9ujcN1Uay7c
zXNNIHX5uzJGew7u&ikV{ZXFAqB^7XpR@Jr&@NcybkS09%d$yS4;@;t?IhhO!-wiF+
zjcPh{-xu9RAXx|%C@Ue-({VDhDx--!X+bqwg4B$ZtM2zxG7HPvmlz|l?M*+d)Q$4X
z`(G4r7S`3tv$UOvS@SeCmW&SkmyG>C?q@@w00Us_zp@jW<(@{Q-a96ITS>4k0N#>U
za;WT-vzYuEeFi*seF$8?a?4DIBZlPQdD3;hxd|{YK$dZ-D=NyzP;;sR+}V7PDRqnf
zsNIB|913>!acDOLszUN(V2nHE;Gpc;g}U-^+0#vIi6L^Mp;zKpfSI#S(BRN)21O{!
zs5mR>?(ds#Uu7n1NCl(gYvu6JX9*khANqVWk0oM6q%Xn3d{t7;M_9^GXfr7_U-R46
zf_a^BOEHIO5<QbuXQQ9N!?dE_?d)OuDH%dmuJUr1mh{bRaa<6G>eX{gQ^Zvc_32>A
z#5kBKnTY3y$8j6;Pp;O@A4d9gp`4c2xN0g&`YyE!eAg=UY5W>$E-uvE6sq2>EdbKc
zv}sJ0>uYf=GJW8$@?=v298K_GtNTZInN!2Mq_+S=MdjBnG@-9ouvlRKXkPSvhUePe
zw$*-J=lTJCkrnpl=4>X@aj+m($g(t-eKJRhTucIPiT7yt&Gr5N*p=sRS7ZOk<05~N
zyU&mnr&rt9C8bl@-@75}vS-r#?()!OT0=`f&yyg@{a3mbJdRp-UPeI+5~t-zoXF#;
z<yO!UK-MLdmJX>twE6P>`zoBk2EcTmGMZeko|Sv*P#hpDwnjkDw<&w}5N@#s+(c5*
z&<y21ehBl0OlzSn()vaK;>x{!&0hr@0}Il60o#8vH}C?6mWVpRu-Z4k%UY|;JeNsj
zAxsN{$Vw@$V|>!wj*o@_x&J9{$%+>V@Hi4r!5{Ov4dLkcE|~7_?8FVTwu<fT?SE<%
zYx@tNG)n9bRb2q68_zKhtk(ep#V{T&E^~2C%iEKc#$|7LLc;rjF$5h+F%`DZB#gK9
zDw|mkzkl>1Vq5IbFxT2l<KtDDYySJ?A%9u-3{t3Z{coPP0#G>t=<AMai5vv%4|(1x
z)_?f5K+I>E^Ro?0<IunUXCrve6_M!M4Qd~pm&ZS3eWo_f0geBE<ul`Ve|&72@&C8}
zk$;ubco;%(NUgc&!T-;0n^CO1xOJwne5PnX6ap8B$*|v)|Igo84SRu3SL5@qswWIk
zr^sRa>fXe9uIT{1@Bh`+^Z!lpeWn=xZ;T}W(1-v3`e~^bfu9lLzc^1)QBtb$gY}A9
zYTxYW7DPQhg7T`+!;oFi<*9_(bxv{?YV6x(0s?s7MY@p(vQc4BlvwK|BqcD6kJkW6
zj8A1?h-5|H!^4VT#_Qr8y%49e+GHIH+55W`H%bndWdm4AY-$>AC9rO0k>z%b>ZFX4
zzmYU5jxYjZ$j7%$(e>kOWN4zCum<K55+NC>obewMVb<5Ns2xXmHO$TY>fg-F!Wgt=
zv*%{yFz==}?=1^3Hp*K0=>3AMh89G={dP%xhBP3n@B8J$hwrj(eGSzg`;E^{#!*By
zo;TDESj`}CHZ!%5JO{^lVL`iMDI>hmJ4v%*8S{7F-)T9d#$4jrhU$)xkYF>0_9enq
zc1yg(hNHp42#-rZoW%WH83SmGb2hK!)LcrbiY<WlgW&IPru9I=SuDS3@V-sPRO~l7
zAs6?X7<5OxRWhuhqADVo2^W*w$&3I#yjti8emE|B;ID?kn!`p28PFzDJYRF#;hNi#
zR~$<ME&G=&X^^66Ahn>?)Bt5p2F2;AEEX=wYm6|UX{4yNRa{puqp+eV-Zmz07tX`P
zS1e%@FG>TShighsd`W2-mXH1SB!??Cvyk4B58Y)blL!?n0XI}WI5jVha*rfRL<Su2
z@ZdMc--Xb`{;W7lJd2e}^R?#@)a>+<g!>*PA~NjkHNXr}SrT_vMg!IdzkH#-s&GEt
zpRN1?g{8ic_3$P~s`7c}-6#;d>}>nmx<znF$j%M{JyUvlWmQ=T@r{^V7jL+{%)2ph
z3X3UsFX1#rDCbke1T9{}<cZ}OCPhd+pl5Qb*mx-OR5aUhpnc%`{rK=u3Hw6LTdJYZ
za`RiqBqcRQd;Z<h=#|@!&7eSh=MP%8MKk;FKf`+RrK9N6%O%2QQlv^MNSnpvq})MN
z#pXjtLIUF7Uq8I+$?wpv5cd`PvWeHkq?SQwQ<y&rrcf{L_w|+<?YysODSgKe)~ygm
zI*(O@@}hWN7(lQcq!0%dLJ<gi(Zf;AH}0iCD>8kmtW$PKQc?no4qqt8yC5mUE^Xj#
zC4iPmEntJ+8Mdf;`@r*n%7t)UV`-pjoDqkZm5r2REj$F36e)W`X(dK~>=4913hy4g
zM^Ck>A*I65gqh%QfeDBklH>WNo~+O#|8|82xS<-EFFYF0{ZMm@jTL3c?-5b1w2AD2
zQ}O<EoV?I{3lFgzFOE$s2;#4$6?OrE$bV4y?n0Xt7sq5@p{Q0@$3$Fx%<k4nf6>$g
zvm5(4E~ipy_>NL3`sUlW_u(<oiJu?+cMlE-P918m($L<(pc9t*AwK8&tb^xP>4`t4
zJSJt+4H!_zTAA-a11fY1ZF+NPLRJ>bPenI1BXtW<C%KDhaGU4iJAw*XzeA<LA3^A&
z2qg;|6wE(SMU#HW%NWOfR8N%Qpni!(zz4(OibTyvZQ$+WQ`GjM=SRskT_rtWaa(Ni
zeFNBgt7$uLs?QbQOoZYbu2TK@;MOo)S$b%*yqRvgv7VeZF^AzBg|@w0*ZUTR_ct`=
z`th3FM=`)hOMCa6GT#7#GYmE5ApNU^boD$MWsR-i8d{7h7-xBmceq^|;^|n4YSaQ%
zGD}dJbk_pDqunIHtB+$(aV4SXG&n+KNz{78b>>LFz)PMZ|ETdA#Hi4t45ZnU#+SB0
zX;Qt7wdy-}(?B)|=u;&9N#vw9N&GT8str2cK0AorH-;R()Gj0FWIoIDq`+l|qjmrx
zzbXU%rNTeX4a~tI=A^>S57sv2$)9w!UT`+eh~;@H&LPzMR2(d_C!}T3EE#00>0Gv-
zbs-vMy&tHx-w$J}HVC0ADnts&o|}W2?Nz%iDlU@`=Vw+#omf~n#AgYiPm><#?Qhjr
zg+NMQ>{c+N3d7PLimPfFP++R6CS>GL$79A%%#YL5w=%pV;ix;KCKXM*yGtfp{sd;g
z?dVU2gVqi*lG9D0qAn`kk81>37Wek|$w_Gjc2>MtEMAj{SN-LtP1_}hxc*AwO#w_k
za+J4EOOfv#O(T;roR+hU0AmBgtU51lPF6;~LQs)WVQZQAjhDTn?dhPJQH*e6y7Bq@
z)_Y1VJD9qXwhxJV3tvF#^RRno2Ed@8_9YJklOsXPD^J8^*wh)8oG~*lDLDm&EOJuJ
zkzR_VZv0xtIP+8>RhPU?APCEs)Ba9z?cpm{TN;cvQrl6bwbZB>U3W`CbWTuJ33^r5
zuc@U~{O0&5lkbK)Aug?4pV*k#o91D?u+}beZ<gSSpyTv3d`~FCgI3{GFJs>?CLm41
zY^1y11o#VtvjlpFhd(D4RkmlNR*pO+1L){JNq+g}!K%N#_7It-^G(Dw;(KF#=_z<}
z3AfZ?P1AwxSOq68)9A~ha++d3z@T`%Tt=+k?7btWW>zQB{u~pIxf?7W;tuS0e9UFT
zM~yyD(bd_)UYtfpq~)?5brS#OeBCu((Tt=F7H8)j;0d!N;vQP5F5}}9@3vjbL%%oz
z@|F`N-Ub(o|CKgod`_jq?t4c~*EE$D$8%y^gcp%Dq-7w8>cr$^1-P&|f5xhH-H(VU
zmnz)z$)s(dSH|4D@Zu68>&TN+APIR@5Y6P<dOpwuvUO`^m~Qti`tgGfqA2!c?NF|f
z0=hNdzKd}A-ou(!utpTptYqWeaO=^AKR&=Opj(vho=UHk<-XkW5?=KqiaWUXsNK&1
zJ@+~-wIFuYASlEU3N+3f4VjqUS}C$(Gc!b{iaI9w9sNqj3yTE17&Hv2Pu{E6u<?!|
z4A?Ppf<`HGez@%n=UBY8Hcz_crlFyh|L}py9cSu+pI^7As79%o^fLMTDOg*p;&*FT
zlol;`_DH)<pxE71_+YSW)Mu{5D-nC8-*~&HZ{ur;N6!2^b%V;^GizxV@UJq7_6tGW
zp6EE9tlDH%O65k`bkIIz!TVaioTrCx<iNyZ@9m8}+H9^S8zVtzmO(!S_<nQNB2!Uw
z^&kgN8=v>^8D<lBD2;}3t!eC=sj3q2%-ghw^O*p2#>K@4Ss&ZFymr4L;~6<{%go5j
zx{Go_S;b*BQd=Ntd;hG;2B*p>C~4HBo33%QP>Qv*2uusV=ug$5fZd;-`mTWA6WA=)
z_@SY(q&_h5j_}EKAljxP#=R$?zQgt@_R0IPOOPz!<QTkiY9v^7`J}7j^$;+AiNbx{
zi{PFNU{xx4BzlmjHTH`{x~1AfaI<*(T(*)~zhd&YAKzZGOLv&47}PIKS;jB%n{aXp
zRB4AJfUs_HmAJ);PZiPxw>Fazpos6}4((Z1YJ;IZGZtUwcUp4q<tuCRnyEt}oB7$4
zuTWpt-A$8IV}C8{AWRM%WDj9z>x5WaS(JP@iU}`xXkCidd7#1?ix?yqPLL{b*J}RG
zO~lG_z$BcVCL#7itfLnmg;LwT{z|IlQelqn=Z_(+Cq0b@u2mXoHoxg`MC98SO1m5B
zzC|hn5<|?y#S(A<h(rmk4DuRAXa*4~DUKK@QCQqv*y<BWb7oR9qP``cKhq745}EzL
zCtCJQ_4-<v^k#76a*4+8!A~vNBQI9bqjf{0gZC|&^6-c%QGO2Mv2c^mFVpv-Uknb~
z#8q>hu+^$xbl(wo=c|f!svx`rVd3FwU%a)uNIrS|5boQm;f8YfGSDE2ASznf2+^#(
z<;&E`$G3F1Qq^KehB~|HUN(vST6>54$r(c=6Z5aM^N_wi?F|reyEr#>(RGo~bEico
zJl^<Czl_`g?{V}5!-Y~OPG5&qqX({Nb^A~=T1DOJ2>i+$vwaS)i%dZbs#6xO41u+W
z`a0jTT*>l>D<?Mll8%0d%(?asy;cbcIZeZOFv>~l3CpD$2K5|!RXF)vXP_rGCVP`=
zd7nZj1&UwyJ?6O#GoR`wO*}N}O}r!&1lb%xwpfSqm0i~YS)wWon$sL)%QpV;2KRxC
zCBAy>bCc*`>|>MmIIM#<`frW}(q`av^jI1SNr4|b`+6qsC@><1@RD(>vsFq-Yi)6+
zR;Yp*AxUebRhD<v9lp=OKmqH+i=pj`JRtWj>9hqIUPCZ3yIw7?IsefRc*99_Yqk%2
zy`T_0sm{-SuIlYrTC4eRJ|@$+3n-a%^n8k|L-Q-kQU;32{Jg&<zQnD=^sUhc1NHV1
zj+$>U5)TC|s|*67Ovd}5!P~SN_309xZxe1@$uxSxO?KBGad8w?idEyt@z%bx;3ZXG
zcESbnlMv*ta=vHB^X0XO4f)ysyWhK>uPB>)0Fj02btJ4JU=M+~2^5tTC{9ec!YOGL
zh?tq-XD;r}d)d4uEbd`+w_ASb3{I3`^4zy!DD0>*;;21Oo_>j?L}pxoqsfrQYnE^K
zea+0nOtWK0+>gE4!M;+9NMvp!QdvP2k3j~bkX&JRpv(su&Ea|*FA6tn;*DAS`aE(F
zm+cH50h<Tw0?3)%KhEmr#*795j34=ND3BYmlJ&M@sP8qikbJdCp!p0X>|wMlCbzV4
z{M3{RZMK0ulsrZhCsPBv^Fa7m=^1nWw>AcBKN3Hv%(vNx%o9tK6?+)CxMlrXa=M>N
zU-y%8@AzJ!`e;mX`>G(_(kqo4E10ycn{)M}_9MY?0sUnC_THb$6ph>AB3}yP56OOn
z;Bl7=h2RJ}I&^tj;D!i)Qm^CfW5$fPdCg4CrLKrlQ-frSnI&_n;6701|F-Uy>TfTA
z^GE*GB<kg6`-<oYp=Ik;c9mF?OlJ6&Hw$X+%lT4Q1T3?LU>?nl@Z*;>CqhC(L*5k3
zoIb|mo3`RO$L055!uJZXLOw9jofvfb5c?s~j-MP#1Q9^1_H*VhO?Y(B{%RcWaAH6{
ze%}^6?HP8xJWY{o;j~|(Zj2$KXjQTKZUsS8rmGQU<0T@V&`yD;^s2#2v=Rcw>KBKH
z$wMwtxZ4Ja@6Aw|C>*RcMVqYj_1kJUC-7W6!w9oeW|swau&-%?{EQrP`8QuI4_c?U
zGV|X~`^=r=kbtodBN`Z18dS+iUr{T*lgKTbI~uMqI9btppozU59q29f|J>12iuZuM
zVmo_REg4=dH5h(XJ%1Hh<?FuCd-MQb6&u&qT?4(*QZ{KZIJdNvUrt?<A0L<Y1?{VG
zFq-SsYQHbglU45%?d7NUHm{FBUdE;UsXpX}5ftW>{w;bS;xIFb3oikfW<@+Aixf%A
zP$ok)%Tprz)0Ta<TYC}>?#4x=JX$M?U3&(G(=gjkuY#$}RmW<3>dP`KhIxE-14pgG
z6=YGVEevvnZn?bK+c>^^b0Xt@SQL$okZ7?m1IxADz+;%L-NCQ=>0B%#(bo8|y6{yb
z&e*A~)kqmcw!^j(BbF<&!ALzkSOX-)_1+C(;FNI0XKC=J`V!RH%EQf~3Y6VRDlfXA
z&SiZUj16MN`=rDovg@&a51tUN59{if72vNv3^1bk8&PsG>rH4VY>FyVFD)(EL_^CH
z`0m#{_VNEnjr)=tX}XNgSX+67=4Skg5{1dnm_5jRI$}smBtH&o1OX}5_L*dYf4<(D
znMn0ng_#nZhOmjCZ?&oBZ~AOZ7F8P6!Fuo_NkZ*&7YCg4NWW{>9-U&>TsG=O?FlI?
zw(&|}U*o;k{uat%{WT^K4{xv}eedF9u_O{V(W!}e5CZftiC;H7aw2+`!@zKRtzBJa
zCb6Q-_&~|ar=WG&QUWIO8Yiisvaq(Gn!RIz)k}r?fmJFj7BfN{gG?Lmt9Q4s>bg>u
zg%;*zis)mCuB}9MImO3suDHEPZ-2hI5U>-3m#gfvnur6m0kWgW&EYpjBF){Nr&gbc
zc3%B_jA;-jDiyAtUhI}$!7pWOy!r0*wgb6{8bN8>410Xm2PTMm5tZ_$PMc$Rh>ncg
ziY;D)rZ<~pueaebGf<KuMB^t`XH;C!AZboXOX-yDIW{t;+n4^HQmPK8w%-1Y>|d+`
z!;M}#99_C_nl3U<^=f(sM-W~hl0i>joL(~`+AVKww(gnCLV~N@>+aCF7(AMQres89
zMLWv<3JmrVk8mng=p!bf{aigZC3}ety#l2RB5f-R%WY(k?2gE_)Skh?s?fpS1_pKO
zw9oS7VY0va$a`^S9KY8D6p{gRP(f<#U0q0a9A+b?@dTy9r6PV-7umA`tiIEtBtbj7
z^L{ctt+S=ubUELa6`7e)p*(VLVK52DcD+S3zbV>}6&lg)Z$m#Wr)zdp-S(ALUE%+x
zPpcwNPIfri3s7ngJM=|%s#xN#@d-He9tmbAySxGoI67l|yOnJApQdd!Gh_`)vNvF@
zrV$IuH2SGnglCKNP=|07O-VU*r&t#QG`26Q>*22+J8^}a7AnuZ?|O-6S8Xh)anu<|
ze3jy4+QvWKaO+YY!&SUJ)#Xsi4<Vgd5lQn(TylI3?tYFbJhg}cWvz|5QQd`QPdC$t
zJ&9MN$n<*;^$UGvk0cl)giQeBq0y^?On>X$O5)2!Ul_9CiTb<!Xwz5Q$7uB-tR}2J
zd%O!_gkGrq1#gXJcMI0Oe1Mr@2agtLX`v#OR=n&E>)IXP)BzK-WQIpzxB+E^PlbDi
z#mS;fho1<MK5ajSKBpWlZ0#d16S2@0UMCN_?O~H=*IH8P*>SM7e<Kh;b5gu^uz&;<
z4)|{PR5Fg`v%G`H?5WxotKW*1&VlB)gH{$vptdLQ1ac}FoYql^Ok^}HR?6mJ^JP3`
z3(#b|ySj(kzo$_&!Lf@+DaDwnqhKRZWubgpv1GOuizp8L%<XLRZG~E~b;;8NyRy0$
zw*#(t%79iAo;LpWr`3ayho*x%h%)}o<mj@Yc8Hym&pRMYZY6VvOO5Mhq44qEn)SCR
z^;$ocL7z6XTu-IqA9Q0{yvNRoY0#{f1x!;!@M(4->r~%og@S=`CfFVbDN;wfQ1x9Z
z4~B0X8o4sa@cG>LzIgQg-VZ@iS}-kmH3-Cn?;{Mu*zF$e7HDXI%L{g6sy}_Wf*kzQ
z;75aiAq1i#>`I~Qb0o97h;GVBn)yYEw)I%0;5H;U?n~e3>DBVdJ;<axR#A+~D0Ax?
z^@wfFA7fA*VcZd?3k&lIQ794Y#G3x47v65@=z=X1U2+9MN+$3#Jp0(}m4pwAA_2$7
z9bejG*-9FOi-&<Jr%-9qt3|NZrK3yIAvGh|i-x%R<Lcp<LEi6~;Z%z#tLO9WP0=@d
zhg|7uJRKea+G;*Mp7NPPS};C`vq>}_o;E7hsWo}j&M#-LAknQA_ziJrzIaf&gIgni
z0dm%zSoGg9u;f*R8=BQN4lXJk@j(k0{9J%Aj5FKVIAz;o{jI-jx}lMOCM!oBUVzMf
zPpG`3qwLU^?BLXVzqb+7f?tS?v_uSvn=x{rb>@y`cxttL{4iRJcIF?I-})w@ClwO?
z8oyx$s4jlx&o%J2d2XSLG<JToR~NCex)$!YWu<mQj&)~rdDoh8L5nV~8iqVCW1|f2
z*8VRlG{fHq!V2%H3U{{S=4{X05Z(pQjfc+xZGR(|Gbh8qj4Ja?;Do{>(God#kK3OM
zvCF!g50nA`c9+w&NN21lSLO9Dp~e08^l3lc_qi~}Gx7HN;214Lao+kp^rbpAGiH7b
zVHi`yQ<OOs>?^gW+gJW>Jc<116%Xf^i>^co!6Rj}?H#l#6djjLS|Hb6dmm^D8MFCQ
z$enMeD4w$)?1$9Q29!lAKR~)zh*dxN*Bl`={66Ba><oMF6UjkD5@GJv9Ry6TDn|B=
zr(8?oR1yf&BK*J^tWKBqapO$@U7Ag%DQ|PF8Ii=x*n;x)qsQ*3I8!&wQv8UAo7P!n
zoM`UXhcQ}pK{3mUNgar`)qX=q0#iGn9_B)0@pp*oaEJ~|PI5?yczB-ZHuL}!+x#Oi
zM?<xxtN4y)bC&x^NYb-?<y>@F@;KssrrDlewS9Z$XS@iIiH1y<$E4et(8Mq1uaDsu
zkv(T>+B=@D-N+3A>n$&}ME&)7ZCL7qT+esB;VFOc^tX>+%|B#r)~-Wvv}#Uzp{AA|
zil1m3Z~dl`=|SaOkV*=h(&562z9rmb^<xtOd%|m7S9`1`;wQcf=a&$8tMVjUj}cqS
z&)b5UwzOlxQW^5<?YITl7H>cI`k*4(LY1RNCHN<f?z~mn87NDhc{O6|R?a6>jAki(
zu<4^65KwkbLko(2k666j^IAj3*@I}7MPtk9)S7wzFI?W^1u&~zCo+I~=hmjXzgD>j
zAdP238~M@W-gd9<phdl4A6Jz_xw>ZVZQ)bHZN`e%%=jKC#66B^E6I`5aJwnQkQgtL
z?ZaV_j*y3&n@lnpCIMc(ry*6{*hUp9T_Fr-Q1=EQK-(?^ipd$PRN~Wf(;0hl>}+T+
z@S@G|zdp%^ckLu=!h12Xvsu}IfkeadY2N#7@&Fu4f)0=DZ7qb>8w!?sR{y5c=H(2s
z?nDTVa-u<Rcwm)6dw^UTeK!t~HJf&uKNjDPrKqEQuqVTuwkYsJ_kMRN$BQcSOKH=}
zBM-3ZGfHf)npxRx1b|cL_o|DDxPE`>p3kQ#?U&;x{!?e}1*-}rZ55VVRnc)O3#dp`
za7=@xwWFhkGd@5dpjKdv%naBFv-R!HUtd!Jl&yv+_>}Za;W21fr@dIXqv3}5J!Vz2
zAREvq$ZFdji%M3|vS&cCv(k~{1+iSLCJce(3D1*MSsG04D#P8J$&%Rt!-dhfS{iXF
z`d&-<RL6d)oLshlG!@@r77^O8v?fZ`WLWcJwNK=XYel>A-*E^e{B!)q$>6HHp(7cX
zDzg;8+UFLCz*wSU*av6ng`QA~ejLzvr8YVo>)YUC&ZB%seC3sC2Zwgw-cjfEyHb3$
zsGxbDnaQ2reBi9ng69CZTM^t^M&Y=R#JROVat@M=qr_DnoNO=L?-UOrH)eSx2k@=G
z|K1p-vIxF|y>$ZECiPB_nUBlxXLV!3z@GlhE}t9Qxe4t`&+VOFHV@yay@Vd$pX?Ul
zj{=*4ez&)AIZ%eZuSGlLgs-+|;~^J$<W&54Q_q((4({ANEnV{XE7P@A)wcmBul_1@
zntgjyS2nYZkmTCMwb`hpD@i^8xR?200X8wN9IbdguC(fnaiBehouPo8U6GIwy1ibr
z=CC{kHG$2VGb%tpAHJoaW}~w|+ql7+#4&u`(HkgVp8}o2dKxY(Y*bj8nEp8zs>Uj~
z6E}HM78#6NK1|p?LGpvA6)&>S4a@(bv)~$N5BZjN;@!x68-AQ85ZonOxU3QX*i$7U
zS-Fz61hU9p09Xs)T9u{ZPN`Ahkb9*0B6U;2A`G#BsdJh#ioNDODLCTW9UNg%33{Of
z({rfy4XVKBteW3syfUoA+m0nTOgMj;;iIcrc7CoV?EvxOP#_UaNMJ&IYyR8Wv>O9#
zIw6B(Ps6gCPdu*+wf&4Nf8&&XT#=DFh5p{MqYU>-@dc3Ht`v0~^QFqdmKF0R@f4NX
z#G}DW5pkwpSJHlRo;YyMElv)7bkY`F-Z}@@1z4{@*IkJVkNxzsPvnnNHuls1R9ai%
z;Os3n&Qp#=f7|q0)AJg<*r@i&%COZ8b*l@Cz!lAR_9xX?FdCQ(nnhIzs(If`p$q>F
zctxT19|>%6wi!YO%_vj6++&$DW<**H?o|@H*kIG1#daUK|A;U#&Z2=EOgmYRiqA@6
z;St5!UZzVs^h+=0u#_S3lG{U`M#!p(x?8_nX7Ph*8`xUiXjYfRR%xsxI2~Q!hV;^d
zmmb3r$X{TgtSt1{cNDNp2mLZNB7cJ^z?WYcDE%c4t8#<umrwtW2kmwb_zvpS1acY<
zJv}^FpWqLu6i}L(@~aDxj@9YrZbqi7tI{5+t9eybg(P!q{Mseah@6ngvulA>+XiAJ
zdP~_*BjM*%1+)_L(6!s*B@)2bQq19UA7(4?zWf~eCgdgO;O<X>!<-Dpn0y?y>Qh50
z3Q2pKn@IHhV|-2g+*{jV9p1gF+18f?PVsn{Yy4gSuFY|98@aF&@G4i*zPRdc5^sBn
zP=fFwX0`v;91@b3y#28(wT*ov8G@)}*TciK%<R-vfIj>(Rl52Tu|016sr{?{1u(EC
zD#-253_B`L&8Sg5gJ>~Swlz~yi3P3v0=7}OgtL{KQOXA-9(^PPh&GIs>4Lc;-xR0M
zn1_UlQh*&kK%nP!j!E3^Nj`|-YV$F{CBP-AST}mZH6`n#SMb5ZtvUPzP-=s22ottN
z;Sc9AiJ)eBD(7U0;7tv!E5R57s4EP0gL_xgIuw1&K63}0gj=q)iU(6YOK$5|6R3=;
zk%LER9?#ROqKvsM7f|n;7D%e_PBTXx+{!~U<4muF?5u<<?MJ{j+@DN^1rtJX?6Zu~
z3>)mTx1dd)Q^fN;X<%r;1y&uQov{9ZZQk~q+kYV7>b)Ck0-#*eMTEzw4h!UqTMqr2
zsaKV4@fBT5cPX`+wGbu7%P(u&7(BDbg`-xEg`9g2h-M8yZ?B`u8ciV;)mJ$s5aX|;
zbaP0s*;j?lS9r6Aq^DL=8Ilo}I%eirdlz4nZKa>Dg}VivrhH=R+e5UW@@AvS%(}4k
zi@J_STet4@DyC{#1=-Z^-<fbf5DwaXG_5mJ7%skX3X{~7oVN^&6euWUJ`kfWL~-DC
z`(_16E^oT5WhjK%0+)SNTXogR9kIi^A*$+1D}(2!wyQUSv8Z&Q{50!?O#juu$LpHT
zJ(h(1ee?TJTQ&!OL9>;RFOJ%{YR1eTwH;(MD?sHZ9Er0d+UMIWUCb;Kqa;B;jS7;|
zvss_<HLa0Y#4>C9Z*KYH8Tn&}!I3(h2M|9+AuoiK``j0F#yd~rzQ?X#*?wWRjELDA
z&JpB_J4IJNy_M1t{n%fYeR<(gF<tZfB8xxmraJL>&G<-!Eo_7l4+ib>`<_Q>MNl?+
ztT}~b2OM`q5B}$hhts`3^OnA#^+&%7OG8O>yuSRF*LRJw3AR;klyNp?*{^zn6)=Q`
zSQo5hXQ9J07V@@hPV|G8aVy;8;P1Yjq<r+K_!EjJ7<DAGhLue{n#?{Ug0UTffL4b^
z4B7{@I&)=;KB9N#=2H$vi(ks$p7krdqwO)V4nlRNVp_)B#IZcPT~!CzXJH;8MecAd
zZu>;!z??Eh4+1#feo0<0#74Pq;x`Qt2in=-_Qps4cKJ|ans_>S*DX_?xi*ieq<oJ<
z*vLRYj0J9Z;K2w%#oMvzbT!V-r7NCr>h9j-#r4U*gMN~k89h)i@Jq2di`KTRRV4E@
z(Jxe(B3W#;+<y{q;~14Qnz_;m2rR5fD)CsF0?3hH_}NMsRg(=fmZDjx1)DAgJc@lN
zI|uLaKfhC=^}zQ51^pb5GYMy0wWniJyCAai5B^zfB`al_QyMmYH+bZU%JR|FiT3M5
zeZi8)H`RHdT7@C86MijC&P#sdl8N(Pp@{^e(_?bLR0dh$NWa?a$*;$+CiJ-ZNs@3G
zzS8&{Q_gd4Yoi_Wa~O69(A4sMH6yUxh$~WX%Ij{%O?X2~)VLe2`1cGQJD(!CW|@DT
zSsnRm`@A{iZZ9Nm9Y8>T!n#E<s6sxkbJudHUf_2&hG-VuYZ@9&Hm$cVlyhwzH~l9F
z^wdySEtcE0|4h^kuM2lVuZ=sRE&CIuZvgwC`4-3Q<U@A452^<#(8agnd9I;+w=w<a
z{7I0|1j__%4#URz3!)XpeIYqO`WhYd>RjqDI!)WDK4jBz>5BmokkZqh3jum<3RsA{
zO9hN7p(%dCZ?K0`_A(Whz|Uq#H$?M_nxpn?+W+;l)tsww-{}thYEL{FXt%<(-adYI
z3#dPLbx?F-LkW3-e7~nrOR<<MeBsGfyd(JF-Y__NW-nYNuQ;4=!89vH$$eSVb!;9)
z{=(!1y>?luon0wyaVdu{Y|P9Zfx{UB6p<1hJ-9!k%D7ool|4ujTA~TK0KO{*8Di6p
z3*Es($#*MatNAG=lQ-2(!?2^1*;9Rf@eN%mjCaj((#GH-^Q7xm{i=l)CdzqIqZN)*
zmFFBK*k4O;ifWOLy4YOCQYr9u@IWd!-Dm&2aHkj1Pl2LQu;f*cB6pz*0{v}YvB<0w
zxV3Am+bTF>pg{SVAa?#(qfz5mkgr`aT~Jk}HPpZ-Vp0zfOMB43$;nd!ar=BnmEJr>
ziK;XzhqJBC`Lf?36$o-RvgoQ_m>8A$jv~48YZlo*o<ag%4ooA`{{jZeH}li$d0e$O
z9{!I?>ia9mp8d5I`_mLy1%*1VY-!=yKRNNw*}PCsvj69qQ8ID@PD~}TerU?-szOqR
zmd3%gaNds{FV@z-63&S}`1XaVvBtb*kK;2~`J?^8Ev39<`OZH>jIS1yg)RUg1<nxf
zsr{SjO1$kJ%iE0fyKY51{bP3zk8eE+;+P(ol<C8L%1IeU>#b39C7KECf90V{+Ws!o
zEEh+6>p1Wp5-W<gmL)JL@qJYAe6f=$DhkH!O?bL^_%oKiuxGAHJU0<|reLXO=W_R4
zj;B=^?U|z!Iu@Xa!%}>86p2pjT3GXQ#cy>Bc0x%D2<W{}`zwcK8T|lT60YuAaw_15
zy98`&WJwJ#oS(=yHjDQ|bu~a5E&L0L6Ad{(4=)TJSOSTVjAR|vNPh%?AQ?@SzhJ2D
zmsU1ku=*qL<rMpDU?;*_M#xANuOz`@K)Lq!l?g&SbGvKtW`*E9tbcE@Y2`mGRR$vp
z-!A}T>v)?#Vx!1t8d!@G)2UI4hE!miflD|M;?mM6lT#hk*o8lFG{cftmEj?{24X4A
zxw*OB=#!hA!ad#Epd6H{I2&atVK`E(o7QMruY!?}Y>Z7_o_kXnndU5cQ%H%Lu=jQ%
zP|xIv>~_4GGw!!`WcF8}1JsFNOT{3c9&qn@HnKBbq9w%8+B%64eU|$AFPYCvs@qsh
zE#hspXb=J-`^2vK6^YHeTqg{*lgGt+*VY9oa_)pxZ)75Kkg3jl+KLl>?GZCo$D~zG
zJ}x1k4QZ-%sE2@oSU?hTnv=r<Q4FIF5w<QqolfI%BEpXV1<pxd<CTImTh+rG0uvrT
zu=OK=<a5TpB)_D1xxr0PBd8)YVwZ9wuvY0&WpiQQ=L4Ua?Ccy4t#Rtc2L0+JD`Q&U
zPqC=bYOCzhp}Y(a=0#&dxxW!G{Wi-RUC&EttWBvgT;1)|qn0ErK!UOzgzrAk_NV;#
zNP(ujhJl+9??XI-i`1{6Vp?6*!vzFdcEdmJ$nSu3=Q~$&RZT(xtAFTEr21*{+~lVs
z>hJPo0dCkWJ_=!W*$T<}&E0(RcMq6bdpx4-w99}a9!=Mqrd)_E?CKsgM^vTsnH&)K
zyun@K;(mLpaLcT^YPP}b*f7Urh0h7)+A8fXsSH~A;cKfzwb}AH9G%sOcIvM}oo={N
z3Er65w{F~j3L_GVzTJ8?%~1kbIa&36`*KcpTHtY$@K@y*>;PKweu)oX>XfGt@E)PV
zEE@+q@+Rgp&^_Y#dgg2=w%xD27I}Y%W#ig~4!%;q=BPRo;4P_neP@A*3xj|91;Js8
z0sKY+e8x1Ns|GDk(osd#)@qUeNYD+9x6En+;h3e-(>puw2N^SXs{@f&F4M?OppjTz
zKl91TmSCmE75YjM+ONOQ?A;4pnJMLVsqTvTiRPR31{0PYt=hUsc@1Evm_M+^5bl*t
zTXN#<%9iW`@Q%tLJA2RTlbVdPPPYQMoqg^#s+$Ib46A+*#hUoRaY0oxrl_GEUQhAY
z4D#cg$n`?Myrh&I<~h1w3G?CRc`m^>er#{eJG^s7V`c)kkN0;~6OsIe02G#p0OJux
z@L7G>uiZ}$qq~<0klSkd+>#e~ymW+N)TPI4FKC;cCcc-Lq=IE);$*^=6_S#^GtHJX
zEgRMkmM@}q)hi+-2<LH@uV?Jpz9WkOOk3YCh-by<#i%$aIjHdHJqw#cBnBCsbofNG
z?Cs{LNe=BxNyzO?dSC;tJg@*S%QP#R3Oo}|X6D{>2;r5YjDp%RtF{sS%Mmk^UGvE~
zuQ0-CGe8iCrtK#RLCZReijGd5iqh}mfgeoufw>T&Za9e13lT@3)9(|&_oL*eQPLsH
z%Uykn=3ayxGhWzVutxEGW*r>dh1N_O3d;8_OGPeGxj<#=#dIPWgKsp{p6Vq08gLa!
zaBN<1YzsK@yrn5b*=!t2lSfD0xg5TH`2u^6-)5vz$+|Y_U{v4A9*HW^q^M%%edQ@l
zOczxLIG?r%UqWV=18n@?ffP^t&yrUyp~osAT$^QcXl2=*DZ;lL6@LQFDS&pl#~sXt
zV~_R0Zz?de23kxP@Wn-hhn^x96z<Fs$SNqPZZue0HJe~)3;cCUbfO;%Md3!-AK*7|
z&&t}`*6nRO2}MGJ-P2%?q)+@W07-QJ$?BND6<apN?gf@6c}Uyw>G9#pg7O{oCF>E9
z$P)80stmzPuxmk*np)Ce=@3?43~%w+#Ty+R9Ui|$wwgQk&8}~*Ke^+Be)kEXFKCA)
zV$pM7N-b=!f!C3yQ?pyU#P@SDJW^w}3tNngp~C1ukT$Pxw9K(~9cv{-gC^+R!r((U
zC*Gn({hTqP&*iIInDKTXFuADR)EaqR^;}+m<T&Z&_`*Iih;el6XGc~jr_=TpWLqd4
zXQ>*A9~;Bo@8@T9t5))Gf$b%?JviP`gSATt=vt6iWhME}g>c(8tA6RX7cXLb_x@F+
zD&-@o#>(cSyU;*9zjQyUDuD)TAg9nJPW^Hh9gZuS8JsP{g#x*j6Z#dU+DVo_UY`zO
zOQTY&WY6PAj%MOnRThmK5DAV0-%uI`B{h=nq=x2aoSsJv%04nWp_vL&cZud7CaS_B
z52l|#w3#wsvV;|`w$XN_;CdfeHS$Qe;YEqSHV|sg4o&oy8Pc|m9&Z+X7!bvhrOKZ=
z!Nz!lfV}m%Y)z^5%l#p5^AT@zRJ2V)8yQ3!|9?n(%cwe+EekZby9al-1ef4JgKRXo
zgy6x0yKLNp1b4RtcXtVH!QI{MedOG8``qr=J>D4a5BtyFl+9Xe)m&><`F<51Yq`MO
z8eS}`aPFU#^LdNip=-GIF{?L8g4pw@pr+@OqF}{AkMsE*2`iGkHj#lNE*1VvLOz_k
zzO?LX3$2o_ORN&1x`M(JUE4U{&sL*R_O>OixNvg_)@XEgnx$F|S2m{#Md)pzctw{9
z*2B&<P#fJ-a|ZfSomA(J96oEjkr?g?S{0u~cF`zUn5BONx>lQwiz7Bk+0F6NFf>y!
zF_FQ$YmmEMf^|vnwr*A;D+NB2F|%=@vy3VTSEQu7lQceLxoa9hO}7;jYBcurHl<6_
zg|Ki<7iwy4o%o5w#BUd&bTjoG^bEuwbGO|<h8S#C+k4oG!XIzZx;it(NoTo-{%D=2
z0JW*n>VRX*Bx~-eq}_x+m6sjflo1qt5rX9EPpK90oh=aapd7KRxPUv|rxmd%2*3TF
z!|In7{$^&J478R8wkql58u=i74M5ba!pHaTS^$IEeBZ|7)ep$)ab#ni^I6&{=<@wN
z7J;uB2{W&ug{UnaZq5CmE0w^0Y5cxjeIG4&K0U%W35o3dS)GE7!y`{FcJFWigvY*0
zn|y8uk;_;QXG$Ly+{520Sf;&f3$<gJKCv6j`Qt-(+qRxOCF&ww$jjDESrRY8poaS2
zMC1&<q|v_lHY5^>YR1l6Ghgr_(n%h1E`Rn6AVtUEcLb}FiHDG1i00ft<(6bDRvvw;
z7EP%@4_xf%MjwHw%pk953B;t8vl;6)M5=G*%fD|lHDK)6^LiAFH7HhdPG9*I-Os0X
z>gO3L>|1qIE{G#<eQ9Ev2`jrS@l_H!6g4IDw>RjplJg5E*d#w(s@>Quvt~#}w}yxW
zwP4{;@%$(Rr;x#V*bR%`ywg8G;A9HN)<7IFty=_s)DTrzq)f<KG-t;H#$qpfwDrEH
zoSVcR0nspQI$@elkExOtjLAv?)7p8KqgbX$uJ=iSUi(SkXWVV?hpO!U)4)WkkO-C_
zY4<~cT7vFYqL^j7Ch2O|(U%eW<YZbR4swH#t{kgQd^i{HN(eqYwYK2}*|g&UEL1SB
z|A=iUO2DhgL3f|hLxV%OAE@!6k0R-OXdfCR)9aX;ejiG`(YqXk=*#N;8oDRT=x#Io
zdE5CB64#RcQ`9JhnU`?oYZH`9*W<3>xGiQKos^1vpR;&x$0_CfVK@A%RyL7kHQYTF
zU&)QZG=GzdE5nj3YxG{8zHHRs@ZwO%8i#E`E1sNu&xB6Jnxil7_Ut*lTsEtaN=dA7
zk}N_7wzcf)Y<i*Hh<Y29fvXFvsqn`USL=dqbvfiU*gto`^Pu%yp$ar!60;L+-vqW2
zz2z9ZWh_IX``{MX)7zc9FlN<tM3x+y`(i<gZZoqON1mhFz0tV?EW$&PL{fYree2$j
zF#8ltzYT)JJl-%}d`aQ8{)5v9Qf$hpJH&Jy8{A>Vz){515hzY;>w=l{^!=JG;5O%r
zqE3ghT#=lUddhYeo_;>d{oiq)tmW1P!YVa8=L%IEdy0B)lF{jB;p5e7ZVq5rIRyMX
zo|<8nqJ!RfH${z*B(J))E7x{zpax+Jt@k>gPr)J>W=C-6u)o74EIv35y>#L^P`*EH
zKh-|IPFxA_83KTV51LK4rRmZU31`Qmpx#ZH=_O}tzeTz6-cb`<j07Iinc&jP9$`%0
zUd<5?nfU7FME#8swVKbMS$~BHhdkdMyV>6r$^nvRX8U7#>}KO|=OaqGqYCJbi~{CT
zpNK+%Xhi;G7H$=6y(_d@Hko(TDOha)kAL0C!Q(F#%*~azN>2_#b^{)_7R$&<BqM`Q
zq<l3P-ki>RFnO;Pn#8RvLnA31NuR=1pVCx)YGI>?y5d0@gA5_jmWyM^@Ns`AJ-*T)
z;#-^?oR76Ju{@k$p(;8$G&F>WFz^vWK?Zai-ZN!8Kq_gm1)GAp;J7WV#oryB!uTs(
zeU&zPq8hXVqju}B*Aw$!&D1%dFMd<vnZQ|IK4VC*z+Jhi24UYP$^h>pLoN^SJy>0p
zxmN>SgopI?CS!Yq@6$26yWeTnq!S--Be!}M6wSdmWB}}eB1jHsJPh62_~35(`k=;~
z(I9CoG*^)Uo{UOyE6H8#XJLh+RD;XKCED_dWFf~9G2^#EACHOvsmjR0$s2ZzD90P4
zt77+Wf46m_gdWv6F1g#ku}m&5V~0$u>W5Ar*oPEt2p@-y1&me?8odL_mg`h+qe7TS
z0e)?k`?wsYp5VQA?3&fV1k_sIJ1)RmT%mONzt>YQV)gLwcs~ry3ztVsQBO2`Oim6u
zFk;~Jl!N|6{FL~}phx}J5CQjr4ZTrzq2&7J)(Jj3hNx3|>c!rj0Rx4k?gl2g4bP`L
z7e`ZDlkirMX*H>hmk?gqr?%x%N8d>618sj(z;6d-q!4Y?1E+9w`YZNtuZ3tJf1r8|
zd44O${Ckn!Awt@z-_Z#sJ@ETWNk^>am;=5W-335}PmCcj(daTc!qM-D1jfYnBaXhX
z%-V~5s=*MJN-}p0KI?w7c@8r$JbVsE9;aTukoWmKkT0@y?n%mTdT_JtRZfru05bSQ
z?B~xfbDDhUdTa9s7(mzcX<p`3&S9dF0~8eJ<iq<d0UnsoVqjv0rP=@f78wO30R<)>
zJB2Ao`@eAYFL;$ukX>UI+sGKdgNKiP(<mke1RgC|K<DY`Y5tzUKUnz}Wd8GmEzldC
z{wF`3zZClc|G!83-@c;?9~5w?;t1w=|6i@Q7(n|2a2~;<=&m>hk%Q8}96M;HJVa;d
z?fPt2`OC+xPlQipPo&R}$IrKVozt@a9?i?1_*dWm+lP-d1)2B&)tJK77oaPrvH?;q
z<O(l6gZ{sUQa9g^5Bx_T0tw-uID$hS3e1--0Y8J52|;+-B!7)Y5L#SXA5FGU`hPa>
zgFp$i>(5Wtz8?VmOyS>V;D4XDKmW50A5t`D=qLk#%71GxC&HVDJ_oh3>3<n)MFcSI
z)8rnth(Lq>n*aYB{6Bjdedf~xv=^94|1Z06lZW1?Xl0wkF!g$Pc#}SMxW)6azg_R*
zVzN5oy^&)mZjs|_<U5CpO{e|KWR=AJPo~$6UT(O}jm<fY`}VQs?<|u<Gv)Rd(r@ko
zSG;?)b=K)wybJy|z*)r1GUlEsHD`HrS-C%mWezB~)!v?OnqT4+aZlgykI27Mu6^X;
zT&`-l3`PWQR^y9w@O<E{&`Clb8_D-lqKVCwzxNqYXp!6CAOCLA4GNg%y4JhfMRHs~
z!@6zBNEqM~NMiYDSu$qm5uo>UdHkF#M`_qH)x#HNo4*`v$Ys1ms<kkiZRp}t$#e6-
zICg$M<Q2OMH%Yj!Qq1ax`+Rrp6|}R^-hutoIyWUO+&Vjdg3ql#X?p=*=<J&{e30b8
zK?E#0NH@YQh?|s3G)J6pW9se=Gv{>SglEoqEO)@r$kS-t@bK7ezxrO4))+%-(s)d>
z{^te;12Od(r$UuN*J%oI8BrUP8(SR&H!nAd+8L9#51jqUEiY%5vp^Gx<=4?!j0X8u
zq+y~t=*FhUZeY1X$9V_W2Zb{q%<RCbt?%gc?0oBlE}ztb<8t=dq^32SMLOHx>gW)8
zP&OYvCz&^k2m1k+Eq#cV&0>1q0=Kfhdc)NcUKu?-CmR%JZS>Hmju?H0@n^}K>0Xwc
zwOei4sZhIcd~q1&*=J~U+syu2l53vt;fzrb4Lrv1ZQQpFn_Q7D-qd5C|BKRnrSvto
zItS5}p4QEzmrS>)?P~#0N+eNkI`H&?04OLV_f4r49|`?+ZGyu+irY?`tBcz+-oWRd
z7nA#M3IP<k5eq(wJSWd1?a3CaXlZU%+Irpv@I_9n@1loGk<Z^hwchYIEK2h6BVOrC
zTkOhPlnp~hQ|#MlW;4_+eWU_emUJ$-2`PQ~l18HG<lx96K5(~1%MdpppwYrf0Yefz
zI%;jwb~iq--$Fw|ANw&OZ?p6bS6(jdxRs=)N%@J|;Nn{%N_uwLm-S@2oHo@0cG+L%
z&6U0es{12L@9I1BD=1Lmw3&-zsOutXD#;^psVF@&yc_{!sBE#nP*f*t>zECja{{hB
zTJ)Q{dYM87aYT$bwE)rJw6_F%*9}C0hg#nZE9ES-PzxPgz6wX84O`<NJ{@U)iP$=W
z84wUsu3gC7-ri-W3zEs%K%z!9;W@ZgMldpcqX+=|r&vxOpXD2Q%lIS-itZjhCh>tC
zI$v9juV2!3T%;D|7q-xo>l-f4E70lR&Uh4ZY4p4=I*~6qs~XW&5|TPfQ^JTnn|LwZ
z6VsGihZumhXM8c<zT46*!pZzXHovO;Mq8UT3Av|hObs&Shwm*K>IFE<6=e`1m4n&{
zGubn;SzDX6kN^wIU9c&Ei(bIQG=4VEHRE7+gXfkTBu5LW6dbqvfXH78SX0yp(#rV1
zmPItjy=%;JJ6ZhW7r5R?jnJ_0$l2nF^ADmqateA<vTNjI;xYoi{~1pOnPxVKhfjUj
zb6jw6y};%6u(+1|o;(_=`hzzPWEX<r?S#`*vc$jJh)%C8K$l}?1n(^X7d#?9(4I~-
zE&GWN1D#^CR1feqBkX)ArQiRaE{?F4T}wZ@!Tg_kd6}xf>SyA;`$Fh1bzA3gU+SI&
zIJJcdAVRoGeQ~82V`uZYs<?X&UQQCV>hdwhREd&VE`4JO%gRWOQ0mH5&n)q{JXNS`
zyz9$pV5k~@Pj^eu{?nF6jOgb#Rf?s`xgaA}Vt+m{&qR^uVClTxUfgQZ@0po8X8{z<
zlko+LCBJ_ECKAL_YP}Xu{Tj=Dg+yOn^+S4U9uur47Q`8)QJss!@QopEc*K^JthRUY
zk{3nnY-(r}uRv4pvz!u|;mTUGV#rs%$cxzqq-qXHAr*636?%2EcX|$@9msG*n3+>e
znIlGn?orfq9~qbxbJg-iGatrj#a#C;cTQK+G_Pk`jEt;|B11M|*122Nf82sxe2PLD
z<5;nItotXF6G)Ps4sLO1xbCU?4Y<Cf>&ugek!>8@kdnwI+9$HYr~CJxP43&I#niOl
zKq4FnByT@(HR~pPXB{p3a)40*ydge5)qRRzwJ8%07?Pf+MUJ_q;oCc|89ch=VP5;k
zOlc?XZ4T~){P5J{`qW;%=ZtrDw;FMp9szsdoa?VUV}^T)35y_PZk_BM-w!u34s6;9
z=>1BtKR<7}y1W<cf%EtlSkU4)WA)vVgd}ZNtU`^_fh6vs`tHr93-PcmsUeEh>#Ei!
z#q0oczcYB&_<qIX!a+BR9!LD>tWoYqExDLsO61v|npow%yEXrXeOmmV3-?0y_rFYb
z_;q4NI92gFN&v0Ekaz2P%E|1PvByi8V|$C_`;mFOvw#Lxa+#^0MGEYmz-=3QxAYpa
zuNjkD>{0GbTQoA?j4<XRMoMmAzD5RfICqdNuQ|H^;zmPBG$m$XtsozA&+Ky+uK2og
z;D0RnC5)pJdo49rCR}FoO*%#j*C$h+9c4U|ylEYMQ(;aP8(ql}#1!^ltvA>8^)Bx$
z9Qn~4+#E4php?oxDp{E&Md=f-W7s2`_}?cZjlfiD=oz@TJUx1uF%gtIQn;%)+RhP>
z@;o=EkS`q4a9Wpp+;{fyHQn#(g=H3Mr;*x{-b(-eSu*U&`H0(UXU_cf`O9}o>VC~H
z_9&`{3JtAAu2~F&7NLwPdS6EnfF;@xoK>(n>bWJv#3rdy=~=3gYU{7JM%gV~$n6F5
zFIlrWjblrw?YTi-#haI|{)W(-#_+H0Ny0W!W^MPVubH(Op|f}C3dwC>32r!)Y&lPI
zzNGMiSw;$>9{3!UUM7i#R_p%0fSE<v!zJ}NOH%I{-^{nAHjU`%aPIwYIBq8lN8g>h
zheXZN1!0Zs2ozO<B|q)rfN{vpbC-Af0uyhK!HXl5)kj|nW{^+|O^OvP!w4@-#vi9f
z+=mr&HP<~b?ukZmT=P83d-;5(ADJV$Su1U?evFm7F2tm`SY8xQlEYKIM1(o;@<%Um
zd?gU6lG4!$Oy(}2kIWbX2dt8|AJjqJks7~5e34GO`x8`{vu$s+muI)dETf=3gyNVk
zd-(LS9zU@W{o3q@puKkt1zoKC^6+qxvmm@rOE7$hNa4K0pV?=i<QUz<WvqUnVmd~N
z`>H7JH+vQf-dYUx<!fF*jnh>8a*Gg@dA`n?2JSPX0v%bin3)-Op4(^K@CgZ?Q14k@
zOaJb)jCnVP)0kiac4zm_nw2eMiGVj6(t0E3yE48-4KJZl#_}dQkt;gJp~djAQ~Mm`
zZQw43Q6Osdfj>ItlVyRYiueHF+!=4DYp&=M2=%~mXH(O<IWtC~Qa&JJk6DeYwDVGc
z#&C;rFJfkB%A4WlI<`*K_F2Q<xVb%bKD(h3M_j6}zQ$s#S~$ejNpZgnXaIGmt^4~a
zL+|#b#YI(#>nQFSDs*wz!`$yZUh-xx*P+m3&^P$@qao96mW`l=2kB2nPxFN}WcoMO
zlTcS%jwCD|KZS;K_(H?t;hBKb?&pIrPgYoUCo<a{22Wh1OeCLg7^uFfSF7{eOt&u~
zc4A6>7f0L@G3tIk2so}2c$7)Ky1BLf4lJR`N3BP5qY{ARTPxeFnO4*{F!EAOe6Je6
z&f?unVCUvB(&d|8;IU7TbD6N*bHt$0sXp^H(eeUoXd!f>yb!a!VP<B=>q+dghrx&K
z471hxJ^VZ8RJg_3GFFATok0a`rL{}`KK<W5N}s<6V_$6Eg_4W?YU-{>^?K-fpFXR>
z-Ldt(|LL`w<gZ6n52&EqZ<jiw9!lTZKjOodvpG1(f+)JxTK5h^+_oym%0XZ<eJBWW
zkYxr>5+863>Cy?p_R7c$mhzEcW+E`~ShA|!DSYly%d6K;L+r*>#Y>=z?f-F!B>?--
z2i0k?s%2^B;#U-k@<iPKjZ~|4(#zG?mRbwHJqsTiCa%#_920w^;*)RNYb>=+Q|@^*
zWA)N$n<hS3C_dghiy`yNVZ*O&a1<Iw?G3fhllD!@dxH_xSbCpE1WGGKqO1-d`;Prp
zcODNp9o-_s=u3+AhkgBUsz}L<nxI2|hNqIwj^pAQD$ZW=MY?B-@Ml!DT?^YJb-+-f
zjNu@)y^3E{)Lh4eWrcl0wG+UXN5zO?N&J=MV9Lt*)qvN*Y26(OyKBmI{hORW4oOMt
zV$A#WF5V%ULoA)j(9kzqPh#hRIMu&;aRyb7wBWa1vB2rt>!M9{O#8B)LnokCVXR~0
zYE>MZQ-X-XmaDwomY?s9<@8P;5U~i)apJ?Y5-lvlZ{>y?Eo#`rif7;0*(PnX4~PKt
z?eAp*?i{)C%FOx9AncO#ThS!JT48Kq^3ZB55{3%r0361WkO_1;%=GlsN+qUvcz+?P
zB1y#Qy!zXYVm#7$Jv!_;QC)i73RENHMv2;2YPy;V3JN2-6?H>hka1TU`Hw-^*ZAOG
zSSGy>0-zHnfp5RcHMAn%J<F3ZHHk-0KYFG|v_2h8P@itUz8XGUB-Pu7*m8qg$|T4F
z70Oh*`UQ5H^gzbEjvm)A8ijQZHyjpiMTQe2iLIE|*05}H1__)$nHvTZnGgq3U+)F3
zf0)^KEO2xiPLt6PTcL&;rS-40tUXeu0=@C)#M<`lofdRE<Xf^E9Z4!|5>|X8x<X*@
z=*hvw;y{gd3<sD`<&|TCrR@agCQ+OB{)SV`8gSFKE%r@cz)%g1Ei2nU>zRzaoeErh
zO(rNnHo7_s4)YK&fTKIf%an^}p0efYN|Drxazm6EBMWXs-FNIkIsA6PWRb0RShhbh
z|5HPcCWV+xN$_Pa+D`w9EpLrX{cLqujyO{Qu)duFojgt>zn8Az<h)ax({JV0@wLK7
z+=e1AOOIWieZA@0QW2e~qc}B2MWjSmx@|nqcsWP0Jt7vh)7UT~1yT1uK6l?ko2Z1D
z9PoZQlOo+dX=w8(y`EgX>ga<Yp*fN9yGbQpsjO(&bw1W-4p>E=c5t>=DVB9dRWWHZ
zqXDUy4(ry+ruMo$!KT3dm_r^O#&Krune}ONhJH__8&joVSwhHlr49h<x^p=7ST9_X
zO@YV|Wl5iaMjibFbe(U-Md;V)jWnZTy-Uwv`@c6d(#wV>zkRv{=$W8;bowXZ)gMJ%
z(;#D?X1a`m79K%o4^v~K1lDRrx*F0YQJzaUGQPmTW^`R-us9v{!(JCTL}o8pJMP}<
zF(N}|UoIK&J|VB-kZZuetIAg(c*r#fuwAWiy|);~o=?Zc)}BL7&=(k#0BCIXyY`@O
zWsPJ!b|rKt6rvZusOr3(KfYVb-O+0O4QG*@uHgca3$5kjx*=s+V);w7v^|87&dCal
z)=rHq^9n3Z*bBLwp?B5Ufs*-<6O3I2fk&Y&yy#3n%B0;UkNUiETOj+|1M`y2B3t1%
z?%4^EEtq41^+qFBm}$q?WhE>ec@#>cZGT^I3vMgvEDw<ev}CU4ERJ#*@Oo`TJ+&<W
zjLIV2;4Sr0Q`hS;-;5e$8HREhEprPOKzWCcsKL!b7m7r36;UOjw~&oVZ%WA0dh%&f
zVsv}88qXVjcK&$|;q-Je`)6k>!t{Ym1q;PN{EsCxppm5mbb1^k3K$uxA!?In!Ox&^
z=Gjs7GBgiIFDL0*_0=xd$@;u<1YHZkFe0^SqDXuxO(5yin?ZDjWYOvNaD)~H(~(aW
zCX*<x16E334_9Ajw|&?|Q;=1OgWBC%lI@r`y5q;r${jG&H>IoI&XO;$mQ@!wL=ob@
z6TrQLQ5zYsTa^8n<LQ+4oZQL3mt_}M=p`E}H8*<~A^82V1!^g_f+_F@L)NvCRw8GX
zXK;Q;eZm$$_xIX3A~Q=j<;&jSnpRA{FYaJOjm}D!d^pSZ;qV7DLwDu#i!+A70|(=W
z#+f9z=d#H3P-W-q@U+mZ3m3(Y9ucce5S!1NyX~{CA%dvlynTk$sP)fSjP(XQV4I;M
zH>N9yC<U@`VYU>KN6zV#@I%sB4gcmN^zVa?SH0ge1<Lc2q}rDfk_Ritf)eyw+iKT_
zd>^q^SDUD$(`R4fNnfjLovQ}W_Bg>T+S=lQuWJ${LkMFFOjxxZi)txHl6uvaWdwb_
zQ2=(Y(r4|{-U|)tL=GDA1`W6^Vv@+Dq$)hGA$Az9es8A!SeUBN%I1QO3rl^ZWz!}W
zga{wkpm_BD+f=v?!?$AAw6|V&nu_$Bt?C{2sZyt*M#XWE@*0F)E~J+ozgQ)~>0WR7
zG<KImpq_iaR;(d!+6M(c_xC#*h#HD_E7SUyjYRIYQ|7TUo?Ll;HbjiKFd@mby115y
ziuPq=)OEo?1ZW6rK5?Qm!!{!cY0ZD3*ny(^^bKJ;V@1dOl9Q-Z0TGs~P=`G-G<26j
zCCa`Yj_9FaP@MBQrx<^_=hvVy`DFRXRn^USWTQ+`c6r8VLnF%qR&{a2C(`MpS_lKO
ztTIO10a|L%!_}|SMbeLB;T2+v5Bs2x1!HI5*KJo!gkMqmPvLc20`OaEt-q(lIzEBt
zyFP*c)#-ZcGE-+8>6)aj;Zq^Nd(&y}rHV-3#K=t1{po?OTRrNBR5|S)UziEbDQfp}
zVY+dhw2?9kRI#OV?OnK$j=>(GuM%Z{MXyo0y00U~dp2~THN+)Rd4gU>auf=QQ4#d2
zW&Oxr-D7{l_S>OlIx~=XNcd07anBDH8Z^|PwUWO9$OQm!l)F_?eJ=u%vO>(6);w6=
zW@dirKA<0t&NDi@S0M@yHO-MN>(y#ZU>L(8oP+y`iMQ&rDbj^~0coK>tpns#Dp*Nq
z;hx~c+gaF|ua5(Zp5`$-eQ>oNF&ny;Y8NY(D=|*Gp>;m7+_iNf1xjgDm&Q5i!eabw
zq%?m{7}<#~Kuw?NN-cOr0;loB2n_mP-1c>-lBZ$M4bNpyD9_z6a9$U!>(6jd{yKJ^
zoy&_}Hf)N|TN*5Uz#S1!*Ut~-2K&q9_$40<*mD^zLN*b5DXu|Rt$Kq?9o}tcxN!1R
zb^&V~C`tqTw9oKNzlWQ@*zmqOg=So3qkzq1QEBnZW<A+(Mmo;GqK7xz*xbzQ9Tyar
zn{3|^Q4#Y0PKD!O6kI_K@8NA`_pr&mtjOXlPhUGuR;RCn%xKitzOCmW7(bbt9!yPz
zj@l>Biz1H5Vy=tZ;okoH84sLi3ucopR>)#FvAhVD&mnWblHy@ilD2PdYK+Umoc4lC
z_c}-34OBD@`he<!MnOH|KE<r+XBg2*iI(OP^_RuuL|qDsGqIcIIko|r9YE<i(vXAn
zbr7jT6AElJbdBF%`Hn8_vt1=G9t(dq5fyc7ZWaXMmo)!Nwvr%`dH#)uohA4;a5<>c
zoF&#Yn453P%L#Os`1d?q>stHK=|_@aNIoh4NZL9HZa`fpHfQbR7tRq740kn1yb#ee
zzNm0cvJ9+<7i;Q>-a~Y1?b?QyY^XdZp0RcD>@#qR3)3ir+c@^Ctw5Cg4jb)T!MTGK
zlz`J>US8(z7%O`8ln5PdxX+Yar8n_9Svn}L&Z0h?b8_qHu0fZj6tfSzZ>AN4VuEQV
zlLAVxd9AuaSN9sygB3+aG?I&6lxCm>)A9F?J+eXlm$DAAh9W0Ag*vu3gd~&vte?u=
z!Y2gXSHicy@nan;yL|neAZ2^i1FApUEA@A)Ia<xO;XK5eH01)aiy<5L#o@O_;ID%G
zw{kj>O`2n^2h6m|h1(t{{6qJu<+LM7v?H|xZ+(R|0Vu}c{@9&i1c=d|wP;M#III{b
z=(Yl1d3t$EO%dp7SQ!Nk8|+Y=?}OX)vHfjxs|tVD0=UqyaaK2=JfEFnHekZ4l}DmE
z9b_TPo>o+-HcT1EIf+5|iq&D}3Lt^ncH&ti;R`oIexmFT4ik69NRGf{jTptwAHt2;
z`r_{yv<EtJ+f7)XhU$E|Yf|TR?)DZC(-U?%poNzO|7MS3J1*?Uq90|&Q#&rVh3_1n
z<<6g?RYLp8btkSK0vip-FqxL{t&-vda@_%|Y3ON~{OdRk=*1zcpnw>5D&zCNNma6Q
zO_n%QRzs~-&;n07yi4|%Iz3+##!|U~FXu#MpTOWJ2D}TEkjo|RsWjKICktREv=$nT
zk|X^uL!ww25u<%?mf<5IMf&8TigmL_^EraCqZY}yL`d4rvE9l@{Iso;gU8<i2JG&#
z?vAX+Molb&myY{U^y_<~tDf#LA^DW6H{0O$#1l;Cj~xpjJe}1D^l)7RgNxJctC8`U
z9<pK^!QFD{bbPM|+4eE}K2p|t!yXIzcwEy%YEnqM%yz8b3N#`)vwVXzhcg8?$?*!J
zIbAw~V>LMz1;dZ31~|QxFr#V~vMI7%y!Me1jvnrMDejvBpoX3J{XCB?{Gqn{rIsCg
zrm$JtRvihMpz)%G;<qE!S>kB?_@-=Eb|i@~6uAJWBx4-Cg(Pv6qgKE2o5$sFj4n1u
zA?rdyaW3ZtQ;2@D-;9Ph(qsFtC3)S#52g?RI57f#a%16tS~kQNIE5JRt4-6%5^kJ$
z!<c^etstx0=xE@Qlk4CPm|DH&M|CrWV?tVlw59;WJ``=CDr4)5k}j=UwkPpTu2SH4
zcyNHLR4$@O<B@`*<8)|G5oTJRUaU45fry)8=Kj!1z2j@932saQ#ou@1sBMa52?s;)
z3O<-HwKV7+0})<Vd4NQ1I^8CHj1Gy}=J;N00r(@KM-~=+1@N{?Mgkp0cPRAS(sRJK
zguZ7#0bx%_Mqv%8ybnNnlfa~fqKw@RMZ}kB#1|XSc3b_Iy)CvgS5KgMECncvP;zwP
zTTkux$s~U%7k5)kY!2P5AZ*snzI_RSu0gTr^1S!3wn5lc#F-Mpuj{1k1;GSstwzmU
z(R2wef?My-HpyE36LvPoDL^X{uo88M)p%E>G&1)E`xUOVO@vW>^B!M2-a+k8HEPlB
zx_-~vrbluc1t^WxNS1~U;`&bEh(=>^1P{+4@}{^Q=)eT<kwyI`r*8mjSB^qQOZQ`A
z89Iv0T`~B~mQi)lvF9Ur(NX*Zm-QdZl)YxH`3@l((}`K#O7=(K{BJ!~w>3=q?A7KG
z`1nJaNrL1~*GdR)uN{MlcW*Bb-j*t`=8v*ZRK)GMRDY3lDPM4wN`=tB_<Az>e5+Rd
zqehI3E)NK^=17(b2hLfM@iyX42W%67ek>7eUQ@lmCd3{#NZN$uc9}5S9c$n{uUlqQ
zspr%J!SxW<rB-eeSrjg$T2Nd{tP{bVB);h?+SAdKpS9w1-cwnNEN#)F)P5CP;*{%4
z+JcuH)Z;T=dW~%0&KwAZhWd^RhMaJ^w$<nuOJkFQ<lT~=jk>EpjYf9I9{UFh2@kxO
zCm;`Z7156uN-2XFMYr9X-plmva&UX<mR!&@lQwI`I8<u2Q-N!cS1UtX6jFsMY@&NK
zT!;Q}!Xcl#ooZSHJNPpcw&YLIoI11-vrFm8&XVwVm9ja|<<cQGNeC3(IQL!~Kguy!
zCiebBOV(-wt=DHk$Xb#9Qsk(IFbb-)wVlW1^|dOd68>ERyhI8)`(UfM!G|iKXheDS
zFey11xHkE^c4pGEDLY^Hm{X3a1#bg(L=!zWidSjIfLNK1GqY;Tqeam$7mRs<BfxW^
zs9|6vIB@Km$lx5qam+FYT*_vthf|P*eCTm)U3PVJ6=Scv8>6!Nv?Ly)$CJbZ{B7X~
z7|jWeqcWMxd3<e5^q!tynu5CjwK))kId@ZoP9Elgj8go7WuoKP9J|KmP4{!8g`(E;
zo&95ulfGz-T%p)_;#4Cq3Ha#5%;w`%#}A{dud4+^01p)Mio)0fJaY>iHYY#6T0Z;L
zl&!WOnAWfuEs)xf4@QVy8==R-S<cy9KPu<iDn@+dGy6jic|>mv#a1b7x`?#F0*j9w
zE&$|h-j^p%k&A~A$W!8Jv4S?WhN9nCiHC$xBF-GHcM5r+O%25C3kG3(IQuLzV>1G|
zv`oXwAt9H5%5i&Qs?)Vf>72TJgeK*IDL!Tvme#<flOJubhlQde_?!E&Bi3INW6wh!
zr*K3=ZpLJ5oZ}kn97S}es-=qy4$bdi5sD8iQ0I!_cY2=bW;U<6=Imq!`L6AGP5#pL
zJO_$SvT9{7HpJWvJcoz3d1-OQ1JuVmU0qxVp-az|g0S5(o4MdKjeWO@)BJ`$Xhv+b
z!}s+<u)cLp<{hX!`3<Esvr49Ikn*FVP8^9ft@?!P9pB9acW+|9?t%LI?n5{iUw2Yn
zjQv#c<G%G+)aW-8nawN;@Q5ZtNq^!AM)3jXa+;4~Rw>QDnPo(C^gSS4;Fb12RW`NV
z<94xH#-?FW6~RfxR1a82NRWe4niA#ZJ8#l}TjbqT5uBid($)0<`RPkOB24;dsK7{!
zbYXab5mNeq`mTWuNj&79Q0^LF0UTe8y3HM^6U!!>q;X8c6e^>`ivgrDn}6OlaOy$6
ze=h*+zJbs7tfx)H+?AKWX^RjV?)5hdD8yG9uPL;#q#&p}w=*60$#Iqi`;18re?Ox3
z%e=Kb46@NUW^OaJGD=5(GdMU*WaSVuY*O&}D+eiEk;ULpK2}QUI1q=Rq@PT;U+)sC
zYB^KjfZy`rom~*J4A*naj<Z_K@I8GbRP3;Yw20M}G`2C#-6TLol7ZB?uVx|ndFI^H
zYOaX7%L^q#D0I;~UmK`*Q0=-;vDW(ZZ>iTJkfW3TwLR&YfRa6*Tp=lWDfHFcNIZ$B
zUKF^WRZZ>pY|sWz=trZ>r!m9)s}urQFks{xfDu6mQNtOiY^sm~N<;jkXOD`qhdAkq
zD_TWksFKTB%dvuDG8EbQj*jZD_?E>)t?1A+tFHHjo)Ly+aUF?KwGvJm2kgPE`I!}y
zSai$UX`u5Ze&~`fKJLfSVzh-F#4hIO@`2l1(6Ph&<m$liW)jSHsj+w6sFKw`;dQK%
zGwCM&<?$=N_B`sK!4Ss9c5G?B(^PMI2`Y=`-|nTJGrG+s$*?46A1Vl)H_HVSDCfKq
zxQ+xExTFXSCp!O#KyAHLyTny#8TUx+tm;w)OMN)BrtHMb1c$YEJaZ4D9NgU7yU}k%
z^=~(lMB@Uc!inLW)SbmyAAmfBKe5;i`;s;4bC&KXe^>rd>Fzf%xKY~O*z4D_27iLg
za#aI2iMZnkyQ?2_<*tMiJm=!=!UKqq(m<w?n%+0oO&2KVI0%qJfR)j)Sp*BC!*<&R
zFw5wFjLD-LJj?jBi8MuRUcwiQvQFm(h!bh?n^^Q2(&&4L<7D7faaCg&K8G*C8SY`l
z*kI8LIxvPg<{DroIA+&CV~?MU)?p15_0Ceb9-CZMu#?Zx97l_`oEIrFOx!WHJIX8a
z(E~@3uUB5Hifoak{0$t6KW1O{xs0mv`_y#z+z7@@dYg#_bD5liCZ;^{T+C5*^a{7W
z^_Xnk{15qQRJMV@Z`ZV0|2~#1Exy>()EtYj-ShKv7sy?-delk^bw}WQ=^&A_e5YR4
z^fBxEmXs2nJ6$7s7GntAzSn_m&>TuP0{n-K>R0WA5?Ldh0kop>K;CD!Fh^M&ZGP1l
zqqLmR^IVF+$$6MFmkwFRvm_E{()MPw9GWtY3^<}BuatlZ1%)05HWEJcZ*G~nJ=2zr
z50Z&BTew((8IU-i^jI;O04VLRagDqp@Sm;E-49)_9Cfhc%P7^59`BFNx^C3tKly@_
zH-Qj+P3R_h8;lZ(9qZWl=t=nE4$SRK_8fQp6U=7Gu#o-Ebasy;8~=(4-rJS0j+{+~
zLN}g$cQ`VBW~+qsr4BlPRRZ6JyhU@<r6kHB>E#6>#o)<jRn6cm+IFt&^2SD4z*-b)
zT~v27Rwd~6PIa>rqSi}XyR(ynvLa9x3pyg}dS5r!-%nT^$#1m!aYt&!Tu7f`7^sQO
zKA9V}Pz>GlT%r9Yo8S+Y-p@{Hw!A7{s^J(Bpjl=yhKA4rjw^;p{nzD^0H}ukfC>XD
zn~+0*(bK@mm?+F`v4(v9G^B82zD`b*cP%0TsTQ=M*uF{xX$OgjzTo>wA;2!6S^7b|
zevd*WFjm}F#X$jW1$Gn0FxeF$PNDDPTFshVNh0-HU3ar?<S+SSCv^mN1#(5cmGNU-
zW3vFSRz9ITIOh6dU0j(k&E|GI+J(x9v2Cq;#i&mO%g3kA4(o*{Yu4<FHUKRH3A44k
zy6peFmt}O6yCIvcpu0`Vv^VQ(!@*wQru_$rh<4bPhfiHbPzlb?g}YX8m9V)*14y3$
zKx(w%o6tDVCYImfw+QfWAjPsEg`$`<@gwY#PsrrV;uG!><_}X!T;nP);TyOqf;I^m
zQ?9&doTEkC131Oh`_`Qfq@`Y=u~UOCns5`j$ud5`oJ+g70KBrIhl}4f*_kfcvjlc9
z-^>NNJU|-?)WoY3AL9q%0mV2%O-+;zo8TGkiG>vjdHqEn%mzi94)5<HGX?u>djYiX
zw@=$=LfpH4(genERIwUFRX!sTvfo6v@r{A+nx+j15uJ=5`~+7H@Lk5=mu3=LXjyKS
zP=$XwFCa9T63y&#!(a+p+%{QknCWtEhmFpPTjb$SCwef*`B`HdUp-cyX~gy(LT8s~
z1Exql)q#0tKjdF{yod@OD5gTpadj@=4~<m^&fp%#wr1fgr>}e}aj!7TvuUs5WgLTH
zA4$L}@?tS`hqm$Eyx7;q&SARxBlpgR`swdYrC1eAI9n=OjnbX4G+jw>SSci~Db5kS
zP*_3ElSSfPG~zOom6-|G=R4EaR2R9+i*H_tLdw60!b7W@vJ8Hf>4mS2#@HQi2j&Jp
zx9)jhZ4Vh0Tj!Fz1TvyIv%<D-9A9X0pn{`an7+G&BwplW3tTrkxBqfMVjAVEY3b`)
zJpY_(f&oyaHci(=00lzG2O>>BQ9JzRU-?l+l{as(I2AvAU64wgr*s|prTuPnRoDi?
z54zh|Pt(@0LQI&6<G7rMfRah$6m=!k8dhW@SEV8|(W=Dq%%^$6bHxv*BmfN6BqV7A
zA$f0;(bN9Cu}_+Y8*NG&=TJu`?2RBQwt^mMvb}|5865WM>gKsBt}*lyN7J&zme8iY
z&W(VS%*JLm&=9f%cTCEeKonZ;yHxv40H|j9cw%vwH}dlLZ-FUu6jVaE)1O-atQ|Hf
zdk_VF=^~_Uo7CVt$d13O6v@f>;2Pxp<vJQj@4AEyZ6Z}^=xy@CLVb$L$rBGL+u5rj
zg~#=MY)>{JxRb>L(hujD2qjZ@umKL+r)d)?aSWVlJji!Y@-`^a7G=B3IO;wL4weO6
za)f}!W7l8({3(z`tG3)?``o6@UK5XSfV*UBR+)?pIr9OoFsxEXOM6uiVE`}8R+$1O
z4Vp}h$hSeEb|ycCNoTXRjR=uKMU{m1z)UfV-w2Z=ajYaMdxW#e!r~fE^nOfjX!uGj
zl<s`!9(@FH^kPUuSitDwC+CGi6W+O|o0nKBn5M>sA)&aWs#7=X%Jo`XEREf{<Q2cW
z&=n<JnKzzcey1pb2T+~;J0g43|B_y6t)$ZwZW>kp*RFnguZyu%eZ2q}!ru_a*AZDG
zhn&Zvt%KC}JkUsS%;=a`<44Pu83F1o6U5r~gF_WF7xqzu9T6-QPqrgJaZf&|_${t^
zT1KfEkZD#gd`+5qMHz`_62@o3s$9NVw_mX|qd~hjI6|<O>8LQwUqY@>zH6D7a>>I@
zE%-voyBU3*%vUNM-hqpvUqpqM?hX1hdmI^{R8<+jzl`aGQeE9zR8s+N8;NZCOkRQ)
z_dqsX86j$ed+`>w^SCZ~btz|BtYplu9gEllf$Qsf#kkgTbJ@>2N^<omKE2`LQTslo
z5^`s7@S4z(hICQYOfPQ_df563x!?gBUF_f~cf71>$NY+fZ^mNKD~N#l4kBIpW_j?B
zdZuFw`Iy^x=@5^5!p>rJ&r_Pe8(O!{5wKUMx^dnS7JCdskMbCax!4;r2y$58ECm50
zO5gT?RARem9*gXgo2Rwy^?gbSf1K)Zm*A<rOeNI~mlmO0d*^qKyM-b3LngQ9TGAN-
zZdm=&10Z8-?o@lZ0m|Zis|>NPH}ea%AFy#hLkTcoTYpp|S`|CIcyYx)W&&o=ull-w
zPx~Gr{=y~dQ*kcmkx8z2PftNzLZpR3;ou+HK>{8H3f5FD220q`Fj)iYQMyaYexJBK
z+wf!zPC%)(+#;ZwR%rX2(Mhq5MxZ@a97X#&L5k`7o_z82Me4Zi-)5J{Q@@qQqHop(
zlN?#=eak-L@*NUxy`g_&b&YVMqvIrcTi?xkp#bbEi5zcgTc*pD!E7!4W|(sRp+mL(
zc-D0q-<(ukN7v{<t}VuSf7xI2cSpsBn%1;&NOe5qtxy{~j9Z!)6+Se}WdmKRE7Zh<
zdyf-RIc*&_sm&=pGh*cebTx6-=EG-uLg*0o^}N0Ondeg5DJT5e7yhkB;|9Em7bhCd
zT+(&JboMC6qVk#5r5<W!;ic-G5L#TtJcpkB5suvnji_Asrc2GPmaft9Y#<=_c6)TJ
zNY(KZIYbnmsoKDin96qAKoB}fZ;>ouh1Y07OQjj@(fbJ)C;~EUMnNglWIgf8#YA^}
z2l1nC{Tj96WDV(2;Zsa96*C3Q@z;_IjF9R3S3w>ddk0bB+8tC!d?ho3OGyjM{DCke
zzENpBGY!<cAE(t?0}ZgZo#+x~xULa+^M2d?sjwQ0>ZzHwyDY1-C@6hYeg5oBTKeTx
z=;^8X+GZ)^iITOV2J2%p4_sNvk#m~hqK6$&z&L9(flesn7Ut+b0^CaDdFY@2dv{DP
zO=#bs<2MYOkcd?~{3t7fpONWh>h@C5TdEFIdTS6C1&mb?AD8f?H@fJuE5Z>cUcO#I
zWyA0`9eN{t@D?tiM~p+s?5v*Eu)6bGgS6Yy)f+Whm5pQn#^zvgUbnfunP?CCp2dlc
z#{WsC&C`JN4Y=CU+r60uPvE?qo_)UM-*G?aJPJMCFF&8R29_$+CjLn2%yYZBt({3I
zq#Cgmb)+(5XNoMEqOM6)QY)+YBr;2{-Sso5w|Tp3ol8ls%M6K0M)n&F1NI;hiv+zz
zhT{7mSh}|#76(rSm~2A25lQh+KeWA8%dXJ35?XQ#RH@8NlxSRK_Ej_vaF6aJTz$lE
zZO&)W3INz=7j(~+4iR)jbd4UoI_nguH`w)LFy3$-nMou|YZsyR8G0FehN)tC87MdL
zfC`9Jy|j^Vohm67NKvzpT_{u^X~eAuPI}qJZ7S9U1ccTs($)#j>=hIBk}R?Bi4w<)
zk$*{_r39snL3iV9%|E#NuqNXt0P-f1WCmO|>=DHu-{DUr;7P;Wh$8`Jwkot4IbN+m
zMSCTb?gVUzm&R2=1|@>XUd{LRR;~4oG>)@vZ=m@o$4d@YiZ{x>f(1O)JT90kowL$x
zxu~a=<=)aZAOwzTUVT&w7z`54u}S^*I@10?wE>ekOsd^85PDk1_sqyYyq(Ve!xT)h
zc8R)%*cj;vtgB`1&e5Ikb6L#zJqGYj(kMeDiF5Xan@&v-Y#I-o<pB>>Q2I6ARVE;4
zG8VGyfw-cg>s?(9t-p`oSkU^o(L)0;$eyo!V2x)U+)s^5DQcu7oC(0#h|3BccXS6%
z`HK-#90IDukw_Xpg(|Cs;=CI!3m4FP@;Zp`>3cXvY=|(?NP0^7uXKGYy=7$5X`&&U
z*z#9C1S;CiioKHZ`ekyQ@0QZ6y<}ll+F7Ja=j>uluj-OMw%te?os|~u7GFJAZeYYV
za*WS-naeqVCrdf7eN6_6Da!;RAZ}f<_=@sPq(pA(jTnUJtZhZ%W+gFEBW4!X$wEb@
z7H7oI&L~g_$kr?+MoP`Z*<o;6GHxkw=&v#Nk*7uQbytRFy*hOI$?X;n!NTP(C7UNC
zB3>1}JIaxo*L8T>_OmkjC?Nmy0cZG;w$N}@t|1>2nj<qb9x>nn?WZcBw9g%n1K_Gc
z3XGsguxHaT9$(%6JZp2?`?_;?7FVfwtNCz*Uvoz_z+;n8UEiviR1R-nThX^f8<Ex(
z{429#sU52S5Qq5U5%2+AcbvOPw9;M!KYhKeUOYhf&ixcE-~py<g0WKtk{p3NujpSg
zi#9n5@6ZmQ9KUyDmF)v*DQ$t^&GyU;o7}|Yd$Z8L<v@R_Ds-S#<byfHIOnY@C;!Mz
z+InA#?&c1)Uf2`)e+U~$k$v1Vg$WmgPSkZ$h%2@yO9u9OP`?dSiyOp!zFf#%{I^mA
zXu-W0{@~zwNVlTT$l2cGs<eY_nfXP`(xyTNXn%&k2n$3<xx?e?y$m4n-$Ec^^h9=O
z^enP-xfJnPSrvS4|6pNV3@YVRH4#WF|Bllbm#&tjTOUD-7=NYj)7A-J{~-?{J1M0=
zcYvJ~(~&?)y((R!|Eoda?=SfxnugtJW}vB-e>LS}jR{e)tluv283RHdzp-+ZJ!2vA
z!_{=#_sH3vAce@f2Xu|&_7Lmu##Z)`KjwH{O&3A6kA0^XA}M5_zj~OZM&Ut<5B9e&
z7j9F2|D2Lz^&exWf{tp1=2eNp+A3VgYqt@uKOp+Xs9MBadgg+zJ9Scqi}9C!CXoq<
zzU6SAsjPw0Ms&fK0`NatB{KLSd`L3TbZtt`FVAsvKIyIK2WOxwMv<}hcm4Wo8KD!b
za=5|&DH(A`hW^?Lyj^&UFye_vtCy&c@&Bb2_(RoF7Nr)AJogr1pr-W+(*F#o^PaR^
zQrxcO#L`oxiw9H(N-c_6x9({_=7<2X0iX`Z1`=UM$7QBGDNDaTy@mL}0s4m|$A?}l
z<A-9tVWDW>VM1~z0o!1Xbn8MUc$&Cc(~08}S426mkVEq22R>Ek0O7*?X`);mfLYj!
za_1j{pJhHs%)?_XQi!w1e@^bd2JQSJG&(><{_i0DfBEWvP1FA__xZ2y{H-?n$1eFF
z>YM*f{`8kz@I@096o<wE`yzaj`^TdGpN#-WaL-@|_J)B8;=cuXfPO3y6`E@G*5Sgh
zN)(v0HA=F54flf$>A#g_f8<F%h2p|?IdaCe_?$99*p!Ju|3f?VXEZe!=%8+{EKvGC
z#s=(2XkZnLAv3gGb6+a$|D*B$YoPcqH0vBO_P=ZNKPXko1)6S8v;x(Tp!`SSZCX8=
zdA_RkTNj=&F87=SlQ&5i3YB!rTvkSuG7GsSWG89xB~}MY03mFDxN=@@Ga7d3np9q?
z9}9-|NfHyjni*sHSsgg?TlmDvCeS=|P^EC%^b3nZlI2zP#$RHmN1pUMdA%6cI8&p0
ze$Ry$aZHkGbmO?z=zl7iz<(*3hRpF~8!5>YnV5ggZ>4Qx{jv~lWnYLPH&mE#b_$()
zx~iS-<>j9GOKK!~0*I4dO)5GA_Z4_bm$ryr_Pl;RpX)aP4uWC}r_6XS>Q6Vlzr}h5
z35NgF8I6{s9UnZ)&HW~Nq3z@waZ1Pga^F^6KepZ$zEE|&&*Ek6Jy72QUNbYXOo*MU
zp|;?@V#Xd+Ddslian36a%}(T$@0|^ryFt6vVBwZTs^<G9qxwBLJ>QZ=(r-2DPC;i+
zpsHZ1DYrf2nY`8GyO6oZK9<3V2<p?V!lyIP);`B|S=9%0DJY0JJx7(;vP?6e@@!vQ
zXE_&?yEg|-4qCgE%T&p{9QN?6b8(a4o-sLP_`afSr~Vqo%b-#ud=q7q1cSoZ7wR%q
z?@=ij8+`{5L#W~r0&1^-v^3Y_u!E{*mb?1?7i};qcT8wFr)Ej{iBM}~i}n278%;Mr
zF&Q|+@eLPH)^WeUaW8jv;Ev@R;}phV!R@Vp)zMSmdpNM3XYRJa-4X#%DT-HHu;7vN
zGn-o!csr$wRgjxoLCa0Z7RK%*JTm)f<7!DEE;%_LFe5stz0{x}8bPjuj8k^=j~WHa
z9>#t@z8e?a4+i?ZBlf&~G;%*d_UQhjI9kJ^f1_to@H4m2*P9yJz{gSf?^=LQULH_y
zc3+zp%I0N#7p)ZE*NWH{|64fKLLen@ncm_uxWPjV;|GKOHR`@mwxO9XQhB&54FDM3
z#mUMdkb@oVQfOvb)J?C<A_%;XddhfaaoPz+fp@7yf|xoN?*=7J>q<?0ypn2fF0ZWV
zUUsoN$w#Nh)?tNUjo;p`tyI#}%$y%*Ffs=caoMQX4yCN6Ej8;87f6mICW{sfi!eYr
zhQsNHpI-qhGKK^A2modLJg>Iwh~#DKQ`L$bHFX!WbZf=(CA}nz**=@qHFoY9%pNOH
zX?ts<8h+JoGXkXOu_T07(MiLv@roP))*5V9^79bvW?rksL4*vi%TQ6pq;AgPPBK2W
zz5Mx{(xh&RtnH`$x%jBq-q-o)R`D&1PkkGJ&UFhT>hZD^2>{YtA^a*d4Q);82OlEO
z+wg0b=nZ(`(+FP4BxJ@V1Lrrg6i&HgOANS!Rac^6I%&ke*2V1O8<gk2Q$RLc1G*l1
zNI{>X*tFz}$Ei3eTd%xuJ{za>K|H$i2f4*Bk{$ElFlurT4ci=3GYD`CWG`+ZBxvg#
zX8!lBkI`Ea_ou@tbJ%#e3SrYFy@}z?LYIKI+8{t?%|9A%IPE!@(^CKs2y;x0w%vR&
z^?8WcP7$-Fn0ew>ZQ#mTqUS_~xArYB6ROny<$=7%pu=ck{UC@vFgk;(Y|^S-H_JHY
zjlKm)XzDI(c?s}7g!Z#zh`a)#2#WPKGzdbfpm4q%aMp`XilZs75XgxJHUkMjaz*hf
zMV+jkJ(|k~D?UNjLD?VG_mn{5ZCC3@*)3ZFlsvS_JJ@k1`j9z}F^}a2b?&c)DqjtA
z^-?@@uZ-qhVwbM(H>||p=ow@G%TK(5aL(=IC%zqtpi;?44#9(dH>A@Kij<k9loj@D
zGY6DX-l^uhv6fL}TzeH4yvuT&fR^=(Iw0qsb$SVN0z$|CshR2!`=gnHXHhVt4u8+X
z?vG<1YWBnB@%HDs7f&2Y$u_RJ>_Hy<o7KXZOyG5Hg<R`*dRgJbv`;?^i9BxoVEt&9
z*KS3lm-Xn>Ouu}IxDE3>4(2<gF(Sf#Ez+8jLi99jl|G7BsHR6#7m=;^3-3`kL322n
z{o1x`WC)K@g@w5S2bmUP^fNEQ%?g5*X(f0p!y5PfEAiJYxf<FT^|!nOpMDUL67mSu
z90@GWI0-$5cK%En*u2G<*^JLI5-UghM2=|$DT_+XO@!VvE4I(|3x<@u2v9Te_~T$Y
zt<M2kB{B8P@-$xeai~RI9sj_r&TAD56dDS2zAm%gD1c&h<d2;6G_?&XF~jlrP=Rq&
zuA-SCQz{<-+2a^d>HcpaR+{v<s%R5<i+TGuP!yYAe+#Mgyr`YZMxb3>ewH}=*`6rx
z<ZjLR)8h{%xI0rb<5(=#EHpGgc{FEbDrsaK>&5?c2X~2~po`@U*+F@7beD4Ug|$$l
z=tO(;oM+h9Wyqc?8wFD}vN25h$E{gPx%9MR4_WN0M+T@pg|#V+H49Lv_4chBh0dc5
zJK5?*{tRpZ9{;Nr*j+MpBL+b)4o?wq9+%O9FKuiCWQ-4(&jljy9!ahTn|qzX30j4q
zY*78sKF?D+T_?G1!;W(Lqm&2ud$|VIiZs}>khxT`P$koHdQLqW3eq0K3L&P}1kX8j
z)r9oBgOk0yy}irT{CV<uk*@QlDj;EN_slqyoz-IB=BJOH1kO{*ZrS0L@eW=eegtMc
z-f5Cv3GOVMPn5TGSIkV~+WtShy=6d@Th~6W2+|=SAf3{w64E_@k`huwh^WMn5<^Ir
zGzdz|ASlRyNOzZnG!jF1ck|zPp7XrtJkNRG5C0FpFFK5P?|bjP_FC6kdtGbk=uK_=
z@8el^+nz@EM7`9nME*AKL<iW&T1=KCwTFw_#6OTLGXn9P56L`NoM>~LUDlUxg*lsU
zo_ae{P1C$tEF=fap)@L6o*p)6TYulq-JszHO5gKRXK(Yf2@TvDw~I*05I%K2y<HN|
z{4e|cxoZa(Jykq29!}2oYeq&ts`<k!HJ{+2hEyvLlZDso#M3AB?s?tkT6{hrQd@mV
zGVicEP*t0egq4bF8}xB}UV1KN$qIctMyEq`$h?LvTnkO5+4-gY6$wh2FeCh3S6;gr
zn!#{MsNhT+dWnY~ge~G{qav5ZV~<}=8lSG*pK`7(55uDA{^i3r+h*!R{cVY9+<PpN
zFi&fV>aKyBGJ;uxFzEWmE*y}xS2|Z))NVx1`bm0rwyu-R-aI00a&0YByiB9<Ug#ZY
zJI643tAEJ<+Fq!sI(0L4IJ*i3B1GlwMrruO%(gTukJfCp2A=IIoeU1qP@vvCz^q6;
z_N|~J>T%p-9THtTFL8P=gJ4dH|MvyvHl{L734R7O`)m8xbvQQ+Vw|-4(psnSlT6(*
z7DlUb!<gJ(L~yLx7!ydD0fM!M>uxOYRuWh-2i7BCy{-zZn(1rp&m6vgtLe&!8=rB$
zGc5F^R!~aG$=Qoy<#9pTcZDz&P9G!u0P+})6@ka0Pv1rk?u$#@KUO;ZDRMNMzD}0K
z)le&I`lb_&r^!}YBC06U+v7I1dfYTU=DcZ?6a(F-rlJRjABQp>?oNwj9YyK~aS%FO
z4`r2<j_C$;D|7!WM(K3BfQD;Lsl?Qw_8PfEii>Hd4KFeYs`^2kz<~rct~a^mj~J}7
zkZDtkVY{CFvgVhbth9PQY)wnTr2B^A82r61s>uj67wo$_K@={wr{dXhUtEO9AGzn*
zn2IOUapps;4hbt!=j40sKEC4HzQa5AjB;JNhF|h|BVXE>qhhyOH`}nY(-`&>8q!Uj
z&3jHqvuP(#hE;*C+l*=&x)RW-hYoEPb4<dy!X3R%hAOeC)z<6bf-UdE&mXi?hQbQu
z5%h9Iq~gYw5&?tz^Xybq2!@wL?DmUQ{YW@3V)<xz=ErLn#|7!Cnp^B6x*1D9v)o)B
z&j&OY?oyC;3iBcu`lS%eqf|DN@i8+=POwKOUShCVNXWfK-J#b<`KuAkw+?JL=s1Gk
zjIHh#%(T>UGGYW?>%BFQP_1<L?6tjDFnU<3^VkR#=b-$nydz{r8PX`#D=Q-Cqamv%
z$kBdTSlDssP&(yt4e1fDC5XiaWn^J?@DJZDu*Y;K{jP*yUXEo(y|KI<?aMIiy%I4o
z4TJGuwGTP_6wYrCOO7sSCLk8?9+oz-NbkAN44L{tlzt&s#^d)0s6T04Gv<R-*)QkL
zL~sMl$tZNZc;yz|V9-4C`EVB$zL`&Y-BkFFy{Ht_H#bP+uSNb8dZ5=nJG0+gW!mXA
zpwFc{Dsaaa|JLLKFC5}HrJ4Mi@89ZaF^Tlr8Y=K{plop5W&v=u(~4`#mmZs0D^@MD
zcW7RZ<P{0h(Ce6ab0S7vz{1Vth^a1EnMWB9Ix>oyT#I}+ZPG@rKuttWM#Y;-+LOn%
zQ)G?KfZplaV$;iND7O7reMi>Eh)Y-4TQ%(Ksqd}{Gwy9tZ|n5+GJkTF2xk_)my;6(
z4#rQ_>-vI(^wPwnD!v7`4tu0>M{gZI;%OZsq(G|D6(5<bvw(1kdd(COUjbV{1L3>J
z6jWO|ZxV30c!0eSES3xPT6!pXE|b#gii3HMhJ$-o58ag7jK@45%@M0*qn%;sO(|{y
zCF+PKRN+S<psMf|EB7YQKao1iInLp{Z!_TIGRmyE^_2p%gObKdE<^p9ig1>CqsM-S
z;;!Iia``ts=<`mF;1I=7c$KQd)yYVBCzMmMl-F3W=V?8mdAS@?dPmP?RuhATGT1pu
z;kqp~z;{Z+4rB1l#Rjrzy1}xB<Sz1#Dq6s7?$#Hk*rLTyan%bpACCylqzHI0YQCBp
zuNdJnT-2XEkKrp2Ivj8X&*{R8*KBGCj=0{aJP<Q4xLFva<T+Y+!D=>pjTcFwhp=5g
z+nSl)`*F&6@cc)WW>$v+=?Jd|!Hhh@BoZY>Gs;CGQNXt0OnW>w0{@Hy)3H=5{JKWD
znSXSe`?Z?CfoO;O73|6aJ!ebWw>;rf$T*wx8my|_kL=?Q80KrS+{6xn((K^V8W!XS
z$`w<FhdIXd8Y?Y`iD{I)<{wVpY_7O&b+|ZvZ+bkiTI0^Aw?TAQHE~Pc2sb;9K>1qf
z6JhFc4JsBEHc36jrZ;iz`S;+e?9r2*Ow9|EiF5gJq^$_JdEX&ioKEPw?zjjtZoI6A
z*6SjYj+ejh$v9Nd#e2@X$OiH;IBsz|YO|c4FG(n20KT&GA)r|0&Ud#vQyGszA-yVd
zn_W@C3Mm7-TB(6+9nWZZ?~FH!w9ZC}t4?L5hp8fRVD>|)qC+%O5+SE8Ry3I2f%@gG
z`;bC=+7gkW*@$g~XUrZIX|%6<dTUwb!`Y(AzLka5bDR^ON>Ea7mPwMYiC15NI3@nn
zU?zs%UEF?kN(9Y?f2o5HRw5z)&|qXqw;1zo`Hh5*2l^9s-8$elN56EqYQv<aVT|S$
zZ93tL(v1L!(!=)>b-MsgzW#W{nGzX#+C78K!b~cme%CO$ru-e!dTl2VEdWi1fPI-T
z)j0Mi)oXc#S;x9CL9{P9xVrzQISz}h7n0eYX0_+VRQJfMwU+V99~vnWip~*DvcUM!
zHMp#IZ|`t6J;;wbZpgOb1{M9}a!HO)v2`uV60KFG48KxxYdT}FB5QTmp3{Zxn_x+b
zo#n`Qw`$Edhc@mPoDB&KVmeXGT0d$AshzE->Tnq`8g$h2d$H@TVFrP_`-G_2M#)yg
z;(Zl@p}3;Vo`Kb3MF|&A$Fzx72H<~E;ZA&_`(gSOZ<V~K=6#PS4l0f-is`XjWJ7Ke
z+~3ZSNK$3Bz(uM&Z2KA-cBp#PjSY2|llgiY0%=xhg3IH0lczVODP(RRs5aXxUT)C;
zFuZ#&rl(nf#x?sfV>9gS3i-9#2aVw(Rc%w)*WJ7sSuQ+^ivu}iud^-pL`uR5l=A9F
z9|`WEQ46M?dzBUj-Zko+c;~2e*ixpKvp68}#uML>_!%Z?rfq$qi-DV2EvGh_xUAHd
z#rTF{m}g&z39>jyiI~l{NaQY8p<ubZu3SUd&j~Lgb}XTsCaV;~v9qQ*(KhM>vpp=r
zKE2WO0DV7R-<(DrVc&F#w}9_4LKPdqE|%7E@xck-*_*suc~umpwP9h+fcJGQ9n-4s
zU_<RupY-UB$Y;ryPW4q!Q6~{0{>J2~THd2?_zDO>Vp6gQlH{i_r}<?+hJRbj!Qmsw
z%n!K3H70iLnQ%yXZ6#uivvtS#kg~6bti!Qap{<XNW(OpBc2)Cl7_s9>cv-}5gbV(f
z%4dp*5ou2jA||oAKDW1(RtOIn7HSwVakEMC-l@oCu9pXQ6E{w82Q^c^7&_Cd6qMwD
zKU0bv9ib8`kf^C@T&)QtGU8mVJo{{{Kz`!oE3oy;N7|2v&qKBb65L)cMgS-dV#x?K
zo457@(cjf#yg;$MK1aB(RFKAPIYJHW>T5$3l75)g;u<}VH(((uzDxU{-XI<9lLDpS
ziS{A-L0^C3xX<9P8=iIB36d>U!K6CE_YwnVaj6^u1tG9^?}7(}?oc3$;%1yC554X3
ze!%S0H(}VMQAA~eC+zcGlSE&i3_MoSCCf-*#q^GNx<K!NHPbHBAVb_-O`XF$?q#FS
z#>P?zol!kMcU?G}E6P5#zyCy0WtJ$Uk@;8Xg>#q}Jwd3q&96YETkIz;)Y5n6zlq75
zX7n~pr?j@LvqTaaIax$Q$6r!6+0pyaUvTWCEPbT=*vifNSR>g(gpE56Q*E5+Sq0;b
zp>$(rYMum9xa7TQEH0nV#Nu2D60eVabsU1_f81dKL6>jy(X==~hI3GY(KDq(w(A#d
zI^RF>e~sn1&iVjqlb4}h#)MAV)YG{p{yb*=!2q&L+NylLnxEbFm0?CIy^P2P+@z3D
zjWAf3LM;X2bNxCV_PB49@>Zy0Lm0zmG_4nU(^?O~&zV*Jr0Jr-*jkHZyOxs>=Yf~A
z{P1o!Y|@RfwB&Jn)587PnRzdsmC??Y$rqXHFV~Pt!@bd)eWbzkq*A?ZMhnyOm>&sz
z9rMWR15r-$>EXLi_2@u?<;)MatUXME?4a32_kj>QuB}HtZ(QA(Oi&;Vm%FzIZpoSk
zHJ@S<Yu{NB#J{<vf60wUY*ky!8f^xLy@l(3yJ-xOH$9{tTMw2+Fu&hj>H&<o=b7j(
zIg(Mg%MK;PofY41MQx&`a+#KM&6s4Sx#c5OMZBX8oU(Gp3L0A`b7JYG-Fyc;ub+x7
zRMM(GMfz<nLoc`RWN$Vqv873q`NZMGfDGsGUAts2JomfZ6s12|m1<?_Tw0;tmNTAC
zmv#>69E`>(H?(Wg(hn8Ih00gIU6sFN;)piOg*r2`Y|!Z4f^I2@h*Jx_QGV!El}gUY
zg8u}StqQ_K_8{Vh$L0#5rygpv$TXZjn)d4OD*7-RA6mh_+`(9el3AW&;UJXKGpi9!
zxUndO+j(f6v!hyywWF+$tVLnlJag2ttBNw=54%wX<2p%rwr!i`R7J~jASK1+S{udr
zX<vbcisO!XR6N6pE&LMth}NC`i?c;^S`w00^knBOO15X|<KasaVs$P0yH8Xlx0<${
zq>7wz(?LC(6Zg{3D(I0K%nsfn(;UQ26WJ?$YZ6BsJBp<oMH-krOkwQ{8C<?QKUALl
zI7t^caa4+b6Pt(>99AFnVjafds`VoY%H~fQ@G`f;r-7Lt>D$6#k3##YqIVtvO-{Ef
z7p7H9`++{Ti^wj>HDIAHgw*w`o15Dbe|?Hzjhop_vj1+3)@51zmsgj+%gH3$eYA~?
zYd*mVnrL)*4VYsx-S;Im&{yCARH46><LqAf>HGT!!+@|dsjyfz@2^xXEN@$3&&zf8
zNz9-z?)p};FXr7Q$-Q9wXDJT><Mro)7#=3y1%tFseZ`LR0AwW`^0T_v`P*%>-Hv`L
z2Jfh%w#`S@gk3*B<dLw$CADn3B|#yUYW{8`c5Z6LReTgx{|iNG7RNpj!bsAnyBp#%
z`$eLWpENJB?FoN|ka|b%<E?uwx`xcA`k%%SQ^7M+TQv^F><&v+P0GVjk8U@)2hY&Y
zP;EwlggY_9G3r(Oma=dOr4#&z(H|8Bq!NWFN@u8F?;VHZY56lXWI~f&gI0BHLfvg0
zyu<K`n-pk3q}Sm5gz^_>J&j(9MuE2<<hoCC3AjNMY*-%T1!1ld8HE3h=69%5;eoft
zNAd9q*i>{E?_RKcv{kP7#s(?R$QwIrc=}nVrJQ;xp@<C{hC!F`+-UERN#dYu@CBnw
z?1?mvRJ32>I>c_dDf%E{;v9t8sXl8)W`v<<9K}Alo0-_P<3odn#B`NB3UN@Af3%Yc
z2<oK$@$y)$KGbBkca&6J?VHL)$3OD$9l1nZOElOj0dBf+qZUiE$ok@~4Sa?$dV6tA
z(n6o!hh^5sA+$rXrJCEU(yxU}LzD8fY=&CDg+J`IO#&(84Y{))d2&n>KU>BC)y5CC
z$<J_K5;O@M8A>eOVswopQW6~QW&W+(mQQv*HHE?GuVvjj5Hvg~(4E1XWpNQoaB{h~
zVwAIB74_}OVWK0Pe?@^vIyIoexB08Oraz37%g+HO|Cu$<dY~u)fY&^uyR~DJP=K<(
zbh)o>6!0+s600dJo6Q04+#)CY!ORa5@`@A@jPIcvwMa(#4Y;LAl9wPiN*5XWGzD_p
zalA_J-f%eUSE`<Wayo7~bHA`y6#*_<C+)mK9g^VY$(1pWxnkGC1;;X0LNlMKK3q(o
zTz&GD_Z|b584((xwX2!z)XVxwV%|VVM;#MBZ|g8M`&!p>xBWq2nZTqINsHh?%TIVC
z9k1I!3*3P4^eu!sxGh7jw&!?WS@!8TQWq-YA0ST&N@D$|e4(kCTNnmxea6bf@w-FO
z&GwlWsbOOu;COT#fT?4VnVK|?GP|l@s`%uJOz}mWubqa1q4k!#;a9gj=*v>Q;hDwJ
zB&SE%(J`d;?c!IWc5o$XSv-UrkL8}W1I`Bl^1;)PoTQk1F0|!qJM}80(ZtU~NERK^
zVSO4klPdg@G1ikzJWhJ&1s~*%iEisN=T)8d<+L6s61%x>F~=*hJ$KISDxfGSkZUrO
z5_354>Xp+as~5$>xn8=239F6b-)i#_3RGsyTd6z$bU9#w_U1MtOu8nv!S>wkfsls%
zI5m4m$3w0fQDVv2^qW6tC2M2o^Lrnz5DO_Fs+v@W8cRlVZmfS(LZGuGX`=#rov5GF
z%l%uh)bk(c2#S=a8%*_~+^SgFe)ThITYvq-+cF<1!#y<$=bU5f-z^Lf&-NO2WDO9X
zwLp(@s`;WKH2F0M!=W@w-(aS=-YR9__Jt`b-HWio<in7u=9PIIntah$QXo)r2ox3D
zuUuBujn}t6?x;*%)Erim#PEfp-ysAtn3IRk$g+|=tkg-*OZ$DS(5|O%Nz9E&bI*12
z<muy5#^Kw!(91w#J;G&6-1cqv$s|Z&xll_v=ctST2tDS~02d<ehtgsd&Tf!Z<1~3|
zqrsgUZw#;7=ZNt*MJmtfG~5~gRF$IDir=?)m{Xv?v6~lOtr767;ndfmZYkB;R=|Bu
zVv+Ib-ezwj`pLxUlj9r!19uC$BqHv$*z76w!6;ZXi5fbeq6Q7^6`>l;KBC{h>xzB^
zg5x}LRW<w;phONL@50l8;21v#w{dA0PSQLLAl3-ttBHDrEFhCrX6JnMs+gMNZ_X7r
z3;q@7YNRn(I=Mt{BF>sKY{rma?=dBreZku>PWwsZBnIx+t`xe_VmdPNe-oNu%!TZN
zRBthYf%LHQoec$(mP{&#K&}sMopF$n=ja(vjdMBf%3lbv6ns$hB6K1CRc79P;K*@Z
zU<D+-18+X44H+ztpguo$zxLa3-=Ir<!XNg~nuc*d0`dhN(0O8<@U`{ka4~%4T?_Rx
zJZ?FzufGWoBjM`mE718?82C%C(`GX~nzeHE`h)x#*NT-4;R01bz?<sC62Ui)PX)ZG
zYQ1due5LY3D99_1G}K#h+;^~Tc*-8u7p#%?-TPx|HK<opSY3P9NOm8w6`VoH^tOJA
zJw0ra^wPbi0pa;-)(wF70TH?&$eqVeIOTA}+|HU>uH4(0XNFg9K^1CqbV+$(&e2$i
zLTgSUE<*UQP6ds@Azs4=r`gZ{hay(4)qBqG@>hYWSvD6~eov0u>GL-tfc*4zw;-IK
z&hstF=u#}g24EBFdXSJa?$~ueLvgc7IH7@$?50iG4R#-640gQnjN!)UZ5~jN)^P6b
zWiq%;@dP*f!s;Ncv9gNuJJ>HLl{AF07}%^|62wJ2Yj5ld_lS)cHtuRFH1xSLLxqyP
zn40fqBnI@|cX+|eOg1nM-@f&dBg~G0E5YiCRo=IK4(7IaF@giZw4aTW^wtD<Cz;C0
zy~Y*uIlnF5k*x*8rhFZ{;a)PXv9Xv8LKre?C|ljaOqvv67Vt`i^9k3xa)>@dtaV<7
z;LuK_9Kw42yvd5WjYMz6{pnjK|M`QO*mQAxMvEuqr(`^B;**bL;BIb`3Pvh6327iJ
zCtNY+`~iD3TBPt=0?`B$V@^+$C815kO0_ARx@<KGJ(`&J`|wrRr277)`7QQx>xdFb
z5i5hJ#cG@h(gio<q5XI@f4dP&dtz~+&|SA$R1r9bf(um@(CLj)kaa3f^bCEcaeeKw
zp@lKnD}4y@{VgL?(7_60Pzp%|-bWcY%P5t309Az;!<?=*G3$5T#9_lHk#XO98tq%X
zllg5k)n-a`#aZQ)g?)cp{|_xRdY;@n!#rF(KO0n>(Bx1eSHv$ND`90A3%J(x9de2X
zcaa5>{*XJEINJ#+9(M{NdU{U|2(51ss*#k9%9uRHUR5h=pEdb}uEW*F5M4sgb$fix
zkJx|Jrr;}=OfVMpBZg1*&wR3CttA@gnwv+ZtrKGmRgJlAF>kQ)<VWV92OJ%9nvGUG
zihW&kN^Ws)vVC~gBT>_rMNN>-#yZE&wj`#D6?^@*KUH1_2)EE1zU$ZAXjv9IyjO5A
z^6T{dFkb)ha=tlm9@C@z(eEZfQS%f{HSG)8^xqSdArx_rrl3O8uvQ!V7Q4INclx*I
zEltb|xm^$LzsNIR%ePr`-08Vbg3D?oYm+x$a@c~ff*3}UPOKa(?8JRI98P`wbCwx}
zGIo%0HHvXZ%<anm;!_f=t#};&rKlyGB?+p};&1xqXq{EbKH}74ph))v;MLaFua*m+
zTE#qg!aDwO?!Cd-NfR?~RbB*IH%zdQ|FadbT&Su7>tn&oTQYD4a#9F!FjQE`vpP}+
zk;6~q414RSD{u#9YGJ(=Zn_z~3#Vo&3Kxlts-F*#cb9XW8ngN}a=bB!3xZZb9k>r3
zo}E$ouqi@vK_R+|;;7TUaE-ZVyM%{MZl73^N(+QmT8sh9Y_-$>On^wVec=-4k=<9$
z&wju51<BW!NoU?#)LPo*KQ*3s-OY>XP3h(|1`aYRiYJ=*G4J%nJB`fn(lc(2no$KB
zcZgXm|EpYRKeHxjAjjsti_0frcDFFhU7j*%#-t04T7ui#CTMV+NqSsgz2{56n9q9o
zQ3<%I40|D=?DDyy&$FOeCGqkV+E!70Efcg^{&i)I*Ka6A%tDA6<;#0-KHXa)fAW2E
zD-GHjlY+Dr&YzyrRVDNz48Dtc@lE5y0Ex48jtct9LFaM?T|reGY%_~XlL1;WA(%pQ
zPJ)BU#r<XfCTpz!x0~<CHW@h5@FoMqFbh({8Qm|g`4ExK{Fw_-HY#y{vxPRO&ji!|
zvEZqh&qZJmK8%>kIsGSO=QEGF*Q^z&x7v`rOXYXhv5{FKkKm(Mow!J_g!kiL&-xZM
z7IVKoxSwa^E;cPJr=@p(Bo6Wl%(+Xeg+6$=sJ}0hQ|1v;8#|4#myK$j-1$YIP4>#L
zo0C^ZlMw??V2EYvm*cH+7%C+r<vb0m1CA6yLUpov{QGg+ZQKvsk=6XmWsSQUhr?x7
zY<cg80dcq|M)fqBNA=d1{xw{uo#!7K${+bcuA6lb-6Y%34)S)N$wAJ86tK?|uVXq$
zJh>uCJwWje*B29ggU7BTJSlZyvWO`fjf0jl7nY*YDQE}K``_`6ds;z-q#tg#4~UJ~
zK&81wZ1rQI_`_~+18Yj#XwodFM2Sh40x1u4`7u#e@8^YbSiD<UD712DU}C>cGO4<b
zS%&oc-Xg9X0usinw~kgfQ>`Mu+z_FVdV&#R%qp{C>J``yPYG)Lm2sKBdEAS?=RWQB
z+ImkZPS8F6F}Tif@3Ak1F=x{{DXLYsrCDVXRHr?98*H<m+wOU9g=Pdlee!ZK8#<~k
z_8GV%f5C|8H?jzRWEO^*Vr=6=(Wc49S4;4Jmgqi;4A_W&J8^z@tb<X{V%_6Jh)y`u
zRedRB@eLK3Ra46Wu|IcWQYT!02-ND>i()iR!XWJzPQHz!`hqGr0ZlJ*J92IX+zpD=
zV$7mbkrG>Aw*P9OUBtLv?&Po)RA^?pR#!MeuXiN-sO9Elh>L7hjN7XH4~+$MMD(i3
z3)!dsviIZcj6+tojoBtjT)dyn)IIyLJiqN>xEUo}MApuED;0NS8tpnv5Q%QPWn53`
zcjG<JR>kPAAYGNf6HnH&H}F)p0M=ATx~1PY{8D2L5aDyXRiWWNVYvwAzA6b;B7co%
zX1m2tJ{5PVjSm3n5L?EX4~6N;4t#Cs^zjgL3fKwr)dFP@K(CDA@q@`k(xfsuoT2WC
zbEvE59uwb_KBGYeUU@S!M)}S#|HHZuWQ(juWMotWuA`Y}3sv7_S~k^#W%yTE^cJo5
zZ^p`OPUBPWuUVMG9zFcDRkeP88<6y63$sma>S;rYo7=<8Y^UBFR}`O;q}5_JUN@rl
zKG1$LO`{$obLV}@qfm$jZQ79p(oV;5*X>u+Q*_tx0E=^1xsIfn*KIBOCnIP#>(COV
zs)i)T;O66%ntj`^z(BulA{S?g*nzcEm73y;;GbeUNW)&VBZ@h4$$J@<&b+2%2i)Q{
zT~<Z!NJ>`*oR$UFhS6Z`w@B}|ciw>Dp!73xoNm!6o0%c=I$2DG*F>({Z=UtZvx;wR
z8q=C=y@MK|=3A#2a5Fni+U;^)uH90fnen2%TjGRcx*4)l>6QOLqa?Q?_TYbn)YhR?
zt##lr&j+v4@P`Gduuaiv`Nyp};a<h|m@OW1-8yqO4`QO*i1^z%T&iUdi4on)oFK{7
zGz}}77%%Q&!IClgiWC{?E{f=4q}jx?qWEw&>|^o2s6Nedm<Gs08_EvGi51!)Hob}6
ztoi0|L^#iSn%yn@UKQUYZ<PM#<}JU)_a+F7;Cmgpj=AvC$<wQZnJq4^DSN#}(M?<4
zoH4g&Ra5oXRzX#UP^)<V)wQL2&XtkoX_-*7>EeNk%9#p2!Kfd!8TLdjJHWSuDAO?)
zH{brki?qKLr})w2U_`^!Z1*1D*xQGKTQ!4S;=YHs(W$rzsv##hjJV_i&gD{X4NLpz
zdPdaA>ep#E4uyLqB}#IJJ?C)c?#9mEkQ7tdvLTAnnRsnS?BZW@gW*%>^deL-?~9q5
zlpp54A5+WFY$FZB#*W>VfMBRaOPA6s5!=tjdQgU_J48P$-)a%*sZ-AggoY8IH}zvz
zIesrRy>c8eh4^Bp`Q9XsMJ=4;JJWvHOJ)G1rYE0ywuN%|4*&tam&{pM{?Y`0^|f7_
z)7|j`l%;%1+9QXWFID$m7$iS<!cI_7Bh@}(!^Geg*~{*DEXH6Nz7ol1>Vo$vVDP~}
zi+*a$Bn2ks{v&I~fTagDqR83xAaaL{R;!ys!^0G9jnkd-TcEUg?c4jqH>1%w*MQ^T
zcG=`1!}G3tR(#U}l#L}aGwr3@qQ&0+ush<HVb>aP+t(fkjb>WA(50t~T=?np&xZ4t
zhr6#;`pAuY6egfHz)`QMKLD&~QfK@@-c|ZhJL8>LOZS8Osz3Rgj~d*v`fYy}NT!23
zMJwI%Cs{I|f6QWZw0*bGl*3`YjXIA$)U`Gu?6PvNyLd>wQ~tT+`w8V2*W`gSSAZt^
z8nEUMPJB06b{%&;-Zkq>zvnWaaju1!Uj;dp1)hV!l_ob2t*IF?oLN$L`R*U>;WF?t
zH_;gDe>FeCU$mxpo5kQBLF(#{&!+(6$+;Yu7kTT*GQ{0>zJ42?P|@5n8SL_=z$XXJ
zS9OD5vLJirqQT`^+S09u(Y}k)nlsjfeHcA)9q|dr@8(4lk_Pq~OF$0XAr%Dln>qBH
z$M&lv*~qe0a;Ghs=&7?gZRR9*_ZiF8ilE&3LgMB{8)oMt$h;14NA%rj;{9>ZHHm9X
zSj%k959``{v`6*|Z%o1u?{#Q91E1S@S>bJWVy!1lW>4eweWU4fo4s6o#%2-LIY8L|
zsG2>$gqM~P+2)vNlvhZ!q?2x5$nba8<Av1n_R_xaL`N3|j%6uv1JP^F<}YEL-@mUl
z@j8cz=DUIuX{cj*o<;C!Ll9&`HSYHMby*h{st*1VSd7LZM)oqXrTIhZcMCw}^SXes
z({<sz>Any#_Ndh`5}Dvzyc#%*>6=@kM5{ezU+2g7>lv|vo{)qH2&&WZ%&J;X#lZI4
zQ~-yEPp0>CgJ7Jz@=ua8&mJd7=Wfh3zIq?P^J^V8`;?*AuoAu_Dw9S(nyh)#U3}GY
zJF+n3#+b~YQp+j}y<WGPL?q4lZ)>qdNw=|Sg7I=ULPv<Yv-pr%(<H;+ezQ0bgC$p9
zlvU&^0;T+vRPfy>@hww2Jgo783STd^Z&%k({lNJ%iS~y`dzAv(luRnIWO=2?V&@Bs
zjBe)fpTj7{&|mbL>I`qOhK5|+Qp+$xIoSA3gRl8~G?-88?__+>HH5KtDDCUGwtant
z@6a03r>Fs{HNwqfN^0`!c>{6ko!@UFzoh0^<?PJeJN`r;rZyG2B`*!<oc!s?VkPJX
zOBMjdYI=2&n|*R|;ui<KHj^3XA|dv0^;JEo=`c|PKn7O&M9;`#jVA|O&&l|PUqlD+
zJ`h{G5UvH8p<DR~$5e4~V+f9ak&GTE3zlRU-}k682Gpai-o{NS>@Xs8;|95&<BZL#
zdj+h<P>`}0bhJxMM4@XLq|a7Ys+rlo(1YzlgY=|?-)lhB7c|P*2?k))>KbK6OHaO1
zai<l0sXHLhmN}+p&nR1laaaN}dc?g|`dBr_mUsYv#n(k%M99q(w<=Oc$PP+Hhy0W(
zDYzol6uh6Z>DlrkcAwc6BuJD5lirpYgXPub+)6bm>F(Q9Ks>u=ofA-)(Gnz~dUlcO
z)a@|`I?nK~8C@kyXmWiMEy@1IQ&A~lsnV&0L-4NPn+(<s&kR>WMhsMVBk3ezp53*+
zu_;9aTUGCK>KZ_Xf<e^Fu2~h#0jZRnENb>iU}n)&XO(dod`EePdPHX9!sSRLNl6P)
zuc~E<t6^=LJs_E?LuoqHV8quyL`xMcvzs<}<P@!rGjPRB`tt3YO6xsPRqFMJ2>f+n
zW!wj@54m<+WrA<)VUADS(wQ%OK$$Av+7S2+>xeuQ_sJKo0!t#nGi#pCrpF-DpcS2m
z)+?j7>f|P-&{T+5b94J1AIwp;kr=b=i@LM=_`}&M75Z%24?@Yp!)jAcwT$v=-}X~`
zcWTvknvUqZ2&LLq2ttUcoR?7kn?c1iaX4EF%ZC-{gsg_of1)54*lucobtem7QL&2S
zeTNnLf(5j&>UjLRhPbC6KqJ0ZLs?(7)J<MggfpSBsrp?LkD3JLr?Rnm^K>B)X{c$>
zn?lT?XRbEFI#~rHa5E(5nxebFz$~&&`+yAgi!9o=#DsA6Jg=;7lAXdpJ#ZXl_)XEs
z978#IijQN2T<WAQHcOpNQz5+nf@I&FhBj0E;jvrYrw3k-L6H)WGzWu#g2(HjpmY(V
zrt5iE*DV=YMkqteLsDw>i|MM~)#A_Rc4{_kTaO5!*wU%Y(H{>!zVd7qv7B0i2>Ts;
z?srRHSjV5`=earw1#Lgg%;ULN?RoDIoDO(-OgUp(rBM2L;q2Z&Xkm8AVwSzv4k9_#
zUOfQW#&o+;JTQg%m$k!<Lae{Uh5JQ{MqY2Q@vXbgk0wu_7M9dqF8t`@M!33UrIyC@
zn(o&JM;jj+CvG2d=TU)HTeY=2+Bqw79s9@p;w1)8T8*KfH3h>&gr=(RFgRu|6ir37
zc3M%hTz6)#E(e4xJ7ou2g2sp4n|fi=IdxL}34YQcve(dWu^)-srZv6Hym<5r)LFD}
zx7|4<qp+o?eqEzZ%l|Q0z|v+Qqrt&EvR-7)=B+ykyVB_(u(2Ti!P<s6K4$E3`A${A
zF}H31mNG*N@-SvH)mw;hKI;acK@DFeyPEyk80D{w1$urojzG!obj-`Mm5dpsGn~&?
zYzc{LXjn>UjUV{rS1^yB9QzYSNcH*m_M=56L0%ru@g==GP!<Igj;Zsjz6Iy}rTYrt
zRxO$o1Qq|YA5Rp852!rC5pT7&?drbgGIN4J1GJ2zM65Nh6LNUo>HN}ZB9t4tf9vW=
zOb1^)Vgd2O6(kOW7A3ggp?_mn{YhMrVnf4vxfD-9M%Yl{VZMk-AR73rqK)A$<L_2L
zR6fyBD%Z<#FhrzH$gX%BWbR78*uXytf*+c1hQ@xL1#mf7n>;2;6uSvz14^r<-u_#&
zp59w}W8<?&3vI|j8%c+8XCl_%9z&>pHfVo9!VPYwYvEob%j{Rg6A5&Oa_ZKkAo@9x
znKC9M#!%?uBEw>_l$gscX%ihVCkq|8>aQzSmj?KOg~P00>jDhUZq6a6+;L51|IZm+
zF={+yvHV$X)8XB8vOfush5tcadc)jsG|M8CzzQ-lZ$$ylhXP=fyx)*`S;dN0a1arj
zA-3VzhY#p(pch1+*6cdiga7E#^RG;4e@4*x<p1I+Q3nn&VO(wMm{i?k#FQ7#$Nh_q
z7?58JyMcb(K^P83psRq-&w^i(`h(U&&>j*OC;RF#?sve@Qc;6kiNtTN&HwsWqBtPs
zy|0$g;#O4!J%vgYUx6ya(Ni%Ee_e<Fg#twRyA6@uf43p}Uw;4hBj5g(Bm$iOO_KhB
z$po4yftkL*18E8)AXoHX?YI8#c@cqW9ZQ)@0isLyKb?X<XoP>G6oHZDIxs13T%`?t
zx*`?*PbaMZV=M+uiC&7UCAbHS-d~wu|HlIUzolkX-b)K%yEoZ*Mb*-vo7GaDryL|p
zcu}K{uogVjZ`_~9F5ZBd4X{NfGX9HWWVyW?%RBH=GEs+R{+}ht=gF-%%Gh66Zv7Uo
zi`r_@pr6bychP7E4?R&6j-Emk5DI#!)oXnXW+o<#$%8mw1FjQ8Uf6sm#^Ni`c@f~p
z$Iqvyhi#|xrcSECwDvsWJu(}9vax)ccSF8gPR+^b+4Cw`wRXJrzRG89ZJ7LXg$qVY
zmrs`{61B%uX_p(D{+_1?0oCxs*U<jM<;A1ovm(o$RHDnKp8mt_)xN&I%Cy<T`83zd
zy{9J<UF=@Qr;}gyaj`ThQ3TS;OarBj5or&pRMW~SjH^9vZA_lbOtcS69BYgexj@U{
ztM0D*JEPX*6CZ!Js-|LFb{l`0G>ZDR9mWV5bv1VZl|8OqM?=RV`uB$s*Xz<9EmW`p
z8g%yXZ@zH~Ik2FV$|DV7hdo!<v;FD~HL-(DNg^KaJ->nyvW9TzxxJo&QOf#91HOl!
zigYE9?GrX?Ja-?8xx2Dfd7ds9{p{Q6*eP+1;7&K#;;vO`C!~=e<}tiLRMv#XVce^v
z{`>IYJ?a9nu`~tWzHa2@g1lrVL)@3~6kXHDqdVB}+~lo@q?NqgomxvgS><)x7Iijn
zO+NndXAl7mHhI3suZi=E6yC<LMYYlwcWsl<31}rKd2lqVi2gdkKV~}iNGj@E7a3Pl
zK7E$`I+&_@;m%xLz1{Jz!%8FR+OwJLO%B0s!!qIs8qwRS5^fYVHQtryz6QcOONFu1
zYXP|vc(xNq^1AUeBl=*$hE6I)|9c`HXufgB<Dcx_fiZ4hjltk=-l`UKZ;>}h2nh`w
z6O#0HnTe0|B05-?EcnVHYPLmkBb^6wyt6n_VjS`HHW9-5Yi%XTzxU?P4?kGxA@6%^
za4fMkDRrfz`mk=x2*an9h27VlfHoA;u6t9eE@swzbW~lfvkS-z=Da+vQm9=-bPypO
zuA*8p4kjV(atVcjfdM`YRfWx)Dv8~bDu#FLzK|ekzhWrtydbknBlKnTtdJzT@x5b0
zW5<ee*>SU4<He1jhW7r&hI(1W##C_?)8qBAt(oTjt-aM#Q)ip1;UCjXtPO>;tc^Z?
zLAATC=MSq~FNDS5C(gsA&zFMFZ6eadsLX;3VW=A0vyPpyt(|%Be}6XC?^m6)M>b4W
zpH^@n^unFa=HByIkCl3R1HoEFD`kz`4ASL=-Y-WAK|(fXLSii&rkPqBYCJJz^z?B!
zIbn^f3A`09I)fVh0u*xpDbMtLnQ!PV6bi8M$r1N?0_n5pIL_tWnC9V)8xzL@Rfcw`
zXbPz&`j4Bz+AJ=$Aki#gah|`Jo~3(pV|{h{tl7%+<;xz-T_a;Js*UlpDxtPuhQ4Fm
z;fiaB&0uk>jft8`cz9;3THw9KeE+PhP0E_Z;ZW)B^RZ#LU(}wLnA;wGlhnf2&)LDx
z;6w<CxjA|7J<wN+A})O;>R5C6bf$}LvUVcChTw09@Q;lSBPBv`X*xP0Ss*X>$Pjz7
zkJ%yZeo^zKu!dR1!%7@K3N~s{FAi!E_mA+XQkZdofzV8ed1DrZL2<F}7f6Vsk%1Au
zl?sE{@rjq~`HI^}(RU$gk=FuEZ<uJ2&}uxFniE^x;_sULhh?tw@aejG*$Kr$O4dM~
z&Q4gvB<PJZ-Pl@WRfXjO^W2uU=%|+sY7FR1YyKE+OZ)R)Hv`xDGRMP+G~s>>uH4*Q
zBf|H3S(>!V=fR7->!Tz<7foW^?(FQGmdg7s;>fc$QCTa>ee)qNRE+8@+hLNPCI7rU
zY`DS-$;BI#9ZBae1cN&FcuavKcs5s_|LZQT3#{4Xg)43NWp*RqKP*lP*wxb-EZ?}6
zViK@4%^ZG2HJ^2aJ@)i~{F((bv{a8lYebjFvmz^A`707``X_+SxwtFy@uiy|ehTr(
zZx&FzGdw>EGMZP&8z}SKO)zz~n6KW>?lD@q*pZ~=9&>oi2zeUjU3lq6o;cUKT$E5@
z^KC{WcTL3p#6#Hm#N+Vv)Wz0GTG?%Gx}kbZsG*a5GqT)izQ}H>ciNWoyQokt7HqOg
zGKfhrGRTo!a)sygY;<;OYNLyABO23&KRMV`sU}_zkC(^j)N7>35=QRcEcJXTkrfhh
zy?ZdyYXsvJ5?YJr0y|&vJLI`3Lq5ApD@F>m9*r%EI<?gurMvhS#n5|~=+esZ8Q6c^
zJMzDeK#3kuaLu>{5o5TTl-ua+HtJw%1g&bx8z;76RW7-rFMCqOr+;NjF4tVHob*(N
z28Rppa719b)Z+A{iJwdfZu1TxJ<l@p1!neh$)~$@p?InmIMT{_j3LrWZcaTr?U5Bz
zOU=94(_Vsib)vvhwi?-O&5X|4)e_qwR*fnSs`CYyZ*ue7DYaglV>Zr5)}YRh&XzN`
zrY3c@bHfrO@i7VMmK)wB)t<c6O^Km%i*i_P;jMg?z9a0mAS3LyVtBZ{PW5{cH;fY8
zeB&sDo2S@mOKZ;Dg2yTcBg7~w_UCyU#!An3M}l2O9$^2E4f~g#e4T$KBl%zd(C{|u
z2}n<*)!U1y$*!%l<=G8{tYSnUZi2+JyIqsQE|qg~rCssZ4hJpNYrC~No@bHQlazo?
zWb(Y(n_6h#r6-PbM`I;4z%pK)O=9u5lP;(+C^8^EKAxuu(h3w}Lo;XEYsr>3EH}A5
z71!<%mx_mJ%RxPlcU>AI{xDg6=VIUIb!au4OgL^nIga*?D_bA#T9wRIWwK=;%axwl
z-aj{(UXDe=7e&>g9JY5@jnY8T4C8<?A+J&R`be-upjOJ*M3NEK{GWE<FEcK{0#=BT
zv~+<Im~0p64&ecz+r*N~Coesm&n+ho^tXe@RGMu*b%X|Ob9M{P^bNvJe40*MP7i79
zSAm-EeqBqO8}(iW=q0u={Anx2n(E<jUb6>nLI&@f*E?qfpBGUB5&=1vC%kHNZ^HfE
zu$a@4UYXtT?8dBK0BLUY4ig4@;h_@Vvsa47#d~}0ja&QYW7S$|tdR19<;cpfjn74B
z#mh)i#k}}B*;|PJdj?Ba)(({*8`W3#+BeQVXT;;|EGx2crNf~tT00f->J5@^oIs6E
zlDf7xnJM1`pbyxk&(8}oez_0^X)#;W(V1Tj-SK)h5dtqjoE;JB*xi^Lo81$1Wce;L
zxA*w$vRafy`%4p{Z4-B8sMo!hCU{7*;M<Y<M^V0Ui#@6SK=8*Na-|vIp(IY$)LiKB
zZJ5ZsCDo#KwhwE{54c)0gIpS86%Q>%n)Uuc8&v~KzkccfK-sWr)DFWyh}L4m=X>Oo
z7$~}fmHFNL=|jfIa^E<gSCPV{DMX0zstZ<A3pcmtD?qN1bd%U>bk)%2!hPuu&7&i@
z_ZRp@l}pbLdR2P9$T!_NyE!63e@AI+5U|?^Mmo~Uv%j!4KhYt66crDv&u)K4XYVqA
zTWB-2s-@y*_jnmN-y<g5KWE|_9x^_J1z5^^$G$H;a6jW~TFI(0PYsQ;O1ofgT5mM2
zai!dtxcIi6=HB=hEcoXxaj61bBHY<k6GNY+VF>2;WKE>dWmCwL_1(L74b8cz&6CU~
z<=7FuDqKd@WbDE_$GPM^lc$~AgT;g8Uzh{mtpdO_{X(VaIE*f!QPD-G=(ucMXWWaE
z#%*M4tA2gUI)ZyJB1M>{JGpY)$f02L>~qodUa3KrjdIcPA4e;qo6}pD$ME<*^+P9s
zXY%}$-fPWDUIgCy^BsS%q~CYs5oG<>7os}gl|2)5{tzeAoU6mSn;>~H#ruWV@PwKA
zp~L-w5|391813t`re(X6vDzbVyUpjW7kk^7i#))9<pbSER9oY<FED<*vRl8ZW@?2@
z=t&ENyGZ1HTuPmfr3=W)(tEk>#dJtvZFGL5XxcMbJ-t3AG(8j!^)DzHvc7}4fdYvU
z7fYBEB0Z@RIE0&?+TxOdl$eAx*xPB*6q<SY`87g3+K47X*u;^A>*>NE=CH`0y6+h1
zK0FwQd7skhAst?V<cSx^IfO%<#kE@^s(4k``iW1A^A3YErbA*E8vrn?P)=Cvk?=qb
z!3qCj7)JQf9Exd*;%NP^1ZaVr2=N0}I%tUlaHB~p20QLW!pOk*S@SOI^hL^mx(B=A
z>0ax^JLOWC7cDR~qdDTJf@=6dAg5750qMI0l+z&>6{0nr<tERztKMkfm#SS--?&8Y
z@#=GV*Yllbcz>8SeXCl^x<sY-V&~S(asQ#0_s(8`-OirSwA>p04$yFeTrBQ1;$9E)
zxN4Wt-k+7Z?ibl@O-=Vwd`SK?7l4O3Wbp1hC9TM7Mxc|wiq1C|wN775B++ph(;}KW
z5OA0LTA@hbG`V!GRQ4`*t-@r|>do0b_D0nzoFj%S;io4{xFkr=`AoXkn=@D7l{{^}
zYjwr#l#m)vF)|G&l6L$~mF*fK=_z`adCn_KC*PZYQKkQKiPo+hp#vkct24G<MxV7^
zZV;69H5T{*qGFyq1)}S9^^MbggFO{E0aVa4`_-(W2Ru3tnpOf`xoSO0XeBF#&*#rl
z0j6!)-sZiSy_L&oJ8|amUD=#ix?|;)o&nBud*kW!V6LU<%brPF+#=u^IQR7FKJB+D
zeSL)@rZQkH)n23fydkvPBLEDSP7Fqyv&xQtTA!XkHWXQI>bg*ic`zGRoplN7Tjl1a
zEan0jOf<U{Jv&iivweEhqqrtibhiG^_L%V#JxeNG?hE4@53*DVFA5#qca1&(%ZK!n
z0~mc;p{3s#e2+ZeY4uLz6%?bG%2G}oj2|rcw$nA_aBM%f6k@kABb1?*4D(QfR(p`?
zB<MEuO;)G$M>n?0dMuPZm;iQ+%7dI&eoQt974CMjB(3ZAPgY8u<VwjO%KULEy>gDo
zmkZy!oPB^z#Fc2d@6WiE!B2I=`&OyQN)1A_Sz9B#wVX9Y&FUNuhsaGq9{JvLEiBEM
z1MNaxr9JLQ3d8a8V>i90pCwiDnuJ|0^h>LMjcDY)3R!HbTljH2!a3h~+KOtuHd~)U
zWhA93su{I&Hkw)B(o<u%)K<7s22f!jYve}c<D%1Z;C#sBy$c!Sn&0-)9yqNHH|==8
zt6R|#moeM5Iy`;QHQkX|M4lgziMr{yt<Qa1^)aC((SLh<5dgAVqbzPC?&qgmxf<Hs
z;E?-MK|O}Iew(u^$7L2+lG&D8dYECvPtza*I$WTvf4gKH!e$Se#^#474s?5Hy=jJq
zCO2n)y416yvqOz8pX%~E%<LWK`!ee|H(g`7c~h;nhE9hnl_kG1+B_S$WI+`~jDd{y
zV?}!JNjM+P6wkbV;i)*jv2ql0Z@PCn(-elsJ2cd=)8XK{fA-RKXHHu04p!DA2D=(v
z0Op}>07XG_=m*bJLMKP>(-jNgMtTWFK8-k{X6-S?DwG6R@l>YceV}ig>$zSEX{k%t
zbYIOmrtCN|PoQ)s3>64~N+3{={t3Kkkvg2sM2kS<Z@MD@(sUdWyk8`iU-%C`xZ#zK
zCRhJM8Wh+X$r$a{nwIl%SZMhnpKab3M^V`{$WIb#uHdf<nNr$H|CetkWuk(C&Ti0A
zo1X+w?(vn29mp*Qo=T)Z&hTHM4SK#i;7C7vXSNUm40Zeequ*{JdMATrg<J8i^S^Zr
zkzNphs5Gty9$x@Rz5f5>js9)_{`T-X%-?uK^A9}Ihh;(Who}~qqP^xC=hG*@`<KP3
z3k1O=!DQD)TYAV#1Q1=BI|~23(pTTmVPR>OY^pFoUK)ym&{Lf0fA8tlH}sHE>CHD_
zLi`w48}iSu2<DAheFL~}&7}Xwh`w<&djwz><XRxg@#ieSyZg{E7m&*6jY;>+YWy3p
zl&DHY#(gb<Zl+igYR-7eH(h=hu?V8KNZ>AOW%<j5|Ba>U(17&dHjU4zbkQQLchIvP
zD%o$cUv9a034vp)8qg;8?`Qw*Lpoyth|q2IT(nCPoF1&>!VfprYd&rE_iS4ov!;^&
zXXH*e;3{(WnDZ)f*YHQ=E&?DmPEJms(?pt(mwV6)<1)`h<5?1le^_}Fa_o^zM{@3i
zJ-YqA=G|$!^K+CJ`Nz0vCE71@rlIr}lvbQzM}n?i{X+80^eFT4qLwnP?w`D7ustI<
zKG0JYZ;7+7wa2TzabN48b>d*3g0=88y-tdF8#d1`YHd8w{u2lKbdhgKhM+tgU6?_L
zZV%n-wDdytXq!;-HsTTN<%j3Yke6mSrlHcx<1?$!vs@Iad<=f<($!Mh|F`$BzH$9B
zycD0To%G6IGhsrXDKWmB+B_eEpKDmJ7vJima(OJRoVl(5+=-_iXC&~{CdutniYNrq
zO)YMKiska82x>@pd4f8tY{}+(`|qv%+Xq3=z42lJMId>0(ONC7Jo!fAKrTcw`{u4i
z?D4=o`J&@4!+Iic*d~+iAQ0*-7<btpea2wlE|NS+yJS6U1^Ypn*7Ud6=8}<et7jxN
z^7~_)kx?}7S>)y5mX-H(&+BOjZ6YExuBzSRqMF`9e=`7xzPR8+NIR#-b9_y+#2rb{
ze+%DxEA7&1w*Q}Y`08v?v2Rg%uOt0yj}|X}R1#SX(X(iOLU6>L<%Y(E8<rCyxj_&p
z)gdZ|Lnj|qMshJlcsbU+{=TnF=bsOlNiMpXS{&+Vzm{5}=Jj?p2dQ+2&eyh+O&}}x
zlVzd=zBlVms?0BZJ)O3!J!15@Xx_Iwob<~_h5l=fe*)`26$=?J5vM#)vLuzn>E7&C
z<EA%dzOe-P&hs7F>m9vA#vGbtwK}`T8c}_0rNQ(p#!P{f%T}*=(kQ#L@9ugUz~jRg
zt;Dr<_?I6Z4-bq;Ox(MqT#Ua!xP4kwO;gN2D{4wOokMu034NpbeFO}63dm%F;Zvf2
zJ9uyEoG@l|bc>lT#XQrzxoaacAZJ|XpKWsEDeGF1#gfHSh_w*Sh<3jOr*i>_Qd+i6
z^$zkl`81XHc;L~e%5VLN@4?>C2r@$0Wy1x)AJIpr8C%SkMl0npGI37GvqjR62oqL9
zwZ6mXY+gBb*_e!8x%9rHkN>C<rHZ7WZ-{x-FL(m*k2gDdcIa6G*xV#zs<1uo<RGS)
z2$l3@;sxY@Xe}4ruieR|&pM8OZ<S7;6>XF08z=VYvvw514w0wu*=9FBvJ0tw^FF(j
z(rtc=tVhQ01)hXx<~~xYS7FNiXDzFjBgHMn-WaQ&sHyWYE}qUUF`7feEIk_56R(@o
z)MFqMuOAfym8?b5TVx=WcbkE@KQXw2m>XoV&{lq$A+R}Q?3So$yBXgQ{XrU<7^A2C
z@qUfTKgtqcY2C<>@qCFf2tP_9ZC15)GJ&{%6Uf3-@9XiDR=zk^r-(%%^P&oihsTkf
zKUjf#(O$v<VL#29t5<>sgNd}49(0_3Y+CKnjP9H1g}nN0j~>ZL5ymNT`6>M=V!!vg
zP6p$|H`XcEK?UgmxC1+xc{uEx3MU*I9iM<4dG7BU+(K~e6(^pWfk;Qi(6x*bZXNC%
zw#K*mgs~Ym)vpZT;o6yVUF)3+hsB|Bdrz6a*lfp3^k7>qC-p&K|1yOCVGV9#oRsW#
z_{Hv=ialb_)#$LP)_ef5fg^_kcaO<TJi8W7-3#U(ZyJ2)N|4rs`W<J`-ZL)WL<egM
zYkZEu?v}2p^DeGl89IBcpDv&it!X8k79`{LLcj_}ItG7G^X#9OwWw~S#JKnxi`Tm#
zeDJDbhA+QD?%=5Gq_P*+upO7rlYtq&hlT~D0=qmn-bSB=Cj!>M9DmU&e90<&wX@~G
z&6U`DST~+HWM<RKRQckm61&t(k8(Vw|Crv@x|&R&ar-Dg{e0_2DCw8I2T48fScDYI
z<uCncP1{}V1C4fMtnQ~d24$-vx7+NiUMZ86z=HpDd`HiMLFgv0h{Msb^>KVAm+Z~n
zzS8p8B&TQUhk=Ca$`zJ1)w&W$(ujc)c1}MM1Gfh?QvW#Hf|qEr2A=rZ9}%ed@QF@i
z7K?P^T&0dr+r<jIo~9k@k#Pb*VOn%>x9^y+VaOd1XKV`D*q2O)gGCM#REyHQ$>BDr
zNSi)auYmJd6RS@*rB63YVCY57TDWF6J5rhbM*P`g)6k-(xpqe1(YJp*uyr|gdft&4
zHT1=NECG1namSvMoLR-wXL_)>DuJvarYv3vc#g2QbjOm_z2R%p9YkaL<7}Hr*QstJ
znpQvAF<t8CK4OP|+_PEM#?{iN>`KtS=f)eS4~?t3?#Zq9el^^Q_`jTjE0ZzAuAU3|
zjKS&Fcw7|%KdB)@ynT>NhpYN3e4!H@B}HY4GzU0xPgJ_&A@qY!Kd{FoRJRa_l}NBC
zrE^~b7`Kb`nNDi1WswcU-R3X8U~_}U)c}D|f0A;)=AR)Jz<L&uQLdSV^jRg3b?%r9
ze5gCL>C_fXP#bffmsWWi9#evqJCRMcfuLv8{+I(9P1IG6>RWmU14~URI|iR0bH~?4
z%!zBAlpJ+XjU>eL6x%dwIx88+$1Ko>==;URXchDw`Tp}lkE0dc|Ii~FU&X5Yb<~3!
z!cJA6@OZ}RHmhfbGPnPoSNBo6;W3hm>&Jm5B1u%9<SA?b!K@Vm1&TlaTF;{K*zE)n
zcdq$rkZoQN3RKMK3kjG2iu}_+DtF;aLDUi2KMiYwiPfZvRmjvZ5c%{&SN0SGa-*%b
zr}6)5?%D&HUgLg-4#r7HoJx|FxkO7!Q6yp%#@I%zG-^Vn+j7ZNx=3!nxmBb#w_z@A
zPSKKz%=L(oIZ1~}l3Y7Ty^qf6ZBFa``}G&|?0LSo=llJ9FWWErm<NmCm<Zgxo)fR&
zAtrGeKXV6KV%%p!GU%zrWu)e^#(WzbLEw7uMiVg{gL2G(%r+w*8Zz;ioP@ItObmG+
zT4MpNO{d_2v+EsWuIE1nP94&Xn@AKC#u8tO;NW-5sRSLPRy_v)2sa<CPB!rle28VP
z%bl>d^9Y>?kma|pL|kJ@n@!!Q^FuQQN_p>YY}z;Cs0E_(5FGP*AZ+)GIwLKXP3_}o
zLUa7ZRU~UPm+ihSFsC*YNK)>|C>k=Z8TVvwLr!f-Zrc6?i8~bf;$&^n(C9<24n$qg
z`hV)+)PPm5x^HowqEfa0h5m<NK5Z%nVHjS)gV()vUfLu5D*Jm^)!^dhLd~QNvg{Xo
znjuBn;a-8ezct$aL}Qjgr^$nO5HEY8I7M%n9dF<mo+6_wklB^QWO$?TG!%#J4rqGO
zgkYJ~O44?k405~xpigd?2c5n-cN=M7GsiOtSeutO!jd$QVXM$wb9Zl-DJ~isr&QRe
zchH+gKhWoHAe~EwRDO%uu?&E;4HZ&8f)E-NU2rr6KO0}&tI#t>WT@VzH`7@dM!s+E
z-jJk%2U^Y=c~H}frTM*gcR=c1##TRe17s8rDyhl2OzW;E3%qFp9~u@(iF01Kf-+(2
zZL4HOYlV|^Y;*T+&8TyVO90;6cE1NJ%l*)iIM6+OOt=K@@8x8Jb5P0axqwvmH%~GS
z<#btadO{j$)uuDDg?%LjpXW@_btY8T+XihG>jhpat95d86_&$YtMh;?5djlz7#b$9
z2g}4D&!te9&c3pk>O4w??=Z;7d_z}*KM5gpy5gPu<?3>nX<aNiBbmC}zybKlQOd6L
zlHV8zZD%U#uV|9k^9EH@vU@@c214;u@l}x7&RWT&;-UN1;oecGsV4T@lySWGohcd9
z*3~v~C;@YgZE6E^+09?z1O3NS%?={ZK^1RT2bp=pw~Xc`B&8z(Ho@w$7fue8ZjJ>}
zvY(a?z6_Qd6jIufIXSkxb<e7zty28gU;G3sV&5Wv5dSiSk=sp-;I*#w;-h<zX0;Cu
zmsMnEjZu{UE?FPiS?#ZHRvU5~>Tc&K)jdw+f7Thn^ZHe9*MD9phw<IyVH<}DWYMZ(
zN=8BpbQ$vYJYA@h@5K&y(&87^ro(p1k-r)RQ^q2LD?Imz$@?vc>!{oaV$LpG4La1e
zA`FWR1HR>vtwt7I14*%(Mx94Wrh!AJy8L_WU=10k*7fc^lb}hucEL~&GisFCHSM3{
zh}JkDk7bNcm>rUJ!YI1H!9TJ&3ivaHTn!^q4odaL)!FcCj3aX5G+*vY%rCjkVPiE=
zt>BS3y5^$*sI20F@7E;94EO_o{Bc0gskg|HFnx#aw=!73U_F231IGwY>wyGf)IzJf
zg(;639r~S2Lx#1@!3{wRQ#Gq~2ZS20?Kf{-mL=~h)I1^fXem`)qq}YH1%_x1W60b#
zstV%fB>>AA?k5CK9dAhbtaJ2bBy@qhWu>P>9O~jQE~n_yHY6!0;iH8*+?A)eW;iC{
z>~`0$m?n~j+N{nrR-c(&ZvRR}OdCW><~Yn}iK@Zw_PT%vJaTLW2C>ML`~WjrP+rX$
zdAs#tekk7cETn44YP$$36s9k)SW6=T^a+I*mL$*(9n=Reqg>-rtt%Pw=cgl~8%_sN
zHSj*MC#mW>!QHFV8rrL%5=ZolQo7r2lxjL;X)m{X)QRCHkti%nw;2QT7hW4IdCLPS
zI3txj2)m+>kDoH}?)%6N;PL{>M|n_Zl9!Kyur8pQ2Ry2(FT4M+tlN40;0mHRLn`Wn
z*{?d!HpqC|EJe_BCiht48FP5g?fm!zI(K3#a^|-D1;f&Do7%E5-t}i6rEO}D)d#dh
zbFCAV>GcxZ?c<aTS4u{!Y=M@{VUUH|xdcR5hC2<C;5m6ir$H=d!Lkf3ayO>3%tKaY
zzJ*M^)@A?eKxbjkzYK3??)qMusZ^D_DlJ8-H1bbL(6!`|<JBPgEe6XQ?QJ!6Jn-IL
zoP7{v{Dlj^LWC>aUA5eG{qg88e{eP|vgF+u+~RgQAu<tVF(5$eu{3HWV^OXHgJ&@_
z6@VSM3)vEV<CMmNz9yIO3>oRQ!CZ$uBA}9ZAGV`wpfD!&YiWnNn2P6oypo)UJq-yC
zDQvvi#I%b#lMQ)Lo`&@PRX6mQ7jD3Qo4WZ6!3erU4%=U0F$8pflKIi5u*?x{;AHn`
zm&QqGpKU>RZL=TSpV2hR8lYe<ILO&grw8j8FV3l-!kEU)OYr|m3aKaH^4P3i4<fC{
zO2Y)oM0>?x1`{!E^zjS2R<lmRp=cDRz}?^ge$(hWfM!XJ(isTbTC2+y(|7nA7<@fM
zcRKU1O-7J}NvksB-Nlx67U|$t7IzzRU`j|o{u{W+90#J`(wl+m%w;0I_PO#?tqkx5
zAi3!_@h16vEVf|0yNL;Wft?8}fFTxPd3h-%S`0%S^+6gc)w*+`@?v`ZJzJlK_8Hja
zP7?wp$C-(?vqY<>^-9<e&5Txn2-tapJjRk<Lz%dYhD^QJWQ!}B6&;}h>>4H@ws9Kr
z0t&DI5wj92wlIcgj)%YWX7JA;Zf-Yb5deO(<xCQ~2NAZZ&x7;<#}_jLN(p(GD9*I*
z<AhAEPJtylEyO<s1IGyuoS5Ttev&;WYSlH(g!qK2OW-*;@<!w(NOl2t^hyFsAIZdw
zj@1qOx9J=zG}tXvrfZVISdb=Z&%;dg)yxIAS<`n^k3oXa32EK1edul5{w$LRgT}u~
z>sSfS$YO+{mv4wxxTp#$IJs3%3LI8^C4smJK0aZ8=mlSIk^VG{##N-VUTZIMCie@M
z==5>+v-1G->1;>stgU$`JheaFZz;Y0j_McE@uBjwD-6o+=FJw$JG~C5sfsB!csTkt
z5oL@llc|$T%rsg%yRO0T3n4t?jqjf={``^Pb@9)?@@x?s@qV+Hf~W)z`%$?^=)<W|
zd6>~7`|^KdFt#u_=I(eZ>h1-11GC!g_G1nG3EQovW;=#CV<APU!GMfhbN;deNqNxb
z3*gXmv1V$y*sO8vUHOePmbC8Cb>OatrJo(TqYV8YwtQ9EfE`dct?N2{Cb2cGxh!yo
z7SvAj&NkbQ&88dCZ`f-j$OrsM3%U9tf7#iKUKVkv{;FTGnRI)86@;SCCBU~5+zYQ_
zb0O@ysY#*_O^1ZT2pXA&+Nm4C!zn)9sOvQrW%OUavQnnw^N|^@O6fugA5f;*w^QAA
zTpCFlPT(pYNP3${$Z1)cPHzI~CHH>cSkv3Gux{vSqi0;no~9%NZWN66>6F#1JqRm=
zjp%~gAc3-z;GHR(P`Z{sMmn}doNd9C)tFCmX1SaP?s9tMOtkmkf4JYoOj;u)Q$J}8
z{KlSXhe4q15>di!N5_?Hd-(nx*N8Dmnbwb5Ly3sz1}V5qR{_U`)N`$_f<95Agc%*@
z*TPRzJtM{W2!$nY;Obx@;gU0aYt*)7a#&0eCd!zth-gHFrwgI80kh#&@uxmK1>I0w
z*MuvB{%ZeWK>;3?^ohV8<ApFxdxL?_cajdGf-a$kDR7)_GA=t}vVKtN_>b&W-j@~@
z6sie?dHDdrU*;Qd2{2l*X4XlTadIc$XUa_ZJ#J&;{iXYo_ku1(gmjQBW|XMa-N@|P
z<FDg%NY-V#v8icyaBU4=n0z)VB`Tdw!^E}dl75_$1iUgTwIZRUfFiOWl$Jr;-9H0M
z?)*I|*u@sbS7!84D(uDpj1-oIoD$*~HiN3(?1($^HUwRV>O2o}&u(?7*XP)6uHw$k
zB@s(&5uh-<5Z1YPPf&b*?KuC*RZ*8<w1UcQ1}ZSfe|GLFs#T4NQJ<dGUxls+8cgy!
zbZttmewo>Bza?eiYs$@R@%1*=V^T=c!GteeN;!9<5MhGTH0=6;^Z{D^+3t^Ng=@|5
zP4_&$#Kx$9OASQNDp~>?l$DNZG$^a*NnS<~2bivS$D)Lc>+koLJEB&0-uFKYFD<~8
z4TU%D)u8JKP6>g~s;+kOz1bEA4wn~Z20R{pMORH5$I>BcU4A4C!au#bR!l+UmP*GZ
z3&bJ^o+d*zzZI5*a;UY6ugIet9v^@^lgEJ0x9xypru?kqN-Wy>MvNdPE0f6JH__c_
z36a-;%N_;8x=)&0BuVD`o}Fpj`wDYUMVz?{a70k;BWV8Nu8wj|P54F2st0Y4K&z*;
z=J|p8U_eEe$F>ueyvSst&ZRdqe>R(VfaksCTMWQ=Oh+bhN2dBsEen(wh))OAONr!U
zkti$z1a4W_zrrwIt4$x)!41IOG0pMr#n2L6iW;hwg)sl%G+|d49CD%)bUTLJ8dm!$
zxbQ~$+&n-vjkeF$z#%tgU`F5k6^_)Pc;13BwQ+r%*o&l;g0NK+{K}NyRSk<x(oG~d
zVagpTtS5+_FK#W(?gz8N3ZsP+E9@j}!t07ZYlBQ5e>1^5B)^Jb!Wk|#ZpTX|;A?`K
zQNQbp6H~b@F=+u}5%SpWPVKT6oCOQuS#~=uUHSMvKX%5&YI=7cR%dF~PZp@=z2Lq7
zQiIv3ZF!m`Ny;2IMt8t76%fXD?L%<V=!|f=&%S5P!!?b6TAWN4?R0%(i<m4YFOBqB
zWk2+B%Fhba0CZa1sTi-bI#2BEJKbYYlcD9~G~6?Qz*Va{&c+vWbq4A)kAm`5%mr=s
z;3W1@z0&Pg;e$ueI>+xy__jGf2lu21R#8a?rM?BYR7M(N(E;A@#qjp&(BK|ZBxtqu
zAX3&k*h$!_Lr_K&YTZhYaPRU!y_mO&ZpgnxufjK@?Kn)uTSajLO2}=<5&U!{t#|?A
zrr_AaPur}zgE&Kh_cAm|iEd}o1GwZzHBKAty4{HE!QR*h>B7;0EmhF&Gx1$f->Lgw
z;ZuFVwjgq-&D;y}FZ@)&9eOk!nOYn&xwl;XX=N%fmJNvwXLtCgC=DCAJw<WM+nbZT
z`X>Ch_d6oBQ_nWswEKbiqQ}xnuf2ssAN;y{VmH})(=yYMqy;FxH=_7gY2Ch!V;TLj
zmrwqU3GDbw7L0gv!h5(8bi?|C3a-4di+K*yu$#tzQO+e&ngqRwBU{A;E2)#z$_h)X
zZc3?1WX7iBxzhTiTd3jEp-579+%6URVVS`X))#@ZzlGvVxY;x6IyO%4Q0aN5TA>;E
zBekH9l_e*?F|}!JOUqTR{Eq5E^U07^yezI<+jUtC=b(ZxS56&n6;)_@GSEA&#M&&-
z{s)fdz@PyLXlUf)1VRT&HLmsj`jg%?9gi_I=#^GKC?CY7B*fNczfihTPnZBh_f<<D
z6@1T*^Ia<W??fWWEk!lBG5L|*r+mcnL+a0!>T!1)=4K|+PfYCbq1HH6ItxzvAJi+H
z=<<(F?MfG;RSDR2t>(^9?g`Jd@d^pFw?!#ZEwBYl)UWxVb=Nhe_qpfr-=py!l1Z{o
zsj&aG1<_wjJa6%JxWidKf`FKS5v<oK{>ZXxJqshD+zQEPxm|>`$A;v3*LU^e`$W-z
z1MAo3{0+YEJXi6w+>Va4JH?lretwU&dC+a|h)uM-HKB0h$H%7eVu=^n67|PzV_Q&x
zhwJkN;Q_0wSFFcc@}49Wvuo^3hbTMZD2@0hVQXyyw}?TlGVhL1i#6}`jr8BE&JT0-
zmGvVC57+Mph5G2Do^P;11(z0WzH(fCf7b=A-^I5AIJU&}<^xHJ2ht-*27!7UYmM?*
zP9$`hD>10xLXJgfr@q;LT~$<G->$4+5AkIYDP55y=y_Kp3&(K&h-`drMM|nszFU8r
zB-ZI|#S>rNqWj+dg*O@QB83zjp7oX8>Z@MfpyrS)Ry>Ko37<|Sy`yU!OWtEM_X&pZ
z^dkq-gE{Ka17e0MLD}D4N+@eBH1)IBXxzOOj6eJuy4Y7?kMD|ylC@^yYt?zM(kPA6
zYc<=#!1D3-gMV&Nf%306OV9C1Tq>h~L6Lv*QSh0a^R6c@KjPJa;&hjaG)>YKW{G96
zQqEz;{5qu$B<!;)B|D$q|KEP1iKRtEiR?4wdrvP~qCaKquYTwd4CWg5@veEgE&6AM
z`I}__(RAe+IT>e@#UZdVOgxHeZ6rO90Fmv9X&}EmR2YpHmqdMas>+IrbO+*M2dY>h
zgX2=NK)#qZxZh2jk<G!DN-1De$~Y+(k#pxehb$!ckcRMn$`Wz5A|+Fg1PxAH+L&kd
zVjbXWU&{WBn+TAjXg+0%8GiZqJVpBHU*=0nl&ezN98lP+J1DZfuVT@r33*(Uh`|`y
z0bOe+Kc)X{BPtTieq<HN3BLuUANY2_fQ+P;v>qvP(!AY>azHJ}{76%Au<z@(iC}ul
z;TYcGiYNo38b|>jL?g*DiNA`zC`w7CHjrhar)r|s&*;no<0+|Cx+#+4)uO2R|56mV
z*h+%TzC-3qkF5JI-LE_(EU&o#)RHxuS8LjikiXjOeko{ye)S*Nd20D;MQad1Dy_#-
z<o~<-H=8y>00V*C9((@y-hW=9+Reon_VP7W-`?_H4LH~V17WloE&Z>7RIXdx?0Ym>
z|8C9PXvweOZoZ4K9+v?+#^pqONyF=8M=$O$llbx>FrB@1-kq)tR04~$+9vbA8$_0t
z=FD{-I^`D;R#w9zRai12FMMA<NByl`7IClJMD1w40m@)1B<H_Fgh%QNDOY;EkYxO`
z?Tey;t0+HMY_9vW6-L*AS(_N%`Hl<bUlHnFZ7@@19X;gy(-kkk8Ko5U-NgMc|5fK1
zmm6x2i`7(7Fj%<N&%j*vXT%f{fu<uZ_g<SX1H}%qMB9wV1K4}V2#bdiU*LU1nnklx
apSPng?5)|4UTVAm{McFV*;%-Q82LY<tWGQd

literal 0
HcmV?d00001

diff --git a/docs/community/index.md b/docs/community/index.md
index 517bf59c5..9519f5b56 100644
--- a/docs/community/index.md
+++ b/docs/community/index.md
@@ -1,9 +1,12 @@
 # Community
 
+![Belonging](belonging.png)
+
 Outlines exists for a community of users who believe software doesn't need to be complicated. Who share the same passion for Large Language Models but don't want to compromise on robustness. Together, we are bringing these powerful models back to the world of software.
 
 ## Connect on Discord
 
 The Outlines community lives on our Discord server. There you can ask questions, share ideas or just chat with people like you. Don't be a stranger and [join us][discord].
 
+
 [discord]: https://discord.gg/UppQmhEpe8
diff --git a/docs/community/versioning.md b/docs/community/versioning.md
index a6e12f218..023b92537 100644
--- a/docs/community/versioning.md
+++ b/docs/community/versioning.md
@@ -13,10 +13,6 @@ Each part of the version number (`major.minor.patch`) conveys information about
 - **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatability with existing use of core interfaces.
 - **Patch Releases** address bug fixes and incorporate low-risk changes to improve stability and performance.
 
-!!! note "Breaking Changes"
-
-    Outlines v1.0 introduced several breaking changes to the core interface. See [the migration guide](/user_guide/migration) for more details.
-
 ## Releases
 
 Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/dottxt-ai/outlines/releases).
diff --git a/docs/examples/atomic_caption.md b/docs/cookbook/atomic_caption.md
similarity index 75%
rename from docs/examples/atomic_caption.md
rename to docs/cookbook/atomic_caption.md
index 08782594f..a2b43e5e2 100644
--- a/docs/examples/atomic_caption.md
+++ b/docs/cookbook/atomic_caption.md
@@ -1,6 +1,5 @@
 # Vision-Language Models with Outlines
-
-This guide demonstrates how to use Outlines with vision-language models, leveraging the outlines `TransformersMultiModal` model. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
+This guide demonstrates how to use Outlines with vision-language models, leveraging the new transformers_vision module. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
 
 We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
 
@@ -12,21 +11,17 @@ pip install outlines transformers torch
 ```
 
 ### Initializing the Model
-We'll use the `outlines.from_transformers` function to initialize our vision-language model. For this function to return a vision multi-modal model we need to pass in a transformers model and a transformers processor that can handle both text and image inputs. Today we'll be using the Pixtral model with the AutoProcessor.
+We'll use the transformers_vision function to initialize our vision-language model. This function is specifically designed to handle models that can process both text and image inputs. Today we'll be using the Pixtral model with the llama tokenizer. (Currently the mistral tokenizer is pending support).
 
 ```python
-import outlines
 import torch
 from transformers import (
-    AutoProcessor,
-    LlavaForConditionalGeneration
+    LlavaForConditionalGeneration,
 )
-
 model_name="mistral-community/pixtral-12b" # original magnet model is able to be loaded without issue
 model_class=LlavaForConditionalGeneration
-processor_class=AutoProcessor
 
-def get_vision_model(model_name: str, model_class, processor_class):
+def get_vision_model(model_name: str, model_class: VisionModel):
     model_kwargs = {
         "torch_dtype": torch.bfloat16,
         "attn_implementation": "flash_attention_2",
@@ -36,25 +31,28 @@ def get_vision_model(model_name: str, model_class, processor_class):
         "device": "cuda",
     }
 
-    model = outlines.from_transformers(
-        model_class.from_pretrained(model_name, **model_kwargs),
-        processor_class.from_pretrained(model_name, **processor_kwargs),
+    model = outlines.models.transformers_vision(
+        model.model_name,
+        model_class=model.model_class,
+        model_kwargs=model_kwargs,
+        processor_kwargs=processor_kwargs,
     )
     return model
-model = get_vision_model(model_name, model_class, processor_class)
+model = get_vision_model(model_name, model_class)
 ```
 
 ### Defining the Schema
-Next, we'll define a schema for the output we expect from our vision multi-modal model. This schema will help structure the model's responses. We use the `outlines.Generator` object to create a generator for our schema that will then be called with our prompt and images.
+Next, we'll define a schema for the output we expect from our vision-language model. This schema will help structure the model's responses.
+
 
 ```python
-from enum import Enum
 from pydantic import BaseModel, Field, confloat, constr
 from pydantic.types import StringConstraints, PositiveFloat
 from typing import List
 from typing_extensions import Annotated
 
-class TagType(Enum):
+from enum import StrEnum
+class TagType(StrEnum):
     ENTITY = "Entity"
     RELATIONSHIP = "Relationship"
     STYLE = "Style"
@@ -88,7 +86,7 @@ class ImageData(BaseModel):
     short_caption: Annotated[str, StringConstraints(min_length=10, max_length=150)]
     dense_caption: Annotated[str, StringConstraints(min_length=100, max_length=2048)]
 
-image_data_generator = outlines.Generator(model, ImageData)
+image_data_generator = outlines.generate.json(model, ImageData)
 ```
 
 This schema defines the structure for image tags, including categories like Entity, Relationship, Style, etc., as well as short and dense captions.
@@ -97,6 +95,7 @@ This schema defines the structure for image tags, including categories like Enti
 
 We'll create a prompt that instructs the model on how to analyze the image and generate the structured output:
 
+
 ```python
 pixtral_instruction = """
 <s>[INST]
@@ -132,34 +131,30 @@ pixtral_instruction = """
 </Examples>
 </TagCategories>
 <ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
-<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG]<image>[/INST]
+<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG][/INST]
 """.strip()
 ```
 
-This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required. It's essential to include the <image> tag in the prompt at the location where the image will be inserted.
+This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required.
 
 ### Generating Structured Output
 Now we can use our model to generate structured output based on an input image:
 
 ```python
-from io import BytesIO
-from urllib.request import urlopen
-from PIL import Image
-
 def img_from_url(url):
     img_byte_stream = BytesIO(urlopen(url).read())
     return Image.open(img_byte_stream).convert("RGB")
 
 image_url="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
 image= img_from_url(image_url)
-result = image_data_generator({
-    "text": pixtral_instruction,
-    "images": image
-})
+result = image_data_generator(
+    pixtral_instruction,
+    [image]
+)
 print(result)
 ```
 
-This code loads an image from a URL, passes it to our vision multi-modal model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
+This code loads an image from a URL, passes it to our vision-language model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
 
 ```json
 {'tags_list': [{'tag': 'astronaut',
@@ -189,6 +184,6 @@ This code loads an image from a URL, passes it to our vision multi-modal model a
 ```
 
 ## Conclusion
-The `TransformersMultiModal` model in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
+The transformers_vision module in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
 
 By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
diff --git a/docs/examples/chain_of_density.md b/docs/cookbook/chain_of_density.md
similarity index 74%
rename from docs/examples/chain_of_density.md
rename to docs/cookbook/chain_of_density.md
index 8a822fc2e..10bda9771 100644
--- a/docs/examples/chain_of_density.md
+++ b/docs/cookbook/chain_of_density.md
@@ -26,12 +26,42 @@ The prompt also asks the model to return a list of JSON objects that contain the
 
 ![Figure 2 in the paper](./images/chain_of_density.png)
 
-We can now implement the prompt provided in the paper. We stored the prompt template in a text file, and we can load it using the `Template` class:
+We can now implement the prompt provided in the paper:
 
 ```python
 from outlines import Template
 
-chain_of_density = Template.from_file("prompt_templates/chain_of_density.txt")
+
+chain_of_density = Template.from_string(
+    """Article: {{ article }}
+
+    You will generate increasingly concise, entity-dense summaries of the above Article.
+
+    Repeat the following 2 steps 5 times.
+
+    Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
+    Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
+
+    A Missing Entity is:
+    - Relevant: to the main story.
+    - Specific: descriptive yet concise (5 words or fewer).
+    - Novel: not in the previous summary.
+    - Faithful: present in the Article.
+    - Anywhere: located anywhere in the Article.
+
+    Guidelines:
+    - The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
+    - Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
+    - Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
+    - The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
+    - Missing entities can appear anywhere in the new summary.
+    - Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
+
+    Remember, use the exact same number of words for each summary.
+
+    Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
+    """
+)
 ```
 
 ??? Note
@@ -54,26 +84,19 @@ class Summaries(BaseModel):
     summaries: conlist(Summary, max_length=5, min_length=5)
 ```
 
-We now generate the prompt by passing the article we want to summarize to the prompt template previously loaded. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use the `Summaries` schema to generate the summaries with structured generation:
+We now generate the prompt by passing the article we want to summarize to the template. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use JSON-structured generation to generate the summaries:
 
 ```python
-import outlines
-import transformers
+model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ")
 
-MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
-
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
-    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
-)
-prompt = chain_of_density(article=article)
-result = model(prompt, Summaries, max_new_tokens=2000)
+prompt = chain_of_density(article)
+result = outlines.generate.json(model, Summaries)(prompt)
 ```
 
 We can now check the results:
 
 ```python
-print(result)
+print(result.model_dump())
 # {'summaries': [
 #     {
 #       'missing_entities': 'English mathematician, cryptanalyst, philosopher',
diff --git a/docs/examples/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
similarity index 63%
rename from docs/examples/chain_of_thought.md
rename to docs/cookbook/chain_of_thought.md
index 42d6c8550..5ac18af25 100644
--- a/docs/examples/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -11,23 +11,20 @@ We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-
 pip install llama-cpp-python
 ```
 
-To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
-
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -41,9 +38,20 @@ model = outlines.from_llamacpp(llm)
     We initialize the model:
 
     ```python
+    import llama_cpp
     from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
     ```
 
 ## Chain of thought
@@ -69,27 +77,30 @@ class Reasoning(BaseModel):
 json_schema = Reasoning.model_json_schema()
 ```
 
+We could generate a response using the json schema but for a change we will use the regex:
+
+```python
+from outlines.fsm.json_schema import convert_json_schema_to_str
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+schema_str = convert_json_schema_to_str(json_schema=json_schema)
+regex_str = build_regex_from_schema(schema_str)
+```
+
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-from outlines import Template
-
-generate_hermes_prompt = Template.from_string(
-    """
-    <|im_start|>system
-    You are a world class AI model who answers questions in JSON
-    Here's the json schema you must adhere to:
-    <schema>
-    {{ json_schema }}
-    </schema>
-    <|im_end|>
-    <|im_start|>user
-    {{ user_prompt }}
-    <|im_end|>
-    <|im_start|>assistant
-    <schema>
-    """
-)
+def generate_hermes_prompt(user_prompt):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON "
+        f"Here's the json schema you must adhere to:\n<schema>\n{json_schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + user_prompt
+        + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
 ```
 
 For a given user prompt:
@@ -98,11 +109,11 @@ For a given user prompt:
 user_prompt = "9.11 and 9.9 -- which is bigger?"
 ```
 
-We can use `outlines.Generator` with the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+we can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-generator = outlines.Generator(model, regex_str)
-prompt = generate_hermes_prompt(json_schema=json_schema, user_prompt=user_prompt)
+generator = generate.regex(model, regex_str)
+prompt = generate_hermes_prompt(user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
diff --git a/docs/cookbook/classification.md b/docs/cookbook/classification.md
new file mode 100644
index 000000000..023e4543a
--- /dev/null
+++ b/docs/cookbook/classification.md
@@ -0,0 +1,106 @@
+# Classification
+
+Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
+
+
+This tutorial shows how one can implement multi-label classification using Outlines. We will use two functionalities of the library: `generate.choice` and `generate.json`.
+
+As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
+
+```python
+import outlines
+
+model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+```
+
+We will use the following prompt template:
+
+```python
+from outlines import Template
+
+
+customer_support = Template.from_string(
+    """You are an experienced customer success manager.
+
+    Given a request from a client, you need to determine when the
+    request is urgent using the label "URGENT" or when it can wait
+    a little with the label "STANDARD".
+
+    # Examples
+
+    Request: "How are you?"
+    Label: STANDARD
+
+    Request: "I need this fixed immediately!"
+    Label: URGENT
+
+    # TASK
+
+    Request: {{ request }}
+    Label: """
+)
+```
+
+## Choosing between multiple choices
+
+Outlines provides a shortcut to do multi-label classification, using the `outlines.generate.choice` function to initialize a generator. Outlines uses multinomial sampling by default, here we will use the greedy sampler to get the label with the highest probability:
+
+```python
+from outlines.samplers import greedy
+
+generator = outlines.generate.choice(model, ["URGENT", "STANDARD"], sampler=greedy())
+```
+Outlines supports batched requests, so we will pass two requests to the model:
+
+```python
+requests = [
+    "My hair is one fire! Please help me!!!",
+    "Just wanted to say hi"
+]
+
+prompts = [customer_support(request) for request in requests]
+```
+
+We can now asks the model to classify the requests:
+
+```python
+labels = generator(prompts)
+print(labels)
+# ['URGENT', 'STANDARD']
+```
+
+Now, you might be in a hurry and don't want to wait until the model finishes completion. After all, you only need to see the first letter of the response to know whether the request is urgent or standard. You can instead stream the response:
+
+```python
+tokens = generator.stream(prompts)
+labels = ["URGENT" if "U" in token else "STANDARD" for token in next(tokens)]
+print(labels)
+# ['URGENT', 'STANDARD']
+```
+
+## Using JSON-structured generation
+
+Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
+
+```python
+from enum import Enum
+from pydantic import BaseModel
+
+
+class Label(str, Enum):
+    urgent = "URGENT"
+    standard = "STANDARD"
+
+
+class Classification(BaseModel):
+    label: Label
+```
+
+and we can use `generate.json` by passing this Pydantic model we just defined, and call the generator:
+
+```python
+generator = outlines.generate.json(model, Classification, sampler=greedy())
+labels = generator(prompts)
+print(labels)
+# [Classification(label=<Label.urgent: 'URGENT'>), Classification(label=<Label.standard: 'STANDARD'>)]
+```
diff --git a/docs/examples/dating_profiles.md b/docs/cookbook/dating_profiles.md
similarity index 92%
rename from docs/examples/dating_profiles.md
rename to docs/cookbook/dating_profiles.md
index 5852e6df9..1cbea556c 100644
--- a/docs/examples/dating_profiles.md
+++ b/docs/cookbook/dating_profiles.md
@@ -3,7 +3,6 @@
 In this example we will see how we can use Outlines to generate synthetic data for a dating application. This example was originally contributed by [Vibhor Kumar](https://github.com/veezbo).
 
 ```python
-import json
 from dataclasses import dataclass
 from enum import Enum
 
@@ -133,21 +132,20 @@ samples: list[Example] = [
 We will use Mosaic's MPT-7B model (requires 13GB of GPU memory) which can fit on a single GPU with a reasonable context window. We initialize it with Outlines:
 
 ```python
-MODEL_NAME = "mosaicml/mpt-7b-8k-instruct"
-
 config = transformers.AutoConfig.from_pretrained(
-    MODEL_NAME, trust_remote_code=True
+    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
 )
 config.init_device = "meta"
-model_kwargs = {
-    "config": config,
-    "trust_remote_code": True,
-    "torch_dtype": torch.bfloat16,
-    "device_map": "cuda",
-}
-tf_model = transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME, **model_kwargs)
-tf_tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
-model = outlines.from_transformers(tf_model, tokenizer=tf_tokenizer)
+model = outlines.models.transformers(
+    model_name="mosaicml/mpt-7b-8k-instruct",
+    device="cuda",
+    model_kwargs={
+        "config": config,
+        "trust_remote_code": True,
+        "torch_dtype": torch.bfloat16,
+        "device_map": {"": 0},
+    },
+)
 ```
 
 ## JSON-structured generation of profiles
@@ -164,9 +162,9 @@ every few  months. I also like weddings because I get to wear those suits, and
 it's  a good excuse for a date. I watch the latest series because I'm paying,
 with my hard-earned money, for every streaming service."""
 
-prompt = dating_profile_prompt(description=new_description, examples=samples)
-profile = model(prompt, DatingProfile)
-parsed_profile = DatingProfile.model_validate_json(json.loads(profile))
+prompt = dating_profile_prompt(new_description, samples)
+profile = outlines.generate.json(model, DatingProfile)(prompt)
+parsed_profile = DatingProfile.model_validate_json(profile)
 ```
 
 ## Results
diff --git a/docs/examples/deploy-using-bentoml.md b/docs/cookbook/deploy-using-bentoml.md
similarity index 91%
rename from docs/examples/deploy-using-bentoml.md
rename to docs/cookbook/deploy-using-bentoml.md
index 5d000cb52..2926df0f6 100644
--- a/docs/examples/deploy-using-bentoml.md
+++ b/docs/cookbook/deploy-using-bentoml.md
@@ -119,22 +119,15 @@ class Outlines:
     bento_model_ref = bentoml.models.get(BENTO_MODEL_TAG)
 
     def __init__(self) -> None:
+
         import outlines
         import torch
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-
-        # Load tokenizer and model from the BentoML model reference path
-        hf_tokenizer = AutoTokenizer.from_pretrained(self.bento_model_ref.path)
-        hf_model = AutoModelForCausalLM.from_pretrained(
+        self.model = outlines.models.transformers(
             self.bento_model_ref.path,
-            torch_dtype=torch.float16,
-            low_cpu_mem_usage=True,
-            device_map="cuda"
+            device="cuda",
+            model_kwargs={"torch_dtype": torch.float16},
         )
 
-        # Then use the loaded model with Outlines
-        self.model = outlines.from_transformers(hf_model, hf_tokenizer)
-
     ...
 ```
 
@@ -149,14 +142,13 @@ We then need to define an HTTP endpoint using `@bentoml.api` to decorate the met
         prompt: str = "Give me a character description.",
         json_schema: t.Optional[str] = DEFAULT_SCHEMA,
     ) -> t.Dict[str, t.Any]:
-        import json
+
         import outlines
-        from outlines.types import JsonSchema
 
-        generator = outlines.Generator(self.model, JsonSchema(json_schema))
+        generator = outlines.generate.json(self.model, json_schema)
         character = generator(prompt)
 
-        return json.loads(character)
+        return character
 ```
 
 Here `@bentoml.api` decorator defines `generate` as an HTTP endpoint that accepts a JSON request body with two fields: `prompt` and `json_schema` (optional, which allows HTTP clients to provide their own JSON schema). The type hints in the function signature will be used to validate incoming JSON requests. You can define as many HTTP endpoints as you want by using `@bentoml.api` to decorate other methods of `Outlines` class.
diff --git a/docs/examples/deploy-using-cerebrium.md b/docs/cookbook/deploy-using-cerebrium.md
similarity index 90%
rename from docs/examples/deploy-using-cerebrium.md
rename to docs/cookbook/deploy-using-cerebrium.md
index bf010d583..6fbaab390 100644
--- a/docs/examples/deploy-using-cerebrium.md
+++ b/docs/cookbook/deploy-using-cerebrium.md
@@ -36,7 +36,7 @@ provider = "aws"
 region = "us-east-1"
 
 [cerebrium.dependencies.pip]
-outline = "==1.0.0"
+outline = "==0.0.37"
 transformers = "==4.38.2"
 datasets = "==2.18.0"
 accelerate = "==0.27.2"
@@ -48,14 +48,9 @@ Running code in Cerebrium is like writing normal python with no special syntax.
 
 ```python
 import outlines
-import transformers
-from outlines.types import JsonSchema
 
 
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
 
 schema = """{
     "title": "Character",
@@ -94,7 +89,7 @@ schema = """{
     }
 }"""
 
-generator = outlines.Generator(model, JsonSchema(schema))
+generator = outlines.generate.json(model, schema)
 ```
 
 On first deploy, it will download the model and store it on disk therefore for subsequent calls it will load the model from disk.
diff --git a/docs/examples/deploy-using-modal.md b/docs/cookbook/deploy-using-modal.md
similarity index 93%
rename from docs/examples/deploy-using-modal.md
rename to docs/cookbook/deploy-using-modal.md
index 23a846a6d..15e200cb5 100644
--- a/docs/examples/deploy-using-modal.md
+++ b/docs/cookbook/deploy-using-modal.md
@@ -68,12 +68,7 @@ When running longer Modal apps, it's recommended to download your language model
 # downloading models, setting up environment variables, etc.
 def import_model():
     import outlines
-    import transformers
-
-    outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(language_model),
-        transformers.AutoTokenizer.from_pretrained(language_model)
-    )
+    outlines.models.transformers(language_model)
 
 # This line tells the container to run the import_model function when it starts.
 outlines_image = outlines_image.run_function(import_model)
@@ -139,22 +134,19 @@ def generate(
     # so we need to import the necessary libraries here. You should
     # do this with any other libraries you might need.
     import outlines
-    import transformers
-    from outlines.types import JsonSchema
 
     # Load the model into memory. The import_model function above
     # should have already downloaded the model, so this call
     # only loads the model into GPU memory.
-    outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(language_model, device_map="cuda"),
-        transformers.AutoTokenizer.from_pretrained(language_model)
+    model = outlines.models.transformers(
+        language_model, device="cuda"
     )
 
     # Generate a character description based on the prompt.
     # We use the .json generation method -- we provide the
     # - model: the model we loaded above
     # - schema: the JSON schema we defined above
-    generator = outlines.Generator(model, JsonSchema(schema))
+    generator = outlines.generate.json(model, schema)
 
     # Make sure you wrap your prompt in instruction tags ([INST] and [/INST])
     # to indicate that the prompt is an instruction. Instruction tags can vary
diff --git a/docs/examples/earnings-reports.md b/docs/cookbook/earnings-reports.md
similarity index 94%
rename from docs/examples/earnings-reports.md
rename to docs/cookbook/earnings-reports.md
index cca1e9513..26c2a9ccf 100644
--- a/docs/examples/earnings-reports.md
+++ b/docs/cookbook/earnings-reports.md
@@ -26,14 +26,16 @@ Choose your language model. We'll use Phi-3 mini, which is small enough to run o
 ```python
 import outlines
 import torch
-import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-tf_model = transformers.AutoModelForCausalLM.from_pretrained(
-    model_name, device_map="cuda", torch_dtype=torch.bfloat16
+model = outlines.models.transformers(
+    model_name,
+    device='auto',
+    model_kwargs={
+        # To reduce memory usage, we'll use bfloat16
+        "torch_dtype": torch.bfloat16,
+    },
 )
-tf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
-model = outlines.from_transformers(tf_model, tf_tokenizer)
 ```
 
 ## Set up the data
@@ -154,7 +156,7 @@ year,revenue,operating_income,net_income,basic_earnings_per_share(
 \d{4},((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),(-?\d+(?:\.\d{1,2})?)){,3}
 ```
 
-Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.Generator` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
+Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.generate.regex` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
 
 ## Prompting the model
 
@@ -232,12 +234,12 @@ def extract_financial_data_prompt(columns_to_extract, income_statement):
 
 Now that we have our prompt and regular expression, we can run the model.
 
-Construct our regex extractor function.
+Construct our regex extractor function. We'll use a greedy sampler, which samples the most likely next token at each step. It's a simple sampler that is more reproducible than multinomial sampling.
 
 ```python
-from outlines.types import Regex
-
-csv_extractor = outlines.Generator(model, Regex(csv_regex))
+csv_extractor = outlines.generate.regex(
+    model, csv_regex, sampler=outlines.samplers.greedy()
+)
 ```
 
 Provide the prompt to the model and run it:
@@ -245,7 +247,7 @@ Provide the prompt to the model and run it:
 ```python
 csv_data = csv_extractor(
     extract_financial_data_prompt(columns_to_extract, income_statement),
-    max_new_tokens=1024,
+    max_tokens=1024,
 )
 
 print(csv_data)
diff --git a/docs/examples/extract_event_details.md b/docs/cookbook/extract_event_details.md
similarity index 94%
rename from docs/examples/extract_event_details.md
rename to docs/cookbook/extract_event_details.md
index 063433ad2..0f87d9586 100644
--- a/docs/examples/extract_event_details.md
+++ b/docs/cookbook/extract_event_details.md
@@ -8,7 +8,7 @@ place, 3 avenue des tanneurs, one hour will be enough see you 😘
 ```
 
 Let see how to extract the event details from the message with the MLX
-library dedicated to Apple Silicon processor (M series).
+library dedicated to  Apple Silicon processor  (M series).
 
 ```python
 --8<-- "docs/cookbook/extract_event_details.py"
diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
new file mode 100644
index 000000000..b51f8d921
--- /dev/null
+++ b/docs/cookbook/extract_event_details.py
@@ -0,0 +1,46 @@
+from datetime import datetime
+
+from pydantic import BaseModel, Field
+
+from outlines import generate, models
+
+# Load the model
+model = models.mlxlm("mlx-community/Hermes-3-Llama-3.1-8B-8bit")
+
+
+# Define the event schema using Pydantic
+class Event(BaseModel):
+    title: str = Field(description="title of the event")
+    location: str
+    start: datetime = Field(
+        default=None, description="date of the event if available in iso format"
+    )
+
+
+# Get the current date and time
+now = datetime.now().strftime("%A %d %B %Y and it's %H:%M")
+
+# Define the prompt
+prompt = f"""
+Today's date and time are {now}
+Given a user message, extract information of the event like date and time in iso format, location and title.
+If the given date is relative, think step by step to find the right date.
+Here is the message:
+"""
+
+# Sample message
+message = """Hello Kitty, my grandmother will be here , I think it's better to postpone our
+appointment to review math lessons to next Friday at 2pm at the same place, 3 avenue des tanneurs, I think that one hour will be enough
+see you 😘 """
+
+# Create the generator
+generator = generate.json(model, Event)
+
+# Extract the event information
+event = generator(prompt + message)
+
+# Print the current date and time
+print(f"Today: {now}")
+
+# Print the extracted event information in JSON format
+print(event.json())
diff --git a/docs/examples/extraction.md b/docs/cookbook/extraction.md
similarity index 85%
rename from docs/examples/extraction.md
rename to docs/cookbook/extraction.md
index cfb39ca5c..dfcca0cfd 100644
--- a/docs/examples/extraction.md
+++ b/docs/cookbook/extraction.md
@@ -7,14 +7,9 @@ Getting LLMs to output the extracted entities in a structured format can be chal
 As always, we start with initializing the model. We will be using a quantized version of Mistal-7B-v0.1 (we're GPU poor):
 
 ```python
-import transformers
 import outlines
 
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name, device_map="cuda"),
-    transformers.AutoTokenizer.from_pretrained(model_name),
-)
+model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
 ```
 
 And we will be using the following prompt template:
@@ -22,6 +17,7 @@ And we will be using the following prompt template:
 ```python
 from outlines import Template
 
+
 take_order = Template.from_string(
     """You are the owner of a pizza parlor. Customers \
     send you orders from which you need to extract:
@@ -74,14 +70,14 @@ orders = [
     "Hi! I would like to order two pepperonni pizzas and would like them in 30mins.",
     "Is it possible to get 12 margheritas?"
 ]
-prompts = [take_order(order=order) for order in orders]
+prompts = [take_order(order) for order in orders]
 
-generator = outlines.Generator(model, Order)
+generator = outlines.generate.json(model, Order)
 
 results = generator(prompts)
 print(results)
-# ['{"pizza": "Pepperoni", "number": 2}',
-# '{"pizza": "Margherita", "number": 12}']
+# [Order(pizza=<Pizza.pepperonni: 'Pepperoni'>, number=2),
+#  Order(pizza=<Pizza.margherita: 'Margherita'>, number=12)]
 ```
 
 There are several ways you could improve this example:
diff --git a/docs/examples/images/chain_of_density.png b/docs/cookbook/images/chain_of_density.png
similarity index 100%
rename from docs/examples/images/chain_of_density.png
rename to docs/cookbook/images/chain_of_density.png
diff --git a/docs/examples/images/coding_structure_diagram.png b/docs/cookbook/images/coding_structure_diagram.png
similarity index 100%
rename from docs/examples/images/coding_structure_diagram.png
rename to docs/cookbook/images/coding_structure_diagram.png
diff --git a/docs/examples/images/knowledge-graph-extraction.png b/docs/cookbook/images/knowledge-graph-extraction.png
similarity index 100%
rename from docs/examples/images/knowledge-graph-extraction.png
rename to docs/cookbook/images/knowledge-graph-extraction.png
diff --git a/docs/examples/images/nvidia-income.png b/docs/cookbook/images/nvidia-income.png
similarity index 100%
rename from docs/examples/images/nvidia-income.png
rename to docs/cookbook/images/nvidia-income.png
diff --git a/docs/examples/images/simtom.png b/docs/cookbook/images/simtom.png
similarity index 100%
rename from docs/examples/images/simtom.png
rename to docs/cookbook/images/simtom.png
diff --git a/docs/examples/images/trader-joes-receipt.jpg b/docs/cookbook/images/trader-joes-receipt.jpg
similarity index 100%
rename from docs/examples/images/trader-joes-receipt.jpg
rename to docs/cookbook/images/trader-joes-receipt.jpg
diff --git a/docs/examples/index.md b/docs/cookbook/index.md
similarity index 82%
rename from docs/examples/index.md
rename to docs/cookbook/index.md
index 0181d137a..c36b98969 100644
--- a/docs/examples/index.md
+++ b/docs/cookbook/index.md
@@ -4,23 +4,15 @@ This part of the documentation provides a few cookbooks that you can browse to g
 
 - [Classification](classification.md): Classify customer requests.
 - [Named Entity Extraction](extraction.md): Extract information from pizza orders.
-- [Dating Profiles](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
+- [Dating Profile](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
 - [Chain Of Density](chain_of_density.md): Summarize documents using chain of density prompting and JSON-structured generation.
 - [Playing Chess](models_playing_chess.md): Make Phi-3 Mini play chess against itself using regex-structured generation.
 - [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
 - [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
-- [Structured Generation Workflow](structured_generation_workflow.md):
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
-- [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
-- [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
 - [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
+- [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
 - [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
-- [Extract Events Details](extract_event_details.md):
-
-Run Outlines on the cloud:
-
-- [BentoML](deploy-using-bentoml.md)
-- [Cerebrium](deploy-using-cerebrium.md)
-- [Modal](deploy-using-modal.md)
+- [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
diff --git a/docs/examples/knowledge_graph_extraction.md b/docs/cookbook/knowledge_graph_extraction.md
similarity index 60%
rename from docs/examples/knowledge_graph_extraction.md
rename to docs/cookbook/knowledge_graph_extraction.md
index f4b21c84c..6a6877756 100644
--- a/docs/examples/knowledge_graph_extraction.md
+++ b/docs/cookbook/knowledge_graph_extraction.md
@@ -8,23 +8,20 @@ We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama
 pip install llama-cpp-python
 ```
 
-To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
-
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -38,9 +35,20 @@ model = outlines.from_llamacpp(llm)
     We initialize the model:
 
     ```python
+    import llama_cpp
     from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
     ```
 
 ## Knowledge Graph Extraction
@@ -84,24 +92,17 @@ schema = KnowledgeGraph.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-from outlines import Template
-
-generate_hermes_prompt = Template.from_string(
-    """
-    <|im_start|>system
-    You are a world class AI model who answers questions in JSON
-    Here's the json schema you must adhere to:
-    <schema>
-    {{ schema }}
-    </schema>
-    <|im_end|>
-    <|im_start|>user
-    {{ user_prompt }}
-    <|im_end|>
-    <|im_start|>assistant
-    <schema>
-    """
-)
+def generate_hermes_prompt(user_prompt):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + user_prompt
+        + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
 ```
 
 For a given user prompt, for example:
@@ -110,26 +111,27 @@ For a given user prompt, for example:
 user_prompt = "Alice loves Bob and she hates Charlie."
 ```
 
-We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-from outlines import Generator
+from outlines import generate, models
 
-generator = Generator(model, KnowledgeGraph)
-prompt = generate_hermes_prompt(schema=schema, user_prompt=user_prompt)
+model = models.LlamaCpp(llm)
+generator = generate.json(model, KnowledgeGraph)
+prompt = generate_hermes_prompt(user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
 We obtain the nodes and edges of the knowledge graph:
 
 ```python
-print(response)
-# {"nodes":[{"id":1,"label":"Alice","property":"loves,hates"},
-# {"id":2,"label":"Bob","property":"loved_by"},
-# {"id":3,"label":"Charlie","property":"hated_by"}],
-# "edges":[{"source":1,"target":2,"label":"loves","property":"love"},
-# {"source":1,"target":3,"label":"hates","property":"hate"}]}
-
+print(response.nodes)
+print(response.edges)
+# [Node(id=1, label='Alice', property='Person'),
+# Node(id=2, label='Bob', property='Person'),
+# Node(id=3, label='Charlie', property='Person')]
+# [Edge(source=1, target=2, label='love', property='Relationship'),
+# Edge(source=1, target=3, label='hate', property='Relationship')]
 ```
 
 ## (Optional) Visualizing the Knowledge Graph
@@ -140,10 +142,10 @@ We can use the [Graphviz library](https://graphviz.readthedocs.io/en/stable/) to
 from graphviz import Digraph
 
 dot = Digraph()
-for node in response["nodes"]:
-    dot.node(str(node["id"]), node["label"], shape='circle', width='1', height='1')
-for edge in response["edges"]:
-    dot.edge(str(edge["source"]), str(edge["target"]), label=edge["label"])
+for node in response.nodes:
+    dot.node(str(node.id), node.label, shape='circle', width='1', height='1')
+for edge in response.edges:
+    dot.edge(str(edge.source), str(edge.target), label=edge.label)
 
 dot.render('knowledge-graph.gv', view=True)
 ```
diff --git a/docs/examples/models_playing_chess.md b/docs/cookbook/models_playing_chess.md
similarity index 77%
rename from docs/examples/models_playing_chess.md
rename to docs/cookbook/models_playing_chess.md
index b8d89958f..a0df2eb2d 100644
--- a/docs/examples/models_playing_chess.md
+++ b/docs/cookbook/models_playing_chess.md
@@ -1,6 +1,6 @@
 # Large language models playing chess
 
-In this example we will make a Phi-3 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
+In this example we will make a Phi-2 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
 
 ## The chessboard
 
@@ -18,33 +18,29 @@ board = chess.Board("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
 
 ## The opponents
 
-Phi-3 will be playing against itself:
+Phi-2 will be playing against itself:
 
 ```python
-import transformers
-import outlines
-
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name),
-    transformers.AutoTokenizer.from_pretrained(model_name),
-)
+from outlines import models
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+
 ```
 
 ## A little help for the language model
 
-To make sure Phi-3 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
+To make sure Phi-2 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
 
 ```python
 import re
-from outlines.types.dsl import either, String
 
 def legal_moves_regex(board):
     """Build a regex that only matches valid moves."""
     legal_moves = list(board.legal_moves)
     legal_modes_str = [board.san(move) for move in legal_moves]
     legal_modes_str = [re.sub(r"[+#]", "", move) for move in legal_modes_str]
-    regex_pattern = either(*[String(move) for move in legal_modes_str])
+    regex_pattern = "|".join(re.escape(move) for move in legal_modes_str)
+    regex_pattern = f"{regex_pattern}"
     return regex_pattern
 ```
 
@@ -62,11 +58,13 @@ We update the prompt at each step so it reflects the state of the board after th
 ## Let's play
 
 ```python
+from outlines import generate
+
 board_state = " "
 turn_number = 0
 while not board.is_game_over():
     regex_pattern = legal_moves_regex(board)
-    structured = model(prompt + board_state, regex_pattern)
+    structured = generate.regex(model, regex_pattern)(prompt + board_state)
     move = board.parse_san(structured)
 
     if turn_number % 2 == 0 :  # It's White's turn
@@ -81,7 +79,7 @@ while not board.is_game_over():
     print(board_state)
 ```
 
-Interestingly enough, Phi-3 hates capturing.
+Interestingly enough, Phi-2 hates capturing.
 
 ```pgn
  e4 e5 1.Nf3 Ne7 3.b4 Nf5 5.Nc3 Ne7 7.Bb5 a6 9.Na4 b6 11.c3 Nec6 13.c4 a5 15.d4 Qg5 17.Nd2 Bb7 19.dxe5
diff --git a/docs/examples/qa-with-citations.md b/docs/cookbook/qa-with-citations.md
similarity index 68%
rename from docs/examples/qa-with-citations.md
rename to docs/cookbook/qa-with-citations.md
index 213942d02..79a2214c3 100644
--- a/docs/examples/qa-with-citations.md
+++ b/docs/cookbook/qa-with-citations.md
@@ -11,19 +11,17 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -37,9 +35,20 @@ model = outlines.from_llamacpp(llm)
     We initialize the model:
 
     ```python
+    import llama_cpp
     from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
     ```
 
 ## Generate Synthetic Data
@@ -65,28 +74,26 @@ class Users(BaseModel):
     users: List[UserDetail]
 ```
 
-We can use a `outlines.Generator` by passing this Pydantic class we just defined, and call the generator:
+We can use a `generate.json` by passing this Pydantic class we just defined, and call the generator:
 
 ```python
-import json
-
-generator = outlines.Generator(model, Users)
+model = models.LlamaCpp(llm)
+generator = generate.json(model, Users)
 response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
-response = json.loads(response)
-print(response['users'])
-# [{'id': 1, 'first_name': 'John', 'last_name': 'Doe', 'age': 25},
-# {'id': 2, 'first_name': 'Jane', 'last_name': 'Doe', 'age': 30},
-# {'id': 3, 'first_name': 'Bob', 'last_name': 'Smith', 'age': 40},
-# {'id': 4, 'first_name': 'Alice', 'last_name': 'Smith', 'age': 35},
-# {'id': 5, 'first_name': 'John', 'last_name': 'Smith', 'age': 20}]
+print(response.users)
+# [UserDetail(id=1, first_name='John', last_name='Doe', age=25),
+# UserDetail(id=2, first_name='Jane', last_name='Doe', age=30),
+# UserDetail(id=3, first_name='Bob', last_name='Smith', age=40),
+# UserDetail(id=4, first_name='Alice', last_name='Smith', age=35),
+# UserDetail(id=5, first_name='John', last_name='Smith', age=20)]
 ```
 
 ```python
-for user in response['users']:
-    print(user['first_name'])
-    print(user['last_name'])
-    print(user['age'])
-    print("#####")
+for user in response.users:
+    print(user.first_name)
+    print(user.last_name)
+    print(user.age)
+    print(#####)
 # John
 # Doe
 # 25
@@ -128,30 +135,23 @@ schema = QuestionAnswer.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-from outlines import Template
-
-hermes_prompt = Template.from_string(
-    """
-    <|im_start|>system
-    You are a world class AI model who answers questions in JSON with correct and exact citations
-    extracted from the `Context`.
-    Here's the json schema you must adhere to:
-    <schema>
-    {{ schema }}
-    </schema>
-    <|im_end|>
-    <|im_start|>user
-    `Context`:
-    {{ context }}
-    `Question`:
-    {{ question }}
-    <|im_end|>
-    <|im_start|>assistant
-    """
-)
+def generate_hermes_prompt(question, context, schema=schema):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON with correct and exact citations "
+        "extracted from the `Context`. "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + "`Context`: "
+        + context
+        + "\n`Question`: "
+        + question + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
 ```
 
-We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
+We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
 
 ```python
 question = "What did the author do during college?"
@@ -161,11 +161,11 @@ I went to an arts high school but in university I studied Computational Mathemat
 As part of coop I worked at many companies including Stitchfix, Facebook.
 I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.
 """
-generator = outlines.Generator(model, QuestionAnswer)
-prompt = hermes_prompt(question=question, context=context, schema=schema)
+generator = generate.json(model, QuestionAnswer)
+prompt = generate_hermes_prompt(question, context)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 print(response)
-# {"question": "What did the author do during college?", "answer": "The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.", "citations": ["I went to an arts high school but in university I studied Computational Mathematics and physics.", "I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years."]}
+# QuestionAnswer(question='What did the author do during college?', answer='The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
 ```
 
 We can do the same for a list of question-context pairs:
@@ -219,13 +219,12 @@ for question, context in [
     (question4, context4),
     (question5, context5),
 ]:
-    prompt = hermes_prompt(question=question, context=context, schema=schema)
-    generator = outlines.Generator(model, QuestionAnswer)
-    response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
-    response = json.loads(response)
-    print(question)
-    print(response['answer'])
-    print(response['citations'])
+    final_prompt = my_final_prompt(question, context)
+    generator = generate.json(model, QuestionAnswer)
+    response = generator(final_prompt, max_tokens=1024, temperature=0, seed=42)
+    display(question)
+    display(response.answer)
+    display(response.citations)
     print("\n\n")
 
 # 'Where was John born?'
diff --git a/docs/examples/react_agent.md b/docs/cookbook/react_agent.md
similarity index 71%
rename from docs/examples/react_agent.md
rename to docs/cookbook/react_agent.md
index bd3d51d6c..15b6ed369 100644
--- a/docs/examples/react_agent.md
+++ b/docs/cookbook/react_agent.md
@@ -15,19 +15,17 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -41,9 +39,20 @@ model = outlines.from_llamacpp(llm)
     We initialize the model:
 
     ```python
+    import llama_cpp
     from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
     ```
 
 ## Build a ReAct agent
@@ -106,16 +115,51 @@ from typing import Union
 
 class Decision(BaseModel):
     Decision: Union[Reason_and_Act, Final_Answer]
+```
+
+We could generate a response using the json schema but we will use the regex and check that everything is working as expected:
+
+```python
+from outlines.fsm.json_schema import convert_json_schema_to_str
+from outlines_core.fsm.json_schema import build_regex_from_schema
 
 json_schema = Decision.model_json_schema()
+schema_str = convert_json_schema_to_str(json_schema=json_schema)
+regex_str = build_regex_from_schema(schema_str)
+print(regex_str)
+# '\\{[ ]?"Decision"[ ]?:[ ]?(\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Thought"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Action"[ ]?:[ ]?("wikipedia"|"calculate")[ ]?,[ ]?"Action_Input"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\}|\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Final_Answer"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\})[ ]?\\}'
 ```
 
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic. We can load a template from a file for that:
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic:
 
 ```python
-from outlines import Template
-
-hermes_prompt = Template.from_file("prompt_templates/react_agent.txt")
+import datetime
+
+def generate_hermes_prompt(question, schema=""):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON with correct Pydantic schema. "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema>\n"
+        "Today is " + datetime.datetime.today().strftime('%Y-%m-%d') + ".\n" +
+        "You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation. "
+        "At the end of the loop you output a Final Answer. "
+        "Use Scratchpad to store the information from the Observation useful to answer the question "
+        "Use Thought to describe your thoughts about the question you have been asked "
+        "and reflect carefully about the Observation if it exists. "
+        "Use Action to run one of the actions available to you. "
+        "Use Action Input to input the arguments of the selected action - then return PAUSE. "
+        "Observation will be the result of running those actions. "
+        "Your available actions are:\n"
+        "calculate:\n"
+        "e.g. calulate: 4**2 / 3\n"
+        "Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary\n"
+        "wikipedia:\n"
+        "e.g. wikipedia: Django\n"
+        "Returns a summary from searching Wikipedia\n"
+        "DO NOT TRY TO GUESS THE ANSWER. Begin! <|im_end|>"
+        "\n<|im_start|>user\n" + question + "<|im_end|>"
+        "\n<|im_start|>assistant\n"
+    )
 ```
 
 We define a ChatBot class
@@ -131,7 +175,7 @@ class ChatBot:
         return result
 
     def execute(self):
-        generator = outlines.Generator(model, Decision)
+        generator = generate.regex(model, regex_str)
         result = generator(self.prompt, max_tokens=1024, temperature=0, seed=42)
         return result
 ```
@@ -150,11 +194,7 @@ def query(question, max_turns=5):
     previous_actions = []
     while i < max_turns:
         i += 1
-        prompt = generate_hermes_prompt(
-            question=question,
-            schema=Decision.model_json_schema(),
-            today=datetime.datetime.today().strftime('%Y-%m-%d')
-        )
+        prompt = generate_hermes_prompt(question=question, schema=Decision.model_json_schema())
         bot = ChatBot(prompt=prompt)
         result = bot(next_prompt)
         json_result = json.loads(result)['Decision']
diff --git a/docs/examples/read-pdfs.md b/docs/cookbook/read-pdfs.md
similarity index 79%
rename from docs/examples/read-pdfs.md
rename to docs/cookbook/read-pdfs.md
index 1854aba42..dbe4ccb02 100644
--- a/docs/examples/read-pdfs.md
+++ b/docs/cookbook/read-pdfs.md
@@ -46,19 +46,17 @@ We've tested this example with [Pixtral 12b](https://huggingface.co/mistral-comm
 To use Pixtral:
 
 ```python
-from transformers import LlavaForConditionalGeneration, LlavaProcessor
+from transformers import LlavaForConditionalGeneration
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
-processor_class = LlavaProcessor
 ```
 
 To use Qwen-2-VL:
 
 ```python
-from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
+from transformers import Qwen2VLForConditionalGeneration
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
-processor_class = AutoProcessor
 ```
 
 You can load your model into memory with:
@@ -66,12 +64,17 @@ You can load your model into memory with:
 ```python
 # This loads the model into memory. On your first run,
 # it will have to download the model, which might take a while.
-model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
-processor_kwargs={"device_map": "cpu"}
-tf_model = model_class.from_pretrained(model_name, **model_kwargs)
-tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-model = outlines.from_transformers(tf_model, tf_processor)
+model = outlines.models.transformers_vision(
+    model_name,
+    model_class=model_class,
+    model_kwargs={
+        "device_map": "auto",
+        "torch_dtype": torch.bfloat16,
+    },
+    processor_kwargs={
+        "device": "auto",
+    },
+)
 ```
 
 ## Convert the PDF to images
@@ -156,7 +159,7 @@ The structured output you can extract is exactly the same as everywhere else in
 
 Suppose you wished to go through each page of the PDF, and extract the page description, key takeaways, and page number.
 
-You can do this by defining a JSON schema, and then using `outlines.Generator` to extract the data.
+You can do this by defining a JSON schema, and then using `outlines.generate.json` to extract the data.
 
 First, define the structure you want to extract:
 
@@ -167,7 +170,7 @@ class PageSummary(BaseModel):
     page_number: int
 ```
 
-Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers processor to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
+Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers `AutoProcessor` to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
 
 Images are denoted with `type: "image"`, and text is denoted with `type: "text"`.
 
@@ -188,15 +191,16 @@ messages = [
                 """
             },
 
-            # This a placeholder, the actual image is passed in when
-            # we call the generator function down below.
+            # Don't need to pass in an image, since we do this
+            # when we call the generator function down below.
             {"type": "image", "image": ""},
         ],
     }
 ]
 
 # Convert the messages to the final prompt
-prompt = tf_processor.apply_chat_template(
+processor = AutoProcessor.from_pretrained(model_name)
+instruction = processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -205,10 +209,10 @@ Now we iterate through each image, and extract the structured information:
 
 ```python
 # Page summarizer function
-page_summary_generator = outlines.Generator(model, PageSummary)
+page_summary_generator = outlines.generate.json(model, PageSummary)
 
 for image in images:
-    result = page_summary_generator({"text": prompt, "images": image})
+    result = page_summary_generator(instruction, [image])
     print(result)
 ```
 
@@ -221,21 +225,19 @@ arXiv identifiers are optionally followed by a version number, i.e. `arXiv:YYMM.
 We can use a regular expression to define this patter:
 
 ```python
-from outlines.types import Regex
-
-paper_regex = Regex(r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?')
+paper_regex = r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?'
 ```
 
 We can build an extractor function from the regex:
 
 ```python
-id_extractor = outlines.Generator(model, paper_regex)
+id_extractor = outlines.generate.regex(model, paper_regex)
 ```
 
 Now, we can extract the arxiv paper identifier from the first image:
 
 ```python
-arxiv_instruction = tf_processor.apply_chat_template(
+arxiv_instruction = processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -254,7 +256,7 @@ arxiv_instruction = tf_processor.apply_chat_template(
 )
 
 # Extract the arxiv paper identifier
-paper_id = id_extractor({"text": arxiv_instruction, "images": images[0]})
+paper_id = id_extractor(arxiv_instruction, [images[0]])
 ```
 
 As of the time of this writing, the arxiv paper identifier is
@@ -267,9 +269,7 @@ Your version number may be different, but the part before `vX` should match.
 
 ### Categorize the paper into one of several categories
 
-`outlines.Generator` also allows the model to select one of several options by providing a Literal type hint with the categories.
-
-Suppose we wanted to categorize the paper into being about "language models", "cell biology", or "other". We would then define the output type as `Literal["llms", "cell biology", "other"]`.
+`outlines.generate.choice` allows the model to select one of several options. Suppose we wanted to categorize the paper into being about "language models", "economics", "structured generation", or "other".
 
 Let's define a few categories we might be interested in:
 
@@ -284,7 +284,7 @@ categories = [
 Now we can construct the prompt:
 
 ```python
-categorization_instruction = tf_processor.apply_chat_template(
+categorization_instruction = processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -310,13 +310,14 @@ categorization_instruction = tf_processor.apply_chat_template(
 Now we can show the model the first page and extract the category:
 
 ```python
-from typing import Literal
-
 # Build the choice extractor
-categorizer = outlines.Generator(model, Literal["llms", "cell biology", "other"])
+categorizer = outlines.generate.choice(
+    model,
+    categories
+)
 
 # Categorize the paper
-category = categorizer({"text": categorization_instruction, "images": images[0]})
+category = categorizer(categorization_instruction, [images[0]])
 print(category)
 ```
 
@@ -331,12 +332,12 @@ llms
 You can provide multiple images to the model by
 
 1. Adding additional image messages
-2. Providing a list of images to the generator
+2. Providing a list of images to the `generate` function
 
 For example, to have two images, you can do:
 
 ```python
-two_image_prompt = tf_processor.apply_chat_template(
+two_image_prompt = processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -354,9 +355,17 @@ two_image_prompt = tf_processor.apply_chat_template(
 )
 
 # Pass two images to the model
-generator = outlines.Generator(model, Literal["hot dog", "not hot dog"])
+generator = outlines.generate.choice(
+    model,
+    ["hot dog", "not hot dog"]
+)
 
-result = generator({"text": two_image_prompt, "images": [images[0], images[1]]})
+result = generator(
+    two_image_prompt,
+
+    # Pass two images to the model
+    [images[0], images[1]]
+)
 print(result)
 ```
 
diff --git a/docs/examples/receipt-digitization.md b/docs/cookbook/receipt-digitization.md
similarity index 71%
rename from docs/examples/receipt-digitization.md
rename to docs/cookbook/receipt-digitization.md
index 8ecb07969..0fa256db7 100644
--- a/docs/examples/receipt-digitization.md
+++ b/docs/cookbook/receipt-digitization.md
@@ -38,20 +38,18 @@ If you want to use Qwen-2-VL, you can do the following:
 
 ```python
 # To use Qwen-2-VL:
-from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
+from transformers import Qwen2VLForConditionalGeneration
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
-processor_class = AutoProcessor
 ```
 
 If you want to use Pixtral, you can do the following:
 
 ```python
 # To use Pixtral:
-from transformers import LlavaForConditionalGeneration, LlavaProcessor
+from transformers import LlavaForConditionalGeneration
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
-processor_class = LlavaProcessor
 ```
 
 ## Load the model
@@ -59,12 +57,17 @@ processor_class = LlavaProcessor
 Load the model into memory:
 
 ```python
-model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
-processor_kwargs={"device_map": "cuda"}
-tf_model = model_class.from_pretrained(model_name, **model_kwargs)
-tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-model = outlines.from_transformers(tf_model, tf_processor)
+model = outlines.models.transformers_vision(
+    model_name,
+    model_class=model_class,
+    model_kwargs={
+        "device_map": "auto",
+        "torch_dtype": torch.bfloat16,
+    },
+    processor_kwargs={
+        "device": "cuda", # set to "cpu" if you don't have a GPU
+    },
+)
 ```
 
 ## Image processing
@@ -163,7 +166,7 @@ class ReceiptSummary(BaseModel):
 
 ## Prepare the prompt
 
-We'll use the `tf_processor` to convert the image and the text prompt into a format that the model can understand. Practically,
+We'll use the `AutoProcessor` to convert the image and the text prompt into a format that the model can understand. Practically,
 this is the code that adds user, system, assistant, and image tokens to the prompt.
 
 ```python
@@ -191,7 +194,8 @@ messages = [
 ]
 
 # Convert the messages to the final prompt
-prompt = tf_processor.apply_chat_template(
+processor = AutoProcessor.from_pretrained(model_name)
+prompt = processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -218,13 +222,17 @@ Return the information in the following JSON schema:
 
 ```python
 # Prepare a function to process receipts
-receipt_summary_generator = outlines.Generator(model, ReceiptSummary)
+receipt_summary_generator = outlines.generate.json(
+    model,
+    ReceiptSummary,
 
-# Generate the receipt summary
-result = receipt_summary_generator(
-    {"text": prompt, "images": image},
-    max_new_tokens=1024
+    # Greedy sampling is a good idea for numeric
+    # data extraction -- no randomness.
+    sampler=outlines.samplers.greedy()
 )
+
+# Generate the receipt summary
+result = receipt_summary_generator(prompt, [image])
 print(result)
 ```
 
@@ -233,31 +241,32 @@ print(result)
 The output should look like this:
 
 ```
-{
-  "store_name": "Trader Joe's",
-  "store_address": "401 Bay Street, San Francisco, CA 94133",
-  "store_number": 0,
-  "items": [
-    {"name": "BANANA EACH", "quantity": 7, "price_per_unit": 0.23, "total_price": 1.61},
-    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "BAREBELLS CARAMEL CASHEW", "quantity": 2, "price_per_unit": 2.29, "total_price": 4.58},
-    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "SPINDRIFT ORANGE MANGO 8", "quantity": 1, "price_per_unit": 7.49, "total_price": 7.49},
-    {"name": "Bottle Deposit", "quantity": 8, "price_per_unit": 0.05, "total_price": 0.4},
-    {"name": "MILK ORGANIC GALLON WHOL", "quantity": 1,"price_per_unit": 6.79,"total_price": 6.79},
-    {"name": "CLASSIC GREEK SALAD", "quantity": 1, "price_per_unit": 3.49, "total_price": 3.49},
-    {"name": "COBB SALAD", "quantity": 1, "price_per_unit": 5.99, "total_price": 5.99},
-    {"name": "PEPPER BELL RED XL EACH", "quantity": 1, "price_per_unit": 1.29, "total_price": 1.29},
-    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
-    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
-  ],
-  "tax": 0.68,
-  "total": 41.98,
-  "date": "2023-11-04",
-  "payment_method": "debit"
-}
+ReceiptSummary(
+    store_name="Trader Joe's",
+    store_address='401 Bay Street, San Francisco, CA 94133',
+    store_number=0,
+    items=[
+        Item(name='BANANA EACH', quantity=7, price_per_unit=0.23, total_price=1.61),
+        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CARAMEL CASHEW', quantity=2, price_per_unit=2.29, total_price=4.58),
+        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='SPINDRIFT ORANGE MANGO 8', quantity=1, price_per_unit=7.49, total_price=7.49),
+        Item(name='Bottle Deposit', quantity=8, price_per_unit=0.05, total_price=0.4),
+        Item(name='MILK ORGANIC GALLON WHOL', quantity=1, price_per_unit=6.79, total_price=6.79),
+        Item(name='CLASSIC GREEK SALAD', quantity=1, price_per_unit=3.49, total_price=3.49),
+        Item(name='COBB SALAD', quantity=1, price_per_unit=5.99, total_price=5.99),
+        Item(name='PEPPER BELL RED XL EACH', quantity=1, price_per_unit=1.29, total_price=1.29),
+        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25),
+        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25)
+    ],
+    tax=0.68,
+    total=41.98,
+    date='2023-11-04',
+    payment_method='debit',
+
+)
 ```
 
 Voila! You've successfully extracted information from a receipt using an LLM.
@@ -275,13 +284,13 @@ class ReceiptSummary(BaseModel):
 which gives you a result like
 
 ```
-{
+ReceiptSummary(
     ...
-    "roast": "You must be a fan of Trader Joe's because you bought enough
+    roast="You must be a fan of Trader Joe's because you bought enough
     items to fill a small grocery bag and still had to pay for a bag fee.
     Maybe you should start using reusable bags to save some money and the
     environment."
-}
+)
 ```
 
 Qwen is not particularly funny, but worth a shot.
diff --git a/docs/examples/simtom.md b/docs/cookbook/simtom.md
similarity index 68%
rename from docs/examples/simtom.md
rename to docs/cookbook/simtom.md
index 53820258e..a80ebc24f 100644
--- a/docs/examples/simtom.md
+++ b/docs/cookbook/simtom.md
@@ -17,7 +17,7 @@ SimToM calls an LLM with two consecutive prompts:
 
 To implement SimToM with Outlines, we will need to:
 
-1. Write the prompts with [prompt templates](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
+1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
 2. Define the JSON object each prompt will return using Pydantic.
 3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/latest/reference/models/transformers/).
 
@@ -30,8 +30,29 @@ The authors have shared their code, prompts and data in [this GitHub repository]
 ```python
 from outlines import Template
 
-perspective_taking = Template.from_file("prompt_templates/simtom_prospective_taking.txt")
-simulation = Template.from_file("prompt_templates/simtom_simulation.txt")
+
+perspective_taking = Template.from_string(
+    """<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
+    Your job is to output only the events that the specified character, {{character}}, knows about.
+
+    Here are a few rules:
+    1. A character knows about all events that they do.
+    2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
+    3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
+
+    Story: {{story}}
+    What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]""" # noqa
+)
+
+simulation = Template.from_string(
+    """<s>[INST] {% for event in events %}
+    {{event}}
+    {% endfor %}
+    You are {{name}}.
+    Based on the above information, answer the following question:
+    {{question}}
+    You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]""" # noqa
+)
 ```
 
 ### JSON Structured Generation
@@ -44,11 +65,13 @@ We will need two Pydantic models for SimToM, one for each prompt:
 from pydantic import BaseModel, Field
 from typing import List
 
+
 class PerspectiveTaking(BaseModel):
     """This is for the first prompt."""
     character: str = Field(description="The character we extract the events for.")
     events: List[str] = Field(description="All events that the character knows about.")
 
+
 class Simulation(BaseModel):
     """This is for the second prompt."""
     answer: str
@@ -74,37 +97,30 @@ character = "Aria"
 We load `Mistral-7B-Instruct-v0.3`, create the prompt using the template we defined earlier, and generate a structured response. As a reminder, the goal of the first call is to get all the events a character, `Aria`, knows about.
 
 ```python
-import transformers
-import outlines
 # Load an LLM from Hugging Face
 MODEL_NAME = "mistral-community/Mistral-7B-Instruct-v0.3"
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
-    transformers.AutoTokenizer.from_pretrained(MODEL_NAME),
-)
+model = outlines.models.transformers(MODEL_NAME, device="cuda")
 
 perspective_prompt = perspective_taking(story=story, character=character)
 
 # Call Mistral 7B with the first prompt
-generator = outlines.Generator(model, PerspectiveTaking)
-perspective = generator(perspective_prompt, max_new_tokens=1024)
+generator = outlines.generate.json(model, PerspectiveTaking)
+perspective = generator(perspective_prompt)
 
-print(perspective)
+print(perspective.model_dump())
 # {'character': 'Aria', 'events': ['1 Aria entered the front_yard.', '3 The grapefruit is in the green_bucket.', '4 Aria moved the grapefruit to the blue_container.']}
 ```
 
 Not bad! We will now generate the second prompt with those events.
 
 ```python
-import json
-
-sim_prompt = simulation(events=json.loads(perspective)["events"], name=character, question=question)
+sim_prompt = simulation(events=perspective.events, name=character, question=question)
 
 # Call Mistral 7B with the second prompt
-generator = outlines.Generator(model, Simulation)
-result = generator(sim_prompt, max_new_tokens=1024)
+generator = outlines.generate.json(model, Simulation)
+result = generator(sim_prompt)
 
-print(result)
+print(result.model_dump())
 # {'answer': 'green_bucket'}
 ```
 
diff --git a/docs/examples/structured_generation_workflow.md b/docs/cookbook/structured_generation_workflow.md
similarity index 79%
rename from docs/examples/structured_generation_workflow.md
rename to docs/cookbook/structured_generation_workflow.md
index 524283f84..b9d7c28a4 100644
--- a/docs/examples/structured_generation_workflow.md
+++ b/docs/cookbook/structured_generation_workflow.md
@@ -2,7 +2,8 @@
 
 This is a condensed version of [Coding for Structured Generation with LLMs](https://blog.dottxt.co/coding-for-structured-generation.html).
 
-For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
+For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this
+with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
 
 ## Unstructured approach
 
@@ -10,30 +11,36 @@ Before diving into how to use structure generation for this task let's start wit
 
 ```python
 import outlines
-import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name),
-    transformers.AutoTokenizer.from_pretrained(model_name)
-)
+model = outlines.models.transformers(model_name)
 ```
 
 Next we need a prompt for this model. Since we're focusing on structured generation, we won't be engaging in any form of "prompt hacking" and will be leaving this prompt untouched for the rest of this example.
 
 ```python
-prompt_phone = """
-    Please generate a realistic phone number for Washington State in the following format
-    (555) 555-5555
-"""
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+messages_phone = [
+            {"role": "user", "content": """
+            Please generate a realistic phone number for Washington State in the following format
+
+            (555) 555-5555
+
+            """}
+]
+
+# This allows us to properly format our prompt for
+# Phi-3 Mini's 'Instruct' interface.
+prompt_phone = tokenizer.apply_chat_template(messages_phone, tokenize=False)
 ```
 
 With our prompt ready we can now generate 10 example phone numbers
 
 ```python
-phone_generator_unstruct = outlines.Generator(model)
-for _ in range(3):
-    print(phone_generator_unstruct(prompt_phone, max_new_tokens=12))
+phone_generator_unstruct = outlines.generate.text(model)
+for _ in range(10):
+    print(phone_generator_unstruct(prompt_phone,max_tokens=12))
 ```
 
 > I'd be happy to help you generate a realistic phone\
@@ -74,9 +81,7 @@ For a simple example like this, we'll just be using a single phone number, for m
 The next step in the process is for use to define a simple regex that we feel correctly models our real data.
 
 ```python
-from outlines.types import Regex
-
-phone_regex_1 = Regex(r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}')
+phone_regex_1 = r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}'
 ```
 
 Next we need to validate this regex against our real data.
@@ -89,8 +94,8 @@ We'll start with a simple method of validation: just checking that our regex mat
 
 ```
 import re
+re.match(phone_regex_1, phone_number)
 
-re.match(phone_regex_1.pattern, phone_number)
 # <re.Match object; span=(0, 14), match='(206) 386-4636'>
 
 ```
@@ -102,9 +107,8 @@ Now that we have a match, we can move on to generating structured output!
 We're ready to see if structured generation can make an improvement over our initial unstructured approach:
 
 ```python
-phone_generator_v1 = outlines.Generator(model, phone_regex_1)
-
-for _ in range(3):
+phone_generator_v1 = outlines.generate.regex(model, phone_regex_1)
+for _ in range(10):
     print(phone_generator_v1(prompt_phone))
 ```
 > (206) 555-1234\
@@ -131,21 +135,21 @@ We've walked through the loop once, so we can go quickly now through each iterat
 We start by improving our structure:
 
 ```python
-phone_regex_2 = Regex(r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}')
+phone_regex_2 = r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}'
 ```
 
 Before rushing to another round of generation, let's validate this new regex. We'll add just a bit more sophistication over our last check:
 
 ```python
-re.match(phone_regex_2.pattern, phone_number)[0] == phone_number
+re.match(phone_regex_2, phone_number)[0] == phone_number
 # True
 ```
 Now that we've validated, let's generate with this new regex!
 
 ```python
-phone_generator_v2 = outlines.Generator(model, phone_regex_2)
-
-for _ in range(3):
+phone_generator_v2 = outlines.generate.regex(model,
+                                             phone_regex_2)
+for _ in range(10):
     print(phone_generator_v2(prompt_phone))
 ```
 
@@ -188,10 +192,10 @@ This prints out:
 Ah! We were missing the last digit, let's fix that and regenerate:
 
 ```python
-phone_regex_3_fixed = Regex(r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]')
-phone_generator_v3 = outlines.Generator(model, phone_regex_3_fixed)
-
-for _ in range(3):
+phone_regex_3_fixed = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]'
+phone_generator_v3 = outlines.generate.regex(model,
+                                             phone_regex_3_fixed)
+for _ in range(10):
     print(phone_generator_v3(prompt_phone))
 ```
 
diff --git a/docs/core_concepts.md b/docs/core_concepts.md
deleted file mode 100644
index 7f53983ec..000000000
--- a/docs/core_concepts.md
+++ /dev/null
@@ -1,13 +0,0 @@
----
-title: Core concepts
----
-
-# Core concepts
-
-Coming soon. This will document various concepts at a high level, so users can understand Outlines before diving into specific implementations.
-
-1. Constrained decoding, tokens, and the basics of logit biasing
-2. Different ways to define output structure (regex, JSON schema, Pydantic models, context-free grammars)
-3. How finite state machines are used to guarantee output structure
-4. `Generator`, `Application`, `Template`,
-5. Prompt engineering vs. structured generation
diff --git a/docs/examples/classification.md b/docs/examples/classification.md
deleted file mode 100644
index 18a933b47..000000000
--- a/docs/examples/classification.md
+++ /dev/null
@@ -1,85 +0,0 @@
-# Classification
-
-Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
-
-
-This tutorial shows how one can implement multi-label classification using Outlines.
-
-As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
-
-```python
-import outlines
-import transformers
-
-MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
-
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
-    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
-)
-```
-
-We will use a prompt template stored in a text file:
-
-```python
-from outlines import Template
-
-customer_support = Template.from_file("prompt_templates/classification.txt")
-```
-
-## Choosing between multiple choices
-
-Outlines provides a convenient way to do multi-label classification, passing a Literal type hint to the `outlines.Generator` object:
-
-```python
-from typing import Literal
-import outlines
-
-generator = outlines.Generator(model, Literal["URGENT", "STANDARD"])
-
-```
-Outlines supports batched requests, so we will pass two requests to the model:
-
-```python
-requests = [
-    "My hair is one fire! Please help me!!!",
-    "Just wanted to say hi"
-]
-
-prompts = [customer_support(request=request) for request in requests]
-```
-
-We can now ask the model to classify the requests:
-
-```python
-labels = generator(prompts)
-print(labels)
-# ['URGENT', 'STANDARD']
-```
-
-## Using JSON-structured generation
-
-Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
-
-```python
-from enum import Enum
-from pydantic import BaseModel
-
-
-class Label(str, Enum):
-    urgent = "URGENT"
-    standard = "STANDARD"
-
-
-class Classification(BaseModel):
-    label: Label
-```
-
-We can then create a generator with the Pydantic model we just defined and call it:
-
-```python
-generator = outlines.Generator(model, Classification)
-labels = generator(prompts)
-print(labels)
-# ['{"label":"URGENT"}', '{ "label": "STANDARD" }']
-```
diff --git a/docs/examples/extract_event_details.py b/docs/examples/extract_event_details.py
deleted file mode 100644
index cd90bc439..000000000
--- a/docs/examples/extract_event_details.py
+++ /dev/null
@@ -1,54 +0,0 @@
-from datetime import datetime
-
-from mlx_lm import load
-from pydantic import BaseModel, Field
-
-import outlines
-from outlines import Generator, Template
-
-
-# Load the model
-model = outlines.from_mlxlm(*load("mlx-community/Hermes-3-Llama-3.1-8B-8bit"))
-
-
-# Define the event schema using Pydantic
-class Event(BaseModel):
-    title: str = Field(description="title of the event")
-    location: str
-    start: datetime = Field(
-        default=None, description="date of the event if available in iso format"
-    )
-
-# Load the prompt template from a string
-prompt_template = Template.from_string(
-    """
-    Today's date and time are {{ now }}
-    Given a user message, extract information of the event like date and time in iso format, location and title.
-    If the given date is relative, think step by step to find the right date.
-    Here is the message:
-    {{ message }}
-    """
-)
-
-# Get the current date and time
-now = datetime.now().strftime("%A %d %B %Y and it's %H:%M")
-
-# Sample message
-message = """Hello Kitty, my grandmother will be here, I think it's better to postpone our
-appointment to review math lessons to next Friday at 2pm at the same place, 3 avenue des tanneurs, I think that one hour will be enough
-see you 😘 """
-
-# Create the generator
-generator = Generator(model, Event)
-
-# Create the prompt
-prompt = prompt_template(now=now, message=message)
-
-# Extract the event information
-event = generator(prompt)
-
-# Print the current date and time
-print(f"Today: {now}")
-
-# Print the extracted event information
-print(event)
diff --git a/docs/examples/prompt_templates/chain_of_density.txt b/docs/examples/prompt_templates/chain_of_density.txt
deleted file mode 100644
index c4f9f55fa..000000000
--- a/docs/examples/prompt_templates/chain_of_density.txt
+++ /dev/null
@@ -1,27 +0,0 @@
-Article: {{ article }}
-
-You will generate increasingly concise, entity-dense summaries of the above Article.
-
-Repeat the following 2 steps 5 times.
-
-Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
-Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
-
-A Missing Entity is:
-- Relevant: to the main story.
-- Specific: descriptive yet concise (5 words or fewer).
-- Novel: not in the previous summary.
-- Faithful: present in the Article.
-- Anywhere: located anywhere in the Article.
-
-Guidelines:
-- The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
-- Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
-- Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
-- The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
-- Missing entities can appear anywhere in the new summary.
-- Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
-
-Remember, use the exact same number of words for each summary.
-
-Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
diff --git a/docs/examples/prompt_templates/classification.txt b/docs/examples/prompt_templates/classification.txt
deleted file mode 100644
index e260a6771..000000000
--- a/docs/examples/prompt_templates/classification.txt
+++ /dev/null
@@ -1,18 +0,0 @@
-You are an experienced customer success manager.
-
-Given a request from a client, you need to determine when the
-request is urgent using the label "URGENT" or when it can wait
-a little with the label "STANDARD".
-
-# Examples
-
-Request: "How are you?"
-Label: STANDARD
-
-Request: "I need this fixed immediately!"
-Label: URGENT
-
-# TASK
-
-Request: {{ request }}
-Label:
diff --git a/docs/examples/prompt_templates/react_agent.txt b/docs/examples/prompt_templates/react_agent.txt
deleted file mode 100644
index 09583f42f..000000000
--- a/docs/examples/prompt_templates/react_agent.txt
+++ /dev/null
@@ -1,27 +0,0 @@
-<|im_start|>system
-You are a world class AI model who answers questions in JSON with correct Pydantic schema.
-Here's the json schema you must adhere to:
-<schema>
-{{ schema }}
-</schema>
-Today is {{ today }}
-You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation.
-At the end of the loop you output a Final Answer.
-Use Scratchpad to store the information from the Observation useful to answer the question
-Use Thought to describe your thoughts about the question you have been asked and reflect carefully about the Observation if it exists.
-Use Action to run one of the actions available to you.
-Use Action Input to input the arguments of the selected action - then return PAUSE.
-Observation will be the result of running those actions.
-Your available actions are:
-calculate:
-e.g. calulate: 4**2 / 3
-Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary
-wikipedia:
-e.g. wikipedia: Django
-Returns a summary from searching Wikipedia
-DO NOT TRY TO GUESS THE ANSWER. Begin!
-<|im_end|>
-<|im_start|>user
-{{ question }}
-<|im_end|>
-<|im_start|>assistant
diff --git a/docs/examples/prompt_templates/simtom_prospective_taking.txt b/docs/examples/prompt_templates/simtom_prospective_taking.txt
deleted file mode 100644
index 8e937ce60..000000000
--- a/docs/examples/prompt_templates/simtom_prospective_taking.txt
+++ /dev/null
@@ -1,10 +0,0 @@
-<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
-Your job is to output only the events that the specified character, {{character}}, knows about.
-
-Here are a few rules:
-1. A character knows about all events that they do.
-2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
-3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
-
-Story: {{story}}
-What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]
diff --git a/docs/examples/prompt_templates/simtom_simulation.txt b/docs/examples/prompt_templates/simtom_simulation.txt
deleted file mode 100644
index ede934783..000000000
--- a/docs/examples/prompt_templates/simtom_simulation.txt
+++ /dev/null
@@ -1,7 +0,0 @@
-<s>[INST] {% for event in events %}
-{{event}}
-{% endfor %}
-You are {{name}}.
-Based on the above information, answer the following question:
-{{question}}
-You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]
diff --git a/docs/features/advanced/logits_processors.md b/docs/features/advanced/logits_processors.md
deleted file mode 100644
index 4e4d9e457..000000000
--- a/docs/features/advanced/logits_processors.md
+++ /dev/null
@@ -1,97 +0,0 @@
----
-title: Logits Processors
----
-
-# Logits Processors
-
-Logits processors are objects that control text generation by modifying the probability distribution of possible next tokens. They do this by adjusting the logits (raw model outputs) at each generation step, effectively biasing the model's token selection.
-Processors can be used to:
-1. Generate structured output (e.g., JSON that follows a specific schema)
-2. Prevent the model from generating specific words or tokens
-3. Implement custom token sampling strategies
-
-## Overview
-
-Outlines uses logits processors with steerable models — models that run locally and allow fine-grained control over the generation process. When using such models in Outlines, the output type provided is turned into a logits processor that is then passed to the inference engine.
-
-There are three models that support logits processors:
-
-- LlamaCpp
-- MLXLM
-- Transformers
-
-Instead of providing an output type that will be turned into a logits processor, it is possible to directly provide a logits processor. To do so, you must create a `Generator` instance using the `processor` keyword argument. You cannot directly call the model with a logits processor.
-
-For instance:
-
-```python
-import transformers
-from outlines import Generator, from_transformers
-from outlines.processors import RegexLogitsProcessor
-
-# Create a model
-model = from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B"),
-    transformers.AutoTokenizer.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
-)
-
-# Create a regex logits processor that only returns hex unicode notations
-logits_processor = RegexLogitsProcessor(r"U\+[0-9A-Fa-f]{4,6}", model.tokenizer, model.tensor_library_name)
-
-# Create a generator with the logits processor and use it to generate text
-generator = Generator(model, processor=logits_processor)
-response = generator("What's the unicode for the hugging face emoji")
-
-print(response) # U+1F917
-```
-
-## Creating Custom Logits Processors
-
-You can create your own logits processor by subclassing the `OutlinesLogitsProcessor` class. This allows you to implement specific logic to modify logits as needed.
-Your logits processor needs to implement the `process_logits` method to modify the logits.
-`process_logits` accepts:
-- `input_ids`: the ids of the tokens of the existing sequences in a 2D tensor.
-- `logits`: the logits for the current generation step in a 2D tensor.
-
-In the example below, we create a custom logits processor to force the model to provide a response using only binary representation (so only the tokens for 0 and 1 are allowed):
-
-```python
-from outlines.processors.base_logits_processor import OutlinesLogitsProcessor, TensorType
-from outlines import Generator, from_transformers
-import transformers
-
-ALLOWED_TOKENS = [15, 16]  # token IDs corresponding to '0' and '1' in the model's vocabulary
-
-# Subclass OutlinesLogitsProcessor
-class BinaryLogitsProcessor(OutlinesLogitsProcessor):
-
-    def process_logits(self, input_ids: TensorType, logits: TensorType) -> TensorType:
-        # Create a mask for all tokens
-        mask = self.tensor_adapter.boolean_ones_like(logits)
-        # Set mask to False for the allowed tokens
-        for token_id in ALLOWED_TOKENS:
-            mask[:, token_id] = False
-        # Set non-allowed tokens to -inf so they are not selected
-        logits[mask] = float("-inf")
-        return logits
-
-# Create a regular model
-tf_tokenizer = transformers.AutoTokenizer.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
-tf_model = transformers.AutoModelForCausalLM.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
-model = from_transformers(tf_model, tf_tokenizer)
-
-# Instantiate your custom logits processor
-logits_processor = BinaryLogitsProcessor(model.tensor_library_name)
-
-prompt = "Write the number 47 in binary. For example, 1010 is the binary representation of 10. Answer just with the binary number composed of 0s and 1s."
-formatted_prompt = tf_tokenizer.apply_chat_template(
-    [{"role": "user", "content": prompt}],
-    tokenize=False
-)
-
-# Create a generator with the custom logits processor instance and use it to generate text
-generator = Generator(model, processor=logits_processor)
-response = generator(formatted_prompt)
-
-print(response) # "101111"
-```
diff --git a/docs/features/core/generator.md b/docs/features/core/generator.md
deleted file mode 100644
index 508ba399c..000000000
--- a/docs/features/core/generator.md
+++ /dev/null
@@ -1,122 +0,0 @@
----
-title: Generator API
----
-
-# Generator
-
-
-The `Generator` class is the core component of Outlines v1. `Generator` accepts a [model](../reference/available_models) and an optional [output type](/api/types). If no output type is provided, the `Generator` will return unstructured text.
-
-!!! note
-
-    `Generator` is new as of Outlines v1, and replaces previous generator constructors:
-
-    - `generate.cfg`
-    - `generate.choice`
-    - `generate.format`
-    - `generate.fsm`
-    - `generate.json`
-    - `generate.regex`
-    - `generate.text`
-
-## Basic Usage
-
-```python
-from outlines import Generator, from_transformers
-import transformers
-
-# Initialize a model
-model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
-model = from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name),
-    transformers.AutoTokenizer.from_pretrained(model_name),
-)
-
-# Create a generator for plain text
-generator = Generator(model)
-result = generator("Write a short poem about AI.")
-
-# Print the result
-print(result)
-```
-
-## Structured Generation
-
-```python
-from pydantic import BaseModel
-from outlines import Generator, from_transformers
-import transformers
-
-# Define a Pydantic model for structured output
-class BookRecommendation(BaseModel):
-    title: str
-    author: str
-    year: int
-
-# Initialize a model
-model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
-model = from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name),
-    transformers.AutoTokenizer.from_pretrained(model_name),
-)
-
-# Create a generator for JSON output
-generator = Generator(model, BookRecommendation)
-
-# Generate a book recommendation
-result = generator("Recommend a science fiction book.")
-
-# Parse the JSON result into a Pydantic model
-book = BookRecommendation.model_validate_json(result)
-print(f"{book.title} by {book.author} ({book.year})")
-```
-
-## Parameters
-
-- `model`: The language model to use for generation
-- `output_type`: Optional. The type of output to generate
-
-## Generation Parameters
-
-When calling the generator, you can pass additional parameters to control the generation process. These parameters are passed through to the underlying model, so they depend on the specific model being used.
-
-Common parameters for most models include:
-- `max_new_tokens`: Maximum number of tokens to generate
-- `temperature`: Controls randomness (higher values = more random)
-- `top_p`: Controls diversity via nucleus sampling
-- `stop_strings`: String or list of strings at which to stop generation
-
-Example:
-```python
-result = generator(
-    "Write a short story.",
-    max_new_tokens=200,
-    temperature=0.7,
-    top_p=0.9,
-    stop_strings=["THE END", "###"]
-)
-```
-
-## Return Value
-
-The generator always returns a raw string containing the generated text. When generating structured outputs, you need to parse this string into the desired format.
-
-Unlike in Outlines v0, where the return type could be a parsed object, in v1 you are responsible for parsing the output when needed:
-
-```python
-# Outlines v1 approach
-from pydantic import BaseModel
-from outlines import Generator
-
-class Person(BaseModel):
-    name: str
-    age: int
-
-generator = Generator(model, Person)
-result = generator("Generate a person:")
-
-# Parse the result yourself
-person = Person.model_validate_json(result)
-```
-
-::: outlines.generator.Generator
diff --git a/docs/features/core/output_types.md b/docs/features/core/output_types.md
deleted file mode 100644
index 3a49c2623..000000000
--- a/docs/features/core/output_types.md
+++ /dev/null
@@ -1,205 +0,0 @@
----
-title: Output Types
----
-
-# Output Types
-
-Outlines provides a simple and intuitive way of defining the output structure of text generation. Possible output formats include basic Python types, multiple-choices, JSON schemas, regular expressions and context-free grammars.
-
-## Overview
-
-Outlines models accept a __prompt__ and an __output type__ when they are invoked, as well as additional inference keyword arguments that are forwarded on to the underlying model.
-
-Output types can be from the general Python ecosystem, including:
-- Most native Python types, such as `int` or `str`
-- Types from the `typing` module, such as `Literal`, `List`, `Dict`, `Enum`, etc
-- Types from popular theird party libraries such as Pydantic or GenSON.
-
-Outlines also provides special classes for certain output structures (more details below):
-- JSON schemas with `JsonSchema`
-- Regular expressions with `Regex`
-- Context-free grammars with `CFG`
-
-The general idea is that you should provide as an output type what you would give as the type hint of the return type of a function.
-
-Consider the following functions for instance:
-
-```python
-from datetime import date
-from typing import Dict, List, Literal, Union
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-    birth_date: date
-    skills: Union[Dict, List[str]]
-
-def give_int() -> int:
-    ...
-
-def pizza_or_burger() -> Literal["pizza", "burger"]:
-    ...
-
-def create_character() -> Character:
-    ...
-```
-
-With an Outlines model, you can generate text that respects the type hints above by providing those as the output type:
-
-```python
-model("How many minutes are there in one hour", int) # "60"
-model("Pizza or burger", Literal["pizza", "burger"]) # "pizza"
-model("Create a character", Character, max_new_tokens=100) # '{"name": "James", "birth_date": "1980-05-10)", "skills": ["archery", "negociation"]}'
-```
-
-An important difference with function type hints though is that an Outlines generator always returns a string.
-You have to cast the response into the type you want yourself.
-
-For instance:
-
-```python
-result = model("Create a character", Character, max_new_tokens=100)
-casted_result = Character.model_validate_json(result)
-print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
-print(casted_result) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
-```
-
-## Output Type Categories
-
-We can group possible output types in several categories based on the use case they correspond to. While most of those types are native python or types coming from well-known third-party libraries, there are three Outlines-specific types: `JsonSchema`, `Regex` and `CFG`. Their use is explained below.
-
-### Basic Python Types
-
-The most straightfoward form of structed generation is to return an answer that conforms to a given basic type such as an int or a python list. You can use the basic Python types and the types from the `typing` library. For instance:
-
-```python
-from typing import Dict
-
-output_type = float # example of valid value: "0.05"
-output_type = bool # example of valid value: "True"
-output_type = Dict[int, str] # example of valid value: "{1: 'hello', 2: 'there'}"
-```
-
-You can combine types to create more complex response formats by relying on collection types and types such as `Union` and `Optional`. Let's consider for instance the output type below used to represent semi-structured data:
-
-```python
-from typing import Dict, List, Optional, Tuple, Union
-
-output_type = Dict[str, Union[int, str, List[Tuple[str, Optional[float]]]]]
-```
-
-Values created with this output type would be dictionnaries with string as keys and values made of either an integer, a string or a list of two elements tuples: a string and either a float or None. Example of a valid response for text generated with this output type (it would be contained in a string):
-
-```json
-{
-    "name": "Alice",
-    "age": 30,
-    "metrics": [("engagement", 0.85), ("satisfaction", None)]
-}
-```
-
-### Multiple Choices
-
-Outlines supports multiple choice classification by using the `Literal` or `Enum` output types. For instance:
-
-```python
-from enum import Enum
-from typing import Literal
-
-class PizzaOrBurger(Enum):
-    pizza = "pizza
-    burger = "burger
-
-# Equivalent multiple-choice output types
-output_type = Literal["pizza", "burger"]
-output_type = PizzaOrBurger
-```
-
-### JSON Schemas
-
-Multiple different common Python types are often used to store information equivalent to a JSON schema. The following can be used in Outlines to generate text that respects a JSON schema:
-
-- A Pydantic class
-- A Dataclass
-- A TypedDict
-- A [GenSON](https://github.com/wolverdude/GenSON) `SchemaBuilder`
-- A Callable (the parameters are turned into the keys and the type hinting is used to define the types of the values)
-
-For instance:
-
-```python
-from dataclasses import dataclass
-
-@dataclass
-class Character:
-    name: str
-    age: int
-
-ouput_type = Character
-
-def character(name: str, age: int):
-    return None
-
-output_type = character
-```
-
-There are two other JSON schema formats that require Outlines-specific classes: JSON schema strings and dictionaries.
-
-As those are contained in regular Python strings or dictionnaries, the associated output format would be ambiguous if they were to be provided directly. As a result, Outlines requires them to be wrapped in a `outlines.types.JsonSchema` object. For instance:
-
-```python
-from outlines.types import JsonSchema
-
-schema_string = '{"type": "object", "properties": {"answer": {"type": "number"}}}'
-output_type = JsonSchema(schema_string)
-
-schema_dict = {
-    "type": "object",
-    "properties": {
-        "answer": {"type": "number"}
-    }
-}
-output_type = JsonSchema(schema_dict)
-```
-
-### Regex Patterns
-
-Outlines provides support for text generation constrained by regular expressions. Since regular expressions are expressed as simple raw string literals, regex strings must wrapped in an `outlines.types.Regex` object to clarify the expected return type.
-
-```python
-from outlines.types import Regex
-
-regex = r"[0-9]{3}"
-output_type = Regex(regex)
-```
-
-The `outlines.types` module contains a few common regex patterns stored in variables you can import and directly use as output types. Common patterns include a sentence, an email address and an [ISBN reference](https://en.wikipedia.org/wiki/ISBN). For instance:
-
-```python
-from outlines.types import sentence
-
-print(type(sentence)) # outlines.types.dsl.Regex
-print(sentence.pattern) # [A-Z].*\s*[.!?]
-```
-
-To help you create complex regex patterns yourself, you can use the Outlines [regex DSL](../../utility/regex_dsl).
-
-### Context-Free Grammars
-
-Outlines allows you to generate text that respects the syntax of a context-free grammar. Context-free grammars are defined using [Lark](https://lark-parser.readthedocs.io/en/latest/index.html), a grammar language. Since grammars are expressed as a string, Large CFG strings should be be wrapped in an `outlines.types.CFG` object. For instance:
-
-```python
-from outlines.types import CFG
-
-grammar_string = """
-    start: expr
-    expr: "{" expr "}" | "[" expr "]" |
-"""
-output_type = CFG(grammar_string)
-```
-
-You can find a few Lark grammar examples in the [grammars module](../../api_reference/grammars.md).
-
-## Output type availability
-
-The output types presented above are not available for all models as some have only limited support for structured outputs. Please refer to the documentation of the specific model you wish to use to know what output types it supports.
diff --git a/docs/features/index.md b/docs/features/index.md
deleted file mode 100644
index 00df02ba3..000000000
--- a/docs/features/index.md
+++ /dev/null
@@ -1,19 +0,0 @@
-# Features
-
-This section presents in details the different features of Outlines.
-
-## Core Concepts
-
-- [Models](./models/index.md)
-- [Output Types](./core/output_types.md)
-- [Generators](./core/generator.md)
-
-## Utility
-
-- [Applications](./utility/application.md)
-- [Templates](./utility/templates.md)
-- [Regex DSL](./utility/regex_dsl.md)
-
-## Advanced
-
-- [Logits Processors](./advanced/logits_processors.md)
diff --git a/docs/features/models/anthropic.md b/docs/features/models/anthropic.md
deleted file mode 100644
index dd0794386..000000000
--- a/docs/features/models/anthropic.md
+++ /dev/null
@@ -1,98 +0,0 @@
----
-title: Anthropic
----
-
-# Anthropic
-
-!!! Installation
-
-    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines. Install all optional dependencies of the `Anthropic` model with: `pip install outlines[anthropic]`.
-
-    You also need to have an Anthropic API key. This API key must either be set as an environment variable called `ANTHROPIC_API_KEY` or be provided to the `anthropic.Anthropic` class when instantiating it.
-
-## Model Initialization
-
-To create an Anthropic model instance, you can use the `from_anthropic` function. It takes 2 arguments:
-
-- `client`: an `anthropic.Anthropic` instance
-- `model_name`: the name of the model you want to use in subsequent model calls (optional)
-
-For instance:
-
-```python
-from anthropic import Anthropic
-import outlines
-
-# Create the Anthropic client
-client = Anthropic()
-
-# Create the model
-model = outlines.from_anthropic(
-    client,
-    "claude-3-5-sonnet-latest"
-)
-```
-
-Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-claude/models) for an up-to-date list of available models.
-
-## Text Generation
-
-Once you've created your Outlines `Anthropic` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
-
-For instance:
-
-```python
-from anthropic import Anthropic
-import outlines
-
-# Create the model
-model = outlines.from_anthropic(
-    Anthropic(),
-    "claude-3-5-sonnet-latest"
-)
-
-# Call it to generate text
-response = model("What's the capital of Latvia?", max_tokens=20)
-print(response) # 'Riga'
-```
-
-Anthropic does not support structured generation, so you cannot provide an output type. It does support streaming and vision inputs though.
-
-For instance:
-
-```python
-import io
-import requests
-import PIL
-from anthropic import Anthropic
-from outlines import from_anthropic
-from outlines.templates import Vision
-
-# Create the model
-model = from_anthropic(
-    Anthropic(),
-    "claude-3-5-sonnet-latest"
-)
-
-# Function to get an image
-def get_image(url):
-    r = requests.get(url)
-    return PIL.Image.open(io.BytesIO(r.content))
-
-# Create the prompt
-prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
-
-# Stream the response
-for chunk in model.stream(prompt, max_tokens=50):
-    print(chunk) # 'This...'
-```
-
-## Inference arguments
-
-When calling the model or streaming, you can provide keyword arguments that will be passed down to the Anthropic client. Make sure to include all the arguments you need to configure the client's behavior to your expected behavior. Some of the most common arguments include `max_tokens`, `temperature`, `stop_sequences` and `top_k`.
-
-See the [Anthropic API documentation](https://docs.anthropic.com/en/api/messages) for the full list of available arguments.
-
-!!! Warning
-
-    You must set a value for `max_tokens` with Anthropic models.
diff --git a/docs/features/models/dottxt.md b/docs/features/models/dottxt.md
deleted file mode 100644
index 306ef2c62..000000000
--- a/docs/features/models/dottxt.md
+++ /dev/null
@@ -1,74 +0,0 @@
----
-title: Dottxt
----
-
-# Dottxt
-
-!!! Installation
-
-    You need to install the `dottxt` python sdk to be able to use the Dottxt API in Outlines. Install all optional dependencies of the `Dottxt` model with: `pip install outlines[dottxt]`.
-
-    You also need to have a Dottxt API key. This API key must either be set as an environment variable called `DOTTXT_API_KEY` or be provided to the `dottxt.client.Dottxt` class when instantiating it.
-
-## Model Initialization
-
-To create an Dottxt model instance, you can use the `from_dottxt` function. It takes 3 arguments:
-
-- `client`: a `dottxt.client.Dottxt` instance
-- `model_name`: the name of the model you want to use in subsequent model calls (optional)
-- `model_revision`: the name of the revision to use for the model selected (optional)
-
-For instance:
-
-```python
-from dottxt.client import Dottxt
-import outlines
-
-# Create client
-client = Dottxt(api_key="...")
-
-# Create the model
-model = outlines.from_dottxt(
-    client,
-    "meta-llama/Llama-3.1-8B",
-    "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b"
-)
-```
-
-Use the `list_models` method of the Dottxt client to get a list of available model names and revisions for your account.
-
-## Text Generation
-
-Dottxt only supports constrained generation with JSON schema output types. You must always provide a value for the `output_type` parameter as unconstrained generation is not available.
-
-For instance:
-
-```python
-from typing import List
-from pydantic import BaseModel
-from dottxt.client import Dottxt
-import outlines
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-# Create the model
-model = outlines.from_dottxt(
-    Dottxt(),
-    "meta-llama/Llama-3.1-8B",
-    "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b"
-)
-
-# Generate structured text
-result = model("Create a character", Character)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-## Inference arguments
-
-You can provide the same optional parameters you would pass to the `dottxt` sdk's client both during the initialization of the `Dottxt` class and when generating text. Some of the common inference arguments include `max_tokens`, `frequency_penalty`, `presence_penalty` and `temperature`.
-
-Consult the [dottxt python sdk GitHub repository](https://github.com/dottxt-ai/dottxt-python) for the full list of parameters.
diff --git a/docs/features/models/gemini.md b/docs/features/models/gemini.md
deleted file mode 100644
index e6daccf26..000000000
--- a/docs/features/models/gemini.md
+++ /dev/null
@@ -1,148 +0,0 @@
-# Gemini
-
-!!! Installation
-
-    You need to install the `google.genai` libray to be able to use the Gemini API in Outlines. Install all optional dependencies of the `Gemini` model with: `pip install outlines[gemini]`.
-
-    You also need to have a Gemini API key. This API key must either be set as an environment variable called `GEMINI_API_KEY` or be provided to the `google.genai.Client` class when instantiating it.
-
-## Model Initialization
-
-To create a Gemini model instance, you can use the `from_gemini` function. It takes 2 arguments:
-
-- `client`: a `google.genai.Client` instance
-- `model_name`: the name of the model you want to use in subsequent model calls (optional)
-
-For instance:
-
-```python
-import outlines
-from google import genai
-
-# Create the client
-client = genai.Client()
-
-# Create the model
-model = outlines.from_gemini(
-    client,
-    "gemini-1.5-flash-latest"
-)
-```
-
-Check the [Gemini documentation](https://github.com/googleapis/python-genai) for an up-to-date list of available models.
-
-## Text Generation
-
-Once you've created your Outlines `Gemini` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
-
-For instance:
-
-```python
-import outlines
-from google import genai
-
-# Create the model
-model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
-
-# Call it to generate text
-result = model("What's the capital of Latvia?", max_output_tokens=20)
-print(result) # 'Riga'
-```
-
-Gemini also supports streaming. For instance:
-
-```python
-import outlines
-from google import genai
-
-# Create the model
-model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", max_output_tokens=20):
-    print(chunk) # 'In...'
-```
-
-## Structured Generation
-
-Gemini provides supports for some forms of structured output: multiple choice, JSON schema (with caveats) and lists of structured objects. To use it, call the model with an `output_type` on top of your prompt.
-
-#### Multiple Choice
-
-```python
-import outlines
-from google import genai
-from enum import Enum
-
-class PizzaOrBurger(Enum):
-    pizza = "pizza"
-    burger = "burger"
-
-# Create the model
-model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
-
-# Call it with the ouput type to generate structured text
-result = model("Pizza or burger?", PizzaOrBurger, max_output_tokens=20)
-print(result) # 'pizza'
-```
-
-#### JSON Schema
-
-Gemini supports only three types of objects used to define a JSON Schema:
-
-- Pydantic classes
-- Dataclasses
-- TypedDicts
-
-```python
-from typing import List
-from pydantic import BaseModel
-from google import genai
-import outlines
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-# Create the model
-model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
-
-# Call it with the ouput type to generate structured text
-result = model("Create a character", Character)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-#### Lists of Structured Objects
-
-A specificity of Gemini is that, despite not supporting regex, it does support a list of structured objects as an output type. To use it, put any of three available types described above in the typing `List` class
-
-```python
-from dataclasses import dataclass
-from google import genai
-import outlines
-
-@dataclass
-class Character:
-    name: str
-    age: int
-    skills: List[str]
-
-# Create the model
-model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
-
-# Call it with the ouput type to generate structured text
-result = model("Create a character", list[Character])
-print(result) # '[{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}, {["name":...'
-```
-
-!!! Attention
-
-    The structured objects must be in a built-in `list`, not a `List` from the `typing` library
-
-## Inference arguments
-
-You can provide the same optional parameters you would pass to the `google.genai.Client` client both during the initialization of the Gemini model and when generating text. Some of the common inference arguments include `max_output_tokens`, `temperature`, and other generation parameters.
-
-Consult the [Google Generative AI documentation](https://github.com/googleapis/python-genai) for the full list of parameters.
diff --git a/docs/features/models/index.md b/docs/features/models/index.md
deleted file mode 100644
index 89c4fa124..000000000
--- a/docs/features/models/index.md
+++ /dev/null
@@ -1,107 +0,0 @@
----
-title: Models
----
-
-# Models
-
-## Overview
-
-Outlines models are objects that wrap an inference client or engine. Models provide a standardized interface to generate structured text.
-
-!!! Warning
-
-    The model loading functions have been modified in v1. While they used to be called `<name_inference_library>`, they are now called `from_<name_inference_library>`. The model classes' names and `__init__` methods are left unchanged.
-
-
-All Outlines model classes have an associated loader function to facilitate initializing a model instance. The name of this function is `from_` plus the name of the model in lower-case letters. For instance, Outlines has a `Transformers` model and an associated `from_transformers` loader function. The parameters to load a model are specific to each provider, please consult the documentation of the model you want to use for more information.
-
-After having created a model instance, you can either directly call it to generate text or first create a reusable generator that you would then call. In either case, you can provide an `output_type` to constrain the format of the generation output. See the [output types section](../core/output_types.md) for more information on constrained generation.
-
-For instance:
-
-```python
-from outlines import from_transformers, Generator
-import transformers
-
-# Create a model
-model = from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-)
-
-# Call it directly
-response = model("How many countries are there in the world", max_new_tokens=20)
-print(result) # 'There are 200 countries in the world.'
-
-# Call it directly with an output_type
-response = model("How many countries are there in the world", int, max_new_tokens=20)
-print(result) # '200'
-
-# Create a generator first and then call it
-generator = Generator(model, int)
-response = generator("How many countries are there in the world")
-print(result) # '200'
-```
-
-## Features Matrix
-
-| | [Anthropic](../../models/anthropic) | [Dottxt](../../models/dottxt) | [Gemini](../../models/gemini) | [LlamaCpp](../../models/llamacpp) | [MLXLM](../../models/mlxlm) | [Ollama](../../models/ollama) | [OpenAI](../../models/openai) | [SGLang](../../models/sglang) | [TGI](../../models/tgi) | [Transformers](../../models/transformers) | [Transformers MultiModal](../../models/transformers_multimodal) | [VLLM](../../models/vllm) | [VLLMOffline](../../models/vllm_offline) |
-|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
-| **Output Types** | | | | | | | | | | | | | |
-| Simple Types | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| JSON Schema | ❌ | ✅ | 🟠 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| Multiple Choice | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| Regex | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| Grammar | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | 🟠 | ❌ | ✅ | ✅ | ✅ | ✅ |
-| **Generation Features** | | | | | | | | | | | | | |
-| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
-| Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
-| Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
-| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ |
-
-## Model Types
-
-Models can be divided into two categories: local models and server-based models.
-
-In the case of local models, the text generation happens within the inference library object used to instantite the model. This gives Outlines direct access to the generation process (through a logits processor) and means all structured generation output types are available.
-
-The local models available are the following:
-
-- LlamaCpp
-- MLXLM
-- Transformers
-- TransformersMultiModal
-- VLLMOffline
-
-In the case of server-based models, the model is initialized with a client that sends a request to a server that is in charge of the actual text generation. As a result, we have limited control over text generation and some output types are not supported. The server on which the text generation happens can either be remote (with OpenAI or Anthopic for instance) or local (with SGLang for instance).
-
-The server-based models available are the following:
-
-- Anthropic
-- Dottxt
-- Gemini
-- Ollama
-- OpenAI
-- SgLang
-- TGI
-- VLLM
-
-Some models have an async version. To use them, just pass the async version of the provider object to their loading function. It will then return a `Async<ModelName>` instance with the same methods and features as the regular sync instance.
-
-For instance:
-
-```python
-from outlines import from_tgi
-from huggingface_hub import AsyncInferenceClient
-
-model = from_tgi(
-    AsyncInferenceClient("http://localhost:8000/v1")
-)
-print(type(model)) # outlines.models.tgi.AsyncTGI
-```
-
-The models that have an async version are the following:
-
-- SgLang
-- TGI
-- VLLM
diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
deleted file mode 100644
index cf7dd3e0c..000000000
--- a/docs/features/models/llamacpp.md
+++ /dev/null
@@ -1,168 +0,0 @@
----
-title: llama.cpp
----
-
-# llama.cpp
-
-Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library](https://github.com/abetlen/llama-cpp-python). Llamacpp allows to run quantized models on machines with limited compute.
-
-!!! Installation
-
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration. Install all optional dependencies of the `LlamaCpp` model with: `pip install outlines[llamacpp]`.
-
-    See the [llama-cpp-python Github page](https://github.com/abetlen/llama-cpp-python?tab=readme-ov-file#supported-backends) for instructions on installing with CUDA, Metal, ROCm and other backends.
-
-
-## Model Initialization
-
-To load the model, you can use the `from_llamacpp` function. The single argument of the function is a `Llama` model instance from the `llama_cpp` library. Consult the [Llama class API reference](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama) for detailed information on how to create a model instance and on the various available parameters.
-
-For instance:
-
-```python
-import outlines
-from llama_cpp import Llama
-
-model = outlines.from_llamacpp(
-    Llama.from_pretrained(
-        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
-    )
-)
-```
-
-## Text Generation
-
-To generate text, you can simply call the model with a prompt.
-
-For instance:
-
-```python
-import outlines
-from llama_cpp import Llama
-
-# Create the model
-model = outlines.from_llamacpp(
-    Llama.from_pretrained(
-        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
-    )
-)
-
-# Call it to generate text
-result = model("What's the capital of Latvia?", max_tokens=20)
-print(result) # 'Riga'
-```
-
-The `LlamaCpp` model also supports streaming. For instance:
-
-```python
-import outlines
-from llama_cpp import Llama
-
-# Create the model
-model = outlines.from_llamacpp(
-    Llama.from_pretrained(
-        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
-    )
-)
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk) # 'In...'
-```
-
-## Structured Generation
-
-The `LlamaCpp` model supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
-
-### Basic Type
-
-```python
-import outlines
-from llama_cpp import Llama
-
-output_type = int
-
-model = outlines.from_llamacpp(
-    Llama.from_pretrained(
-        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
-    )
-)
-
-result = model("How many countries are there in the world?", output_type)
-print(result) # '200'
-```
-
-### JSON Schema
-
-```python
-from typing import List
-from pydantic import BaseModel
-import outlines
-from llama_cpp import Llama
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-model = outlines.from_llamacpp(
-    Llama.from_pretrained(
-        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
-    )
-)
-
-result = model("Create a character.", output_type=Character, max_tokens=200)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-### Multiple Choice
-
-```python
-from typing import Literal
-import outlines
-from llama_cpp import Llama
-
-output_type = Literal["Paris", "London", "Rome", "Berlin"]
-
-model = outlines.from_llamacpp(
-    Llama.from_pretrained(
-        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
-    )
-)
-
-result = model("What is the capital of France?", output_type)
-print(result) # 'Paris'
-```
-
-### Regex
-
-```python
-from outlines.types import Regex
-import outlines
-from llama_cpp import Llama
-
-output_type = Regex(r"\d{3}-\d{2}-\d{4}")
-
-model = outlines.from_llamacpp(
-    Llama.from_pretrained(
-        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
-        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
-    )
-)
-
-result = model("Generate a fake social security number.", output_type)
-print(result) # '782-32-3789'
-```
-
-## Inference Arguments
-
-When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `__call__` method of the `llama_cpp.Llama` model. Some common inference arguments include `max_tokens`, `temperature`, `frequency_penalty` and `top_p`.
-
-See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__) for more information on inference parameters.
diff --git a/docs/features/models/mlxlm.md b/docs/features/models/mlxlm.md
deleted file mode 100644
index d2cba95fc..000000000
--- a/docs/features/models/mlxlm.md
+++ /dev/null
@@ -1,189 +0,0 @@
----
-title: mlx-lm
----
-
-# mlx-lm
-
-Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
-
-!!! Note "Installation"
-
-    You need a device that [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration.
-
-    You need to install the `mlx` and `mlx-lm` libraries to be able to use mlx in Outlines. Install all optional dependencies of the `MLXLM` model with: `pip install outlines[mlxlm]`.
-
-## Model Initialization
-
-To create a MLXLM model instance, you can use the `from_mlxlm` function. It takes 2 arguments:
-
-- `model`: an `mlx.nn.Module` instance
-- `tokenizer`: a `transformers.PreTrainedTokenizer` instance
-
-However, we recommend you simply pass on the output of the `mlx_lm.load` function (it takes a model name as an argument).
-
-For instance:
-
-```python
-import outlines
-import mlx_lm
-
-# Create the model
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-```
-
-## Text Generation
-
-To generate text, you can simply call the model with a prompt.
-
-For instance:
-
-```python
-import outlines
-import mlx_lm
-
-# Load the model
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-
-# Call it to generate text
-result = model("What's the capital of Latvia?", max_tokens=20)
-print(result) # 'Riga'
-```
-
-The `MLXLM` model also supports streaming. For instance:
-
-```python
-import outlines
-import mlx_lm
-
-# Load the model
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk) # 'In...'
-```
-
-## Structured Generation
-
-As a local model, `MLXLM` supports all forms of structured generation available in Outlines.
-
-### Basic Type
-
-```python
-import outlines
-import mlx_lm
-
-output_type = int
-
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-
-result = model("How many countries are there in the world?", output_type)
-print(result) # '200'
-```
-
-### JSON Schema
-
-```python
-from pydantic import BaseModel
-from typing import List
-import outlines
-import mlx_lm
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-
-result = model("Create a character.", output_type=Character)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-### Multiple Choice
-
-```python
-from typing import Literal
-import outlines
-import mlx_lm
-
-output_type = Literal["Paris", "London", "Rome", "Berlin"]
-
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-
-result = model("What is the capital of France?", output_type)
-print(result) # 'Paris'
-```
-
-### Regex
-
-```python
-from outlines.types import Regex
-import outlines
-import mlx_lm
-
-output_type = Regex(r"\d{3}-\d{2}-\d{4}")
-
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-
-result = model("Generate a fake social security number.", output_type)
-print(result) # '782-32-3789'
-```
-
-### Context-Free Grammar
-
-```python
-from outlines.types import CFG
-import outlines
-import mlx_lm
-
-arithmetic_grammar = """
-?start: sum
-
-?sum: product
-| sum "+" product   -> add
-| sum "-" product   -> sub
-
-?product: atom
-| product "*" atom  -> mul
-| product "/" atom  -> div
-
-?atom: NUMBER           -> number
-| "-" atom         -> neg
-| "(" sum ")"
-
-%import common.NUMBER
-%import common.WS_INLINE
-
-%ignore WS_INLINE
-"""
-output_type = CFG(arithmetic_grammar)
-
-model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-)
-
-result = model("Write an addition.", output_type, max_tokens=20)
-print(result) # '23 + 48'
-```
-
-## Inference Arguments
-
-When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `mlx_lm.generate` function used to generate text.
-
-See the [MLXLM documentation](https://github.com/ml-explore/mlx-lm) for more information on inference parameters.
diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
deleted file mode 100644
index eaad4c3e2..000000000
--- a/docs/features/models/ollama.md
+++ /dev/null
@@ -1,103 +0,0 @@
----
-title: Ollama
----
-
-# Ollama
-
-!!! Installation
-
-    To be able to use Ollama in Outlines, you must install both Ollama and the optional dependency libraries of the model.
-
-    - To download Ollama: https://ollama.com/download
-    - To install the ollama python sdk: `pip install outlines[ollama]`
-
-    Consult the [`ollama` documentation](https://github.com/ollama/ollama-python) for detailed information on installation and client initialization.
-
-## Model Initialization
-
-To create an Ollama model instance, you can use the `from_ollama` function. It takes 2 arguments:
-
-- `client`: an `ollama.Client` instance
-- `model_name`: the name of the model you want to use
-
-For instance:
-
-```python
-import ollama
-import outlines
-
-# Create the client
-client = ollama.Client()
-
-# Create the model
-model = outlines.from_ollama(
-    client,
-    "tinyllama",
-)
-```
-
-You can find the list of available models on the [Ollama library](https://ollama.com/library).
-
-## Text Generation
-
-Once you've created your Outlines `Ollama` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
-
-For instance:
-
-```python
-import ollama
-import outlines
-
-# Create the model
-model = outlines.from_ollama(ollama.Client(), "tinyllama")
-
-# Call it to generate text
-response = model("What's the capital of Latvia?")
-print(response) # 'Riga'
-```
-
-Ollama also supports streaming. For instance:
-
-```python
-import ollama
-import outlines
-
-# Create the model
-model = outlines.from_ollama(ollama.Client(), "tinyllama")
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat"):
-    print(chunk) # 'In...'
-```
-
-## Structured Generation
-
-Ollama only provides support for structured generation based on a JSON schema. To use it, call the model with a JSON schema object as an `output_type` on top of your prompt.
-
-For instance:
-
-```python
-from typing import List
-from pydantic import BaseModel
-import ollama
-import outlines
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-# Create the model
-model = outlines.from_ollama(ollama.Client(), "tinyllama")
-
-# Call it with the output type to generate structured text
-result = model("Create a character", Character)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-## Inference arguments
-
-When calling the model, you can provide keyword arguments that will be passed down to the `generate` method of the Ollama client.
-
-Consult the [Ollama REST API documentation](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-completion) for the full list of inference parameters.
diff --git a/docs/features/models/openai.md b/docs/features/models/openai.md
deleted file mode 100644
index e4532d213..000000000
--- a/docs/features/models/openai.md
+++ /dev/null
@@ -1,134 +0,0 @@
-# OpenAI
-
-!!! Installation
-
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
-
-    You also need to have an OpenAI API key. This API key must either be set as an environment variable called `OPENAI_API_KEY` or be provided to the `openai.OpenAI` class when instantiating it.
-
-## Model Initialization
-
-To create an OpenAI model instance, you can use the `from_openai` function. It takes 2 arguments:
-
-- `client`: an `openai.OpenAI` or `openai.AzureOpenAI` instance
-- `model_name`: the name of the model you want to use
-
-For instance:
-
-```python
-import outlines
-import openai
-
-# Create the client
-client = openai.OpenAI()
-
-# Create the model
-model = outlines.from_openai(
-    client,
-    "gpt-4o"
-)
-```
-
-Check the [OpenAI documentation](https://platform.openai.com/docs/models) for an up-to-date list of available models. As shown above, you can use Azure OpenAI in Outlines the same way you would use OpenAI, just provide an `openai.AzureOpenAI` instance to the Outlines model class.
-
-## Text Generation
-
-Once you've created your Outlines `OpenAI` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
-
-For instance:
-
-```python
-import openai
-import outlines
-
-# Create the model
-model = outlines.from_openai(
-    openai.OpenAI(),
-    "gpt-4o"
-)
-
-# Call it to generate text
-response = model("What's the capital of Latvia?", max_tokens=20)
-print(response) # 'Riga'
-```
-
-OpenAI supports streaming and vision inputs.
-
-For instance:
-
-```python
-import io
-import requests
-import PIL
-import openai
-import outlines
-from outlines.templates import Vision
-
-# Create the model
-model = outlines.from_openai(
-    openai.OpenAI(),
-    "gpt-4o"
-)
-
-# Function to get an image
-def get_image(url):
-    r = requests.get(url)
-    return PIL.Image.open(io.BytesIO(r.content))
-
-# Create the prompt
-prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
-
-# Stream the response
-for chunk in model.stream(prompt, max_tokens=50, stop="."):
-    print(chunk) # 'This...'
-```
-
-## Structured Generation
-
-OpenAI provides supports for some forms of structured output: JSON schemas and JSON syntax. To use it, call the model with an `output_type` on top of your prompt.
-
-#### JSON Schema
-
-```python
-from typing import List
-from pydantic import BaseModel
-import openai
-import outlines
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-# Create the model
-model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
-
-# Call it with the output type to generate structured text
-result = model("Create a character, use the json format.", Character, top_p=0.1)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-#### JSON Syntax
-
-What we mean by JSON syntax is what is sometimes called JSON mode, meaning that the model will return a valid JSON, but you do not get to specify its structure. To use this JSON mode, provide the `dict` type as an output type.
-
-```python
-import openai
-import outlines
-
-# Create the model
-model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
-
-# Call it with the output type to generate structured text
-result = model("Create a character, use the json format.", dict, temperature=0.5)
-print(result) # '{"first_name": "Henri", "last_name": "Smith", "height": "170"}'
-```
-
-## Inference arguments
-
-When calling the model, you can provide keyword arguments that will be passed down to the `chat.completions.create` method of the OpenAI client. Some of the most common arguments include `max_tokens`, `temperature`, `stop` and `top_p`.
-
-Another keyword argument of interest is `n`. If set with an integer value superior to 1, OpenAI will generate several sample responses and you will receive a list of strings as a response to your model call.
-
-See the [OpenAI API documentation](http://platform.openai.com/docs/api-reference/chat/create) for the full list of available arguments.
diff --git a/docs/features/models/openai_compatible.md b/docs/features/models/openai_compatible.md
deleted file mode 100644
index 4925d53d4..000000000
--- a/docs/features/models/openai_compatible.md
+++ /dev/null
@@ -1,87 +0,0 @@
-# OpenAI-Compatible APIs
-
-Many inference providers offer OpenAI-compatible APIs, allowing you to use the familiar OpenAI SDK while connecting to different backends. Outlines allows you can leverage various providers while maintaining consistent code.
-
-## What are OpenAI-Compatible APIs?
-
-OpenAI-compatible APIs implement the same REST endpoints and request/response formats as OpenAI's API, but serve different models or run on different infrastructure. This allows you to use the `openai` Python library with any compatible provider by simply changing the `base_url`.
-
-!!! Installation
-
-    You need to install the `openai` library to be able to use the OpenAI-compatible APIs in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
-
-## General Usage Pattern
-
-The standard approach is to use the OpenAI SDK with a custom base URL:
-
-```python
-import openai
-import outlines
-
-# Point to your OpenAI-compatible endpoint
-client = openai.OpenAI(
-    base_url="https://your-provider.com/v1",  # Custom endpoint
-    api_key="your-api-key"
-)
-
-# Use with Outlines
-model = outlines.from_openai(client, "model-name")
-```
-
-## Important: Provider-Specific Parameters
-
-!!! Warning "API-Specific Parameters"
-
-    Some providers require additional parameters in the API request for structured generation to work properly. These are typically passed as extra arguments when calling the model.
-
-    For example, some providers may need special parameters in the request body to enable guided generation or specify constraints. Always consult your provider's documentation for structured generation requirements.
-
-## Popular OpenAI-Compatible Providers
-
-Many providers offer OpenAI-compatible endpoints:
-
-- **Groq**
-- **Together AI**
-- **Anyscale**
-- **Fireworks AI**
-- **Perplexity**
-- **Local servers** (LocalAI, etc.)
-
-## Configuration Examples
-
-### Basic Setup
-```python
-import openai
-import outlines
-
-# Generic OpenAI-compatible setup
-client = openai.OpenAI(
-    base_url="https://api.your-provider.com/v1",
-    api_key="your-api-key"
-)
-
-model = outlines.from_openai(client, "provider-model-name")
-```
-
-### With Authentication Headers
-```python
-import openai
-import outlines
-
-# Some providers need custom headers
-client = openai.OpenAI(
-    base_url="https://api.your-provider.com/v1",
-    api_key="your-api-key",
-    default_headers={"Custom-Header": "value"}
-)
-
-model = outlines.from_openai(client, "provider-model-name")
-```
-
-## Related Documentation
-
-For specific implementations that use OpenAI-compatible APIs:
-
-- [SGLang](sglang.md): Local inference server with OpenAI-compatible endpoints
-- [vLLM](vllm.md): High-performance inference with OpenAI-compatible API
-- [OpenAI](openai.md): The original OpenAI API implementation
diff --git a/docs/features/models/sglang.md b/docs/features/models/sglang.md
deleted file mode 100644
index 95a6eb7b5..000000000
--- a/docs/features/models/sglang.md
+++ /dev/null
@@ -1,171 +0,0 @@
----
-title: SGLang
----
-
-# SGLang
-
-## Prerequisites
-
-The Outlines `SGLang` model is intended to be used along with an SGLang instance running on a separate server (can be local or remote). Make sure you have a SGLang server running and accessible before using the `SGLang` model. As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies of the `SGLang` model with: `pip install outlines[sglang]`.
-
-When launching your SGLang server, you can specify the backend engine to use for structured generation through the `grammar-backend` cli argument. Add `--grammar-backend outlines` to your command to use Outlines instead of the default engine.
-
-## Model Initialization
-
-To load the model, you can use the `from_sglang` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Make sure the value of the `base_url` argument of the `OpenAI` client points to your running SGLang server. Consult the [SGLang documentation](https://docs.sglang.ai/backend/send_request.html) on using an OpenAI client with an SGLang server for more information.
-
-Based on whether the `openai` client instance is synchronous or asynchronous, you will receive a `SGLang` or `AsyncSGLang` model instance.
-
-For instance:
-
-```python
-import openai
-import outlines
-
-# Create the OpenAI client
-sync_openai_client = openai.OpenAI(base_url="http://localhost:11434")
-async_openai_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
-
-# Create a sync model
-sync_model = outlines.from_sglang(sync_openai_client)
-print(type(sync_model)) # <class 'outlines.models.sglang.SGLang'>
-
-# Create an async model
-async_model = outlines.from_sglang(async_openai_client)
-print(type(async_model)) # <class 'outlines.models.sglang.AsyncSGLang'>
-```
-
-## Text Generation
-
-To generate text, you can simply call the model with a prompt.
-
-For instance:
-
-```python
-import openai
-import outlines
-
-# Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-# Call it to generate text
-result = model("Write a short story about a cat.", max_tokens=100)
-print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
-```
-
-The `SGLang` model also supports streaming. For instance:
-
-```python
-import openai
-import outlines
-
-# Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk) # 'In ...'
-```
-
-## Structured Generation
-
-SGLang supports all output types available in Outlines (context-free grammars with caveats though, see the subsection below for more details). Simply provide an `output_type` after the prompt when calling the model.
-
-### Simple Type
-
-```python
-import openai
-import outlines
-
-output_type = int
-
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-result = model("How many countries are there in the world?", output_type)
-print(result) # '200'
-```
-
-### JSON Schema
-
-```python
-import openai
-import outlines
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-result = model("Create a character.", Character, frequency_penalty=1.5)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-### Multiple Choice
-
-```python
-from typing import Literal
-import openai
-import outlines
-
-output_type = Literal["Paris", "London", "Rome", "Berlin"]
-
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-result = model("What is the capital of France?", output_type, temperature=0)
-print(result) # 'Paris'
-```
-
-### Regex
-
-```python
-import openai
-import outlines
-from outlines.types import Regex
-
-output_type = Regex(r"\d{3}-\d{2}-\d{4}")
-
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-result = model("Generate a fake social security number.", output_type, top_p=0.1)
-print(result) # '782-32-3789'
-```
-
-### Context-Free Grammar
-
-SGLang supports grammars, but expects an EBNF format instead of the Lark format Outlines uses. Thus, to use a context-free grammar with SGLang, provide a string using the EBNF syntax to the Outlines `CFG` object.
-
-```python
-import openai
-import outlines
-from outlines.types import CFG
-
-ebnf_grammar = """
-root ::= answer
-answer ::= "yes" | "no"
-"""
-output_type = CFG(ebnf_grammar)
-
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-result = model("Is the weather good today?", output_type)
-print(result) # 'yes'
-```
-
-## Inference Arguments
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.
-
-An optional parameter of particular interest is `extra_body`, which is a dictionary containing arguments that are specific to SGLang and are not part of the standard `openai` interface.
-
-See the [SGLang documentation](https://docs.sglang.ai/backend/openai_api_completions.html) on parameters for the OpenAI-compatible server for more information on inference parameters.
diff --git a/docs/features/models/tgi.md b/docs/features/models/tgi.md
deleted file mode 100644
index 8fb807580..000000000
--- a/docs/features/models/tgi.md
+++ /dev/null
@@ -1,137 +0,0 @@
----
-title: TGI
----
-
-# TGI
-
-## Prerequisites
-
-The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed. Install all optional dependencoes of the `TGI` model with: `pip install outlines[tgi]`
-
-## Model Initialization
-
-To load the model, you can use the `from_tgi` function. The argument of the function is either an `InferenceClient` or `AsyncInferenceClient` instance from the `huggingface_hub` library. Consult the [HuggingFace documentation](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client) for more information on their inference client.
-
-Based on whether the inference client instance is synchronous or asynchronous, you will receive a `TGI` or an `AsyncTGI` model instance.
-
-For instance:
-
-```python
-import outlines
-import huggingface_hub
-
-# Create the inference client
-client = huggingface_hub.InferenceClient("http://localhost:11434")
-async_client = huggingface_hub.AsyncInferenceClient("http://localhost:11434")
-
-# Create a sync model
-sync_model = outlines.from_tgi(client)
-print(type(sync_model))  # <class 'outlines.models.tgi.TGI'>
-
-# Create an async model
-async_model = outlines.from_tgi(async_client)
-print(type(async_model))  # <class 'outlines.models.tgi.AsyncTGI'>
-```
-
-## Text Generation
-
-To generate text, you can simply call the model with a prompt.
-
-For instance:
-
-```python
-import outlines
-import huggingface_hub
-
-# Create the model
-client = huggingface_hub.InferenceClient("http://localhost:11434")
-model = outlines.from_tgi(client)
-
-# Call it to generate text
-result = model("Write a short story about a cat.", stop_sequences=["."])
-print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
-```The `TGI` model supports streaming. For instance:
-
-```python
-import outlines
-import huggingface_hub
-
-# Create the model
-client = huggingface_hub.InferenceClient("http://localhost:11434")
-model = outlines.from_tgi(client)
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", stop_sequences=["."]):
-    print(chunk) # 'In ...'
-```
-
-## Structured Generation
-
-TGI supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
-
-### Simple Type
-
-```python
-import outlines
-import huggingface_hub
-
-output_type = int
-
-tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
-model = outlines.from_tgi(tgi_client)
-
-result = model("How many countries are there in the world?", output_type)
-print(result) # '200'
-```### JSON Schema
-
-```python
-import outlines
-import huggingface_hub
-from typing import List
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
-model = outlines.from_tgi(tgi_client)
-
-result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```### Multiple Choice
-
-```python
-import outlines
-import huggingface_hub
-from typing import Literal
-
-output_type = Literal["Paris", "London", "Rome", "Berlin"]
-
-tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
-model = outlines.from_tgi(tgi_client)
-
-result = model("What is the capital of France?", output_type, temperature=0)
-print(result) # 'Paris'
-```### Regex
-
-```python
-import outlines
-import huggingface_hub
-from outlines.types import Regex
-
-output_type = Regex(r"\d{3}-\d{2}-\d{4}")
-
-tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
-model = outlines.from_tgi(tgi_client)
-
-result = model("Generate a fake social security number.", output_type, top_p=0.1)
-print(result) # '782-32-3789'
-```
-## Inference parameters
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `text_generation` method of the TGI client.
-
-Common parameters include `max_new_tokens`, `stop_sequences`, `temperature`, `top_k`, `top_p`, and others as specified in the [TGI inference client documentation](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client#huggingface_hub.InferenceClient).
diff --git a/docs/features/models/transformers.md b/docs/features/models/transformers.md
deleted file mode 100644
index 9dc1cab79..000000000
--- a/docs/features/models/transformers.md
+++ /dev/null
@@ -1,207 +0,0 @@
----
-title: Transformers
----
-
-# Transformers
-
-!!! Installation
-
-    You need to install the `transformers` library to be able to use the Transformers in Outlines. Install all optional dependencies of the `Transformers` model with: `pip install outlines[transformers]`.
-
-    See the [HuggingFace documentation](https://huggingface.co/docs/transformers/en/installation) for more information on installing `transformers` with CPU, GPU...
-
-## Model Initialization
-
-To load the model, you can use the `from_transformers` function. It takes 2 arguments:
-
-- `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
-- `tokenizer_or_processor`: a `transformers` tokenizer (created with `AutoTokenizer` for instance, it must be an instance of either `PreTrainedTokenizer` or `PreTrainedTokenizerFast`)
-
-For instance:
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-# Create the transformers model and tokenizer
-hf_model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-hf_tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-
-# Create the Outlines model
-model = outlines.from_transformers(hf_model, hf_tokenizer)
-```
-
-If you provide a processor instead of a tokenizer for the second argument of the `from_transformers` function, you would get a `TransformersMultiModal` instance. See the [TransformersMultiModal model documentation](./transformers_multimodal.md) for more information on using multimodal models in Outlines.
-
-## Text Generation
-
-To generate text, you can simply call the model with a prompt.
-
-For instance:
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-# Create model
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-# Call it to generate text
-result = model("What's the capital of Latvia?", max_new_tokens=20)
-print(result) # 'Riga'
-```
-
-The `Transformers` model also supports batch generation. To use it, provide a list of prompts instead of a single prompt. You will receive as result a list of completions. For instance:
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-# Create model
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-# Create a list of prompts that will be used in a single batch
-prompts = [
-    "What's the capital of Lithuania?",
-    "What's the capital of Latvia?",
-    "What's the capital of Estonia?"
-]
-
-# Call it to generate text
-result = model(prompts, max_new_tokens=20)
-print(result) # ['Vilnius', 'Riga', 'Tallinn']
-```
-
-## Structured Generation
-
-As a local model, `Transformers` supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
-
-### Simple Type
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-output_type = int
-
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("How many countries are there in the world?", output_type, max_new_tokens=5)
-print(result) # '200'
-```
-
-### JSON Schema
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from pydantic import BaseModel
-from typing import List
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("Create a character.", output_type=Character, max_new_tokens=200, repetition_penalty=0.5)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-### Multiple Choice
-
-```python
-from typing import Literal
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-output_type = Literal["Paris", "London", "Rome", "Berlin"]
-
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("What is the capital of France?", output_type, max_new_tokens=10, temperature=0)
-print(result) # 'Paris'
-```
-
-### Regex
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from outlines.types import Regex
-
-output_type = Regex(r"\d{3}-\d{2}-\d{4}")
-
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("Generate a fake social security number.", output_type, max_new_tokens=20, top_p=0.5)
-print(result) # '782-32-3789'
-```
-
-### Context-Free Grammar
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from outlines.types import CFG
-
-arithmetic_grammar = """
-?start: sum
-
-?sum: product
-| sum "+" product   -> add
-| sum "-" product   -> sub
-
-?product: atom
-| product "*" atom  -> mul
-| product "/" atom  -> div
-
-?atom: NUMBER           -> number
-| "-" atom         -> neg
-| "(" sum ")"
-
-%import common.NUMBER
-%import common.WS_INLINE
-
-%ignore WS_INLINE
-"""
-output_type = CFG(arithmetic_grammar)
-
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("Write an addition.", output_type, max_new_tokens=100)
-print(result) # '23 + 48'
-```
-
-## Inference Arguments
-
-When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `generate` method of the `transformers` model. Some common inference arguments include `max_new_tokens`, `temperature`, `repetition_penalty` and `top_p`.
-
-See the [transformers documentation](https://huggingface.co/docs/transformers/en/main_classes/text_generation) for more information on inference parameters.
-
-!!! Warning
-
-    The `max_new_tokens` inference parameter has a default value of 20. This is insufficient for most tasks and will result in the generation output not respecting the output type (because the response is truncated). We recommend you always provide a value for this argument.
diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
deleted file mode 100644
index b2b846f6c..000000000
--- a/docs/features/models/transformers_multimodal.md
+++ /dev/null
@@ -1,89 +0,0 @@
----
-title: Transformers MultiModal
----
-
-# Transformers MultiModal
-
-The Outlines `TransformersMultiModal` model inherits from `Transformers` and shares most of its interface. Please start by reading the [Transformers documentation](./transformers.md) as this document only focuses on the specificities of `TransformersMultiModal` compared to `Transformers`.
-
-On top of the difference of arguments to provide to the `from_transformers` function as described below, the only other specificity of the `TransformersMultiModal` model is the format of the prompt to use when calling the model. For all other elements such as the output types and inference arguments, it works exactly the same way as `Transformers`.
-
-## Model Initialization
-
-To load the model, you can use the `from_transformers` function. It takes 2 arguments:
-
-- `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
-- `tokenizer_or_processor`: a `transformers` processor (created with `AutoProcessor` for instance, it must be an instance of `ProcessorMixin`)
-
-For instance:
-
-```python
-import outlines
-from transformers import AutoModelForCausalLM, AutoProcessor
-
-# Create the transformers model and processor
-hf_model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-hf_processor = AutoProcessor.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-
-# Create the Outlines model
-model = outlines.from_transformers(hf_model, hf_processor)
-```
-
-## Model Input
-
-Instead of a string or a list of strings (for batch generation), you should provide a dictionnary or a list of dictionnaries as a prompt when calling the `TransformersMultiModal` model. The dictionnary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
-
-```python
-{
-    "text": Union[str, List[str]]
-    "<other_keys_depending_on_your_processor>": Union[Any, List[Any]]
-}
-```
-
-Some common keys to include for processors are `images` for vision models or `audios` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets or if you are using batch generation).
-
-Here's an example of using a vision multimodal model:
-
-```python
-from io import BytesIO
-from urllib.request import urlopen
-
-from PIL import Image
-from pydantic import BaseModel
-from transformers import (
-    LlavaForConditionalGeneration,
-    AutoProcessor,
-)
-
-import outlines
-
-TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-
-class Animal(BaseModel):
-    specie: str
-    color: str
-    weight: int
-
-def get_image_from_url():
-    img_byte_stream = BytesIO(urlopen(IMAGE_URL).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-# Create a model
-model = outlines.from_transformers(
-    LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
-    AutoProcessor.from_pretrained(TEST_MODEL),
-)
-
-# Call it with a model input dict containing a text prompt and an image + an output type
-result = model(
-    {"text": "<image>Describe this animal.", "images": get_image_from_url(IMAGE_URL)},
-    Animal,
-)
-print(result) # '{"specie": "cat", "color": "white", "weight": 4}'
-print(Animal.model_validate_json(result)) # specie=cat, color=white, weight=4
-```
-
-!!! Warning
-
-    Make sure your prompt (the value of the `text` key) contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are images present in the value of the `images` key.
diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
deleted file mode 100644
index 21a048f9f..000000000
--- a/docs/features/models/vllm.md
+++ /dev/null
@@ -1,186 +0,0 @@
----
-title: vLLM
----
-
-# vLLM
-
-## Prerequisites
-
-The Outlines `VLLM` model is intended to be used along with a vLLM instance running on a separate server (can be local or remote). Make sure you have a vLLM server running and accessible before using the `VLLM` model. As the vLLM client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies for the `VLLM` model with: `pip install openai`.
-
-If you want to use the vllm offline inference mode instead of the server mode, please refer to the [VLLMOffline](./vllm_offline.md) model documentation.
-
-## Model Initialization
-
-To load the model, you can use the `from_vllm` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Make sure the value of the `base_url` argument of the `OpenAI` client points to your running vLLM server. Consult the [vLLM documentation](https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html) on using an OpenAI client with a vLLM server for more information.
-
-Based on whether the `openai` client instance is synchronous or asynchronous, you will receive a `VLLM` or `AsyncVLLM` model instance.
-
-For instance:
-
-```python
-import openai
-import outlines
-
-# Create the OpenAI client
-sync_openai_client = openai.OpenAI(base_url="http://localhost:8080")
-async_openai_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
-
-# Create a sync model
-sync_model = outlines.from_vllm(sync_openai_client)
-print(type(sync_model)) # <class 'outlines.models.vllm.VLLM'>
-
-# Create an async model
-async_model = outlines.from_vllm(async_openai_client)
-print(type(async_model)) # <class 'outlines.models.vllm.AsyncVLLM'>
-```
-
-## Text Generation
-
-To generate text, you can simply call the model with a prompt.
-
-For instance:
-
-```python
-import openai
-import outlines
-
-# Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
-model = outlines.from_vllm(openai_client)
-
-# Call it to generate text
-result = model("Write a short story about a cat.", max_tokens=100)
-print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
-```
-
-The `VLLM` model also supports streaming. For instance:
-
-```python
-import openai
-import outlines
-
-# Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
-model = outlines.from_vllm(openai_client)
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk) # 'In ...'
-```
-
-## Structured Generation
-
-vLLM supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
-
-### Simple Type
-
-```python
-import openai
-import outlines
-
-output_type = int
-
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
-model = outlines.from_vllm(openai_client)
-
-result = model("How many countries are there in the world?", output_type)
-print(result) # '200'
-```
-
-### JSON Schema
-
-```python
-import openai
-import outlines
-from typing import List
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
-model = outlines.from_vllm(openai_client)
-
-result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-### Multiple Choice
-
-```python
-from typing import Literal
-import openai
-import outlines
-
-output_type = Literal["Paris", "London", "Rome", "Berlin"]
-
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
-model = outlines.from_vllm(openai_client)
-
-result = model("What is the capital of France?", output_type, temperature=0)
-print(result) # 'Paris'
-```
-
-### Regex
-
-```python
-import openai
-import outlines
-from outlines.types import Regex
-
-output_type = Regex(r"\d{3}-\d{2}-\d{4}")
-
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
-model = outlines.from_vllm(openai_client)
-
-result = model("Generate a fake social security number.", output_type, top_p=0.1)
-print(result) # '782-32-3789'
-```
-
-### Context-Free Grammar
-
-```python
-import openai
-import outlines
-from outlines.types import CFG
-
-arithmetic_grammar = """
-?start: sum
-
-?sum: product
-| sum "+" product   -> add
-| sum "-" product   -> sub
-
-?product: atom
-| product "*" atom  -> mul
-| product "/" atom  -> div
-
-?atom: NUMBER           -> number
-| "-" atom         -> neg
-| "(" sum ")"
-
-%import common.NUMBER
-%import common.WS_INLINE
-
-%ignore WS_INLINE
-"""
-output_type = CFG(arithmetic_grammar)
-
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
-model = outlines.from_vllm(openai_client)
-
-result = model("Write an addition.", output_type, extra_body={"guided_decoding_backend": "outlines"})
-print(result) # '23 + 48'
-```
-
-## Inference Arguments
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.
-
-An optional parameter of particular interest is `extra_body`, which is a dictionary containing arguments that are specific to vLLM and are not part of the standard `openai` interface. Among those, `guided_decoding_backend` allows you to select the library used by the vLLM server to control structured generation. You can use the value `outlines` to generated structured text with Outlines.
-
-See the [vLLM documentation](https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html#extra-parameters) on extra parameters for the OpenAI-compatible server for more information on inference parameters.
diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
deleted file mode 100644
index 38c476e59..000000000
--- a/docs/features/models/vllm_offline.md
+++ /dev/null
@@ -1,176 +0,0 @@
----
-title: vLLM Offline
----
-
-# vLLM Offline
-
-Outlines provides an integration with [vLLM](https://docs.vllm.ai/en/latest/) using the [vllm library](https://github.com/vllm-project/vllm). This model allows you to use vLLM in the "Offline Inference" mode, meaning that text generation happens within the model, there is no separate server. If you want to use vLLM with a server, see the [VLLM model documentation](./vllm.md).
-
-!!! Note "Installation"
-
-    You need to install the `vllm` library to be able to use the `VLLMOffline` model: `pip install vllm`. Due to a library version conflict between outlines and vllm, you MUST install `vllm` before installing `outlines`.
-
-    When installing `outlines` (after having first installed `vllm`), you may encounter the following error: `ERROR: pip's dependency resolver does not currently take into account all the packages that are installed`. You can safely ignore it.
-
-    See the [vLLM documentation](https://docs.vllm.ai/en/latest/getting_started/installation/index.html) for instructions on how to install vLLM for CPU, ROCm...
-
-## Model Initialization
-
-To load the model, you can use the `from_vllm_offline` function. The single argument of the function is a `LLM` model instance from the `vllm` library. You will then receive a `VLLMOffline` model instance you can use to generate text.
-
-Consult the [LLM class API reference](https://docs.vllm.ai/en/latest/api/vllm/index.html#vllm.LLM) for detailed information on how to create an `LLM` instance and on the various available parameters.
-
-For instance:
-
-```python
-import outlines
-from vllm import LLM
-
-# Create the model
-model = outlines.from_vllm_offline(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-```
-
-!!! Note
-
-    When initializing the `vllm.LLM` object, you can specify a `guided_decoding_backend` to choose what library will be used by vLLM to constrain the generation. Consult the [vLLM documentation](https://docs.vllm.ai/en/v0.8.2/features/structured_outputs.html) on structured output for the list of possible values.
-
-## Text Generation
-
-To generate text, you can simply call the model with a prompt.
-
-For instance:
-
-```python
-import outlines
-from vllm import LLM, SamplingParams
-
-# Create the model
-model = outlines.from_vllm_offline(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-
-# Call it to generate text
-result = model("What's the capital of Latvia?", sampling_params=SamplingParams(max_tokens=20))
-print(result) # 'Riga'
-```
-
-## Structured Generation
-
-The `VLLMOffline` model supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
-
-### Simple Type
-
-```python
-import outlines
-from vllm import LLM
-
-output_type = int
-
-model = outlines.from_vllm_offline(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("How many countries are there in the world?", output_type)
-print(result) # '200'
-```
-
-### JSON Schema
-
-```python
-import outlines
-from vllm import LLM, SamplingParams
-from typing import List
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-    age: int
-    skills: List[str]
-
-model = outlines.from_vllm_offline(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("Create a character.", output_type=Character, sampling_params=SamplingParams(frequency_penalty=1.5, max_tokens=200))
-print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
-print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
-```
-
-### Multiple Choice
-
-```python
-from typing import Literal
-import outlines
-from vllm import LLM, SamplingParams
-
-output_type = Literal["Paris", "London", "Rome", "Berlin"]
-
-model = outlines.from_vllm_offline(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("What is the capital of France?", output_type, sampling_params=SamplingParams(temperature=0))
-print(result) # 'Paris'
-```
-
-### Regex
-
-```python
-import outlines
-from vllm import LLM, SamplingParams
-from outlines.types import Regex
-
-output_type = Regex(r"\d{3}-\d{2}-\d{4}")
-
-model = outlines.from_vllm_offline(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("Generate a fake social security number.", output_type, sampling_params=SamplingParams(top_p=0.1))
-print(result) # '782-32-3789'
-```
-
-### Context-Free Grammar
-
-```python
-import outlines
-from vllm import LLM, SamplingParams
-from outlines.types import CFG
-
-arithmetic_grammar = """
-?start: sum
-
-?sum: product
-| sum "+" product   -> add
-| sum "-" product   -> sub
-
-?product: atom
-| product "*" atom  -> mul
-| product "/" atom  -> div
-
-?atom: NUMBER           -> number
-| "-" atom         -> neg
-| "(" sum ")"
-
-%import common.NUMBER
-%import common.WS_INLINE
-
-%ignore WS_INLINE
-"""
-output_type = CFG(arithmetic_grammar)
-
-model = outlines.from_vllm_offline(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-
-result = model("Write an addition.", output_type)
-print(result) # '23 + 48'
-```
-
-## Inference Arguments
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `generate` method of the `LLM` model instance. An argument of particular interest is `sampling_params`. It takes as a value a `vllm.SamplingParams` instance containing parameters such as max_tokens or temperature.
-
-See the [vLLM documentation](https://docs.vllm.ai/en/latest/api/vllm/sampling_params.html#vllm.sampling_params.SamplingParams) on sampling parameters for more information on inference parameters.
diff --git a/docs/features/utility/application.md b/docs/features/utility/application.md
deleted file mode 100644
index 4faccb10d..000000000
--- a/docs/features/utility/application.md
+++ /dev/null
@@ -1,65 +0,0 @@
----
-title: Application
----
-
-# Application
-
-The `Application` class enables you to encapsulate a prompt template and an output type into a reusable component.
-
-## Overview
-
-An `Application` combines a prompt template with an output type, creating a reusable component that can be applied to different models.
-
-Applications are useful for simplifying repeated tasks where you have a well-defined `Template` and a fixed output type, such as classification tasks or data extraction.
-
-To create an `Application` instance, initialize the class with a prompt template and an output type. You can then call the application with a model and the variables defined in your template in a dictionary.
-
-For instance:
-
-```python
-from typing import Literal
-import transformers
-from outlines import Application, Template, from_transformers
-
-# Create a template
-template_str = "Is {{ name }} a boy or a girl name?"""
-template = Template.from_string(template_str)
-
-# Create a model
-model = from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-# Create the application and call it to generate text
-application = Application(template, Literal["boy", "girl"])
-response = application(model, {"name": "Alice"}, max_new_tokens=10)
-
-print(response) # "girl"
-```
-
-Instead of providing an Outlines `Template` instance, you can provide a `Callable` that returns a string. The parameters of the callable are used as the variables of the template such that you must provide values for them in the dictionary when calling the application.
-
-For instance, we can create the same example as above using a a function instead of a template:
-
-```python
-from typing import Literal
-import transformers
-from outlines import Application, from_transformers
-
-# Create a function that will be used as a template
-def template(name: str) -> str:
-    return f"Is {name} a boy or a girl name?"
-
-# Create a model
-model = from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-# Create the application with the function template and call it to generate text
-application = Application(template, Literal["boy", "girl"])
-response = application(model, {"name": "Alice"}, max_new_tokens=10)
-
-print(response) # "girl"
-```
diff --git a/docs/features/utility/template.md b/docs/features/utility/template.md
deleted file mode 100644
index 631114f95..000000000
--- a/docs/features/utility/template.md
+++ /dev/null
@@ -1,85 +0,0 @@
----
-title: Template
----
-
-# Template
-
-Outlines templates provide a way of creating reusable prompt structures with placeholders for dynamic content.
-
-## Overview
-
-To create a `Template` instance, you can use two class methods:
-- `from_string`: Creates a template from a string containing a Jinja2 template
-- `from_file`: Creates a template from a file containing a Jinja2 template
-
-After creating a template, you can call it with the variables required by the template as keyword arguments.
-
-For instance:
-
-```python
-from outlines import Template
-
-# Create a template from a string
-template_str = """
-Hello, {{ name }}!
-The weather today is {{ weather }}.
-"""
-template = Template.from_string(template_str)
-
-# Create a template from a file, assuming the content of template_str is put into a file
-template = Template.from_file("path_to/my_file.txt")
-
-# Call the template to render the prompt
-prompt: str = template(name="Alice", weather="sunny")
-print(prompt)  # "Hello, Alice!\nThe weather today is sunny."
-```
-
-## Composite Templates
-
-Templates can be nested and composed to create complex prompt structures:
-
-```python
-from outlines import Template
-
-# Create component templates
-user_template = Template.from_string("User: {{ query }}")
-system_template = Template.from_string("System: {{ instruction }}")
-
-# Create a composite template
-chat_template = Template.from_string("""
-{{ system }}
-{{ user }}
-""")
-
-# Fill in nested templates
-prompt = chat_template(
-    system=system_template(instruction="You are a helpful assistant."),
-    user=user_template(query="What is machine learning?")
-)
-
-print(prompt)
-# System: You are a helpful assistant.
-#
-# User: What is machine learning?
-```
-
-## Custom Filters
-
-You can add custom filters to your Outlines template to extend the templating functionality. To do so, provide as second argument a dictionnary with filter names as keys and filter functions as values. The filter can then be used in your jinja2 template following the regular syntax. When rendering a prompt, the function will be applied to the associated variable.
-
-For instance:
-
-```python
-from outlines import Template
-
-def uppercase(text: str) -> str:
-    return text.upper()
-
-# Add custom filter when creating template
-template = Template.from_string(
-    "Hello {{ name | uppercase }}!",
-    filters={"uppercase": uppercase}
-)
-prompt = template(name="alice")
-print(prompt)  # "Hello ALICE!"
-```
diff --git a/docs/guide/core_concepts.md b/docs/guide/core_concepts.md
deleted file mode 100644
index 7f53983ec..000000000
--- a/docs/guide/core_concepts.md
+++ /dev/null
@@ -1,13 +0,0 @@
----
-title: Core concepts
----
-
-# Core concepts
-
-Coming soon. This will document various concepts at a high level, so users can understand Outlines before diving into specific implementations.
-
-1. Constrained decoding, tokens, and the basics of logit biasing
-2. Different ways to define output structure (regex, JSON schema, Pydantic models, context-free grammars)
-3. How finite state machines are used to guarantee output structure
-4. `Generator`, `Application`, `Template`,
-5. Prompt engineering vs. structured generation
diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
deleted file mode 100644
index 80a6c2456..000000000
--- a/docs/guide/getting_started.md
+++ /dev/null
@@ -1,363 +0,0 @@
----
-title: Getting Started
----
-
-# Getting Started
-
-## Installation
-
-We recommend using `uv` to install Outlines. You can find `uv` installation instructions [here](https://github.com/astral-sh/uv).
-
-```bash
-uv pip install 'outlines[transformers]'
-```
-
-or the classic `pip`:
-
-```bash
-pip install 'outlines[transformers]'
-```
-
-For more information, see the [installation guide](./installation).
-
-## Creating a Model
-
-Outlines contains a variety of models that wrap LLM inference engines/clients. For each of them, you need to install the model's associated library as described in the [installation guide](../installation).
-
-The full list of available models along with detailed explanation on how to use them can be found in the [models page](../features/models/index.md) of the Features section of the documentation.
-
-For a quick start, you can find below an example of how to initialize all supported models in Outlines:
-
-=== "Anthropic"
-
-    ```python
-    import outlines
-    from anthropic import Anthropic
-
-    # Create an Anthropic client
-    anthropic_client = Anthropic()
-
-    # Create an Outlines model
-    model = outlines.from_anthropic(anthropic_client, "claude-3-haiku-20240307")
-    ```
-
-=== "Dottxt"
-
-    ```python
-    import outlines
-    from dottxt.client import Dottxt
-
-    # Create an Dottxt client
-    dottxt_client = Dottxt()
-
-    # Create an Outlines model
-    model = outlines.from_dottxt(dottxt_client)
-    ```
-
-=== "Gemini"
-
-    ```python
-    import outlines
-    from google.generativeai import GenerativeModel
-
-    # Create a Gemini client
-    gemini_client = GenerativeModel()
-
-    # Create an Outlines model
-    model = outlines.from_gemini(gemini_client, "gemini-1-5-flash")
-    ```
-
-=== "llama.cpp"
-
-    ```python
-    import outlines
-    from llama_cpp import Llama
-
-    # Model to use, it will be downloaded from the HuggingFace hub
-    repo_id = "TheBloke/Llama-2-13B-chat-GGUF"
-    file_name = "llama-2-13b-chat.Q4_K_M.gguf"
-
-    # Create a Llama.cpp model
-    llama_cpp_model = Llama.from_pretrained(repo_id, file_name)
-
-    # Create an Outlines model
-    model = outlines.from_llamacpp(llama_cpp_model)
-    ```
-
-=== "mlx-lm"
-
-    ```python
-    import outlines
-    import mlx_lm
-
-    # Create an MLXLM model with the output of mlx_lm.load
-    # The model will be downloaded from the HuggingFace hub
-    model = outlines.from_mlxlm(
-        **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-    )
-    ```
-
-=== "Ollama"
-
-    ```python
-    import outlines
-    from ollama import Client
-
-    # Create an Ollama client
-    ollama_client = Client()
-
-    # Create an Outlines model, the model must be available on your system
-    model = outlines.from_ollama(ollama_client, "tinyllama")
-    ```
-
-=== "OpenAI"
-
-    ```python
-    import outlines
-    from openai import OpenAI
-
-    # Create an OpenAI client instance
-    openai_client = OpenAI()
-
-    # Create an Outlines model
-    model = outlines.from_openai(openai_client, "gpt-4o")
-    ```
-
-=== "SgLang"
-
-    ```python
-    import outlines
-    from openai import OpenAI
-
-    # You must have a separete SgLang server running
-    # Create an OpenAI client with the base URL of the SgLang server
-    openai_client = OpenAI(base_url="http://localhost:11434/v1")
-
-    # Create an Outlines model
-    model = outlines.from_sglang(openai_client)
-    ```
-
-=== "TGI"
-
-    ```python
-    # SgLang
-
-    import outlines
-    from huggingface_hub import InferenceClient
-
-    # You must have a separete TGI server running
-    # Create an InferenceClient client with the base URL of the TGI server
-    tgi_client = InferenceClient("http://localhost:8080")
-
-    # Create an Outlines model
-    model = outlines.from_tgi(tgi_client)
-    ```
-
-=== "Transformers"
-
-    ```python
-    import outlines
-    from transformers
-
-    # Define the model you want to use
-    model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
-
-    # Create a HuggingFace model and tokenizer
-    hf_model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
-    hf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
-
-    # Create an Outlines model
-    model = outlines.from_transformers(hf_model, hf_tokenizer)
-    ```
-
-=== "vLLM (offline)"
-
-    ```python
-    import outlines
-    from vllm import LLM
-
-    # Create a vLLM model
-    vllm_model = LLM("microsoft/Phi-3-mini-4k-instruct")
-
-    # Create an Outlines model
-    model = outlines.from_vllm_offline(vllm_model)
-    ```
-
-=== "vLLM (online)"
-
-    ```python
-    import outlines
-    from openai import OpenAI
-
-    # You must have a separete vLLM server running
-    # Create an OpenAI client with the base URL of the VLLM server
-    openai_client = OpenAI(base_url="http://localhost:11434/v1")
-
-    # Create an Outlines model
-    model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
-    ```
-
-## Generating Text
-
-Once you have created the Outlines model for your inference engine/client, you are already all set to generate text! Models are callable such that you can simply call them with a text prompt. For instance:
-
-```python
-model = <your_model_as_defined_above>
-
-# Call the model to generate text
-result = model("Write a short story about a cat.")
-print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
-```
-
-Most models also support streaming through the use of a `streaming` method. You can directly use with a prompt just like regular text generation. For instance:
-
-```python
-model = <your_model_as_defined_above>
-
-# Stream text
-for chunk in model.streaming("Write a short story about a cat.")
-    print(chunk) # 'In ...'
-```
-
-## Structured Generation
-
-Outlines follows a simple pattern that mirrors Python's own type system for structured output. Simply specify the desired output type as you would when using type hinting with a function, and Outlines will ensure your data matches that structure exactly.
-
-Supported output types can be organized in 5 categories:
-
-- [Basic Types](../../features/core/output_types#basic-python-types): `int`, `float`, `bool`...
-- [Multiple Choices](../../features/core/output_types#multiple-choices): using `Literal` or `Enum`
-- [JSON Schemas](../../features/core/output_types#json-schemas): using a wide range of possible objects including Pydantic models and dataclasses
-- [Regex](../../features/core/output_types#regex-patterns): through the Outlines's `Regex` object
-- [Context-free Grammarq](../../features/core/output_types#context-free-grammars): through the Outlines's `CFG` object
-
-Consult the section on [Output Types](../../features/core/output_types.md) in the features documention for more detailed information on all supported types for each output type category.
-
-In the meantime, you can find below examples of using each of the five output type categories:
-
-=== "Basic Yypes"
-
-    ```python
-    model = <your_model_as_defined_above>
-
-    # Generate an integer
-    result = model("How many countries are there in the world?", int)
-    print(result) # '200'
-    ```
-
-=== "Multiple Choice"
-
-    ```python
-    from enum import Enum
-
-    # Define our multiple choice output type
-    class PizzaOrBurger(Enum):
-        pizza = "pizza
-        burger = "burger
-
-    model = <your_model_as_defined_above>
-
-    # Generate text corresponding to either of the choices defined above
-    result = model("What do you want to eat, a pizza or a burger?", PizzaOrBurger)
-    print(result) # 'pizza'
-    ```
-
-=== "JSON Schemas"
-
-    ```python
-    from datetime import date
-    from typing import Dict, List, Union
-    from pydantic import BaseModel
-
-    model = <your_model_as_defined_above>
-
-    # Define the class we will use as an output type
-    class Character(BaseModel):
-        name: str
-        birth_date: date
-        skills: Union[Dict, List[str]]
-
-    # Generate a character
-    result = model("Create a character", Character)
-    print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
-    print(Characte.model_validate_json(result);) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
-    ```
-
-=== "Regex"
-
-    ```python
-    from outlines.types import Regex
-
-    model = <your_model_as_defined_above>
-
-    # Define our regex for a 3 digit number
-    output_type = Regex(r"[0-9]{3}")
-
-    # Generate the number
-    result = model("Write a 3 digit number", output_type)
-    print(result) # '236'
-    ```
-
-=== "Context-free Grammars"
-
-    ```python
-    from outlines.types import CFG
-
-    model = <your_model_as_defined_above>
-
-    # Define your Lark grammar as string
-    arithmetic_grammar = """
-        ?start: sum
-
-        ?sum: product
-            | sum "+" product   -> add
-            | sum "-" product   -> sub
-
-        ?product: atom
-            | product "*" atom  -> mul
-            | product "/" atom  -> div
-
-        ?atom: NUMBER           -> number
-            | "-" atom         -> neg
-            | "(" sum ")"
-
-        %import common.NUMBER
-        %import common.WS_INLINE
-
-        %ignore WS_INLINE
-    """
-
-    # Generate an arithmetic operation
-    result = model("Write an arithmetic operation", CFG(grammar_string))
-    print(result) # '2 + 3'
-    ```
-
-It's important to note that not all output types are available for all models due to limitations in the underline inference libraries. The [Models](../features/models/index.md) section of the features documentation includes a features matrix to easily visualize output type availabilities.
-
-## Generators
-
-Generators are an important type of objects in Outlines that are used to encapsulate a model and an output type. After having created a generator, you can call it using a similar interface to a model and it will generate text conforming to the output type you initially provided.
-
-This feature is useful if you want to generate text several times for given model and output type. Not only does it prevent having to include the same output type at each call, but it also allows us to compile the output type only once instead of doing it at each generation (which is important for local models as this operation can be expensive).
-
-For instance:
-
-```python
-from typing import Literal
-from outlines import Generator
-
-model = <your_model_as_defined_above>
-
-# Create a generator
-generator = Generator(model, Literal["pizza", "burger"])
-
-# Call it as you would call a model
-result = generator("What do you want to eat, a pizza or a burger?")
-print(result) # pizza
-```
-
-You can find more information on generators in the dedicated page on [Generators](../features/core/generator.md) in the features documentation.
-
-## Other features
-
-On top of more detailed explanation on the concepts already discussed here, the [Features](../features/index.md) section of the documentation contains information on additional Outlines features such as applications, prompt templates, the regex DSL...
diff --git a/docs/guide/installation.md b/docs/guide/installation.md
deleted file mode 100644
index c626841fd..000000000
--- a/docs/guide/installation.md
+++ /dev/null
@@ -1,73 +0,0 @@
----
-title: Installation
----
-
-# Installation
-
-## Dependency Management
-
-We recommend using modern Python packaging tools such as `uv` for managing python dependencies.
-
-### uv (Recommended)
-
-```bash
-# Install uv
-curl -LsSf https://astral.sh/uv/install.sh | sh
-
-# Create a virtual environment and install Outlines
-uv venv
-source .venv/bin/activate
-uv pip install outlines
-```
-
-or with basic pip:
-
-```bash
-pip install outlines
-```
-
-
-
-
-
-## Optional Dependencies
-
-To use Outlines models, you need to install the Python libraries for the associated inference engines/clients. Such libraries are not part of the general installation as you should only install the libraries needed for the specific models you want to use.
-
-Outlines models with the installation of their associated additional depencies:
-
-- [Anthropic](features/models/anthropic.md): `pip install anthropic`
-- [Dottxt](features/models/dottxt.md): `pip install dottxt`
-- [Gemini](features/models/gemini.md): `pip install google-generativeai`
-- [Llamacpp](features/models/llamacpp.md): `pip install llama-cpp-python`
-- [Mlx-lm](features/models/mlxlm.md): `pip install mlx mlx-lm`
-- [Ollama](features/models/ollama.md): `pip install ollama` (after having downloaded Ollama in your system)
-- [OpenAI](features/models/openai.md): `pip install openai`
-- [SGLang](features/models/sglang.md): `pip install openai`
-- [TGI](features/models/tgi.md): `pip install huggingface_hub`
-- [Transformers](features/models/transformers.md): `pip install transformers`
-- [TransformersMultiModal](features/models/transformers_multimodal.md): `pip install transformers`
-- [vLLM (online server)](features/models/vllm.md): `pip install openai`
-- [vLLM (offline)](features/models/vllm_offline.md): `pip install vllm`
-
-If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
-
-!!! warning "Hardware Requirements"
-
-    If you are using a local model, your model may require specific hardware. Please check the documentation for these libraries.
-
-    Some libraries like `vllm` and `llama-cpp-python` require specific hardware, such as a compatible GPU. `mlx-lm` on its side is designed for Apple Silicon, so it may not be appropriate for your use case if you are on a different platform.
-
-## Bleeding Edge
-
-You can install the latest version of Outlines from the repository's `main` branch:
-
-```sh
-pip install git+https://github.com/dottxt-ai/outlines.git@main
-```
-
-This can be useful, for instance, when a fix has been merged but not yet released.
-
-## Installing for Development
-
-See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs/guide/migration.md b/docs/guide/migration.md
deleted file mode 100644
index df766d8c8..000000000
--- a/docs/guide/migration.md
+++ /dev/null
@@ -1,246 +0,0 @@
-# Outlines 1.0 migration guide
-
-Outlines 1.0 introduces some breaking changes that affect the way you use the library. You are likely concerned by all of the following sections, so please read this document carefully until the end.
-
-This guide will help you migrate your code to the new version.
-
-All previous functionality will be supported until Outlines version 1.1.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
-
-## Removed or modified features
-- [Generate functions](#generate-functions)
-- [Models](#models)
-- [Samplers](#samplers)
-- [Functions](#functions)
-- [Text generation return types](#text-generation-return-types)
-- [Inference arguments](#inference-arguments)
-
-### Generate functions
-
-The whole `generate` module has been removed. That includes the functions `generate.cfg`, `generate.choice`, `generate.format`,`generate.fsm`, `generate.json`, `generate.regex` and `generate.text`.
-
-You should replace these functions by the [`Generator`](../features/core/generator.md) object along with the right output type as an argument (on top of the model). The output type can either be a python type or be an object from the `outlines.types` module. You can find more information about the output types in the [Output Types](../features/core/output_types.md) section of the features documentation.
-
-Associated v1 output types for each deprecated function:
-- `generate.cfg` -> `outlines.types.CFG`
-- `generate.choice` -> `typing.Literal` or `typing.Union`
-- `generate.format` -> native python types (`str`, `int` etc.)
-- `generate.fsm` -> `outlines.types.FSM`
-- `generate.json` -> `pydantic.BaseModel`, `typing.TypedDict`, `dataclasses.dataclass`, `genson.schema.SchemaBuilder` or `outlines.types.JsonSchema`
-- `generate.regex` -> `outlines.types.Regex`
-- `generate.text` -> no output type (`None`)
-
-For instance, instead of:
-
-```python
-from outlines import generate
-
-model = ...
-generator = generate.choice(model, ["foo", "bar"])
-```
-
-You should now use:
-
-```python
-from typing import Literal
-from outlines import Generator
-
-model = ...
-generator = Generator(model, Literal["foo", "bar"])
-```
-
-### Models
-
-The model classes found in the `outlines.models` module are maintained but there are a few important changes to be aware of.
-
-The functions used to created a model have been replaced by equivalent functions named with a `from_` prefix. The function `outlines.models.transformers` has been replaced by `outlines.from_transformers` for instance. On top of this change of name, the arguments have been modified. You should refer to the [models documentation](../features/models/index.md) for more details, but the overall idea is that you now need to provide a model/client instance from the inference library the Outlines model is wrapping.
-
-For instance, instead of:
-
-```python
-from outlines import models
-
-model = models.llamacpp(
-    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-)
-```
-
-You should now do:
-
-```python
-from llama_cpp import Llama
-from outlines import from_llamacpp
-
-llamacpp_model = Llama.from_pretrained(
-    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-)
-model = from_llamacpp(llamacpp_model)
-```
-
-The `load_lora` methods that are present on the `VLLM` and `LlamaCpp` models have been removed. You should now handle lora loading through the `Llama` instance in the case of the `LlamaCpp` model or provide it as a keyword argument when calling the model in the case of the `VLLM` model.
-
-For instance, instead of:
-
-```python
-from outlines import from_vllm
-from vllm import LLM
-
-model = from_vllm(
-    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
-)
-model.load_lora("path/to/lora/file")
-
-response = model("foo")
-```
-
-You should now do:
-
-```python
-from outlines import from_vllm
-from vllm import LLM
-from vllm.lora.request import LoRARequest
-
-model = from_vllm(
-    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
-)
-lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
-
-response = model("foo", lora_request=lora_request)
-```
-
-The `ExLlamav2` model has been removed without replacement. This inference library is not fully compatible with Outlines, so we decided to remove it. You can still use it until final deprecation, but we recommend you to migrate to a different inference library right now.
-
-### Samplers
-
-The `outlines.samplers` module has been removed without replacement. You should now use the arguments of the inference library model to control the sampling. Depending on the model you use, this could be done at initialization or when calling the model to generate text (so when calling the outlines model or a generator).
-
-For instance, instead of:
-
-```python
-from outlines import generate
-
-model = <transformers_model>
-
-generator = generate.text(model, samplers.beam_search(2))
-response = generator("foo")
-```
-
-You should now do:
-
-```python
-from outlines import Generator
-
-model = <transformers_model>
-
-generator = Generator(model)
-response = generator("foo", num_beams=2)
-```
-
-### Functions
-
-The `outlines.function` module has been removed. It is replaced by the `outlines.applications` module. An [`Application`](../features/utility/application.md) serves a similar purpose as a `Function`: it encapsulates a prompt template and an output type. A difference is that can `Application` is not instantiated with a model name. Instead, you should provide a model instance along with the prompt when calling it.
-
-For instance, instead of:
-
-```python
-from outlines import Function
-
-prompt_template = ...
-output_type = ...
-
-fn = Function(
-    prompt_template,
-    output_type,
-    "hf-internal-testing/tiny-random-GPTJForCausalLM",
-)
-
-result = fn("foo")
-```
-
-You should now do:
-
-```python
-from outlines import Application
-
-prompt_template = ...
-output_type = ...
-
-application = Application(
-    prompt_template,
-    output_type,
-)
-
-model = ...
-
-result = application(model, "foo")
-```
-
-### Text generation return types
-
-In the previous version of Outlines, the return type of the generators depended on the output type provided. For instance, if you passed a Pydantic model to the `generate.json` function, the return type was a Pydantic model instance. In the v1, the return type of a generator is always a `str`, the raw text generated by the model. You are responsible for parsing the text into the desired format.
-
-For instance, instead of:
-
-```python
-from pydantic import BaseModel
-from outlines import generate
-
-class Foo(BaseModel):
-    bar: str
-
-model = ...
-
-generator = generate.json(model, Foo)
-
-result = generator("foo")
-print(result.bar)
-```
-
-You should now do:
-
-```python
-from pydantic import BaseModel
-from outlines import Generator
-
-class Foo(BaseModel):
-    bar: str
-
-model = ...
-
-generator = Generator(model, Foo)
-
-result = generator("foo")
-result = Foo.model_validate_json(result) # parse the text into the Pydantic model instance
-print(result.bar)
-```
-
-The [Output Types](../features/core/output_types.md) section of the features documentation includes extensive details on available output types.
-
-### Inference arguments
-
-In the previous version of Outlines, some of the inference arguments were standardized across the models and were provided as positional arguments to the generator or through the sampling params dictionary. Additionally, various default values were added by outlines to the inference library models. This is no longer the case. You should refer to the documentation of the inference library you use to find the right arguments for your use case and pass them as keyword arguments to the outlines generator when calling it.
-
-For instance, instead of:
-
-```python
-from outlines import generate
-
-model = <transformers_model>
-
-generator = generate.text(model)
-
-result = generator("foo", 256, ".", 10) # 256 tokens, stop at "." and seed 10
-```
-
-You should now do:
-
-```python
-from outlines import Generator
-
-model = <transformers_model>
-
-generator = Generator(model)
-
-result = generator("foo", max_new_tokens=256, stop_strings=".", seed=10)
-```
diff --git a/docs/guide/selecting_an_inference_backend.md b/docs/guide/selecting_an_inference_backend.md
deleted file mode 100644
index 2d52941fd..000000000
--- a/docs/guide/selecting_an_inference_backend.md
+++ /dev/null
@@ -1,5 +0,0 @@
-This guide should provide a general overview of the available models in the [API reference](/api/models/).
-
-## Models
-
--  [Anthropic](/api/models/anthropic)
diff --git a/docs/index.md b/docs/index.md
index 1ae1999bb..3692d32c9 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,162 +1,8 @@
 ---
-title: Welcome to Outlines!
+title: Outlines
+template: home.html # Note that this is managed in overrides/home.html
 hide:
   - navigation
+  - toc
+  - feedback
 ---
-
-#
-
-<figure markdown>
-![](/assets/images/logo-light-mode.svg#only-light){ width="500" }
-![](/assets/images/logo-dark-mode.svg#only-dark){ width="500" }
-</figure>
-
-LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
-
-Outlines guarantees structured outputs during generation — directly from any LLM.
-
-- **Works with any model** - Same code runs across OpenAI, Ollama, vLLM, and more
-- **Simple integration** - Just pass your desired output type: `model(prompt, output_type)`
-- **Guaranteed valid structure** - No more parsing headaches or broken JSON
-- **Provider independence** - Switch models without changing code
-- **Rich structure definition** - Use Json Schema, regular expressions or context-free grammars
-
-<figure markdown>
-[Get Started](guide/getting_started){ .md-button .md-button--primary }
-[View Examples](examples/){ .md-button }
-[API Reference](api_reference/){ .md-button }
-[GitHub](https://github.com/dottxt-ai/outlines){ .md-button }
-</figure>
-
-## See it in action
-
-```python
-from pydantic import BaseModel
-from typing import Literal
-import outlines
-import openai
-
-class Customer(BaseModel):
-    name: str
-    urgency: Literal["high", "medium", "low"]
-    issue: str
-
-client = openai.OpenAI()
-model = outlines.from_openai(client, "gpt-4o")
-
-customer = model(
-    "Alice needs help with login issues ASAP",
-    Customer
-)
-# ✓ Always returns valid Customer object
-# ✓ No parsing, no errors, no retries
-```
-
-## Quick install
-
-```shell
-pip install outlines
-```
-
-## Features
-
-<div class="grid cards" markdown>
-
-- :material-shield-check: **Reliable** - Guaranteed schema compliance -- always valid JSON.
-- :material-puzzle: **Feature-rich** - Supports a large proportion of the JSON Schema spec, along with regex and context-free grammars.
-- :material-lightning-bolt: **Fast** - Microseconds of overhead vs seconds of retries. Compilation happens once, not every request.
-- :material-lightbulb: **Simple** - Outlines is a low-abstraction library. Write code the way you normally do with LLMs. No agent frameworks needed.
-
-</div>
-
-## Supported inference APIs, libraries & servers
-
-- [vLLM](reference/models/vllm.md)
-- [Transformers](reference/models/transformers.md)
-- [llama.cpp](reference/models/llamacpp.md)
-- [Ollama](reference/models/ollama.md)
-- [MLX-LM](reference/models/mlxlm.md)
-- [SgLang](reference/models/sglang.md)
-- [TGI](reference/models/tgi.md)
-- [OpenAI](reference/models/openai.md)
-- [Anthropic](reference/models/anthropic.md)
-- [Gemini](reference/models/gemini.md)
-
-## Who is using Outlines?
-
-Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) use Outlines. Prominent companies and organizations that use Outlines include:
-
-<div class="grid cards" markdown>
-  <div class="row"><img src="../logos/amazon.png" width="200"></div>
-  <div class="row"><img src="../logos/apple.png" width="200"></div>
-  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
-  <div class="row"><img src="../logos/canoe.png" width="200"></div>
-  <div class="row"><img src="../logos/cisco.png" width="200"></div>
-  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
-  <div class="row"><img src="../logos/databricks.png" width="200"></div>
-  <div class="row"><img src="../logos/datadog.png" width="200"></div>
-  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
-  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
-  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
-  <div class="row"><img src="../logos/harvard.png" width="200"></div>
-  <div class="row"><img src="../logos/hf.png" width="200"></div>
-  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
-  <div class="row"><img src="../logos/meta.png" width="200"></div>
-  <div class="row"><img src="../logos/mit.png" width="200"></div>
-  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
-  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
-  <div class="row"><img src="../logos/nyu.png" width="200"></div>
-  <div class="row"><img src="../logos/safran.png" width="200"></div>
-  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
-  <div class="row"><img src="../logos/shopify.png" width="200"></div>
-  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
-  <div class="row"><img src="../logos/tinder.png" width="200"></div>
-  <div class="row"><img src="../logos/upenn.png" width="200"></div>
-</div>
-
-Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
-
-Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
-
-
-## Outlines people
-
-Outlines would not be what it is today without a community of dedicated developers:
-
-<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
-  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
-</a>
-
-## About .txt
-
-Outlines is built with ❤️ by [.txt](https://dottxt.co).
-
-.txt solves the critical problem of reliable structured output generation for large language models. Our [commercially-licensed libraries][dottxt-doc] ensure 100% compliance with JSON Schema, regular expressions and context-free grammars while adding only microseconds of latency. Unlike open-source alternatives, we offer superior reliability, performance, and enterprise support.
-
-
-## Acknowledgements
-
-<div class="grid" markdown>
-
-<figure markdown>
-  <a href="https://www.normalcomputing.ai">
-  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
-  </a>
-</figure>
-
-</div>
-
-Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
-
-[discord]: https://discord.gg/R9DSu34mGd
-[aesara]: https://github.com/aesara-devs
-[blackjax]: https://github.com/blackjax-devs/blackjax
-[pythological]: https://github.com/pythological
-[hy]: https://hylang.org/
-[.txt]: https://dottxt.co
-[vllm]: https://github.com/vllm-project/vllm
-[tgi]: https://github.com/huggingface/text-generation-inference
-[lorax]: https://github.com/predibase/lorax
-[xinference]: https://github.com/xorbitsai/inference
-[sglang]: https://github.com/sgl-project/sglang/
-[dottxt-doc]: https://docs.dottxt.co
diff --git a/docs/installation.md b/docs/installation.md
new file mode 100644
index 000000000..f88bc7125
--- /dev/null
+++ b/docs/installation.md
@@ -0,0 +1,49 @@
+---
+title: Installation
+---
+
+# Installation
+
+You can install Outlines with `pip`:
+
+```sh
+pip install outlines
+```
+
+Outlines supports OpenAI, Transformers, Mamba, llama.cpp, and ExLlamaV2, but **you will need to install them manually**:
+
+```sh
+pip install openai
+pip install transformers datasets accelerate torch
+pip install llama-cpp-python
+pip install exllamav2 transformers torch
+pip install mamba_ssm transformers torch
+pip install vllm
+```
+
+If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
+
+## Optional Dependencies
+
+Outlines provides multiple optional dependency sets to support different backends and use cases. You can install them as needed using:
+
+- `pip install "outlines[vllm]"` for [vLLM](https://github.com/vllm-project/vllm), optimized for high-throughput inference.
+- `pip install "outlines[transformers]"` for [Hugging Face Transformers](https://huggingface.co/docs/transformers/index).
+- `pip install "outlines[mlx]"` for [MLX-LM](https://github.com/ml-explore/mlx-lm), optimized for Apple silicon.
+- `pip install "outlines[openai]"` to use OpenAI’s API.
+- `pip install "outlines[llamacpp]"` for [llama.cpp](https://github.com/ggerganov/llama.cpp), a lightweight LLM inference engine.
+- `pip install "outlines[exllamav2]"` for [ExLlamaV2](https://github.com/turboderp/exllamav2), optimized for NVIDIA GPUs.
+
+## Bleeding Edge
+
+You can install the latest version of Outlines from the repository's `main` branch:
+
+```sh
+pip install git+https://github.com/dottxt-ai/outlines.git@main
+```
+
+This can be useful, for instance, when a fix has been merged but not yet released.
+
+## Installing for Development
+
+See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs/logos/cohere.png b/docs/logos/cohere.png
deleted file mode 100644
index 85f6e6d9904b0d49302a2d933181279e974a7c86..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 113536
zcmeEvcRbbmAAg-FBXP_^=qR#M*UCO+Mksqz%FLEMI@hfZWfdvAtn7q{aJo?<dy|p9
zPxj{b{;2M~<#YP|9>2f8e_Z!Gbl#uW`~4cv*L;5jTvAn_AY&xkvu6**1w~o)J$v>i
zLI08L2fumH;CUAOZ?Chu!udVfb<E>?_OR`_AbU>J-FT{p{Be(t9{-lx^b<qRM217O
zWOCukwvF0Bw&onfiEZbjI-604o;-O)LgwyoMkkt(KqQGpO<cZAOH?VGmm2)pTrGNE
zuey@C+O5Dx<-~P+_imqLneEtfo1^wx&fvnI$w{?SPd2;rmbZN7Bs@s2Wx18&EwV<w
zd@S_hxXnsLLdAx@wP!B^<^vh6B5|8`6#PH#tV5F|Ng&{setRnBJTZ~_?Ky5$0tCRf
z*h1&_|85I(REl)f8E0Ylx3lgRx|)gL2}r0;YNFY+PUZ~29PbBN1eZPG{h<>s?Zs$5
zeG^DSm<UPI8^ZhB^FJm4{rVph{>OxG_Vzzb_@5^HPZR#}7ymPd|Cz)8%;A6L0IcAD
zO!)t10_HG_-`bHhr@zM)4g|+MLaC~qaqyIt@>y+1j{j9vOXH-(glc+kIYJ?N)hxZ@
zeO2x;ryHNRgp;b6*dMi5y!Z0!U7s9Yv@#1U+Q!G;kNuA*v`B_ll#<TUI+n-RY)g*H
zAjdBSpAC0n4y=~kHVAfkPCXuHlIpEc=jfHJF8bJ+Ia6@{t@(*V!t6c13Bt>M+KoYc
zJE@0hB?6H~r=h2~BWF*A{=Wli>U$5beo9uF58EeN_X%kspCrIu$<98iAfI&iL(dfr
z<3lm3)3G02l>Su>GzL*~uKlf-?}%SWco~)Gp^+F$ZB!2n*H&8ReYIpI@AS(VewjF_
za1IDnrD&-X7r!Jk20-Uo(=nYK>UiQWT4yIeKNZttrct4XoTWfIY6owOxu|_TZaUG&
zY5K>%$z0Fzudl0SDY=H#L^OsqxaLwdg4Osa-ls}A*lud{v^+P4V9Rt9WO$Q)2gJT*
zM<#Sd?$)E@%2xevEQRY;!s4j7SQtcOkBUkQt9!F^|NC$vl7&!Qs&j?YrNy+ECQkOz
z5KZ;@w%g-O(#<jgmJ!f~yN<JkTGJ%`vBSOuVvL}mW2K0!l&4+N$_VXjZO!%(<fwcF
zyZ&=Cu74~CeDJ*7Zz{W{m{cT$>eClxEM!?z`s}G(x)ou}EME^j>%|{?F)h_GnW!CY
z%b@NM6Ls~iu65Q>8cm`d9ge9BsH*wWJfcOHzr(k0;Jg>H$@7(xYi^q%ZSISCMOGSh
zp>THey6kqa4aBl=WsqfNuM6`*JE3KJvH4mj#&xA3Z2p<LsK^XyOkK%xvud9v&QXc_
zKWSvbMTDDtMPG_y5cbXK7x@ci(H6_TWf%9DpFrP2my?hrtqpV3o`a-cNj}wUbYyMp
zKz-W5gNC8bWo+t(RHAhu*blURMKnLQxWlP_I}CGt6%rp3zMQIvw{MwXWtV{-(ZkOc
zI<Y0*bPU>Jra!KY$4w%$iFsowe4f}t)IHGUoUBA;WS7C+8i(Wm<tK}N|8`2%dZbJ`
z?d+DnGUsQf?x`uG28d=t2@#Qc&rR0iW01Dpr0x*b^HHkvIG&o%hq{9rh`DK9nY**6
z9}4`!9OL3#=kb(!d_-KBxyx;`0KYjGrOdq}2^(rE#jJpOb4W?X<S<9YU2m)nq($6|
zs`;X!+KsAokd=7DV9_Y>-_4OQ>-1-1H_EUs?;1xMnWP;ygSID&#CahmV_uPyBz38#
zvq1}7qH%Kd9wYA#%!tk9V+E<Xq=itn)2~<-cX;#9CXZ4p&W1F8a;{aH%3>D%lEajj
z=jJ+g?E`e!7GS{YCgop9sn{Cg5e(DCQy=$PeH~D&wG|y0pdJ6XDVmn{lO#-ofOrFa
z8b66J_k`MQ;0hMivU205A!1+I5C|?ce@M+e^-XKY3vYC6+|GM7q}26@-Gmug)3<K|
zDJdB~0%+~<W@m$@r9@IJTy~!t-^^Bh+;(c_vo9X5@)Q~{1m|0bzJ*Tjvulw6-#m%I
z$?52nJN7G<@uDQz)xF(hV>JKF4f+N>KeiZ4w{lh1Q%AN%*k|URU#@*$S#li<UJ!a#
zBzRW&OtT<()|fJeuXQ58KXqt&pQt3ef&@D%t|ii+F%a7Nk4F&|tFtL=7$}X`tIOJ6
z@thgKRvC3ZPXYhj0HNNgGUhBKVBjZ&Oo?k8Q`+2prNE=216Ih~v`;MNoc|sW$jlC@
z7+)KfQ=g_Me%j|9Ur$qn6!e&B@EE)DW1G-pjw41*e8w+uvi7)KsLN$npTaVS@O&J5
zfBdHh8PiqG24D?qMBC|e%D*I*irp!L1V;*rIh^6KJ4W?^c1L07%l2>&3}8F3Kc9(Y
z;W;j==8L9MZP(umLGj~_NAgv`2LA>sf4h<UBBaukv>j4XzH?tN;^q8+(WxC_sJax{
zAIko1J8PFrO#&AlX;88}=_zr;f02r*R|Jf+)I+ni{_%tY;$cG|&WYq^JMO8)2-ZR&
zP}zqpoler`C($Q-AU-6o<b_I>ww&@<R`B$up9b|PQx(7#q1n!z&3vHNJ@oXDZPEOk
zQn1CSqFM-6CFvgl?R-FgB)0n~y8Q*G!|k<HU+a@LME{->o$wcU2(@u>7$A7*FWjn*
ziD%OV<>>cUS07$N)?=}?axsY<{}nb+3M&k=NdRqLZIz#Ug7EfMT%6oIbf$CQOo&Ix
z(3!p#APmx5X+&Zk#6NM8;6X_~r2P3A;MC}?Yr)mg!Lq)Sd<ARe1uvWRA?1q#%GYjp
zu^c>Q41o#zIFaWRh!0`bFlq><|K&W=B%elN{OldgeNChfn(YUxB`%@ahE*V=!8sDY
zdgD-Yhr-1BQ?Ta85|n}vN>y}*gZ$@b?7_5>c%M|*=o%0;*y44?+hwQRl7vXh14R+(
z4U7X{Zy-ht4i4_y_8V?UPNObC&CwypTTWRjlS2pnSLqUX5%G1%a>t=y_Zq&%HJ$1|
z@<mCh#!SG2wtKTag73cI#(c?mA9q-JRRL8=34E8fRt3j@%m{j1f)S!{6x-Jg+<}2e
z{miC!Nnkp_D2yrqGcmpy3Gj87r?}vw;=)HId`xwQs^{6~G)i8`@d4BLkLF=sk~6C!
z+ea`989|dHQkPv1&eO0pcmYxOmIxdLx|3jwc&B5$mG!!Fqmg!1$>c$fZ6?_-&p;77
z$$rvtnr~mK|Lou#b|4Kh09P`As%dHT5OkqU*H>0P-rbLnQ4}pbt!~({kIyFTCuAU@
zI{OND*GVbIC??M902{RX5+LAidqqFMJ<Er<S4OQ*6C>QO1iHJ*#<Zloto|=cp#x3w
zJJ@Ir&;>~@t}fck9RRD&z5+Gh8gN`REBTEh<wGqL{+tq=a=zze*V?35S}lJT%Pn-p
z-Ka3X3AF9j=4nWpwh)$-AU7dqQy>gD75KAdhl!CM5-3$mnF}X=k|{VZ3sMp37w7D$
zXJ$V0PO|BxFew07a0H^r8?p$Z>SN!0{a8ceXfiE77VBJdrhw}w+Xc@!z)*$HJgDjG
zx_#4`H5DKPW|1B^yJd>A@!*6uw1>;Arvh@rO&Gc2XpqtEx^=F<YDGCw6mq<{T77<W
zxxQ}O%rD5C5%B*4FgWGHWF7GB&uXUEtf#EPqWM3Dr6a%8JLl|pm;ao`{UnB6bf7lk
z?OPpw!}U5pIXSQoRq_?`PpbFCL%!`JDpN{)c;$(r;}=Hd&R?tc;cEStf>(NKqxWyE
zmr47!znu%RzltDg1F=tm@mw7gw2*(Z`nU%sRH8;S$PoK089Q+BlWOMBR3Wz?p^Q7v
zTGT|FEzfYE!2W6Z*h0DWb$!6c%ya!|gW@cXE~a0jVg}@pvoD0gU+#p2|2Ykt123lA
zsZKigtUx&>ge`#}*riL$5C$K$xlVhxcF>q{l}o4_(jlX@_1+8r!q8i|Kq9A#Zl3Wm
zW%?BRjZcBXx{z4tK{U(vrye*t21WczHP~2!YPN|vjS;zZ{}i=>?Jp1M#~>ipmYz%A
z=rWjC%;^Vl5L%pg<GO{oP%8*8aT7#M4J{o4iBC+T)DLE5|7G-xYJwPrbVEaR_4>W$
zd9N<*%f}EzN+3+*-lW+BR(q1hh)9*qw!D?8zlV*Kt$5g$K{@*0_;JVHgZV-!1vRz;
z_BD%qK97aO2wEmtS`EHnv-u}aOGx>iXO!3&6rHP=tw_}WY$pmB^j|)r6CNT&N8Jss
zzw57|&-NKDcbzk}K(aNM05jPtR%-?pQ~8g_U3eCQ-;!-%FD}x8T>5_i2RyY*o9)Bg
za=53KS<03t6g=n#-9lRnSua7BMvSpFGfNdz3${Ux+MovR)4$UHPhJDI4Zc722WFx0
z@`ZXU$j23Do|y);hru>Zm9tHQ9hD%u<Gm`a?o#q?x+E<OV9#c`aW?KcPQMM&Bh<iR
z>QTe4Qa60>yBpxJFQ`GtuN0mM$%h>AZct5{oziaOb@<>1RAm6)je_uR^#46jgA8Kz
zVT7i~+pY1}<ugxH3Y2c)#7Ie!+DLmrM7t1=^B5dtVik5@iK1P#@Ij?-&$P?_%+*`C
zTw<pRR*IAn<z8u@chmR5HolsIsLvj<OB)77cAC}<rxROVL++ouYSMmTCE$f9#;;a%
zyS?(s>ZDh;GX+3zB=95(kB^=QxdGWhHSKa9R4)~>uPH|$>fW#We{mH<i?>ZlW|ewj
zfw$0k^Wi+utYbj4DutLH01PKTINjx8m#*mx!nwwou8Cc;^fc@b4xi_tHy`oiL^*}j
zKLdvOZ$cI>vy)45&Dm&F%vYbu{T1-`A~NWz4mh#I=*D=or|KBIl>+&v0st#%H!Erc
z5gF~-?FnTyvDvYa$W-&KaYe_g7zHliTz*!~F-{EsaJb(MM^9fhX^QLg2qJw3D#Zdj
z^8<Sk*Qq;7Oa9m>xAX8Fa=$4nk$y2TdKW}A$YZ2?^7B{kF4xxeHolKi0!kzU^ogaI
zH;bCBK@aogW3GklP`g5fhM3pO=AWzr+Kdm^*Ku>}EgWW?$!@%6{Gi+|oDA@O5>!B0
zCD5jp?RCx?$#d_qE2qx>B7CsemQJ^t%eY`=h_9Tw#7(4IU+)Zps8}B`p|w3x6hS1$
z_2Q<})F-97oB&#2-%6bIu8>dtX%iqp<QmxAw~r~#XZ|LB#qi09Fc0Pr5D8R!uY<t3
zKLf!|lbG{~s}zM2HiMF|pC2%W#Z4n;WrFlKjgE@e9s|gl0SwY;^{6TU7xIH@_kHb?
z7E-V)DT_0et3Oi)0Qi7x<86wHNwIk!D_&f)X*EbCffg`N-7xPMDV2KyuBzpFU+6pz
zFJGZ5_UmHcwoiSXRjSBaYM<~cpNX{u&pQuA&M&x?ZlMdwZureC)!sc6KUYmVD**BU
zgY_1MpZ5+R<1fTrjbH}ePc1ooU#8?a8kPa|vhu6@sM#FiabH_H-0N1`H7Ytae`47V
z4~jKFKGHi<?=#N1Jl$Q`5_@za5Q_9jlC%@efwPc2aXQ~-^mUs0YALp(S0l*g(J!21
zyJU=c-;cXv(c|4CE~R6Pct|tG56sW6C)J9Cswf`!$a=AYg7X7^3U-?j$<+UA&;l?b
z95>2Sd$`-r$0s+nxQ%x&;sbC{9*4odB;y(?#<Ihkauh1e)($M6xc*}nzzHq0%Mh!N
zIUJWW%)6Aq@3+V#4RGNK*aoFZTpKCXiYLZaCn&aAA}Vs5vU>R|i~>+%!Uxtg>`m6*
zmU=C-%!CA7JeS}N*f`hoI)D^xq{k!<pR{s2d>|Ed)y(9O)yZEJ@D@4%pje+)Q)3MS
zaFZ7Uj@(yGumDa{((5K8P(Ue6SU@AIWy%5**T4<#-CE$o=Wtq1!se;=c1BrTlLOHj
zh|gQ^Mp_?7Tt{{YpK)z#4DX9Uo#LtaH1bpWwxbKW^i|T`&uh2I$U&$Bp@+CzvIc@^
zhV!(b=-LMD>>AdE`|z(D-h)vm`y_0=Xc6h;BttWH4&s(^)FaYxJ=0Z<%e5`u2C~IT
zpAP+;cJ7cXGDtuwy9EiG_s!!a0Hf`|Qn$X4nH_mo4UNq^RoJg-hFk39jz1msYdr84
zX`(9&w<r4+F90^U<bVLt3oma0;xe9ojmvR}UzHo=Cq!G4ax6yAYef_REX4w$%)DVR
zLb5sJ;2LBXUvu&KadWHspa0dS+0gyRhz_OFme8k6;!|XAp_hSMD#ok2fRC~_VwuGS
z<a})ciw=_PqPPbWK{lkwZ;)AAYL`5j#g0x807UD~ir)rt578aYbd9h9w;j){x63mF
zs&<%|c_xPPRdEwq0B2#qH6MlcTy(&Qij4}n<5MZ@RuY-5U`Y#+CUtnVM-CjYOKUHG
znIa|({24u9u)$%vV`$SXP#H)US3l89j||bZ;amSDeuj?nnLTvN)ym51!vPEX4F(jz
zGQeYNpUqg+gat(R`}e9=akTc>s(LQT{d;!n1cIo^f@h!O$1RWx8KWIfkW%SB1<P+=
zdIWgyi+LK*IL(zT32LG185uv9Np=Jcfc8Lin#V`T5nMTc3%3T4r0qjb8i>|TO9(C9
zsaoY!qTIV%R0$dgx1?~_Qy1$?b2JW!-Q9!PhCl#h{VtlOmIHww6vWoDX&uigP`Vz5
zkr(z|*nHjlPa7tG12oKogykt7gia%t=Xmcr5^jMS7*dxx9QP}9p-m-;gOg|n$(tA5
zax7@-I4%hTMt9nMenZrD?pv+N{ezirbL5!80dIv}@gxbea>5ZV6~UBI^I0I|BVesX
zOzioDd_<Y7iB(Q-|4*t9d=AErp(?C-o=L#jetJoMQ6Yg90R5)JJ#)5(WN1yK@%>l3
zA-aq^RAX(na{RRj>KHmf7f{J9OQ$)Rg{!)@Sm<1FPnq1*VYeLPkX2m9xo3^GhvTet
z6Nx+_IPs56xsK@JePQdp;8^>1Asf50WH+N>|LZmZe_7bRN;4L$sa*yX4+Dy4veT(g
z0yNH^02P7)tcx|*-h;cM5VnS95Fd-2s{6DpD?1d&B0Eo143WJZ`RoLu^yKM3MAtGz
zoBSp#dlz@=FhA1;nj_lV@;Ggq1PKBt9#GNFNwp?Y;&k}3ay^8RIrEJqDk$8}ec$fs
zqJ=c{R+hZ&R+QAyI$feRIgn(XnmTl%ffe!YVJkim+lh6T^}BeDY!{REB9=*#!i`H$
zOBP0hNNE}gVx|w5=@8dx6lFA8<5fg`&!F~K3%~K$ZDmTR&^e)?@zp-=kNNM9UwRHu
z?2>UMDTsH!SBqnUv9X&Xu>rf~N#Y%g3jw$*-4E=e3uu7Nd|os<jY!_rX!b8dJx?`R
z&+}>fMSvgtnL)SEPe7Hyu=9DV&2{<H;+O&uV6|~uN@E_798{b3_8h8?m#=uI^vBCH
zySGd7nbSc~Zc6`oYhUnRz`p||)E+fwz^Ienzj(jwkJN<;?DKGIhj+VmoN7f4liYGY
zmb|^aj~w7%8*ph^(VgcK1VMH%$XS7VDjh_dn0%xw;g{V)#>QgUf2q#n!@#EvDNxGG
z=Kw;jt+7TVBM(0*o?>!hDLRIVQ;&K|0f(aQ2+^_!PB%=7%C_DP@-40tk!+-bYK9E}
z*h!*SDpTIp0$L-e0YRva`)@k(J+cEVeydbbxo)#qZXt6ER}Hw#<NPuZZAEk^Ro?Fp
zr?o}w;ghuu`#~oDD=0J+<H`kjlo?n$@|N?m><j$St6=qTi8Bq&*%aK;boT*D$+d8i
zWbEj%Z+(%U>6^+SSZ%c*-8nSXxn4<^^$H8(0wVPjW!{N2MQZ)eMkVII>{AH-B$tGQ
z2?79*jyp7)MGE`jr>3SFAGzvXOt=Qn?152{BdNqC(`)T@_Ey^5A}QFs@!f|87*vc7
zUP-Lf-@L2eXLSHG2f(-5xt-(0*P94-jn*1xffV-20CyOEQR$YVRc8Y$$Jh9C%KnTc
zS~CE4+qY~ZhM?qzaC<&%K7;-5_<g;kyR>o-MwlcC%c=`htM`t&64jXaUO!}yn~pUb
z;se9sC+T%V%q79)O>ovnq)%N{Ya8^0iC(|U><oq)Y*)$Qqx^9Yv_6_uR_!_K{&z3-
za0-lPzlA=AC=}7?^)&U|@?%gpx{Q$I22GpBd(FZaZQ@hR<x`ja{r9bd&WiT3zS=Gs
zgXC5j5znGni+A5%peQWNueplw;RO#EtPx_wtbaE}tC@RMA<3}R6^Ix?+PpTi6Dw9(
z9Xj@J_N|+Ndlit2zmc*LhdpbSRWS8a+5S8GOotJo1Jm~k*4$->BJeLCp#6X`ugx2)
z$bsk~OCvE-E6Pe(JqG*Nui6QDkvj74pj9%@DISJ)gTk6gDHj<~sFcx2)Jcf}!NERT
zo^-0+vf7KtA{%PCZ^3J5;%A@1l*QSg2Gq;9SS63DNXPVA`r6BwL7iK$szrx5pdq$q
zOWrKvo5EAvNQAjb1*#|eirGnNPln#%l=l-?_q>d;IJw&`0B%nr3gcWI#-Bm_vh&F}
zO5|w*6l~qi5&<B1r(w5pcPUAcMiEXo$D}aKjy+eid~Yt_OkoEdK9ZzB5myA}Z(`S}
zj@<G7G=D#OE_rh4PkVnh#cyKI)Mp#rFuNx(UtT_e5T+ix4BT0Wam*`H0Et_zk#R0f
zVJi3l+V1DDr(Q%TM5T#(o37n4C@#~Wptei|NFVfA2|1Sky-gO(yXhh8?Uh>Z$^ysl
z4EmMSM5g`xE4cVK^V`|Qe+ChGUIb@1V@NSVER!0ZbzJTAZ`{aGoRx(IgoPMf*?Uog
zG7(U5UosqJ<aj{_Dm2^;H;%pN1$F5l$uq$t%W@To?IjGOyImtF=!ju}NEs|9_)U83
zN=OzSf&CZGC>Ii4y*BaEzQ&`UpMBsUGaTa<UZLy8=PIAPe%%{)ewip3l-tD0!sv*K
z<@<Y2Po!Yob=w7#fv?^zvNn{1!oehadqa`sfCAdwNOU0(S@Gtb1%-ZNOqfwCx$}vw
z^M{8oz;Jv_9AhBjGox$(fVQy3?cotDt`uqjV$hVp%S|x{XnLs0#(ds4ZT)NY9P$C6
zDycry&et+x(l#JmL%MJu9M&RRDe*d8k;r}ht!pi$*M&}t>=qM%o}sGRdrSf%yy2$W
z?2kayqV>13h!dXVpvrwaHLjkM_a|b-n}6z_e>*lti#%!NgSq}%+di))U&mIV8^=K`
z2MTeehi#FdU%uoMQMHKeko~p9ytCT}BuS6$F1F|Jf*67~4D3}l(zc}+dARJQ7-OyO
zp~Q_{ehqL8I0Vtr0I1ldEd0uG-4N)~D8r36y)=iTx;!paM=*i>6n`_v3{EgzY+ozi
zVkr8&cWNX|)?DWjcz|;=a-@i2a|i)-ijHQdI(z`O_Tp}#f^KO6u1r2_OA;BToMpOF
z2BM|KS}n$;cH&pl>4$_gHm)a~&26`6l-|u>5L;1IZD(DM-Y#&}RQCrZ6%e|t?~hSq
zL%mX+{;d8n!UyU22UINDtrDMk9CQSlZXd9AE~GxzFbQ~TTz+nI`3<h^W9~dpqqm8y
zSi$JsGcZ62sM4==Fgy}Aaex}})2A|7FscA&#nPsE5tLMiy9(w%Yo@)CM%CT=xwrj&
z8yGY4p+4)Dk({S836a2wmwJ17;;kTfR!bjRD8ssIs!=}kfZMd#u@|M6Gx}Nr{A11~
zkO06zrdv~=4F=T=x)8N-OPRyNX}b>!Z1p_G&$4AgP4|m_4T(!EP>9KFS((jKUt4C6
zhUhXLmZ0J)%k8reVuj@}jbfNd;iYIly>;$%*#u|6s5(9I0Q64ZC&-_Z*YA%BLs&zn
zvb@}HIplH$YTNM5<i+5MA@OXI30|V#X=yX51rOto6eRNsL_dUKGngLO*aR_)$!T$M
zul(r<y5pA;z}C@*$pB(O&!=02Rm!{C3pHQ`Q-1;ie?bpQFT)@QGci)8@FI1A3V&9P
zqY<j6>r2!AcUgVrda*7S#_x!nhC8+#$>)f9oaj9P{F)20vi10-zW{9#Eg^`um)rG~
z!9tKYWZ#1G;vjM%dy@mkfOME9x#WJ_UWha@bx&X83>%zyDgGF0I{m7Ym?iR|Dt_I}
z+U+arJQ<1!n}E|x(dml7bAS0bVZ%N!d=^HHLCZK4_ZAr6=(j@IC7<i0`6}S3!b-P6
zWQ+T%xp<fTuAji$L|CKO#W3S7KefbbO&v?8(FKh`>*+h(R4a@iA<xYXpA%<)_U9d#
zz3xTaCf9t{)u9p3zrlQfO4lDOq;P*l4HYGa#vFndbdb4E4(&c8_>I`{uYTMns%l-*
zBFo8#Vn+`n_<&mIbL3$_z5bPvkgMvvrSW}i^Wj~EIOz1CRl?O|&)YI$-1Jc(RxW`K
z7y*xd8?`$XChMV@g*IwiMK*@`r?H`b_&mwxaD0E02h@$tJ!B#N05qtH)IXef32MsZ
zF@B55oU!@&R;v#$_x=6L9Kv^-nml<44D*;xx|b!AN`TC)dXVkKehly@lq|jrTmati
zdNM~0V3H&dK%c@)gCcN>E1kw&G&Hj_{XjD{93k_t71Zywl9_uO{@LY!KDZ>}`b4a6
zn$)H-*{2Lik|rr_?YUp>3Dlu6QX#258;YpmRKl_xtvddJA$7F3W2>5*TDKHPXVKO5
zk*?`V^@OFM*(ub9faZXksmm@M+>0>ccJRu3xv&Dcom}PhW^Gc?P4gt>@2Rq(*_qQi
zF?Urq=EW-|SGcGEoxp~nE$@7sz3p<YHR43WOJFqD|6r=-{hn<zuYeh%blBn_knJRK
za0OZ{$Fb)#ClnMtIuG1sppEb~aiVVlnUy~)7CYa4S&af_<Dh}b9cor~Mk9yb^UeT~
z_DQx-{bS+RiSBS@_&q|!$^8kd#*)XsA*#%#b%7dC2OrPe_trVMuh$k^k+bB}S?P?h
zN=(L3G@Yv)T^}DCE2+P7;Sk$bDiBT7`-7UIJV!=-HPUZCybhI~{84mSx{CLlVFl{D
z{W<vp85NYdb~~}?-p8sChwD6m-E1t?iXpG9?|C!Vr8-sSk%Nd2U|`FGfrv$Rr(K&X
zcy`4NIco_`x9m3dvFjjUO9cU2d1sBlSZN{Tczpbd`4ztvq*f~PmJ|ZX$;Hm=!7|t7
zA~ay6-?ecz{=vlI*ZwTfy|LpY%rd5yFFkDoi^$-ueNCS2gEEuAV*7Hf#Qt$Ea9lN`
zNOSRHQ1bVzYSm55@sR-BZrK{R_5woT`Od*70HIS*cYUln*Ea2meIR!4Y4{V0V3^0R
z-z#|CzSwOzGVTkAE%${T5<l+nrDE8EGG(Msu44SCSkoE!)z5KrsY$~8wp;Igz|5J;
zg6Z7|W1xU*oO?|5`ZqH8CC8JJRaoz1odaOF3$(YKc&=X9dcM4=ParD08`yY{EBl%r
z1LjM>L^pfoa!$>cLoj6lki=LJAF|W$@_wau;U<0QR4Qpw^%mk2>pKlI{=gW&1H`Zj
zSd9v0H-4>~QJ%s(ALaaG1XNc6^VBG@FO1vj3`#iDj42ozLfy3<ol5Mjl-@jWLRyVY
zYwp0Mk=ergCQcLY-_qFK!{${_nf^Nau=5Nh?$9QFP?PjcOKN5+c?X=+A%F2~8=|JK
zLt|d-O(%?+!{p&5sRv0i(=APnMloV-twwwUSQi6opmmt;^se6V4k?s3KUWT@+#=Fc
zFNmkw5Xq;!k9*|YyUdCSU+xnsd<LH!T<A9`f04gB)opHp*H4{4YCyf$@0iJhd%`1<
zh;EMeb=D~xVc4_&FqH;~L!UQuX0{d0inloEe+)9Xwv&P;F9Ygo(IjItY|;xRFW5eC
zG72nmX0^}O@!bVz$Gm)_8hbd3aXlF{Z?C{=$}L<wsYKr8jPiTrm}fvw<UXc%YkePs
zDds|P8?`<*hvy^(LR6~bahgdBMb_9;X<)&d9yuw+@wlZ?W5=#9?-ZmNks%1!m=iR=
z=&K@JW(VS4Z+er^LXGJ2TD95KhodrkLs6$D(^CXt*J12vs%B^2U)&6|vr(4b1nVY3
zyaX%#LDHn{HzX}}WO4l#>|W9RMTpfgQ;9G8+S{aWevY6A(IALQP>+X8*<akWUbqku
zV;6+2Ep0APh4q*(d=a8^(F`{8{D$R+Vm{{pW$-QSd{BS0cV*?h#sSJr6<LYoAzCm#
z2dCjY2)E?7Z}S+SX^K}bKk6es&`>EJRja};C*i(UjNxUK_$NM&2*K49z!)uU6*5OJ
zv(@BvFyGl?O2{j?r(I{Yx~8+(WEij<-t%??7D05Aanrf^t7n2bJH>Tg_X#6RI01B>
zTvPf{o_n+<g{qs%czY0@)v9D*rsDm=XC^U^q`dEyGK1prJDeHL$jf{^{rlmq*^5}J
z?Rjgxt^I}8YEWz9o}XUsoO(xNCQZv#o?QR_4|rg1lS^p2|7EkiouPf$;qDTY6yUS-
zWA+*C6b{l>8?k58Vh}0tB%_I)-Xx#V@bit%;MP$GID3J=Ea!bMm<L$MHCRfemGKdZ
zyaKBhFUk-mTXX$i1nf(fN==OWgkyRsA<-69KBzza$C<V2m?r*+n71pi42I$w>Cca!
zc-gFzXXt+Gsl2HlIoc((b~Nq5A=wJk_#ks;c=v@t8(X%F6y7|?u0Q4ob{t}2`lOqc
zg;n|(W$$VsSJT590=P#)%XR69V&9wX;@2Sz9(TGnp(*Zd@tJPYyQ%5rNKxN(`vZb7
zy(0iB#pl>QOakW@=_<qHIc5Z?;(LwK#92GFN>A%tT~%K6c$e%;dIwI06ZRqE%O}JA
zd_O@m#)=2(oMhw4R%$k+N;`a6Cg04?2BjYA^)LV>ed&6uU87YynC}vVuIuZ2{KU_q
zeIYR{4j91V=Ljoz_m^v3Sa(rUST<3OJ;qT4>HFlHrHVes18nkfOWu@hkSmerKm0tY
zJOHcYr)Dh%yVNBDo%f)*bKPQO>vcwP)5UsFb@d!c7cBN!UwZp4L~6JVdp2_aE@$*@
ztJ-X-?v(>Bn~~yP>CJ1P)6&W>qW7mL0Djle7!$2OFaI(+{armUGXR~^06Ik%=bSmZ
z_$N?e2D=L(XTFUV=xw>QupNTkER`-R#_$3I|ICx_a;a4Dg70bgi6#AhwI8zp6Fz`d
zPq7{9&%%OW8VR5(luc8`<?H2Gy!Fti{qdb*5G}MW>Lk5|!gKkg_ORVC81x&=O@@_7
z`7B>;*`b?w*?p2AyqQRw<TlBZq0y0$N=_`4s~{bY?yZ2e08p}wnRnx_`-Gn?t8;0o
z1*T(HcC@q-cvMh0V}vb+gGM?vCfv9DjO4PJQB#nxr!3efJA>AHVFiv4+0|h=e1lWw
z@OI8XzOfE6aA_x3*Ck7a<>v*bfEL~Cylpwkg5evCj45{f`j;AL4n2YQbD)GQ$Zq5b
zZsCmfv_{w!Llf7vg&G6B0Mh)Jk<z+X5sC0wgx3#%Sr&etX9p4O4jSWx;(#UAK3e-#
zFNM|JLjXITpogo!X6fc=sf(=+A}zZ#%1PMuyT|a6K(Uyp?$v;&C_)s)g_k%j>1Sr|
zl2P+xICL8X9~|j9AGv%FkrPneQd~FaXgZgaDO0%K1`5cNeU7H)M+LPTb#)(Xv;u_U
ziQ)ITVm0sg+SYX`I~%X*KY0qIEXzC%I3K9N_!|41diOlZjt2EoDRO1&KF%&L5h-pR
z_JsO%8hxE6bl>ZOcsbar1T=JW$);sOL$Wp3y#0GO(<cGsIyV6n#VW(+`9z*d{lP$8
zmDFx#cHRWFxh(P$89e`KK|LSGz!h8A+A_#@y$})MWft2DpG}sHN5y860An6@)vbwP
z?xniM`3mYt9YLlJ5<1>L6({`7?7{Dqwagq)h|Juw3tVx81{+~)t050j2-=_Ip0B)u
z+#nrCn%`wsHl2)!jMuRewhzMYJx^d^3Bq4lIpmFpD}L0=4H3)mZ4h?z0NvrZ%wAfg
zn#?aZ3Vx86JQdnZDPlg-+cl^Gd=tQTOp%gw5w1#eaV}=i2pTwqiB1yuiMH}DXW%F9
zT{w6d3m8MCEyt?#V=O=~9$cni2ZO<I>bA)iTHf%yHl79fnl<H&%sL27rbV=7&m!0#
znt`cf6H54`lW0wmO{7rB+)j5aG2ng^G+57cCPJ(XSE(VcCLP5SAAr>eF7G0Q9!D@S
zeraa}w#g-%bL7RLe%Q&vcXv|X_Z}AAK1s-r2w<35Z+vRupU6wQ-^>9hMzckTmVqud
zzjRL1jb29Nz7Hn0@VO1i!eexq&3Ik2{gcgm{0q8$UDLM!oAy-&7b%#!zx33Ql?VcD
zbvR)zq@q<<(<{eUB{xN5)7ntLpzg($^3Z=PeLv4}_rDFd3@EHKXq**;4(CLijp4i7
zpdv-a>55xfS;1KxIOM~tE)hn+buAR&MwtrM*w;PJzsphrBLdvQ(fUZBs%2;8He7kU
zGGTK*Fw{=>T(i!8=PF6Qmn{RhX~#{X-RVVCSH9NtP*uNcAq45Tp9c1jRAYX?3xx22
zc?jQaHS52%fuGK@)a!l;O|@mHE0q}ZTJ9sjpCokP8;hsHep+h6Ipg1N(nv^bEeKq7
z7@SK@RDDJOR+Et{7ZbJpysc7%#sUF7o~th{g&>&fyN|We&TD5Ou+uA;<c*H^QybEY
zUy1bX^MDv@6RLaIUo2K14apKbuzd_Ae3}3ZIFSy;aMD|h<nXt*r1+fN3zEOfSZFdU
zL_KhnUfu9A0g_-$s;>crJ&tZ7X(iv#!{|ZY_sU`!ABy|Lk<Q2PS^)0Mu-dgu^zFS3
zFc=I9Kb?k-2yO_{+45(y5nyQ`I!{keY!eq(YBCjZXHhQ`98$NhNvluz%u;M^xU_EZ
zTMC%D6v)C&*h8ac!OYquzVf^&@Q%Yr^R-OfofhgrK5fgF4j)yLEIh;U1EAAcygKM<
zR}QFcrX*(-+3MVUt`{J0wfEq!w+6nW&@#hH257UCDw6O5^?u|33S_szoAua@g;#W%
z(AYW*PdxSd>p+)KHMplFx>Vw?-fng<Y7t^Ko&iQzeY}!3M3V3Bfq}T4v(Y^ta{03L
z<Td3nkTuTj??M5eQF=_yu<g;+Kv-bBh4v#!@>VhP+$vl%JFJljxhQ%`FFgI%@spGo
z_zi;D`>mwQ(d*~LfKvA2qKLs1-J;@|;*s-Y1Zb3#dKZD(yh1s}^|<dtCyvs4<Adjz
z;o%w#`^{L#<PHA?#+@&ueFhPiQ{Q$}D(e$wAGct^SA+?fkPVs7zp0$GU$HbfWbvF1
zFkSoN^)LH~Pjv}-zD1wd<U0b#a<Gf92?#F9k~5Mb%Fln$OJR-6+_I9rpOc2Ig^?3j
z)X)wPObE38dDcA@-yPLKv)wdAkq4hR{m0Etks*n%l*r4+2>3A~mWyn1aR$$8O;mpn
zOa}+dEE83b@M|T$v3oFdCk%N>;#~|lX{L8aXfzkNB4ha)gj(suI%}!RLRBjJRkNCC
zTQ~;@V&EoHZ;!8EKNcFV_wq;DIqdV(0?L0TpBD-%d;;As)G!fRsZN$u#ZF$ZD$jLS
zWT6KbDdp1>nCbe`f&6!OL$;h{V;Y*|@4;3SpzDv)UC&#$6F%*^9&f~^wZzacumQ?!
z(yXG}SMPQ6!sTILuYZsyEtCX}zh;Pdk}#D=4pry$AN#p`Z~S#wB7bB2d9!|pIZOw4
z$DZ6x61BHa9UKN1LL!!D4C1`qtrs}iQ@0!E<eA{rEnx~_0w9Pp^61k~^MIK5I$xcX
zpx)*OT1Qucd)$j?%;b&rF|F^de|8c~Go{BY_s+G*&?J_qD+%z~=Myj<00cwc7tGAJ
z_^*U(fKq3jm%YaX=rx^}Jy`gr5xUI>XG1S|LkCua#@7r?e=M?^2urIe;~T|L7Uq8h
z_mN<iIY?ueXQP0Jjkr|J1-#v8X`$dEh&{zn90pz2&cMPCH2C}7VbO&<fjGC?1J4uZ
z3EiTPsH*9^X6H~D%?q!c(B06g-10LT<%chd0_7ZjvrMNZ`Nw4+71dd1o?g?POV}{S
z1DEBbg{&7p{LS@^k7du&+;*lxto*l`krtVj{&Mi=L-)TWo=d6su^KG%S%1V9p#~UQ
zCT=oYA@eY?UM;Nm%>V^G%oyB`CEca6OxYOP*F3+?r2cIxY@SX{LPD>U4^?M0yPq6h
zpF!V>SUv2N<fZ6!XT@QVBt_usbOdL?_@<Xrg-Nz$E}@;85`)?rIlaxiU(uh*j~1Q)
zg{eQ4Z!~yNO|y@MtE}8;v1PYUUh$ji@p2#5G2-8<?|R~@kI|sDztvBP)T9b-*jFnO
zA%i3`c)2gx!qSxt%f@DWh#RGDd1TmH?%btwvS?6~J<Jt5b@8t9TclX$lfQ9axQ98;
zTb(1z&rgn3HP7i2kV}ZP;`}&uysxIlZ{5hIv@fTCg(coMu4hL^80<!%uiK0ZbuG>5
zQ0v}U)|a!6u>tXxa^QZo{@slZ=u#pa)R>MVdVC8-3Kr^p^ceQ;ODq07MCEl~Mde1F
zXt_Jqpo>7fOu|exwp>zrhW0zTKHtIGXoJcVUWYcsCO8{lNdcrdMsC=PUz!mflC;&?
z3#gG9HtEYVPv)zAL~g%)WFfZelF&ET$;cimV(hyZChj(4c;lj(oIUO7&^anQj!PE{
z`0q1^R1nDKh_q!)L+5<l)ard)g_d4fcn}{<s`e7GiMuh?@}W<>aLLcEmeQ3#)FP<B
zMj4E+zY%Y*dPRkj<glX0fj)+lVuLvWiuLCR5>#dzT`0Q!q_eZNtFLSrY!g^<rbjxB
znuOzEgj?@d6XdLP+lIpdcs)m(5vdQQ*1C=MHa-43u=k`X+wif^Cr`H>*$!eJuZ3@J
zgG+eUWJ!vCf_<@@HLViwdWV4maQDS#DnGmW(&uv0Szn_=ZYSZvL-0a>#HJmdScN+?
z!$i?z-vQsFLN}Y#MkVItz$Fk^A(|Ett5-yil$2UzE{QJjN8U0~jF^;aJE78ORgOY!
zUHc8Oi)`hIeX(BW-zkZdWn(m(<8dRkSpyFwBjqcMqE(;5POpF&`T*SedQb1W_Mzi<
z!{u@4bU~e5sdMsbw(*bk9pA3S%xJ>HsbvB#b*-m&lc}SmF@hqA(PtAWm%8~1+tK2%
ze-R!oOs9#;+VD%hPSe58>PLuxvN(`WhJWc$2@xBqqEPUYL<Q@nk^~dTS4Kz=0(y$`
z)*>aj7)^hC`QxaSo~qTG6dafS^Rkxz3Zm^?|DDRonKu@bw{^1KH3l1U0+YD;C;z)#
zu+uMC_e)1!RuPCDYmtX9_<NZMmTr>QT4knTG-dj59cdw(m7lGB-aK7wg{@XWg?r24
zg-|1bJXZY=B*DgJfS%BP{u;;1YwP@|O4f)IA@Fl5Yz@aTE?3LN*XuZh>^yb$rQIpJ
z^{lx}5H-`=ux1TgF9J6Z$Sy9Hl+LWlJFMxG9ahXrcD+z1=4(BG^pEM&d2seBtbW}>
zzaw%Qvfy<Y?QMC{6_jnm){2|t=jgT%q*b#wy%_;7xrBoVu5DjG&zYPWD!Dra$kU<v
zOruRPPE!IQk^@y4_@!EGk(t$bQUZ1xL=8svi%$1=c_b={yvf99zGK0SwN%ik*V)_a
z+TJ8UQx=4i*U5W&?<vt<2-l(qG_Uu+v^BZ-3b=^4ORo&)2plkx$haB5w3To<h{?DO
zoqmq8|1Xv3nIgA$iwCxxeG~+rze<NUzDs{#m{(cpi=FFv^>`h%FXPbGbdEKzZk}oN
z^&E27=yduB+cw#YsHIj6tB=LkNv7##MfX57m!-R`%K7Cao}G)(BB0MsWHEjCCDl_F
zf@#_^fRvD?a*;*u_!((xrr<Xy)MpP;yuxu<+Y`LNCiljotlP`wp_oW%#ur%`V|1gH
zG`I`e5L#?a0Q$g){nSMLg7523$m>kmw=gXZlFB9Bx~ZEp%(EHE6ekI!dV-2<o8`W4
zOXn!KKd+?kSEq1Cd{Ddi*~;PWMi%z4721f7fb=#H@zZWz%Y1$pN$;L8WfS84NR1ch
zda|-eyP9<(_$2&P1Bsw=KMz36)F`djowfi~9Jb|BC7F=Z`!%)T#v*~Wryp~;Uf#X_
z#acGP=`doy`nyZKvn-V<DheO&Oy=CjvMV!dKrb|b(G|-_xH8|u(LMu_u{w$*Yv0bq
zhl8mp;pW!H8j|oUo<Bm*xOrt+%RO3h0PuB4AqQb|Pj%fE++IXnmV*xiF(V`^w>6z-
z*50S=R~IV#x^CO$GIXYIvwrcY(O^_M?2OSZ^deFAq??ztW@!g?jwVo!WG6aZ5i!v-
zVd}lwnCB=0GR`9^>%lJtTWJrQ7Yc=@&u%h2oEDJ_Ol>cEW(~^%wo57)djKM=Y343z
zIw(`N0_uXmOfIo$9@M}ox&WSEUi$GPz!6mFp9E6+5Uru3u`eyE43n$TP5Cjas3>Kz
z1@7tqh)k@>y$kxLdkVVfl1zeu_BrKqFW5F(ZClC4=sm4^M&LkZL`}(?GgFd`8aj=q
zL(|9EPl`=bDpdGe>#Ok*5Ya0f18`@h=|<<)hCx)pr;vf_lbX7qRx~-V<rArOJexq}
zr9lw9GikDZVWv61OJ&A}P3v7Zzu2^eTm_T0uF44l#O~r!0WtoT#6Wsw2owX=E6cbz
zx=R($80U5j5Ioe}hxMM<2ZU3ypKaaI?CHgA3fqd!J*H?kIf_(h%ZipFAS(vUo07g*
zrYvP9$wPeo4C;4<(_0u4V`p6h1r5>~{w6@KnW(oNzr@m+-?VeAyl&R%{B2=S8mzTW
zBY<a%)MG#?0z9>vMwWn{TQ5lks;&zl9pj@~)$hWWL9CC5veyHzA|{lcRDBFuv>+QS
zK3x%SO(E?4T0TN^cRat7c!-e=t$f(Jh!>gERY~<HKdjg~cKyEVdp`lWwY!X3>bD6{
z14B%2T(1-_$1rMZj0^9O6AV=KQkIp-B}_}7Y)EIJ#5uM7Wt92>I-pPW?-b=MHVlyC
zkJmJt;BF)Ue3^hqIRH{~Vy?~O{QzNvJdPx$21}nUXo`z!xePb|3@zggox?Auj&2p#
z(cLV0eG8+hVqDRoxxOm;fyYK&0e*gr>Ul0Yc;o)<na|PIhL;Nrfc~})U7*#NCTIV^
z{+m^WIL-3z$+nVbt|HK}-g=##t)LMx_86tmk*j#m*m9Xb`fYvcwBl-RN$TjS=N>AH
zcEJ9%{6`%OEe3Q9!upK-xw&Cb6G#ockEn@_MXWq|`7OZZ*>Vu}TYseutXrvJZz8w8
zBu5NyEta=%ij1GvpN&tOSfaJ)D3T`T_XmDlk_#JHSra74eqM)n12^4z^A`W!=NI^X
zTSykipTzCk$SJY-SK`u4X>%(w_vwIE1%c9uImwW3VY{`w7GF6hINfZa!$wO5`+m{#
zoB<(1Yeq=Cv-+ktuQ^Xxd!j)=Jb*X7VSUxPZncsw=a|%g1I&ND3P8k%01+9!AtK`&
zOrz?$57z)v==rnV_kMSDT2Q9H55%*iwa?Rsmx(scr~hsL<u4l5ju*?>B3ql!6S)m#
z;HD!1G*~`B2hXnwlxjq2>gjBMEK@4SG&XYttND9hPF5y?yAuQm)>2o^z6W-uSIeE1
zWumCHLUCWUKut~5So;gvGxz`#M2hOck@fWp#vLG1%mI-2h4YdNNAHm^R*l~-s;hqz
z|HRAoo%K#0ayuRbC#);$m)$ZQh#Lt+3s(@L$r|)27WFo|)9>4X?;4{y94$5;>BO?C
zC^vWyc1&h3f`YE<wNUZtJX<%JNs(8tFq(2XzAjE1h3ZN<IhNkJ1ZX`)__qy6y2h6b
zHr;8nzhS#CEzsDf@ySXmb|XCSJwr4+<V10W0;pG%9WrmTx3d9jdN(aCHXhK9m(d0B
zLvNh<R~i1X8bOWzIp0FJLUV0<SFnVvcYBtG#*OjwF~5>O8|4V(5FF$~E2e%a%MBd8
z%8ju<W&vmf3raAW#P(Z7X|oUYMQw9TU7)vh!K5z<orqCTI(QhZ((q|%zdGHU<;Jvt
z$fOt(&9`GH0+N16tiIO0wJ?Pjev**ZLy^SoJ`aFV5;=aE?*6Gi2pmqIExs3DSNg1x
zh=R|aO{?Hu-<#K3hA&6EuxHb)8;1#mz2@Xe%xX{K$76eL+%|tD5}Mvo%8)8Kv-iwl
z@64;YmGG<fBEr-?S2h-=)-MaH0h7GyW_+<Ao<+1*lc1a3gV8no?AK0dSK6O(G72*O
zTyg#sjx7$St=Pay`bc;wlT3l2b&FX=rVoSBtUOgVb{`+8XhwIB?-V=}s8)1A@ELzN
ziVGs(1tJE#Y(*pC?+Y5Seel?P(#xe{VQK9LRHDZl1VP@i8(z0j_h4r{09`v;kX}TA
z^nz7E0AMY*3tld@CjdK6CP%;PeJ9@)!rP54VqRbrjKb{wXN|=`CU)b^Zx($%*`PG8
z<Em1>#S?4}GD@wn-g|W`UZ|!-ZdyG8WdJsZMBG)oo_T+P?y~!dPk?)*D_(vk;Co$G
zEaiNd93gQYAX-`~DNVd2r~{;lax=NXSnhR0F;bggc!uszLBx-_c~PR93_(^M_<3_|
zs6tlB!$58z@B)0Vn#hj(1&BE%kr$jHt~o7Nwp1Jmn~x_P1ETv6Ah=>X<KuitAB;c@
z7@D~U1r?83bcV_*2*mfSBtzk<dXM8ILw?W!)rA0Tk>782R>uFr@5T1$sa~hI7NI*l
zAag7CaLy>YR$qp-)H;@Z9}Z4{x%YD*)`8aula1m73Wr%|m0Zj%57p07(_%}<5?V8Y
z;F9_rKXde6awjC8qV3#|v)d0fy7UQV5_>QPBuRcVZWew=mGdCxIc$UEh+mb<HXE&p
zgvXs*xLRT-i5%%>FS<wOAdu7Q-}=AhMtc#3REnHSC3*OYrW;VX^0Qx~mERwz8J*wk
zESmJ7{=MZ6pP69gTtNe%K8R%#m#)`{GXxSaK!kS1wY*z=KRF=rpd2$OKD|zflDa(i
zcmO{80;eZU!Y+Su^9qp%nR)t<y{U`!04djKVWdQgJ0UJ>0~ZzvUh@!#xdWoTkA3LI
zBij$L3xN+l&_ZukgIo4li7nZEW5dOsi4h3qgx@{@x~%L?%}Oj9&`bW{;N5`-!-Vp^
zp;qWd{9{e(raf}DT=_R^$T2Qf<>FqHyhMjUNExwlls0xr2Q^@KB7_U}Y9}CF^!j^{
z@q!AJJ6D<hi74BD#{)LRD)HvbXP(CIi?Rh~aoR&6kABNGu7R-mp5FC_Z)*;Jt5&Q3
zPCDeg;H9mbSq3>W-^FtyeRCt`UHuVdUo`?U#bpsmf!x>^Lyp0d9;Y3q19Y~JDQuW~
z&?w0hz>h<y&_UL90m*IO?!lEZAk+X>nMN!8eeB)Hv(J8g#o!MS_iKauK@Ueu@<tPm
zeU$_nM_qpVYWik;*kBytvIkXB1ezzc1RwRJ^dJRTGgpq$bm>}W|GVMEhT91Oog!}&
zHovmI*mXTNfOMuEmz$QqhJO6eV<yAsot89P0iTHPMI2!Gx}5mtzCWZ13wrnCgdM#Z
zhwdpcs__z1xl4Qfq^sYZeGE8op|7%|rQ%xMYl9~DtAwn_pN#j^O!?%2LgTMm642h*
zw3?PrgWAs^?P|U9ZwOIAg!0Z7L4-`?&xmdnD~`MUZh;N7AfpIk*#oJd+`5%l%vR5S
zbM``NaCV8nZfc;@<J~7yy1Y;y{C`J)2O^LNa;gp{P<q&@iV%zi1a<n^H|^>v`J%kG
z3%ft_Cn>F*4pfN_4yI`LOZ}@+mY_iBs<=vBu6QmTlps390hsl-+iZh|S}|VHX6OO6
zJPrI^G9nUY%>*Ir)%Hid2o?y|J@%Pt&}@s3ZrJ~(aAVA7(u%L!bdVi-ud(OsbEv!U
z^$M*9O*Cq>FyeF+{KXyMb^y2y#N4*Sf3zg|{0NALv`_ww&}>QA{G`%h75FM86&t{+
zaCHk`PuKmBfXGTQKLSz01s`YPIf{Gjn_uIy7>rvD5w(;jpgIS&{#`$*t|NYCfeZI<
zp6v>Qheoui9(YySvsF9%l^Xq_!S|#<!M|1m2QG~2I&q(a8XkkGif*HQ&y|ZWEAbN9
zL9yju_{6dM{IW@dzdM-#=3K_j?VeG)by|9e`Cj{g)33y?j1vwmqd32If@Gj?q_xfc
zHNy8;fnl!DO4(SxJr>K})_+m~4(&lShoK`M3O6o3E$Q_>P~^pnGmxVwnHW?RpY_Y%
z&y>v#D%G%#0`#r2h?Xel_>JCNpGmYNVOGMSCoY}0*{HwAsj_insy^Om(CnxJ0qA?g
z89L|;;#8kUg;nh;ps2vzSsS7%HB%J6pZvtV6YvxLz-uJqctDm!{(?$35ES{;!`r?_
zKNFa>L3(^(WEtP}X0h*>`QirV9OFh#PjmM&uh&!JAgHa^S-=qM%85!*rdNiarw2nS
z`!KX76Kc!8jHH-k5UufPSqGIdP3HTFD{M?N9tHP0;q~N&H-<R}l)~c*H*RYSY5|d7
z6Ok5cm69BG*ZX)ifIv-*sEj|<*~aCHdAHqhqDKJJ3N+J%*0RSg#?XbSM6_3R!(ZxR
z1Bc@Wp{^I7R|IhoXku`8E`G1dM)E#73%^5jzdR<qDZWmv-tPT5Q&AxDnnrzpdj;`n
zzYECJ%(s_=;SxVclr{7n<;Q$S2J*bNbEj_83+h9sYX|WWJ_u{V^&4H;i&+Lq<K0BI
z7f^UT9GEq?{H;mWz|CsrMTnC(Ftn>y^F6N%cjU$ZosI_J<UVQVjcp~ul5^NBU|QN5
z`%ve*TyT<v7vuoeudilo_%vFXAdzl*!)7-KR56HvUJgucx!}2K42EnFeLIS&e3GNP
zT|SVNuqh;rBSfXsr3IFIQv}A!{{RP!c%9Xi{;at!<JIdRLg8#R(J4@acQff!&wxt!
z<Gld)#<OLi7Z~rhRY7C6)P<(o<DCMGlzYJ+dI0u)wYXK!lC3~2+Z8?w!bFhsj<iWi
zpUs=D;pux@odSa9h$&*|$N37_aw+;9cnd`5>dfas#=3;9D?sOua`eB5WKyIsXz_*H
z6~N%L2!F+p-kw;W`Y%&g_#j7F3jHv`!t0)8K6=X=pkZDR$0gsBK=Wj-F!yqRwn|<;
z4LpD{$DylNMdt;ab0pswWpYqmVbgl41H9NvawL4MSOvV6J@#$YfRP(0!E~#fXz+3e
zbl&ALSF1wg@(J9JV@^TWtuB%aIG6i8?$%Z9mxqoaN%Xt3JjW%p7H`iu7y6`wo-NBu
z07;?eDBZ;OrM;ue-~H0G9W1=V_971?jaNz5JO0E4L1^nm0zobnJj^zrF)>YC<jr}b
zLNAJ|GmZmXP<p`wdVrv1(*_?P;X;YZI>3~MP*Uj)ng5q#JO&5v1P8ab1kF+%Jzy;n
zn8=b+&_t>9CSWKsv@6=6FcEfa#)1bS0u&`;>e%lT>miZ4Se~@x`U4KFwwtz~p5FG}
z<6|no)F;(<;T7w&A!bs(segt+#eTcZp|F5yZ+R$p4-<fZo(K}yF+Yqvc%4#vts97r
zkk=y&oxuWq=ecg_1^`hED0r8`hzgyrDy28C+rBC@k^cHUkcZ7d-{*s%`F3I`9=@7|
zdqM0p^yK!8p+(1fR|swyoJMJh9{g}UHn7`p!EX}wBCJlk<2D{m$3unb_6vZ+vHkXw
z9nYFs{wkl=dTBsFlZEINa}%X|#M9nWXz^g6u;|p66eQ?Fn8@BZNf2xkQV{D0RwGaK
z{;}K`)uxeg98$8^SbQZw&Qhe~HNrj!+zZUj0tvSF(Ts7;Q$YWndK5q^2j18?_Mnm)
z-r9&JxW1Bu(sc7;W{TDmz>8`ZCgg$r>N*M)M+rOE=iqg3zJ<#bYpqSs??d}5iTGaB
zxt0r$O?xn&qz6$0TfRrNa+LX;nSfpPsDf;s9p#%ucnu?IQq^dk+i09-qEdReA>^$X
z1iP&KB;__N3G-nQdH8<PTc1-=y<4|I=O$@tXNOOpHE&tVZpac|Cxom*+Ut!i#c4kj
zv(}W|p{C{11qNh#e#44@e+pcN|1J{(o8*%FoJja+`Xt}GP0s?UBcOz%^CsQHX@1#2
zuLy#U|B??Tr^i8M<5Qe5Uc5Eu*|}Cod0owbD^A0%+$JpK^i!b&BJ0hiewm&Ws~J!+
z@}(Eg_w<IKAAbI7XM~4eUN+m6Vmt%N(_TjNaC9@s1rGj(A1}~K<6q{R^YA&Kp3ie?
z2kLpn2Jr!kA-{3K#owyZ8B8W~Vo>TiMgKiACm2m?YM7eXoY|hr&T=+MH}_7Pg!)r2
zqj|-q`H-^}gn7YzG8DDh>g~4l8WgXcYIivJ6mYN&yRt}lRXXQa6A@lxw?0V&{x|8K
zjVH3BfkA(?%YPi@?lZJuKsf5vyvR1}UcGS<KLvvS!jv6d>;NghPS_wN;UMUFI64mR
zHX59xdF_e~RLpxhc{KV)IdAIOQo=elK{90%lyqQ9%wp-mgzTgInH^$ZKs}FPINVtM
zr7k?c-9pQO!MD7T66g9HMZ+^pJB++{N^H6>1gk?h3<ZDnmU(tK-@T=NvCytG8{sH-
zBNXDvDX&a{Ehf>X=-(*GpXfTZvHDM?iDZ|r-+(g4k258YnWaV5#c~KA?<ILsM`Qd<
ztme|=IUx<mJeQ@~thSYc-R%h*^)P)$$zy}%)&5Vf_#?$rTz+ILabjEjElOU5C2%qq
ztu)Iy-&`uq?<m{IxTDAiPP<MF;$scgrF951Q;*om`^EPBhv4@kwf-;V!|y@I>&4~9
zE`_ihnf;B9RdYGWFy8-c?_0AP&u<}-M&b=Z-}D>&SLR2w9Y4lxUcYzZ*FTZ+y`CR_
znQhw<gk=oZFLKaeq9KR=Rq{LZ6D|={qa6CBTSc11PobRjP#@3rzk-ZmMHqDRUt)`F
zND~8K9ra$sFjbKhg}iU0UwRI6<gv=9#K7O$=8u>FiscFw^*O6=4-u$_D}nco)f<h~
zZEBtPo9qIv4-9C6%E@2}s!3P0`2vAkWs_Xs4FFMKNa0o7kr)-zf3joWOo1{!g(zMB
zL}miFf?BD+#!qCl`CM4<gG|Aj3{2Xogbj=n&i6ADe?(KdU__7O1q!%Py8px1b;ncr
z{qee8GAf(CN^VxRmSp7?GDG&JGO}fF-G)TjBV;5aWQ6Q-E6P^(CNq0wuitsDk?QvQ
z=lkmWa-Zj%^V#G5InTS)so2;Hn{%E&A2bV=PhpRE<j$ccH_R=HZikgKTD}H6$EW@H
zLvM{pp1}QC+9P{<>;;AviM|wiBg<0ngj{!%$U1v=)F<E!_}ELD)dX6?9OQjm7RM&c
zq0e_<iA~;w@7Y>)b4Pp^Ont)Q)<o0dYXBiv`goV_A0Z|+A%LuM_eQ!-QPYlJh6#g2
ziJEo9@)8@<nN#E}?hcm=+y4QrU}c|qnnLNzx2T|z+ave=SS;BLPh8b9?k_OMvvxh~
zr7RRTS)SXL?$%b3Lp4scS{50Nh;A@wZ5s#c5*c=G1D{T$^*Ud6R+YkDcW9=p;1zPd
zWiWNL+5EI1Y8sOBpN!;@vuyMY=NO-#`RjK4?bp1GMS2PF1MF)T>#18NFcT;{Pw0{o
zdix`rXdOO(MM~$p)ISLP80oW>cWcX*r~@DjQjEWe;e*sosfS0-rh}ip|NGDl=Krv#
zJ#QbNd+xqJ)gWZw`9<ZHg$AiA4;0Yytab`HyuIKi<71;PG4S*s0-ti)Zy<Y9+QNxD
zC+7@m0TevcWj8u`oozBVs)G7g{(aITYT(Yt)yHA^y{G1Ka_Hp<fb`01eo7U7X4g>~
zN~itrvMl;6j=amV2fy~Ef|2LWi$|?BoJp7=%cW-CPs!_5IU7hHKmYG4@bM@#j|Q)x
zZfnhi{-@3c8F&={^?}dbExU#9g1`P*cSU{2J0dUX`vUaUwk&h92vJrLU+1;^Z^=4+
z)S_CNzHpRlbi)%Xtx&h{2F2@je$+*AKP#TN3MOcPnwLa^)Sc1i%s)VcC!^4vph=eW
zB=>PMzz2eT_B&HgW|s!hP2KGGjQxkoRKc&<jIdF;JAUz@aaPIcVudwBQQNF5MdYH3
z2>HK@_v{)7Ehpi@!;_<nX{w9zgm{Vov6oy%rDD^eLD&EE|2!z$*XelnxK>-{MsH|t
zO)ESZiUa!B*!ai{MF&TM_cs}WS8?igu>ilBiVPdn2FYv%2-Yp6bo!-W3s4(nR?tW0
zh5-K%Xde6%YRP^RA?nuQVn4+pwcce2H-`AxKhB<TYVAcus(r}*#eZKA4x)lkfXk)j
z_^uF5{3agA(9`q<fJ<1*wpx{bO7HcLe##TligT^eqc=oEt~4cX@}deLDOfk~mg(F=
z2Cw^+(PIHK+(%F1O*}1seX6CmV1Z8(9?lMSIoHXR(6RL&@08j<n?z_1fO4)s;&b`-
zM8CV+7oI-iy9JP!)&W=)Oatei9VtsPDCqHjdELIYMjO}T016LeGNlgyo~M?jpXcA5
zxn&fZ2js)ApnMoc4!(rL2JJNT>?PjK*q04b|L4IpbTh~YZ&6u<tucd}BHa+sgROGX
zIf-=T`TuUwiD=R~&QxsHEJ7cY=z>zGL9W|xFK7>nZ+=M`Tw=i5*?Uga7L9waVLha$
z@C4Vv`{=7sK(NVvYBeZu`h2Fv?7zE}Hz+Ge?x_Qq*`@!k+QV`oKm}^)rcK8d1&Po<
zA*nI<z?GhcTs1YtukMvoFl8J8=1N~TwKY-Huz!PO<<_`ui8c1}Ti+%4PBWPn03dY0
z>0n%OFHE;w;~WfdV!g0j&<{%GxcJ*C7tfoFWVkXA6;OkFuc2zwQA$gau3bnL`HyfX
z_lzI?)~9Rsbwv0AWH2en9Md&9xhdN)xpB}<tJWW>`v+EMi&FfaoqNI3IEg>Ffg<A~
zI9@z&>bS}sg)<vAb@V!#t;EOvflNCD@*(%BHig^s!w10?At50Ad~cItyOH@LJKfjw
z`8f5*@4u<tu5y3^U5VwuQ0vP)>yf(h>lc`1!7vrBJR4HOm$6*YS$5E2#_8pxG<J9A
zxn$AhuyDt&#C8NlOSCKm)&;rZd+CgYRhMn2)!`I5t%>#Fi>iIXC#Za#%IZMl)hGis
z20tbRS_&D<E#|XrgUjAll3Z&4CNatAp;q&}%`)G}jVTi&Kr1UitKiPm(W=-+Hy4qj
zJ2!}Iu|v?q+5K*Ri7b8ZFbBSPw|FTGNZ%)5<ETCQsV8Qu6?`|sHu4u%xGz_cV5Jv=
z_?_**kc#YNSH3j~7@>c{rHim#_Ato-tz6pW>A;>_wa09*->pT>;@8^+x@86;gV$SB
z!qKcCZ%u`b8Jnrue6|;jdk4)R{bSAaQF~{1H0l8AJ1}xbTdkGsPeh*FDhz7KQduC5
zz*@J8c?EQ(7L@tiJD`~lRV6?mr6HQBVSv0S4VHJ>u(St7Hms}}403b6<<wH@zCe!m
zHULyPSwEzG&bbpgOD7ht^SpQxD?kR700voQBLIRcK51itDFf<bnY<E+yKriGP^|a+
zG4B)U|25d<U@t^B;-u9-bolr&q{k;g_=hZ*S~?P^tl?0xMin_*FB<1e<fze!)z8M<
zf~_Pgn+qC)K;j9tvZfXcO6$RA4cLUTe>{t(@|V$GIi`#i$_%RbqYW+M0>-VpS)muT
zuDQcZfX{lb+#t+n2vC3JEb+C%9d(zma^!9}`V1A-lF-;g4X$P5Mz3)x@P!Oe<y3IT
zG4!?j+{w}~K@g&jb$kVtK=@OmHQ~!dcw}+GDDf(pstMrCdrG02`Oi@<bG{?x<>@rT
zmqM^JRSm|Q(+w_!P=IDQbJfb<Gs+zUk6U$p%9IDd-mPaRmo!`GXt6&GNkl8Ko*ObE
zgC3-BZs>MHd)^=jp^f>GohwpC7W$9t=kYE|BfdZwQ5EjD?+d`UCV;grVB@uk0ZC4g
zr%lVN(x2FGzMy8atVIkdfRR3}uc`;j@q@&H<rN8i*Tf78C^MdI`AyTVSbK&jCj3E9
zu0&5U&_$BvNy<%wS3$0NxHJE`>FzYfQ%_}_0dK4~5ulXFE9TdHFW>IRxNGHEVhu7x
zhBTm*{)6D5c~IGN5Lv&iHdlc)4+Ni51PD37xjB=l0>~;Z)b+anuK{KB->7cdeBL@o
zCzu*LTaML=gTQ!kq{6-EYvfwWEpuZo82Anjm{$>~=I&1VksVDzVxih7OoKJ1LCxX2
z@F#u%<^N7DcAbmhlNifUAj%N!{n*wdQ5rN~A;wCrV~SN}PnLkh!;}0Bz}d^vZW`Eu
zFS&9<I;3FFz0|XZqcrCBeTU^%Z(s$gSp0gp?@l`RPg&c$6uw-Eqk|A~pcCyzbSHAL
z(7G&LIq$&_th%6@hOA>LHU)r#8qbPqtqIWGYY_t_Z`ZCTR>qJ~r0o)-sWP~PUGrCP
zN@!9qr@EV4q^;j%WZ1uI`k?31zgJC0Ga8+pYg^$Mad%51bP>CcbOgail{HnhjMYa4
z*~`Y$J^WfUBqu_O+hABprYMT|()C_!B67X0eYzxv)qe=AJ`A1N9gH3p^V6MDpQY}{
zs!4siu!kq_4PVP$IxhgnQ-n}s9ZOVqBA=Gl_-|$W3kr_rz)6!PNd!2sX<=b7lLPp}
z>h|Q0!sM5slOrlc9?MwCo0kMWb_&KV&RTC$y#tpc69FOi>Kzb{wwlwrzR00{^Pii%
zkCgG(b{!n*b=$0q<fj7PSCXRz4^wyPgE~4Nom(2Nk|MUoFXTD;y;=}55xou0_QLp0
zOmdKkqSl-S2<7K24Hy1}$q#4AuqHOLvbSF}sePNyjar4~FZz7ED;E{xF6aJhH8t3p
zbsd-5^2*jiKwFXH*SBLGq<5hb-_fbH(-AVOl}Arp=S;s!L4mEJan}bulzLU*7J#i>
zvt~7sjL8QK<j!9M8m=eOmCj__g!P0yuuidtzDg*#RxUO%7Q2sW>ln{be&(BO9ro-^
zk)Ywmw%nT}WZmo&LSP76v!m2OoD5;}ZbSiWk(*VVjrNVPH`ojcauPoGWgZb-kfFa;
z+L{)A4S;RrL9F8^(!9JkOsFC%l0yA79$OtO<}{2~)M=rI-*rfKvFKiXiHD7HAcU!e
z##<yn5$y?Gf7?H%TcAlP?s%Iq2<)7qCF}%)9|J|gF-sh&>=F#vAX+6>JQho>2=P@B
zb#lZ^z<6+!zt?Fq7&KNR4xnf;YjTrEKGpHA;5Sy3punbE$!9EWeSSI<cDK7HYuh{p
zpT>eZ#mxa)Ph4dk#_+`^yPXqL=x=}GJ2@?m?fGy{zk!EI@7){Mf8uPl+YoHv0{|hQ
zbgj0?cD~j6HU!HD0^xW;nusBn&+9c^Qrq~&mhIk~%6Qwmq{Vq_AC<J0%|)#3bAUE&
z*s6ZGp+KmSv&9cxWP$-E)%IM!*D-L4_~C~y#>!GO*fv{r#?n>to#ST7mrYUA9sy<@
z0bsb91HpE~%<qV+dDWGL($C2rHRA8YMrCAQ2#l6m;-LZ;=sw|KK5@DYbzxnEX0*X~
zf0X`#Ntbg&nJCmH)`O}zsK>4hNQ&P_szD~uicGlX-1czar7{jPY;_Ri08YbUd3BdL
z41pa84#4r)ffX}u*BjEAL7x#?J!Y@kwfx+(O?5My6kGHpXO41F0-w-<W{avK6~GOO
zJN}vXnJGxdB1u-xJ$d)|uMhQN;?1_6s?o4Q0|PE$pe(3QV9<^4WHD(vN5bc5GT!2Z
zG7?d2se4Qj?C@rmODPy~t>dx~sl+3G142{9U~`lI)V|2Lt`}RIY>kAxD@(*~MNiXh
z4Sjyzy53%wJ5_WFx^2s2XN+{}LdSol459@V|EOOhE5T=dOON5MxsUgAP`M=NeFGiK
z7+lEC<dt<Mp9Pb-tfB(BpDC2EZM$y<hf|70&>q0nhG%j!p=+#;L4m|T=sBm3vRNZP
zuH<{7&Dd;>YCih5$a?MP^4E23`L^TSY`vc)f&85W7@V=M*OfNuKq0A>zMoTx^E5w(
zu76m%Pm~q&4_hiOfXS3eg-<>05TXMEx~%Oj-!<g_7)=dfnt5I^>brI!%R;+jqy&^_
z(a?aZ$AD^wmQMt=P*QN?VgoMuvt5wtzUnLLJ#kwoT`{NDi3t1JI_+Z0!+g+48)lGf
z#W}AROwyjp`o79W-~S_AMF|^{8TAl9&S4rcmjzt}K?$W&Eg&&8E*@;1>56UaAtjO3
z{6tTKO{&577zXP{x>i3yH#*q*(t&lK0}zoh8Uhi?h)DCqBzs|#4iU`=g(8<;8m={O
zL5r>|BIrYhRZ!W)Lqm1w$d^dF{;j)U?C=8Jb>$W2PHYiDMih)CfhQX%1kp?Do}uyo
zKDC2|@}OFlaF58hFb6{sH)()4z>+z|+yvhCwZBHa`iu6Q!kgv6jL4YtK)8bQlRpvi
zfpX8YmklR~ET~ej6*ic|&^&bIqZ^}9?n^3}-T+sCiU3ZRdr)+JuIY}ugS)wFw|h?d
zX{xlpM{kkcIOd)w)6HDZY_zAfx2KG2D}w|+8RWA{8~mD955LJcJ9}L=sw8y5RviH!
z`?4}UVRUqJ1{%O)*0~IQz|GPUn0egdXYcCD0<8wKPAIA=VX>IDq|16D>oNdOlFJv|
z06vvw0GtZ<K$L~4<9b<HB(8mhj<@IQ6N(zx9u5X|BNY)zL(Z<H5AL)PJVdNo*BxMR
zYET|)&Pd_nBh#Rt+cLs_3xMKHT^*72Nq`77g9ZRol2wrgV1Jl4$SsRoJD_}?w>bq^
zq$HUs!_{w&b|NZj&9wPJHRw11K?VDs&^ozRbI14c+*qM2J>67O#@HAZgZbtGh~nz%
zHC}{usy)s)4}PJW1eh!poczrm*vn|TIfZ{XQzfcp5s{bQfYnQPB3S`KV0jiIa*6jg
z5K`<=N(z<~D8>|~BhGsXOxx8x_0-I<gf2JGRRt@43V7RgMKsY>RHQ2&`+Srf6)OFD
z9oJ-i&4!s4!6;;-JE2QzeykmjG~TkUnsq;+u49IUuj7m};NFwBv?W%(I@4|M?o2y9
zwuI)KT7bnePA!+R>iLm0KqQlzoyc&n><?oF5nRq5l|nz#YTD)gQ;i>Vh!dvt7l26x
zVe^qV>UAd2Weg;_SBgva2FrbLpkk2bOP=oQ%2SkGx}epoIr(V<V3|;vP*Hfnpar4w
z?LT*I5k8ZnK34F*ttBb}pFP1d*XJ3C>JQiPPR}I-w?ZRbxuAAuJQH!;d9G3@ii(QM
z+tVNS0@n3kz<9-a&bRr=lrB&7xF4~;;|%H*F;#%#N+rdo^}vvHZlY(PM@BFQ7F&N&
z9b8vdo9a)sP~W{cdtI0r+B$@&=U!I>m-|Q$zbrgzD2L@6D`+nqE1$>y;-$GJ28+UM
zJ!Rmd+;~k;@phQX<&D$cR4eckK2KI#gs>G4`wU>Tl@A@hgW}6EPKXbdYhc&f;$UpL
z_XVBcA*`9b6zQ^Q-O4l2Im#6k(QY81g--h&Ja^Rnz~M2fKwlc~rjEzhK2d?{q*#{p
za`_B(3SL6A&~*flRnI3+wu0GnDfzgymYkT|SQJp1bLKhuD8(1e{Amm5^BML-ovOf|
zqM*qt?xiV<mZBHb?lOZvsj=RP;-b?x=IO8iyy@ock(Sg}ODG)$kV#O2>q=8*^X>lm
z=8jL7<<C~|Rj5=i7A&Y;-4)njH=Ko4)uQi=8>!Q)!*ih;m}nmok~yXluCsjYZc>u7
ziVRpbPtBSkM7O2D{yA&|z)y_cpgO>hR4ITaE8R*RPIw*Y%g-7A;Tm_N6qec4?T_F+
z{IJ=`1x<J<mv{$b_@n^{*Ymj+4o0QF&@~*&z!G2Hy^r*%7!flWEZ+dI#iE0*N+7{N
z7XZVRdcu{L=u~95gI(3ibXAxPD;A9wCarkKab4Z;xX97UMyL{Y`O$q;&>5=>sKZxQ
zzS-w}>i?WBT1JEI{-`&vkX%+y_uLH@>A5EopkB5l)WPUTQ7hF2P4#K|@(Zn8s6Z@E
z?TD7H^Pr8#)uyjnX2A2H7{O*ao+hc$QUsC_c?Ah|y>03E+rl=<Bg%bR5Rv(iBD|kS
zBnK<9LckP%Z}ssPqJ}OiBGek=){$7MO}dw1qgtsaxz2l1`y^;V3Xlf~`-Dxkabfog
z4{Y~Cpu!2jR9D>9m-!}A7OxU0nn9R^WQ3A>N3#^#)t%vM`+|#fZnqa?t+C8)qdt*~
z&-S!zI{060y2Yms%o<7w^a#}%+xSfRL0EZKv?%+Smd)R$@=YT6d}Zs&m74DEeDAS#
zIW|b^q1mNVOL?puUN?MK{M#%C^-12{Q#zJ`<tjh$+2&=@Cdr|<ka;Ztb|#h3+4dMs
zq-iN*yo0eVsnp-2qjH9*6I<66WfH+q(S}^4^BAaLp&&VGp1l5~EAhok2f4WQ@24$H
z__4@QHM#h-@ep*(^X)FsF$2Jw!Gm;FC-1y%+yrMCI&T>C{I6&dARAfi$`HjFLKoKj
z!!Id9*Nr^~wR+<+Zd&ci+s3yQ2G?^wb(R}rTavWx6mR4$;`i(qplT@i)-bMoJh(j@
zhmCJCU8JnLDp->Gug6dw_=7LJ#SN)Nnx_mX==Lm{RH_ibR+#&&C&#-@V?7{oh5wyP
zKvRA~23-H`Jcj5PNFW{6`<`>&J99f6bW_<qYA^pr*~*<j9eVi2{s?%*#1;F_MB_$D
z%r)D2!9d}}-sz#<RW4Rrc{PR>p|JsIk8%3ZH*_9O{dxN#vn+J;L@h+7&w@npL!T~1
zF5mG!acR9(?o!s@SZJgIPQgRv@p#;uH;Ubg@rmyygP2vp!tHUfchb{VvAWXtABRHz
zF1=q=&~=Y{;iJ5owaJ?mZNh|V0oLJe4tO?e+P!^}c6D&E_&VhPdaLoP`r>WEzkOaf
zQwEpSy0wY>W|o%5NJ&%n+$yBhXd;kj6nW?qf~rP{r$c`xmHKX{>X#aPB}Wy5?<bG=
zZ*cWObDI(J>aq|o^UMw&VoAT#4*O!ubgIp3{hK~l_P*3)$L+^V!CFNJdCZUV&eaQ9
z);Wf@7Qe-y-7{=}IgZy@nPY<OOsmWNy>C2`viJT23nv!2q83o8{3SBm_k47LdQMe@
zsy`A;nN^67ex%^Y`TGXi7*1Gg+{yIXHH+oy%@%vtEYA3!(5X=0fg3v7n|+R-Qejfi
zf8fJ&mbZUj0S8J{CBfF4<F0Gz?zPZB)8lQ(&CVpw=874?v#5?M=ic|jL?HL@JKbuD
zp}5&g#aA;>JJ-Z+am#R82A%bG+GWeV{X_YoT`WM!5?MeDqT@d^tplm*f_!w88Ub%E
zjK<P1d}z58AYx|b+>z8KYQ%3QJ0g7{angq5UXtg1LQV!<n@8}ncqQZ02Pr7nTaHqF
zHBPY=W~`p?>UIB=JyPRt{FtHMipwG`CA4_Vb;~Wfe%xLK#Z|vyDkWy3PIvzR3~x{v
z$j%jVg(BSO0Lj_!3N#1d#K;5t|I@%Q!+$~R$3}3dO>VA*tirv5?FYrcgKj4k3MG{W
zvW*N%*xw%@7yIo&%zFaZ`<$+!>ZVUDbf&b{>-0tsCff)@KVyD-;84=X)r8v%V+pG|
zQhetr^jV&Q4I!Xi{rXSbRkTfMp<8B7YL-?&;;ZvCB+%{%*nrW^C`nQ+PDt7e<m`Uo
zl^G<0WpR!SNy^Q<Lfdk)$}|an=wztdgaBBP<8Xc{3#|KOIY{w5QK<611-F;>b3mbF
zksN`YTcg@tv&jBVd18nrF$^%!X#W^hbi*V(k-NZIV=nv~-hN_j(@y*?8f|~=J?kBz
zA0;INPs?)RjzeccGQf|Ei}oKUC60|}eB)-bK_77J2J!^!uhZ`RtCGU9m~kElby!F%
zD{mmTl0)(WX7|kaXZ^eyjLHP0Z11>cNqgVcez^SC=I<lrakN9%%Jb()sCum@Yu&|-
zC7ez}bQMtmOuJam7?rd+uONS@m(f;4fY6T`%L!ud3yu?TO}KyJ)ah+sQ6@dnfDio&
zu?h$f>wf*6q-vw$mPDxeGo`y9G4imtb<B32lF!rzMD}*K@N)StFh7P^O3Zc;SN<Je
zn(CwXfbt6gK2Jr}uVH(XEaR}&=Ii%QZWP?);JDR4b`wKkL<m^<wxj8{V_L3Gu94Sl
zuk$mG{voOmf8+2SRPL^RFK8LcWk~(c74wUJE;N#L)9`pDuF~^`Zi8emz021y^!R<#
z5~OFlIa^EZb8R=zOT7Q_0NS-5JJ@w4UwuV(Sl7kLGNhZc9*hKzg<v7~rAib}xs*EJ
z`*Nr<`T7YAyBG-0`bqXDU5`_+RoKu9Fs3JcavTf0Na301-{YV*ENd5Ne3@Pm(_nVp
za2o8IW<)a}<jalpbb>>+p?cPrM6pOY<5#w<l!u!$VkRA-781(b9hhUqT?M?RmdSra
zV-h}QVLw_%dR^B9+STuyN7TVA@-EivkEEN7FKfN@_QO~JL^q&zgX<=ag--Y7$>U+^
zp=DXhmS&g!eEI+ij$d@!AT%iSFjcC0-p^+ELM#DB)+KR{lg9n*^HwV?c(+Lqc15Q~
zu$}PtrJ<d8CZP$ZR!24HoklE*rWSebbfDU{DYX9hz#*1XUJnRgem$!XK36?q6lTY^
zY+(AkeZU;GOrh^)XuXpl|6IqD&d#-dHB{5iJ8S{>1ZYG_-WaEC=?t&lx_Rji85SB`
zfprJCb#%AnO9y;8b4dbYV~CHx1QzOp<Glr9oCN7G`Ps23eP2mxp)y7=sQlME#J$ky
zJ|jowt<i#<p_D**uT$GQC<+9pi1LlaSLAMvHfRWlHLJLeMFO(ntD5S1|E(=ejq^u&
zDqn2amP|QSu>Z#k`U_i3a?Pr}W(lmie=+a?iX6X)v%Sdl68~DyqK3BkO^f7hwVH7U
zQ)shi$(3RX?sc3yUt0PmBJ)3pIloz0gd8<$q}JwMU8Z#_hx&?>5nTq`_K&_1>mn$g
zTZ=1sjU^bhif@i81}Cup^?aN}A4!<O!t>S1_v0h!(|ZmAG1Cgn-K;6OR^OuwGTl58
z<Gt)gdf#{qh5sTRGR6`%&bSd%m}+hGJx5o6`v?6jK>njfDA$^rZ`k(wPxRi=dGy!n
zi2L`mKOS29)*(t`g5Hr%R4$b+qCD^>DT+lQL9wVnrTvc8b2f<s*p$bEzR#(8s8v<%
zY~k9f&l!&_@}+I!`QZZ68(kjqw0mnDWtW`aOa9HF?Ajm71OB&1rq7W_OFhAIYOIAK
z9r<X7wuMZf_yW*9$w*l9jfK*npG!FEQTp40wO@%)0{CqYGiN@Q+^o+{jV>(_lJ(9!
zvVF=V5?w@a_-h}7Z<UGBRwBe6r6FZm$7|{7DE_2^5v8R>?u(0sDXH^y7lprW`*46+
zISW6YhFBufh)I4ob^Zdsa~t8i-&v251V38TWI&qIXM@=dxs`+Z6x-AXtUD5<*|$8L
z<UAzM#%`1z@Z|L0567wX;aEqL!LK}-ej+)42WbyIjJk#geC#7ulVj>J$w`}ROj_rE
zoX9`;x6&NXxi2(mo~2W-lyjx4r-x@^NOhOMo7e86$caSt;K>nVhfAc!Dk|V~cz>Q7
zk3kb{P8d(v^4_+~e<rpG6IdW)1|cWy28S<aA#DemsMRuXNwA1q3@mNsYFMqdwd<GH
zINeXmcX-pc1#EWzEzcmN?R<8=)s}qRHGntz^Z5Y8j931!fmOS=TXV@=IRkm@q3zAx
zBLzB>*eH3mrNhJHX0g7YuauK4!I(VBU#}qGn#r!=8tXJ5Y_Yd1QCgmC+oe#C3^;qe
z=pfuUd|dWkYeNMNJ@H>hFB1cSk!IGh3gw=!Jv~w0vX%imf-4PqEP>Xx++_2&r4k7W
zSk*?(6s0kACB=DmPKVt2&f+%Z4xmao(MYdE`T!X1-ODS2`ubrgm8IuiddR=sB{B%U
z%4kA(Xsxn9wEC2f{MpPMyMlIy2q@wC772Ww(8g?G9H*}-^1)yJ7MTQs6ce{gp@n^r
z*<!XmzPJ-V?(U)iQXQ>;WO|wa%fGFb-a%*ODJqpgRxX|9p3Uq2JcgJ6577De17NsM
z$J~>eh5PW|>9}>Pj+UjzXJS`1f4z!?a`}C}TgUsn`2svc2I?Ie7d?j@k9P&3B2Tlc
zMkcN373PC~DQ*lBwrP!HcW)`SZV0F2{;p=npn>cW9I2W6M0NM>-3kGX7U$?evL~g~
zY=2ta`$%cH6Qd3}ZWv(;hx($W^m#I}9U0m5f;RA66;JxgMiBjjvt5WJc1jff-_fx!
z#d2|2ZSBvc`W2<MDyk<1cuPBEMrK2y=D2|P{4to)jfGzB(Zh(pIY5Ww#*s^?Q>~=A
zW$Em!Lw7=+0J8-YIS7(LHt;2tFMZ0p^Qk}8AJU78B*>G_^)RPOGa1jjlc-{SaEEqf
z${^@4_2rEG;ApGES0CB(i`bZdJtb@7Y_;80yQd~%r*|}KnGo{S<nPGdm?v+kNNiCS
z_Q9|38iQS*2mI7VCSry+MMDL2C3lvLgAx&^I=*u^OiWCwotC->v@6drC6J5mKDE1d
z@$CD8%!FJMCl*{fS69Z@$h3m3cD6T@1K8$p_BPARhf!NGLzwf~kni}<|1ft*C%E#C
zk^QNdz_KeXk&uJ;ZX*B=-n})Bwk+qshd-r7Jof%L2a(O=YA&tTg7sMj<v6mPr;lg@
zm|tIQ9z83JgX8ML>+xy-i!7}E6{tVg_QE;Evm`Z{+Be2cL(?fgfwbKroWlu-aFp*T
z--K<>J`}9jP?y(#>rH%W_td*S(SaaOFZT(hc4nIGiloG;dg{0x@}jOm8BzAxaVpK2
zgd{f)DQkifk`sT4D>@pd$FH6H)kkkf4q76do&8S$QBBgi*&mM4mR`oBOLJH8J^$SN
zTO5}xy--|>j^LsYu9m|R4%>V80>&o!$DpYy9*V5)vTd(v+7^j~Gx=@TmOpbjL@kV|
z`#XwrRJ$qvLF-)nibM?m_6~%BtLL`ipL+HJm>bu?xE9ZHx??+=+dIQO3Wvt-^aUi<
zoO(&#snhw}{PFYwd;jqgM`8{k*R5Krn4mD1K`yoT(2i*M$sBA|Fn3EjoX-aIbj68}
z<nFFutWW|6ZE>@OEeScPM%$Z}%d>PsVw%N>_|l(=Fv0-F+0B^i-NHHb!&A+se<$rh
z9sycf-sk6e`7)~VhV7TbnT3qAE<`>)DSt>%15=-qRKoa!XIgo~t3G0l%F!h|+mjPG
z==Q0^p=Cl*1eH8&)XTHv&e9)ZipjY4#d*k=@F;&58Bx*p2TFvYCDe(1dOMi}$^n>b
zuaSIzC&L{PEM#=GGz^_v`eP5o;{IsXe4Pf@<#*Lh4WtQgc$9X25mN>^T$cl;I^Y<I
zvb+eq>Qe#PhTrXO?L>0FG?9)gqdQJCACYI-9ji|yc7<ctvLgJ!J4hYh2CZhH2Zsv+
z59LyL{>2KugN>AdoNQe6H3RQmlK5Q-=rKnm<^?L?&rb>)0+CMs=51?N?%wgtQ-4YU
z2}NFKmwS%8`C#qj<Yd050R!dEPvWG338^|J@Z8q-?p<bDg_c5@xu-@GvWCCAeUu{J
z&%CDLt7pA|2=*84lq--E5EG{=@C3+F$)9~`?^c-w80Eg|su$d0(H<K_6wpz&^Z6~k
z_Kx;zW?W(y=#%y+^Atrcd({}8Og)|u(ca%gp1}P3_K^WN1zH7n%``@iN^nZ>>}*UN
z65+v@j5qAA#uPYq#9Ij&QgIHE{=V&l@K~m=XBtV>EZvp>(M35<I_@ruvIpTx`Mf_X
z%EQB>rbW8m_<~Y=^5tD`wrhX;*=HUU>hxG#GIZ4#k-Eou&2D%7bg=%<C|Bc%)|V?!
zT?Ha&mIL+H;wVx5f1Ldrm1VbyLFi8(H}gu!_P(Zl@o4uaXdNhAZ}gW$4J5RxJVq%?
zRmG9sZ`l3wjxo7O@xy-Q2FgZ4rE_lTjRynOE`82IInqOlfz>iRV1jEew&Ofy5&0wF
zLtQ2;`q8kidC7jXcrv$$c!W-GPepg4tz?|!6XO;EqfLtRa(@`lm`AY17?vWvnw0#O
zU^bP-kh6Fi+mZ)Zi$ejL9dg=~z6I#G)z$7QF{9$RPp_kbUH;hrQ<94O?2Bg$<5J!a
zOqFHF?gg_m#~|M$AkTF+KY!{g+Dw<3n{VOxfj@+wT)vN@GQHMm_<=JOSEqW#bJY{O
z7Tz-)FyT!ko}(n4T;pZ4<OqAo{w*>_-h+SSJEZKo1v;;4+vneJiUK?QVMLmHhm8(m
z2i@i)55rMH_=-HTYrR%mLihhzAK`ZxbtN!YD!j2dr-+|u{2hFk{=_DL{_n0_`8c8x
z15$maFqzt=g11k8_W^#C2c264UbdRpt*qwMg*G&VNA1!FT?h1MQrfdLniVCnuuvH6
zJNG_-COEl2=#N^*aZqBh$7y)Ekrgb}+1Y9RN&Q*Gu8l&$p-M~ND|sVuu|0RwbHeJ0
zgA&u!J(69!#(HNtKT8x(*G7S~n&oWxIRB{|4k}}NanUEB4BYHj%w>|1H|DcT?U(o&
z^#eNmwRd;Ab0lF(hqCzU^A;T+Ay@2CM8LCS&JVy?z#D>NKP8OyfxKnBgelh>*Pcpc
zPgb^H#Gr^c(>nLnw5}k>ho7{=Uw!xe$Pa)L6wg8)2LgwrSt_13%wY3sC$HH34Z06U
z)2T9!E%TJ7-E_ac#^F+!{V!G>0f|>XFH-jmh@5Rr7$c65?{aC|2!RkrvqRTB)P`#+
zk5w?$$ZMBN<RIlx{BEkodm8qIN4A@d3U{f(M*JG<%|p9VwjD?GP2#Ld_n5!ZDlm{3
zQTN-mBM0GgXF#&<9`ZIw7ZpAn_@&UWMZ(Gcfr<R@r4sQJwDtLj){!grkRdS=jp5nd
zpPV?5qQ_18lOp(v(QZ{Y2|k%i6rM>wuK#NZ@Gn#oL0h_}o6*pG=jNT}WTp0|^KvJ5
zcgPHd{d9;dQ7d?Ic%oaAK%SC36L0EWOdW5}2*-(|7ed0+t*=^shY}Z=o7Owk67UN~
zNSEc^F8RJjDm7-;zi_#AGAr=E-S=ZCApQ;=XjPl~rbhFf@%IK5g<pZ#ZgIFS;U3i+
z<K(S02_a8sqPgFhk56S8Hf6aa)tbYh!&ynz6Y8|fSfm~()`%|N#iO)EjeTBqJnwpL
zjJNW4HHwq#vkgqcXD_;IJMMKxp)I;<@99AxI-*nJc;shL;Ryx$oDr0XHM=f;=K!Sa
zGbeb~+BL$WOn9{I*COzzBz6q|5@837&~1A?xK9AmR``>6{m(9=gk<I+I>KEBVF>6)
zhjI(;4Z2rqTH8-iD!Er_0)Ko5CB7*Gbf)AuJ{D00Dx62A{73zmhv|1i-EArL6FC@~
zBN<K=6wL9XU~0Jc%scwXT^6GlKy_QP!Kfi#55`@4#@D|7ZQ#$TfXBLbesgGu6&PnD
z4U<;OLWxAAu@QctPMF6Y9nJ$fFb<qFA~}c}Yn0`cyw%2Gw1ii78ut4+7*L@%a?WOS
zy=+cpi;|W<SFoo;Mo`ryBrw77=B^ZpHh9LGRxs@@fl(P`CeGA_75yO!i)>K#Og_`D
zWVFXB1`VJ~nf0M|jKpt@-$s}lcuQq)YwXzw%zGq=+osYOulg0{r7VOT4wUme_UrYe
zAfJH234jeyT%xJmEIuHW#qj&?r4P1}8Q3{GidqgOKbF$DVIZNtXT(N<3<vjQB|4&T
zk|PqeSeL8`ijG$QE`|4~egsWqPUZM$3}`=E|FyEa7AQQ(9)>+8%NQG8H5uf~ved{p
z|J&NcLI*>&1IDXfKHy#A)Xw2{J1YKc*N7l?U9l4!m;P%5X_G)sNwZetVe1pEzn>nT
zB2R4CdeMb3S8u2)xEOy@oMVr}s9eB+@;fLb3Wu7S`h0!JLKpMs08WqB(5|h=h%<_r
z;Afu6J>o&mDdXay56rj>zvK-09jFpT&kGu2IWsdewds=Ao&QPBp50?HI55%}t|?pT
zl9sulVwc;yXl3lAK)$EveP&SIR2Lz~fI_`etS@;T8fIhVpQDGV>F@n2rVWT;DrLvZ
zv~rqU@wwZFb%lN%;w}_a59j-_zJiHeycp!m(phpX)0Zim?)|Q?L+?RE!qQ(%jG?Wp
zm2=Y^&PT$tf2F(*XMjNzpv5Csi&=czQPtD)EmvZ7jn1Oh?iatHgOuQ$@-vbVUC9wm
zp-daAT<YKFdi_p<cU1C9g_nuz%DZlu551mHu9Wq>X&U=GKio+(Oa#BKzxU>ALGH7b
z=rrI?Iw%;^uErdI9THCRx>-;gKR}ovAocs>j+8UCR)w1RB`F1+7S6o&-dQL2WCr;W
z(2{ZeeiD`0K@a(FS7;mdBo*xbzEuDpM^jfCW2C*fmGnN+9EbACS8whS!5qnTnS;fo
zL-v<mQ7-rbTt-*>E?Ly}`}7W*+-$kJ8>$kEZ-=CUwe$>VIo4}Wu9EL@<siI)t?zud
zQCj{`y@f`OadWb%6U$wnJ@NIk0OVSs)E9(%FrPge#7Y}4BSCwrpR~(k?<bHIJR;gh
z5k~1Tmp`r^Ds&|XC-3l{OdLQ(0u$B|M(j?Dq9bUEQqpMe<afdmPNIv=+F!LDrO5lD
zWzEKx%c1?=Qc2e7zkR!qI|6o5E@wal$32afGbUEm-1E|q1V1FREY?p@Rx_7T0X2nV
zQwC|jWCp4G1DJ6S^DmKj<kB$(uWrR%|H&Ra9{0fP<H!9{13G@8LX;JpIXsgEm~w)`
z%h+Fk_Xi3{5%wB^2itN-b#!!O_PC1WN*beouYw+bfVQjb7|M6-_Exbu`rXa^7YmRG
z3$Ts3>y_8=$VJegfScwwVU42}DY{>S5l>;I{wqZ#n^BL)JKZ<cNeQ<*j2K2D+`tR^
z?F|@YUO`pF#{qRWZrQT#H)7~C{GZ|L*R7*-4Q9P`CObGZ_ockZStyXI504~19)eU-
z*s#B``XPhzyTLw5AQle-Jz8Oja22(icFh<os>oaQ%S13p0wW&l6L?+%IP)mGocwzK
zvMu{Sx3-@<rBaCJed&Zh!$u7PYc(1>?orA8A|YPc)%(3oKA%|pNl;sKa?P6d1^HTR
zEC+rgVFN>7gw_X_>P78u-RNt$1iM%Ox6W!KV;Pmb&7zh;fyQg=zXrr(;1(AW^Gu8g
zQWW-uB`9)&w&$-dx@&2Cp<!W{{grg19svcGthg#q?+W9-VcX1^8e}TH+R=D?pTNcu
zwq*ryRTx;$Pn(qvJ&wzbr~Re8bCB|GJu`ZtiyEkVVW;|HzLVj(tmTd5eM$^M`_Kuy
zN-iOUMteJ`L`zBCbNXc;7<2#~=jva0;8R_Gb*rkUJ(J{6>D|Z|Gs!#E`fYyu!#Ioe
zGE**u6uLm64VhYi{4Z)If&4IdRqVih20ZEH^xjmLAg;iP*AhR35AV?ksP^eTtpwev
z7b<I=DruVS5k7U<zp6;$DZrd$M+p>@o%#CptDj?<=^8gDzu>(g=6ywBq!m0Te{CVE
zQnSGPTjrBM1@HYnzbFf;^AW>Ip-$YA-~|&2RgSS=@`50?M?IgqvXn~V+G++hG6%Fh
z_IXykBcL$aq+6bM3UwT$sjZRb=B?sYWe84^eL5kYj7*oB9RA)lRnArrpdPS4@z^c2
zQNwRDTWZc{6Fr`y3AkQ<_KN?$7kH-Lucjp=ui;9^OVfOBIhyj0<C^6@8qG0-Al>%`
z`(tDioC!qPBv+63a2jpB(WfWc<M{y?KG}~8jkQb>xe4zF(jM8q-d|4p1WI1qrq?Ku
zhz?j_L_#sYU%KB#mJSJtz32zHq?4fB7sfHOS%T9&nOl7F3$ETi6BTnDsJ7Runp1c|
zs@t9|>f7Y`=HSZ&o%{6L4~}}PNi!VY`aW*{m_*bgd7FJrvlu_{3WB8S4m=Slv~hSs
zt<_4+-Ez94`_Ax=#Nl&1&Q-Pc2WjKxo35tvWWU@yxC;WPg%|oX@j=vWf~Jctt{-YS
zL&?dlbr$zI8AJk%H!!^-Y&g|Kt0X{Hxcn#i(_guyoE0$Aj6yINLe2#>Hk?Eoc$i0$
zT4}wzeP5#obs4WM%b*~$RLFS+Q+9Y#ruR!Gk%%gAj93>wJ>i)9ux8NW3lh-eY|Vq2
zf7)OBD3e6TYrq=U(^~5E-x0{D?qeq%sGqX-9~KdS6R^lRw|rTem^{kQAoR*_E#uNY
zTHK_xG&0e(3toCcSkM_NL^bqQZ=VfAq(P?E$!hRALKb%P=+UrLdMAU3@f1h(fc?3b
z=R>s5v`4_<`NCuFTaxdyAKk;daed!j`dPtV($0vP!{^w6g79?Lqy;NRjl`=R+h=Q-
z@dfC!*s?a-aa?#5gM1#d|I%>@NXLJYCD6hi62IsYu$Aw-ni%JL_}+=oU$z86Y>Rq3
z7<sDW!Fwq_i{QbVMABFGZ6N@(Tv?6INYcVRSS+u73M$Uh*JoqAK|L(r^~=ZKM;hU1
z$~wldJ&<mv&}!{oUu*EVh3Q)E?6ZLbyrp|Y)*0TT+~?`xF?WvMDa~(O+H+rY;u(f4
zm7=-&W<^ViZ@RBh?i|11Zb=u!2MkaT<hF0fZM-8oI~8@0QquhmAUR!7$jq-lN}*Yt
zmO5aM)ARhd;8qusZ(RJ(LwIpFFs*w6mn6-BhL*_PqAv&o{HZD=3d)6RhYDZOxhune
ziIsRhFm*vf2b#omRp3nAE;iCDp8jT5H`;^8^Y^(Q6pT=SYl$thiapPMHZv8c$NRS}
zAc(*g=I&RJ;>C%^B;;>uOs&3(lUC<z-0e|;;PpI+e<O#(H8$6C^iz~mtFANscXc0;
zkh+QB9d?psICzK|9Z+3MY^it<!BBkozkNLOLd+Zvr7Tz}vQ4d~<@)Gm`di}z2LKS1
z4k(6*AVy(lE=sraYbdvHj?xD71iSr`oOse1PU}LK8k6=D%E0hzB$H2g?~m73KsXaj
zE2exE;R+SFo9`@bMdZ7ykZ!R{^pb&3$?j}9h8Wh_wc%8)H*CkHz9@J4lJ1S&KS3^v
z`T2p;r<&CEcA|)Lu;%BswEpc2MIX7&iAnC28e_;|r6D$>5&Ffg7$}Pr$$i0n#u*ut
z0q8+6UOk2vcWtMyo+s1Lq)49dE4@R{;AmP@y1bK~FVG0>y|9j>w%ZiOY%`V}2xnEI
z@Fn<d;D!VyO|46j4HYj$&v5SPJMtTR{j}OyAqU~sx1&GWl(Ws+e?zEA0!2-ZTQ~JV
z&h2(bEdp%x<~sHP*>pW&f&tx#`r2{rqF5786U!5OuLqPZQw8$<jL7!~o`{leRS*kB
z`jpz@9SNPlG9klQC%M5rBMLg)o^g(@?UO3$7F#POk9&Q#flWIUM<m16uisOY_0Ys)
zK*kt2ZAhH@p)N5kgQVGqh?@vUM#rIKYwE|%$Jd9_x<4ub&LfN9Q){!s=O}`e+oP5A
zb%!Xzo~aC9Q^I|v5?(}iMA0a%ipE_@?t7Z#h2GiHeCKwVPW71t=jDR1;+z4YTT}15
z&^p!U)0Irq9VTZiCTrPb?Y%01m;$}JNd|)7_hhGufE#?#q_~w!A7>)X$9t>!&BHyR
z0@5t4(PieN@0|X_ymV6E5*!HihCsmPTj2CDrFV%>z8|k#I@0w;yrc2ZiTN+nqNix8
zdh|1C+-1|wou6bms(E3i^yvCh#kvEyAc*0_GJMRZwBQLpt}5rzb?arN{4E)SQX;`+
zt>b#`?h}rOuh545gkz#>anNf_YNFwNOv3yesWhLR#qm?^{G@U?97G4Z9FjYLa+Bbx
zc6D3oAF`L6gH3!000%>J5_Z{tZ(Q@itNvX0an|5cUH9QlKJB?#&gVQvzFOMs`Ba-9
z;(GHZ94n)P*BUc*O^7`-K4xZeh(#A}y{=4^*@hp=gS3`4{)#ib-+5GJt?d(Bb14Rz
zz|66xy;3bH6BN|0yR9gVck<~-s`7#u3Hb=8rM|~WJZ$K5*x>=;cGWOiJ@Y$5ZZXn_
z#V$K_#TGM<vIIt1w;Ej7xOCB(X`qC$X2FbiHR8$o74ydx7-g1o0{%8$=nyIR;5#NK
ziP~6Pj>>lzlV?&IV%gL8G7Vi|j-|Rku9#@3tUmx=d+n@*FR}^e(#x%%S3~fb)@S)g
zeF(GyYiu6HEr|-+d(AC~chHSW1PUH^%3}${zjVrWs_i+;`50D+o@fM!e~*$rt0qAr
zgkao_<3fF?4#wy8^<Hm(rO9%~5B8cG5ISk8%o)A;h?m6*Gooy>+$1b~qF+q*@GDbA
z_MHO@U%oGiNE9ZvGxl_ET(vhNc%Xad$7YOP1XJC~HO(S^e@w>#K?@n(R~6KFh_ehR
z2tOjIFBtmbXZvXm^bL%3V*c}(6rAyKfW8)7p0$tTlxf#|?h|!R{;Qd<*M*+Q9@^rS
zBP|sQULWaatmLUOeVu=a_2+Z?>NPjlA1XwXH{5KgF&=eY?Y{f>d3L~^N}S~Iv9TL=
zsd0i*r#ueqE`&_RS-z;4>>E9YH+WNbz&EO$O=XmO_-b&aZ{S<Wh6(}m5AsimXT<bB
zTzAa-Tr(YJ*@NnS`bHEVGMlJeP=+dh2&}n3@f+TGvNh&Tp-%loF*@SYJK7Jy*7985
zvE><K(w5{KN^s(e!28u^11lQ7+2|DEEaSjd@k=wVvg;)ZQO^{zrAtlw$P^32hR0#^
z&YM$3v%W8C6!ehbF(l$0Y`qNePE6_?yq`txLh&X8(Z%TL5U-IH|6OeBZH7LlkF}IP
ztJC`nU){Db7ZF9kU?yCSF1}T!qQPM*Y}v+PxK}nVGQqxi`bncy7&SQNbqYGWeQ7Mw
ztBo>Qn&(p*AaI0u4p2e8Xs$_O&v;Cwlx=(iRb9%&LRL-sXfMo-W%+8D5HS?q+uY)4
zc#%E;JQ>TB@qIm!C{tQ)d~Qsr*lGIB8+``ZWOz=sZElz{TT?e#$+@X1#rKI2!jAY3
z5chLe1Y+s^Y@mAjnSxL)j+L@F`(MeKoGwc2Mh-fwSj8SHV4;?7I|4LiKH62NN5G)W
zbIw=2?~KzM-zpo+8a5K4c33ClhWu164yxC#$?@<eQ@bsuU|FUDY`IIl&mdl$a0~z?
zJO139-&~fNa}UgR=96qB-3{I<ypidZ7B=k<rgG|ZhV?}$uO2{&5OncOmyiruloWpM
zmJJ#ekqC6q5T*0w<A~3ZEa>jC7$2|iahu2`a^v7AABE5>BA~vg*d*%-dSe;FF$pV!
zvuQOh`7#fF;c5G5a889}$HdZQSvG%+PrnT%^+l0rkNMp7NZ8h!8eF&MN^C|9@^}o2
zx|2S#vXvG>W8;VI7_;f2!Wv2y3hc|zk$`<~vdeV~q+Xp06?Y<H-IeJBuq|2iptn`S
z(0E5k`ez()yya=|;i<Xxo~!P?U)VFq7NXO?cM>te<<ImviJ~?%(>kj%+bP60=dW_b
z)!qLHUWCL4Mj@CuI+X;edl&Zp=+W0s7dqV;vi=Kr@GAAA*6vbBQuz8Sl*-(KVXjVL
z$;L=g05iUr<YD+6^UPOKjWZ6%-5T_1Loc58mC7QANsu_)?OSkr)joHp?nEIgOZ*Qg
z4iZrU<ihwzM1}$?dDM0~7o9dtrOY4S58H`N#6!>#RvdOXge{;L%c<6U7kH^p6b^kK
z4bR}-T4o8OW+=AE8^0$6I#bIJBlr%rHt^`>w$_|16cp`Ff}94qC|Kj>i*~>Q0wz$i
zRTQ%8RT*4#m3U2<Lqxcjk;PM~mJSrY$_(~aexF-<bM-O<<nr)L^JYeuA`Gdott3kG
zF|g)WppegSa6NA3Mwle6y0kx<e&){jSqlIyz+tX~h4g$LAb@c@i6XlAay?2~ZCF0=
z3hztXyghu8=L_`sd6VFF{ci7&qeaS|w1AxGW2nQN9die(R}yO`6asn(K~I8Q8e!PL
zHE29OK3%%fW<guSW<_|t7O=<92563i(5=<ez(qlcnKcpRi;I`^o0E2(1U{){t9l@Z
zTv}{uVR57iF5?f=qz7Qm{*va5X)ybG&|q|c>AGc0zcv7p1t8N4m8*&-IdCG2S}epp
zx<I*r1wwuFe@f(oNl?5IOv%YixA3(nTwB`LDfAOZnN7L5em1QO&#Ald)Kr`Wyek_m
zOeLU2rapYJQkKGf#{8>W*L~zcoSuND?^E7PZm#QduqjODhrA8-Fou~~fREYI(w!Ss
zIEMEzlaL|OV=ogYzJ*#iU#sM<&lc_BHN!%36jU1pE<2BJ4u4P7w+o3|qPS<0bL{}?
z75=)+oC!scmgCR-H6gxi#ykjuEnWg_*kJUzqa@<R$mY>TpRBJV5`5=>HJy=)a3=<i
zdql%&T1BQ>ir4%>7f|v{Mv}gQ<JOOF(Pu}q&POt{Fl$lUT|WSB_I%5|lv0&d$Z=&|
z!`b4*XTB~-uo$3@V!uz&H4y#hU;=OYCD(5?ZirC}xxENcl3^yVayVE8u960U$4T{d
zHGmQGLqE*Qb(3otxp76S;RT*3JwTKnk&XG%ZIOvi=UXj0G<~v19@E%mfTmZ@*ZETd
zgW?Sfq`tzJ5NkTPq{XKp-_F_dS$|g<0ekmVn&-Z=e=Y__)*W?pMChIGd5Y+BD{7oH
zBwN=n0;#qqw=@f}u~p-=AM)RF`AT>fs&yGbU6q!=L>?Fm)eDzZ7BTp(<qHzDGQ2no
zJ0U)RFxMBh2e;5__jH%6=YKT&1e$Zihf!<8-8{F;PE1EJHL3+{(Yqr#k0oBe4xi)n
z3tYeCXxUud8Kt`TDuhUXzT^-g<e5o;d{KP=bQ-`B$+fq`;$`+qAJ%pA3Gi3UiHwLX
zN7K+7nQ7(K7Sh5xe02BG{Dh`27i!|pW@Oda7DzjPhYaVsM4xDA8qvqDb^PATKS&fI
z3waT$2!3K|1Z4_4Cf(**Tf>4*B5CTvEbo=IJ!<I?ESs3y-C)Y7Zqe+WWd<ztGe;$=
zM<sV%;%oeoV9h`iVScoO{yw@6XM)PWZS-dP^T`ZKLVGts6)1+DB?l*{x2ER?CqQCq
zJW)M@CB}2p;vHEBG%|Lx9R?Znsf$Eh^Ia9=g`J(bXVOkRCPwPY!ItWstXw`<m9RIC
zFch1}@mQY&r$(6FTj?Sp7^da5o~Xuv6C^;S9-K`U4@y@@G%)$p<ERy?P<J)?l?*Dq
zUB@LJj!c*4f%9tRfQEHPT<~k?=Q8M_obL}f6lzU9;!x8b{Y&b&VinCmP(xwG0hCWD
zHEn7f4-4Qypj$ukgKhyqL4k1qZaApbcj$TIGY-O!wAIaJT&(kT)ep95R(7*bBD6Nk
zbeaNGyk-+a1{@yfT}SxBA`C2CRZ5@eoXz5r;5&;k^<b4PxOB?UW+VQEFC08!!cbmO
zrUvBrrL6A+9zu?M+u<9!{Y(^^R1Fe@Ofx2CqGRY&XUeKB?Bp|lF2@=RJke8Nw6bk8
zKY}yNH8Lm#G2?d9-A9hWCrBi+s<hoV3g|U_CZx0=oD`A{GAjMYS%70HCw^o8cFnWK
zjm}8^+~OUg9bdwng!T?dt8ubyFZK>K<1iuxFg8>Gp+NJT;*iOm^R>f=9x@k~4x;b?
zi<2lUM;$A<Mp?!S-;Sj`DTz|Y^8$;o1J_tS*M<iQW%1;1jNYy(-d%lF*`rc*a$>i3
z=17I7UvT3U*JT7x=Ii2O!O$fdRWMt1n=xbdqhZg90@wDqdI@T>l$l4OV3>!nh-F2;
zYN4JFlMRgOkAga5C8!DnKeLVl7(~z<%gM=kYhU^;I&j=MV3BQZ(R8;G=}3`b$XF_=
z7TpW!V-aWVZJ>|*2Q6lG2Y#g58t{FICtJDF-bQm8MIs)7whJ3xU|S!3H$t=Q(oxk2
z98y0c$hEacMAd;s0^mZ>6c0Ar3fJoe$hNY6lHH!qW&yHpkZth8dfKPnQh5%J)2bL{
zSu@yJv=CF>qS9Mrlo)78z3L3!@(SszH^0iUM24$yPC$^3-j5XcDi+?m#(-(Jw!7vn
z52wTJiWEvRGYa4F0r7)0{)lj&k#*^L1!cFP^t3IaO>)r6%%?!b;R)D$XyoQ?KXSpu
zPcbS;`zKKECW4>hS4w-=zZ^-BUppdq5>){;A}78P9{~qqKAV)ZX(5(Zaw{Q=agyZA
zF}>}86%ht=%Zkv+$+0c34{Y~U1|Ir2Dy>~Z%x7-q+1H7afj*~diZvnsStg2(m&&5t
z<bRly_wqO|$uCcWjDf?x+=#fu*Ci+*mO0Nbsqg^LbQiJQ;LTZEuWudN)9F@Ch*4f5
z&7y^`2%Z_`Lt97EyM`UX)`ReVR@mfcx(aRcnlUm7JWc}aF2%QprvXxrL3qcMeA7AG
z$TL{<<jQ}El7x`#f|B}M`}TV6dgW|EFh)T2ut-GY018IXW%<zXn7g@2@tRMoiMt4C
zDV03ZA_Tas3IbD}M$P)gjYl68_3Ax!@ffu4M?YfR^kf0h)tkg`Qd2p`JMEjoIM+hr
zKE?VyJ+jRR2I&}`hUarL4N!fawB<P;AQXJthdh|$s`x1z@cGy+d`u%OfO#fl(0cU~
ziICvpiM8UOvIjQ69shusG1cw|m_f=a*QmP0lG|4pAS-=W)Ofq0A}5VPsm#{a8(3RA
za_F3EZzlxL2wWZy8vv$CIyy-4RCgQ$RJ@gBo$;tKWYIV`{==eVq&9%*mNoF}0Jxsi
zbdnqN*$~DMc76*nIvz0QU?$!q^F#o1hHq3ew&ts9TW>}zRE69((21{?fk9x|PFS74
zI)C=L8g%h=Ik_SPkHck~;SW{Uc66Pmpfz5Rv&-PI76dYR_BasRhu`{vVjyD<z?u!E
z`OB=`qVnqi>=9;)DK`L7VGss5n;NRar)r=H#uB}tSfH;*#B@K2*3%f|Pw>8KH<a$U
zbJ^q`7kH(g73z!T;^fl{UwU3z9UcRg<RGvl6vUzNz`LS8;GOIhuuXs28OpHc1&YGf
zTswFI0Osm_u{M-1Z4R(JY&x<SFc(?|Zl3qy!V@#s#eqyq4?S93KyHvDro+a`+gg3D
z*iRG`rBwlAaQ{HMXa9dtUm^C)*-`Jn>xxAsI<F&Y2b~jQ?XNIkf)r*@eyZO3q!$Ve
z1!mqY%p?Jktg`N>71@b_FfXpKb?<;hUsE~EbOS!IWov`9Oz-8f=r3m5RV<_keEE^0
z$gu~|)E~MNzv*JLyXstezA5#!ZIOz*hRzn9vR>0ud22E5!a@<r0)&2=IpbbSNGZ{w
z$U6;Z7@pt9jLpT|g{j%sro6!`SrRI7LDPEWu$?%4A9)+liE;5R3usYGkYXU<-I*g*
z6IlUvvhqATsFt$`-8s>;zEueDK>`($KsDs!6v65h4kHg~_a_)6B61Y2>C_z;Ju&NK
zzIc`6@InwLK0n|?7i7^l;FZ3>=97ZRCrFT8WzjLPgs$;w6058lM*ZUO^WqBi3({gs
zttyij7kL7d{jLBpaX=lV610CYVx@C#URLUo(+ry>2D1-%8x_2+!&Gz9KD83pjq?lS
zfPmizHShBBazGbAHFfm|V`5YMq6Q<9t81QdbhtoBfGsSfo5{=9=}fQThQKGbD&~FE
zP)d&0;;;|isQt0zcFo{|X7UX?jSfhko)YKP>qSSLwVn`udRG`96mcP>WbDl>z(j=H
zaR9S2`ch$&HuyZCiZF{X1ydG5oMlIacDqXG7Nn(^IWA==_k*sn_%n1u^p%AuQ>NvB
z2-k0Mu4D?&=%9`A!86V6K4_%5r-bp^IrFhX39|>7O*YPZ&;=-lCnBP2)tGI6m_Udv
zmurr%)bk|4IgBq?ezEouiryZ<fNU;jp$}1+rykWD5#PKgp<jWq`ekb9=zH}pcKxkH
zd2DqY6(ENy2bFB4lZJK>3%yt^H3yo=tO%|cCg4gGQrTKx!ZdZn^Y07k5u}X<>tvOr
z&c_Gc0e4TrHyLIsWVKIrUx*#KfLb>V_1SJjBh+A1CCuaWb&!pkoiEj+;V>hvfLf(R
zR+-SmGBL+a6{_>Z(4yl7B~dmSBh2;#==vI@P)#9OGvT(WY%OQvsRRKFm!Z&Y*b{B)
z5q5>mDEE7M1p$U@B$N)#U0aSHTfcC5T3(L@Qu`}~1_RjyiZ`NHg~;L2JfqSQYvk4z
z+if^7L``5HQEGQjOzidvd7#6%9G1OBYuAH{-_<7>97-Q*#DA!L{{&_~{hTvw7lcec
zIu1lXb0z_pJER+s+o9ILX2?CL;z4@Xpk-L8Nac-wgz1%dKD#fG!uTGCGAGqsYgSV4
z0W{Pr$!WPIn=_%JoWPYoN*K_iFo%R=`i{);XuAt0j(S^%I4@6LtPVbp=M2`c0U?#>
z2)p?I*!t?QD7Wry89@h7Iuwr5oq~YU(%p#CA|a)Klz_lkAW9A0iZn<!0wV?uN=U;<
zibzN|-+pFL;r)I8oXhJl^URL5@3q#w?zJC?3#2f4HQlE9>y5*LS`6_!nW?7K)8?2)
zM?@uj*N|q@)W0niQtr~ke3=B!X&j`^=(x}(5P<43TLqPH)nGSW4*mK|lG(ETA1%Nm
zO}hJ|IK-cs`7V)4gEAo~AOyBT<v}3m53x^lcj(wHVHHvy<kA>u#&Ez|a3Hj$D4(X_
zHthRWR74_H$O_!Hf%!a3=;Ec~`d#juPUVvUjTQLKn7k#3J}hFcL8N`Z%zC<iUcNt+
z2OwoC_!&ko{*Mp>s&HVcdEA?MlXP5GMC+J{fxQ|<XKaYPRO75}-)JH~mUI?)oMjlT
z%{>Xu$U2ARZkG9rTPR%YLWlWFF%85icK)}K3xCQ8QaK835bTl`4M22SU;>D&M)oBe
zByZ<y9{k>sA$notyNo%BzmO;{<d{RTyoqCy$3oWyV~~4illAu$XN!wjud+>vDt0;l
zg&-fo%2q19bpziCz~|TUjJ+rU@_#u<h*#Pf;0mo5=p>}gyukFJq>Y9}ieBJ79K+!9
zQy`>bUGHnSf2ef$lmyxWe7>$~QB<8qJRuxDh~U|=YPl>sZ5HN`@}w^v(@-Ey3iP+8
zT{p>XLTYLQhl+u!ClacMJSX;_hkVr)0?;*f@VG{^;=%`xx_g`<HcAkHq@zL@B3tkN
zygOGZ6a|X}F^CD7=bdkYY}0MlvDv0N)-iX-86hwQ_`!&_o7*TZxVRI%XG+sDKMWFt
z-;lVnV1<zfI+Pq30(^caSd?SJ&V7IbsUXiue+%>7<Z#!LxRK$tx{ZSCs;Y2&@bM-0
zLz;6x=zybpfc3&7GVp%^I?7Xc>fb$B`B4yBn;qzlnTRx)2>zD=i1kxS#e}?kbJ@g`
znKcfEWMUY3PT+GCYfLof>cNc~%dNdj&1z}@Rl`xA2aG~pi0e5^td(n`Fj)vV67=+f
z?X-J_3&W2MiGCsgg@A+_dgEqPxWGhTrNP9W>Bfzc^L(MGO5uFI_#gZ1^TMkjZZJia
zeTKPl##9yAH+Qu^UsSyY$dhPPMzs3W-W$av>x}bAePa7dcMoIN0@!}mqhej)*AEm|
zs<~ePbUD0c0S<9!%tq!eD?Q4H(<f8nX*MrDx{Vx(h7@$t+PJ1}TW<w;_#(!NtKB|v
z$qz)S1Ka-wBQ{wYzER=ytW}E7cJW3GgE+KYp#7q}^;fQ9ipY8pb4(g7%ogOhQJ-})
z1;JmnWU&29`wn9q0=XiyACR1?=K=sf2+RRroUv~i!O(*<(9#Qaom_d!#sy490m?vA
z-?H7zy3ed0MG?u0UvubwuHbm^T+qJxU^;x}*UD$~!lWHIB{=|-vy#tLL1;#S6~S1P
zYOHFpwXQmw&c{5G&13|@)phfz8qt<VP6MD5S(9kpIv?b3s|4as-*vrDnXcnCcSgHv
zd9h%nW+eA;{8-QAeUH80I6}L4F%5ha2ZVLvr!lq>PI}I>EamBSo8oN0U0hVrt~aNE
zqAUsILH+BeyHq0Fo?LVP1jSr@gWnv$!F*1F;T)|SmbmWPAukJ%rn5t<Pao!v@U1#n
z`D#Yzr#;B<BlKuI0XSUrlV=>>HiAq)t!^b*vZ>HSeaY#8P$xO^-F<VoKBz3$7`!!%
zdJ3h$=U$>FV#G~<U-#Q`jcX6P{D6PM1v(GFlrNFp7#QmtbH#AY3USW?5*`3vlOhx&
zo|opvwae&BnVeoLzVIRYg9T}P82A=y2(*MOo5c>8R2kpxl~BF=;0!wyhp=P$2lKe|
z*6UZ5{lp@L4)uMASbPWPn;$;L-CleSUbX|at3U=DK9U5!K{=mwn`j_i4_!ulkMxu4
zGIjoS4LbM1Af>)uHDEIBYv5rt!<22Sp$*C$HU6#huL+Zr$PBN1u+(M7docC{4+F;}
zf#uTIGZz1QI9hr0YlyNahMxA339^C01Znzpn0|1kS~*G+X>Ts=Ci9C>8?4nhFaJ`l
zW_{!qpA7K=1p1d2o;HEtBb~bcviDp~an}sXL2;UMx~AMAX{^FqSghv$Ib5)@&Ul#_
zI5>bs-~xVCH3(q=zEC6ZZIVlO7H)kC7<Nmo(m=X;8W%@^!oLMqp~<oka>2E8_^Pmh
zubH>EH*dKk97HDJWS;0$9y+*85|eFpT8q+Cro$fgbZXBItITDB`Yt+Le3E`jU}80?
zAB8ZxRz!)Z3_YnD{!Qgorq-yO6XJOO5%8-J`}mf2lS0qD+zxUu+V~1YU@f1aba^!O
z@6smOx(;0m6cs-I_729z{6XKuHp{{ESHVUz<H8F!fVyp&5g-lUOKuFtg9$Ow(6nq2
zc#WnIzEdcsiLhgyr;XW#{4f%}`+m~i*zm%W^<2{WA@{Ky{oG;}<%SGkx_1yeHWF7Q
zZ(Er~&)x@`gCE6_m=H^vehiio5>yv%^WXm4+KMS}+i$Wb5W9d_YYAn_#rp9Z=*KqS
z=qM)NeU{PZD5R+xrJE3y183|z;jTv4$bzrU=U)c)gQ6m5{gkgmpg(GQgp%@${8-Ta
z;06ZKNzPTHbXmo$8rq78&pyOo0ArTHUm_ooX)eap{G}a$vVuk-CvWC4(jY-^z8KNB
zJAYwF1{eI82bGfq!%B_G`;ldIj6F5qLmB}HdA}UeQVo3ZzZ{e&2x{r(usnYlE6)9g
z?jTI#QPZ`lW+XWccFXw+ivR|hB8^{dzqAdw@hw+YH>%l=IpxN(7GygN@J)N9W0-Ai
z^MW?(eVaZvkpo{ip)=s<p(5s4YR9BJCv?s0-OI|9vW8C*u#<wcA=CRMil0No7E8Oi
zE4p7)fkg-cyOzV==78C?u*;ZkApMZ?=GOrcEVQ1V#&DGbSl+TdgsNO9wxo-ZtNv-h
z3vs8nR(0DnBRAn6_0ZYTnVe>{5w^l~=j49dp<^#H6=BZRILDIe+?uA^X2+(%`1zC`
zII}8nb;TVK!Wc_rp@<c*Qu)N?Z5a|dfJ^9pCX2PnV>haAFKWrSUsMP-WLe#4HIiTG
z{p6ju+Z=)K*SPqJOm^5;ihj@Uf{RhV{v%$y8Mhu}t)hcX<H7AgnNhlXU&sF2nENP_
zP7fmm>L$wHCmFFJ#4eu3m5H8*J1ztDgcKUC1Go_(k*;dH315Ld5;#Quy0%d$=IS<9
zV!8DUnE@sC(r8WC==l53UybJ4TtigZRL1*8#4lim4v#cN^F+k6wf$F-x2mn=FYSf5
z`P|v7GkN-OW>6ct1IMBm_k`rN*K|`SQ!)fDu-N+G-!2jYVuya-S<(op|KcSa@&g{f
z+_SD94PqM{8hSwN)1&c19G+&3-d9Z?6N{KXc$DUcyv#fGUk|o%jvfh{-rj#w5W3T+
z^o)rD#sQIXk7P4|9|GRe2iFc5<yz#%#ka<uk!>LaY52T*Nof)j-H+1XXAEB{7AvZK
zXN?iAI-LL`7Ix#gxt+#Lw`w;GupESkXn>9R&CS-aHy@`hGu~|ngcS4#dVUI=lJyY*
z37d=Cy|Uf`E@B%1#=n3-|LjPJ7#2U-<580twu>F?J_W->_*08zD5B_{_s9#}vm^3^
z4GQoxswk0Ghmx9PxTU5SI&`|Ys&V0_OY7|qhS5+j8G@0}wS837B{IOK()O7p`HOcF
zL)b0#2d^W3S5wa?HU+9pyMu)>766g^tsp#8`}7n$KsjAVb8UI!n|D?f(HXUXXBRIz
zP`xu)J~>6h)qVG?nnn&YLFMkH2npUTau<n<hD$eC&m{jcEOxz5wgW|1T`WtePe#ON
zT|^a^JC|msT$3amG1Ee_gP}GIH;x_zZ%9Ubba$1Z4+1fIswbx<hNX}ORCDiHI`5d^
zrk8~G;~uu})(Xt?4Io&9-E6tXjG`2Om2yvcsaS_9@6*O4yACihaaTM5I$MOnuisFE
z7tlhV74e*2nkoNQLwqK6l~j3GR8*Ty4WraTL=wI`r^H4lbbt6EhsP^Pm^>+(txNuv
zH90U$aB5R|v_`Zl{7CxyjbAw<9P+sS<ElX1>P4~To!Z|{z9T|!Vv6T}Cb6`4(nxlo
zTfkf6s^4JiBOA=62c{bqmmtTM3Zw1tv!KJ9G7Fncao^nugX(^m_(2bGF@grNZtiTn
z*{8rYfRt;Bx#<x2KjKMxTJK%6x=KsS`Ja7$(E^Rl;H6TUO`_cTsM&kW^Q6CNY(q9q
zD<`n{X%HlbHBhU2aYULDq{X;o#3(eG8IDwsM_p$63SF4e&+pxe-vuP$*`#N&6C+Uy
z^-p-{HvZtYUv%>4%kt=h+l!{CiM-a;(6LTt2?P&k183NC3X)P`auO0PExAwRsTQyC
zA8}<hX$#m!P-&+!wx_k15>|6@S8ppKrTb0>H5BJ%u_tp{8s@MUvP3+h4@w-=iEmpO
zEBwhw5zl&mvvzT3-p%<f@8XX4!p;h9!Hea~%4>Kxl=R0EpJNx~qanJCXL0{@x`R!W
zf>|O`qAbg$B1wIpvDqVPs6(JutLw#nF^a~>qV~!+b}K%cIjjBC%EV1;r<)h=mo%!^
zy*Hs%D}S#>JLT1ub~;1N`_}@U&(vwnNQqOYA(X#F%FbQur>N<;?;jV;lNxWK^8$!b
zFgsDW2dj;9@Or;q|CtsX#fvypIuqLC7z+WbmsN>+y)xa--g_^3Dm^?Y<HlZ|c9f~P
zHDBZwH-}K`hzD}OHOfb&#X+U{-Tu9U0sRz#l<{H?m)_RP@-#cGb8D*yDcPnAy3DyT
z_CcTbcXT@?*8C1Oqz)E$#MDxkWRVy5TFr=6mornQbefH6Exp>NLNg=1@+Q+28>5vp
z_dHZ3cvl)fM=P^vIx1~7JwvKL@Ocp|Dd}dp{VXNlxXPAnz>^Y;QyOgxdrG;nVE=~6
zCb~`Xj!zU{%Af5%N6v3lWfgE*Lwu?9WW@>Kg{COY8T3sxHLz|U*gDpG!Ec~{=GNXF
z`@Xgq9@P={OC3j%<<I3IJsRkM?(()P*7Y4G7jsQ;S(!k`6qU~J8eN)0Ac4#pT2XP?
zSB0zjCbf458`bu;m1|C7qA%z#4zlKNf9l?AO)P2jN&^mC9LAm2GSS3wxD%M|0RO=u
z$TPhz-k|FnV%dmyE>!98&m@KZs`W9uJnaLY%WxXOttOzKgMl(Gm+^l(%2A%U1geW;
z+SzQH(XP+b3NChviT~BOi3gxQTwfLu;kmO*DFe3qDwspW%7o+L3SdYP0GHC>CccG{
zK1Ib&;gtxMg!4+=7FcWXmtMsaR`T*@O{lxNt$)WKyo%$7Qg26c^Z(2tf}%2eE@L~%
zro*?Ot;1<HI6^$64-4hs@iS!o4%Ul~cJ`G-AUTeWJwcfLyz2L#uKhp1bQbU1dY<>J
zlF858ch|KeOs{`ne|f0HK}^D5#;)#XIBdsHLRWT7k%JADRMo(u=HCs?DnvQ928_kJ
zByntcGxr*J{@3_d`Orzh9yide?s(Qf2CU8y8%oA*=bh&tS<zuQ11_BQn7o*{^)Zr7
z^3HMzOKob)#Q8skD7d&0%1t3dVvp|&@K>AQJO_^jvHh9kp;6_qJIxipHt0=vaKK-~
zQRd4UzoYQiV1}s7Im#TFk|FEmPTRb4IP*W2gv(U;vrAw1C*9cEL+jgP(V79L$zKAU
z1=dhd*Kvnk?d}_Gx1zz%mxG^o913JR9QdCl;N*6)&#yFkSSvZ55vt^6`?S#+llMYD
zp~Z{i(H(G8CF(`Q{c&!T3};7$>Yob;u!cbIgpgVMqAegM+f{`82@6#f|6?cS_Au`B
zO=Eg~A_y;}$N!3Du|EvnytMmkT+F7uUto>SD!ZFytiXru{vkmDepwJlq$DDKkS0Ix
zChQawNhL4<6knj{pQ&LMK-w7h`b@w~wa>5I*ZbOCtCg#%7@8US<B|c6hg5b|?Qi=)
zF!vHL_utuvw)^KWXT_oh^UHJ<-Qyo58r!INnS46s%zt=9=}DeRH=Dd|pP$Wb2$Utj
zZGcWvhe_C<iY#^!1A|d-tH$rTt%=G`(}g?q^;5k0+uUF1h|s?8?V;=aVWWJHU6USk
z<WcD~SJ}s7V(P#9NJ)L8#Nh|e8*7``*>lQnlypre{;VLq8iK8TH%Z+wSmfFHP(8Sw
zng3J0wf4B<m#lnLLEBJ|UA6xlot@eA%T>7`Q2$6b6FIeKgNkUi?X_)1a;Zlmq=BJe
zAvcFv0O9e+0KmTxIRtqf#GP(@NlYczmBsYju6+3Ocf=jILu6)1OX=e8k~<$+SYT1m
zLAq+Bad^%S7t?}6tBLrbR&l7ZD04I($KQ^uNCR_O6|ae52QU2xp;MP}Q(1r&8E!zO
z)guFV#R{|cT6A>cO8UO6%QCMML2~A=tTdHtDYGZgRIGyOp@%D1*nfy-{KByKM7n!_
z&nd<xiYT3`D6gNX?^O}i?Gj^2w=e=K{69D`g%_0F)Aas+XVRYZl;R~E#1g=yrygwB
zO3&Yc*l{`yExYlR>Wpv*B;D;A6*~LjZx0Q<1}uLlGq~-F0|f6cba|@(k0-=FhaTKv
z&to=bY`~c`pFyQ8NTwAR^mia@cxqDlev?ISrd`_~6mNfs3;h4XqjWpo!n@55TiOaW
ziEGw2N?wlbCrSTqn9?Kv*&+4N%kTQC^CF*AyULlX`9GydIeBbN=@HO4HkKnZfuE8^
znj)oP|Jvw;DEYh7)g5O;doSaDAk9hzM!{(zh`j>p@9{_nl1%l*?TVN@*JsV6yd3(i
z;%)zS{R^AGz^eQedT=DDbd)QtfE48qDuFa074O@0fSEXB66aQb3Mwh-&Y70K`-Q--
zwD-O5y`6)yAU%yJCM*o&6yv}IA%E8C1nd@kxIEQUD<@5mC5hhF>s+=HHtxe5?K23T
z>s<*6RB3)s&cK|hz{#8$N<Su4XVySx<R|<zd)IBcGPzxu*2hYOQSf150~Zjvmsz|Y
z>XzS9$DJTQk8lPiMMF(a{1<jH1Gz=`@}W_k_SX3YQIm|31`Sm}${gAX;w@YpUEFus
zhIT>zG+=%U!-6K>9_eBPTSdl%f%SBqa#R{s9gmqsep2e;{78&&YHsW9krzlzCO*6v
z>?j|^avdCPi-P3we=7j!7=4_@WxZS9wr{ITt~+jxlJs7j{I~x7@&0>#_cr<`eoTZ`
zX#0=^Mu1JcGQw)|XN=H30#Y-VxAtBD2d!eqW@~4_Ak#N1j=$cC*$~CoVM*J#zeHM3
z5*Q4=-*3z!f9(7Du==gt?IaPYo1%TjHZ7~EmFsDL?_cppb>r>Q2@ZTNuy_Qp_=eJ_
zDc0G4`vmNvh@9Q{oiFj2j-h3SE^~XJ$u)w1Yb>6EqT<_1h^M~1e~uM*oE{(pQH9dc
z&OVaBDCt|DYInAOG_ke+!c|@?;#B+-cLLO!C-r2Xa%;ao<Uu!@9OFraAOSG{)RXHl
zH9B-Oas(91pFXgdF%O85;Y-{8+d~_|%DZFc?n}1@o~bX<lUM;e)~=H%{%faTF93mf
zosa9Hw#oom1Rd22naICbLBd`D4q;sRYSENLWH(7xK3Kh<UV_Ijz1U1Z{oGwLj@eN1
zqCL{_1Q$>7Z{@wyA?$OC1vixw%sWQ}v`-ybcyNW@tyu~09A3BZw4xN!Pm8UL-VuP<
zIXs&=oM;xV3nhL-t@TE|*=J;5slj$XXR3w$_cw~eAfJOOCeD28`_pFFEhlZ;QKEC~
z?=wI^d>$5RWV`w*1J8RDaR%ytV4y6cWrC{me~+@92D*G|&h$=tkA<L^QP;~RO8UPp
z(qH-tEWc(hQ}bT({ptMB0P-MaxPy`Z)1LBRPYi4MXSgJ>g(!31JK@SR%EI>>0+ujL
z!68^^k~?jAB$Na%!tfC6;1~vVIfi4x^*>)`WYYR^K~%*$MmhdM+!x%xet`tk@v0<;
zdXo3`EMerAe}nLd?11#dpRZvSvxP&e(sS=(UApV@xKZ7X`&W-#LyNE>>>m;wcs3V)
zjh+B}In_zk@?S3%urBl=1S@9pB8}Lj*f=-m80r744PX`nOxAx+JA9~h=8y~Bgb-kH
zqzj`r_%nnYX00DQbw(MK$o0G~rO*Esm3}GGzsh$jvvEQ4S5rQdBK<XbOJIBqV?+Z-
zY(Sd+4x+a;z`7=rH)1#I)nl@M{dC}1FN}5%k6u}i+TN9+zX87d-t_GIV_&vJ$lUua
z6DLXPqEP(r%rPsX+D5<jd~-5!q3OPC2hH34_T;}s1@n$4pokt@VSVsZ6=`bPO@HR7
z#Vq1&KG5-Dd<0?c>Ke(v5rH|&ke+KJUc9@^rz9#IsW$oVq(Fuy)oj^aposVXKcVKh
z`E`SHrbWJ+!v|aKx-n_gxJRN`U?>jisCQi5-!^TXNZ{%KDes8LTs(SQ;0osaQa(;9
z`}cTjIb*f3Ei>2wpttZ;t}j1Mpt`(Iz>X_J1_s{c7zW9|{|@jGxW;~~n)+MRPp^?x
z2Ibv4jzA*@+&??*kE+Bo_Q<^~6mfI2>+FPYz-iK~!V~pHIsE+pQUwSqN&!^R`PS6<
z>38vdRna&a;p}UotYU<0Xf;0IBDA2;!-f3l`ao+w9PaHjJ)NB($=BB`bhhDd+yNH0
zg&ECJ`*v%1)h}}PO2{>H9k6SxpBh^MQLTp~wKLy+*(%dSNS{x;*)&z}=O-7xB0aBg
zOEKj^<@}q-&d7lfq<f_O={Rd8Nt5Di<mY`FZ+G$fi{YzjlbR~)PmB{R2ia{O4~r<R
z$Et;@c*!j*ruC#)r3dI02ShGQ<>+m^5|&tfW#Te2{rk+sA^IX&L8SCuBtykFU9E=Y
zt=PkVIgJI+68g?w<ki!OE5)a8k04#bNx+&vw`T<6Pd!}}b?1=$8@FJU>}xDSO637M
zgX;S|^{>3*()_|!Q|+udC-S6JtIfI=r?zX=HdGAxR@QWKvSuYMP4@h_C&WVzNv}Hy
zH<P&xDPh6h32U3v7b1^d5~_9sRUgwVTz9_#lucn)V8``Yl|n$*!X5n0)GfZgta_Q~
zL8EDJ)$9ReXLM<18ns-m`apNc!?ND$gDrRY1D@bLE01Q^%<>-^bMv&EJC}*DxbR8=
zmK$@nHY2haCt^Kc8~5{(h)0VOM<i)*IC|ej|Jfy6#0;48Cq+Q8S@56<jW)DRW_yvf
zDt8NzyHUt_t}UV>UehWUTILakSrO!cg34Qc+f{cXuilX;+UCPNInw;a=lMTc00Z;h
z^xO=!fkOM3G!139DT$=+f96X9*D?R_gQ|Aizhm{`wG1XbMqptq<MABNrFZe(*^lZu
z<`0>@88pk*Q()<A>RIjSygP0ngnKb!$1XjOe&bs3=BuD|Bgd(huXbl_4%IdjlsCF{
zhjaDraj}f~y<1%RcdG}eNRuor+7t&XetJX5lgbX9S5SolD0qeOIXGU=aSgM)?N_U?
ze9|0W4K-qA$fCKYlHJ88G+tk5WiWE_hRQew2AP%2-Rd?OVtBt3=Q022%Eky7FivMl
z4IwBvuqyT8ku(pON3Tp$98h?hz(%j$W8-j(Z6k^rdW{STyX;m&^!>XQu5k03Y@9D=
ziP7qp+ZFv*V*}?@uDD=+;-fBDw2gVK`dgD@YNC?KXKnu`B_JWJ#7jQM6a32q^=M!>
zamhL2w|N!#wJ25OeZaan!0I+@#ac5jFlb)j-TICs;vV|`A4Rf%C{Lzq$)ptW-fsAs
z&rZ>j3=(@keGaT*J-@Rjd({n;0_;w^9El*btPv@ft!Rlv;$pUpLuAnpphOf07-Kx4
z_}WTJa+Y6&idRJ2DErnmA!2#|LHx=Cx#((Tx9vKU9uM95NBo$}9_W9<a#ubH$1H0v
zUh_SITrIrl(P-Oy#@g%muaY2UK#uyU5g45vBYLLnvTe^8=Rm@gSBaXN5PriK6rRtU
zw!3=G`?_v#MEa4rzSN%P;~2+)OJC^K{N=^v_EW}Q_9vVJ|2lsFYJso>72)liy`lAY
zY#*I0gEB#w<2lOpmISVWnrYAN{<ZDVv?^RR%4+NG#DGtMhcHvEHnr}%=`j|p4Odfu
z#)|k!aE6!Gt2WMrrTcDgQrqPt6JTUTZ)*d7<XK=k>)j_{ZR9tX14E>L!0J3ld1>DI
z#k7sF42UtS6H5qgF|sV#D33*LmC?kxyydB}kWK3%r-E?&)i2oNt?wmjb;ddm2Z(2+
zk34^z9WBb#oxC4&<@y_4Pk@^ZHRNOrB(^w<SDcBP$Cy?;RJ~NwT1g~DVGFz^_LJM?
zI_o@h3$#Ype`7)gdIp$?+lb!5_)TxWo68YoI&twwf}==Pv~$dSU!r4Wh4E<@B3LI#
zZi70(5mGZmiA^uErjN3%Y$%Jmbe(oN0gIsC3~8%>-9=<pk<2G0yftwJ)89+4hZytm
z-Rsblyji{OSC>A=@Y?#{4ToSW7zit8r4oj^c6WQj_&8VxK$Jbnn2nGw!+SRo8Q1lB
zg2nqlSBg5w4E}@zjode}=6YLm&nlKw8DsAJ+6W`>KAWGvvaNb9l1n_^ID#9}=4z;e
zOA(H{oK$c<9pF}3+^*6h&JlRp%oQO!V_e^*>7C)y;i!WVt5&N_pZq4V^my}I+Iip@
z!_ENKXUSQhUZ(4%<>d^PzrOMWjFQsxN)*k&y@je0x~v}*M4F|*V&u%Erby4F`16T>
zKXEfhKjXSGAMKct_*3aG;3k-Nnz|Xrw$!FYyg$;OVEEQeF`Ss8_}%Rn#&vL-BfCk$
zo-1v}njl$}$L4YlrU~NJPmN#^%Q?{|bR*XiE{Z8Hur>d3<7WT~R*C(T@6)O6f)Gc=
zAjrxukybKopxfGYg-uvy=PzZ;{o#J0FyIjx{{U@5@mZ-eoD{b#m;j~;8z__FwlJYz
z7!m0!(?uq4+j~9Gn4-XCWN=DRGmDC<vrx_7Hen_s$0%mPQ1PSQv?|xDuPyJ9=_TYG
zu@SwSsD=LGiIeK5liktLoW$0~FbJN}$m~E=Ij*sQTbq=;HC_1XASO5mb^eTdt@EwX
zb2Y5%CXgl>E+>sHr~LF-hsA`=b_V{D<!JC6PUW;UKJ9SZBTbb=QVB$X*GvgtlVpW}
zg}zmg8LeeaH*L10*O<r=PF^XS(JCadR`AVj>P@(+ra4$v`ctr6?||v5fj(yS;A=-)
zujxtf5WyS;L-+3PEPT+0tA73g2k~u(NjL?f7=N=NTBOM}UX8~}Eb=i)U<Il_CfiTD
zD?NJ7jCL`n6xzWC(i=&1+oi3Nlp2i4BXKW~KYKjPt&-RqwJc4yDgz6luNr?9j!Taw
zoJpZC>9+Ech44pO53i25gwGHH5gutMF`{?HI(*UGyV6}nuWeZV-Bfb|=1j?7;UQcm
z;=`A}_e1|7ScYB^8%CHu^-@T*pe3M#a&sgT5%7=tD7TW~dUMs&`9b#bn75*ok$|Fy
zy+N;fjp_29{iPJ(YW{)Wk_civ`ocls1|svqmGz|Cn@>?`0HM_5!%mXGK_HV@`bLLo
zrpYxrZKOd>nG@H7p-^F2m0M(~pa>&5m45XM;L_`>tTm=x7lmHFzjM@*H(&U-Qj6Ho
z*5F)w^&h~cme)<^;keA`7n-9t_a_^t)}570U_lBLK;y9lF6CsKJ70NaK_2RXfa}RX
z!ERx;>l;f!E@w%bBT<My>Its!-J-71)8)0h8{}RIF(jVxCkQztiQpgQpb_W(R2XCq
z-V@?sEz3ekQ{a75HOx0YaETBP!fS#qbJH_2FZ9?8w@zPFiPxSVXggZ|K+I4(ZEz(F
z(MtJ~!y-U^MCj(VQN>rEVWYjvWR9zN0<9|XIulo30{OuKd8MXK!PNH3E6jR|SD?%n
zztpt-qTRiJCgtcIwJ<8B0fxByw7wUeWK!j&0S4jnJ`aTA{ThwLboDwdPM*k@K&yTG
zk#r%<G%_>VP;+eNaj+6USW6I&J@s*}OXFfmx1&H59o;&9UK~1ymJDM4^J8bNjZMW%
zg@94`@Pcj*|IP%1%mH0dmc`AKdxYh>AIHh55zK_U*YguWA<76z`UC1%+|`GcIJ&vf
zcj00^+xOc?I1OnE+%`$1t$n39fp+t`j1%=+ge#^>Pf^$L-o07lj^jg?ipgQ9o1OH$
z(vQ`r>(2L{!mu{1T>P<ozTeVT*u04rk3<Om(n5H<#5765VkJxIv^iI!ZvKY2=9HQ=
z<(GIq+Nv+oRwljNUV{5H5tjI|`F&Ytma3=kYFu|7WQ4S7#&<zvpvtnNguH+m`k?`X
zBh<Dwkg>j+nHrxmTn1IMjQlc~mfpd9P3OXC){3Mgc3HE)!ODQ;7qqCRl%*StT<Cs=
zkznn!YBf<e*3~v<haEW%Ku3uuoUX&<-67$zelpxqdMO^H>pABo9+P_B#WCxz4%Lxd
zXFDqij04UrkK*llxg+UlB6Ut#<+c|XsSEDsIsrTN#igmcqAXXB+&>C}3<tcr*yaq)
zP)D&KeZXYueY$h?9Ach&DR-uK)<Cy9EwhdjxHtN+D>z&&wXI&;2RzQb9Kc+QD9z7P
z4&|m}Ua<tW4Lr&PPeI6Vkga5CY7VI?z49F3R(v4MBE1$4&f=Y|S81B9JvR^&R{?4>
zvX~#FgBL86ul`Jt;}pIHG8lvnyzu2X)6v~F#@n5XbC}qf7z{Cu0{fzW-atd%d>A(R
z)vK^gifQ(TbN$E1kVC}T=xkN!tQc-lL~6%I0}EioiPm<mR{wXsB&bO^AR=AA04|(T
z82-`#5G9`Jo<s-%!p#92&F+^{z8V?p0(9-p;LCgnsisx@xrKUh)3Q5&)N!T;xwV<y
za_Hw;w!I@RT&WP%9_O$fh`W)w?G8AIK@I+kiU3f+@KW<!!!2iB5*yEh;hn(gTg$2V
z<gPtt4jY8tNWsXBGEl$-TbOmj{0HSu5W8grYA*MNJNi7uUjtYo^ZMPBAMDTPFrt{P
zQ(?K7s}9Qmn|TIT@^;lep~aWS1Yoebh~C;yKsFs2`|ZuK3f*oksuPgJuU{RZ%3J8s
zNh{P!-*~d~&onxNEE~u)wepGPW)C_o1;2pH5M;>NCj$(W_J4nI3IEe-g_$uY8_Gag
zzeD&gTS-{ZP;eL96|kieH#io<=gz)=3cV*o&IFyZIul1YU))+QR}0?)(&TL@a9%Ol
zc34gL0vyx7r7PF<FF+1z6sAdb$Z%$hXfEc!JsWokEx+kU-8uu}^E85CrrG~m*ZkPB
zPpksb?jYfr=@)c@vu1WFPjHW7jTRwJok|YX&#wv1GMru>3gdA2=6JNM{_;D^06EMp
zU-U$<>6AX%l474Sz5jS4h``-dufB-+B(@o-jBikVtrqP%0^wR3(Bl&0FG;)@yZm|8
zB3oY!XVGD+=U|sGPP2~}Cjw0BJq&0ZjDdAXzCQ)pbr75cU~Kn<W^0f?bFV~e9~~XZ
zs}p7{l<0I^s^?Gj%&BD(e-Lawc-v01=7)5Zb2S1*Dd4xzhISHr7m9w!;Nz0OubZnc
zN(4l4uSJ-jBs<P$aiWWj;!DmJ<o?upCa(exYFhq_!q)_zimg=wJj6nDK%@#TbFbHG
z=`3P7DWz%YtpX%{@_46KWbdnOX9I5~8u;NKQ2;)y17r2Rty}&*Ae=zGKL>&n8llG^
zW_Mh1@9WuCWv&$o+4}*C<IW^YPorsXPzA8Mf>X*55iRla`#l17ef(tM$EZdbQvTUz
zpy>NUYdMUG8=(HJk<+%Iee#9jO!FQg(*3!XLn<&}0ld2G>T1FH)Pg<9xHVhQp*Tf$
zqwH5*jGD6JO_rOtt2!qCJ7O#(fico5qId2J*K(S^y`u|NG^eyCUw*{*w6Ql`F8@fG
z2}n`7n`qy-lv^Vw$6mVN0~Q%ncD}j4O%fHbpVnl#pjAVm_K&;=v-pCiwr#B|yjS<-
zKC)mNoVA>L9UtL@n~$PdduPp<FHn7_kpp|nOc)z|!{gwAGAs1n1BZI3CgZy~*HOVS
z>MK6_m%{>!`2e<%JXlcmFm#NiKr}c-6DkQ2<9_BqU*Mrb&gv-Ze%>Mog+k%i1iP6E
ziW((WTzyYz!i>QQjt2f&tD{Y{K}G&pybU)%<u_T<Y7mLgHQO<uWPq8sy-f7yi}!mz
z+!18#1cUroC(jaF)x#dI?K&K&e7ocWd%FAOG|LWw%27k~N4dizRfMbi_8q-POD-%f
z-Dv~C)TiA?*&<hQ%!Z3oRJIM23sQ!a!8=%Ba9pmX^Q~)UBo3mWze-vi@00;WXgM;!
zz!v`OnCoQ58(w=ly?)uRNeWqE5ch>UjdP+U`x42L{M!}gdgJ_?m$GE>z(%6CaO<+q
zcWV^o7l@yMu8UnV0Nm=g`I+o$eWbhGF*=YW)*ls7K43OjTKT(^+%v|&0wOt{#i*n=
z6K<w{<E>huJQ<NWi3`NpA*e;>gRrSncb{umB7s?@<9^HAT=Zj)`sv4FyP?vKHa=!V
zVDm7nKkL4_?Uh3rP25Mg=g_<J@rF|Cf@C3@YPLZT7-&-SOZ9@_;S6Y+U^YNU?=HQ(
zpkFOtJUz+z<?pT6(B^{_#SiSz!E13FXO$i(f!;gSY$lAeD9*r?UdPFyTP}MQA}In>
z{cm<u?@-BK5{(15S1iMxPJObU^xBxz90wCQVn)&eE8V~a@O^+mA=4?6u*nyOArU(V
z5e}!NF`r8&+pM-0_bqmNXv1_7xR(DA4a=_yC%;7{0-@9qh(&P1^XyxMe_06>G;k~6
zLh1o@D}$jMu8(%BOqExTU@)X=k$!?}{8->2%4N@<DKb6_=uNIcy!D8HiQxOb2gj=^
zAk641id$c;U{EOsNu}kO-}sRC%&uSbuhoyTC!oMxJjQ?SR`0;`^0Y;7@X{M;AH2Ip
zb#+&#ZcB-%2#yaSt&duymMkIxFioUFiJd=AfK&v}312R&&fg>HifV$&L*(eqy;(Jg
z_+w;8`&O@=@wxP<`8fzWhDa*c#>W<sGy^*PvsUK+sC6h-iXwB`$_3R7AL?io97VJ-
zl@4hoGh?&?N=(Buox7KhLw7XdDm1IW8h?{JRdt%5YgzV4l#r+#RRi_7?ah2C=k9CD
zKh^(T0SZ2w^&Axg+-QKkK-}wVAfO=_r-5tp;4H@aMYes8)~?aBX9cX)34|y9W(>Co
zN=p3moQ`dTIih##HJ9-^0+(D1`&zI=_m{&~_&|4W+JyM3O^RZ!2vV5=058F;aFdQU
z&SE0qjG`znG<=8lh77kYzju{7(cC1)QBB4~l##>$QhQVh7+vi1T|KI#0R?rn7GbOI
zT}kG5x!Ed=R66>%td-v4y(?LpKo(@V5lutK_7d6VQ-xoifK(0AX3L=uJBJv$PC={^
z<@zFR`(xzs_g;Wks;Aco3_LT4ZL<eg6X{cf*vrl3)O912n2{9M)>dzFD<m}m)pn0o
zpJX@U;b{MSJvn+|j2frvK}*J%(x8NJIy**D4A};>J>jHIvp=RCwx2cOvfdeh2)|(D
zV;PF2j&Gf+zSe;QO^^`lqa3<ZpmwZ!r=BiCvImmhKlS--gbWFOs7-LhQTt2Bf?VBZ
zuxmG&xos%>4kwik<<!)YA)rrMDUq8DBW3S!t$`QN!Bug&U?Z`2`v92$f>;u$6t>A7
zVMgX82m0X-2L`S7E}qI-_O`@KVvW(IeSO<aSLgs<v|vN8O!wLrY~O2ITj-B^1Hv47
zl4Adz=5K=h>j5~MfaSp<7#^&V)U`=BQ;8=t7iT>7)JM3+)rD6R_AMWA>gtn$RPa<0
z*;0tEbxZPQ#-!pW>^1v{A8?;Gm%SJTHeO{?ptjFZT^zdA@+`azX`1}<^O4(Bft7{f
z8E-$F{`yQJuT_O%hS>;WG^6?=y6xD}gc^hT+2Qw^UGJQ|&5zMZp?a+)_PG4P*PL)q
zt$fM_L~NoQGv-FKL(Bn<{*^nI+UJA&SVtI?_iie&%$DCC?k!oNz5FRZywnd$e4z6Z
z;=jW<cfLuUjbE8GumJUQ^mS<EfYonDCBdWc6_|w*0`af}X}iWRd?@xmK~wOA1wE5e
znpytRN3koh5E%|#g{V~R@7LvWlA6^7u@CRjfs$UXHWfVOZu_{$14lu^AN39Qr=o(H
zaOUmKov+HzAlw<g?_ji@O8ko_*jF)s&%y|40c~F--f&Hx4Z2%9U)miMb<rlthPB;G
ze3M@4dSYRJ-`D}o+JYDDYnoB(tlVa#v&aWL|0cWKbrf~^!kb7xgD%N%o@-F`2T+gE
zb2BMd$z1w+K=U-e)GkLv2?^1RiqbKY?j&1!o#Ei68FBT>Z)&s|(Wiru+XGA41uvJs
zE*8BEnp%Ti>^08egqYfvBJZ>-m7y0$2ys12P+>WUO50xQ3O{y5AfOBUqk!0_c6TTC
zBP2K4UA*FcHf1|bn~7;`Vrcc#f*l7Gk5Tg{-5c3UdEu?T&7ftEC^NCo3~7tk4@<mf
zYdqI#E{Q#E*oAy4iwz<!Ha-?7a|T9DT8~QmwDD{AlzPz|bGx#k$2N4VE|LByHTKg~
ztJXS+pvSw;$G2Y1Mf-o=*4jzcUwW;41uEI2KZ07b-n&JS>+?SEm1}bDPI@WtG#(l>
zXiOf=%k-1)vg@Ws*unE=YiA0P>Au|0{W=OxT@gI$@m?9jc!0e1m2rTxMl86OP^itD
zp;r(W!>(o=gw1L}m;6Ct^3o*`lF+<hjg+P*K{Ryjc<7Bp)JI1WF_U3Cd;dHX;Y`q9
z7U%Hc(u7ZJBO_GaPo<7Y>zBM|nq=KKMn9{52=D;A*ly+@*yI0DmdDLu1W7kvjH?|~
zKRQHe>;u^gP+5-aCh=4v`&I86T@00g#mH2jEq2gHQI{Dr_0@dGoTtFo{&Th8x)0u-
z6MjPm)2BSwMm=WN<)`Py{#VO|c`U+(pt7T>H1i-mQPJi44e+prYT_<&S%9ngVXHXU
zw~ShFX(5La-exXl^b3PG7*kSJvPg;FR@_QLx)&_5ha4#hgO%ZA{gW5;pCHPw0!Jj?
z3<oD!6+&vhtvEOak3O2B8RF^(s!!j2@lcn5Q40sSdFO03LQbd4ibqlP1Q@!~0~0TZ
zdDOqqQurq;C|5iL{-G73PRtkN4AFOk`?bFfiBQZ^Jy8TgOh|W?xJ=eM%vmFNtTfHZ
zgkJsxRIv*!i5=Nxpm?o6M{B-cf{S%e{+P>A>78NqUtY#38unSylJ_Mxz}2L&THm!f
zKaQ<^a?d1#KO2l)#}jxy{1J|m8&_~f)vhuYg>eQj+C-|O#AYf~+^q(PCg~wy0!7DI
z6Z?^ejdEO<ZAr6&rC)peFplFXXm<s9^gr+ikppR#X0J%cRgVe_D~*{RLy*kpq)V0@
zXn;xr9ZL!AEzBlExYg`(l1`%<wnyqIiFm)ew7^=g+jS=G_}F6ceHbIu%_vGVdcr@-
z1chEBfsHyha9L&3sSL6NN<YDXZ&ndD2KjaccY+k;eNGR4pSoYm((jLZl=j2HW#vXy
z%qQ@hp!tf!2zDPKNjfp)Tsoa{3-fC}>}$C}nt=(DoD(%y&cU8TQh*EhigU?yV^fd9
zDWGub9L>ZF5})dcFAC|reOEFC?)`kJclai0J!RCGRTX0Ky6w5!Bydx7$nax95zT>B
zF=f89lH7k@7zRuZTG{6Wl_@cx_$BbQ?*X40z?EC0sunsA4g<o&*JuGwqyIw=t;8^I
z_0RQrpYTS~z;tN_sdqBVp^e4=r5|6ldi*8Fo$<|EfPQN<3RYq>>nYaCPY(RZ0vf3S
z!q4NL5;8pSGIIa)Y*5}@09HoL&dpP2=w_(9!_0i>&JIs5C!hJWw@{hTzQ+{XrPqc~
zcE{45SWmtr&(qkvwpfVe)%YcGcAiabhtQ6dfLe$+NR&TFUD~?!+d&cPlf_OGdLVic
z&)D!wqWZ=-86BAC@2b|O%fCCw1&||*14QBY>ROof7$mF&^r)ZW{NpfCb0Y+T&Wh^Y
zmY<r|M=J)9vi30SP3S=X(cVd>_*b+iyv|;eii*63RRJRAsrVm$tlr@pTiEc#LBI)z
zk#xyLZ+w0Y>Akz$AQTAEq}W+NX6#Vle~+Vg(F^@(WWTbOZ`mgn!j*S1{Ta(0(YeeQ
z=!PCqV2mp38R~oV+<~*usY@pjB)9~%pV9?*;0%Q4kIwf5!j2$zPF8(>wm|PbYY8BJ
zWO7`}6L-(xnDq>@Y!Lcm0|;{$m~V{r%ersX*Cj7k-S?zLEa5?-bc6J-WyqfmnyNC#
zRuPm5`z(uRHzN1+i1MC8C{@UIFN@DEiESe{qA&3?W;#)%OZHF<9d`XxjP6M?V8#a@
zCj;;(p0gjBEwBY4JmGIr+x;&vI?j06IQ{crW1^g=dNT0=`?EK>U1eNyILuc*F!$F}
z+mbiyR<7V93ULWq14j8OqHd)hyS5x`j3bVp2kX#glAWfN=@t}KjYX2DFrbroM5s$5
z_5Nr9RJ|nlv15reK|M@fJG1LO&^aZI3nO8ld?AbM!jb!*H{NK8;cT=D&Ii+Fo19>c
zQ3XD;iutNdF0Ag(K8uSO;h~U*lX<ByZ-IVpf6#<dH03sF)^l)7R-K!#_gG$$Jxbb~
zK)k};B?g?R6fWg+esUnM5MmHEr3v--DD_dw;0v9(nS8iCWU#Rc;{)vo&lCev_#c&W
zQ>h!$^gliRhJVHb{i}@ZT<>sO!|Hq)P3Q13h}<|snKFS>qAtCht&z*Dj)SEF5kZGW
z{%M9E@Lq6=Xlg#By)&i4rDH4*N#GOzn!$vj?k%M!ER**EmX3eWCWUyk=V>1}U9JRR
zvA0kN$2XZG<;6MhoYAt}pBW*0zp`czDT9B<?-e(A1?87Wbvl#i4_DHxfo0n%(2czx
z9f*&4$5VIK7W0k}k}MRR{^IHWnN!bBwkvYumU9$VS8TF~Bo()xe(>*E(7(XVjnfx?
ziY%6PUfprfxIXV$X2n2Ru#@+I)~rWVbHSib9FBuI4oCp1Z`aQJ7tyv{8d427nr(yF
zC|@QH$~E4}<1V;`2i^6xePH$Mlw{0G4s>3;z7?fRVtKS7=6SN6VhW6i1%A`KXllDg
zmPkn$w;b^N#D0p)9<Ptf1_4ivBRG{X*frlT2Uu!7*G1mS$<cx3v&xHSdCwuRLt}TU
zoZ9wF{5b9!k^xmeiMVds{jzadKBBOLpbK;X<PQnX&sL%TJSuB)^s7t0TI*v1`EwZD
zVkR86uG&9|JR`kv2cKeDwnEAMPN_}=rnLWe;j;3j_nLmRPP_hk;OCYp@YWdYLJLbA
z`u^^_Z`Q%v{S`ka4rQqf_<4oKenv`&rbi%l2T4ao{3_j#<JJau(0#=cp-$|u!f_D}
zRldXe$lhv!DatA<%5e$VV(gua0qvUy@_sQ=6OQvY5i)g~y2=4^|ND+t1Sfv%Y*$A4
zYGcmjkV{rjY+?N+?ZQkFPYQB%Hf=2fBc}xS+6?z)U$)AJbPIB1<bt<gPe)(Bv2?%P
zV5j@{3QP>-$G-AE5RPyhzW0c4#V{DaK<_3jReY4X;=df^-RCZ+vHDbIY1FR3+{G(A
zIO^HwuHn>)az2Vtusn))S=lF~6mv3mBWOBg>~wsZeFwy54)g@-YSoIRPL)FDlgeQ3
zRqwXpCqZ*dIgSGHesoPSKB_JyvTpUL4ieIzp*q7k$t}x5Z!zdQ{*G;13E84<%Hv{n
z%fggONIY=*!c^AoZ+?5|T>s<F8rS3tsj$^&chXr8oAeK%Q_Bd)H$=~Ik4-S?X^-V9
z*>=81=|%396t+^yGR%sDtZ7|-ZdObp=Xdi3K0q~W%UY(Ve9onG{2%WCU%H=`^V<0N
zdBFF*rdbhusNG6DxobNbYbLk`J_=rJ8mZFQ-<6vmY;)C6UinoRU7hywo7Bj-yyGa&
z4g+lgsx!%AI|#rR-}V7U^?Bx_t{>@A#}*;85y4?IM>qxlPMO<A`*TZonMMWwf{>tg
zeVt=Lx*apxx>2Y?$mfm1#Cfr6z!#7|Q60ChD|Dkv-xWdrME0k2V1BxY^hdez!FYe^
zYo$qGoU(oKzmmcUHK20*Ut5=^r@-keR`d3{#qQj6NfJD4;{G^?^61rrIb<#I8SF<H
z`NEcE9hy%_5no!SK3|JzZOiTY(NDMZJ>|n2kB7|1$Yw2EHK~r$bYfx_qN|_TQ1_lR
z&O~lzf{LYFoFC6`k`tKY@lhVl+PYTH!E5z{^a2!Grn4U0aV3j5^8O5Pq=DK@xBJam
zzCvCuytoi8BT!oWbozDz%y)gDyY$23`UvjY_JBjgLo~rRZ${H@%2<w*U0w$2xihs*
znFD*s428Ngct>Y^0zpG~qnX&j@XKZ{O{4j)JMaL5M0shFKQHjKTH0ya`FUY@-;dYP
zq`kIgHCYD;keJ-if`mVsal75uo-!U@J8-zcB-%JJNh67ln?7<AYy!eSf?~R}sp_pb
z_g-Me#^GhkOKcZbDl@uz*lSygEWODxI0M@bWL;~1quAulfPc?%W4Qra@U^<{J#iYa
z<nc3*_+)0{0bsbMyp$C`U_1CgC3ASpLE_{A<&ZB0p$xD)N&V&4$8cHr@%<$Nx@iE4
zwG&z_92bBZWGaL0OShPzBzw03>k?-H7OY5E5Jud62i5F&toEz81Y6?;EVTk&&C35?
zx&}pFPq?)Bh8}2?e}C;nj{G$XEO=8KWQGOF2ATS69<4oa=*BDX!6Is>BwLBSga3At
zDe@rSzAM~ZklXC^?06b-iRzFcJ7@4xUx-SlsJdf0>NnuWRD6B6G$xs`^LRw%sYR>n
zJ;Y)lN#m;0Ag}W$cR>E{yR&h*4>K=+CbD^K3!Atqmg_hmI;=I2v4&L``*}Br>Q*#>
zKA<4)sGn{IPaB>7wxL1uqz_ZWDT#_Je>M?uQ_}10O(pro4$kC$Owx)7BDOdTN8jV8
z#dCX1#iK1q*^8k<gp{(oh9x1%G18m_eiP;9ks(p*(L~z3?gy>e<=rIzls>OCg}t4_
z;5<wTr(0!iVMg;h%kpxo*>=Z^#l-SkmV<O^@%sDSSC72x1>@(`%v(d&#4dUc-lhEt
zhnSO1553+ez@?ZWNA-GNpTVY7^vpQjLvL#iw!;|1S>nO?ma>s%7u#3Uj-FNwG8c;o
zThaQS3odQe0CmNIjRTMQhd>Nq{P3_Q4=%u?Gs-_$dP1vJe+DXYmhsl+oKb`-sz-et
z>OMR+(y=8Y<|IWO2zi$lv+VEzTGCXRnv@~V9I9z=;OxAb+TJ%S{0*mi2ho$}5eMR%
zuduf`{lz6lUi0MwAuC$VAkSdAk2VZZf%c>B%Af?RIvwM%qnwQWAU!F}<3|2L)bl$k
z%|UQbmoyWem8HCzcszy#_q^~HvPP(jTq8Ri);Rvd(RZTr`clk&3zT%9PaL8gllhEc
z{$dIvdCaT#aBRcixw!Sy%H5Cf<w=eosyf)4`jI!c27fmW-SOf2{N_ARDh^Mi83H-m
z1Ek+woq(mo_Zb$EFYT<&+_bIvRm@YAtMEDqG`~>~8Mhd?!UK+0mw-v_s+*Ti-Y2^8
zWz$}%0TR8h@9BvrVM_$^%3}IH=T2f_EHDvEC@Sy@BkVbW?the}+f?Z6dx*(90_@jd
z)4o2s-V-gj`XQG3R661B%9o$6|L^g#U^e&ClSGySF@Z%-5GN(YjoLiVCz-1s2tv@u
zR7Akl*}!9W1=R;W6Z+#rbXfm}?pIexJjvO6R`qkINLJQm?40C`P`staV{C6m5a70D
zkSfHeik2C!#Qsm{$Ips3>$IdD3zhE&$D*MwB`5=U`w@Y;ydNY)*q9U4)HUzOo&7d+
z(D=jP^)(i}Z*FfB_J8D67j-s=L?wPW&Yyz%AjQ&{b#d^{cQUAIQ@EIEF^lEvENjy^
zC}|tsV-5DBJPPP#b~<0EUhvXWFfXj|d=z(z^S)^2ae#u0hoXKEttEI2&m=w+enSEE
z-8d~~q&B}Rv3i~+e^|NzTlZ1XyhNe&G^l>35Wba*kn#OdgRFf+W^*5t%pbx9_-WJ%
z;Y+WaE&p{1f-6tC09iL&lYK+L%xAfS)3M5cOiC=rqUApf=7K9^yJzxh4tD$3Qnha4
z-mTr-sObrIs4~;Z;Xdv1xA7B$R-oL8U%C5a+F_!1XpM~6Kp*Ule$|I_(=Vp|6z=Xi
zt@GzUroc@6D``J{@UOd7QK52d0^njgsN2QBx;0lQ^y;yb3>K1XQx0@h!(*I4RQS|e
zgi?7LFz(uJOdHDa`r$bLVMWx^@9=mFjSp<bO~T#DNr9H)UX;(l^LMRc|MRp6YNy$;
z_>x&&RzET}J*b+WZh2CoPt1_2^<(svFFVl1t02)`g86m53?Ip}pNe?T#$w(JYVSrB
zDVaZgb^qeWU;FNl$EZ=@cN#!YK)EoY#~kNT{u1O=BmZ(P&@R{wn_E~^m(dpu(FDXh
zyN&gQ3GevKqvr%Xk&^-4wU|(1yFcQ=;?(NON|-HhP2u~w#XDDJGx+2fxXxp9QqYip
zv2|_nypv?A{>@imeEBh@-cgML$Jjqir<0L;-q>$cxBC!W?U0)YN(0;ksBq|qf4GVb
zall(>EB*StPSr)GwYZLegbRa&|Ir;g!;lmtO(#0e`|gLIgKP}%zd>@&$@|}dg7H=4
zS>`leA?(Pq3jIgys_OdmJmHOs(&kQEP8C%i3iO<$*gqel0-~R=a=^LW@$6XO!>@r<
zmJF0%Ukr`*kA7V~88=#3yZHN$6BG}SZV4@8IXU*3Z2;1}Dd~s~5MaXie+$<@!e7#n
zQT{BMzbkA!4xC@IrMRW1JomIzq?DxfIk5*xO9E(t08hOB^Q@)L9pE`W+&Af{cm$uf
zXgY=k%2b;cKx-?}e1gHlhJkW<K9VDW&u%s%H(jmK`A@L5tB=w-ZL|?pkINhfO3Jd-
zO>ep4<{>@B5o>$Gjvi|Tm~!4iKtP#0&0?Qp=_%W<biJSXv3+TndvQ*{gvgfoZr(+I
z5Tfbmry&Fldqo8y@kSHpfSM3;uYdound?8duI3#-mWj9h92nV;Pmx^F9&L-!=cVq{
z7{5xyP5~u|_j6STM2R@&NR~qPPY-9qyPEntL&yKME@DOq8rVVsB}Dy<Jb+mfR2V5I
zIv<BV#51;OeT;MuasKnNLy)-iZGk1L8hmE%RR(Gkbl^qK#$_`C#K*=4R0eqEzrfoP
z5SjDkq?w03##XA67gjW*2BxgU!X6*a#GDIuz0}x@>2foK#8MHfgO+6($CD-ShC1!`
zc5bz?i-@%6FPdaHKN~D0@qkNb&yDdxS6kxX!pPVMewUZtvyy%usB;ZV1)G_@C;jjE
z>Vt|pj-a=quF(fKa-XC=P9SE;c~n^@K8VdIv#m~nDkZcUKnqmg;c{|ONn^}X#H7~e
z$D0Lr+hWQ)&~2vwLJA-<@0tZn1N8^Bm|uEeH@yx<2L#roD!VNVLk+PXe8R?$jK5uL
zurK<~oS74dKkxlcu=did>YRc~wtx2$c&}o(ab$Cbly-E)m9#uyuq`d;Ki?W%`+A{t
zwn$1j{-Oe-4)&+R;0SQeJy=t1AE}x?EbE)E@!ix`7GsV@{5&!^fLBd0vv$kwgL5vZ
zH0#?8lphi{I--jTUXH7h-S(`|!vs8VL$oi*bGUyzd6G~neQQhN9x`*^zU#E}v5A6b
zrQ_zeT6X&w?WxRajDuQ!1d4D?e=+vpc3drQkI&>`^i&RGNx27!iXEkbvcDJ^b9Ten
zkIfQn6(#>3XLqzD{}<QiA)jP8t@{=UHLpl#_Y_!Me<6yUfSfF9&v5Si6;_aqUnK=}
zgJ(_JRq_Wh9{=;K9L7RP5IE^sk;#?=GOx$o`RPHSM=5tiw#054Gd6Pvs?KR|!wr<g
zCC8W_h85F6HNEOzYo9BC`$Uee7D*i{0#v>XH!nrRggw=3TNoBIT8Q}ASr9l-tAioj
z=v83r_^oRu6zLx=FQlM;PRY`_Rj4*@T-H2VWCrWjqw>n~4c}nBV`qT9P)d~%beP^~
zKJzHQ$hB2rq4%8gaFUuR*BZBt&R|m~vU{q)$#1F)*4h|#{@yr-ilBvs3!OYz0g0J<
zh4H#mIDlVPNWoupn2otCr4W^cEr9}&D-mEAL*LLO4|h^kZrc?MdF(B{LAs-k2?s$f
z)f2%46cp|6Gu3WLQ|Vl)a$3A%mEq9cA48vEq4S*PaBv2w-IZ?=PVTORVY>j%e11DA
zOUWF0a~^cs>3wiS#PX}&ey9D;rhZ)pde}Xn#K>vm4OPBihc#C4vcl6mef9P|s$mBQ
z;EIYDGc~I>kG|s=Is*+rO&uy!A+jv!f3gOM2M{JrauYyG=O$3=n7X)fh>P$@Bh={a
zsyh3YhnzFL3WrRFonTV-(TxP>cZdaBRBm>!{qTVjQcF0Gw-7IdY}GXX8g0?EfF16<
zLjbCeyU(x7UA*%Rf?&7@ykqx;&HP$y+AKnW`IveWght)C!?l@JwMFESg}}=_b)zF!
znYi%Qrg;~MU)!d7I07vW1RGFYmiz*$OFg?2GuIW(*OtAB+%!jYl5EwmT=3sBAN$V#
zz#fp?Uz<}t;7?J2VQRF~BuyppjK#QG27ea(!}=xbq71qFL{4ax;+^si5N*#B8{|<s
zjzr8#C<m*Ix8FB$g+k<@Sh_M}hG_Nz+X4gES;cp!eq${vFc;MuJtxJGKZWed4de6G
z>>7&bib+1UdO**5x6QKq-t%jrH}rww{+^p0T9e;}J;(eYi3ftM>^DovFQlq)W9=Q|
zod+?Q*E?x^<+=#R2g6PO=Sg_it%_$KTl8~vMFBC^_BQjPL)A61gwen;ChUkTxV4tI
z%3aYp=R{t*mG!nhGo9ctqMSoU9xu9Cu%f$60M1HY%B9>A<xXStd5R6|4;UK$A6r))
z)`a)<bq*;NCLkz1RJueRB}}DTLXb8PK?Ugsrvejc=~hY_MY^U^qJ*T<lG4)6-a9sR
z{J!shK9A$>oO4fp&N+8<gzgIx0?lr*_24|4bdQfWcdv%|Z&PcX0t^nQR-dj~dHuuR
zTl@U{Y(@u2FW952n0``#3Bx_5J_uQj8r!^k26xbj#ae4c@rw~-Tto|Vz{G3x*j}m@
zV991YOM@{2cie`ZC${sHSF2`Se9Z~gn{*hG+9q7IoO%DgP;|&kS`gf*l!nziRog=T
zKJBv?GEtV4lFPsPwE}0vZBu&eqSyL>^M`Lv{&rcKI$sW+!fxl#M}5bO7OX7KzUXRq
za<!$XPoB*AvRnREK*vj6vZr-F<_}=?w9O_7X4__|ptD&H-yhyk@b-{yiEfrK2n<=C
zMb~j&#M=NlP5$e#%~Lej#TOkY@y(bJ!7KYiIe0zeFEj|fPE~Y{GQ%`)eB<fH>j=RH
z2RWG$8vIBHYLq->o7&*q@1;??AghrN1Fzzu9b}3R;@?>Q<GN8$O8<zv<|nG$jh>J^
z$pRC{CA=q1*nnynw9PDBYSQ{+N_Gd=1g1}}+c()!gCTQJ)yEpTj+B@TUH+fAX$qB7
zmYR4>5>S2YEcNA@D?CqFA<}uNipe}OLgFF#+fc}VjQwhR$oDJ*n{6lM#;}~C2%J6+
zvrk_SK$H|oo0`&eFT(<Hh1A?FFo0>d3;7%oxJAN3fPulltrCY|K`QS&6~__o1a;G<
zMK|Lci?m*$fkC);hVcatE_hNK;r;Bl0M_bJaG}oW13zplQ<6Ct;u{E;X~1U2*&ABL
z8yY|zunI8nmu9P-_fo*SGQ{UR1__*GYVn1{*j=hSgr|4>d7nEq#bySqFq&mDc2i2?
z5w`=7pJY)cVCa_9mj->h<qX)dFuv7x13z={k1`o8jM@Bs@Zf?-%;ZRPjJI~pTqk{D
zg=+3J{6uUC;j4dFL)LEOdN~Pwq{n4tk@k<)6&3eCw!;QmM)*D7bK*a7qj(bnb>TG<
z`mT5MMRR>!{>;*Rq~QuTzO9^Bb^#<nu4PDe6MLCIx<Jsg*Ste*PsSgVA<B*pM}l^E
z&0L3sffxJ6W!MYVKU1S1*|k79tSwR19Q1-yD8I<N9^VCJ3c0sWRKWEg7vlf?KDDK{
z>-P4}6mE(-VI_rgV|Z;YyKPxoZIkU7A?HHrQ@D5A?#ySJ9k?r94XBC}%wXmGAJzIr
zQ8M!mylY_b?IHEvCEBNhPNCb_Y5<w+SsubQNg9|w9osZHpi%<*7{lJ@Jm7uLPVaYM
z)ZdF9@3@`q6|(MaS&+f+dX*iqt0=f2=wfg(aQF`-`X&|0v*K!CJEm@V67{2nAB@3N
zpPAJ2|6h6W4GjWdBFNA7X^9!AM_w_mG)I^yi3?o`!9DuBClUjiNAY0NP8mb4mDFlb
zp`2FvQAM8TfB0J<UpGD5TY_Cr7|XB(SD5Kzd-AUHL`8hcrtg<WwDJU{LFSOC3>M7o
zI(3O3UP|Q8FS<Y55pqKWaeH5p1XSJy1<E8VfA<#*py*w}{wQB^E^o;+-kugyq~I?=
z)<zVlQ=a%Xa5ufskUIR@7rv(v2{eaE!>%b6yiACHq99{Y(0(^PLXinc^6qNZ<D`b9
z95#w876t0E7|5(xI&ErXhK5q?9qDSu$~7lcTGPqZ-#eab64*t_`yO3eKPrq>yG|-C
zMwPnM(ZJ6R=!TZuI?jfCd;p9^@;$n3=`eRN!x^GmNUt_4%e0+q??pz9|6eHXA4z?X
z!CZu;6Ym;Xr;;_hPUzZKgLD&SvC4O+ZE8>NodP|^8Qhj(u{Ed8JQs!aD1r2^ZVSq%
zNHbgM_U}8(^Z=xE;En2$q})(m+WcUB_P+eQvGK3I@ZkxceqCwE-WQshLd?^yPZS=?
z%KhyJEt8;c4x%-<ZvW#9{=ac#gU@k=Vyaw|9c4lw0d6NKGeX@Vwszw~^XGnhC4^Q)
zM(v$6ln)UTw~#j<x!EEnq>Zw2`)SQFym0kl=E=Py{xEsu%24@=+-$>T+Us^^kQ17`
zc8pUw?yGf2g+ua~EVY{OETAq^8pi6!kC67u?2M7y_fk^Agi9+7)M)-3gA*F~?TYPV
z8BZ?xd24@Q8P%|{wcA9%+gKymGSPb!J;vBVyK>9DT$>WTEW~u<nBy?D^b;4P9s7rO
zKR9Sa<9Q>T=)?8939#70zS=$|1n4@mjvgKV`oXNjI*j-{O;5R=stKbnrNc8r&@y_@
z$@<q$Cy@Ei|I7jm6+E3zI=7$WYeQLpW2vh%>EYhm8`6hLOBb|Y8)1vM6c8`EEvRi=
zq@%sD8d#jND?gDKZ0Bhr+@^4uAbvcFS+wgdfWm5>u1A$YkgBVhK~Yb+!i(cQW%?>s
zBQAGbe~D*^p%mk*nPu10phfoE1Zc;{7aYSYqWe@M2?7zz&9lyBHb5!zh&%!E15ZGT
z?6m$$4WgG!b0$cy5n2_mA77p`lWDMX;#U#902<x2lbgk|k2`-E(^o*t0#XXne1Vqn
ztdnxIs8tJTy59b9+fN?PK(;npju`txuign=*Z#^~@~?&T1JN7_1J!;!Fyafvyn))*
zOd!r%a{6ahR~6IYj!TCl%y-$~#u~q)h6x|$rBj%U1_q#feeblHhk`Szu$5gM?dSV$
zRE|qX*Scey+4>pP@25HZDHO{LzJ8(R=DT8=+uB6L>9Hlo>bn`g9iA3gRZZ$OUbm9z
z%djS)gX7JOppe%n+$#p&Y!t6ttHMw%7X|aQ-e$68?~L1Ul<7+y60YUIfBL$>Z$w6?
z%VJ|*`41}87r~NUsp+$;gRMdOmc(}c7zu+ermoa=+<p-(<AXf`IQ`AZo>;l?EUE!(
zJcNEtU^VN60%b;Fuiy=%!jUNoLzkZ)uby>Xc;2aF6T6RBKX@<)w#lb1Oz;4I<5N8a
zx?g9G!OmF6d%BJ;H{VJR3&ej~uMh(O-IS%>?X0ZaWQ-rlxMOo-f>xPxEr}t)yAasD
zLOzOq0dB&amZu}icY#@xWcR@6__DQZ${_zEnQw$od6KErB)7T5Q>`V3o#LhWRWxC8
zB1FLO$HU_N>h>7ob=|tdGxc$4-NjIEZB(KZ0#S7dZp}x0)Tpy!&o`H=(n=X`&!I|J
zhh4-LyTbb&U+iO)f-`s|V5O;$o?f7vf%F{^+sMTDS2~!DEsDcW4zF&@vTT)7g^t*}
zCQ97z;Hb*-69b{P>y4#!lch357KOS(F_Vq3G@yunpmOzw0^R!0rLF*kY_mL_=<9oU
z6F_=F1ypmIZ*OTSg@=;?H#GkO&6z!#D}n!S0%aOVR9JB*LUl*&;LCuArzta@NO=s1
zdl-(|t6m;?E=wRvcWAR?im7)nG?QN*XjuguzSX=umL#T3OcI=Obo<@rc*YAqCY!!@
zfT-o#r61;TCunF`x$a56{UdhDKHGG+R7&hJ<BmYqKpS+mh(1>xCjIVxyV^d3LVhE$
zZEWe!x0uS2Z4p!3;7g&{>~V1%uMHVqAqn6-fw$yVYg8OdjN{C%O!b#cCubCs7?3ZO
z%ropoDalA0Vph4lk<A<2j2G2+9JoIQDk^wQ3jDe{F>Wsbf5<~j-stko``c^1TSnI_
zKT+DZV7I@$9Y<p&w-q_}>jfXAGR$jyR8<SB9Z@k7+<ql9DK9RTagaGAQNPvs7J-+Y
z6Adl~J#Xs;4Q7j{k53sn{2bmi7Icby)B)QcrJiF@#!?Gm{94%Sp(VFrxZ_&lVnys=
zHkw2iVZfGJoHkpY*gOAjirIK5lXBdv3K1gZy#7*czlq*}9-Db_Tiv27S5SGtron<;
zr&YJPDcV}L#)$HiIv@VKlLjH&SS=6}%s2PAo#_|D&V+@U*t<FN?BiDB6v{REF7{e}
z_aYmKTkunxMX~fjgzNokP0T$wfzzinB~pdr^~Z868jCqw`^CKOrYvtg+b=V55S(h=
zlqoZYD-}n7OlonYs-F;wQNUjl_>t_uiuMqT-b7EQVE&G(9L#_bWuA(dag)zJGveXM
z*R$@h43AEO<GjUh_NA`Lba}OTKGX9d1SPvnF?xYYO=4iR^m38LRuL|ckigzXAKmBu
zwdIBuU6WUX8&$MIE(7ga4|Z|v2@2pIDo5i@?XW=b$ou!nORrFC--fn8f-!;R(%f|5
zBi1T54La0v_e$NUR^~ZXopIm`hm^;EL^FpB&p_m82u__L1=(Dk;5f7Kl0c5p3wrtH
z^aA#MUb+$d&SvGcC2wC6H#J~~ml{5Zx=erVh1VBQFsPo!-<WAIM9f-QVs*Qj&ESoP
z0dO@RNo{K^9_$zMImkp8W1+70-59?kx7c+szF(W7l{tpLk@%lIju(ND9!gp%r+JCA
z_sR+@Er^d&EHJ7wSBcE|Af%JJfjhqRX8Y*<EMZlLG#qON<Kth=;lm&H9aOED)BF;5
zP_IDB32B;;o#M!#!^D#20S-R#?JHpi8`-*yuM8(NN!%Oz;x_bU`m!E>lOcqv!#u&c
z3OW7KVy!3q2IWkx-F|M;E^yr2&)6vlcUy87;;gh`fJLT=H;DZM=c<><o&1vLHu|5`
z9{0s73PT`fC8JZ!Q8dSXqZ)Vl#T>%|GXHy&nELj2pQbhJnx%?XJU38))reBzODCt>
zu6(%loFIxL@=1~U@fSW>q<&fBOF4IRpUu7r#8j1MSANM56djl7jDB<KB~Zs1W}uEQ
zInsG4PAD@GE=><YvlXN^VPyL$zh{!>+e1M;g$*`hdrcCDV6z6^^cRbdXqz7gmtSd?
z5WlgZu<(J1sR&F>&GX5MVr(xs<u?16$Ug1yBR9MTG6v6ah*Tynr#xfr(RDJ!&5ekR
zdyvVHjKVFBtM}7G=7Sec5~nZ7Lg;Shz?(}s`W!m9PvI!IZDr+tO~&&okua3m#>`V@
z-3BgXJ8rHY<4z>LWHN0<wz2J=mDu#=>^2ph=-Mlprh?J2?SerHdE9z@HU!JONuIJE
z@V;`X{*z^Tkat+ie#<@+JUp|R)<Y5=lL7yP6DGTzT^b#%faJxQV+6(JL(vvkfIYpN
zh3)lM8qN1<tj_TUu{6Yk+{$Dn5gjv->_4fzW7WKUS*9Y%^lc5n+I-FiH>>9K=P~F@
z1Z@9@2?e&)fluAm;&Zp@PGy`+#?ipNH1l;%L0BtaZ<ugrpN=;y_yETciX*~FVdKcY
zyHtW~Y0gJ!A(+?3c$&1sZ{Rd5)WvbLGNkP#<v*oxhFB}-G~4-ftEF4Hv1<`u2AEhT
z>tMX#7W70a)tCGGCy>mi!WMoJ=|T*0GpnO#=l!d~_J*LDrTtgw0L{%rSWah&s1=?C
ziv-`@fYnyZ)L{AkUBgdu!x;Ve`eS+VYE@VNM~cAi#Re*G+RaXOMy0uG;Y6ScM4%Hg
z>_y8LAxd!8zoENl3dN}yjV0pCX1#Y{Z+)Qq)FI5+)V%UqujVE<k2vTCe_*+YFqdM5
zQ4tYwlYDf&`+c%q8_p_7z3vP;!G{%2?Dwu#D5iziM+16T!`%F)(ux5$z2oD{4zS7Z
zMAAsW&}(KbC4ah~u2o$5I{;lmO1{rfxr#9|f{Bi85gAnBM8|y9g%4K|sg@*aS}SYh
zg_h5R$8U+0?X_;|%0l8q8iRozTV~Oa-YU6&_`1nY9dbQ|9KSqlq@7)z?yuEE9ofAv
z)qtaxb2@rF){c`1cyxeLvJ9OM<=as{MWm`IVR}&w>yEC#(#x;vaZD|Zj;jdho&IAb
zKPmd+^y8C#dTlGp9eOEE&}dTs;wO{jM0%#-ifZX#EKKCup`~L1O*qq_`|LXTq%gb-
zJYv9BUj;;4liwR?S8e6CAOhR?F!I3EkytKaCTu&hmBKEr8xwElpujD=8Q|g$8h^#$
zSNJMBrm|0Z_AbnRe<*0racgItA}Ws9VCkq8Tns$gJLa`bLmLAD`ktR&2$*?_#(PgW
zwqSi`Y`{gCH6VqTFq<iA14}|iNnm8>gu=%1W?Dw|FQ*$To18<e5}W2fB9%^Pdfe20
zw)ah8`rw`Pl9uA)%5`<_#ZSJyItXGGsDGtnWTtj%6k`LoRHW}`cA6Zd%-D6!EsOE)
z$Rwx38&xVjc#n6E#<6!PQZ7ZFsyNGPpN_T4qXsq^IlEIIW^~ubD6QN8(rih};x}h6
z)W5v8Cv)Tm315oOmQQ~?<$Mrr5aa*C6-hx5-A={})r=)UNl&@V*e{TE>U6vh)+b5G
z+WV>`I7oaxc{BhyT!Jlx{dDXMAeJL^lPf=c;7!x=+XT%+)PUFnQ<h5l3ORBuMtGGv
z3b^`Nb-TFISy5&;_5b<v-+rLFwql*;zHk71^k7nMT8ejv<#OMdy%^DnJ9AFL(Mez?
zkhUt+s$`~&IMq(FhH$5I<_t=1R8`Sw-(v%`9=d7da*Z-0yth<wPi-^dU@*TgGOoAu
zH>X%_-qHWaP!LtPMYY}Vyr3Q{Up<u}!T`wlP?B4c>D}QsNvO}lVK_3h9D6D0WopL%
z%LPYpwWhVo&Lc0&jB^U$?=Mf~CKY*iP_!E2T{j&K?(d&EIyEQKaBtig`DE6;PmDXq
z*Ly)iSK~>iW&Q{-@wC9`ni*vV-CIKG#B&D6#6(>01?+ZQd#ulYbm=9g6o4Ic9+|0M
z4}c3lj4lst11QfmJG@b8E8}!={zrf5`iqc7lOz8T0zE3uGo#97QE*FBIRr(ar*JgP
zX`_}uCBk|fpAfx?0WH=?Cd-XRFM?tkHs1YL2JFF_o<p0eS|&5{{1*=M0)q7h&2)={
z^)E(K;#J9ZPnFt0k*O%Q*!QS>icmQcOv;101R!V3oxb8qM=k?;$r`>`++;6g;y1@@
z@`XK3&eeT?BJf6mK-Kxa+<m(67&ArEHrUPhQo49koae|@Z~{Z)mtWw7BYg7s6J+;J
z$nSnoKHiKN@MvoKG@Hq|&-Z8Rv(?|9ioIW5I3-STL>8RE^;OyW7JH#RFDL#C9o!k%
zcevCHKAih$KAU6z0fubYE0w}41L1>Cc2PKnJB+UgQf8cC6T7mfk30fEGTWqcMH6`e
z3R?gDdu>w_>sb%=e%r5D{Yj5CADkc!=RY|X1dH>99Ks875qGZFFtAB2m`&qviD?BQ
z4}B2AOIwl#5+YM?CnH7_xHhuZ`8TP!KqiAv>x3XUqfDOfFZ}%>Y-GZB@P_<8&!iWE
z3{>;5dpP1ZS9bn5_-si2pr;nGgN5P<ke@H27l#R7+HMvnK&^P^glU8>-T)xWeBMm1
zK{C#Oy*HVc-|us;fGPLoN(p-7NX?2n>+ufzVGufUl@#C3mYD5Gx_$?X)8NZEZdUK4
z;ru)`KG0yupx_#U2Kp~xGaD>=|0@F68-(R5BQJ{aCo|Q>E8ABK0d_>>-uJ#A1Q!?B
zM$muxM`5Hv6(XNKl)%dY(6v+SRX}>bYw@;Bg7XEtSM8mer7vo3mj86>3fbooBlv?y
zE&!}#P=ECjy&MntCe7X6i)#SO`5)pvI|IE*SG(%|p9{!ECZOxBMYC2>V*_@c?|}f4
zSE^N={(A-04|>h+d;OZDs<;5(UR(IRu}_i!iFLg{Z$IG{w;q~YT^Xm`9B7O&DUrt>
z9*o2VUMe^~H)X(U3CHVRV?8wb^EdnM@ck@tJ#wkM=GJh$Mf^G&&<Dt+x({6cdmD0(
zcLymmf@Er10@KQK@v1;LC7%|>^!EPM<SxC&8|Hq%m+54$fuiiY+n;dD4x~h}@{}^8
zRywGm<)o^U(VnsWA;Od!!d=_mZJk-^#%&(Y9Raekw*#L51M}<z2E2iqb9HcxI|@{Y
zr#p|`E1Oo@XA;UVJ;uoa&WmTOi^2+;X@NV;)SCxe9QkY1cn|4+1=u?66wGec+oTC*
z&GH`-$RTzsayeDc>;h>u_L7!iCy-uVCgkQEI5j`$Dl^mX(x0p~Wh5RmSGM0MZj#s<
zY781mY~@cdai!q$$^=}Y@t-Sj+byiGk9jWmA0a*xMLjo`noZ%?VioTb!@3CrM?|qI
z@ZW1-fki$ne=tkczk@24dxn)`X`=Z>EDe;z@shDOg~fd%&qUNqNVLRAw#~{-=kF3F
zVg5)U4d^dyx>@lhuZJ=*r;H%+`)*`N1HwK0begfS{u_55j+uex|1yKcdUt%3-k!IP
za2@@1{FlnU8}X%8uyORznuUmu;SlHVy#RLmA8~GCElQobKN@HTnw<Q92%w?K@NB#B
z($2mS_(+}n;)4IRaL6*Y{avrIqy&7P;Xi**4MBh;H2E`_E5Rv}IT80qOon1JI!h;h
z_f9!$n9h4s1}RrMOYg6sxxz4sAa9axdaa@UZd>St<`)FxucMS1s6`&NJw6-R9$mF`
zaPRC!8&COleH6amRU0YSVQjF^(~5>68`HZ1hDj1<ONlF2c0h-R5oQxaR3gM6cQq|%
zOR=pz7WRy|22U1j;p+dO3)B!&D?j^UU&L8!0hv?m*ZEA@3vZQuFa+pHo=lAgsg{ur
z{(F<45EF%*=AZ~Rb2E<SLVAHOu%OJD$A<|x`d+Dv%z)zKsBUJ%M;1+-cv$`W;%@kp
zna~;KJN?PiRoUvR)0pF5`gEx$_mX{`jjQHLU)DUo&+s)4pyCjn2z4X9$EV@h;3&)R
z$2(OCvF*C@r%mYkt@T#$-{=CB3ek`Ejo`Fe33YVh-*p|`mhX)HDnoo2S0Yl7S#)<)
zwcf;_YV<fZ38!{aG^Ju9^sp+PGSm4#&g?bD4D<rNQg2zT9D?*y(zBkb+_n)SK8j7q
z6183GdVA_avf3*zFAz#LN4^*Hl#1E?W|el=u^*Hnv?0#WG;HLx2VPR=zA>d#JDY%I
zb$WiG*t1;o>dAe`-Wcuo8`d2~VIpN6(p)%YY>@%gyNAqr=C3D}F9=o1Jwm6hb<@IL
z0aZEcou&1RHF;)<xZi|=QKT44+U`X3$jdQYo|Xcc*b1zGHPzN`q?}_>YwMaaK@F3_
zAl3i|qg0NrxbkYEhjRWaOg*nvL5>)hLzY5q-92{3Kf9$Xlp20Wt-J($_Y;qcK*w@m
z3o}GImVgQDm^$wZ3i_+$Ah21#2s4%?AX0?tUhXuZ0}%A4yy$PHKb|9!7sRxyUUW*v
z37;NikSy|Q-ft;QWqAepSuR2EC9Jvog2Mk6lvwY>G}$Je*FX5~j{Tx^()ia^z$M+x
z(~#okI7YE%BFN<xc?x}c*Bs(lPQ08iZOP&qy?T^rFqWBWU&;lXRnRo!!ObB9eTAa~
z&NrBC?e1QL>whU56~cc`gBf*RIKHTQhq?3Qn~Im9p)eSTE)R2pd?xm38kz^CbMsQD
zoa*0#uGMOGsn`x9Y?6uy_t1lgsVPf{v!1#x;bCQx)SH0aLN7DO?bob3s1COW9_==f
z$4avt4}libDWTuCUnX)ysu{$#tDcN~6jR^x51H`bMQ)3&1m3K3YS9x>pgft18xA*1
zxV*p6RRg^7_P^x6kni20nd)ON0J=-sqoeeNM?V`d?#Wa`6y;GG(E~cfk@t@rnyq~F
z(<Xy{p7qnW0Ah!W`bNHP<B?pz`p6tqS%(CR-L0DA7+~14KDTB<dzTKj-i(-wGu|pL
z*1i*?8MLtQBDZQQWNUtT1s%n;!uFB44GbO&2?8k=Rxh`qO#qiCb1Pkyd|50x@k_4&
zEA9JTlb7IdL?I^XIn7jboPno7&TJ?XC-dY_Sl^R_5B!l)<+WHQB~DczTxx8;_z3~r
zma&X;>?bWTylrHj{r#h3YTLQfhi6F=oWjU4?EN1}$+F4zR8<MPsNGm>ER{T{95m)~
zbj;qW*<B1ihX}hHL#Q%jkC*TjiU)CPCfT&SbDG#L&{DEyb|PIOX3mrWLh8CvqgAA`
z(<|gUDeR8Rmt$FrZYRW-p3{6FI&+zn018@?r#QRy47fNJ`tYosw<{lK4Yu37Qkou|
zK+I$+byOC@az*nrovkd$<#=x~&ZJg2j9bI&X-Ioz9SD+o1QBaimk=-T{&Ha)1&}F5
z_^{ONW>a@!%Q>RWmfB%hZDCuFjOCKj#O3CTtYt0$a!poPm?i?Z6;N>|5sovZ_dq|w
zxC&W(2}=)~q>uD!-yg>GULykwX$m)^*Q%|aPcZYo0#5ZHD?jKZ^Jw;5Cu)(<Dr}qt
z+Uj)lop;X{`l$PY4;o>!V}tfjiTQ*^oDCMtLZof0jiufUUkw9iB?`1yxf`c!*YBr|
zH5Ab1j;C$yW38{8pqKjwqJKY>m`Xlsl7UFuePCf8&y$6t%tXNY(#ZP;(3T_an5fLm
zJ&j!cGYcRs9`E$|!=Ub{ItnV_)jl@e)o;Iy;23Uz)E@UHrd9`5GKourws2`sD$k+l
zkydayB-?jJf0$$SQVu2Z|ND{;<n9uM;!cc)F5$Wyt^K9He=01XJ3Agl5+4Y);tjT~
zaM>B}S^BPR0^;Ul>$u{b#`uwO11kaIvkb%a<5!;Y<}F&{S~JD6N_qOKNAV8y&v`>d
zkkUvm$4?O%LGLoE0PmLvMG@vzw6J>G^9g!*mxCN6X){UfnoD1)%zokl`|Xyp`SA@I
zv2XN5vbA(GYSrI7NriL$RdygM-Im!_A?nK+vCS*xTnea))X5iwh}p*67xcUYpZga1
z#E%HFLq5h_M(Kil$?lrJ?;bH?U_O8ko?x)AU$_topk*J=NS!<{(7U+82~XS7y|!<r
z#lH|a5b0W<P&Qd27*zO?;V784X{~QAU^F?&U0s&6P(wg1H~CwqFA2nB1F2Tavt4&+
ze$B;sS*6(je&MrC9N`Y2Iw>ovL9Kgp#Y~&)1&%Tkp2SNcG&FC^i4O~2TmZ^@AJ-Y(
zB{#JB7CB#j7h9FLocCIZ*n2aWpqM64GM;)lf64^45Zebjnbrx#R@<K&HWr`M(GYs#
zztWLPgNnUQgQ<yUI#2ln+wwA#=@E*W$|TPCO&>saZ|?-WtWSHGfw)M+cphZyo9qn*
z)sKo5oJ!vkQOEnp%+f(^(ps_eFuPE<v815<fH=Cxa^GQwU_@j7;k^!HF@dF+rD|O2
zzpdENG`WG+=Ka^ruOdt72W^rY9^9!7NRn{@J!hE&G!@nR=flKYz9~W)mt!%)oXsp)
z)=rjo`@NK92$!-{NW)ol|II;oz#8&YaxPVOM{4>V(H<Ipu@u8vZnY2#`zg<NOk3ZE
z$TMZ~1`q05t6a(`TZ*1p1sN5^_K$`Fk#czL4U#~vQ+TCpr9y=zh7`>&yfEClDfa9X
zD7Ku&(dksL!uMh4ZxWD!>0OBK0j<gpD)yp4b3c&>{Pt=wF!#J<yZw!f<DM$MlsIxV
zDUv0udC+4;NPqIJE2`9Jsij4%(qbiCo7j_JzLRg)1DrHSwO7-6<}4zAkX>iA1)1Q{
z^$ETrb$!(T9RX9!I#I$6tlY?fj>P9|KENy=___DDp0eCGzu-qUN?TodQ)||V_gtWG
z1ox0cN(A}EuOl0^>4O*MidGFi&1BHy`4zf@Vqygu0~M>S?(B8$)`x=I1~S|0JYP!i
zH3A`s*c>mp&>uQqyc?I17u4Q*cgG9ohp^{C^JD@(e3>MOn=t&7Z;L{tUZ3#ht)3g_
zg09ZRS*M2GmH%d-PIM7uVb|E?E6lDYuiiKNDqx3CZbkEP$*qk{oK*f{3PPkyrfb-c
zwWv>!W_U=}AIbY@%$t%b`d2Mygj<UX@p=9^n3@>U?iTUYA%y9=NE(hdul2q@@B<aw
zF;D7ggYYg(F9RdvsV8pTQk}hA=9UbmQs~DyUSBMOiW5gx@Px?5JCJw_n<?^6m;yC9
zJknE~xiJ%<?*Dun;lcZcnvk;bp*YqCxgVB~cHQz0y~_>pd@!gP7DsIRz$0x_0+`9|
zhJP_I{11cDi5{Cfw{FcwQnF}C;yEytg-rT^ZqrNNvMKPy^|(M3_OoP&0ILi2LA>k<
z-}V4?AM|Efqu^t=>cgyXVc_YBxvM*UaiwSBfzwyH&k%0v7WwuyONRuLY=t&Hc}&o(
zL107yk!krSBLkP)js|@04j7rayXy%w_qaP8q}!qB_0;1sbOfMCF9PZ5X*s)*Y-VIr
zq5-j+tzEJm&GDu<a|BNgzB}s2@w-s;kxY)~>W&qU+G#~O90ts@VYb}V6#WBzkAK83
znGgm}LR}|E?RV!Y&Com}uSm+ub1$7)tVv(r@4z3D*rw<{yr)tXa}@OyaCl~&Y4k@x
zP+*uK|7C3$J`+ZzQucxFngst&$xB(f-?z8?HqY4-0q#P#P}Xa0sLwAEpgRJaXsyB_
z?j$Z~3BQoyOJi?JHgc;{-$Jd=!1ox@^T7-Pb)AT`8Tj^uV*RZLi3h`5BDphUAe&w4
zeVVkILVQ{)G{8~bu7SN&FJna#{n1*=$8}Zkw>b0&40~9CVUJ=er>|IJcT}ZB-X*cr
z=e&N2+E<8G^@!?Fmq%qG2rpw%*bDCiPth5JDEwP}dGMf}-S}&?vxVZYAHFR>#&kVt
zt)|1ZevoA~>QE{*;8kfs<qDD06*`TI8(Cqz*zIYjF#DeSInXgh79IpbO@363_!v{{
zC9UaDYy2_4=JZ7owsBg1`NT&dP4gjJYpKSw9jf&nF8L3&1!Hli8S^33^JKH*1r-g2
z3G$aXQ6dR?lPGOw6t9y}@lE(zZg#WG*?oH72x5W3#q*ZcMelm8bZ|W%nY`Fo5iARg
zGmPuq8@y6r*h$)QJ9_#bYw-ijV6TbR8PmA0I@e41TTqvSA<v+vfTrihxWIlI{vbkm
z`R?mIX`rsoCohV$tzMdc4bB2(YZLeI6H8UIQ^b<zGh$0$Z$BDqSzWfws22Eg?<k3u
z=55=Q_4ebkCkS|=Xc(r(Y5rxiTet2{ZRH`q7gg?q7q-?sxZzzo_s_KxE3By^g!JrU
z%!qoA`^<2tn#6fWoX)x~Ryb^yG)%_)JF_GJxqAm#Q=(}m@USgdS((jROEq&IV)FW`
z9uG2)wcFrh3EhlWOH*`<%(FpNjlXyX-wn3<4KPUS_<4evFC5dCTgy2bH>jK0`jJQ{
zkftk;DhbOKpYS|T<-g->Zkj~Mq|k@+6_pUgj-^vrcm}2fJlQ&F8$#Oqnro-wwZffo
zt=dF>g1RIJ*|;#i5_`Yp_SK<4;0BdeHyR4st7*B7G~az6CVV3)i7gX4d6hN1$jFvA
zc#7idtf8}m+Ct-$+P!nc+_@({o)26^>4gFiG&IW3j?22=Ij>>key!qZa=b^$KZ2r4
zxjwP*MRfiiIZ1BMM(Sr{@qCSszu?5q+4KiQtc}$QxCql9^hOL4IQJ#xLP#0OlV9NL
z$vQ)}#CUx-0S3EgZyBn=;HHMmcAwsn`wB!P)!i^1tf|@k{Ib~<;}t0K<VWOLfJmKB
zzY^cr6b*{+L2hlXn*^WwiINHJT9zc@wEk2r%m~crMFpbOeoO=>{z-*#vEJ!)u;zOJ
z06$EV0buY6P33Dm-=ihn>OR3$-LwBqNX+RCy#Uw`Tgn1;$+l`_o`F;QH<RaRx4`<E
zG@=%R=@AkueZ~yOTesWtqE3FQH~HLvYnJh1;=yo_rVNwTV=g#B7nNZOU|^zhDKwSI
zA?*~%YPaX*Q$(_a!K2MY+WPwtL=3CuvD9%<02%=x|3dAIC-GmJqNSgsO?dy!F7(o`
zdv{D#(6_FWeKmU2^mM5vrp3Rd<}AL~xu+ne{?Muuq=9cbTdD?CZ=Wx?rppbEQ^H&`
zYQgKrxy;>7f_{8CzHa4igEeYD6Un64#qX;uC{%3slSZues**H%Ux52v3E^(ewazp;
z;0<~Ld?^vhI!OgCy;^~O^L;D5As-n}{4$ckI~m^-4fD91Pr-Z}WEe4CYK@aE_13mD
zM2e!J14bLh-1?k{4%S;s?3U}qe=~dv%m2#YKdq5D{$p}YPD!0bNko-!!Qqsn>xG~&
zhzO^DRFDn%3r=^-)=)({94;jWVF00GwM!QtWL8E~pvC|$#IKu;oB+<71hzm*;&++?
zug{OTP|tVd)0YHF4>mhffZ@ME74C<uEt9H*#1MBi$Z*_P^K`&x(o}^~CA(-dQJ^=U
zY*-W`w-?Q^3QVGpc94F&1YmqwdFU#byUH=C?ch$tm>VGaW+NOBSa}&HovP*IM7$%1
zu;eLm3GY3ZMz^A<^*~hUQ0kzt=R}wb1fb>Vxk9-Jj$n;qbh!<TV@@ae7b8I`c~i#o
zhcj;!5iWoVrNj+4?`VHEr}laVh~uc4m!AfS)jPENO=-&v0xoz2$)>K7#%wLUF7a&p
z<ZaZPY<Iu(>e@KRty}&l{fQDu<Z}|+x%hWCmuz$+oT1<30SJ6ym{jlKR)x!mc6L*D
zZc*SCfjA^o08H@QKCF4;TN?oU_-CFonTH9KV2FJb`n!>JM*_(cj3rQ*>&d#o<sgN;
zAtRZ|W>Z7Ru6OeuvvyAu{JsRyoh!5YBr~^Rc_op8$PYB0&7a3qiOz-Ho!xv#5?QPV
zxU6*zqD%eU6dfs7Bg<Ph^JlD26crbwDzUmwwfccPGS?PN!%)E%8rn$cqyG5;)LS}*
zkfJ<ml?0KZ1Bd5vL3EHyUYDyyjg^Mm!@{22{wzLsvf>CH*|CtHAiqQAo`(ZJ!qCes
zJjC2_J;yj8WL;Z7F}_4Zp3aLR*&7BQ;p5Z?!(%2gt3k$JF55-yef1Gdwy}v+FgzyN
zaKbWDolRa1WNa<+HHR7_jc*Y#Tc^q<EBVkPc~y7hVzCP;2M|9)_bc6|0+H%d+}p|B
ztz#GPKFF8$m#oxcf<=97b*(Rk3i$OHFVE&l)jPy4MdKc*9o|^yUaD$6>CGDsSj|Y7
zUi~!A-%_fa{R6v#*3%6M@yF{(BVRsep}2?{3la#=(BoUD(`NIo1oL+PE$Q<m0zAa(
zYpMe&uIYKsnv>{)oxZM9YX%tvgzCHsx(cel3-AP^S0VoRMzs=8a6*KB`HsTV(5eBU
zTTZ={g4jV5h2@x=%WeBmpD!mGHpW((v?lns5-6c_pmmgag*B)z&QnunbDARH4(*<d
zG~`f%>DMQs&K-_)afT#%?f{j+M$8My)m-xJ*5M?QBN?A8o7dcZgy({nq-$^Tn&8-b
zkw0EFH13Omd2Rrc)|uzLXZL&ou2_r*aa(!iu*kJ)5?oP~7FM)Yp6g0{-Fg8?zvVY8
z0iUVz(tOIH$8bH!w=J>>@nQ(gIIXvD{qb`@jU27hJ9osf%(`Km`6<cA*C$qoL!6#h
zP{Y<v>Kyj?A5X=-*xQ9jOStNdVZ&82uf#vbY{Z%G$Sk`ivfPB`?0p)uPBFTz4$7>=
zlO_jV3hE}6-@mCmD(Y3b7RO061&KKTt+J_s(MWHORJ&mFjwKa^LO+7{30R->#qE}(
zm%md4-o+2nYS^Rb!RnBzEETtj4b(kS+h*X(@qg00Ef_pf3%bvUz{*39QDDyNsK2AV
zx1HUH9mir~fB>p_Pb#K-Bvp*^1G57by{w+*eCdc~KN5gMwfKcO=9D~su}^Y2NI&(Q
z!SfrzwA^_`VOvlub#TG269f9<#MU2%OKX2y@W+ov`EmVD1JNxX{oxKz5hlIZV-xM+
zRNY(1JuL4RuPy$MMko(Y@CY(HFVSeL+ppO?`w1Dq|3}$3nK=;Vj&d@|gASd9ew3BG
zDV!ytdD`8d3KdQ^(_gsW7PchITRKgcsl-e0I%lw8CL+zasO*##jc5gM$v*>>T;PaA
z2Td94+_0bCc&{)<vmV(oC7JF1_KZL%aLJF1WS9(6RNfKAaSa!gU`}OwJgfV#a^SAx
z80Uqd27tB5?LwbY2FHrb2chea=00u2%Ln5{KDs@PtDs+9HoX7=6IOzLwAOn6gXHD&
z{b>jYOb32~UI*==!Yd``48Is45?I7c41KcC;nVEnl<pVw<G0ets@Dr^g7lvKBK3;w
z*exHNokG8%V-Gq)bSc_~Y_*Bqh)&PaVXaenNZ?dST>UT00H-P_Q*-F>H3L<=u!pLv
zBr?MtXMQw3K}tix)(KWlZIbF=6A6!^^i?5u8=LN1U;25|JmC??8OW#OplcDajNZhp
zVtKJiSj)^762`IUf30|)2e8~zlLNz4s4CoY2CnW{PK|;S(MK)%4B0rTIE_%RqB@4U
zi5n-<a)VYbinUZaEwA=d!v<WTyZSh`k+me6_HC=pb$;p=73`F_G5*o8HaFkC>k>rz
zQYXtMpI%iuKL2iHaeySIJ@hp)NcasG=+O9}mtSA*)WyTF{y9iNmhk3zcjpHah;<Bf
zyoZvt0hWF=8}*ThIR7k6cbA;_&zic??m9Y5UovhsD}JX{d}93M3i^5F)%$N?d!B`X
z6nJ1Y3Gtv)zSo*W-BG2Kxh{_q^p>Lck9?qbC{t+FO{z0i=XhG~6%FH>S!)kBWB3r}
z19UL;>HF7e_xw6WlyZ5|UfioiqAnUU^OP>CSIq00+GOg}+1KLutY%M}N5IJWz`Ix1
zwhC>cyW?sBaXbUDM^0ap+VXF|b!+(0A?{o*JbBV6FA&6^k06Sk`*I684Qy8~!X_z0
zhK)bXWDx+Uy%WP}OU@o+veDd`+3MH+Jb6Ti0wngDl3HF@z;vy&TusYzdJH}`n);L2
zR-B&wRdLgw#9!((DE4XxRYFW8QXG4TjYJh?++3-Jcpx>`Xc=gp`3{(G_*m*@un4|5
zW>w<Gho!n3d&@Tx%?IwXChRQ%QdL>m2*_{O{4Sd!!qjk?8XlS`Qxb_zjRX2T=3Y#f
z4z@l@(V`ZAx1>hER>K&G>J>KTLfGHndS3*64Z+@YW_BQ&bxpbg!M@wA`|O+kFhloM
zw7({<NfIj6r*wM)EK<p&NE?EwD*S~+QT8#AcCow^aL!6By)@4H>}Js5L%|A8c*41v
zs!bx;L&Fjk3u;kTX^W+|FH#e(tTzIdb;h|Pr4&GaNea$V0*YQjQjR?nyf~jW>*g7I
z7<H2p7uu1{la!z4vRH{Od5o@sTiWSUT1IgM=t5?J;z*~E-sq!bsn}8*y5t_YJTsx<
z-%Z5u3%{NFZa<LdMMr_UhBG+a2;#NnF2dd+E)_%Bl8-ic6}%E&8MRG~D4GHS)baY+
z8!d=HG|P|nL<L$u*t4em<EUAX<(i+N?`H>CY)g?<NiYAjmDIYC=Gv_Mr3*y)$D1@{
z;JU3Espl|gxO*kX(8OeBgJaD+N_EDY3q;Ii{fx5sM(~QuK#%gUF`eCMC$ZjXEIIO^
z`4$BrvmN%0`>uEf(V1M2k!qFB9h>}xL9}cy$e>jql34mnQ@c3cW!F>*6b3y?oSI?U
zpte@2%x*;jv-w)k!@K}CtwO-JzA-*EyB_~H8ARS7aRftJ{ANc5oeF`E&fKYT5;aPQ
zmlDz3+z~!2geOh!39{<U*<&hcpj5a??#s+B^azzDusuw*=6YY673Re3EOL~X!XRXz
zm(f;^_0H(?UbTus$yaBXIf4GzxXxbv8WSr*>P7Y8P+i=I^*w(wbPU=y5Be{<aJU9A
z7KsUyV9m{^aI;XNHWMm@qAzBtMv?L2Aa+nb=87G}X<3lZzRrLRVQx$f+ZDcDA$>Rv
zmkOXjb$FLa?3!}#fG5Xkh%D%N7_L7`ahCaJ`<ZCH1@oK4wpp+*-(^Y_H#`UsyY*uk
zsRlZ$cR+KJM{_%<@z^~f{B7StEd%h=W6m$HT*34)E=M0Pd0sIN7PD#&^l}lGN(&^8
z-He6ibkb?hMsQ8oxx~ll1_Oi?udOE|A)g0pL{+pVebV6qrK1PvA+o@K6^h2%S<zic
zG$it_M+}EDle-2MgK3&2e@393+8$$g!0lWo*9)Hpn*_-rZR)yaEOvYD1f@essFAK4
z$sdFqqAgh;t{AV+EuVV{+Za7qbS>uZZ6840qHwpHt?60eOmO&mMR-I}c4!Nn>ki*l
zW<#VW(uVAwBE=gRHH>@r!o&HvnkUGUU`202T_vOp_BQQ^iy09DkOX>NsqQ^sT-B>P
z($bHb;R(nZsDr<>!4c_T4=zDJqQVv*mQmy>-=m70#1CcV38ADYfnw)9M@O!eX4?{R
zO{>i&@7n28H=4D#qJ~?^UHq_)C0XneVL9$HOr!9};q2`$1eT%}l(~Rfy9>ZswD*HB
zAYIt50sy(<m$G3Y6a<$U4Z-|q*+CcecFOk0zshw)$5ho-Mcir4?CNaBNt6^Az$sfz
zw@p}|kP6cz>ae4RDGWW9dJNP&zDINR>Gh#QVWRIJdxxplt&xL0w%hPpdXBdER{VVk
z(>fafcw5BBj<FPuOK2wK{rKrl)-w>X>Ryd0zJ04HW2GSQ(t0~=n_WX37)aZtA47-0
zki&#bpdIO;BI%w2Y6zi>T{XnL);dX)&m2RYp>%icO%M9+wN~!LvG_F7{qyC$kL?91
zXd_#&mXf-DwZ)?K7wYSB5MPjlES8F<Ac2QH3*Wx#)3SHJ{vpWCN`UA$3ZMG^hK{ae
z>SOInrSYz|Om@yMbltx=Amj)J)WX4l+8|^3hogqvfJUWD=2#z)Bbm$9NNoP;OKO;+
zn3im&c1hFBjct=K+_D_`cW|a0%bhx_Ts6c-<g~r%{%n_cALU{|ox!0KFiWPOu^-XZ
zezes>fd#R9%k6b4Z1s65%<<BU1sMSbH^E}3LvOdod09_q@A~tEP8iPKrvNU}9jm#S
zNM)@uV7PNrRnGxXnOXgtXZ%Z($P|pmP50VI{M?qtVLX0kM#jT|8Y45b`2gl(RI+!9
z(V~`OYjH&QQIZg35PhgpYs#F}N%iJ&@++ovz-?z?7=5$AN?vKmNSsYnr$6pKh0-8m
zm(I_A-+3+^Kk%K79yDwc=|iw3!?#ZnS~|G=jE-bAwZm<669CuQexIT(B!SV@E%irA
zIt&y&gjX$<dIix^87`h7-E-Z@8&KOB6$x>J#^btb5MCH?Va{GD&V5RAWyCJ*9#k{M
zJjYsPZg0%;h#Uw5Lb>6Yz`?foqn)%Z!7sJ!g|!CCh4?KguFBQIVho#x8H!h6bJ3%F
zrgyDW!6uKV#p2>W{po}$RGR=`KqtP{!1n8(SpXN6D?_+usr&JW(;$B*rPm~-*V@G&
zZb?Sim(3FP=O*iWi!pTjnISgZA4ePyDlop?gQ6Qz@A+3K619<#FSsWi&VDapX#w7m
z&30EshDkX0@Hno(01u##I;C}i=1`l=U3M_737~RBI@;tC7}k^^QX>wUlFjkL;<(Lx
zFm_;bhy46mAtbo>gTvQz><7dvxos_V9||ntXas#rCbaS90%+}I)mzyf3M#~;n_$Ds
z=c>ob69dsbdRH0#!AVGA5&#^!rgybS&BffRd+7(p;ep?^7QIdfUTAoSUViiTR8Kzc
z;x&X(?v7xhFzMFSlfIZxTpq<a)&N>A68w`-t`eF0k*h;pOq1q_^Jk@n{qMR`GXNB#
z<ISjY528ga@f0`nw)_$NI~vFhsHVH<!};}(ev6fHH6FBJKE&#4$j<J2sTiJrdow^4
zD;pM;H^+P2u3!7414#u#vplsTbi5#RdFrIR&(X_B!RI~7)D4rbEb6UHY>J>f@|pa&
zi3=Zvh1Ztq7DRQV8MQ*Sv9-#s;>plP1=B*Dp3^b4Z-Iv%Zo@Wi-WIQVi&M%Zf5PIa
z+R5ZEDqwLqvU+mSVcJT3lLPj%dSWwx=xGID)T3egj2Bg{jD~IM$-;2$qh)o`n1eus
zKVg%Ja_vUC_}FTQ+Ft1BS<?DaIVK4z0Q|P~-F_X|<gfjT=}oIJ8M>MF+5E)0SPkgm
z(;&LCeYfF?Ew<`AAy#e4boavE0zsG|q0-GB)k3Q$%>(b-eTh$VJ{g3r`C2-n&^6Qw
z;g;GgwD$b@Eofl%*AAM@AGBLKL@QSh+-)joXAKDRbUlz7j{l3D<IXyk4#xd?x2CAy
z#Nd#ycWC>Yo#Wue#y_?L7V=ZpJ9;?gw%^d0*e(}?pI7Aqu^1FZY;&Sv%?0i0a!QFz
zzBiiomJCv9*A*-sSZ*|qUZG)l4XWthk+21S_M#5Izp)x4O5uKYFlWnisw0PLdtCIG
z^<8i^QirlmG&BVmz!x{N1XA!H{#A^Lv9|?5YyixR1R*<d^ExZ?Ipr4EdXC|g%%;06
ze~3H?A%T@_VtO2;JsP8>ZLeNlL|-=j1&+PSnRoY-q{PmE!sO@9xaPesi-r=4DYh*z
zkoD^I#(El8T1aBm@6Ddwf!$JcNj=x*G@%2C(~Xai7hvV_Rl1e<+!wlpZ}*_bz$~$Y
z9-K4+9$c)(N-yTY4nG@VJ8*K35Ft~PQ|k5lAXY}9cZ;1gQD7a%PAr{D?pZab_PmSg
z-;zcMf&<9ylh)c-3WCb-LT3w4Yj7R(`o`B`7bc-%)eI(*2lVH-Q})t63bH=x;yG!t
z*|Je(x~N(*sE<Hve_{R^+jX0-S0lT@PZr~akn{v%GM%GHC9z@lQdjrUMGlx=5xDnv
zpAAbmP+*#DC1j*F44-E-Ks8<O_Ec?!hOBf)AxUr#%nVR-8*8Jk`Qw8v?qjCf5K!EC
zI(pn*%)Ca@8j1V{KrFviI9<cbSVn2O5cLcUm)xEkZX-`I$|x+qR`wl2!xe=bh~ONb
z^XxZ~y;IHc-qf1oFBXI3hE_3WtTk+=jpTC4>a&v{Q3Re{p>?F=taY^Ky=+BR*7WG6
zJ;dFcdVK;+^x3nAl3fPRqG0=sB^L!kq>sbWubVP);G7?xZH5Sp(=GU=G|#Tl>LXdN
zq@aypol7SolIKP<02mczT#Y{8da?JG!x-IndeY$z`^)CJ6&wm`NzRZ+lSexqU8*zH
z=iqZD752yf6?9y27$;jnX)D}x)d_)UB$R>;ocdV)jpZ|Ns<O;xcxJP?08Z>&!N@=p
zFu`HXtpUGA2KMeVjk<T&QIZP_v^HyvN6e`IDd-OUy4BoRVzK{}bhck7SQ5!>HaKAw
zy1hM-_rk4z{WDry&`=bGlRVg&=B^faIJnd{b>z-*NG%O;ms02CcQs_Pd5y)N@7>`Z
z`~HSQT?&<)9SCNp$%dC?$RrHVcoNSz9+@u&EzSy~&(*bWo)E!F3&~C*1seB>{CCd1
zBGXne&8;jEmiKh4tvmPkcEGF@Mc;8TX|U!sHLq*X-~rJA+l85%l=P-mlI067V33fW
z&5vv^LYzU-BINvh)0V^&bK3OovTNMI4W5r-x5`tDwf@b^fX_0##;C#;o`L1HpsO<(
zbf`9v6@{-(HquD~EKj!P#+VX6KTd&Z_1RL*!TS$_-1c%m$+R_aYx1CxtjjAJ6Y`H?
z(1exm6aDrj=ti~Kl$LVUC>z)}1)gY$3L{hxdbU&mt)MDMH{K)i8Y;|_^Wk7+&g=G*
zFOWY;z{q5E0J@%D>dj-^|KnBYFqon+j&Yv(96ML-Bs8XTMne?@Tg$5Z?CFhpqC4M3
z^H*<&+<q5)&>NRK;@)8e>#HJ-$FHkFbZ9{jv75DUHSG|qQV6SY?Z(%?7LEk9UzKZ8
zB&$thM21(g%wx<VuB|pPw#);P{^)v{W7R{};eojTdIz9Nna%WWqEfj)sf+(nf3+lP
zNTfCP<E1X<w{sk@w^h`rXMgpbO$fmL-Sg^`A<{phwzrcEGSrj&z@_WBcRpzeGfy{R
z;I4d2c=rbNk$iMsX4E&mOQOvC#aU{UFx~ps>D=<QoZ{78F*wWlG-w?j`NZ85E9ySy
zVvKy;`;G6l3nzc1KsSM>>oigCjNl56Wo!v*Fo<yO6W&v2>X6|`fXSJ>2CAwDA?({p
zvHYw*M?Y;wh|u83Ux~=g2TUbnAu9!5%e*^T8OKozI4rFiA^5chU~`eM$&=pXr}qXD
zX?59ZAM^}}RhJw*!yz;Po?F6A%DVHxnJ*0--z$Ijyc_`Y%icw<N`uEq0O;Ua5RJw>
z0$GgDbWZOY*{Etd>&fXZ_w$<KB=||05)agN9f-oWt^fry04lXxLpzxavuU}-8BTb6
z2S^qZO0vs+jA@}{^T%4b#b*T5Y2=V2E-4h1;|Xqwc6qcsJS6z3YF^>#y}fRlEZur6
zz==!W{2YZNL6w<X@#rTzP35{TTGa2~I~?HW>Qm5EETyXvCZa#T(pYgXF}<VV0xfv$
z5aZh_1*18PUIMnet$(l=K&lm!*aG_f>@4C690?wvb#pRa`xWLI8NvBB<)6M*nXpZ4
zuO5v}PiXQ!fZ7Gtw9>6rNJlaHBiba8)~oE}bS}=1Ay=SQ=sveQJqv*?+Scc3U~j50
z1My4hYTT6)@GfmHeHsEB1qI$<s1KN@pI_4*CWtpK<*E@Ov(W(fLy=sK%lSP^L5^<Q
z0U*1Y^YUY)3m70eaTi|S%kgngtZ~af^iTV!PBAG`YrD=zwGe@uQadB0a?neRDXqxZ
zmktE>S#-r@J}EO!lTj3uqvoky3H@{|t@ogHa=;t?&Jz1*@~eKl_UNAb2mEs7V48q}
zfcHnXlqFC@P$irTej<OW7!5$MF7$5^{rcITO~AFIu*Vg^X>)_VfH(&zu<}W?b1dti
zZam@wVY~sEbPIx5ZmY52v@XLA_%)dUMsdJt-%B!kz$l_h5H5|{+hg^ANhlv}V!mfX
z>tun>EKG7Q*23O88s5Fg`6nqtl7r2OpU461cda5x(<-Qi>}d^G;5XW&Z(bm6Fi5bA
z)GL^uhqryo6nnLYZ~--Go7TF&X06GNOe(`cEvt$&F2hE!x4C6nG5&vJ_<4}K#Z<+5
z-P{q=G7(h0dzn1U5agJnRC;6U03L}_n4E_X*Jnwl|G?b~+dzbV{KO+*hMN#wBO&Zk
z<@4nMbcxnV`UUpCq@qYxXUf0mCb&&es`cv*`M-v}i33_VmhfKx(3ss_7N$xtq0v{Z
z|1x@SWd-0RwQj(=wJbmhC<Q3hckhsCCXLi|VouCZ{c}5P8nh_djm~LCtvntj7O_7m
zrlX=cNHIY|$m)-F3%7^vCD<S#jjW?zON_uFLu?Reo$^p4B|S^1>w**a@Z89SQj5LX
z5GMmUu1CMiUnLk=cdYsgJ?(<l;dbP+6RRaZO^nyc%Z#qc*My8I@^j;u>&rz32FWsQ
zl)N8l8hi~rtTTmS>2%pqn#zIH5chB2oIN2be=sbDObo(KP}&6-(2o^n-G!2YnHcXd
zDj5qXfMC>KNG_n-peMp`Bv`2rH3Fx<F!`j|lLP?$Bm9p#y<bebRk~Zq^tn}>QNgcg
z0PzbUQG#_fF`Qjs^wI9n%&YSUA<Dpf^dX@fwXg<r>FLkH-z3&0NcSd^!LNzBi?5Dx
zrmQB^+SWzJ96)(e+6!*C1<B51d^v`Fw~P!f?KxMReF)PxETCe$`xn3Z#=CqBxGiuK
z{8l`{GT$<xBqf&KD=g%W_1Ww-pefjbhYQ|1&eF8cSFXfBQedb9C<r+LSq88DT_*86
zLzs&Ij^R0pX3Fd|<I9OKFatoYtp>EqR$ZLs(AbZF7IYZaRx<6v86XB8_D2pW7t~dH
z2Uj<gRBn?&pqm2t9zuM(8@#?{xw7QRU#~QB`jMU5<PT66RoTV)T;xTa!$JN(!|zC@
zJWdUx!daFQ$0LX3wwG-i<#__Cw=R=YGd#zPS(mx@m&FF|E}^4X6c4EWRRbk~;0e2$
zmlw~cNf|9(uHn>B1h%ClHNAc$yjV^8lI>#Wv5}=wFcy>a07B}eVv_CRH}B8W`{}c?
z>3c9P={?v8tCr}YE&Jn{$ih^HK21h!FRx6fcvRjgtvw@j5d_WKOC6oe=Mm-!$6@ms
zbI;C#rXtdUM2dO3OR5%}H@iiyW|{*0FuxSXSl@FrhukR{yT5nznsOOFF=&L%>B(v@
z>fm+N+aU8Vz|~pRp5Av`%v&U-&VsP<admG^%8&h#*~}Mp)!9ogYSA1d2O(y(zl=-P
zTYrsJ2coyagsZ>)mE|G&5_VZ}50G_#Iy9eh@E{}$OdjZ8cm<J#H5jC`(un~FK6fol
z#TM8<Xw%xd^SRy!rsX`XTW9Nem4>ZV7o~3E?D5dc8tV&&gFX1O8F1Eu%Xg4JNTezp
zv8YKJ^B`={J1XGxY;eoUk&!Gf#g@d{sRE2$b!E02qWmA=CM#)GhuRvjCT?picg=4k
zP;Ps69|?X;ayj!l`9VI*qo)t9w9R&D_GZ>P&O2r)9KZh}N0XOM@$u=_>+)I1X0-nO
zOBz(CKfEs*d)=jv(7XITC;XZ7cRJZCqFiqtK7ojRbuFfOm~;Ki=daltUB#ZIxrV3S
zRjpOd4~lr5Ub@qzf-QWvZM1fgKS~lxww#Aa7uaaKNy=o8>CA9(#TEtseXYav3Di<`
zK*)T{zuWmm75Q<8BPbTo)?Oj&xll>-W+LA6ijO#}g01_!l>-e<5GkXX?iOoHjXp<{
zaj{M4Bm|LDHXC0!T6}iof%)I4xa^JTR?1j17mV8osZ!uQeGoFpY05r!=yf<`rH@2J
z#1Jd_`KA614Q_DcdG#R=vma&j2*C)jk@#NKw0){qKxNUVXr=A%lH5=|W!PwPYJuCp
z^9iOko$G+9bl`=(npLMNsoyB8N${OyF;LnWHm>MX#$A?aHA$mE<GaG@;=v#{T#dKI
zmA}x|3tk1hkvRHlYyC~+U*%p-#N1QbN&e9rxXWSvt=sF<%P5Q&*qL2iXmv%E-PsLS
zH2%K%ngXz9sswGE{_?KSOZ1FxdDsCI>cskTqru%{4nEkkpR-Xu8h_xT)?28UzHlp#
zpM6VrjR%KgGGPq3XD?7ES4`=1UQGO7d)FD(bhh+0h7~Lzc(H&2t^_tB1VNNetQfjM
zK@!?tL=X^^D&6R+YehpSQbJT(L_m;834!dYumR~sP$0N~fly5h5L({z53tX3@BRLM
zk}vQie@@PsGxIyYnKLt~Mj#4z_(_ZdgVk<<4W$+PjVr3po#6oH;T|C2e~u9{w}(1u
z$*W{U*Nlo)4?@<tV-!{8#)pOh(-OL(d^0&q%r)i>uvI3>kJteZ&@MWn4vg#Pl%r5#
ziMxda_g-DGz&GdM^?L9zQ!a!Dd8PX`k=xh(lk4hz8*SZ~N*^9RJMR4)*pvuVi4S@F
zN`Z4*CA_jbOE;gn=n2iR2sV@1DfAT1iHwSB4WYgo16#XFzIJBVjxXGt)X+0Ew_TNG
z_PKsOV(5GKy$gQizjqa=quu2IZC6{{@wj1I?)J)I$JTt`__?hOHv4vi1u3Y%Leh+?
z_gC}y4=mN7%yi;x71ngq>WZT^nir=zT!+m8qOZyy6rRvt2jC@bt})}F%~`AS2qySb
zBB(9jG4rd`Iw=Y`KH$?X{Q)|R*OdFtMbT_!0n~RWLcM?9LeTihdIcc}cVu(%hBPhx
zZ#^FV$;te*qu0>(fUTw&?krbxp*&0rJw9R9AB`~yX+;GA_<+yhC{?umc+uUe0u^I$
zi=zJON=G6hi*%0|{8K^YpR2>h$mR-mzirD#rG<39t?ouW9#TcSi@?Cs?8bOa*<4Go
z>w2awpNQcvj$C8HH<I>iCB4KS_#?*<V$vqEU6q0LuPf36PY^T4SU0u)*`+Q^LPQuf
zJ?VpKSk3ww-dfaNG(`e{ujY<t8wR%L-s5DX5)%f@SW;jj-uKW9RMx`D^7FlJC0UP^
z;F6oNnH+Z1{M*opu__b`wK@A*=jw!Oq+U)Ywy;euekl_5m7Ryls%TTexbbzPQ!#R2
z+{H-}&kG|xnbT7YslFWi5Tcj8zrLX7pp=&p5zN{CXGjukWY$ZYz$p^^vY1ssILSPW
z<r57!O40J*ZtEv|DEiDP5X5cCy%)Q9g*7~c6e>T0po0RNY-;TFBIzkkC^s+p0POuY
zL6Kc1m~Hefu;OAg3#hBne99eItdU_9Gk_d5ZQI-S?y`=oO0y!cs6KsqRuh~{l$bol
z4VZEAy8<b!DFN$>*K#t_sXAs17!2XUMrWhI@hREcCSOCSPQ=W%v69x+V8xtkKiik;
z)#RejXGF7)gZ-}21R+HKqw7n~Qnb=bwL_XdSImO@c0%YzlAmpzWj(_8a0xv(6o_$_
z%&bYklU*&!br5lXFK2Z@>q5eO!xARKZE;sWxWorm9Fy8I4CVS(DL!E`{sl?)s0C*n
zWAY|xrUEK?GWl1TX?o>L=gJ9@V9Lf3%#0!n%cvatdMsz$a4OCciSjwFvS-yO#z7Ys
zJlE0&r_Uc4&xP!ED^$2}n%H<TA8iu0^mu1uhMLeUUy};Xts6)?ZfZSZ)XgAyhz5LB
z<_!E$LHAr`dqg-!dgz}Szp91_`bfNr0KfUxtfpPzUndq#-Uk6Fb_DE(@w&tBx&*gX
zSe=pb(RQlnAvorlXC%ig)T`~aT*HKCoF{A`5I^nCe_?bXxPDk3LvDE&1$GEfF(6L`
zO}oIMse+cE3!0|W8V?$gyMXFIZ9B3Sv?dP&fc+zgSlNM}6?`5XM4<btC3_yVEciQ>
zPpFBLCZ^aDhw?uKhFUl!o>x!?9|URG8iF#)j@Z;OI25#Pcl2(k7S=c!kcS&m!f2S8
zik*4mRqqM*w@y`lKX{x{4X;Z982fCgZa)+(i9lT*NI@y<p1G}6$Z8TSg&tl#w&srZ
zD`pLLsQagW>(s$kL_a<UvaYj%2jNp^u7a-K(2n5OR!!D<E~dEo1J$<ZCzApl%FmDA
zBKYL*hja(|>WbhJ3dOpsDzlSwT4^-}t>gp+Z_#IZrvB#ZZ4I+0@fIV%*9rj87%?!G
zn7E;xZ);|d-k3@>&d-wX0N$%eVoAp+th)GR_hoaDV+$ll#`YcaeOTK4s#-@&L0{j4
zWkH|^;7di>OVd3kbj?1CjktL`FyiO~YI{<(?2<lI>$6(3Ii$EXgtc39oz5i|(u0|}
zJjm<A$d+AfNYtg%UjD**8XrgiTDrVpo8@y^E(oWCUM(%7iVCMeL(sY7Kn-kvLR{cW
z084AaaJw8Wce58t<m6PE0|{}Sxs@-ROF}*BBVT~d$OQ6+sEF=MaSDbilQz72CX6o6
z)jgxiUR7anU2box{(`CI=gw!VUB^pp@rH^SQnRV`q>b&mO~xuRjG0za<-hCRN&-Ej
zY^coMkDsYHqFHzt+POWK0XJ~P#_M6EN8d+Jtf-iz6@j__B%QuJJ*w{tKD*!F5zGwA
zOG%V=$fwE!lM~@AoYLDpat_`345;%i!W=8O?pffs3foOKUJ!5A4uP>3eW+oo1|eK*
z*=i4M?LT0886%-}8n!t7f#+MJiU)x9ZP|!ahQrk`6fNUR5z-2-Pq=wmYInMd;F1fs
z(fND9Ir$L=p*Nv#V^2v0=({Dz$HZjMl+stKlT@^S1tr@&Zue)e1<X!luar^85eQ6B
zVBbu4Nm><ylYBBMzW7gdq}t9;-{3!;rVJ{#YTm2o(L@J-7lZ&aM%%-=r^fwE$KE7$
z;%fSI1hUpF#LD$N+8qU-D9m!6W;U<eZXy^PD4l#i%79v?%>!Vgv4fRVUj@KISPPz*
zZIhqvT_J!}-lv_?p5@s+mk<S+7sUXU>JT;bXw!J5HtU&cTS~muY6c`TcI%G$$RWC6
zdW)Iyx3qz(I*6e$fN#aUS!^U3<w((bGE+hd#Q2ccLRlK~>UL(2{JkzZ6-|F#jO??A
zR6)BN$a)Fu;kL)j?AxJ{`0??T5^VPj2~#^?&h>JRlHg`^Oe@rlX$L5WZP%YNws4P?
z;QlS-WDbG}&CG0S0>_m%Oz@_?@usKnlg3Z^<yGM6VMl`E#~#yCdwMK!5V`p22(VJ|
ziq686CsjFVWy8a^v=<3FfIU6{TF1U`(fmfL{RIEQdHsM>NR&VRaJ1x>WOqW!*>cfe
zY*au*+e0bS?kmt3&Fcu(zHcg3X!jY>{3|A9jq>7=A?f|AtbjCE(Y?_t1`frqvY|d4
z`vkq#Dj4xj^2{+dFo3lL<t?-Hb{@AlYTWFxSVNY!ca=oc=V9x8v4K}brD8jWR<VvS
zimliiv`s0<TKPPvY9p)#0#>-TC$CEx?A&#J)uPEyuBq)e=Jytn;F4A;=!V3A$vijv
z#KOp#%-%2V(4v#&2)3@F{HJYU$|b6-JmM}L#2)9B@I#3+#WmG1_)A()+}!#4d%$eb
z84&01X1=>rie^lGjNL~Mi_0YGe(pU3@&sc_(BaYKGT+R~%Wr=Xmm7N*GPC8~qORVO
zj=DrlDlN?d5djE3&;b%$P-^LzQF8ffZDeHOXYlH)Fre6xoWZm3h<yN&B5_#P^g)AV
zT(HVw&bMk7YG`!!87J)N5r-QX^0Q-EVM>g=iI_{&o%2=CjxiX$msZB#2oKG<a11&0
zH$*Z0k0{Fn(o+O>^4{${oM>Zv7T?CUbm54F&9LZ@0${y+;RP(a@aS^yaoTUt1l=!7
z9CC@M9cb-cyp?Gjds+VA3PMjG=}*Okd)=z9Gw@OlnKw1&WI;020T_aO9(I7h08DFp
zS~?bG#rvHRTc-hFs|3yRB08VZaw;uoF}MjO)B@=9a}F8Bq_>u5sx?iFyw_Suvz<H9
zuR#?QyU;^L-{KQZFW{?xON7$PJRp;omi&GPz1zS8H^IJggqX!%Qk^D5%i)0s8DmEo
zd>CAh01AHi?J^-Wv*@qU!^alHSLo>3X0C?IPw{Tziv1@3e9k@rWsWa`)3u{8K0eZ2
zf-B51(MkV;z%=eW2Y)a1J%{-wbAOX2JuzXk;>r2f*V`evHv1B+^WK=6rAH9npD7&b
zUG8t6^0QqbRnPY;*@(~0RwY6M&x4RQtvZmrSNlCIWaRn)(JsyB#+o$>WX$m+XVNlo
zZl>=~JKI78J=ciXc-IQ0QTLbp=U>H4fUqdrRTfy^R&QrurYEX5av6)Gcr7c9o7C7o
zs-s%^(Ch6DT|JJhn;_qU83#OT-j^ekCJC+tUpb>CEtQxw9WMV(3iyYPJ}&7`Th0#s
zy?Lf2Kg|?cgSH+lRf{oj#WOhgil#81is^S)Hhp(Fi8on;yq0vQD?H+JkhT=L3xYf(
z7(PWVG)XS`DxqTbdj@7?zZI|ej#=*=xa8Ok=0jCq9mC4(mr;|w^HXU_Vtzni0C&$b
zZ1KF;Ou=hQ-d$DmF~JSjmDX?vttqL4Z}kMcb#G$%g%=)t_|#o;jR&dRGspw7Xf=Q<
z{A}0cb~(y0{4BP~=alLJ$xU~e(<ELE>C557du+bo2r^$BAhpV8Zq5U78hzn7(QsNO
zzNW9WTV@5vWZ*u$_gKCcjbxTe_paF;y~XJ`c<VlBvgFhAo3u)^6<Ey)^Q>UECnUBd
z_m2iEq&&evvgb+4;T)qVCnsTF0PWdhIq|M1!4w?>!oioonX%&Mimx?8FQO@nuo?BC
zCd<+)kLhsp*q4hTP*DYQ26Rsn{tg}YhP-FTg4||al1u_PfoJMRUw}Tyunfxvd>t<&
zwx`~*cR3$H$*MoQ18+-3fY5TA(ikmfrQB?ofMyBw$``J!%VETJI?0GG$oK~X5(`%V
zbyyvL6-l0O4+IXV5M|v^A8&jm(Esv2fPT5c+wF6m#s6MRSl8?zZl*fyL1-%Ii+JbF
z4NwceXiaXeJ$roFmKT4yjJNJ<Lkf5Rk^5TXB+f&`_c&O9bxAH?@;d>|aQ`EyZJhjs
z<Tiez>}NUeTWD6r{jZli-&e$XpHxALm6F-F(GM3gZ|^%ob|QvjvL_egU03e4$DT9i
zuv{Z?Xd^nR5pu7aq2P6A+_@t#GaNVp&3wn)KE%v$j-8WLnPS7M$T;?69NX2wS%?rT
zE1JTVFr81ND;G!BK5UCOoc3boOh+mU#S{#^7j98tTSvbwW9-c+9rS&799&;fo&u=Y
zy!-w+cmxu#+vLRg&;<tz*TnPwsL9angj*XH2D;?-)qkwcQ>8uK`F=T1Ye23mcw~2+
zM#%RMIKCq0;jWLrDlK1Fty8xvCE@yB(<fp$pdPc-v!AO0st*Z^;GFp+b1-g;_iG@f
zCFDq_E>UOt&K;Xn&`h#_JzVJ&Zz7mC0THL>Wk9;2`s!jwCr4f2i41R45p*sWqK>7a
z8KtVRE|!d_#y>p@x=Xzn;#@;;DFiwqPaTus8h9BWTt9yG3f3)+r0(%^Gmd~WBV+qg
zN6fzU>F{!9Xm4`<EmlZCXH03l#=W<V=x%mCUvmaWmI$q`lyLFYM2WNen8wq<vxeo|
z(a-+EVO?`_Ieq9RRNlbM1SgucLjDb@5_5{H_9T*;E<p_;lm=aFa$aA_a-lk<esaNs
zEg4x`YPYPM3w8us6n*^kvA2SoNKaQ{j{CA4*X_0Ap=gPD6|@Bbi=#xl=pF&pwGl8N
zvY2w-y>_d2D%!<(QUW>jT@~dGI0O#WY|UHBe2J>>TnZWeyl)vJ)a(FuyP0O&->zA@
z(%O}=&davH<>B&2GyoN+<LWv*%N9D)UdL51c0SF7&Qm%7k578Gmn6Si0lqo@`!4+`
zoR5iS*22*ytx#D)w0y464l6P?m|gS~(kFU9)-qN?P_r>0;Nu_b#~Qv)`e0c%uw1N8
zFqNOhS(qmKm+$LcXL^vcU>=ia8~Hl~h*ba(xv8h6EGXrGV_Y=Pkv2oI_*mZkuWAW?
z*Wd-#n#!jtlz*xJ$TmPT&Ms%fnaFpQaXaHJMk-aPP1bciWY>x1!r51GFoe*#v@_2V
z>Dc5H{5@oTwoDvwW9oy5NR8AmXl+sD2K?(VD<6)d{b*tK;FBoH?{@u@_Vs7kOXs`X
z5{4YFM&4BVlfInya$m}?>U*Mjd&?JlIwB`A#Vc#@GGK8KzTTCXz%-a!@V-24X_0Bb
zMU6RT>&J~koF3{ku+AiANXn`JZ6Yc$zu-OVh#f6d9V*92C-*3(o4UhfR0TJvL)=Z{
zRTa*aU&w1m3$*|i0aEc2y_=feaQjlYMMb*3FN=yHC%<ExDH<L41AJ(+0%1MT+08;Y
z-$IgUa+w{rMr6IoPX8JF<8aX0*geDiwRdePZgB2j7uSES`ohT`Od=Yqvkoj0T^dhH
z^$>|No@4VzdyD69c)q>Wax!b{O0%=sr?3oMIOj2FUyDk>)+E|D*#xRC7g}q_5F!G6
z%J;BBvc6V~te=>z9cVI*shw>Kfk6|tx&>(R`CP;JHH7u@Al8&l!jTisk8x15+h5CA
z97;yz?zCo^6XzapN{~sq*V%h8c?(V)v?GK{`Bemu+WX`;Y0*q|k+#-4442I`F+Ih|
z?TGZcKZ+eCOSF&iFD%uo_70qbq>~H>aEC#MfT-@gni!6D8^xMF7jW!(LF{PDq-dje
zd3aU3B?B5LW^s<cNH0Y0RSbQTo3>IS$dr*3M`Yeg8%?u%aW+V#89rX*L_Bt6nJTnL
zArT+dhbn>%KQ<X~T0|D&gzq^h8WHQIZb0bV(EeCx{*{nMO!3O>^vWc~<M9qf2MD&~
z!uk^<u1UoM9z%auFBoovCM#xrHC3W^>)V%};C(nG|AwYSqL*rTSN~n;LYhF39ai(S
zr}mE05g(IUT$Iy;mTcK3+T-#N4Os(jVm31$omFj~gbC7gl~P8GG7?+nN=QYe!6RfM
z=}XL{9z|GN1^VtI1zV|tt(cix#eQ$!tfkqle3wnvMnVeCA6FFOa7$;p!#;;J=oN1q
zh*A=WeJY@645JZ08)~X=ae1O&RW;s}UvffsBEauFq-kjI?R?v?18Tv5ecA7M#fh}b
z;tvLo#@qAWsj6F_kPb6VMhk<nA_P^G=FiEfze`h89+RqGy<@wo4TXJ84bvlzp48Pq
zs0aJl<RWZO1^#uzWK`V0nvw&XK?r-gTB-44PUr-k3gNq;hx0zR_GRKbRRs_d?)d$H
zB#2mEmEdaFn*7ohr;qfU>3Zg4gXO3W9nT=`C>(P&6Xmz}kZih1{$cjF0zIS^7%elg
zpM1?wD`|B#NbTu_Gr%)5d>w40KIVi)y!Ht^ij7fY9T_n08-O-xYCs}1X>eojpf20V
zcu3dN@leLAt2LU4tUs4I^a@qp;!Q&{`oGs^CpuA|B;d%8h_&jGF*X4q-!K}E6PDa+
zwIM!I5QJHDK53MJgxrdYj)tohuG%()REHShH`=DA7wL1qVe7wPOte@Yl>)Z@uO=X_
zyk<kYCZ^m#?{E?^l}_t(zI7{r9Mt$+6td>P8R*<3M1!ooPv+Vvbu?u+Ai$L##JR3a
zej9Zt$9~iU8|YWe;T-$a5?GCo(q$#Z2U@l=5VWWRwE6d&5>O<|PId)5Ct@|<tXZ3&
z=m46{{pB5PHQw$)<aHI9$s~%L9TIBw_7~_wP;!%*WdE8*iooZ4S`v*ca(Zw&5u<lD
z#tEZPurhIp=qD)0q_5lO<TYl@hW3p%^Q`;Q4OZ&9tqu1xcO}>;C8U@mgjChIQ-QNF
zN@g8#i+Ezu47W-H5tf~56dw_3*q!44=7M*yW%<>&Cic5q7sd7hR;niH=NeLbA7>tC
ztsY0EHNMO;-cM6<ts|RlR3ZmU+j6@^%JclE%HVnL7feTD&mtpZh}=Hm*f!D=@6Vtc
zk?7Jr1{EQle|yxra+X6A$p<0Scpjp?7ItY{7vCgLT7T?|fwp-z{-IS)&~YD=pGTC)
zq_@Ofi7CgxB>~cNEY(+;m~z9urf=UX8_PNZrY7bgBQzk!&$3vTY81~y`%Kufx(?r7
zbnsbpz)o1Rrd2-^D>O$I@D{?86}<tO_SK2$LR{!2I=ylvAZ)C`)AgS$X%GX+i1?ho
zUMAM}`;-n&rpZp)w&4p>U?fU4E>UZ<WR;{emgr1Rw;<ZEqa`?Ay4%MhH%?@=P4<wH
zS?$vP5wNCpNV`evx-K?<dcmbo_46p%?c=48rdSP8ABGVIe)nFYC5~r>g5m~x>4VOv
z<KX(O`Hhp8Nz~YG?S<vg)dN7deBSK<3~wbN?|=SSjluqt<M2NZHcJAzp1gkq+%ZFl
z@jvea8W9Oj-A_jNm9L=_pK$;6;E##?x9EOI<cCClSj~Uvzz-bxZ;1UNkslKIar*wJ
z4*cLFz*+kbD)vJnKP0l8N%4clulgPTZ+^|Jh@8IWRq`#5xi#PibHe6$!BO|?{|Dl<
Bsw4ma

diff --git a/docs/logos/vllm.png b/docs/logos/vllm.png
deleted file mode 100644
index 1ead9972879c29e17160f1d3ead150482467b1e8..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 89976
zcmeFaWn5Hi_dg6HD4_y^q9QP&gpx}4C?e9`Eh^I8ISL|@2GU_rA|c(3prTSrcS%YO
z-SAv{XwUhb`~UKJ;kiHOb8q(SeZ}haUF+J%S58Lk7{OTrEG(>J;<s)nU||vFV__ZA
z!aEG^)UozI!NS6(Hxdz%GZ7IJu`so;R<zVNxO?r6#qGNaV%M+m^73F|F+4TU(^I&`
z#PC)Bgr45l8pcZm)(#4fANwoldDNo5z(2uj;a}m=VOm<fr-^$juyEMUo^&nxg2H>v
zI`*Kt{6!^UzJh|^bL1Q~yjF&V;j+G-UNu%`NR=c3Vw?)DdGQiOwGP&8a%@VxI~Ld<
z8qSZC>Twb;P-a8~!wuo9hk|QDaqh7r$NMFHGH?;5NhNUjNZ7O37w@oUN*=w6kN2kY
zb%*0gC@@MOvTfY2-Ic%ReAD;A1-7>eCk1Ao&D)asANBV?a~jciTz;LG_*{(tZA0SQ
zc%H<O7e1nT%U&RngGAS>--aUfk;e}m+Enrii#UJip{LNHic;Qh@uGL-3{Ho8+Mnn@
zgmvCkLN!)GxCc<~#$7e>d(zTam%%k277Y6Y)**0(4gQDaV}yl+xyHhx1OJbO1&f32
z+#$@5JG66sNDKPWSYt>HJc4Vaq-LuoEyZVGVaB2lQJ=-p%o2J4OTdv2T$<gr)rUKp
znVQ@1ISSHXp5O!5&}~*4IOY*s6G0j^X*sxvh4o!H7YiE;8;uYF91a(-zH^UH;l|DF
z*TG+cH1}<7E%{hk9UL539N1YbtPNQKt9e=3uCiXe$_$=hwsA7I)pukzx4E!2$j&%7
z?%Ehw8(G>KS(wA2arJLo*x3rw&_EOI{<F1CTcdk>Gnw0LZwqXY75ats3JV+SzrA+X
z(dhqp4f<v4HO#)YrW1e$<CC>Ex(gNzElcQ%0A_*%_x6@20FA;YXXJR-RPBb5*<EuR
z=pA-8HZ}p)fBorSUsd0Gm51xVoBMv*ep7%IBGf*@ZE*#27qCT$K!A0Z#X<x<eZLH`
zun<_{H?Ao;VvlzbW`{Jn!O=FyUz2=*NxyG@b>(BD|1mCcvFv-I?*ghc*p=kHE=F@j
zzioIsWtAh%ZkCwN;dbY$Swn(&Km?a$)nKV3&IPg?@`p)~^?EOED6w6W>xKv6Eq1mI
zKQfV@8gm`)oE2P{djGj_(Y42YZL`d6DsOqR%`jHs+e0jD7!DpW9O3!*;x7^YO2S_)
z_^SnfwcxK7{MCZLTJTp3{%XNrE%>Vif3@JR7W~zMzgqBD3;t@sUoH5n1%I{RuNM5(
zg1=huR}21X!Cx)-|CtsHNX12EIeJSTtY^L^S1WOz5C8QgAd=%=b=cI@#6<LI1l^0O
z!K{jw>AoKIzSi)*u`_+G6y=MCULFP&vmg4F!_f2D==6m}PYE6!eBD*5KCKC-`4_`E
zQ>(E)iX}~Q31>?e8cUmNz0(EhgIY=#sD<5UR%48F+FdfGCbVQzCoIQrOsq=8E10CD
z*Tlrcl;+C398Fgo`xP=J($+1Pm|mt)q8QT>xR_)2eR^ZK<RgQvda91)N8BFAhRAN)
zq*D)%K*iYs5tOT8X;r1{I|h7W_>X*4q21V*7^k6d!&n}x{wPZi#Z~8p)(-}iJ}yyK
zujaGLnQJJ@r%P9h(Q9*x8}nUb-qr@?Wew$mO=0b2Xt$nqouyLyx)z5`^3nxC<KdOE
zsm-(B=X+M?Jtk743Gc2>wi)-deD@NYph}%`XNz(f)OoLxZNGMBxzyz&@suyR*nU%$
zU#BRP(C2cctkIkFAj?vfQv0$c2^acRhNj7rBGDW6tCJ)m{BHBmGAf=An^YPnu1ub<
zSakcaNKNRo+CJ(*HA%Yqy*aR@)Tw5=)L<P~hJPZvTcbZGJygs3+auzRX}NBwG+l7Q
zaiPtyg^JI1BvbK0ns(Jn+Hl{PFFM_xa<XVBwwP;SeSLC(vpBu;+muPm!bZWX%@wVb
z9%1e|XeZz2^PIltl%<bfnrLRpTc4QMZZR2_iJ9_CR_T3F5SK1>KhL8(IV5?oezCY&
zcBz!?d(fTI(V~r(6~4QrqrxgCc^0>Md3&Oqi%%)^yb(@&F1q@?QD&}XG}dCJ&PUPB
zraHOO#6~Jr#w=#kLn=mK`J4E2I?ANaz1Oy-@3DH^!s}cN7x$IegGrIhnz_Zpzo<8t
z-bHyuhO<>%_hP)x7Icpej)a~=1;d7l&O03zG%BTd@dY2qz#SEMwUg>2m*UOPHZ0tQ
zC#Xc>_0b@2`pu>9*gfDC_z{>`5LF5K5cLx3t`8Bl5NCTk#an@^mn}QY^0X<hyRg0K
z!}r@ZRzI#HTS`RA?h1}Q{=<AFX~M^cX5H*m6_A);2Bc7~p;_3ZERY;tJcVb|?e!@{
z*)$oUs@l@i#MNeIl*o4h;fZ(%^CM^7NNHg*x5yA}0#dr&PLO>Q1zT+98S_3k4~Hn+
z)i{;%<%wI5f;O1xlM)D^@!ww5E0sKKQ7So<a0>igPljg`gWF(q96{-pu3P3l{#;hb
zNs>~Tj22Az4%R2lYcW1K853em%$KeimzY@Db7f3LackYg46i+LCq>UG^OEc=1zHnA
zANO*|_pow0`YT{QVz{LO%5_)_^U7@!w&6jOw^IefF&z~T<QE5*&vN}d@mNJ~v_8J9
z%w=)%{%j}@v`Wl4rMOI%b&E^BF$eZVLyA|l>U3T(^GYa0u-jy8J)e1gThoDwqhR)y
zbuIbn+C>ZFi80Q<S6oGzHK0fEEWVuSX=_`ixH80IxIH`;@;07Yua3Xf3*4O<G4tG@
zkz)yEwhX%SK3?1t3mJx6?QkWiv=fk<I1~{!TdWhAbv~BcqB~l@FUhK^EgwG=VG0Wp
zS{bR#HS0K~***g(x3yq~U{8uE*s%nq1Dzjm0;DyWIv4M^3`Uwfx=7<d@bDX^-oj|C
zR@YE0>5~wTn0gf@6{tLMy3#dres@l}4)d$hCcT6Br$Iy9?PcSvso9~M`<6X=R|ber
z(_#~Lis7qd9gG(AIB=NUPajh@I{|T~8K>Gooq}W<dJV-u;HcQb&>wiSp>Ok2e1_#*
zT83w1Yrx?QE_~wqYI+R`(pja`0TyaRjTv|NjwxL}IV;IG=e)hUt8~4>*=Hz-`}S4<
z$pG>dj;rgs9A7OY5eRuOo#+{HC(Lw~u%Y5UuS_TR&reQTv}GCAd3?KP`<ey;zlTH*
zMM{W?Rg&N43g+M81>y=0lO^jHp6?G1a-Rf`|8qg;zjDWumo@g6TmL8D`pyEIscb(N
zp-S0CoF`1^@vORyfe22MS{Ca-YoJM6BR&>E`Sn}*XEE(*5?E*)9A@yv--<O>kE!A{
z3H3xIYKtYbKyB#usSVq3d~%Ok&*n<{CZAnTshe=U8ly2Oc;5+@!`)@;;NX$$#BA)l
zyLf~%OovRiPN(i5f8xxM?UhGcw8e2TVj0YiJ*!=NSSlZSDp(B_6&l7e|D-aIlF`o2
zt}DbD4ud!+j_fLOC>z^Yo$_EO-VtYPm?$y&wIj(*`QrzsAb#!{{ng;CSpkp$xDFV;
z*nrgry!44queNIXagWPXNvOB?q@s1ygCF?B48f?i<=HYvKt<SHGQ-P*=*R3f>R?|O
zqUkW}Ay-ob1$Jq;bzhwv4HIh?yX6W^{NpH?gPkk~W1XFdfx_#*IP5!(?yHCo3=CvF
zYTd36X3v(=+uT@JG`1)-QupDb+2IMC?K|vLe$ds7p#x0uM6lz&YkKFhj1w0k@%i;r
z9P@X8Rd8l7(^V(%FLw?^t^^<e((6NWCHH^6HVHb%I};nx5TOG7BBl538SN&$mftgF
zl^y#G<nu6YF+<G9Ya}Ojgl>ylr=?J0jmPPJVJs1uILcM^?avp4&@)GA^=c_3=&6Ux
z)4s)Ln9S%7T+om|0hhah4ApTSJu*}m>tR4*^PF^NJVY_rO(p-rRr=ixf(zt1g4S?O
zpNDjJAl@Zuf(f|82{0U2#J16qIWozq@8ca$4czgMf^^>TOJFz+8uSklqV?|X5*N+-
zP#LhP?GdvjfbmxmAM@FjG|UCZwYa>zmJ(OebDS7Pw4Gw>x$EM3Lk!|$4Jq3&XWiAB
zm-Z9w${GdMgVYIU9+u)0KURcSUogze%iGk5Y9nVK1JRWA753*Rx3?1~gw6V-WjLN=
zXM0;y-qJ)7p2NQhFK(@5Mn+x+`?w{Kvg5}tU1q_K!2GET<apyW2kiqO0J_I*ro9q4
z5=g+_{pEW5e5f~?COYzhGO}+&6xiC;%mqE9i~ARew|lps-{R=>rXR*0f~fZR8p@Tt
zf}!duv<72fS{|&O6WCmvjbEKA+jJm&7RZ4|{1u6CRIC~8tKp3Oel6+x#a$lA<pIBC
z_dI#`#@0nWDITj<L8xqyE=Eaz&;arBi?b7fNR~_(HmgAWAmQ+j?(Pj=wuE01?MYw2
zn%EwrTFN%&iA=kT+j<C&Ok0I*5d}zR4{YO2<kU2&-~n>opp#2h8lNO@3zG~5k6s}N
zo4S3@ju&VcTm~@z!<Ti?G02<>?0(CWkPuU_i`NK8_L?xK*&#O}zYo6qEDAfYkQR?#
zbM2B1bAj-~*;19}QvnwdicnB+2~6^=FLLTxL9!?iKMve#C!^m%GClg_i)DC(vUR7U
zn{y<<4^#L8>vIh)LJO@1hSLl&2ws6Fp&|Y|!_N~s$``w{bME|okT=^#75*c`Q`Gbr
z>Af^$O(b_{bGq&(VK|`_0tb%_NQiGjeOsH8q5g?yc)LVFVuNi^V|_jkv2{nPfzB#p
z`%!1|UG0Z#1M(i&K}2#g^VTzQO|aRanpG_|Prlry_1{5QNFE&f$<Ad1A^|iC*Xj<V
z<jz<~H4sAt*$oUz@2n5QYNwNEk{l;WUMtA3Ap}7j=$!om@@z-}z~>nO=ZBw4t$uy^
z5Xog0!THt52N<JZHF&X8f`G@yn#X&U6yFXxjO1-moC@rqu7bgL?-oV~qR8<0+Y3Ww
z(GA2pb6NK<ZYzf*i1wtqT|=v-q3yiAq*p7c=wZ7W!|yOTp@05&*v|>kF~~eyqI*C3
zO%_d+I(&Ya=elbT`$>V7cQs%!*xBPQB@bSl7Dd{PHAm;%t$G<t^_hi3+7sag^E<`7
z5@}EwC>rJ@aLI?etS#w6Oy5h$z<`~wc>%Q$>E!`3vVW+n486F4M~^=VhBg~F3W$~P
z+&CC|<mzSMLPk&=7v6g;hk6t+j`vS;F5lBLeu#yv@2l{s)6KVh#a+9kzRT6e*HJaL
zWED)}2iDH;1d*d*$Ey3{;|oj>fAq56PDD=35a!97u6Jl9!d4|+XZ?#=N3OI-jlL|z
zkXw4oGlL$ky(r`jJLNqO63h*()Ues7Pk$t*c#X`E<~>o%i`cTY+(5m%Z%P$ITbHv%
zq7Oe(aB*1;tzem)6!#Nqow)#fQ6HbRjG^H$YoIlE#GVZgJ_BMcc6Kt@`7OhNkpe)F
z3c?o^21^PO7%@h=5{6C92)e(9-a|eGa&^-@q~#f-%6rGz+Gu{(DwDN-Is}{oMkKV<
zYw^fLoa+I}UL=Y<1%lIV*d6VCDsgZ!jJD`uVW+no_f&W*A}A;r4{Su&UGhbmfOlGW
z*(z`?;cjA~WBoxI9Vy*AO9TSP*h@q=x-u4hfXR4`jJ-OgI#&iEIfDlQCJO{cgl!2O
z5*zm=v*yAD)?#!y|6zeqIAHxF616h-^*Zym%)4O<GUTGD#3Hi07YDafGLFl=q7QE3
zHtmrNv-Z@J40*Xlx99F97T@-N!vw6j1keS&{2B-RA4-6hh!tSS=PLCYFE}m_hDs8i
zVp^(Ca(D|2J^?-PMv!pAn>W8~yd|Zs?^%jRBpE2T)9}Py_1N3!0m>mRz!tk4dIpai
zKbN2Ml&g^z0T+TS)aKlyV-&UitTtzMVjR3rEWr7U3uDh5SUpk;-%C5PvufD*6AnhQ
z#6fT?-q;9<08)SeSPdTIZ?}c42X2`Gt17Fa36%RbSyBy&wzjszr?8@^y%jpjch(&w
z{(+@#{dD>+Qq=F1cAjBf%(pk40U#b?c#K+@=wS6&>hq~%zQtF+t5tBhP{^NIU6MMe
zxb-KKlr9h2UifM_)npu^DdZ#5h=X?nD3>G`dZ_4*xiJXR;nKkAM69Gvt@eI=7!PcP
zTf)5H*i$TIF}_z<M^#r>moC9(Wu4|oznNUzIjUn^ulTlYLmV-PZOCt*rrk-y_ttYz
zP}RxRGW0%<XIM4`R;yDpOq1MFhr6~{bcC{KuUxp~^U7t<7$oDX^~Rhn3EZO<L>a_7
ztWVT(ab`X%88a6zDO=B=Das4O24CRJd`GP{uP|vaGi&9egKQ!_Q6PcwMfHF7+gun7
z)ioE_<=a!#U{PQm46T$t-8e|Bf5?qD*~stTE3F&rtMsR?@`{rH&xU$>LCoFq=WEdx
zuFl}Ydv6S=ff(!@Q4ENm_YnaufwX?1&K+hOuh7>V_v$wg+WQ|o$LwL2v)V8YW$=1%
zDTr9tuiSUS0k)q>aAo98S=UUU-7QZ93(iX>^%;7f<CNb;-#^%kT-Y?A@c89>`vWum
z%T7>OFiJ|w*zt)8N@3I`AE0V#aSNlzqoboY1#iQWcZHpJgfs)yMFiV#@IlNG>glrV
z8W4IPng$;3xy%wCsR1cnurLspjW6#nkyv!(rp}=&h&G)S&o%&W?oFarWItgs(VmmS
zwtVZYJJLAkcZ2qCL&fd04-9dh!eL~Va`FF*?ro>SKz1GLt=-uZJi#C;=7~tbF>ZN?
z9zDL9qP#x37v~~gLioVU$m4*6%2a$42x>a-z$hNhxGM+jqA0ZP^T60=eq`&RwNZo;
z03)EV^^LM+Qx<Mxn(<=PZgUq##b{tKoJ=^n)gWlSyg5}fi}=auJs!V>tio>*IM$xD
z?pdL|$A}-x$CZ?<r&pxa7#LJ**W%!<`5^L-M>RJ$FSKl0eR9})ssB_4N^o*I^k96r
zUxFaMUc&d1?Lj-6Ni3+3+Jl_jfZAhix=r0z>>M!V7tjOGWj9(Z1$iz+ch7?mZkvs8
z7ubg0vyv+$BnSC=S_#$fbIm9Ul2RlLKWvJCBTQjiHy)kLU!5!}J8f@4#cp~H7mt|X
zEn;as-AmRcza;pZ_&al?wPWK>o(hx_8&IFMH%?+NAl;hXF$WcP#h{pN&iF(fM3*>E
z;7nl0-@SnT0N}?l3a+U68VmQ?b@_{oAP~@@M|axPX%{%p|K={8^g6UhAR353{U1Yj
zEY8;U24vx;y}xvXJSymJL%c{*1o;zvSU>b$uqF^sbDy+PWL-4>Vp=;(t+a*OLjde)
zz}LD%iwX-T@pRcK6MOggm92p~W}WhR9+MaCO}+QP4nJf_90#6?(Fx90v#Z8hf6W7y
z6rSNQ)h+pwRoC4LCs|@I_(Ny|C2jc}&vuab`%J?&qwc36%oc|h78Y1r>}|}=t28bk
z;8ieSb7O_w*BoDpkWuW8fK&o8rmVMi;=wh-KX{TClKxOmQ8vf##JJEWu(S?*^dBA-
z&Q9`PtH62&@*j0l2oSgEDd7g5<Gt!M-T%2kEQd5;&meqF2F`?+l1G2<)<0Jc?jWpi
zoSmH|fR)+2L%!JWpDYE^W6P<M#SJI50U;%lVHyCtY306TEVl~1y0s6qGxp{s=Mvuw
z2RE@RwmW3;_G~fHd=>RZ2!tK>)1_{!2}2BbUY&pkBjCkP?<xRT)aQhf@*doYAOqrl
z;xn7%vmFJ%j6`q|gfxhUmmo-u7>K?rH$wpF*3^inxPjEdI>sg@MIhc2rMg={!Vm(P
z+;4&|3vv7Qt4EB@dg3VT*PpzUJBI`%2NoHwt6S=G2n2hF?X6y57BC<UXL2nW2OgDt
zklB3t1f1<E2rpk+0a}%DX4XsW!KIHCf#=|<)NcMyio9i#J0A2x<cHf~+(0Tpjo-j8
zxgHiF1+axQv0B!>8nM##2@CF5w9;hqsLf+fc-~oWhr`c65E*krn)0B{3V!BE&ZF?4
zg&E>XHlRz3f_e1zn$D1^6+2Gn=)YPey359K9(X-ZL~DtgDF7S|Dehla-Wy3Cj7{|q
zn*URAW@s!&e!QxGfJ?~qpr*>x_c;m=m&?4<@fWGAz!iJ1&nNf5B;r&cdOwesd^+ge
z5CX6!<D7G{xmPhMni@c4KLi#RLl4Cz!Ss=<%=&r5)C|V%Gc?4Jh$Xi)fWGxmX?E`+
zvGCv<fX)|<fA47X)-sVVaLUU$7iTWDjlF~zr$Mh*Y7(6j1N{ou2BxHNLw(9^s=FjU
z<McvZCRO4RkfU%E@D1r2fb3pRLsXvZN*tmS0DM6%Yj<V;l;?fo1pRd8WSFqWhFE!8
z9Vz?Sm<%xFE5wo@i-Wy=_m%SgC;Nm$2P}HT0|)Eis&{R|&IRHWu8w1$cReVM<;dQ<
z!gYlN&Ss4y*LAh}y}sBZe0PBJg}*ZJi64eoEVlWFN(Lq>_Fdn;*b9=tU@Ihg<q?3{
z43v_UirHTL2lg2oTMF!{JOB8GtGzqTDUd>@pTvv(g8mFZqZX=#s9m~XA+O;SGYH}N
zSNx}pLnJptT6nB>#GhnTS-aV8H4=-3yaQWZm?&a%Jy8)vpOOj{EVenv1LgHLop%rB
z_5YlKmt<soe61nFZ<WIOL>e1DF^?#m$$T^AjaX2O^(SD(cWHZ+9$=f6*K7aV?CII!
ze1FsMUd4|t0|d+U7lH(4e!d>0rG|-;<5@^K{Q_cs@1&sY@*WrO=K`UwIn_k9yLimS
zX>EXSoLG89PFg@~xsB4~o71DXduX>n8&HzG6$cU?iwai{=fLZyE+a>9PqFH(eOq5$
z6q~}*8vshtw2uB-@M1eWBwzv&3i61m@xD#%?rz_O+)-pDwMX%c#pZ%VmfzC=COk4-
zc=7qWt{~Ah`&_5=Xz%?9RxsqObk42+6lvsBJjW9IixS_9X^8EuHY+pkFJ!A8UwVXv
ztR@hc`yo;`w>Dd6$R&i_Bg)rI5DngOvi)@PKSu_Xk|B7nnKxNYAL}}^jsTHqF2eD`
z_tA-o36Bs=bhsmD@FF15h7P7T{uuG_2}@St;O}gemqR3+!f0e5_9AjY?uP%RVyE5$
z1YZPMU^DE_%M>%e8GkM?I`|h!YrfSl=AM=2ytSAw>;o9ngJInLOwUaz?lVd@dt}-Z
zqJ3mW=w;(HzgK7P3pb1@xR>4D0;*L3_+H#`{`f_C9v?oaGzR{^DdoXz3YrzDX>M+=
zN3NgCmm}hiKM4fp8x>X40hz|{V_%fnrDz;JAPG0N{I<{Ap;VkQOsCj9B-<TYHpu^G
zwIfzmK=}*{sRE*6jb!cjrBu!%uLyH3dfc5pWIj8FTqE=vb1ulrnhjhtq`|<iTPAIk
z0JxX5vhIWC8Vg_-b8~Y`6t@VY>@W%2C_wM(=UcXFJNM%O+QXa!&}n-s*1SwzH(;%@
zQGt~F-8t5S-?@pjzpL+PIiU9=Qgr|1|I|vbc7J>(-O@EP*3zZ^_%no(JUFjUCYAKP
zEh<`FXs3`9*vF##QoyW!S!;y0jN#}X_Ia1D%;Y>E3`++4mMO~5i_-DIISfJ?9;dNq
zw36<osOQI_n%edd|3?bGZ`$o=GtbLuO1S`!n0NQWS$X2eno#Z2W43F`o!cCzjT`dx
zk0^fbEG!NSu#jAjIDU5`%*8YwD95o{UOa7XzKxtfVc$1&!ESn!Z%>r&(dSM6@<9U{
z_X4xkc<ToU-6Fprt=o=s^#XDCee20`Z^E9~aL=p6Uq6%UMj;uzb|(o|#rDAeI`-8!
za%Uks``mp5bAe0F12A_<Rb4l`k6ETnC7_nH_#2qk4LO6<Bmg*GfC0E`>EXj8L|pYZ
zUi}6&lK6Os;$M{B-^H))k~#Hc2C5{d+j%bzI5^4ngGfG{do>X`se%1JPr8l9he`lo
z#e-bC{{&qy39`GpJBy!>&z<u^-yIr%O^cHG`*bi8oE`Qe{7sHz?|ciu97-D}e&FwO
zQ9IA=-|c5|$1%;Gl`2&&tIw1Dh7kf4-h||}c!XVblHCs?K%xGmV=wZE`@A=~&Ml1+
z%lYnw+&tw(>Bt4f0|*r}w7d9DsDQtJbYKZ6Uv*Z|!vUB)h|>gx?pPoksdo6*TjTJp
zvk?SvhzNj&k&Kl&Rx#(T<&lI2X7mCuLQ?W2)AB5yV(3+*5^lNOnm<S%WaH1?_`*j7
z%nFlN_9X*-{M{rOZU};sx;9UIsMfGc@qb=C#h-(iQ<W$a?aYx`JSLF`6sJP}!^GRd
z%K%nFh`>x*(xPYQ8e}zHY!s|!2EN6QHYD~?#w>8*P&nMBu=^yqk-3eR$q+2b=2CCf
zGY@Wh|GZzb5d~U8S6mjxgYNNteetj($3)c#WFuvv)~5^Ui31RhBHzMlYisMoo}=+j
z9E;7JF6wihsJV3^L&0ljFM9U@4(qdEg$dsNscO4-uEN>O+OumfDJCef6HD9^0l-cn
zGgF};e-<`@KLUwUx?E$SQv?A2EP@vbBFV+0(iWGsJ=W(>t#Q9i+Rj}7Kgxh3Uq4a0
zPwO$;^M+tHq4vCtH;kBuhY@nNo_Wb7Og1ih60ZnWCSX__)x%`$Uh~xigfsDjRiQo2
z{<+!LJW;){zJQB^fQ9@=SV?-q!`Tft-1240$aIIU?wY65$Dqo-h;rn9^6gok9TK0`
zL5<~9P?Ws;G&kBX1=YfI?W&)v6586^Z^)~;x(uZRIp2f|srKWo(*`LjsfBUjW<v$3
zHnqHowzD<dn^OYAaQhu0L7L((JrvLVBEoTSRy`i<#hD6xUQCcHO3c=CQlD&aYY8{6
z!N-rg-dPxLi#_X|K-e(AUMHYPKxnZk2Gx6jfo`<7fEp(#`%J#jNU}zqVZT>R*fQ)k
zDk>KhQ&AVP+Sh+HQEH26$ip~miAmT3%73_)KZQ6h2d1jyG-yaK9nsh>W7DkJ`gwix
zk?6C;VJ6eQ6JNbiY1&0B8o8!zK|?gHr*L@9JM-HoCpNQiwMTdR3g9$=z8B7#CLd%h
zC><H|n|zKa!Xz4SD&Qvt444%Gfj+UYS@nNfXE&@}lNdiPGPwDPR91=0v=#mF)ps6%
zS6xM>yU(Y@@&o6Q`5pJY-THh>k4Lk*;RJ)8vBGYl0u)JMi)7Pw|0{y6Aebx7{dA1G
z6V0!GZP0|s#r+8avg8#My)yiR!4?4Pl<SQT)#0-M+9TYg$L^gh1Xx@j#cShk<>{s4
z&_xyJFS5tQ!@wZ9$(^O&)sBC^fcwFIRR($JfAK$~TXox1-S^UXultvwnn=Stv{Y8a
z3=dGi3vOIwvSz)0S^8U_+muVso8c&%+UD)r?sg=U2(m<Xf=6JL|CHE$ci_}8&@|DI
z9_79=x{Sva&oEWGI>`@Ga|4CnI(aw*v=G!{Q&?1Vq&>A*&MA;19Aw{xgD*9OG+06>
znuQdyNnQ}f43dy3?idF-N5EA=;ZG;_z|$=;EoK`q#F*mSKzcv3eYd`dh`SF<>k0@M
z@xmv~)pqEz8>29ZZ_Q~*)8^N&ddV!TKF)MkERGX4)||RJuU%xHlaO_8&bmoiRZ!4r
zX24<qu5^A^bOz)ASI-@f2tKf3(0AcS-B0Nf4x&#u9grlcH_WOSRb!(;Vmll;zG*m8
zGW~@@SgS*{)xZ)H-5oE=7dadC^$e_nz^msKcQ&ZXf6KtVa2frP$Q><`Glqs=BC>Z4
z?mxblusG&;Mz*TDIQ|g=|AWE2BuNugt~H!30+G)x74dDF-YE|YE_T`o$4CtG(aI4c
z#i*PHL^^WK7;kX}BwhiwlN_r3RGzgEC)_7U0G@NxBNscc(*sTp^DO)qET`)suz<_2
z{C&cW9YHEjW*?I_(XS)JNgPf3H5swg#hVHeomt*v9!ljgf-d%EiwnP#IMU>Ij1e~M
z8Vcqhe0Xne%$5*+xW$ikm(IV0LuLU7dp4SSt2aQo_q893^V}20v{nj}jSJsgK+BP0
zjK{{ZL*U!@An+*G=<V5D@A3FOT}dtct4Zva+d{2F=`6Z2S&{m&s-7)7k%h#}y|+JU
zo429IU{WAVHdp1!{}iT!q(!`im52;#cZfRHM)BHPQNu<*`I5#@PENAAh;$O)k|8;b
zfUngmL2V+#Sc$I%A+zhjqvhc82>bL*p`euZzA{$p$FBcbBpT4{eJ07GPGwIf6fwng
zlY1pzGIV1EP>_YHOD9_<P7S7+XC9pkkX0MM5ak(D_az^Xr6o`jDP?maG~Sh2fN6MA
zc8U$XCHvF7{=v;VO7L+y*W19tQ`g*g496QQ>RC4j<*p8P5`Q?}03vR|_eh_^P}(`K
z_DtdmyM@q-`31)+zQ$b?8j%gcSuy^(69-Cb*f4o~kS!=xFhi054Mx3EKl>3{8BatY
zu2xYbk0U_6$D*#@mB{gxK9!aX#h4`!Dv)96EB+Y8fCoa`U;V(RZVt=e=L?mVi4iE6
z>s~BfoYXqZdVfcRke!e}n|vgNIv_;E1)$LJ)dTA-@Mp+dApClIG@An>&;X@!7E@pu
zzq~dBSyG#ecMX_r2EIP6e=X{_!PUGrC>=+IL_Z2H6SnHVA*f8Cz0j;_v3?`JmWt8#
z*|Xhf@2kHjIsVmumWmIo0vIObO`HVTSoaoOa;+b?NWIfno6_CwzCcBER@c`jF8=UD
zO?DM|SVAVg4MFPt&Uvz6_+|I{q{F5NrFg(dfVf>Yx%5vo-2-cmZxUN5<o{XFKPa>y
zm|7;m8-o$mMR1NaEc@oqKc(Waj3YZWxgk2{(tH8Ah6iO<iaD|Z=%pA`bvXMb3zM~n
z?_R$xQU#Rlx-T28|7qis@#hv6+LswP-6%0>cM2eI>a%7dNX+hN_X<uq)fv)BN4%J4
zQw*^9_yAYf0kk|!$#UWWvyg;>DsQ#1!kZOmj-?OkONX(&B0H78sb9t4e11nx5si>M
z39dc+L!SPiE@BARdCb=m^t7#3+^$Y<RHV<%EOMA)N!2QtA-Y|qS&DO$2L1IyP-iai
zGEd0@URV2{VIO)VTiL2=m}#e-H)g77wJS(?0EiFL3y%LEm`E6od4Auc9^SF{S+^w$
zCbjta`8_VU!m1aWwB>POxTo;SNDXs=O?O8szGut<v1f>R;b=m~?{{zapk^c;)C7QY
zL;6p`9^7A?4l*OSek^DHbwI}2()2G~yrSc*l={;Qeu7X(6i^M#{Ii22lFn>MA8Pld
z;j-E2siXFMg}?f0u1i+T+Zx*XCWcwIGM7c&UM$=RJMY{?QA_=c;8qZ-t+5sVM9^Li
zc|!i^@B=G<AV6nMg321j>77nIL|i^hzojoI8hG>1(i!;S-qG(i+17f-s_8i5@nHqz
zcx8l%cDA;zEQ`JzmyqrAHUKE0Hc_yN*sIh2zb4eor$(gfdv-nd^TvSqS^&^W>R0~O
zeL}12bmFuG)XF;5Q|vU``s({9+sSs5bZy<DK)Pz)pLYEEy6<*TDKf}+?^LS(Fn#+t
zKwJbemr0DfJ|{GI(f-N#St5QF53X@7vy|2IAyG_)h$YIT?3|n~^*I5t2;GT7UXbO)
zrfZ9dX^8Re$PwtmBeSIMk&-ssoqyjPTcOI9m%E2UCmXXCV3Jla)K-o2;v=FwuqmXH
z%4J;S%u(rEI2v^5YW6y<GwwumIqxg{9XQ-p<`V1pr;6#XQYD~wb^ZTB2T_ZlUCcOq
ztMi1R&~rI$<N}j0w8~pFavg3<Ly;C6Z5Qnf-jM=m?|SRUmFzI<myAevBsxp?c-bL`
zdEvzzr3erUADQTmsg=E+pPu!Sk<z(j+q;}0MaRDOKla#icOn0N0orC26QtB_KA%(_
zH`DR^Y+?jXkr=wJkLi|ua18(U7=?y`8hP=+$0O9P9cbEL3?ylxtfhm@7V6|;xtY6e
z`RhB7qp4G#vHD*`t$^v1_Zzncw6aWA+yd=5i)k6p=fm`$@Jh#7!e)o-Hm}Rd%J$7j
zMEv;H)sZ4Tma<nMj5`CtY9jI!2et4Y0u!u`lw>Km(*~l7pnF`Xqd2+l>nmA#rfoB&
zN4;opV9caE)&pM$K+$!ogn5kZ7iGz6*Obt?v{0gZ0#bdu8@K@n7%a|1Ceb~~fhGo0
z7_V@t(j4vq-DfBTj9W5b62UNz1;DVvB=)o$p9h#f$<M~POfE~be(vIX!1GA+<+V7F
z+Oq~x@_*@ni%QMwUa{$OHsW^?CfVIiGGYD|X*WEaeL}dCaUa;E2;SUrEAvl@EJfp)
zflzWSrggsjL5ga+&?0E?FS=UZ7=P{+3VG-V9n_d{h$KNs?>hwW2l-HTk{@%-+;}Km
zVBXS`dUkhaQP9m(!f@dd(ZRj`V<D$OAR@6gGv1bEL6PlA8lU0flAM~Fn(J32dwq6R
z&JhWu1nsam)uX!vG_aRBuXAR)JyF{~yV&~81>!HLc?aA3P{$+WN5JU|*kaVNFw6fH
z$A+V1*FMR1T7ASNznF4PJoN~bV=ptm{kWs*#Zd`%1tgj-&<bEuWt2uM>5{9U!fiA9
zE+wsS*pB|(UZw+TZ@80}K5o^U_Nl`CXn=~&a|btDjbRGv3Mbfl;&IY7!2&txdUfKj
zloguz1Xyi3X**T_BuGAux{IBl;_KtXg5^r8X<Exq(KiF}*r|C2Gh9~b%jc(CQHGRI
z&1}nG{&2?jU^6NypgOoyfYMD4Qv6iJKtwY%dZkrWw|t*t-pRfhmlxjEX<2bBHk`vq
zfjyE*OS3AJHMS6RCFPN`T)`*7gZ0%=7<6=W2owHtfelAH=!_PQrLFqVxp2%Eb`7+0
z?CMSMAy7#j&<pnZZ^sGjA_-m**;FjA&5&wDgT@t2C#V+;G_L4gzGt8LL|botWu99I
zAl?QB4kLA&36bV2Qu!>UjhmfY85L%`>cpURfbb3c@c$5``{4+H7?6TY*FxRHX?z4^
zoi24h--xkR8;swkV+#Yd&yxC|Kdn~vh7I_IK-RCI*1IwMVK92owIcG^clAJ7({17X
zk08iVpMlePD*_?}wM*tt&6tkr0Uc0@>fr^a4dM%A?~p;(;1-bLiP<-;-U;~GT6&<z
zb*d{vzryRKL)6dX9z%xiLot@rA6WMtGD5&_L*NCK>%NYjJz)b!1vU+*gir(TY9CJb
zbSa0bFkU>515@GXI8BcIg`-2SknJWs^&;v+Ky;l|P3Tl?{^vl+6P5P2`F#y2D(xAp
z7_Mu>Z;(EwalSz+N_PI;*`&D4ckHvm51%r7Wji=<^4nC&)>XXea_O<DUEQe7aIl*S
zXWyX*8QzORn#v55X>ymjk)J2E4HN22(n}Vq=f_&6qQf7MJu^W9we14dk+p`i-2U_8
zL<ctkVP{!k*DDG4Hd^64#_+h(T@85#P9C^q4vX3E+KPX__<;+PmX@vq-T&4)<hIe_
zEFg=g@D`mE{w8&e;9CwV^KMnQO(mtnxnL8oi6@(_l|uF+%MDDq`~$l^!`LuiV1@**
zD4vIU)c@TnZ)sH0O9W8(OXE6g;VaHTH62%Dxy;(>%WGR6wm-27M4ZI|!R=hp?GIs(
zl!^o@L0)!~nVDI4d<J=RS0@p1&tV(!-Qqkof6RoO!+Rn8fKRaP`&Kk>D}$O(voY6z
z`~spBcL6jrR?6^=z7<_oxnHyr!l-()Ma!mUu_Ms>B_b17B~_!zwc;4}x#)4Gl+Z4t
z#OMW-C-xzor0&Nzjnfg1vmvWt6Q56PE9V^zU;?FN)BFEe!%mpJZ9jsYkqWBHa^0WZ
zs>7~$G%K$g#`k=CPVYfm9G>Q(@>^bP1E)Na!@Ur+p{|`Qeg1^v?Ah1gz@(1a#&X{C
zh7i8S5I5D!VtZM_MW}5KM>eqWzo$jSncyDh8;NrX5?p8v(dL&ZbkQxeHL`7x=5}N5
zQvNEG4l`+fe?|Z((#zuJGJ`Jq7{`8g5uf93)|J0X_cDXvi$GPLb5L0C1(|zn*s+n|
z4)Ps=m$ttjunBp7F8(VZh2LZu@9(Ckt2To9l7@i>k-X+K?a2IM<EL^<PUPC)C~53S
ze^-Zze8iPq%X*a#$m<t6CEh<I2-SgGE>cvOM{TzSRNaw8&9i9ECZB~wEh!BJGV005
z9+177B~2!co}B=d8UM-rUxBO%j3#YE+EB~$dV*B=%;sR1zUF%6Ih*SG^@R?L?%b*M
zJs=SdRrl%b?hX9W$^8#HE0f?^kae5|k)4+EVaaJYDCEqSd1-gLB;?~H4@>p_6r*1E
z%lIm2s618JIg_aD73=U>TZ~811c_#E)=1NkeK5VXW8@9IrFaIC1~_z#@;_B39wjeE
z$+;Z&@OUOVD?r>49I*WjGWu@WT#|k=<R?5Y?{YypnW&}bpDrr>x^dJ0^an0ijn8JB
zG8O2(k=tnjb&JAqc7vpUpAvKkmZyD*ET8=1@qF+sqkP9(@#icAgg!;8uUXxQM<kE{
z1lpto$AxIdZg<}hnkt?ZqI9o4Sy->tvp3ge+FtTy3Gv^%HI;2qx%D35=r~Mu>$aiT
zZRlyaOV}+%JI~E9S@!$VZ<xorEqr6YW;y{P%Bc;IVMb9<FnA)4;g(0Tou!isL+f%%
zotGD(2IW@GK%V}`;l2_QdxbHac4D+;-m!Cu|4C8M1TzGSH@-{U&6exlnQtZhuG!Iu
za`-!Adhy$0PoC7}D*jcNtibdqUvNnAe$TjSdno}pP`0{^o;8zPea(16IVZN`<Tn-i
z-7WiqmQX<}MVj;fi<coNP}^Q+J#^&w`zCbq>BUz*RKHh2jVSizlRLEG{NP-lUm=JK
zwIYhj9pusA1lSuQw-FK5iT><;y&xwp0bbH1iYn~R|L+%agvgAl>EFOP>+1M@Xy6z+
zsj4--ni1R@@sLAWL&n+-R6B20NAo+BMZKWVytkzEB;7mKgu>tu?#1hIYD~?vG$i6X
zVq(nHOC8A>gH^GIodRGs05+%{oFJ%lR_NEqa`O7HvVA?<m^YD1_%Gf?PgDvmd`Jgw
zEWN!V!MGUUQ4VMT3JJY<_m(!&2|43dm8_Vkb8e&7HFe?=)cf@FQexE1*utkLdxqy<
zZ4sgz_dC&JbEn(0Bp!lRV<PO-%uL4JS*0%JiZ7{<+&JIUwaGBFfset`9Mog==P}NJ
z6Sus7A>I&(+I!=`@Z1mI2_Mr-g#k0&RaVfz<#eXj0T_)V)iHvr=%b;nuj&D!KU=52
zxG+y?tQ4*{#g^p%MrTLD3P5o{Fi$bvazE>{b>r6JtS8yy4N8ULm#RaE8tXK(K1{9;
zHXk|eYi{nrO0#SBcE%Yc;Q@GZSv@%BId@U(p_E10`oa*c<jZsN4__S<*Gq&!&Z-XN
zLYWOVPW(KUy{HjTL|GdyZc5eJnHp&db*}|DDeaw#01m0}MqF`*TleewD?Q=zTDGHI
z-YzO>td!YiaSs>$Uw@(cY12`<$o+@sLck?4VL`0bmou}ON{aC&mrnT&YIkmSuuO)$
zY|=Hv0T!Kq07S0*DePuht@j#}<U^01X46wXWumxunr{KB9B-UU!5n_s*(&8J2H_X3
zrsAVX)LN=qjz1~AU0xRJ7<YeoF5=T%Lr85=<OI2);2TJ^#=O<2c+_dkacP7ouGZA$
z@5$_#?Gk&Nm?X!?53<kCQ>En1TSHz(pHVeUAUpN+FrJc#9so#sG|SxGI3G!@odyxG
zU#-yUxCO0{JpQB<%B?y&nP(gHACN(neX`*mB-jCRAjw6g;eIFVh*j6&TchVTcbn-k
zzudXG_M{%W=!uZ+R6VP`UF|NwwL>}tF0RDM@e~p5iJTUlplI8AaYFt^2E&h<xu(8p
zr7OQBtyOZZN$H3uO}Jztxh5p>B~$K=M%gv1vsch8-sO};?L%7ap*F`ZoW=tscXeV0
zR#Ojs(a{Xu?!x(})VjKk=;XV0Q>H0t=6$-)SBdWDmq5GUQQDmgBXWkRej)IE3E9Mw
zz_i-Um9cbT0C6`Rv7hHzq0eAMQdksEOM=FO)}~P{%BiLlRSxI5k@Uv=k%*Ej$zSX!
zcF&W?T?c+r6;FYCzs0$gf|_ZO^jR}E?~Cz>Qg<Bh3mxn3h~Ycl#s8$-=2>V5DdVH<
zfM_>-2)>7W2tt$?0}!I*aJ2krDzNCxzY@V`_ghq3FmMBZTncX5{nG{jkiy+<1p-vy
zGYk!r?ciyhuLcvwyPWzO2$I9LaLJJUTaFWfQE<IjC|fJ-A1{lPcU&nQe*K0&GrqQH
zA&?4>Kg_JF%@Z8!`aKU$ImcvK_LV0@M>{~E-2P3$e_TTbcIJ-gfzNRCQQev@o~_OD
zb2-`s&|FyyGc(Flq(dCYCxo32J#N3CMt~=akzbA%*^Qlw233<-NrOq~eEs&A86{wq
z9VSZ=`{&>#R#UxvELtF)XCgh3OeZWW&e8f6Y0zULN)3n4flB&`Z{79fR~JEioUIS)
zXbqr{m}g#jx`#9=1o1)5mgud@(icp3=R5rTZsmY>a9|MAl_f$XIG3<kGUY~Lt>#=L
zsZ(Tc#sS8N%_%}6U?++37B5mjel(K6HpG`jXys>+5a<0Lak$|OH}~+2f>UTq?v{ph
z$9*K3xtNilbk?@(g7@aekg_*zi?E7-q0K9sro~Ar(bc2x%RDs4lv7lWIRc!0V=y@}
zMxzMy`%RTOtwjK=*3R0?OmOud%`w48ku&sd0o1}9tk0fGTW5hj_%BNty^faF>@U=t
zaX?Y+sUj6<-|tpWNnxM&idB?f%Jvl|^l-Q>`MlP@^<Lz~fgkrChzp)^f22_r()jHu
z`4$)Dl(EamltVi7_Ompp6p??vdvl0BZ=-UN?vzgCish&xMeexH@hM6bay!4*-TJC&
z4#EHdCJhB2!Y~Heq;AKn7D@Z1?)`0#+m4O!oL!?lDRdO=_B+3k{(*{{c#wVk;OFyG
z5OO_)$V91d>obAEc;~<dSSRv&+-Dv4#PsU}2zLt;<=ndkN9fVR`mkkVi%y*?-3++o
z-9q0<y*?Nbz>oj?)2oL^EPNqB(*)J~kV<azP9A$h?YZ%`+0tlA2Y*B=PI;}eUC!AK
z%t3fWoCa`sP0zTFL-ZI>hHIH{#{$9hc;2$op}7cB$TH$dNzNQ@opZ=q0xy>rVx7P%
zi|}^{nM{-?N@<;Tmuce_G$Bo)+Li!>3}neiKQZoOA#$9+>&Rol;S#wO58ztVJic|-
zFRHvxH%)cREk2P08t$5mtd`JyRUHJD&^H_Q^8y}x6G`N;R8iIi+;DxEbCdP)*adQ^
zg+w>K-ShDGctrjY4Q~B%oLHLWp>cr0_oehK*E)2+tRjE-de?bgg(4xkL}mGXLl4Nq
z1sX>e2A<V>Z*rvMFTCy`*vuW6|IT=Hq=2{mP0P#t!1Sa1xXbyIJha2RBW;;VWFx+p
z3sourWx9hDLLEa$V+5RL^1yk8?)2{!XIaEx>A;MC%<Fa@bFopfj*NdK8@lB0l4*8=
z7kv0)YlFd8LC&PerC+~?A^~p_2-DYhsmMwXsa+qH_U7y>lO!)u`CV!t6<=y$-cH=1
zEFw~|`ROqJek1}j;9I3$=yc6D017HbxY^aagFdd;%d^9q(sno5C8%%{WTIF6uD=Sd
ztnX2o)%T&yHTv|JguTbU9H~U$#eKOk{;HMuPRMo{YIpU%X15D4L8+TDzL)pJe0s48
z-(aW>L6wKCw*K!2@$yq?R7C~Joc52w@%cKxV<!-C-(iD6scAF8kG3jx%H)<MDSQRk
z5C!3hutSHBsx{Tu&#Xqh?3v62%_Vc|=vDD_dv4-eanVp800kdNIU738#4Klk4A=eq
zd*AMe<?APRp{-y>KsOuCt|$BG1}B6KdP_>`vQpYich=AL6s1|tR<MdHcU(2?`4V8|
z6$wrv=3Zs_>U2=FK)QbfM}qQks>I0Grx(h8g0Hl^^ABmhc8W#obeBz?V63dk#*9oF
z3Y4teGxSB)=%wlM9(=!R51P|%0-&Q)Ib=*;c2A<7od;yQI&|&S-YsxA>?Cl-x&`J9
zF6sH{OG^%hWWekd^t(5?78a@Y1jzSHnt6;{e4WnI&x<ooI5|eQBOkj?LCy%7k?%Ij
z3tB)7;#X@lJV@zf)><sflEj*SE67gZsHADNL6r&NbnV06&N=}ALHF3A0XHpU3U{Ve
z5D)S>4egF@MnETB8ce^1>?s+tlfbl7jqm>Vhu$uwf+c-Fd1XV6GdK25qtF}EMKtpx
zKMii1(~I}*-xGwBq~Y8QI6EH?ke8u%LxHDOmwpAc2(O_l$ZfNu9brW$pc7A<Q@@*1
z9PId1D;L4Z`;kUP*>$_LCn`)k8jdgr>#3_zV%({!LUP7Ike&PTQH|@Jv3cES4abF*
z%GYnW^GcA;p)d126irMwfdn~naNWW}w&Ud=QJ`>m1Jy$ib|B_^H)h5GZS`vbLQ%2d
zH8@gOTg5A4x(wysqTocn`3EEn59Glu^6B6^N9!U|YUVO|+1YCh<K0vq4P4wbZ44aM
zM%rAL<#*y@y+xeP@oquk)cyMZgBI4OUlZ3xND%6=+FoAOKowZqnh`Jg$jBklPX1DR
zV$YZSr4W$1X;F|52<&Wjhvab&VVp}$MN3C3&VH%~ZcM1*#}8uas9%o(wll8-j&rv0
z(#bh{<3P}aER#-g{`HdhqDV_>E<RB2+@J-h%x;hc1)ofoJvjk8$RK8}q(X#RYqw)d
zmgeR!3Ppl+Z9CmjQj^;)FI4G2udi4<cIL=lXW|uTTo+ybXl{4)D(2fo`<+YN0?Y+O
zrUvvNGJV}{_-y=WT7Lf%zX~_jG16y((!RAU#qNdDQ9RSCzA~9lzD2uC7KD1cd`L;*
zxg+Ld9h?8=cK<H9x@mf;gDv{d$M%!K_yRH2k2!bra=k+-`<slRXGO2ev_6z_IgJ}P
z4kawf7EZKf4f|{7I2;ISAqfl?Mb1R7gD*v`e`9J^`4$14;l~Cc8ejY~x8IDqFmPtz
z>ls<$b=x{u<&s%r3JQuU(7I_eydSF%hMNAet^~c?XG7ufklhWc)Vbuke5B>1hKHR?
zAN9-PnCk0~Kt<)(1_w7c>tjEw7~nmBkByBPbLLuMwus2V0b)mMb`GOqXqBy4^;aOj
z`JiUdin$mPoQ{a^(+T4w;4fbiQi)qD1Rd(SXDPy-djg%l2hMrF9DJ*Fceh_;KnW#G
zWy3f9eJ>_N#O1>JUP<M7OXlgIvm*V*Hq7itL+jF$%!JGWx=Tt=9~&)7P7R(Z5Y?~D
zTvA3`mGJ!In9QwkL_WR3Ibru=?M{Osv&qhaRUlbQtzm45_{Gw_Ksg0hkVmLaujP{R
z3j6?NrV((gVAJ}zr0QOaW-X+vZC;G~+Fg<D1TP#YG-jMft97d{y<h7K_<2&6<&s^x
z+0E%UL>9&Y?H?BB>wdYM_{TZsuL&rCXgD{l_H1snQo=IT-B#aSKIQjQSykK4C|`I;
zvAJE0?lRsBUDoIHvbip+k?D^b09(_5lPMed{8Sq`!|pY%l6ZB;u&v^*DKxq>KHYDn
zMzRsh*Q#eY<B7Cg4W~n<q*%2Uze>s*B=V<NR}6t<d{WlH;MXy$EB`!~l6e`F`Fb$m
zoyu5>k~ZYX&qFT#vBZ30>^LEldgg$3UYC6bf0nM1t9MtSo!`wMs)=DKfe^gyC1D|t
zLOC{4TxtIS;BOdjkdwyI|77)v?0VN+dGsRF<zvjW++ET4xQdM`WP&U_EJ4QQm%G{B
zg7b_{Zo}h>5%0GxgiJ{T0M+%cSGF80CFv6Ch!ER!??WOM`}fMFuR+d#0#Zu{cZ+W=
z#Zxjlpg6y3=d|eI@p|O#a~WrA00!!ycAGOj5nCVEU`p9LUx|f`z=Pbuo!<{XA1UQ5
zLH~$Kl<*;|&&?D}njg*Ogu)`8GAAEO`_`d+_@7Y`HY}R#)@SVyAoK~~-xfY~1M`S3
z$6o<rX=)$J13L@Cr)$aE+uQlZp~Ujltl*u*yKZi7bH1{!{m=V`r7pWu-4oNH-DP6V
z2~e|CnZaZ)+x7#V1xrnhcQX{yC>~8w1Rq#&es4Q4blq7}&=O%{I@9;+JL~O-IQy*`
z#-CclEz9xg*X}`};kEc^R*+}}`~=OstiN4}G=@a=zQ4F6+?{QtsPo+{s}@pUtJaeB
zk6aI*+0|cPyHn!>J{GfOEs`Ms&rXp}e}~OrUC%BP3u!o^$8o*!drEWB_1Qj863riX
zEeRxttX)b61`T&IARxeIqp@6dcWc`h$Tb450IX&okPq;d!6s5%hpNM=wR4P{&<mn|
z-`}8s|6Ihg@LI|eXfDl)0G+(TSuU%!cdk0UXL+uCY3GF8k88l9@#E1QFqixUAYNOa
za?eHkSx8+fc5ZTupElu|JpHtLYSGM}m2UjpoP;yu3*P;P8QNuBCQLtZ&;v9o>D}e8
zQL_MGTu?d9VyxkNra>jS-rNrJ)!*QQUzb}AoQqa~0A{R2gRiJfjHV)n_IAey9qF#D
z$=+vTJy8^pjh%90B`aNwKI7fOTADH!a&jySmpy+mA<3gI)k?I6fd9zJ{s&R%Nr(@x
zO9m6!7ZEe;EtkO;B`83a5*;blQdC=D1bt{ngCKt>N4Bi7G~>>VV-T7o3H~dB-@ein
zr~~wi)6mCz(uJN|B)mx?&b7x5aJT1O(~>yFA~0)ZY@HA7N^Z4(`nn4{$4TOLj!!D+
zMDGN^Tb>5(8DIH4{YB}_!!d$X@-CH?jb|-cYIn&vxy>FkgQHXCWl+5PXT0_`90diF
z;PX6XQOaE(tX0!>D4}#kctnlyrxS5L)LC0(J6bRB;;iN`m!8#b>AteD+lQW$2fyQ{
zqYaY1xl6@U`7X!yOEQR$0Ri_MfjeIj+HY7jJ4;`p;t_YwmwVl<t&Xy$!T&f%kEq3g
zaLhM10}9U@-gU_D1SqPzeN6=CL%vSAht(jh%jlv9jW8DB%oz{811R|yW&w!V8q_wf
zmsh^)%JD8jhbBMCKE>I)MVyp_a7<Dt8<?;Ede8;f$!H6^Z^c6TPedLab>f7=Z*gM0
zP3jif^F+LujUm&>EK9~DwB0a-M+Opt0tp7$4CJns_XA-YXCWxTl!d%`7IO9i8W_5m
z9Du@{>+}df;w2J+!qrHaL0~vB^(7ifSl_W@lLq&Ut~XuZ;)pF&d6Q`{+D#llkk|~c
zI<J@Mm)u7B`Tf#e4W0FyFR+RRsHZz=x>W!Y>2y8mDD7(M0pZNVhigT~<GWu>+>PF%
zVe$4~XfHFF9z|<`!`Q6y;<Rkx0Eu6-%O`3PMkb3VgHPMo$qDP1dr7_%Fum<#Vq~-&
zX}M{qc$?Ie|AW#-r+Lvu!(9r#gA!BDoPd4hB*bNa33RBYGL~!}a1u2gT}wj=&8-E%
zD72jXa1FL&lXeA$_aq*nqNG0Zw;`wxT3`NAnyCvQfDx$A$awX&=w0>RZ}hQiKPrfX
z1>jYjxp}$z6^}%r<9{BxAiv)=1jzT$w?cBxC}X+_w<WNM7QOPN{-Yu_Q|DspsJG9s
zfHL3P;JlCti{Fqw*c~M@qJHvk>hJ-y0l005(?@4BRN_or798*2h|U<jF=dZS`28)y
zso^P_542Q7X7K|gk^0eU`gBq7otdUMhCKr-e+OBlioLr$aPa_PVq(-<z5JP?R4@l!
z{T`$UT;wxV^>V8Ij~2RG=@5>++3uU`i@II+f}i7@xJXQe*G$VmnY<-*EF?haPN+m<
zqAq0D3A~nJOdy`489fj;4e~zn8ql@m$IGDXTSN?fE~kf7=g^Ufe%rL;L{tJiD`+3#
zwdYzVQ`PpC$Pavd>7mE_d-1=9BGi%uN1AcHtIFKW`UZ`NPRaIU(CW@ly>FkHoanl}
zk#m}=R!7MlQIxe1_|J0Nns_hj5K-|eNCPyN6EBpuhoeA^&Hcn=(|l_iZ)GMcvAxBn
z^ar(S_g521Y~$VT-$37*zRgp6HT>OT$k#MP^)v7BQ)xSa5+uem%*0eLIlJdg`+CZ*
z#G~6!E`|twNSHWQW#iq{DZKIAVs&nm{gNuf{senlr4-<%C*(|lK-<zs4kxsM51Oib
z05W@A<;G>43xWYuy%Xvh;adngV!b8`zC!qhy_V;fg^g73YJEi(p-F-D=4GsHi}?bC
zYy)=t%qK`G%WL{{h)UE}njP+uHEUU`j;5=;uTx&0xSU)3(4{k$O3+z=g#6b(QYudr
zHBR?z%wi{cW3kMGG$<FhoTa;rU~#H2hjP%g8EXIC$YR!X&A#WPBfK#kq}$NinlE5t
z^L_`k0%W0t4wLQATtJszJi?K!dHvhx_`3Cm>hcxc+DmH<&sr4KZ-&x$rMx0OHd(@J
z9`zZN!{?hl0Y*ehOKe|dLv&G|o#qTA8Y2XtY?&t4M2`{OB!5J{j73TB141Fy@b1ix
zDuBN>+H9;X58unX4i4NBPj(zHTxd7Zv=@k}oB|(grNp3}vuZ^VyVe44oT0aGGLU9p
z=^X)A1W=%O1$$Lq?%G^Jd3$WP*3!i55TS4HQt^fK5GG2hEh?WVhTW+s&j0DE;Bvf3
zg~7si+Okmd>L-!oyl>Ttj@Dn9X;Rf)@0bxRvAx4s)dkL3EqIuK7`LcHnywwH^EiO6
z57v9H!x+AQs+;hy_P)AfiIT}cbQID+XSAmoIQ=-~A{i^HDxosVO;P<O!lgnH*?LBe
z>}dhB@{OI1XRCfRHgd3oe15-zV+KTN;*_&YMt%vPSj@>gH&@evGg-E|flWxxA~0`8
zf9`7(BHyRO;df)QM|43spS;}SWY-&T<Z5`+_H{Ddu2IDK7wm6_k(la9W66T^=!oC&
zDczNB|MD855~Z`jx&eWzr53*)wcZ3=)jmye@gF2Z^qRrwqb$oU0Hm~#`|EhIaxYX5
z&MroN1{oiV2@KTbHab(%OCxeJ`IPG*gS*u2GPSOhc-G{_B_#S@(o&sTjne1cisiT#
zN*K>!vY%n!{x6YYK%?Bb7{*CPp^a1%g2nDxt^bd%_l~FffB(l{i83<EDp3>-L@9)D
zN~L6HCM!us_MV+el8lV3>`fUVdn6LF$=-W!$M(CP=O{kkcfbC6-P~S|=XpFIkLz(=
z_v?PYU%Ds5(Cuk=JEK~LpL(jP4A50`&hXFx#DI<_@^!k+eBtaWF2ISA0MP$H&I!8)
z;q!QXg_UcNj|(dnju<Rix2Rf?A}UXhS_S#UHZ{f#XqO=Ela4JVwW8#d6we0_9ze5R
zC_5&kg%+Lrck^`v?_#~<&K|}c*b3ks7mMjbE`_D3W6rB)p{xuK&c%Ly@p!PKQ`ix+
z{QC9l?K`)>e%YG`A}`vM*(>2z*lAIZjgjjw4-7qB<5$y5Q+fye2IB<YS&L4N3m1^m
zzP?s?JD;0PvqWeyg1jJOduv@d;6skSPH5m-PFGEk^WUf`CtWN|!}=zHZO_I_WJ@(v
z)g86)G$>fsSEQz?wq@tHG>ga`(0P`D+@s!WOB1{T|J-;S5J0Mvpqp#9%CY~WwTK3Y
zP7YpxqK}ZijS{R^o)YF|Z7VLbPCUGYK)`X6-Q9cwpZ}Z!{4Z>9sauXHXz$sCu;ZRj
znpF8X-zzdx)vRc8iAv`&s4z&7zW@;B3PUIOQLUNE`*5{(;Fo;o2$PxRp+1OwAfrCg
zkPenvLSAigGXBE_OBSEa0Uhv9Zvl8CU47J0E*snkpMr3dwxi5}mx8i=5y&NiCJ<U2
zD1+4G8~MlMqZEL(`OQksLi*+3HKVwLMhKGu%;*2Y4-|~E2hl)k$K<WIl}HuxWv0$o
zuHU0gZo3mJizl*f_-@VIQ<c|2%X+pjQ&wx-1?5N3xj%s`yAOHN@H6nF+?u?&jXau-
z0_$G?cXZt2KRmL2DY^v%34ar@=u<3Atg*}-;$;_y-e!Y=V%ygE56_d^x<;0IMT(NA
zCQN1Hxc+RqXT;fUU1vOfb{CS6`8wmZ#nGiYx0sZz{%N#6Qbshl{AOWFG8n3Zc%mYm
zqE_H|{9i{mToq5}qsb3{;EcB=x8@&!Zr6FFlbEvbz=>wU5n(xWTC}LHYo~qn_`O(&
zDKNUfVx%D2;kt+EbP!6dg^N{se|gc2Ub5)a)PnUFEHSkkf3Atga}(z;FcawVVtOco
zwZnuQrak_V8!^^#pqWHoU*iR`vP(Pl$AR#P<h<!u&{jR^rS;;MuUdD31Jo|GI<eth
z?2kc=6Sm+QnCBY*w`M~<A;{JK%C&nA03myTo(+%E?0Y*pAY71OJcK~>O*G!#r4tik
z_1&91KeScA6H3s(oB$^@-LtY_Ff0ULB{KuK10=vFY!@YVzOha>bTKhAlRF3+ZbdE)
zqC=@0;(qz|tMmJx)P_1($G#nZe9490^v>L9lxJbCPc`>neGa3S%|3x_9mlu3*gc9-
z5Gi3ctCH@W-PDuRT=h(56Hm`B;5W-sA9x)vDw0tBt)%nei4(ar9fwT4Q7;H81zxdS
z6->ZxTJ&2&h-<H8aCDmRMBtfYR+02IGLbB(<XcJbe>G}=s0p1HSFP+ILh3-_9vC%X
zDLTKd4duL1$>#6sRch(KtM?!JE5N5l9p#8N3L7~gG9@xftG!tmUu!tX0UN6mqeEqA
zGaB<$gZedhX)s>3#OE{`z}i>CS{e84{u1uD(aDFOu7|9uUVb8Pdjb0BkkCrWYo}6?
z&jZ{cifCG8-roD^Fb;Zeh9kqXpE^bGpF3c>c=_i_DV>5%+<t9!PLJD=)ZY6vG_Ln}
zbEg<hpqRZU!y0?nd-u<MIWMi&T4;HHtd4DlO=x~9!TI2{vXX&iOWvAUS7!#3WnFA+
ztj)hF1blvWW9VLVJ8aCAr%W1u|HvHxed9R9OyjDKwoYtkYWC)gipl_Q`%m$2&!`%p
zQS)+knC9rSqkOv+3J3;M_FPu@dZ(Z{l3>&R(wU1+cKeSPKu3{K!ffehDZg%&bHAD0
zJIULzSY>*Ggv;#hS4-e6G|uz#<L)u~lL?OcqLfA-93oNUoRk}&=YEj(=$84!=PXBF
zQ5K!pCS#{GleNV$^`GL`^JIPk{r*@843ZG;a_6|R+Yj<0L0UwYs^9K@x)(iIwu;%Q
zVxF3`x5QG17EPy(eh((8sJP%6NPjV$J26E+Z+_Nmsb$Es%21%Pjg{D1;Lh39@A_AV
z{kmNxxsj9_SFOHEo=`nocLjccnj(0(peKCL^=EJDcd|T3XsA7#x#9IjTtB!J8mHS_
z>>mshpV8c&i7)=7oL>0FG=}{AxjY0eFCr<*D(aRIncID*^9ZZ4`^m$C-5=>Or|QY;
z7Ehl2y47Np%*GzFCO!9&R7KDCjb_iOIwVy$2C<&N*adfctda8WZ^C|*mO-cXT1<du
zNm#vJ@^Wh%Xo@3|7@TyB{X#m6GcWVvN9b9SUslmLNK6vueJlhLJEs=(nHzs%FSLUd
zpoH#}EZBnq$cZvuD;_n&l$4M=+3RaI%ah-9Dtw}4@q=4z9X}|Tr}O~C)&<vpoQ{2X
zc`U}@^Co$9&`ZQJ$@AJ3GPjs*j!v`55SIt((gTP(xVwt8=%cf&W!~%bS|K}IfM@5s
zrtciw{bHtwVuS3Avg_`T^q8c(14q7%kT`929dllyt-Y{0r9Ywf-uY6;k<ktZ{S48R
z*3XNy<j1&5Co>7gcFWwIEgX)A#1RQiTB|&tawQ>X>eoYSZ!QL~aY{yt>BGDa9hRDl
zBvGu~QV~*NLiNfzAxnzfQaABjX$f0i<d&X2e|;x>pr`>NPr>`b^f#Cg4&G(FmNB~A
zTBRbq`CYjzvnaP<F}*BhBw@|G*|zuj@77n)O>jB^mlcX1zjOcH@@o}XqcuyZ;w3}O
zMh318z)idIQ&ea^?E7odB%2d=r8F9%?`|KN7FoVy8){Q)1Q&hW1*Ozt!}UJri*S_)
zfD`M=qi%J|@!fl?PHi7=UI!b?p7uJLhoEGC%$klGoBLW@wR7{f<vsH@vlPGFthVBc
zAd)M*Wt_(3Cj$l2hh2B=4|nq6pLwYBod0PKbT5J9@#bB%@hO84DZwSVO>o89PN@0X
zk~~3BiE;ZrJ>_q)d8guJ1k^M2NVCg-LtJ|okzwRR*t9wvk9RN|B;WDF8#+kM`I$9j
zHi^$!TevX!;${VlOphFm_xEVbvrY<?rMd_a+8_UDfm6f>U|Ew<kW$KjD*9we1?*ex
zY~HpDNlP&W$>90wz?e^@d8d6M{^nNPN7KrUZ!ZrDtGX7-Hi;T`2lz(vy=UFo6pZrl
zU~zi6%)3s<@T`4(GF?eYyeCa6RLgX>w}MKZkrz;rN=jOCxl*>K*T%-Di865*IJR;d
z+<)k`u)9V7><2$eLX4Jzb5|&5%0mFx#@>sb7(2FgJfwQOK~;XW+r)Z*3Cmb=Bv)HY
zON%IYy%b*DaC-L5O8d^z-}Tv%OV?NKhBoQDXzt_pX7cB|`}2*|RfK1zon!WQ2Bt(C
zTCC|!_s|JQ181OC3sCVpuBg>-dSbGEXRu#$G~@Y%YTVcr+()^>@Q-Y#g0!?tCN?Y=
zB8Vu@Y$$@*j!D;MNHDBL&2GvIWFzz+tXxbtPp<wHRns#xbbLdAUM7AOru*A8llI=u
z$@0<#Mvb<uE%h#Y03c-m?Kfk8v30KjTZJG?NR>Zd`jf%7ZBi)8tnqd0x03SV?j!$5
z+3J??ayrx_jCu<1LA%*PtEDm4?iDmjxrM+@JuW8i?WDFa7zMZyJGV-C{!;4`NO%f7
z-LPL{(^EO?IR!prR(~hNXz;@6>><4-u_wOW(s-7TG4e^&&8P5E)NKE!^0z0u^0_(v
zbkb{b<Ecm|vMzniMoQzMf5tP_Qxr7WPB{SZsy0}C;4c_YOmZ|T9~k^K3Y7&Wlw1FI
zMMXutWgx=;VMF*o0{Z^w_q%^SS`otQVok`e-5;6f0OZiL{Q*M$S{@}yRopc>1KqC9
zArZQ8V!ZEkytf7O$_$H~21JF|WQ?@qOYD#CwsUq0=+w(-xcx@}cX{VxX<3sdOw<WU
z#!K1@4;CYY8Rq0K5Fl{KclYB1ye;wC9?f)iAZ(?5-PNIGRI0#sa^O!Q1o(#!)Camj
z7m)E5JLf*$8|f{NOzV(r&sHZmx1`uJu9eW$o=#*|577(?exP-bIPc{q57M2dB6{WG
zI*pEgS{RKDA0d>0x_t~*z#a>r(f75h=>d`?E+Qf#dtdyfTA$jJE4mHQRV6(Zu6NFr
z-~IF;Qx63W7n&YKTS{wh4xi@92gAU;Ezg7Yd)kR87yN6TQxyIhyD3~74Iw$&_S(vg
zG0q}^sMM|gJh2{|XR2Q(s`V|ZH+f4xbjJa63eH2*O~wtU2Z<HM2XLu_Jb=+jr@lrY
z$2B|_FWH8BNUE$C!LAp<3s%kd?PL9bK~SudDSdfdahO~DTitaHXr(&AU@kqH>l%=s
zJ00$h?x*A`p!;*;@d?dk5g58ech@)q2(x{fdRgVP{z7x#C)_3+EmZo1X^KK)#+$wr
zpt5FNvqjc_7*?ob({!VrnwI|!B+Y+$nu<<?=I%=8s3HMzSZgp2Yvh>b4$9T&5txkL
z4xu9ri`VR#^RS}wgc{HlVZbLGCKSeO(%-8#xVuBT7!`wLDNj~$4=?`+2gM2gCgGwl
z=EJpyte-Yof=T>^F?tCITe{nj_zK{?bdR=Jj!uVd3vDQNR!3afy=eaKl;_3!4$>R9
zfDc5`7d%z%TsIg@Q)D$%>F4q!_a{EB70h*VV4aK%>phg@{{Xr->MrX^!AZQ4?|0Yk
zu@vJIWszlKg}*Kbu*YTOx+|W}RW`UdTIg$52&<`NPglgqoc6aTyaGKCns(@3S7C!)
zAao|4jzsxZ2bvR6{499R+hNhq(rep7bVUOGgtXnhmi+eG>;0DrTUH(^U9{8er~GX4
zBAD06qlB!#=(7o34!uyP(KOAZ)BCD^t8=zMvv!jNu7A&zaqC`|pE)9JCVN4&y97X4
zoiAA)nh-n9k(0O5WR-HEbG5Fj-s?)sX^aQS6^y5P?+IIn)B!>LJsTN!dPlj?vi?gS
zy*<bUCQ~OwL85V<W$ZV5wQ~9sh9Y3%?$`f-k~|0`Hy7b$kS4E~*InH1t3J1VwA;5K
zIXIQiA?MsK?+%p*5zmO$hjI68jL}+eE;TSIb}QQV&csrm$X#!1V9Ie~4cIzAWlT_z
zM;&>+AS<JdBcJHeCd|Mo%B^NRyBnSUgoZ>`3YuSr4B(_YlEnX%_+SU*|3DqRBH*fM
zGF`G+y5>rGR!z2yiBh1+i%BF>%Y$!N$5kRV*VP7JM%5_l4+6mjwvcwiv2xdUH+VN|
z-2+zA;CYR?Sfd*DYYr3lV@{I@t*m`U@vTiikNq8=oILdcnj4PCP6XJl)baSF@BS`s
z-=Gso=z$DMNyazmmFOA#bstj5URF_2af!JwlQE2FC{ALEw%3dHHwHPeEC8guXmkMb
zC0myDCPfZ{N7MZy4L9p?pNb+wxQu<#Y=Jwm6$NHG`eHHPS_B_t&mX5G=By1;zh<%U
z+rH^@r7Fo47lLUMC%WUu<4Vh}N`dB~>`}d-*A>Z+Sg`|;jgzey`j6n8<YmuN@UXNZ
z35wqMkrdMss0rPHcRlCBN*-L*S28_i0mO$wcK@;&=|xbNMK;2GIROFsWvsF6&U*-<
z0u76ganPUjgovF8d%n(k!Uw27c9bM3k`D3ajuXStPgqm*NG3b7nJj<KhA$%7c_z&)
z-=WOVeFJ-1o>0+b64<wQ2KNIsW7G)I!Rf8X8Rml>S=h;GA2LbR{`2>&!BcOh5L)D9
zV#@8R-qZ0iQL(S9MbxIeYZ}`U!aBx2^8TG^bmTO=;774HPVK%VODV={sbdb9bmtj+
zGd;~}YK*8;&Xa-Zr8tYP_y?BT=}AYMmwC1{P1f%r2yEErO8Nh(ZGL(b!?ybEew?>v
zdhpWyT<|tL$8Nf<CLYk3cOKgnS)q#Pnq=9mhE4{h`w79d?U<q?)PHs_EE7TouLvXw
z-2D+!b+&d-eWSgnzA^po#zdKw-&S9yMacN=sCSSGXZ}!Vt*e_|_-7u#O$~SLWydSP
zjt(03di@QY-q!q#XEM%{=EPnq2g}t-D?~3C_M?+6WC3J7Hg-ybTe30h%_n!VO(-p|
zkwD9Bq)l;mNY@$sP-TUq7)XNM;Bs&^rK-P$hg@IF^e0MwA)PkhEcH<mkIM$-W?+vf
zDi0`d?Gh|F2}Se+#)k`vnHd=mL&r}gdTlhs+>1*6Z~eIZQC5EuEH=`SZ)3)x-%~gW
zXTaCWXxnYdb2X^gkXWfIGx&MX&c0|v&<CDj8S1~UkdH<wHDjxU-alp0lE^pG;rIY;
zHli{k>1Ynw6KQt-e{>M>I~W%=0R#x{Y9`W`QZQuD2+XG(8ocy_Odo>PZ#YO-Ee(Q9
zrxPDa@eTQ|f}^2hY|>f_=kDU<BW9Q_fr@{%C#W+3k_Jvoaa;%ua(zTquZ?Gi>7rVi
zRx?Q7o_%0@ZgNYzDu&1Jf6)bIk?_B{B-fX_hBRPm`REEbD5Ws-em8%@1IVBe_?q#A
zR487YSX*1eW|L$X`Fqrp3VX>L_;6<y9NN99qOn8W);Vo2lo8x1V8!A@1B>I$6T-ko
zVZOn5?V|pA7qRP4*(2C>0b1e18OvHJeIzp9@qDhYW;@L48u*%ETAd7I?j7(T(v&za
zd;6Th^nZyr(vPsOO9pZcB~g!aG<QMkfru0I2twp>xr7zFr}s`g>}jHqqq@7%;HhVE
zV~T~a_Pq$f_mAe;>i`-0x&egE8dqfZbm2jAX1DbiOBS1g<V-)z2iJ7<BEn{nw>{`r
zrO&(_D>;SC>RbcZVx&>^>doZb!lmCT!dU<TK5WiUdW864NVOF>DDr#kK2572l$%yN
zUl5l)a#LID=`Q93oza*rC24YzD(}kc`O$g}1I-bC<i7r!WA6Ni-f8K?>ePx8nfa-f
zz-bNBefso{WhBv05R7iQr7l=w@T0xQsa|M$BAJg)0!QSHij!u)F0qkc?R}mDOXb$J
zlGC-S{CUFu{)InRSR0}~7BEXV^5;fGe}pWRNRXEO0z+W>7r&0{=*@MbH@Pk4LiYlw
zPrO-Ue@@m;ynD0nBsWbDG=6$9d4}50SGehgh&Ggm2|GDrVAO%Rc*BCYts^okDF;}I
zUuu}eo51T<h+#T`#z#nJeh)UPuy1?z&cv^Zzt=S*1}~Ub^k6|Y7v|7^kMM)mvwqyX
zDFe{)zp-pa6Fj5oen-vzBlk3J_^UgvIyp;-7+CdG<ecwt?&@%AsTdB_vxG#!$hy|9
z+e{?(t?Z%Q6ZHz|@TU^J$FlyhOAm$IwKR7A^rLx~tnBP4{pHbE;u46Z@_Cb%lH7_d
z@r2r~1P-Yq#Q@;W4`;XpGKq0tD31o#eE%?_zxZ0x%3Xh4=LDoEg)n61=rPLSMGkyW
z4ZBj7aICoJT@6Y44oYoElplWKR=mZw>n{nt1VB%AdI`94Lr$#Mm8FbRE|GKd@%?a7
z5vKO^+TrZ}uN!MEMUGfL$=|5-C)_*+*3;KAggAaCunj2C(3+cXNTtlNIKuGpgN6dH
zcBO9>e=ECB-;8Zb-dj-yi>6X>nz%h=1EM3NAHp9-?-bk_OtbBiKAvyC(qIyj*p^tE
zn=$98-2^Ab$Mb0Km7UPE`)b&Cz|u1}H(o_9QJ;{PF3*2Fa$4rRUcB2lUF2FkQ>j%_
z3W>UC-vmX`)J)x1p+q;j8+mKO^qkEtE;BvZ-;N(3bTDVTH56Q6(Pz9OV1LqiprfY9
zg#9C1m}mF9b&+E+6x+9ZhIouR*~vCtOu9F6IW{(jTbO3u7nnSSOQ}U3uGW4X7;@ZN
z9C1$(`P^~RZooiSCGT09fNPs&WsfRb9j0qry~(a9{rkP*u;e3z+vLlu)9*^It2P%o
zrZX9%8ehk6e9!(Y+_z@Z26@ruBb7@vEce2d7TZWeJ{}IUsmJN(rAeTdEt!<pZ4o4i
zIH5G%F6K_@noc&kBsb%T$<uK^dg;;q)|h-9@u$gBJ)m`ZiKs}(55DR!Bd!0|uMQ^a
zhs{Q6x>8<Na~(U;Ek3W<7pgazd@7J#O{V#Xadpy%43nJ%?j~{-j*h*@;R!1~D2D+7
zTWYletF>jTE0yX-Dp^Wfw(Djc4-B0QbdqSZgo{2`*YsD8SkmOBYjv#YPIP_^nD5D>
zGg=Q|C^9ta5%>xNVe!SB-J6e{E04<W&K;^NODURiIKf9FvsMG3<pV3rl?&y7Dw_?d
zfVVcvlHluRyKfh<Jg#QjG7o-^<AeVUdhXO%_UoyVah?VEgQyddbaJ{ARWFUDn%qHB
zxv*I)vx^Jy5ezA>HRLff+gme5Y#l?JQf6jmi+vzs>CYF4FRB92IQL`gf4ChBP6nUe
z-2Kj8f#YD}qXmVsTU7f}VlE;z3XQd?@^?OF!crn^)zA%`GM9lhpdsG|u(0CiO;+p#
zn0BaJ^?k~5lfS&4p-+R}IT4!q%pJEB!gB?tv$cv4eW)@z*|ffFeyEB(dv!S2#<`6Q
znwSpOTBWZaiszrwKSR+i-rr-v-N?J+k_AHAA}SVLo<ZDMXs3uzT6aua4;6Kabf*B}
z^w0OjFIejRP34kh1(@nm)_*@mrF?$&^$@w10{PL4y-WW>_nXv;f7ee1yfjYHX#_vE
z*-Xn<!~D@o?6-mU+dNYx;4?xO8B$jJk#AuH#E`l==9b@q5|J{R(I~mzGN};1hsRO@
z<WNE71*YTJSY8*4YCP!Vph9T<8C$aa%tu+zHf#qD#h?Q@<<I~<_dZ`roBA$NtNwcX
z8l#8b{}T7^xD*R?&v|SkO1;MS2BhivK-D8t1&32jkdjl2!0cPgpCH(@Q(*O;>Sr#8
z?yyX@e{QUsONA_ET0lpa=MVLaA|C0ia-(>eaX8y2RJMu2q|k~|UbYpDv;f|*!`2qh
z77MivqjVng<R!mDIPaBHEh#;V$yKQTXNoBj37ct-ay|Z;3#8S}jP`-pXZiCUgJ9{j
znvyZ3I%RJJW@ts~MM93J-StM~r{+rHG;Skyp_oL$<hk7xWl$OF)N8ec?Z{;aD653%
zHrx-te)%Pct2N{tA<TQnJDAwZO0z!rrGJH$Oefx#c(UVYQL&)!@?)_oA09Lr$tzys
zwbnVZBwg%1s0%s6y^ePkI6}C&d+{-yk8@D~sai>l92@6m?-s&JN44Xl-+nkUJ;(c+
zcb|PkWJqjAI#AzbS5!OBvd5WkMgR3y@REe2s*%s=3j%th=}BH4e_A#zGT9$G;E!Sz
zQ_p+yECX^5PnYApjy<1d-w`zC61YXx%4Mr`X^U&z36qd1I^mq1NHYbw(PxwEHD{&5
zH}X*PU%YYPLHs8GJGBLuuxGl0+PM1ObZ0bQ`fTn4KB&8@8Xt;qoAzXzXkD=wZ&H`n
z`gFzPaQA<+evrLKUs`%Q+?i$6b1q+Z(1^7?p@uMniNf)9M|>3N$sGd$aPm_!_6^HX
zz|Q-7d?$Q+;cPekw85tu>APYT0jP=Bzn_8?Xd_HcPAGvv24WrCm=Y3BUoHj1%&KE6
zzvg9o^vLz{&4~bYPf>xf80)o-@_q(?@|N9euAL1*_>Hdc@O<UHmSuwAvSVP+v_4Fu
zLF??|u3!rujjJx$YgQ-3;`c{@!V7$pBB}S_*!vL7UE@eJE}7wHmF6{k7_r_vAXHy(
z3Brv#MaABJ_3xAVTR7s1q^T!Apv<;O)28f}#w|qt!7cp{(T}Od(S`x@hBuCDmGw-4
zVOQ?c_6*(a2W3q6FEv3G>9FO-Ss3v4VIEfnq@TBumVv#ZN5r3r8Ax$$354PG8dZ86
z66Z=g>a<A2+g>I(*T{SwkDAu{k<HwF>KQqgz$5W)V)dy%fKv@XkD6~n$%4H2CG{bN
z*!D+hqbKfYpTlA0L^QBDX9w@#23R$rG(dK)P8>|%q%mktiWmc>AzfJS<WG4V9pIId
z1$rRWCX3V2aU=JkkMSN15`}|9-)ZkQJ{Se?PE0@1;gH8hLq5v*q~nTDkDC9AGXt&m
z<x>=fUM?7&l|~tKufldITg-w$yQF;~{%dZ-ceYo2+u8Z^KTHFp^PMN_*dQ({0LJ(g
zue(lO#VHEOn^8+Z?P%BOTCVn3UV5JwpL)@%4|v2UD^>d#*sWpj)JyPoEs`@Bd}x~m
zSzp*12(B<S2IwjFPZ1GJbYS@G40DVfw=x>VkM8DqVo*7$-Iahr0LYnQB{_6;IC3H~
zKI1E>6g@yeMRdq!IrHbtZ03Sg+0XBQ-!4ox#?^BED<3C}eh{92F$<LR=|j%BTNxKF
z!bz!hkzhKI!Qi)<J|v+4sD;0FQNs$cFo=aY=)ku5^xXSXmQ20w`|_T<3OwA}zR2<8
zj|30I2OF}!F3Jk0k*9_6@SPty%#Yh|oV~0PUP4G=?N(euU7t3=TL3=Hg!vNTIZhjR
z7;t&(+WPvM?0<>{!pD*twSpM?K?JA`$SGlcSO*tf)I>_rRxtH6k3|<^*C^VBNLN9w
z=}bvh+Y<@lR8<4BO#FR(w<!jCKz8Gq|2{}#FBLEr*eAMe`sAm+!x8N0VLxi?alaCE
zxv!!H+7jV*4wu@Kj5hKZ0YA#Vs-KzUg8GQEW>lvALs7hw2QjBzM|ul;eu0QJM34);
zM&X6pe(R1@vsK=bnhJ-3)9J?!>!k!kGd8B}Q?iyk15`0pF61X2oys4$?$rT#Ba(XA
z`|7w)JAl_0YaY=NTHaPxaA5g0v$O<y);OZCLc9K0Gkv4d80Aw#J$Ha<{b-Yq->+)%
zSd>4jh5c5zbH9^p)e!p9j-RmVG^rA3<dJ%&=D0dEiny(ch3a(|{zk$JL&TOymhgv`
zlyc9L>fHqe0WE9Ie`7WZTxl-dJL&BeNx$=fSJC@UZVYiV{ot53k*=Y<lU1BLq9$Rr
z@5_uWwzgu3hGT1~`Bwf2Xt`#k8w9=BB|ilrzubEP0J2@F*laAyCc;oDb^Sc!9MYjZ
z=~8{$woLR1=ObYJ6?NxY=r0pbXxy4{+Foxd0(-y0&o8TA?N*?S(wh6ot=JAF3Shr9
zSezJ}<<>dOrqTX>WT16~QJMiA=P!EJVBR2tzxDhHZR^%=g44A?p*1mj`0gLQyS5|N
zEs*_IK_kMzm-6ywU?IMn23HFbtr9LP{>g5kNyMD(=`iuc&FdqO`<|_;?c%SOqr{^=
zOqf!X`_?>~a%c5Xl~CwNt>TM?VNpc`%BTFclZ&$!8@~%W=vIOupb}l%66-Td5L5ah
zPdnQ6nSyjOlf|%80x%Rd5JyPmqt4GA6Z=0-8L%bp0D{Gt!_D5gDeNl78c`t*7lAPF
zC%b?H<{xbVmQI`2-w${OkO?l<T@a-=DgDK9{rApzTI`l!K-4xO4`1NGBvXR%aiQGI
z^<pidkj_KeU~jJj7Ap5?*xiLX+m_phM0$(&O+;losFu9^<@Vi)Ms#F=B|;2$Y%KQ`
zrAIp)7{WmdaM$XGQN=CuWgfGDdQ{4xT%cRV;PTIkQRofK2pM(fr`bd(K@;UfW1Q4>
zdVW)fLI?!t)koP*ba`%fR2;5iME412ZQto5`Lln~HHdrbG45yBO^3Qc5ahwYdm;LX
zoBOL8&kVxJ*Tu6VQ#l34?N^_+15cvd3FtqG_D|!z!6WKzS)?7Kk&X%0B#P@yudRO?
zkvy&rj!#7?W#{n$z@P|0!>7v`WI1dG*BZPk`Yq4tL90g{k>-$zE?}k*#~n}m!PAO^
zDW=rByY5cxbC<k|kF?6TfGbIqpXRn6t|i=<@PqC!3_3=;sI!JDq^8I|8aaF3b*jUI
zU>YiH`zz;HD|d!D?w}vY7Ykp*gK0t=ETE({S~@7%_D1s{rd#tY=JNdymM)E?ZYy?D
zL08697z_ix9YEaANR05>0=_l)?v8<kG!NR?rtQqQ18aGq&iq&F`96aUp2EE85+Erb
zIWybzWz>!6-d%no%A>9NW=&0I`VSRSNa?Q{Kl^77yA3o3)+!JSwmsWEr<(df-Y)O&
zOwb#scHac?D9pB=>#vN6$%u;(ak(8AV%csa8&sHa`Ha_6n82}Mj-j2<t$NOD%;}v)
zL~t^0yFDVP%*)0cweQrS3=S^^?z9&w`+b}KH}3{b20iAhR(Qzeq;L52>58`XbLF?B
zRC}Ws_P)Cl`rM9=`lAn`6Vw3^F<Apz?BmeHIGGKa5y~&hs#MN)wOwby0BhmX@0Rzu
zh*^)1@l4Y&Bg1t!&m}W~?mN3vO^(CDWuXUfw%vD<P}@;Xe0+n2=fU#{PN*=EK{q|a
z{CuLY!wegq0EF7N+r1u;G)VO*tCSNC-eSBK{i*8S=W(R94jQlQ?jfsbdEEFQ{!<i=
zI@Senk75{feOg+9nCs-?+#8r5hxjDhZ-61jcxX#bmx9$wQx(>_r~dyqkiifHxTB<g
zT>FV@{mRL~)!E&CarS6lwxXQ84!+*$*`NLW%3t{s+#PIhU%2(14iH+PzYHcUHZeDT
z1v+-~BbAX59CV{hVO&sE!k&mmVp03)-k_EfESFT7kYQ6mztK`O4@nRjsR_yJ0cMW9
zT<lxovR7e(OFm~La<A%JSj2|PMBP>u(kT9~<CpmXDO@RreKV@(0Q#rxAICYFG|6gs
zYH<b~{Q<Ub7tVfb#b;$q)llxDWvg@N;9^geDsq>ZF-)5X2_rIcY!A>p5VH$n{y7ik
zsBrZn$10$Ae>~1#$Lt?b_UKa3Ma>dz&Fj{&mL^VMb|z#FS_JZ+L(ztc<}2Hb{OgnH
zMPIv^u!t2l)B-z_0%{COuEPzukM*MgzngBYx-=)XF-I6w2}xLqh>xQb2gkHx!C`mb
z+i+!-&vbVKr`zeC*0ViMEq}J(C_mcMX=6Zr5=fgZq+ahsVLr}p5pV@r4t?i)AFfM7
z$b|j1V?w(UPHzYsYbgyDnp>^a?ZRHbDg#imK>y*+jL{=Zbw%qsQ~jZD0=@9jHAJ7<
zC<^RW+iNnLS%tBOz-_7{Yo;0@_dQlG7WA5|O(VuubZ;Nd<P+~)!j>f9u+-}eQ$ozt
z@{SK)F$Q0SdT`t0tf}Of9x7YYtl!$l7g8hqfq{p>wFc%BaBj>!=Okpl8D3Z@bU*Cq
zdkO{Jor8Lw$~axX+2@QlHul!3p>xUYHST$SD|bjPm;+bCJ0FU*Qm91FvYl>YmI#kX
zrkd(9x^M6R85TwIPX?kX3xyETS95rg_PLO`Dq2{@wgwv&uEqmjg1#y6>s|fJ0MhkP
zyOl-9$S>JT3U4mjRT|C5uN5D_5mj*M4>AZe%k3@&?(C<)0TsT?^!Q}a+NA=P%8#fs
z5Ks0A=3H5mJ>zy?t_**9riacViL-3UYUE(a@lfO%^+dIPE2sRqDDE$gEFic=g2eDu
zBi7$X_H1vEfA0CJhb7R3a_INHFR6^nDnX1;m6xq9*Jqg2-IXYx^tuNeTtyWfdS8*Y
zepboC>ZTXwgRdR=5M+*f8Q2H32A3u)P6FKZXoAkoB{vx384ak_$rm3w{4)~O*jOdz
zuFNhj;Rmq<Y(A6`+?V<vgcvV_eQwfuPaP{s>{WoKe{5?y^6;0&DPS#IjlZR4TfS#@
zE8=ySAMM1*X07Yl(3M}|*I&{Mc15^NV9fH|&=!BB;XiW$q4?u-_8*$@VsC(O5m#sM
zSGyF&{lMK|S7mTlN)x(8bT9J^-Vh>zU!3k*ugN|-LNo{j?08JH5IN=N+wg@3AA-%|
z2<Lx8*jo@S3G@Ko9xLFN7X7Xm&e27jnMZ;Dy4{jNLodW_KAWN2csq`JTWm<R@D>$F
zm~O6g;7y7}>@fjE(IU4yL3}Rmecz=-4(Y|&)}lKP)aGHnY73qe2U}>?M0yRyo5)a7
z=cjkVs?(?-<g3yBW5rO6S$LWeMkc^=3Hx;cb&xphlGg32_Dp>j^YFWmZir0xsHQ)W
zcr>3Ne+u#Rz<NHaeg47AfxZfA4J!Sdey@)^9RO3NUjN94;_0Vwca{w;rdADjgDLoU
zV-$qEoD-oBWpghwv{imlKu&5@_GbL|?-fav<mdGLqHulzJ7EYHJsOy1VK#=@a-Yk|
zLMDc`-yBsCL_~SP6PR4j|N6*nGR((y326Lq3bgR6e6E2$3+TqC1E1(W*zJjWA=2N=
zR};i=!*2wzYo%3pDXa%HVsJV?7KiKjiwaWu`@K5Caxwq>mW>vm3!L}l8oU$UK(R75
z;KOz*0h^lM`2y97w~nYGa=@hT0zE=ZZ(QFBzvM;FXOK&;3+FY8s<k7AWM(kK^n`Ar
z6{sTAALKS26nxEV@`==FI9pJ{?Cc(^U>n5pF6DBHA+qm(#-OwY6amhu3RE<2$5&IM
zb}+8~k)eH|2Wd?M&{U9+Fy@pL;y*{oEYHYJXF|@L;B%T2`PKLUc(Sr{7P!R5^gNbE
z>IiA1rp`n_o6}OWViYXSf{|9q5i|Y`k5;UlMNIq#xA-99V<2aJSv2DH`J;PWwa}Kt
z*4Z7B=QOdw1ngrhz!4{><3zdRC6df|G`L%yP<fbrv1%#d8HV7%V&@Uk$pV(-C{h>@
zsZRY5QxCZ1zo3*>@>XAvaI838=_f#Z<ug-_^VRJiPl1Q<>^BoA7=CTGdwnMpFhGik
zC%+wUKmK!zz-@%G!#Cab3|MA##i{p{k37{|o8N9~zLl1ADqSx90b%sr@ce=P7iIMU
zSBBVI#z`i7n&<?c0=LiUwTpL5E9!sv=n#3jsjh;L2uC-V1X`C8g6G3kWU|kek+@Q|
zl*@WGDdidQm1otIcTsp)P$pYFmwR8r9W*K!G^`I=o;?2}yg+|M%nW>76>!YK)-$tE
z-|hEYZ@{J|sDAv{9QGIH?QHvFS(J%LXNs73BEYIa>K*A3FNJ+4sY8DU`e2*4<~PB?
z;Pit>*c+=FC{|O!f=2yDVU32O&K7>3!5DV?o1vBOOESc#R7!K7O$xJruEw_)_mPHD
z_uZN_x*P_W4JTv0@I$`&Xw^|Cs#&vjPpZN{SpvoAyXVfALu49pK623y!xOgs9gS!)
zzEY3Y8nf0*IfiH36r;$W%^oAkt240<E-svcSOZ+k@~>s!ItG4Ro%`o*ARmXm25N&L
z%@55ux@xXc@5_@Ly|o9(Pjr?=MHj6Do%9ye&E;WdGri~IoDtmZOZ2c}bv}|;HqQ>b
zyCY>dc7NfzS(Bm9gS+xZ;<#~FR!s3pu##V{Lf}_?E{@sK{t_K7wa9tP`v7JlzCg}A
zc0}YUAIOezJys3AF1bs>`}<{EU5lY$P0K}cNddm0Ix;-XeXB*zb|zg`S3kZn^5;;s
z9qi08b2)P=knT)ct<d_-H~=7x^a5T)U?UhV8jZe!s~`l`xZ|38J$x*HbK6Spt^LL`
znltxF0}m6D-^jh7L>KZNtfd=RAhhw6E_VR>=H&S=O$$nc7B4Q$aSc%?S(*Iqdxi&n
zz|k$xvcBtCc97j7!vsxWO%B~|d7_cr!wLEj9fz3Db4uFm<Q%YQPvhXBrd}PvwP3!S
zkyf;79iD#axp@OjkI2!!B$IignZAmQn;f*QMNayYM?p{U>!0ild`oCfs26D?kT_}H
z)u~xaJV;;Hp4lvTPmwimR3T}HJxmKasjzTU?_)9eXGK9$yP5q%_h=}v3rlJ#^>g@{
z&jjEW3#K5bzx~3WK8}mGQxvIgb^>0nDB>fFwK!=dyJh41pT8ChL8-2cx@)>0*IGEi
zZ$6q1iBJQ(w3yj^W~Ja{le}V5nmZPd4LJgwNc`acS5Uy!>RjBKrd4#5N4B+0*(HnU
z8W<}TjMRp0r#<UA+cyL9{Vl|F?(UtKSiV=hx(`n{imItLG9TI<Q2HDdhag#0!8khs
z@J|5Gx#I^I)?*ChqKNUvXq8ewO;?Sj+#parpK7#CD&zUOb)<XYyGXEvYM@Kup6#)-
z+dywjlU0O(&j-J#k4-CHMFj5NUnb@eu500NkiKNsW@1B^2ow#AbC41zjV0e$-rfE4
zh$PN&L~RKBqh3<9pvUHXjHXEMDKb2KJrKt^E&{E@-1=GiMcW2~*H|W1$;<Mva$r*J
za8LhkduU%sG8#lm(*bu&2XRuQ4O(A7dhW%@zKgp192z3W9kT(3W{#F^qz6k`&o1m1
z;Fc)q-F+m&|LYI{(?h$<3f6nICh0=l$K-CkB`Aq2JF(JVNkz-Jl!zr4dNejQncw>d
z2#(6de<a+%p*j?Owgc#96LX-9wvn;rN>+^p<f^B|mKJky>p9}rt0dt3_#HjZKG%}q
z%Y|O=3?A>H0HSn|SN1N52m6eST6hvKg$-#j@2V3_8_v#_HjKWDEH@tu&Srhm2~wJ=
zW~=*?ZxAyqzJF;L|3h06`dXP70mb&_N*+V~Dp?NT2pF5K)e&K(d(PWo1?&I$Onf4D
zc6wrWGE-Igiy~VeU69GQTVMCRJpMK9f=addjY}F0y|29n&+yuKliwgB^5UfJr^sIY
zS$65bI}xM48&6N}yJ3+3h4kq0Nx#R+9Nw1AmcyYeR23hVVWdD{qoeH!;r5UP;hFXA
z?e4kI=0K~hh1FW8EelMKlau%P=tt8eF#2o)?!|lo>Iy1eq7-W?tErtT86v=&=cLo5
znpX(1;Sz=ukF~mC%sEsygA5dz?C{!XvY(j^HMb?BmaAz3#=5&{92w>s=iJ=agG1cr
ze5l_Dn8<rqGX|$CvQlGEUvkW$trQ*(DgM%h{0UVD!y04^s5jg|cgKVC6wqmDX-1*o
zu-c*L&Z+lsBTva3C|mi$Tig8hv?)a3>RB+iv0XqUqNDfI!^0zRdWy3eQYaSV=O}wm
z;rC%o2cvucxZHz(Lw<JfTWv2Kns7%gJbyN8n<?=&&o7oDDz0<k#lJ|%2Pm7hNR)ED
zM;I^iT%mq9JuzV?b@3k>=>O1sgr#;D*pBlUSbw9JJZ~`SU$_`EkY(7l7R~%$;nHWL
zf~(VvVW2k(S_rJQA6@|$ex4ZC^A!{UFHsv$0zLR%k>GkNSUJxDOee=@fY;$H6?fu`
zm7mCqb69!T^RYuY5{4bVf|-u>bCl$~kC?Rvl{K_GzX>kt)JNV@Coj7iN`juvV=$G8
zJ%#*scrUCZffL?21j8hN{ykgWBb}V40Z-vAGq4Tw9?{o03^PJbBE~1%b>ke<bM_%~
z`7?yk{_EeLJ$ojz@#G!t52MQX;X);XVP%brwPgvoP7P$9N$}n?#awat1ObIo_DZHx
ziksa&ng^~4l-b(%hUS}CSR&+z7A+<{X_lr*hUCU(W{QNV2CbI~`f0~4b&B5JWHlyd
zK5Xd&cV@T{5Es@u`_muJjK`Xtg3aW87@B`Sx;~lFzTrJ21y{qCwwZ9cwT}IAdYIr4
zXfpU>{rq}tCqCEaK(55E!ly$w@&Mq;;rjaendnQrSdZbIME)W#@SH}5S#YW4-~$vb
z<{@WFzbt#I?kQXCw~amzi1ANkF?JTfzs=3dCuu`on8)nN`gT4BW@O)&<Qw{t?UT1I
zh0g4P30mT`7z}<1o#oZ(?(SuTJWxP-{NnfhWnCR%6DM?E(@Kx;Mlv#r)Nh;r;ix(A
zi82|p=Bk~7<b)ZrjE-}fvmT$DPe0&`%lUFPHIlx&x4WZb9X%<3W`Li-zSK0tSQ1%j
zv^bc<YU}0T1TPB|#?`?GL+`7wQ5ey8+d+c8ZK<zQJ!}xRF?hDNtI%Qnt-|B&fs@-S
zA5$Er^hq9EQVbe-)AKy-wB!CC{*51_I``DR@v6lKFzugjViOqj4MJbB^64=;zJ)Ae
znNO#PqEQ+X9;2>vPFoZK@*3vGm`~+8WXl+g6YtS~i0_{lU&B{OQQPc>cVY8R`rZAr
zXU|4Ct@=ByCOuI+&MST1Ves@x8}$vV^f<B26%))dZK+X7fTRP=d843ss;|#RaD)C&
zONHtNJl75OooWt_!&lEw>+A7!s{!>nX|89a{fnnmSSQB$>+d=rR_e?<OE=zqGSR9W
zy0<6D2>2w!f8x!Nd{jHiI!}q3f_HK54HPNYX&waPqW5+53dX$xtjvopeWT*Kyj0EG
zV?PAbko<Wr_aP&6?y9ep&?<17d9c6REIlh^(sMs_-dE}|lOo@qL{B;xISYKQxMaim
zh*IWwXP4?}ot1$48=rF@`A`ImC7X3P)K_^2bbM?(eDrt_@tKbL`nlwlvRluBUdZC@
zc?~<)=P;Y`+E`qXg|)SH;rGF<4VXvCyao?Y-^9D$e&zegt#ya><)Lmaxn3xd)erZa
zB()?Wav#`hgfn(H^_!reJ`=8?;l+xcIz5rnSSBcc?^K{YF?+VzJ&TKBF$sw>cx+WZ
z-y(oUDRkt}|MD&1Z0^H@=x_fNyd`FgiluW-_D7xE98D>Am!#o|s!K>n@Pi#z9@hRY
zE&a+cXxbS*s}-!hRIXG`-rf6zuLul`$&b6@zMeBxU0H!wRq`keN#{Dsd|sq)xbs3y
zrTN&A_}ijn(`2!Tws}s&>v<<(Dd$@S)BEpk;DVth5HiX34`^a5)es<LR^J=<nb3>X
zljvO?3EO`C?RtU0jdn;fP{?qMO^ozI!EykthONXUUk7V7UAFvC<C>RM#4+7nc=jG2
zEwiF<7Ri^ozbdUO_h7F*si4DgB?eI%Ph;Kkgkb=Yghu10-w5wB<NYHyG**kJ<Ks?e
z`@_5`0{WE4LyyX+>&4M4+msko5Rria@@#jUVVALfxQ^qIhEw3}-@ji4<hu)FH>d<^
zPYrNPr?~h0YN)f?4{{VOI1H;lwHL%MEW~lD7hbcS_X*pu3go=G51)+Ujgwlor_FO0
zPF${s?k8S}iO<$gF-86iBL&sD&-_+a<=&T_1dg7A_`pX}%6sl!mc-54;00I7bH2q&
ze4YJ8Fz<@N`G(A2N6(#j23IR9et%6{%J|sW-CG>Gl8+ueIt8=Sgf~^(f6Y~?Z_RJF
z7CP9ScKJ1^G-oy|ar*RWfy;sU(R<R)0C`w3WgGe(%#Vg$6?1|ZKZsFUuOvYExar_Q
zmBbL`dk+dfQ!DXN8U8w9Rn>4lDYNPk5BBl*hFG)`V|S+`hQ<Hkyabd9h?%ex4X@xB
zsp#v|PjbJN4;1jz2hO|!iquWS>0f1DHD-MG+)ydPbMz+{RAQeEC4qNYwKMO-r>&5n
z7X*hHrPGVaze<D^>Kj$};gOuHQ%k%ldT5eNp%5Gy@?QMIq6uEPQn(zed<r;f`dgWs
zU$GwCAhp7DS&6EzRw}Bszo{%-YfdQ~J}bO>_w?(^#iIzdbDfss;fi#uo7Z0F8y9K_
z`l(m`^umiMj`D976|B>$BW3Il!cM;W-(Iwr>3rN}8qTFYddfOZxZ!qT`uAe>*=v17
z-K_6_5EvLrHV>19)z-85iS=&B+yDN|3XAtx2}*R2GP8oD{nh7@9j8{z&AH0$_3aw|
z2Ph~fY(_%j_KA?b-rlI)-tbs)R8u64a3OWAohgK@yW6_;`3w4^^z`pI3Kko9srGtX
zQk;u^PHRPyGx-)5fz2q-`6-O`3-NN*<3FK(qyFk?5oB~<lwZ=b=NBv4us*G8X~FP*
z`(K?s+~GmY7aPkqVFsA3<A#;blb^MzP>PFoAi55t0I6Hxs)h*_5E2p!V=^_F*7jGm
zTqrIs{_v10>aRT`d<fZjjhE|8%GY-EsMnC&ThwhO6t#(W|Ni|dCu^t85rOohcv?dh
z8WD5l!rNgQ|Ld&=J6?k7=<(z0Kfpz^;Ir%f!6%~#?Lg3EzpsZaaS-3kkK;xgVM`x_
z<IHSrZ6isK^K9G>)srPb?<eTrZkZb5rzqV~c%d}G5=pK#MiCcm1A>bo3GgV>$u~6A
z0U|)JGz{u(vZ(3abm8y6gtyI7*C@n5ikT)gMTUqL{^QwD%LIvSoJ>X3rCl9tbFT+z
zhqFq4$S;R&uZ9(5b(&ViEWNWvdUlI2Q7t_$)NH}L<`%8$;!<~ax81zleM$W}Gl4jg
zB=(&ng-XIVFgEhbu)&@V6d69v$GNql>g&hE7`?KKSB6LZT+cG0<w!m~ym-u6C?F*E
z{U0xY<{TKl^<$;%`<NW<DZc+bp*INW;r)dPtPc9b5z9<obK8h>T8qP!F$8staQzY&
z9r1KrjG76q;<I@-H97eXiFXq=pClyqTm$s%@G`%hD8!|W_)iYkl{wJRyAU>=qbQ=N
z&(SqAmvrwdA!bcAeafY@+S{wqr(Q>Viii<1_>I?I4f}f5u)=G?L0N#a{i_oHQsGxH
zUYp>7=wx%oTjFX3SRD$G^czl&)Udb%-XA)bvJE5YhD{|;XaWDEa}J#ByF$y`n|6D!
z&x^m|@noNz?86ZeP>hV>Zg&6u)Wq9|7L>{8FDJ);LC=w2EF;qQL~(O-bK9vef|!`1
zNajcRI*s|1A9QZfryw_@*3E1NHWK;~N%k<c-R6RVL+7_m5$F%wvlv-$M}=m)t#8wA
zZ>u*tx?fFRd=I_p&Bn&Et&8O=loz5#U>+>zn;!ALZl%kizCU*RC|+FXpsjDdUp~g{
zsBD<e&8Blq->ca$B8t_}!waNoHJ2L7*X5jhY)cv(vhiZNpuQw@GeUQ{l%cag{#5kf
z-oya+I#5;k{qE&<yfC%B{ewh6pKIaPQp!mah;d<R_lRKBB#?ZBS1YviU8~K0ftubz
zy|Z95&n8*s8U7P|doKTxKUQzOMX=v6;(rns_SD{=BimCSrA)z5RhY6hgFBao0NsZF
zqq1*$3jVc>$z2JK)PiH(d*bJxW6cL7#j3U!tT?a7#}d6DohUSfxT~+5b*(3x&CVP~
z$k&>ia5|qU0#<BFM}eKyId)7rV)i;xhPj#PS+Oa`w#&6Bb|T7dn+n{yUw$cvU&*yi
z6rsiN<?*3W8;@>bR&FlcWhVS}=AuIA$En#`kENw0)sz3ue#-FX4h&*b#I&-;C#Jca
z4rBL3dw`=P6?$P6vY#}>uHjePqqoMB2~rN@BFIQ8#kbN)bbGGJ70!LfI|V6m_Z2!9
z-`pg{t@<tjZEFrM|8Vy~5|8kW=O}JEhC^1yGvniHwZ*7CL(O+q7GnI(USK2IPt495
zpEA?YZ5SIH8-0D;(7MM!wNrR|wRU?6%*uRYz65zsKm$yjzR2cL?)jI;_u6*L5x4e+
zR%*ZB>6jNGv1^(|XUlNyvL6#nT<a7%ESMJ*YW~>!20CWbF0!zuH*V5U1(dWkRI!{f
zG(<4fl^~4yf>_xV>^2J5g$h$5iW^Htjur>D`7nWr%I7DEgZl^Ef0BPz&09X{UCX?A
ztn_%RRQuOw9ceC==H(s?r%$fcxg=k0N;@UE9DP<$S8Ai@>|21(_pI5QQ?O(tf0m5k
zfT#V|whE>~l35ys?%+{$N`J1+W<!RjCBkiC<RrRrDuUKxq?gsVxE9Qg=H!q?L!Sa{
zm|D|s2OH*Mgh6Gr8+~<k(g(OCc*-DyzQx;ZJ<!9z>V~l6orEr3GKm*)wl#oEx1|&#
zf}u?uwWpdmVJY<cud)hV3RdHi@48E|U!QRuNi+m#!&dFuQf?C92XRt3{W9oTaqz1j
zu6_KPo#dV0*_UIOr6mXUf7VH;7f5Cp+yap@5dZ$qUrk|aO>v<@+}xS{#jgg|E8LQc
zw$}Vlt|%QY5*iNBibQ<+)=vFS9_}YDH5yIzPs*;^$plfo{{au2;d$Q<zM(OVHxw1R
zotzUkbO=u-;pqYL88^pg6CRQBpSyDiGs+~|{K+TDLcf5;sldyBW<bpB#$Z2Im>3s#
zwvW9**daehaIN>CuyvjO1*q7zM`{GR4m<=SpPABN2TZgR^&Z2B=cp)1_l+{*al~fS
zH`wr;tuNTJ4AC<^_j4kreL|E7(NPEDB9oVU<Gy;7bB?~U`bK+0{Qo#Z<MzES%{l)n
zVdzu46@?7bH+pxJgEwH6<#eY_skjh4h%?fOV{7zQk+ne-{PmYZ=iW_t2leFdxpkP?
zG+x?$*?NL&5av*v3!7U%XhBnPOXe+O9SXfd;M%Zqb8-9tnQlBi@jFrpf~4{3X-CnE
zcTm#=tbSs%Arv{H@lnMqI%#%{7TJ8xkWJi$VBdUp##S*bI34z=S2PqO-fP2#Us0X$
z5xGyQgZ#}QeAy-b@+^X$4Px%;xFF7i?L@!;*lt?mW_iVeOO9kyl^gx6d!SPY4*A(C
zw81_*8-G86SWJCFO<%Lo)!9qxl#Iq+#l2H1iEd=JrM&njSRj7cH=Xv}{F;(7W9I45
zbb8OX{sYrDoKJ<ftb!=O@)`nFbHA*?(8uKx;K{4n4!HBu6e?FjL@}>m@y};bWn<n$
z55a!`xtASVW;M=L$dp}LXo_^^mx+zo>5H6b*Mm`%SreBpbAHbc>ugqY{jfYk^ns6y
z%k=*TBJ(BuWp3grU?MEMCBSklhSwb^5~2t(aC-=94O3em1DPj^Fbzexv^LO1<w}Sy
zVYFEBS+f{DKIXsI9z4#Al~`P6YICHnOi(H*;lpi`PDdwghp6%^YK8beUZ08jTC8tB
zN5ZfEX1gOS<sZ`2_`Bf`?)!$z5x;YHcNl=Xru#$XpzMbx;O<h%v-<gJH@BOECtf6i
zd}34tlG=h8dq*rOrAhyc1v%#p75}XV&oXgn$r%z~QT6%cGg6w?c!88}1fKaC=B2DP
z{9@tg`#_9@fQ6;S#e4xI1faje$)zC5cOOiR<Hy}kT&IEE78v7=b7VW`i`ozn(*_e6
zPIkHax1Uvdzf^JL*%%ue>v!$Hy4M5zAtA(aH*L8c0MiAfyPDt=@;#}?%L4#wMs7T}
z=}x+(*fYmnqh&R;ZCC?Z`pew5FZ??Y5fgZT1f@0v#BzPX8s~qhihZUWotknFUHCY%
zaeLy)6-r3OaeHzfo-PzMf}h!NZC^$&nSH9AmvfC941<cLzKX_8w{~Upk$&8d-m2Mz
zG{hl7j1@AbTpDLzVpIFr3FYfz^4R_)6;i?!yQP@J2U8(ehX#pSuj#<Z+7@Vr3-gQd
z#Zs>K_PcRzjl*!c=>#S@9L9_b3w*Tj2qPq)u3?|5P7fVsVWMvO<79@d+QMO&V6`$_
zNzxD$oZrBdnt1J)cr*Ze`_*R2xc+w}3&owZVg%J8X8>VbcXn7mw-{GuO-;?ysrF3H
zdt3+2DDS+4(ZrelY@{Si2XTwQXo84t3h{S9pjFCqDb`~G^NN0N34f_-44)Z%hEYff
zT@S1|=UuEsdc1SMb9l4-R3MCc@`28!?vYQL$VCVCHOW>956CKZPU=D8Qd0WmAF7?|
zR=rw<XJ$#zI)W3*18%gOg`2cJd#5cS`DL5x<dUKw_N_d9`<!E1LX-O`55aB6)C=<h
zRy-}q+RC0}5(Fbbp`lxn|HiC+Nswt}PQ3j5LG$b}EqZ7e@q=LIZ+h@$o-K6_?<DES
zTW1Wx<LQVtQ+<q!1YDeu9S{3U?<1wiHq+0r#S|Vm^!QCDG5c5-C8?!^%XmlHqq4Uy
zH?Hpyo9?DDd8mA}4cB!4R&2(|`J&&%A03w4Fl87N_v+INd2N3-89q?ip{SSR^c=+g
z2n1h_zsnq~mrg#(!VBb84sRNqyMJk8rF60C{`)t~vTPR>$gGTJCMP#kLfmBYUfpGh
z^^upIA4YI>v;vU(GG#j+B3t-Rtd_jl3%moTDr`Cxd?f^LF8D;~#EGZ;Ew^#@C)lv;
zeFQezd7A~?Nyj5ZPI?Q$x%;k=!AgHt=-_?k!NEUxo&3m3PH}79<GRSKEX)sVm8W;u
zm^_+><r4O9S^&QNDJ6UZo8SNmpx^x$E;}NTPCTMdaTVsFo#2NnzED;qo&wJqaG%>W
zk$2^oXzuUaiN*Plc!bi1`CvccKgg)fO~NF~)9?uCSEJTRq+^au(44!~JR1M4>)Qe_
z+bmW7@zNQ4@c6q%3>-MKdtdZ`K|`god}V?tGEkcHEoMPsoUes1fo9fBpt<VwpZX8A
z23NV%HeCfSyMBz6-{PEMgCk6{n|5)vT{9mxuHi2@jnAXIWUNGI@0wa|^h;(}+im5@
z?Nsc4?jdC9Pov>qHH;2g755O66=T3dM%9KQg+|z?0lB!n|En3K%w~nfmAGAamt?d*
zbdfkf_&j>fbu=h?n8%0*sVN*Lpo=y0>o)h7-EwIs?WbV8$cw(mS+LkFC@5$qaMwH;
zAH)u`^V#3Zw;oc*9Nmp)_;jb>Vi7q?idBu^LtOwD9r?EH*8z{iKZD`$Pqk;T@Lx8v
zNQg@VoFSq2IW!7#X$ky@kM#~XOeig{l*wXH<Hlylkd`BeKh6H0rDU=z$oFBi0gBFG
z3+ez7k#E&d9tUJHMrW@S=u*;LU2fH4Lt6c$L~)w7d!9y?wYRLxm*ZUqZr_LEi^UB2
zFu2vOL|to+rol@mnskj!v-?_0)2CEVz~9bV|7EM|`fI3ZH_!b$xE~<s=N*wpvP^Gi
ziyeBxIkZX)&iK=rb7qW%A=6+WevsIDU>;3ZxP!XCg)-gly`nmTm;(Upj7L(l5|x!@
z%zyvz9l73hD^Xg7#cnZk(gvZ`w-E{Pf*UDmMrFKTFtG(RtN%j^eZbgjs6yi96YvO3
zDY8PXx^0BT9~qjRIeU?xpH6V#SZy{(&Y@qOc)%W4d2T?wvk<6If)W+h`G*DV*j4xs
z{eft-zne>sG6tp5_@Po?*6THuEIYL2j{P#!(CAhDHv+}Q4hWP363O3g*BDdFh9#k(
zjIABuChuVmnJDhZ9VToL{pk*OdXBP~Gj<cA7fINamMV+HoIY?dUlc~`Wwu`yk`Z)t
zjJhf?gX{qfSJ(sU#E8LJ_!IQXNV;OB{_i8$Ew4-s)$DlXj%b`pMWgI3wYo4c8@|%2
zL|knqCnpz0THqaT@283Fd4PCP2zRzkA9s?SCKhS_{!&#{sPYGLB`uGj-w{r>A7>Dl
zGrsNZ_v92_0YhWGcXz?J9Po@XSO3Rhhp_jKq%<U-4Z%gO|EThEbjL&TE4u-^u<6K_
zQI_2qi6|L3<ecM`NjRN5GxNoa`Jk+`S2A@QS-d(pa<e*BY$S8zwLxF#H&k(M(Y0#M
z&6c-pTO#g))oQee#_+8k^T(%f#Q(Z$5bB28=&)`q4J^RWOkVPmFvUY?dpd}B#BBMA
zGyai*$AMh@?Zqfmv>PuE9Mawk`I&OS_N$>)vcYaYurJQdj(c`%u<<Ecg0eo-{Y0bn
z>XD;vyxA^Wix^B1FV+9)dtiSDzI#4lsnK$45$J~$2V+iH&_BQ?Qo4{xO+<Lw`u>ML
z1}{~Q(?9Od<QGPo*?aHfc3_P<Q%Gu5Bn?Ywh)Q`G4o<hqs&+??iTlW)>%SSFPK=0%
zFnXMEO_%ac`g1!*T{VUrWZR>npg%#!qKlg|gbE+RhBs0$qJANjte<1Ae`a!G!Yw&>
z{Qu+ZzvH?7zyEQ(9uX2rRw`7sj7TWs(Lj+AB4qT;mQ6N2(v}e+WMn3@kR27-p+ebv
zZ$jqx+#f@p)${#%{eI8C9+#eVzu#}Sv(LGnb6$2{uf1-BgOK6QIEu2oVeC$fYsMgg
z^0T+=c6x;>VKVWlb90NSC(U-yswrdcNVvulODvyhoP1EgAVIam_Fub1a3@Dc67}#+
zU=us}pf7^;$i$bw*R9DFR<3dsyOeI8k3l6`2t0o;2iIi*#rzn=k0=$k{|A+0`Mvn~
zt4f7uWSCdatrzCuzlD$d1DAv@`m=80S)5T%-EEy7C;aI_oo$n*u}%({{?b{0^fluA
z{CuPQoZ`Y)v5dU^C=bg-H%uI1-W+sYQB0WX9mMGi%(ES!-@n}9wbpDj%g!>584n_c
z{aO^*tgf1;=Q;@lsXKU;lhg`D)rO33B7&>D6BzBEluT~qdsrS`nH47|-G78i#N!*S
zqmAd6cQEIB6ooYTEQB1<w3Tfd^_SY@!=~=Zcvs}mWP~f)K`0gNy&Q{n3wOZG9MFF0
zrOAJKt(FGLQ51Ji=JIR58FQ_zdH;=Q$B{0ka~OKPVYq8fe<<=`=(Lk?6ii1UeH7aF
z?%kW!((N8GGdK5!?IP=^qqH=0<29^TQENIH#Sucd%fsv^gLUDu-bplAb>}aP3FVsr
z6~7Xg($A}hx<|b@e5OH_iE>g$@@o<k&;63!*kp{{ixvx;l26M1y*OHL6mpOsGu((k
zku<02Ec-d9rkglOjGHPgpKtrxTOggRy=h`8X^j$Snx8+#@KY_GCYIc~)Ldv0c}6q>
zDniE<)FQ(lo@ITkg?Tk&Adr)mmgYEOKx6Qi+MvH3hMX_AYXO2KTV2DWRRv2U$x)M$
z#z&j9f_{A=BN9a(7A9=ZsL=!$LM%}1;2jju`q*1B7n>cbfhrFnbbHUcw!FM76&&*<
z9&d`V${0^JmfcwILF0-|S>m41wm6DCqA*`DCtWLhq?7S@GzMoVuz#8J5pdTXO|v^V
zw`~sLu8wTee~jYUeIrWi<vlXuTS7}q2Wsx3wUZcmncEIrxM5(>=L<E*_vEEkFO%q)
zo=a{4n{H=E2M?r%T@=aT?LX2iPjJ8!S@wcWH)yL3qreP`4~jL<&&?Se*3@KD(FX)B
zzi&Z~2BmYo*RKm8rFtq|Qahn(y&unQT8F_b6`3#ANLe6ypxz0?EWse*(A@FS&%AVn
z^bw_^bcDq_iOSw-Z;5W5sh?LblHu=QbE@FOjt!53`FIJW^q;|wV%Ny2+Z=A(jpyZj
z(I<#!npn9bs$Sz?tHKY27zsId3;Su&3@BIiR9Eez+?NcQh(CI2xu0hmRZ;B)w<_@6
zyZnNJ&fR~{h_aw;gzh8Tg>*kRG+gw5ID2UTsY27Gcdxl!if7#9$!M{`3zOq&@g}W?
zTR_luLQ$v9n9Wuo9yMKlF2|zfie~udxws%|U^X>@z_)MT){*>W&I}@MB`_VyN&+Cn
z6Wq#`d7M9>w>Dk;N0L&}LHaVzO%oO|XCWi4bh|wcM}d;(;3`W!d076LPe7QKce#7j
zv`)q2{0{qUCE3I7uD7@zJ$lr3068kU9c5E4090?nLQqHzq@l{d!^5V8J5sbdBBeGQ
zQ}m^t%`OjfX-IKSRM@z}DQ=aG_y+hR7Ny%(JM{iMPQdyu0>g`^nXQ@Uk9H5G6iFHg
zuaiso?a=tx-r9;vFD~PscRRRNCuGX>Ns)j8`o;GZi6zi7@YVxbF|OAcJN!)M)=eKW
zRdQ{CpBJPzF<F)hNVU3IxHnEKkOI#$u~I|rpq1lg`Z-Ox&*wvt2t^<^GBVN;x#Pt=
zG)!D0-i~Fh3Zx6RL*ueL^28~aBT}e$j%Yn?`(dI!br#E=7d0W(M{cXF|7nLe45ZL3
zH0lnI5lhw<^V<#B-lMEc8^?d@4Weh)K#lKZKc)kRPk?qY3?l8<z6?}lGK$7Mx8)R|
zbF|*IIy^eMY{vQVkLBBN25DLZ6)ovn#g%y5DvOqulg)u&C;wRf)O+%ELAGIt%GtrM
zz5lH`s4^(<7*0OwO&~BA*81v%Cw(FDU3H!1H_p5MxtT@&<a4Ax@6U1Hm<DDkW%;<L
zF*|nJPx#TzkC6xIX+nxS7^_!nl%iFVa!CZGP8(x8zPE!-7tA=Pf{!mosPBuoO~4F-
zUSWrbvAAHeE3mw$xy<~Bbv663%@YvwXs3ECEko4t;PnDj{9}%>vCFmtQHBt|(o1*w
zMkz|!<*Pg~Yt-~Bc}d)@)MH^~wT@ni@iSJDDEYL>JHQJ)=c5v67IFGO#__C6QcF(5
z?cFWn1_bwTk4r&GYZE6JTmVjZ#QPK`Dk5IVMIt)!iIyCicy$LO4ShGe;g&HY{#kX>
zM?oL<wYj>wuD_zqd-`W%<4wTV_ptAF1#hA_yq|IA*>XRWxqG|1x>m7epkL}XOA+4J
zq#6DI$9Ye^z>Tmgp$@x`gief6FCw2*^s(1IDjDih6KDn<K1l@JCwe*BWhw_>3Cx3_
zsk;9)H+EMV<-x8nM^c(yyG5TNtJOL)!J+tYVBYC83f9myWJaxyHM2tmg|nnM=6w!4
zCw4`R4I6xb>j&ySB&f#t0MxN0mO4mgHN#&r9u`30K|^7b6mMd$@JZ1i-V^37Y$yFp
zCF6y?s4AG3r1BR18K3wV;GlA}>CA@pOCohvG{z!YU7zLoSEV2yvvxkVT$l-S2og5I
zD+Ouy$<vK+Dx&Ks%7e>g=K~Ua(2|9GNpO&V`E#P<7NhyfuBm4#oyO=9w?@y+`CIYH
zlPA}YdtCgaI-GHufDzb4NVudf7^>nrJ2f@+Zr2juQ4APkPLD6|c&Oz|a3u&FZ$LeC
z`@$8z<QVLG(H@}j>gu!xQUj)IgF0)p6-G=`kYc@PSABiG&)-8O(iIhOABcCsV6}N>
z&!0E>sc9igKyJBr|NcUu2G$FkQOgscVC&igxN&29`{Y5y-BOKpUp|mxdhO^Ilh4in
z<3;Q=D*grI0bz!YuD1r20j-Qu1-mzqjG!<2w6R64Gi~CBI;D-G^TE;N^vHbY?Ivfq
zolgGYugef*2V3%y?m|BHkD~8GP(W|f_(f;-Oke|<wqx8$S(qf1;b!glwNaOX&Bk%%
zeVqYkg$?Q+6{%z%rHx`Y^TTyd9$KcY2$Q~s5u+OAE_oe|m;uBEW95f9WreYV#R~<w
z$~{P!WGwj9B~8U+=}VNy0!%4bKdd3mh3i+q>Sl}@l78L5kFp4a1Jh>?iw55=NnN<a
zC;yMGosq15zp_)Oi?J~P6jUE$``0$vZ-jH><hugo`SRZ6<cR2;r{xR})rDR6*PTwS
z<cN&Wn6wAAP;Yh3(|Ug$KMIXy9N0^4c}JxmPSO@b;iZwJrn_ILC&w#<$2P~YF+0bp
z>Zgd)UO3`0Js$X1QVXGu*Sj!@%S2z{Tj%WPTN^LmKZTYjLVEbdlhsp~%;edH&eIXE
zw0-J%S&0qffUEouK|V$`i$Mjimnb0xxZG_a=o0s&q0sm<Q|>i+hx2!5LtFxhPqcD3
z>|QPK$~*#ov+Q3#frxa!ihhFKBCW=i>CpoR4%`!;csH;k1yG*S(~N3iC%qdsu?pu)
zkZx%mX9ryG+D^oodna9ewQ9k_FEYzrgOXfT<pYuWP6g@lZckHp9*=2bs}JnL&~%th
zdV9St#gGP@gF}Ytr8y4kGJnUXC@AO{_U_H=ha_Ne@;lfE&SKs3fJ0GY`hn7lJxk41
zoJ8XY_LaBG$g$mC5pM*vc(MLl<%pEkkHUe&*3l*7t=#6jR$v(@6E9MenLFe6^DWe5
z#{Hb@xBcSbUn;ZayncfB@+A$OAMY0@#UePIJW$B~<LYa(tF!yczHj~~zXfUMo|n}3
z>aGsZB(TNbB2`On{JiMis6hu9o+d85O@$ojX~d@9+jCz}e6FGH-iK){0H99>{23AX
zBr4=+q@=k6#5eb|D|=!6iyMf4k$zC^+%_Ub(eB$-$V#N=zs2RG`_)0seE2Mi&#=WI
zj?J+&x#zZ8*5$!3a+GO#ZYZMM4=l%)6}fs1!t{dOyAV}Rko8SJsA!%#$5tS2Bh5~l
zdMXU0{MT+=XL?K4;{Sc;>1cn{ye+NmcUt@i3vsOyerZ3tX<WLS#oq;jbJgOrc?wgJ
zBr&QtCMqqQcbMNSshUxvSvuASX>7;QPfc5{pP_mhi(XleaAmPOwg;=fz9T?`j<8C6
zpNPpLblatsFEC~PVL^^hxA+(*EJu5Ug|2?sPB<k_+_#fC+2ow^zl&#oEe?p#M^kF;
zGmuilETu=OBM~JX1CJ8ZO3A^uqr<~LJMTGtJF}UNgj@K^MOEBjwGJH_db)$dipdvQ
zCZ4P(zH(S(H}UBu+mv*_S<(%9R_fWKa-1-LhYy@T(qCz43$B#K%G(igJI!Y?jO+p@
zj?wN<E-qeo{SRTQIo}1VWtRqX16>UjImps9yY3(My$_Mwt(<1JTj<T4UULRW?A#@W
zUp{_L5lGxOw(po-(y898>rZtGV!D~Hd#|ju2~LkGWMvpzsZDf9nfFZZW4pntm1R<%
za*C3glw9q@K`a9OCVBvwL0Eiy2XcB0BeIGE+J#tHsT}D#3m`YsNl_-I8&!e`!o;`5
z$4j4XAo13QoD6`{|I}7^7AyH%QCvvLPulz9O*M~vp%d2@-lAw;mgW<y2Y-mFK@hb~
zi_$FoSDI7D&Z2&ht064~BZGF>Vv_;1wDBD#XH&pE9B>QJ8s$+3o^=@5UlQG1>#O##
zLva1##Ze6r3>|r{-}is4zu*<cFb#}XJ<%+UFAJb4!1grXWR1Iu%287DEV}Bomoo%_
z2&(V$U0hwQusK1T+StVK=x9DnZeD2`Aj=B#(gGvwFJqvmsg0r4^5pfK|56bq8KCYc
zwhQAY_19}5nW#~nuisxbnk^^r230@QD$Y0G4Y`R%z!t(Zg2JNUhSJTiBD`EVsswQl
z2Od9uyrg)e=^D}Y?b{(Cv;OHCr4!-rU`1Jh29hBUa_{2PzhOvuXV#~dN*ar>vzH=6
z-J`$URLn6l{$jr*izKz9=+gO>&Itek{w3><*4IT%OXOk`5g2ERo<6-DS8OZU)>{B5
z_(e~$YhykLLg*|sup1pqI6izcNh|4yP4|$uGZt`ZAf9k}v0%Dv+*xqNKIVOMag6xh
zV?!e&E~=31#4r*iBqk;c-zIa0f@#kiR~A3PX=Lp3=ObX*J4c{8KRAFu;#j%Bvd6v6
zn>~V{KMLPo;Ty~=4gT8H-hKd(VH07&7mV=66?Y^1Ss?koO~4Ib>NQZ+cUl}f5aC_t
zbnSR%hyxwbThGq9<k$X*`vdn}7CT1!Onkz2?)zV0eTan~;H~fMlzIhlfVvvj=v%w2
zVR@B@m5Ip^39+%VBqn%=TJj~Z@7$JxIJRU-MeO@%RRunVdm32s_FCiuz1@jZ0w6p+
z`yNDuAD1<5WZ+A=i5ibL%K~>NV+P1u@K+26C0o~*8p`;dSDM6pd3^!q;OiUv*i22K
z0Eg88oAGq_d&>E*uT$3bRdf@JTTxo|Zj@Mo$B!dCj%zoBabw_U4C}?p8eVq!x{$Pb
zqb|~<OrljaJhD`UMSv0NZYJa%$?I|}r|Z{f;&^Z^^c!^;?2kKC2$18UNS<c_RoU_9
z6|mlFA+5W+ds6tS$r)^#1~2$5oZ#sB6S{2tX@UJSU#ontnthWRG@JYmfMPGtzwx03
z82wI42FHg>;n>R$<Auyz<{ckwADSu5%aH%TU4fmKfxxj#E(}idq~-QBgTt-*SSdK*
z{&)<4e~Zp7%pXY&yYkD~=#z&;jazD>8P1V@P7KJExfzkvMW{LnE9`44$)jT6xhIl~
zV0zf-%eXhC9q^xj6MdIRMn=X`3`tZ}DOV+(`vEiH+uoQpzD!RzwD<fbK~+8nKGiR%
z?N3_JZpG%|XAMEU8@t6U-DVv)ZhV|5zcn%T@!O(z29U>mEes_dv=0i}bAlMp;8u!Z
zYM{Dlcw?9qYvSx^pey9?Q)`qR&C_AD2ZA%PPB@$=l}#?bkXIKm+%dtVHg0g}*O94y
zQ#k%9K<mn-L5q1E*VtZmrcc_JMoi<d#?{5*h{yA#s2+N+gneS^uQt^5rXMd9u;q@E
zNV%h#j(3Dh$+_yQP5FU^y#R)1Jxo=c&~-qpvOb((UqV6IacNTx<F93w9uT2xcX=5O
z?hzTZ%*%P5MogDbkpaPr;8S-;$AxuV#f28eB4?fcz8|p^{>Xxq&pi4bco}j)2}M*O
zS<s%ZY{8#zBQ-xkijyk&qMz|;F!(phEYJ}|p7l)*mv4~<9BL6KPy5P=a*h~{W1e(E
z2AcZUISbzvfs5=KZ2n|p;5W|i#u~E}a%9SL)QSEls`Ca#zD9%41tIm8`eggJG3Q5{
z+aWRjaK?iN30C|9;=fiJ%3u22Xy!2FTif_pjg2i-jcn(Qac{c0v0eRH{2g6u%1G?;
z%Mfnda2N!otewn`HpL0fLVefzfQwhvuV;%eH@?`!i<l$i5S{cl2I4>5`Z7q-F8N;I
zoQ28zk#6gs7`^V$0xU!fnx0}~dk;*L!%AipSAHleQnmK0@NtBh<B;?hFVDeU$ak2$
z0fZAGUtNPJ8{?VLUd0JomV0!Uv&}BqL)O1&p<Y(v*Bm~u^-TaZ?9j5x#M@9_r03(#
zH?e1CZt{j_D63Uewt?MFycPQcx`?oKON=@h#Nynhvj{67O@EHi^oLJTpUXzK-T$W!
zfi*^ru<p449UN;MuP!bo7~c{L3q$WF>QQ2tpUgjWBnuC_#6&{}@U64>)*4I-a_5af
z@{*H_16NH~-VU)<8XxjHM*fKMj^5}4J*p*-FmwNX1PR_m<U?dB(9O~g{j$>Bc?nuj
zSf_(klFpu&R5SiHM*TNChIgBUBxhqzzEN8i()F4P=M?yy!!a`XjIuVv0xF?_&sp#s
zvH7*PP>uSxx1fYcODVW+SbWRqP8GgQ$K40ZOIKRI4Lm+goBGX&97bkezq+v#;NR7<
z=piI9Ufrv#x4Ne_7D@;rgK1XQ#{H-DQ}%b{bTl^p7+91(C^BvLs^=lWBmJe4D!kEc
zNh?UFhclOPhr<!AEKZjJp{Soy0}}pQH+T?!)T2Rl?YnP^pUDjP5UA1F>c}P4j0~P{
zf0I4){@5T#P9L%Q&O4o3j~+WV_n=QG=oAz2%jqcbrg(NNXIn?#9Bs!Jonf?o?m)k@
zLu0TyCYP(s3+6tqqP1qt9dwa&|Ggg9&{i#p^q-qQ0U~Y+B;v;JZ0S1<o~jz(HCOJ;
zhk+0E)UFp;9MXzvTImZaL~FM_S$*+tYWqT&H~qT6&+CTJSh*Ec$bLLtvLY2!$VRU2
zCi*aOk3xGoe@L~@e!;+N6qJ-QNQ1|ufuQsGJr(IUJ)8R-E%0-PNo!6)Yy#OXJo6i9
zM7z~ktxDg>3~}}*mYHZ%8NWXLETP)J1VKn>xMpDH$89)X6z%1ffB6^j-f@_!vmpF3
zOx!o#p*nSSM^RsZ{99<TaGbnK>57k9aAbQx1`^It^Z=*G0parGuF|)LB7+~{*e*h@
zsShMELL3(Ip9XreIY*&v+jXJ6y}d(ikJ?`fIcOF!jFa6$Hq^8$+5D$!05Q>E2iPwS
zwr6*TH<=0+Ys6&js_Z$1(@iXXC@<OdyA%_{n%`7hRK!FB6&4hWd%9%ZWV`gCatK$x
z*y~g5G^I&^{BNcpnneLT#X@Esm*MzMMODSrnto(5#{^`%b|Ht@{CEG|a7rY|k3`^h
z(|fII+z#$A34=_dMQ0XU{tMGR45G8B2#cVh`xuFLC+PUDIN^%psdNCxY3#wGnlP-8
z)D2J;!R>rz@js7INm46XnW@%m`Tjk$a%C{kURG8%3x?u^)G*W-?;qUkpG0dytyV<7
zDHlgRG&gZo`AU;soZ!*(@zPxt%$5rB+%xB5OCILBx@vdLlC-q6So-QQv_ySe>3_0s
zE5G=F5f?4Cyp;*~<l0-k*y%`pa?rq|QJwTOg$(KuE5>b=D`_eH5}j^QKWmITgjJW_
z%55Y!Yl<+WwvuUp8&LqtS_RAg@@4_~*UabUioPVBmp#Nq0}&~4CiHhXTm2Owuod-y
zyvAOjc&3U+k47Z*oOp;j8&$LTm!j1>`8hdqZQ?`&>{RQ;n$MWe*~RY0N8<q2<CiAZ
zw(iH@%EdVE0^@hx;2BTO$?tNk>4gpf^<WM!Djgl2PEssFq92m~jS~LB$>CTSkrpp(
z_%XAY#oGpFDux3GR!3lHSf<3T?5`>xAnTixdihn4d3Nc>24<{PP=UgVf?Nk4+>8DP
z+vgRM;-UHv*;COV)U=HMr|H^;FRG!>;#3aSH@}k7YAfJ>B5w%#w)O95G!}nKOs=+V
zKhmXSbK<qB2OW}!xv(%VH)|}oZS#eX0OYX9<D@bE)O+a|#Fs=M4?49ZeE5fijon0n
zg}mi<3(JpJvZ0k&VvV!6c(zB`GWg2DKjoZwH{@9MNIHxpC4S6#FuOVk6Hz-HU`~nq
z@y<&xe@!|?=9_?+-@BxS{}cgZFyw<5f|vJv^i;{y&ktJ`q!rzTp@rZx)78aAk^$;i
z`jM#VeO&h6!V;f<L1dR@70G+)9o~z4Ir<t*ux0Eup+U{k-VawF^Mg5ALY+3BJ=W}^
zmfJ$gvK130l>P<cDcSB=hrfsD3{dkN`E&7SjipNaa;zTMuJ@E3teUUiPh|GJtE01X
zo!O2-1%Rxnv><HR9Y#nt17m;)n~ew$?jio$uwB$kddt~*)fU$<9Fo%yf)E54^QU(z
zTo`m3@yo`SpkZC^5O&jj<%WSyTJfcLZpqsNZnC4imgA}E+V_#u4@Lq<m`Wbx(O>?W
zDT3|(wOa<j=#AUA_W!XG0=W!#a9?I6kSweFlA!qo85p{@T$}sR-Dhtt=Vn5gq=9*<
z{ULk;|Jev2MD;D2SI(-rW9XYJ^YsRUgi6FMHV~MEq_yKmeyU4e7`(%=nQsQTyCSQv
zl$AEKWx%LyM?V-j<6s%4bYf{JHWp^T%`$(6PHbBaKp0O1Z){om-F@Oh4sA#tpnM{y
zT_f*nMAPjlA;ED-=|45KvYITv^;ZZ#)wxu3Tgg8fZDp$#NeUUCN}IqdF1NAh4u9t0
zJ9!MJ!f|2mF$8NV<IRXqELS{(7$8*p^<9oaQ~BXb6wF_5iXqLp6nFKOlhl)Q249%O
z$7z?i+Kya~T9m%>{{+7gQBsRSJk%iy@bLO;nrlWsXZjm5BY&h6f3l^p-9)EoCy;R7
zj<96Bt#f(;AL55Ht<8yVy09mbt>cNffQO#pvHe^zFHqIn=g&$U2-2+f6QN!;0K9!6
zxkuB*zvd)ZOfiePonAKOv%6R$WMc>X%zuXGCLHi#tP_Q1PYo82_A1@90qT7bINz`0
zKDdv9ua!UJ^BZA2kP)$S|02G0Shb-yCqFxCfK5(3=hgE{eM-=B|CRG$*@Hsdm(l8e
zTXx$B-b!#kWJ7dahNEiedKotJ$Y&(xWz>*X#M_i=h&!#soXyw^e{@s@C7qW!ltgn`
zlhVq1a&mJ03U^=#u7T$_2hNV}{>exs*kU^%WG8vlkfU_pk9D!Me$112wga1qO@uUK
zQxl-j4-diFE~+nSIDBNDy_|1hXJ4DnG);@~B0eD(Wb(aa!i!nF)Nuh8-L>OVnsM?l
z(a=8i-gU%2M~i`SskITt@8pE$9H?jXwLo)Xjk=_$D1*+8?}q8s*SAt{CUQOMz@0SR
zX}q{$wh<ZS=St1^n0ICww>EpM-@NA@YK6t&kk9DEj0_w`5}XdVb4cAszp5&=i$pm;
z`!y;3S0W4MyHxeh_%)?!FSPS}OnY<0^JKQX&KrvHmV{NEj#d^8xbhf|#4P@N;L3c-
zW_pcJ=~Bahk{(Mh*Z1D_?}tUCXe11m)94gUUe=JM+()oxr8>_#=pNcUL%c(!s29g=
z#jjyC66P3$-|;q=`mo)EwJ?;y{O-%6WXEKngSfWMODykbY_`Xfg`!FK)TpnI{-t7I
zd@B7_FLpt~&KDSQhBPiC293_P*2y>iSfA9}_}51%oGT6%VX;uQHuQ0x<bPYMO5g<|
zq8^dvZ`HH{ws;ZZybHYi@YSc6cwQc0K`@b&-KT}PsT|9>eU#Mno;A$Hx1%tT^TQ@_
zR#w&t&}Ds{40#C)1+stWDs3M&g;NXT+Rfi&OmB7we<gXMhY`Oh;u{g3hntWhWXr?G
zh{>LlZ_c(mM2(lWfwN)7Pb_I34V|V@J(k>G^2#9;-^`ncna308*N<pv$VTOmw&0r1
zZld2Hk8W~l3>;Ttex%r%ml!D&p-KbT#3cPBaB|mq{)&VB^RY(20p<xf)Fw0`AT=4+
z!5*QG%W+}e>Z7R(3dOl5!pV-7sMs@s@Al9gxjZ$FB%6HbUmW*1cNpskV7@{pP`u@N
zRxI@uV91I&9^Sv{7??4n@2)*N5Wm&D-T>oXE<VX0CNBN>9zUC=8+fm+_Imc00>qx4
zY)AD^xMTY_Sr>LvP)IOf3)NqtP#q8!?~l#U9lHpv6TkAcSU(v}P6VZB;)(?P1%BoS
zt2vkccqcwUxJ`8TUR_g}#KC2{3$CRR3$h;cmRs%B6Cs^dkOMvLg~@8%!>@xIrGqyX
zc>Qr<gd_ROZ4Gx($vYB&&!~JxREU;(%q!=WuC<{tiFX5SJ`rC`zEdL^(hIpznK<+A
zZyn-3RP_eYyB~gpPezXtt+ZK2WQP)*{yfM@OKi0GqN6W@>medHbF}vf)9g8N=G^hO
z!y#FKNH$cAX7r)Fb22w4C#N$w?=vfIbsym|o8@hTwa(=sStmV~VZ2LGi4;~=7>@~h
z0nYaobRl0nTz*XbEs@O9yo`H9kn+XWyZ7!n-T{xIo^@-x--wsPt~f~%WBAuBTmV8`
z>eHFTBgFzB`zEi_+4kFOAxMC>!T#6humA_hNCtvgGu86B+h_aLZ^EiVbbDnIGOMAe
zC^>nwHeRKYyk5+mk`tc>-uUi6f1E?07ky*3_(L0<gD7ApuR2Skw&#h}lvN-E=G6XA
z*(KkUoxGNz_|IYM6D|5*^2S02)#goM(Fp+4Y_K+j82ECZyO>1_by4A4p@>UFm^Zmx
z%a6SZyy9VVu9wt*)IJD&Ly+jFnQ;Gh>~?tQt+E>UR{Qd2;_v#Z_v%CP)7;I@-UVp{
z=d|wYQbP**mn1#LUX@Rf!AO^z=n-TIcYDA&Ec^g^vs{{q)c%!1{`EezIb6JDY7V}w
z1hW&Nba&aZ^T^408NUK!*ZEQ9Q?o?s&?-YR2~B_U{}CqrPEQ>qJzh4zio43_7r>mx
zuKk|5V)-u^uYZadXEc<O!q7s{y`|swr#Hg<VVw0%P3opSQ}`-N5|X1X6Fi5_+5tvB
ztQlpX`n&cE1(d3Uojm(h+<REg!$k#tjaqNZu}CX4T8m{kn2ypJhgf$_;@!-zSogQD
zCCD)}h0fd7w%6>%{eM`T3CkTmNMEAHX<&)d0ak=r0QFbkP2r=hDGkYuW|Pn^J!Ir#
z)iy?+!VvbTMJ}`XB6pA^(|V5QJf|!f)wv`9L38Qgw%Vc3K*f)-%2eR*Tz2j>B-mNM
z))ML~T8PeV4rug@96l7FB$?rpJ545;#dE@=ryCE3gnmr#;qmbzzW-6FTY^#_#SM`H
zC7^78d0t+$R4bh(<(+NSqLL@>?OdJbfE{{M2>|7{e+N1K;voFxDdzyoI(!ksci?fp
zMw|ALopb;*B&6>KQ>5Mq23znDFP((h6fO&Q5>k%n<lAc_Ib2?<7jN__ag@cx!QBMt
zZk@#<%aS1y(_*t=`t71v0H2X-Y?Q=`&nU$yfJ|qlR&>vT=k1kamBeJTNGkh~Pg#Cp
z9!lM#a`7thL+T?O{+D0~jUG%<%XgYV3VE0?c$wLTb;Gh2tbr*_H_h`4tqQ{TfV0$_
zfs|)!wYlM$!V<BqpHwW{cT3o`ej4lJ`O<`KV-5tA>&oPR6uN%XwistfPD%1ha^UN>
z(va1c*jIP6HAk(<@UW^TI~y|VMAQL*-Jbse5Fn@ug37BwX7e~ti{bGoocW+Y9i+C&
zIYfz-@9FikJ<@(dZ|=2NBzqwm<?Xs~)=THh^M{U*z%-Hlk4*aS-~1E#W+^Ew>JK6M
zA&B?ZSua~J4~*!KWIoaV2X#%@9%(Rhb4^lv$sQc{fNVbtRk2OE-SieeF}Go+iv1b4
z!6#rb1(T4k0T%k-XyMJ-aYW!CeyW|YPxl+ceF$bJf)$&P&R@U`^~v5y#NwF%OR|%_
zzP;TYsd?h0^!JMay6Q8bY%&kH<_&ZYs^Jxd6ciGx>a<=G>UYY(LqyZtRIT!hXI9Yn
zKM*6$VkY(iE0p(h!&}CQja2Y)?ytwgIiie$q2~&j9%r{_%FfkN7v|653Bm;FCj>B0
zaKvB8B7zK)h}Kgg?IOO2W1LiNTP^l9%!Y}0J<jCIxUTQ`wf!Pl(DVlw3YPSD1p99k
z*@~KllnYdYWC%>KazX!HZ<;&#!!TkwVcAcItu<&UI$!0ri1{6S7G{TeWeaF44UJ~K
z{%5D0iFO<mm`zALP4n^`awnWfmFRhh9wF%X_hmx!551#Z=93-xSRjB77<QP}X>4Ur
zD=cDfS@>2iak`5h#dD9{bGrZPIuirKV$IcE_zvDbzWZPLw_X_((!LFah)vKZj_$Qy
zKHDn)CFt?>&#d(g>JNe`#wZhxiHo~A0;M7PSJOTjfiqg@Ll9v-$Oy;IaiE6qLVip8
zlYhMdyr+DmOK)MR0-rjncnabn%zrs)pyI?f(BQ2JwL#S*;_arykZODOu;s)0p(XVs
zwHdK3#7>f&o4ShsbuFyC0<%l%pT)2ofV|OvJK_-#!U337C<kn1e`N6M*RPMI+QNLV
z7(7<Q;<#S{j=Oj#HiFm#1K7@thn`alz<x9EYMSG0i5)&({KalKjT<KeV*d5WowInM
zC`6jZCmAJYm1b3x)JJodpQqW)D;f*CI_E$|LwxVE1n9TqLG0`QO$6+FD3G3yt53EQ
z>VPY9`gJcItd<)c-^!iVcav|032n{KzEcg2KlQ!qy0AS^2SEk@k1FpjsE-Z-xovot
zlE{M%!0u^)o$WFCsR2HXom-5pWi`j~t47!!brsWjd-?d!B)#iylqF*qeA!v3)*mIa
z3#4*kMGP$u%3G_-0t*Z{rwqB_Q*NqwZMt#PB%}{JTmJA-ywHz=7gCph3zJ%w8M#k9
zm*b&&TM;h<dvpmlbo}Av<u$nrOA~pB>I_V?`>-nM2ZXPUJcGDSYCJvS&@7gPy`2tR
zmr1zr5?@IU9V;+l3f~~*uZp-L+TRow)INnbb&BQ@SCrK7+)cQp31b1*UH>n2^~<(`
zuK;r*I1}Fyh(`kMnMa0f^4r!dHkV7u1A<EPM0EC(K1f}ZO;Q-@Nc#VyHvfo}A1^Cy
ztTCsghUmWI6|`K44V~(&A$<~72JO2A<suqN*P9kU*1pBL-_R%S`H^|sh~1Xg#ckXK
zy+crXgwNcd0MWlusW;K^xheSr7T>E+6zrXB+jn4A)jcP(;3jW{+jSE`(P{usQy|P5
zuxDz(`@~qUNNO&L%Dr$^c(6ZI!*mWTra3x9V<${hLP3nJc)Q^5-=M;dbcgcZ$1g6x
ztdKqN6}1WBn@H7eZsmwE&apf$WT-@liL$fZrChTKc(>o51@qo>Zx&tItA`%O^OnzU
zfI3~<?}M#Rz%id*0twf2W^|c)^Pnnu--+mVwrm+$o=RVvHOU&Y91IK$jQ>}HKt;O4
zNfj40gPLK&^}tkydagX(#$@sW!HVyTdXZ)}6urGV2-@b`TMG&b)&*hx>?YXlzx2~4
zviKhaB`cJ5Sp`?nCwk3`ZDBMnxOLY^&uqv<3Pw9s4mH1-F!@;VvBn$06I_;4QmQM*
z0%s~>z=74gA}og7Ldg{ny#a3a$iEw06%K%cjs%}Qjbj3|E|RKStMc%Sv6Y@RK6*x-
zxJABkr~=deZq9L{?dOO8QGFypp1*@e-3yykfKyWP0>1XS4&$^*wia*u@!duFyRYp%
z9EFladV|Irm2G;<VXAl}Vn_pM?t69+M^&63FviaCXQ?Lz4@8EO5sT@64WUKIol61O
z4Q`um4E(-*ZQ8t+k~rlNU5_1QLl-qIdq>7ZcN!ecG(NMADVb0C&^4l#aeHE7B4_*G
z!JIgy5q4OgU&HQ-58ls#CIo)zz1jl9_!E7nzij=HxkzI9Qe65iWOfMivkQ8<f^vQ!
zcq>Gj(7Q(eCGgb+KpJEKF$nm#IAqXQQ1u}$+*kfm9XR4Zvq^aSuVxb}lbh(IEH@SU
zg%@A9RC(Rab(c3>J+*7`Zpl^%lfE>3UXvgZw;kf|rI=V?TDlLR%bybQ`=4kXgw!n0
zlUl+EXnaOGbT9{$x_#@wi2R_~qc3(yD>7|1afdoSGLkCaDYqHRiTz%KOVMGCrTxgX
zuv<G8^K|O(Xkd{R@EMTbI1c%Z&ybGBeiTd;Ml9cc-PL<`qS8ZSFwNr|^^Z?XJtmKd
zkp`_9q;W$M0R!zIYH4ld;W~F_2lk&vCxL-g=oiK!e&f!8<&gG1Nc-FM!E})0xrs(x
ztav%6QHT~_3Ml5&swMflZz$d!Z9t*7X5eZ+@<~FNXa`qH$pw;y`lvhP^d1t}Zre~O
zrm}vi=D_X=e1T-6Wbj)W@DzfO(6c$-{%Ra<U<7HFI&eekkK+W4MOO$azCUrKXHDMc
zr~A>jc{8_u`gDznlVdw_U|ev}<-H6H-{}6@;rfSMM&}`S*W5frxI0ba9UW!v>fu_u
z)RP<Ev`(bz>X;Yi_K>9<QQGc3F*KB$2E!e1{jFH@r-l#X2KuWyQxN(ep6YZN>$HVD
zn<mMW6=S4g$eUfOn9SZ9cZiuO`Kp&IV3bI_ugBf^?ZSU&T;MXa7SVMgxoWuzeCrKo
z3fj3$zVqU@o_)%T5!$hhuJa=LENA-E^~*s1-mI;yts})$e<qBXi%O*$`*>)T`fKFG
z2;k8UNPy2sAv-}>f7IfOwHq7`k_6K<_`Kur{B?+ICB=JrYD9!4X&|>rC3fUM)Kd@V
z*pGsWYed0w6A!lMC!D!6ei_zs9sVm)g>X{}bTEgivg9s4zyBK9zH&V!@CZ(}Fwe-J
zD7k((X{Y;-YE`m&@_6r_25E)8ZP8d99V8xa+<03AiO1kzZ(0r9B%9v_knkepsQwP1
zv@+k`n!&%#_MkGikD>#g<m`g&#S7-%qBCh0sS)+pzf&XdK=!sTUuIwZj|w;#QNYwj
zfrJZtj~k$6t!8;h&gVtn@`7SiLGkDNj~_jj#i*$H$lQB>pe+REwQ~9WNPe&dQx5sr
z83yWI$Uld~%9Tmm+!t~LN0abO>`~ASFv)|j<zX^z%q&u*Gqul#j;SjV$Y=|($U_jq
zR+HS2)d4kM<`^OuKgG9L&-u4eDt}znv9mzry8?`T>u($<kJ{=iT)jEj{(XVZG1o1H
zQ~I-t_0D+VmV_h=UarHu<V5EW^kLq=iV%IZ@1%pdyh(T#W&J$!bK-xZPmL`?W47q=
z&+Hf<8q3`p3M$_?1M<zfSR6)3*vcM8(s9WhVtd%uZ<}Z5%2%mb%jY=oX{>OfZJ9!U
zvPPnKcG2B9*Z7#-0Ab>6v<QmrWLXhU7VCL-B}w`Nqc3*{k@IF;sqM&-6chWId`HIr
z%0c4ylf^YwmNbGk=Mf(ookW!47$0U?3HBrUE2^nbMX9*&H-(L1M+6^ZLugVF*om2=
zSx7olE-CQnElHbkn}z4-#jKUD-z~K-7FVFE8yjop1_lS$IfeeOXj%ZWWE3AY`El9=
z{fW4cv+5_&UM8IH0k~cF&e_a`5dP6mLTgQd4ZBW$$hFIP{q-qBvb0E@PSu44SUad>
zfH>OGoT!d%TvZovYXh~&E>CI=bN*o!u-t)pOf_Y+aPC6jDZITjq&}x)!z^V#rWeFD
zSk&h-V?Vy?nv|$0Mr5gd*_(UsdiI_*qQ$6>HiZ`{cWV;dhRwqATWlqBhUHpyxe4o4
zLhy)*AsZ_m-KJ<`bu{VfwVAmOiGjsUht6KCBo3xm7a|@Q7?6Cn7iL!dm(TU<`TEg7
zhh0y!jd54D1v5|7v*;_|OQNx6=-SZlo$n;(GM*Gt9g_2?mHu*Cn&I}kPoLJ(Yrqqp
z;qvdZyDWn<6E58PMWbAaACXhAlmpH8EBQVr1XP5heSpE%?xGVecu$YiV8@u=&wdg?
zFBuV*38SITMa%K>{G@ctSHwo;>1Xd5@fou_Pj$Wf5{lHwVRF|3`r)x~TO`oO7@?X9
zj0Dp*$5r}JKNFqrbT}-abQeFsW=^lfJ7#U)C>~ABlEHy+S>>5>PaCBma^2y+q@l=L
ziJV1zM4!L3_C88ihOuI+mQI0AnckLags=RU3i$0xhrA%S=#Fk=#ou(a28tY3>IN2>
zC3B9mo)=uN814&>={xM#L4HI-Lug+_RMh$vqrXnM#S>tIWEc#%-ICU}2M7$(LVtf!
z)W&oLw&^u^_&k14BdS;6%)}q*x3)x19~Q!M(0cj37W@0QwzeokgoXd#<Y&l*uiwVe
ze30<<LYNTCr6eWk^J=3h>Nc^(Kih(GXhcqiI`Cw8j@YV2<Oej?EWLmKJ}-r(AEp*i
ztp!L2G{5S1yV9|(5-(?r#3_zGj0f6tIvh&bkUHiE`iubzPs&HXe@;czO_73G5e^w_
z`3+A0jon&Sw9MVsH2!{Ku2pqr1WEU{;>B*&YwKANmyh<-h-%eJ78R)^wK12!qIlo$
z_ON2p!l{jG`}dCuq+BEx#qxn5x^XCStCe`AlVz(TcF&h8h7F>OiGi6wzZgktH!qC0
zF5F2anfPy#+a98#bbTGd!(n%o0Y|YCYtS-`X~i7`4QD}CI&I#}5#qZ#`{ow;{<#|Y
zhJ~nNwJxIgrg$`U`x}4vTg%_Reap!oJLmBK_BFO(B$4z4$K_fAAGq*lOnuZ_@j*rO
zFin;XcB^JbeV3PTs5)qEU?~x{H{FoU)yULz__VseEebq9WiA7Ce6_C?Xwxbo{#V<!
z1+yE<cJ4$Rwf=bfZ@VKdF;}V0ex<|Kh*`Q@N$?|w65Hev^R<~zbf5{0wks$otZMy_
z%={cl(H?U-KoHmy^e_0<hOFp@XciQ=Qc5gcIo4>(Hr>DbvhL;T3pZR`mgEtk@6%(c
zT&MEE4Y0E31r-yfjP1A)I0PzYcPQtF-@910b3k&lE9D1)sfM&zJjTPDIVe-xzBruu
zV8t?FPwMHw=klRNCNaA-L^_(fIcD;^BbQM7VOd}C;Qvc$3J_sLeaHJ5Zx!(2Z)k<g
z>Ee@Ov7?9ZWJJqjMyiV9#miRvgevQDZeP4?hSB~oFhBjOi6oNk{7mZ9$0?Vj=d4Q6
zZHRSQSgA%LRm0sG9YKZ({rI}dzj+UXB6rFxC<){d#=Rr@DavLgzdFdPyhEzzyMz5L
z*Kbe7Z_KET^B#`O>|VI2pO+u@eF=0x`v0l}kQU7|{muCLHd@~b<y}7U^ZI_-eXhw8
zF4;{@G2zUv9&;ZpO1<_V2H?gUtIoISSR?s|NXMc_rP@Qvcke@H4p#AYE!@Z()FW&u
z!lXgUXcA9AM2Ej42L|(F7L^=H%*~-#SgdYn-xqFikNikziJGBMzUd7Kp;1@}=P3d<
zAG4c^)U)yZy)zryy&Y_F{;Vn%W-cRJ5+-uO&JaI0Vna!nbU_;)^Rw(of(4SGeUVnU
zo+GnJvyi*zk*p$HasBziZ?<Gd_*102J5N_@?=RnVcU{MHZyb*2{!I*=ZV@<(nd;A}
zqAF;{Z5jiECjv%v(1?2{K4;HTA6n9=+D7kJl%A;-OMjrFVxN?A{`{F(5!=H|PfBbZ
zgfo*vLqnB}|3@{QkB}$|wT+<D$8@5#iUy^8r?@68*gUu%TDtb!d2^aLNDZhT9Qw1T
zr{@gTWD@@V-%f!u4|ic+({g=c$ri!|xzgX)TLAvrR^CytKsTc|+hBZr-GjGsUDU@~
z6t&DQ5>qzplk&-Yc6RnH`{Veyzl<^3xPAN&fDKOXfAo%u@Cu@Qs&B<J6P+H@@9s%P
zur{2fukCv?Qu$?C;-bMlS4$3)if!hDbMtcXGohls=4Y5*PLR#1l<=2Hm_(j%;9jcI
zc>B7R?|eWD=uo4gyt(Ng%O^b(LPncMrhk<Gh`LzhvFfO}YNM6cp|sj!c{H`^#>Yvw
zXY;RSwcrn%QpP_%U0RKJ<@4i9!sqht8^V(=*Ik}pT6NJ(y)kZX4o|r8d8u37YSxUI
z;;2a0fq!y*dd<e?NAyxPm>sV%|2RI{eELF>pM}Ts!nOV9(yjKHOBCCSpKMK%N?Qor
z9Z1()o}cPhTU(oIGk!f&GV<e_@a({kw_DTPukhrn(?u4vE?JGmMlAhUjF9>wdAe0&
zu&_f`f}Phqx7gwlMRV&*J-_tzo$2InRz{j{tX)`6FCRaWEM-X>7~I%<<BeTU=>w~K
z&aP}eHq|uW!jfywZPf-p$lTm{b>Z4#j?<5Y3#&FeE)^GRYp!aMCQGf4n15-E()=-M
z(cJ8{)T?Zj?6K;xx^VVP4aK!qG2g6p>tw&-5r*)u#ailhiAUMLcB-dK`et>1TvRT1
zflZtRjS@NE$JNyrQr0hAOL1(DUol~`{*l+njo$ljGYkImbr69Ni;w;IsuGPNi?3d>
z&BXJC)m|R{wWvBDWm}t**C?1tl{yxd!#xXdCOtnREOc>hZq5oT3ei(2<GM&6$;EfO
zCTfIMt8c93w=2^6kx>nWmyC>zv_GF-Z_lATcEx4usLpT{Y71%K=@OR^Z&LyM2r_+p
z;zB*Kd^_VloM}ey0@FNlFd>1!dLO8P^CCOR+|zd@uY3G(?S9|iEN+*kmuj9viptSw
z7X0!wJbc;ZmrX0#LiGN@M2Fvjo0)vbL-cduzPp!3@nFMW0;Wu~mH7qDRl0t>+x0}3
zMWW1oroM`whN7pNp4~IoM65E=5p~bFDK_sxZOV2%%_~>Lqi8n-NPrZhl{&Ec?G$<x
zn^m|A4)s3n;!gyxV#7VWLp6nwJD6}0x1j{oD5n;KXWgLJhZoO!m~w=SIg}bFEI1S|
zh!F=TCx<@p^YdGG*<XI%D0&OtO2L^vzPH%YK7{K!+IgwOMMTXbf?$~mcLi0ru5Cf5
zFT}DE{^TM}@!F4lLECsH`5sCKunKV~ainv}NzM!vflA}eYyg$UZ}dy0?L5a)`0<Cw
z*v%9N>=dC7donsG>U8TiPJ;h|i6LD?$8!pCVY`*!O3_r8#pg@t-}{Z}6d#WMv0^so
zZmhc(gJJAZhA)3UmNQZ9@v_slns?VO!<#4A>SZw*nx%SdL$T=za9j0CxQjfFNp-Y1
zjdW#0P@m!$q3}>RU5LNox3{Ckh}@!8e4T^Ya2%~n4Mslw!hZ4j*~9*p&xAk9My9Sh
zsXv;T+0iEM$rYAjFfr{ez7k&1L9C}M6N<mIZFCIeIa?i*a4QG|thbJ~msC37gAf||
z5?7M4Fs1K!HS9_=6=ULO{gMzJ&2TI0{7hQk|MB=txC^u7f%b5XYrTwbOB(Jp)||Sa
zsOYgN=GDK6I=(_drH!2hCvXhjQ$OVJ2W$#ALn~D70YUMnflK;;XAg;bN06OgFh}k^
zsoKKH$6qR+KO59{Izw|g&B*iKH4l%qUH3%K>JS*glIlgH8Lu8mr}8^oSE#@(VE*0D
zj-A0b8x!~fs)@+hnr904y9qCJIjF4TIZA)#B)w}xlR%5p{Pj7JPln9(5|5q_3z4Aq
zT+1FB_$eQHSj=s4YR@KHwT*^UuP<e1#z`IQ?O>jhUY@-BK1~(Rx*Mhi{TvKC|5tlC
zLV>|rfko1$r)uu=D@Wg#<%f;UtoAm(+UK}VjD*}DU)M>xbNcrBtoh08BD#CXe`mZ#
z1maGkn^D)gn;Z`Jo0hm}MCJu^!Ll{rQU~$pO@+Rd*irPG72QKVkcgPNj1}C;F`swT
z6nC$!GfG$T#BQfj{Q0oh{e<)63!hde5)SLEF!#dsv9U2(5|Kj^1X{Q3Ax(EEVs2#$
z!nR*ATm=e}Dn&lW-?=l@RL3bMqjwDW7d$6-x_E4kOMoD{XiuW|!;Rs+_a@fz*0MU9
z)fc3|b8c>CIwtuURozXrkfQyqYmGzmC~lh$T$dLeX?jBu^GOr?Rpc8=N<b7gMSL3`
zT(mVxQkkpmi5wQ01+SlaS+Z6jxfXvBLZu}7*kU`I)yDm8hwC{7L~Phj9COW^{+zb3
zi%Hzcr5Qy}P*NXgU+m0rEyA7#g8|xYf1iTeVue?W2!Eh$zz^h~XsaSjv80Mm%+;V|
z@|30I>M^=@!Z|;<eERslw9##JCOALf=Ukpt*YjaQbA>;bKkI{qYeGTT`4lBD$S+UI
zxNrACzz+YCZUHyY$vZcNp7k1-vY_Gx{cj3uDtae}FZK?TL~}$bbbOB;IHn4_2|!<J
zg6(!cj&(TcNcDQo$v7?Eh5Ru_h5UANfnt`$>|XsM^F4A11l*Pr;*skrDb5kjusvqu
zCb2)NkkW(LfMzi6gU}C%dlpl}_mF6Kup1K<M0R;h|6{+%ALUmlQ~TYZ-&Az`Sr^ms
zTu)x5YscyEhd}GFtOSMX=#ey6ui~N{X{roZim+i~c3vzQG&oaA@Iw8WxAH%n5u%qN
z6ac|!zW>6p#yJ)z)|x}<@^wmxJ|^e~kA&F;p*wpnKYP2M_kyA1sU~_lx=*xx&SZEN
zy+Z}6>@@2kGVEv6gMn>E$!?yGg+Gxt@QbN8em^`I#w_BVoV%)RyE-FsRNJi3?tN-A
z6Z`n`(Ovl^4I^d=Wu>L1+q&a&S%s7dWYQ-^g@=a+9kfrzZ9WiC24O(wIpDhj391A#
zKXKsy-g%~uO)et;+LGq7Q*Zg2Z0^!~AGR}j)Ra#;J>&!8Y$4HBo}>5Uen~Q_n%q2H
z(!|!v%xr{Q;o%hmeRk@wxYwSXJH>$eh@s4zKho+-!Zsn?addow3^$;@@S29e*-2>8
z#{A<p*8PdIFJHZRbE8_X-*DMo_pzKr8f$*455+`aU|{DF;Xr(z6LW?9(H-qH5)*6e
zbw!ha=44_q+tlZa<5gu_V9_TlWYF7!`-@jljX@-Dso5Gr#f?@M<%OhjHLH6o9)WF?
z9cYRjp044j2#t@ACwFE@KT0)}iT*q|FO2P2cBi(Ev5?JZD)7xE0{LT+f;ajOz7K0Y
z8j7nf{bEpb10sM+JFD?6BP{hS{*gVB9zzp3!dDoZ+MSHcJ6&$fx}QvUz7WlqaYZM!
z(_+aEZt{!*IW2)KGY6F`%LN|cPBaHTM6-sj4_=KC(?emQYtIo0{GlX(Gh6Ja@dYI+
zn1;3%bjEYPqK)TqC9*q#MeIyq^&%3F?v;N_<ja?0NNGYy0N~$t2F!vexDr&W4%p&|
z7lh2e^1K9w0x>;A4z5H+CXWbSh)VcSHg0;Uud`<6bslbaHf<$?O`r|qN)ZPP7M2*<
zFw?no?XUS=y_O#zmVGNE=e!r+>IA9csWQdM@?fhLLolzH&k$_NiS$zw_=s=|q;pqF
zNVp|98d>nVhh-h*y}p`|e)mgO<frO$#qJ67XU+$NG%(0^5LxQv()bv!hg6Haw{wxO
zx+X%v50+43s(j0tun8U*O?G+mp=h%0A6c56$}Fzo6<bJRdGLpW2%5BXp<Tn9{+`Ff
zW%bt_Q9cvxE?tzE1E?JSl!;v1-Vd;7#;?t(krq`(Lz;MSNly7_qB_h%d4gtfJ6N~;
z%HmVzk2D2r{jcIVh(hN1R|5`&ZJQfsq~g4HbC+7x(YlRE-EAz^UEfxB-SPJ8R2Fzj
zZc`LKb+BWu@=N9Dh~Fw}hH96IgT-L{J|-mv1^1tKz4-`_n&Id=+qzQwU^k*{(34<`
zibNU@;YEQOGnE(Gtil#*Zi-KJr6L%U@mRtd9OcUx>o9de=dQGzT_@jV(#JGEb8-F_
zTkok$oWK*_@OjK-V9IyD*vy4Z{`(ZM+c}99F#!~EXG6ts#`TXb@bX_ggS?#v>7hEU
zl-@^hX;O*leBFD@5BBhD%l^ojYZx?>YUyOklB@zNJ2&q%TDaaWM5fo|B(O4BH4&di
zNX^-@>3`5mOZcFVJSSQSL{RdbS@6H#1;AE1D1}uEewiU8uf`4`@@HexpXkWN_$@xL
zu55uaMd*OfGbB}DS0f}ORQ&i?W(wlzsioGt#rtc#4k8ze_Jn&}{n7Xb3@~~p5|dpG
z?k9|MH#w+WG>VbhRmP!xdoGppd_dkkrAEt9Z4yQ%XG|;oYmr<an8dM}2XZ&j#;Ehg
z-B{2di!3}QM`~|V!8~@!D2o-G>yuGwSaWEI34s;y>>o{pyiz|Pe&dyFSk<zq;3Zim
zp;$im9-Mi8-Rr_93=QcnKW)2DmzZv_`)hD;aAe&M6+)r5!VA54D{<l4tibM%SW>yu
zW_0`r(&)wN$eZYQ#P0`hbe!Z|+%Jz;VJ1qD=%0Ufmf&aY;LTX~VP5s@e(|E1!9IW^
zDrxn@sfDEChTY)nMs?^WsY&kX7)$k0RF}bn9-dq&dGLnK6Lk)8BiN9XoIy<dA1S$^
z5HLln0gZh4^XlIw`mEZ<BVXQmwBq2sFNHmUiLnd2UMso#1cTYr8n~I5nCQvzt7gT*
zf$DZxFi^bv4=#Hgjojti1~Yua&o9PQW&|FQ7iy#?R3WnHM5pfhd7hMuNze8j(KOw!
zrRB6}|GL<i*X)Hp>b(8(>3OmVJ3G7e9Q_CKxNTfioEhlFfI%nBmq14B_vS)OIeYfk
zO9)~xgbW9>ByJ>_L7KBj@H$daRY$=ckyt0~IlB^n>PJG<#XXn9iuTtke$}EpDQ>K%
z1rcng;5kC|0QltU2R$~~4@YsMk?ZLpOxL?C-)XYcVK}^d3qc9seOqeC2<ygF1{2_f
zEb{uDqf*l*_kWtHU*WG}7SnNc<!@<_IZE6jhNS)7YVL)XJ@eGerXvlPc6-qob`HFC
zB3mAT<HwD>T*EW3{g+>cX#lT!)ublyB0)@NkRkbz#^v7#;=YaUelKUJ=uCnmMRe;b
zlIGrvmda<ii0KC!Ru~%_ugm|^DQI6J=rwWjy?YJUbsEi&&@jUcuckQt0GHJtY95Mz
z8!u5Irx_-e;8!clU{=%Dzg;^x$-f}9P+Qy5%H+-$He$@&!lZ+FHASCj)gDsxZWY!J
zNs8rbGUC&+$}(FyT7qqw`VnI^1IDOIWKi%=V}z(NU2ijjU02?MnJ(*fm(W?Qw3|NJ
ze0;|iwmWZ0+XRs~mQ~Y2ifOIXbe?Z9Mg0lha1(9k1O>hIpHXbH%UyOzK`<8i5_}+>
z^nrNXfjVB1fGXE+y!4|diRdQZu2*IhExzl>&Mk@$ObmE1t|5L1f{2EZk>p)rZQ-!Y
zL^<$K2;o_Iw4{3V-gYNNx8W)?64W5!q$)xi@4)}SP(u|=x2H0MpxvT_N>(rh9jz}5
z!(YQ!gpZ8SUFn*?Gv@UjAyY=z&d-PI)mX@K67K9ksLL{m>UCz;*ZpW*R9$_6fdp7)
zo)!G|o3&xhTZ5H<S6{~v^=0&ujez(bx}ZP5$Wgq{H)N&glbNBcUXpafwL0VS2b_r|
z2Jv+5kP;C3`QhF!sm=OX3M036*WfJn1kr=Q9Fc)$f25O@ksiis;m+3t0k>xra33O>
z9Di|o-#evZ@x2z;-|Xg$dxFXt*Y^4NUW7-`a@QRx$G4An#}br<zbr7tfQ|Yi?DSR_
zq&$-SVjeGEf@I&SFC}~r#yd3iYM>w!j$!WvYy1W@BxkNr%N94vKj6>}8!@zN;eLvw
zafD*OfB){cDUH*oT5`A8?UVOKtj)GUZ1#_@IDN87lSJDiH)mPJMo31$%`MeW>y4zy
zj1HC8L5sQ8i?(=Kh_JR?^;(YFR6giaW;sv2Z_QSe|1P#@Q-tXHCROu4I6)9_p}HS6
zLxk|UH+mc9PQirdQMnkU8_MyfL;ba)OA|bS3e~hvwDm1}<m|1D4)5vidgqlW2;eu)
zH)xfyIRhAC51V*PZLV-x-jJxPacQ<wp#VYE7XUK(Mx;6)2uvbqMZ)Q5kp0@fqW-NY
z2`YxDb?x<hlG?LMQBe;9SSO@aTR5XjZ1*MRXmAJB-P=`VPJBU64tD}H8PWSu1`e=_
zk%(2qyoPYKYHcTBq<y^)+KJbYUqj^P2wq{_**z8WVzYFu6VWNTO~hvit!fmEcUR74
zwjY1)fq_zglVB!9P|Mdh=_W<u0M==b>xzRPGZ03Ic7$_vXqw^Y`x1@>K8QLJcsL9o
zh&Jqf4k@#QO07+vNU=S$sM<~BD1xdbY7u+q4q2hPutrBh!V?@RMQdivdBL-Y=nQFr
z-!J@pJ3)_b(5#;hWV3CS-<6;tT2b;GOu1CDmBkhX?|n-eeO?XyY^*$>A;D^`&{Z<m
zZ_?RkL|5Jdo6kQP5pxj?QQu|YJpz&dT?eodiP*2$p&OfxN%S*>_v|_h31|5hSki*o
z!`z`-raU}6o7uhITcoa$-i1QgZ?EEo3vPnGy_*72mPFL~fNxVFU-FJ(SFsL7fSc$T
zto4Tn{sh^X{Q6U2Awi3|Qss^>-fMnO_rvqey_XGvgU0hJ3K0RMZErJzLx-UXM~`Ja
zOv&P6F2BnYEC;|ATJUP^%r>%<k|0tCYP^Nhw;g%Dae;W=F-6i`wQBMN7r1^O#>;V~
zG_gmtWhg6~L+{PIBGt^-rR5i?voqhVCWKoq8EIbV!@Dgssw^wQjQiFeie<k>&Vtzv
z26@}v0{lxbl*DG)TJaya-xH2=2N@>l|Am_X*+j1s^)$CTie;@uD!PakOEMK@Nq#?1
z`-+WPRIr#y!KYcvX#ZN_&*7=9Kko#e+Cg~CPh`W_x}Cvo<VE#GaG9=WcsCOa%W`tO
zV1$zr^=1g4Q$)7^+3JH%3Ss^?+RK%*idI(5FCW=dlP8CoDk&+gZ_3J}*NCq#R<G~h
zi*-Hh`9bk@D3jx#gVzplTouH0j0}emwmL&;bNaz@TB*&9?+tS)Loe3?wfILTwJglN
zmP6zo95%)AYP2${oBy+u)Ei|X?!Cw#eN!lL$q9e>heFzlia-+Adw5@MyjdGX(zy~?
z%*<4n;qbgHB2hQe;^NB}HK<<5cLqbglStNg|4&6kT-kP?d!NdhBo_$MCrtyPdw8Z9
z!GW3Kf&=O6rZ^waCnHz~!u81=hFddD(2614TE~vkY4_|GbVY2*jhW_q8tza;nHz}(
zWv?{3n_pyGE|z<{ThUXSKlw49X)!&dw<)ggDXWxW$38;e5wb25d>bK5%kHUW>z0eu
z_`&~APO-T1oZm~+M{wv749yv-`9aV9+>%W%XMp9kXnA<@N@m_`C6quVM;76Dyq=Lc
zQz9ZIdoA)2cJ*&Ql35*0B*11jLa^vLM10q{wh;a#!x3Zs@N7Wo`D)d+C>Q4&lLuaC
zI)N)@X5<S~E%iSA+HJr<v31{?%>U0XsE;V=7V%u=kKz3JrCsZM0H;e8*+fA@E*+84
zN0He0@71AnKqzPSUw*s(F(P)EYfoixl9dn6ibGkV;0CJ^en4;#`D-Gm4q2(qlSdjc
zIg?Elp^cS8pN$N24mFdWh2UJK`)8oTve#4yD$M}Yow6Wi>c>HB(t_b481*+&HvfM`
zTSGY|JyMHHd#*;!OpcEYR)!cm>drhZ2?oP%G3t?dGBYzX%1nKL;7jR$B?@4*i5Y*P
z5H^h%A2a~B3jUJ!1{r{S<}xMKE`o>?&9e&}m0VRC5iT!JA8&OYS-G$p5JhB3b(xYQ
z*SI|)A%Ust;YC6`$3Tmj3NbBYKFJ+u_zdB?3}axr$YC`?08C&QQmR}?vC}GTOrgX?
zlB#+3J<mK(vo6>2t#r{TOVc~XSohVtB92_*>v1QtcaKNC4G7Y-0Km&n1eElLCHnJE
z$`J2W)Q*dgm@cU&=KI9EvMlO*{QY^Ih9vJV4KlUWwa=C06iEQ+h{YVa9TyLIuff*#
zKRlkO>X3kL@Y!E*j(ShbNyU{qdE>R#T`SjzhG&(E#^&bNy!|Y9$$|+dk&=SV=U6-O
z9KDz`JwDa%SFEihaX!E)XSw_TYwz4+n!3UOZV)GP3uGA+){1UIhS>(JQwl=?2kK;O
z3_(F4B817K;IQDMl+ub*nM!$thXxAFDKrDy@+h=CYS9S<gKcbw1zNdlZ~_G?6|Z2s
z+wQnoYQUIff1mr;&3ALp`}dviob#P?f;V=?ewR==_&y%5Yz;}EMT?`9Omgc;1Cl|?
zf3(vS&x%2QCQY~K=n3wZrJOFA5#M*uQL$g#6f!%i@!%ki3W~UjT#KII3q0F;+i(du
z9veN8pOu=%B!fk8cR3Y}*$Uq9V{+onF^*rLDniLrR5A%1<XTAdQ4wt;-?M6zhiDGx
zb>P(_dok)l`ygYJ5*O)veZAtEH1Svf#yKu0R`htrrmqVYiapEu^f3x<il6Dr*`V1{
zz;2oW$MlSl1{p0)r+s2(>d@afIx>4+Fqlr1tW6k>VD#<(I3=E%dQaObo7Uy&V-9;3
z)ZW*B_(KC-h$J5solEtvV+(v+1!Y-{2fC>a)w|UZk#9=xa*&nGc{~F-JH)xfuST`|
zTAWN=0Sm|T>Zqfub>)nol=X6o=&o2t1*KNSpP;s*+#3{j6s7A=gd)D%PBzN)cW&dx
zK$$4Noyr4?^le!=F<37hXGrft97`f>1&LTXhrm3BQ+7m%Qb0ZfF;3glF#jhl`xt!`
zo}GHcZt6a<U|{uRg@dtjasfDA?!R=!8^dahwmj1}<MnjkTYF_wXsSA?c0Jyw{wf(w
zos>=~xV?FODV=3{G_>64*MSQcuhp=n7aIs_x;~pUMHW`mxvcU7hh1}6_ue8*O1Ef9
z&(NAltU)1v+?$$GsV0K+&2FL1R7^kiB#AnsaofgTcQUkZ_IYpO?DB`_jnR)TC`VPl
zzp!A=yEdYAX}YkI#d-Yh&j9(jt#eE%Z9Ml;7g_+A>d$e^EHe3(OjK{6Y^$hB#)cor
zWHRH*)wW$M6D@z)2xh6B3?K&Xk1+`C(Y_x?UhKvMllOw{-U@Ep9=xeI$-}#<StULF
zaM<HctVE?+A7Z4YkjXmQTtulw16XX-r4iWg0p{06Nj=L<Vc=RJYCzCrH=JR%DDSd?
zwdT2#<yS-kL8oi%_w9e%erXV|iY9bZt}ft^^FMw4^T=~mHmc*wA&){4Uti}^Fd7Or
zBJ{qOmwntVom~1@GVdNcFaQqh(qDLK-Ez5>u}Idq7HK^)%npdaBJ#?02}+c*zL-0g
zIW7@xp7}l_<MR&MvkXKpKSHq9cFjqKkDM|@6Pk-xnLCz$ZS+?b&Gt^Kgqb(aFNXh1
z&-lcO75Ma^9bfMKpwS_>F(~U8(2g6<)Yxfjulv;6QwAPB4U=?UUDV~~;<O~$EGY*4
znf0PL$HAtutfZ%>M}Uf)o$LeR5)&UfpV_^`?1}K&#5w_EZ1m-|i0pT4*usO3c)X!I
zR#PADo(qn|w!2-{l=dIeWv#IyV+A#fDYMqQ6}j|Jj}`jUb>PMaO#8LL`O_ogRVPBu
zhMT;0d9MmIk0@T%0L+bK=*r1=Rr5CN#lh|NajSUwW0H3d7U{$Lbyt-2&CCZ4M8U;F
zZ5Hir{&{)b2=~PO)f>#U#5hsY4SZ)ALU5<Ec=Di~Q|ol;Y-Br0eH#0O=eJHG0{#<)
zw(KO)i*cYOcjoGVlGktmKP_*9Py&&GTtEsS1&{(r0i*y@04abJKnfrQkOD{nqySO?
pDS#9}3LpiL0!YFCrl8h-!6cShZH@{^lCJ<y(9Vz@+<>@Ke*;LjI~f1~

diff --git a/docs/quickstart.md b/docs/quickstart.md
new file mode 100644
index 000000000..ec445dcba
--- /dev/null
+++ b/docs/quickstart.md
@@ -0,0 +1,228 @@
+---
+title: Quickstart
+---
+
+# Quickstart
+
+After [installing Outlines](installation.md), the fastest way to get to up to speed with the library is to get acquainted with its few core elements. We advise you to take a quick look at this page to see everything Outlines has to offer before diving in the [documentation](reference/index.md).
+
+
+## Core elements
+
+### Models
+
+The first step when writing a program with Outlines is to initialize a model. Weights will be loaded on the device at this step:
+
+```python
+import outlines
+
+model = outlines.models.transformers(
+    "microsoft/Phi-3-mini-4k-instruct",
+    device="cuda"  # optional device argument, default is cpu
+)
+```
+
+Outlines supports a wide variety of inference engines and model weight types. More details on different models can be found in the [Outlines Models](./reference/models/models.md) documentation page.
+
+### Generation
+
+Once the model is initialized you can build an `outlines.generate` generator. This generator can be called with a prompt directly.
+
+([Outlines Structured Generation Full Documentation](./reference/generation/generation.md))
+
+=== "Text"
+
+    ```python
+    generator = outlines.generate.text(model)
+
+    result = generator("Question: What's 2+2? Answer:", max_tokens=100)
+    print(result)
+    # The answer is 4
+
+	# Outlines also supports streaming output
+    stream = generator.stream("What's 2+2?", max_tokens=4)
+    for i in range(5):
+        token = next(stream)
+        print(repr(token))
+	# '2'
+	# '+'
+	# '2'
+	# ' equals'
+	# '4'
+    ```
+
+=== "Structured"
+
+    Along with typical language model generation behavior via, `outlines.generate.text`, Outlines supports structured generation, which guarantees the tokens generated by the model will follow a predefined structure. Structures can be defined by a regex pattern, JSON schema, python object type, or a Lark grammar defining a parsable language such as SQL or Python.
+
+	Example: using pydantic to enforce a JSON schema
+
+    ```python
+    from enum import Enum
+    from pydantic import BaseModel, constr, conint
+
+    class Character(BaseModel):
+        name: constr(max_length=10)
+        age: conint(gt=18, lt=99)
+        armor: (Enum('Armor', {'leather': 'leather', 'chainmail': 'chainmail', 'plate': 'plate'}))
+        strength: conint(gt=1, lt=100)
+
+    generator = outlines.generate.json(model, Character)
+
+    character = generator(
+        "Generate a new character for my awesome game: "
+        + "name, age (between 1 and 99), armor and strength. "
+        )
+    print(character)
+    # Character(name='Zara', age=25, armor=<Armor.leather: 'leather'>, strength=85)
+	```
+
+## [Deploy using vLLM and FastAPI](./reference/serve/vllm.md)
+
+Outlines can be deployed as a LLM service using [vLLM][vllm]{:target="_blank"} and [FastAPI][fastapi]{:target="_blank"}. The server supports asynchronous processing of incoming requests, and benefits from the performance of vLLM.
+
+First start the server:
+
+```python
+python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
+```
+
+Or you can start the server with Outlines' official Docker image:
+
+```bash
+docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
+```
+
+This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used.
+
+
+You can then query the model in shell by passing a prompt and a [JSON Schema][jsonschema]{:target="_blank"} specification for the structure of the output:
+
+```bash
+curl http://127.0.0.1:8000/generate \
+    -d '{
+        "prompt": "Question: What is a language model? Answer:",
+        "schema": {"type": "string"}
+        }'
+```
+
+Or use the [requests][requests]{:target="_blank"} library from another python program. You can read the [vLLM documentation][vllm]{:target="_blank"} for more details.
+
+## Utilities
+
+### [Prompt templates](./reference/prompting.md)
+
+Prompting can lead to messy code. Outlines' prompt functions are python functions that contain a template for the prompt in their docstring. We use a powerful templating language to allow you to loop over lists, dictionaries, add conditionals, etc. directly from the prompt. When called, a prompt function returns the rendered template:
+
+```python
+from outlines import Template
+
+few_shots = Template.from_string(
+    """{{ instructions }}
+
+    Examples
+    --------
+
+    {% for example in examples %}
+    Q: {{ example.question }}
+    A: {{ example.answer }}
+
+    {% endfor %}
+    Question
+    --------
+
+    Q: {{ question }}
+    A:
+    """
+)
+
+instructions = "Please answer the following question following the examples"
+examples = [
+    {"question": "2+2=?", "answer":4},
+    {"question": "3+3=?", "answer":6}
+]
+question = "4+4 = ?"
+
+prompt = few_shots(instructions, examples, question)
+print(prompt)
+# Please answer the following question following the examples
+
+# Examples
+# --------
+
+# Q: 2+2=?
+# A: 4
+
+# Q: 3+3=?
+# A: 6
+
+# Question
+# --------
+
+# Q: 4+4 = ?
+# A:
+```
+
+### Outlines functions
+
+Once you are done experimenting with a prompt and an output structure, it is useful to be able to encapsulate all of these in a single function that can be called from other parts of the program. This is what `outlines.Function` allows you to do:
+
+=== "function.py"
+
+    ```python
+    from pydantic import BaseModel
+
+    import outlines
+
+
+    tell_a_joke = outlines.Template.from_string(
+        """Tell me a joke about {{ topic }}."""
+    )
+
+    class Joke(BaseModel):
+        setup: str
+        punchline: str
+
+    generate_joke = outlines.Function(
+        tell_a_joke,
+        Joke,
+        "microsoft/Phi-3-mini-4k-instruct"
+    )
+    ```
+
+=== "Call a function"
+
+    ```python
+    from .function import generate_joke
+
+    response = generate_joke("baseball")
+
+    # haha
+    # Joke(setup='Why was the baseball in a bad mood?', punchline='Because it got hit around a lot.')
+    ```
+
+=== "Call a function stored on GitHub"
+
+    You can load a function that is stored on a repository on GitHub directly from Outlines. Say `Someone` stores a function in `joke.py` at the root of the `TheirRepo` repository:
+
+    ```python
+    import outlines
+
+    joke = outlines.Function.from_github("Someone/TheirRepo/joke")
+    response = joke("baseball")
+    ```
+    It make it easier for the community to collaborate on the infinite number of use cases enabled by these models!
+
+
+## Going further
+
+If you need more inspiration you can take a look at the [cookbook](cookbook/index.md) or watch [Remi Louf's AI Engineer World’s Fair Presentation on Outlines](https://www.youtube.com/live/R0X7mPagRiE?t=775s). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/dottxt-ai/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
+
+
+[pydantic]: https://docs.pydantic.dev/latest
+[jsonschema]: https://json-schema.org/
+[fastapi]: https://fastapi.tiangolo.com/
+[cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
+[ebnf]: https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form
+[requests]: https://requests.readthedocs.io/en/latest/
+[vllm]: https://docs.vllm.ai/en/latest/index.html
diff --git a/docs/guide/chat_templating.md b/docs/reference/chat_templating.md
similarity index 100%
rename from docs/guide/chat_templating.md
rename to docs/reference/chat_templating.md
diff --git a/docs/reference/generation/cfg.md b/docs/reference/generation/cfg.md
new file mode 100644
index 000000000..f0104ec63
--- /dev/null
+++ b/docs/reference/generation/cfg.md
@@ -0,0 +1,154 @@
+# Grammar-structured generation
+
+You can pass any context-free grammar in the EBNF format and Outlines will generate an output that is valid to this grammar:
+
+```python
+from outlines import models, generate
+
+arithmetic_grammar = """
+    ?start: expression
+
+    ?expression: term (("+" | "-") term)*
+
+    ?term: factor (("*" | "/") factor)*
+
+    ?factor: NUMBER
+           | "-" factor
+           | "(" expression ")"
+
+    %import common.NUMBER
+"""
+
+model = models.transformers("WizardLM/WizardMath-7B-V1.1")
+generator = generate.cfg(model, arithmetic_grammar)
+sequence = generator(
+  "Alice had 4 apples and Bob ate 2. "
+  + "Write an expression for Alice's apples:"
+)
+
+print(sequence)
+# (8-2)
+```
+
+###### Disclaimer
+
+!!! Note "Experimental"
+
+    Outlines current **community-contributed** implementation of CFG-structured generation is experimental. This does not reflect the performance of [.txt](https://dottxt.co)'s product, where we have optimized grammar-structured generation to be as fast as regex-structured generation. Additionally, it does not fully align with the approach described in our [technical report](https://arxiv.org/pdf/2307.09702), aside from its use of incremental/partial parsing. This feature is still a work in progress, requiring performance enhancements and bug fixes for an ideal implementation. For more details, please see our [grammar-related open issues on GitHub](https://github.com/dottxt-ai/outlines/issues?q=is%3Aissue+is%3Aopen+label%3Agrammar).
+
+!!! Note "Greedy"
+
+    To mitigate performance issues, CFG-structured generation will use rejection sampling and iterate over the candidate tokens highest logit first,, completing once a single valid token ID is selected. This is effectively greedy generation.
+
+## Ready-to-use grammars
+
+Outlines contains a (small) library of grammars that can be imported and use directly. We can rewrite the previous example as:
+
+```python
+from outlines import models, generate
+
+arithmetic_grammar = outlines.grammars.arithmetic
+
+model = models.transformers("WizardLM/WizardMath-7B-V1.1")
+generator = generate.cfg(model, arithmetic_grammar)
+sequence = generator(
+  "Alice had 4 apples and Bob ate 2. "
+  + "Write an expression for Alice's apples:"
+)
+
+print(sequence)
+# (8-2)
+```
+
+The following grammars are currently available:
+
+- Arithmetic grammar via `outlines.grammars.arithmetic`
+- JSON grammar via `outlines.grammars.json`
+
+If you would like more grammars to be added to the repository, please open an [issue](https://github.com/dottxt-ai/outlines/issues) or a [pull request](https://github.com/dottxt-ai/outlines/pulls).
+
+
+## Grammar guide
+
+A grammar is a list of rules and terminals that define a *language*:
+
+- Terminals define the vocabulary of the language; they may be a string, regular expression or combination of these and other terminals.
+- Rules define the structure of that language; they are a list of terminals and rules.
+
+Outlines uses the [Lark library](https://github.com/lark-parser/lark) to make Large Language Models generate text in a language of a grammar, it thus uses grammars defined in a format that Lark understands, based on the [EBNF syntax](https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form). Read the [Lark documentation](https://lark-parser.readthedocs.io/en/stable/grammar.html) for more details on grammar, the following is a small primer that should help get your started.
+
+In the following we will define a [LOGO-like toy language](https://github.com/lark-parser/lark/blob/master/examples/turtle_dsl.py) for python's [turtle](https://docs.python.org/3/library/turtle.html) library.
+
+### Terminals
+
+A turtle can take 4 different `MOVEMENT` move instructions: forward (`f`), backward (`b`), turn right (`r`) and turn left (`l`). It can take `NUMBER` number of steps in each direction, and draw lines in a specified `COLOR`. These define the vocabulary of our language:
+
+```ebnf
+MOVEMENT: "f"|"b"|"r"|"l"
+COLOR: LETTER+
+
+%import common.LETTER
+%import common.INT -> NUMBER
+%import common.WS
+%ignore WS
+```
+
+The lines that start with `%` are called "directive". They allow to import pre-defined terminals and rules, such as `LETTER` and `NUMBER`. `LETTER+` is a regular expressions, and indicates that a `COLOR` is made of at least one `LETTER`. The last two lines specify that we will ignore white spaces (`WS`) in the grammar.
+
+### Rules
+
+We now need to define our rules, by decomposing instructions we can send to the turtle via our python program. At each line of the program, we can either choose a direction and execute a given number of steps, change the color used to draw the pattern. We can also choose to start filling, make a series of moves, and stop filling. We can also choose to repeat a series of move.
+
+We can easily write the first two rules:
+
+```ebnf
+instruction: MOVEMENT NUMBER   -> movement
+           | "c" COLOR [COLOR] -> change_color
+```
+
+where `movement` and `change_color` represent aliases for the rules. A whitespace implied concatenating the elements, and `|` choosing either of the elements. The `fill` and `repeat` rules are slightly more complex, since they apply to a code block, which is made of instructions. We thus define a new `code_block`  rule that refers to `instruction` and finish implementing our rules:
+
+```ebnf
+instruction: MOVEMENT NUMBER            -> movement
+           | "c" COLOR [COLOR]          -> change_color
+           | "fill" code_block          -> fill
+           | "repeat" NUMBER code_block -> repeat
+
+code_block: "{" instruction "}"
+```
+
+We can now write the full grammar:
+
+```ebnf
+start: instruction+
+
+instruction: MOVEMENT NUMBER            -> movement
+            | "c" COLOR [COLOR]          -> change_color
+            | "fill" code_block          -> fill
+            | "repeat" NUMBER code_block -> repeat
+
+code_block: "{" instruction+ "}"
+
+MOVEMENT: "f"|"b"|"l"|"r"
+COLOR: LETTER+
+
+%import common.LETTER
+%import common.INT -> NUMBER
+%import common.WS
+%ignore WS
+```
+
+Notice the `start` rule, which defines the starting point of the grammar, i.e. the rule with which a program must start. This full grammars allows us to parse programs such as:
+
+```python
+c red yellow
+    fill { repeat 36 {
+        f200 l170
+    }}
+```
+
+The result of the parse, the parse tree, can then easily be translated into a Python program that uses the `turtle` library to draw a pattern.
+
+### Next steps
+
+This section provides a very brief overview of grammars and their possibilities. Check out the [Lark documentation](https://lark-parser.readthedocs.io/en/stable/index.html) for more thorough explanations and more examples.
diff --git a/docs/reference/generation/choices.md b/docs/reference/generation/choices.md
new file mode 100644
index 000000000..aed5af5ab
--- /dev/null
+++ b/docs/reference/generation/choices.md
@@ -0,0 +1,16 @@
+# Multiple choices
+
+Oultines allows you to make sure the generated text is chosen between different options:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.choice(model, ["skirt", "dress", "pen", "jacket"])
+answer = generator("Pick the odd word out: skirt, dress, pen, jacket")
+
+```
+
+!!! Note "Performance"
+
+    `generation.choice` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate from the same list of choices several times make sure that you only call `generate.choice` once.
diff --git a/docs/reference/generation/creating_grammars.md b/docs/reference/generation/creating_grammars.md
new file mode 100644
index 000000000..78e41282a
--- /dev/null
+++ b/docs/reference/generation/creating_grammars.md
@@ -0,0 +1,99 @@
+# Overview
+
+Outlines allows the use of [Lark](https://github.com/lark-parser/lark) grammars to guide generation. These grammars are used to construct parsers that filter out incompatible tokens during the generation process The result is a generation that adheres to the grammar's production rules.
+
+# Primer on Creating Grammars
+
+To create grammars for Outlines, a solid understanding of Lark grammars is necessary. Here's how you can get started:
+
+- Read Lark's grammars documentations [here](https://lark-parser.readthedocs.io/en/latest/grammar.html).
+- Review Outlines' existing grammars [here](/outlines/grammars).
+
+
+# Compatibility With Outlines
+
+It's important to note that not all Lark grammars work with Outlines. Changes may be necessary to ensure compatability.
+
+### LALR(1) Parser
+
+Outlines utilizes Larks LALR(1) parser, meaning the grammar must be unambiguous at least up to the next token (one token lookahead). Read Lark's official LALR(1) parser documentation [here](https://lark-parser.readthedocs.io/en/stable/parsers.html#lalr-1).
+
+If your grammar is ambiguous, you will recieve the following error at runtime:
+
+```
+GrammarError: Reduce/Reduce collision in Terminal('B') between the following rules:
+```
+
+### Regex Terminal Restrictions
+
+Outlines converts terminals to finite state machines using the [Interegular](https://github.com/MegaIng/interegular/) library. Not all regular expressions work with Interegular, mitigation is described in the subsections which follow.
+
+
+#### Avoid Lookarounds
+
+Examples of removing lookaround while maintaining the same functionality
+
+##### Example: Escaped String
+
+From Outlines' modified `ESCAPED_STRING` in [common.lark](/outlines/grammars/common.lark).
+
+Before:
+```
+_STRING_INNER: /.*?/
+_STRING_ESC_INNER: _STRING_INNER /(?<!\\)(\\\\)*?/
+
+ESCAPED_STRING : "\"" _STRING_ESC_INNER "\""
+```
+
+After:
+```
+_NON_CONTROL_CHAR: /([^"\\\x00-\x1F\x7F-\x9F])/
+_ESCAPED_CHAR: /\\/ (_NON_CONTROL_CHAR | /\\/ | /"/)
+ESCAPED_STRING_INNER: _NON_CONTROL_CHAR | _ESCAPED_CHAR
+ESCAPED_STRING: /"/ ESCAPED_STRING_INNER* /"/
+```
+
+#### Avoid Backreferences
+
+Backreferences, for example `([ab]^*)\1`, cannot be simulated by a finite state machine, and will result in an error if used.
+
+# Creating a Valid Grammar
+
+You can use Outlines' test suite to verify your grammar.
+
+### 1) Create Your Grammar
+
+Create your grammar file named `your_new_grammar.lark`, adhering to the guidelines provided above. Add it to `outlines/grammars/` (ensure attribution is included and license is compatible).
+
+Update `outlines/grammars.py` with a line including your grammar.
+
+### 2) Test Your Grammar
+
+Test grammar for false negatives, ensure sample grammars can be generated:
+- Add valid example outputs which are compliant with the grammar to `tests/benchmark/cfg_samples/your_new_grammar/`
+- Run the tests for your grammar via `pytest -s tests/fsm/test_cfg_guide.py::test_cfg_grammar_sample -k "your_new_grammar"`
+
+Test grammar for false positives, ensure invalid outputs aren't generated.
+
+Currently there isn't a builtin false positive testing utility. It is recommended you smoke test via
+```
+from outlines import models, generate, grammars
+model = models.transformers("mistralai/Mistral-7B-v0.1")
+generator = generate.cfg(model, grammars.your_new_grammar)
+result = generator(<your prompt to generate output for your grammar>)
+print(result)
+```
+
+# Converting
+There are a few tools available for converting from other grammars to lark. These tools serve as a starting point. However, you will typically need to make additional adjustments to ensure full compatibility and proper functioning within Outlines.
+
+Tools:
+- Larks built in "Nearley-to-Lark" converter https://lark-parser.readthedocs.io/en/latest/tools.html
+- Convert ANTLR4 to Lark (Note, most antlr4 grammars are not LALR(1) compatible, so will require additional tweaking) https://github.com/kaby76/Domemtech.Trash/blob/main/src/trconvert/readme.md
+- Extract EBNF from Yacc files https://www.bottlecaps.de/rr/ui
+
+Reference Grammars:
+- Github Lark Grammars https://github.com/search?q=path%3A*.lark&type=code
+- Github Nearley Grammars https://github.com/search?q=path%3A*.ne+%22-%3E%22&type=code
+- Antlr4 grammars https://github.com/antlr/grammars-v4/
+- Grammar zoo https://slebok.github.io/zoo/index.html#html
diff --git a/docs/reference/generation/custom_fsm_ops.md b/docs/reference/generation/custom_fsm_ops.md
new file mode 100644
index 000000000..5c4be96fa
--- /dev/null
+++ b/docs/reference/generation/custom_fsm_ops.md
@@ -0,0 +1,37 @@
+# Custom FSM Operations
+
+Outlines is fast because it compiles regular expressions into an index ahead of inference. To do so we use the equivalence between regular expressions and Finite State Machines (FSMs), and the library [interegular](https://github.com/MegaIng/interegular) to perform the translation.
+
+Alternatively, one can pass a FSM built using `integular` directly to structure the generation.
+
+## Example
+
+### Using the `difference` operation
+
+In the following example we build a fsm which recognizes only the strings valid to the first regular expression but not the second. In particular, it will prevent the words "pink" and "elephant" from being generated:
+
+```python
+import interegular
+from outlines import models, generate
+
+
+list_of_strings_pattern = """\["[^"\s]*"(?:,"[^"\s]*")*\]"""
+pink_elephant_pattern = """.*(pink|elephant).*"""
+
+list_of_strings_fsm = interegular.parse_pattern(list_of_strings_pattern).to_fsm()
+pink_elephant_fsm = interegular.parse_pattern(pink_elephant_pattern).to_fsm()
+
+difference_fsm = list_of_strings_fsm - pink_elephant_fsm
+
+difference_fsm_fsm.accepts('["a","pink","elephant"]')
+# False
+difference_fsm_fsm.accepts('["a","blue","donkey"]')
+# True
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.fsm(model, difference_fsm)
+response = generator("Don't talk about pink elephants")
+```
+
+To see the other operations available, consult [interegular's documentation](https://github.com/MegaIng/interegular/blob/master/interegular/fsm.py).
diff --git a/docs/reference/generation/format.md b/docs/reference/generation/format.md
new file mode 100644
index 000000000..749baa8b6
--- /dev/null
+++ b/docs/reference/generation/format.md
@@ -0,0 +1,23 @@
+# Type constraints
+
+We can ask completions to be restricted to valid python types:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.format(model, int)
+answer = generator("When I was 6 my sister was half my age. Now I’m 70 how old is my sister?")
+print(answer)
+# 67
+```
+
+The following types are currently available:
+
+- int
+- float
+- bool
+- datetime.date
+- datetime.time
+- datetime.datetime
+- We also provide [custom types](types.md)
diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
new file mode 100644
index 000000000..930ad9d22
--- /dev/null
+++ b/docs/reference/generation/generation.md
@@ -0,0 +1,216 @@
+---
+title: Generation
+---
+
+# Generation
+
+Once an [Outlines model](../models/models.md) is constructed you can use `outlines.generate` to generate text. Standard LLM generation is possible via `outlines.generate.text`, along with a variety of structured generation methods described below. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
+
+Before generating text, you must construct an `outlines.model`. Example:
+
+```python
+import outlines
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+```
+### Text generator
+
+```python
+generator = outlines.generate.text(model)
+
+result = generator("Question: What's 2+2? Answer:", max_tokens=100)
+print(result)
+# The answer is 4
+
+# Outlines also supports streaming output
+stream = generator.stream("What's 2+2?", max_tokens=4)
+for i in range(5):
+	token = next(stream)
+	print(repr(token))
+# '2'
+# '+'
+# '2'
+# ' equals'
+# '4'
+```
+
+### [Multi-label classification](./choices.md)
+
+Outlines allows you to do multi-label classification by guiding the model so it can only output either of the specified choices:
+
+```python
+import outlines
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = outlines.generate.choice(model, ["Blue", "Red", "Yellow"])
+
+color = generator("What is the closest color to Indigo? ")
+print(color)
+# Blue
+```
+
+### [JSON-structured generation](./json.md)
+
+Outlines can guide models so that they output valid JSON **100%** of the time. You can either specify the structure using [Pydantic][pydantic]{:target="_blank"} or a string that contains a [JSON Schema][jsonschema]{:target="_blank"}:
+
+=== "Pydantic"
+
+    ```python
+    from enum import Enum
+    from pydantic import BaseModel, constr, conint
+
+    import outlines
+
+    class Armor(str, Enum):
+        leather = "leather"
+        chainmail = "chainmail"
+        plate = "plate"
+
+
+    class Character(BaseModel):
+        name: constr(max_length=10)
+        age: conint(gt=18, lt=99)
+        armor: Armor
+        strength: conint(gt=1, lt=100)
+
+    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+    generator = outlines.generate.json(model, Character)
+
+    character = generator(
+        "Generate a new character for my awesome game: "
+        + "name, age (between 1 and 99), armor and strength. "
+        )
+    print(character)
+    # name='Orla' age=21 armor=<Armor.plate: 'plate'> strength=8
+    ```
+
+=== "JSON Schema"
+
+    ```python
+    import outlines
+
+    schema = """{
+        "$defs": {
+            "Armor": {
+                "enum": ["leather", "chainmail", "plate"],
+                "title": "Armor",
+                "type": "string"
+            }
+        },
+        "properties": {
+            "name": {"maxLength": 10, "title": "Name", "type": "string"},
+            "age": {"title": "Age", "type": "integer"},
+            "armor": {"$ref": "#/$defs/Armor"},
+            "strength": {"title": "Strength", "type": "integer"}\
+        },
+        "required": ["name", "age", "armor", "strength"],
+        "title": "Character",
+        "type": "object"
+    }"""
+
+    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+    generator = outlines.generate.json(model, schema)
+    character = generator(
+        "Generate a new character for my awesome game: "
+        + "name, age (between 1 and 99), armor and strength. "
+        )
+    print(character)
+    # {'name': 'Yuki', 'age': 24, 'armor': 'plate', 'strength': 3}
+    ```
+
+!!! Note
+
+    We advise you to constrain the length of the strings fields when first testing your schema, especially with small models.
+
+### [Grammar-structured generation](./cfg.md)
+
+Outlines also allows to generate text that is valid to any [context-free grammar][cfg]{:target="_blank"} (CFG) in the [EBNF format][ebnf]{:target="_blank"}. Grammars can be intimidating, but they are a very powerful tool! Indeed, they determine the syntax of every programming language, valid chess moves, molecule structure, can help with procedural graphics generation, etc.
+
+Here we show a simple example of a grammar that defines arithmetic operations:
+
+```python
+from outlines import models, generate
+
+arithmetic_grammar = """
+    ?start: sum
+
+    ?sum: product
+        | sum "+" product   -> add
+        | sum "-" product   -> sub
+
+    ?product: atom
+        | product "*" atom  -> mul
+        | product "/" atom  -> div
+
+    ?atom: NUMBER           -> number
+         | "-" atom         -> neg
+         | "(" sum ")"
+
+    %import common.NUMBER
+    %import common.WS_INLINE
+
+    %ignore WS_INLINE
+"""
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = generate.cfg(model, arithmetic_grammar, max_tokens=100)
+
+result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
+print(result)
+# 5+5+5+5+5
+```
+
+
+EBNF grammars can be cumbersome to write. This is why Outlines provides grammar definitions in the `outlines.grammars.` module
+
+```python
+from outlines import models, generate, grammars
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = generate.cfg(model, grammars.arithmetic, max_tokens=100)
+
+result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
+print(result)
+# 5+5+5+5+5
+```
+
+The available grammars are listed [here](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
+
+
+### [Regex-structured generation](./regex.md)
+
+Slightly simpler, but no less useful, Outlines can generate text that is in the language of a [regular expression](https://www.regular-expressions.info/tutorial.html). For instance to force the model to generate IP addresses:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+
+regex_str = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
+generator = generate.regex(model, regex_str)
+
+result = generator("What is the IP address of localhost?\nIP: ")
+print(result)
+# 127.0.0.100
+```
+
+### [Generate a given Python type](./types.md)
+
+We provide a shortcut to regex-structured generation for simple use cases. Pass a Python type to the `outlines.generate.format` function and the LLM will output text that matches this type:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+generator = generate.format(model, int)
+
+result = generator("What is 2+2?")
+print(result)
+# 4
+```
+
+
+[jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
+[pydantic]: https://docs.pydantic.dev/latest
+[cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
+[ebnf]: https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form
diff --git a/docs/reference/generation/json.md b/docs/reference/generation/json.md
new file mode 100644
index 000000000..05dc69654
--- /dev/null
+++ b/docs/reference/generation/json.md
@@ -0,0 +1,178 @@
+# JSON structured generation
+
+Outlines can make any open source model return a JSON object that follows a structure that is specified by the user. This is useful whenever we want the output of the model to be processed by code downstream: code does not understand natural language but rather the structured language it has been programmed to understand.
+
+There are mostly two reasons why someone would want to get an output formatted as JSON from a LLM:
+
+1. Parse the answer (e.g. with Pydantic), store it somewhere, return it to a user, etc.
+2. Call a function with the result
+
+Outlines has you covered in both cases! Indeed, to define the structure of the JSON you want the model to follow you can either provide a Pydantic model, or a function. No need to duplicate code!
+
+## Using Pydantic
+
+Outlines can infer the structure of the output from a Pydantic model. The result is an instance of the model that contains the values returned by the LLM:
+
+```python
+from pydantic import BaseModel
+
+from outlines import models, generate
+
+
+class User(BaseModel):
+    name: str
+    last_name: str
+    id: int
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.json(model, User)
+result = generator(
+    "Create a user profile with the fields name, last_name and id"
+)
+print(result)
+# User(name="John", last_name="Doe", id=11)
+```
+
+!!! Note "JSON and whitespaces"
+
+    By default Outlines prevents the model from generating json with syntactic newlines, tabs, or multiple spaces. The default `whitespace_pattern` is `r"[ ]?"`. Small models tend to enter an infinite repetition loop if the `whitespace_pattern` allows infinite spacing. If you would like to allow the model to generate multiple tabs, newlines, and spaces, you can set the whitespace pattern as follows:
+
+    ```python
+    generator = generate.json(model, User, whitespace_pattern=r"[\n\t ]*")
+    ```
+
+!!! Note "Performance"
+
+    `generation.json` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate several times with the same schema make sure that you only call `generate.json` once.
+
+!!! Tip "Custom types"
+
+    Outlines provides [custom Pydantic types](types.md) so you do not have to write regular expressions for common types, such as phone numbers or zip codes.
+
+## Using a JSON Schema
+
+Instead of a Pydantic model you can pass a string that represents a [JSON Schema](https://json-schema.org/) specification to `generate.json`:
+
+```python
+from pydantic import BaseModel
+
+from outlines import models
+from outlines import generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+
+schema = """
+{
+  "title": "User",
+  "type": "object",
+  "properties": {
+    "name": {"type": "string"},
+    "last_name": {"type": "string"},
+    "id": {"type": "integer"}
+  },
+  "required": ["name", "last_name", "id"]
+}
+"""
+
+generator = generate.json(model, schema)
+result = generator(
+    "Create a user profile with the fields name, last_name and id"
+)
+print(result)
+# User(name="John", last_name="Doe", id=11)
+```
+
+## From a function's signature
+
+Outlines can infer the structure of the output from the signature of a function. The result is a dictionary, and can be passed directly to the function using the usual dictionary expansion syntax `**`:
+
+```python
+from outlines import models
+from outlines import generate
+
+def add(a: int, b: int):
+    return a + b
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.json(model, add)
+result = generator("Return two integers named a and b respectively. a is odd and b even.")
+
+print(add(**result))
+# 3
+```
+
+A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
+
+
+## From a dynamic JSON schema builder - GenSON
+
+Outlines integrated [GenSON](https://github.com/wolverdude/GenSON) builders to be able to dynamicly declare JSON schemas. It can be used as follow:
+
+```python
+from genson import SchemaBuilder
+
+from outlines import models
+from outlines import generate
+
+builder = SchemaBuilder()
+builder.add_schema({"type": "object", "properties": {}})
+builder.add_object({"name": "Toto", "age": 5})
+
+model = models.transformers(
+    "HuggingFaceTB/SmolLM2-135M",
+    device="auto",
+)
+generator = generate.json(model, builder)
+
+res = generator("Return a json of a young boy")
+print(res)
+# {"name": "Ben", "age": 10}
+```
+
+Anytime you are updating the schema through the builder, you need to redifine the outline generator to include these changes. From the the previous example:
+
+```python
+from genson import SchemaBuilder
+
+from outlines import models
+from outlines import generate
+
+builder = SchemaBuilder()
+builder.add_schema({"type": "object", "properties": {}})
+builder.add_object({"name": "Toto", "age": 5})
+
+model = models.transformers(
+    "HuggingFaceTB/SmolLM2-135M",
+    device="auto",
+)
+generator = generate.json(model, builder)
+
+res = generator("Return a json of a young boy")
+print(res)
+# {"name": "Ben", "age": 10}
+
+builder.add_object({"hobby": "sports"})
+generator = generate.json(model, builder)
+
+res = generator("Return a json of a youg boy whose hobby is coding")
+print(res)
+# {"name": "Ben", "age": 10, "hobby": "coding"}
+```
+
+!!! Note
+
+    Beware of [GenSON](https://github.com/wolverdude/GenSON)'s behavior regarding dynamic amending of schemas through their builder. Here is an example of how you could lose `required` informations and generate json with missing fields:
+
+    ```python
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"name": "Toto", "age": 5})
+
+    print(builder.to_schema())
+    # {'$schema': 'http://json-schema.org/schema#', 'type': 'object', 'properties': {'name': {'type': 'string'}, 'age': {'type': 'integer'}}, 'required': ['age', 'name']}
+
+    builder.add_object({"hobby": "sport"})
+    print(builder.to_schema())
+    # {'name': {'type': 'string'}, 'age': {'type': 'integer'}, 'hobby': {'type': 'string'}}}
+    ```
diff --git a/docs/reference/generation/regex.md b/docs/reference/generation/regex.md
new file mode 100644
index 000000000..23df35567
--- /dev/null
+++ b/docs/reference/generation/regex.md
@@ -0,0 +1,27 @@
+# Regular expressions
+
+Outlines can guarantee that the text generated by the LLM will be valid to a regular expression:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+
+generator = generate.regex(
+    model,
+    r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)",
+)
+
+prompt = "What is the IP address of the Google DNS servers? "
+answer = generator(prompt, max_tokens=30)
+
+print(answer)
+# What is the IP address of the Google DNS servers?
+# 2.2.6.1
+```
+
+If you find yourself using `generate.regex` to restrict the answers' type you can take a look at [type-structured generation](types.md) instead.
+
+!!! Note "Performance"
+
+    `generate.regex` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate several times using the same regular expression make sure that you only call `generate.regex` once.
diff --git a/docs/reference/generation/structured_generation_explanation.md b/docs/reference/generation/structured_generation_explanation.md
new file mode 100644
index 000000000..aa27a7a85
--- /dev/null
+++ b/docs/reference/generation/structured_generation_explanation.md
@@ -0,0 +1,54 @@
+# How does Outlines work?
+
+
+Language models generate text token by token, using the previous token sequence as input and sampled logits as output. This document explains the structured generation process, where only legal tokens are considered for the next step based on a predefined automata, e.g. a regex-defined [finite-state machine](https://en.wikipedia.org/wiki/Finite-state_machine) (FSM) or [Lark](https://lark-parser.readthedocs.io/en/stable/) grammar.`
+
+
+## Worked Example
+
+Let's consider a worked example with a pattern for whole and decimal numbers:
+
+`^\d*(\.\d+)?$`.
+
+### Creating Automata
+
+The pattern is first converted into an automata. Below is a brief explanation of the automata conversion and its representation.
+
+**Automata Diagram:**
+
+```mermaid
+graph LR
+    node0("1-9") --> node1("1-9")
+    node1 --> node1
+    node1 --> nodeEND{{END}}
+    node1 --> nodePeriod(".")
+    nodePeriod --> node2("1-9")
+    node2 --> node2
+    node2 --> nodeEND{{END}}
+```
+
+### Generating a Token
+
+Let's assume that we're in the middle of generation, and so far "748" has been generated. Here is the automata with the current state highlighted in green, with the legal next characters being another number (1-9), a dot (.), or end of sequence.
+
+```mermaid
+graph LR
+    node0("1-9") --> node1("1-9")
+    node1 --> node1
+    node1 --> nodeEND{{END}}
+    node1 --> nodePeriod(".")
+    nodePeriod --> node2("1-9")
+    node2 --> node2
+    node2 --> nodeEND{{END}}
+
+    style node1 fill:#090
+```
+
+Generating a token requires the following steps:
+
+- Feed the previous input sequence ("748") into the language model.
+- Language model runs a forward pass and produces token logits.
+- Outlines logits processor sets the probability of illegal tokens to 0%.
+- A token is sampled from the set of legal tokens.
+
+![Generation and Logits Processing Flow Chart](../../assets/images/logits_processing_diagram.svg)
diff --git a/docs/reference/generation/types.md b/docs/reference/generation/types.md
new file mode 100644
index 000000000..eb6d7382b
--- /dev/null
+++ b/docs/reference/generation/types.md
@@ -0,0 +1,83 @@
+# Custom types
+
+Outlines provides custom Pydantic types so you can focus on your use case rather than on writing regular expressions:
+
+| Category | Type | Import | Description |
+|:--------:|:----:|:-------|:------------|
+| ISBN | 10 & 13 | `outlines.types.ISBN` | There is no guarantee that the [check digit][wiki-isbn] will be correct |
+| Airport | IATA | `outlines.types.airports.IATA` | Valid [airport IATA codes][wiki-airport-iata] |
+| Country | alpha-2 code | `outlines.types.airports.Alpha2` | Valid [country alpha-2 codes][wiki-country-alpha-2] |
+|  | alpha-3 code | `outlines.types.countries.Alpha3` | Valid [country alpha-3 codes][wiki-country-alpha-3] |
+|  | numeric code | `outlines.types.countries.Numeric` | Valid [country numeric codes][wiki-country-numeric] |
+|  | name | `outlines.types.countries.Name` | Valid country names |
+|  | flag | `outlines.types.countries.Flag` | Valid flag emojis |
+| | email | `outlines.types.Email` | Valid email address |
+
+Some types require localization. We currently only support US types, but please don't hesitate to create localized versions of the different types and open a Pull Request. Localized types are specified using `types.locale` in the following way:
+
+```python
+from outlines import types
+
+types.locale("us").ZipCode
+types.locale("us").PhoneNumber
+```
+
+Here are the localized types that are currently available:
+
+| Category | Locale | Import | Description |
+|:--------:|:----:|:-------|:------------|
+| Zip code | US | `ZipCode` | Generate US Zip(+4) codes |
+| Phone number  | US | `PhoneNumber` | Generate valid US phone numbers |
+
+
+You can use these types in Pydantic schemas for JSON-structured generation:
+
+```python
+from pydantic import BaseModel
+
+from outlines import models, generate, types
+
+# Specify the locale for types
+locale = types.locale("us")
+
+class Client(BaseModel):
+    name: str
+    phone_number: locale.PhoneNumber
+    zip_code: locale.ZipCode
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.json(model, Client)
+result = generator(
+    "Create a client profile with the fields name, phone_number and zip_code"
+)
+print(result)
+# name='Tommy' phone_number='129-896-5501' zip_code='50766'
+```
+
+Or simply with `outlines.generate.format`:
+
+```python
+from pydantic import BaseModel
+
+from outlines import models, generate, types
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.format(model, types.locale("us").PhoneNumber)
+result = generator(
+    "Return a US Phone number: "
+)
+print(result)
+# 334-253-2630
+```
+
+
+We plan on adding many more custom types. If you have found yourself writing regular expressions to generate fields of a given type, or if you could benefit from more specific types don't hesite to [submit a PR](https://github.com/dottxt-ai/outlines/pulls) or [open an issue](https://github.com/dottxt-ai/outlines/issues/new/choose).
+
+
+[wiki-isbn]: https://en.wikipedia.org/wiki/ISBN#Check_digits
+[wiki-airport-iata]: https://en.wikipedia.org/wiki/IATA_airport_code
+[wiki-country-alpha-2]: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2
+[wiki-country-alpha-3]: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3
+[wiki-country-numeric]: https://en.wikipedia.org/wiki/ISO_3166-1_numeric
diff --git a/docs/reference/index.md b/docs/reference/index.md
new file mode 100644
index 000000000..0cd42ba88
--- /dev/null
+++ b/docs/reference/index.md
@@ -0,0 +1,15 @@
+# Reference
+
+## Structured generation
+
+While LLM capabilities are increasingly impressive, we can make their output more reliable by steering the generation. Outlines thus offers mechanisms to specify high level constraints on text completions by generative language models.
+
+Stopping sequence
+By default, language models stop generating tokens after and <EOS> token was generated, or after a set maximum number of tokens. Their output can be verbose, and for practical purposes it is often necessary to stop the generation after a given sequence has been found instead. You can use the stop_at keyword argument when calling the model with a prompt:
+
+```python
+import outlines.models as models
+
+complete = models.openai("gpt-4o-mini")
+expert = complete("Name an expert in quantum gravity.", stop_at=["\n", "."])
+```
diff --git a/docs/reference/models/exllamav2.md b/docs/reference/models/exllamav2.md
new file mode 100644
index 000000000..a4e727840
--- /dev/null
+++ b/docs/reference/models/exllamav2.md
@@ -0,0 +1,15 @@
+# ExllamaV2
+
+The `outlines.models.exllamav2` model requires a Logits Processor component for compatibility with Outlines structured generation. While ExLlamaV2 doesn't natively support this feature, a third-party fork provides the necessary functionality. You can install it with the following command:
+
+```bash
+pip install git+https://github.com/lapp0/exllamav2@sampler-logits-processor
+```
+
+Install other requirements:
+
+```bash
+pip install transformers torch
+```
+
+*Coming soon*
diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
new file mode 100644
index 000000000..51b62eca8
--- /dev/null
+++ b/docs/reference/models/llamacpp.md
@@ -0,0 +1,230 @@
+# Llama.cpp
+
+Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library][llamacpp]. Llamacpp allows to run quantized models on machines with limited compute.
+
+!!! Note "Installation"
+
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration. See the [installation section](#installation) for instructions to install `llama-cpp-python` with CUDA, Metal, ROCm and other backends. To get started quickly you can also run:
+
+    ```bash
+    pip install "outlines[llamacpp]"
+    ```
+
+## Load the model
+
+You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+
+```python
+from outlines import models
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+```
+
+This will download the model files to the hub cache folder and load the weights in memory.
+
+You can also initialize the model by passing the path to the weights on your machine. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
+
+```python
+from outlines import models
+from llama_cpp import Llama
+
+llm = Llama("./phi-2.Q4_K_M.gguf")
+model = models.LlamaCpp(llm)
+```
+
+If you need more control, you can pass the same keyword arguments to the model as you would pass in the [llama-ccp-library][llamacpp]:
+
+```python
+from outlines import models
+
+model = models.llamacpp(
+    "TheBloke/phi-2-GGUF",
+    "phi-2.Q4_K_M.gguf"
+    n_ctx=512,  # to set the context length value
+)
+```
+
+**Main parameters:**
+
+| Parameters | Type | Description | Default |
+|------------|------|-------------|---------|
+| `n_gpu_layers`| `int` | Number of layers to offload to GPU. If -1, all layers are offloaded | `0` |
+| `split_mode` | `int` | How to split the model across GPUs. `1` for layer-wise split, `2` for row-wise split | `1` |
+| `main_gpu` | `int` | Main GPU | `0` |
+| `tensor_split` | `Optional[List[float]]` | How split tensors should be distributed across GPUs. If `None` the model is not split. | `None` |
+| `n_ctx` | `int` | Text context. Inference from the model if set to `0` | `0` |
+| `n_threads` | `Optional[int]` | Number of threads to use for generation. All available threads if set to `None`.| `None` |
+| `verbose` | `bool` | Print verbose outputs to `stderr` | `False` |
+
+See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters.
+
+### Load the model on GPU
+
+!!! Note
+
+    [Make sure](#cuda) that you installed `llama-cpp-python` with GPU support.
+
+ To load the model on GPU, pass `n_gpu_layers=-1`:
+
+```python
+from outlines import models
+
+model = models.llamacpp(
+    "TheBloke/phi-2-GGUF",
+    "phi-2.Q4_K_M.gguf",
+    n_gpu_layers=-1,  # to use GPU acceleration
+)
+```
+
+This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
+
+### Load LoRA adapters
+
+You can load LoRA adapters dynamically:
+
+```python
+from outlines import models, generate
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+generator = generate.text(model)
+answer_1 = generator("prompt")
+
+model.load_lora("./path/to/adapter.gguf")
+answer_2 = generator("prompt")
+```
+
+To load another adapter you need to re-initialize the model. Otherwise the adapter will be added on top of the previous one:
+
+```python
+from outlines import models
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+model.load_lora("./path/to/adapter1.gguf")  # Load first adapter
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+model.load_lora("./path/to/adapter2.gguf")  # Load second adapter
+```
+
+## Generate text
+
+In addition to the parameters described in the [text generation section](../text.md) you can pass extra keyword arguments, for instance to set sampling parameters not exposed in Outlines' public API:
+
+```python
+from outlines import models, generate
+
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+generator = generate.text(model)
+
+answer = generator("A prompt", presence_penalty=0.8)
+```
+
+**Extra keyword arguments:**
+
+The value of the keyword arguments you pass to the generator suspersede the values set when initializing the sampler or generator. All extra sampling methods and repetition penalties are disabled by default.
+
+| Parameters | Type | Description | Default |
+|------------|------|-------------|---------|
+| `suffix` | `Optional[str]` | A suffix to append to the generated text. If `None` no suffix is added. | `None` |
+| `echo` | `bool` | Whether to preprend the prompt to the completion. | `False` |
+| `seed` | `int` | The random seed to use for sampling. | `None` |
+| `max_tokens` | `Optional[int]` | The maximum number of tokens to generate. If `None` the maximum number of tokens depends on `n_ctx`. | `16` |
+| `frequence_penalty` | `float` | The penalty to apply to tokens based on their frequency in the past 64 tokens. | `0.0` |
+| `presence_penalty` | `float` | The penalty to apply to tokens based on their presence in the past 64 tokens. | `0.0` |
+| `repeat_penalty` | `float` | The penalty to apply to repeated tokens in the past 64 tokens. | `1.` |
+| `stopping_criteria` | `Optional[StoppingCriteriaList]` | A list of stopping criteria to use. | `None`
+| `logits_processor` | `Optional[LogitsProcessorList]` | A list of logits processors to use. The logits processor used for structured generation will be added to this list. | `None`
+| `temperature` | `float` | The temperature to use for sampling | `1.0` |
+| `top_p` | `float` | The top-p value to use for [nucleus sampling][degeneration]. | `1.` |
+| `min_p` | `float` | The min-p value to use for [minimum-p sampling][minimum-p]. | `0.` |
+| `typical_p` | `float` | The p value to use for [locally typical sampling][locally-typical]. | `1.0` |
+| `stop` | `Optional[Union[str, List[str]]]` | A list of strings that stop generation when encountered. | `[]` |
+| `top_k` |  `int` | The top-k value used for [top-k sampling][top-k]. Negative value to consider all logit values. | `-1.` |
+| `tfs_z` | `float` | The [tail-free sampling][tail-free] parameter. | `1.0` |
+| `mirostat_mode` | `int` | The [mirostat sampling][mirostat] mode. | `0` |
+| `mirostat_tau` | `float` | The target cross-entropy for [mirostat sampling][mirostat].| `5.0` |
+| `mirostat_eta` | `float` | The learning rate used to update `mu` in [mirostat sampling][mirostat]. | `0.1` |
+
+See the [llama-cpp-python documentation][llama-cpp-python-call] for the full and up-to-date list of parameters and the [llama.cpp code][llama-cpp-sampling-params] for the default values of other
+sampling parameters.
+
+### Streaming
+
+
+## Installation
+
+You need to install the `llama-cpp-python` library to use the llama.cpp integration.
+
+### CPU
+
+For a *CPU-only* installation run:
+
+```bash
+pip install llama-cpp-python
+```
+
+!!! Warning
+
+    Do not run this command if you want support for BLAS, Metal or CUDA. Follow the instructions below instead.
+
+### CUDA
+
+```bash
+CMAKE_ARGS="-DLLAMA_CUDA=on" pip install llama-cpp-python
+```
+
+It is also possible to install pre-built wheels with CUDA support (Python 3.10 and above):
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/<cuda-version>
+```
+
+Where `<cuda-version>` is one of the following, depending on the version of CUDA installed on your system:
+
+- `cu121` for CUDA 12.1
+- `cu122` for CUDA 12.2
+- `cu123` CUDA 12.3
+
+### Metal
+
+```bash
+CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
+```
+
+It is also possible to install pre-build wheels with Metal support (Python 3.10 or above, MacOS 11.0 and above):
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/metal
+```
+
+### OpenBLAS
+
+```bash
+CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
+```
+
+### Other backend
+
+`llama.cpp` supports many other backends. Refer to the [llama.cpp documentation][llama-cpp-python-install] to use the following backends:
+
+- CLBast (OpenCL)
+- hipBLAS (ROCm)
+- Vulkan
+- Kompute
+- SYCL
+
+
+
+
+[llamacpp]: https://github.com/abetlen/llama-cpp-python
+[llama-cpp-python-call]: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
+[llama-cpp-python-install]: https://github.com/abetlen/llama-cpp-python/tree/08b16afe11e7b42adec2fed0a781123383476045?tab=readme-ov-file#supported-backends
+[llama-cpp-sampling-params]: https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
+[mirostat]: https://arxiv.org/abs/2007.14966
+[degeneration]: https://arxiv.org/abs/1904.09751
+[top-k]: https://arxiv.org/abs/1805.04833
+[minimum-p]: https://github.com/ggerganov/llama.cpp/pull/3841
+[locally-typical]: https://arxiv.org/abs/2202.00666
+[tail-free]: https://www.trentonbricken.com/Tail-Free-Sampling
diff --git a/docs/reference/models/mlxlm.md b/docs/reference/models/mlxlm.md
new file mode 100644
index 000000000..d435b9c1f
--- /dev/null
+++ b/docs/reference/models/mlxlm.md
@@ -0,0 +1,93 @@
+# mlx-lm
+
+Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
+
+!!! Note "Installation"
+
+    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To get started quickly you can also run:
+
+    ```bash
+    pip install "outlines[mlxlm]"
+    ```
+
+
+## Load the model
+
+You can initialize the model by passing the name of the repository on the HuggingFace Hub. The official repository for mlx-lm supported models is [mlx-community](https://huggingface.co/mlx-community).
+
+```python
+from outlines import models
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+```
+
+This will download the model files to the hub cache folder and load the weights in memory.
+
+The arguments `model_config` and `tokenizer_config` are available to modify loading behavior. For example, per the `mlx-lm` [documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms#supported-models), you must set an eos_token for `qwen/Qwen-7B`. In outlines you may do so via
+
+```
+model = models.mlxlm(
+    "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit",
+    tokenizer_config={"eos_token": "<|endoftext|>", "trust_remote_code": True},
+)
+```
+
+**Main parameters:**
+
+(Subject to change. Table based on [mlx-lm.load docstring](https://github.com/ml-explore/mlx-examples/blob/main/llms/mlx_lm/utils.py#L429))
+
+| Parameters         | Type   | Description                                                                                      | Default |
+|--------------------|--------|--------------------------------------------------------------------------------------------------|---------|
+| `tokenizer_config` | `dict` | Configuration parameters specifically for the tokenizer. Defaults to an empty dictionary.        | `{}`    |
+| `model_config`     | `dict` | Configuration parameters specifically for the model. Defaults to an empty dictionary.            | `{}`    |
+| `adapter_path`     | `str`  | Path to the LoRA adapters. If provided, applies LoRA layers to the model.                        | `None`  |
+| `lazy`             | `bool` | If False, evaluate the model parameters to make sure they are loaded in memory before returning. | `False` |
+
+
+## Generate text
+
+You may generate text using the parameters described in the [text generation documentation](../text.md).
+
+With the loaded model, you can generate text or perform structured generation, e.g.
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+generator = generate.text(model)
+
+answer = generator("A prompt", temperature=2.0)
+```
+
+## Streaming
+
+You may creating a streaming iterable with minimal changes
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+generator = generate.text(model)
+
+for token_str in generator.text("A prompt", temperature=2.0):
+    print(token_str)
+```
+
+## Structured
+
+You may perform structured generation with mlxlm to guarantee your output will match a regex pattern, json schema, or lark grammar.
+
+Example: Phone number generation with pattern `"\\+?[1-9][0-9]{7,14}"`:
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+
+phone_number_pattern = "\\+?[1-9][0-9]{7,14}"
+generator = generate.regex(model, phone_number_pattern)
+
+model_output = generator("What's Jennys Number?\n")
+print(model_output)
+# '8675309'
+```
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
new file mode 100644
index 000000000..eab70f2a1
--- /dev/null
+++ b/docs/reference/models/models.md
@@ -0,0 +1,56 @@
+---
+title: Models
+---
+
+# Models
+
+Outlines supports generation using a number of inference engines (`outlines.models`). Loading a model using outlines follows a similar interface between inference engines:
+
+```python
+import outlines
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+model = outlines.models.transformers_vision("llava-hf/llava-v1.6-mistral-7b-hf")
+model = outlines.models.vllm("microsoft/Phi-3-mini-128k-instruct")
+model = outlines.models.llamacpp(
+    "microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf"
+)
+model = outlines.models.exllamav2("bartowski/Phi-3-mini-128k-instruct-exl2")
+model = outlines.models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
+
+model = outlines.models.openai(
+    "gpt-4o-mini",
+    api_key=os.environ["OPENAI_API_KEY"]
+)
+```
+
+
+# Feature Matrix
+|                   | [Transformers](transformers.md) | [Transformers Vision](transformers_vision.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md)* |
+|-------------------|--------------|---------------------|------|-----------|-----------|-------|---------|
+| **Device**        |              |                     |      |           |           |       |         |
+| Cuda              | ✅           | ✅                  | ✅   | ✅        | ✅        | ❌    | N/A     |
+| Apple Silicon     | ✅           | ✅                  | ❌   | ✅        | ✅        | ✅    | N/A     |
+| x86 / AMD64       | ✅           | ✅                  | ❌   | ✅        | ✅        | ❌    | N/A     |
+| **Sampling**      |              |                     |      |           |           |       |         |
+| Greedy            | ✅           | ✅                  | ✅   | ✅*       | ✅        | ✅    | ❌      |
+| Multinomial       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Multiple Samples  | ✅           | ✅                  |      | ❌        |           | ❌    | ✅      |
+| Beam Search       | ✅           | ✅                  | ✅   | ❌        | ✅        | ❌    | ❌      |
+| **Generation**    |              |                     |      |           |           |       |         |
+| Batch             | ✅           | ✅                  | ✅   | ❌        | ?         | ❌    | ❌      |
+| Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
+| Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| **Structured**    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| JSON Schema       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Choice            | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Regex             | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
+| Grammar           | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
+
+
+## Caveats
+
+- OpenAI doesn't support structured generation due to limitations in their API and server implementation.
+- `outlines.generate` ["Structured"](../generation/generation.md) includes methods such as `outlines.generate.regex`, `outlines.generate.json`, `outlines.generate.cfg`, etc.
+- MLXLM only supports Apple Silicon.
+- llama.cpp greedy sampling available via multinomial with `temperature = 0.0`.
diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
new file mode 100644
index 000000000..0777ad8b7
--- /dev/null
+++ b/docs/reference/models/openai.md
@@ -0,0 +1,204 @@
+# OpenAI and compatible APIs
+
+!!! Installation
+
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively:
+
+    ```bash
+    pip install "outlines[openai]"
+    ```
+
+## OpenAI models
+
+Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.openai`:
+
+```python
+from outlines import models
+
+
+model = models.openai("gpt-4o-mini")
+model = models.openai("gpt-4o")
+```
+
+Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.AsyncOpenAI` as keyword arguments:
+
+```python
+import os
+from outlines import models
+
+
+model = models.openai(
+    "gpt-4o-mini",
+    api_key=os.environ["OPENAI_API_KEY"]
+)
+```
+
+The following table enumerates the possible parameters. Refer to the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list.
+
+**Parameters:**
+
+| **Parameters** | **Type** | **Description** | **Default** |
+|----------------|:---------|:----------------|:------------|
+| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
+| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
+| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
+| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if no specified. | `None` |
+| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
+| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
+| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
+| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
+| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
+
+## Azure OpenAI models
+
+Outlines also supports Azure OpenAI models:
+
+```python
+from outlines import models
+
+
+model = models.azure_openai(
+    "azure-deployment-name",
+    "gpt-4o-mini",
+    api_version="2024-07-18",
+    azure_endpoint="https://example-endpoint.openai.azure.com",
+)
+```
+
+!!! Question "Why do I need to specify model and deployment name?"
+
+    The model name is needed to load the correct tokenizer for the model. The tokenizer is necessary for structured generation.
+
+
+You can pass any parameter you would pass to `openai.AsyncAzureOpenAI`. You can consult the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/lib/azure.py) for an up-to-date list.
+
+**Parameters:**
+
+
+| **Parameters** | **Type** | **Description** | **Default** |
+|----------------|:---------|:----------------|:------------|
+| `azure_endpoint` | `str` | Azure endpoint, including the resource. Infered from `AZURE_OPENAI_ENDPOINT` if not specified | `None` |
+| `api_version` | `str` | API version. Infered from `AZURE_OPENAI_API_KEY` if not specified | `None` |
+| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
+| `azure_ad_token` | `str` | Azure active directory token. Inference from `AZURE_OPENAI_AD_TOKEN` if not specified | `None` |
+| `azure_ad_token_provider` | `AzureADTokenProvider` | A function that returns an Azure Active Directory token | `None` |
+| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
+| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
+| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if not specified. | `None` |
+| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
+| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
+| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
+| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
+| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
+
+## Models that follow the OpenAI standard
+
+Outlines supports models that follow the OpenAI standard. You will need to initialize the OpenAI client properly configured and pass it to `outlines.models.openai`
+
+```python
+import os
+from openai import AsyncOpenAI
+from outlines import models
+from outlines.models.openai import OpenAIConfig
+
+
+client = AsyncOpenAI(
+    api_key=os.environ.get("PROVIDER_KEY"),
+    base_url="http://other.provider.server.com"
+)
+config = OpenAIConfig("model_name")
+model = models.openai(client, config)
+```
+
+!!! Warning
+
+    You need to pass the async client to be able to do batch inference.
+
+## Structured Generation Support
+
+Outlines provides support for [OpenAI Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode) via `outlines.generate.json`, `outlines.generate.choice`
+
+```python
+from pydantic import BaseModel, ConfigDict
+import outlines.models as models
+from outlines import generate
+
+model = models.openai("gpt-4o-mini")
+
+class Person(BaseModel):
+    model_config = ConfigDict(extra='forbid')  # required for openai
+    first_name: str
+    last_name: str
+    age: int
+
+generator = generate.json(model, Person)
+generator("current indian prime minister on january 1st 2023")
+# Person(first_name='Narendra', last_name='Modi', age=72)
+
+generator = generate.choice(model, ["Chicken", "Egg"])
+print(generator("Which came first?"))
+# Chicken
+```
+
+!!! Warning
+
+    Structured generation support only provided to OpenAI-compatible endpoints which conform to OpenAI's standard. Additionally, `generate.regex` and `generate.cfg` are not supported.
+
+
+## Advanced configuration
+
+For more advanced configuration option, such as support proxy, please consult the [OpenAI SDK's documentation](https://github.com/openai/openai-python):
+
+
+```python
+from openai import AsyncOpenAI, DefaultHttpxClient
+from outlines import models
+from outlines.models.openai import OpenAIConfig
+
+
+client = AsyncOpenAI(
+    base_url="http://my.test.server.example.com:8083",
+    http_client=DefaultHttpxClient(
+        proxies="http://my.test.proxy.example.com",
+        transport=httpx.HTTPTransport(local_address="0.0.0.0"),
+    ),
+)
+config = OpenAIConfig("model_name")
+model = models.openai(client, config)
+```
+
+It is possible to specify the values for `seed`, `presence_penalty`, `frequence_penalty`, `top_p` by passing an instance of `OpenAIConfig` when initializing the model:
+
+```python
+from outlines.models.openai import OpenAIConfig
+from outlines import models
+
+
+config = OpenAIConfig(
+    presence_penalty=1.,
+    frequency_penalty=1.,
+    top_p=.95,
+    seed=0,
+)
+model = models.openai("gpt-4o-mini", config)
+```
+
+## Monitoring API use
+
+It is important to be able to track your API usage when working with OpenAI's API. The number of prompt tokens and completion tokens is directly accessible via the model instance:
+
+```python
+from openai import AsyncOpenAI
+import outlines.models
+
+
+model = models.openai("gpt-4o")
+
+print(model.prompt_tokens)
+# 0
+
+print(model.completion_tokens)
+# 0
+```
+
+These numbers are updated every time you call the model.
diff --git a/docs/reference/models/tgi.md b/docs/reference/models/tgi.md
new file mode 100644
index 000000000..3f0568103
--- /dev/null
+++ b/docs/reference/models/tgi.md
@@ -0,0 +1,3 @@
+# Text-generation-inference (TGI)
+
+TGI uses Outlines to provide structured generation, see [their documentation](https://huggingface.co/docs/text-generation-inference/en/basic_tutorials/using_guidance).
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
new file mode 100644
index 000000000..3649b9b13
--- /dev/null
+++ b/docs/reference/models/transformers.md
@@ -0,0 +1,148 @@
+# transformers
+
+
+!!! Installation
+
+    You need to install the `transformer`, `datasets` and `torch` libraries to be able to use these models in Outlines, or alternatively:
+
+    ```bash
+    pip install "outlines[transformers]"
+    ```
+
+
+Outlines provides an integration with the `torch` implementation of causal models in the [transformers][transformers] library. You can initialize the model by passing its name:
+
+```python
+from outlines import models
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+```
+
+If you need more fine-grained control you can also initialize the model and tokenizer separately:
+
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from outlines import models
+
+llm = AutoModelForCausalLM.from_pretrained("gpt2", output_attentions=True)
+tokenizer = AutoTokenizer.from_pretrained("gpt2")
+model = models.Transformers(llm, tokenizer)
+```
+
+## Using Logits Processors
+
+There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
+
+1. Use Outlines generation wrapper, `outlines.models.transformers`
+2. Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
+
+Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
+
+### Using `outlines.models.transformers`
+
+```python
+import outlines
+
+time_regex_pattern = r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?"
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+generator = outlines.generate.regex(model, time_regex_pattern)
+
+output = generator("The the best time to visit a dentist is at ")
+print(output)
+# 2:30 pm
+```
+
+### Using models initialized via the `transformers`  library
+
+```python
+import outlines
+import transformers
+
+
+model_uri = "microsoft/Phi-3-mini-4k-instruct"
+
+outlines_tokenizer = outlines.models.TransformerTokenizer(
+    transformers.AutoTokenizer.from_pretrained(model_uri)
+)
+phone_number_logits_processor = outlines.processors.RegexLogitsProcessor(
+    "\\+?[1-9][0-9]{7,14}",  # phone number pattern
+    outlines_tokenizer,
+)
+
+generator = transformers.pipeline('text-generation', model=model_uri)
+
+output = generator(
+    "Jenny gave me her number it's ",
+	logits_processor=transformers.LogitsProcessorList([phone_number_logits_processor])
+)
+print(output)
+# [{'generated_text': "Jenny gave me her number it's 2125550182"}]
+# not quite 8675309 what we expected, but it is a valid phone number
+```
+
+[transformers]: https://github.com/huggingface/transformers
+
+
+## Alternative Model Classes
+
+`outlines.models.transformers` defaults to `transformers.AutoModelForCausalLM`, which is the appropriate class for most standard large language models, including Llama 3, Mistral, Phi-3, etc.
+
+However other variants with unique behavior can be used as well by passing the appropriate class.
+
+### Mamba
+
+[Mamba](https://github.com/state-spaces/mamba) is a transformers alternative which employs memory efficient, linear-time decoding.
+
+To use Mamba with outlines you must first install the necessary requirements:
+```
+pip install causal-conv1d>=1.2.0 mamba-ssm torch transformers
+```
+
+Then you can either create an Mamba-2 Outlines model via
+```python
+import outlines
+
+model = outlines.models.mamba("state-spaces/mamba-2.8b-hf")
+```
+
+or explicitly with
+```python
+import outlines
+from transformers import MambaForCausalLM
+
+model = outlines.models.transformers(
+    "state-spaces/mamba-2.8b-hf",
+    model_class=MambaForCausalLM
+)
+```
+
+
+
+Read [`transformers`'s documentation](https://huggingface.co/docs/transformers/en/model_doc/mamba) for more information.
+
+### Encoder-Decoder Models
+
+You can use encoder-decoder (seq2seq) models like T5 and BART with Outlines.
+
+Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
+
+T5 Example:
+```python
+import outlines
+from transformers import AutoModelForSeq2SeqLM
+
+model_pile_t5 = outlines.models.transformers(
+    model_name="EleutherAI/pile-t5-large",
+    model_class=AutoModelForSeq2SeqLM,
+)
+```
+
+Bart Example:
+```python
+model_bart = outlines.models.transformers(
+    model_name="facebook/bart-large",
+    model_class=AutoModelForSeq2SeqLM,
+)
+```
diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
new file mode 100644
index 000000000..451e54a8f
--- /dev/null
+++ b/docs/reference/models/transformers_vision.md
@@ -0,0 +1,115 @@
+# Transformers Vision
+
+Outlines allows seamless use of [vision models](https://huggingface.co/learn/computer-vision-course/en/unit4/multimodal-models/tasks-models-part1).
+
+`outlines.models.transformers_vision` shares interfaces with, and is based on [outlines.models.transformers](./transformers.md).
+
+Tasks supported include
+
+- image + text -> text
+- video + text -> text
+
+
+
+## Example: Using [Llava-Next](https://huggingface.co/docs/transformers/en/model_doc/llava_next) Vision Models
+
+Install dependencies
+`pip install torchvision pillow flash-attn`
+
+Create the model
+```python
+import outlines
+from transformers import LlavaNextForConditionalGeneration
+
+model = outlines.models.transformers_vision(
+    "llava-hf/llava-v1.6-mistral-7b-hf",
+    model_class=LlavaNextForConditionalGeneration,
+	device="cuda",
+)
+```
+
+Create convenience function to load a `PIL.Image` from URL
+```python
+from PIL import Image
+from io import BytesIO
+from urllib.request import urlopen
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+```
+
+### Describing an image
+
+```python
+description_generator = outlines.generate.text(model)
+description_generator(
+    "<image> detailed description:",
+    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")]
+)
+```
+
+> This is a color photograph featuring a Siamese cat with striking blue eyes. The cat has a creamy coat and a light eye color, which is typical for the Siamese breed. Its features include elongated ears, a long, thin tail, and a striking coat pattern. The cat is sitting in an indoor setting, possibly on a cat tower or a similar raised platform, which is covered with a beige fabric, providing a comfortable and soft surface for the cat to rest or perch. The surface of the wall behind the cat appears to be a light-colored stucco or plaster.
+
+#### Multiple Images
+
+To include multiple images in your prompt you simply add more `<image>` tokens to the prompt
+
+```python
+image_urls = [
+	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-3.png",  # triangle
+	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-11.png",  # hexagon
+]
+description_generator = outlines.generate.text(model)
+description_generator(
+    "<image><image>What shapes are present?",
+    list(map(img_from_url, image_urls)),
+)
+```
+
+> There are two shapes present. One shape is a hexagon and the other shape is an triangle. '
+
+
+### Classifying an Image
+
+```python
+pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
+planet_generator = outlines.generate.regex(model, pattern)
+
+planet_generator(
+    "What planet is this: <image>",
+    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg")]
+)
+```
+
+> Saturn
+
+
+### Extracting Structured Image data
+
+```python
+from pydantic import BaseModel
+from typing import List, Optional
+
+class ImageData(BaseModel):
+    caption: str
+    tags_list: List[str]
+    object_list: List[str]
+    is_photo: bool
+
+image_data_generator = outlines.generate.json(model, ImageData)
+
+image_data_generator(
+    "<image> detailed JSON metadata:",
+    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg")]
+)
+```
+
+> `ImageData(caption='An astronaut on the moon', tags_list=['moon', 'space', 'nasa', 'americanflag'], object_list=['moon', 'moon_surface', 'space_suit', 'americanflag'], is_photo=True)`
+
+
+## Resources
+
+### Choosing a model
+- https://mmbench.opencompass.org.cn/leaderboard
+- https://huggingface.co/spaces/WildVision/vision-arena
diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
new file mode 100644
index 000000000..8789b588e
--- /dev/null
+++ b/docs/reference/models/vllm.md
@@ -0,0 +1,235 @@
+# vLLM
+
+
+!!! Note "Installation"
+
+    You need to install the `vllm` library to use the vLLM integration. See the [installation section](#installation) for instructions to install vLLM for CPU or ROCm. To get started you can also run:
+
+    ```bash
+    pip install "outlines[vllm]"
+    ```
+
+## Load the model
+
+Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
+
+
+```python
+from outlines import models
+
+model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
+```
+
+Or alternatively:
+
+```python
+import vllm
+from outlines import models
+
+llm = vllm.LLM("microsoft/Phi-3-mini-4k-instruct")
+model = models.VLLM(llm)
+```
+
+
+Models are loaded from the [HuggingFace hub](https://huggingface.co/).
+
+
+!!! Warning "Device"
+
+    The default installation of vLLM only allows to load models on GPU. See the [installation instructions](#installation) to run models on CPU.
+
+
+You can pass any parameter that you would normally pass to `vllm.LLM`, as keyword arguments:
+
+```python
+from outlines import models
+
+model = models.vllm(
+    "microsoft/Phi-3-mini-4k-instruct",
+    trust_remote_code=True,
+    gpu_memory_utilization=0.7
+)
+```
+
+**Main parameters:**
+
+| **Parameters** | **Type** | **Description** | **Default** |
+|----------------|:---------|:----------------|:------------|
+| `tokenizer_mode`| `str`  | "auto" will use the fast tokenizer if available and "slow" will always use the slow tokenizer. | `auto`
+| `trust_remote_code`| `bool` | Trust remote code when downloading the model and tokenizer. | `False` |
+| `tensor_parallel_size`| `int` | The number of GPUs to use for distributed execution with tensor parallelism.| `1` |
+| `dtype`| `str` | The data type for the model weights and activations. Currently, we support `float32`, `float16`, and `bfloat16`. If `auto`, we use the `torch_dtype` attribute specified in the model config file. However, if the `torch_dtype` in the config is `float32`, we will use `float16` instead.| `auto` |
+| `quantization`| `Optional[str]` | The method used to quantize the model weights. Currently, we support "awq", "gptq" and "squeezellm". If None, we first check the `quantization_config` attribute in the model config file. If that is None, we assume the model weights are not quantized and use `dtype` to determine the data type of the weights.| `None` |
+| `revision`| `Optional[str]` | The specific model version to use. It can be a branch name, a tag name, or a commit id.| `None` |
+| `tokenizer_revision`| `Optional[str]`| The specific tokenizer version to use. It can be a branch name, a tag name, or a commit id.| `None` |
+| `gpu_memory_utilization`| `float` | The ratio (between 0 and 1) of GPU memory to reserve for the model weights, activations, and KV cache. Higher values will increase the KV cache size and thus improve the model's throughput. However, if the value is too high, it may cause out-of-memory (OOM) errors.| `0.9` |
+| `swap_space`| `int` | The size (GiB) of CPU memory per GPU to use as swap space. This can be used for temporarily storing the states of the requests when their `best_of` sampling parameters are larger than 1. If all requests will have `best_of=1`, you can safely set this to 0. Otherwise, too small values may cause out-of-memory (OOM) errors.| 4 |
+| `enforce_eager`| `bool` | Whether to enforce eager execution. If True, we will disable CUDA graph and always execute the model in eager mode. If False, we will use CUDA graph and eager execution in hybrid.| `False` |
+| `enable_lora` | `bool` | Whether to enable loading LoRA adapters | `False` |
+
+See the [vLLM code](https://github.com/vllm-project/vllm/blob/8f44facdddcf3c704f7d6a2719b6e85efc393449/vllm/entrypoints/llm.py#L72) for a list of all the available parameters.
+
+### Use quantized models
+
+vLLM supports AWQ, GPTQ and SqueezeLLM quantized models:
+
+
+```python
+from outlines import models
+
+model = models.vllm("TheBloke/Llama-2-7B-Chat-AWQ", quantization="awq")
+model = models.vllm("TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", quantization="gptq")
+model = models.vllm("https://huggingface.co/squeeze-ai-lab/sq-llama-30b-w4-s5", quantization="squeezellm")
+```
+
+!!! Warning "Dependencies"
+
+    To use AWQ model you need to install the autoawq library `pip install autoawq`.
+
+    To use GPTQ models you need to install the autoGTPQ and optimum libraries `pip install auto-gptq optimum`.
+
+
+### Multi-GPU usage
+
+To run multi-GPU inference with vLLM you need to set the `tensor_parallel_size` argument to the number of GPUs available when initializing the model. For instance to run inference on 2 GPUs:
+
+
+```python
+from outlines import models
+
+model = models.vllm(
+    "microsoft/Phi-3-mini-4k-instruct"
+    tensor_parallel_size=2
+)
+```
+
+### Load LoRA adapters
+
+You can load LoRA adapters and alternate between them dynamically:
+
+```python
+from outlines import models
+
+model = models.vllm("facebook/opt-350m", enable_lora=True)
+model.load_lora("ybelkaa/opt-350m-lora")  # Load LoRA adapter
+model.load_lora(None)  # Unload LoRA adapter
+```
+
+## Generate text
+
+In addition to the parameters described in the [text generation section](../text.md) you can pass an instance of `SamplingParams` directly to any generator via the `sampling_params` keyword argument:
+
+```python
+from vllm.sampling_params import SamplingParams
+from outlines import models, generate
+
+
+model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+
+params = SamplingParams(n=2, frequency_penalty=1., min_tokens=2)
+answer = generator("A prompt", sampling_params=params)
+```
+
+This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
+
+!!! Note
+
+    The values passed via the `SamplingParams` instance supersede the other arguments to the generator or the samplers.
+
+**`SamplingParams` attributes:**
+
+| Parameters | Type             | Description            | Default |
+|:-----------|------------------|:-----------------------|---------|
+| `n` | `int` | Number of output sequences to return for the given prompt. | `1` |
+| `best_of` | `Optional[int]` | Number of output sequences that are generated from the prompt. From these `best_of` sequences, the top `n` sequences are returned. `best_of` must be greater than or equal to `n`. This is treated as the beam width when `use_beam_search` is True. By default, `best_of` is set to `n`. | `None` |
+| `presence_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens.| `0.0` |
+| `frequency_penalty` | `float` | Float that penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens. | `0.0`
+| `repetition_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the prompt and the generated text so far. Values > 1 encourage the model to use new tokens, while values < 1 encourage the model to repeat tokens. | `1.0` |
+| `temperature` | `float` | Float that controls the randomness of the sampling. Lower values make the model more deterministic, while higher values make the model more random. Zero means greedy sampling. | `1.0` |
+| `top_p` | `float` |  Float that controls the cumulative probability of the top tokens to consider. Must be in (0, 1]. Set to 1 to consider all tokens. | `1.0` |
+| `top_k` | `int` |  Integer that controls the number of top tokens to consider. Set to -1 to consider all tokens. | `-1` |
+| `min_p` |`float` |  Float that represents the minimum probability for a token to be considered, relative to the probability of the most likely token. Must be in [0, 1]. Set to 0 to disable this. | `0.0` |
+| `seed` | `Optional[int]` | Random seed to use for the generation. | `None` |
+| `use_beam_search` | `bool` |  Whether to use beam search instead of sampling. | `False` |
+| `length_penalty` | `float` | Float that penalizes sequences based on their length. Used in beam search.  | `1.0` |
+| `early_stopping` | `Union[bool, str]` |  Controls the stopping condition for beam search. It accepts the following values: `True`, where the generation stops as soon as there are `best_of` complete candidates; `False`, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; `"never"`, where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm). | `False` |
+| `stop` | `Optional[Union[str, List[str]]]` |  List of strings that stop the generation when they are generated. The returned output will not contain the stop strings. | `None` |
+| `stop_token_ids` | `Optional[List[int]]` |  List of tokens that stop the generation when they are generated. The returned output will contain the stop tokens unless the stop tokens are special tokens. | `None` |
+| `include_stop_str_in_output` | `bool` |  Whether to include the stop strings in output text. Defaults to False. | `False` |
+| `ignore_eos` | `bool` |  Whether to ignore the EOS token and continue generating tokens after the EOS token is generated. | `False` |
+| `max_tokens` | `int` |  Maximum number of tokens to generate per output sequence. | `16` |
+| `min_tokens` | `int` | Minimum number of tokens to generate per output sequence before EOS or stop_token_ids can be generated | `0` |
+| `skip_special_tokens` | `bool` | Whether to skip special tokens in the output. | `True` |
+| `spaces_between_special_tokens` | `bool` |  Whether to add spaces between special tokens in the output.  Defaults to True. | `True` |
+
+### Streaming
+
+!!! Warning
+
+    Streaming is not available for the offline vLLM integration.
+
+
+## Installation
+
+By default the vLLM library is installed with pre-commpiled C++ and CUDA binaries and will only run on GPU:
+
+```python
+pip install vllm
+```
+
+### CPU
+
+You need to have the `gcc` compiler installed on your system. Then you will need to install vLLM from source. First clone the repository:
+
+```bash
+git clone https://github.com/vllm-project/vllm.git
+cd vllm
+```
+
+Install the Python packages needed for the installation:
+
+```bash
+pip install --upgrade pip
+pip install wheel packaging ninja setuptools>=49.4.0 numpy
+pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+and finally run:
+
+```bash
+VLLM_TARGET_DEVICE=cpu python setup.py install
+```
+
+See the [vLLM documentation][vllm-install-cpu] for more details, alternative installation methods (Docker) and performance tips.
+
+### ROCm
+
+
+You will need to install vLLM from source. First install Pytorch on ROCm:
+
+```bash
+pip install torch==2.2.0.dev20231206+rocm5.7 --index-url https://download.pytorch.org/whl/nightly/rocm5.7 # tested version
+```
+
+You will then need to install flash attention for ROCm following [these instructions][rocm-flash-attention]. You can then install `xformers=0.0.23` and apply the patches needed to adapt Flash Attention for ROCm:
+
+```bash
+pip install xformers==0.0.23 --no-deps
+bash patch_xformers.rocm.sh
+```
+
+And finally build vLLM:
+
+```bash
+cd vllm
+pip install -U -r requirements-rocm.txt
+python setup.py install # This may take 5-10 minutes.
+```
+
+See the [vLLM documentation][vllm-install-rocm] for alternative installation methods (Docker).
+
+
+[vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
+[vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
+[rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
diff --git a/docs/reference/prompting.md b/docs/reference/prompting.md
new file mode 100644
index 000000000..986ca82fc
--- /dev/null
+++ b/docs/reference/prompting.md
@@ -0,0 +1,395 @@
+# Prompt templating
+
+Outlines provides a powerful domain-specific language to write and manage
+prompts, via what we call *prompt functions*.  Prompt functions are Python
+functions that contain a template for the prompt in their docstring, and their
+arguments correspond to the variables used in the prompt. When called, a prompt
+function returns the template rendered with the values of the arguments.
+
+The aim of prompt functions is to solve several recurrent problems with prompting:
+
+1. **Building complex prompts quickly leads to messy code.** This problem has
+   already been solved in the web development community by using templating, so
+   why not use it here?
+2. **Composing prompts is difficult.** Why not just compose functions?
+3. **Separating prompts from code.** Encapsulation in functions allows a clean
+   separation between prompts and code. Moreover, like any function, prompt
+   functions can be imported from other modules.
+
+Outlines uses the [Jinja templating
+engine](https://jinja.palletsprojects.com/en/3.1.x/) to render prompts, which
+allows to easily compose complex prompts.
+
+!!! warning "Prompt rendering"
+
+    Prompt functions are opinionated when it comes to prompt rendering. These opinions are meant to avoid common prompting errors, but can have unintended consequences if you are doing something unusual. We advise to always print the prompt before using it. You can also [read the
+    reference](#formatting-conventions) section if you want to know more.
+
+## Your first prompt
+
+The following snippet showcases a very simple prompt. The variables between
+curly brackets `{{  }}` are placeholders for the values of the arguments you
+will pass to the prompt function.
+
+=== "Code"
+
+    ```python title="greetings.py"
+    from outlines import Template
+
+    prompt = """Hello, {{ name }}!
+    {{ question }}"""
+
+    greetings = Template.from_string(prompt)
+    prompt = greetings("user", "How are you?")
+    print(prompt)
+    ```
+
+=== "Output"
+
+    ```text
+    Hello, user!
+    How are you?
+    ```
+
+If a variable is missing in the function's arguments, Jinja2 will throw an `UndefinedError` exception:
+
+=== "Code"
+
+    ```python
+    from outlines import Template
+
+    prompt = """Hello, {{ surname }}!"""
+    greetings = Template.from_string(prompt)
+    prompt = greetings("user")
+    ```
+
+=== "Output"
+
+    ```text
+    Traceback (most recent call last):
+      File "<stdin>", line 9, in <module>
+      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 38, in __call__
+          return render(self.template, **bound_arguments.arguments)
+      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 213, in render
+          return jinja_template.render(**values)
+      File "/home/remi/micromamba/envs/outlines/lib/python3.9/site-packages/jinja2/environment.py", line 1301, in render
+          self.environment.handle_exception()
+      File "/home/remi/micromamba/envs/outlines/lib/python3.9/site-packages/jinja2/environment.py", line 936, in handle_exception
+          raise rewrite_traceback_stack(source=source)
+      File "<template>", line 1, in top-level template code
+      jinja2.exceptions.UndefinedError: 'surname' is undefined
+    ```
+
+## Importing prompts from files
+
+Outlines allows you to read a prompt template from a text file. This way you can build "white space perfect" prompts, and version them independently from your code. We have found ourselves gravitating around this pattern a lot since Outlines came out:
+
+=== "prompt.txt"
+    ```text
+    """Hello, {{ name }}!
+    {{ question }}
+    """
+    ```
+
+=== "generate.py"
+
+    ```python
+    from outlines import Template
+
+    greetings = Template.from_file("prompt.txt")
+    prompt = greetings("John Doe", "How are you today?")
+    ```
+
+=== "Output"
+
+    ```text
+    Hello, John Doe!
+    How are you today?
+    ```
+
+## Few-shot prompting
+
+Few-shot prompting can lead to messy code. Prompt functions allow you to loop
+over lists or dictionaries from the template. In the following example we
+demonstrate how we can generate a prompt by passing a list of dictionaries with
+keys `question` and `answer` to the prompt function:
+
+=== "Code"
+
+    ```text title="prompt.txt"
+    {{ instructions }}
+
+    Examples
+    --------
+
+    {% for example in examples %}
+    Q: {{ example.question }}
+    A: {{ example.answer }}
+
+    {% endfor %}
+    Question
+    --------
+
+    Q: {{ question }}
+    A:
+    ```
+
+    ```python title="render.py"
+    from outlines import Template
+
+    instructions = "Please answer the following question following the examples"
+    examples = [
+        {"question": "2+2=?", "answer":4},
+        {"question": "3+3=?", "answer":6}
+    ]
+    question = "4+4 = ?"
+
+    few_shots = Template.from_file("prompt.txt")
+    prompt = few_shots(instructions, examples, question)
+    print(prompt)
+    ```
+
+=== "Output"
+
+    ```text
+    Please answer the following question following the examples
+
+    Examples
+    --------
+
+    Q: 2+2=?
+    A: 4
+
+    Q: 3+3=?
+    A: 6
+
+    Question
+    --------
+
+    Q: 4+4 = ?
+    A:
+    ```
+
+## Conditionals, filters, etc.
+
+Jinja2 has many features beyond looping that are not described here:
+conditionals, filtering, formatting, etc. Please refer to the [Jinja
+documentation](https://jinja.palletsprojects.com/en/3.1.x/>) for more
+information about the syntax of the templating language. The Jinja syntax is
+powerful, and we recommend you take some time to read their documentation if you
+are building complex prompts.
+
+
+## Tools
+
+Several projects (e.g.[Toolformer](https://arxiv.org/abs/2302.04761), [ViperGPT](https://viper.cs.columbia.edu/), [AutoGPT](https://github.com/Significant-Gravitas/Auto-GPT), etc.) have shown that we can "teach" language models to use external functions by describing what these functions do in the prompt. In these projects the same information is often repeated twice: the function implementation, name, docstring, or arguments are copy-pasted in the prompt. This is cumbersome and error prone; you can directly pull this information from within an Outlines prompt function:
+
+=== "Code"
+
+    ```python
+    from outlines import Template
+
+    def my_tool(arg1: str, arg2: int):
+        """Tool description.
+
+        The rest of the docstring
+        """
+        pass
+
+    prompt = """{{ question }}
+
+    COMMANDS
+    1. {{ tool | name }}: {{ tool | description }}, args: {{ tool | args }}
+
+    {{ tool | source }}
+    """
+
+    tool_prompt = Template.from_string(prompt)
+    prompt = tool_prompt("Can you do something?", my_tool)
+    print(prompt)
+    ```
+
+=== "Output"
+
+    ```text
+    Can you do something?
+
+    COMMANDS
+    1. my_tool: Tool description., args: arg1: str, arg2: int
+
+    def my_tool(arg1: str, arg2: int):
+        """Tool description.
+
+        The rest of the docstring
+        """
+        pass
+    ```
+
+## JSON response format
+
+To build reliable chains with language models we often need to instruct them the
+format in which we would like them to return their response.
+
+Without prompt templating, the information is repeated twice between creating
+the parsing function (e.g. a Pydantic model), and writing the desired schema in
+the prompt. This can lead to errors that are hard to debug.
+
+Outlines allows you to directly pull the JSON schema of a pydantic model, or
+pretty print a dictionary from within an Outlines prompt function
+
+=== "Code"
+
+    ```python
+    from pydantic import BaseModel, Field
+
+   from outlines import Template
+
+    class MyResponse(BaseModel):
+        field1: int = Field(description="an int")
+        field2: str
+
+
+    my_prompt = Template.from_string("""{{ response_model | schema }}""")
+    prompt = my_prompt(MyResponse)
+    print(prompt)
+    # {
+    #   "field1": "an int",
+    #   "field2": "<field2>"
+    # }
+    ```
+
+=== "Output"
+
+    ```python
+    response = {
+        "field1": "<field1>",
+        "field2": "a string"
+    }
+
+    my_prompt(MyResponse)
+    # {
+    #   "field1": "<field1>",
+    #   "field2": "a string"
+    # }
+    ```
+
+## Formatting conventions
+
+Prompt templates are opinionated when it comes to rendering a template read from
+a string, and these opinions are meant to avoid prompting mistakes and help with
+formatting.
+
+### Whitespaces
+
+If you have experience working with strings between triple quotes you know that
+indenting has an influence on the string's formatting. Prompt functions adopt
+a few conventions so you don't have to think about indents when writing prompt.
+
+First, whether you start the prompt right after the triple quotes or on the line
+below does not matter for formatting:
+
+=== "Code"
+
+    ```python
+    from outlines import Template
+
+
+    prompt1 = Template.from_string(
+    """My prompt
+    """
+    )
+
+    prompt2 = Template.from_string(
+    """
+    My prompt
+    """
+    )
+
+    print(prompt1())
+    print(prompt2())
+    ```
+
+=== "Output"
+
+    ```text
+    My prompt
+    My prompt
+    ```
+
+Indentation is relative to the second line of the docstring, and leading spaces are removed:
+
+=== "Code"
+
+    ```python
+    from outlines import Template
+
+    example1 = Template.from_string(
+    """First line
+    Second line
+    """
+    )
+
+    example2 = Template.from_string(
+    """
+        Second line
+        Third line
+    """
+    )
+
+    example3 = Template.from_string(
+    """
+        Second line
+          Third line
+    """
+    )
+
+    print(example1())
+    print(example2())
+    print(example3())
+    ```
+
+=== "Output"
+
+    ```text
+    First line
+    Second line
+
+    Second line
+    Third line
+
+    Second line
+      Third line
+    ```
+
+Trailing whitespaces are not removed, unless they follow a linebreak symbol `\` (see [linebreaks](#linebreaks)).
+
+### Linebreaks
+
+You can use the backslash `\` to break a long line of text. It will render as a single line:
+
+=== "Code"
+
+    ```python
+    from outlines import Template
+
+    example = Template.from_string(
+    """
+    Break in \
+    several lines \
+    But respect the indentation
+        on line breaks.
+    And after everything \
+    Goes back to normal
+    """
+    )
+
+    print(example())
+    ```
+
+=== "Output"
+
+    ```text
+    Break in several lines But respect the indentation
+        on line breaks.
+    And after everything Goes back to normal
+    ```
diff --git a/docs/features/utility/regex_dsl.md b/docs/reference/regex_dsl.md
similarity index 99%
rename from docs/features/utility/regex_dsl.md
rename to docs/reference/regex_dsl.md
index 740a0772b..f42b6dd34 100644
--- a/docs/features/utility/regex_dsl.md
+++ b/docs/reference/regex_dsl.md
@@ -1,8 +1,4 @@
----
-title: Regex DSL
----
-
-# Regex DSL
+# DSL to express constraints
 
 This library provides a Domain-Specific Language (DSL) to construct regular expressions in a more intuitive and modular way. It allows you to create complex regexes using simple building blocks that represent literal strings, patterns, and various quantifiers. Additionally, these custom regex types can be used directly as types in [Pydantic](https://pydantic-docs.helpmanual.io/) schemas to enforce pattern constraints during text generation.
 
diff --git a/docs/reference/samplers.md b/docs/reference/samplers.md
new file mode 100644
index 000000000..9cba5cefa
--- /dev/null
+++ b/docs/reference/samplers.md
@@ -0,0 +1,178 @@
+# Samplers
+
+Outlines offers different sequence sampling algorithms, and we will integrate more in the future. You can read [this blog post](https://huggingface.co/blog/how-to-generate) for an overview of the different sampling algorithm.
+
+Samplers provide control over the sampling process, allowing you to influence the output of the model. This can include controlling randomness (temperature), biasing towards certain tokens (top-k, top-p), or sequence generation (beam search).
+
+## Multinomial sampling
+
+[Multinomial sampling](https://en.wikipedia.org/wiki/Multinomial_distribution) is the default sampling algorithm in Outlines.
+
+As an example, suppose we have only two possible tokens: "H" and "T". For a fixed prompt such as "Flip a coin, did you get heads or tails?" The language model calculates probability for each token:
+
+| Token | Probability |
+|-------|-------------|
+| "H"   | 0.5         |
+| "T"   | 0.5         |
+
+You'd expect to receive "H" 50% of the time and "T" 50% of the time.
+
+### Parameters
+
+- `samples`: Number of samples to generate (default: 1)
+- `top_k`: Only consider the top k tokens (optional)
+- `top_p`: Only consider the top tokens with cumulative probability >= p (optional)
+- `temperature`: Controls randomness of sampling (optional)
+
+### Default behavior
+
+Outlines defaults to the multinomial sampler without top-p or top-k sampling, and temperature equal to 1.
+
+Not specifying a sampler is equivalent to:
+
+```python
+from outlines import models, generate, samplers
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+sampler = samplers.multinomial()
+
+generator = generate.text(model, sampler)
+answer = generator("What is 2+2?")
+
+print(answer)
+# 4
+```
+
+### Batching
+
+You can ask the generator to take multiple samples by passing the number of samples when initializing the sampler:
+
+```python
+from outlines import models, generate, samplers
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+sampler = samplers.multinomial(3)
+
+generator = generate.text(model, sampler)
+answer = generator("What is 2+2?")
+
+print(answer)
+# [4, 4, 4]
+```
+
+If you ask multiple samples for a batch of prompts the returned array will be of shape `(num_samples, num_batches)`:
+
+```python
+from outlines import models, generate, samplers
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+sampler = samplers.multinomial(3)
+
+generator = generate.text(model, sampler)
+answer = generator(["What is 2+2?", "What is 3+3?"])
+
+print(answer)
+# [[4, 4, 4], [6, 6, 6]]
+```
+
+### Temperature
+
+You can control the temperature with
+
+```python
+from outlines import models, generate, samplers
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+sampler = samplers.multinomial(3, temperature=0.5)
+
+generator = generate.text(model, sampler)
+answer = generator(["What is 2+2?", "What is 3+3?"])
+
+print(answer)
+```
+
+If you would like to use `temperature=0.0`, please use `sampler=samplers.greedy()` instead.
+
+### Top-k sampling
+
+You can ask Outlines to only consider the top-k logits at each step by specifying the value of the `top-k` keyword argument when initializing the sampler.
+
+```python
+sampler = samplers.multinomial(3, top_k=10)
+```
+
+### Top-p sampling
+
+You can ask Outlines to only consider the highest probability tokens such that their cumulative probability is greater than a threshold `p`. Specify the `top_p` keyword argument when initializing the sampler:
+
+
+```python
+sampler = samplers.multinomial(3, top_p=0.95)
+```
+
+## Greedy sampler
+
+Greedy sampling selects the token with the highest probability at each step. It's deterministic and always produces the same output for a given input.
+
+To use the greedy sampler, initialize the generator with the sampler:
+
+
+```python
+from outlines import models, generate, samplers
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+sampler = samplers.greedy()
+
+generator = generate.text(model, sampler)
+answer = generator("What is 2+2?")
+
+print(answer)
+# 4
+```
+
+You cannot ask for multiple samples with the greedy sampler since it does not clear what the result should be. Only the most likely token can be returned.
+
+
+## Beam Search
+
+Beam search maintains multiple candidate sequences at each step, potentially finding better overall sequences than greedy or multinomial sampling.
+
+To use Beam Search, initialize the generator with the sampler:
+
+```python
+from outlines import models, generate, samplers
+
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+sampler = samplers.beam_search(beams=5)
+
+generator = generate.text(model, sampler)
+answer = generator("What is 2+2?")
+
+print(answer)
+# 4
+```
+!!! Warning "Compatibility"
+
+    Only models from the `transformers`  and `exllamav2 ` libraries are compatible with Beam Search.
+
+### Parameters
+
+- `beams`: Number of beams to use (default: 1)
+
+## Sampler Comparison
+
+Here's a table comparing the different samplers:
+
+| Sampler | Pros | Cons | Use Cases |
+|---------|------|------|-----------|
+| Greedy | Deterministic, fast | May produce repetitive text | When you need consistent, predictable output |
+| Multinomial | Balances exploration and exploitation | Results may vary between runs | General-purpose text generation, creative tasks |
+| Beam Search | Can find globally better sequences | More computationally expensive | When sequence quality is critical, e.g., translation |
+
+For most use cases, we recommend using the default [multinomial sampler](#multinomial-sampling).
diff --git a/docs/reference/serve/lmstudio.md b/docs/reference/serve/lmstudio.md
new file mode 100644
index 000000000..db383186a
--- /dev/null
+++ b/docs/reference/serve/lmstudio.md
@@ -0,0 +1,89 @@
+# Serve with LM Studio
+
+!!! tip "Would rather not self-host?"
+
+    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
+
+[LM Studio](https://lmstudio.ai/) is an application that runs local LLMs. It flexibly mixes GPU and CPU compute in hardware-constrained environments.
+
+As of [LM Studio 0.3.4](https://lmstudio.ai/blog/lmstudio-v0.3.4), it natively supports Outlines for structured text generation, using an OpenAI-compatible endpoint.
+
+## Setup
+
+1. Install LM Studio by visiting their [downloads page](https://lmstudio.ai/download).
+2. Enable the LM Studio [server functionality](https://lmstudio.ai/docs/basics/server).
+3. Download [a model](https://lmstudio.ai/docs/basics#1-download-an-llm-to-your-computer).
+4. Install Python dependencies.
+```bash
+pip install pydantic openai
+```
+
+## Calling the server
+
+By default, LM Studio will serve from `http://localhost:1234`. If you are serving on a different port or host, make sure to change the `base_url` argument in `OpenAI` to the relevant location.
+
+```python
+class Testing(BaseModel):
+    """
+    A class representing a testing schema.
+    """
+    name: str
+    age: int
+
+openai_client = openai.OpenAI(
+    base_url="http://0.0.0.0:1234/v1",
+    api_key="dopeness"
+)
+
+# Make a request to the local LM Studio server
+response = openai_client.beta.chat.completions.parse(
+    model="hugging-quants/Llama-3.2-1B-Instruct-Q8_0-GGUF",
+    messages=[
+        {"role": "system", "content": "You are like so good at whatever you do."},
+        {"role": "user", "content": "My name is Cameron and I am 28 years old. What's my name and age?"}
+    ],
+    response_format=Testing
+)
+```
+
+You should receive a `ParsedChatCompletion[Testing]` object back:
+
+```python
+ParsedChatCompletion[Testing](
+    id='chatcmpl-3hykyf0fxus7jc90k6gwlw',
+    choices=[
+        ParsedChoice[Testing](
+            finish_reason='stop',
+            index=0,
+            logprobs=None,
+            message=ParsedChatCompletionMessage[Testing](
+                content='{ "age": 28, "name": "Cameron" }',
+                refusal=None,
+                role='assistant',
+                function_call=None,
+                tool_calls=[],
+                parsed=Testing(name='Cameron', age=28)
+            )
+        )
+    ],
+    created=1728595622,
+    model='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-Q4_K_M.gguf',
+    object='chat.completion',
+    service_tier=None,
+    system_fingerprint='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-
+Q4_K_M.gguf',
+    usage=CompletionUsage(
+        completion_tokens=17,
+        prompt_tokens=47,
+        total_tokens=64,
+        completion_tokens_details=None,
+        prompt_tokens_details=None
+    )
+)
+```
+
+You can retrieve your `Testing` object with
+
+```python
+response.choices[0].message.parsed
+```
diff --git a/docs/reference/serve/vllm.md b/docs/reference/serve/vllm.md
new file mode 100644
index 000000000..14277a526
--- /dev/null
+++ b/docs/reference/serve/vllm.md
@@ -0,0 +1,73 @@
+# Serve with vLLM
+
+!!! tip "Would rather not self-host?"
+
+    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
+
+Outlines can be deployed as an LLM service using the vLLM inference engine and a FastAPI server. vLLM is not installed by default so will need to install Outlines with:
+
+```bash
+pip install outlines[serve]
+```
+
+You can then start the server with:
+
+```bash
+python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
+```
+
+This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used. The `--model` argument allows you to specify any model of your choosing.
+
+To run inference on multiple GPUs you must pass the `--tensor-parallel-size` argument when initializing the server. For instance, to run inference on 2 GPUs:
+
+
+```bash
+python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct" --tensor-parallel-size 2
+```
+
+
+### Alternative Method: Via Docker
+
+You can install and run the server with Outlines' official Docker image using the command
+
+```bash
+docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
+```
+
+## Querying Endpoint
+
+You can then query the model in shell by passing a prompt and either
+
+1. a [JSON Schema][jsonschema]{:target="_blank"} specification or
+2. a [Regex][regex]{:target="_blank"} pattern
+
+with the `schema` or `regex` parameters, respectively, to the `/generate` endpoint. If both are specified, the schema will be used. If neither is specified, the generated text will be unconstrained.
+
+For example, to generate a string that matches the schema `{"type": "string"}` (any string):
+
+```bash
+curl http://127.0.0.1:8000/generate \
+    -d '{
+        "prompt": "What is the capital of France?",
+        "schema": {"type": "string", "maxLength": 5}
+        }'
+```
+
+To generate a string that matches the regex `(-)?(0|[1-9][0-9]*)(\.[0-9]+)?([eE][+-][0-9]+)?` (a number):
+
+```bash
+curl http://127.0.0.1:8000/generate \
+    -d '{
+        "prompt": "What is Pi? Give me the first 15 digits: ",
+        "regex": "(-)?(0|[1-9][0-9]*)(\\.[0-9]+)?([eE][+-][0-9]+)?"
+        }'
+```
+
+Instead of `curl`, you can also use the [requests][requests]{:target="_blank"} library from another python program.
+
+Please consult the [vLLM documentation][vllm]{:target="_blank"} for details on additional request parameters. You can also [read the code](https://github.com/dottxt-ai/outlines/blob/main/outlines/serve/serve.py) in case you need to customize the solution to your needs.
+
+[requests]: https://requests.readthedocs.io/en/latest/
+[vllm]: https://docs.vllm.ai/en/latest/index.html
+[jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
+[regex]: https://www.regular-expressions.info/tutorial.html
diff --git a/docs/reference/text.md b/docs/reference/text.md
new file mode 100644
index 000000000..f36a67da0
--- /dev/null
+++ b/docs/reference/text.md
@@ -0,0 +1,86 @@
+# Text generation
+
+Outlines provides a unified interface to generate text with many language models, API-based and local. The same pattern is used throughout the library:
+
+1. Instantiate a generator by calling `outlines.generate.text` with the model to be used.
+2. Call the generator with the prompt and (optionally) some generation parameters.
+
+
+```python
+from outlines import models, generate
+
+model = models.openai("gpt-4o-mini")
+generator = generate.text(model)
+answer = generator("What is 2+2?")
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+answer = generator("What is 2+2?")
+```
+
+By default Outlines uses the multinomial sampler with `temperature=1`. See [this section](samplers.md) to learn how to use different samplers.
+
+## Streaming
+
+Outlines allows you to stream the model's response by calling the `.stream` method of the generator with the prompt:
+
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+
+tokens = generator.stream("What is 2+2?")
+for token in tokens:
+    print(token)
+```
+
+## Parameters
+
+### Limit the number of tokens generated
+
+To limit the number of tokens generated you can pass the `max_tokens` positional argument to the generator:
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+
+answer = generator("What is 2+2?", 5)
+answer = generator("What is 2+2?", max_tokens=5)
+```
+
+### Stop after a given string is generated
+
+You can also ask the model to stop generating text after a given string has been generated, for instance a period or a line break. You can pass a string or a line of string for the `stop_at` argument:
+
+
+```python
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+
+answer = generator("What is 2+2?", stop_at=".")
+answer = generator("What is 2+2?", stop_at=[".", "\n"])
+```
+
+*The stopping string will be included in the response.*
+
+
+### Seed the generation
+
+It can be useful to seed the generation in order to get reproducible results:
+
+```python
+import torch
+from outlines import models, generate
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+
+seed = 789001
+
+answer = generator("What is 2+2?", seed=seed)
+```
diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index 381f5aa11..e71a6124f 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -1,22 +1,149 @@
 @font-face {
-    font-family: "Source Code Pro Custom", monospace;
-    src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
+  font-family: 'Source Code Pro Custom', monospace;
+  src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
 }
 
 :root > * {
-    /* Accent color shades */
-    /* --md-accent-fg-color: #d36237;
-    --md-accent-fg-color--transparent: #d36237;
-    --md-accent-bg-color: hsla(0, 0%, 100%, 1);
-    --md-accent-bg-color--light: hsla(0, 0%, 100%, 0.7); */
-
-    /* Primary color shades */
-    /* --md-accent-fg-color: hsla(#d36237, 1);
-    --md-accent-fg-color--transparent: hsla(#d36237, 0.1);
-    --md-accent-bg-color: #d36237
-    --md-accent-bg-color--light: hsla(#d36237, 0.7); */
-
-    --md-text-font-family: "Inter";
-    --md-code-font: "Source Code Pro Custom";
-    /* --md-typeset-a-color: #d36237; /*this is the brand color */
+  --md-default-bg-color: #FFFFFF;
+  --md-code-bg-color: #2E3440;
+  --md-code-fg-color: #FFFFFF;
+  --md-text-font-family: "Inter";
+  --md-code-font: "Source Code Pro Custom";
+  --md-typeset-a-color: #d16626; /*this is the brand color*/
+
+  /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
+  --md-mermaid-label-fg-color: #000000;
+  --md-mermaid-edge-color: #000000;
+}
+
+.index-pre-code {
+  max-width: 700px;
+  left: 50%;
+}
+
+.index-pre-code pre>code {
+  text-align: left;
+}
+
+.md-clipboard::after {
+  color: #FFFFFF;
+  transition: color 0.3s ease-in-out;
+}
+
+.md-clipboard:hover::after {
+  color: #D8DEE9;
+}
+
+.md-source-file {
+  text-align: center;
+  padding: 24px 0;
+}
+
+.md-typeset pre>code {
+  border-radius: .2rem;
+  box-shadow: 10px 5px 5px #D8DEE9;
+}
+
+.md-typeset p > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset strong > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-content p > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset td > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset li > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset code {
+  font-weight: 500;
+}
+
+.md-typeset pre {
+  margin-left: .5rem;
+  margin-right: .5rem;
+  margin-top: 2rem;
+  margin-bottom: 2rem;
+}
+
+.language-python {
+  background: #FFFFFF ! important
+}
+
+.language-bash {
+  background: #FFFFFF ! important
+}
+
+.language-toml {
+  background: #FFFFFF ! important
+}
+
+.language-text {
+  background: #FFFFFF ! important
+}
+
+.language-json {
+  background: #FFFFFF ! important
+}
+
+h1.title {
+  color: #FFFFFF;
+  margin: 0px 0px 5px;
+}
+
+h2.subtitle {
+  margin: 5px 0px 25px;
+  font-size: 1rem;
+  max-width: 540px;
+  margin: 0 auto;
+}
+
+.md-typeset {
+  line-height: 24px;
+  font-weight: 400;
+}
+
+.md-typeset h1 {
+  font-weight: bold;
+  color: #000000;
+}
+
+.md-typeset h2 {
+  font-weight: bold;
+  color: #000000;
+}
+
+span.md-ellipsis {
+  color: black;
+}
+
+.md-nav__link--active {
+  background-color: #ECEFF4;
+}
+
+.md-typeset ol li {
+  margin-bottom: .1rem;
+}
+
+.md-typeset ul li {
+  margin-bottom: .1rem;
 }
diff --git a/docs/welcome.md b/docs/welcome.md
new file mode 100644
index 000000000..728a30a97
--- /dev/null
+++ b/docs/welcome.md
@@ -0,0 +1,183 @@
+---
+title: Welcome to Outlines!
+---
+
+Outlines is a Python library that allows you to use Large Language Model in a simple and robust way (with structured generation). It is built by [.txt][.txt]{:target="_blank"}, and is already used in production by many companies.
+
+## What models do you support?
+
+We support [Openai](reference/models/openai.md), but the true power of Outlines is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [exllama2](reference/models/exllamav2.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
+
+## What are the main features?
+
+<div class="grid cards" markdown>
+-   :material-code-json:{ .lg .middle } __Make LLMs generate valid JSON__
+
+    ---
+
+    No more invalid JSON outputs, 100% guaranteed
+
+    [:octicons-arrow-right-24: Generate JSON](reference/generation/json.md)
+
+-   :material-keyboard-outline:{ .lg .middle } __JSON mode for vLLM__
+
+    ---
+
+    Deploy a LLM service using Outlines' JSON structured generation and vLLM
+
+    [:octicons-arrow-right-24: Deploy outlines](reference/serve/vllm.md)
+
+
+-   :material-regex:{ .lg .middle } __Make LLMs follow a Regex__
+
+    ---
+
+    Generate text that parses correctly 100% of the time
+
+    [:octicons-arrow-right-24: Guide LLMs](reference/generation/regex.md)
+
+-    :material-chat-processing-outline:{ .lg .middle } __Powerful Prompt Templating__
+
+     ---
+
+     Better manage your prompts' complexity with prompt templating
+
+    [:octicons-arrow-right-24: Learn more](reference/prompting.md)
+</div>
+
+## Why use Outlines?
+
+
+Outlines is built at [.txt][.txt] by engineers with decades of experience in software engineering, machine learning (Bayesian Statistics and NLP), and compilers. [.txt][.txt] is a VC-backed company fully focused on the topic of structured generation and is committed to make the community benefit from its experience.
+
+We are also open source veterans and have authored/maintained many libraries over the years: the [Aesara][aesara]{:target="_blank"} and [Pythological][pythological]{:target="_blank"} ecosystems, [Blackjax][blackjax]{:target="_blank"} and [Hy][hy]{:target="_blank"} among many others.
+.
+
+Outlines does not use unnecessary abstractions that tend to get in your way. We have a laser focus on reliable text generation with LLMs, a clear roadmap to push the state of the art in this area and a commitment to clean and robust code.
+
+And last but not least, unlike alternatives, Outlines' structured generation introduces **no overhead** during inference.
+
+
+## Who is using Outlines?
+
+Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) are using Outlines. Some of the prominent companies and organizations that are using Outlines include:
+
+<head>
+  <style>
+  .row {
+      display: inline-block;
+      width: 100%;
+      margin-bottom: 50px;
+      margin-top: 0px !important;
+      break-inside: avoid;
+  }
+
+  /* Create two equal columns that sits next to each other */
+  .column {
+      column-count: 3;
+      column-gap: 20px;
+      padding: 20px;
+  }
+
+  </style>
+</head>
+<body>
+
+<div class="column">
+  <div class="row"><img src="../logos/amazon.png" width="200"></div>
+  <div class="row"><img src="../logos/apple.png" width="200"></div>
+  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
+  <div class="row"><img src="../logos/canoe.png" width="200"></div>
+  <div class="row"><img src="../logos/cisco.png" width="200"></div>
+  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
+  <div class="row"><img src="../logos/databricks.png" width="200"></div>
+  <div class="row"><img src="../logos/datadog.png" width="200"></div>
+  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
+  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
+  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
+  <div class="row"><img src="../logos/harvard.png" width="200"></div>
+  <div class="row"><img src="../logos/hf.png" width="200"></div>
+  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
+  <div class="row"><img src="../logos/meta.png" width="200"></div>
+  <div class="row"><img src="../logos/mit.png" width="200"></div>
+  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
+  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
+  <div class="row"><img src="../logos/nyu.png" width="200"></div>
+  <div class="row"><img src="../logos/safran.png" width="200"></div>
+  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
+  <div class="row"><img src="../logos/shopify.png" width="200"></div>
+  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
+  <div class="row"><img src="../logos/tinder.png" width="200"></div>
+  <div class="row"><img src="../logos/upenn.png" width="200"></div>
+</div>
+
+</body>
+
+Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
+
+Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
+
+## Philosophy
+
+**Outlines**  is a library for neural text generation. You can think of it as a
+more flexible replacement for the `generate` method in the
+[transformers](https://github.com/huggingface/transformers) library.
+
+**Outlines**  helps developers *structure text generation* to build robust
+interfaces with external systems. It provides generation methods that
+guarantee that the output will match a regular expressions, or follow
+a JSON schema.
+
+**Outlines**  provides *robust prompting primitives* that separate the prompting
+from the execution logic and lead to simple implementations of few-shot
+generations, ReAct, meta-prompting, agents, etc.
+
+**Outlines**  is designed as a *library* that is meant to be compatible the
+broader ecosystem, not to replace it. We use as few abstractions as possible,
+and generation can be interleaved with control flow, conditionals, custom Python
+functions and calls to other libraries.
+
+**Outlines**  is *compatible with every auto-regressive model*. It only interfaces with models
+via the next-token logits distribution.
+
+## Outlines people
+
+Outlines would not be what it is today without a community of dedicated developers:
+
+<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
+</a>
+
+## Acknowledgements
+
+<div class="grid" markdown>
+
+
+<figure markdown>
+  <a href="http://www.dottxt.co">
+  ![Normal Computing logo](assets/images/dottxt.png){ width="150" }
+  </a>
+</figure>
+
+<figure markdown>
+  <a href="https://www.normalcomputing.ai">
+  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
+  </a>
+</figure>
+
+</div>
+
+Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
+
+
+[discord]: https://discord.gg/R9DSu34mGd
+[aesara]: https://github.com/aesara-devs
+[blackjax]: https://github.com/blackjax-devs/blackjax
+[pythological]: https://github.com/pythological
+[hy]: https://hylang.org/
+[.txt]: https://dottxt.co
+[vllm]: https://github.com/vllm-project/vllm
+[tgi]: https://github.com/huggingface/text-generation-inference
+[lorax]: https://github.com/predibase/lorax
+[xinference]: https://github.com/xorbitsai/inference
+[sglang]: https://github.com/sgl-project/sglang/
diff --git a/examples/babyagi.py b/examples/babyagi.py
index 4a817906d..a8f1de8ba 100644
--- a/examples/babyagi.py
+++ b/examples/babyagi.py
@@ -8,14 +8,13 @@
 from collections import deque
 from typing import Deque, List
 
-from openai import OpenAI
-
 import outlines
+import outlines.models as models
 from outlines import Template
 
 
-model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
-complete = outlines.Generator(model)
+model = models.openai("gpt-4o-mini")
+complete = outlines.generate.text(model)
 
 ## Load the prompts
 perform_task_ppt = Template.from_file("prompts/babyagi_perform_task.txt")
@@ -77,14 +76,11 @@ def one_cycle(objective: str, task_list, next_task_id: int):
 
     task = task_list.popleft()
 
-    prompt = perform_task_ppt(objective=objective, task=task)
+    prompt = perform_task_ppt(objective, task)
     result = complete(prompt)
 
     prompt = create_tasks_ppt(
-        objective=objective,
-        task=first_task["task_name"],
-        result=result,
-        previous_tasks=[first_task["task_name"]],
+        objective, first_task["task_name"], result, [first_task["task_name"]]
     )
     new_tasks = complete(prompt)
 
@@ -95,9 +91,7 @@ def one_cycle(objective: str, task_list, next_task_id: int):
         task_list.append({"task_id": next_task_id, "task_name": task})
 
     prompt = prioritize_tasks_ppt(
-        objective=objective,
-        tasks=[task["task_name"] for task in task_list],
-        next_task_id=next_task_id,
+        objective, [task["task_name"] for task in task_list], next_task_id
     )
     prioritized_tasks = complete(prompt)
 
diff --git a/examples/beam-cloud/app.py b/examples/beam-cloud/app.py
index 781bc1d3d..fb6c2cb2b 100644
--- a/examples/beam-cloud/app.py
+++ b/examples/beam-cloud/app.py
@@ -1,21 +1,14 @@
-from typing import Literal
-
 from beam import Image, endpoint, env
 
-
 if env.is_remote():
     import outlines
 
 
 # Pre-load models when the container first starts
 def load_models():
-    from transformers import AutoModelForCausalLM, AutoTokenizer
     import outlines
 
-    model = outlines.models.from_transformers(
-        AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-        AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    )
+    model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
     return model
 
 
@@ -41,6 +34,6 @@ def predict(context, **inputs):
     # Unpack cached model from context
     model = context.on_start_value
     # Inference
-    generator = outlines.Generator(model, Literal["Positive", "Negative"])
+    generator = outlines.generate.choice(model, ["Positive", "Negative"])
     answer = generator(prompt)
     return {"answer": answer}
diff --git a/examples/bentoml/service.py b/examples/bentoml/service.py
index 370bac901..98d46894a 100644
--- a/examples/bentoml/service.py
+++ b/examples/bentoml/service.py
@@ -1,7 +1,7 @@
 import typing as t
 
 import bentoml
-from import_model import BENTO_MODEL_TAG, MODEL_ID
+from import_model import BENTO_MODEL_TAG
 
 DEFAULT_SCHEMA = """{
     "title": "Character",
@@ -55,17 +55,13 @@ class Outlines:
 
     def __init__(self) -> None:
         import torch
-        from transformers import AutoModelForCausalLM, AutoTokenizer
 
         import outlines
 
-        self.model = outlines.from_transformers(
-            AutoTokenizer.from_pretrained(MODEL_ID),
-            AutoModelForCausalLM.from_pretrained(
-                MODEL_ID,
-                torch_dtype=torch.float16,
-                low_cpu_mem_usage=True,
-            )
+        self.model = outlines.models.transformers(
+            self.bento_model_ref.path,
+            device="cuda",
+            model_kwargs={"torch_dtype": torch.float16},
         )
 
     @bentoml.api
@@ -76,7 +72,7 @@ async def generate(
     ) -> t.Dict[str, t.Any]:
         import outlines
 
-        generator = outlines.Generator(self.model, outlines.json_schema(json_schema))
+        generator = outlines.generate.json(self.model, json_schema)
         character = generator(prompt)
 
         return character
diff --git a/examples/cerebrium/main.py b/examples/cerebrium/main.py
index 4e1987197..b61cdb6e5 100644
--- a/examples/cerebrium/main.py
+++ b/examples/cerebrium/main.py
@@ -1,13 +1,6 @@
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
 import outlines
 
-
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
-    AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
-)
-
+model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
 
 schema = {
     "title": "Character",
@@ -36,13 +29,14 @@
     },
 }
 
+generator = outlines.generate.json(model, schema)
+
 
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
-    character = model(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
-        outlines.json_schema(schema),
+    character = generator(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
     )
 
     print(character)
diff --git a/examples/cfg.py b/examples/cfg.py
index bd3b58a2e..99edf35f1 100644
--- a/examples/cfg.py
+++ b/examples/cfg.py
@@ -1,8 +1,5 @@
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-import outlines
-from outlines.types import CFG
-
+import outlines.generate as generate
+import outlines.models as models
 
 nlamb_grammar = r"""
     start: sentence
@@ -78,20 +75,15 @@
     %ignore WS
 """
 
-model_name = "hf-internal-testing/tiny-random-gpt2"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name),
-)
-
+model = models.transformers("hf-internal-testing/tiny-random-gpt2")
 batch_size = 10
 for grammar in [nlamb_grammar, calc_grammar, dyck_grammar, json_grammar]:
-    generator = outlines.Generator(model, CFG(grammar))
-    sequences = generator([" "] * batch_size, max_tokens=model.model.config.n_positions)
+    generator = generate.cfg(model, grammar, max_tokens=model.model.config.n_positions)
+    sequences = generator([" "] * batch_size)
     for seq in sequences:
         try:
-            parse = generator.logits_processor.guide.parser.parse(seq)
+            parse = generator.fsm.parser.parse(seq)
             assert parse is not None
             print("SUCCESS", seq)
-        except Exception:  # will also fail if goes over max_new_tokens / context window
+        except Exception:  # will also fail if goes over max_tokens / context window
             print("FAILURE", seq)
diff --git a/examples/dating_profile.py b/examples/dating_profile.py
index ff488698e..d26984223 100644
--- a/examples/dating_profile.py
+++ b/examples/dating_profile.py
@@ -6,7 +6,7 @@
 from pydantic import BaseModel, conlist
 
 import outlines
-from outlines import Template
+from outlines import models, Template
 
 
 class QuestionChoice(str, Enum):
@@ -86,17 +86,26 @@ class Example:
 # Below requires ~13GB of GPU memory
 # https://huggingface.co/mosaicml/mpt-7b-8k-instruct
 # Motivation: Reasonably large model that fits on a single GPU and has been fine-tuned for a larger context window
-model_name = "mosaicml/mpt-7b-8k-instruct"
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name),
-    transformers.AutoTokenizer.from_pretrained(model_name),
+config = transformers.AutoConfig.from_pretrained(
+    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
+)
+config.init_device = "meta"
+model = models.transformers(
+    model_name="mosaicml/mpt-7b-8k-instruct",
+    device="cuda",
+    model_kwargs={
+        "config": config,
+        "trust_remote_code": True,
+        "torch_dtype": torch.bfloat16,
+        "device_map": {"": 0},
+    },
 )
 
 new_description = "I'm a laid-back lawyer who spends a lot of his free-time gaming. I work in a corporate office, but ended up here after the start-up I cofounded got acquired, so still play ping pong with my cool coworkers every day. I have a bar at home where I make cocktails, which is great for entertaining friends. I secretly like to wear suits and get a new one tailored every few months. I also like weddings because I get to wear those suits, and it's a good excuse for a date. I watch the latest series because I'm paying, with my hard-earned money, for every streaming service."
 
 dating_profile_prompt = Template.from_file("prompts/dating_profile.txt")
 prompt = dating_profile_prompt(description=new_description, examples=samples)
-profile = model(prompt, outlines.json_schema(DatingProfile), max_tokens=500)  # type: ignore
+profile = outlines.generate.json(model, DatingProfile)(prompt)  # type: ignore
 print(profile)
 
 # Sample generated profiles
diff --git a/examples/llamacpp_example.py b/examples/llamacpp_example.py
index f6afc40ed..22d0da3ba 100644
--- a/examples/llamacpp_example.py
+++ b/examples/llamacpp_example.py
@@ -1,7 +1,6 @@
 from enum import Enum
 
 from pydantic import BaseModel, constr
-from llama_cpp import Llama
 
 import outlines
 
@@ -31,10 +30,10 @@ class Character(BaseModel):
 
 if __name__ == "__main__":
     # curl -L -o mistral-7b-instruct-v0.2.Q5_K_M.gguf https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q5_K_M.gguf
-    model = outlines.from_llamacpp(Llama("./mistral-7b-instruct-v0.2.Q5_K_M.gguf"))
+    model = outlines.models.llamacpp("./mistral-7b-instruct-v0.2.Q5_K_M.gguf")
 
     # Construct structured sequence generator
-    generator = outlines.Generator(model, Character)
+    generator = outlines.generate.json(model, Character)
 
     # Draw a sample
     seed = 789005
diff --git a/examples/llamacpp_processor.py b/examples/llamacpp_processor.py
index a0bf4aa3f..909df38fc 100644
--- a/examples/llamacpp_processor.py
+++ b/examples/llamacpp_processor.py
@@ -3,7 +3,7 @@
 from llama_cpp import Llama, LogitsProcessorList
 from pydantic import BaseModel, constr
 
-from outlines.processors import JSONLogitsProcessor
+from outlines.generate.processors import JSONLogitsProcessor
 from outlines.models.llamacpp import LlamaCppTokenizer
 
 
@@ -36,7 +36,7 @@ class Character(BaseModel):
 
     prompt = "Instruct: You are a leading role play gamer. You have seen thousands of different characters and their attributes.\nPlease return a JSON object with common attributes of an RPG character. Give me a character description\nOutput:"
 
-    logits_processor = JSONLogitsProcessor(Character, tokenizer, tensor_library_name="numpy")
+    logits_processor = JSONLogitsProcessor(Character, tokenizer)
 
     json_str = llama.create_completion(
         prompt,
diff --git a/examples/math_generate_code.py b/examples/math_generate_code.py
index a5a53f084..c65537ddb 100644
--- a/examples/math_generate_code.py
+++ b/examples/math_generate_code.py
@@ -1,11 +1,9 @@
 """Example from https://dust.tt/spolu/a/d12ac33169"""
 
-import openai
-
 import outlines
+import outlines.models as models
 from outlines import Template
 
-
 examples = [
     {"question": "What is 37593 * 67?", "code": "37593 * 67"},
     {
@@ -37,8 +35,8 @@ def execute_code(code):
     return result
 
 
-prompt = answer_with_code_prompt(question=question, examples=examples)
-model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
-answer = model(prompt)
+prompt = answer_with_code_prompt(question, examples)
+model = models.openai("gpt-4o-mini")
+answer = outlines.generate.text(model)(prompt)
 result = execute_code(answer)
 print(f"It takes Carla {result:.0f} minutes to download the file.")
diff --git a/examples/meta_prompting.py b/examples/meta_prompting.py
index ed4b7455a..18f1ad41a 100644
--- a/examples/meta_prompting.py
+++ b/examples/meta_prompting.py
@@ -12,15 +12,11 @@
 
 import argparse
 
-import openai
-
 import outlines
+import outlines.models as models
 from outlines import Template
 
 
-client = openai.OpenAI()
-
-
 def split_into_steps(question, model_name: str):
     solve = Template.from_string(
         """{{question}}
@@ -28,15 +24,16 @@ def split_into_steps(question, model_name: str):
         """
     )
 
-    model = outlines.from_openai(client, model_name)
+    model = models.openai(model_name)
+    generator = outlines.generate.text(model)
 
-    prompt = solve(question=question)
-    answer = model(prompt, max_tokens=500)
+    prompt = solve(question)
+    answer = generator(prompt, 500)
     prompt += (
         answer
         + "\n what is the only option that displays the same type of relationship as : :?"
     )
-    answer = model(prompt, max_tokens=500)
+    answer = generator(prompt, 500)
     completed = prompt + answer
 
     return completed
@@ -52,12 +49,13 @@ def fill_in_the_blanks(question, model_name: str):
 
     solve = Template.from_string("""{{memory}}. Let's begin.""")
 
-    model = outlines.from_openai(client, model_name)
+    model = models.openai(model_name)
+    generator = outlines.generate.text(model)
 
-    prompt = determine_goal(question=question)
-    answer = model(prompt, stop=["."])
-    prompt = solve(memory=prompt + answer)
-    answer = model(prompt, max_tokens=500)
+    prompt = determine_goal(question)
+    answer = generator(prompt, stop_at=["."])
+    prompt = solve(prompt + answer)
+    answer = generator(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -91,12 +89,13 @@ def ask_an_expert(question, model_name: str):
         """
     )
 
-    model = outlines.from_openai(client, model_name)
+    model = models.openai(model_name)
+    generator = outlines.generate.text(model)
 
-    prompt = find_expert(question=question)
-    expert = model(prompt, stop=['"'])
-    prompt = get_answer(question=question, expert=expert, memory=prompt+expert)
-    answer = model(prompt, max_tokens=500)
+    prompt = find_expert(question)
+    expert = generator(prompt, stop_at=['"'])
+    prompt = get_answer(question, expert, prompt + expert)
+    answer = generator(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -118,12 +117,13 @@ def ask_an_expert_simple(question, model_name: str):
         """
     )
 
-    model = outlines.from_openai(client, model_name)
+    model = models.openai(model_name)
+    generator = outlines.generate.text(model)
 
-    prompt = find_expert(question=question)
-    expert = model(prompt, stop=["\n", "."])
-    prompt = get_answer(expert=expert, memory=prompt+expert)
-    answer = model(prompt, max_tokens=500)
+    prompt = find_expert(question)
+    expert = generator(prompt, stop_at=["\n", "."])
+    prompt = get_answer(expert, prompt + expert)
+    answer = generator(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
diff --git a/examples/modal_example.py b/examples/modal_example.py
index fe0e6bc67..34d4ee606 100644
--- a/examples/modal_example.py
+++ b/examples/modal_example.py
@@ -4,7 +4,7 @@
 
 
 outlines_image = modal.Image.debian_slim(python_version="3.11").pip_install(
-    "outlines==1.0.0",
+    "outlines==0.0.37",
     "transformers==4.38.2",
     "datasets==2.18.0",
     "accelerate==0.27.2",
@@ -12,11 +12,9 @@
 
 
 def import_model():
-    from transformers import AutoModelForCausalLM, AutoTokenizer
+    import outlines
 
-    model_id = "mistralai/Mistral-7B-Instruct-v0.2"
-    _ = AutoTokenizer.from_pretrained(model_id)
-    _ = AutoModelForCausalLM.from_pretrained(model_id)
+    outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
 
 
 outlines_image = outlines_image.run_function(import_model)
@@ -60,22 +58,17 @@ def import_model():
 }"""
 
 
-@app.function(image=outlines_image, gpu="A100-40GB")
+@app.function(image=outlines_image, gpu=modal.gpu.A100(memory=80))
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
     import outlines
-    from transformers import AutoModelForCausalLM, AutoTokenizer
 
-    model_id = "mistralai/Mistral-7B-Instruct-v0.2"
-    model = outlines.from_transformers(
-        tokenizer=AutoTokenizer.from_pretrained(model_id),
-        model=AutoModelForCausalLM.from_pretrained(model_id, device="cuda"),
-    )
+    model = outlines.models.transformers("mistralai/Mistral-7B-v0.1", device="cuda")
 
-    character = model(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
-        outlines.json_schema(schema),
+    generator = outlines.generate.json(model, schema)
+    character = generator(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
     )
 
     print(character)
diff --git a/examples/parsing.py b/examples/parsing.py
index c8aac5a2e..9e08eae14 100644
--- a/examples/parsing.py
+++ b/examples/parsing.py
@@ -36,6 +36,7 @@
     start="file_input",
 )
 
+
 class ParserLogitsProcessor(LogitsProcessor):
     """Bias invalid token scores according to a running parse state."""
 
diff --git a/examples/pick_odd_one_out.py b/examples/pick_odd_one_out.py
index 36ec9dfe7..f1a992411 100644
--- a/examples/pick_odd_one_out.py
+++ b/examples/pick_odd_one_out.py
@@ -10,37 +10,34 @@
 
 """
 
-import json
+import outlines
+import outlines.models as models
 
-import openai
 
-import outlines
-from outlines import Generator
-from outlines.types import JsonSchema
+build_ooo_prompt = outlines.Template.from_string(
+    """
+    Pick the odd word out: skirt, dress, pen, jacket.
+    skirt is clothing, dress is clothing, pen is an object, jacket is clothing.
+    So the odd one is pen.
+
+    Pick the odd word out: Spain, France, German, England, Singapore.
+    Spain is a country, France is a country, German is a language, ...
+    So the odd one is German.
 
+    Pick the odd word out: {{ options | join(", ") }}.
 
-build_ooo_prompt = outlines.Template.from_file("prompts/pick_odd_one_out.txt")
+    """
+)
 
 options = ["sea", "mountains", "plains", "sock"]
-options_schema = JsonSchema({
-    "type": "object",
-    "properties": {
-        "result": {
-            "type": "string",
-            "enum": options
-        }
-    },
-    "required": ["result"]
-})
-
-model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
-gen_text = Generator(model)
-gen_choice = Generator(model, options_schema)
-
-prompt = build_ooo_prompt(options=options)
-reasoning = gen_text(prompt, stop=["Pick the odd word", "So the odd one"])
+
+model = models.openai("gpt-4o-mini")
+gen_text = outlines.generate.text(model)
+gen_choice = outlines.generate.choice(model, options)
+
+prompt = build_ooo_prompt(options)
+reasoning = gen_text(prompt, stop_at=["Pick the odd word", "So the odd one"])
 prompt += reasoning
-raw_result = gen_choice(prompt)
-result = json.loads(raw_result)["result"]
+result = gen_choice(prompt)
 prompt += result
 print(result)
diff --git a/examples/prompts/babyagi_create_task.txt b/examples/prompts/babyagi_create_task.txt
deleted file mode 100644
index 434063aad..000000000
--- a/examples/prompts/babyagi_create_task.txt
+++ /dev/null
@@ -1,9 +0,0 @@
-Objective: {{ objective }}
-Current Task: {{ task }}
-Result: {{ result }}
-Previous Tasks: {{ previous_tasks }}
-
-Based on the result, create a list of new tasks that will help achieve the objective.
-Please provide the tasks in the following format:
-1. [Task description]
-2. [Task description]
diff --git a/examples/prompts/babyagi_perform_task.txt b/examples/prompts/babyagi_perform_task.txt
deleted file mode 100644
index bd8098924..000000000
--- a/examples/prompts/babyagi_perform_task.txt
+++ /dev/null
@@ -1,5 +0,0 @@
-Objective: {{ objective }}
-Task: {{ task }}
-
-Please perform the task and provide a concise result in the following format:
-Result: [Your concise result here]
diff --git a/examples/prompts/babyagi_prioritize_task.txt b/examples/prompts/babyagi_prioritize_task.txt
deleted file mode 100644
index f42c8b05f..000000000
--- a/examples/prompts/babyagi_prioritize_task.txt
+++ /dev/null
@@ -1,7 +0,0 @@
-Tasks: {{ tasks }}
-Next Task ID: {{ next_task_id }}
-
-Please prioritize the tasks based on their importance and urgency to achieve the objective.
-Provide the prioritized tasks in the following format:
-1. [Task ID]. [Task description]
-2. [Task ID]. [Task description]
diff --git a/examples/prompts/dating_profile.txt b/examples/prompts/dating_profile.txt
deleted file mode 100644
index 129be4582..000000000
--- a/examples/prompts/dating_profile.txt
+++ /dev/null
@@ -1,11 +0,0 @@
-You are a world-renowned matchmaker who understands the modern dating market. Your job is to generate dating app profiles for male clients interested in women based on a provided description. The profiles should be authentic, show off their strengths, and maximize their likelihood of getting matches on dating apps.
-Here are some examples of past clients that you have successfully created profiles for:
-{% for example in examples %}
-Description:
-{{ example.description }}
-Profile:
-{{ example.profile }}
-{% endfor %}
-Here is the new client who you need to create a profile for:
-Description: {{ description }}
-Profile:
diff --git a/examples/prompts/pick_odd_one_out.txt b/examples/prompts/pick_odd_one_out.txt
deleted file mode 100644
index 4a00ceb00..000000000
--- a/examples/prompts/pick_odd_one_out.txt
+++ /dev/null
@@ -1,9 +0,0 @@
-Pick the odd word out: skirt, dress, pen, jacket.
-skirt is clothing, dress is clothing, pen is an object, jacket is clothing.
-So the odd one is pen.
-
-Pick the odd word out: Spain, France, German, England, Singapore.
-Spain is a country, France is a country, German is a language, ...
-So the odd one is German.
-
-Pick the odd word out: {{ options | join(", ") }}.
diff --git a/examples/prompts/self_consistency.txt b/examples/prompts/self_consistency.txt
deleted file mode 100644
index 29f1f8a74..000000000
--- a/examples/prompts/self_consistency.txt
+++ /dev/null
@@ -1,6 +0,0 @@
-{% for example in examples %}
-Q: {{ example.question }}
-A: {{ example.answer }}
-{% endfor %}
-Q: {{ question }}
-A:
diff --git a/examples/react.py b/examples/react.py
index b7c807726..9c1954e88 100644
--- a/examples/react.py
+++ b/examples/react.py
@@ -11,14 +11,12 @@
 
 """
 
-import json
-
 import requests  # type: ignore
-from openai import OpenAI
 
 import outlines
-from outlines import Generator, Template
-from outlines.types import JsonSchema
+from outlines import Template
+import outlines.generate as generate
+import outlines.models as models
 
 
 build_reAct_prompt = Template.from_string(
@@ -49,56 +47,32 @@ def search_wikipedia(query: str):
     return ".".join(list(page.values())[0]["extract"].split(".")[:2])
 
 
-prompt = build_reAct_prompt(question="Where is Apple Computers headquarted? ")
-model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
-
-# Define JSON schemas for mode and action
-mode_schema = JsonSchema({
-    "type": "object",
-    "properties": {
-        "result": {
-            "type": "string",
-            "enum": ["Tho", "Act"]
-        }
-    },
-    "required": ["result"]
-})
-action_schema = JsonSchema({
-    "type": "object",
-    "properties": {
-        "result": {
-            "type": "string",
-            "enum": ["Search", "Finish"]
-        }
-    },
-    "required": ["result"]
-})
-
-mode_generator = Generator(model, mode_schema)
-action_generator = Generator(model, action_schema)
-text_generator = Generator(model)
+prompt = build_reAct_prompt("Where is Apple Computers headquarted? ")
+model = models.openai("gpt-4o-mini")
+
+mode_generator = generate.choice(model, choices=["Tho", "Act"])
+action_generator = generate.choice(model, choices=["Search", "Finish"])
+text_generator = generate.text(model)
 
 for i in range(1, 10):
-    mode_output = mode_generator(prompt, max_tokens=128)
-    mode = json.loads(mode_output)["result"]  # Extract the result from the JSON output
-    prompt = add_mode(i=i, mode=mode, result="", prompt=prompt)
+    mode = mode_generator(prompt, max_tokens=128)
+    prompt = add_mode(i, mode, "", prompt)
 
     if mode == "Tho":
-        thought = text_generator(prompt, stop="\n", max_tokens=128)
+        thought = text_generator(prompt, stop_at="\n", max_tokens=128)
         prompt += f"{thought}"
     elif mode == "Act":
-        action_output = action_generator(prompt, max_tokens=128)
-        action = json.loads(action_output)["result"]  # Extract the result from the JSON output
+        action = action_generator(prompt, max_tokens=128)
         prompt += f"{action} '"
 
-        subject = text_generator(prompt, stop=["'"], max_tokens=128)
+        subject = text_generator(prompt, stop_at=["'"], max_tokens=128)
         # Apple Computers headquartered
         subject = " ".join(subject.split()[:2])
         prompt += f"{subject}'"
 
         if action == "Search":
             result = search_wikipedia(subject)
-            prompt = add_mode(i=i, mode="Obs", result=result, prompt=prompt)
+            prompt = add_mode(i, "Obs", result, prompt)
         else:
             break
 
diff --git a/examples/sampling.ipynb b/examples/sampling.ipynb
index 75561451c..d6b50e33d 100644
--- a/examples/sampling.ipynb
+++ b/examples/sampling.ipynb
@@ -7,15 +7,14 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "#OK\n",
     "import functools as ft\n",
     "import re\n",
     "\n",
     "import numpy as np\n",
     "import matplotlib.pylab as plt\n",
-    "import openai\n",
     "\n",
-    "import outlines"
+    "import outlines.models as models\n",
+    "import outlines.text as text"
    ]
   },
   {
@@ -41,7 +40,7 @@
    "id": "2a3514d6-d5d7-46e9-9b69-1251d337e094",
    "metadata": {},
    "source": [
-    "In this example we will look at completion results for questions similar to those in the GSM8K dataset, using few-shots prompts with 5 examples. We first use `outlines.Template` to build the few-shot prompt. Outlines uses the Jinja2 templating engine to render the object when the function is called with the variables' values; it thus allows you to build complex prompts very easily."
+    "In this example we will look at completion results for questions from the GSM8K dataset, using few-shots prompts with 5 examples. We first use `outlines.text.prompt` to build the few-shot prompt. `outlines.text.prompt` is a decorator around \"prompt functions\" which contain the prompt template in its docstring. Outlines uses the Jinja2 templating engine to render the prompt when the function is called with the variables' values; it thus allows you to build complex prompts very easily."
    ]
   },
   {
@@ -87,7 +86,8 @@
     "]\n",
     "\n",
     "\n",
-    "few_shot_prompt = outlines.Template.from_string(\n",
+    "@text.prompt\n",
+    "def few_shot_prompt(question, examples):\n",
     "    \"\"\"\n",
     "    {% for example in examples %}\n",
     "    Q: {{ example.question }}\n",
@@ -96,9 +96,9 @@
     "    Q: {{ question }}\n",
     "    A:\n",
     "    \"\"\"\n",
-    ")\n",
     "\n",
-    "# Template instances can be partially evaluated because they are callable objects\n",
+    "\n",
+    "# Prompt functions can be partially evaluated like any other function\n",
     "gsm8k_prompt = ft.partial(few_shot_prompt, examples=examples)"
    ]
   },
@@ -107,7 +107,7 @@
    "id": "1eae0ec8-89f0-43fc-b055-6fcd64cbc03b",
    "metadata": {},
    "source": [
-    "## When `gpt-4o-mini` is uncertain"
+    "## When `text-davinci-003` is uncertain"
    ]
   },
   {
@@ -115,7 +115,7 @@
    "id": "a273ed78-e813-467e-85f3-16d7f283ba87",
    "metadata": {},
    "source": [
-    "Let us now sample 20 completions with the `gpt-4o-mini` model. Outlines is sampling first, and allows to draw several samples with both OpenAI and `transformers` models easily:"
+    "Let us now sample 20 completions with the `text-davinci-003` model. Outlines is sampling first, and allows to draw several samples with both OpenAI and `transformers` models easily:"
    ]
   },
   {
@@ -125,11 +125,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4o\")\n",
+    "model = models.text_completion.openai(\"text-davinci-003\", max_tokens=128)\n",
     "\n",
-    "question = \"When I was 6, my sister was half the age of my brother. When I was 14, my sister was 3 years younger than my brother. Now I'm 70, how old is my sister now?\"\n",
-    "prompt = gsm8k_prompt(question=question)\n",
-    "answers = model(prompt, n=20, max_tokens=512)"
+    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
+    "prompt = gsm8k_prompt(question)\n",
+    "answers = model(prompt, samples=20)"
    ]
   },
   {
@@ -201,7 +201,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqsAAAHgCAYAAACCbCTDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAAsTAAALEwEAmpwYAAAVwklEQVR4nO3df7DldX3f8dcbFtmVLb9cDDgLrB0DjNWUlWQzWBaDtEJg0BBiJk5VsKLFKRSsU7u2M8wKZouDjsg4Y0chBn9MTfgh3QkapQiCHSJWFigBAlMgggUkJlVJDXXh0z/2ILuwF9LZe+95372Px8yZvff7PeznvcB893m/53vOt8YYAQCAjnaZ9gAAADATsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtLXkRfb7XCsAAOZDbW+jM6sAALQlVgEAaEusAgvGddddl5NPPjn7779/dt9997ziFa/Icccdl69+9avTHg3YiTn2TNeLXbMK0MIHP/jBXHjhhVm5cmXe/OY3Z8WKFXn88cfzve99LzfccENOOOGEaY8I7IQce6avxnjB91B5gxUwdZ/97Gfz3ve+N6eeemo+85nP5CUveck2+3/+859nt912m9J0wM7KsWfebfcNVmIVaO3JJ5/MgQcemGXLluW+++573l8WAHPBsWcqthurLgMAWrv22mvz+OOP55xzzskuu+ySa665JnfeeWeWLl2aNWvW5Mgjj5z2iMBOyLGnD7EKtPbd7343SbJ06dKsXr06d9555zb7jz766FxxxRXZb7/9pjEesJNy7OnDpwEArf3whz9Mklx44YWpqtx000356U9/mjvuuCNvetObcuONN+atb33rlKcEdjaOPX2IVaC1p59+OkmyZMmSbNy4MUcddVSWL1+e1772tfnKV76SlStX5lvf+lZuvvnmKU8K7Ewce/oQq0Bre++9d5Jk9erVWbVq1Tb7XvrSl+a4445Lktxyyy3zPBmwM3Ps6UOsAq0deuihSZ79i+O59tlnnyTJz372s/kaCVgEHHv6EKtAa8cee2yqKnfdddcvXpbb2jNvenjlK18536MBOzHHnj7EKtDawQcfnJNOOinf//7388lPfnKbfd/4xjfy9a9/PXvvvXeOP/74KU0I7Iwce/pwUwCgvYcffjivf/3r89BDD+XYY4/N6tWr88ADD+Tqq69OVeXLX/5yTjnllGmPCexkHHvmnTtYAQvX448/nvPOOy8bN27MI488kj333DNr167Nhz70oaxZs2ba4wE7KceeeSVWAQBoa7ux6ppVAADaEqsAALQlVgEAaGvJtAcAeDGr1l0z474HLzhxHicBFhPHnh6cWQUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCAAvaF7/4xVRVqiqXXHLJtMdhlolVAGDBeuihh3LmmWdm+fLl0x6FOSJWAYAFaYyRd73rXXnZy16WM844Y9rjMEfEKgCwIF188cX55je/mc997nPZY489pj0Oc0SsAgALzt13351169bl7LPPztFHHz3tcZhDYhUAWFA2b96cd7zjHTnooIOyYcOGaY/DHFsy7QEAAP5/nHfeedm0aVO+/e1vZ9myZdMehznmzCoAsGB85zvfyYYNG/KBD3wgRx555LTHYR6IVQBgQdi8eXPe+c535pBDDsn5558/7XGYJ2IVAFgQnnjiidx77725++67s3Tp0l/cCKCq8uEPfzhJ8p73vCdVlXPOOWe6wzJrXLMKACwIu+++e9797ndvd9+tt96aTZs25aijjsqhhx7qEoGdiFgFABaEZcuWzXg71fXr12fTpk059dRTc/rpp8/zZMwllwEAANCWWAUAoC2xCgAseOvXr88YwyUAOyGxCgBAW2IVAIC2xCoAAG356CoAYMFYte6aF9z/4AUnztMkzBdnVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAAIvEj370o1xyySU5+eST86pXvSrLli3LXnvtlaOOOiqXXnppnn766WmP+DxLpj0AAADz4/LLL8/73ve+HHDAATnmmGNy0EEH5bHHHstVV12V008/PV/72tdy+eWXp6qmPeoviFUAgEXikEMOycaNG3PiiSdml12efYF9w4YNWbNmTa688spcddVVOeWUU6Y45bZcBgAAsEi88Y1vzEknnbRNqCbJ/vvvnzPOOCNJcsMNN0xhspmJVQAAsttuuyVJlizp9cK7WAUAWOQ2b96cz3/+80mS448/fsrTbEusAgAscuvWrcudd96ZE044Iccdd9y0x9mGWAUAWMQuvvjifPzjH89hhx2WL3zhC9Me53nEKgDAIvWpT30qZ599dl796lfn+uuvz7777jvtkZ5HrAIALEIXXXRRzjrrrLzmNa/J9ddfn/3333/aI22XWAUAWGQ++tGP5v3vf38OP/zwXH/99Xn5y18+7ZFmJFYBABaR888/P+vWrcsRRxyR6667LitWrJj2SC+o1wdpAQAwZy677LKce+652XXXXbN27dpcfPHFz3vOqlWrctppp83/cDMQqwAAi8QDDzyQJHnqqady0UUXbfc5b3jDG1rFqssAAAAWifXr12eM8YIPt1sFAIC/J7EKAEBbYhUAgLa8wQoAYBFZte6aGfc9eMGJ8zjJ348zqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoq2WsXnHFFTnrrLOydu3a7LnnnqmqvP3tb5/2WAAAc0L7zGzJtAfYno985CO5/fbbs3z58qxcuTL33HPPtEcCAJgz2mdmLc+sfuITn8i9996bn/zkJ/n0pz897XEAAOaU9plZyzOrxxxzzLRHAACYN9pnZi3PrAIAQCJWAQBoTKwCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2mp5U4Crr746V199dZLk0UcfTZLcfPPNOe2005IkK1asyMc+9rEpTQcAMLu0z8xaxuptt92Wyy67bJtt999/f+6///4kycEHH7xo/4MBADsf7TOzlpcBrF+/PmOMGR8PPvjgtEcEAJg12mdmLWMVAAASsQoAQGNiFQCAtlq+wSpJVq27ZsZ9D15w4jxOAgAwt3TPzJxZBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG3VGGPmnVV/mmTF/I0zoxVJ/mraQwBtOCYAz5jP44G15tZfjTGOf+7GF4zVLqrqv48xfnXacwA9OCYAz5jP44G1psNlAAAAtCVWAQBoa6HE6memPQDQimMC8Iz5PB5YawoWxDWrAAAsTgvlzCoAAItQ+1itqr2r6oqquqeq7q6qI6c9EzA/qmppVd1SVbdX1Z9X1Ycn2/+wqh6oqtsmj8OnPCowD16oCarqA1U1qmpWPnJze2tV1R9tddx5sKpum4V1Dt3q97ytqn5SVedU1b5VdW1V3Tf5dZ85XOv8qrpjsu0bVfWKHV1rNrW/DKCqLkty0xjjkqp6SZKXjjH+95THAuZBVVWSPcYYT1TVbkm+neTsJGck+ZMxxhVTHRCYVzM1QVUdmOSSJIclOWKMscOfGfpi/VFVH0/y4zHGeTu61la/565JfpDk15P8qyR/Pca4oKrWJdlnjPHv5mitvxlj/GSy/V8nefUY44zZWmtHtT6zWlV7JTk6yaVJMsb4v0IVFo+xxROTb3ebPHr/hA3MiRdpgk8k+WBm6fjwYv0x+UH6d5P859lYbyvHJvmfY4y/TPKWJJdNtl+W5Lfmaq1nQnVijzQ7zraO1SSvTPJ4ks9V1aaquqSq9pj2UMD8qapdJy+1/TDJtWOM70x2/f7kZatPVNXu05sQmCfbbYKqekuSH4wxbp/rtbbavzbJY2OM+2ZxzST5vTwbwL80xnhk8vWjSX5pDtdKVf1+VT2U5J8nOXeW19oh3WN1SZLXJfn0GGN1kr9Nsm66IwHzaYzx1Bjj8CQrk6ypqtck+VC2vNz3a0n2TTJrL40BbW2vCdYn+feZ/bh6sf54W2b5rOrkUoM3J7n8ufvGlms2Z+1s5/bWGmP8hzHGgUm+lOTM2VprNnSP1YeTPLzVmZQrsuV/HmCRmbwEd32S48cYj0wuEXgyyeeSrJnqcMB8mKkJXpnk9qp6MFt+qL21qvafo7VSVUuS/HaSP9rBNZ7rN5PcOsZ4bPL9Y1V1wGTNA7Ll1aW5WmtrX0pyyiyutcNax+oY49EkD1XVoZNNxya5a4ojAfOoqvarqr0nXy9L8s+S3LPVAbyy5TquO6c1IzA/ZmiCW8cYLx9jrBpjrMqWyHzd5LmzvdYz/fFPk9wzxnh4R9bYjueerd2Y5NTJ16cm+S9ztVZV/fJW+96S5J5ZXGuHLYRPAzg8W97h95Ik9yd51xjjb6Y6FDAvqupXsuWNBbtmyw/XfzzGOK+qvplkvySV5LYkZ2z1RixgJ/ViTTA5u/qrs/RpANtdq6r+MMmfjTH+046usdVaeyT5fpJ/OMb48WTby5L8cZKDkvxlkt8dY/z1HK11ZZJDkzw9WeuMMcYPdnSt2dI+VgEAWLxaXwYAAMDiJlYBAGhLrAIA0JZYBQCgLbEKAEBbYhXgOarqt6pqVNVh054FYLETqwDP97Yk3578OhWTu+QALHpiFWArVbU8yVFJ3p3k9ybbfqOqbqiqK6rqnqr60uTuWamqC6rqrqq6o6o+VlW7VtUDtcXeVfVUVR09ee6NVfXLVbVHVf1BVd1SVZuq6i2T/adV1cbJTQ+um86/AYBe/OQOsK23JPnTMca9VfWjqjpisn11kn+U5H8l+W9J/klV3Z3k5CSHjTFGVe09xniqqv4iyauz5Z7ltyZZW1XfSXLgGOO+qtqQ5JtjjH8xuZ3sLVX1XyfrvC7Jr8zGnWoAdgbOrAJs621Jvjz5+st59lKAW8YYD48xns6WW7yuSvLjJH+X5NKq+u0k/2fy3JuSHD15/MdsOVP7a0m+O9n/piTrquq2JDckWZott1RMkmuFKsCznFkFmKiqfZO8Mclrq2ok2TXJSHJNkie3eupTSZaMMTZX1Zokxyb5nSRnTv75G5O8L8krkpyb5N8m+Y1sidgkqSSnjDH+4jnr/3qSv52TPxzAAuXMKsCzfifJF8YYB48xVo0xDkzyQJK123vy5PrWvcYYX03y/iT/eLLrliSvT/L0GOPvsuVM7L/MlohNkq8nOWur615Xz9GfB2DBE6sAz3pbkq88Z9uVmflTAf5Bkj+pqjuy5dMD/k2SjDGeTPJQkj+bPO+myXP/x+T785PsluSOqvrzyfcAbEeNMaY9AwAAbJczqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2vp/jdj4sUZoV2sAAAAASUVORK5CYII=\n",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
@@ -222,7 +222,7 @@
    "id": "661a1135-ac2d-4a49-a786-d04a7ba68b48",
    "metadata": {},
    "source": [
-    "We see that there is an important variabilty in the answers given by `gpt-4o-mini`. Depending on the number of samples taken, even self-consistency sampling may lead to the wrong result here."
+    "We see that there is an important variabilty in the answers given by `text-davinci-003`. Depending on the number of samples taken, even self-consistency sampling may lead to the wrong result here."
    ]
   },
   {
@@ -248,7 +248,7 @@
    "id": "0b15b230-b667-4c9c-8a5d-366dd61de9b7",
    "metadata": {},
    "source": [
-    "## `gpt-4o-mini` on an easier question"
+    "## `text-davinci-003` on an easier question"
    ]
   },
   {
@@ -266,9 +266,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
+    "question = \"John buys 2 pairs of shoes for each of his 3 children. They cost $60 each. How much did he pay?\"\n",
     "prompt = gsm8k_prompt(question)\n",
-    "answers = model(question, samples=20, max_tokens=512)"
+    "answers = model(question, samples=20)"
    ]
   },
   {
@@ -286,7 +286,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
@@ -316,7 +316,7 @@
    "id": "22f31872-aab7-4a68-b9f2-d335a4f1a875",
    "metadata": {},
    "source": [
-    "## How `gpt-4` compares to `gpt-4o-mini`\n",
+    "## How `gpt-4` compares to `text-davinci-003`\n",
     "\n",
     "Let us now look at how GPT4 fares on the original question:"
    ]
@@ -328,11 +328,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4\")\n",
+    "model = models.text_completion.openai(\"gpt-4\", max_tokens=128)\n",
     "\n",
-    "question = \"When I was 6, my sister was half the age of my brother. When I was 14, my sister was 3 years younger than my brother. Now I'm 70, how old is my sister now?\"\n",
+    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
     "prompt = gsm8k_prompt(question)\n",
-    "answers = model(prompt, samples=20, max_tokens=512)"
+    "answers = model(prompt, samples=20)"
    ]
   },
   {
@@ -350,7 +350,7 @@
     },
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqwAAAHgCAYAAABgsD+6AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAAsTAAALEwEAmpwYAAAQI0lEQVR4nO3dYahf9X3H8c+vua6J3Ui4hKKDOpVMxTmn1Tlwatt0rKKUziZgfJa5B2tg0TkY5EEZLepWYYITwTJ0o5Wi08ypXYdjMxG76OpkLc51jQUdU4ZiHaagREg8e5C/0WhiNWlyP7l9vSDk/s/v/Lnffx6Ed05+95wxTVMAAKDVhxZ6AAAAeC+CFQCAaoIVAIBqghUAgGqCFQCAaoIVAIBqcz9h3T2vAAA4EsaBFlxhBQCgmmAFAKCaYAV4H15++eXcdtttueyyy7Jq1aosW7Ysy5cvzwUXXJDbb789b7zxxn7f9+ijj+aSSy7J/Px8li1bljPPPDM33XRTdu/efYQ/AcDRa/yER7PawwqQ5Ktf/Wo2bNiQ448/Pp/61Kdywgkn5MUXX8y9996bHTt2ZM2aNbnnnnsyxltbsO6///6sWbMmS5cuzeWXX575+fl885vfzPbt27N27drcc889C/iJAOoccA+rYAV4H7Zs2ZJXX301l156aT70obf+c+qFF17Ieeedl+eeey6bN2/OmjVrkiQ//vGPs2rVquzYsSPbtm3LueeemyTZuXNnVq9encceeyx33nln1q1btyCfB6CQH7oCOBSrV6/OZz/72X1iNUmOO+64fOELX0iSPPzww3uPb968OS+99FLWrVu3N1aTZOnSpbnuuuuSJLfeeuvhHxxgERCsAIfomGOOSZLMzb11p8AtW7YkSS6++OJ3nX/RRRfl2GOPzaOPPprXX3/9yAwJcBQTrACHYNeuXfn617+eZN843b59e5LklFNOedd75ubmctJJJ2XXrl155plnjsygAEcxwQpwCDZt2pSnnnoql1xyST7zmc/sPb5jx44kyfLly/f7vjePv/LKK4d9RoCjnWAFOEg333xzbrzxxpx22mm54447FnocgEVLsAIchFtuuSVXX311Tj/99GzdujXz8/P7rL95BfXNK63v9ObxFStWHNY5ARYDwQrwAd10003ZuHFjzjjjjGzdujXHHXfcu8459dRTkyRPP/30u9Z27dqVZ599NnNzczn55JMP+7wARzvBCvAB3HDDDbnmmmty1llnZevWrfnoRz+63/NWr16dJHnwwQfftfbII4/ktddey/nnn58Pf/jDh3VegMVAsAK8T9dee202bdqUc845Jw899FBWrlx5wHPXrl2blStX5q677soTTzyx9/jOnTvzxS9+MUmyYcOGwz4zwGLgSVcA78PXvva1rF+/PkuWLMnGjRv3+9P/J554YtavX7/39X333Ze1a9dm6dKlWbduXebn5/PAAw/sfTTr3Xffvc+jXAF+xnk0K8Ch+NKXvpQvf/nL73nOJz7xiX2edpUk27Zty/XXX5/HHnssO3fuzKpVq3LllVfmqquuypIlSw7jxABHHcEKAEC1AwarPawAAFQTrAAAVBOsAABUm1voAQ7kxE3fWugRAAB+pvz3Vy5d6BH2yxVWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKqNaZoOvDjGg0lWHrlxABaFlUl+tNBDABxlfjRN08X7W3jPYAXggxtjPDFN07kLPQfAYmFLAAAA1QQrAADVBCvAT99fLvQAAIuJPawAAFRzhRUAgGpzCz0AwNFsjLEiyW1JzkgyJbkyyR8mOXV2yookr0zTdNaRnw5gcRCsAIfmL5I8OE3T2jHGzyU5dpqmy99cHGPcmGTHgk0HsAjYwwpwkMYYy5N8L8nJ037+Mh1jjCT/k2T1NE0/PMLjASwa9rACHLyTkryU5K/HGN8dY9w2xvjI29YvTPKiWAU4NIIV4ODNJfl4klunaTo7yatJNr1t/Yokdy7EYACLiWAFOHjPJ3l+mqbvzF5vzp6AzRhjLsnnk/zNAs0GsGgIVoCDNE3TC0meG2O8eUeATyf5/uzr30ryg2manl+Q4QAWEXcJADg0G5N8Y3aHgGeS/O7s+LrYDgDwU+EuAQAAVLMlAACAaoIVAIBqghUAgGqCFQCAaoIVAIBqghXgHcYYvzPGmMYYpy30LAAIVoD9uSLJv8x+XxCzJ2UBEMEKsI8xxs8nuSDJ72XPzf8zxvjkGOPhMcbmMcYPxhjfGGOM2dpXxhjfH2M8Ocb48zHGkjHGs2OPFWOM3WOMi2bnPjLG+OUxxkfGGH81xnh8jPHdMcbnZuvrxxgPjDG2JHloYf4EAPr4FzzAvj6X5MFpmp4eY7w8xjhndvzsJL+S5H+TbEvym2OM/0pyWZLTpmmaxhgrpmnaPcbYnuT0JCcl+fckF44xvpPkY9M0/XCM8adJtkzTdOUYY0WSx8cY/zz7Ph9PcuY0Tf93pD4wQDtXWAH2dUWSu2Zf35W3tgU8Pk3T89M0vZHke0lOTLIjyc4kt48xPp/ktdm5305y0ezXn2XPFdtfT/Jvs/XfTrJpjPG9JA8nWZrkhNnaP4lVgH25wgowM8aYT7I6ya+OMaYkS5JMSb6V5PW3nbo7ydw0TbvGGOcl+XSStUn+YPb+R5JsSPKLSf4kyR8n+WT2hGySjCRrpmna/o7v/xtJXj0sHw7gKOYKK8Bb1ia5Y5qmX5qm6cRpmj6W5NkkF+7v5Nl+1+XTNP1DkmuS/Nps6fEk5yd5Y5qmndlzRfb3sydkk+Qfk2x82z7Ysw/T5wFYFAQrwFuuSPJ37zj2tznw3QJ+IcnfjzGezJ67CvxRkkzT9HqS55L86+y8b8/O/Y/Z62uTHJPkyTHGf85eA3AAY5qmhZ4BAAAOyBVWAACqCVYAAKoJVgAAqglWAACqCVYAAKoJVgAAqglWAACqCVYAAKr9P8bb7HZA9fu3AAAAAElFTkSuQmCC",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
diff --git a/examples/self_consistency.py b/examples/self_consistency.py
index 061ff594c..4acba9d58 100644
--- a/examples/self_consistency.py
+++ b/examples/self_consistency.py
@@ -1,9 +1,9 @@
 import re
 
 import numpy as np
-import openai
 
 import outlines
+import outlines.models as models
 from outlines import Template
 
 examples = [
@@ -44,12 +44,21 @@
 question = "When I was 6 my sister was half my age. Now I’m 70 how old is my sister?"
 
 
-few_shots = Template.from_file("prompts/self_consistency.txt")
+few_shots = Template.from_string(
+    """
+    {% for example in examples %}
+    Q: {{ example.question }}
+    A: {{ example.answer }}
+    {% endfor %}
+    Q: {{ question }}
+    A:
+    """
+)
 
-model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
-generator = outlines.Generator(model)
-prompt = few_shots(question=question, examples=examples)
-answers = generator(prompt, n=10)
+model = models.openai("gpt-4o-mini")
+generator = outlines.generate.text(model)
+prompt = few_shots(question, examples)
+answers = generator(prompt, samples=10)
 
 digits = []
 for answer in answers:
@@ -62,7 +71,7 @@
         print(f"Could not parse the completion: '{answer}'")
 
 unique_digits, counts = np.unique(digits, return_counts=True)
-results = {int(d): int(c) for d, c in zip(unique_digits, counts)}
+results = {d: c for d, c in zip(unique_digits, counts)}
 print(results)
 
 max_count = max(results.values())
diff --git a/examples/simulation_based_inference.ipynb b/examples/simulation_based_inference.ipynb
index e1f888324..6ad52e58e 100644
--- a/examples/simulation_based_inference.ipynb
+++ b/examples/simulation_based_inference.ipynb
@@ -20,9 +20,8 @@
     "import requests\n",
     "import re\n",
     "\n",
-    "import openai\n",
-    "\n",
-    "import outlines\n",
+    "import outlines.models as models\n",
+    "import outlines.text as text\n",
     "\n",
     "random.seed(0)"
    ]
@@ -50,7 +49,7 @@
    "id": "aabb4db6-fd94-4c42-ab7f-97c3de45b2cc",
    "metadata": {},
    "source": [
-    "In this example we will use GPT 4 mini to solve problems from the GSM-8K dataset. The state-of-the-art performance on this dataset is obtained using few-shot prompting with 5 examples. However, it is not clear how one should select these examples. Here, we will use **simulation-based inference** to try to infer which examples we should be using to get the best out of the model's abilities to solve the problem.\n",
+    "In this example we will use GPT 3.5 to solve problems from the GSM-8K dataset. The state-of-the-art performance on this dataset is obtained using few-shot prompting with 5 examples. However, it is not clear how one should select these examples. Here, we will use **simulation-based inference** to try to infer which examples we should be using to get the best out of the model's abilities to solve the problem.\n",
     "\n",
     "Let's start with downloading the dataset:"
    ]
@@ -113,7 +112,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "few_shots = outlines.Template.from_string(\n",
+    "@text.prompt\n",
+    "def few_shots(question, examples):\n",
     "    \"\"\"\n",
     "    {% for example in examples %}\n",
     "    Q: {{ example.question }}\n",
@@ -122,16 +122,16 @@
     "    Q: {{ question }}\n",
     "    A:\n",
     "    \"\"\"\n",
-    ")\n",
     "\n",
-    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4o-mini\")\n",
+    "\n",
+    "model = models.text_completion.openai(\"text-davinci-003\", max_tokens=128)\n",
     "\n",
     "\n",
     "# TODO: This could largely benefit from vectorization in #52\n",
     "def one_train_example(problem, example_set):\n",
     "    example_ids = random.choices(range(0, len(example_set)), k=5)\n",
     "    examples = [example_set[i] for i in example_ids]\n",
-    "    prompt = few_shots(question=problem[\"question\"], examples=examples)\n",
+    "    prompt = few_shots(problem[\"question\"], examples)\n",
     "    answers_raw = model(prompt, samples=20)\n",
     "\n",
     "    samples = []\n",
@@ -203,7 +203,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtoAAAHgCAYAAACb58plAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAAsTAAALEwEAmpwYAAAbiElEQVR4nO3da7BldXnn8d8jLV7wAmqHIsBMk5FxtIyJpEW8jhFjQBxxLEQdo5SFYVKjxktubZIqy5kXg1OWGmdSZFBUrFHUoJZEGJRBvE2NaItEFHTsIEozKK0i3pIo+syLszDHtpHT0P+9zuXzqTp19l5r7X2ebRfdX9f5772quwMAAOxbd5p7AAAAWI+ENgAADCC0AQBgAKENAAADCG0AABhAaAMAwACb5h5ghOOOO64vvPDCuccAAGD9q1vbsS7PaH/jG9+YewQAADa4dRnaAAAwt2GhXVVvqqobqupzy7bdp6ouqqovTd8PmrZXVb2+qnZU1Wer6qhljzllOv5LVXXKqHkBAGBfGnlG+y1Jjttt27YkF3f3kUkunu4nyfFJjpy+TktyRrIU5klekeThSY5O8opb4hwAAFazYaHd3R9N8q3dNp+Y5Ozp9tlJnrps+1t7ySeSHFhVhyT57SQXdfe3uvvGJBfl5+MdAABWnUWv0T64u6+fbn8tycHT7UOTXLvsuJ3Ttlvb/nOq6rSq2l5V23ft2rVvpwYAgL0025shu7uT9D58vjO7e2t3b928efO+eloAALhdFh3aX5+WhGT6fsO0/bokhy877rBp261tBwCAVW3RoX1ekls+OeSUJO9btv2506ePHJPkpmmJyQeSPLGqDpreBPnEaRsAAKxqw64MWVXnJHlckvtV1c4sfXrI6UneVVWnJvlKkpOnwy9I8qQkO5L8IMnzkqS7v1VV/ynJp6bj/mN37/4GSwAAWHVqaan0+rJ169bevn373GMAALD+baxLsAMAwNyENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhl2CHQCA1W/LtvPnHuEOu+b0E+YeYY+c0QYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAm+YeANayLdvOn3uEfeKa00+YewQAWHec0QYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGCATXMPwPqwZdv5c4+wT1xz+glzjwAArBPOaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAAf7wcAe+BjS4E7apYz2lX10qr6fFV9rqrOqaq7VtURVXVpVe2oqndW1f7TsXeZ7u+Y9m+ZY2YAANgbCw/tqjo0ye8n2drdD06yX5JnJnlVktd29/2T3Jjk1Okhpya5cdr+2uk4AABY1eZao70pyd2qalOSuye5Psnjk5w77T87yVOn2ydO9zPtP7aqanGjAgDA3lt4aHf3dUleneSrWQrsm5J8Osm3u/vm6bCdSQ6dbh+a5NrpsTdPx993kTMDAMDemmPpyEFZOkt9RJJfTnJAkuP2wfOeVlXbq2r7rl277ujTAQDAHTLH0pEnJPlyd+/q7h8leU+SRyU5cFpKkiSHJbluun1dksOTZNp/7yTf3P1Ju/vM7t7a3Vs3b948+jUAAMAvNEdofzXJMVV192mt9bFJrkxySZKTpmNOSfK+6fZ50/1M+z/U3b3AeQEAYK/NsUb70iy9qfGyJFdMM5yZ5E+SvKyqdmRpDfZZ00POSnLfafvLkmxb9MwAALC3ZrlgTXe/Iskrdtt8dZKj93DsPyR5+iLmAgCAfcUl2AEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGGDT3AOsN1u2nT/3CHfYNaefMPcIAABrnjPaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADLBp7gEAWN22bDt/7hH2iWtOP2HuEdaM9fBn7s+b1cAZbQAAGEBoAwDAAEIbAAAGsEYb2GvrYf1mYg0nAGM5ow0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYYNPcAwAArAZbtp0/9wh32DWnnzD3CCzjjDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABpgltKvqwKo6t6q+UFVXVdUjquo+VXVRVX1p+n7QdGxV1eurakdVfbaqjppjZgAA2BtzndH+iyQXdve/SvJrSa5Ksi3Jxd19ZJKLp/tJcnySI6ev05KcsfhxAQBg7yw8tKvq3kkem+SsJOnuH3b3t5OcmOTs6bCzkzx1un1ikrf2kk8kObCqDlno0AAAsJfmOKN9RJJdSd5cVZ+pqjdW1QFJDu7u66djvpbk4On2oUmuXfb4ndM2AABYteYI7U1JjkpyRnc/NMn380/LRJIk3d1Jem+etKpOq6rtVbV9165d+2xYAAC4PeYI7Z1Jdnb3pdP9c7MU3l+/ZUnI9P2Gaf91SQ5f9vjDpm0/o7vP7O6t3b118+bNw4YHAICVWHhod/fXklxbVQ+YNh2b5Mok5yU5Zdp2SpL3TbfPS/Lc6dNHjkly07IlJgAAsCptmunnvijJ26pq/yRXJ3lelqL/XVV1apKvJDl5OvaCJE9KsiPJD6ZjAQBgVZsltLv78iRb97Dr2D0c20leMHomAADYl1wZEgAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAH2OrSr6qCqesiIYQAAYL1YUWhX1Yer6l5VdZ8klyV5Q1W9ZuxoAACwdq30jPa9u/s7SZ6W5K3d/fAkTxg3FgAArG0rDe1NVXVIkpOTvH/gPAAAsC6sNLRfmeQDSXZ096eq6leSfGncWAAAsLZtWuFx13f3T98A2d1XW6MNAAC3bqVntP/rCrcBAAC5jTPaVfWIJI9MsrmqXrZs172S7DdyMAAAWMtua+nI/knuMR13z2Xbv5PkpFFDAQDAWvcLQ7u7P5LkI1X1lu7+yoJmAgCANW+lb4a8S1WdmWTL8sd09+NHDAUAAGvdSkP7r5P8VZI3JvnxuHEAAGB9WGlo39zdZwydBAAA1pGVfrzf31TVf6iqQ6rqPrd8DZ0MAADWsJWe0T5l+v5Hy7Z1kl/Zt+MAAMD6sKLQ7u4jRg8CAADryYpCu6qeu6ft3f3WfTsOAACsDytdOvKwZbfvmuTYJJclEdoAALAHK1068qLl96vqwCTvGDEQAACsByv91JHdfT+JddsAAHArVrpG+2+y9CkjSbJfkgcmedeooQAAYK1b6RrtVy+7fXOSr3T3zgHzAADAurCipSPd/ZEkX0hyzyQHJfnhyKEAAGCtW1FoV9XJST6Z5OlJTk5yaVWdNHIwAABYy1a6dOTPkjysu29IkqranOR/JTl31GAAALCWrfRTR+50S2RPvrkXjwUAgA1npWe0L6yqDyQ5Z7r/jCQXjBkJAADWvl8Y2lV1/yQHd/cfVdXTkjx62vV/krxt9HAAALBW3dYZ7dcleXmSdPd7krwnSarqV6d9/2bgbACrypZt5889wj5xzeknzD0CwIZwW+usD+7uK3bfOG3bMmQiAABYB24rtA/8Bfvutg/nAACAdeW2Qnt7Vf3u7hur6vlJPj1mJAAAWPtua432S5K8t6qenX8K661J9k/ybwfOBQAAa9ovDO3u/nqSR1bVbyZ58LT5/O7+0PDJAABgDVvR52h39yVJLhk8CwAArBuu7ggAAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMMFtoV9V+VfWZqnr/dP+Iqrq0qnZU1Turav9p+12m+zum/VvmmhkAAFZqzjPaL05y1bL7r0ry2u6+f5Ibk5w6bT81yY3T9tdOxwEAwKo2S2hX1WFJTkjyxul+JXl8knOnQ85O8tTp9onT/Uz7j52OBwCAVWuuM9qvS/LHSX4y3b9vkm93983T/Z1JDp1uH5rk2iSZ9t80Hf8zquq0qtpeVdt37do1cHQAALhtCw/tqnpykhu6+9P78nm7+8zu3trdWzdv3rwvnxoAAPbaphl+5qOSPKWqnpTkrknuleQvkhxYVZums9aHJbluOv66JIcn2VlVm5LcO8k3Fz82AACs3MLPaHf3y7v7sO7ekuSZST7U3c9OckmSk6bDTknyvun2edP9TPs/1N29wJEBAGCvrabP0f6TJC+rqh1ZWoN91rT9rCT3nba/LMm2meYDAIAVm2PpyE9194eTfHi6fXWSo/dwzD8kefpCBwMAgDtoNZ3RBgCAdUNoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADDAwkO7qg6vqkuq6sqq+nxVvXjafp+quqiqvjR9P2jaXlX1+qraUVWfraqjFj0zAADsrTnOaN+c5A+6+0FJjknygqp6UJJtSS7u7iOTXDzdT5Ljkxw5fZ2W5IzFjwwAAHtn4aHd3dd392XT7e8muSrJoUlOTHL2dNjZSZ463T4xyVt7ySeSHFhVhyx2agAA2DuzrtGuqi1JHprk0iQHd/f1066vJTl4un1okmuXPWzntA0AAFat2UK7qu6R5N1JXtLd31m+r7s7Se/l851WVduravuuXbv24aQAALD3ZgntqrpzliL7bd39nmnz129ZEjJ9v2Hafl2Sw5c9/LBp28/o7jO7e2t3b928efO44QEAYAXm+NSRSnJWkqu6+zXLdp2X5JTp9ilJ3rds+3OnTx85JslNy5aYAADAqrRphp/5qCTPSXJFVV0+bfvTJKcneVdVnZrkK0lOnvZdkORJSXYk+UGS5y10WgAAuB0WHtrd/fEkdSu7j93D8Z3kBUOHAgCAfcyVIQEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAYQ2gAAMIDQBgCAAYQ2AAAMILQBAGAAoQ0AAAMIbQAAGEBoAwDAAEIbAAAGENoAADCA0AYAgAGENgAADCC0AQBgAKENAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtAAAYQGgDAMAAQhsAAAZYM6FdVcdV1RerakdVbZt7HgAA+EXWRGhX1X5J/jLJ8UkelORZVfWgeacCAIBbtyZCO8nRSXZ099Xd/cMk70hy4swzAQDArVoroX1okmuX3d85bQMAgFWpunvuGW5TVZ2U5Ljufv50/zlJHt7dL1x2zGlJTpvuPiDJFxc+6OLcL8k35h5iBl73xuJ1byxe98ayUV93snFf+3p+3d/o7uP2tGPToie5na5Lcviy+4dN236qu89McuYih5pLVW3v7q1zz7FoXvfG4nVvLF73xrJRX3eycV/7Rn3da2XpyKeSHFlVR1TV/kmemeS8mWcCAIBbtSbOaHf3zVX1wiQfSLJfkjd19+dnHgsAAG7VmgjtJOnuC5JcMPccq8SGWCKzB173xuJ1byxe98ayUV93snFf+4Z83WvizZAAALDWrJU12gAAsKYI7TVko16GvqreVFU3VNXn5p5lUarq8Kq6pKqurKrPV9WL555pUarqrlX1yar62+m1v3LumRalqvarqs9U1fvnnmWRquqaqrqiqi6vqu1zz7MoVXVgVZ1bVV+oqquq6hFzzzRaVT1g+nO+5es7VfWSuedahKp66fR32ueq6pyquuvcMy1CVb14es2f3yh/1stZOrJGTJeh/79JfitLF+z5VJJndfeVsw62AFX12CTfS/LW7n7w3PMsQlUdkuSQ7r6squ6Z5NNJnrpB/rwryQHd/b2qunOSjyd5cXd/YubRhquqlyXZmuRe3f3kuedZlKq6JsnW7l6vn7G7R1V1dpKPdfcbp0/Uunt3f3vmsRZm+nftuixdF+Mrc88zUlUdmqW/yx7U3X9fVe9KckF3v2Xeycaqqgdn6WreRyf5YZILk/xed++YdbAFckZ77diwl6Hv7o8m+dbccyxSd1/f3ZdNt7+b5KpskKuh9pLvTXfvPH2t+zMCVXVYkhOSvHHuWRivqu6d5LFJzkqS7v7hRorsybFJ/m69R/Yym5Lcrao2Jbl7kv838zyL8MAkl3b3D7r75iQfSfK0mWdaKKG9drgM/QZVVVuSPDTJpTOPsjDTEorLk9yQ5KLu3giv/XVJ/jjJT2aeYw6d5INV9enpKr8bwRFJdiV587Rc6I1VdcDcQy3YM5OcM/cQi9Dd1yV5dZKvJrk+yU3d/cF5p1qIzyV5TFXdt6runuRJ+dkLEK57QhtWsaq6R5J3J3lJd39n7nkWpbt/3N2/nqWrwB49/fpx3aqqJye5obs/PfcsM3l0dx+V5PgkL5iWi613m5IcleSM7n5oku8n2Ujvvdk/yVOS/PXcsyxCVR2Upd9CH5Hkl5McUFW/M+9U43X3VUleleSDWVo2cnmSH88506IJ7bXjNi9Dz/oyrU9+d5K3dfd75p5nDtOv0i9JctzMo4z2qCRPmdYqvyPJ46vqf8w70uJMZ/vS3TckeW+WlsqtdzuT7Fz225pzsxTeG8XxSS7r7q/PPciCPCHJl7t7V3f/KMl7kjxy5pkWorvP6u7f6O7HJrkxS+832zCE9trhMvQbyPSGwLOSXNXdr5l7nkWqqs1VdeB0+25ZegPwF2YdarDufnl3H9bdW7L03/aHunvdn+1Kkqo6YHrDb6alE0/M0q+b17Xu/lqSa6vqAdOmY5Os+zc7L/OsbJBlI5OvJjmmqu4+/f1+bJbee7PuVdUvTd//WZbWZ7993okWa81cGXKj28iXoa+qc5I8Lsn9qmpnkld091nzTjXco5I8J8kV01rlJPnT6Qqp690hSc6ePpHgTkne1d0b6uPuNpiDk7x3qT2yKcnbu/vCeUdamBcledt08uTqJM+beZ6FmP4P1W8l+fdzz7Io3X1pVZ2b5LIkNyf5TDbOlRLfXVX3TfKjJC/YaG/69fF+AAAwgKUjAAAwgNAGAIABhDYAAAwgtAEAYAChDQAAAwhtgFWuqn5cVZcv+5rlCoJVdU1V3e92PO63q+qVVXWfqvqfI2YDWI18jjbA6vf30yXp16rHZOkKn49J8vGZZwFYGGe0Adagqrp3VX3xlisLVtU5VfW70+0zqmp7VX2+ql657DHXVNV/ns6Kb6+qo6rqA1X1d1X1e9Mxj6uqj1bV+dPz/1VV/dy/FVX1O1X1yem5/vt0gaHdj3nGdMGl30/yuiRvSPK8qnJVW2BDENoAq9/ddls68ozuvinJC5O8paqemeSg7n7DdPyfdffWJA9J8q+r6iHLnuur09nxjyV5S5KTkhyT5JXLjjk6S1ctfFCSf5Glyyb/VFU9MMkzkjxqeq4fJ3n27kN39zuTPDTJ57r7V5NckeSh3f2U2/8/BcDaYekIwOq3x6Uj3X1RVT09yV8m+bVlu06uqtOy9Hf8IVkK5s9O+245m3xFknt093eTfLeq/rGqDpz2fbK7r06WzpQneXSSc5c9/7FJfiPJp6ZLp98tyQ23Mvu/zNLlxZPkgOnnAWwIQhtgjZqWdDwwyQ+SHJRkZ1UdkeQPkzysu2+sqrckueuyh/3j9P0ny27fcv+WfxN6tx+1+/1KcnZ3v/w25tue5H5JNlXVlUkOmZaSvKi7P3bbrxBgbbN0BGDtemmSq5L8uyRvrqo7J7lXku8nuamqDk5y/O143qOr6ogp5J+Rn38D48VJTqqqX0qS6dNE/vnuTzItXzk/yYlJ/kuWlrT8usgGNgqhDbD67b5G+/TpTZDPT/IHU7h+NMmfd/ffJvlMki8keXuS/307ft6nkvy3LEX8l5O8d/nO7r4yyZ8n+WBVfTbJRVlaorInRyW5PEufOPKR2zELwJpV3bv/RhCAjaqqHpfkD7v7yTOPArDmOaMNAAADOKMNAAADOKMNAAADCG0AABhAaAMAwABCGwAABhDaAAAwgNAGAIAB/j/+cpD1bixMkQAAAABJRU5ErkJggg==\n",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
diff --git a/examples/transformers_integration.py b/examples/transformers_integration.py
new file mode 100644
index 000000000..16e2b8ec3
--- /dev/null
+++ b/examples/transformers_integration.py
@@ -0,0 +1,25 @@
+"""Example of integrating `outlines` with `transformers`."""
+
+from pydantic import BaseModel
+from transformers import pipeline
+
+from outlines.integrations.transformers import JSONPrefixAllowedTokens
+
+
+class Person(BaseModel):
+    first_name: str
+    surname: str
+
+
+pipe = pipeline("text-generation", model="mistralai/Mistral-7B-v0.1")
+prefix_allowed_tokens_fn = JSONPrefixAllowedTokens(
+    schema=Person, tokenizer_or_pipe=pipe, whitespace_pattern=r" ?"
+)
+results = pipe(
+    ["He is Tom Jones", "She saw Linda Smith"],
+    return_full_text=False,
+    do_sample=False,
+    max_new_tokens=50,
+    prefix_allowed_tokens_fn=prefix_allowed_tokens_fn,
+)
+print(results)
diff --git a/examples/vllm_offline_integration.py b/examples/vllm_integration.py
similarity index 80%
rename from examples/vllm_offline_integration.py
rename to examples/vllm_integration.py
index e320c6184..ee1b24a33 100644
--- a/examples/vllm_offline_integration.py
+++ b/examples/vllm_integration.py
@@ -4,7 +4,7 @@
 from pydantic import BaseModel
 from transformers import AutoTokenizer
 
-from outlines.models.vllm_offline import adapt_tokenizer
+from outlines.models.vllm import adapt_tokenizer
 from outlines.processors import JSONLogitsProcessor
 
 
@@ -17,10 +17,7 @@ class Person(BaseModel):
 llm = vllm.LLM(model=MODEL_ID, max_model_len=512)
 tokenizer = adapt_tokenizer(AutoTokenizer.from_pretrained(MODEL_ID))
 logits_processor = JSONLogitsProcessor(
-    schema=Person,
-    tokenizer=tokenizer,
-    tensor_library_name="torch",
-    whitespace_pattern=r" ?"
+    schema=Person, tokenizer=tokenizer, whitespace_pattern=r" ?"
 )
 result = llm.generate(
     ["He is Tom Jones", "She saw Linda Smith"],
diff --git a/mkdocs.yml b/mkdocs.yml
index ce075062b..2e77a0440 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -2,7 +2,7 @@
 site_name: Outlines
 site_author: The Outlines developers
 site_description: >-
-    Structured text generation with LLMs
+  Structured text generation with LLMs
 
 # Repository
 repo_name: dottxt-ai/outlines
@@ -11,225 +11,173 @@ repo_url: https://github.com/dottxt-ai/outlines
 # Copyright
 copyright: Copyright &copy; 2023- The Outlines Developers
 
-# Documentation directory
-docs_dir: docs
-
 # Configuration
 theme:
-    name: material
-    palette:
-        # Palette toggle for light mode
-        - media: "(prefers-color-scheme: light)"
-          scheme: default
-          primary: white
-    logo: assets/images/logo-square.svg
-    favicon: assets/images/logo-square.png
-    icon:
-        repo: fontawesome/brands/github
-    features:
-        - content.code.copy
-        - navigation.expand
-        - navigation.tabs
-        - navigation.sections
-        - header.autohide
-        - announce.dismiss
-    font:
-        text: Inter
-        code: Source Code Pro
+  name: material
+  custom_dir: docs/overrides
+  palette:
+    - scheme: default
+      primary: white
+  logo: assets/images/logo-square.svg
+  favicon: assets/images/logo-simple.png
+  icon:
+    repo: fontawesome/brands/github
+  features:
+    - content.code.copy
+    - navigation.expand
+    - navigation.tabs
+    - navigation.sections
+    - header.autohide
+    - announce.dismiss
+  font:
+    text: Inter
+    code: Source Code Pro
 
 # Additional configuration
 extra:
-    social:
-        - icon: fontawesome/brands/github
-          link: https://github.com/dottxt-ai
-        - icon: fontawesome/brands/twitter
-          link: https://twitter.com/remilouf
-    generator: false
-    analytics:
-        provider: google
-        property: !ENV GOOGLE_ANALYTICS_KEY
-    version:
-        provider: mike
-        default: latest
-        alias: true
+  social:
+    - icon: fontawesome/brands/github
+      link: https://github.com/dottxt-ai
+    - icon: fontawesome/brands/twitter
+      link: https://twitter.com/remilouf
+  generator: false
+  analytics:
+    provider: google
+    property: !ENV GOOGLE_ANALYTICS_KEY
+  version:
+    provider: mike
+    default: latest
+    alias: true
 
 # Extensions
 markdown_extensions:
-    - admonition
-    - def_list
-    - attr_list
-    - md_in_html
-    - pymdownx.highlight:
-          anchor_linenums: true
-          line_spans: __span
-          pygments_lang_class: true
-          noclasses: True
-    - pymdownx.superfences:
-          custom_fences:
-              - name: mermaid
-                class: mermaid
-                format: !!python/name:pymdownx.superfences.fence_code_format
-    - pymdownx.tabbed:
-          alternate_style: true
-    - pymdownx.inlinehilite
-    - pymdownx.details
-    - pymdownx.emoji:
-          emoji_index: !!python/name:material.extensions.emoji.twemoji
-          emoji_generator: !!python/name:material.extensions.emoji.to_svg
-    - pymdownx.snippets:
+  - admonition
+  - def_list
+  - attr_list
+  - md_in_html
+  - pymdownx.highlight:
+      anchor_linenums: true
+      line_spans: __span
+      pygments_lang_class: true
+      noclasses: True
+      pygments_style: nord
+  - pymdownx.superfences:
+      custom_fences:
+        - name: mermaid
+          class: mermaid
+          format: !!python/name:pymdownx.superfences.fence_code_format
+  - pymdownx.tabbed:
+      alternate_style: true
+  - pymdownx.inlinehilite
+  - pymdownx.details
+  - pymdownx.emoji:
+      emoji_index: !!python/name:material.extensions.emoji.twemoji
+      emoji_generator: !!python/name:material.extensions.emoji.to_svg
+  - pymdownx.snippets:
 
 extra_css:
-    - stylesheets/extra.css
+  - stylesheets/extra.css
 
 plugins:
-    - blog
-    - mkdocstrings:
-          update_features_models_documentation default_handler: python
-          handlers:
-              python:
-                  options:
-                      docstring_style: numpy
-                      show_submodules: true
-    - search
-    - section-index
-    - social:
-          cards_layout_options:
-              color: #173a58
-    - redirects:
-          redirect_maps:
-              "welcome.md": "index.md"
-              "reference/chat_templating.md": "guide/chat_templating.md"
-              "blog.md": "https://blog.dottxt.co"
-              "cookbook/index.md": "examples/index.md"
-              "cookbook/classification.md": "examples/classification.md"
-              "cookbook/extraction.md": "examples/extraction.md"
-              "cookbook/dating_profiles.md": "examples/dating_profiles.md"
-              "cookbook/chain_of_density.md": "examples/chain_of_density.md"
-              "cookbook/models_playing_chess.md": "examples/models_playing_chess.md"
-              "cookbook/simtom.md": "examples/simtom.md"
-              "cookbook/qa-with-citations.md": "examples/qa-with-citations.md"
-              "cookbook/knowledge_graph_extraction.md": "examples/knowledge_graph_extraction.md"
-              "cookbook/structured_generation_workflow.md": "examples/structured_generation_workflow.md"
-              "cookbook/chain_of_thought.md": "examples/chain_of_thought.md"
-              "cookbook/react_agent.md": "examples/react_agent.md"
-              "cookbook/atomic_caption.md": "examples/atomic_caption.md"
-              "cookbook/read-pdfs.md": "examples/read-pdfs.md"
-              "cookbook/earnings-reports.md": "examples/earnings-reports.md"
-              "cookbook/receipt-digitization.md": "examples/receipt-digitization.md"
-              "cookbook/extract_event_details.md": "examples/extract_event_details.md"
-              "cookbook/deploy-using-bentoml.md": "examples/deploy-using-bentoml.md"
-              "cookbook/deploy-using-cerebrium.md": "examples/deploy-using-cerebrium.md"
-              "cookbook/deploy-using-modal.md": "examples/deploy-using-modal.md"
-
-              # Root files
-              "quickstart.md": "getting_started/index.md"
-
-              # API to features redirects
-              "api/index.md": "features/index.md"
-              "api/json_schema.md": "features/types/json_schema.md"
-              "api/regex.md": "features/types/regex.md"
-              "api/applications.md": "features/utility/application.md"
-              "api/templates.md": "features/utility/template.md"
-              "api/guide.md": "features/advanced/guide.md"
-              "api/parsing.md": "features/advanced/parsing.md"
-              "api/models.md": "features/available_models.md"
-
-              # API subdirectories to features
-              "api/types/": "features/types/index.md"
-              "api/core/": "features/index.md"
-              "api/advanced/": "features/index.md"
-
-    - git-committers:
-        repository: dottxt-ai/outlines
-        branch: main
-    - git-revision-date-localized:
-        enable_creation_date: true
-        type: timeago
-
-    - gen-files:
-        scripts:
-        - scripts/gen_ref_pages.py
-    - literate-nav:
-        nav_file: SUMMARY.md
+  - blog
+  - mkdocstrings:
+      default_handler: python
+      handlers:
+        python:
+          options:
+            docstring_style: numpy
+            show_submodules: true
+  - search
+  - section-index
+  - social:
+      cards_layout_options:
+        color: #173a58
+  - git-committers:
+      repository: dottxt-ai/outlines
+      branch: main
+  - git-revision-date-localized:
+      enable_creation_date: true
+      type: timeago
 
 nav:
-    - Home: index.md
-
-    - Guide:
-          - Getting Started: guide/getting_started.md
-          - Installation: guide/installation.md
-          - Migrating to v1: guide/migration.md
-          - Licence: licence.md
-          - Utility:
-                - Chat Templating for Instruct Models: guide/chat_templating.md
-
-    - Features:
-          - Overview: features/index.md
-
-          - Core:
-                - Models:
-                    - Overview: features/models/index.md
-                    - Anthropic: features/models/anthropic.md
-                    - Dottxt: features/models/dottxt.md
-                    - Gemini: features/models/gemini.md
-                    - Llamacpp: features/models/llamacpp.md
-                    - Mlx-lm: features/models/mlxlm.md
-                    - Ollama: features/models/ollama.md
-                    - OpenAI: features/models/openai.md
-                    - OpenAI compatible API: features/models/openai_compatible.md
-                    - SGLang: features/models/sglang.md
-                    - TGI: features/models/tgi.md
-                    - Transformers: features/models/transformers.md
-                    - TransformersMultiModal: features/models/transformers_multimodal.md
-                    - vLLM (online server): features/models/vllm.md
-                    - vLLM (offline): features/models/vllm_offline.md
-                - Output Types:
-                    - Overview: features/core/output_types.md
-                    - Basic Types: features/core/output_types#basic-python-types
-                    - Multiple-Choices: features/core/output_types#multiple-choices
-                    - JSON: features/core/output_types#json-schemas
-                    - Regex: features/core/output_types#regex-patterns
-                    - Context-free Grammars: features/core/output_types#context-free-grammars
-                - Generator: features/core/generator.md
-
-          - Utilities:
-                - Application: features/utility/application.md
-                - Regex DSL: features/utility/regex_dsl.md
-                - Template: features/utility/template.md
-
-          - Advanced:
-                - Logits Processors: features/advanced/logits_processors.md
-
-    - API Reference: api_reference/
-
-    - Examples:
-          - examples/index.md
-          - Classification: examples/classification.md
-          - Named Entity Extraction: examples/extraction.md
-          - Dating Profiles: examples/dating_profiles.md
-          - Chain of Density: examples/chain_of_density.md
-          - Playing chess: examples/models_playing_chess.md
-          - SimTom: examples/simtom.md
-          - Q&A with Citations: examples/qa-with-citations.md
-          - Knowledge Graph Extraction: examples/knowledge_graph_extraction.md
-          - Structured Generation Workflow: examples/structured_generation_workflow.md
-          - Chain of Thought (CoT): examples/chain_of_thought.md
-          - ReAct Agent: examples/react_agent.md
-          - Vision-Language Models: examples/atomic_caption.md
-          - Structured Generation from PDFs: examples/read-pdfs.md
-          - Earnings Reports to CSV: examples/earnings-reports.md
-          - Receipt Digitization: examples/receipt-digitization.md
-          - Extract Events Details: examples/extract_event_details.md
-          - Run on the cloud:
-                - BentoML: examples/deploy-using-bentoml.md
-                - Cerebrium: examples/deploy-using-cerebrium.md
-                - Modal: examples/deploy-using-modal.md
-    - Community:
-        - community/index.md
-        - Feedback 🫶: community/feedback.md
-        - Our Discord Server ☕: https://discord.com/invite/R9DSu34mGd
-        - How to Contribute 🏗️: community/contribute.md
-        - Community Projects 👏: community/examples.md
-        - Versioning Guide 📌: community/versioning.md
-
-    - Blog: https://blog.dottxt.co
+  - Home: index.md
+  - Getting Started:
+      - Welcome to Outlines: welcome.md
+      - Installation: installation.md
+      - Quickstart: quickstart.md
+      - Licence: licence.md
+  - Cookbook:
+      - cookbook/index.md
+      - Classification: cookbook/classification.md
+      - Named Entity Extraction: cookbook/extraction.md
+      - Generate synthetic data: cookbook/dating_profiles.md
+      - Summarize a document: cookbook/chain_of_density.md
+      - Playing chess: cookbook/models_playing_chess.md
+      - Perspective-taking prompting: cookbook/simtom.md
+      - Question-answering with citations: cookbook/qa-with-citations.md
+      - Knowledge Graph Extraction: cookbook/knowledge_graph_extraction.md
+      - Structured Generation Workflow: cookbook/structured_generation_workflow.md
+      - Chain of Thought (CoT): cookbook/chain_of_thought.md
+      - ReAct Agent: cookbook/react_agent.md
+      - Vision-Language Models: cookbook/atomic_caption.md
+      - Structured Generation from PDFs: cookbook/read-pdfs.md
+      - Earnings reports to CSV: cookbook/earnings-reports.md
+      - Digitizing receipts with vision models: cookbook/receipt-digitization.md
+      - Extract events details from text: cookbook/extract_event_details.md
+      - Run on the cloud:
+          - BentoML: cookbook/deploy-using-bentoml.md
+          - Cerebrium: cookbook/deploy-using-cerebrium.md
+          - Modal: cookbook/deploy-using-modal.md
+  - Docs:
+      - reference/index.md
+      - Generation:
+          - Overview: reference/generation/generation.md
+          - Chat templating: reference/chat_templating.md
+          - Regex DSL: reference/regex_dsl.md
+          - Text: reference/text.md
+          - Samplers: reference/samplers.md
+          - Structured generation:
+              - How does it work?: reference/generation/structured_generation_explanation.md
+              - Classification: reference/generation/choices.md
+              - Regex: reference/generation/regex.md
+              - Type constraints: reference/generation/format.md
+              - JSON (function calling): reference/generation/json.md
+              - Grammar: reference/generation/cfg.md
+              - Creating Grammars: reference/generation/creating_grammars.md
+              - Custom FSM operations: reference/generation/custom_fsm_ops.md
+      - Utilities:
+          - Serve with vLLM: reference/serve/vllm.md
+          - Serve with LM Studio: reference/serve/lmstudio.md
+          - Custom types: reference/generation/types.md
+          - Prompt templating: reference/prompting.md
+          - Outlines functions: reference/functions.md
+      - Models:
+          - Overview: reference/models/models.md
+          - Open source:
+              - Transformers: reference/models/transformers.md
+              - Transformers Vision: reference/models/transformers_vision.md
+              - Llama.cpp: reference/models/llamacpp.md
+              - vLLM: reference/models/vllm.md
+              - TGI: reference/models/tgi.md
+              - ExllamaV2: reference/models/exllamav2.md
+              - MLX: reference/models/mlxlm.md
+              - Mamba: reference/models/transformers/#mamba
+          - API:
+              - OpenAI: reference/models/openai.md
+  - API Reference:
+      - api/index.md
+      - api/models.md
+      - api/templates.md
+      - api/json_schema.md
+      - api/guide.md
+      - api/parsing.md
+      - api/regex.md
+      - api/samplers.md
+  - Community:
+      - community/index.md
+      - Feedback 🫶: community/feedback.md
+      - Chat with us ☕: https://discord.com/invite/R9DSu34mGd
+      - How to contribute 🏗️: community/contribute.md
+      - Your projects 👏: community/examples.md
+      - Versioning Guide 📌: community/versioning.md
+  - Blog: https://blog.dottxt.co
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 1eebb8bf0..628c88f0e 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -1,83 +1,22 @@
 """Outlines is a Generative Model Programming Framework."""
 
-import sys
-from types import ModuleType
-
+import outlines.generate
 import outlines.grammars
+import outlines.models
 import outlines.processors
 import outlines.types
-from outlines.applications import Application
+from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
-from outlines.generator import Generator
-from outlines.models import (
-    from_dottxt,
-    from_openai,
-    from_transformers,
-    from_gemini,
-    from_anthropic,
-    from_ollama,
-    from_llamacpp,
-    from_mlxlm,
-    from_sglang,
-    from_tgi,
-    from_vllm,
-    from_vllm_offline,
-)
+from outlines.function import Function
 from outlines.templates import Template, prompt
-from outlines.types import regex, json_schema, cfg
-from outlines.templates import Vision
-
-from .v0_legacy import (
-    generate,
-    samplers,
-    models as legacy_models,
-    function,
-)
-from .v0_legacy.function import Function
-
-model_list = [
-    "from_anthropic",
-    "from_dottxt",
-    "from_gemini",
-    "from_llamacpp",
-    "from_mlxlm",
-    "from_ollama",
-    "from_openai",
-    "from_tgi",
-    "from_transformers",
-    "from_vllm",
-]
 
 __all__ = [
     "clear_cache",
     "disable_cache",
     "get_cache",
-    "Application",
-    "Generator",
-    "regex",
-    "json_schema",
-    "cfg",
+    "Function",
     "prompt",
-    "Template",
+    "Prompt",
+    "vectorize",
     "grammars",
-] + model_list
-
-
-# v0 legacy
-
-generate_module = ModuleType("generate")
-generate_module.__dict__.update(generate.__dict__)
-
-function_module = ModuleType("function")
-function_module.__dict__.update(function.__dict__)
-
-samplers_module = ModuleType("samplers")
-samplers_module.__dict__.update(samplers.__dict__)
-
-outlines.models.__dict__.update(legacy_models.__dict__)
-
-sys.modules["outlines.generate"] = generate_module
-sys.modules["outlines.samplers"] = samplers_module
-sys.modules["outlines.function"] = function_module
-
-__all__ += ["Function"]
+]
diff --git a/outlines/applications.py b/outlines/applications.py
deleted file mode 100644
index 4b4b34187..000000000
--- a/outlines/applications.py
+++ /dev/null
@@ -1,107 +0,0 @@
-"""Encapsulate a prompt template and an output type into a reusable object."""
-
-from typing import Any, Callable, Dict, Optional, Union
-
-from outlines.generator import (
-    BlackBoxGenerator,
-    Generator,
-    SteerableGenerator,
-)
-from outlines.models.base import Model
-from outlines.templates import Template
-
-
-class Application:
-    """
-    Application is a class that encapsulates a prompt template and an
-    output type. It can be called to generate a response by providing a
-    model, the values to be substituted in the template in a dictionary
-    and optional inference parameters.
-
-    Parameters
-    ----------
-    template : Union[Template, Callable]
-        A callable that takes arguments and returns a prompt string.
-    output_type : Any
-        The expected output type of the generated response.
-
-    Examples
-    --------
-    ```python
-    from pydantic import BaseModel
-    from transformers import AutoModelForCausalLM, AutoTokenizer
-    from outlines import models, Application
-    from outlines.types import JsonType
-    from outlines.templates import Template
-
-    class OutputModel(BaseModel):
-        result: int
-
-    model = models.from_transformers(
-        AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-        AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-    )
-
-    template_string = "What is 2 times {{ num }}?"
-    template = Template.from_string(template_string)
-
-    application = Application(template, JsonType(OutputModel))
-
-    result = application(model, {"num": 3}, max_new_tokens=20)
-    print(result)  # Expected output: { "result" : 6 }
-    ```
-
-    """
-    def __init__(
-        self,
-        template: Union[Template, Callable],
-        output_type: Optional[Any] = None,
-    ):
-        """
-        Parameters
-        ----------
-        template
-            The template to use to build the prompt.
-        output_type
-            The output type provided to the generator.
-
-        """
-        self.template = template
-        self.output_type = output_type
-        self.generator: Optional[Union[
-            BlackBoxGenerator, SteerableGenerator
-        ]] = None
-        self.model: Optional[Model] = None
-
-    def __call__(
-        self,
-        model: Model,
-        template_vars: Dict[str, Any],
-        **inference_kwargs
-    ) -> Any:
-        """
-        Parameters
-        ----------
-        model
-            The model to use to generate the response.
-        template_vars
-            The variables to be substituted in the template.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-        Returns
-        -------
-        Any
-            The generated response.
-        """
-        if model is None:
-            raise ValueError("you must provide a model")
-        # We save the generator to avoid creating a new one for each call.
-        # If the model has changed since the last call, we create a new
-        # generator.
-        if model != self.model:
-            self.model = model
-            self.generator = Generator(model, self.output_type)  # type: ignore
-
-        prompt = self.template(**template_vars)
-        assert self.generator is not None
-        return self.generator(prompt, **inference_kwargs)
diff --git a/outlines/v0_legacy/base.py b/outlines/base.py
similarity index 67%
rename from outlines/v0_legacy/base.py
rename to outlines/base.py
index e545b5bcf..29d42c54c 100644
--- a/outlines/v0_legacy/base.py
+++ b/outlines/base.py
@@ -4,10 +4,28 @@
 import inspect
 from typing import Callable, Optional
 
+import numpy as np
+
+# Import required functions based on NumPy version
+np_major_version = int(np.__version__.split(".")[0])
+if np_major_version >= 2:
+    from numpy.lib._function_base_impl import (
+        _calculate_shapes,
+        _parse_gufunc_signature,
+        _parse_input_dimensions,
+        _update_dim_sizes,
+    )
+else:
+    from numpy.lib.function_base import (
+        _calculate_shapes,
+        _parse_gufunc_signature,
+        _parse_input_dimensions,
+        _update_dim_sizes,
+    )
 
 # Allow nested loops for running in notebook. We don't enable it globally as it
 # may interfere with other libraries that use asyncio.
-if hasattr(builtins, "__IPYTHON__"): # pragma: no cover
+if hasattr(builtins, "__IPYTHON__"):
     try:
         import nest_asyncio
 
@@ -33,65 +51,21 @@ class vectorize:
     """
 
     def __init__(self, func: Callable, signature: Optional[str] = None):
-        self.setup_numpy(signature)
         self.func = func
         self.signature = signature
         self.is_coroutine_fn = inspect.iscoroutinefunction(func)
 
         functools.update_wrapper(self, func)
 
-    def setup_numpy(self, signature: Optional[str] = None):
-        """Setup NumPy and import required functions based on NumPy version.
-
-        As numpy is an optional dependency only needed for the legacy OpenAI
-        model, we put the import in a try block to make sure no error is raised
-        if it's not installed.
-        """
-        try:
-            import numpy as np
-            self.np = np
-            # Import required functions based on NumPy version
-            np_major_version = int(np.__version__.split(".")[0])
-            if np_major_version >= 2:
-                from numpy.lib._function_base_impl import (
-                    _calculate_shapes,
-                    _parse_gufunc_signature,
-                    _parse_input_dimensions,
-                    _update_dim_sizes,
-                )
-            else: # pragma: no cover
-                from numpy.lib.function_base import (
-                    _calculate_shapes,
-                    _parse_gufunc_signature,
-                    _parse_input_dimensions,
-                    _update_dim_sizes,
-                )
-            self._calculate_shapes = _calculate_shapes
-            self._parse_gufunc_signature = _parse_gufunc_signature
-            self._parse_input_dimensions = _parse_input_dimensions
-            self._update_dim_sizes = _update_dim_sizes
-            if signature is not None:
-                # Parse the signature string into a Python data structure.
-                # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
-                self._in_and_out_core_dimensions = self._parse_gufunc_signature(signature)
-            else:
-                self._in_and_out_core_dimensions = None
-        except (ModuleNotFoundError, ImportError): # pragma: no cover
-            self.np = None
-            self._calculate_shapes = None
-            self._parse_gufunc_signature = None
-            self._parse_input_dimensions = None
-            self._update_dim_sizes = None
+        if signature is not None:
+            # Parse the signature string into a Python data structure.
+            # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
+            self._in_and_out_core_dimensions = _parse_gufunc_signature(signature)
+        else:
             self._in_and_out_core_dimensions = None
 
     def __call__(self, *args, **kwargs):
         """Call the vectorized function."""
-        # Raise an error ourselves if NumPy is not installed.
-        if not self.np: # pragma: no cover
-            raise ImportError(
-                "NumPy is required to use the legacy version of the OpenAI "
-                "model."
-            )
         if not args and not kwargs:
             return self.call_thunk()
         elif self.signature is not None:
@@ -130,14 +104,14 @@ def call_no_signature(self, *args, **kwargs):
 
         """
         # Convert args and kwargs to arrays
-        args = [self.np.array(arg) for arg in args]
-        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
+        args = [np.array(arg) for arg in args]
+        kwargs = {key: np.array(value) for key, value in kwargs.items()}
 
         # Broadcast args and kwargs
-        broadcast_shape = self.np.broadcast(*args, *list(kwargs.values())).shape
-        args = [self.np.broadcast_to(arg, broadcast_shape) for arg in args]
+        broadcast_shape = np.broadcast(*args, *list(kwargs.values())).shape
+        args = [np.broadcast_to(arg, broadcast_shape) for arg in args]
         kwargs = {
-            key: self.np.broadcast_to(value, broadcast_shape)
+            key: np.broadcast_to(value, broadcast_shape)
             for key, value in kwargs.items()
         }
 
@@ -153,9 +127,9 @@ def call_no_signature(self, *args, **kwargs):
             results if isinstance(results, tuple) else (results,) for results in outputs
         ]
         outputs = tuple(
-            [self.np.asarray(x).reshape(broadcast_shape).squeeze() for x in zip(*outputs)]
+            [np.asarray(x).reshape(broadcast_shape).squeeze() for x in zip(*outputs)]
         )
-        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
+        outputs = tuple([x.item() if np.ndim(x) == 0 else x for x in outputs])
 
         n_results = len(list(outputs))
 
@@ -175,25 +149,25 @@ def call_with_signature(self, *args, **kwargs):
             )
 
         # Convert args and kwargs to arrays
-        args = [self.np.asarray(arg) for arg in args]
-        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
+        args = [np.asarray(arg) for arg in args]
+        kwargs = {key: np.array(value) for key, value in kwargs.items()}
 
         # Find the arguments' broadcast shape, and map placeholder
         # variables in the signature to the number of dimensions
         # they correspond to given the arguments.
-        broadcast_shape, dim_sizes = self._parse_input_dimensions(
+        broadcast_shape, dim_sizes = _parse_input_dimensions(
             args + list(kwargs.values()), input_core_dims
         )
 
         # Calculate the shape to which each of the arguments should be broadcasted
         # and reshape them accordingly.
-        input_shapes = self._calculate_shapes(broadcast_shape, dim_sizes, input_core_dims)
+        input_shapes = _calculate_shapes(broadcast_shape, dim_sizes, input_core_dims)
         args = [
-            self.np.broadcast_to(arg, shape, subok=True)
+            np.broadcast_to(arg, shape, subok=True)
             for arg, shape in zip(args, input_shapes)
         ]
         kwargs = {
-            key: self.np.broadcast_to(value, broadcast_shape)
+            key: np.broadcast_to(value, broadcast_shape)
             for key, value in kwargs.items()
         }
 
@@ -221,18 +195,18 @@ def call_with_signature(self, *args, **kwargs):
         # number of dimensions of the outputs accordingly.
         for results, core_dims in zip(flat_outputs, output_core_dims):
             for result in results:
-                self._update_dim_sizes(dim_sizes, result, core_dims)
+                _update_dim_sizes(dim_sizes, result, core_dims)
 
         # Calculate the shape to which each of the outputs should be broadcasted
         # and reshape them.
-        shapes = self._calculate_shapes(broadcast_shape, dim_sizes, output_core_dims)
+        shapes = _calculate_shapes(broadcast_shape, dim_sizes, output_core_dims)
         outputs = tuple(
             [
-                self.np.hstack(results).reshape(shape).squeeze()
+                np.hstack(results).reshape(shape).squeeze()
                 for shape, results in zip(shapes, zip(*outputs))
             ]
         )
-        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
+        outputs = tuple([x.item() if np.ndim(x) == 0 else x for x in outputs])
 
         return outputs[0] if n_results == 1 else outputs
 
@@ -252,7 +226,7 @@ def vectorize_call(self, broadcast_shape, args, kwargs):
 
         """
         outputs = []
-        for index in self.np.ndindex(*broadcast_shape):
+        for index in np.ndindex(*broadcast_shape):
             current_args = tuple(arg[index] for arg in args)
             current_kwargs = {key: value[index] for key, value in kwargs.items()}
             outputs.append(self.func(*current_args, **current_kwargs))
@@ -278,7 +252,7 @@ def vectorize_call_coroutine(self, broadcast_shape, args, kwargs):
 
         async def create_and_gather_tasks():
             tasks = []
-            for index in self.np.ndindex(*broadcast_shape):
+            for index in np.ndindex(*broadcast_shape):
                 current_args = tuple(arg[index] for arg in args)
                 current_kwargs = {key: value[index] for key, value in kwargs.items()}
                 tasks.append(self.func(*current_args, **current_kwargs))
@@ -294,3 +268,32 @@ async def create_and_gather_tasks():
             loop.close()
 
         return outputs
+
+
+def _update_arrays_type(arrays, results):
+    """Update the dtype of arrays.
+
+    String arrays contain strings of fixed length. Here they are initialized with
+    the type of the first results, so that if the next results contain longer
+    strings they will be truncated when added to the output arrays. Here we
+    update the type if the current results contain longer strings than in the
+    current output array.
+
+    Parameters
+    ----------
+    arrays
+        Arrays that contain the vectorized function's results.
+    results
+        The current output of the function being vectorized.
+
+    """
+
+    updated_arrays = []
+    for array, result in zip(arrays, results):
+        if array.dtype.type == np.str_:
+            if array.dtype < np.array(result).dtype:
+                array = array.astype(np.array(result).dtype)
+
+        updated_arrays.append(array)
+
+    return tuple(updated_arrays)
diff --git a/outlines/caching.py b/outlines/caching.py
index 7cb2ed169..6882bac6b 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -1,5 +1,3 @@
-"""Caching and memoization of function calls."""
-
 import asyncio
 import contextlib
 import functools
@@ -14,7 +12,7 @@
 _caching_enabled = True
 
 
-class CloudpickleDisk(Disk): # pragma: no cover
+class CloudpickleDisk(Disk):
     def __init__(self, directory, compress_level=1, **kwargs):
         self.compress_level = compress_level
         super().__init__(directory, **kwargs)
@@ -59,11 +57,11 @@ def get_cache():
     if outlines_cache_dir:
         # OUTLINES_CACHE_DIR takes precendence
         cache_dir = outlines_cache_dir
-    elif xdg_cache_home:  # pragma: no cover
+    elif xdg_cache_home:
         cache_dir = os.path.join(xdg_cache_home, ".cache", "outlines")
     elif home_dir != "/":
         cache_dir = os.path.join(home_dir, ".cache", "outlines")
-    else:  # pragma: no cover
+    else:
         # home_dir may be / inside a docker container without existing user
         tempdir = tempfile.gettempdir()
         cache_dir = os.path.join(tempdir, ".cache", "outlines")
@@ -110,7 +108,7 @@ def decorator(cached_function: Callable):
 
         base = (full_name(cached_function),)
 
-        if asyncio.iscoroutinefunction(cached_function):  # pragma: no cover
+        if asyncio.iscoroutinefunction(cached_function):
 
             async def wrapper(*args, **kwargs):
                 if not _caching_enabled:
diff --git a/outlines/fsm/__init__.py b/outlines/fsm/__init__.py
index 4e38718ef..e69de29bb 100644
--- a/outlines/fsm/__init__.py
+++ b/outlines/fsm/__init__.py
@@ -1 +0,0 @@
-"""Finite state machines builder for CFG generation."""
diff --git a/outlines/processors/guide.py b/outlines/fsm/guide.py
similarity index 58%
rename from outlines/processors/guide.py
rename to outlines/fsm/guide.py
index bebddd9cb..6b97d7729 100644
--- a/outlines/processors/guide.py
+++ b/outlines/fsm/guide.py
@@ -1,13 +1,9 @@
-"""Guides to control generation in steerable models.
-
-Logits processors rely on guides to control the generation process.
-"""
-
 import collections
 import copy
 import warnings
-from typing import TYPE_CHECKING, Any, Generator, ValuesView, Union
+from typing import TYPE_CHECKING, Any, Generator, Union
 
+import torch
 from lark.indenter import DedentError
 from lark.lexer import UnexpectedCharacters, UnexpectedToken
 from outlines_core.fsm.guide import Generate
@@ -19,7 +15,6 @@
 )
 
 from outlines import grammars
-from outlines.caching import cache
 from outlines.fsm.parsing import PartialLark, PartialParserState
 
 if TYPE_CHECKING:
@@ -32,106 +27,59 @@
 class Guide(CoreGuide):
     """Base definition of a generation guide.
 
-    A generation guide defines the behavior of a finite-state machine that
-    guides a text generation procedure. Unlike the DFAs built from regular
-    expressions guides, it can also emit a `Write` instructions which tells
-    the model that it can append a sequence of tokens (or token word) instead
-    of generating it.
+    A generation guide defines the behavior of a finite-state machine that guides
+    a text generation procedure. Unlike the DFAs built from regular expressions
+    guides can also emit a `Write` instructions which tells the model that it can
+    append a sequence of tokens (or token word) instead of generating it.
 
     """
+
     initial_state: Any
 
 
 class StopAtEOSGuide(Guide):
     """Guide to generate tokens until the EOS token has been generated."""
+
     final_state = 1
+    start_state = 0  # TODO: remove start_state, use only initial_state
     initial_state = 0
 
     def __init__(self, tokenizer: "Tokenizer"):
-        """
-        Parameters
-        ----------
-        tokenizer
-            The tokenizer used to convert tokens to ids.
+        """Initialize the generation guide.
+
+        model
+            The logit generator used to generate the next token.
 
         """
         self.eos_token_id = tokenizer.eos_token_id
         self.vocabulary = tokenizer.vocabulary.values()
 
     def get_next_instruction(self, state: int) -> Instruction:
-        """Return the next instruction.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        Instruction
-            An `Instruction` instance.
-
-        """
         if self.is_final_state(state):
             return Write([self.eos_token_id])
         return Generate(None)
 
     def get_next_state(self, state: int, token_id: int) -> int:
-        """Return the next state.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-        token_id
-            The id of the token that was just generated.
-
-        Returns
-        -------
-        int
-            The next state.
-
-        """
         if token_id == self.eos_token_id or state == self.final_state:
             return self.final_state
 
         return self.initial_state
 
-    def is_final_state(self, state: int) -> bool:
-        """Return whether the given state is a final state.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether the given state is a final state.
-
-        """
+    def is_final_state(self, state: int):
         return state == self.final_state
 
-    def copy(self) -> "StopAtEOSGuide":
-        """Return itself as there is no need to copy."""
+    def copy(self):
         return self
 
 
-@cache()
 def cached_create_states_mapping(regex_string, tokenizer, *args, **kwargs):
-    """Wrap the uncached create_states_mapping function in a cache."""
-    return uncached_create_states_mapping(
-        regex_string, tokenizer, *args, **kwargs
-    )
+    return uncached_create_states_mapping(regex_string, tokenizer, *args, **kwargs)
 
 
 class RegexGuide(CoreRegexGuide):
-    """Guide to generate text in the language of a regular expression.
-
-    This class is a wrapper around the CoreRegexGuide class that adds a cache
-    to the create_states_mapping function.
-
+    """
+    Guide to generate text in the language of a regular expression.
+    CoreRegexGuide with outlines cache
     """
 
     @classmethod
@@ -141,23 +89,6 @@ def from_regex(
         tokenizer,
         **kwargs,
     ):
-        """Create a RegexGuide from a regular expression.
-
-        Parameters
-        ----------
-        regex_string
-            The regular expression to generate text from.
-        tokenizer
-            The tokenizer to use to convert tokens to ids.
-        kwargs
-            Additional keyword arguments to pass to the CoreRegexGuide constructor.
-
-        Returns
-        -------
-        RegexGuide
-            A RegexGuide instance.
-
-        """
         return super().from_regex(
             regex_string,
             tokenizer,
@@ -170,25 +101,15 @@ def from_regex(
 
 
 class CFGGuide(Guide):
-    """Guide to generate text that is in the language of a context-free Lark
-    grammar.
-
-    """
+    """Guide to generate text that is in the language of a context-free Lark grammar."""
 
-    def __init__(self, cfg_string: str, tokenizer: "Tokenizer"):
+    def __init__(self, cfg_string: str, tokenizer):
         """
-        Parameters
-        ----------
-        cfg_string
-            The context-free grammar to generate text from.
-        tokenizer
-            The tokenizer to use to convert tokens to ids.
-
+        Construct the PartialLark parser and set the empty initial_state (PartialParserState)
         """
         warnings.warn(
-            "Outlines' public *community-contributed* CFG structured generation "
-            "is experimental. Please review "
-            "https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
+            "Outlines' public *community-contributed* CFG structured generation is experimental. "
+            "Please review https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
         )
 
         self.cfg_string = cfg_string
@@ -221,32 +142,25 @@ def get_next_instruction(self, state: CFGState) -> Instruction:
 
         Returns
         -------
-        Instruction
-            A `Generate` instance that contains the model and the allowed token
-            ids.
+        A `Generate` instance that contains the model and the allowed token ids.
 
         """
-        import torch
 
         if state.parser_state is None:
             return Write(torch.tensor([self.eos_token_id]))
 
         valid_tokens = list(
-            self.iter_valid_token_ids(
-                state, self.tokenizer.vocabulary.values()
-            )
+            self.iter_valid_token_ids(state, self.tokenizer.vocabulary.values())
         )
-
         if len(valid_tokens) == 1:
             return Write(torch.tensor(valid_tokens))
-
         return Generate(torch.tensor(valid_tokens))
 
     def iter_valid_token_ids(
-        self, state: CFGState, candidate_token_ids: ValuesView[int]
+        self, state: CFGState, candidate_token_ids: list
     ) -> Generator[int, None, None]:
-        """Iterate over the given token_ids and yield those that are valid for
-        the current parser state.
+        """
+        Iterate over the given token_ids and yield those that are valid for the current parser state.
 
         Parameters
         ----------
@@ -259,8 +173,11 @@ def iter_valid_token_ids(
         ------
         int
             Valid token ids.
-
         """
+        if state.parser_state is None:
+            yield self.eos_token_id
+            return
+
         for token_id in candidate_token_ids:
             if token_id == self.eos_token_id:
                 if self.can_terminate_state(state):
@@ -279,10 +196,9 @@ def iter_valid_token_ids(
                     pass
 
     def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
-        """Update the state of the guide.
-
-        Decode the token_id, and calculate the new parser_state with the token
-        applied.
+        """
+        Update the state of the guide.
+        Decode the token_id, and calculate the new parser_state with the token applied.
 
         Parameters
         ----------
@@ -293,8 +209,7 @@ def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
 
         Returns
         -------
-        CFGState
-            The guides new PartialParserState
+        The guides new PartialParserState
 
         """
         if state.parser_state is None or token_id == self.eos_token_id:
@@ -306,8 +221,7 @@ def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
     def _get_parser_state_token_applied(
         self, state: CFGState, token_id: int
     ) -> PartialParserState:
-        """Apply the given token_id to the parser state.
-
+        """
         Don't mutate `parser_state`, copy to protect
 
         Get the token string
@@ -315,19 +229,6 @@ def _get_parser_state_token_applied(
           - else: normalized (with possibly leading whitespace)
 
         Don't allow empty ("") tokens, raise ValueError
-
-        Parameters
-        ----------
-        state
-            The guide's current PartialParserState, or None if complete
-        token_id
-            The id of the token that was just generated.
-
-        Returns
-        -------
-        PartialParserState
-            The parser state with the token applied.
-
         """
         parser_state = copy.copy(state.parser_state)  # prevent side effects
 
@@ -351,38 +252,12 @@ def _get_parser_state_token_applied(
         return parser_state
 
     def is_final_state(self, state: CFGState) -> bool:
-        """Return whether the given state is a final state.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether the given state is a final state.
-
-        """
-        # TODO: remove this method, use can_terminate_state and
-        # must_terminate_state here and in RegexGuide per
-        # https://github.com/dottxt-ai/outlines/issues/885
+        # TODO: remove this method, use can_terminate_state and must_terminate_state
+        # here and in RegexGuide per https://github.com/dottxt-ai/outlines/issues/885
         return self.can_terminate_state(state)
 
     def can_terminate_state(self, state: CFGState) -> bool:
-        """Return whether generation is allowed to terminate.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether generation is allowed to terminate.
-
-        """
+        """Generation is allowed to terminate"""
         if state.parser_state is not None:
             try:
                 copy.copy(state.parser_state).feed_eof()
@@ -391,32 +266,11 @@ def can_terminate_state(self, state: CFGState) -> bool:
         return True
 
     def must_terminate_state(self, state: CFGState) -> bool:
-        """Indicate whether generation must terminate as there are no legal
-        continuations.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether generation must terminate.
-
-        """
-        return (
-            state.parser_state is None or
-            set(state.parser_state.accepts()).issubset({"$END"})
+        """Generation must terminate, no legal continuations"""
+        return state.parser_state is None or set(state.parser_state.accepts()).issubset(
+            {"$END"}
         )
 
     def copy(self) -> "CFGGuide":
-        """Create a copy of the Guide.
-
-        Returns
-        -------
-        CFGGuide
-            A copy of the Guide.
-
-        """
+        """Create a copy of the Guide."""
         return CFGGuide(self.cfg_string, self.tokenizer)
diff --git a/outlines/fsm/json_schema.py b/outlines/fsm/json_schema.py
new file mode 100644
index 000000000..578ee7626
--- /dev/null
+++ b/outlines/fsm/json_schema.py
@@ -0,0 +1,83 @@
+import inspect
+import json
+import warnings
+from enum import Enum
+from typing import Callable, Type, Union
+
+from pydantic import BaseModel, create_model
+
+
+def convert_json_schema_to_str(json_schema: Union[dict, str, Type[BaseModel]]) -> str:
+    """Convert a JSON schema to a string.
+
+    Parameters
+    ----------
+    json_schema
+        The JSON schema.
+
+    Returns
+    -------
+    str
+        The JSON schema converted to a string.
+
+    Raises
+    ------
+    ValueError
+        If the schema is not a dictionary, a string or a Pydantic class.
+    """
+    if isinstance(json_schema, dict):
+        schema_str = json.dumps(json_schema)
+    elif isinstance(json_schema, str):
+        schema_str = json_schema
+    elif issubclass(json_schema, BaseModel):
+        schema_str = json.dumps(json_schema.model_json_schema())
+    else:
+        raise ValueError(
+            f"Cannot parse schema {json_schema}. The schema must be either "
+            + "a Pydantic class, a dictionary or a string that contains the JSON "
+            + "schema specification"
+        )
+    return schema_str
+
+
+def get_schema_from_signature(fn: Callable) -> dict:
+    """Turn a function signature into a JSON schema.
+
+    Every JSON object valid to the output JSON Schema can be passed
+    to `fn` using the ** unpacking syntax.
+
+    """
+    signature = inspect.signature(fn)
+    arguments = {}
+    for name, arg in signature.parameters.items():
+        if arg.annotation == inspect._empty:
+            raise ValueError("Each argument must have a type annotation")
+        else:
+            arguments[name] = (arg.annotation, ...)
+
+    try:
+        fn_name = fn.__name__
+    except Exception as e:
+        fn_name = "Arguments"
+        warnings.warn(
+            f"The function name could not be determined. Using default name 'Arguments' instead. For debugging, here is exact error:\n{e}",
+            category=UserWarning,
+        )
+    model = create_model(fn_name, **arguments)
+
+    return model.model_json_schema()
+
+
+def get_schema_from_enum(myenum: type[Enum]) -> dict:
+    if len(myenum) == 0:
+        raise ValueError(
+            f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
+        )
+    choices = [
+        get_schema_from_signature(elt.value.func)
+        if callable(elt.value)
+        else {"const": elt.value}
+        for elt in myenum
+    ]
+    schema = {"title": myenum.__name__, "oneOf": choices}
+    return schema
diff --git a/outlines/fsm/parsing.py b/outlines/fsm/parsing.py
index 3831c57c1..e48fb69e4 100644
--- a/outlines/fsm/parsing.py
+++ b/outlines/fsm/parsing.py
@@ -1,5 +1,3 @@
-"""Build a parser from a grammar to create a finite state machine."""
-
 from copy import copy, deepcopy
 from dataclasses import dataclass
 from functools import lru_cache
diff --git a/outlines/fsm/types.py b/outlines/fsm/types.py
new file mode 100644
index 000000000..f6409aa66
--- /dev/null
+++ b/outlines/fsm/types.py
@@ -0,0 +1,76 @@
+import datetime
+from enum import EnumMeta
+from typing import Any, Protocol, Tuple, Type
+
+from outlines.types import Regex, boolean as boolean_regex, date as date_regex
+from outlines.types import datetime as datetime_regex
+from outlines.types import (
+    integer as integer_regex,
+    number as number_regex,
+    time as time_regex,
+)
+
+
+class FormatFunction(Protocol):
+    def __call__(self, sequence: str) -> Any: ...
+
+
+def python_types_to_regex(python_type: Type) -> Tuple[Regex, FormatFunction]:
+    # If it is a custom type
+    if isinstance(python_type, Regex):
+        custom_regex_str = python_type.pattern
+
+        def custom_format_fn(sequence: str) -> str:
+            return str(sequence)
+
+        return Regex(custom_regex_str), custom_format_fn
+
+    if isinstance(python_type, EnumMeta):
+        values = python_type.__members__.keys()
+        enum_regex_str: str = "(" + "|".join(values) + ")"
+
+        def enum_format_fn(sequence: str) -> str:
+            return str(sequence)
+
+        return Regex(enum_regex_str), enum_format_fn
+
+    if python_type is float:
+
+        def float_format_fn(sequence: str) -> float:
+            return float(sequence)
+
+        return number_regex, float_format_fn
+    elif python_type is int:
+
+        def int_format_fn(sequence: str) -> int:
+            return int(sequence)
+
+        return integer_regex, int_format_fn
+    elif python_type is bool:
+
+        def bool_format_fn(sequence: str) -> bool:
+            return bool(sequence)
+
+        return boolean_regex, bool_format_fn
+    elif python_type == datetime.date:
+
+        def date_format_fn(sequence: str) -> datetime.date:
+            return datetime.datetime.strptime(sequence, "%Y-%m-%d").date()
+
+        return date_regex, date_format_fn
+    elif python_type == datetime.time:
+
+        def time_format_fn(sequence: str) -> datetime.time:
+            return datetime.datetime.strptime(sequence, "%H:%M:%S").time()
+
+        return time_regex, time_format_fn
+    elif python_type == datetime.datetime:
+
+        def datetime_format_fn(sequence: str) -> datetime.datetime:
+            return datetime.datetime.strptime(sequence, "%Y-%m-%d %H:%M:%S")
+
+        return datetime_regex, datetime_format_fn
+    else:
+        raise NotImplementedError(
+            f"The Python type {python_type} is not supported. Please open an issue."
+        )
diff --git a/outlines/v0_legacy/function.py b/outlines/function.py
similarity index 58%
rename from outlines/v0_legacy/function.py
rename to outlines/function.py
index 465df53f9..aff21d68f 100644
--- a/outlines/v0_legacy/function.py
+++ b/outlines/function.py
@@ -1,58 +1,16 @@
 import importlib.util
-import warnings
 from dataclasses import dataclass
 from typing import TYPE_CHECKING, Callable, Optional, Tuple, Union
 
 import requests
 
-from outlines.v0_legacy import generate
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-from outlines.v0_legacy.models import transformers
+from outlines import generate, models
 
-if TYPE_CHECKING: # pragma: no cover
+if TYPE_CHECKING:
+    from outlines.generate.api import SequenceGenerator
     from outlines.templates import Template
 
 
-def function_warning():
-    warnings.warn("""
-        The `Function` class is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should use the `outlines.Application` class that
-        implements a similar functionality.
-        An `Application` is initialized with a prompt template and an
-        output type. It can then be called with a model and a prompt
-        to generate a response that follows the format specified by the
-        output type.
-        For example:
-        ```python
-        from pydantic import BaseModel
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-        from outlines import models, Application
-        from outlines.types import JsonType
-        from outlines.templates import Template
-
-        class OutputModel(BaseModel):
-            result: int
-
-        model = models.from_transformers(
-            AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-            AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-        )
-
-        template_string = "What is 2 times {{ num }}?"
-        template = Template.from_string(template_string)
-
-        application = Application(template, JsonType(OutputModel))
-
-        result = application(model, num=3)
-        print(result)  # Expected output: { "result" : 6 }
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2
-    )
-
-
 @dataclass
 class Function:
     """Represents an Outlines function.
@@ -63,18 +21,15 @@ class Function:
     prompt template.
 
     """
+
     prompt_template: "Template"
     schema: Union[str, Callable, object]
     model_name: str
-    generator: Optional["GeneratorV0Adapter"] = None
-
-    def __post_init__(self):
-        function_warning()
+    generator: Optional["SequenceGenerator"] = None
 
     @classmethod
     def from_github(cls, program_path: str, function_name: str = "fn"):
         """Load a function stored on GitHub"""
-        function_warning()
         program_content = download_from_github(program_path)
         function = extract_function_from_file(program_content, function_name)
 
@@ -82,7 +37,7 @@ def from_github(cls, program_path: str, function_name: str = "fn"):
 
     def init_generator(self):
         """Load the model and initialize the generator."""
-        model = transformers(self.model_name)
+        model = models.transformers(self.model_name)
         self.generator = generate.json(model, self.schema)
 
     def __call__(self, *args, **kwargs):
@@ -100,18 +55,11 @@ def __call__(self, *args, **kwargs):
             Values to pass to the prompt template as keyword arguments.
 
         """
-        # We do not want to raise the deprecation warning about the `json`
-        # function or the transformers function here.
-        with warnings.catch_warnings():
-            warnings.filterwarnings(
-                "ignore",
-                category=DeprecationWarning
-            )
-            if self.generator is None:
-                self.init_generator()
+        if self.generator is None:
+            self.init_generator()
 
-            prompt = self.prompt_template(*args, **kwargs)
-            return self.generator(prompt)
+        prompt = self.prompt_template(*args, **kwargs)
+        return self.generator(prompt)
 
 
 def download_from_github(short_path: str):
@@ -146,10 +94,11 @@ def download_from_github(short_path: str):
 
 def extract_function_from_file(content: str, function_name: str) -> Tuple[Callable]:
     """Extract a function object from a downloaded file."""
+
     spec = importlib.util.spec_from_loader(
         "outlines_function", loader=None, origin="github"
     )
-    if spec is not None:  # pragma: no cover
+    if spec is not None:
         module = importlib.util.module_from_spec(spec)
         exec(content, module.__dict__)
 
diff --git a/outlines/v0_legacy/generate/__init__.py b/outlines/generate/__init__.py
similarity index 62%
rename from outlines/v0_legacy/generate/__init__.py
rename to outlines/generate/__init__.py
index 617f7b2e3..f28cbd80d 100644
--- a/outlines/v0_legacy/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -1,3 +1,4 @@
+from .api import SequenceGenerator
 from .cfg import cfg
 from .choice import choice
 from .format import format
@@ -5,13 +6,3 @@
 from .json import json
 from .regex import regex
 from .text import text
-
-__all__ = [
-    "cfg",
-    "choice",
-    "format",
-    "fsm",
-    "json",
-    "regex",
-    "text",
-]
diff --git a/outlines/generate/api.py b/outlines/generate/api.py
new file mode 100644
index 000000000..5d3c52a8a
--- /dev/null
+++ b/outlines/generate/api.py
@@ -0,0 +1,609 @@
+import datetime
+from copy import copy
+from dataclasses import dataclass
+from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
+
+from outlines.generate.generator import sequence_generator
+
+if TYPE_CHECKING:
+    import torch
+
+FormattedOutput = Union[
+    str, int, float, bool, datetime.date, datetime.time, datetime.datetime
+]
+
+
+class SequenceGenerator:
+    def __init__(
+        self,
+        fsm,
+        model,
+        sampler,
+        device,
+    ):
+        self.fsm = fsm
+        self.model = model
+        self.sampler = sampler
+        self.tokenizer = model.tokenizer
+        self.device = device
+        self.num_samples = sampler.samples
+
+    def get_generated_token_ids(
+        self,
+        prompt_token_ids: "torch.Tensor",
+        token_ids: "torch.Tensor",
+    ) -> List["torch.Tensor"]:
+        """Get the tokens generated so far.
+
+        Parameters
+        ----------
+        prompt_token_ids
+            Tensor that contains the token ids of the sequences' prompts.
+        token_ids
+            The generated token ids.
+
+        Returns
+        -------
+        A tensor that contains the token ids that have been generated so far.
+
+        """
+        prompt_lengths = [len(prompt) for prompt in prompt_token_ids]
+        token_ids = [
+            cur_token_ids[length:]
+            for cur_token_ids, length in zip(token_ids, prompt_lengths)
+        ]
+
+        return token_ids
+
+    def is_stop_sequence_found(
+        self, generated_sequences: List[str], stop_sequences: List[str]
+    ) -> bool:
+        """Determine whether one of the stop sequences has been generated.
+
+        Parameters
+        ----------
+        generated_sequences
+            The list of sequences generated so far.
+        stop_sequences
+            The list that contains the sequence which stop the generation when
+            found.
+
+        Returns
+        -------
+        True if at least one of the stop sequences has been found in each generated
+        sequence.
+
+        """
+        return all(
+            [
+                any([seq in generated for seq in stop_sequences])
+                for generated in generated_sequences
+            ]
+        )
+
+    def strip_stop_sequences(
+        self, sequence: str, stop_sequences: Optional[List[str]]
+    ) -> str:
+        """Remove the stop sequences from the generated sequences.
+
+        Parameters
+        ----------
+        sequence
+            One of the generated sequences.
+        stop_sequences
+            The list that contains the sequence which stop the generation when
+            found.
+
+        """
+        if stop_sequences:
+            match_indexes = [sequence.find(seq) for seq in stop_sequences]
+            if any([index != -1 for index in match_indexes]):
+                # select the stop_sequence that is found first in the sequence
+                min_match_index_value = min([i for i in match_indexes if i != -1])
+                min_match_index_pos = match_indexes.index(min_match_index_value)
+                sequence = sequence[
+                    : match_indexes[min_match_index_pos]
+                    + len(stop_sequences[min_match_index_pos])
+                ]
+
+        return sequence
+
+    def format_sequence(self, sequence: str) -> FormattedOutput:
+        """Translate the generated sequence to another type.
+
+        This method is for instance overridden when generating JSON to either
+        return a dictionnary or a Pydantic model.
+
+        Parameters
+        ----------
+        sequence
+            A generated sequences.
+
+        Returns
+        -------
+        The formatted sequence.
+
+        """
+        return sequence
+
+    def __call__(
+        self,
+        prompts: Union[str, List[str]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        rng: Optional["torch.Generator"] = None,
+    ) -> Union[FormattedOutput, List[FormattedOutput], List[List[FormattedOutput]]]:
+        """Generate the full text sequence.
+
+        Since `SequenceGenerator.stream` calls the tokenizer at every step this
+        method loops over the generator returned by `sequence_generator` itself
+        so the tokenizer is called only once after all token ids have been
+        generated.
+
+        Parameters
+        ----------
+        prompts
+            A string or list of strings that are passed to the model before
+            generating the first token.
+        max_tokens
+            An integer representing maximum number of tokens that will be generated
+            (per prompt)
+        stop_at
+            A string or list of strings at which the text generated will stop
+        rng
+            The random number generator. Defaults to a non-seeded `torch.Generator`
+            instance.
+
+        Returns
+        -------
+        The generation(s), potentially cast to another type.
+        """
+        import torch
+
+        if isinstance(prompts, str):
+            prompts = [prompts]
+
+        if isinstance(stop_at, str):
+            stop_at = [stop_at]
+
+        stop_sequences = stop_at
+        num_samples = self.num_samples
+
+        if rng is None:
+            rng = torch.Generator(device=self.device)
+            rng.seed()
+
+        prompt_token_ids, attention_masks = self.tokenizer.encode(prompts)
+        prompt_token_ids = prompt_token_ids.to(self.device)
+        attention_masks = attention_masks.to(self.device)
+
+        # To draw multiple samples we repeat the prompt as many times
+        # as there are samples. We copy the FSMs and initialize the
+        # FSM states.
+        num_samples = self.num_samples
+        batch_size = len(prompts)
+
+        prompt_token_ids = torch.repeat_interleave(prompt_token_ids, num_samples, dim=0)
+        attention_masks = torch.repeat_interleave(attention_masks, num_samples, dim=0)
+        fsm_states = [0 for _ in range(batch_size * num_samples)]
+        fsms = [self.fsm.copy() for _ in range(batch_size * num_samples)]
+        weights = torch.zeros(
+            (batch_size * num_samples), dtype=torch.float, device=self.device
+        )
+
+        states = sequence_generator(
+            self.model,
+            self.sampler,
+            fsms,
+            prompt_token_ids,
+            weights,
+            attention_masks,
+            fsm_states,
+            rng=rng,
+        )
+
+        while True:
+            try:
+                last_state = next(states)
+                if max_tokens or stop_sequences:
+                    token_ids = last_state.token_ids
+                    generated_token_ids = self.get_generated_token_ids(
+                        prompt_token_ids, token_ids
+                    )
+                    if max_tokens and len(generated_token_ids[0]) >= max_tokens:
+                        break
+                    if stop_sequences and self.is_stop_sequence_found(
+                        self.tokenizer.decode(generated_token_ids), stop_sequences
+                    ):
+                        break
+            except StopIteration:
+                break
+
+        token_ids = last_state.token_ids
+        generated_token_ids = self.get_generated_token_ids(prompt_token_ids, token_ids)
+
+        generated = self.tokenizer.decode(generated_token_ids)
+        stripped = [
+            self.strip_stop_sequences(sequence, stop_sequences)
+            for sequence in generated
+        ]
+        formatted = [self.format_sequence(sequence) for sequence in stripped]
+
+        # We reshape the output to (batch_size, sample_size)
+        output: List[List[FormattedOutput]] = list()
+        for i in range(0, batch_size * num_samples, num_samples):
+            output.append(formatted[i : i + num_samples])
+
+        # We remove leading dimensions for the output
+        if batch_size == 1 and num_samples == 1:
+            return output[0][0]
+        elif batch_size == 1:
+            return output[0]
+        elif num_samples == 1:
+            return [samples[0] for samples in output]
+        else:
+            return output
+
+    def stream(
+        self,
+        prompts: Union[str, List[str]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        rng: Optional["torch.Generator"] = None,
+    ) -> Iterator[Union[List[str], str, List[List[str]]]]:
+        """Generate the text sequence one token at a time.
+
+        Since `Tokenizer.decode` strips the whitespaces from the tokens we have no
+        choice but to decode the generated token ids at each step and compare the
+        current decoded strings to the previously decoded strings.
+
+        Parameters
+        ----------
+        prompts
+            A string or list of strings that are passed to the model before
+            generating the first token.
+        max_tokens
+            An integer representing maximum number of tokens that will be generated
+            (per prompt)
+        stop_at
+            A string or list of strings at which the text generated will stop
+        rng
+            The random number generator. Defaults to a non-seeded `torch.Generator`
+            instance.
+
+        Returns
+        -------
+        A string or list of strings that contain the generated text.
+
+        """
+        import torch
+
+        if isinstance(prompts, str):
+            prompts = [prompts]
+
+        if isinstance(stop_at, str):
+            stop_at = [stop_at]
+
+        stop_sequences = stop_at
+        num_samples = self.num_samples
+
+        prompt_token_ids, attention_masks = self.tokenizer.encode(prompts)
+        prompt_token_ids = prompt_token_ids.to(self.device)
+        attention_masks = attention_masks.to(prompt_token_ids.device)
+
+        # To draw multiple samples we repeat the prompt as many times
+        # as there are samples. We copy the FSMs and initialize the
+        # FSM states.
+        num_samples = self.num_samples
+        batch_size = len(prompts)
+
+        prompt_token_ids = torch.repeat_interleave(prompt_token_ids, num_samples, dim=0)
+        attention_masks = torch.repeat_interleave(attention_masks, num_samples, dim=0)
+        fsm_states = [0 for _ in range(batch_size * num_samples)]
+        fsms = [self.fsm.copy() for _ in range(batch_size * num_samples)]
+        weights = torch.zeros(
+            (batch_size * num_samples),
+            dtype=torch.float,
+            device=prompt_token_ids.device,
+        )
+
+        if rng is None:
+            rng = torch.Generator(device=prompt_token_ids.device)
+            rng.seed()
+
+        states = sequence_generator(
+            self.model,
+            self.sampler,
+            fsms,
+            prompt_token_ids,
+            weights,
+            attention_masks,
+            fsm_states,
+            rng=rng,
+        )
+
+        def token_generator() -> Iterator[Union[List[str], str, List[List[str]]]]:
+            previously_generated_sequences = [
+                "" for _ in range(batch_size)
+            ] * num_samples
+            num_generated = 0
+            is_stop_at_reached = [False for _ in range(batch_size)] * num_samples
+            while True:
+                if (max_tokens and num_generated >= max_tokens) or all(
+                    is_stop_at_reached
+                ):
+                    return
+                try:
+                    sequence = next(states)
+                    num_generated += 1
+                except StopIteration:
+                    return
+                generated_token_ids = sequence.token_ids[:, -num_generated:]
+                generated_sequences = self.tokenizer.decode(generated_token_ids)
+                if stop_sequences:
+                    is_stop_at_reached = [
+                        stop
+                        or self.is_stop_sequence_found(
+                            [generated_sequence], stop_sequences
+                        )
+                        for generated_sequence, stop in zip(
+                            generated_sequences, is_stop_at_reached
+                        )
+                    ]
+
+                    generated_sequences = [
+                        (
+                            self.format_sequence(
+                                self.strip_stop_sequences(sequence, stop_sequences)
+                            )
+                            if stop
+                            else sequence
+                        )
+                        for sequence, stop in zip(
+                            generated_sequences, is_stop_at_reached
+                        )
+                    ]
+                next_tokens = [
+                    token[len(sequence) :]
+                    for token, sequence, stop in zip(
+                        generated_sequences,
+                        previously_generated_sequences,
+                        is_stop_at_reached,
+                    )
+                ]
+                previously_generated_sequences = generated_sequences
+                # We reshape the output to (batch_size, sample_size)
+                output: List[List[str]] = list()
+                for i in range(0, batch_size * num_samples, num_samples):
+                    output.append(next_tokens[i : i + num_samples])
+
+                # We remove leading dimensions for the output
+                if batch_size == 1 and num_samples == 1:
+                    yield output[0][0]
+                elif batch_size == 1:
+                    yield output[0]
+                elif num_samples == 1:
+                    yield [samples[0] for samples in output]
+                else:
+                    yield output
+
+        return token_generator()
+
+
+@dataclass(frozen=True)
+class GenerationParameters:
+    """Generation parameters used in Outlines' public API."""
+
+    max_tokens: Optional[int]
+    stop_at: Optional[Union[str, List[str]]]
+    seed: Optional[int]
+
+
+@dataclass(frozen=True)
+class SamplingParameters:
+    """Sampling parameters available in Outlines."""
+
+    sampler: str
+    num_samples: int = 1
+    top_p: Optional[float] = None
+    top_k: Optional[int] = None
+    temperature: Optional[float] = None
+
+
+class SequenceGeneratorAdapter:
+    """Class used to unify the interface to the model providers'
+    generation functions.
+
+    Attributes
+    ----------
+    model
+        The wrapped model.
+    logits_processor
+        The logits processor to use to generate text.
+    sampler
+        The sampler to use to generate text.
+
+    """
+
+    def __init__(self, model, logits_processor, sampler):
+        self.model = model
+        self.logits_processor = logits_processor
+
+        self.sampling_params = sampler.sampling_params
+
+    def prepare_generation_parameters(
+        self,
+        max_tokens: Optional[int],
+        stop_at: Optional[Union[str, List[str]]],
+        seed: Optional[int],
+    ):
+        if isinstance(stop_at, str):
+            stop_at = [stop_at]
+
+        generation_params = GenerationParameters(
+            max_tokens,
+            stop_at,
+            seed,
+        )
+
+        return generation_params
+
+    def format_sequence(self, sequence: str) -> FormattedOutput:
+        """Translate the generated sequence to another type.
+
+        This method is for instance overridden when generating JSON to either
+        return a dictionnary or a Pydantic model.
+
+        Parameters
+        ----------
+        sequence
+            A generated sequences.
+
+        Returns
+        -------
+        The formatted sequence.
+
+        """
+        return sequence
+
+    def _format(self, sequences):
+        """Apply formatting to every string in a completion."""
+        if isinstance(sequences, list):
+            return [self._format(sequence) for sequence in sequences]
+        else:
+            return self.format_sequence(sequences)
+
+    def __call__(
+        self,
+        prompts: Union[str, List[str]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        """Generate text from a prompt of list of prompts."""
+
+        generation_params = self.prepare_generation_parameters(
+            max_tokens, stop_at, seed
+        )
+
+        completions = self.model.generate(
+            prompts,
+            generation_params,
+            copy(self.logits_processor),
+            self.sampling_params,
+            **model_specific_params,
+        )
+
+        return self._format(completions)
+
+    def stream(
+        self,
+        prompts: Union[str, List[str]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        """Return a text generator from a prompt or a list of prompts."""
+        generation_params = self.prepare_generation_parameters(
+            max_tokens, stop_at, seed
+        )
+        return self.model.stream(
+            prompts,
+            generation_params,
+            copy(self.logits_processor),
+            self.sampling_params,
+            **model_specific_params,
+        )
+
+
+class VisionSequenceGeneratorAdapter(SequenceGeneratorAdapter):
+    def __call__(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[str, Any],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        """
+        Generate text from a prompt of list of prompts.
+
+        Media: A URI to construct media or media object itself. Used as AutoProcessor argument.
+        """
+        prompts, media = self._validate_prompt_media_types(prompts, media)
+
+        generation_params = self.prepare_generation_parameters(
+            max_tokens, stop_at, seed
+        )
+
+        completions = self.model.generate(
+            prompts,
+            media,
+            generation_params,
+            copy(self.logits_processor),
+            self.sampling_params,
+            **model_specific_params,
+        )
+
+        return self._format(completions)
+
+    def stream(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: List[Union[str, Any, List[Union[str, Any]]]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        """Return a text generator from a prompt or a list of prompts."""
+        prompts, media = self._validate_prompt_media_types(prompts, media)
+        generation_params = self.prepare_generation_parameters(
+            max_tokens, stop_at, seed
+        )
+        return self.model.stream(
+            prompts,
+            media,
+            generation_params,
+            copy(self.logits_processor),
+            self.sampling_params,
+            **model_specific_params,
+        )
+
+    @classmethod
+    def _validate_prompt_media_types(
+        cls,
+        prompts: Union[str, List[str]],
+        media: Union[str, Any, List[Union[str, Any]]],
+    ) -> Union[Any, List[Any]]:
+        """
+        Prepare media as PIL.Image and ensure for every prompt str there is one List[PIL.Image]
+        """
+
+        def valid_types(prompts, media):
+            from PIL import Image  # type: ignore
+
+            if isinstance(prompts, list):
+                if not isinstance(media, list) or len(prompts) != len(media):
+                    return False
+                for subprompt, submedia in zip(prompts, media):
+                    if not isinstance(subprompt, str) or not all(
+                        isinstance(m, Image.Image) for m in submedia
+                    ):
+                        return False
+            elif isinstance(prompts, str):
+                if not all(isinstance(m, Image.Image) for m in media):
+                    return False
+            return True
+
+        if not valid_types(prompts, media):
+            raise TypeError(
+                "Expected (prompts, media) to be of type "
+                "(str, List[Image])), or (List[str], List[List[Image]]) "
+                f"instead got prompts={prompts}, media={media}"
+            )
+
+        return prompts, media
diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
new file mode 100644
index 000000000..b677040d5
--- /dev/null
+++ b/outlines/generate/cfg.py
@@ -0,0 +1,54 @@
+from functools import singledispatch
+
+from outlines.generate.api import (
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import LlamaCpp, OpenAI, TransformersVision
+from outlines.samplers import Sampler, multinomial
+
+
+@singledispatch
+def cfg(
+    model, cfg_str: str, sampler: Sampler = multinomial()
+) -> SequenceGeneratorAdapter:
+    """Generate text in the language of a Context-Free Grammar
+
+    Arguments
+    ---------
+    model:
+        An `outlines.model` instance.
+    sampler:
+        The sampling algorithm to use to generate token ids from the logits
+        distribution.
+
+    Returns
+    -------
+    A `SequenceGeneratorAdapter` instance that generates text.
+
+    """
+    from outlines.processors import CFGLogitsProcessor
+
+    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
+    return SequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@cfg.register(TransformersVision)
+def cfg_vision(model, cfg_str: str, sampler: Sampler = multinomial()):
+    from outlines.processors import CFGLogitsProcessor
+
+    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
+    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@cfg.register(LlamaCpp)
+def cfg_llamacpp(model, cfg_str: str, sampler: Sampler = multinomial()):
+    raise NotImplementedError("Not yet available due to bug in llama_cpp tokenizer")
+
+
+@cfg.register(OpenAI)
+def cfg_openai(model, cfg_str: str, sampler: Sampler = multinomial()):
+    raise NotImplementedError(
+        "Cannot use grammar-structured generation with an OpenAI model"
+        + "due to the limitations of the OpenAI API."
+    )
diff --git a/outlines/generate/choice.py b/outlines/generate/choice.py
new file mode 100644
index 000000000..afb998f52
--- /dev/null
+++ b/outlines/generate/choice.py
@@ -0,0 +1,59 @@
+import json as pyjson
+import re
+from enum import Enum
+from functools import singledispatch
+from typing import Callable, List, Union
+
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+from outlines.fsm.json_schema import get_schema_from_enum
+from outlines.generate.api import SequenceGeneratorAdapter
+from outlines.models import OpenAI
+from outlines.samplers import Sampler, multinomial
+
+from .json import json
+from .regex import regex
+
+
+@singledispatch
+def choice(
+    model, choices: Union[List[str], type[Enum]], sampler: Sampler = multinomial()
+) -> SequenceGeneratorAdapter:
+    if isinstance(choices, type(Enum)):
+        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
+    else:
+        choices = [re.escape(choice) for choice in choices]  # type: ignore
+        regex_str = r"(" + r"|".join(choices) + r")"
+
+    generator = regex(model, regex_str, sampler)
+    if isinstance(choices, type(Enum)):
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    else:
+        generator.format_sequence = lambda x: x
+
+    return generator
+
+
+@choice.register(OpenAI)
+def choice_openai(
+    model: OpenAI, choices: List[str], sampler: Sampler = multinomial()
+) -> Callable:
+    """
+    Call OpenAI API with response_format of a dict:
+    {"result": <one of choices>}
+    """
+
+    choices_schema = pyjson.dumps(
+        {
+            "type": "object",
+            "properties": {"result": {"type": "string", "enum": choices}},
+            "additionalProperties": False,
+            "required": ["result"],
+        }
+    )
+    generator = json(model, choices_schema, sampler)
+
+    def generate_choice(*args, **kwargs):
+        return generator(*args, **kwargs)["result"]
+
+    return generate_choice
diff --git a/outlines/generate/format.py b/outlines/generate/format.py
new file mode 100644
index 000000000..56ed10ac1
--- /dev/null
+++ b/outlines/generate/format.py
@@ -0,0 +1,48 @@
+from functools import singledispatch
+
+from outlines.fsm.types import python_types_to_regex
+from outlines.generate.api import SequenceGeneratorAdapter
+from outlines.models import OpenAI
+from outlines.samplers import Sampler, multinomial
+
+from .regex import regex
+
+
+@singledispatch
+def format(
+    model, python_type, sampler: Sampler = multinomial()
+) -> SequenceGeneratorAdapter:
+    """Generate structured data that can be parsed as a Python type.
+
+    Parameters
+    ----------
+    model:
+        An instance of `Transformer` that represents a model from the
+        `transformers` library.
+    python_type:
+        A Python type. The output of the generator must be parseable into
+        this type.
+    sampler:
+        The sampling algorithm to use to generate token ids from the logits
+        distribution.
+
+    Returns
+    -------
+    A `SequenceGenerator` instance that generates text constrained by the Python type
+    and translates this text into the corresponding type.
+
+    """
+    regex_str, format_fn = python_types_to_regex(python_type)
+    regex_str = regex_str.pattern
+    generator = regex(model, regex_str, sampler)
+    generator.format_sequence = format_fn
+
+    return generator
+
+
+@format.register(OpenAI)
+def format_openai(model, python_type, sampler: Sampler = multinomial()):
+    raise NotImplementedError(
+        "Cannot use Python type-structured generation with an OpenAI model"
+        + " due to the limitations of the OpenAI API."
+    )
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
new file mode 100644
index 000000000..1950812d2
--- /dev/null
+++ b/outlines/generate/fsm.py
@@ -0,0 +1,31 @@
+from functools import singledispatch
+
+import interegular
+
+from outlines.fsm.guide import RegexGuide
+from outlines.generate.api import (
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import TransformersVision
+from outlines.samplers import Sampler, multinomial
+
+
+@singledispatch
+def fsm(
+    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
+) -> SequenceGeneratorAdapter:
+    from outlines.processors import GuideLogitsProcessor
+
+    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
+    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
+    return SequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@fsm.register(TransformersVision)
+def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()):
+    from outlines.processors import GuideLogitsProcessor
+
+    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
+    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
+    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
diff --git a/outlines/generate/generator.py b/outlines/generate/generator.py
new file mode 100644
index 000000000..e506aa035
--- /dev/null
+++ b/outlines/generate/generator.py
@@ -0,0 +1,312 @@
+import dataclasses
+import math
+from typing import TYPE_CHECKING, Callable, Iterable, Iterator, List, Optional, Tuple
+
+if TYPE_CHECKING:
+    import torch
+
+    from outlines.fsm.guide import Guide
+
+
+class ContextLengthExceededError(Exception):
+    pass
+
+
+@dataclasses.dataclass(frozen=True)
+class GenerationState:
+    token_ids: "torch.Tensor"
+    kv_cache: "torch.Tensor"
+    logits: "torch.Tensor"
+    weights: "torch.Tensor"
+    fsm_states: List[int]
+
+
+def sequence_generator(
+    model: Callable,
+    sampler: Callable,
+    fsms: List["Guide"],
+    token_ids: "torch.Tensor",
+    sequence_weights: "torch.Tensor",
+    attention_masks: "torch.Tensor",
+    fsm_states: List[int],
+    rng: "torch.Generator",
+) -> Iterator[GenerationState]:
+    """Generates sequences of tokens.
+
+    Parameters
+    ----------
+    model
+        A callable that generates a probability distribution over the
+        vocabulary when passed a tensor of token ids.
+    sampler
+        A callable that returns the next token ids, their ancestor sequence and
+        the updated sequence weights when passed a distribution over the
+        vocabulary.
+    token_ids
+        A tensor of token ids on which the sequence distribution is conditioned, of
+        shape ``(n_seqs, n_prompt_tokens)``
+    sequence_weights
+        A tensor that contains the initial weights of the sequences, of shape
+        ``(n_seqs,)``
+    attention_masks
+        A tensor of tensors that represent the tokens considered at the attention
+        layer, of shape ``(n_seqs, n_prompt_tokens)``.
+    fsms
+        List of finite-state machines that drive the text generation,
+        one for each sequence in the batch.
+    fsm_states
+        The initial states of the finite-state machine for each sequence in the batch.
+
+    Yields
+    ------
+    A new sequence.
+
+    """
+    import torch
+
+    if rng is None:
+        rng = torch.Generator()
+
+    kv_cache = None
+
+    while True:
+        try:
+            logits, kv_cache = model(token_ids, attention_masks, kv_cache)
+        except IndexError:  # Exceeding the context length
+            raise ContextLengthExceededError(
+                "The input length exceeds the context length of the model."
+            )
+
+        allowed_tokens = get_allowed_tokens(fsms, fsm_states)
+        biased_logits = bias_logits(logits, allowed_tokens)
+        next_token_ids, ancestors, sequence_weights = sampler(
+            biased_logits, sequence_weights, rng
+        )
+
+        token_ids = update_token_ids(token_ids, next_token_ids, ancestors)
+        attention_masks = update_attention_masks(attention_masks, ancestors)
+        kv_cache = reorder_kv_cache(kv_cache, ancestors)
+        if len(ancestors) > 1:
+            fsms = reorder_fsms(fsms, ancestors)
+            fsm_states = reorder_fsm_states(fsm_states, ancestors)
+
+        fsm_states = get_next_fsm_states(fsms, fsm_states, next_token_ids)
+        is_finished = is_generation_finished(fsms, fsm_states)
+
+        if is_finished:
+            yield GenerationState(
+                token_ids,
+                kv_cache,
+                logits,
+                sequence_weights,
+                fsm_states,
+            )
+            return
+
+        yield GenerationState(
+            token_ids,
+            kv_cache,
+            logits,
+            sequence_weights,
+            fsm_states,
+        )
+
+
+def get_next_fsm_states(
+    fsms: List["Guide"], fsm_states: List[int], next_token_ids: "torch.Tensor"
+) -> List[int]:
+    """
+
+    Parameters
+    ----------
+    fsm
+        The finite-state machine used to monitor this batch.
+    next_token_ids
+        The tokens that were just generated.
+
+    Returns
+    -------
+    A `torch.Tensor` object that represents the next logit mask.
+
+    """
+    return [
+        fsm.get_next_state(fsm_state, int(token_id[0]))
+        for fsm, fsm_state, token_id in zip(fsms, fsm_states, next_token_ids)
+    ]
+
+
+def get_allowed_tokens(
+    fsms: List["Guide"], fsm_states: List[int]
+) -> List[Optional[Iterable[int]]]:
+    """Get the new instructions for each sequence from the finite-state machine.
+
+    Parameters
+    ----------
+    fsm
+        The finite-state machine used to monitor this batch.
+    fsm_states
+        The FSM states corresponding to each sequence in the batch.
+
+    Returns
+    -------
+    A nested list that contains the ids of the logits to keep.
+
+    """
+    return [
+        fsm.get_next_instruction(state).tokens for fsm, state in zip(fsms, fsm_states)
+    ]
+
+
+def is_generation_finished(fsms: List["Guide"], fsm_states: List[int]) -> bool:
+    """Determine if the generation is finished.
+
+    A generation is considered finished if the FSM of every sequence in the
+    batch is in a final state.
+
+    A better solution is to return finished sequences as soon as their FSM
+    is in a final state.
+
+    Parameters
+    ----------
+    fsm
+        The finite-state machine used to monitor this batch.
+    fsm_states
+        The FSM states corresponding to each sequence in the batch.
+
+    Returns
+    -------
+    Whether all sequences are finished sampling.
+
+    """
+    return all([fsm.is_final_state(state) for fsm, state in zip(fsms, fsm_states)])
+
+
+def update_token_ids(
+    token_ids: "torch.Tensor", next_token_ids: "torch.Tensor", ancestors: "torch.Tensor"
+) -> "torch.Tensor":
+    """Append the sampled tokens to the running sequence of tokens.
+
+    Parameters
+    ----------
+    token_ids
+        The current token sequences
+    next_token_ids
+        The tokens that were just generated and that we need to append
+        to the existing sequences.
+    ancestors
+        The sequences to which the token ids need to be added.
+
+    Returns
+    -------
+    A new sequence of token ids that contains the tokens that were
+    just generated.
+
+    """
+    import torch
+
+    token_ids = torch.index_select(token_ids, 0, ancestors)
+    return torch.concatenate([token_ids, next_token_ids], dim=-1)
+
+
+def update_attention_masks(
+    attention_masks: "torch.Tensor", ancestors: "torch.Tensor"
+) -> "torch.Tensor":
+    """Expand the attention masks.
+
+    Parameters
+    ----------
+    attention_masks
+        The attention masks for each sequence in the batch.
+    ancestors
+        The sequences to which the token ids need to be added.
+
+    Returns
+    -------
+    The attention masks padded with 1s.
+
+    """
+    import torch
+
+    attention_masks = torch.index_select(attention_masks, 0, ancestors)
+    return torch.concatenate(
+        [
+            attention_masks,
+            torch.ones(
+                attention_masks.shape[:-1] + (1,), device=attention_masks.device
+            ),
+        ],
+        axis=-1,
+    )
+
+
+def reorder_fsms(fsms: List["Guide"], ancestors: "torch.Tensor") -> List["Guide"]:
+    reordered_fsms = []
+    for ancestor in ancestors:
+        reordered_fsms.append(fsms[ancestor].copy())
+
+    return reordered_fsms
+
+
+def reorder_fsm_states(fsm_states: List[int], ancestors: "torch.Tensor") -> List[int]:
+    reordered_states = []
+    for ancestor in ancestors:
+        reordered_states.append(fsm_states[ancestor])
+
+    return reordered_states
+
+
+def reorder_kv_cache(
+    kv_cache: Optional[Tuple], ancestors: "torch.Tensor"
+) -> Optional[Tuple]:
+    """Re-order the KV-cache based on the ancestors.
+
+    In transformers, the object that stores the KV-cache is a tuple who elements
+    are the key cache and the value cache. Each of these caches are tuples where
+    each element correpond to a layer. To each layer corresponds a tensor whose
+    first dimension is the batch size.
+
+    """
+    import torch
+
+    if kv_cache is None:
+        return None
+
+    new_kv_cache: Tuple = tuple()
+    for cache_item in kv_cache:
+        new_cache_item: Tuple = tuple()
+        for layer in cache_item:
+            layer = torch.index_select(layer, 0, ancestors.to(layer.device))
+            new_cache_item += (layer,)
+        new_kv_cache += (new_cache_item,)
+
+    return new_kv_cache
+
+
+def bias_logits(logits: "torch.Tensor", allowed_token_ids: List) -> "torch.Tensor":
+    """Mask the logits.
+
+    The function iterates over a nested list where each list corresponds to the
+    indices that need to be masked for each row in the array.
+
+    Parameters
+    ----------
+    logits
+        Two dimensional tensor that contains the next-token probability
+        distribution.
+    allowed_token_ids
+        A list that contains the tokens that can be generated by the model.
+
+    Returns
+    -------
+    A view of the original logits tensor where some values are masked.
+
+    """
+    import torch
+
+    biased_logits = torch.full_like(logits, -math.inf, device=logits.device)
+    for i, ids in enumerate(allowed_token_ids):
+        if ids is not None:
+            biased_logits[i, ids] = logits[i, ids]
+        else:
+            biased_logits[i] = logits[i]
+    return biased_logits
diff --git a/outlines/generate/json.py b/outlines/generate/json.py
new file mode 100644
index 000000000..ab05df1f4
--- /dev/null
+++ b/outlines/generate/json.py
@@ -0,0 +1,123 @@
+import json as pyjson
+from enum import Enum
+from functools import singledispatch
+from typing import Callable, Optional, Union
+
+from genson import SchemaBuilder
+from outlines_core.fsm.json_schema import build_regex_from_schema
+from pydantic import BaseModel
+
+from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
+from outlines.generate.api import SequenceGeneratorAdapter
+from outlines.models import OpenAI
+from outlines.samplers import Sampler, multinomial
+
+from .regex import regex
+
+
+@singledispatch
+def json(
+    model,
+    schema_object: Union[str, object, Callable],
+    sampler: Sampler = multinomial(),
+    whitespace_pattern: Optional[str] = None,
+) -> SequenceGeneratorAdapter:
+    """
+    Generate structured JSON data with a `Transformer` model based on a specified JSON Schema.
+
+    Parameters
+    ----------
+    model:
+        An instance of `Transformer` that represents a model from the
+        `transformers` library.
+    schema_object:
+        The JSON Schema to generate data for. Can be a JSON string, a Pydantic model, or a callable
+        that returns a JSON schema.
+    sampler:
+        The sampling algorithm to use to generate token ids from the logits
+        distribution.
+    whitespace_pattern
+        Pattern to use for JSON syntactic whitespace (doesn't impact string literals)
+        Example: allow only a single space or newline with `whitespace_pattern=r"[\n ]?"`
+
+    Returns
+    -------
+    A `SequenceGenerator` instance that generates text constrained by the schema_object and
+    transforms the result if BaseModel is used.
+
+    """
+    if isinstance(schema_object, type(BaseModel)):
+        schema = pyjson.dumps(schema_object.model_json_schema())
+        regex_str = build_regex_from_schema(schema, whitespace_pattern)
+        generator = regex(model, regex_str, sampler)
+        generator.format_sequence = lambda x: schema_object.parse_raw(x)
+    elif isinstance(schema_object, type(Enum)):
+        schema = pyjson.dumps(get_schema_from_enum(schema_object))
+        regex_str = build_regex_from_schema(schema, whitespace_pattern)
+        generator = regex(model, regex_str, sampler)
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    elif isinstance(schema_object, SchemaBuilder):
+        schema = schema_object.to_json()
+        regex_str = build_regex_from_schema(schema, whitespace_pattern)
+        generator = regex(model, regex_str, sampler)
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    elif callable(schema_object):
+        schema = pyjson.dumps(get_schema_from_signature(schema_object))
+        regex_str = build_regex_from_schema(schema, whitespace_pattern)
+        generator = regex(model, regex_str, sampler)
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    elif isinstance(schema_object, str):
+        schema = schema_object
+        regex_str = build_regex_from_schema(schema, whitespace_pattern)
+        generator = regex(model, regex_str, sampler)
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    else:
+        raise ValueError(
+            f"Cannot parse schema {schema_object}. The schema must be either "
+            + "a Pydantic object, a function or a string that contains the JSON "
+            + "Schema specification"
+        )
+
+    return generator
+
+
+@json.register(OpenAI)
+def json_openai(
+    model, schema_object: Union[str, object], sampler: Sampler = multinomial()
+):
+    if not isinstance(sampler, multinomial):
+        raise NotImplementedError(
+            r"The OpenAI API does not support any other sampling algorithm "
+            + "than the multinomial sampler."
+        )
+
+    if isinstance(schema_object, type(BaseModel)):
+        schema = schema_object.model_json_schema()
+        schema["additionalProperties"] = False
+        schema = pyjson.dumps(schema)
+        format_sequence = lambda x: schema_object.parse_raw(x)
+    elif isinstance(schema_object, str):
+        schema = schema_object
+        format_sequence = lambda x: pyjson.loads(x)
+    else:
+        raise ValueError(
+            f"Cannot parse schema {schema_object}. The schema must be either "
+            + "a Pydantic object, a function or a string that contains the JSON "
+            + "Schema specification"
+        )
+
+    # create copied, patched model with normalized json schema set
+    generator = model.new_with_replacements(
+        response_format={
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": pyjson.loads(schema),
+            },
+        }
+    )
+
+    generator.format_sequence = format_sequence
+
+    return generator
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
new file mode 100644
index 000000000..326701f4b
--- /dev/null
+++ b/outlines/generate/regex.py
@@ -0,0 +1,66 @@
+from functools import singledispatch
+
+from outlines.generate.api import (
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import OpenAI, TransformersVision
+from outlines.samplers import Sampler, multinomial
+from outlines.types import Regex
+
+
+@singledispatch
+def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
+    """Generate structured text in the language of a regular expression.
+
+    Parameters
+    ----------
+    model:
+        An instance of `Transformer` that represents a model from the
+        `transformers` library.
+    regex_str:
+        The regular expression that the output must follow.
+    sampler:
+        The sampling algorithm to use to generate token ids from the logits
+        distribution.
+
+    Returns
+    -------
+    A `SequenceGeneratorAdapter` instance that generates text constrained by the
+    regular expression.
+
+    """
+    from outlines.processors import RegexLogitsProcessor
+
+    if isinstance(regex_str, Regex):
+        regex_str = regex_str.pattern
+
+    logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
+    return SequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@regex.register(TransformersVision)
+def regex_vision(
+    model,
+    regex_str: str | Regex,
+    sampler: Sampler = multinomial(),
+):
+    from outlines.processors import RegexLogitsProcessor
+
+    if isinstance(regex_str, Regex):
+        regex_str = regex_str.pattern
+
+    logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
+    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
+
+
+@regex.register(OpenAI)
+def regex_openai(
+    model: OpenAI,
+    regex_str: str,
+    sampler: Sampler = multinomial(),
+):
+    raise NotImplementedError(
+        "Cannot use regex-structured generation with an OpenAI model"
+        + "due to the limitations of the OpenAI API."
+    )
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
new file mode 100644
index 000000000..32530d0c4
--- /dev/null
+++ b/outlines/generate/text.py
@@ -0,0 +1,50 @@
+from functools import singledispatch
+
+from outlines.generate.api import (
+    SequenceGeneratorAdapter,
+    VisionSequenceGeneratorAdapter,
+)
+from outlines.models import OpenAI, TransformersVision
+from outlines.samplers import Sampler, multinomial
+
+
+@singledispatch
+def text(model, sampler: Sampler = multinomial()) -> SequenceGeneratorAdapter:
+    """Generate text with a `Transformer` model.
+
+    Note
+    ----
+    Python 3.11 allows dispatching on Union types and
+    this should greatly simplify the code.
+
+    Arguments
+    ---------
+    model:
+        An instance of `Transformer` that represents a model from the
+        `transformers` library.
+    sampler:
+        The sampling algorithm to use to generate token ids from the logits
+        distribution.
+
+    Returns
+    -------
+    A `SequenceGeneratorAdapter` instance that generates text.
+
+    """
+    return SequenceGeneratorAdapter(model, None, sampler)
+
+
+@text.register(TransformersVision)
+def text_vision(model, sampler: Sampler = multinomial()):
+    return VisionSequenceGeneratorAdapter(model, None, sampler)
+
+
+@text.register(OpenAI)
+def text_openai(model: OpenAI, sampler: Sampler = multinomial()) -> OpenAI:
+    if not isinstance(sampler, multinomial):
+        raise NotImplementedError(
+            r"The OpenAI API does not support any other sampling algorithm "
+            + "than the multinomial sampler."
+        )
+
+    return model
diff --git a/outlines/generator.py b/outlines/generator.py
deleted file mode 100644
index 1623fe85e..000000000
--- a/outlines/generator.py
+++ /dev/null
@@ -1,347 +0,0 @@
-"""Encapsulate a model and an output type into a reusable object."""
-
-from typing import (
-    Any,
-    AsyncIterator,
-    Iterator,
-    Optional,
-    Union,
-)
-
-from outlines.models import (
-    AsyncBlackBoxModel,
-    BlackBoxModel,
-    SteerableModel,
-)
-from outlines.models.base import AsyncModel, Model
-from outlines.processors import (
-    CFGLogitsProcessor,
-    GuideLogitsProcessor,
-    OutlinesLogitsProcessor,
-    RegexLogitsProcessor,
-)
-from outlines.processors.guide import RegexGuide
-from outlines.types import CFG, FSM
-from outlines.types.dsl import python_types_to_terms, to_regex
-
-
-class BlackBoxGenerator:
-    """Synchronous generator for which we don't control constrained
-    generation.
-
-    The output type provided is not compiled into a logits processor, but is
-    instead directly passed on to the model.
-
-    """
-    output_type: Optional[Any]
-
-    def __init__(self, model: BlackBoxModel, output_type: Optional[Any]):
-        """
-        Parameters
-        ----------
-        model
-            An instance of an Outlines model.
-        output_type
-            The output type that will be used to constrain the generation.
-
-        """
-        self.model = model
-        self.output_type = output_type
-
-        if isinstance(self.output_type, FSM):
-            raise NotImplementedError(
-                "FSM generation is not supported for API-based models"
-            )
-
-    def __call__(self, prompt: Any, **inference_kwargs) -> Any:
-        """Generate a response from the model.
-
-        Parameters
-        ----------
-        prompt
-            The prompt to use to generate a response.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        return self.model.generate(
-            prompt, self.output_type, **inference_kwargs
-        )
-
-    def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
-        """Generate a stream of responses from the model.
-
-        Parameters
-        ----------
-        prompt
-            The prompt to use to generate a response.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        return self.model.generate_stream(
-            prompt, self.output_type, **inference_kwargs
-        )
-
-
-class AsyncBlackBoxGenerator:
-    """Asynchronous generator for which we don't control constrained
-    generation.
-
-    The output type provided is not compiled into a logits processor, but is
-    instead directly passed on to the model.
-
-    """
-    output_type: Optional[Any]
-
-    def __init__(self, model: AsyncBlackBoxModel, output_type: Optional[Any]):
-        """
-        Parameters
-        ----------
-        model
-            An instance of an Outlines model.
-        output_type
-            The output type that will be used to constrain the generation.
-
-        """
-        self.model = model
-        self.output_type = output_type
-
-        if isinstance(self.output_type, FSM):
-            raise NotImplementedError(
-                "FSM generation is not supported for API-based models"
-            )
-
-    async def __call__(self, prompt: Any, **inference_kwargs) -> Any:
-        """Generate a response from the model.
-
-        Parameters
-        ----------
-        prompt
-            The prompt to use to generate a response.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        return await self.model.generate(
-            prompt, self.output_type, **inference_kwargs
-        )
-
-    async def stream(
-        self, prompt: Any, **inference_kwargs
-    ) -> AsyncIterator[Any]:
-        """Generate a stream of responses from the model.
-
-        Parameters
-        ----------
-        prompt
-            The prompt to use to generate a response.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        async for chunk in self.model.generate_stream(  # pragma: no cover
-            prompt, self.output_type, **inference_kwargs
-        ):
-            yield chunk
-
-
-class SteerableGenerator:
-    """Represents a generator for which we control constrained generation.
-
-    The generator is responsible for building and storing the logits processor
-    (which can be quite expensive to build), and then passing it to the model
-    when the generator is called.
-
-    The argument defining constrained generation can be of 2 types associated
-    to different methods to create an instance of the generator:
-    - `output_type` (through `__init__`): an output type as defined in the
-      `outlines.types` module
-    - `processor` (through `from_processor`): an already built logits processor
-       as defined in the `outlines.processors` module
-
-    The 2 parameters are mutually exclusive.
-
-    """
-    logits_processor: Optional[OutlinesLogitsProcessor]
-
-    def __init__(self, model: SteerableModel, output_type: Optional[Any]):
-        """
-        Parameters
-        ----------
-        model
-            An instance of an Outlines model.
-        output_type
-            The output type expressed as a Python type
-
-        """
-        self.model = model
-        if output_type is None:
-            self.logits_processor = None
-        else:
-            term = python_types_to_terms(output_type)
-            if isinstance(term, CFG):
-                cfg_string = term.definition
-                self.logits_processor = CFGLogitsProcessor(
-                    cfg_string,
-                    self.model.tokenizer,
-                    self.model.tensor_library_name,
-                )
-            elif isinstance(term, FSM):
-                guide = RegexGuide.from_interegular_fsm(
-                    term.fsm,
-                    self.model.tokenizer,
-                )
-                self.logits_processor = GuideLogitsProcessor(
-                    self.model.tokenizer, guide, self.model.tensor_library_name
-                )
-            else:
-                regex_string = to_regex(term)
-                self.logits_processor = RegexLogitsProcessor(
-                    regex_string,
-                    self.model.tokenizer,
-                    self.model.tensor_library_name,
-                )
-
-    @classmethod
-    def from_processor(
-        cls, model: SteerableModel, processor: OutlinesLogitsProcessor
-    ):
-        """Create a generator from a logits processor.
-
-        Parameters
-        ----------
-        model
-            An instance of an Outlines model.
-        processor
-            An instance of an OutlinesLogitsProcessor.
-
-        """
-        if not isinstance(processor, OutlinesLogitsProcessor):
-            raise TypeError(
-                "The processor argument must be an instance of "
-                "OutlinesLogitsProcessor"
-            )
-        instance = cls.__new__(cls)
-        instance.model = model
-        instance.logits_processor = processor
-
-        return instance
-
-    def __call__(self, prompt: Any, **inference_kwargs) -> Any:
-        """Generate a response from the model.
-
-        Parameters
-        ----------
-        prompt
-            The prompt to use to generate a response.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        return self.model.generate(
-            prompt, self.logits_processor, **inference_kwargs
-        )
-
-    def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
-        """Generate a stream of responses from the model.
-
-        Parameters
-        ----------
-        prompt
-            The prompt to use to generate a response.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        return self.model.generate_stream(
-            prompt, self.logits_processor, **inference_kwargs
-        )
-
-
-def Generator(
-    model: Union[Model, AsyncModel],
-    output_type: Optional[Any] = None,
-    *,
-    processor: Optional[OutlinesLogitsProcessor] = None,
-) -> Union[SteerableGenerator, BlackBoxGenerator, AsyncBlackBoxGenerator]:
-    """Create a generator for the given model and output parameters.
-
-    The 2 parameters output_type and processor are mutually exclusive. The
-    parameters processor is only supported for SteerableModel instances
-    (typically local models) and is intended to be only used by advanced users.
-
-    Parameters
-    ----------
-    model
-        An instance of an Outlines model.
-    output_type
-        The output type expressed as a Python type or a type defined in the
-        outlines.types.dsl module.
-    processor
-        An instance of an OutlinesLogitsProcessor.
-
-    Returns
-    -------
-    Union[SteerableGenerator, BlackBoxGenerator, AsyncBlackBoxGenerator]
-        A generator instance.
-
-    """
-    provided_output_params = sum(
-        param is not None
-        for param in [output_type, processor]
-    )
-    if provided_output_params > 1:
-        raise ValueError(
-            "At most one of output_type or processor can be provided"
-        )
-
-    if isinstance(model, SteerableModel): # type: ignore
-        if processor is not None:
-            return SteerableGenerator.from_processor(model, processor) # type: ignore
-        else:
-            return SteerableGenerator(model, output_type) # type: ignore
-    else:
-        if processor is not None:
-            raise NotImplementedError(
-                "This model does not support logits processors"
-            )
-        if isinstance(model, AsyncBlackBoxModel): # type: ignore
-            return AsyncBlackBoxGenerator(model, output_type) # type: ignore
-        elif isinstance(model, BlackBoxModel): # type: ignore
-            return BlackBoxGenerator(model, output_type) # type: ignore
-        else:
-            raise ValueError(
-                "The model argument must be an instance of "
-                "SteerableModel, BlackBoxModel or AsyncBlackBoxModel"
-            )
diff --git a/outlines/grammars.py b/outlines/grammars.py
index e2050148a..f0c122964 100644
--- a/outlines/grammars.py
+++ b/outlines/grammars.py
@@ -1,29 +1,10 @@
-"""A few common Lark grammars."""
-
 from pathlib import Path
 
 GRAMMAR_PATH = Path(__file__).parent / "grammars"
 
 
-def read_grammar(
-    grammar_file_name: str,
-    base_grammar_path: Path = GRAMMAR_PATH,
-) -> str:
-    """Read grammar file from default grammar path.
-
-    Parameters
-    ----------
-    grammar_file_name
-        The name of the grammar file to read.
-    base_grammar_path
-        The path to the directory containing the grammar file.
-
-    Returns
-    -------
-    str
-        The contents of the grammar file.
-
-    """
+def read_grammar(grammar_file_name, base_grammar_path=GRAMMAR_PATH):
+    """Read grammar file from default grammar path"""
     full_path = base_grammar_path / grammar_file_name
     with open(full_path) as file:
         return file.read()
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index d818c5db5..fe6f861ac 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -8,39 +8,12 @@
 
 from typing import Union
 
-from .anthropic import Anthropic, from_anthropic
-from .base import Model, ModelTypeAdapter
-from .dottxt import Dottxt, from_dottxt
-from .gemini import Gemini, from_gemini
-from .llamacpp import LlamaCpp, from_llamacpp
-from .mlxlm import MLXLM, from_mlxlm
-from .ollama import Ollama, from_ollama
-from .openai import OpenAI, from_openai
-from .sglang import AsyncSGLang, SGLang, from_sglang
-from .tgi import AsyncTGI, TGI, from_tgi
-from .transformers import (
-    Transformers,
-    TransformerTokenizer,
-    TransformersMultiModal,
-    from_transformers,
-)
-from .vllm_offline import VLLMOffline, from_vllm_offline
-from .vllm import AsyncVLLM, VLLM, from_vllm
+from .exllamav2 import ExLlamaV2Model, exl2
+from .llamacpp import LlamaCpp, llamacpp
+from .mlxlm import MLXLM, mlxlm
+from .openai import OpenAI, azure_openai, openai
+from .transformers import Transformers, TransformerTokenizer, mamba, transformers
+from .transformers_vision import TransformersVision, transformers_vision
+from .vllm import VLLM, vllm
 
-SteerableModel = Union[LlamaCpp, MLXLM, Transformers]
-BlackBoxModel = Union[
-    Anthropic,
-    Dottxt,
-    Gemini,
-    Ollama,
-    OpenAI,
-    SGLang,
-    TGI,
-    VLLM,
-    VLLMOffline,
-]
-AsyncBlackBoxModel = Union[
-    AsyncTGI,
-    AsyncSGLang,
-    AsyncVLLM,
-]
+LogitsGenerator = Union[Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM]
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
deleted file mode 100644
index 4ed3828d9..000000000
--- a/outlines/models/anthropic.py
+++ /dev/null
@@ -1,227 +0,0 @@
-"""Integration with Anthropic's API."""
-
-from typing import TYPE_CHECKING, Any, Iterator, Optional, Union
-
-from outlines.models.base import Model, ModelTypeAdapter
-from outlines.templates import Vision
-
-if TYPE_CHECKING:
-    from anthropic import Anthropic as AnthropicClient
-
-__all__ = ["Anthropic", "from_anthropic"]
-
-
-class AnthropicTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `Anthropic` model.
-
-    `AnthropicTypeAdapter` is responsible for preparing the arguments to
-    Anthropic's `messages.create` method: the input (prompt and possibly
-    image).
-    Anthropic does not support defining the output type, so
-    `format_output_type` is not implemented.
-
-    """
-
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
-        """Generate the `messages` argument to pass to the client.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        dict
-            The `messages` argument to pass to the client.
-
-        """
-        if isinstance(model_input, str):
-            return self.format_str_model_input(model_input)
-        elif isinstance(model_input, Vision):
-            return self.format_vision_model_input(model_input)
-        raise TypeError(
-            f"The input type {input} is not available with Anthropic. "
-            "The only available types are `str` and `Vision`."
-        )
-
-    def format_str_model_input(self, model_input: str) -> dict:
-        return {
-            "messages": [
-                {
-                    "role": "user",
-                    "content": model_input,
-                }
-            ]
-        }
-
-    def format_vision_model_input(self, model_input: Vision) -> dict:
-        return {
-            "messages": [
-                {
-                    "role": "user",
-                    "content": [
-                        {
-                            "type": "image",
-                            "source": {
-                                "type": "base64",
-                                "media_type": model_input.image_format,
-                                "data": model_input.image_str,
-                            },
-                        },
-                        {"type": "text", "text": model_input.prompt},
-                    ],
-                }
-            ]
-        }
-
-    def format_output_type(self, output_type):
-        """Not implemented for Anthropic."""
-        if output_type is None:
-            return {}
-        else:
-            raise NotImplementedError(
-                f"The output type {output_type} is not available with "
-                "Anthropic."
-            )
-
-
-class Anthropic(Model):
-    """Thin wrapper around the `anthropic.Anthropic` client.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `anthropic.Anthropic` client.
-
-    """
-    def __init__(
-        self, client: "AnthropicClient", model_name: Optional[str] = None
-    ):
-        """
-        Parameters
-        ----------
-        client
-            An `anthropic.Anthropic` client.
-        model_name
-            The name of the model to use.
-
-        """
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = AnthropicTypeAdapter()
-
-    def generate(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> str:
-        """Generate text using Anthropic.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            As structured generation is not supported by Anthropic, the value
-            of this argument must be `None`. Otherwise, an error will be
-            raised at runtime.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        str
-            The response generated by the model.
-
-        """
-        messages = self.type_adapter.format_input(model_input)
-
-        if output_type is not None:
-            raise NotImplementedError(
-                f"The type {output_type} is not available with Anthropic."
-            )
-
-        if (
-            "model" not in inference_kwargs
-            and self.model_name is not None
-        ):
-            inference_kwargs["model"] = self.model_name
-
-        completion = self.client.messages.create(
-            **messages,
-            **inference_kwargs,
-        )
-        return completion.content[0].text
-
-    def generate_stream(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Iterator[str]:
-        """Stream text using Anthropic.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            As structured generation is not supported by Anthropic, the value
-            of this argument must be `None`. Otherwise, an error will be
-            raised at runtime.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
-
-        """
-        messages = self.type_adapter.format_input(model_input)
-
-        if output_type is not None:
-            raise NotImplementedError(
-                f"The type {output_type} is not available with Anthropic."
-            )
-
-        if (
-            "model" not in inference_kwargs
-            and self.model_name is not None
-        ):
-            inference_kwargs["model"] = self.model_name
-
-        stream = self.client.messages.create(
-            **messages,
-            stream=True,
-            **inference_kwargs,
-        )
-
-        for chunk in stream:
-            if (
-                chunk.type == "content_block_delta"
-                and chunk.delta.type == "text_delta"
-            ):
-                yield chunk.delta.text
-
-
-def from_anthropic(
-    client: "AnthropicClient", model_name: Optional[str] = None
-) -> Anthropic:
-    """Create an Outlines `Anthropic` model instance from an
-    `anthropic.Anthropic` client instance.
-
-    Parameters
-    ----------
-    client
-        An `anthropic.Anthropic` client instance.
-    model_name
-        The name of the model to use.
-
-    Returns
-    -------
-    Anthropic
-        An Outlines `Anthropic` model instance.
-
-    """
-    return Anthropic(client, model_name)
diff --git a/outlines/models/base.py b/outlines/models/base.py
deleted file mode 100644
index 08beece18..000000000
--- a/outlines/models/base.py
+++ /dev/null
@@ -1,377 +0,0 @@
-"""Base classes for all models and model type adapters."""
-
-from abc import ABC, abstractmethod
-from typing import Any, AsyncIterator, Iterator, Optional
-
-
-class ModelTypeAdapter(ABC):
-    """Base class for all model type adapters.
-
-    A type adapter instance must be given as a value to the `type_adapter`
-    attribute when instantiating a model.
-    The type adapter is responsible for formatting the input and output types
-    passed to the model to match the specific format expected by the
-    associated model.
-
-    """
-
-    @abstractmethod
-    def format_input(self, model_input: Any) -> Any:
-        """Format the user input to the expected format of the model.
-
-        For API-based models, it typically means creating the `messages`
-        argument passed to the client. For local models, it can mean casting
-        the input from str to list for instance.
-        This method is also used to validate that the input type provided by
-        the user is supported by the model.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        Any
-            The formatted input to be passed to the model.
-
-        """
-        ...
-
-    @abstractmethod
-    def format_output_type(self, output_type: Optional[Any] = None) -> Any:
-        """Format the output type to the expected format of the model.
-
-        For API-based models, this typically means creating a `response_format`
-        argument. For local models, it means formatting the logits processor to
-        create the object type expected by the model.
-
-        Parameters
-        ----------
-        output_type
-            The output type provided by the user.
-
-        Returns
-        -------
-        Any
-            The formatted output type to be passed to the model.
-
-        """
-        ...
-
-class Model(ABC):
-    """Base class for all synchronous models.
-
-    This class defines a shared `__call__` method that can be used to call the
-    model directly.
-    All models inheriting from this class must define a `type_adapter`
-    attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
-    attribute are used in the `generate` method to format the input and output
-    types received by the model.
-    Additionally, local models must define a `tensor_library_name` attribute.
-
-    """
-    type_adapter: ModelTypeAdapter
-    tensor_library_name: str
-
-    def __call__(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> Any:
-        """Call the model.
-
-        Users can call the model directly, in which case we will create a
-        generator instance with the output type provided and call it.
-        Thus, those commands are equivalent:
-        ```python
-        generator = Generator(model, Foo)
-        generator("prompt")
-        ```
-        and
-        ```python
-        model("prompt", Foo)
-        ```
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        from outlines import Generator
-
-        return Generator(self, output_type)(model_input, **inference_kwargs)
-
-    def stream(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> Iterator[Any]:
-        """Stream a response from the model.
-
-        Users can use the `stream` method from the model directly, in which
-        case we will create a generator instance with the output type provided
-        and then invoke its `stream` method.
-        Thus, those commands are equivalent:
-        ```python
-        generator = Generator(model, Foo)
-        for chunk in generator("prompt"):
-            print(chunk)
-        ```
-        and
-        ```python
-        for chunk in model.stream("prompt", Foo):
-            print(chunk)
-        ```
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Iterator[Any]
-            A stream of responses from the model.
-
-        """
-        from outlines import Generator
-
-        generator = Generator(self, output_type)
-        return generator.stream(model_input, **inference_kwargs) # type: ignore
-
-    @abstractmethod
-    def generate(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> Any:
-        """Generate a response from the model.
-
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
-        This method is not intended to be used directly by end users.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        ...
-
-    @abstractmethod
-    def generate_stream(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> Iterator[Any]:
-        """Generate a stream of responses from the model.
-
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
-        This method is not intended to be used directly by end users.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Iterator[Any]
-            A stream of responses from the model.
-
-        """
-        ...
-
-class AsyncModel(ABC):
-    """Base class for all asynchronous models.
-
-    This class defines a shared `__call__` method that can be used to call the
-    model directly.
-    All models inheriting from this class must define a `type_adapter`
-    attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
-    attribute are used in the `generate` method to format the input and output
-    types received by the model.
-    Additionally, local models must define a `tensor_library_name` attribute.
-
-    """
-    type_adapter: ModelTypeAdapter
-    tensor_library_name: str
-
-    async def __call__(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> Any:
-        """Call the model.
-
-        Users can call the model directly, in which case we will create a
-        generator instance with the output type provided and call it.
-        Thus, those commands are equivalent:
-        ```python
-        generator = Generator(model, Foo)
-        await generator("prompt")
-        ```
-        and
-        ```python
-        await model("prompt", Foo)
-        ```
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        from outlines import Generator
-
-        generator = Generator(self, output_type)
-        return await generator(model_input, **inference_kwargs)
-
-    async def stream(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> AsyncIterator[Any]:
-        """Stream a response from the model.
-
-        Users can use the `stream` method from the model directly, in which
-        case we will create a generator instance with the output type provided
-        and then invoke its `stream` method.
-        Thus, those commands are equivalent:
-        ```python
-        generator = Generator(model, Foo)
-        async for chunk in generator("prompt"):
-            print(chunk)
-        ```
-        and
-        ```python
-        async for chunk in model.stream("prompt", Foo):
-            print(chunk)
-        ```
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        AsyncIterator[Any]
-            A stream of responses from the model.
-
-        """
-        from outlines import Generator
-
-        generator = Generator(self, output_type)
-
-        async for chunk in generator.stream(model_input, **inference_kwargs):  # type: ignore
-            yield chunk
-
-    @abstractmethod
-    async def generate(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> Any:
-        """Generate a response from the model.
-
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
-        This method is not intended to be used directly by end users.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        Any
-            The response generated by the model.
-
-        """
-        ...
-
-    @abstractmethod
-    async def generate_stream(
-        self,
-        model_input: Any,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any
-    ) -> AsyncIterator[Any]:
-        """Generate a stream of responses from the model.
-
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
-        This method is not intended to be used directly by end users.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-        output_type
-            The output type provided by the user.
-        **inference_kwargs
-            Additional keyword arguments to pass to the model.
-
-        Returns
-        -------
-        AsyncIterator[Any]
-            A coroutine that will produce an async iterator of responses from the model.
-
-        """
-        ...
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
deleted file mode 100644
index 62c40a7c5..000000000
--- a/outlines/models/dottxt.py
+++ /dev/null
@@ -1,215 +0,0 @@
-"""Integration with Dottxt's API."""
-
-import json
-from typing import TYPE_CHECKING, Any, Optional
-
-from pydantic import TypeAdapter
-
-from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import CFG, JsonSchema, Regex
-from outlines.types.utils import (
-    is_dataclass,
-    is_genson_schema_builder,
-    is_pydantic_model,
-    is_typed_dict,
-)
-
-if TYPE_CHECKING:
-    from dottxt import Dottxt as DottxtClient
-
-__all__ = ["Dottxt", "from_dottxt"]
-
-
-class DottxtTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `Dottxt` model."""
-
-    def format_input(self, model_input: str) -> str:
-        """Format the prompt to pass to the client.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        str
-            The input to pass to the client.
-
-        """
-        if isinstance(model_input, str):
-            return model_input
-        raise TypeError(
-            f"The input type {model_input} is not available with Dottxt. "
-            "The only available type is `str`."
-        )
-
-    def format_output_type(self, output_type: Optional[Any] = None) -> str:
-        """Format the output type to pass to the client.
-
-        TODO: `int`, `float` and other Python types could be supported via
-        JSON Schema.
-
-        Parameters
-        ----------
-        output_type
-            The output type provided by the user.
-
-        Returns
-        -------
-        str
-            The output type to pass to the client.
-
-        """
-        # Unsupported languages
-        if output_type is None:
-            raise TypeError(
-                "You must provide an output type. Dottxt only supports "
-                "constrained generation."
-            )
-        elif isinstance(output_type, Regex):
-            raise TypeError(
-                "Regex-based structured outputs will soon be available with "
-                "Dottxt. Use an open source model in the meantime."
-            )
-        elif isinstance(output_type, CFG):
-            raise TypeError(
-                "CFG-based structured outputs will soon be available with "
-                "Dottxt. Use an open source model in the meantime."
-            )
-
-        elif isinstance(output_type, JsonSchema):
-            return output_type.schema
-        elif is_dataclass(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return json.dumps(schema)
-        elif is_typed_dict(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return json.dumps(schema)
-        elif is_pydantic_model(output_type):
-            schema = output_type.model_json_schema()
-            return json.dumps(schema)
-        elif is_genson_schema_builder(output_type):
-            return output_type.to_json()
-        else:
-            type_name = getattr(output_type, "__name__", output_type)
-            raise TypeError(
-                f"The type `{type_name}` is not supported by Dottxt. "
-                "Consider using a local mode instead."
-            )
-
-
-class Dottxt(Model):
-    """Thin wrapper around the `dottxt.client.Dottxt` client.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `dottxt.client.Dottxt` client.
-
-    """
-
-    def __init__(
-        self,
-        client: "DottxtClient",
-        model_name: Optional[str] = None,
-        model_revision: Optional[str] = None,
-    ):
-        """
-        Parameters
-        ----------
-        client
-            A `dottxt.Dottxt` client.
-        model_name
-            The name of the model to use.
-        model_revision
-            The revision of the model to use.
-
-        """
-        self.client = client
-        self.model_name = model_name
-        self.model_revision = model_revision
-        self.type_adapter = DottxtTypeAdapter()
-
-    def generate(
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> str:
-        """Generate text using Dottxt.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. The
-            output type must be of a type that can be converted to a JSON
-            schema.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        str
-            The text generated by the model.
-
-        """
-        prompt = self.type_adapter.format_input(model_input)
-        json_schema = self.type_adapter.format_output_type(output_type)
-
-        if (
-            "model_name" not in inference_kwargs
-            and self.model_name is not None
-        ):
-            inference_kwargs["model_name"] = self.model_name
-
-        if (
-            "model_revision" not in inference_kwargs
-            and self.model_revision is not None
-        ):
-            inference_kwargs["model_revision"] = self.model_revision
-
-        completion = self.client.json(
-            prompt,
-            json_schema,
-            **inference_kwargs,
-        )
-        return completion.data
-
-    def generate_stream(
-        self,
-        model_input,
-        output_type=None,
-        **inference_kwargs,
-    ):
-        """Not available for Dottxt."""
-        raise NotImplementedError(
-            "Dottxt does not support streaming. Call the model/generator for "
-            + "regular generation instead."
-        )
-
-
-def from_dottxt(
-    client: "DottxtClient",
-    model_name: Optional[str] = None,
-    model_revision: Optional[str] = None,
-) -> Dottxt:
-    """Create an Outlines `Dottxt` model instance from a `dottxt.Dottxt`
-    client instance.
-
-    Parameters
-    ----------
-    client
-        A `dottxt.Dottxt` client instance.
-    model_name
-        The name of the model to use.
-    model_revision
-        The revision of the model to use.
-
-    Returns
-    -------
-    Dottxt
-        An Outlines `Dottxt` model instance.
-
-    """
-    return Dottxt(client, model_name, model_revision)
diff --git a/outlines/v0_legacy/models/exllamav2.py b/outlines/models/exllamav2.py
similarity index 79%
rename from outlines/v0_legacy/models/exllamav2.py
rename to outlines/models/exllamav2.py
index 683512002..efd70f96a 100644
--- a/outlines/v0_legacy/models/exllamav2.py
+++ b/outlines/models/exllamav2.py
@@ -1,30 +1,18 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import (
-    TYPE_CHECKING,
-    Dict,
-    Iterator,
-    List,
-    Optional,
-    Tuple,
-    TypedDict,
-    Union
-)
+import dataclasses
+from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, TypedDict, Union
 
-if TYPE_CHECKING:
-    import torch.LongTensor
-    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler  # type: ignore
+import torch
+from typing_extensions import Unpack
 
+from outlines.generate.api import GenerationParameters, SamplingParameters
 
-def exllamav2_params_adapter(sampling_params: dict, generation_params: dict, kwargs: dict) -> dict:
-    """Argument names mapping is done in the model class for the `exllamav2` model."""
-    kwargs.update(**sampling_params, **generation_params)
-    return kwargs
+if TYPE_CHECKING:
+    import torch.LongTensor
+    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler
 
 
 class ExllamaV2Params(TypedDict, total=False):
+    max_tokens: int
     stop_conditions: Optional[List[Union[int, str]]]
     seed: Optional[int]
     gen_settings: "ExLlamaV2Sampler.Settings"
@@ -42,8 +30,6 @@ def convert_token_to_string(self, token):
         return token
 
     def decode(self, token_ids: "torch.LongTensor") -> List[str]:
-        import torch
-
         decoded = self.exl2_tokenizer.decode(
             torch.tensor(token_ids),
             decode_special_tokens=False,
@@ -53,7 +39,7 @@ def decode(self, token_ids: "torch.LongTensor") -> List[str]:
         return decoded
 
 
-class ExLlamaV2Model():
+class ExLlamaV2Model:
     """Represents a `exl2` model."""
 
     def __init__(
@@ -62,37 +48,18 @@ def __init__(
         tokenizer: "OutlinesExLlamaV2Tokenizer",
         max_seq_len: int,
     ):
-        warnings.warn("""
-            The `ExLlamaV2Model` model is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.1.0.
-            As the `exllamav2` library is not compatible with some key features
-            of `outlines`, we decided to remove support for it.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
         self.generator = generator
         self.tokenizer = tokenizer
         self.max_seq_len = max_seq_len
 
-    def __call__(self, *args, **kwargs):
-        raise NotImplementedError(
-            "As a v0 model, the `exllamav2` model does not support the "
-            + "`__call__` method."
-        )
-
-    def stream(self, *args, **kwargs):
-        raise NotImplementedError(
-            "As a v0 model, the `exllamav2` model does not support the "
-            + "`stream` method."
-        )
-
     def prepare_generation_parameters(
         self,
         prompts: Union[str, List[str]],
-        logits_processor,
-        kwargs
-    ) -> Tuple[Dict, Union[str, List[str]]]:
+        generation_parameters: GenerationParameters,
+        sampling_parameters: SamplingParameters,
+        structure_logits_processor,
+        **exllamav2_params: Unpack[ExllamaV2Params],
+    ) -> Tuple[ExllamaV2Params, Union[str, List[str]]]:
         """Prepare the generation parameters.
 
         `exllamav2` uses different default values
@@ -100,12 +67,10 @@ def prepare_generation_parameters(
         """
         from exllamav2.generator import ExLlamaV2Sampler
 
-        exllamav2_params = {}
-
         if isinstance(prompts, str):
             prompts = [prompts]
+        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
 
-        max_tokens = kwargs.get("max_tokens")
         if max_tokens is None:
             max_tokens = []
             for prompt in prompts:
@@ -120,40 +85,28 @@ def prepare_generation_parameters(
                 max_tokens for _ in range(len(prompts))
             ]
 
-        stop_at = kwargs.get("stop_at")
         stop_conditions = [self.generator.tokenizer.eos_token_id]
-        if isinstance(stop_at, str):
-            stop_conditions.append(stop_at)
-        elif isinstance(stop_at, list):
-            for stop_at in stop_at:
+        if isinstance(generation_parameters.stop_at, str):
+            stop_conditions.append(generation_parameters.stop_at)
+        elif isinstance(generation_parameters.stop_at, list):
+            for stop_at in generation_parameters.stop_at:
                 stop_conditions.append(stop_at)
         exllamav2_params["stop_conditions"] = stop_conditions
-
-        exllamav2_params["seed"] = kwargs.get("seed")
+        exllamav2_params["seed"] = seed
 
         gen_settings = ExLlamaV2Sampler.Settings()
-
-        temperature = kwargs.get("temperature")
-        if temperature is not None:
-            gen_settings.temperature = temperature
-
-        top_p = kwargs.get("top_p")
-        if top_p is not None:
-            gen_settings.top_p = top_p
-
-        top_k = kwargs.get("top_k")
-        if top_k is not None:
-            gen_settings.top_k = top_k
-
-        gen_settings.logits_processor = logits_processor
-
+        if sampling_parameters.temperature is not None:
+            gen_settings.temperature = sampling_parameters.temperature
+        if sampling_parameters.top_p is not None:
+            gen_settings.top_p = sampling_parameters.top_p
+        if sampling_parameters.top_k is not None:
+            gen_settings.top_k = sampling_parameters.top_k
+        gen_settings.logits_processor = structure_logits_processor
         exllamav2_params["gen_settings"] = gen_settings
-
-        num_samples = kwargs.get("num_samples")
-        if num_samples is not None and num_samples > 1:
-            prompts = prompts * num_samples
+        if sampling_parameters.num_samples > 1:
+            prompts = prompts * sampling_parameters.num_samples
             exllamav2_params["max_new_tokens"] = (
-                exllamav2_params["max_new_tokens"] * num_samples
+                exllamav2_params["max_new_tokens"] * sampling_parameters.num_samples
             )
 
         if len(prompts) == 1:
@@ -162,7 +115,7 @@ def prepare_generation_parameters(
         return exllamav2_params, prompts
 
     def reformat_output(
-        self, output: Union[str, List[str]], sampling_parameters
+        self, output: Union[str, List[str]], sampling_parameters: SamplingParameters
     ):
         """
         The purpose of this function is to reformat the output from exllamav2's output format to outline's output format.
@@ -195,13 +148,16 @@ def reformat_output(
     def generate(
         self,
         prompts: Union[str, List[str]],
-        logits_processor,
-        **kwargs
+        generation_parameters: GenerationParameters,
+        structure_logits_processor,
+        sampling_parameters: SamplingParameters,
+        **exllamav2_params: Unpack[ExllamaV2Params],
     ) -> Union[str, List[str]]:
         exllamav2_params, prompts = self.prepare_generation_parameters(
             prompts,
-            logits_processor,
-            kwargs
+            generation_parameters,
+            sampling_parameters,
+            structure_logits_processor,
         )
         """
         In exllamav2, it needs the max amount of new tokens generated.
@@ -221,20 +177,23 @@ def generate(
             seed=exllamav2_params["seed"],
         )
 
-        return self.reformat_output(output, kwargs)
+        return self.reformat_output(output, sampling_parameters)
 
-    def generate_stream(
+    def stream(
         self,
         prompts: Union[str, List[str]],
-        logits_processor,
-        **kwargs
+        generation_parameters: GenerationParameters,
+        structure_logits_processor,
+        sampling_parameters: SamplingParameters,
+        **exllamav2_params: Unpack[ExllamaV2Params],
     ) -> Iterator[Union[str, List[str]]]:
         from exllamav2.generator import ExLlamaV2DynamicJob
 
         exllamav2_params, prompts = self.prepare_generation_parameters(
             prompts,
-            logits_processor,
-            kwargs
+            generation_parameters,
+            sampling_parameters,
+            structure_logits_processor,
         )
 
         order = {}
@@ -278,7 +237,7 @@ def token_generator() -> Iterator[str]:
                         next_text[idx] = text
                     if r["eos"]:
                         next_text[idx] = ""
-                yield self.reformat_output(next_text, kwargs)
+                yield self.reformat_output(next_text, sampling_parameters)
             return
 
         return token_generator()
@@ -332,7 +291,7 @@ def exl2(
 
     """
     try:
-        from exllamav2 import (  # type: ignore
+        from exllamav2 import (
             ExLlamaV2,
             ExLlamaV2Cache,
             ExLlamaV2Cache_Q4,
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
deleted file mode 100644
index c66c984af..000000000
--- a/outlines/models/gemini.py
+++ /dev/null
@@ -1,296 +0,0 @@
-"""Integration with Gemini's API."""
-
-from typing import (
-    TYPE_CHECKING,
-    Any,
-    Iterator,
-    Optional,
-    Union,
-    get_args,
-)
-
-from outlines.models.base import Model, ModelTypeAdapter
-from outlines.templates import Vision
-from outlines.types import CFG, JsonSchema, Regex
-from outlines.types.utils import (
-    is_dataclass,
-    is_enum,
-    get_enum_from_literal,
-    is_genson_schema_builder,
-    is_literal,
-    is_pydantic_model,
-    is_typed_dict,
-    is_typing_list,
-)
-
-if TYPE_CHECKING:
-    from google.genai import Client
-
-__all__ = ["Gemini", "from_gemini"]
-
-
-class GeminiTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `Gemini` model.
-
-    `GeminiTypeAdapter` is responsible for preparing the arguments to Gemini's
-    client `models.generate_content` method: the input (prompt and possibly
-    image), as well as the output type (either JSON or multiple choice).
-
-    """
-
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
-        """Generate the `contents` argument to pass to the client.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        dict
-            The `contents` argument to pass to the client.
-
-        """
-        if isinstance(model_input, str):
-            return {"contents": [model_input]}
-        elif isinstance(model_input, Vision):
-            from google.genai import types
-
-            image_part = types.Part.from_bytes(
-                data=model_input.image_str,
-                mime_type=model_input.image_format
-            ),
-            return {"contents": [model_input.prompt, image_part]}
-        else:
-            raise TypeError(
-                f"The input type {input} is not available with Gemini. "
-                "The only available types are `str` and `Vision`."
-            )
-
-    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
-        """Generate the `generation_config` argument to pass to the client.
-
-        Parameters
-        ----------
-        output_type
-            The output type provided by the user.
-
-        Returns
-        -------
-        dict
-            The `generation_config` argument to pass to the client.
-
-        """
-
-        # Unsupported output pytes
-        if isinstance(output_type, Regex):
-            raise TypeError(
-                "Neither regex-based structured outputs nor the `pattern` "
-                "keyword in Json Schema are available with Gemini. Use an "
-                "open source model or dottxt instead."
-            )
-        elif isinstance(output_type, CFG):
-            raise TypeError(
-                "CFG-based structured outputs are not available with Gemini. "
-                "Use an open source model or dottxt instead."
-            )
-        elif is_genson_schema_builder(output_type):
-            raise TypeError(
-                "The Gemini SDK does not accept Genson schema builders as an "
-                "input. Pass a Pydantic model, typed dict or dataclass "
-                "instead."
-            )
-        elif isinstance(output_type, JsonSchema):
-            raise TypeError(
-                "The Gemini SDK does not accept Json Schemas as an input. "
-                "Pass a Pydantic model, typed dict or dataclass instead."
-            )
-
-        if output_type is None:
-            return {}
-
-        # Structured types
-        elif is_dataclass(output_type):
-            return self.format_json_output_type(output_type)
-        elif is_typed_dict(output_type):
-            return self.format_json_output_type(output_type)
-        elif is_pydantic_model(output_type):
-            return self.format_json_output_type(output_type)
-
-        # List of structured types
-        elif is_typing_list(output_type):
-            return self.format_list_output_type(output_type)
-
-        # Multiple choice types
-        elif is_enum(output_type):
-            return self.format_enum_output_type(output_type)
-        elif is_literal(output_type):
-            enum = get_enum_from_literal(output_type)
-            return self.format_enum_output_type(enum)
-
-        else:
-            type_name = getattr(output_type, "__name__", output_type)
-            raise TypeError(
-                f"The type `{type_name}` is not supported by Gemini. "
-                "Consider using a local model or dottxt instead."
-            )
-
-    def format_enum_output_type(self, output_type: Optional[Any]) -> dict:
-        return {
-            "response_mime_type": "text/x.enum",
-            "response_schema": output_type,
-        }
-
-    def format_json_output_type(self, output_type: Optional[Any]) -> dict:
-        return {
-            "response_mime_type": "application/json",
-            "response_schema": output_type,
-        }
-
-    def format_list_output_type(self, output_type: Optional[Any]) -> dict:
-        args = get_args(output_type)
-
-        if len(args) == 1:
-            item_type = args[0]
-
-            # Check if list item type is supported
-            if (
-                is_pydantic_model(item_type)
-                or is_typed_dict(item_type)
-                or is_dataclass(item_type)
-            ):
-                return {
-                    "response_mime_type": "application/json",
-                    "response_schema": output_type,
-                }
-
-            else:
-                raise TypeError(
-                    "The only supported types for list items are Pydantic "
-                    + "models, typed dicts and dataclasses."
-                )
-
-        raise TypeError(
-            f"Gemini only supports homogenous lists: "
-            "list[BaseModel], list[TypedDict] or list[dataclass]. "
-            f"Got {output_type} instead."
-        )
-
-
-class Gemini(Model):
-    """Thin wrapper around the `google.genai.Client` client.
-
-    This wrapper is used to convert the input and output types specified by
-    the users at a higher level to arguments to the `google.genai.Client`
-    client.
-
-    """
-
-    def __init__(self, client: "Client", model_name: Optional[str] = None):
-        """
-        Parameters
-        ----------
-        client
-            A `google.genai.Client` instance.
-        model_name
-            The name of the model to use.
-
-        """
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = GeminiTypeAdapter()
-
-    def generate(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs,
-    ) -> str:
-        """Generate a response from the model.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. The
-            output type must be of a type that can be converted to a JSON
-            schema, a list of such types, or a multiple choice type.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        str
-            The response generated by the model.
-
-        """
-        contents = self.type_adapter.format_input(model_input)
-        generation_config = self.type_adapter.format_output_type(output_type)
-
-        completion = self.client.models.generate_content(
-            **contents,
-            model=inference_kwargs.pop("model", self.model_name),
-            config={**generation_config, **inference_kwargs}
-        )
-
-        return completion.text
-
-    def generate_stream(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs,
-    ) -> Iterator[str]:
-        """Generate a stream of responses from the model.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. The
-            output type must be of a type that can be converted to a JSON
-            schema, a list of such types, or a multiple choice type.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
-
-        """
-        contents = self.type_adapter.format_input(model_input)
-        generation_config = self.type_adapter.format_output_type(output_type)
-
-        stream = self.client.models.generate_content_stream(
-            **contents,
-            model=inference_kwargs.pop("model", self.model_name),
-            config={**generation_config, **inference_kwargs},
-        )
-
-        for chunk in stream:
-            if hasattr(chunk, "text") and chunk.text:
-                yield chunk.text
-
-
-def from_gemini(client: "Client", model_name: Optional[str] = None) -> Gemini:
-    """Create an Outlines `Gemini` model instance from a
-    `google.genai.Client` instance.
-
-    Parameters
-    ----------
-    client
-        A `google.genai.Client` instance.
-    model_name
-        The name of the model to use.
-
-    Returns
-    -------
-    Gemini
-        An Outlines `Gemini` model instance.
-
-    """
-    return Gemini(client, model_name)
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 92d5fe15f..7c7ad64df 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,28 +1,26 @@
-"""Integration with the `llama-cpp-python` library."""
-
+import dataclasses
+import pickle
 import warnings
-from functools import singledispatchmethod
 from typing import (
     TYPE_CHECKING,
-    Any,
     Dict,
     Iterator,
     List,
     Optional,
     Set,
     Tuple,
+    TypedDict,
     Union,
 )
 
-from outlines.models.base import Model, ModelTypeAdapter
+from typing_extensions import Unpack
+
+from outlines.generate.api import GenerationParameters, SamplingParameters
 from outlines.models.tokenizer import Tokenizer
-from outlines.processors import CFGLogitsProcessor, OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     from llama_cpp import Llama, LogitsProcessorList
 
-__all__ = ["LlamaCpp", "from_llamacpp"]
-
 
 class LlamaCppTokenizer(Tokenizer):
     def __init__(self, model: "Llama"):
@@ -35,8 +33,7 @@ def __init__(self, model: "Llama"):
 
         self.tokenizer = model.tokenizer()
 
-        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613
-        # is resolved
+        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
         self._hf_tokenizer = None
         try:
             self.vocabulary = model.tokenizer_.hf_tokenizer.get_vocab()
@@ -50,8 +47,7 @@ def __init__(self, model: "Llama"):
         # ensure stable ordering of vocabulary
         self.vocabulary = {
             tok: tok_id
-            for tok, tok_id
-            in sorted(self.vocabulary.items(), key=lambda x: x[1])
+            for tok, tok_id in sorted(self.vocabulary.items(), key=lambda x: x[1])
         }
 
         self._hash = None
@@ -61,19 +57,14 @@ def decode(self, token_ids: List[int]) -> List[str]:
         return [decoded_bytes.decode("utf-8", errors="ignore")]
 
     def encode(
-        self,
-        prompt: Union[str, List[str]],
-        add_bos: bool = True,
-        special: bool = True,
+        self, prompt: Union[str, List[str]], add_bos: bool = True, special: bool = True
     ) -> Tuple[List[int], List[int]]:
         if isinstance(prompt, list):
             raise NotImplementedError(
                 "llama-cpp-python tokenizer doesn't support batch tokenization"
             )
         token_ids = self.tokenizer.tokenize(
-            prompt.encode("utf-8", errors="ignore"),
-            add_bos=add_bos,
-            special=special,
+            prompt.encode("utf-8", errors="ignore"), add_bos=add_bos, special=special
         )
         # generate attention mask, missing from llama-cpp-python
         attention_mask = [
@@ -86,10 +77,7 @@ def convert_token_to_string(self, token: str) -> str:
             from transformers.file_utils import SPIECE_UNDERLINE
 
             token_str = self._hf_tokenizer.convert_tokens_to_string([token])
-            if (
-                token.startswith(SPIECE_UNDERLINE)
-                or token == "<0x20>"
-            ):  # pragma: no cover
+            if token.startswith(SPIECE_UNDERLINE) or token == "<0x20>":
                 token_str = " " + token_str
             return token_str
         else:
@@ -101,15 +89,8 @@ def __eq__(self, other):
         return self.__getstate__() == other.__getstate__()
 
     def __hash__(self):
-        # We create a custom hash as pickle.dumps(self) is not stable
         if self._hash is None:
-            self._hash = hash((
-                tuple(sorted(self.vocabulary.items())),
-                self.eos_token_id,
-                self.eos_token,
-                self.pad_token_id,
-                tuple(sorted(self.special_tokens)),
-            ))
+            self._hash = hash(pickle.dumps(self))
         return self._hash
 
     def __getstate__(self):
@@ -126,160 +107,241 @@ def __setstate__(self, state):
         raise NotImplementedError("Cannot load a pickled llamacpp tokenizer")
 
 
-class LlamaCppTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `LlamaCpp` model.
-
-    `LlamaCppTypeAdapter` is responsible for preparing the arguments to
-    `llama-cpp-python`'s `Llama.__call__` method: the input (a string prompt),
-    as well as the logits processor (an instance of `LogitsProcessorList`).
+class LlamaCppParams(TypedDict, total=False):
+    suffix: Optional[str]
+    temperature: float
+    top_p: float
+    min_p: float
+    typical_p: float
+    seed: int
+    max_tokens: int
+    logits_processor: "LogitsProcessorList"
+    stop: Optional[Union[str, List[str]]]
+    frequence_penalty: float
+    presence_penalty: float
+    repeat_penalty: float
+    top_k: int
+    tfs_z: float
+    mirostat_mode: int
+    mirostat_tau: float
+    mirostat_eta: float
+    stream: bool
+
+
+class LlamaCpp:
+    """Represents a model provided by the `llama-cpp-python` library.
+
+    We wrap models from model providing libraries in order to give all of
+    them the same interface in Outlines and allow users to easily switch
+    between providers. This class wraps the `llama_cpp.Llama` class from the
+    `llama-cpp-python` library.
 
     """
 
-    @singledispatchmethod
-    def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        str
-            The formatted input to be passed to the model.
-
-        """
-        raise NotImplementedError(
-            f"The input type {input} is not available. "
-            "The `llama-cpp-python` library does not support batch inference. "
-        )
-
-    @format_input.register(str)
-    def format_str_input(self, model_input: str) -> str:
-        return model_input
+    def __init__(self, model: "Llama"):
+        self.model = model
 
-    def format_output_type(
-        self, output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> "LogitsProcessorList":
-        """Generate the logits processor argument to pass to the model.
+    @property
+    def tokenizer(self):
+        return LlamaCppTokenizer(self.model)
 
-        Parameters
-        ----------
-        output_type
-            The logits processor provided.
+    def prepare_generation_parameters(
+        self,
+        generation_parameters: GenerationParameters,
+        sampling_parameters: SamplingParameters,
+        structure_logits_processor,
+        **llama_cpp_params: Unpack[LlamaCppParams],
+    ):
+        """Prepare the generation parameters.
 
-        Returns
-        -------
-        LogitsProcessorList
-            The logits processor to pass to the model.
+        `llama-cpp-python` uses different default values
 
         """
         from llama_cpp import LogitsProcessorList
 
-        return LogitsProcessorList([output_type])
-
-
-class LlamaCpp(Model):
-    """Thin wrapper around the `llama_cpp.Llama` model.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `llama_cpp.Llama` model.
-    """
-
-    tensor_library_name = "numpy"
+        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
+
+        # We update `llama_cpp_params` with the values the user passed to the
+        # generator.
+        if "stop" not in llama_cpp_params:
+            llama_cpp_params["stop"] = stop_at
+        if "seed" not in llama_cpp_params:
+            llama_cpp_params["seed"] = seed
+
+        # Somehow `llama-cpp-python` generates `max_tokens + 1`  tokens
+        if "max_tokens" not in llama_cpp_params:
+            if max_tokens is None:
+                llama_cpp_params["max_tokens"] = -1  # indicates unlimited tokens
+            else:
+                llama_cpp_params["max_tokens"] = max_tokens - 1
+        else:
+            llama_cpp_params["max_tokens"] = llama_cpp_params["max_tokens"] - 1
 
-    def __init__(self, model: "Llama"):
-        """
-        Parameters
-        ----------
-        model
-            A `llama_cpp.Llama` model instance.
+        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
+            sampling_parameters
+        )
 
-        """
-        self.model = model
-        self.tokenizer = LlamaCppTokenizer(self.model)
-        self.type_adapter = LlamaCppTypeAdapter()
+        # We update the `llama_cpp_params` with the sampling values that
+        # were specified by the user via the `Sampler` class, unless they
+        # are also specified in `llama_cpp_params`. We also disable other
+        # sampling methods that are enabled by default and reset the temperature
+        # value.
+        #
+        # See https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
+        # for the default values in `llama.cpp` and indications to disable the sampling modes.
+        # Mirostat sampling, tail-free sampling and all penalties are disabled by default.
+        #
+        # See https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
+        # for default values in `llama-cpp-python`
+        if sampler == "beam_search":
+            raise NotImplementedError(
+                "The `llama_cpp_python` library does not support Beam Search."
+            )
+        if num_samples != 1:
+            raise NotImplementedError(
+                "The `llama_cpp_python` library does not allow to take several samples."
+            )
+        if "top_p" not in llama_cpp_params:
+            if top_p is not None:
+                llama_cpp_params["top_p"] = top_p
+            else:
+                llama_cpp_params["top_p"] = 1.0
+
+        if "min_p" not in llama_cpp_params:
+            llama_cpp_params["min_p"] = 0.0
+
+        if "top_k" not in llama_cpp_params:
+            if top_k is not None:
+                llama_cpp_params["top_k"] = top_k
+            else:
+                llama_cpp_params["top_k"] = -1
+
+        if "temperature" not in llama_cpp_params:
+            if temperature is not None:
+                llama_cpp_params["temperature"] = temperature
+            else:
+                llama_cpp_params["temperature"] = 1.0
+
+        if "repeat_penalty" not in llama_cpp_params:
+            llama_cpp_params["repeat_penalty"] = 1.0
+
+        # The choice to stream or not should happen via the high-level API
+        llama_cpp_params["stream"] = False
+
+        if structure_logits_processor is not None:
+            if "logits_processor" in llama_cpp_params:
+                llama_cpp_params["logits_processor"].append(structure_logits_processor)
+            else:
+                llama_cpp_params["logits_processor"] = LogitsProcessorList(
+                    [structure_logits_processor]
+                )
+
+        return llama_cpp_params
 
     def generate(
         self,
-        model_input: str,
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-        **inference_kwargs: Any,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        structure_logits_processor,
+        sampling_parameters: SamplingParameters,
+        **llama_cpp_params: Unpack[LlamaCppParams],
     ) -> str:
         """Generate text using `llama-cpp-python`.
 
         Parameters
         ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The logits processor the model will use to constrain the format of
-            the generated text.
-        **inference_kwargs
-            Additional keyword arguments to pass to the `Llama.__call__`
-            method of the `llama-cpp-python` library.
+        prompts
+            A prompt or list of prompts.
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
+        llama_cpp_params
+            Keyword arguments that can be passed to
+            `llama_cpp_python.Llama.__call__`.  The values in `llama_cpp_params`
+            supersede the values of the parameters in `generation_parameters` and
+            `sampling_parameters`.  See the `llama_cpp_python` documentation for
+            a list of possible values: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
 
         Returns
         -------
-        str
-            The text generated by the model.
+        The generated text.
 
         """
-        if isinstance(output_type, CFGLogitsProcessor):
+        if not isinstance(prompts, str):
             raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp due to bug in "
-                "the llama_cpp tokenizer"
+                "The `llama-cpp-python` library does not support batch inference."
             )
 
-        completion = self.model(
-            self.type_adapter.format_input(model_input),
-            logits_processor=self.type_adapter.format_output_type(output_type),
-            **inference_kwargs,
+        llama_cpp_params = self.prepare_generation_parameters(
+            generation_parameters,
+            sampling_parameters,
+            structure_logits_processor,
+            **llama_cpp_params,
         )
+        completion = self.model(prompts, **llama_cpp_params)
         result = completion["choices"][0]["text"]
 
         self.model.reset()
 
         return result
 
-    def generate_stream(
+    def stream(
         self,
-        model_input: str,
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-        **inference_kwargs: Any,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        structure_logits_processor,
+        sampling_parameters: SamplingParameters,
+        **llama_cpp_params: Unpack[LlamaCppParams],
     ) -> Iterator[str]:
         """Stream text using `llama-cpp-python`.
 
         Parameters
         ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The logits processor the model will use to constrain the format of
-            the generated text.
-        **inference_kwargs
-            Additional keyword arguments to pass to the `Llama.__call__`
-            method of the `llama-cpp-python` library.
+        prompts
+            A prompt or list of prompts.
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
+        llama_cpp_params
+            Keyword arguments that can be passed to
+            `llama_cpp_python.Llama.__call__`.  The values in `llama_cpp_params`
+            supersede the values of the parameters in `generation_parameters` and
+            `sampling_parameters`.  See the `llama_cpp_python` documentation for
+            a list of possible values: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
 
         Returns
         -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
+        A generator that return strings.
 
         """
-        if isinstance(output_type, CFGLogitsProcessor):
+
+        if not isinstance(prompts, str):
             raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp due to bug in "
-                "the llama_cpp tokenizer"
+                "The `llama-cpp-python` library does not support batch inference."
             )
 
-        generator = self.model(
-            self.type_adapter.format_input(model_input),
-            logits_processor=self.type_adapter.format_output_type(output_type),
-            stream=True,
-            **inference_kwargs,
+        llama_cpp_params = self.prepare_generation_parameters(
+            generation_parameters,
+            sampling_parameters,
+            structure_logits_processor,
+            **llama_cpp_params,
         )
+        llama_cpp_params["stream"] = True
+        generator = self.model(prompts, **llama_cpp_params)
 
         def token_generator() -> Iterator[str]:
             while True:
@@ -292,37 +354,54 @@ def token_generator() -> Iterator[str]:
 
         return token_generator()
 
-    def load_lora(self, adapter_path: str) -> None:  # pragma: no cover
-        """Load a LoRA adapter. Deprecated since v1.0.0."""
-        warnings.warn("""
-            The `load_lora` method is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.1.0.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
+    def load_lora(self, adapter_path: str):
         if self.model._model.apply_lora_from_file(
             adapter_path,
             1.0,
         ):
-            raise RuntimeError(
-                f"Failed to apply LoRA from lora path: {adapter_path}"
-            )
+            raise RuntimeError(f"Failed to apply LoRA from lora path: {adapter_path}")
 
 
-def from_llamacpp(model: "Llama"):
-    """Create an Outlines `LlamaCpp` model instance from a
-    `llama_cpp.Llama` instance.
+def llamacpp(
+    repo_id: str, filename: Optional[str] = None, **llamacpp_model_params
+) -> LlamaCpp:
+    """Load a model from the `llama-cpp-python` library.
+
+    We use the `Llama.from_pretrained` classmethod that downloads models
+    directly from the HuggingFace hub, instead of asking users to specify
+    a path to the downloaded model. One can still load a local model
+    by initializing `llama_cpp.Llama` directly.
 
     Parameters
     ----------
-    model
-        A `llama_cpp.Llama` instance.
-
-    Returns
-    -------
-    LlamaCpp
-        An Outlines `LlamaCpp` model instance.
+    repo_id
+        The name of the model repository.
+    filename:
+        A filename of glob pattern to match the model file in the repo.
+    llama_cpp_model_params
+        Llama-specific model parameters. See the `llama-cpp-python` documentation
+        for the full list: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__
 
     """
+    from llama_cpp import Llama
+
+    # Default to using the model's full context length
+    if "n_ctx" not in llamacpp_model_params:
+        llamacpp_model_params["n_ctx"] = 0
+
+    if "verbose" not in llamacpp_model_params:
+        llamacpp_model_params["verbose"] = False
+
+    # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
+    if "tokenizer" not in llamacpp_model_params:
+        warnings.warn(
+            "The pre-tokenizer in `llama.cpp` handles unicode improperly "
+            + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
+            + "Outlines may raise a `RuntimeError` when building the regex index.\n"
+            + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
+            + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
+        )
+
+    model = Llama.from_pretrained(repo_id, filename, **llamacpp_model_params)
+
     return LlamaCpp(model)
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index bea4ad293..843107d66 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -1,185 +1,247 @@
-"""Integration with the `mlx_lm` library."""
+import dataclasses
+from typing import TYPE_CHECKING, Generator, Iterator, List, Optional, Tuple, Union
 
-from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Iterator, List, Optional
-
-from outlines.models.base import Model, ModelTypeAdapter
-from outlines.models.transformers import TransformerTokenizer
-from outlines.processors import OutlinesLogitsProcessor
+from .transformers import TransformerTokenizer
 
 if TYPE_CHECKING:
+    import mlx.core as mx
     import mlx.nn as nn
     from transformers import PreTrainedTokenizer
 
-__all__ = ["MLXLM", "from_mlxlm"]
-
-
-class MLXLMTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `MLXLM` model."""
-
-    @singledispatchmethod
-    def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
+    from outlines.generate.api import GenerationParameters, SamplingParameters
+    from outlines.processors import OutlinesLogitsProcessor
 
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        str
-            The formatted input to be passed to the model.
-
-        """
-        raise NotImplementedError(
-            f"The input type {input} is not available. "
-            "The `mlx_lm` library does not support batch inference."
-        )
-
-    @format_input.register(str)
-    def format_str_input(self, model_input: str):
-        return model_input
-
-    def format_output_type(
-        self, output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> Optional[List[OutlinesLogitsProcessor]]:
-        """Generate the logits processor argument to pass to the model.
-
-        Parameters
-        ----------
-        output_type
-            The logits processor provided.
-
-        Returns
-        -------
-        Optional[list[OutlinesLogitsProcessor]]
-            The logits processor argument to be passed to the model.
-
-        """
-        if not output_type:
-            return None
-        return [output_type]
-
-
-class MLXLM(Model):
-    """Thin wrapper around an `mlx_lm` model.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `mlx_lm` library.
 
+class MLXLM:
+    """
+    Represents an `mlx_lm` model
     """
-
-    tensor_library_name = "mlx"
 
     def __init__(
         self,
         model: "nn.Module",
         tokenizer: "PreTrainedTokenizer",
     ):
-        """
-        Parameters
-        ----------
-        model
-            An instance of an `mlx_lm` model.
-        tokenizer
-            An instance of an `mlx_lm` tokenizer or of a compatible
-            `transformers` tokenizer.
-
-        """
         self.model = model
-        # self.mlx_tokenizer is used by the mlx-lm in its generate function
-        self.mlx_tokenizer = tokenizer
-        # self.tokenizer is used by the logits processor
-        self.tokenizer = TransformerTokenizer(tokenizer._tokenizer)
-        self.type_adapter = MLXLMTypeAdapter()
+        self.mlx_tokenizer = tokenizer  # returns mlx tensors, used for encode()
+        self.tokenizer = TransformerTokenizer(
+            tokenizer._tokenizer
+        )  # _tokenizer is HF Tokenizer
 
     def generate(
         self,
-        model_input: str,
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-        **kwargs,
+        prompts: Union[str, List[str]],
+        generation_parameters: "GenerationParameters",
+        logits_processor,
+        sampling_parameters: "SamplingParameters",
     ) -> str:
-        """Generate text using `mlx-lm`.
+        streamer = self.stream(
+            prompts, generation_parameters, logits_processor, sampling_parameters
+        )
+        return "".join(list(streamer))
+
+    def stream(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: "GenerationParameters",
+        logits_processor,
+        sampling_parameters: "SamplingParameters",
+    ) -> Iterator[str]:
+        """Generate text using `mlx_lm`.
 
         Parameters
         ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The logits processor the model will use to constrain the format of
-            the generated text.
-        kwargs
-            Additional keyword arguments to pass to the `mlx-lm` library.
+        prompts
+            A prompt or list of prompts.
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
 
         Returns
         -------
-        str
-            The text generated by the model.
-
+        The generated text.
         """
-        from mlx_lm import generate
-
-        return generate(
-            self.model,
-            self.mlx_tokenizer,
-            self.type_adapter.format_input(model_input),
-            logits_processors=self.type_adapter.format_output_type(output_type),
-            **kwargs,
+        import mlx.core as mx
+
+        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
+        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
+            sampling_parameters
         )
+        if max_tokens is None:
+            max_tokens = int(1e9)
+
+        if not isinstance(prompts, str):
+            raise NotImplementedError(
+                "The `mlx-lm` library does not support batch inference."
+            )
+        if sampler == "beam_search":
+            raise NotImplementedError(
+                "The `mlx-lm` library does not support Beam Search."
+            )
+        if num_samples != 1:
+            raise NotImplementedError(
+                "The `mlx-lm` library does not allow to take several samples."
+            )
+        if top_k is not None:
+            raise NotImplementedError("The `mlx-lm` library does not support top_k.")
+        if seed is not None:
+            raise NotImplementedError("The `mlx-lm` library does not support seed.")
+        if stop_at is not None:
+            raise NotImplementedError("The `mlx-lm` library does not support stop_at.")
+
+        generate_kwargs = {
+            "temp": temperature,
+            "top_p": top_p,
+            "sampler": sampler,
+            "logits_processor": logits_processor,
+        }
+
+        # Adapted from
+        # https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L267
+        prompt_tokens = mx.array(self.mlx_tokenizer.encode(prompts))
+
+        detokenizer = self.mlx_tokenizer.detokenizer
+        detokenizer.reset()
+
+        for (token, prob), n in zip(
+            self.generate_step(prompt_tokens, **generate_kwargs),
+            range(max_tokens),
+        ):
+            if token == self.tokenizer.eos_token_id:
+                break
+            detokenizer.add_token(token)
+            yield detokenizer.last_segment
 
-    def generate_stream(
+        detokenizer.finalize()
+        yield detokenizer.last_segment
+
+    def generate_step(
         self,
-        model_input: str,
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-        **kwargs,
-    ) -> Iterator[str]:
-        """Stream text using `mlx-lm`.
+        prompt: "mx.array",
+        temp: Optional[float],
+        top_p: Optional[float],
+        sampler: str,
+        logits_processor: "OutlinesLogitsProcessor",
+    ) -> Generator[Tuple[int, float], None, None]:
+        """
+        Adapted from
+        https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L129
+
+        A generator producing token ids based on the given prompt from the model.
 
         Parameters
         ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The logits processor the model will use to constrain the format of
-            the generated text.
-        kwargs
-            Additional keyword arguments to pass to the `mlx-lm` library.
+        prompt
+            The input prompt.
+        temp
+            The temperature for sampling, if 0 the argmax is used.
+        top_p
+            Nulceus sampling, higher means model considers more less likely words.
+        sampler
+            The sampler string defined by SequenceGeneratorAdapter
+        logits_processor
+            Augment logits before sampling.
+        """
+        import mlx.core as mx
+        import mlx_lm
 
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
+        temperature: float = temp or 1.0
 
-        """
-        from mlx_lm import stream_generate
-
-        for gen_response in stream_generate(
-            self.model,
-            self.mlx_tokenizer,
-            self.type_adapter.format_input(model_input),
-            logits_processors=self.type_adapter.format_output_type(output_type),
-            **kwargs,
-        ):
-            yield gen_response.text
+        def sample(logits: "mx.array") -> Tuple["mx.array", float]:
+            softmax_logits = mx.softmax(logits)
+
+            if temperature == 0.0 or sampler == "greedy":
+                token = mx.argmax(logits, axis=-1)
+            elif sampler == "multinomial":
+                if top_p is not None and top_p > 0 and top_p < 1.0:
+                    token = mlx_lm.sample_utils.top_p_sampling(
+                        logits, top_p, temperature
+                    )
+                else:
+                    token = mx.random.categorical(logits * (1 / temperature))
+            else:
+                raise ValueError(f"Invalid mlx-lm sampler: `{sampler}`")
+
+            prob = softmax_logits[0, token]
+            return token, prob
+
+        cache = mlx_lm.models.cache.make_prompt_cache(self.model)
 
+        # kv cache contains processed input IDs, we pass the unprocessed inputs and cache to model()
+        unprocessed_input_ids = prompt
+        generated_ids: List[int] = []
 
-def from_mlxlm(model: "nn.Module", tokenizer: "PreTrainedTokenizer") -> MLXLM:
-    """Create an Outlines `MLXLM` model instance from an `mlx_lm` model and a
-    tokenizer.
+        while True:
+            logits = self.model(unprocessed_input_ids[None], cache=cache)
+            logits = logits[:, -1, :]
+
+            if logits_processor is not None:
+                # convert to logits_processor 1d expectation, apply, then convert back
+                logits_1d = logits.reshape(-1)
+                logits_1d = logits_processor(generated_ids, logits_1d)
+                logits = logits_1d.reshape(1, -1)
+
+            new_token_single, prob = sample(logits)
+            new_token = new_token_single.item()
+            yield new_token, prob
+
+            generated_ids.append(new_token)
+            unprocessed_input_ids = new_token_single
+
+
+def mlxlm(
+    model_name: str,
+    tokenizer_config: dict = {},
+    model_config: dict = {},
+    adapter_path: Optional[str] = None,
+    lazy: bool = False,
+):
+    """Instantiate a model from the `mlx_lm` library and its tokenizer.
+
+    Signature adapted from
+    https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L422
 
     Parameters
     ----------
-    model
-        An instance of an `mlx_lm` model.
-    tokenizer
-        An instance of an `mlx_lm` tokenizer or of a compatible
-        transformers tokenizer.
+    Args:
+        path_or_hf_repo (Path): The path or the huggingface repository to load the model from.
+        tokenizer_config (dict, optional): Configuration parameters specifically for the tokenizer.
+            Defaults to an empty dictionary.
+        model_config(dict, optional): Configuration parameters specifically for the model.
+            Defaults to an empty dictionary.
+        adapter_path (str, optional): Path to the LoRA adapters. If provided, applies LoRA layers
+            to the model. Default: ``None``.
+        lazy (bool): If False eval the model parameters to make sure they are
+            loaded in memory before returning, otherwise they will be loaded
+            when needed. Default: ``False``
 
     Returns
     -------
-    MLXLM
-        An Outlines `MLXLM` model instance.
+    A `MLXLM` model instance.
 
     """
+    try:
+        import mlx.core as mx
+        import mlx_lm
+    except ImportError:
+        raise ImportError(
+            "The `mlx_lm` library needs to be installed in order to use `mlx_lm` models."
+        )
+    if not mx.metal.is_available():
+        raise RuntimeError("You cannot use `mlx_lm` without Apple Silicon (Metal)")
+
+    model, tokenizer = mlx_lm.load(
+        model_name,
+        tokenizer_config=tokenizer_config,
+        model_config=model_config,
+        adapter_path=adapter_path,
+        lazy=lazy,
+    )
     return MLXLM(model, tokenizer)
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
deleted file mode 100644
index 862c71f69..000000000
--- a/outlines/models/ollama.py
+++ /dev/null
@@ -1,213 +0,0 @@
-"""Integration with the `ollama` library."""
-
-import json
-from typing import TYPE_CHECKING, Any, Iterator, Optional
-
-from pydantic import TypeAdapter
-
-from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import CFG, JsonSchema, Regex
-from outlines.types.utils import (
-    is_dataclass,
-    is_genson_schema_builder,
-    is_pydantic_model,
-    is_typed_dict,
-)
-
-if TYPE_CHECKING:
-    from ollama import Client as OllamaClient
-
-__all__ = ["Ollama", "from_ollama"]
-
-
-class OllamaTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `Ollama` model."""
-
-    def format_input(self, model_input: str) -> str:
-        """Generate the prompt argument to pass to the model.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        str
-            The formatted input to be passed to the model.
-
-        """
-        if isinstance(model_input, str):
-            return model_input
-        raise TypeError(
-            f"The input type {model_input} is not available. "
-            "Ollama does not support batch inference."
-        )
-
-    def format_output_type(self, output_type: Optional[Any] = None) -> Optional[str]:
-        """Format the output type to pass to the client.
-
-        TODO: `int`, `float` and other Python types could be supported via
-        JSON Schema.
-
-        Parameters
-        ----------
-        output_type
-            The output type provided by the user.
-
-        Returns
-        -------
-        Optional[str]
-            The formatted output type to be passed to the model.
-
-        """
-        if isinstance(output_type, Regex):
-            raise TypeError(
-                "Regex-based structured outputs are not supported by Ollama. "
-                "Use an open source model in the meantime."
-            )
-        elif isinstance(output_type, CFG):
-            raise TypeError(
-                "CFG-based structured outputs are not supported by Ollama. "
-                "Use an open source model in the meantime."
-            )
-
-        if output_type is None:
-            return None
-        elif isinstance(output_type, JsonSchema):
-            return json.loads(output_type.schema)
-        elif is_dataclass(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return schema
-        elif is_typed_dict(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return schema
-        elif is_pydantic_model(output_type):
-            schema = output_type.model_json_schema()
-            return schema
-        elif is_genson_schema_builder(output_type):
-            return output_type.to_json()
-        else:
-            type_name = getattr(output_type, "__name__", output_type)
-            raise TypeError(
-                f"The type `{type_name}` is not supported by Ollama. "
-                "Consider using a local model instead."
-            )
-
-
-class Ollama(Model):
-    """Thin wrapper around the `ollama.Client` client.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `ollama.Client` client.
-
-    """
-
-    def __init__(
-        self,client: "OllamaClient", model_name: Optional[str] = None,
-    ):
-        """
-        Parameters
-        ----------
-        client
-            The `ollama.Client` client.
-        model_name
-            The name of the model to use.
-
-        """
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = OllamaTypeAdapter()
-
-    def generate(self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **kwargs: Any,
-    ) -> str:
-        """Generate text using Ollama.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. The
-            output type must be of a type that can be converted to a JSON
-            schema.
-        **kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        str
-            The text generated by the model.
-
-        """
-        if "model" not in kwargs and self.model_name is not None:
-            kwargs["model"] = self.model_name
-
-        response = self.client.generate(
-            prompt=self.type_adapter.format_input(model_input),
-            format=self.type_adapter.format_output_type(output_type),
-            **kwargs,
-        )
-        return response.response
-
-    def generate_stream(
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **kwargs: Any,
-    ) -> Iterator[str]:
-        """Stream text using Ollama.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. The
-            output type must be of a type that can be converted to a JSON
-            schema.
-        **kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
-
-        """
-        if "model" not in kwargs and self.model_name is not None:
-            kwargs["model"] = self.model_name
-
-        response = self.client.generate(
-            prompt=self.type_adapter.format_input(model_input),
-            format=self.type_adapter.format_output_type(output_type),
-            stream=True,
-            **kwargs,
-        )
-        for chunk in response:
-            yield chunk.response
-
-
-def from_ollama(
-    client: "OllamaClient", model_name: Optional[str] = None
-) -> Ollama:
-    """Create an Outlines `Ollama` model instance from an `ollama.Client`
-    client.
-
-    Parameters
-    ----------
-    client
-        A `ollama.Client` client instance.
-    model_name
-        The name of the model to use.
-
-    Returns
-    -------
-    Ollama
-        An Outlines `Ollama` model instance.
-
-    """
-    return Ollama(client, model_name)
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 2112e1974..e46107bd4 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -2,419 +2,15 @@
 
 import copy
 import functools
-import json
-import warnings
-from typing import (
-    TYPE_CHECKING,
-    Any,
-    Callable,
-    Dict,
-    Iterator,
-    List,
-    Optional,
-    Union,
-)
 from dataclasses import asdict, dataclass, field, replace
+from typing import Callable, Dict, List, Optional, Tuple, Union
 
-from pydantic import BaseModel, TypeAdapter
+import numpy as np
 
+from outlines.base import vectorize
 from outlines.caching import cache
-from outlines.models.base import Model, ModelTypeAdapter
-from outlines.templates import Vision
-from outlines.types import JsonSchema, Regex, CFG
-from outlines.types.utils import (
-    is_dataclass,
-    is_typed_dict,
-    is_pydantic_model,
-    is_genson_schema_builder,
-    is_native_dict
-)
-from outlines.v0_legacy.base import vectorize
 
-if TYPE_CHECKING:
-    from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
-
-__all__ = ["OpenAI", "from_openai"]
-
-
-class OpenAITypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `OpenAI` model.
-
-    `OpenAITypeAdapter` is responsible for preparing the arguments to OpenAI's
-    `completions.create` methods: the input (prompt and possibly image), as
-    well as the output type (only JSON).
-
-    """
-
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
-        """Generate the `messages` argument to pass to the client.
-
-        Parameters
-        ----------
-        model_input
-            The input provided by the user.
-
-        Returns
-        -------
-        dict
-            The formatted input to be passed to the client.
-
-        """
-        if isinstance(model_input, str):
-            return self.format_str_model_input(model_input)
-        elif isinstance(model_input, Vision):
-            return self.format_vision_model_input(model_input)
-        raise TypeError(
-            f"The input type {input} is not available with OpenAI. "
-            "The only available types are `str` and `Vision`."
-        )
-
-    def format_str_model_input(self, model_input: str) -> dict:
-        """Generate the `messages` argument to pass to the client when the user
-        only passes a prompt.
-
-        """
-        return {
-            "messages": [
-                {
-                    "role": "user",
-                    "content": model_input,
-                }
-            ]
-        }
-
-    def format_vision_model_input(self, model_input: Vision) -> dict:
-        """Generate the `messages` argument to pass to the client when the user
-        passes a prompt and an image.
-
-        """
-        return {
-            "messages": [
-                {
-                    "role": "user",
-                    "content": [
-                        {"type": "text", "text": model_input.prompt},
-                        {
-                            "type": "image_url",
-                            "image_url": {
-                                "url": f"data:{model_input.image_format};base64,{model_input.image_str}"  # noqa: E702
-                            },
-                        },
-                    ],
-                }
-            ]
-        }
-
-    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
-        """Generate the `response_format` argument to the client based on the
-        output type specified by the user.
-
-        TODO: `int`, `float` and other Python types could be supported via
-        JSON Schema.
-
-        Parameters
-        ----------
-        output_type
-            The output type provided by the user.
-
-        Returns
-        -------
-        dict
-            The formatted output type to be passed to the client.
-
-        """
-        # Unsupported languages
-        if isinstance(output_type, Regex):
-            raise TypeError(
-                "Neither regex-based structured outputs nor the `pattern` keyword "
-                "in Json Schema are available with OpenAI. Use an open source "
-                "model or dottxt instead."
-            )
-        elif isinstance(output_type, CFG):
-            raise TypeError(
-                "CFG-based structured outputs are not available with OpenAI. "
-                "Use an open source model or dottxt instead."
-            )
-
-        if output_type is None:
-            return {}
-        elif is_native_dict(output_type):
-            return self.format_json_mode_type()
-        elif is_dataclass(output_type):
-            output_type = TypeAdapter(output_type).json_schema()
-            return self.format_json_output_type(output_type)
-        elif is_typed_dict(output_type):
-            output_type = TypeAdapter(output_type).json_schema()
-            return self.format_json_output_type(output_type)
-        elif is_pydantic_model(output_type):
-            output_type = output_type.model_json_schema()
-            return self.format_json_output_type(output_type)
-        elif is_genson_schema_builder(output_type):
-            schema = json.loads(output_type.to_json())
-            return self.format_json_output_type(schema)
-        elif isinstance(output_type, JsonSchema):
-            return self.format_json_output_type(json.loads(output_type.schema))
-        else:
-            type_name = getattr(output_type, "__name__", output_type)
-            raise TypeError(
-                f"The type `{type_name}` is not available with OpenAI. "
-                "Use an open source model or dottxt instead."
-            )
-
-    def format_json_output_type(self, schema: dict) -> dict:
-        """Generate the `response_format` argument to the client when the user
-        specified a `Json` output type.
-
-        """
-        # OpenAI requires `additionalProperties` to be set
-        if "additionalProperties" not in schema:
-            schema["additionalProperties"] = False
-
-        return {
-            "response_format": {
-                "type": "json_schema",
-                "json_schema": {
-                    "name": "default",
-                    "strict": True,
-                    "schema": schema,
-                },
-            }
-        }
-
-    def format_json_mode_type(self) -> dict:
-        """Generate the `response_format` argument to the client when the user
-        specified the output type should be a JSON but without specifying the
-        schema (also called "JSON mode").
-
-        """
-        return {"response_format": {"type": "json_object"}}
-
-
-class OpenAI(Model):
-    """Thin wrapper around the `openai.OpenAI` client.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `openai.OpenAI` client.
-
-    """
-
-    def __init__(
-        self,
-        client: Union["OpenAIClient", "AzureOpenAIClient"],
-        model_name: Optional[Union[str, "OpenAIConfig"]] = None,
-        **kwargs
-    ):
-        """Initialize the OpenAI model.
-
-        To provide temporary backwards compatibility with Outlines v0,
-        the class can be instantiated with a `OpenAIConfig` instance as
-        a value for the `model_name` argument. This is deprecated and will
-        be removed in v1.1.0. Please provide a model name instead.
-
-        Parameters
-        ----------
-        client
-            The `openai.OpenAI` client.
-        model_name
-            The name of the model to use.
-
-        """
-
-        # legacy mode
-        if isinstance(model_name, OpenAIConfig) or kwargs.get("config"):
-            warnings.warn("""
-                The `openai` function is deprecated starting from v1.0.0.
-                Do not use it. Support for it will be removed in v1.1.0.
-                Instead, you should instantiate a `OpenAI` model with the
-                `outlines.from_openai` function that takes an openai library
-                client and a model name as arguments. Similarly, you cannot
-                instantiate a `OpenAI` model directly with a `OpenAIConfig`
-                instance anymore, but must provide a client and a model name
-                instead.
-                For example:
-                ```python
-                from openai import OpenAI as OpenAIClient
-                from outlines import from_openai
-                client = OpenAIClient()
-                model = from_openai(client, "gpt-4o")
-                ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-            )
-            config = (
-                model_name
-                if isinstance(model_name, OpenAIConfig)
-                else kwargs.pop("config")
-            )
-            self.legacy_instance = OpenAILegacy(
-                client, config, kwargs.get("system_prompt")
-            )
-        # regular mode
-        else:
-            self.client = client
-            self.model_name = model_name
-            self.type_adapter = OpenAITypeAdapter()
-
-    def generate(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Union[type[BaseModel], str]] = None,
-        **inference_kwargs: Any,
-    ) -> Union[str, list[str]]:
-        """Generate text using OpenAI.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. The
-            output type must be of a type that can be converted to a JSON
-            schema or an empty dictionary.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Union[str, list[str]]
-            The text generated by the model.
-
-        """
-        import openai
-
-        messages = self.type_adapter.format_input(model_input)
-        response_format = self.type_adapter.format_output_type(output_type)
-
-        if "model" not in inference_kwargs and self.model_name is not None:
-            inference_kwargs["model"] = self.model_name
-
-        try:
-            result = self.client.chat.completions.create(
-                **messages,
-                **response_format,
-                **inference_kwargs,
-            )
-        except openai.BadRequestError as e:
-            if e.body["message"].startswith("Invalid schema"):
-                raise TypeError(
-                    f"OpenAI does not support your schema: {e.body['message']}. "
-                    "Try a local model or dottxt instead."
-                )
-            else:
-                raise e
-
-        messages = [choice.message for choice in result.choices]
-        for message in messages:
-            if message.refusal is not None:
-                raise ValueError(
-                    f"OpenAI refused to answer the request: {message.refusal}"
-                )
-
-        if len(messages) == 1:
-            return messages[0].content
-        else:
-            return [message.content for message in messages]
-
-    def generate_stream(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Union[type[BaseModel], str]] = None,
-        **inference_kwargs,
-    ) -> Iterator[str]:
-        """Stream text using OpenAI.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. The
-            output type must be of a type that can be converted to a JSON
-            schema or an empty dictionary.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
-
-        """
-        import openai
-
-        messages = self.type_adapter.format_input(model_input)
-        response_format = self.type_adapter.format_output_type(output_type)
-
-        if "model" not in inference_kwargs and self.model_name is not None:
-            inference_kwargs["model"] = self.model_name
-
-        stream = self.client.chat.completions.create(
-            stream=True,
-            **messages,
-            **response_format,
-            **inference_kwargs
-        )
-
-        for chunk in stream:
-            if chunk.choices and chunk.choices[0].delta.content is not None:
-                yield chunk.choices[0].delta.content
-
-    ### Legacy !!!
-
-    def __call__(self, *args, **kwargs):
-        if hasattr(self, "legacy_instance"):
-            return self.legacy_instance(*args, **kwargs)
-        else:
-            return super().__call__(*args, **kwargs)
-
-    def stream(self, *args, **kwargs):
-        if hasattr(self, "legacy_instance"):
-            return self.legacy_instance.stream(*args, **kwargs)
-        else:
-            return super().stream(*args, **kwargs)
-
-    def new_with_replacements(self, **kwargs):
-        if hasattr(self, "legacy_instance"):
-            return self.legacy_instance.new_with_replacements(self, **kwargs)
-        raise NotImplementedError("This method is only available in legacy mode")
-
-    def __str__(self):
-        if hasattr(self, "legacy_instance"):
-            return str(self.legacy_instance)
-        else:
-            return super().__str__()
-
-    def __repr__(self):
-        if hasattr(self, "legacy_instance"):
-            return repr(self.legacy_instance)
-        else:
-            return super().__repr__()
-
-def from_openai(
-    client: Union["OpenAIClient", "AzureOpenAIClient"],
-    model_name: Optional[str] = None,
-) -> OpenAI:
-    """Create an Outlines `OpenAI` model instance from an `openai.OpenAI`
-    client.
-
-    Parameters
-    ----------
-    client
-        An `openai.OpenAI` client instance.
-    model_name
-        The name of the model to use.
-
-    Returns
-    -------
-    OpenAI
-        An Outlines `OpenAI` model instance.
-
-    """
-    return OpenAI(client, model_name)
-
-
-### Legacy !!!
+__all__ = ["OpenAI", "openai", "azure_openai"]
 
 
 @dataclass(frozen=True)
@@ -470,7 +66,7 @@ class OpenAIConfig:
     user: str = field(default_factory=str)
 
 
-class OpenAILegacy():
+class OpenAI:
     """An object that represents the OpenAI API."""
 
     def __init__(
@@ -495,7 +91,6 @@ def __init__(
 
         self.client = client
         self.config = config
-        self.system_prompt = system_prompt
 
         # We count the total number of prompt and generated tokens as returned
         # by the OpenAI API, summed over all the requests performed with this
@@ -503,7 +98,7 @@ def __init__(
         self.prompt_tokens = 0
         self.completion_tokens = 0
 
-        self.format_sequence = lambda x: x
+        self.format_sequence = lambda seq: seq
 
     def __call__(
         self,
@@ -514,7 +109,7 @@ def __call__(
         system_prompt: Optional[str] = None,
         temperature: Optional[float] = None,
         samples: Optional[int] = None,
-    ):
+    ) -> np.ndarray:
         """Call the OpenAI API to generate text.
 
         Parameters
@@ -554,7 +149,7 @@ def __call__(
         )  # type: ignore
 
         response, prompt_tokens, completion_tokens = generate_chat(
-            prompt, system_prompt or self.system_prompt, self.client, config
+            prompt, system_prompt, self.client, config
         )
         self.prompt_tokens += prompt_tokens
         self.completion_tokens += completion_tokens
@@ -566,11 +161,9 @@ def stream(self, *args, **kwargs):
             "Streaming is currently not supported for the OpenAI API"
         )
 
-    def new_with_replacements(self, model, **kwargs):
-        new_instance = copy.copy(model)
-        new_instance.legacy_instance.config = replace(
-            new_instance.legacy_instance.config, **kwargs
-        )
+    def new_with_replacements(self, **kwargs):
+        new_instance = copy.copy(self)
+        new_instance.config = replace(new_instance.config, **kwargs)
         return new_instance
 
     def __str__(self):
@@ -586,7 +179,7 @@ async def generate_chat(
     system_prompt: Union[str, None],
     client,
     config: OpenAIConfig,
-):
+) -> Tuple[np.ndarray, int, int]:
     """Call OpenAI's Chat Completion API.
 
     Parameters
@@ -607,7 +200,6 @@ async def generate_chat(
     A tuple that contains the model's response(s) and usage statistics.
 
     """
-    import numpy as np
 
     @error_handler
     @cache()
@@ -658,3 +250,54 @@ def call(*args, **kwargs):
             raise e
 
     return call
+
+
+@functools.singledispatch
+def openai(model_or_client, *args, **kwargs):
+    return OpenAI(model_or_client, *args, **kwargs)
+
+
+@openai.register(str)
+def openai_model(
+    model_name: str,
+    config: Optional[OpenAIConfig] = None,
+    **openai_client_params,
+):
+    try:
+        from openai import AsyncOpenAI
+    except ImportError:
+        raise ImportError(
+            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
+        )
+
+    if config is not None:
+        config = replace(config, model=model_name)  # type: ignore
+    else:
+        config = OpenAIConfig(model=model_name)
+
+    client = AsyncOpenAI(**openai_client_params)
+
+    return OpenAI(client, config)
+
+
+def azure_openai(
+    deployment_name: str,
+    model_name: Optional[str] = None,
+    config: Optional[OpenAIConfig] = None,
+    **azure_openai_client_params,
+):
+    try:
+        from openai import AsyncAzureOpenAI
+    except ImportError:
+        raise ImportError(
+            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
+        )
+
+    if config is not None:
+        config = replace(config, model=deployment_name)  # type: ignore
+    if config is None:
+        config = OpenAIConfig(model=deployment_name)
+
+    client = AsyncAzureOpenAI(**azure_openai_client_params)
+
+    return OpenAI(client, config)
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
deleted file mode 100644
index 0c2a51da1..000000000
--- a/outlines/models/sglang.py
+++ /dev/null
@@ -1,372 +0,0 @@
-"""Integration with an SGLang server."""
-
-import json
-import warnings
-from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
-
-from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
-from outlines.models.openai import OpenAITypeAdapter
-from outlines.types.dsl import (
-    CFG,
-    JsonSchema,
-    python_types_to_terms,
-    to_regex,
-)
-from outlines.templates import Vision
-
-if TYPE_CHECKING:
-    from openai import AsyncOpenAI, OpenAI
-
-__all__ = ["AsyncSGLang", "SGLang", "from_sglang"]
-
-
-class SGLangTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `SGLang` and `AsyncSGLang` models."""
-
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
-        """Generate the prompt argument to pass to the client.
-
-        We rely on the OpenAITypeAdapter to format the input as the sglang
-        server expects input in the same format as OpenAI.
-
-        Parameters
-        ----------
-        model_input
-            The input passed by the user.
-
-        Returns
-        -------
-        dict
-            The formatted input to be passed to the client.
-
-        """
-        return OpenAITypeAdapter().format_input(model_input)
-
-    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
-        """Generate the structured output argument to pass to the client.
-
-        Parameters
-        ----------
-        output_type
-            The structured output type provided.
-
-        Returns
-        -------
-        dict
-            The formatted output type to be passed to the client.
-
-        """
-        if output_type is None:
-            return {}
-
-        term = python_types_to_terms(output_type)
-        if isinstance(term, CFG):
-            warnings.warn(
-                "SGLang grammar-based structured outputs expects an EBNF "
-                "grammar instead of a Lark grammar as is generally used in "
-                "Outlines. The grammar cannot be used as a structured output "
-                "type with an outlines backend, it is only compatible with "
-                "the sglang and llguidance backends."
-            )
-            return {"extra_body": {"ebnf": term.definition}}
-        elif isinstance(term, JsonSchema):
-            return OpenAITypeAdapter().format_json_output_type(
-                json.loads(term.schema)
-            )
-        else:
-            return {"extra_body": {"regex": to_regex(term)}}
-
-
-class SGLang(Model):
-    """Thin wrapper around the `openai.OpenAI` client used to communicate with
-    an SGLang server.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `openai.OpenAI` client for the
-    SGLang server.
-
-    """
-
-    def __init__(self, client, model_name: Optional[str] = None):
-        """
-        Parameters
-        ----------
-        client
-            An `openai.OpenAI` client instance.
-        model_name
-            The name of the model to use.
-
-        """
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = SGLangTypeAdapter()
-
-    def generate(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Union[str, list[str]]:
-        """Generate text using SGLang.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Union[str, list[str]]
-            The text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input,
-            output_type,
-            **inference_kwargs,
-        )
-
-        response = self.client.chat.completions.create(**client_args)
-
-        messages = [choice.message for choice in response.choices]
-        for message in messages:
-            if message.refusal is not None:  # pragma: no cover
-                raise ValueError(
-                    f"The SGLang server refused to answer the request: "
-                    f"{message.refusal}"
-                )
-
-        if len(messages) == 1:
-            return messages[0].content
-        else:
-            return [message.content for message in messages]
-
-    def generate_stream(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Iterator[str]:
-        """Stream text using SGLang.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
-        )
-
-        stream = self.client.chat.completions.create(
-            **client_args, stream=True,
-        )
-
-        for chunk in stream:  # pragma: no cover
-            if chunk.choices and chunk.choices[0].delta.content is not None:
-                yield chunk.choices[0].delta.content
-
-    def _build_client_args(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> dict:
-        """Build the arguments to pass to the SGLang client."""
-        messages = self.type_adapter.format_input(model_input)
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        inference_kwargs.update(output_type_args)
-
-        if "model" not in inference_kwargs and self.model_name is not None:
-            inference_kwargs["model"] = self.model_name
-
-        client_args = {
-            **messages,
-            **inference_kwargs,
-        }
-
-        return client_args
-
-
-class AsyncSGLang(AsyncModel):
-    """Thin async wrapper around the `openai.OpenAI` client used to communicate
-    with an SGLang server.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `openai.OpenAI` client for the
-    SGLang server.
-
-    """
-
-    def __init__(self, client, model_name: Optional[str] = None):
-        """
-        Parameters
-        ----------
-        client
-            An `openai.AsyncOpenAI` client instance.
-        model_name
-            The name of the model to use.
-
-        Parameters
-        ----------
-        client
-            An `openai.AsyncOpenAI` client instance.
-
-        """
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = SGLangTypeAdapter()
-
-    async def generate(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Union[str, list[str]]:
-        """Generate text using `sglang`.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Union[str, list[str]]
-            The text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
-        )
-
-        response = await self.client.chat.completions.create(**client_args)
-
-        messages = [choice.message for choice in response.choices]
-        for message in messages:
-            if message.refusal is not None:  # pragma: no cover
-                raise ValueError(
-                    f"The sglang server refused to answer the request: "
-                    f"{message.refusal}"
-                )
-
-        if len(messages) == 1:
-            return messages[0].content
-        else:
-            return [message.content for message in messages]
-
-    async def generate_stream( # type: ignore
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> AsyncIterator[str]:
-        """Return a text generator.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        AsyncIterator[str]
-            An async iterator that yields the text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
-        )
-
-        stream = await self.client.chat.completions.create(
-            **client_args,
-            stream=True,
-        )
-
-        async for chunk in stream:  # pragma: no cover
-            if chunk.choices and chunk.choices[0].delta.content is not None:
-                yield chunk.choices[0].delta.content
-
-    def _build_client_args(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> dict:
-        """Build the arguments to pass to the SGLang client."""
-        messages = self.type_adapter.format_input(model_input)
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        inference_kwargs.update(output_type_args)
-
-        if "model" not in inference_kwargs and self.model_name is not None:
-            inference_kwargs["model"] = self.model_name
-
-        client_args = {
-            **messages,
-            **inference_kwargs,
-        }
-
-        return client_args
-
-
-def from_sglang(
-    client: Union["OpenAI", "AsyncOpenAI"],
-    model_name: Optional[str] = None,
-) -> Union[SGLang, AsyncSGLang]:
-    """Create a `SGLang` or `AsyncSGLang` instance from an `openai.OpenAI` or
-    `openai.AsyncOpenAI` instance.
-
-    Parameters
-    ----------
-    client
-        An `openai.OpenAI` or `openai.AsyncOpenAI` instance.
-    model_name
-        The name of the model to use.
-
-    Returns
-    -------
-    Union[SGLang, AsyncSGLang]
-        An Outlines `SGLang` or `AsyncSGLang` model instance.
-
-    """
-    from openai import AsyncOpenAI, OpenAI
-
-    if isinstance(client, OpenAI):
-        return SGLang(client, model_name)
-    elif isinstance(client, AsyncOpenAI):
-        return AsyncSGLang(client, model_name)
-    else:
-        raise ValueError(
-            f"Unsupported client type: {type(client)}.\n"
-            "Please provide an OpenAI or AsyncOpenAI instance."
-        )
diff --git a/outlines/models/tgi.py b/outlines/models/tgi.py
deleted file mode 100644
index d9f3ffd75..000000000
--- a/outlines/models/tgi.py
+++ /dev/null
@@ -1,340 +0,0 @@
-"""Integration with a TGI server."""
-
-import json
-from functools import singledispatchmethod
-from typing import (
-    TYPE_CHECKING,
-    Any,
-    AsyncIterator,
-    Iterator,
-    Optional,
-    Union,
-)
-
-from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
-from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
-
-if TYPE_CHECKING:
-    from huggingface_hub import AsyncInferenceClient, InferenceClient
-
-__all__ = ["AsyncTGI", "TGI", "from_tgi"]
-
-
-class TGITypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `TGI` and `AsyncTGI` models."""
-
-    @singledispatchmethod
-    def format_input(self, model_input):
-        """Generate the prompt argument to pass to the client.
-
-        Argument
-        --------
-        model_input
-            The input passed by the user.
-
-        Returns
-        -------
-        str
-            The formatted input to be passed to the model.
-
-        """
-        raise NotImplementedError(
-            f"The input type {input} is not available with TGI. "
-            + "Please provide a string."
-        )
-
-    @format_input.register(str)
-    def format_str_input(self, model_input: str) -> str:
-        return model_input
-
-    @format_input.register(list)
-    def format_list_input(self, model_input: list):
-        raise NotImplementedError("TGI does not support batch inference.")
-
-    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
-        """Generate the structured output argument to pass to the client.
-
-        Argument
-        --------
-        output_type
-            The structured output type provided.
-
-        Returns
-        -------
-        dict
-            The structured output argument to pass to the client.
-
-        """
-        if output_type is None:
-            return {}
-
-        term = python_types_to_terms(output_type)
-        if isinstance(term, CFG):
-            raise NotImplementedError(
-                "TGI does not support CFG-based structured outputs."
-            )
-        elif isinstance(term, JsonSchema):
-            return {
-                "grammar": {
-                    "type": "json",
-                    "value": json.loads(term.schema),
-                }
-            }
-        else:
-            return {
-                "grammar": {
-                    "type": "regex",
-                    "value": to_regex(term),
-                }
-            }
-
-
-class TGI(Model):
-    """Thin wrapper around a `huggingface_hub.InferenceClient` client used to
-    communicate with a `TGI` server.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the
-    `huggingface_hub.InferenceClient` client.
-
-    """
-
-    def __init__(self, client):
-        """
-        Parameters
-        ----------
-        client
-            A huggingface `InferenceClient` client instance.
-
-        """
-        self.client = client
-        self.type_adapter = TGITypeAdapter()
-
-    def generate(
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> str:
-        """Generate text using TGI.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types except `CFG` are supported provided your server uses
-            a backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        str
-            The text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input,
-            output_type,
-            **inference_kwargs,
-        )
-
-        return self.client.text_generation(**client_args)
-
-    def generate_stream(
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Iterator[str]:
-        """Stream text using TGI.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types except `CFG` are supported provided your server uses
-            a backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
-        )
-
-        stream = self.client.text_generation(
-            **client_args, stream=True,
-        )
-
-        for chunk in stream:  # pragma: no cover
-            yield chunk
-
-    def _build_client_args(
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> dict:
-        """Build the arguments to pass to the TGI client."""
-        prompt = self.type_adapter.format_input(model_input)
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        inference_kwargs.update(output_type_args)
-
-        client_args = {
-            "prompt": prompt,
-            **inference_kwargs,
-        }
-
-        return client_args
-
-
-class AsyncTGI(AsyncModel):
-    """Thin async wrapper around a `huggingface_hub.AsyncInferenceClient`
-    client used to communicate with a `TGI` server.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the
-    `huggingface_hub.AsyncInferenceClient` client.
-
-    """
-
-    def __init__(self, client):
-        """
-        Parameters
-        ----------
-        client
-            A huggingface `AsyncInferenceClient` client instance.
-
-        """
-        self.client = client
-        self.type_adapter = TGITypeAdapter()
-
-    async def generate(
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> str:
-        """Generate text using TGI.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types except `CFG` are supported provided your server uses
-            a backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        str
-            The text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
-        )
-
-        response = await self.client.text_generation(**client_args)
-
-        return response
-
-    async def generate_stream( # type: ignore
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> AsyncIterator[str]:
-        """Stream text using TGI.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types except `CFG` are supported provided your server uses
-            a backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        AsyncIterator[str]
-            An async iterator that yields the text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
-        )
-
-        stream = await self.client.text_generation(
-            **client_args, stream=True
-        )
-
-        async for chunk in stream:  # pragma: no cover
-            yield chunk
-
-    def _build_client_args(
-        self,
-        model_input: str,
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> dict:
-        """Build the arguments to pass to the TGI client."""
-        prompt = self.type_adapter.format_input(model_input)
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        inference_kwargs.update(output_type_args)
-
-        client_args = {
-            "prompt": prompt,
-            **inference_kwargs,
-        }
-
-        return client_args
-
-
-def from_tgi(
-    client: Union["InferenceClient", "AsyncInferenceClient"],
-) -> Union[TGI, AsyncTGI]:
-    """Create an Outlines `TGI` or `AsyncTGI` model instance from an
-    `huggingface_hub.InferenceClient` or `huggingface_hub.AsyncInferenceClient`
-    instance.
-
-    Parameters
-    ----------
-    client
-        An `huggingface_hub.InferenceClient` or
-        `huggingface_hub.AsyncInferenceClient` instance.
-
-    Returns
-    -------
-    Union[TGI, AsyncTGI]
-        An Outlines `TGI` or `AsyncTGI` model instance.
-
-    """
-    from huggingface_hub import AsyncInferenceClient, InferenceClient
-
-    if isinstance(client, InferenceClient):
-        return TGI(client)
-    elif isinstance(client, AsyncInferenceClient):
-        return AsyncTGI(client)
-    else:
-        raise ValueError(
-            f"Unsupported client type: {type(client)}.\n"
-            + "Please provide an HuggingFace InferenceClient "
-            + "or AsyncInferenceClient instance."
-        )
diff --git a/outlines/models/tokenizer.py b/outlines/models/tokenizer.py
index 5463c5eaa..1a5708d85 100644
--- a/outlines/models/tokenizer.py
+++ b/outlines/models/tokenizer.py
@@ -1,9 +1,7 @@
-from typing import Dict, Hashable, List, Protocol, Set, Tuple, Union, TYPE_CHECKING
+from typing import Dict, Hashable, List, Protocol, Set, Tuple, Union
 
-
-if TYPE_CHECKING:
-    import numpy as np
-    from numpy.typing import NDArray
+import numpy as np
+from numpy.typing import NDArray
 
 
 class Tokenizer(Hashable, Protocol):
@@ -15,11 +13,11 @@ class Tokenizer(Hashable, Protocol):
 
     def encode(
         self, prompt: Union[str, List[str]]
-    ) -> "Tuple['NDArray[np.int64]', 'NDArray[np.int64]']":
+    ) -> Tuple[NDArray[np.int64], NDArray[np.int64]]:
         """Translate the input prompts into arrays of token ids and attention mask."""
         ...
 
-    def decode(self, token_ids: "NDArray[np.int64]") -> List[str]:
+    def decode(self, token_ids: NDArray[np.int64]) -> List[str]:
         """Translate an array of token ids to a string or list of strings."""
         ...
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 01c7b34a5..6b204ec6e 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,22 +1,20 @@
-"""Integration with the `transformers` library."""
+import dataclasses
+import inspect
+from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, Union
 
-from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union
-
-from outlines.models.base import Model, ModelTypeAdapter
+from outlines.generate.api import GenerationParameters, SamplingParameters
 from outlines.models.tokenizer import Tokenizer
-from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     import torch
-    from transformers import (
-        PreTrainedTokenizer,
-        PreTrainedModel,
-        ProcessorMixin,
-        LogitsProcessorList,
-    )
+    from transformers import PreTrainedModel, PreTrainedTokenizer
+
+    from outlines.processors import OutlinesLogitsProcessor
+
+__all__ = ["transformers"]
 
-__all__ = ["Transformers", "TransformersMultiModal", "from_transformers"]
+
+KVCacheType = Tuple[Tuple["torch.DoubleTensor", "torch.DoubleTensor"], ...]
 
 
 def get_llama_tokenizer_types():
@@ -27,28 +25,28 @@ def get_llama_tokenizer_types():
     """
     try:
         from transformers.models.llama import LlamaTokenizer
-    except ImportError:  # pragma: no cover
+    except ImportError:
 
         class LlamaTokenizer:  # type: ignore
             pass
 
     try:
         from transformers.models.llama import LlamaTokenizerFast
-    except ImportError:  # pragma: no cover
+    except ImportError:
 
         class LlamaTokenizerFast:  # type: ignore
             pass
 
     try:
         from transformers.models.code_llama import CodeLlamaTokenizer
-    except ImportError:  # pragma: no cover
+    except ImportError:
 
         class CodeLlamaTokenizer:  # type: ignore
             pass
 
     try:
         from transformers.models.code_llama import CodeLlamaTokenizerFast
-    except ImportError:  # pragma: no cover
+    except ImportError:
 
         class CodeLlamaTokenizerFast:  # type: ignore
             pass
@@ -128,190 +126,229 @@ def __setstate__(self, state):
         self.__init__(state["tokenizer"])
 
 
-class TransformersTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `Transformers` model."""
+class Transformers:
+    """Represents a `transformers` model."""
+
+    def __init__(
+        self,
+        model: "PreTrainedModel",
+        tokenizer: "PreTrainedTokenizer",
+    ):
+        self.model = model
+        self.tokenizer = TransformerTokenizer(tokenizer)
 
-    @singledispatchmethod
-    def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
+    def forward(
+        self,
+        input_ids: "torch.LongTensor",
+        attention_mask: "torch.LongTensor",
+        past_key_values: Optional[Tuple] = None,
+    ) -> Tuple["torch.FloatTensor", Optional[KVCacheType]]:
+        """Compute a forward pass through the transformer model.
 
         Parameters
         ----------
-        model_input
-            The input passed by the user.
+        input_ids
+            The input token ids.  Must be one or two dimensional.
+        attention_mask
+            The attention mask.  Must be one or two dimensional.
+        past_key_values
+            A tuple of tuples containing the cached key and value tensors for each
+            attention head.
 
         Returns
         -------
-        str
-            The formatted input to be passed to the model.
+        The computed logits and the new cached key and value tensors.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available."
-            "Please use a string or a list of strings."
-        )
+        try:
+            import torch
+        except ImportError:
+            ImportError(
+                "The `torch` library needs to be installed to use `transformers` models."
+            )
+        assert 0 < input_ids.ndim < 3
+
+        if past_key_values:
+            input_ids = input_ids[..., -1].unsqueeze(-1)
+
+        with torch.inference_mode():
+            output = self.model(
+                input_ids,
+                attention_mask=attention_mask,
+                return_dict=True,
+                output_attentions=False,
+                output_hidden_states=False,
+                past_key_values=past_key_values,
+            )
 
-    @format_input.register(str)
-    def format_str_input(self, model_input: str) -> str:
-        return model_input
+        return output.logits, output.past_key_values
+
+    def __call__(
+        self,
+        input_ids: "torch.LongTensor",
+        attention_mask: "torch.LongTensor",
+        past_key_values: Optional[Tuple] = None,
+    ) -> "torch.FloatTensor":
+        logits, kv_cache = self.forward(input_ids, attention_mask, past_key_values)
+        next_token_logits = logits[..., -1, :]
 
-    @format_input.register(list)
-    def format_list_input(self, model_input: List[str]) -> List[str]:
-        return model_input
+        return next_token_logits, kv_cache
 
-    def format_output_type(
+    def generate(
         self,
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> Optional["LogitsProcessorList"]:
-        """Generate the logits processor argument to pass to the model.
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Union[str, List[str], List[List[str]]]:
+        """Generate text using `transformers`.
 
         Parameters
         ----------
-        output_type
-            The logits processor provided.
+        prompts
+            A prompt or list of prompts.
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
 
         Returns
         -------
-        Optional[LogitsProcessorList]
-            The logits processor to pass to the model.
-
+        The generated text
         """
-        from transformers import LogitsProcessorList
-
-        if output_type is not None:
-            return LogitsProcessorList([output_type])
-        return None
+        if isinstance(prompts, str):
+            # convert to 2d
+            input_ids, attention_mask = self.tokenizer.encode([prompts])
+        else:
+            input_ids, attention_mask = self.tokenizer.encode(prompts)
 
+        inputs = {
+            "input_ids": input_ids.to(self.model.device),
+            "attention_mask": attention_mask.to(self.model.device),
+        }
+        if (
+            "attention_mask"
+            not in inspect.signature(self.model.forward).parameters.keys()
+        ):
+            del inputs["attention_mask"]
 
-class Transformers(Model):
-    """Thin wrapper around a `transformers` model and a `transformers`
-    tokenizer.
+        generation_kwargs = self._get_generation_kwargs(
+            prompts,
+            generation_parameters,
+            logits_processor,
+            sampling_parameters,
+        )
+        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
 
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `transformers` model and
-    tokenizer.
+        # if single str input and single sample per input, convert to a 1D output
+        if isinstance(prompts, str):
+            generated_ids = generated_ids.squeeze(0)
 
-    """
+        return self._decode_generation(generated_ids)
 
-    def __init__(
+    def stream(
         self,
-        model: "PreTrainedModel",
-        tokenizer: "PreTrainedTokenizer",
-    ):
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Iterator[Union[str, List[str]]]:
         """
-        Parameters:
-        ----------
-        model
-            A `PreTrainedModel`, or any model that is compatible with the
-            `transformers` API for models.
-        tokenizer
-            A `PreTrainedTokenizer`, or any tokenizer that is compatible with
-            the `transformers` API for tokenizers.
+        Temporary stream stand-in which implements stream() signature
+        and equivalent behaviour but isn't yielded until generation completes.
 
+        TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
         """
-        # We need to handle the cases in which jax/flax or tensorflow
-        # is not available in the environment.
-        try:
-            from transformers import FlaxPreTrainedModel
-        except ImportError:  # pragma: no cover
-            FlaxPreTrainedModel = None
-
-        try:
-            from transformers import TFPreTrainedModel
-        except ImportError:  # pragma: no cover
-            TFPreTrainedModel = None
-
-        tokenizer.padding_side = "left"
-        self.model = model
-        self.transformer_tokenizer = tokenizer
-        self.tokenizer = TransformerTokenizer(tokenizer)
-        self.type_adapter = TransformersTypeAdapter()
-
-        if (
-            FlaxPreTrainedModel is not None
-            and isinstance(model, FlaxPreTrainedModel)
-        ):
-            self.tensor_library_name = "jax"
-        elif (
-            TFPreTrainedModel is not None
-            and isinstance(model, TFPreTrainedModel)
-        ):
-            self.tensor_library_name = "tensorflow"
+        if isinstance(prompts, str):
+            # convert to 2d
+            input_ids, attention_mask = self.tokenizer.encode([prompts])
         else:
-            self.tensor_library_name = "torch"
-
-    def _prepare_model_inputs(
-        self,
-        model_input: Union[str, List[str], dict],
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> Tuple[Union[str, List[str]], dict]:
-        """Turn the user input into arguements to pass to the model"""
-        prompts = self.type_adapter.format_input(model_input)
-        input_ids, attention_mask = self.tokenizer.encode(prompts)
+            input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
             "input_ids": input_ids.to(self.model.device),
             "attention_mask": attention_mask.to(self.model.device),
         }
+        if (
+            "attention_mask"
+            not in inspect.signature(self.model.forward).parameters.keys()
+        ):
+            del inputs["attention_mask"]
 
-        return prompts, inputs
-
-    def generate(
-        self,
-        model_input: Union[str, List[str], dict],
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-        **inference_kwargs: Any,
-    ) -> Union[str, List[str]]:
-        """Generate text using `transformers`.
+        generation_kwargs = self._get_generation_kwargs(
+            prompts,
+            generation_parameters,
+            logits_processor,
+            sampling_parameters,
+        )
+        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
 
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response. For
-            multi-modal models, the input should be a dictionary containing the
-            `text` key with a value of type `Union[str, List[str]]` and the
-            other keys required by the model.
-        output_type
-            The logits processor the model will use to constrain the format of
-            the generated text.
-        inference_kwargs
-            Additional keyword arguments to pass to the `generate` method
-            of the `transformers` model.
+        # if single str input and single sample per input, convert to a 1D output
+        if isinstance(prompts, str):
+            generated_ids = generated_ids.squeeze(0)
 
-        Returns
-        -------
-        Union[str, List[str]]
-            The text generated by the model.
+        for i in range(generated_ids.size(-1)):
+            output_group_ids = generated_ids.select(-1, i).unsqueeze(-1)
+            yield self._decode_generation(output_group_ids)
 
+    def _get_generation_kwargs(
+        self,
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> dict:
         """
-        prompts, inputs = self._prepare_model_inputs(model_input, output_type)
-        logits_processor = self.type_adapter.format_output_type(output_type)
+        Convert outlines generation parameters into model.generate kwargs
+        """
+        from transformers import GenerationConfig, LogitsProcessorList, set_seed
 
-        generated_ids = self._generate_output_seq(
-            prompts, inputs, logits_processor=logits_processor, **inference_kwargs
+        max_new_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
+        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
+            sampling_parameters
         )
+        if max_new_tokens is None:
+            max_new_tokens = int(2**30)
 
-        # if single str input, convert to a 1D outputt
-        if isinstance(prompts, str):
-            generated_ids = generated_ids.squeeze(0)
-
-        return self._decode_generation(generated_ids)
-
-    def generate_stream(self, model_input, output_type, **inference_kwargs):
-        """Not available for `transformers` models.
+        # global seed, not desirable
+        if seed is not None:
+            set_seed(seed)
 
-        TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
+        if logits_processor is not None:
+            logits_processor_list = LogitsProcessorList([logits_processor])
+        else:
+            logits_processor_list = None
+
+        generation_config = GenerationConfig(
+            max_new_tokens=max_new_tokens,
+            stop_strings=stop_at,
+            num_return_sequences=(num_samples or 1),
+            top_p=top_p,
+            top_k=top_k,
+            temperature=temperature,
+            do_sample=(sampler == "multinomial"),
+            num_beams=(num_samples if sampler == "beam_search" else 1),
+            eos_token_id=self.tokenizer.eos_token_id,
+            pad_token_id=self.tokenizer.pad_token_id,
+        )
 
-        """
-        raise NotImplementedError(
-            "Streaming is not implemented for Transformers models."
+        return dict(
+            logits_processor=logits_processor_list,
+            generation_config=generation_config,
+            tokenizer=self.tokenizer.tokenizer,
         )
 
-    def _generate_output_seq(self, prompts, inputs, **inference_kwargs):
+    def _generate_output_seq(
+        self, prompts, inputs, generation_config, **generation_kwargs
+    ):
         input_ids = inputs["input_ids"]
         output_ids = self.model.generate(
-            **inputs,
-            tokenizer=self.transformer_tokenizer,
-            **inference_kwargs,
+            **inputs, generation_config=generation_config, **generation_kwargs
         )
 
         # encoder-decoder returns output_ids only, decoder-only returns full seq ids
@@ -321,10 +358,12 @@ def _generate_output_seq(self, prompts, inputs, **inference_kwargs):
             generated_ids = output_ids[:, input_ids.shape[1] :]
 
         # if batch list inputs AND multiple samples per input, convert generated_id to 3D view
-        num_samples = inference_kwargs.get("num_return_sequences", 1)
+        num_samples = generation_config.num_return_sequences or 1
+
         if num_samples > 1 and isinstance(prompts, list):
             batch_size = input_ids.size(0)
-            generated_ids = generated_ids.view(batch_size, num_samples, -1)
+            num_return_sequences = generation_config.num_return_sequences or 1
+            generated_ids = generated_ids.view(batch_size, num_return_sequences, -1)
 
         return generated_ids
 
@@ -338,164 +377,82 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
                 self.tokenizer.decode(generated_ids[i])
                 for i in range(len(generated_ids))
             ]
-        else:  # pragma: no cover
+        else:
             raise TypeError(
                 f"Generated outputs aren't 1D, 2D or 3D, but instead are {generated_ids.shape}"
             )
 
 
-class TransformersMultiModalTypeAdapter(ModelTypeAdapter):
-    """Type adapter for `TransformersMultiModal` model."""
-
-    @singledispatchmethod
-    def format_input(self, model_input):
-        """Generate the prompt arguments to pass to the model.
-
-        Argument
-        --------
-        model_input
-            The input passed by the user.
-
-        Returns
-        -------
-        dict
-            The formatted input to be passed to the model.
-
-        """
-        raise NotImplementedError(
-            f"The input type {input} is not available. Please provide a "
-            + "dictionary containing at least the 'text' key with a value "
-            + "of type Union[str, List[str]]. You should also include the "
-            + "other keys required by your processor (for instance, 'images' "
-            + "or 'audios')."
-            + "Make sure that the text is correctly formatted for the model "
-            + "(e.g. include <image> or <|AUDIO|> tags) and that the number "
-            + "of text tags match the number of additional assets provided."
-        )
-
-    @format_input.register(dict)
-    def format_list_input(self, model_input: dict) -> dict:
-        if "text" not in model_input:
-            raise ValueError(
-                "The input must contain the 'text' key along with the other "
-                + "keys required by your processor."
-            )
-        return model_input
-
-    def format_output_type(
-        self,
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> Optional["LogitsProcessorList"]:
-        """Generate the logits processor argument to pass to the model.
-
-        Argument
-        --------
-        output_type
-            The logits processor provided.
+def transformers(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    tokenizer_kwargs: dict = {},
+    model_class=None,
+    tokenizer_class=None,
+):
+    """Instantiate a model from the `transformers` library and its tokenizer.
 
-        Returns
-        -------
-        Optional[LogitsProcessorList]
-            The logits processor to pass to the model.
-
-        """
-        from transformers import LogitsProcessorList
-
-        if output_type is not None:
-            return LogitsProcessorList([output_type])
-        return None
-
-
-class TransformersMultiModal(Transformers):
-    """Thin wrapper around a `transformers` model and a `transformers`
-    processor.
+    Parameters
+    ----------
+    model_name
+        The name of the model as listed on Hugging Face's model page.
+    device
+        The device(s) on which the model should be loaded. This overrides
+        the `device_map` entry in `model_kwargs` when provided.
+    model_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the model.
+    tokenizer_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the tokenizer.
 
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `transformers` model and
-    processor.
+    Returns
+    -------
+    A `TransformersModel` model instance.
 
     """
+    if model_class is None or tokenizer_class is None:
+        try:
+            from transformers import AutoModelForCausalLM, AutoTokenizer
+        except ImportError:
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if model_class is None:
+        model_class = AutoModelForCausalLM
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
 
-    def __init__(self, model: "PreTrainedModel", processor):
-        """Create a TransformersMultiModal model instance
-
-        We rely on the `__init__` method of the `Transformers` class to handle
-        most of the initialization and then add elements specific to vision
-        models.
-
-        Parameters
-        ----------
-        model
-            A `PreTrainedModel`, or any model that is compatible with the
-            `transformers` API for models.
-        processor
-            A `ProcessorMixin` instance.
-
-        """
-        self.processor = processor
-        self.processor.padding_side = "left"
-        self.processor.pad_token = "[PAD]"
-
-        tokenizer: "PreTrainedTokenizer" = self.processor.tokenizer
+    if device is not None:
+        model_kwargs["device_map"] = device
 
-        super().__init__(model, tokenizer)
+    model = model_class.from_pretrained(model_name, **model_kwargs)
 
-        self.type_adapter = TransformersMultiModalTypeAdapter()
+    tokenizer_kwargs.setdefault("padding_side", "left")
+    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
 
-    def _prepare_model_inputs(
-        self,
-        model_input: Union[str, List[str], dict],
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> Tuple[Union[str, List[str]], dict]:
-        """Turn the user input into arguements to pass to the model"""
-        model_input = self.type_adapter.format_input(model_input)
-        inputs = self.processor(
-            **model_input, padding=True, return_tensors="pt"
-        ).to(self.model.device)
-
-        return model_input["text"], inputs
-
-
-def from_transformers(
-    model: "PreTrainedModel",
-    tokenizer_or_processor: Union["PreTrainedTokenizer", "ProcessorMixin"],
-) -> Union[Transformers, TransformersMultiModal]:
-    """Create an Outlines `Transformers` or `TransformersMultiModal` model
-    instance from a `PreTrainedModel` instance and a `PreTrainedTokenizer` or
-    `ProcessorMixin` instance.
-
-    `outlines` supports `PreTrainedModelForCausalLM`,
-    `PreTrainedMambaForCausalLM`, `PreTrainedModelForSeq2Seq` and any model
-    that implements the `transformers` model API.
+    return Transformers(model, tokenizer)
 
-    Parameters
-    ----------
-    model
-        A `transformers.PreTrainedModel` instance.
-    tokenizer_or_processor
-        A `transformers.PreTrainedTokenizer` or
-        `transformers.ProcessorMixin` instance.
 
-    Returns
-    -------
-    Union[Transformers, TransformersMultiModal]
-        An Outlines `Transformers` or `TransformersMultiModal` model instance.
-
-    """
-    from transformers import (
-        PreTrainedTokenizer, PreTrainedTokenizerFast, ProcessorMixin)
+def mamba(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    tokenizer_kwargs: dict = {},
+):
+    try:
+        from transformers import MambaForCausalLM
 
-    if isinstance(
-        tokenizer_or_processor, (PreTrainedTokenizer, PreTrainedTokenizerFast)
-    ):
-        tokenizer = tokenizer_or_processor
-        return Transformers(model, tokenizer)
-    elif isinstance(tokenizer_or_processor, ProcessorMixin):
-        processor = tokenizer_or_processor
-        return TransformersMultiModal(model, processor)
-    else:
-        raise ValueError(
-            "We could determine whether the model passed to `from_transformers`"
-            + " is a text-2-text or a multi-modal model. Please provide a "
-            + "a transformers tokenizer or processor."
+    except ImportError:
+        raise ImportError(
+            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
         )
+
+    return transformers(
+        model_name=model_name,
+        device=device,
+        model_kwargs=model_kwargs,
+        tokenizer_kwargs=tokenizer_kwargs,
+        model_class=MambaForCausalLM,
+    )
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
new file mode 100644
index 000000000..c8a86536e
--- /dev/null
+++ b/outlines/models/transformers_vision.py
@@ -0,0 +1,138 @@
+from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
+
+from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.models import Transformers
+
+if TYPE_CHECKING:
+    from outlines.processors import OutlinesLogitsProcessor
+
+
+class TransformersVision(Transformers):
+    def __init__(self, model, tokenizer, processor):
+        super().__init__(model, tokenizer)
+        self.processor = processor
+
+    def generate(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[List[Any], List[List[Any]]],
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Union[str, List[str], List[List[str]]]:
+        """Generate text using `transformers`.
+
+        Parameters
+        ----------
+        prompts
+            A prompt or list of prompts.
+        media
+            A List[PIL.Image] or List[List[PIL.Image]]
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
+
+        Returns
+        -------
+        The generated text
+        """
+        inputs = self.processor(
+            text=prompts, images=media, padding=True, return_tensors="pt"
+        ).to(self.model.device)
+
+        generation_kwargs = self._get_generation_kwargs(
+            prompts,
+            generation_parameters,
+            logits_processor,
+            sampling_parameters,
+        )
+        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
+
+        # if single str input and single sample per input, convert to a 1D output
+        if isinstance(prompts, str):
+            # Should always be true until NotImplementedError above is fixed
+            generated_ids = generated_ids.squeeze(0)
+
+        return self._decode_generation(generated_ids)
+
+    def stream(  # type: ignore
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[Any, List[Any]],  # TODO: docstring
+        generation_parameters: GenerationParameters,
+        logits_processor: Optional["OutlinesLogitsProcessor"],
+        sampling_parameters: SamplingParameters,
+    ) -> Iterator[Union[str, List[str]]]:
+        raise NotImplementedError
+
+
+def transformers_vision(
+    model_name: str,
+    model_class,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    processor_kwargs: dict = {},
+    tokenizer_class=None,
+    processor_class=None,
+):
+    """Instantiate a model from the `transformers` library and its tokenizer.
+
+    Parameters
+    ----------
+    model_name
+        The name of the model as listed on Hugging Face's model page.
+    model_class
+        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
+        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
+    device
+        The device(s) on which the model should be loaded. This overrides
+        the `device_map` entry in `model_kwargs` when provided.
+    model_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the model.
+    processor_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the processor.
+
+    Returns
+    -------
+    A `TransformersModel` model instance.
+
+    """
+    if processor_class is None or tokenizer_class is None:
+        try:
+            from transformers import AutoProcessor, AutoTokenizer
+        except ImportError:
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if processor_class is None:
+        processor_class = AutoProcessor
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
+
+    if device is not None:
+        model_kwargs["device_map"] = device
+
+    model = model_class.from_pretrained(model_name, **model_kwargs)
+
+    processor_kwargs.setdefault("padding_side", "left")
+    processor_kwargs.setdefault("pad_token", "[PAD]")
+    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+    if tokenizer_class is None:
+        if getattr(processor, "tokenizer", None):
+            tokenizer = processor.tokenizer
+        else:
+            tokenizer = AutoTokenizer.from_pretrained(model_name, **processor_kwargs)
+    else:
+        tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
+
+    return TransformersVision(model, tokenizer, processor)
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index f787e7072..b9b035d1f 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,362 +1,227 @@
-"""Integration with a vLLM server."""
+import dataclasses
+from typing import TYPE_CHECKING, List, Optional, Union
 
-import json
-from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
-
-from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
-from outlines.models.openai import OpenAITypeAdapter
-from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
-from outlines.templates import Vision
+from outlines.generate.api import GenerationParameters, SamplingParameters
 
 if TYPE_CHECKING:
-    from openai import AsyncOpenAI, OpenAI
-
-__all__ = ["VLLM", "AsyncVLLM", "from_vllm"]
-
+    from transformers import PreTrainedTokenizerBase
+    from vllm import LLM
+    from vllm.sampling_params import SamplingParams
 
-class VLLMTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `VLLM` and `AsyncVLLM` models."""
 
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
-        """Generate the prompt argument to pass to the client.
+class VLLM:
+    """Represents a vLLM model.
 
-        We rely on the OpenAITypeAdapter to format the input as the vLLM server
-        expects input in the same format as OpenAI.
-
-        Parameters
-        ----------
-        model_input
-            The input passed by the user.
-
-        Returns
-        -------
-        dict
-            The formatted input to be passed to the model.
-
-        """
-        return OpenAITypeAdapter().format_input(model_input)
+    We wrap models from model providing libraries in order to give all of
+    them the same interface in Outlines and allow users to easily switch
+    between providers. This class wraps the `vllm.LLM` class from the
+    `vllm` library.
 
-    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
-        """Generate the structured output argument to pass to the client.
+    """
 
-        Parameters
-        ----------
-        output_type
-            The structured output type provided.
+    def __init__(self, model: "LLM"):
+        self.model = model
+        self.lora_request = None
 
-        Returns
-        -------
-        dict
-            The structured output argument to pass to the model.
+        self.tokenizer = self._get_tokenizer()
 
-        """
-        if output_type is None:
-            return {}
-
-        term = python_types_to_terms(output_type)
-        if isinstance(term, CFG):
-            return {"guided_grammar": term.definition}
-        elif isinstance(term, JsonSchema):
-            extra_body = {"guided_json": json.loads(term.schema)}
-            if term.whitespace_pattern:
-                extra_body["whitespace_pattern"] = term.whitespace_pattern
-            return extra_body
+    def _get_tokenizer(self):
+        if hasattr(self.model, "get_tokenizer"):
+            tokenizer = self.model.get_tokenizer()
+        elif hasattr(self.model, "tokenizer"):
+            if hasattr(self.model.tokenizer, "tokenizer"):
+                tokenizer = self.model.tokenizer.tokenizer
+            else:
+                tokenizer = self.model.tokenizer
         else:
-            return {"guided_regex": to_regex(term)}
-
-
-class VLLM(Model):
-    """Thin wrapper around the `openai.OpenAI` client used to communicate with
-    a `vllm` server.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `openai.OpenAI` client for the
-    `vllm` server.
-    """
-
-    def __init__(
-        self,
-        client: "OpenAI",
-        model_name: Optional[str] = None,
-    ):
-        """
-        Parameters
-        ----------
-        client
-            An `openai.OpenAI` client instance.
-
-        """
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = VLLMTypeAdapter()
+            raise ValueError(
+                "The provided LLM instance neither has a "
+                "`tokenizer` attribute or a `get_tokenizer` method."
+            )
+        return adapt_tokenizer(tokenizer=tokenizer)
 
     def generate(
         self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Union[str, list[str]]:
+        prompts: Union[str, List[str]],
+        generation_parameters: GenerationParameters,
+        logits_processor,
+        sampling_parameters: SamplingParameters,
+        *,
+        sampling_params: Optional["SamplingParams"] = None,
+        use_tqdm: bool = True,
+    ):
         """Generate text using vLLM.
 
         Parameters
         ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
+        prompts
+            A prompt or list of prompts.
+        generation_parameters
+            An instance of `GenerationParameters` that contains the prompt,
+            the maximum number of tokens, stop sequences and seed. All the
+            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        logits_processor
+            The logits processor to use when generating text.
+        sampling_parameters
+            An instance of `SamplingParameters`, a dataclass that contains
+            the name of the sampler to use and related parameters as available
+            in Outlines.
+        sampling_params
+            An instance of `vllm.sampling_params.SamplingParams`. The values
+            passed via this dataclass supersede the values of the parameters
+            in `generation_parameters` and `sampling_parameters`. See the
+            vLLM documentation for more details: https://docs.vllm.ai/en/latest/dev/sampling_params.html.
+        use_tqdm
+            A boolean in order to display progress bar while inferencing
 
         Returns
         -------
-        Union[str, list[str]]
-            The text generated by the model.
+        The generated text, of shape `(n_batch, n_samples)`. If there are only
+        one batch and several samples, the list is of shape `(n_samples)`. If
+        this is a batch with several sequences but only one sample the list is
+        of shape `(n_batch)`. If there is only one sequence and one sample, a
+        string is returned.
 
         """
-        client_args = self._build_client_args(
-            model_input,
-            output_type,
-            **inference_kwargs,
+        from vllm.sampling_params import SamplingParams
+
+        if sampling_params is None:
+            sampling_params = SamplingParams()
+
+        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
+
+        # We only update the values in `sampling_params` if they
+        # are specified by the user when calling the generator.
+        if max_tokens is not None:
+            sampling_params.max_tokens = max_tokens
+        if stop_at is not None:
+            if isinstance(stop_at, str):
+                stop_at = [stop_at]
+            sampling_params.stop = stop_at
+        if seed is not None:
+            sampling_params.seed = seed
+
+        sampling_params.logits_processors = (
+            [logits_processor] if logits_processor is not None else []
         )
 
-        response = self.client.chat.completions.create(**client_args)
-
-        messages = [choice.message for choice in response.choices]
-        for message in messages:
-            if message.refusal is not None:  # pragma: no cover
-                raise ValueError(
-                    f"The vLLM server refused to answer the request: "
-                    f"{message.refusal}"
-                )
-
-        if len(messages) == 1:
-            return messages[0].content
-        else:
-            return [message.content for message in messages]
-
-    def generate_stream(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Iterator[str]:
-        """Stream text using vLLM.
-
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Iterator[str]
-            An iterator that yields the text generated by the model.
-
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
+        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
+            sampling_parameters
         )
 
-        stream = self.client.chat.completions.create(
-            **client_args, stream=True,
+        # We only update the values in `sampling_params` that
+        # were not specified by the user.
+        if sampling_params.n == 1:
+            sampling_params.n = num_samples
+            sampling_params.best_of = num_samples
+        if top_p is not None and sampling_params.top_p == 1.0:
+            sampling_params.top_p = top_p
+        if top_k is not None and sampling_params.top_k == -1:
+            sampling_params.top_k = top_k
+            # TODO: remove this if statement once fixed
+            # https://github.com/vllm-project/vllm/issues/5404#issuecomment-2175972897
+            if top_k == 1:
+                sampling_params.repetition_penalty = 0
+        if temperature is not None and sampling_params.temperature == 1.0:
+            sampling_params.temperature = temperature
+        if sampler == "beam_search":
+            sampling_params.use_beam_search = True
+
+        results = self.model.generate(
+            prompts,
+            sampling_params=sampling_params,
+            lora_request=self.lora_request,
+            use_tqdm=use_tqdm,
         )
+        results = [[sample.text for sample in batch.outputs] for batch in results]
 
-        for chunk in stream:  # pragma: no cover
-            if chunk.choices and chunk.choices[0].delta.content is not None:
-                yield chunk.choices[0].delta.content
+        batch_size = len(results)
+        sample_size = len(results[0])
 
-    def _build_client_args(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> dict:
-        """Build the arguments to pass to the OpenAI client."""
-        messages = self.type_adapter.format_input(model_input)
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        extra_body = inference_kwargs.pop("extra_body", {})
-        extra_body.update(output_type_args)
-
-        if "model" not in inference_kwargs and self.model_name is not None:
-            inference_kwargs["model"] = self.model_name
-
-        client_args = {
-            **messages,
-            **inference_kwargs,
-        }
-        if extra_body:
-            client_args["extra_body"] = extra_body
-
-        return client_args
-
-
-class AsyncVLLM(AsyncModel):
-    """Thin async wrapper around the `openai.OpenAI` client used to communicate
-    with a `vllm` server.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `openai.OpenAI` client for the
-    `vllm` server.
-    """
+        if batch_size == 1 and sample_size == 1:
+            return results[0][0]
+        elif batch_size == 1:
+            return results[0]
+        elif sample_size == 1:
+            return [batch[0] for batch in results]
 
-    def __init__(
-        self,
-        client: "AsyncOpenAI",
-        model_name: Optional[str] = None,
-    ):
-        """
-        Parameters
-        ----------
-        client
-            An `openai.AsyncOpenAI` client instance.
+        return results
 
-        """
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = VLLMTypeAdapter()
+    def stream(self, *args, **kwargs):
+        """Return a text generator.
 
-    async def generate(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Union[str, list[str]]:
-        """Generate text using vLLM.
+        Streaming is not yet available for `vllm.LLM`.
 
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Returns
-        -------
-        Union[str, list[str]]
-            The text generated by the model.
+        TODO: Implement the streaming functionality ourselves.
 
         """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
+        raise NotImplementedError(
+            "Streaming is not available for the vLLM integration."
         )
 
-        response = await self.client.chat.completions.create(**client_args)
+    def load_lora(self, adapter_path: Optional[str]):
+        from vllm.lora.request import LoRARequest
 
-        messages = [choice.message for choice in response.choices]
-        for message in messages:
-            if message.refusal is not None:  # pragma: no cover
-                raise ValueError(
-                    f"The vLLM server refused to answer the request: "
-                    f"{message.refusal}"
-                )
-
-        if len(messages) == 1:
-            return messages[0].content
+        if adapter_path is None:
+            self.lora_request = None
         else:
-            return [message.content for message in messages]
+            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
 
-    async def generate_stream( # type: ignore
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> AsyncIterator[str]:
-        """Stream text using vLLM.
 
-        Parameters
-        ----------
-        model_input
-            The prompt based on which the model will generate a response.
-        output_type
-            The desired format of the response generated by the model. All
-            output types available in Outlines are supported provided your
-            server uses a structured generation backend that supports them.
-        inference_kwargs
-            Additional keyword arguments to pass to the client.
+def vllm(model_name: str, **vllm_model_params):
+    """Load a vLLM model.
 
-        Returns
-        -------
-        AsyncIterator[str]
-            An async iterator that yields the text generated by the model.
-        """
-        client_args = self._build_client_args(
-            model_input, output_type, **inference_kwargs,
-        )
+    Parameters
+    ---------
+    model_name
+        The name of the model to load from the HuggingFace hub.
+    vllm_model_params
+        vLLM-specific model parameters. See the vLLM code for the full list:
+        https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/llm.py
 
-        stream = await self.client.chat.completions.create(
-            **client_args,
-            stream=True,
-        )
+    """
+    from vllm import LLM
 
-        async for chunk in stream:  # pragma: no cover
-            if chunk.choices and chunk.choices[0].delta.content is not None:
-                yield chunk.choices[0].delta.content
+    model = LLM(model_name, **vllm_model_params)
 
-    def _build_client_args(
-        self,
-        model_input: Union[str, Vision],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> dict:
-        """Build the arguments to pass to the OpenAI client."""
-        messages = self.type_adapter.format_input(model_input)
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        extra_body = inference_kwargs.pop("extra_body", {})
-        extra_body.update(output_type_args)
-
-        if "model" not in inference_kwargs and self.model_name is not None:
-            inference_kwargs["model"] = self.model_name
-
-        client_args = {
-            **messages,
-            **inference_kwargs,
-        }
-        if extra_body:
-            client_args["extra_body"] = extra_body
-
-        return client_args
-
-
-def from_vllm(
-    client: Union["OpenAI", "AsyncOpenAI"],
-    model_name: Optional[str] = None,
-) -> Union[VLLM, AsyncVLLM]:
-    """Create an Outlines `VLLM` or `AsyncVLLM` model instance from an
-    `openai.OpenAI` or `openai.AsyncOpenAI` instance.
+    return VLLM(model)
+
+
+def adapt_tokenizer(tokenizer: "PreTrainedTokenizerBase") -> "PreTrainedTokenizerBase":
+    """Adapt a tokenizer to use to compile the FSM.
+
+    The API of Outlines tokenizers is slightly different to that of `transformers`. In
+    addition we need to handle the missing spaces to Llama's tokenizer to be able to
+    compile FSMs for this model.
 
     Parameters
     ----------
-    client
-        An `openai.OpenAI` or `openai.AsyncOpenAI` instance.
-    model_name
-        The name of the model to use.
+    tokenizer
+        The tokenizer of the model.
 
     Returns
     -------
-    Union[VLLM, AsyncVLLM]
-        An Outlines `VLLM` or `AsyncVLLM` model instance.
-
+    PreTrainedTokenizerBase
+        The adapted tokenizer.
     """
-    from openai import AsyncOpenAI, OpenAI
-
-    if isinstance(client, OpenAI):
-        return VLLM(client, model_name)
-    elif isinstance(client, AsyncOpenAI):
-        return AsyncVLLM(client, model_name)
-    else:
-        raise ValueError(
-            f"Unsupported client type: {type(client)}.\n"
-            "Please provide an OpenAI or AsyncOpenAI instance."
-        )
+    from transformers import SPIECE_UNDERLINE
+
+    tokenizer.vocabulary = tokenizer.get_vocab()
+    tokenizer.special_tokens = set(tokenizer.all_special_tokens)
+
+    def convert_token_to_string(token: Union[str, bytes]) -> str:
+        string = tokenizer.convert_tokens_to_string([token])
+
+        # A hack to handle missing spaces to HF's Llama tokenizers
+        if (
+            type(token) is str
+            and token.startswith(SPIECE_UNDERLINE)
+            or token == "<0x20>"
+        ):
+            return " " + string
+
+        return string
+
+    tokenizer.convert_token_to_string = convert_token_to_string
+
+    return tokenizer
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
deleted file mode 100644
index f970b2f9a..000000000
--- a/outlines/models/vllm_offline.py
+++ /dev/null
@@ -1,206 +0,0 @@
-"""Integration with the `vllm` library (offline mode)."""
-
-import json
-import warnings
-from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Any, List, Optional, Union
-
-from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
-
-if TYPE_CHECKING:
-    from transformers import PreTrainedTokenizerBase
-    from vllm import LLM
-
-__all__ = ["VLLMOffline", "from_vllm_offline"]
-
-
-class VLLMOfflineTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `VLLMOffline` model."""
-
-    @singledispatchmethod
-    def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
-
-        Argument
-        --------
-        model_input
-            The input passed by the user.
-
-        """
-        raise NotImplementedError(
-            f"The input type {input} is not available. "
-            "Please use a string or a list of strings."
-        )
-
-    @format_input.register(str)
-    def format_str_input(self, model_input: str) -> str:
-        return model_input
-
-    @format_input.register(list)
-    def format_list_input(self, model_input: List[str]) -> List[str]:
-        return model_input
-
-    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
-        """Generate the structured output argument to pass to the model.
-
-        For vLLM, the structured output definition is set in the
-        `GuidedDecodingParams` constructor that is provided as a value to the
-        `guided_decoding` parameter of the `SamplingParams` constructor, itself
-        provided as a value to the `sampling_params` parameter of the `generate`
-        method.
-
-        Parameters
-        ----------
-        output_type
-            The structured output type provided.
-
-        Returns
-        -------
-        dict
-            The arguments to provide to the `GuidedDecodingParams` constructor.
-
-        """
-        if output_type is None:
-            return {}
-
-        term = python_types_to_terms(output_type)
-        if isinstance(term, CFG):
-            return {"grammar": term.definition}
-        elif isinstance(term, JsonSchema):
-            guided_decoding_params = {"json": json.loads(term.schema)}
-            if term.whitespace_pattern:
-                guided_decoding_params["whitespace_pattern"] = term.whitespace_pattern
-            return guided_decoding_params
-        else:
-            return {"regex": to_regex(term)}
-
-
-class VLLMOffline(Model):
-    """Thin wrapper around a `vllm.LLM` model.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `vllm.LLM` model.
-
-    """
-
-    def __init__(self, model: "LLM"):
-        """Create a VLLM model instance.
-
-        Parameters
-        ----------
-        model
-            A `vllm.LLM` model instance.
-
-        """
-        self.model = model
-        self.type_adapter = VLLMOfflineTypeAdapter()
-        self.lora_request = None # v0 legacy, to be removed
-
-    def generate(
-        self,
-        model_input: Union[str, List[str]],
-        output_type: Optional[Any] = None,
-        **inference_kwargs: Any,
-    ) -> Union[str, List[str], List[List[str]]]:
-        """Generate text using vLLM.
-
-        Parameters
-        ----------
-        prompt
-            The prompt based on which the model will generate a response.
-        output_type
-            The logits processor the model will use to constrain the format of
-            the generated text.
-        inference_kwargs
-            Additional keyword arguments to pass to the `generate` method
-            in the `vllm.LLM` model.
-
-        Returns
-        -------
-        Union[str, List[str], List[List[str]]]
-            The text generated by the model.
-
-        """
-        from vllm.sampling_params import GuidedDecodingParams, SamplingParams
-
-        sampling_params = inference_kwargs.pop("sampling_params", None)
-
-        if sampling_params is None:
-            sampling_params = SamplingParams()
-
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        if output_type_args:
-            sampling_params.guided_decoding = GuidedDecodingParams(**output_type_args)
-
-        results = self.model.generate(
-            self.type_adapter.format_input(model_input),
-            sampling_params=sampling_params,
-            lora_request=self.lora_request, # v0 legacy, to be removed
-            **inference_kwargs,
-        )
-        results = [[sample.text for sample in batch.outputs] for batch in results]
-
-        batch_size = len(results)
-        sample_size = len(results[0])
-
-        if batch_size == 1 and sample_size == 1:
-            return results[0][0]
-        elif batch_size == 1:
-            return results[0]
-        elif sample_size == 1:
-            return [batch[0] for batch in results]
-
-        return results
-
-    def generate_stream(self, model_input, output_type, **inference_kwargs):
-        """Not available for `vllm.LLM`.
-
-        TODO: Implement the streaming functionality ourselves.
-
-        """
-        raise NotImplementedError(
-            "Streaming is not available for the vLLM integration."
-        )
-
-    def load_lora(self, adapter_path: Optional[str]) -> None:
-        """Load a LoRA adapter. Deprecated since v1.0.0.
-
-        Use the `lora_request` argument when calling the model or generator
-        instead.
-
-        """
-        warnings.warn("""
-            The `load_lora` method is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.1.0.
-            Please use the v1 of the `outlines` library by using the
-            `outlines.from_vllm` function to create a `VLLM` model
-            instance.
-            In the v1, you must pass the `lora_request` argument as
-            a keyword argument when calling the model or generator.
-            """)
-
-        from vllm.lora.request import LoRARequest
-
-        if adapter_path is None:
-            self.lora_request = None
-        else:
-            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
-
-
-def from_vllm_offline(model: "LLM") -> VLLMOffline:
-    """Create an Outlines `VLLMOffline` model instance from a `vllm.LLM`
-    instance.
-
-    Parameters
-    ----------
-    model
-        A `vllm.LLM` instance.
-
-    Returns
-    -------
-    VLLMOffline
-        An Outlines `VLLMOffline` model instance.
-
-    """
-    return VLLMOffline(model)
diff --git a/outlines/processors/__init__.py b/outlines/processors/__init__.py
index f07ce0cbb..f0f0f829b 100644
--- a/outlines/processors/__init__.py
+++ b/outlines/processors/__init__.py
@@ -1,5 +1,3 @@
-"""Processors and guides to control generation in steerable models."""
-
 from .structured import (
     CFGLogitsProcessor,
     GuideLogitsProcessor,
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index bde129b2c..d6fe346e0 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -1,121 +1,159 @@
-"""Base class for logits processors."""
-
 from abc import abstractmethod
-from typing import TypeVar
+from typing import TYPE_CHECKING, List, Protocol, Type, Union
+
+import numpy as np
+import torch
+from numpy.typing import NDArray
+
+if TYPE_CHECKING:
+    import mlx.core as mx
+
+
+Array = Union[NDArray, torch.Tensor, List, "mx.array"]
 
-from outlines.processors.tensor_adapters import (
-    TensorAdapterImplementation,
-    tensor_adapters,
-)
 
-TensorType = TypeVar('TensorType')
+def is_mlx_array_type(array_type):
+    try:
+        import mlx.core as mx
+    except ImportError:
+        return False
+    return issubclass(array_type, mx.array)
 
 
-class OutlinesLogitsProcessor:
-    """Base class for logits processors.
-    This class implements a shared `__call__` method is called by the models
-    and returns the processed logits. It relies on the `process_logits` method
-    that must be implemented by the subclasses to do the actual processing. The
-    `tensor_adapter` attribute, created at initialization based on the
-    tensor library name specified in the constructor, is used to manipulate the
-    tensors using the appropriate library for the model (numpy, torch...).
+def is_jax_array_type(array_type):
+    try:
+        import jaxlib
+    except ImportError:
+        return False
+    return issubclass(array_type, jaxlib.xla_extension.ArrayImpl) or isinstance(
+        array_type, jaxlib.xla_extension.ArrayImpl
+    )
+
+
+class OutlinesLogitsProcessor(Protocol):
     """
-    tensor_adapter: TensorAdapterImplementation
+    Base class for logits processors which normalizes types of logits:
+    - ndarray (used by llama-cpp-python), converted to torch.Tensor
+    - mlx.core.array (used by mlx-lm), converted to torch.Tensor
+    - torch.Tensor (used by everything else)
 
-    def __init__(self, tensor_library_name: str):
-        """
-        Parameters
-        ----------
-        tensor_library_name
-            The name of the library to use to manipulate tensors. Possible
-            values are "jax", "mlx", "numpy", "tensorflow" and "torch". You
-            must choose the library that your model is using.
-        """
-        tensor_adapter_class = tensor_adapters.get(tensor_library_name)
-        if tensor_adapter_class is None:
-            raise NotImplementedError(
-                f"Library {tensor_library_name} is not available"
-            )
-        self.tensor_adapter = tensor_adapter_class()  # type: ignore
+    Normalization of types and conversion to torch.Tensor
+    doesn't move memory, it just casts the type.
+
+    Normalizing the types allows all logits processors inheriting from this class
+    to implement a single method for all the business logit: `process_logits()`
+    """
 
     @abstractmethod
     def process_logits(
-        self, input_ids: TensorType, logits: TensorType
-    ) -> TensorType:
-        """Main method to implement for logits processors subclasses.
-        This method applies a mask on the logits to bias the generation.
-        It is called by the `__call__` method that standardizes the shape of
-        `input_ids` and `logits` to ensure they are 2D tensors.
-        Elements to keep in mind when designing universal logits processors:
-        - logits processors are only used once and never re-applied for a new
-        sequence generator
-        - Some models only pass output_ids, some models such as llamacpp and
-        transformers prefix with input_ids
-        - Some sampling methods, such as beam search, result in unstable
-        sequence ordering in models like vLLM
-        Parameters
-        ----------
-        input_ids
-            The ids of the tokens of the existing sequences in a 2D tensor.
-        logits
-            The logits for the current generation step in a 2D tensor.
-        Returns
-        -------
-        TensorType
-            The processed logits as a 2D tensor.
+        self, input_ids: List[List[int]], logits: torch.Tensor
+    ) -> torch.Tensor:
+        """
+        input_ids and logits are always 2D tensors for handling a batch of sequences.
+
+        - input_ids -> List[List[tokens]]
+        - logits -> 2D_Tensor[logit floats]
+
+        Important to keep in mind when designing universal logits processors
+        - logits processors are only used once and never re-applied for a new sequence generator
+        - Some models only pass output_ids, some models such as llamacpp and transformers prefix with input_ids
+        - Some sampling methods, such as beam search, result in unstable sequence ordering in models like vLLM
         """
-        ...
+        pass
 
+    @torch.no_grad()
     def __call__(
-        self, input_ids: TensorType, logits: TensorType
-    ) -> TensorType:
-        """Entrypoint for logits processors, this is the method that is
-        called by the model.
-        Because different models use different structures to store the
-        input_ids and logits, we standardize their format to 2D tensors
-        before calling the `process_logits` method. After processing, the
-        logits are cast back to the original array library type before being
-        returned.
-        Parameters
-        ----------
-        input_ids
-            The ids of the tokens of the existing sequences in a tensor.
-        logits
-            The logits for the current generation step in a tensor.
-        Returns
-        -------
-        TensorType
-            The processed logits as a tensor.
+        self,
+        input_ids: Array,
+        logits: Array,
+    ) -> Array:
         """
-        # if input_ids is 1D and logits is 2D with a single sequence,
-        # reshape input_ids to 2D (needed for mlx-lm)
-        if (
-            len(self.tensor_adapter.shape(input_ids)) == 1
-            and len(self.tensor_adapter.shape(logits)) == 2
-            and self.tensor_adapter.shape(logits)[0] == 1
-        ):
-            input_ids = self.tensor_adapter.unsqueeze(input_ids)
-
-        assert (
-            self.tensor_adapter.shape(logits)[:-1]
-            == self.tensor_adapter.shape(input_ids)[:-1]
-        )
-
-        # Guarantee passed as 2D Tensors, then covert back to original
-        # (1D or 2D) shape
-        if len(self.tensor_adapter.shape(logits)) == 2:
-            processed_logits = self.process_logits(input_ids, logits)
-        elif len(self.tensor_adapter.shape(logits)) == 1:
-            processed_logits = self.tensor_adapter.squeeze(
-                self.process_logits(
-                    self.tensor_adapter.unsqueeze(input_ids),
-                    self.tensor_adapter.unsqueeze(logits),
-                ),
-            )
+        Apply logits processor
+
+        1) Unify type
+        - convert input_ids: either ndarray, mlx array, List[int], or Tensor -> List[List[int]]
+        - convert logits: either ndarray, mlx array, or Tensor -> 2D float Tensor
+        2) Unify shape, ensure logits and input_ids are 2D
+        3) Call self.process_logits() to perform business logic
+        4) Cast logits back to original array library type
+        """
+        # ensure logits are torch Tensors
+        torch_logits = self._to_torch(logits)
+        input_ids = self._to_torch(input_ids)
+
+        assert torch_logits.shape[:-1] == input_ids.shape[:-1]
+
+        # Guarantee passed as 2D Tensors, then covert back to original (1D or 2D) shape
+        if len(torch_logits.shape) == 2:
+            processed_logits = self.process_logits(input_ids, torch_logits)
+        elif len(torch_logits.shape) == 1:
+            processed_logits = self.process_logits(
+                input_ids.unsqueeze(0), torch_logits.unsqueeze(0)
+            ).squeeze(0)
+
+        # return logits as passed array type
+        return self._from_torch(processed_logits, type(logits))
+
+    @staticmethod
+    def _to_torch(tensor_like: Array) -> torch.Tensor:
+        """Convert various types to torch.Tensor."""
+        if isinstance(tensor_like, torch.Tensor):
+            return tensor_like
+
+        elif isinstance(tensor_like, np.ndarray):
+            return torch.from_numpy(tensor_like)
+
+        elif isinstance(tensor_like, (list, tuple)):
+            return torch.tensor(tensor_like)
+
+        elif is_mlx_array_type(type(tensor_like)):
+            import mlx.core as mx
+
+            # https://ml-explore.github.io/mlx/build/html/usage/numpy.html#pytorch
+            if tensor_like.dtype == mx.bfloat16:
+                tensor_like = tensor_like.astype(mx.float32)
+            return torch.from_dlpack(np.array(tensor_like, copy=False))
+
+        elif is_jax_array_type(type(tensor_like)):
+            import jax
+
+            torch_tensor = torch.from_dlpack(jax.dlpack.to_dlpack(tensor_like))
+            return torch_tensor
+
         else:
-            raise ValueError(
-                f"Logits shape {self.tensor_adapter.shape(logits)} is not "
-                + "supported"
+            raise TypeError(
+                "LogitsProcessor must be called with either np.NDArray, "
+                "torch.Tensor, list, or mlx.core.array typed logits. "
+                f"Logits type: `{type(tensor_like)}`"
             )
 
-        return processed_logits
+    @staticmethod
+    def _from_torch(tensor: torch.Tensor, target_type: Type) -> Array:
+        """Convert torch.Tensor to the specified target type."""
+        if target_type == torch.Tensor:
+            return tensor
+
+        elif target_type == np.ndarray:
+            return tensor.detach().numpy()
+
+        elif target_type is list:
+            return tensor.detach().tolist()
+
+        elif target_type is tuple:
+            return tuple(tensor.detach().tolist())
+
+        elif is_mlx_array_type(target_type):
+            import mlx.core as mx
+
+            # numpy doesn't support bfloat16, mlx doesn't support direct conversion from torch
+            return mx.array(tensor.float().numpy())
+
+        elif is_jax_array_type(target_type):
+            import jax
+
+            return jax.dlpack.from_dlpack(tensor)
+
+        else:
+            raise TypeError(
+                f"Failed to convert torch tensors to target_type `{target_type}`"
+            )
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index f1a55b902..8ce69c32a 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -1,4 +1,4 @@
-"""Logits processors for structured generation.
+"""
  _______________________________
 / Don't want to self-host?       \
 \\ Try .json at http://dottxt.co /
@@ -27,90 +27,79 @@
 import math
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
 
-from pydantic import BaseModel
+import torch
 from outlines_core.fsm.json_schema import build_regex_from_schema
+from pydantic import BaseModel
+
+from outlines.fsm.guide import CFGGuide, Guide, RegexGuide
+from outlines.fsm.json_schema import convert_json_schema_to_str
 
-from outlines.processors.base_logits_processor import (
-    OutlinesLogitsProcessor,
-    TensorType
-)
-from outlines.processors.guide import (
-    CFGGuide,
-    Guide,
-    RegexGuide
-)
-from outlines.types import JsonSchema
+from .base_logits_processor import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     from outlines.models.tokenizer import Tokenizer
 
 
 class GuideLogitsProcessor(OutlinesLogitsProcessor):
-    """Bias generation using a guide.
+    """Bias generation using a finite
 
     Attributes
     ----------
     tokenizer
-        The outlines tokenizer used to convert tokens to ids.
+        The tokenizer used to convert tokens to ids.
     guide
-        The outlines guide used to bias the logits.
+        The `outlines.fsm.Guide` which is used to bias the logits.
     """
+
     tokenizer: "Tokenizer"
     guide: Guide
     _guide_states: Dict[int, Any]
     _seq_start_idx: Optional[int]
 
-    def __init__(
-        self, tokenizer: "Tokenizer", guide: Guide, tensor_library_name: str
-    ):
-        """
+    def __init__(self, tokenizer: "Tokenizer", guide: Guide):
+        """A Guide-based logits processor.
+
         Parameters
         ----------
         tokenizer
             The tokenizer used to convert tokens to ids.
         guide
-            The `outlines.processors.guide.Guide` that is used to bias the
-            logits.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
-
+            The `outlines.fsm.Guide. which is used to bias the logits.
         """
-        super().__init__(tensor_library_name=tensor_library_name)
         self.tokenizer = tokenizer
         self.guide = guide
         self._guide_states = {hash(tuple([])): self.guide.initial_state}
         self._seq_start_idx = None
 
     def process_logits(
-        self, input_ids: TensorType, logits: TensorType
-    ) -> TensorType:
+        self, input_ids: torch.LongTensor, logits: torch.FloatTensor
+    ) -> torch.Tensor:
         """Use the Guide to bias the logits before sampling the next token.
 
         Parameters
         ----------
         input_ids
-            The ids of the tokens of the existing sequences.
+            The input token ids.
         logits
-            The logits for the current generation step.
+            The logits.
 
         Returns
         -------
-        TensorType
+        torch.Tensor
             The biased logits.
-
         """
         if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0]) # type: ignore
+            self._seq_start_idx = len(input_ids[0])
 
         sequence_states: List[int] = []  # vector of states corresponding to `input_ids`
 
-        for seq_ids in input_ids: # type: ignore
+        for seq_ids in input_ids:
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
+            curr_state_key = hash(tuple(gen_ids.tolist()))
 
             if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
-                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
+                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
+                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])
@@ -121,165 +110,136 @@ def process_logits(
             allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
             allowed_tokens_batch.append(allowed_tokens)
             batch_indices.append(
-                self.tensor_adapter.full_like(allowed_tokens, i)
+                torch.full_like(allowed_tokens, i)
             )  # Store batch index for each allowed token
 
-        device = self.tensor_adapter.get_device(logits)
-        allowed_tokens_concat = self.tensor_adapter.to_device(
-            self.tensor_adapter.concatenate(allowed_tokens_batch),
-            device
-        )
-        batch_indices_concat = self.tensor_adapter.to_device(
-            self.tensor_adapter.concatenate(batch_indices),
-            device
-        )
-
-        mask = self.tensor_adapter.boolean_ones_like(logits)
+        allowed_tokens_concat = torch.cat(allowed_tokens_batch).to(logits.device)
+        batch_indices_concat = torch.cat(batch_indices).to(logits.device)
+
+        mask = torch.ones_like(logits, dtype=torch.bool)
         mask[batch_indices_concat, allowed_tokens_concat] = False
-        logits = self.tensor_adapter.apply_mask(logits, mask, float("-inf"))
+        logits.masked_fill_(mask, float("-inf"))
 
         return logits
 
     def copy(self) -> "GuideLogitsProcessor":
         """Return a copy of the logits processor."""
-        return GuideLogitsProcessor(
-            tokenizer=self.tokenizer,
-            guide=self.guide.copy(),
-            tensor_library_name=self.tensor_adapter.library_name
-        )
+        return GuideLogitsProcessor(tokenizer=self.tokenizer, guide=self.guide.copy())
 
 
 class RegexLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a regular expression."""
-    guide: RegexGuide
+    """Bias generation based on a regular expression.
+
+    Attributes
+    ----------
+    tokenizer
+        The tokenizer used to convert tokens to ids.
+    guide
+        The `outlines.fsm.RegexGuide. which is used to bias the logits.
+    """
+
+    def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
+        """Compile the RegexGuide that drives the regex-guided generation.
 
-    def __init__(
-        self,
-        regex_string: str,
-        tokenizer: "Tokenizer",
-        tensor_library_name: str,
-    ):
-        """
         Parameters
         ----------
         regex_string
-            A string that represents a regular expression.
+            A string that represents a regular expression
         tokenizer
-            An Outlines tokenizer.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
-
+            An Outlines tokenizer
         """
-        # Build a guide from the regex string and then pass it to the
-        # GuideLogitsProcessor superclass.
         guide = RegexGuide.from_regex(regex_string, tokenizer)
-        super().__init__(tokenizer=tokenizer, guide=guide, tensor_library_name=tensor_library_name)
+        super().__init__(tokenizer=tokenizer, guide=guide)
 
 
 class JSONLogitsProcessor(RegexLogitsProcessor):
-    """Bias generation based on a JSON schema."""
+    """Bias generation based on a JSON schema.
+
+    Attributes
+    ----------
+    tokenizer
+        The tokenizer used to convert tokens to ids.
+    guide
+        The `outlines.fsm.RegexGuide. which is used to bias the logits.
+    """
+
     def __init__(
         self,
         schema: Union[dict, Type[BaseModel], str],
         tokenizer: "Tokenizer",
-        tensor_library_name: str,
         whitespace_pattern: Optional[str] = None,
     ):
-        """
+        """Compile the Guide that drives the JSON-guided generation.
+
         Parameters
         ----------
         schema
             A JSON schema that encodes the structure we want the model to generate.
         tokenizer
             The tokenizer used to convert tokens to ids.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
         whitespace_pattern
             Pattern to use for JSON syntactic whitespace (doesn't impact string
             literals). For example, to allow only a single space or newline with
-            `whitespace_pattern=r"[\n ]?"`.
-
+            `whitespace_pattern=r"[\n ]?"`
         """
-        # Convert the JSON schema into a regex string and then pass it to the
-        # RegexLogitsProcessor superclass.
-        schema_str = JsonSchema(schema).schema
+        schema_str = convert_json_schema_to_str(json_schema=schema)
         regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(
-            regex_string=regex_string,
-            tokenizer=tokenizer,
-            tensor_library_name=tensor_library_name
-        )
+        super().__init__(regex_string=regex_string, tokenizer=tokenizer)
 
 
 class CFGLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a context-free grammar."""
+    """Bias generation based on a context-free grammar.
+
+    Attributes
+    ----------
+    tokenizer
+        The tokenizer used to convert tokens to ids.
+    guide
+        The `outlines.fsm.CFGGuide. which is used to bias the logits.
+    """
+
     guide: CFGGuide
 
-    def __init__(
-        self, cfg_str: str, tokenizer: "Tokenizer", tensor_library_name: str
-    ):
-        """
+    def __init__(self, cfg_str: str, tokenizer: "Tokenizer"):
+        """Compile the CFGGuide that drives the CFG-guided generation.
+
         Parameters
         ----------
         cfg_str
-            A string that represents a grammar.
+            A string that represents a grammar
         tokenizer
             The tokenizer used to convert tokens to ids.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
-
         """
-        # Build a guide from the CFG string and then pass it to the
-        # GuideLogitsProcessor superclass.
         cfg_guide = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
-        super().__init__(
-            tokenizer=tokenizer,
-            guide=cfg_guide,
-            tensor_library_name=tensor_library_name
-        )
+        super().__init__(tokenizer=tokenizer, guide=cfg_guide)
 
     def process_logits(
-        self, input_ids: TensorType, logits: TensorType
-    ) -> TensorType:
-        """Same behavior as GuideLogitsProcessor, but uses rejection
-        sampling.
-
-        Parameters
-        ----------
-        input_ids
-            The ids of the tokens of the existing sequences.
-        logits
-            The logits for the current generation step.
-
-        Returns
-        -------
-        TensorType
-            The biased logits.
-
-        """
+        self, input_ids: torch.LongTensor, logits: torch.Tensor
+    ) -> torch.Tensor:
+        """Same behavior as GuideLogitsProcessor, but uses rejection sampling"""
         if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0]) # type: ignore
+            self._seq_start_idx = len(input_ids[0])
 
         sequence_states: List = []  # vector of states corresponding to `input_ids`
 
-        for seq_ids in input_ids: # type: ignore
+        for seq_ids in input_ids:
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
+            curr_state_key = hash(tuple(gen_ids.tolist()))
 
-            if curr_state_key not in self._guide_states: # pragma: no cover
-                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
-                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
+            if curr_state_key not in self._guide_states:
+                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
+                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])
 
-        mask = self.tensor_adapter.full_like(logits, -math.inf)
+        mask = torch.full_like(logits, -math.inf)
         for i, guide_state in enumerate(sequence_states):
             first_legal_token = next(
                 self.guide.iter_valid_token_ids(
-                    guide_state, self.tensor_adapter.argsort_descending(logits[i]) # type: ignore
+                    guide_state, torch.argsort(logits[i], descending=True)
                 )
             )
-            mask[i, [first_legal_token]] = logits[i, [first_legal_token]] # type: ignore
+            mask[i, [first_legal_token]] = logits[i, [first_legal_token]]
 
         return mask
diff --git a/outlines/processors/tensor_adapters/__init__.py b/outlines/processors/tensor_adapters/__init__.py
deleted file mode 100644
index 8e3afe7f6..000000000
--- a/outlines/processors/tensor_adapters/__init__.py
+++ /dev/null
@@ -1,26 +0,0 @@
-"""Library specific objects to manipulate tensors."""
-
-from typing import Union
-
-from .jax import JAXTensorAdapter
-from .mlx import MLXTensorAdapter
-from .numpy import NumpyTensorAdapter
-from .tensorflow import TensorFlowTensorAdapter
-from .torch import TorchTensorAdapter
-
-
-tensor_adapters = {
-    "jax": JAXTensorAdapter,
-    "mlx": MLXTensorAdapter,
-    "numpy": NumpyTensorAdapter,
-    "tensorflow": TensorFlowTensorAdapter,
-    "torch": TorchTensorAdapter,
-}
-
-TensorAdapterImplementation = Union[
-    JAXTensorAdapter,
-    MLXTensorAdapter,
-    NumpyTensorAdapter,
-    TensorFlowTensorAdapter,
-    TorchTensorAdapter,
-]
diff --git a/outlines/processors/tensor_adapters/base.py b/outlines/processors/tensor_adapters/base.py
deleted file mode 100644
index ffaac58df..000000000
--- a/outlines/processors/tensor_adapters/base.py
+++ /dev/null
@@ -1,257 +0,0 @@
-"""Base class for tensor adapters."""
-
-from abc import ABC, abstractmethod
-from typing import TYPE_CHECKING, TypeVar, Any, Union
-
-if TYPE_CHECKING:
-    import torch
-
-TensorType = TypeVar('TensorType')
-
-
-class TensorAdapter(ABC):
-    """Abstract base class for tensor adapters.
-
-    This class defines the interface for tensor adapters that are used to
-    manipulate tensors in different libraries. Concrete implementations of
-    this class should provide specific implementations for each method as
-    well as providing a `library_name` attribute.
-
-    TODO: Update the version of outlines-core used to receive plain arrays
-    instead of torch tensors. In the meantime, implementations of this class
-    must make sure that their `full_like` and `concatenate` methods can
-    handle torch tensors.
-
-    """
-    library_name: str
-
-    @abstractmethod
-    def shape(self, tensor: TensorType) -> list[int]:
-        """Get the shape of the tensor.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to get the shape of.
-
-        Returns
-        -------
-        list[int]
-            The shape of the tensor. The list contains as many elements as
-            there are dimensions in the tensor.
-
-        """
-        ...
-
-    @abstractmethod
-    def unsqueeze(self, tensor: TensorType) -> TensorType:
-        """Add a dimension to the tensor at axis 0.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to add a dimension to.
-
-        Returns
-        -------
-        TensorType
-            The tensor with an additional dimension.
-
-        """
-        ...
-
-    @abstractmethod
-    def squeeze(self, tensor: TensorType) -> TensorType:
-        """Remove a dimension from the tensor at axis 0.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to remove a dimension from.
-
-        Returns
-        -------
-        TensorType
-            The tensor with one less dimension.
-
-        """
-        ...
-
-    @abstractmethod
-    def to_list(self, tensor: TensorType) -> list:
-        """Convert the tensor to a list.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to convert to a list.
-
-        Returns
-        -------
-        list
-            The tensor as a list.
-
-        """
-        ...
-
-    @abstractmethod
-    def to_scalar(self, tensor: TensorType) -> Any:
-        """Return the only element of the tensor.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to return the only element of.
-
-        Returns
-        -------
-        Any
-            The only element of the tensor.
-
-        """
-        ...
-
-    @abstractmethod
-    def full_like(self, tensor: "torch.Tensor", fill_value: Any) -> TensorType: # type: ignore
-        """Create a tensor with the same shape as the input tensor filled
-        with a scalar value.
-
-        ATTENTION: This method receives a torch tensor regardless of the
-        library used.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to create a new tensor with the same shape.
-        fill_value
-            The value to fill the new tensor with.
-
-        Returns
-        -------
-        TensorType
-            A tensor with the same shape as the input tensor filled with the
-            specified value.
-
-        """
-        ...
-
-    @abstractmethod
-    def concatenate(
-        self, tensors: list[Union["torch.Tensor", TensorType]]
-    ) -> TensorType:
-        """Concatenate a list of tensors along axis 0.
-
-        ATTENTION: This method can either receive a list of torch tensors or
-        a list of tensors from the library used.
-
-        Parameters
-        ----------
-        tensors
-            The list of tensors to concatenate.
-
-        Returns
-        -------
-        TensorType
-            The concatenated tensor.
-
-        """
-        ...
-
-    @abstractmethod
-    def get_device(self, tensor: TensorType) -> str:
-        """Get the name of the tensor's device.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to get the device of.
-
-        Returns
-        -------
-        str
-            The name of the tensor's device.
-
-        """
-        ...
-
-    @abstractmethod
-    def to_device(self, tensor: TensorType, device: str) -> TensorType:
-        """Move the tensor to a specified device.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to move to a specified device.
-        device
-            The name of the device to move the tensor to.
-
-        Returns
-        -------
-        TensorType
-            The tensor moved to the specified device.
-
-        """
-        ...
-
-    @abstractmethod
-    def boolean_ones_like(self, tensor: TensorType) -> TensorType:
-        """Create a boolean ones tensor with the same shape as the input
-        tensor.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to create a boolean ones tensor with the same shape.
-
-        Returns
-        -------
-        TensorType
-            A boolean ones tensor with the same shape as the input tensor.
-
-        """
-        ...
-
-    @abstractmethod
-    def apply_mask(
-        self, tensor: TensorType, mask: TensorType, value: Any
-    ) -> TensorType:
-        """Fill the elements of the tensor where the mask is True with the
-        specified value.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to fill.
-        mask
-            The mask to apply to the tensor.
-        value
-            The value to fill the tensor with.
-
-        Returns
-        -------
-        TensorType
-            The tensor with the mask applied.
-
-        """
-        ...
-
-    @abstractmethod
-    def argsort_descending(
-        self, tensor: TensorType
-    ) -> TensorType:
-        """Return the indices that would sort the tensor in descending order
-        along axis -1.
-
-        Parameters
-        ----------
-        tensor
-            The tensor to sort.
-
-        Returns
-        -------
-        TensorType
-            The indices that would sort the tensor in descending order along
-            axis -1.
-
-        """
-        ...
diff --git a/outlines/processors/tensor_adapters/jax.py b/outlines/processors/tensor_adapters/jax.py
deleted file mode 100644
index 30cd2ecc9..000000000
--- a/outlines/processors/tensor_adapters/jax.py
+++ /dev/null
@@ -1,50 +0,0 @@
-"""Tensor adapter for the `jax` library."""
-
-from outlines.processors.tensor_adapters.base import TensorAdapter
-
-
-class JAXTensorAdapter(TensorAdapter):
-    library_name = "jax"
-
-    def __init__(self):
-        import jax
-
-        self.jax = jax
-
-    def shape(self, tensor):
-        return tensor.shape
-
-    def unsqueeze(self, tensor):
-        return self.jax.numpy.expand_dims(tensor, axis=0)
-
-    def squeeze(self, tensor):
-        return self.jax.numpy.squeeze(tensor, axis=0)
-
-    def to_list(self, tensor):
-        return tensor.tolist()
-
-    def to_scalar(self, tensor):
-        return tensor.item()
-
-    def full_like(self, tensor, fill_value):
-        return self.jax.numpy.full_like(tensor, fill_value)
-
-    def concatenate(self, tensors):
-        return self.jax.numpy.concatenate(tensors, axis=0)
-
-    def to_device(self, tensor, device):
-        return tensor
-
-    def get_device(self, tensor):
-        return None
-
-    def boolean_ones_like(self, tensor):
-        return self.jax.numpy.ones_like(tensor, dtype=bool)
-
-    def apply_mask(self, tensor, mask, value):
-        result = tensor.copy()
-        result = result.at[mask].set(value)
-        return result
-
-    def argsort_descending(self, tensor):
-        return self.jax.numpy.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/mlx.py b/outlines/processors/tensor_adapters/mlx.py
deleted file mode 100644
index 5714808d4..000000000
--- a/outlines/processors/tensor_adapters/mlx.py
+++ /dev/null
@@ -1,60 +0,0 @@
-"""Tensor adapter for the `mlx` library."""
-
-from outlines.processors.tensor_adapters.base import TensorAdapter
-
-
-class MLXTensorAdapter(TensorAdapter):
-    library_name = "mlx"
-
-    def __init__(self):
-        import mlx.core
-
-        self.mlx = mlx.core
-
-    def shape(self, tensor):
-        return tensor.shape
-
-    def unsqueeze(self, tensor):
-        return self.mlx.expand_dims(tensor, 0)
-
-    def squeeze(self, tensor):
-        if tensor.shape[0] == 1:
-            return tensor[0]
-        return tensor
-
-    def to_list(self, tensor):
-        return tensor.tolist()
-
-    def to_scalar(self, tensor):
-        return tensor.item()
-
-    def full_like(self, tensor, fill_value):
-        # Compatible with receiving a torch tensor
-        return self.mlx.full(tensor.shape, fill_value)
-
-    def concatenate(self, tensors):
-        # Can handle both torch and mlx tensors
-        return self.mlx.concatenate(
-            [
-                self.mlx.array(t) if not isinstance(t, self.mlx.array) else t
-                for t in tensors
-            ],
-            axis=0
-        )
-
-    def get_device(self, tensor):
-        return None
-
-    def to_device(self, tensor, device):
-        return tensor
-
-    def boolean_ones_like(self, tensor):
-        return self.mlx.ones(tensor.shape, dtype=self.mlx.bool_)
-
-    def apply_mask(self, tensor, mask, value):
-        result = tensor.astype(tensor.dtype)
-        result = self.mlx.where(mask, self.mlx.array(value), result)
-        return result
-
-    def argsort_descending(self, tensor):
-        return self.mlx.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/numpy.py b/outlines/processors/tensor_adapters/numpy.py
deleted file mode 100644
index 831220444..000000000
--- a/outlines/processors/tensor_adapters/numpy.py
+++ /dev/null
@@ -1,50 +0,0 @@
-"""Tensor adapter for the `numpy` library."""
-
-from outlines.processors.tensor_adapters.base import TensorAdapter
-
-
-class NumpyTensorAdapter(TensorAdapter):
-    library_name = "numpy"
-
-    def __init__(self):
-        import numpy
-
-        self.numpy = numpy
-
-    def shape(self, tensor):
-        return tensor.shape
-
-    def unsqueeze(self, tensor):
-        return self.numpy.expand_dims(tensor, axis=0)
-
-    def squeeze(self, tensor):
-        return self.numpy.squeeze(tensor, axis=0)
-
-    def to_list(self, tensor):
-        return tensor.tolist()
-
-    def to_scalar(self, tensor):
-        return tensor.item()
-
-    def full_like(self, tensor, fill_value):
-        return self.numpy.full_like(tensor, fill_value)
-
-    def concatenate(self, tensors):
-        return self.numpy.concatenate(tensors, axis=0)
-
-    def get_device(self, tensor):
-        return None
-
-    def to_device(self, tensor, device):
-        return tensor
-
-    def boolean_ones_like(self, tensor):
-        return self.numpy.ones_like(tensor, dtype=bool)
-
-    def apply_mask(self, tensor, mask, value):
-        result = tensor.copy()
-        result[mask] = value
-        return result
-
-    def argsort_descending(self, tensor):
-        return self.numpy.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/tensorflow.py b/outlines/processors/tensor_adapters/tensorflow.py
deleted file mode 100644
index 382f847b6..000000000
--- a/outlines/processors/tensor_adapters/tensorflow.py
+++ /dev/null
@@ -1,50 +0,0 @@
-"""Tensor adapter for the `tensorflow` library."""
-
-from outlines.processors.tensor_adapters.base import TensorAdapter
-
-
-class TensorFlowTensorAdapter(TensorAdapter):
-    library_name = "tensorflow"
-
-    def __init__(self):
-        import tensorflow as tf
-
-        self.tf = tf
-
-    def shape(self, tensor):
-        return tensor.shape
-
-    def unsqueeze(self, tensor):
-        return self.tf.expand_dims(tensor, axis=0)
-
-    def squeeze(self, tensor):
-        return self.tf.squeeze(tensor, axis=0)
-
-    def to_list(self, tensor):
-        return tensor.numpy().tolist()
-
-    def to_scalar(self, tensor):
-        return tensor.numpy().item()
-
-    def full_like(self, tensor, fill_value):
-        return self.tf.fill(self.tf.shape(tensor), fill_value)
-
-    def concatenate(self, tensors):
-        return self.tf.concat(tensors, axis=0)
-
-    def get_device(self, tensor):
-        return tensor.device
-
-    def to_device(self, tensor, device):
-        # TensorFlow handles device placement differently
-        with self.tf.device(device):
-            return self.tf.identity(tensor)
-
-    def boolean_ones_like(self, tensor):
-        return self.tf.ones_like(tensor, dtype=self.tf.bool)
-
-    def apply_mask(self, tensor, mask, value):
-        return self.tf.where(mask, self.tf.constant(value, dtype=tensor.dtype), tensor)
-
-    def argsort_descending(self, tensor):
-        return self.tf.argsort(tensor, direction='DESCENDING')
diff --git a/outlines/processors/tensor_adapters/torch.py b/outlines/processors/tensor_adapters/torch.py
deleted file mode 100644
index 4b624e960..000000000
--- a/outlines/processors/tensor_adapters/torch.py
+++ /dev/null
@@ -1,48 +0,0 @@
-"""Tensor adapter for the `torch` library."""
-
-from outlines.processors.tensor_adapters.base import TensorAdapter
-
-
-class TorchTensorAdapter(TensorAdapter):
-    library_name = "torch"
-
-    def __init__(self):
-        import torch
-
-        self.torch = torch
-
-    def shape(self, tensor):
-        return tensor.shape
-
-    def unsqueeze(self, tensor):
-        return tensor.unsqueeze(0)
-
-    def squeeze(self, tensor):
-        return tensor.squeeze(0)
-
-    def to_list(self, tensor):
-        return tensor.tolist()
-
-    def to_scalar(self, tensor):
-        return tensor.item()
-
-    def full_like(self, tensor, fill_value):
-        return self.torch.full_like(tensor, fill_value)
-
-    def concatenate(self, tensors):
-        return self.torch.cat(tensors, dim=0)
-
-    def get_device(self, tensor):
-        return tensor.device
-
-    def to_device(self, tensor, device):
-        return tensor.to(device)
-
-    def boolean_ones_like(self, tensor):
-        return self.torch.ones_like(tensor, dtype=self.torch.bool)
-
-    def apply_mask(self, tensor, mask, value):
-        return self.torch.masked_fill(tensor, mask, value)
-
-    def argsort_descending(self, tensor):
-        return self.torch.argsort(tensor, descending=True)
diff --git a/outlines/release_note.md b/outlines/release_note.md
deleted file mode 100644
index 35812623b..000000000
--- a/outlines/release_note.md
+++ /dev/null
@@ -1,392 +0,0 @@
-# Release Note
-
-### Why a new major version?
-
-The v1 intends on making Outlines more closely focused on constrained generation. To do so, we delegate a wider range of tasks to the users and inference libraries. On top of making Outlines leaner, this design provides more flexibility to the users and let them use interfaces they are already familiar with.
-
-Our approach is inspired by the unix best practices — each element does one thing well, and we compose those functional elements.
-
-As this new version deprecates some previously available features of Outlines, we have written a migration guide that gives detailed information on how to upgrade your v0 code to v1.
-
-### Deprecated
-
-All deprecated features listed below will be removed in version 1.1.0. Until then, a warning will be displayed with information on how to migrate your code to v1.
-
-- The model loader functions from the `models` module (`transformers`, `openai`, etc.) have been deprecated. They are replaced by equivalent functions prefixed with `from_` such as `from_transformers`, `from_openai`, etc. The new loader functions accept different arguments compared to the old ones. They now typically require an instance of an engine/client from the associated inference library. This change was made to avoid duplicating inference library logic and to give users more control over inference engine/client initialization.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
-
-```python
-# v0
-from outlines import models
-from transformers import BertForSequenceClassification, BertTokenizer
-
-model = models.transformers(
-    model_name="prajjwal1/bert-tiny",
-    model_class=BertForSequenceClassification,
-    tokenizer_class=BertTokenizer,
-    model_kwargs={"use_cache": False},
-    tokenizer_kwargs={"model_max_length": 512},
-)
-
-# v1
-import outlines
-from transformers import BertForSequenceClassification, BertTokenizer
-
-hf_model = BertForSequenceClassification.from_pretrained("prajjwal1/bert-tiny", use_cache=False)
-hf_tokenizer = BertTokenizer.from_pretrained("prajjwal1/bert-tiny", model_max_length=512)
-model = outlines.from_transformers(hf_model, hf_tokenizer)
-```
-
-- The `generate` module and the associated functions (`json`, `choice`…) have been deprecated. They are replaced by the `Generator` constructor. While you had to select the right generate function for your output type, you can now provide any output type supported by Outlines to the unique `Generator` object.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/generator)
-
-
-```python
-# v0
-from pydantic import BaseModel
-from outlines import generate, models
-
-class Character(BaseModel):
-	name: str
-
-model = models.openai("gpt-4o")
-generator = generate.json(model, Character)
-
-# v1
-from openai import OpenAI
-from pydantic import BaseModel
-from outlines import Generator, from_openai
-
-class Character(BaseModel):
-	name: str
-
-model = from_openai(OpenAI())
-generator = Generator(model, Character)
-```
-
-- The `TransformersVision` model has been deprecated. It's replaced by `TransformersMultiModal`, which is more general as it supports additional input types beyond images, such as audio. When calling it, instead of providing the prompt and image assets separately, both should now be included in a single dictionary. The model is loaded with `from_transformers` just like the `Transformers` model, but the second argument must be a processor instead of a tokenizer.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models/transformers_multimodal)
-
-
-```python
-# v0
-from io import BytesIO
-from urllib.request import urlopen
-from PIL import Image
-from transformers import LlavaForConditionalGeneration
-from outlines import models, generate
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-model = models.transformers_vision(
-    model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-    model_class=LlavaForConditionalGeneration,
-)
-generator = generate.text(model)
-result = generator(
-    "Describe the image <image>",
-    img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
-)
-
-# v1
-from io import BytesIO
-from urllib.request import urlopen
-from PIL import Image
-from transformers import LlavaForConditionalGeneration, AutoProcessor
-import outlines
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-model = outlines.from_transformers(
-	LlavaForConditionalGeneration.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration"),
-	AutoProcessor.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration")
-)
-image = img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
-result = model({"text": "Describe the image <image>", "images": image})
-```
-
-- The `Exllamav2` model has been deprecated without replacement because its interface is not fully compatible with Outlines. We had to implement cumbersome patching to make it work, so we decided to remove it entirely.
-
-- The `function` module and the associated `Function` class have been deprecated. They are replaced by the `Application` class, which serves a similar purpose to `Function`. There are two notable differences: an `Application` is not initialized with a model (a model must be provided when calling the object), and template variables must be provided in a dictionary instead of as keyword arguments when calling the `Application`.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/utility/application)
-
-
-```python
-# v0
-from pydantic import BaseModel
-from outlines import Function, Template
-
-class Character(BaseModel):
-	name: str
-
-template = Template.from_string("Create a {{ gender }} character.")
-fn = Function(template, Character, "hf-internal-testing/tiny-random-GPTJForCausalLM")
-response = fn(gender="female")
-
-# v1
-from pydantic import BaseModel
-from outlines import Application, Template, from_transformers
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-class Character(BaseModel):
-	name: str
-
-model = from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-template = Template.from_string("Create a {{ gender }} character.")
-app = Application(template, Character)
-response = app(model, {"gender": "female"})
-```
-
-- The `samplers` module and the associated objects (`multinomial`, `greedy`…) have been deprecated. You should now use the inference arguments specific to the inference library your  model is based on to control the sampling.
-
-```python
-# v0
-from outlines import generate, models, samplers
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model, samplers.beam_search(2))
-response = generator("Write a short story about a cat", max_tokens=10)
-
-# v1
-from outlines import Generator, from_transformers
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model = from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-response = model("Write a short story about a cat", num_beams=2)
-```
-
-- The `load_lora` methods on the `VLLM` and `LlamaCpp` models have been deprecated. You should now load through the `Llama` instance provided when initializing the model in the case of the `LlamaCpp` model, and provide it as a keyword argument when calling the model in the case of the `VLLM` model.
-
-```python
-# v0
-from outlines import models
-from vllm import LLM
-
-model = models.vllm("erwanf/gpt2-mini")
-model.load_lora("path/to/lora/file")
-response = model("Write a short story about a cat.")
-
-#v1
-from outlines import from_vllm
-from vllm import LLM
-from vllm.lora.request import LoRARequest
-
-model = from_vllm(
-    LLM("microsoft/Phi-3-mini-4k-instruct")
-)
-lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
-response = model("Write a short story about a cat.", lora_request=lora_request)
-```
-
-### Modified
-
-Some objects are maintained, but their interface or behavior has been modified.
-
-- The interface of `Model` classes (`Transformers`, `OpenAI`, etc.) has been significantly modified. Models can now be called directly with a prompt and an output type without having to create a generator first. Additionally, all models have a `stream` method that can be invoked directly by the user.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
-
-
-```python
-# v0
-from pydantic import BaseModel
-from outlines import generate, models
-
-class Character(BaseModel):
-		name: str
-
-model = models.openai("gpt-4o")
-generator = generate.json(model, Character)
-result = generator("Create a character")
-
-# v1
-from openai import OpenAI
-from pydantic import BaseModel
-from outlines import from_openai
-
-class Character(BaseModel):
-	name: str
-
-model = from_openai(OpenAI(), "gpt-4o")
-result = model("Create a character", Character)
-```
-
-- The interface of the `__init__` method of the `OpenAI` model class has been modified. While it previously accepted a client and an `OpenAIConfig` object instance, it now accepts a client and a model name. The inference arguments from the config object should now be specified when calling the model to more closely align with the OpenAI Python library's functionality. If you provide an `OpenAIConfig` instance when initializing the model, a deprecation warning will appear and your model will behave like a v0 model.
-We recommend using the `from_openai` function instead of initializing models directly.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models/openai)
-
-
-```python
-# v0
-from outlines.models.openai import OpenAI, OpenAIConfig
-from openai import OpenAI as OpenAIClient
-
-model = OpenAI(
-	OpenAIClient(),
-	OpenAIConfig(model="gpt-4o", stop=["."])
-)
-
-# v1
-import outlines
-from openai import OpenAI
-
-model = outlines.from_openai(OpenAIClient(), "gpt-4o")
-```
-
-- The return type of text generation is now consistently a string (or list/lists of strings for multiple samples or batching). In v0, Outlines automatically cast the inference result into the type provided by the user for constrained generation, but we have removed this behavior. This change was made to create more consistent behavior and to give users more freedom in deciding how to handle the generation result.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
-
-```python
-# v0
-from pydantic import BaseModel
-from outlines import generate, models
-
-class Character(BaseModel):
-	name: str
-
-model = models.openai("gpt-4o")
-generator = generate.json(model, Character)
-result = generator("Create a character")
-print(result) # name='James'
-
-# v1
-import openai
-from pydantic import BaseModel
-from outlines import from_openai
-
-class Character(BaseModel):
-		name: str
-
-model = from_openai(OpenAI())
-result = model("Create a character", Character)
-print(result) # {"name": "James"}
-print(Character.model_validate_json(result)) # name='James'
-```
-
-- While Outlines was trying to standardize inference argument names across models in v0, we decided to stop doing so and to directly pass on the inference arguments provided by the user to the inference engine/client. Our objective is to let the user use all arguments they are accustomed to with their inference library instead of having to learn Outlines-defined arguments. The deprecation of the `samplers` mentioned above is a part of this change of approach.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
-
-```python
-# v0
-from outlines import generate, models
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-result = generator("Create a character", max_tokens=256, stop_at=".")
-
-# v1
-from outlines import from_transformers
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model = from_transformers(
-	AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-	AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-result = model("Create a character", max_new_tokens=256, stop_strings=".")
-```
-
-### Added features
-
-- There are 8 additional models available. All of them are loaded with an associated `from_` function that accepts an inference engine/client instance.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
-    - `Dottxt`
-    - `Anthopic`
-    - `Gemini`
-    - `Ollama`
-    - `SGLang`
-    - `TGI`
-    - `TransformersMultiModel`
-    - `VLLM`
-- Some server-based models now have an async version. To create an async model, just provide an async client instance when using the loader function. The async models are the following.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
-    - `AsyncSGLang`
-    - `AsyncTGI`
-    - `AsyncVLLM`
-
-```python
-import outlines
-from huggingface_hub import AsyncInferenceClient
-
-async_model = outlines.from_tgi(AsyncInferenceClient("http://localhost:11434"))
-```
-
-- As explained previously, the `Generator` constructor has been added. It accepts a model and an output type as arguments and returns a generator object that can be used to generate text by providing a prompt and inference arguments. The interest of a generator is that it's reusable such that the user does not have to specify the output type they want each time and the output type compilation (when applicable) happens only once.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/generator)
-
-```python
-# direct model calling
-from typing import Literal
-from outlines import from_transformers
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model = from_transformers(
-		AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
-		AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
-)
-result = model("Pizza or burger", Literal["pizza", "burger"])
-
-# using a generator
-from outlines import Generator, from_transformers
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model = from_transformers(
-		AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-		AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-generator = Generator(model, Literal["pizza", "burger"])
-result = generator("Pizza or burger")
-```
-
-- As explained previously, the `Application` class has been added. An `Application` is initialized with a prompt template and an output type. The application object returned can then be called with a model, a dictionary containing values for the template variables and inference arguments. The objective of this object is to let users easily switch from a model to another for a given set of prompt and output type.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/utility/application)
-
-```python
-from pydantic import BaseModel
-from outlines import Application, Template
-
-class Character(BaseModel):
-	name: str
-
-template = Template.from_string("Create a {{ gender }} character.")
-app = Application(template, Character)
-response = app(model, {"gender": "female"})
-```
-
-- The regex DSL and the associated `Term` classes and functions have been added. Terms (`Regex`, `String`…) can be used as output types to generate text with models or generators (they are turned into a regex). The term functions (`either`, `optional`, `at_least`…) are useful to build more complex regex patterns by combining terms. On top of the objects related to regex patterns, there are also 3 terms that are intended to be used by themselves as output types: `JsonSchema`, `CFG` and `FSM`.
-[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/ouput_types)
-
-```python
-# term used directly as an output type
-from outlines import from_transformers
-from outlines.types import JsonSchema
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model = from_transformers(
-		AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
-		AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
-)
-json_schema = '{"type": "object", "properties": {"answer": {"type": "number"}}}'
-result = model("What's 2 + 2? Respond in a json", JsonSchema(json_schema))
-
-# creating a complex regex pattern
-from outlines import from_transformers
-from outlines.types import at_least, either, integer, optional
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model = from_transformers(
-	AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
-	AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
-)
-regex_term = "I have " + integer + either("dog", "cat") + optional("s")
-result = model("How many pets do you have", regex_term)
-```
diff --git a/outlines/samplers.py b/outlines/samplers.py
new file mode 100644
index 000000000..3fef673b1
--- /dev/null
+++ b/outlines/samplers.py
@@ -0,0 +1,353 @@
+import math
+from dataclasses import dataclass
+from typing import TYPE_CHECKING, Callable, Optional, Protocol, Tuple
+
+if TYPE_CHECKING:
+    import torch
+
+
+class Sampler(Protocol):
+    samples: int
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        rng: "torch.Generator",
+    ) -> "torch.DoubleTensor": ...
+
+
+@dataclass(frozen=True)
+class SamplingParameters:
+    """Sampling parameters available in Outlines."""
+
+    sampler: str
+    num_samples: int = 1
+    top_p: Optional[float] = None
+    top_k: Optional[int] = None
+    temperature: Optional[float] = None
+
+
+class GreedySampler:
+    """Greedy Sampling algorithm.
+
+    Greedy sampling consists in choosing the token with the largest
+    likelihood at every step.
+
+    We don't allow more than one sample. We could attribute this a meaning, for
+    instance the k-th sample represents the k-th most likely token. In which
+    case it would be equivalent to beam search without the sequence weights.
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence.
+
+    """
+
+    def __init__(self):
+        self.samples = 1
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        _,
+    ) -> "torch.DoubleTensor":
+        """Call the greedy sampler.
+
+        Parameters
+        ----------
+        next_token_logits
+            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
+            probability distribution of the next token over the vocabulary.
+        sequence_weights
+            A tensor of shape ``(n_seqs,)`` that represents the cumulative
+            weight of each sequence.
+        rng
+            A random number generator.
+
+        Returns
+        -------
+        A tuple with an array that contains the ids of the sampled tokens of
+        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
+        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
+        cumulative weights of each sequence of shape ``(n_seqs,)``.
+
+        """
+        import torch
+
+        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
+        next_token_ids = torch.argmax(logprobs, dim=-1, keepdim=True)
+
+        ancestors = torch.arange(
+            next_token_logits.shape[0], device=next_token_logits.device
+        )
+        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
+
+        return next_token_ids, ancestors, weights
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters("greedy", self.samples, None, None, 0.0)
+
+
+greedy = GreedySampler
+
+
+class MultinomialSampler:
+    """Multinomial sampling algorithm.
+
+    Multinomial sampling consists in randomly sampling the next token assuming
+    its distribution is a Categorical distribution parametrized by the
+    next-token logits.
+
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence.
+
+    """
+
+    def __init__(
+        self,
+        samples: int = 1,
+        *,
+        top_k: Optional[int] = None,
+        top_p: Optional[float] = None,
+        temperature: Optional[float] = None,
+    ):
+        self.samples = samples
+        self.top_k = top_k
+        self.top_p = top_p
+        self.temperature = temperature
+
+        self.logits_processors = []
+        if top_k is not None:
+            self.logits_processors.append(keep_top_k_logits(top_k))
+        elif top_p is not None:
+            self.logits_processors.append(keep_top_p_logits(top_p))
+
+        if temperature is not None:
+            self.logits_processors.append(rescale_logits(temperature))
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        rng: "torch.Generator",
+    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
+        """Call the multinomial sampler.
+
+        Parameters
+        ----------
+        next_token_logits
+            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
+            probability distribution of the next token over the vocabulary.
+        sequence_weights
+            A tensor of shape ``(n_seqs,)`` that represents the cumulative
+            weight of each sequence.
+        rng
+            A random number generator.
+
+        Returns
+        -------
+        A tuple with an array that contains the ids of the sampled tokens of
+        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
+        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
+        cumulative weights of each sequence of shape ``(n_seqs,)``.
+
+        """
+        import torch
+
+        altered_next_token_logits = next_token_logits
+        for logit_processor in self.logits_processors:
+            altered_next_token_logits = logit_processor(next_token_logits)
+
+        probs = torch.nn.functional.softmax(altered_next_token_logits, dim=-1)
+        next_token_ids = torch.multinomial(probs, num_samples=1, generator=rng)
+
+        logprobs = torch.nn.functional.log_softmax(altered_next_token_logits, dim=-1)
+        ancestors = torch.arange(
+            altered_next_token_logits.shape[0], device=next_token_logits.device
+        )
+        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
+
+        return next_token_ids, ancestors, weights
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters(
+            "multinomial",
+            self.samples,
+            self.top_p,
+            self.top_k,
+            self.temperature,
+        )
+
+
+multinomial = MultinomialSampler
+
+
+def keep_top_k_logits(k: int) -> Callable[["torch.Tensor"], "torch.Tensor"]:
+    """Build a function that masks logits values smaller than the top `k` ones.
+
+    Parameters
+    ----------
+    k
+        The ranking below which logit values are replaced by `-math.inf`.
+
+    """
+    import torch
+
+    if not isinstance(k, int) or k < 1:
+        raise ValueError(f"`k` must be a strictly positive integers, got {k} instead.")
+
+    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
+        num_to_keep = min(k, logits.size(-1))
+        mask_idx = logits < torch.topk(logits, num_to_keep)[0][..., -1, None]
+        return logits.masked_fill(mask_idx, -math.inf)
+
+    return logits_processor
+
+
+def keep_top_p_logits(p: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
+    """Build a function that masks the lowest probability tokens whose
+    cumulative probability is below a certain threshold.
+
+    Parameters
+    ----------
+    p
+        The value of the threshold. We keep the highest probability tokens whose
+        cumulative distribution is greater than or equal to `p` and mask the
+        others. Its value must be between 0 (excluded) and 1 (included).
+
+    """
+    import torch
+
+    if p <= 0.0 or p > 1.0:
+        raise ValueError(
+            f"`p` must be a floating point number between 0 (excluded) and 1 (included), got {p} instead."
+        )
+
+    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
+        sorted_logits, sorted_idx = torch.sort(logits, descending=False)
+        cumulative_probabilties = torch.nn.functional.softmax(
+            sorted_logits, dim=-1
+        ).cumsum(dim=-1)
+
+        sorted_masked_idx = cumulative_probabilties <= (1 - p)
+        mask_idx = torch.scatter(sorted_masked_idx, 1, sorted_idx, sorted_masked_idx)
+        return logits.masked_fill(mask_idx, -math.inf)
+
+    return logits_processor
+
+
+def rescale_logits(temperature: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
+    """Build a function that rescales the token probabilities exponentially.
+
+    Parameters
+    ----------
+    temperature
+        The value by which we rescale the logits.
+
+    """
+
+    if not isinstance(temperature, float) or temperature < 0.0:
+        raise ValueError(
+            f"`temperature` must be a strictly positive floating point number, got {temperature} instead."
+        )
+    elif temperature == 0.0:
+        raise ValueError(
+            "Please use the greedy sampler instead of setting the temperature to 0."
+        )
+
+    def logits_processor(logits: "torch.Tensor") -> "torch.Tensor":
+        return logits / temperature
+
+    return logits_processor
+
+
+class BeamSearchSampler:
+    """Beam Search sampling algorithm.
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence. Equivalent to the
+        number of beams.
+    """
+
+    def __init__(self, beams: int = 1):
+        self.samples = beams
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        _,
+    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
+        """Call the beam search sampler.
+
+        Parameters
+        ----------
+        next_token_logits
+            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
+            probability distribution of the next token over the vocabulary.
+        sequence_weights
+            A tensor of shape ``(n_seqs,)`` that represents the cumulative
+            weight of each sequence.
+        rng
+            A random number generator.
+
+        Returns
+        -------
+        A tuple with an array that contains the ids of the sampled tokens of
+        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
+        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
+        cumulative weights of each sequence of shape ``(n_seqs,)``.
+
+        """
+        import torch
+
+        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
+        weights = logprobs + sequence_weights.unsqueeze(1).expand_as(next_token_logits)
+
+        # Flatten scores to (n_batch, n_samples * vocab_size)
+        # and find the top-k weights for each batch.
+        batch_size = next_token_logits.shape[0] // self.samples
+        vocab_size = next_token_logits.shape[-1]
+        weights = weights.view(batch_size, self.samples * vocab_size)
+
+        # If the weights are all equal to 0 we are at the beginning of the search
+        # and thus only need to sample from one set of token logits for each
+        # batch.
+        if torch.all(sequence_weights == 0):
+            weights = weights[:, :vocab_size]
+
+        weights, indices = torch.topk(
+            weights, self.samples, dim=1, largest=True, sorted=True
+        )
+
+        ancestors = torch.div(indices, vocab_size, rounding_mode="floor")
+        next_token_ids = indices % vocab_size
+
+        # Re-shape the weights, next_token_ids and ancestors to (n_batch * n_samples, 1)
+        first_batch_idx = torch.arange(
+            0, batch_size * self.samples, self.samples, device=next_token_logits.device
+        ).unsqueeze(1)
+        ancestors = ancestors + first_batch_idx
+
+        ancestors = ancestors.view(self.samples * batch_size)
+        weights = weights.view(self.samples * batch_size)
+        next_token_ids = next_token_ids.view(self.samples * batch_size, 1)
+
+        return next_token_ids, ancestors, weights
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
+
+
+beam_search = BeamSearchSampler
diff --git a/outlines/serve/__init__.py b/outlines/serve/__init__.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/outlines/serve/serve.py b/outlines/serve/serve.py
new file mode 100644
index 000000000..61d3ed7af
--- /dev/null
+++ b/outlines/serve/serve.py
@@ -0,0 +1,140 @@
+#  _______________________________
+# / Don't want to self-host?      \
+# \ Try .json at http://dottxt.co /
+#  -------------------------------
+#        \   ^__^
+#         \  (oo)\_______
+#            (__)\       )\/\
+#                ||----w |
+#                ||     ||
+#
+#
+# Copyright 2024- the Outlines developers
+# Copyright 2023 the vLLM developers
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+
+#     http://www.apache.org/licenses/LICENSE-2.0
+
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import argparse
+import json
+from typing import AsyncGenerator
+
+import uvicorn
+from fastapi import FastAPI, Request
+from fastapi.responses import JSONResponse, Response, StreamingResponse
+from vllm.engine.arg_utils import AsyncEngineArgs
+from vllm.engine.async_llm_engine import AsyncLLMEngine
+from vllm.sampling_params import SamplingParams
+from vllm.utils import random_uuid
+
+from outlines.models.vllm import adapt_tokenizer
+from outlines.processors import JSONLogitsProcessor, RegexLogitsProcessor
+
+TIMEOUT_KEEP_ALIVE = 5  # seconds.
+TIMEOUT_TO_PREVENT_DEADLOCK = 1  # seconds.
+app = FastAPI()
+engine = None
+tokenizer = None
+
+
+@app.get("/health")
+async def health() -> Response:
+    """Health check."""
+    return Response(status_code=200)
+
+
+@app.post("/generate")
+async def generate(request: Request) -> Response:
+    """Generate completion for the request.
+
+    The request should be a JSON object with the following fields:
+    - prompt: the prompt to use for the generation.
+    - schema: the JSON schema to use for the generation (if regex is not provided).
+    - regex: the regex to use for the generation (if schema is not provided).
+    - stream: whether to stream the results or not.
+    - other fields: the sampling parameters (See `SamplingParams` for details).
+    """
+    assert engine is not None
+
+    request_dict = await request.json()
+    prompt = request_dict.pop("prompt")
+    stream = request_dict.pop("stream", False)
+
+    json_schema = request_dict.pop("schema", None)
+    regex_string = request_dict.pop("regex", None)
+    if json_schema is not None:
+        logits_processors = [JSONLogitsProcessor(json_schema, tokenizer)]
+    elif regex_string is not None:
+        logits_processors = [RegexLogitsProcessor(regex_string, tokenizer)]
+    else:
+        logits_processors = []
+
+    sampling_params = SamplingParams(
+        **request_dict,
+        logits_processors=logits_processors,  # type: ignore
+    )
+    request_id = random_uuid()
+
+    results_generator = engine.generate(prompt, sampling_params, request_id)  # type: ignore
+
+    # Streaming case
+    async def stream_results() -> AsyncGenerator[bytes, None]:
+        async for request_output in results_generator:
+            prompt = request_output.prompt
+            text_outputs = [prompt + output.text for output in request_output.outputs]
+            ret = {"text": text_outputs}
+            yield (json.dumps(ret) + "\0").encode("utf-8")
+
+    if stream:
+        return StreamingResponse(stream_results())
+
+    # Non-streaming case
+    final_output = None
+    async for request_output in results_generator:
+        if await request.is_disconnected():
+            # Abort the request if the client disconnects.
+            await engine.abort(request_id)  # type: ignore
+            return Response(status_code=499)
+        final_output = request_output
+
+    assert final_output is not None
+    prompt = final_output.prompt
+    text_outputs = [prompt + output.text for output in final_output.outputs]
+    ret = {"text": text_outputs}
+    return JSONResponse(ret)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--host", type=str, default=None)
+    parser.add_argument("--port", type=int, default=8000)
+    parser.add_argument("--ssl-keyfile", type=str, default=None)
+    parser.add_argument("--ssl-certfile", type=str, default=None)
+    parser = AsyncEngineArgs.add_cli_args(parser)
+    args = parser.parse_args()
+
+    # Adds the `engine_use_ray`,  `disable_log_requests` and `max_log_len`
+    # arguments
+    engine_args: AsyncEngineArgs = AsyncEngineArgs.from_cli_args(args)  # type: ignore
+
+    # Sets default for the model (`facebook/opt-125m`)
+    engine = AsyncLLMEngine.from_engine_args(engine_args)
+    tokenizer = adapt_tokenizer(tokenizer=engine.engine.tokenizer.tokenizer)
+
+    uvicorn.run(
+        app,
+        host=args.host,
+        port=args.port,
+        log_level="debug",
+        timeout_keep_alive=TIMEOUT_KEEP_ALIVE,
+        ssl_keyfile=args.ssl_keyfile,
+        ssl_certfile=args.ssl_certfile,
+    )
diff --git a/outlines/templates.py b/outlines/templates.py
index cd6a4fbf5..16588e813 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -1,6 +1,3 @@
-"""Create templates to easily build prompts."""
-
-import base64
 import functools
 import inspect
 import json
@@ -8,46 +5,12 @@
 import re
 import textwrap
 from dataclasses import dataclass
-from io import BytesIO
 from pathlib import Path
 from typing import Any, Callable, Dict, Optional, Type, cast
 import warnings
 
 import jinja2
-from PIL import Image
-from pydantic import BaseModel
-
-
-@dataclass
-class Vision:
-    """Contains the input for a vision model.
-
-    Provide an instance of this class as the `model_input` argument to a model
-    that supports vision.
-
-    Parameters
-    ----------
-    prompt
-        The prompt to use to generate the response.
-    image
-        The image to use to generate the response.
-
-    """
-    prompt: str
-    image: Image.Image
-
-    def __post_init__(self):
-        image = self.image
-
-        if not image.format:
-            raise TypeError(
-                "Could not read the format of the image passed to the model."
-            )
-
-        buffer = BytesIO()
-        image.save(buffer, format=image.format)
-        self.image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
-        self.image_format = f"image/{image.format.lower()}"
+import pydantic
 
 
 @dataclass
@@ -58,6 +21,7 @@ class Template:
     template can be accessed by callers.
 
     """
+
     template: jinja2.Template
     signature: Optional[inspect.Signature]
 
@@ -66,8 +30,7 @@ def __call__(self, *args, **kwargs) -> str:
 
         Returns
         -------
-        str
-            The rendered template as a Python string.
+        The rendered template as a Python ``str``.
 
         """
         if self.signature is not None:
@@ -79,8 +42,7 @@ def __call__(self, *args, **kwargs) -> str:
 
     @classmethod
     def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
-        """Create a `Template` instance from a string containing a Jinja
-        template.
+        """Create a `Template` instance from a string containing a Jinja template.
 
         Parameters
         ----------
@@ -89,20 +51,16 @@ def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
 
         Returns
         -------
-        Template
-            An instance of the class with the provided content as a template.
-
+        An instance of the class with the provided content as a template.
         """
         return cls(build_template_from_string(content, filters), None)
 
     @classmethod
     def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
-        """Create a `Template` instance from a file containing a Jinja
-        template.
+        """Create a `Template` instance from a file containing a Jinja template.
 
-        Note: This method does not allow to include and inheritance to
-        reference files that are outside the folder or subfolders of the file
-        given to `from_file`.
+        Note: This method does not allow to include and inheritance to reference files
+        that are outside the folder or subfolders of the file given to `from_file`.
 
         Parameters
         ----------
@@ -112,14 +70,10 @@ def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
         Returns
         -------
         Template
-            An instance of the Template class with the template loaded from the
-            file.
-
+            An instance of the Template class with the template loaded from the file.
         """
-        # We don't use a `Signature` here because it seems not feasible to
-        # infer one from a Jinja2 environment that is
-        # split across multiple files (since e.g. we support features like
-        # Jinja2 includes and template inheritance)
+        # We don't use a `Signature` here because it seems not feasible to infer one from a Jinja2 environment that is
+        # split across multiple files (since e.g. we support features like Jinja2 includes and template inheritance)
         return cls(build_template_from_file(path, filters), None)
 
 
@@ -249,12 +203,12 @@ def create_jinja_env(
     - `source`: get a function's source code
     - `signature`: get a function's signature
     - `args`: get a function's arguments
-    - `schema`: display a JSON Schema
+    - `schema`: isplay a JSON Schema
 
     Users may pass additional filters, and/or override existing ones.
 
-    Parameters
-    ----------
+    Arguments
+    ---------
     loader
        An optional `BaseLoader` instance
     filters
@@ -333,7 +287,7 @@ def get_fn_source(fn: Callable):
     re_search = re.search(re.compile(r"(\bdef\b.*)", re.DOTALL), source)
     if re_search is not None:
         source = re_search.group(0)
-    else:  # pragma: no cover
+    else:
         raise TypeError("Could not read the function's source code")
 
     return source
@@ -346,7 +300,7 @@ def get_fn_signature(fn: Callable):
 
     source = textwrap.dedent(inspect.getsource(fn))
     re_search = re.search(re.compile(r"\(([^)]+)\)"), source)
-    if re_search is None:  # pragma: no cover
+    if re_search is None:
         signature = ""
     else:
         signature = re_search.group(1)
@@ -367,9 +321,12 @@ def get_schema_dict(model: Dict):
     return json.dumps(model, indent=2)
 
 
-@get_schema.register(type(BaseModel))
-def get_schema_pydantic(model: Type[BaseModel]):
+@get_schema.register(type(pydantic.BaseModel))
+def get_schema_pydantic(model: Type[pydantic.BaseModel]):
     """Return the schema of a Pydantic model."""
+    if not isinstance(model, type(pydantic.BaseModel)):
+        raise TypeError("The `schema` filter only applies to Pydantic models.")
+
     if hasattr(model, "model_json_schema"):
         def_key = "$defs"
         raw_schema = model.model_json_schema()
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 85af9adc0..d7f40309a 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,40 +1,21 @@
-"""Output types for structured generation and regex DSL."""
-
-from dataclasses import dataclass, is_dataclass
-from enum import Enum, EnumMeta
-from typing import Union
-
-from jsonschema import Draft202012Validator as Validator
-from jsonschema.exceptions import SchemaError
-from pydantic import BaseModel, TypeAdapter
-from typing_extensions import _TypedDictMeta  # type: ignore
+from enum import Enum
 
 from . import airports, countries, locale
 from outlines.types.dsl import (
     Regex,
-    CFG,
-    FSM,
-    JsonSchema,
-    regex,
-    cfg,
-    fsm,
     json_schema,
-    optional,
+    regex,
     either,
+    optional,
     exactly,
     at_least,
     at_most,
     between,
-    zero_or_more,
     one_or_more,
-    # deprecated
-    repeat,
-    times,
+    zero_or_more,
 )
 
-
 # Python types
-string = Regex(r'"[^"]*"')
 integer = Regex(r"[+-]?(0|[1-9][0-9]*)")
 boolean = Regex("(True|False)")
 number = Regex(rf"{integer.pattern}(\.[0-9]+)?([eE][+-][0-9]+)?")
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 1a0a0ca1c..73de3876c 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,77 +1,12 @@
-"""Regular expression DSL and output types for structured generation.
-
-This module contains elements related to three logical steps in the use of
-output types for structured generation:
-
-1. Definition of `Term` classes that contain output type definitions. That
-   includes both terms intended to be used by themselves such as `JsonSchema`
-   or `CFG` and terms that are part of the regular expression DSL such as
-   `Alternatives` or `KleeneStar` (and the related functions).
-2. Conversion of Python types into `Term` instances (`python_types_to_terms`).
-3. Conversion of a `Term` instance into a regular expression (`to_regex`).
-
-"""
-
-import json
+import json as json
 import re
-import sys
-import warnings
 from dataclasses import dataclass
-from enum import EnumMeta
-from types import FunctionType
-from typing import (
-    Any,
-    List,
-    Optional as OptionalType,
-    Union,
-    get_args,
-)
-import interegular
-import jsonschema
-from genson import SchemaBuilder
-from outlines_core.fsm.json_schema import build_regex_from_schema
-from pydantic import (
-    BaseModel,
-    GetCoreSchemaHandler,
-    GetJsonSchemaHandler,
-    TypeAdapter,
-)
+from typing import Any, List, Union
+
+from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler
 from pydantic.json_schema import JsonSchemaValue
 from pydantic_core import core_schema as cs
-
-import outlines.types as types
-from outlines import grammars
-from outlines.types.utils import (
-    get_schema_from_signature,
-    is_int,
-    is_int_instance,
-    is_float,
-    is_float_instance,
-    is_str,
-    is_str_instance,
-    is_bool,
-    is_datetime,
-    is_date,
-    is_time,
-    is_native_dict,
-    is_dict_instance,
-    is_dataclass,
-    is_typed_dict,
-    is_pydantic_model,
-    is_genson_schema_builder,
-    is_literal,
-    is_union,
-    is_enum,
-    is_callable,
-    is_typing_list,
-    is_typing_tuple,
-    is_typing_dict,
-)
-
-if sys.version_info >= (3, 12):  # pragma: no cover
-    from typing import _TypedDictMeta  # type: ignore
-else:  # pragma: no cover
-    from typing_extensions import _TypedDictMeta  # type: ignore
+from outlines_core.fsm.json_schema import build_regex_from_schema
 
 
 class Term:
@@ -96,30 +31,18 @@ class also handles validation.
 
     """
 
-    def __add__(self: "Term", other: "Term") -> "Sequence":
-        if is_str_instance(other):
-            other = String(str(other))
+    def __add__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
+        if isinstance(other, str):
+            other = String(other)
 
         return Sequence([self, other])
 
-    def __radd__(self: "Term", other: "Term") -> "Sequence":
-        if is_str_instance(other):
-            other = String(str(other))
+    def __radd__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
+        if isinstance(other, str):
+            other = String(other)
 
         return Sequence([other, self])
 
-    def __or__(self: "Term", other: "Term") -> "Alternatives":
-        if is_str_instance(other):
-            other = String(str(other))
-
-        return Alternatives([self, other])
-
-    def __ror__(self: "Term", other: "Term") -> "Alternatives":
-        if is_str_instance(other):
-            other = String(str(other))
-
-        return Alternatives([other, self])
-
     def __get_validator__(self, _core_schema):
         def validate(input_value):
             return self.validate(input_value)
@@ -201,16 +124,6 @@ def one_or_more(self) -> "KleenePlus":
     def zero_or_more(self) -> "KleeneStar":
         return zero_or_more(self)
 
-    # deprecated
-    def times(self, count: int) -> "QuantifyExact":
-        return times(self, count)
-
-    # deprecated
-    def repeat(self, min_count: int, max_count: int) -> Union[
-        "QuantifyMinimum", "QuantifyMaximum", "QuantifyBetween"
-    ]:
-        return repeat(self, min_count, max_count)
-
 
 @dataclass
 class String(Term):
@@ -225,14 +138,6 @@ def __repr__(self):
 
 @dataclass
 class Regex(Term):
-    """Class representing a regular expression.
-
-    Parameters
-    ----------
-    pattern
-        The regular expression as a string.
-
-    """
     pattern: str
 
     def _display_node(self) -> str:
@@ -242,119 +147,22 @@ def __repr__(self):
         return f"Regex(pattern='{self.pattern}')"
 
 
-@dataclass
-class CFG(Term):
-    """Class representing a context-free grammar.
-
-    Parameters
-    ----------
-    definition
-        The definition of the context-free grammar as a string.
-
-    """
-    definition: str
-
-    def _display_node(self) -> str:
-        return f"CFG('{self.definition}')"
-
-    def __repr__(self):
-        return f"CFG(definition='{self.definition}')"
-
-    def __eq__(self, other):
-        if not isinstance(other, CFG):
-            return False
-        return self.definition == other.definition
-
-    @classmethod
-    def from_file(cls, path: str) -> "CFG":
-        """Create a CFG instance from a file containing a CFG definition.
-
-        Parameters
-        ----------
-        path : str
-            The path to the file containing the CFG definition.
-        Returns
-        -------
-        CFG
-            A CFG instance.
-
-        """
-        with open(path, "r") as f:
-            definition = f.read()
-        return cls(definition)
-
-
-@dataclass
-class FSM(Term):
-    """Class representing a finite state machine.
-
-    Parameters
-    ----------
-    fsm
-        The finite state machine to store. This object must be an instance of
-        `interegular.fsm.FSM`.
-
-    """
-    fsm: interegular.fsm.FSM
-
-    def _display_node(self) -> str:
-        return f"FSM({self.fsm.__repr__()})"
-
-    def __repr__(self):
-        return f"FSM(fsm={self.fsm.__repr__()})"
-
-
 class JsonSchema(Term):
-    """Class representing a JSON schema.
-
-    The JSON schema object from which to instantiate the class can be a
-    dictionary, a string, a Pydantic model, a typed dict, a dataclass, or a
-    genSON schema builder.
-
-    """
-    def __init__(
-        self,
-        schema: Union[
-            dict, str, type[BaseModel], _TypedDictMeta, type, SchemaBuilder
-        ],
-        whitespace_pattern: OptionalType[str] = None,
-    ):
-        """
-        Parameters
-        ----------
-        schema
-            The object containing the JSON schema.
-        whitespace_pattern
-            The pattern to use to match whitespace characters.
-
-        """
-        schema_str: str
-
-        if is_dict_instance(schema):
+    def __init__(self, schema: Union[dict, str, type[BaseModel]]):
+        if isinstance(schema, dict):
             schema_str = json.dumps(schema)
-        elif is_str_instance(schema):
-            schema_str = str(schema)
-        elif is_pydantic_model(schema):
-            schema_str = json.dumps(schema.model_json_schema())  # type: ignore
-        elif is_typed_dict(schema):
-            schema_str = json.dumps(TypeAdapter(schema).json_schema())
-        elif is_dataclass(schema):
-            schema_str = json.dumps(TypeAdapter(schema).json_schema())
-        elif is_genson_schema_builder(schema):
-            schema_str = schema.to_json()  # type: ignore
+        elif isinstance(schema, str):
+            schema_str = schema
+        elif issubclass(schema, BaseModel):
+            schema_str = json.dumps(schema.model_json_schema())
         else:
             raise ValueError(
-                f"Cannot parse schema {schema}. The schema must be either "
-                + "a Pydantic class, typed dict, a dataclass, a genSON schema "
-                + "builder or a string or dict that contains the JSON schema "
-                + "specification"
+                f"Cannot parse schema {json_schema}. The schema must be either "
+                + "a Pydantic class, a dictionary or a string that contains the JSON "
+                + "schema specification"
             )
 
         self.schema = schema_str
-        self.whitespace_pattern = whitespace_pattern
-
-    def __post_init__(self):
-        jsonschema.Draft7Validator.check_schema(json.loads(self.schema))
 
     def _display_node(self) -> str:
         return f"JsonSchema('{self.schema}')"
@@ -362,35 +170,6 @@ def _display_node(self) -> str:
     def __repr__(self):
         return f"JsonSchema(schema='{self.schema}')"
 
-    def __eq__(self, other):
-        if not isinstance(other, JsonSchema):
-            return False
-        try:
-            self_dict = json.loads(self.schema)
-            other_dict = json.loads(other.schema)
-            return self_dict == other_dict
-        except json.JSONDecodeError:  # pragma: no cover
-            return self.schema == other.schema
-
-    @classmethod
-    def from_file(cls, path: str) -> "JsonSchema":
-        """Create a JsonSchema instance from a .json file containing a JSON
-        schema.
-
-        Parameters
-        ----------
-        path:
-            The path to the file containing the JSON schema.
-        Returns
-        -------
-        JsonSchema
-            A JsonSchema instance.
-
-        """
-        with open(path, "r") as f:
-            schema = json.load(f)
-        return cls(schema)
-
 
 @dataclass
 class KleeneStar(Term):
@@ -543,14 +322,6 @@ def regex(pattern: str):
     return Regex(pattern)
 
 
-def cfg(definition: str):
-    return CFG(definition)
-
-
-def fsm(fsm: interegular.fsm.FSM):
-    return FSM(fsm)
-
-
 def json_schema(schema: Union[str, dict, type[BaseModel]]):
     return JsonSchema(schema)
 
@@ -560,7 +331,6 @@ def either(*terms: Union[str, Term]):
 
     This factory function automatically translates string arguments
     into `String` objects.
-
     """
     terms = [String(arg) if isinstance(arg, str) else arg for arg in terms]
     return Alternatives(terms)
@@ -604,306 +374,11 @@ def one_or_more(term: Union[Term, str]) -> KleenePlus:
     return KleenePlus(term)
 
 
-def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
-    """Convert Python types to Outlines DSL terms that constrain LLM output.
-
-    Parameters
-    ----------
-    ptype
-        The Python type to convert
-    recursion_depth
-        Current recursion depth to prevent infinite recursion
-
-    Returns
-    -------
-    Term
-        The corresponding DSL `Term` instance.
-
-    """
-    if recursion_depth > 10:
-        raise RecursionError(
-            f"Maximum recursion depth exceeded when converting {ptype}. "
-            "This might be due to a recursive type definition."
-        )
-
-    # First handle Term instances
-    if isinstance(ptype, Term):
-        return ptype
-
-    # Basic types
-    if is_int(ptype):
-        return types.integer
-    elif is_float(ptype):
-        return types.number
-    elif is_bool(ptype):
-        return types.boolean
-    elif is_str(ptype):
-        return types.string
-    elif is_native_dict(ptype):
-        return CFG(grammars.json)
-    elif is_time(ptype):
-        return types.time
-    elif is_date(ptype):
-        return types.date
-    elif is_datetime(ptype):
-        return types.datetime
-
-    # Basic type instances
-    if is_str_instance(ptype):
-        return String(ptype)
-    elif is_int_instance(ptype) or is_float_instance(ptype):
-        return Regex(str(ptype))
-
-    # Structured types
-    structured_type_checks = [
-        lambda x: is_dataclass(x),
-        lambda x: is_typed_dict(x),
-        lambda x: is_pydantic_model(x),
-    ]
-    if any(check(ptype) for check in structured_type_checks):
-        schema = TypeAdapter(ptype).json_schema()
-        return JsonSchema(schema)
-
-    elif is_genson_schema_builder(ptype):
-        schema = ptype.to_json()
-        return JsonSchema(schema)
-
-    if is_enum(ptype):
-        return Alternatives(
-            [
-                python_types_to_terms(member, recursion_depth + 1)
-                for member in _get_enum_members(ptype)
-            ]
-        )
-
-    args = get_args(ptype)
-    if is_literal(ptype):
-        return _handle_literal(args)
-    elif is_union(ptype):
-        return _handle_union(args, recursion_depth)
-    elif is_typing_list(ptype):
-        return _handle_list(args, recursion_depth)
-    elif is_typing_tuple(ptype):
-        return _handle_tuple(args, recursion_depth)
-    elif is_typing_dict(ptype):
-        return _handle_dict(args, recursion_depth)
-
-    if is_callable(ptype):
-        return JsonSchema(get_schema_from_signature(ptype))
-
-    type_name = getattr(ptype, "__name__", ptype)
-    raise TypeError(
-        f"Type {type_name} is currently not supported. Please open an issue: "
-        "https://github.com/dottxt-ai/outlines/issues"
-    )
-
-
-def _get_enum_members(ptype: EnumMeta) -> List[Any]:
-    regular_members = [member.value for member in ptype]  # type: ignore
-    function_members = []
-    for key, value in ptype.__dict__.items():
-        if (
-            isinstance(value, FunctionType)
-            and not (key.startswith('__') and key.endswith('__'))
-            and key != '_generate_next_value_'  # Skip this specific method that causes issues
-        ):
-            function_members.append(value)
-    return regular_members + function_members
-
-
-def _handle_literal(args: tuple) -> Alternatives:
-    return Alternatives([python_types_to_terms(arg) for arg in args])
-
-
-def _handle_union(args: tuple, recursion_depth: int) -> Alternatives:
-    # Handle the Optional[T] type
-    if len(args) == 2 and (type(None) in args or None in args):
-        other_ptype = next(arg for arg in args if arg not in (type(None), None))
-        return Alternatives(
-            [
-                python_types_to_terms(other_ptype, recursion_depth + 1),
-                String("None"),
-            ]
-        )
-    return Alternatives(
-        [python_types_to_terms(arg, recursion_depth + 1) for arg in args]
-    )
-
-
-def _handle_list(args: tuple, recursion_depth: int) -> Sequence:
-    if args is None or len(args) > 1:
-        raise TypeError(
-            f"Only homogeneous lists are supported. Got multiple type arguments {args}."
-        )
-    item_type = python_types_to_terms(args[0], recursion_depth + 1)
-    return Sequence(
-        [
-            String("["),
-            item_type,
-            KleeneStar(Sequence([String(", "), item_type])),
-            String("]"),
-        ]
-    )
-
-
-def _handle_tuple(args: tuple, recursion_depth: int) -> Union[Sequence, String]:
-    if len(args) == 0 or args == ((),):
-        return String("()")
-    elif len(args) == 2 and args[1] is Ellipsis:
-        item_term = python_types_to_terms(args[0], recursion_depth + 1)
-        return Sequence(
-            [
-                String("("),
-                item_term,
-                KleeneStar(Sequence([String(", "), item_term])),
-                String(")"),
-            ]
-        )
-    else:
-        items = [python_types_to_terms(arg, recursion_depth + 1) for arg in args]
-        separator = String(", ")
-        elements = []
-        for i, item in enumerate(items):
-            elements.append(item)
-            if i < len(items) - 1:
-                elements.append(separator)
-        return Sequence([String("("), *elements, String(")")])
-
-
-def _handle_dict(args: tuple, recursion_depth: int) -> Sequence:
-    if args is None or len(args) != 2:
-        raise TypeError(f"Dict must have exactly two type arguments. Got {args}.")
-    # Add dict support with key:value pairs
-    key_type = python_types_to_terms(args[0], recursion_depth + 1)
-    value_type = python_types_to_terms(args[1], recursion_depth + 1)
-    return Sequence(
-        [
-            String("{"),
-            Optional(
-                Sequence(
-                    [
-                        key_type,
-                        String(":"),
-                        value_type,
-                        KleeneStar(
-                            Sequence([String(", "), key_type, String(":"), value_type])
-                        ),
-                    ]
-                )
-            ),
-            String("}"),
-        ]
-    )
-
-
-# deprecated
-def repeat(term: Term, min_count: int, max_count: int) -> Union[
-    QuantifyMinimum, QuantifyMaximum, QuantifyBetween
-]:
-    if min_count is None and max_count is None:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `between`, `at_least` or `at_most` instead.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        raise ValueError(
-            "You must provide a value for at least `min_count` or `max_count`"
-        )
-    if max_count is None:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `at_least` instead.
-            For instance:
-            ```python
-            from outlines.types import Regex, at_least
-            digit = Regex(r'[0-9]')
-            # 2 ways of doing the same thing
-            at_least_5_digits = digit.at_least(5)
-            at_least_5_digits = at_least(digit, 5)
-            ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        return at_least(min_count, term)
-    if min_count is None:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `at_most` instead.
-            For instance:
-            ```python
-            from outlines.types import Regex, at_most
-            digit = Regex(r'[0-9]')
-            # 2 ways of doing the same thing
-            at_most_5_digits = digit.at_most(5)
-            at_most_5_digits = at_most(digit, 5)
-            ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        return at_most(max_count, term)
-    else:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `between` instead.
-            For instance:
-            ```python
-            from outlines.types import Regex, between
-            digit = Regex(r'[0-9]')
-            # 2 ways of doing the same thing
-            five_to_eight_digits = digit.between(5, 8)
-            five_to_eight_digits = between(digit, 5, 8)
-            ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        return between(min_count, max_count, term)
-
-
-# deprecated
-def times(term: Term, count: int) -> QuantifyExact:
-    warnings.warn("""
-        The `times` function/method is deprecated starting from v0.2.2.
-        Do not use it. Support for it will be removed in v1.5.0.
-        Use `exactly` instead.
-        For instance:
-        ```python
-        from outlines.types import Regex, exactly
-        digit = Regex(r'[0-9]')
-        # 2 ways of doing the same thing
-        five_digits = digit.exactly(5)
-        five_digits = exactly(digit, 5)
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-    return exactly(count, term)
-
-
 def to_regex(term: Term) -> str:
     """Convert a term to a regular expression.
 
     We only consider self-contained terms that do not refer to another rule.
 
-    Parameters
-    ----------
-    term
-        The term to convert to a regular expression.
-
-    Returns
-    -------
-    str
-        The regular expression as a string.
-
     """
     match term:
         case String():
@@ -911,7 +386,7 @@ def to_regex(term: Term) -> str:
         case Regex():
             return f"({term.pattern})"
         case JsonSchema():
-            regex_str = build_regex_from_schema(term.schema, term.whitespace_pattern)
+            regex_str = build_regex_from_schema(term.schema)
             return f"({regex_str})"
         case KleeneStar():
             return f"({to_regex(term.term)})*"
diff --git a/outlines/types/locale/__init__.py b/outlines/types/locale/__init__.py
index 97c9abff0..511631d84 100644
--- a/outlines/types/locale/__init__.py
+++ b/outlines/types/locale/__init__.py
@@ -1,3 +1 @@
-"""Locale-specific regex patterns."""
-
 from . import us
diff --git a/outlines/types/locale/us.py b/outlines/types/locale/us.py
index 275aa492e..0bda82d44 100644
--- a/outlines/types/locale/us.py
+++ b/outlines/types/locale/us.py
@@ -1,5 +1,3 @@
-"""Locale-specific regex patterns for the United States."""
-
 from outlines.types.dsl import Regex
 
 zip_code = Regex(r"\d{5}(?:-\d{4})?")
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
deleted file mode 100644
index 64537643e..000000000
--- a/outlines/types/utils.py
+++ /dev/null
@@ -1,206 +0,0 @@
-"""Utility functions for the types module."""
-
-import dataclasses
-import datetime
-import inspect
-import sys
-import warnings
-from enum import Enum, EnumMeta
-from typing import (
-    Annotated,
-    Any,
-    Callable,
-    Dict,
-    Literal,
-    List,
-    NewType,
-    Tuple,
-    Union,
-    get_args,
-    get_origin,
-)
-
-import interegular
-from genson import SchemaBuilder
-from pydantic import BaseModel, create_model
-
-if sys.version_info >= (3, 12): # pragma: no cover
-    from typing import _TypedDictMeta  # type: ignore
-else: # pragma: no cover
-    from typing_extensions import _TypedDictMeta  # type: ignore
-
-
-# Type identification
-
-
-def is_int(value: Any) -> bool:
-    return (
-        value is int
-        or get_origin(value) is int
-        or (get_origin(value) is Annotated and get_args(value)[0] is int)
-        or (hasattr(value, "__supertype__") and value.__supertype__ is int)
-    )
-
-
-def is_int_instance(value: Any) -> bool:
-    return isinstance(value, int)
-
-
-def is_float(value: Any) -> bool:
-    return (
-        value is float
-        or get_origin(value) is float
-        or (get_origin(value) is Annotated and get_args(value)[0] is float)
-        or (hasattr(value, "__supertype__") and value.__supertype__ is float)
-    )
-
-
-def is_float_instance(value: Any) -> bool:
-    return isinstance(value, float)
-
-
-def is_str(value: Any) -> bool:
-    return (
-        value is str
-        or get_origin(value) is str
-        or (get_origin(value) is Annotated and get_args(value)[0] is str)
-        or (hasattr(value, "__supertype__") and value.__supertype__ is str)
-    )
-
-
-def is_str_instance(value: Any) -> bool:
-    return isinstance(value, str)
-
-
-def is_bool(value: Any) -> bool:
-    return (
-        value is bool
-        or get_origin(value) is bool
-        or (get_origin(value) is Annotated and get_args(value)[0] is bool)
-        or (hasattr(value, "__supertype__") and value.__supertype__ is bool)
-    )
-
-
-def is_dict_instance(value: Any) -> bool:
-    return isinstance(value, dict)
-
-
-def is_datetime(value: Any) -> bool:
-    return value is datetime.datetime or get_origin(value) is datetime.datetime
-
-
-def is_date(value: Any) -> bool:
-    return value is datetime.date or get_origin(value) is datetime.date
-
-
-def is_time(value: Any) -> bool:
-    return value is datetime.time or get_origin(value) is datetime.time
-
-
-def is_native_dict(value: Any) -> bool:
-    return value is dict
-
-
-def is_typing_dict(value: Any) -> bool:
-    return get_origin(value) is dict
-
-
-def is_typing_list(value: Any) -> bool:
-    return get_origin(value) is list
-
-
-def is_typing_tuple(value: Any) -> bool:
-    return get_origin(value) is tuple
-
-
-def is_union(value: Any) -> bool:
-    return get_origin(value) is Union
-
-
-def is_literal(value: Any) -> bool:
-    return get_origin(value) is Literal
-
-
-def is_dataclass(value: Any) -> bool:
-    return isinstance(value, type) and dataclasses.is_dataclass(value)
-
-
-def is_typed_dict(value: Any) -> bool:
-    return isinstance(value, _TypedDictMeta)
-
-
-def is_pydantic_model(value):
-    # needed because generic type cannot be used with `issubclass`    # for Python versions < 3.11
-    if get_origin(value) is not None:
-        return False
-
-    return isinstance(value, type) and issubclass(value, BaseModel)
-
-
-def is_genson_schema_builder(value: Any) -> bool:
-    return isinstance(value, SchemaBuilder)
-
-
-def is_enum(value: Any) -> bool:
-    return isinstance(value, EnumMeta)
-
-
-def is_callable(value: Any) -> bool:
-    return callable(value) and not isinstance(value, type)
-
-
-def is_interegular_fsm(value: Any) -> bool:
-    return isinstance(value, interegular.fsm.FSM)
-
-
-# Type conversion
-
-
-def get_enum_from_literal(value) -> Enum:
-    return Enum(
-        value.__name__,
-        {str(arg): arg for arg in get_args(value)}
-    )
-
-
-def get_schema_from_signature(fn: Callable) -> dict:
-    """Turn a function signature into a JSON schema.
-
-    Every JSON object valid to the output JSON Schema can be passed
-    to `fn` using the ** unpacking syntax.
-
-    """
-    signature = inspect.signature(fn)
-    arguments = {}
-    for name, arg in signature.parameters.items():
-        if arg.annotation == inspect._empty:
-            raise ValueError("Each argument must have a type annotation")
-        else:
-            arguments[name] = (arg.annotation, ...)
-
-    try:
-        fn_name = fn.__name__
-    except Exception as e:
-        fn_name = "Arguments"
-        warnings.warn(
-            f"The function name could not be determined. Using default name 'Arguments' instead. For debugging, here is exact error:\n{e}",
-            category=UserWarning,
-        )
-    model = create_model(fn_name, **arguments)
-
-    return model.model_json_schema()
-
-
-def get_schema_from_enum(myenum: type[Enum]) -> dict:
-    if len(myenum) == 0:
-        raise ValueError(
-            f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
-        )
-    choices = [
-        get_schema_from_signature(elt.value.func)
-        if callable(elt.value)
-        else {"const": elt.value}
-        for elt in myenum
-    ]
-    schema = {"title": myenum.__name__, "oneOf": choices}
-    return schema
diff --git a/outlines/v0_legacy/generate/api.py b/outlines/v0_legacy/generate/api.py
deleted file mode 100644
index 625bac105..000000000
--- a/outlines/v0_legacy/generate/api.py
+++ /dev/null
@@ -1,175 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-from dataclasses import asdict
-from typing import Any, Dict, List, Optional, Union
-
-from outlines.generator import Generator
-from outlines.v0_legacy.models.exllamav2 import ExLlamaV2Model, exllamav2_params_adapter
-from outlines.v0_legacy.models.transformers import transformers_params_adapter
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
-from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
-from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
-from outlines.models import (
-    LlamaCpp,
-    MLXLM,
-    Transformers,
-    VLLMOffline,
-)
-
-V0GeneratorModel = Union[
-    ExLlamaV2Model,
-    LlamaCpp,
-    MLXLM,
-    Transformers,
-    TransformersVision,
-    VLLMOffline,
-]
-
-
-class GeneratorV0Adapter:
-    """Object used to bridge the gap between the legacy `generate` functions
-    and the current `Generator` object.
-
-    The legacy `generate` functions return an instance of this class. When
-    initialized, it stores an instance of the current `Generator` object
-    and calls it when it's called.
-    """
-
-    def __init__(self, model, output_type, sampler):
-        if not isinstance(model, V0GeneratorModel):  # type: ignore
-            raise ValueError(
-                "You can only use the v0 API with models that were already"
-                + f"available in v0. Got {type(model)}."
-            )
-        self.model = model
-        self.generator = Generator(model, output_type)
-        self.sampling_params = asdict(sampler.sampling_params)
-
-    def format_sequence(self, sequence: str):
-        return sequence
-
-    def create_inference_params(
-        self, max_tokens, stop_at, seed, **model_specific_params
-    ):
-        """Turn the arguments received by the generator into arguments
-        for the model associated to it.
-        """
-        generation_params = {
-            "max_tokens": max_tokens,
-            "stop_at": stop_at,
-            "seed": seed,
-        }
-
-        if isinstance(self.model, LlamaCpp):
-            adapter_function = llamacpp_params_adapter
-        elif isinstance(self.model, VLLMOffline): # pragma: no cover
-            adapter_function = vllm_offline_params_adapter
-        elif isinstance(self.model, Transformers):
-            adapter_function = transformers_params_adapter
-        elif isinstance(self.model, ExLlamaV2Model): # pragma: no cover
-            adapter_function = exllamav2_params_adapter
-        elif isinstance(self.model, MLXLM): # pragma: no cover
-            adapter_function = mlxlm_params_adapter
-
-        inference_params = adapter_function(
-            self.sampling_params,
-            generation_params,
-            model_specific_params
-        )
-        return inference_params
-
-    def __call__(
-        self,
-        prompts: Union[str, List[str], Dict[str, Any]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        result = self.generator(
-            prompts,
-            **self.create_inference_params(
-                max_tokens, stop_at, seed, **model_specific_params
-            )
-        )
-        if isinstance(result, list):
-            return [self.format_sequence(r) for r in result]
-        else:
-            return self.format_sequence(result)
-
-    def stream(
-        self,
-        prompts: Union[str, List[str], Dict[str, Any]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        return self.generator.stream(
-            prompts,
-            **self.create_inference_params(
-                max_tokens, stop_at, seed, **model_specific_params
-            )
-        )
-
-
-class GeneratorVisionV0Adapter(GeneratorV0Adapter):
-    """This object does the same as `GeneratorV0Adapter` but for the
-    `TransformersVision` model. The only difference between the two is that
-    it receives a media argument in addition to the prompts.
-    """
-
-    @staticmethod
-    def merge_prompts_and_media(
-        prompts: Union[str, List[str]],
-        media: Union[Any, List[Any]],
-    ) -> Dict[str, Any]:
-        """Merge prompts and media into a single dictionary
-
-        This function is needed as the `TransformersMultiModal` model requires
-        the prompt and assets to be contained in a single argument while the
-        legacy `TransformersVision` model requires separate arguments for the
-        prompt and assets.
-        """
-        return {
-            "text": prompts,
-            "images": media,
-        }
-
-    def __call__(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        merged_input = self.merge_prompts_and_media(prompts, media)
-        return super().__call__(
-            merged_input,
-            max_tokens,
-            stop_at,
-            seed,
-            **model_specific_params
-        )
-
-    def stream(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        merged_input = self.merge_prompts_and_media(prompts, media)
-        return super().stream(
-            merged_input,
-            max_tokens,
-            stop_at,
-            seed,
-            **model_specific_params
-        )
diff --git a/outlines/v0_legacy/generate/cfg.py b/outlines/v0_legacy/generate/cfg.py
deleted file mode 100644
index 0f33b6efb..000000000
--- a/outlines/v0_legacy/generate/cfg.py
+++ /dev/null
@@ -1,80 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models.llamacpp import LlamaCpp
-from outlines.models.openai import OpenAI
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.types.dsl import CFG
-
-
-def cfg(
-    model, cfg_str: str, sampler: Sampler = multinomial()
-) -> GeneratorV0Adapter:
-    """Generate text in the language of a Context-Free Grammar
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    from outlines.types import CFG
-    cfg_str = "..."  # Context-Free Grammar as a string
-    generator = Generator(model, CFG(cfg_str))
-    ```
-
-    You can then call the generator created with a prompt to generate text
-    that matches the Context-Free Grammar.
-
-    Arguments
-    ---------
-    model:
-        The Outlines v0 model to use to generate text.
-    cfg_str:
-        The Context-Free Grammar to use to generate text.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-
-    warnings.warn("""
-        The `cfg` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import CFG
-        cfg_str = '...'  # Context-Free Grammar as a string
-        generator = Generator(model, CFG(cfg_str))
-        ```
-        You can then call the generator created with a prompt to generate
-        text in the language of a Context-Free Grammar.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use CFG-based structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API."
-        )
-    elif isinstance(model, LlamaCpp):
-        raise NotImplementedError(
-            "Not yet available due to bug in llama_cpp tokenizer"
-        )
-    elif isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, CFG(cfg_str), sampler)
-    return GeneratorV0Adapter(model, CFG(cfg_str), sampler)
diff --git a/outlines/v0_legacy/generate/choice.py b/outlines/v0_legacy/generate/choice.py
deleted file mode 100644
index eff2fcffd..000000000
--- a/outlines/v0_legacy/generate/choice.py
+++ /dev/null
@@ -1,126 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import re
-import json as pyjson
-import warnings
-from enum import Enum
-from typing import Callable, List, Union
-
-from outlines.models.openai import OpenAI
-from outlines.types.utils import get_schema_from_enum
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
-from outlines.v0_legacy.generate.json import json
-from outlines.v0_legacy.generate.regex import regex
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-
-def choice(
-    model,
-    choices: Union[List[str], type[Enum]],
-    sampler: Sampler = multinomial(),
-) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter, Callable]:
-    """Generate a choice from a list of options.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from typing import Literal
-    from outlines import Generator
-    generator = Generator(model, Literal["foo", "bar"])
-    ```
-
-    You can then call the generator created with a prompt to generate a
-    choice from the list of options.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    choices:
-        A list of options to choose from.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate a choice from the list of options.
-
-    """
-    warnings.warn("""
-        The `choice` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from typing import Literal
-        generator = Generator(model, Literal['foo', 'bar'])
-        ```
-        You can then call the generator created with a prompt to generate
-        a choice from the list of options.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        if isinstance(choices, list):
-            return choice_openai(model, choices, sampler)
-        raise ValueError(
-            "The `choice` function with OpenAI only supports a list of "
-            + "strings as choices."
-        )
-
-    if isinstance(choices, type(Enum)):
-        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
-    else:
-        choices = [re.escape(choice) for choice in choices]  # type: ignore
-        regex_str = r"(" + r"|".join(choices) + r")"
-
-    # We do not want to raise the deprecation warning through the regex function
-    with warnings.catch_warnings():
-        warnings.filterwarnings(
-            "ignore",
-            category=DeprecationWarning,
-        )
-        generator = regex(model, regex_str, sampler)
-
-    if isinstance(choices, type(Enum)):
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    else:
-        generator.format_sequence = lambda x: x
-
-    return generator
-
-
-def choice_openai(
-    model,
-    choices: List[str],
-    sampler: Sampler = multinomial(),
-) -> Callable:
-
-    choices_schema = pyjson.dumps(
-        {
-            "type": "object",
-            "properties": {"result": {"type": "string", "enum": choices}},
-            "additionalProperties": False,
-            "required": ["result"],
-        }
-    )
-
-    # We do not want to raise the deprecation warning through the json function
-    with warnings.catch_warnings():
-        warnings.filterwarnings(
-            "ignore",
-            category=DeprecationWarning
-        )
-        generator = json(model, choices_schema, sampler)
-
-    def generate_choice(*args, **kwargs): # pragma: no cover
-        return generator(*args, **kwargs)["result"]
-
-    return generate_choice
diff --git a/outlines/v0_legacy/generate/format.py b/outlines/v0_legacy/generate/format.py
deleted file mode 100644
index cd574ac4e..000000000
--- a/outlines/v0_legacy/generate/format.py
+++ /dev/null
@@ -1,70 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models.openai import OpenAI
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.samplers import Sampler, multinomial
-
-
-def format(
-    model, python_type, sampler: Sampler = multinomial()
-) -> GeneratorV0Adapter:
-    """Generate structured data that can be parsed as a Python type.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead::
-
-    ```python
-    from outlines import Generator
-    generator = Generator(model, int)
-    ```
-
-    You can then call the generator created with a prompt to generate a
-    integer.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    python_type:
-        A Python type. The output of the generator must be parseable into
-        this type.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `SequenceGenerator` instance that generates text constrained by the Python type
-    and translates this text into the corresponding type.
-
-    """
-    warnings.warn("""
-        The `format` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        generator = Generator(model, int)
-        ```
-        You can then call the generator created with a prompt to generate
-        an integer.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use Python type-structured generation with an OpenAI model"
-            + " due to the limitations of the OpenAI API."
-        )
-    elif isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, python_type, sampler)
-    return GeneratorV0Adapter(model, python_type, sampler)
diff --git a/outlines/v0_legacy/generate/fsm.py b/outlines/v0_legacy/generate/fsm.py
deleted file mode 100644
index 23a5477fe..000000000
--- a/outlines/v0_legacy/generate/fsm.py
+++ /dev/null
@@ -1,77 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-import interegular
-
-from outlines.models.openai import OpenAI
-from outlines.types import FSM
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-
-
-def fsm(
-    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
-) -> GeneratorV0Adapter:
-    """Generate text that follows a Finite State Machine.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    import interegular
-    from outlines import Generator
-    from outlines.types import FSM
-    fsm = FSM(interegular.parse_pattern(r"[a-z]+").to_fsm())
-    generator = Generator(model, fsm)
-    ```
-
-    You can then call the generator created with a prompt to generate text
-    that follows the Finite State Machine.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    fsm:
-        The interregular `FSM` to use to generate text.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text that follows the Finite State Machine.
-
-    """
-    warnings.warn("""
-        The `fsm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import FSM
-        fsm_str = '...'  # Finite State Machine as a string
-        generator = Generator(model, FSM(fsm_str))
-        ```
-        You can then call the generator created with a prompt to generate
-        text that follows the Finite State Machine.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use FSM-based generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API."
-        )
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, FSM(fsm), sampler)
-    return GeneratorV0Adapter(model, FSM(fsm), sampler)
diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
deleted file mode 100644
index 6260b6471..000000000
--- a/outlines/v0_legacy/generate/json.py
+++ /dev/null
@@ -1,145 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import json as pyjson
-import warnings
-from typing import Callable, Optional, Union
-
-from pydantic import BaseModel
-
-from outlines.models.openai import OpenAI
-from outlines.types.dsl import JsonSchema
-from outlines.types.utils import get_schema_from_signature, is_callable
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-
-
-def json(
-    model,
-    schema_object: Union[str, object, Callable],
-    sampler: Sampler = multinomial(),
-    whitespace_pattern: Optional[str] = None,
-) -> GeneratorV0Adapter:
-    """Generate structured text that follows a JSON Schema.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    from outlines.types import JsonSchema
-    generator = Generator(model, JsonSchema(schema_object))
-    ```
-
-    You can then call the generator created with a prompt to generate text that
-    matches the JSON Schema.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    schema_object:
-        The JSON Schema to generate data for. Can be a JSON string, a Pydantic
-        model, or a callable that returns a JSON schema.
-    sampler:
-        The sampler defining the sampling parameters.
-    whitespace_pattern
-        Pattern to use for JSON syntactic whitespace (doesn't impact string
-        literals).
-        Example: allow only a single space or newline with
-        `whitespace_pattern=r"[\n ]?"`.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-    warnings.warn("""
-        The `json` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import JsonSchema
-        schema_str = '...'  # JSON Schema as a string
-        generator = Generator(model, JsonSchema(schema_str))
-        ```
-        You can then call the generator created with a prompt to generate
-        JSON data that matches the schema.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        return json_openai(model, schema_object, sampler)
-
-    if is_callable(schema_object):
-        json_schema = JsonSchema(get_schema_from_signature(schema_object))  # type: ignore
-    else:
-        json_schema = JsonSchema(schema_object, whitespace_pattern)
-
-    if isinstance(model, TransformersVision):
-        generator = GeneratorVisionV0Adapter(model, json_schema, sampler)
-    else:
-        generator = GeneratorV0Adapter(model, json_schema, sampler)  # type: ignore
-
-    if isinstance(schema_object, type(BaseModel)):
-        setattr(
-            generator,
-            'format_sequence',
-            lambda x: schema_object.model_validate_json(x)
-        )
-    else:
-        setattr(generator, 'format_sequence', lambda x: pyjson.loads(x))
-
-    return generator
-
-
-def json_openai(
-    model,
-    schema_object: Union[str, object, Callable],
-    sampler: Sampler = multinomial(),
-):
-    if not isinstance(sampler, multinomial):
-        raise NotImplementedError(
-            "The OpenAI API does not support any other sampling algorithm "
-            + "than the multinomial sampler."
-        )
-
-    if isinstance(schema_object, type(BaseModel)):
-        schema = schema_object.model_json_schema()
-        schema["additionalProperties"] = False
-        schema = pyjson.dumps(schema)
-        format_sequence = lambda x: schema_object.model_validate_json(x)
-    elif isinstance(schema_object, str):
-        schema = schema_object
-        format_sequence = lambda x: pyjson.loads(x)
-    else:
-        raise ValueError(
-            f"Cannot parse schema {schema_object}. The schema must be either "
-            + "a Pydantic object or a string that contains the JSON Schema "
-            + "specification"
-        )
-
-    # create copied, patched model with normalized json schema set
-    generator = model.new_with_replacements(
-        response_format={
-            "type": "json_schema",
-            "json_schema": {
-                "name": "default",
-                "strict": True,
-                "schema": pyjson.loads(schema),
-            },
-        }
-    )
-
-    generator.format_sequence = format_sequence
-
-    return generator
diff --git a/outlines/v0_legacy/generate/regex.py b/outlines/v0_legacy/generate/regex.py
deleted file mode 100644
index 261f72037..000000000
--- a/outlines/v0_legacy/generate/regex.py
+++ /dev/null
@@ -1,67 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models.openai import OpenAI
-from outlines.types import Regex
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.samplers import Sampler, multinomial
-
-
-def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
-    """Generate structured text in the language of a regular expression.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    from outlines.types import Regex
-    generator = Generator(model, Regex(r"^[a-z]+$"))
-    ```
-
-    You can then call the generator created with a prompt to generate text
-    that matches the regular expression.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    regex_str:
-        The regular expression that the output must follow.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-    warnings.warn("""
-        The `regex` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import Regex
-        generator = Generator(model, Regex(r'regex_str'))
-        ```
-        You can then call the generator created with a prompt to generate
-        text that matches the regular expression.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use regex-structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API."
-        )
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, regex_str, sampler)
-    return GeneratorV0Adapter(model, regex_str, sampler)
diff --git a/outlines/v0_legacy/generate/text.py b/outlines/v0_legacy/generate/text.py
deleted file mode 100644
index 148cf131d..000000000
--- a/outlines/v0_legacy/generate/text.py
+++ /dev/null
@@ -1,64 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import Union
-
-from outlines.models.openai import OpenAI
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.samplers import Sampler, multinomial
-
-
-def text(model, sampler: Sampler = multinomial()) -> Union[
-    GeneratorV0Adapter, GeneratorVisionV0Adapter, OpenAI
-]:
-    """Generate unconstrained text.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    generator = Generator(model)
-    ```
-
-    You can then call the generator created with a prompt to generate text.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-    warnings.warn("""
-        The `text` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        generator = Generator(model)
-        ```
-        You can then call the generator created with a prompt to generate
-        text.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        return model
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, None, sampler)
-    return GeneratorV0Adapter(model, None, sampler)
diff --git a/outlines/v0_legacy/models/__init__.py b/outlines/v0_legacy/models/__init__.py
deleted file mode 100644
index 1dc012c37..000000000
--- a/outlines/v0_legacy/models/__init__.py
+++ /dev/null
@@ -1,21 +0,0 @@
-from .exllamav2 import ExLlamaV2Model, exl2
-from .llamacpp import llamacpp
-from .mlxlm import mlxlm
-from .openai import azure_openai, openai
-from .transformers import mamba, transformers
-from .transformers_vision import TransformersVision, transformers_vision
-from .vllm_offline import vllm
-
-__all__ = [
-    "ExLlamaV2Model",
-    "TransformersVision",
-    "azure_openai",
-    "exl2",
-    "llamacpp",
-    "mamba",
-    "mlxlm",
-    "openai",
-    "transformers",
-    "transformers_vision",
-    "vllm"
-]
diff --git a/outlines/v0_legacy/models/llamacpp.py b/outlines/v0_legacy/models/llamacpp.py
deleted file mode 100644
index 6c29630d5..000000000
--- a/outlines/v0_legacy/models/llamacpp.py
+++ /dev/null
@@ -1,147 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import Optional
-
-from outlines.models import LlamaCpp
-
-
-def llamacpp_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `llamacpp` model
-    to the new `Generator` API.
-    """
-    # generation_params
-    if (
-        generation_params.get("stop_at") is not None
-        and kwargs.get("stop") is None
-    ):
-        kwargs["stop"] = generation_params["stop_at"]
-
-    if (
-        generation_params.get("seed") is not None
-        and kwargs.get("seed") is None
-    ):
-        kwargs["seed"] = generation_params["seed"]
-
-    if (
-        generation_params.get("max_tokens") is not None
-        and kwargs.get("max_tokens") is None
-    ):
-        kwargs["max_tokens"] = generation_params["max_tokens"]
-    else:
-        kwargs["max_tokens"] = -1
-
-    # sampling_params
-    if sampling_params.get("sampler") == "beam_search":
-        raise NotImplementedError(
-            "The `llama_cpp_python` library does not support Beam Search."
-        )
-
-    if sampling_params.get("num_samples") != 1:
-        raise NotImplementedError(
-            "The `llama_cpp_python` library does not allow to take several "
-            + "samples."
-        )
-
-    if (
-        sampling_params.get("top_p") is not None
-        and kwargs.get("top_p") is None
-    ):
-        kwargs["top_p"] = sampling_params["top_p"]
-    else:
-        kwargs["top_p"] = 1.0
-
-    if (
-        sampling_params.get("top_k") is not None
-        and kwargs.get("top_k") is None
-    ):
-        kwargs["top_k"] = sampling_params["top_k"]
-    else:
-        kwargs["top_k"] = -1
-
-    if (
-        sampling_params.get("temperature") is not None
-        and kwargs.get("temperature") is None
-    ):
-        kwargs["temperature"] = sampling_params["temperature"]
-    else:
-        kwargs["temperature"] = 1.0
-
-    # model specific params
-    if kwargs.get("min_p") is None:
-        kwargs["min_p"] = 0.0
-
-    if kwargs.get("repeat_penalty") is None:
-        kwargs["repeat_penalty"] = 1.0
-
-    # The choice to stream or not should happen via the high-level API
-    kwargs.pop("stream", None)
-
-    return kwargs
-
-
-def llamacpp(
-    repo_id: str, filename: Optional[str] = None, **llamacpp_model_params
-) -> LlamaCpp:
-    """Load a model from the `llama-cpp-python` library.
-
-    We use the `Llama.from_pretrained` classmethod that downloads models
-    directly from the HuggingFace hub, instead of asking users to specify
-    a path to the downloaded model. One can still load a local model
-    by initializing `llama_cpp.Llama` directly.
-
-    Parameters
-    ----------
-    repo_id
-        The name of the model repository.
-    filename:
-        A filename of glob pattern to match the model file in the repo.
-    llama_cpp_model_params
-        Llama-specific model parameters. See the `llama-cpp-python` documentation
-        for the full list: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__
-
-    """
-    from llama_cpp import Llama
-
-    warnings.warn("""
-        The `llamacpp` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `LlamaCpp` model with the
-        `outlines.from_llamacpp` function that takes a `llama_cpp.Llama`
-        instance as argument. For example:
-        ```python
-        from llama_cpp import Llama
-        from outlines import from_llamacpp
-        model = from_llamacpp(Llama.from_pretrained(
-            repo_id="<repo_id>",
-            filename="<filename>",
-        ))
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    # Default to using the model's full context length
-    if "n_ctx" not in llamacpp_model_params:
-        llamacpp_model_params["n_ctx"] = 0
-
-    if "verbose" not in llamacpp_model_params:
-        llamacpp_model_params["verbose"] = False
-
-    # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
-    if "tokenizer" not in llamacpp_model_params:
-        warnings.warn(
-            "The pre-tokenizer in `llama.cpp` handles unicode improperly "
-            + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
-            + "Outlines may raise a `RuntimeError` when building the regex index.\n"
-            + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
-            + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
-        )
-
-    model = Llama.from_pretrained(repo_id, filename, **llamacpp_model_params)
-
-    return LlamaCpp(model)
diff --git a/outlines/v0_legacy/models/mlxlm.py b/outlines/v0_legacy/models/mlxlm.py
deleted file mode 100644
index a69594173..000000000
--- a/outlines/v0_legacy/models/mlxlm.py
+++ /dev/null
@@ -1,153 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-from typing import Optional
-import warnings
-
-from outlines.models import MLXLM
-
-
-def mlxlm_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `mlxlm` model
-    to the new `Generator` API.
-    """
-    # generation_params
-    if generation_params.get("stop_at") is not None:
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support `stop_at`."
-        )
-
-    if generation_params.get("seed") is not None:
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support `seed`."
-        )
-
-    if (
-        generation_params.get("max_tokens") is not None
-        and kwargs.get("max_tokens") is None
-    ):
-        kwargs["max_tokens"] = generation_params["max_tokens"]
-    else:
-        kwargs["max_tokens"] = int(1e9)
-
-    # sampling_params
-    if sampling_params.get("sampler") == "beam_search":
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support Beam Search."
-        )
-
-    if sampling_params.get("num_samples") != 1:
-        raise NotImplementedError(
-            "The `mlx-lm` library does not allow to take several samples."
-        )
-
-    if (
-        sampling_params.get("top_k") is not None
-        or kwargs.get("top_k") is not None
-    ):
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support top_k."
-        )
-
-    if (
-        sampling_params.get("top_p") is not None
-        or kwargs.get("top_p") is not None
-    ):
-        kwargs.pop("top_p", None)
-        warnings.warn(
-            """
-            The `top_p` parameter is not available anymore.
-            The argument will be ignored. Sorry for the inconvenience.
-            Please migrate to the v1 of Outlines.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-
-    if (
-        sampling_params.get("temperature") is not None
-        or kwargs.get("temperature") is not None
-    ):
-        kwargs.pop("temperature", None)
-        warnings.warn(
-            """
-            The `temperature` parameter is not available anymore.
-            The argument will be ignored. Sorry for the inconvenience.
-            Please migrate to the v1 of Outlines.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-
-    return kwargs
-
-
-def mlxlm(
-    model_name: str,
-    tokenizer_config: dict = {},
-    model_config: dict = {},
-    adapter_path: Optional[str] = None,
-    lazy: bool = False,
-):
-    """Instantiate a model from the `mlx_lm` library and its tokenizer.
-
-    Signature adapted from
-    https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L422
-
-    Parameters
-    ----------
-    Args:
-        path_or_hf_repo (Path): The path or the huggingface repository to load the model from.
-        tokenizer_config (dict, optional): Configuration parameters specifically for the tokenizer.
-            Defaults to an empty dictionary.
-        model_config(dict, optional): Configuration parameters specifically for the model.
-            Defaults to an empty dictionary.
-        adapter_path (str, optional): Path to the LoRA adapters. If provided, applies LoRA layers
-            to the model. Default: ``None``.
-        lazy (bool): If False eval the model parameters to make sure they are
-            loaded in memory before returning, otherwise they will be loaded
-            when needed. Default: ``False``
-
-    Returns
-    -------
-    A `MLXLM` model instance.
-
-    """
-    warnings.warn("""
-        The `mlxlm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `MLXLM` model with the
-        `outlines.from_mlxlm` function that takes a mlxlm model
-        and a tokenizer as arguments. For example:
-        ```python
-        from mlx_lm import load
-        from outlines import from_mlxlm
-        model = from_mlxlm(
-            load("<model_name>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    try:
-        import mlx.core as mx
-        import mlx_lm
-    except ImportError:
-        raise ImportError(
-            "The `mlx_lm` library needs to be installed in order to use `mlx_lm` models."
-        )
-    if not mx.metal.is_available():
-        raise RuntimeError("You cannot use `mlx_lm` without Apple Silicon (Metal)")
-
-    model, tokenizer = mlx_lm.load(
-        model_name,
-        tokenizer_config=tokenizer_config,
-        model_config=model_config,
-        adapter_path=adapter_path,
-        lazy=lazy,
-    )
-    return MLXLM(model, tokenizer)
diff --git a/outlines/v0_legacy/models/openai.py b/outlines/v0_legacy/models/openai.py
deleted file mode 100644
index 46546b003..000000000
--- a/outlines/v0_legacy/models/openai.py
+++ /dev/null
@@ -1,64 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import functools
-from dataclasses import replace
-from typing import Optional
-
-from outlines.models.openai import OpenAI, OpenAIConfig
-
-
-@functools.singledispatch
-def openai(model_or_client, *args, **kwargs):
-    if len(args) == 1 and isinstance(args[0], OpenAIConfig):
-        return OpenAI(model_or_client, args[0], **kwargs)
-    elif kwargs.get("config"):
-        return OpenAI(model_or_client, kwargs.pop("config"), **kwargs)
-    else:
-        return OpenAI(model_or_client, OpenAIConfig(), **kwargs)
-
-
-@openai.register(str)
-def openai_model(
-    model_name: str,
-    config: Optional[OpenAIConfig] = None,
-    **openai_client_params,
-):
-    try:
-        from openai import AsyncOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
-        )
-
-    if config is not None:
-        config = replace(config, model=model_name)  # type: ignore
-    else:
-        config = OpenAIConfig(model=model_name)
-
-    client = AsyncOpenAI(**openai_client_params)
-
-    return OpenAI(client, config)
-
-
-def azure_openai(
-    deployment_name: str,
-    model_name: Optional[str] = None,
-    config: Optional[OpenAIConfig] = None,
-    **azure_openai_client_params,
-):
-    try:
-        from openai import AsyncAzureOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
-        )
-
-    if config is not None:
-        config = replace(config, model=deployment_name)  # type: ignore
-    if config is None:
-        config = OpenAIConfig(model=deployment_name)
-
-    client = AsyncAzureOpenAI(**azure_openai_client_params)
-
-    return OpenAI(client, config)
diff --git a/outlines/v0_legacy/models/transformers.py b/outlines/v0_legacy/models/transformers.py
deleted file mode 100644
index eb936c491..000000000
--- a/outlines/v0_legacy/models/transformers.py
+++ /dev/null
@@ -1,196 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import Optional
-
-from outlines.models import Transformers
-
-
-def transformers_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `transformers` model
-    to the new `Generator` API.
-    """
-    from transformers import set_seed
-
-    # generation_params
-    if (
-        generation_params.get("stop_at") is not None
-        and kwargs.get("stop_strings") is None
-    ):
-        kwargs["stop_strings"] = generation_params["stop_at"]
-
-    if (
-        generation_params.get("max_tokens") is not None
-        and kwargs.get("max_new_tokens") is None
-    ):
-        kwargs["max_new_tokens"] = generation_params["max_tokens"]
-    else:
-        kwargs["max_new_tokens"] = int(2**30)
-
-    if (
-        generation_params.get("seed") is not None
-        or kwargs.get("seed") is not None
-    ):
-        seed = (
-            generation_params.get("seed")
-            if generation_params.get("seed") is not None
-            else kwargs.get("seed")
-        )
-        set_seed(seed) # global seed
-
-    # sampling_params
-    if sampling_params.get("sampler") == "multinomial":
-        kwargs["do_sample"] = True
-    elif sampling_params.get("sampler") == "beam_search":
-        kwargs["num_beams"] = sampling_params.get("num_samples", 1)
-
-    if (
-        sampling_params.get("num_samples") is not None
-        and kwargs.get("num_return_sequences") is None
-    ):
-        kwargs["num_return_sequences"] = sampling_params["num_samples"]
-    else:
-        kwargs["num_return_sequences"] = 1
-
-    if (
-        sampling_params.get("top_p") is not None
-        and kwargs.get("top_p") is None
-    ):
-        kwargs["top_p"] = sampling_params["top_p"]
-
-    if (
-        sampling_params.get("top_k") is not None
-        and kwargs.get("top_k") is None
-    ):
-        kwargs["top_k"] = sampling_params["top_k"]
-
-    if (
-        sampling_params.get("temperature") is not None
-        and kwargs.get("temperature") is None
-    ):
-        kwargs["temperature"] = sampling_params["temperature"]
-
-    return kwargs
-
-
-def transformers(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-    model_class=None,
-    tokenizer_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
-
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    tokenizer_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the tokenizer.
-
-    Returns
-    -------
-    A `TransformersModel` model instance.
-
-    """
-    warnings.warn("""
-        The `transformers` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `Transformers` model with the
-        `outlines.from_transformers` function that takes a transformers model
-        and a tokenizer as arguments. For example:
-        ```python
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-        from outlines import from_transformers
-        model = from_transformers(
-            AutoModelForCausalLM.from_pretrained("<model_name>"),
-            AutoTokenizer.from_pretrained("<model_name>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if model_class is None or tokenizer_class is None:
-        try:
-            from transformers import AutoModelForCausalLM, AutoTokenizer
-        except ImportError:  # pragma: no cover
-            raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
-            )
-    if model_class is None:
-        model_class = AutoModelForCausalLM
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
-
-    if device is not None:
-        model_kwargs["device_map"] = device
-
-    model = model_class.from_pretrained(model_name, **model_kwargs)
-
-    tokenizer_kwargs.setdefault("padding_side", "left")
-    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
-
-    return Transformers(model, tokenizer)
-
-
-def mamba(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-):
-    warnings.warn("""
-        The `mamba` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `Transformers` model with the
-        `outlines.from_transformers` function that takes a transformers model
-        and a tokenizer as arguments. For Mamba, you would use a mamba model
-        and a mamba tokenizer. For example:
-        ```python
-        from transformers import MambaForCausalLM, MambaTokenizer
-        from outlines import from_transformers
-        model = from_transformers(
-            MambaForCausalLM.from_pretrained("<repo_id>"),
-            MambaTokenizer.from_pretrained("<repo_id>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    try:
-        from transformers import MambaForCausalLM
-
-    except ImportError:  # pragma: no cover
-        raise ImportError(
-            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
-        )
-
-    # We do not want to raise the deprecation warning through the transformers function
-    with warnings.catch_warnings():
-        warnings.filterwarnings(
-            "ignore",
-            category=DeprecationWarning,
-        )
-        return transformers(
-            model_name=model_name,
-            device=device,
-            model_kwargs=model_kwargs,
-            tokenizer_kwargs=tokenizer_kwargs,
-            model_class=MambaForCausalLM,
-    )
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
deleted file mode 100644
index 1c7657a19..000000000
--- a/outlines/v0_legacy/models/transformers_vision.py
+++ /dev/null
@@ -1,107 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-from typing import TYPE_CHECKING, Optional
-import warnings
-
-from outlines.models.transformers import (
-    Transformers,
-    TransformersMultiModalTypeAdapter,
-)
-
-
-class TransformersVision(Transformers):
-    def __init__(self, model, tokenizer, processor):
-        self.processor = processor
-        self.processor.padding_side = "left"
-        self.processor.pad_token = "[PAD]"
-
-        super().__init__(model, tokenizer)
-
-        self.type_adapter = TransformersMultiModalTypeAdapter()
-
-
-    def _prepare_model_inputs(self, model_input, output_type):
-        model_input = self.type_adapter.format_input(model_input)
-        inputs = self.processor(
-            **model_input, padding=True, return_tensors="pt"
-        ).to(self.model.device)
-
-        return model_input["text"], inputs
-
-
-def transformers_vision(
-    model_name: str,
-    model_class,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    processor_kwargs: dict = {},
-    tokenizer_class=None,
-    processor_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
-
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    model_class
-        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
-        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    processor_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the processor.
-
-    Returns
-    -------
-    A `TransformersModel` model instance.
-
-    """
-    warnings.warn("""
-        The `transformers_vision` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `TransformersMultiModal` model with
-        the `outlines.from_transformers` function that takes a transformers
-        model and a processor as arguments. For example:
-        ```python
-        from transformers import AutoModelForCausalLM, AutoProcessor
-        from outlines import from_transformers
-        model = from_transformers(
-            AutoModelForCausalLM.from_pretrained("<model_name>"),
-            AutoProcessor.from_pretrained("<model_name>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if processor_class is None or tokenizer_class is None:
-        try:
-            from transformers import AutoProcessor, AutoTokenizer
-        except ImportError:  # pragma: no cover
-            raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
-            )
-    if processor_class is None:
-        processor_class = AutoProcessor
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
-
-    if device is not None:
-        model_kwargs["device_map"] = device
-
-    model = model_class.from_pretrained(model_name, **model_kwargs)
-
-    processor_kwargs.setdefault("padding_side", "left")
-    processor_kwargs.setdefault("pad_token", "[PAD]")
-    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-    tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
-
-    return TransformersVision(model, tokenizer, processor)
diff --git a/outlines/v0_legacy/models/vllm_offline.py b/outlines/v0_legacy/models/vllm_offline.py
deleted file mode 100644
index 8efb7cfc0..000000000
--- a/outlines/v0_legacy/models/vllm_offline.py
+++ /dev/null
@@ -1,102 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models import VLLMOffline
-
-
-def vllm_offline_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `vllm` model
-    to the new `Generator` API.
-    """
-    from vllm.sampling_params import SamplingParams
-
-    # vllm has a `sampling_params` argument that is a `SamplingParams` object
-    # this is unrelated to the `sampling_params` argument in the legacy
-    # `Generator` API
-    sampling_params_kwarg = kwargs.get("sampling_params", SamplingParams())
-
-    # generation_params
-    if generation_params.get("stop_at") is not None:
-        if isinstance(generation_params["stop_at"], str):
-            sampling_params_kwarg.stop_strings = [generation_params["stop_at"]]
-        else:
-            sampling_params_kwarg.stop_strings = generation_params["stop_at"]
-
-    if generation_params.get("seed") is not None:
-        sampling_params_kwarg.seed = generation_params["seed"]
-
-    if generation_params.get("max_tokens") is not None:
-        sampling_params_kwarg.max_tokens = generation_params["max_tokens"]
-
-    # sampling_params
-    if sampling_params.get("sampler") == "beam_search":
-        sampling_params_kwarg.use_beam_search = True
-
-    if (
-        sampling_params.get("num_samples") is not None
-        and sampling_params_kwarg.n == 1
-    ):
-        sampling_params_kwarg.n = sampling_params["num_samples"]
-        sampling_params_kwarg.best_of = sampling_params["num_samples"]
-
-    if (
-        sampling_params.get("top_p") is not None
-        and sampling_params_kwarg.top_p == 1.0
-    ):
-        sampling_params_kwarg.top_p = sampling_params["top_p"]
-
-    if (
-        sampling_params.get("top_k") is not None
-        and sampling_params_kwarg.top_k == -1
-    ):
-        sampling_params_kwarg.top_k = sampling_params["top_k"]
-        if sampling_params.get("top_k") == 1:
-            sampling_params_kwarg.repetition_penalty = 0
-
-    if (
-        sampling_params.get("temperature") is not None
-        and sampling_params_kwarg.temperature == 1.0
-    ):
-        sampling_params_kwarg.temperature = sampling_params["temperature"]
-
-    return {"sampling_params": sampling_params_kwarg, **kwargs}
-
-
-def vllm(model_name: str, **vllm_model_params):
-    """Load an offline vLLM model.
-
-    Parameters
-    ---------
-    model_name
-        The name of the model to load from the HuggingFace hub.
-    vllm_model_params
-        vLLM-specific model parameters. See the vLLM code for the full list:
-        https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/llm.py
-
-    """
-    warnings.warn("""
-        The `vllm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `VLLMOffline` model with the
-        `outlines.from_vllm_offline` function that takes a vLLM model name as
-        argument.
-        For example:
-        ```python
-        from vllm import LLM
-        from outlines import from_vllm_offline
-        model = from_vllm_offline(LLM("<model_name>"))
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    from vllm import LLM
-
-    model = LLM(model_name, **vllm_model_params)
-
-    return VLLMOffline(model)
diff --git a/outlines/v0_legacy/samplers.py b/outlines/v0_legacy/samplers.py
deleted file mode 100644
index 9f1ff5cc6..000000000
--- a/outlines/v0_legacy/samplers.py
+++ /dev/null
@@ -1,118 +0,0 @@
-from dataclasses import dataclass
-from typing import Optional, Protocol
-
-
-@dataclass(frozen=True)
-class SamplingParameters:
-    """Sampling parameters available in Outlines."""
-    sampler: str
-    num_samples: int = 1
-    top_p: Optional[float] = None
-    top_k: Optional[int] = None
-    temperature: Optional[float] = None
-
-
-class Sampler(Protocol):
-    """
-    Protocol for all samplers.
-
-    As outlines now delegates the handling of the generation loop
-    to the models, the samplers are only used to define the sampling
-    parameters that are passed to the model.
-    """
-
-    samples: int
-
-    def sampling_params(self) -> SamplingParameters:
-        ...
-
-
-class GreedySampler:
-    """Greedy Sampling algorithm.
-
-    Greedy sampling consists in choosing the token with the largest
-    likelihood at every step.
-
-    We don't allow more than one sample. We could attribute this a meaning, for
-    instance the k-th sample represents the k-th most likely token. In which
-    case it would be equivalent to beam search without the sequence weights.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(self):
-        self.samples = 1
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("greedy", self.samples, None, None, 0.0)
-
-
-greedy = GreedySampler
-
-
-class MultinomialSampler:
-    """Multinomial sampling algorithm.
-
-    Multinomial sampling consists in randomly sampling the next token assuming
-    its distribution is a Categorical distribution parametrized by the
-    next-token logits.
-
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(
-        self,
-        samples: int = 1,
-        *,
-        top_k: Optional[int] = None,
-        top_p: Optional[float] = None,
-        temperature: Optional[float] = None,
-    ):
-        self.samples = samples
-        self.top_k = top_k
-        self.top_p = top_p
-        self.temperature = temperature
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters(
-            "multinomial",
-            self.samples,
-            self.top_p,
-            self.top_k,
-            self.temperature,
-        )
-
-
-multinomial = MultinomialSampler
-
-
-class BeamSearchSampler:
-    """Beam Search sampling algorithm.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence. Equivalent to the
-        number of beams.
-    """
-
-    def __init__(self, beams: int = 1):
-        self.samples = beams
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
-
-
-beam_search = BeamSearchSampler
diff --git a/pyproject.toml b/pyproject.toml
index 9216216d2..dfed39973 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -28,42 +28,36 @@ dependencies = [
    "jinja2",
    "lark",
    "nest_asyncio",
+   "numpy",
    "cloudpickle",
    "diskcache",
    "pydantic>=2.0",
    "referencing",
    "jsonschema",
    "requests",
-   "pillow",
    "tqdm",
    "typing_extensions",
    "iso3166",
    "airportsdata",
+   "torch",
    "outlines_core==0.1.26",
    "genson",
-   "torch",
 ]
 dynamic = ["version"]
 
 [project.optional-dependencies]
-anthropic = ["anthropic"]
-dottxt = ["dottxt"]
-gemini = ["google-genai"]
-llamacpp = ["huggingface-hub", "llama-cpp-python"]
-mlxlm = ["datasets", "mlx", "mlx-lm"]
-ollama = ["ollama"]
+vllm = ["vllm", "transformers", "numpy<2"]
+transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
+mlxlm = ["mlx-lm", "datasets"]
 openai = ["openai"]
-sglang = ["openai"]
-tgi = ["huggingface_hub"]
-transformers = ["accelerate", "datasets", "transformers"]
-vllm = ["openai"]
+llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
+exllamav2 = ["exllamav2"]
 test = [
     "pre-commit",
     "pytest",
     "pytest-benchmark",
     "pytest-cov",
     "pytest-mock",
-    "pytest-asyncio",
     "coverage[toml]>=5.1",
     "diff-cover",
     "accelerate",
@@ -74,22 +68,18 @@ test = [
     "huggingface_hub",
     "openai>=1.0.0",
     "datasets",
-    "anthropic",
-    "google-genai",
-    "transformers<=4.52.1",
+    "transformers",
     "pillow",
+    "exllamav2",
     "jax",
-    "flax",
-    "numpy>=2.0.0,<2.2.0",
-    "torch",
-    "tensorflow",
-    "tf-keras",
-    "ollama",
-    "dottxt",
-    "sentencepiece",
-    "mkdocs_gen_files",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
+serve = [
+    "vllm>=0.3.0",
+    "uvicorn",
+    "fastapi",
+    "pydantic>=2.0",
+]
 
 [project.urls]
 homepage = "https://github.com/dottxt-ai/outlines"
@@ -118,8 +108,6 @@ filterwarnings = [
     "ignore::FutureWarning:huggingface_hub.*",
     "ignore::UserWarning",
     "ignore::DeprecationWarning:pyairports.*",
-    "ignore::DeprecationWarning:jax.*",
-    "ignore::DeprecationWarning:flax.*",
 ]
 
 [tool.mypy]
@@ -128,13 +116,13 @@ enable_incomplete_feature = ["Unpack"]
 
 [[tool.mypy.overrides]]
 module = [
+    "exllamav2.*",
     "jax",
     "jaxlib",
     "jax.numpy",
     "jinja2",
     "jsonschema.*",
-    "anthropic.*",
-    "google.*",
+    "openai.*",
     "mamba_ssm.*",
     "mlx_lm.*",
     "mlx.*",
@@ -153,55 +141,36 @@ module = [
     "lark.*",
     "interegular.*",
     "datasets.*",
-    "openai.*",
     "requests.*",
     "responses.*",
     "vllm.*",
+    "uvicorn.*",
+    "fastapi.*",
     "iso3166.*",
     "airportsdata.*",
     "outlines_core.*",
     "genson",
-    "ollama.*",
-    "dottxt.*",
-    "tensorflow",
-    "tensorflow.*",
-    "tf-keras",
-    "tf-keras.*",
-    "mkdocs_gen_files.*",
 ]
 ignore_missing_imports = true
 
 [tool.coverage.run]
-# we omit the files that require a GPU or Apple Silicon
-# as well as the models that make API calls
 omit = [
     "outlines/_version.py",
-    "outlines/models/anthropic.py",
-    "outlines/models/dottxt.py",
-    "outlines/models/gemini.py",
-    "outlines/models/mlxlm.py",
-    "outlines/models/openai.py",
-    "outlines/models/vllm_offline.py",
-    "outlines/processors/tensor_adapters/mlx.py",
-    "outlines/v0_legacy/models/anthropic.py",
-    "outlines/v0_legacy/models/dottxt.py",
-    "outlines/v0_legacy/models/exllamav2.py",
-    "outlines/v0_legacy/models/gemini.py",
-    "outlines/v0_legacy/models/mlxlm.py",
-    "outlines/v0_legacy/models/openai.py",
-    "outlines/v0_legacy/models/vllm_offline.py",
     "tests/*",
 ]
 branch = true
 relative_files = true
 
 [tool.coverage.report]
-show_missing = true
+omit = [
+    "tests/*",
+]
 exclude_lines = [
     "pragma: no cover",
     "if TYPE_CHECKING:",
     "\\.\\.\\.",
 ]
+show_missing = true
 
 [tool.diff_cover]
 compare_branch = "origin/main"
diff --git a/requirements-doc.txt b/requirements-doc.txt
index 91f621d6c..05f9f36ec 100644
--- a/requirements-doc.txt
+++ b/requirements-doc.txt
@@ -6,7 +6,4 @@ mkdocs-section-index
 mkdocstrings[python]
 mkdocs-git-committers-plugin-2
 mkdocs-git-revision-date-localized-plugin
-mkdocs-redirects
-mkdocs-gen-files
-mkdocs-literate-nav
 mike
diff --git a/scripts/gen_ref_pages.py b/scripts/gen_ref_pages.py
deleted file mode 100644
index 48c226442..000000000
--- a/scripts/gen_ref_pages.py
+++ /dev/null
@@ -1,53 +0,0 @@
-"""Generate the API reference pages and navigation automatically.
-
-This script is based on the `gen_ref_pages.py` script in the
-[mkdocstrings](https://mkdocstrings.github.io/recipes/#automatic-code-reference-pages) project.
-
-To exclude a file or module from being included in the generated API reference,
-add a part of its path to the `EXCLUDED_FILES` list.
-"""
-
-from pathlib import Path
-
-import mkdocs_gen_files
-
-
-CODEBASE_DIR_NAME = "outlines"
-OUTPUT_DIR_NAME = "api_reference"
-EXCLUDED_FILES = ["_version", "v0_legacy"]
-
-
-nav = mkdocs_gen_files.Nav()
-root = Path(__file__).parent.parent
-src = root / CODEBASE_DIR_NAME
-
-for path in sorted(src.rglob("*.py")):
-    module_path = path.relative_to(src).with_suffix("")
-    doc_path = path.relative_to(src).with_suffix(".md")
-    full_doc_path = Path(OUTPUT_DIR_NAME, doc_path)
-
-    parts = tuple(module_path.parts)
-
-    if any(part in EXCLUDED_FILES for part in parts):
-        continue
-
-    if parts[-1] == "__init__":
-        if len(parts) == 1:
-            doc_path = Path("index.md")
-            full_doc_path = Path(OUTPUT_DIR_NAME, doc_path)
-            parts = (CODEBASE_DIR_NAME,)
-        else:
-            parts = parts[:-1]
-            doc_path = doc_path.with_name("index.md")
-            full_doc_path = full_doc_path.with_name("index.md")
-
-    nav[parts] = doc_path.as_posix()
-
-    with mkdocs_gen_files.open(full_doc_path, "w") as fd:
-        ident = ".".join(parts)
-        fd.write(f"::: {CODEBASE_DIR_NAME}.{ident}")
-
-    mkdocs_gen_files.set_edit_path(full_doc_path, path.relative_to(root))
-
-with mkdocs_gen_files.open(f"{OUTPUT_DIR_NAME}/SUMMARY.md", "w") as nav_file:
-    nav_file.writelines(nav.build_literate_nav())
diff --git a/tests/processors/test_cfg_guide.py b/tests/fsm/test_cfg_guide.py
similarity index 97%
rename from tests/processors/test_cfg_guide.py
rename to tests/fsm/test_cfg_guide.py
index 6091fd4c2..d92afa625 100644
--- a/tests/processors/test_cfg_guide.py
+++ b/tests/fsm/test_cfg_guide.py
@@ -4,9 +4,8 @@
 import pytest
 from transformers import AutoTokenizer
 
-from outlines import grammars
-from outlines.models.transformers import TransformerTokenizer
-from outlines.processors.guide import CFGGuide
+from outlines import grammars, models
+from outlines.fsm.guide import CFGGuide
 
 
 @pytest.fixture
@@ -342,12 +341,12 @@ def decode(self, token_ids):
 
 @pytest.fixture(scope="session")
 def tokenizer_sentencepiece_gpt2():
-    return TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
+    return models.TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
 
 
 @pytest.fixture(scope="session")
 def tokenizer_sentencepiece_llama1():
-    return TransformerTokenizer(
+    return models.TransformerTokenizer(
         AutoTokenizer.from_pretrained(
             "trl-internal-testing/tiny-random-LlamaForCausalLM"
         )
@@ -356,14 +355,16 @@ def tokenizer_sentencepiece_llama1():
 
 @pytest.fixture(scope="session")
 def tokenizer_tiktoken_llama3():
-    return TransformerTokenizer(
+    return models.TransformerTokenizer(
         AutoTokenizer.from_pretrained("yujiepan/llama-3-tiny-random")
     )
 
 
 @pytest.fixture(scope="session")
 def tokenizer_character_level_byt5():
-    return TransformerTokenizer(AutoTokenizer.from_pretrained("google/byt5-small"))
+    return models.TransformerTokenizer(
+        AutoTokenizer.from_pretrained("google/byt5-small")
+    )
 
 
 # Collects all samples within cfg_samples/ and makes adding
diff --git a/tests/processors/test_guide.py b/tests/fsm/test_guide.py
similarity index 85%
rename from tests/processors/test_guide.py
rename to tests/fsm/test_guide.py
index 0238e37d3..bf25c43c4 100644
--- a/tests/processors/test_guide.py
+++ b/tests/fsm/test_guide.py
@@ -1,23 +1,6 @@
 import pytest
 
-import llama_cpp
-import transformers
-
-import outlines
-from outlines.processors.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
-from outlines import caching
-
-try:
-    import mlx_lm
-    HAS_MLX = True
-except ImportError:
-    HAS_MLX = False
-
-try:
-    import vllm
-    HAS_VLLM = True
-except ImportError:
-    HAS_VLLM = False
+from outlines.fsm.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
 
 
 def assert_expected_tensor_ids(tensor, ids):
@@ -34,7 +17,7 @@ class MockTokenizer:
 
     fsm = StopAtEOSGuide(MockTokenizer())
 
-    instruction = fsm.get_next_instruction(fsm.initial_state)
+    instruction = fsm.get_next_instruction(fsm.start_state)
     assert isinstance(instruction, Generate)
     assert instruction.tokens is None
 
@@ -42,13 +25,11 @@ class MockTokenizer:
     assert isinstance(instruction, Write)
     assert instruction.tokens == [2]
 
-    assert fsm.get_next_state(fsm.initial_state, 2) == fsm.final_state
-    assert fsm.get_next_state(fsm.initial_state, 1) == fsm.initial_state
-    assert fsm.is_final_state(fsm.initial_state) is False
+    assert fsm.get_next_state(fsm.start_state, 2) == fsm.final_state
+    assert fsm.get_next_state(fsm.start_state, 1) == fsm.start_state
+    assert fsm.is_final_state(fsm.start_state) is False
     assert fsm.is_final_state(fsm.final_state) is True
 
-    assert isinstance(fsm.copy(), StopAtEOSGuide)
-
 
 def test_regex_vocabulary_error():
     class MockTokenizer:
@@ -200,46 +181,6 @@ def convert_token_to_string(self, token):
     assert fsm.is_final_state(state)
 
 
-def test_regex_guide_caching():
-    assert caching._caching_enabled
-
-    cache = caching.get_cache()
-    cache.clear()
-    _, _ = cache.stats(enable=True, reset=True) # (hits, misses)
-
-    regex = r"[0-9]{3}"
-
-    models = [
-        outlines.from_transformers(
-            transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-            transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini")
-        ),
-        outlines.from_llamacpp(llama_cpp.Llama.from_pretrained(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        ))
-    ]
-    if HAS_MLX:
-        models.append(outlines.from_mlxlm(
-            *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-        ))
-    if HAS_VLLM:
-        models.append(outlines.from_vllm_offline(vllm.LLM("TinyLlama/TinyLlama-1.1B-Chat-v1.0")))
-
-    for i, model in enumerate(models):
-        # First call for each model should be a miss
-        RegexGuide.from_regex(regex, model.tokenizer)
-        expected_misses = i + 1
-        expected_hits = i
-        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
-
-        # Second call for each model
-        RegexGuide.from_regex(regex, model.tokenizer)
-        expected_misses = i + 1
-        expected_hits = i + 1
-        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
-
-
 def test_cfg():
     class MockTokenizer:
         vocabulary = {"{": 1, "}": 2, "[": 3, "]": 4, "eos": 5}
@@ -297,8 +238,6 @@ def decode(self, token_ids):
     assert guide.must_terminate_state(state)
     assert guide.can_terminate_state(state)
 
-    assert isinstance(guide.copy(), CFGGuide)
-
 
 def test_cfg_early_termination():
     class MockTokenizer:
@@ -526,14 +465,3 @@ def decode(self, token_ids):
     state = guide.get_next_state(state, token_id=4)
     assert guide.must_terminate_state(state)
     assert guide.can_terminate_state(state)
-    assert guide.is_final_state(state)
-
-    cfg_str = """
-        start: CHAR
-        CHAR: /./
-    """
-
-    guide = CFGGuide(cfg_str, tokenizer)
-
-    state = guide.get_next_state(guide.initial_state, token_id=1)
-    assert guide.get_next_state(state, token_id=4).parser_state is None
diff --git a/tests/fsm/test_json_schema.py b/tests/fsm/test_json_schema.py
new file mode 100644
index 000000000..23864e029
--- /dev/null
+++ b/tests/fsm/test_json_schema.py
@@ -0,0 +1,78 @@
+import json
+from contextlib import nullcontext
+from enum import Enum
+from functools import partial
+from typing import List
+
+import pytest
+from outlines_core.fsm.json_schema import build_regex_from_schema
+from pydantic import BaseModel, constr
+
+from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
+
+
+def test_function_basic():
+    def test_function(foo: str, bar: List[int]):
+        pass
+
+    result = get_schema_from_signature(test_function)
+    assert result["type"] == "object"
+    assert list(result["properties"].keys()) == ["foo", "bar"]
+    assert result["properties"]["foo"]["type"] == "string"
+    assert result["properties"]["bar"]["type"] == "array"
+    assert result["properties"]["bar"]["items"]["type"] == "integer"
+
+
+def test_function_no_type():
+    def test_function(foo, bar: List[int]):
+        pass
+
+    with pytest.raises(ValueError):
+        get_schema_from_signature(test_function)
+
+
+def test_from_pydantic():
+    class User(BaseModel):
+        user_id: int
+        name: str
+        maxlength_name: constr(max_length=10)
+        minlength_name: constr(min_length=10)
+        value: float
+        is_true: bool
+
+    schema = json.dumps(User.model_json_schema())
+    regex_str = build_regex_from_schema(schema)
+    assert isinstance(regex_str, str)
+
+
+def add(a: float, b: float) -> float:
+    return a + b
+
+
+class MyEnum(Enum):
+    add = partial(add)
+    a = "a"
+    b = 2
+
+
+# if you don't register your function as callable, you will get an empty enum
+class EmptyEnum(Enum):
+    add = add
+
+
+@pytest.mark.parametrize(
+    "enum,expectation",
+    [
+        (MyEnum, nullcontext()),
+        (EmptyEnum, pytest.raises(ValueError)),
+    ],
+)
+def test_enum_schema(enum, expectation):
+    with expectation:
+        schema = get_schema_from_enum(enum)
+        regex_str = build_regex_from_schema(json.dumps(schema))
+        assert isinstance(regex_str, str)
+        assert schema["title"] == enum.__name__
+        assert len(schema["oneOf"]) == len(enum)
+        for elt in schema["oneOf"]:
+            assert type(elt) in [int, float, bool, type(None), str, dict]
diff --git a/tests/fsm/test_types.py b/tests/fsm/test_types.py
new file mode 100644
index 000000000..4017a817c
--- /dev/null
+++ b/tests/fsm/test_types.py
@@ -0,0 +1,22 @@
+import datetime as pydatetime
+
+import pytest
+
+from outlines.fsm.types import python_types_to_regex
+from outlines import types
+
+
+@pytest.mark.parametrize(
+    "python_type,custom_type",
+    [
+        (int, types.integer),
+        (float, types.number),
+        (bool, types.boolean),
+        (pydatetime.date, types.date),
+        (pydatetime.time, types.time),
+        (pydatetime.datetime, types.datetime),
+    ],
+)
+def test_python_types(python_type, custom_type):
+    test_regex, _ = python_types_to_regex(python_type)
+    assert custom_type.pattern == test_regex.pattern
diff --git a/tests/generate/__init__.py b/tests/generate/__init__.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
new file mode 100644
index 000000000..abd9c72a4
--- /dev/null
+++ b/tests/generate/conftest.py
@@ -0,0 +1,66 @@
+from importlib import reload
+
+import pytest
+import torch
+
+
+def is_metal_available():
+    try:
+        import mlx.core as mx
+        import mlx_lm  # noqa: F401
+
+        assert mx.metal.is_available()
+    except (ImportError, AssertionError):
+        return False
+    return True
+
+
+def pytest_collection_modifyitems(config, items):
+    """
+    If mlxlm and Metal aren't available, skip mlxlm tests
+    If CUDA isn't available, skip vllm and transformers_vision
+    """
+    if not torch.cuda.is_available():
+        skip_marker = pytest.mark.skip(
+            reason="Skipping test because CUDA is not available"
+        )
+        for item in items:
+            if "model_fixture" in item.fixturenames:
+                model_param = item.callspec.params.get("model_fixture", None)
+                if (
+                    model_param.startswith("model_transformers_vision")
+                    or model_param.startswith("model_vllm")
+                    or model_param.startswith("model_exllamav2")
+                ):
+                    item.add_marker(skip_marker)
+
+    if not is_metal_available():
+        skip_marker = pytest.mark.skip(
+            reason="Skipping test because mlx-lm or Metal are not available"
+        )
+        for item in items:
+            if "model_fixture" in item.fixturenames:
+                model_param = item.callspec.params.get("model_fixture", None)
+                if model_param.startswith("model_mlxlm"):
+                    item.add_marker(skip_marker)
+
+
+@pytest.fixture
+def temp_cache_dir():
+    import os
+    import tempfile
+
+    import outlines.caching
+    import outlines.fsm.guide
+
+    with tempfile.TemporaryDirectory() as tempdir:
+        os.environ["OUTLINES_CACHE_DIR"] = tempdir
+        outlines.caching.get_cache.cache_clear()
+        reload(outlines)
+        reload(outlines.fsm.guide)
+        cache_status = outlines.caching._caching_enabled
+        try:
+            outlines.caching._caching_enabled = True
+            yield
+        finally:
+            outlines.caching._caching_enabled = cache_status
diff --git a/tests/generate/test_api.py b/tests/generate/test_api.py
new file mode 100644
index 000000000..7188022f5
--- /dev/null
+++ b/tests/generate/test_api.py
@@ -0,0 +1,33 @@
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image  # type: ignore
+
+from outlines.generate.api import VisionSequenceGeneratorAdapter
+
+IMG_URI = "https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg"
+PIL_IMG = Image.open(BytesIO(urlopen(IMG_URI).read())).convert("RGB")
+
+
+@pytest.mark.parametrize(
+    "prompts,media,type_error",
+    [
+        ("single prompt", [PIL_IMG], False),
+        (["prompt0", "prompt1"], [[PIL_IMG], [PIL_IMG]], False),
+        ("single prompt", [PIL_IMG, PIL_IMG], False),
+        (["prompt0", "prompt1"], [[PIL_IMG, PIL_IMG], [PIL_IMG]], False),
+        ("single prompt", "this isn't an image, it's a string", True),
+        ("single prompt", PIL_IMG, True),
+        (["prompt0", "prompt1"], [PIL_IMG], True),
+        (["prompt0", "prompt1"], [[PIL_IMG]], True),
+        (["prompt0", "prompt1"], [[[PIL_IMG]], [[PIL_IMG]]], True),
+    ],
+)
+def test_vision_sequence_generator_validate_types(prompts, media, type_error):
+    """Ensure inputs are validated correctly"""
+    if type_error:
+        with pytest.raises(TypeError):
+            VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
+    else:
+        VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
new file mode 100644
index 000000000..3d36faa5d
--- /dev/null
+++ b/tests/generate/test_generate.py
@@ -0,0 +1,439 @@
+import contextlib
+import re
+from enum import Enum
+
+import pytest
+
+import outlines.generate as generate
+import outlines.models as models
+import outlines.samplers as samplers
+
+##########################################
+# Model Fixtures
+##########################################
+
+
+@pytest.fixture(scope="session")
+def model_llamacpp(tmp_path_factory):
+    return models.llamacpp(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+
+
+@pytest.fixture(scope="session")
+def model_exllamav2(tmp_path_factory):
+    from huggingface_hub import snapshot_download
+
+    tmp_dir = tmp_path_factory.mktemp("model_download")
+    model_path = snapshot_download(
+        repo_id="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4.6-exl2",
+        cache_dir=tmp_dir,
+    )
+
+    return models.exl2(
+        model_path=model_path,
+        cache_q4=True,
+        paged=False,
+    )
+
+
+@pytest.fixture(scope="session")
+def model_mlxlm(tmp_path_factory):
+    return models.mlxlm("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
+
+
+@pytest.fixture(scope="session")
+def model_mlxlm_phi3(tmp_path_factory):
+    return models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
+
+
+@pytest.fixture(scope="session")
+def model_transformers_random(tmp_path_factory):
+    return models.transformers("hf-internal-testing/tiny-random-gpt2", device="cpu")
+
+
+@pytest.fixture(scope="session")
+def model_transformers_opt125m(tmp_path_factory):
+    return models.transformers("facebook/opt-125m", device="cpu")
+
+
+@pytest.fixture(scope="session")
+def model_mamba(tmp_path_factory):
+    return models.mamba(model_name="state-spaces/mamba-130m-hf", device="cpu")
+
+
+@pytest.fixture(scope="session")
+def model_bart(tmp_path_factory):
+    from transformers import AutoModelForSeq2SeqLM
+
+    return models.transformers(
+        "facebook/bart-base", device="cpu", model_class=AutoModelForSeq2SeqLM
+    )
+
+
+@pytest.fixture(scope="session")
+def model_transformers_vision(tmp_path_factory):
+    import torch
+    from transformers import LlavaNextForConditionalGeneration
+
+    return models.transformers_vision(
+        "llava-hf/llava-v1.6-mistral-7b-hf",
+        model_class=LlavaNextForConditionalGeneration,
+        device="cuda",
+        model_kwargs=dict(
+            torch_dtype=torch.bfloat16,
+            load_in_4bit=True,
+            low_cpu_mem_usage=True,
+        ),
+    )
+
+
+@pytest.fixture(scope="session")
+def model_vllm(tmp_path_factory):
+    return models.vllm("facebook/opt-125m", gpu_memory_utilization=0.1)
+
+
+# TODO: exllamav2 failing in main, address in https://github.com/dottxt-ai/outlines/issues/808
+# TODO: t5 tokenizer doesn't work with streaming
+"""
+@pytest.fixture(scope="session")
+def model_exllamav2(tmp_path_factory):
+    return models.exllamav2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        device="cpu"
+    )
+
+@pytest.fixture(scope="session")
+def model_t5(tmp_path_factory):
+    from transformers import AutoModelForSeq2SeqLM
+
+    return models.transformers(
+        "EleutherAI/pile-t5-base", device="cpu", model_class=AutoModelForSeq2SeqLM
+    )
+"""
+
+
+ALL_MODEL_FIXTURES = (
+    "model_llamacpp",
+    "model_exllamav2",
+    "model_mlxlm",
+    "model_mlxlm_phi3",
+    "model_transformers_random",
+    "model_transformers_opt125m",
+    "model_mamba",
+    "model_bart",
+    "model_transformers_vision",
+    "model_vllm",
+)
+
+
+class MyEnum(Enum):
+    foo = "foo"
+    bar = "bar"
+    baz = "baz"
+
+
+ALL_SAMPLE_CHOICES_FIXTURES = (
+    ["foo", "bar", "baz"],
+    MyEnum,
+)
+
+
+##########################################
+# Stuctured Generation Inputs
+##########################################
+
+
+@pytest.fixture()
+def sample_schema():
+    from pydantic import BaseModel, conint, conlist, constr
+
+    class SampleSchema(BaseModel):
+        title: constr(max_length=10)
+        numbers: conlist(conint(strict=True), min_length=3, max_length=3)
+        labels: conlist(constr(min_length=1, max_length=5), min_length=3, max_length=3)
+
+    return SampleSchema
+
+
+@pytest.fixture()
+def sample_choices():
+    return ["foo", "bar", "baz"]
+
+
+@pytest.fixture()
+def sample_lark_grammar():
+    # from https://github.com/lark-parser/lark/blob/master/docs/grammar.md
+    return """
+    ?start: hello_world "!" number
+    hello_world: ("hello" | "world") ~ 3
+    number: ("0".."9") ~ 5
+    thanks: "Thank"i " for testing!"
+    """
+
+
+REGEX_PATTERNS = [
+    "a b c d e",  # ensure proper tokenizer whitespace prefix handling
+    "(123456789)|(abcdefghijklmnop)",  # ensure consistent correct sequence handling during batch
+    r"([a-z]{10})@([a-z]{5})\.([a-z]{3})",  # email example
+]
+
+
+###########################################
+# Model/Generator Pair Behavior Definitions
+###########################################
+
+
+def enforce_not_implemented(model_fixture, *task_names):
+    """
+    Per `NOT_IMPLEMENTED`, mapping, if a model hasn't implemented a task,
+    assert an NotImplementedError is raised. Otherwise, run normally
+    """
+    NOT_IMPLEMENTED = {
+        "stream": ["model_transformers_vision", "model_vllm"],
+        "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
+        "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
+        "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
+        "cfg": ["model_llamacpp"],  # TODO: fix llama_cpp tokenizer
+    }
+    for task_name in task_names:
+        if model_fixture in NOT_IMPLEMENTED.get(task_name, []):
+            return pytest.raises(NotImplementedError)
+    else:
+        return contextlib.nullcontext()
+
+
+def get_inputs(fixture_name, batch_size=None):
+    """Get generator kwargs, just the prompt by default, but include images for transformers_visian"""
+    from io import BytesIO
+    from urllib.request import urlopen
+
+    from PIL import Image  # type: ignore
+
+    prompts = ["abcd", "efgh", "1234", "5678", "foo", "bar", "baz", "bif"]
+    prompts = prompts[0] if batch_size is None else prompts[:batch_size]
+
+    if fixture_name.endswith("_vision"):
+        img_url = "https://python-pillow.org/pillow-perf/static/space_pil_lanczos.png"
+        img = Image.open(BytesIO(urlopen(img_url).read())).convert("RGB")
+
+        if batch_size is None:
+            return {"prompts": f"<image> {prompts}", "media": [img]}
+        else:
+            return {
+                "prompts": [f"<image> {p}" for p in prompts],
+                "media": [[img] for _ in range(batch_size)],
+            }
+
+    else:
+        return {"prompts": prompts}
+
+
+###########################################
+# Tests
+###########################################
+
+
+@pytest.mark.parametrize("sampler_name", ("greedy", "multinomial", "beam_search"))
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_text(request, model_fixture, sampler_name):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.text(model, getattr(samplers, sampler_name)())
+    with enforce_not_implemented(model_fixture, sampler_name):
+        res = generator(**get_inputs(model_fixture), max_tokens=10)
+        assert isinstance(res, str)
+
+
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_regex(request, model_fixture, pattern):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.regex(model, pattern)
+    res = generator(**get_inputs(model_fixture), max_tokens=20)
+    assert re.fullmatch(pattern, res) is not None, res
+
+
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_fsm(request, model_fixture, pattern):
+    import interegular
+
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.fsm(model, interegular.parse_pattern(pattern).to_fsm())
+    res = generator(**get_inputs(model_fixture))
+    assert re.fullmatch(pattern, res) is not None, res
+
+
+@pytest.mark.skip(
+    "Fix issues with JSON, some models fail this test https://github.com/dottxt-ai/outlines/issues/985"
+)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_json(request, model_fixture, sample_schema):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.json(model, sample_schema)
+    # asserts valid within call
+    generator(**get_inputs(model_fixture), max_tokens=100)
+
+
+def test_integrate_genson_generate_json(request):
+    from genson import SchemaBuilder
+
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"name": "Toto", "age": 5})
+
+    model = request.getfixturevalue("model_transformers_opt125m")
+
+    generator = generate.json(model, builder)
+    res = generator("Return a json of a young boy")
+
+    assert "name" in res
+    assert "age" in res
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
+def test_generate_choice(request, model_fixture, sample_choices):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.choice(model, sample_choices)
+    res = generator(**get_inputs(model_fixture))
+    if isinstance(sample_choices, type(Enum)):
+        assert res in [elt.value for elt in sample_choices]
+    else:
+        assert res in sample_choices
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
+def test_generate_choice_twice(request, model_fixture, sample_choices):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.choice(model, sample_choices)
+    res = generator(**get_inputs(model_fixture))
+    if isinstance(sample_choices, type(Enum)):
+        assert res in [elt.value for elt in sample_choices]
+    else:
+        assert res in sample_choices
+
+    res = generator(**get_inputs(model_fixture))
+    if isinstance(sample_choices, type(Enum)):
+        assert res in [elt.value for elt in sample_choices]
+    else:
+        assert res in sample_choices
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_format_bool(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.format(model, bool)
+    res = generator(**get_inputs(model_fixture))
+    assert isinstance(res, bool)
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_cfg(request, model_fixture, sample_lark_grammar):
+    from lark import Lark
+
+    from outlines import grammars
+
+    model = request.getfixturevalue(model_fixture)
+    with enforce_not_implemented(model_fixture, "cfg"):
+        generator = generate.cfg(model, sample_lark_grammar)
+        res = generator(**get_inputs(model_fixture))
+        # validate legal with the grammar via lark
+        # TODO: cleanup PartialLark so doesn't modify Lark globally
+        import importlib
+
+        import lark.lark
+
+        importlib.reload(lark.lark)
+        Lark(
+            sample_lark_grammar, parser="lalr", import_paths=[grammars.GRAMMAR_PATH]
+        ).parse(res)
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_text_stream(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.text(model)
+    with enforce_not_implemented(model_fixture, "stream"):
+        for token in generator.stream(**get_inputs(model_fixture), max_tokens=10):
+            assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_regex_stream(request, model_fixture, pattern):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.regex(model, pattern)
+    with enforce_not_implemented(model_fixture, "stream"):
+        output = ""
+        for token in generator.stream(**get_inputs(model_fixture), max_tokens=20):
+            output += token
+        assert re.fullmatch(pattern, output) is not None, output
+
+
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_batch_text(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.text(model)
+    with enforce_not_implemented(model_fixture, "batch"):
+        res = generator(**get_inputs(model_fixture, 2), max_tokens=10)
+        assert isinstance(res, list)
+        assert isinstance(res[0], str)
+
+
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_regex_batch(request, model_fixture, pattern):
+    """Ensure batch requests work and fsm order is maintained"""
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.regex(model, pattern)
+    with enforce_not_implemented(model_fixture, "batch"):
+        outputs = generator(**get_inputs(model_fixture, 4), max_tokens=20)
+        for output in outputs:
+            assert re.fullmatch(pattern, output) is not None, output
+
+
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_regex_batch_stream(request, model_fixture, pattern):
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.regex(model, pattern)
+    with enforce_not_implemented(model_fixture, "batch", "stream"):
+        outputs = ["", ""]
+        for tokens in generator.stream(**get_inputs(model_fixture, 2), max_tokens=20):
+            outputs[0] += tokens[0]
+            outputs[1] += tokens[1]
+        for output in outputs:
+            assert re.fullmatch(pattern, output) is not None, output
+
+
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+def test_generate_regex_single_multinomial(request, model_fixture, pattern):
+    """Ensure batch requests work and fsm order is maintained"""
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.regex(model, pattern, sampler=samplers.multinomial(4))
+    with enforce_not_implemented(model_fixture, "multiple_samples"):
+        output_sample_groups = generator(**get_inputs(model_fixture), max_tokens=40)
+        for output in output_sample_groups:
+            assert re.fullmatch(pattern, output) is not None, output
+
+
+@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
+@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
+@pytest.mark.parametrize("sampler_name", ("multinomial", "beam_search"))
+def test_generate_regex_batch_multi_sample(
+    request, model_fixture, pattern, sampler_name
+):
+    """Ensure batch requests work and fsm order is maintained"""
+    model = request.getfixturevalue(model_fixture)
+    generator = generate.regex(
+        model, pattern, sampler=getattr(samplers, sampler_name)(4)
+    )
+    with enforce_not_implemented(model_fixture, "batch", "multiple_samples"):
+        output_batch_groups = generator(**get_inputs(model_fixture, 4), max_tokens=40)
+        for output_sample_groups in output_batch_groups:
+            for output in output_sample_groups:
+                assert re.fullmatch(pattern, output) is not None, output
diff --git a/tests/generate/test_generator.py b/tests/generate/test_generator.py
new file mode 100644
index 000000000..5a2edf8dc
--- /dev/null
+++ b/tests/generate/test_generator.py
@@ -0,0 +1,497 @@
+import math
+from typing import Generator
+
+import pytest
+import torch
+
+from outlines.fsm.guide import Generate
+from outlines.generate.api import SequenceGenerator
+from outlines.generate.generator import (
+    bias_logits,
+    get_allowed_tokens,
+    get_next_fsm_states,
+    is_generation_finished,
+    sequence_generator,
+    update_attention_masks,
+    update_token_ids,
+)
+
+
+def test_sequence_generator_class():
+    class MockFSM:
+        first_state = 0
+
+        def get_next_state(self, state, next_token_ids):
+            return 4
+
+        def get_next_instruction(self, *_):
+            return Generate([4])
+
+        def is_final_state(self, _):
+            return True
+
+        def copy(self):
+            return self
+
+    class MockTokenizer:
+        def encode(self, _):
+            # Input: "test"
+            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
+
+        def decode(self, tokens):
+            return ["testx"[i] for i in tokens]
+
+    class MockModel:
+        def __init__(self):
+            self.tokenizer = MockTokenizer()
+
+        def __call__(*_):
+            return torch.tensor([[0, 1, 2, 3, 4]], dtype=torch.float), None
+
+    class sampler:
+        def __init__(self):
+            self.samples = 1
+
+        def __call__(self, biased_logits, *_):
+            return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
+
+    # Stream
+    generator = SequenceGenerator(MockFSM(), MockModel(), sampler(), "cpu")
+    assert generator.device == "cpu"
+    assert isinstance(generator.tokenizer, MockTokenizer)
+    assert isinstance(generator.fsm, MockFSM)
+
+    sequence = generator.stream("test")
+    assert isinstance(sequence, Generator)
+
+    next(sequence)
+
+    with pytest.raises(StopIteration):
+        next(sequence)
+
+    # Call
+    generator = SequenceGenerator(MockFSM(), MockModel(), sampler(), "cpu")
+    result = generator("test")
+    assert result == "x"
+
+
+def test_sequence_generator_1d_single_iteration():
+    class MockFSM:
+        def get_next_state(self, state, next_token_ids):
+            return 0
+
+        def get_next_instruction(self, _):
+            return Generate([0, 1, 2, 3])
+
+        def is_final_state(self, _):
+            return True
+
+        def copy(self):
+            return self
+
+    class MockTokenizer:
+        def encode(self, _):
+            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
+
+        def decode(self, x):
+            return x
+
+    class MockModel:
+        def __init__(self):
+            self.tokenizer = MockTokenizer()
+
+        def __call__(*_):
+            return torch.tensor([[0, 1, 2, 3]], dtype=torch.float), None
+
+    def sampler(biased_logits, *_):
+        return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
+
+    token_ids, attention_mask = (
+        torch.tensor([[0, 1, 2, 3]]),
+        torch.tensor([[1, 1, 1, 1]]),
+    )
+    init_fsm_states = [0]
+    sequence = sequence_generator(
+        MockModel(),
+        sampler,
+        [MockFSM()],
+        token_ids,
+        torch.tensor([0]),
+        attention_mask,
+        init_fsm_states,
+        rng=torch.Generator(),
+    )
+    result = next(sequence)
+
+    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3]]))
+    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
+
+    with pytest.raises(StopIteration):
+        next(sequence)
+
+
+def test_sequence_generator_1d_several_iterations():
+    class MockFSM:
+        def get_next_state(self, state, next_token_ids):
+            return state + 1
+
+        def get_next_instruction(self, _):
+            return Generate([0, 1, 2, 3])
+
+        def is_final_state(self, state):
+            if state < 2:
+                return False
+            else:
+                return True
+
+        def copy(self):
+            return self
+
+    class MockTokenizer:
+        def encode(self, _):
+            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
+
+        def decode(self, x):
+            return x
+
+    class MockModel:
+        def __init__(self):
+            self.tokenizer = MockTokenizer()
+
+        def __call__(*_):
+            return torch.tensor([[0, 1, 2, 3]], dtype=torch.float), None
+
+    def sampler(biased_logits, *_):
+        return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
+
+    token_ids, attention_mask = (
+        torch.tensor([[0, 1, 2, 3]]),
+        torch.tensor([[1, 1, 1, 1]]),
+    )
+    init_fsm_states = [0]
+    sequence = sequence_generator(
+        MockModel(),
+        sampler,
+        [MockFSM()],
+        token_ids,
+        torch.tensor([0]),
+        attention_mask,
+        init_fsm_states,
+        rng=torch.Generator(),
+    )
+
+    result = next(sequence)
+    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3]]))
+    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
+
+    result = next(sequence)
+    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3, 3]]))
+    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
+
+    with pytest.raises(StopIteration):
+        next(sequence)
+
+
+def test_sequence_generator_2d_single_iteration():
+    class MockFSM:
+        def get_next_state(self, state, next_token_ids):
+            return 0
+
+        def get_next_instruction(self, _):
+            return Generate([0, 1, 2, 3])
+
+        def is_final_state(self, _):
+            return True
+
+        def copy(self):
+            return self
+
+    class MockTokenizer:
+        def encode(self, _):
+            return torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]), torch.tensor(
+                [[1, 1, 1, 1], [1, 1, 1, 1]]
+            )
+
+        def decode(self, x):
+            return x
+
+    class MockModel:
+        def __init__(self):
+            self.tokenizer = MockTokenizer()
+
+        def __call__(*_):
+            return torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float), None
+
+    def sampler(biased_logits, *_):
+        return (
+            torch.argmax(biased_logits, keepdims=True, dim=-1),
+            torch.tensor([0, 1]),
+            None,
+        )
+
+    token_ids, attention_mask = (
+        torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]),
+        torch.tensor([[1, 1, 1, 1], [1, 1, 1, 1]]),
+    )
+    init_fsm_states = [0, 0]
+    fsms = [MockFSM(), MockFSM()]
+    sequence = sequence_generator(
+        MockModel(),
+        sampler,
+        fsms,
+        token_ids,
+        torch.tensor([0, 0]),
+        attention_mask,
+        init_fsm_states,
+        rng=torch.Generator(),
+    )
+
+    result = next(sequence)
+    assert torch.equal(
+        result.token_ids, torch.tensor([[0, 1, 2, 3, 3], [4, 5, 6, 7, 2]])
+    )
+    assert torch.equal(
+        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
+    )
+
+    with pytest.raises(StopIteration):
+        next(sequence)
+
+
+def test_sequence_generator_2d_several_iterations():
+    class MockFSM:
+        def get_next_state(self, state, next_token_ids):
+            return state + 1
+
+        def get_next_instruction(self, _):
+            return Generate([0, 1, 2, 3])
+
+        def is_final_state(self, state):
+            if state < 2:
+                return False
+            else:
+                return True
+
+        def copy(self):
+            return self
+
+    class MockTokenizer:
+        def encode(self, _):
+            return torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]), torch.tensor(
+                [[1, 1, 1, 1], [1, 1, 1, 1]]
+            )
+
+        def decode(self, x):
+            return x
+
+    class MockModel:
+        def __init__(self):
+            self.tokenizer = MockTokenizer()
+
+        def __call__(*_):
+            return torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float), None
+
+    def sampler(biased_logits, *_):
+        return (
+            torch.argmax(biased_logits, keepdims=True, dim=-1),
+            torch.tensor([0, 1]),
+            None,
+        )
+
+    token_ids, attention_mask = (
+        torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]),
+        torch.tensor([[1, 1, 1, 1], [1, 1, 1, 1]]),
+    )
+    init_fsm_states = [0, 0]
+    fsms = [MockFSM(), MockFSM()]
+    sequence = sequence_generator(
+        MockModel(),
+        sampler,
+        fsms,
+        token_ids,
+        torch.tensor([0, 0]),
+        attention_mask,
+        init_fsm_states,
+        rng=torch.Generator(),
+    )
+
+    result = next(sequence)
+    assert torch.equal(
+        result.token_ids, torch.tensor([[0, 1, 2, 3, 3], [4, 5, 6, 7, 2]])
+    )
+    assert torch.equal(
+        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
+    )
+
+    result = next(sequence)
+    assert torch.equal(
+        result.token_ids, torch.tensor([[0, 1, 2, 3, 3, 3], [4, 5, 6, 7, 2, 2]])
+    )
+    assert torch.equal(
+        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
+    )
+
+    with pytest.raises(StopIteration):
+        next(sequence)
+
+
+def test_get_next_fsm_states():
+    class MockFSM:
+        def get_next_state(self, state, next_token_ids):
+            return 0
+
+        def copy(self):
+            return self
+
+    result = get_next_fsm_states([MockFSM()], [0], torch.tensor([[0]]))
+    assert result == [0]
+
+    result = get_next_fsm_states(
+        [MockFSM(), MockFSM()], [0, 0], torch.tensor([[0], [0]])
+    )
+    assert result == [0, 0]
+
+
+def test_get_get_next_instructions():
+    class MockFSM:
+        def get_next_instruction(self, _):
+            return Generate([1, 2, 3, 4])
+
+    result = get_allowed_tokens([MockFSM()], [0])
+    assert result == [[1, 2, 3, 4]]
+
+    result = get_allowed_tokens([MockFSM(), MockFSM()], [0, 1])
+    assert result == [[1, 2, 3, 4], [1, 2, 3, 4]]
+
+
+def test_is_generation_finished():
+    class MockFSMFinished:
+        def is_final_state(self, _):
+            return True
+
+    result = is_generation_finished([MockFSMFinished(), MockFSMFinished()], [1, 1])
+    assert result is True
+
+    class MockFSMNotFinished:
+        def is_final_state(self, state):
+            if state == 0:
+                return False
+            else:
+                return True
+
+    result = is_generation_finished(
+        [MockFSMNotFinished(), MockFSMNotFinished()], [0, 1]
+    )
+    assert result is False
+
+
+@pytest.mark.parametrize(
+    "token_ids,next_token_ids,ancestors,expected_result",
+    [
+        (
+            torch.tensor([[1]]),
+            torch.tensor([[2]]),
+            torch.tensor([0]),
+            torch.tensor([[1, 2]]),
+        ),
+        (
+            torch.tensor([[1], [2]]),
+            torch.tensor([[3], [4]]),
+            torch.tensor([0, 1]),
+            torch.tensor([[1, 3], [2, 4]]),
+        ),
+        (
+            torch.tensor([[1], [2]]),
+            torch.tensor([[3], [4]]),
+            torch.tensor([0, 0]),
+            torch.tensor([[1, 3], [1, 4]]),
+        ),
+        (
+            torch.tensor([[1], [2]]),
+            torch.tensor([[3], [4]]),
+            torch.tensor([1, 0]),
+            torch.tensor([[2, 3], [1, 4]]),
+        ),
+        (
+            torch.tensor([[1, 2], [3, 5]]),
+            torch.tensor([[3], [4]]),
+            torch.tensor([1, 0]),
+            torch.tensor([[3, 5, 3], [1, 2, 4]]),
+        ),
+    ],
+)
+def test_update_token_ids(token_ids, next_token_ids, ancestors, expected_result):
+    result = update_token_ids(token_ids, next_token_ids, ancestors)
+    assert torch.equal(result, expected_result)
+
+
+@pytest.mark.parametrize(
+    "attention_masks,ancestors,expected_result",
+    [
+        (
+            torch.tensor([[1, 1]], dtype=torch.float),
+            torch.tensor([0]),
+            torch.tensor([[1, 1, 1]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[1, 1], [1, 1]], dtype=torch.float),
+            torch.tensor([0, 0]),
+            torch.tensor([[1, 1, 1], [1, 1, 1]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
+            torch.tensor([0, 1]),
+            torch.tensor([[0, 1, 1], [1, 1, 1]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
+            torch.tensor([1, 0]),
+            torch.tensor([[1, 1, 1], [0, 1, 1]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
+            torch.tensor([1, 1]),
+            torch.tensor([[1, 1, 1], [1, 1, 1]], dtype=torch.float),
+        ),
+    ],
+)
+def test_expand_attention_masks(attention_masks, ancestors, expected_result):
+    result = update_attention_masks(attention_masks, ancestors)
+    assert torch.equal(result, expected_result)
+
+
+@pytest.mark.parametrize(
+    "logits,indices_to_mask,expected",
+    [
+        (
+            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
+            [[0, 1, 2, 3]],
+            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
+            [[0, 2, 3]],
+            torch.tensor([[1, -math.inf, 3, 4]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
+            [[0, 2]],
+            torch.tensor([[1, -math.inf, 3, -math.inf]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[1, 2, 3], [4, 5, 6]], dtype=torch.float),
+            [[1, 2], [0, 1]],
+            torch.tensor([[-math.inf, 2, 3], [4, 5, -math.inf]], dtype=torch.float),
+        ),
+        (
+            torch.tensor([[1, 2, 3], [4, 5, 6]], dtype=torch.float),
+            [[0, 2], [1]],
+            torch.tensor(
+                [[1, -math.inf, 3], [-math.inf, 5, -math.inf]], dtype=torch.float
+            ),
+        ),
+    ],
+)
+def test_bias_logits(logits, indices_to_mask, expected):
+    masked_logits = bias_logits(logits, indices_to_mask)
+    assert torch.equal(masked_logits, expected)
diff --git a/tests/generate/test_integration_exllamav2.py b/tests/generate/test_integration_exllamav2.py
new file mode 100644
index 000000000..12c4143b3
--- /dev/null
+++ b/tests/generate/test_integration_exllamav2.py
@@ -0,0 +1,363 @@
+import importlib
+from unittest.mock import patch
+
+import pytest
+
+import outlines.models as models
+from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.models.exllamav2 import ExLlamaV2Model
+
+
+@pytest.fixture(scope="session")
+def model_exllamav2(tmp_path_factory):
+    return models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=False,
+    )
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_import_error(request, model_fixture):
+    with patch.dict("sys.modules", {"exllamav2": None}):
+        with pytest.raises(ImportError):
+            models.exl2(
+                model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+                cache_q4=True,
+                paged=False,
+            )
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_attributes(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    assert hasattr(model, "generator")
+    assert hasattr(model, "tokenizer")
+    assert model.tokenizer.convert_token_to_string(1) == 1
+    assert hasattr(model, "max_seq_len")
+    assert isinstance(model.max_seq_len, int)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_prompt_types(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+    prompt = ["test"]
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_no_max_tokens(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_test_stop_at(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_generate_multisampling(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    output = model.generate(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    assert isinstance(output, list)
+    assert isinstance(output[0], str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_prepare_generation_parameters(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    exllamav2_params, prompts = model.prepare_generation_parameters(
+        prompt, generation_params, sampling_params, structure_logits_processor
+    )
+    assert isinstance(exllamav2_params, dict)
+    assert isinstance(prompts, list)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_prompt_types(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+    prompt = ["test"]
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_no_max_tokens(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_test_stop_at(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_multisampling(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, list)
+        assert isinstance(token[0], str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_model_stream_seed(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    prompt = "test"
+    generation_params = GenerationParameters(max_tokens=10, seed=1, stop_at=None)
+    structure_logits_processor = None
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+        0.9,
+        50,
+        1.0,
+    )
+    generator = model.stream(
+        prompt, generation_params, structure_logits_processor, sampling_params
+    )
+    for token in generator:
+        assert isinstance(token, str)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_reformat_output(request, model_fixture):
+    model = request.getfixturevalue(model_fixture)
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+    )
+    output = "test"
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert reformatted_output == output
+    output = ["test"]
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert reformatted_output == output[0]
+    output = ["test", "test"]
+    sampling_params = SamplingParameters(
+        "multinomial",
+        1,
+    )
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert len(reformatted_output) == 2
+    assert reformatted_output[0] == "test"
+    assert reformatted_output[1] == "test"
+    output = ["test", "test"]
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert len(reformatted_output) == 2
+    assert reformatted_output[0] == "test"
+    assert reformatted_output[1] == "test"
+    output = ["test", "test", "test", "test"]
+    sampling_params = SamplingParameters(
+        "multinomial",
+        2,
+    )
+    reformatted_output = model.reformat_output(output, sampling_params)
+    assert len(reformatted_output) == 2
+    assert reformatted_output[0] == ["test", "test"]
+    assert reformatted_output[1] == ["test", "test"]
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_max_chunk_size(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=False,
+        max_chunk_size=128,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_cache_default(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        paged=False,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def is_flash_attn_available():
+    try:
+        importlib.import_module("flash_attn")
+    except (ImportError, AssertionError):
+        return False
+    return True
+
+
+@pytest.mark.skipif(not is_flash_attn_available(), reason="flash-attn is not installed")
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_paged(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=True,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_draft_model(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        cache_q4=True,
+        paged=False,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_draft_model_cache_default(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        paged=False,
+    )
+    assert isinstance(model, ExLlamaV2Model)
+
+
+@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
+def test_exl2_set_max_seq_len(request, model_fixture):
+    model = models.exl2(
+        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+        max_seq_len=2048,
+        paged=False,
+        cache_q4=True,
+    )
+    assert isinstance(model, ExLlamaV2Model)
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
new file mode 100644
index 000000000..fd5be2171
--- /dev/null
+++ b/tests/generate/test_integration_llamacpp.py
@@ -0,0 +1,374 @@
+import datetime
+import re
+
+import pytest
+from pydantic import BaseModel, constr
+
+import outlines.generate as generate
+import outlines.models as models
+import outlines.samplers as samplers
+
+TEST_MODEL = "./llama-test-model/TinyMistral-248M-v2-Instruct.Q4_K_M.gguf"
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    return models.llamacpp(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+
+
+@pytest.mark.parametrize(
+    "generator_type,params",
+    (
+        (generate.text, []),
+        (generate.regex, ("[0-9]",)),
+        # (generate.cfg, (grammars.arithmetic,)),  # Awaiting CFG fix
+    ),
+)
+def test_llamacpp_generation_api(model, generator_type, params):
+    generator = generator_type(model, *params)
+
+    res = generator("test", max_tokens=10)
+    assert isinstance(res, str)
+
+    res = generator("test", max_tokens=10)
+    assert isinstance(res, str)
+
+    res = generator("test", stop_at=".")
+    assert isinstance(res, str)
+
+    res = generator("test", stop_at=[".", "ab"])
+    assert isinstance(res, str)
+
+    res = generator("test", stop_at=[".", "ab"])
+    assert isinstance(res, str)
+
+    res1 = generator("test", seed=1, max_tokens=10)
+    res2 = generator("test", seed=1, max_tokens=10)
+    assert isinstance(res1, str)
+    assert isinstance(res2, str)
+    assert res1 == res2
+
+
+def test_llama_cpp_streaming_api(model):
+    generator = generate.text(model)
+    token_generator = generator.stream("test", max_tokens=10)
+    tokens = [token for token in token_generator]
+    assert len(tokens) <= 10
+    assert isinstance(tokens[0], str)
+
+
+@pytest.mark.xfail(reason="Batch inference is not available in `llama-cpp-python`.")
+def test_llamacpp_batch_inference(model):
+    generator = generate.text(model)
+    res = generator(["test", "test1"])
+    assert len(res) == 2
+
+
+def test_llamacpp_sampling_params(model):
+    generator = generate.text(model)
+
+    params = {
+        "frequency_penalty": 1.0,
+        "presence_penalty": 1.0,
+    }
+    res = generator("test", seed=1, max_tokens=10, **params)
+    assert isinstance(res, str)
+
+
+def test_llamacpp_greedy_sampling(model):
+    sampler = samplers.greedy()
+    generator = generate.text(model, sampler)
+    res = generator("test", max_tokens=20)
+    assert isinstance(res, str)
+
+
+def test_llamacpp_multinomial_sampling(model):
+    sampler = samplers.multinomial()
+    generator = generate.text(model, sampler)
+    res = generator("test", max_tokens=10)
+    assert isinstance(res, str)
+
+    sampler = samplers.multinomial(1, temperature=1.0)
+    generator = generate.text(model, sampler)
+    res = generator("test", max_tokens=10)
+    assert isinstance(res, str)
+
+    sampler = samplers.multinomial(1, top_k=1)
+    generator = generate.text(model, sampler)
+    res = generator("test", max_tokens=10)
+    assert isinstance(res, str)
+
+    sampler = samplers.multinomial(1, top_p=0.5)
+    generator = generate.text(model, sampler)
+    res = generator("test", max_tokens=10)
+    assert isinstance(res, str)
+
+
+def test_llamacpp_several_samples(model):
+    sampler = samplers.multinomial(3)
+    generator = generate.text(model, sampler)
+    with pytest.raises(NotImplementedError, match="allow to take several samples"):
+        generator("test")
+
+
+def test_llamacpp_beam_search(model):
+    sampler = samplers.beam_search(1)
+    generator = generate.text(model, sampler)
+
+    with pytest.raises(NotImplementedError, match="does not support Beam Search"):
+        generator("test")
+
+
+def test_llamacpp_text_stop(model):
+    prompt = (
+        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
+    )
+    sequence = generate.text(model)(prompt, stop_at="a", max_tokens=100)
+    assert isinstance(sequence, str)
+    assert sequence.find("a") == -1
+
+
+def test_llamacpp_regex(model):
+    prompt = (
+        "<|im_start|>user\nWrite an email address<|im_end|>\n<|im_start|>assistant\n"
+    )
+    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
+    generator = generate.regex(model, regex_str)
+
+    # One prompt
+    sequence = generator(prompts=prompt)
+    assert isinstance(sequence, str)
+    assert re.fullmatch(pattern=regex_str, string=sequence) is not None
+
+
+def test_llamacpp_integer(model):
+    prompt = (
+        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
+    )
+    sequence = generate.format(model, int)(prompt, max_tokens=10)
+    assert isinstance(sequence, int)
+    assert sequence != ""
+    int(sequence)
+
+
+def test_llamacpp_float(model):
+    prompt = (
+        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
+    )
+    sequence = generate.format(model, float)(prompt, max_tokens=10)
+    assert isinstance(sequence, float)
+
+    assert sequence != ""
+    float(sequence)
+
+
+def test_llamacpp_bool(model):
+    prompt = (
+        "<|im_start|>user\nIs this True or False?<|im_end|>\n<|im_start|>assistant\n"
+    )
+    sequence = generate.format(model, bool)(prompt, max_tokens=10)
+    assert isinstance(sequence, bool)
+
+    assert sequence != ""
+    bool(sequence)
+
+
+def test_llamacpp_date(model):
+    prompt = (
+        "<|im_start|>user\nWhat day is it today?<|im_end|>\n<|im_start|>assistant\n"
+    )
+    sequence = generate.format(model, datetime.date)(prompt, max_tokens=20, seed=10)
+    assert isinstance(sequence, datetime.date)
+
+
+def test_llamacpp_time(model):
+    prompt = "<|im_start|>user\nWhat time is it?<|im_end|>\n<|im_start|>assistant\n"
+    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10)
+    assert isinstance(sequence, datetime.time)
+
+
+def test_llamacpp_datetime(model):
+    prompt = "<|im_start|>user\nWhat time is it?<|im_end|>\n<|im_start|>assistant\n"
+    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20)
+    assert isinstance(sequence, datetime.datetime)
+
+
+def test_llamacpp_choice(model):
+    prompt = (
+        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
+    )
+    sequence = generate.choice(model, ["test", "choice"])(prompt)
+    assert sequence == "test" or sequence == "choice"
+
+
+def test_llamacpp_json_basic(model):
+    prompt = "<|im_start|>user\nOutput some JSON<|im_end|>\n<|im_start|>assistant\n"
+
+    class Spam(BaseModel):
+        spam: constr(max_length=10)
+        fuzz: bool
+
+    result = generate.json(model, Spam, whitespace_pattern="")(
+        prompt, max_tokens=100, temperature=0.0, seed=1
+    )
+    assert isinstance(result, BaseModel)
+    assert isinstance(result.spam, str)
+    assert isinstance(result.fuzz, bool)
+    assert len(result.spam) <= 10
+
+
+def test_llamacpp_json_schema(model):
+    prompt = "<|im_start|>user\nOutput some JSON<|im_end|>\n<|im_start|>assistant\n"
+
+    schema = """{
+      "title": "spam",
+      "type": "object",
+      "properties": {
+           "foo" : {"type": "boolean"},
+           "bar": {"type": "string", "maxLength": 4}
+        },
+      "required": ["foo", "bar"]
+      }
+    """
+
+    result = generate.json(model, schema, whitespace_pattern="")(
+        prompt, max_tokens=100, temperature=0, seed=10
+    )
+    assert isinstance(result, dict)
+    assert isinstance(result["foo"], bool)
+    assert isinstance(result["bar"], str)
+
+
+@pytest.mark.parametrize(
+    "repo,model_path,hf_tokenizer_uri",
+    [
+        ("Qwen/Qwen1.5-0.5B-Chat-GGUF", "*q2*.gguf", "Qwen/Qwen1.5-0.5B-Chat"),
+        ("TheBloke/phi-2-GGUF", "*Q2*.gguf", "microsoft/phi-2"),
+    ],
+)
+def test_byte_tokenizer_regression(repo, model_path, hf_tokenizer_uri):
+    """Reproduce https://github.com/dottxt-ai/outlines/issues/820"""
+    import llama_cpp
+
+    model = models.llamacpp(
+        repo,
+        model_path,
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            hf_tokenizer_uri
+        ),
+    )
+    generator = generate.choice(model, ["skirt", "dress", "pen", "jacket"])
+    generator("Pick the odd word out: skirt, dress, pen, jacket")
+
+
+def test_llama_cpp_pre_tokenizer_remains_broken():
+    """If fails, llama.cpp pre-tokenizer is fixed -> revert #892, remove `with pytest.raises`"""
+    repo = "Qwen/Qwen1.5-0.5B-Chat-GGUF"
+    model_path = "*q2*.gguf"
+
+    model = models.llamacpp(repo, model_path)
+    with pytest.raises(RuntimeError):
+        generate.choice(model, ["skirt", "dress", "pen", "jacket"])
+
+
+@pytest.mark.skip("Caching for guide was temporarily turned off")
+def test_RegexGuide_caching(model, temp_cache_dir):
+    import llama_cpp
+
+    import outlines.caching
+    from outlines.fsm.guide import cached_create_states_mapping
+
+    assert outlines.caching._caching_enabled
+
+    regex = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
+    prompt = "What is the IP address of the Google DNS servers? "
+
+    cache = outlines.caching.get_cache()
+
+    # Returns (hits, misses)
+    _ = cache.stats(enable=True)
+    assert cache.statistics
+
+    assert cached_create_states_mapping.__memory__ is cache
+
+    generator = generate.regex(model, regex, sampler=samplers.greedy())
+    assert cache.stats() == (0, 1)
+
+    model_2 = models.llamacpp(
+        "Qwen/Qwen1.5-0.5B-Chat-GGUF",
+        "*q2*.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "Qwen/Qwen1.5-0.5B-Chat"
+        ),
+    )
+    generator_2 = generate.regex(model_2, regex, sampler=samplers.greedy())
+    assert cache.stats() == (0, 2)
+
+    # These two different models and tokenizers should not have the same state
+    # mapping results
+    assert (
+        generator.logits_processor.guide.states_to_token_maps
+        != generator_2.logits_processor.guide.states_to_token_maps
+    )
+
+    generator_3 = generate.regex(model_2, regex, sampler=samplers.greedy())
+    assert cache.stats() == (1, 2)
+    assert (
+        generator_2.logits_processor.guide.states_to_token_maps
+        == generator_3.logits_processor.guide.states_to_token_maps
+    )
+
+    # Just for fun...
+    structured = generator(prompt, max_tokens=30)
+    structured_2 = generator_2(prompt, max_tokens=30)
+
+    assert re.fullmatch(regex, structured)
+    assert re.fullmatch(regex, structured_2)
+    assert structured != structured_2
+
+
+@pytest.mark.xfail(
+    reason="Some versions of the Hermes-2-Pro-Llama-3 model have a broken config"
+)
+def test_tokenizer_vocabulary_decode_sanity():
+    """Assert the decoded newline token (198) is the same as the normalized vocab token"""
+    import llama_cpp
+
+    model = models.llamacpp(
+        "bartowski/Meta-Llama-3-8B-Instruct-GGUF",
+        "Meta-Llama-3-8B-Instruct-IQ1_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B",
+        ),
+    )
+    tokenizer = generate.regex(model, "a").logits_processor.tokenizer
+
+    decoded_nl_token = tokenizer.decode([198])[0]
+    vocab_nl_token = tokenizer.convert_token_to_string(
+        [token for token, token_id in tokenizer.vocabulary.items() if token_id == 198][
+            0
+        ]
+    )
+    assert decoded_nl_token == vocab_nl_token
+
+
+def test_no_length_constraint_when_unset():
+    """Assert that models.llamacpp doesn't have an implicit max_tokens preventing full sequence generation"""
+    import llama_cpp
+
+    model = models.llamacpp(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "Locutusque/TinyMistral-248M-Instruct"
+        ),
+    )
+
+    long_pattern = "abcdefg" * 10
+    generator = generate.regex(model, long_pattern)
+
+    output = generator("a")
+    assert re.match(long_pattern, output)
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
new file mode 100644
index 000000000..92c5d789c
--- /dev/null
+++ b/tests/generate/test_integration_transformers.py
@@ -0,0 +1,569 @@
+import datetime
+import re
+from enum import Enum
+from functools import partial
+from typing import List, Union
+
+import pytest
+import torch
+from outlines_core.fsm.regex import reduced_vocabulary
+from pydantic import BaseModel, constr
+
+import outlines.generate as generate
+import outlines.models as models
+from outlines.models.transformers import Transformers, TransformerTokenizer
+from outlines.samplers import beam_search, greedy, multinomial
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    return models.transformers(
+        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
+    )
+
+
+def test_transformers_integration_text(model):
+    sequence = generate.text(model)(
+        "Write a short sentence ", seed=10000, max_tokens=10
+    )
+    assert isinstance(sequence, str)
+    assert model.tokenizer.eos_token not in sequence
+
+    sequence = generate.text(model)(
+        "Write a short sentence ",
+        max_tokens=20,
+        stop_at="a",
+        seed=10000,
+    )
+    assert isinstance(sequence, str)
+
+    prompts = ["Write a short sentence ", "And another one "]
+    sequence = generate.text(model)(
+        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
+    )
+    assert isinstance(sequence, list)
+    assert len(sequence) == 2
+    assert isinstance(sequence[0], str)
+
+
+def test_transformers_integration_text_multiple_samples(model):
+    sampler = multinomial(2)
+
+    sequence = generate.text(model, sampler=sampler)(
+        "Write a short sentence ", seed=10000
+    )
+    assert isinstance(sequence, list)
+    assert len(sequence) == 2
+    assert model.tokenizer.eos_token not in sequence
+
+    prompts = ["Write a short sentence ", "And another one "]
+    sequence = generate.text(model, sampler=sampler)(
+        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
+    )
+    assert isinstance(sequence, list)
+    assert len(sequence) == 2
+    assert isinstance(sequence[0], list)
+    assert len(sequence) == 2
+    assert isinstance(sequence[0][0], str)
+
+
+def test_transformers_integration_streaming(model):
+    sequence = generate.text(model).stream(
+        "Write a short sentence ", max_tokens=10, stop_at=[".", ","], seed=10000
+    )
+
+    token = next(sequence)
+    assert isinstance(token, str)
+
+    remaining = "".join([token for token in sequence])
+    assert isinstance(remaining, str)
+
+    sequence = generate.text(model).stream(
+        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
+    )
+    tokens = next(sequence)
+    assert isinstance(tokens, list)
+    assert isinstance(tokens[0], str)
+    assert isinstance(tokens[1], str)
+
+
+def test_transformers_integration_streaming_batch_samples(model):
+    sampler = multinomial(samples=2)
+
+    sequence = generate.text(model, sampler=sampler).stream(
+        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
+    )
+    tokens = next(sequence)
+    assert isinstance(tokens, list)
+    assert len(tokens) == 2
+    assert isinstance(tokens[0], list)
+    assert len(tokens[0]) == 2
+    assert isinstance(tokens[0], list)
+    assert len(tokens[1]) == 2
+
+
+def test_transformers_integration_streaming_batch_beam_search(model):
+    sampler = beam_search(beams=2)
+
+    sequence = generate.regex(model, r"ab[cd]e", sampler=sampler).stream(
+        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=["c", "d"], seed=10000
+    )
+    tokens = next(sequence)
+    assert isinstance(tokens, list)
+    assert len(tokens) == 2
+    assert isinstance(tokens[0], list)
+    assert len(tokens[0]) == 2
+    assert isinstance(tokens[0], list)
+    assert len(tokens[1]) == 2
+
+
+def test_transformers_integration_text_stop(model):
+    prompt = "Write a short sentence "
+    sequence = generate.text(model)(prompt, stop_at="a", seed=10000)
+    assert sequence[len(prompt) :].find("a") == -1
+
+
+def test_transformers_various_regexes(model):
+    prompt = "Write an email address"
+    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
+    generator = generate.regex(model, regex_str)
+
+    # One prompt
+    sequence = generator(prompt, seed=0)
+    assert re.fullmatch(regex_str, sequence) is not None
+
+
+def test_transformers_various_regexes_prompt_list(model):
+    prompt = "Write an email address"
+    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
+    generator = generate.regex(model, regex_str)
+
+    # Two prompts
+    sequence = generator([prompt, prompt], seed=0)
+    assert re.fullmatch(regex_str, sequence[0]) is not None
+    assert re.fullmatch(regex_str, sequence[1]) is not None
+
+
+def test_transformers_various_regexes_prompt_list_multiple_samples(model):
+    sampler = multinomial(samples=2)
+    prompt = "Write an email address"
+    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
+    generator = generate.regex(model, regex_str, sampler=sampler)
+
+    # Two prompts
+    sequence = generator([prompt, prompt], seed=0, max_tokens=500)
+    assert isinstance(sequence, list)
+    assert len(sequence) == 2
+    assert re.fullmatch(regex_str, sequence[0][0]) is not None
+    assert re.fullmatch(regex_str, sequence[0][1]) is not None
+    assert re.fullmatch(regex_str, sequence[1][0]) is not None
+    assert re.fullmatch(regex_str, sequence[1][1]) is not None
+
+
+def test_transformers_various_regexes_prompt_list_beam_search(model):
+    sampler = beam_search(5)
+    prompt_1 = "Write an email address"
+    prompt_2 = "Random"
+    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
+    generator = generate.regex(model, regex_str, sampler=sampler)
+
+    # Two prompts
+    sequence = generator([prompt_1, prompt_2], seed=0)
+    assert isinstance(sequence, list)
+    assert len(sequence) == 2
+    assert len(sequence[0]) == 5
+    assert re.fullmatch(regex_str, sequence[0][0]) is not None
+    assert re.fullmatch(regex_str, sequence[0][1]) is not None
+    assert re.fullmatch(regex_str, sequence[1][0]) is not None
+    assert re.fullmatch(regex_str, sequence[1][1]) is not None
+
+
+def test_transformers_integration_integer(model):
+    prompt = "Write a short sentence"
+    sequence = generate.format(model, int)(prompt, max_tokens=10, seed=0)
+
+    assert isinstance(sequence, int)
+
+
+def test_transformers_integration_integer_array(model):
+    prompts = ["Give me a number", "And another one"]
+    sequence = generate.format(model, int)(prompts, max_tokens=10, seed=0)
+    assert isinstance(sequence, list)
+    assert len(sequence) == 2
+    assert isinstance(sequence[0], int)
+    assert isinstance(sequence[1], int)
+
+
+def test_transformers_integration_float(model):
+    prompt = "Write a short sentence"
+    sequence = generate.format(model, float)(prompt, max_tokens=10, seed=0)
+
+    assert sequence != ""
+    assert isinstance(sequence, float)
+
+
+def test_transformers_integration_bool(model):
+    prompt = "Is this True or False?"
+    sequence = generate.format(model, bool)(prompt, max_tokens=10, seed=0)
+
+    assert sequence != ""
+    assert isinstance(sequence, bool)
+
+
+def test_transformers_integration_date(model):
+    prompt = "What day is it today?"
+    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10, seed=0)
+
+    assert sequence != ""
+    assert isinstance(sequence, datetime.date)
+
+
+def test_transformers_integration_time(model):
+    prompt = "What time is it?"
+    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10, seed=0)
+
+    assert sequence != ""
+    assert isinstance(sequence, datetime.time)
+
+
+def test_transformers_integration_datetime(model):
+    prompt = "What time is it?"
+    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20, seed=0)
+
+    assert sequence != 0
+    assert isinstance(sequence, datetime.datetime)
+
+
+def test_transformers_integration_choice(model):
+    prompt = "Write a short sentence "
+    sequence = generate.choice(model, ["test", "choice"])(prompt, seed=0)
+
+    assert sequence == "test" or sequence == "choice"
+
+
+def test_transformers_integration_with_pad_token():
+    model_name = "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM"
+    model = models.transformers(model_name, device="meta")
+    assert model.tokenizer.pad_token_id == 1
+    assert model.tokenizer.pad_token == "<pad>"
+
+
+def test_transformers_json_basic(model):
+    prompt = "Output some JSON "
+
+    class Spam(BaseModel):
+        foo: int
+        bar: float
+        spam: constr(max_length=10)
+        fuzz: bool
+
+    result = generate.json(model, Spam)(prompt, max_tokens=500, seed=0)
+    assert isinstance(result, BaseModel)
+    assert isinstance(result.foo, int)
+    assert isinstance(result.bar, float)
+    assert isinstance(result.spam, str)
+    assert isinstance(result.fuzz, bool)
+    assert len(result.spam) <= 10
+
+
+def test_transformers_json_schema(model):
+    prompt = "Output some JSON "
+
+    schema = """{
+      "title": "spam",
+      "type": "object",
+      "properties": {
+           "foo" : {"type": "integer"},
+           "bar": {"type": "string", "maxLength": 4}
+        },
+      "required": ["foo", "bar"]
+      }
+    """
+
+    result = generate.json(model, schema)(prompt, max_tokens=500, seed=0)
+    assert isinstance(result, dict)
+    assert isinstance(result["foo"], int)
+    assert isinstance(result["bar"], str)
+
+
+def test_transformers_json_batch(model):
+    prompts = ["Output some JSON ", "Output more JSON"]
+
+    class Spam(BaseModel):
+        foo: int
+        bar: float
+        spam: constr(max_length=10)
+        fuzz: bool
+
+    result = generate.json(model, Spam)(prompts, max_tokens=500, seed=0)
+    assert isinstance(result[0], BaseModel)
+    assert isinstance(result[1], BaseModel)
+
+
+def test_transformers_json_batch_multiple_samples(model):
+    sampler = multinomial(samples=2)
+    prompts = ["Output some JSON ", "Output more JSON"]
+
+    class Spam(BaseModel):
+        foo: int
+        bar: float
+        spam: constr(max_length=10)
+        fuzz: bool
+
+    result = generate.json(model, Spam, sampler=sampler)(
+        prompts, max_tokens=500, seed=0
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0][0], BaseModel)
+    assert isinstance(result[0][1], BaseModel)
+    assert isinstance(result[1][0], BaseModel)
+    assert isinstance(result[1][1], BaseModel)
+
+
+def test_transformers_json_str_enum(model):
+    prompt = "Output some JSON "
+
+    class Name(str, Enum):
+        john = "John"
+        marc = "Marc"
+        michel = "Michel"
+
+    class User(BaseModel):
+        user_id: int
+        name: Name
+
+    result = generate.json(model, User)(prompt, seed=0)
+    assert isinstance(result, BaseModel)
+    assert isinstance(result.user_id, int)
+    assert result.name in ["John", "Marc", "Michel"]
+
+
+def test_transformers_json_int_enum(model):
+    prompt = "Output some JSON "
+
+    class Id(int, Enum):
+        one = 1
+        two = 2
+
+    class User(BaseModel):
+        user_id: Id
+
+    result = generate.json(model, User)(prompt, seed=0)
+    assert isinstance(result, BaseModel)
+    assert isinstance(result.user_id, int)
+    assert result.user_id in [1, 2]
+
+
+def add(a: int, b: int) -> int:
+    return a + b
+
+
+def mul(c: float, d: float) -> float:
+    return c * d
+
+
+def test_transformers_json_function_enum(model):
+    prompt = "Output some JSON "
+
+    class Operation(Enum):
+        add = partial(add)
+        mul = partial(mul)
+
+    result = generate.json(model, Operation)(prompt, seed=0)
+    assert isinstance(result, dict)
+    assert len(result) == 2
+    for k, v in result.items():
+        assert k in ["a", "b", "c", "d"]
+        assert isinstance(v, (int, float))
+
+
+def test_transformers_json_array(model):
+    prompt = "Output some JSON "
+
+    class User(BaseModel):
+        user_id: int
+        value: List[float]
+
+    result = generate.json(model, User)(prompt, seed=0)
+    assert isinstance(result, BaseModel)
+    assert isinstance(result.user_id, int)
+    assert isinstance(result.value, list)
+    for value in result.value:
+        assert isinstance(value, float) or isinstance(value, int)
+
+
+@pytest.mark.xfail(reason="The implementation of `anyOf` is incorrect")
+def test_transformers_json_union(model):
+    prompt = "Output some JSON "
+
+    class Spam(BaseModel):
+        foo: int
+        bar: Union[constr(max_length=10), float]
+
+    result = generate.json(model, Spam)(prompt, max_tokens=100, seed=4)
+    assert isinstance(result, BaseModel)
+    assert (
+        isinstance(result.bar, int)
+        or isinstance(result.bar, float)
+        or isinstance(result.bar, str)
+    )
+
+
+def test_transformers_json_function(model):
+    prompt = "Output arguments for the function"
+
+    def function(foo: int, bar: List[int]):
+        return foo + sum(bar)
+
+    sequence = generate.json(model, function)(prompt, max_tokens=100, seed=4)
+    assert isinstance(sequence, dict)
+    assert isinstance(function(**sequence), int)
+
+
+def test_transformers_logits_vocab_size(model):
+    # Artificially increase the weights/logits size relative
+    # to the vocabulary
+    model.model.resize_token_embeddings(pad_to_multiple_of=3)
+
+    assert len(model.tokenizer.vocabulary) == 1024
+    assert model.model.base_model.wte.weight.shape[0] == 1026
+
+    generator = generate.choice(model, ["True", "False"])
+
+    sequence = generator("blah", seed=101)
+    assert sequence == "False"
+
+
+def test_transformers_json_custom_ws(model):
+    prompt = "Output some JSON with newlines"  # try to force model to use newlines
+
+    schema = """{
+      "title": "spam",
+      "type": "object",
+      "properties": {
+           "foo" : {"type": "integer"},
+           "bar": {"type": "integer"}
+        },
+      "required": ["foo", "bar"]
+      }
+    """
+
+    generator = generate.json(model, schema, whitespace_pattern=r"[ ]?")
+    generator.format_sequence = lambda x: x  # patch to return raw text
+    assert "\n" not in generator(prompt, max_tokens=500, seed=0)
+
+
+def test_transformers_reduced_vocabulary_caching():
+    from transformers import AutoTokenizer
+
+    hf_tokenizer = AutoTokenizer.from_pretrained("gpt2")
+    tokenizer = TransformerTokenizer(hf_tokenizer)
+    tokenizer2 = TransformerTokenizer(hf_tokenizer)
+
+    # TODO: We might actually want only one copy of a given tokenizer.
+    assert tokenizer is not tokenizer2
+
+    vocab = reduced_vocabulary(tokenizer)
+    vocab2 = reduced_vocabulary(tokenizer2)
+
+    assert vocab2 is vocab
+
+
+@pytest.mark.skip(reason="Custom Sampler Disabled in Transformers Integration")
+def test_custom_sampler(model):
+    seen = False
+    target_token_ids = model.tokenizer.encode(["c"])[0]
+
+    class biased_sampler:
+        def __init__(self, samples: int = 1):
+            self.samples = samples
+
+        def __call__(
+            logits: torch.DoubleTensor, samples: int, *_
+        ) -> torch.DoubleTensor:
+            nonlocal seen
+
+            if not seen:
+                seen = True
+                return target_token_ids, torch.tensor([0]), None
+            else:
+                return (
+                    torch.tensor([[model.tokenizer.eos_token_id]]),
+                    torch.tensor([0]),
+                    None,
+                )
+
+    generator = generate.choice(model, ["a", "b", "c"], sampler=biased_sampler(1))
+    sequence = generator(
+        """What is 1+1?
+    a. 3
+    b. 4
+    c. 2"""
+    )
+
+    assert sequence == "c"
+
+
+def test_transformers_use_existing_model_and_tokenizer():
+    from transformers import AutoModelForCausalLM, AutoTokenizer
+
+    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
+    hf_tokenizer = AutoTokenizer.from_pretrained(model_name)
+    hf_model = AutoModelForCausalLM.from_pretrained(model_name)
+    model = Transformers(hf_model, hf_tokenizer)
+    sequence = generate.text(model)("Write a short sentence ", seed=10000)
+    assert isinstance(sequence, str)
+
+
+@pytest.mark.skip("Caching for guide was temporarily turned off")
+def test_RegexGuide_caching(temp_cache_dir):
+    import outlines.caching
+    from outlines.fsm.guide import cached_create_states_mapping
+
+    assert outlines.caching._caching_enabled
+
+    regex = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
+    prompt = "What is the IP address of the Google DNS servers? "
+
+    cache = outlines.caching.get_cache()
+
+    # Returns (hits, misses)
+    _ = cache.stats(enable=True)
+    assert cache.statistics
+
+    assert cached_create_states_mapping.__memory__ is cache
+
+    model = models.transformers(
+        "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM", device="cpu"
+    )
+    generator = generate.regex(model, regex, sampler=greedy())
+    assert cache.stats() == (0, 1)
+
+    model_2 = models.transformers(
+        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
+    )
+    generator_2 = generate.regex(model_2, regex, sampler=greedy())
+    assert cache.stats() == (0, 2)
+
+    # These two different models and tokenizers should not have the same state
+    # mapping results
+    assert (
+        generator.logits_processor.guide.states_to_token_maps
+        != generator_2.logits_processor.guide.states_to_token_maps
+    )
+
+    generator_3 = generate.regex(model_2, regex, sampler=greedy())
+    assert cache.stats() == (1, 2)
+    assert (
+        generator_2.logits_processor.guide.states_to_token_maps
+        == generator_3.logits_processor.guide.states_to_token_maps
+    )
+
+    # Just for fun...
+    structured = generator(prompt, max_tokens=30)
+    structured_2 = generator_2(prompt, max_tokens=30)
+
+    assert re.fullmatch(regex, structured)
+    assert re.fullmatch(regex, structured_2)
+    assert structured != structured_2
diff --git a/tests/generate/test_integration_transformers_vision.py b/tests/generate/test_integration_transformers_vision.py
new file mode 100644
index 000000000..ee4f84c06
--- /dev/null
+++ b/tests/generate/test_integration_transformers_vision.py
@@ -0,0 +1,116 @@
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image
+from transformers import AutoProcessor, LlavaForConditionalGeneration
+
+import outlines
+from outlines.models.transformers_vision import transformers_vision
+
+IMAGE_URLS = [
+    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
+    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
+    "https://upload.wikimedia.org/wikipedia/commons/b/be/Tamias-rufus-001.jpg",
+]
+
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    return transformers_vision(
+        "trl-internal-testing/tiny-LlavaForConditionalGeneration",
+        model_class=LlavaForConditionalGeneration,
+        device="cpu",
+    )
+
+
+@pytest.fixture(scope="session")
+def processor(tmp_path_factory):
+    return AutoProcessor.from_pretrained("llava-hf/llava-interleave-qwen-0.5b-hf")
+
+
+def test_single_image_text_gen(model, processor):
+    conversation = [
+        {
+            "role": "user",
+            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
+        },
+    ]
+    generator = outlines.generate.text(model)
+    sequence = generator(
+        processor.apply_chat_template(conversation),
+        [img_from_url(IMAGE_URLS[0])],
+        seed=10000,
+        max_tokens=10,
+    )
+    assert isinstance(sequence, str)
+
+
+def test_multi_image_text_gen(model, processor):
+    """If the length of image tags and number of images we pass are > 1 and equal,
+    we should yield a successful generation.
+    """
+    conversation = [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "What do all these have in common?"},
+            ]
+            + [{"type": "image"} for _ in range(len(IMAGE_URLS))],
+        },
+    ]
+    generator = outlines.generate.text(model)
+    sequence = generator(
+        processor.apply_chat_template(conversation),
+        [img_from_url(i) for i in IMAGE_URLS],
+        seed=10000,
+        max_tokens=10,
+    )
+    assert isinstance(sequence, str)
+
+
+def test_mismatched_image_text_gen(model, processor):
+    """If the length of image tags and number of images we pass are unequal,
+    we should raise an error.
+    """
+    conversation = [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "I'm passing 3 images, but only 1 image tag"},
+                {"type": "image"},
+            ],
+        },
+    ]
+    generator = outlines.generate.text(model)
+    with pytest.raises(ValueError):
+        _ = generator(
+            processor.apply_chat_template(conversation),
+            [img_from_url(i) for i in IMAGE_URLS],
+            seed=10000,
+            max_tokens=10,
+        )
+
+
+def test_single_image_choice(model, processor):
+    conversation = [
+        {
+            "role": "user",
+            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
+        },
+    ]
+    choices = ["cat", "dog"]
+    generator = outlines.generate.choice(model, choices)
+    sequence = generator(
+        processor.apply_chat_template(conversation),
+        [img_from_url(IMAGE_URLS[0])],
+        seed=10000,
+        max_tokens=10,
+    )
+    assert isinstance(sequence, str)
+    assert sequence in choices
diff --git a/tests/generate/test_integration_vllm.py b/tests/generate/test_integration_vllm.py
new file mode 100644
index 000000000..d812d3cf2
--- /dev/null
+++ b/tests/generate/test_integration_vllm.py
@@ -0,0 +1,243 @@
+import datetime
+import re
+
+import pytest
+import torch
+from pydantic import BaseModel, constr
+
+try:
+    from vllm.sampling_params import SamplingParams
+except ImportError:
+    pass
+
+import outlines.generate as generate
+import outlines.grammars as grammars
+import outlines.models as models
+import outlines.samplers as samplers
+
+pytestmark = pytest.mark.skipif(
+    not torch.cuda.is_available(), reason="vLLM models can only be run on GPU."
+)
+
+
+@pytest.fixture(scope="module")
+def model():
+    return models.vllm("gpt2", gpu_memory_utilization=0.5)
+
+
+@pytest.mark.parametrize(
+    "generator_type,params", ((generate.text, []), (generate.regex, ("[0-9]",)))
+)
+def test_vllm_generation_api(model, generator_type, params):
+    generator = generator_type(model, *params)
+
+    res = generator("test")
+    assert isinstance(res, str)
+
+    res = generator("test", max_tokens=10)
+    assert isinstance(res, str)
+
+    res = generator("test", stop_at=".")
+    assert isinstance(res, str)
+
+    res = generator("test", stop_at=[".", "ab"])
+    assert isinstance(res, str)
+
+    res = generator("test", stop_at=[".", "ab"])
+    assert isinstance(res, str)
+
+    res1 = generator("test", seed=1)
+    res2 = generator("test", seed=1)
+    assert isinstance(res1, str)
+    assert isinstance(res2, str)
+    assert res1 == res2
+
+    res = generator(["test", "test1"])
+    assert len(res) == 2
+
+
+def test_vllm_sampling_params(model):
+    generator = generate.text(model)
+
+    sampling_params = SamplingParams(n=2)
+    res = generator("test", sampling_params=sampling_params)
+    assert len(res) == 2
+    assert isinstance(res[0], str)
+    assert isinstance(res[1], str)
+
+    sampling_params = SamplingParams(seed=2)
+    res1 = generator("test", sampling_params=sampling_params)
+    res2 = generator("test", sampling_params=sampling_params)
+    assert res1 == res2
+
+
+def test_vllm_greedy_sampling(model):
+    sampler = samplers.greedy()
+    generator = generate.text(model, sampler)
+    res = generator("test")
+    assert isinstance(res, str)
+
+
+def test_vllm_multinomial_sampling(model):
+    sampler = samplers.multinomial()
+    generator = generate.text(model, sampler)
+    res = generator("test")
+    assert isinstance(res, str)
+
+    sampler = samplers.multinomial(3)
+    generator = generate.text(model, sampler)
+    res = generator("test")
+    assert len(res) == 3
+    assert isinstance(res[0], str)
+    assert isinstance(res[1], str)
+
+    sampler = samplers.multinomial(2, top_k=1)
+    generator = generate.text(model, sampler)
+    res = generator("test")
+    assert res[0] == res[1]
+
+    sampler = samplers.multinomial(1, top_p=0.5)
+    generator = generate.text(model, sampler)
+    res = generator("test")
+    assert isinstance(res, str)
+
+    sampler = samplers.multinomial(2, temperature=0.00001)
+    generator = generate.text(model, sampler)
+    res = generator("test")
+    assert res[0] == res[1]
+
+
+def test_vllm_beam_search(model):
+    sampler = samplers.beam_search(1)
+    generator = generate.text(model, sampler)
+    res1 = generator("test")
+    sampler = samplers.greedy()
+    generator = generate.text(model, sampler)
+    res2 = generator("test")
+    assert res1 == res2
+
+    sampler = samplers.beam_search(2)
+    generator = generate.text(model, sampler)
+    res = generator("test")
+    assert len(res) == 2
+    assert res[0] != res[1]
+
+
+def test_vllm_text_stop(model):
+    prompt = "Write a short sentence containing 'You': "
+    sequence = generate.text(model)(prompt, max_tokens=100, seed=10)
+    assert sequence.find("news") != -1
+
+    sequence = generate.text(model)(prompt, stop_at="news", max_tokens=100, seed=10)
+    assert isinstance(sequence, str)
+    assert sequence.find("news") == -1
+
+
+def test_vllm_regex(model):
+    prompt = "Write an email address: "
+    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
+    generator = generate.regex(model, regex_str)
+
+    # One prompt
+    sequence = generator(prompts=prompt)
+    assert isinstance(sequence, str)
+    assert re.fullmatch(pattern=regex_str, string=sequence) is not None
+
+
+def test_vllm_integer(model):
+    prompt = "Give me an integer: "
+    sequence = generate.format(model, int)(prompt, max_tokens=10)
+    assert isinstance(sequence, int)
+    assert sequence != ""
+    int(sequence)
+
+
+def test_vllm_float(model):
+    prompt = "Give me a floating-point number: "
+    sequence = generate.format(model, float)(prompt, max_tokens=10)
+    assert isinstance(sequence, float)
+
+    assert sequence != ""
+    float(sequence)
+
+
+def test_vllm_bool(model):
+    prompt = "Is this True or False? "
+    sequence = generate.format(model, bool)(prompt, max_tokens=10)
+    assert isinstance(sequence, bool)
+
+    assert sequence != ""
+    bool(sequence)
+
+
+def test_vllm_date(model):
+    prompt = "What day is it today? "
+    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10)
+    assert isinstance(sequence, datetime.date)
+
+
+def test_vllm_time(model):
+    prompt = "What time is it? "
+    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10)
+    assert isinstance(sequence, datetime.time)
+
+
+def test_vllm_datetime(model):
+    prompt = "What time is it? "
+    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20)
+    assert isinstance(sequence, datetime.datetime)
+
+
+def test_vllm_choice(model):
+    prompt = "Which one between 'test' and 'choice'? "
+    sequence = generate.choice(model, ["test", "choice"])(prompt)
+    assert sequence == "test" or sequence == "choice"
+
+
+def test_vllm_json_basic(model):
+    prompt = "Output some JSON. "
+
+    class Spam(BaseModel):
+        spam: constr(max_length=10)
+        fuzz: bool
+
+    sampling_params = SamplingParams(temperature=0)
+    result = generate.json(model, Spam, whitespace_pattern="")(
+        prompt, max_tokens=100, seed=1, sampling_params=sampling_params
+    )
+    assert isinstance(result, BaseModel)
+    assert isinstance(result.spam, str)
+    assert isinstance(result.fuzz, bool)
+    assert len(result.spam) <= 10
+
+
+def test_vllm_json_schema(model):
+    prompt = "Output some JSON. "
+
+    schema = """{
+      "title": "spam",
+      "type": "object",
+      "properties": {
+           "foo" : {"type": "boolean"},
+           "bar": {"type": "string", "maxLength": 4}
+        },
+      "required": ["foo", "bar"]
+      }
+    """
+
+    sampling_params = SamplingParams(temperature=0)
+    result = generate.json(model, schema, whitespace_pattern="")(
+        prompt, max_tokens=100, seed=10, sampling_params=sampling_params
+    )
+    assert isinstance(result, dict)
+    assert isinstance(result["foo"], bool)
+    assert isinstance(result["bar"], str)
+
+
+@pytest.mark.xfail(
+    reason="The CFG logits processor for vLLM has not been implemented yet."
+)
+def test_vllm_cfg(model):
+    prompt = "<|im_start|>user\nOutput a short and valid JSON object with two keys.<|im_end|>\n><|im_start|>assistant\n"
+    result = generate.cfg(model, grammars.arithmetic)(prompt, seed=11)
+    assert isinstance(result, str)
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
deleted file mode 100644
index 66b3c74db..000000000
--- a/tests/models/test_anthropic.py
+++ /dev/null
@@ -1,106 +0,0 @@
-import io
-from typing import Generator
-
-from anthropic import Anthropic as AnthropicClient
-from PIL import Image
-import pytest
-
-import outlines
-from outlines.models.anthropic import Anthropic
-from outlines.templates import Vision
-
-MODEL_NAME = "claude-3-haiku-20240307"
-
-
-@pytest.fixture(scope="session")
-def model():
-    return Anthropic(AnthropicClient(), MODEL_NAME)
-
-
-@pytest.fixture(scope="session")
-def model_no_model_name():
-    return Anthropic(AnthropicClient())
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-    image.format = "PNG"
-
-    return image
-
-
-def test_init_from_client():
-    client = AnthropicClient()
-
-    # With model name
-    model = outlines.from_anthropic(client, MODEL_NAME)
-    assert isinstance(model, Anthropic)
-    assert model.client == client
-    assert model.model_name == MODEL_NAME
-
-    # Without model name
-    model = outlines.from_anthropic(client)
-    assert isinstance(model, Anthropic)
-    assert model.client == client
-    assert model.model_name is None
-
-
-def test_anthropic_wrong_inference_parameters():
-    with pytest.raises(TypeError, match="got an unexpected"):
-        model = Anthropic(AnthropicClient(), MODEL_NAME)
-        model.generate("prompt", foo=10, max_tokens=1024)
-
-
-def test_anthropic_wrong_input_type():
-    class Foo:
-        def __init__(self, foo):
-            self.foo = foo
-
-    with pytest.raises(TypeError, match="is not available"):
-        model = Anthropic(AnthropicClient(), MODEL_NAME)
-        model.generate(Foo("prompt"))
-
-
-def test_anthropic_wrong_output_type():
-    class Foo:
-        def __init__(self, foo):
-            self.foo = foo
-
-    with pytest.raises(NotImplementedError, match="is not available"):
-        model = Anthropic(AnthropicClient(), MODEL_NAME)
-        model.generate("prompt", Foo(1))
-
-
-@pytest.mark.api_call
-def test_anthropic_simple_call(model):
-    result = model.generate("Respond with one word. Not more.", max_tokens=1024)
-    assert isinstance(result, str)
-
-
-@pytest.mark.xfail(reason="Anthropic requires the `max_tokens` parameter to be set")
-@pytest.mark.api_call
-def test_anthropic_direct_call(model_no_model_name):
-    result = model_no_model_name(
-        "Respond with one word. Not more.",
-        model_name=MODEL_NAME,
-        max_tokens=1024,
-    )
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_anthropic_simple_vision(model, image):
-    result = model.generate(
-        Vision("What does this logo represent?", image), max_tokens=1024
-    )
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_anthopic_streaming(model):
-    result = model.stream("Respond with one word. Not more.", max_tokens=1024)
-    assert isinstance(result, Generator)
-    assert isinstance(next(result), str)
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
deleted file mode 100644
index 0bd1277af..000000000
--- a/tests/models/test_dottxt.py
+++ /dev/null
@@ -1,131 +0,0 @@
-import json
-import os
-
-import pytest
-from dottxt.client import Dottxt as DottxtClient
-from pydantic import BaseModel
-
-import outlines
-from outlines import Generator
-from outlines.models.dottxt import Dottxt
-
-
-MODEL_NAME = "dottxt/dottxt-v1-alpha"
-MODEL_REVISION = "d06c86726aadd8dadb92c5b9b9e3ce8ef246c471"
-
-
-class User(BaseModel):
-    first_name: str
-    last_name: str
-    user_id: int
-
-
-@pytest.fixture(scope="session")
-def api_key():
-    """Get the Dottxt API key from the environment, providing a default value
-    if not found.
-
-    This fixture should be used for tests that do not make actual api calls,
-    but still require to initialize the Dottxt client.
-
-    """
-    api_key = os.getenv("DOTTXT_API_KEY")
-    if not api_key:
-        return "MOCK_API_KEY"
-    return api_key
-
-
-@pytest.fixture(scope="session")
-def model_name_and_revision(api_key):
-    client = DottxtClient(api_key=api_key)
-    model_list = client.list_models()
-    return (model_list[0].name, model_list[0].revision)
-
-
-@pytest.fixture(scope="session")
-def model(api_key, model_name_and_revision):
-    client = DottxtClient(api_key=api_key)
-    return Dottxt(
-        client,
-        model_name_and_revision[0],
-        model_name_and_revision[1],
-    )
-
-
-@pytest.fixture(scope="session")
-def model_no_model_name(api_key):
-    client = DottxtClient(api_key=api_key)
-    return Dottxt(client)
-
-
-@pytest.mark.api_call
-def test_dottxt_init_from_client(api_key, model_name_and_revision):
-    client = DottxtClient(api_key=api_key)
-
-    # Without model name
-    model = outlines.from_dottxt(client)
-    assert isinstance(model, Dottxt)
-    assert model.client == client
-    assert model.model_name is None
-
-    # With model name
-    model = outlines.from_dottxt(
-        client,
-        model_name_and_revision[0],
-        model_name_and_revision[1],
-    )
-    assert isinstance(model, Dottxt)
-    assert model.client == client
-    assert model.model_name == model_name_and_revision[0]
-    assert model.model_revision == model_name_and_revision[1]
-
-
-def test_dottxt_wrong_output_type(model_no_model_name):
-    with pytest.raises(TypeError, match="You must provide an output type"):
-        model_no_model_name("prompt")
-
-
-def test_dottxt_wrong_input_type(model_no_model_name):
-    with pytest.raises(TypeError, match="is not available"):
-        model_no_model_name(["prompt"], User)
-
-
-@pytest.mark.api_call
-def test_dottxt_wrong_inference_parameters(model_no_model_name):
-    with pytest.raises(TypeError, match="got an unexpected"):
-        model_no_model_name("prompt", User, foo=10)
-
-
-@pytest.mark.api_call
-def test_dottxt_direct_pydantic_call(model_no_model_name):
-    result = model_no_model_name("Create a user", User)
-    assert "first_name" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_dottxt_direct_jsonschema_call(
-    model_no_model_name, model_name_and_revision
-):
-    result = model_no_model_name(
-        "Create a user",
-        User,
-        model_name=model_name_and_revision[0],
-        model_revision=model_name_and_revision[1],
-    )
-    assert "first_name" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_dottxt_generator_pydantic_call(model):
-    generator = Generator(model, User)
-    result = generator("Create a user")
-    assert "first_name" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_dottxt_streaming(model):
-    with pytest.raises(
-        NotImplementedError,
-        match="Dottxt does not support streaming"
-    ):
-        model.stream("Create a user", User)
diff --git a/tests/models/test_dottxt_type_adapter.py b/tests/models/test_dottxt_type_adapter.py
deleted file mode 100644
index 0253af1ea..000000000
--- a/tests/models/test_dottxt_type_adapter.py
+++ /dev/null
@@ -1,137 +0,0 @@
-import io
-import json
-import pytest
-import sys
-from dataclasses import dataclass
-
-from PIL import Image
-from genson import SchemaBuilder
-from pydantic import BaseModel
-
-from outlines.models.dottxt import DottxtTypeAdapter
-from outlines.templates import Vision
-from outlines.types import cfg, json_schema, regex
-
-if sys.version_info >= (3, 12):
-    from typing import TypedDict
-else:
-    from typing_extensions import TypedDict
-
-
-@pytest.fixture
-def schema():
-    return {
-        "properties": {
-            "user_id": {"title": "User Id", "type": "integer"},
-            "name": {"title": "Name", "type": "string"},
-        },
-        "required": ["user_id", "name"],
-        "title": "User",
-        "type": "object",
-    }
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-@pytest.fixture
-def adapter():
-    return DottxtTypeAdapter()
-
-
-def test_dottxt_type_adapter_input_text(adapter):
-    message = "prompt"
-    result = adapter.format_input(message)
-    assert result == message
-
-
-def test_dottxt_type_adapter_input_invalid(adapter, image):
-    prompt = Vision("prompt", image)
-    with pytest.raises(TypeError, match="The input type"):
-        _ = adapter.format_input(prompt)
-
-
-def test_dottxt_type_adapter_output_invalid(adapter):
-    with pytest.raises(TypeError, match="You must provide an output type"):
-        adapter.format_output_type(None)
-
-    with pytest.raises(TypeError, match="The type `str` is not supported"):
-        adapter.format_output_type(str)
-
-    with pytest.raises(TypeError, match="The type `int` is not supported"):
-        adapter.format_output_type(int)
-
-    with pytest.raises(TypeError, match="Regex-based structured outputs will soon be"):
-        adapter.format_output_type(regex("[0-9]"))
-
-    with pytest.raises(TypeError, match="CFG-based structured outputs will soon be"):
-        adapter.format_output_type(cfg(""))
-
-
-def test_dottxt_type_adapter_output_dataclass(adapter, schema):
-    @dataclass
-    class User:
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == json.dumps(schema)
-
-
-def test_dottxt_type_adapter_output_typed_dict(adapter, schema):
-    class User(TypedDict):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == json.dumps(schema)
-
-
-def test_dottxt_type_adapter_output_pydantic(adapter, schema):
-    class User(BaseModel):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == json.dumps(schema)
-
-
-def test_dottxt_type_adapter_output_genson_schema_builder(adapter, schema):
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"hi": "there"})
-    builder.add_object({"hi": 5})
-
-    result = adapter.format_output_type(builder)
-    result_dict = json.loads(result)
-    assert isinstance(result_dict, dict)
-    expected_schema = {
-        "$schema": "http://json-schema.org/schema#",
-        "type": "object",
-        "properties": {"hi": {"type": ["integer", "string"]}},
-        "required": ["hi"],
-    }
-    assert result_dict == expected_schema
-
-
-def test_dottxt_type_adapter_json_schema_str(adapter, schema):
-    schema_str = json.dumps(schema)
-    result = adapter.format_output_type(json_schema(schema_str))
-    assert result == json.dumps(schema)
-
-
-def test_dottxt_type_adapter_json_schema_dict(adapter, schema):
-    result = adapter.format_output_type(json_schema(schema))
-    assert result == json.dumps(schema)
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
deleted file mode 100644
index 5df7b2669..000000000
--- a/tests/models/test_gemini.py
+++ /dev/null
@@ -1,221 +0,0 @@
-import io
-import json
-import sys
-from dataclasses import dataclass
-from enum import Enum
-from typing import Generator, Literal
-
-import pytest
-from PIL import Image
-from google.genai import Client
-from pydantic import BaseModel, ValidationError
-
-import outlines
-from outlines.models.gemini import Gemini
-from outlines.templates import Vision
-
-if sys.version_info >= (3, 12):
-    from typing import TypedDict
-else:
-    from typing_extensions import TypedDict
-
-MODEL_NAME = "gemini-1.5-flash-latest"
-
-
-@pytest.fixture(scope="session")
-def model():
-    return Gemini(Client(), MODEL_NAME)
-
-
-@pytest.fixture(scope="session")
-def model_no_model_name():
-    return Gemini(Client())
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-@pytest.mark.api_call
-def test_gemini_init_from_client():
-    client = Client()
-
-    # Without model name
-    model = outlines.from_gemini(client)
-    assert isinstance(model, Gemini)
-    assert model.client == client
-    assert model.model_name is None
-
-    # With model name
-    model = outlines.from_gemini(client, MODEL_NAME)
-    assert isinstance(model, Gemini)
-    assert model.client == client
-    assert model.model_name == MODEL_NAME
-
-
-@pytest.mark.api_call
-def test_gemini_wrong_inference_parameters(model):
-    with pytest.raises(ValidationError):
-        model.generate("prompt", foo=10)
-
-
-@pytest.mark.api_call
-def test_gemini_simple_call(model):
-    result = model.generate("Respond with one word. Not more.")
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_gemini_direct_call(model_no_model_name):
-    result = model_no_model_name(
-        "Respond with one word. Not more.",
-        model=MODEL_NAME
-    )
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_gemini_simple_vision(model, image):
-    result = model.generate(Vision("What does this logo represent?", image))
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_gemini_simple_pydantic(model):
-    class Foo(BaseModel):
-        bar: int
-
-    result = model.generate("foo?", Foo)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_gemini_simple_vision_pydantic(model, image):
-    class Logo(BaseModel):
-        name: int
-
-    result = model.generate(Vision("What does this logo represent?", image), Logo)
-    assert isinstance(result, str)
-    assert "name" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_gemini_nested_pydantic(model):
-    class Bar(BaseModel):
-        fu: str
-
-    class Foo(BaseModel):
-        sna: int
-        bar: Bar
-
-    result = model.generate("foo?", Foo)
-    assert isinstance(result, str)
-    assert "sna" in json.loads(result)
-    assert "bar" in json.loads(result)
-    assert "fu" in json.loads(result)["bar"]
-
-
-@pytest.mark.xfail(
-    reason="The Gemini SDK's serialization method does not support Json Schema strings."
-)
-@pytest.mark.api_call
-def test_gemini_simple_json_schema_string(model):
-    schema = "{'properties': {'bar': {'title': 'Bar', 'type': 'integer'}}, 'required': ['bar'], 'title': 'Foo', 'type': 'object'}"
-    result = model.generate("foo?", schema)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-@pytest.mark.xfail(
-    reason="The Gemini SDK's serialization method does not support Json Schema dictionaries."
-)
-@pytest.mark.api_call
-def test_gemini_simple_json_schema_dict(model):
-    schema = {
-        "properties": {"bar": {"type": "integer"}},
-        "required": ["bar"],
-        "type": "object",
-    }
-    result = model.generate("foo?", schema)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_gemini_simple_typed_dict(model):
-    class Foo(TypedDict):
-        bar: int
-
-    result = model.generate("foo?", Foo)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_gemini_simple_dataclass(model):
-    @dataclass
-    class Foo:
-        bar: int
-
-    result = model.generate("foo?", Foo)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_gemini_simple_choice_enum(model):
-    class Foo(Enum):
-        bar = "Bar"
-        foor = "Foo"
-
-    result = model.generate("foo?", Foo)
-    assert isinstance(result, str)
-    assert result == "Foo" or result == "Bar"
-
-
-@pytest.mark.api_call
-def test_gemini_sample_choice_literal(model):
-    result = model.generate("foo?", Literal["Foo", "Bar"])
-    assert isinstance(result, str)
-    assert result == "Foo" or result == "Bar"
-
-
-@pytest.mark.xfail(
-    reason="Gemini supports lists for choices but we do not as it is semantically incorrect."
-)
-@pytest.mark.api_call
-def test_gemini_simple_choice_list(model):
-    choices = ["Foo", "Bar"]
-    result = model.generate("foo?", choices)
-    assert isinstance(result, str)
-    assert result == "Foo" or result == "Bar"
-
-
-@pytest.mark.api_call
-def test_gemini_simple_list_pydantic(model):
-    class Foo(BaseModel):
-        bar: int
-
-    result = model.generate("foo?", list[Foo])
-    assert isinstance(json.loads(result), list)
-    assert isinstance(json.loads(result)[0], dict)
-    assert "bar" in json.loads(result)[0]
-
-
-@pytest.mark.api_call
-def test_gemini_streaming(model):
-    result = model.stream("Respond with one word. Not more.")
-    assert isinstance(result, Generator)
-    assert isinstance(next(result), str)
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
deleted file mode 100644
index 46834eae9..000000000
--- a/tests/models/test_gemini_type_adapter.py
+++ /dev/null
@@ -1,204 +0,0 @@
-import io
-import pytest
-import sys
-from dataclasses import dataclass
-from enum import Enum, EnumMeta
-from typing import Literal
-
-from PIL import Image
-from genson import SchemaBuilder
-from google.genai import types
-from pydantic import BaseModel
-
-from outlines import cfg, json_schema, regex
-from outlines.models.gemini import GeminiTypeAdapter
-from outlines.templates import Vision
-
-if sys.version_info >= (3, 12):
-    from typing import TypedDict, is_typeddict
-else:
-    from typing_extensions import TypedDict, is_typeddict
-
-
-@pytest.fixture
-def schema():
-    return {
-        "properties": {
-            "user_id": {"title": "User Id", "type": "integer"},
-            "name": {"title": "Name", "type": "string"},
-        },
-        "required": ["user_id", "name"],
-        "title": "User",
-        "type": "object",
-    }
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-@pytest.fixture
-def adapter():
-    return GeminiTypeAdapter()
-
-
-def test_gemini_type_adapter_input_text(adapter):
-    message = "prompt"
-    result = adapter.format_input(message)
-    assert result == {"contents": [message]}
-
-
-def test_gemini_type_adapter_input_vision(adapter, image):
-    input_message = Vision("hello", image)
-    result = adapter.format_input(input_message)
-    image_part = types.Part.from_bytes(
-        data=input_message.image_str,
-        mime_type=input_message.image_format
-    ),
-    assert result == {"contents": [input_message.prompt, image_part]}
-
-
-def test_dottxt_type_adapter_input_invalid(adapter):
-    @dataclass
-    class Audio:
-        file: str
-
-    prompt = Audio(
-        "file",
-    )
-    with pytest.raises(TypeError, match="The input type"):
-        _ = adapter.format_input(prompt)
-
-
-def test_gemini_type_adapter_output_invalid(adapter):
-    with pytest.raises(TypeError, match="The type `str` is not supported"):
-        adapter.format_output_type(str)
-
-    with pytest.raises(TypeError, match="The type `int` is not supported"):
-        adapter.format_output_type(int)
-
-    with pytest.raises(TypeError, match="Neither regex-based"):
-        adapter.format_output_type(regex("[0-9]"))
-
-    with pytest.raises(TypeError, match="CFG-based structured outputs"):
-        adapter.format_output_type(cfg(""))
-
-    with pytest.raises(TypeError, match="The Gemini SDK does not accept"):
-        adapter.format_output_type(SchemaBuilder())
-
-    with pytest.raises(TypeError, match="The Gemini SDK does not"):
-        adapter.format_output_type(json_schema(""))
-
-
-def test_gemini_type_adapter_output_none(adapter):
-    result = adapter.format_output_type(None)
-    assert result == {}
-
-
-def test_gemini_type_adapter_output_dataclass(adapter, schema):
-    @dataclass
-    class User:
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == {
-        "response_mime_type": "application/json",
-        "response_schema": User,
-    }
-
-
-def test_gemini_type_adapter_output_list_dataclass(adapter, schema):
-    class User(BaseModel):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(list[User])
-    assert result == {
-        "response_mime_type": "application/json",
-        "response_schema": list[User],
-    }
-
-
-def test_gemini_type_adapter_output_typed_dict(adapter, schema):
-    class User(TypedDict):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == {
-        "response_mime_type": "application/json",
-        "response_schema": User,
-    }
-
-
-def test_gemini_type_adapter_output_list_typed_dict(adapter, schema):
-    class User(BaseModel):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(list[User])
-    assert result == {
-        "response_mime_type": "application/json",
-        "response_schema": list[User],
-    }
-
-
-def test_gemini_type_adapter_output_pydantic(adapter, schema):
-    class User(BaseModel):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == {
-        "response_mime_type": "application/json",
-        "response_schema": User,
-    }
-
-
-def test_gemini_type_adapter_output_list_pydantic(adapter, schema):
-    class User(BaseModel):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(list[User])
-    assert result == {
-        "response_mime_type": "application/json",
-        "response_schema": list[User],
-    }
-
-
-def test_gemini_type_adapter_output_enum(adapter):
-    class Foo(Enum):
-        Bar = "bar"
-        Fuzz = "fuzz"
-
-    result = adapter.format_output_type(Foo)
-    assert result == {
-        "response_mime_type": "text/x.enum",
-        "response_schema": Foo,
-    }
-
-
-def test_gemini_type_adapter_output_literal(adapter):
-    Foo = Literal["bar", "fuzz"]
-    result = adapter.format_output_type(Foo)
-
-    assert isinstance(result, dict)
-    assert len(result) == 2
-    assert result["response_mime_type"] == "text/x.enum"
-    assert isinstance(result["response_schema"], EnumMeta)
-    assert len(result["response_schema"].__members__) == 2
-    assert result["response_schema"].bar.value == "bar"
-    assert result["response_schema"].fuzz.value == "fuzz"
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
deleted file mode 100644
index 6a77528f6..000000000
--- a/tests/models/test_llamacpp.py
+++ /dev/null
@@ -1,134 +0,0 @@
-import json
-from enum import Enum
-
-import pytest
-from llama_cpp import Llama
-from pydantic import BaseModel
-
-from outlines.models.llamacpp import (
-    LlamaCpp,
-    LlamaCppTokenizer,
-    LlamaCppTypeAdapter,
-    from_llamacpp
-)
-from outlines.processors import RegexLogitsProcessor
-from outlines.types.dsl import Regex, CFG
-
-
-def test_load_model():
-    model = from_llamacpp(
-        Llama.from_pretrained(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        )
-    )
-
-    assert isinstance(model, LlamaCpp)
-    assert isinstance(model.model, Llama)
-    assert isinstance(model.tokenizer, LlamaCppTokenizer)
-    assert isinstance(model.type_adapter, LlamaCppTypeAdapter)
-    assert model.tensor_library_name == "numpy"
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    return LlamaCpp(
-        Llama.from_pretrained(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        )
-    )
-
-
-def test_llamacpp_simple(model):
-    result = model.generate("Respond with one word. Not more.", None)
-    assert isinstance(result, str)
-
-
-def test_llamacpp_regex(model):
-    regex_str = Regex(r"[0-9]").pattern
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer, model.tensor_library_name)
-    result = model.generate("Respond with one word. Not more.", logits_processor)
-    assert isinstance(result, str)
-
-
-def test_llamacpp_json(model):
-    class Foo(BaseModel):
-        bar: str
-
-    result = model("foo? Respond with one word.", Foo, max_tokens=1000)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-def test_llamacpp_choice(model):
-    class Foo(Enum):
-        bar = "Bar"
-        foor = "Foo"
-
-    result = model("foo?", Foo)
-    assert result == "Foo" or result == "Bar"
-
-
-def test_llamacpp_cfg(model):
-    with pytest.raises(
-        NotImplementedError,
-        match="CFG generation is not supported for LlamaCpp"
-    ):
-        model("Respond with one word. Not more.", CFG('start: "a"'))
-
-
-def test_llamacpp_text_stop(model):
-    result = model.generate("Write the letter a.", None, stop="a", max_tokens=100)
-    assert "a" not in result
-
-
-def test_llamacpp_stream_simple(model):
-    generator = model.stream("Respond with one word. Not more.", None)
-
-    for x in generator:
-        assert isinstance(x, str)
-
-
-def test_llamacpp_stream_regex(model):
-    generator = model.stream("Respond with one word. Not more.", Regex(r"[0-9]"))
-
-    x = next(generator)
-    assert isinstance(x, str)
-
-
-def test_llamacpp_stream_json(model):
-    class Foo(BaseModel):
-        bar: int
-
-    generator = model.stream("foo?", Foo)
-
-    x = next(generator)
-    assert x == "{"
-
-
-def test_llamacpp_stream_cfg(model):
-    with pytest.raises(
-        NotImplementedError,
-        match="CFG generation is not supported for LlamaCpp"
-    ):
-        model.stream("Respond with one word. Not more.", CFG('start: "a"'))
-
-
-def test_llamacpp_stream_choice(model):
-    class Foo(Enum):
-        bar = "Bar"
-        foor = "Foo"
-
-    generator = model.stream("foo?", Foo)
-
-    x = next(generator)
-    assert x[0] in ("B", "F")
-
-
-def test_llamacpp_stream_text_stop(model):
-    generator = model.stream("Write the letter a.", None, stop="a", max_tokens=100)
-
-    result = next(generator)
-    assert isinstance(result, str)
-    assert result != "a"
diff --git a/tests/models/test_llamacpp_tokenizer.py b/tests/models/test_llamacpp_tokenizer.py
deleted file mode 100644
index dbf2507f3..000000000
--- a/tests/models/test_llamacpp_tokenizer.py
+++ /dev/null
@@ -1,133 +0,0 @@
-import pytest
-import sys
-from unittest.mock import patch
-
-import llama_cpp
-import transformers
-
-from outlines.models.llamacpp import LlamaCppTokenizer
-
-
-@pytest.fixture
-def model():
-    model = llama_cpp.Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-    )
-    setattr(
-        model.tokenizer_,
-        "hf_tokenizer",
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return model
-
-
-@pytest.fixture
-def model_no_hf_tokenizer():
-    model = llama_cpp.Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-    )
-    del model.tokenizer_
-    return model
-
-
-@pytest.fixture
-def different_model():
-    model = llama_cpp.Llama.from_pretrained(
-        "TheBloke/phi-2-GGUF",
-        "phi-2.Q4_K_M.gguf",
-    )
-    return model
-
-
-@pytest.fixture
-def tokenizer(model):
-    return LlamaCppTokenizer(model)
-
-
-@pytest.fixture
-def another_tokenizer(model):
-    return LlamaCppTokenizer(model)
-
-
-@pytest.fixture
-def tokenizer_no_hf_tokenizer(model_no_hf_tokenizer):
-    return LlamaCppTokenizer(model_no_hf_tokenizer)
-
-
-@pytest.fixture
-def different_tokenizer(different_model):
-    return LlamaCppTokenizer(different_model)
-
-
-def test_llama_cpp_tokenizer_init(tokenizer, tokenizer_no_hf_tokenizer):
-    # regular case
-    assert tokenizer.eos_token_id is not None
-    assert tokenizer.pad_token_id is not None
-    assert isinstance(tokenizer.vocabulary, dict)
-
-    # tokenizer with no hf_tokenizer
-    assert tokenizer_no_hf_tokenizer.eos_token_id is not None
-    assert tokenizer_no_hf_tokenizer.pad_token_id is not None
-    assert isinstance(tokenizer_no_hf_tokenizer.vocabulary, dict)
-
-
-def test_llama_cpp_tokenizer_encode(tokenizer):
-    # batch case
-    with pytest.raises(NotImplementedError):
-        token_ids, attention_mask = tokenizer.encode(["foo", "bar"])
-
-    # regular case
-    token_ids, attention_mask = tokenizer.encode("Hello, world!")
-    assert token_ids is not None
-    assert attention_mask is not None
-    assert len(token_ids) == len(attention_mask)
-
-
-def test_llama_cpp_tokenizer_decode(tokenizer):
-    token_ids, _ = tokenizer.encode("Hello, world!")
-    decoded_text = tokenizer.decode(token_ids)
-    assert isinstance(decoded_text, list)
-    assert "".join(decoded_text).strip() == "Hello, world!"
-
-
-def test_llama_cpp_tokenizer_convert_token_to_string(
-    tokenizer,
-    tokenizer_no_hf_tokenizer
-):
-    # with self._hf_tokenizer
-    token_str = tokenizer.convert_token_to_string("<0x20>")
-    assert isinstance(token_str, str)
-
-    # without self._hf_tokenizer
-    token_str = tokenizer_no_hf_tokenizer.convert_token_to_string("<0x20>")
-    assert isinstance(token_str, str)
-
-
-def test_llama_cpp_tokenizer_eq(tokenizer, another_tokenizer, different_tokenizer):
-    assert not tokenizer == 1
-    assert tokenizer == another_tokenizer
-    assert tokenizer != different_tokenizer
-
-
-def test_llama_cpp_tokenizer_hash(tokenizer, another_tokenizer, different_tokenizer):
-    assert isinstance(hash(tokenizer), int)
-    assert hash(tokenizer) == hash(another_tokenizer)
-    assert hash(tokenizer) != hash(different_tokenizer)
-
-
-def test_llama_cpp_tokenizer_getstate(tokenizer):
-    state = tokenizer.__getstate__()
-    assert isinstance(state, tuple)
-    assert len(state) == 5
-    assert isinstance(state[0], dict)
-    assert isinstance(state[1], int)
-    assert isinstance(state[2], str)
-    assert isinstance(state[3], int)
-    assert isinstance(state[4], list)
-
-
-def test_llama_cpp_tokenizer_setstate(tokenizer):
-    with pytest.raises(NotImplementedError):
-        tokenizer.__setstate__(None)
diff --git a/tests/models/test_llamacpp_type_adapter.py b/tests/models/test_llamacpp_type_adapter.py
deleted file mode 100644
index e49586d5d..000000000
--- a/tests/models/test_llamacpp_type_adapter.py
+++ /dev/null
@@ -1,41 +0,0 @@
-import pytest
-
-import llama_cpp
-from llama_cpp import LogitsProcessorList
-
-from outlines.models.llamacpp import LlamaCppTypeAdapter, LlamaCppTokenizer
-from outlines.processors.structured import RegexLogitsProcessor
-
-
-@pytest.fixture
-def adapter():
-    return LlamaCppTypeAdapter()
-
-
-@pytest.fixture
-def logits_processor():
-    model = llama_cpp.Llama.from_pretrained(
-        "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-    )
-    return RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=LlamaCppTokenizer(model),
-        tensor_library_name="numpy",
-    )
-
-
-def test_llamacpp_type_adapter_format_input(adapter):
-    # anything else than a string
-    with pytest.raises(NotImplementedError):
-        adapter.format_input(["Hello, world!"])
-
-    # string case
-    assert adapter.format_input("Hello, world!") == "Hello, world!"
-
-
-def test_llamacpp_type_adapter_format_output_type(adapter, logits_processor):
-    formatted = adapter.format_output_type(logits_processor)
-    assert isinstance(formatted, LogitsProcessorList)
-    assert len(formatted) == 1
-    assert isinstance(formatted[0], RegexLogitsProcessor)
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index 8ec889230..20e59da81 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -1,20 +1,9 @@
 import pytest
-import re
-from enum import Enum
-from typing import Generator
-
-import outlines
-from outlines.types import Regex
-from outlines.models.mlxlm import (
-    MLXLM,
-    MLXLMTypeAdapter,
-    from_mlxlm
-)
+
+from outlines.models.mlxlm import mlxlm
 from outlines.models.transformers import TransformerTokenizer
-from pydantic import BaseModel
 
 try:
-    import mlx_lm
     import mlx.core as mx
 
     HAS_MLX = mx.metal.is_available()
@@ -26,94 +15,86 @@
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_model_initialization():
-    model = from_mlxlm(*mlx_lm.load(TEST_MODEL))
-    assert isinstance(model, MLXLM)
-    assert isinstance(model.model, mlx_lm.models.llama.Model)
-    assert isinstance(
-        model.mlx_tokenizer, mlx_lm.tokenizer_utils.TokenizerWrapper
-    )
+def test_mlxlm_model():
+    model = mlxlm(TEST_MODEL)
+    assert hasattr(model, "model")
+    assert hasattr(model, "tokenizer")
     assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert isinstance(model.type_adapter, MLXLMTypeAdapter)
-    assert model.tensor_library_name == "mlx"
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    model, tokenizer = mlx_lm.load(TEST_MODEL)
-    return outlines.from_mlxlm(model, tokenizer)
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_tokenizer(model):
+def test_mlxlm_tokenizer():
+    model = mlxlm(TEST_MODEL)
+
     # Test single string encoding/decoding
     test_text = "Hello, world!"
-    token_ids, _ = model.tokenizer.encode(test_text)
-    token_ids = mx.array(token_ids)
+    token_ids = mx.array(model.mlx_tokenizer.encode(test_text))
     assert isinstance(token_ids, mx.array)
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_simple(model):
-    result = model.generate("Respond with one word. Not more.", None)
-    assert isinstance(result, str)
-
-
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_call(model):
-    result = model("Respond with one word. Not more.")
-    assert isinstance(result, str)
+def test_mlxlm_generate():
+    from outlines.generate.api import GenerationParameters, SamplingParameters
 
+    model = mlxlm(TEST_MODEL)
+    prompt = "Write a haiku about programming:"
 
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_invalid_input_type(model):
-    with pytest.raises(NotImplementedError, match="is not available"):
-        model(["Respond with one word. Not more."])
+    # Test with basic generation parameters
+    gen_params = GenerationParameters(max_tokens=50, stop_at=None, seed=None)
 
+    # Test with different sampling parameters
+    sampling_params = SamplingParameters(
+        sampler="multinomial", num_samples=1, top_p=0.9, top_k=None, temperature=0.7
+    )
 
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_invalid_inference_kwargs(model):
-    with pytest.raises(TypeError):
-        model("Respond with one word. Not more.", foo="bar")
+    # Test generation
+    output = model.generate(prompt, gen_params, None, sampling_params)
+    assert isinstance(output, str)
+    assert len(output) > 0
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_inference_kwargs(model):
-    result = model("Write a short story about a cat.", max_tokens=2)
-    assert isinstance(result, str)
-    assert len(result) < 20
+def test_mlxlm_stream():
+    from outlines.generate.api import GenerationParameters, SamplingParameters
 
+    model = mlxlm(TEST_MODEL)
+    prompt = "Count from 1 to 5:"
 
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_regex(model):
-    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]", result)
+    gen_params = GenerationParameters(max_tokens=20, stop_at=None, seed=None)
 
+    sampling_params = SamplingParameters(
+        sampler="greedy",  # Use greedy sampling for deterministic output
+        num_samples=1,
+        top_p=None,
+        top_k=None,
+        temperature=0.0,
+    )
 
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_JsonType(model):
-    class Character(BaseModel):
-        name: str
+    # Test streaming
+    stream = model.stream(prompt, gen_params, None, sampling_params)
+    tokens = list(stream)
+    assert len(tokens) > 0
+    assert all(isinstance(token, str) for token in tokens)
 
-    result = model("Create a character with a name.", Character)
-    assert "name" in result
+    # Test that concatenated streaming output matches generate output
+    streamed_text = "".join(tokens)
+    generated_text = model.generate(prompt, gen_params, None, sampling_params)
+    assert streamed_text == generated_text
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_choice(model):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
+def test_mlxlm_errors():
+    model = mlxlm(TEST_MODEL)
 
-    result = model("Cat or dog?", Foo)
-    assert result in ["cat", "dog"]
+    # Test batch inference (should raise NotImplementedError)
+    with pytest.raises(NotImplementedError):
+        from outlines.generate.api import GenerationParameters, SamplingParameters
 
+        gen_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
+        sampling_params = SamplingParameters("multinomial", 1, None, None, 1.0)
+        model.generate(["prompt1", "prompt2"], gen_params, None, sampling_params)
 
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_stream_text_stop(model):
-    generator = model.stream(
-        "Respond with one word. Not more.", None, max_tokens=100
-    )
-    assert isinstance(generator, Generator)
-    assert isinstance(next(generator), str)
+    # Test beam search (should raise NotImplementedError)
+    with pytest.raises(NotImplementedError):
+        sampling_params = SamplingParameters("beam_search", 1, None, None, 1.0)
+        model.generate("test prompt", gen_params, None, sampling_params)
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
deleted file mode 100644
index 408fa2ceb..000000000
--- a/tests/models/test_ollama.py
+++ /dev/null
@@ -1,101 +0,0 @@
-import json
-from enum import Enum
-from typing import Annotated
-
-import pytest
-from ollama import Client as OllamaClient
-from pydantic import BaseModel, Field
-
-import outlines
-from outlines.models import Ollama
-
-
-MODEL_NAME = "tinyllama"
-
-
-@pytest.fixture
-def model():
-    return Ollama(OllamaClient(), MODEL_NAME)
-
-
-@pytest.fixture
-def model_no_model_name():
-    return Ollama(OllamaClient())
-
-
-def test_init_from_client():
-    client = OllamaClient()
-
-    # With model name
-    model = outlines.from_ollama(client, MODEL_NAME)
-    assert isinstance(model, Ollama)
-    assert model.client == client
-    assert model.model_name == MODEL_NAME
-
-    # Without model name
-    model = outlines.from_ollama(client)
-    assert isinstance(model, Ollama)
-    assert model.client == client
-    assert model.model_name is None
-
-
-def test_wrong_inference_parameters(model):
-    with pytest.raises(TypeError, match="got an unexpected"):
-        model.generate(
-            "Respond with one word. Not more.", None, foo=10
-        )
-
-
-def test_ollama_simple(model):
-    result = model.generate(
-        "Respond with one word. Not more.", None
-    )
-    assert isinstance(result, str)
-
-
-def test_ollama_direct(model_no_model_name):
-    result = model_no_model_name(
-        "Respond with one word. Not more.",
-        None,
-        model=MODEL_NAME,
-    )
-    assert isinstance(result, str)
-
-
-def test_ollama_json(model):
-    class Foo(BaseModel):
-        foo: Annotated[str, Field(max_length=1)]
-
-    result = model("Respond with one word. Not more.", Foo)
-    assert isinstance(result, str)
-    assert "foo" in json.loads(result)
-
-
-def test_ollama_wrong_output_type(model):
-    class Foo(Enum):
-        bar = "Bar"
-        foor = "Foo"
-
-    with pytest.raises(TypeError, match="is not supported"):
-        model.generate("foo?", Foo)
-
-
-def test_ollama_wrong_input_type(model):
-    with pytest.raises(TypeError, match="is not available"):
-        model.generate(["foo?", "bar?"], None)
-
-
-def test_ollama_stream(model):
-    generator = model.stream("Write a sentence about a cat.")
-    assert isinstance(next(generator), str)
-
-
-def test_ollama_stream_json(model_no_model_name):
-    class Foo(BaseModel):
-        foo: Annotated[str, Field(max_length=2)]
-
-    generator = model_no_model_name.stream("Create a character.", Foo, model=MODEL_NAME)
-    generated_text = []
-    for text in generator:
-        generated_text.append(text)
-    assert "foo" in json.loads("".join(generated_text))
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
deleted file mode 100644
index 58b6c6aa9..000000000
--- a/tests/models/test_ollama_type_adapter.py
+++ /dev/null
@@ -1,131 +0,0 @@
-import io
-import json
-import pytest
-import sys
-from dataclasses import dataclass
-
-from genson import SchemaBuilder
-from PIL import Image
-from pydantic import BaseModel
-
-from outlines.models.ollama import OllamaTypeAdapter
-from outlines.templates import Vision
-from outlines.types import cfg, json_schema, regex
-
-if sys.version_info >= (3, 12):
-    from typing import TypedDict
-else:
-    from typing_extensions import TypedDict
-
-
-@pytest.fixture
-def schema():
-    return {
-        "properties": {
-            "user_id": {"title": "User Id", "type": "integer"},
-            "name": {"title": "Name", "type": "string"},
-        },
-        "required": ["user_id", "name"],
-        "title": "User",
-        "type": "object",
-    }
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-@pytest.fixture
-def adapter():
-    return OllamaTypeAdapter()
-
-
-def test_ollama_type_adapter_input_text(adapter):
-    message = "prompt"
-    result = adapter.format_input(message)
-    assert result == message
-
-
-def test_ollama_type_adapter_input_invalid(adapter, image):
-    prompt = Vision("prompt", image)
-    with pytest.raises(TypeError, match="The input type"):
-        _ = adapter.format_input(prompt)
-
-
-def test_ollama_type_adapter_output_invalid(adapter):
-    with pytest.raises(TypeError, match="The type `str` is not supported"):
-        adapter.format_output_type(str)
-
-    with pytest.raises(TypeError, match="The type `int` is not supported"):
-        adapter.format_output_type(int)
-
-    with pytest.raises(TypeError, match="Regex-based structured outputs are not"):
-        adapter.format_output_type(regex("[0-9]"))
-
-    with pytest.raises(TypeError, match="CFG-based structured outputs are not"):
-        adapter.format_output_type(cfg(""))
-
-
-def test_ollama_type_adapter_output_dataclass(adapter, schema):
-    @dataclass
-    class User:
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == schema
-
-
-def test_ollama_type_adapter_output_typed_dict(adapter, schema):
-    class User(TypedDict):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == schema
-
-
-def test_ollama_type_adapter_output_pydantic(adapter, schema):
-    class User(BaseModel):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert result == schema
-
-
-def test_ollama_type_adapter_output_genson_schema_builder(adapter, schema):
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"hi": "there"})
-    builder.add_object({"hi": 5})
-
-    result = adapter.format_output_type(builder)
-    assert result == json.dumps({
-        "$schema": "http://json-schema.org/schema#",
-        "type": "object",
-        "properties": {"hi": {"type": ["integer", "string"]}},
-        "required": ["hi"]
-    })
-
-
-def test_ollama_type_adapter_json_schema_str(adapter, schema):
-    schema_str = json.dumps(schema)
-    result = adapter.format_output_type(json_schema(schema_str))
-    assert result == schema
-
-
-def test_ollama_type_adapter_json_schema_dict(adapter, schema):
-    result = adapter.format_output_type(json_schema(schema))
-    assert result == schema
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 455087707..f4b97f36a 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,173 +1,125 @@
-import io
+import importlib
 import json
-import os
-from typing import Annotated, Generator
+from contextlib import contextmanager
+from unittest import mock
+from unittest.mock import MagicMock
 
 import pytest
-import requests
-from PIL import Image
-from openai import OpenAI as OpenAIClient
-from pydantic import BaseModel, Field
-
-import outlines
-from outlines.models.openai import OpenAI
-from outlines.templates import Vision
-from outlines.types import json_schema
-
-MODEL_NAME = "gpt-4o-mini-2024-07-18"
-
-
-@pytest.fixture(scope="session")
-def api_key():
-    """Get the OpenAI API key from the environment, providing a default value if not found.
-
-    This fixture should be used for tests that do not make actual api calls,
-    but still require to initialize the OpenAI client.
-
-    """
-    api_key = os.getenv("OPENAI_API_KEY")
-    if not api_key:
-        return "MOCK_VALUE"
-    return api_key
-
-
-@pytest.fixture(scope="session")
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-@pytest.fixture(scope="session")
-def model(api_key):
-    return OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
-
-
-@pytest.fixture(scope="session")
-def model_no_model_name(api_key):
-    return OpenAI(OpenAIClient(api_key=api_key))
-
-
-def test_init_from_client(api_key):
-    client = OpenAIClient(api_key=api_key)
-
-    # With model name
-    model = outlines.from_openai(client, "gpt-4o")
-    assert isinstance(model, OpenAI)
-    assert model.client == client
-    assert model.model_name == "gpt-4o"
-
-    # Without model name
-    model = outlines.from_openai(client)
-    assert isinstance(model, OpenAI)
-    assert model.client == client
-    assert model.model_name is None
-
-
-def test_openai_wrong_inference_parameters(model):
-    with pytest.raises(TypeError, match="got an unexpected"):
-        model.generate("prompt", foo=10)
-
-
-def test_openai_wrong_input_type(model):
-    class Foo:
-        def __init__(self, foo):
-            self.foo = foo
-
-    with pytest.raises(TypeError, match="is not available"):
-        model.generate(Foo("prompt"))
-
-
-def test_openai_wrong_output_type(model):
-    class Foo:
-        def __init__(self, foo):
-            self.foo = foo
-
-    with pytest.raises(TypeError, match="is not available"):
-        model.generate("prompt", Foo(1))
-
-
-@pytest.mark.api_call
-def test_openai_simple_call(model):
-    result = model.generate("Respond with one word. Not more.")
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_openai_simple_call_multiple_samples(model):
-    result = model.generate("Respond with one word. Not more.", n=2)
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert isinstance(result[0], str)
-    assert isinstance(result[1], str)
-
-
-@pytest.mark.api_call
-def test_openai_direct_call(model_no_model_name):
-    result = model_no_model_name(
-        "Respond with one word. Not more.",
-        model=MODEL_NAME,
-    )
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_openai_simple_vision(image, model):
-    result = model.generate(Vision("What does this logo represent?", image))
-    assert isinstance(result, str)
-
-
-@pytest.mark.api_call
-def test_openai_simple_pydantic(model):
-    class Foo(BaseModel):
-        bar: int
-
-    result = model.generate("foo?", Foo)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_openai_simple_pydantic_refusal(model):
-    class Foo(BaseModel):
-        bar: Annotated[str, Field(int, pattern=r"^\d+$")]
-
-    with pytest.raises(TypeError, match="OpenAI does not support your schema"):
-        _ = model.generate("foo?", Foo)
-
-
-@pytest.mark.api_call
-def test_openai_simple_vision_pydantic(image, model):
-    class Logo(BaseModel):
-        name: int
-
-    result = model.generate(Vision("What does this logo represent?", image), Logo)
-    assert isinstance(result, str)
-    assert "name" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_openai_simple_json_schema(model):
-    class Foo(BaseModel):
-        bar: int
-
-    schema = json.dumps(Foo.model_json_schema())
-
-    result = model.generate("foo?", json_schema(schema))
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
-@pytest.mark.api_call
-def test_openai_streaming(model):
-    result = model.stream("Respond with one word. Not more.")
-    assert isinstance(result, Generator)
-    assert isinstance(next(result), str)
+from openai import AsyncOpenAI
+
+from outlines import generate
+from outlines.models.openai import OpenAI, OpenAIConfig
+
+
+def module_patch(path):
+    """Patch functions that have the same name as the module in which they're implemented."""
+    target = path
+    components = target.split(".")
+    for i in range(len(components), 0, -1):
+        try:
+            # attempt to import the module
+            imported = importlib.import_module(".".join(components[:i]))
+
+            # module was imported, let's use it in the patch
+            patch = mock.patch(path)
+            patch.getter = lambda: imported
+            patch.attribute = ".".join(components[i:])
+            return patch
+        except Exception:
+            continue
+
+    # did not find a module, just return the default mock
+    return mock.patch(path)
+
+
+def test_openai_call():
+    with module_patch("outlines.models.openai.generate_chat") as mocked_generate_chat:
+        mocked_generate_chat.return_value = ["foo"], 1, 2
+        async_client = MagicMock(spec=AsyncOpenAI, api_key="key")
+
+        model = OpenAI(
+            async_client,
+            OpenAIConfig(max_tokens=10, temperature=0.5, n=2, stop=["."]),
+        )
+
+        assert model("bar")[0] == "foo"
+        assert model.prompt_tokens == 1
+        assert model.completion_tokens == 2
+        mocked_generate_chat_args = mocked_generate_chat.call_args
+        mocked_generate_chat_arg_config = mocked_generate_chat_args[0][3]
+        assert isinstance(mocked_generate_chat_arg_config, OpenAIConfig)
+        assert mocked_generate_chat_arg_config.max_tokens == 10
+        assert mocked_generate_chat_arg_config.temperature == 0.5
+        assert mocked_generate_chat_arg_config.n == 2
+        assert mocked_generate_chat_arg_config.stop == ["."]
+
+        model("bar", samples=3)
+        mocked_generate_chat_args = mocked_generate_chat.call_args
+        mocked_generate_chat_arg_config = mocked_generate_chat_args[0][3]
+        assert mocked_generate_chat_arg_config.n == 3
+
+
+@contextmanager
+def patched_openai(completion, **oai_config):
+    """Create a patched openai whose chat completions always returns `completion`"""
+    with module_patch("outlines.models.openai.generate_chat") as mocked_generate_chat:
+        mocked_generate_chat.return_value = completion, 1, 2
+        async_client = MagicMock(spec=AsyncOpenAI, api_key="key")
+        model = OpenAI(
+            async_client,
+            OpenAIConfig(max_tokens=10, temperature=0.5, n=2, stop=["."]),
+        )
+        yield model
+
+
+def test_openai_choice_call():
+    with patched_openai(completion='{"result": "foo"}') as model:
+        generator = generate.choice(model, ["foo", "bar"])
+        assert generator("hi") == "foo"
+
+
+def test_openai_choice_call_invalid_server_response():
+    with patched_openai(completion="not actual json") as model:
+        generator = generate.choice(model, ["foo", "bar"])
+        with pytest.raises(json.decoder.JSONDecodeError):
+            generator("hi")
+
+
+def test_openai_json_call_pydantic():
+    from pydantic import BaseModel, ConfigDict, ValidationError
+
+    class Person(BaseModel):
+        model_config = ConfigDict(extra="forbid")  # required for openai
+        first_name: str
+        last_name: str
+        age: int
+
+    completion = '{"first_name": "Usain", "last_name": "Bolt", "age": 38}'
+
+    # assert success for valid response
+    with patched_openai(completion=completion) as model:
+        generator = generate.json(model, Person)
+        assert generator("fastest person") == Person.parse_raw(completion)
+
+    # assert fail for non-json response
+    with patched_openai(completion="usain bolt") as model:
+        generator = generate.json(model, Person)
+        with pytest.raises(ValidationError):
+            assert generator("fastest person")
+
+
+def test_openai_json_call_str():
+    person_schema = '{"additionalProperties": false, "properties": {"first_name": {"title": "First Name", "type": "string"}, "last_name": {"title": "Last Name", "type": "string"}, "age": {"title": "Age", "type": "integer"}}, "required": ["first_name", "last_name", "age"], "title": "Person", "type": "object"}'
+
+    output = {"first_name": "Usain", "last_name": "Bolt", "age": 38}
+
+    # assert success for valid response
+    with patched_openai(completion=json.dumps(output)) as model:
+        generator = generate.json(model, person_schema)
+        assert generator("fastest person") == output
+
+    # assert fail for non-json response
+    with patched_openai(completion="usain bolt") as model:
+        generator = generate.json(model, person_schema)
+        with pytest.raises(json.decoder.JSONDecodeError):
+            assert generator("fastest person")
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
deleted file mode 100644
index fcafbb8de..000000000
--- a/tests/models/test_openai_type_adapter.py
+++ /dev/null
@@ -1,195 +0,0 @@
-import io
-import json
-import pytest
-import sys
-from dataclasses import dataclass
-from typing import Literal
-
-from genson import SchemaBuilder
-from PIL import Image
-from pydantic import BaseModel
-
-from outlines import cfg, json_schema, regex, Vision
-from outlines.models.openai import OpenAITypeAdapter
-
-if sys.version_info >= (3, 12):
-    from typing import TypedDict
-else:
-    from typing_extensions import TypedDict
-
-
-@pytest.fixture
-def schema():
-    return {
-        "properties": {
-            "user_id": {"title": "User Id", "type": "integer"},
-            "name": {"title": "Name", "type": "string"},
-        },
-        "required": ["user_id", "name"],
-        "title": "User",
-        "type": "object",
-        "additionalProperties": False,
-    }
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-@pytest.fixture
-def adapter():
-    return OpenAITypeAdapter()
-
-
-def test_openai_type_adapter_input_text(adapter):
-    message = "prompt"
-    result = adapter.format_input(message)
-    assert isinstance(result, dict)
-    assert len(result["messages"]) == 1
-    assert result["messages"][0]["role"] == "user"
-    assert result["messages"][0]["content"] == message
-
-
-def test_openai_type_adapter_input_vision(adapter, image):
-    input_message = Vision("hello", image)
-    result = adapter.format_input(input_message)
-    assert isinstance(result, dict)
-
-    messages = result["messages"]
-    assert len(messages) == 1
-
-    message = messages[0]
-    assert message["role"] == "user"
-    assert len(message["content"]) == 2
-    assert message["content"][0]["type"] == "text"
-    assert message["content"][0]["text"] == "hello"
-
-    assert message["content"][1]["type"] == "image_url"
-    assert (
-        message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message.image_str}"
-    )
-
-
-def test_dottxt_type_adapter_input_invalid(adapter):
-    @dataclass
-    class Audio:
-        file: str
-
-    prompt = Audio(
-        "file",
-    )
-    with pytest.raises(TypeError, match="The input type"):
-        _ = adapter.format_input(prompt)
-
-
-def test_openai_type_adapter_output_invalid(adapter):
-    with pytest.raises(TypeError, match="The type `str` is not available"):
-        adapter.format_output_type(str)
-
-    with pytest.raises(TypeError, match="The type `int` is not available"):
-        adapter.format_output_type(int)
-
-    with pytest.raises(TypeError, match="The type `Literal` is not available"):
-        adapter.format_output_type(Literal[1, 2])
-
-    with pytest.raises(TypeError, match="Neither regex-based"):
-        adapter.format_output_type(regex("[0-9]"))
-
-    with pytest.raises(TypeError, match="CFG-based structured outputs"):
-        adapter.format_output_type(cfg(""))
-
-    class Foo(BaseModel):
-        bar: str
-
-    with pytest.raises(TypeError, match="The type `list` is not available"):
-        adapter.format_output_type(list[Foo])
-
-
-def test_openai_type_adapter_output_none(adapter):
-    result = adapter.format_output_type(None)
-    assert result == {}
-
-
-def test_openai_type_adapter_json_mode(adapter):
-    result = adapter.format_output_type(dict)
-    assert result == {"response_format": {"type": "json_object"}}
-
-
-def test_openai_type_adapter_dataclass(adapter, schema):
-    @dataclass
-    class User:
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert isinstance(result, dict)
-    assert result["response_format"]["json_schema"]["strict"] is True
-    assert result["response_format"]["json_schema"]["schema"] == schema
-
-
-def test_openai_type_adapter_typed_dict(adapter, schema):
-    class User(TypedDict):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert isinstance(result, dict)
-    assert result["response_format"]["json_schema"]["strict"] is True
-    assert result["response_format"]["json_schema"]["schema"] == schema
-
-
-def test_openai_type_adapter_pydantic(adapter, schema):
-    class User(BaseModel):
-        user_id: int
-        name: str
-
-    result = adapter.format_output_type(User)
-    assert isinstance(result, dict)
-    assert result["response_format"]["json_schema"]["strict"] is True
-    assert result["response_format"]["json_schema"]["schema"] == schema
-
-
-def test_openai_type_adapter_genson_schema_builder(adapter, schema):
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"hi": "there"})
-    builder.add_object({"hi": 5})
-
-    result = adapter.format_output_type(builder)
-    assert isinstance(result, dict)
-    assert result["response_format"]["json_schema"]["strict"] is True
-    expected_schema = {
-        "$schema": "http://json-schema.org/schema#",
-        "type": "object",
-        "properties": {"hi": {"type": ["integer", "string"]}},
-        "required": ["hi"],
-        "additionalProperties": False  # OpenAI adds this
-    }
-    assert result["response_format"]["json_schema"]["schema"] == expected_schema
-
-
-def test_openai_type_adapter_json_schema_str(adapter, schema):
-    schema_str = json.dumps(schema)
-    result = adapter.format_output_type(json_schema(schema_str))
-    assert isinstance(result, dict)
-    assert result["response_format"]["json_schema"]["strict"] is True
-    assert result["response_format"]["json_schema"]["schema"] == schema
-
-
-def test_openai_type_adapter_json_schema_dict(adapter, schema):
-    result = adapter.format_output_type(json_schema(schema))
-    assert isinstance(result, dict)
-    assert result["response_format"]["json_schema"]["strict"] is True
-    assert result["response_format"]["json_schema"]["schema"] == schema
diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
deleted file mode 100644
index 60960f463..000000000
--- a/tests/models/test_sglang.py
+++ /dev/null
@@ -1,272 +0,0 @@
-# ATTENTION: When running this test with an actual SGLang server, use the
-# llguidance backend (--grammar-backend llguidance)
-# The outlines backend does not support the EBNF grammar. The xgrammar
-# backend is slow and buggy.
-
-import os
-import re
-import warnings
-from typing import AsyncGenerator, Generator
-
-import pytest
-from openai import AsyncOpenAI, OpenAI
-
-from outlines.models.sglang import SGLang, AsyncSGLang, from_sglang
-from outlines.types.dsl import CFG, Regex, JsonSchema
-from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
-
-
-EBNF_YES_NO_GRAMMAR = """
-root ::= answer
-answer ::= "yes" | "no"
-"""
-
-
-# If the SGLANG_SERVER_URL environment variable is set, use the real SGLang server
-# Otherwise, use the mock server
-sglang_server_url = os.environ.get("SGLANG_SERVER_URL")
-sglang_model_name = os.environ.get(
-    "SGLANG_MODEL_NAME", "qwen/qwen2.5-0.5b-instruct"
-)
-if sglang_server_url:
-    openai_client = OpenAI(base_url=sglang_server_url)
-    async_openai_client = AsyncOpenAI(base_url=sglang_server_url)
-else:
-    warnings.warn("No SGLang server URL provided, using mock server")
-    openai_client = MockOpenAIClient()
-    async_openai_client = MockAsyncOpenAIClient()
-
-mock_responses = [
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Respond with a single word.'}
-            ],
-            'model': sglang_model_name,
-        },
-        "foo"
-    ),
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Respond with a single word.'}
-            ],
-            'model': sglang_model_name,
-            'stream': True
-        },
-        ["foo", "bar"]
-    ),
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Respond with a single word.'}
-            ],
-            'n': 2,
-            'model': sglang_model_name,
-        },
-        ["foo", "bar"]
-    ),
-    (
-        {
-            'messages': [{'role': "user", 'content': 'foo?'}],
-            'model': sglang_model_name,
-            'max_tokens': 10,
-            'response_format': {
-                'type': 'json_schema',
-                'json_schema': {
-                    'name': 'default',
-                    'strict': True,
-                    'schema': {
-                        'type': 'object',
-                        'properties': {'bar': {'type': 'string'}},
-                        'additionalProperties': False
-                    }
-                }
-            }
-        },
-        '{"foo": "bar"}'
-    ),
-    (
-        {
-            'messages': [{'role': "user", 'content': 'foo?'}],
-            'model': sglang_model_name,
-            'max_tokens': 10,
-            'extra_body': {
-                'regex': '([0-9]{3})',
-            },
-        },
-        "123"
-    ),
-    (
-        {
-            'messages': [{'role': "user", 'content': 'foo?'}],
-            'model': sglang_model_name,
-            'max_tokens': 10,
-            'extra_body': {
-                'ebnf': EBNF_YES_NO_GRAMMAR,
-            },
-        },
-        "yes"
-    ),
-]
-
-
-# If the SGLANG_SERVER_URL environment variable is not set, add the mock
-# responses to the mock clients
-if not sglang_server_url:
-    async_openai_client.add_mock_responses(mock_responses)
-    openai_client.add_mock_responses(mock_responses)
-
-
-@pytest.fixture
-def sync_model():
-    return SGLang(openai_client, model_name=sglang_model_name)
-
-
-@pytest.fixture
-def sync_model_no_model_name():
-    return SGLang(openai_client)
-
-
-@pytest.fixture
-def async_model():
-    return AsyncSGLang(async_openai_client, model_name=sglang_model_name)
-
-
-@pytest.fixture
-def async_model_no_model_name():
-    return AsyncSGLang(async_openai_client)
-
-
-def test_sglang_init():
-    # We do not rely on the mock server here because we need an object
-    # of type OpenAI and AsyncOpenAI to test the init function.
-    openai_client = OpenAI(base_url="http://localhost:11434")
-    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
-
-    # Sync with model name
-    model = from_sglang(openai_client, sglang_model_name)
-    assert isinstance(model, SGLang)
-    assert model.client == openai_client
-    assert model.model_name == sglang_model_name
-
-    # Sync without model name
-    model = from_sglang(openai_client)
-    assert isinstance(model, SGLang)
-    assert model.client == openai_client
-    assert model.model_name is None
-
-    # Async with model name
-    model = from_sglang(async_openai_client, sglang_model_name)
-    assert isinstance(model, AsyncSGLang)
-    assert model.client == async_openai_client
-    assert model.model_name == sglang_model_name
-
-    # Async without model name
-    model = from_sglang(async_openai_client)
-    assert isinstance(model, AsyncSGLang)
-    assert model.client == async_openai_client
-    assert model.model_name is None
-
-    with pytest.raises(ValueError, match="Unsupported client type"):
-        from_sglang("foo")
-
-
-def test_sglang_sync_simple_call(sync_model):
-    result = sync_model("Respond with a single word.",)
-    assert isinstance(result, str)
-
-
-def test_sglang_sync_streaming(sync_model_no_model_name):
-    result = sync_model_no_model_name.stream(
-        "Respond with a single word.",
-        model=sglang_model_name,
-    )
-    assert isinstance(result, Generator)
-    assert isinstance(next(result), str)
-
-
-def test_sglang_sync_multiple_samples(sync_model):
-    result = sync_model("Respond with a single word.", n=2)
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert isinstance(result[0], str)
-    assert isinstance(result[1], str)
-
-
-def test_sglang_sync_json(sync_model):
-    json_string = (
-        '{"type": "object", "properties":'
-        + ' {"bar": {"type": "string"}}}'
-    )
-    result = sync_model("foo?", JsonSchema(json_string), max_tokens=10)
-    assert isinstance(result, str)
-    assert "bar" in result
-
-
-def test_sglang_sync_regex(sync_model):
-    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]{3}", result)
-
-
-def test_sglang_sync_cfg(sync_model):
-    with pytest.warns(
-        UserWarning,
-        match="SGLang grammar-based structured outputs expects an EBNF"
-    ):
-        result = sync_model("foo?", CFG(EBNF_YES_NO_GRAMMAR), max_tokens=10)
-        assert isinstance(result, str)
-        assert result in ["yes", "no"]
-
-
-@pytest.mark.asyncio
-async def test_sglang_async_simple_call(async_model):
-    result = await async_model("Respond with a single word.",)
-    assert isinstance(result, str)
-
-
-@pytest.mark.asyncio
-async def test_sglang_async_streaming(async_model_no_model_name):
-    result = async_model_no_model_name.stream(
-        "Respond with a single word.",
-        model=sglang_model_name,
-    )
-    assert isinstance(result, AsyncGenerator)
-    async for chunk in result:
-        assert isinstance(chunk, str)
-        break  # Just check the first chunk
-
-
-@pytest.mark.asyncio
-async def test_sglang_async_multiple_samples(async_model):
-    result = await async_model("Respond with a single word.", n=2)
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert isinstance(result[0], str)
-    assert isinstance(result[1], str)
-
-
-@pytest.mark.asyncio
-async def test_sglang_async_json(async_model):
-    json_string = (
-        '{"type": "object", "properties":'
-        + ' {"bar": {"type": "string"}}}'
-    )
-    result = await async_model("foo?", JsonSchema(json_string), max_tokens=10)
-    assert isinstance(result, str)
-    assert "bar" in result
-
-
-@pytest.mark.asyncio
-async def test_sglang_async_regex(async_model):
-    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]{3}", result)
-
-
-@pytest.mark.asyncio
-async def test_sglang_async_cfg(async_model):
-    result = await async_model("foo?", CFG(EBNF_YES_NO_GRAMMAR), max_tokens=10)
-    assert isinstance(result, str)
-    assert result in ["yes", "no"]
diff --git a/tests/models/test_sglang_type_adapter.py b/tests/models/test_sglang_type_adapter.py
deleted file mode 100644
index 7530174f9..000000000
--- a/tests/models/test_sglang_type_adapter.py
+++ /dev/null
@@ -1,145 +0,0 @@
-import io
-import json
-import pytest
-from dataclasses import dataclass
-
-from PIL import Image
-from outlines.models.sglang import SGLangTypeAdapter
-from outlines.types import CFG, JsonSchema
-
-from outlines import Vision
-
-
-CFG_STRING = """
-?start: expr
-?expr: NUMBER
-"""
-
-JSON_SCHEMA_STRING = """
-{
-    "type": "object",
-    "properties": {
-        "answer": {"type": "number"}
-    }
-}
-"""
-
-
-@pytest.fixture
-def type_adapter():
-    return SGLangTypeAdapter()
-
-@pytest.fixture
-def cfg_instance():
-    return CFG(CFG_STRING)
-
-@pytest.fixture
-def json_schema_instance():
-    return JsonSchema(JSON_SCHEMA_STRING)
-
-@pytest.fixture
-def json_schema_whitespace_instance():
-    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-def test_sglang_type_adapter_input_text(type_adapter):
-    message = "prompt"
-    result = type_adapter.format_input(message)
-    assert isinstance(result, dict)
-    assert len(result["messages"]) == 1
-    assert result["messages"][0]["role"] == "user"
-    assert result["messages"][0]["content"] == message
-
-
-def test_sglang_type_adapter_input_vision(type_adapter, image):
-    input_message = Vision("hello", image)
-    result = type_adapter.format_input(input_message)
-    assert isinstance(result, dict)
-
-    messages = result["messages"]
-    assert len(messages) == 1
-
-    message = messages[0]
-    assert message["role"] == "user"
-    assert len(message["content"]) == 2
-    assert message["content"][0]["type"] == "text"
-    assert message["content"][0]["text"] == "hello"
-
-    assert message["content"][1]["type"] == "image_url"
-    assert (
-        message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message.image_str}"
-    )
-
-
-def test_sglang_type_adapter_input_invalid(type_adapter):
-    @dataclass
-    class Audio:
-        file: str
-
-    prompt = Audio(
-        "file",
-    )
-    with pytest.raises(TypeError, match="The input type"):
-        _ = type_adapter.format_input(prompt)
-
-
-def test_sglang_type_adapter_output_type(
-    type_adapter,
-    cfg_instance,
-    json_schema_instance,
-    json_schema_whitespace_instance,
-):
-    assert type_adapter.format_output_type(None) == {}
-    with pytest.warns(
-        UserWarning,
-        match="SGLang grammar-based structured outputs expects an EBNF"
-    ):
-        assert type_adapter.format_output_type(cfg_instance) == {
-            "extra_body": {"ebnf": CFG_STRING}
-        }
-    assert type_adapter.format_output_type(json_schema_instance) == {
-        "response_format": {
-            "type": "json_schema",
-            "json_schema": {
-                "name": "default",
-                "strict": True,
-                "schema": {
-                    **json.loads(JSON_SCHEMA_STRING),
-                    "additionalProperties": False,
-                },
-            },
-        }
-    }
-    # whitespace pattern is ignored
-    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
-        "response_format": {
-            "type": "json_schema",
-            "json_schema": {
-                "name": "default",
-                "strict": True,
-                "schema": {
-                    **json.loads(JSON_SCHEMA_STRING),
-                    "additionalProperties": False,
-                },
-            },
-        }
-    }
-    assert type_adapter.format_output_type(int) == {
-        "extra_body": {"regex": "([+-]?(0|[1-9][0-9]*))"}
-    }
diff --git a/tests/models/test_tgi.py b/tests/models/test_tgi.py
deleted file mode 100644
index 52704f6c0..000000000
--- a/tests/models/test_tgi.py
+++ /dev/null
@@ -1,180 +0,0 @@
-import os
-import re
-import warnings
-from typing import AsyncGenerator, Generator
-
-import pytest
-from huggingface_hub import InferenceClient, AsyncInferenceClient
-
-from outlines.models.tgi import TGI, AsyncTGI, from_tgi
-from outlines.types.dsl import CFG, Regex, JsonSchema
-from tests.test_utils.mock_tgi_client import MockTGIInferenceClient, MockAsyncTGIInferenceClient
-
-
-YES_NO_GRAMMAR = """
-?start: answer
-
-answer: "yes" | "no"
-"""
-
-# If the TGI_SERVER_URL environment variable is set, use the real TGI server
-# Otherwise, use the mock server
-tgi_server_url = os.environ.get("TGI_SERVER_URL")
-if tgi_server_url:
-    tgi_client = InferenceClient(tgi_server_url)
-    async_tgi_client = AsyncInferenceClient(tgi_server_url)
-else:
-    warnings.warn("No TGI server URL provided, using mock server")
-    tgi_client = MockTGIInferenceClient()
-    async_tgi_client = MockAsyncTGIInferenceClient()
-
-mock_responses = [
-    (
-        {
-            'prompt': 'Respond with a single word.',
-            'max_new_tokens': 10,
-        },
-        "foo"
-    ),
-    (
-        {
-            'prompt': 'Respond with a single word.',
-            'max_new_tokens': 10,
-            'stream': True
-        },
-        ["foo", "bar"]
-    ),
-    (
-        {
-            'prompt': 'foo?',
-            'max_new_tokens': 10,
-            'grammar': {
-                'type': 'json',
-                'value': {
-                    'type': 'object',
-                    'properties': {
-                        'bar': {'type': 'string'}
-                    },
-                    'required': ['bar']
-                }
-            }
-        },
-        '{"foo": "bar"}'
-    ),
-    (
-        {
-            'prompt': 'foo?',
-            'max_new_tokens': 10,
-            'grammar': {
-                'type': 'regex',
-                'value': '([0-9]{3})',
-            },
-        },
-        "123"
-    ),
-]
-
-# If the TGI_SERVER_URL environment variable is not set, add the mock
-# responses to the mock clients
-if not tgi_server_url:
-    async_tgi_client.add_mock_responses(mock_responses)
-    tgi_client.add_mock_responses(mock_responses)
-
-
-@pytest.fixture
-def sync_model():
-    return TGI(tgi_client)
-
-
-@pytest.fixture
-def async_model():
-    return AsyncTGI(async_tgi_client)
-
-
-def test_tgi_init():
-    model = from_tgi(
-        InferenceClient("http://localhost:11434"),
-    )
-    assert isinstance(model, TGI)
-
-    model = from_tgi(
-        AsyncInferenceClient("http://localhost:11434"),
-    )
-    assert isinstance(model, AsyncTGI)
-
-    with pytest.raises(ValueError, match="Unsupported client type"):
-        from_tgi("foo")
-
-
-def test_tgi_sync_simple_call(sync_model):
-    result = sync_model("Respond with a single word.", max_new_tokens=10)
-    assert isinstance(result, str)
-
-
-def test_tgi_sync_streaming(sync_model):
-    result = sync_model.stream(
-        "Respond with a single word.",
-        max_new_tokens=10,
-    )
-    assert isinstance(result, Generator)
-    assert isinstance(next(result), str)
-
-
-def test_tgi_sync_json(sync_model):
-    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
-    result = sync_model("foo?", JsonSchema(json_string), max_new_tokens=10)
-    assert isinstance(result, str)
-    assert "bar" in result
-
-
-def test_tgi_sync_regex(sync_model):
-    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_new_tokens=10)
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]{3}", result)
-
-
-def test_tgi_sync_cfg(sync_model):
-    with pytest.raises(
-        NotImplementedError,
-        match="TGI does not support CFG-based structured outputs",
-    ):
-        sync_model("foo?", CFG(YES_NO_GRAMMAR), max_new_tokens=10)
-
-
-@pytest.mark.asyncio
-async def test_tgi_async_simple_call(async_model):
-    result = await async_model("Respond with a single word.", max_new_tokens=10)
-    assert isinstance(result, str)
-
-
-@pytest.mark.asyncio
-async def test_tgi_async_streaming(async_model):
-    result = async_model.stream("Respond with a single word.", max_new_tokens=10)
-    assert isinstance(result, AsyncGenerator)
-    async for chunk in result:
-        assert isinstance(chunk, str)
-        break  # Just check the first chunk
-
-
-@pytest.mark.asyncio
-async def test_tgi_async_json(async_model):
-    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
-    result = await async_model("foo?", JsonSchema(json_string), max_new_tokens=10)
-    assert isinstance(result, str)
-    assert "bar" in result
-
-
-@pytest.mark.asyncio
-async def test_tgi_async_regex(async_model):
-    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_new_tokens=10)
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]{3}", result)
-
-
-@pytest.mark.asyncio
-async def test_tgi_async_cfg(async_model):
-    with pytest.raises(
-        NotImplementedError,
-        match="TGI does not support CFG-based structured outputs",
-    ):
-        await async_model("foo?", CFG(YES_NO_GRAMMAR), max_new_tokens=10)
diff --git a/tests/models/test_tgi_model_adapter.py b/tests/models/test_tgi_model_adapter.py
deleted file mode 100644
index 543ab75ff..000000000
--- a/tests/models/test_tgi_model_adapter.py
+++ /dev/null
@@ -1,93 +0,0 @@
-import json
-import pytest
-
-from outlines.models.tgi import TGITypeAdapter
-from outlines.types import CFG, JsonSchema
-
-
-CFG_STRING = """
-?start: expr
-?expr: NUMBER
-"""
-
-JSON_SCHEMA_STRING = """
-{
-    "type": "object",
-    "properties": {
-        "answer": {"type": "number"}
-    }
-}
-"""
-
-
-@pytest.fixture
-def type_adapter():
-    return TGITypeAdapter()
-
-@pytest.fixture
-def cfg_instance():
-    return CFG(CFG_STRING)
-
-@pytest.fixture
-def json_schema_instance():
-    return JsonSchema(JSON_SCHEMA_STRING)
-
-@pytest.fixture
-def json_schema_whitespace_instance():
-    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
-
-
-def test_tgi_type_adapter_input_text(type_adapter):
-    message = "prompt"
-    assert message == type_adapter.format_input(message)
-
-
-def test_tgi_type_adapter_input_invalid(type_adapter):
-    with pytest.raises(
-        NotImplementedError,
-        match="TGI does not support batch inference.",
-    ):
-        type_adapter.format_input(["foo", "bar"])
-    with pytest.raises(
-        NotImplementedError,
-        match="is not available with TGI",
-    ):
-        type_adapter.format_input({"foo": "bar"})
-
-
-def test_tgi_type_adapter_output_type(
-    type_adapter,
-    json_schema_instance,
-    json_schema_whitespace_instance,
-):
-    assert type_adapter.format_output_type(None) == {}
-    assert type_adapter.format_output_type(json_schema_instance) == {
-        "grammar": {
-            "type": "json",
-            "value": json.loads(JSON_SCHEMA_STRING),
-        }
-    }
-    # whitespace_pattern is ignored
-    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
-        "grammar": {
-            "type": "json",
-            "value": json.loads(JSON_SCHEMA_STRING),
-        }
-    }
-    assert type_adapter.format_output_type(int) == {
-        "grammar": {
-            "type": "regex",
-            "value": "([+-]?(0|[1-9][0-9]*))",
-        }
-    }
-
-
-def test_tgi_type_adapter_output_type_invalid(
-    type_adapter,
-    cfg_instance,
-):
-    with pytest.raises(
-        NotImplementedError,
-        match="TGI does not support CFG-based structured outputs.",
-    ):
-        type_adapter.format_output_type(cfg_instance)
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index cd6f60811..1404d287a 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -1,206 +1,117 @@
-import re
-from enum import Enum
-
-from pydantic import BaseModel
 import pytest
-import transformers
-
-import outlines
-from outlines.models.transformers import (
-    Transformers,
-    TransformerTokenizer,
-    TransformersTypeAdapter,
-)
-from outlines.types import Regex
-
+import torch
+from transformers import AutoTokenizer
+from transformers.models.gpt2 import GPT2TokenizerFast
 
-TEST_MODEL = "erwanf/gpt2-mini"
-TEST_MODEL_MAMBA = "hf-internal-testing/tiny-random-MambaForCausalLM"
-TEST_MODEL_BART = "trl-internal-testing/tiny-BartModel"
+from outlines.models.transformers import TransformerTokenizer, transformers
 
+TEST_MODEL = "hf-internal-testing/tiny-random-GPTJForCausalLM"
 
-def test_transformers_instantiate_invalid():
-    with pytest.raises(ValueError):
-        outlines.from_transformers(
-            transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
-            int,
-        )
 
+def test_tokenizer():
+    tokenizer = AutoTokenizer.from_pretrained(TEST_MODEL, padding_side="left")
+    tokenizer = TransformerTokenizer(tokenizer)
+    assert tokenizer.eos_token_id == 0
+    assert tokenizer.pad_token_id == 0
+    assert isinstance(tokenizer.tokenizer, GPT2TokenizerFast)
 
-def test_transformers_instantiate_simple():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
-        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-    )
-    assert isinstance(model, Transformers)
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert isinstance(model.type_adapter, TransformersTypeAdapter)
-    assert model.tensor_library_name == "torch"
-
+    token_ids, attention_mask = tokenizer.encode("Test")
+    assert token_ids.ndim == 2
+    assert token_ids.shape[0] == 1
+    assert isinstance(token_ids, torch.LongTensor)
+    assert token_ids.shape == attention_mask.shape
 
-def test_transformers_instantiate_flax_model():
-    model = outlines.from_transformers(
-        transformers.FlaxAutoModelForCausalLM.from_pretrained(TEST_MODEL),
-        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-    )
-    assert isinstance(model, Transformers)
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert isinstance(model.type_adapter, TransformersTypeAdapter)
-    assert model.tensor_library_name == "jax"
+    token_ids, attention_mask = tokenizer.encode(["Test", "Test"])
+    assert token_ids.ndim == 2
+    assert token_ids.shape[0] == 2
+    assert isinstance(token_ids, torch.LongTensor)
+    assert token_ids.shape == attention_mask.shape
 
+    token_ids, attention_mask = tokenizer.encode(["Test", "A long sentence"])
+    assert token_ids.shape == attention_mask.shape
+    assert attention_mask[0][0] == tokenizer.pad_token_id
 
-def test_transformers_instantiate_tensorflow_model():
-    model = outlines.from_transformers(
-        transformers.TFAutoModelForCausalLM.from_pretrained(TEST_MODEL),
-        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-    )
-    assert isinstance(model, Transformers)
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert isinstance(model.type_adapter, TransformersTypeAdapter)
-    assert model.tensor_library_name == "tensorflow"
-
-
-def test_transformers_instantiate_mamba():
-    model = outlines.from_transformers(
-        transformers.MambaForCausalLM.from_pretrained(TEST_MODEL_MAMBA),
-        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-    )
-    assert isinstance(model, Transformers)
-
-
-def test_transformers_instantiate_tokenizer_kwargs():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
-        transformers.AutoTokenizer.from_pretrained(
-            TEST_MODEL, additional_special_tokens=["<t1>", "<t2>"]
-        ),
-    )
-    assert "<t1>" in model.tokenizer.special_tokens
-    assert "<t2>" in model.tokenizer.special_tokens
-
-
-@pytest.fixture
-def model():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
-        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-    )
-    return model
+    text = tokenizer.decode(torch.tensor([[0, 1, 2]]))
+    isinstance(text, str)
 
+    text = tokenizer.decode(torch.tensor([[0, 1, 2], [3, 4, 5]]))
+    isinstance(text, list)
+    isinstance(text[0], str)
+    isinstance(text[1], str)
 
-@pytest.fixture
-def model_bart():
-    model = outlines.from_transformers(
-        transformers.BartForConditionalGeneration.from_pretrained(TEST_MODEL_BART),
-        transformers.BartTokenizer.from_pretrained(TEST_MODEL_BART),
+    tokenizer = AutoTokenizer.from_pretrained(
+        TEST_MODEL, additional_special_tokens=["<t1>", "<t2>"]
     )
-    return model
+    tokenizer = TransformerTokenizer(tokenizer)
+    assert "<t1>" in tokenizer.special_tokens
+    assert "<t2>" in tokenizer.special_tokens
 
 
-def test_transformers_simple(model):
-    result = model.generate("Respond with one word. Not more.", None)
-    assert isinstance(result, str)
+def test_llama_tokenizer():
+    tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/llama-tokenizer")
+    tokenizer = TransformerTokenizer(tokenizer)
 
+    # Broken
+    assert tokenizer.tokenizer.convert_tokens_to_string(["▁baz"]) == "baz"
+    assert tokenizer.tokenizer.convert_tokens_to_string(["<0x20>"]) == ""
+    assert tokenizer.tokenizer.convert_tokens_to_string(["▁▁▁"]) == "  "
 
-def test_transformers_call(model, model_bart):
-    result = model("Respond with one word. Not more.")
-    assert isinstance(result, str)
+    # Not broken
+    assert tokenizer.convert_token_to_string("▁baz") == " baz"
+    assert tokenizer.convert_token_to_string("<0x20>") == " "
+    assert tokenizer.convert_token_to_string("▁▁▁") == "   "
 
-    result = model_bart("Respond with one word. Not more.")
-    assert isinstance(result, str)
 
+def test_model():
+    model = transformers(TEST_MODEL, device="cpu")
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert model.model.device.type == "cpu"
 
-def test_transformers_inference_kwargs(model):
-    result = model("Respond with one word. Not more.", max_new_tokens=100)
-    assert isinstance(result, str)
-
-
-def test_transformers_invalid_inference_kwargs(model):
-    with pytest.raises(ValueError):
-        model("Respond with one word. Not more.", foo="bar")
-
-
-def test_transformers_regex(model):
-    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]", result)
-
-
-def test_transformers_json(model):
-    class Character(BaseModel):
-        name: str
-
-    result = model("Create a character with a name.", Character)
-    assert "name" in result
-
-
-def test_transformers_choice(model):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
-
-    result = model("Cat or dog?", Foo)
-    assert result in ["cat", "dog"]
-
+    model = transformers(TEST_MODEL, model_kwargs={"device_map": "cpu"})
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert model.model.device.type == "cpu"
 
-def test_transformers_batch_samples(model):
-    result = model("Respond with one word. Not more.")
-    assert isinstance(result, str)
-    result = model(
-        "Respond with one word. Not more.", num_return_sequences=2, num_beams=2
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    result = model(
-        ["Respond with one word. Not more.", "Respond with one word. Not more."]
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    result = model(
-        ["Respond with one word. Not more.", "Respond with one word. Not more."],
-        num_return_sequences=2,
-        num_beams=2,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
-
-
-def test_transformers_batch_samples_constrained(model):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
-
-    result = model("Cat or dog?", Foo, num_return_sequences=2, num_beams=2)
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert result[0] in ["cat", "dog"]
-    assert result[1] in ["cat", "dog"]
-    result = model(
-        ["Cat or dog?", "Cat or dog?"],
-        Foo,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert result[0] in ["cat", "dog"]
-    assert result[1] in ["cat", "dog"]
-    result = model(
-        ["Cat or dog?", "Cat or dog?"],
-        Foo,
-        num_return_sequences=2,
-        num_beams=2,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
-        assert item[0] in ["cat", "dog"]
-        assert item[1] in ["cat", "dog"]
-
-
-def test_transformers_streaming(model):
-    with pytest.raises(NotImplementedError, match="Streaming is not implemented"):
-        model.stream("Respond with one word. Not more.")
+    model = transformers(TEST_MODEL, device="cpu", model_kwargs={"device_map": "cuda"})
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert model.model.device.type == "cpu"
+
+    input_ids = torch.tensor([[0, 1, 2]])
+    logits, kv_cache = model(input_ids, torch.ones_like(input_ids))
+    assert logits.type() == "torch.FloatTensor"
+    assert logits.ndim == 2
+    assert logits.shape[0] == 1
+    assert len(kv_cache) == model.model.config.n_layer
+    assert len(kv_cache[0]) == 2
+    assert kv_cache[0][0].shape[1] == model.model.config.n_head
+    assert kv_cache[0][0].shape[2] == 3  # number of tokens
+
+    input_ids = torch.tensor([[0, 1, 2], [3, 4, 5], [6, 7, 8]])
+    logits, kv_cache = model(input_ids, torch.ones_like(input_ids))
+    assert logits.type() == "torch.FloatTensor"
+    assert logits.ndim == 2
+    assert logits.shape[0] == 3
+    assert len(kv_cache) == model.model.config.n_layer
+    assert len(kv_cache[0]) == 2
+    assert kv_cache[0][0].shape[1] == model.model.config.n_head
+    assert kv_cache[0][0].shape[2] == 3  # number of tokens
+
+    with pytest.raises(AssertionError):
+        input_ids = torch.tensor([[[0, 1, 2], [3, 4, 5]], [[6, 7, 8], [0, 1, 2]]])
+        logits = model(input_ids, torch.ones_like(input_ids))
+
+
+def test_tokenizer_eq_hash():
+    tokenizer_hf = AutoTokenizer.from_pretrained("gpt2")
+
+    tokenizer = TransformerTokenizer(tokenizer_hf)
+    tokenizer_2 = TransformerTokenizer(tokenizer_hf)
+
+    assert tokenizer == tokenizer_2
+    assert hash(tokenizer) == hash(tokenizer_2)
+
+    tokenizer_hf_2 = AutoTokenizer.from_pretrained("gpt2")
+    tokenizer_hf_2.add_tokens(["test_token"])
+
+    tokenizer_3 = TransformerTokenizer(tokenizer_hf_2)
+    assert tokenizer != tokenizer_3
+    assert hash(tokenizer) != hash(tokenizer_3)
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
deleted file mode 100644
index 448bec221..000000000
--- a/tests/models/test_transformers_multimodal.py
+++ /dev/null
@@ -1,194 +0,0 @@
-# we only test vision models here as audio models are too heavy to run on CI
-
-import re
-from enum import Enum
-from io import BytesIO
-from urllib.request import urlopen
-
-import pytest
-from PIL import Image
-from pydantic import BaseModel
-from transformers import (
-    LlavaForConditionalGeneration,
-    AutoProcessor,
-)
-
-import outlines
-from outlines.models.transformers import (
-    TransformersMultiModal,
-    TransformerTokenizer,
-    TransformersMultiModalTypeAdapter,
-)
-from outlines.types import Regex
-
-TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-IMAGE_URLS = [
-    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
-    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
-]
-
-
-@pytest.fixture
-def images():
-    def img_from_url(url):
-        img_byte_stream = BytesIO(urlopen(url).read())
-        return Image.open(img_byte_stream).convert("RGB")
-
-    return [img_from_url(url) for url in IMAGE_URLS]
-
-
-@pytest.fixture
-def model():
-    return outlines.from_transformers(
-        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
-        AutoProcessor.from_pretrained(TEST_MODEL),
-    )
-
-
-def test_transformers_vision_instantiate_simple():
-    model = outlines.from_transformers(
-        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
-        AutoProcessor.from_pretrained(TEST_MODEL),
-    )
-    assert isinstance(model, TransformersMultiModal)
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert isinstance(model.type_adapter, TransformersMultiModalTypeAdapter)
-    assert model.tensor_library_name == "torch"
-
-
-def test_transformers_vision_simple(model, images):
-    result = model.generate(
-        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
-        None,
-        max_new_tokens=2,
-    )
-    assert isinstance(result, str)
-
-
-def test_transformers_vision_call(model, images):
-    result = model(
-        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
-        max_new_tokens=2,
-    )
-    assert isinstance(result, str)
-
-
-def test_transformers_vision_wrong_number_images(model, images):
-    with pytest.raises(ValueError):
-        model(
-            {
-                "text": "<image>Describe this image in one sentence:",
-                "images": [images[0], images[1]],
-            }
-        )
-
-
-def test_transformers_vision_wrong_input_type(model):
-    with pytest.raises(NotImplementedError):
-        model.generate("invalid input", None)
-
-
-def test_transformers_inference_kwargs(model, images):
-    result = model(
-        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
-        max_new_tokens=2,
-    )
-    assert isinstance(result, str)
-
-
-def test_transformers_invalid_inference_kwargs(model, images):
-    with pytest.raises(ValueError):
-        model(
-            {
-                "text": "<image>Describe this image in one sentence:",
-                "images": images[0],
-            },
-            foo="bar",
-        )
-
-
-def test_transformers_several_images(model, images):
-    result = model(
-        {
-            "text": "<image><image>Describe this image in one sentence:",
-            "images": [images[0], images[1]],
-        },
-        max_new_tokens=2,
-    )
-    assert isinstance(result, str)
-
-
-def test_transformers_vision_json(model, images):
-    class Foo(BaseModel):
-        name: str
-
-    result = model(
-        {"text": "<image>Give a name to this animal.", "images": images[0]},
-        Foo,
-        max_new_tokens=10,
-    )
-    assert "name" in result
-
-
-def test_transformers_vision_regex(model, images):
-    result = model(
-        {"text": "<image>How old is it?", "images": images[0]},
-        Regex(r"[0-9]")
-    )
-
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]", result)
-
-
-def test_transformers_vision_choice(model, images):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
-
-    result = model(
-        {"text": "<image>Is it a cat or a dog?", "images": images[0]},
-        Foo,
-    )
-
-    assert isinstance(result, str)
-    assert result in ["cat", "dog"]
-
-
-def test_transformers_vision_batch_samples(model, images):
-    result = model(
-        {"text": "<image>Describe this image in one sentence.", "images": images[0]},
-        num_return_sequences=2,
-        num_beams=2,
-        max_new_tokens=2,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    result = model(
-        {
-            "text": [
-                "<image>Describe this image in one sentence.",
-                "<image>Describe this image in one sentence.",
-            ],
-            "images": [images[0], images[1]],
-        },
-        max_new_tokens=2,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    result = model(
-        {
-            "text": [
-                "<image>Describe this image in one sentence.",
-                "<image>Describe this image in one sentence.",
-            ],
-            "images": [images[0], images[1]],
-        },
-        num_return_sequences=2,
-        num_beams=2,
-        max_new_tokens=2,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
deleted file mode 100644
index 28569d25f..000000000
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ /dev/null
@@ -1,48 +0,0 @@
-import pytest
-
-import outlines
-import transformers
-from transformers import LogitsProcessorList
-
-from outlines.models.transformers import TransformersMultiModalTypeAdapter
-from outlines.processors.structured import RegexLogitsProcessor
-
-
-@pytest.fixture
-def adapter():
-    return TransformersMultiModalTypeAdapter()
-
-
-@pytest.fixture
-def logits_processor():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=model.tokenizer,
-        tensor_library_name="torch",
-    )
-
-
-def test_transformers_multimodal_type_adapter_format_input(adapter):
-    with pytest.raises(NotImplementedError):
-        adapter.format_input("hello")
-
-    with pytest.raises(ValueError):
-        adapter.format_input({"foo": "bar"})
-
-    assert adapter.format_input({"text": "foo"}) == {"text": "foo"}
-
-
-def test_transformers_multimodal_type_adapter_format_output_type(
-    adapter, logits_processor
-):
-    formatted = adapter.format_output_type(logits_processor)
-    assert isinstance(formatted, LogitsProcessorList)
-    assert len(formatted) == 1
-    assert isinstance(formatted[0], RegexLogitsProcessor)
-
-    formatted = adapter.format_output_type(None)
-    assert formatted is None
diff --git a/tests/models/test_transformers_tokenizer.py b/tests/models/test_transformers_tokenizer.py
deleted file mode 100644
index bf3452302..000000000
--- a/tests/models/test_transformers_tokenizer.py
+++ /dev/null
@@ -1,142 +0,0 @@
-import pytest
-
-import transformers
-
-from outlines.models.transformers import (
-    get_llama_tokenizer_types,
-    TransformerTokenizer,
-)
-
-
-TEST_MODEL = "erwanf/gpt2-mini"
-TEST_MODEL_SEQ2SEQ = "hf-internal-testing/tiny-random-t5"
-
-
-@pytest.fixture
-def tokenizer():
-    return transformers.AutoTokenizer.from_pretrained(TEST_MODEL)
-
-
-@pytest.fixture
-def tokenizer_no_pad_token_id(tokenizer):
-    tokenizer.pad_token_id = None
-    return tokenizer
-
-
-@pytest.fixture
-def tokenizer_seq2seq():
-    return transformers.AutoTokenizer.from_pretrained(TEST_MODEL_SEQ2SEQ)
-
-
-@pytest.fixture
-def transformer_tokenizer(tokenizer):
-    return TransformerTokenizer(tokenizer)
-
-
-@pytest.fixture
-def another_transformer_tokenizer(tokenizer):
-    return TransformerTokenizer(tokenizer)
-
-
-@pytest.fixture
-def transformer_tokenizer_seq2seq(tokenizer_seq2seq):
-    return TransformerTokenizer(tokenizer_seq2seq)
-
-
-def test_get_llama_tokenizer_types():
-    tokenizer_types = get_llama_tokenizer_types()
-    assert tokenizer_types[0] is transformers.models.llama.LlamaTokenizer
-    assert tokenizer_types[1] is transformers.models.llama.LlamaTokenizerFast
-    assert tokenizer_types[2] is transformers.models.code_llama.CodeLlamaTokenizer
-    assert tokenizer_types[3] is transformers.models.code_llama.CodeLlamaTokenizerFast
-
-
-def test_transformer_tokenizer_init(
-    tokenizer,
-    tokenizer_no_pad_token_id
-):
-    # tokenizer with a pad_token_id
-    transformer_tokenizer = TransformerTokenizer(tokenizer)
-    assert transformer_tokenizer.tokenizer == tokenizer
-    assert transformer_tokenizer.eos_token_id == tokenizer.eos_token_id
-    assert transformer_tokenizer.pad_token_id == tokenizer.pad_token_id
-    assert transformer_tokenizer.special_tokens == set(tokenizer.all_special_tokens)
-    assert transformer_tokenizer.vocabulary == tokenizer.get_vocab()
-
-    # tokenizer with no pad_token_id
-    transformer_tokenizer_no_pad_token_id = TransformerTokenizer(tokenizer_no_pad_token_id)
-    assert transformer_tokenizer_no_pad_token_id.tokenizer == tokenizer_no_pad_token_id
-    assert transformer_tokenizer_no_pad_token_id.eos_token_id == tokenizer_no_pad_token_id.eos_token_id
-    assert transformer_tokenizer_no_pad_token_id.pad_token_id == tokenizer_no_pad_token_id.eos_token_id
-    assert transformer_tokenizer_no_pad_token_id.special_tokens == set(tokenizer_no_pad_token_id.all_special_tokens)
-    assert transformer_tokenizer_no_pad_token_id.vocabulary == tokenizer_no_pad_token_id.get_vocab()
-
-
-def test_transformer_tokenizer_encode(transformer_tokenizer):
-    input_ids, attention_mask = transformer_tokenizer.encode("Hello, world!")
-    assert input_ids is not None
-    assert attention_mask is not None
-    assert input_ids.shape == attention_mask.shape
-
-
-def test_transformer_tokenizer_decode(transformer_tokenizer):
-    input_ids, _ = transformer_tokenizer.encode("Hello, world!")
-    decoded_text = transformer_tokenizer.decode(input_ids)
-    assert isinstance(decoded_text, list)
-    assert "Hello, world!" in decoded_text[0]
-
-
-def test_transformer_tokenizer_convert_token_to_string(transformer_tokenizer):
-    # regular
-    transformer_tokenizer.is_llama = False
-    token = transformer_tokenizer.tokenizer.tokenize("Hello")[0]
-    string = transformer_tokenizer.convert_token_to_string(token)
-    assert isinstance(string, str)
-    assert "Hello" in string
-
-    # is_llama + <0x20>
-    transformer_tokenizer.is_llama = True
-    string = transformer_tokenizer.convert_token_to_string("<0x20>")
-    assert isinstance(string, str)
-    assert " " in string
-
-
-def test_transformer_tokenizer_eq(
-    transformer_tokenizer,
-    another_transformer_tokenizer,
-    transformer_tokenizer_seq2seq,
-):
-    # different types of object
-    assert transformer_tokenizer.__eq__(1) == NotImplemented
-
-    # regular case
-    assert transformer_tokenizer == another_transformer_tokenizer
-    assert transformer_tokenizer != transformer_tokenizer_seq2seq
-
-    # with model name and kwargs attributes
-    transformer_tokenizer.model_name = "foo"
-    transformer_tokenizer.kwargs = {"foo": "bar"}
-    another_transformer_tokenizer.model_name = "foo"
-    another_transformer_tokenizer.kwargs = {"foo": "bar"}
-    assert transformer_tokenizer == another_transformer_tokenizer
-
-
-def test_transformer_tokenizer_hash(
-    transformer_tokenizer,
-    another_transformer_tokenizer,
-    transformer_tokenizer_seq2seq,
-):
-    assert isinstance(hash(transformer_tokenizer), int)
-    assert hash(transformer_tokenizer) == hash(another_transformer_tokenizer)
-    assert hash(transformer_tokenizer) != hash(transformer_tokenizer_seq2seq)
-
-
-def test_transformer_tokenizer_getstate_setstate(
-    transformer_tokenizer,
-    another_transformer_tokenizer,
-):
-    state = transformer_tokenizer.__getstate__()
-    assert "tokenizer" in state
-
-    another_transformer_tokenizer.__setstate__(state)
-    assert another_transformer_tokenizer == transformer_tokenizer
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
deleted file mode 100644
index 44f6ec4e6..000000000
--- a/tests/models/test_transformers_type_adapter.py
+++ /dev/null
@@ -1,50 +0,0 @@
-import pytest
-
-import transformers
-from transformers import LogitsProcessorList
-
-import outlines
-from outlines.models.transformers import TransformersTypeAdapter
-from outlines.processors.structured import RegexLogitsProcessor
-
-
-@pytest.fixture
-def adapter():
-    return TransformersTypeAdapter()
-
-
-@pytest.fixture
-def logits_processor():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=model.tokenizer,
-        tensor_library_name="torch",
-    )
-
-
-def test_transformers_type_adapter_format_input(adapter):
-    with pytest.raises(NotImplementedError):
-        adapter.format_input(int)
-
-    assert adapter.format_input("Hello, world!") == "Hello, world!"
-
-    assert adapter.format_input(["Hello, world!", "Hello, world!"]) == [
-        "Hello, world!",
-        "Hello, world!",
-    ]
-
-
-def test_transformers_type_adapter_format_output_type(
-    adapter, logits_processor
-):
-    formatted = adapter.format_output_type(logits_processor)
-    assert isinstance(formatted, LogitsProcessorList)
-    assert len(formatted) == 1
-    assert isinstance(formatted[0], RegexLogitsProcessor)
-
-    formatted = adapter.format_output_type(None)
-    assert formatted is None
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
deleted file mode 100644
index 76a9da814..000000000
--- a/tests/models/test_vllm.py
+++ /dev/null
@@ -1,254 +0,0 @@
-import os
-import re
-import warnings
-from typing import AsyncGenerator, Generator
-
-import pytest
-from openai import AsyncOpenAI, OpenAI
-
-from outlines.models.vllm import VLLM, AsyncVLLM, from_vllm
-from outlines.types.dsl import CFG, Regex, JsonSchema
-from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
-
-
-YES_NO_GRAMMAR = """
-?start: answer
-
-answer: "yes" | "no"
-"""
-
-
-# If the VLLM_SERVER_URL environment variable is set, use the real vLLM server
-# Otherwise, use the mock server
-vllm_server_url = os.environ.get("VLLM_SERVER_URL")
-vllm_model_name = os.environ.get(
-    "VLLM_MODEL_NAME", "microsoft/Phi-3-mini-4k-instruct"
-)
-if vllm_server_url:
-    openai_client = OpenAI(base_url=vllm_server_url)
-    async_openai_client = AsyncOpenAI(base_url=vllm_server_url)
-else:
-    warnings.warn("No VLLM server URL provided, using mock server")
-    openai_client = MockOpenAIClient()
-    async_openai_client = MockAsyncOpenAIClient()
-
-mock_responses = [
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Respond with a single word.'}
-            ],
-            'model': vllm_model_name,
-        },
-        "foo"
-    ),
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Respond with a single word.'}
-            ],
-            'model': vllm_model_name,
-            'stream': True
-        },
-        ["foo", "bar"]
-    ),
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Respond with a single word.'}
-            ],
-            'n': 2,
-            'model': vllm_model_name,
-        },
-        ["foo", "bar"]
-    ),
-    (
-        {
-            'messages': [{'role': "user", 'content': 'foo?'}],
-            'model': vllm_model_name,
-            'max_tokens': 10,
-            'extra_body': {
-            'guided_json': {
-                'type': 'object',
-                'properties': {
-                    'bar': {'type': 'string'}
-                }
-            },
-            }
-        },
-        '{"foo": "bar"}'
-    ),
-    (
-        {
-            'messages': [{'role': "user", 'content': 'foo?'}],
-            'model': vllm_model_name,
-            'max_tokens': 10,
-            'extra_body': {
-                'guided_regex': '([0-9]{3})',
-            },
-        },
-        "123"
-    ),
-    (
-        {
-            'messages': [{'role': "user", 'content': 'foo?'}],
-            'model': vllm_model_name,
-            'max_tokens': 10,
-            'extra_body': {
-                'guided_grammar': YES_NO_GRAMMAR,
-            },
-        },
-        "yes"
-    ),
-]
-
-
-# If the VLLM_SERVER_URL environment variable is not set, add the mock
-# responses to the mock clients
-if not vllm_server_url:
-    async_openai_client.add_mock_responses(mock_responses)
-    openai_client.add_mock_responses(mock_responses)
-
-
-@pytest.fixture
-def sync_model():
-    return VLLM(openai_client, vllm_model_name)
-
-
-@pytest.fixture
-def sync_model_no_model_name():
-    return VLLM(openai_client)
-
-
-@pytest.fixture
-def async_model():
-    return AsyncVLLM(async_openai_client, vllm_model_name)
-
-
-@pytest.fixture
-def async_model_no_model_name():
-    return AsyncVLLM(async_openai_client)
-
-
-def test_vllm_init():
-    # We do not rely on the mock server here because we need an object
-    # of type OpenAI and AsyncOpenAI to test the init function.
-    openai_client = OpenAI(base_url="http://localhost:11434")
-    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
-
-    # Sync with model name
-    model = from_vllm(openai_client, vllm_model_name)
-    assert isinstance(model, VLLM)
-    assert model.client == openai_client
-    assert model.model_name == vllm_model_name
-
-    # Sync without model name
-    model = from_vllm(openai_client)
-    assert isinstance(model, VLLM)
-    assert model.client == openai_client
-    assert model.model_name is None
-
-    # Async with model name
-    model = from_vllm(async_openai_client, vllm_model_name)
-    assert isinstance(model, AsyncVLLM)
-    assert model.client == async_openai_client
-    assert model.model_name == vllm_model_name
-
-    # Async without model name
-    model = from_vllm(async_openai_client)
-    assert isinstance(model, AsyncVLLM)
-    assert model.client == async_openai_client
-    assert model.model_name is None
-
-    with pytest.raises(ValueError, match="Unsupported client type"):
-        from_vllm("foo")
-
-
-def test_vllm_sync_simple_call(sync_model):
-    result = sync_model("Respond with a single word.",)
-    assert isinstance(result, str)
-
-
-def test_vllm_sync_streaming(sync_model_no_model_name):
-    result = sync_model_no_model_name.stream(
-        "Respond with a single word.",
-        model=vllm_model_name,
-    )
-    assert isinstance(result, Generator)
-    assert isinstance(next(result), str)
-
-
-def test_vllm_sync_multiple_samples(sync_model):
-    result = sync_model("Respond with a single word.", n=2)
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert isinstance(result[0], str)
-    assert isinstance(result[1], str)
-
-
-def test_vllm_sync_json(sync_model):
-    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}}'
-    result = sync_model("foo?", JsonSchema(json_string), max_tokens=10)
-    assert isinstance(result, str)
-    assert "bar" in result
-
-
-def test_vllm_sync_regex(sync_model):
-    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]{3}", result)
-
-
-def test_vllm_sync_cfg(sync_model):
-    result = sync_model("foo?", CFG(YES_NO_GRAMMAR), max_tokens=10)
-    assert isinstance(result, str)
-    assert result in ["yes", "no"]
-
-
-@pytest.mark.asyncio
-async def test_vllm_async_simple_call(async_model):
-    result = await async_model("Respond with a single word.",)
-    assert isinstance(result, str)
-
-
-@pytest.mark.asyncio
-async def test_vllm_async_streaming(async_model_no_model_name):
-    result = async_model_no_model_name.stream(
-        "Respond with a single word.",
-        model=vllm_model_name,
-    )
-    assert isinstance(result, AsyncGenerator)
-    async for chunk in result:
-        assert isinstance(chunk, str)
-        break  # Just check the first chunk
-
-
-@pytest.mark.asyncio
-async def test_vllm_async_multiple_samples(async_model):
-    result = await async_model("Respond with a single word.", n=2)
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert isinstance(result[0], str)
-    assert isinstance(result[1], str)
-
-
-@pytest.mark.asyncio
-async def test_vllm_async_json(async_model):
-    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}}'
-    result = await async_model("foo?", JsonSchema(json_string), max_tokens=10)
-    assert isinstance(result, str)
-    assert "bar" in result
-
-
-@pytest.mark.asyncio
-async def test_vllm_async_regex(async_model):
-    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]{3}", result)
-
-
-@pytest.mark.asyncio
-async def test_vllm_async_cfg(async_model):
-    result = await async_model("foo?", CFG(YES_NO_GRAMMAR), max_tokens=10)
-    assert isinstance(result, str)
-    assert result in ["yes", "no"]
diff --git a/tests/models/test_vllm_offline.py b/tests/models/test_vllm_offline.py
deleted file mode 100644
index 01d5d6a59..000000000
--- a/tests/models/test_vllm_offline.py
+++ /dev/null
@@ -1,117 +0,0 @@
-import re
-from enum import Enum
-
-import pytest
-from pydantic import BaseModel
-
-try:
-    from vllm import LLM, SamplingParams
-    HAS_VLLM = True
-except ImportError:
-    HAS_VLLM = False
-
-import outlines
-from outlines.models.vllm_offline import (
-    VLLMOffline,
-    VLLMOfflineTypeAdapter,
-    from_vllm_offline
-)
-from outlines.types import Regex
-
-
-TEST_MODEL = "erwanf/gpt2-mini"
-
-pytestmark = pytest.mark.skipif(
-    not HAS_VLLM,
-    reason="vLLM models can only be run on GPU."
-)
-
-
-def test_vllm_model_initialization():
-    model = from_vllm_offline(LLM(TEST_MODEL))
-    assert isinstance(model, VLLMOffline)
-    assert isinstance(model.model, LLM)
-    assert isinstance(model.type_adapter, VLLMOfflineTypeAdapter)
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    model = outlines.from_vllm_offline(LLM(TEST_MODEL))
-    return model
-
-
-def test_vllm_simple(model):
-    result = model.generate("Respond with one word. Not more.", None)
-    assert isinstance(result, str)
-
-
-def test_vllm_call(model):
-    result = model("Respond with one word. Not more.")
-    assert isinstance(result, str)
-
-
-def test_vllm_inference_kwargs(model):
-    result = model("Write a short story about a cat.", sampling_params=SamplingParams(max_tokens=2), use_tqdm=True)
-    assert isinstance(result, str)
-    assert len(result) <= 20
-
-
-def test_vllm_invalid_inference_kwargs(model):
-    with pytest.raises(TypeError):
-        model("Respond with one word. Not more.", foo="bar")
-
-
-def test_vllm_regex(model):
-    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
-    assert isinstance(result, str)
-    assert re.match(r"[0-9]", result)
-
-
-def test_vllm_json(model):
-    class Character(BaseModel):
-        name: str
-
-    result = model("Create a character with a name.", Character)
-    assert "name" in result
-
-
-def test_vllm_choice(model):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
-
-    result = model("Cat or dog?", Foo)
-    assert result in ["cat", "dog"]
-
-
-def test_vllm_batch_samples(model):
-    result = model(
-        "Respond with one word. Not more.",
-        sampling_params=SamplingParams(n=2)
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-
-    result = model(
-        ["Respond with one word. Not more.", "Respond with one word. Not more."]
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-
-    result = model(
-        ["Respond with one word. Not more.", "Respond with one word. Not more."],
-        sampling_params=SamplingParams(n=2)
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
-
-
-def test_vllm_streaming(model):
-    with pytest.raises(
-        NotImplementedError,
-        match="Streaming is not available for the vLLM integration."
-    ):
-        model.stream("Respond with one word. Not more.")
diff --git a/tests/models/test_vllm_offline_type_adapter.py b/tests/models/test_vllm_offline_type_adapter.py
deleted file mode 100644
index b8a18b6d1..000000000
--- a/tests/models/test_vllm_offline_type_adapter.py
+++ /dev/null
@@ -1,84 +0,0 @@
-import io
-import json
-from dataclasses import dataclass
-
-import pytest
-
-from outlines.models.vllm_offline import VLLMOfflineTypeAdapter
-from outlines.types import CFG, JsonSchema, Regex
-
-
-CFG_STRING = """
-?start: expr
-?expr: NUMBER
-"""
-
-JSON_SCHEMA_STRING = """
-{
-    "type": "object",
-    "properties": {
-        "answer": {"type": "number"}
-    }
-}
-"""
-
-
-@pytest.fixture
-def type_adapter():
-    return VLLMOfflineTypeAdapter()
-
-@pytest.fixture
-def cfg_instance():
-    return CFG(CFG_STRING)
-
-@pytest.fixture
-def json_schema_instance():
-    return JsonSchema(JSON_SCHEMA_STRING)
-
-@pytest.fixture
-def json_schema_whitespace_instance():
-    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
-
-@pytest.fixture
-def regex_instance():
-    return Regex(r"[0-9]+")
-
-
-def test_vllm_type_adapter_input_text(type_adapter):
-    # string
-    message = "prompt"
-    result = type_adapter.format_input(message)
-    assert message == result
-
-    # list of strings
-    messages = ["foo", "bar"]
-    result = type_adapter.format_input(messages)
-    assert messages == result
-
-
-def test_vllm_type_adapter_input_invalid(type_adapter):
-    with pytest.raises(NotImplementedError):
-        _ = type_adapter.format_input({"foo": "bar"})
-
-
-def test_vllm_type_adapter_output_type(
-    type_adapter,
-    cfg_instance,
-    json_schema_instance,
-    json_schema_whitespace_instance,
-    regex_instance,
-):
-    assert type_adapter.format_output_type(None) == {}
-    assert type_adapter.format_output_type(cfg_instance) == {
-        "grammar": CFG_STRING
-    }
-    assert type_adapter.format_output_type(json_schema_instance) == {
-        "json": json.loads(JSON_SCHEMA_STRING)
-    }
-    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
-        "json": json.loads(JSON_SCHEMA_STRING),
-        "whitespace_pattern": "\n"
-    }
-    assert type_adapter.format_output_type(regex_instance) == {
-        "regex": "([0-9]+)"
-    }
diff --git a/tests/models/test_vllm_type_adapter.py b/tests/models/test_vllm_type_adapter.py
deleted file mode 100644
index 46d021a8c..000000000
--- a/tests/models/test_vllm_type_adapter.py
+++ /dev/null
@@ -1,121 +0,0 @@
-import io
-import json
-import pytest
-from dataclasses import dataclass
-
-from PIL import Image
-from outlines.models.vllm import VLLMTypeAdapter
-from outlines.types import CFG, JsonSchema
-
-from outlines import Vision
-
-
-CFG_STRING = """
-?start: expr
-?expr: NUMBER
-"""
-
-JSON_SCHEMA_STRING = """
-{
-    "type": "object",
-    "properties": {
-        "answer": {"type": "number"}
-    }
-}
-"""
-
-
-@pytest.fixture
-def type_adapter():
-    return VLLMTypeAdapter()
-
-@pytest.fixture
-def cfg_instance():
-    return CFG(CFG_STRING)
-
-@pytest.fixture
-def json_schema_instance():
-    return JsonSchema(JSON_SCHEMA_STRING)
-
-@pytest.fixture
-def json_schema_whitespace_instance():
-    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
-
-
-@pytest.fixture
-def image():
-    width, height = 1, 1
-    white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = Image.open(buffer)
-
-    return image
-
-
-def test_vllm_type_adapter_input_text(type_adapter):
-    message = "prompt"
-    result = type_adapter.format_input(message)
-    assert isinstance(result, dict)
-    assert len(result["messages"]) == 1
-    assert result["messages"][0]["role"] == "user"
-    assert result["messages"][0]["content"] == message
-
-
-def test_vllm_type_adapter_input_vision(type_adapter, image):
-    input_message = Vision("hello", image)
-    result = type_adapter.format_input(input_message)
-    assert isinstance(result, dict)
-
-    messages = result["messages"]
-    assert len(messages) == 1
-
-    message = messages[0]
-    assert message["role"] == "user"
-    assert len(message["content"]) == 2
-    assert message["content"][0]["type"] == "text"
-    assert message["content"][0]["text"] == "hello"
-
-    assert message["content"][1]["type"] == "image_url"
-    assert (
-        message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message.image_str}"
-    )
-
-
-def test_vllm_type_adapter_input_invalid(type_adapter):
-    @dataclass
-    class Audio:
-        file: str
-
-    prompt = Audio(
-        "file",
-    )
-    with pytest.raises(TypeError, match="The input type"):
-        _ = type_adapter.format_input(prompt)
-
-
-def test_vllm_type_adapter_output_type(
-    type_adapter,
-    cfg_instance,
-    json_schema_instance,
-    json_schema_whitespace_instance,
-):
-    assert type_adapter.format_output_type(None) == {}
-    assert type_adapter.format_output_type(cfg_instance) == {
-        "guided_grammar": CFG_STRING
-    }
-    assert type_adapter.format_output_type(json_schema_instance) == {
-        "guided_json": json.loads(JSON_SCHEMA_STRING)
-    }
-    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
-        "guided_json": json.loads(JSON_SCHEMA_STRING),
-        "whitespace_pattern": "\n"
-    }
-    assert type_adapter.format_output_type(int) == {
-        "guided_regex": "([+-]?(0|[1-9][0-9]*))"
-    }
diff --git a/tests/processors/test_base_processor.py b/tests/processors/test_base_processor.py
index 3b729410a..d2a1e1af2 100644
--- a/tests/processors/test_base_processor.py
+++ b/tests/processors/test_base_processor.py
@@ -7,96 +7,74 @@
 
 from outlines.processors.base_logits_processor import OutlinesLogitsProcessor
 
+arrays = {
+    "list": [[1.0, 2.0], [3.0, 4.0]],
+    "np": np.array([[1, 2], [3, 4]], dtype=np.float32),
+    "jax": jnp.array([[1, 2], [3, 4]], dtype=jnp.float32),
+    "torch": torch.tensor([[1, 2], [3, 4]], dtype=torch.float32),
+}
+
 try:
     import mlx.core as mx
-    HAS_MLX = True
+
+    arrays["mlx"] = mx.array([[1, 2], [3, 4]], dtype=mx.float32)
+    arrays["mlx_bfloat16"] = mx.array([[1, 2], [3, 4]], dtype=mx.bfloat16)
 except ImportError:
-    HAS_MLX = False
+    pass
 
+try:
+    import jax.numpy as jnp
 
-libraries = ["numpy", "torch", "jax"]
-if HAS_MLX:
-    libraries.append("mlx")
+    arrays["jax"] = jnp.array([[1, 2], [3, 4]], dtype=jnp.float32)
+except ImportError:
+    pass
 
-# we check the accepted shapes:
-# - both 1D
-# - both 2D
-# - input_ids 1D and logits 2D with a single sequence
-# we raise an error if the shapes are not accepted:
-# - input_ids 2D and logits 1D
-# - input_ids 1D and logits 2D, but with multiple sequences
-# - both 3D
-arrays = {
-    "numpy": [
-        (np.array([1, 2], dtype=np.float32), np.array([1, 2], dtype=np.int32), None),
-        (np.array([[1, 2], [3, 4]], dtype=np.float32), np.array([[1, 2], [3, 4]], dtype=np.int32), None),
-        (np.array([1, 2], dtype=np.float32), np.array([[1, 2]], dtype=np.int32), None),
-        (np.array([[1, 2]], dtype=np.float32), np.array([1, 2], dtype=np.int32), AssertionError),
-        (np.array([1, 2], dtype=np.float32), np.array([[1, 2], [3, 4]], dtype=np.int32), AssertionError),
-        (np.array([[[1, 2]]], dtype=np.float32), np.array([[[1, 2]]], dtype=np.int32), ValueError),
-    ],
-    "torch": [
-        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([1, 2], dtype=torch.int32), None),
-        (torch.tensor([[1, 2], [3, 4]], dtype=torch.float32), torch.tensor([[1, 2], [3, 4]], dtype=torch.int32), None),
-        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([[1, 2]], dtype=torch.int32), None),
-        (torch.tensor([[1, 2]], dtype=torch.float32), torch.tensor([1, 2], dtype=torch.int32), AssertionError),
-        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([[1, 2], [3, 4]], dtype=torch.int32), AssertionError),
-        (torch.tensor([[[1, 2]]], dtype=torch.float32), torch.tensor([[[1, 2]]], dtype=torch.int32), ValueError),
-    ],
-    "jax": [
-        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), None),
-        (jnp.array([[1, 2], [3, 4]], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), None),
-        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2]], dtype=jnp.int32), None),
-        (jnp.array([[1, 2]], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), AssertionError),
-        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), AssertionError),
-        (jnp.array([[[1, 2]]], dtype=jnp.float32), jnp.array([[[1, 2]]], dtype=jnp.int32), ValueError),
-    ],
-}
-if HAS_MLX:
-    arrays["mlx"] = [
-        (mx.array([1, 2], dtype=mx.float32), mx.array([1, 2], dtype=mx.int32), None),
-        (mx.array([[1, 2], [3, 4]], dtype=mx.float32), mx.array([[1, 2], [3, 4]], dtype=mx.int32), None),
-        (mx.array([1, 2], dtype=mx.float32), mx.array([[1, 2]], dtype=mx.int32), None),
-        (mx.array([[1, 2]], dtype=mx.float32), mx.array([1, 2], dtype=mx.int32), AssertionError),
-        (mx.array([1, 2], dtype=mx.float32), mx.array([[1, 2], [3, 4]], dtype=mx.int32), AssertionError),
-        (mx.array([[[1, 2]]], dtype=mx.float32), mx.array([[[1, 2]]], dtype=mx.int32), ValueError),
-    ]
 
+# Mock implementation of the abstract class for testing
 class MockLogitsProcessor(OutlinesLogitsProcessor):
-    def process_logits(self, input_ids, logits):
-        # check that input_ids and logits received are 2D tensors
-        assert len(self.tensor_adapter.shape(input_ids)) == 2
-        assert len(self.tensor_adapter.shape(logits)) == 2
-        return logits
-
-
-@pytest.mark.parametrize("library", libraries)
-def test_base_logits_processor_init(library):
-    processor = MockLogitsProcessor(library)
-    assert processor.tensor_adapter is not None
-    with pytest.raises(NotImplementedError):
-        processor = MockLogitsProcessor("foo")
-
-
-@pytest.mark.parametrize("library", libraries)
-def test_base_logits_processor_call(library):
-    processor = MockLogitsProcessor(library)
-    input_values = arrays[library]
-    for input_value in input_values:
-        input_ids, logits, expected_error = input_value
-        if expected_error is not None:
-            with pytest.raises(expected_error):
-                processor(input_ids, logits)
-        else:
-            original_shape = processor.tensor_adapter.shape(logits)
-            processed_logits = processor(input_ids, logits)
-            # we check that the shape of logits is preserved
-            assert processor.tensor_adapter.shape(processed_logits) == original_shape
-
-
-@pytest.mark.parametrize("library", libraries)
-def test_base_logits_processor_init_library_name(library):
-    processor = MockLogitsProcessor(library)
-    assert processor.tensor_adapter is not None
-    with pytest.raises(NotImplementedError):
-        processor = MockLogitsProcessor("foo")
+    def process_logits(
+        self, input_ids: List[List[int]], logits: torch.Tensor
+    ) -> torch.Tensor:
+        # For testing purposes, let's just return logits multiplied by 2
+        return logits * 2
+
+
+@pytest.fixture
+def processor():
+    """Fixture for creating an instance of the MockLogitsProcessor."""
+    return MockLogitsProcessor()
+
+
+@pytest.mark.parametrize("array_type", arrays.keys())
+def test_to_torch(array_type, processor):
+    data = arrays[array_type]
+    torch_tensor = processor._to_torch(data)
+    assert isinstance(torch_tensor, torch.Tensor)
+    assert torch.allclose(
+        torch_tensor.cpu(), torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
+    )
+
+
+@pytest.mark.parametrize("array_type", arrays.keys())
+def test_from_torch(array_type, processor):
+    torch_tensor = torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
+    data = processor._from_torch(torch_tensor, type(arrays[array_type]))
+    assert isinstance(data, type(arrays[array_type]))
+    if array_type == "mlx_bfloat16":
+        # For bfloat16, we expect the output to be float32 due to the conversion
+        assert data.dtype == mx.float32
+        assert np.allclose(np.array(data), np.array([[1, 2], [3, 4]], dtype=np.float32))
+    else:
+        assert np.allclose(data, arrays[array_type])
+
+
+@pytest.mark.parametrize("array_type", arrays.keys())
+def test_call(array_type, processor):
+    input_ids = arrays[array_type]
+    logits = arrays[array_type]
+    processed_logits = processor(input_ids, logits)
+
+    assert isinstance(processed_logits, type(arrays[array_type]))
+    assert np.allclose(
+        np.array(processed_logits), np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
+    )
diff --git a/tests/processors/test_structured.py b/tests/processors/test_structured.py
deleted file mode 100644
index 54afae2b2..000000000
--- a/tests/processors/test_structured.py
+++ /dev/null
@@ -1,135 +0,0 @@
-import pytest
-from typing import Dict
-
-import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM
-
-import outlines
-from outlines.processors.guide import RegexGuide
-from outlines.processors.structured import (
-    GuideLogitsProcessor,
-    RegexLogitsProcessor,
-    JSONLogitsProcessor,
-    CFGLogitsProcessor,
-)
-
-
-arithmetic_grammar = """
-?start: sum
-
-?sum: product
-| sum "+" product   -> add
-| sum "-" product   -> sub
-
-?product: atom
-| product "*" atom  -> mul
-| product "/" atom  -> div
-
-?atom: NUMBER           -> number
-| "-" atom         -> neg
-| "(" sum ")"
-
-%import common.NUMBER
-%import common.WS_INLINE
-
-%ignore WS_INLINE
-"""
-
-@pytest.fixture
-def tokenizer():
-    model = outlines.from_transformers(
-        AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return model.tokenizer
-
-
-def test_structured_init_guide_logits_processor(tokenizer):
-    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
-    processor = GuideLogitsProcessor(
-        tokenizer=tokenizer,
-        guide=guide,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, GuideLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is guide
-    assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_init_guide_logits_processor_copy(tokenizer):
-    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
-    processor = GuideLogitsProcessor(
-        tokenizer=tokenizer,
-        guide=guide,
-        tensor_library_name="torch",
-    )
-    processor_copy = processor.copy()
-    assert processor_copy is not processor
-    assert processor_copy.tokenizer is tokenizer
-    assert processor_copy.guide is guide
-    assert processor_copy.tensor_adapter.library_name == "torch"
-
-
-def test_structured_guide_logits_processor_call(tokenizer):
-    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
-    processor = GuideLogitsProcessor(
-        tokenizer=tokenizer,
-        guide=guide,
-        tensor_library_name="torch",
-    )
-    vocab_token_ids = list(tokenizer.vocabulary.values())
-    logits = torch.randn(1, len(vocab_token_ids))
-    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
-    output = processor(input_ids, logits)
-    assert output.shape == (1, len(vocab_token_ids))
-
-
-def test_structured_init_json_logits_processor(tokenizer):
-    processor = JSONLogitsProcessor(
-        schema={"type": "object", "properties": {"name": {"type": "string"}}},
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, JSONLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is not None
-    assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_init_regex_logits_processor(tokenizer):
-    processor = RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, RegexLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is not None
-    assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_init_cfg_logits_processor(tokenizer):
-    processor = CFGLogitsProcessor(
-        cfg_str=arithmetic_grammar,
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, CFGLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is not None
-    assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_cfg_logits_processor_call(tokenizer):
-    processor = CFGLogitsProcessor(
-        cfg_str=arithmetic_grammar,
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    vocab_token_ids = list(tokenizer.vocabulary.values())
-    logits = torch.randn(1, len(vocab_token_ids))
-    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
-    output = processor(input_ids, logits)
-    assert output.shape == (1, len(vocab_token_ids))
-    processor(input_ids, logits)
diff --git a/tests/processors/test_tensor_adapters.py b/tests/processors/test_tensor_adapters.py
deleted file mode 100644
index 4a5bcc1e1..000000000
--- a/tests/processors/test_tensor_adapters.py
+++ /dev/null
@@ -1,276 +0,0 @@
-import pytest
-from pytest import mark
-
-import jax
-import jax.numpy as jnp
-import numpy as np
-import tensorflow as tf
-import torch
-
-from outlines.processors.tensor_adapters import (
-    NumpyTensorAdapter,
-    TorchTensorAdapter,
-    MLXTensorAdapter,
-    JAXTensorAdapter,
-    TensorFlowTensorAdapter,
-)
-
-try:
-    import mlx_lm
-    import mlx.core as mx
-
-    HAS_MLX = mx.metal.is_available()
-except ImportError:
-    HAS_MLX = False
-
-
-adapters = {
-    "jax": JAXTensorAdapter(),
-    "numpy": NumpyTensorAdapter(),
-    "tensorflow": TensorFlowTensorAdapter(),
-    "torch": TorchTensorAdapter(),
-}
-if HAS_MLX:
-    adapters["mlx"] = MLXTensorAdapter()
-
-frameworks = ["jax", "numpy", "tensorflow", "torch", "mlx"]
-
-def create_tensor(framework, shape, dtype=None):
-    if framework == "torch":
-        return torch.randn(*shape)
-    elif framework == "numpy":
-        return np.random.randn(*shape)
-    elif framework == "jax":
-        key = jax.random.PRNGKey(0)
-        return jax.random.normal(key, shape=shape)
-    elif framework == "tensorflow":
-        return tf.random.normal(shape)
-    elif framework == "mlx":
-        if not HAS_MLX:
-            pytest.skip("MLX not available")
-        return mx.random.normal(shape)
-
-def compare_tensors(framework, tensor1, tensor2):
-    if framework == "torch":
-        return torch.allclose(tensor1, tensor2)
-    elif framework == "numpy":
-        return np.array_equal(tensor1, tensor2)
-    elif framework == "jax":
-        return jax.numpy.array_equal(tensor1, tensor2)
-    elif framework == "tensorflow":
-        return tf.reduce_all(tf.equal(tensor1, tensor2))
-    elif framework == "mlx":
-        if not HAS_MLX:
-            pytest.skip("MLX not available")
-        return mx.array_equal(tensor1, tensor2)
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_shape(framework):
-    # 1d tensor
-    tensor_1d = create_tensor(framework, (2,))
-    result_1d = adapters[framework].shape(tensor_1d)
-    assert len(result_1d) == 1
-    assert result_1d[0] == 2
-
-    # 2d tensor
-    tensor_2d = create_tensor(framework, (2, 3))
-    result_2d = adapters[framework].shape(tensor_2d)
-    assert len(result_2d) == 2
-    assert result_2d[0] == 2
-    assert result_2d[1] == 3
-
-    # 3d tensor
-    tensor_3d = create_tensor(framework, (2, 2, 3))
-    result_3d = adapters[framework].shape(tensor_3d)
-    assert len(result_3d) == 3
-    assert result_3d[0] == 2
-    assert result_3d[1] == 2
-    assert result_3d[2] == 3
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_unsqueeze(framework):
-    # 1d tensor
-    tensor_1d = create_tensor(framework, (2,))
-    result_1d = adapters[framework].unsqueeze(tensor_1d)
-    assert result_1d.shape == (1, 2)
-
-    # 2d tensor
-    tensor_2d = create_tensor(framework, (2, 3))
-    result_2d = adapters[framework].unsqueeze(tensor_2d)
-    assert result_2d.shape == (1, 2, 3)
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_squeeze(framework):
-    # 1d tensor
-    tensor_1d = create_tensor(framework, (1,))
-    result_1d = adapters[framework].squeeze(tensor_1d)
-    with pytest.raises(TypeError):
-        len(result_1d)
-
-    # 2d tensor
-    tensor_2d = create_tensor(framework, (1, 2))
-    result_2d = adapters[framework].squeeze(tensor_2d)
-    assert result_2d.shape == (2,)
-
-    # 3d tensor
-    tensor_3d = create_tensor(framework, (1, 2, 3))
-    result_3d = adapters[framework].squeeze(tensor_3d)
-    assert result_3d.shape == (2, 3)
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_to_list(framework):
-    # 1d tensor
-    tensor_1d = create_tensor(framework, (2,))
-    result_1d = adapters[framework].to_list(tensor_1d)
-    assert isinstance(result_1d, list)
-    assert len(result_1d) == 2
-
-    # 2d tensor
-    tensor_2d = create_tensor(framework, (2, 3))
-    result_2d = adapters[framework].to_list(tensor_2d)
-    assert isinstance(result_2d, list)
-    assert len(result_2d) == 2
-    assert len(result_2d[0]) == 3
-    assert len(result_2d[1]) == 3
-
-    # 3d tensor
-    tensor_3d = create_tensor(framework, (2, 2, 3))
-    result_3d = adapters[framework].to_list(tensor_3d)
-    assert isinstance(result_3d, list)
-    assert len(result_3d) == 2
-    assert len(result_3d[0]) == 2
-    assert len(result_3d[1]) == 2
-    assert len(result_3d[0][0]) == 3
-    assert len(result_3d[0][1]) == 3
-    assert len(result_3d[1][0]) == 3
-    assert len(result_3d[1][1]) == 3
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_to_scalar(framework):
-    # multi-elements tensor, should raise an error
-    tensor_multi = create_tensor(framework, (2, 3))
-    if framework == "torch":
-        with pytest.raises(RuntimeError):
-            adapters[framework].to_scalar(tensor_multi)
-    else:
-        with pytest.raises(ValueError):
-            adapters[framework].to_scalar(tensor_multi)
-
-    # single-element tensor
-    tensor_single = create_tensor(framework, (1, 1))
-    scalar = adapters[framework].to_scalar(tensor_single)
-    assert isinstance(scalar, float)
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_full_like(framework):
-    tensor = create_tensor(framework, (2, 3))
-    result = adapters[framework].full_like(tensor, 0)
-    assert result.shape == (2, 3)
-    for i in range(2):
-        for j in range(3):
-            assert result[i, j] == 0
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_concatenate(framework):
-    # 1d tensors
-    tensor1 = create_tensor(framework, (2,))
-    tensor2 = create_tensor(framework, (2,))
-    result = adapters[framework].concatenate([tensor1, tensor2])
-    assert result.shape == (4,)
-    assert result[0] == tensor1[0]
-    assert result[1] == tensor1[1]
-    assert result[2] == tensor2[0]
-    assert result[3] == tensor2[1]
-
-    # 2d tensors
-    tensor1 = create_tensor(framework, (2, 3))
-    tensor2 = create_tensor(framework, (2, 3))
-    result = adapters[framework].concatenate([tensor1, tensor2])
-    assert result.shape == (4, 3)
-    for i in range(2):
-        for j in range(3):
-            assert result[i, j] == tensor1[i, j]
-            assert result[i + 2, j] == tensor2[i, j]
-
-    # 3d tensors
-    tensor1 = create_tensor(framework, (2, 2, 3))
-    tensor2 = create_tensor(framework, (2, 2, 3))
-    result = adapters[framework].concatenate([tensor1, tensor2])
-    assert result.shape == (4, 2, 3)
-    for i in range(2):
-        for j in range(2):
-            for k in range(3):
-                assert result[i, j, k] == tensor1[i, j, k]
-                assert result[i + 2, j, k] == tensor2[i, j, k]
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_get_to_device(framework):
-    tensor = create_tensor(framework, (2, 3))
-    device = adapters[framework].get_device(tensor)
-    device_tensor = adapters[framework].to_device(tensor, device)
-
-    if framework == "torch":
-        assert isinstance(device_tensor.device.type, str)
-        assert compare_tensors(framework, device_tensor, tensor)
-    else:
-        assert compare_tensors(framework, device_tensor, tensor)
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_boolean_ones_like(framework):
-    tensor = create_tensor(framework, (2, 3))
-    ones = adapters[framework].boolean_ones_like(tensor)
-
-    assert ones.shape == (2, 3)
-    for i in range(2):
-        for j in range(3):
-            assert ones[i, j]
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_apply_mask(framework):
-    tensor = create_tensor(framework, (2, 3))
-
-    if framework == "torch":
-        mask = torch.randn(2, 3) > 0
-    elif framework == "numpy":
-        mask = np.random.randn(2, 3) > 0
-    elif framework == "jax":
-        key = jax.random.PRNGKey(0)
-        mask = jax.random.normal(key, shape=(2, 3)) > 0
-    elif framework == "tensorflow":
-        mask = tf.random.normal((2, 3)) > 0
-    elif framework == "mlx":
-        if not HAS_MLX:
-            pytest.skip("MLX not available")
-        mask = mx.random.normal((2, 3)) > 0
-
-    masked = adapters[framework].apply_mask(tensor, mask, float("-inf"))
-
-    assert masked.shape == (2, 3)
-    for i in range(2):
-        for j in range(3):
-            if mask[i, j]:
-                assert masked[i, j] == float("-inf")
-            else:
-                assert masked[i, j] == tensor[i, j]
-
-
-@pytest.mark.parametrize("framework", frameworks)
-def test_tensor_adapter_argsort_descending(framework):
-    tensor = create_tensor(framework, (2, 3))
-    indices = adapters[framework].argsort_descending(tensor)
-
-    assert indices.shape == (2, 3)
-    for i in range(2):
-        sorted_values = [tensor[i][idx] for idx in indices[i]]
-        for j in range(len(sorted_values) - 1):
-            assert sorted_values[j] >= sorted_values[j + 1]
diff --git a/tests/test_applications.py b/tests/test_applications.py
deleted file mode 100644
index 5fed7bf23..000000000
--- a/tests/test_applications.py
+++ /dev/null
@@ -1,93 +0,0 @@
-from typing import Any
-
-import jinja2
-import pytest
-import transformers
-
-from outlines import from_transformers
-from outlines.applications import Application
-from outlines.templates import Template
-
-
-@pytest.fixture(scope="session")
-def model():
-    return from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("gpt2"),
-        transformers.AutoTokenizer.from_pretrained("gpt2"),
-    )
-
-
-@pytest.fixture(scope="session")
-def another_model():
-    return from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("gpt2"),
-        transformers.AutoTokenizer.from_pretrained("gpt2"),
-    )
-
-
-def test_application_initialization():
-    template = Template.from_string("Test {{ value }}")
-    output_type = None
-    application = Application(template, output_type)
-
-    assert application.template == template
-    assert application.output_type == output_type
-    assert application.model is None
-    assert application.generator is None
-
-
-def test_application_generator_no_model():
-    template = Template.from_string("Test {{ value }}")
-    output_type = None
-    application = Application(template, output_type)
-
-    with pytest.raises(ValueError):
-        application(None, {"value": "example"})
-
-
-def test_application_template_call(model):
-    template = Template.from_string("Test {{ value }}")
-    output_type = None
-    application = Application(template, output_type)
-    result = application(model, {"value": "example"}, max_new_tokens=10)
-
-    assert isinstance(result, str)
-
-
-def test_application_callable_call(model):
-    def template(value):
-        return f"Test {value}"
-
-    output_type = None
-    application = Application(template, output_type)
-    result = application(model, {"value": "example"}, max_new_tokens=10)
-
-    assert isinstance(result, str)
-
-
-def test_application_template_error(model):
-    template = Template.from_string("Test {{ value }}")
-    output_type = None
-    application = Application(template, output_type)
-
-    with pytest.raises(jinja2.exceptions.UndefinedError):
-        application(model, {"foo": "bar"})
-
-
-def test_application_generator_reuse(model, another_model):
-    template = Template.from_string("Test {{ value }}")
-    output_type = None
-    application = Application(template, output_type)
-
-    application(model, {"value": "example"}, max_new_tokens=10)
-    first_generator = application.generator
-    first_model = application.model
-
-    application(model, {"value": "example"}, max_new_tokens=10)
-    assert application.model == first_model
-    assert application.generator == first_generator
-
-    application(another_model, {"value": "example"}, max_new_tokens=10)
-    assert application.model == another_model
-    assert application.model != first_model
-    assert application.generator != first_generator
diff --git a/tests/v0_legacy/test_base_legacy.py b/tests/test_base.py
similarity index 99%
rename from tests/v0_legacy/test_base_legacy.py
rename to tests/test_base.py
index f445b4f5d..24c84f7d9 100644
--- a/tests/v0_legacy/test_base_legacy.py
+++ b/tests/test_base.py
@@ -2,7 +2,7 @@
 import pytest
 from numpy.testing import assert_array_equal
 
-from outlines.v0_legacy.base import vectorize
+from outlines.base import vectorize
 
 
 def test_vectorize_docstring():
diff --git a/tests/test_cache.py b/tests/test_cache.py
index 149d856c0..eb4ec406e 100644
--- a/tests/test_cache.py
+++ b/tests/test_cache.py
@@ -1,19 +1,9 @@
 import os
 import tempfile
 import unittest
-from importlib import reload
 
 import diskcache
 import pytest
-from diskcache import Cache, UNKNOWN
-from outlines.caching import CloudpickleDisk
-
-
-@pytest.fixture
-def temp_dir():
-    """Create a temporary directory for testing."""
-    directory = tempfile.mkdtemp()
-    yield directory
 
 
 @pytest.fixture
@@ -198,24 +188,3 @@ def fn():
     # scope has exited, cache is enabled again
     fn()
     assert mock.call_count == 2
-
-
-@pytest.fixture
-def temp_cache_dir():
-    import os
-    import tempfile
-
-    import outlines.caching
-    import outlines.processors.guide
-
-    with tempfile.TemporaryDirectory() as tempdir:
-        os.environ["OUTLINES_CACHE_DIR"] = tempdir
-        outlines.caching.get_cache.cache_clear()
-        reload(outlines)
-        reload(outlines.processors.guide)
-        cache_status = outlines.caching._caching_enabled
-        try:
-            outlines.caching._caching_enabled = True
-            yield
-        finally:
-            outlines.caching._caching_enabled = cache_status
diff --git a/tests/v0_legacy/test_function_legacy.py b/tests/test_function.py
similarity index 59%
rename from tests/v0_legacy/test_function_legacy.py
rename to tests/test_function.py
index 1fd03ebb9..8bb12976e 100644
--- a/tests/v0_legacy/test_function_legacy.py
+++ b/tests/test_function.py
@@ -4,76 +4,25 @@
 from requests.exceptions import HTTPError
 
 import outlines
-from outlines.v0_legacy.function import (
-    Function,
-    download_from_github,
-    extract_function_from_file,
-)
+from outlines.function import Function, download_from_github, extract_function_from_file
 
 
 def test_function_basic():
-    with pytest.warns(DeprecationWarning, match="The @prompt decorator"):
-        with pytest.warns(
-            DeprecationWarning, match="The `Function` class is deprecated"
-        ):
-            @outlines.prompt
-            def test_template(text: str):
-                """{{ text }}"""
-
-            class Foo(BaseModel):
-                id: int
+    with pytest.deprecated_call(match="The @prompt decorator"):
 
-            fn = Function(
-                test_template,
-                Foo,
-                "erwanf/gpt2-mini",
-            )
+        @outlines.prompt
+        def test_template(text: str):
+            """{{ text }}"""
 
-            assert fn.generator is None
+    class Foo(BaseModel):
+        id: int
 
-            result = fn("test")
-            assert isinstance(result, BaseModel)
-            assert fn.generator is not None
-            fn("test")
+    fn = Function(test_template, Foo, "hf-internal-testing/tiny-random-GPTJForCausalLM")
 
+    assert fn.generator is None
 
-@responses.activate
-def test_function_from_github():
-    content = """
-import outlines
-from pydantic import BaseModel
-
-model = "gpt2"
-
-
-prompt = outlines.Template.from_string("{{ text }}")
-
-
-class User(BaseModel):
-    id: int
-    name: str
-
-
-fn = outlines.Function(
-    prompt,
-    User,
-    "gpt2",
-)
-"""
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
-        body=content,
-        status=200,
-    )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `Function` class is deprecated",
-    ):
-        fn = Function.from_github("dottxt-ai/outlines/program", "fn")
-        assert (
-            str(type(fn)) == "<class 'outlines.v0_legacy.function.Function'>"
-        )
+    result = fn("test")
+    assert isinstance(result, BaseModel)
 
 
 def test_download_from_github_invalid():
@@ -154,10 +103,8 @@ class User(BaseModel):
     "gpt2",
 )
     """
-    with pytest.raises(
-        DeprecationWarning,
-        match="The @prompt decorator is deprecated",
-    ):
+
+    with pytest.deprecated_call(match="The @prompt decorator"):
         fn = extract_function_from_file(content, "function")
         assert (
             str(type(fn)) == "<class 'outlines.function.Function'>"
@@ -184,9 +131,7 @@ class User(BaseModel):
     "gpt2",
 )
     """
-    with pytest.raises(
-        DeprecationWarning,
-        match="The @prompt decorator is deprecated",
-    ):
+
+    with pytest.deprecated_call(match="The @prompt decorator"):
         with pytest.raises(AttributeError, match="Could not find"):
             extract_function_from_file(content, "function")
diff --git a/tests/test_generator.py b/tests/test_generator.py
deleted file mode 100644
index 5ce54dc4b..000000000
--- a/tests/test_generator.py
+++ /dev/null
@@ -1,255 +0,0 @@
-import pytest
-from typing import AsyncGenerator, Generator as TypingGenerator, Literal
-
-import interegular
-import transformers
-
-import outlines
-from outlines.generator import (
-    BlackBoxGenerator,
-    SteerableGenerator,
-    Generator,
-    AsyncBlackBoxGenerator,
-)
-from outlines.models import AsyncVLLM, VLLM
-from outlines.processors import (
-    OutlinesLogitsProcessor,
-    RegexLogitsProcessor,
-)
-from outlines.types import CFG, FSM
-from tests.test_utils.mock_openai_client import (
-    MockAsyncOpenAIClient,
-    MockOpenAIClient,
-)
-
-
-MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
-
-
-# We used the mocked vllm model to test the black box generator
-async_openai_client = MockAsyncOpenAIClient()
-openai_client = MockOpenAIClient()
-mock_responses = [
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Write a very short sentence'}
-            ],
-            'model': MODEL_NAME,
-            'max_tokens': 10,
-            'extra_body': {'guided_regex': '("[^"]*")'},
-        },
-        "Mock response"
-    ),
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': 'Write a very short sentence'}
-            ],
-            'model': MODEL_NAME,
-            'max_tokens': 10,
-            'extra_body': {'guided_regex': '("[^"]*")'},
-            'stream': True,
-        },
-        ["Mock", "response"]
-    ),
-]
-async_openai_client.add_mock_responses(mock_responses)
-openai_client.add_mock_responses(mock_responses)
-
-
-@pytest.fixture(scope="session")
-def steerable_model():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return model
-
-
-@pytest.fixture(scope="session")
-def sample_processor():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    processor = RegexLogitsProcessor(
-        regex_string="[0-9]{3}",
-        tokenizer=model.tokenizer,
-        tensor_library_name=model.tensor_library_name,
-    )
-    return processor
-
-
-@pytest.fixture(scope="module")
-def black_box_sync_model():
-    return VLLM(openai_client, MODEL_NAME)
-
-
-@pytest.fixture(scope="module")
-def black_box_async_model():
-    return AsyncVLLM(async_openai_client, MODEL_NAME)
-
-
-# SteerableGenerator
-
-
-def test_steerable_generator_init_valid_processor(steerable_model, sample_processor):
-    generator = SteerableGenerator.from_processor(steerable_model, sample_processor)
-    assert generator.logits_processor == sample_processor
-    assert generator.model == steerable_model
-
-
-def test_steerable_generator_init_invalid_processor(steerable_model):
-    with pytest.raises(TypeError):
-        SteerableGenerator.from_processor(steerable_model, Literal["foo", "bar"])
-
-
-def test_steerable_generator_init_cfg_output_type(steerable_model):
-    generator = SteerableGenerator(steerable_model, CFG('start: "a"'))
-    assert generator.model == steerable_model
-    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
-
-
-def test_steerable_generator_init_fsm_output_type(steerable_model):
-    generator = SteerableGenerator(
-        steerable_model,
-        FSM(interegular.parse_pattern(r"abc").to_fsm())
-    )
-    assert generator.model == steerable_model
-    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
-
-
-def test_steerable_generator_init_other_output_type(steerable_model):
-    generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
-    assert generator.model == steerable_model
-    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
-
-
-def test_steerable_generator_init_invalid_output_type(steerable_model, sample_processor):
-    with pytest.raises(ValueError):
-        SteerableGenerator(steerable_model, sample_processor)
-
-
-def test_steerable_generator_call(steerable_model):
-    generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
-    result = generator("foo", max_new_tokens=10)
-    assert isinstance(result, str)
-
-
-def test_steerable_generator_stream(steerable_model):
-    with pytest.raises(NotImplementedError):
-        generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
-        result = generator.stream("foo", max_tokens=10)
-        assert isinstance(result, TypingGenerator)
-        assert isinstance(next(result), str)
-
-
-# BlackBoxGenerator
-
-
-def test_black_box_generator_init_valid(black_box_sync_model):
-    generator = BlackBoxGenerator(black_box_sync_model, Literal["foo", "bar"])
-    assert generator.model == black_box_sync_model
-    assert generator.output_type == Literal["foo", "bar"]
-
-
-def test_black_box_generator_init_invalid(black_box_sync_model):
-    with pytest.raises(NotImplementedError):
-        BlackBoxGenerator(black_box_sync_model, FSM("foo"))
-
-
-def test_black_box_generator_call(black_box_sync_model):
-    generator = BlackBoxGenerator(black_box_sync_model, str)
-    result = generator("Write a very short sentence", max_tokens=10)
-    assert isinstance(result, str)
-
-
-def test_black_box_generator_stream(black_box_sync_model):
-    generator = BlackBoxGenerator(black_box_sync_model, str)
-    result = generator.stream("Write a very short sentence", max_tokens=10)
-    assert isinstance(result, TypingGenerator)
-    assert isinstance(next(result), str)
-
-
-
-# AsyncBlackBoxGenerator
-
-
-def test_async_black_box_generator_init_valid(black_box_async_model):
-    generator = AsyncBlackBoxGenerator(black_box_async_model, Literal["foo", "bar"])
-    assert generator.model == black_box_async_model
-    assert generator.output_type == Literal["foo", "bar"]
-
-
-def test_async_black_box_generator_init_invalid(black_box_async_model):
-    with pytest.raises(NotImplementedError):
-        AsyncBlackBoxGenerator(black_box_async_model, FSM("foo"))
-
-
-@pytest.mark.asyncio
-async def test_async_black_box_generator_call(black_box_async_model):
-    generator = AsyncBlackBoxGenerator(black_box_async_model, str)
-    result = await generator("Write a very short sentence", max_tokens=10)
-    assert isinstance(result, str)
-
-
-@pytest.mark.asyncio
-async def test_async_black_box_generator_stream(black_box_async_model):
-    generator = AsyncBlackBoxGenerator(black_box_async_model, str)
-    result = generator.stream("Write a very short sentence", max_tokens=10)
-    assert isinstance(result, AsyncGenerator)
-    async for chunk in result:
-        assert isinstance(chunk, str)
-        break  # Just check the first chunk
-
-
-# Generator
-
-
-def test_generator_init_no_model():
-    with pytest.raises(ValueError):
-        Generator(None, Literal["foo", "bar"])
-
-
-def test_generator_init_multiple_output_type(steerable_model, sample_processor):
-    with pytest.raises(ValueError):
-        Generator(steerable_model, Literal["foo", "bar"], processor=sample_processor)
-
-
-def test_generator_steerable_output_type(steerable_model):
-    generator = Generator(steerable_model, Literal["foo", "bar"])
-    assert isinstance(generator, SteerableGenerator)
-    assert generator.model == steerable_model
-    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
-
-
-def test_generator_steerable_processor(steerable_model, sample_processor):
-    generator = Generator(steerable_model, processor=sample_processor)
-    assert isinstance(generator, SteerableGenerator)
-    assert generator.model == steerable_model
-    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
-
-
-def test_generator_black_box_sync_output_type(black_box_sync_model):
-    generator = Generator(black_box_sync_model, Literal["foo", "bar"])
-    assert isinstance(generator, BlackBoxGenerator)
-    assert generator.model == black_box_sync_model
-    assert generator.output_type == Literal["foo", "bar"]
-
-
-def test_generator_black_box_sync_processor(black_box_sync_model, sample_processor):
-    with pytest.raises(NotImplementedError):
-        Generator(black_box_sync_model, processor=sample_processor)
-
-
-def test_generator_black_box_async_output_type(black_box_async_model):
-    generator = Generator(black_box_async_model, Literal["foo", "bar"])
-    assert isinstance(generator, AsyncBlackBoxGenerator)
-    assert generator.model == black_box_async_model
-    assert generator.output_type == Literal["foo", "bar"]
-
-
-def test_generator_black_box_async_processor(black_box_async_model, sample_processor):
-    with pytest.raises(NotImplementedError):
-        Generator(black_box_async_model, processor=sample_processor)
diff --git a/tests/test_grammars.py b/tests/test_grammars.py
index 69486684b..3d704f094 100644
--- a/tests/test_grammars.py
+++ b/tests/test_grammars.py
@@ -1,7 +1,7 @@
 import pytest
 
 import outlines.grammars as grammars
-from outlines.processors.guide import CFGGuide
+from outlines.fsm.guide import CFGGuide
 
 
 @pytest.mark.parametrize("grammar", [grammars.json, grammars.arithmetic])
diff --git a/tests/test_samplers.py b/tests/test_samplers.py
new file mode 100644
index 000000000..10a7be26f
--- /dev/null
+++ b/tests/test_samplers.py
@@ -0,0 +1,277 @@
+import math
+
+import pytest
+import torch
+
+from outlines.samplers import (
+    BeamSearchSampler,
+    GreedySampler,
+    MultinomialSampler,
+    beam_search,
+    greedy,
+    keep_top_k_logits,
+    keep_top_p_logits,
+    multinomial,
+    rescale_logits,
+)
+
+
+def compute_logprobs(logits):
+    return torch.nn.functional.log_softmax(logits, dim=-1)
+
+
+def test_aliases():
+    assert greedy == GreedySampler
+    assert multinomial == MultinomialSampler
+    assert beam_search == BeamSearchSampler
+
+
+def test_greedy():
+    sampler = GreedySampler()
+    logits = torch.tensor([[1.0, 2.0, 5.0]])
+    weights = torch.tensor([0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[2]]))
+    assert ancestors.equal(torch.tensor([0]))
+    assert weights.equal(logprobs[..., 2])
+
+    sampler = GreedySampler()
+    logits = torch.tensor([[10.0, 0.0, 3.0], [-math.inf, 2.0, 5.0]])
+    weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [2]]))
+    assert ancestors.equal(torch.tensor([0, 1]))
+    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
+
+    params = sampler.sampling_params
+    assert params.sampler == "greedy"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 0.0
+
+
+def test_multinomial():
+    rng = torch.Generator()
+    rng.manual_seed(239)
+
+    sampler = MultinomialSampler()
+    logits = torch.tensor([[1.0, 4.0, 5.0]])
+    weights = torch.tensor([0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[2]]))
+    assert ancestors.equal(torch.tensor([0]))
+    assert weights.equal(torch.tensor([logprobs[:, 2]]))
+
+    sampler = MultinomialSampler()
+    logits = torch.tensor([[10.0, 0.0, 9.0], [-math.inf, 4.0, 5.0]])
+    weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [2]]))
+    assert ancestors.equal(torch.tensor([0, 1]))
+    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
+
+    sampler = MultinomialSampler(samples=5, top_k=10, top_p=0.9, temperature=0.8)
+    params = sampler.sampling_params
+    assert params.sampler == "multinomial"
+    assert params.num_samples == 5
+    assert params.top_p == 0.9
+    assert params.top_k == 10
+    assert params.temperature == 0.8
+
+
+def test_multinomial_init():
+    sampler = MultinomialSampler()
+    assert sampler.logits_processors == []
+
+    sampler = MultinomialSampler(3)
+    assert sampler.logits_processors == []
+
+    sampler = MultinomialSampler(top_k=1)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(top_p=0.95)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(top_k=1, top_p=0.95)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(temperature=1.0)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(top_k=1, temperature=1.0)
+    assert len(sampler.logits_processors) == 2
+
+    sampler = MultinomialSampler(top_p=0.95, temperature=1.0)
+    assert len(sampler.logits_processors) == 2
+
+
+def test_top_k():
+    with pytest.raises(ValueError, match="`k` must be a strictly"):
+        keep_top_k_logits(-1)
+
+    with pytest.raises(ValueError, match="`k` must be a strictly"):
+        keep_top_k_logits(0.1)
+
+    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0]])
+
+    logits_processor = keep_top_k_logits(1)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
+
+    logits_processor = keep_top_k_logits(10)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[1.0, 2.0, 3.0, 4.0]]))
+
+    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
+    logits_processor = keep_top_k_logits(2)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [[-math.inf, -math.inf, 3.0, 4.0], [-math.inf, -math.inf, 7.0, 8.0]]
+        )
+    )
+
+
+def test_top_p():
+    with pytest.raises(ValueError, match="`p` must be a floating point"):
+        keep_top_p_logits(-0.1)
+
+    with pytest.raises(ValueError, match="`p` must be a floating point"):
+        keep_top_p_logits(0.0)
+
+    with pytest.raises(ValueError, match="`p` must be a floating point"):
+        keep_top_p_logits(1.1)
+
+    logits = torch.tensor([[1.0, 1.01, 1.02, 4.0]])
+
+    logits_processor = keep_top_p_logits(0.1)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
+
+    logits_processor = keep_top_p_logits(0.95)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[-math.inf, 1.01, 1.02, 4.0]]))
+
+    logits_processor = keep_top_p_logits(1.0)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[1.0, 1.01, 1.02, 4.0]]))
+
+    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
+
+    logits_processor = keep_top_p_logits(0.1)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [
+                [-math.inf, -math.inf, -math.inf, 4.0],
+                [-math.inf, -math.inf, -math.inf, 8.0],
+            ]
+        )
+    )
+
+    logits_processor = keep_top_p_logits(0.95)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [
+                [-math.inf, 2.0, 3.0, 4.0],
+                [-math.inf, 6.0, 7.0, 8.0],
+            ]
+        )
+    )
+
+    logits_processor = keep_top_p_logits(1.0)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [
+                [1.0, 2.0, 3.0, 4.0],
+                [5.0, 6.0, 7.0, 8.0],
+            ]
+        )
+    )
+
+
+def test_rescale():
+    with pytest.raises(ValueError, match="`temperature` must"):
+        rescale_logits(1)
+
+    with pytest.raises(ValueError, match="`temperature` must"):
+        rescale_logits(-0.1)
+
+    with pytest.raises(ValueError, match="Please use the greedy sampler"):
+        rescale_logits(0.0)
+
+
+def test_beam_search():
+    # Two beams, single sequence
+    sampler = BeamSearchSampler(2)
+    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
+    init_weights = torch.tensor([0, 1.0])
+    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [1]]))
+    assert ancestors.equal(torch.tensor([1, 0]))
+    assert weights.equal(
+        torch.tensor([init_weights[1] + logprobs[1][0], logprobs[0][1]])
+    )
+
+    # Make sure that initial samples are different
+    sampler = BeamSearchSampler(2)
+    logits = torch.tensor([[0.0, 1.0], [0.0, 1.0]])
+    init_weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[1], [0]]))
+    assert ancestors.equal(torch.tensor([0, 0]))
+    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[0][0]]))
+
+    # One beam, batch of two sequences. Reduces to Greedy Search.
+    sampler = BeamSearchSampler(1)
+    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
+    weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[1], [0]]))
+    assert ancestors.equal(torch.tensor([0, 1]))
+    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[1][0]]))
+
+    # Two beams, batch of two sequences
+    sampler = BeamSearchSampler(2)
+    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0], [3.0, 2.0], [0.0, 1.0]])
+    init_weights = torch.tensor([0, 0, 2.0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [1], [0], [1]]))
+    assert ancestors.equal(torch.tensor([1, 0, 2, 2]))
+    assert weights.equal(
+        torch.tensor(
+            [
+                logprobs[1][0],
+                logprobs[0][1],
+                init_weights[2] + logprobs[2][0],
+                init_weights[2] + logprobs[2][1],
+            ]
+        )
+    )
+
+    sampler = BeamSearchSampler(beams=3)
+    params = sampler.sampling_params
+    assert params.sampler == "beam_search"
+    assert params.num_samples == 3
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 1.0
diff --git a/tests/test_templates.py b/tests/test_templates.py
index cf9d0ed5f..2fda3caa3 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -1,75 +1,12 @@
-import base64
-import json
 import os
 import tempfile
 from typing import Dict, List
 
 import pytest
-from PIL import Image
-from io import BytesIO
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.templates import (
-    Template,
-    build_template_from_string,
-    Vision,
-    get_fn_name,
-    get_fn_args,
-    get_fn_description,
-    get_fn_source,
-    get_fn_signature,
-    get_schema,
-)
-
-
-def sample_function(x, y=2):
-    """This is a sample function."""
-    return x + y
-
-def function_with_annotations(x: int, y: str) -> str:
-    """Function with annotations."""
-    return f"{x} {y}"
-
-def function_with_no_docstring(x, y):
-    return x * y
-
-class CallableClass:
-    def __call__(self):
-        pass
-
-class PydanticClass(BaseModel):
-    foo: str
-
-
-def test_vision_initialization():
-    # Create a simple image for testing
-    image = Image.new("RGB", (10, 10), color="red")
-    image.format = "PNG"
-
-    # Initialize the Vision object
-    vision = Vision(prompt="Test prompt", image=image)
-
-    # Check that the prompt is set correctly
-    assert vision.prompt == "Test prompt"
-
-    # Check that the image is encoded correctly
-    buffer = BytesIO()
-    image.save(buffer, format=image.format)
-    expected_image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
-    assert vision.image_str == expected_image_str
-
-    # Check that the image format is set correctly
-    assert vision.image_format == "image/png"
-
-
-def test_vision_invalid_image_format():
-    # Create an image without a format
-    image = Image.new("RGB", (10, 10), color="blue")
-
-    # Expect a TypeError when the image format is not set
-    with pytest.raises(TypeError, match="Could not read the format"):
-        Vision(prompt="Test prompt", image=image)
+from outlines.templates import Template, build_template_from_string
 
 
 def render(content: str, **kwargs):
@@ -504,60 +441,3 @@ def test_template_from_str_with_extra_linebreaks():
     """
     template = build_template_from_string(content)
     assert template.render(name="World") == "Hello, World!\n"
-
-
-def test_get_fn_name():
-    with pytest.raises(TypeError):
-        get_fn_name(1)
-    assert get_fn_name(sample_function) == "sample_function"
-    assert get_fn_name(function_with_annotations) == "function_with_annotations"
-    no_name_func = lambda x: x
-    assert get_fn_name(no_name_func) == "<lambda>"
-    assert get_fn_name(CallableClass()) == "CallableClass"
-
-
-def test_get_fn_args():
-    with pytest.raises(TypeError):
-        get_fn_args(1)
-    assert get_fn_args(sample_function) == "x, y=2"
-    assert get_fn_args(function_with_annotations) == "x: int, y: str"
-
-
-def test_get_fn_description():
-    with pytest.raises(TypeError):
-        get_fn_description(1)
-    assert get_fn_description(sample_function) == "This is a sample function."
-    assert get_fn_description(function_with_annotations) == "Function with annotations."
-    assert get_fn_description(function_with_no_docstring) == ""
-
-
-def test_get_fn_source():
-    with pytest.raises(TypeError, match="The `source` filter only applies to callables."):
-        get_fn_source(1)
-    source = (
-        'def sample_function(x, y=2):\n'
-        '    """This is a sample function."""\n'
-        '    return x + y'
-    )
-    assert get_fn_source(sample_function).strip() == source
-
-
-def test_get_fn_signature():
-    with pytest.raises(TypeError, match="The `source` filter only applies to callables."):
-        get_fn_signature(1)
-    sample_function_signature = "x, y=2"
-    assert get_fn_signature(sample_function) == sample_function_signature
-    function_with_annotations_signature = "x: int, y: str"
-    assert get_fn_signature(function_with_annotations) == function_with_annotations_signature
-
-
-def test_get_schema():
-    with pytest.raises(NotImplementedError):
-        get_schema(1)
-
-    dict_schema = {"foo": "bar"}
-    dict_schema_output = get_schema(dict_schema)
-    assert dict_schema_output == '{\n  "foo": "bar"\n}'
-
-    pydantic_schema_output = get_schema(PydanticClass)
-    assert pydantic_schema_output == '{\n  "foo": "<foo>"\n}'
diff --git a/tests/test_utils/mock_openai_client.py b/tests/test_utils/mock_openai_client.py
deleted file mode 100644
index 5f466e30f..000000000
--- a/tests/test_utils/mock_openai_client.py
+++ /dev/null
@@ -1,117 +0,0 @@
-from typing import List, Dict, Any, Optional
-from unittest.mock import MagicMock
-
-from tests.test_utils.utils import hash_dict
-
-
-class MockChoice:
-    def __init__(
-        self,
-        content: str,
-        finish_reason: str = "stop",
-        refusal: Optional[str] = None
-    ):
-        self.message = MagicMock()
-        self.message.content = content
-        self.message.refusal = refusal
-        self.finish_reason = finish_reason
-        self.delta = MagicMock()
-        self.delta.content = content
-
-
-class MockCompletionResponse:
-    def __init__(self, choices: List[MockChoice]):
-        self.choices = choices
-
-
-class MockStreamingChunk:
-    def __init__(self, content: Optional[str] = None):
-        self.choices = []
-        if content is not None:
-            choice = MagicMock()
-            delta = MagicMock()
-            delta.content = content
-            choice.delta = delta
-            self.choices = [choice]
-
-
-class MockOpenAIClient:
-    """Mock for OpenAI client that can be used to test vLLM integration"""
-
-    def __init__(self):
-        self.chat = MagicMock()
-        self.chat.completions = MagicMock()
-        self.chat.completions.create = MagicMock()
-
-        # The method that will be called by the model when it makes a request
-        def _create(**kwargs):
-            # Hash the arguments to create a unique key
-            request_key = hash_dict(kwargs)
-            response = self._mock_responses.get(request_key)
-            if not response:
-                raise ValueError(f"No response found for {kwargs}")
-            if kwargs.get("stream", False):
-                return self._create_streaming_response(response)
-            else:
-                return self._create_standard_response(response)
-
-        self.chat.completions.create.side_effect = _create
-        self._mock_responses: Dict[str, Any] = {}
-
-    def add_mock_responses(self, mocks: list):
-        for kwargs, response in mocks:
-            request_key = hash_dict(kwargs)
-            self._mock_responses[request_key] = response
-
-    def _create_standard_response(self, response):
-        if isinstance(response, str):
-            response = [response]
-        choices = [MockChoice(content=chunk) for chunk in response]
-        return MockCompletionResponse(choices=choices)
-
-    def _create_streaming_response(self, response):
-        chunks = [MockStreamingChunk(content=chunk) for chunk in response]
-        return iter(chunks)
-
-
-class MockAsyncOpenAIClient:
-    """Mock for AsyncOpenAI client that can be used to test AsyncVLLM integration"""
-
-    def __init__(self):
-        self.chat = MagicMock()
-        self.chat.completions = MagicMock()
-        self.chat.completions.create = MagicMock()
-
-        # The method that will be called by the model when it makes a request
-        async def _async_create(**kwargs):
-            # Hash the arguments to create a unique key
-            request_key = hash_dict(kwargs)
-            response = self._mock_responses.get(request_key)
-            if not response:
-                raise ValueError(f"No response found for {kwargs}")
-            if kwargs.get("stream", False):
-                return self._create_async_streaming_response(response)
-            else:
-                return await self._create_async_standard_response(response)
-
-        self.chat.completions.create.side_effect = _async_create
-        self._mock_responses: Dict[str, Any] = {}
-
-    def add_mock_responses(self, mocks: list):
-        for kwargs, response in mocks:
-            request_key = hash_dict(kwargs)
-            self._mock_responses[request_key] = response
-
-    async def _create_async_standard_response(self, response):
-        """Create an async standard (non-streaming) response"""
-        if isinstance(response, str):
-            response = [response]
-        choices = [MockChoice(content=chunk) for chunk in response]
-        return MockCompletionResponse(choices=choices)
-
-    async def _create_async_streaming_response(self, response):
-        """Create an async streaming response generator"""
-        chunks = [MockStreamingChunk(content=chunk) for chunk in response]
-
-        for chunk in chunks:
-            yield chunk
diff --git a/tests/test_utils/mock_tgi_client.py b/tests/test_utils/mock_tgi_client.py
deleted file mode 100644
index fef054682..000000000
--- a/tests/test_utils/mock_tgi_client.py
+++ /dev/null
@@ -1,63 +0,0 @@
-from typing import Any, Dict
-from unittest.mock import MagicMock
-
-from tests.test_utils.utils import hash_dict
-
-
-class MockTGIInferenceClient:
-    """Mock for TGI `InferenceClient` that can be used to test the TGI model"""
-
-    def __init__(self):
-        self.text_generation = MagicMock()
-
-        # The method that will be called by the model when it makes a request
-        def _create(**kwargs):
-            # Hash the arguments to create a unique key
-            request_key = hash_dict(kwargs)
-            response = self._mock_responses.get(request_key)
-            if not response:
-                raise ValueError(f"No response found for {kwargs}")
-            if kwargs.get("stream", False):
-                return iter(response)
-            else:
-                return response
-
-        self.text_generation.side_effect = _create
-        self._mock_responses: Dict[str, Any] = {}
-
-    def add_mock_responses(self, mocks: list):
-        for kwargs, response in mocks:
-            request_key = hash_dict(kwargs)
-            self._mock_responses[request_key] = response
-
-
-class MockAsyncTGIInferenceClient:
-    """Mock for TGI `InferenceClient` that can be used to test the TGI model"""
-
-    def __init__(self):
-        self.text_generation = MagicMock()
-
-        # The method that will be called by the model when it makes a request
-        async def _async_create(**kwargs):
-            # Hash the arguments to create a unique key
-            request_key = hash_dict(kwargs)
-            response = self._mock_responses.get(request_key)
-            if not response:
-                raise ValueError(f"No response found for {kwargs}")
-            if kwargs.get("stream", False):
-                return self._create_async_streaming_response(response)
-            else:
-                return response
-
-        self.text_generation.side_effect = _async_create
-        self._mock_responses: Dict[str, Any] = {}
-
-    def add_mock_responses(self, mocks: list):
-        for kwargs, response in mocks:
-            request_key = hash_dict(kwargs)
-            self._mock_responses[request_key] = response
-
-    async def _create_async_streaming_response(self, response):
-        """Create an async streaming response generator"""
-        for chunk in response:
-            yield chunk
diff --git a/tests/test_utils/utils.py b/tests/test_utils/utils.py
deleted file mode 100644
index d3a3fa969..000000000
--- a/tests/test_utils/utils.py
+++ /dev/null
@@ -1,23 +0,0 @@
-import hashlib
-import pickle
-import sys
-
-
-def hash_dict(d) -> str:
-    def make_hashable(obj):
-        if isinstance(obj, (bool, int, float, str, type(None))):
-            if isinstance(obj, str):
-                return sys.intern(obj)
-            return obj
-        if isinstance(obj, dict):
-            return tuple(sorted(
-                (sys.intern(k) if isinstance(k, str) else k, make_hashable(v))
-                for k, v in obj.items()
-            ))
-        if isinstance(obj, (list, tuple)):
-            return tuple(make_hashable(e) for e in obj)
-        return str(obj)
-
-    hashable_obj = make_hashable(d)
-    pickled_obj = pickle.dumps(hashable_obj, protocol=4)
-    return hashlib.sha256(pickled_obj).hexdigest()
diff --git a/tests/types/test_custom_types.py b/tests/types/test_custom_types.py
index 691b681b8..46df9ba4a 100644
--- a/tests/types/test_custom_types.py
+++ b/tests/types/test_custom_types.py
@@ -4,6 +4,7 @@
 from pydantic import BaseModel
 
 from outlines import types
+from outlines.fsm.types import python_types_to_regex
 from outlines.types.dsl import to_regex
 
 
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index 4e13b1d38..8351ef319 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -1,25 +1,6 @@
-import datetime
-import json
-import sys
-import tempfile
-from dataclasses import dataclass
-from enum import Enum
-from typing import (
-    Dict,
-    List,
-    Literal,
-    Tuple,
-    Union,
-    get_args,
-    Optional as PyOptional
-)
-
-import interegular
 import pytest
-from genson import SchemaBuilder
 from pydantic import BaseModel
 
-from outlines import grammars, types
 from outlines.types.dsl import (
     Alternatives,
     JsonSchema,
@@ -35,14 +16,6 @@
     String,
     Term,
     either,
-    CFG,
-    FSM,
-    _handle_dict,
-    _handle_list,
-    _handle_literal,
-    _handle_tuple,
-    _handle_union,
-    json_schema,
     one_or_more,
     zero_or_more,
     optional,
@@ -51,48 +24,34 @@
     at_least,
     exactly,
     regex,
-    python_types_to_terms,
-    # deprecated
-    repeat,
-    times,
+    json_schema,
 )
 
-if sys.version_info >= (3, 12):
-    from typing import TypedDict
-else:
-    from typing_extensions import TypedDict
-
 
 def test_dsl_init():
     string = String("test")
     assert string.value == "test"
     assert repr(string) == "String(value='test')"
-    assert string.display_ascii_tree() == "└── String('test')\n"
 
     regex = Regex("[0-9]")
     assert regex.pattern == "[0-9]"
     assert repr(regex) == "Regex(pattern='[0-9]')"
-    assert regex.display_ascii_tree() == "└── Regex('[0-9]')\n"
 
     schema = JsonSchema('{ "type": "string" }')
     assert schema.schema == '{ "type": "string" }'
     assert repr(schema) == 'JsonSchema(schema=\'{ "type": "string" }\')'
-    assert schema.display_ascii_tree() == "└── JsonSchema('{ \"type\": \"string\" }')\n"
 
     kleene_star = KleeneStar(string)
     assert kleene_star.term == string
     assert repr(kleene_star) == "KleeneStar(term=String(value='test'))"
-    assert kleene_star.display_ascii_tree() == "└── KleeneStar(*)\n    └── String('test')\n"
 
     kleene_plus = KleenePlus(string)
     assert kleene_plus.term == string
     assert repr(kleene_plus) == "KleenePlus(term=String(value='test'))"
-    assert kleene_plus.display_ascii_tree() == "└── KleenePlus(+)\n    └── String('test')\n"
 
     optional = Optional(string)
     assert optional.term == string
     assert repr(optional) == "Optional(term=String(value='test'))"
-    assert optional.display_ascii_tree() == "└── Optional(?)\n    └── String('test')\n"
 
     alternatives = Alternatives([string, regex])
     assert alternatives.terms[0] == string
@@ -101,7 +60,6 @@ def test_dsl_init():
         repr(alternatives)
         == "Alternatives(terms=[String(value='test'), Regex(pattern='[0-9]')])"
     )
-    assert alternatives.display_ascii_tree() == "└── Alternatives(|)\n    ├── String('test')\n    └── Regex('[0-9]')\n"
 
     sequence = Sequence([string, regex])
     assert sequence.terms[0] == string
@@ -110,25 +68,21 @@ def test_dsl_init():
         repr(sequence)
         == "Sequence(terms=[String(value='test'), Regex(pattern='[0-9]')])"
     )
-    assert sequence.display_ascii_tree() == "└── Sequence\n    ├── String('test')\n    └── Regex('[0-9]')\n"
 
     exact = QuantifyExact(string, 3)
     assert exact.term == string
     assert exact.count == 3
     assert repr(exact) == "QuantifyExact(term=String(value='test'), count=3)"
-    assert exact.display_ascii_tree() == "└── Quantify({3})\n    └── String('test')\n"
 
     minimum = QuantifyMinimum(string, 3)
     assert minimum.term == string
     assert minimum.min_count == 3
     assert repr(minimum) == "QuantifyMinimum(term=String(value='test'), min_count=3)"
-    assert minimum.display_ascii_tree() == "└── Quantify({3,})\n    └── String('test')\n"
 
     maximum = QuantifyMaximum(string, 3)
     assert maximum.term == string
     assert maximum.max_count == 3
     assert repr(maximum) == "QuantifyMaximum(term=String(value='test'), max_count=3)"
-    assert maximum.display_ascii_tree() == "└── Quantify({,3})\n    └── String('test')\n"
 
     between = QuantifyBetween(string, 1, 3)
     assert between.term == string
@@ -138,7 +92,6 @@ def test_dsl_init():
         repr(between)
         == "QuantifyBetween(term=String(value='test'), min_count=1, max_count=3)"
     )
-    assert between.display_ascii_tree() == "└── Quantify({1,3})\n    └── String('test')\n"
 
     with pytest.raises(
         ValueError, match="`max_count` must be greater than `min_count`"
@@ -146,44 +99,6 @@ def test_dsl_init():
         QuantifyBetween(string, 3, 1)
 
 
-def test_dsl_term_methods():
-    a = String("a")
-    b = Regex("[0-9]")
-    c = "c"
-
-    assert a + b == Sequence([a, b])
-    assert a + c == Sequence([a, String(c)])
-    assert a.__radd__(b) == Sequence([b, a])
-    assert a.__radd__(c) == Sequence([String(c), a])
-
-    assert a | b == Alternatives([a, b])
-    assert a | c == Alternatives([a, String(c)])
-    assert a.__ror__(b) == Alternatives([b, a])
-    assert a.__ror__(c) == Alternatives([String(c), a])
-
-    core_schema = a.__get_pydantic_core_schema__("", "")
-    validator = a.__get_validator__(core_schema)
-    assert validator("a") == "a"
-    with pytest.raises(
-        ValueError,
-        match="Input should be in the language of the regular expression",
-    ):
-        validator("b")
-
-    assert a.__get_pydantic_json_schema__("", "") == {"type": "string", "pattern": "a"}
-
-    assert a.matches("a")
-    assert not a.matches("b")
-
-    assert a.display_ascii_tree() == "└── String('a')\n"
-
-    with pytest.raises(NotImplementedError):
-        Term()._display_node()
-
-    assert a._display_children("") == ""
-
-    assert a.__str__() == "└── String('a')\n"
-
 def test_dsl_sequence():
     a = String("a")
     b = String("b")
@@ -229,7 +144,7 @@ def test_dsl_alternatives():
 def test_dsl_optional():
     a = String("a")
 
-    opt = a.optional()
+    opt = optional(a)
     assert isinstance(opt, Optional)
 
     opt = optional("a")
@@ -243,7 +158,7 @@ def test_dsl_optional():
 def test_dsl_exactly():
     a = String("a")
 
-    rep = a.exactly(2)
+    rep = exactly(2, a)
     assert isinstance(rep, QuantifyExact)
     assert rep.count == 2
 
@@ -258,7 +173,7 @@ def test_dsl_exactly():
 def test_dsl_at_least():
     a = String("a")
 
-    rep = a.at_least(2)
+    rep = at_least(2, a)
     assert isinstance(rep, QuantifyMinimum)
     assert rep.min_count == 2
 
@@ -273,7 +188,7 @@ def test_dsl_at_least():
 def test_dsl_at_most():
     a = String("a")
 
-    rep = a.at_most(2)
+    rep = at_most(2, a)
     assert isinstance(rep, QuantifyMaximum)
     assert rep.max_count == 2
 
@@ -288,7 +203,7 @@ def test_dsl_at_most():
 def test_between():
     a = String("a")
 
-    rep = a.between(1, 2)
+    rep = between(1, 2, a)
     assert isinstance(rep, QuantifyBetween)
     assert rep.min_count == 1
     assert rep.max_count == 2
@@ -304,7 +219,7 @@ def test_between():
 def test_dsl_zero_or_more():
     a = String("a")
 
-    rep = a.zero_or_more()
+    rep = zero_or_more(a)
     assert isinstance(rep, KleeneStar)
 
     rep = zero_or_more("a")
@@ -318,7 +233,7 @@ def test_dsl_zero_or_more():
 def test_dsl_one_or_more():
     a = String("a")
 
-    rep = a.one_or_more()
+    rep = one_or_more(a)
     assert isinstance(rep, KleenePlus)
 
     rep = one_or_more("a")
@@ -337,49 +252,6 @@ def test_dsl_aliases():
     assert isinstance(test, JsonSchema)
 
 
-def test_dsl_repeat():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `repeat` function/method is deprecated",
-    ):
-        a = String("a")
-
-        # error missing min_count or max_count
-        with pytest.raises(
-            ValueError,
-            match="You must provide a value for at least `min_count` or "
-            "`max_count`",
-        ):
-            repeat(a, None, None)
-
-        # at_least
-        assert repeat(a, 2, None) == at_least(2, a)
-        assert repeat("a", 2, None) == at_least(2, "a")
-        assert a.repeat(2, None) == a.at_least(2)
-
-        # at_most
-        assert repeat(a, None, 2) == at_most(2, a)
-        assert repeat("a", None, 2) == at_most(2, "a")
-        assert a.repeat(None, 2) == a.at_most(2)
-
-        # between
-        assert repeat(a, 1, 2) == between(1, 2, a)
-        assert repeat("a", 1, 2) == between(1, 2, "a")
-        assert a.repeat(1, 2) == a.between(1, 2)
-
-
-def test_dsl_times():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `times` function/method is deprecated",
-    ):
-        a = String("a")
-
-        assert times(a, 2) == exactly(2, a)
-        assert times("a", 2) == exactly(2, "a")
-        assert a.times(2) == a.exactly(2)
-
-
 def test_dsl_term_pydantic_simple():
     a = String("a")
 
@@ -425,367 +297,3 @@ def test_dsl_display():
         tree
         == "└── Sequence\n    ├── KleeneStar(*)\n    │   └── Alternatives(|)\n    │       ├── String('a')\n    │       └── String('b')\n    └── Regex('[0-9]')\n"
     )
-
-
-def test_cfg():
-    cfg_string = """
-?start: expr
-?expr: NUMBER
-"""
-    cfg = types.cfg(cfg_string)
-    assert isinstance(cfg, CFG)
-    assert cfg.definition.strip() == "?start: expr\n?expr: NUMBER"
-    assert cfg._display_node() == "CFG('\n?start: expr\n?expr: NUMBER\n')"
-    assert cfg.__repr__() == "CFG(definition='\n?start: expr\n?expr: NUMBER\n')"
-    assert cfg == types.cfg(cfg_string)
-    assert not cfg == "a"
-
-
-def test_fsm():
-    interegular_fsm = interegular.parse_pattern(r"a").to_fsm()
-    fsm = types.fsm(interegular_fsm)
-    assert isinstance(fsm, FSM)
-    assert fsm.fsm == interegular_fsm
-    assert fsm._display_node() == f"FSM({fsm.fsm.__repr__()})"
-    assert fsm.__repr__() == f"FSM(fsm={fsm.fsm.__repr__()})"
-
-
-def test_json_schema():
-    # init dict
-    schema = types.json_schema({"type": "string"})
-    assert schema.schema == '{"type": "string"}'
-
-    # init str
-    schema = types.json_schema('{"type": "string"}')
-    assert schema.schema == '{"type": "string"}'
-
-    # init Pydantic model
-    class PydanticModel(BaseModel):
-        foo: str
-    schema = types.json_schema(PydanticModel)
-    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "PydanticModel", "type": "object"}'
-
-    # init TypedDict
-    class SomeTypedDict(TypedDict):
-        foo: str
-    schema = types.json_schema(SomeTypedDict)
-    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "SomeTypedDict", "type": "object"}'
-
-    # init dataclass
-    @dataclass
-    class DataClass:
-        foo: str
-    schema = types.json_schema(DataClass)
-    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "DataClass", "type": "object"}'
-
-    # init SchemaBuilder
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    schema = types.json_schema(builder)
-    assert schema.schema == '{"$schema": "http://json-schema.org/schema#", "type": "object"}'
-
-    # init unsupported type
-    with pytest.raises(ValueError, match="Cannot parse schema"):
-        types.json_schema(1)
-
-    # other methods
-    schema = types.json_schema('{"type": "string"}')
-    schema.__post_init__()
-    assert schema._display_node() == "JsonSchema('{\"type\": \"string\"}')"
-    assert schema.__repr__() == "JsonSchema(schema='{\"type\": \"string\"}')"
-    assert schema == types.json_schema('{"type": "string"}')
-    assert not schema == "a"
-
-
-def test_dsl_cfg_from_file():
-    grammar_content = """
-    ?start: expression
-    ?expression: term (("+" | "-") term)*
-    ?term: factor (("*" | "/") factor)*
-    ?factor: NUMBER
-    """
-    with tempfile.NamedTemporaryFile(mode='w', suffix='.txt', delete=True) as temp_file:
-        temp_file.write(grammar_content)
-        temp_file.flush()
-        temp_file_path = temp_file.name
-        cfg = CFG.from_file(temp_file_path)
-        assert cfg == CFG(grammar_content)
-
-
-def test_dsl_json_schema_from_file():
-    schema_content = """
-    {
-        "type": "object",
-        "properties": {
-            "name": {
-                "type": "string"
-            }
-        }
-    }
-    """
-    with tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=True) as temp_file:
-        temp_file.write(schema_content)
-        temp_file.flush()
-        temp_file_path = temp_file.name
-        schema = JsonSchema.from_file(temp_file_path)
-        assert schema == JsonSchema(schema_content)
-
-
-def test_dsl_python_types_to_terms():
-    with pytest.raises(RecursionError):
-        python_types_to_terms(None, 11)
-
-    term = Term()
-    assert python_types_to_terms(term) == term
-
-    assert python_types_to_terms(int) == types.integer
-    assert python_types_to_terms(float) == types.number
-    assert python_types_to_terms(bool) == types.boolean
-    assert python_types_to_terms(str) == types.string
-    assert python_types_to_terms(datetime.time) == types.time
-    assert python_types_to_terms(datetime.date) == types.date
-    assert python_types_to_terms(datetime.datetime) == types.datetime
-    assert python_types_to_terms(dict) == types.CFG(grammars.json)
-
-    string_instance = "a"
-    assert python_types_to_terms(string_instance) == String(string_instance)
-    int_instance = 1
-    assert python_types_to_terms(int_instance) == Regex(r"1")
-    float_instance = 1.0
-    assert python_types_to_terms(float_instance) == Regex(r"1.0")
-
-    @dataclass
-    class DataClass:
-        a: int
-        b: str
-
-    assert python_types_to_terms(DataClass) == JsonSchema(
-        {
-            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
-            "required": ["a", "b"],
-            "title": "DataClass",
-            "type": "object",
-        }
-    )
-
-    class SomeTypedDict(TypedDict):
-        a: int
-        b: str
-
-    assert python_types_to_terms(SomeTypedDict) == JsonSchema(
-        {
-            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
-            "required": ["a", "b"],
-            "title": "SomeTypedDict",
-            "type": "object",
-        }
-    )
-
-    class PydanticModel(BaseModel):
-        a: int
-        b: str
-
-    assert python_types_to_terms(PydanticModel) == JsonSchema(
-        {
-            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
-            "required": ["a", "b"],
-            "title": "PydanticModel",
-            "type": "object",
-        }
-    )
-
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"hi": "there"})
-    builder.add_object({"hi": 5})
-    assert python_types_to_terms(builder) == JsonSchema(
-        {
-            "$schema": "http://json-schema.org/schema#",
-            "type": "object",
-            "properties": {"hi": {"type": ["integer", "string"]}},
-            "required": ["hi"]
-        }
-    )
-
-    interegular_fsm = interegular.parse_pattern(r"abc").to_fsm()
-    assert python_types_to_terms(types.fsm(interegular_fsm)).fsm is interegular_fsm
-
-    def func(a: int, b: str):
-        return (a, b)
-
-    assert python_types_to_terms(func) == JsonSchema(
-        {
-            "type": "object",
-            "properties": {
-                "a": {"title": "A", "type": "integer"},
-                "b": {"title": "B", "type": "string"},
-            },
-            "required": ["a", "b"],
-            "title": "func",
-        }
-    )
-
-    class SomeEnum(Enum):
-        a = "a"
-        b = int
-        c = func
-
-    result = python_types_to_terms(SomeEnum)
-    assert isinstance(result, Alternatives)
-    assert len(result.terms) == 3
-    assert result.terms[0] == String("a")
-    assert result.terms[1] == types.integer
-    assert isinstance(result.terms[2], JsonSchema)
-    schema_dict = json.loads(result.terms[2].schema)
-    assert schema_dict == {
-        "properties": {
-            "a": {"title": "A", "type": "integer"},
-            "b": {"title": "B", "type": "string"},
-        },
-        "required": ["a", "b"],
-        "title": "func",
-        "type": "object",
-    }
-
-    # for generic types we only test the dispatch as the functions that
-    # convert to terms are tested in distinct tests below
-    assert python_types_to_terms(Literal["a", "b"]) == _handle_literal(("a", "b"))
-    assert python_types_to_terms(Union[int, str]) == _handle_union((int, str), recursion_depth=0)
-    assert python_types_to_terms(list[int]) == _handle_list((int,), recursion_depth=0)
-    assert python_types_to_terms(tuple[int, str]) == _handle_tuple((int, str), recursion_depth=0)
-    assert python_types_to_terms(dict[int, str]) == _handle_dict((int, str), recursion_depth=0)
-
-
-def test_dsl_handle_literal():
-    literal = Literal["a", 1]
-    result = _handle_literal(get_args(literal))
-    assert isinstance(result, Alternatives)
-    assert len(result.terms) == 2
-    assert result.terms[0] == String("a")
-    assert result.terms[1] == Regex(r"1")
-
-
-def test_dsl_handle_union():
-    # test simple Union
-    simple_union = Union[int, str]
-    result = _handle_union(get_args(simple_union), recursion_depth=0)
-    assert isinstance(result, Alternatives)
-    assert len(result.terms) == 2
-    assert result.terms[0] == types.integer
-    assert result.terms[1] == types.string
-
-    # test with Optional[T]
-    optional_type = PyOptional[int]
-    result = _handle_union(get_args(optional_type), recursion_depth=0)
-    assert isinstance(result, Alternatives)
-    assert len(result.terms) == 2
-    assert result.terms[0] == types.integer
-    assert result.terms[1] == String("None")
-
-    # test with more complex types
-    class TestModel(BaseModel):
-        field: str
-
-    class TestEnum(Enum):
-        a = "a"
-        b = "b"
-
-    complex_union = Union[TestModel, TestEnum]
-    result = _handle_union(get_args(complex_union), recursion_depth=0)
-    assert isinstance(result, Alternatives)
-    assert len(result.terms) == 2
-    assert isinstance(result.terms[0], JsonSchema)
-    assert isinstance(result.terms[1], Alternatives)
-    assert len(result.terms[1].terms) == 2
-    assert result.terms[1].terms[0] == String("a")
-    assert result.terms[1].terms[1] == String("b")
-
-
-def test_dsl_handle_list():
-    with pytest.raises(TypeError):
-        _handle_list((int, str), recursion_depth=0)
-
-    # simple type
-    list_type = list[int]
-    result = _handle_list(get_args(list_type), recursion_depth=0)
-    assert isinstance(result, Sequence)
-    assert len(result.terms) == 4
-    assert result.terms[0] == String("[")
-    assert result.terms[1] == types.integer
-    assert isinstance(result.terms[2], KleeneStar)
-    assert result.terms[2].term == Sequence([String(", "), types.integer])
-    assert result.terms[3] == String("]")
-
-    # more complex type
-    list_type = list[Union[int, str]]
-    result = _handle_list(get_args(list_type), recursion_depth=0)
-    assert isinstance(result, Sequence)
-    assert len(result.terms) == 4
-    assert result.terms[0] == String("[")
-    assert result.terms[1] == _handle_union(get_args(Union[int, str]), recursion_depth=0)
-    assert isinstance(result.terms[2], KleeneStar)
-    assert result.terms[2].term == Sequence([String(", "), _handle_union(get_args(Union[int, str]), recursion_depth=0)])
-    assert result.terms[3] == String("]")
-
-
-def test_dsl_handle_tuple():
-    # empty tuple
-    tuple_type = Tuple[()]
-    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
-    assert isinstance(result, String)
-    assert result.value == "()"
-
-    # tuple with ellipsis
-    tuple_type = tuple[int, ...]
-    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
-    assert isinstance(result, Sequence)
-    assert len(result.terms) == 4
-    assert result.terms[0] == String("(")
-    assert result.terms[1] == types.integer
-    assert isinstance(result.terms[2], KleeneStar)
-    assert result.terms[2].term == Sequence([String(", "), types.integer])
-    assert result.terms[3] == String(")")
-
-    # tuple with fixed length
-    tuple_type = tuple[int, str]
-    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
-    assert isinstance(result, Sequence)
-    assert len(result.terms) == 5
-    assert result.terms[0] == String("(")
-    assert result.terms[1] == types.integer
-    assert result.terms[2] == String(", ")
-    assert result.terms[3] == types.string
-    assert result.terms[4] == String(")")
-
-    # tuple with fixed length and complex types
-    tuple_type = tuple[int, Union[str, int]]
-    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
-    assert isinstance(result, Sequence)
-    assert len(result.terms) == 5
-    assert result.terms[0] == String("(")
-    assert result.terms[1] == types.integer
-    assert result.terms[2] == String(", ")
-    assert result.terms[3] == _handle_union(get_args(Union[str, int]), recursion_depth=0)
-    assert result.terms[4] == String(")")
-
-
-def test_dsl_handle_dict():
-    # args of incorrect length
-    with pytest.raises(TypeError):
-        incorrect_dict_type = dict[int, str, int]
-        _handle_dict(get_args(incorrect_dict_type), recursion_depth=0)
-
-    # correct type
-    dict_type = dict[int, str]
-    result = _handle_dict(get_args(dict_type), recursion_depth=0)
-    assert isinstance(result, Sequence)
-    assert len(result.terms) == 3
-    assert result.terms[0] == String("{")
-    assert isinstance(result.terms[1], Optional)
-    assert isinstance(result.terms[1].term, Sequence)
-    assert len(result.terms[1].term.terms) == 4
-    assert result.terms[1].term.terms[0] == types.integer
-    assert result.terms[1].term.terms[1] == String(":")
-    assert result.terms[1].term.terms[2] == types.string
-    assert result.terms[1].term.terms[3] == KleeneStar(Sequence([String(", "), types.integer, String(":"), types.string]))
-    assert result.terms[2] == String("}")
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
deleted file mode 100644
index 4d73312d6..000000000
--- a/tests/types/test_types_utils.py
+++ /dev/null
@@ -1,409 +0,0 @@
-import datetime
-import pytest
-import sys
-from dataclasses import dataclass
-from enum import Enum
-from functools import partial
-from typing import (
-    Annotated,
-    Any,
-    Dict,
-    List,
-    Literal,
-    NewType,
-    Optional,
-    Tuple,
-    Union
-)
-
-import interegular
-from genson import SchemaBuilder
-from pydantic import BaseModel, constr
-
-from outlines.types.dsl import JsonSchema
-from outlines.types.utils import (
-    get_enum_from_literal,
-    get_schema_from_enum,
-    get_schema_from_signature,
-    is_bool,
-    is_callable,
-    is_date,
-    is_dataclass,
-    is_datetime,
-    is_enum,
-    is_float,
-    is_float_instance,
-    is_genson_schema_builder,
-    is_int,
-    is_int_instance,
-    is_interegular_fsm,
-    is_literal,
-    is_native_dict,
-    is_pydantic_model,
-    is_str,
-    is_str_instance,
-    is_time,
-    is_typed_dict,
-    is_typing_dict,
-    is_typing_list,
-    is_typing_tuple,
-    is_union
-)
-
-if sys.version_info >= (3, 12):
-    from typing import TypedDict
-else:
-    from typing_extensions import TypedDict
-
-
-# Type identification
-
-
-@pytest.fixture
-def sample_enum():
-    class SampleEnum(Enum):
-        A = 1
-        B = 2
-
-    return SampleEnum
-
-@pytest.fixture
-def sample_complex_enum():
-    def add_func(a: float, b: float) -> float:
-        return a + b
-
-    class SampleComplexEnum(Enum):
-        add = partial(add_func)
-        a = "a"
-        b = 2
-
-    return SampleComplexEnum
-
-@pytest.fixture
-def sample_empty_enum():
-    def add_func(a: float, b: float) -> float:
-        return a + b
-
-    # the enum is empty because the function is not registered as callable
-    class SampleEmptyEnum(Enum):
-        add = add_func
-
-    return SampleEmptyEnum
-
-@pytest.fixture
-def sample_class():
-    class SampleClass:
-        pass
-
-    return SampleClass
-
-@pytest.fixture
-def sample_dataclass():
-    @dataclass
-    class SampleDataclass:
-        field1: str
-        field2: int
-
-    return SampleDataclass
-
-@pytest.fixture
-def sample_typed_dict():
-    class SampleTypedDict(TypedDict):
-        name: str
-        age: int
-
-    return SampleTypedDict
-
-@pytest.fixture
-def sample_pydantic_model():
-    class SamplePydanticModel(BaseModel):
-        name: str
-        age: int
-
-    return SamplePydanticModel
-
-@pytest.fixture
-def sample_schema_builder():
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"hi": "there"})
-    builder.add_object({"hi": 5})
-    return builder
-
-@pytest.fixture
-def sample_function():
-    def sample_function(foo: str, bar: List[int]):
-        pass
-
-    return sample_function
-
-@pytest.fixture
-def sample_function_missing_type():
-    def sample_function(foo, bar: List[int]):
-        pass
-
-    return sample_function
-
-
-def test_is_int():
-    assert is_int(int)
-    assert not is_int(float)
-    assert not is_int(1)
-    assert not is_int(List[int])
-    assert not is_int(Dict[int, int])
-    assert is_int(Annotated[int, "some metadata"])
-    assert not is_int(Annotated[str, "some metadata"])
-    assert is_int(NewType("UserId", int))
-    assert not is_int(NewType("UserId", str))
-
-
-def test_is_int_instance():
-    assert is_int_instance(1)
-    assert not is_int_instance(1.0)
-    assert not is_int_instance("1")
-    assert not is_int_instance(int)
-
-
-def test_is_float():
-    assert is_float(float)
-    assert not is_float(int)
-    assert not is_float(1.0)
-    assert not is_float(List[float])
-    assert not is_float(Dict[float, float])
-    assert is_float(Annotated[float, "some metadata"])
-    assert not is_float(Annotated[int, "some metadata"])
-    assert is_float(NewType("UserId", float))
-    assert not is_float(NewType("UserId", int))
-
-
-def test_is_float_instance():
-    assert is_float_instance(1.0)
-    assert not is_float_instance(1)
-    assert not is_float_instance("1.0")
-    assert not is_float_instance(float)
-
-
-def test_is_str():
-    assert is_str(str)
-    assert not is_str(int)
-    assert not is_str("hello")
-    assert not is_str(List[str])
-    assert not is_str(Dict[str, str])
-    assert is_str(Annotated[str, "some metadata"])
-    assert not is_str(Annotated[int, "some metadata"])
-    assert is_str(NewType("UserId", str))
-    assert not is_str(NewType("UserId", int))
-
-
-def test_is_str_instance():
-    assert is_str_instance("hello")
-    assert is_str_instance("")
-    assert is_str_instance("123")
-    assert not is_str_instance(123)
-    assert not is_str_instance(str)
-
-
-def test_is_bool():
-    assert is_bool(bool)
-    assert not is_bool(int)
-    assert not is_bool(True)
-    assert is_bool(Annotated[bool, "some metadata"])
-    assert not is_bool(Annotated[int, "some metadata"])
-    assert is_bool(NewType("UserId", bool))
-    assert not is_bool(NewType("UserId", int))
-
-
-def test_is_datetime():
-    assert is_datetime(datetime.datetime)
-    assert not is_datetime(datetime.date)
-    assert not is_datetime(datetime.time)
-    assert not is_datetime(datetime.datetime.now())
-
-
-def test_is_date():
-    assert is_date(datetime.date)
-    assert not is_date(datetime.datetime)
-    assert not is_date(datetime.time)
-    assert not is_date(datetime.date.today())
-
-
-def test_is_time():
-    assert is_time(datetime.time)
-    assert not is_time(datetime.datetime)
-    assert not is_time(datetime.date)
-    assert not is_time(datetime.time(12, 30))
-
-
-def test_is_native_dict():
-    assert is_native_dict(dict)
-    assert not is_native_dict({})
-    assert not is_native_dict({"key": "value"})
-    assert not is_native_dict(list)
-    assert not is_native_dict(dict[str, int])
-
-
-def test_is_typing_dict():
-    assert is_typing_dict(dict[str, int])
-    assert is_typing_dict(Dict[int, str])
-    assert not is_typing_dict(dict)
-    assert not is_typing_dict({})
-
-
-def test_is_typing_list():
-    assert is_typing_list(list[int])
-    assert is_typing_list(List[int])
-    assert not is_typing_list(list)
-    assert not is_typing_list([])
-    assert not is_typing_list(dict)
-
-
-def test_is_typing_tuple():
-    assert is_typing_tuple(tuple[int, str])
-    assert is_typing_tuple(Tuple[int, str])
-    assert not is_typing_tuple(tuple)
-    assert not is_typing_tuple(())
-    assert not is_typing_tuple(list)
-
-
-def test_is_union():
-    assert is_union(Union[int, str])
-    assert is_union(Optional[int])
-    assert not is_union(list)
-    assert not is_union(["a", "b"])
-    assert not is_union(Literal[int, str])
-
-
-def test_is_literal():
-    assert is_literal(Literal["a", "b"])
-    assert not is_literal(str)
-    assert not is_literal("a")
-    assert not is_literal(["a", "b"])
-    assert not is_literal(Union[str, int])
-
-
-def test_is_dataclass(
-    sample_dataclass,
-    sample_class,
-    sample_typed_dict,
-    sample_pydantic_model
-):
-    assert is_dataclass(sample_dataclass)
-    assert not is_dataclass(sample_dataclass(field1="test", field2=123))
-    assert not is_dataclass(dict)
-    assert not is_dataclass(sample_class)
-    assert not is_dataclass(sample_typed_dict)
-    assert not is_dataclass(sample_pydantic_model)
-
-
-def test_is_typed_dict(
-    sample_typed_dict,
-    sample_class,
-    sample_dataclass,
-    sample_pydantic_model
-):
-    assert is_typed_dict(sample_typed_dict)
-    assert not is_typed_dict(sample_typed_dict(name="test", age=30))
-    assert not is_typed_dict(dict)
-    assert not is_typed_dict(sample_class)
-    assert not is_typed_dict(sample_dataclass)
-    assert not is_typed_dict(sample_pydantic_model)
-
-
-def test_is_pydantic_model(
-    sample_pydantic_model,
-    sample_class,
-    sample_dataclass,
-    sample_typed_dict
-):
-    assert is_pydantic_model(sample_pydantic_model)
-    assert not is_pydantic_model(sample_pydantic_model(name="test", age=30))  # Instance
-    assert not is_pydantic_model(dict)
-    assert not is_pydantic_model(sample_class)
-    assert not is_pydantic_model(sample_dataclass)
-    assert not is_pydantic_model(sample_typed_dict)
-
-
-def test_is_genson_schema_builder(
-    sample_schema_builder,
-    sample_class,
-    sample_dataclass,
-    sample_typed_dict,
-    sample_pydantic_model
-):
-    assert is_genson_schema_builder(sample_schema_builder)
-    assert not is_genson_schema_builder(dict)
-    assert not is_genson_schema_builder(str)
-    assert not is_genson_schema_builder({"type": 'object', "properties": {}})
-    assert not is_genson_schema_builder('{"type": "object", "properties": {}}')
-    assert not is_genson_schema_builder(sample_class)
-    assert not is_genson_schema_builder(sample_dataclass)
-    assert not is_genson_schema_builder(sample_typed_dict)
-    assert not is_genson_schema_builder(sample_pydantic_model)
-
-
-def test_is_enum(sample_enum):
-    assert is_enum(sample_enum)
-    assert not is_enum(sample_enum.A)
-    assert not is_enum(dict)
-    assert not is_enum(Literal["a", "b"])
-    assert not is_enum(["a", "b"])
-
-
-def test_is_callable(sample_function, sample_class, sample_dataclass, sample_typed_dict, sample_pydantic_model):
-    assert is_callable(sample_function)
-    assert is_callable(lambda x: x)
-    assert not is_callable(dict)
-    assert not is_callable(sample_class)
-    assert not is_callable(sample_dataclass)
-    assert not is_callable(sample_typed_dict)
-    assert not is_callable(sample_pydantic_model)
-
-
-def test_is_interegular_fsm():
-    fsm = interegular.parse_pattern(r"[0-9]+").to_fsm()
-    assert is_interegular_fsm(fsm)
-    assert not is_interegular_fsm(dict)
-
-
-# Type conversion
-
-
-def test_get_enum_from_literal(sample_enum):
-    basic_enum = get_enum_from_literal(Literal["a", "b"])
-    assert(is_enum(basic_enum))
-    assert basic_enum.a.value == "a"
-    assert basic_enum.b.value == "b"
-
-    complex_enum = get_enum_from_literal(Literal["a", 1, True, None, sample_enum.A])
-    assert is_enum(complex_enum)
-    assert complex_enum.a.value == "a"
-    assert getattr(complex_enum, "1").value == 1
-    assert getattr(complex_enum, "True").value
-    assert getattr(complex_enum, "None").value is None
-    assert getattr(complex_enum, "SampleEnum.A").value == sample_enum.A
-
-
-def test_get_schema_from_signature(sample_function, sample_function_missing_type):
-    result = get_schema_from_signature(sample_function)
-    assert result["type"] == "object"
-    assert list(result["properties"].keys()) == ["foo", "bar"]
-    assert result["properties"]["foo"]["type"] == "string"
-    assert result["properties"]["bar"]["type"] == "array"
-    assert result["properties"]["bar"]["items"]["type"] == "integer"
-
-    # in case of a function missing type annotations
-    with pytest.raises(ValueError):
-        get_schema_from_signature(sample_function_missing_type)
-
-
-def test_get_schema_from_enum(sample_complex_enum, sample_empty_enum):
-    schema = get_schema_from_enum(sample_complex_enum)
-    assert JsonSchema(schema)
-    assert schema["title"] == sample_complex_enum.__name__
-    assert len(schema["oneOf"]) == len(sample_complex_enum)
-    for elt in schema["oneOf"]:
-        assert type(elt) in [int, float, bool, type(None), str, dict]
-
-    # in case of an empty enum because the function member is not registered as callable
-    with pytest.raises(ValueError):
-        get_schema_from_enum(sample_empty_enum)
diff --git a/tests/v0_legacy/generate/test_api_legacy.py b/tests/v0_legacy/generate/test_api_legacy.py
deleted file mode 100644
index 75cf72ff0..000000000
--- a/tests/v0_legacy/generate/test_api_legacy.py
+++ /dev/null
@@ -1,186 +0,0 @@
-from io import BytesIO
-from dataclasses import asdict
-from typing import Generator
-from urllib.request import urlopen
-
-import pytest
-import transformers
-from PIL import Image
-from anthropic import Anthropic as AnthropicClient
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers(
-            "erwanf/gpt2-mini"
-        )
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=transformers.LlavaForConditionalGeneration,
-            processor_class=transformers.AutoProcessor,
-        )
-
-
-@pytest.fixture(scope="session")
-def llama_cpp_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        return models.llamacpp(
-            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        )
-
-
-def create_generator_v0_adapter(model):
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(
-            model,
-            int,
-            samplers.greedy(),
-        )
-    else:
-        return GeneratorV0Adapter(
-            model,
-            int,
-            samplers.greedy(),
-        )
-
-
-def test_generator_v0_adapter_init(transformers_model):
-    # wrong model type
-    model = models.from_anthropic(AnthropicClient(), "claude-3-haiku-20240307")
-    with pytest.raises(
-        ValueError,
-        match="You can only use the v0 API with models that were already"
-        + "available in v0. Got <class 'outlines.models.anthropic.Anthropic'>.",
-    ):
-        GeneratorV0Adapter(model, None, samplers.greedy())
-
-    # unknown output type
-    with pytest.raises(
-        TypeError,
-        match="is currently not supported",
-    ):
-        GeneratorV0Adapter(
-            transformers_model, type, samplers.greedy()
-        )
-
-    # valid initialization
-    generator_v0_adapter = GeneratorV0Adapter(
-        transformers_model,
-        int,
-        samplers.greedy(),
-    )
-    assert isinstance(generator_v0_adapter, GeneratorV0Adapter)
-    assert generator_v0_adapter.model == transformers_model
-    assert generator_v0_adapter.sampling_params == asdict(samplers.greedy().sampling_params)
-    assert isinstance(generator_v0_adapter.generator, SteerableGenerator)
-    assert generator_v0_adapter.generator.model == transformers_model
-    assert isinstance(generator_v0_adapter.generator.logits_processor, RegexLogitsProcessor)
-
-
-def test_generator_v0_adapter_create_inference_params(
-    llama_cpp_model,
-    transformers_vision_model,
-    transformers_model,
-):
-    args = {
-        "max_tokens": 10,
-        "stop_at": "\n",
-        "seed": 42
-    }
-    kwargs = {
-        "foo": "bar",
-    }
-
-    result = (
-        create_generator_v0_adapter(llama_cpp_model)
-        .create_inference_params(**args, **kwargs)
-    )
-    assert isinstance(result, dict)
-
-    result = (
-        create_generator_v0_adapter(transformers_vision_model)
-        .create_inference_params(**args, **kwargs)
-    )
-    assert isinstance(result, dict)
-
-    result = (
-        create_generator_v0_adapter(transformers_model)
-        .create_inference_params(**args, **kwargs)
-    )
-    assert isinstance(result, dict)
-
-
-def test_generator_vision_v0_adapter_merge_prompts_and_media(transformers_vision_model):
-    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
-    prompts = ["Hello, world!", "Goodbye, world!"]
-    media = [
-        img_from_url(
-            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-        ),
-        img_from_url(
-            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-        )
-    ]
-    result = generator_v0_adapter.merge_prompts_and_media(prompts, media)
-    assert isinstance(result, dict)
-    assert result["text"] == prompts
-    assert result["images"] == media
-
-
-def test_generator_v0_adapter_call_stream(llama_cpp_model, transformers_model):
-    generator_v0_adapter = create_generator_v0_adapter(transformers_model)
-    result = generator_v0_adapter("Hello, world!", max_tokens=10)
-    assert isinstance(result, str)
-    result = generator_v0_adapter(["Hello, world!", "Goodbye, world!"], max_tokens=10)
-    assert isinstance(result, list)
-    for r in result:
-        assert isinstance(r, str)
-
-    generator_v0_adapter = create_generator_v0_adapter(llama_cpp_model)
-    gen = generator_v0_adapter.stream("Hello, world!", max_tokens=10)
-    assert isinstance(gen, Generator)
-
-
-def test_generator_v0_adapter_vision_call_stream(transformers_vision_model):
-    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
-    media = img_from_url(
-        "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-    )
-    result = generator_v0_adapter("Hello, world!<image>", media, max_tokens=10)
-    assert isinstance(result, str)
-    with pytest.raises(
-        NotImplementedError,
-        match="Streaming is not implemented for Transformers models",
-    ):
-        gen = generator_v0_adapter.stream("Hello, world!<image>", media, max_tokens=10)
-        assert isinstance(gen, Generator)
diff --git a/tests/v0_legacy/generate/test_cfg_legacy.py b/tests/v0_legacy/generate/test_cfg_legacy.py
deleted file mode 100644
index 2b8a2734e..000000000
--- a/tests/v0_legacy/generate/test_cfg_legacy.py
+++ /dev/null
@@ -1,111 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import grammars, models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import CFGLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.cfg import cfg
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def llama_cpp_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        return models.llamacpp(
-            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf"
-        )
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_cfg_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use CFG-based structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API.",
-        ):
-            cfg(openai_model, grammars.arithmetic)
-
-
-def test_cfg_legacy_llama_cpp(llama_cpp_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Not yet available due to bug in llama_cpp tokenizer",
-        ):
-            cfg(llama_cpp_model, grammars.arithmetic)
-
-
-def test_cfg_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        generator = cfg(transformers_vision_model, grammars.arithmetic)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
-
-
-def test_cfg_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        generator = cfg(transformers_model, grammars.arithmetic, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_choice_legacy.py b/tests/v0_legacy/generate/test_choice_legacy.py
deleted file mode 100644
index fbc4650e2..000000000
--- a/tests/v0_legacy/generate/test_choice_legacy.py
+++ /dev/null
@@ -1,97 +0,0 @@
-import json as pyjson
-from dataclasses import asdict
-from enum import Enum
-from typing import Callable
-
-import pytest
-from pydantic import BaseModel
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.choice import choice
-
-
-class EnumChoice(Enum):
-    foo = "foo"
-    bar = "bar"
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-def test_choice_legacy_standard_enum(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `choice` function is deprecated",
-    ):
-        generator = choice(transformers_model, EnumChoice, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_choice_legacy_standard_list_strings(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `choice` function is deprecated",
-    ):
-        generator = choice(transformers_model, ["foo", "bar"])
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert generator.format_sequence("foo") == "foo"
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_choice_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `choice` function is deprecated",
-    ):
-        with pytest.raises(
-            ValueError,
-            match="The `choice` function with OpenAI only supports a list of "
-            + "strings as choices.",
-        ):
-            choice(openai_model, EnumChoice)
-
-        generator = choice(openai_model, ["foo", "bar"])
-        assert isinstance(generator, Callable)
diff --git a/tests/v0_legacy/generate/test_format_legacy.py b/tests/v0_legacy/generate/test_format_legacy.py
deleted file mode 100644
index 81503fb8f..000000000
--- a/tests/v0_legacy/generate/test_format_legacy.py
+++ /dev/null
@@ -1,93 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.format import format
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_format_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `format` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use Python type-structured generation with an OpenAI model"
-            + " due to the limitations of the OpenAI API.",
-        ):
-            format(openai_model, int)
-
-
-def test_format_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `format` function is deprecated",
-    ):
-        generator = format(transformers_vision_model, int)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        RegexLogitsProcessor
-    )
-
-
-def test_format_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `format` function is deprecated",
-    ):
-        generator = format(transformers_model, int, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        RegexLogitsProcessor
-    )
diff --git a/tests/v0_legacy/generate/test_fsm_legacy.py b/tests/v0_legacy/generate/test_fsm_legacy.py
deleted file mode 100644
index fe6e5ba2e..000000000
--- a/tests/v0_legacy/generate/test_fsm_legacy.py
+++ /dev/null
@@ -1,104 +0,0 @@
-from dataclasses import asdict
-
-import interegular
-import pytest
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import GuideLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.fsm import fsm
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_fsm_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `fsm` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use FSM-based generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API.",
-        ):
-            fsm(
-                openai_model,
-                interegular.parse_pattern(r"[a-z]+").to_fsm()
-            )
-
-
-def test_fsm_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `fsm` function is deprecated",
-    ):
-        generator = fsm(
-            transformers_vision_model,
-            interegular.parse_pattern(r"[a-z]+").to_fsm()
-        )
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        GuideLogitsProcessor
-    )
-
-
-def test_fsm_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `fsm` function is deprecated",
-    ):
-        generator = fsm(
-            transformers_model,
-            interegular.parse_pattern(r"[a-z]+").to_fsm(),
-            samplers.greedy()
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        GuideLogitsProcessor
-    )
diff --git a/tests/v0_legacy/generate/test_json_legacy.py b/tests/v0_legacy/generate/test_json_legacy.py
deleted file mode 100644
index d09d0347b..000000000
--- a/tests/v0_legacy/generate/test_json_legacy.py
+++ /dev/null
@@ -1,200 +0,0 @@
-import json as pyjson
-import pytest
-from dataclasses import asdict
-
-from pydantic import BaseModel
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.json import json
-
-
-JSON_SCHEMA_STRING = """
-{
-    "type": "object",
-    "properties": {
-        "foo": {
-            "type": "string"
-        }
-    },
-    "required": ["foo"]
-}
-"""
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_json_legacy_standard_callable(transformers_model):
-    def foo(bar: str) -> str:
-        return bar
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_model, foo, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_standard_json_string(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_model, JSON_SCHEMA_STRING)
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_standard_pydantic(transformers_model):
-    class Foo(BaseModel):
-        foo: str
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_model, Foo)
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"foo": "value"}')
-        == Foo.model_validate_json('{"foo": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_vision_model, JSON_SCHEMA_STRING)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_openai_wrong_sampler(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="The OpenAI API does not support any other sampling "
-            + "algorithm than the multinomial sampler.",
-        ):
-            json(openai_model, JSON_SCHEMA_STRING, samplers.greedy())
-
-
-def test_json_legacy_openai_pydantic(openai_model):
-    class Foo(BaseModel):
-        foo: str
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(openai_model, Foo)
-    assert isinstance(generator, models.OpenAI)
-    assert (
-        generator.format_sequence('{"foo": "value"}')
-        == Foo.model_validate_json('{"foo": "value"}')
-    )
-
-
-def test_json_legacy_openai_json_string(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(openai_model, JSON_SCHEMA_STRING)
-    assert isinstance(generator, models.OpenAI)
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-
-
-def test_json_legacy_openai_not_supported(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        with pytest.raises(ValueError, match="Cannot parse schema"):
-            json(openai_model, int)
diff --git a/tests/v0_legacy/generate/test_regex_legacy.py b/tests/v0_legacy/generate/test_regex_legacy.py
deleted file mode 100644
index 86da00c35..000000000
--- a/tests/v0_legacy/generate/test_regex_legacy.py
+++ /dev/null
@@ -1,90 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.regex import regex
-
-
-REGEX_STRING = r"^[a-z]+$"
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_regex_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `regex` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use regex-structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API.",
-        ):
-            regex(openai_model, REGEX_STRING)
-
-
-def test_regex_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `regex` function is deprecated",
-    ):
-        generator = regex(transformers_vision_model, REGEX_STRING)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
-
-
-def test_regex_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `regex` function is deprecated",
-    ):
-        generator = regex(transformers_model, REGEX_STRING, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_text_legacy.py b/tests/v0_legacy/generate/test_text_legacy.py
deleted file mode 100644
index e7412b295..000000000
--- a/tests/v0_legacy/generate/test_text_legacy.py
+++ /dev/null
@@ -1,83 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.text import text
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_text_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = text(openai_model)
-    assert generator == openai_model
-    assert isinstance(generator, models.OpenAI)
-
-
-def test_text_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = text(transformers_vision_model)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert generator.generator.logits_processor is None
-
-
-def test_text_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = text(transformers_model, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert generator.generator.logits_processor is None
diff --git a/tests/v0_legacy/models/test_exllamav2_legacy.py b/tests/v0_legacy/models/test_exllamav2_legacy.py
deleted file mode 100644
index 06fa9e17e..000000000
--- a/tests/v0_legacy/models/test_exllamav2_legacy.py
+++ /dev/null
@@ -1,76 +0,0 @@
-import warnings
-
-import pytest
-
-try:
-    import exllamav2  # type: ignore
-    HAS_EXV2 = True
-except ImportError:
-    HAS_EXV2 = False
-    warnings.warn(
-        "Skipping exllamav2 legacy tests because exllamav2 is not available."
-    )
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-pytestmark = pytest.mark.skipif(
-    not HAS_EXV2,
-    reason="Exllamav2 is not available."
-)
-
-
-@pytest.fixture(scope="session")
-def model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `exllamav2` function is deprecated",
-    ):
-        model = models.exl2(
-            model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-            max_seq_len=1024,
-            cache_q4=False,
-        )
-    return model
-
-
-def test_exllamav2_legacy_init(model):
-    assert isinstance(model, models.ExllamaV2)
-    assert hasattr(model, "generator")
-    assert hasattr(model, "tokenizer")
-    assert model.tokenizer.convert_token_to_string(1) == 1
-    assert hasattr(model, "max_seq_len")
-    assert isinstance(model.max_seq_len, int)
-
-
-def test_exllamav2_legacy_call(model):
-    # minimal
-    generator = generate.text(model)
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator("Respond with a single word.")
-    assert isinstance(result, str)
-
-    # with args
-    generator = generate.text(
-        model,
-        samplers.multinomial(samples=2, temperature=10, top_p=0.5, top_k=2)
-    )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(["Write a short story.", "Write a poem."], 2, "e", 2)
-    assert isinstance(result, list)
-    for sequence in result:
-        assert isinstance(sequence, list)
-        for s in sequence:
-            assert isinstance(s, str)
-            assert len(s) < 20
-
-    # stream
-    generator = generate.text(model)
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator.stream("Respond with a single word.", 2)
-    assert isinstance(result, GeneratorV0Adapter)
-    for r in result:
-        assert isinstance(r, str)
diff --git a/tests/v0_legacy/models/test_llamacpp_legacy.py b/tests/v0_legacy/models/test_llamacpp_legacy.py
deleted file mode 100644
index aead5bd86..000000000
--- a/tests/v0_legacy/models/test_llamacpp_legacy.py
+++ /dev/null
@@ -1,169 +0,0 @@
-import warnings
-from dataclasses import asdict
-
-import pytest
-from llama_cpp import Llama
-from llama_cpp.llama_tokenizer import LlamaHFTokenizer, LlamaTokenizer # type: ignore
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-
-def test_llamacpp_legacy_init():
-    # without tokenizer
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        with pytest.warns(
-            match="The pre-tokenizer in `llama.cpp` handles unicode improperly"
-        ):
-            model = models.llamacpp(
-                repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-                filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-            )
-    assert isinstance(model, models.LlamaCpp)
-    assert isinstance(model.model, Llama)
-    assert not model.model.verbose
-
-    # with tokenizer
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        model = models.llamacpp(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-            tokenizer=LlamaHFTokenizer.from_pretrained("distilbert-base-uncased"),
-            n_ctx=0,
-            verbose=True,
-        )
-    assert isinstance(model, models.LlamaCpp)
-    assert isinstance(model.model, Llama)
-    assert model.model.verbose
-    assert isinstance(model.model.tokenizer(), LlamaTokenizer)
-
-
-def test_llamacpp_legacy_params_adapter_generation_params():
-    result = llamacpp_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert not result.get("stop")
-    assert not result.get("seed")
-    assert result.get("max_tokens") == -1
-
-    result = llamacpp_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
-        {}
-    )
-    assert result.get("stop") == "foo"
-    assert result.get("seed") == 2
-    assert result.get("max_tokens") == 10
-
-
-def test_llamacpp_legacy_params_adapter_sampling_params():
-    with pytest.raises(
-        NotImplementedError,
-        match="does not support Beam Search"
-    ):
-        result = llamacpp_params_adapter(
-            asdict(samplers.beam_search().sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    with pytest.raises(
-        NotImplementedError,
-        match="does not allow to take several samples"
-    ):
-        result = llamacpp_params_adapter(
-            asdict(samplers.multinomial(samples=2).sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    result = llamacpp_params_adapter(
-        {"sampler": "multinomial", "num_samples": 1, "top_p": None, "top_k": None, "temperature": None},
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("top_p") == 1.0
-    assert result.get("top_k") == -1
-    assert result.get("temperature") == 1.0
-
-    result = llamacpp_params_adapter(
-        {"sampler": "greedy", "num_samples": 1, "top_p": 2, "top_k": 0, "temperature": 0.5},
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("top_p") == 2
-    assert result.get("top_k") == 0
-    assert result.get("temperature") == 0.5
-
-
-def test_llamacpp_legacy_params_adapter_model_specific_params():
-    result = llamacpp_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("min_p") == 0.0
-    assert result.get("repeat_penalty") == 1.0
-
-    result = llamacpp_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
-        {"min_p": 0.5, "repeat_penalty": 1.5, "stream": True}
-    )
-    assert result.get("min_p") == 0.5
-    assert result.get("repeat_penalty") == 1.5
-    assert not hasattr(result, "stream")
-
-
-def test_llamacpp_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        model = models.llamacpp(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                top_p=0.9,
-                top_k=10,
-                temperature=0.7
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!",
-        10,
-        "foo",
-        2,
-        min_p=0.5,
-        repeat_penalty=1.5,
-    )
-    assert isinstance(result, str)
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!",
-        10,
-        "foo",
-        2,
-        min_p=0.5,
-        repeat_penalty=1.5,
-    )
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_mlxlm_legacy.py b/tests/v0_legacy/models/test_mlxlm_legacy.py
deleted file mode 100644
index e70943ef2..000000000
--- a/tests/v0_legacy/models/test_mlxlm_legacy.py
+++ /dev/null
@@ -1,193 +0,0 @@
-import warnings
-from dataclasses import asdict
-
-import pytest
-
-try:
-    import mlx.core as mx
-    import mlx_lm  # noqa: F401
-
-    assert mx.metal.is_available()
-    HAS_MLX = True
-except ImportError:
-    HAS_MLX = False
-    warnings.warn("Skipping MLXLM legacy tests because MLX is not available.")
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
-
-pytestmark = pytest.mark.skipif(
-    not HAS_MLX,
-    reason="MLX tests require Apple Silicon."
-)
-
-
-def test_mlxlm_legacy_init():
-    # minimal
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mlxlm` function is deprecated",
-    ):
-        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
-    assert isinstance(model, models.MLXLM)
-    assert isinstance(model.model, mlx_lm.models.llama.Model)
-
-    # with optional arguments
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mlxlm` function is deprecated",
-    ):
-        model = models.mlxlm(
-            "mlx-community/SmolLM-135M-Instruct-4bit",
-            tokenizer_config={"max_length": 100},
-            model_config={"max_length": 100},
-            adapter_path=None,
-            lazy=False,
-        )
-    assert isinstance(model, models.MLXLM)
-    assert isinstance(model.model, mlx_lm.models.llama.Model)
-    assert isinstance(model.mlx_tokenizer, mlx_lm.tokenizer_utils.TokenizerWrapper)
-
-
-def test_mlxlm_legacy_params_adapter_generation_params():
-    # default
-    result = mlxlm_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("stop") is None
-    assert result.get("seed") is None
-    assert result.get("max_tokens") == int(1e9)
-
-    # max_tokens
-    result = mlxlm_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": 2},
-        {}
-    )
-    assert result.get("max_tokens") == 2
-
-    # stop_at
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support `stop_at`."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial().sampling_params),
-            {"stop_at": "foo", "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # seed
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support `seed`."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial().sampling_params),
-            {"stop_at": None, "seed": 2, "max_tokens": None},
-            {}
-        )
-
-
-def test_mlxlm_legacy_params_adapter_sampling_params():
-    # beam search
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support Beam Search."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.beam_search().sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # num_samples
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not allow to take several samples."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial(samples=2).sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # top_k
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support top_k."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial(top_k=2).sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # temperature
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `temperature` parameter is not available",
-    ):
-        result = mlxlm_params_adapter(
-            {
-                "sampler": "multinomial",
-                "num_samples": 1,
-                "top_p": None,
-                "top_k": None,
-                "temperature": 0.5
-            },
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-        assert result.get("temperature") is None
-
-    # top_p
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `top_p` parameter is not available",
-    ):
-        result = mlxlm_params_adapter(
-            {
-                "sampler": "multinomial",
-                "num_samples": 1,
-                "top_p": 2,
-                "top_k": None,
-                "temperature": None
-            },
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-        assert result.get("top_p") is None
-
-
-def test_mlxlm_legacy_params_adapter_model_specific_params():
-    result = mlxlm_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {"length_penalty": 0.5}
-    )
-    assert result.get("length_penalty") == 0.5
-
-
-def test_mlxlm_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mlxlm` function is deprecated",
-    ):
-        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
-
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial()
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator("Hello, world!", 10)
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_openai_legacy.py b/tests/v0_legacy/models/test_openai_legacy.py
deleted file mode 100644
index c07443dab..000000000
--- a/tests/v0_legacy/models/test_openai_legacy.py
+++ /dev/null
@@ -1,159 +0,0 @@
-import os
-
-import pytest
-from openai import AsyncAzureOpenAI, AsyncOpenAI
-
-from outlines import models, generate
-from outlines.models.openai import OpenAI, OpenAIConfig, OpenAILegacy
-
-
-# Set the OPENAI_API_KEY environment variable to "MOCK_VALUE" if it is not set
-# not to raise an error when the OpenAI client is initialized
-api_key = os.getenv("OPENAI_API_KEY")
-if not api_key:
-    os.environ["OPENAI_API_KEY"] = "MOCK_VALUE"
-azure_api_key = os.getenv("AZURE_OPENAI_API_KEY")
-if not azure_api_key:
-    os.environ["AZURE_OPENAI_API_KEY"] = "MOCK_VALUE"
-
-
-def test_openai_legacy_init():
-    # through the openai function with a client and no config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        client = AsyncOpenAI()
-        model = models.openai(client, system_prompt="You are a helpful assistant.")
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.config == OpenAIConfig()
-    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # through the openai function with a client and config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            client,
-            OpenAIConfig(stop=["."]),
-            system_prompt="You are a helpful assistant.",
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.config == OpenAIConfig(stop=["."])
-    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # through the openai function with model name and no config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            "gpt-4o",
-            max_retries=1,
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.client.max_retries == 1
-    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o")
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # through the openai function with model name and config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            "gpt-4o",
-            OpenAIConfig(stop=["."]),
-            max_retries=1,
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.client.max_retries == 1
-    assert model.legacy_instance.config == OpenAIConfig(
-        model="gpt-4o", stop=["."]
-    )
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # directly through the OpenAI class with client, config and system prompt
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.OpenAI(
-            AsyncOpenAI(),
-            OpenAIConfig(model="gpt-4o", n=1),
-            system_prompt="You are a helpful assistant.",
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o", n=1)
-    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-
-def test_azure_openai_legacy_init():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.azure_openai(
-            "foo",
-            max_retries=1,
-            base_url="https://foo.openai.azure.com/",
-            api_version="2023-03-15-preview",
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncAzureOpenAI)
-    assert model.legacy_instance.client.max_retries == 1
-    assert model.legacy_instance.config == OpenAIConfig(model="foo")
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-
-@pytest.mark.api_call
-def test_openai_legacy_call_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            "gpt-4o",
-            max_retries=1,
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(model)
-    result = generator(
-        "Hello, world!",
-        10,
-        ["."],
-        system_prompt="You are a helpful assistant.",
-        temperature=0.5,
-        samples=1
-    )
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_legacy.py b/tests/v0_legacy/models/test_transformers_legacy.py
deleted file mode 100644
index 2c34c2517..000000000
--- a/tests/v0_legacy/models/test_transformers_legacy.py
+++ /dev/null
@@ -1,172 +0,0 @@
-from dataclasses import asdict
-
-import pytest
-from transformers import (
-    GPT2LMHeadModel,
-    GPT2TokenizerFast,
-    BertForSequenceClassification,
-    BertTokenizer,
-    MambaForCausalLM,
-)
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.models.transformers import transformers_params_adapter
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-
-def test_transformers_legacy_init():
-    # Test with default model and tokenizer classes
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="erwanf/gpt2-mini",
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, GPT2LMHeadModel)
-    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
-
-    # Test with non default model and tokenizer classes
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="bert-base-uncased",
-            model_class=BertForSequenceClassification,
-            tokenizer_class=BertTokenizer,
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, BertForSequenceClassification)
-    assert isinstance(model.tokenizer.tokenizer, BertTokenizer)
-
-    # Test with model and tokenizer kwargs and device
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="erwanf/gpt2-mini",
-            model_kwargs={"use_cache": False},
-            tokenizer_kwargs={"model_max_length": 512},
-            device="cpu",
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, GPT2LMHeadModel)
-    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
-    assert model.tokenizer.tokenizer.model_max_length == 512
-    assert not model.model.config.use_cache
-    assert model.model.device.type == "cpu"
-
-
-def test_mamba_legacy_init():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mamba` function is deprecated",
-    ):
-        model = models.mamba(
-            model_name="erwanf/gpt2-mini",
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, MambaForCausalLM)
-    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
-
-
-def test_transformers_legacy_params_adapter_generation_params():
-    result = transformers_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert not result.get("stop")
-    assert not result.get("seed")
-    assert result.get("max_new_tokens") == 2**30
-
-    result = transformers_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
-        {}
-    )
-    assert result.get("stop_strings") == "foo"
-    assert not result.get("seed")
-    assert result.get("max_new_tokens") == 10
-
-
-def test_transformers_legacy_params_adapter_sampling_params():
-    result = transformers_params_adapter(
-        {
-            "sampler": "multinomial",
-            "num_samples": None,
-            "top_p": None,
-            "top_k": None,
-            "temperature": None,
-        },
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {},
-    )
-    assert result.get("do_sample")
-    assert result.get("num_return_sequences") == 1
-    assert result.get("top_p") is None
-    assert result.get("top_k") is None
-    assert result.get("temperature") is None
-
-    result = transformers_params_adapter(
-        {
-            "sampler": "beam_search",
-            "num_samples": 2,
-            "top_p": 2,
-            "top_k": 0,
-            "temperature": 0.5,
-        },
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {},
-    )
-    assert result.get("num_beams") == 2
-    assert result.get("num_return_sequences") == 2
-    assert result.get("top_p") == 2
-    assert result.get("top_k") == 0
-    assert result.get("temperature") == 0.5
-
-
-def test_transformers_legacy_params_adapter_model_specific_params():
-    result = transformers_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {"length_penalty": 1.0}
-    )
-    assert result.get("length_penalty") == 1
-    assert not result.get("foo")
-
-
-def test_transformers_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="erwanf/gpt2-mini",
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                samples=1,
-                top_p=0.9,
-                top_k=10,
-                temperature=0.7
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!",
-        10,
-        "foo",
-        2,
-        length_penalty=0.5,
-    )
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_vision_legacy.py b/tests/v0_legacy/models/test_transformers_vision_legacy.py
deleted file mode 100644
index bcaf4ea3e..000000000
--- a/tests/v0_legacy/models/test_transformers_vision_legacy.py
+++ /dev/null
@@ -1,110 +0,0 @@
-from io import BytesIO
-from urllib.request import urlopen
-
-import pytest
-from PIL import Image
-from transformers import (
-    AutoProcessor,
-    LlavaForConditionalGeneration,
-    LlamaTokenizer,
-    LlamaTokenizerFast,
-    LlavaProcessor,
-)
-
-from outlines import models, generate, samplers
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-
-def test_transformers_vision_legacy_init():
-    # Test with only model name and model class
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-        )
-    assert isinstance(model, models.TransformersVision)
-    assert isinstance(model.model, LlavaForConditionalGeneration)
-    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
-    assert isinstance(model.processor, LlavaProcessor)
-
-    # Test with model, tokenizer, and processor classes
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-            tokenizer_class=LlamaTokenizer,
-            processor_class=AutoProcessor,
-        )
-    assert isinstance(model, models.TransformersVision)
-    assert isinstance(model.model, LlavaForConditionalGeneration)
-    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizer)
-    assert isinstance(model.processor, LlavaProcessor)
-
-    # Test with model, tokenizer, processor kwargs, and device
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-            model_kwargs={"output_attentions": True},
-            processor_kwargs={"num_additional_image_tokens": 1},
-            device="cpu",
-        )
-    assert isinstance(model, models.TransformersVision)
-    assert isinstance(model.model, LlavaForConditionalGeneration)
-    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
-    assert isinstance(model.processor, LlavaProcessor)
-    assert model.model.config.output_attentions
-    assert model.processor.num_additional_image_tokens == 1
-    assert model.model.device.type == "cpu"
-
-
-def test_transformers_vision_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                samples=1,
-                top_p=0.9,
-                top_k=10,
-                temperature=0.7
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!<image>",
-        img_from_url(
-            "https://upload.wikimedia.org/wikipedia/commons"
-            + "/2/25/Siam_lilacpoint.jpg"
-        ),
-        10,
-        "foo",
-        2,
-        length_penalty=0.5,
-    )
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_vllm_offline_legacy.py b/tests/v0_legacy/models/test_vllm_offline_legacy.py
deleted file mode 100644
index 318eeddb7..000000000
--- a/tests/v0_legacy/models/test_vllm_offline_legacy.py
+++ /dev/null
@@ -1,181 +0,0 @@
-import warnings
-from dataclasses import asdict
-
-import pytest
-
-try:
-    from vllm import LLM
-    HAS_VLLM = True
-except ImportError:
-    HAS_VLLM = False
-    warnings.warn("Skipping vLLM legacy tests because vLLM is not available.")
-
-from outlines import generate, models, samplers
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
-
-pytestmark = pytest.mark.skipif(
-    not HAS_VLLM,
-    reason="vLLM models can only be run on GPU."
-)
-
-
-@pytest.fixture
-def model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `vllm` function is deprecated",
-    ):
-        model_instance = models.vllm(
-            "erwanf/gpt2-mini",
-            seed=1,
-        )
-    return model_instance
-
-
-def test_vllm_legacy_init(model):
-    assert isinstance(model, models.VLLM)
-    assert isinstance(model.model, LLM)
-
-
-def test_vllm_legacy_params_adapter_generation_params():
-    # default
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].stop_strings is None
-    assert result["sampling_params"].seed is None
-    assert result["sampling_params"].max_tokens is None
-
-    # max_tokens
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": 2},
-        {}
-    )
-    assert result["sampling_params"].max_tokens == 2
-
-    # seed
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": 2, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].seed == 2
-
-    # stop_at str
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].stop_strings == ["foo"]
-
-    # stop_at list
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": ["foo", "bar"], "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].stop_strings == ["foo", "bar"]
-
-
-def test_vllm_legacy_params_adapter_sampling_params():
-    # default
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].n == 1
-    assert result["sampling_params"].top_p == 1.0
-    assert result["sampling_params"].top_k == -1
-    assert result["sampling_params"].temperature == 1.0
-
-    # beam search
-    result = vllm_offline_params_adapter(
-        asdict(samplers.beam_search().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].use_beam_search is True
-
-    # num_samples
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(samples=2).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].n == 2
-    assert result["sampling_params"].best_of == 2
-
-    # top_p
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(top_p=0.5).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].top_p == 0.5
-
-    # top_k
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(top_k=2).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].top_k == 2
-
-    # temperature
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(temperature=0.5).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].temperature == 0.5
-
-
-def test_vllm_legacy_params_adapter_model_specific_params():
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {"use_tqdm": False}
-    )
-    assert not result.get("use_tqdm")
-
-
-def test_vllm_legacy_call_generation(model):
-    # default
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial()
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator("Hello, world!")
-    assert isinstance(result, str)
-
-    # with args
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                samples=2, top_p=0.5, top_k=2, temperature=0.5
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!", 10, stop_at="foo", seed=1, use_tqdm=False
-    )
-    assert isinstance(result, list)
-    for completion in result:
-        assert isinstance(completion, str)
diff --git a/tests/v0_legacy/test_samplers_legacy.py b/tests/v0_legacy/test_samplers_legacy.py
deleted file mode 100644
index a4d750116..000000000
--- a/tests/v0_legacy/test_samplers_legacy.py
+++ /dev/null
@@ -1,60 +0,0 @@
-from outlines.v0_legacy.samplers import (
-    BeamSearchSampler,
-    GreedySampler,
-    MultinomialSampler,
-    beam_search,
-    greedy,
-    multinomial,
-)
-
-
-def test_aliases():
-    assert greedy == GreedySampler
-    assert multinomial == MultinomialSampler
-    assert beam_search == BeamSearchSampler
-
-
-def test_greedy():
-    sampler = GreedySampler()
-    params = sampler.sampling_params
-    assert params.sampler == "greedy"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 0.0
-
-
-def test_multinomial():
-    sampler = MultinomialSampler()
-    params = sampler.sampling_params
-    assert params.sampler == "multinomial"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature is None
-
-    sampler = multinomial(samples=5, top_k=10, top_p=0.9, temperature=0.8)
-    params = sampler.sampling_params
-    assert params.sampler == "multinomial"
-    assert params.num_samples == 5
-    assert params.top_p == 0.9
-    assert params.top_k == 10
-    assert params.temperature == 0.8
-
-
-def test_beam_search():
-    sampler = BeamSearchSampler()
-    params = sampler.sampling_params
-    assert params.sampler == "beam_search"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 1.0
-
-    sampler = beam_search(beams=3)
-    params = sampler.sampling_params
-    assert params.sampler == "beam_search"
-    assert params.num_samples == 3
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 1.0
diff --git a/uv.lock b/uv.lock
index 2d927f136..f1deefbe6 100644
--- a/uv.lock
+++ b/uv.lock
@@ -625,6 +625,28 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/02/cc/b7e31358aac6ed1ef2bb790a9746ac2c69bcb3c8588b41616914eb106eaf/exceptiongroup-1.2.2-py3-none-any.whl", hash = "sha256:3111b9d131c238bec2f8f516e123e14ba243563fb135d3fe885990585aa7795b", size = 16453 },
 ]
 
+[[package]]
+name = "exllamav2"
+version = "0.2.7"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "fastparquet" },
+    { name = "ninja" },
+    { name = "numpy" },
+    { name = "pandas" },
+    { name = "pillow" },
+    { name = "pygments" },
+    { name = "regex" },
+    { name = "rich" },
+    { name = "safetensors" },
+    { name = "sentencepiece" },
+    { name = "torch" },
+    { name = "websockets" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b5/04/300012908311c9a904d44c74a037821e7cde530c648afff4ad6bed4e22c3/exllamav2-0.2.7-py3-none-any.whl", hash = "sha256:410d1a69fed329f6f1c10de6988a3b6b33e1fa715aa4fe32c545edd1abd1254b", size = 1512625 },
+]
+
 [[package]]
 name = "fastapi"
 version = "0.115.6"
@@ -1895,6 +1917,9 @@ dependencies = [
 ]
 
 [package.optional-dependencies]
+exllamav2 = [
+    { name = "exllamav2" },
+]
 llamacpp = [
     { name = "datasets" },
     { name = "llama-cpp-python" },
@@ -1921,6 +1946,7 @@ test = [
     { name = "datasets" },
     { name = "diff-cover", version = "9.2.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.9.17'" },
     { name = "diff-cover", version = "9.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.9.17'" },
+    { name = "exllamav2" },
     { name = "huggingface-hub" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
@@ -1963,6 +1989,8 @@ requires-dist = [
     { name = "datasets", marker = "extra == 'transformers'" },
     { name = "diff-cover", marker = "extra == 'test'" },
     { name = "diskcache" },
+    { name = "exllamav2", marker = "extra == 'exllamav2'" },
+    { name = "exllamav2", marker = "extra == 'test'" },
     { name = "fastapi", marker = "extra == 'serve'" },
     { name = "genson" },
     { name = "huggingface-hub", marker = "extra == 'test'" },

From 0bf58b5c61c10bdfd1d4dca5d1d4f14598258f27 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 09:46:06 +0100
Subject: [PATCH 194/505] Add `Vision` type for prompt + image input to vision
 models

---
 outlines/templates.py | 27 +++++++++++++++++++++++++++
 pyproject.toml        |  1 +
 2 files changed, 28 insertions(+)

diff --git a/outlines/templates.py b/outlines/templates.py
index 16588e813..325bcfe04 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -1,3 +1,4 @@
+import base64
 import functools
 import inspect
 import json
@@ -5,12 +6,38 @@
 import re
 import textwrap
 from dataclasses import dataclass
+from io import BytesIO
 from pathlib import Path
 from typing import Any, Callable, Dict, Optional, Type, cast
 import warnings
 
 import jinja2
+from jinja2 import Environment, StrictUndefined
+from PIL import Image
 import pydantic
+from pydantic import BaseModel
+
+
+@dataclass
+class Vision:
+    """Represents the input for a vision model."""
+
+    prompt: str
+    image: Image.Image
+
+    def __post_init__(self):
+        image = self.image
+        buffer = BytesIO()
+        image.save(buffer, format=image.format)
+        self.image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
+
+        image_format = image.format
+        if image_format is not None:
+            self.image_format = f"image/{image_format.lower()}"
+        else:
+            raise TypeError(
+                "Could not read the format of the image passed to the model."
+            )
 
 
 @dataclass
diff --git a/pyproject.toml b/pyproject.toml
index dfed39973..35078a6fa 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -35,6 +35,7 @@ dependencies = [
    "referencing",
    "jsonschema",
    "requests",
+   "pillow",
    "tqdm",
    "typing_extensions",
    "iso3166",

From ef754cd03735b40f94c381aa5b61a7b7c46d8570 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 09:47:10 +0100
Subject: [PATCH 195/505] Add `Json` output type

---
 outlines/types/__init__.py | 30 ++++++++++++++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index d7f40309a..18d41384d 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,4 +1,9 @@
+from dataclasses import dataclass
 from enum import Enum
+import json
+from typing import Union
+
+from pydantic import BaseModel
 
 from . import airports, countries, locale
 from outlines.types.dsl import (
@@ -15,6 +20,7 @@
     zero_or_more,
 )
 
+
 # Python types
 integer = Regex(r"[+-]?(0|[1-9][0-9]*)")
 boolean = Regex("(True|False)")
@@ -49,3 +55,27 @@
 isbn = Regex(
     r"(?:ISBN(?:-1[03])?:? )?(?=[0-9X]{10}$|(?=(?:[0-9]+[- ]){3})[- 0-9X]{13}$|97[89][0-9]{10}$|(?=(?:[0-9]+[- ]){4})[- 0-9]{17}$)(?:97[89][- ]?)?[0-9]{1,5}[- ]?[0-9]+[- ]?[0-9]+[- ]?[0-9X]"
 )
+
+
+@dataclass
+class Json:
+    """Represents a JSON object.
+
+    The structure of JSON object can be defined using a JSON Schema
+    specification represented as a string or a dictionary, or a Pydantic
+    BaseModel.
+
+    Attributes
+    ----------
+    definition
+        The object used to define the structure of the JSON object.
+
+    """
+
+    def __init__(self, definition: Union[str, dict, BaseModel]):
+        if isinstance(definition, type(BaseModel)):
+            definition = definition.model_json_schema()
+        if isinstance(definition, str):
+            definition = json.loads(definition)
+
+        self.definition = definition

From 0f39894af590afd3a8883a48d4efdcd959bb6eea Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 09:48:09 +0100
Subject: [PATCH 196/505] Refactor the `OpenAI` interface

---
 outlines/models/__init__.py |   2 +-
 outlines/models/openai.py   | 402 ++++++++++++------------------------
 pyproject.toml              |   2 +-
 tests/models/test_openai.py | 226 ++++++++++----------
 4 files changed, 234 insertions(+), 398 deletions(-)

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index fe6f861ac..ef1a883b9 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -11,7 +11,7 @@
 from .exllamav2 import ExLlamaV2Model, exl2
 from .llamacpp import LlamaCpp, llamacpp
 from .mlxlm import MLXLM, mlxlm
-from .openai import OpenAI, azure_openai, openai
+from .openai import AzureOpenAI, OpenAI
 from .transformers import Transformers, TransformerTokenizer, mamba, transformers
 from .transformers_vision import TransformersVision, transformers_vision
 from .vllm import VLLM, vllm
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index e46107bd4..03ea0b5ed 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,303 +1,157 @@
 """Integration with OpenAI's API."""
 
-import copy
-import functools
-from dataclasses import asdict, dataclass, field, replace
-from typing import Callable, Dict, List, Optional, Tuple, Union
-
-import numpy as np
-
-from outlines.base import vectorize
-from outlines.caching import cache
-
-__all__ = ["OpenAI", "openai", "azure_openai"]
-
-
-@dataclass(frozen=True)
-class OpenAIConfig:
-    """Represents the parameters of the OpenAI API.
-
-    The information was last fetched on 2023/11/20. We document below the
-    properties that are specific to the OpenAI API. Not all these properties are
-    supported by Outlines.
-
-    Parameters
-    ----------
-    model
-        The name of the model. Available models can be found on OpenAI's website.
-    frequency_penalty
-        Number between 2.0 and -2.0. Positive values penalize new tokens based on
-        their existing frequency in the text,
-    logit_bias
-        Modifies the likelihood of specified tokens to appear in the completion.
-        Number between -100 (forbid) and +100 (only allows).
-    n
-        The number of completions to return for each prompt.
-    presence_penalty
-        Similar to frequency penalty.
-    response_format
-        Specifies the format the model must output. `{"type": "json_object"}`
-        enables JSON mode.
-    seed
-        Two completions with the same `seed` value should return the same
-        completion. This is however not guaranteed.
-    stop
-        Up to 4 words where the API will stop the completion.
-    temperature
-        Number between 0 and 2. Higher values make the output more random, while
-        lower values make it more deterministic.
-    top_p
-        Number between 0 and 1. Parameter for nucleus sampling.
-    user
-        A unique identifier for the end-user.
-    """
+from functools import singledispatchmethod
+from types import NoneType
+from typing import Optional, Union
 
-    model: str = ""
-    frequency_penalty: float = 0
-    logit_bias: Dict[int, int] = field(default_factory=dict)
-    max_tokens: Optional[int] = None
-    n: int = 1
-    presence_penalty: float = 0
-    response_format: Optional[Dict[str, str]] = None
-    seed: Optional[int] = None
-    stop: Optional[Union[str, List[str]]] = None
-    temperature: float = 1.0
-    top_p: int = 1
-    user: str = field(default_factory=str)
+from pydantic import BaseModel
 
+from outlines.prompts import Vision
+from outlines.types import Json
 
-class OpenAI:
-    """An object that represents the OpenAI API."""
+__all__ = ["OpenAI"]
 
-    def __init__(
-        self,
-        client,
-        config,
-        system_prompt: Optional[str] = None,
-    ):
-        """Create an `OpenAI` instance.
-
-        This class supports the standard OpenAI API, the Azure OpeanAI API as
-        well as compatible APIs that rely on the OpenAI client.
-
-        Parameters
-        ----------
-        client
-            An instance of the API's async client.
-        config
-            An instance of `OpenAIConfig`. Can be useful to specify some
-            parameters that cannot be set by calling this class' methods.
-        """
 
-        self.client = client
-        self.config = config
+class OpenAIBase:
+    """Base class for the OpenAI clients.
 
-        # We count the total number of prompt and generated tokens as returned
-        # by the OpenAI API, summed over all the requests performed with this
-        # model instance.
-        self.prompt_tokens = 0
-        self.completion_tokens = 0
+    `OpenAI` base is responsible for preparing the arguments to OpenAI's
+    `completions.create` methods: the input (prompt and possibly image), as well
+    as the output type (only JSON).
 
-        self.format_sequence = lambda seq: seq
+    """
 
-    def __call__(
-        self,
-        prompt: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[List[str], str]] = None,
-        *,
-        system_prompt: Optional[str] = None,
-        temperature: Optional[float] = None,
-        samples: Optional[int] = None,
-    ) -> np.ndarray:
-        """Call the OpenAI API to generate text.
-
-        Parameters
-        ----------
-        prompt
-            A string or list of strings that will be used to prompt the model
-        max_tokens
-            The maximum number of tokens to generate
-        stop_at
-            A string or array of strings which, such that the generation stops
-            when they are generated.
-        system_prompt
-            The content of the system message that precedes the user's prompt.
-        temperature
-            The value of the temperature used to sample tokens
-        samples
-            The number of completions to generate for each prompt
-        stop_at
-            Up to 4 words where the API will stop the completion.
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the `messages` argument to pass to the client.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
 
         """
-        if max_tokens is None:
-            max_tokens = self.config.max_tokens
-        if stop_at is None:
-            stop_at = self.config.stop
-        if temperature is None:
-            temperature = self.config.temperature
-        if samples is None:
-            samples = self.config.n
-
-        config = replace(
-            self.config,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            n=samples,
-            stop=stop_at,
-        )  # type: ignore
-
-        response, prompt_tokens, completion_tokens = generate_chat(
-            prompt, system_prompt, self.client, config
+        raise NotImplementedError(
+            f"The input type {input} is not available with OpenAI. The only available types are `str` and `Vision`."
         )
-        self.prompt_tokens += prompt_tokens
-        self.completion_tokens += completion_tokens
 
-        return self.format_sequence(response)
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        """Generate the `messages` argument to pass to the client when the user
+        only passes a prompt.
 
-    def stream(self, *args, **kwargs):
+        """
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": model_input,
+                }
+            ]
+        }
+
+    @format_input.register(Vision)
+    def format_vision_input(self, model_input: Vision):
+        """Generate the `messages` argument to pass to the client when the user
+        passes a prompt and an image.
+
+        """
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": model_input.prompt},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:{model_input.image_format};base64,{model_input.image_str}"
+                            },
+                        },
+                    ],
+                }
+            ]
+        }
+
+    @singledispatchmethod
+    def format_output_type(self, output_type):
+        """Generate the `response_format` argument to the client based on the
+        output type specified by the user.
+
+        """
         raise NotImplementedError(
-            "Streaming is currently not supported for the OpenAI API"
+            f"The type {output_type} is not available with OpenAI. The only output type available is `Json`."
         )
 
-    def new_with_replacements(self, **kwargs):
-        new_instance = copy.copy(self)
-        new_instance.config = replace(new_instance.config, **kwargs)
-        return new_instance
-
-    def __str__(self):
-        return self.__class__.__name__ + " API"
-
-    def __repr__(self):
-        return str(self.config)
-
-
-@functools.partial(vectorize, signature="(),(),(),()->(s),(),()")
-async def generate_chat(
-    prompt: str,
-    system_prompt: Union[str, None],
-    client,
-    config: OpenAIConfig,
-) -> Tuple[np.ndarray, int, int]:
-    """Call OpenAI's Chat Completion API.
-
-    Parameters
-    ----------
-    prompt
-        The prompt we use to start the generation. Passed to the model
-        with the "user" role.
-    system_prompt
-        The system prompt, passed to the model with the "system" role
-        before the prompt.
-    client
-        The API client
-    config
-        An `OpenAIConfig` instance.
-
-    Returns
-    -------
-    A tuple that contains the model's response(s) and usage statistics.
+    @format_output_type.register(NoneType)
+    def format_none_output_type(self, _: None):
+        """Generate the `response_format` argument to the client when no
+        output type is specified by the user.
 
-    """
+        """
+        return {}
 
-    @error_handler
-    @cache()
-    async def call_api(prompt, system_prompt, config):
-        responses = await client.chat.completions.create(
-            messages=system_message + user_message,
-            **asdict(config),  # type: ignore
-        )
-        return responses.model_dump()
-
-    system_message = (
-        [{"role": "system", "content": system_prompt}] if system_prompt else []
-    )
-    user_message = [{"role": "user", "content": prompt}]
-
-    responses = await call_api(prompt, system_prompt, config)
-
-    results = np.array(
-        [responses["choices"][i]["message"]["content"] for i in range(config.n)]
-    )
-    usage = responses["usage"]
-
-    return results, usage["prompt_tokens"], usage["completion_tokens"]
-
-
-def error_handler(api_call_fn: Callable) -> Callable:
-    """Handle OpenAI API errors and missing API key."""
-
-    def call(*args, **kwargs):
-        import openai
-
-        try:
-            return api_call_fn(*args, **kwargs)
-        except (
-            openai.APITimeoutError,
-            openai.InternalServerError,
-            openai.RateLimitError,
-        ) as e:
-            raise OSError(f"Could not connect to the OpenAI API: {e}")
-        except (
-            openai.AuthenticationError,
-            openai.BadRequestError,
-            openai.ConflictError,
-            openai.PermissionDeniedError,
-            openai.NotFoundError,
-            openai.UnprocessableEntityError,
-        ) as e:
-            raise e
-
-    return call
-
-
-@functools.singledispatch
-def openai(model_or_client, *args, **kwargs):
-    return OpenAI(model_or_client, *args, **kwargs)
-
-
-@openai.register(str)
-def openai_model(
-    model_name: str,
-    config: Optional[OpenAIConfig] = None,
-    **openai_client_params,
-):
-    try:
-        from openai import AsyncOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
-        )
+    @format_output_type.register(Json)
+    def format_json_output_type(self, output_type: Json):
+        """Generate the `response_format` argument to the client when the user
+        specified a `Json` output type.
+
+        """
+        # OpenAI requires `additionalProperties` to be set
+        if "additionalProperties" not in output_type.definition:
+            output_type.definition["additionalProperties"] = False
 
-    if config is not None:
-        config = replace(config, model=model_name)  # type: ignore
-    else:
-        config = OpenAIConfig(model=model_name)
+        return {
+            "response_format": {
+                "type": "json_schema",
+                "json_schema": {
+                    "name": "default",
+                    "strict": True,
+                    "schema": output_type.definition,
+                },
+            }
+        }
 
-    client = AsyncOpenAI(**openai_client_params)
 
-    return OpenAI(client, config)
+class OpenAI(OpenAIBase):
+    """Thin wrapper around the `openai.OpenAI` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client.
+
+    """
+
+    def __init__(self, model_name: str, *args, **kwargs):
+        from openai import OpenAI
 
+        self.client = OpenAI(*args, **kwargs)
+        self.model_name = model_name
 
-def azure_openai(
-    deployment_name: str,
-    model_name: Optional[str] = None,
-    config: Optional[OpenAIConfig] = None,
-    **azure_openai_client_params,
-):
-    try:
-        from openai import AsyncAzureOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Union[type[BaseModel], str]] = None,
+        **inference_kwargs,
+    ):
+        messages = self.format_input(model_input)
+        response_format = self.format_output_type(output_type)
+        result = self.client.chat.completions.create(
+            model=self.model_name, **messages, **response_format, **inference_kwargs
         )
 
-    if config is not None:
-        config = replace(config, model=deployment_name)  # type: ignore
-    if config is None:
-        config = OpenAIConfig(model=deployment_name)
+        return result.choices[0].message.content
+
+
+class AzureOpenAI(OpenAI):
+    """Thin wrapper around the `openai.AzureOpenAI` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.AzureOpenAI` client.
+
+    """
 
-    client = AsyncAzureOpenAI(**azure_openai_client_params)
+    def __init__(self, deployment_name: str, *args, **kwargs):
+        from openai import AzureOpenAI
 
-    return OpenAI(client, config)
+        self.client = AzureOpenAI(*args, **kwargs)
+        self.model_name = deployment_name
diff --git a/pyproject.toml b/pyproject.toml
index 35078a6fa..76b8fce2f 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -123,7 +123,6 @@ module = [
     "jax.numpy",
     "jinja2",
     "jsonschema.*",
-    "openai.*",
     "mamba_ssm.*",
     "mlx_lm.*",
     "mlx.*",
@@ -142,6 +141,7 @@ module = [
     "lark.*",
     "interegular.*",
     "datasets.*",
+    "openai.*",
     "requests.*",
     "responses.*",
     "vllm.*",
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index f4b97f36a..0a6873251 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,125 +1,107 @@
-import importlib
+import io
 import json
-from contextlib import contextmanager
-from unittest import mock
-from unittest.mock import MagicMock
 
+import PIL
 import pytest
-from openai import AsyncOpenAI
-
-from outlines import generate
-from outlines.models.openai import OpenAI, OpenAIConfig
-
-
-def module_patch(path):
-    """Patch functions that have the same name as the module in which they're implemented."""
-    target = path
-    components = target.split(".")
-    for i in range(len(components), 0, -1):
-        try:
-            # attempt to import the module
-            imported = importlib.import_module(".".join(components[:i]))
-
-            # module was imported, let's use it in the patch
-            patch = mock.patch(path)
-            patch.getter = lambda: imported
-            patch.attribute = ".".join(components[i:])
-            return patch
-        except Exception:
-            continue
-
-    # did not find a module, just return the default mock
-    return mock.patch(path)
-
-
-def test_openai_call():
-    with module_patch("outlines.models.openai.generate_chat") as mocked_generate_chat:
-        mocked_generate_chat.return_value = ["foo"], 1, 2
-        async_client = MagicMock(spec=AsyncOpenAI, api_key="key")
-
-        model = OpenAI(
-            async_client,
-            OpenAIConfig(max_tokens=10, temperature=0.5, n=2, stop=["."]),
-        )
-
-        assert model("bar")[0] == "foo"
-        assert model.prompt_tokens == 1
-        assert model.completion_tokens == 2
-        mocked_generate_chat_args = mocked_generate_chat.call_args
-        mocked_generate_chat_arg_config = mocked_generate_chat_args[0][3]
-        assert isinstance(mocked_generate_chat_arg_config, OpenAIConfig)
-        assert mocked_generate_chat_arg_config.max_tokens == 10
-        assert mocked_generate_chat_arg_config.temperature == 0.5
-        assert mocked_generate_chat_arg_config.n == 2
-        assert mocked_generate_chat_arg_config.stop == ["."]
-
-        model("bar", samples=3)
-        mocked_generate_chat_args = mocked_generate_chat.call_args
-        mocked_generate_chat_arg_config = mocked_generate_chat_args[0][3]
-        assert mocked_generate_chat_arg_config.n == 3
-
-
-@contextmanager
-def patched_openai(completion, **oai_config):
-    """Create a patched openai whose chat completions always returns `completion`"""
-    with module_patch("outlines.models.openai.generate_chat") as mocked_generate_chat:
-        mocked_generate_chat.return_value = completion, 1, 2
-        async_client = MagicMock(spec=AsyncOpenAI, api_key="key")
-        model = OpenAI(
-            async_client,
-            OpenAIConfig(max_tokens=10, temperature=0.5, n=2, stop=["."]),
-        )
-        yield model
-
-
-def test_openai_choice_call():
-    with patched_openai(completion='{"result": "foo"}') as model:
-        generator = generate.choice(model, ["foo", "bar"])
-        assert generator("hi") == "foo"
-
-
-def test_openai_choice_call_invalid_server_response():
-    with patched_openai(completion="not actual json") as model:
-        generator = generate.choice(model, ["foo", "bar"])
-        with pytest.raises(json.decoder.JSONDecodeError):
-            generator("hi")
-
-
-def test_openai_json_call_pydantic():
-    from pydantic import BaseModel, ConfigDict, ValidationError
-
-    class Person(BaseModel):
-        model_config = ConfigDict(extra="forbid")  # required for openai
-        first_name: str
-        last_name: str
-        age: int
-
-    completion = '{"first_name": "Usain", "last_name": "Bolt", "age": 38}'
-
-    # assert success for valid response
-    with patched_openai(completion=completion) as model:
-        generator = generate.json(model, Person)
-        assert generator("fastest person") == Person.parse_raw(completion)
-
-    # assert fail for non-json response
-    with patched_openai(completion="usain bolt") as model:
-        generator = generate.json(model, Person)
-        with pytest.raises(ValidationError):
-            assert generator("fastest person")
-
-
-def test_openai_json_call_str():
-    person_schema = '{"additionalProperties": false, "properties": {"first_name": {"title": "First Name", "type": "string"}, "last_name": {"title": "Last Name", "type": "string"}, "age": {"title": "Age", "type": "integer"}}, "required": ["first_name", "last_name", "age"], "title": "Person", "type": "object"}'
-
-    output = {"first_name": "Usain", "last_name": "Bolt", "age": 38}
-
-    # assert success for valid response
-    with patched_openai(completion=json.dumps(output)) as model:
-        generator = generate.json(model, person_schema)
-        assert generator("fastest person") == output
-
-    # assert fail for non-json response
-    with patched_openai(completion="usain bolt") as model:
-        generator = generate.json(model, person_schema)
-        with pytest.raises(json.decoder.JSONDecodeError):
-            assert generator("fastest person")
+import requests
+from pydantic import BaseModel
+
+from outlines.models.openai import OpenAI
+from outlines.prompts import Vision
+from outlines.types import Json
+
+MODEL_NAME = "gpt-4o-mini-2024-07-18"
+
+
+def test_openai_wrong_init_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        OpenAI(MODEL_NAME, foo=10)
+
+
+def test_openai_wrong_inference_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model = OpenAI(MODEL_NAME)
+        model.generate("prompt", foo=10)
+
+
+def test_openai_wrong_input_type():
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model = OpenAI(MODEL_NAME)
+        model.generate(Foo("prompt"))
+
+
+def test_openai_wrong_output_type():
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model = OpenAI(MODEL_NAME)
+        model.generate("prompt", Foo(1))
+
+
+@pytest.mark.api_call
+def test_openai_simple_call():
+    model = OpenAI(MODEL_NAME)
+    result = model.generate("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_openai_simple_vision():
+    model = OpenAI(MODEL_NAME)
+
+    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
+    r = requests.get(url, stream=True)
+    if r.status_code == 200:
+        image = PIL.Image.open(io.BytesIO(r.content))
+
+    result = model.generate(Vision("What does this logo represent?", image))
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_openai_simple_pydantic():
+    model = OpenAI(MODEL_NAME)
+
+    class Foo(BaseModel):
+        bar: int
+
+    result = model.generate("foo?", Json(Foo))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_openai_simple_vision_pydantic():
+    model = OpenAI(MODEL_NAME)
+
+    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
+    r = requests.get(url, stream=True)
+    if r.status_code == 200:
+        image = PIL.Image.open(io.BytesIO(r.content))
+
+    class Logo(BaseModel):
+        name: int
+
+    result = model.generate(Vision("What does this logo represent?", image), Json(Logo))
+    assert isinstance(result, str)
+    assert "name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_openai_simple_json_schema():
+    model = OpenAI(MODEL_NAME)
+
+    class Foo(BaseModel):
+        bar: int
+
+    schema = json.dumps(Foo.model_json_schema())
+
+    result = model.generate("foo?", Json(schema))
+    assert isinstance(result, str)
+    json.loads(result)

From eacc0a92ec602fd30f0128c5ddd446d340f920ee Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Tue, 5 Nov 2024 17:27:10 +0100
Subject: [PATCH 197/505] Update tests/models/test_openai.py

Co-authored-by: Victoria Terenina <torymur@gmail.com>
---
 tests/models/test_openai.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 0a6873251..a3d8a176b 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -104,4 +104,4 @@ class Foo(BaseModel):
 
     result = model.generate("foo?", Json(schema))
     assert isinstance(result, str)
-    json.loads(result)
+    assert "bar" in json.loads(result)

From e0451192ab43c2d4f17b37025dac5698da3d7f76 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 08:47:50 +0100
Subject: [PATCH 198/505] Add Anthropic integration

---
 outlines/models/__init__.py    |  1 +
 outlines/models/anthropic.py   | 88 ++++++++++++++++++++++++++++++++++
 pyproject.toml                 |  2 +
 tests/models/test_anthropic.py | 63 ++++++++++++++++++++++++
 4 files changed, 154 insertions(+)
 create mode 100644 outlines/models/anthropic.py
 create mode 100644 tests/models/test_anthropic.py

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index ef1a883b9..4813742f4 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -8,6 +8,7 @@
 
 from typing import Union
 
+from .anthropic import Anthropic
 from .exllamav2 import ExLlamaV2Model, exl2
 from .llamacpp import LlamaCpp, llamacpp
 from .mlxlm import MLXLM, mlxlm
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
new file mode 100644
index 000000000..e55b2e3e2
--- /dev/null
+++ b/outlines/models/anthropic.py
@@ -0,0 +1,88 @@
+"""Integration with Anthropic's API."""
+from functools import singledispatchmethod
+from typing import Union
+
+from outlines.prompts import Vision
+
+__all__ = ["Anthropic"]
+
+
+class AnthropicBase:
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the `messages` argument to pass to the client.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available with Anthropic. The only available types are `str` and `Vision`."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        """Generate the `messages` argument to pass to the client when the user
+        only passes a prompt.
+
+        """
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": model_input,
+                }
+            ]
+        }
+
+    @format_input.register(Vision)
+    def format_vision_input(self, model_input: Vision):
+        """Generate the `messages` argument to pass to the client when the user
+        passes a prompt and an image.
+
+        """
+        return {
+            "messages": [
+                {
+                    "role": "user",
+                    "content": [
+                        {
+                            "type": "image",
+                            "source": {
+                                "type": "base64",
+                                "media_type": model_input.image_format,
+                                "data": model_input.image_str,
+                            },
+                        },
+                        {"type": "text", "text": model_input.prompt},
+                    ],
+                }
+            ]
+        }
+
+
+class Anthropic(AnthropicBase):
+    def __init__(self, model_name: str, *args, **kwargs):
+        from anthropic import Anthropic
+
+        self.client = Anthropic(*args, **kwargs)
+        self.model_name = model_name
+
+    def generate(
+        self, model_input: Union[str, Vision], output_type=None, **inference_kwargs
+    ):
+        messages = self.format_input(model_input)
+
+        if output_type is not None:
+            raise NotImplementedError(
+                f"The type {output_type} is not available with Anthropic."
+            )
+
+        completion = self.client.messages.create(
+            **messages,
+            model=self.model_name,
+            **inference_kwargs,
+        )
+        return completion.content[0].text
diff --git a/pyproject.toml b/pyproject.toml
index 76b8fce2f..17c7a468c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -69,6 +69,7 @@ test = [
     "huggingface_hub",
     "openai>=1.0.0",
     "datasets",
+    "anthropic",
     "transformers",
     "pillow",
     "exllamav2",
@@ -123,6 +124,7 @@ module = [
     "jax.numpy",
     "jinja2",
     "jsonschema.*",
+    "anthropic.*",
     "mamba_ssm.*",
     "mlx_lm.*",
     "mlx.*",
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
new file mode 100644
index 000000000..44a176084
--- /dev/null
+++ b/tests/models/test_anthropic.py
@@ -0,0 +1,63 @@
+import io
+
+import PIL
+import pytest
+import requests
+
+from outlines.models.anthropic import Anthropic
+from outlines.prompts import Vision
+
+MODEL_NAME = "claude-3-haiku-20240307"
+
+
+def test_anthropic_wrong_init_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        Anthropic(MODEL_NAME, foo=10)
+
+
+def test_anthropic_wrong_inference_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model = Anthropic(MODEL_NAME)
+        model.generate("prompt", foo=10, max_tokens=1024)
+
+
+def test_anthropic_wrong_input_type():
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model = Anthropic(MODEL_NAME)
+        model.generate(Foo("prompt"))
+
+
+def test_anthropic_wrong_output_type():
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model = Anthropic(MODEL_NAME)
+        model.generate("prompt", Foo(1))
+
+
+@pytest.mark.api_call
+def test_anthropic_simple_call():
+    model = Anthropic(MODEL_NAME)
+    result = model.generate("Respond with one word. Not more.", max_tokens=1024)
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_anthropic_simple_vision():
+    model = Anthropic(MODEL_NAME)
+
+    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
+    r = requests.get(url, stream=True)
+    if r.status_code == 200:
+        image = PIL.Image.open(io.BytesIO(r.content))
+
+    result = model.generate(
+        Vision("What does this logo represent?", image), max_tokens=1024
+    )
+    assert isinstance(result, str)

From 3a74d6e94e4adc73817bcdac2c9b3d2454ff77c3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 13:39:13 +0100
Subject: [PATCH 199/505] Add `Choice` output type

---
 outlines/types/__init__.py | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 18d41384d..3b0fa984d 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,6 +1,6 @@
 from dataclasses import dataclass
-from enum import Enum
 import json
+from enum import Enum, EnumMeta
 from typing import Union
 
 from pydantic import BaseModel
@@ -79,3 +79,14 @@ def __init__(self, definition: Union[str, dict, BaseModel]):
             definition = json.loads(definition)
 
         self.definition = definition
+
+
+@dataclass
+class Choice:
+    """Represents a multiple choice"""
+
+    definition: Union[EnumMeta, list[str]]
+
+    def __post_init__(self):
+        if isinstance(self.definition, list):
+            self.definition = Enum("Definition", [(x, x) for x in self.definition])

From 11080cbe6fd69f0dc04aa241f475267b52b53eed Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 08:54:21 +0100
Subject: [PATCH 200/505] Add Gemini integration

---
 outlines/models/__init__.py |   1 +
 outlines/models/gemini.py   | 121 +++++++++++++++++++++++
 outlines/types/__init__.py  |   2 +
 pyproject.toml              |   2 +
 tests/models/test_gemini.py | 188 ++++++++++++++++++++++++++++++++++++
 5 files changed, 314 insertions(+)
 create mode 100644 outlines/models/gemini.py
 create mode 100644 tests/models/test_gemini.py

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 4813742f4..d51e5e483 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -10,6 +10,7 @@
 
 from .anthropic import Anthropic
 from .exllamav2 import ExLlamaV2Model, exl2
+from .gemini import Gemini
 from .llamacpp import LlamaCpp, llamacpp
 from .mlxlm import MLXLM, mlxlm
 from .openai import AzureOpenAI, OpenAI
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
new file mode 100644
index 000000000..7614861c2
--- /dev/null
+++ b/outlines/models/gemini.py
@@ -0,0 +1,121 @@
+"""Integration with Gemini's API."""
+from enum import EnumMeta
+from functools import singledispatchmethod
+from types import NoneType
+from typing import Optional, Union
+
+from pydantic import BaseModel
+from typing_extensions import _TypedDictMeta  # type: ignore
+
+from outlines.prompts import Vision
+from outlines.types import Choice, Json
+
+__all__ = ["Gemini"]
+
+
+class GeminiBase:
+    """Base class for the Gemini clients.
+
+    `GeminiBase` is responsible for preparing the arguments to Gemini's
+    `generate_contents` methods: the input (prompt and possibly image), as well
+    as the output type (only JSON).
+
+    """
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the `messages` argument to pass to the client.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available with Gemini. The only available types are `str` and `Vision`."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        """Generate the `messages` argument to pass to the client when the user
+        only passes a prompt.
+
+        """
+        return {"contents": [model_input]}
+
+    @format_input.register(Vision)
+    def format_vision_input(self, model_input: Vision):
+        """Generate the `messages` argument to pass to the client when the user
+        passes a prompt and an image.
+
+        """
+        return {"contents": [model_input.prompt, model_input.image]}
+
+    @singledispatchmethod
+    def format_output_type(self, output_type):
+        if output_type.__origin__ == list:
+            if len(output_type.__args__) == 1 and isinstance(
+                output_type.__args__[0], Json
+            ):
+                return {
+                    "response_mime_type": "application/json",
+                    "response_schema": list[
+                        output_type.__args__[0].original_definition
+                    ],
+                }
+            else:
+                raise TypeError
+        else:
+            raise NotImplementedError
+
+    @format_output_type.register(NoneType)
+    def format_none_output_type(self, output_type):
+        return {}
+
+    @format_output_type.register(Json)
+    def format_json_output_type(self, output_type):
+        if issubclass(output_type.original_definition, BaseModel):
+            return {
+                "response_mime_type": "application/json",
+                "response_schema": output_type.original_definition,
+            }
+        elif isinstance(output_type.original_definition, _TypedDictMeta):
+            return {
+                "response_mime_type": "application/json",
+                "response_schema": output_type.original_definition,
+            }
+        else:
+            raise NotImplementedError
+
+    @format_output_type.register(Choice)
+    def format_enum_output_type(self, output_type):
+        return {
+            "response_mime_type": "text/x.enum",
+            "response_schema": output_type.definition,
+        }
+
+
+class Gemini(GeminiBase):
+    def __init__(self, model_name: str, *args, **kwargs):
+        import google.generativeai as genai
+
+        self.client = genai.GenerativeModel(model_name, *args, **kwargs)
+
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Union[Json, EnumMeta]] = None,
+        **inference_kwargs,
+    ):
+        import google.generativeai as genai
+
+        contents = self.format_input(model_input)
+        generation_config = genai.GenerationConfig(
+            **self.format_output_type(output_type)
+        )
+        completion = self.client.generate_content(
+            generation_config=generation_config, **contents, **inference_kwargs
+        )
+
+        return completion.text
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 3b0fa984d..5e669c2a0 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -73,6 +73,8 @@ class Json:
     """
 
     def __init__(self, definition: Union[str, dict, BaseModel]):
+        self.original_definition = definition
+
         if isinstance(definition, type(BaseModel)):
             definition = definition.model_json_schema()
         if isinstance(definition, str):
diff --git a/pyproject.toml b/pyproject.toml
index 17c7a468c..d73a93b48 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -70,6 +70,7 @@ test = [
     "openai>=1.0.0",
     "datasets",
     "anthropic",
+    "google-generativeai",
     "transformers",
     "pillow",
     "exllamav2",
@@ -125,6 +126,7 @@ module = [
     "jinja2",
     "jsonschema.*",
     "anthropic.*",
+    "google.*",
     "mamba_ssm.*",
     "mlx_lm.*",
     "mlx.*",
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
new file mode 100644
index 000000000..1f30d9bdb
--- /dev/null
+++ b/tests/models/test_gemini.py
@@ -0,0 +1,188 @@
+import io
+import json
+from enum import Enum
+from typing import List
+
+import PIL
+import pytest
+import requests
+from pydantic import BaseModel
+from typing_extensions import TypedDict
+
+from outlines.models.gemini import Gemini
+from outlines.prompts import Vision
+from outlines.types import Choice, Json
+
+MODEL_NAME = "gemini-1.5-flash-latest"
+
+
+def test_gemini_wrong_init_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        Gemini(MODEL_NAME, foo=10)
+
+
+def test_gemini_wrong_inference_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model = Gemini(MODEL_NAME)
+        model.generate("prompt", foo=10)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_call():
+    model = Gemini(MODEL_NAME)
+    result = model.generate("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_vision():
+    model = Gemini(MODEL_NAME)
+
+    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
+    r = requests.get(url, stream=True)
+    if r.status_code == 200:
+        image = PIL.Image.open(io.BytesIO(r.content))
+
+    result = model.generate(Vision("What does this logo represent?", image))
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_pydantic():
+    model = Gemini(MODEL_NAME)
+
+    class Foo(BaseModel):
+        bar: int
+
+    result = model.generate("foo?", Json(Foo))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.xfail(reason="Vision models do not work with structured outputs.")
+@pytest.mark.api_call
+def test_gemini_simple_vision_pydantic():
+    model = Gemini(MODEL_NAME)
+
+    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
+    r = requests.get(url, stream=True)
+    if r.status_code == 200:
+        image = PIL.Image.open(io.BytesIO(r.content))
+
+    class Logo(BaseModel):
+        name: int
+
+    result = model.generate(Vision("What does this logo represent?", image), Logo)
+    assert isinstance(result, str)
+    assert "name" in json.loads(result)
+
+
+@pytest.mark.xfail(reason="Gemini seems to be unable to follow nested schemas.")
+@pytest.mark.api_call
+def test_gemini_nested_pydantic():
+    model = Gemini(MODEL_NAME)
+
+    class Bar(BaseModel):
+        fu: str
+
+    class Foo(BaseModel):
+        sna: int
+        bar: Bar
+
+    result = model.generate("foo?", Json(Foo))
+    assert isinstance(result, str)
+    assert "sna" in json.loads(result)
+    assert "bar" in json.loads(result)
+    assert "fu" in json.loads(result)["bar"]
+
+
+@pytest.mark.xfail(
+    reason="The Gemini SDK's serialization method does not support Json Schema dictionaries."
+)
+@pytest.mark.api_call
+def test_gemini_simple_json_schema_dict():
+    model = Gemini(MODEL_NAME)
+
+    schema = {
+        "properties": {"bar": {"title": "Bar", "type": "integer"}},
+        "required": ["bar"],
+        "title": "Foo",
+        "type": "object",
+    }
+    result = model.generate("foo?", Json(schema))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.xfail(
+    reason="The Gemini SDK's serialization method does not support Json Schema strings."
+)
+@pytest.mark.api_call
+def test_gemini_simple_json_schema_string():
+    model = Gemini(MODEL_NAME)
+
+    schema = "{'properties': {'bar': {'title': 'Bar', 'type': 'integer'}}, 'required': ['bar'], 'title': 'Foo', 'type': 'object'}"
+    result = model.generate("foo?", Json(schema))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_typed_dict():
+    model = Gemini(MODEL_NAME)
+
+    class Foo(TypedDict):
+        bar: int
+
+    result = model.generate("foo?", Json(Foo))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_gemini_simple_choice_enum():
+    model = Gemini(MODEL_NAME)
+
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    result = model.generate("foo?", Choice(Foo))
+    assert isinstance(result, str)
+    assert result == "Foo" or result == "Bar"
+
+
+@pytest.mark.api_call
+def test_gemini_simple_choice_list():
+    model = Gemini(MODEL_NAME)
+
+    choices = ["Foo", "Bar"]
+    result = model.generate("foo?", Choice(choices))
+    assert isinstance(result, str)
+    assert result == "Foo" or result == "Bar"
+
+
+@pytest.mark.api_call
+def test_gemini_simple_list_pydantic():
+    model = Gemini(MODEL_NAME)
+
+    class Foo(BaseModel):
+        bar: int
+
+    result = model.generate("foo?", list[Json(Foo)])
+    assert isinstance(json.loads(result), list)
+    assert isinstance(json.loads(result)[0], dict)
+    assert "bar" in json.loads(result)[0]
+
+
+@pytest.mark.api_call
+def test_gemini_simple_list_annotation_pydantic():
+    model = Gemini(MODEL_NAME)
+
+    class Foo(BaseModel):
+        bar: int
+
+    result = model.generate("foo?", List[Json(Foo)])
+    assert isinstance(json.loads(result), list)
+    assert isinstance(json.loads(result)[0], dict)
+    assert "bar" in json.loads(result)[0]

From 44357d59fbe384a46bd6c6cb9f4e9f1f989ad403 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Fri, 22 Nov 2024 14:18:17 +0100
Subject: [PATCH 201/505] Add `Json` method to convert to Json Schema

---
 outlines/models/gemini.py  |  9 +++++----
 outlines/models/openai.py  |  8 +++++---
 outlines/types/__init__.py | 36 ++++++++++++++++++++++++++----------
 3 files changed, 36 insertions(+), 17 deletions(-)

diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 7614861c2..a11369a2d 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -75,15 +75,16 @@ def format_none_output_type(self, output_type):
 
     @format_output_type.register(Json)
     def format_json_output_type(self, output_type):
-        if issubclass(output_type.original_definition, BaseModel):
+        """Gemini only accepts Pydantic models and TypeDicts to define the JSON structure."""
+        if issubclass(output_type.definition, BaseModel):
             return {
                 "response_mime_type": "application/json",
-                "response_schema": output_type.original_definition,
+                "response_schema": output_type.definition,
             }
-        elif isinstance(output_type.original_definition, _TypedDictMeta):
+        elif isinstance(output_type.definition, _TypedDictMeta):
             return {
                 "response_mime_type": "application/json",
-                "response_schema": output_type.original_definition,
+                "response_schema": output_type.definition,
             }
         else:
             raise NotImplementedError
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 03ea0b5ed..cc97f5f2c 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -97,9 +97,11 @@ def format_json_output_type(self, output_type: Json):
         specified a `Json` output type.
 
         """
+        schema = output_type.to_json_schema()
+
         # OpenAI requires `additionalProperties` to be set
-        if "additionalProperties" not in output_type.definition:
-            output_type.definition["additionalProperties"] = False
+        if "additionalProperties" not in schema:
+            schema["additionalProperties"] = False
 
         return {
             "response_format": {
@@ -107,7 +109,7 @@ def format_json_output_type(self, output_type: Json):
                 "json_schema": {
                     "name": "default",
                     "strict": True,
-                    "schema": output_type.definition,
+                    "schema": schema,
                 },
             }
         }
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 5e669c2a0..c81e09603 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -3,7 +3,10 @@
 from enum import Enum, EnumMeta
 from typing import Union
 
-from pydantic import BaseModel
+from jsonschema import Draft202012Validator as Validator
+from jsonschema.exceptions import SchemaError
+from pydantic import BaseModel, TypeAdapter
+from typing_extensions import _TypedDictMeta  # type: ignore
 
 from . import airports, countries, locale
 from outlines.types.dsl import (
@@ -72,15 +75,28 @@ class Json:
 
     """
 
-    def __init__(self, definition: Union[str, dict, BaseModel]):
-        self.original_definition = definition
-
-        if isinstance(definition, type(BaseModel)):
-            definition = definition.model_json_schema()
-        if isinstance(definition, str):
-            definition = json.loads(definition)
-
-        self.definition = definition
+    definition: Union[str, dict]
+
+    def to_json_schema(self):
+        if isinstance(self.definition, str):
+            schema = json.loads(self.definition)
+        elif isinstance(self.definition, dict):
+            schema = self.definition
+        elif isinstance(self.definition, type(BaseModel)):
+            schema = self.definition.model_json_schema()
+        elif isinstance(self.definition, _TypedDictMeta):
+            schema = TypeAdapter(self.definition).json_schema()
+        else:
+            raise TypeError(
+                "The Json definition must be a JSON Schema string, dictionary or Pydantic model."
+            )
+
+        try:
+            Validator.check_schema(schema)
+        except SchemaError as e:
+            raise e
+
+        return schema
 
 
 @dataclass

From 224174dcf305b24d01094e4278aa4941a1283866 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Fri, 22 Nov 2024 16:25:07 +0100
Subject: [PATCH 202/505] Add `List` type

---
 outlines/models/gemini.py   | 24 +++++++++---------------
 outlines/types/__init__.py  |  5 +++++
 tests/models/test_gemini.py | 18 ++----------------
 3 files changed, 16 insertions(+), 31 deletions(-)

diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index a11369a2d..1d6bbfd1e 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -8,7 +8,7 @@
 from typing_extensions import _TypedDictMeta  # type: ignore
 
 from outlines.prompts import Vision
-from outlines.types import Choice, Json
+from outlines.types import Choice, Json, List
 
 __all__ = ["Gemini"]
 
@@ -54,20 +54,14 @@ def format_vision_input(self, model_input: Vision):
 
     @singledispatchmethod
     def format_output_type(self, output_type):
-        if output_type.__origin__ == list:
-            if len(output_type.__args__) == 1 and isinstance(
-                output_type.__args__[0], Json
-            ):
-                return {
-                    "response_mime_type": "application/json",
-                    "response_schema": list[
-                        output_type.__args__[0].original_definition
-                    ],
-                }
-            else:
-                raise TypeError
-        else:
-            raise NotImplementedError
+        raise NotImplementedError
+
+    @format_output_type.register(List)
+    def format_list_output_type(self, output_type):
+        return {
+            "response_mime_type": "application/json",
+            "response_schema": list[output_type.definition.definition],
+        }
 
     @format_output_type.register(NoneType)
     def format_none_output_type(self, output_type):
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index c81e09603..d81d6f2e2 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -99,6 +99,11 @@ def to_json_schema(self):
         return schema
 
 
+@dataclass
+class List:
+    definition: list
+
+
 @dataclass
 class Choice:
     """Represents a multiple choice"""
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 1f30d9bdb..4718b12d0 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -1,7 +1,6 @@
 import io
 import json
 from enum import Enum
-from typing import List
 
 import PIL
 import pytest
@@ -11,7 +10,7 @@
 
 from outlines.models.gemini import Gemini
 from outlines.prompts import Vision
-from outlines.types import Choice, Json
+from outlines.types import Choice, Json, List
 
 MODEL_NAME = "gemini-1.5-flash-latest"
 
@@ -169,20 +168,7 @@ def test_gemini_simple_list_pydantic():
     class Foo(BaseModel):
         bar: int
 
-    result = model.generate("foo?", list[Json(Foo)])
-    assert isinstance(json.loads(result), list)
-    assert isinstance(json.loads(result)[0], dict)
-    assert "bar" in json.loads(result)[0]
-
-
-@pytest.mark.api_call
-def test_gemini_simple_list_annotation_pydantic():
-    model = Gemini(MODEL_NAME)
-
-    class Foo(BaseModel):
-        bar: int
-
-    result = model.generate("foo?", List[Json(Foo)])
+    result = model.generate("foo?", List(Json(Foo)))
     assert isinstance(json.loads(result), list)
     assert isinstance(json.loads(result)[0], dict)
     assert "bar" in json.loads(result)[0]

From ac836ea9df88614770b7d86e645cdbcd774e4c0e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 29 Nov 2024 11:47:27 +0100
Subject: [PATCH 203/505] Refactor the llama.cpp interface

---
 outlines/models/__init__.py                 |   2 +-
 outlines/models/llamacpp.py                 | 319 ++++-------------
 tests/generate/test_integration_llamacpp.py | 374 --------------------
 tests/models/test_llamacpp.py               | 118 ++++++
 4 files changed, 183 insertions(+), 630 deletions(-)
 delete mode 100644 tests/generate/test_integration_llamacpp.py
 create mode 100644 tests/models/test_llamacpp.py

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index d51e5e483..7bfb09fb0 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -11,7 +11,7 @@
 from .anthropic import Anthropic
 from .exllamav2 import ExLlamaV2Model, exl2
 from .gemini import Gemini
-from .llamacpp import LlamaCpp, llamacpp
+from .llamacpp import LlamaCpp
 from .mlxlm import MLXLM, mlxlm
 from .openai import AzureOpenAI, OpenAI
 from .transformers import Transformers, TransformerTokenizer, mamba, transformers
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 7c7ad64df..0e597c5a7 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,25 +1,11 @@
-import dataclasses
 import pickle
 import warnings
-from typing import (
-    TYPE_CHECKING,
-    Dict,
-    Iterator,
-    List,
-    Optional,
-    Set,
-    Tuple,
-    TypedDict,
-    Union,
-)
-
-from typing_extensions import Unpack
-
-from outlines.generate.api import GenerationParameters, SamplingParameters
+from typing import TYPE_CHECKING, Dict, Iterator, List, Set, Tuple, Union
+
 from outlines.models.tokenizer import Tokenizer
 
 if TYPE_CHECKING:
-    from llama_cpp import Llama, LogitsProcessorList
+    from llama_cpp import Llama
 
 
 class LlamaCppTokenizer(Tokenizer):
@@ -107,185 +93,77 @@ def __setstate__(self, state):
         raise NotImplementedError("Cannot load a pickled llamacpp tokenizer")
 
 
-class LlamaCppParams(TypedDict, total=False):
-    suffix: Optional[str]
-    temperature: float
-    top_p: float
-    min_p: float
-    typical_p: float
-    seed: int
-    max_tokens: int
-    logits_processor: "LogitsProcessorList"
-    stop: Optional[Union[str, List[str]]]
-    frequence_penalty: float
-    presence_penalty: float
-    repeat_penalty: float
-    top_k: int
-    tfs_z: float
-    mirostat_mode: int
-    mirostat_tau: float
-    mirostat_eta: float
-    stream: bool
-
-
 class LlamaCpp:
-    """Represents a model provided by the `llama-cpp-python` library.
-
-    We wrap models from model providing libraries in order to give all of
-    them the same interface in Outlines and allow users to easily switch
-    between providers. This class wraps the `llama_cpp.Llama` class from the
-    `llama-cpp-python` library.
-
-    """
-
-    def __init__(self, model: "Llama"):
-        self.model = model
+    """Wraps a model provided by the `llama-cpp-python` library."""
 
-    @property
-    def tokenizer(self):
-        return LlamaCppTokenizer(self.model)
+    def __init__(self, model_path: Union[str, "Llama"], **kwargs):
+        from llama_cpp import Llama
 
-    def prepare_generation_parameters(
-        self,
-        generation_parameters: GenerationParameters,
-        sampling_parameters: SamplingParameters,
-        structure_logits_processor,
-        **llama_cpp_params: Unpack[LlamaCppParams],
-    ):
-        """Prepare the generation parameters.
+        if isinstance(model_path, Llama):
+            self.model = model_path
+        else:
+            # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
+            if "tokenizer" not in kwargs:
+                warnings.warn(
+                    "The pre-tokenizer in `llama.cpp` handles unicode improperly "
+                    + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
+                    + "Outlines may raise a `RuntimeError` when building the regex index.\n"
+                    + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
+                    + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
+                )
 
-        `llama-cpp-python` uses different default values
+            self.model = Llama(model_path, **kwargs)
 
-        """
-        from llama_cpp import LogitsProcessorList
+        self.tokenizer = LlamaCppTokenizer(self.model)
 
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-
-        # We update `llama_cpp_params` with the values the user passed to the
-        # generator.
-        if "stop" not in llama_cpp_params:
-            llama_cpp_params["stop"] = stop_at
-        if "seed" not in llama_cpp_params:
-            llama_cpp_params["seed"] = seed
-
-        # Somehow `llama-cpp-python` generates `max_tokens + 1`  tokens
-        if "max_tokens" not in llama_cpp_params:
-            if max_tokens is None:
-                llama_cpp_params["max_tokens"] = -1  # indicates unlimited tokens
-            else:
-                llama_cpp_params["max_tokens"] = max_tokens - 1
-        else:
-            llama_cpp_params["max_tokens"] = llama_cpp_params["max_tokens"] - 1
-
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
-        )
+    @classmethod
+    def from_pretrained(cls, repo_id, filename, **kwargs):
+        """Download the model weights from Hugging Face and create a `Llama` instance"""
+        from llama_cpp import Llama
 
-        # We update the `llama_cpp_params` with the sampling values that
-        # were specified by the user via the `Sampler` class, unless they
-        # are also specified in `llama_cpp_params`. We also disable other
-        # sampling methods that are enabled by default and reset the temperature
-        # value.
-        #
-        # See https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
-        # for the default values in `llama.cpp` and indications to disable the sampling modes.
-        # Mirostat sampling, tail-free sampling and all penalties are disabled by default.
-        #
-        # See https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
-        # for default values in `llama-cpp-python`
-        if sampler == "beam_search":
-            raise NotImplementedError(
-                "The `llama_cpp_python` library does not support Beam Search."
-            )
-        if num_samples != 1:
-            raise NotImplementedError(
-                "The `llama_cpp_python` library does not allow to take several samples."
+        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
+        if "tokenizer" not in kwargs:
+            warnings.warn(
+                "The pre-tokenizer in `llama.cpp` handles unicode improperly "
+                + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
+                + "Outlines may raise a `RuntimeError` when building the regex index.\n"
+                + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
+                + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
             )
-        if "top_p" not in llama_cpp_params:
-            if top_p is not None:
-                llama_cpp_params["top_p"] = top_p
-            else:
-                llama_cpp_params["top_p"] = 1.0
-
-        if "min_p" not in llama_cpp_params:
-            llama_cpp_params["min_p"] = 0.0
-
-        if "top_k" not in llama_cpp_params:
-            if top_k is not None:
-                llama_cpp_params["top_k"] = top_k
-            else:
-                llama_cpp_params["top_k"] = -1
-
-        if "temperature" not in llama_cpp_params:
-            if temperature is not None:
-                llama_cpp_params["temperature"] = temperature
-            else:
-                llama_cpp_params["temperature"] = 1.0
-
-        if "repeat_penalty" not in llama_cpp_params:
-            llama_cpp_params["repeat_penalty"] = 1.0
-
-        # The choice to stream or not should happen via the high-level API
-        llama_cpp_params["stream"] = False
-
-        if structure_logits_processor is not None:
-            if "logits_processor" in llama_cpp_params:
-                llama_cpp_params["logits_processor"].append(structure_logits_processor)
-            else:
-                llama_cpp_params["logits_processor"] = LogitsProcessorList(
-                    [structure_logits_processor]
-                )
 
-        return llama_cpp_params
+            model = Llama.from_pretrained(repo_id, filename, **kwargs)
+            return cls(model)
 
-    def generate(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **llama_cpp_params: Unpack[LlamaCppParams],
-    ) -> str:
+    def generate(self, prompt: str, logits_processor, **inference_kwargs) -> str:
         """Generate text using `llama-cpp-python`.
 
-        Parameters
-        ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        Arguments
+        ---------
+        prompt
+            A prompt.
         logits_processor
             The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-        llama_cpp_params
-            Keyword arguments that can be passed to
-            `llama_cpp_python.Llama.__call__`.  The values in `llama_cpp_params`
-            supersede the values of the parameters in `generation_parameters` and
-            `sampling_parameters`.  See the `llama_cpp_python` documentation for
-            a list of possible values: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
+        inference_kwargs
+            The inference kwargs that can be passed to the `Llama.__call__` method
+            in the `llama-cpp-python` library.
 
         Returns
         -------
         The generated text.
 
         """
-        if not isinstance(prompts, str):
+        from llama_cpp import LogitsProcessorList
+
+        if not isinstance(prompt, str):
             raise NotImplementedError(
                 "The `llama-cpp-python` library does not support batch inference."
             )
 
-        llama_cpp_params = self.prepare_generation_parameters(
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
-            **llama_cpp_params,
+        completion = self.model(
+            prompt,
+            logits_processor=LogitsProcessorList([logits_processor]),
+            **inference_kwargs,
         )
-        completion = self.model(prompts, **llama_cpp_params)
         result = completion["choices"][0]["text"]
 
         self.model.reset()
@@ -293,55 +171,38 @@ def generate(
         return result
 
     def stream(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **llama_cpp_params: Unpack[LlamaCppParams],
+        self, prompt: str, logits_processor, **inference_kwargs
     ) -> Iterator[str]:
         """Stream text using `llama-cpp-python`.
 
-        Parameters
-        ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        Arguments
+        ---------
+        prompt
+            A prompt.
         logits_processor
             The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-        llama_cpp_params
-            Keyword arguments that can be passed to
-            `llama_cpp_python.Llama.__call__`.  The values in `llama_cpp_params`
-            supersede the values of the parameters in `generation_parameters` and
-            `sampling_parameters`.  See the `llama_cpp_python` documentation for
-            a list of possible values: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
+        inference_kwargs
+            The inference kwargs that can be passed to the `Llama.__call__` method
+            in the `llama-cpp-python` library.
 
         Returns
         -------
         A generator that return strings.
 
         """
+        from llama_cpp import LogitsProcessorList
 
-        if not isinstance(prompts, str):
+        if not isinstance(prompt, str):
             raise NotImplementedError(
                 "The `llama-cpp-python` library does not support batch inference."
             )
 
-        llama_cpp_params = self.prepare_generation_parameters(
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
-            **llama_cpp_params,
+        generator = self.model(
+            prompt,
+            logits_processor=LogitsProcessorList([logits_processor]),
+            stream=True,
+            **inference_kwargs,
         )
-        llama_cpp_params["stream"] = True
-        generator = self.model(prompts, **llama_cpp_params)
 
         def token_generator() -> Iterator[str]:
             while True:
@@ -353,55 +214,3 @@ def token_generator() -> Iterator[str]:
                     return
 
         return token_generator()
-
-    def load_lora(self, adapter_path: str):
-        if self.model._model.apply_lora_from_file(
-            adapter_path,
-            1.0,
-        ):
-            raise RuntimeError(f"Failed to apply LoRA from lora path: {adapter_path}")
-
-
-def llamacpp(
-    repo_id: str, filename: Optional[str] = None, **llamacpp_model_params
-) -> LlamaCpp:
-    """Load a model from the `llama-cpp-python` library.
-
-    We use the `Llama.from_pretrained` classmethod that downloads models
-    directly from the HuggingFace hub, instead of asking users to specify
-    a path to the downloaded model. One can still load a local model
-    by initializing `llama_cpp.Llama` directly.
-
-    Parameters
-    ----------
-    repo_id
-        The name of the model repository.
-    filename:
-        A filename of glob pattern to match the model file in the repo.
-    llama_cpp_model_params
-        Llama-specific model parameters. See the `llama-cpp-python` documentation
-        for the full list: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__
-
-    """
-    from llama_cpp import Llama
-
-    # Default to using the model's full context length
-    if "n_ctx" not in llamacpp_model_params:
-        llamacpp_model_params["n_ctx"] = 0
-
-    if "verbose" not in llamacpp_model_params:
-        llamacpp_model_params["verbose"] = False
-
-    # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
-    if "tokenizer" not in llamacpp_model_params:
-        warnings.warn(
-            "The pre-tokenizer in `llama.cpp` handles unicode improperly "
-            + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
-            + "Outlines may raise a `RuntimeError` when building the regex index.\n"
-            + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
-            + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
-        )
-
-    model = Llama.from_pretrained(repo_id, filename, **llamacpp_model_params)
-
-    return LlamaCpp(model)
diff --git a/tests/generate/test_integration_llamacpp.py b/tests/generate/test_integration_llamacpp.py
deleted file mode 100644
index fd5be2171..000000000
--- a/tests/generate/test_integration_llamacpp.py
+++ /dev/null
@@ -1,374 +0,0 @@
-import datetime
-import re
-
-import pytest
-from pydantic import BaseModel, constr
-
-import outlines.generate as generate
-import outlines.models as models
-import outlines.samplers as samplers
-
-TEST_MODEL = "./llama-test-model/TinyMistral-248M-v2-Instruct.Q4_K_M.gguf"
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    return models.llamacpp(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-    )
-
-
-@pytest.mark.parametrize(
-    "generator_type,params",
-    (
-        (generate.text, []),
-        (generate.regex, ("[0-9]",)),
-        # (generate.cfg, (grammars.arithmetic,)),  # Awaiting CFG fix
-    ),
-)
-def test_llamacpp_generation_api(model, generator_type, params):
-    generator = generator_type(model, *params)
-
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=".")
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res1 = generator("test", seed=1, max_tokens=10)
-    res2 = generator("test", seed=1, max_tokens=10)
-    assert isinstance(res1, str)
-    assert isinstance(res2, str)
-    assert res1 == res2
-
-
-def test_llama_cpp_streaming_api(model):
-    generator = generate.text(model)
-    token_generator = generator.stream("test", max_tokens=10)
-    tokens = [token for token in token_generator]
-    assert len(tokens) <= 10
-    assert isinstance(tokens[0], str)
-
-
-@pytest.mark.xfail(reason="Batch inference is not available in `llama-cpp-python`.")
-def test_llamacpp_batch_inference(model):
-    generator = generate.text(model)
-    res = generator(["test", "test1"])
-    assert len(res) == 2
-
-
-def test_llamacpp_sampling_params(model):
-    generator = generate.text(model)
-
-    params = {
-        "frequency_penalty": 1.0,
-        "presence_penalty": 1.0,
-    }
-    res = generator("test", seed=1, max_tokens=10, **params)
-    assert isinstance(res, str)
-
-
-def test_llamacpp_greedy_sampling(model):
-    sampler = samplers.greedy()
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=20)
-    assert isinstance(res, str)
-
-
-def test_llamacpp_multinomial_sampling(model):
-    sampler = samplers.multinomial()
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(1, temperature=1.0)
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(1, top_k=1)
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(1, top_p=0.5)
-    generator = generate.text(model, sampler)
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-
-def test_llamacpp_several_samples(model):
-    sampler = samplers.multinomial(3)
-    generator = generate.text(model, sampler)
-    with pytest.raises(NotImplementedError, match="allow to take several samples"):
-        generator("test")
-
-
-def test_llamacpp_beam_search(model):
-    sampler = samplers.beam_search(1)
-    generator = generate.text(model, sampler)
-
-    with pytest.raises(NotImplementedError, match="does not support Beam Search"):
-        generator("test")
-
-
-def test_llamacpp_text_stop(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.text(model)(prompt, stop_at="a", max_tokens=100)
-    assert isinstance(sequence, str)
-    assert sequence.find("a") == -1
-
-
-def test_llamacpp_regex(model):
-    prompt = (
-        "<|im_start|>user\nWrite an email address<|im_end|>\n<|im_start|>assistant\n"
-    )
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # One prompt
-    sequence = generator(prompts=prompt)
-    assert isinstance(sequence, str)
-    assert re.fullmatch(pattern=regex_str, string=sequence) is not None
-
-
-def test_llamacpp_integer(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, int)(prompt, max_tokens=10)
-    assert isinstance(sequence, int)
-    assert sequence != ""
-    int(sequence)
-
-
-def test_llamacpp_float(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, float)(prompt, max_tokens=10)
-    assert isinstance(sequence, float)
-
-    assert sequence != ""
-    float(sequence)
-
-
-def test_llamacpp_bool(model):
-    prompt = (
-        "<|im_start|>user\nIs this True or False?<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, bool)(prompt, max_tokens=10)
-    assert isinstance(sequence, bool)
-
-    assert sequence != ""
-    bool(sequence)
-
-
-def test_llamacpp_date(model):
-    prompt = (
-        "<|im_start|>user\nWhat day is it today?<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.format(model, datetime.date)(prompt, max_tokens=20, seed=10)
-    assert isinstance(sequence, datetime.date)
-
-
-def test_llamacpp_time(model):
-    prompt = "<|im_start|>user\nWhat time is it?<|im_end|>\n<|im_start|>assistant\n"
-    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10)
-    assert isinstance(sequence, datetime.time)
-
-
-def test_llamacpp_datetime(model):
-    prompt = "<|im_start|>user\nWhat time is it?<|im_end|>\n<|im_start|>assistant\n"
-    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20)
-    assert isinstance(sequence, datetime.datetime)
-
-
-def test_llamacpp_choice(model):
-    prompt = (
-        "<|im_start|>user\nWrite a short sentence<|im_end|>\n<|im_start|>assistant\n"
-    )
-    sequence = generate.choice(model, ["test", "choice"])(prompt)
-    assert sequence == "test" or sequence == "choice"
-
-
-def test_llamacpp_json_basic(model):
-    prompt = "<|im_start|>user\nOutput some JSON<|im_end|>\n<|im_start|>assistant\n"
-
-    class Spam(BaseModel):
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam, whitespace_pattern="")(
-        prompt, max_tokens=100, temperature=0.0, seed=1
-    )
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.spam, str)
-    assert isinstance(result.fuzz, bool)
-    assert len(result.spam) <= 10
-
-
-def test_llamacpp_json_schema(model):
-    prompt = "<|im_start|>user\nOutput some JSON<|im_end|>\n<|im_start|>assistant\n"
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "boolean"},
-           "bar": {"type": "string", "maxLength": 4}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    result = generate.json(model, schema, whitespace_pattern="")(
-        prompt, max_tokens=100, temperature=0, seed=10
-    )
-    assert isinstance(result, dict)
-    assert isinstance(result["foo"], bool)
-    assert isinstance(result["bar"], str)
-
-
-@pytest.mark.parametrize(
-    "repo,model_path,hf_tokenizer_uri",
-    [
-        ("Qwen/Qwen1.5-0.5B-Chat-GGUF", "*q2*.gguf", "Qwen/Qwen1.5-0.5B-Chat"),
-        ("TheBloke/phi-2-GGUF", "*Q2*.gguf", "microsoft/phi-2"),
-    ],
-)
-def test_byte_tokenizer_regression(repo, model_path, hf_tokenizer_uri):
-    """Reproduce https://github.com/dottxt-ai/outlines/issues/820"""
-    import llama_cpp
-
-    model = models.llamacpp(
-        repo,
-        model_path,
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            hf_tokenizer_uri
-        ),
-    )
-    generator = generate.choice(model, ["skirt", "dress", "pen", "jacket"])
-    generator("Pick the odd word out: skirt, dress, pen, jacket")
-
-
-def test_llama_cpp_pre_tokenizer_remains_broken():
-    """If fails, llama.cpp pre-tokenizer is fixed -> revert #892, remove `with pytest.raises`"""
-    repo = "Qwen/Qwen1.5-0.5B-Chat-GGUF"
-    model_path = "*q2*.gguf"
-
-    model = models.llamacpp(repo, model_path)
-    with pytest.raises(RuntimeError):
-        generate.choice(model, ["skirt", "dress", "pen", "jacket"])
-
-
-@pytest.mark.skip("Caching for guide was temporarily turned off")
-def test_RegexGuide_caching(model, temp_cache_dir):
-    import llama_cpp
-
-    import outlines.caching
-    from outlines.fsm.guide import cached_create_states_mapping
-
-    assert outlines.caching._caching_enabled
-
-    regex = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-    prompt = "What is the IP address of the Google DNS servers? "
-
-    cache = outlines.caching.get_cache()
-
-    # Returns (hits, misses)
-    _ = cache.stats(enable=True)
-    assert cache.statistics
-
-    assert cached_create_states_mapping.__memory__ is cache
-
-    generator = generate.regex(model, regex, sampler=samplers.greedy())
-    assert cache.stats() == (0, 1)
-
-    model_2 = models.llamacpp(
-        "Qwen/Qwen1.5-0.5B-Chat-GGUF",
-        "*q2*.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "Qwen/Qwen1.5-0.5B-Chat"
-        ),
-    )
-    generator_2 = generate.regex(model_2, regex, sampler=samplers.greedy())
-    assert cache.stats() == (0, 2)
-
-    # These two different models and tokenizers should not have the same state
-    # mapping results
-    assert (
-        generator.logits_processor.guide.states_to_token_maps
-        != generator_2.logits_processor.guide.states_to_token_maps
-    )
-
-    generator_3 = generate.regex(model_2, regex, sampler=samplers.greedy())
-    assert cache.stats() == (1, 2)
-    assert (
-        generator_2.logits_processor.guide.states_to_token_maps
-        == generator_3.logits_processor.guide.states_to_token_maps
-    )
-
-    # Just for fun...
-    structured = generator(prompt, max_tokens=30)
-    structured_2 = generator_2(prompt, max_tokens=30)
-
-    assert re.fullmatch(regex, structured)
-    assert re.fullmatch(regex, structured_2)
-    assert structured != structured_2
-
-
-@pytest.mark.xfail(
-    reason="Some versions of the Hermes-2-Pro-Llama-3 model have a broken config"
-)
-def test_tokenizer_vocabulary_decode_sanity():
-    """Assert the decoded newline token (198) is the same as the normalized vocab token"""
-    import llama_cpp
-
-    model = models.llamacpp(
-        "bartowski/Meta-Llama-3-8B-Instruct-GGUF",
-        "Meta-Llama-3-8B-Instruct-IQ1_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B",
-        ),
-    )
-    tokenizer = generate.regex(model, "a").logits_processor.tokenizer
-
-    decoded_nl_token = tokenizer.decode([198])[0]
-    vocab_nl_token = tokenizer.convert_token_to_string(
-        [token for token, token_id in tokenizer.vocabulary.items() if token_id == 198][
-            0
-        ]
-    )
-    assert decoded_nl_token == vocab_nl_token
-
-
-def test_no_length_constraint_when_unset():
-    """Assert that models.llamacpp doesn't have an implicit max_tokens preventing full sequence generation"""
-    import llama_cpp
-
-    model = models.llamacpp(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "Locutusque/TinyMistral-248M-Instruct"
-        ),
-    )
-
-    long_pattern = "abcdefg" * 10
-    generator = generate.regex(model, long_pattern)
-
-    output = generator("a")
-    assert re.match(long_pattern, output)
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
new file mode 100644
index 000000000..ffb0110cd
--- /dev/null
+++ b/tests/models/test_llamacpp.py
@@ -0,0 +1,118 @@
+import json
+from enum import Enum
+
+import pytest
+from pydantic import BaseModel
+
+from outlines.models import LlamaCpp
+from outlines.processors import RegexLogitsProcessor
+from outlines.types import Choice, Json, Regex
+
+
+def test_load_model():
+    model = LlamaCpp.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+
+    assert isinstance(model, LlamaCpp)
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    return LlamaCpp.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+
+
+def test_llamacpp_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_llamacpp_regex(model):
+    regex_str = Regex(r"[0-9]").to_regex()
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
+    result = model.generate("Respond with one word. Not more.", logits_processor)
+    assert isinstance(result, str)
+
+
+def test_llamacpp_json(model):
+    class Foo(BaseModel):
+        bar: str
+
+    regex_str = Json(Foo).to_regex()
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
+    result = model.generate(
+        "foo? Respond with one word.", logits_processor, max_tokens=1000
+    )
+
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+def test_llamacpp_choice(model):
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    regex_str = Choice(Foo).to_regex()
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
+    result = model.generate("foo?", logits_processor)
+
+    assert result == "Foo" or result == "Bar"
+
+
+def test_llamacpp_text_stop(model):
+    result = model.generate("Write the letter a.", None, stop="a", max_tokens=100)
+    assert "a" not in result
+
+
+def test_llamacpp_stream_text_stop(model):
+    generator = model.stream("Write the letter a.", None, stop="a", max_tokens=100)
+
+    result = next(generator)
+    assert isinstance(result, str)
+    assert result != "a"
+
+
+def test_llamacpp_stream_simple(model):
+    generator = model.stream("Respond with one word. Not more.", None)
+
+    x = next(generator)
+    assert isinstance(x, str)
+
+
+def test_llamacpp_stream_regex(model):
+    regex_str = Regex(r"[0-9]").to_regex()
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
+    generator = model.stream("Respond with one word. Not more.", logits_processor)
+
+    x = next(generator)
+    assert isinstance(x, str)
+
+
+def test_llamacpp_stream_json(model):
+    class Foo(BaseModel):
+        bar: int
+
+    regex_str = Json(Foo).to_regex()
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
+    generator = model.stream("foo?", logits_processor)
+
+    x = next(generator)
+    assert x == "{"
+
+
+def test_llamacpp_stream_choice(model):
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    regex_str = Choice(Foo).to_regex()
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
+    generator = model.stream("foo?", logits_processor)
+
+    x = next(generator)
+    assert isinstance(x, str)

From 3349f4942d9d433958db2fd39d33f64d1693f562 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 29 Nov 2024 18:46:37 +0100
Subject: [PATCH 204/505] Add `LocalModel` and `APIModel` types

---
 outlines/models/__init__.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 7bfb09fb0..620108bb9 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -19,3 +19,6 @@
 from .vllm import VLLM, vllm
 
 LogitsGenerator = Union[Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM]
+
+LocalModel = LlamaCpp
+APIModel = Union[AzureOpenAI, OpenAI, Anthropic, Gemini]

From b4d8245c7c56addf790658beb6617a91cc195da3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 29 Nov 2024 20:46:08 +0100
Subject: [PATCH 205/505] Add `Generator` function and builder

---
 outlines/generate/__init__.py | 71 +++++++++++++++++++++++++++++++++++
 1 file changed, 71 insertions(+)

diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
index f28cbd80d..12041af49 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -1,3 +1,9 @@
+from dataclasses import dataclass
+from typing import TYPE_CHECKING, Any, Optional, Union, cast, get_args
+
+from outlines.models import APIModel, LocalModel
+from outlines.types import Choice, Json, List, Regex
+
 from .api import SequenceGenerator
 from .cfg import cfg
 from .choice import choice
@@ -6,3 +12,68 @@
 from .json import json
 from .regex import regex
 from .text import text
+
+if TYPE_CHECKING:
+    from outlines.processors import RegexLogitsProcessor
+
+
+@dataclass
+class APIGenerator:
+    """Represents an API-based generator.
+
+    Attributes
+    ----------
+    model
+        An instance of a model wrapper.
+    output_type
+        The output type.
+
+    """
+
+    model: APIModel
+    output_type: Optional[Union[Json, List, Choice, Regex]] = None
+
+    def __call__(self, prompt, **inference_kwargs):
+        return self.model.generate(prompt, self.output_type, **inference_kwargs)
+
+
+@dataclass
+class LocalGenerator:
+    """Represents a local model-based generator.
+
+    We use this class to keep track of the logits processor which can be quite
+    expensive to build.
+
+    Attributes
+    ----------
+    model
+        An instance of a model wrapper.
+    output_type
+        The output type.
+
+    """
+
+    model: LocalModel
+    output_type: Optional[Union[Json, List, Choice, Regex]]
+
+    def __post_init__(self):
+        if self.output_type is None:
+            self.logits_processor = None
+        else:
+            regex_string = self.output_type.to_regex()
+            self.logits_processor = RegexLogitsProcessor(
+                regex_string, self.model.tokenizer
+            )
+
+    def __call__(self, prompt, **inference_kwargs):
+        return self.model.generate(prompt, self.logits_processor, **inference_kwargs)
+
+
+def Generator(
+    model: Union[LocalModel, APIModel],
+    output_type: Optional[Union[Json, List, Choice, Regex]] = None,
+):
+    if isinstance(model, APIModel):  # type: ignore
+        return APIGenerator(model, output_type)  # type: ignore
+    else:
+        return LocalGenerator(model, output_type)  # type: ignore

From 01509d588b7b19855743c94ef45cb9f72f6a7567 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 29 Nov 2024 18:46:02 +0100
Subject: [PATCH 206/505] Add `to_regex` method to the different types

---
 outlines/types/__init__.py | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index d81d6f2e2..5ff95f662 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -5,6 +5,7 @@
 
 from jsonschema import Draft202012Validator as Validator
 from jsonschema.exceptions import SchemaError
+from outlines_core.json_schema import build_regex_from_schema
 from pydantic import BaseModel, TypeAdapter
 from typing_extensions import _TypedDictMeta  # type: ignore
 
@@ -76,6 +77,7 @@ class Json:
     """
 
     definition: Union[str, dict]
+    whitespace_pattern: str = " "
 
     def to_json_schema(self):
         if isinstance(self.definition, str):
@@ -98,11 +100,21 @@ def to_json_schema(self):
 
         return schema
 
+    def to_regex(self):
+        schema = self.to_json_schema()
+        schema_str = json.dumps(schema)
+        return build_regex_from_schema(schema_str, self.whitespace_pattern)
+
 
 @dataclass
 class List:
     definition: list
 
+    def to_regex(self):
+        raise NotImplementedError(
+            "Structured generation for lists of objects are not implemented yet."
+        )
+
 
 @dataclass
 class Choice:
@@ -113,3 +125,14 @@ class Choice:
     def __post_init__(self):
         if isinstance(self.definition, list):
             self.definition = Enum("Definition", [(x, x) for x in self.definition])
+
+    def to_list(self):
+        if isinstance(self.definition, list):
+            return self.definition
+        else:
+            return [x.value for x in self.definition]
+
+    def to_regex(self):
+        choices = self.to_list()
+        regex_str = r"(" + r"|".join(choices) + r")"
+        return regex_str

From 7a707965f04fd3a96e35889e8a4d0bcb7905e404 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 2 Dec 2024 09:03:46 +0100
Subject: [PATCH 207/505] Support dataclasses to define Json Schema

---
 outlines/types/__init__.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 5ff95f662..9246c96a7 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,5 +1,5 @@
-from dataclasses import dataclass
 import json
+from dataclasses import dataclass, is_dataclass
 from enum import Enum, EnumMeta
 from typing import Union
 
@@ -88,6 +88,8 @@ def to_json_schema(self):
             schema = self.definition.model_json_schema()
         elif isinstance(self.definition, _TypedDictMeta):
             schema = TypeAdapter(self.definition).json_schema()
+        elif is_dataclass(self.definition):
+            schema = TypeAdapter(self.definition).json_schema()
         else:
             raise TypeError(
                 "The Json definition must be a JSON Schema string, dictionary or Pydantic model."

From bd2545b83c980b15bea39c39f240e4899babe95e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 2 Dec 2024 11:23:58 +0100
Subject: [PATCH 208/505] Update the documentation of the llama.cpp integration

---
 docs/reference/models/llamacpp.md | 100 ++++++------------------------
 1 file changed, 19 insertions(+), 81 deletions(-)

diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
index 51b62eca8..52e892b76 100644
--- a/docs/reference/models/llamacpp.md
+++ b/docs/reference/models/llamacpp.md
@@ -12,50 +12,38 @@ Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/l
 
 ## Load the model
 
-You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+To load a model you can use the same interface as you would using `llamap-cpp-python` directly. The default method is to initialize the model by passing the path to the weights on your machine. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
 
 ```python
 from outlines import models
 
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+llm = models.LlamaCpp("./phi-2.Q4_K_M.gguf")
 ```
 
-This will download the model files to the hub cache folder and load the weights in memory.
+You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 
-You can also initialize the model by passing the path to the weights on your machine. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
 
 ```python
 from outlines import models
-from llama_cpp import Llama
 
-llm = Llama("./phi-2.Q4_K_M.gguf")
-model = models.LlamaCpp(llm)
+model = models.LlamaCpp.from_pretrained("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
 ```
 
-If you need more control, you can pass the same keyword arguments to the model as you would pass in the [llama-ccp-library][llamacpp]:
+This will download the model files to the hub cache folder and load the weights in memory.
+
+
+You can pass the same keyword arguments to the model as you would pass in the [llama-ccp-library][llamacpp]:
 
 ```python
 from outlines import models
 
-model = models.llamacpp(
+model = models.LlamaCpp(
     "TheBloke/phi-2-GGUF",
     "phi-2.Q4_K_M.gguf"
     n_ctx=512,  # to set the context length value
 )
 ```
 
-**Main parameters:**
-
-| Parameters | Type | Description | Default |
-|------------|------|-------------|---------|
-| `n_gpu_layers`| `int` | Number of layers to offload to GPU. If -1, all layers are offloaded | `0` |
-| `split_mode` | `int` | How to split the model across GPUs. `1` for layer-wise split, `2` for row-wise split | `1` |
-| `main_gpu` | `int` | Main GPU | `0` |
-| `tensor_split` | `Optional[List[float]]` | How split tensors should be distributed across GPUs. If `None` the model is not split. | `None` |
-| `n_ctx` | `int` | Text context. Inference from the model if set to `0` | `0` |
-| `n_threads` | `Optional[int]` | Number of threads to use for generation. All available threads if set to `None`.| `None` |
-| `verbose` | `bool` | Print verbose outputs to `stderr` | `False` |
-
 See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters.
 
 ### Load the model on GPU
@@ -69,87 +57,39 @@ See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io
 ```python
 from outlines import models
 
-model = models.llamacpp(
+model = models.LlamaCpp(
     "TheBloke/phi-2-GGUF",
     "phi-2.Q4_K_M.gguf",
     n_gpu_layers=-1,  # to use GPU acceleration
 )
 ```
 
-This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
 
-### Load LoRA adapters
+## Generate text
+
 
-You can load LoRA adapters dynamically:
+To generate text you must first create a `Generator` object by passing the model instance and, possibley, the expected output type:
 
 ```python
 from outlines import models, generate
 
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-generator = generate.text(model)
-answer_1 = generator("prompt")
 
-model.load_lora("./path/to/adapter.gguf")
-answer_2 = generator("prompt")
+model = models.LlamaCpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+generator = Generator(model)
 ```
 
-To load another adapter you need to re-initialize the model. Otherwise the adapter will be added on top of the previous one:
+You can pass to the generator the same keyword arguments you would pass in `llama-cpp-python`:
 
 ```python
-from outlines import models
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-model.load_lora("./path/to/adapter1.gguf")  # Load first adapter
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-model.load_lora("./path/to/adapter2.gguf")  # Load second adapter
+answer = generator("A prompt", presence_penalty=0.8)
 ```
 
-## Generate text
-
-In addition to the parameters described in the [text generation section](../text.md) you can pass extra keyword arguments, for instance to set sampling parameters not exposed in Outlines' public API:
+You can also stream the tokens:
 
 ```python
-from outlines import models, generate
-
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-generator = generate.text(model)
-
-answer = generator("A prompt", presence_penalty=0.8)
+tokens = generator.stream("A prompt")
 ```
 
-**Extra keyword arguments:**
-
-The value of the keyword arguments you pass to the generator suspersede the values set when initializing the sampler or generator. All extra sampling methods and repetition penalties are disabled by default.
-
-| Parameters | Type | Description | Default |
-|------------|------|-------------|---------|
-| `suffix` | `Optional[str]` | A suffix to append to the generated text. If `None` no suffix is added. | `None` |
-| `echo` | `bool` | Whether to preprend the prompt to the completion. | `False` |
-| `seed` | `int` | The random seed to use for sampling. | `None` |
-| `max_tokens` | `Optional[int]` | The maximum number of tokens to generate. If `None` the maximum number of tokens depends on `n_ctx`. | `16` |
-| `frequence_penalty` | `float` | The penalty to apply to tokens based on their frequency in the past 64 tokens. | `0.0` |
-| `presence_penalty` | `float` | The penalty to apply to tokens based on their presence in the past 64 tokens. | `0.0` |
-| `repeat_penalty` | `float` | The penalty to apply to repeated tokens in the past 64 tokens. | `1.` |
-| `stopping_criteria` | `Optional[StoppingCriteriaList]` | A list of stopping criteria to use. | `None`
-| `logits_processor` | `Optional[LogitsProcessorList]` | A list of logits processors to use. The logits processor used for structured generation will be added to this list. | `None`
-| `temperature` | `float` | The temperature to use for sampling | `1.0` |
-| `top_p` | `float` | The top-p value to use for [nucleus sampling][degeneration]. | `1.` |
-| `min_p` | `float` | The min-p value to use for [minimum-p sampling][minimum-p]. | `0.` |
-| `typical_p` | `float` | The p value to use for [locally typical sampling][locally-typical]. | `1.0` |
-| `stop` | `Optional[Union[str, List[str]]]` | A list of strings that stop generation when encountered. | `[]` |
-| `top_k` |  `int` | The top-k value used for [top-k sampling][top-k]. Negative value to consider all logit values. | `-1.` |
-| `tfs_z` | `float` | The [tail-free sampling][tail-free] parameter. | `1.0` |
-| `mirostat_mode` | `int` | The [mirostat sampling][mirostat] mode. | `0` |
-| `mirostat_tau` | `float` | The target cross-entropy for [mirostat sampling][mirostat].| `5.0` |
-| `mirostat_eta` | `float` | The learning rate used to update `mu` in [mirostat sampling][mirostat]. | `0.1` |
-
-See the [llama-cpp-python documentation][llama-cpp-python-call] for the full and up-to-date list of parameters and the [llama.cpp code][llama-cpp-sampling-params] for the default values of other
-sampling parameters.
-
-### Streaming
-
 
 ## Installation
 
@@ -216,8 +156,6 @@ CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-
 - SYCL
 
 
-
-
 [llamacpp]: https://github.com/abetlen/llama-cpp-python
 [llama-cpp-python-call]: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
 [llama-cpp-python-install]: https://github.com/abetlen/llama-cpp-python/tree/08b16afe11e7b42adec2fed0a781123383476045?tab=readme-ov-file#supported-backends

From 504a180af5934a70260e74601f0b09cc3342335d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 2 Dec 2024 11:24:26 +0100
Subject: [PATCH 209/505] Update the documentation of the OpenAI integration

---
 docs/reference/models/openai.md | 173 ++++++++++----------------------
 1 file changed, 51 insertions(+), 122 deletions(-)

diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
index 0777ad8b7..eae8dde39 100644
--- a/docs/reference/models/openai.md
+++ b/docs/reference/models/openai.md
@@ -2,7 +2,7 @@
 
 !!! Installation
 
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively:
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively you can run:
 
     ```bash
     pip install "outlines[openai]"
@@ -10,44 +10,30 @@
 
 ## OpenAI models
 
-Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.openai`:
+Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.OpenAI`:
 
 ```python
 from outlines import models
 
 
-model = models.openai("gpt-4o-mini")
-model = models.openai("gpt-4o")
+model = models.OpenAI("gpt-4o-mini")
+model = models.OpenAI("gpt-4o")
 ```
 
-Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.AsyncOpenAI` as keyword arguments:
+Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.OpenAI` as keyword arguments:
 
 ```python
 import os
 from outlines import models
 
 
-model = models.openai(
+model = models.OpenAI(
     "gpt-4o-mini",
     api_key=os.environ["OPENAI_API_KEY"]
 )
 ```
 
-The following table enumerates the possible parameters. Refer to the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list.
-
-**Parameters:**
-
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
-| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
-| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
-| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if no specified. | `None` |
-| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
-| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
-| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
-| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
-| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
+Refer to the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list of the initialization parameters.
 
 ## Azure OpenAI models
 
@@ -57,93 +43,14 @@ Outlines also supports Azure OpenAI models:
 from outlines import models
 
 
-model = models.azure_openai(
+model = models.AzureOpenAI(
     "azure-deployment-name",
-    "gpt-4o-mini",
     api_version="2024-07-18",
     azure_endpoint="https://example-endpoint.openai.azure.com",
 )
 ```
 
-!!! Question "Why do I need to specify model and deployment name?"
-
-    The model name is needed to load the correct tokenizer for the model. The tokenizer is necessary for structured generation.
-
-
-You can pass any parameter you would pass to `openai.AsyncAzureOpenAI`. You can consult the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/lib/azure.py) for an up-to-date list.
-
-**Parameters:**
-
-
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `azure_endpoint` | `str` | Azure endpoint, including the resource. Infered from `AZURE_OPENAI_ENDPOINT` if not specified | `None` |
-| `api_version` | `str` | API version. Infered from `AZURE_OPENAI_API_KEY` if not specified | `None` |
-| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
-| `azure_ad_token` | `str` | Azure active directory token. Inference from `AZURE_OPENAI_AD_TOKEN` if not specified | `None` |
-| `azure_ad_token_provider` | `AzureADTokenProvider` | A function that returns an Azure Active Directory token | `None` |
-| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
-| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
-| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if not specified. | `None` |
-| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
-| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
-| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
-| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
-| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
-
-## Models that follow the OpenAI standard
-
-Outlines supports models that follow the OpenAI standard. You will need to initialize the OpenAI client properly configured and pass it to `outlines.models.openai`
-
-```python
-import os
-from openai import AsyncOpenAI
-from outlines import models
-from outlines.models.openai import OpenAIConfig
-
-
-client = AsyncOpenAI(
-    api_key=os.environ.get("PROVIDER_KEY"),
-    base_url="http://other.provider.server.com"
-)
-config = OpenAIConfig("model_name")
-model = models.openai(client, config)
-```
-
-!!! Warning
-
-    You need to pass the async client to be able to do batch inference.
-
-## Structured Generation Support
-
-Outlines provides support for [OpenAI Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode) via `outlines.generate.json`, `outlines.generate.choice`
-
-```python
-from pydantic import BaseModel, ConfigDict
-import outlines.models as models
-from outlines import generate
-
-model = models.openai("gpt-4o-mini")
-
-class Person(BaseModel):
-    model_config = ConfigDict(extra='forbid')  # required for openai
-    first_name: str
-    last_name: str
-    age: int
-
-generator = generate.json(model, Person)
-generator("current indian prime minister on january 1st 2023")
-# Person(first_name='Narendra', last_name='Modi', age=72)
-
-generator = generate.choice(model, ["Chicken", "Egg"])
-print(generator("Which came first?"))
-# Chicken
-```
-
-!!! Warning
-
-    Structured generation support only provided to OpenAI-compatible endpoints which conform to OpenAI's standard. Additionally, `generate.regex` and `generate.cfg` are not supported.
-
+You can pass any parameter you would pass to `openai.AzureOpenAI`. You can consult the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/lib/azure.py) for an up-to-date list.
 
 ## Advanced configuration
 
@@ -163,42 +70,64 @@ client = AsyncOpenAI(
         transport=httpx.HTTPTransport(local_address="0.0.0.0"),
     ),
 )
-config = OpenAIConfig("model_name")
-model = models.openai(client, config)
 ```
 
-It is possible to specify the values for `seed`, `presence_penalty`, `frequence_penalty`, `top_p` by passing an instance of `OpenAIConfig` when initializing the model:
+## Models that follow the OpenAI standard
+
+Outlines supports models that follow the OpenAI standard. You will need to initialize the OpenAI client properly configured and pass it to `outlines.models.OpenAI`
 
 ```python
-from outlines.models.openai import OpenAIConfig
+import os
+from openai import AsyncOpenAI
 from outlines import models
+from outlines.models.openai import OpenAIConfig
 
 
-config = OpenAIConfig(
-    presence_penalty=1.,
-    frequency_penalty=1.,
-    top_p=.95,
-    seed=0,
+model = models.OpenAI(
+    "model_name",
+    api_key=os.environ.get("PROVIDER_KEY"),
+    base_url="http://other.provider.server.com"
 )
-model = models.openai("gpt-4o-mini", config)
 ```
 
-## Monitoring API use
+## Text generation
 
-It is important to be able to track your API usage when working with OpenAI's API. The number of prompt tokens and completion tokens is directly accessible via the model instance:
+To generate text using an OpenAI model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
 
 ```python
-from openai import AsyncOpenAI
-import outlines.models
+from outlines import models, Generator
+
+model = models.OpenAI("gpt-4o-mini")
+generator = Generator(model)
+result = generator("Prompt", seed=10)
+```
+
+See the [OpenAI SDK documentation](https://github.com/openai/openai-python/blob/6974a981aec1814b5abba429a8ea21be9ac58538/src/openai/types/completion_create_params.py#L13) for the list of available arguments.
+
+### Structured Generation Support
 
+Outlines provides support for [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode). Currently only JSON-Schema is supported:
 
-model = models.openai("gpt-4o")
+```python
+from pydantic import BaseModel
+from outlines import models, Generator
+from outlines.types import Json
+
+model = models.OpenAI("gpt-4o-mini")
 
-print(model.prompt_tokens)
-# 0
+class Person(BaseModel):
+    first_name: str
+    last_name: str
+    age: int
 
-print(model.completion_tokens)
-# 0
+generator = Generator(model, Json(Person))
+generator("current indian prime minister on january 1st 2023")
+# Person(first_name='Narendra', last_name='Modi', age=72)
 ```
 
-These numbers are updated every time you call the model.
+The following objects can be used to define the structure of the Json object:
+- A string that represents a Json Schema
+- A dictionary that represents a Json Schema
+- A Pydantic model
+- A TypedDict
+- A dataclass

From 18e0fcffbb95998a1f5647dc1b2f64e4992dadb6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 2 Dec 2024 11:25:31 +0100
Subject: [PATCH 210/505] Update the documentation for the Gemini integration

---
 docs/reference/models/gemini.md | 88 +++++++++++++++++++++++++++++++++
 pyproject.toml                  |  1 +
 2 files changed, 89 insertions(+)
 create mode 100644 docs/reference/models/gemini.md

diff --git a/docs/reference/models/gemini.md b/docs/reference/models/gemini.md
new file mode 100644
index 000000000..07ec9aa5a
--- /dev/null
+++ b/docs/reference/models/gemini.md
@@ -0,0 +1,88 @@
+# Gemini
+
+!!! Installation
+
+    You need to install the `google-generativeai` library to be able to use the Gemini API in Outlines. Or alternatively you can run:
+
+    ```bash
+    pip install "outlines[gemini]"
+    ```
+
+## Gemini models
+
+Outlines supports models available via the Gemini API, e.g. Gemini 1.5. You can initialize the model by passing the model name to `outlines.models.Gemini`:
+
+```python
+from outlines import models
+
+model = models.Gemini("gemini-1-5-flash")
+model = models.Gemini("gemini-1-5-pro")
+```
+
+Check the [Gemini documentation](https://ai.google.dev/gemini-api/docs/models/gemini) for an up-to-date list of available models.
+
+## Text generation
+
+To generate text using a Gemini model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
+
+```python
+from outlines import models, Generator
+
+model = models.Gemini("gemini-1-5-flash")
+generator = Generator(model)
+result = generator("Prompt", max_tokens=1024)
+```
+
+### Structured generation
+
+Gemini provides support for structured outputs.
+
+#### Json Schema
+
+Outlines provides support for JSON Schema-based structured generation with the Gemini models:
+
+```python
+from collections import TypedDict
+from outlines import Generator, models
+from outlines.types import Json
+
+model = models.Gemini("gemini-1-5-flash")
+
+class Person(TypedDict):
+    first_name: str
+    last_name: str
+    age: int
+
+generator = Generator(model, Json(Person))
+generator("current indian prime minister on january 1st 2023")
+# Person(first_name='Narendra', last_name='Modi', age=72)
+```
+
+Because of the current limitations of the Gemini SDK only The following objects can be used to define the structure of the Json object:
+- A Pydantic model
+- A TypedDict
+
+#### Multiple choices
+
+Outlines provides support for multiple-choices structured generation. Enums and lists of choices are supported:
+
+```python
+from enum import Enum
+from outlines import Generator, models
+from outlines.types import Choice
+
+model = models.Gemini("gemini-1-5-flash")
+
+class Foo(Enum):
+    foo = "Foo"
+    fizz = "Fizz"
+    fuzz = "Fuzz"
+
+generator = Generator(model, Choice(Foo))
+generator("current indian prime minister on january 1st 2023")
+# Person(first_name='Narendra', last_name='Modi', age=72)
+```
+
+The following objects can be used to define the choices:
+- An Enum object
+- A Python list
diff --git a/pyproject.toml b/pyproject.toml
index d73a93b48..cad7a2c70 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -51,6 +51,7 @@ vllm = ["vllm", "transformers", "numpy<2"]
 transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
 mlxlm = ["mlx-lm", "datasets"]
 openai = ["openai"]
+gemini = ["google-generativeai"]
 llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
 exllamav2 = ["exllamav2"]
 test = [

From a757f70c80c6a0ed8af852a21b99930cf893f2de Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 2 Dec 2024 11:26:01 +0100
Subject: [PATCH 211/505] Update the documentation for the Anthropic
 integration

---
 docs/reference/models/anthropic.md | 45 ++++++++++++++++++++++++++++++
 pyproject.toml                     |  1 +
 2 files changed, 46 insertions(+)
 create mode 100644 docs/reference/models/anthropic.md

diff --git a/docs/reference/models/anthropic.md b/docs/reference/models/anthropic.md
new file mode 100644
index 000000000..ffc510f43
--- /dev/null
+++ b/docs/reference/models/anthropic.md
@@ -0,0 +1,45 @@
+# Anthropic
+
+!!! Installation
+
+    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines. Or alternatively you can run:
+
+    ```bash
+    pip install "outlines[anthropic]"
+    ```
+
+## Anthropic models
+
+Outlines supports models available via the Anthropic API, e.g. Claude 3.5 Haiku or Claude 3.5 Sonner. You can initialize the model by passing the model name to `outlines.models.Anthropic`:
+
+```python
+from outlines import models
+
+model = models.Anthropic("claude-3-5-haiku-latest")
+model = models.Anthropic("claude-3-5-sonnet-latest")
+```
+
+Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-claude/models) for an up-to-date list of available models. You can pass any paramater you would pass to the Anthropic SDK as keyword arguments:
+
+```python
+model = models.Anthropic(
+    "claude-3.5-haiku-latest",
+    api_key="<my api key>"
+)
+```
+
+## Text generation
+
+To generate text using an Anthropic model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
+
+```python
+from outlines import models, Generator
+
+model = models.Anthropic("claude-3-5-haiku-latest")
+generator = Generator(model)
+result = generator("Prompt", max_tokens=1024)
+```
+
+See the [Anthropic SDK documentation](https://github.com/anthropics/anthropic-sdk-python/blob/main/src/anthropic/resources/messages.py) for the list of available arguments.
+
+The Anthropic API currently does not support structured generation.
diff --git a/pyproject.toml b/pyproject.toml
index cad7a2c70..3a09fc84d 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -51,6 +51,7 @@ vllm = ["vllm", "transformers", "numpy<2"]
 transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
 mlxlm = ["mlx-lm", "datasets"]
 openai = ["openai"]
+anthropic = ["anthropic"]
 gemini = ["google-generativeai"]
 llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
 exllamav2 = ["exllamav2"]

From 7b5017ec641155a73546114b489755df0beb53b0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 16 Dec 2024 13:41:42 +0100
Subject: [PATCH 212/505] Run tests when PR against `v1.0` branch

---
 .github/workflows/tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 46d3d5ff6..796f7cc89 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -2,7 +2,7 @@ name: Tests
 
 on:
   pull_request:
-    branches: [main]
+    branches: [main,v1.0]
   push:
     branches: [main]
 

From 017ac784c22854468230401728111496d72c9061 Mon Sep 17 00:00:00 2001
From: Robin Picard <robinpicard@device-36.home>
Date: Thu, 13 Feb 2025 10:43:28 +0100
Subject: [PATCH 213/505] Fix the failing tests on branch v1.0

---
 tests/generate/test_generate.py |  8 ++++++--
 tests/models/test_openai.py     | 31 +++++++++++++++++++++++--------
 2 files changed, 29 insertions(+), 10 deletions(-)

diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 3d36faa5d..c7e9bdc32 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -15,10 +15,14 @@
 
 @pytest.fixture(scope="session")
 def model_llamacpp(tmp_path_factory):
-    return models.llamacpp(
+    from llama_cpp import Llama
+
+    llm = Llama.from_pretrained(
         repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
         filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        verbose=False,
     )
+    return models.LlamaCpp(llm)
 
 
 @pytest.fixture(scope="session")
@@ -115,7 +119,7 @@ def model_t5(tmp_path_factory):
 
 
 ALL_MODEL_FIXTURES = (
-    "model_llamacpp",
+    # "model_llamacpp",  # temporary disabled due to the v1 model refactoring
     "model_exllamav2",
     "model_mlxlm",
     "model_mlxlm_phi3",
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index a3d8a176b..d2b8d1a40 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,5 +1,6 @@
 import io
 import json
+import os
 
 import PIL
 import pytest
@@ -13,34 +14,48 @@
 MODEL_NAME = "gpt-4o-mini-2024-07-18"
 
 
-def test_openai_wrong_init_parameters():
+@pytest.fixture
+def api_key():
+    """Get the OpenAI API key from the environment, providing a default value if not found.
+
+    This fixture should be used for tests that do not make actual api calls,
+    but still require to initialize the OpenAI client.
+
+    """
+    api_key = os.getenv("OPENAI_API_KEY")
+    if not api_key:
+        return "MOCK_VALUE"
+    return api_key
+
+
+def test_openai_wrong_init_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
-        OpenAI(MODEL_NAME, foo=10)
+        OpenAI(MODEL_NAME, api_key=api_key, foo=10)
 
 
-def test_openai_wrong_inference_parameters():
+def test_openai_wrong_inference_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = OpenAI(MODEL_NAME)
+        model = OpenAI(MODEL_NAME, api_key=api_key)
         model.generate("prompt", foo=10)
 
 
-def test_openai_wrong_input_type():
+def test_openai_wrong_input_type(api_key):
     class Foo:
         def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = OpenAI(MODEL_NAME)
+        model = OpenAI(MODEL_NAME, api_key=api_key)
         model.generate(Foo("prompt"))
 
 
-def test_openai_wrong_output_type():
+def test_openai_wrong_output_type(api_key):
     class Foo:
         def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = OpenAI(MODEL_NAME)
+        model = OpenAI(MODEL_NAME, api_key=api_key)
         model.generate("prompt", Foo(1))
 
 

From aebccdf868329e3412208b004052b6b06bbf7cd4 Mon Sep 17 00:00:00 2001
From: Robin Picard <robinpicard@device-36.home>
Date: Thu, 13 Feb 2025 15:21:16 +0100
Subject: [PATCH 214/505] Create the CFG type to be used as an output type

---
 outlines/generate/__init__.py | 32 +++++++++++++++++++++-----------
 outlines/models/llamacpp.py   | 11 +++++++++++
 outlines/types/__init__.py    |  7 +++++++
 3 files changed, 39 insertions(+), 11 deletions(-)

diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
index 12041af49..2fb99a030 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -1,8 +1,9 @@
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, Any, Optional, Union, cast, get_args
+from typing import Any, Optional, Union, cast, get_args
 
-from outlines.models import APIModel, LocalModel
-from outlines.types import Choice, Json, List, Regex
+from outlines.models import APIModel, LlamaCpp, LocalModel
+from outlines.processors import CFGLogitsProcessor, RegexLogitsProcessor
+from outlines.types import CFG, Choice, Json, List, Regex
 
 from .api import SequenceGenerator
 from .cfg import cfg
@@ -13,9 +14,6 @@
 from .regex import regex
 from .text import text
 
-if TYPE_CHECKING:
-    from outlines.processors import RegexLogitsProcessor
-
 
 @dataclass
 class APIGenerator:
@@ -33,6 +31,12 @@ class APIGenerator:
     model: APIModel
     output_type: Optional[Union[Json, List, Choice, Regex]] = None
 
+    def __post_init__(self):
+        if isinstance(self.output_type, CFG):
+            raise NotImplementedError(
+                "CFG generation is not supported for API-based models"
+            )
+
     def __call__(self, prompt, **inference_kwargs):
         return self.model.generate(prompt, self.output_type, **inference_kwargs)
 
@@ -60,10 +64,16 @@ def __post_init__(self):
         if self.output_type is None:
             self.logits_processor = None
         else:
-            regex_string = self.output_type.to_regex()
-            self.logits_processor = RegexLogitsProcessor(
-                regex_string, self.model.tokenizer
-            )
+            if isinstance(self.output_type, CFG):
+                cfg_string = self.output_type.definition
+                self.logits_processor = CFGLogitsProcessor(
+                    cfg_string, self.model.tokenizer
+                )
+            else:
+                regex_string = self.output_type.to_regex()
+                self.logits_processor = RegexLogitsProcessor(
+                    regex_string, self.model.tokenizer
+                )
 
     def __call__(self, prompt, **inference_kwargs):
         return self.model.generate(prompt, self.logits_processor, **inference_kwargs)
@@ -71,7 +81,7 @@ def __call__(self, prompt, **inference_kwargs):
 
 def Generator(
     model: Union[LocalModel, APIModel],
-    output_type: Optional[Union[Json, List, Choice, Regex]] = None,
+    output_type: Optional[Union[Json, List, Choice, Regex, CFG]] = None,
 ):
     if isinstance(model, APIModel):  # type: ignore
         return APIGenerator(model, output_type)  # type: ignore
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 0e597c5a7..d4682b644 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -3,6 +3,7 @@
 from typing import TYPE_CHECKING, Dict, Iterator, List, Set, Tuple, Union
 
 from outlines.models.tokenizer import Tokenizer
+from outlines.processors import CFGLogitsProcessor
 
 if TYPE_CHECKING:
     from llama_cpp import Llama
@@ -159,6 +160,11 @@ def generate(self, prompt: str, logits_processor, **inference_kwargs) -> str:
                 "The `llama-cpp-python` library does not support batch inference."
             )
 
+        if isinstance(logits_processor, CFGLogitsProcessor):
+            raise NotImplementedError(
+                "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
+            )
+
         completion = self.model(
             prompt,
             logits_processor=LogitsProcessorList([logits_processor]),
@@ -197,6 +203,11 @@ def stream(
                 "The `llama-cpp-python` library does not support batch inference."
             )
 
+        if isinstance(logits_processor, CFGLogitsProcessor):
+            raise NotImplementedError(
+                "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
+            )
+
         generator = self.model(
             prompt,
             logits_processor=LogitsProcessorList([logits_processor]),
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 9246c96a7..01bc251da 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -138,3 +138,10 @@ def to_regex(self):
         choices = self.to_list()
         regex_str = r"(" + r"|".join(choices) + r")"
         return regex_str
+
+
+@dataclass
+class CFG:
+    """Represents a Context-Free Grammar as a string."""
+
+    definition: str

From 7a4231751866e5edd0330c82df22f59735b065d6 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Sun, 16 Feb 2025 10:29:46 +0100
Subject: [PATCH 215/505] Create a base class for the models (#1385)

Updated on 2025-02-11

The objective of this PR is to create a base class all models would
inherit from. This base class defines the interface for models and
includes a `__call__` method that allows models to be called with a
prompt and output type directly following the suggestion in #1359 .

Formatting methods for the input and output types are implemented in a
separate class inheriting from the base class `ModelTypeAdapter`. Each
model must set a `type_adapter` attribute in its initialization. The
rationale for separating those from the model class is that the accepted
types for each model and the formatting logic are quite different
concerns compared to the specifics of the model's implementation.
Separating them also allows us to have an easier to read documentation
of the accepted types for each model.

Another change proposed in this PR is to set a mandatory `model_type`
attribute in each model class. This would allow us not to have to rely
on a static list of models through the `APIModel` and `LocalModel`
classes.

The use of the base model class is implemented for the models that have
already been refactored: OpenAI, Gemini, Anthropic and Llamacpp.

---------

Co-authored-by: Robin Picard <robinpicard@device-36.home>
---
 outlines/models/__init__.py    |  1 +
 outlines/models/anthropic.py   | 25 +++++++++--
 outlines/models/base.py        | 82 ++++++++++++++++++++++++++++++++++
 outlines/models/gemini.py      | 17 ++++---
 outlines/models/llamacpp.py    | 74 +++++++++++++++++++++---------
 outlines/models/openai.py      | 23 ++++++----
 tests/models/test_anthropic.py |  7 +++
 tests/models/test_gemini.py    |  7 +++
 tests/models/test_llamacpp.py  |  5 +++
 tests/models/test_openai.py    |  7 +++
 10 files changed, 208 insertions(+), 40 deletions(-)
 create mode 100644 outlines/models/base.py

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 620108bb9..761de9cc5 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -9,6 +9,7 @@
 from typing import Union
 
 from .anthropic import Anthropic
+from .base import Model, ModelTypeAdapter
 from .exllamav2 import ExLlamaV2Model, exl2
 from .gemini import Gemini
 from .llamacpp import LlamaCpp
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index e55b2e3e2..830c0cf77 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -2,12 +2,23 @@
 from functools import singledispatchmethod
 from typing import Union
 
+from outlines.models.base import Model, ModelTypeAdapter
 from outlines.prompts import Vision
 
 __all__ = ["Anthropic"]
 
 
-class AnthropicBase:
+class AnthropicTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the Anthropic clients.
+
+    `AnthropicTypeAdapter` is responsible for preparing the arguments to
+    Anthropic's `messages.create` methods: the input (prompt and possibly
+    image).
+    Anthropic does not support defining the output type, so
+    `format_output_type` is not implemented.
+
+    """
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the `messages` argument to pass to the client.
@@ -62,18 +73,26 @@ def format_vision_input(self, model_input: Vision):
             ]
         }
 
+    def format_output_type(self, output_type):
+        """Not implemented for Anthropic."""
+        raise NotImplementedError(
+            f"The output type {output_type} is not available with Anthropic."
+        )
+
 
-class Anthropic(AnthropicBase):
+class Anthropic(Model):
     def __init__(self, model_name: str, *args, **kwargs):
         from anthropic import Anthropic
 
         self.client = Anthropic(*args, **kwargs)
         self.model_name = model_name
+        self.model_type = "api"
+        self.type_adapter = AnthropicTypeAdapter()
 
     def generate(
         self, model_input: Union[str, Vision], output_type=None, **inference_kwargs
     ):
-        messages = self.format_input(model_input)
+        messages = self.type_adapter.format_input(model_input)
 
         if output_type is not None:
             raise NotImplementedError(
diff --git a/outlines/models/base.py b/outlines/models/base.py
new file mode 100644
index 000000000..9d30522c4
--- /dev/null
+++ b/outlines/models/base.py
@@ -0,0 +1,82 @@
+from abc import ABC, abstractmethod
+
+
+class ModelTypeAdapter(ABC):
+    """Base class for all model type adapters.
+
+    A type adapter instance must be given as a value to the `type_adapter`
+    attribute when instantiating a model.
+    The type adapter is responsible for formatting the input and output types
+    passed to the model to match the specific format expected by the
+    associated model.
+
+    """
+
+    @abstractmethod
+    def format_input(self, model_input):
+        """Format the user input to the expected format of the model.
+
+        For API-based models, it typically means creating the `messages`
+        argument passed to the client. For local models, it can mean casting
+        the input from str to list for instance.
+        This method is also used to validate that the input type provided by
+        the user is supported by the model.
+
+        """
+        ...
+
+    @abstractmethod
+    def format_output_type(self, output_type):
+        """Format the output type to the expected format of the model.
+
+        For API-based models, this typically means creating a `response_format`
+        argument. For local models, it means formatting the logits processor to
+        create the object type expected by the model.
+
+        """
+        ...
+
+
+class Model(ABC):
+    """Base class for all models.
+
+    This class defines a shared `__call__` method that can be used to call the
+    model directly.
+    All models inheriting from this class must define a `type_adapter`
+    attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
+    attribute are used in the `generate` method to format the input and output
+    types received by the model.
+
+    """
+
+    type_adapter: ModelTypeAdapter
+
+    def __call__(self, model_input, output_type=None, **inference_kwargs):
+        """Call the model.
+
+        Users can call the model directly, in which case we will create a
+        generator instance with the output type provided and call it.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        generator("prompt")
+        ```
+        and
+        ```python
+        model("prompt", Foo)
+        ```
+
+        """
+        from outlines.generate import Generator
+
+        return Generator(self, output_type)(model_input, **inference_kwargs)
+
+    @abstractmethod
+    def generate(self, model_input, output_type=None, **inference_kwargs):
+        """Generate a response from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+
+        """
+        ...
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 1d6bbfd1e..5ab6a515c 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -7,17 +7,18 @@
 from pydantic import BaseModel
 from typing_extensions import _TypedDictMeta  # type: ignore
 
+from outlines.models.base import Model, ModelTypeAdapter
 from outlines.prompts import Vision
 from outlines.types import Choice, Json, List
 
 __all__ = ["Gemini"]
 
 
-class GeminiBase:
-    """Base class for the Gemini clients.
+class GeminiTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the Gemini clients.
 
-    `GeminiBase` is responsible for preparing the arguments to Gemini's
-    `generate_contents` methods: the input (prompt and possibly image), as well
+    `GeminiTypeAdapter` is responsible for preparing the arguments to Gemini's
+    `generate_content` methods: the input (prompt and possibly image), as well
     as the output type (only JSON).
 
     """
@@ -91,11 +92,13 @@ def format_enum_output_type(self, output_type):
         }
 
 
-class Gemini(GeminiBase):
+class Gemini(Model):
     def __init__(self, model_name: str, *args, **kwargs):
         import google.generativeai as genai
 
         self.client = genai.GenerativeModel(model_name, *args, **kwargs)
+        self.model_type = "api"
+        self.type_adapter = GeminiTypeAdapter()
 
     def generate(
         self,
@@ -105,9 +108,9 @@ def generate(
     ):
         import google.generativeai as genai
 
-        contents = self.format_input(model_input)
+        contents = self.type_adapter.format_input(model_input)
         generation_config = genai.GenerationConfig(
-            **self.format_output_type(output_type)
+            **self.type_adapter.format_output_type(output_type)
         )
         completion = self.client.generate_content(
             generation_config=generation_config, **contents, **inference_kwargs
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index d4682b644..8823f3358 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,7 +1,9 @@
 import pickle
 import warnings
+from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Dict, Iterator, List, Set, Tuple, Union
 
+from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
 from outlines.processors import CFGLogitsProcessor
 
@@ -94,7 +96,49 @@ def __setstate__(self, state):
         raise NotImplementedError("Cannot load a pickled llamacpp tokenizer")
 
 
-class LlamaCpp:
+class LlamaCppTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `llama-cpp-python` library.
+
+    `LlamaCppTypeAdapter` is responsible for preparing the arguments to
+    `llama-cpp-python`'s `Llama.__call__` method: the input (a string prompt),
+    as well as the logits processor (an instance of `LogitsProcessorList`).
+
+    """
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. "
+            "The `llama-cpp-python` library does not support batch inference. "
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        return model_input
+
+    def format_output_type(self, output_type):
+        """Generate the logits processor argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The logits processor provided.
+
+        """
+        from llama_cpp import LogitsProcessorList
+
+        return LogitsProcessorList([output_type])
+
+
+class LlamaCpp(Model):
     """Wraps a model provided by the `llama-cpp-python` library."""
 
     def __init__(self, model_path: Union[str, "Llama"], **kwargs):
@@ -116,6 +160,8 @@ def __init__(self, model_path: Union[str, "Llama"], **kwargs):
             self.model = Llama(model_path, **kwargs)
 
         self.tokenizer = LlamaCppTokenizer(self.model)
+        self.model_type = "local"
+        self.type_adapter = LlamaCppTypeAdapter()
 
     @classmethod
     def from_pretrained(cls, repo_id, filename, **kwargs):
@@ -135,7 +181,7 @@ def from_pretrained(cls, repo_id, filename, **kwargs):
             model = Llama.from_pretrained(repo_id, filename, **kwargs)
             return cls(model)
 
-    def generate(self, prompt: str, logits_processor, **inference_kwargs) -> str:
+    def generate(self, model_input, logits_processor, **inference_kwargs):
         """Generate text using `llama-cpp-python`.
 
         Arguments
@@ -153,21 +199,14 @@ def generate(self, prompt: str, logits_processor, **inference_kwargs) -> str:
         The generated text.
 
         """
-        from llama_cpp import LogitsProcessorList
-
-        if not isinstance(prompt, str):
-            raise NotImplementedError(
-                "The `llama-cpp-python` library does not support batch inference."
-            )
-
         if isinstance(logits_processor, CFGLogitsProcessor):
             raise NotImplementedError(
                 "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
             )
 
         completion = self.model(
-            prompt,
-            logits_processor=LogitsProcessorList([logits_processor]),
+            self.type_adapter.format_input(model_input),
+            logits_processor=self.type_adapter.format_output_type(logits_processor),
             **inference_kwargs,
         )
         result = completion["choices"][0]["text"]
@@ -177,7 +216,7 @@ def generate(self, prompt: str, logits_processor, **inference_kwargs) -> str:
         return result
 
     def stream(
-        self, prompt: str, logits_processor, **inference_kwargs
+        self, model_input, logits_processor, **inference_kwargs
     ) -> Iterator[str]:
         """Stream text using `llama-cpp-python`.
 
@@ -196,21 +235,14 @@ def stream(
         A generator that return strings.
 
         """
-        from llama_cpp import LogitsProcessorList
-
-        if not isinstance(prompt, str):
-            raise NotImplementedError(
-                "The `llama-cpp-python` library does not support batch inference."
-            )
-
         if isinstance(logits_processor, CFGLogitsProcessor):
             raise NotImplementedError(
                 "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
             )
 
         generator = self.model(
-            prompt,
-            logits_processor=LogitsProcessorList([logits_processor]),
+            self.type_adapter.format_input(model_input),
+            logits_processor=self.type_adapter.format_output_type(logits_processor),
             stream=True,
             **inference_kwargs,
         )
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index cc97f5f2c..378bb830d 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -6,18 +6,19 @@
 
 from pydantic import BaseModel
 
+from outlines.models.base import Model, ModelTypeAdapter
 from outlines.prompts import Vision
 from outlines.types import Json
 
 __all__ = ["OpenAI"]
 
 
-class OpenAIBase:
-    """Base class for the OpenAI clients.
+class OpenAITypeAdapter(ModelTypeAdapter):
+    """Type adapter for the OpenAI clients.
 
-    `OpenAI` base is responsible for preparing the arguments to OpenAI's
-    `completions.create` methods: the input (prompt and possibly image), as well
-    as the output type (only JSON).
+    `OpenAITypeAdapter` is responsible for preparing the arguments to OpenAI's
+    `completions.create` methods: the input (prompt and possibly image), as
+    well as the output type (only JSON).
 
     """
 
@@ -65,7 +66,7 @@ def format_vision_input(self, model_input: Vision):
                         {
                             "type": "image_url",
                             "image_url": {
-                                "url": f"data:{model_input.image_format};base64,{model_input.image_str}"
+                                "url": f"data:{model_input.image_format};base64,{model_input.image_str}"  # noqa: E702
                             },
                         },
                     ],
@@ -115,7 +116,7 @@ def format_json_output_type(self, output_type: Json):
         }
 
 
-class OpenAI(OpenAIBase):
+class OpenAI(Model):
     """Thin wrapper around the `openai.OpenAI` client.
 
     This wrapper is used to convert the input and output types specified by the
@@ -128,6 +129,8 @@ def __init__(self, model_name: str, *args, **kwargs):
 
         self.client = OpenAI(*args, **kwargs)
         self.model_name = model_name
+        self.model_type = "api"
+        self.type_adapter = OpenAITypeAdapter()
 
     def generate(
         self,
@@ -135,8 +138,8 @@ def generate(
         output_type: Optional[Union[type[BaseModel], str]] = None,
         **inference_kwargs,
     ):
-        messages = self.format_input(model_input)
-        response_format = self.format_output_type(output_type)
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
         result = self.client.chat.completions.create(
             model=self.model_name, **messages, **response_format, **inference_kwargs
         )
@@ -157,3 +160,5 @@ def __init__(self, deployment_name: str, *args, **kwargs):
 
         self.client = AzureOpenAI(*args, **kwargs)
         self.model_name = deployment_name
+        self.model_type = "api"
+        self.type_adapter = OpenAITypeAdapter()
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 44a176084..1d5b5c533 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -48,6 +48,13 @@ def test_anthropic_simple_call():
     assert isinstance(result, str)
 
 
+@pytest.mark.api_call
+def test_anthropic_direct_call():
+    model = Anthropic(MODEL_NAME)
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
 @pytest.mark.api_call
 def test_anthropic_simple_vision():
     model = Anthropic(MODEL_NAME)
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 4718b12d0..fb7a79def 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -33,6 +33,13 @@ def test_gemini_simple_call():
     assert isinstance(result, str)
 
 
+@pytest.mark.api_call
+def test_gemini_direct_call():
+    model = Gemini(MODEL_NAME)
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
 @pytest.mark.api_call
 def test_gemini_simple_vision():
     model = Gemini(MODEL_NAME)
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index ffb0110cd..e9af3cd2a 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -31,6 +31,11 @@ def test_llamacpp_simple(model):
     assert isinstance(result, str)
 
 
+def test_llamacpp_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
 def test_llamacpp_regex(model):
     regex_str = Regex(r"[0-9]").to_regex()
     logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index d2b8d1a40..d7d074573 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -66,6 +66,13 @@ def test_openai_simple_call():
     assert isinstance(result, str)
 
 
+@pytest.mark.api_call
+def test_openai_direct_call():
+    model = OpenAI(MODEL_NAME)
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
 @pytest.mark.api_call
 def test_openai_simple_vision():
     model = OpenAI(MODEL_NAME)

From 5b962f50b3a565e1427f10d6cd782e6695c25eda Mon Sep 17 00:00:00 2001
From: Robin Picard <robinpicard@device-36.home>
Date: Sun, 16 Feb 2025 21:24:27 +0100
Subject: [PATCH 216/505] Create the Ollama model

---
 .github/workflows/tests.yml     |  4 ++
 docs/reference/models/ollama.md | 72 +++++++++++++++++++++++++
 outlines/models/__init__.py     |  7 ++-
 outlines/models/ollama.py       | 95 +++++++++++++++++++++++++++++++++
 pyproject.toml                  |  6 +++
 tests/models/test_ollama.py     | 76 ++++++++++++++++++++++++++
 6 files changed, 258 insertions(+), 2 deletions(-)
 create mode 100644 docs/reference/models/ollama.md
 create mode 100644 outlines/models/ollama.py
 create mode 100644 tests/models/test_ollama.py

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 796f7cc89..0565d0f3c 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -29,6 +29,10 @@ jobs:
       uses: actions/setup-python@v4
       with:
         python-version: ${{ matrix.python-version }}
+    - name: Install Ollama
+      run: |
+        curl -fsSL https://ollama.com/install.sh | sh
+        ollama --version
     - name: Set up test environment
       run: |
         python -m pip install --upgrade pip
diff --git a/docs/reference/models/ollama.md b/docs/reference/models/ollama.md
new file mode 100644
index 000000000..0d2961832
--- /dev/null
+++ b/docs/reference/models/ollama.md
@@ -0,0 +1,72 @@
+# Ollama
+
+!!! Installation
+
+    To be able to use Ollama in Outlines, you must install both Ollama and the `ollama` python sdk.
+
+    - To download Ollama: https://ollama.com/download
+    - To install the ollama python sdk: `pip install ollama`
+
+## Ollama models
+
+You must provide a model name when instantiating the `outlines.models.Ollama` class. This model must be available on your system.
+```python
+from outlines.models import Ollama
+
+model = Ollama("tinyllama")
+```
+
+To download a new model from the Ollama model hub, you can use the following command (it will return an `Ollama` instance):
+```python
+from outlines.models import Ollama
+
+Ollama.from_pretrained("llama3.1:8b")
+```
+
+You can find the list of available models on the [Ollama library](https://ollama.com/library).
+
+## Generate text
+
+As with other models, you can either first create a `Generator` object and then call it
+```python
+from outlines.models import Ollama
+from outlines.generate import Generator
+
+model = Ollama("tinyllama")
+generator = Generator(model)
+answer = generator("Write a sentence about a cat.")
+```
+or directly call the model
+```python
+from outlines.models import Ollama
+
+model = Ollama("tinyllama")
+answer = model("Write a sentence about a cat.")
+```
+
+The input of the generation must be a string. Batch generation is not supported.
+The only output type supported is `Json`.
+```python
+from outlines.models import Ollama
+from outlines.types import Json
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+
+model = Ollama("tinyllama")
+answer = model("Create a character.", output_type=Json(Character))
+```
+
+You can also stream the tokens:
+```python
+from outlines.models import Ollama
+
+model = Ollama("tinyllama")
+tokens = model.stream("Write a sentence about a cat.")
+```
+
+## Optional parameters
+
+You can provide the same optional parameters you would pass to the `ollama` sdk's client both during the initialization of the `Ollama` class and when generating text.
+Consult the [ollama python sdk Github repository](https://github.com/ollama/ollama-python) for the full list of parameters.
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 761de9cc5..ea228c5c6 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -14,12 +14,15 @@
 from .gemini import Gemini
 from .llamacpp import LlamaCpp
 from .mlxlm import MLXLM, mlxlm
+from .ollama import Ollama
 from .openai import AzureOpenAI, OpenAI
 from .transformers import Transformers, TransformerTokenizer, mamba, transformers
 from .transformers_vision import TransformersVision, transformers_vision
 from .vllm import VLLM, vllm
 
-LogitsGenerator = Union[Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM]
+LogitsGenerator = Union[
+    Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM, Ollama
+]
 
 LocalModel = LlamaCpp
-APIModel = Union[AzureOpenAI, OpenAI, Anthropic, Gemini]
+APIModel = Union[AzureOpenAI, OpenAI, Anthropic, Gemini, Ollama]
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
new file mode 100644
index 000000000..c97460200
--- /dev/null
+++ b/outlines/models/ollama.py
@@ -0,0 +1,95 @@
+from functools import singledispatchmethod
+from types import NoneType
+from typing import Iterator
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.types import Json
+
+
+class OllamaTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the Ollama model."""
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. "
+            "Ollama does not support batch inference."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        return model_input
+
+    @singledispatchmethod
+    def format_output_type(self, output_type):
+        """Generate the `format` argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The output type passed by the user.
+        """
+        raise NotImplementedError(
+            f"The output type {input} is not available. "
+            "Ollama only supports structured output with `Json`."
+        )
+
+    @format_output_type.register(NoneType)
+    def format_none_output_type(self, output_type: None):
+        return ""
+
+    @format_output_type.register(Json)
+    def format_json_output_type(self, output_type: Json):
+        return output_type.to_json_schema()
+
+
+class Ollama(Model):
+    """Thin wrapper around the `ollama` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `ollama` client.
+
+    """
+
+    def __init__(self, model_name: str, *args, **kwargs):
+        from ollama import Client
+
+        self.client = Client(*args, **kwargs)
+        self.model_name = model_name
+        self.type_adapter = OllamaTypeAdapter()
+
+    @classmethod
+    def from_pretrained(cls, model_name: str, *args, **kwargs):
+        """Download the model weights from Ollama and create a `Ollama` instance."""
+        from ollama import pull
+
+        pull(model_name)
+        return cls(model_name, *args, **kwargs)
+
+    def generate(self, model_input, output_type=None, **kwargs) -> str:
+        response = self.client.generate(
+            model=self.model_name,
+            prompt=self.type_adapter.format_input(model_input),
+            format=self.type_adapter.format_output_type(output_type),
+            **kwargs,
+        )
+        return response.response
+
+    def stream(self, model_input, output_type=None, **kwargs) -> Iterator[str]:
+        response = self.client.generate(
+            model=self.model_name,
+            prompt=self.type_adapter.format_input(model_input),
+            format=self.type_adapter.format_output_type(output_type),
+            stream=True,
+            **kwargs,
+        )
+        for chunk in response:
+            yield chunk.response
diff --git a/pyproject.toml b/pyproject.toml
index 3a09fc84d..6ad04f601 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -55,6 +55,7 @@ anthropic = ["anthropic"]
 gemini = ["google-generativeai"]
 llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
 exllamav2 = ["exllamav2"]
+ollama = ["ollama"]
 test = [
     "pre-commit",
     "pytest",
@@ -77,6 +78,7 @@ test = [
     "pillow",
     "exllamav2",
     "jax",
+    "ollama"
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
 serve = [
@@ -156,7 +158,11 @@ module = [
     "iso3166.*",
     "airportsdata.*",
     "outlines_core.*",
+<<<<<<< HEAD
     "genson",
+=======
+    "ollama.*",
+>>>>>>> 5af6b93 (Create the Ollama model)
 ]
 ignore_missing_imports = true
 
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
new file mode 100644
index 000000000..ed05f3e21
--- /dev/null
+++ b/tests/models/test_ollama.py
@@ -0,0 +1,76 @@
+import json
+from enum import Enum
+
+import pytest
+from pydantic import BaseModel
+
+from outlines.models import Ollama
+from outlines.types import Choice, Json
+
+MODEL_NAME = "tinyllama"
+
+
+def test_pull_model():
+    model = Ollama.from_pretrained(MODEL_NAME)
+    assert isinstance(model, Ollama)
+
+
+def test_ollama_wrong_init_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        Ollama(MODEL_NAME, foo=10)
+
+
+def test_wrong_inference_parameters():
+    with pytest.raises(TypeError, match="got an unexpected"):
+        Ollama(MODEL_NAME).generate("Respond with one word. Not more.", None, foo=10)
+
+
+def test_ollama_simple():
+    result = Ollama(MODEL_NAME).generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_ollama_direct():
+    result = Ollama(MODEL_NAME)("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_ollama_json():
+    class Foo(BaseModel):
+        foo: str
+
+    result = Ollama(MODEL_NAME)("Respond with one word. Not more.", Json(Foo))
+    assert isinstance(result, str)
+    assert "foo" in json.loads(result)
+
+
+def test_ollama_wrong_output_type():
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    with pytest.raises(NotImplementedError, match="is not available"):
+        Ollama(MODEL_NAME).generate("foo?", Choice(Foo))
+
+
+def test_ollama_wrong_input_type():
+    with pytest.raises(NotImplementedError, match="is not available"):
+        Ollama(MODEL_NAME).generate(["foo?", "bar?"], None)
+
+
+def test_ollama_stream():
+    model = Ollama(MODEL_NAME)
+    generator = model.stream("Write a sentence about a cat.")
+    assert isinstance(next(generator), str)
+
+
+def test_ollama_stream_json():
+    class Foo(BaseModel):
+        foo: str
+
+    model = Ollama(MODEL_NAME)
+    generator = model.stream("Create a character.", Json(Foo))
+    generated_text = []
+    for text in generator:
+        generated_text.append(text)
+    assert "foo" in json.loads("".join(generated_text))

From 8734560210f89bdd77746bb6c8c878fa28589048 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Fri, 21 Feb 2025 22:36:52 +0100
Subject: [PATCH 217/505] Refactor the Transformers and TransformersVision
 model (#1430)

The objective of this commit is to adapt `Transfomers` and `TransformersVision` to the
model interface used in v1.0. The main changes are:
- make the models inherit from the base class `Model` and create a
`ModelTypeAdapter` for them
- modify the signature of the `generate` method and make it the only way of
generating text with the model.
- modify the handling of model/tokenizer optional arguments to avoid manipulating
them ourselves (giving to the user the responsibility of choosing those they need).
- delete the tests for those models in the `generate` directory and test them instead
in the `models` directory
---
 docs/reference/models/transformers.md         | 150 ++---
 docs/reference/models/transformers_vision.md  | 128 ++--
 outlines/models/__init__.py                   |   6 +-
 outlines/models/transformers.py               | 406 ++++---------
 outlines/models/transformers_vision.py        | 240 ++++----
 outlines/types/__init__.py                    |   2 +-
 outlines/types/dsl.py                         |   3 +
 pyproject.toml                                |   3 -
 tests/generate/test_generate.py               |  49 +-
 .../generate/test_integration_transformers.py | 569 ------------------
 .../test_integration_transformers_vision.py   | 116 ----
 tests/models/test_transformers.py             | 246 ++++----
 tests/models/test_transformers_vision.py      | 235 ++++++++
 13 files changed, 794 insertions(+), 1359 deletions(-)
 delete mode 100644 tests/generate/test_integration_transformers.py
 delete mode 100644 tests/generate/test_integration_transformers_vision.py
 create mode 100644 tests/models/test_transformers_vision.py

diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 3649b9b13..f24591436 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -1,96 +1,63 @@
-# transformers
+# Transformers
 
 
 !!! Installation
 
-    You need to install the `transformer`, `datasets` and `torch` libraries to be able to use these models in Outlines, or alternatively:
+    You need to install the `transformer` library to be able to use these models in Outlines, or alternatively:
 
     ```bash
     pip install "outlines[transformers]"
     ```
 
+## Create a `Transformers` model
 
-Outlines provides an integration with the `torch` implementation of causal models in the [transformers][transformers] library. You can initialize the model by passing its name:
-
+The only mandatory argument to instantiate a `Transformers` model is the name of the model to use.
 ```python
 from outlines import models
 
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+model = models.Transformers("microsoft/Phi-3-mini-4k-instruct")
 ```
 
-If you need more fine-grained control you can also initialize the model and tokenizer separately:
+The model name must be a valid `transformers` model name. You can find a list of all in the HuggingFace library [here](https://huggingface.co/models).
+
+When instantiating a `Transformers` model as such, the class creates a model from the transformers libray using the class `AutoModelForCausalLM` by default (`transformers.AutoModelForCausalLM.from_pretrained(model_name, **model_kwargs)`).
 
+You can also provide keyword arguments in an optional `model_kwargs` parameter. Those will be passed to the `from_pretrained` method of the model class. One such argument is `device_map`, which allows you to specify the device on which the model will be loaded.
 
+For instance:
 ```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
 from outlines import models
 
-llm = AutoModelForCausalLM.from_pretrained("gpt2", output_attentions=True)
-tokenizer = AutoTokenizer.from_pretrained("gpt2")
-model = models.Transformers(llm, tokenizer)
+model = models.Transformers("microsoft/Phi-3-mini-4k-instruct", model_kwargs={"device_map": "cuda"})
 ```
 
-## Using Logits Processors
-
-There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
+## Alternative model classes
 
-1. Use Outlines generation wrapper, `outlines.models.transformers`
-2. Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
-
-Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
-
-### Using `outlines.models.transformers`
+If the model you want to use is not compatible with `AutoModelForCausalLM`, you must provide a value for the `model_class` parameter. This value must be a valid `transformers` model class.
 
+For instance:
 ```python
-import outlines
-
-time_regex_pattern = r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?"
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
-generator = outlines.generate.regex(model, time_regex_pattern)
+from outlines import models
+from transformers import AutoModelForSeq2SeqLM
 
-output = generator("The the best time to visit a dentist is at ")
-print(output)
-# 2:30 pm
+model = models.Transformers("facebook/bart-large", model_class=AutoModelForSeq2SeqLM)
 ```
 
-### Using models initialized via the `transformers`  library
-
-```python
-import outlines
-import transformers
-
+When you instantiate a `Transformers` model, the class also creates a `Tokenizer` instance from the `AutoTokenizer` class. You can provide keyword arguments in an optional `tokenizer_kwargs` parameter. Those will be passed to the `from_pretrained` method of the tokenizer class as such: `tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)`.
 
-model_uri = "microsoft/Phi-3-mini-4k-instruct"
+Similarly, if your model is not compatible with `AutoTokenizer`, you must provide a value for the `tokenizer_class` parameter.
 
-outlines_tokenizer = outlines.models.TransformerTokenizer(
-    transformers.AutoTokenizer.from_pretrained(model_uri)
-)
-phone_number_logits_processor = outlines.processors.RegexLogitsProcessor(
-    "\\+?[1-9][0-9]{7,14}",  # phone number pattern
-    outlines_tokenizer,
-)
-
-generator = transformers.pipeline('text-generation', model=model_uri)
+```python
+from outlines import models
+from transformers import T5ForConditionalGeneration, T5Tokenizer
 
-output = generator(
-    "Jenny gave me her number it's ",
-	logits_processor=transformers.LogitsProcessorList([phone_number_logits_processor])
+model_pile_t5 = models.Transformers(
+    model_name="EleutherAI/pile-t5-large",
+    model_class=T5ForConditionalGeneration,
+    tokenizer_class=T5Tokenizer
 )
-print(output)
-# [{'generated_text': "Jenny gave me her number it's 2125550182"}]
-# not quite 8675309 what we expected, but it is a valid phone number
 ```
 
-[transformers]: https://github.com/huggingface/transformers
-
-
-## Alternative Model Classes
-
-`outlines.models.transformers` defaults to `transformers.AutoModelForCausalLM`, which is the appropriate class for most standard large language models, including Llama 3, Mistral, Phi-3, etc.
-
-However other variants with unique behavior can be used as well by passing the appropriate class.
-
 ### Mamba
 
 [Mamba](https://github.com/state-spaces/mamba) is a transformers alternative which employs memory efficient, linear-time decoding.
@@ -100,25 +67,14 @@ To use Mamba with outlines you must first install the necessary requirements:
 pip install causal-conv1d>=1.2.0 mamba-ssm torch transformers
 ```
 
-Then you can either create an Mamba-2 Outlines model via
-```python
-import outlines
-
-model = outlines.models.mamba("state-spaces/mamba-2.8b-hf")
-```
-
-or explicitly with
+Then you can create an `Mamba` Outlines model via:
 ```python
-import outlines
-from transformers import MambaForCausalLM
+from outlines import models
 
-model = outlines.models.transformers(
-    "state-spaces/mamba-2.8b-hf",
-    model_class=MambaForCausalLM
-)
+model = models.Mamba("state-spaces/mamba-2.8b-hf", model_kwargs={"device_map": "cuda"}, tokenizer_kwargs={"padding_side": "left"})
 ```
 
-
+Alternatively, you can use the `Transformers` class to create an `Mamba` model by providing the appropriate `model_class` and `tokenizer_class` arguments.
 
 Read [`transformers`'s documentation](https://huggingface.co/docs/transformers/en/model_doc/mamba) for more information.
 
@@ -128,21 +84,43 @@ You can use encoder-decoder (seq2seq) models like T5 and BART with Outlines.
 
 Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
 
-T5 Example:
+## Use the model to generate text
+
+Once you have created a `Transformers` model, you can use it to generate text by calling the instance of the model.
 ```python
-import outlines
-from transformers import AutoModelForSeq2SeqLM
+model("Hello, how are you?")
+```
 
-model_pile_t5 = outlines.models.transformers(
-    model_name="EleutherAI/pile-t5-large",
-    model_class=AutoModelForSeq2SeqLM,
-)
+You can also first create a `Generator` and then call it.
+```python
+from outlines import Generator
+
+generator = Generator(model)
+generator("Hello, how are you?")
 ```
 
-Bart Example:
+`Transformers` models typically support batching and the generation of several samples at once.
+
+For instance:
 ```python
-model_bart = outlines.models.transformers(
-    model_name="facebook/bart-large",
-    model_class=AutoModelForSeq2SeqLM,
-)
+model(["Hello, how are you?", "Respond with one word. Not more."], num_return_sequences=2, num_beams=2)
+```
+
+This would generate two sequences for each prompt, for a total of four sequences (two lists of 2 elements each in a list).
+
+## Use the model to generate structured data
+
+`Transformers` models can generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
+
+Supported types include `Json`, `Choice`, `Regex` and `CFG`.
+
+For instance:
+```python
+from outlines.types import Json
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+
+model("Create a character with a name.", Json(Character))
 ```
diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
index 451e54a8f..03ba6a5c9 100644
--- a/docs/reference/models/transformers_vision.md
+++ b/docs/reference/models/transformers_vision.md
@@ -2,33 +2,28 @@
 
 Outlines allows seamless use of [vision models](https://huggingface.co/learn/computer-vision-course/en/unit4/multimodal-models/tasks-models-part1).
 
-`outlines.models.transformers_vision` shares interfaces with, and is based on [outlines.models.transformers](./transformers.md).
+`outlines.models.Transformers_vision` shares interfaces with, and is based on [outlines.models.Transformers](./transformers.md).
 
-Tasks supported include
+## Create a `TransformersVision` model
 
-- image + text -> text
-- video + text -> text
+`TransformersVision` models inherit from `Transformers` and accept the same initialization parameters.
 
+In addition, they also accept the optional parameters `processor_class` and `processor_kwargs`. Those are used to create a `Processor` instance that is then used to preprocess the images. By default, `AutoProcessor` is used to create the processor as such: `AutoProcessor.from_pretrained(model_name, **processor_kwargs)`.
 
+If your model is not compatible with `AutoProcessor`, you must provide a value for the `processor_class` parameter.
+For instance:
+```python
+from outlines import models
+from transformers import CLIPModel, CLIPProcessor
 
-## Example: Using [Llava-Next](https://huggingface.co/docs/transformers/en/model_doc/llava_next) Vision Models
+model = models.TransformersVision("openai/clip-vit-base-patch32", model_class=CLIPModel, processor_class=CLIPProcessor)
+```
 
-Install dependencies
-`pip install torchvision pillow flash-attn`
+## Use the model to generate text from prompts and images
 
-Create the model
-```python
-import outlines
-from transformers import LlavaNextForConditionalGeneration
-
-model = outlines.models.transformers_vision(
-    "llava-hf/llava-v1.6-mistral-7b-hf",
-    model_class=LlavaNextForConditionalGeneration,
-	device="cuda",
-)
-```
+When calling the model, the prompt argument you provide must be a dictionary with a key `"prompts"` and a key `"images"`. The associated values must be a string or a list of strings for the prompts, and a PIL image or a list of PIL images for the images. Your prompts must include `<image>` tags tokens to indicate where the image should be inserted. There must be as many `<image>` tags as there are images.
 
-Create convenience function to load a `PIL.Image` from URL
+For easier use, we recommend you to create a convenience function to load a `PIL.Image` from URL.
 ```python
 from PIL import Image
 from io import BytesIO
@@ -39,56 +34,65 @@ def img_from_url(url):
     return Image.open(img_byte_stream).convert("RGB")
 ```
 
-### Describing an image
-
+You can then call the model with your prompts and images to generate text.
 ```python
-description_generator = outlines.generate.text(model)
-description_generator(
-    "<image> detailed description:",
-    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")]
-)
+from transformers import LlavaForConditionalGeneration
+from outlines import models
+
+model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
+prompt = {
+    "prompts": "<image> detailed description:",
+    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
+}
+model(prompt)
 ```
 
-> This is a color photograph featuring a Siamese cat with striking blue eyes. The cat has a creamy coat and a light eye color, which is typical for the Siamese breed. Its features include elongated ears, a long, thin tail, and a striking coat pattern. The cat is sitting in an indoor setting, possibly on a cat tower or a similar raised platform, which is covered with a beige fabric, providing a comfortable and soft surface for the cat to rest or perch. The surface of the wall behind the cat appears to be a light-colored stucco or plaster.
-
-#### Multiple Images
-
-To include multiple images in your prompt you simply add more `<image>` tokens to the prompt
-
+You can include several images per prompt by adding more `<image>` tags to the prompt. Batching is also supported.
 ```python
-image_urls = [
-	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-3.png",  # triangle
-	"https://cdn1.byjus.com/wp-content/uploads/2020/08/ShapeArtboard-1-copy-11.png",  # hexagon
-]
-description_generator = outlines.generate.text(model)
-description_generator(
-    "<image><image>What shapes are present?",
-    list(map(img_from_url, image_urls)),
-)
+from transformers import LlavaForConditionalGeneration
+from outlines import models
+
+model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
+prompt = {
+    "prompts": ["<image><image>detailed description:", "<image><image>. What animals are present?"],
+    "images": [
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
+    ]
+}
+model(prompt)
 ```
 
-> There are two shapes present. One shape is a hexagon and the other shape is an triangle. '
+Here we have two prompts, each expecting two images. We correspondingly provide four images. This will generate two descriptions, one for each prompt.
+
+### Use the model for structured generation
 
+You can use the model to generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
 
-### Classifying an Image
+Supported types include `Json`, `Choice`, `Regex` and `CFG`.
 
+For instance to do classification, you can use the `Regex` type:
 ```python
-pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
-planet_generator = outlines.generate.regex(model, pattern)
+from outlines import models
+from outlines.types import Regex
+from transformers import LlavaForConditionalGeneration
 
-planet_generator(
-    "What planet is this: <image>",
-    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg")]
-)
+model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
+pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
+prompt = {
+    "prompts": "<image>detailed description:",
+    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
+}
+model(prompt, Regex(pattern))
 ```
 
-> Saturn
-
-
-### Extracting Structured Image data
-
+Another example could be to generated a structured description of an image using the `Json` type:
 ```python
+from outlines import models
 from pydantic import BaseModel
+from transformers import LlavaForConditionalGeneration
 from typing import List, Optional
 
 class ImageData(BaseModel):
@@ -97,17 +101,15 @@ class ImageData(BaseModel):
     object_list: List[str]
     is_photo: bool
 
-image_data_generator = outlines.generate.json(model, ImageData)
-
-image_data_generator(
-    "<image> detailed JSON metadata:",
-    [img_from_url("https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg")]
-)
+model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
+pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
+prompt = {
+    "prompts": "<image>detailed description:",
+    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
+}
+model(prompt, Json(ImageData))
 ```
 
-> `ImageData(caption='An astronaut on the moon', tags_list=['moon', 'space', 'nasa', 'americanflag'], object_list=['moon', 'moon_surface', 'space_suit', 'americanflag'], is_photo=True)`
-
-
 ## Resources
 
 ### Choosing a model
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index ea228c5c6..4254e7cf6 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -16,13 +16,13 @@
 from .mlxlm import MLXLM, mlxlm
 from .ollama import Ollama
 from .openai import AzureOpenAI, OpenAI
-from .transformers import Transformers, TransformerTokenizer, mamba, transformers
-from .transformers_vision import TransformersVision, transformers_vision
+from .transformers import Mamba, Transformers, TransformerTokenizer
+from .transformers_vision import TransformersVision
 from .vllm import VLLM, vllm
 
 LogitsGenerator = Union[
     Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM, Ollama
 ]
 
-LocalModel = LlamaCpp
+LocalModel = Union[LlamaCpp, Transformers]
 APIModel = Union[AzureOpenAI, OpenAI, Anthropic, Gemini, Ollama]
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 6b204ec6e..3d87443f6 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,17 +1,14 @@
-import dataclasses
-import inspect
-from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, Union
+from functools import singledispatchmethod
+from typing import TYPE_CHECKING, List, Tuple, Union
 
-from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
 
 if TYPE_CHECKING:
     import torch
-    from transformers import PreTrainedModel, PreTrainedTokenizer
+    from transformers import PreTrainedTokenizer
 
-    from outlines.processors import OutlinesLogitsProcessor
-
-__all__ = ["transformers"]
+__all__ = ["Transformers", "Mamba"]
 
 
 KVCacheType = Tuple[Tuple["torch.DoubleTensor", "torch.DoubleTensor"], ...]
@@ -126,230 +123,130 @@ def __setstate__(self, state):
         self.__init__(state["tokenizer"])
 
 
-class Transformers:
-    """Represents a `transformers` model."""
+class TransformersTypeAdapter(ModelTypeAdapter):
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
 
-    def __init__(
-        self,
-        model: "PreTrainedModel",
-        tokenizer: "PreTrainedTokenizer",
-    ):
-        self.model = model
-        self.tokenizer = TransformerTokenizer(tokenizer)
-
-    def forward(
-        self,
-        input_ids: "torch.LongTensor",
-        attention_mask: "torch.LongTensor",
-        past_key_values: Optional[Tuple] = None,
-    ) -> Tuple["torch.FloatTensor", Optional[KVCacheType]]:
-        """Compute a forward pass through the transformer model.
-
-        Parameters
-        ----------
-        input_ids
-            The input token ids.  Must be one or two dimensional.
-        attention_mask
-            The attention mask.  Must be one or two dimensional.
-        past_key_values
-            A tuple of tuples containing the cached key and value tensors for each
-            attention head.
-
-        Returns
-        -------
-        The computed logits and the new cached key and value tensors.
+        Argument
+        --------
+        model_input
+            The input passed by the user.
 
         """
-        try:
-            import torch
-        except ImportError:
-            ImportError(
-                "The `torch` library needs to be installed to use `transformers` models."
-            )
-        assert 0 < input_ids.ndim < 3
-
-        if past_key_values:
-            input_ids = input_ids[..., -1].unsqueeze(-1)
-
-        with torch.inference_mode():
-            output = self.model(
-                input_ids,
-                attention_mask=attention_mask,
-                return_dict=True,
-                output_attentions=False,
-                output_hidden_states=False,
-                past_key_values=past_key_values,
-            )
+        raise NotImplementedError(
+            f"The input type {input} is not available."
+            "Please use a string or a list of strings."
+        )
 
-        return output.logits, output.past_key_values
+    @format_input.register(str)
+    def format_str_input(self, model_input):
+        return model_input
 
-    def __call__(
-        self,
-        input_ids: "torch.LongTensor",
-        attention_mask: "torch.LongTensor",
-        past_key_values: Optional[Tuple] = None,
-    ) -> "torch.FloatTensor":
-        logits, kv_cache = self.forward(input_ids, attention_mask, past_key_values)
-        next_token_logits = logits[..., -1, :]
+    @format_input.register(list)
+    def format_list_input(self, model_input):
+        return model_input
 
-        return next_token_logits, kv_cache
+    def format_output_type(self, output_type):
+        """Generate the logits processor argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The logits processor provided.
 
-    def generate(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Union[str, List[str], List[List[str]]]:
-        """Generate text using `transformers`.
-
-        Parameters
-        ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-
-        Returns
-        -------
-        The generated text
         """
-        if isinstance(prompts, str):
-            # convert to 2d
-            input_ids, attention_mask = self.tokenizer.encode([prompts])
-        else:
-            input_ids, attention_mask = self.tokenizer.encode(prompts)
+        from transformers import LogitsProcessorList
 
-        inputs = {
-            "input_ids": input_ids.to(self.model.device),
-            "attention_mask": attention_mask.to(self.model.device),
-        }
-        if (
-            "attention_mask"
-            not in inspect.signature(self.model.forward).parameters.keys()
-        ):
-            del inputs["attention_mask"]
-
-        generation_kwargs = self._get_generation_kwargs(
-            prompts,
-            generation_parameters,
-            logits_processor,
-            sampling_parameters,
-        )
-        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
+        if output_type is not None:
+            return LogitsProcessorList([output_type])
+        return None
 
-        # if single str input and single sample per input, convert to a 1D output
-        if isinstance(prompts, str):
-            generated_ids = generated_ids.squeeze(0)
 
-        return self._decode_generation(generated_ids)
+class Transformers(Model):
+    """Represents a `transformers` model."""
 
-    def stream(
+    def __init__(
         self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Iterator[Union[str, List[str]]]:
-        """
-        Temporary stream stand-in which implements stream() signature
-        and equivalent behaviour but isn't yielded until generation completes.
+        model_name: str,
+        model_class=None,
+        model_kwargs: dict = {},
+        tokenizer_class=None,
+        tokenizer_kwargs: dict = {},
+    ):
+        """Create a Transformers model instance
+
+        Parameters:
+        ----------
+        model_name
+            The name of the transformers model to use;
+        model_class
+            The Transformers model class from which to create the model.
+            If not provided,`AutoModelForCausalLM` will be used.
+            If you gave the name of a non-causal language model,
+            you must provide a value for this parameter.
+        model_kwargs
+            A dictionary of keyword arguments to pass to the `from_pretrained`
+            method of the model class.
+        tokenizer_class
+            The Transformers tokenizer class from which to create the tokenizer.
+            If not provided,`AutoTokenizer` will be used.
+            If you gave the name of a model that is not compatible with `AutoTokenizer`,
+            you must provide a value for this parameter.
+        tokenizer_kwargs
+            A dictionary of keyword arguments to pass to the `from_pretrained`
+            method of the tokenizer class.
 
-        TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
         """
-        if isinstance(prompts, str):
-            # convert to 2d
-            input_ids, attention_mask = self.tokenizer.encode([prompts])
-        else:
-            input_ids, attention_mask = self.tokenizer.encode(prompts)
+        if model_class is None or tokenizer_class is None:
+            try:
+                from transformers import AutoModelForCausalLM, AutoTokenizer
+            except ImportError:
+                raise ImportError(
+                    "The `transformers` library needs to be installed in order to use `transformers` models."
+                )
+        if model_class is None:
+            model_class = AutoModelForCausalLM
+        if tokenizer_class is None:
+            tokenizer_class = AutoTokenizer
+        self.model = model_class.from_pretrained(model_name, **model_kwargs)
+        tokenizer_kwargs.setdefault("padding_side", "left")
+        self.tokenizer = TransformerTokenizer(
+            tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
+        )
+        self.type_adapter = TransformersTypeAdapter()
+
+    def generate(self, model_input, output_type, **inference_kwargs):
+        prompts = self.type_adapter.format_input(model_input)
+        input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
             "input_ids": input_ids.to(self.model.device),
             "attention_mask": attention_mask.to(self.model.device),
         }
-        if (
-            "attention_mask"
-            not in inspect.signature(self.model.forward).parameters.keys()
-        ):
-            del inputs["attention_mask"]
-
-        generation_kwargs = self._get_generation_kwargs(
-            prompts,
-            generation_parameters,
-            logits_processor,
-            sampling_parameters,
+
+        logits_processor = self.type_adapter.format_output_type(output_type)
+
+        generated_ids = self._generate_output_seq(
+            prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )
-        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
 
-        # if single str input and single sample per input, convert to a 1D output
-        if isinstance(prompts, str):
+        # if single str input, convert to a 1D outputt
+        if isinstance(model_input, str):
             generated_ids = generated_ids.squeeze(0)
 
-        for i in range(generated_ids.size(-1)):
-            output_group_ids = generated_ids.select(-1, i).unsqueeze(-1)
-            yield self._decode_generation(output_group_ids)
+        return self._decode_generation(generated_ids)
 
-    def _get_generation_kwargs(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> dict:
+    def stream(self, model_input, output_type, **inference_kwargs):
         """
-        Convert outlines generation parameters into model.generate kwargs
+        TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
         """
-        from transformers import GenerationConfig, LogitsProcessorList, set_seed
-
-        max_new_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
-        )
-        if max_new_tokens is None:
-            max_new_tokens = int(2**30)
-
-        # global seed, not desirable
-        if seed is not None:
-            set_seed(seed)
-
-        if logits_processor is not None:
-            logits_processor_list = LogitsProcessorList([logits_processor])
-        else:
-            logits_processor_list = None
-
-        generation_config = GenerationConfig(
-            max_new_tokens=max_new_tokens,
-            stop_strings=stop_at,
-            num_return_sequences=(num_samples or 1),
-            top_p=top_p,
-            top_k=top_k,
-            temperature=temperature,
-            do_sample=(sampler == "multinomial"),
-            num_beams=(num_samples if sampler == "beam_search" else 1),
-            eos_token_id=self.tokenizer.eos_token_id,
-            pad_token_id=self.tokenizer.pad_token_id,
+        raise NotImplementedError(
+            "Streaming is not implemented for Transformers models."
         )
 
-        return dict(
-            logits_processor=logits_processor_list,
-            generation_config=generation_config,
-            tokenizer=self.tokenizer.tokenizer,
-        )
-
-    def _generate_output_seq(
-        self, prompts, inputs, generation_config, **generation_kwargs
-    ):
+    def _generate_output_seq(self, prompts, inputs, **inference_kwargs):
         input_ids = inputs["input_ids"]
-        output_ids = self.model.generate(
-            **inputs, generation_config=generation_config, **generation_kwargs
-        )
+        output_ids = self.model.generate(**inputs, **inference_kwargs)
 
         # encoder-decoder returns output_ids only, decoder-only returns full seq ids
         if self.model.config.is_encoder_decoder:
@@ -358,12 +255,10 @@ def _generate_output_seq(
             generated_ids = output_ids[:, input_ids.shape[1] :]
 
         # if batch list inputs AND multiple samples per input, convert generated_id to 3D view
-        num_samples = generation_config.num_return_sequences or 1
-
+        num_samples = inference_kwargs.get("num_return_sequences", 1)
         if num_samples > 1 and isinstance(prompts, list):
             batch_size = input_ids.size(0)
-            num_return_sequences = generation_config.num_return_sequences or 1
-            generated_ids = generated_ids.view(batch_size, num_return_sequences, -1)
+            generated_ids = generated_ids.view(batch_size, num_samples, -1)
 
         return generated_ids
 
@@ -383,76 +278,41 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
             )
 
 
-def transformers(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-    model_class=None,
-    tokenizer_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
-
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    tokenizer_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the tokenizer.
-
-    Returns
-    -------
-    A `TransformersModel` model instance.
+class Mamba(Transformers):
+    """Represents a Mamba model."""
 
-    """
-    if model_class is None or tokenizer_class is None:
+    def __init__(
+        self,
+        model_name: str,
+        model_kwargs: dict = {},
+        tokenizer_kwargs: dict = {},
+    ):
+        """
+        Create a Mamba model instance
+
+        Parameters:
+        ----------
+        model_name
+            The name of the transformers model to use. It will be passed to
+            the `from_pretrained` method of the `MambaForCausalLM` class.
+        model_kwargs
+            A dictionary of keyword arguments to pass to the `from_pretrained`
+            method of the `MambaForCausalLM` class.
+        tokenizer_kwargs
+            A dictionary of keyword arguments to pass to the `from_pretrained`
+            method of the `AutoTokenizer` class.
+        """
         try:
-            from transformers import AutoModelForCausalLM, AutoTokenizer
+            from transformers import MambaForCausalLM
+
         except ImportError:
             raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
+                "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
             )
-    if model_class is None:
-        model_class = AutoModelForCausalLM
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
-
-    if device is not None:
-        model_kwargs["device_map"] = device
-
-    model = model_class.from_pretrained(model_name, **model_kwargs)
 
-    tokenizer_kwargs.setdefault("padding_side", "left")
-    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
-
-    return Transformers(model, tokenizer)
-
-
-def mamba(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-):
-    try:
-        from transformers import MambaForCausalLM
-
-    except ImportError:
-        raise ImportError(
-            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
+        return super().__init__(
+            model_name=model_name,
+            model_class=MambaForCausalLM,
+            model_kwargs=model_kwargs,
+            tokenizer_kwargs=tokenizer_kwargs,
         )
-
-    return transformers(
-        model_name=model_name,
-        device=device,
-        model_kwargs=model_kwargs,
-        tokenizer_kwargs=tokenizer_kwargs,
-        model_class=MambaForCausalLM,
-    )
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
index c8a86536e..b1c1c1de4 100644
--- a/outlines/models/transformers_vision.py
+++ b/outlines/models/transformers_vision.py
@@ -1,138 +1,146 @@
-from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
+from functools import singledispatchmethod
 
-from outlines.generate.api import GenerationParameters, SamplingParameters
 from outlines.models import Transformers
+from outlines.models.base import ModelTypeAdapter
 
-if TYPE_CHECKING:
-    from outlines.processors import OutlinesLogitsProcessor
+
+class TransformersVisionTypeAdapter(ModelTypeAdapter):
+    """Type adapter for TransformersVision models."""
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. Please provide a "
+            "dictionary with the following format: "
+            "{'prompts': Union[str, List[str]], 'images': Union[Any, List[Any]]}"
+            "Make sure the number of image tags in the prompts is equal to the "
+            "number of images provided."
+        )
+
+    @format_input.register(dict)
+    def format_list_input(self, model_input):
+        if "prompts" not in model_input or "images" not in model_input:
+            raise ValueError(
+                "The input must contain the following keys: 'prompts' and 'images'."
+            )
+        return model_input["prompts"], model_input["images"]
+
+    def format_output_type(self, output_type):
+        """Generate the logits processor argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The logits processor provided.
+
+        """
+        from transformers import LogitsProcessorList
+
+        if output_type is not None:
+            return LogitsProcessorList([output_type])
+        return None
 
 
 class TransformersVision(Transformers):
-    def __init__(self, model, tokenizer, processor):
-        super().__init__(model, tokenizer)
-        self.processor = processor
+    """Represents a `transformers` model with a vision processor."""
 
-    def generate(  # type: ignore
+    def __init__(
         self,
-        prompts: Union[str, List[str]],
-        media: Union[List[Any], List[List[Any]]],
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Union[str, List[str], List[List[str]]]:
-        """Generate text using `transformers`.
+        model_name: str,
+        model_class,
+        model_kwargs: dict = {},
+        tokenizer_class=None,
+        tokenizer_kwargs: dict = {},
+        processor_class=None,
+        processor_kwargs: dict = {},
+    ):
+        """Create a TransformersVision model instance
+
+        We rely on the `__init__` method of the `Transformers` class to handle
+        most of the initialization and then add elements specific to vision
+        models.
 
         Parameters
         ----------
-        prompts
-            A prompt or list of prompts.
-        media
-            A List[PIL.Image] or List[List[PIL.Image]]
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-
-        Returns
-        -------
-        The generated text
+        model_name
+            The name of the transformers model to use;
+        model_class
+            The Transformers model class from which to create the model.
+        model_kwargs
+            A dictionary of keyword arguments to pass to the `from_pretrained`
+            method of the model class.
+        tokenizer_class
+            The Transformers tokenizer class from which to create the tokenizer.
+            If not provided,`AutoTokenizer` will be used.
+            If you gave the name of a model that is not compatible with `AutoTokenizer`,
+            you must provide a value for this parameter.
+        tokenizer_kwargs
+            A dictionary of keyword arguments to pass to the `from_pretrained`
+            method of the tokenizer class.
+        processor_class
+            The Transformers processor class from which to create the processor.
+            If not provided,`AutoProcessor` will be used.
+            If you gave the name of a model that is not compatible with `AutoProcessor`,
+            you must provide a value for this parameter.
+        processor_kwargs
+            A dictionary of keyword arguments to pass to the `from_pretrained`
+            method of the processor class.
+
         """
+        if processor_class is None:
+            try:
+                from transformers import AutoProcessor
+
+                processor_class = AutoProcessor
+            except ImportError:
+                raise ImportError(
+                    "The `transformers` library needs to be installed in order to use `transformers` models."
+                )
+
+        processor_kwargs.setdefault("padding_side", "left")
+        processor_kwargs.setdefault("pad_token", "[PAD]")
+        self.processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+        if tokenizer_class is None and getattr(self.processor, "tokenizer", None):
+            tokenizer_class = type(self.processor.tokenizer)
+
+        super().__init__(
+            model_name,
+            model_class,
+            model_kwargs,
+            tokenizer_class,
+            tokenizer_kwargs,
+        )
+
+        self.type_adapter = TransformersVisionTypeAdapter()
+
+    def generate(self, model_input, output_type, **inference_kwargs):
+        prompts, images = self.type_adapter.format_input(model_input)
+
         inputs = self.processor(
-            text=prompts, images=media, padding=True, return_tensors="pt"
+            text=prompts, images=images, padding=True, return_tensors="pt"
         ).to(self.model.device)
+        logits_processor = self.type_adapter.format_output_type(output_type)
 
-        generation_kwargs = self._get_generation_kwargs(
-            prompts,
-            generation_parameters,
-            logits_processor,
-            sampling_parameters,
+        generated_ids = self._generate_output_seq(
+            prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )
-        generated_ids = self._generate_output_seq(prompts, inputs, **generation_kwargs)
 
-        # if single str input and single sample per input, convert to a 1D output
+        # if single str input, convert to a 1D outputt
         if isinstance(prompts, str):
-            # Should always be true until NotImplementedError above is fixed
             generated_ids = generated_ids.squeeze(0)
 
         return self._decode_generation(generated_ids)
 
-    def stream(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[Any, List[Any]],  # TODO: docstring
-        generation_parameters: GenerationParameters,
-        logits_processor: Optional["OutlinesLogitsProcessor"],
-        sampling_parameters: SamplingParameters,
-    ) -> Iterator[Union[str, List[str]]]:
-        raise NotImplementedError
-
-
-def transformers_vision(
-    model_name: str,
-    model_class,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    processor_kwargs: dict = {},
-    tokenizer_class=None,
-    processor_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
-
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    model_class
-        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
-        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    processor_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the processor.
-
-    Returns
-    -------
-    A `TransformersModel` model instance.
-
-    """
-    if processor_class is None or tokenizer_class is None:
-        try:
-            from transformers import AutoProcessor, AutoTokenizer
-        except ImportError:
-            raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
-            )
-    if processor_class is None:
-        processor_class = AutoProcessor
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
-
-    if device is not None:
-        model_kwargs["device_map"] = device
-
-    model = model_class.from_pretrained(model_name, **model_kwargs)
-
-    processor_kwargs.setdefault("padding_side", "left")
-    processor_kwargs.setdefault("pad_token", "[PAD]")
-    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-    if tokenizer_class is None:
-        if getattr(processor, "tokenizer", None):
-            tokenizer = processor.tokenizer
-        else:
-            tokenizer = AutoTokenizer.from_pretrained(model_name, **processor_kwargs)
-    else:
-        tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
-
-    return TransformersVision(model, tokenizer, processor)
+    def stream(self, model_input, output_type, **inference_kwargs):
+        raise NotImplementedError(
+            "Streaming is not implemented for TransformersVision models."
+        )
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 01bc251da..ebc99a0f7 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -5,7 +5,7 @@
 
 from jsonschema import Draft202012Validator as Validator
 from jsonschema.exceptions import SchemaError
-from outlines_core.json_schema import build_regex_from_schema
+from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel, TypeAdapter
 from typing_extensions import _TypedDictMeta  # type: ignore
 
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 73de3876c..7f8ce0c16 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -146,6 +146,9 @@ def _display_node(self) -> str:
     def __repr__(self):
         return f"Regex(pattern='{self.pattern}')"
 
+    def to_regex(self) -> str:
+        return self.pattern
+
 
 class JsonSchema(Term):
     def __init__(self, schema: Union[dict, str, type[BaseModel]]):
diff --git a/pyproject.toml b/pyproject.toml
index 6ad04f601..02e76eb43 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -158,11 +158,8 @@ module = [
     "iso3166.*",
     "airportsdata.*",
     "outlines_core.*",
-<<<<<<< HEAD
     "genson",
-=======
     "ollama.*",
->>>>>>> 5af6b93 (Create the Ollama model)
 ]
 ignore_missing_imports = true
 
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index c7e9bdc32..aaa4509bf 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -54,25 +54,25 @@ def model_mlxlm_phi3(tmp_path_factory):
 
 @pytest.fixture(scope="session")
 def model_transformers_random(tmp_path_factory):
-    return models.transformers("hf-internal-testing/tiny-random-gpt2", device="cpu")
+    return models.Transformers("hf-internal-testing/tiny-random-gpt2")
 
 
 @pytest.fixture(scope="session")
 def model_transformers_opt125m(tmp_path_factory):
-    return models.transformers("facebook/opt-125m", device="cpu")
+    return models.Transformers("facebook/opt-125m")
 
 
 @pytest.fixture(scope="session")
 def model_mamba(tmp_path_factory):
-    return models.mamba(model_name="state-spaces/mamba-130m-hf", device="cpu")
+    return models.Mamba(model_name="state-spaces/mamba-130m-hf")
 
 
 @pytest.fixture(scope="session")
 def model_bart(tmp_path_factory):
     from transformers import AutoModelForSeq2SeqLM
 
-    return models.transformers(
-        "facebook/bart-base", device="cpu", model_class=AutoModelForSeq2SeqLM
+    return models.Transformers(
+        "facebook/bart-base", model_class=AutoModelForSeq2SeqLM
     )
 
 
@@ -123,11 +123,11 @@ def model_t5(tmp_path_factory):
     "model_exllamav2",
     "model_mlxlm",
     "model_mlxlm_phi3",
-    "model_transformers_random",
-    "model_transformers_opt125m",
-    "model_mamba",
-    "model_bart",
-    "model_transformers_vision",
+    # "model_transformers_random",
+    # "model_transformers_opt125m",
+    # "model_mamba",
+    # "model_bart",
+    # "model_transformers_vision",
     "model_vllm",
 )
 
@@ -280,20 +280,21 @@ def test_generate_json(request, model_fixture, sample_schema):
     generator(**get_inputs(model_fixture), max_tokens=100)
 
 
-def test_integrate_genson_generate_json(request):
-    from genson import SchemaBuilder
-
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"name": "Toto", "age": 5})
-
-    model = request.getfixturevalue("model_transformers_opt125m")
-
-    generator = generate.json(model, builder)
-    res = generator("Return a json of a young boy")
-
-    assert "name" in res
-    assert "age" in res
+# TODO: add support for genson in the Regex type of v1.0
+#def test_integrate_genson_generate_json(request):
+#    from genson import SchemaBuilder
+#
+#    builder = SchemaBuilder()
+#    builder.add_schema({"type": "object", "properties": {}})
+#    builder.add_object({"name": "Toto", "age": 5})
+#
+#    model = request.getfixturevalue("model_transformers_opt125m")
+#
+#    generator = generate.json(model, builder)
+#    res = generator("Return a json of a young boy")
+#
+#    assert "name" in res
+#    assert "age" in res
 
 
 @pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
diff --git a/tests/generate/test_integration_transformers.py b/tests/generate/test_integration_transformers.py
deleted file mode 100644
index 92c5d789c..000000000
--- a/tests/generate/test_integration_transformers.py
+++ /dev/null
@@ -1,569 +0,0 @@
-import datetime
-import re
-from enum import Enum
-from functools import partial
-from typing import List, Union
-
-import pytest
-import torch
-from outlines_core.fsm.regex import reduced_vocabulary
-from pydantic import BaseModel, constr
-
-import outlines.generate as generate
-import outlines.models as models
-from outlines.models.transformers import Transformers, TransformerTokenizer
-from outlines.samplers import beam_search, greedy, multinomial
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    return models.transformers(
-        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
-    )
-
-
-def test_transformers_integration_text(model):
-    sequence = generate.text(model)(
-        "Write a short sentence ", seed=10000, max_tokens=10
-    )
-    assert isinstance(sequence, str)
-    assert model.tokenizer.eos_token not in sequence
-
-    sequence = generate.text(model)(
-        "Write a short sentence ",
-        max_tokens=20,
-        stop_at="a",
-        seed=10000,
-    )
-    assert isinstance(sequence, str)
-
-    prompts = ["Write a short sentence ", "And another one "]
-    sequence = generate.text(model)(
-        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0], str)
-
-
-def test_transformers_integration_text_multiple_samples(model):
-    sampler = multinomial(2)
-
-    sequence = generate.text(model, sampler=sampler)(
-        "Write a short sentence ", seed=10000
-    )
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert model.tokenizer.eos_token not in sequence
-
-    prompts = ["Write a short sentence ", "And another one "]
-    sequence = generate.text(model, sampler=sampler)(
-        prompts, max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0], list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0][0], str)
-
-
-def test_transformers_integration_streaming(model):
-    sequence = generate.text(model).stream(
-        "Write a short sentence ", max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-
-    token = next(sequence)
-    assert isinstance(token, str)
-
-    remaining = "".join([token for token in sequence])
-    assert isinstance(remaining, str)
-
-    sequence = generate.text(model).stream(
-        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    tokens = next(sequence)
-    assert isinstance(tokens, list)
-    assert isinstance(tokens[0], str)
-    assert isinstance(tokens[1], str)
-
-
-def test_transformers_integration_streaming_batch_samples(model):
-    sampler = multinomial(samples=2)
-
-    sequence = generate.text(model, sampler=sampler).stream(
-        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=[".", ","], seed=10000
-    )
-    tokens = next(sequence)
-    assert isinstance(tokens, list)
-    assert len(tokens) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[0]) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[1]) == 2
-
-
-def test_transformers_integration_streaming_batch_beam_search(model):
-    sampler = beam_search(beams=2)
-
-    sequence = generate.regex(model, r"ab[cd]e", sampler=sampler).stream(
-        ["Prompt1", "Prompt2"], max_tokens=10, stop_at=["c", "d"], seed=10000
-    )
-    tokens = next(sequence)
-    assert isinstance(tokens, list)
-    assert len(tokens) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[0]) == 2
-    assert isinstance(tokens[0], list)
-    assert len(tokens[1]) == 2
-
-
-def test_transformers_integration_text_stop(model):
-    prompt = "Write a short sentence "
-    sequence = generate.text(model)(prompt, stop_at="a", seed=10000)
-    assert sequence[len(prompt) :].find("a") == -1
-
-
-def test_transformers_various_regexes(model):
-    prompt = "Write an email address"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # One prompt
-    sequence = generator(prompt, seed=0)
-    assert re.fullmatch(regex_str, sequence) is not None
-
-
-def test_transformers_various_regexes_prompt_list(model):
-    prompt = "Write an email address"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # Two prompts
-    sequence = generator([prompt, prompt], seed=0)
-    assert re.fullmatch(regex_str, sequence[0]) is not None
-    assert re.fullmatch(regex_str, sequence[1]) is not None
-
-
-def test_transformers_various_regexes_prompt_list_multiple_samples(model):
-    sampler = multinomial(samples=2)
-    prompt = "Write an email address"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str, sampler=sampler)
-
-    # Two prompts
-    sequence = generator([prompt, prompt], seed=0, max_tokens=500)
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert re.fullmatch(regex_str, sequence[0][0]) is not None
-    assert re.fullmatch(regex_str, sequence[0][1]) is not None
-    assert re.fullmatch(regex_str, sequence[1][0]) is not None
-    assert re.fullmatch(regex_str, sequence[1][1]) is not None
-
-
-def test_transformers_various_regexes_prompt_list_beam_search(model):
-    sampler = beam_search(5)
-    prompt_1 = "Write an email address"
-    prompt_2 = "Random"
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str, sampler=sampler)
-
-    # Two prompts
-    sequence = generator([prompt_1, prompt_2], seed=0)
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert len(sequence[0]) == 5
-    assert re.fullmatch(regex_str, sequence[0][0]) is not None
-    assert re.fullmatch(regex_str, sequence[0][1]) is not None
-    assert re.fullmatch(regex_str, sequence[1][0]) is not None
-    assert re.fullmatch(regex_str, sequence[1][1]) is not None
-
-
-def test_transformers_integration_integer(model):
-    prompt = "Write a short sentence"
-    sequence = generate.format(model, int)(prompt, max_tokens=10, seed=0)
-
-    assert isinstance(sequence, int)
-
-
-def test_transformers_integration_integer_array(model):
-    prompts = ["Give me a number", "And another one"]
-    sequence = generate.format(model, int)(prompts, max_tokens=10, seed=0)
-    assert isinstance(sequence, list)
-    assert len(sequence) == 2
-    assert isinstance(sequence[0], int)
-    assert isinstance(sequence[1], int)
-
-
-def test_transformers_integration_float(model):
-    prompt = "Write a short sentence"
-    sequence = generate.format(model, float)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, float)
-
-
-def test_transformers_integration_bool(model):
-    prompt = "Is this True or False?"
-    sequence = generate.format(model, bool)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, bool)
-
-
-def test_transformers_integration_date(model):
-    prompt = "What day is it today?"
-    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, datetime.date)
-
-
-def test_transformers_integration_time(model):
-    prompt = "What time is it?"
-    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10, seed=0)
-
-    assert sequence != ""
-    assert isinstance(sequence, datetime.time)
-
-
-def test_transformers_integration_datetime(model):
-    prompt = "What time is it?"
-    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20, seed=0)
-
-    assert sequence != 0
-    assert isinstance(sequence, datetime.datetime)
-
-
-def test_transformers_integration_choice(model):
-    prompt = "Write a short sentence "
-    sequence = generate.choice(model, ["test", "choice"])(prompt, seed=0)
-
-    assert sequence == "test" or sequence == "choice"
-
-
-def test_transformers_integration_with_pad_token():
-    model_name = "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM"
-    model = models.transformers(model_name, device="meta")
-    assert model.tokenizer.pad_token_id == 1
-    assert model.tokenizer.pad_token == "<pad>"
-
-
-def test_transformers_json_basic(model):
-    prompt = "Output some JSON "
-
-    class Spam(BaseModel):
-        foo: int
-        bar: float
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam)(prompt, max_tokens=500, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.foo, int)
-    assert isinstance(result.bar, float)
-    assert isinstance(result.spam, str)
-    assert isinstance(result.fuzz, bool)
-    assert len(result.spam) <= 10
-
-
-def test_transformers_json_schema(model):
-    prompt = "Output some JSON "
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "integer"},
-           "bar": {"type": "string", "maxLength": 4}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    result = generate.json(model, schema)(prompt, max_tokens=500, seed=0)
-    assert isinstance(result, dict)
-    assert isinstance(result["foo"], int)
-    assert isinstance(result["bar"], str)
-
-
-def test_transformers_json_batch(model):
-    prompts = ["Output some JSON ", "Output more JSON"]
-
-    class Spam(BaseModel):
-        foo: int
-        bar: float
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam)(prompts, max_tokens=500, seed=0)
-    assert isinstance(result[0], BaseModel)
-    assert isinstance(result[1], BaseModel)
-
-
-def test_transformers_json_batch_multiple_samples(model):
-    sampler = multinomial(samples=2)
-    prompts = ["Output some JSON ", "Output more JSON"]
-
-    class Spam(BaseModel):
-        foo: int
-        bar: float
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    result = generate.json(model, Spam, sampler=sampler)(
-        prompts, max_tokens=500, seed=0
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert isinstance(result[0][0], BaseModel)
-    assert isinstance(result[0][1], BaseModel)
-    assert isinstance(result[1][0], BaseModel)
-    assert isinstance(result[1][1], BaseModel)
-
-
-def test_transformers_json_str_enum(model):
-    prompt = "Output some JSON "
-
-    class Name(str, Enum):
-        john = "John"
-        marc = "Marc"
-        michel = "Michel"
-
-    class User(BaseModel):
-        user_id: int
-        name: Name
-
-    result = generate.json(model, User)(prompt, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.user_id, int)
-    assert result.name in ["John", "Marc", "Michel"]
-
-
-def test_transformers_json_int_enum(model):
-    prompt = "Output some JSON "
-
-    class Id(int, Enum):
-        one = 1
-        two = 2
-
-    class User(BaseModel):
-        user_id: Id
-
-    result = generate.json(model, User)(prompt, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.user_id, int)
-    assert result.user_id in [1, 2]
-
-
-def add(a: int, b: int) -> int:
-    return a + b
-
-
-def mul(c: float, d: float) -> float:
-    return c * d
-
-
-def test_transformers_json_function_enum(model):
-    prompt = "Output some JSON "
-
-    class Operation(Enum):
-        add = partial(add)
-        mul = partial(mul)
-
-    result = generate.json(model, Operation)(prompt, seed=0)
-    assert isinstance(result, dict)
-    assert len(result) == 2
-    for k, v in result.items():
-        assert k in ["a", "b", "c", "d"]
-        assert isinstance(v, (int, float))
-
-
-def test_transformers_json_array(model):
-    prompt = "Output some JSON "
-
-    class User(BaseModel):
-        user_id: int
-        value: List[float]
-
-    result = generate.json(model, User)(prompt, seed=0)
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.user_id, int)
-    assert isinstance(result.value, list)
-    for value in result.value:
-        assert isinstance(value, float) or isinstance(value, int)
-
-
-@pytest.mark.xfail(reason="The implementation of `anyOf` is incorrect")
-def test_transformers_json_union(model):
-    prompt = "Output some JSON "
-
-    class Spam(BaseModel):
-        foo: int
-        bar: Union[constr(max_length=10), float]
-
-    result = generate.json(model, Spam)(prompt, max_tokens=100, seed=4)
-    assert isinstance(result, BaseModel)
-    assert (
-        isinstance(result.bar, int)
-        or isinstance(result.bar, float)
-        or isinstance(result.bar, str)
-    )
-
-
-def test_transformers_json_function(model):
-    prompt = "Output arguments for the function"
-
-    def function(foo: int, bar: List[int]):
-        return foo + sum(bar)
-
-    sequence = generate.json(model, function)(prompt, max_tokens=100, seed=4)
-    assert isinstance(sequence, dict)
-    assert isinstance(function(**sequence), int)
-
-
-def test_transformers_logits_vocab_size(model):
-    # Artificially increase the weights/logits size relative
-    # to the vocabulary
-    model.model.resize_token_embeddings(pad_to_multiple_of=3)
-
-    assert len(model.tokenizer.vocabulary) == 1024
-    assert model.model.base_model.wte.weight.shape[0] == 1026
-
-    generator = generate.choice(model, ["True", "False"])
-
-    sequence = generator("blah", seed=101)
-    assert sequence == "False"
-
-
-def test_transformers_json_custom_ws(model):
-    prompt = "Output some JSON with newlines"  # try to force model to use newlines
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "integer"},
-           "bar": {"type": "integer"}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    generator = generate.json(model, schema, whitespace_pattern=r"[ ]?")
-    generator.format_sequence = lambda x: x  # patch to return raw text
-    assert "\n" not in generator(prompt, max_tokens=500, seed=0)
-
-
-def test_transformers_reduced_vocabulary_caching():
-    from transformers import AutoTokenizer
-
-    hf_tokenizer = AutoTokenizer.from_pretrained("gpt2")
-    tokenizer = TransformerTokenizer(hf_tokenizer)
-    tokenizer2 = TransformerTokenizer(hf_tokenizer)
-
-    # TODO: We might actually want only one copy of a given tokenizer.
-    assert tokenizer is not tokenizer2
-
-    vocab = reduced_vocabulary(tokenizer)
-    vocab2 = reduced_vocabulary(tokenizer2)
-
-    assert vocab2 is vocab
-
-
-@pytest.mark.skip(reason="Custom Sampler Disabled in Transformers Integration")
-def test_custom_sampler(model):
-    seen = False
-    target_token_ids = model.tokenizer.encode(["c"])[0]
-
-    class biased_sampler:
-        def __init__(self, samples: int = 1):
-            self.samples = samples
-
-        def __call__(
-            logits: torch.DoubleTensor, samples: int, *_
-        ) -> torch.DoubleTensor:
-            nonlocal seen
-
-            if not seen:
-                seen = True
-                return target_token_ids, torch.tensor([0]), None
-            else:
-                return (
-                    torch.tensor([[model.tokenizer.eos_token_id]]),
-                    torch.tensor([0]),
-                    None,
-                )
-
-    generator = generate.choice(model, ["a", "b", "c"], sampler=biased_sampler(1))
-    sequence = generator(
-        """What is 1+1?
-    a. 3
-    b. 4
-    c. 2"""
-    )
-
-    assert sequence == "c"
-
-
-def test_transformers_use_existing_model_and_tokenizer():
-    from transformers import AutoModelForCausalLM, AutoTokenizer
-
-    model_name = "hf-internal-testing/tiny-random-GPTJForCausalLM"
-    hf_tokenizer = AutoTokenizer.from_pretrained(model_name)
-    hf_model = AutoModelForCausalLM.from_pretrained(model_name)
-    model = Transformers(hf_model, hf_tokenizer)
-    sequence = generate.text(model)("Write a short sentence ", seed=10000)
-    assert isinstance(sequence, str)
-
-
-@pytest.mark.skip("Caching for guide was temporarily turned off")
-def test_RegexGuide_caching(temp_cache_dir):
-    import outlines.caching
-    from outlines.fsm.guide import cached_create_states_mapping
-
-    assert outlines.caching._caching_enabled
-
-    regex = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-    prompt = "What is the IP address of the Google DNS servers? "
-
-    cache = outlines.caching.get_cache()
-
-    # Returns (hits, misses)
-    _ = cache.stats(enable=True)
-    assert cache.statistics
-
-    assert cached_create_states_mapping.__memory__ is cache
-
-    model = models.transformers(
-        "hf-internal-testing/tiny-random-XLMRobertaXLForCausalLM", device="cpu"
-    )
-    generator = generate.regex(model, regex, sampler=greedy())
-    assert cache.stats() == (0, 1)
-
-    model_2 = models.transformers(
-        "hf-internal-testing/tiny-random-GPTJForCausalLM", device="cpu"
-    )
-    generator_2 = generate.regex(model_2, regex, sampler=greedy())
-    assert cache.stats() == (0, 2)
-
-    # These two different models and tokenizers should not have the same state
-    # mapping results
-    assert (
-        generator.logits_processor.guide.states_to_token_maps
-        != generator_2.logits_processor.guide.states_to_token_maps
-    )
-
-    generator_3 = generate.regex(model_2, regex, sampler=greedy())
-    assert cache.stats() == (1, 2)
-    assert (
-        generator_2.logits_processor.guide.states_to_token_maps
-        == generator_3.logits_processor.guide.states_to_token_maps
-    )
-
-    # Just for fun...
-    structured = generator(prompt, max_tokens=30)
-    structured_2 = generator_2(prompt, max_tokens=30)
-
-    assert re.fullmatch(regex, structured)
-    assert re.fullmatch(regex, structured_2)
-    assert structured != structured_2
diff --git a/tests/generate/test_integration_transformers_vision.py b/tests/generate/test_integration_transformers_vision.py
deleted file mode 100644
index ee4f84c06..000000000
--- a/tests/generate/test_integration_transformers_vision.py
+++ /dev/null
@@ -1,116 +0,0 @@
-from io import BytesIO
-from urllib.request import urlopen
-
-import pytest
-from PIL import Image
-from transformers import AutoProcessor, LlavaForConditionalGeneration
-
-import outlines
-from outlines.models.transformers_vision import transformers_vision
-
-IMAGE_URLS = [
-    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
-    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
-    "https://upload.wikimedia.org/wikipedia/commons/b/be/Tamias-rufus-001.jpg",
-]
-
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    return transformers_vision(
-        "trl-internal-testing/tiny-LlavaForConditionalGeneration",
-        model_class=LlavaForConditionalGeneration,
-        device="cpu",
-    )
-
-
-@pytest.fixture(scope="session")
-def processor(tmp_path_factory):
-    return AutoProcessor.from_pretrained("llava-hf/llava-interleave-qwen-0.5b-hf")
-
-
-def test_single_image_text_gen(model, processor):
-    conversation = [
-        {
-            "role": "user",
-            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
-        },
-    ]
-    generator = outlines.generate.text(model)
-    sequence = generator(
-        processor.apply_chat_template(conversation),
-        [img_from_url(IMAGE_URLS[0])],
-        seed=10000,
-        max_tokens=10,
-    )
-    assert isinstance(sequence, str)
-
-
-def test_multi_image_text_gen(model, processor):
-    """If the length of image tags and number of images we pass are > 1 and equal,
-    we should yield a successful generation.
-    """
-    conversation = [
-        {
-            "role": "user",
-            "content": [
-                {"type": "text", "text": "What do all these have in common?"},
-            ]
-            + [{"type": "image"} for _ in range(len(IMAGE_URLS))],
-        },
-    ]
-    generator = outlines.generate.text(model)
-    sequence = generator(
-        processor.apply_chat_template(conversation),
-        [img_from_url(i) for i in IMAGE_URLS],
-        seed=10000,
-        max_tokens=10,
-    )
-    assert isinstance(sequence, str)
-
-
-def test_mismatched_image_text_gen(model, processor):
-    """If the length of image tags and number of images we pass are unequal,
-    we should raise an error.
-    """
-    conversation = [
-        {
-            "role": "user",
-            "content": [
-                {"type": "text", "text": "I'm passing 3 images, but only 1 image tag"},
-                {"type": "image"},
-            ],
-        },
-    ]
-    generator = outlines.generate.text(model)
-    with pytest.raises(ValueError):
-        _ = generator(
-            processor.apply_chat_template(conversation),
-            [img_from_url(i) for i in IMAGE_URLS],
-            seed=10000,
-            max_tokens=10,
-        )
-
-
-def test_single_image_choice(model, processor):
-    conversation = [
-        {
-            "role": "user",
-            "content": [{"type": "text", "text": "What is this?"}, {"type": "image"}],
-        },
-    ]
-    choices = ["cat", "dog"]
-    generator = outlines.generate.choice(model, choices)
-    sequence = generator(
-        processor.apply_chat_template(conversation),
-        [img_from_url(IMAGE_URLS[0])],
-        seed=10000,
-        max_tokens=10,
-    )
-    assert isinstance(sequence, str)
-    assert sequence in choices
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 1404d287a..249893ff9 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -1,117 +1,153 @@
+import re
+from enum import Enum
+
 import pytest
-import torch
-from transformers import AutoTokenizer
-from transformers.models.gpt2 import GPT2TokenizerFast
+from pydantic import BaseModel
+from transformers import AutoModelForSeq2SeqLM
+
+from outlines.models.transformers import Mamba, Transformers
+from outlines.types import Choice, Json, Regex
+
+TEST_MODEL = "erwanf/gpt2-mini"
+TEST_MODEL_SEQ2SEQ = "hf-internal-testing/tiny-random-t5"
+TEST_MODEL_MAMBA = "hf-internal-testing/tiny-random-MambaForCausalLM"
+
+
+def test_transformers_instantiate_simple():
+    model = Transformers(TEST_MODEL)
+    assert isinstance(model, Transformers)
+
+
+def test_transformers_instantiate_wrong_kwargs():
+    with pytest.raises(TypeError):
+        Transformers(TEST_MODEL, model_kwargs={"foo": "bar"})
+
+
+def test_transformers_instantiate_other_model_class():
+    model = Transformers(
+        model_name=TEST_MODEL_SEQ2SEQ, model_class=AutoModelForSeq2SeqLM
+    )
+    assert isinstance(model, Transformers)
+
+
+def test_transformers_instantiate_mamba():
+    model = Mamba(
+        model_name=TEST_MODEL_MAMBA,
+    )
+    assert isinstance(model, Mamba)
+    assert isinstance(model, Transformers)
+
+
+def test_transformers_instantiate_tokenizer_kwargs():
+    model = Transformers(
+        TEST_MODEL,
+        tokenizer_kwargs={"additional_special_tokens": ["<t1>", "<t2>"]}
+    )
+    assert "<t1>" in model.tokenizer.special_tokens
+    assert "<t2>" in model.tokenizer.special_tokens
+
+
+@pytest.fixture
+def model():
+    return Transformers(TEST_MODEL)
+
+
+def test_transformers_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_transformers_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
 
-from outlines.models.transformers import TransformerTokenizer, transformers
+def test_transformers_inference_kwargs(model):
+    result = model("Respond with one word. Not more.", max_new_tokens=100)
+    assert isinstance(result, str)
 
-TEST_MODEL = "hf-internal-testing/tiny-random-GPTJForCausalLM"
 
+def test_transformers_invalid_inference_kwargs(model):
+    with pytest.raises(ValueError):
+        model("Respond with one word. Not more.", foo="bar")
 
-def test_tokenizer():
-    tokenizer = AutoTokenizer.from_pretrained(TEST_MODEL, padding_side="left")
-    tokenizer = TransformerTokenizer(tokenizer)
-    assert tokenizer.eos_token_id == 0
-    assert tokenizer.pad_token_id == 0
-    assert isinstance(tokenizer.tokenizer, GPT2TokenizerFast)
 
-    token_ids, attention_mask = tokenizer.encode("Test")
-    assert token_ids.ndim == 2
-    assert token_ids.shape[0] == 1
-    assert isinstance(token_ids, torch.LongTensor)
-    assert token_ids.shape == attention_mask.shape
+def test_transformers_regex(model):
+    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
 
-    token_ids, attention_mask = tokenizer.encode(["Test", "Test"])
-    assert token_ids.ndim == 2
-    assert token_ids.shape[0] == 2
-    assert isinstance(token_ids, torch.LongTensor)
-    assert token_ids.shape == attention_mask.shape
 
-    token_ids, attention_mask = tokenizer.encode(["Test", "A long sentence"])
-    assert token_ids.shape == attention_mask.shape
-    assert attention_mask[0][0] == tokenizer.pad_token_id
+def test_transformers_json(model):
+    class Character(BaseModel):
+        name: str
 
-    text = tokenizer.decode(torch.tensor([[0, 1, 2]]))
-    isinstance(text, str)
+    result = model("Create a character with a name.", Json(Character))
+    assert "name" in result
 
-    text = tokenizer.decode(torch.tensor([[0, 1, 2], [3, 4, 5]]))
-    isinstance(text, list)
-    isinstance(text[0], str)
-    isinstance(text[1], str)
 
-    tokenizer = AutoTokenizer.from_pretrained(
-        TEST_MODEL, additional_special_tokens=["<t1>", "<t2>"]
+def test_transformers_choice(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Choice(Foo))
+    assert result in ["cat", "dog"]
+
+
+def test_transformers_batch_samples(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+    result = model(
+        "Respond with one word. Not more.", num_return_sequences=2, num_beams=2
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."]
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."],
+        num_return_sequences=2,
+        num_beams=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+
+
+def test_transformers_batch_samples_constrained(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Choice(Foo), num_return_sequences=2, num_beams=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Choice(Foo),
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Choice(Foo),
+        num_return_sequences=2,
+        num_beams=2,
     )
-    tokenizer = TransformerTokenizer(tokenizer)
-    assert "<t1>" in tokenizer.special_tokens
-    assert "<t2>" in tokenizer.special_tokens
-
-
-def test_llama_tokenizer():
-    tokenizer = AutoTokenizer.from_pretrained("hf-internal-testing/llama-tokenizer")
-    tokenizer = TransformerTokenizer(tokenizer)
-
-    # Broken
-    assert tokenizer.tokenizer.convert_tokens_to_string(["▁baz"]) == "baz"
-    assert tokenizer.tokenizer.convert_tokens_to_string(["<0x20>"]) == ""
-    assert tokenizer.tokenizer.convert_tokens_to_string(["▁▁▁"]) == "  "
-
-    # Not broken
-    assert tokenizer.convert_token_to_string("▁baz") == " baz"
-    assert tokenizer.convert_token_to_string("<0x20>") == " "
-    assert tokenizer.convert_token_to_string("▁▁▁") == "   "
-
-
-def test_model():
-    model = transformers(TEST_MODEL, device="cpu")
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.model.device.type == "cpu"
-
-    model = transformers(TEST_MODEL, model_kwargs={"device_map": "cpu"})
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.model.device.type == "cpu"
-
-    model = transformers(TEST_MODEL, device="cpu", model_kwargs={"device_map": "cuda"})
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert model.model.device.type == "cpu"
-
-    input_ids = torch.tensor([[0, 1, 2]])
-    logits, kv_cache = model(input_ids, torch.ones_like(input_ids))
-    assert logits.type() == "torch.FloatTensor"
-    assert logits.ndim == 2
-    assert logits.shape[0] == 1
-    assert len(kv_cache) == model.model.config.n_layer
-    assert len(kv_cache[0]) == 2
-    assert kv_cache[0][0].shape[1] == model.model.config.n_head
-    assert kv_cache[0][0].shape[2] == 3  # number of tokens
-
-    input_ids = torch.tensor([[0, 1, 2], [3, 4, 5], [6, 7, 8]])
-    logits, kv_cache = model(input_ids, torch.ones_like(input_ids))
-    assert logits.type() == "torch.FloatTensor"
-    assert logits.ndim == 2
-    assert logits.shape[0] == 3
-    assert len(kv_cache) == model.model.config.n_layer
-    assert len(kv_cache[0]) == 2
-    assert kv_cache[0][0].shape[1] == model.model.config.n_head
-    assert kv_cache[0][0].shape[2] == 3  # number of tokens
-
-    with pytest.raises(AssertionError):
-        input_ids = torch.tensor([[[0, 1, 2], [3, 4, 5]], [[6, 7, 8], [0, 1, 2]]])
-        logits = model(input_ids, torch.ones_like(input_ids))
-
-
-def test_tokenizer_eq_hash():
-    tokenizer_hf = AutoTokenizer.from_pretrained("gpt2")
-
-    tokenizer = TransformerTokenizer(tokenizer_hf)
-    tokenizer_2 = TransformerTokenizer(tokenizer_hf)
-
-    assert tokenizer == tokenizer_2
-    assert hash(tokenizer) == hash(tokenizer_2)
-
-    tokenizer_hf_2 = AutoTokenizer.from_pretrained("gpt2")
-    tokenizer_hf_2.add_tokens(["test_token"])
-
-    tokenizer_3 = TransformerTokenizer(tokenizer_hf_2)
-    assert tokenizer != tokenizer_3
-    assert hash(tokenizer) != hash(tokenizer_3)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+        assert item[0] in ["cat", "dog"]
+        assert item[1] in ["cat", "dog"]
diff --git a/tests/models/test_transformers_vision.py b/tests/models/test_transformers_vision.py
new file mode 100644
index 000000000..44ca4e050
--- /dev/null
+++ b/tests/models/test_transformers_vision.py
@@ -0,0 +1,235 @@
+import re
+from enum import Enum
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image
+from pydantic import BaseModel
+from transformers import (
+    Blip2ForConditionalGeneration,
+    CLIPModel,
+    CLIPProcessor,
+    LlavaForConditionalGeneration,
+)
+
+from outlines.models.transformers_vision import TransformersVision
+from outlines.types import Choice, Json, Regex
+
+TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+TEST_CLIP_MODEL = "openai/clip-vit-base-patch32"
+IMAGE_URLS = [
+    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
+    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
+]
+
+
+@pytest.fixture
+def images():
+    def img_from_url(url):
+        img_byte_stream = BytesIO(urlopen(url).read())
+        return Image.open(img_byte_stream).convert("RGB")
+
+    return [img_from_url(url) for url in IMAGE_URLS]
+
+
+@pytest.fixture
+def model():
+    return TransformersVision(
+        model_name=TEST_MODEL, model_class=LlavaForConditionalGeneration
+    )
+
+
+def test_transformers_vision_instantiate_simple():
+    model = TransformersVision(
+        model_name=TEST_MODEL,
+        model_class=Blip2ForConditionalGeneration,
+    )
+    assert isinstance(model, TransformersVision)
+
+
+def test_transformers_vision_instantiate_other_processor_class():
+    model = TransformersVision(
+        model_name=TEST_CLIP_MODEL, model_class=CLIPModel, processor_class=CLIPProcessor
+    )
+    assert isinstance(model, TransformersVision)
+
+
+def test_transformers_vision_simple(model, images):
+    result = model.generate(
+        {"prompts": "<image>Describe this image in one sentence:", "images": images[0]},
+        None,
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_vision_call(model, images):
+    result = model(
+        {"prompts": "<image>Describe this image in one sentence:", "images": images[0]}
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_vision_wrong_number_images(model, images):
+    with pytest.raises(ValueError):
+        a = model(
+            {
+                "prompts": "<image>Describe this image in one sentence:",
+                "images": [images[0], images[1]],
+            }
+        )
+        print(a)
+
+
+def test_transformers_vision_wrong_input_type(model):
+    with pytest.raises(NotImplementedError):
+        model.generate("invalid input", None)
+
+
+def test_transformers_inference_kwargs(model, images):
+    result = model(
+        {"prompts": "<image>Describe this image in one sentence:", "images": images[0]},
+        max_new_tokens=100,
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_invalid_inference_kwargs(model, images):
+    with pytest.raises(ValueError):
+        model(
+            {
+                "prompts": "<image>Describe this image in one sentence:",
+                "images": images[0],
+            },
+            foo="bar",
+        )
+
+
+def test_transformers_several_images(model, images):
+    result = model(
+        {
+            "prompts": "<image><image>Describe this image in one sentence:",
+            "images": [images[0], images[1]],
+        }
+    )
+    assert isinstance(result, str)
+
+
+def test_transformers_vision_json(model, images):
+    class Foo(BaseModel):
+        name: str
+
+    result = model(
+        {"prompts": "<image>Give a name to this animal.", "images": images[0]},
+        Json(Foo),
+    )
+    assert "name" in result
+
+
+def test_transformers_vision_regex(model, images):
+    result = model(
+        {"prompts": "<image>How old is it?", "images": images[0]}, Regex(r"[0-9]")
+    )
+
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
+
+
+def test_transformers_vision_choice(model, images):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model(
+        {"prompts": "<image>Is it a cat or a dog?", "images": images[0]}, Choice(Foo)
+    )
+
+    assert isinstance(result, str)
+    assert result in ["cat", "dog"]
+
+
+def test_transformers_vision_batch_samples(model, images):
+    result = model(
+        {"prompts": "<image>Describe this image in one sentence.", "images": images[0]},
+        num_return_sequences=2,
+        num_beams=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        {
+            "prompts": [
+                "<image>Describe this image in one sentence.",
+                "<image>Describe this image in one sentence.",
+            ],
+            "images": [images[0], images[1]],
+        }
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    result = model(
+        {
+            "prompts": [
+                "<image>Describe this image in one sentence.",
+                "<image>Describe this image in one sentence.",
+            ],
+            "images": [images[0], images[1]],
+        },
+        num_return_sequences=2,
+        num_beams=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+
+
+def test_transformers_vision_batch_samples_constrained(model, images):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model(
+        {"prompts": "<image>Describe this image in one sentence.", "images": images[0]},
+        Choice(Foo),
+        num_return_sequences=2,
+        num_beams=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert item in ["cat", "dog"]
+    result = model(
+        {
+            "prompts": [
+                "<image>Describe this image in one sentence.",
+                "<image>Describe this image in one sentence.",
+            ],
+            "images": [images[0], images[1]],
+        },
+        Choice(Foo),
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert item in ["cat", "dog"]
+    result = model(
+        {
+            "prompts": [
+                "<image>Describe this image in one sentence.",
+                "<image>Describe this image in one sentence.",
+            ],
+            "images": [images[0], images[1]],
+        },
+        Choice(Foo),
+        num_return_sequences=2,
+        num_beams=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+        assert item[0] in ["cat", "dog"]
+        assert item[1] in ["cat", "dog"]

From 7df30c855589566d130a5607f3843b67cde17d5e Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Fri, 21 Feb 2025 23:05:46 +0100
Subject: [PATCH 218/505] Create the Dottxt model

---
 docs/reference/models/dottxt.md | 51 ++++++++++++++++++++
 outlines/models/__init__.py     |  3 +-
 outlines/models/dottxt.py       | 84 +++++++++++++++++++++++++++++++++
 pyproject.toml                  |  5 +-
 tests/models/test_dottxt.py     | 69 +++++++++++++++++++++++++++
 5 files changed, 210 insertions(+), 2 deletions(-)
 create mode 100644 docs/reference/models/dottxt.md
 create mode 100644 outlines/models/dottxt.py
 create mode 100644 tests/models/test_dottxt.py

diff --git a/docs/reference/models/dottxt.md b/docs/reference/models/dottxt.md
new file mode 100644
index 000000000..0437891f8
--- /dev/null
+++ b/docs/reference/models/dottxt.md
@@ -0,0 +1,51 @@
+# Dottxt
+
+!!! Installation
+
+    To be able to use Dottxt in Outlines, you must install the `dottxt` python sdk.
+
+    ```bash
+    pip install dottxt
+    ```
+
+    You also need to have a Dottxt API key. This API key must either be set as an environment variable called `DOTTXT_API_KEY` or be provided to the `outlines.models.Dottxt` class when instantiating it.
+
+## Generate text
+
+Dottxt only supports constrained generation with the `Json` output type. The input of the generation must be a string. Batch generation is not supported.
+Thus, you must always provide an output type.
+
+You can either create a `Generator` object and call it afterward:
+
+```python
+from outlines.models import Dottxt
+from outlines.generate import Generator
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+
+model = Dottxt()
+generator = Generator(model, Character)
+result = generator("Create a character")
+```
+
+or call the model directly with the output type:
+
+```python
+from outlines.models import Dottxt
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+
+model = Dottxt()
+result = model("Create a character", Character)
+```
+
+In any case, compilation for a given output type happens only once (the first time it is used to generate text).
+
+## Optional parameters
+
+You can provide the same optional parameters you would pass to the `dottxt` sdk's client both during the initialization of the `Dottxt` class and when generating text.
+Consult the [dottxt python sdk Github repository](https://github.com/dottxt-ai/dottxt-python) for the full list of parameters.
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 4254e7cf6..7adf0d0b5 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -10,6 +10,7 @@
 
 from .anthropic import Anthropic
 from .base import Model, ModelTypeAdapter
+from .dottxt import Dottxt
 from .exllamav2 import ExLlamaV2Model, exl2
 from .gemini import Gemini
 from .llamacpp import LlamaCpp
@@ -25,4 +26,4 @@
 ]
 
 LocalModel = Union[LlamaCpp, Transformers]
-APIModel = Union[AzureOpenAI, OpenAI, Anthropic, Gemini, Ollama]
+APIModel = Union[AzureOpenAI, OpenAI, Anthropic, Gemini, Ollama, Dottxt]
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
new file mode 100644
index 000000000..3f0ca2245
--- /dev/null
+++ b/outlines/models/dottxt.py
@@ -0,0 +1,84 @@
+"""Integration with Dottxt's API."""
+import json
+from functools import singledispatchmethod
+from types import NoneType
+from typing import Optional
+
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.types import Json
+
+__all__ = ["Dottxt"]
+
+
+class DottxtTypeAdapter(ModelTypeAdapter):
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the `messages` argument to pass to the client.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available with Dottxt. The only available type is `str`."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        """Generate the `messages` argument to pass to the client when the user
+        only passes a prompt.
+
+        """
+        return model_input
+
+    @singledispatchmethod
+    def format_output_type(self, output_type):
+        """Format the output type to pass to the client."""
+        raise NotImplementedError(
+            f"The input type {input} is not available with Dottxt."
+        )
+
+    @format_output_type.register(Json)
+    def format_json_output_type(self, output_type: Json):
+        """Format the output type to pass to the client."""
+        schema = output_type.to_json_schema()
+        return json.dumps(schema)
+
+    @format_output_type.register(NoneType)
+    def format_none_output_type(self, output_type: None):
+        """Format the output type to pass to the client."""
+        raise NotImplementedError(
+            "You must provide an output type. Dottxt only supports constrained generation."
+        )
+
+
+class Dottxt(Model):
+    """Thin wrapper around the `dottxt.client.Dottxt` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `dottxt.client.Dottxt` client.
+
+    """
+
+    def __init__(self, model_name: Optional[str] = None, *args, **kwargs):
+        from dottxt.client import Dottxt
+
+        self.client = Dottxt(*args, **kwargs)
+        self.model_name = model_name
+        self.type_adapter = DottxtTypeAdapter()
+
+    def generate(self, model_input, output_type=None, **inference_kwargs):
+        prompt = self.type_adapter.format_input(model_input)
+        json_schema = self.type_adapter.format_output_type(output_type)
+
+        if self.model_name:
+            inference_kwargs["model_name"] = self.model_name
+
+        completion = self.client.json(
+            prompt,
+            json_schema,
+            **inference_kwargs,
+        )
+        return completion.data
diff --git a/pyproject.toml b/pyproject.toml
index 02e76eb43..9b6d5174b 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -56,6 +56,7 @@ gemini = ["google-generativeai"]
 llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
 exllamav2 = ["exllamav2"]
 ollama = ["ollama"]
+dottxt = ["dottxt"]
 test = [
     "pre-commit",
     "pytest",
@@ -78,7 +79,8 @@ test = [
     "pillow",
     "exllamav2",
     "jax",
-    "ollama"
+    "ollama",
+    "dottxt",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
 serve = [
@@ -160,6 +162,7 @@ module = [
     "outlines_core.*",
     "genson",
     "ollama.*",
+    "dottxt.*",
 ]
 ignore_missing_imports = true
 
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
new file mode 100644
index 000000000..fe1aa9be5
--- /dev/null
+++ b/tests/models/test_dottxt.py
@@ -0,0 +1,69 @@
+import json
+import os
+
+import pytest
+from pydantic import BaseModel
+
+from outlines.generate import Generator
+from outlines.models.dottxt import Dottxt
+from outlines.types import Json
+
+
+class User(BaseModel):
+    first_name: str
+    last_name: str
+    user_id: int
+
+
+@pytest.fixture
+def api_key():
+    """Get the Dottxt API key from the environment, providing a default value if not found.
+
+    This fixture should be used for tests that do not make actual api calls,
+    but still require to initialize the Dottxt client.
+
+    """
+    api_key = os.getenv("DOTTXT_API_KEY")
+    if not api_key:
+        return "MOCK_API_KEY"
+    return api_key
+
+
+def test_dottxt_wrong_init_parameters(api_key):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        Dottxt(api_key=api_key, foo=10)
+
+
+def test_dottxt_wrong_output_type(api_key):
+    with pytest.raises(NotImplementedError, match="must provide an output type"):
+        model = Dottxt(api_key=api_key)
+        model("prompt")
+
+
+@pytest.mark.api_call
+def test_dottxt_wrong_input_type(api_key):
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model = Dottxt(api_key=api_key)
+        model(["prompt"], Json(User))
+
+
+@pytest.mark.api_call
+def test_dottxt_wrong_inference_parameters(api_key):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model = Dottxt(api_key=api_key)
+        model("prompt", Json(User), foo=10)
+
+
+@pytest.mark.api_call
+def test_dottxt_direct_call(api_key):
+    model = Dottxt(api_key=api_key, model_name="meta-llama/Llama-3.1-8B-Instruct")
+    result = model("Create a user", Json(User))
+    assert "first_name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_dottxt_generator_call(api_key):
+    model = Dottxt(api_key=api_key, model_name="meta-llama/Llama-3.1-8B-Instruct")
+    generator = Generator(model, Json(User))
+    result = generator("Create a user")
+    assert "first_name" in json.loads(result)

From 80fc21f48ab1de46b494c2216f4bc57e2bb74bfb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:24:46 +0100
Subject: [PATCH 219/505] Misc format fixes

---
 outlines/types/__init__.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index ebc99a0f7..a2ddccfbc 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -12,16 +12,17 @@
 from . import airports, countries, locale
 from outlines.types.dsl import (
     Regex,
-    json_schema,
+    JsonSchema,
     regex,
+    json_schema,
     either,
     optional,
     exactly,
     at_least,
     at_most,
     between,
-    one_or_more,
     zero_or_more,
+    one_or_more,
 )
 
 

From e5a7a91f026c6d9963cf8eb2bbd5c1f244e8bb4c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:50:04 +0100
Subject: [PATCH 220/505] Rename `Prompt` to `Template`

The `Prompt` class really represents a prompt *template* and not a
prompt. I thus renamed it to `Template` to make it less confusing to the
user.
---
 README.md | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/README.md b/README.md
index b20fc821c..d58d207f5 100644
--- a/README.md
+++ b/README.md
@@ -410,12 +410,35 @@ labelling = outlines.Template.from_file("prompt.txt")
 prompt = labelling("Just awesome", examples)
 ```
 
+<<<<<<< HEAD
 This helps:
 
 - Keep content separate from the code
 - Design "white space perfect" prompts
 
 It is more maintainable and means prompts can be versioned separately from the code.
+=======
+You can also load a template from a text file. Assume you have the following prompt template defined in `prompt.txt`:
+
+``` text
+You are a sentiment-labelling assistant.
+
+{% for example in examples %}
+{{ example[0] }} // {{ example[1] }}
+{% endfor %}
+{{ to_label }} //
+```
+
+You can load it with:
+
+``` python
+import outlines
+
+labelling = outlines.Template.from_file("prompt.txt")
+prompt = labelling("Just awesome", examples)
+
+```
+>>>>>>> 8b15cf3 (Rename `Prompt` to `Template`)
 
 ## Join us
 

From 403cfc3fd97ffee2c58cbea6df87933fbba4f082 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:58:57 +0100
Subject: [PATCH 221/505] Add deprecation warning for the `prompt` decorator

---
 README.md                      | 11 ++++++++++-
 outlines/models/anthropic.py   |  3 ++-
 outlines/models/dottxt.py      |  1 +
 outlines/models/gemini.py      |  3 ++-
 outlines/models/openai.py      |  2 +-
 tests/models/test_anthropic.py |  2 +-
 tests/models/test_gemini.py    |  2 +-
 tests/models/test_openai.py    |  2 +-
 tests/test_function.py         |  1 -
 tests/test_templates.py        | 13 +------------
 10 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/README.md b/README.md
index d58d207f5..3b726216f 100644
--- a/README.md
+++ b/README.md
@@ -406,6 +406,7 @@ examples = [
     ("The waiter was rude", "Negative")
 ]
 
+<<<<<<< HEAD
 labelling = outlines.Template.from_file("prompt.txt")
 prompt = labelling("Just awesome", examples)
 ```
@@ -434,12 +435,20 @@ You can load it with:
 ``` python
 import outlines
 
+=======
+>>>>>>> 732de37 (Add deprecation warning for the `prompt` decorator)
 labelling = outlines.Template.from_file("prompt.txt")
 prompt = labelling("Just awesome", examples)
-
 ```
 >>>>>>> 8b15cf3 (Rename `Prompt` to `Template`)
 
+This helps:
+
+- Keep content separate from the code
+- Design "white space perfect" prompts
+
+It is more maintainable and means prompts can be versioned separately from the code.
+
 ## Join us
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 830c0cf77..7084215d4 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -1,9 +1,10 @@
 """Integration with Anthropic's API."""
+
 from functools import singledispatchmethod
 from typing import Union
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.prompts import Vision
+from outlines.templates import Vision
 
 __all__ = ["Anthropic"]
 
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index 3f0ca2245..b7fb3722b 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -1,4 +1,5 @@
 """Integration with Dottxt's API."""
+
 import json
 from functools import singledispatchmethod
 from types import NoneType
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 5ab6a515c..401bd401a 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -1,4 +1,5 @@
 """Integration with Gemini's API."""
+
 from enum import EnumMeta
 from functools import singledispatchmethod
 from types import NoneType
@@ -8,7 +9,7 @@
 from typing_extensions import _TypedDictMeta  # type: ignore
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.prompts import Vision
+from outlines.templates import Vision
 from outlines.types import Choice, Json, List
 
 __all__ = ["Gemini"]
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 378bb830d..8efa980e6 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -7,7 +7,7 @@
 from pydantic import BaseModel
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.prompts import Vision
+from outlines.templates import Vision
 from outlines.types import Json
 
 __all__ = ["OpenAI"]
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 1d5b5c533..cd80ea704 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -5,7 +5,7 @@
 import requests
 
 from outlines.models.anthropic import Anthropic
-from outlines.prompts import Vision
+from outlines.templates import Vision
 
 MODEL_NAME = "claude-3-haiku-20240307"
 
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index fb7a79def..72e488084 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -9,7 +9,7 @@
 from typing_extensions import TypedDict
 
 from outlines.models.gemini import Gemini
-from outlines.prompts import Vision
+from outlines.templates import Vision
 from outlines.types import Choice, Json, List
 
 MODEL_NAME = "gemini-1.5-flash-latest"
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index d7d074573..25f399402 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -8,7 +8,7 @@
 from pydantic import BaseModel
 
 from outlines.models.openai import OpenAI
-from outlines.prompts import Vision
+from outlines.templates import Vision
 from outlines.types import Json
 
 MODEL_NAME = "gpt-4o-mini-2024-07-18"
diff --git a/tests/test_function.py b/tests/test_function.py
index 8bb12976e..d66f3911c 100644
--- a/tests/test_function.py
+++ b/tests/test_function.py
@@ -9,7 +9,6 @@
 
 def test_function_basic():
     with pytest.deprecated_call(match="The @prompt decorator"):
-
         @outlines.prompt
         def test_template(text: str):
             """{{ text }}"""
diff --git a/tests/test_templates.py b/tests/test_templates.py
index 2fda3caa3..aa9939278 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -114,7 +114,6 @@ def test_render_jinja():
 
 def test_prompt_basic():
     with pytest.deprecated_call():
-
         @outlines.prompt
         def test_tpl(variable):
             """{{variable}} test"""
@@ -142,7 +141,6 @@ def test_single_quote_tpl(variable):
 
 def test_prompt_kwargs():
     with pytest.deprecated_call():
-
         @outlines.prompt
         def test_kwarg_tpl(var, other_var="other"):
             """{{var}} and {{other_var}}"""
@@ -186,7 +184,6 @@ def with_description():
         pass
 
     with pytest.deprecated_call():
-
         @outlines.prompt
         def name_description_ppt(fn):
             """
@@ -203,7 +200,6 @@ def with_signature(one: int, two: List[str], three: float = 1.0):
         pass
 
     with pytest.deprecated_call():
-
         @outlines.prompt
         def name_signature_ppt(fn):
             """
@@ -211,15 +207,12 @@ def name_signature_ppt(fn):
             """
 
         rendered = name_signature_ppt(with_signature)
-        assert (
-            rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
-        )
+        assert rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
 
     def test_function_call(one, two=2):
         return one + two
 
     with pytest.deprecated_call():
-
         @outlines.prompt
         def source_ppt(fn):
             """
@@ -236,7 +229,6 @@ class SimpleResponse(BaseModel):
         two: str
 
     with pytest.deprecated_call():
-
         @outlines.prompt
         def source_ppt(model):
             "{{model | schema }}"
@@ -269,7 +261,6 @@ def test_prompt_dict_response():
     response = {"one": "a description", "two": ""}
 
     with pytest.deprecated_call():
-
         @outlines.prompt
         def source_ppt(model):
             "{{model | schema }}"
@@ -315,7 +306,6 @@ def with_all(
         pass
 
     with pytest.deprecated_call():
-
         @outlines.prompt
         def args_prompt(fn):
             """args: {{ fn | args }}"""
@@ -345,7 +335,6 @@ def reverse(s: str) -> str:
         return s[::-1]
 
     with pytest.deprecated_call():
-
         @outlines.prompt(filters=dict(reverse=reverse))
         def test_tpl(variable):
             """{{ variable | reverse }} test"""

From 1fc74efea09cc355e0aa8cd18a2f8943e82a803e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 09:46:06 +0100
Subject: [PATCH 222/505] Add `Vision` type for prompt + image input to vision
 models

---
 outlines/__init__.py    |  2 +-
 outlines/templates.py   |  4 ++++
 tests/test_function.py  |  1 +
 tests/test_templates.py | 13 ++++++++++++-
 4 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 628c88f0e..df20a5dd9 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -16,7 +16,7 @@
     "get_cache",
     "Function",
     "prompt",
-    "Prompt",
+    "Template",
     "vectorize",
     "grammars",
 ]
diff --git a/outlines/templates.py b/outlines/templates.py
index 325bcfe04..6e771651c 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -8,7 +8,11 @@
 from dataclasses import dataclass
 from io import BytesIO
 from pathlib import Path
+<<<<<<< HEAD
 from typing import Any, Callable, Dict, Optional, Type, cast
+=======
+from typing import Any, Callable, Dict, List, Optional, Type, cast
+>>>>>>> 8797d19 (Add `Vision` type for prompt + image input to vision models)
 import warnings
 
 import jinja2
diff --git a/tests/test_function.py b/tests/test_function.py
index d66f3911c..8bb12976e 100644
--- a/tests/test_function.py
+++ b/tests/test_function.py
@@ -9,6 +9,7 @@
 
 def test_function_basic():
     with pytest.deprecated_call(match="The @prompt decorator"):
+
         @outlines.prompt
         def test_template(text: str):
             """{{ text }}"""
diff --git a/tests/test_templates.py b/tests/test_templates.py
index aa9939278..2fda3caa3 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -114,6 +114,7 @@ def test_render_jinja():
 
 def test_prompt_basic():
     with pytest.deprecated_call():
+
         @outlines.prompt
         def test_tpl(variable):
             """{{variable}} test"""
@@ -141,6 +142,7 @@ def test_single_quote_tpl(variable):
 
 def test_prompt_kwargs():
     with pytest.deprecated_call():
+
         @outlines.prompt
         def test_kwarg_tpl(var, other_var="other"):
             """{{var}} and {{other_var}}"""
@@ -184,6 +186,7 @@ def with_description():
         pass
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def name_description_ppt(fn):
             """
@@ -200,6 +203,7 @@ def with_signature(one: int, two: List[str], three: float = 1.0):
         pass
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def name_signature_ppt(fn):
             """
@@ -207,12 +211,15 @@ def name_signature_ppt(fn):
             """
 
         rendered = name_signature_ppt(with_signature)
-        assert rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
+        assert (
+            rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
+        )
 
     def test_function_call(one, two=2):
         return one + two
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def source_ppt(fn):
             """
@@ -229,6 +236,7 @@ class SimpleResponse(BaseModel):
         two: str
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def source_ppt(model):
             "{{model | schema }}"
@@ -261,6 +269,7 @@ def test_prompt_dict_response():
     response = {"one": "a description", "two": ""}
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def source_ppt(model):
             "{{model | schema }}"
@@ -306,6 +315,7 @@ def with_all(
         pass
 
     with pytest.deprecated_call():
+
         @outlines.prompt
         def args_prompt(fn):
             """args: {{ fn | args }}"""
@@ -335,6 +345,7 @@ def reverse(s: str) -> str:
         return s[::-1]
 
     with pytest.deprecated_call():
+
         @outlines.prompt(filters=dict(reverse=reverse))
         def test_tpl(variable):
             """{{ variable | reverse }} test"""

From d1d45333932f81ba95a7a03350fb98d1c2cf8987 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 09:48:09 +0100
Subject: [PATCH 223/505] Refactor the `OpenAI` interface

---
 outlines/models/openai.py      | 2 +-
 tests/models/test_anthropic.py | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 8efa980e6..b809742e7 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,9 +1,9 @@
 """Integration with OpenAI's API."""
-
 from functools import singledispatchmethod
 from types import NoneType
 from typing import Optional, Union
 
+
 from pydantic import BaseModel
 
 from outlines.models.base import Model, ModelTypeAdapter
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index cd80ea704..3aabba53d 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -51,7 +51,7 @@ def test_anthropic_simple_call():
 @pytest.mark.api_call
 def test_anthropic_direct_call():
     model = Anthropic(MODEL_NAME)
-    result = model("Respond with one word. Not more.")
+    result = model("Respond with one word. Not more.", max_tokens=1024)
     assert isinstance(result, str)
 
 

From 1c644f9d951fe751f71d335cbb570294ffc2bd92 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 29 Nov 2024 11:47:27 +0100
Subject: [PATCH 224/505] Refactor the llama.cpp interface

---
 tests/models/test_llamacpp.py | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index e9af3cd2a..ffb0110cd 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -31,11 +31,6 @@ def test_llamacpp_simple(model):
     assert isinstance(result, str)
 
 
-def test_llamacpp_call(model):
-    result = model("Respond with one word. Not more.")
-    assert isinstance(result, str)
-
-
 def test_llamacpp_regex(model):
     regex_str = Regex(r"[0-9]").to_regex()
     logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)

From a0ab514c17157b2b55f49359ae2279ca00c1260d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 16:54:09 +0100
Subject: [PATCH 225/505] Fix formatting

---
 outlines/generate/__init__.py |  8 ++++----
 outlines/models/gemini.py     |  6 +++---
 outlines/models/ollama.py     |  6 +++---
 outlines/models/openai.py     |  8 ++++----
 outlines/types/__init__.py    |  4 ++--
 tests/models/test_gemini.py   | 16 ++++++++--------
 tests/models/test_llamacpp.py |  6 +++---
 tests/models/test_ollama.py   |  6 +++---
 tests/models/test_openai.py   |  8 ++++----
 9 files changed, 34 insertions(+), 34 deletions(-)

diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
index 2fb99a030..0d4f3c8e3 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -3,7 +3,7 @@
 
 from outlines.models import APIModel, LlamaCpp, LocalModel
 from outlines.processors import CFGLogitsProcessor, RegexLogitsProcessor
-from outlines.types import CFG, Choice, Json, List, Regex
+from outlines.types import CFG, Choice, JsonType, List, Regex
 
 from .api import SequenceGenerator
 from .cfg import cfg
@@ -29,7 +29,7 @@ class APIGenerator:
     """
 
     model: APIModel
-    output_type: Optional[Union[Json, List, Choice, Regex]] = None
+    output_type: Optional[Union[JsonType, List, Choice, Regex]] = None
 
     def __post_init__(self):
         if isinstance(self.output_type, CFG):
@@ -58,7 +58,7 @@ class LocalGenerator:
     """
 
     model: LocalModel
-    output_type: Optional[Union[Json, List, Choice, Regex]]
+    output_type: Optional[Union[JsonType, List, Choice, Regex]]
 
     def __post_init__(self):
         if self.output_type is None:
@@ -81,7 +81,7 @@ def __call__(self, prompt, **inference_kwargs):
 
 def Generator(
     model: Union[LocalModel, APIModel],
-    output_type: Optional[Union[Json, List, Choice, Regex, CFG]] = None,
+    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG]] = None,
 ):
     if isinstance(model, APIModel):  # type: ignore
         return APIGenerator(model, output_type)  # type: ignore
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 401bd401a..d1dd454b3 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -10,7 +10,7 @@
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
-from outlines.types import Choice, Json, List
+from outlines.types import JsonType, Choice, List
 
 __all__ = ["Gemini"]
 
@@ -69,7 +69,7 @@ def format_list_output_type(self, output_type):
     def format_none_output_type(self, output_type):
         return {}
 
-    @format_output_type.register(Json)
+    @format_output_type.register(JsonType)
     def format_json_output_type(self, output_type):
         """Gemini only accepts Pydantic models and TypeDicts to define the JSON structure."""
         if issubclass(output_type.definition, BaseModel):
@@ -104,7 +104,7 @@ def __init__(self, model_name: str, *args, **kwargs):
     def generate(
         self,
         model_input: Union[str, Vision],
-        output_type: Optional[Union[Json, EnumMeta]] = None,
+        output_type: Optional[Union[JsonType, EnumMeta]] = None,
         **inference_kwargs,
     ):
         import google.generativeai as genai
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index c97460200..2e5037c4f 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -3,7 +3,7 @@
 from typing import Iterator
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import Json
+from outlines.types import JsonType
 
 
 class OllamaTypeAdapter(ModelTypeAdapter):
@@ -46,8 +46,8 @@ def format_output_type(self, output_type):
     def format_none_output_type(self, output_type: None):
         return ""
 
-    @format_output_type.register(Json)
-    def format_json_output_type(self, output_type: Json):
+    @format_output_type.register(JsonType)
+    def format_json_output_type(self, output_type: JsonType):
         return output_type.to_json_schema()
 
 
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index b809742e7..bca72a6e7 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -7,8 +7,8 @@
 from pydantic import BaseModel
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.templates import Vision
-from outlines.types import Json
+from outlines.prompts import Vision
+from outlines.types import JsonType
 
 __all__ = ["OpenAI"]
 
@@ -92,8 +92,8 @@ def format_none_output_type(self, _: None):
         """
         return {}
 
-    @format_output_type.register(Json)
-    def format_json_output_type(self, output_type: Json):
+    @format_output_type.register(JsonType)
+    def format_json_output_type(self, output_type: JsonType):
         """Generate the `response_format` argument to the client when the user
         specified a `Json` output type.
 
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index a2ddccfbc..3b69673cc 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -63,7 +63,7 @@
 
 
 @dataclass
-class Json:
+class JsonType:
     """Represents a JSON object.
 
     The structure of JSON object can be defined using a JSON Schema
@@ -77,7 +77,7 @@ class Json:
 
     """
 
-    definition: Union[str, dict]
+    definition: Union[str, dict, type[BaseModel]]
     whitespace_pattern: str = " "
 
     def to_json_schema(self):
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 72e488084..9d2a1e9e3 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -9,8 +9,8 @@
 from typing_extensions import TypedDict
 
 from outlines.models.gemini import Gemini
-from outlines.templates import Vision
-from outlines.types import Choice, Json, List
+from outlines.prompts import Vision
+from outlines.types import Choice, JsonType, List
 
 MODEL_NAME = "gemini-1.5-flash-latest"
 
@@ -60,7 +60,7 @@ def test_gemini_simple_pydantic():
     class Foo(BaseModel):
         bar: int
 
-    result = model.generate("foo?", Json(Foo))
+    result = model.generate("foo?", JsonType(Foo))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -95,7 +95,7 @@ class Foo(BaseModel):
         sna: int
         bar: Bar
 
-    result = model.generate("foo?", Json(Foo))
+    result = model.generate("foo?", JsonType(Foo))
     assert isinstance(result, str)
     assert "sna" in json.loads(result)
     assert "bar" in json.loads(result)
@@ -115,7 +115,7 @@ def test_gemini_simple_json_schema_dict():
         "title": "Foo",
         "type": "object",
     }
-    result = model.generate("foo?", Json(schema))
+    result = model.generate("foo?", JsonType(schema))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -128,7 +128,7 @@ def test_gemini_simple_json_schema_string():
     model = Gemini(MODEL_NAME)
 
     schema = "{'properties': {'bar': {'title': 'Bar', 'type': 'integer'}}, 'required': ['bar'], 'title': 'Foo', 'type': 'object'}"
-    result = model.generate("foo?", Json(schema))
+    result = model.generate("foo?", JsonType(schema))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -140,7 +140,7 @@ def test_gemini_simple_typed_dict():
     class Foo(TypedDict):
         bar: int
 
-    result = model.generate("foo?", Json(Foo))
+    result = model.generate("foo?", JsonType(Foo))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -175,7 +175,7 @@ def test_gemini_simple_list_pydantic():
     class Foo(BaseModel):
         bar: int
 
-    result = model.generate("foo?", List(Json(Foo)))
+    result = model.generate("foo?", List(JsonType(Foo)))
     assert isinstance(json.loads(result), list)
     assert isinstance(json.loads(result)[0], dict)
     assert "bar" in json.loads(result)[0]
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index ffb0110cd..9e384e52d 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -6,7 +6,7 @@
 
 from outlines.models import LlamaCpp
 from outlines.processors import RegexLogitsProcessor
-from outlines.types import Choice, Json, Regex
+from outlines.types import Choice, JsonType, Regex
 
 
 def test_load_model():
@@ -42,7 +42,7 @@ def test_llamacpp_json(model):
     class Foo(BaseModel):
         bar: str
 
-    regex_str = Json(Foo).to_regex()
+    regex_str = JsonType(Foo).to_regex()
     logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
     result = model.generate(
         "foo? Respond with one word.", logits_processor, max_tokens=1000
@@ -97,7 +97,7 @@ def test_llamacpp_stream_json(model):
     class Foo(BaseModel):
         bar: int
 
-    regex_str = Json(Foo).to_regex()
+    regex_str = JsonType(Foo).to_regex()
     logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
     generator = model.stream("foo?", logits_processor)
 
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index ed05f3e21..795abd027 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -5,7 +5,7 @@
 from pydantic import BaseModel
 
 from outlines.models import Ollama
-from outlines.types import Choice, Json
+from outlines.types import Choice, JsonType
 
 MODEL_NAME = "tinyllama"
 
@@ -39,7 +39,7 @@ def test_ollama_json():
     class Foo(BaseModel):
         foo: str
 
-    result = Ollama(MODEL_NAME)("Respond with one word. Not more.", Json(Foo))
+    result = Ollama(MODEL_NAME)("Respond with one word. Not more.", JsonType(Foo))
     assert isinstance(result, str)
     assert "foo" in json.loads(result)
 
@@ -69,7 +69,7 @@ class Foo(BaseModel):
         foo: str
 
     model = Ollama(MODEL_NAME)
-    generator = model.stream("Create a character.", Json(Foo))
+    generator = model.stream("Create a character.", JsonType(Foo))
     generated_text = []
     for text in generator:
         generated_text.append(text)
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 25f399402..c03e99f26 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -9,7 +9,7 @@
 
 from outlines.models.openai import OpenAI
 from outlines.templates import Vision
-from outlines.types import Json
+from outlines.types import JsonType
 
 MODEL_NAME = "gpt-4o-mini-2024-07-18"
 
@@ -93,7 +93,7 @@ def test_openai_simple_pydantic():
     class Foo(BaseModel):
         bar: int
 
-    result = model.generate("foo?", Json(Foo))
+    result = model.generate("foo?", JsonType(Foo))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -110,7 +110,7 @@ def test_openai_simple_vision_pydantic():
     class Logo(BaseModel):
         name: int
 
-    result = model.generate(Vision("What does this logo represent?", image), Json(Logo))
+    result = model.generate(Vision("What does this logo represent?", image), JsonType(Logo))
     assert isinstance(result, str)
     assert "name" in json.loads(result)
 
@@ -124,6 +124,6 @@ class Foo(BaseModel):
 
     schema = json.dumps(Foo.model_json_schema())
 
-    result = model.generate("foo?", Json(schema))
+    result = model.generate("foo?", JsonType(schema))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)

From 92b5aaffb29f502487f079e5cd7962d4a372ae09 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 16:54:25 +0100
Subject: [PATCH 226/505] Initialize `OpenAI` from client instance

---
 outlines/__init__.py                     |  2 ++
 outlines/function.py                     |  8 +++---
 outlines/models/__init__.py              |  6 ++---
 outlines/models/dottxt.py                |  6 ++---
 outlines/models/openai.py                | 34 ++++++++++--------------
 pyproject.toml                           |  6 -----
 tests/models/test_dottxt.py              | 10 +++----
 tests/models/test_gemini.py              |  2 +-
 tests/models/test_openai.py              | 32 +++++++++++++---------
 tests/models/test_transformers.py        |  4 +--
 tests/models/test_transformers_vision.py |  4 +--
 tests/test_function.py                   |  7 ++---
 12 files changed, 60 insertions(+), 61 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index df20a5dd9..308548ff2 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -10,6 +10,8 @@
 from outlines.function import Function
 from outlines.templates import Template, prompt
 
+from outlines.models import from_openai
+
 __all__ = [
     "clear_cache",
     "disable_cache",
diff --git a/outlines/function.py b/outlines/function.py
index aff21d68f..1f5e80595 100644
--- a/outlines/function.py
+++ b/outlines/function.py
@@ -4,7 +4,9 @@
 
 import requests
 
-from outlines import generate, models
+import outlines
+from outlines import models
+from outlines.types import JsonType
 
 if TYPE_CHECKING:
     from outlines.generate.api import SequenceGenerator
@@ -37,8 +39,8 @@ def from_github(cls, program_path: str, function_name: str = "fn"):
 
     def init_generator(self):
         """Load the model and initialize the generator."""
-        model = models.transformers(self.model_name)
-        self.generator = generate.json(model, self.schema)
+        model = models.Transformers(self.model_name)
+        self.generator = outlines.generate.Generator(model, JsonType(self.schema))
 
     def __call__(self, *args, **kwargs):
         """Call the function.
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 7adf0d0b5..150129884 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -16,8 +16,8 @@
 from .llamacpp import LlamaCpp
 from .mlxlm import MLXLM, mlxlm
 from .ollama import Ollama
-from .openai import AzureOpenAI, OpenAI
-from .transformers import Mamba, Transformers, TransformerTokenizer
+from .openai import from_openai, OpenAI
+from .transformers import Transformers, TransformerTokenizer, Mamba
 from .transformers_vision import TransformersVision
 from .vllm import VLLM, vllm
 
@@ -26,4 +26,4 @@
 ]
 
 LocalModel = Union[LlamaCpp, Transformers]
-APIModel = Union[AzureOpenAI, OpenAI, Anthropic, Gemini, Ollama, Dottxt]
+APIModel = Union[OpenAI, Anthropic, Gemini, Ollama, Dottxt]
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index b7fb3722b..4ba2f84a2 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -6,7 +6,7 @@
 from typing import Optional
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import Json
+from outlines.types import JsonType
 
 __all__ = ["Dottxt"]
 
@@ -41,8 +41,8 @@ def format_output_type(self, output_type):
             f"The input type {input} is not available with Dottxt."
         )
 
-    @format_output_type.register(Json)
-    def format_json_output_type(self, output_type: Json):
+    @format_output_type.register(JsonType)
+    def format_json_output_type(self, output_type: JsonType):
         """Format the output type to pass to the client."""
         schema = output_type.to_json_schema()
         return json.dumps(schema)
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index bca72a6e7..5bf3e6a44 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,15 +1,19 @@
 """Integration with OpenAI's API."""
+
 from functools import singledispatchmethod
 from types import NoneType
-from typing import Optional, Union
+from typing import Optional, Union, TYPE_CHECKING
 
 
 from pydantic import BaseModel
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.prompts import Vision
+from outlines.templates import Vision
 from outlines.types import JsonType
 
+if TYPE_CHECKING:
+    from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
+
 __all__ = ["OpenAI"]
 
 
@@ -124,12 +128,13 @@ class OpenAI(Model):
 
     """
 
-    def __init__(self, model_name: str, *args, **kwargs):
+    def __init__(
+        self, client: Union["OpenAIClient", "AzureOpenAIClient"], model_name: str
+    ):
         from openai import OpenAI
 
-        self.client = OpenAI(*args, **kwargs)
+        self.client = client
         self.model_name = model_name
-        self.model_type = "api"
         self.type_adapter = OpenAITypeAdapter()
 
     def generate(
@@ -147,18 +152,7 @@ def generate(
         return result.choices[0].message.content
 
 
-class AzureOpenAI(OpenAI):
-    """Thin wrapper around the `openai.AzureOpenAI` client.
-
-    This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `openai.AzureOpenAI` client.
-
-    """
-
-    def __init__(self, deployment_name: str, *args, **kwargs):
-        from openai import AzureOpenAI
-
-        self.client = AzureOpenAI(*args, **kwargs)
-        self.model_name = deployment_name
-        self.model_type = "api"
-        self.type_adapter = OpenAITypeAdapter()
+def from_openai(
+    client: Union["OpenAIClient", "AzureOpenAIClient"], model_name: str
+) -> OpenAI:
+    return OpenAI(client, model_name)
diff --git a/pyproject.toml b/pyproject.toml
index 9b6d5174b..fa9f5e724 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -49,13 +49,7 @@ dynamic = ["version"]
 [project.optional-dependencies]
 vllm = ["vllm", "transformers", "numpy<2"]
 transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
-mlxlm = ["mlx-lm", "datasets"]
-openai = ["openai"]
-anthropic = ["anthropic"]
-gemini = ["google-generativeai"]
-llamacpp = ["llama-cpp-python", "transformers", "datasets", "numpy<2"]
 exllamav2 = ["exllamav2"]
-ollama = ["ollama"]
 dottxt = ["dottxt"]
 test = [
     "pre-commit",
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index fe1aa9be5..4c58392a5 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -6,7 +6,7 @@
 
 from outlines.generate import Generator
 from outlines.models.dottxt import Dottxt
-from outlines.types import Json
+from outlines.types import JsonType
 
 
 class User(BaseModel):
@@ -44,26 +44,26 @@ def test_dottxt_wrong_output_type(api_key):
 def test_dottxt_wrong_input_type(api_key):
     with pytest.raises(NotImplementedError, match="is not available"):
         model = Dottxt(api_key=api_key)
-        model(["prompt"], Json(User))
+        model(["prompt"], JsonType(User))
 
 
 @pytest.mark.api_call
 def test_dottxt_wrong_inference_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
         model = Dottxt(api_key=api_key)
-        model("prompt", Json(User), foo=10)
+        model("prompt", JsonType(User), foo=10)
 
 
 @pytest.mark.api_call
 def test_dottxt_direct_call(api_key):
     model = Dottxt(api_key=api_key, model_name="meta-llama/Llama-3.1-8B-Instruct")
-    result = model("Create a user", Json(User))
+    result = model("Create a user", JsonType(User))
     assert "first_name" in json.loads(result)
 
 
 @pytest.mark.api_call
 def test_dottxt_generator_call(api_key):
     model = Dottxt(api_key=api_key, model_name="meta-llama/Llama-3.1-8B-Instruct")
-    generator = Generator(model, Json(User))
+    generator = Generator(model, JsonType(User))
     result = generator("Create a user")
     assert "first_name" in json.loads(result)
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 9d2a1e9e3..432a5df56 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -9,7 +9,7 @@
 from typing_extensions import TypedDict
 
 from outlines.models.gemini import Gemini
-from outlines.prompts import Vision
+from outlines.templates import Vision
 from outlines.types import Choice, JsonType, List
 
 MODEL_NAME = "gemini-1.5-flash-latest"
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index c03e99f26..b8c857213 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -5,8 +5,10 @@
 import PIL
 import pytest
 import requests
+from openai import OpenAI as OpenAIClient
 from pydantic import BaseModel
 
+import outlines
 from outlines.models.openai import OpenAI
 from outlines.templates import Vision
 from outlines.types import JsonType
@@ -28,14 +30,16 @@ def api_key():
     return api_key
 
 
-def test_openai_wrong_init_parameters(api_key):
-    with pytest.raises(TypeError, match="got an unexpected"):
-        OpenAI(MODEL_NAME, api_key=api_key, foo=10)
+def test_init_from_client():
+    client = OpenAIClient()
+    model = outlines.from_openai(client, "gpt-4o")
+    assert isinstance(model, OpenAI)
+    assert model.client == client
 
 
 def test_openai_wrong_inference_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = OpenAI(MODEL_NAME, api_key=api_key)
+        model = OpenAI(OpenAIClient(), MODEL_NAME)
         model.generate("prompt", foo=10)
 
 
@@ -45,7 +49,7 @@ def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = OpenAI(MODEL_NAME, api_key=api_key)
+        model = OpenAI(OpenAIClient(), MODEL_NAME)
         model.generate(Foo("prompt"))
 
 
@@ -55,27 +59,27 @@ def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = OpenAI(MODEL_NAME, api_key=api_key)
+        model = OpenAI(OpenAIClient, MODEL_NAME)
         model.generate("prompt", Foo(1))
 
 
 @pytest.mark.api_call
 def test_openai_simple_call():
-    model = OpenAI(MODEL_NAME)
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
     result = model.generate("Respond with one word. Not more.")
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
 def test_openai_direct_call():
-    model = OpenAI(MODEL_NAME)
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
     result = model("Respond with one word. Not more.")
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
 def test_openai_simple_vision():
-    model = OpenAI(MODEL_NAME)
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
 
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)
@@ -88,7 +92,7 @@ def test_openai_simple_vision():
 
 @pytest.mark.api_call
 def test_openai_simple_pydantic():
-    model = OpenAI(MODEL_NAME)
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
 
     class Foo(BaseModel):
         bar: int
@@ -100,7 +104,7 @@ class Foo(BaseModel):
 
 @pytest.mark.api_call
 def test_openai_simple_vision_pydantic():
-    model = OpenAI(MODEL_NAME)
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
 
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)
@@ -110,14 +114,16 @@ def test_openai_simple_vision_pydantic():
     class Logo(BaseModel):
         name: int
 
-    result = model.generate(Vision("What does this logo represent?", image), JsonType(Logo))
+    result = model.generate(
+        Vision("What does this logo represent?", image), JsonType(Logo)
+    )
     assert isinstance(result, str)
     assert "name" in json.loads(result)
 
 
 @pytest.mark.api_call
 def test_openai_simple_json_schema():
-    model = OpenAI(MODEL_NAME)
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
 
     class Foo(BaseModel):
         bar: int
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 249893ff9..f52b89002 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -6,7 +6,7 @@
 from transformers import AutoModelForSeq2SeqLM
 
 from outlines.models.transformers import Mamba, Transformers
-from outlines.types import Choice, Json, Regex
+from outlines.types import Choice, JsonType, Regex
 
 TEST_MODEL = "erwanf/gpt2-mini"
 TEST_MODEL_SEQ2SEQ = "hf-internal-testing/tiny-random-t5"
@@ -82,7 +82,7 @@ def test_transformers_json(model):
     class Character(BaseModel):
         name: str
 
-    result = model("Create a character with a name.", Json(Character))
+    result = model("Create a character with a name.", JsonType(Character))
     assert "name" in result
 
 
diff --git a/tests/models/test_transformers_vision.py b/tests/models/test_transformers_vision.py
index 44ca4e050..477a2f456 100644
--- a/tests/models/test_transformers_vision.py
+++ b/tests/models/test_transformers_vision.py
@@ -14,7 +14,7 @@
 )
 
 from outlines.models.transformers_vision import TransformersVision
-from outlines.types import Choice, Json, Regex
+from outlines.types import Choice, JsonType, Regex
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
 TEST_CLIP_MODEL = "openai/clip-vit-base-patch32"
@@ -121,7 +121,7 @@ class Foo(BaseModel):
 
     result = model(
         {"prompts": "<image>Give a name to this animal.", "images": images[0]},
-        Json(Foo),
+        JsonType(Foo),
     )
     assert "name" in result
 
diff --git a/tests/test_function.py b/tests/test_function.py
index 8bb12976e..cc8d7493e 100644
--- a/tests/test_function.py
+++ b/tests/test_function.py
@@ -1,6 +1,7 @@
+from typing import Annotated
 import pytest
 import responses
-from pydantic import BaseModel
+from pydantic import BaseModel, Field
 from requests.exceptions import HTTPError
 
 import outlines
@@ -15,14 +16,14 @@ def test_template(text: str):
             """{{ text }}"""
 
     class Foo(BaseModel):
-        id: int
+        id: Annotated[str, Field(max_length=5)]
 
     fn = Function(test_template, Foo, "hf-internal-testing/tiny-random-GPTJForCausalLM")
 
     assert fn.generator is None
 
     result = fn("test")
-    assert isinstance(result, BaseModel)
+    Foo.parse_raw(result)
 
 
 def test_download_from_github_invalid():

From be5e5c2e87236c1e7d93b14e724143ea055b84d2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 17:04:31 +0100
Subject: [PATCH 227/505] Initialize `Gemini` from client instance

---
 docs/reference/models/gemini.md |  3 ---
 outlines/__init__.py            |  2 +-
 outlines/models/__init__.py     |  2 +-
 outlines/models/gemini.py       | 19 ++++++++++------
 outlines/models/openai.py       |  2 +-
 tests/models/test_gemini.py     | 40 ++++++++++++++++++---------------
 6 files changed, 37 insertions(+), 31 deletions(-)

diff --git a/docs/reference/models/gemini.md b/docs/reference/models/gemini.md
index 07ec9aa5a..472f1839a 100644
--- a/docs/reference/models/gemini.md
+++ b/docs/reference/models/gemini.md
@@ -4,9 +4,6 @@
 
     You need to install the `google-generativeai` library to be able to use the Gemini API in Outlines. Or alternatively you can run:
 
-    ```bash
-    pip install "outlines[gemini]"
-    ```
 
 ## Gemini models
 
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 308548ff2..28b7e6a69 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -10,7 +10,7 @@
 from outlines.function import Function
 from outlines.templates import Template, prompt
 
-from outlines.models import from_openai
+from outlines.models import from_openai, from_gemini
 
 __all__ = [
     "clear_cache",
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 150129884..d35481cdb 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -12,7 +12,7 @@
 from .base import Model, ModelTypeAdapter
 from .dottxt import Dottxt
 from .exllamav2 import ExLlamaV2Model, exl2
-from .gemini import Gemini
+from .gemini import from_gemini, Gemini
 from .llamacpp import LlamaCpp
 from .mlxlm import MLXLM, mlxlm
 from .ollama import Ollama
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index d1dd454b3..f0fd8d78a 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -3,7 +3,7 @@
 from enum import EnumMeta
 from functools import singledispatchmethod
 from types import NoneType
-from typing import Optional, Union
+from typing import Optional, Union, TYPE_CHECKING
 
 from pydantic import BaseModel
 from typing_extensions import _TypedDictMeta  # type: ignore
@@ -12,7 +12,11 @@
 from outlines.templates import Vision
 from outlines.types import JsonType, Choice, List
 
-__all__ = ["Gemini"]
+
+if TYPE_CHECKING:
+    from google.generativeai import GenerativeModel as GeminiClient
+
+__all__ = ["Gemini", "from_gemini"]
 
 
 class GeminiTypeAdapter(ModelTypeAdapter):
@@ -94,11 +98,8 @@ def format_enum_output_type(self, output_type):
 
 
 class Gemini(Model):
-    def __init__(self, model_name: str, *args, **kwargs):
-        import google.generativeai as genai
-
-        self.client = genai.GenerativeModel(model_name, *args, **kwargs)
-        self.model_type = "api"
+    def __init__(self, client: "GeminiClient"):
+        self.client = client
         self.type_adapter = GeminiTypeAdapter()
 
     def generate(
@@ -118,3 +119,7 @@ def generate(
         )
 
         return completion.text
+
+
+def from_gemini(client: "GeminiClient"):
+    return Gemini(client)
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 5bf3e6a44..6038ad1a1 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -14,7 +14,7 @@
 if TYPE_CHECKING:
     from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
 
-__all__ = ["OpenAI"]
+__all__ = ["OpenAI", "from_openai"]
 
 
 class OpenAITypeAdapter(ModelTypeAdapter):
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 432a5df56..cc54bd71a 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -7,7 +7,9 @@
 import requests
 from pydantic import BaseModel
 from typing_extensions import TypedDict
+import google.generativeai as genai
 
+import outlines
 from outlines.models.gemini import Gemini
 from outlines.templates import Vision
 from outlines.types import Choice, JsonType, List
@@ -15,34 +17,36 @@
 MODEL_NAME = "gemini-1.5-flash-latest"
 
 
-def test_gemini_wrong_init_parameters():
-    with pytest.raises(TypeError, match="got an unexpected"):
-        Gemini(MODEL_NAME, foo=10)
-
-
 def test_gemini_wrong_inference_parameters():
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = Gemini(MODEL_NAME)
+        model = Gemini(genai.GenerativeModel(MODEL_NAME))
         model.generate("prompt", foo=10)
 
 
+def test_gemini_init_from_client():
+    client = genai.GenerativeModel(MODEL_NAME)
+    model = outlines.from_gemini(client)
+    assert isinstance(model, Gemini)
+    assert model.client == client
+
+
 @pytest.mark.api_call
 def test_gemini_simple_call():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
     result = model.generate("Respond with one word. Not more.")
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
 def test_gemini_direct_call():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
     result = model("Respond with one word. Not more.")
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
 def test_gemini_simple_vision():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)
@@ -55,7 +59,7 @@ def test_gemini_simple_vision():
 
 @pytest.mark.api_call
 def test_gemini_simple_pydantic():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     class Foo(BaseModel):
         bar: int
@@ -68,7 +72,7 @@ class Foo(BaseModel):
 @pytest.mark.xfail(reason="Vision models do not work with structured outputs.")
 @pytest.mark.api_call
 def test_gemini_simple_vision_pydantic():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)
@@ -86,7 +90,7 @@ class Logo(BaseModel):
 @pytest.mark.xfail(reason="Gemini seems to be unable to follow nested schemas.")
 @pytest.mark.api_call
 def test_gemini_nested_pydantic():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     class Bar(BaseModel):
         fu: str
@@ -107,7 +111,7 @@ class Foo(BaseModel):
 )
 @pytest.mark.api_call
 def test_gemini_simple_json_schema_dict():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     schema = {
         "properties": {"bar": {"title": "Bar", "type": "integer"}},
@@ -125,7 +129,7 @@ def test_gemini_simple_json_schema_dict():
 )
 @pytest.mark.api_call
 def test_gemini_simple_json_schema_string():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     schema = "{'properties': {'bar': {'title': 'Bar', 'type': 'integer'}}, 'required': ['bar'], 'title': 'Foo', 'type': 'object'}"
     result = model.generate("foo?", JsonType(schema))
@@ -135,7 +139,7 @@ def test_gemini_simple_json_schema_string():
 
 @pytest.mark.api_call
 def test_gemini_simple_typed_dict():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     class Foo(TypedDict):
         bar: int
@@ -147,7 +151,7 @@ class Foo(TypedDict):
 
 @pytest.mark.api_call
 def test_gemini_simple_choice_enum():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     class Foo(Enum):
         bar = "Bar"
@@ -160,7 +164,7 @@ class Foo(Enum):
 
 @pytest.mark.api_call
 def test_gemini_simple_choice_list():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     choices = ["Foo", "Bar"]
     result = model.generate("foo?", Choice(choices))
@@ -170,7 +174,7 @@ def test_gemini_simple_choice_list():
 
 @pytest.mark.api_call
 def test_gemini_simple_list_pydantic():
-    model = Gemini(MODEL_NAME)
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
 
     class Foo(BaseModel):
         bar: int

From 6b1bf705406a2c8898b5b712847202838926ab97 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 17:12:55 +0100
Subject: [PATCH 228/505] Initialize `Anthropic` from client instance

---
 outlines/__init__.py           |  5 ++++-
 outlines/models/__init__.py    |  2 +-
 outlines/models/anthropic.py   | 18 +++++++++++-------
 tests/models/test_anthropic.py | 27 ++++++++++++++++-----------
 4 files changed, 32 insertions(+), 20 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 28b7e6a69..9169e6f2e 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -10,13 +10,16 @@
 from outlines.function import Function
 from outlines.templates import Template, prompt
 
-from outlines.models import from_openai, from_gemini
+from outlines.models import from_openai, from_gemini, from_anthropic
 
 __all__ = [
     "clear_cache",
     "disable_cache",
     "get_cache",
     "Function",
+    "from_anthropic",
+    "from_gemini",
+    "from_openai",
     "prompt",
     "Template",
     "vectorize",
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index d35481cdb..daf619516 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -8,7 +8,7 @@
 
 from typing import Union
 
-from .anthropic import Anthropic
+from .anthropic import from_anthropic, Anthropic
 from .base import Model, ModelTypeAdapter
 from .dottxt import Dottxt
 from .exllamav2 import ExLlamaV2Model, exl2
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 7084215d4..c7ed702dd 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -1,12 +1,15 @@
 """Integration with Anthropic's API."""
 
 from functools import singledispatchmethod
-from typing import Union
+from typing import Union, TYPE_CHECKING
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
 
-__all__ = ["Anthropic"]
+if TYPE_CHECKING:
+    from anthropic import Anthropic as AnthropicClient
+
+__all__ = ["Anthropic", "from_anthropic"]
 
 
 class AnthropicTypeAdapter(ModelTypeAdapter):
@@ -82,12 +85,9 @@ def format_output_type(self, output_type):
 
 
 class Anthropic(Model):
-    def __init__(self, model_name: str, *args, **kwargs):
-        from anthropic import Anthropic
-
-        self.client = Anthropic(*args, **kwargs)
+    def __init__(self, client: "AnthropicClient", model_name: str):
+        self.client = client
         self.model_name = model_name
-        self.model_type = "api"
         self.type_adapter = AnthropicTypeAdapter()
 
     def generate(
@@ -106,3 +106,7 @@ def generate(
             **inference_kwargs,
         )
         return completion.content[0].text
+
+
+def from_anthropic(client: "AnthropicClient", model_name: str) -> Anthropic:
+    return Anthropic(client, model_name)
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 3aabba53d..8aef257cc 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -1,23 +1,20 @@
 import io
 
+from anthropic import Anthropic as AnthropicClient
 import PIL
 import pytest
 import requests
 
+import outlines
 from outlines.models.anthropic import Anthropic
 from outlines.templates import Vision
 
 MODEL_NAME = "claude-3-haiku-20240307"
 
 
-def test_anthropic_wrong_init_parameters():
-    with pytest.raises(TypeError, match="got an unexpected"):
-        Anthropic(MODEL_NAME, foo=10)
-
-
 def test_anthropic_wrong_inference_parameters():
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = Anthropic(MODEL_NAME)
+        model = Anthropic(AnthropicClient(), MODEL_NAME)
         model.generate("prompt", foo=10, max_tokens=1024)
 
 
@@ -27,37 +24,45 @@ def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = Anthropic(MODEL_NAME)
+        model = Anthropic(AnthropicClient(), MODEL_NAME)
         model.generate(Foo("prompt"))
 
 
+def test_init_from_client():
+    client = AnthropicClient()
+    model = outlines.from_anthropic(client, MODEL_NAME)
+    assert isinstance(model, Anthropic)
+    assert model.client == client
+
+
 def test_anthropic_wrong_output_type():
     class Foo:
         def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = Anthropic(MODEL_NAME)
+        model = Anthropic(AnthropicClient(), MODEL_NAME)
         model.generate("prompt", Foo(1))
 
 
 @pytest.mark.api_call
 def test_anthropic_simple_call():
-    model = Anthropic(MODEL_NAME)
+    model = Anthropic(AnthropicClient(), MODEL_NAME)
     result = model.generate("Respond with one word. Not more.", max_tokens=1024)
     assert isinstance(result, str)
 
 
+@pytest.mark.xfail(reason="Anthropic requires the `max_tokens` parameter to be set")
 @pytest.mark.api_call
 def test_anthropic_direct_call():
-    model = Anthropic(MODEL_NAME)
+    model = Anthropic(AnthropicClient(), MODEL_NAME)
     result = model("Respond with one word. Not more.", max_tokens=1024)
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
 def test_anthropic_simple_vision():
-    model = Anthropic(MODEL_NAME)
+    model = Anthropic(AnthropicClient(), MODEL_NAME)
 
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)

From 900b9ff0dba6a49d26538fbc4d03a13cdb9311d0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 17:25:19 +0100
Subject: [PATCH 229/505] Initialize `Ollama` from client instance

---
 outlines/__init__.py        |  3 ++-
 outlines/models/__init__.py |  2 +-
 outlines/models/ollama.py   | 26 ++++++++++++-------------
 tests/models/test_ollama.py | 39 +++++++++++++++++++++----------------
 4 files changed, 38 insertions(+), 32 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 9169e6f2e..893794235 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -10,7 +10,7 @@
 from outlines.function import Function
 from outlines.templates import Template, prompt
 
-from outlines.models import from_openai, from_gemini, from_anthropic
+from outlines.models import from_openai, from_gemini, from_anthropic, from_ollama
 
 __all__ = [
     "clear_cache",
@@ -19,6 +19,7 @@
     "Function",
     "from_anthropic",
     "from_gemini",
+    "from_ollama",
     "from_openai",
     "prompt",
     "Template",
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index daf619516..64d992e6b 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -15,7 +15,7 @@
 from .gemini import from_gemini, Gemini
 from .llamacpp import LlamaCpp
 from .mlxlm import MLXLM, mlxlm
-from .ollama import Ollama
+from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
 from .transformers import Transformers, TransformerTokenizer, Mamba
 from .transformers_vision import TransformersVision
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 2e5037c4f..2eec91879 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,10 +1,16 @@
 from functools import singledispatchmethod
 from types import NoneType
-from typing import Iterator
+from typing import Iterator, TYPE_CHECKING
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.types import JsonType
 
+if TYPE_CHECKING:
+    from ollama import Client as OllamaClient
+
+
+__all__ = ["Ollama", "from_ollama"]
+
 
 class OllamaTypeAdapter(ModelTypeAdapter):
     """Type adapter for the Ollama model."""
@@ -59,21 +65,11 @@ class Ollama(Model):
 
     """
 
-    def __init__(self, model_name: str, *args, **kwargs):
-        from ollama import Client
-
-        self.client = Client(*args, **kwargs)
+    def __init__(self, client: "OllamaClient", model_name: str):
+        self.client = client
         self.model_name = model_name
         self.type_adapter = OllamaTypeAdapter()
 
-    @classmethod
-    def from_pretrained(cls, model_name: str, *args, **kwargs):
-        """Download the model weights from Ollama and create a `Ollama` instance."""
-        from ollama import pull
-
-        pull(model_name)
-        return cls(model_name, *args, **kwargs)
-
     def generate(self, model_input, output_type=None, **kwargs) -> str:
         response = self.client.generate(
             model=self.model_name,
@@ -93,3 +89,7 @@ def stream(self, model_input, output_type=None, **kwargs) -> Iterator[str]:
         )
         for chunk in response:
             yield chunk.response
+
+
+def from_ollama(client: "OllamaClient", model_name: str) -> Ollama:
+    return Ollama(client, model_name)
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index 795abd027..68afa5640 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -2,36 +2,39 @@
 from enum import Enum
 
 import pytest
+from ollama import Client as OllamaClient
 from pydantic import BaseModel
 
+import outlines
 from outlines.models import Ollama
 from outlines.types import Choice, JsonType
 
 MODEL_NAME = "tinyllama"
+CLIENT = OllamaClient()
 
 
-def test_pull_model():
-    model = Ollama.from_pretrained(MODEL_NAME)
-    assert isinstance(model, Ollama)
-
-
-def test_ollama_wrong_init_parameters():
+def test_wrong_inference_parameters():
     with pytest.raises(TypeError, match="got an unexpected"):
-        Ollama(MODEL_NAME, foo=10)
+        Ollama(CLIENT, MODEL_NAME).generate(
+            "Respond with one word. Not more.", None, foo=10
+        )
 
 
-def test_wrong_inference_parameters():
-    with pytest.raises(TypeError, match="got an unexpected"):
-        Ollama(MODEL_NAME).generate("Respond with one word. Not more.", None, foo=10)
+def test_init_from_client():
+    model = outlines.from_ollama(CLIENT, MODEL_NAME)
+    assert isinstance(model, Ollama)
+    assert model.client == CLIENT
 
 
 def test_ollama_simple():
-    result = Ollama(MODEL_NAME).generate("Respond with one word. Not more.", None)
+    result = Ollama(CLIENT, MODEL_NAME).generate(
+        "Respond with one word. Not more.", None
+    )
     assert isinstance(result, str)
 
 
 def test_ollama_direct():
-    result = Ollama(MODEL_NAME)("Respond with one word. Not more.", None)
+    result = Ollama(CLIENT, MODEL_NAME)("Respond with one word. Not more.", None)
     assert isinstance(result, str)
 
 
@@ -39,7 +42,9 @@ def test_ollama_json():
     class Foo(BaseModel):
         foo: str
 
-    result = Ollama(MODEL_NAME)("Respond with one word. Not more.", JsonType(Foo))
+    result = Ollama(CLIENT, MODEL_NAME)(
+        "Respond with one word. Not more.", JsonType(Foo)
+    )
     assert isinstance(result, str)
     assert "foo" in json.loads(result)
 
@@ -50,16 +55,16 @@ class Foo(Enum):
         foor = "Foo"
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        Ollama(MODEL_NAME).generate("foo?", Choice(Foo))
+        Ollama(CLIENT, MODEL_NAME).generate("foo?", Choice(Foo))
 
 
 def test_ollama_wrong_input_type():
     with pytest.raises(NotImplementedError, match="is not available"):
-        Ollama(MODEL_NAME).generate(["foo?", "bar?"], None)
+        Ollama(CLIENT, MODEL_NAME).generate(["foo?", "bar?"], None)
 
 
 def test_ollama_stream():
-    model = Ollama(MODEL_NAME)
+    model = Ollama(CLIENT, MODEL_NAME)
     generator = model.stream("Write a sentence about a cat.")
     assert isinstance(next(generator), str)
 
@@ -68,7 +73,7 @@ def test_ollama_stream_json():
     class Foo(BaseModel):
         foo: str
 
-    model = Ollama(MODEL_NAME)
+    model = Ollama(CLIENT, MODEL_NAME)
     generator = model.stream("Create a character.", JsonType(Foo))
     generated_text = []
     for text in generator:

From fc084e4714630a39b950db910a28c055bb94ab1e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 18:58:29 +0100
Subject: [PATCH 230/505] Initialize `LlamaCpp` from model instance

---
 docs/reference/models/llamacpp.md |  2 +-
 outlines/__init__.py              |  3 ++-
 outlines/models/__init__.py       |  2 +-
 outlines/models/llamacpp.py       | 40 +++++--------------------------
 tests/models/test_llamacpp.py     | 22 ++++++++++-------
 5 files changed, 24 insertions(+), 45 deletions(-)

diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
index 52e892b76..9ea0c5c09 100644
--- a/docs/reference/models/llamacpp.md
+++ b/docs/reference/models/llamacpp.md
@@ -12,7 +12,7 @@ Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/l
 
 ## Load the model
 
-To load a model you can use the same interface as you would using `llamap-cpp-python` directly. The default method is to initialize the model by passing the path to the weights on your machine. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
+You can use `outlines.from_llamacpp` to load a `llama-cpp-python` model. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
 
 ```python
 from outlines import models
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 893794235..2809f4637 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -10,7 +10,7 @@
 from outlines.function import Function
 from outlines.templates import Template, prompt
 
-from outlines.models import from_openai, from_gemini, from_anthropic, from_ollama
+from outlines.models import from_openai, from_gemini, from_anthropic, from_ollama, from_llamacpp
 
 __all__ = [
     "clear_cache",
@@ -19,6 +19,7 @@
     "Function",
     "from_anthropic",
     "from_gemini",
+    "from_llamacpp",
     "from_ollama",
     "from_openai",
     "prompt",
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 64d992e6b..c39b05a3e 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -13,7 +13,7 @@
 from .dottxt import Dottxt
 from .exllamav2 import ExLlamaV2Model, exl2
 from .gemini import from_gemini, Gemini
-from .llamacpp import LlamaCpp
+from .llamacpp import LlamaCpp, from_llamacpp
 from .mlxlm import MLXLM, mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 8823f3358..3657b669c 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -141,46 +141,14 @@ def format_output_type(self, output_type):
 class LlamaCpp(Model):
     """Wraps a model provided by the `llama-cpp-python` library."""
 
-    def __init__(self, model_path: Union[str, "Llama"], **kwargs):
+    def __init__(self, model: "Llama"):
         from llama_cpp import Llama
 
-        if isinstance(model_path, Llama):
-            self.model = model_path
-        else:
-            # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
-            if "tokenizer" not in kwargs:
-                warnings.warn(
-                    "The pre-tokenizer in `llama.cpp` handles unicode improperly "
-                    + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
-                    + "Outlines may raise a `RuntimeError` when building the regex index.\n"
-                    + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
-                    + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
-                )
-
-            self.model = Llama(model_path, **kwargs)
-
+        self.model = model
         self.tokenizer = LlamaCppTokenizer(self.model)
         self.model_type = "local"
         self.type_adapter = LlamaCppTypeAdapter()
 
-    @classmethod
-    def from_pretrained(cls, repo_id, filename, **kwargs):
-        """Download the model weights from Hugging Face and create a `Llama` instance"""
-        from llama_cpp import Llama
-
-        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
-        if "tokenizer" not in kwargs:
-            warnings.warn(
-                "The pre-tokenizer in `llama.cpp` handles unicode improperly "
-                + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
-                + "Outlines may raise a `RuntimeError` when building the regex index.\n"
-                + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
-                + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
-            )
-
-            model = Llama.from_pretrained(repo_id, filename, **kwargs)
-            return cls(model)
-
     def generate(self, model_input, logits_processor, **inference_kwargs):
         """Generate text using `llama-cpp-python`.
 
@@ -257,3 +225,7 @@ def token_generator() -> Iterator[str]:
                     return
 
         return token_generator()
+
+
+def from_llamacpp(model: "Llama"):
+    return LlamaCpp(model)
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 9e384e52d..225ec558d 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -2,6 +2,7 @@
 from enum import Enum
 
 import pytest
+from llama_cpp import Llama
 from pydantic import BaseModel
 
 from outlines.models import LlamaCpp
@@ -10,19 +11,24 @@
 
 
 def test_load_model():
-    model = LlamaCpp.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    model = LlamaCpp(
+        Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
     )
 
     assert isinstance(model, LlamaCpp)
+    assert isinstance(model.model, Llama)
 
 
 @pytest.fixture(scope="session")
 def model(tmp_path_factory):
-    return LlamaCpp.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    return LlamaCpp(
+        Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
     )
 
 
@@ -32,7 +38,7 @@ def test_llamacpp_simple(model):
 
 
 def test_llamacpp_regex(model):
-    regex_str = Regex(r"[0-9]").to_regex()
+    regex_str = Regex(r"[0-9]").pattern
     logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
     result = model.generate("Respond with one word. Not more.", logits_processor)
     assert isinstance(result, str)
@@ -85,7 +91,7 @@ def test_llamacpp_stream_simple(model):
 
 
 def test_llamacpp_stream_regex(model):
-    regex_str = Regex(r"[0-9]").to_regex()
+    regex_str = Regex(r"[0-9]").pattern
     logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
     generator = model.stream("Respond with one word. Not more.", logits_processor)
 

From 51fe015249d41cc849436d07336d24adc6c7c184 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 19:07:47 +0100
Subject: [PATCH 231/505] Initialize `VLLM` from model instance

---
 outlines/__init__.py                    | 26 ++++++++++++++++++-------
 outlines/models/__init__.py             |  2 +-
 outlines/models/vllm.py                 | 20 ++++---------------
 pyproject.toml                          |  1 -
 tests/generate/test_integration_vllm.py |  4 +++-
 5 files changed, 27 insertions(+), 26 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 2809f4637..7f6360f0e 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -10,20 +10,32 @@
 from outlines.function import Function
 from outlines.templates import Template, prompt
 
-from outlines.models import from_openai, from_gemini, from_anthropic, from_ollama, from_llamacpp
+from outlines.models import (
+    from_openai,
+    from_gemini,
+    from_anthropic,
+    from_ollama,
+    from_llamacpp,
+    from_vllm,
+)
 
-__all__ = [
-    "clear_cache",
-    "disable_cache",
-    "get_cache",
-    "Function",
+
+models = [
     "from_anthropic",
     "from_gemini",
     "from_llamacpp",
     "from_ollama",
     "from_openai",
+    "from_vllm",
+]
+
+__all__ = [
+    "clear_cache",
+    "disable_cache",
+    "get_cache",
+    "Function",
     "prompt",
     "Template",
     "vectorize",
     "grammars",
-]
+] + models
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index c39b05a3e..77b0c60dc 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -19,7 +19,7 @@
 from .openai import from_openai, OpenAI
 from .transformers import Transformers, TransformerTokenizer, Mamba
 from .transformers_vision import TransformersVision
-from .vllm import VLLM, vllm
+from .vllm import VLLM, from_vllm
 
 LogitsGenerator = Union[
     Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM, Ollama
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index b9b035d1f..8c709a5e2 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -9,6 +9,9 @@
     from vllm.sampling_params import SamplingParams
 
 
+__all__ = ["VLLM", "from_vllm"]
+
+
 class VLLM:
     """Represents a vLLM model.
 
@@ -168,22 +171,7 @@ def load_lora(self, adapter_path: Optional[str]):
             self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
 
 
-def vllm(model_name: str, **vllm_model_params):
-    """Load a vLLM model.
-
-    Parameters
-    ---------
-    model_name
-        The name of the model to load from the HuggingFace hub.
-    vllm_model_params
-        vLLM-specific model parameters. See the vLLM code for the full list:
-        https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/llm.py
-
-    """
-    from vllm import LLM
-
-    model = LLM(model_name, **vllm_model_params)
-
+def from_vllm(model: "LLM") -> VLLM:
     return VLLM(model)
 
 
diff --git a/pyproject.toml b/pyproject.toml
index fa9f5e724..c3fd2c053 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -47,7 +47,6 @@ dependencies = [
 dynamic = ["version"]
 
 [project.optional-dependencies]
-vllm = ["vllm", "transformers", "numpy<2"]
 transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
 exllamav2 = ["exllamav2"]
 dottxt = ["dottxt"]
diff --git a/tests/generate/test_integration_vllm.py b/tests/generate/test_integration_vllm.py
index d812d3cf2..faa8a404a 100644
--- a/tests/generate/test_integration_vllm.py
+++ b/tests/generate/test_integration_vllm.py
@@ -4,12 +4,14 @@
 import pytest
 import torch
 from pydantic import BaseModel, constr
+from vllm import LLM
 
 try:
     from vllm.sampling_params import SamplingParams
 except ImportError:
     pass
 
+import outlines
 import outlines.generate as generate
 import outlines.grammars as grammars
 import outlines.models as models
@@ -22,7 +24,7 @@
 
 @pytest.fixture(scope="module")
 def model():
-    return models.vllm("gpt2", gpu_memory_utilization=0.5)
+    return outlines.from_vllm(LLM("gpt2", gpu_memory_utilization=0.5))
 
 
 @pytest.mark.parametrize(

From 0058aef7cae6bcd6e24a2bea38bdfd7b448135e6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 21 Feb 2025 19:22:33 +0100
Subject: [PATCH 232/505] Initialize `MLXLM` from model and tokenizer instances

---
 docs/cookbook/extract_event_details.py |  9 +++--
 outlines/__init__.py                   |  2 +
 outlines/models/__init__.py            |  4 +-
 outlines/models/mlxlm.py               | 52 ++------------------------
 tests/models/test_mlxlm.py             | 26 ++++++-------
 5 files changed, 27 insertions(+), 66 deletions(-)

diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
index b51f8d921..2104bcac9 100644
--- a/docs/cookbook/extract_event_details.py
+++ b/docs/cookbook/extract_event_details.py
@@ -1,11 +1,14 @@
 from datetime import datetime
 
+from mlx_lm import load
 from pydantic import BaseModel, Field
 
-from outlines import generate, models
+import outlines
+from outlines.generate import Generator
+from outlines.types import JsonType
 
 # Load the model
-model = models.mlxlm("mlx-community/Hermes-3-Llama-3.1-8B-8bit")
+model = outlines.from_mlxlm(*load("mlx-community/Hermes-3-Llama-3.1-8B-8bit"))
 
 
 # Define the event schema using Pydantic
@@ -34,7 +37,7 @@ class Event(BaseModel):
 see you 😘 """
 
 # Create the generator
-generator = generate.json(model, Event)
+generator = Generator(model, JsonType(Event))
 
 # Extract the event information
 event = generator(prompt + message)
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 7f6360f0e..23c01cd4f 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -16,6 +16,7 @@
     from_anthropic,
     from_ollama,
     from_llamacpp,
+    from_mlxlm,
     from_vllm,
 )
 
@@ -24,6 +25,7 @@
     "from_anthropic",
     "from_gemini",
     "from_llamacpp",
+    "from_mlxlm",
     "from_ollama",
     "from_openai",
     "from_vllm",
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 77b0c60dc..c1a2a8e34 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -14,7 +14,7 @@
 from .exllamav2 import ExLlamaV2Model, exl2
 from .gemini import from_gemini, Gemini
 from .llamacpp import LlamaCpp, from_llamacpp
-from .mlxlm import MLXLM, mlxlm
+from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
 from .transformers import Transformers, TransformerTokenizer, Mamba
@@ -25,5 +25,5 @@
     Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM, Ollama
 ]
 
-LocalModel = Union[LlamaCpp, Transformers]
+LocalModel = Union[LlamaCpp, Transformers, MLXLM, VLLM]
 APIModel = Union[OpenAI, Anthropic, Gemini, Ollama, Dottxt]
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 843107d66..380fb0150 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -12,6 +12,9 @@
     from outlines.processors import OutlinesLogitsProcessor
 
 
+__all__ = ["MLXLM", "from_mlxlm"]
+
+
 class MLXLM:
     """
     Represents an `mlx_lm` model
@@ -196,52 +199,5 @@ def sample(logits: "mx.array") -> Tuple["mx.array", float]:
             unprocessed_input_ids = new_token_single
 
 
-def mlxlm(
-    model_name: str,
-    tokenizer_config: dict = {},
-    model_config: dict = {},
-    adapter_path: Optional[str] = None,
-    lazy: bool = False,
-):
-    """Instantiate a model from the `mlx_lm` library and its tokenizer.
-
-    Signature adapted from
-    https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L422
-
-    Parameters
-    ----------
-    Args:
-        path_or_hf_repo (Path): The path or the huggingface repository to load the model from.
-        tokenizer_config (dict, optional): Configuration parameters specifically for the tokenizer.
-            Defaults to an empty dictionary.
-        model_config(dict, optional): Configuration parameters specifically for the model.
-            Defaults to an empty dictionary.
-        adapter_path (str, optional): Path to the LoRA adapters. If provided, applies LoRA layers
-            to the model. Default: ``None``.
-        lazy (bool): If False eval the model parameters to make sure they are
-            loaded in memory before returning, otherwise they will be loaded
-            when needed. Default: ``False``
-
-    Returns
-    -------
-    A `MLXLM` model instance.
-
-    """
-    try:
-        import mlx.core as mx
-        import mlx_lm
-    except ImportError:
-        raise ImportError(
-            "The `mlx_lm` library needs to be installed in order to use `mlx_lm` models."
-        )
-    if not mx.metal.is_available():
-        raise RuntimeError("You cannot use `mlx_lm` without Apple Silicon (Metal)")
-
-    model, tokenizer = mlx_lm.load(
-        model_name,
-        tokenizer_config=tokenizer_config,
-        model_config=model_config,
-        adapter_path=adapter_path,
-        lazy=lazy,
-    )
+def from_mlxlm(model: "nn.Module", tokenizer: "PreTrainedTokenizer") -> MLXLM:
     return MLXLM(model, tokenizer)
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index 20e59da81..10280b756 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -1,9 +1,10 @@
 import pytest
 
-from outlines.models.mlxlm import mlxlm
+import outlines
 from outlines.models.transformers import TransformerTokenizer
 
 try:
+    import mlx_lm
     import mlx.core as mx
 
     HAS_MLX = mx.metal.is_available()
@@ -14,18 +15,21 @@
 TEST_MODEL = "mlx-community/SmolLM-135M-Instruct-4bit"
 
 
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    model, tokenizer = mlx_lm.load(TEST_MODEL)
+    return outlines.from_mlxlm(model, tokenizer)
+
+
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_model():
-    model = mlxlm(TEST_MODEL)
+def test_mlxlm_model(model):
     assert hasattr(model, "model")
     assert hasattr(model, "tokenizer")
     assert isinstance(model.tokenizer, TransformerTokenizer)
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_tokenizer():
-    model = mlxlm(TEST_MODEL)
-
+def test_mlxlm_tokenizer(model):
     # Test single string encoding/decoding
     test_text = "Hello, world!"
     token_ids = mx.array(model.mlx_tokenizer.encode(test_text))
@@ -33,10 +37,9 @@ def test_mlxlm_tokenizer():
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_generate():
+def test_mlxlm_generate(model):
     from outlines.generate.api import GenerationParameters, SamplingParameters
 
-    model = mlxlm(TEST_MODEL)
     prompt = "Write a haiku about programming:"
 
     # Test with basic generation parameters
@@ -54,10 +57,9 @@ def test_mlxlm_generate():
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_stream():
+def test_mlxlm_stream(model):
     from outlines.generate.api import GenerationParameters, SamplingParameters
 
-    model = mlxlm(TEST_MODEL)
     prompt = "Count from 1 to 5:"
 
     gen_params = GenerationParameters(max_tokens=20, stop_at=None, seed=None)
@@ -83,9 +85,7 @@ def test_mlxlm_stream():
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_errors():
-    model = mlxlm(TEST_MODEL)
-
+def test_mlxlm_errors(model):
     # Test batch inference (should raise NotImplementedError)
     with pytest.raises(NotImplementedError):
         from outlines.generate.api import GenerationParameters, SamplingParameters

From 05ca83da7300d73db5e586327642ea32f113c8bb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 22 Feb 2025 11:28:04 +0100
Subject: [PATCH 233/505] Initialize `Transformers` from model and tokenizer
 instances

The `Mamba` class is not necessary anymore. More generally, this makes
outlines compatible with any model class that implements the
`transformers` API.
---
 outlines/__init__.py              |   3 +-
 outlines/models/__init__.py       |   2 +-
 outlines/models/transformers.py   | 105 +++++++-----------------------
 pyproject.toml                    |   3 -
 tests/generate/test_generate.py   |   6 +-
 tests/models/test_transformers.py |  41 +++++-------
 6 files changed, 46 insertions(+), 114 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 23c01cd4f..51b21ecf2 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -12,6 +12,7 @@
 
 from outlines.models import (
     from_openai,
+    from_transformers,
     from_gemini,
     from_anthropic,
     from_ollama,
@@ -28,7 +29,7 @@
     "from_mlxlm",
     "from_ollama",
     "from_openai",
-    "from_vllm",
+    "from_transformersfrom_vllm",
 ]
 
 __all__ = [
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index c1a2a8e34..1e6863113 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -17,7 +17,7 @@
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
-from .transformers import Transformers, TransformerTokenizer, Mamba
+from .transformers import Transformers, TransformerTokenizer, from_transformers
 from .transformers_vision import TransformersVision
 from .vllm import VLLM, from_vllm
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 3d87443f6..74f0d3b1a 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -6,12 +6,10 @@
 
 if TYPE_CHECKING:
     import torch
-    from transformers import PreTrainedTokenizer
+    from transformers import PreTrainedTokenizer, PreTrainedModel
 
-__all__ = ["Transformers", "Mamba"]
 
-
-KVCacheType = Tuple[Tuple["torch.DoubleTensor", "torch.DoubleTensor"], ...]
+__all__ = ["Transformers", "from_transformers"]
 
 
 def get_llama_tokenizer_types():
@@ -168,52 +166,28 @@ class Transformers(Model):
 
     def __init__(
         self,
-        model_name: str,
-        model_class=None,
-        model_kwargs: dict = {},
-        tokenizer_class=None,
-        tokenizer_kwargs: dict = {},
+        model: "PreTrainedModel",
+        tokenizer: "PreTrainedTokenizer",
     ):
-        """Create a Transformers model instance
+        """Create a Transformers model instance.
+
+        `outlines` supports `PreTrainedModelForCausalLM`,
+        `PreTrainedMambaForCausalLM`, `PreTrainedModelForSeq2Seq` and any model
+        that implements the `transformers` model API.
 
         Parameters:
         ----------
-        model_name
-            The name of the transformers model to use;
-        model_class
-            The Transformers model class from which to create the model.
-            If not provided,`AutoModelForCausalLM` will be used.
-            If you gave the name of a non-causal language model,
-            you must provide a value for this parameter.
-        model_kwargs
-            A dictionary of keyword arguments to pass to the `from_pretrained`
-            method of the model class.
-        tokenizer_class
-            The Transformers tokenizer class from which to create the tokenizer.
-            If not provided,`AutoTokenizer` will be used.
-            If you gave the name of a model that is not compatible with `AutoTokenizer`,
-            you must provide a value for this parameter.
-        tokenizer_kwargs
-            A dictionary of keyword arguments to pass to the `from_pretrained`
-            method of the tokenizer class.
+        model
+            A `PreTrainedModel`, or any model that is compatible with the
+            `transformers` API for models.
+        tokenizer
+            A `PreTrainedTokenizer`, or any tokenizer that is compatible with
+            the `transformers` API for tokenizers.
 
         """
-        if model_class is None or tokenizer_class is None:
-            try:
-                from transformers import AutoModelForCausalLM, AutoTokenizer
-            except ImportError:
-                raise ImportError(
-                    "The `transformers` library needs to be installed in order to use `transformers` models."
-                )
-        if model_class is None:
-            model_class = AutoModelForCausalLM
-        if tokenizer_class is None:
-            tokenizer_class = AutoTokenizer
-        self.model = model_class.from_pretrained(model_name, **model_kwargs)
-        tokenizer_kwargs.setdefault("padding_side", "left")
-        self.tokenizer = TransformerTokenizer(
-            tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
-        )
+        tokenizer.padding_size = "left"
+        self.model = model
+        self.tokenizer = TransformerTokenizer(tokenizer)
         self.type_adapter = TransformersTypeAdapter()
 
     def generate(self, model_input, output_type, **inference_kwargs):
@@ -278,41 +252,8 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
             )
 
 
-class Mamba(Transformers):
-    """Represents a Mamba model."""
-
-    def __init__(
-        self,
-        model_name: str,
-        model_kwargs: dict = {},
-        tokenizer_kwargs: dict = {},
-    ):
-        """
-        Create a Mamba model instance
-
-        Parameters:
-        ----------
-        model_name
-            The name of the transformers model to use. It will be passed to
-            the `from_pretrained` method of the `MambaForCausalLM` class.
-        model_kwargs
-            A dictionary of keyword arguments to pass to the `from_pretrained`
-            method of the `MambaForCausalLM` class.
-        tokenizer_kwargs
-            A dictionary of keyword arguments to pass to the `from_pretrained`
-            method of the `AutoTokenizer` class.
-        """
-        try:
-            from transformers import MambaForCausalLM
-
-        except ImportError:
-            raise ImportError(
-                "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
-            )
-
-        return super().__init__(
-            model_name=model_name,
-            model_class=MambaForCausalLM,
-            model_kwargs=model_kwargs,
-            tokenizer_kwargs=tokenizer_kwargs,
-        )
+def from_transformers(
+    model: Union["PreTrainedModel", object],
+    tokenizer: Union["PreTrainedTokenizer", object],
+):
+    return Transformers(model, tokenizer)
diff --git a/pyproject.toml b/pyproject.toml
index c3fd2c053..2c5382e79 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -47,9 +47,6 @@ dependencies = [
 dynamic = ["version"]
 
 [project.optional-dependencies]
-transformers = ["transformers", "accelerate", "datasets", "numpy<2"]
-exllamav2 = ["exllamav2"]
-dottxt = ["dottxt"]
 test = [
     "pre-commit",
     "pytest",
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index aaa4509bf..64dee6e4b 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -71,9 +71,7 @@ def model_mamba(tmp_path_factory):
 def model_bart(tmp_path_factory):
     from transformers import AutoModelForSeq2SeqLM
 
-    return models.Transformers(
-        "facebook/bart-base", model_class=AutoModelForSeq2SeqLM
-    )
+    return models.Transformers("facebook/bart-base", model_class=AutoModelForSeq2SeqLM)
 
 
 @pytest.fixture(scope="session")
@@ -281,7 +279,7 @@ def test_generate_json(request, model_fixture, sample_schema):
 
 
 # TODO: add support for genson in the Regex type of v1.0
-#def test_integrate_genson_generate_json(request):
+# def test_integrate_genson_generate_json(request):
 #    from genson import SchemaBuilder
 #
 #    builder = SchemaBuilder()
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index f52b89002..f2fb90424 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -3,9 +3,10 @@
 
 import pytest
 from pydantic import BaseModel
-from transformers import AutoModelForSeq2SeqLM
+import transformers
 
-from outlines.models.transformers import Mamba, Transformers
+import outlines
+from outlines.models.transformers import Transformers
 from outlines.types import Choice, JsonType, Regex
 
 TEST_MODEL = "erwanf/gpt2-mini"
@@ -14,42 +15,36 @@
 
 
 def test_transformers_instantiate_simple():
-    model = Transformers(TEST_MODEL)
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+    )
     assert isinstance(model, Transformers)
 
 
-def test_transformers_instantiate_wrong_kwargs():
-    with pytest.raises(TypeError):
-        Transformers(TEST_MODEL, model_kwargs={"foo": "bar"})
-
-
 def test_transformers_instantiate_other_model_class():
-    model = Transformers(
-        model_name=TEST_MODEL_SEQ2SEQ, model_class=AutoModelForSeq2SeqLM
+    model = outlines.from_transformers(
+        transformers.AutoModelForSeq2SeqLM.from_pretrained(TEST_MODEL_SEQ2SEQ),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
     )
     assert isinstance(model, Transformers)
 
 
 def test_transformers_instantiate_mamba():
-    model = Mamba(
-        model_name=TEST_MODEL_MAMBA,
+    model = outlines.from_transformers(
+        transformers.MambaForCausalLM.from_pretrained(TEST_MODEL_MAMBA),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
     )
-    assert isinstance(model, Mamba)
     assert isinstance(model, Transformers)
 
 
-def test_transformers_instantiate_tokenizer_kwargs():
-    model = Transformers(
-        TEST_MODEL,
-        tokenizer_kwargs={"additional_special_tokens": ["<t1>", "<t2>"]}
-    )
-    assert "<t1>" in model.tokenizer.special_tokens
-    assert "<t2>" in model.tokenizer.special_tokens
-
-
 @pytest.fixture
 def model():
-    return Transformers(TEST_MODEL)
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+    )
+    return model
 
 
 def test_transformers_simple(model):

From c22936a6a461b923bd45a83ebaea0efd415d1672 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 22 Feb 2025 12:55:40 +0100
Subject: [PATCH 234/505] This is a combination of 2 commits.

Merge `transformers` vision models with text-2-text models
---
 examples/bentoml/service.py              |  14 ++-
 outlines/models/__init__.py              |   8 +-
 outlines/models/transformers.py          | 131 +++++++++++++++++++-
 outlines/models/transformers_vision.py   | 146 -----------------------
 tests/models/test_transformers_vision.py |  22 ++--
 5 files changed, 150 insertions(+), 171 deletions(-)

diff --git a/examples/bentoml/service.py b/examples/bentoml/service.py
index 98d46894a..884904a73 100644
--- a/examples/bentoml/service.py
+++ b/examples/bentoml/service.py
@@ -1,7 +1,7 @@
 import typing as t
 
 import bentoml
-from import_model import BENTO_MODEL_TAG
+from import_model import BENTO_MODEL_TAG, MODEL_ID
 
 DEFAULT_SCHEMA = """{
     "title": "Character",
@@ -55,13 +55,17 @@ class Outlines:
 
     def __init__(self) -> None:
         import torch
+        from transformers import AutoModelForCausalLM, AutoTokenizer
 
         import outlines
 
-        self.model = outlines.models.transformers(
-            self.bento_model_ref.path,
-            device="cuda",
-            model_kwargs={"torch_dtype": torch.float16},
+        self.model = outlines.from_transformers(
+            AutoTokenizer.from_pretrained(MODEL_ID),
+            AutoModelForCausalLM.from_pretrained(
+                MODEL_ID,
+                torch_dtype=torch.float16,
+                low_cpu_mem_usage=True,
+            )
         )
 
     @bentoml.api
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 1e6863113..a0b4a85d3 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -17,8 +17,12 @@
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
-from .transformers import Transformers, TransformerTokenizer, from_transformers
-from .transformers_vision import TransformersVision
+from .transformers import (
+    Transformers,
+    TransformerTokenizer,
+    TransformersVision,
+    from_transformers,
+)
 from .vllm import VLLM, from_vllm
 
 LogitsGenerator = Union[
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 74f0d3b1a..b65d0cefb 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -190,7 +190,7 @@ def __init__(
         self.tokenizer = TransformerTokenizer(tokenizer)
         self.type_adapter = TransformersTypeAdapter()
 
-    def generate(self, model_input, output_type, **inference_kwargs):
+    def _prepare_model_inputs(self, model_input, output_type):
         prompts = self.type_adapter.format_input(model_input)
         input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
@@ -198,6 +198,10 @@ def generate(self, model_input, output_type, **inference_kwargs):
             "attention_mask": attention_mask.to(self.model.device),
         }
 
+        return prompts, inputs
+
+    def generate(self, model_input, output_type, **inference_kwargs):
+        prompts, inputs = self._prepare_model_inputs(model_input, output_type)
         logits_processor = self.type_adapter.format_output_type(output_type)
 
         generated_ids = self._generate_output_seq(
@@ -205,7 +209,7 @@ def generate(self, model_input, output_type, **inference_kwargs):
         )
 
         # if single str input, convert to a 1D outputt
-        if isinstance(model_input, str):
+        if isinstance(prompts, str):
             generated_ids = generated_ids.squeeze(0)
 
         return self._decode_generation(generated_ids)
@@ -252,8 +256,125 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
             )
 
 
+class TransformersVisionTypeAdapter(ModelTypeAdapter):
+    """Type adapter for TransformersVision models."""
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. Please provide a "
+            "dictionary with the following format: "
+            "{'prompts': Union[str, List[str]], 'images': Union[Any, List[Any]]}"
+            "Make sure the number of image tags in the prompts is equal to the "
+            "number of images provided."
+        )
+
+    @format_input.register(dict)
+    def format_list_input(self, model_input):
+        if "prompts" not in model_input or "images" not in model_input:
+            raise ValueError(
+                "The input must contain the following keys: 'prompts' and 'images'."
+            )
+        return model_input["prompts"], model_input["images"]
+
+    def format_output_type(self, output_type):
+        """Generate the logits processor argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The logits processor provided.
+
+        """
+        from transformers import LogitsProcessorList
+
+        if output_type is not None:
+            return LogitsProcessorList([output_type])
+        return None
+
+
+class TransformersVision(Transformers):
+    """Represents a `transformers` model with a vision processor."""
+
+    def __init__(self, model: "PreTrainedModel", processor):
+        """Create a TransformersVision model instance
+
+        We rely on the `__init__` method of the `Transformers` class to handle
+        most of the initialization and then add elements specific to vision
+        models.
+
+        """
+        self.processor = processor
+        self.processor.padding_side = "left"
+        self.processor.pad_token = "[PAD]"
+
+        tokenizer: "PreTrainedTokenizer" = self.processor.tokenizer
+
+        super().__init__(model, tokenizer)
+
+        self.type_adapter = TransformersVisionTypeAdapter()
+
+    def _prepare_model_inputs(self, model_input, output_type):
+        prompts, images = self.type_adapter.format_input(model_input)
+        inputs = self.processor(
+            text=prompts, images=images, padding=True, return_tensors="pt"
+        ).to(self.model.device)
+
+        return prompts, inputs
+
+
 def from_transformers(
-    model: Union["PreTrainedModel", object],
-    tokenizer: Union["PreTrainedTokenizer", object],
+    model: "PreTrainedModel",
+    tokenizer_or_processor: "PreTrainedTokenizer",
 ):
-    return Transformers(model, tokenizer)
+    from transformers import (
+        PreTrainedTokenizer,
+        PreTrainedTokenizerFast,
+        Blip2Processor,
+        LlavaProcessor,
+        IdeficsProcessor,
+        CLIPProcessor,
+        Qwen2_5_VLProcessor,
+        Qwen2VLProcessor,
+        NougatProcessor,
+        LlavaNextProcessor,
+        PixtralProcessor,
+        PaliGemmaProcessor,
+    )
+
+    vision_processors = (
+        Blip2Processor,
+        LlavaProcessor,
+        IdeficsProcessor,
+        CLIPProcessor,
+        Qwen2_5_VLProcessor,
+        Qwen2VLProcessor,
+        NougatProcessor,
+        LlavaNextProcessor,
+        PixtralProcessor,
+        PaliGemmaProcessor,
+    )
+
+    if isinstance(
+        tokenizer_or_processor, (PreTrainedTokenizer, PreTrainedTokenizerFast)
+    ):
+        tokenizer = tokenizer_or_processor
+        return Transformers(model, tokenizer)
+    elif isinstance(tokenizer_or_processor, vision_processors):
+        processor = tokenizer_or_processor
+        return TransformersVision(model, processor)
+    else:
+        raise ValueError(
+            "We could determine whether the model passed to `from_transformers`"
+            + " is a text-2-text of vision language model. If you passed a"
+            + " vision language model please open an issue pasting the error"
+            + " with the name of the vision language model."
+        )
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
index b1c1c1de4..e69de29bb 100644
--- a/outlines/models/transformers_vision.py
+++ b/outlines/models/transformers_vision.py
@@ -1,146 +0,0 @@
-from functools import singledispatchmethod
-
-from outlines.models import Transformers
-from outlines.models.base import ModelTypeAdapter
-
-
-class TransformersVisionTypeAdapter(ModelTypeAdapter):
-    """Type adapter for TransformersVision models."""
-
-    @singledispatchmethod
-    def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
-
-        Argument
-        --------
-        model_input
-            The input passed by the user.
-
-        """
-        raise NotImplementedError(
-            f"The input type {input} is not available. Please provide a "
-            "dictionary with the following format: "
-            "{'prompts': Union[str, List[str]], 'images': Union[Any, List[Any]]}"
-            "Make sure the number of image tags in the prompts is equal to the "
-            "number of images provided."
-        )
-
-    @format_input.register(dict)
-    def format_list_input(self, model_input):
-        if "prompts" not in model_input or "images" not in model_input:
-            raise ValueError(
-                "The input must contain the following keys: 'prompts' and 'images'."
-            )
-        return model_input["prompts"], model_input["images"]
-
-    def format_output_type(self, output_type):
-        """Generate the logits processor argument to pass to the model.
-
-        Argument
-        --------
-        output_type
-            The logits processor provided.
-
-        """
-        from transformers import LogitsProcessorList
-
-        if output_type is not None:
-            return LogitsProcessorList([output_type])
-        return None
-
-
-class TransformersVision(Transformers):
-    """Represents a `transformers` model with a vision processor."""
-
-    def __init__(
-        self,
-        model_name: str,
-        model_class,
-        model_kwargs: dict = {},
-        tokenizer_class=None,
-        tokenizer_kwargs: dict = {},
-        processor_class=None,
-        processor_kwargs: dict = {},
-    ):
-        """Create a TransformersVision model instance
-
-        We rely on the `__init__` method of the `Transformers` class to handle
-        most of the initialization and then add elements specific to vision
-        models.
-
-        Parameters
-        ----------
-        model_name
-            The name of the transformers model to use;
-        model_class
-            The Transformers model class from which to create the model.
-        model_kwargs
-            A dictionary of keyword arguments to pass to the `from_pretrained`
-            method of the model class.
-        tokenizer_class
-            The Transformers tokenizer class from which to create the tokenizer.
-            If not provided,`AutoTokenizer` will be used.
-            If you gave the name of a model that is not compatible with `AutoTokenizer`,
-            you must provide a value for this parameter.
-        tokenizer_kwargs
-            A dictionary of keyword arguments to pass to the `from_pretrained`
-            method of the tokenizer class.
-        processor_class
-            The Transformers processor class from which to create the processor.
-            If not provided,`AutoProcessor` will be used.
-            If you gave the name of a model that is not compatible with `AutoProcessor`,
-            you must provide a value for this parameter.
-        processor_kwargs
-            A dictionary of keyword arguments to pass to the `from_pretrained`
-            method of the processor class.
-
-        """
-        if processor_class is None:
-            try:
-                from transformers import AutoProcessor
-
-                processor_class = AutoProcessor
-            except ImportError:
-                raise ImportError(
-                    "The `transformers` library needs to be installed in order to use `transformers` models."
-                )
-
-        processor_kwargs.setdefault("padding_side", "left")
-        processor_kwargs.setdefault("pad_token", "[PAD]")
-        self.processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-        if tokenizer_class is None and getattr(self.processor, "tokenizer", None):
-            tokenizer_class = type(self.processor.tokenizer)
-
-        super().__init__(
-            model_name,
-            model_class,
-            model_kwargs,
-            tokenizer_class,
-            tokenizer_kwargs,
-        )
-
-        self.type_adapter = TransformersVisionTypeAdapter()
-
-    def generate(self, model_input, output_type, **inference_kwargs):
-        prompts, images = self.type_adapter.format_input(model_input)
-
-        inputs = self.processor(
-            text=prompts, images=images, padding=True, return_tensors="pt"
-        ).to(self.model.device)
-        logits_processor = self.type_adapter.format_output_type(output_type)
-
-        generated_ids = self._generate_output_seq(
-            prompts, inputs, logits_processor=logits_processor, **inference_kwargs
-        )
-
-        # if single str input, convert to a 1D outputt
-        if isinstance(prompts, str):
-            generated_ids = generated_ids.squeeze(0)
-
-        return self._decode_generation(generated_ids)
-
-    def stream(self, model_input, output_type, **inference_kwargs):
-        raise NotImplementedError(
-            "Streaming is not implemented for TransformersVision models."
-        )
diff --git a/tests/models/test_transformers_vision.py b/tests/models/test_transformers_vision.py
index 477a2f456..57dedcb69 100644
--- a/tests/models/test_transformers_vision.py
+++ b/tests/models/test_transformers_vision.py
@@ -11,9 +11,11 @@
     CLIPModel,
     CLIPProcessor,
     LlavaForConditionalGeneration,
+    AutoProcessor,
 )
 
-from outlines.models.transformers_vision import TransformersVision
+import outlines
+from outlines.models.transformers import TransformersVision
 from outlines.types import Choice, JsonType, Regex
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
@@ -35,22 +37,16 @@ def img_from_url(url):
 
 @pytest.fixture
 def model():
-    return TransformersVision(
-        model_name=TEST_MODEL, model_class=LlavaForConditionalGeneration
+    return outlines.from_transformers(
+        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+        AutoProcessor.from_pretrained(TEST_MODEL),
     )
 
 
 def test_transformers_vision_instantiate_simple():
-    model = TransformersVision(
-        model_name=TEST_MODEL,
-        model_class=Blip2ForConditionalGeneration,
-    )
-    assert isinstance(model, TransformersVision)
-
-
-def test_transformers_vision_instantiate_other_processor_class():
-    model = TransformersVision(
-        model_name=TEST_CLIP_MODEL, model_class=CLIPModel, processor_class=CLIPProcessor
+    model = outlines.from_transformers(
+        Blip2ForConditionalGeneration.from_pretrained(TEST_MODEL),
+        AutoProcessor.from_pretrained(TEST_MODEL),
     )
     assert isinstance(model, TransformersVision)
 

From a0b2d2bfac525be97a5911a74ab83805ac4b3220 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 22 Feb 2025 11:06:01 +0100
Subject: [PATCH 235/505] Update the model documentation

---
 docs/cookbook/extract_event_details.py |   1 +
 docs/reference/models/anthropic.md     |  35 +++--
 docs/reference/models/gemini.md        |  39 +++--
 docs/reference/models/llamacpp.md      | 124 +++------------
 docs/reference/models/mlxlm.md         |  57 +++----
 docs/reference/models/models.md        |  96 ++++++------
 docs/reference/models/ollama.md        |   2 +
 docs/reference/models/openai.md        |  86 ++++-------
 docs/reference/models/transformers.md  |  80 ++--------
 docs/reference/models/vllm.md          | 203 ++-----------------------
 outlines/__init__.py                   |   1 +
 11 files changed, 184 insertions(+), 540 deletions(-)

diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
index 2104bcac9..a3231cd00 100644
--- a/docs/cookbook/extract_event_details.py
+++ b/docs/cookbook/extract_event_details.py
@@ -7,6 +7,7 @@
 from outlines.generate import Generator
 from outlines.types import JsonType
 
+
 # Load the model
 model = outlines.from_mlxlm(*load("mlx-community/Hermes-3-Llama-3.1-8B-8bit"))
 
diff --git a/docs/reference/models/anthropic.md b/docs/reference/models/anthropic.md
index ffc510f43..d3db67ddb 100644
--- a/docs/reference/models/anthropic.md
+++ b/docs/reference/models/anthropic.md
@@ -2,44 +2,43 @@
 
 !!! Installation
 
-    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines. Or alternatively you can run:
-
-    ```bash
-    pip install "outlines[anthropic]"
-    ```
+    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines: `pip install anthropic`.
 
 ## Anthropic models
 
 Outlines supports models available via the Anthropic API, e.g. Claude 3.5 Haiku or Claude 3.5 Sonner. You can initialize the model by passing the model name to `outlines.models.Anthropic`:
 
 ```python
-from outlines import models
+from anthropic import Anthropic
+import outlines
 
-model = models.Anthropic("claude-3-5-haiku-latest")
-model = models.Anthropic("claude-3-5-sonnet-latest")
+model = outlines.from_anthropic(Anthropic("claude-3-5-haiku-latest"))
+model = outlines.from_anthropic(Anthropic("claude-3-5-sonnet-latest"))
 ```
 
-Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-claude/models) for an up-to-date list of available models. You can pass any paramater you would pass to the Anthropic SDK as keyword arguments:
-
-```python
-model = models.Anthropic(
-    "claude-3.5-haiku-latest",
-    api_key="<my api key>"
-)
-```
+Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-claude/models) for an up-to-date list of available models.
 
 ## Text generation
 
 To generate text using an Anthropic model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
 
 ```python
-from outlines import models, Generator
+from anthropic import Anthropic
+import outlines
 
-model = models.Anthropic("claude-3-5-haiku-latest")
+model = outlines.from_anthropic(Anthropic("claude-3-5-haiku-latest"))
 generator = Generator(model)
 result = generator("Prompt", max_tokens=1024)
+
+# Call the model directly
+result = model("Prompt", max_tokens=1024)
 ```
 
+
+!!! Warning
+
+    You must set a value for `max_tokens` with Anthropic models.
+
 See the [Anthropic SDK documentation](https://github.com/anthropics/anthropic-sdk-python/blob/main/src/anthropic/resources/messages.py) for the list of available arguments.
 
 The Anthropic API currently does not support structured generation.
diff --git a/docs/reference/models/gemini.md b/docs/reference/models/gemini.md
index 472f1839a..ba83d59ec 100644
--- a/docs/reference/models/gemini.md
+++ b/docs/reference/models/gemini.md
@@ -2,7 +2,7 @@
 
 !!! Installation
 
-    You need to install the `google-generativeai` library to be able to use the Gemini API in Outlines. Or alternatively you can run:
+    You need to install the `google-generativeai` library to be able to use the Gemini API in Outlines: `pip install google-generativeai`
 
 
 ## Gemini models
@@ -10,10 +10,11 @@
 Outlines supports models available via the Gemini API, e.g. Gemini 1.5. You can initialize the model by passing the model name to `outlines.models.Gemini`:
 
 ```python
-from outlines import models
+import outlines
+import google.generativeai as genai
 
-model = models.Gemini("gemini-1-5-flash")
-model = models.Gemini("gemini-1-5-pro")
+model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
+model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-pro"))
 ```
 
 Check the [Gemini documentation](https://ai.google.dev/gemini-api/docs/models/gemini) for an up-to-date list of available models.
@@ -23,11 +24,15 @@ Check the [Gemini documentation](https://ai.google.dev/gemini-api/docs/models/ge
 To generate text using a Gemini model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
 
 ```python
-from outlines import models, Generator
+import outlines
+import google.generativeai as genai
 
-model = models.Gemini("gemini-1-5-flash")
-generator = Generator(model)
+model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
+generator = outlines.Generator(model)
 result = generator("Prompt", max_tokens=1024)
+
+# Call the model directly
+result = model("Prompt", max_tokens=1024)
 ```
 
 ### Structured generation
@@ -40,17 +45,22 @@ Outlines provides support for JSON Schema-based structured generation with the G
 
 ```python
 from collections import TypedDict
-from outlines import Generator, models
-from outlines.types import Json
 
-model = models.Gemini("gemini-1-5-flash")
+import google.generativeai as genai
+
+import outlines
+from outlines import Generator
+from outlines.types import JsonType
+
+
+model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
 
 class Person(TypedDict):
     first_name: str
     last_name: str
     age: int
 
-generator = Generator(model, Json(Person))
+generator = Generator(model, JsonType(Person))
 generator("current indian prime minister on january 1st 2023")
 # Person(first_name='Narendra', last_name='Modi', age=72)
 ```
@@ -65,10 +75,13 @@ Outlines provides support for multiple-choices structured generation. Enums and
 
 ```python
 from enum import Enum
-from outlines import Generator, models
+
+import google.generativeai as genai
+
+from outlines import Generator
 from outlines.types import Choice
 
-model = models.Gemini("gemini-1-5-flash")
+model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
 
 class Foo(Enum):
     foo = "Foo"
diff --git a/docs/reference/models/llamacpp.md b/docs/reference/models/llamacpp.md
index 9ea0c5c09..313f58eb9 100644
--- a/docs/reference/models/llamacpp.md
+++ b/docs/reference/models/llamacpp.md
@@ -2,49 +2,36 @@
 
 Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library][llamacpp]. Llamacpp allows to run quantized models on machines with limited compute.
 
-!!! Note "Installation"
+!!! Note "Documentation"
 
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration. See the [installation section](#installation) for instructions to install `llama-cpp-python` with CUDA, Metal, ROCm and other backends. To get started quickly you can also run:
+    To be able to use llama.cpp in Outlines, you must install the `llama-cpp-python` library, `pip install llama-cpp-python`
+
+    Consult the [`llama-cpp-python` documentation](https://llama-cpp-python.readthedocs.io/en/latest) for detailed informations about how to initialize model and the available options.
 
-    ```bash
-    pip install "outlines[llamacpp]"
-    ```
 
 ## Load the model
 
 You can use `outlines.from_llamacpp` to load a `llama-cpp-python` model. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
 
 ```python
-from outlines import models
+from llama_cpp import Llama
+import outlines
 
-llm = models.LlamaCpp("./phi-2.Q4_K_M.gguf")
+llm = outlines.from_llamacpp(Llama("./phi-2.Q4_K_M.gguf"))
 ```
 
 You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 
 
 ```python
-from outlines import models
+from llama_cpp import Llama
+import outlines
 
-model = models.LlamaCpp.from_pretrained("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+model = outlines.from_llamacpp(Llama.from_pretrained("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf"))
 ```
 
-This will download the model files to the hub cache folder and load the weights in memory.
-
-
-You can pass the same keyword arguments to the model as you would pass in the [llama-ccp-library][llamacpp]:
-
-```python
-from outlines import models
-
-model = models.LlamaCpp(
-    "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf"
-    n_ctx=512,  # to set the context length value
-)
-```
+This will download the model files to the hub cache folder and load the weights in memory. See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters you can pass to the `Llama` class.
 
-See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters.
 
 ### Load the model on GPU
 
@@ -55,13 +42,14 @@ See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io
  To load the model on GPU, pass `n_gpu_layers=-1`:
 
 ```python
-from outlines import models
+from llama_cpp import Llama
+import outlines
 
-model = models.LlamaCpp(
+model = outlines.from_llamacpp(Llama(
     "TheBloke/phi-2-GGUF",
     "phi-2.Q4_K_M.gguf",
     n_gpu_layers=-1,  # to use GPU acceleration
-)
+))
 ```
 
 
@@ -71,11 +59,15 @@ model = models.LlamaCpp(
 To generate text you must first create a `Generator` object by passing the model instance and, possibley, the expected output type:
 
 ```python
-from outlines import models, generate
+import outlines
 
 
-model = models.LlamaCpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-generator = Generator(model)
+model = outlines.from_llamacpp(Llama(
+    "TheBloke/phi-2-GGUF",
+    "phi-2.Q4_K_M.gguf",
+    n_gpu_layers=-1,  # to use GPU acceleration
+))
+generator = outlines.Generator(model)
 ```
 
 You can pass to the generator the same keyword arguments you would pass in `llama-cpp-python`:
@@ -91,78 +83,6 @@ tokens = generator.stream("A prompt")
 ```
 
 
-## Installation
-
-You need to install the `llama-cpp-python` library to use the llama.cpp integration.
-
-### CPU
-
-For a *CPU-only* installation run:
-
-```bash
-pip install llama-cpp-python
-```
-
-!!! Warning
-
-    Do not run this command if you want support for BLAS, Metal or CUDA. Follow the instructions below instead.
-
-### CUDA
-
-```bash
-CMAKE_ARGS="-DLLAMA_CUDA=on" pip install llama-cpp-python
-```
-
-It is also possible to install pre-built wheels with CUDA support (Python 3.10 and above):
-
-```bash
-pip install llama-cpp-python \
-  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/<cuda-version>
-```
-
-Where `<cuda-version>` is one of the following, depending on the version of CUDA installed on your system:
-
-- `cu121` for CUDA 12.1
-- `cu122` for CUDA 12.2
-- `cu123` CUDA 12.3
-
-### Metal
-
-```bash
-CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
-```
-
-It is also possible to install pre-build wheels with Metal support (Python 3.10 or above, MacOS 11.0 and above):
-
-```bash
-pip install llama-cpp-python \
-  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/metal
-```
-
-### OpenBLAS
-
-```bash
-CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
-```
-
-### Other backend
-
-`llama.cpp` supports many other backends. Refer to the [llama.cpp documentation][llama-cpp-python-install] to use the following backends:
-
-- CLBast (OpenCL)
-- hipBLAS (ROCm)
-- Vulkan
-- Kompute
-- SYCL
-
-
 [llamacpp]: https://github.com/abetlen/llama-cpp-python
 [llama-cpp-python-call]: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
 [llama-cpp-python-install]: https://github.com/abetlen/llama-cpp-python/tree/08b16afe11e7b42adec2fed0a781123383476045?tab=readme-ov-file#supported-backends
-[llama-cpp-sampling-params]: https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
-[mirostat]: https://arxiv.org/abs/2007.14966
-[degeneration]: https://arxiv.org/abs/1904.09751
-[top-k]: https://arxiv.org/abs/1805.04833
-[minimum-p]: https://github.com/ggerganov/llama.cpp/pull/3841
-[locally-typical]: https://arxiv.org/abs/2202.00666
-[tail-free]: https://www.trentonbricken.com/Tail-Free-Sampling
diff --git a/docs/reference/models/mlxlm.md b/docs/reference/models/mlxlm.md
index d435b9c1f..33953c11f 100644
--- a/docs/reference/models/mlxlm.md
+++ b/docs/reference/models/mlxlm.md
@@ -4,11 +4,9 @@ Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx
 
 !!! Note "Installation"
 
-    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To get started quickly you can also run:
+    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration: `pip install mlx mlx-lm`.
 
-    ```bash
-    pip install "outlines[mlxlm]"
-    ```
+    Consult the [`mlx-lm` documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms) for detailed informations about how to initialize OpenAI clients and the available options.
 
 
 ## Load the model
@@ -16,34 +14,14 @@ Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx
 You can initialize the model by passing the name of the repository on the HuggingFace Hub. The official repository for mlx-lm supported models is [mlx-community](https://huggingface.co/mlx-community).
 
 ```python
-from outlines import models
+from mlx_lm import load
+import outlines
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
 ```
 
 This will download the model files to the hub cache folder and load the weights in memory.
 
-The arguments `model_config` and `tokenizer_config` are available to modify loading behavior. For example, per the `mlx-lm` [documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms#supported-models), you must set an eos_token for `qwen/Qwen-7B`. In outlines you may do so via
-
-```
-model = models.mlxlm(
-    "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit",
-    tokenizer_config={"eos_token": "<|endoftext|>", "trust_remote_code": True},
-)
-```
-
-**Main parameters:**
-
-(Subject to change. Table based on [mlx-lm.load docstring](https://github.com/ml-explore/mlx-examples/blob/main/llms/mlx_lm/utils.py#L429))
-
-| Parameters         | Type   | Description                                                                                      | Default |
-|--------------------|--------|--------------------------------------------------------------------------------------------------|---------|
-| `tokenizer_config` | `dict` | Configuration parameters specifically for the tokenizer. Defaults to an empty dictionary.        | `{}`    |
-| `model_config`     | `dict` | Configuration parameters specifically for the model. Defaults to an empty dictionary.            | `{}`    |
-| `adapter_path`     | `str`  | Path to the LoRA adapters. If provided, applies LoRA layers to the model.                        | `None`  |
-| `lazy`             | `bool` | If False, evaluate the model parameters to make sure they are loaded in memory before returning. | `False` |
-
-
 ## Generate text
 
 You may generate text using the parameters described in the [text generation documentation](../text.md).
@@ -51,12 +29,11 @@ You may generate text using the parameters described in the [text generation doc
 With the loaded model, you can generate text or perform structured generation, e.g.
 
 ```python
-from outlines import models, generate
+from mlx_lm import load
+import outlines
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-generator = generate.text(model)
-
-answer = generator("A prompt", temperature=2.0)
+model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
+answer = model("A prompt", temperature=2.0)
 ```
 
 ## Streaming
@@ -64,10 +41,11 @@ answer = generator("A prompt", temperature=2.0)
 You may creating a streaming iterable with minimal changes
 
 ```python
-from outlines import models, generate
+from mlx_lm import load
+import outlines
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-generator = generate.text(model)
+model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
+generator = outlines.Generator(model)
 
 for token_str in generator.text("A prompt", temperature=2.0):
     print(token_str)
@@ -80,14 +58,13 @@ You may perform structured generation with mlxlm to guarantee your output will m
 Example: Phone number generation with pattern `"\\+?[1-9][0-9]{7,14}"`:
 
 ```python
-from outlines import models, generate
+from mlx_lm import load
+import outlines
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
 
 phone_number_pattern = "\\+?[1-9][0-9]{7,14}"
-generator = generate.regex(model, phone_number_pattern)
-
-model_output = generator("What's Jennys Number?\n")
+model_output = model("What's Jennys Number?\n", outlines.Regex(phone_number_pattern))
 print(model_output)
 # '8675309'
 ```
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
index eab70f2a1..5d85c67f8 100644
--- a/docs/reference/models/models.md
+++ b/docs/reference/models/models.md
@@ -4,53 +4,49 @@ title: Models
 
 # Models
 
-Outlines supports generation using a number of inference engines (`outlines.models`). Loading a model using outlines follows a similar interface between inference engines:
-
-```python
-import outlines
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-model = outlines.models.transformers_vision("llava-hf/llava-v1.6-mistral-7b-hf")
-model = outlines.models.vllm("microsoft/Phi-3-mini-128k-instruct")
-model = outlines.models.llamacpp(
-    "microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf"
-)
-model = outlines.models.exllamav2("bartowski/Phi-3-mini-128k-instruct-exl2")
-model = outlines.models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
-
-model = outlines.models.openai(
-    "gpt-4o-mini",
-    api_key=os.environ["OPENAI_API_KEY"]
-)
-```
-
-
-# Feature Matrix
-|                   | [Transformers](transformers.md) | [Transformers Vision](transformers_vision.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md)* |
-|-------------------|--------------|---------------------|------|-----------|-----------|-------|---------|
-| **Device**        |              |                     |      |           |           |       |         |
-| Cuda              | ✅           | ✅                  | ✅   | ✅        | ✅        | ❌    | N/A     |
-| Apple Silicon     | ✅           | ✅                  | ❌   | ✅        | ✅        | ✅    | N/A     |
-| x86 / AMD64       | ✅           | ✅                  | ❌   | ✅        | ✅        | ❌    | N/A     |
-| **Sampling**      |              |                     |      |           |           |       |         |
-| Greedy            | ✅           | ✅                  | ✅   | ✅*       | ✅        | ✅    | ❌      |
-| Multinomial       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Multiple Samples  | ✅           | ✅                  |      | ❌        |           | ❌    | ✅      |
-| Beam Search       | ✅           | ✅                  | ✅   | ❌        | ✅        | ❌    | ❌      |
-| **Generation**    |              |                     |      |           |           |       |         |
-| Batch             | ✅           | ✅                  | ✅   | ❌        | ?         | ❌    | ❌      |
-| Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
-| Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| **Structured**    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| JSON Schema       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Choice            | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Regex             | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
-| Grammar           | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
-
-
-## Caveats
-
-- OpenAI doesn't support structured generation due to limitations in their API and server implementation.
-- `outlines.generate` ["Structured"](../generation/generation.md) includes methods such as `outlines.generate.regex`, `outlines.generate.json`, `outlines.generate.cfg`, etc.
-- MLXLM only supports Apple Silicon.
-- llama.cpp greedy sampling available via multinomial with `temperature = 0.0`.
+This section provides detailed information about using structured generation with a number of inference engines.
+
+## Supported providers
+
+### Open Source models
+
+Open source models offer more flexibility for structured generation as we have control over the sampling loop:
+
+- [transformers](transformers.md) - Run open-source models locally.
+- [llama-cpp-python](llamacpp.md) - Python bindings for llama.cpp.
+- [mlx-lm](mlxlm.md) - Run open-source models on Metal hardware.
+- [vllm](vlm.md) - Run open-source models on the vLLM engine.
+- [exllamaV2](exllamav2.md)
+
+Ollama only supports Json Schema:
+
+- [ollama](ollama.md) - Python client library for Ollama.
+
+### Cloud AI providers
+
+OpenAI has recently integrated [structured outputs][structured-outputs] in its API, and only JSON Schema-based structured generation is available. Google's Gemini API supports both Json Schema and multiple choices:
+
+- [OpenAI](openai.md) - GPT-4o, o1, o3-mini and other OpenAI models.
+- [Azure OpenAI](openai.md) - Microsoft's Azure-hosted OpenAI models.
+- [Gemini](gemini.md) - Run Google's Gemini model.
+
+
+## Structure generation coverage
+
+Integrations differ in their coverage of structured generation. Here is a summary:
+
+
+|                   | [Transformers](transformers.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md) | [Gemini](gemini.md)
+|-------------------|--------------|------|-----------|-----------|-------|---------|-------|
+| **Supported HW**  |              |      |           |           |       |         |       |
+| CUDA              | ✅           | ✅  | ✅        | ✅        | ❌    | N/A     | N/A   |
+| Apple Silicon     | ✅           | ❌  | ✅        | ✅        | ✅    | N/A     | N/A   |
+| x86 / AMD64       | ✅           | ❌  | ✅        | ✅        | ❌    | N/A     | N/A   |
+| **Structure**     |               |     |           |           |       |         |       |
+| JSON Schema       | ✅           | ✅  | ✅        | ✅        | ✅    | ✅      | ✅   |
+| Choice            | ✅           | ✅  | ✅        | ✅        | ✅    | ❌      | ✅   |
+| Regex             | ✅           | ✅  | ✅        | ✅        | ✅    | ❌      | ❌   |
+| Grammar           | ✅           | ✅  | ✅        | ✅        | ✅    | ❌      | ❌   |
+
+
+[structured-outputs]: https://platform.openai.com/docs/guides/structured-outputs
diff --git a/docs/reference/models/ollama.md b/docs/reference/models/ollama.md
index 0d2961832..4cc3bcd2d 100644
--- a/docs/reference/models/ollama.md
+++ b/docs/reference/models/ollama.md
@@ -7,6 +7,8 @@
     - To download Ollama: https://ollama.com/download
     - To install the ollama python sdk: `pip install ollama`
 
+    Consult the [`ollama` documentation](https://github.com/ollama/ollama-python) for detailed informations about how to initialize models and the available options.
+
 ## Ollama models
 
 You must provide a model name when instantiating the `outlines.models.Ollama` class. This model must be available on your system.
diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
index eae8dde39..354994861 100644
--- a/docs/reference/models/openai.md
+++ b/docs/reference/models/openai.md
@@ -1,56 +1,34 @@
 # OpenAI and compatible APIs
 
-!!! Installation
+!!! Note "Documentation"
 
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively you can run:
+    To be able to use OpenAI in Outlines, you must install the OpenAI Python SDK with `pip install openai`
 
-    ```bash
-    pip install "outlines[openai]"
-    ```
+    Consult the [OpenAI SDK documentation](https://github.com/openai/openai-python) for detailed informations about how to initialize OpenAI clients and the available options.
 
 ## OpenAI models
 
 Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.OpenAI`:
 
 ```python
-from outlines import models
+from openai import OpenAI
+import outlines
 
+# OpenAI models
+client = OpenAI()
+model = outlines.from_openai(client, "gpt-4o-mini")
+model = outlines.from_openai(client, "gpt-4o")
 
-model = models.OpenAI("gpt-4o-mini")
-model = models.OpenAI("gpt-4o")
-```
-
-Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.OpenAI` as keyword arguments:
-
-```python
-import os
-from outlines import models
-
-
-model = models.OpenAI(
-    "gpt-4o-mini",
-    api_key=os.environ["OPENAI_API_KEY"]
-)
-```
-
-Refer to the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list of the initialization parameters.
-
-## Azure OpenAI models
-
-Outlines also supports Azure OpenAI models:
-
-```python
-from outlines import models
-
-
-model = models.AzureOpenAI(
-    "azure-deployment-name",
+# OpenAI models deployed on Azure
+client = AzureOpenAI(
     api_version="2024-07-18",
     azure_endpoint="https://example-endpoint.openai.azure.com",
 )
+model = outlines.from_openai(client, "azure-deployment-name")
 ```
 
-You can pass any parameter you would pass to `openai.AzureOpenAI`. You can consult the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/lib/azure.py) for an up-to-date list.
+Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models, and the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list of the initialization parameters.
+
 
 ## Advanced configuration
 
@@ -59,17 +37,17 @@ For more advanced configuration option, such as support proxy, please consult th
 
 ```python
 from openai import AsyncOpenAI, DefaultHttpxClient
-from outlines import models
-from outlines.models.openai import OpenAIConfig
-
+import outlines
 
 client = AsyncOpenAI(
+    api_key="my key",
     base_url="http://my.test.server.example.com:8083",
     http_client=DefaultHttpxClient(
         proxies="http://my.test.proxy.example.com",
         transport=httpx.HTTPTransport(local_address="0.0.0.0"),
     ),
 )
+model = outlines.from_openai(client, "model_name")
 ```
 
 ## Models that follow the OpenAI standard
@@ -79,27 +57,25 @@ Outlines supports models that follow the OpenAI standard. You will need to initi
 ```python
 import os
 from openai import AsyncOpenAI
-from outlines import models
-from outlines.models.openai import OpenAIConfig
-
+import outlines
 
-model = models.OpenAI(
-    "model_name",
+client = AsyncOpenAI(
     api_key=os.environ.get("PROVIDER_KEY"),
-    base_url="http://other.provider.server.com"
+    base_url="http://other.provider.server.com",
 )
+model = outlines.from_openai(client, "model_name")
 ```
 
-## Text generation
+## Calling the model
 
-To generate text using an OpenAI model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
+You can call the model directly. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
 
 ```python
-from outlines import models, Generator
+from openai import OpenAI
+import outlines
 
-model = models.OpenAI("gpt-4o-mini")
-generator = Generator(model)
-result = generator("Prompt", seed=10)
+model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
+result = model("Prompt", seed=10)
 ```
 
 See the [OpenAI SDK documentation](https://github.com/openai/openai-python/blob/6974a981aec1814b5abba429a8ea21be9ac58538/src/openai/types/completion_create_params.py#L13) for the list of available arguments.
@@ -109,11 +85,13 @@ See the [OpenAI SDK documentation](https://github.com/openai/openai-python/blob/
 Outlines provides support for [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode). Currently only JSON-Schema is supported:
 
 ```python
+from openai import OpenAI
 from pydantic import BaseModel
-from outlines import models, Generator
-from outlines.types import Json
 
-model = models.OpenAI("gpt-4o-mini")
+import outlines
+from outlines.types import JsonType
+
+model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
 
 class Person(BaseModel):
     first_name: str
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index f24591436..724a724c3 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -1,92 +1,34 @@
 # Transformers
 
+!!! Note "Documentation"
 
-!!! Installation
+    To be able to use Transformers models in Outlines, you must install the `transformers` library, `pip install transformers`
 
-    You need to install the `transformer` library to be able to use these models in Outlines, or alternatively:
-
-    ```bash
-    pip install "outlines[transformers]"
-    ```
+    Consult the [`transformers` documentation](https://huggingface.co/docs/transformers/en/index) for detailed informations about how to initialize models and the available options.
 
 ## Create a `Transformers` model
 
-The only mandatory argument to instantiate a `Transformers` model is the name of the model to use.
-```python
-from outlines import models
-
-model = models.Transformers("microsoft/Phi-3-mini-4k-instruct")
-```
-
-The model name must be a valid `transformers` model name. You can find a list of all in the HuggingFace library [here](https://huggingface.co/models).
-
-When instantiating a `Transformers` model as such, the class creates a model from the transformers libray using the class `AutoModelForCausalLM` by default (`transformers.AutoModelForCausalLM.from_pretrained(model_name, **model_kwargs)`).
-
-You can also provide keyword arguments in an optional `model_kwargs` parameter. Those will be passed to the `from_pretrained` method of the model class. One such argument is `device_map`, which allows you to specify the device on which the model will be loaded.
-
-For instance:
-```python
-from outlines import models
-
-model = models.Transformers("microsoft/Phi-3-mini-4k-instruct", model_kwargs={"device_map": "cuda"})
-```
-
-## Alternative model classes
-
-If the model you want to use is not compatible with `AutoModelForCausalLM`, you must provide a value for the `model_class` parameter. This value must be a valid `transformers` model class.
-
-For instance:
-```python
-from outlines import models
-from transformers import AutoModelForSeq2SeqLM
-
-model = models.Transformers("facebook/bart-large", model_class=AutoModelForSeq2SeqLM)
-```
-
-When you instantiate a `Transformers` model, the class also creates a `Tokenizer` instance from the `AutoTokenizer` class. You can provide keyword arguments in an optional `tokenizer_kwargs` parameter. Those will be passed to the `from_pretrained` method of the tokenizer class as such: `tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)`.
-
-Similarly, if your model is not compatible with `AutoTokenizer`, you must provide a value for the `tokenizer_class` parameter.
+You can use `outlines.from_transformers` to load a `transformers` model and tokenizer:
 
 ```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 from outlines import models
-from transformers import T5ForConditionalGeneration, T5Tokenizer
 
-model_pile_t5 = models.Transformers(
-    model_name="EleutherAI/pile-t5-large",
-    model_class=T5ForConditionalGeneration,
-    tokenizer_class=T5Tokenizer
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = models.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
 )
 ```
 
-### Mamba
-
-[Mamba](https://github.com/state-spaces/mamba) is a transformers alternative which employs memory efficient, linear-time decoding.
-
-To use Mamba with outlines you must first install the necessary requirements:
-```
-pip install causal-conv1d>=1.2.0 mamba-ssm torch transformers
-```
-
-Then you can create an `Mamba` Outlines model via:
-```python
-from outlines import models
-
-model = models.Mamba("state-spaces/mamba-2.8b-hf", model_kwargs={"device_map": "cuda"}, tokenizer_kwargs={"padding_side": "left"})
-```
-
-Alternatively, you can use the `Transformers` class to create an `Mamba` model by providing the appropriate `model_class` and `tokenizer_class` arguments.
-
-Read [`transformers`'s documentation](https://huggingface.co/docs/transformers/en/model_doc/mamba) for more information.
-
-### Encoder-Decoder Models
-
-You can use encoder-decoder (seq2seq) models like T5 and BART with Outlines.
+The model name must be a valid `transformers` model name. You can find a list of all in the HuggingFace library [here](https://huggingface.co/models). We currently support `CausalLM`, `Seq2Seq`, `Mamba` and vision models.
 
 Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
 
 ## Use the model to generate text
 
 Once you have created a `Transformers` model, you can use it to generate text by calling the instance of the model.
+
 ```python
 model("Hello, how are you?")
 ```
diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index 8789b588e..1636410a4 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -3,11 +3,9 @@
 
 !!! Note "Installation"
 
-    You need to install the `vllm` library to use the vLLM integration. See the [installation section](#installation) for instructions to install vLLM for CPU or ROCm. To get started you can also run:
+    You need to install the `vllm` library to use the vLLM integration: `pip install vllm`. The default installation only works on machines with a GPU, follow the [installation section][vllm-install-cpu] for instructions to install vLLM for CPU or ROCm.
 
-    ```bash
-    pip install "outlines[vllm]"
-    ```
+    Consult the [vLLM documentation][vllm-docs] for detailed informations about how to initialize OpenAI clients and the available options.
 
 ## Load the model
 
@@ -15,106 +13,19 @@ Outlines supports models available via vLLM's offline batched inference interfac
 
 
 ```python
-from outlines import models
-
-model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
+import outlines
+from vllm import LLM
+model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
 ```
 
-Or alternatively:
-
-```python
-import vllm
-from outlines import models
-
-llm = vllm.LLM("microsoft/Phi-3-mini-4k-instruct")
-model = models.VLLM(llm)
-```
-
-
 Models are loaded from the [HuggingFace hub](https://huggingface.co/).
 
 
 !!! Warning "Device"
 
-    The default installation of vLLM only allows to load models on GPU. See the [installation instructions](#installation) to run models on CPU.
-
-
-You can pass any parameter that you would normally pass to `vllm.LLM`, as keyword arguments:
-
-```python
-from outlines import models
-
-model = models.vllm(
-    "microsoft/Phi-3-mini-4k-instruct",
-    trust_remote_code=True,
-    gpu_memory_utilization=0.7
-)
-```
-
-**Main parameters:**
-
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `tokenizer_mode`| `str`  | "auto" will use the fast tokenizer if available and "slow" will always use the slow tokenizer. | `auto`
-| `trust_remote_code`| `bool` | Trust remote code when downloading the model and tokenizer. | `False` |
-| `tensor_parallel_size`| `int` | The number of GPUs to use for distributed execution with tensor parallelism.| `1` |
-| `dtype`| `str` | The data type for the model weights and activations. Currently, we support `float32`, `float16`, and `bfloat16`. If `auto`, we use the `torch_dtype` attribute specified in the model config file. However, if the `torch_dtype` in the config is `float32`, we will use `float16` instead.| `auto` |
-| `quantization`| `Optional[str]` | The method used to quantize the model weights. Currently, we support "awq", "gptq" and "squeezellm". If None, we first check the `quantization_config` attribute in the model config file. If that is None, we assume the model weights are not quantized and use `dtype` to determine the data type of the weights.| `None` |
-| `revision`| `Optional[str]` | The specific model version to use. It can be a branch name, a tag name, or a commit id.| `None` |
-| `tokenizer_revision`| `Optional[str]`| The specific tokenizer version to use. It can be a branch name, a tag name, or a commit id.| `None` |
-| `gpu_memory_utilization`| `float` | The ratio (between 0 and 1) of GPU memory to reserve for the model weights, activations, and KV cache. Higher values will increase the KV cache size and thus improve the model's throughput. However, if the value is too high, it may cause out-of-memory (OOM) errors.| `0.9` |
-| `swap_space`| `int` | The size (GiB) of CPU memory per GPU to use as swap space. This can be used for temporarily storing the states of the requests when their `best_of` sampling parameters are larger than 1. If all requests will have `best_of=1`, you can safely set this to 0. Otherwise, too small values may cause out-of-memory (OOM) errors.| 4 |
-| `enforce_eager`| `bool` | Whether to enforce eager execution. If True, we will disable CUDA graph and always execute the model in eager mode. If False, we will use CUDA graph and eager execution in hybrid.| `False` |
-| `enable_lora` | `bool` | Whether to enable loading LoRA adapters | `False` |
-
-See the [vLLM code](https://github.com/vllm-project/vllm/blob/8f44facdddcf3c704f7d6a2719b6e85efc393449/vllm/entrypoints/llm.py#L72) for a list of all the available parameters.
-
-### Use quantized models
-
-vLLM supports AWQ, GPTQ and SqueezeLLM quantized models:
+    The default installation of vLLM only allows to load models on GPU. See the [installation instructions][vllm-install-cpu] to run models on CPU.
 
 
-```python
-from outlines import models
-
-model = models.vllm("TheBloke/Llama-2-7B-Chat-AWQ", quantization="awq")
-model = models.vllm("TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", quantization="gptq")
-model = models.vllm("https://huggingface.co/squeeze-ai-lab/sq-llama-30b-w4-s5", quantization="squeezellm")
-```
-
-!!! Warning "Dependencies"
-
-    To use AWQ model you need to install the autoawq library `pip install autoawq`.
-
-    To use GPTQ models you need to install the autoGTPQ and optimum libraries `pip install auto-gptq optimum`.
-
-
-### Multi-GPU usage
-
-To run multi-GPU inference with vLLM you need to set the `tensor_parallel_size` argument to the number of GPUs available when initializing the model. For instance to run inference on 2 GPUs:
-
-
-```python
-from outlines import models
-
-model = models.vllm(
-    "microsoft/Phi-3-mini-4k-instruct"
-    tensor_parallel_size=2
-)
-```
-
-### Load LoRA adapters
-
-You can load LoRA adapters and alternate between them dynamically:
-
-```python
-from outlines import models
-
-model = models.vllm("facebook/opt-350m", enable_lora=True)
-model.load_lora("ybelkaa/opt-350m-lora")  # Load LoRA adapter
-model.load_lora(None)  # Unload LoRA adapter
-```
-
 ## Generate text
 
 In addition to the parameters described in the [text generation section](../text.md) you can pass an instance of `SamplingParams` directly to any generator via the `sampling_params` keyword argument:
@@ -124,112 +35,16 @@ from vllm.sampling_params import SamplingParams
 from outlines import models, generate
 
 
-model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
+model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
 params = SamplingParams(n=2, frequency_penalty=1., min_tokens=2)
-answer = generator("A prompt", sampling_params=params)
+answer = model("A prompt", sampling_params=params)
 ```
 
-This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
-
-!!! Note
-
-    The values passed via the `SamplingParams` instance supersede the other arguments to the generator or the samplers.
-
-**`SamplingParams` attributes:**
-
-| Parameters | Type             | Description            | Default |
-|:-----------|------------------|:-----------------------|---------|
-| `n` | `int` | Number of output sequences to return for the given prompt. | `1` |
-| `best_of` | `Optional[int]` | Number of output sequences that are generated from the prompt. From these `best_of` sequences, the top `n` sequences are returned. `best_of` must be greater than or equal to `n`. This is treated as the beam width when `use_beam_search` is True. By default, `best_of` is set to `n`. | `None` |
-| `presence_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens.| `0.0` |
-| `frequency_penalty` | `float` | Float that penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens. | `0.0`
-| `repetition_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the prompt and the generated text so far. Values > 1 encourage the model to use new tokens, while values < 1 encourage the model to repeat tokens. | `1.0` |
-| `temperature` | `float` | Float that controls the randomness of the sampling. Lower values make the model more deterministic, while higher values make the model more random. Zero means greedy sampling. | `1.0` |
-| `top_p` | `float` |  Float that controls the cumulative probability of the top tokens to consider. Must be in (0, 1]. Set to 1 to consider all tokens. | `1.0` |
-| `top_k` | `int` |  Integer that controls the number of top tokens to consider. Set to -1 to consider all tokens. | `-1` |
-| `min_p` |`float` |  Float that represents the minimum probability for a token to be considered, relative to the probability of the most likely token. Must be in [0, 1]. Set to 0 to disable this. | `0.0` |
-| `seed` | `Optional[int]` | Random seed to use for the generation. | `None` |
-| `use_beam_search` | `bool` |  Whether to use beam search instead of sampling. | `False` |
-| `length_penalty` | `float` | Float that penalizes sequences based on their length. Used in beam search.  | `1.0` |
-| `early_stopping` | `Union[bool, str]` |  Controls the stopping condition for beam search. It accepts the following values: `True`, where the generation stops as soon as there are `best_of` complete candidates; `False`, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; `"never"`, where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm). | `False` |
-| `stop` | `Optional[Union[str, List[str]]]` |  List of strings that stop the generation when they are generated. The returned output will not contain the stop strings. | `None` |
-| `stop_token_ids` | `Optional[List[int]]` |  List of tokens that stop the generation when they are generated. The returned output will contain the stop tokens unless the stop tokens are special tokens. | `None` |
-| `include_stop_str_in_output` | `bool` |  Whether to include the stop strings in output text. Defaults to False. | `False` |
-| `ignore_eos` | `bool` |  Whether to ignore the EOS token and continue generating tokens after the EOS token is generated. | `False` |
-| `max_tokens` | `int` |  Maximum number of tokens to generate per output sequence. | `16` |
-| `min_tokens` | `int` | Minimum number of tokens to generate per output sequence before EOS or stop_token_ids can be generated | `0` |
-| `skip_special_tokens` | `bool` | Whether to skip special tokens in the output. | `True` |
-| `spaces_between_special_tokens` | `bool` |  Whether to add spaces between special tokens in the output.  Defaults to True. | `True` |
-
-### Streaming
-
 !!! Warning
 
     Streaming is not available for the offline vLLM integration.
 
-
-## Installation
-
-By default the vLLM library is installed with pre-commpiled C++ and CUDA binaries and will only run on GPU:
-
-```python
-pip install vllm
-```
-
-### CPU
-
-You need to have the `gcc` compiler installed on your system. Then you will need to install vLLM from source. First clone the repository:
-
-```bash
-git clone https://github.com/vllm-project/vllm.git
-cd vllm
-```
-
-Install the Python packages needed for the installation:
-
-```bash
-pip install --upgrade pip
-pip install wheel packaging ninja setuptools>=49.4.0 numpy
-pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/cpu
-```
-
-and finally run:
-
-```bash
-VLLM_TARGET_DEVICE=cpu python setup.py install
-```
-
-See the [vLLM documentation][vllm-install-cpu] for more details, alternative installation methods (Docker) and performance tips.
-
-### ROCm
-
-
-You will need to install vLLM from source. First install Pytorch on ROCm:
-
-```bash
-pip install torch==2.2.0.dev20231206+rocm5.7 --index-url https://download.pytorch.org/whl/nightly/rocm5.7 # tested version
-```
-
-You will then need to install flash attention for ROCm following [these instructions][rocm-flash-attention]. You can then install `xformers=0.0.23` and apply the patches needed to adapt Flash Attention for ROCm:
-
-```bash
-pip install xformers==0.0.23 --no-deps
-bash patch_xformers.rocm.sh
-```
-
-And finally build vLLM:
-
-```bash
-cd vllm
-pip install -U -r requirements-rocm.txt
-python setup.py install # This may take 5-10 minutes.
-```
-
-See the [vLLM documentation][vllm-install-rocm] for alternative installation methods (Docker).
-
-
+[vllm-docs]:https://docs.vllm.ai/en/latest/
 [vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
 [vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
 [rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 51b21ecf2..2f33aa0f6 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -8,6 +8,7 @@
 from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function
+from outlines.generate import Generator
 from outlines.templates import Template, prompt
 
 from outlines.models import (

From d6d950fefd5edd6f4a74349c21e4dbd6157b02b2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 22 Feb 2025 15:09:02 +0100
Subject: [PATCH 236/505] Update the README

---
 README.md            | 110 ++++++++++++++++++++++++++++++-------------
 outlines/__init__.py |   1 +
 2 files changed, 78 insertions(+), 33 deletions(-)

diff --git a/README.md b/README.md
index 3b726216f..e8f4fdbce 100644
--- a/README.md
+++ b/README.md
@@ -84,10 +84,15 @@ Please see [the documentation](https://dottxt-ai.github.io/outlines/latest/refer
 You can reduce the completion to a choice between multiple possibilities:
 
 ``` python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
 model_name = "HuggingFaceTB/SmolLM2-360M-Instruct"
-model = outlines.models.transformers(model_name)
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
 
 # You must apply the chat template tokens to the prompt!
 # See below for an example.
@@ -104,9 +109,7 @@ Text: I really really really want pizza.
 <|im_start|>assistant
 """
 
-generator = outlines.generate.choice(model, ["Pizza", "Pasta", "Salad", "Dessert"])
-answer = generator(prompt)
-
+answer = model(prompt, outlines.Choice(["Pizza", "Pasta", "Salad", "Dessert"]))
 # Likely answer: Pizza
 ```
 
@@ -121,8 +124,7 @@ class Food(str, Enum):
     salad = "Salad"
     dessert = "Dessert"
 
-generator = outlines.generate.choice(model, Food)
-answer = generator(prompt)
+answer = model(prompt, outlines.Choice(Food))
 # Likely answer: Pizza
 ````
 
@@ -132,9 +134,14 @@ You can instruct the model to only return integers or floats:
 
 
 ``` python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 import outlines
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
+model_name = "WizardLM/WizardMath-7B-V1.1"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
 
 prompt = "<s>result of 9 + 9 = 18</s><s>result of 1 + 2 = "
 answer = outlines.generate.format(model, int)(prompt)
@@ -142,8 +149,7 @@ print(answer)
 # 3
 
 prompt = "sqrt(2)="
-generator = outlines.generate.format(model, float)
-answer = generator(prompt, max_tokens=10)
+answer = model(prompt, outlines.types.number, max_tokens=10)
 print(answer)
 # 1.41421356
 ```
@@ -155,9 +161,14 @@ Outlines also comes with fast regex-structured generation. In fact, the `choice`
 hood:
 
 ``` python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 import outlines
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
 
 prompt = """
 <|im_start|>system You are a helpful assistant.
@@ -171,15 +182,13 @@ The IP address of a Google DNS server is
 
 """
 
-generator = outlines.generate.text(model)
-unstructured = generator(prompt, max_tokens=30)
+unstructured = model(prompt, max_tokens=10)
 
-generator = outlines.generate.regex(
-    model,
-    r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)",
-    sampler=outlines.samplers.greedy(),
+structured = model(
+    prompt,
+    outlines.Regex(r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"),
+    max_tokens=30
 )
-structured = generator(prompt, max_tokens=30)
 
 print(unstructured)
 # 8.8.8.8
@@ -200,6 +209,7 @@ Outlines users can guide the generation process so the output is *guaranteed* to
 ```python
 from enum import Enum
 from pydantic import BaseModel, constr
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 import outlines
 
@@ -226,20 +236,25 @@ class Character(BaseModel):
     strength: int
 
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
 
 # Construct structured sequence generator
-generator = outlines.generate.json(model, Character)
 
 # Draw a sample
 seed = 789001
 
-character = generator("Give me a character description", seed=seed)
+prompt = "Give me a character description"
+character = model(prompt, outlines.JsonType(Character), seed=seed)
 
 print(repr(character))
 # Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
 
-character = generator("Give me an interesting character description")
+prompt = "Give me an interesting character description"
+character = model(prompt, outlines.JsonType(Character), seed=seed)
 
 print(repr(character))
 # Character(name='Vivian Thr', age=44, armor=<Armor.plate: 'plate'>, weapon=<Weapon.crossbow: 'crossbow'>, strength=125)
@@ -252,6 +267,8 @@ The method works with union types, optional types, arrays, nested schemas, etc.
 Sometimes you just want to be able to pass a JSON Schema instead of a Pydantic model. We've got you covered:
 
 ``` python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
 schema = '''{
@@ -291,9 +308,14 @@ schema = '''{
     }
 }'''
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = outlines.generate.json(model, schema)
-character = generator("Give me a character description")
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
+
+prompt = "Give me a character description"
+character = model(prompt, outlines.JsonType(schema))
 ```
 
 ### Using context-free grammars to guide generation
@@ -301,6 +323,8 @@ character = generator("Give me a character description")
 Formal grammars rule the world, and Outlines makes them rule LLMs too. You can pass any context-free grammar in the EBNF format and Outlines will generate an output that is valid to this grammar:
 
 ``` python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
 arithmetic_grammar = """
@@ -317,9 +341,14 @@ arithmetic_grammar = """
     %import common.NUMBER
 """
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = outlines.generate.cfg(model, arithmetic_grammar)
-sequence = generator("Alice had 4 apples and Bob ate 2. Write an expression for Alice's apples:")
+model_name = "WizardLM/WizardMath-7B-V1.1"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
+
+prompt = "Alice had 4 apples and Bob ate 2. Write an expression for Alice's apples:"
+sequence = model(prompt, outlines.types.Cfg(arithmetic_grammar))
 
 print(sequence)
 # (8-2)
@@ -332,17 +361,24 @@ This was a very simple grammar, and you can use `outlines.generate.cfg` to gener
 Outlines can infer the structure of the output from the signature of a function. The result is a dictionary, and can be passed directly to the function using the usual dictionary expansion syntax `**`:
 
 ```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
 
 def add(a: int, b: int):
     return a + b
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = outlines.generate.json(model, add)
+
+model_name = "WizardLM/WizardMath-7B-V1.1"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
+generator = outlines.generate.json(model, outlines.types.JsonType(add))
 result = generator("Return json with two integers named a and b respectively. a is odd and b even.")
 
-print(add(**result))
+# print(add(**result))
 # 3
 ```
 
@@ -354,6 +390,8 @@ You can also embed various functions into an enum to generate params:
 from enum import Enum
 from functools import partial
 
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
 
@@ -367,11 +405,15 @@ class Operation(Enum):
     add = partial(add)
     mul = partial(mul)
 
-model = outlines.models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = outlines.generate.json(model, Operation)
+model_name = "WizardLM/WizardMath-7B-V1.1"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
+generator = outlines.generate.json(model, outlines.types.JsonType(Operation))
 result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
 
-print(result)
+#print(result)
 # {'c': -3.14, 'd': 1.5}
 ```
 
@@ -397,6 +439,8 @@ You are a sentiment-labelling assistant.
 You can then load it and call it with:
 
 ``` python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
 examples = [
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 2f33aa0f6..af62bde19 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -5,6 +5,7 @@
 import outlines.models
 import outlines.processors
 import outlines.types
+from outlines.types import Choice, Regex, JsonType
 from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function

From 760b1678d3021107ef7511ef152d27cc93dfd1ee Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 22 Feb 2025 15:31:53 +0100
Subject: [PATCH 237/505] Update the examples

---
 examples/babyagi.py            |  7 ++++---
 examples/cerebrium/main.py     | 14 +++++++++-----
 examples/cfg.py                | 17 ++++++++++++-----
 examples/dating_profile.py     | 24 +++++++++++++-----------
 examples/llamacpp_example.py   |  5 +++--
 examples/math_generate_code.py |  8 +++++---
 examples/meta_prompting.py     | 34 +++++++++++++++++-----------------
 examples/modal_example.py      | 21 ++++++++++++++-------
 examples/pick_odd_one_out.py   |  5 +++--
 examples/react.py              | 14 +++++++-------
 examples/self_consistency.py   |  6 +++---
 outlines/__init__.py           |  5 +++++
 outlines/templates.py          |  4 ----
 13 files changed, 95 insertions(+), 69 deletions(-)

diff --git a/examples/babyagi.py b/examples/babyagi.py
index a8f1de8ba..3644ecf51 100644
--- a/examples/babyagi.py
+++ b/examples/babyagi.py
@@ -8,13 +8,14 @@
 from collections import deque
 from typing import Deque, List
 
+from openai import OpenAI
+
 import outlines
-import outlines.models as models
 from outlines import Template
 
 
-model = models.openai("gpt-4o-mini")
-complete = outlines.generate.text(model)
+model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
+complete = outlines.Generator(model)
 
 ## Load the prompts
 perform_task_ppt = Template.from_file("prompts/babyagi_perform_task.txt")
diff --git a/examples/cerebrium/main.py b/examples/cerebrium/main.py
index b61cdb6e5..5ea1b61f3 100644
--- a/examples/cerebrium/main.py
+++ b/examples/cerebrium/main.py
@@ -1,6 +1,11 @@
 import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
+    AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
+)
 
-model = outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
 
 schema = {
     "title": "Character",
@@ -29,14 +34,13 @@
     },
 }
 
-generator = outlines.generate.json(model, schema)
-
 
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
-    character = generator(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
+    character = model(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
+        outlines.JsonType(schema),
     )
 
     print(character)
diff --git a/examples/cfg.py b/examples/cfg.py
index 99edf35f1..22fdf7ed0 100644
--- a/examples/cfg.py
+++ b/examples/cfg.py
@@ -1,5 +1,7 @@
-import outlines.generate as generate
-import outlines.models as models
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+import outlines
+from outlines.types import Cfg
 
 nlamb_grammar = r"""
     start: sentence
@@ -75,11 +77,16 @@
     %ignore WS
 """
 
-model = models.transformers("hf-internal-testing/tiny-random-gpt2")
+model_name = "hf-internal-testing/tiny-random-gpt2"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name),
+)
+
 batch_size = 10
 for grammar in [nlamb_grammar, calc_grammar, dyck_grammar, json_grammar]:
-    generator = generate.cfg(model, grammar, max_tokens=model.model.config.n_positions)
-    sequences = generator([" "] * batch_size)
+    generator = outlines.Generator(model, Cfg(grammar))
+    sequences = generator([" "] * batch_size, max_tokens=model.model.config.n_positions)
     for seq in sequences:
         try:
             parse = generator.fsm.parser.parse(seq)
diff --git a/examples/dating_profile.py b/examples/dating_profile.py
index d26984223..2f1a0e1b0 100644
--- a/examples/dating_profile.py
+++ b/examples/dating_profile.py
@@ -6,7 +6,7 @@
 from pydantic import BaseModel, conlist
 
 import outlines
-from outlines import models, Template
+from outlines import Template
 
 
 class QuestionChoice(str, Enum):
@@ -86,26 +86,28 @@ class Example:
 # Below requires ~13GB of GPU memory
 # https://huggingface.co/mosaicml/mpt-7b-8k-instruct
 # Motivation: Reasonably large model that fits on a single GPU and has been fine-tuned for a larger context window
+model_name = "mosaicml/mpt-7b-8k-instruct"
 config = transformers.AutoConfig.from_pretrained(
     "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
 )
 config.init_device = "meta"
-model = models.transformers(
-    model_name="mosaicml/mpt-7b-8k-instruct",
-    device="cuda",
-    model_kwargs={
-        "config": config,
-        "trust_remote_code": True,
-        "torch_dtype": torch.bfloat16,
-        "device_map": {"": 0},
-    },
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(
+        model_name,
+        device="cuda",
+        config=config,
+        trust_remote_code=True,
+        torch_dtype=torch.bfloat16,
+        device_map={"": 0},
+    ),
+    transformers.AutoTokenizer.from_pretrained(model_name),
 )
 
 new_description = "I'm a laid-back lawyer who spends a lot of his free-time gaming. I work in a corporate office, but ended up here after the start-up I cofounded got acquired, so still play ping pong with my cool coworkers every day. I have a bar at home where I make cocktails, which is great for entertaining friends. I secretly like to wear suits and get a new one tailored every few months. I also like weddings because I get to wear those suits, and it's a good excuse for a date. I watch the latest series because I'm paying, with my hard-earned money, for every streaming service."
 
 dating_profile_prompt = Template.from_file("prompts/dating_profile.txt")
 prompt = dating_profile_prompt(description=new_description, examples=samples)
-profile = outlines.generate.json(model, DatingProfile)(prompt)  # type: ignore
+profile = model(prompt, outlines.JsonType(DatingProfile))  # type: ignore
 print(profile)
 
 # Sample generated profiles
diff --git a/examples/llamacpp_example.py b/examples/llamacpp_example.py
index 22d0da3ba..cf64a934a 100644
--- a/examples/llamacpp_example.py
+++ b/examples/llamacpp_example.py
@@ -1,6 +1,7 @@
 from enum import Enum
 
 from pydantic import BaseModel, constr
+from llama_cpp import Llama
 
 import outlines
 
@@ -30,10 +31,10 @@ class Character(BaseModel):
 
 if __name__ == "__main__":
     # curl -L -o mistral-7b-instruct-v0.2.Q5_K_M.gguf https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q5_K_M.gguf
-    model = outlines.models.llamacpp("./mistral-7b-instruct-v0.2.Q5_K_M.gguf")
+    model = outlines.from_llamacpp(Llama("./mistral-7b-instruct-v0.2.Q5_K_M.gguf"))
 
     # Construct structured sequence generator
-    generator = outlines.generate.json(model, Character)
+    generator = outlines.Generator(model, outlines.JsonType(Character))
 
     # Draw a sample
     seed = 789005
diff --git a/examples/math_generate_code.py b/examples/math_generate_code.py
index c65537ddb..9dd21e95d 100644
--- a/examples/math_generate_code.py
+++ b/examples/math_generate_code.py
@@ -1,9 +1,11 @@
 """Example from https://dust.tt/spolu/a/d12ac33169"""
 
+import openai
+
 import outlines
-import outlines.models as models
 from outlines import Template
 
+
 examples = [
     {"question": "What is 37593 * 67?", "code": "37593 * 67"},
     {
@@ -36,7 +38,7 @@ def execute_code(code):
 
 
 prompt = answer_with_code_prompt(question, examples)
-model = models.openai("gpt-4o-mini")
-answer = outlines.generate.text(model)(prompt)
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
+answer = model(prompt)
 result = execute_code(answer)
 print(f"It takes Carla {result:.0f} minutes to download the file.")
diff --git a/examples/meta_prompting.py b/examples/meta_prompting.py
index 18f1ad41a..510b4befe 100644
--- a/examples/meta_prompting.py
+++ b/examples/meta_prompting.py
@@ -12,11 +12,15 @@
 
 import argparse
 
+import openai
+
 import outlines
-import outlines.models as models
 from outlines import Template
 
 
+client = openai.OpenAI()
+
+
 def split_into_steps(question, model_name: str):
     solve = Template.from_string(
         """{{question}}
@@ -24,16 +28,15 @@ def split_into_steps(question, model_name: str):
         """
     )
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
     prompt = solve(question)
-    answer = generator(prompt, 500)
+    answer = model(prompt, 500)
     prompt += (
         answer
         + "\n what is the only option that displays the same type of relationship as : :?"
     )
-    answer = generator(prompt, 500)
+    answer = model(prompt, 500)
     completed = prompt + answer
 
     return completed
@@ -49,13 +52,12 @@ def fill_in_the_blanks(question, model_name: str):
 
     solve = Template.from_string("""{{memory}}. Let's begin.""")
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
     prompt = determine_goal(question)
-    answer = generator(prompt, stop_at=["."])
+    answer = model(prompt, stop_at=["."])
     prompt = solve(prompt + answer)
-    answer = generator(prompt, max_tokens=500)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -89,13 +91,12 @@ def ask_an_expert(question, model_name: str):
         """
     )
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
     prompt = find_expert(question)
-    expert = generator(prompt, stop_at=['"'])
+    expert = model(prompt, stop_at=['"'])
     prompt = get_answer(question, expert, prompt + expert)
-    answer = generator(prompt, max_tokens=500)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -117,13 +118,12 @@ def ask_an_expert_simple(question, model_name: str):
         """
     )
 
-    model = models.openai(model_name)
-    generator = outlines.generate.text(model)
+    model = outlines.from_openai(client, model_name)
 
     prompt = find_expert(question)
-    expert = generator(prompt, stop_at=["\n", "."])
+    expert = model(prompt, stop_at=["\n", "."])
     prompt = get_answer(expert, prompt + expert)
-    answer = generator(prompt, max_tokens=500)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
diff --git a/examples/modal_example.py b/examples/modal_example.py
index 34d4ee606..44b66e15c 100644
--- a/examples/modal_example.py
+++ b/examples/modal_example.py
@@ -4,7 +4,7 @@
 
 
 outlines_image = modal.Image.debian_slim(python_version="3.11").pip_install(
-    "outlines==0.0.37",
+    "outlines==1.0.0",
     "transformers==4.38.2",
     "datasets==2.18.0",
     "accelerate==0.27.2",
@@ -12,9 +12,11 @@
 
 
 def import_model():
-    import outlines
+    from transformers import AutoModelForCausalLM, AutoTokenizer
 
-    outlines.models.transformers("mistralai/Mistral-7B-Instruct-v0.2")
+    model_id = "mistralai/Mistral-7B-Instruct-v0.2"
+    _ = AutoTokenizer.from_pretrained(model_id)
+    _ = AutoModelForCausalLM.from_pretrained(model_id)
 
 
 outlines_image = outlines_image.run_function(import_model)
@@ -63,12 +65,17 @@ def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
     import outlines
+    from transformers import AutoModelForCausalLM, AutoTokenizer
 
-    model = outlines.models.transformers("mistralai/Mistral-7B-v0.1", device="cuda")
+    model_id = "mistralai/Mistral-7B-Instruct-v0.2"
+    model = outlines.from_transformers(
+        tokenizer=AutoTokenizer.from_pretrained(model_id),
+        model=AutoModelForCausalLM.from_pretrained(model_id, device="cuda"),
+    )
 
-    generator = outlines.generate.json(model, schema)
-    character = generator(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
+    character = model(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
+        outlines.JsonType(schema),
     )
 
     print(character)
diff --git a/examples/pick_odd_one_out.py b/examples/pick_odd_one_out.py
index f1a992411..55127c5f7 100644
--- a/examples/pick_odd_one_out.py
+++ b/examples/pick_odd_one_out.py
@@ -10,8 +10,9 @@
 
 """
 
+import openai
+
 import outlines
-import outlines.models as models
 
 
 build_ooo_prompt = outlines.Template.from_string(
@@ -31,7 +32,7 @@
 
 options = ["sea", "mountains", "plains", "sock"]
 
-model = models.openai("gpt-4o-mini")
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
 gen_text = outlines.generate.text(model)
 gen_choice = outlines.generate.choice(model, options)
 
diff --git a/examples/react.py b/examples/react.py
index 9c1954e88..472875cc1 100644
--- a/examples/react.py
+++ b/examples/react.py
@@ -13,10 +13,10 @@
 
 import requests  # type: ignore
 
+from openai import OpenAI
+
 import outlines
-from outlines import Template
-import outlines.generate as generate
-import outlines.models as models
+from outlines import Generator, Choice, Template
 
 
 build_reAct_prompt = Template.from_string(
@@ -48,11 +48,11 @@ def search_wikipedia(query: str):
 
 
 prompt = build_reAct_prompt("Where is Apple Computers headquarted? ")
-model = models.openai("gpt-4o-mini")
+model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
 
-mode_generator = generate.choice(model, choices=["Tho", "Act"])
-action_generator = generate.choice(model, choices=["Search", "Finish"])
-text_generator = generate.text(model)
+mode_generator = Generator(model, Choice(["Tho", "Act"]))
+action_generator = Generator(model, Choice(["Search", "Finish"]))
+text_generator = Generator(model)
 
 for i in range(1, 10):
     mode = mode_generator(prompt, max_tokens=128)
diff --git a/examples/self_consistency.py b/examples/self_consistency.py
index 4acba9d58..1b578a4d3 100644
--- a/examples/self_consistency.py
+++ b/examples/self_consistency.py
@@ -1,9 +1,9 @@
 import re
 
 import numpy as np
+import openai
 
 import outlines
-import outlines.models as models
 from outlines import Template
 
 examples = [
@@ -55,8 +55,8 @@
     """
 )
 
-model = models.openai("gpt-4o-mini")
-generator = outlines.generate.text(model)
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
+generator = outlines.Generator(model)
 prompt = few_shots(question, examples)
 answers = generator(prompt, samples=10)
 
diff --git a/outlines/__init__.py b/outlines/__init__.py
index af62bde19..c44abd725 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -5,6 +5,7 @@
 import outlines.models
 import outlines.processors
 import outlines.types
+from outlines.generate import Generator
 from outlines.types import Choice, Regex, JsonType
 from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
@@ -39,6 +40,10 @@
     "disable_cache",
     "get_cache",
     "Function",
+    "Generator",
+    "JsonType",
+    "Cfg",
+    "Regex",
     "prompt",
     "Template",
     "vectorize",
diff --git a/outlines/templates.py b/outlines/templates.py
index 6e771651c..325bcfe04 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -8,11 +8,7 @@
 from dataclasses import dataclass
 from io import BytesIO
 from pathlib import Path
-<<<<<<< HEAD
 from typing import Any, Callable, Dict, Optional, Type, cast
-=======
-from typing import Any, Callable, Dict, List, Optional, Type, cast
->>>>>>> 8797d19 (Add `Vision` type for prompt + image input to vision models)
 import warnings
 
 import jinja2

From 54b8f3b8ec54f7931bfb612092bc3f23e0249178 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 22 Feb 2025 16:33:48 +0100
Subject: [PATCH 238/505] Fix tests

---
 .github/workflows/tests.yml             |  1 +
 outlines/__init__.py                    |  5 +--
 outlines/function.py                    | 12 +++--
 tests/fsm/test_cfg_guide.py             | 11 ++---
 tests/generate/test_generate.py         | 60 +++++++++++++++++--------
 tests/generate/test_integration_vllm.py | 12 ++---
 tests/models/test_openai.py             | 10 ++---
 tests/test_function.py                  |  4 ++
 8 files changed, 74 insertions(+), 41 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 0565d0f3c..e8e70910c 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -33,6 +33,7 @@ jobs:
       run: |
         curl -fsSL https://ollama.com/install.sh | sh
         ollama --version
+        ollama pull tinyllama
     - name: Set up test environment
       run: |
         python -m pip install --upgrade pip
diff --git a/outlines/__init__.py b/outlines/__init__.py
index c44abd725..581f76b79 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -5,7 +5,6 @@
 import outlines.models
 import outlines.processors
 import outlines.types
-from outlines.generate import Generator
 from outlines.types import Choice, Regex, JsonType
 from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
@@ -25,7 +24,7 @@
 )
 
 
-models = [
+model_list = [
     "from_anthropic",
     "from_gemini",
     "from_llamacpp",
@@ -48,4 +47,4 @@
     "Template",
     "vectorize",
     "grammars",
-] + models
+] + model_list
diff --git a/outlines/function.py b/outlines/function.py
index 1f5e80595..d7f6dec68 100644
--- a/outlines/function.py
+++ b/outlines/function.py
@@ -5,8 +5,6 @@
 import requests
 
 import outlines
-from outlines import models
-from outlines.types import JsonType
 
 if TYPE_CHECKING:
     from outlines.generate.api import SequenceGenerator
@@ -39,8 +37,14 @@ def from_github(cls, program_path: str, function_name: str = "fn"):
 
     def init_generator(self):
         """Load the model and initialize the generator."""
-        model = models.Transformers(self.model_name)
-        self.generator = outlines.generate.Generator(model, JsonType(self.schema))
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+
+        model = outlines.from_transformers(
+            AutoModelForCausalLM.from_pretrained(self.model_name),
+            AutoTokenizer.from_pretrained(self.model_name),
+        )
+
+        self.generator = outlines.Generator(model, outlines.JsonType(self.schema))
 
     def __call__(self, *args, **kwargs):
         """Call the function.
diff --git a/tests/fsm/test_cfg_guide.py b/tests/fsm/test_cfg_guide.py
index d92afa625..b925fefea 100644
--- a/tests/fsm/test_cfg_guide.py
+++ b/tests/fsm/test_cfg_guide.py
@@ -4,7 +4,8 @@
 import pytest
 from transformers import AutoTokenizer
 
-from outlines import grammars, models
+from outlines import grammars
+from outlines.models.transformers import TransformerTokenizer
 from outlines.fsm.guide import CFGGuide
 
 
@@ -341,12 +342,12 @@ def decode(self, token_ids):
 
 @pytest.fixture(scope="session")
 def tokenizer_sentencepiece_gpt2():
-    return models.TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
+    return TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
 
 
 @pytest.fixture(scope="session")
 def tokenizer_sentencepiece_llama1():
-    return models.TransformerTokenizer(
+    return TransformerTokenizer(
         AutoTokenizer.from_pretrained(
             "trl-internal-testing/tiny-random-LlamaForCausalLM"
         )
@@ -355,14 +356,14 @@ def tokenizer_sentencepiece_llama1():
 
 @pytest.fixture(scope="session")
 def tokenizer_tiktoken_llama3():
-    return models.TransformerTokenizer(
+    return TransformerTokenizer(
         AutoTokenizer.from_pretrained("yujiepan/llama-3-tiny-random")
     )
 
 
 @pytest.fixture(scope="session")
 def tokenizer_character_level_byt5():
-    return models.TransformerTokenizer(
+    return TransformerTokenizer(
         AutoTokenizer.from_pretrained("google/byt5-small")
     )
 
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 64dee6e4b..1bf3f2a41 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -4,8 +4,8 @@
 
 import pytest
 
+import outlines
 import outlines.generate as generate
-import outlines.models as models
 import outlines.samplers as samplers
 
 ##########################################
@@ -22,11 +22,12 @@ def model_llamacpp(tmp_path_factory):
         filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
         verbose=False,
     )
-    return models.LlamaCpp(llm)
+    return outlines.from_llamacpp(llm)
 
 
 @pytest.fixture(scope="session")
 def model_exllamav2(tmp_path_factory):
+    from outlines.models.exllamav2 import exl2
     from huggingface_hub import snapshot_download
 
     tmp_dir = tmp_path_factory.mktemp("model_download")
@@ -35,7 +36,7 @@ def model_exllamav2(tmp_path_factory):
         cache_dir=tmp_dir,
     )
 
-    return models.exl2(
+    return exl2(
         model_path=model_path,
         cache_q4=True,
         paged=False,
@@ -44,56 +45,79 @@ def model_exllamav2(tmp_path_factory):
 
 @pytest.fixture(scope="session")
 def model_mlxlm(tmp_path_factory):
-    return models.mlxlm("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
+    from mlx_lm import load
+
+    return outlines.from_mlxlm(*load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit"))
 
 
 @pytest.fixture(scope="session")
 def model_mlxlm_phi3(tmp_path_factory):
-    return models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
+    from mlx_lm import load
+
+    return outlines.from_mlxlm(*load("mlx-community/Phi-3-mini-4k-instruct-4bit"))
 
 
 @pytest.fixture(scope="session")
 def model_transformers_random(tmp_path_factory):
-    return models.Transformers("hf-internal-testing/tiny-random-gpt2")
+    from transformers import AutoModelForCausalLM, AutoTokenizer
+
+    return outlines.from_transformers(
+        AutoModelForCausalLM.fromt_pretrained("hf-internal-testing/tiny-random-gpt2"),
+        AutoTokenizer.fromt_pretrained("hf-internal-testing/tiny-random-gpt2"),
+    )
 
 
 @pytest.fixture(scope="session")
 def model_transformers_opt125m(tmp_path_factory):
-    return models.Transformers("facebook/opt-125m")
+    from transformers import AutoModelForCausalLM, AutoTokenizer
+
+    return outlines.from_transformers(
+        AutoModelForCausalLM.fromt_pretrained("facebook/opt-125m"),
+        AutoTokenizer.fromt_pretrained("facebook/opt-125m"),
+    )
 
 
 @pytest.fixture(scope="session")
 def model_mamba(tmp_path_factory):
-    return models.Mamba(model_name="state-spaces/mamba-130m-hf")
+    from transformers import MambaModel, AutoTokenizer
+
+    return outlines.from_transformers(
+        MambaModel.from_pretrained(model_name="state-spaces/mamba-130m-hf"),
+        AutoTokenizer.from_pretrained(model_name="state-spaces/mamba-130m-hf"),
+    )
 
 
 @pytest.fixture(scope="session")
 def model_bart(tmp_path_factory):
-    from transformers import AutoModelForSeq2SeqLM
+    from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
 
-    return models.Transformers("facebook/bart-base", model_class=AutoModelForSeq2SeqLM)
+    return outlines.from_transformers(
+        AutoModelForSeq2SeqLM.from_pretrained("facebook/bart-base"),
+        AutoTokenizer.from_pretrained("facebook/bart-base"),
+    )
 
 
 @pytest.fixture(scope="session")
 def model_transformers_vision(tmp_path_factory):
     import torch
-    from transformers import LlavaNextForConditionalGeneration
+    from transformers import LlavaNextForConditionalGeneration, AutoTokenizer
 
-    return models.transformers_vision(
-        "llava-hf/llava-v1.6-mistral-7b-hf",
-        model_class=LlavaNextForConditionalGeneration,
-        device="cuda",
-        model_kwargs=dict(
+    return outlines.from_transformers(
+        LlavaNextForConditionalGeneration.from_pretrained(
+            "llava-hf/llava-v1.6-mistral-7b-hf",
             torch_dtype=torch.bfloat16,
             load_in_4bit=True,
-            low_cpu_mem_usage=True,
+            low_mem_cpu_usage=True,
         ),
+        AutoTokenizer.from_pretrained("llava-hf/llava-v1.6-mistral-7b-hf"),
     )
 
 
 @pytest.fixture(scope="session")
 def model_vllm(tmp_path_factory):
-    return models.vllm("facebook/opt-125m", gpu_memory_utilization=0.1)
+    from vllm import LLM
+
+    return outlines.from_vllm(LLM("facebook/opt-125m", gpu_memory_utilization=0.1))
 
 
 # TODO: exllamav2 failing in main, address in https://github.com/dottxt-ai/outlines/issues/808
diff --git a/tests/generate/test_integration_vllm.py b/tests/generate/test_integration_vllm.py
index faa8a404a..7b9acc240 100644
--- a/tests/generate/test_integration_vllm.py
+++ b/tests/generate/test_integration_vllm.py
@@ -4,12 +4,6 @@
 import pytest
 import torch
 from pydantic import BaseModel, constr
-from vllm import LLM
-
-try:
-    from vllm.sampling_params import SamplingParams
-except ImportError:
-    pass
 
 import outlines
 import outlines.generate as generate
@@ -17,6 +11,12 @@
 import outlines.models as models
 import outlines.samplers as samplers
 
+try:
+    from vllm import LLM
+    from vllm.sampling_params import SamplingParams
+except ImportError:
+    pass
+
 pytestmark = pytest.mark.skipif(
     not torch.cuda.is_available(), reason="vLLM models can only be run on GPU."
 )
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index b8c857213..57ed323de 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -30,8 +30,8 @@ def api_key():
     return api_key
 
 
-def test_init_from_client():
-    client = OpenAIClient()
+def test_init_from_client(api_key):
+    client = OpenAIClient(api_key=api_key)
     model = outlines.from_openai(client, "gpt-4o")
     assert isinstance(model, OpenAI)
     assert model.client == client
@@ -39,7 +39,7 @@ def test_init_from_client():
 
 def test_openai_wrong_inference_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = OpenAI(OpenAIClient(), MODEL_NAME)
+        model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate("prompt", foo=10)
 
 
@@ -49,7 +49,7 @@ def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = OpenAI(OpenAIClient(), MODEL_NAME)
+        model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate(Foo("prompt"))
 
 
@@ -59,7 +59,7 @@ def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = OpenAI(OpenAIClient, MODEL_NAME)
+        model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate("prompt", Foo(1))
 
 
diff --git a/tests/test_function.py b/tests/test_function.py
index cc8d7493e..42e121c60 100644
--- a/tests/test_function.py
+++ b/tests/test_function.py
@@ -1,4 +1,6 @@
 from typing import Annotated
+
+import json
 import pytest
 import responses
 from pydantic import BaseModel, Field
@@ -24,6 +26,8 @@ class Foo(BaseModel):
 
     result = fn("test")
     Foo.parse_raw(result)
+    assert isinstance(json.loads(result), dict)
+    assert "id" in json.loads(result)
 
 
 def test_download_from_github_invalid():

From 0f6a79842d69d4496e7cca33fb08a79cb8e3a943 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 22 Feb 2025 17:39:31 +0100
Subject: [PATCH 239/505] Do not run tests that require API key in CI

---
 .github/workflows/tests.yml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index e8e70910c..9390f8769 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -51,7 +51,7 @@ jobs:
         echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
-        uv run pytest -x --cov=outlines
+        uv run pytest -x --cov=outlines -m 'not api_call'
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
     - name: Upload coverage data

From f285233443d5c1bb3debfb96d3d25bb3dfe17480 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 24 Feb 2025 17:14:20 +0100
Subject: [PATCH 240/505] Initialize `Dottxt` from client instance and model
 name

---
 outlines/__init__.py        |  5 ++++-
 outlines/models/__init__.py |  2 +-
 outlines/models/dottxt.py   | 26 +++++++++++++++++++++-----
 tests/models/test_dottxt.py | 26 ++++++++++++++++++++------
 4 files changed, 46 insertions(+), 13 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 581f76b79..4aacb6c7e 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -13,6 +13,7 @@
 from outlines.templates import Template, prompt
 
 from outlines.models import (
+    from_dottxt,
     from_openai,
     from_transformers,
     from_gemini,
@@ -26,12 +27,14 @@
 
 model_list = [
     "from_anthropic",
+    "from_dottxt",
     "from_gemini",
     "from_llamacpp",
     "from_mlxlm",
     "from_ollama",
     "from_openai",
-    "from_transformersfrom_vllm",
+    "from_transformers",
+    "from_vllm",
 ]
 
 __all__ = [
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index a0b4a85d3..ee50356b0 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -10,7 +10,7 @@
 
 from .anthropic import from_anthropic, Anthropic
 from .base import Model, ModelTypeAdapter
-from .dottxt import Dottxt
+from .dottxt import Dottxt, from_dottxt
 from .exllamav2 import ExLlamaV2Model, exl2
 from .gemini import from_gemini, Gemini
 from .llamacpp import LlamaCpp, from_llamacpp
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index 4ba2f84a2..576a4c4cd 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -3,11 +3,14 @@
 import json
 from functools import singledispatchmethod
 from types import NoneType
-from typing import Optional
+from typing import Optional, TYPE_CHECKING
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.types import JsonType
 
+if TYPE_CHECKING:
+    from dottxt import Dottxt as DottxtClient
+
 __all__ = ["Dottxt"]
 
 
@@ -63,11 +66,15 @@ class Dottxt(Model):
 
     """
 
-    def __init__(self, model_name: Optional[str] = None, *args, **kwargs):
-        from dottxt.client import Dottxt
-
-        self.client = Dottxt(*args, **kwargs)
+    def __init__(
+        self,
+        client: "Dottxt",
+        model_name: Optional[str] = None,
+        model_revision: str = "",
+    ):
+        self.client = client
         self.model_name = model_name
+        self.model_revision = model_revision
         self.type_adapter = DottxtTypeAdapter()
 
     def generate(self, model_input, output_type=None, **inference_kwargs):
@@ -76,6 +83,7 @@ def generate(self, model_input, output_type=None, **inference_kwargs):
 
         if self.model_name:
             inference_kwargs["model_name"] = self.model_name
+            inference_kwargs["model_revision"] = self.model_revision
 
         completion = self.client.json(
             prompt,
@@ -83,3 +91,11 @@ def generate(self, model_input, output_type=None, **inference_kwargs):
             **inference_kwargs,
         )
         return completion.data
+
+
+def from_dottxt(
+    client: "DottxtClient",
+    model_name: Optional[str] = None,
+    model_revision: str = "",
+):
+    return Dottxt(client, model_name, model_revision)
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index 4c58392a5..6c3ea612b 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -4,6 +4,9 @@
 import pytest
 from pydantic import BaseModel
 
+from dottxt.client import Dottxt as DottxtClient
+
+import outlines
 from outlines.generate import Generator
 from outlines.models.dottxt import Dottxt
 from outlines.types import JsonType
@@ -31,39 +34,50 @@ def api_key():
 
 def test_dottxt_wrong_init_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
-        Dottxt(api_key=api_key, foo=10)
+        client = DottxtClient(api_key=api_key)
+        Dottxt(client, foo=10)
 
 
 def test_dottxt_wrong_output_type(api_key):
     with pytest.raises(NotImplementedError, match="must provide an output type"):
-        model = Dottxt(api_key=api_key)
+        client = DottxtClient(api_key=api_key)
+        model = Dottxt(client)
         model("prompt")
 
+def test_dottxt_init_from_client(api_key):
+    client = DottxtClient(api_key=api_key)
+    model = outlines.from_dottxt(client)
+    assert isinstance(model, Dottxt)
+    assert model.client == client
 
 @pytest.mark.api_call
 def test_dottxt_wrong_input_type(api_key):
     with pytest.raises(NotImplementedError, match="is not available"):
-        model = Dottxt(api_key=api_key)
+        client = DottxtClient(api_key=api_key)
+        model = Dottxt(client)
         model(["prompt"], JsonType(User))
 
 
 @pytest.mark.api_call
 def test_dottxt_wrong_inference_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = Dottxt(api_key=api_key)
+        client = DottxtClient(api_key=api_key)
+        model = Dottxt(client)
         model("prompt", JsonType(User), foo=10)
 
 
 @pytest.mark.api_call
 def test_dottxt_direct_call(api_key):
-    model = Dottxt(api_key=api_key, model_name="meta-llama/Llama-3.1-8B-Instruct")
+    client = DottxtClient(api_key=api_key)
+    model = Dottxt(client, model_name="meta-llama/Llama-3.1-8B-Instruct")
     result = model("Create a user", JsonType(User))
     assert "first_name" in json.loads(result)
 
 
 @pytest.mark.api_call
 def test_dottxt_generator_call(api_key):
-    model = Dottxt(api_key=api_key, model_name="meta-llama/Llama-3.1-8B-Instruct")
+    client = DottxtClient(api_key=api_key)
+    model = Dottxt(client, model_name="meta-llama/Llama-3.1-8B-Instruct")
     generator = Generator(model, JsonType(User))
     result = generator("Create a user")
     assert "first_name" in json.loads(result)

From 4e5afaf0d27e904dbf6975a1089e7035bd9bd2a7 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Mon, 24 Feb 2025 21:21:07 +0100
Subject: [PATCH 241/505] Rename Local/API into Steerable/BlackBox for
 models/generators (#1439)

---
 outlines/generate/__init__.py | 22 +++++++++++-----------
 outlines/models/__init__.py   |  4 ++--
 2 files changed, 13 insertions(+), 13 deletions(-)

diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
index 0d4f3c8e3..10bf78553 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 from typing import Any, Optional, Union, cast, get_args
 
-from outlines.models import APIModel, LlamaCpp, LocalModel
+from outlines.models import BlackBoxModel, SteerableModel
 from outlines.processors import CFGLogitsProcessor, RegexLogitsProcessor
 from outlines.types import CFG, Choice, JsonType, List, Regex
 
@@ -16,8 +16,8 @@
 
 
 @dataclass
-class APIGenerator:
-    """Represents an API-based generator.
+class BlackBoxGenerator:
+    """Represents a generator for which we don't control constrained generation.
 
     Attributes
     ----------
@@ -28,7 +28,7 @@ class APIGenerator:
 
     """
 
-    model: APIModel
+    model: BlackBoxModel
     output_type: Optional[Union[JsonType, List, Choice, Regex]] = None
 
     def __post_init__(self):
@@ -42,8 +42,8 @@ def __call__(self, prompt, **inference_kwargs):
 
 
 @dataclass
-class LocalGenerator:
-    """Represents a local model-based generator.
+class SteerableGenerator:
+    """Represents a generator for which we control constrained generation.
 
     We use this class to keep track of the logits processor which can be quite
     expensive to build.
@@ -57,7 +57,7 @@ class LocalGenerator:
 
     """
 
-    model: LocalModel
+    model: SteerableModel
     output_type: Optional[Union[JsonType, List, Choice, Regex]]
 
     def __post_init__(self):
@@ -80,10 +80,10 @@ def __call__(self, prompt, **inference_kwargs):
 
 
 def Generator(
-    model: Union[LocalModel, APIModel],
+    model: Union[SteerableModel, BlackBoxModel],
     output_type: Optional[Union[JsonType, List, Choice, Regex, CFG]] = None,
 ):
-    if isinstance(model, APIModel):  # type: ignore
-        return APIGenerator(model, output_type)  # type: ignore
+    if isinstance(model, BlackBoxModel):  # type: ignore
+        return BlackBoxGenerator(model, output_type)  # type: ignore
     else:
-        return LocalGenerator(model, output_type)  # type: ignore
+        return SteerableGenerator(model, output_type)  # type: ignore
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index ee50356b0..6a9ba753f 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -29,5 +29,5 @@
     Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM, Ollama
 ]
 
-LocalModel = Union[LlamaCpp, Transformers, MLXLM, VLLM]
-APIModel = Union[OpenAI, Anthropic, Gemini, Ollama, Dottxt]
+SteerableModel = Union[LlamaCpp, Transformers, MLXLM, VLLM]
+BlackBoxModel = Union[OpenAI, Anthropic, Gemini, Ollama, Dottxt]

From 547569235a5bdb0e383040c7a5dea36dd443afcc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Tue, 5 Nov 2024 09:48:09 +0100
Subject: [PATCH 242/505] Refactor the `OpenAI` interface

---
 outlines/models/openai.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 6038ad1a1..58f96740d 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,10 +1,10 @@
 """Integration with OpenAI's API."""
-
 from functools import singledispatchmethod
 from types import NoneType
 from typing import Optional, Union, TYPE_CHECKING
 
 
+
 from pydantic import BaseModel
 
 from outlines.models.base import Model, ModelTypeAdapter

From dd215b4fbe164dff83f181142baa1e643c6210fb Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 24 Feb 2025 20:35:21 +0100
Subject: [PATCH 243/505] Refactor the MLXLM model

---
 outlines/models/mlxlm.py                     | 273 +++++++------------
 outlines/processors/base_logits_processor.py |   9 +
 tests/models/test_mlxlm.py                   |  98 +++----
 3 files changed, 157 insertions(+), 223 deletions(-)

diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 380fb0150..deeb42d59 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -1,202 +1,125 @@
-import dataclasses
-from typing import TYPE_CHECKING, Generator, Iterator, List, Optional, Tuple, Union
+from functools import singledispatchmethod
+from typing import TYPE_CHECKING
 
-from .transformers import TransformerTokenizer
+from outlines.models.base import Model, ModelTypeAdapter
+from outlines.models.transformers import TransformerTokenizer
 
 if TYPE_CHECKING:
-    import mlx.core as mx
     import mlx.nn as nn
     from transformers import PreTrainedTokenizer
 
-    from outlines.generate.api import GenerationParameters, SamplingParameters
-    from outlines.processors import OutlinesLogitsProcessor
-
 
 __all__ = ["MLXLM", "from_mlxlm"]
 
 
-class MLXLM:
-    """
-    Represents an `mlx_lm` model
-    """
+class MLXLMTypeAdapter(ModelTypeAdapter):
+    """Type adapter for `mlx_lm` models."""
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available. "
+            "The `mlx_lm` library does not support batch inference."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        return model_input
+
+    def format_output_type(self, output_type):
+        """Generate the logits processor argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The logits processor provided.
+
+        """
+        if not output_type:
+            return None
+        return [output_type]
+
+
+class MLXLM(Model):
+    """Represents an `mlx_lm` model."""
 
     def __init__(
         self,
         model: "nn.Module",
         tokenizer: "PreTrainedTokenizer",
     ):
-        self.model = model
-        self.mlx_tokenizer = tokenizer  # returns mlx tensors, used for encode()
-        self.tokenizer = TransformerTokenizer(
-            tokenizer._tokenizer
-        )  # _tokenizer is HF Tokenizer
+        """Create a MLXLM model instance.
 
-    def generate(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: "GenerationParameters",
-        logits_processor,
-        sampling_parameters: "SamplingParameters",
-    ) -> str:
-        streamer = self.stream(
-            prompts, generation_parameters, logits_processor, sampling_parameters
-        )
-        return "".join(list(streamer))
+        Arguments
+        ---------
+        model
+            An instance of an mlx-lm model.
+        tokenizer
+            An instance of an mlx-lm tokenizer or of a compatible
+            transformers tokenizer.
 
-    def stream(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: "GenerationParameters",
-        logits_processor,
-        sampling_parameters: "SamplingParameters",
-    ) -> Iterator[str]:
-        """Generate text using `mlx_lm`.
-
-        Parameters
-        ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
-        logits_processor
-            The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-
-        Returns
-        -------
-        The generated text.
         """
-        import mlx.core as mx
+        self.model = model
+        # self.mlx_tokenizer is used by the mlx-lm in its generate function
+        self.mlx_tokenizer = tokenizer
+        # self.tokenizer is used by the logits processor
+        self.tokenizer = TransformerTokenizer(tokenizer._tokenizer)
+        self.type_adapter = MLXLMTypeAdapter()
+
+    def generate(self, model_input, output_type, **kwargs):
+        """Generate text using `mlx-lm`.
+
+        Arguments
+        ---------
+        model_input
+            The prompt passed by the user.
+        output_type
+            The logits processor provided.
+        kwargs
+            The inference kwargs that can be passed to the `mlx-lm` library.
 
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
+        """
+        from mlx_lm import generate
+
+        return generate(
+            self.model,
+            self.mlx_tokenizer,
+            self.type_adapter.format_input(model_input),
+            logits_processors=self.type_adapter.format_output_type(output_type),
+            **kwargs,
         )
-        if max_tokens is None:
-            max_tokens = int(1e9)
-
-        if not isinstance(prompts, str):
-            raise NotImplementedError(
-                "The `mlx-lm` library does not support batch inference."
-            )
-        if sampler == "beam_search":
-            raise NotImplementedError(
-                "The `mlx-lm` library does not support Beam Search."
-            )
-        if num_samples != 1:
-            raise NotImplementedError(
-                "The `mlx-lm` library does not allow to take several samples."
-            )
-        if top_k is not None:
-            raise NotImplementedError("The `mlx-lm` library does not support top_k.")
-        if seed is not None:
-            raise NotImplementedError("The `mlx-lm` library does not support seed.")
-        if stop_at is not None:
-            raise NotImplementedError("The `mlx-lm` library does not support stop_at.")
-
-        generate_kwargs = {
-            "temp": temperature,
-            "top_p": top_p,
-            "sampler": sampler,
-            "logits_processor": logits_processor,
-        }
-
-        # Adapted from
-        # https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L267
-        prompt_tokens = mx.array(self.mlx_tokenizer.encode(prompts))
-
-        detokenizer = self.mlx_tokenizer.detokenizer
-        detokenizer.reset()
-
-        for (token, prob), n in zip(
-            self.generate_step(prompt_tokens, **generate_kwargs),
-            range(max_tokens),
-        ):
-            if token == self.tokenizer.eos_token_id:
-                break
-            detokenizer.add_token(token)
-            yield detokenizer.last_segment
 
-        detokenizer.finalize()
-        yield detokenizer.last_segment
+    def stream(self, model_input, output_type, **kwargs):
+        """Stream text using `mlx-lm`.
+
+        Arguments
+        ---------
+        model_input
+            The prompt passed by the user.
+        output_type
+            The logits processor provided.
+        kwargs
+            The inference kwargs that can be passed to the `mlx-lm` library.
 
-    def generate_step(
-        self,
-        prompt: "mx.array",
-        temp: Optional[float],
-        top_p: Optional[float],
-        sampler: str,
-        logits_processor: "OutlinesLogitsProcessor",
-    ) -> Generator[Tuple[int, float], None, None]:
-        """
-        Adapted from
-        https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L129
-
-        A generator producing token ids based on the given prompt from the model.
-
-        Parameters
-        ----------
-        prompt
-            The input prompt.
-        temp
-            The temperature for sampling, if 0 the argmax is used.
-        top_p
-            Nulceus sampling, higher means model considers more less likely words.
-        sampler
-            The sampler string defined by SequenceGeneratorAdapter
-        logits_processor
-            Augment logits before sampling.
         """
-        import mlx.core as mx
-        import mlx_lm
-
-        temperature: float = temp or 1.0
-
-        def sample(logits: "mx.array") -> Tuple["mx.array", float]:
-            softmax_logits = mx.softmax(logits)
-
-            if temperature == 0.0 or sampler == "greedy":
-                token = mx.argmax(logits, axis=-1)
-            elif sampler == "multinomial":
-                if top_p is not None and top_p > 0 and top_p < 1.0:
-                    token = mlx_lm.sample_utils.top_p_sampling(
-                        logits, top_p, temperature
-                    )
-                else:
-                    token = mx.random.categorical(logits * (1 / temperature))
-            else:
-                raise ValueError(f"Invalid mlx-lm sampler: `{sampler}`")
-
-            prob = softmax_logits[0, token]
-            return token, prob
-
-        cache = mlx_lm.models.cache.make_prompt_cache(self.model)
-
-        # kv cache contains processed input IDs, we pass the unprocessed inputs and cache to model()
-        unprocessed_input_ids = prompt
-        generated_ids: List[int] = []
-
-        while True:
-            logits = self.model(unprocessed_input_ids[None], cache=cache)
-            logits = logits[:, -1, :]
-
-            if logits_processor is not None:
-                # convert to logits_processor 1d expectation, apply, then convert back
-                logits_1d = logits.reshape(-1)
-                logits_1d = logits_processor(generated_ids, logits_1d)
-                logits = logits_1d.reshape(1, -1)
-
-            new_token_single, prob = sample(logits)
-            new_token = new_token_single.item()
-            yield new_token, prob
-
-            generated_ids.append(new_token)
-            unprocessed_input_ids = new_token_single
+        from mlx_lm import stream_generate
+
+        for token in stream_generate(
+            self.model,
+            self.mlx_tokenizer,
+            self.type_adapter.format_input(model_input),
+            logits_processors=self.type_adapter.format_output_type(output_type),
+            **kwargs,
+        ):
+            yield token
 
 
 def from_mlxlm(model: "nn.Module", tokenizer: "PreTrainedTokenizer") -> MLXLM:
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index d6fe346e0..a74e7bf1f 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -81,6 +81,15 @@ def __call__(
         torch_logits = self._to_torch(logits)
         input_ids = self._to_torch(input_ids)
 
+        # if input_ids is 1D and logits is 2D with a single sequence,
+        # reshape input_ids to 2D (needed for mlx-lm)
+        if (
+            len(input_ids.shape) == 1
+            and len(torch_logits.shape) == 2
+            and torch_logits.shape[0] == 1
+        ):
+            input_ids = input_ids.unsqueeze(0)
+
         assert torch_logits.shape[:-1] == input_ids.shape[:-1]
 
         # Guarantee passed as 2D Tensors, then covert back to original (1D or 2D) shape
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index 10280b756..3a4d64681 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -1,11 +1,18 @@
 import pytest
+import re
+from enum import Enum
 
 import outlines
+from outlines.types import Choice, JsonType, Regex
 from outlines.models.transformers import TransformerTokenizer
+from pydantic import BaseModel
+from transformers import PreTrainedTokenizer
 
 try:
     import mlx_lm
     import mlx.core as mx
+    from mlx_lm import load
+    from mlx_lm.tokenizer_utils import TokenizerWrapper
 
     HAS_MLX = mx.metal.is_available()
 except ImportError:
@@ -22,8 +29,9 @@ def model(tmp_path_factory):
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_model(model):
+def test_mlxlm_model_initialization(model):
     assert hasattr(model, "model")
+    assert hasattr(model, "mlx_tokenizer")
     assert hasattr(model, "tokenizer")
     assert isinstance(model.tokenizer, TransformerTokenizer)
 
@@ -32,69 +40,63 @@ def test_mlxlm_model(model):
 def test_mlxlm_tokenizer(model):
     # Test single string encoding/decoding
     test_text = "Hello, world!"
-    token_ids = mx.array(model.mlx_tokenizer.encode(test_text))
+    token_ids, _ = model.tokenizer.encode(test_text)
+    token_ids = mx.array(token_ids)
     assert isinstance(token_ids, mx.array)
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_generate(model):
-    from outlines.generate.api import GenerationParameters, SamplingParameters
+def test_mlxlm_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
 
-    prompt = "Write a haiku about programming:"
 
-    # Test with basic generation parameters
-    gen_params = GenerationParameters(max_tokens=50, stop_at=None, seed=None)
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_invalid_input_type(model):
+    with pytest.raises(NotImplementedError, match="is not available"):
+        model(["Respond with one word. Not more."])
 
-    # Test with different sampling parameters
-    sampling_params = SamplingParameters(
-        sampler="multinomial", num_samples=1, top_p=0.9, top_k=None, temperature=0.7
-    )
 
-    # Test generation
-    output = model.generate(prompt, gen_params, None, sampling_params)
-    assert isinstance(output, str)
-    assert len(output) > 0
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_invalid_inference_kwargs(model):
+    with pytest.raises(TypeError):
+        model("Respond with one word. Not more.", foo="bar")
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_stream(model):
-    from outlines.generate.api import GenerationParameters, SamplingParameters
+def test_mlxlm_inference_kwargs(model):
+    result = model("Write a short story about a cat.", max_tokens=2)
+    assert isinstance(result, str)
+    assert len(result) < 20
 
-    prompt = "Count from 1 to 5:"
 
-    gen_params = GenerationParameters(max_tokens=20, stop_at=None, seed=None)
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_regex(model):
+    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
 
-    sampling_params = SamplingParameters(
-        sampler="greedy",  # Use greedy sampling for deterministic output
-        num_samples=1,
-        top_p=None,
-        top_k=None,
-        temperature=0.0,
-    )
 
-    # Test streaming
-    stream = model.stream(prompt, gen_params, None, sampling_params)
-    tokens = list(stream)
-    assert len(tokens) > 0
-    assert all(isinstance(token, str) for token in tokens)
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_JsonType(model):
+    class Character(BaseModel):
+        name: str
 
-    # Test that concatenated streaming output matches generate output
-    streamed_text = "".join(tokens)
-    generated_text = model.generate(prompt, gen_params, None, sampling_params)
-    assert streamed_text == generated_text
+    result = model("Create a character with a name.", JsonType(Character))
+    assert "name" in result
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_errors(model):
-    # Test batch inference (should raise NotImplementedError)
-    with pytest.raises(NotImplementedError):
-        from outlines.generate.api import GenerationParameters, SamplingParameters
-
-        gen_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
-        sampling_params = SamplingParameters("multinomial", 1, None, None, 1.0)
-        model.generate(["prompt1", "prompt2"], gen_params, None, sampling_params)
-
-    # Test beam search (should raise NotImplementedError)
-    with pytest.raises(NotImplementedError):
-        sampling_params = SamplingParameters("beam_search", 1, None, None, 1.0)
-        model.generate("test prompt", gen_params, None, sampling_params)
+def test_mlxlm_choice(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Choice(Foo))
+    assert result in ["cat", "dog"]

From 4e8dbfb52b0b503f2285bfedc50673ed1b0f0b35 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Tue, 25 Feb 2025 20:45:14 +0100
Subject: [PATCH 244/505] Remove the custom sampling loop using
 SequenceGenerator

---
 outlines/function.py             |   4 +-
 outlines/generate/__init__.py    |   2 +-
 outlines/generate/api.py         | 379 -----------------------
 outlines/generate/generator.py   | 312 -------------------
 tests/generate/test_generator.py | 497 -------------------------------
 5 files changed, 3 insertions(+), 1191 deletions(-)
 delete mode 100644 outlines/generate/generator.py
 delete mode 100644 tests/generate/test_generator.py

diff --git a/outlines/function.py b/outlines/function.py
index d7f6dec68..cabaf1fef 100644
--- a/outlines/function.py
+++ b/outlines/function.py
@@ -7,7 +7,7 @@
 import outlines
 
 if TYPE_CHECKING:
-    from outlines.generate.api import SequenceGenerator
+    from outlines.generate import SequenceGeneratorAdapter
     from outlines.templates import Template
 
 
@@ -25,7 +25,7 @@ class Function:
     prompt_template: "Template"
     schema: Union[str, Callable, object]
     model_name: str
-    generator: Optional["SequenceGenerator"] = None
+    generator: Optional["SequenceGeneratorAdapter"] = None
 
     @classmethod
     def from_github(cls, program_path: str, function_name: str = "fn"):
diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
index 10bf78553..bc91b6124 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -5,7 +5,7 @@
 from outlines.processors import CFGLogitsProcessor, RegexLogitsProcessor
 from outlines.types import CFG, Choice, JsonType, List, Regex
 
-from .api import SequenceGenerator
+from .api import SequenceGeneratorAdapter
 from .cfg import cfg
 from .choice import choice
 from .format import format
diff --git a/outlines/generate/api.py b/outlines/generate/api.py
index 5d3c52a8a..d9833eb12 100644
--- a/outlines/generate/api.py
+++ b/outlines/generate/api.py
@@ -3,8 +3,6 @@
 from dataclasses import dataclass
 from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
 
-from outlines.generate.generator import sequence_generator
-
 if TYPE_CHECKING:
     import torch
 
@@ -13,383 +11,6 @@
 ]
 
 
-class SequenceGenerator:
-    def __init__(
-        self,
-        fsm,
-        model,
-        sampler,
-        device,
-    ):
-        self.fsm = fsm
-        self.model = model
-        self.sampler = sampler
-        self.tokenizer = model.tokenizer
-        self.device = device
-        self.num_samples = sampler.samples
-
-    def get_generated_token_ids(
-        self,
-        prompt_token_ids: "torch.Tensor",
-        token_ids: "torch.Tensor",
-    ) -> List["torch.Tensor"]:
-        """Get the tokens generated so far.
-
-        Parameters
-        ----------
-        prompt_token_ids
-            Tensor that contains the token ids of the sequences' prompts.
-        token_ids
-            The generated token ids.
-
-        Returns
-        -------
-        A tensor that contains the token ids that have been generated so far.
-
-        """
-        prompt_lengths = [len(prompt) for prompt in prompt_token_ids]
-        token_ids = [
-            cur_token_ids[length:]
-            for cur_token_ids, length in zip(token_ids, prompt_lengths)
-        ]
-
-        return token_ids
-
-    def is_stop_sequence_found(
-        self, generated_sequences: List[str], stop_sequences: List[str]
-    ) -> bool:
-        """Determine whether one of the stop sequences has been generated.
-
-        Parameters
-        ----------
-        generated_sequences
-            The list of sequences generated so far.
-        stop_sequences
-            The list that contains the sequence which stop the generation when
-            found.
-
-        Returns
-        -------
-        True if at least one of the stop sequences has been found in each generated
-        sequence.
-
-        """
-        return all(
-            [
-                any([seq in generated for seq in stop_sequences])
-                for generated in generated_sequences
-            ]
-        )
-
-    def strip_stop_sequences(
-        self, sequence: str, stop_sequences: Optional[List[str]]
-    ) -> str:
-        """Remove the stop sequences from the generated sequences.
-
-        Parameters
-        ----------
-        sequence
-            One of the generated sequences.
-        stop_sequences
-            The list that contains the sequence which stop the generation when
-            found.
-
-        """
-        if stop_sequences:
-            match_indexes = [sequence.find(seq) for seq in stop_sequences]
-            if any([index != -1 for index in match_indexes]):
-                # select the stop_sequence that is found first in the sequence
-                min_match_index_value = min([i for i in match_indexes if i != -1])
-                min_match_index_pos = match_indexes.index(min_match_index_value)
-                sequence = sequence[
-                    : match_indexes[min_match_index_pos]
-                    + len(stop_sequences[min_match_index_pos])
-                ]
-
-        return sequence
-
-    def format_sequence(self, sequence: str) -> FormattedOutput:
-        """Translate the generated sequence to another type.
-
-        This method is for instance overridden when generating JSON to either
-        return a dictionnary or a Pydantic model.
-
-        Parameters
-        ----------
-        sequence
-            A generated sequences.
-
-        Returns
-        -------
-        The formatted sequence.
-
-        """
-        return sequence
-
-    def __call__(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        rng: Optional["torch.Generator"] = None,
-    ) -> Union[FormattedOutput, List[FormattedOutput], List[List[FormattedOutput]]]:
-        """Generate the full text sequence.
-
-        Since `SequenceGenerator.stream` calls the tokenizer at every step this
-        method loops over the generator returned by `sequence_generator` itself
-        so the tokenizer is called only once after all token ids have been
-        generated.
-
-        Parameters
-        ----------
-        prompts
-            A string or list of strings that are passed to the model before
-            generating the first token.
-        max_tokens
-            An integer representing maximum number of tokens that will be generated
-            (per prompt)
-        stop_at
-            A string or list of strings at which the text generated will stop
-        rng
-            The random number generator. Defaults to a non-seeded `torch.Generator`
-            instance.
-
-        Returns
-        -------
-        The generation(s), potentially cast to another type.
-        """
-        import torch
-
-        if isinstance(prompts, str):
-            prompts = [prompts]
-
-        if isinstance(stop_at, str):
-            stop_at = [stop_at]
-
-        stop_sequences = stop_at
-        num_samples = self.num_samples
-
-        if rng is None:
-            rng = torch.Generator(device=self.device)
-            rng.seed()
-
-        prompt_token_ids, attention_masks = self.tokenizer.encode(prompts)
-        prompt_token_ids = prompt_token_ids.to(self.device)
-        attention_masks = attention_masks.to(self.device)
-
-        # To draw multiple samples we repeat the prompt as many times
-        # as there are samples. We copy the FSMs and initialize the
-        # FSM states.
-        num_samples = self.num_samples
-        batch_size = len(prompts)
-
-        prompt_token_ids = torch.repeat_interleave(prompt_token_ids, num_samples, dim=0)
-        attention_masks = torch.repeat_interleave(attention_masks, num_samples, dim=0)
-        fsm_states = [0 for _ in range(batch_size * num_samples)]
-        fsms = [self.fsm.copy() for _ in range(batch_size * num_samples)]
-        weights = torch.zeros(
-            (batch_size * num_samples), dtype=torch.float, device=self.device
-        )
-
-        states = sequence_generator(
-            self.model,
-            self.sampler,
-            fsms,
-            prompt_token_ids,
-            weights,
-            attention_masks,
-            fsm_states,
-            rng=rng,
-        )
-
-        while True:
-            try:
-                last_state = next(states)
-                if max_tokens or stop_sequences:
-                    token_ids = last_state.token_ids
-                    generated_token_ids = self.get_generated_token_ids(
-                        prompt_token_ids, token_ids
-                    )
-                    if max_tokens and len(generated_token_ids[0]) >= max_tokens:
-                        break
-                    if stop_sequences and self.is_stop_sequence_found(
-                        self.tokenizer.decode(generated_token_ids), stop_sequences
-                    ):
-                        break
-            except StopIteration:
-                break
-
-        token_ids = last_state.token_ids
-        generated_token_ids = self.get_generated_token_ids(prompt_token_ids, token_ids)
-
-        generated = self.tokenizer.decode(generated_token_ids)
-        stripped = [
-            self.strip_stop_sequences(sequence, stop_sequences)
-            for sequence in generated
-        ]
-        formatted = [self.format_sequence(sequence) for sequence in stripped]
-
-        # We reshape the output to (batch_size, sample_size)
-        output: List[List[FormattedOutput]] = list()
-        for i in range(0, batch_size * num_samples, num_samples):
-            output.append(formatted[i : i + num_samples])
-
-        # We remove leading dimensions for the output
-        if batch_size == 1 and num_samples == 1:
-            return output[0][0]
-        elif batch_size == 1:
-            return output[0]
-        elif num_samples == 1:
-            return [samples[0] for samples in output]
-        else:
-            return output
-
-    def stream(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        rng: Optional["torch.Generator"] = None,
-    ) -> Iterator[Union[List[str], str, List[List[str]]]]:
-        """Generate the text sequence one token at a time.
-
-        Since `Tokenizer.decode` strips the whitespaces from the tokens we have no
-        choice but to decode the generated token ids at each step and compare the
-        current decoded strings to the previously decoded strings.
-
-        Parameters
-        ----------
-        prompts
-            A string or list of strings that are passed to the model before
-            generating the first token.
-        max_tokens
-            An integer representing maximum number of tokens that will be generated
-            (per prompt)
-        stop_at
-            A string or list of strings at which the text generated will stop
-        rng
-            The random number generator. Defaults to a non-seeded `torch.Generator`
-            instance.
-
-        Returns
-        -------
-        A string or list of strings that contain the generated text.
-
-        """
-        import torch
-
-        if isinstance(prompts, str):
-            prompts = [prompts]
-
-        if isinstance(stop_at, str):
-            stop_at = [stop_at]
-
-        stop_sequences = stop_at
-        num_samples = self.num_samples
-
-        prompt_token_ids, attention_masks = self.tokenizer.encode(prompts)
-        prompt_token_ids = prompt_token_ids.to(self.device)
-        attention_masks = attention_masks.to(prompt_token_ids.device)
-
-        # To draw multiple samples we repeat the prompt as many times
-        # as there are samples. We copy the FSMs and initialize the
-        # FSM states.
-        num_samples = self.num_samples
-        batch_size = len(prompts)
-
-        prompt_token_ids = torch.repeat_interleave(prompt_token_ids, num_samples, dim=0)
-        attention_masks = torch.repeat_interleave(attention_masks, num_samples, dim=0)
-        fsm_states = [0 for _ in range(batch_size * num_samples)]
-        fsms = [self.fsm.copy() for _ in range(batch_size * num_samples)]
-        weights = torch.zeros(
-            (batch_size * num_samples),
-            dtype=torch.float,
-            device=prompt_token_ids.device,
-        )
-
-        if rng is None:
-            rng = torch.Generator(device=prompt_token_ids.device)
-            rng.seed()
-
-        states = sequence_generator(
-            self.model,
-            self.sampler,
-            fsms,
-            prompt_token_ids,
-            weights,
-            attention_masks,
-            fsm_states,
-            rng=rng,
-        )
-
-        def token_generator() -> Iterator[Union[List[str], str, List[List[str]]]]:
-            previously_generated_sequences = [
-                "" for _ in range(batch_size)
-            ] * num_samples
-            num_generated = 0
-            is_stop_at_reached = [False for _ in range(batch_size)] * num_samples
-            while True:
-                if (max_tokens and num_generated >= max_tokens) or all(
-                    is_stop_at_reached
-                ):
-                    return
-                try:
-                    sequence = next(states)
-                    num_generated += 1
-                except StopIteration:
-                    return
-                generated_token_ids = sequence.token_ids[:, -num_generated:]
-                generated_sequences = self.tokenizer.decode(generated_token_ids)
-                if stop_sequences:
-                    is_stop_at_reached = [
-                        stop
-                        or self.is_stop_sequence_found(
-                            [generated_sequence], stop_sequences
-                        )
-                        for generated_sequence, stop in zip(
-                            generated_sequences, is_stop_at_reached
-                        )
-                    ]
-
-                    generated_sequences = [
-                        (
-                            self.format_sequence(
-                                self.strip_stop_sequences(sequence, stop_sequences)
-                            )
-                            if stop
-                            else sequence
-                        )
-                        for sequence, stop in zip(
-                            generated_sequences, is_stop_at_reached
-                        )
-                    ]
-                next_tokens = [
-                    token[len(sequence) :]
-                    for token, sequence, stop in zip(
-                        generated_sequences,
-                        previously_generated_sequences,
-                        is_stop_at_reached,
-                    )
-                ]
-                previously_generated_sequences = generated_sequences
-                # We reshape the output to (batch_size, sample_size)
-                output: List[List[str]] = list()
-                for i in range(0, batch_size * num_samples, num_samples):
-                    output.append(next_tokens[i : i + num_samples])
-
-                # We remove leading dimensions for the output
-                if batch_size == 1 and num_samples == 1:
-                    yield output[0][0]
-                elif batch_size == 1:
-                    yield output[0]
-                elif num_samples == 1:
-                    yield [samples[0] for samples in output]
-                else:
-                    yield output
-
-        return token_generator()
-
-
 @dataclass(frozen=True)
 class GenerationParameters:
     """Generation parameters used in Outlines' public API."""
diff --git a/outlines/generate/generator.py b/outlines/generate/generator.py
deleted file mode 100644
index e506aa035..000000000
--- a/outlines/generate/generator.py
+++ /dev/null
@@ -1,312 +0,0 @@
-import dataclasses
-import math
-from typing import TYPE_CHECKING, Callable, Iterable, Iterator, List, Optional, Tuple
-
-if TYPE_CHECKING:
-    import torch
-
-    from outlines.fsm.guide import Guide
-
-
-class ContextLengthExceededError(Exception):
-    pass
-
-
-@dataclasses.dataclass(frozen=True)
-class GenerationState:
-    token_ids: "torch.Tensor"
-    kv_cache: "torch.Tensor"
-    logits: "torch.Tensor"
-    weights: "torch.Tensor"
-    fsm_states: List[int]
-
-
-def sequence_generator(
-    model: Callable,
-    sampler: Callable,
-    fsms: List["Guide"],
-    token_ids: "torch.Tensor",
-    sequence_weights: "torch.Tensor",
-    attention_masks: "torch.Tensor",
-    fsm_states: List[int],
-    rng: "torch.Generator",
-) -> Iterator[GenerationState]:
-    """Generates sequences of tokens.
-
-    Parameters
-    ----------
-    model
-        A callable that generates a probability distribution over the
-        vocabulary when passed a tensor of token ids.
-    sampler
-        A callable that returns the next token ids, their ancestor sequence and
-        the updated sequence weights when passed a distribution over the
-        vocabulary.
-    token_ids
-        A tensor of token ids on which the sequence distribution is conditioned, of
-        shape ``(n_seqs, n_prompt_tokens)``
-    sequence_weights
-        A tensor that contains the initial weights of the sequences, of shape
-        ``(n_seqs,)``
-    attention_masks
-        A tensor of tensors that represent the tokens considered at the attention
-        layer, of shape ``(n_seqs, n_prompt_tokens)``.
-    fsms
-        List of finite-state machines that drive the text generation,
-        one for each sequence in the batch.
-    fsm_states
-        The initial states of the finite-state machine for each sequence in the batch.
-
-    Yields
-    ------
-    A new sequence.
-
-    """
-    import torch
-
-    if rng is None:
-        rng = torch.Generator()
-
-    kv_cache = None
-
-    while True:
-        try:
-            logits, kv_cache = model(token_ids, attention_masks, kv_cache)
-        except IndexError:  # Exceeding the context length
-            raise ContextLengthExceededError(
-                "The input length exceeds the context length of the model."
-            )
-
-        allowed_tokens = get_allowed_tokens(fsms, fsm_states)
-        biased_logits = bias_logits(logits, allowed_tokens)
-        next_token_ids, ancestors, sequence_weights = sampler(
-            biased_logits, sequence_weights, rng
-        )
-
-        token_ids = update_token_ids(token_ids, next_token_ids, ancestors)
-        attention_masks = update_attention_masks(attention_masks, ancestors)
-        kv_cache = reorder_kv_cache(kv_cache, ancestors)
-        if len(ancestors) > 1:
-            fsms = reorder_fsms(fsms, ancestors)
-            fsm_states = reorder_fsm_states(fsm_states, ancestors)
-
-        fsm_states = get_next_fsm_states(fsms, fsm_states, next_token_ids)
-        is_finished = is_generation_finished(fsms, fsm_states)
-
-        if is_finished:
-            yield GenerationState(
-                token_ids,
-                kv_cache,
-                logits,
-                sequence_weights,
-                fsm_states,
-            )
-            return
-
-        yield GenerationState(
-            token_ids,
-            kv_cache,
-            logits,
-            sequence_weights,
-            fsm_states,
-        )
-
-
-def get_next_fsm_states(
-    fsms: List["Guide"], fsm_states: List[int], next_token_ids: "torch.Tensor"
-) -> List[int]:
-    """
-
-    Parameters
-    ----------
-    fsm
-        The finite-state machine used to monitor this batch.
-    next_token_ids
-        The tokens that were just generated.
-
-    Returns
-    -------
-    A `torch.Tensor` object that represents the next logit mask.
-
-    """
-    return [
-        fsm.get_next_state(fsm_state, int(token_id[0]))
-        for fsm, fsm_state, token_id in zip(fsms, fsm_states, next_token_ids)
-    ]
-
-
-def get_allowed_tokens(
-    fsms: List["Guide"], fsm_states: List[int]
-) -> List[Optional[Iterable[int]]]:
-    """Get the new instructions for each sequence from the finite-state machine.
-
-    Parameters
-    ----------
-    fsm
-        The finite-state machine used to monitor this batch.
-    fsm_states
-        The FSM states corresponding to each sequence in the batch.
-
-    Returns
-    -------
-    A nested list that contains the ids of the logits to keep.
-
-    """
-    return [
-        fsm.get_next_instruction(state).tokens for fsm, state in zip(fsms, fsm_states)
-    ]
-
-
-def is_generation_finished(fsms: List["Guide"], fsm_states: List[int]) -> bool:
-    """Determine if the generation is finished.
-
-    A generation is considered finished if the FSM of every sequence in the
-    batch is in a final state.
-
-    A better solution is to return finished sequences as soon as their FSM
-    is in a final state.
-
-    Parameters
-    ----------
-    fsm
-        The finite-state machine used to monitor this batch.
-    fsm_states
-        The FSM states corresponding to each sequence in the batch.
-
-    Returns
-    -------
-    Whether all sequences are finished sampling.
-
-    """
-    return all([fsm.is_final_state(state) for fsm, state in zip(fsms, fsm_states)])
-
-
-def update_token_ids(
-    token_ids: "torch.Tensor", next_token_ids: "torch.Tensor", ancestors: "torch.Tensor"
-) -> "torch.Tensor":
-    """Append the sampled tokens to the running sequence of tokens.
-
-    Parameters
-    ----------
-    token_ids
-        The current token sequences
-    next_token_ids
-        The tokens that were just generated and that we need to append
-        to the existing sequences.
-    ancestors
-        The sequences to which the token ids need to be added.
-
-    Returns
-    -------
-    A new sequence of token ids that contains the tokens that were
-    just generated.
-
-    """
-    import torch
-
-    token_ids = torch.index_select(token_ids, 0, ancestors)
-    return torch.concatenate([token_ids, next_token_ids], dim=-1)
-
-
-def update_attention_masks(
-    attention_masks: "torch.Tensor", ancestors: "torch.Tensor"
-) -> "torch.Tensor":
-    """Expand the attention masks.
-
-    Parameters
-    ----------
-    attention_masks
-        The attention masks for each sequence in the batch.
-    ancestors
-        The sequences to which the token ids need to be added.
-
-    Returns
-    -------
-    The attention masks padded with 1s.
-
-    """
-    import torch
-
-    attention_masks = torch.index_select(attention_masks, 0, ancestors)
-    return torch.concatenate(
-        [
-            attention_masks,
-            torch.ones(
-                attention_masks.shape[:-1] + (1,), device=attention_masks.device
-            ),
-        ],
-        axis=-1,
-    )
-
-
-def reorder_fsms(fsms: List["Guide"], ancestors: "torch.Tensor") -> List["Guide"]:
-    reordered_fsms = []
-    for ancestor in ancestors:
-        reordered_fsms.append(fsms[ancestor].copy())
-
-    return reordered_fsms
-
-
-def reorder_fsm_states(fsm_states: List[int], ancestors: "torch.Tensor") -> List[int]:
-    reordered_states = []
-    for ancestor in ancestors:
-        reordered_states.append(fsm_states[ancestor])
-
-    return reordered_states
-
-
-def reorder_kv_cache(
-    kv_cache: Optional[Tuple], ancestors: "torch.Tensor"
-) -> Optional[Tuple]:
-    """Re-order the KV-cache based on the ancestors.
-
-    In transformers, the object that stores the KV-cache is a tuple who elements
-    are the key cache and the value cache. Each of these caches are tuples where
-    each element correpond to a layer. To each layer corresponds a tensor whose
-    first dimension is the batch size.
-
-    """
-    import torch
-
-    if kv_cache is None:
-        return None
-
-    new_kv_cache: Tuple = tuple()
-    for cache_item in kv_cache:
-        new_cache_item: Tuple = tuple()
-        for layer in cache_item:
-            layer = torch.index_select(layer, 0, ancestors.to(layer.device))
-            new_cache_item += (layer,)
-        new_kv_cache += (new_cache_item,)
-
-    return new_kv_cache
-
-
-def bias_logits(logits: "torch.Tensor", allowed_token_ids: List) -> "torch.Tensor":
-    """Mask the logits.
-
-    The function iterates over a nested list where each list corresponds to the
-    indices that need to be masked for each row in the array.
-
-    Parameters
-    ----------
-    logits
-        Two dimensional tensor that contains the next-token probability
-        distribution.
-    allowed_token_ids
-        A list that contains the tokens that can be generated by the model.
-
-    Returns
-    -------
-    A view of the original logits tensor where some values are masked.
-
-    """
-    import torch
-
-    biased_logits = torch.full_like(logits, -math.inf, device=logits.device)
-    for i, ids in enumerate(allowed_token_ids):
-        if ids is not None:
-            biased_logits[i, ids] = logits[i, ids]
-        else:
-            biased_logits[i] = logits[i]
-    return biased_logits
diff --git a/tests/generate/test_generator.py b/tests/generate/test_generator.py
deleted file mode 100644
index 5a2edf8dc..000000000
--- a/tests/generate/test_generator.py
+++ /dev/null
@@ -1,497 +0,0 @@
-import math
-from typing import Generator
-
-import pytest
-import torch
-
-from outlines.fsm.guide import Generate
-from outlines.generate.api import SequenceGenerator
-from outlines.generate.generator import (
-    bias_logits,
-    get_allowed_tokens,
-    get_next_fsm_states,
-    is_generation_finished,
-    sequence_generator,
-    update_attention_masks,
-    update_token_ids,
-)
-
-
-def test_sequence_generator_class():
-    class MockFSM:
-        first_state = 0
-
-        def get_next_state(self, state, next_token_ids):
-            return 4
-
-        def get_next_instruction(self, *_):
-            return Generate([4])
-
-        def is_final_state(self, _):
-            return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            # Input: "test"
-            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
-
-        def decode(self, tokens):
-            return ["testx"[i] for i in tokens]
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3, 4]], dtype=torch.float), None
-
-    class sampler:
-        def __init__(self):
-            self.samples = 1
-
-        def __call__(self, biased_logits, *_):
-            return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
-
-    # Stream
-    generator = SequenceGenerator(MockFSM(), MockModel(), sampler(), "cpu")
-    assert generator.device == "cpu"
-    assert isinstance(generator.tokenizer, MockTokenizer)
-    assert isinstance(generator.fsm, MockFSM)
-
-    sequence = generator.stream("test")
-    assert isinstance(sequence, Generator)
-
-    next(sequence)
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-    # Call
-    generator = SequenceGenerator(MockFSM(), MockModel(), sampler(), "cpu")
-    result = generator("test")
-    assert result == "x"
-
-
-def test_sequence_generator_1d_single_iteration():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return 0
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, _):
-            return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3]]),
-        torch.tensor([[1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        [MockFSM()],
-        token_ids,
-        torch.tensor([0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-    result = next(sequence)
-
-    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3]]))
-    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_sequence_generator_1d_several_iterations():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return state + 1
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, state):
-            if state < 2:
-                return False
-            else:
-                return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3]]), torch.tensor([[1, 1, 1, 1]])
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return torch.argmax(biased_logits, keepdims=True), torch.tensor([0]), None
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3]]),
-        torch.tensor([[1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        [MockFSM()],
-        token_ids,
-        torch.tensor([0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-
-    result = next(sequence)
-    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3]]))
-    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
-
-    result = next(sequence)
-    assert torch.equal(result.token_ids, torch.tensor([[0, 1, 2, 3, 3, 3]]))
-    assert torch.equal(result.logits, torch.tensor([[0, 1, 2, 3]]))
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_sequence_generator_2d_single_iteration():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return 0
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, _):
-            return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]), torch.tensor(
-                [[1, 1, 1, 1], [1, 1, 1, 1]]
-            )
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return (
-            torch.argmax(biased_logits, keepdims=True, dim=-1),
-            torch.tensor([0, 1]),
-            None,
-        )
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]),
-        torch.tensor([[1, 1, 1, 1], [1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0, 0]
-    fsms = [MockFSM(), MockFSM()]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        fsms,
-        token_ids,
-        torch.tensor([0, 0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-
-    result = next(sequence)
-    assert torch.equal(
-        result.token_ids, torch.tensor([[0, 1, 2, 3, 3], [4, 5, 6, 7, 2]])
-    )
-    assert torch.equal(
-        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
-    )
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_sequence_generator_2d_several_iterations():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return state + 1
-
-        def get_next_instruction(self, _):
-            return Generate([0, 1, 2, 3])
-
-        def is_final_state(self, state):
-            if state < 2:
-                return False
-            else:
-                return True
-
-        def copy(self):
-            return self
-
-    class MockTokenizer:
-        def encode(self, _):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]), torch.tensor(
-                [[1, 1, 1, 1], [1, 1, 1, 1]]
-            )
-
-        def decode(self, x):
-            return x
-
-    class MockModel:
-        def __init__(self):
-            self.tokenizer = MockTokenizer()
-
-        def __call__(*_):
-            return torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float), None
-
-    def sampler(biased_logits, *_):
-        return (
-            torch.argmax(biased_logits, keepdims=True, dim=-1),
-            torch.tensor([0, 1]),
-            None,
-        )
-
-    token_ids, attention_mask = (
-        torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]]),
-        torch.tensor([[1, 1, 1, 1], [1, 1, 1, 1]]),
-    )
-    init_fsm_states = [0, 0]
-    fsms = [MockFSM(), MockFSM()]
-    sequence = sequence_generator(
-        MockModel(),
-        sampler,
-        fsms,
-        token_ids,
-        torch.tensor([0, 0]),
-        attention_mask,
-        init_fsm_states,
-        rng=torch.Generator(),
-    )
-
-    result = next(sequence)
-    assert torch.equal(
-        result.token_ids, torch.tensor([[0, 1, 2, 3, 3], [4, 5, 6, 7, 2]])
-    )
-    assert torch.equal(
-        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
-    )
-
-    result = next(sequence)
-    assert torch.equal(
-        result.token_ids, torch.tensor([[0, 1, 2, 3, 3, 3], [4, 5, 6, 7, 2, 2]])
-    )
-    assert torch.equal(
-        result.logits, torch.tensor([[0, 1, 2, 3], [4, 5, 7, 6]], dtype=torch.float)
-    )
-
-    with pytest.raises(StopIteration):
-        next(sequence)
-
-
-def test_get_next_fsm_states():
-    class MockFSM:
-        def get_next_state(self, state, next_token_ids):
-            return 0
-
-        def copy(self):
-            return self
-
-    result = get_next_fsm_states([MockFSM()], [0], torch.tensor([[0]]))
-    assert result == [0]
-
-    result = get_next_fsm_states(
-        [MockFSM(), MockFSM()], [0, 0], torch.tensor([[0], [0]])
-    )
-    assert result == [0, 0]
-
-
-def test_get_get_next_instructions():
-    class MockFSM:
-        def get_next_instruction(self, _):
-            return Generate([1, 2, 3, 4])
-
-    result = get_allowed_tokens([MockFSM()], [0])
-    assert result == [[1, 2, 3, 4]]
-
-    result = get_allowed_tokens([MockFSM(), MockFSM()], [0, 1])
-    assert result == [[1, 2, 3, 4], [1, 2, 3, 4]]
-
-
-def test_is_generation_finished():
-    class MockFSMFinished:
-        def is_final_state(self, _):
-            return True
-
-    result = is_generation_finished([MockFSMFinished(), MockFSMFinished()], [1, 1])
-    assert result is True
-
-    class MockFSMNotFinished:
-        def is_final_state(self, state):
-            if state == 0:
-                return False
-            else:
-                return True
-
-    result = is_generation_finished(
-        [MockFSMNotFinished(), MockFSMNotFinished()], [0, 1]
-    )
-    assert result is False
-
-
-@pytest.mark.parametrize(
-    "token_ids,next_token_ids,ancestors,expected_result",
-    [
-        (
-            torch.tensor([[1]]),
-            torch.tensor([[2]]),
-            torch.tensor([0]),
-            torch.tensor([[1, 2]]),
-        ),
-        (
-            torch.tensor([[1], [2]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([0, 1]),
-            torch.tensor([[1, 3], [2, 4]]),
-        ),
-        (
-            torch.tensor([[1], [2]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([0, 0]),
-            torch.tensor([[1, 3], [1, 4]]),
-        ),
-        (
-            torch.tensor([[1], [2]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([1, 0]),
-            torch.tensor([[2, 3], [1, 4]]),
-        ),
-        (
-            torch.tensor([[1, 2], [3, 5]]),
-            torch.tensor([[3], [4]]),
-            torch.tensor([1, 0]),
-            torch.tensor([[3, 5, 3], [1, 2, 4]]),
-        ),
-    ],
-)
-def test_update_token_ids(token_ids, next_token_ids, ancestors, expected_result):
-    result = update_token_ids(token_ids, next_token_ids, ancestors)
-    assert torch.equal(result, expected_result)
-
-
-@pytest.mark.parametrize(
-    "attention_masks,ancestors,expected_result",
-    [
-        (
-            torch.tensor([[1, 1]], dtype=torch.float),
-            torch.tensor([0]),
-            torch.tensor([[1, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([0, 0]),
-            torch.tensor([[1, 1, 1], [1, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([0, 1]),
-            torch.tensor([[0, 1, 1], [1, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([1, 0]),
-            torch.tensor([[1, 1, 1], [0, 1, 1]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[0, 1], [1, 1]], dtype=torch.float),
-            torch.tensor([1, 1]),
-            torch.tensor([[1, 1, 1], [1, 1, 1]], dtype=torch.float),
-        ),
-    ],
-)
-def test_expand_attention_masks(attention_masks, ancestors, expected_result):
-    result = update_attention_masks(attention_masks, ancestors)
-    assert torch.equal(result, expected_result)
-
-
-@pytest.mark.parametrize(
-    "logits,indices_to_mask,expected",
-    [
-        (
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-            [[0, 1, 2, 3]],
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-            [[0, 2, 3]],
-            torch.tensor([[1, -math.inf, 3, 4]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3, 4]], dtype=torch.float),
-            [[0, 2]],
-            torch.tensor([[1, -math.inf, 3, -math.inf]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3], [4, 5, 6]], dtype=torch.float),
-            [[1, 2], [0, 1]],
-            torch.tensor([[-math.inf, 2, 3], [4, 5, -math.inf]], dtype=torch.float),
-        ),
-        (
-            torch.tensor([[1, 2, 3], [4, 5, 6]], dtype=torch.float),
-            [[0, 2], [1]],
-            torch.tensor(
-                [[1, -math.inf, 3], [-math.inf, 5, -math.inf]], dtype=torch.float
-            ),
-        ),
-    ],
-)
-def test_bias_logits(logits, indices_to_mask, expected):
-    masked_logits = bias_logits(logits, indices_to_mask)
-    assert torch.equal(masked_logits, expected)

From 15e28661405cdb829c900cfca9af9c1e1aeb6860 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Wed, 26 Feb 2025 15:08:09 +0100
Subject: [PATCH 245/505] Create nightly tests run in Github actions (#1447)

Co-authored-by: Robin Picard <robin@dottxt.co>
---
 .github/workflows/tests_nightly.yml | 60 +++++++++++++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 .github/workflows/tests_nightly.yml

diff --git a/.github/workflows/tests_nightly.yml b/.github/workflows/tests_nightly.yml
new file mode 100644
index 000000000..c34a707fa
--- /dev/null
+++ b/.github/workflows/tests_nightly.yml
@@ -0,0 +1,60 @@
+name: Nightly Tests
+
+# Run every day at 6:00 AM UTC
+on:
+  schedule:
+    - cron: "0 6 * * *"
+
+jobs:
+  tests:
+    name: Run the tests
+    runs-on: ubuntu-latest
+    env:
+      OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+      ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
+      GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
+      DOTTXT_API_KEY: ${{ secrets.DOTTXT_API_KEY }}
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.10"]
+
+    steps:
+    - uses: actions/checkout@v3
+    - name: Set up Python ${{ matrix.python-version }}
+      uses: actions/setup-python@v4
+      with:
+        python-version: ${{ matrix.python-version }}
+    - name: Install Ollama
+      run: |
+        curl -fsSL https://ollama.com/install.sh | sh
+        ollama --version
+        ollama pull tinyllama
+    - name: Set up test environment
+      run: |
+        python -m pip install --upgrade pip
+        pip install uv
+        uv venv
+        uv pip install -e .[test]
+    - name: Create matrix id
+      id: matrix-id
+      env:
+        MATRIX_CONTEXT: ${{ toJson(matrix) }}
+      run: |
+        echo $MATRIX_CONTEXT
+        export MATRIX_ID=`echo $MATRIX_CONTEXT | md5sum | cut -c 1-32`
+        echo $MATRIX_ID
+        echo "::set-output name=id::$MATRIX_ID"
+    - name: Run tests
+      run: |
+        uv run pytest -x --cov=outlines
+      env:
+        COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
+    - name: Slack notification on failure
+      if: failure()
+      uses: 8398a7/action-slack@v3
+      with:
+        status: ${{ job.status }}
+        fields: commit,author,ref
+      env:
+        SLACK_WEBHOOK_URL: ${{ secrets.SLACK_TESTS_ERRORS_WEBHOOK_URL }}

From 17bf3e3ba70ac277ae3ce953b59896fc6dd37351 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Thu, 27 Feb 2025 00:32:08 +0100
Subject: [PATCH 246/505] Add support for Transformers audio models (#1452)

This PR turn the class `TransformersVision` into
`TransformersMultiModal` that accommodates both vision and audio models.
We should be able to use this same class for video models with minimal
change, but I did not find a good model to test it out.
---
 docs/reference/models/transformers.md         |   2 +-
 .../models/transformers_multimodal.md         | 196 ++++++++++++++++++
 docs/reference/models/transformers_vision.md  | 117 -----------
 outlines/generate/cfg.py                      |   4 +-
 outlines/generate/fsm.py                      |   4 +-
 outlines/generate/regex.py                    |   4 +-
 outlines/generate/text.py                     |   4 +-
 outlines/models/__init__.py                   |   2 +-
 outlines/models/transformers.py               |  76 +++----
 outlines/models/transformers_vision.py        |   0
 ...ion.py => test_transformers_multimodal.py} |  36 ++--
 11 files changed, 256 insertions(+), 189 deletions(-)
 create mode 100644 docs/reference/models/transformers_multimodal.md
 delete mode 100644 docs/reference/models/transformers_vision.md
 delete mode 100644 outlines/models/transformers_vision.py
 rename tests/models/{test_transformers_vision.py => test_transformers_multimodal.py} (81%)

diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index 724a724c3..c3b397297 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -21,7 +21,7 @@ model = models.from_transformers(
 )
 ```
 
-The model name must be a valid `transformers` model name. You can find a list of all in the HuggingFace library [here](https://huggingface.co/models). We currently support `CausalLM`, `Seq2Seq`, `Mamba` and vision models.
+The model name must be a valid `transformers` model name. You can find a list of all in the HuggingFace library [here](https://huggingface.co/models). We currently support `CausalLM`, `Seq2Seq`, `Mamba`, vision and audio models.
 
 Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
 
diff --git a/docs/reference/models/transformers_multimodal.md b/docs/reference/models/transformers_multimodal.md
new file mode 100644
index 000000000..cea644204
--- /dev/null
+++ b/docs/reference/models/transformers_multimodal.md
@@ -0,0 +1,196 @@
+# Transformers Multimodal
+
+Outlines allows seamless use of [multimodal models](https://huggingface.co/learn/multimodal-models-course/en/unit1/introduction).
+
+`outlines.models.TransformersMultimodal` shares interfaces with, and is based on [outlines.models.Transformers](./transformers.md).
+
+## Create a `TransformersMultimodal` model
+
+`TransformersMultimodal` models inherit from `Transformers` and accept similar initialization parameters, with the exception that you should provide a processor instead of a tokenizer.
+
+You can use `outlines.from_transformers` to load a `transformers` model and processor:
+
+```python
+from transformers import CLIPModel, CLIPProcessor
+from outlines import models
+
+model_name = "openai/clip-vit-base-patch32"
+model = models.from_transformers(
+    CLIPModel.from_pretrained(model_name),
+    CLIPProcessor.from_pretrained(model_name)
+)
+```
+
+You must provider a processor adapted to your model. We currently only support vision and audio models.
+
+## Use the model to generate text from prompts and images/audios
+
+When calling the model, the prompt argument you provide must be a dictionary using the following format:
+```
+{
+    "text": Union[str, List[str]],
+    "images": Optional[Union[Any, List[Any]]],
+    "audios": Optional[Union[Any, List[Any]]],
+}
+```
+
+The `text` key is required. Either the `images` or `audios` key must be provided.
+The format and values of those keys must correspond to what you would provide to the `__call__` method of the processor if you were to call it directly. Thus, the text prompt must include the appropriate tags to indicate where the images or audios should be inserted (e.g. `<image>` or `<|AUDIO|>`).
+
+### Vision models
+
+For easier use, we recommend you to create a convenience function to load a `PIL.Image` from URL.
+```python
+from PIL import Image
+from io import BytesIO
+from urllib.request import urlopen
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+```
+
+You can then call the model with your prompts and images to generate text.
+```python
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
+from outlines import models
+
+MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+
+model = models.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
+    LlavaProcessor.from_pretrained(MODEL_NAME)
+)
+prompt = {
+    "text": "<image> detailed description:",
+    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
+}
+model(prompt)
+```
+
+You can include several images per prompt by adding more `<image>` tags to the prompt. Batching is also supported.
+```python
+from transformers import LlavaForConditionalGeneration
+from outlines import models
+
+MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+
+model = models.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
+    LlavaProcessor.from_pretrained(MODEL_NAME)
+)
+prompt = {
+    "text": ["<image><image>detailed description:", "<image><image>. What animals are present?"],
+    "images": [
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
+        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
+    ]
+}
+model(prompt)
+```
+
+Here we have two prompts, each expecting two images. We correspondingly provide four images. This will generate two descriptions, one for each prompt.
+
+### Audio models
+
+For easier use, we recommend you to create a convenience function to turn a `.wav` file from a URL into an audio array.
+```python
+import soundfile
+from io import BytesIO
+from urllib.request import urlopen
+
+def audio_from_url(url):
+    audio_data = BytesIO(urlopen(url).read())
+    audio_array, _ = soundfile.read(audio_data)
+    return audio_array
+```
+
+You can then call the model with your prompt and audio files to generate text.
+```python
+from transformers import Qwen2AudioForConditionalGeneration, AutoProcessor
+from outlines import models
+
+MODEL_NAME = "Qwen/Qwen2-Audio-7B-Instruct"
+URL = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/guess_age_gender.wav"
+
+tf_model = Qwen2AudioForConditionalGeneration.from_pretrained(MODEL_NAME)
+tf_processor = AutoProcessor.from_pretrained(MODEL_NAME)
+
+model = models.TransformersMultiModal(tf_model, tf_processor)
+
+text = """<|im_start|>user
+Audio 1: <|audio_bos|><|AUDIO|><|audio_eos|>
+<|im_end|>
+<|im_start|>user
+Write a sentence about what you hear in the audio.<|im_end|>
+<|im_start|>assistant
+"""
+prompt = {
+    "text": text,
+    "audios": audio_from_url(URL)
+}
+model(prompt)
+```
+
+You can include multiple audio files by providing a list for the `audios` argument and adding more of the appropriate tags to the prompt. Batching is also supported.
+
+## Use the model for structured generation
+
+You can use the model to generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
+
+Supported types include `Json`, `Choice`, `Regex` and `CFG`.
+
+For instance to do classification, you can use the `Regex` type:
+```python
+from outlines import models
+from outlines.types import Regex
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
+
+MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+
+model = models.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
+    LlavaProcessor.from_pretrained(MODEL_NAME)
+)
+pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
+prompt = {
+    "text": "<image>detailed description:",
+    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
+}
+model(prompt, Regex(pattern))
+```
+
+Another example could be to generated a structured description of an image using the `Json` type:
+```python
+from outlines import models
+from pydantic import BaseModel
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
+from typing import List, Optional
+
+MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+
+class ImageData(BaseModel):
+    caption: str
+    tags_list: List[str]
+    object_list: List[str]
+    is_photo: bool
+
+model = models.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
+    LlavaProcessor.from_pretrained(MODEL_NAME)
+)
+pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
+prompt = {
+    "text": "<image>detailed description:",
+    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
+}
+model(prompt, Json(ImageData))
+```
+
+## Resources
+
+### Choosing a model
+- https://mmbench.opencompass.org.cn/leaderboard
+- https://huggingface.co/spaces/WildVision/vision-arena
diff --git a/docs/reference/models/transformers_vision.md b/docs/reference/models/transformers_vision.md
deleted file mode 100644
index 03ba6a5c9..000000000
--- a/docs/reference/models/transformers_vision.md
+++ /dev/null
@@ -1,117 +0,0 @@
-# Transformers Vision
-
-Outlines allows seamless use of [vision models](https://huggingface.co/learn/computer-vision-course/en/unit4/multimodal-models/tasks-models-part1).
-
-`outlines.models.Transformers_vision` shares interfaces with, and is based on [outlines.models.Transformers](./transformers.md).
-
-## Create a `TransformersVision` model
-
-`TransformersVision` models inherit from `Transformers` and accept the same initialization parameters.
-
-In addition, they also accept the optional parameters `processor_class` and `processor_kwargs`. Those are used to create a `Processor` instance that is then used to preprocess the images. By default, `AutoProcessor` is used to create the processor as such: `AutoProcessor.from_pretrained(model_name, **processor_kwargs)`.
-
-If your model is not compatible with `AutoProcessor`, you must provide a value for the `processor_class` parameter.
-For instance:
-```python
-from outlines import models
-from transformers import CLIPModel, CLIPProcessor
-
-model = models.TransformersVision("openai/clip-vit-base-patch32", model_class=CLIPModel, processor_class=CLIPProcessor)
-```
-
-## Use the model to generate text from prompts and images
-
-When calling the model, the prompt argument you provide must be a dictionary with a key `"prompts"` and a key `"images"`. The associated values must be a string or a list of strings for the prompts, and a PIL image or a list of PIL images for the images. Your prompts must include `<image>` tags tokens to indicate where the image should be inserted. There must be as many `<image>` tags as there are images.
-
-For easier use, we recommend you to create a convenience function to load a `PIL.Image` from URL.
-```python
-from PIL import Image
-from io import BytesIO
-from urllib.request import urlopen
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-```
-
-You can then call the model with your prompts and images to generate text.
-```python
-from transformers import LlavaForConditionalGeneration
-from outlines import models
-
-model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
-prompt = {
-    "prompts": "<image> detailed description:",
-    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
-}
-model(prompt)
-```
-
-You can include several images per prompt by adding more `<image>` tags to the prompt. Batching is also supported.
-```python
-from transformers import LlavaForConditionalGeneration
-from outlines import models
-
-model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
-prompt = {
-    "prompts": ["<image><image>detailed description:", "<image><image>. What animals are present?"],
-    "images": [
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
-    ]
-}
-model(prompt)
-```
-
-Here we have two prompts, each expecting two images. We correspondingly provide four images. This will generate two descriptions, one for each prompt.
-
-### Use the model for structured generation
-
-You can use the model to generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
-
-Supported types include `Json`, `Choice`, `Regex` and `CFG`.
-
-For instance to do classification, you can use the `Regex` type:
-```python
-from outlines import models
-from outlines.types import Regex
-from transformers import LlavaForConditionalGeneration
-
-model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
-pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
-prompt = {
-    "prompts": "<image>detailed description:",
-    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
-}
-model(prompt, Regex(pattern))
-```
-
-Another example could be to generated a structured description of an image using the `Json` type:
-```python
-from outlines import models
-from pydantic import BaseModel
-from transformers import LlavaForConditionalGeneration
-from typing import List, Optional
-
-class ImageData(BaseModel):
-    caption: str
-    tags_list: List[str]
-    object_list: List[str]
-    is_photo: bool
-
-model = models.TransformersVision("trl-internal-testing/tiny-LlavaForConditionalGeneration", model_class=LlavaForConditionalGeneration)
-pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
-prompt = {
-    "prompts": "<image>detailed description:",
-    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
-}
-model(prompt, Json(ImageData))
-```
-
-## Resources
-
-### Choosing a model
-- https://mmbench.opencompass.org.cn/leaderboard
-- https://huggingface.co/spaces/WildVision/vision-arena
diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
index b677040d5..0ffbda6a3 100644
--- a/outlines/generate/cfg.py
+++ b/outlines/generate/cfg.py
@@ -4,7 +4,7 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import LlamaCpp, OpenAI, TransformersVision
+from outlines.models import LlamaCpp, OpenAI, TransformersMultiModal
 from outlines.samplers import Sampler, multinomial
 
 
@@ -33,7 +33,7 @@ def cfg(
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
-@cfg.register(TransformersVision)
+@cfg.register(TransformersMultiModal)
 def cfg_vision(model, cfg_str: str, sampler: Sampler = multinomial()):
     from outlines.processors import CFGLogitsProcessor
 
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
index 1950812d2..81f974bd2 100644
--- a/outlines/generate/fsm.py
+++ b/outlines/generate/fsm.py
@@ -7,7 +7,7 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import TransformersVision
+from outlines.models import TransformersMultiModal
 from outlines.samplers import Sampler, multinomial
 
 
@@ -22,7 +22,7 @@ def fsm(
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
-@fsm.register(TransformersVision)
+@fsm.register(TransformersMultiModal)
 def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()):
     from outlines.processors import GuideLogitsProcessor
 
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
index 326701f4b..477a7bf5e 100644
--- a/outlines/generate/regex.py
+++ b/outlines/generate/regex.py
@@ -4,7 +4,7 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import OpenAI, TransformersVision
+from outlines.models import OpenAI, TransformersMultiModal
 from outlines.samplers import Sampler, multinomial
 from outlines.types import Regex
 
@@ -39,7 +39,7 @@ def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
     return SequenceGeneratorAdapter(model, logits_processor, sampler)
 
 
-@regex.register(TransformersVision)
+@regex.register(TransformersMultiModal)
 def regex_vision(
     model,
     regex_str: str | Regex,
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
index 32530d0c4..e2e4cbe5d 100644
--- a/outlines/generate/text.py
+++ b/outlines/generate/text.py
@@ -4,7 +4,7 @@
     SequenceGeneratorAdapter,
     VisionSequenceGeneratorAdapter,
 )
-from outlines.models import OpenAI, TransformersVision
+from outlines.models import OpenAI, TransformersMultiModal
 from outlines.samplers import Sampler, multinomial
 
 
@@ -34,7 +34,7 @@ def text(model, sampler: Sampler = multinomial()) -> SequenceGeneratorAdapter:
     return SequenceGeneratorAdapter(model, None, sampler)
 
 
-@text.register(TransformersVision)
+@text.register(TransformersMultiModal)
 def text_vision(model, sampler: Sampler = multinomial()):
     return VisionSequenceGeneratorAdapter(model, None, sampler)
 
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 6a9ba753f..fb982d3b2 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -20,7 +20,7 @@
 from .transformers import (
     Transformers,
     TransformerTokenizer,
-    TransformersVision,
+    TransformersMultiModal,
     from_transformers,
 )
 from .vllm import VLLM, from_vllm
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index b65d0cefb..ceaf39b7d 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -6,7 +6,7 @@
 
 if TYPE_CHECKING:
     import torch
-    from transformers import PreTrainedTokenizer, PreTrainedModel
+    from transformers import PreTrainedTokenizer, PreTrainedModel, ProcessorMixin
 
 
 __all__ = ["Transformers", "from_transformers"]
@@ -256,12 +256,12 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
             )
 
 
-class TransformersVisionTypeAdapter(ModelTypeAdapter):
-    """Type adapter for TransformersVision models."""
+class TransformersMultiModalTypeAdapter(ModelTypeAdapter):
+    """Type adapter for TransformersMultiModal models."""
 
     @singledispatchmethod
     def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
+        """Generate the prompt arguments to pass to the model.
 
         Argument
         --------
@@ -272,18 +272,27 @@ def format_input(self, model_input):
         raise NotImplementedError(
             f"The input type {input} is not available. Please provide a "
             "dictionary with the following format: "
-            "{'prompts': Union[str, List[str]], 'images': Union[Any, List[Any]]}"
-            "Make sure the number of image tags in the prompts is equal to the "
-            "number of images provided."
+            "{'text': Union[str, List[str]], 'images': Optional[Union[Any, "
+            "List[Any]]], 'audios': Optional[Union[Any, List[Any]]]}"
+            "Either 'images' or 'audios' must be provided."
+            "Make sure the text is formatted correctly for the model "
+            "(e.g. include <image> or <|AUDIO|> tags)."
         )
 
     @format_input.register(dict)
     def format_list_input(self, model_input):
-        if "prompts" not in model_input or "images" not in model_input:
+        if (
+            "text" not in model_input
+            or (
+                "images" not in model_input
+                and "audios" not in model_input
+            )
+        ):
             raise ValueError(
-                "The input must contain the following keys: 'prompts' and 'images'."
+                "The input must contain the 'text' key along with either "
+                "'images' or 'audios'."
             )
-        return model_input["prompts"], model_input["images"]
+        return model_input
 
     def format_output_type(self, output_type):
         """Generate the logits processor argument to pass to the model.
@@ -301,11 +310,11 @@ def format_output_type(self, output_type):
         return None
 
 
-class TransformersVision(Transformers):
+class TransformersMultiModal(Transformers):
     """Represents a `transformers` model with a vision processor."""
 
     def __init__(self, model: "PreTrainedModel", processor):
-        """Create a TransformersVision model instance
+        """Create a TransformersMultiModal model instance
 
         We rely on the `__init__` method of the `Transformers` class to handle
         most of the initialization and then add elements specific to vision
@@ -320,47 +329,25 @@ def __init__(self, model: "PreTrainedModel", processor):
 
         super().__init__(model, tokenizer)
 
-        self.type_adapter = TransformersVisionTypeAdapter()
+        self.type_adapter = TransformersMultiModalTypeAdapter()
 
     def _prepare_model_inputs(self, model_input, output_type):
-        prompts, images = self.type_adapter.format_input(model_input)
+        model_input = self.type_adapter.format_input(model_input)
         inputs = self.processor(
-            text=prompts, images=images, padding=True, return_tensors="pt"
+            **model_input, padding=True, return_tensors="pt"
         ).to(self.model.device)
 
-        return prompts, inputs
+        return model_input["text"], inputs
 
 
 def from_transformers(
     model: "PreTrainedModel",
-    tokenizer_or_processor: "PreTrainedTokenizer",
+    tokenizer_or_processor: Union["PreTrainedTokenizer", "ProcessorMixin"],
 ):
     from transformers import (
         PreTrainedTokenizer,
         PreTrainedTokenizerFast,
-        Blip2Processor,
-        LlavaProcessor,
-        IdeficsProcessor,
-        CLIPProcessor,
-        Qwen2_5_VLProcessor,
-        Qwen2VLProcessor,
-        NougatProcessor,
-        LlavaNextProcessor,
-        PixtralProcessor,
-        PaliGemmaProcessor,
-    )
-
-    vision_processors = (
-        Blip2Processor,
-        LlavaProcessor,
-        IdeficsProcessor,
-        CLIPProcessor,
-        Qwen2_5_VLProcessor,
-        Qwen2VLProcessor,
-        NougatProcessor,
-        LlavaNextProcessor,
-        PixtralProcessor,
-        PaliGemmaProcessor,
+        ProcessorMixin
     )
 
     if isinstance(
@@ -368,13 +355,12 @@ def from_transformers(
     ):
         tokenizer = tokenizer_or_processor
         return Transformers(model, tokenizer)
-    elif isinstance(tokenizer_or_processor, vision_processors):
+    elif isinstance(tokenizer_or_processor, ProcessorMixin):
         processor = tokenizer_or_processor
-        return TransformersVision(model, processor)
+        return TransformersMultiModal(model, processor)
     else:
         raise ValueError(
             "We could determine whether the model passed to `from_transformers`"
-            + " is a text-2-text of vision language model. If you passed a"
-            + " vision language model please open an issue pasting the error"
-            + " with the name of the vision language model."
+            + " is a text-2-text or a multi-modal model. Please provide a "
+            + "a transformers tokenizer or processor."
         )
diff --git a/outlines/models/transformers_vision.py b/outlines/models/transformers_vision.py
deleted file mode 100644
index e69de29bb..000000000
diff --git a/tests/models/test_transformers_vision.py b/tests/models/test_transformers_multimodal.py
similarity index 81%
rename from tests/models/test_transformers_vision.py
rename to tests/models/test_transformers_multimodal.py
index 57dedcb69..9292bdf6a 100644
--- a/tests/models/test_transformers_vision.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -1,3 +1,5 @@
+# we only test vision models here as audio models are too heavy to run on CI
+
 import re
 from enum import Enum
 from io import BytesIO
@@ -15,7 +17,7 @@
 )
 
 import outlines
-from outlines.models.transformers import TransformersVision
+from outlines.models.transformers import TransformersMultiModal
 from outlines.types import Choice, JsonType, Regex
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
@@ -48,12 +50,12 @@ def test_transformers_vision_instantiate_simple():
         Blip2ForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
-    assert isinstance(model, TransformersVision)
+    assert isinstance(model, TransformersMultiModal)
 
 
 def test_transformers_vision_simple(model, images):
     result = model.generate(
-        {"prompts": "<image>Describe this image in one sentence:", "images": images[0]},
+        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
         None,
     )
     assert isinstance(result, str)
@@ -61,7 +63,7 @@ def test_transformers_vision_simple(model, images):
 
 def test_transformers_vision_call(model, images):
     result = model(
-        {"prompts": "<image>Describe this image in one sentence:", "images": images[0]}
+        {"text": "<image>Describe this image in one sentence:", "images": images[0]}
     )
     assert isinstance(result, str)
 
@@ -70,7 +72,7 @@ def test_transformers_vision_wrong_number_images(model, images):
     with pytest.raises(ValueError):
         a = model(
             {
-                "prompts": "<image>Describe this image in one sentence:",
+                "text": "<image>Describe this image in one sentence:",
                 "images": [images[0], images[1]],
             }
         )
@@ -84,7 +86,7 @@ def test_transformers_vision_wrong_input_type(model):
 
 def test_transformers_inference_kwargs(model, images):
     result = model(
-        {"prompts": "<image>Describe this image in one sentence:", "images": images[0]},
+        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
         max_new_tokens=100,
     )
     assert isinstance(result, str)
@@ -94,7 +96,7 @@ def test_transformers_invalid_inference_kwargs(model, images):
     with pytest.raises(ValueError):
         model(
             {
-                "prompts": "<image>Describe this image in one sentence:",
+                "text": "<image>Describe this image in one sentence:",
                 "images": images[0],
             },
             foo="bar",
@@ -104,7 +106,7 @@ def test_transformers_invalid_inference_kwargs(model, images):
 def test_transformers_several_images(model, images):
     result = model(
         {
-            "prompts": "<image><image>Describe this image in one sentence:",
+            "text": "<image><image>Describe this image in one sentence:",
             "images": [images[0], images[1]],
         }
     )
@@ -116,7 +118,7 @@ class Foo(BaseModel):
         name: str
 
     result = model(
-        {"prompts": "<image>Give a name to this animal.", "images": images[0]},
+        {"text": "<image>Give a name to this animal.", "images": images[0]},
         JsonType(Foo),
     )
     assert "name" in result
@@ -124,7 +126,7 @@ class Foo(BaseModel):
 
 def test_transformers_vision_regex(model, images):
     result = model(
-        {"prompts": "<image>How old is it?", "images": images[0]}, Regex(r"[0-9]")
+        {"text": "<image>How old is it?", "images": images[0]}, Regex(r"[0-9]")
     )
 
     assert isinstance(result, str)
@@ -137,7 +139,7 @@ class Foo(Enum):
         dog = "dog"
 
     result = model(
-        {"prompts": "<image>Is it a cat or a dog?", "images": images[0]}, Choice(Foo)
+        {"text": "<image>Is it a cat or a dog?", "images": images[0]}, Choice(Foo)
     )
 
     assert isinstance(result, str)
@@ -146,7 +148,7 @@ class Foo(Enum):
 
 def test_transformers_vision_batch_samples(model, images):
     result = model(
-        {"prompts": "<image>Describe this image in one sentence.", "images": images[0]},
+        {"text": "<image>Describe this image in one sentence.", "images": images[0]},
         num_return_sequences=2,
         num_beams=2,
     )
@@ -154,7 +156,7 @@ def test_transformers_vision_batch_samples(model, images):
     assert len(result) == 2
     result = model(
         {
-            "prompts": [
+            "text": [
                 "<image>Describe this image in one sentence.",
                 "<image>Describe this image in one sentence.",
             ],
@@ -165,7 +167,7 @@ def test_transformers_vision_batch_samples(model, images):
     assert len(result) == 2
     result = model(
         {
-            "prompts": [
+            "text": [
                 "<image>Describe this image in one sentence.",
                 "<image>Describe this image in one sentence.",
             ],
@@ -187,7 +189,7 @@ class Foo(Enum):
         dog = "dog"
 
     result = model(
-        {"prompts": "<image>Describe this image in one sentence.", "images": images[0]},
+        {"text": "<image>Describe this image in one sentence.", "images": images[0]},
         Choice(Foo),
         num_return_sequences=2,
         num_beams=2,
@@ -198,7 +200,7 @@ class Foo(Enum):
         assert item in ["cat", "dog"]
     result = model(
         {
-            "prompts": [
+            "text": [
                 "<image>Describe this image in one sentence.",
                 "<image>Describe this image in one sentence.",
             ],
@@ -212,7 +214,7 @@ class Foo(Enum):
         assert item in ["cat", "dog"]
     result = model(
         {
-            "prompts": [
+            "text": [
                 "<image>Describe this image in one sentence.",
                 "<image>Describe this image in one sentence.",
             ],

From c53651bc5ed10802bf6aa0f878e209b71436f767 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Thu, 27 Feb 2025 15:10:28 +0100
Subject: [PATCH 247/505] Refactor the VLLM model (#1453)

Modify the `VLLM` class to inherit from the base `Model` class and
implement the standard interface of models for the v1.0.
This PR also creates tests for the `VLLM` model in the `tests/models`
directory and remove the previous tests for it in `tests/generate`.
---
 docs/reference/models/vllm.md           |  31 ++-
 outlines/models/vllm.py                 | 159 +++++++--------
 tests/generate/test_generate.py         |   7 +-
 tests/generate/test_integration_vllm.py | 245 ------------------------
 tests/models/test_vllm.py               | 135 +++++++++++++
 5 files changed, 227 insertions(+), 350 deletions(-)
 delete mode 100644 tests/generate/test_integration_vllm.py
 create mode 100644 tests/models/test_vllm.py

diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index 1636410a4..7698cd31e 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -11,10 +11,10 @@
 
 Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
 
-
 ```python
 import outlines
 from vllm import LLM
+
 model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
 ```
 
@@ -28,18 +28,37 @@ Models are loaded from the [HuggingFace hub](https://huggingface.co/).
 
 ## Generate text
 
-In addition to the parameters described in the [text generation section](../text.md) you can pass an instance of `SamplingParams` directly to any generator via the `sampling_params` keyword argument:
+To generate text, you can just call the model with a prompt as argument:
 
 ```python
-from vllm.sampling_params import SamplingParams
-from outlines import models, generate
+import outlines
+from vllm import LLM
 
+model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
+answer = model("Write a short story about a cat.")
+```
+
+You can also use structured generation with the `VLLM` model by providing an output type after the prompt:
+
+```python
+import outlines
+from vllm import LLM
+from outlines.types import JsonType
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
 
 model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
-params = SamplingParams(n=2, frequency_penalty=1., min_tokens=2)
-answer = model("A prompt", sampling_params=params)
+answer = model("Create a character.", output_type=JsonType(Character))
 ```
 
+The VLLM model supports batch generation. To use it, you can pass a list of strings as prompt instead of a single string.
+
+## Optional parameters
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `LLM.generate` method of the `vllm` library. An optional parameter of particular interest is `sampling_params`, which is an instance of `SamplingParams`. You can find more information about it in the [vLLM documentation][https://docs.vllm.ai/en/latest/api/inference_params.html].
+
 !!! Warning
 
     Streaming is not available for the offline vLLM integration.
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 8c709a5e2..3f10f6433 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,7 +1,9 @@
 import dataclasses
+from functools import singledispatchmethod
 from typing import TYPE_CHECKING, List, Optional, Union
 
 from outlines.generate.api import GenerationParameters, SamplingParameters
+from outlines.models.base import Model, ModelTypeAdapter
 
 if TYPE_CHECKING:
     from transformers import PreTrainedTokenizerBase
@@ -12,21 +14,59 @@
 __all__ = ["VLLM", "from_vllm"]
 
 
-class VLLM:
-    """Represents a vLLM model.
+class VLLMTypeAdapter(ModelTypeAdapter):
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
 
-    We wrap models from model providing libraries in order to give all of
-    them the same interface in Outlines and allow users to easily switch
-    between providers. This class wraps the `vllm.LLM` class from the
-    `vllm` library.
+        Argument
+        --------
+        model_input
+            The input passed by the user.
 
-    """
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available."
+            "Please use a string or a list of strings."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input):
+        return model_input
+
+    @format_input.register(list)
+    def format_list_input(self, model_input):
+        return model_input
+
+    def format_output_type(self, output_type):
+        """Generate the logits processor argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The logits processor provided.
+
+        """
+        if output_type is not None:
+            return [output_type]
+        return []
+
+
+class VLLM(Model):
+    """Represents a `vllm` model."""
 
     def __init__(self, model: "LLM"):
-        self.model = model
-        self.lora_request = None
+        """Create a VLLM model instance.
+
+        Parameters
+        ----------
+        model
+            A `vllm.LLM` model instance.
 
+        """
+        self.model = model
         self.tokenizer = self._get_tokenizer()
+        self.type_adapter = VLLMTypeAdapter()
 
     def _get_tokenizer(self):
         if hasattr(self.model, "get_tokenizer"):
@@ -43,98 +83,35 @@ def _get_tokenizer(self):
             )
         return adapt_tokenizer(tokenizer=tokenizer)
 
-    def generate(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        logits_processor,
-        sampling_parameters: SamplingParameters,
-        *,
-        sampling_params: Optional["SamplingParams"] = None,
-        use_tqdm: bool = True,
-    ):
-        """Generate text using vLLM.
+    def generate(self, model_input, output_type, **inference_kwargs):
+        """Generate text using `vllm`.
 
-        Parameters
-        ----------
-        prompts
-            A prompt or list of prompts.
-        generation_parameters
-            An instance of `GenerationParameters` that contains the prompt,
-            the maximum number of tokens, stop sequences and seed. All the
-            arguments to `SequenceGeneratorAdapter`'s `__cal__` method.
+        Arguments
+        ---------
+        prompt
+            The text prompt provided by the user.
         logits_processor
             The logits processor to use when generating text.
-        sampling_parameters
-            An instance of `SamplingParameters`, a dataclass that contains
-            the name of the sampler to use and related parameters as available
-            in Outlines.
-        sampling_params
-            An instance of `vllm.sampling_params.SamplingParams`. The values
-            passed via this dataclass supersede the values of the parameters
-            in `generation_parameters` and `sampling_parameters`. See the
-            vLLM documentation for more details: https://docs.vllm.ai/en/latest/dev/sampling_params.html.
-        use_tqdm
-            A boolean in order to display progress bar while inferencing
+        inference_kwargs
+            The inference kwargs that can be passed to the `LLM.generate` method
+            in the `vllm` library.
 
         Returns
         -------
-        The generated text, of shape `(n_batch, n_samples)`. If there are only
-        one batch and several samples, the list is of shape `(n_samples)`. If
-        this is a batch with several sequences but only one sample the list is
-        of shape `(n_batch)`. If there is only one sequence and one sample, a
-        string is returned.
+        The generated text.
 
         """
         from vllm.sampling_params import SamplingParams
 
+        sampling_params = inference_kwargs.pop("sampling_params", None)
         if sampling_params is None:
             sampling_params = SamplingParams()
-
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-
-        # We only update the values in `sampling_params` if they
-        # are specified by the user when calling the generator.
-        if max_tokens is not None:
-            sampling_params.max_tokens = max_tokens
-        if stop_at is not None:
-            if isinstance(stop_at, str):
-                stop_at = [stop_at]
-            sampling_params.stop = stop_at
-        if seed is not None:
-            sampling_params.seed = seed
-
-        sampling_params.logits_processors = (
-            [logits_processor] if logits_processor is not None else []
-        )
-
-        sampler, num_samples, top_p, top_k, temperature = dataclasses.astuple(
-            sampling_parameters
-        )
-
-        # We only update the values in `sampling_params` that
-        # were not specified by the user.
-        if sampling_params.n == 1:
-            sampling_params.n = num_samples
-            sampling_params.best_of = num_samples
-        if top_p is not None and sampling_params.top_p == 1.0:
-            sampling_params.top_p = top_p
-        if top_k is not None and sampling_params.top_k == -1:
-            sampling_params.top_k = top_k
-            # TODO: remove this if statement once fixed
-            # https://github.com/vllm-project/vllm/issues/5404#issuecomment-2175972897
-            if top_k == 1:
-                sampling_params.repetition_penalty = 0
-        if temperature is not None and sampling_params.temperature == 1.0:
-            sampling_params.temperature = temperature
-        if sampler == "beam_search":
-            sampling_params.use_beam_search = True
+        sampling_params.logits_processors = self.type_adapter.format_output_type(output_type)
 
         results = self.model.generate(
-            prompts,
+            self.type_adapter.format_input(model_input),
             sampling_params=sampling_params,
-            lora_request=self.lora_request,
-            use_tqdm=use_tqdm,
+            **inference_kwargs,
         )
         results = [[sample.text for sample in batch.outputs] for batch in results]
 
@@ -150,7 +127,7 @@ def generate(
 
         return results
 
-    def stream(self, *args, **kwargs):
+    def stream(self, model_input, output_type, **inference_kwargs):
         """Return a text generator.
 
         Streaming is not yet available for `vllm.LLM`.
@@ -162,14 +139,6 @@ def stream(self, *args, **kwargs):
             "Streaming is not available for the vLLM integration."
         )
 
-    def load_lora(self, adapter_path: Optional[str]):
-        from vllm.lora.request import LoRARequest
-
-        if adapter_path is None:
-            self.lora_request = None
-        else:
-            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
-
 
 def from_vllm(model: "LLM") -> VLLM:
     return VLLM(model)
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 1bf3f2a41..84c8b761c 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -143,17 +143,16 @@ def model_t5(tmp_path_factory):
 ALL_MODEL_FIXTURES = (
     # "model_llamacpp",  # temporary disabled due to the v1 model refactoring
     "model_exllamav2",
-    "model_mlxlm",
-    "model_mlxlm_phi3",
+    # "model_mlxlm",
+    # "model_mlxlm_phi3",
     # "model_transformers_random",
     # "model_transformers_opt125m",
     # "model_mamba",
     # "model_bart",
     # "model_transformers_vision",
-    "model_vllm",
+    # "model_vllm",
 )
 
-
 class MyEnum(Enum):
     foo = "foo"
     bar = "bar"
diff --git a/tests/generate/test_integration_vllm.py b/tests/generate/test_integration_vllm.py
deleted file mode 100644
index 7b9acc240..000000000
--- a/tests/generate/test_integration_vllm.py
+++ /dev/null
@@ -1,245 +0,0 @@
-import datetime
-import re
-
-import pytest
-import torch
-from pydantic import BaseModel, constr
-
-import outlines
-import outlines.generate as generate
-import outlines.grammars as grammars
-import outlines.models as models
-import outlines.samplers as samplers
-
-try:
-    from vllm import LLM
-    from vllm.sampling_params import SamplingParams
-except ImportError:
-    pass
-
-pytestmark = pytest.mark.skipif(
-    not torch.cuda.is_available(), reason="vLLM models can only be run on GPU."
-)
-
-
-@pytest.fixture(scope="module")
-def model():
-    return outlines.from_vllm(LLM("gpt2", gpu_memory_utilization=0.5))
-
-
-@pytest.mark.parametrize(
-    "generator_type,params", ((generate.text, []), (generate.regex, ("[0-9]",)))
-)
-def test_vllm_generation_api(model, generator_type, params):
-    generator = generator_type(model, *params)
-
-    res = generator("test")
-    assert isinstance(res, str)
-
-    res = generator("test", max_tokens=10)
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=".")
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res = generator("test", stop_at=[".", "ab"])
-    assert isinstance(res, str)
-
-    res1 = generator("test", seed=1)
-    res2 = generator("test", seed=1)
-    assert isinstance(res1, str)
-    assert isinstance(res2, str)
-    assert res1 == res2
-
-    res = generator(["test", "test1"])
-    assert len(res) == 2
-
-
-def test_vllm_sampling_params(model):
-    generator = generate.text(model)
-
-    sampling_params = SamplingParams(n=2)
-    res = generator("test", sampling_params=sampling_params)
-    assert len(res) == 2
-    assert isinstance(res[0], str)
-    assert isinstance(res[1], str)
-
-    sampling_params = SamplingParams(seed=2)
-    res1 = generator("test", sampling_params=sampling_params)
-    res2 = generator("test", sampling_params=sampling_params)
-    assert res1 == res2
-
-
-def test_vllm_greedy_sampling(model):
-    sampler = samplers.greedy()
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert isinstance(res, str)
-
-
-def test_vllm_multinomial_sampling(model):
-    sampler = samplers.multinomial()
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(3)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert len(res) == 3
-    assert isinstance(res[0], str)
-    assert isinstance(res[1], str)
-
-    sampler = samplers.multinomial(2, top_k=1)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert res[0] == res[1]
-
-    sampler = samplers.multinomial(1, top_p=0.5)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert isinstance(res, str)
-
-    sampler = samplers.multinomial(2, temperature=0.00001)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert res[0] == res[1]
-
-
-def test_vllm_beam_search(model):
-    sampler = samplers.beam_search(1)
-    generator = generate.text(model, sampler)
-    res1 = generator("test")
-    sampler = samplers.greedy()
-    generator = generate.text(model, sampler)
-    res2 = generator("test")
-    assert res1 == res2
-
-    sampler = samplers.beam_search(2)
-    generator = generate.text(model, sampler)
-    res = generator("test")
-    assert len(res) == 2
-    assert res[0] != res[1]
-
-
-def test_vllm_text_stop(model):
-    prompt = "Write a short sentence containing 'You': "
-    sequence = generate.text(model)(prompt, max_tokens=100, seed=10)
-    assert sequence.find("news") != -1
-
-    sequence = generate.text(model)(prompt, stop_at="news", max_tokens=100, seed=10)
-    assert isinstance(sequence, str)
-    assert sequence.find("news") == -1
-
-
-def test_vllm_regex(model):
-    prompt = "Write an email address: "
-    regex_str = r"([a-z]{10})@([a-z]{5})\.([a-z]{3})"
-    generator = generate.regex(model, regex_str)
-
-    # One prompt
-    sequence = generator(prompts=prompt)
-    assert isinstance(sequence, str)
-    assert re.fullmatch(pattern=regex_str, string=sequence) is not None
-
-
-def test_vllm_integer(model):
-    prompt = "Give me an integer: "
-    sequence = generate.format(model, int)(prompt, max_tokens=10)
-    assert isinstance(sequence, int)
-    assert sequence != ""
-    int(sequence)
-
-
-def test_vllm_float(model):
-    prompt = "Give me a floating-point number: "
-    sequence = generate.format(model, float)(prompt, max_tokens=10)
-    assert isinstance(sequence, float)
-
-    assert sequence != ""
-    float(sequence)
-
-
-def test_vllm_bool(model):
-    prompt = "Is this True or False? "
-    sequence = generate.format(model, bool)(prompt, max_tokens=10)
-    assert isinstance(sequence, bool)
-
-    assert sequence != ""
-    bool(sequence)
-
-
-def test_vllm_date(model):
-    prompt = "What day is it today? "
-    sequence = generate.format(model, datetime.date)(prompt, max_tokens=10)
-    assert isinstance(sequence, datetime.date)
-
-
-def test_vllm_time(model):
-    prompt = "What time is it? "
-    sequence = generate.format(model, datetime.time)(prompt, max_tokens=10)
-    assert isinstance(sequence, datetime.time)
-
-
-def test_vllm_datetime(model):
-    prompt = "What time is it? "
-    sequence = generate.format(model, datetime.datetime)(prompt, max_tokens=20)
-    assert isinstance(sequence, datetime.datetime)
-
-
-def test_vllm_choice(model):
-    prompt = "Which one between 'test' and 'choice'? "
-    sequence = generate.choice(model, ["test", "choice"])(prompt)
-    assert sequence == "test" or sequence == "choice"
-
-
-def test_vllm_json_basic(model):
-    prompt = "Output some JSON. "
-
-    class Spam(BaseModel):
-        spam: constr(max_length=10)
-        fuzz: bool
-
-    sampling_params = SamplingParams(temperature=0)
-    result = generate.json(model, Spam, whitespace_pattern="")(
-        prompt, max_tokens=100, seed=1, sampling_params=sampling_params
-    )
-    assert isinstance(result, BaseModel)
-    assert isinstance(result.spam, str)
-    assert isinstance(result.fuzz, bool)
-    assert len(result.spam) <= 10
-
-
-def test_vllm_json_schema(model):
-    prompt = "Output some JSON. "
-
-    schema = """{
-      "title": "spam",
-      "type": "object",
-      "properties": {
-           "foo" : {"type": "boolean"},
-           "bar": {"type": "string", "maxLength": 4}
-        },
-      "required": ["foo", "bar"]
-      }
-    """
-
-    sampling_params = SamplingParams(temperature=0)
-    result = generate.json(model, schema, whitespace_pattern="")(
-        prompt, max_tokens=100, seed=10, sampling_params=sampling_params
-    )
-    assert isinstance(result, dict)
-    assert isinstance(result["foo"], bool)
-    assert isinstance(result["bar"], str)
-
-
-@pytest.mark.xfail(
-    reason="The CFG logits processor for vLLM has not been implemented yet."
-)
-def test_vllm_cfg(model):
-    prompt = "<|im_start|>user\nOutput a short and valid JSON object with two keys.<|im_end|>\n><|im_start|>assistant\n"
-    result = generate.cfg(model, grammars.arithmetic)(prompt, seed=11)
-    assert isinstance(result, str)
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
new file mode 100644
index 000000000..5b5aa96ec
--- /dev/null
+++ b/tests/models/test_vllm.py
@@ -0,0 +1,135 @@
+import re
+from enum import Enum
+
+import pytest
+import torch
+from pydantic import BaseModel
+
+try:
+    from vllm import LLM, SamplingParams
+except ImportError:
+    pass
+
+import outlines
+from outlines.models.vllm import VLLM
+from outlines.types import Choice, JsonType, Regex
+
+
+TEST_MODEL = "erwanf/gpt2-mini"
+
+pytestmark = pytest.mark.skipif(
+    not torch.cuda.is_available(),
+    reason="vLLM models can only be run on GPU."
+)
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    model = outlines.from_vllm(LLM(TEST_MODEL))
+    return model
+
+
+def test_vllm_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_vllm_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+def test_vllm_inference_kwargs(model):
+    result = model("Write a short story about a cat.", sampling_params=SamplingParams(max_tokens=2), use_tqdm=True)
+    assert isinstance(result, str)
+    assert len(result) <= 20
+
+
+def test_vllm_invalid_inference_kwargs(model):
+    with pytest.raises(TypeError):
+        model("Respond with one word. Not more.", foo="bar")
+
+
+def test_vllm_regex(model):
+    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
+
+
+def test_vllm_json(model):
+    class Character(BaseModel):
+        name: str
+
+    result = model("Create a character with a name.", JsonType(Character))
+    assert "name" in result
+
+
+def test_vllm_choice(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Choice(Foo))
+    assert result in ["cat", "dog"]
+
+
+def test_vllm_batch_samples(model):
+    result = model(
+        "Respond with one word. Not more.",
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."]
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."],
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+
+
+def test_vllm_batch_samples_constrained(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model(
+        "Cat or dog?",
+        Choice(Foo),
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Choice(Foo),
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Choice(Foo),
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+        assert item[0] in ["cat", "dog"]
+        assert item[1] in ["cat", "dog"]

From 99af914833ad07905aaa8b10f6a51ad9a0685451 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Thu, 27 Feb 2025 15:41:02 +0100
Subject: [PATCH 248/505] Remove the model Exllamav2

---
 README.md                                    |   2 +-
 docs/installation.md                         |   5 +-
 docs/reference/models/exllamav2.md           |  15 -
 docs/reference/models/models.md              |  23 +-
 docs/welcome.md                              |   2 +-
 mkdocs.yml                                   |   1 -
 outlines/models/__init__.py                  |   3 +-
 outlines/models/exllamav2.py                 | 359 ------------------
 pyproject.toml                               |   2 -
 tests/generate/conftest.py                   |   1 -
 tests/generate/test_generate.py              |  27 --
 tests/generate/test_integration_exllamav2.py | 363 -------------------
 uv.lock                                      |  28 --
 13 files changed, 16 insertions(+), 815 deletions(-)
 delete mode 100644 docs/reference/models/exllamav2.md
 delete mode 100644 outlines/models/exllamav2.py
 delete mode 100644 tests/generate/test_integration_exllamav2.py

diff --git a/README.md b/README.md
index e8f4fdbce..4e9448249 100644
--- a/README.md
+++ b/README.md
@@ -29,7 +29,7 @@ Need a high-performance commercial solution for structured outputs? Email us at
 
 ## Features
 
-- 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, exllama2, mamba
+- 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, mamba
 - 🔥 Fast [JSON generation](#efficient-json-generation-following-a-pydantic-model) following a JSON schema or a Pydantic model
 - 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
 - 📝 Generate text that follows a [regex](#efficient-regex-structured-generation) or a [context-free grammar](#using-context-free-grammars-to-guide-generation)
diff --git a/docs/installation.md b/docs/installation.md
index f88bc7125..04ecd8237 100644
--- a/docs/installation.md
+++ b/docs/installation.md
@@ -10,13 +10,13 @@ You can install Outlines with `pip`:
 pip install outlines
 ```
 
-Outlines supports OpenAI, Transformers, Mamba, llama.cpp, and ExLlamaV2, but **you will need to install them manually**:
+Outlines supports OpenAI, Transformers, Mamba and llama.cpp, but **you will need to install them manually**:
 
 ```sh
 pip install openai
 pip install transformers datasets accelerate torch
 pip install llama-cpp-python
-pip install exllamav2 transformers torch
+pip install transformers torch
 pip install mamba_ssm transformers torch
 pip install vllm
 ```
@@ -32,7 +32,6 @@ Outlines provides multiple optional dependency sets to support different backend
 - `pip install "outlines[mlx]"` for [MLX-LM](https://github.com/ml-explore/mlx-lm), optimized for Apple silicon.
 - `pip install "outlines[openai]"` to use OpenAI’s API.
 - `pip install "outlines[llamacpp]"` for [llama.cpp](https://github.com/ggerganov/llama.cpp), a lightweight LLM inference engine.
-- `pip install "outlines[exllamav2]"` for [ExLlamaV2](https://github.com/turboderp/exllamav2), optimized for NVIDIA GPUs.
 
 ## Bleeding Edge
 
diff --git a/docs/reference/models/exllamav2.md b/docs/reference/models/exllamav2.md
deleted file mode 100644
index a4e727840..000000000
--- a/docs/reference/models/exllamav2.md
+++ /dev/null
@@ -1,15 +0,0 @@
-# ExllamaV2
-
-The `outlines.models.exllamav2` model requires a Logits Processor component for compatibility with Outlines structured generation. While ExLlamaV2 doesn't natively support this feature, a third-party fork provides the necessary functionality. You can install it with the following command:
-
-```bash
-pip install git+https://github.com/lapp0/exllamav2@sampler-logits-processor
-```
-
-Install other requirements:
-
-```bash
-pip install transformers torch
-```
-
-*Coming soon*
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
index 5d85c67f8..f0fbc20ad 100644
--- a/docs/reference/models/models.md
+++ b/docs/reference/models/models.md
@@ -16,7 +16,6 @@ Open source models offer more flexibility for structured generation as we have c
 - [llama-cpp-python](llamacpp.md) - Python bindings for llama.cpp.
 - [mlx-lm](mlxlm.md) - Run open-source models on Metal hardware.
 - [vllm](vlm.md) - Run open-source models on the vLLM engine.
-- [exllamaV2](exllamav2.md)
 
 Ollama only supports Json Schema:
 
@@ -36,17 +35,17 @@ OpenAI has recently integrated [structured outputs][structured-outputs] in its A
 Integrations differ in their coverage of structured generation. Here is a summary:
 
 
-|                   | [Transformers](transformers.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md) | [Gemini](gemini.md)
-|-------------------|--------------|------|-----------|-----------|-------|---------|-------|
-| **Supported HW**  |              |      |           |           |       |         |       |
-| CUDA              | ✅           | ✅  | ✅        | ✅        | ❌    | N/A     | N/A   |
-| Apple Silicon     | ✅           | ❌  | ✅        | ✅        | ✅    | N/A     | N/A   |
-| x86 / AMD64       | ✅           | ❌  | ✅        | ✅        | ❌    | N/A     | N/A   |
-| **Structure**     |               |     |           |           |       |         |       |
-| JSON Schema       | ✅           | ✅  | ✅        | ✅        | ✅    | ✅      | ✅   |
-| Choice            | ✅           | ✅  | ✅        | ✅        | ✅    | ❌      | ✅   |
-| Regex             | ✅           | ✅  | ✅        | ✅        | ✅    | ❌      | ❌   |
-| Grammar           | ✅           | ✅  | ✅        | ✅        | ✅    | ❌      | ❌   |
+|                   | [Transformers](transformers.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md) | [Gemini](gemini.md) |
+|-------------------|--------------|------|-----------|-------|---------|-------|
+| **Supported HW**  |              |      |           |       |         |       |
+| CUDA              | ✅           | ✅  | ✅        | ❌    | N/A     | N/A   |
+| Apple Silicon     | ✅           | ❌  | ✅        | ✅    | N/A     | N/A   |
+| x86 / AMD64       | ✅           | ❌  | ✅        | ❌    | N/A     | N/A   |
+| **Structure**     |               |     |           |       |         |       |
+| JSON Schema       | ✅           | ✅  | ✅        | ✅    | ✅      | ✅   |
+| Choice            | ✅           | ✅  | ✅        | ✅    | ❌      | ✅   |
+| Regex             | ✅           | ✅  | ✅        | ✅    | ❌      | ❌   |
+| Grammar           | ✅           | ✅  | ✅        | ✅    | ❌      | ❌   |
 
 
 [structured-outputs]: https://platform.openai.com/docs/guides/structured-outputs
diff --git a/docs/welcome.md b/docs/welcome.md
index 728a30a97..bd7377f5f 100644
--- a/docs/welcome.md
+++ b/docs/welcome.md
@@ -6,7 +6,7 @@ Outlines is a Python library that allows you to use Large Language Model in a si
 
 ## What models do you support?
 
-We support [Openai](reference/models/openai.md), but the true power of Outlines is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [exllama2](reference/models/exllamav2.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
+We support [Openai](reference/models/openai.md), but the true power of Outlines is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
 
 ## What are the main features?
 
diff --git a/mkdocs.yml b/mkdocs.yml
index 2e77a0440..430840ac1 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -159,7 +159,6 @@ nav:
               - Llama.cpp: reference/models/llamacpp.md
               - vLLM: reference/models/vllm.md
               - TGI: reference/models/tgi.md
-              - ExllamaV2: reference/models/exllamav2.md
               - MLX: reference/models/mlxlm.md
               - Mamba: reference/models/transformers/#mamba
           - API:
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index fb982d3b2..9bd06ead6 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -11,7 +11,6 @@
 from .anthropic import from_anthropic, Anthropic
 from .base import Model, ModelTypeAdapter
 from .dottxt import Dottxt, from_dottxt
-from .exllamav2 import ExLlamaV2Model, exl2
 from .gemini import from_gemini, Gemini
 from .llamacpp import LlamaCpp, from_llamacpp
 from .mlxlm import MLXLM, from_mlxlm
@@ -26,7 +25,7 @@
 from .vllm import VLLM, from_vllm
 
 LogitsGenerator = Union[
-    Transformers, LlamaCpp, OpenAI, ExLlamaV2Model, MLXLM, VLLM, Ollama
+    Transformers, LlamaCpp, OpenAI, MLXLM, VLLM, Ollama
 ]
 
 SteerableModel = Union[LlamaCpp, Transformers, MLXLM, VLLM]
diff --git a/outlines/models/exllamav2.py b/outlines/models/exllamav2.py
deleted file mode 100644
index efd70f96a..000000000
--- a/outlines/models/exllamav2.py
+++ /dev/null
@@ -1,359 +0,0 @@
-import dataclasses
-from typing import TYPE_CHECKING, Iterator, List, Optional, Tuple, TypedDict, Union
-
-import torch
-from typing_extensions import Unpack
-
-from outlines.generate.api import GenerationParameters, SamplingParameters
-
-if TYPE_CHECKING:
-    import torch.LongTensor
-    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler
-
-
-class ExllamaV2Params(TypedDict, total=False):
-    max_tokens: int
-    stop_conditions: Optional[List[Union[int, str]]]
-    seed: Optional[int]
-    gen_settings: "ExLlamaV2Sampler.Settings"
-    max_new_tokens: List[int]
-
-
-class OutlinesExLlamaV2Tokenizer:
-    def __init__(self, tokenizer):
-        self.exl2_tokenizer = tokenizer
-        self.vocabulary = self.exl2_tokenizer.get_piece_to_id_dict()
-        self.special_tokens = set(self.exl2_tokenizer.extended_piece_to_id)
-        self.eos_token_id = self.exl2_tokenizer.eos_token_id
-
-    def convert_token_to_string(self, token):
-        return token
-
-    def decode(self, token_ids: "torch.LongTensor") -> List[str]:
-        decoded = self.exl2_tokenizer.decode(
-            torch.tensor(token_ids),
-            decode_special_tokens=False,
-        )
-        if isinstance(decoded, str):
-            return [decoded]
-        return decoded
-
-
-class ExLlamaV2Model:
-    """Represents a `exl2` model."""
-
-    def __init__(
-        self,
-        generator: "ExLlamaV2DynamicGenerator",
-        tokenizer: "OutlinesExLlamaV2Tokenizer",
-        max_seq_len: int,
-    ):
-        self.generator = generator
-        self.tokenizer = tokenizer
-        self.max_seq_len = max_seq_len
-
-    def prepare_generation_parameters(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        sampling_parameters: SamplingParameters,
-        structure_logits_processor,
-        **exllamav2_params: Unpack[ExllamaV2Params],
-    ) -> Tuple[ExllamaV2Params, Union[str, List[str]]]:
-        """Prepare the generation parameters.
-
-        `exllamav2` uses different default values
-
-        """
-        from exllamav2.generator import ExLlamaV2Sampler
-
-        if isinstance(prompts, str):
-            prompts = [prompts]
-        max_tokens, stop_at, seed = dataclasses.astuple(generation_parameters)
-
-        if max_tokens is None:
-            max_tokens = []
-            for prompt in prompts:
-                ids = self.generator.tokenizer.encode(
-                    prompt, encode_special_tokens=True
-                )
-                prompt_tokens = ids.shape[-1]
-                max_tokens.append(self.max_seq_len - prompt_tokens)
-            exllamav2_params["max_new_tokens"] = max_tokens
-        else:
-            exllamav2_params["max_new_tokens"] = [
-                max_tokens for _ in range(len(prompts))
-            ]
-
-        stop_conditions = [self.generator.tokenizer.eos_token_id]
-        if isinstance(generation_parameters.stop_at, str):
-            stop_conditions.append(generation_parameters.stop_at)
-        elif isinstance(generation_parameters.stop_at, list):
-            for stop_at in generation_parameters.stop_at:
-                stop_conditions.append(stop_at)
-        exllamav2_params["stop_conditions"] = stop_conditions
-        exllamav2_params["seed"] = seed
-
-        gen_settings = ExLlamaV2Sampler.Settings()
-        if sampling_parameters.temperature is not None:
-            gen_settings.temperature = sampling_parameters.temperature
-        if sampling_parameters.top_p is not None:
-            gen_settings.top_p = sampling_parameters.top_p
-        if sampling_parameters.top_k is not None:
-            gen_settings.top_k = sampling_parameters.top_k
-        gen_settings.logits_processor = structure_logits_processor
-        exllamav2_params["gen_settings"] = gen_settings
-        if sampling_parameters.num_samples > 1:
-            prompts = prompts * sampling_parameters.num_samples
-            exllamav2_params["max_new_tokens"] = (
-                exllamav2_params["max_new_tokens"] * sampling_parameters.num_samples
-            )
-
-        if len(prompts) == 1:
-            prompts = prompts[0]
-
-        return exllamav2_params, prompts
-
-    def reformat_output(
-        self, output: Union[str, List[str]], sampling_parameters: SamplingParameters
-    ):
-        """
-        The purpose of this function is to reformat the output from exllamav2's output format to outline's output format.
-
-        For exllamav2, it mainly accepts only a list or a string(they also do cfg sampling with tuples but we will ignore this for now).
-        The exllamav2's logic is:
-
-        1. If the prompt is a string, return a string. This is the same as outlines
-        2. If a prompt is a list, return a list. This is not the same as outlines output in that if the list is only one element, the string is expected to be outputted.
-        3. There is no such thing as num_samples, so the prompts had to be duplicated by num_samples times. Then, we had the function output a list of lists
-        """
-        if isinstance(output, str):
-            return output
-        if len(output) == 1:
-            return output[0]
-        if sampling_parameters.num_samples > 1:
-            if len(output) == sampling_parameters.num_samples:
-                return output
-            assert len(output) % sampling_parameters.num_samples == 0
-            num_items_per_sample = len(output) // sampling_parameters.num_samples
-            new_output = []
-            for i in range(sampling_parameters.num_samples):
-                curr_sample = []
-                for j in range(num_items_per_sample):
-                    curr_sample.append(output[i * num_items_per_sample + j])
-                new_output.append(curr_sample)
-            return new_output
-        return output
-
-    def generate(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **exllamav2_params: Unpack[ExllamaV2Params],
-    ) -> Union[str, List[str]]:
-        exllamav2_params, prompts = self.prepare_generation_parameters(
-            prompts,
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
-        )
-        """
-        In exllamav2, it needs the max amount of new tokens generated.
-        The reason exllamav2_params["max_new_tokens"] is a list is because in prepare_generation_parameters
-        the max amount of tokens that can be generated by the model for each prompt(by encoding with tokenizer) is calculated.
-        The minimum is picked because otherwise it might be possible for one of the
-        prompts to exceed the max sequence length.
-        """
-        output = self.generator.generate(
-            prompt=prompts,
-            gen_settings=exllamav2_params["gen_settings"],
-            max_new_tokens=min(exllamav2_params["max_new_tokens"]),
-            completion_only=True,
-            encode_special_tokens=True,
-            stop_conditions=exllamav2_params["stop_conditions"],
-            add_bos=False,
-            seed=exllamav2_params["seed"],
-        )
-
-        return self.reformat_output(output, sampling_parameters)
-
-    def stream(
-        self,
-        prompts: Union[str, List[str]],
-        generation_parameters: GenerationParameters,
-        structure_logits_processor,
-        sampling_parameters: SamplingParameters,
-        **exllamav2_params: Unpack[ExllamaV2Params],
-    ) -> Iterator[Union[str, List[str]]]:
-        from exllamav2.generator import ExLlamaV2DynamicJob
-
-        exllamav2_params, prompts = self.prepare_generation_parameters(
-            prompts,
-            generation_parameters,
-            sampling_parameters,
-            structure_logits_processor,
-        )
-
-        order = {}
-        if isinstance(prompts, str):
-            prompts = [prompts]
-        batch_size = len(prompts)
-        seed = exllamav2_params["seed"]
-        for idx, p in enumerate(prompts):
-            input_ids = self.generator.tokenizer.encode(
-                p, encode_special_tokens=True, add_bos=False
-            )
-
-            job = ExLlamaV2DynamicJob(
-                input_ids=input_ids,
-                max_new_tokens=exllamav2_params["max_new_tokens"][idx],
-                min_new_tokens=0,
-                seed=seed,
-                stop_conditions=exllamav2_params["stop_conditions"],
-                gen_settings=exllamav2_params["gen_settings"],
-                token_healing=False,
-                decode_special_tokens=False,
-            )
-
-            if seed is not None:
-                seed += 1
-
-            serial = self.generator.enqueue(job)
-            order[serial] = idx
-
-        # Collect outputs until all jobs finish
-
-        next_text = [""] * batch_size
-
-        def token_generator() -> Iterator[str]:
-            while self.generator.num_remaining_jobs():
-                results = self.generator.iterate()
-                for r in results:
-                    idx = order[r["serial"]]
-                    if r["stage"] == "streaming":
-                        text = r.get("text", "")
-                        next_text[idx] = text
-                    if r["eos"]:
-                        next_text[idx] = ""
-                yield self.reformat_output(next_text, sampling_parameters)
-            return
-
-        return token_generator()
-
-
-def exl2(
-    model_path: str,
-    draft_model_path: Optional[str] = None,
-    max_seq_len: Optional[int] = None,
-    cache_q4: bool = False,
-    paged: bool = True,
-    max_chunk_size: Optional[int] = None,
-) -> ExLlamaV2Model:
-    """
-    Load an ExLlamaV2 model.
-
-    Parameters
-    ----------
-    model_path (str)
-        Path to the model directory.
-    device (str)
-        Device to load the model on. Pass in 'cuda' for GPU or 'cpu' for CPU
-    max_seq_len (Optional[int], optional)
-        Maximum sequence length. Defaults to None.
-    scale_pos_emb (Optional[float], optional)
-        Scale factor for positional embeddings. Defaults to None.
-    scale_alpha_value (Optional[float], optional)
-        Scale alpha value. Defaults to None.
-    no_flash_attn (Optional[bool], optional)
-        Disable flash attention. Defaults to None.
-    num_experts_per_token (Optional[int], optional)
-        Number of experts per token. Defaults to None.
-    cache_q4 (bool, optional)
-        Use Q4 cache. Defaults to False.
-    tokenizer_kwargs (dict, optional)
-        Additional keyword arguments for the tokenizer. Defaults to {}.
-    gpu_split (str)
-        \"auto\", or VRAM allocation per GPU in GB. Auto will use exllama's autosplit feature
-    low_mem (bool, optional)
-        Enable VRAM optimizations, potentially trading off speed
-    verbose (bool, optional)
-        Enable if you want debugging statements
-
-    Returns
-    -------
-    An `ExLlamaV2Model` instance.
-
-    Raises
-    ------
-    `ImportError` if the `exllamav2` library is not installed.
-
-    """
-    try:
-        from exllamav2 import (
-            ExLlamaV2,
-            ExLlamaV2Cache,
-            ExLlamaV2Cache_Q4,
-            ExLlamaV2Config,
-            ExLlamaV2Tokenizer,
-        )
-        from exllamav2.generator import ExLlamaV2DynamicGenerator
-
-    except ImportError:
-        raise ImportError(
-            "The `exllamav2`, `transformers` and `torch` libraries needs to be installed in order to use `exllamav2` models. "
-            "Please run `pip install transformers torch git+https://github.com/lapp0/exllamav2@sampler-logits-processor` "
-            "Documentation: https://dottxt-ai.github.io/outlines/latest/reference/models/exllamav2/"
-        )
-    config = ExLlamaV2Config(model_path)
-    if max_chunk_size is not None:
-        config.max_input_len = max_chunk_size
-        config.max_attention_size = max_chunk_size**2
-
-    config.arch_compat_overrides()
-    model = ExLlamaV2(config)
-    if max_seq_len is None:
-        max_seq_len = -1
-    if cache_q4:
-        cache = ExLlamaV2Cache_Q4(model, max_seq_len=max_seq_len, lazy=True)
-    else:
-        cache = ExLlamaV2Cache(model, max_seq_len=max_seq_len, lazy=True)
-    model.load_autosplit(cache, progress=True)
-
-    print("Loading tokenizer...")
-    tokenizer = ExLlamaV2Tokenizer(config)
-    max_batch_size = 4 if paged else 1
-
-    draft_model = None
-    draft_cache = None
-    if draft_model_path is not None:
-        draft_config = ExLlamaV2Config(draft_model_path)
-        draft_model = ExLlamaV2(draft_config)
-
-        if cache_q4:
-            draft_cache = ExLlamaV2Cache_Q4(
-                draft_model, max_seq_len=max_seq_len, lazy=True
-            )
-        else:
-            draft_cache = ExLlamaV2Cache(
-                draft_model, max_seq_len=max_seq_len, lazy=True
-            )
-
-    # Initialize the generator with all default parameters
-    generator = ExLlamaV2DynamicGenerator(
-        model=model,
-        cache=cache,
-        draft_model=draft_model,
-        draft_cache=draft_cache,
-        tokenizer=tokenizer,
-        max_batch_size=max_batch_size,
-        use_ngram_draft=False,
-        max_chunk_size=max_chunk_size,
-        paged=paged,
-    )
-    max_seq_len = cache.max_seq_len
-
-    outlines_tokenizer = OutlinesExLlamaV2Tokenizer(tokenizer)
-    outlines_exl2_model = ExLlamaV2Model(generator, outlines_tokenizer, max_seq_len)
-    return outlines_exl2_model
diff --git a/pyproject.toml b/pyproject.toml
index 2c5382e79..8881a9501 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -67,7 +67,6 @@ test = [
     "google-generativeai",
     "transformers",
     "pillow",
-    "exllamav2",
     "jax",
     "ollama",
     "dottxt",
@@ -115,7 +114,6 @@ enable_incomplete_feature = ["Unpack"]
 
 [[tool.mypy.overrides]]
 module = [
-    "exllamav2.*",
     "jax",
     "jaxlib",
     "jax.numpy",
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
index abd9c72a4..8e5db2584 100644
--- a/tests/generate/conftest.py
+++ b/tests/generate/conftest.py
@@ -30,7 +30,6 @@ def pytest_collection_modifyitems(config, items):
                 if (
                     model_param.startswith("model_transformers_vision")
                     or model_param.startswith("model_vllm")
-                    or model_param.startswith("model_exllamav2")
                 ):
                     item.add_marker(skip_marker)
 
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
index 84c8b761c..42471fba9 100644
--- a/tests/generate/test_generate.py
+++ b/tests/generate/test_generate.py
@@ -25,24 +25,6 @@ def model_llamacpp(tmp_path_factory):
     return outlines.from_llamacpp(llm)
 
 
-@pytest.fixture(scope="session")
-def model_exllamav2(tmp_path_factory):
-    from outlines.models.exllamav2 import exl2
-    from huggingface_hub import snapshot_download
-
-    tmp_dir = tmp_path_factory.mktemp("model_download")
-    model_path = snapshot_download(
-        repo_id="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4.6-exl2",
-        cache_dir=tmp_dir,
-    )
-
-    return exl2(
-        model_path=model_path,
-        cache_q4=True,
-        paged=False,
-    )
-
-
 @pytest.fixture(scope="session")
 def model_mlxlm(tmp_path_factory):
     from mlx_lm import load
@@ -120,16 +102,8 @@ def model_vllm(tmp_path_factory):
     return outlines.from_vllm(LLM("facebook/opt-125m", gpu_memory_utilization=0.1))
 
 
-# TODO: exllamav2 failing in main, address in https://github.com/dottxt-ai/outlines/issues/808
 # TODO: t5 tokenizer doesn't work with streaming
 """
-@pytest.fixture(scope="session")
-def model_exllamav2(tmp_path_factory):
-    return models.exllamav2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        device="cpu"
-    )
-
 @pytest.fixture(scope="session")
 def model_t5(tmp_path_factory):
     from transformers import AutoModelForSeq2SeqLM
@@ -142,7 +116,6 @@ def model_t5(tmp_path_factory):
 
 ALL_MODEL_FIXTURES = (
     # "model_llamacpp",  # temporary disabled due to the v1 model refactoring
-    "model_exllamav2",
     # "model_mlxlm",
     # "model_mlxlm_phi3",
     # "model_transformers_random",
diff --git a/tests/generate/test_integration_exllamav2.py b/tests/generate/test_integration_exllamav2.py
deleted file mode 100644
index 12c4143b3..000000000
--- a/tests/generate/test_integration_exllamav2.py
+++ /dev/null
@@ -1,363 +0,0 @@
-import importlib
-from unittest.mock import patch
-
-import pytest
-
-import outlines.models as models
-from outlines.generate.api import GenerationParameters, SamplingParameters
-from outlines.models.exllamav2 import ExLlamaV2Model
-
-
-@pytest.fixture(scope="session")
-def model_exllamav2(tmp_path_factory):
-    return models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=False,
-    )
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_import_error(request, model_fixture):
-    with patch.dict("sys.modules", {"exllamav2": None}):
-        with pytest.raises(ImportError):
-            models.exl2(
-                model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-                cache_q4=True,
-                paged=False,
-            )
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_attributes(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    assert hasattr(model, "generator")
-    assert hasattr(model, "tokenizer")
-    assert model.tokenizer.convert_token_to_string(1) == 1
-    assert hasattr(model, "max_seq_len")
-    assert isinstance(model.max_seq_len, int)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_prompt_types(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-    prompt = ["test"]
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_no_max_tokens(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_test_stop_at(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_generate_multisampling(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    output = model.generate(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    assert isinstance(output, list)
-    assert isinstance(output[0], str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_prepare_generation_parameters(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    exllamav2_params, prompts = model.prepare_generation_parameters(
-        prompt, generation_params, sampling_params, structure_logits_processor
-    )
-    assert isinstance(exllamav2_params, dict)
-    assert isinstance(prompts, list)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_prompt_types(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-    prompt = ["test"]
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_no_max_tokens(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=None, stop_at=None, seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_test_stop_at(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-    generation_params = GenerationParameters(max_tokens=10, stop_at=["stop"], seed=None)
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_multisampling(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, stop_at="stop", seed=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, list)
-        assert isinstance(token[0], str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_model_stream_seed(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    prompt = "test"
-    generation_params = GenerationParameters(max_tokens=10, seed=1, stop_at=None)
-    structure_logits_processor = None
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-        0.9,
-        50,
-        1.0,
-    )
-    generator = model.stream(
-        prompt, generation_params, structure_logits_processor, sampling_params
-    )
-    for token in generator:
-        assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_reformat_output(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-    )
-    output = "test"
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert reformatted_output == output
-    output = ["test"]
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert reformatted_output == output[0]
-    output = ["test", "test"]
-    sampling_params = SamplingParameters(
-        "multinomial",
-        1,
-    )
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert len(reformatted_output) == 2
-    assert reformatted_output[0] == "test"
-    assert reformatted_output[1] == "test"
-    output = ["test", "test"]
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert len(reformatted_output) == 2
-    assert reformatted_output[0] == "test"
-    assert reformatted_output[1] == "test"
-    output = ["test", "test", "test", "test"]
-    sampling_params = SamplingParameters(
-        "multinomial",
-        2,
-    )
-    reformatted_output = model.reformat_output(output, sampling_params)
-    assert len(reformatted_output) == 2
-    assert reformatted_output[0] == ["test", "test"]
-    assert reformatted_output[1] == ["test", "test"]
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_max_chunk_size(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=False,
-        max_chunk_size=128,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_cache_default(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        paged=False,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def is_flash_attn_available():
-    try:
-        importlib.import_module("flash_attn")
-    except (ImportError, AssertionError):
-        return False
-    return True
-
-
-@pytest.mark.skipif(not is_flash_attn_available(), reason="flash-attn is not installed")
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_paged(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=True,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_draft_model(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        cache_q4=True,
-        paged=False,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_draft_model_cache_default(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        draft_model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        paged=False,
-    )
-    assert isinstance(model, ExLlamaV2Model)
-
-
-@pytest.mark.parametrize("model_fixture", ["model_exllamav2"])
-def test_exl2_set_max_seq_len(request, model_fixture):
-    model = models.exl2(
-        model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-        max_seq_len=2048,
-        paged=False,
-        cache_q4=True,
-    )
-    assert isinstance(model, ExLlamaV2Model)
diff --git a/uv.lock b/uv.lock
index f1deefbe6..2d927f136 100644
--- a/uv.lock
+++ b/uv.lock
@@ -625,28 +625,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/02/cc/b7e31358aac6ed1ef2bb790a9746ac2c69bcb3c8588b41616914eb106eaf/exceptiongroup-1.2.2-py3-none-any.whl", hash = "sha256:3111b9d131c238bec2f8f516e123e14ba243563fb135d3fe885990585aa7795b", size = 16453 },
 ]
 
-[[package]]
-name = "exllamav2"
-version = "0.2.7"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "fastparquet" },
-    { name = "ninja" },
-    { name = "numpy" },
-    { name = "pandas" },
-    { name = "pillow" },
-    { name = "pygments" },
-    { name = "regex" },
-    { name = "rich" },
-    { name = "safetensors" },
-    { name = "sentencepiece" },
-    { name = "torch" },
-    { name = "websockets" },
-]
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/b5/04/300012908311c9a904d44c74a037821e7cde530c648afff4ad6bed4e22c3/exllamav2-0.2.7-py3-none-any.whl", hash = "sha256:410d1a69fed329f6f1c10de6988a3b6b33e1fa715aa4fe32c545edd1abd1254b", size = 1512625 },
-]
-
 [[package]]
 name = "fastapi"
 version = "0.115.6"
@@ -1917,9 +1895,6 @@ dependencies = [
 ]
 
 [package.optional-dependencies]
-exllamav2 = [
-    { name = "exllamav2" },
-]
 llamacpp = [
     { name = "datasets" },
     { name = "llama-cpp-python" },
@@ -1946,7 +1921,6 @@ test = [
     { name = "datasets" },
     { name = "diff-cover", version = "9.2.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.9.17'" },
     { name = "diff-cover", version = "9.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.9.17'" },
-    { name = "exllamav2" },
     { name = "huggingface-hub" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
@@ -1989,8 +1963,6 @@ requires-dist = [
     { name = "datasets", marker = "extra == 'transformers'" },
     { name = "diff-cover", marker = "extra == 'test'" },
     { name = "diskcache" },
-    { name = "exllamav2", marker = "extra == 'exllamav2'" },
-    { name = "exllamav2", marker = "extra == 'test'" },
     { name = "fastapi", marker = "extra == 'serve'" },
     { name = "genson" },
     { name = "huggingface-hub", marker = "extra == 'test'" },

From dd3fe3fa5653793adae0daffb5cd3697ca96d698 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Thu, 27 Feb 2025 16:36:35 +0100
Subject: [PATCH 249/505] Delete generate tests

---
 pyproject.toml                  |   1 +
 tests/generate/conftest.py      |  41 ---
 tests/generate/test_generate.py | 438 --------------------------------
 3 files changed, 1 insertion(+), 479 deletions(-)
 delete mode 100644 tests/generate/test_generate.py

diff --git a/pyproject.toml b/pyproject.toml
index 8881a9501..25fd8e5dc 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -70,6 +70,7 @@ test = [
     "jax",
     "ollama",
     "dottxt",
+    "sentencepiece",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
 serve = [
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
index 8e5db2584..ef7e40eed 100644
--- a/tests/generate/conftest.py
+++ b/tests/generate/conftest.py
@@ -1,47 +1,6 @@
 from importlib import reload
 
 import pytest
-import torch
-
-
-def is_metal_available():
-    try:
-        import mlx.core as mx
-        import mlx_lm  # noqa: F401
-
-        assert mx.metal.is_available()
-    except (ImportError, AssertionError):
-        return False
-    return True
-
-
-def pytest_collection_modifyitems(config, items):
-    """
-    If mlxlm and Metal aren't available, skip mlxlm tests
-    If CUDA isn't available, skip vllm and transformers_vision
-    """
-    if not torch.cuda.is_available():
-        skip_marker = pytest.mark.skip(
-            reason="Skipping test because CUDA is not available"
-        )
-        for item in items:
-            if "model_fixture" in item.fixturenames:
-                model_param = item.callspec.params.get("model_fixture", None)
-                if (
-                    model_param.startswith("model_transformers_vision")
-                    or model_param.startswith("model_vllm")
-                ):
-                    item.add_marker(skip_marker)
-
-    if not is_metal_available():
-        skip_marker = pytest.mark.skip(
-            reason="Skipping test because mlx-lm or Metal are not available"
-        )
-        for item in items:
-            if "model_fixture" in item.fixturenames:
-                model_param = item.callspec.params.get("model_fixture", None)
-                if model_param.startswith("model_mlxlm"):
-                    item.add_marker(skip_marker)
 
 
 @pytest.fixture
diff --git a/tests/generate/test_generate.py b/tests/generate/test_generate.py
deleted file mode 100644
index 42471fba9..000000000
--- a/tests/generate/test_generate.py
+++ /dev/null
@@ -1,438 +0,0 @@
-import contextlib
-import re
-from enum import Enum
-
-import pytest
-
-import outlines
-import outlines.generate as generate
-import outlines.samplers as samplers
-
-##########################################
-# Model Fixtures
-##########################################
-
-
-@pytest.fixture(scope="session")
-def model_llamacpp(tmp_path_factory):
-    from llama_cpp import Llama
-
-    llm = Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        verbose=False,
-    )
-    return outlines.from_llamacpp(llm)
-
-
-@pytest.fixture(scope="session")
-def model_mlxlm(tmp_path_factory):
-    from mlx_lm import load
-
-    return outlines.from_mlxlm(*load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit"))
-
-
-@pytest.fixture(scope="session")
-def model_mlxlm_phi3(tmp_path_factory):
-    from mlx_lm import load
-
-    return outlines.from_mlxlm(*load("mlx-community/Phi-3-mini-4k-instruct-4bit"))
-
-
-@pytest.fixture(scope="session")
-def model_transformers_random(tmp_path_factory):
-    from transformers import AutoModelForCausalLM, AutoTokenizer
-
-    return outlines.from_transformers(
-        AutoModelForCausalLM.fromt_pretrained("hf-internal-testing/tiny-random-gpt2"),
-        AutoTokenizer.fromt_pretrained("hf-internal-testing/tiny-random-gpt2"),
-    )
-
-
-@pytest.fixture(scope="session")
-def model_transformers_opt125m(tmp_path_factory):
-    from transformers import AutoModelForCausalLM, AutoTokenizer
-
-    return outlines.from_transformers(
-        AutoModelForCausalLM.fromt_pretrained("facebook/opt-125m"),
-        AutoTokenizer.fromt_pretrained("facebook/opt-125m"),
-    )
-
-
-@pytest.fixture(scope="session")
-def model_mamba(tmp_path_factory):
-    from transformers import MambaModel, AutoTokenizer
-
-    return outlines.from_transformers(
-        MambaModel.from_pretrained(model_name="state-spaces/mamba-130m-hf"),
-        AutoTokenizer.from_pretrained(model_name="state-spaces/mamba-130m-hf"),
-    )
-
-
-@pytest.fixture(scope="session")
-def model_bart(tmp_path_factory):
-    from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
-
-    return outlines.from_transformers(
-        AutoModelForSeq2SeqLM.from_pretrained("facebook/bart-base"),
-        AutoTokenizer.from_pretrained("facebook/bart-base"),
-    )
-
-
-@pytest.fixture(scope="session")
-def model_transformers_vision(tmp_path_factory):
-    import torch
-    from transformers import LlavaNextForConditionalGeneration, AutoTokenizer
-
-    return outlines.from_transformers(
-        LlavaNextForConditionalGeneration.from_pretrained(
-            "llava-hf/llava-v1.6-mistral-7b-hf",
-            torch_dtype=torch.bfloat16,
-            load_in_4bit=True,
-            low_mem_cpu_usage=True,
-        ),
-        AutoTokenizer.from_pretrained("llava-hf/llava-v1.6-mistral-7b-hf"),
-    )
-
-
-@pytest.fixture(scope="session")
-def model_vllm(tmp_path_factory):
-    from vllm import LLM
-
-    return outlines.from_vllm(LLM("facebook/opt-125m", gpu_memory_utilization=0.1))
-
-
-# TODO: t5 tokenizer doesn't work with streaming
-"""
-@pytest.fixture(scope="session")
-def model_t5(tmp_path_factory):
-    from transformers import AutoModelForSeq2SeqLM
-
-    return models.transformers(
-        "EleutherAI/pile-t5-base", device="cpu", model_class=AutoModelForSeq2SeqLM
-    )
-"""
-
-
-ALL_MODEL_FIXTURES = (
-    # "model_llamacpp",  # temporary disabled due to the v1 model refactoring
-    # "model_mlxlm",
-    # "model_mlxlm_phi3",
-    # "model_transformers_random",
-    # "model_transformers_opt125m",
-    # "model_mamba",
-    # "model_bart",
-    # "model_transformers_vision",
-    # "model_vllm",
-)
-
-class MyEnum(Enum):
-    foo = "foo"
-    bar = "bar"
-    baz = "baz"
-
-
-ALL_SAMPLE_CHOICES_FIXTURES = (
-    ["foo", "bar", "baz"],
-    MyEnum,
-)
-
-
-##########################################
-# Stuctured Generation Inputs
-##########################################
-
-
-@pytest.fixture()
-def sample_schema():
-    from pydantic import BaseModel, conint, conlist, constr
-
-    class SampleSchema(BaseModel):
-        title: constr(max_length=10)
-        numbers: conlist(conint(strict=True), min_length=3, max_length=3)
-        labels: conlist(constr(min_length=1, max_length=5), min_length=3, max_length=3)
-
-    return SampleSchema
-
-
-@pytest.fixture()
-def sample_choices():
-    return ["foo", "bar", "baz"]
-
-
-@pytest.fixture()
-def sample_lark_grammar():
-    # from https://github.com/lark-parser/lark/blob/master/docs/grammar.md
-    return """
-    ?start: hello_world "!" number
-    hello_world: ("hello" | "world") ~ 3
-    number: ("0".."9") ~ 5
-    thanks: "Thank"i " for testing!"
-    """
-
-
-REGEX_PATTERNS = [
-    "a b c d e",  # ensure proper tokenizer whitespace prefix handling
-    "(123456789)|(abcdefghijklmnop)",  # ensure consistent correct sequence handling during batch
-    r"([a-z]{10})@([a-z]{5})\.([a-z]{3})",  # email example
-]
-
-
-###########################################
-# Model/Generator Pair Behavior Definitions
-###########################################
-
-
-def enforce_not_implemented(model_fixture, *task_names):
-    """
-    Per `NOT_IMPLEMENTED`, mapping, if a model hasn't implemented a task,
-    assert an NotImplementedError is raised. Otherwise, run normally
-    """
-    NOT_IMPLEMENTED = {
-        "stream": ["model_transformers_vision", "model_vllm"],
-        "batch": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-        "beam_search": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-        "multiple_samples": ["model_llamacpp", "model_mlxlm", "model_mlxlm_phi3"],
-        "cfg": ["model_llamacpp"],  # TODO: fix llama_cpp tokenizer
-    }
-    for task_name in task_names:
-        if model_fixture in NOT_IMPLEMENTED.get(task_name, []):
-            return pytest.raises(NotImplementedError)
-    else:
-        return contextlib.nullcontext()
-
-
-def get_inputs(fixture_name, batch_size=None):
-    """Get generator kwargs, just the prompt by default, but include images for transformers_visian"""
-    from io import BytesIO
-    from urllib.request import urlopen
-
-    from PIL import Image  # type: ignore
-
-    prompts = ["abcd", "efgh", "1234", "5678", "foo", "bar", "baz", "bif"]
-    prompts = prompts[0] if batch_size is None else prompts[:batch_size]
-
-    if fixture_name.endswith("_vision"):
-        img_url = "https://python-pillow.org/pillow-perf/static/space_pil_lanczos.png"
-        img = Image.open(BytesIO(urlopen(img_url).read())).convert("RGB")
-
-        if batch_size is None:
-            return {"prompts": f"<image> {prompts}", "media": [img]}
-        else:
-            return {
-                "prompts": [f"<image> {p}" for p in prompts],
-                "media": [[img] for _ in range(batch_size)],
-            }
-
-    else:
-        return {"prompts": prompts}
-
-
-###########################################
-# Tests
-###########################################
-
-
-@pytest.mark.parametrize("sampler_name", ("greedy", "multinomial", "beam_search"))
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_text(request, model_fixture, sampler_name):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model, getattr(samplers, sampler_name)())
-    with enforce_not_implemented(model_fixture, sampler_name):
-        res = generator(**get_inputs(model_fixture), max_tokens=10)
-        assert isinstance(res, str)
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex(request, model_fixture, pattern):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    res = generator(**get_inputs(model_fixture), max_tokens=20)
-    assert re.fullmatch(pattern, res) is not None, res
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_fsm(request, model_fixture, pattern):
-    import interegular
-
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.fsm(model, interegular.parse_pattern(pattern).to_fsm())
-    res = generator(**get_inputs(model_fixture))
-    assert re.fullmatch(pattern, res) is not None, res
-
-
-@pytest.mark.skip(
-    "Fix issues with JSON, some models fail this test https://github.com/dottxt-ai/outlines/issues/985"
-)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_json(request, model_fixture, sample_schema):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.json(model, sample_schema)
-    # asserts valid within call
-    generator(**get_inputs(model_fixture), max_tokens=100)
-
-
-# TODO: add support for genson in the Regex type of v1.0
-# def test_integrate_genson_generate_json(request):
-#    from genson import SchemaBuilder
-#
-#    builder = SchemaBuilder()
-#    builder.add_schema({"type": "object", "properties": {}})
-#    builder.add_object({"name": "Toto", "age": 5})
-#
-#    model = request.getfixturevalue("model_transformers_opt125m")
-#
-#    generator = generate.json(model, builder)
-#    res = generator("Return a json of a young boy")
-#
-#    assert "name" in res
-#    assert "age" in res
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
-def test_generate_choice(request, model_fixture, sample_choices):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.choice(model, sample_choices)
-    res = generator(**get_inputs(model_fixture))
-    if isinstance(sample_choices, type(Enum)):
-        assert res in [elt.value for elt in sample_choices]
-    else:
-        assert res in sample_choices
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-@pytest.mark.parametrize("sample_choices", ALL_SAMPLE_CHOICES_FIXTURES)
-def test_generate_choice_twice(request, model_fixture, sample_choices):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.choice(model, sample_choices)
-    res = generator(**get_inputs(model_fixture))
-    if isinstance(sample_choices, type(Enum)):
-        assert res in [elt.value for elt in sample_choices]
-    else:
-        assert res in sample_choices
-
-    res = generator(**get_inputs(model_fixture))
-    if isinstance(sample_choices, type(Enum)):
-        assert res in [elt.value for elt in sample_choices]
-    else:
-        assert res in sample_choices
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_format_bool(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.format(model, bool)
-    res = generator(**get_inputs(model_fixture))
-    assert isinstance(res, bool)
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_cfg(request, model_fixture, sample_lark_grammar):
-    from lark import Lark
-
-    from outlines import grammars
-
-    model = request.getfixturevalue(model_fixture)
-    with enforce_not_implemented(model_fixture, "cfg"):
-        generator = generate.cfg(model, sample_lark_grammar)
-        res = generator(**get_inputs(model_fixture))
-        # validate legal with the grammar via lark
-        # TODO: cleanup PartialLark so doesn't modify Lark globally
-        import importlib
-
-        import lark.lark
-
-        importlib.reload(lark.lark)
-        Lark(
-            sample_lark_grammar, parser="lalr", import_paths=[grammars.GRAMMAR_PATH]
-        ).parse(res)
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_text_stream(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model)
-    with enforce_not_implemented(model_fixture, "stream"):
-        for token in generator.stream(**get_inputs(model_fixture), max_tokens=10):
-            assert isinstance(token, str)
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_stream(request, model_fixture, pattern):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    with enforce_not_implemented(model_fixture, "stream"):
-        output = ""
-        for token in generator.stream(**get_inputs(model_fixture), max_tokens=20):
-            output += token
-        assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_batch_text(request, model_fixture):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.text(model)
-    with enforce_not_implemented(model_fixture, "batch"):
-        res = generator(**get_inputs(model_fixture, 2), max_tokens=10)
-        assert isinstance(res, list)
-        assert isinstance(res[0], str)
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_batch(request, model_fixture, pattern):
-    """Ensure batch requests work and fsm order is maintained"""
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    with enforce_not_implemented(model_fixture, "batch"):
-        outputs = generator(**get_inputs(model_fixture, 4), max_tokens=20)
-        for output in outputs:
-            assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_batch_stream(request, model_fixture, pattern):
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern)
-    with enforce_not_implemented(model_fixture, "batch", "stream"):
-        outputs = ["", ""]
-        for tokens in generator.stream(**get_inputs(model_fixture, 2), max_tokens=20):
-            outputs[0] += tokens[0]
-            outputs[1] += tokens[1]
-        for output in outputs:
-            assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-def test_generate_regex_single_multinomial(request, model_fixture, pattern):
-    """Ensure batch requests work and fsm order is maintained"""
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(model, pattern, sampler=samplers.multinomial(4))
-    with enforce_not_implemented(model_fixture, "multiple_samples"):
-        output_sample_groups = generator(**get_inputs(model_fixture), max_tokens=40)
-        for output in output_sample_groups:
-            assert re.fullmatch(pattern, output) is not None, output
-
-
-@pytest.mark.parametrize("pattern", REGEX_PATTERNS)
-@pytest.mark.parametrize("model_fixture", ALL_MODEL_FIXTURES)
-@pytest.mark.parametrize("sampler_name", ("multinomial", "beam_search"))
-def test_generate_regex_batch_multi_sample(
-    request, model_fixture, pattern, sampler_name
-):
-    """Ensure batch requests work and fsm order is maintained"""
-    model = request.getfixturevalue(model_fixture)
-    generator = generate.regex(
-        model, pattern, sampler=getattr(samplers, sampler_name)(4)
-    )
-    with enforce_not_implemented(model_fixture, "batch", "multiple_samples"):
-        output_batch_groups = generator(**get_inputs(model_fixture, 4), max_tokens=40)
-        for output_sample_groups in output_batch_groups:
-            for output in output_sample_groups:
-                assert re.fullmatch(pattern, output) is not None, output

From c6c4c4f94656f4fac5d7acab9431cd1c76baaeaa Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Fri, 28 Feb 2025 15:37:55 +0100
Subject: [PATCH 250/505] Remove the generate functions (#1462)

---
 docs/cookbook/extract_event_details.py  |   2 +-
 docs/reference/generation/cfg.md        | 154 --------------------
 docs/reference/generation/choices.md    |  16 ---
 docs/reference/generation/format.md     |  23 ---
 docs/reference/generation/generation.md |  72 ++++------
 docs/reference/generation/json.md       | 178 ------------------------
 docs/reference/generation/regex.md      |  27 ----
 docs/reference/generation/types.md      |  28 +---
 docs/reference/models/dottxt.md         |   2 +-
 docs/reference/models/ollama.md         |   2 +-
 docs/reference/text.md                  |  86 ------------
 outlines/__init__.py                    |   2 +-
 outlines/generate/__init__.py           |  87 ------------
 outlines/generate/cfg.py                |  54 -------
 outlines/generate/choice.py             |  59 --------
 outlines/generate/format.py             |  48 -------
 outlines/generate/regex.py              |  66 ---------
 outlines/generate/text.py               |  50 -------
 outlines/generator.py                   |  80 +++++++++++
 outlines/models/base.py                 |   2 +-
 tests/models/test_dottxt.py             |   2 +-
 21 files changed, 119 insertions(+), 921 deletions(-)
 delete mode 100644 docs/reference/generation/cfg.md
 delete mode 100644 docs/reference/generation/choices.md
 delete mode 100644 docs/reference/generation/format.md
 delete mode 100644 docs/reference/generation/json.md
 delete mode 100644 docs/reference/generation/regex.md
 delete mode 100644 docs/reference/text.md
 delete mode 100644 outlines/generate/cfg.py
 delete mode 100644 outlines/generate/choice.py
 delete mode 100644 outlines/generate/format.py
 delete mode 100644 outlines/generate/regex.py
 delete mode 100644 outlines/generate/text.py
 create mode 100644 outlines/generator.py

diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
index a3231cd00..8a60cf90b 100644
--- a/docs/cookbook/extract_event_details.py
+++ b/docs/cookbook/extract_event_details.py
@@ -4,7 +4,7 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.generate import Generator
+from outlines import Generator
 from outlines.types import JsonType
 
 
diff --git a/docs/reference/generation/cfg.md b/docs/reference/generation/cfg.md
deleted file mode 100644
index f0104ec63..000000000
--- a/docs/reference/generation/cfg.md
+++ /dev/null
@@ -1,154 +0,0 @@
-# Grammar-structured generation
-
-You can pass any context-free grammar in the EBNF format and Outlines will generate an output that is valid to this grammar:
-
-```python
-from outlines import models, generate
-
-arithmetic_grammar = """
-    ?start: expression
-
-    ?expression: term (("+" | "-") term)*
-
-    ?term: factor (("*" | "/") factor)*
-
-    ?factor: NUMBER
-           | "-" factor
-           | "(" expression ")"
-
-    %import common.NUMBER
-"""
-
-model = models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = generate.cfg(model, arithmetic_grammar)
-sequence = generator(
-  "Alice had 4 apples and Bob ate 2. "
-  + "Write an expression for Alice's apples:"
-)
-
-print(sequence)
-# (8-2)
-```
-
-###### Disclaimer
-
-!!! Note "Experimental"
-
-    Outlines current **community-contributed** implementation of CFG-structured generation is experimental. This does not reflect the performance of [.txt](https://dottxt.co)'s product, where we have optimized grammar-structured generation to be as fast as regex-structured generation. Additionally, it does not fully align with the approach described in our [technical report](https://arxiv.org/pdf/2307.09702), aside from its use of incremental/partial parsing. This feature is still a work in progress, requiring performance enhancements and bug fixes for an ideal implementation. For more details, please see our [grammar-related open issues on GitHub](https://github.com/dottxt-ai/outlines/issues?q=is%3Aissue+is%3Aopen+label%3Agrammar).
-
-!!! Note "Greedy"
-
-    To mitigate performance issues, CFG-structured generation will use rejection sampling and iterate over the candidate tokens highest logit first,, completing once a single valid token ID is selected. This is effectively greedy generation.
-
-## Ready-to-use grammars
-
-Outlines contains a (small) library of grammars that can be imported and use directly. We can rewrite the previous example as:
-
-```python
-from outlines import models, generate
-
-arithmetic_grammar = outlines.grammars.arithmetic
-
-model = models.transformers("WizardLM/WizardMath-7B-V1.1")
-generator = generate.cfg(model, arithmetic_grammar)
-sequence = generator(
-  "Alice had 4 apples and Bob ate 2. "
-  + "Write an expression for Alice's apples:"
-)
-
-print(sequence)
-# (8-2)
-```
-
-The following grammars are currently available:
-
-- Arithmetic grammar via `outlines.grammars.arithmetic`
-- JSON grammar via `outlines.grammars.json`
-
-If you would like more grammars to be added to the repository, please open an [issue](https://github.com/dottxt-ai/outlines/issues) or a [pull request](https://github.com/dottxt-ai/outlines/pulls).
-
-
-## Grammar guide
-
-A grammar is a list of rules and terminals that define a *language*:
-
-- Terminals define the vocabulary of the language; they may be a string, regular expression or combination of these and other terminals.
-- Rules define the structure of that language; they are a list of terminals and rules.
-
-Outlines uses the [Lark library](https://github.com/lark-parser/lark) to make Large Language Models generate text in a language of a grammar, it thus uses grammars defined in a format that Lark understands, based on the [EBNF syntax](https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form). Read the [Lark documentation](https://lark-parser.readthedocs.io/en/stable/grammar.html) for more details on grammar, the following is a small primer that should help get your started.
-
-In the following we will define a [LOGO-like toy language](https://github.com/lark-parser/lark/blob/master/examples/turtle_dsl.py) for python's [turtle](https://docs.python.org/3/library/turtle.html) library.
-
-### Terminals
-
-A turtle can take 4 different `MOVEMENT` move instructions: forward (`f`), backward (`b`), turn right (`r`) and turn left (`l`). It can take `NUMBER` number of steps in each direction, and draw lines in a specified `COLOR`. These define the vocabulary of our language:
-
-```ebnf
-MOVEMENT: "f"|"b"|"r"|"l"
-COLOR: LETTER+
-
-%import common.LETTER
-%import common.INT -> NUMBER
-%import common.WS
-%ignore WS
-```
-
-The lines that start with `%` are called "directive". They allow to import pre-defined terminals and rules, such as `LETTER` and `NUMBER`. `LETTER+` is a regular expressions, and indicates that a `COLOR` is made of at least one `LETTER`. The last two lines specify that we will ignore white spaces (`WS`) in the grammar.
-
-### Rules
-
-We now need to define our rules, by decomposing instructions we can send to the turtle via our python program. At each line of the program, we can either choose a direction and execute a given number of steps, change the color used to draw the pattern. We can also choose to start filling, make a series of moves, and stop filling. We can also choose to repeat a series of move.
-
-We can easily write the first two rules:
-
-```ebnf
-instruction: MOVEMENT NUMBER   -> movement
-           | "c" COLOR [COLOR] -> change_color
-```
-
-where `movement` and `change_color` represent aliases for the rules. A whitespace implied concatenating the elements, and `|` choosing either of the elements. The `fill` and `repeat` rules are slightly more complex, since they apply to a code block, which is made of instructions. We thus define a new `code_block`  rule that refers to `instruction` and finish implementing our rules:
-
-```ebnf
-instruction: MOVEMENT NUMBER            -> movement
-           | "c" COLOR [COLOR]          -> change_color
-           | "fill" code_block          -> fill
-           | "repeat" NUMBER code_block -> repeat
-
-code_block: "{" instruction "}"
-```
-
-We can now write the full grammar:
-
-```ebnf
-start: instruction+
-
-instruction: MOVEMENT NUMBER            -> movement
-            | "c" COLOR [COLOR]          -> change_color
-            | "fill" code_block          -> fill
-            | "repeat" NUMBER code_block -> repeat
-
-code_block: "{" instruction+ "}"
-
-MOVEMENT: "f"|"b"|"l"|"r"
-COLOR: LETTER+
-
-%import common.LETTER
-%import common.INT -> NUMBER
-%import common.WS
-%ignore WS
-```
-
-Notice the `start` rule, which defines the starting point of the grammar, i.e. the rule with which a program must start. This full grammars allows us to parse programs such as:
-
-```python
-c red yellow
-    fill { repeat 36 {
-        f200 l170
-    }}
-```
-
-The result of the parse, the parse tree, can then easily be translated into a Python program that uses the `turtle` library to draw a pattern.
-
-### Next steps
-
-This section provides a very brief overview of grammars and their possibilities. Check out the [Lark documentation](https://lark-parser.readthedocs.io/en/stable/index.html) for more thorough explanations and more examples.
diff --git a/docs/reference/generation/choices.md b/docs/reference/generation/choices.md
deleted file mode 100644
index aed5af5ab..000000000
--- a/docs/reference/generation/choices.md
+++ /dev/null
@@ -1,16 +0,0 @@
-# Multiple choices
-
-Oultines allows you to make sure the generated text is chosen between different options:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.choice(model, ["skirt", "dress", "pen", "jacket"])
-answer = generator("Pick the odd word out: skirt, dress, pen, jacket")
-
-```
-
-!!! Note "Performance"
-
-    `generation.choice` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate from the same list of choices several times make sure that you only call `generate.choice` once.
diff --git a/docs/reference/generation/format.md b/docs/reference/generation/format.md
deleted file mode 100644
index 749baa8b6..000000000
--- a/docs/reference/generation/format.md
+++ /dev/null
@@ -1,23 +0,0 @@
-# Type constraints
-
-We can ask completions to be restricted to valid python types:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.format(model, int)
-answer = generator("When I was 6 my sister was half my age. Now I’m 70 how old is my sister?")
-print(answer)
-# 67
-```
-
-The following types are currently available:
-
-- int
-- float
-- bool
-- datetime.date
-- datetime.time
-- datetime.datetime
-- We also provide [custom types](types.md)
diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
index 930ad9d22..c2168f9e3 100644
--- a/docs/reference/generation/generation.md
+++ b/docs/reference/generation/generation.md
@@ -4,34 +4,30 @@ title: Generation
 
 # Generation
 
-Once an [Outlines model](../models/models.md) is constructed you can use `outlines.generate` to generate text. Standard LLM generation is possible via `outlines.generate.text`, along with a variety of structured generation methods described below. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
+Once an [Outlines model](../models/models.md) is constructed you can use a `Generator` to generate text. The second argument to the `Generator` constructor is the output type. Standard LLM generation is possible by not passing any output type. Otherwise, you can provide a variety of different output types to benefit from structured generation. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
 
 Before generating text, you must construct an `outlines.model`. Example:
 
 ```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
 import outlines
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
 ```
+
 ### Text generator
 
 ```python
-generator = outlines.generate.text(model)
+generator = outlines.Generator(model)
 
 result = generator("Question: What's 2+2? Answer:", max_tokens=100)
 print(result)
 # The answer is 4
-
-# Outlines also supports streaming output
-stream = generator.stream("What's 2+2?", max_tokens=4)
-for i in range(5):
-	token = next(stream)
-	print(repr(token))
-# '2'
-# '+'
-# '2'
-# ' equals'
-# '4'
 ```
 
 ### [Multi-label classification](./choices.md)
@@ -40,9 +36,9 @@ Outlines allows you to do multi-label classification by guiding the model so it
 
 ```python
 import outlines
+from outlines.types import Choice
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = outlines.generate.choice(model, ["Blue", "Red", "Yellow"])
+generator = outlines.Generator(model, Choice(["Blue", "Red", "Yellow"]))
 
 color = generator("What is the closest color to Indigo? ")
 print(color)
@@ -59,6 +55,7 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
     from enum import Enum
     from pydantic import BaseModel, constr, conint
 
+    from outlines.types import JsonType
     import outlines
 
     class Armor(str, Enum):
@@ -73,8 +70,7 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
         armor: Armor
         strength: conint(gt=1, lt=100)
 
-    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-    generator = outlines.generate.json(model, Character)
+    generator = outlines.Generator(model, JsonType(Character))
 
     character = generator(
         "Generate a new character for my awesome game: "
@@ -88,6 +84,7 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
 
     ```python
     import outlines
+    from outlines.types import JsonType
 
     schema = """{
         "$defs": {
@@ -108,8 +105,7 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
         "type": "object"
     }"""
 
-    model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-    generator = outlines.generate.json(model, schema)
+    generator = outlines.Generator(model, JsonType(schema))
     character = generator(
         "Generate a new character for my awesome game: "
         + "name, age (between 1 and 99), armor and strength. "
@@ -129,7 +125,8 @@ Outlines also allows to generate text that is valid to any [context-free grammar
 Here we show a simple example of a grammar that defines arithmetic operations:
 
 ```python
-from outlines import models, generate
+import outlines
+from outlines.types import CFG
 
 arithmetic_grammar = """
     ?start: sum
@@ -152,8 +149,7 @@ arithmetic_grammar = """
     %ignore WS_INLINE
 """
 
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = generate.cfg(model, arithmetic_grammar, max_tokens=100)
+generator = outlines.Generator(model, CFG(arithmetic_grammar, max_tokens=100))
 
 result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
 print(result)
@@ -164,10 +160,11 @@ print(result)
 EBNF grammars can be cumbersome to write. This is why Outlines provides grammar definitions in the `outlines.grammars.` module
 
 ```python
-from outlines import models, generate, grammars
+import outlines
+from outlines import grammars
+from outlines.types import CFG
 
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = generate.cfg(model, grammars.arithmetic, max_tokens=100)
+generator = outlines.Generator(model, CFG(grammars.arithmetic, max_tokens=100))
 
 result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
 print(result)
@@ -182,34 +179,17 @@ The available grammars are listed [here](https://github.com/dottxt-ai/outlines/t
 Slightly simpler, but no less useful, Outlines can generate text that is in the language of a [regular expression](https://www.regular-expressions.info/tutorial.html). For instance to force the model to generate IP addresses:
 
 ```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
+import outlines
+from outlines.types import Regex
 
 regex_str = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-generator = generate.regex(model, regex_str)
+generator = outlines.Generator(model, Regex(regex_str))
 
 result = generator("What is the IP address of localhost?\nIP: ")
 print(result)
 # 127.0.0.100
 ```
 
-### [Generate a given Python type](./types.md)
-
-We provide a shortcut to regex-structured generation for simple use cases. Pass a Python type to the `outlines.generate.format` function and the LLM will output text that matches this type:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-128k-instruct")
-generator = generate.format(model, int)
-
-result = generator("What is 2+2?")
-print(result)
-# 4
-```
-
-
 [jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
 [pydantic]: https://docs.pydantic.dev/latest
 [cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
diff --git a/docs/reference/generation/json.md b/docs/reference/generation/json.md
deleted file mode 100644
index 05dc69654..000000000
--- a/docs/reference/generation/json.md
+++ /dev/null
@@ -1,178 +0,0 @@
-# JSON structured generation
-
-Outlines can make any open source model return a JSON object that follows a structure that is specified by the user. This is useful whenever we want the output of the model to be processed by code downstream: code does not understand natural language but rather the structured language it has been programmed to understand.
-
-There are mostly two reasons why someone would want to get an output formatted as JSON from a LLM:
-
-1. Parse the answer (e.g. with Pydantic), store it somewhere, return it to a user, etc.
-2. Call a function with the result
-
-Outlines has you covered in both cases! Indeed, to define the structure of the JSON you want the model to follow you can either provide a Pydantic model, or a function. No need to duplicate code!
-
-## Using Pydantic
-
-Outlines can infer the structure of the output from a Pydantic model. The result is an instance of the model that contains the values returned by the LLM:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models, generate
-
-
-class User(BaseModel):
-    name: str
-    last_name: str
-    id: int
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.json(model, User)
-result = generator(
-    "Create a user profile with the fields name, last_name and id"
-)
-print(result)
-# User(name="John", last_name="Doe", id=11)
-```
-
-!!! Note "JSON and whitespaces"
-
-    By default Outlines prevents the model from generating json with syntactic newlines, tabs, or multiple spaces. The default `whitespace_pattern` is `r"[ ]?"`. Small models tend to enter an infinite repetition loop if the `whitespace_pattern` allows infinite spacing. If you would like to allow the model to generate multiple tabs, newlines, and spaces, you can set the whitespace pattern as follows:
-
-    ```python
-    generator = generate.json(model, User, whitespace_pattern=r"[\n\t ]*")
-    ```
-
-!!! Note "Performance"
-
-    `generation.json` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate several times with the same schema make sure that you only call `generate.json` once.
-
-!!! Tip "Custom types"
-
-    Outlines provides [custom Pydantic types](types.md) so you do not have to write regular expressions for common types, such as phone numbers or zip codes.
-
-## Using a JSON Schema
-
-Instead of a Pydantic model you can pass a string that represents a [JSON Schema](https://json-schema.org/) specification to `generate.json`:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models
-from outlines import generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
-schema = """
-{
-  "title": "User",
-  "type": "object",
-  "properties": {
-    "name": {"type": "string"},
-    "last_name": {"type": "string"},
-    "id": {"type": "integer"}
-  },
-  "required": ["name", "last_name", "id"]
-}
-"""
-
-generator = generate.json(model, schema)
-result = generator(
-    "Create a user profile with the fields name, last_name and id"
-)
-print(result)
-# User(name="John", last_name="Doe", id=11)
-```
-
-## From a function's signature
-
-Outlines can infer the structure of the output from the signature of a function. The result is a dictionary, and can be passed directly to the function using the usual dictionary expansion syntax `**`:
-
-```python
-from outlines import models
-from outlines import generate
-
-def add(a: int, b: int):
-    return a + b
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.json(model, add)
-result = generator("Return two integers named a and b respectively. a is odd and b even.")
-
-print(add(**result))
-# 3
-```
-
-A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
-
-
-## From a dynamic JSON schema builder - GenSON
-
-Outlines integrated [GenSON](https://github.com/wolverdude/GenSON) builders to be able to dynamicly declare JSON schemas. It can be used as follow:
-
-```python
-from genson import SchemaBuilder
-
-from outlines import models
-from outlines import generate
-
-builder = SchemaBuilder()
-builder.add_schema({"type": "object", "properties": {}})
-builder.add_object({"name": "Toto", "age": 5})
-
-model = models.transformers(
-    "HuggingFaceTB/SmolLM2-135M",
-    device="auto",
-)
-generator = generate.json(model, builder)
-
-res = generator("Return a json of a young boy")
-print(res)
-# {"name": "Ben", "age": 10}
-```
-
-Anytime you are updating the schema through the builder, you need to redifine the outline generator to include these changes. From the the previous example:
-
-```python
-from genson import SchemaBuilder
-
-from outlines import models
-from outlines import generate
-
-builder = SchemaBuilder()
-builder.add_schema({"type": "object", "properties": {}})
-builder.add_object({"name": "Toto", "age": 5})
-
-model = models.transformers(
-    "HuggingFaceTB/SmolLM2-135M",
-    device="auto",
-)
-generator = generate.json(model, builder)
-
-res = generator("Return a json of a young boy")
-print(res)
-# {"name": "Ben", "age": 10}
-
-builder.add_object({"hobby": "sports"})
-generator = generate.json(model, builder)
-
-res = generator("Return a json of a youg boy whose hobby is coding")
-print(res)
-# {"name": "Ben", "age": 10, "hobby": "coding"}
-```
-
-!!! Note
-
-    Beware of [GenSON](https://github.com/wolverdude/GenSON)'s behavior regarding dynamic amending of schemas through their builder. Here is an example of how you could lose `required` informations and generate json with missing fields:
-
-    ```python
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"name": "Toto", "age": 5})
-
-    print(builder.to_schema())
-    # {'$schema': 'http://json-schema.org/schema#', 'type': 'object', 'properties': {'name': {'type': 'string'}, 'age': {'type': 'integer'}}, 'required': ['age', 'name']}
-
-    builder.add_object({"hobby": "sport"})
-    print(builder.to_schema())
-    # {'name': {'type': 'string'}, 'age': {'type': 'integer'}, 'hobby': {'type': 'string'}}}
-    ```
diff --git a/docs/reference/generation/regex.md b/docs/reference/generation/regex.md
deleted file mode 100644
index 23df35567..000000000
--- a/docs/reference/generation/regex.md
+++ /dev/null
@@ -1,27 +0,0 @@
-# Regular expressions
-
-Outlines can guarantee that the text generated by the LLM will be valid to a regular expression:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
-generator = generate.regex(
-    model,
-    r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)",
-)
-
-prompt = "What is the IP address of the Google DNS servers? "
-answer = generator(prompt, max_tokens=30)
-
-print(answer)
-# What is the IP address of the Google DNS servers?
-# 2.2.6.1
-```
-
-If you find yourself using `generate.regex` to restrict the answers' type you can take a look at [type-structured generation](types.md) instead.
-
-!!! Note "Performance"
-
-    `generate.regex` computes an index that helps Outlines guide generation. This can take some time, but only needs to be done once. If you want to generate several times using the same regular expression make sure that you only call `generate.regex` once.
diff --git a/docs/reference/generation/types.md b/docs/reference/generation/types.md
index eb6d7382b..ba712171e 100644
--- a/docs/reference/generation/types.md
+++ b/docs/reference/generation/types.md
@@ -35,7 +35,7 @@ You can use these types in Pydantic schemas for JSON-structured generation:
 ```python
 from pydantic import BaseModel
 
-from outlines import models, generate, types
+from outlines import models, types, Generator
 
 # Specify the locale for types
 locale = types.locale("us")
@@ -46,8 +46,12 @@ class Client(BaseModel):
     zip_code: locale.ZipCode
 
 
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.json(model, Client)
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
+generator = Generator(model, types.JsonType(Client))
 result = generator(
     "Create a client profile with the fields name, phone_number and zip_code"
 )
@@ -55,24 +59,6 @@ print(result)
 # name='Tommy' phone_number='129-896-5501' zip_code='50766'
 ```
 
-Or simply with `outlines.generate.format`:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models, generate, types
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.format(model, types.locale("us").PhoneNumber)
-result = generator(
-    "Return a US Phone number: "
-)
-print(result)
-# 334-253-2630
-```
-
-
 We plan on adding many more custom types. If you have found yourself writing regular expressions to generate fields of a given type, or if you could benefit from more specific types don't hesite to [submit a PR](https://github.com/dottxt-ai/outlines/pulls) or [open an issue](https://github.com/dottxt-ai/outlines/issues/new/choose).
 
 
diff --git a/docs/reference/models/dottxt.md b/docs/reference/models/dottxt.md
index 0437891f8..2ac07de6c 100644
--- a/docs/reference/models/dottxt.md
+++ b/docs/reference/models/dottxt.md
@@ -19,7 +19,7 @@ You can either create a `Generator` object and call it afterward:
 
 ```python
 from outlines.models import Dottxt
-from outlines.generate import Generator
+from outlines import Generator
 from pydantic import BaseModel
 
 class Character(BaseModel):
diff --git a/docs/reference/models/ollama.md b/docs/reference/models/ollama.md
index 4cc3bcd2d..f6805f55c 100644
--- a/docs/reference/models/ollama.md
+++ b/docs/reference/models/ollama.md
@@ -32,7 +32,7 @@ You can find the list of available models on the [Ollama library](https://ollama
 As with other models, you can either first create a `Generator` object and then call it
 ```python
 from outlines.models import Ollama
-from outlines.generate import Generator
+from outlines import Generator
 
 model = Ollama("tinyllama")
 generator = Generator(model)
diff --git a/docs/reference/text.md b/docs/reference/text.md
deleted file mode 100644
index f36a67da0..000000000
--- a/docs/reference/text.md
+++ /dev/null
@@ -1,86 +0,0 @@
-# Text generation
-
-Outlines provides a unified interface to generate text with many language models, API-based and local. The same pattern is used throughout the library:
-
-1. Instantiate a generator by calling `outlines.generate.text` with the model to be used.
-2. Call the generator with the prompt and (optionally) some generation parameters.
-
-
-```python
-from outlines import models, generate
-
-model = models.openai("gpt-4o-mini")
-generator = generate.text(model)
-answer = generator("What is 2+2?")
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-answer = generator("What is 2+2?")
-```
-
-By default Outlines uses the multinomial sampler with `temperature=1`. See [this section](samplers.md) to learn how to use different samplers.
-
-## Streaming
-
-Outlines allows you to stream the model's response by calling the `.stream` method of the generator with the prompt:
-
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
-tokens = generator.stream("What is 2+2?")
-for token in tokens:
-    print(token)
-```
-
-## Parameters
-
-### Limit the number of tokens generated
-
-To limit the number of tokens generated you can pass the `max_tokens` positional argument to the generator:
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
-answer = generator("What is 2+2?", 5)
-answer = generator("What is 2+2?", max_tokens=5)
-```
-
-### Stop after a given string is generated
-
-You can also ask the model to stop generating text after a given string has been generated, for instance a period or a line break. You can pass a string or a line of string for the `stop_at` argument:
-
-
-```python
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
-
-answer = generator("What is 2+2?", stop_at=".")
-answer = generator("What is 2+2?", stop_at=[".", "\n"])
-```
-
-*The stopping string will be included in the response.*
-
-
-### Seed the generation
-
-It can be useful to seed the generation in order to get reproducible results:
-
-```python
-import torch
-from outlines import models, generate
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
-seed = 789001
-
-answer = generator("What is 2+2?", seed=seed)
-```
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 4aacb6c7e..3c736ca5b 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -9,7 +9,7 @@
 from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function
-from outlines.generate import Generator
+from outlines.generator import Generator
 from outlines.templates import Template, prompt
 
 from outlines.models import (
diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
index bc91b6124..1e29217de 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -1,89 +1,2 @@
-from dataclasses import dataclass
-from typing import Any, Optional, Union, cast, get_args
-
-from outlines.models import BlackBoxModel, SteerableModel
-from outlines.processors import CFGLogitsProcessor, RegexLogitsProcessor
-from outlines.types import CFG, Choice, JsonType, List, Regex
-
 from .api import SequenceGeneratorAdapter
-from .cfg import cfg
-from .choice import choice
-from .format import format
 from .fsm import fsm
-from .json import json
-from .regex import regex
-from .text import text
-
-
-@dataclass
-class BlackBoxGenerator:
-    """Represents a generator for which we don't control constrained generation.
-
-    Attributes
-    ----------
-    model
-        An instance of a model wrapper.
-    output_type
-        The output type.
-
-    """
-
-    model: BlackBoxModel
-    output_type: Optional[Union[JsonType, List, Choice, Regex]] = None
-
-    def __post_init__(self):
-        if isinstance(self.output_type, CFG):
-            raise NotImplementedError(
-                "CFG generation is not supported for API-based models"
-            )
-
-    def __call__(self, prompt, **inference_kwargs):
-        return self.model.generate(prompt, self.output_type, **inference_kwargs)
-
-
-@dataclass
-class SteerableGenerator:
-    """Represents a generator for which we control constrained generation.
-
-    We use this class to keep track of the logits processor which can be quite
-    expensive to build.
-
-    Attributes
-    ----------
-    model
-        An instance of a model wrapper.
-    output_type
-        The output type.
-
-    """
-
-    model: SteerableModel
-    output_type: Optional[Union[JsonType, List, Choice, Regex]]
-
-    def __post_init__(self):
-        if self.output_type is None:
-            self.logits_processor = None
-        else:
-            if isinstance(self.output_type, CFG):
-                cfg_string = self.output_type.definition
-                self.logits_processor = CFGLogitsProcessor(
-                    cfg_string, self.model.tokenizer
-                )
-            else:
-                regex_string = self.output_type.to_regex()
-                self.logits_processor = RegexLogitsProcessor(
-                    regex_string, self.model.tokenizer
-                )
-
-    def __call__(self, prompt, **inference_kwargs):
-        return self.model.generate(prompt, self.logits_processor, **inference_kwargs)
-
-
-def Generator(
-    model: Union[SteerableModel, BlackBoxModel],
-    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG]] = None,
-):
-    if isinstance(model, BlackBoxModel):  # type: ignore
-        return BlackBoxGenerator(model, output_type)  # type: ignore
-    else:
-        return SteerableGenerator(model, output_type)  # type: ignore
diff --git a/outlines/generate/cfg.py b/outlines/generate/cfg.py
deleted file mode 100644
index 0ffbda6a3..000000000
--- a/outlines/generate/cfg.py
+++ /dev/null
@@ -1,54 +0,0 @@
-from functools import singledispatch
-
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import LlamaCpp, OpenAI, TransformersMultiModal
-from outlines.samplers import Sampler, multinomial
-
-
-@singledispatch
-def cfg(
-    model, cfg_str: str, sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    """Generate text in the language of a Context-Free Grammar
-
-    Arguments
-    ---------
-    model:
-        An `outlines.model` instance.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGeneratorAdapter` instance that generates text.
-
-    """
-    from outlines.processors import CFGLogitsProcessor
-
-    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@cfg.register(TransformersMultiModal)
-def cfg_vision(model, cfg_str: str, sampler: Sampler = multinomial()):
-    from outlines.processors import CFGLogitsProcessor
-
-    logits_processor = CFGLogitsProcessor(cfg_str, tokenizer=model.tokenizer)
-    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@cfg.register(LlamaCpp)
-def cfg_llamacpp(model, cfg_str: str, sampler: Sampler = multinomial()):
-    raise NotImplementedError("Not yet available due to bug in llama_cpp tokenizer")
-
-
-@cfg.register(OpenAI)
-def cfg_openai(model, cfg_str: str, sampler: Sampler = multinomial()):
-    raise NotImplementedError(
-        "Cannot use grammar-structured generation with an OpenAI model"
-        + "due to the limitations of the OpenAI API."
-    )
diff --git a/outlines/generate/choice.py b/outlines/generate/choice.py
deleted file mode 100644
index afb998f52..000000000
--- a/outlines/generate/choice.py
+++ /dev/null
@@ -1,59 +0,0 @@
-import json as pyjson
-import re
-from enum import Enum
-from functools import singledispatch
-from typing import Callable, List, Union
-
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-from outlines.fsm.json_schema import get_schema_from_enum
-from outlines.generate.api import SequenceGeneratorAdapter
-from outlines.models import OpenAI
-from outlines.samplers import Sampler, multinomial
-
-from .json import json
-from .regex import regex
-
-
-@singledispatch
-def choice(
-    model, choices: Union[List[str], type[Enum]], sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    if isinstance(choices, type(Enum)):
-        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
-    else:
-        choices = [re.escape(choice) for choice in choices]  # type: ignore
-        regex_str = r"(" + r"|".join(choices) + r")"
-
-    generator = regex(model, regex_str, sampler)
-    if isinstance(choices, type(Enum)):
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    else:
-        generator.format_sequence = lambda x: x
-
-    return generator
-
-
-@choice.register(OpenAI)
-def choice_openai(
-    model: OpenAI, choices: List[str], sampler: Sampler = multinomial()
-) -> Callable:
-    """
-    Call OpenAI API with response_format of a dict:
-    {"result": <one of choices>}
-    """
-
-    choices_schema = pyjson.dumps(
-        {
-            "type": "object",
-            "properties": {"result": {"type": "string", "enum": choices}},
-            "additionalProperties": False,
-            "required": ["result"],
-        }
-    )
-    generator = json(model, choices_schema, sampler)
-
-    def generate_choice(*args, **kwargs):
-        return generator(*args, **kwargs)["result"]
-
-    return generate_choice
diff --git a/outlines/generate/format.py b/outlines/generate/format.py
deleted file mode 100644
index 56ed10ac1..000000000
--- a/outlines/generate/format.py
+++ /dev/null
@@ -1,48 +0,0 @@
-from functools import singledispatch
-
-from outlines.fsm.types import python_types_to_regex
-from outlines.generate.api import SequenceGeneratorAdapter
-from outlines.models import OpenAI
-from outlines.samplers import Sampler, multinomial
-
-from .regex import regex
-
-
-@singledispatch
-def format(
-    model, python_type, sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    """Generate structured data that can be parsed as a Python type.
-
-    Parameters
-    ----------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    python_type:
-        A Python type. The output of the generator must be parseable into
-        this type.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGenerator` instance that generates text constrained by the Python type
-    and translates this text into the corresponding type.
-
-    """
-    regex_str, format_fn = python_types_to_regex(python_type)
-    regex_str = regex_str.pattern
-    generator = regex(model, regex_str, sampler)
-    generator.format_sequence = format_fn
-
-    return generator
-
-
-@format.register(OpenAI)
-def format_openai(model, python_type, sampler: Sampler = multinomial()):
-    raise NotImplementedError(
-        "Cannot use Python type-structured generation with an OpenAI model"
-        + " due to the limitations of the OpenAI API."
-    )
diff --git a/outlines/generate/regex.py b/outlines/generate/regex.py
deleted file mode 100644
index 477a7bf5e..000000000
--- a/outlines/generate/regex.py
+++ /dev/null
@@ -1,66 +0,0 @@
-from functools import singledispatch
-
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import OpenAI, TransformersMultiModal
-from outlines.samplers import Sampler, multinomial
-from outlines.types import Regex
-
-
-@singledispatch
-def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
-    """Generate structured text in the language of a regular expression.
-
-    Parameters
-    ----------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    regex_str:
-        The regular expression that the output must follow.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGeneratorAdapter` instance that generates text constrained by the
-    regular expression.
-
-    """
-    from outlines.processors import RegexLogitsProcessor
-
-    if isinstance(regex_str, Regex):
-        regex_str = regex_str.pattern
-
-    logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@regex.register(TransformersMultiModal)
-def regex_vision(
-    model,
-    regex_str: str | Regex,
-    sampler: Sampler = multinomial(),
-):
-    from outlines.processors import RegexLogitsProcessor
-
-    if isinstance(regex_str, Regex):
-        regex_str = regex_str.pattern
-
-    logits_processor = RegexLogitsProcessor(regex_str, tokenizer=model.tokenizer)
-    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@regex.register(OpenAI)
-def regex_openai(
-    model: OpenAI,
-    regex_str: str,
-    sampler: Sampler = multinomial(),
-):
-    raise NotImplementedError(
-        "Cannot use regex-structured generation with an OpenAI model"
-        + "due to the limitations of the OpenAI API."
-    )
diff --git a/outlines/generate/text.py b/outlines/generate/text.py
deleted file mode 100644
index e2e4cbe5d..000000000
--- a/outlines/generate/text.py
+++ /dev/null
@@ -1,50 +0,0 @@
-from functools import singledispatch
-
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import OpenAI, TransformersMultiModal
-from outlines.samplers import Sampler, multinomial
-
-
-@singledispatch
-def text(model, sampler: Sampler = multinomial()) -> SequenceGeneratorAdapter:
-    """Generate text with a `Transformer` model.
-
-    Note
-    ----
-    Python 3.11 allows dispatching on Union types and
-    this should greatly simplify the code.
-
-    Arguments
-    ---------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-
-    Returns
-    -------
-    A `SequenceGeneratorAdapter` instance that generates text.
-
-    """
-    return SequenceGeneratorAdapter(model, None, sampler)
-
-
-@text.register(TransformersMultiModal)
-def text_vision(model, sampler: Sampler = multinomial()):
-    return VisionSequenceGeneratorAdapter(model, None, sampler)
-
-
-@text.register(OpenAI)
-def text_openai(model: OpenAI, sampler: Sampler = multinomial()) -> OpenAI:
-    if not isinstance(sampler, multinomial):
-        raise NotImplementedError(
-            r"The OpenAI API does not support any other sampling algorithm "
-            + "than the multinomial sampler."
-        )
-
-    return model
diff --git a/outlines/generator.py b/outlines/generator.py
new file mode 100644
index 000000000..051308ed4
--- /dev/null
+++ b/outlines/generator.py
@@ -0,0 +1,80 @@
+from dataclasses import dataclass
+from typing import Any, Optional, Union, cast, get_args
+
+from outlines.models import BlackBoxModel, SteerableModel
+from outlines.processors import CFGLogitsProcessor, RegexLogitsProcessor
+from outlines.types import CFG, Choice, JsonType, List, Regex
+
+
+@dataclass
+class BlackBoxGenerator:
+    """Represents a generator for which we don't control constrained generation.
+
+    Attributes
+    ----------
+    model
+        An instance of a model wrapper.
+    output_type
+        The output type.
+
+    """
+
+    model: BlackBoxModel
+    output_type: Optional[Union[JsonType, List, Choice, Regex]] = None
+
+    def __post_init__(self):
+        if isinstance(self.output_type, CFG):
+            raise NotImplementedError(
+                "CFG generation is not supported for API-based models"
+            )
+
+    def __call__(self, prompt, **inference_kwargs):
+        return self.model.generate(prompt, self.output_type, **inference_kwargs)
+
+
+@dataclass
+class SteerableGenerator:
+    """Represents a generator for which we control constrained generation.
+
+    We use this class to keep track of the logits processor which can be quite
+    expensive to build.
+
+    Attributes
+    ----------
+    model
+        An instance of a model wrapper.
+    output_type
+        The output type.
+
+    """
+
+    model: SteerableModel
+    output_type: Optional[Union[JsonType, List, Choice, Regex]]
+
+    def __post_init__(self):
+        if self.output_type is None:
+            self.logits_processor = None
+        else:
+            if isinstance(self.output_type, CFG):
+                cfg_string = self.output_type.definition
+                self.logits_processor = CFGLogitsProcessor(
+                    cfg_string, self.model.tokenizer
+                )
+            else:
+                regex_string = self.output_type.to_regex()
+                self.logits_processor = RegexLogitsProcessor(
+                    regex_string, self.model.tokenizer
+                )
+
+    def __call__(self, prompt, **inference_kwargs):
+        return self.model.generate(prompt, self.logits_processor, **inference_kwargs)
+
+
+def Generator(
+    model: Union[SteerableModel, BlackBoxModel],
+    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG]] = None,
+):
+    if isinstance(model, BlackBoxModel):  # type: ignore
+        return BlackBoxGenerator(model, output_type)  # type: ignore
+    else:
+        return SteerableGenerator(model, output_type)  # type: ignore
diff --git a/outlines/models/base.py b/outlines/models/base.py
index 9d30522c4..62c2e8dcb 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -67,7 +67,7 @@ def __call__(self, model_input, output_type=None, **inference_kwargs):
         ```
 
         """
-        from outlines.generate import Generator
+        from outlines import Generator
 
         return Generator(self, output_type)(model_input, **inference_kwargs)
 
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index 6c3ea612b..618827917 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -7,7 +7,7 @@
 from dottxt.client import Dottxt as DottxtClient
 
 import outlines
-from outlines.generate import Generator
+from outlines import Generator
 from outlines.models.dottxt import Dottxt
 from outlines.types import JsonType
 

From 670b49e97f892087f54bfe66a32543615ad1db7b Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Mon, 3 Mar 2025 10:54:35 +0100
Subject: [PATCH 251/505] Remove the vectorize function and the associted
 base;py file

---
 outlines/__init__.py |   2 -
 outlines/base.py     | 299 -------------------------------------------
 tests/test_base.py   | 275 ---------------------------------------
 3 files changed, 576 deletions(-)
 delete mode 100644 outlines/base.py
 delete mode 100644 tests/test_base.py

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 3c736ca5b..dbeb376d0 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -6,7 +6,6 @@
 import outlines.processors
 import outlines.types
 from outlines.types import Choice, Regex, JsonType
-from outlines.base import vectorize
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function
 from outlines.generator import Generator
@@ -48,6 +47,5 @@
     "Regex",
     "prompt",
     "Template",
-    "vectorize",
     "grammars",
 ] + model_list
diff --git a/outlines/base.py b/outlines/base.py
deleted file mode 100644
index 29d42c54c..000000000
--- a/outlines/base.py
+++ /dev/null
@@ -1,299 +0,0 @@
-import asyncio
-import builtins
-import functools
-import inspect
-from typing import Callable, Optional
-
-import numpy as np
-
-# Import required functions based on NumPy version
-np_major_version = int(np.__version__.split(".")[0])
-if np_major_version >= 2:
-    from numpy.lib._function_base_impl import (
-        _calculate_shapes,
-        _parse_gufunc_signature,
-        _parse_input_dimensions,
-        _update_dim_sizes,
-    )
-else:
-    from numpy.lib.function_base import (
-        _calculate_shapes,
-        _parse_gufunc_signature,
-        _parse_input_dimensions,
-        _update_dim_sizes,
-    )
-
-# Allow nested loops for running in notebook. We don't enable it globally as it
-# may interfere with other libraries that use asyncio.
-if hasattr(builtins, "__IPYTHON__"):
-    try:
-        import nest_asyncio
-
-        nest_asyncio.apply()
-    except ImportError:
-        print(
-            "Couldn't patch nest_asyncio because it's not installed. Running in the notebook might be have issues"
-        )
-
-
-class vectorize:
-    """Returns an object that acts like a function but takes arrays as an input.
-
-    The vectorized function evaluates `func` over successive tuples of the input
-    chararrays and returns a single NumPy chararrays or a tuple of NumPy chararrays.
-
-    Its behavior is similar to NumPy's `vectorize` for Python functions: the function
-    being vectorized is executed in a `for` loop. Coroutines, however, are executed
-    concurrently.
-
-    Part of the code was adapted from `numpy.lib.function_base`.
-
-    """
-
-    def __init__(self, func: Callable, signature: Optional[str] = None):
-        self.func = func
-        self.signature = signature
-        self.is_coroutine_fn = inspect.iscoroutinefunction(func)
-
-        functools.update_wrapper(self, func)
-
-        if signature is not None:
-            # Parse the signature string into a Python data structure.
-            # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
-            self._in_and_out_core_dimensions = _parse_gufunc_signature(signature)
-        else:
-            self._in_and_out_core_dimensions = None
-
-    def __call__(self, *args, **kwargs):
-        """Call the vectorized function."""
-        if not args and not kwargs:
-            return self.call_thunk()
-        elif self.signature is not None:
-            return self.call_with_signature(*args, **kwargs)
-        else:
-            return self.call_no_signature(*args, **kwargs)
-
-    def call_thunk(self):
-        """Call a vectorized thunk.
-
-        Thunks have no arguments and can thus be called directly.
-
-        """
-        if self.is_coroutine_fn:
-            loop = asyncio.new_event_loop()
-            try:
-                outputs = loop.run_until_complete(self.func())
-            finally:
-                loop.close()
-        else:
-            outputs = self.func()
-
-        return outputs
-
-    def call_no_signature(self, *args, **kwargs):
-        """Call functions and coroutines when no signature is specified.
-
-        When no signature is specified we assume that all of the function's
-        inputs and outputs are scalars (core dimension of zero). We first
-        broadcast the input arrays, then iteratively apply the function over the
-        elements of the broadcasted arrays and finally reshape the results to
-        match the input shape.
-
-        Functions are executed in a for loop, coroutines are executed
-        concurrently.
-
-        """
-        # Convert args and kwargs to arrays
-        args = [np.array(arg) for arg in args]
-        kwargs = {key: np.array(value) for key, value in kwargs.items()}
-
-        # Broadcast args and kwargs
-        broadcast_shape = np.broadcast(*args, *list(kwargs.values())).shape
-        args = [np.broadcast_to(arg, broadcast_shape) for arg in args]
-        kwargs = {
-            key: np.broadcast_to(value, broadcast_shape)
-            for key, value in kwargs.items()
-        }
-
-        # Execute functions in a loop, and coroutines concurrently
-        if self.is_coroutine_fn:
-            outputs = self.vectorize_call_coroutine(broadcast_shape, args, kwargs)
-        else:
-            outputs = self.vectorize_call(broadcast_shape, args, kwargs)
-
-        # `outputs` is a flat array or a tuple of flat arrays. We reshape the arrays
-        # to match the input shape.
-        outputs = [
-            results if isinstance(results, tuple) else (results,) for results in outputs
-        ]
-        outputs = tuple(
-            [np.asarray(x).reshape(broadcast_shape).squeeze() for x in zip(*outputs)]
-        )
-        outputs = tuple([x.item() if np.ndim(x) == 0 else x for x in outputs])
-
-        n_results = len(list(outputs))
-
-        return outputs[0] if n_results == 1 else outputs
-
-    def call_with_signature(self, *args, **kwargs):
-        """Call functions and coroutines when a signature is specified."""
-        input_core_dims, output_core_dims = self._in_and_out_core_dimensions
-
-        # Make sure that the numbers of arguments passed is compatible with
-        # the signature.
-        num_args = len(args) + len(kwargs)
-        if num_args != len(input_core_dims):
-            raise TypeError(
-                "wrong number of positional arguments: "
-                "expected %r, got %r" % (len(input_core_dims), len(args))
-            )
-
-        # Convert args and kwargs to arrays
-        args = [np.asarray(arg) for arg in args]
-        kwargs = {key: np.array(value) for key, value in kwargs.items()}
-
-        # Find the arguments' broadcast shape, and map placeholder
-        # variables in the signature to the number of dimensions
-        # they correspond to given the arguments.
-        broadcast_shape, dim_sizes = _parse_input_dimensions(
-            args + list(kwargs.values()), input_core_dims
-        )
-
-        # Calculate the shape to which each of the arguments should be broadcasted
-        # and reshape them accordingly.
-        input_shapes = _calculate_shapes(broadcast_shape, dim_sizes, input_core_dims)
-        args = [
-            np.broadcast_to(arg, shape, subok=True)
-            for arg, shape in zip(args, input_shapes)
-        ]
-        kwargs = {
-            key: np.broadcast_to(value, broadcast_shape)
-            for key, value in kwargs.items()
-        }
-
-        n_out = len(output_core_dims)
-
-        if self.is_coroutine_fn:
-            outputs = self.vectorize_call_coroutine(broadcast_shape, args, kwargs)
-        else:
-            outputs = self.vectorize_call(broadcast_shape, args, kwargs)
-
-        outputs = [
-            results if isinstance(results, tuple) else (results,) for results in outputs
-        ]
-
-        flat_outputs = list(zip(*outputs))
-        n_results = len(flat_outputs)
-
-        if n_out != n_results:
-            raise ValueError(
-                f"wrong number of outputs from the function, expected {n_out}, got {n_results}"
-            )
-
-        # The number of dimensions of the outputs are not necessarily known in
-        # advance. The following iterates over the results and updates the
-        # number of dimensions of the outputs accordingly.
-        for results, core_dims in zip(flat_outputs, output_core_dims):
-            for result in results:
-                _update_dim_sizes(dim_sizes, result, core_dims)
-
-        # Calculate the shape to which each of the outputs should be broadcasted
-        # and reshape them.
-        shapes = _calculate_shapes(broadcast_shape, dim_sizes, output_core_dims)
-        outputs = tuple(
-            [
-                np.hstack(results).reshape(shape).squeeze()
-                for shape, results in zip(shapes, zip(*outputs))
-            ]
-        )
-        outputs = tuple([x.item() if np.ndim(x) == 0 else x for x in outputs])
-
-        return outputs[0] if n_results == 1 else outputs
-
-    def vectorize_call(self, broadcast_shape, args, kwargs):
-        """Run the function in a for loop.
-
-        A possible extension would be to parallelize the calls.
-
-        Parameters
-        ----------
-        broadcast_shape
-            The brodcast shape of the input arrays.
-        args
-            The function's broadcasted arguments.
-        kwargs
-            The function's broadcasted keyword arguments.
-
-        """
-        outputs = []
-        for index in np.ndindex(*broadcast_shape):
-            current_args = tuple(arg[index] for arg in args)
-            current_kwargs = {key: value[index] for key, value in kwargs.items()}
-            outputs.append(self.func(*current_args, **current_kwargs))
-
-        return outputs
-
-    def vectorize_call_coroutine(self, broadcast_shape, args, kwargs):
-        """Run coroutines concurrently.
-
-        Creates as many tasks as needed and executes them in a new event
-        loop.
-
-        Parameters
-        ----------
-        broadcast_shape
-            The brodcast shape of the input arrays.
-        args
-            The function's broadcasted arguments.
-        kwargs
-            The function's broadcasted keyword arguments.
-
-        """
-
-        async def create_and_gather_tasks():
-            tasks = []
-            for index in np.ndindex(*broadcast_shape):
-                current_args = tuple(arg[index] for arg in args)
-                current_kwargs = {key: value[index] for key, value in kwargs.items()}
-                tasks.append(self.func(*current_args, **current_kwargs))
-
-            outputs = await asyncio.gather(*tasks)
-
-            return outputs
-
-        loop = asyncio.new_event_loop()
-        try:
-            outputs = loop.run_until_complete(create_and_gather_tasks())
-        finally:
-            loop.close()
-
-        return outputs
-
-
-def _update_arrays_type(arrays, results):
-    """Update the dtype of arrays.
-
-    String arrays contain strings of fixed length. Here they are initialized with
-    the type of the first results, so that if the next results contain longer
-    strings they will be truncated when added to the output arrays. Here we
-    update the type if the current results contain longer strings than in the
-    current output array.
-
-    Parameters
-    ----------
-    arrays
-        Arrays that contain the vectorized function's results.
-    results
-        The current output of the function being vectorized.
-
-    """
-
-    updated_arrays = []
-    for array, result in zip(arrays, results):
-        if array.dtype.type == np.str_:
-            if array.dtype < np.array(result).dtype:
-                array = array.astype(np.array(result).dtype)
-
-        updated_arrays.append(array)
-
-    return tuple(updated_arrays)
diff --git a/tests/test_base.py b/tests/test_base.py
deleted file mode 100644
index 24c84f7d9..000000000
--- a/tests/test_base.py
+++ /dev/null
@@ -1,275 +0,0 @@
-import numpy as np
-import pytest
-from numpy.testing import assert_array_equal
-
-from outlines.base import vectorize
-
-
-def test_vectorize_docstring():
-    def test(x):
-        """This is a test docstring"""
-        return x
-
-    fn = vectorize(test)
-    assert fn.__doc__ == "This is a test docstring"
-    assert fn.__name__ == "test"
-
-    async def test(x):
-        """This is a test docstring"""
-        return x
-
-    fn = vectorize(test)
-    assert fn.__doc__ == "This is a test docstring"
-    assert fn.__name__ == "test"
-
-
-def test_vectorize_thunk():
-    def thunk():
-        """A thunk"""
-        return 1
-
-    fn = vectorize(thunk)
-    assert fn() == 1
-
-    async def thunk():
-        """A thunk"""
-        return 1
-
-    fn = vectorize(thunk)
-    assert fn() == 1
-
-
-def test_vectorize_scalar_simple():
-    def passthrough(x):
-        """A passthrough function."""
-        return x
-
-    fn = vectorize(passthrough)
-
-    out_vector = fn(["one", "two", "three"])
-    assert_array_equal(out_vector, ["one", "two", "three"])
-
-    out_array = fn([["one", "two"], ["three", "four"]])
-    assert_array_equal(out_array, [["one", "two"], ["three", "four"]])
-
-    async def passthrough(x):
-        """A passthrough function."""
-        return x
-
-    fn = vectorize(passthrough)
-    assert fn.__doc__ == "A passthrough function."
-
-    out_vector = fn(["one", "two", "three"])
-    assert_array_equal(out_vector, ["one", "two", "three"])
-
-    out_array = fn([["one", "two"], ["three", "four"]])
-    assert_array_equal(out_array, [["one", "two"], ["three", "four"]])
-
-
-def test_vectorize_scalar_multiple_outputs():
-    def passthrough_multiple_outputs(x):
-        return x, x
-
-    fn = vectorize(passthrough_multiple_outputs)
-    out = fn(["one", "two", "three"])
-    assert len(out) == 2
-    assert_array_equal(out[0], ["one", "two", "three"])
-    assert_array_equal(out[1], ["one", "two", "three"])
-
-    async def passthrough_multiple_outputs(x):
-        return x, x
-
-    fn = vectorize(passthrough_multiple_outputs)
-    out = fn(["one", "two", "three"])
-    assert len(out) == 2
-    assert_array_equal(out[0], ["one", "two", "three"])
-    assert_array_equal(out[1], ["one", "two", "three"])
-
-
-def test_vectorize_scalar_args():
-    def passthrough_args(*args):
-        """A passthrough function."""
-        result = ""
-        for arg in args:
-            result += arg
-        return result
-
-    fn = vectorize(passthrough_args)
-
-    out_array = fn(["one", "two"], ["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn([["one", "two"], ["three", "four"]], ["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-    async def passthrough_args(*args):
-        """A passthrough function."""
-        result = ""
-        for arg in args:
-            result += arg
-        return result
-
-    fn = vectorize(passthrough_args)
-
-    out_array = fn(["one", "two"], ["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn([["one", "two"], ["three", "four"]], ["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-
-def test_vectorize_scalar_kwargs():
-    def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs)
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-    async def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs)
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-
-def test_signature_invalid():
-    with pytest.raises(ValueError):
-        vectorize(lambda x: x, "(m,)->()")
-
-    with pytest.raises(TypeError, match="wrong number of positional"):
-        fn = vectorize(lambda x, y: x, "()->()")
-        fn(1, 2)
-
-    with pytest.raises(ValueError, match="wrong number of outputs"):
-
-        def test_multioutput(x, y):
-            return x, y
-
-        fn = vectorize(test_multioutput, "(),()->()")
-        fn(1, 2)
-
-
-def test_vectorize_simple():
-    def test(x):
-        return x
-
-    fn = vectorize(test, "(m)->(m)")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-
-def test_vectorize_kwargs():
-    def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs, "(),()->()")
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-    async def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs, "(),()->()")
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-
-def test_vectorize_reduce():
-    def test(x):
-        return x[0]
-
-    fn = vectorize(test, "(m)->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, ["one", "four"])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["o", "t", "t"], ["f", "f", "s"]])
-
-
-def test_vectorize_expand():
-    def test(x):
-        return np.array([x for i in range(3)])
-
-    fn = vectorize(test, "()->(s)")
-    out = fn(["one", "two"])
-    assert_array_equal(out, [["one", "one", "one"], ["two", "two", "two"]])
-
-
-def test_vectorize_coroutine_simple():
-    async def test(x):
-        return x
-
-    fn = vectorize(test, "(m)->(m)")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-
-def test_vectorize_coroutine_reduce():
-    async def test(x):
-        return x[0]
-
-    fn = vectorize(test, "(m)->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, ["one", "four"])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["o", "t", "t"], ["f", "f", "s"]])
-
-
-def test_vectorize_coroutine_expand():
-    async def test(x):
-        return np.array([x for i in range(3)])
-
-    fn = vectorize(test, "()->(s)")
-    out = fn(["one", "two"])
-    assert_array_equal(out, [["one", "one", "one"], ["two", "two", "two"]])

From a8f247dfb32ecb849262c4d84034cec0c648b36d Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Mon, 3 Mar 2025 11:45:41 +0100
Subject: [PATCH 252/505] Move the fsm output type into Generator and remove
 fsm.py

---
 docs/reference/generation/custom_fsm_ops.md | 12 +++++---
 outlines/generate/__init__.py               |  1 -
 outlines/generate/fsm.py                    | 31 ---------------------
 outlines/generator.py                       | 16 +++++++++--
 4 files changed, 21 insertions(+), 39 deletions(-)
 delete mode 100644 outlines/generate/fsm.py

diff --git a/docs/reference/generation/custom_fsm_ops.md b/docs/reference/generation/custom_fsm_ops.md
index 5c4be96fa..0dc7713ed 100644
--- a/docs/reference/generation/custom_fsm_ops.md
+++ b/docs/reference/generation/custom_fsm_ops.md
@@ -12,8 +12,8 @@ In the following example we build a fsm which recognizes only the strings valid
 
 ```python
 import interegular
-from outlines import models, generate
-
+from outlines import from_transformers, Generator
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 list_of_strings_pattern = """\["[^"\s]*"(?:,"[^"\s]*")*\]"""
 pink_elephant_pattern = """.*(pink|elephant).*"""
@@ -28,9 +28,13 @@ difference_fsm_fsm.accepts('["a","pink","elephant"]')
 difference_fsm_fsm.accepts('["a","blue","donkey"]')
 # True
 
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = from_transformers(
+    AutoModelForCausalLM.from_pretrained(model_name),
+    AutoTokenizer.from_pretrained(model_name)
+)
 
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.fsm(model, difference_fsm)
+generator = Generator(model, difference_fsm)
 response = generator("Don't talk about pink elephants")
 ```
 
diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
index 1e29217de..8f0cb009f 100644
--- a/outlines/generate/__init__.py
+++ b/outlines/generate/__init__.py
@@ -1,2 +1 @@
 from .api import SequenceGeneratorAdapter
-from .fsm import fsm
diff --git a/outlines/generate/fsm.py b/outlines/generate/fsm.py
deleted file mode 100644
index 81f974bd2..000000000
--- a/outlines/generate/fsm.py
+++ /dev/null
@@ -1,31 +0,0 @@
-from functools import singledispatch
-
-import interegular
-
-from outlines.fsm.guide import RegexGuide
-from outlines.generate.api import (
-    SequenceGeneratorAdapter,
-    VisionSequenceGeneratorAdapter,
-)
-from outlines.models import TransformersMultiModal
-from outlines.samplers import Sampler, multinomial
-
-
-@singledispatch
-def fsm(
-    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
-) -> SequenceGeneratorAdapter:
-    from outlines.processors import GuideLogitsProcessor
-
-    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
-    return SequenceGeneratorAdapter(model, logits_processor, sampler)
-
-
-@fsm.register(TransformersMultiModal)
-def fsm_vision(model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()):
-    from outlines.processors import GuideLogitsProcessor
-
-    guide = RegexGuide.from_interegular_fsm(fsm, model.tokenizer)
-    logits_processor = GuideLogitsProcessor(tokenizer=model.tokenizer, guide=guide)
-    return VisionSequenceGeneratorAdapter(model, logits_processor, sampler)
diff --git a/outlines/generator.py b/outlines/generator.py
index 051308ed4..7a806b3a7 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -1,8 +1,11 @@
 from dataclasses import dataclass
 from typing import Any, Optional, Union, cast, get_args
 
+import interegular
+
+from outlines.fsm.guide import RegexGuide
 from outlines.models import BlackBoxModel, SteerableModel
-from outlines.processors import CFGLogitsProcessor, RegexLogitsProcessor
+from outlines.processors import CFGLogitsProcessor, GuideLogitsProcessor, RegexLogitsProcessor
 from outlines.types import CFG, Choice, JsonType, List, Regex
 
 
@@ -27,6 +30,10 @@ def __post_init__(self):
             raise NotImplementedError(
                 "CFG generation is not supported for API-based models"
             )
+        elif isinstance(self.output_type, interegular.fsm.FSM):
+            raise NotImplementedError(
+                "FSM generation is not supported for API-based models"
+            )
 
     def __call__(self, prompt, **inference_kwargs):
         return self.model.generate(prompt, self.output_type, **inference_kwargs)
@@ -49,7 +56,7 @@ class SteerableGenerator:
     """
 
     model: SteerableModel
-    output_type: Optional[Union[JsonType, List, Choice, Regex]]
+    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG, interegular.fsm.FSM]]
 
     def __post_init__(self):
         if self.output_type is None:
@@ -60,6 +67,9 @@ def __post_init__(self):
                 self.logits_processor = CFGLogitsProcessor(
                     cfg_string, self.model.tokenizer
                 )
+            elif isinstance(self.output_type, interegular.fsm.FSM):
+                guide = RegexGuide.from_interegular_fsm(self.output_type, self.model.tokenizer)
+                self.logits_processor = GuideLogitsProcessor(tokenizer=self.model.tokenizer, guide=guide)
             else:
                 regex_string = self.output_type.to_regex()
                 self.logits_processor = RegexLogitsProcessor(
@@ -72,7 +82,7 @@ def __call__(self, prompt, **inference_kwargs):
 
 def Generator(
     model: Union[SteerableModel, BlackBoxModel],
-    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG]] = None,
+    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG, interegular.fsm.FSM]] = None,
 ):
     if isinstance(model, BlackBoxModel):  # type: ignore
         return BlackBoxGenerator(model, output_type)  # type: ignore

From 6ba06f703dc981ec436b3d4c5767a522d5fbafc7 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin@dottxt.co>
Date: Mon, 3 Mar 2025 12:08:08 +0100
Subject: [PATCH 253/505] Remove Sampler, SamplingParameters and
 GenerationParameters

---
 docs/api/samplers.md          |   1 -
 docs/reference/samplers.md    | 178 -----------------
 mkdocs.yml                    |   1 -
 outlines/__init__.py          |   1 -
 outlines/function.py          |   3 +-
 outlines/generate/__init__.py |   1 -
 outlines/generate/api.py      | 230 ----------------------
 outlines/models/vllm.py       |   1 -
 outlines/samplers.py          | 353 ----------------------------------
 tests/generate/test_api.py    |  33 ----
 tests/test_samplers.py        | 277 --------------------------
 11 files changed, 1 insertion(+), 1078 deletions(-)
 delete mode 100644 docs/api/samplers.md
 delete mode 100644 docs/reference/samplers.md
 delete mode 100644 outlines/generate/__init__.py
 delete mode 100644 outlines/generate/api.py
 delete mode 100644 outlines/samplers.py
 delete mode 100644 tests/generate/test_api.py
 delete mode 100644 tests/test_samplers.py

diff --git a/docs/api/samplers.md b/docs/api/samplers.md
deleted file mode 100644
index 2b9b34234..000000000
--- a/docs/api/samplers.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.samplers
diff --git a/docs/reference/samplers.md b/docs/reference/samplers.md
deleted file mode 100644
index 9cba5cefa..000000000
--- a/docs/reference/samplers.md
+++ /dev/null
@@ -1,178 +0,0 @@
-# Samplers
-
-Outlines offers different sequence sampling algorithms, and we will integrate more in the future. You can read [this blog post](https://huggingface.co/blog/how-to-generate) for an overview of the different sampling algorithm.
-
-Samplers provide control over the sampling process, allowing you to influence the output of the model. This can include controlling randomness (temperature), biasing towards certain tokens (top-k, top-p), or sequence generation (beam search).
-
-## Multinomial sampling
-
-[Multinomial sampling](https://en.wikipedia.org/wiki/Multinomial_distribution) is the default sampling algorithm in Outlines.
-
-As an example, suppose we have only two possible tokens: "H" and "T". For a fixed prompt such as "Flip a coin, did you get heads or tails?" The language model calculates probability for each token:
-
-| Token | Probability |
-|-------|-------------|
-| "H"   | 0.5         |
-| "T"   | 0.5         |
-
-You'd expect to receive "H" 50% of the time and "T" 50% of the time.
-
-### Parameters
-
-- `samples`: Number of samples to generate (default: 1)
-- `top_k`: Only consider the top k tokens (optional)
-- `top_p`: Only consider the top tokens with cumulative probability >= p (optional)
-- `temperature`: Controls randomness of sampling (optional)
-
-### Default behavior
-
-Outlines defaults to the multinomial sampler without top-p or top-k sampling, and temperature equal to 1.
-
-Not specifying a sampler is equivalent to:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial()
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# 4
-```
-
-### Batching
-
-You can ask the generator to take multiple samples by passing the number of samples when initializing the sampler:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial(3)
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# [4, 4, 4]
-```
-
-If you ask multiple samples for a batch of prompts the returned array will be of shape `(num_samples, num_batches)`:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial(3)
-
-generator = generate.text(model, sampler)
-answer = generator(["What is 2+2?", "What is 3+3?"])
-
-print(answer)
-# [[4, 4, 4], [6, 6, 6]]
-```
-
-### Temperature
-
-You can control the temperature with
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.multinomial(3, temperature=0.5)
-
-generator = generate.text(model, sampler)
-answer = generator(["What is 2+2?", "What is 3+3?"])
-
-print(answer)
-```
-
-If you would like to use `temperature=0.0`, please use `sampler=samplers.greedy()` instead.
-
-### Top-k sampling
-
-You can ask Outlines to only consider the top-k logits at each step by specifying the value of the `top-k` keyword argument when initializing the sampler.
-
-```python
-sampler = samplers.multinomial(3, top_k=10)
-```
-
-### Top-p sampling
-
-You can ask Outlines to only consider the highest probability tokens such that their cumulative probability is greater than a threshold `p`. Specify the `top_p` keyword argument when initializing the sampler:
-
-
-```python
-sampler = samplers.multinomial(3, top_p=0.95)
-```
-
-## Greedy sampler
-
-Greedy sampling selects the token with the highest probability at each step. It's deterministic and always produces the same output for a given input.
-
-To use the greedy sampler, initialize the generator with the sampler:
-
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.greedy()
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# 4
-```
-
-You cannot ask for multiple samples with the greedy sampler since it does not clear what the result should be. Only the most likely token can be returned.
-
-
-## Beam Search
-
-Beam search maintains multiple candidate sequences at each step, potentially finding better overall sequences than greedy or multinomial sampling.
-
-To use Beam Search, initialize the generator with the sampler:
-
-```python
-from outlines import models, generate, samplers
-
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-sampler = samplers.beam_search(beams=5)
-
-generator = generate.text(model, sampler)
-answer = generator("What is 2+2?")
-
-print(answer)
-# 4
-```
-!!! Warning "Compatibility"
-
-    Only models from the `transformers`  and `exllamav2 ` libraries are compatible with Beam Search.
-
-### Parameters
-
-- `beams`: Number of beams to use (default: 1)
-
-## Sampler Comparison
-
-Here's a table comparing the different samplers:
-
-| Sampler | Pros | Cons | Use Cases |
-|---------|------|------|-----------|
-| Greedy | Deterministic, fast | May produce repetitive text | When you need consistent, predictable output |
-| Multinomial | Balances exploration and exploitation | Results may vary between runs | General-purpose text generation, creative tasks |
-| Beam Search | Can find globally better sequences | More computationally expensive | When sequence quality is critical, e.g., translation |
-
-For most use cases, we recommend using the default [multinomial sampler](#multinomial-sampling).
diff --git a/mkdocs.yml b/mkdocs.yml
index 430840ac1..8ec7ac838 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -135,7 +135,6 @@ nav:
           - Chat templating: reference/chat_templating.md
           - Regex DSL: reference/regex_dsl.md
           - Text: reference/text.md
-          - Samplers: reference/samplers.md
           - Structured generation:
               - How does it work?: reference/generation/structured_generation_explanation.md
               - Classification: reference/generation/choices.md
diff --git a/outlines/__init__.py b/outlines/__init__.py
index dbeb376d0..1fce45e96 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -1,6 +1,5 @@
 """Outlines is a Generative Model Programming Framework."""
 
-import outlines.generate
 import outlines.grammars
 import outlines.models
 import outlines.processors
diff --git a/outlines/function.py b/outlines/function.py
index cabaf1fef..250d1784a 100644
--- a/outlines/function.py
+++ b/outlines/function.py
@@ -7,7 +7,6 @@
 import outlines
 
 if TYPE_CHECKING:
-    from outlines.generate import SequenceGeneratorAdapter
     from outlines.templates import Template
 
 
@@ -25,7 +24,7 @@ class Function:
     prompt_template: "Template"
     schema: Union[str, Callable, object]
     model_name: str
-    generator: Optional["SequenceGeneratorAdapter"] = None
+    generator = None
 
     @classmethod
     def from_github(cls, program_path: str, function_name: str = "fn"):
diff --git a/outlines/generate/__init__.py b/outlines/generate/__init__.py
deleted file mode 100644
index 8f0cb009f..000000000
--- a/outlines/generate/__init__.py
+++ /dev/null
@@ -1 +0,0 @@
-from .api import SequenceGeneratorAdapter
diff --git a/outlines/generate/api.py b/outlines/generate/api.py
deleted file mode 100644
index d9833eb12..000000000
--- a/outlines/generate/api.py
+++ /dev/null
@@ -1,230 +0,0 @@
-import datetime
-from copy import copy
-from dataclasses import dataclass
-from typing import TYPE_CHECKING, Any, Iterator, List, Optional, Union
-
-if TYPE_CHECKING:
-    import torch
-
-FormattedOutput = Union[
-    str, int, float, bool, datetime.date, datetime.time, datetime.datetime
-]
-
-
-@dataclass(frozen=True)
-class GenerationParameters:
-    """Generation parameters used in Outlines' public API."""
-
-    max_tokens: Optional[int]
-    stop_at: Optional[Union[str, List[str]]]
-    seed: Optional[int]
-
-
-@dataclass(frozen=True)
-class SamplingParameters:
-    """Sampling parameters available in Outlines."""
-
-    sampler: str
-    num_samples: int = 1
-    top_p: Optional[float] = None
-    top_k: Optional[int] = None
-    temperature: Optional[float] = None
-
-
-class SequenceGeneratorAdapter:
-    """Class used to unify the interface to the model providers'
-    generation functions.
-
-    Attributes
-    ----------
-    model
-        The wrapped model.
-    logits_processor
-        The logits processor to use to generate text.
-    sampler
-        The sampler to use to generate text.
-
-    """
-
-    def __init__(self, model, logits_processor, sampler):
-        self.model = model
-        self.logits_processor = logits_processor
-
-        self.sampling_params = sampler.sampling_params
-
-    def prepare_generation_parameters(
-        self,
-        max_tokens: Optional[int],
-        stop_at: Optional[Union[str, List[str]]],
-        seed: Optional[int],
-    ):
-        if isinstance(stop_at, str):
-            stop_at = [stop_at]
-
-        generation_params = GenerationParameters(
-            max_tokens,
-            stop_at,
-            seed,
-        )
-
-        return generation_params
-
-    def format_sequence(self, sequence: str) -> FormattedOutput:
-        """Translate the generated sequence to another type.
-
-        This method is for instance overridden when generating JSON to either
-        return a dictionnary or a Pydantic model.
-
-        Parameters
-        ----------
-        sequence
-            A generated sequences.
-
-        Returns
-        -------
-        The formatted sequence.
-
-        """
-        return sequence
-
-    def _format(self, sequences):
-        """Apply formatting to every string in a completion."""
-        if isinstance(sequences, list):
-            return [self._format(sequence) for sequence in sequences]
-        else:
-            return self.format_sequence(sequences)
-
-    def __call__(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """Generate text from a prompt of list of prompts."""
-
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-
-        completions = self.model.generate(
-            prompts,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-        return self._format(completions)
-
-    def stream(
-        self,
-        prompts: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """Return a text generator from a prompt or a list of prompts."""
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-        return self.model.stream(
-            prompts,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-
-class VisionSequenceGeneratorAdapter(SequenceGeneratorAdapter):
-    def __call__(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[str, Any],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """
-        Generate text from a prompt of list of prompts.
-
-        Media: A URI to construct media or media object itself. Used as AutoProcessor argument.
-        """
-        prompts, media = self._validate_prompt_media_types(prompts, media)
-
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-
-        completions = self.model.generate(
-            prompts,
-            media,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-        return self._format(completions)
-
-    def stream(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: List[Union[str, Any, List[Union[str, Any]]]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        """Return a text generator from a prompt or a list of prompts."""
-        prompts, media = self._validate_prompt_media_types(prompts, media)
-        generation_params = self.prepare_generation_parameters(
-            max_tokens, stop_at, seed
-        )
-        return self.model.stream(
-            prompts,
-            media,
-            generation_params,
-            copy(self.logits_processor),
-            self.sampling_params,
-            **model_specific_params,
-        )
-
-    @classmethod
-    def _validate_prompt_media_types(
-        cls,
-        prompts: Union[str, List[str]],
-        media: Union[str, Any, List[Union[str, Any]]],
-    ) -> Union[Any, List[Any]]:
-        """
-        Prepare media as PIL.Image and ensure for every prompt str there is one List[PIL.Image]
-        """
-
-        def valid_types(prompts, media):
-            from PIL import Image  # type: ignore
-
-            if isinstance(prompts, list):
-                if not isinstance(media, list) or len(prompts) != len(media):
-                    return False
-                for subprompt, submedia in zip(prompts, media):
-                    if not isinstance(subprompt, str) or not all(
-                        isinstance(m, Image.Image) for m in submedia
-                    ):
-                        return False
-            elif isinstance(prompts, str):
-                if not all(isinstance(m, Image.Image) for m in media):
-                    return False
-            return True
-
-        if not valid_types(prompts, media):
-            raise TypeError(
-                "Expected (prompts, media) to be of type "
-                "(str, List[Image])), or (List[str], List[List[Image]]) "
-                f"instead got prompts={prompts}, media={media}"
-            )
-
-        return prompts, media
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 3f10f6433..58121562e 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -2,7 +2,6 @@
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, List, Optional, Union
 
-from outlines.generate.api import GenerationParameters, SamplingParameters
 from outlines.models.base import Model, ModelTypeAdapter
 
 if TYPE_CHECKING:
diff --git a/outlines/samplers.py b/outlines/samplers.py
deleted file mode 100644
index 3fef673b1..000000000
--- a/outlines/samplers.py
+++ /dev/null
@@ -1,353 +0,0 @@
-import math
-from dataclasses import dataclass
-from typing import TYPE_CHECKING, Callable, Optional, Protocol, Tuple
-
-if TYPE_CHECKING:
-    import torch
-
-
-class Sampler(Protocol):
-    samples: int
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        rng: "torch.Generator",
-    ) -> "torch.DoubleTensor": ...
-
-
-@dataclass(frozen=True)
-class SamplingParameters:
-    """Sampling parameters available in Outlines."""
-
-    sampler: str
-    num_samples: int = 1
-    top_p: Optional[float] = None
-    top_k: Optional[int] = None
-    temperature: Optional[float] = None
-
-
-class GreedySampler:
-    """Greedy Sampling algorithm.
-
-    Greedy sampling consists in choosing the token with the largest
-    likelihood at every step.
-
-    We don't allow more than one sample. We could attribute this a meaning, for
-    instance the k-th sample represents the k-th most likely token. In which
-    case it would be equivalent to beam search without the sequence weights.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(self):
-        self.samples = 1
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        _,
-    ) -> "torch.DoubleTensor":
-        """Call the greedy sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
-        next_token_ids = torch.argmax(logprobs, dim=-1, keepdim=True)
-
-        ancestors = torch.arange(
-            next_token_logits.shape[0], device=next_token_logits.device
-        )
-        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
-
-        return next_token_ids, ancestors, weights
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("greedy", self.samples, None, None, 0.0)
-
-
-greedy = GreedySampler
-
-
-class MultinomialSampler:
-    """Multinomial sampling algorithm.
-
-    Multinomial sampling consists in randomly sampling the next token assuming
-    its distribution is a Categorical distribution parametrized by the
-    next-token logits.
-
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(
-        self,
-        samples: int = 1,
-        *,
-        top_k: Optional[int] = None,
-        top_p: Optional[float] = None,
-        temperature: Optional[float] = None,
-    ):
-        self.samples = samples
-        self.top_k = top_k
-        self.top_p = top_p
-        self.temperature = temperature
-
-        self.logits_processors = []
-        if top_k is not None:
-            self.logits_processors.append(keep_top_k_logits(top_k))
-        elif top_p is not None:
-            self.logits_processors.append(keep_top_p_logits(top_p))
-
-        if temperature is not None:
-            self.logits_processors.append(rescale_logits(temperature))
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        rng: "torch.Generator",
-    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
-        """Call the multinomial sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        altered_next_token_logits = next_token_logits
-        for logit_processor in self.logits_processors:
-            altered_next_token_logits = logit_processor(next_token_logits)
-
-        probs = torch.nn.functional.softmax(altered_next_token_logits, dim=-1)
-        next_token_ids = torch.multinomial(probs, num_samples=1, generator=rng)
-
-        logprobs = torch.nn.functional.log_softmax(altered_next_token_logits, dim=-1)
-        ancestors = torch.arange(
-            altered_next_token_logits.shape[0], device=next_token_logits.device
-        )
-        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
-
-        return next_token_ids, ancestors, weights
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters(
-            "multinomial",
-            self.samples,
-            self.top_p,
-            self.top_k,
-            self.temperature,
-        )
-
-
-multinomial = MultinomialSampler
-
-
-def keep_top_k_logits(k: int) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that masks logits values smaller than the top `k` ones.
-
-    Parameters
-    ----------
-    k
-        The ranking below which logit values are replaced by `-math.inf`.
-
-    """
-    import torch
-
-    if not isinstance(k, int) or k < 1:
-        raise ValueError(f"`k` must be a strictly positive integers, got {k} instead.")
-
-    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
-        num_to_keep = min(k, logits.size(-1))
-        mask_idx = logits < torch.topk(logits, num_to_keep)[0][..., -1, None]
-        return logits.masked_fill(mask_idx, -math.inf)
-
-    return logits_processor
-
-
-def keep_top_p_logits(p: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that masks the lowest probability tokens whose
-    cumulative probability is below a certain threshold.
-
-    Parameters
-    ----------
-    p
-        The value of the threshold. We keep the highest probability tokens whose
-        cumulative distribution is greater than or equal to `p` and mask the
-        others. Its value must be between 0 (excluded) and 1 (included).
-
-    """
-    import torch
-
-    if p <= 0.0 or p > 1.0:
-        raise ValueError(
-            f"`p` must be a floating point number between 0 (excluded) and 1 (included), got {p} instead."
-        )
-
-    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
-        sorted_logits, sorted_idx = torch.sort(logits, descending=False)
-        cumulative_probabilties = torch.nn.functional.softmax(
-            sorted_logits, dim=-1
-        ).cumsum(dim=-1)
-
-        sorted_masked_idx = cumulative_probabilties <= (1 - p)
-        mask_idx = torch.scatter(sorted_masked_idx, 1, sorted_idx, sorted_masked_idx)
-        return logits.masked_fill(mask_idx, -math.inf)
-
-    return logits_processor
-
-
-def rescale_logits(temperature: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that rescales the token probabilities exponentially.
-
-    Parameters
-    ----------
-    temperature
-        The value by which we rescale the logits.
-
-    """
-
-    if not isinstance(temperature, float) or temperature < 0.0:
-        raise ValueError(
-            f"`temperature` must be a strictly positive floating point number, got {temperature} instead."
-        )
-    elif temperature == 0.0:
-        raise ValueError(
-            "Please use the greedy sampler instead of setting the temperature to 0."
-        )
-
-    def logits_processor(logits: "torch.Tensor") -> "torch.Tensor":
-        return logits / temperature
-
-    return logits_processor
-
-
-class BeamSearchSampler:
-    """Beam Search sampling algorithm.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence. Equivalent to the
-        number of beams.
-    """
-
-    def __init__(self, beams: int = 1):
-        self.samples = beams
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        _,
-    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
-        """Call the beam search sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
-        weights = logprobs + sequence_weights.unsqueeze(1).expand_as(next_token_logits)
-
-        # Flatten scores to (n_batch, n_samples * vocab_size)
-        # and find the top-k weights for each batch.
-        batch_size = next_token_logits.shape[0] // self.samples
-        vocab_size = next_token_logits.shape[-1]
-        weights = weights.view(batch_size, self.samples * vocab_size)
-
-        # If the weights are all equal to 0 we are at the beginning of the search
-        # and thus only need to sample from one set of token logits for each
-        # batch.
-        if torch.all(sequence_weights == 0):
-            weights = weights[:, :vocab_size]
-
-        weights, indices = torch.topk(
-            weights, self.samples, dim=1, largest=True, sorted=True
-        )
-
-        ancestors = torch.div(indices, vocab_size, rounding_mode="floor")
-        next_token_ids = indices % vocab_size
-
-        # Re-shape the weights, next_token_ids and ancestors to (n_batch * n_samples, 1)
-        first_batch_idx = torch.arange(
-            0, batch_size * self.samples, self.samples, device=next_token_logits.device
-        ).unsqueeze(1)
-        ancestors = ancestors + first_batch_idx
-
-        ancestors = ancestors.view(self.samples * batch_size)
-        weights = weights.view(self.samples * batch_size)
-        next_token_ids = next_token_ids.view(self.samples * batch_size, 1)
-
-        return next_token_ids, ancestors, weights
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
-
-
-beam_search = BeamSearchSampler
diff --git a/tests/generate/test_api.py b/tests/generate/test_api.py
deleted file mode 100644
index 7188022f5..000000000
--- a/tests/generate/test_api.py
+++ /dev/null
@@ -1,33 +0,0 @@
-from io import BytesIO
-from urllib.request import urlopen
-
-import pytest
-from PIL import Image  # type: ignore
-
-from outlines.generate.api import VisionSequenceGeneratorAdapter
-
-IMG_URI = "https://upload.wikimedia.org/wikipedia/en/a/a9/Example.jpg"
-PIL_IMG = Image.open(BytesIO(urlopen(IMG_URI).read())).convert("RGB")
-
-
-@pytest.mark.parametrize(
-    "prompts,media,type_error",
-    [
-        ("single prompt", [PIL_IMG], False),
-        (["prompt0", "prompt1"], [[PIL_IMG], [PIL_IMG]], False),
-        ("single prompt", [PIL_IMG, PIL_IMG], False),
-        (["prompt0", "prompt1"], [[PIL_IMG, PIL_IMG], [PIL_IMG]], False),
-        ("single prompt", "this isn't an image, it's a string", True),
-        ("single prompt", PIL_IMG, True),
-        (["prompt0", "prompt1"], [PIL_IMG], True),
-        (["prompt0", "prompt1"], [[PIL_IMG]], True),
-        (["prompt0", "prompt1"], [[[PIL_IMG]], [[PIL_IMG]]], True),
-    ],
-)
-def test_vision_sequence_generator_validate_types(prompts, media, type_error):
-    """Ensure inputs are validated correctly"""
-    if type_error:
-        with pytest.raises(TypeError):
-            VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
-    else:
-        VisionSequenceGeneratorAdapter._validate_prompt_media_types(prompts, media)
diff --git a/tests/test_samplers.py b/tests/test_samplers.py
deleted file mode 100644
index 10a7be26f..000000000
--- a/tests/test_samplers.py
+++ /dev/null
@@ -1,277 +0,0 @@
-import math
-
-import pytest
-import torch
-
-from outlines.samplers import (
-    BeamSearchSampler,
-    GreedySampler,
-    MultinomialSampler,
-    beam_search,
-    greedy,
-    keep_top_k_logits,
-    keep_top_p_logits,
-    multinomial,
-    rescale_logits,
-)
-
-
-def compute_logprobs(logits):
-    return torch.nn.functional.log_softmax(logits, dim=-1)
-
-
-def test_aliases():
-    assert greedy == GreedySampler
-    assert multinomial == MultinomialSampler
-    assert beam_search == BeamSearchSampler
-
-
-def test_greedy():
-    sampler = GreedySampler()
-    logits = torch.tensor([[1.0, 2.0, 5.0]])
-    weights = torch.tensor([0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[2]]))
-    assert ancestors.equal(torch.tensor([0]))
-    assert weights.equal(logprobs[..., 2])
-
-    sampler = GreedySampler()
-    logits = torch.tensor([[10.0, 0.0, 3.0], [-math.inf, 2.0, 5.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [2]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
-
-    params = sampler.sampling_params
-    assert params.sampler == "greedy"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 0.0
-
-
-def test_multinomial():
-    rng = torch.Generator()
-    rng.manual_seed(239)
-
-    sampler = MultinomialSampler()
-    logits = torch.tensor([[1.0, 4.0, 5.0]])
-    weights = torch.tensor([0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[2]]))
-    assert ancestors.equal(torch.tensor([0]))
-    assert weights.equal(torch.tensor([logprobs[:, 2]]))
-
-    sampler = MultinomialSampler()
-    logits = torch.tensor([[10.0, 0.0, 9.0], [-math.inf, 4.0, 5.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [2]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
-
-    sampler = MultinomialSampler(samples=5, top_k=10, top_p=0.9, temperature=0.8)
-    params = sampler.sampling_params
-    assert params.sampler == "multinomial"
-    assert params.num_samples == 5
-    assert params.top_p == 0.9
-    assert params.top_k == 10
-    assert params.temperature == 0.8
-
-
-def test_multinomial_init():
-    sampler = MultinomialSampler()
-    assert sampler.logits_processors == []
-
-    sampler = MultinomialSampler(3)
-    assert sampler.logits_processors == []
-
-    sampler = MultinomialSampler(top_k=1)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_p=0.95)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_k=1, top_p=0.95)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(temperature=1.0)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_k=1, temperature=1.0)
-    assert len(sampler.logits_processors) == 2
-
-    sampler = MultinomialSampler(top_p=0.95, temperature=1.0)
-    assert len(sampler.logits_processors) == 2
-
-
-def test_top_k():
-    with pytest.raises(ValueError, match="`k` must be a strictly"):
-        keep_top_k_logits(-1)
-
-    with pytest.raises(ValueError, match="`k` must be a strictly"):
-        keep_top_k_logits(0.1)
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0]])
-
-    logits_processor = keep_top_k_logits(1)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
-
-    logits_processor = keep_top_k_logits(10)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[1.0, 2.0, 3.0, 4.0]]))
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
-    logits_processor = keep_top_k_logits(2)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [[-math.inf, -math.inf, 3.0, 4.0], [-math.inf, -math.inf, 7.0, 8.0]]
-        )
-    )
-
-
-def test_top_p():
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(-0.1)
-
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(0.0)
-
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(1.1)
-
-    logits = torch.tensor([[1.0, 1.01, 1.02, 4.0]])
-
-    logits_processor = keep_top_p_logits(0.1)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
-
-    logits_processor = keep_top_p_logits(0.95)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, 1.01, 1.02, 4.0]]))
-
-    logits_processor = keep_top_p_logits(1.0)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[1.0, 1.01, 1.02, 4.0]]))
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
-
-    logits_processor = keep_top_p_logits(0.1)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [-math.inf, -math.inf, -math.inf, 4.0],
-                [-math.inf, -math.inf, -math.inf, 8.0],
-            ]
-        )
-    )
-
-    logits_processor = keep_top_p_logits(0.95)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [-math.inf, 2.0, 3.0, 4.0],
-                [-math.inf, 6.0, 7.0, 8.0],
-            ]
-        )
-    )
-
-    logits_processor = keep_top_p_logits(1.0)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [1.0, 2.0, 3.0, 4.0],
-                [5.0, 6.0, 7.0, 8.0],
-            ]
-        )
-    )
-
-
-def test_rescale():
-    with pytest.raises(ValueError, match="`temperature` must"):
-        rescale_logits(1)
-
-    with pytest.raises(ValueError, match="`temperature` must"):
-        rescale_logits(-0.1)
-
-    with pytest.raises(ValueError, match="Please use the greedy sampler"):
-        rescale_logits(0.0)
-
-
-def test_beam_search():
-    # Two beams, single sequence
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
-    init_weights = torch.tensor([0, 1.0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [1]]))
-    assert ancestors.equal(torch.tensor([1, 0]))
-    assert weights.equal(
-        torch.tensor([init_weights[1] + logprobs[1][0], logprobs[0][1]])
-    )
-
-    # Make sure that initial samples are different
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [0.0, 1.0]])
-    init_weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[1], [0]]))
-    assert ancestors.equal(torch.tensor([0, 0]))
-    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[0][0]]))
-
-    # One beam, batch of two sequences. Reduces to Greedy Search.
-    sampler = BeamSearchSampler(1)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[1], [0]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[1][0]]))
-
-    # Two beams, batch of two sequences
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0], [3.0, 2.0], [0.0, 1.0]])
-    init_weights = torch.tensor([0, 0, 2.0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [1], [0], [1]]))
-    assert ancestors.equal(torch.tensor([1, 0, 2, 2]))
-    assert weights.equal(
-        torch.tensor(
-            [
-                logprobs[1][0],
-                logprobs[0][1],
-                init_weights[2] + logprobs[2][0],
-                init_weights[2] + logprobs[2][1],
-            ]
-        )
-    )
-
-    sampler = BeamSearchSampler(beams=3)
-    params = sampler.sampling_params
-    assert params.sampler == "beam_search"
-    assert params.num_samples == 3
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 1.0

From a59177bd976bc1301618f53befcc8f660e223cac Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 7 Mar 2025 15:09:03 +0100
Subject: [PATCH 254/505] Add stream method to Generator classes

---
 outlines/generator.py             |  6 +++
 outlines/models/anthropic.py      | 19 ++++++++++
 outlines/models/base.py           | 35 +++++++++++++++++
 outlines/models/dottxt.py         |  6 +++
 outlines/models/gemini.py         | 25 ++++++++++++
 outlines/models/llamacpp.py       | 20 +++++-----
 outlines/models/mlxlm.py          |  4 +-
 outlines/models/ollama.py         |  2 +-
 outlines/models/openai.py         | 20 ++++++++++
 outlines/models/transformers.py   |  2 +-
 outlines/models/vllm.py           |  2 +-
 tests/models/test_anthropic.py    | 23 +++++++----
 tests/models/test_dottxt.py       | 11 +++++-
 tests/models/test_gemini.py       | 63 +++++++++++++------------------
 tests/models/test_llamacpp.py     | 12 ++----
 tests/models/test_mlxlm.py        | 10 +++++
 tests/models/test_openai.py       | 11 +++++-
 tests/models/test_transformers.py |  5 +++
 tests/models/test_vllm.py         |  5 +++
 19 files changed, 210 insertions(+), 71 deletions(-)

diff --git a/outlines/generator.py b/outlines/generator.py
index 7a806b3a7..0238fa598 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -38,6 +38,9 @@ def __post_init__(self):
     def __call__(self, prompt, **inference_kwargs):
         return self.model.generate(prompt, self.output_type, **inference_kwargs)
 
+    def stream(self, prompt, **inference_kwargs):
+        return self.model.generate_stream(prompt, self.output_type, **inference_kwargs)
+
 
 @dataclass
 class SteerableGenerator:
@@ -79,6 +82,9 @@ def __post_init__(self):
     def __call__(self, prompt, **inference_kwargs):
         return self.model.generate(prompt, self.logits_processor, **inference_kwargs)
 
+    def stream(self, prompt, **inference_kwargs):
+        return self.model.generate_stream(prompt, self.logits_processor, **inference_kwargs)
+
 
 def Generator(
     model: Union[SteerableModel, BlackBoxModel],
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index c7ed702dd..5fee17e9a 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -107,6 +107,25 @@ def generate(
         )
         return completion.content[0].text
 
+    def generate_stream(
+        self, model_input: Union[str, Vision], output_type=None, **inference_kwargs
+    ):
+        messages = self.type_adapter.format_input(model_input)
+
+        if output_type is not None:
+            raise NotImplementedError(
+                f"The type {output_type} is not available with Anthropic."
+            )
+
+        completion = self.client.messages.stream(
+            **messages,
+            model=self.model_name,
+            **inference_kwargs,
+        )
+        for chunk in completion:
+            if chunk.type == "content_block_delta":
+                yield chunk.delta.text
+
 
 def from_anthropic(client: "AnthropicClient", model_name: str) -> Anthropic:
     return Anthropic(client, model_name)
diff --git a/outlines/models/base.py b/outlines/models/base.py
index 62c2e8dcb..a70fba250 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -71,12 +71,47 @@ def __call__(self, model_input, output_type=None, **inference_kwargs):
 
         return Generator(self, output_type)(model_input, **inference_kwargs)
 
+    def stream(self, model_input, output_type=None, **inference_kwargs):
+        """Stream a response from the model.
+
+        Users can use the `stream` method from the model directly, in which
+        case we will create a generator instance with the output type provided
+        and then invoke its `stream` method.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        for chunk in generator("prompt"):
+            print(chunk)
+        ```
+        and
+        ```python
+        for chunk in model.stream("prompt", Foo):
+            print(chunk)
+        ```
+
+        """
+        from outlines import Generator
+
+        return Generator(self, output_type).stream(model_input, **inference_kwargs)
+
     @abstractmethod
     def generate(self, model_input, output_type=None, **inference_kwargs):
         """Generate a response from the model.
 
         The output_type argument contains a logits processor for local models
         while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
+
+        """
+        ...
+
+    @abstractmethod
+    def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+        """Generate a stream of responses from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
 
         """
         ...
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index 576a4c4cd..c725222a6 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -92,6 +92,12 @@ def generate(self, model_input, output_type=None, **inference_kwargs):
         )
         return completion.data
 
+    def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+        raise NotImplementedError(
+            "Dottxt does not support streaming. Call the model/generator for "
+            + "regular generation instead."
+        )
+
 
 def from_dottxt(
     client: "DottxtClient",
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index f0fd8d78a..1f6a9e37d 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -114,12 +114,37 @@ def generate(
         generation_config = genai.GenerationConfig(
             **self.type_adapter.format_output_type(output_type)
         )
+
         completion = self.client.generate_content(
             generation_config=generation_config, **contents, **inference_kwargs
         )
 
         return completion.text
 
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Union[JsonType, Choice, List]] = None,
+        **inference_kwargs,
+    ):
+        import google.generativeai as genai
+
+        contents = self.type_adapter.format_input(model_input)
+        generation_config = genai.GenerationConfig(
+            **self.type_adapter.format_output_type(output_type)
+        )
+
+        stream = self.client.generate_content(
+            generation_config=generation_config,
+            **contents,
+            **inference_kwargs,
+            stream=True,
+        )
+
+        for chunk in stream:
+            if hasattr(chunk, "text") and chunk.text:
+                yield chunk.text
+
 
 def from_gemini(client: "GeminiClient"):
     return Gemini(client)
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 3657b669c..b761a926a 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -149,14 +149,14 @@ def __init__(self, model: "Llama"):
         self.model_type = "local"
         self.type_adapter = LlamaCppTypeAdapter()
 
-    def generate(self, model_input, logits_processor, **inference_kwargs):
+    def generate(self, model_input, output_type, **inference_kwargs):
         """Generate text using `llama-cpp-python`.
 
         Arguments
         ---------
         prompt
             A prompt.
-        logits_processor
+        output_type
             The logits processor to use when generating text.
         inference_kwargs
             The inference kwargs that can be passed to the `Llama.__call__` method
@@ -167,14 +167,14 @@ def generate(self, model_input, logits_processor, **inference_kwargs):
         The generated text.
 
         """
-        if isinstance(logits_processor, CFGLogitsProcessor):
+        if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
                 "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
             )
 
         completion = self.model(
             self.type_adapter.format_input(model_input),
-            logits_processor=self.type_adapter.format_output_type(logits_processor),
+            logits_processor=self.type_adapter.format_output_type(output_type),
             **inference_kwargs,
         )
         result = completion["choices"][0]["text"]
@@ -183,16 +183,16 @@ def generate(self, model_input, logits_processor, **inference_kwargs):
 
         return result
 
-    def stream(
-        self, model_input, logits_processor, **inference_kwargs
-    ) -> Iterator[str]:
+    def generate_stream(
+        self, model_input, output_type, **inference_kwargs
+    ):
         """Stream text using `llama-cpp-python`.
 
         Arguments
         ---------
         prompt
             A prompt.
-        logits_processor
+        output_type
             The logits processor to use when generating text.
         inference_kwargs
             The inference kwargs that can be passed to the `Llama.__call__` method
@@ -203,14 +203,14 @@ def stream(
         A generator that return strings.
 
         """
-        if isinstance(logits_processor, CFGLogitsProcessor):
+        if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
                 "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
             )
 
         generator = self.model(
             self.type_adapter.format_input(model_input),
-            logits_processor=self.type_adapter.format_output_type(logits_processor),
+            logits_processor=self.type_adapter.format_output_type(output_type),
             stream=True,
             **inference_kwargs,
         )
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index deeb42d59..2191dc20d 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -74,7 +74,7 @@ def __init__(
         self.tokenizer = TransformerTokenizer(tokenizer._tokenizer)
         self.type_adapter = MLXLMTypeAdapter()
 
-    def generate(self, model_input, output_type, **kwargs):
+    def generate(self, model_input, output_type=None, **kwargs):
         """Generate text using `mlx-lm`.
 
         Arguments
@@ -97,7 +97,7 @@ def generate(self, model_input, output_type, **kwargs):
             **kwargs,
         )
 
-    def stream(self, model_input, output_type, **kwargs):
+    def generate_stream(self, model_input, output_type=None, **kwargs):
         """Stream text using `mlx-lm`.
 
         Arguments
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 2eec91879..ee45660bd 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -79,7 +79,7 @@ def generate(self, model_input, output_type=None, **kwargs) -> str:
         )
         return response.response
 
-    def stream(self, model_input, output_type=None, **kwargs) -> Iterator[str]:
+    def generate_stream(self, model_input, output_type=None, **kwargs) -> Iterator[str]:
         response = self.client.generate(
             model=self.model_name,
             prompt=self.type_adapter.format_input(model_input),
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 58f96740d..80774a8d3 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -151,6 +151,26 @@ def generate(
 
         return result.choices[0].message.content
 
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Union[type[BaseModel], str]] = None,
+        **inference_kwargs,
+    ):
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+        stream = self.client.chat.completions.create(
+            model=self.model_name,
+            stream=True,
+            **messages,
+            **response_format,
+            **inference_kwargs
+        )
+
+        for chunk in stream:
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
 
 def from_openai(
     client: Union["OpenAIClient", "AzureOpenAIClient"], model_name: str
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index ceaf39b7d..61d646364 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -214,7 +214,7 @@ def generate(self, model_input, output_type, **inference_kwargs):
 
         return self._decode_generation(generated_ids)
 
-    def stream(self, model_input, output_type, **inference_kwargs):
+    def generate_stream(self, model_input, output_type, **inference_kwargs):
         """
         TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
         """
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 58121562e..cedf049e3 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -126,7 +126,7 @@ def generate(self, model_input, output_type, **inference_kwargs):
 
         return results
 
-    def stream(self, model_input, output_type, **inference_kwargs):
+    def generate_stream(self, model_input, output_type, **inference_kwargs):
         """Return a text generator.
 
         Streaming is not yet available for `vllm.LLM`.
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 8aef257cc..75be2e413 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -1,4 +1,5 @@
 import io
+from typing import Generator
 
 from anthropic import Anthropic as AnthropicClient
 import PIL
@@ -12,6 +13,11 @@
 MODEL_NAME = "claude-3-haiku-20240307"
 
 
+@pytest.fixture(scope="session")
+def model():
+    return Anthropic(AnthropicClient(), MODEL_NAME)
+
+
 def test_anthropic_wrong_inference_parameters():
     with pytest.raises(TypeError, match="got an unexpected"):
         model = Anthropic(AnthropicClient(), MODEL_NAME)
@@ -46,24 +52,20 @@ def __init__(self, foo):
 
 
 @pytest.mark.api_call
-def test_anthropic_simple_call():
-    model = Anthropic(AnthropicClient(), MODEL_NAME)
+def test_anthropic_simple_call(model):
     result = model.generate("Respond with one word. Not more.", max_tokens=1024)
     assert isinstance(result, str)
 
 
 @pytest.mark.xfail(reason="Anthropic requires the `max_tokens` parameter to be set")
 @pytest.mark.api_call
-def test_anthropic_direct_call():
-    model = Anthropic(AnthropicClient(), MODEL_NAME)
+def test_anthropic_direct_call(model):
     result = model("Respond with one word. Not more.", max_tokens=1024)
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
-def test_anthropic_simple_vision():
-    model = Anthropic(AnthropicClient(), MODEL_NAME)
-
+def test_anthropic_simple_vision(model):
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)
     if r.status_code == 200:
@@ -73,3 +75,10 @@ def test_anthropic_simple_vision():
         Vision("What does this logo represent?", image), max_tokens=1024
     )
     assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_anthopic_streaming(model):
+    result = model.stream("Respond with one word. Not more.", max_tokens=1024)
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index 618827917..8d649dfd0 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -69,7 +69,7 @@ def test_dottxt_wrong_inference_parameters(api_key):
 @pytest.mark.api_call
 def test_dottxt_direct_call(api_key):
     client = DottxtClient(api_key=api_key)
-    model = Dottxt(client, model_name="meta-llama/Llama-3.1-8B-Instruct")
+    model = Dottxt(client)
     result = model("Create a user", JsonType(User))
     assert "first_name" in json.loads(result)
 
@@ -77,7 +77,14 @@ def test_dottxt_direct_call(api_key):
 @pytest.mark.api_call
 def test_dottxt_generator_call(api_key):
     client = DottxtClient(api_key=api_key)
-    model = Dottxt(client, model_name="meta-llama/Llama-3.1-8B-Instruct")
+    model = Dottxt(client)
     generator = Generator(model, JsonType(User))
     result = generator("Create a user")
     assert "first_name" in json.loads(result)
+
+
+def test_dottxt_streaming(api_key):
+    client = DottxtClient(api_key=api_key)
+    model = Dottxt(client)
+    with pytest.raises(NotImplementedError, match="Dottxt does not support streaming"):
+        model.stream("Create a user", JsonType(User))
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index cc54bd71a..3560b310a 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -1,7 +1,7 @@
 import io
 import json
 from enum import Enum
-
+from typing import Generator
 import PIL
 import pytest
 import requests
@@ -17,9 +17,13 @@
 MODEL_NAME = "gemini-1.5-flash-latest"
 
 
-def test_gemini_wrong_inference_parameters():
+@pytest.fixture(scope="session")
+def model():
+    return Gemini(genai.GenerativeModel(MODEL_NAME))
+
+
+def test_gemini_wrong_inference_parameters(model):
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = Gemini(genai.GenerativeModel(MODEL_NAME))
         model.generate("prompt", foo=10)
 
 
@@ -31,23 +35,19 @@ def test_gemini_init_from_client():
 
 
 @pytest.mark.api_call
-def test_gemini_simple_call():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
+def test_gemini_simple_call(model):
     result = model.generate("Respond with one word. Not more.")
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
-def test_gemini_direct_call():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
+def test_gemini_direct_call(model):
     result = model("Respond with one word. Not more.")
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
-def test_gemini_simple_vision():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_vision(model):
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)
     if r.status_code == 200:
@@ -58,9 +58,7 @@ def test_gemini_simple_vision():
 
 
 @pytest.mark.api_call
-def test_gemini_simple_pydantic():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_pydantic(model):
     class Foo(BaseModel):
         bar: int
 
@@ -71,9 +69,7 @@ class Foo(BaseModel):
 
 @pytest.mark.xfail(reason="Vision models do not work with structured outputs.")
 @pytest.mark.api_call
-def test_gemini_simple_vision_pydantic():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_vision_pydantic(model):
     url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
     r = requests.get(url, stream=True)
     if r.status_code == 200:
@@ -89,9 +85,7 @@ class Logo(BaseModel):
 
 @pytest.mark.xfail(reason="Gemini seems to be unable to follow nested schemas.")
 @pytest.mark.api_call
-def test_gemini_nested_pydantic():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_nested_pydantic(model):
     class Bar(BaseModel):
         fu: str
 
@@ -110,9 +104,7 @@ class Foo(BaseModel):
     reason="The Gemini SDK's serialization method does not support Json Schema dictionaries."
 )
 @pytest.mark.api_call
-def test_gemini_simple_json_schema_dict():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_json_schema_dict(model):
     schema = {
         "properties": {"bar": {"title": "Bar", "type": "integer"}},
         "required": ["bar"],
@@ -128,9 +120,7 @@ def test_gemini_simple_json_schema_dict():
     reason="The Gemini SDK's serialization method does not support Json Schema strings."
 )
 @pytest.mark.api_call
-def test_gemini_simple_json_schema_string():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_json_schema_string(model):
     schema = "{'properties': {'bar': {'title': 'Bar', 'type': 'integer'}}, 'required': ['bar'], 'title': 'Foo', 'type': 'object'}"
     result = model.generate("foo?", JsonType(schema))
     assert isinstance(result, str)
@@ -138,9 +128,7 @@ def test_gemini_simple_json_schema_string():
 
 
 @pytest.mark.api_call
-def test_gemini_simple_typed_dict():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_typed_dict(model):
     class Foo(TypedDict):
         bar: int
 
@@ -150,9 +138,7 @@ class Foo(TypedDict):
 
 
 @pytest.mark.api_call
-def test_gemini_simple_choice_enum():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_choice_enum(model):
     class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
@@ -163,9 +149,7 @@ class Foo(Enum):
 
 
 @pytest.mark.api_call
-def test_gemini_simple_choice_list():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_choice_list(model):
     choices = ["Foo", "Bar"]
     result = model.generate("foo?", Choice(choices))
     assert isinstance(result, str)
@@ -173,9 +157,7 @@ def test_gemini_simple_choice_list():
 
 
 @pytest.mark.api_call
-def test_gemini_simple_list_pydantic():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_list_pydantic(model):
     class Foo(BaseModel):
         bar: int
 
@@ -183,3 +165,10 @@ class Foo(BaseModel):
     assert isinstance(json.loads(result), list)
     assert isinstance(json.loads(result)[0], dict)
     assert "bar" in json.loads(result)[0]
+
+
+@pytest.mark.api_call
+def test_gemini_streaming(model):
+    result = model.stream("Respond with one word. Not more.")
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 225ec558d..71cd3053b 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -91,9 +91,7 @@ def test_llamacpp_stream_simple(model):
 
 
 def test_llamacpp_stream_regex(model):
-    regex_str = Regex(r"[0-9]").pattern
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
-    generator = model.stream("Respond with one word. Not more.", logits_processor)
+    generator = model.stream("Respond with one word. Not more.", Regex(r"[0-9]"))
 
     x = next(generator)
     assert isinstance(x, str)
@@ -103,9 +101,7 @@ def test_llamacpp_stream_json(model):
     class Foo(BaseModel):
         bar: int
 
-    regex_str = JsonType(Foo).to_regex()
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
-    generator = model.stream("foo?", logits_processor)
+    generator = model.stream("foo?", JsonType(Foo))
 
     x = next(generator)
     assert x == "{"
@@ -116,9 +112,7 @@ class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
 
-    regex_str = Choice(Foo).to_regex()
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
-    generator = model.stream("foo?", logits_processor)
+    generator = model.stream("foo?", Choice(Foo))
 
     x = next(generator)
     assert isinstance(x, str)
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index 3a4d64681..a6b49bf92 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -1,6 +1,7 @@
 import pytest
 import re
 from enum import Enum
+from typing import Generator
 
 import outlines
 from outlines.types import Choice, JsonType, Regex
@@ -100,3 +101,12 @@ class Foo(Enum):
 
     result = model("Cat or dog?", Choice(Foo))
     assert result in ["cat", "dog"]
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_stream_text_stop(model):
+    generator = model.stream(
+        "Respond with one word. Not more.", None, max_tokens=100
+    )
+    assert isinstance(generator, Generator)
+    assert isinstance(next(generator), str)
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 57ed323de..92fbccddb 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,7 +1,7 @@
 import io
 import json
 import os
-
+from typing import Generator
 import PIL
 import pytest
 import requests
@@ -133,3 +133,12 @@ class Foo(BaseModel):
     result = model.generate("foo?", JsonType(schema))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_openai_streaming():
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
+
+    result = model.stream("Respond with one word. Not more.")
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index f2fb90424..ea3a601e9 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -146,3 +146,8 @@ class Foo(Enum):
         assert len(item) == 2
         assert item[0] in ["cat", "dog"]
         assert item[1] in ["cat", "dog"]
+
+
+def test_transformers_streaming(model):
+    with pytest.raises(NotImplementedError, match="Streaming is not implemented"):
+        model.stream("Respond with one word. Not more.")
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index 5b5aa96ec..2a94ee4c9 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -133,3 +133,8 @@ class Foo(Enum):
         assert len(item) == 2
         assert item[0] in ["cat", "dog"]
         assert item[1] in ["cat", "dog"]
+
+
+def test_vllm_streaming(model):
+    with pytest.raises(NotImplementedError, match="Streaming is not implemented"):
+        model.stream("Respond with one word. Not more.")

From 93f70f5e17b962f9063cf7973b0990c9a2227302 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 11 Mar 2025 10:21:46 +0100
Subject: [PATCH 255/505] Remove the function python_types_to_regex

---
 outlines/fsm/types.py            | 76 --------------------------------
 tests/fsm/test_types.py          | 22 ---------
 tests/types/test_custom_types.py |  1 -
 3 files changed, 99 deletions(-)
 delete mode 100644 outlines/fsm/types.py
 delete mode 100644 tests/fsm/test_types.py

diff --git a/outlines/fsm/types.py b/outlines/fsm/types.py
deleted file mode 100644
index f6409aa66..000000000
--- a/outlines/fsm/types.py
+++ /dev/null
@@ -1,76 +0,0 @@
-import datetime
-from enum import EnumMeta
-from typing import Any, Protocol, Tuple, Type
-
-from outlines.types import Regex, boolean as boolean_regex, date as date_regex
-from outlines.types import datetime as datetime_regex
-from outlines.types import (
-    integer as integer_regex,
-    number as number_regex,
-    time as time_regex,
-)
-
-
-class FormatFunction(Protocol):
-    def __call__(self, sequence: str) -> Any: ...
-
-
-def python_types_to_regex(python_type: Type) -> Tuple[Regex, FormatFunction]:
-    # If it is a custom type
-    if isinstance(python_type, Regex):
-        custom_regex_str = python_type.pattern
-
-        def custom_format_fn(sequence: str) -> str:
-            return str(sequence)
-
-        return Regex(custom_regex_str), custom_format_fn
-
-    if isinstance(python_type, EnumMeta):
-        values = python_type.__members__.keys()
-        enum_regex_str: str = "(" + "|".join(values) + ")"
-
-        def enum_format_fn(sequence: str) -> str:
-            return str(sequence)
-
-        return Regex(enum_regex_str), enum_format_fn
-
-    if python_type is float:
-
-        def float_format_fn(sequence: str) -> float:
-            return float(sequence)
-
-        return number_regex, float_format_fn
-    elif python_type is int:
-
-        def int_format_fn(sequence: str) -> int:
-            return int(sequence)
-
-        return integer_regex, int_format_fn
-    elif python_type is bool:
-
-        def bool_format_fn(sequence: str) -> bool:
-            return bool(sequence)
-
-        return boolean_regex, bool_format_fn
-    elif python_type == datetime.date:
-
-        def date_format_fn(sequence: str) -> datetime.date:
-            return datetime.datetime.strptime(sequence, "%Y-%m-%d").date()
-
-        return date_regex, date_format_fn
-    elif python_type == datetime.time:
-
-        def time_format_fn(sequence: str) -> datetime.time:
-            return datetime.datetime.strptime(sequence, "%H:%M:%S").time()
-
-        return time_regex, time_format_fn
-    elif python_type == datetime.datetime:
-
-        def datetime_format_fn(sequence: str) -> datetime.datetime:
-            return datetime.datetime.strptime(sequence, "%Y-%m-%d %H:%M:%S")
-
-        return datetime_regex, datetime_format_fn
-    else:
-        raise NotImplementedError(
-            f"The Python type {python_type} is not supported. Please open an issue."
-        )
diff --git a/tests/fsm/test_types.py b/tests/fsm/test_types.py
deleted file mode 100644
index 4017a817c..000000000
--- a/tests/fsm/test_types.py
+++ /dev/null
@@ -1,22 +0,0 @@
-import datetime as pydatetime
-
-import pytest
-
-from outlines.fsm.types import python_types_to_regex
-from outlines import types
-
-
-@pytest.mark.parametrize(
-    "python_type,custom_type",
-    [
-        (int, types.integer),
-        (float, types.number),
-        (bool, types.boolean),
-        (pydatetime.date, types.date),
-        (pydatetime.time, types.time),
-        (pydatetime.datetime, types.datetime),
-    ],
-)
-def test_python_types(python_type, custom_type):
-    test_regex, _ = python_types_to_regex(python_type)
-    assert custom_type.pattern == test_regex.pattern
diff --git a/tests/types/test_custom_types.py b/tests/types/test_custom_types.py
index 46df9ba4a..691b681b8 100644
--- a/tests/types/test_custom_types.py
+++ b/tests/types/test_custom_types.py
@@ -4,7 +4,6 @@
 from pydantic import BaseModel
 
 from outlines import types
-from outlines.fsm.types import python_types_to_regex
 from outlines.types.dsl import to_regex
 
 

From e58a078b68641d5e4256bd3ff24ee36e646e76a0 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 11 Mar 2025 12:31:29 +0100
Subject: [PATCH 256/505] Move the guides to the processors directory

---
 benchmarks/bench_cfg_guide.py               | 2 +-
 benchmarks/bench_json_schema.py             | 2 +-
 benchmarks/bench_regex_guide.py             | 2 +-
 docs/api/guide.md                           | 2 +-
 outlines/generator.py                       | 2 +-
 outlines/{fsm => processors}/guide.py       | 0
 outlines/processors/structured.py           | 6 +++---
 tests/generate/conftest.py                  | 4 ++--
 tests/{fsm => processors}/test_cfg_guide.py | 6 ++----
 tests/{fsm => processors}/test_guide.py     | 2 +-
 tests/test_grammars.py                      | 2 +-
 11 files changed, 14 insertions(+), 16 deletions(-)
 rename outlines/{fsm => processors}/guide.py (100%)
 rename tests/{fsm => processors}/test_cfg_guide.py (99%)
 rename tests/{fsm => processors}/test_guide.py (99%)

diff --git a/benchmarks/bench_cfg_guide.py b/benchmarks/bench_cfg_guide.py
index 477104435..a4d288700 100644
--- a/benchmarks/bench_cfg_guide.py
+++ b/benchmarks/bench_cfg_guide.py
@@ -4,7 +4,7 @@
 
 import outlines.grammars
 from outlines.caching import cache_disabled
-from outlines.fsm.guide import CFGGuide
+from outlines.processors.guide import CFGGuide
 from outlines.models.transformers import TransformerTokenizer
 
 random.seed(42)
diff --git a/benchmarks/bench_json_schema.py b/benchmarks/bench_json_schema.py
index 3a1f72cb6..6ed223ef4 100644
--- a/benchmarks/bench_json_schema.py
+++ b/benchmarks/bench_json_schema.py
@@ -1,7 +1,7 @@
 from outlines_core.fsm.json_schema import build_regex_from_schema
 
 from outlines.caching import cache_disabled
-from outlines.fsm.guide import RegexGuide
+from outlines.processors.guide import RegexGuide
 
 from .common import setup_tokenizer  # noqa: E402
 
diff --git a/benchmarks/bench_regex_guide.py b/benchmarks/bench_regex_guide.py
index fa23a724f..14b15da89 100644
--- a/benchmarks/bench_regex_guide.py
+++ b/benchmarks/bench_regex_guide.py
@@ -1,5 +1,5 @@
 from outlines.caching import cache_disabled
-from outlines.fsm.guide import RegexGuide
+from outlines.processors.guide import RegexGuide
 
 from .common import setup_tokenizer
 
diff --git a/docs/api/guide.md b/docs/api/guide.md
index 1c3160c8c..08d2f7f0f 100644
--- a/docs/api/guide.md
+++ b/docs/api/guide.md
@@ -1 +1 @@
-::: outlines.fsm.guide
+::: outlines.processors.guide
diff --git a/outlines/generator.py b/outlines/generator.py
index 0238fa598..462066531 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -3,7 +3,7 @@
 
 import interegular
 
-from outlines.fsm.guide import RegexGuide
+from outlines.processors.guide import RegexGuide
 from outlines.models import BlackBoxModel, SteerableModel
 from outlines.processors import CFGLogitsProcessor, GuideLogitsProcessor, RegexLogitsProcessor
 from outlines.types import CFG, Choice, JsonType, List, Regex
diff --git a/outlines/fsm/guide.py b/outlines/processors/guide.py
similarity index 100%
rename from outlines/fsm/guide.py
rename to outlines/processors/guide.py
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 8ce69c32a..225d39c3e 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -31,7 +31,7 @@
 from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel
 
-from outlines.fsm.guide import CFGGuide, Guide, RegexGuide
+from outlines.processors.guide import CFGGuide, Guide, RegexGuide
 from outlines.fsm.json_schema import convert_json_schema_to_str
 
 from .base_logits_processor import OutlinesLogitsProcessor
@@ -48,7 +48,7 @@ class GuideLogitsProcessor(OutlinesLogitsProcessor):
     tokenizer
         The tokenizer used to convert tokens to ids.
     guide
-        The `outlines.fsm.Guide` which is used to bias the logits.
+        The `outlines.processors.guide` which is used to bias the logits.
     """
 
     tokenizer: "Tokenizer"
@@ -64,7 +64,7 @@ def __init__(self, tokenizer: "Tokenizer", guide: Guide):
         tokenizer
             The tokenizer used to convert tokens to ids.
         guide
-            The `outlines.fsm.Guide. which is used to bias the logits.
+            The `outlines.processors.guide. which is used to bias the logits.
         """
         self.tokenizer = tokenizer
         self.guide = guide
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
index ef7e40eed..b067aa46a 100644
--- a/tests/generate/conftest.py
+++ b/tests/generate/conftest.py
@@ -9,13 +9,13 @@ def temp_cache_dir():
     import tempfile
 
     import outlines.caching
-    import outlines.fsm.guide
+    import outlines.processors.guide
 
     with tempfile.TemporaryDirectory() as tempdir:
         os.environ["OUTLINES_CACHE_DIR"] = tempdir
         outlines.caching.get_cache.cache_clear()
         reload(outlines)
-        reload(outlines.fsm.guide)
+        reload(outlines.processors.guide)
         cache_status = outlines.caching._caching_enabled
         try:
             outlines.caching._caching_enabled = True
diff --git a/tests/fsm/test_cfg_guide.py b/tests/processors/test_cfg_guide.py
similarity index 99%
rename from tests/fsm/test_cfg_guide.py
rename to tests/processors/test_cfg_guide.py
index b925fefea..6091fd4c2 100644
--- a/tests/fsm/test_cfg_guide.py
+++ b/tests/processors/test_cfg_guide.py
@@ -6,7 +6,7 @@
 
 from outlines import grammars
 from outlines.models.transformers import TransformerTokenizer
-from outlines.fsm.guide import CFGGuide
+from outlines.processors.guide import CFGGuide
 
 
 @pytest.fixture
@@ -363,9 +363,7 @@ def tokenizer_tiktoken_llama3():
 
 @pytest.fixture(scope="session")
 def tokenizer_character_level_byt5():
-    return TransformerTokenizer(
-        AutoTokenizer.from_pretrained("google/byt5-small")
-    )
+    return TransformerTokenizer(AutoTokenizer.from_pretrained("google/byt5-small"))
 
 
 # Collects all samples within cfg_samples/ and makes adding
diff --git a/tests/fsm/test_guide.py b/tests/processors/test_guide.py
similarity index 99%
rename from tests/fsm/test_guide.py
rename to tests/processors/test_guide.py
index bf25c43c4..0f8cc9974 100644
--- a/tests/fsm/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -1,6 +1,6 @@
 import pytest
 
-from outlines.fsm.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
+from outlines.processors.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
 
 
 def assert_expected_tensor_ids(tensor, ids):
diff --git a/tests/test_grammars.py b/tests/test_grammars.py
index 3d704f094..69486684b 100644
--- a/tests/test_grammars.py
+++ b/tests/test_grammars.py
@@ -1,7 +1,7 @@
 import pytest
 
 import outlines.grammars as grammars
-from outlines.fsm.guide import CFGGuide
+from outlines.processors.guide import CFGGuide
 
 
 @pytest.mark.parametrize("grammar", [grammars.json, grammars.arithmetic])

From c49bcd9b741a4104f38ab494da9d54b236e26e79 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 11 Mar 2025 15:45:22 +0100
Subject: [PATCH 257/505] Modify TransformersMultiModalTypeAdapter to make it
 more flexible

---
 .../models/transformers_multimodal.md         | 19 +++++++++-----
 outlines/models/transformers.py               | 25 ++++++++-----------
 2 files changed, 23 insertions(+), 21 deletions(-)

diff --git a/docs/reference/models/transformers_multimodal.md b/docs/reference/models/transformers_multimodal.md
index cea644204..199d38d09 100644
--- a/docs/reference/models/transformers_multimodal.md
+++ b/docs/reference/models/transformers_multimodal.md
@@ -21,21 +21,28 @@ model = models.from_transformers(
 )
 ```
 
-You must provider a processor adapted to your model. We currently only support vision and audio models.
+You must provider a processor adapted to your model. We currently offer official support for vision and audio models only, but you can use this class for any type of multimodal model (for instance video) as long as you provide the appropriate processor for your transformers model.
 
-## Use the model to generate text from prompts and images/audios
+## Use the model to generate text from prompts and assets
 
 When calling the model, the prompt argument you provide must be a dictionary using the following format:
 ```
 {
     "text": Union[str, List[str]],
-    "images": Optional[Union[Any, List[Any]]],
-    "audios": Optional[Union[Any, List[Any]]],
+    "...": Any
 }
 ```
 
-The `text` key is required. Either the `images` or `audios` key must be provided.
-The format and values of those keys must correspond to what you would provide to the `__call__` method of the processor if you were to call it directly. Thus, the text prompt must include the appropriate tags to indicate where the images or audios should be inserted (e.g. `<image>` or `<|AUDIO|>`).
+The `text` key is mandatory. Other keys are optional and depend on the processor you are using.
+The format and values of those keys must correspond to what you would provide to the `__call__` method of the processor if you were to call it directly. Thus, the text prompt must include the appropriate tags to indicate where the assets should be inserted (e.g. `<image>` or `<|AUDIO|>`).
+
+Example of a correctly formatted prompt for a vision model:
+```
+{
+    "text": "<image>Describe this image in one sentence:",
+    "images": PIL.Image.Image
+}
+```
 
 ### Vision models
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 61d646364..69bbd85bc 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -271,26 +271,21 @@ def format_input(self, model_input):
         """
         raise NotImplementedError(
             f"The input type {input} is not available. Please provide a "
-            "dictionary with the following format: "
-            "{'text': Union[str, List[str]], 'images': Optional[Union[Any, "
-            "List[Any]]], 'audios': Optional[Union[Any, List[Any]]]}"
-            "Either 'images' or 'audios' must be provided."
-            "Make sure the text is formatted correctly for the model "
-            "(e.g. include <image> or <|AUDIO|> tags)."
+            + "dictionary containing at least the 'text' key with a value "
+            + "of type Union[str, List[str]]. You should also include the "
+            + "other keys required by your processor (for instance, 'images' "
+            + "or 'audios')."
+            + "Make sure that the text is correctly formatted for the model "
+            + "(e.g. include <image> or <|AUDIO|> tags) and that the number "
+            + "of text tags match the number of additional assets provided."
         )
 
     @format_input.register(dict)
     def format_list_input(self, model_input):
-        if (
-            "text" not in model_input
-            or (
-                "images" not in model_input
-                and "audios" not in model_input
-            )
-        ):
+        if "text" not in model_input:
             raise ValueError(
-                "The input must contain the 'text' key along with either "
-                "'images' or 'audios'."
+                "The input must contain the 'text' key along with the other "
+                + "keys required by your processor."
             )
         return model_input
 

From d0b9be1356fc6d936190b0529b248f061997526e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sun, 23 Feb 2025 23:01:12 +0100
Subject: [PATCH 258/505] Update the README

---
 README.md | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/README.md b/README.md
index 4e9448249..611224dee 100644
--- a/README.md
+++ b/README.md
@@ -163,6 +163,7 @@ hood:
 ``` python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import outlines
+from outlines import regex
 
 model_name = "microsoft/Phi-3-mini-4k-instruct"
 model = outlines.from_transformers(
@@ -243,12 +244,11 @@ model = outlines.from_transformers(
 )
 
 # Construct structured sequence generator
+generator = outlines.Generator(model, Character)
 
 # Draw a sample
 seed = 789001
-
-prompt = "Give me a character description"
-character = model(prompt, outlines.JsonType(Character), seed=seed)
+character = generator("Give me a character description", seed=seed)
 
 print(repr(character))
 # Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
@@ -369,16 +369,17 @@ import outlines
 def add(a: int, b: int):
     return a + b
 
-
 model_name = "WizardLM/WizardMath-7B-V1.1"
 model = outlines.from_transformers(
     AutoModelForCausalLM.from_pretrained(model_name),
     AutoTokenizer.from_pretrained(model_name)
 )
-generator = outlines.generate.json(model, outlines.types.JsonType(add))
-result = generator("Return json with two integers named a and b respectively. a is odd and b even.")
+result = model(
+    "Return json with two integers named a and b respectively. a is odd and b even.",
+    add
+)
 
-# print(add(**result))
+print(add(**result))
 # 3
 ```
 
@@ -410,9 +411,8 @@ model = outlines.from_transformers(
     AutoModelForCausalLM.from_pretrained(model_name),
     AutoTokenizer.from_pretrained(model_name)
 )
-generator = outlines.generate.json(model, outlines.types.JsonType(Operation))
+generator = outlines.Generator(model, outlines.types.JsonType(Operation))
 result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
-
 #print(result)
 # {'c': -3.14, 'd': 1.5}
 ```

From 48a33110ec269439b66da07c1c3e3a498a2f04a6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 24 Feb 2025 00:52:15 +0100
Subject: [PATCH 259/505] Misc type fixes

---
 outlines/models/openai.py         | 1 +
 tests/models/test_transformers.py | 8 ++++++++
 2 files changed, 9 insertions(+)

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 80774a8d3..a3f738221 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,4 +1,5 @@
 """Integration with OpenAI's API."""
+
 from functools import singledispatchmethod
 from types import NoneType
 from typing import Optional, Union, TYPE_CHECKING
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index ea3a601e9..1c7602b78 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -38,6 +38,14 @@ def test_transformers_instantiate_mamba():
     assert isinstance(model, Transformers)
 
 
+def test_transformers_instantiate_tokenizer_kwargs():
+    model = Transformers(
+        TEST_MODEL, tokenizer_kwargs={"additional_special_tokens": ["<t1>", "<t2>"]}
+    )
+    assert "<t1>" in model.tokenizer.special_tokens
+    assert "<t2>" in model.tokenizer.special_tokens
+
+
 @pytest.fixture
 def model():
     model = outlines.from_transformers(

From d0f78d78dbc2958da55ddcdf57ef2f51e2002e68 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 24 Feb 2025 23:32:24 +0100
Subject: [PATCH 260/505] Use fixture to instantiate `Gemini` model

---
 tests/models/test_gemini.py | 36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 3560b310a..36addf14d 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -2,12 +2,13 @@
 import json
 from enum import Enum
 from typing import Generator
+
 import PIL
+import google.generativeai as genai
 import pytest
 import requests
 from pydantic import BaseModel
 from typing_extensions import TypedDict
-import google.generativeai as genai
 
 import outlines
 from outlines.models.gemini import Gemini
@@ -62,7 +63,7 @@ def test_gemini_simple_pydantic(model):
     class Foo(BaseModel):
         bar: int
 
-    result = model.generate("foo?", JsonType(Foo))
+    result = model.generate("foo?", Foo)
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -93,7 +94,7 @@ class Foo(BaseModel):
         sna: int
         bar: Bar
 
-    result = model.generate("foo?", JsonType(Foo))
+    result = model.generate("foo?", Foo)
     assert isinstance(result, str)
     assert "sna" in json.loads(result)
     assert "bar" in json.loads(result)
@@ -111,7 +112,7 @@ def test_gemini_simple_json_schema_dict(model):
         "title": "Foo",
         "type": "object",
     }
-    result = model.generate("foo?", JsonType(schema))
+    result = model.generate("foo?", schema)
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -122,7 +123,7 @@ def test_gemini_simple_json_schema_dict(model):
 @pytest.mark.api_call
 def test_gemini_simple_json_schema_string(model):
     schema = "{'properties': {'bar': {'title': 'Bar', 'type': 'integer'}}, 'required': ['bar'], 'title': 'Foo', 'type': 'object'}"
-    result = model.generate("foo?", JsonType(schema))
+    result = model.generate("foo?", schema)
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -137,8 +138,22 @@ class Foo(TypedDict):
     assert "bar" in json.loads(result)
 
 
+@pytest.mark.api_call
+def test_gemini_simple_dataclass():
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
+
+    @dataclass
+    class Foo:
+        bar: int
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
 @pytest.mark.api_call
 def test_gemini_simple_choice_enum(model):
+
     class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
@@ -148,6 +163,17 @@ class Foo(Enum):
     assert result == "Foo" or result == "Bar"
 
 
+@pytest.mark.api_call
+def test_gemini_sample_choice_literal():
+    model = Gemini(genai.GenerativeModel(MODEL_NAME))
+    result = model.generate("foo?", Literal["Foo", "Bar"])
+    assert isinstance(result, str)
+    assert result == "Foo" or result == "Bar"
+
+
+@pytest.mark.xfail(
+    reason="Gemini supports lists for choices but we do not as it is semantically incorrect."
+)
 @pytest.mark.api_call
 def test_gemini_simple_choice_list(model):
     choices = ["Foo", "Bar"]

From 820660bbacd976c466e0fec9d7b91c6c382ff5bd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 24 Feb 2025 00:52:32 +0100
Subject: [PATCH 261/505] Handle `Gemini` output types solely in
 `GeminiTypeAdapter`

We are currently using intermediate `Json`, `Choice` and `List` classes
which, in addition to being cumbersome, are not useful abstractions.
This commit bypasses these abstractions to handle the output types
directly at the level of `GeminiTypeAdapter`.
---
 outlines/models/gemini.py                | 145 ++++++++++-------
 outlines/models/openai.py                |   1 -
 tests/models/test_dottxt.py              |   2 +
 tests/models/test_gemini.py              |  50 +++---
 tests/models/test_gemini_type_adapter.py | 192 +++++++++++++++++++++++
 5 files changed, 307 insertions(+), 83 deletions(-)
 create mode 100644 tests/models/test_gemini_type_adapter.py

diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 1f6a9e37d..c1ac1b075 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -1,16 +1,22 @@
 """Integration with Gemini's API."""
 
-from enum import EnumMeta
-from functools import singledispatchmethod
+from dataclasses import is_dataclass
+from enum import EnumMeta, Enum
 from types import NoneType
-from typing import Optional, Union, TYPE_CHECKING
-
-from pydantic import BaseModel
-from typing_extensions import _TypedDictMeta  # type: ignore
-
+from typing import (
+    get_args,
+    get_origin,
+    Optional,
+    Union,
+    Any,
+    TYPE_CHECKING,
+    Literal,
+)
+from typing_extensions import is_typeddict
+from pydantic import BaseModel, TypeAdapter
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
-from outlines.types import JsonType, Choice, List
+from outlines.types import Regex, CFG, JsonSchema
 
 
 if TYPE_CHECKING:
@@ -28,7 +34,6 @@ class GeminiTypeAdapter(ModelTypeAdapter):
 
     """
 
-    @singledispatchmethod
     def format_input(self, model_input):
         """Generate the `messages` argument to pass to the client.
 
@@ -38,64 +43,88 @@ def format_input(self, model_input):
             The input passed by the user.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available with Gemini. The only available types are `str` and `Vision`."
-        )
-
-    @format_input.register(str)
-    def format_str_input(self, model_input: str):
-        """Generate the `messages` argument to pass to the client when the user
-        only passes a prompt.
-
-        """
-        return {"contents": [model_input]}
-
-    @format_input.register(Vision)
-    def format_vision_input(self, model_input: Vision):
-        """Generate the `messages` argument to pass to the client when the user
-        passes a prompt and an image.
-
-        """
-        return {"contents": [model_input.prompt, model_input.image]}
+        if isinstance(model_input, str):
+            return {"contents": [model_input]}
+        elif isinstance(model_input, Vision):
+            return {"contents": [model_input.prompt, model_input.image]}
+        else:
+            raise TypeError(
+                f"The input type {input} is not available with Gemini. The only available types are `str` and `Vision`."
+            )
 
-    @singledispatchmethod
     def format_output_type(self, output_type):
-        raise NotImplementedError
+        # TODO: `int`, `float` and other Python types could be supported via JSON Schema.
+
+        # Unsupported languages
+        if isinstance(output_type, Regex):
+            raise TypeError(
+                "Neither regex-based structured outputs nor the `pattern` keyword in Json Schema are available with Gemini. Use an open source model or dottxt instead."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs are not available with Gemini. Use an open source model or dottxt instead."
+            )
+        elif isinstance(output_type, JsonSchema):
+            raise TypeError(
+                "The Gemini SDK does not accept Json Schemas as an input. Pass a Pydantic model, typed dict or dataclass instead."
+            )
+
+        if output_type is None:
+            return {}
+        elif is_dataclass(output_type):  # structured types
+            return self.format_json_output_type(output_type)
+        elif is_typeddict(output_type):
+            return self.format_json_output_type(output_type)
+        elif isinstance(output_type, type(BaseModel)):
+            return self.format_json_output_type(output_type)
+        if isinstance(output_type, EnumMeta):  # Enum types
+            return self.format_enum_output_type(output_type)
+        elif get_origin(output_type) is Literal:
+            out = Enum("EnumFromLiteral", [(arg, arg) for arg in get_args(output_type)])
+            return self.format_enum_output_type(out)
+        elif get_origin(output_type) is list:  # List of objects
+            return self.format_list_output_type(output_type)
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not supported by Gemini. "
+                "Consider using a local model or dottxt instead."
+            )
 
-    @format_output_type.register(List)
-    def format_list_output_type(self, output_type):
+    def format_enum_output_type(self, output_type):
         return {
-            "response_mime_type": "application/json",
-            "response_schema": list[output_type.definition.definition],
+            "response_mime_type": "text/x.enum",
+            "response_schema": output_type,
         }
 
-    @format_output_type.register(NoneType)
-    def format_none_output_type(self, output_type):
-        return {}
-
-    @format_output_type.register(JsonType)
     def format_json_output_type(self, output_type):
-        """Gemini only accepts Pydantic models and TypeDicts to define the JSON structure."""
-        if issubclass(output_type.definition, BaseModel):
-            return {
-                "response_mime_type": "application/json",
-                "response_schema": output_type.definition,
-            }
-        elif isinstance(output_type.definition, _TypedDictMeta):
-            return {
-                "response_mime_type": "application/json",
-                "response_schema": output_type.definition,
-            }
-        else:
-            raise NotImplementedError
-
-    @format_output_type.register(Choice)
-    def format_enum_output_type(self, output_type):
         return {
-            "response_mime_type": "text/x.enum",
-            "response_schema": output_type.definition,
+            "response_mime_type": "application/json",
+            "response_schema": output_type,
         }
 
+    def format_list_output_type(self, output_type):
+        args = get_args(output_type)
+
+        if len(args) == 1:
+            item_type = args[0]
+
+            # Check if list item type is supported
+            if (
+                isinstance(item_type, type(BaseModel))
+                or is_typeddict(item_type)
+                or is_dataclass(item_type)
+            ):
+                return {
+                    "response_mime_type": "application/json",
+                    "response_schema": output_type,
+                }
+
+        raise TypeError(
+            f"Gemini only supports homogenous lists: list[BaseModel], list[TypedDict] or list[dataclass]. "
+            f"Got {output_type} instead."
+        )
+
 
 class Gemini(Model):
     def __init__(self, client: "GeminiClient"):
@@ -105,7 +134,7 @@ def __init__(self, client: "GeminiClient"):
     def generate(
         self,
         model_input: Union[str, Vision],
-        output_type: Optional[Union[JsonType, EnumMeta]] = None,
+        output_type: Optional[Any] = None,
         **inference_kwargs,
     ):
         import google.generativeai as genai
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index a3f738221..887744741 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -5,7 +5,6 @@
 from typing import Optional, Union, TYPE_CHECKING
 
 
-
 from pydantic import BaseModel
 
 from outlines.models.base import Model, ModelTypeAdapter
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index 8d649dfd0..2d995829e 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -44,12 +44,14 @@ def test_dottxt_wrong_output_type(api_key):
         model = Dottxt(client)
         model("prompt")
 
+
 def test_dottxt_init_from_client(api_key):
     client = DottxtClient(api_key=api_key)
     model = outlines.from_dottxt(client)
     assert isinstance(model, Dottxt)
     assert model.client == client
 
+
 @pytest.mark.api_call
 def test_dottxt_wrong_input_type(api_key):
     with pytest.raises(NotImplementedError, match="is not available"):
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 36addf14d..6955c9db2 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -1,19 +1,20 @@
 import io
 import json
+from dataclasses import dataclass
 from enum import Enum
-from typing import Generator
+from typing import Generator, Literal
 
-import PIL
 import google.generativeai as genai
 import pytest
 import requests
+from PIL import Image
 from pydantic import BaseModel
 from typing_extensions import TypedDict
 
 import outlines
 from outlines.models.gemini import Gemini
 from outlines.templates import Vision
-from outlines.types import Choice, JsonType, List
+
 
 MODEL_NAME = "gemini-1.5-flash-latest"
 
@@ -23,6 +24,21 @@ def model():
     return Gemini(genai.GenerativeModel(MODEL_NAME))
 
 
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
 def test_gemini_wrong_inference_parameters(model):
     with pytest.raises(TypeError, match="got an unexpected"):
         model.generate("prompt", foo=10)
@@ -48,12 +64,7 @@ def test_gemini_direct_call(model):
 
 
 @pytest.mark.api_call
-def test_gemini_simple_vision(model):
-    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
-    r = requests.get(url, stream=True)
-    if r.status_code == 200:
-        image = PIL.Image.open(io.BytesIO(r.content))
-
+def test_gemini_simple_vision(model, image):
     result = model.generate(Vision("What does this logo represent?", image))
     assert isinstance(result, str)
 
@@ -71,11 +82,6 @@ class Foo(BaseModel):
 @pytest.mark.xfail(reason="Vision models do not work with structured outputs.")
 @pytest.mark.api_call
 def test_gemini_simple_vision_pydantic(model):
-    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
-    r = requests.get(url, stream=True)
-    if r.status_code == 200:
-        image = PIL.Image.open(io.BytesIO(r.content))
-
     class Logo(BaseModel):
         name: int
 
@@ -133,15 +139,13 @@ def test_gemini_simple_typed_dict(model):
     class Foo(TypedDict):
         bar: int
 
-    result = model.generate("foo?", JsonType(Foo))
+    result = model.generate("foo?", Foo)
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
 
 @pytest.mark.api_call
-def test_gemini_simple_dataclass():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
-
+def test_gemini_simple_dataclass(model):
     @dataclass
     class Foo:
         bar: int
@@ -153,19 +157,17 @@ class Foo:
 
 @pytest.mark.api_call
 def test_gemini_simple_choice_enum(model):
-
     class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
 
-    result = model.generate("foo?", Choice(Foo))
+    result = model.generate("foo?", Foo)
     assert isinstance(result, str)
     assert result == "Foo" or result == "Bar"
 
 
 @pytest.mark.api_call
-def test_gemini_sample_choice_literal():
-    model = Gemini(genai.GenerativeModel(MODEL_NAME))
+def test_gemini_sample_choice_literal(model):
     result = model.generate("foo?", Literal["Foo", "Bar"])
     assert isinstance(result, str)
     assert result == "Foo" or result == "Bar"
@@ -177,7 +179,7 @@ def test_gemini_sample_choice_literal():
 @pytest.mark.api_call
 def test_gemini_simple_choice_list(model):
     choices = ["Foo", "Bar"]
-    result = model.generate("foo?", Choice(choices))
+    result = model.generate("foo?", choices)
     assert isinstance(result, str)
     assert result == "Foo" or result == "Bar"
 
@@ -187,7 +189,7 @@ def test_gemini_simple_list_pydantic(model):
     class Foo(BaseModel):
         bar: int
 
-    result = model.generate("foo?", List(JsonType(Foo)))
+    result = model.generate("foo?", list[Foo])
     assert isinstance(json.loads(result), list)
     assert isinstance(json.loads(result)[0], dict)
     assert "bar" in json.loads(result)[0]
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
new file mode 100644
index 000000000..6b3925f08
--- /dev/null
+++ b/tests/models/test_gemini_type_adapter.py
@@ -0,0 +1,192 @@
+from dataclasses import dataclass
+from enum import EnumMeta
+import json
+from typing import Literal, get_args
+from typing_extensions import TypedDict, is_typeddict
+from enum import Enum
+import io
+
+from PIL import Image
+from pydantic import BaseModel
+import pytest
+
+from outlines import regex, cfg, json_schema
+from outlines.models.gemini import GeminiTypeAdapter
+from outlines.templates import Vision
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return GeminiTypeAdapter()
+
+
+def test_gemini_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert result == {"contents": [message]}
+
+
+def test_gemini_type_adapter_input_vision(adapter, image):
+    input_message = Vision("hello", image)
+    result = adapter.format_input(input_message)
+    assert result == {"contents": [input_message.prompt, input_message.image]}
+
+
+def test_dottxt_type_adapter_input_invalid(adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_gemini_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="The type `str` is not supported"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not supported"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="Neither regex-based"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs"):
+        adapter.format_output_type(cfg(""))
+
+    with pytest.raises(TypeError, match="The Gemini SDK does not"):
+        adapter.format_output_type(json_schema(""))
+
+
+def test_gemini_type_adapter_output_none(adapter):
+    result = adapter.format_output_type(None)
+    assert result == {}
+
+
+def test_gemini_type_adapter_output_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": User,
+    }
+
+
+def test_gemini_type_adapter_output_list_dataclass(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(list[User])
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": list[User],
+    }
+
+
+def test_gemini_type_adapter_output_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": User,
+    }
+
+
+def test_gemini_type_adapter_output_list_typed_dict(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(list[User])
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": list[User],
+    }
+
+
+def test_gemini_type_adapter_output_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": User,
+    }
+
+
+def test_gemini_type_adapter_output_list_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(list[User])
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": list[User],
+    }
+
+
+def test_gemini_type_adapter_output_enum(adapter):
+    class Foo(Enum):
+        Bar = "bar"
+        Fuzz = "fuzz"
+
+    result = adapter.format_output_type(Foo)
+    assert result == {
+        "response_mime_type": "text/x.enum",
+        "response_schema": Foo,
+    }
+
+
+def test_gemini_type_adapter_output_literal(adapter):
+    Foo = Literal["bar", "fuzz"]
+    result = adapter.format_output_type(Foo)
+
+    assert isinstance(result, dict)
+    assert len(result) == 2
+    assert result["response_mime_type"] == "text/x.enum"
+    assert isinstance(result["response_schema"], EnumMeta)
+    assert len(result["response_schema"].__members__) == 2
+    assert result["response_schema"].bar.value == "bar"
+    assert result["response_schema"].fuzz.value == "fuzz"

From 7f155d6464ae39296811abd57ba141ac8f6ffc98 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Tue, 25 Feb 2025 09:16:15 +0100
Subject: [PATCH 262/505] Handle `Dottxt` output types in `DottxtTypeAdapter`

---
 outlines/models/anthropic.py             |   9 +-
 outlines/models/dottxt.py                |  73 +++++++++------
 tests/models/test_dottxt.py              |  31 ++++---
 tests/models/test_dottxt_type_adapter.py | 113 +++++++++++++++++++++++
 4 files changed, 182 insertions(+), 44 deletions(-)
 create mode 100644 tests/models/test_dottxt_type_adapter.py

diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 5fee17e9a..75123c4af 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -79,9 +79,12 @@ def format_vision_input(self, model_input: Vision):
 
     def format_output_type(self, output_type):
         """Not implemented for Anthropic."""
-        raise NotImplementedError(
-            f"The output type {output_type} is not available with Anthropic."
-        )
+        if output_type is None:
+            return {}
+        else:
+            raise NotImplementedError(
+                f"The output type {output_type} is not available with Anthropic."
+            )
 
 
 class Anthropic(Model):
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index c725222a6..db3385a73 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -1,12 +1,16 @@
 """Integration with Dottxt's API."""
 
+from dataclasses import is_dataclass
 import json
 from functools import singledispatchmethod
-from types import NoneType
 from typing import Optional, TYPE_CHECKING
 
+from pydantic import BaseModel, TypeAdapter
+from typing_extensions import _TypedDictMeta  # type: ignore
+
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import JsonType
+from outlines.types import Regex, CFG, JsonSchema
+
 
 if TYPE_CHECKING:
     from dottxt import Dottxt as DottxtClient
@@ -15,7 +19,6 @@
 
 
 class DottxtTypeAdapter(ModelTypeAdapter):
-    @singledispatchmethod
     def format_input(self, model_input):
         """Generate the `messages` argument to pass to the client.
 
@@ -25,37 +28,49 @@ def format_input(self, model_input):
             The input passed by the user.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available with Dottxt. The only available type is `str`."
+        if isinstance(model_input, str):
+            return model_input
+        raise TypeError(
+            f"The input type {model_input} is not available with Dottxt. The only available type is `str`."
         )
 
-    @format_input.register(str)
-    def format_str_input(self, model_input: str):
-        """Generate the `messages` argument to pass to the client when the user
-        only passes a prompt.
-
-        """
-        return model_input
-
-    @singledispatchmethod
     def format_output_type(self, output_type):
-        """Format the output type to pass to the client."""
-        raise NotImplementedError(
-            f"The input type {input} is not available with Dottxt."
-        )
+        """Format the output type to pass to the client.
 
-    @format_output_type.register(JsonType)
-    def format_json_output_type(self, output_type: JsonType):
-        """Format the output type to pass to the client."""
-        schema = output_type.to_json_schema()
-        return json.dumps(schema)
+        TODO: `int`, `float` and other Python types could be supported via JSON Schema.
+        """
 
-    @format_output_type.register(NoneType)
-    def format_none_output_type(self, output_type: None):
-        """Format the output type to pass to the client."""
-        raise NotImplementedError(
-            "You must provide an output type. Dottxt only supports constrained generation."
-        )
+        # Unsupported languages
+        if output_type is None:
+            raise TypeError(
+                "You must provide an output type. Dottxt only supports constrained generation."
+            )
+        elif isinstance(output_type, Regex):
+            raise TypeError(
+                "Regex-based structured outputs will soon be available with Dottxt. Use an open source model in the meantime."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs will soon be available with Dottxt. Use an open source model in the meantime."
+            )
+
+        elif isinstance(output_type, JsonSchema):
+            return output_type.schema
+        elif is_dataclass(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return json.dumps(schema)
+        elif isinstance(output_type, _TypedDictMeta):
+            schema = TypeAdapter(output_type).json_schema()
+            return json.dumps(schema)
+        elif isinstance(output_type, type(BaseModel)):
+            schema = output_type.model_json_schema()
+            return json.dumps(schema)
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not supported by Dottxt. "
+                "Consider using a local mode instead."
+            )
 
 
 class Dottxt(Model):
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index 2d995829e..7d4d8eb23 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -2,14 +2,12 @@
 import os
 
 import pytest
-from pydantic import BaseModel
-
 from dottxt.client import Dottxt as DottxtClient
+from pydantic import BaseModel
 
 import outlines
 from outlines import Generator
 from outlines.models.dottxt import Dottxt
-from outlines.types import JsonType
 
 
 class User(BaseModel):
@@ -39,7 +37,7 @@ def test_dottxt_wrong_init_parameters(api_key):
 
 
 def test_dottxt_wrong_output_type(api_key):
-    with pytest.raises(NotImplementedError, match="must provide an output type"):
+    with pytest.raises(TypeError, match="You must provide an output type"):
         client = DottxtClient(api_key=api_key)
         model = Dottxt(client)
         model("prompt")
@@ -54,10 +52,10 @@ def test_dottxt_init_from_client(api_key):
 
 @pytest.mark.api_call
 def test_dottxt_wrong_input_type(api_key):
-    with pytest.raises(NotImplementedError, match="is not available"):
+    with pytest.raises(TypeError, match="You must provide an output type"):
         client = DottxtClient(api_key=api_key)
         model = Dottxt(client)
-        model(["prompt"], JsonType(User))
+        model(["prompt"], User)
 
 
 @pytest.mark.api_call
@@ -65,28 +63,37 @@ def test_dottxt_wrong_inference_parameters(api_key):
     with pytest.raises(TypeError, match="got an unexpected"):
         client = DottxtClient(api_key=api_key)
         model = Dottxt(client)
-        model("prompt", JsonType(User), foo=10)
+        model("prompt", User, foo=10)
+
+
+@pytest.mark.api_call
+def test_dottxt_direct_pydantic_call(api_key):
+    client = DottxtClient(api_key=api_key)
+    model = Dottxt(client)
+    result = model("Create a user", User)
+    assert "first_name" in json.loads(result)
 
 
 @pytest.mark.api_call
-def test_dottxt_direct_call(api_key):
+def test_dottxt_direct_jsonschema_call(api_key):
     client = DottxtClient(api_key=api_key)
     model = Dottxt(client)
-    result = model("Create a user", JsonType(User))
+    result = model("Create a user", User)
     assert "first_name" in json.loads(result)
 
 
 @pytest.mark.api_call
-def test_dottxt_generator_call(api_key):
+def test_dottxt_generator_pydantic_call(api_key):
     client = DottxtClient(api_key=api_key)
     model = Dottxt(client)
-    generator = Generator(model, JsonType(User))
+    generator = Generator(model, User)
     result = generator("Create a user")
     assert "first_name" in json.loads(result)
 
 
+@pytest.mark.api_call
 def test_dottxt_streaming(api_key):
     client = DottxtClient(api_key=api_key)
     model = Dottxt(client)
     with pytest.raises(NotImplementedError, match="Dottxt does not support streaming"):
-        model.stream("Create a user", JsonType(User))
+        model.stream("Create a user", User)
diff --git a/tests/models/test_dottxt_type_adapter.py b/tests/models/test_dottxt_type_adapter.py
new file mode 100644
index 000000000..ea8269adc
--- /dev/null
+++ b/tests/models/test_dottxt_type_adapter.py
@@ -0,0 +1,113 @@
+import io
+from dataclasses import dataclass
+import json
+import pytest
+from typing_extensions import TypedDict
+
+from PIL import Image
+from pydantic import BaseModel
+
+from outlines.models.dottxt import DottxtTypeAdapter
+from outlines.types import regex, cfg, json_schema
+from outlines.templates import Vision
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return DottxtTypeAdapter()
+
+
+def test_dottxt_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert result == message
+
+
+def test_dottxt_type_adapter_input_invalid(adapter, image):
+    prompt = Vision("prompt", image)
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_dottxt_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="You must provide an output type"):
+        adapter.format_output_type(None)
+
+    with pytest.raises(TypeError, match="The type `str` is not supported"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not supported"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="Regex-based structured outputs will soon be"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs will soon be"):
+        adapter.format_output_type(cfg(""))
+
+
+def test_dottxt_type_adapter_output_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_output_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_output_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_json_schema_str(adapter, schema):
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(json_schema(schema_str))
+    assert result == json.dumps(schema)
+
+
+def test_dottxt_type_adapter_json_schema_dict(adapter, schema):
+    result = adapter.format_output_type(json_schema(schema))
+    assert result == json.dumps(schema)

From b072209214e9784c887a6cca047c3b9e518bcd83 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Tue, 25 Feb 2025 09:20:32 +0100
Subject: [PATCH 263/505] Simplify handling model inputs in
 `AnthropicTypeAdapter`

---
 outlines/models/anthropic.py | 22 +++++++---------------
 1 file changed, 7 insertions(+), 15 deletions(-)

diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 75123c4af..2d3efaea2 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -1,6 +1,5 @@
 """Integration with Anthropic's API."""
 
-from functools import singledispatchmethod
 from typing import Union, TYPE_CHECKING
 
 from outlines.models.base import Model, ModelTypeAdapter
@@ -23,7 +22,6 @@ class AnthropicTypeAdapter(ModelTypeAdapter):
 
     """
 
-    @singledispatchmethod
     def format_input(self, model_input):
         """Generate the `messages` argument to pass to the client.
 
@@ -33,16 +31,15 @@ def format_input(self, model_input):
             The input passed by the user.
 
         """
-        raise NotImplementedError(
+        if isinstance(model_input, str):
+            return self.format_str_model_input(model_input)
+        elif isinstance(model_input, Vision):
+            return self.format_vision_model_input(model_input)
+        raise TypeError(
             f"The input type {input} is not available with Anthropic. The only available types are `str` and `Vision`."
         )
 
-    @format_input.register(str)
-    def format_str_input(self, model_input: str):
-        """Generate the `messages` argument to pass to the client when the user
-        only passes a prompt.
-
-        """
+    def format_str_model_input(self, model_input):
         return {
             "messages": [
                 {
@@ -52,12 +49,7 @@ def format_str_input(self, model_input: str):
             ]
         }
 
-    @format_input.register(Vision)
-    def format_vision_input(self, model_input: Vision):
-        """Generate the `messages` argument to pass to the client when the user
-        passes a prompt and an image.
-
-        """
+    def format_vision_model_input(self, model_input):
         return {
             "messages": [
                 {

From 003a8200b193eb9e9a031d8de6511e4a9bb4de41 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Tue, 25 Feb 2025 09:28:25 +0100
Subject: [PATCH 264/505] Handle `OpenAI` output types in `OpenAITypeHandler`

---
 outlines/__init__.py                     |   2 +
 outlines/models/gemini.py                |   1 -
 outlines/models/ollama.py                |   7 +-
 outlines/models/openai.py                |  93 +++++++++----
 outlines/types/__init__.py               |   5 +-
 tests/models/test_openai.py              |  58 +++++---
 tests/models/test_openai_type_adapter.py | 165 +++++++++++++++++++++++
 7 files changed, 274 insertions(+), 57 deletions(-)
 create mode 100644 tests/models/test_openai_type_adapter.py

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 1fce45e96..2b2ad836f 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -9,6 +9,8 @@
 from outlines.function import Function
 from outlines.generator import Generator
 from outlines.templates import Template, prompt
+from outlines.types import regex, json_schema, cfg
+from outlines.templates import Vision
 
 from outlines.models import (
     from_dottxt,
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index c1ac1b075..699e392ec 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -2,7 +2,6 @@
 
 from dataclasses import is_dataclass
 from enum import EnumMeta, Enum
-from types import NoneType
 from typing import (
     get_args,
     get_origin,
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index ee45660bd..8ca0b3175 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,5 +1,4 @@
 from functools import singledispatchmethod
-from types import NoneType
 from typing import Iterator, TYPE_CHECKING
 
 from outlines.models.base import Model, ModelTypeAdapter
@@ -48,9 +47,9 @@ def format_output_type(self, output_type):
             "Ollama only supports structured output with `Json`."
         )
 
-    @format_output_type.register(NoneType)
-    def format_none_output_type(self, output_type: None):
-        return ""
+    # @format_output_type.register(NoneType)
+    # def format_none_output_type(self, output_type: None):
+    #     return ""
 
     @format_output_type.register(JsonType)
     def format_json_output_type(self, output_type: JsonType):
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 887744741..71020e958 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,15 +1,15 @@
 """Integration with OpenAI's API."""
 
-from functools import singledispatchmethod
-from types import NoneType
+from dataclasses import is_dataclass
+import json
 from typing import Optional, Union, TYPE_CHECKING
+from typing_extensions import is_typeddict
 
-
-from pydantic import BaseModel
+from pydantic import BaseModel, TypeAdapter
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
-from outlines.types import JsonType
+from outlines.types import JsonSchema, Regex, CFG
 
 if TYPE_CHECKING:
     from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
@@ -26,7 +26,6 @@ class OpenAITypeAdapter(ModelTypeAdapter):
 
     """
 
-    @singledispatchmethod
     def format_input(self, model_input):
         """Generate the `messages` argument to pass to the client.
 
@@ -36,12 +35,15 @@ def format_input(self, model_input):
             The input passed by the user.
 
         """
-        raise NotImplementedError(
+        if isinstance(model_input, str):
+            return self.format_str_model_input(model_input)
+        elif isinstance(model_input, Vision):
+            return self.format_vision_model_input(model_input)
+        raise TypeError(
             f"The input type {input} is not available with OpenAI. The only available types are `str` and `Vision`."
         )
 
-    @format_input.register(str)
-    def format_str_input(self, model_input: str):
+    def format_str_model_input(self, model_input: str):
         """Generate the `messages` argument to pass to the client when the user
         only passes a prompt.
 
@@ -55,8 +57,7 @@ def format_str_input(self, model_input: str):
             ]
         }
 
-    @format_input.register(Vision)
-    def format_vision_input(self, model_input: Vision):
+    def format_vision_model_input(self, model_input: Vision):
         """Generate the `messages` argument to pass to the client when the user
         passes a prompt and an image.
 
@@ -78,31 +79,48 @@ def format_vision_input(self, model_input: Vision):
             ]
         }
 
-    @singledispatchmethod
     def format_output_type(self, output_type):
         """Generate the `response_format` argument to the client based on the
         output type specified by the user.
 
-        """
-        raise NotImplementedError(
-            f"The type {output_type} is not available with OpenAI. The only output type available is `Json`."
-        )
-
-    @format_output_type.register(NoneType)
-    def format_none_output_type(self, _: None):
-        """Generate the `response_format` argument to the client when no
-        output type is specified by the user.
+        TODO: `int`, `float` and other Python types could be supported via JSON Schema.
 
         """
-        return {}
 
-    @format_output_type.register(JsonType)
-    def format_json_output_type(self, output_type: JsonType):
+        # Unsupported languages
+        if isinstance(output_type, Regex):
+            raise TypeError(
+                "Neither regex-based structured outputs nor the `pattern` keyword in Json Schema are available with OpenAI. Use an open source model or dottxt instead."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs are not available with OpenAI. Use an open source model or dottxt instead."
+            )
+
+        if output_type is None:
+            return {}
+        elif is_dataclass(output_type):
+            output_type = TypeAdapter(output_type).json_schema()
+            return self.format_json_output_type(output_type)
+        elif is_typeddict(output_type):
+            output_type = TypeAdapter(output_type).json_schema()
+            return self.format_json_output_type(output_type)
+        elif isinstance(output_type, type(BaseModel)):
+            output_type = output_type.model_json_schema()
+            return self.format_json_output_type(output_type)
+        elif isinstance(output_type, JsonSchema):
+            return self.format_json_output_type(json.loads(output_type.schema))
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not available with OpenAI. Use an open source model or dottxt instead."
+            )
+
+    def format_json_output_type(self, schema: dict):
         """Generate the `response_format` argument to the client when the user
         specified a `Json` output type.
 
         """
-        schema = output_type.to_json_schema()
 
         # OpenAI requires `additionalProperties` to be set
         if "additionalProperties" not in schema:
@@ -143,13 +161,30 @@ def generate(
         output_type: Optional[Union[type[BaseModel], str]] = None,
         **inference_kwargs,
     ):
+        import openai
+
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
-        result = self.client.chat.completions.create(
-            model=self.model_name, **messages, **response_format, **inference_kwargs
-        )
 
-        return result.choices[0].message.content
+        try:
+            result = self.client.chat.completions.create(
+                model=self.model_name, **messages, **response_format, **inference_kwargs
+            )
+        except openai.BadRequestError as e:
+            if e.body["message"].startswith("Invalid schema"):
+                raise TypeError(
+                    f"OpenAI does not support your schema: {e.body['message']}. Try a local model or dottxt instead."
+                )
+            else:
+                raise e
+
+        message = result.choices[0].message
+        if message.refusal is not None:
+            raise ValueError(
+                f"OpenAI refused to answer the request: {result.choices[0].refusal}"
+            )
+
+        return message.content
 
     def generate_stream(
         self,
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 3b69673cc..a181c77fd 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -15,8 +15,8 @@
     JsonSchema,
     regex,
     json_schema,
-    either,
     optional,
+    either,
     exactly,
     at_least,
     at_most,
@@ -146,3 +146,6 @@ class CFG:
     """Represents a Context-Free Grammar as a string."""
 
     definition: str
+
+
+cfg = CFG
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 92fbccddb..df9182269 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,17 +1,18 @@
 import io
 import json
 import os
-from typing import Generator
-import PIL
+from typing import Annotated, Generator
+
 import pytest
 import requests
+from PIL import Image
 from openai import OpenAI as OpenAIClient
-from pydantic import BaseModel
+from pydantic import BaseModel, Field
 
 import outlines
 from outlines.models.openai import OpenAI
 from outlines.templates import Vision
-from outlines.types import JsonType
+from outlines.types import json_schema
 
 MODEL_NAME = "gpt-4o-mini-2024-07-18"
 
@@ -30,6 +31,21 @@ def api_key():
     return api_key
 
 
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
 def test_init_from_client(api_key):
     client = OpenAIClient(api_key=api_key)
     model = outlines.from_openai(client, "gpt-4o")
@@ -58,7 +74,7 @@ class Foo:
         def __init__(self, foo):
             self.foo = foo
 
-    with pytest.raises(NotImplementedError, match="is not available"):
+    with pytest.raises(TypeError, match="is not available"):
         model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate("prompt", Foo(1))
 
@@ -78,14 +94,8 @@ def test_openai_direct_call():
 
 
 @pytest.mark.api_call
-def test_openai_simple_vision():
+def test_openai_simple_vision(image):
     model = OpenAI(OpenAIClient(), MODEL_NAME)
-
-    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
-    r = requests.get(url, stream=True)
-    if r.status_code == 200:
-        image = PIL.Image.open(io.BytesIO(r.content))
-
     result = model.generate(Vision("What does this logo represent?", image))
     assert isinstance(result, str)
 
@@ -97,26 +107,30 @@ def test_openai_simple_pydantic():
     class Foo(BaseModel):
         bar: int
 
-    result = model.generate("foo?", JsonType(Foo))
+    result = model.generate("foo?", Foo)
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
 
 @pytest.mark.api_call
-def test_openai_simple_vision_pydantic():
+def test_openai_simple_pydantic_refusal():
     model = OpenAI(OpenAIClient(), MODEL_NAME)
 
-    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
-    r = requests.get(url, stream=True)
-    if r.status_code == 200:
-        image = PIL.Image.open(io.BytesIO(r.content))
+    class Foo(BaseModel):
+        bar: Annotated[str, Field(int, pattern=r"^\d+$")]
+
+    with pytest.raises(TypeError, match="OpenAI does not support your schema"):
+        _ = model.generate("foo?", Foo)
+
+
+@pytest.mark.api_call
+def test_openai_simple_vision_pydantic(image):
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
 
     class Logo(BaseModel):
         name: int
 
-    result = model.generate(
-        Vision("What does this logo represent?", image), JsonType(Logo)
-    )
+    result = model.generate(Vision("What does this logo represent?", image), Logo)
     assert isinstance(result, str)
     assert "name" in json.loads(result)
 
@@ -130,7 +144,7 @@ class Foo(BaseModel):
 
     schema = json.dumps(Foo.model_json_schema())
 
-    result = model.generate("foo?", JsonType(schema))
+    result = model.generate("foo?", json_schema(schema))
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
new file mode 100644
index 000000000..923695e1c
--- /dev/null
+++ b/tests/models/test_openai_type_adapter.py
@@ -0,0 +1,165 @@
+import io
+import json
+from typing import Literal
+from typing_extensions import TypedDict
+from dataclasses import dataclass
+
+from PIL import Image
+from pydantic import BaseModel
+import pytest
+
+from outlines import json_schema, regex, cfg, Vision
+from outlines.models.openai import OpenAITypeAdapter
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+        "additionalProperties": False,
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return OpenAITypeAdapter()
+
+
+def test_openai_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert isinstance(result, dict)
+    assert len(result["messages"]) == 1
+    assert result["messages"][0]["role"] == "user"
+    assert result["messages"][0]["content"] == message
+
+
+def test_openai_type_adapter_input_vision(adapter, image):
+    input_message = Vision("hello", image)
+    result = adapter.format_input(input_message)
+    assert isinstance(result, dict)
+
+    messages = result["messages"]
+    assert len(messages) == 1
+
+    message = messages[0]
+    assert message["role"] == "user"
+    assert len(message["content"]) == 2
+    assert message["content"][0]["type"] == "text"
+    assert message["content"][0]["text"] == "hello"
+
+    assert message["content"][1]["type"] == "image_url"
+    assert (
+        message["content"][1]["image_url"]["url"]
+        == f"data:image/png;base64,{input_message.image_str}"
+    )
+
+
+def test_dottxt_type_adapter_input_invalid(adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_openai_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="The type `str` is not available"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not available"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="The type `Literal` is not available"):
+        adapter.format_output_type(Literal[1, 2])
+
+    with pytest.raises(TypeError, match="Neither regex-based"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs"):
+        adapter.format_output_type(cfg(""))
+
+    class Foo(BaseModel):
+        bar: str
+
+    with pytest.raises(TypeError, match="The type `list` is not available"):
+        adapter.format_output_type(list[Foo])
+
+
+def test_openai_type_adapter_output_none(adapter):
+    result = adapter.format_output_type(None)
+    assert result == {}
+
+
+def test_openai_type_adapter_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_json_schema_str(adapter, schema):
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(json_schema(schema_str))
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema
+
+
+def test_openai_type_adapter_json_schema_dict(adapter, schema):
+    result = adapter.format_output_type(json_schema(schema))
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    assert result["response_format"]["json_schema"]["schema"] == schema

From ab72637971b540cb9a87964ce6697895caa74f4f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 27 Feb 2025 10:59:25 +0100
Subject: [PATCH 265/505] Handle `Ollama` output types in `OllamaTypeAdapter`

---
 outlines/models/ollama.py                |  66 ++++++++------
 tests/models/test_ollama.py              |  21 +++--
 tests/models/test_ollama_type_adapter.py | 110 +++++++++++++++++++++++
 3 files changed, 160 insertions(+), 37 deletions(-)
 create mode 100644 tests/models/test_ollama_type_adapter.py

diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 8ca0b3175..d5a348c16 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,8 +1,12 @@
-from functools import singledispatchmethod
+from dataclasses import is_dataclass
+import json
 from typing import Iterator, TYPE_CHECKING
 
+from pydantic import BaseModel, TypeAdapter
+from typing_extensions import _TypedDictMeta  # type: ignore
+
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import JsonType
+from outlines.types import Regex, CFG, JsonSchema
 
 if TYPE_CHECKING:
     from ollama import Client as OllamaClient
@@ -14,7 +18,6 @@
 class OllamaTypeAdapter(ModelTypeAdapter):
     """Type adapter for the Ollama model."""
 
-    @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
 
@@ -24,36 +27,47 @@ def format_input(self, model_input):
             The input passed by the user.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available. "
+        if isinstance(model_input, str):
+            return model_input
+        raise TypeError(
+            f"The input type {model_input} is not available. "
             "Ollama does not support batch inference."
         )
 
-    @format_input.register(str)
-    def format_str_input(self, model_input: str):
-        return model_input
-
-    @singledispatchmethod
     def format_output_type(self, output_type):
-        """Generate the `format` argument to pass to the model.
+        """Format the output type to pass to the client.
 
-        Argument
-        --------
-        output_type
-            The output type passed by the user.
+        TODO: `int`, `float` and other Python types could be supported via JSON Schema.
         """
-        raise NotImplementedError(
-            f"The output type {input} is not available. "
-            "Ollama only supports structured output with `Json`."
-        )
-
-    # @format_output_type.register(NoneType)
-    # def format_none_output_type(self, output_type: None):
-    #     return ""
 
-    @format_output_type.register(JsonType)
-    def format_json_output_type(self, output_type: JsonType):
-        return output_type.to_json_schema()
+        if isinstance(output_type, Regex):
+            raise TypeError(
+                "Regex-based structured outputs are not supported by Ollama. Use an open source model in the meantime."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs are not supported by Ollama. Use an open source model in the meantime."
+            )
+
+        if output_type is None:
+            return None
+        elif isinstance(output_type, JsonSchema):
+            return json.loads(output_type.schema)
+        elif is_dataclass(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return schema
+        elif isinstance(output_type, _TypedDictMeta):
+            schema = TypeAdapter(output_type).json_schema()
+            return schema
+        elif isinstance(output_type, type(BaseModel)):
+            schema = output_type.model_json_schema()
+            return schema
+        else:
+            type_name = getattr(output_type, "__name__", output_type)
+            raise TypeError(
+                f"The type `{type_name}` is not supported by Ollama. "
+                "Consider using a local model instead."
+            )
 
 
 class Ollama(Model):
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index 68afa5640..9f4050e9e 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -1,13 +1,14 @@
 import json
 from enum import Enum
+from typing import Annotated
 
 import pytest
 from ollama import Client as OllamaClient
-from pydantic import BaseModel
+from pydantic import BaseModel, Field
 
 import outlines
 from outlines.models import Ollama
-from outlines.types import Choice, JsonType
+
 
 MODEL_NAME = "tinyllama"
 CLIENT = OllamaClient()
@@ -40,11 +41,9 @@ def test_ollama_direct():
 
 def test_ollama_json():
     class Foo(BaseModel):
-        foo: str
+        foo: Annotated[str, Field(max_length=1)]
 
-    result = Ollama(CLIENT, MODEL_NAME)(
-        "Respond with one word. Not more.", JsonType(Foo)
-    )
+    result = Ollama(CLIENT, MODEL_NAME)("Respond with one word. Not more.", Foo)
     assert isinstance(result, str)
     assert "foo" in json.loads(result)
 
@@ -54,12 +53,12 @@ class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
 
-    with pytest.raises(NotImplementedError, match="is not available"):
-        Ollama(CLIENT, MODEL_NAME).generate("foo?", Choice(Foo))
+    with pytest.raises(TypeError, match="is not supported"):
+        Ollama(CLIENT, MODEL_NAME).generate("foo?", Foo)
 
 
 def test_ollama_wrong_input_type():
-    with pytest.raises(NotImplementedError, match="is not available"):
+    with pytest.raises(TypeError, match="is not available"):
         Ollama(CLIENT, MODEL_NAME).generate(["foo?", "bar?"], None)
 
 
@@ -71,10 +70,10 @@ def test_ollama_stream():
 
 def test_ollama_stream_json():
     class Foo(BaseModel):
-        foo: str
+        foo: Annotated[str, Field(max_length=2)]
 
     model = Ollama(CLIENT, MODEL_NAME)
-    generator = model.stream("Create a character.", JsonType(Foo))
+    generator = model.stream("Create a character.", Foo)
     generated_text = []
     for text in generator:
         generated_text.append(text)
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
new file mode 100644
index 000000000..614266c35
--- /dev/null
+++ b/tests/models/test_ollama_type_adapter.py
@@ -0,0 +1,110 @@
+import io
+from dataclasses import dataclass
+import json
+import pytest
+from typing_extensions import TypedDict
+
+from PIL import Image
+from pydantic import BaseModel
+
+from outlines.models.ollama import OllamaTypeAdapter
+from outlines.types import regex, cfg, json_schema
+from outlines.templates import Vision
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+    }
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return OllamaTypeAdapter()
+
+
+def test_ollama_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert result == message
+
+
+def test_ollama_type_adapter_input_invalid(adapter, image):
+    prompt = Vision("prompt", image)
+    with pytest.raises(TypeError, match="The input type"):
+        _ = adapter.format_input(prompt)
+
+
+def test_ollama_type_adapter_output_invalid(adapter):
+    with pytest.raises(TypeError, match="The type `str` is not supported"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type `int` is not supported"):
+        adapter.format_output_type(int)
+
+    with pytest.raises(TypeError, match="Regex-based structured outputs are not"):
+        adapter.format_output_type(regex("[0-9]"))
+
+    with pytest.raises(TypeError, match="CFG-based structured outputs are not"):
+        adapter.format_output_type(cfg(""))
+
+
+def test_ollama_type_adapter_output_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_ollama_type_adapter_output_typed_dict(adapter, schema):
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_ollama_type_adapter_output_pydantic(adapter, schema):
+    class User(BaseModel):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert result == json.dumps(schema)
+
+
+def test_ollama_type_adapter_json_schema_str(adapter, schema):
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(json_schema(schema_str))
+    assert result == json.dumps(schema)
+
+
+def test_ollama_type_adapter_json_schema_dict(adapter, schema):
+    result = adapter.format_output_type(json_schema(schema))
+    assert result == json.dumps(schema)

From 525df5a4eac8cb1d72f244be3da460987a1d920f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Tue, 25 Feb 2025 10:02:00 +0100
Subject: [PATCH 266/505] Convert Python types to `Term`s

---
 outlines/__init__.py  |   1 -
 outlines/generator.py |   6 +-
 outlines/types/dsl.py | 174 +++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 176 insertions(+), 5 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 2b2ad836f..07cb89f70 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -4,7 +4,6 @@
 import outlines.models
 import outlines.processors
 import outlines.types
-from outlines.types import Choice, Regex, JsonType
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function
 from outlines.generator import Generator
diff --git a/outlines/generator.py b/outlines/generator.py
index 462066531..9546f5ef2 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -59,9 +59,13 @@ class SteerableGenerator:
     """
 
     model: SteerableModel
-    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG, interegular.fsm.FSM]]
+    output_type: Optional[Any]
 
     def __post_init__(self):
+        # If CFG -> CFG
+        # if dict -> CFG
+        # if Any -> None
+        # Else -> Regex
         if self.output_type is None:
             self.logits_processor = None
         else:
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 7f8ce0c16..cef66a443 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,9 +1,14 @@
-import json as json
+from dataclasses import is_dataclass
+import datetime
+from enum import EnumMeta
+import json
 import re
 from dataclasses import dataclass
-from typing import Any, List, Union
+from typing import Any, List, Union, get_origin, Literal, get_args
+from typing_extensions import _TypedDictMeta  # type: ignore
+import outlines.types as types
 
-from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler
+from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
 from pydantic.json_schema import JsonSchemaValue
 from pydantic_core import core_schema as cs
 from outlines_core.fsm.json_schema import build_regex_from_schema
@@ -377,6 +382,169 @@ def one_or_more(term: Union[Term, str]) -> KleenePlus:
     return KleenePlus(term)
 
 
+def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
+    """Convert Python types to Outlines DSL terms that constrain LLM output.
+
+    Arguments
+    ---------
+    ptype
+        The Python type to convert
+    recursion_depth
+        Current recursion depth to prevent infinite recursion
+
+    Returns
+    -------
+    The corresponding DSL `Term` instance.
+    """
+    if recursion_depth > 10:
+        raise RecursionError(
+            f"Maximum recursion depth exceeded when converting {ptype}. "
+            "This might be due to a recursive type definition."
+        )
+
+    # First handle Term instances
+    if isinstance(ptype, Term):
+        return ptype
+
+    # Basic types
+    if ptype is int or get_origin(ptype) is int:
+        return types.integer
+    elif ptype is float or get_origin(ptype) is float:
+        return types.number
+    elif ptype is bool or get_origin(ptype) is bool:
+        return types.boolean
+    elif ptype is str or get_origin(ptype) is str:
+        return String(ptype)
+    elif ptype is datetime.time:
+        return types.time
+    elif ptype is datetime.date:
+        return types.date
+    elif ptype is datetime.datetime:
+        return types.datetime
+
+    # Structured types
+    structured_type_checks = [
+        lambda x: is_dataclass(x),
+        lambda x: isinstance(x, _TypedDictMeta),
+        lambda x: isinstance(x, BaseModel),
+    ]
+    if any(check(ptype) for check in structured_type_checks):
+        schema = TypeAdapter(ptype).json_schema()
+        return JsonSchema(schema)
+
+    # Enums
+    if isinstance(ptype, EnumMeta):
+        return Alternatives(
+            [python_types_to_terms(arg.value, recursion_depth + 1) for arg in ptype]  # type: ignore
+        )
+
+    # Generic types
+    origin = get_origin(ptype)
+    args = get_origin(ptype)
+
+    if origin is Literal:
+        return _handle_literal(args)
+    elif origin is Union:
+        return _handle_union(args, recursion_depth)
+    elif origin is list:
+        return _handle_list(args, recursion_depth)
+    elif origin is tuple:
+        return _handle_tuple(args, recursion_depth)
+    elif ptype is dict:
+        return _handle_dict(args, recursion_depth)
+    else:
+        type_name = getattr(ptype, "__name__", ptype)
+        raise TypeError(
+            f"Type {type_name} is currently not supported. Please open an issue: "
+            "https://github.com/dottxt-ai/outlines/issues"
+        )
+
+
+def _handle_literal(args: tuple) -> Alternatives:
+    return Alternatives([python_types_to_terms(arg) for arg in args])
+
+
+def _handle_union(args: tuple, recursion_depth: int) -> Alternatives:
+    # Handle the Optional[T] type
+    if len(args) == 2 and (type(None) in args or None in args):
+        other_ptype = next(arg for arg in args if arg not in (type(None), None))
+        return Alternatives(
+            [
+                python_types_to_terms(other_ptype, recursion_depth + 1),
+                String("None"),
+            ]
+        )
+    return Alternatives(
+        [python_types_to_terms(arg, recursion_depth + 1) for arg in args]
+    )
+
+
+def _handle_list(args: tuple, recursion_depth: int) -> Sequence:
+    if args is None or len(args) > 1:
+        raise TypeError(
+            f"Only homogeneous lists are supported. Got multiple type arguments {args}."
+        )
+    item_type = python_types_to_terms(args[0], recursion_depth + 1)
+    return Sequence(
+        [
+            String("["),
+            item_type,
+            KleeneStar(Sequence([String(","), item_type])),
+            String("]"),
+        ]
+    )
+
+
+def _handle_tuple(args: tuple, recursion_depth: int) -> Sequence:
+    if args is None:
+        return String("()")
+    elif len(args) == 2 and args[1] is Ellipsis:
+        item_term = python_types_to_terms(args[0], recursion_depth + 1)
+        return Sequence(
+            [
+                String("("),
+                item_term,
+                KleeneStar(Sequence([String(","), item_term])),
+                String(")"),
+            ]
+        )
+    else:
+        items = [python_types_to_terms(arg, recursion_depth + 1) for arg in args]
+        separator = String(", ")
+        elements = []
+        for i, item in enumerate(items):
+            elements.append(item)
+            if i < len(items) - 1:
+                elements.append(separator)
+        return Sequence([String("("), *elements, String(")")])
+
+
+def _handle_dict(args: tuple, recursion_depth: int) -> Sequence:
+    if args is None or len(args) != 2:
+        raise TypeError(f"Dict must have exactly two type arguments. Got {args}.")
+    # Add dict support with key:value pairs
+    key_type = python_types_to_terms(args[0], recursion_depth + 1)
+    value_type = python_types_to_terms(args[1], recursion_depth + 1)
+    return Sequence(
+        [
+            String("{"),
+            Optional(
+                Sequence(
+                    [
+                        key_type,
+                        String(":"),
+                        value_type,
+                        KleeneStar(
+                            Sequence([String(","), key_type, String(":"), value_type])
+                        ),
+                    ]
+                )
+            ),
+            String("}"),
+        ]
+    )
+
+
 def to_regex(term: Term) -> str:
     """Convert a term to a regular expression.
 

From 87cf4f3a60e1fa8c2abad0af645461fa51410785 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 12:33:23 +0100
Subject: [PATCH 267/505] Use jsonschema to validate JSON schemas passed by the
 user

---
 outlines/types/dsl.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index cef66a443..48cd0cca7 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -8,6 +8,7 @@
 from typing_extensions import _TypedDictMeta  # type: ignore
 import outlines.types as types
 
+import jsonschema
 from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
 from pydantic.json_schema import JsonSchemaValue
 from pydantic_core import core_schema as cs
@@ -172,6 +173,9 @@ def __init__(self, schema: Union[dict, str, type[BaseModel]]):
 
         self.schema = schema_str
 
+    def __post_init__(self):
+        jsonschema.Draft7Validator.check_schema(json.loads(self.schema))
+
     def _display_node(self) -> str:
         return f"JsonSchema('{self.schema}')"
 
@@ -426,7 +430,7 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     structured_type_checks = [
         lambda x: is_dataclass(x),
         lambda x: isinstance(x, _TypedDictMeta),
-        lambda x: isinstance(x, BaseModel),
+        lambda x: issubclass(x, BaseModel),
     ]
     if any(check(ptype) for check in structured_type_checks):
         schema = TypeAdapter(ptype).json_schema()

From ba93829971db58d9556ddfc2ba6967bb93489b83 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 12:54:49 +0100
Subject: [PATCH 268/505] Turn CFG into a Term

---
 outlines/types/__init__.py | 13 +++----------
 outlines/types/dsl.py      | 15 +++++++++++++++
 2 files changed, 18 insertions(+), 10 deletions(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index a181c77fd..5e190dcaf 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -12,8 +12,10 @@
 from . import airports, countries, locale
 from outlines.types.dsl import (
     Regex,
+    CFG,
     JsonSchema,
     regex,
+    cfg,
     json_schema,
     optional,
     either,
@@ -21,6 +23,7 @@
     at_least,
     at_most,
     between,
+    repeat,
     zero_or_more,
     one_or_more,
 )
@@ -139,13 +142,3 @@ def to_regex(self):
         choices = self.to_list()
         regex_str = r"(" + r"|".join(choices) + r")"
         return regex_str
-
-
-@dataclass
-class CFG:
-    """Represents a Context-Free Grammar as a string."""
-
-    definition: str
-
-
-cfg = CFG
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 48cd0cca7..53d08acf6 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -156,6 +156,17 @@ def to_regex(self) -> str:
         return self.pattern
 
 
+@dataclass
+class CFG(Term):
+    definition: str
+
+    def _display_node(self) -> str:
+        return f"CFG('{self.definition}')"
+
+    def __repr__(self):
+        return f"CFG(definition='{self.definition}')"
+
+
 class JsonSchema(Term):
     def __init__(self, schema: Union[dict, str, type[BaseModel]]):
         if isinstance(schema, dict):
@@ -334,6 +345,10 @@ def regex(pattern: str):
     return Regex(pattern)
 
 
+def cfg(definition: str):
+    return CFG(definition)
+
+
 def json_schema(schema: Union[str, dict, type[BaseModel]]):
     return JsonSchema(schema)
 

From ebdb6f39cc37798783e135752691a7db45c653c7 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 14:42:36 +0100
Subject: [PATCH 269/505] Add support for json mode

---
 outlines/models/openai.py                | 10 ++++++++++
 outlines/types/dsl.py                    |  3 +++
 tests/models/test_openai_type_adapter.py |  5 +++++
 3 files changed, 18 insertions(+)

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 71020e958..8430c8f37 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -99,6 +99,8 @@ def format_output_type(self, output_type):
 
         if output_type is None:
             return {}
+        elif output_type is dict:
+            return self.format_json_mode_type()
         elif is_dataclass(output_type):
             output_type = TypeAdapter(output_type).json_schema()
             return self.format_json_output_type(output_type)
@@ -137,6 +139,14 @@ def format_json_output_type(self, schema: dict):
             }
         }
 
+    def format_json_mode_type(self):
+        """Generate the `response_format` argument to the client when the user
+        specified the output type should be a JSON but without specifying the
+        schema (also called "JSON mode").
+
+        """
+        return {"response_format": {"type": "json_object"}}
+
 
 class OpenAI(Model):
     """Thin wrapper around the `openai.OpenAI` client.
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 53d08acf6..9f0b7e171 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -7,6 +7,7 @@
 from typing import Any, List, Union, get_origin, Literal, get_args
 from typing_extensions import _TypedDictMeta  # type: ignore
 import outlines.types as types
+from outlines import grammars
 
 import jsonschema
 from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
@@ -434,6 +435,8 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
         return types.boolean
     elif ptype is str or get_origin(ptype) is str:
         return String(ptype)
+    elif ptype is dict:
+        return CFG(grammars.json)
     elif ptype is datetime.time:
         return types.time
     elif ptype is datetime.date:
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
index 923695e1c..c3498334c 100644
--- a/tests/models/test_openai_type_adapter.py
+++ b/tests/models/test_openai_type_adapter.py
@@ -116,6 +116,11 @@ def test_openai_type_adapter_output_none(adapter):
     assert result == {}
 
 
+def test_openai_type_adapter_json_mode(adapter):
+    result = adapter.format_output_type(dict)
+    assert result == {"response_format": {"type": "json_object"}}
+
+
 def test_openai_type_adapter_dataclass(adapter, schema):
     @dataclass
     class User:

From 59d87083132c14c151d79dc04fc4ef68ecb7c9e1 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 16:41:18 +0100
Subject: [PATCH 270/505] Add support for callables and for callables in enums

---
 outlines/types/dsl.py | 27 ++++++++++++++++++++++++---
 1 file changed, 24 insertions(+), 3 deletions(-)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 9f0b7e171..163919194 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -4,10 +4,12 @@
 import json
 import re
 from dataclasses import dataclass
-from typing import Any, List, Union, get_origin, Literal, get_args
+from typing import Any, List, Union, get_origin, Literal, get_args, Callable
 from typing_extensions import _TypedDictMeta  # type: ignore
 import outlines.types as types
 from outlines import grammars
+from types import FunctionType
+from outlines.fsm.json_schema import get_schema_from_signature
 
 import jsonschema
 from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
@@ -448,7 +450,7 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     structured_type_checks = [
         lambda x: is_dataclass(x),
         lambda x: isinstance(x, _TypedDictMeta),
-        lambda x: issubclass(x, BaseModel),
+        lambda x: isinstance(x, type) and issubclass(x, BaseModel),
     ]
     if any(check(ptype) for check in structured_type_checks):
         schema = TypeAdapter(ptype).json_schema()
@@ -457,9 +459,16 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     # Enums
     if isinstance(ptype, EnumMeta):
         return Alternatives(
-            [python_types_to_terms(arg.value, recursion_depth + 1) for arg in ptype]  # type: ignore
+            [
+                python_types_to_terms(member, recursion_depth + 1)
+                for member in _get_enum_members(ptype)
+            ]
         )
 
+    # Callables
+    if callable(ptype) and not isinstance(ptype, type):
+        return JsonSchema(get_schema_from_signature(ptype))
+
     # Generic types
     origin = get_origin(ptype)
     args = get_origin(ptype)
@@ -482,6 +491,18 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
         )
 
 
+def _get_enum_members(ptype: EnumMeta) -> List[Any]:
+    regular_members = [member.value for member in ptype]  # type: ignore
+    function_members = [
+        value for key, value in ptype.__dict__.items()
+        if (
+            isinstance(value, FunctionType)
+            and not (key.startswith('__') and key.endswith('__'))
+        )
+    ]
+    return regular_members + function_members
+
+
 def _handle_literal(args: tuple) -> Alternatives:
     return Alternatives([python_types_to_terms(arg) for arg in args])
 

From e590b9d1a6399f1d8a732d82870daf4caf91d296 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 16:52:47 +0100
Subject: [PATCH 271/505] Add support for Interegular FSM

---
 outlines/types/dsl.py | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 163919194..a5c799d8c 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -10,6 +10,7 @@
 from outlines import grammars
 from types import FunctionType
 from outlines.fsm.json_schema import get_schema_from_signature
+import interegular
 
 import jsonschema
 from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
@@ -469,6 +470,10 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     if callable(ptype) and not isinstance(ptype, type):
         return JsonSchema(get_schema_from_signature(ptype))
 
+    # Interegular FSM are passed on as is and are handled by the Generator
+    elif isinstance(ptype, interegular.fsm.FSM):
+        return ptype
+
     # Generic types
     origin = get_origin(ptype)
     args = get_origin(ptype)

From 3e9804bbdef97cc9a5f291df58d8ec027f9a4edd Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 17:13:46 +0100
Subject: [PATCH 272/505] Add support for genSON

---
 outlines/types/dsl.py | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index a5c799d8c..196260851 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -11,6 +11,7 @@
 from types import FunctionType
 from outlines.fsm.json_schema import get_schema_from_signature
 import interegular
+from genson import SchemaBuilder
 
 import jsonschema
 from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
@@ -457,6 +458,11 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
         schema = TypeAdapter(ptype).json_schema()
         return JsonSchema(schema)
 
+    # genSON SchemaBuilder
+    elif isinstance(ptype, SchemaBuilder):
+        schema = ptype.to_json()
+        return JsonSchema(schema)
+
     # Enums
     if isinstance(ptype, EnumMeta):
         return Alternatives(

From e29b1187e7e9618aa650927edd2485634d30c902 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 17:34:25 +0100
Subject: [PATCH 273/505] Fix error with strings in python_types_to_terms

---
 outlines/types/__init__.py | 1 +
 outlines/types/dsl.py      | 2 ++
 2 files changed, 3 insertions(+)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 5e190dcaf..52b28f647 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -30,6 +30,7 @@
 
 
 # Python types
+string = Regex(r'"[^"]*"')
 integer = Regex(r"[+-]?(0|[1-9][0-9]*)")
 boolean = Regex("(True|False)")
 number = Regex(rf"{integer.pattern}(\.[0-9]+)?([eE][+-][0-9]+)?")
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 196260851..1a51e1252 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -438,6 +438,8 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     elif ptype is bool or get_origin(ptype) is bool:
         return types.boolean
     elif ptype is str or get_origin(ptype) is str:
+        return types.string
+    elif isinstance(ptype, str):
         return String(ptype)
     elif ptype is dict:
         return CFG(grammars.json)

From 654fe282e929ac5151513466d416b9f111d3024e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 18:43:55 +0100
Subject: [PATCH 274/505] Add support for basic type instances

---
 outlines/types/dsl.py | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 1a51e1252..919ebe682 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -439,8 +439,6 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
         return types.boolean
     elif ptype is str or get_origin(ptype) is str:
         return types.string
-    elif isinstance(ptype, str):
-        return String(ptype)
     elif ptype is dict:
         return CFG(grammars.json)
     elif ptype is datetime.time:
@@ -450,6 +448,12 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     elif ptype is datetime.datetime:
         return types.datetime
 
+    # Basic type instances
+    if isinstance(ptype, str):
+        return String(ptype)
+    elif isinstance(ptype, (int, float)):
+        return Regex(str(ptype))
+
     # Structured types
     structured_type_checks = [
         lambda x: is_dataclass(x),

From 23a56cc3802650ac4475a2cb479696ed52d93471 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 18:47:24 +0100
Subject: [PATCH 275/505] Update the Generator to use native types

---
 outlines/generator.py | 18 ++++++++++--------
 outlines/types/dsl.py |  3 ---
 2 files changed, 10 insertions(+), 11 deletions(-)

diff --git a/outlines/generator.py b/outlines/generator.py
index 9546f5ef2..2aeb853cd 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -6,7 +6,8 @@
 from outlines.processors.guide import RegexGuide
 from outlines.models import BlackBoxModel, SteerableModel
 from outlines.processors import CFGLogitsProcessor, GuideLogitsProcessor, RegexLogitsProcessor
-from outlines.types import CFG, Choice, JsonType, List, Regex
+from outlines.types import CFG
+from outlines.types.dsl import python_types_to_terms, to_regex
 
 
 @dataclass
@@ -23,7 +24,7 @@ class BlackBoxGenerator:
     """
 
     model: BlackBoxModel
-    output_type: Optional[Union[JsonType, List, Choice, Regex]] = None
+    output_type: Optional[Any] = None
 
     def __post_init__(self):
         if isinstance(self.output_type, CFG):
@@ -69,16 +70,17 @@ def __post_init__(self):
         if self.output_type is None:
             self.logits_processor = None
         else:
-            if isinstance(self.output_type, CFG):
-                cfg_string = self.output_type.definition
+            term = python_types_to_terms(self.output_type)
+            if isinstance(term, CFG):
+                cfg_string = term.definition
                 self.logits_processor = CFGLogitsProcessor(
                     cfg_string, self.model.tokenizer
                 )
-            elif isinstance(self.output_type, interegular.fsm.FSM):
-                guide = RegexGuide.from_interegular_fsm(self.output_type, self.model.tokenizer)
+            elif isinstance(term, interegular.fsm.FSM):
+                guide = RegexGuide.from_interegular_fsm(term, self.model.tokenizer)
                 self.logits_processor = GuideLogitsProcessor(tokenizer=self.model.tokenizer, guide=guide)
             else:
-                regex_string = self.output_type.to_regex()
+                regex_string = to_regex(term)
                 self.logits_processor = RegexLogitsProcessor(
                     regex_string, self.model.tokenizer
                 )
@@ -92,7 +94,7 @@ def stream(self, prompt, **inference_kwargs):
 
 def Generator(
     model: Union[SteerableModel, BlackBoxModel],
-    output_type: Optional[Union[JsonType, List, Choice, Regex, CFG, interegular.fsm.FSM]] = None,
+    output_type: Optional[Any] = None,
 ):
     if isinstance(model, BlackBoxModel):  # type: ignore
         return BlackBoxGenerator(model, output_type)  # type: ignore
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 919ebe682..715f69939 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -157,9 +157,6 @@ def _display_node(self) -> str:
     def __repr__(self):
         return f"Regex(pattern='{self.pattern}')"
 
-    def to_regex(self) -> str:
-        return self.pattern
-
 
 @dataclass
 class CFG(Term):

From 77d2e5a24c1984c59697880d853ebb3be7431774 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 5 Mar 2025 23:07:46 +0100
Subject: [PATCH 276/505] Create unit tests for python_types_to_terms

---
 outlines/types/dsl.py   |  72 +++++-----
 tests/types/test_dsl.py | 294 +++++++++++++++++++++++++++++++++++++++-
 2 files changed, 334 insertions(+), 32 deletions(-)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 715f69939..943d4dd09 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,22 +1,21 @@
-from dataclasses import is_dataclass
 import datetime
-from enum import EnumMeta
 import json
 import re
-from dataclasses import dataclass
-from typing import Any, List, Union, get_origin, Literal, get_args, Callable
-from typing_extensions import _TypedDictMeta  # type: ignore
-import outlines.types as types
-from outlines import grammars
+from dataclasses import dataclass, is_dataclass
+from enum import EnumMeta
 from types import FunctionType
-from outlines.fsm.json_schema import get_schema_from_signature
-import interegular
-from genson import SchemaBuilder
+from typing import Any, Callable, List, Literal, Union, get_args, get_origin, _TypedDictMeta # type: ignore
 
+import interegular
 import jsonschema
+from genson import SchemaBuilder
 from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
 from pydantic.json_schema import JsonSchemaValue
 from pydantic_core import core_schema as cs
+
+import outlines.types as types
+from outlines import grammars
+from outlines.fsm.json_schema import get_schema_from_signature
 from outlines_core.fsm.json_schema import build_regex_from_schema
 
 
@@ -195,6 +194,16 @@ def _display_node(self) -> str:
     def __repr__(self):
         return f"JsonSchema(schema='{self.schema}')"
 
+    def __eq__(self, other):
+        if not isinstance(other, JsonSchema):
+            return False
+        try:
+            self_dict = json.loads(self.schema)
+            other_dict = json.loads(other.schema)
+            return self_dict == other_dict
+        except json.JSONDecodeError:
+            return self.schema == other.schema
+
 
 @dataclass
 class KleeneStar(Term):
@@ -475,17 +484,9 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
             ]
         )
 
-    # Callables
-    if callable(ptype) and not isinstance(ptype, type):
-        return JsonSchema(get_schema_from_signature(ptype))
-
-    # Interegular FSM are passed on as is and are handled by the Generator
-    elif isinstance(ptype, interegular.fsm.FSM):
-        return ptype
-
     # Generic types
     origin = get_origin(ptype)
-    args = get_origin(ptype)
+    args = get_args(ptype)
 
     if origin is Literal:
         return _handle_literal(args)
@@ -495,14 +496,22 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
         return _handle_list(args, recursion_depth)
     elif origin is tuple:
         return _handle_tuple(args, recursion_depth)
-    elif ptype is dict:
+    elif origin is dict:
         return _handle_dict(args, recursion_depth)
-    else:
-        type_name = getattr(ptype, "__name__", ptype)
-        raise TypeError(
-            f"Type {type_name} is currently not supported. Please open an issue: "
-            "https://github.com/dottxt-ai/outlines/issues"
-        )
+
+    # Callables
+    if callable(ptype) and not isinstance(ptype, type):
+        return JsonSchema(get_schema_from_signature(ptype))
+
+    # Interegular FSM are passed on as is and are handled by the Generator
+    elif isinstance(ptype, interegular.fsm.FSM):
+        return ptype
+
+    type_name = getattr(ptype, "__name__", ptype)
+    raise TypeError(
+        f"Type {type_name} is currently not supported. Please open an issue: "
+        "https://github.com/dottxt-ai/outlines/issues"
+    )
 
 
 def _get_enum_members(ptype: EnumMeta) -> List[Any]:
@@ -518,6 +527,7 @@ def _get_enum_members(ptype: EnumMeta) -> List[Any]:
 
 
 def _handle_literal(args: tuple) -> Alternatives:
+    print(args)
     return Alternatives([python_types_to_terms(arg) for arg in args])
 
 
@@ -546,14 +556,14 @@ def _handle_list(args: tuple, recursion_depth: int) -> Sequence:
         [
             String("["),
             item_type,
-            KleeneStar(Sequence([String(","), item_type])),
+            KleeneStar(Sequence([String(", "), item_type])),
             String("]"),
         ]
     )
 
 
-def _handle_tuple(args: tuple, recursion_depth: int) -> Sequence:
-    if args is None:
+def _handle_tuple(args: tuple, recursion_depth: int) -> Union[Sequence, String]:
+    if len(args) == 0:
         return String("()")
     elif len(args) == 2 and args[1] is Ellipsis:
         item_term = python_types_to_terms(args[0], recursion_depth + 1)
@@ -561,7 +571,7 @@ def _handle_tuple(args: tuple, recursion_depth: int) -> Sequence:
             [
                 String("("),
                 item_term,
-                KleeneStar(Sequence([String(","), item_term])),
+                KleeneStar(Sequence([String(", "), item_term])),
                 String(")"),
             ]
         )
@@ -592,7 +602,7 @@ def _handle_dict(args: tuple, recursion_depth: int) -> Sequence:
                         String(":"),
                         value_type,
                         KleeneStar(
-                            Sequence([String(","), key_type, String(":"), value_type])
+                            Sequence([String(", "), key_type, String(":"), value_type])
                         ),
                     ]
                 )
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index 8351ef319..cf7a314db 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -1,6 +1,24 @@
+import datetime
+import json
+from dataclasses import dataclass
+from enum import Enum
+from typing import (
+    Dict,
+    List,
+    Literal,
+    Tuple,
+    TypedDict,
+    Union,
+    get_args,
+    Optional as PyOptional
+)
+
+import interegular
 import pytest
+from genson import SchemaBuilder
 from pydantic import BaseModel
 
+from outlines import grammars, types
 from outlines.types.dsl import (
     Alternatives,
     JsonSchema,
@@ -16,6 +34,12 @@
     String,
     Term,
     either,
+    _handle_dict,
+    _handle_list,
+    _handle_literal,
+    _handle_tuple,
+    _handle_union,
+    json_schema,
     one_or_more,
     zero_or_more,
     optional,
@@ -24,7 +48,8 @@
     at_least,
     exactly,
     regex,
-    json_schema,
+    python_types_to_terms,
+    cfg,
 )
 
 
@@ -297,3 +322,270 @@ def test_dsl_display():
         tree
         == "└── Sequence\n    ├── KleeneStar(*)\n    │   └── Alternatives(|)\n    │       ├── String('a')\n    │       └── String('b')\n    └── Regex('[0-9]')\n"
     )
+
+
+def test_dsl_python_types_to_terms():
+    with pytest.raises(RecursionError):
+        python_types_to_terms(None, 11)
+
+    term = Term()
+    assert python_types_to_terms(term) == term
+
+    assert python_types_to_terms(int) == types.integer
+    assert python_types_to_terms(float) == types.number
+    assert python_types_to_terms(bool) == types.boolean
+    assert python_types_to_terms(str) == types.string
+    assert python_types_to_terms(datetime.time) == types.time
+    assert python_types_to_terms(datetime.date) == types.date
+    assert python_types_to_terms(datetime.datetime) == types.datetime
+    assert python_types_to_terms(dict) == types.CFG(grammars.json)
+
+    string_instance = "a"
+    assert python_types_to_terms(string_instance) == String(string_instance)
+    int_instance = 1
+    assert python_types_to_terms(int_instance) == Regex(r"1")
+    float_instance = 1.0
+    assert python_types_to_terms(float_instance) == Regex(r"1.0")
+
+    @dataclass
+    class DataClass:
+        a: int
+        b: str
+
+    assert python_types_to_terms(DataClass) == JsonSchema(
+        {
+            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
+            "required": ["a", "b"],
+            "title": "DataClass",
+            "type": "object",
+        }
+    )
+
+    class SomeTypedDict(TypedDict):
+        a: int
+        b: str
+
+    assert python_types_to_terms(SomeTypedDict) == JsonSchema(
+        {
+            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
+            "required": ["a", "b"],
+            "title": "SomeTypedDict",
+            "type": "object",
+        }
+    )
+
+    class PydanticModel(BaseModel):
+        a: int
+        b: str
+
+    assert python_types_to_terms(PydanticModel) == JsonSchema(
+        {
+            "properties": {"a": {"title": "A", "type": "integer"}, "b": {"title": "B", "type": "string"}},
+            "required": ["a", "b"],
+            "title": "PydanticModel",
+            "type": "object",
+        }
+    )
+
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+    assert python_types_to_terms(builder) == JsonSchema(
+        {
+            "$schema": "http://json-schema.org/schema#",
+            "type": "object",
+            "properties": {"hi": {"type": ["integer", "string"]}},
+            "required": ["hi"]
+        }
+    )
+
+    interegular_fsm = interegular.fsm.FSM(
+        alphabet={},
+        states={},
+        initial={},
+        finals={},
+        map={},
+        __no_validation__=True,
+    )
+    assert python_types_to_terms(interegular_fsm) is interegular_fsm
+
+    def func(a: int, b: str):
+        return (a, b)
+
+    assert python_types_to_terms(func) == JsonSchema(
+        {
+            "type": "object",
+            "properties": {
+                "a": {"title": "A", "type": "integer"},
+                "b": {"title": "B", "type": "string"},
+            },
+            "required": ["a", "b"],
+            "title": "func",
+        }
+    )
+
+    class SomeEnum(Enum):
+        a = "a"
+        b = int
+        c = func
+
+    result = python_types_to_terms(SomeEnum)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 3
+    assert result.terms[0] == String("a")
+    assert result.terms[1] == types.integer
+    assert isinstance(result.terms[2], JsonSchema)
+    schema_dict = json.loads(result.terms[2].schema)
+    assert schema_dict == {
+        "properties": {
+            "a": {"title": "A", "type": "integer"},
+            "b": {"title": "B", "type": "string"},
+        },
+        "required": ["a", "b"],
+        "title": "func",
+        "type": "object",
+    }
+
+    # for generic types we only test the dispatch as the functions that
+    # convert to terms are tested in distinct tests below
+    assert python_types_to_terms(Literal["a", "b"]) == _handle_literal(("a", "b"))
+    assert python_types_to_terms(Union[int, str]) == _handle_union((int, str), recursion_depth=0)
+    assert python_types_to_terms(list[int]) == _handle_list((int,), recursion_depth=0)
+    assert python_types_to_terms(tuple[int, str]) == _handle_tuple((int, str), recursion_depth=0)
+    assert python_types_to_terms(dict[int, str]) == _handle_dict((int, str), recursion_depth=0)
+
+
+def test_dsl_handle_literal():
+    literal = Literal["a", 1]
+    result = _handle_literal(get_args(literal))
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert result.terms[0] == String("a")
+    assert result.terms[1] == Regex(r"1")
+
+
+def test_dsl_handle_union():
+    # test simple Union
+    simple_union = Union[int, str]
+    result = _handle_union(get_args(simple_union), recursion_depth=0)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert result.terms[0] == types.integer
+    assert result.terms[1] == types.string
+
+    # test with Optional[T]
+    optional_type = PyOptional[int]
+    result = _handle_union(get_args(optional_type), recursion_depth=0)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert result.terms[0] == types.integer
+    assert result.terms[1] == String("None")
+
+    # test with more complex types
+    class TestModel(BaseModel):
+        field: str
+
+    class TestEnum(Enum):
+        a = "a"
+        b = "b"
+
+    complex_union = Union[TestModel, TestEnum]
+    result = _handle_union(get_args(complex_union), recursion_depth=0)
+    assert isinstance(result, Alternatives)
+    assert len(result.terms) == 2
+    assert isinstance(result.terms[0], JsonSchema)
+    assert isinstance(result.terms[1], Alternatives)
+    assert len(result.terms[1].terms) == 2
+    assert result.terms[1].terms[0] == String("a")
+    assert result.terms[1].terms[1] == String("b")
+
+
+def test_dsl_handle_list():
+    with pytest.raises(TypeError):
+        _handle_list((int, str), recursion_depth=0)
+
+    # simple type
+    list_type = list[int]
+    result = _handle_list(get_args(list_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 4
+    assert result.terms[0] == String("[")
+    assert result.terms[1] == types.integer
+    assert isinstance(result.terms[2], KleeneStar)
+    assert result.terms[2].term == Sequence([String(", "), types.integer])
+    assert result.terms[3] == String("]")
+
+    # more complex type
+    list_type = list[Union[int, str]]
+    result = _handle_list(get_args(list_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 4
+    assert result.terms[0] == String("[")
+    assert result.terms[1] == _handle_union(get_args(Union[int, str]), recursion_depth=0)
+    assert isinstance(result.terms[2], KleeneStar)
+    assert result.terms[2].term == Sequence([String(", "), _handle_union(get_args(Union[int, str]), recursion_depth=0)])
+    assert result.terms[3] == String("]")
+
+
+def test_dsl_handle_tuple():
+    # empty tuple
+    tuple_type = Tuple[()]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, String)
+    assert result.value == "()"
+
+    # tuple with ellipsis
+    tuple_type = tuple[int, ...]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 4
+    assert result.terms[0] == String("(")
+    assert result.terms[1] == types.integer
+    assert isinstance(result.terms[2], KleeneStar)
+    assert result.terms[2].term == Sequence([String(", "), types.integer])
+    assert result.terms[3] == String(")")
+
+    # tuple with fixed length
+    tuple_type = tuple[int, str]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 5
+    assert result.terms[0] == String("(")
+    assert result.terms[1] == types.integer
+    assert result.terms[2] == String(", ")
+    assert result.terms[3] == types.string
+    assert result.terms[4] == String(")")
+
+    # tuple with fixed length and complex types
+    tuple_type = tuple[int, Union[str, int]]
+    result = _handle_tuple(get_args(tuple_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 5
+    assert result.terms[0] == String("(")
+    assert result.terms[1] == types.integer
+    assert result.terms[2] == String(", ")
+    assert result.terms[3] == _handle_union(get_args(Union[str, int]), recursion_depth=0)
+    assert result.terms[4] == String(")")
+
+
+def test_dsl_handle_dict():
+    # args of incorrect length
+    with pytest.raises(TypeError):
+        incorrect_dict_type = dict[int, str, int]
+        _handle_dict(get_args(incorrect_dict_type), recursion_depth=0)
+
+    # correct type
+    dict_type = dict[int, str]
+    result = _handle_dict(get_args(dict_type), recursion_depth=0)
+    assert isinstance(result, Sequence)
+    assert len(result.terms) == 3
+    assert result.terms[0] == String("{")
+    assert isinstance(result.terms[1], Optional)
+    assert isinstance(result.terms[1].term, Sequence)
+    assert len(result.terms[1].term.terms) == 4
+    assert result.terms[1].term.terms[0] == types.integer
+    assert result.terms[1].term.terms[1] == String(":")
+    assert result.terms[1].term.terms[2] == types.string
+    assert result.terms[1].term.terms[3] == KleeneStar(Sequence([String(", "), types.integer, String(":"), types.string]))
+    assert result.terms[2] == String("}")

From 69783b37bb90e1d2453fa9bf8eaf657a42357f96 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 6 Mar 2025 15:55:00 +0100
Subject: [PATCH 277/505] Remove the Outlines types Choice, JsonType and List

---
 README.md                                     | 11 +--
 docs/cookbook/extract_event_details.py        |  3 +-
 docs/reference/generation/generation.md       | 13 ++-
 docs/reference/generation/types.md            |  2 +-
 docs/reference/models/gemini.md               |  6 +-
 docs/reference/models/models.md               |  1 -
 docs/reference/models/openai.md               |  3 +-
 docs/reference/models/transformers.md         |  4 +-
 .../models/transformers_multimodal.md         |  2 +-
 docs/reference/models/vllm.md                 |  3 +-
 examples/cerebrium/main.py                    |  2 +-
 examples/dating_profile.py                    |  2 +-
 examples/llamacpp_example.py                  |  2 +-
 examples/modal_example.py                     |  2 +-
 outlines/__init__.py                          |  6 +-
 outlines/function.py                          |  3 +-
 outlines/models/gemini.py                     | 14 +++-
 outlines/types/__init__.py                    | 79 -------------------
 outlines/types/dsl.py                         | 12 ++-
 tests/models/test_anthropic.py                |  2 +-
 tests/models/test_dottxt.py                   |  2 +-
 tests/models/test_gemini.py                   | 32 +++-----
 tests/models/test_gemini_type_adapter.py      | 13 +++
 tests/models/test_llamacpp.py                 | 18 ++---
 tests/models/test_mlxlm.py                    |  6 +-
 tests/models/test_ollama_type_adapter.py      | 10 +--
 tests/models/test_transformers.py             | 19 +++--
 tests/models/test_transformers_multimodal.py  | 12 +--
 tests/models/test_vllm.py                     | 12 +--
 29 files changed, 114 insertions(+), 182 deletions(-)

diff --git a/README.md b/README.md
index 611224dee..0e84ecc4e 100644
--- a/README.md
+++ b/README.md
@@ -84,6 +84,7 @@ Please see [the documentation](https://dottxt-ai.github.io/outlines/latest/refer
 You can reduce the completion to a choice between multiple possibilities:
 
 ``` python
+from typing import Literal
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
 import outlines
@@ -109,7 +110,7 @@ Text: I really really really want pizza.
 <|im_start|>assistant
 """
 
-answer = model(prompt, outlines.Choice(["Pizza", "Pasta", "Salad", "Dessert"]))
+answer = model(prompt, Literal["Pizza", "Pasta", "Salad", "Dessert"])
 # Likely answer: Pizza
 ```
 
@@ -124,7 +125,7 @@ class Food(str, Enum):
     salad = "Salad"
     dessert = "Dessert"
 
-answer = model(prompt, outlines.Choice(Food))
+answer = model(prompt, Food)
 # Likely answer: Pizza
 ````
 
@@ -254,7 +255,7 @@ print(repr(character))
 # Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
 
 prompt = "Give me an interesting character description"
-character = model(prompt, outlines.JsonType(Character), seed=seed)
+character = model(prompt, Character, seed=seed)
 
 print(repr(character))
 # Character(name='Vivian Thr', age=44, armor=<Armor.plate: 'plate'>, weapon=<Weapon.crossbow: 'crossbow'>, strength=125)
@@ -315,7 +316,7 @@ model = outlines.from_transformers(
 )
 
 prompt = "Give me a character description"
-character = model(prompt, outlines.JsonType(schema))
+character = model(prompt, outlines.json_schema(schema))
 ```
 
 ### Using context-free grammars to guide generation
@@ -411,7 +412,7 @@ model = outlines.from_transformers(
     AutoModelForCausalLM.from_pretrained(model_name),
     AutoTokenizer.from_pretrained(model_name)
 )
-generator = outlines.Generator(model, outlines.types.JsonType(Operation))
+generator = outlines.Generator(model, Operation)
 result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
 #print(result)
 # {'c': -3.14, 'd': 1.5}
diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
index 8a60cf90b..a4c7f9188 100644
--- a/docs/cookbook/extract_event_details.py
+++ b/docs/cookbook/extract_event_details.py
@@ -5,7 +5,6 @@
 
 import outlines
 from outlines import Generator
-from outlines.types import JsonType
 
 
 # Load the model
@@ -38,7 +37,7 @@ class Event(BaseModel):
 see you 😘 """
 
 # Create the generator
-generator = Generator(model, JsonType(Event))
+generator = Generator(model, Event)
 
 # Extract the event information
 event = generator(prompt + message)
diff --git a/docs/reference/generation/generation.md b/docs/reference/generation/generation.md
index c2168f9e3..ad030fffd 100644
--- a/docs/reference/generation/generation.md
+++ b/docs/reference/generation/generation.md
@@ -30,15 +30,15 @@ print(result)
 # The answer is 4
 ```
 
-### [Multi-label classification](./choices.md)
+### [Multi-label classification]
 
 Outlines allows you to do multi-label classification by guiding the model so it can only output either of the specified choices:
 
 ```python
+from typing import Literal
 import outlines
-from outlines.types import Choice
 
-generator = outlines.Generator(model, Choice(["Blue", "Red", "Yellow"]))
+generator = outlines.Generator(model, Literal["Blue", "Red", "Yellow"])
 
 color = generator("What is the closest color to Indigo? ")
 print(color)
@@ -55,7 +55,6 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
     from enum import Enum
     from pydantic import BaseModel, constr, conint
 
-    from outlines.types import JsonType
     import outlines
 
     class Armor(str, Enum):
@@ -70,7 +69,7 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
         armor: Armor
         strength: conint(gt=1, lt=100)
 
-    generator = outlines.Generator(model, JsonType(Character))
+    generator = outlines.Generator(model, Character)
 
     character = generator(
         "Generate a new character for my awesome game: "
@@ -84,7 +83,7 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
 
     ```python
     import outlines
-    from outlines.types import JsonType
+    from outlines.types import JsonSchema
 
     schema = """{
         "$defs": {
@@ -105,7 +104,7 @@ Outlines can guide models so that they output valid JSON **100%** of the time. Y
         "type": "object"
     }"""
 
-    generator = outlines.Generator(model, JsonType(schema))
+    generator = outlines.Generator(model, JsonSchema(schema))
     character = generator(
         "Generate a new character for my awesome game: "
         + "name, age (between 1 and 99), armor and strength. "
diff --git a/docs/reference/generation/types.md b/docs/reference/generation/types.md
index ba712171e..037db313b 100644
--- a/docs/reference/generation/types.md
+++ b/docs/reference/generation/types.md
@@ -51,7 +51,7 @@ model = outlines.from_transformers(
     AutoModelForCausalLM.from_pretrained(model_name),
     AutoTokenizer.from_pretrained(model_name)
 )
-generator = Generator(model, types.JsonType(Client))
+generator = Generator(model, Client)
 result = generator(
     "Create a client profile with the fields name, phone_number and zip_code"
 )
diff --git a/docs/reference/models/gemini.md b/docs/reference/models/gemini.md
index ba83d59ec..6ce530fbb 100644
--- a/docs/reference/models/gemini.md
+++ b/docs/reference/models/gemini.md
@@ -50,7 +50,6 @@ import google.generativeai as genai
 
 import outlines
 from outlines import Generator
-from outlines.types import JsonType
 
 
 model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
@@ -60,7 +59,7 @@ class Person(TypedDict):
     last_name: str
     age: int
 
-generator = Generator(model, JsonType(Person))
+generator = Generator(model, Person)
 generator("current indian prime minister on january 1st 2023")
 # Person(first_name='Narendra', last_name='Modi', age=72)
 ```
@@ -79,7 +78,6 @@ from enum import Enum
 import google.generativeai as genai
 
 from outlines import Generator
-from outlines.types import Choice
 
 model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
 
@@ -88,7 +86,7 @@ class Foo(Enum):
     fizz = "Fizz"
     fuzz = "Fuzz"
 
-generator = Generator(model, Choice(Foo))
+generator = Generator(model, Foo)
 generator("current indian prime minister on january 1st 2023")
 # Person(first_name='Narendra', last_name='Modi', age=72)
 ```
diff --git a/docs/reference/models/models.md b/docs/reference/models/models.md
index f0fbc20ad..d9d581e96 100644
--- a/docs/reference/models/models.md
+++ b/docs/reference/models/models.md
@@ -43,7 +43,6 @@ Integrations differ in their coverage of structured generation. Here is a summar
 | x86 / AMD64       | ✅           | ❌  | ✅        | ❌    | N/A     | N/A   |
 | **Structure**     |               |     |           |       |         |       |
 | JSON Schema       | ✅           | ✅  | ✅        | ✅    | ✅      | ✅   |
-| Choice            | ✅           | ✅  | ✅        | ✅    | ❌      | ✅   |
 | Regex             | ✅           | ✅  | ✅        | ✅    | ❌      | ❌   |
 | Grammar           | ✅           | ✅  | ✅        | ✅    | ❌      | ❌   |
 
diff --git a/docs/reference/models/openai.md b/docs/reference/models/openai.md
index 354994861..95c03b6ac 100644
--- a/docs/reference/models/openai.md
+++ b/docs/reference/models/openai.md
@@ -89,7 +89,6 @@ from openai import OpenAI
 from pydantic import BaseModel
 
 import outlines
-from outlines.types import JsonType
 
 model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
 
@@ -98,7 +97,7 @@ class Person(BaseModel):
     last_name: str
     age: int
 
-generator = Generator(model, Json(Person))
+generator = Generator(model, Person)
 generator("current indian prime minister on january 1st 2023")
 # Person(first_name='Narendra', last_name='Modi', age=72)
 ```
diff --git a/docs/reference/models/transformers.md b/docs/reference/models/transformers.md
index c3b397297..0c5600b03 100644
--- a/docs/reference/models/transformers.md
+++ b/docs/reference/models/transformers.md
@@ -54,7 +54,7 @@ This would generate two sequences for each prompt, for a total of four sequences
 
 `Transformers` models can generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
 
-Supported types include `Json`, `Choice`, `Regex` and `CFG`.
+You can use most common Python types along with the Outlines DSL types `JsonSchema`, `Regex` and `CFG`.
 
 For instance:
 ```python
@@ -64,5 +64,5 @@ from pydantic import BaseModel
 class Character(BaseModel):
     name: str
 
-model("Create a character with a name.", Json(Character))
+model("Create a character with a name.", Character)
 ```
diff --git a/docs/reference/models/transformers_multimodal.md b/docs/reference/models/transformers_multimodal.md
index 199d38d09..85ed8b8b2 100644
--- a/docs/reference/models/transformers_multimodal.md
+++ b/docs/reference/models/transformers_multimodal.md
@@ -147,7 +147,7 @@ You can include multiple audio files by providing a list for the `audios` argume
 
 You can use the model to generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
 
-Supported types include `Json`, `Choice`, `Regex` and `CFG`.
+You can use most common Python types along with the Outlines DSL types `JsonSchema`, `Regex` and `CFG`.
 
 For instance to do classification, you can use the `Regex` type:
 ```python
diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index 7698cd31e..1b88b8d7e 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -43,14 +43,13 @@ You can also use structured generation with the `VLLM` model by providing an out
 ```python
 import outlines
 from vllm import LLM
-from outlines.types import JsonType
 from pydantic import BaseModel
 
 class Character(BaseModel):
     name: str
 
 model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
-answer = model("Create a character.", output_type=JsonType(Character))
+answer = model("Create a character.", output_type=Character)
 ```
 
 The VLLM model supports batch generation. To use it, you can pass a list of strings as prompt instead of a single string.
diff --git a/examples/cerebrium/main.py b/examples/cerebrium/main.py
index 5ea1b61f3..bce3f441c 100644
--- a/examples/cerebrium/main.py
+++ b/examples/cerebrium/main.py
@@ -40,7 +40,7 @@ def generate(
 ):
     character = model(
         f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
-        outlines.JsonType(schema),
+        outlines.json_schema(schema),
     )
 
     print(character)
diff --git a/examples/dating_profile.py b/examples/dating_profile.py
index 2f1a0e1b0..5576ca50a 100644
--- a/examples/dating_profile.py
+++ b/examples/dating_profile.py
@@ -107,7 +107,7 @@ class Example:
 
 dating_profile_prompt = Template.from_file("prompts/dating_profile.txt")
 prompt = dating_profile_prompt(description=new_description, examples=samples)
-profile = model(prompt, outlines.JsonType(DatingProfile))  # type: ignore
+profile = model(prompt, outlines.json_schema(DatingProfile))  # type: ignore
 print(profile)
 
 # Sample generated profiles
diff --git a/examples/llamacpp_example.py b/examples/llamacpp_example.py
index cf64a934a..f6afc40ed 100644
--- a/examples/llamacpp_example.py
+++ b/examples/llamacpp_example.py
@@ -34,7 +34,7 @@ class Character(BaseModel):
     model = outlines.from_llamacpp(Llama("./mistral-7b-instruct-v0.2.Q5_K_M.gguf"))
 
     # Construct structured sequence generator
-    generator = outlines.Generator(model, outlines.JsonType(Character))
+    generator = outlines.Generator(model, Character)
 
     # Draw a sample
     seed = 789005
diff --git a/examples/modal_example.py b/examples/modal_example.py
index 44b66e15c..3ec6c2c6f 100644
--- a/examples/modal_example.py
+++ b/examples/modal_example.py
@@ -75,7 +75,7 @@ def generate(
 
     character = model(
         f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]",
-        outlines.JsonType(schema),
+        outlines.json_schema(schema),
     )
 
     print(character)
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 07cb89f70..0c5ed5454 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -42,9 +42,9 @@
     "get_cache",
     "Function",
     "Generator",
-    "JsonType",
-    "Cfg",
-    "Regex",
+    "regex",
+    "json_schema",
+    "cfg",
     "prompt",
     "Template",
     "grammars",
diff --git a/outlines/function.py b/outlines/function.py
index 250d1784a..be1a8095d 100644
--- a/outlines/function.py
+++ b/outlines/function.py
@@ -5,6 +5,7 @@
 import requests
 
 import outlines
+from outlines.types import JsonSchema
 
 if TYPE_CHECKING:
     from outlines.templates import Template
@@ -43,7 +44,7 @@ def init_generator(self):
             AutoTokenizer.from_pretrained(self.model_name),
         )
 
-        self.generator = outlines.Generator(model, outlines.JsonType(self.schema))
+        self.generator = outlines.Generator(model, JsonSchema(self.schema))
 
     def __call__(self, *args, **kwargs):
         """Call the function.
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 699e392ec..1b790a119 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -76,7 +76,11 @@ def format_output_type(self, output_type):
             return self.format_json_output_type(output_type)
         elif isinstance(output_type, type(BaseModel)):
             return self.format_json_output_type(output_type)
-        if isinstance(output_type, EnumMeta):  # Enum types
+        # json schema as a dict is accepted but the title keyword is not supported ?!
+        # another restriction: the dict cannot be put in a list
+        elif isinstance(output_type, dict):
+            return self.format_json_output_type(output_type)
+        elif isinstance(output_type, EnumMeta):  # Enum types
             return self.format_enum_output_type(output_type)
         elif get_origin(output_type) is Literal:
             out = Enum("EnumFromLiteral", [(arg, arg) for arg in get_args(output_type)])
@@ -119,6 +123,12 @@ def format_list_output_type(self, output_type):
                     "response_schema": output_type,
                 }
 
+            elif isinstance(item_type, dict):
+                raise TypeError(
+                    "JSON schema dict output type is not supported with lists. "
+                    "Use a Pydantic model, typed dict or dataclass instead."
+                )
+
         raise TypeError(
             f"Gemini only supports homogenous lists: list[BaseModel], list[TypedDict] or list[dataclass]. "
             f"Got {output_type} instead."
@@ -152,7 +162,7 @@ def generate(
     def generate_stream(
         self,
         model_input: Union[str, Vision],
-        output_type: Optional[Union[JsonType, Choice, List]] = None,
+        output_type: Optional[Any] = None,
         **inference_kwargs,
     ):
         import google.generativeai as genai
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 52b28f647..1872f4193 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -64,82 +64,3 @@
 isbn = Regex(
     r"(?:ISBN(?:-1[03])?:? )?(?=[0-9X]{10}$|(?=(?:[0-9]+[- ]){3})[- 0-9X]{13}$|97[89][0-9]{10}$|(?=(?:[0-9]+[- ]){4})[- 0-9]{17}$)(?:97[89][- ]?)?[0-9]{1,5}[- ]?[0-9]+[- ]?[0-9]+[- ]?[0-9X]"
 )
-
-
-@dataclass
-class JsonType:
-    """Represents a JSON object.
-
-    The structure of JSON object can be defined using a JSON Schema
-    specification represented as a string or a dictionary, or a Pydantic
-    BaseModel.
-
-    Attributes
-    ----------
-    definition
-        The object used to define the structure of the JSON object.
-
-    """
-
-    definition: Union[str, dict, type[BaseModel]]
-    whitespace_pattern: str = " "
-
-    def to_json_schema(self):
-        if isinstance(self.definition, str):
-            schema = json.loads(self.definition)
-        elif isinstance(self.definition, dict):
-            schema = self.definition
-        elif isinstance(self.definition, type(BaseModel)):
-            schema = self.definition.model_json_schema()
-        elif isinstance(self.definition, _TypedDictMeta):
-            schema = TypeAdapter(self.definition).json_schema()
-        elif is_dataclass(self.definition):
-            schema = TypeAdapter(self.definition).json_schema()
-        else:
-            raise TypeError(
-                "The Json definition must be a JSON Schema string, dictionary or Pydantic model."
-            )
-
-        try:
-            Validator.check_schema(schema)
-        except SchemaError as e:
-            raise e
-
-        return schema
-
-    def to_regex(self):
-        schema = self.to_json_schema()
-        schema_str = json.dumps(schema)
-        return build_regex_from_schema(schema_str, self.whitespace_pattern)
-
-
-@dataclass
-class List:
-    definition: list
-
-    def to_regex(self):
-        raise NotImplementedError(
-            "Structured generation for lists of objects are not implemented yet."
-        )
-
-
-@dataclass
-class Choice:
-    """Represents a multiple choice"""
-
-    definition: Union[EnumMeta, list[str]]
-
-    def __post_init__(self):
-        if isinstance(self.definition, list):
-            self.definition = Enum("Definition", [(x, x) for x in self.definition])
-
-    def to_list(self):
-        if isinstance(self.definition, list):
-            return self.definition
-        else:
-            return [x.value for x in self.definition]
-
-    def to_regex(self):
-        choices = self.to_list()
-        regex_str = r"(" + r"|".join(choices) + r")"
-        return regex_str
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 943d4dd09..de4c654c5 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -169,16 +169,20 @@ def __repr__(self):
 
 
 class JsonSchema(Term):
-    def __init__(self, schema: Union[dict, str, type[BaseModel]]):
+    def __init__(self, schema: Union[dict, str, type[BaseModel], _TypedDictMeta, type]):
         if isinstance(schema, dict):
             schema_str = json.dumps(schema)
         elif isinstance(schema, str):
             schema_str = schema
-        elif issubclass(schema, BaseModel):
-            schema_str = json.dumps(schema.model_json_schema())
+        elif isinstance(schema, type) and issubclass(schema, BaseModel):
+            schema_str = json.dumps(schema.model_json_schema())  # type: ignore
+        elif isinstance(schema, _TypedDictMeta):
+            schema_str = json.dumps(TypeAdapter(schema).json_schema())
+        elif is_dataclass(schema):
+            schema_str = json.dumps(TypeAdapter(schema).json_schema())
         else:
             raise ValueError(
-                f"Cannot parse schema {json_schema}. The schema must be either "
+                f"Cannot parse schema {schema}. The schema must be either "
                 + "a Pydantic class, a dictionary or a string that contains the JSON "
                 + "schema specification"
             )
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 75be2e413..56ed242c6 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -29,7 +29,7 @@ class Foo:
         def __init__(self, foo):
             self.foo = foo
 
-    with pytest.raises(NotImplementedError, match="is not available"):
+    with pytest.raises(TypeError, match="is not available"):
         model = Anthropic(AnthropicClient(), MODEL_NAME)
         model.generate(Foo("prompt"))
 
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index 7d4d8eb23..ae32bfcac 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -52,7 +52,7 @@ def test_dottxt_init_from_client(api_key):
 
 @pytest.mark.api_call
 def test_dottxt_wrong_input_type(api_key):
-    with pytest.raises(TypeError, match="You must provide an output type"):
+    with pytest.raises(TypeError, match="is not available"):
         client = DottxtClient(api_key=api_key)
         model = Dottxt(client)
         model(["prompt"], User)
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 6955c9db2..0f92efd48 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -79,9 +79,8 @@ class Foo(BaseModel):
     assert "bar" in json.loads(result)
 
 
-@pytest.mark.xfail(reason="Vision models do not work with structured outputs.")
 @pytest.mark.api_call
-def test_gemini_simple_vision_pydantic(model):
+def test_gemini_simple_vision_pydantic(model, image):
     class Logo(BaseModel):
         name: int
 
@@ -90,7 +89,6 @@ class Logo(BaseModel):
     assert "name" in json.loads(result)
 
 
-@pytest.mark.xfail(reason="Gemini seems to be unable to follow nested schemas.")
 @pytest.mark.api_call
 def test_gemini_nested_pydantic(model):
     class Bar(BaseModel):
@@ -107,22 +105,6 @@ class Foo(BaseModel):
     assert "fu" in json.loads(result)["bar"]
 
 
-@pytest.mark.xfail(
-    reason="The Gemini SDK's serialization method does not support Json Schema dictionaries."
-)
-@pytest.mark.api_call
-def test_gemini_simple_json_schema_dict(model):
-    schema = {
-        "properties": {"bar": {"title": "Bar", "type": "integer"}},
-        "required": ["bar"],
-        "title": "Foo",
-        "type": "object",
-    }
-    result = model.generate("foo?", schema)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
 @pytest.mark.xfail(
     reason="The Gemini SDK's serialization method does not support Json Schema strings."
 )
@@ -144,6 +126,18 @@ class Foo(TypedDict):
     assert "bar" in json.loads(result)
 
 
+@pytest.mark.api_call
+def test_gemini_simple_json_schema_dict(model):
+    schema = {
+        "properties": {"bar": {"type": "integer"}},
+        "required": ["bar"],
+        "type": "object",
+    }
+    result = model.generate("foo?", schema)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
 @pytest.mark.api_call
 def test_gemini_simple_dataclass(model):
     @dataclass
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
index 6b3925f08..6dc8131a6 100644
--- a/tests/models/test_gemini_type_adapter.py
+++ b/tests/models/test_gemini_type_adapter.py
@@ -94,6 +94,19 @@ def test_gemini_type_adapter_output_none(adapter):
     assert result == {}
 
 
+def test_gemini_type_adapter_output_dict(adapter, schema):
+    json_schema = {
+        "properties": {"bar": {"type": "integer"}},
+        "required": ["bar"],
+        "type": "object",
+    }
+    result = adapter.format_output_type(json_schema)
+    assert result == {
+        "response_mime_type": "application/json",
+        "response_schema": json_schema,
+    }
+
+
 def test_gemini_type_adapter_output_dataclass(adapter, schema):
     @dataclass
     class User:
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 71cd3053b..fffc82331 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -7,7 +7,7 @@
 
 from outlines.models import LlamaCpp
 from outlines.processors import RegexLogitsProcessor
-from outlines.types import Choice, JsonType, Regex
+from outlines.types.dsl import Regex, python_types_to_terms, to_regex
 
 
 def test_load_model():
@@ -48,12 +48,7 @@ def test_llamacpp_json(model):
     class Foo(BaseModel):
         bar: str
 
-    regex_str = JsonType(Foo).to_regex()
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
-    result = model.generate(
-        "foo? Respond with one word.", logits_processor, max_tokens=1000
-    )
-
+    result = model("foo? Respond with one word.", Foo, max_tokens=1000)
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -63,10 +58,7 @@ class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
 
-    regex_str = Choice(Foo).to_regex()
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
-    result = model.generate("foo?", logits_processor)
-
+    result = model("foo?", Foo)
     assert result == "Foo" or result == "Bar"
 
 
@@ -101,7 +93,7 @@ def test_llamacpp_stream_json(model):
     class Foo(BaseModel):
         bar: int
 
-    generator = model.stream("foo?", JsonType(Foo))
+    generator = model.stream("foo?", Foo)
 
     x = next(generator)
     assert x == "{"
@@ -112,7 +104,7 @@ class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
 
-    generator = model.stream("foo?", Choice(Foo))
+    generator = model.stream("foo?", Foo)
 
     x = next(generator)
     assert isinstance(x, str)
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index a6b49bf92..a5ce111fa 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -4,7 +4,7 @@
 from typing import Generator
 
 import outlines
-from outlines.types import Choice, JsonType, Regex
+from outlines.types import Regex
 from outlines.models.transformers import TransformerTokenizer
 from pydantic import BaseModel
 from transformers import PreTrainedTokenizer
@@ -89,7 +89,7 @@ def test_mlxlm_JsonType(model):
     class Character(BaseModel):
         name: str
 
-    result = model("Create a character with a name.", JsonType(Character))
+    result = model("Create a character with a name.", Character)
     assert "name" in result
 
 
@@ -99,7 +99,7 @@ class Foo(Enum):
         cat = "cat"
         dog = "dog"
 
-    result = model("Cat or dog?", Choice(Foo))
+    result = model("Cat or dog?", Foo)
     assert result in ["cat", "dog"]
 
 
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
index 614266c35..783c49592 100644
--- a/tests/models/test_ollama_type_adapter.py
+++ b/tests/models/test_ollama_type_adapter.py
@@ -78,7 +78,7 @@ class User:
         name: str
 
     result = adapter.format_output_type(User)
-    assert result == json.dumps(schema)
+    assert result == schema
 
 
 def test_ollama_type_adapter_output_typed_dict(adapter, schema):
@@ -87,7 +87,7 @@ class User(TypedDict):
         name: str
 
     result = adapter.format_output_type(User)
-    assert result == json.dumps(schema)
+    assert result == schema
 
 
 def test_ollama_type_adapter_output_pydantic(adapter, schema):
@@ -96,15 +96,15 @@ class User(BaseModel):
         name: str
 
     result = adapter.format_output_type(User)
-    assert result == json.dumps(schema)
+    assert result == schema
 
 
 def test_ollama_type_adapter_json_schema_str(adapter, schema):
     schema_str = json.dumps(schema)
     result = adapter.format_output_type(json_schema(schema_str))
-    assert result == json.dumps(schema)
+    assert result == schema
 
 
 def test_ollama_type_adapter_json_schema_dict(adapter, schema):
     result = adapter.format_output_type(json_schema(schema))
-    assert result == json.dumps(schema)
+    assert result == schema
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 1c7602b78..1ffb8db21 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -7,7 +7,7 @@
 
 import outlines
 from outlines.models.transformers import Transformers
-from outlines.types import Choice, JsonType, Regex
+from outlines.types import Regex
 
 TEST_MODEL = "erwanf/gpt2-mini"
 TEST_MODEL_SEQ2SEQ = "hf-internal-testing/tiny-random-t5"
@@ -39,8 +39,11 @@ def test_transformers_instantiate_mamba():
 
 
 def test_transformers_instantiate_tokenizer_kwargs():
-    model = Transformers(
-        TEST_MODEL, tokenizer_kwargs={"additional_special_tokens": ["<t1>", "<t2>"]}
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(
+            TEST_MODEL, additional_special_tokens=["<t1>", "<t2>"]
+        ),
     )
     assert "<t1>" in model.tokenizer.special_tokens
     assert "<t2>" in model.tokenizer.special_tokens
@@ -85,7 +88,7 @@ def test_transformers_json(model):
     class Character(BaseModel):
         name: str
 
-    result = model("Create a character with a name.", JsonType(Character))
+    result = model("Create a character with a name.", Character)
     assert "name" in result
 
 
@@ -94,7 +97,7 @@ class Foo(Enum):
         cat = "cat"
         dog = "dog"
 
-    result = model("Cat or dog?", Choice(Foo))
+    result = model("Cat or dog?", Foo)
     assert result in ["cat", "dog"]
 
 
@@ -128,14 +131,14 @@ class Foo(Enum):
         cat = "cat"
         dog = "dog"
 
-    result = model("Cat or dog?", Choice(Foo), num_return_sequences=2, num_beams=2)
+    result = model("Cat or dog?", Foo, num_return_sequences=2, num_beams=2)
     assert isinstance(result, list)
     assert len(result) == 2
     assert result[0] in ["cat", "dog"]
     assert result[1] in ["cat", "dog"]
     result = model(
         ["Cat or dog?", "Cat or dog?"],
-        Choice(Foo),
+        Foo,
     )
     assert isinstance(result, list)
     assert len(result) == 2
@@ -143,7 +146,7 @@ class Foo(Enum):
     assert result[1] in ["cat", "dog"]
     result = model(
         ["Cat or dog?", "Cat or dog?"],
-        Choice(Foo),
+        Foo,
         num_return_sequences=2,
         num_beams=2,
     )
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 9292bdf6a..a701ec9bc 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -18,7 +18,7 @@
 
 import outlines
 from outlines.models.transformers import TransformersMultiModal
-from outlines.types import Choice, JsonType, Regex
+from outlines.types import Regex
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
 TEST_CLIP_MODEL = "openai/clip-vit-base-patch32"
@@ -119,7 +119,7 @@ class Foo(BaseModel):
 
     result = model(
         {"text": "<image>Give a name to this animal.", "images": images[0]},
-        JsonType(Foo),
+        Foo,
     )
     assert "name" in result
 
@@ -139,7 +139,7 @@ class Foo(Enum):
         dog = "dog"
 
     result = model(
-        {"text": "<image>Is it a cat or a dog?", "images": images[0]}, Choice(Foo)
+        {"text": "<image>Is it a cat or a dog?", "images": images[0]}, Foo
     )
 
     assert isinstance(result, str)
@@ -190,7 +190,7 @@ class Foo(Enum):
 
     result = model(
         {"text": "<image>Describe this image in one sentence.", "images": images[0]},
-        Choice(Foo),
+        Foo,
         num_return_sequences=2,
         num_beams=2,
     )
@@ -206,7 +206,7 @@ class Foo(Enum):
             ],
             "images": [images[0], images[1]],
         },
-        Choice(Foo),
+        Foo,
     )
     assert isinstance(result, list)
     assert len(result) == 2
@@ -220,7 +220,7 @@ class Foo(Enum):
             ],
             "images": [images[0], images[1]],
         },
-        Choice(Foo),
+        Foo,
         num_return_sequences=2,
         num_beams=2,
     )
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index 2a94ee4c9..eff98a037 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -12,7 +12,7 @@
 
 import outlines
 from outlines.models.vllm import VLLM
-from outlines.types import Choice, JsonType, Regex
+from outlines.types import Regex
 
 
 TEST_MODEL = "erwanf/gpt2-mini"
@@ -59,7 +59,7 @@ def test_vllm_json(model):
     class Character(BaseModel):
         name: str
 
-    result = model("Create a character with a name.", JsonType(Character))
+    result = model("Create a character with a name.", Character)
     assert "name" in result
 
 
@@ -68,7 +68,7 @@ class Foo(Enum):
         cat = "cat"
         dog = "dog"
 
-    result = model("Cat or dog?", Choice(Foo))
+    result = model("Cat or dog?", Foo)
     assert result in ["cat", "dog"]
 
 
@@ -104,7 +104,7 @@ class Foo(Enum):
 
     result = model(
         "Cat or dog?",
-        Choice(Foo),
+        Foo,
         sampling_params=SamplingParams(n=2)
     )
     assert isinstance(result, list)
@@ -114,7 +114,7 @@ class Foo(Enum):
 
     result = model(
         ["Cat or dog?", "Cat or dog?"],
-        Choice(Foo),
+        Foo,
     )
     assert isinstance(result, list)
     assert len(result) == 2
@@ -123,7 +123,7 @@ class Foo(Enum):
 
     result = model(
         ["Cat or dog?", "Cat or dog?"],
-        Choice(Foo),
+        Foo,
         sampling_params=SamplingParams(n=2)
     )
     assert isinstance(result, list)

From 8ce2967771257909db23633beb3ba80476422f20 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 11 Mar 2025 20:23:16 +0100
Subject: [PATCH 278/505] Create type utils functions

---
 outlines/generator.py                    |   4 -
 outlines/models/dottxt.py                |   8 +-
 outlines/models/gemini.py                |  53 ++--
 outlines/models/ollama.py                |   9 +-
 outlines/models/openai.py                |  19 +-
 outlines/types/dsl.py                    | 126 ++++++---
 outlines/types/utils.py                  | 150 +++++++++++
 tests/models/test_dottxt_type_adapter.py |  23 +-
 tests/models/test_gemini.py              |  23 +-
 tests/models/test_gemini_type_adapter.py |  30 +--
 tests/models/test_ollama_type_adapter.py |  20 +-
 tests/models/test_openai_type_adapter.py |  27 +-
 tests/types/test_types_utils.py          | 327 +++++++++++++++++++++++
 13 files changed, 702 insertions(+), 117 deletions(-)
 create mode 100644 outlines/types/utils.py
 create mode 100644 tests/types/test_types_utils.py

diff --git a/outlines/generator.py b/outlines/generator.py
index 2aeb853cd..088b6d113 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -63,10 +63,6 @@ class SteerableGenerator:
     output_type: Optional[Any]
 
     def __post_init__(self):
-        # If CFG -> CFG
-        # if dict -> CFG
-        # if Any -> None
-        # Else -> Regex
         if self.output_type is None:
             self.logits_processor = None
         else:
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index db3385a73..61fc9b19a 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -1,6 +1,5 @@
 """Integration with Dottxt's API."""
 
-from dataclasses import is_dataclass
 import json
 from functools import singledispatchmethod
 from typing import Optional, TYPE_CHECKING
@@ -10,6 +9,7 @@
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.types import Regex, CFG, JsonSchema
+from outlines.types.utils import is_dataclass, is_typed_dict, is_pydantic_model, is_genson_schema_builder
 
 
 if TYPE_CHECKING:
@@ -59,12 +59,14 @@ def format_output_type(self, output_type):
         elif is_dataclass(output_type):
             schema = TypeAdapter(output_type).json_schema()
             return json.dumps(schema)
-        elif isinstance(output_type, _TypedDictMeta):
+        elif is_typed_dict(output_type):
             schema = TypeAdapter(output_type).json_schema()
             return json.dumps(schema)
-        elif isinstance(output_type, type(BaseModel)):
+        elif is_pydantic_model(output_type):
             schema = output_type.model_json_schema()
             return json.dumps(schema)
+        elif is_genson_schema_builder(output_type):
+            return output_type.to_json()
         else:
             type_name = getattr(output_type, "__name__", output_type)
             raise TypeError(
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 1b790a119..0201e6484 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -1,6 +1,5 @@
 """Integration with Gemini's API."""
 
-from dataclasses import is_dataclass
 from enum import EnumMeta, Enum
 from typing import (
     get_args,
@@ -11,12 +10,23 @@
     TYPE_CHECKING,
     Literal,
 )
-from typing_extensions import is_typeddict
+
 from pydantic import BaseModel, TypeAdapter
+from typing_extensions import is_typeddict
+
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
 from outlines.types import Regex, CFG, JsonSchema
-
+from outlines.types.utils import (
+    is_dataclass,
+    is_typed_dict,
+    is_pydantic_model,
+    is_genson_schema_builder,
+    is_enum,
+    is_literal,
+    is_typing_list,
+    literal_to_enum
+)
 
 if TYPE_CHECKING:
     from google.generativeai import GenerativeModel as GeminiClient
@@ -63,6 +73,10 @@ def format_output_type(self, output_type):
             raise TypeError(
                 "CFG-based structured outputs are not available with Gemini. Use an open source model or dottxt instead."
             )
+        elif is_genson_schema_builder(output_type):
+            raise TypeError(
+                "The Gemini SDK does not accept Genson schema builders as an input. Pass a Pydantic model, typed dict or dataclass instead."
+            )
         elif isinstance(output_type, JsonSchema):
             raise TypeError(
                 "The Gemini SDK does not accept Json Schemas as an input. Pass a Pydantic model, typed dict or dataclass instead."
@@ -70,22 +84,21 @@ def format_output_type(self, output_type):
 
         if output_type is None:
             return {}
-        elif is_dataclass(output_type):  # structured types
-            return self.format_json_output_type(output_type)
-        elif is_typeddict(output_type):
+        # structured types
+        elif is_dataclass(output_type):
             return self.format_json_output_type(output_type)
-        elif isinstance(output_type, type(BaseModel)):
+        elif is_typed_dict(output_type):
             return self.format_json_output_type(output_type)
-        # json schema as a dict is accepted but the title keyword is not supported ?!
-        # another restriction: the dict cannot be put in a list
-        elif isinstance(output_type, dict):
+        elif is_pydantic_model(output_type):
             return self.format_json_output_type(output_type)
-        elif isinstance(output_type, EnumMeta):  # Enum types
+        # enum types
+        elif is_enum(output_type):
             return self.format_enum_output_type(output_type)
-        elif get_origin(output_type) is Literal:
-            out = Enum("EnumFromLiteral", [(arg, arg) for arg in get_args(output_type)])
-            return self.format_enum_output_type(out)
-        elif get_origin(output_type) is list:  # List of objects
+        elif is_literal(output_type):
+            enum = literal_to_enum(output_type)
+            return self.format_enum_output_type(enum)
+        # list of objects
+        elif is_typing_list(output_type):
             return self.format_list_output_type(output_type)
         else:
             type_name = getattr(output_type, "__name__", output_type)
@@ -114,8 +127,8 @@ def format_list_output_type(self, output_type):
 
             # Check if list item type is supported
             if (
-                isinstance(item_type, type(BaseModel))
-                or is_typeddict(item_type)
+                is_pydantic_model(item_type)
+                or is_typed_dict(item_type)
                 or is_dataclass(item_type)
             ):
                 return {
@@ -123,10 +136,10 @@ def format_list_output_type(self, output_type):
                     "response_schema": output_type,
                 }
 
-            elif isinstance(item_type, dict):
+            else:
                 raise TypeError(
-                    "JSON schema dict output type is not supported with lists. "
-                    "Use a Pydantic model, typed dict or dataclass instead."
+                    "The only supported types for list items are Pydantic "
+                    + "models, typed dicts and dataclasses."
                 )
 
         raise TypeError(
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index d5a348c16..04526f3aa 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,4 +1,3 @@
-from dataclasses import is_dataclass
 import json
 from typing import Iterator, TYPE_CHECKING
 
@@ -7,6 +6,8 @@
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.types import Regex, CFG, JsonSchema
+from outlines.types.utils import is_dataclass, is_typed_dict, is_pydantic_model, is_genson_schema_builder
+
 
 if TYPE_CHECKING:
     from ollama import Client as OllamaClient
@@ -56,12 +57,14 @@ def format_output_type(self, output_type):
         elif is_dataclass(output_type):
             schema = TypeAdapter(output_type).json_schema()
             return schema
-        elif isinstance(output_type, _TypedDictMeta):
+        elif is_typed_dict(output_type):
             schema = TypeAdapter(output_type).json_schema()
             return schema
-        elif isinstance(output_type, type(BaseModel)):
+        elif is_pydantic_model(output_type):
             schema = output_type.model_json_schema()
             return schema
+        elif is_genson_schema_builder(output_type):
+            return output_type.to_json()
         else:
             type_name = getattr(output_type, "__name__", output_type)
             raise TypeError(
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 8430c8f37..6c0f200e0 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,15 +1,21 @@
 """Integration with OpenAI's API."""
 
-from dataclasses import is_dataclass
 import json
 from typing import Optional, Union, TYPE_CHECKING
-from typing_extensions import is_typeddict
 
 from pydantic import BaseModel, TypeAdapter
+from typing_extensions import is_typeddict
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
 from outlines.types import JsonSchema, Regex, CFG
+from outlines.types.utils import (
+    is_dataclass,
+    is_typed_dict,
+    is_pydantic_model,
+    is_genson_schema_builder,
+    is_native_dict
+)
 
 if TYPE_CHECKING:
     from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
@@ -99,17 +105,20 @@ def format_output_type(self, output_type):
 
         if output_type is None:
             return {}
-        elif output_type is dict:
+        elif is_native_dict(output_type):
             return self.format_json_mode_type()
         elif is_dataclass(output_type):
             output_type = TypeAdapter(output_type).json_schema()
             return self.format_json_output_type(output_type)
-        elif is_typeddict(output_type):
+        elif is_typed_dict(output_type):
             output_type = TypeAdapter(output_type).json_schema()
             return self.format_json_output_type(output_type)
-        elif isinstance(output_type, type(BaseModel)):
+        elif is_pydantic_model(output_type):
             output_type = output_type.model_json_schema()
             return self.format_json_output_type(output_type)
+        elif is_genson_schema_builder(output_type):
+            schema = json.loads(output_type.to_json())
+            return self.format_json_output_type(schema)
         elif isinstance(output_type, JsonSchema):
             return self.format_json_output_type(json.loads(output_type.schema))
         else:
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index de4c654c5..20e297cd5 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,7 +1,7 @@
 import datetime
 import json
 import re
-from dataclasses import dataclass, is_dataclass
+from dataclasses import dataclass
 from enum import EnumMeta
 from types import FunctionType
 from typing import Any, Callable, List, Literal, Union, get_args, get_origin, _TypedDictMeta # type: ignore
@@ -16,6 +16,32 @@
 import outlines.types as types
 from outlines import grammars
 from outlines.fsm.json_schema import get_schema_from_signature
+from outlines.types.utils import (
+    is_int,
+    is_int_instance,
+    is_float,
+    is_float_instance,
+    is_str,
+    is_str_instance,
+    is_bool,
+    is_datetime,
+    is_date,
+    is_time,
+    is_native_dict,
+    is_dict_instance,
+    is_dataclass,
+    is_typed_dict,
+    is_pydantic_model,
+    is_genson_schema_builder,
+    is_literal,
+    is_union,
+    is_enum,
+    is_callable,
+    is_typing_list,
+    is_typing_tuple,
+    is_typing_dict,
+    is_interegular_fsm,
+)
 from outlines_core.fsm.json_schema import build_regex_from_schema
 
 
@@ -41,18 +67,30 @@ class also handles validation.
 
     """
 
-    def __add__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
-        if isinstance(other, str):
-            other = String(other)
+    def __add__(self: "Term", other: "Term") -> "Sequence":
+        if is_str_instance(other):
+            other = String(str(other))
 
         return Sequence([self, other])
 
-    def __radd__(self: "Term", other: Union[str, "Term"]) -> "Sequence":
-        if isinstance(other, str):
-            other = String(other)
+    def __radd__(self: "Term", other: "Term") -> "Sequence":
+        if is_str_instance(other):
+            other = String(str(other))
 
         return Sequence([other, self])
 
+    def __or__(self: "Term", other: "Term") -> "Alternatives":
+        if is_str_instance(other):
+            other = String(str(other))
+
+        return Alternatives([self, other])
+
+    def __ror__(self: "Term", other: "Term") -> "Alternatives":
+        if is_str_instance(other):
+            other = String(str(other))
+
+        return Alternatives([other, self])
+
     def __get_validator__(self, _core_schema):
         def validate(input_value):
             return self.validate(input_value)
@@ -169,22 +207,27 @@ def __repr__(self):
 
 
 class JsonSchema(Term):
-    def __init__(self, schema: Union[dict, str, type[BaseModel], _TypedDictMeta, type]):
-        if isinstance(schema, dict):
+    def __init__(self, schema: Union[dict, str, type[BaseModel], _TypedDictMeta, type, SchemaBuilder]):
+        schema_str: str
+
+        if is_dict_instance(schema):
             schema_str = json.dumps(schema)
-        elif isinstance(schema, str):
-            schema_str = schema
-        elif isinstance(schema, type) and issubclass(schema, BaseModel):
+        elif is_str_instance(schema):
+            schema_str = str(schema)
+        elif is_pydantic_model(schema):
             schema_str = json.dumps(schema.model_json_schema())  # type: ignore
-        elif isinstance(schema, _TypedDictMeta):
+        elif is_typed_dict(schema):
             schema_str = json.dumps(TypeAdapter(schema).json_schema())
         elif is_dataclass(schema):
             schema_str = json.dumps(TypeAdapter(schema).json_schema())
+        elif is_genson_schema_builder(schema):
+            schema_str = schema.to_json()  # type: ignore
         else:
             raise ValueError(
                 f"Cannot parse schema {schema}. The schema must be either "
-                + "a Pydantic class, a dictionary or a string that contains the JSON "
-                + "schema specification"
+                + "a Pydantic class, typed dict, a dataclass, a genSON schema "
+                + "builder or a string or dict that contains the JSON schema "
+                + "specification"
             )
 
         self.schema = schema_str
@@ -441,46 +484,44 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
         return ptype
 
     # Basic types
-    if ptype is int or get_origin(ptype) is int:
+    if is_int(ptype):
         return types.integer
-    elif ptype is float or get_origin(ptype) is float:
+    elif is_float(ptype):
         return types.number
-    elif ptype is bool or get_origin(ptype) is bool:
+    elif is_bool(ptype):
         return types.boolean
-    elif ptype is str or get_origin(ptype) is str:
+    elif is_str(ptype):
         return types.string
-    elif ptype is dict:
+    elif is_native_dict(ptype):
         return CFG(grammars.json)
-    elif ptype is datetime.time:
+    elif is_time(ptype):
         return types.time
-    elif ptype is datetime.date:
+    elif is_date(ptype):
         return types.date
-    elif ptype is datetime.datetime:
+    elif is_datetime(ptype):
         return types.datetime
 
     # Basic type instances
-    if isinstance(ptype, str):
+    if is_str_instance(ptype):
         return String(ptype)
-    elif isinstance(ptype, (int, float)):
+    elif is_int_instance(ptype) or is_float_instance(ptype):
         return Regex(str(ptype))
 
     # Structured types
     structured_type_checks = [
         lambda x: is_dataclass(x),
-        lambda x: isinstance(x, _TypedDictMeta),
-        lambda x: isinstance(x, type) and issubclass(x, BaseModel),
+        lambda x: is_typed_dict(x),
+        lambda x: is_pydantic_model(x),
     ]
     if any(check(ptype) for check in structured_type_checks):
         schema = TypeAdapter(ptype).json_schema()
         return JsonSchema(schema)
 
-    # genSON SchemaBuilder
-    elif isinstance(ptype, SchemaBuilder):
+    elif is_genson_schema_builder(ptype):
         schema = ptype.to_json()
         return JsonSchema(schema)
 
-    # Enums
-    if isinstance(ptype, EnumMeta):
+    if is_enum(ptype):
         return Alternatives(
             [
                 python_types_to_terms(member, recursion_depth + 1)
@@ -488,27 +529,25 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
             ]
         )
 
-    # Generic types
-    origin = get_origin(ptype)
     args = get_args(ptype)
-
-    if origin is Literal:
+    if is_literal(ptype):
         return _handle_literal(args)
-    elif origin is Union:
+    elif is_union(ptype):
         return _handle_union(args, recursion_depth)
-    elif origin is list:
+    elif is_typing_list(ptype):
         return _handle_list(args, recursion_depth)
-    elif origin is tuple:
+    elif is_typing_tuple(ptype):
         return _handle_tuple(args, recursion_depth)
-    elif origin is dict:
+    elif is_typing_dict(ptype):
         return _handle_dict(args, recursion_depth)
 
-    # Callables
-    if callable(ptype) and not isinstance(ptype, type):
+    if is_callable(ptype):
         return JsonSchema(get_schema_from_signature(ptype))
 
-    # Interegular FSM are passed on as is and are handled by the Generator
-    elif isinstance(ptype, interegular.fsm.FSM):
+    if isinstance(ptype, interegular.fsm.FSM):
+        return ptype
+
+    if is_interegular_fsm(ptype):
         return ptype
 
     type_name = getattr(ptype, "__name__", ptype)
@@ -531,7 +570,6 @@ def _get_enum_members(ptype: EnumMeta) -> List[Any]:
 
 
 def _handle_literal(args: tuple) -> Alternatives:
-    print(args)
     return Alternatives([python_types_to_terms(arg) for arg in args])
 
 
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
new file mode 100644
index 000000000..21d36281a
--- /dev/null
+++ b/outlines/types/utils.py
@@ -0,0 +1,150 @@
+import dataclasses
+import datetime
+from enum import Enum, EnumMeta
+from typing import ( # type: ignore
+    Annotated,
+    Any,
+    Dict,
+    Literal,
+    List,
+    NewType,
+    Tuple,
+    Union,
+    get_args,
+    get_origin,
+    _TypedDictMeta,
+)
+
+import interegular
+from genson import SchemaBuilder
+from pydantic import BaseModel
+from typing_extensions import _TypedDictMeta as Ext_TypedDictMeta  # type: ignore
+
+
+# Type identification
+
+
+def is_int(value: Any) -> bool:
+    return (
+        value is int
+        or get_origin(value) is int
+        or (get_origin(value) is Annotated and get_args(value)[0] is int)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is int)
+    )
+
+
+def is_int_instance(value: Any) -> bool:
+    return isinstance(value, int)
+
+
+def is_float(value: Any) -> bool:
+    return (
+        value is float
+        or get_origin(value) is float
+        or (get_origin(value) is Annotated and get_args(value)[0] is float)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is float)
+    )
+
+
+def is_float_instance(value: Any) -> bool:
+    return isinstance(value, float)
+
+
+def is_str(value: Any) -> bool:
+    return (
+        value is str
+        or get_origin(value) is str
+        or (get_origin(value) is Annotated and get_args(value)[0] is str)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is str)
+    )
+
+
+def is_str_instance(value: Any) -> bool:
+    return isinstance(value, str)
+
+
+def is_bool(value: Any) -> bool:
+    return (
+        value is bool
+        or get_origin(value) is bool
+        or (get_origin(value) is Annotated and get_args(value)[0] is bool)
+        or (hasattr(value, "__supertype__") and value.__supertype__ is bool)
+    )
+
+
+def is_dict_instance(value: Any) -> bool:
+    return isinstance(value, dict)
+
+
+def is_datetime(value: Any) -> bool:
+    return value is datetime.datetime or get_origin(value) is datetime.datetime
+
+
+def is_date(value: Any) -> bool:
+    return value is datetime.date or get_origin(value) is datetime.date
+
+
+def is_time(value: Any) -> bool:
+    return value is datetime.time or get_origin(value) is datetime.time
+
+
+def is_native_dict(value: Any) -> bool:
+    return value is dict
+
+
+def is_typing_dict(value: Any) -> bool:
+    return get_origin(value) is dict
+
+
+def is_typing_list(value: Any) -> bool:
+    return get_origin(value) is list
+
+
+def is_typing_tuple(value: Any) -> bool:
+    return get_origin(value) is tuple
+
+
+def is_union(value: Any) -> bool:
+    return get_origin(value) is Union
+
+
+def is_literal(value: Any) -> bool:
+    return get_origin(value) is Literal
+
+
+def is_dataclass(value: Any) -> bool:
+    return isinstance(value, type) and dataclasses.is_dataclass(value)
+
+
+def is_typed_dict(value: Any) -> bool:
+    return (
+        isinstance(value, _TypedDictMeta)
+        or isinstance(value, Ext_TypedDictMeta)
+    )
+
+
+def is_pydantic_model(value: Any) -> bool:
+    return isinstance(value, type) and issubclass(value, BaseModel)
+
+
+def is_genson_schema_builder(value: Any) -> bool:
+    return isinstance(value, SchemaBuilder)
+
+
+def is_enum(value: Any) -> bool:
+    return isinstance(value, EnumMeta)
+
+
+def is_callable(value: Any) -> bool:
+    return callable(value) and not isinstance(value, type)
+
+
+def is_interegular_fsm(value: Any) -> bool:
+    return isinstance(value, interegular.fsm.FSM)
+
+
+# Type conversion
+
+
+def literal_to_enum(value) -> Enum:
+    return Enum(value.__name__, [(arg, arg) for arg in get_args(value)])
diff --git a/tests/models/test_dottxt_type_adapter.py b/tests/models/test_dottxt_type_adapter.py
index ea8269adc..c1e84c3ad 100644
--- a/tests/models/test_dottxt_type_adapter.py
+++ b/tests/models/test_dottxt_type_adapter.py
@@ -1,15 +1,16 @@
 import io
-from dataclasses import dataclass
 import json
 import pytest
+from dataclasses import dataclass
 from typing_extensions import TypedDict
 
 from PIL import Image
+from genson import SchemaBuilder
 from pydantic import BaseModel
 
 from outlines.models.dottxt import DottxtTypeAdapter
-from outlines.types import regex, cfg, json_schema
 from outlines.templates import Vision
+from outlines.types import cfg, json_schema, regex
 
 
 @pytest.fixture
@@ -102,6 +103,24 @@ class User(BaseModel):
     assert result == json.dumps(schema)
 
 
+def test_dottxt_type_adapter_output_genson_schema_builder(adapter, schema):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    result_dict = json.loads(result)
+    assert isinstance(result_dict, dict)
+    expected_schema = {
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"],
+    }
+    assert result_dict == expected_schema
+
+
 def test_dottxt_type_adapter_json_schema_str(adapter, schema):
     schema_str = json.dumps(schema)
     result = adapter.format_output_type(json_schema(schema_str))
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 0f92efd48..2697fe812 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -116,16 +116,9 @@ def test_gemini_simple_json_schema_string(model):
     assert "bar" in json.loads(result)
 
 
-@pytest.mark.api_call
-def test_gemini_simple_typed_dict(model):
-    class Foo(TypedDict):
-        bar: int
-
-    result = model.generate("foo?", Foo)
-    assert isinstance(result, str)
-    assert "bar" in json.loads(result)
-
-
+@pytest.mark.xfail(
+    reason="The Gemini SDK's serialization method does not support Json Schema dictionaries."
+)
 @pytest.mark.api_call
 def test_gemini_simple_json_schema_dict(model):
     schema = {
@@ -138,6 +131,16 @@ def test_gemini_simple_json_schema_dict(model):
     assert "bar" in json.loads(result)
 
 
+@pytest.mark.api_call
+def test_gemini_simple_typed_dict(model):
+    class Foo(TypedDict):
+        bar: int
+
+    result = model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
 @pytest.mark.api_call
 def test_gemini_simple_dataclass(model):
     @dataclass
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
index 6dc8131a6..da9524850 100644
--- a/tests/models/test_gemini_type_adapter.py
+++ b/tests/models/test_gemini_type_adapter.py
@@ -1,16 +1,16 @@
-from dataclasses import dataclass
-from enum import EnumMeta
+import io
 import json
+import pytest
+from dataclasses import dataclass
+from enum import Enum, EnumMeta
 from typing import Literal, get_args
-from typing_extensions import TypedDict, is_typeddict
-from enum import Enum
-import io
 
 from PIL import Image
+from genson import SchemaBuilder
 from pydantic import BaseModel
-import pytest
+from typing_extensions import TypedDict, is_typeddict
 
-from outlines import regex, cfg, json_schema
+from outlines import cfg, json_schema, regex
 from outlines.models.gemini import GeminiTypeAdapter
 from outlines.templates import Vision
 
@@ -85,6 +85,9 @@ def test_gemini_type_adapter_output_invalid(adapter):
     with pytest.raises(TypeError, match="CFG-based structured outputs"):
         adapter.format_output_type(cfg(""))
 
+    with pytest.raises(TypeError, match="The Gemini SDK does not accept"):
+        adapter.format_output_type(SchemaBuilder())
+
     with pytest.raises(TypeError, match="The Gemini SDK does not"):
         adapter.format_output_type(json_schema(""))
 
@@ -94,19 +97,6 @@ def test_gemini_type_adapter_output_none(adapter):
     assert result == {}
 
 
-def test_gemini_type_adapter_output_dict(adapter, schema):
-    json_schema = {
-        "properties": {"bar": {"type": "integer"}},
-        "required": ["bar"],
-        "type": "object",
-    }
-    result = adapter.format_output_type(json_schema)
-    assert result == {
-        "response_mime_type": "application/json",
-        "response_schema": json_schema,
-    }
-
-
 def test_gemini_type_adapter_output_dataclass(adapter, schema):
     @dataclass
     class User:
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
index 783c49592..d019f014e 100644
--- a/tests/models/test_ollama_type_adapter.py
+++ b/tests/models/test_ollama_type_adapter.py
@@ -1,15 +1,16 @@
 import io
-from dataclasses import dataclass
 import json
 import pytest
+from dataclasses import dataclass
 from typing_extensions import TypedDict
 
+from genson import SchemaBuilder
 from PIL import Image
 from pydantic import BaseModel
 
 from outlines.models.ollama import OllamaTypeAdapter
-from outlines.types import regex, cfg, json_schema
 from outlines.templates import Vision
+from outlines.types import cfg, json_schema, regex
 
 
 @pytest.fixture
@@ -99,6 +100,21 @@ class User(BaseModel):
     assert result == schema
 
 
+def test_ollama_type_adapter_output_genson_schema_builder(adapter, schema):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    assert result == json.dumps({
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"]
+    })
+
+
 def test_ollama_type_adapter_json_schema_str(adapter, schema):
     schema_str = json.dumps(schema)
     result = adapter.format_output_type(json_schema(schema_str))
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
index c3498334c..0e14cae08 100644
--- a/tests/models/test_openai_type_adapter.py
+++ b/tests/models/test_openai_type_adapter.py
@@ -1,14 +1,14 @@
 import io
 import json
-from typing import Literal
-from typing_extensions import TypedDict
+import pytest
 from dataclasses import dataclass
+from typing import Literal, TypedDict
 
+from genson import SchemaBuilder
 from PIL import Image
 from pydantic import BaseModel
-import pytest
 
-from outlines import json_schema, regex, cfg, Vision
+from outlines import cfg, json_schema, regex, Vision
 from outlines.models.openai import OpenAITypeAdapter
 
 
@@ -155,6 +155,25 @@ class User(BaseModel):
     assert result["response_format"]["json_schema"]["schema"] == schema
 
 
+def test_openai_type_adapter_genson_schema_builder(adapter, schema):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    assert isinstance(result, dict)
+    assert result["response_format"]["json_schema"]["strict"] is True
+    expected_schema = {
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"],
+        "additionalProperties": False  # OpenAI adds this
+    }
+    assert result["response_format"]["json_schema"]["schema"] == expected_schema
+
+
 def test_openai_type_adapter_json_schema_str(adapter, schema):
     schema_str = json.dumps(schema)
     result = adapter.format_output_type(json_schema(schema_str))
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
new file mode 100644
index 000000000..e724551bc
--- /dev/null
+++ b/tests/types/test_types_utils.py
@@ -0,0 +1,327 @@
+import datetime
+import pytest
+from dataclasses import dataclass
+from enum import Enum
+from typing import (
+    Annotated,
+    Any,
+    Dict,
+    List,
+    Literal,
+    NewType,
+    Optional,
+    Tuple,
+    TypedDict,
+    Union
+)
+
+import interegular
+from genson import SchemaBuilder
+from pydantic import BaseModel
+
+from outlines.types.utils import (
+    is_bool,
+    is_callable,
+    is_date,
+    is_dataclass,
+    is_datetime,
+    is_enum,
+    is_float,
+    is_float_instance,
+    is_genson_schema_builder,
+    is_int,
+    is_int_instance,
+    is_interegular_fsm,
+    is_literal,
+    is_native_dict,
+    is_pydantic_model,
+    is_str,
+    is_str_instance,
+    is_time,
+    is_typed_dict,
+    is_typing_dict,
+    is_typing_list,
+    is_typing_tuple,
+    is_union
+)
+
+
+@pytest.fixture
+def sample_enum():
+    class SampleEnum(Enum):
+        A = 1
+        B = 2
+
+    return SampleEnum
+
+@pytest.fixture
+def sample_class():
+    class SampleClass:
+        pass
+
+    return SampleClass
+
+@pytest.fixture
+def sample_dataclass():
+    @dataclass
+    class SampleDataclass:
+        field1: str
+        field2: int
+
+    return SampleDataclass
+
+@pytest.fixture
+def sample_typed_dict():
+    class SampleTypedDict(TypedDict):
+        name: str
+        age: int
+
+    return SampleTypedDict
+
+@pytest.fixture
+def sample_pydantic_model():
+    class SamplePydanticModel(BaseModel):
+        name: str
+        age: int
+
+    return SamplePydanticModel
+
+@pytest.fixture
+def sample_schema_builder():
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+    return builder
+
+@pytest.fixture
+def sample_function():
+    def sample_function():
+        pass
+
+    return sample_function
+
+@pytest.fixture
+def sample_interegular_fsm():
+    pattern = interegular.parse_pattern(r"[a-z]{3}-[0-9]{2}")
+    return pattern.to_fsm()
+
+
+def test_is_int():
+    assert is_int(int)
+    assert not is_int(float)
+    assert not is_int(1)
+    assert not is_int(List[int])
+    assert not is_int(Dict[int, int])
+    assert is_int(Annotated[int, "some metadata"])
+    assert not is_int(Annotated[str, "some metadata"])
+    assert is_int(NewType("UserId", int))
+    assert not is_int(NewType("UserId", str))
+
+
+def test_is_int_instance():
+    assert is_int_instance(1)
+    assert not is_int_instance(1.0)
+    assert not is_int_instance("1")
+    assert not is_int_instance(int)
+
+
+def test_is_float():
+    assert is_float(float)
+    assert not is_float(int)
+    assert not is_float(1.0)
+    assert not is_float(List[float])
+    assert not is_float(Dict[float, float])
+    assert is_float(Annotated[float, "some metadata"])
+    assert not is_float(Annotated[int, "some metadata"])
+    assert is_float(NewType("UserId", float))
+    assert not is_float(NewType("UserId", int))
+
+
+def test_is_float_instance():
+    assert is_float_instance(1.0)
+    assert not is_float_instance(1)
+    assert not is_float_instance("1.0")
+    assert not is_float_instance(float)
+
+
+def test_is_str():
+    assert is_str(str)
+    assert not is_str(int)
+    assert not is_str("hello")
+    assert not is_str(List[str])
+    assert not is_str(Dict[str, str])
+    assert is_str(Annotated[str, "some metadata"])
+    assert not is_str(Annotated[int, "some metadata"])
+    assert is_str(NewType("UserId", str))
+    assert not is_str(NewType("UserId", int))
+
+
+def test_is_str_instance():
+    assert is_str_instance("hello")
+    assert is_str_instance("")
+    assert is_str_instance("123")
+    assert not is_str_instance(123)
+    assert not is_str_instance(str)
+
+
+def test_is_bool():
+    assert is_bool(bool)
+    assert not is_bool(int)
+    assert not is_bool(True)
+    assert is_bool(Annotated[bool, "some metadata"])
+    assert not is_bool(Annotated[int, "some metadata"])
+    assert is_bool(NewType("UserId", bool))
+    assert not is_bool(NewType("UserId", int))
+
+
+def test_is_datetime():
+    assert is_datetime(datetime.datetime)
+    assert not is_datetime(datetime.date)
+    assert not is_datetime(datetime.time)
+    assert not is_datetime(datetime.datetime.now())
+
+
+def test_is_date():
+    assert is_date(datetime.date)
+    assert not is_date(datetime.datetime)
+    assert not is_date(datetime.time)
+    assert not is_date(datetime.date.today())
+
+
+def test_is_time():
+    assert is_time(datetime.time)
+    assert not is_time(datetime.datetime)
+    assert not is_time(datetime.date)
+    assert not is_time(datetime.time(12, 30))
+
+
+def test_is_native_dict():
+    assert is_native_dict(dict)
+    assert not is_native_dict({})
+    assert not is_native_dict({"key": "value"})
+    assert not is_native_dict(list)
+    assert not is_native_dict(dict[str, int])
+
+
+def test_is_typing_dict():
+    assert is_typing_dict(dict[str, int])
+    assert is_typing_dict(Dict[int, str])
+    assert not is_typing_dict(dict)
+    assert not is_typing_dict({})
+
+
+def test_is_typing_list():
+    assert is_typing_list(list[int])
+    assert is_typing_list(List[int])
+    assert not is_typing_list(list)
+    assert not is_typing_list([])
+    assert not is_typing_list(dict)
+
+
+def test_is_typing_tuple():
+    assert is_typing_tuple(tuple[int, str])
+    assert is_typing_tuple(Tuple[int, str])
+    assert not is_typing_tuple(tuple)
+    assert not is_typing_tuple(())
+    assert not is_typing_tuple(list)
+
+
+def test_is_union():
+    assert is_union(Union[int, str])
+    assert is_union(Optional[int])
+    assert not is_union(list)
+    assert not is_union(["a", "b"])
+    assert not is_union(Literal[int, str])
+
+
+def test_is_literal():
+    assert is_literal(Literal["a", "b"])
+    assert not is_literal(str)
+    assert not is_literal("a")
+    assert not is_literal(["a", "b"])
+    assert not is_literal(Union[str, int])
+
+
+def test_is_dataclass(
+    sample_dataclass,
+    sample_class,
+    sample_typed_dict,
+    sample_pydantic_model
+):
+    assert is_dataclass(sample_dataclass)
+    assert not is_dataclass(sample_dataclass(field1="test", field2=123))
+    assert not is_dataclass(dict)
+    assert not is_dataclass(sample_class)
+    assert not is_dataclass(sample_typed_dict)
+    assert not is_dataclass(sample_pydantic_model)
+
+
+def test_is_typed_dict(
+    sample_typed_dict,
+    sample_class,
+    sample_dataclass,
+    sample_pydantic_model
+):
+    assert is_typed_dict(sample_typed_dict)
+    assert not is_typed_dict(sample_typed_dict(name="test", age=30))
+    assert not is_typed_dict(dict)
+    assert not is_typed_dict(sample_class)
+    assert not is_typed_dict(sample_dataclass)
+    assert not is_typed_dict(sample_pydantic_model)
+
+
+def test_is_pydantic_model(
+    sample_pydantic_model,
+    sample_class,
+    sample_dataclass,
+    sample_typed_dict
+):
+    assert is_pydantic_model(sample_pydantic_model)
+    assert not is_pydantic_model(sample_pydantic_model(name="test", age=30))  # Instance
+    assert not is_pydantic_model(dict)
+    assert not is_pydantic_model(sample_class)
+    assert not is_pydantic_model(sample_dataclass)
+    assert not is_pydantic_model(sample_typed_dict)
+
+
+def test_is_genson_schema_builder(
+    sample_schema_builder,
+    sample_class,
+    sample_dataclass,
+    sample_typed_dict,
+    sample_pydantic_model
+):
+    assert is_genson_schema_builder(sample_schema_builder)
+    assert not is_genson_schema_builder(dict)
+    assert not is_genson_schema_builder(str)
+    assert not is_genson_schema_builder({"type": 'object', "properties": {}})
+    assert not is_genson_schema_builder('{"type": "object", "properties": {}}')
+    assert not is_genson_schema_builder(sample_class)
+    assert not is_genson_schema_builder(sample_dataclass)
+    assert not is_genson_schema_builder(sample_typed_dict)
+    assert not is_genson_schema_builder(sample_pydantic_model)
+
+
+def test_is_enum(sample_enum):
+    assert is_enum(sample_enum)
+    assert not is_enum(sample_enum.A)
+    assert not is_enum(dict)
+    assert not is_enum(Literal["a", "b"])
+    assert not is_enum(["a", "b"])
+
+
+def test_is_callable(sample_function, sample_class, sample_dataclass, sample_typed_dict, sample_pydantic_model):
+    assert is_callable(sample_function)
+    assert is_callable(lambda x: x)
+    assert not is_callable(dict)
+    assert not is_callable(sample_class)
+    assert not is_callable(sample_dataclass)
+    assert not is_callable(sample_typed_dict)
+    assert not is_callable(sample_pydantic_model)
+
+
+def test_is_interegular_fsm(sample_interegular_fsm):
+    assert is_interegular_fsm(sample_interegular_fsm)
+    assert not is_interegular_fsm({})
+    assert not is_interegular_fsm("")

From f76adf457c763b21e5fbcdda79b23a3da0829cff Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 11 Mar 2025 22:48:23 +0100
Subject: [PATCH 279/505] Fix python version incompatibility issues

---
 outlines/models/anthropic.py             |  8 +++++---
 outlines/types/dsl.py                    | 19 +++++++++++++------
 outlines/types/utils.py                  | 21 +++++++++++++--------
 tests/models/test_dottxt_type_adapter.py |  7 ++++++-
 tests/models/test_gemini.py              |  7 ++++++-
 tests/models/test_gemini_type_adapter.py |  7 ++++++-
 tests/models/test_ollama_type_adapter.py |  7 ++++++-
 tests/models/test_openai.py              |  2 +-
 tests/models/test_openai_type_adapter.py |  8 +++++++-
 tests/processors/test_base_processor.py  | 13 ++++++++++---
 tests/types/test_dsl.py                  |  7 ++++++-
 tests/types/test_types_utils.py          |  7 ++++++-
 12 files changed, 85 insertions(+), 28 deletions(-)

diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 2d3efaea2..a225b1a87 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -112,13 +112,15 @@ def generate_stream(
                 f"The type {output_type} is not available with Anthropic."
             )
 
-        completion = self.client.messages.stream(
+        stream = self.client.messages.create(
             **messages,
             model=self.model_name,
+            stream=True,
             **inference_kwargs,
         )
-        for chunk in completion:
-            if chunk.type == "content_block_delta":
+
+        for chunk in stream:
+            if chunk.type == "content_block_delta" and chunk.delta.type == "text_delta":
                 yield chunk.delta.text
 
 
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 20e297cd5..c6e582f58 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,10 +1,11 @@
 import datetime
 import json
 import re
+import sys
 from dataclasses import dataclass
 from enum import EnumMeta
 from types import FunctionType
-from typing import Any, Callable, List, Literal, Union, get_args, get_origin, _TypedDictMeta # type: ignore
+from typing import Any, Callable, List, Literal, Union, get_args, get_origin
 
 import interegular
 import jsonschema
@@ -44,6 +45,11 @@
 )
 from outlines_core.fsm.json_schema import build_regex_from_schema
 
+if sys.version_info >= (3, 12):
+    from typing import _TypedDictMeta  # type: ignore
+else:
+    from typing_extensions import _TypedDictMeta  # type: ignore
+
 
 class Term:
     """Represents types defined with a regular expression.
@@ -559,13 +565,14 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
 
 def _get_enum_members(ptype: EnumMeta) -> List[Any]:
     regular_members = [member.value for member in ptype]  # type: ignore
-    function_members = [
-        value for key, value in ptype.__dict__.items()
+    function_members = []
+    for key, value in ptype.__dict__.items():
         if (
             isinstance(value, FunctionType)
             and not (key.startswith('__') and key.endswith('__'))
-        )
-    ]
+            and key != '_generate_next_value_'  # Skip this specific method that causes issues
+        ):
+            function_members.append(value)
     return regular_members + function_members
 
 
@@ -605,7 +612,7 @@ def _handle_list(args: tuple, recursion_depth: int) -> Sequence:
 
 
 def _handle_tuple(args: tuple, recursion_depth: int) -> Union[Sequence, String]:
-    if len(args) == 0:
+    if len(args) == 0 or args == ((),):
         return String("()")
     elif len(args) == 2 and args[1] is Ellipsis:
         item_term = python_types_to_terms(args[0], recursion_depth + 1)
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
index 21d36281a..d195ad26e 100644
--- a/outlines/types/utils.py
+++ b/outlines/types/utils.py
@@ -1,7 +1,8 @@
 import dataclasses
 import datetime
+import sys
 from enum import Enum, EnumMeta
-from typing import ( # type: ignore
+from typing import (
     Annotated,
     Any,
     Dict,
@@ -12,13 +13,16 @@
     Union,
     get_args,
     get_origin,
-    _TypedDictMeta,
 )
 
 import interegular
 from genson import SchemaBuilder
 from pydantic import BaseModel
-from typing_extensions import _TypedDictMeta as Ext_TypedDictMeta  # type: ignore
+
+if sys.version_info >= (3, 12):
+    from typing import _TypedDictMeta  # type: ignore
+else:
+    from typing_extensions import _TypedDictMeta  # type: ignore
 
 
 # Type identification
@@ -117,13 +121,14 @@ def is_dataclass(value: Any) -> bool:
 
 
 def is_typed_dict(value: Any) -> bool:
-    return (
-        isinstance(value, _TypedDictMeta)
-        or isinstance(value, Ext_TypedDictMeta)
-    )
+    return isinstance(value, _TypedDictMeta)
+
 
+def is_pydantic_model(value):
+    # needed because generic type cannot be used with `issubclass`    # for Python versions < 3.11
+    if get_origin(value) is not None:
+        return False
 
-def is_pydantic_model(value: Any) -> bool:
     return isinstance(value, type) and issubclass(value, BaseModel)
 
 
diff --git a/tests/models/test_dottxt_type_adapter.py b/tests/models/test_dottxt_type_adapter.py
index c1e84c3ad..0253af1ea 100644
--- a/tests/models/test_dottxt_type_adapter.py
+++ b/tests/models/test_dottxt_type_adapter.py
@@ -1,8 +1,8 @@
 import io
 import json
 import pytest
+import sys
 from dataclasses import dataclass
-from typing_extensions import TypedDict
 
 from PIL import Image
 from genson import SchemaBuilder
@@ -12,6 +12,11 @@
 from outlines.templates import Vision
 from outlines.types import cfg, json_schema, regex
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
 
 @pytest.fixture
 def schema():
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 2697fe812..9dfe7a70e 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -1,5 +1,6 @@
 import io
 import json
+import sys
 from dataclasses import dataclass
 from enum import Enum
 from typing import Generator, Literal
@@ -9,12 +10,16 @@
 import requests
 from PIL import Image
 from pydantic import BaseModel
-from typing_extensions import TypedDict
 
 import outlines
 from outlines.models.gemini import Gemini
 from outlines.templates import Vision
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
 
 MODEL_NAME = "gemini-1.5-flash-latest"
 
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
index da9524850..a2d6b4a5d 100644
--- a/tests/models/test_gemini_type_adapter.py
+++ b/tests/models/test_gemini_type_adapter.py
@@ -1,6 +1,7 @@
 import io
 import json
 import pytest
+import sys
 from dataclasses import dataclass
 from enum import Enum, EnumMeta
 from typing import Literal, get_args
@@ -8,12 +9,16 @@
 from PIL import Image
 from genson import SchemaBuilder
 from pydantic import BaseModel
-from typing_extensions import TypedDict, is_typeddict
 
 from outlines import cfg, json_schema, regex
 from outlines.models.gemini import GeminiTypeAdapter
 from outlines.templates import Vision
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict, is_typeddict
+else:
+    from typing_extensions import TypedDict, is_typeddict
+
 
 @pytest.fixture
 def schema():
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
index d019f014e..58b6c6aa9 100644
--- a/tests/models/test_ollama_type_adapter.py
+++ b/tests/models/test_ollama_type_adapter.py
@@ -1,8 +1,8 @@
 import io
 import json
 import pytest
+import sys
 from dataclasses import dataclass
-from typing_extensions import TypedDict
 
 from genson import SchemaBuilder
 from PIL import Image
@@ -12,6 +12,11 @@
 from outlines.templates import Vision
 from outlines.types import cfg, json_schema, regex
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
 
 @pytest.fixture
 def schema():
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index df9182269..5b90508a6 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -64,7 +64,7 @@ class Foo:
         def __init__(self, foo):
             self.foo = foo
 
-    with pytest.raises(NotImplementedError, match="is not available"):
+    with pytest.raises(TypeError, match="is not available"):
         model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate(Foo("prompt"))
 
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
index 0e14cae08..fcafbb8de 100644
--- a/tests/models/test_openai_type_adapter.py
+++ b/tests/models/test_openai_type_adapter.py
@@ -1,8 +1,9 @@
 import io
 import json
 import pytest
+import sys
 from dataclasses import dataclass
-from typing import Literal, TypedDict
+from typing import Literal
 
 from genson import SchemaBuilder
 from PIL import Image
@@ -11,6 +12,11 @@
 from outlines import cfg, json_schema, regex, Vision
 from outlines.models.openai import OpenAITypeAdapter
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
 
 @pytest.fixture
 def schema():
diff --git a/tests/processors/test_base_processor.py b/tests/processors/test_base_processor.py
index d2a1e1af2..e135571e3 100644
--- a/tests/processors/test_base_processor.py
+++ b/tests/processors/test_base_processor.py
@@ -75,6 +75,13 @@ def test_call(array_type, processor):
     processed_logits = processor(input_ids, logits)
 
     assert isinstance(processed_logits, type(arrays[array_type]))
-    assert np.allclose(
-        np.array(processed_logits), np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
-    )
+
+    # Convert to numpy array for comparison, handling PyTorch tensors properly
+    if array_type == "torch":
+        expected = np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
+        actual = processed_logits.detach().cpu().numpy()
+        assert np.allclose(actual, expected)
+    else:
+        assert np.allclose(
+            np.array(processed_logits), np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
+        )
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index cf7a314db..cbc9b07fc 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -1,5 +1,6 @@
 import datetime
 import json
+import sys
 from dataclasses import dataclass
 from enum import Enum
 from typing import (
@@ -7,7 +8,6 @@
     List,
     Literal,
     Tuple,
-    TypedDict,
     Union,
     get_args,
     Optional as PyOptional
@@ -52,6 +52,11 @@
     cfg,
 )
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
 
 def test_dsl_init():
     string = String("test")
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
index e724551bc..1887f37a2 100644
--- a/tests/types/test_types_utils.py
+++ b/tests/types/test_types_utils.py
@@ -1,5 +1,6 @@
 import datetime
 import pytest
+import sys
 from dataclasses import dataclass
 from enum import Enum
 from typing import (
@@ -11,7 +12,6 @@
     NewType,
     Optional,
     Tuple,
-    TypedDict,
     Union
 )
 
@@ -45,6 +45,11 @@
     is_union
 )
 
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
 
 @pytest.fixture
 def sample_enum():

From 6a6a3574dd3875dad0f0a72d7b7e83a47ce65269 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 12 Mar 2025 10:54:23 +0100
Subject: [PATCH 280/505] Fix function get_enum_from_literal

---
 outlines/models/gemini.py       |  4 ++--
 outlines/types/utils.py         | 10 ++++++++--
 tests/types/test_types_utils.py | 28 +++++++++++++++++++++++++++-
 3 files changed, 37 insertions(+), 5 deletions(-)

diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 0201e6484..f3939af08 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -25,7 +25,7 @@
     is_enum,
     is_literal,
     is_typing_list,
-    literal_to_enum
+    get_enum_from_literal
 )
 
 if TYPE_CHECKING:
@@ -95,7 +95,7 @@ def format_output_type(self, output_type):
         elif is_enum(output_type):
             return self.format_enum_output_type(output_type)
         elif is_literal(output_type):
-            enum = literal_to_enum(output_type)
+            enum = get_enum_from_literal(output_type)
             return self.format_enum_output_type(enum)
         # list of objects
         elif is_typing_list(output_type):
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
index d195ad26e..99ffc5dcc 100644
--- a/outlines/types/utils.py
+++ b/outlines/types/utils.py
@@ -1,10 +1,13 @@
 import dataclasses
 import datetime
+import inspect
 import sys
+import warnings
 from enum import Enum, EnumMeta
 from typing import (
     Annotated,
     Any,
+    Callable,
     Dict,
     Literal,
     List,
@@ -151,5 +154,8 @@ def is_interegular_fsm(value: Any) -> bool:
 # Type conversion
 
 
-def literal_to_enum(value) -> Enum:
-    return Enum(value.__name__, [(arg, arg) for arg in get_args(value)])
+def get_enum_from_literal(value) -> Enum:
+    return Enum(
+        value.__name__,
+        {str(arg): arg for arg in get_args(value)}
+    )
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
index 1887f37a2..7fa668179 100644
--- a/tests/types/test_types_utils.py
+++ b/tests/types/test_types_utils.py
@@ -1,8 +1,10 @@
 import datetime
 import pytest
 import sys
+from contextlib import nullcontext
 from dataclasses import dataclass
 from enum import Enum
+from functools import partial
 from typing import (
     Annotated,
     Any,
@@ -17,9 +19,12 @@
 
 import interegular
 from genson import SchemaBuilder
-from pydantic import BaseModel
+from pydantic import BaseModel, constr
 
 from outlines.types.utils import (
+    get_enum_from_literal,
+    get_schema_from_enum,
+    get_schema_from_signature,
     is_bool,
     is_callable,
     is_date,
@@ -51,6 +56,9 @@
     from typing_extensions import TypedDict
 
 
+# Type identification
+
+
 @pytest.fixture
 def sample_enum():
     class SampleEnum(Enum):
@@ -330,3 +338,21 @@ def test_is_interegular_fsm(sample_interegular_fsm):
     assert is_interegular_fsm(sample_interegular_fsm)
     assert not is_interegular_fsm({})
     assert not is_interegular_fsm("")
+
+
+# Type conversion
+
+
+def test_get_enum_from_literal(sample_enum):
+    basic_enum = get_enum_from_literal(Literal["a", "b"])
+    assert(is_enum(basic_enum))
+    assert basic_enum.a.value == "a"
+    assert basic_enum.b.value == "b"
+
+    complex_enum = get_enum_from_literal(Literal["a", 1, True, None, sample_enum.A])
+    assert is_enum(complex_enum)
+    assert complex_enum.a.value == "a"
+    assert getattr(complex_enum, "1").value == 1
+    assert getattr(complex_enum, "True").value
+    assert getattr(complex_enum, "None").value is None
+    assert getattr(complex_enum, "SampleEnum.A").value == sample_enum.A

From b7860af5f6a5d1434b2e2f3406b0ae97b12fbd0b Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 12 Mar 2025 11:31:07 +0100
Subject: [PATCH 281/505] Move json_schema from fsm to types

---
 outlines/fsm/json_schema.py       | 83 -------------------------------
 outlines/processors/structured.py |  4 +-
 outlines/types/__init__.py        |  1 -
 outlines/types/dsl.py             |  4 +-
 outlines/types/utils.py           | 45 ++++++++++++++++-
 tests/fsm/test_json_schema.py     | 78 -----------------------------
 tests/types/test_types_utils.py   | 60 +++++++++++++++++++++-
 7 files changed, 106 insertions(+), 169 deletions(-)
 delete mode 100644 outlines/fsm/json_schema.py
 delete mode 100644 tests/fsm/test_json_schema.py

diff --git a/outlines/fsm/json_schema.py b/outlines/fsm/json_schema.py
deleted file mode 100644
index 578ee7626..000000000
--- a/outlines/fsm/json_schema.py
+++ /dev/null
@@ -1,83 +0,0 @@
-import inspect
-import json
-import warnings
-from enum import Enum
-from typing import Callable, Type, Union
-
-from pydantic import BaseModel, create_model
-
-
-def convert_json_schema_to_str(json_schema: Union[dict, str, Type[BaseModel]]) -> str:
-    """Convert a JSON schema to a string.
-
-    Parameters
-    ----------
-    json_schema
-        The JSON schema.
-
-    Returns
-    -------
-    str
-        The JSON schema converted to a string.
-
-    Raises
-    ------
-    ValueError
-        If the schema is not a dictionary, a string or a Pydantic class.
-    """
-    if isinstance(json_schema, dict):
-        schema_str = json.dumps(json_schema)
-    elif isinstance(json_schema, str):
-        schema_str = json_schema
-    elif issubclass(json_schema, BaseModel):
-        schema_str = json.dumps(json_schema.model_json_schema())
-    else:
-        raise ValueError(
-            f"Cannot parse schema {json_schema}. The schema must be either "
-            + "a Pydantic class, a dictionary or a string that contains the JSON "
-            + "schema specification"
-        )
-    return schema_str
-
-
-def get_schema_from_signature(fn: Callable) -> dict:
-    """Turn a function signature into a JSON schema.
-
-    Every JSON object valid to the output JSON Schema can be passed
-    to `fn` using the ** unpacking syntax.
-
-    """
-    signature = inspect.signature(fn)
-    arguments = {}
-    for name, arg in signature.parameters.items():
-        if arg.annotation == inspect._empty:
-            raise ValueError("Each argument must have a type annotation")
-        else:
-            arguments[name] = (arg.annotation, ...)
-
-    try:
-        fn_name = fn.__name__
-    except Exception as e:
-        fn_name = "Arguments"
-        warnings.warn(
-            f"The function name could not be determined. Using default name 'Arguments' instead. For debugging, here is exact error:\n{e}",
-            category=UserWarning,
-        )
-    model = create_model(fn_name, **arguments)
-
-    return model.model_json_schema()
-
-
-def get_schema_from_enum(myenum: type[Enum]) -> dict:
-    if len(myenum) == 0:
-        raise ValueError(
-            f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
-        )
-    choices = [
-        get_schema_from_signature(elt.value.func)
-        if callable(elt.value)
-        else {"const": elt.value}
-        for elt in myenum
-    ]
-    schema = {"title": myenum.__name__, "oneOf": choices}
-    return schema
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 225d39c3e..806172eaa 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -32,7 +32,7 @@
 from pydantic import BaseModel
 
 from outlines.processors.guide import CFGGuide, Guide, RegexGuide
-from outlines.fsm.json_schema import convert_json_schema_to_str
+from outlines.types import JsonSchema
 
 from .base_logits_processor import OutlinesLogitsProcessor
 
@@ -182,7 +182,7 @@ def __init__(
             literals). For example, to allow only a single space or newline with
             `whitespace_pattern=r"[\n ]?"`
         """
-        schema_str = convert_json_schema_to_str(json_schema=schema)
+        schema_str = JsonSchema(schema).schema
         regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
         super().__init__(regex_string=regex_string, tokenizer=tokenizer)
 
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 1872f4193..29f1d7a0d 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -5,7 +5,6 @@
 
 from jsonschema import Draft202012Validator as Validator
 from jsonschema.exceptions import SchemaError
-from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel, TypeAdapter
 from typing_extensions import _TypedDictMeta  # type: ignore
 
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index c6e582f58..82d7cd1a7 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -10,14 +10,15 @@
 import interegular
 import jsonschema
 from genson import SchemaBuilder
+from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
 from pydantic.json_schema import JsonSchemaValue
 from pydantic_core import core_schema as cs
 
 import outlines.types as types
 from outlines import grammars
-from outlines.fsm.json_schema import get_schema_from_signature
 from outlines.types.utils import (
+    get_schema_from_signature,
     is_int,
     is_int_instance,
     is_float,
@@ -43,7 +44,6 @@
     is_typing_dict,
     is_interegular_fsm,
 )
-from outlines_core.fsm.json_schema import build_regex_from_schema
 
 if sys.version_info >= (3, 12):
     from typing import _TypedDictMeta  # type: ignore
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
index 99ffc5dcc..e1be024ba 100644
--- a/outlines/types/utils.py
+++ b/outlines/types/utils.py
@@ -20,7 +20,7 @@
 
 import interegular
 from genson import SchemaBuilder
-from pydantic import BaseModel
+from pydantic import BaseModel, create_model
 
 if sys.version_info >= (3, 12):
     from typing import _TypedDictMeta  # type: ignore
@@ -159,3 +159,46 @@ def get_enum_from_literal(value) -> Enum:
         value.__name__,
         {str(arg): arg for arg in get_args(value)}
     )
+
+
+def get_schema_from_signature(fn: Callable) -> dict:
+    """Turn a function signature into a JSON schema.
+
+    Every JSON object valid to the output JSON Schema can be passed
+    to `fn` using the ** unpacking syntax.
+
+    """
+    signature = inspect.signature(fn)
+    arguments = {}
+    for name, arg in signature.parameters.items():
+        if arg.annotation == inspect._empty:
+            raise ValueError("Each argument must have a type annotation")
+        else:
+            arguments[name] = (arg.annotation, ...)
+
+    try:
+        fn_name = fn.__name__
+    except Exception as e:
+        fn_name = "Arguments"
+        warnings.warn(
+            f"The function name could not be determined. Using default name 'Arguments' instead. For debugging, here is exact error:\n{e}",
+            category=UserWarning,
+        )
+    model = create_model(fn_name, **arguments)
+
+    return model.model_json_schema()
+
+
+def get_schema_from_enum(myenum: type[Enum]) -> dict:
+    if len(myenum) == 0:
+        raise ValueError(
+            f"Your enum class {myenum.__name__} has 0 members. If you are working with an enum of functions, do not forget to register them as callable (using `partial` for instance)"
+        )
+    choices = [
+        get_schema_from_signature(elt.value.func)
+        if callable(elt.value)
+        else {"const": elt.value}
+        for elt in myenum
+    ]
+    schema = {"title": myenum.__name__, "oneOf": choices}
+    return schema
diff --git a/tests/fsm/test_json_schema.py b/tests/fsm/test_json_schema.py
deleted file mode 100644
index 23864e029..000000000
--- a/tests/fsm/test_json_schema.py
+++ /dev/null
@@ -1,78 +0,0 @@
-import json
-from contextlib import nullcontext
-from enum import Enum
-from functools import partial
-from typing import List
-
-import pytest
-from outlines_core.fsm.json_schema import build_regex_from_schema
-from pydantic import BaseModel, constr
-
-from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
-
-
-def test_function_basic():
-    def test_function(foo: str, bar: List[int]):
-        pass
-
-    result = get_schema_from_signature(test_function)
-    assert result["type"] == "object"
-    assert list(result["properties"].keys()) == ["foo", "bar"]
-    assert result["properties"]["foo"]["type"] == "string"
-    assert result["properties"]["bar"]["type"] == "array"
-    assert result["properties"]["bar"]["items"]["type"] == "integer"
-
-
-def test_function_no_type():
-    def test_function(foo, bar: List[int]):
-        pass
-
-    with pytest.raises(ValueError):
-        get_schema_from_signature(test_function)
-
-
-def test_from_pydantic():
-    class User(BaseModel):
-        user_id: int
-        name: str
-        maxlength_name: constr(max_length=10)
-        minlength_name: constr(min_length=10)
-        value: float
-        is_true: bool
-
-    schema = json.dumps(User.model_json_schema())
-    regex_str = build_regex_from_schema(schema)
-    assert isinstance(regex_str, str)
-
-
-def add(a: float, b: float) -> float:
-    return a + b
-
-
-class MyEnum(Enum):
-    add = partial(add)
-    a = "a"
-    b = 2
-
-
-# if you don't register your function as callable, you will get an empty enum
-class EmptyEnum(Enum):
-    add = add
-
-
-@pytest.mark.parametrize(
-    "enum,expectation",
-    [
-        (MyEnum, nullcontext()),
-        (EmptyEnum, pytest.raises(ValueError)),
-    ],
-)
-def test_enum_schema(enum, expectation):
-    with expectation:
-        schema = get_schema_from_enum(enum)
-        regex_str = build_regex_from_schema(json.dumps(schema))
-        assert isinstance(regex_str, str)
-        assert schema["title"] == enum.__name__
-        assert len(schema["oneOf"]) == len(enum)
-        for elt in schema["oneOf"]:
-            assert type(elt) in [int, float, bool, type(None), str, dict]
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
index 7fa668179..0410d5aa7 100644
--- a/tests/types/test_types_utils.py
+++ b/tests/types/test_types_utils.py
@@ -1,7 +1,6 @@
 import datetime
 import pytest
 import sys
-from contextlib import nullcontext
 from dataclasses import dataclass
 from enum import Enum
 from functools import partial
@@ -21,6 +20,7 @@
 from genson import SchemaBuilder
 from pydantic import BaseModel, constr
 
+from outlines.types.dsl import JsonSchema
 from outlines.types.utils import (
     get_enum_from_literal,
     get_schema_from_enum,
@@ -67,6 +67,29 @@ class SampleEnum(Enum):
 
     return SampleEnum
 
+@pytest.fixture
+def sample_complex_enum():
+    def add_func(a: float, b: float) -> float:
+        return a + b
+
+    class SampleComplexEnum(Enum):
+        add = partial(add_func)
+        a = "a"
+        b = 2
+
+    return SampleComplexEnum
+
+@pytest.fixture
+def sample_empty_enum():
+    def add_func(a: float, b: float) -> float:
+        return a + b
+
+    # the enum is empty because the function is not registered as callable
+    class SampleEmptyEnum(Enum):
+        add = add_func
+
+    return SampleEmptyEnum
+
 @pytest.fixture
 def sample_class():
     class SampleClass:
@@ -109,7 +132,14 @@ def sample_schema_builder():
 
 @pytest.fixture
 def sample_function():
-    def sample_function():
+    def sample_function(foo: str, bar: List[int]):
+        pass
+
+    return sample_function
+
+@pytest.fixture
+def sample_function_missing_type():
+    def sample_function(foo, bar: List[int]):
         pass
 
     return sample_function
@@ -356,3 +386,29 @@ def test_get_enum_from_literal(sample_enum):
     assert getattr(complex_enum, "True").value
     assert getattr(complex_enum, "None").value is None
     assert getattr(complex_enum, "SampleEnum.A").value == sample_enum.A
+
+
+def test_get_schema_from_signature(sample_function, sample_function_missing_type):
+    result = get_schema_from_signature(sample_function)
+    assert result["type"] == "object"
+    assert list(result["properties"].keys()) == ["foo", "bar"]
+    assert result["properties"]["foo"]["type"] == "string"
+    assert result["properties"]["bar"]["type"] == "array"
+    assert result["properties"]["bar"]["items"]["type"] == "integer"
+
+    # in case of a function missing type annotations
+    with pytest.raises(ValueError):
+        get_schema_from_signature(sample_function_missing_type)
+
+
+def test_get_schema_from_enum(sample_complex_enum, sample_empty_enum):
+    schema = get_schema_from_enum(sample_complex_enum)
+    assert JsonSchema(schema)
+    assert schema["title"] == sample_complex_enum.__name__
+    assert len(schema["oneOf"]) == len(sample_complex_enum)
+    for elt in schema["oneOf"]:
+        assert type(elt) in [int, float, bool, type(None), str, dict]
+
+    # in case of an empty enum because the function member is not registered as callable
+    with pytest.raises(ValueError):
+        get_schema_from_enum(sample_empty_enum)

From d9aaf4fb92d742a89defac026c494f96f24e8828 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 13 Mar 2025 17:23:10 +0100
Subject: [PATCH 282/505] Make Generator accept logits processors as output
 paramater + test it

---
 outlines/generator.py                        | 122 ++++++++++-----
 outlines/processors/base_logits_processor.py |   3 +-
 outlines/types/__init__.py                   |   3 +-
 outlines/types/dsl.py                        |  21 ++-
 tests/test_generator.py                      | 153 +++++++++++++++++++
 tests/types/test_dsl.py                      |  11 +-
 tests/types/test_types_utils.py              |  12 --
 7 files changed, 256 insertions(+), 69 deletions(-)
 create mode 100644 tests/test_generator.py

diff --git a/outlines/generator.py b/outlines/generator.py
index 088b6d113..0d9c86ea9 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -1,37 +1,35 @@
-from dataclasses import dataclass
-from typing import Any, Optional, Union, cast, get_args
+from typing import Any, Optional, Union
 
-import interegular
-
-from outlines.processors.guide import RegexGuide
 from outlines.models import BlackBoxModel, SteerableModel
-from outlines.processors import CFGLogitsProcessor, GuideLogitsProcessor, RegexLogitsProcessor
-from outlines.types import CFG
+from outlines.processors import (
+    CFGLogitsProcessor,
+    GuideLogitsProcessor,
+    RegexLogitsProcessor,
+    OutlinesLogitsProcessor,
+)
+from outlines.processors.guide import RegexGuide
+from outlines.types import CFG, FSM
 from outlines.types.dsl import python_types_to_terms, to_regex
 
 
-@dataclass
 class BlackBoxGenerator:
     """Represents a generator for which we don't control constrained generation.
 
-    Attributes
-    ----------
-    model
-        An instance of a model wrapper.
-    output_type
-        The output type.
-
+    This type of generator only accepts an output type as an argument defining
+    constrained generation. This output type is not modified and thus only
+    passed through to the model.
     """
+    output_type: Optional[Any]
 
-    model: BlackBoxModel
-    output_type: Optional[Any] = None
+    def __init__(self, model, output_type: Optional[Any]):
+        self.model = model
+        self.output_type = output_type
 
-    def __post_init__(self):
         if isinstance(self.output_type, CFG):
             raise NotImplementedError(
                 "CFG generation is not supported for API-based models"
             )
-        elif isinstance(self.output_type, interegular.fsm.FSM):
+        elif isinstance(self.output_type, FSM):
             raise NotImplementedError(
                 "FSM generation is not supported for API-based models"
             )
@@ -43,37 +41,35 @@ def stream(self, prompt, **inference_kwargs):
         return self.model.generate_stream(prompt, self.output_type, **inference_kwargs)
 
 
-@dataclass
 class SteerableGenerator:
     """Represents a generator for which we control constrained generation.
 
-    We use this class to keep track of the logits processor which can be quite
-    expensive to build.
-
-    Attributes
-    ----------
-    model
-        An instance of a model wrapper.
-    output_type
-        The output type.
-
+    The generator is responsible for building and storing the logits processor
+    (which can be quite expensive to build), and then to pass it to the model
+    when the generator is called. The argument defining constrained generation
+    can be of 2 types associated to different methods to create an instance of
+    the generator:
+    - output_type (through `__init__`): a Python type, will be used to build a
+      logits processor
+    - processor (through `from_processor`): an already built
+      logits processor
     """
+    logits_processor: Optional[OutlinesLogitsProcessor]
 
-    model: SteerableModel
-    output_type: Optional[Any]
+    def __init__(self, model, output_type: Optional[Any]):
+        self.model = model
 
-    def __post_init__(self):
-        if self.output_type is None:
+        if output_type is None:
             self.logits_processor = None
         else:
-            term = python_types_to_terms(self.output_type)
+            term = python_types_to_terms(output_type)
             if isinstance(term, CFG):
                 cfg_string = term.definition
                 self.logits_processor = CFGLogitsProcessor(
                     cfg_string, self.model.tokenizer
                 )
-            elif isinstance(term, interegular.fsm.FSM):
-                guide = RegexGuide.from_interegular_fsm(term, self.model.tokenizer)
+            elif isinstance(term, FSM):
+                guide = RegexGuide.from_interegular_fsm(term.fsm, self.model.tokenizer)
                 self.logits_processor = GuideLogitsProcessor(tokenizer=self.model.tokenizer, guide=guide)
             else:
                 regex_string = to_regex(term)
@@ -81,6 +77,16 @@ def __post_init__(self):
                     regex_string, self.model.tokenizer
                 )
 
+    @classmethod
+    def from_processor(cls, model, processor: OutlinesLogitsProcessor):
+        if not isinstance(processor, OutlinesLogitsProcessor):
+            raise TypeError("The processor argument must be an instance of OutlinesLogitsProcessor")
+        instance = cls.__new__(cls)
+        instance.model = model
+        instance.logits_processor = processor
+
+        return instance
+
     def __call__(self, prompt, **inference_kwargs):
         return self.model.generate(prompt, self.logits_processor, **inference_kwargs)
 
@@ -91,8 +97,44 @@ def stream(self, prompt, **inference_kwargs):
 def Generator(
     model: Union[SteerableModel, BlackBoxModel],
     output_type: Optional[Any] = None,
+    *,
+    processor: Optional[OutlinesLogitsProcessor] = None,
 ):
-    if isinstance(model, BlackBoxModel):  # type: ignore
-        return BlackBoxGenerator(model, output_type)  # type: ignore
+    """
+    Create a generator for the given model and output parameters.
+
+    The 2 parameters output_type and processor are mutually exclusive. The
+    parameters processor is only supported for SteerableModel instances
+    (typically local models) and is intended to be only used by advanced users.
+
+    Parameters
+    ----------
+    model
+        An instance of an Outlines model.
+    output_type
+        The output type expressed as a Python type or a type defined in the
+        outlines.types.dsl module.
+    processor
+        An instance of an OutlinesLogitsProcessor.
+
+    Returns
+    -------
+    A generator.
+    """
+    provided_output_params = sum(
+        param is not None
+        for param in [output_type, processor]
+    )
+    if provided_output_params > 1:
+        raise ValueError("At most one of output_type or processor can be provided")
+
+    if isinstance(model, BlackBoxModel): # type: ignore
+        if processor is not None:
+            raise NotImplementedError("This model does not support logits processors")
+        else:
+            return BlackBoxGenerator(model, output_type)
     else:
-        return SteerableGenerator(model, output_type)  # type: ignore
+        if processor is not None:
+            return SteerableGenerator.from_processor(model, processor)
+        else:
+            return SteerableGenerator(model, output_type)
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index a74e7bf1f..d9549d780 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -1,5 +1,5 @@
 from abc import abstractmethod
-from typing import TYPE_CHECKING, List, Protocol, Type, Union
+from typing import TYPE_CHECKING, List, Protocol, Type, Union, runtime_checkable
 
 import numpy as np
 import torch
@@ -30,6 +30,7 @@ def is_jax_array_type(array_type):
     )
 
 
+@runtime_checkable
 class OutlinesLogitsProcessor(Protocol):
     """
     Base class for logits processors which normalizes types of logits:
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 29f1d7a0d..d653f2a06 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -12,9 +12,11 @@
 from outlines.types.dsl import (
     Regex,
     CFG,
+    FSM,
     JsonSchema,
     regex,
     cfg,
+    fsm,
     json_schema,
     optional,
     either,
@@ -22,7 +24,6 @@
     at_least,
     at_most,
     between,
-    repeat,
     zero_or_more,
     one_or_more,
 )
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 82d7cd1a7..31495e63c 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -212,6 +212,17 @@ def __repr__(self):
         return f"CFG(definition='{self.definition}')"
 
 
+@dataclass
+class FSM(Term):
+    fsm: interegular.fsm.FSM
+
+    def _display_node(self) -> str:
+        return f"FSM({self.fsm.__repr__()})"
+
+    def __repr__(self):
+        return f"FSM(fsm={self.fsm.__repr__()})"
+
+
 class JsonSchema(Term):
     def __init__(self, schema: Union[dict, str, type[BaseModel], _TypedDictMeta, type, SchemaBuilder]):
         schema_str: str
@@ -413,6 +424,10 @@ def cfg(definition: str):
     return CFG(definition)
 
 
+def fsm(fsm: interegular.fsm.FSM):
+    return FSM(fsm)
+
+
 def json_schema(schema: Union[str, dict, type[BaseModel]]):
     return JsonSchema(schema)
 
@@ -550,12 +565,6 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     if is_callable(ptype):
         return JsonSchema(get_schema_from_signature(ptype))
 
-    if isinstance(ptype, interegular.fsm.FSM):
-        return ptype
-
-    if is_interegular_fsm(ptype):
-        return ptype
-
     type_name = getattr(ptype, "__name__", ptype)
     raise TypeError(
         f"Type {type_name} is currently not supported. Please open an issue: "
diff --git a/tests/test_generator.py b/tests/test_generator.py
new file mode 100644
index 000000000..fb8fc8b90
--- /dev/null
+++ b/tests/test_generator.py
@@ -0,0 +1,153 @@
+import os
+import pytest
+from typing import Literal, Generator as TypingGenerator
+
+import interegular
+import transformers
+from openai import OpenAI
+
+import outlines
+from outlines.generator import BlackBoxGenerator, SteerableGenerator, Generator
+from outlines.models import BlackBoxModel, SteerableModel
+from outlines.processors import (
+    OutlinesLogitsProcessor,
+    RegexLogitsProcessor,
+)
+from outlines.types import CFG, FSM
+
+
+@pytest.fixture(scope="session")
+def steerable_model():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return model
+
+
+@pytest.fixture(scope="session")
+def sample_processor():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    processor = RegexLogitsProcessor(
+        regex_string="[0-9]{3}",
+        tokenizer=model.tokenizer,
+    )
+    return processor
+
+
+@pytest.fixture(scope="session")
+def black_box_model():
+    api_key = os.getenv("OPENAI_API_KEY")
+    if not api_key:
+        api_key = "MOCK_API_KEY"
+    client = OpenAI(api_key=api_key)
+    model = outlines.from_openai(client, model_name="gpt-4o-mini")
+    return model
+
+
+# SteerableGenerator
+
+
+def test_steerable_generator_init_valid_processor(steerable_model, sample_processor):
+    generator = SteerableGenerator.from_processor(steerable_model, sample_processor)
+    assert generator.logits_processor == sample_processor
+    assert generator.model == steerable_model
+
+
+def test_steerable_generator_init_invalid_processor(steerable_model):
+    with pytest.raises(TypeError):
+        SteerableGenerator.from_processor(steerable_model, Literal["foo", "bar"])
+
+
+def test_steerable_generator_init_valid_output_type(steerable_model):
+    generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_steerable_generator_init_invalid_output_type(steerable_model, sample_processor):
+    with pytest.raises(TypeError):
+        SteerableGenerator(steerable_model, sample_processor)
+
+
+def test_steerable_generator_call(steerable_model):
+    generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
+    result = generator("foo", max_new_tokens=10)
+    assert isinstance(result, str)
+
+
+def test_steerable_generator_stream(steerable_model):
+    with pytest.raises(NotImplementedError):
+        generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
+        result = generator.stream("foo", max_tokens=10)
+        assert isinstance(result, TypingGenerator)
+        assert isinstance(next(result), str)
+
+
+# BlackBoxGenerator
+
+
+def test_black_box_generator_init_valid(black_box_model):
+    generator = BlackBoxGenerator(black_box_model, Literal["foo", "bar"])
+    assert generator.model == black_box_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_black_box_generator_init_invalid(black_box_model):
+    with pytest.raises(NotImplementedError):
+        BlackBoxGenerator(black_box_model, CFG("foo"))
+
+    with pytest.raises(NotImplementedError):
+        BlackBoxGenerator(black_box_model, FSM("foo"))
+
+
+@pytest.mark.api_call
+def test_black_box_generator_call(black_box_model):
+    generator = BlackBoxGenerator(black_box_model, dict)
+    result = generator("write a very short json object", max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_black_box_generator_stream(black_box_model):
+    generator = BlackBoxGenerator(black_box_model, dict)
+    result = generator.stream("write a very short json object", max_tokens=10)
+    assert isinstance(result, TypingGenerator)
+    assert isinstance(next(result), str)
+
+
+# Generator
+
+
+def test_generator_init_multiple_output_type(steerable_model, sample_processor):
+    with pytest.raises(ValueError):
+        Generator(steerable_model, Literal["foo", "bar"], processor=sample_processor)
+
+
+def test_generator_steerable_output_type(steerable_model):
+    generator = Generator(steerable_model, Literal["foo", "bar"])
+    assert isinstance(generator, SteerableGenerator)
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_generator_steerable_processor(steerable_model, sample_processor):
+    generator = Generator(steerable_model, processor=sample_processor)
+    assert isinstance(generator, SteerableGenerator)
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_generator_black_box_output_type(black_box_model):
+    generator = Generator(black_box_model, Literal["foo", "bar"])
+    assert isinstance(generator, BlackBoxGenerator)
+    assert generator.model == black_box_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_generator_black_box_processor(black_box_model, sample_processor):
+    with pytest.raises(NotImplementedError):
+        Generator(black_box_model, processor=sample_processor)
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index cbc9b07fc..f2973aa77 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -405,15 +405,8 @@ class PydanticModel(BaseModel):
         }
     )
 
-    interegular_fsm = interegular.fsm.FSM(
-        alphabet={},
-        states={},
-        initial={},
-        finals={},
-        map={},
-        __no_validation__=True,
-    )
-    assert python_types_to_terms(interegular_fsm) is interegular_fsm
+    interegular_fsm = interegular.parse_pattern(r"abc").to_fsm()
+    assert python_types_to_terms(types.fsm(interegular_fsm)).fsm is interegular_fsm
 
     def func(a: int, b: str):
         return (a, b)
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
index 0410d5aa7..f32aae924 100644
--- a/tests/types/test_types_utils.py
+++ b/tests/types/test_types_utils.py
@@ -36,7 +36,6 @@
     is_genson_schema_builder,
     is_int,
     is_int_instance,
-    is_interegular_fsm,
     is_literal,
     is_native_dict,
     is_pydantic_model,
@@ -144,11 +143,6 @@ def sample_function(foo, bar: List[int]):
 
     return sample_function
 
-@pytest.fixture
-def sample_interegular_fsm():
-    pattern = interegular.parse_pattern(r"[a-z]{3}-[0-9]{2}")
-    return pattern.to_fsm()
-
 
 def test_is_int():
     assert is_int(int)
@@ -364,12 +358,6 @@ def test_is_callable(sample_function, sample_class, sample_dataclass, sample_typ
     assert not is_callable(sample_pydantic_model)
 
 
-def test_is_interegular_fsm(sample_interegular_fsm):
-    assert is_interegular_fsm(sample_interegular_fsm)
-    assert not is_interegular_fsm({})
-    assert not is_interegular_fsm("")
-
-
 # Type conversion
 
 

From 2c5edef53aec5166358186fb3673d9a3dd0ba5ac Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 13 Mar 2025 17:23:23 +0100
Subject: [PATCH 283/505] Delete tests/generate and move the content to
 tests/test_cache.py

---
 tests/generate/__init__.py |  0
 tests/generate/conftest.py | 24 ------------------------
 tests/test_cache.py        | 22 ++++++++++++++++++++++
 3 files changed, 22 insertions(+), 24 deletions(-)
 delete mode 100644 tests/generate/__init__.py
 delete mode 100644 tests/generate/conftest.py

diff --git a/tests/generate/__init__.py b/tests/generate/__init__.py
deleted file mode 100644
index e69de29bb..000000000
diff --git a/tests/generate/conftest.py b/tests/generate/conftest.py
deleted file mode 100644
index b067aa46a..000000000
--- a/tests/generate/conftest.py
+++ /dev/null
@@ -1,24 +0,0 @@
-from importlib import reload
-
-import pytest
-
-
-@pytest.fixture
-def temp_cache_dir():
-    import os
-    import tempfile
-
-    import outlines.caching
-    import outlines.processors.guide
-
-    with tempfile.TemporaryDirectory() as tempdir:
-        os.environ["OUTLINES_CACHE_DIR"] = tempdir
-        outlines.caching.get_cache.cache_clear()
-        reload(outlines)
-        reload(outlines.processors.guide)
-        cache_status = outlines.caching._caching_enabled
-        try:
-            outlines.caching._caching_enabled = True
-            yield
-        finally:
-            outlines.caching._caching_enabled = cache_status
diff --git a/tests/test_cache.py b/tests/test_cache.py
index eb4ec406e..9987747b0 100644
--- a/tests/test_cache.py
+++ b/tests/test_cache.py
@@ -1,6 +1,7 @@
 import os
 import tempfile
 import unittest
+from importlib import reload
 
 import diskcache
 import pytest
@@ -188,3 +189,24 @@ def fn():
     # scope has exited, cache is enabled again
     fn()
     assert mock.call_count == 2
+
+
+@pytest.fixture
+def temp_cache_dir():
+    import os
+    import tempfile
+
+    import outlines.caching
+    import outlines.processors.guide
+
+    with tempfile.TemporaryDirectory() as tempdir:
+        os.environ["OUTLINES_CACHE_DIR"] = tempdir
+        outlines.caching.get_cache.cache_clear()
+        reload(outlines)
+        reload(outlines.processors.guide)
+        cache_status = outlines.caching._caching_enabled
+        try:
+            outlines.caching._caching_enabled = True
+            yield
+        finally:
+            outlines.caching._caching_enabled = cache_status

From 26b3488cd7818bd9f2ff74ee3401c8c01d8127f8 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Sun, 16 Mar 2025 23:08:19 +0100
Subject: [PATCH 284/505] Create the Application class

---
 docs/api/applications.md       |  1 +
 docs/reference/applications.md | 40 ++++++++++++++
 outlines/__init__.py           |  2 +
 outlines/applications.py       | 63 ++++++++++++++++++++++
 tests/test_applications.py     | 95 ++++++++++++++++++++++++++++++++++
 5 files changed, 201 insertions(+)
 create mode 100644 docs/api/applications.md
 create mode 100644 docs/reference/applications.md
 create mode 100644 outlines/applications.py
 create mode 100644 tests/test_applications.py

diff --git a/docs/api/applications.md b/docs/api/applications.md
new file mode 100644
index 000000000..b44ddf566
--- /dev/null
+++ b/docs/api/applications.md
@@ -0,0 +1 @@
+::: outlines.applications
diff --git a/docs/reference/applications.md b/docs/reference/applications.md
new file mode 100644
index 000000000..10c036cc0
--- /dev/null
+++ b/docs/reference/applications.md
@@ -0,0 +1,40 @@
+# Applications
+
+Outlines offers a convenient way of encapsulating a prompt template and an output type in a single object called an `Application`. After instantiating an `Application`, it can be called to generate a response by providing a model and optional arguments used to create the prompt from the template.
+
+## Create an Application
+
+To create an Application, you need to provide 2 arguments:
+
+- A template: either an instance of `outlines.templates.Template` or a callable that takes arguments and returns a prompt
+- An output type: a type instance from `outlines.types` that is used to define the structure of the output
+
+```python
+from pydantic import BaseModel
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from outlines import models, Application
+from outlines.types import JsonType
+from outlines.templates import Template
+
+class OutputModel(BaseModel):
+    result: int
+
+model = models.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+template_string = "What is 2 times {{ num }}?"
+template = Template.from_str(template_string)
+
+application = Application(template, JsonType(OutputModel))
+```
+
+## Call an Application
+
+Once the Application is instantiated, it can be called. The first argument is the model to use to generate the response. The remaining arguments are passed to the template to create the prompt.
+
+```python
+result = application(model, num=3)
+print(result) # Expected output: { "result" : 6 }
+```
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 0c5ed5454..8f87e0357 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -4,6 +4,7 @@
 import outlines.models
 import outlines.processors
 import outlines.types
+from outlines.applications import Application
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.function import Function
 from outlines.generator import Generator
@@ -40,6 +41,7 @@
     "clear_cache",
     "disable_cache",
     "get_cache",
+    "Application",
     "Function",
     "Generator",
     "regex",
diff --git a/outlines/applications.py b/outlines/applications.py
new file mode 100644
index 000000000..8dbb69741
--- /dev/null
+++ b/outlines/applications.py
@@ -0,0 +1,63 @@
+from typing import Any, Callable, Optional, Union
+
+from outlines.generator import BlackBoxGenerator, Generator, SteerableGenerator
+from outlines.templates import Template
+from outlines.models import BlackBoxModel, SteerableModel
+
+
+class Application:
+    """
+    Application is a class that encapsulates a prompt template and an
+    output type. It can be called to generate a response by providing a
+    model and optional arguments used to create the prompt from the template.
+
+    Parameters
+    ----------
+    template : Union[Template, Callable]
+        A callable that takes arguments and returns a prompt string.
+    output_type : Any
+        The expected output type of the generated response.
+
+    Examples
+    --------
+    from pydantic import BaseModel
+    from transformers import AutoModelForCausalLM, AutoTokenizer
+    from outlines import models, Application
+    from outlines.types import JsonType
+    from outlines.templates import Template
+
+    class OutputModel(BaseModel):
+        result: int
+
+    model = models.from_transformers(
+        AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+        AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+    )
+
+    template_string = "What is 2 times {{ num }}?"
+    template = Template.from_str(template_string)
+
+    application = Application(template, JsonType(OutputModel))
+
+    result = application(model, num=3)
+    print(result)  # Expected output: { "result" : 6 }
+    """
+    def __init__(self, template: Union[Template, Callable], output_type: Any):
+        self.template = template
+        self.output_type = output_type
+        self.model: Optional[Union[BlackBoxModel, SteerableModel]] = None
+        self.generator: Optional[Union[BlackBoxGenerator, SteerableGenerator]] = None
+
+    def __call__(self, model: Union[BlackBoxModel, SteerableModel], *args, **kwargs):
+        if model is None:
+            raise ValueError("you must provide a model")
+        # We save the generator to avoid creating a new one for each call.
+        # If the model has changed since the last call, we create a new
+        # generator.
+        elif model != self.model:
+            self.model = model
+            self.generator = Generator(model, self.output_type)
+
+        prompt = self.template(*args, **kwargs)
+        assert self.generator is not None
+        return self.generator(prompt)
diff --git a/tests/test_applications.py b/tests/test_applications.py
new file mode 100644
index 000000000..2f2d4f056
--- /dev/null
+++ b/tests/test_applications.py
@@ -0,0 +1,95 @@
+import pytest
+
+import jinja2
+
+from outlines.applications import Application
+from outlines.templates import Template
+from outlines.models import Model
+from typing import Any
+
+
+@pytest.fixture(scope="session")
+def model():
+    class MockModel(Model):
+        type_adapter = None
+
+        def generate(self, model_input: str, output_type=None, **kwargs) -> Any:
+            return model_input
+
+        def generate_stream(self, model_input: str, output_type=None, **kwargs) -> Any:
+            return model_input
+
+    return MockModel()
+
+
+@pytest.fixture(scope="session")
+def another_model():
+    class MockModel(Model):
+        type_adapter = None
+
+        def generate(self, model_input: str, output_type=None, **kwargs) -> Any:
+            return model_input
+
+        def generate_stream(self, model_input: str, output_type=None, **kwargs) -> Any:
+            return model_input
+
+    return MockModel()
+
+
+def test_application_initialization():
+    template = Template.from_str("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    assert application.template == template
+    assert application.output_type == output_type
+    assert application.model is None
+    assert application.generator is None
+
+
+def test_application_template_call(model):
+    template = Template.from_str("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+    result = application(model, value="example")
+
+    assert result == "Test example"
+
+
+def test_application_callable_call(model):
+    def template(value):
+        return f"Test {value}"
+
+    output_type = None
+    application = Application(template, output_type)
+    result = application(model, value="example")
+
+    assert result == "Test example"
+
+
+def test_application_template_error(model):
+    template = Template.from_str("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    with pytest.raises(jinja2.exceptions.UndefinedError):
+        application(model, foo="bar")
+
+
+def test_application_generator_reuse(model, another_model):
+    template = Template.from_str("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    application(model, value="example")
+    first_generator = application.generator
+    first_model = application.model
+
+    application(model, value="example")
+    assert application.model == first_model
+    assert application.generator == first_generator
+
+    application(another_model, value="example")
+    assert application.model == another_model
+    assert application.model != first_model
+    assert application.generator != first_generator

From 209f0ca33ca0e981d80856ca3ba2412fd904b274 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Sun, 16 Mar 2025 23:14:12 +0100
Subject: [PATCH 285/505] Remove the Function class

---
 outlines/__init__.py   |   2 -
 outlines/function.py   | 123 -----------------------------------
 tests/test_function.py | 142 -----------------------------------------
 3 files changed, 267 deletions(-)
 delete mode 100644 outlines/function.py
 delete mode 100644 tests/test_function.py

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 8f87e0357..a2c7be86f 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -6,7 +6,6 @@
 import outlines.types
 from outlines.applications import Application
 from outlines.caching import clear_cache, disable_cache, get_cache
-from outlines.function import Function
 from outlines.generator import Generator
 from outlines.templates import Template, prompt
 from outlines.types import regex, json_schema, cfg
@@ -42,7 +41,6 @@
     "disable_cache",
     "get_cache",
     "Application",
-    "Function",
     "Generator",
     "regex",
     "json_schema",
diff --git a/outlines/function.py b/outlines/function.py
deleted file mode 100644
index be1a8095d..000000000
--- a/outlines/function.py
+++ /dev/null
@@ -1,123 +0,0 @@
-import importlib.util
-from dataclasses import dataclass
-from typing import TYPE_CHECKING, Callable, Optional, Tuple, Union
-
-import requests
-
-import outlines
-from outlines.types import JsonSchema
-
-if TYPE_CHECKING:
-    from outlines.templates import Template
-
-
-@dataclass
-class Function:
-    """Represents an Outlines function.
-
-    Functions are a convenient way to encapsulate a prompt template, a language
-    model and a Pydantic model that define the output structure. Once defined,
-    the function can be called with arguments that will be used to render the
-    prompt template.
-
-    """
-
-    prompt_template: "Template"
-    schema: Union[str, Callable, object]
-    model_name: str
-    generator = None
-
-    @classmethod
-    def from_github(cls, program_path: str, function_name: str = "fn"):
-        """Load a function stored on GitHub"""
-        program_content = download_from_github(program_path)
-        function = extract_function_from_file(program_content, function_name)
-
-        return function
-
-    def init_generator(self):
-        """Load the model and initialize the generator."""
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-
-        model = outlines.from_transformers(
-            AutoModelForCausalLM.from_pretrained(self.model_name),
-            AutoTokenizer.from_pretrained(self.model_name),
-        )
-
-        self.generator = outlines.Generator(model, JsonSchema(self.schema))
-
-    def __call__(self, *args, **kwargs):
-        """Call the function.
-
-        .. warning::
-
-           This currently does not support batching.
-
-        Parameters
-        ----------
-        args
-            Values to pass to the prompt template as positional arguments.
-        kwargs
-            Values to pass to the prompt template as keyword arguments.
-
-        """
-        if self.generator is None:
-            self.init_generator()
-
-        prompt = self.prompt_template(*args, **kwargs)
-        return self.generator(prompt)
-
-
-def download_from_github(short_path: str):
-    """Download the file in which the function is stored on GitHub."""
-    GITHUB_BASE_URL = "https://raw.githubusercontent.com"
-    BRANCH = "main"
-
-    path = short_path.split("/")
-    if len(path) < 3:
-        raise ValueError(
-            "Please provide a valid path in the form {USERNAME}/{REPO_NAME}/{PATH_TO_FILE}."
-        )
-    elif short_path[-3:] == ".py":
-        raise ValueError("Do not append the `.py` extension to the program name.")
-
-    username = path[0]
-    repo = path[1]
-    path_to_file = path[2:]
-
-    url = "/".join([GITHUB_BASE_URL, username, repo, BRANCH] + path_to_file) + ".py"
-    result = requests.get(url)
-
-    if result.status_code == 200:
-        return result.text
-    elif result.status_code == 404:
-        raise ValueError(
-            f"Program could not be found at {url}. Please make sure you entered the GitHub username, repository name and path to the program correctly."
-        )
-    else:
-        result.raise_for_status()
-
-
-def extract_function_from_file(content: str, function_name: str) -> Tuple[Callable]:
-    """Extract a function object from a downloaded file."""
-
-    spec = importlib.util.spec_from_loader(
-        "outlines_function", loader=None, origin="github"
-    )
-    if spec is not None:
-        module = importlib.util.module_from_spec(spec)
-        exec(content, module.__dict__)
-
-        try:
-            fn = getattr(module, function_name)
-        except AttributeError:
-            raise AttributeError(
-                "Could not find an `outlines.Function` instance in the remote file. Make sure that the path you specified is correct."
-            )
-
-        if not isinstance(fn, module.outlines.Function):
-            raise TypeError(
-                f"The `{function_name}` variable in the program must be an instance of `outlines.Function`"
-            )
-
-    return fn
diff --git a/tests/test_function.py b/tests/test_function.py
deleted file mode 100644
index 42e121c60..000000000
--- a/tests/test_function.py
+++ /dev/null
@@ -1,142 +0,0 @@
-from typing import Annotated
-
-import json
-import pytest
-import responses
-from pydantic import BaseModel, Field
-from requests.exceptions import HTTPError
-
-import outlines
-from outlines.function import Function, download_from_github, extract_function_from_file
-
-
-def test_function_basic():
-    with pytest.deprecated_call(match="The @prompt decorator"):
-
-        @outlines.prompt
-        def test_template(text: str):
-            """{{ text }}"""
-
-    class Foo(BaseModel):
-        id: Annotated[str, Field(max_length=5)]
-
-    fn = Function(test_template, Foo, "hf-internal-testing/tiny-random-GPTJForCausalLM")
-
-    assert fn.generator is None
-
-    result = fn("test")
-    Foo.parse_raw(result)
-    assert isinstance(json.loads(result), dict)
-    assert "id" in json.loads(result)
-
-
-def test_download_from_github_invalid():
-    with pytest.raises(ValueError, match="Please provide"):
-        download_from_github("outlines/program")
-
-    with pytest.raises(ValueError, match="Do not append"):
-        download_from_github("dottxt-ai/outlines/program.py")
-
-
-@responses.activate
-def test_download_from_github_success():
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
-        body="import outlines\n",
-        status=200,
-    )
-
-    file = download_from_github("dottxt-ai/outlines/program")
-    assert file == "import outlines\n"
-
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/foo/bar/program.py",
-        body="import outlines\n",
-        status=200,
-    )
-
-    file = download_from_github("dottxt-ai/outlines/foo/bar/program")
-    assert file == "import outlines\n"
-
-
-@responses.activate
-def test_download_from_github_error():
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/foo/bar/main/program.py",
-        json={"error": "not found"},
-        status=404,
-    )
-
-    with pytest.raises(ValueError, match="Program could not be found at"):
-        download_from_github("foo/bar/program")
-
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/foo/bar/main/program.py",
-        json={"error": "Internal Server Error"},
-        status=500,
-    )
-
-    with pytest.raises(HTTPError, match="500 Server Error"):
-        download_from_github("foo/bar/program")
-
-
-def test_extract_function_from_file():
-    content = """
-import outlines
-from pydantic import BaseModel
-
-model = "gpt2"
-
-
-@outlines.prompt
-def prompt():
-    '''Hello'''
-
-
-class User(BaseModel):
-    id: int
-    name: str
-
-
-function = outlines.Function(
-    prompt,
-    User,
-    "gpt2",
-)
-    """
-
-    with pytest.deprecated_call(match="The @prompt decorator"):
-        fn = extract_function_from_file(content, "function")
-        assert (
-            str(type(fn)) == "<class 'outlines.function.Function'>"
-        )  # because imported via `exec`
-
-
-def test_extract_function_from_file_no_function():
-    content = """
-import outlines
-from pydantic import BaseModel
-
-@outlines.prompt
-def prompt():
-    '''Hello'''
-
-
-class User(BaseModel):
-    id: int
-    name: str
-
-program = outlines.Function(
-    prompt,
-    User,
-    "gpt2",
-)
-    """
-
-    with pytest.deprecated_call(match="The @prompt decorator"):
-        with pytest.raises(AttributeError, match="Could not find"):
-            extract_function_from_file(content, "function")

From 5bd6592c58de37b292b8ecb3bf3d0a9822f3b3d9 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 17 Mar 2025 13:28:57 +0100
Subject: [PATCH 286/505] Add from_file class methods to CFG and JsonSchema

---
 outlines/types/dsl.py   | 32 ++++++++++++++++++++++++++++++++
 tests/types/test_dsl.py | 37 +++++++++++++++++++++++++++++++++++++
 2 files changed, 69 insertions(+)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 31495e63c..a04f2be64 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -211,6 +211,24 @@ def _display_node(self) -> str:
     def __repr__(self):
         return f"CFG(definition='{self.definition}')"
 
+    def __eq__(self, other):
+        if not isinstance(other, CFG):
+            return False
+        return self.definition == other.definition
+
+    @classmethod
+    def from_file(cls, path: str):
+        """Create a CFG instance from a file containing a CFG definition.
+
+        Parameters
+        ----------
+        path : str
+            The path to the file containing the CFG definition.
+        """
+        with open(path, "r") as f:
+            definition = f.read()
+        return cls(definition)
+
 
 @dataclass
 class FSM(Term):
@@ -268,6 +286,20 @@ def __eq__(self, other):
         except json.JSONDecodeError:
             return self.schema == other.schema
 
+    @classmethod
+    def from_file(cls, path: str):
+        """Create a JsonSchema instance from a .json file containing a JSON
+        schema.
+
+        Parameters
+        ----------
+        path : str
+            The path to the file containing the JSON schema.
+        """
+        with open(path, "r") as f:
+            schema = json.load(f)
+        return cls(schema)
+
 
 @dataclass
 class KleeneStar(Term):
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index f2973aa77..e92755f3f 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -1,6 +1,8 @@
 import datetime
 import json
+import os
 import sys
+import tempfile
 from dataclasses import dataclass
 from enum import Enum
 from typing import (
@@ -34,6 +36,7 @@
     String,
     Term,
     either,
+    CFG,
     _handle_dict,
     _handle_list,
     _handle_literal,
@@ -329,6 +332,40 @@ def test_dsl_display():
     )
 
 
+def test_dsl_cfg_from_file():
+    grammar_content = """
+    ?start: expression
+    ?expression: term (("+" | "-") term)*
+    ?term: factor (("*" | "/") factor)*
+    ?factor: NUMBER
+    """
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.txt', delete=True) as temp_file:
+        temp_file.write(grammar_content)
+        temp_file.flush()
+        temp_file_path = temp_file.name
+        cfg = CFG.from_file(temp_file_path)
+        assert cfg == CFG(grammar_content)
+
+
+def test_dsl_json_schema_from_file():
+    schema_content = """
+    {
+        "type": "object",
+        "properties": {
+            "name": {
+                "type": "string"
+            }
+        }
+    }
+    """
+    with tempfile.NamedTemporaryFile(mode='w', suffix='.json', delete=True) as temp_file:
+        temp_file.write(schema_content)
+        temp_file.flush()
+        temp_file_path = temp_file.name
+        schema = JsonSchema.from_file(temp_file_path)
+        assert schema == JsonSchema(schema_content)
+
+
 def test_dsl_python_types_to_terms():
     with pytest.raises(RecursionError):
         python_types_to_terms(None, 11)

From 32a46fe08a6cadc9457268507a0fbc396e770922 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 18 Mar 2025 10:31:33 +0100
Subject: [PATCH 287/505] Run coverage check on target branch instead of always
 main

---
 .github/workflows/tests.yml | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 9390f8769..c6e37d684 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -87,16 +87,30 @@ jobs:
         with:
           name: coverage-data
 
-      - name: Fetch main for coverage diff
+      - name: Determine base branch for comparison
+        id: base-branch
         run: |
-          git fetch --no-tags --prune origin main
+          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
+            # For PRs, use the remote tracking branch
+            COMPARE_BRANCH="origin/${{ github.base_ref }}"
+            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
+          else
+            # For push events, compare against the parent commit
+            COMPARE_BRANCH="${{ github.event.before }}"
+            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
+          fi
+          echo "Using $COMPARE_BRANCH for coverage comparison"
+
+      - name: Fetch base branch for coverage diff
+        run: |
+          git fetch --no-tags --prune origin ${COMPARE_BRANCH#origin/}
 
       - name: Combine coverage & fail if it's <100%.
         run: |
           python -m coverage combine
           python -m coverage html --skip-covered --skip-empty
           python -m coverage xml
-          diff-cover coverage.xml --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
+          diff-cover coverage.xml --compare-branch=$COMPARE_BRANCH --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
 
       - name: Upload HTML report if check failed.
         uses: actions/upload-artifact@v4

From c474aadaa0caf76627df01b7d6aa86e641c350a0 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 18 Mar 2025 18:49:20 +0100
Subject: [PATCH 288/505] Add caching to the function
 cached_create_states_mapping

---
 outlines/processors/guide.py   |  2 ++
 tests/processors/test_guide.py | 56 ++++++++++++++++++++++++++++++++++
 2 files changed, 58 insertions(+)

diff --git a/outlines/processors/guide.py b/outlines/processors/guide.py
index 6b97d7729..d46228fe9 100644
--- a/outlines/processors/guide.py
+++ b/outlines/processors/guide.py
@@ -15,6 +15,7 @@
 )
 
 from outlines import grammars
+from outlines.caching import cache
 from outlines.fsm.parsing import PartialLark, PartialParserState
 
 if TYPE_CHECKING:
@@ -72,6 +73,7 @@ def copy(self):
         return self
 
 
+@cache()
 def cached_create_states_mapping(regex_string, tokenizer, *args, **kwargs):
     return uncached_create_states_mapping(regex_string, tokenizer, *args, **kwargs)
 
diff --git a/tests/processors/test_guide.py b/tests/processors/test_guide.py
index 0f8cc9974..add142d26 100644
--- a/tests/processors/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -1,6 +1,23 @@
 import pytest
 
+import llama_cpp
+import transformers
+
+import outlines
 from outlines.processors.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
+from outlines import caching
+
+try:
+    import mlx_lm
+    HAS_MLX = True
+except ImportError:
+    HAS_MLX = False
+
+try:
+    import vllm
+    HAS_VLLM = True
+except ImportError:
+    HAS_VLLM = False
 
 
 def assert_expected_tensor_ids(tensor, ids):
@@ -181,6 +198,45 @@ def convert_token_to_string(self, token):
     assert fsm.is_final_state(state)
 
 
+def test_regex_guide_caching():
+    assert caching._caching_enabled
+
+    cache = caching.get_cache()
+    _, _ = cache.stats(enable=True, reset=True) # (hits, misses)
+
+    regex = r"[0-9]{3}"
+
+    models = [
+        outlines.from_transformers(
+            transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+            transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini")
+        ),
+        outlines.from_llamacpp(llama_cpp.Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        ))
+    ]
+    if HAS_MLX:
+        models.append(outlines.from_mlxlm(
+            *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+        ))
+    if HAS_VLLM:
+        models.append(outlines.from_vllm(vllm.LLM("erwanf/gpt2-mini")))
+
+    for i, model in enumerate(models):
+        # First call for each model should be a miss
+        RegexGuide.from_regex(regex, model.tokenizer)
+        expected_misses = i + 1
+        expected_hits = i
+        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
+
+        # Second call for each model
+        RegexGuide.from_regex(regex, model.tokenizer)
+        expected_misses = i + 1
+        expected_hits = i + 1
+        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
+
+
 def test_cfg():
     class MockTokenizer:
         vocabulary = {"{": 1, "}": 2, "[": 3, "]": 4, "eos": 5}

From ffe1b9018d5b26a11d2fefb29743ec5f19272b9e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 24 Mar 2025 10:24:23 +0100
Subject: [PATCH 289/505] Create the TensorAdapter classes

---
 .../processors/tensor_adapters/__init__.py    |  24 ++
 outlines/processors/tensor_adapters/base.py   |  72 +++++
 outlines/processors/tensor_adapters/jax.py    |  48 +++
 outlines/processors/tensor_adapters/mlx.py    |  50 ++++
 outlines/processors/tensor_adapters/numpy.py  |  48 +++
 .../processors/tensor_adapters/tensorflow.py  |  48 +++
 outlines/processors/tensor_adapters/torch.py  |  46 +++
 tests/processors/test_tensor_adapters.py      | 276 ++++++++++++++++++
 8 files changed, 612 insertions(+)
 create mode 100644 outlines/processors/tensor_adapters/__init__.py
 create mode 100644 outlines/processors/tensor_adapters/base.py
 create mode 100644 outlines/processors/tensor_adapters/jax.py
 create mode 100644 outlines/processors/tensor_adapters/mlx.py
 create mode 100644 outlines/processors/tensor_adapters/numpy.py
 create mode 100644 outlines/processors/tensor_adapters/tensorflow.py
 create mode 100644 outlines/processors/tensor_adapters/torch.py
 create mode 100644 tests/processors/test_tensor_adapters.py

diff --git a/outlines/processors/tensor_adapters/__init__.py b/outlines/processors/tensor_adapters/__init__.py
new file mode 100644
index 000000000..a61fa1ea5
--- /dev/null
+++ b/outlines/processors/tensor_adapters/__init__.py
@@ -0,0 +1,24 @@
+from typing import Union
+
+from .jax import JAXTensorAdapter
+from .mlx import MLXTensorAdapter
+from .numpy import NumpyTensorAdapter
+from .tensorflow import TensorFlowTensorAdapter
+from .torch import TorchTensorAdapter
+
+
+tensor_adapters = {
+    "jax": JAXTensorAdapter,
+    "mlx": MLXTensorAdapter,
+    "numpy": NumpyTensorAdapter,
+    "torch": TorchTensorAdapter,
+    "tensorflow": TensorFlowTensorAdapter,
+}
+
+TensorAdapterImplementation = Union[
+    JAXTensorAdapter,
+    MLXTensorAdapter,
+    NumpyTensorAdapter,
+    TorchTensorAdapter,
+    TensorFlowTensorAdapter,
+]
diff --git a/outlines/processors/tensor_adapters/base.py b/outlines/processors/tensor_adapters/base.py
new file mode 100644
index 000000000..4ea7b4459
--- /dev/null
+++ b/outlines/processors/tensor_adapters/base.py
@@ -0,0 +1,72 @@
+from abc import ABC, abstractmethod
+
+
+class TensorAdapter(ABC):
+    """Abstract base class for tensor adapters.
+
+    This class defines the interface for tensor adapters that are used to
+    manipulate tensors in different libraries. Concrete implementations of
+    this class should provide specific implementations for each method as
+    well as providing a library_name attribute.
+    """
+    library_name: str
+
+    @abstractmethod
+    def shape(self, tensor):
+        """Get the shape of the tensor"""
+        ...
+
+    @abstractmethod
+    def unsqueeze(self, tensor):
+        """Add a dimension to the tensor at the specified position"""
+        ...
+
+    @abstractmethod
+    def squeeze(self, tensor):
+        """Remove a dimension from the tensor at the specified position"""
+        ...
+
+    @abstractmethod
+    def to_list(self, tensor):
+        """Convert the tensor to a list"""
+        ...
+
+    @abstractmethod
+    def to_scalar(self, tensor):
+        """Convert the tensor to a scalar"""
+        ...
+
+    @abstractmethod
+    def full_like(self, tensor, fill_value):
+        """Create a tensor with the same shape as the input tensor filled with a scalar value"""
+        ...
+
+    @abstractmethod
+    def concatenate(self, tensors):
+        """Concatenate a list of tensors along a specified dimension"""
+        ...
+
+    @abstractmethod
+    def get_device(self, tensor):
+        """Get the device of the tensor"""
+        ...
+
+    @abstractmethod
+    def to_device(self, tensor, device):
+        """Move the tensor to a specified device"""
+        ...
+
+    @abstractmethod
+    def boolean_ones_like(self, tensor):
+        """Create a boolean ones tensor with the same shape as the input tensor"""
+        ...
+
+    @abstractmethod
+    def apply_mask(self, tensor, mask, value):
+        """Fill the elements of the tensor where the mask is True with the specified value"""
+        ...
+
+    @abstractmethod
+    def argsort_descending(self, tensor):
+        """Return the indices that would sort the tensor in descending order"""
+        ...
diff --git a/outlines/processors/tensor_adapters/jax.py b/outlines/processors/tensor_adapters/jax.py
new file mode 100644
index 000000000..73b4539a4
--- /dev/null
+++ b/outlines/processors/tensor_adapters/jax.py
@@ -0,0 +1,48 @@
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class JAXTensorAdapter(TensorAdapter):
+    library_name = "jax"
+
+    def __init__(self):
+        import jax
+
+        self.jax = jax
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.jax.numpy.expand_dims(tensor, axis=0)
+
+    def squeeze(self, tensor):
+        return self.jax.numpy.squeeze(tensor, axis=0)
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        return self.jax.numpy.full_like(tensor, fill_value)
+
+    def concatenate(self, tensors):
+        return self.jax.numpy.concatenate(tensors, axis=0)
+
+    def to_device(self, tensor, device):
+        return tensor
+
+    def get_device(self, tensor):
+        return None
+
+    def boolean_ones_like(self, tensor):
+        return self.jax.numpy.ones_like(tensor, dtype=bool)
+
+    def apply_mask(self, tensor, mask, value):
+        result = tensor.copy()
+        result = result.at[mask].set(value)
+        return result
+
+    def argsort_descending(self, tensor):
+        return self.jax.numpy.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/mlx.py b/outlines/processors/tensor_adapters/mlx.py
new file mode 100644
index 000000000..47811e183
--- /dev/null
+++ b/outlines/processors/tensor_adapters/mlx.py
@@ -0,0 +1,50 @@
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class MLXTensorAdapter(TensorAdapter):
+    library_name = "mlx"
+
+    def __init__(self):
+        import mlx.core
+
+        self.mlx = mlx.core
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.mlx.expand_dims(tensor, 0)
+
+    def squeeze(self, tensor):
+        if tensor.shape[0] == 1:
+            return tensor[0]
+        return tensor
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        return self.mlx.full(tensor.shape, fill_value, dtype=tensor.dtype)
+
+    def concatenate(self, tensors):
+        return self.mlx.concat(tensors, axis=0)
+
+    def get_device(self, tensor):
+        return None
+
+    def to_device(self, tensor, device):
+        return tensor
+
+    def boolean_ones_like(self, tensor):
+        return self.mlx.ones(tensor.shape, dtype=self.mlx.bool_)
+
+    def apply_mask(self, tensor, mask, value):
+        result = tensor.astype(tensor.dtype)
+        result = self.mlx.where(mask, self.mlx.array(value), result)
+        return result
+
+    def argsort_descending(self, tensor):
+        return self.mlx.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/numpy.py b/outlines/processors/tensor_adapters/numpy.py
new file mode 100644
index 000000000..9bc2fb159
--- /dev/null
+++ b/outlines/processors/tensor_adapters/numpy.py
@@ -0,0 +1,48 @@
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class NumpyTensorAdapter(TensorAdapter):
+    library_name = "numpy"
+
+    def __init__(self):
+        import numpy
+
+        self.numpy = numpy
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.numpy.expand_dims(tensor, axis=0)
+
+    def squeeze(self, tensor):
+        return self.numpy.squeeze(tensor, axis=0)
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        return self.numpy.full_like(tensor, fill_value)
+
+    def concatenate(self, tensors):
+        return self.numpy.concatenate(tensors, axis=0)
+
+    def get_device(self, tensor):
+        return None
+
+    def to_device(self, tensor, device):
+        return tensor
+
+    def boolean_ones_like(self, tensor):
+        return self.numpy.ones_like(tensor, dtype=bool)
+
+    def apply_mask(self, tensor, mask, value):
+        result = tensor.copy()
+        result[mask] = value
+        return result
+
+    def argsort_descending(self, tensor):
+        return self.numpy.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/tensorflow.py b/outlines/processors/tensor_adapters/tensorflow.py
new file mode 100644
index 000000000..4d2f93a6b
--- /dev/null
+++ b/outlines/processors/tensor_adapters/tensorflow.py
@@ -0,0 +1,48 @@
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class TensorFlowTensorAdapter(TensorAdapter):
+    library_name = "tensorflow"
+
+    def __init__(self):
+        import tensorflow as tf
+
+        self.tf = tf
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return self.tf.expand_dims(tensor, axis=0)
+
+    def squeeze(self, tensor):
+        return self.tf.squeeze(tensor, axis=0)
+
+    def to_list(self, tensor):
+        return tensor.numpy().tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.numpy().item()
+
+    def full_like(self, tensor, fill_value):
+        return self.tf.fill(self.tf.shape(tensor), fill_value)
+
+    def concatenate(self, tensors):
+        return self.tf.concat(tensors, axis=0)
+
+    def get_device(self, tensor):
+        return tensor.device
+
+    def to_device(self, tensor, device):
+        # TensorFlow handles device placement differently
+        with self.tf.device(device):
+            return self.tf.identity(tensor)
+
+    def boolean_ones_like(self, tensor):
+        return self.tf.ones_like(tensor, dtype=self.tf.bool)
+
+    def apply_mask(self, tensor, mask, value):
+        return self.tf.where(mask, self.tf.constant(value, dtype=tensor.dtype), tensor)
+
+    def argsort_descending(self, tensor):
+        return self.tf.argsort(tensor, direction='DESCENDING')
diff --git a/outlines/processors/tensor_adapters/torch.py b/outlines/processors/tensor_adapters/torch.py
new file mode 100644
index 000000000..6570b2f5f
--- /dev/null
+++ b/outlines/processors/tensor_adapters/torch.py
@@ -0,0 +1,46 @@
+from outlines.processors.tensor_adapters.base import TensorAdapter
+
+
+class TorchTensorAdapter(TensorAdapter):
+    library_name = "torch"
+
+    def __init__(self):
+        import torch
+
+        self.torch = torch
+
+    def shape(self, tensor):
+        return tensor.shape
+
+    def unsqueeze(self, tensor):
+        return tensor.unsqueeze(0)
+
+    def squeeze(self, tensor):
+        return tensor.squeeze(0)
+
+    def to_list(self, tensor):
+        return tensor.tolist()
+
+    def to_scalar(self, tensor):
+        return tensor.item()
+
+    def full_like(self, tensor, fill_value):
+        return self.torch.full_like(tensor, fill_value)
+
+    def concatenate(self, tensors):
+        return self.torch.cat(tensors, dim=0)
+
+    def get_device(self, tensor):
+        return tensor.device
+
+    def to_device(self, tensor, device):
+        return tensor.to(device)
+
+    def boolean_ones_like(self, tensor):
+        return self.torch.ones_like(tensor, dtype=self.torch.bool)
+
+    def apply_mask(self, tensor, mask, value):
+        return self.torch.masked_fill(tensor, mask, value)
+
+    def argsort_descending(self, tensor):
+        return self.torch.argsort(tensor, descending=True)
diff --git a/tests/processors/test_tensor_adapters.py b/tests/processors/test_tensor_adapters.py
new file mode 100644
index 000000000..4a5bcc1e1
--- /dev/null
+++ b/tests/processors/test_tensor_adapters.py
@@ -0,0 +1,276 @@
+import pytest
+from pytest import mark
+
+import jax
+import jax.numpy as jnp
+import numpy as np
+import tensorflow as tf
+import torch
+
+from outlines.processors.tensor_adapters import (
+    NumpyTensorAdapter,
+    TorchTensorAdapter,
+    MLXTensorAdapter,
+    JAXTensorAdapter,
+    TensorFlowTensorAdapter,
+)
+
+try:
+    import mlx_lm
+    import mlx.core as mx
+
+    HAS_MLX = mx.metal.is_available()
+except ImportError:
+    HAS_MLX = False
+
+
+adapters = {
+    "jax": JAXTensorAdapter(),
+    "numpy": NumpyTensorAdapter(),
+    "tensorflow": TensorFlowTensorAdapter(),
+    "torch": TorchTensorAdapter(),
+}
+if HAS_MLX:
+    adapters["mlx"] = MLXTensorAdapter()
+
+frameworks = ["jax", "numpy", "tensorflow", "torch", "mlx"]
+
+def create_tensor(framework, shape, dtype=None):
+    if framework == "torch":
+        return torch.randn(*shape)
+    elif framework == "numpy":
+        return np.random.randn(*shape)
+    elif framework == "jax":
+        key = jax.random.PRNGKey(0)
+        return jax.random.normal(key, shape=shape)
+    elif framework == "tensorflow":
+        return tf.random.normal(shape)
+    elif framework == "mlx":
+        if not HAS_MLX:
+            pytest.skip("MLX not available")
+        return mx.random.normal(shape)
+
+def compare_tensors(framework, tensor1, tensor2):
+    if framework == "torch":
+        return torch.allclose(tensor1, tensor2)
+    elif framework == "numpy":
+        return np.array_equal(tensor1, tensor2)
+    elif framework == "jax":
+        return jax.numpy.array_equal(tensor1, tensor2)
+    elif framework == "tensorflow":
+        return tf.reduce_all(tf.equal(tensor1, tensor2))
+    elif framework == "mlx":
+        if not HAS_MLX:
+            pytest.skip("MLX not available")
+        return mx.array_equal(tensor1, tensor2)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_shape(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (2,))
+    result_1d = adapters[framework].shape(tensor_1d)
+    assert len(result_1d) == 1
+    assert result_1d[0] == 2
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (2, 3))
+    result_2d = adapters[framework].shape(tensor_2d)
+    assert len(result_2d) == 2
+    assert result_2d[0] == 2
+    assert result_2d[1] == 3
+
+    # 3d tensor
+    tensor_3d = create_tensor(framework, (2, 2, 3))
+    result_3d = adapters[framework].shape(tensor_3d)
+    assert len(result_3d) == 3
+    assert result_3d[0] == 2
+    assert result_3d[1] == 2
+    assert result_3d[2] == 3
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_unsqueeze(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (2,))
+    result_1d = adapters[framework].unsqueeze(tensor_1d)
+    assert result_1d.shape == (1, 2)
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (2, 3))
+    result_2d = adapters[framework].unsqueeze(tensor_2d)
+    assert result_2d.shape == (1, 2, 3)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_squeeze(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (1,))
+    result_1d = adapters[framework].squeeze(tensor_1d)
+    with pytest.raises(TypeError):
+        len(result_1d)
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (1, 2))
+    result_2d = adapters[framework].squeeze(tensor_2d)
+    assert result_2d.shape == (2,)
+
+    # 3d tensor
+    tensor_3d = create_tensor(framework, (1, 2, 3))
+    result_3d = adapters[framework].squeeze(tensor_3d)
+    assert result_3d.shape == (2, 3)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_to_list(framework):
+    # 1d tensor
+    tensor_1d = create_tensor(framework, (2,))
+    result_1d = adapters[framework].to_list(tensor_1d)
+    assert isinstance(result_1d, list)
+    assert len(result_1d) == 2
+
+    # 2d tensor
+    tensor_2d = create_tensor(framework, (2, 3))
+    result_2d = adapters[framework].to_list(tensor_2d)
+    assert isinstance(result_2d, list)
+    assert len(result_2d) == 2
+    assert len(result_2d[0]) == 3
+    assert len(result_2d[1]) == 3
+
+    # 3d tensor
+    tensor_3d = create_tensor(framework, (2, 2, 3))
+    result_3d = adapters[framework].to_list(tensor_3d)
+    assert isinstance(result_3d, list)
+    assert len(result_3d) == 2
+    assert len(result_3d[0]) == 2
+    assert len(result_3d[1]) == 2
+    assert len(result_3d[0][0]) == 3
+    assert len(result_3d[0][1]) == 3
+    assert len(result_3d[1][0]) == 3
+    assert len(result_3d[1][1]) == 3
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_to_scalar(framework):
+    # multi-elements tensor, should raise an error
+    tensor_multi = create_tensor(framework, (2, 3))
+    if framework == "torch":
+        with pytest.raises(RuntimeError):
+            adapters[framework].to_scalar(tensor_multi)
+    else:
+        with pytest.raises(ValueError):
+            adapters[framework].to_scalar(tensor_multi)
+
+    # single-element tensor
+    tensor_single = create_tensor(framework, (1, 1))
+    scalar = adapters[framework].to_scalar(tensor_single)
+    assert isinstance(scalar, float)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_full_like(framework):
+    tensor = create_tensor(framework, (2, 3))
+    result = adapters[framework].full_like(tensor, 0)
+    assert result.shape == (2, 3)
+    for i in range(2):
+        for j in range(3):
+            assert result[i, j] == 0
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_concatenate(framework):
+    # 1d tensors
+    tensor1 = create_tensor(framework, (2,))
+    tensor2 = create_tensor(framework, (2,))
+    result = adapters[framework].concatenate([tensor1, tensor2])
+    assert result.shape == (4,)
+    assert result[0] == tensor1[0]
+    assert result[1] == tensor1[1]
+    assert result[2] == tensor2[0]
+    assert result[3] == tensor2[1]
+
+    # 2d tensors
+    tensor1 = create_tensor(framework, (2, 3))
+    tensor2 = create_tensor(framework, (2, 3))
+    result = adapters[framework].concatenate([tensor1, tensor2])
+    assert result.shape == (4, 3)
+    for i in range(2):
+        for j in range(3):
+            assert result[i, j] == tensor1[i, j]
+            assert result[i + 2, j] == tensor2[i, j]
+
+    # 3d tensors
+    tensor1 = create_tensor(framework, (2, 2, 3))
+    tensor2 = create_tensor(framework, (2, 2, 3))
+    result = adapters[framework].concatenate([tensor1, tensor2])
+    assert result.shape == (4, 2, 3)
+    for i in range(2):
+        for j in range(2):
+            for k in range(3):
+                assert result[i, j, k] == tensor1[i, j, k]
+                assert result[i + 2, j, k] == tensor2[i, j, k]
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_get_to_device(framework):
+    tensor = create_tensor(framework, (2, 3))
+    device = adapters[framework].get_device(tensor)
+    device_tensor = adapters[framework].to_device(tensor, device)
+
+    if framework == "torch":
+        assert isinstance(device_tensor.device.type, str)
+        assert compare_tensors(framework, device_tensor, tensor)
+    else:
+        assert compare_tensors(framework, device_tensor, tensor)
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_boolean_ones_like(framework):
+    tensor = create_tensor(framework, (2, 3))
+    ones = adapters[framework].boolean_ones_like(tensor)
+
+    assert ones.shape == (2, 3)
+    for i in range(2):
+        for j in range(3):
+            assert ones[i, j]
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_apply_mask(framework):
+    tensor = create_tensor(framework, (2, 3))
+
+    if framework == "torch":
+        mask = torch.randn(2, 3) > 0
+    elif framework == "numpy":
+        mask = np.random.randn(2, 3) > 0
+    elif framework == "jax":
+        key = jax.random.PRNGKey(0)
+        mask = jax.random.normal(key, shape=(2, 3)) > 0
+    elif framework == "tensorflow":
+        mask = tf.random.normal((2, 3)) > 0
+    elif framework == "mlx":
+        if not HAS_MLX:
+            pytest.skip("MLX not available")
+        mask = mx.random.normal((2, 3)) > 0
+
+    masked = adapters[framework].apply_mask(tensor, mask, float("-inf"))
+
+    assert masked.shape == (2, 3)
+    for i in range(2):
+        for j in range(3):
+            if mask[i, j]:
+                assert masked[i, j] == float("-inf")
+            else:
+                assert masked[i, j] == tensor[i, j]
+
+
+@pytest.mark.parametrize("framework", frameworks)
+def test_tensor_adapter_argsort_descending(framework):
+    tensor = create_tensor(framework, (2, 3))
+    indices = adapters[framework].argsort_descending(tensor)
+
+    assert indices.shape == (2, 3)
+    for i in range(2):
+        sorted_values = [tensor[i][idx] for idx in indices[i]]
+        for j in range(len(sorted_values) - 1):
+            assert sorted_values[j] >= sorted_values[j + 1]

From d1c143c219c88fcb8ace7fd38db4c67afc581a22 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 24 Mar 2025 10:25:14 +0100
Subject: [PATCH 290/505] Modify the logits processors to use the tensor
 adapters

---
 outlines/processors/base_logits_processor.py | 195 +++++--------------
 outlines/processors/structured.py            | 116 +++++------
 tests/processors/test_base_processor.py      | 155 ++++++++-------
 tests/processors/test_guide.py               |   3 +-
 tests/processors/test_structured.py          | 134 +++++++++++++
 5 files changed, 334 insertions(+), 269 deletions(-)
 create mode 100644 tests/processors/test_structured.py

diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index d9549d780..ba4df1036 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -1,169 +1,80 @@
 from abc import abstractmethod
-from typing import TYPE_CHECKING, List, Protocol, Type, Union, runtime_checkable
+from typing import TypeVar
 
-import numpy as np
-import torch
-from numpy.typing import NDArray
+from outlines.processors.tensor_adapters import tensor_adapters, TensorAdapterImplementation
 
-if TYPE_CHECKING:
-    import mlx.core as mx
+TensorType = TypeVar('TensorType')
 
 
-Array = Union[NDArray, torch.Tensor, List, "mx.array"]
+class OutlinesLogitsProcessor:
+    """Base class for logits processors.
 
-
-def is_mlx_array_type(array_type):
-    try:
-        import mlx.core as mx
-    except ImportError:
-        return False
-    return issubclass(array_type, mx.array)
-
-
-def is_jax_array_type(array_type):
-    try:
-        import jaxlib
-    except ImportError:
-        return False
-    return issubclass(array_type, jaxlib.xla_extension.ArrayImpl) or isinstance(
-        array_type, jaxlib.xla_extension.ArrayImpl
-    )
-
-
-@runtime_checkable
-class OutlinesLogitsProcessor(Protocol):
+    This class standardizes the shape of input_ids and logits received and
+    returned by the logits processor. At initialization, it creates a
+    tensor_adapter attribute which is used to manipulate the tensors
+    using the appropriate library for the model (numpy, torch...).
     """
-    Base class for logits processors which normalizes types of logits:
-    - ndarray (used by llama-cpp-python), converted to torch.Tensor
-    - mlx.core.array (used by mlx-lm), converted to torch.Tensor
-    - torch.Tensor (used by everything else)
-
-    Normalization of types and conversion to torch.Tensor
-    doesn't move memory, it just casts the type.
+    tensor_adapter: TensorAdapterImplementation
 
-    Normalizing the types allows all logits processors inheriting from this class
-    to implement a single method for all the business logit: `process_logits()`
-    """
+    def __init__(self, tensor_library_name: str):
+        tensor_adapter_class = tensor_adapters.get(tensor_library_name)
+        if tensor_adapter_class is None:
+            raise NotImplementedError(f"Library {tensor_library_name} is not available")
+        self.tensor_adapter = tensor_adapter_class()  # type: ignore
 
     @abstractmethod
     def process_logits(
-        self, input_ids: List[List[int]], logits: torch.Tensor
-    ) -> torch.Tensor:
-        """
-        input_ids and logits are always 2D tensors for handling a batch of sequences.
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Main method to implement for logits processors.
 
-        - input_ids -> List[List[tokens]]
-        - logits -> 2D_Tensor[logit floats]
+        This method applies a mask on the logits to bias the generation.
+        It is called by the __call__ method that standardizes the shape of
+        input_ids and logits to ensure they are 2D tensors.
 
         Important to keep in mind when designing universal logits processors
-        - logits processors are only used once and never re-applied for a new sequence generator
-        - Some models only pass output_ids, some models such as llamacpp and transformers prefix with input_ids
-        - Some sampling methods, such as beam search, result in unstable sequence ordering in models like vLLM
+        - logits processors are only used once and never re-applied for a new
+        sequence generator
+        - Some models only pass output_ids, some models such as llamacpp and
+        transformers prefix with input_ids
+        - Some sampling methods, such as beam search, result in unstable
+        sequence ordering in models like vLLM
         """
-        pass
+        ...
 
-    @torch.no_grad()
     def __call__(
-        self,
-        input_ids: Array,
-        logits: Array,
-    ) -> Array:
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Entrypoint for logits processors, this is the method that is called
+        by the model.
+
+        Because different models use different libraries, we need to standardize
+        the shape of input_ids and logits to ensure they are 2D tensors before
+        calling the unified process_logits method. After processing, the logits
+        are cast back to the original array library type before being returned.
         """
-        Apply logits processor
-
-        1) Unify type
-        - convert input_ids: either ndarray, mlx array, List[int], or Tensor -> List[List[int]]
-        - convert logits: either ndarray, mlx array, or Tensor -> 2D float Tensor
-        2) Unify shape, ensure logits and input_ids are 2D
-        3) Call self.process_logits() to perform business logic
-        4) Cast logits back to original array library type
-        """
-        # ensure logits are torch Tensors
-        torch_logits = self._to_torch(logits)
-        input_ids = self._to_torch(input_ids)
-
         # if input_ids is 1D and logits is 2D with a single sequence,
         # reshape input_ids to 2D (needed for mlx-lm)
         if (
-            len(input_ids.shape) == 1
-            and len(torch_logits.shape) == 2
-            and torch_logits.shape[0] == 1
+            len(self.tensor_adapter.shape(input_ids)) == 1
+            and len(self.tensor_adapter.shape(logits)) == 2
+            and self.tensor_adapter.shape(logits)[0] == 1
         ):
-            input_ids = input_ids.unsqueeze(0)
+            input_ids = self.tensor_adapter.unsqueeze(input_ids)
 
-        assert torch_logits.shape[:-1] == input_ids.shape[:-1]
+        assert self.tensor_adapter.shape(logits)[:-1] == self.tensor_adapter.shape(input_ids)[:-1]
 
         # Guarantee passed as 2D Tensors, then covert back to original (1D or 2D) shape
-        if len(torch_logits.shape) == 2:
-            processed_logits = self.process_logits(input_ids, torch_logits)
-        elif len(torch_logits.shape) == 1:
-            processed_logits = self.process_logits(
-                input_ids.unsqueeze(0), torch_logits.unsqueeze(0)
-            ).squeeze(0)
-
-        # return logits as passed array type
-        return self._from_torch(processed_logits, type(logits))
-
-    @staticmethod
-    def _to_torch(tensor_like: Array) -> torch.Tensor:
-        """Convert various types to torch.Tensor."""
-        if isinstance(tensor_like, torch.Tensor):
-            return tensor_like
-
-        elif isinstance(tensor_like, np.ndarray):
-            return torch.from_numpy(tensor_like)
-
-        elif isinstance(tensor_like, (list, tuple)):
-            return torch.tensor(tensor_like)
-
-        elif is_mlx_array_type(type(tensor_like)):
-            import mlx.core as mx
-
-            # https://ml-explore.github.io/mlx/build/html/usage/numpy.html#pytorch
-            if tensor_like.dtype == mx.bfloat16:
-                tensor_like = tensor_like.astype(mx.float32)
-            return torch.from_dlpack(np.array(tensor_like, copy=False))
-
-        elif is_jax_array_type(type(tensor_like)):
-            import jax
-
-            torch_tensor = torch.from_dlpack(jax.dlpack.to_dlpack(tensor_like))
-            return torch_tensor
-
-        else:
-            raise TypeError(
-                "LogitsProcessor must be called with either np.NDArray, "
-                "torch.Tensor, list, or mlx.core.array typed logits. "
-                f"Logits type: `{type(tensor_like)}`"
+        if len(self.tensor_adapter.shape(logits)) == 2:
+            processed_logits = self.process_logits(input_ids, logits)
+        elif len(self.tensor_adapter.shape(logits)) == 1:
+            processed_logits = self.tensor_adapter.squeeze(
+                self.process_logits(
+                    self.tensor_adapter.unsqueeze(input_ids),
+                    self.tensor_adapter.unsqueeze(logits),
+                ),
             )
-
-    @staticmethod
-    def _from_torch(tensor: torch.Tensor, target_type: Type) -> Array:
-        """Convert torch.Tensor to the specified target type."""
-        if target_type == torch.Tensor:
-            return tensor
-
-        elif target_type == np.ndarray:
-            return tensor.detach().numpy()
-
-        elif target_type is list:
-            return tensor.detach().tolist()
-
-        elif target_type is tuple:
-            return tuple(tensor.detach().tolist())
-
-        elif is_mlx_array_type(target_type):
-            import mlx.core as mx
-
-            # numpy doesn't support bfloat16, mlx doesn't support direct conversion from torch
-            return mx.array(tensor.float().numpy())
-
-        elif is_jax_array_type(target_type):
-            import jax
-
-            return jax.dlpack.from_dlpack(tensor)
-
         else:
-            raise TypeError(
-                f"Failed to convert torch tensors to target_type `{target_type}`"
-            )
+            raise ValueError(f"Logits shape {self.tensor_adapter.shape(logits)} is not supported")
+
+        return processed_logits
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 806172eaa..965806621 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -25,16 +25,15 @@
 """
 
 import math
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union, cast
 
-import torch
 from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel
 
 from outlines.processors.guide import CFGGuide, Guide, RegexGuide
 from outlines.types import JsonSchema
 
-from .base_logits_processor import OutlinesLogitsProcessor
+from .base_logits_processor import OutlinesLogitsProcessor, TensorType
 
 if TYPE_CHECKING:
     from outlines.models.tokenizer import Tokenizer
@@ -56,7 +55,7 @@ class GuideLogitsProcessor(OutlinesLogitsProcessor):
     _guide_states: Dict[int, Any]
     _seq_start_idx: Optional[int]
 
-    def __init__(self, tokenizer: "Tokenizer", guide: Guide):
+    def __init__(self, tokenizer: "Tokenizer", guide: Guide, tensor_library_name: str):
         """A Guide-based logits processor.
 
         Parameters
@@ -65,15 +64,18 @@ def __init__(self, tokenizer: "Tokenizer", guide: Guide):
             The tokenizer used to convert tokens to ids.
         guide
             The `outlines.processors.guide. which is used to bias the logits.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
         """
+        super().__init__(tensor_library_name=tensor_library_name)
         self.tokenizer = tokenizer
         self.guide = guide
         self._guide_states = {hash(tuple([])): self.guide.initial_state}
         self._seq_start_idx = None
 
     def process_logits(
-        self, input_ids: torch.LongTensor, logits: torch.FloatTensor
-    ) -> torch.Tensor:
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
         """Use the Guide to bias the logits before sampling the next token.
 
         Parameters
@@ -85,21 +87,20 @@ def process_logits(
 
         Returns
         -------
-        torch.Tensor
-            The biased logits.
+        The biased logits.
         """
         if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0])
+            self._seq_start_idx = len(input_ids[0]) # type: ignore
 
         sequence_states: List[int] = []  # vector of states corresponding to `input_ids`
 
-        for seq_ids in input_ids:
+        for seq_ids in input_ids: # type: ignore
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(gen_ids.tolist()))
+            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
 
             if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
-                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
+                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
+                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])
@@ -110,63 +111,64 @@ def process_logits(
             allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
             allowed_tokens_batch.append(allowed_tokens)
             batch_indices.append(
-                torch.full_like(allowed_tokens, i)
+                self.tensor_adapter.full_like(allowed_tokens, i)
             )  # Store batch index for each allowed token
 
-        allowed_tokens_concat = torch.cat(allowed_tokens_batch).to(logits.device)
-        batch_indices_concat = torch.cat(batch_indices).to(logits.device)
-
-        mask = torch.ones_like(logits, dtype=torch.bool)
+        device = self.tensor_adapter.get_device(logits)
+        allowed_tokens_concat = self.tensor_adapter.to_device(
+            self.tensor_adapter.concatenate(allowed_tokens_batch),
+            device
+        )
+        batch_indices_concat = self.tensor_adapter.to_device(
+            self.tensor_adapter.concatenate(batch_indices),
+            device
+        )
+
+        mask = self.tensor_adapter.boolean_ones_like(logits)
         mask[batch_indices_concat, allowed_tokens_concat] = False
-        logits.masked_fill_(mask, float("-inf"))
+        logits = self.tensor_adapter.apply_mask(logits, mask, float("-inf"))
 
         return logits
 
     def copy(self) -> "GuideLogitsProcessor":
         """Return a copy of the logits processor."""
-        return GuideLogitsProcessor(tokenizer=self.tokenizer, guide=self.guide.copy())
+        return GuideLogitsProcessor(
+            tokenizer=self.tokenizer,
+            guide=self.guide.copy(),
+            tensor_library_name=self.tensor_adapter.library_name
+        )
 
 
 class RegexLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a regular expression.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    guide
-        The `outlines.fsm.RegexGuide. which is used to bias the logits.
-    """
-
-    def __init__(self, regex_string: str, tokenizer: "Tokenizer"):
+    """Bias generation based on a regular expression."""
+    def __init__(
+        self,
+        regex_string: str,
+        tokenizer: "Tokenizer",
+        tensor_library_name: str,
+    ):
         """Compile the RegexGuide that drives the regex-guided generation.
 
         Parameters
         ----------
         regex_string
-            A string that represents a regular expression
+            A string that represents a regular expression.
         tokenizer
-            An Outlines tokenizer
+            An Outlines tokenizer.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
         """
         guide = RegexGuide.from_regex(regex_string, tokenizer)
-        super().__init__(tokenizer=tokenizer, guide=guide)
+        super().__init__(tokenizer=tokenizer, guide=guide, tensor_library_name=tensor_library_name)
 
 
 class JSONLogitsProcessor(RegexLogitsProcessor):
-    """Bias generation based on a JSON schema.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    guide
-        The `outlines.fsm.RegexGuide. which is used to bias the logits.
-    """
-
+    """Bias generation based on a JSON schema."""
     def __init__(
         self,
         schema: Union[dict, Type[BaseModel], str],
         tokenizer: "Tokenizer",
+        tensor_library_name: str,
         whitespace_pattern: Optional[str] = None,
     ):
         """Compile the Guide that drives the JSON-guided generation.
@@ -177,14 +179,16 @@ def __init__(
             A JSON schema that encodes the structure we want the model to generate.
         tokenizer
             The tokenizer used to convert tokens to ids.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
         whitespace_pattern
             Pattern to use for JSON syntactic whitespace (doesn't impact string
             literals). For example, to allow only a single space or newline with
-            `whitespace_pattern=r"[\n ]?"`
+            `whitespace_pattern=r"[\n ]?"`.
         """
         schema_str = JsonSchema(schema).schema
         regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(regex_string=regex_string, tokenizer=tokenizer)
+        super().__init__(regex_string=regex_string, tokenizer=tokenizer, tensor_library_name=tensor_library_name)
 
 
 class CFGLogitsProcessor(GuideLogitsProcessor):
@@ -200,7 +204,7 @@ class CFGLogitsProcessor(GuideLogitsProcessor):
 
     guide: CFGGuide
 
-    def __init__(self, cfg_str: str, tokenizer: "Tokenizer"):
+    def __init__(self, cfg_str: str, tokenizer: "Tokenizer", tensor_library_name: str):
         """Compile the CFGGuide that drives the CFG-guided generation.
 
         Parameters
@@ -211,11 +215,11 @@ def __init__(self, cfg_str: str, tokenizer: "Tokenizer"):
             The tokenizer used to convert tokens to ids.
         """
         cfg_guide = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
-        super().__init__(tokenizer=tokenizer, guide=cfg_guide)
+        super().__init__(tokenizer=tokenizer, guide=cfg_guide, tensor_library_name=tensor_library_name)
 
     def process_logits(
-        self, input_ids: torch.LongTensor, logits: torch.Tensor
-    ) -> torch.Tensor:
+        self, input_ids, logits
+    ):
         """Same behavior as GuideLogitsProcessor, but uses rejection sampling"""
         if self._seq_start_idx is None:
             self._seq_start_idx = len(input_ids[0])
@@ -224,20 +228,20 @@ def process_logits(
 
         for seq_ids in input_ids:
             gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(gen_ids.tolist()))
+            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
 
-            if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(gen_ids[:-1].tolist()))]
-                curr_state = self.guide.get_next_state(prev_state, gen_ids[-1].item())
+            if curr_state_key not in self._guide_states: # pragma: no cover
+                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
+                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
                 self._guide_states[curr_state_key] = curr_state
 
             sequence_states.append(self._guide_states[curr_state_key])
 
-        mask = torch.full_like(logits, -math.inf)
+        mask = self.tensor_adapter.full_like(logits, -math.inf)
         for i, guide_state in enumerate(sequence_states):
             first_legal_token = next(
                 self.guide.iter_valid_token_ids(
-                    guide_state, torch.argsort(logits[i], descending=True)
+                    guide_state, self.tensor_adapter.argsort_descending(logits[i])
                 )
             )
             mask[i, [first_legal_token]] = logits[i, [first_legal_token]]
diff --git a/tests/processors/test_base_processor.py b/tests/processors/test_base_processor.py
index e135571e3..3b729410a 100644
--- a/tests/processors/test_base_processor.py
+++ b/tests/processors/test_base_processor.py
@@ -7,81 +7,96 @@
 
 from outlines.processors.base_logits_processor import OutlinesLogitsProcessor
 
-arrays = {
-    "list": [[1.0, 2.0], [3.0, 4.0]],
-    "np": np.array([[1, 2], [3, 4]], dtype=np.float32),
-    "jax": jnp.array([[1, 2], [3, 4]], dtype=jnp.float32),
-    "torch": torch.tensor([[1, 2], [3, 4]], dtype=torch.float32),
-}
-
 try:
     import mlx.core as mx
-
-    arrays["mlx"] = mx.array([[1, 2], [3, 4]], dtype=mx.float32)
-    arrays["mlx_bfloat16"] = mx.array([[1, 2], [3, 4]], dtype=mx.bfloat16)
+    HAS_MLX = True
 except ImportError:
-    pass
+    HAS_MLX = False
 
-try:
-    import jax.numpy as jnp
 
-    arrays["jax"] = jnp.array([[1, 2], [3, 4]], dtype=jnp.float32)
-except ImportError:
-    pass
+libraries = ["numpy", "torch", "jax"]
+if HAS_MLX:
+    libraries.append("mlx")
 
+# we check the accepted shapes:
+# - both 1D
+# - both 2D
+# - input_ids 1D and logits 2D with a single sequence
+# we raise an error if the shapes are not accepted:
+# - input_ids 2D and logits 1D
+# - input_ids 1D and logits 2D, but with multiple sequences
+# - both 3D
+arrays = {
+    "numpy": [
+        (np.array([1, 2], dtype=np.float32), np.array([1, 2], dtype=np.int32), None),
+        (np.array([[1, 2], [3, 4]], dtype=np.float32), np.array([[1, 2], [3, 4]], dtype=np.int32), None),
+        (np.array([1, 2], dtype=np.float32), np.array([[1, 2]], dtype=np.int32), None),
+        (np.array([[1, 2]], dtype=np.float32), np.array([1, 2], dtype=np.int32), AssertionError),
+        (np.array([1, 2], dtype=np.float32), np.array([[1, 2], [3, 4]], dtype=np.int32), AssertionError),
+        (np.array([[[1, 2]]], dtype=np.float32), np.array([[[1, 2]]], dtype=np.int32), ValueError),
+    ],
+    "torch": [
+        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([1, 2], dtype=torch.int32), None),
+        (torch.tensor([[1, 2], [3, 4]], dtype=torch.float32), torch.tensor([[1, 2], [3, 4]], dtype=torch.int32), None),
+        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([[1, 2]], dtype=torch.int32), None),
+        (torch.tensor([[1, 2]], dtype=torch.float32), torch.tensor([1, 2], dtype=torch.int32), AssertionError),
+        (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([[1, 2], [3, 4]], dtype=torch.int32), AssertionError),
+        (torch.tensor([[[1, 2]]], dtype=torch.float32), torch.tensor([[[1, 2]]], dtype=torch.int32), ValueError),
+    ],
+    "jax": [
+        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), None),
+        (jnp.array([[1, 2], [3, 4]], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), None),
+        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2]], dtype=jnp.int32), None),
+        (jnp.array([[1, 2]], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), AssertionError),
+        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), AssertionError),
+        (jnp.array([[[1, 2]]], dtype=jnp.float32), jnp.array([[[1, 2]]], dtype=jnp.int32), ValueError),
+    ],
+}
+if HAS_MLX:
+    arrays["mlx"] = [
+        (mx.array([1, 2], dtype=mx.float32), mx.array([1, 2], dtype=mx.int32), None),
+        (mx.array([[1, 2], [3, 4]], dtype=mx.float32), mx.array([[1, 2], [3, 4]], dtype=mx.int32), None),
+        (mx.array([1, 2], dtype=mx.float32), mx.array([[1, 2]], dtype=mx.int32), None),
+        (mx.array([[1, 2]], dtype=mx.float32), mx.array([1, 2], dtype=mx.int32), AssertionError),
+        (mx.array([1, 2], dtype=mx.float32), mx.array([[1, 2], [3, 4]], dtype=mx.int32), AssertionError),
+        (mx.array([[[1, 2]]], dtype=mx.float32), mx.array([[[1, 2]]], dtype=mx.int32), ValueError),
+    ]
 
-# Mock implementation of the abstract class for testing
 class MockLogitsProcessor(OutlinesLogitsProcessor):
-    def process_logits(
-        self, input_ids: List[List[int]], logits: torch.Tensor
-    ) -> torch.Tensor:
-        # For testing purposes, let's just return logits multiplied by 2
-        return logits * 2
-
-
-@pytest.fixture
-def processor():
-    """Fixture for creating an instance of the MockLogitsProcessor."""
-    return MockLogitsProcessor()
-
-
-@pytest.mark.parametrize("array_type", arrays.keys())
-def test_to_torch(array_type, processor):
-    data = arrays[array_type]
-    torch_tensor = processor._to_torch(data)
-    assert isinstance(torch_tensor, torch.Tensor)
-    assert torch.allclose(
-        torch_tensor.cpu(), torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
-    )
-
-
-@pytest.mark.parametrize("array_type", arrays.keys())
-def test_from_torch(array_type, processor):
-    torch_tensor = torch.tensor([[1, 2], [3, 4]], dtype=torch.float32)
-    data = processor._from_torch(torch_tensor, type(arrays[array_type]))
-    assert isinstance(data, type(arrays[array_type]))
-    if array_type == "mlx_bfloat16":
-        # For bfloat16, we expect the output to be float32 due to the conversion
-        assert data.dtype == mx.float32
-        assert np.allclose(np.array(data), np.array([[1, 2], [3, 4]], dtype=np.float32))
-    else:
-        assert np.allclose(data, arrays[array_type])
-
-
-@pytest.mark.parametrize("array_type", arrays.keys())
-def test_call(array_type, processor):
-    input_ids = arrays[array_type]
-    logits = arrays[array_type]
-    processed_logits = processor(input_ids, logits)
-
-    assert isinstance(processed_logits, type(arrays[array_type]))
-
-    # Convert to numpy array for comparison, handling PyTorch tensors properly
-    if array_type == "torch":
-        expected = np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
-        actual = processed_logits.detach().cpu().numpy()
-        assert np.allclose(actual, expected)
-    else:
-        assert np.allclose(
-            np.array(processed_logits), np.array([[2.0, 4.0], [6.0, 8.0]], dtype=np.float32)
-        )
+    def process_logits(self, input_ids, logits):
+        # check that input_ids and logits received are 2D tensors
+        assert len(self.tensor_adapter.shape(input_ids)) == 2
+        assert len(self.tensor_adapter.shape(logits)) == 2
+        return logits
+
+
+@pytest.mark.parametrize("library", libraries)
+def test_base_logits_processor_init(library):
+    processor = MockLogitsProcessor(library)
+    assert processor.tensor_adapter is not None
+    with pytest.raises(NotImplementedError):
+        processor = MockLogitsProcessor("foo")
+
+
+@pytest.mark.parametrize("library", libraries)
+def test_base_logits_processor_call(library):
+    processor = MockLogitsProcessor(library)
+    input_values = arrays[library]
+    for input_value in input_values:
+        input_ids, logits, expected_error = input_value
+        if expected_error is not None:
+            with pytest.raises(expected_error):
+                processor(input_ids, logits)
+        else:
+            original_shape = processor.tensor_adapter.shape(logits)
+            processed_logits = processor(input_ids, logits)
+            # we check that the shape of logits is preserved
+            assert processor.tensor_adapter.shape(processed_logits) == original_shape
+
+
+@pytest.mark.parametrize("library", libraries)
+def test_base_logits_processor_init_library_name(library):
+    processor = MockLogitsProcessor(library)
+    assert processor.tensor_adapter is not None
+    with pytest.raises(NotImplementedError):
+        processor = MockLogitsProcessor("foo")
diff --git a/tests/processors/test_guide.py b/tests/processors/test_guide.py
index add142d26..2d46a2d2d 100644
--- a/tests/processors/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -202,6 +202,7 @@ def test_regex_guide_caching():
     assert caching._caching_enabled
 
     cache = caching.get_cache()
+    cache.clear()
     _, _ = cache.stats(enable=True, reset=True) # (hits, misses)
 
     regex = r"[0-9]{3}"
@@ -221,7 +222,7 @@ def test_regex_guide_caching():
             *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
         ))
     if HAS_VLLM:
-        models.append(outlines.from_vllm(vllm.LLM("erwanf/gpt2-mini")))
+        models.append(outlines.from_vllm(vllm.LLM("TinyLlama/TinyLlama-1.1B-Chat-v1.0")))
 
     for i, model in enumerate(models):
         # First call for each model should be a miss
diff --git a/tests/processors/test_structured.py b/tests/processors/test_structured.py
new file mode 100644
index 000000000..00007b3d3
--- /dev/null
+++ b/tests/processors/test_structured.py
@@ -0,0 +1,134 @@
+import pytest
+from typing import Dict
+
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM
+
+import outlines
+from outlines.processors.guide import RegexGuide
+from outlines.processors.structured import (
+    GuideLogitsProcessor,
+    RegexLogitsProcessor,
+    JSONLogitsProcessor,
+    CFGLogitsProcessor,
+)
+
+
+arithmetic_grammar = """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+
+@pytest.fixture
+def tokenizer():
+    model = outlines.from_transformers(
+        AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return model.tokenizer
+
+
+def test_structured_init_guide_logits_processor(tokenizer):
+    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
+    processor = GuideLogitsProcessor(
+        tokenizer=tokenizer,
+        guide=guide,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, GuideLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is guide
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_init_guide_logits_processor_copy(tokenizer):
+    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
+    processor = GuideLogitsProcessor(
+        tokenizer=tokenizer,
+        guide=guide,
+        tensor_library_name="torch",
+    )
+    processor_copy = processor.copy()
+    assert processor_copy is not processor
+    assert processor_copy.tokenizer is tokenizer
+    assert processor_copy.guide is guide
+    assert processor_copy.tensor_adapter.library_name == "torch"
+
+
+def test_structured_guide_logits_processor_call(tokenizer):
+    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
+    processor = GuideLogitsProcessor(
+        tokenizer=tokenizer,
+        guide=guide,
+        tensor_library_name="torch",
+    )
+    vocab_token_ids = list(tokenizer.vocabulary.values())
+    logits = torch.randn(1, len(vocab_token_ids))
+    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
+    output = processor(input_ids, logits)
+    assert output.shape == (1, len(vocab_token_ids))
+
+
+def test_structured_init_json_logits_processor(tokenizer):
+    processor = JSONLogitsProcessor(
+        schema={"type": "object", "properties": {"name": {"type": "string"}}},
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, JSONLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is not None
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_init_regex_logits_processor(tokenizer):
+    processor = RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, RegexLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is not None
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_init_cfg_logits_processor(tokenizer):
+    processor = CFGLogitsProcessor(
+        cfg_str=arithmetic_grammar,
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    assert isinstance(processor, CFGLogitsProcessor)
+    assert processor.tokenizer is tokenizer
+    assert processor.guide is not None
+    assert processor.tensor_adapter.library_name == "torch"
+
+
+def test_structured_cfg_logits_processor_call(tokenizer):
+    processor = CFGLogitsProcessor(
+        cfg_str=arithmetic_grammar,
+        tokenizer=tokenizer,
+        tensor_library_name="torch",
+    )
+    vocab_token_ids = list(tokenizer.vocabulary.values())
+    logits = torch.randn(1, len(vocab_token_ids))
+    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
+    output = processor(input_ids, logits)
+    assert output.shape == (1, len(vocab_token_ids))

From 9401569875c41d97648b53aa9831f4321474ae2d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 24 Mar 2025 13:36:01 +0100
Subject: [PATCH 291/505] Modify models and generators to implement the use of
 tensor adapters

---
 outlines/generator.py                        |  8 +++---
 outlines/models/base.py                      |  3 ++-
 outlines/models/llamacpp.py                  |  4 +--
 outlines/models/mlxlm.py                     |  1 +
 outlines/models/tokenizer.py                 | 12 +++++----
 outlines/models/transformers.py              | 25 ++++++++++++++++++
 outlines/models/vllm.py                      |  1 +
 outlines/serve/serve.py                      |  5 ++--
 pyproject.toml                               | 12 +++++++--
 tests/models/test_llamacpp.py                | 14 +++++++---
 tests/models/test_mlxlm.py                   | 24 +++++++++++------
 tests/models/test_transformers.py            | 27 +++++++++++++++++---
 tests/models/test_transformers_multimodal.py | 15 ++++++-----
 tests/models/test_vllm.py                    | 21 ++++++++++++---
 tests/test_generator.py                      | 20 +++++++++++++--
 15 files changed, 152 insertions(+), 40 deletions(-)

diff --git a/outlines/generator.py b/outlines/generator.py
index 0d9c86ea9..65842f4c9 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -66,15 +66,17 @@ def __init__(self, model, output_type: Optional[Any]):
             if isinstance(term, CFG):
                 cfg_string = term.definition
                 self.logits_processor = CFGLogitsProcessor(
-                    cfg_string, self.model.tokenizer
+                    cfg_string, self.model.tokenizer, self.model.tensor_library_name
                 )
             elif isinstance(term, FSM):
                 guide = RegexGuide.from_interegular_fsm(term.fsm, self.model.tokenizer)
-                self.logits_processor = GuideLogitsProcessor(tokenizer=self.model.tokenizer, guide=guide)
+                self.logits_processor = GuideLogitsProcessor(
+                    self.model.tokenizer, guide, self.model.tensor_library_name
+                )
             else:
                 regex_string = to_regex(term)
                 self.logits_processor = RegexLogitsProcessor(
-                    regex_string, self.model.tokenizer
+                    regex_string, self.model.tokenizer, self.model.tensor_library_name
                 )
 
     @classmethod
diff --git a/outlines/models/base.py b/outlines/models/base.py
index a70fba250..375f2549d 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -46,10 +46,11 @@ class Model(ABC):
     attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
     attribute are used in the `generate` method to format the input and output
     types received by the model.
+    Additionally, local models must define a `tensor_library_name` attribute.
 
     """
-
     type_adapter: ModelTypeAdapter
+    tensor_library_name: str
 
     def __call__(self, model_input, output_type=None, **inference_kwargs):
         """Call the model.
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index b761a926a..2f42ddca8 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,7 +1,7 @@
 import pickle
 import warnings
 from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Dict, Iterator, List, Set, Tuple, Union
+from typing import TYPE_CHECKING, Dict, Iterator, List, Optional, Set, Tuple, Union
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
@@ -140,13 +140,13 @@ def format_output_type(self, output_type):
 
 class LlamaCpp(Model):
     """Wraps a model provided by the `llama-cpp-python` library."""
+    tensor_library_name = "numpy"
 
     def __init__(self, model: "Llama"):
         from llama_cpp import Llama
 
         self.model = model
         self.tokenizer = LlamaCppTokenizer(self.model)
-        self.model_type = "local"
         self.type_adapter = LlamaCppTypeAdapter()
 
     def generate(self, model_input, output_type, **inference_kwargs):
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 2191dc20d..fda8f1a08 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -50,6 +50,7 @@ def format_output_type(self, output_type):
 
 class MLXLM(Model):
     """Represents an `mlx_lm` model."""
+    tensor_library_name = "mlx"
 
     def __init__(
         self,
diff --git a/outlines/models/tokenizer.py b/outlines/models/tokenizer.py
index 1a5708d85..5463c5eaa 100644
--- a/outlines/models/tokenizer.py
+++ b/outlines/models/tokenizer.py
@@ -1,7 +1,9 @@
-from typing import Dict, Hashable, List, Protocol, Set, Tuple, Union
+from typing import Dict, Hashable, List, Protocol, Set, Tuple, Union, TYPE_CHECKING
 
-import numpy as np
-from numpy.typing import NDArray
+
+if TYPE_CHECKING:
+    import numpy as np
+    from numpy.typing import NDArray
 
 
 class Tokenizer(Hashable, Protocol):
@@ -13,11 +15,11 @@ class Tokenizer(Hashable, Protocol):
 
     def encode(
         self, prompt: Union[str, List[str]]
-    ) -> Tuple[NDArray[np.int64], NDArray[np.int64]]:
+    ) -> "Tuple['NDArray[np.int64]', 'NDArray[np.int64]']":
         """Translate the input prompts into arrays of token ids and attention mask."""
         ...
 
-    def decode(self, token_ids: NDArray[np.int64]) -> List[str]:
+    def decode(self, token_ids: "NDArray[np.int64]") -> List[str]:
         """Translate an array of token ids to a string or list of strings."""
         ...
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 69bbd85bc..622bede3b 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -185,11 +185,36 @@ def __init__(
             the `transformers` API for tokenizers.
 
         """
+        # We need to handle the cases in which jax/flax or tensorflow
+        # is not available in the environment.
+        try:
+            from transformers import FlaxPreTrainedModel
+        except ImportError:
+            FlaxPreTrainedModel = None
+
+        try:
+            from transformers import TFPreTrainedModel
+        except ImportError:
+            TFPreTrainedModel = None
+
         tokenizer.padding_size = "left"
         self.model = model
         self.tokenizer = TransformerTokenizer(tokenizer)
         self.type_adapter = TransformersTypeAdapter()
 
+        if (
+            FlaxPreTrainedModel is not None
+            and isinstance(model, FlaxPreTrainedModel)
+        ):
+            self.tensor_library_name = "jax"
+        elif (
+            TFPreTrainedModel is not None
+            and isinstance(model, TFPreTrainedModel)
+        ):
+            self.tensor_library_name = "tensorflow"
+        else:
+            self.tensor_library_name = "torch"
+
     def _prepare_model_inputs(self, model_input, output_type):
         prompts = self.type_adapter.format_input(model_input)
         input_ids, attention_mask = self.tokenizer.encode(prompts)
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index cedf049e3..3dfe44be3 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -53,6 +53,7 @@ def format_output_type(self, output_type):
 
 class VLLM(Model):
     """Represents a `vllm` model."""
+    tensor_library_name = "torch"
 
     def __init__(self, model: "LLM"):
         """Create a VLLM model instance.
diff --git a/outlines/serve/serve.py b/outlines/serve/serve.py
index 61d3ed7af..41fdf885a 100644
--- a/outlines/serve/serve.py
+++ b/outlines/serve/serve.py
@@ -70,10 +70,11 @@ async def generate(request: Request) -> Response:
 
     json_schema = request_dict.pop("schema", None)
     regex_string = request_dict.pop("regex", None)
+    tensor_library_name = request_dict.pop("tensor_library_name", "torch")
     if json_schema is not None:
-        logits_processors = [JSONLogitsProcessor(json_schema, tokenizer)]
+        logits_processors = [JSONLogitsProcessor(json_schema, tokenizer, tensor_library_name)]
     elif regex_string is not None:
-        logits_processors = [RegexLogitsProcessor(regex_string, tokenizer)]
+        logits_processors = [RegexLogitsProcessor(regex_string, tokenizer, tensor_library_name)]
     else:
         logits_processors = []
 
diff --git a/pyproject.toml b/pyproject.toml
index 25fd8e5dc..457c72940 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -28,7 +28,6 @@ dependencies = [
    "jinja2",
    "lark",
    "nest_asyncio",
-   "numpy",
    "cloudpickle",
    "diskcache",
    "pydantic>=2.0",
@@ -40,7 +39,6 @@ dependencies = [
    "typing_extensions",
    "iso3166",
    "airportsdata",
-   "torch",
    "outlines_core==0.1.26",
    "genson",
 ]
@@ -68,6 +66,11 @@ test = [
     "transformers",
     "pillow",
     "jax",
+    "flax",
+    "numpy",
+    "torch",
+    "tensorflow",
+    "tf-keras",
     "ollama",
     "dottxt",
     "sentencepiece",
@@ -107,6 +110,7 @@ filterwarnings = [
     "ignore::FutureWarning:huggingface_hub.*",
     "ignore::UserWarning",
     "ignore::DeprecationWarning:pyairports.*",
+    "ignore::DeprecationWarning:jax.*",
 ]
 
 [tool.mypy]
@@ -152,6 +156,10 @@ module = [
     "genson",
     "ollama.*",
     "dottxt.*",
+    "tensorflow",
+    "tensorflow.*",
+    "tf-keras",
+    "tf-keras.*",
 ]
 ignore_missing_imports = true
 
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index fffc82331..f5e196d15 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -5,13 +5,18 @@
 from llama_cpp import Llama
 from pydantic import BaseModel
 
-from outlines.models import LlamaCpp
+from outlines.models.llamacpp import (
+    LlamaCpp,
+    LlamaCppTokenizer,
+    LlamaCppTypeAdapter,
+    from_llamacpp
+)
 from outlines.processors import RegexLogitsProcessor
 from outlines.types.dsl import Regex, python_types_to_terms, to_regex
 
 
 def test_load_model():
-    model = LlamaCpp(
+    model = from_llamacpp(
         Llama.from_pretrained(
             repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
             filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
@@ -20,6 +25,9 @@ def test_load_model():
 
     assert isinstance(model, LlamaCpp)
     assert isinstance(model.model, Llama)
+    assert isinstance(model.tokenizer, LlamaCppTokenizer)
+    assert isinstance(model.type_adapter, LlamaCppTypeAdapter)
+    assert model.tensor_library_name == "numpy"
 
 
 @pytest.fixture(scope="session")
@@ -39,7 +47,7 @@ def test_llamacpp_simple(model):
 
 def test_llamacpp_regex(model):
     regex_str = Regex(r"[0-9]").pattern
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer)
+    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer, model.tensor_library_name)
     result = model.generate("Respond with one word. Not more.", logits_processor)
     assert isinstance(result, str)
 
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index a5ce111fa..b293cf576 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -5,6 +5,11 @@
 
 import outlines
 from outlines.types import Regex
+from outlines.models.mlxlm import (
+    MLXLM,
+    MLXLMTypeAdapter,
+    from_mlxlm
+)
 from outlines.models.transformers import TransformerTokenizer
 from pydantic import BaseModel
 from transformers import PreTrainedTokenizer
@@ -23,20 +28,23 @@
 TEST_MODEL = "mlx-community/SmolLM-135M-Instruct-4bit"
 
 
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_model_initialization():
+    model = from_mlxlm(*mlx_lm.load(TEST_MODEL))
+    assert isinstance(model, MLXLM)
+    assert isinstance(model.model, mlx_lm.MLXLM)
+    assert isinstance(model.mlx_tokenizer, mlx_lm.TokenizerWrapper)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, MLXLMTypeAdapter)
+    assert model.tensor_library_name == "mlx"
+
+
 @pytest.fixture(scope="session")
 def model(tmp_path_factory):
     model, tokenizer = mlx_lm.load(TEST_MODEL)
     return outlines.from_mlxlm(model, tokenizer)
 
 
-@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_model_initialization(model):
-    assert hasattr(model, "model")
-    assert hasattr(model, "mlx_tokenizer")
-    assert hasattr(model, "tokenizer")
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-
-
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
 def test_mlxlm_tokenizer(model):
     # Test single string encoding/decoding
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 1ffb8db21..49e88f538 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -6,7 +6,11 @@
 import transformers
 
 import outlines
-from outlines.models.transformers import Transformers
+from outlines.models.transformers import (
+    Transformers,
+    TransformerTokenizer,
+    TransformersTypeAdapter,
+)
 from outlines.types import Regex
 
 TEST_MODEL = "erwanf/gpt2-mini"
@@ -20,14 +24,31 @@ def test_transformers_instantiate_simple():
         transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
     )
     assert isinstance(model, Transformers)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersTypeAdapter)
+    assert model.tensor_library_name == "torch"
 
 
-def test_transformers_instantiate_other_model_class():
+def test_transformers_instantiate_flax_model():
     model = outlines.from_transformers(
-        transformers.AutoModelForSeq2SeqLM.from_pretrained(TEST_MODEL_SEQ2SEQ),
+        transformers.FlaxAutoModelForCausalLM.from_pretrained(TEST_MODEL),
         transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
     )
     assert isinstance(model, Transformers)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersTypeAdapter)
+    assert model.tensor_library_name == "jax"
+
+
+def test_transformers_instantiate_tensorflow_model():
+    model = outlines.from_transformers(
+        transformers.TFAutoModelForCausalLM.from_pretrained(TEST_MODEL),
+        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+    )
+    assert isinstance(model, Transformers)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersTypeAdapter)
+    assert model.tensor_library_name == "tensorflow"
 
 
 def test_transformers_instantiate_mamba():
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index a701ec9bc..7f64e9821 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -9,19 +9,19 @@
 from PIL import Image
 from pydantic import BaseModel
 from transformers import (
-    Blip2ForConditionalGeneration,
-    CLIPModel,
-    CLIPProcessor,
     LlavaForConditionalGeneration,
     AutoProcessor,
 )
 
 import outlines
-from outlines.models.transformers import TransformersMultiModal
+from outlines.models.transformers import (
+    TransformersMultiModal,
+    TransformerTokenizer,
+    TransformersMultiModalTypeAdapter,
+)
 from outlines.types import Regex
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-TEST_CLIP_MODEL = "openai/clip-vit-base-patch32"
 IMAGE_URLS = [
     "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
     "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
@@ -47,10 +47,13 @@ def model():
 
 def test_transformers_vision_instantiate_simple():
     model = outlines.from_transformers(
-        Blip2ForConditionalGeneration.from_pretrained(TEST_MODEL),
+        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
     assert isinstance(model, TransformersMultiModal)
+    assert isinstance(model.tokenizer, TransformerTokenizer)
+    assert isinstance(model.type_adapter, TransformersMultiModalTypeAdapter)
+    assert model.tensor_library_name == "torch"
 
 
 def test_transformers_vision_simple(model, images):
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index eff98a037..f5a32d997 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -7,21 +7,36 @@
 
 try:
     from vllm import LLM, SamplingParams
+    HAS_VLLM = True
 except ImportError:
-    pass
+    HAS_VLLM = False
 
 import outlines
-from outlines.models.vllm import VLLM
+from outlines.models.vllm import (
+    VLLM,
+    VLLMTypeAdapter,
+    from_vllm
+)
 from outlines.types import Regex
 
 
 TEST_MODEL = "erwanf/gpt2-mini"
 
 pytestmark = pytest.mark.skipif(
-    not torch.cuda.is_available(),
+    not HAS_VLLM,
     reason="vLLM models can only be run on GPU."
 )
 
+
+def test_vllm_model_initialization():
+    model = from_vllm(LLM(TEST_MODEL))
+    assert isinstance(model, VLLM)
+    assert isinstance(model.model, LLM)
+    assert hasattr(model, "tokenizer")
+    assert isinstance(model.type_adapter, VLLMTypeAdapter)
+    assert model.tensor_library_name == "torch"
+
+
 @pytest.fixture(scope="session")
 def model(tmp_path_factory):
     model = outlines.from_vllm(LLM(TEST_MODEL))
diff --git a/tests/test_generator.py b/tests/test_generator.py
index fb8fc8b90..eae1d1632 100644
--- a/tests/test_generator.py
+++ b/tests/test_generator.py
@@ -34,6 +34,7 @@ def sample_processor():
     processor = RegexLogitsProcessor(
         regex_string="[0-9]{3}",
         tokenizer=model.tokenizer,
+        tensor_library_name=model.tensor_library_name,
     )
     return processor
 
@@ -62,14 +63,29 @@ def test_steerable_generator_init_invalid_processor(steerable_model):
         SteerableGenerator.from_processor(steerable_model, Literal["foo", "bar"])
 
 
-def test_steerable_generator_init_valid_output_type(steerable_model):
+def test_steerable_generator_init_cfg_output_type(steerable_model):
+    generator = SteerableGenerator(steerable_model, CFG('start: "a"'))
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_steerable_generator_init_fsm_output_type(steerable_model):
+    generator = SteerableGenerator(
+        steerable_model,
+        FSM(interegular.parse_pattern(r"abc").to_fsm())
+    )
+    assert generator.model == steerable_model
+    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
+
+
+def test_steerable_generator_init_other_output_type(steerable_model):
     generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
     assert generator.model == steerable_model
     assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
 
 
 def test_steerable_generator_init_invalid_output_type(steerable_model, sample_processor):
-    with pytest.raises(TypeError):
+    with pytest.raises(ValueError):
         SteerableGenerator(steerable_model, sample_processor)
 
 

From 6794eb6e3747afeb71a30670c924a8066e71f5a0 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 25 Mar 2025 09:07:42 +0100
Subject: [PATCH 292/505] Add coverage exclusion for mlx, vllm and serve

---
 outlines/models/transformers.py | 4 ++--
 pyproject.toml                  | 5 +++++
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 622bede3b..dafedf8d6 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -189,12 +189,12 @@ def __init__(
         # is not available in the environment.
         try:
             from transformers import FlaxPreTrainedModel
-        except ImportError:
+        except ImportError:  # pragma: no cover
             FlaxPreTrainedModel = None
 
         try:
             from transformers import TFPreTrainedModel
-        except ImportError:
+        except ImportError:  # pragma: no cover
             TFPreTrainedModel = None
 
         tokenizer.padding_size = "left"
diff --git a/pyproject.toml b/pyproject.toml
index 457c72940..6956d4b96 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -111,6 +111,7 @@ filterwarnings = [
     "ignore::UserWarning",
     "ignore::DeprecationWarning:pyairports.*",
     "ignore::DeprecationWarning:jax.*",
+    "ignore::DeprecationWarning:flax.*",
 ]
 
 [tool.mypy]
@@ -167,6 +168,10 @@ ignore_missing_imports = true
 omit = [
     "outlines/_version.py",
     "tests/*",
+    "outlines/models/mlxlm.py",
+    "outlines/models/vllm.py",
+    "outlines/processors/tensor_adapters/mlx.py",
+    "outlines/serve/serve.py",
 ]
 branch = true
 relative_files = true

From 1e60375883d48e8926d486aa303b567b3e0acbfd Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 25 Mar 2025 13:02:08 +0100
Subject: [PATCH 293/505] Fix issue in Templates, change from_str to
 from_string

---
 docs/reference/applications.md | 2 +-
 outlines/applications.py       | 2 +-
 tests/test_applications.py     | 8 ++++----
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/docs/reference/applications.md b/docs/reference/applications.md
index 10c036cc0..93124daa4 100644
--- a/docs/reference/applications.md
+++ b/docs/reference/applications.md
@@ -25,7 +25,7 @@ model = models.from_transformers(
 )
 
 template_string = "What is 2 times {{ num }}?"
-template = Template.from_str(template_string)
+template = Template.from_string(template_string)
 
 application = Application(template, JsonType(OutputModel))
 ```
diff --git a/outlines/applications.py b/outlines/applications.py
index 8dbb69741..8d4ae149b 100644
--- a/outlines/applications.py
+++ b/outlines/applications.py
@@ -35,7 +35,7 @@ class OutputModel(BaseModel):
     )
 
     template_string = "What is 2 times {{ num }}?"
-    template = Template.from_str(template_string)
+    template = Template.from_string(template_string)
 
     application = Application(template, JsonType(OutputModel))
 
diff --git a/tests/test_applications.py b/tests/test_applications.py
index 2f2d4f056..09eb5bd97 100644
--- a/tests/test_applications.py
+++ b/tests/test_applications.py
@@ -37,7 +37,7 @@ def generate_stream(self, model_input: str, output_type=None, **kwargs) -> Any:
 
 
 def test_application_initialization():
-    template = Template.from_str("Test {{ value }}")
+    template = Template.from_string("Test {{ value }}")
     output_type = None
     application = Application(template, output_type)
 
@@ -48,7 +48,7 @@ def test_application_initialization():
 
 
 def test_application_template_call(model):
-    template = Template.from_str("Test {{ value }}")
+    template = Template.from_string("Test {{ value }}")
     output_type = None
     application = Application(template, output_type)
     result = application(model, value="example")
@@ -68,7 +68,7 @@ def template(value):
 
 
 def test_application_template_error(model):
-    template = Template.from_str("Test {{ value }}")
+    template = Template.from_string("Test {{ value }}")
     output_type = None
     application = Application(template, output_type)
 
@@ -77,7 +77,7 @@ def test_application_template_error(model):
 
 
 def test_application_generator_reuse(model, another_model):
-    template = Template.from_str("Test {{ value }}")
+    template = Template.from_string("Test {{ value }}")
     output_type = None
     application = Application(template, output_type)
 

From f54479d060074f33a32bff3608a675d82d6c6d9e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 25 Mar 2025 14:24:12 +0100
Subject: [PATCH 294/505] Modify the OpenAI model to handle multiple sample

---
 outlines/models/openai.py   | 16 ++++++++++------
 tests/models/test_openai.py | 10 ++++++++++
 2 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 6c0f200e0..192a06fb7 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -197,13 +197,17 @@ def generate(
             else:
                 raise e
 
-        message = result.choices[0].message
-        if message.refusal is not None:
-            raise ValueError(
-                f"OpenAI refused to answer the request: {result.choices[0].refusal}"
-            )
+        messages = [choice.message for choice in result.choices]
+        for message in messages:
+            if message.refusal is not None:
+                raise ValueError(
+                    f"OpenAI refused to answer the request: {message.refusal}"
+                )
 
-        return message.content
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
 
     def generate_stream(
         self,
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 5b90508a6..992276054 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -86,6 +86,16 @@ def test_openai_simple_call():
     assert isinstance(result, str)
 
 
+@pytest.mark.api_call
+def test_openai_simple_call_multiple_samples():
+    model = OpenAI(OpenAIClient(), MODEL_NAME)
+    result = model.generate("Respond with one word. Not more.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
 @pytest.mark.api_call
 def test_openai_direct_call():
     model = OpenAI(OpenAIClient(), MODEL_NAME)

From 88ef134ae57e6caf1663a980150c9d27f35110c5 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 14:21:03 +0200
Subject: [PATCH 295/505] Restore the load_lora method to LlamaCpp with a
 warning

---
 outlines/models/llamacpp.py | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 2f42ddca8..9412665e4 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -143,8 +143,6 @@ class LlamaCpp(Model):
     tensor_library_name = "numpy"
 
     def __init__(self, model: "Llama"):
-        from llama_cpp import Llama
-
         self.model = model
         self.tokenizer = LlamaCppTokenizer(self.model)
         self.type_adapter = LlamaCppTypeAdapter()
@@ -226,6 +224,22 @@ def token_generator() -> Iterator[str]:
 
         return token_generator()
 
+    def load_lora(self, adapter_path: str):
+        warnings.warn("""
+            The `load_lora` method is deprecated starting from v1.0.0.
+            Support for it will be removed in v1.5.0.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        if self.model._model.apply_lora_from_file(
+            adapter_path,
+            1.0,
+        ):
+            raise RuntimeError(
+                f"Failed to apply LoRA from lora path: {adapter_path}"
+            )
+
 
 def from_llamacpp(model: "Llama"):
     return LlamaCpp(model)

From 719df7f8ed7e1e1a71177f9de8e053dbe147521c Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 14:22:12 +0200
Subject: [PATCH 296/505] Fix error in the generate_stream method of MLXLM

---
 outlines/models/mlxlm.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index fda8f1a08..238e16ffc 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -113,14 +113,14 @@ def generate_stream(self, model_input, output_type=None, **kwargs):
         """
         from mlx_lm import stream_generate
 
-        for token in stream_generate(
+        for gen_response in stream_generate(
             self.model,
             self.mlx_tokenizer,
             self.type_adapter.format_input(model_input),
             logits_processors=self.type_adapter.format_output_type(output_type),
             **kwargs,
         ):
-            yield token
+            yield gen_response.text
 
 
 def from_mlxlm(model: "nn.Module", tokenizer: "PreTrainedTokenizer") -> MLXLM:

From b3711dd4fe4c26a259f099a25482b48ba8b661dc Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 14:25:16 +0200
Subject: [PATCH 297/505] Restore the load_lora method to VLLM with a warning

---
 outlines/models/vllm.py | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 3dfe44be3..4f928eac0 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,4 +1,4 @@
-import dataclasses
+import warnings
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, List, Optional, Union
 
@@ -7,7 +7,6 @@
 if TYPE_CHECKING:
     from transformers import PreTrainedTokenizerBase
     from vllm import LLM
-    from vllm.sampling_params import SamplingParams
 
 
 __all__ = ["VLLM", "from_vllm"]
@@ -67,6 +66,7 @@ def __init__(self, model: "LLM"):
         self.model = model
         self.tokenizer = self._get_tokenizer()
         self.type_adapter = VLLMTypeAdapter()
+        self.lora_request = None # v0 legacy, to be removed
 
     def _get_tokenizer(self):
         if hasattr(self.model, "get_tokenizer"):
@@ -111,6 +111,7 @@ def generate(self, model_input, output_type, **inference_kwargs):
         results = self.model.generate(
             self.type_adapter.format_input(model_input),
             sampling_params=sampling_params,
+            lora_request=self.lora_request, # v0 legacy, to be removed
             **inference_kwargs,
         )
         results = [[sample.text for sample in batch.outputs] for batch in results]
@@ -139,6 +140,24 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
             "Streaming is not available for the vLLM integration."
         )
 
+    def load_lora(self, adapter_path: Optional[str]):
+        warnings.warn("""
+            The `load_lora` method is deprecated starting from v1.0.0.
+            Support for it will be removed in v1.5.0.
+            Please use the v1 of the `outlines` library by using the
+            `outlines.from_vllm` function to create a `VLLM` model
+            instance.
+            In the v1, you must pass the `lora_request` argument as
+            a keyword argument when calling the model or generator.
+            """)
+
+        from vllm.lora.request import LoRARequest
+
+        if adapter_path is None:
+            self.lora_request = None
+        else:
+            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
+
 
 def from_vllm(model: "LLM") -> VLLM:
     return VLLM(model)

From 4418153aa3b3b49cfc7439fe614a57e65a8bce37 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 14:26:03 +0200
Subject: [PATCH 298/505] Small fixes in the Transformers model

---
 outlines/models/transformers.py | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index dafedf8d6..8a48961cc 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -197,8 +197,9 @@ def __init__(
         except ImportError:  # pragma: no cover
             TFPreTrainedModel = None
 
-        tokenizer.padding_size = "left"
+        tokenizer.padding_side = "left"
         self.model = model
+        self.transformer_tokenizer = tokenizer
         self.tokenizer = TransformerTokenizer(tokenizer)
         self.type_adapter = TransformersTypeAdapter()
 
@@ -249,7 +250,11 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
 
     def _generate_output_seq(self, prompts, inputs, **inference_kwargs):
         input_ids = inputs["input_ids"]
-        output_ids = self.model.generate(**inputs, **inference_kwargs)
+        output_ids = self.model.generate(
+            **inputs,
+            tokenizer=self.transformer_tokenizer,
+            **inference_kwargs,
+        )
 
         # encoder-decoder returns output_ids only, decoder-only returns full seq ids
         if self.model.config.is_encoder_decoder:

From fb269343fd6e43bba49ce80f3a0c7ff5b6efdc9e Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 14:30:01 +0200
Subject: [PATCH 299/505] Add a whitespace_pattern to JsonSchema

---
 outlines/types/dsl.py | 28 ++++++++++++++++++++++------
 1 file changed, 22 insertions(+), 6 deletions(-)

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index a04f2be64..6b16c5897 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,17 +1,26 @@
-import datetime
 import json
 import re
 import sys
 from dataclasses import dataclass
 from enum import EnumMeta
 from types import FunctionType
-from typing import Any, Callable, List, Literal, Union, get_args, get_origin
-
+from typing import (
+    Any,
+    List,
+    Optional as OptionalType,
+    Union,
+    get_args,
+)
 import interegular
 import jsonschema
 from genson import SchemaBuilder
 from outlines_core.fsm.json_schema import build_regex_from_schema
-from pydantic import BaseModel, GetCoreSchemaHandler, GetJsonSchemaHandler, TypeAdapter
+from pydantic import (
+    BaseModel,
+    GetCoreSchemaHandler,
+    GetJsonSchemaHandler,
+    TypeAdapter,
+)
 from pydantic.json_schema import JsonSchemaValue
 from pydantic_core import core_schema as cs
 
@@ -242,7 +251,13 @@ def __repr__(self):
 
 
 class JsonSchema(Term):
-    def __init__(self, schema: Union[dict, str, type[BaseModel], _TypedDictMeta, type, SchemaBuilder]):
+    def __init__(
+        self,
+        schema: Union[
+            dict, str, type[BaseModel], _TypedDictMeta, type, SchemaBuilder
+        ],
+        whitespace_pattern: OptionalType[str] = None,
+    ):
         schema_str: str
 
         if is_dict_instance(schema):
@@ -266,6 +281,7 @@ def __init__(self, schema: Union[dict, str, type[BaseModel], _TypedDictMeta, typ
             )
 
         self.schema = schema_str
+        self.whitespace_pattern = whitespace_pattern
 
     def __post_init__(self):
         jsonschema.Draft7Validator.check_schema(json.loads(self.schema))
@@ -714,7 +730,7 @@ def to_regex(term: Term) -> str:
         case Regex():
             return f"({term.pattern})"
         case JsonSchema():
-            regex_str = build_regex_from_schema(term.schema)
+            regex_str = build_regex_from_schema(term.schema, term.whitespace_pattern)
             return f"({regex_str})"
         case KleeneStar():
             return f"({to_regex(term.term)})*"

From 4cf766ab73e53a34df1f2e1a2ad7f6e25b19ee9d Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 15:07:46 +0200
Subject: [PATCH 300/505] Restore the base, function and samplers files in
 v0_legacy

---
 outlines/v0_legacy/base.py     | 276 ++++++++++++++++++++++++++
 outlines/v0_legacy/function.py | 169 ++++++++++++++++
 outlines/v0_legacy/samplers.py | 353 +++++++++++++++++++++++++++++++++
 3 files changed, 798 insertions(+)
 create mode 100644 outlines/v0_legacy/base.py
 create mode 100644 outlines/v0_legacy/function.py
 create mode 100644 outlines/v0_legacy/samplers.py

diff --git a/outlines/v0_legacy/base.py b/outlines/v0_legacy/base.py
new file mode 100644
index 000000000..8d767391f
--- /dev/null
+++ b/outlines/v0_legacy/base.py
@@ -0,0 +1,276 @@
+import asyncio
+import builtins
+import functools
+import inspect
+from typing import Callable, Optional
+
+
+# Allow nested loops for running in notebook. We don't enable it globally as it
+# may interfere with other libraries that use asyncio.
+if hasattr(builtins, "__IPYTHON__"):
+    try:
+        import nest_asyncio
+
+        nest_asyncio.apply()
+    except ImportError:
+        print(
+            "Couldn't patch nest_asyncio because it's not installed. Running in the notebook might be have issues"
+        )
+
+
+class vectorize:
+    """Returns an object that acts like a function but takes arrays as an input.
+
+    The vectorized function evaluates `func` over successive tuples of the input
+    chararrays and returns a single NumPy chararrays or a tuple of NumPy chararrays.
+
+    Its behavior is similar to NumPy's `vectorize` for Python functions: the function
+    being vectorized is executed in a `for` loop. Coroutines, however, are executed
+    concurrently.
+
+    Part of the code was adapted from `numpy.lib.function_base`.
+
+    """
+
+    def __init__(self, func: Callable, signature: Optional[str] = None):
+        # imports are in the __init__ as numpy is an optional dependency
+        import numpy as np
+        self.np = np
+        # Import required functions based on NumPy version
+        np_major_version = int(np.__version__.split(".")[0])
+        if np_major_version >= 2:
+            from numpy.lib._function_base_impl import (
+                _calculate_shapes,
+                _parse_gufunc_signature,
+                _parse_input_dimensions,
+                _update_dim_sizes,
+            )
+        else:
+            from numpy.lib.function_base import (
+                _calculate_shapes,
+                _parse_gufunc_signature,
+                _parse_input_dimensions,
+                _update_dim_sizes,
+            )
+        self._calculate_shapes = _calculate_shapes
+        self._parse_gufunc_signature = _parse_gufunc_signature
+        self._parse_input_dimensions = _parse_input_dimensions
+        self._update_dim_sizes = _update_dim_sizes
+        #
+        self.func = func
+        self.signature = signature
+        self.is_coroutine_fn = inspect.iscoroutinefunction(func)
+
+        functools.update_wrapper(self, func)
+
+        if signature is not None:
+            # Parse the signature string into a Python data structure.
+            # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
+            self._in_and_out_core_dimensions = self._parse_gufunc_signature(signature)
+        else:
+            self._in_and_out_core_dimensions = None
+
+    def __call__(self, *args, **kwargs):
+        """Call the vectorized function."""
+        if not args and not kwargs:
+            return self.call_thunk()
+        elif self.signature is not None:
+            return self.call_with_signature(*args, **kwargs)
+        else:
+            return self.call_no_signature(*args, **kwargs)
+
+    def call_thunk(self):
+        """Call a vectorized thunk.
+
+        Thunks have no arguments and can thus be called directly.
+
+        """
+        if self.is_coroutine_fn:
+            loop = asyncio.new_event_loop()
+            try:
+                outputs = loop.run_until_complete(self.func())
+            finally:
+                loop.close()
+        else:
+            outputs = self.func()
+
+        return outputs
+
+    def call_no_signature(self, *args, **kwargs):
+        """Call functions and coroutines when no signature is specified.
+
+        When no signature is specified we assume that all of the function's
+        inputs and outputs are scalars (core dimension of zero). We first
+        broadcast the input arrays, then iteratively apply the function over the
+        elements of the broadcasted arrays and finally reshape the results to
+        match the input shape.
+
+        Functions are executed in a for loop, coroutines are executed
+        concurrently.
+
+        """
+        # Convert args and kwargs to arrays
+        args = [self.np.array(arg) for arg in args]
+        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
+
+        # Broadcast args and kwargs
+        broadcast_shape = self.np.broadcast(*args, *list(kwargs.values())).shape
+        args = [self.np.broadcast_to(arg, broadcast_shape) for arg in args]
+        kwargs = {
+            key: self.np.broadcast_to(value, broadcast_shape)
+            for key, value in kwargs.items()
+        }
+
+        # Execute functions in a loop, and coroutines concurrently
+        if self.is_coroutine_fn:
+            outputs = self.vectorize_call_coroutine(broadcast_shape, args, kwargs)
+        else:
+            outputs = self.vectorize_call(broadcast_shape, args, kwargs)
+
+        # `outputs` is a flat array or a tuple of flat arrays. We reshape the arrays
+        # to match the input shape.
+        outputs = [
+            results if isinstance(results, tuple) else (results,) for results in outputs
+        ]
+        outputs = tuple(
+            [self.np.asarray(x).reshape(broadcast_shape).squeeze() for x in zip(*outputs)]
+        )
+        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
+
+        n_results = len(list(outputs))
+
+        return outputs[0] if n_results == 1 else outputs
+
+    def call_with_signature(self, *args, **kwargs):
+        """Call functions and coroutines when a signature is specified."""
+        input_core_dims, output_core_dims = self._in_and_out_core_dimensions
+
+        # Make sure that the numbers of arguments passed is compatible with
+        # the signature.
+        num_args = len(args) + len(kwargs)
+        if num_args != len(input_core_dims):
+            raise TypeError(
+                "wrong number of positional arguments: "
+                "expected %r, got %r" % (len(input_core_dims), len(args))
+            )
+
+        # Convert args and kwargs to arrays
+        args = [self.np.asarray(arg) for arg in args]
+        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
+
+        # Find the arguments' broadcast shape, and map placeholder
+        # variables in the signature to the number of dimensions
+        # they correspond to given the arguments.
+        broadcast_shape, dim_sizes = self._parse_input_dimensions(
+            args + list(kwargs.values()), input_core_dims
+        )
+
+        # Calculate the shape to which each of the arguments should be broadcasted
+        # and reshape them accordingly.
+        input_shapes = self._calculate_shapes(broadcast_shape, dim_sizes, input_core_dims)
+        args = [
+            self.np.broadcast_to(arg, shape, subok=True)
+            for arg, shape in zip(args, input_shapes)
+        ]
+        kwargs = {
+            key: self.np.broadcast_to(value, broadcast_shape)
+            for key, value in kwargs.items()
+        }
+
+        n_out = len(output_core_dims)
+
+        if self.is_coroutine_fn:
+            outputs = self.vectorize_call_coroutine(broadcast_shape, args, kwargs)
+        else:
+            outputs = self.vectorize_call(broadcast_shape, args, kwargs)
+
+        outputs = [
+            results if isinstance(results, tuple) else (results,) for results in outputs
+        ]
+
+        flat_outputs = list(zip(*outputs))
+        n_results = len(flat_outputs)
+
+        if n_out != n_results:
+            raise ValueError(
+                f"wrong number of outputs from the function, expected {n_out}, got {n_results}"
+            )
+
+        # The number of dimensions of the outputs are not necessarily known in
+        # advance. The following iterates over the results and updates the
+        # number of dimensions of the outputs accordingly.
+        for results, core_dims in zip(flat_outputs, output_core_dims):
+            for result in results:
+                self._update_dim_sizes(dim_sizes, result, core_dims)
+
+        # Calculate the shape to which each of the outputs should be broadcasted
+        # and reshape them.
+        shapes = self._calculate_shapes(broadcast_shape, dim_sizes, output_core_dims)
+        outputs = tuple(
+            [
+                self.np.hstack(results).reshape(shape).squeeze()
+                for shape, results in zip(shapes, zip(*outputs))
+            ]
+        )
+        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
+
+        return outputs[0] if n_results == 1 else outputs
+
+    def vectorize_call(self, broadcast_shape, args, kwargs):
+        """Run the function in a for loop.
+
+        A possible extension would be to parallelize the calls.
+
+        Parameters
+        ----------
+        broadcast_shape
+            The brodcast shape of the input arrays.
+        args
+            The function's broadcasted arguments.
+        kwargs
+            The function's broadcasted keyword arguments.
+
+        """
+        outputs = []
+        for index in self.np.ndindex(*broadcast_shape):
+            current_args = tuple(arg[index] for arg in args)
+            current_kwargs = {key: value[index] for key, value in kwargs.items()}
+            outputs.append(self.func(*current_args, **current_kwargs))
+
+        return outputs
+
+    def vectorize_call_coroutine(self, broadcast_shape, args, kwargs):
+        """Run coroutines concurrently.
+
+        Creates as many tasks as needed and executes them in a new event
+        loop.
+
+        Parameters
+        ----------
+        broadcast_shape
+            The brodcast shape of the input arrays.
+        args
+            The function's broadcasted arguments.
+        kwargs
+            The function's broadcasted keyword arguments.
+
+        """
+
+        async def create_and_gather_tasks():
+            tasks = []
+            for index in self.np.ndindex(*broadcast_shape):
+                current_args = tuple(arg[index] for arg in args)
+                current_kwargs = {key: value[index] for key, value in kwargs.items()}
+                tasks.append(self.func(*current_args, **current_kwargs))
+
+            outputs = await asyncio.gather(*tasks)
+
+            return outputs
+
+        loop = asyncio.new_event_loop()
+        try:
+            outputs = loop.run_until_complete(create_and_gather_tasks())
+        finally:
+            loop.close()
+
+        return outputs
diff --git a/outlines/v0_legacy/function.py b/outlines/v0_legacy/function.py
new file mode 100644
index 000000000..e0e10c3a0
--- /dev/null
+++ b/outlines/v0_legacy/function.py
@@ -0,0 +1,169 @@
+import importlib.util
+import warnings
+from dataclasses import dataclass
+from typing import TYPE_CHECKING, Callable, Optional, Tuple, Union
+
+import requests
+
+from outlines.v0_legacy import generate
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.models import transformers
+
+if TYPE_CHECKING:
+    from outlines.templates import Template
+
+
+@dataclass
+class Function:
+    """Represents an Outlines function.
+
+    Functions are a convenient way to encapsulate a prompt template, a language
+    model and a Pydantic model that define the output structure. Once defined,
+    the function can be called with arguments that will be used to render the
+    prompt template.
+
+    """
+
+    prompt_template: "Template"
+    schema: Union[str, Callable, object]
+    model_name: str
+    generator: Optional["GeneratorV0Adapter"] = None
+
+    def __post_init__(self):
+        _warn()
+
+    @classmethod
+    def from_github(cls, program_path: str, function_name: str = "fn"):
+        """Load a function stored on GitHub"""
+        _warn()
+        program_content = download_from_github(program_path)
+        function = extract_function_from_file(program_content, function_name)
+
+        return function
+
+    def init_generator(self):
+        """Load the model and initialize the generator."""
+        model = transformers(self.model_name)
+        self.generator = generate.json(model, self.schema)
+
+    def __call__(self, *args, **kwargs):
+        """Call the function.
+
+        .. warning::
+
+           This currently does not support batching.
+
+        Parameters
+        ----------
+        args
+            Values to pass to the prompt template as positional arguments.
+        kwargs
+            Values to pass to the prompt template as keyword arguments.
+
+        """
+        # We do not want to raise the deprecation warning about the `json`
+        # function or the transformers function here.
+        with warnings.catch_warnings():
+            warnings.filterwarnings(
+                "ignore",
+                category=DeprecationWarning
+            )
+            if self.generator is None:
+                self.init_generator()
+
+            prompt = self.prompt_template(*args, **kwargs)
+            return self.generator(prompt)
+
+
+def _warn():
+    warnings.warn("""
+        The `Function` class is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should use the `outlines.Application` class that
+        implements a similar functionality.
+        An `Application` is initialized with a prompt template and an
+        output type. It can then be called with a model and a prompt
+        to generate a response that follows the format specified by the
+        output type.
+        For example:
+        ```python
+        from pydantic import BaseModel
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+        from outlines import models, Application
+        from outlines.types import JsonType
+        from outlines.templates import Template
+
+        class OutputModel(BaseModel):
+            result: int
+
+        model = models.from_transformers(
+            AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+            AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+        )
+
+        template_string = "What is 2 times {{ num }}?"
+        template = Template.from_string(template_string)
+
+        application = Application(template, JsonType(OutputModel))
+
+        result = application(model, num=3)
+        print(result)  # Expected output: { "result" : 6 }
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2
+    )
+
+
+def download_from_github(short_path: str):
+    """Download the file in which the function is stored on GitHub."""
+    GITHUB_BASE_URL = "https://raw.githubusercontent.com"
+    BRANCH = "main"
+
+    path = short_path.split("/")
+    if len(path) < 3:
+        raise ValueError(
+            "Please provide a valid path in the form {USERNAME}/{REPO_NAME}/{PATH_TO_FILE}."
+        )
+    elif short_path[-3:] == ".py":
+        raise ValueError("Do not append the `.py` extension to the program name.")
+
+    username = path[0]
+    repo = path[1]
+    path_to_file = path[2:]
+
+    url = "/".join([GITHUB_BASE_URL, username, repo, BRANCH] + path_to_file) + ".py"
+    result = requests.get(url)
+
+    if result.status_code == 200:
+        return result.text
+    elif result.status_code == 404:
+        raise ValueError(
+            f"Program could not be found at {url}. Please make sure you entered the GitHub username, repository name and path to the program correctly."
+        )
+    else:
+        result.raise_for_status()
+
+
+def extract_function_from_file(content: str, function_name: str) -> Tuple[Callable]:
+    """Extract a function object from a downloaded file."""
+    spec = importlib.util.spec_from_loader(
+        "outlines_function", loader=None, origin="github"
+    )
+    if spec is not None:
+        module = importlib.util.module_from_spec(spec)
+        exec(content, module.__dict__)
+
+        try:
+            fn = getattr(module, function_name)
+        except AttributeError:
+            raise AttributeError(
+                "Could not find an `outlines.Function` instance in the remote file. Make sure that the path you specified is correct."
+            )
+
+        if not isinstance(fn, module.outlines.Function):
+            raise TypeError(
+                f"The `{function_name}` variable in the program must be an instance of `outlines.Function`"
+            )
+
+    return fn
diff --git a/outlines/v0_legacy/samplers.py b/outlines/v0_legacy/samplers.py
new file mode 100644
index 000000000..2ae22161d
--- /dev/null
+++ b/outlines/v0_legacy/samplers.py
@@ -0,0 +1,353 @@
+
+import math
+from dataclasses import dataclass
+from typing import TYPE_CHECKING, Callable, Optional, Protocol, Tuple
+
+if TYPE_CHECKING:
+    import torch
+
+
+class Sampler(Protocol):
+    samples: int
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        rng: "torch.Generator",
+    ) -> "torch.DoubleTensor": ...
+
+
+@dataclass(frozen=True)
+class SamplingParameters:
+    """Sampling parameters available in Outlines."""
+    sampler: str
+    num_samples: int = 1
+    top_p: Optional[float] = None
+    top_k: Optional[int] = None
+    temperature: Optional[float] = None
+
+
+class GreedySampler:
+    """Greedy Sampling algorithm.
+
+    Greedy sampling consists in choosing the token with the largest
+    likelihood at every step.
+
+    We don't allow more than one sample. We could attribute this a meaning, for
+    instance the k-th sample represents the k-th most likely token. In which
+    case it would be equivalent to beam search without the sequence weights.
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence.
+
+    """
+
+    def __init__(self):
+        self.samples = 1
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        _,
+    ) -> "torch.DoubleTensor":
+        """Call the greedy sampler.
+
+        Parameters
+        ----------
+        next_token_logits
+            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
+            probability distribution of the next token over the vocabulary.
+        sequence_weights
+            A tensor of shape ``(n_seqs,)`` that represents the cumulative
+            weight of each sequence.
+        rng
+            A random number generator.
+
+        Returns
+        -------
+        A tuple with an array that contains the ids of the sampled tokens of
+        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
+        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
+        cumulative weights of each sequence of shape ``(n_seqs,)``.
+
+        """
+        import torch
+
+        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
+        next_token_ids = torch.argmax(logprobs, dim=-1, keepdim=True)
+
+        ancestors = torch.arange(
+            next_token_logits.shape[0], device=next_token_logits.device
+        )
+        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
+
+        return next_token_ids, ancestors, weights
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters("greedy", self.samples, None, None, 0.0)
+
+
+greedy = GreedySampler
+
+
+class MultinomialSampler:
+    """Multinomial sampling algorithm.
+
+    Multinomial sampling consists in randomly sampling the next token assuming
+    its distribution is a Categorical distribution parametrized by the
+    next-token logits.
+
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence.
+
+    """
+
+    def __init__(
+        self,
+        samples: int = 1,
+        *,
+        top_k: Optional[int] = None,
+        top_p: Optional[float] = None,
+        temperature: Optional[float] = None,
+    ):
+        self.samples = samples
+        self.top_k = top_k
+        self.top_p = top_p
+        self.temperature = temperature
+
+        self.logits_processors = []
+        if top_k is not None:
+            self.logits_processors.append(keep_top_k_logits(top_k))
+        elif top_p is not None:
+            self.logits_processors.append(keep_top_p_logits(top_p))
+
+        if temperature is not None:
+            self.logits_processors.append(rescale_logits(temperature))
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        rng: "torch.Generator",
+    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
+        """Call the multinomial sampler.
+
+        Parameters
+        ----------
+        next_token_logits
+            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
+            probability distribution of the next token over the vocabulary.
+        sequence_weights
+            A tensor of shape ``(n_seqs,)`` that represents the cumulative
+            weight of each sequence.
+        rng
+            A random number generator.
+
+        Returns
+        -------
+        A tuple with an array that contains the ids of the sampled tokens of
+        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
+        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
+        cumulative weights of each sequence of shape ``(n_seqs,)``.
+
+        """
+        import torch
+
+        altered_next_token_logits = next_token_logits
+        for logit_processor in self.logits_processors:
+            altered_next_token_logits = logit_processor(next_token_logits)
+
+        probs = torch.nn.functional.softmax(altered_next_token_logits, dim=-1)
+        next_token_ids = torch.multinomial(probs, num_samples=1, generator=rng)
+
+        logprobs = torch.nn.functional.log_softmax(altered_next_token_logits, dim=-1)
+        ancestors = torch.arange(
+            altered_next_token_logits.shape[0], device=next_token_logits.device
+        )
+        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
+
+        return next_token_ids, ancestors, weights
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters(
+            "multinomial",
+            self.samples,
+            self.top_p,
+            self.top_k,
+            self.temperature,
+        )
+
+
+multinomial = MultinomialSampler
+
+
+def keep_top_k_logits(k: int) -> Callable[["torch.Tensor"], "torch.Tensor"]:
+    """Build a function that masks logits values smaller than the top `k` ones.
+
+    Parameters
+    ----------
+    k
+        The ranking below which logit values are replaced by `-math.inf`.
+
+    """
+    import torch
+
+    if not isinstance(k, int) or k < 1:
+        raise ValueError(f"`k` must be a strictly positive integers, got {k} instead.")
+
+    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
+        num_to_keep = min(k, logits.size(-1))
+        mask_idx = logits < torch.topk(logits, num_to_keep)[0][..., -1, None]
+        return logits.masked_fill(mask_idx, -math.inf)
+
+    return logits_processor
+
+
+def keep_top_p_logits(p: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
+    """Build a function that masks the lowest probability tokens whose
+    cumulative probability is below a certain threshold.
+
+    Parameters
+    ----------
+    p
+        The value of the threshold. We keep the highest probability tokens whose
+        cumulative distribution is greater than or equal to `p` and mask the
+        others. Its value must be between 0 (excluded) and 1 (included).
+
+    """
+    import torch
+
+    if p <= 0.0 or p > 1.0:
+        raise ValueError(
+            f"`p` must be a floating point number between 0 (excluded) and 1 (included), got {p} instead."
+        )
+
+    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
+        sorted_logits, sorted_idx = torch.sort(logits, descending=False)
+        cumulative_probabilties = torch.nn.functional.softmax(
+            sorted_logits, dim=-1
+        ).cumsum(dim=-1)
+
+        sorted_masked_idx = cumulative_probabilties <= (1 - p)
+        mask_idx = torch.scatter(sorted_masked_idx, 1, sorted_idx, sorted_masked_idx)
+        return logits.masked_fill(mask_idx, -math.inf)
+
+    return logits_processor
+
+
+def rescale_logits(temperature: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
+    """Build a function that rescales the token probabilities exponentially.
+
+    Parameters
+    ----------
+    temperature
+        The value by which we rescale the logits.
+
+    """
+
+    if not isinstance(temperature, float) or temperature < 0.0:
+        raise ValueError(
+            f"`temperature` must be a strictly positive floating point number, got {temperature} instead."
+        )
+    elif temperature == 0.0:
+        raise ValueError(
+            "Please use the greedy sampler instead of setting the temperature to 0."
+        )
+
+    def logits_processor(logits: "torch.Tensor") -> "torch.Tensor":
+        return logits / temperature
+
+    return logits_processor
+
+
+class BeamSearchSampler:
+    """Beam Search sampling algorithm.
+
+    Attributes
+    ----------
+    samples
+        The number of samples taken for each input sequence. Equivalent to the
+        number of beams.
+    """
+
+    def __init__(self, beams: int = 1):
+        self.samples = beams
+
+    def __call__(
+        self,
+        next_token_logits: "torch.DoubleTensor",
+        sequence_weights: "torch.DoubleTensor",
+        _,
+    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
+        """Call the beam search sampler.
+
+        Parameters
+        ----------
+        next_token_logits
+            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
+            probability distribution of the next token over the vocabulary.
+        sequence_weights
+            A tensor of shape ``(n_seqs,)`` that represents the cumulative
+            weight of each sequence.
+        rng
+            A random number generator.
+
+        Returns
+        -------
+        A tuple with an array that contains the ids of the sampled tokens of
+        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
+        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
+        cumulative weights of each sequence of shape ``(n_seqs,)``.
+
+        """
+        import torch
+
+        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
+        weights = logprobs + sequence_weights.unsqueeze(1).expand_as(next_token_logits)
+
+        # Flatten scores to (n_batch, n_samples * vocab_size)
+        # and find the top-k weights for each batch.
+        batch_size = next_token_logits.shape[0] // self.samples
+        vocab_size = next_token_logits.shape[-1]
+        weights = weights.view(batch_size, self.samples * vocab_size)
+
+        # If the weights are all equal to 0 we are at the beginning of the search
+        # and thus only need to sample from one set of token logits for each
+        # batch.
+        if torch.all(sequence_weights == 0):
+            weights = weights[:, :vocab_size]
+
+        weights, indices = torch.topk(
+            weights, self.samples, dim=1, largest=True, sorted=True
+        )
+
+        ancestors = torch.div(indices, vocab_size, rounding_mode="floor")
+        next_token_ids = indices % vocab_size
+
+        # Re-shape the weights, next_token_ids and ancestors to (n_batch * n_samples, 1)
+        first_batch_idx = torch.arange(
+            0, batch_size * self.samples, self.samples, device=next_token_logits.device
+        ).unsqueeze(1)
+        ancestors = ancestors + first_batch_idx
+
+        ancestors = ancestors.view(self.samples * batch_size)
+        weights = weights.view(self.samples * batch_size)
+        next_token_ids = next_token_ids.view(self.samples * batch_size, 1)
+
+        return next_token_ids, ancestors, weights
+
+    @property
+    def sampling_params(self):
+        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
+
+
+beam_search = BeamSearchSampler

From 94dc81a7cc61ba7be5940a3faa8408af8fd27af6 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 15:13:35 +0200
Subject: [PATCH 301/505] Restore the v0 models' loading functions and update
 the OpenAI model

---
 outlines/models/openai.py                     | 331 ++++++++++++++-
 outlines/v0_legacy/models/__init__.py         |  21 +
 outlines/v0_legacy/models/exllamav2.py        | 400 ++++++++++++++++++
 outlines/v0_legacy/models/llamacpp.py         | 147 +++++++
 outlines/v0_legacy/models/mlxlm.py            | 153 +++++++
 outlines/v0_legacy/models/openai.py           |  64 +++
 outlines/v0_legacy/models/transformers.py     | 196 +++++++++
 .../v0_legacy/models/transformers_vision.py   | 117 +++++
 outlines/v0_legacy/models/vllm.py             | 101 +++++
 9 files changed, 1522 insertions(+), 8 deletions(-)
 create mode 100644 outlines/v0_legacy/models/__init__.py
 create mode 100644 outlines/v0_legacy/models/exllamav2.py
 create mode 100644 outlines/v0_legacy/models/llamacpp.py
 create mode 100644 outlines/v0_legacy/models/mlxlm.py
 create mode 100644 outlines/v0_legacy/models/openai.py
 create mode 100644 outlines/v0_legacy/models/transformers.py
 create mode 100644 outlines/v0_legacy/models/transformers_vision.py
 create mode 100644 outlines/v0_legacy/models/vllm.py

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 192a06fb7..84c219aeb 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,11 +1,23 @@
 """Integration with OpenAI's API."""
 
+import copy
+import functools
 import json
-from typing import Optional, Union, TYPE_CHECKING
+import warnings
+from typing import (
+    Any,
+    Callable,
+    Dict,
+    List,
+    Optional,
+    TYPE_CHECKING,
+    Union
+)
+from dataclasses import asdict, dataclass, field, replace
 
 from pydantic import BaseModel, TypeAdapter
-from typing_extensions import is_typeddict
 
+from outlines.caching import cache
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
 from outlines.types import JsonSchema, Regex, CFG
@@ -16,6 +28,8 @@
     is_genson_schema_builder,
     is_native_dict
 )
+from outlines.v0_legacy.base import vectorize
+
 
 if TYPE_CHECKING:
     from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
@@ -166,13 +180,40 @@ class OpenAI(Model):
     """
 
     def __init__(
-        self, client: Union["OpenAIClient", "AzureOpenAIClient"], model_name: str
+        self,
+        client: Union["OpenAIClient", "AzureOpenAIClient"],
+        model_name: Any,
+        **kwargs
     ):
-        from openai import OpenAI
-
-        self.client = client
-        self.model_name = model_name
-        self.type_adapter = OpenAITypeAdapter()
+        # legacy mode
+        if isinstance(model_name, OpenAIConfig) or kwargs.get("config"):
+            warnings.warn("""
+                The `openai` function is deprecated starting from v1.0.0.
+                Do not use it. Support for it will be removed in v1.5.0.
+                Instead, you should instantiate a `OpenAI` model with the
+                `outlines.from_openai` function that takes an openai library
+                client and a model name as arguments. Similarly, you cannot
+                instantiate a `OpenAI` model directly with a `OpenAIConfig`
+                instance anymore, but must provide a client and a model name
+                instead.
+                For example:
+                ```python
+                from openai import OpenAI as OpenAIClient
+                from outlines import from_openai
+                client = OpenAIClient()
+                model = from_openai(client, "gpt-4o")
+                ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+            )
+            config = model_name if isinstance(model_name, OpenAIConfig) else kwargs.pop("config")
+            self.legacy_instance = OpenAILegacy(client, config, kwargs.get("system_prompt"))
+        # regular mode
+        else:
+            self.client = client
+            self.model_name = model_name
+            self.type_adapter = OpenAITypeAdapter()
 
     def generate(
         self,
@@ -229,8 +270,282 @@ def generate_stream(
             if chunk.choices and chunk.choices[0].delta.content is not None:
                 yield chunk.choices[0].delta.content
 
+    ### Legacy !!!
+
+    def __call__(self, *args, **kwargs):
+        if hasattr(self, "legacy_instance"):
+            return self.legacy_instance(*args, **kwargs)
+        else:
+            return super().__call__(*args, **kwargs)
+
+    def stream(self, *args, **kwargs):
+        if hasattr(self, "legacy_instance"):
+            return self.legacy_instance.stream(*args, **kwargs)
+        else:
+            return super().stream(*args, **kwargs)
+
+    def new_with_replacements(self, **kwargs):
+        if hasattr(self, "legacy_instance"):
+            return self.legacy_instance.new_with_replacements(**kwargs)
+        raise NotImplementedError("This method is only available in legacy mode")
+
+    def __str__(self):
+        if hasattr(self, "legacy_instance"):
+            return str(self.legacy_instance)
+        else:
+            return super().__str__()
+
+    def __repr__(self):
+        if hasattr(self, "legacy_instance"):
+            return repr(self.legacy_instance)
+        else:
+            return super().__repr__()
 
 def from_openai(
     client: Union["OpenAIClient", "AzureOpenAIClient"], model_name: str
 ) -> OpenAI:
     return OpenAI(client, model_name)
+
+
+### Legacy !!!
+
+
+@dataclass(frozen=True)
+class OpenAIConfig:
+    """Represents the parameters of the OpenAI API.
+
+    The information was last fetched on 2023/11/20. We document below the
+    properties that are specific to the OpenAI API. Not all these properties are
+    supported by Outlines.
+
+    Parameters
+    ----------
+    model
+        The name of the model. Available models can be found on OpenAI's website.
+    frequency_penalty
+        Number between 2.0 and -2.0. Positive values penalize new tokens based on
+        their existing frequency in the text,
+    logit_bias
+        Modifies the likelihood of specified tokens to appear in the completion.
+        Number between -100 (forbid) and +100 (only allows).
+    n
+        The number of completions to return for each prompt.
+    presence_penalty
+        Similar to frequency penalty.
+    response_format
+        Specifies the format the model must output. `{"type": "json_object"}`
+        enables JSON mode.
+    seed
+        Two completions with the same `seed` value should return the same
+        completion. This is however not guaranteed.
+    stop
+        Up to 4 words where the API will stop the completion.
+    temperature
+        Number between 0 and 2. Higher values make the output more random, while
+        lower values make it more deterministic.
+    top_p
+        Number between 0 and 1. Parameter for nucleus sampling.
+    user
+        A unique identifier for the end-user.
+    """
+
+    model: str = ""
+    frequency_penalty: float = 0
+    logit_bias: Dict[int, int] = field(default_factory=dict)
+    max_tokens: Optional[int] = None
+    n: int = 1
+    presence_penalty: float = 0
+    response_format: Optional[Dict[str, str]] = None
+    seed: Optional[int] = None
+    stop: Optional[Union[str, List[str]]] = None
+    temperature: float = 1.0
+    top_p: int = 1
+    user: str = field(default_factory=str)
+
+
+class OpenAILegacy():
+    """An object that represents the OpenAI API."""
+
+    def __init__(
+        self,
+        client,
+        config,
+        system_prompt: Optional[str] = None,
+    ):
+        """Create an `OpenAI` instance.
+
+        This class supports the standard OpenAI API, the Azure OpeanAI API as
+        well as compatible APIs that rely on the OpenAI client.
+
+        Parameters
+        ----------
+        client
+            An instance of the API's async client.
+        config
+            An instance of `OpenAIConfig`. Can be useful to specify some
+            parameters that cannot be set by calling this class' methods.
+        """
+
+        self.client = client
+        self.config = config
+        self.system_prompt = system_prompt
+
+        # We count the total number of prompt and generated tokens as returned
+        # by the OpenAI API, summed over all the requests performed with this
+        # model instance.
+        self.prompt_tokens = 0
+        self.completion_tokens = 0
+
+        self.format_sequence = lambda x: x
+
+    def __call__(
+        self,
+        prompt: Union[str, List[str]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[List[str], str]] = None,
+        *,
+        system_prompt: Optional[str] = None,
+        temperature: Optional[float] = None,
+        samples: Optional[int] = None,
+    ):
+        """Call the OpenAI API to generate text.
+
+        Parameters
+        ----------
+        prompt
+            A string or list of strings that will be used to prompt the model
+        max_tokens
+            The maximum number of tokens to generate
+        stop_at
+            A string or array of strings which, such that the generation stops
+            when they are generated.
+        system_prompt
+            The content of the system message that precedes the user's prompt.
+        temperature
+            The value of the temperature used to sample tokens
+        samples
+            The number of completions to generate for each prompt
+        stop_at
+            Up to 4 words where the API will stop the completion.
+
+        """
+        if max_tokens is None:
+            max_tokens = self.config.max_tokens
+        if stop_at is None:
+            stop_at = self.config.stop
+        if temperature is None:
+            temperature = self.config.temperature
+        if samples is None:
+            samples = self.config.n
+
+        config = replace(
+            self.config,
+            max_tokens=max_tokens,
+            temperature=temperature,
+            n=samples,
+            stop=stop_at,
+        )  # type: ignore
+
+        response, prompt_tokens, completion_tokens = generate_chat(
+            prompt, system_prompt or self.system_prompt, self.client, config
+        )
+        self.prompt_tokens += prompt_tokens
+        self.completion_tokens += completion_tokens
+
+        return self.format_sequence(response)
+
+    def stream(self, *args, **kwargs):
+        raise NotImplementedError(
+            "Streaming is currently not supported for the OpenAI API"
+        )
+
+    def new_with_replacements(self, **kwargs):
+        new_instance = copy.copy(self)
+        new_instance.config = replace(new_instance.config, **kwargs)
+        return new_instance
+
+    def __str__(self):
+        return self.__class__.__name__ + " API"
+
+    def __repr__(self):
+        return str(self.config)
+
+
+@functools.partial(vectorize, signature="(),(),(),()->(s),(),()")
+async def generate_chat(
+    prompt: str,
+    system_prompt: Union[str, None],
+    client,
+    config: OpenAIConfig,
+):
+    """Call OpenAI's Chat Completion API.
+
+    Parameters
+    ----------
+    prompt
+        The prompt we use to start the generation. Passed to the model
+        with the "user" role.
+    system_prompt
+        The system prompt, passed to the model with the "system" role
+        before the prompt.
+    client
+        The API client
+    config
+        An `OpenAIConfig` instance.
+
+    Returns
+    -------
+    A tuple that contains the model's response(s) and usage statistics.
+
+    """
+    import numpy as np
+
+    @error_handler
+    @cache()
+    async def call_api(prompt, system_prompt, config):
+        responses = await client.chat.completions.create(
+            messages=system_message + user_message,
+            **asdict(config),  # type: ignore
+        )
+        return responses.model_dump()
+
+    system_message = (
+        [{"role": "system", "content": system_prompt}] if system_prompt else []
+    )
+    user_message = [{"role": "user", "content": prompt}]
+
+    responses = await call_api(prompt, system_prompt, config)
+
+    results = np.array(
+        [responses["choices"][i]["message"]["content"] for i in range(config.n)]
+    )
+    usage = responses["usage"]
+
+    return results, usage["prompt_tokens"], usage["completion_tokens"]
+
+
+def error_handler(api_call_fn: Callable) -> Callable:
+    """Handle OpenAI API errors and missing API key."""
+
+    def call(*args, **kwargs):
+        import openai
+
+        try:
+            return api_call_fn(*args, **kwargs)
+        except (
+            openai.APITimeoutError,
+            openai.InternalServerError,
+            openai.RateLimitError,
+        ) as e:
+            raise OSError(f"Could not connect to the OpenAI API: {e}")
+        except (
+            openai.AuthenticationError,
+            openai.BadRequestError,
+            openai.ConflictError,
+            openai.PermissionDeniedError,
+            openai.NotFoundError,
+            openai.UnprocessableEntityError,
+        ) as e:
+            raise e
+
+    return call
diff --git a/outlines/v0_legacy/models/__init__.py b/outlines/v0_legacy/models/__init__.py
new file mode 100644
index 000000000..15d3a944e
--- /dev/null
+++ b/outlines/v0_legacy/models/__init__.py
@@ -0,0 +1,21 @@
+from .exllamav2 import ExLlamaV2Model, exl2
+from .llamacpp import llamacpp
+from .mlxlm import mlxlm
+from .openai import azure_openai, openai
+from .transformers import mamba, transformers
+from .transformers_vision import TransformersVision, transformers_vision
+from .vllm import vllm
+
+__all__ = [
+    "ExLlamaV2Model",
+    "TransformersVision",
+    "azure_openai",
+    "exl2",
+    "llamacpp",
+    "mamba",
+    "mlxlm",
+    "openai",
+    "transformers",
+    "transformers_vision",
+    "vllm"
+]
diff --git a/outlines/v0_legacy/models/exllamav2.py b/outlines/v0_legacy/models/exllamav2.py
new file mode 100644
index 000000000..f208fc555
--- /dev/null
+++ b/outlines/v0_legacy/models/exllamav2.py
@@ -0,0 +1,400 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import (
+    TYPE_CHECKING,
+    Dict,
+    Iterator,
+    List,
+    Optional,
+    Tuple,
+    TypedDict,
+    Union
+)
+
+if TYPE_CHECKING:
+    import torch.LongTensor
+    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler
+
+
+def exllamav2_params_adapter(sampling_params: dict, generation_params: dict, kwargs: dict) -> dict:
+    """Argument names mapping is done in the model class for the `exllamav2` model."""
+    kwargs.update(**sampling_params, **generation_params)
+    return kwargs
+
+
+class ExllamaV2Params(TypedDict, total=False):
+    stop_conditions: Optional[List[Union[int, str]]]
+    seed: Optional[int]
+    gen_settings: "ExLlamaV2Sampler.Settings"
+    max_new_tokens: List[int]
+
+
+class OutlinesExLlamaV2Tokenizer:
+    def __init__(self, tokenizer):
+        self.exl2_tokenizer = tokenizer
+        self.vocabulary = self.exl2_tokenizer.get_piece_to_id_dict()
+        self.special_tokens = set(self.exl2_tokenizer.extended_piece_to_id)
+        self.eos_token_id = self.exl2_tokenizer.eos_token_id
+
+    def convert_token_to_string(self, token):
+        return token
+
+    def decode(self, token_ids: "torch.LongTensor") -> List[str]:
+        import torch
+
+        decoded = self.exl2_tokenizer.decode(
+            torch.tensor(token_ids),
+            decode_special_tokens=False,
+        )
+        if isinstance(decoded, str):
+            return [decoded]
+        return decoded
+
+
+class ExLlamaV2Model():
+    """Represents a `exl2` model."""
+
+    def __init__(
+        self,
+        generator: "ExLlamaV2DynamicGenerator",
+        tokenizer: "OutlinesExLlamaV2Tokenizer",
+        max_seq_len: int,
+    ):
+        warnings.warn("""
+            The `ExLlamaV2Model` model is deprecated starting from v1.0.0.
+            Support for it will be removed in v1.5.0.
+            As the `exllamav2` library is not compatible with some key features
+            of `outlines`, we decided to remove support for it.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        self.generator = generator
+        self.tokenizer = tokenizer
+        self.max_seq_len = max_seq_len
+
+    def __call__(self, *args, **kwargs):
+        raise NotImplementedError(
+            "As a v0 model, the `exllamav2` model does not support the "
+            + "`__call__` method."
+        )
+
+    def stream(self, *args, **kwargs):
+        raise NotImplementedError(
+            "As a v0 model, the `exllamav2` model does not support the "
+            + "`stream` method."
+        )
+
+    def prepare_generation_parameters(
+        self,
+        prompts: Union[str, List[str]],
+        logits_processor,
+        kwargs
+    ) -> Tuple[Dict, Union[str, List[str]]]:
+        """Prepare the generation parameters.
+
+        `exllamav2` uses different default values
+
+        """
+        from exllamav2.generator import ExLlamaV2Sampler
+
+        exllamav2_params = {}
+
+        if isinstance(prompts, str):
+            prompts = [prompts]
+
+        max_tokens = kwargs.get("max_tokens")
+        if max_tokens is None:
+            max_tokens = []
+            for prompt in prompts:
+                ids = self.generator.tokenizer.encode(
+                    prompt, encode_special_tokens=True
+                )
+                prompt_tokens = ids.shape[-1]
+                max_tokens.append(self.max_seq_len - prompt_tokens)
+            exllamav2_params["max_new_tokens"] = max_tokens
+        else:
+            exllamav2_params["max_new_tokens"] = [
+                max_tokens for _ in range(len(prompts))
+            ]
+
+        stop_at = kwargs.get("stop_at")
+        stop_conditions = [self.generator.tokenizer.eos_token_id]
+        if isinstance(stop_at, str):
+            stop_conditions.append(stop_at)
+        elif isinstance(stop_at, list):
+            for stop_at in stop_at:
+                stop_conditions.append(stop_at)
+        exllamav2_params["stop_conditions"] = stop_conditions
+
+        exllamav2_params["seed"] = kwargs.get("seed")
+
+        gen_settings = ExLlamaV2Sampler.Settings()
+
+        temperature = kwargs.get("temperature")
+        if temperature is not None:
+            gen_settings.temperature = temperature
+
+        top_p = kwargs.get("top_p")
+        if top_p is not None:
+            gen_settings.top_p = top_p
+
+        top_k = kwargs.get("top_k")
+        if top_k is not None:
+            gen_settings.top_k = top_k
+
+        gen_settings.logits_processor = logits_processor
+
+        exllamav2_params["gen_settings"] = gen_settings
+
+        num_samples = kwargs.get("num_samples")
+        if num_samples is not None and num_samples > 1:
+            prompts = prompts * num_samples
+            exllamav2_params["max_new_tokens"] = (
+                exllamav2_params["max_new_tokens"] * num_samples
+            )
+
+        if len(prompts) == 1:
+            prompts = prompts[0]
+
+        return exllamav2_params, prompts
+
+    def reformat_output(
+        self, output: Union[str, List[str]], sampling_parameters
+    ):
+        """
+        The purpose of this function is to reformat the output from exllamav2's output format to outline's output format.
+
+        For exllamav2, it mainly accepts only a list or a string(they also do cfg sampling with tuples but we will ignore this for now).
+        The exllamav2's logic is:
+
+        1. If the prompt is a string, return a string. This is the same as outlines
+        2. If a prompt is a list, return a list. This is not the same as outlines output in that if the list is only one element, the string is expected to be outputted.
+        3. There is no such thing as num_samples, so the prompts had to be duplicated by num_samples times. Then, we had the function output a list of lists
+        """
+        if isinstance(output, str):
+            return output
+        if len(output) == 1:
+            return output[0]
+        if sampling_parameters.num_samples > 1:
+            if len(output) == sampling_parameters.num_samples:
+                return output
+            assert len(output) % sampling_parameters.num_samples == 0
+            num_items_per_sample = len(output) // sampling_parameters.num_samples
+            new_output = []
+            for i in range(sampling_parameters.num_samples):
+                curr_sample = []
+                for j in range(num_items_per_sample):
+                    curr_sample.append(output[i * num_items_per_sample + j])
+                new_output.append(curr_sample)
+            return new_output
+        return output
+
+    def generate(
+        self,
+        prompts: Union[str, List[str]],
+        logits_processor,
+        **kwargs
+    ) -> Union[str, List[str]]:
+        exllamav2_params, prompts = self.prepare_generation_parameters(
+            prompts,
+            logits_processor,
+            kwargs
+        )
+        """
+        In exllamav2, it needs the max amount of new tokens generated.
+        The reason exllamav2_params["max_new_tokens"] is a list is because in prepare_generation_parameters
+        the max amount of tokens that can be generated by the model for each prompt(by encoding with tokenizer) is calculated.
+        The minimum is picked because otherwise it might be possible for one of the
+        prompts to exceed the max sequence length.
+        """
+        output = self.generator.generate(
+            prompt=prompts,
+            gen_settings=exllamav2_params["gen_settings"],
+            max_new_tokens=min(exllamav2_params["max_new_tokens"]),
+            completion_only=True,
+            encode_special_tokens=True,
+            stop_conditions=exllamav2_params["stop_conditions"],
+            add_bos=False,
+            seed=exllamav2_params["seed"],
+        )
+
+        return self.reformat_output(output, kwargs)
+
+    def generate_stream(
+        self,
+        prompts: Union[str, List[str]],
+        logits_processor,
+        **kwargs
+    ) -> Iterator[Union[str, List[str]]]:
+        from exllamav2.generator import ExLlamaV2DynamicJob
+
+        exllamav2_params, prompts = self.prepare_generation_parameters(
+            prompts,
+            logits_processor,
+            kwargs
+        )
+
+        order = {}
+        if isinstance(prompts, str):
+            prompts = [prompts]
+        batch_size = len(prompts)
+        seed = exllamav2_params["seed"]
+        for idx, p in enumerate(prompts):
+            input_ids = self.generator.tokenizer.encode(
+                p, encode_special_tokens=True, add_bos=False
+            )
+
+            job = ExLlamaV2DynamicJob(
+                input_ids=input_ids,
+                max_new_tokens=exllamav2_params["max_new_tokens"][idx],
+                min_new_tokens=0,
+                seed=seed,
+                stop_conditions=exllamav2_params["stop_conditions"],
+                gen_settings=exllamav2_params["gen_settings"],
+                token_healing=False,
+                decode_special_tokens=False,
+            )
+
+            if seed is not None:
+                seed += 1
+
+            serial = self.generator.enqueue(job)
+            order[serial] = idx
+
+        # Collect outputs until all jobs finish
+
+        next_text = [""] * batch_size
+
+        def token_generator() -> Iterator[str]:
+            while self.generator.num_remaining_jobs():
+                results = self.generator.iterate()
+                for r in results:
+                    idx = order[r["serial"]]
+                    if r["stage"] == "streaming":
+                        text = r.get("text", "")
+                        next_text[idx] = text
+                    if r["eos"]:
+                        next_text[idx] = ""
+                yield self.reformat_output(next_text, kwargs)
+            return
+
+        return token_generator()
+
+
+def exl2(
+    model_path: str,
+    draft_model_path: Optional[str] = None,
+    max_seq_len: Optional[int] = None,
+    cache_q4: bool = False,
+    paged: bool = True,
+    max_chunk_size: Optional[int] = None,
+) -> ExLlamaV2Model:
+    """
+    Load an ExLlamaV2 model.
+
+    Parameters
+    ----------
+    model_path (str)
+        Path to the model directory.
+    device (str)
+        Device to load the model on. Pass in 'cuda' for GPU or 'cpu' for CPU
+    max_seq_len (Optional[int], optional)
+        Maximum sequence length. Defaults to None.
+    scale_pos_emb (Optional[float], optional)
+        Scale factor for positional embeddings. Defaults to None.
+    scale_alpha_value (Optional[float], optional)
+        Scale alpha value. Defaults to None.
+    no_flash_attn (Optional[bool], optional)
+        Disable flash attention. Defaults to None.
+    num_experts_per_token (Optional[int], optional)
+        Number of experts per token. Defaults to None.
+    cache_q4 (bool, optional)
+        Use Q4 cache. Defaults to False.
+    tokenizer_kwargs (dict, optional)
+        Additional keyword arguments for the tokenizer. Defaults to {}.
+    gpu_split (str)
+        \"auto\", or VRAM allocation per GPU in GB. Auto will use exllama's autosplit feature
+    low_mem (bool, optional)
+        Enable VRAM optimizations, potentially trading off speed
+    verbose (bool, optional)
+        Enable if you want debugging statements
+
+    Returns
+    -------
+    An `ExLlamaV2Model` instance.
+
+    Raises
+    ------
+    `ImportError` if the `exllamav2` library is not installed.
+
+    """
+    try:
+        from exllamav2 import (
+            ExLlamaV2,
+            ExLlamaV2Cache,
+            ExLlamaV2Cache_Q4,
+            ExLlamaV2Config,
+            ExLlamaV2Tokenizer,
+        )
+        from exllamav2.generator import ExLlamaV2DynamicGenerator
+
+    except ImportError:
+        raise ImportError(
+            "The `exllamav2`, `transformers` and `torch` libraries needs to be installed in order to use `exllamav2` models. "
+            "Please run `pip install transformers torch git+https://github.com/lapp0/exllamav2@sampler-logits-processor` "
+            "Documentation: https://dottxt-ai.github.io/outlines/latest/reference/models/exllamav2/"
+        )
+    config = ExLlamaV2Config(model_path)
+    if max_chunk_size is not None:
+        config.max_input_len = max_chunk_size
+        config.max_attention_size = max_chunk_size**2
+
+    config.arch_compat_overrides()
+    model = ExLlamaV2(config)
+    if max_seq_len is None:
+        max_seq_len = -1
+    if cache_q4:
+        cache = ExLlamaV2Cache_Q4(model, max_seq_len=max_seq_len, lazy=True)
+    else:
+        cache = ExLlamaV2Cache(model, max_seq_len=max_seq_len, lazy=True)
+    model.load_autosplit(cache, progress=True)
+
+    print("Loading tokenizer...")
+    tokenizer = ExLlamaV2Tokenizer(config)
+    max_batch_size = 4 if paged else 1
+
+    draft_model = None
+    draft_cache = None
+    if draft_model_path is not None:
+        draft_config = ExLlamaV2Config(draft_model_path)
+        draft_model = ExLlamaV2(draft_config)
+
+        if cache_q4:
+            draft_cache = ExLlamaV2Cache_Q4(
+                draft_model, max_seq_len=max_seq_len, lazy=True
+            )
+        else:
+            draft_cache = ExLlamaV2Cache(
+                draft_model, max_seq_len=max_seq_len, lazy=True
+            )
+
+    # Initialize the generator with all default parameters
+    generator = ExLlamaV2DynamicGenerator(
+        model=model,
+        cache=cache,
+        draft_model=draft_model,
+        draft_cache=draft_cache,
+        tokenizer=tokenizer,
+        max_batch_size=max_batch_size,
+        use_ngram_draft=False,
+        max_chunk_size=max_chunk_size,
+        paged=paged,
+    )
+    max_seq_len = cache.max_seq_len
+
+    outlines_tokenizer = OutlinesExLlamaV2Tokenizer(tokenizer)
+    outlines_exl2_model = ExLlamaV2Model(generator, outlines_tokenizer, max_seq_len)
+    return outlines_exl2_model
diff --git a/outlines/v0_legacy/models/llamacpp.py b/outlines/v0_legacy/models/llamacpp.py
new file mode 100644
index 000000000..86233a63b
--- /dev/null
+++ b/outlines/v0_legacy/models/llamacpp.py
@@ -0,0 +1,147 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import Optional
+
+from outlines.models import LlamaCpp
+
+
+def llamacpp_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `llamacpp` model
+    to the new `Generator` API.
+    """
+    # generation_params
+    if (
+        generation_params.get("stop_at") is not None
+        and kwargs.get("stop") is None
+    ):
+        kwargs["stop"] = generation_params["stop_at"]
+
+    if (
+        generation_params.get("seed") is not None
+        and kwargs.get("seed") is None
+    ):
+        kwargs["seed"] = generation_params["seed"]
+
+    if (
+        generation_params.get("max_tokens") is not None
+        and kwargs.get("max_tokens") is None
+    ):
+        kwargs["max_tokens"] = generation_params["max_tokens"]
+    else:
+        kwargs["max_tokens"] = -1
+
+    # sampling_params
+    if sampling_params.get("sampler") == "beam_search":
+        raise NotImplementedError(
+            "The `llama_cpp_python` library does not support Beam Search."
+        )
+
+    if sampling_params.get("num_samples") != 1:
+        raise NotImplementedError(
+            "The `llama_cpp_python` library does not allow to take several "
+            + "samples."
+        )
+
+    if (
+        sampling_params.get("top_p") is not None
+        and kwargs.get("top_p") is None
+    ):
+        kwargs["top_p"] = sampling_params["top_p"]
+    else:
+        kwargs["top_p"] = 1.0
+
+    if (
+        sampling_params.get("top_k") is not None
+        and kwargs.get("top_k") is None
+    ):
+        kwargs["top_k"] = sampling_params["top_k"]
+    else:
+        kwargs["top_k"] = -1
+
+    if (
+        sampling_params.get("temperature") is not None
+        and kwargs.get("temperature") is None
+    ):
+        kwargs["temperature"] = sampling_params["temperature"]
+    else:
+        kwargs["temperature"] = 1.0
+
+    # model specific params
+    if kwargs.get("min_p") is None:
+        kwargs["min_p"] = 0.0
+
+    if kwargs.get("repeat_penalty") is None:
+        kwargs["repeat_penalty"] = 1.0
+
+    # The choice to stream or not should happen via the high-level API
+    kwargs["stream"] = False
+
+    return kwargs
+
+
+def llamacpp(
+    repo_id: str, filename: Optional[str] = None, **llamacpp_model_params
+) -> LlamaCpp:
+    """Load a model from the `llama-cpp-python` library.
+
+    We use the `Llama.from_pretrained` classmethod that downloads models
+    directly from the HuggingFace hub, instead of asking users to specify
+    a path to the downloaded model. One can still load a local model
+    by initializing `llama_cpp.Llama` directly.
+
+    Parameters
+    ----------
+    repo_id
+        The name of the model repository.
+    filename:
+        A filename of glob pattern to match the model file in the repo.
+    llama_cpp_model_params
+        Llama-specific model parameters. See the `llama-cpp-python` documentation
+        for the full list: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__
+
+    """
+    from llama_cpp import Llama
+
+    warnings.warn("""
+        The `llamacpp` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should instantiate a `LlamaCpp` model with the
+        `outlines.from_llamacpp` function that takes a `llama_cpp.Llama`
+        instance as argument. For example:
+        ```python
+        from llama_cpp import Llama
+        from outlines import from_llamacpp
+        model = from_llamacpp(Llama.from_pretrained(
+            repo_id="<repo_id>",
+            filename="<filename>",
+        ))
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    # Default to using the model's full context length
+    if "n_ctx" not in llamacpp_model_params:
+        llamacpp_model_params["n_ctx"] = 0
+
+    if "verbose" not in llamacpp_model_params:
+        llamacpp_model_params["verbose"] = False
+
+    # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
+    if "tokenizer" not in llamacpp_model_params:
+        warnings.warn(
+            "The pre-tokenizer in `llama.cpp` handles unicode improperly "
+            + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
+            + "Outlines may raise a `RuntimeError` when building the regex index.\n"
+            + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
+            + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
+        )
+
+    model = Llama.from_pretrained(repo_id, filename, **llamacpp_model_params)
+
+    return LlamaCpp(model)
diff --git a/outlines/v0_legacy/models/mlxlm.py b/outlines/v0_legacy/models/mlxlm.py
new file mode 100644
index 000000000..76ba44465
--- /dev/null
+++ b/outlines/v0_legacy/models/mlxlm.py
@@ -0,0 +1,153 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+from typing import Optional
+import warnings
+
+from outlines.models import MLXLM
+
+
+def mlxlm_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `mlxlm` model
+    to the new `Generator` API.
+    """
+    # generation_params
+    if generation_params.get("stop_at") is not None:
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support `stop_at`."
+        )
+
+    if generation_params.get("seed") is not None:
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support `seed`."
+        )
+
+    if (
+        generation_params.get("max_tokens") is not None
+        and kwargs.get("max_tokens") is None
+    ):
+        kwargs["max_tokens"] = generation_params["max_tokens"]
+    else:
+        kwargs["max_tokens"] = int(1e9)
+
+    # sampling_params
+    if sampling_params.get("sampler") == "beam_search":
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support Beam Search."
+        )
+
+    if sampling_params.get("num_samples") != 1:
+        raise NotImplementedError(
+            "The `mlx-lm` library does not allow to take several samples."
+        )
+
+    if (
+        sampling_params.get("top_k") is not None
+        or kwargs.get("top_k") is not None
+    ):
+        raise NotImplementedError(
+            "The `mlx-lm` library does not support top_k."
+        )
+
+    if (
+        sampling_params.get("top_p") is not None
+        or kwargs.get("top_p") is not None
+    ):
+        kwargs.pop("top_p", None)
+        warnings.warn(
+            """
+            The `top_p` parameter is not available anymore.
+            The argument will be ignored. Sorry for the inconvenience.
+            Please migrate to the v1 of Outlines.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+
+    if (
+        sampling_params.get("temperature") is not None
+        or kwargs.get("temperature") is not None
+    ):
+        kwargs.pop("temperature", None)
+        warnings.warn(
+            """
+            The `temperature` parameter is not available anymore.
+            The argument will be ignored. Sorry for the inconvenience.
+            Please migrate to the v1 of Outlines.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+
+    return kwargs
+
+
+def mlxlm(
+    model_name: str,
+    tokenizer_config: dict = {},
+    model_config: dict = {},
+    adapter_path: Optional[str] = None,
+    lazy: bool = False,
+):
+    """Instantiate a model from the `mlx_lm` library and its tokenizer.
+
+    Signature adapted from
+    https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L422
+
+    Parameters
+    ----------
+    Args:
+        path_or_hf_repo (Path): The path or the huggingface repository to load the model from.
+        tokenizer_config (dict, optional): Configuration parameters specifically for the tokenizer.
+            Defaults to an empty dictionary.
+        model_config(dict, optional): Configuration parameters specifically for the model.
+            Defaults to an empty dictionary.
+        adapter_path (str, optional): Path to the LoRA adapters. If provided, applies LoRA layers
+            to the model. Default: ``None``.
+        lazy (bool): If False eval the model parameters to make sure they are
+            loaded in memory before returning, otherwise they will be loaded
+            when needed. Default: ``False``
+
+    Returns
+    -------
+    A `MLXLM` model instance.
+
+    """
+    warnings.warn("""
+        The `mlxlm` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should instantiate a `MLXLM` model with the
+        `outlines.from_mlxlm` function that takes a mlxlm model
+        and a tokenizer as arguments. For example:
+        ```python
+        from mlx_lm import load
+        from outlines import from_mlxlm
+        model = from_mlxlm(
+            load("<model_name>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    try:
+        import mlx.core as mx
+        import mlx_lm
+    except ImportError:
+        raise ImportError(
+            "The `mlx_lm` library needs to be installed in order to use `mlx_lm` models."
+        )
+    if not mx.metal.is_available():
+        raise RuntimeError("You cannot use `mlx_lm` without Apple Silicon (Metal)")
+
+    model, tokenizer = mlx_lm.load(
+        model_name,
+        tokenizer_config=tokenizer_config,
+        model_config=model_config,
+        adapter_path=adapter_path,
+        lazy=lazy,
+    )
+    return MLXLM(model, tokenizer)
diff --git a/outlines/v0_legacy/models/openai.py b/outlines/v0_legacy/models/openai.py
new file mode 100644
index 000000000..46546b003
--- /dev/null
+++ b/outlines/v0_legacy/models/openai.py
@@ -0,0 +1,64 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import functools
+from dataclasses import replace
+from typing import Optional
+
+from outlines.models.openai import OpenAI, OpenAIConfig
+
+
+@functools.singledispatch
+def openai(model_or_client, *args, **kwargs):
+    if len(args) == 1 and isinstance(args[0], OpenAIConfig):
+        return OpenAI(model_or_client, args[0], **kwargs)
+    elif kwargs.get("config"):
+        return OpenAI(model_or_client, kwargs.pop("config"), **kwargs)
+    else:
+        return OpenAI(model_or_client, OpenAIConfig(), **kwargs)
+
+
+@openai.register(str)
+def openai_model(
+    model_name: str,
+    config: Optional[OpenAIConfig] = None,
+    **openai_client_params,
+):
+    try:
+        from openai import AsyncOpenAI
+    except ImportError:
+        raise ImportError(
+            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
+        )
+
+    if config is not None:
+        config = replace(config, model=model_name)  # type: ignore
+    else:
+        config = OpenAIConfig(model=model_name)
+
+    client = AsyncOpenAI(**openai_client_params)
+
+    return OpenAI(client, config)
+
+
+def azure_openai(
+    deployment_name: str,
+    model_name: Optional[str] = None,
+    config: Optional[OpenAIConfig] = None,
+    **azure_openai_client_params,
+):
+    try:
+        from openai import AsyncAzureOpenAI
+    except ImportError:
+        raise ImportError(
+            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
+        )
+
+    if config is not None:
+        config = replace(config, model=deployment_name)  # type: ignore
+    if config is None:
+        config = OpenAIConfig(model=deployment_name)
+
+    client = AsyncAzureOpenAI(**azure_openai_client_params)
+
+    return OpenAI(client, config)
diff --git a/outlines/v0_legacy/models/transformers.py b/outlines/v0_legacy/models/transformers.py
new file mode 100644
index 000000000..b275eca12
--- /dev/null
+++ b/outlines/v0_legacy/models/transformers.py
@@ -0,0 +1,196 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import Optional
+
+from transformers import set_seed
+
+from outlines.models import Transformers
+
+
+def transformers_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `transformers` model
+    to the new `Generator` API.
+    """
+    # generation_params
+    if (
+        generation_params.get("stop_at") is not None
+        and kwargs.get("stop_strings") is None
+    ):
+        kwargs["stop_strings"] = generation_params["stop_at"]
+
+    if (
+        generation_params.get("max_tokens") is not None
+        and kwargs.get("max_new_tokens") is None
+    ):
+        kwargs["max_new_tokens"] = generation_params["max_tokens"]
+    else:
+        kwargs["max_new_tokens"] = int(2**30)
+
+    if (
+        generation_params.get("seed") is not None
+        or kwargs.get("seed") is not None
+    ):
+        seed = (
+            generation_params.get("seed")
+            if generation_params.get("seed") is not None
+            else kwargs.get("seed")
+        )
+        set_seed(seed) # global seed
+
+    # sampling_params
+    if sampling_params.get("sampler") == "multinomial":
+        kwargs["do_sample"] = True
+    elif sampling_params.get("sampler") == "beam_search":
+        kwargs["num_beams"] = sampling_params.get("num_samples", 1)
+
+    if (
+        sampling_params.get("num_samples") is not None
+        and kwargs.get("num_return_sequences") is None
+    ):
+        kwargs["num_return_sequences"] = sampling_params["num_samples"]
+    else:
+        kwargs["num_return_sequences"] = 1
+
+    if (
+        sampling_params.get("top_p") is not None
+        and kwargs.get("top_p") is None
+    ):
+        kwargs["top_p"] = sampling_params["top_p"]
+
+    if (
+        sampling_params.get("top_k") is not None
+        and kwargs.get("top_k") is None
+    ):
+        kwargs["top_k"] = sampling_params["top_k"]
+
+    if (
+        sampling_params.get("temperature") is not None
+        and kwargs.get("temperature") is None
+    ):
+        kwargs["temperature"] = sampling_params["temperature"]
+
+    return kwargs
+
+
+def transformers(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    tokenizer_kwargs: dict = {},
+    model_class=None,
+    tokenizer_class=None,
+):
+    """Instantiate a model from the `transformers` library and its tokenizer.
+
+    Parameters
+    ----------
+    model_name
+        The name of the model as listed on Hugging Face's model page.
+    device
+        The device(s) on which the model should be loaded. This overrides
+        the `device_map` entry in `model_kwargs` when provided.
+    model_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the model.
+    tokenizer_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the tokenizer.
+
+    Returns
+    -------
+    A `TransformersModel` model instance.
+
+    """
+    warnings.warn("""
+        The `transformers` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should instantiate a `Transformers` model with the
+        `outlines.from_transformers` function that takes a transformers model
+        and a tokenizer as arguments. For example:
+        ```python
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+        from outlines import from_transformers
+        model = from_transformers(
+            AutoModelForCausalLM.from_pretrained("<model_name>"),
+            AutoTokenizer.from_pretrained("<model_name>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if model_class is None or tokenizer_class is None:
+        try:
+            from transformers import AutoModelForCausalLM, AutoTokenizer
+        except ImportError:
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if model_class is None:
+        model_class = AutoModelForCausalLM
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
+
+    if device is not None:
+        model_kwargs["device_map"] = device
+
+    model = model_class.from_pretrained(model_name, **model_kwargs)
+
+    tokenizer_kwargs.setdefault("padding_side", "left")
+    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
+
+    return Transformers(model, tokenizer)
+
+
+def mamba(
+    model_name: str,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    tokenizer_kwargs: dict = {},
+):
+    warnings.warn("""
+        The `mamba` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should instantiate a `Transformers` model with the
+        `outlines.from_transformers` function that takes a transformers model
+        and a tokenizer as arguments. For Mamba, you would use a mamba model
+        and a mamba tokenizer. For example:
+        ```python
+        from transformers import MambaForCausalLM, MambaTokenizer
+        from outlines import from_transformers
+        model = from_transformers(
+            MambaForCausalLM.from_pretrained("<repo_id>"),
+            MambaTokenizer.from_pretrained("<repo_id>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    try:
+        from transformers import MambaForCausalLM
+
+    except ImportError:
+        raise ImportError(
+            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
+        )
+
+    # We do not want to raise the deprecation warning through the transformers function
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            category=DeprecationWarning,
+        )
+        return transformers(
+            model_name=model_name,
+            device=device,
+            model_kwargs=model_kwargs,
+            tokenizer_kwargs=tokenizer_kwargs,
+            model_class=MambaForCausalLM,
+    )
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
new file mode 100644
index 000000000..96edf313e
--- /dev/null
+++ b/outlines/v0_legacy/models/transformers_vision.py
@@ -0,0 +1,117 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+from typing import TYPE_CHECKING, Optional
+import warnings
+
+from outlines.models.transformers import (
+    Transformers,
+    TransformersMultiModalTypeAdapter,
+)
+
+if TYPE_CHECKING:
+    from outlines.processors import OutlinesLogitsProcessor
+
+
+class TransformersVision(Transformers):
+    def __init__(self, model, tokenizer, processor):
+        self.processor = processor
+        self.processor.padding_side = "left"
+        self.processor.pad_token = "[PAD]"
+
+        super().__init__(model, tokenizer)
+
+        self.type_adapter = TransformersMultiModalTypeAdapter()
+
+
+    def _prepare_model_inputs(self, model_input, output_type):
+        model_input = self.type_adapter.format_input(model_input)
+        inputs = self.processor(
+            **model_input, padding=True, return_tensors="pt"
+        ).to(self.model.device)
+
+        return model_input["text"], inputs
+
+
+def transformers_vision(
+    model_name: str,
+    model_class,
+    device: Optional[str] = None,
+    model_kwargs: dict = {},
+    processor_kwargs: dict = {},
+    tokenizer_class=None,
+    processor_class=None,
+):
+    """Instantiate a model from the `transformers` library and its tokenizer.
+
+    Parameters
+    ----------
+    model_name
+        The name of the model as listed on Hugging Face's model page.
+    model_class
+        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
+        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
+    device
+        The device(s) on which the model should be loaded. This overrides
+        the `device_map` entry in `model_kwargs` when provided.
+    model_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the model.
+    processor_kwargs
+        A dictionary that contains the keyword arguments to pass to the
+        `from_pretrained` method when loading the processor.
+
+    Returns
+    -------
+    A `TransformersModel` model instance.
+
+    """
+    warnings.warn("""
+        The `transformers_vision` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should instantiate a `TransformersMultiModal` model with
+        the `outlines.from_transformers` function that takes a transformers
+        model and a processor as arguments. For example:
+        ```python
+        from transformers import AutoModelForCausalLM, AutoProcessor
+        from outlines import from_transformers
+        model = from_transformers(
+            AutoModelForCausalLM.from_pretrained("<model_name>"),
+            AutoProcessor.from_pretrained("<model_name>")
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if processor_class is None or tokenizer_class is None:
+        try:
+            from transformers import AutoProcessor, AutoTokenizer
+        except ImportError:
+            raise ImportError(
+                "The `transformers` library needs to be installed in order to use `transformers` models."
+            )
+    if processor_class is None:
+        processor_class = AutoProcessor
+    if tokenizer_class is None:
+        tokenizer_class = AutoTokenizer
+
+    if device is not None:
+        model_kwargs["device_map"] = device
+
+    model = model_class.from_pretrained(model_name, **model_kwargs)
+
+    processor_kwargs.setdefault("padding_side", "left")
+    processor_kwargs.setdefault("pad_token", "[PAD]")
+    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+    if tokenizer_class is None:
+        if getattr(processor, "tokenizer", None):
+            tokenizer = processor.tokenizer
+        else:
+            tokenizer = AutoTokenizer.from_pretrained(model_name, **processor_kwargs)
+    else:
+        tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
+
+    return TransformersVision(model, tokenizer, processor)
diff --git a/outlines/v0_legacy/models/vllm.py b/outlines/v0_legacy/models/vllm.py
new file mode 100644
index 000000000..6b3309724
--- /dev/null
+++ b/outlines/v0_legacy/models/vllm.py
@@ -0,0 +1,101 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models import VLLM
+
+
+def vllm_params_adapter(
+    sampling_params: dict, generation_params: dict, kwargs: dict
+) -> dict:
+    """Adapt the parameters of the legacy generator for the `vllm` model
+    to the new `Generator` API.
+    """
+    from vllm.sampling_params import SamplingParams
+
+    # vllm has a `sampling_params` argument that is a `SamplingParams` object
+    # this is unrelated to the `sampling_params` argument in the legacy
+    # `Generator` API
+    sampling_params_kwarg = kwargs.get("sampling_params", SamplingParams())
+
+    # generation_params
+    if generation_params.get("stop_at") is not None:
+        if isinstance(generation_params["stop_at"], str):
+            sampling_params_kwarg.stop_strings = [generation_params["stop_at"]]
+        else:
+            sampling_params_kwarg.stop_strings = generation_params["stop_at"]
+
+    if generation_params.get("seed") is not None:
+        sampling_params_kwarg.seed = generation_params["seed"]
+
+    if generation_params.get("max_tokens") is not None:
+        sampling_params_kwarg.max_tokens = generation_params["max_tokens"]
+
+    # sampling_params
+    if sampling_params.get("sampler") == "beam_search":
+        sampling_params_kwarg.use_beam_search = True
+
+    if (
+        sampling_params.get("num_samples") is not None
+        and sampling_params_kwarg.n == 1
+    ):
+        sampling_params_kwarg.n = sampling_params["num_samples"]
+        sampling_params_kwarg.best_of = sampling_params["num_samples"]
+
+    if (
+        sampling_params.get("top_p") is not None
+        and sampling_params_kwarg.top_p == 1.0
+    ):
+        sampling_params_kwarg.top_p = sampling_params["top_p"]
+
+    if (
+        sampling_params.get("top_k") is not None
+        and sampling_params_kwarg.top_k == -1
+    ):
+        sampling_params_kwarg.top_k = sampling_params["top_k"]
+        if sampling_params.get("top_k") == 1:
+            sampling_params_kwarg.repetition_penalty = 0
+
+    if (
+        sampling_params.get("temperature") is not None
+        and sampling_params_kwarg.temperature == 1.0
+    ):
+        sampling_params_kwarg.temperature = sampling_params["temperature"]
+
+    return {"sampling_params": sampling_params_kwarg, **kwargs}
+
+
+def vllm(model_name: str, **vllm_model_params):
+    """Load a vLLM model.
+
+    Parameters
+    ---------
+    model_name
+        The name of the model to load from the HuggingFace hub.
+    vllm_model_params
+        vLLM-specific model parameters. See the vLLM code for the full list:
+        https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/llm.py
+
+    """
+    warnings.warn("""
+        The `vllm` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should instantiate a `VLLM` model with the
+        `outlines.from_vllm` function that takes a vLLM model name as argument.
+        For example:
+        ```python
+        from vllm import LLM
+        from outlines import from_vllm
+        model = from_vllm(LLM("<model_name>"))
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    from vllm import LLM
+
+    model = LLM(model_name, **vllm_model_params)
+
+    return VLLM(model)

From fb68b2b05e180204352f07c07a583de4d55a7f2c Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 15:43:42 +0200
Subject: [PATCH 302/505] Restore the generate functions

---
 outlines/generate/json.py               | 123 -----------------
 outlines/v0_legacy/generate/__init__.py |  17 +++
 outlines/v0_legacy/generate/api.py      | 176 ++++++++++++++++++++++++
 outlines/v0_legacy/generate/cfg.py      |  79 +++++++++++
 outlines/v0_legacy/generate/choice.py   | 123 +++++++++++++++++
 outlines/v0_legacy/generate/format.py   |  69 ++++++++++
 outlines/v0_legacy/generate/fsm.py      |  76 ++++++++++
 outlines/v0_legacy/generate/json.py     | 145 +++++++++++++++++++
 outlines/v0_legacy/generate/regex.py    |  66 +++++++++
 outlines/v0_legacy/generate/text.py     |  62 +++++++++
 10 files changed, 813 insertions(+), 123 deletions(-)
 delete mode 100644 outlines/generate/json.py
 create mode 100644 outlines/v0_legacy/generate/__init__.py
 create mode 100644 outlines/v0_legacy/generate/api.py
 create mode 100644 outlines/v0_legacy/generate/cfg.py
 create mode 100644 outlines/v0_legacy/generate/choice.py
 create mode 100644 outlines/v0_legacy/generate/format.py
 create mode 100644 outlines/v0_legacy/generate/fsm.py
 create mode 100644 outlines/v0_legacy/generate/json.py
 create mode 100644 outlines/v0_legacy/generate/regex.py
 create mode 100644 outlines/v0_legacy/generate/text.py

diff --git a/outlines/generate/json.py b/outlines/generate/json.py
deleted file mode 100644
index ab05df1f4..000000000
--- a/outlines/generate/json.py
+++ /dev/null
@@ -1,123 +0,0 @@
-import json as pyjson
-from enum import Enum
-from functools import singledispatch
-from typing import Callable, Optional, Union
-
-from genson import SchemaBuilder
-from outlines_core.fsm.json_schema import build_regex_from_schema
-from pydantic import BaseModel
-
-from outlines.fsm.json_schema import get_schema_from_enum, get_schema_from_signature
-from outlines.generate.api import SequenceGeneratorAdapter
-from outlines.models import OpenAI
-from outlines.samplers import Sampler, multinomial
-
-from .regex import regex
-
-
-@singledispatch
-def json(
-    model,
-    schema_object: Union[str, object, Callable],
-    sampler: Sampler = multinomial(),
-    whitespace_pattern: Optional[str] = None,
-) -> SequenceGeneratorAdapter:
-    """
-    Generate structured JSON data with a `Transformer` model based on a specified JSON Schema.
-
-    Parameters
-    ----------
-    model:
-        An instance of `Transformer` that represents a model from the
-        `transformers` library.
-    schema_object:
-        The JSON Schema to generate data for. Can be a JSON string, a Pydantic model, or a callable
-        that returns a JSON schema.
-    sampler:
-        The sampling algorithm to use to generate token ids from the logits
-        distribution.
-    whitespace_pattern
-        Pattern to use for JSON syntactic whitespace (doesn't impact string literals)
-        Example: allow only a single space or newline with `whitespace_pattern=r"[\n ]?"`
-
-    Returns
-    -------
-    A `SequenceGenerator` instance that generates text constrained by the schema_object and
-    transforms the result if BaseModel is used.
-
-    """
-    if isinstance(schema_object, type(BaseModel)):
-        schema = pyjson.dumps(schema_object.model_json_schema())
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: schema_object.parse_raw(x)
-    elif isinstance(schema_object, type(Enum)):
-        schema = pyjson.dumps(get_schema_from_enum(schema_object))
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    elif isinstance(schema_object, SchemaBuilder):
-        schema = schema_object.to_json()
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    elif callable(schema_object):
-        schema = pyjson.dumps(get_schema_from_signature(schema_object))
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    elif isinstance(schema_object, str):
-        schema = schema_object
-        regex_str = build_regex_from_schema(schema, whitespace_pattern)
-        generator = regex(model, regex_str, sampler)
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    else:
-        raise ValueError(
-            f"Cannot parse schema {schema_object}. The schema must be either "
-            + "a Pydantic object, a function or a string that contains the JSON "
-            + "Schema specification"
-        )
-
-    return generator
-
-
-@json.register(OpenAI)
-def json_openai(
-    model, schema_object: Union[str, object], sampler: Sampler = multinomial()
-):
-    if not isinstance(sampler, multinomial):
-        raise NotImplementedError(
-            r"The OpenAI API does not support any other sampling algorithm "
-            + "than the multinomial sampler."
-        )
-
-    if isinstance(schema_object, type(BaseModel)):
-        schema = schema_object.model_json_schema()
-        schema["additionalProperties"] = False
-        schema = pyjson.dumps(schema)
-        format_sequence = lambda x: schema_object.parse_raw(x)
-    elif isinstance(schema_object, str):
-        schema = schema_object
-        format_sequence = lambda x: pyjson.loads(x)
-    else:
-        raise ValueError(
-            f"Cannot parse schema {schema_object}. The schema must be either "
-            + "a Pydantic object, a function or a string that contains the JSON "
-            + "Schema specification"
-        )
-
-    # create copied, patched model with normalized json schema set
-    generator = model.new_with_replacements(
-        response_format={
-            "type": "json_schema",
-            "json_schema": {
-                "name": "default",
-                "strict": True,
-                "schema": pyjson.loads(schema),
-            },
-        }
-    )
-
-    generator.format_sequence = format_sequence
-
-    return generator
diff --git a/outlines/v0_legacy/generate/__init__.py b/outlines/v0_legacy/generate/__init__.py
new file mode 100644
index 000000000..617f7b2e3
--- /dev/null
+++ b/outlines/v0_legacy/generate/__init__.py
@@ -0,0 +1,17 @@
+from .cfg import cfg
+from .choice import choice
+from .format import format
+from .fsm import fsm
+from .json import json
+from .regex import regex
+from .text import text
+
+__all__ = [
+    "cfg",
+    "choice",
+    "format",
+    "fsm",
+    "json",
+    "regex",
+    "text",
+]
diff --git a/outlines/v0_legacy/generate/api.py b/outlines/v0_legacy/generate/api.py
new file mode 100644
index 000000000..c5be8592a
--- /dev/null
+++ b/outlines/v0_legacy/generate/api.py
@@ -0,0 +1,176 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+from dataclasses import asdict
+from typing import Any, Dict, List, Optional, Union
+
+from outlines.generator import Generator
+from outlines.v0_legacy.models.exllamav2 import ExLlamaV2Model, exllamav2_params_adapter
+from outlines.v0_legacy.models.transformers import transformers_params_adapter
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
+from outlines.v0_legacy.models.vllm import vllm_params_adapter
+from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
+from outlines.models import (
+    LlamaCpp,
+    MLXLM,
+    Transformers,
+    VLLM
+)
+
+V0GeneratorModel = Union[
+    ExLlamaV2Model,
+    LlamaCpp,
+    MLXLM,
+    Transformers,
+    TransformersVision,
+    VLLM,
+]
+
+
+class GeneratorV0Adapter:
+    """Object used to bridge the gap between the legacy `generate` functions
+    and the current `Generator` object.
+
+    The legacy `generate` functions return an instance of this class. When
+    initialized, it stores an instance of the current `Generator` object
+    and calls it when it's called.
+    """
+
+    def __init__(self, model, output_type, sampler):
+        if not isinstance(model, V0GeneratorModel):  # type: ignore[misc]
+            raise ValueError(
+                "You can only use the v0 API with models that were already"
+                + f"available in v0. Got {type(model)}."
+            )
+        self.model = model
+        self.generator = Generator(model, output_type)
+        self.sampling_params = asdict(sampler.sampling_params)
+
+    def format_sequence(self, sequence: str):
+        return sequence
+
+    def create_inference_params(
+        self, max_tokens, stop_at, seed, **model_specific_params
+    ):
+        """Turn the arguments received by the generator into arguments
+        for the model associated to it.
+        """
+        generation_params = {
+            "max_tokens": max_tokens,
+            "stop_at": stop_at,
+            "seed": seed,
+        }
+
+        if isinstance(self.model, LlamaCpp):
+            adapter_function = llamacpp_params_adapter
+        elif isinstance(self.model, VLLM):
+            adapter_function = vllm_params_adapter
+        elif isinstance(self.model, Transformers):
+            adapter_function = transformers_params_adapter
+        elif isinstance(self.model, ExLlamaV2Model):
+            adapter_function = exllamav2_params_adapter
+        elif isinstance(self.model, MLXLM):
+            adapter_function = mlxlm_params_adapter
+        else:
+            raise ValueError(
+                "You can only use the v0 API with models that were already"
+                + f"available in v0. Got {type(self.model)}."
+            )
+
+        inference_params = adapter_function(
+            self.sampling_params,
+            generation_params,
+            model_specific_params
+        )
+        return inference_params
+
+    def __call__(
+        self,
+        prompts: Union[str, List[str], dict],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        result = self.generator(
+            prompts,
+            **self.create_inference_params(max_tokens, stop_at, seed, **model_specific_params)
+        )
+        if isinstance(result, list):
+            return [self.format_sequence(r) for r in result]
+        else:
+            return self.format_sequence(result)
+
+    def stream(
+        self,
+        prompts: Union[str, List[str], dict],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        return self.generator.stream(
+            prompts,
+            **self.create_inference_params(max_tokens, stop_at, seed, **model_specific_params)
+        )
+
+
+class GeneratorVisionV0Adapter(GeneratorV0Adapter):
+    """This object does the same as `GeneratorV0Adapter` but for the
+    `TransformersVision` model. The only difference between the two is that
+    it receives a media argument in addition to the prompts.
+    """
+
+    @staticmethod
+    def merge_prompts_and_media(
+        prompts: Union[str, List[str]],
+        media: Union[Any, List[Any]],
+    ) -> Dict[str, Any]:
+        """Merge prompts and media into a single dictionary
+
+        This function is needed as the `TransformersMultiModal` model requires
+        the prompt and assets to be contained in a single argument while the
+        legacy `TransformersVision` model requires separate arguments for the
+        prompt and assets.
+        """
+        return {
+            "text": prompts,
+            "images": media,
+        }
+
+    def __call__( # type: ignore[override]
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[Any, List[Any]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        merged_input = self.merge_prompts_and_media(prompts, media)
+        return super().__call__(
+            merged_input,
+            max_tokens,
+            stop_at,
+            seed,
+            **model_specific_params
+        )
+
+    def stream( # type: ignore[override]
+        self,
+        prompts: Union[str, List[str]],
+        media: Union[Any, List[Any]],
+        max_tokens: Optional[int] = None,
+        stop_at: Optional[Union[str, List[str]]] = None,
+        seed: Optional[int] = None,
+        **model_specific_params,
+    ):
+        merged_input = self.merge_prompts_and_media(prompts, media)
+        return super().stream(
+            merged_input,
+            max_tokens,
+            stop_at,
+            seed,
+            **model_specific_params
+        )
diff --git a/outlines/v0_legacy/generate/cfg.py b/outlines/v0_legacy/generate/cfg.py
new file mode 100644
index 000000000..9f59d08f2
--- /dev/null
+++ b/outlines/v0_legacy/generate/cfg.py
@@ -0,0 +1,79 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models.llamacpp import LlamaCpp
+from outlines.models.openai import OpenAI
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.types.dsl import CFG
+
+
+def cfg(
+    model, cfg_str: str, sampler: Sampler = multinomial()
+) -> GeneratorV0Adapter:
+    """Generate text in the language of a Context-Free Grammar
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Instead, use the `Generator` object instead as such:
+
+    ```python
+    from outlines import Generator
+    from outlines.types import CFG
+    cfg_str = "..."  # Context-Free Grammar as a string
+    generator = Generator(model, CFG(cfg_str))
+    ```
+
+    You can then call the generator created with a prompt to generate text
+    that matches the Context-Free Grammar.
+
+    Arguments
+    ---------
+    model:
+        The Outlines v0 model to use to generate text.
+    cfg_str:
+        The Context-Free Grammar to use to generate text.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+
+    warnings.warn("""
+        The `cfg` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, use the `Generator` object instead as such:
+        ```python
+        from outlines import Generator
+        from outlines.types import CFG
+        cfg_str = '...'  # Context-Free Grammar as a string
+        generator = Generator(model, CFG(cfg_str))
+        ```
+        You can then call the generator created with a prompt to generate
+        text in the language of a Context-Free Grammar.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use CFG-based structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API."
+        )
+    elif isinstance(model, LlamaCpp):
+        raise NotImplementedError(
+            "Not yet available due to bug in llama_cpp tokenizer"
+        )
+    elif isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, CFG(cfg_str), sampler)
+    return GeneratorV0Adapter(model, CFG(cfg_str), sampler)
diff --git a/outlines/v0_legacy/generate/choice.py b/outlines/v0_legacy/generate/choice.py
new file mode 100644
index 000000000..22cb68f91
--- /dev/null
+++ b/outlines/v0_legacy/generate/choice.py
@@ -0,0 +1,123 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import re
+import json as pyjson
+import warnings
+from enum import Enum
+from typing import List, Union
+
+from outlines.models.openai import OpenAI
+from outlines.types.utils import get_schema_from_enum
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.generate.json import json
+from outlines.v0_legacy.generate.regex import regex
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+
+def choice(
+    model,
+    choices: Union[List[str], type[Enum]],
+    sampler: Sampler = multinomial(),
+) -> GeneratorV0Adapter:
+    """Generate a choice from a list of options.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Instead, use the `Generator` object instead as such:
+
+    ```python
+    from typing import Literal
+    from outlines import Generator
+    generator = Generator(model, Literal["foo", "bar"])
+    ```
+
+    You can then call the generator created with a prompt to generate a
+    choice from the list of options.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    choices:
+        A list of options to choose from.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate a choice from the list of options.
+
+    """
+    warnings.warn("""
+        The `choice` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, use the `Generator` object instead as such:
+        ```python
+        from outlines import Generator
+        from typing import Literal
+        generator = Generator(model, Literal['foo', 'bar'])
+        ```
+        You can then call the generator created with a prompt to generate
+        a choice from the list of options.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        return choice_openai(model, choices, sampler)
+
+    if isinstance(choices, type(Enum)):
+        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
+    else:
+        choices = [re.escape(choice) for choice in choices]  # type: ignore
+        regex_str = r"(" + r"|".join(choices) + r")"
+
+    # We do not want to raise the deprecation warning through the regex function
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            category=DeprecationWarning,
+        )
+        generator = regex(model, regex_str, sampler)
+
+    if isinstance(choices, type(Enum)):
+        generator.format_sequence = lambda x: pyjson.loads(x)
+    else:
+        generator.format_sequence = lambda x: x
+
+    return generator
+
+
+def choice_openai(
+    model,
+    choices: Union[List[str], type[Enum]],
+    sampler: Sampler = multinomial(),
+):
+
+    if isinstance(choices, type(Enum)):
+        choices = [choice.value for choice in choices]
+
+    choices_schema = pyjson.dumps(
+        {
+            "type": "object",
+            "properties": {"result": {"type": "string", "enum": choices}},
+            "additionalProperties": False,
+            "required": ["result"],
+        }
+    )
+
+    # We do not want to raise the deprecation warning through the json function
+    with warnings.catch_warnings():
+        warnings.filterwarnings(
+            "ignore",
+            category=DeprecationWarning
+        )
+        generator = json(model, choices_schema, sampler)
+
+    def generate_choice(*args, **kwargs):
+        return generator(*args, **kwargs)["result"]
+
+    return generate_choice
diff --git a/outlines/v0_legacy/generate/format.py b/outlines/v0_legacy/generate/format.py
new file mode 100644
index 000000000..b1b49eec4
--- /dev/null
+++ b/outlines/v0_legacy/generate/format.py
@@ -0,0 +1,69 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models.openai import OpenAI
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.samplers import Sampler, multinomial
+
+
+def format(
+    model, python_type, sampler: Sampler = multinomial()
+) -> GeneratorV0Adapter:
+    """Generate structured data that can be parsed as a Python type.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Instead, use the `Generator` object instead as such for instance:
+
+    ```python
+    from outlines import Generator
+    generator = Generator(model, int)
+    ```
+
+    You can then call the generator created with a prompt to generate a
+    integer.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    python_type:
+        A Python type. The output of the generator must be parseable into
+        this type.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `SequenceGenerator` instance that generates text constrained by the Python type
+    and translates this text into the corresponding type.
+
+    """
+    warnings.warn("""
+        The `format` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, use the `Generator` object instead as such:
+        ```python
+        from outlines import Generator
+        generator = Generator(model, int)
+        ```
+        You can then call the generator created with a prompt to generate
+        an integer.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use Python type-structured generation with an OpenAI model"
+            + " due to the limitations of the OpenAI API."
+        )
+    elif isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, python_type, sampler)
+    return GeneratorV0Adapter(model, python_type, sampler)
diff --git a/outlines/v0_legacy/generate/fsm.py b/outlines/v0_legacy/generate/fsm.py
new file mode 100644
index 000000000..faff9962f
--- /dev/null
+++ b/outlines/v0_legacy/generate/fsm.py
@@ -0,0 +1,76 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+import interegular
+
+from outlines.models.openai import OpenAI
+from outlines.types import FSM
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+
+
+def fsm(
+    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
+) -> GeneratorV0Adapter:
+    """Generate text that follows a Finite State Machine.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Instead, use the `Generator` object instead as such:
+
+    ```python
+    import interegular
+    from outlines import Generator
+    from outlines.types import FSM
+    fsm = FSM(interegular.parse_pattern(r"[a-z]+").to_fsm())
+    generator = Generator(model, fsm)
+    ```
+
+    You can then call the generator created with a prompt to generate text
+    that follows the Finite State Machine.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    fsm:
+        The interregular `FSM` to use to generate text.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text that follows the Finite State Machine.
+
+    """
+    warnings.warn("""
+        The `fsm` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, use the `Generator` object instead as such:
+        ```python
+        from outlines import Generator
+        from outlines.types import FSM
+        fsm_str = '...'  # Finite State Machine as a string
+        generator = Generator(model, FSM(fsm_str))
+        ```
+        You can then call the generator created with a prompt to generate
+        text that follows the Finite State Machine.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use FSM-based generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API."
+        )
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, FSM(fsm), sampler)
+    return GeneratorV0Adapter(model, FSM(fsm), sampler)
diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
new file mode 100644
index 000000000..1e3c9f5a0
--- /dev/null
+++ b/outlines/v0_legacy/generate/json.py
@@ -0,0 +1,145 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import json as pyjson
+import warnings
+from typing import Callable, Optional, Union
+
+from pydantic import BaseModel
+
+from outlines.models.openai import OpenAI
+from outlines.types.dsl import JsonSchema
+from outlines.types.utils import get_schema_from_signature, is_callable
+from outlines.v0_legacy.samplers import Sampler, multinomial
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+
+
+def json(
+    model,
+    schema_object: Union[str, object, Callable],
+    sampler: Sampler = multinomial(),
+    whitespace_pattern: Optional[str] = None,
+) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter]:
+    """Generate structured text that follows a JSON Schema.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Instead, use the `Generator` object instead as such:
+
+    ```python
+    from outlines import Generator
+    from outlines.types import JsonSchema
+    generator = Generator(model, JsonSchema(schema_object))
+    ```
+
+    You can then call the generator created with a prompt to generate text that
+    matches the JSON Schema.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    schema_object:
+        The JSON Schema to generate data for. Can be a JSON string, a Pydantic
+        model, or a callable that returns a JSON schema.
+    sampler:
+        The sampler defining the sampling parameters.
+    whitespace_pattern
+        Pattern to use for JSON syntactic whitespace (doesn't impact string
+        literals).
+        Example: allow only a single space or newline with
+        `whitespace_pattern=r"[\n ]?"`.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+    warnings.warn("""
+        The `json` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, use the `Generator` object instead as such:
+        ```python
+        from outlines import Generator
+        from outlines.types import JsonSchema
+        schema_str = '...'  # JSON Schema as a string
+        generator = Generator(model, JsonSchema(schema_str))
+        ```
+        You can then call the generator created with a prompt to generate
+        JSON data that matches the schema.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        return json_openai(model, schema_object, sampler, whitespace_pattern)
+
+    if is_callable(schema_object):
+        json_schema = str(get_schema_from_signature(schema_object))  # type: ignore
+    else:
+        json_schema = JsonSchema(schema_object, whitespace_pattern) # type: ignore
+
+    if isinstance(model, TransformersVision):
+        generator = GeneratorVisionV0Adapter(model, json_schema, sampler)
+    else:
+        generator = GeneratorV0Adapter(model, json_schema, sampler)  # type: ignore
+
+    if isinstance(schema_object, type(BaseModel)):
+        setattr(
+            generator,
+            'format_sequence',
+            lambda x: schema_object.parse_raw(x)
+        )
+    else:
+        setattr(generator, 'format_sequence', lambda x: pyjson.loads(x))
+
+    return generator
+
+
+def json_openai(
+    model,
+    schema_object: Union[str, object, Callable],
+    sampler: Sampler = multinomial(),
+    whitespace_pattern: Optional[str] = None,
+):
+    if not isinstance(sampler, multinomial):
+        raise NotImplementedError(
+            "The OpenAI API does not support any other sampling algorithm "
+            + "than the multinomial sampler."
+        )
+
+    if isinstance(schema_object, type(BaseModel)):
+        schema = schema_object.model_json_schema()
+        schema["additionalProperties"] = False
+        schema = pyjson.dumps(schema)
+        format_sequence = lambda x: schema_object.parse_raw(x)
+    elif isinstance(schema_object, str):
+        schema = schema_object
+        format_sequence = lambda x: pyjson.loads(x)
+    else:
+        raise ValueError(
+            f"Cannot parse schema {schema_object}. The schema must be either "
+            + "a Pydantic object, a function or a string that contains the JSON "
+            + "Schema specification"
+        )
+
+    # create copied, patched model with normalized json schema set
+    generator = model.new_with_replacements(
+        response_format={
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": pyjson.loads(schema),
+            },
+        }
+    )
+
+    generator.format_sequence = format_sequence
+
+    return generator
diff --git a/outlines/v0_legacy/generate/regex.py b/outlines/v0_legacy/generate/regex.py
new file mode 100644
index 000000000..7d8df6105
--- /dev/null
+++ b/outlines/v0_legacy/generate/regex.py
@@ -0,0 +1,66 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+
+from outlines.models.openai import OpenAI
+from outlines.types import Regex
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.samplers import Sampler, multinomial
+
+
+def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
+    """Generate structured text in the language of a regular expression.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Instead, use the `Generator` object instead as such:
+
+    ```python
+    from outlines import Generator
+    from outlines.types import Regex
+    generator = Generator(model, Regex(r"^[a-z]+$"))
+    ```
+
+    You can then call the generator created with a prompt to generate text
+    that matches the regular expression.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    regex_str:
+        The regular expression that the output must follow.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+    warnings.warn("""
+        The `regex` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, use the `Generator` object instead as such:
+        ```python
+        from outlines import Generator
+        from outlines.types import Regex
+        generator = Generator(model, Regex(r'regex_str'))
+        ```
+        You can then call the generator created with a prompt to generate
+        text that matches the regular expression.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        raise NotImplementedError(
+            "Cannot use regex-structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API."
+        )
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, regex_str, sampler)
+    return GeneratorV0Adapter(model, regex_str, sampler)
diff --git a/outlines/v0_legacy/generate/text.py b/outlines/v0_legacy/generate/text.py
new file mode 100644
index 000000000..bcc6fe081
--- /dev/null
+++ b/outlines/v0_legacy/generate/text.py
@@ -0,0 +1,62 @@
+### Legacy API: file exists for backward compatibility with v0
+### Do not use it as an example of how to write new code
+
+import warnings
+from typing import Union
+from outlines.models.openai import OpenAI
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.samplers import Sampler, multinomial
+
+
+def text(
+        model, sampler: Sampler = multinomial()
+    ) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter, OpenAI]:
+    """Generate unconstrained text.
+
+    This function is deprecated starting from v1.0.0. Do not use it.
+    Instead, use the `Generator` object instead as such:
+
+    ```python
+    from outlines import Generator
+    generator = Generator(model)
+    ```
+
+    You can then call the generator created with a prompt to generate text.
+
+    Parameters
+    ----------
+    model:
+        The Outlines v0 model to use to generate text.
+    sampler:
+        The sampler defining the sampling parameters.
+
+    Returns
+    -------
+    A `GeneratorV0Adapter` instance that can be called with a prompt to
+    generate text.
+
+    """
+    warnings.warn("""
+        The `text` function is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, use the `Generator` object instead as such:
+        ```python
+        from outlines import Generator
+        generator = Generator(model)
+        ```
+        You can then call the generator created with a prompt to generate
+        text.
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+
+    if isinstance(model, OpenAI):
+        return model
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(model, None, sampler)
+    return GeneratorV0Adapter(model, None, sampler)

From 2764241e5fb01a92a0a0d9ccdd0717c8708c91fc Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 15:44:28 +0200
Subject: [PATCH 303/505] Include legacy code in the root level __init__

---
 outlines/__init__.py | 32 ++++++++++++++++++++++++++++++--
 1 file changed, 30 insertions(+), 2 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index a2c7be86f..14c90d090 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -1,7 +1,9 @@
 """Outlines is a Generative Model Programming Framework."""
 
+import sys
+from types import ModuleType
+
 import outlines.grammars
-import outlines.models
 import outlines.processors
 import outlines.types
 from outlines.applications import Application
@@ -23,7 +25,6 @@
     from_vllm,
 )
 
-
 model_list = [
     "from_anthropic",
     "from_dottxt",
@@ -49,3 +50,30 @@
     "Template",
     "grammars",
 ] + model_list
+
+
+# v0 legacy
+
+# flake8: noqa: E402
+from .v0_legacy import generate
+from .v0_legacy import samplers
+from .v0_legacy import models as legacy_models
+from .v0_legacy import function
+from .v0_legacy.function import Function
+
+generate_module = ModuleType("generate")
+generate_module.__dict__.update(generate.__dict__)
+
+function_module = ModuleType("function")
+function_module.__dict__.update(function.__dict__)
+
+samplers_module = ModuleType("samplers")
+samplers_module.__dict__.update(samplers.__dict__)
+
+outlines.models.__dict__.update(legacy_models.__dict__)
+
+sys.modules["outlines.generate"] = generate_module
+sys.modules["outlines.samplers"] = samplers_module
+sys.modules["outlines.function"] = function_module
+
+__all__ += ["Function"]

From c54fd9ad9a90efe118966e15c347a91014732b67 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 31 Mar 2025 15:48:55 +0200
Subject: [PATCH 304/505] Add tests for v0 legacy code

---
 tests/models/test_llamacpp.py                 |   2 +-
 tests/models/test_mlxlm.py                    |   9 +-
 tests/models/test_vllm.py                     |   6 +-
 tests/v0_legacy/generate/test_api_legacy.py   | 149 ++++++++++
 tests/v0_legacy/generate/test_cfg_legacy.py   | 111 +++++++
 .../v0_legacy/generate/test_choice_legacy.py  |  90 ++++++
 .../v0_legacy/generate/test_format_legacy.py  |  93 ++++++
 tests/v0_legacy/generate/test_fsm_legacy.py   | 104 +++++++
 tests/v0_legacy/generate/test_json_legacy.py  | 191 ++++++++++++
 tests/v0_legacy/generate/test_regex_legacy.py |  90 ++++++
 tests/v0_legacy/generate/test_text_legacy.py  |  83 ++++++
 .../v0_legacy/models/test_exllamav2_legacy.py |  68 +++++
 .../v0_legacy/models/test_llamacpp_legacy.py  | 167 +++++++++++
 tests/v0_legacy/models/test_mlxlm_legacy.py   | 193 ++++++++++++
 tests/v0_legacy/models/test_openai_legacy.py  | 166 +++++++++++
 .../models/test_transformers_legacy.py        | 172 +++++++++++
 .../models/test_transformers_vision_legacy.py | 110 +++++++
 tests/v0_legacy/models/test_vllm_legacy.py    | 181 ++++++++++++
 tests/v0_legacy/test_base_legacy.py           | 275 +++++++++++++++++
 tests/v0_legacy/test_function_legacy.py       | 151 ++++++++++
 tests/v0_legacy/test_samplers_legacy.py       | 277 ++++++++++++++++++
 21 files changed, 2680 insertions(+), 8 deletions(-)
 create mode 100644 tests/v0_legacy/generate/test_api_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_cfg_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_choice_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_format_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_fsm_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_json_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_regex_legacy.py
 create mode 100644 tests/v0_legacy/generate/test_text_legacy.py
 create mode 100644 tests/v0_legacy/models/test_exllamav2_legacy.py
 create mode 100644 tests/v0_legacy/models/test_llamacpp_legacy.py
 create mode 100644 tests/v0_legacy/models/test_mlxlm_legacy.py
 create mode 100644 tests/v0_legacy/models/test_openai_legacy.py
 create mode 100644 tests/v0_legacy/models/test_transformers_legacy.py
 create mode 100644 tests/v0_legacy/models/test_transformers_vision_legacy.py
 create mode 100644 tests/v0_legacy/models/test_vllm_legacy.py
 create mode 100644 tests/v0_legacy/test_base_legacy.py
 create mode 100644 tests/v0_legacy/test_function_legacy.py
 create mode 100644 tests/v0_legacy/test_samplers_legacy.py

diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index f5e196d15..af6802679 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -12,7 +12,7 @@
     from_llamacpp
 )
 from outlines.processors import RegexLogitsProcessor
-from outlines.types.dsl import Regex, python_types_to_terms, to_regex
+from outlines.types.dsl import Regex
 
 
 def test_load_model():
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index b293cf576..8ec889230 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -12,13 +12,10 @@
 )
 from outlines.models.transformers import TransformerTokenizer
 from pydantic import BaseModel
-from transformers import PreTrainedTokenizer
 
 try:
     import mlx_lm
     import mlx.core as mx
-    from mlx_lm import load
-    from mlx_lm.tokenizer_utils import TokenizerWrapper
 
     HAS_MLX = mx.metal.is_available()
 except ImportError:
@@ -32,8 +29,10 @@
 def test_mlxlm_model_initialization():
     model = from_mlxlm(*mlx_lm.load(TEST_MODEL))
     assert isinstance(model, MLXLM)
-    assert isinstance(model.model, mlx_lm.MLXLM)
-    assert isinstance(model.mlx_tokenizer, mlx_lm.TokenizerWrapper)
+    assert isinstance(model.model, mlx_lm.models.llama.Model)
+    assert isinstance(
+        model.mlx_tokenizer, mlx_lm.tokenizer_utils.TokenizerWrapper
+    )
     assert isinstance(model.tokenizer, TransformerTokenizer)
     assert isinstance(model.type_adapter, MLXLMTypeAdapter)
     assert model.tensor_library_name == "mlx"
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index f5a32d997..4efb93f4c 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -2,7 +2,6 @@
 from enum import Enum
 
 import pytest
-import torch
 from pydantic import BaseModel
 
 try:
@@ -151,5 +150,8 @@ class Foo(Enum):
 
 
 def test_vllm_streaming(model):
-    with pytest.raises(NotImplementedError, match="Streaming is not implemented"):
+    with pytest.raises(
+        NotImplementedError,
+        match="Streaming is not available for the vLLM integration."
+    ):
         model.stream("Respond with one word. Not more.")
diff --git a/tests/v0_legacy/generate/test_api_legacy.py b/tests/v0_legacy/generate/test_api_legacy.py
new file mode 100644
index 000000000..b139f20ea
--- /dev/null
+++ b/tests/v0_legacy/generate/test_api_legacy.py
@@ -0,0 +1,149 @@
+from io import BytesIO
+from urllib.request import urlopen
+from dataclasses import asdict
+
+import pytest
+import transformers
+from PIL import Image
+from anthropic import Anthropic as AnthropicClient
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator
+from outlines.models import TransformersVision  # type: ignore
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers(
+            "erwanf/gpt2-mini"
+        )
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            transformers.AutoModelForCausalLM
+        )
+
+
+@pytest.fixture(scope="session")
+def llama_cpp_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        return models.llamacpp(
+            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+
+
+def create_generator_v0_adapter(model):
+    if isinstance(model, TransformersVision):
+        return GeneratorVisionV0Adapter(
+            model,
+            int,
+            samplers.greedy(),
+        )
+    else:
+        return GeneratorV0Adapter(
+            model,
+            int,
+            samplers.greedy(),
+        )
+
+
+def test_generator_v0_adapter_init(transformers_model):
+    # wrong model type
+    model = models.from_anthropic(AnthropicClient(), "claude-3-haiku-20240307")
+    with pytest.raises(
+        ValueError,
+        match="You can only use the v0 API with models that were already"
+        + "available in v0. Got <class 'outlines.models.anthropic.Anthropic'>.",
+    ):
+        adapter = GeneratorV0Adapter(model, None, samplers.greedy())
+
+    # unknown output type
+    with pytest.raises(
+        TypeError,
+        match="is currently not supported",
+    ):
+        adapter = GeneratorV0Adapter(
+            transformers_model, type, samplers.greedy()
+        )
+
+    # valid initialization
+    adapter = GeneratorV0Adapter(
+        transformers_model,
+        int,
+        samplers.greedy(),
+    )
+    assert isinstance(adapter, GeneratorV0Adapter)
+    assert adapter.model == transformers_model
+    assert adapter.sampling_params == asdict(samplers.greedy().sampling_params)
+    assert isinstance(adapter.generator, SteerableGenerator)
+    assert adapter.generator.model == transformers_model
+    assert isinstance(adapter.generator.logits_processor, RegexLogitsProcessor)
+
+
+def test_generator_v0_adapter_create_inference_params(
+    llama_cpp_model,
+    transformers_vision_model,
+    transformers_model,
+):
+    args = {
+        "max_tokens": 10,
+        "stop_at": "\n",
+        "seed": 42
+    }
+    kwargs = {
+        "foo": "bar",
+    }
+
+    generator = create_generator_v0_adapter(llama_cpp_model)
+    result = generator.create_inference_params(**args, **kwargs)
+    assert isinstance(result, dict)
+
+    generator = create_generator_v0_adapter(transformers_vision_model)
+    result = generator.create_inference_params(**args, **kwargs)
+    assert isinstance(result, dict)
+
+    generator = create_generator_v0_adapter(transformers_model)
+    result = generator.create_inference_params(**args, **kwargs)
+    assert isinstance(result, dict)
+
+
+def test_generator_vision_v0_adapter_merge_prompts_and_media(transformers_vision_model):
+    generator = create_generator_v0_adapter(transformers_vision_model)
+    prompts = ["Hello, world!", "Goodbye, world!"]
+    media = [
+        img_from_url(
+            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+        ),
+        img_from_url(
+            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+        )
+    ]
+    result = generator.merge_prompts_and_media(prompts, media)
+    assert isinstance(result, dict)
+    assert result["text"] == prompts
+    assert result["images"] == media
diff --git a/tests/v0_legacy/generate/test_cfg_legacy.py b/tests/v0_legacy/generate/test_cfg_legacy.py
new file mode 100644
index 000000000..2b8a2734e
--- /dev/null
+++ b/tests/v0_legacy/generate/test_cfg_legacy.py
@@ -0,0 +1,111 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import grammars, models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import CFGLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.cfg import cfg
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def llama_cpp_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        return models.llamacpp(
+            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf"
+        )
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_cfg_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use CFG-based structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API.",
+        ):
+            cfg(openai_model, grammars.arithmetic)
+
+
+def test_cfg_legacy_llama_cpp(llama_cpp_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Not yet available due to bug in llama_cpp tokenizer",
+        ):
+            cfg(llama_cpp_model, grammars.arithmetic)
+
+
+def test_cfg_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        generator = cfg(transformers_vision_model, grammars.arithmetic)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
+
+
+def test_cfg_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `cfg` function is deprecated",
+    ):
+        generator = cfg(transformers_model, grammars.arithmetic, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_choice_legacy.py b/tests/v0_legacy/generate/test_choice_legacy.py
new file mode 100644
index 000000000..fafe8bbce
--- /dev/null
+++ b/tests/v0_legacy/generate/test_choice_legacy.py
@@ -0,0 +1,90 @@
+import json as pyjson
+from dataclasses import asdict
+from enum import Enum
+from typing import Callable
+
+import pytest
+from pydantic import BaseModel
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.choice import choice
+
+
+class EnumChoice(Enum):
+    foo = "foo"
+    bar = "bar"
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+def test_choice_legacy_standard_enum(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `choice` function is deprecated",
+    ):
+        generator = choice(transformers_model, EnumChoice, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_choice_legacy_standard_list_strings(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `choice` function is deprecated",
+    ):
+        generator = choice(transformers_model, ["foo", "bar"])
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert generator.format_sequence("foo") == "foo"
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_choice_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `choice` function is deprecated",
+    ):
+        generator = choice(openai_model, ["foo", "bar"])
+        assert isinstance(generator, Callable)
diff --git a/tests/v0_legacy/generate/test_format_legacy.py b/tests/v0_legacy/generate/test_format_legacy.py
new file mode 100644
index 000000000..81503fb8f
--- /dev/null
+++ b/tests/v0_legacy/generate/test_format_legacy.py
@@ -0,0 +1,93 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.format import format
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_format_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `format` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use Python type-structured generation with an OpenAI model"
+            + " due to the limitations of the OpenAI API.",
+        ):
+            format(openai_model, int)
+
+
+def test_format_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `format` function is deprecated",
+    ):
+        generator = format(transformers_vision_model, int)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        RegexLogitsProcessor
+    )
+
+
+def test_format_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `format` function is deprecated",
+    ):
+        generator = format(transformers_model, int, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        RegexLogitsProcessor
+    )
diff --git a/tests/v0_legacy/generate/test_fsm_legacy.py b/tests/v0_legacy/generate/test_fsm_legacy.py
new file mode 100644
index 000000000..fe6e5ba2e
--- /dev/null
+++ b/tests/v0_legacy/generate/test_fsm_legacy.py
@@ -0,0 +1,104 @@
+from dataclasses import asdict
+
+import interegular
+import pytest
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import GuideLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.fsm import fsm
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_fsm_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `fsm` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use FSM-based generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API.",
+        ):
+            fsm(
+                openai_model,
+                interegular.parse_pattern(r"[a-z]+").to_fsm()
+            )
+
+
+def test_fsm_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `fsm` function is deprecated",
+    ):
+        generator = fsm(
+            transformers_vision_model,
+            interegular.parse_pattern(r"[a-z]+").to_fsm()
+        )
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        GuideLogitsProcessor
+    )
+
+
+def test_fsm_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `fsm` function is deprecated",
+    ):
+        generator = fsm(
+            transformers_model,
+            interegular.parse_pattern(r"[a-z]+").to_fsm(),
+            samplers.greedy()
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(
+        generator.generator.logits_processor,
+        GuideLogitsProcessor
+    )
diff --git a/tests/v0_legacy/generate/test_json_legacy.py b/tests/v0_legacy/generate/test_json_legacy.py
new file mode 100644
index 000000000..fa4ac28ff
--- /dev/null
+++ b/tests/v0_legacy/generate/test_json_legacy.py
@@ -0,0 +1,191 @@
+import json as pyjson
+import pytest
+from dataclasses import asdict
+
+from pydantic import BaseModel
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.json import json
+
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "foo": {
+            "type": "string"
+        }
+    },
+    "required": ["foo"]
+}
+"""
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_json_legacy_standard_callable(transformers_model):
+    def foo(bar: str) -> str:
+        return bar
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_model, foo, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_standard_json_string(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_model, JSON_SCHEMA_STRING)
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_standard_pydantic(transformers_model):
+    class Foo(BaseModel):
+        foo: str
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_model, Foo)
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert (
+        generator.format_sequence('{"foo": "value"}')
+        == Foo.model_validate_json('{"foo": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(transformers_vision_model, JSON_SCHEMA_STRING)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
+    assert isinstance(
+        generator.generator.logits_processor, RegexLogitsProcessor
+    )
+
+
+def test_json_legacy_openai_wrong_sampler(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="The OpenAI API does not support any other sampling "
+            + "algorithm than the multinomial sampler.",
+        ):
+            json(openai_model, JSON_SCHEMA_STRING, samplers.greedy())
+
+
+def test_json_legacy_openai_pydantic(openai_model):
+    class Foo(BaseModel):
+        foo: str
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(openai_model, Foo)
+    assert isinstance(generator, models.OpenAI)
+    assert (
+        generator.format_sequence('{"foo": "value"}')
+        == Foo.model_validate_json('{"foo": "value"}')
+    )
+
+
+def test_json_legacy_openai_json_string(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        generator = json(openai_model, JSON_SCHEMA_STRING)
+    assert isinstance(generator, models.OpenAI)
+    assert (
+        generator.format_sequence('{"key": "value"}')
+        == pyjson.loads('{"key": "value"}')
+    )
diff --git a/tests/v0_legacy/generate/test_regex_legacy.py b/tests/v0_legacy/generate/test_regex_legacy.py
new file mode 100644
index 000000000..86da00c35
--- /dev/null
+++ b/tests/v0_legacy/generate/test_regex_legacy.py
@@ -0,0 +1,90 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.processors import RegexLogitsProcessor
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.regex import regex
+
+
+REGEX_STRING = r"^[a-z]+$"
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_regex_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `regex` function is deprecated",
+    ):
+        with pytest.raises(
+            NotImplementedError,
+            match="Cannot use regex-structured generation with an OpenAI model"
+            + "due to the limitations of the OpenAI API.",
+        ):
+            regex(openai_model, REGEX_STRING)
+
+
+def test_regex_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `regex` function is deprecated",
+    ):
+        generator = regex(transformers_vision_model, REGEX_STRING)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
+
+
+def test_regex_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `regex` function is deprecated",
+    ):
+        generator = regex(transformers_model, REGEX_STRING, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_text_legacy.py b/tests/v0_legacy/generate/test_text_legacy.py
new file mode 100644
index 000000000..e7412b295
--- /dev/null
+++ b/tests/v0_legacy/generate/test_text_legacy.py
@@ -0,0 +1,83 @@
+import pytest
+from dataclasses import asdict
+
+from transformers import AutoModelForCausalLM
+
+from outlines import models, samplers
+from outlines.generator import SteerableGenerator, Generator
+from outlines.v0_legacy.generate.api import (
+    GeneratorV0Adapter,
+    GeneratorVisionV0Adapter
+)
+from outlines.v0_legacy.generate.text import text
+
+
+@pytest.fixture(scope="session")
+def openai_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        return models.openai("gpt-4o")
+
+
+@pytest.fixture(scope="session")
+def transformers_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        return models.transformers("erwanf/gpt2-mini")
+
+
+@pytest.fixture(scope="session")
+def transformers_vision_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        return models.transformers_vision(
+            "erwanf/gpt2-mini",
+            AutoModelForCausalLM
+        )
+
+
+def test_text_legacy_openai(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = text(openai_model)
+    assert generator == openai_model
+    assert isinstance(generator, models.OpenAI)
+
+
+def test_text_legacy_transformers_vision(transformers_vision_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = text(transformers_vision_model)
+    assert isinstance(generator, GeneratorVisionV0Adapter)
+    assert generator.model == transformers_vision_model
+    assert generator.sampling_params == asdict(
+        samplers.multinomial().sampling_params
+    )
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_vision_model
+    assert generator.generator.logits_processor is None
+
+
+def test_text_legacy_standard_model(transformers_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = text(transformers_model, samplers.greedy())
+    assert isinstance(generator, GeneratorV0Adapter)
+    assert generator.model == transformers_model
+    assert generator.sampling_params == asdict(
+        samplers.greedy().sampling_params)
+    assert isinstance(generator.generator, SteerableGenerator)
+    assert generator.generator.model == transformers_model
+    assert generator.generator.logits_processor is None
diff --git a/tests/v0_legacy/models/test_exllamav2_legacy.py b/tests/v0_legacy/models/test_exllamav2_legacy.py
new file mode 100644
index 000000000..34afc7e9c
--- /dev/null
+++ b/tests/v0_legacy/models/test_exllamav2_legacy.py
@@ -0,0 +1,68 @@
+import warnings
+
+import pytest
+
+try:
+    import exllamav2  # noqa: F401
+    HAS_EXV2 = True
+except ImportError:
+    HAS_EXV2 = False
+    warnings.warn(
+        "Skipping exllamav2 legacy tests because exllamav2 is not available."
+    )
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.models.exllamav2 import exllamav2_params_adapter
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+pytestmark = pytest.mark.skipif(
+    not HAS_EXV2,
+    reason="Exllamav2 is not available."
+)
+
+
+@pytest.fixture(scope="session")
+def model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `exllamav2` function is deprecated",
+    ):
+        model = models.exl2(
+            model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
+            max_seq_len=1024,
+            cache_q4=False,
+        )
+    return model
+
+
+def test_exllamav2_legacy_init(model):
+    assert isinstance(model, models.ExllamaV2)
+    assert hasattr(model, "generator")
+    assert hasattr(model, "tokenizer")
+    assert model.tokenizer.convert_token_to_string(1) == 1
+    assert hasattr(model, "max_seq_len")
+    assert isinstance(model.max_seq_len, int)
+
+
+def test_exllamav2_legacy_call(model):
+    # minimal
+    generator = generate.text(model)
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator("Respond with a single word.")
+    assert isinstance(result, str)
+
+    # with args
+    generator = generate.text(
+        model,
+        samplers.multinomial(samples=2, temperature=10, top_p=0.5, top_k=2)
+    )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(["Write a short story.", "Write a poem."], 10, "e", 2)
+    assert isinstance(result, list)
+    for sequence in result:
+        assert isinstance(sequence, list)
+        for s in sequence:
+            assert isinstance(s, str)
+            assert len(s) < 20
diff --git a/tests/v0_legacy/models/test_llamacpp_legacy.py b/tests/v0_legacy/models/test_llamacpp_legacy.py
new file mode 100644
index 000000000..318df4bef
--- /dev/null
+++ b/tests/v0_legacy/models/test_llamacpp_legacy.py
@@ -0,0 +1,167 @@
+import warnings
+from dataclasses import asdict
+
+import pytest
+from llama_cpp import Llama
+from llama_cpp.llama_tokenizer import LlamaHFTokenizer, LlamaTokenizer
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+
+def test_llamacpp_legacy_init():
+    # without tokenizer
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        with pytest.warns(
+            match="The pre-tokenizer in `llama.cpp` handles unicode improperly"
+        ):
+            model = models.llamacpp(
+                repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+                filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            )
+    assert isinstance(model, models.LlamaCpp)
+    assert isinstance(model.model, Llama)
+    assert not model.model.verbose
+
+    # with tokenizer
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        model = models.llamacpp(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            tokenizer=LlamaHFTokenizer.from_pretrained("distilbert-base-uncased"),
+        )
+    assert isinstance(model, models.LlamaCpp)
+    assert isinstance(model.model, Llama)
+    assert not model.model.verbose
+    assert isinstance(model.model.tokenizer(), LlamaTokenizer)
+
+
+def test_llamacpp_legacy_params_adapter_generation_params():
+    result = llamacpp_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert not result.get("stop")
+    assert not result.get("seed")
+    assert result.get("max_tokens") == -1
+
+    result = llamacpp_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
+        {}
+    )
+    assert result.get("stop") == "foo"
+    assert result.get("seed") == 2
+    assert result.get("max_tokens") == 10
+
+
+def test_llamacpp_legacy_params_adapter_sampling_params():
+    with pytest.raises(
+        NotImplementedError,
+        match="does not support Beam Search"
+    ):
+        result = llamacpp_params_adapter(
+            asdict(samplers.beam_search().sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    with pytest.raises(
+        NotImplementedError,
+        match="does not allow to take several samples"
+    ):
+        result = llamacpp_params_adapter(
+            asdict(samplers.multinomial(samples=2).sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    result = llamacpp_params_adapter(
+        {"sampler": "multinomial", "num_samples": 1, "top_p": None, "top_k": None, "temperature": None},
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("top_p") == 1.0
+    assert result.get("top_k") == -1
+    assert result.get("temperature") == 1.0
+
+    result = llamacpp_params_adapter(
+        {"sampler": "greedy", "num_samples": 1, "top_p": 2, "top_k": 0, "temperature": 0.5},
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("top_p") == 2
+    assert result.get("top_k") == 0
+    assert result.get("temperature") == 0.5
+
+
+def test_llamacpp_legacy_params_adapter_model_specific_params():
+    result = llamacpp_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("min_p") == 0.0
+    assert result.get("repeat_penalty") == 1.0
+
+    result = llamacpp_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
+        {"min_p": 0.5, "repeat_penalty": 1.5, "stream": True}
+    )
+    assert result.get("min_p") == 0.5
+    assert result.get("repeat_penalty") == 1.5
+    assert not result.get("stream")
+
+
+def test_llamacpp_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `llamacpp` function is deprecated",
+    ):
+        model = models.llamacpp(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                top_p=0.9,
+                top_k=10,
+                temperature=0.7
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!",
+        10,
+        "foo",
+        2,
+        min_p=0.5,
+        repeat_penalty=1.5,
+    )
+    assert isinstance(result, str)
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!",
+        10,
+        "foo",
+        2,
+        min_p=0.5,
+        repeat_penalty=1.5,
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_mlxlm_legacy.py b/tests/v0_legacy/models/test_mlxlm_legacy.py
new file mode 100644
index 000000000..e70943ef2
--- /dev/null
+++ b/tests/v0_legacy/models/test_mlxlm_legacy.py
@@ -0,0 +1,193 @@
+import warnings
+from dataclasses import asdict
+
+import pytest
+
+try:
+    import mlx.core as mx
+    import mlx_lm  # noqa: F401
+
+    assert mx.metal.is_available()
+    HAS_MLX = True
+except ImportError:
+    HAS_MLX = False
+    warnings.warn("Skipping MLXLM legacy tests because MLX is not available.")
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
+
+pytestmark = pytest.mark.skipif(
+    not HAS_MLX,
+    reason="MLX tests require Apple Silicon."
+)
+
+
+def test_mlxlm_legacy_init():
+    # minimal
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mlxlm` function is deprecated",
+    ):
+        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
+    assert isinstance(model, models.MLXLM)
+    assert isinstance(model.model, mlx_lm.models.llama.Model)
+
+    # with optional arguments
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mlxlm` function is deprecated",
+    ):
+        model = models.mlxlm(
+            "mlx-community/SmolLM-135M-Instruct-4bit",
+            tokenizer_config={"max_length": 100},
+            model_config={"max_length": 100},
+            adapter_path=None,
+            lazy=False,
+        )
+    assert isinstance(model, models.MLXLM)
+    assert isinstance(model.model, mlx_lm.models.llama.Model)
+    assert isinstance(model.mlx_tokenizer, mlx_lm.tokenizer_utils.TokenizerWrapper)
+
+
+def test_mlxlm_legacy_params_adapter_generation_params():
+    # default
+    result = mlxlm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result.get("stop") is None
+    assert result.get("seed") is None
+    assert result.get("max_tokens") == int(1e9)
+
+    # max_tokens
+    result = mlxlm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": 2},
+        {}
+    )
+    assert result.get("max_tokens") == 2
+
+    # stop_at
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support `stop_at`."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial().sampling_params),
+            {"stop_at": "foo", "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # seed
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support `seed`."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial().sampling_params),
+            {"stop_at": None, "seed": 2, "max_tokens": None},
+            {}
+        )
+
+
+def test_mlxlm_legacy_params_adapter_sampling_params():
+    # beam search
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support Beam Search."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.beam_search().sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # num_samples
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not allow to take several samples."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial(samples=2).sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # top_k
+    with pytest.raises(
+        NotImplementedError,
+        match="The `mlx-lm` library does not support top_k."
+    ):
+        result = mlxlm_params_adapter(
+            asdict(samplers.multinomial(top_k=2).sampling_params),
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+
+    # temperature
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `temperature` parameter is not available",
+    ):
+        result = mlxlm_params_adapter(
+            {
+                "sampler": "multinomial",
+                "num_samples": 1,
+                "top_p": None,
+                "top_k": None,
+                "temperature": 0.5
+            },
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+        assert result.get("temperature") is None
+
+    # top_p
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `top_p` parameter is not available",
+    ):
+        result = mlxlm_params_adapter(
+            {
+                "sampler": "multinomial",
+                "num_samples": 1,
+                "top_p": 2,
+                "top_k": None,
+                "temperature": None
+            },
+            {"stop_at": None, "seed": None, "max_tokens": None},
+            {}
+        )
+        assert result.get("top_p") is None
+
+
+def test_mlxlm_legacy_params_adapter_model_specific_params():
+    result = mlxlm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {"length_penalty": 0.5}
+    )
+    assert result.get("length_penalty") == 0.5
+
+
+def test_mlxlm_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mlxlm` function is deprecated",
+    ):
+        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
+
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial()
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator("Hello, world!", 10)
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_openai_legacy.py b/tests/v0_legacy/models/test_openai_legacy.py
new file mode 100644
index 000000000..13e3c750f
--- /dev/null
+++ b/tests/v0_legacy/models/test_openai_legacy.py
@@ -0,0 +1,166 @@
+import os
+
+import pytest
+from openai import AsyncAzureOpenAI, AsyncOpenAI
+
+from outlines import models, generate
+from outlines.models.openai import OpenAI, OpenAIConfig, OpenAILegacy
+
+
+# Set the OPENAI_API_KEY environment variable to "MOCK_VALUE" if it is not set
+# not to raise an error when the OpenAI client is initialized
+api_key = os.getenv("OPENAI_API_KEY")
+if not api_key:
+    os.environ["OPENAI_API_KEY"] = "MOCK_VALUE"
+azure_api_key = os.getenv("AZURE_OPENAI_API_KEY")
+if not azure_api_key:
+    os.environ["AZURE_OPENAI_API_KEY"] = "MOCK_VALUE"
+
+
+def test_openai_legacy_init():
+    # through the openai function with a client and no config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        client = AsyncOpenAI()
+        model = models.openai(client, system_prompt="You are a helpful assistant.")
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.config == OpenAIConfig()
+    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # through the openai function with a client and config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            client,
+            OpenAIConfig(stop=["."]),
+            system_prompt="You are a helpful assistant.",
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.config == OpenAIConfig(stop=["."])
+    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # through the openai function with model name and no config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            "gpt-4o",
+            max_retries=1,
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.client.max_retries == 1
+    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o")
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # through the openai function with model name and config
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            "gpt-4o",
+            OpenAIConfig(stop=["."]),
+            max_retries=1,
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.client.max_retries == 1
+    assert model.legacy_instance.config == OpenAIConfig(
+        model="gpt-4o", stop=["."]
+    )
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # directly through the OpenAI class with client, config and system prompt
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.OpenAI(
+            AsyncOpenAI(),
+            OpenAIConfig(model="gpt-4o", n=1),
+            system_prompt="You are a helpful assistant.",
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
+    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o", n=1)
+    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+    # directly through the OpenAI class without config
+    with pytest.raises(TypeError):
+        models.OpenAI(
+            AsyncOpenAI(),
+            system_prompt="You are a helpful assistant.",
+        )
+
+
+def test_azure_openai_legacy_init():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.azure_openai(
+            "foo",
+            max_retries=1,
+            base_url="https://foo.openai.azure.com/",
+            api_version="2023-03-15-preview",
+        )
+    assert isinstance(model, models.OpenAI)
+    assert isinstance(model.legacy_instance, OpenAILegacy)
+    assert isinstance(model.legacy_instance.client, AsyncAzureOpenAI)
+    assert model.legacy_instance.client.max_retries == 1
+    assert model.legacy_instance.config == OpenAIConfig(model="foo")
+    assert not hasattr(model, "client")
+    assert not hasattr(model, "model_name")
+    assert not hasattr(model, "type_adapter")
+
+
+@pytest.mark.api_call
+def test_openai_legacy_call_model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `openai` function is deprecated",
+    ):
+        model = models.openai(
+            "gpt-4o",
+            max_retries=1,
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(model)
+    result = generator(
+        "Hello, world!",
+        10,
+        ["."],
+        system_prompt="You are a helpful assistant.",
+        temperature=0.5,
+        samples=1
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_legacy.py b/tests/v0_legacy/models/test_transformers_legacy.py
new file mode 100644
index 000000000..91b25a571
--- /dev/null
+++ b/tests/v0_legacy/models/test_transformers_legacy.py
@@ -0,0 +1,172 @@
+from dataclasses import asdict
+
+import pytest
+from transformers import (
+    GPT2LMHeadModel,
+    GPT2TokenizerFast,
+    BertForSequenceClassification,
+    BertTokenizer,
+    MambaForCausalLM,
+)
+
+from outlines import models, samplers, generate
+from outlines.v0_legacy.models.transformers import transformers_params_adapter
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+
+def test_transformers_legacy_init():
+    # Test with default model and tokenizer classes
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="erwanf/gpt2-mini",
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, GPT2LMHeadModel)
+    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
+
+    # Test with non default model and tokenizer classes
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="prajjwal1/bert-tiny",
+            model_class=BertForSequenceClassification,
+            tokenizer_class=BertTokenizer,
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, BertForSequenceClassification)
+    assert isinstance(model.tokenizer.tokenizer, BertTokenizer)
+
+    # Test with model and tokenizer kwargs and device
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="erwanf/gpt2-mini",
+            model_kwargs={"use_cache": False},
+            tokenizer_kwargs={"model_max_length": 512},
+            device="cpu",
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, GPT2LMHeadModel)
+    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
+    assert model.tokenizer.tokenizer.model_max_length == 512
+    assert not model.model.config.use_cache
+    assert model.model.device.type == "cpu"
+
+
+def test_mamba_legacy_init():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `mamba` function is deprecated",
+    ):
+        model = models.mamba(
+            model_name="erwanf/gpt2-mini",
+        )
+    assert isinstance(model, models.Transformers)
+    assert isinstance(model.model, MambaForCausalLM)
+    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
+
+
+def test_transformers_legacy_params_adapter_generation_params():
+    result = transformers_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert not result.get("stop")
+    assert not result.get("seed")
+    assert result.get("max_new_tokens") == 2**30
+
+    result = transformers_params_adapter(
+         asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
+        {}
+    )
+    assert result.get("stop_strings") == "foo"
+    assert not result.get("seed")
+    assert result.get("max_new_tokens") == 10
+
+
+def test_transformers_legacy_params_adapter_sampling_params():
+    result = transformers_params_adapter(
+        {
+            "sampler": "multinomial",
+            "num_samples": None,
+            "top_p": None,
+            "top_k": None,
+            "temperature": None,
+        },
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {},
+    )
+    assert result.get("do_sample")
+    assert result.get("num_return_sequences") == 1
+    assert result.get("top_p") is None
+    assert result.get("top_k") is None
+    assert result.get("temperature") is None
+
+    result = transformers_params_adapter(
+        {
+            "sampler": "beam_search",
+            "num_samples": 2,
+            "top_p": 2,
+            "top_k": 0,
+            "temperature": 0.5,
+        },
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {},
+    )
+    assert result.get("num_beams") == 2
+    assert result.get("num_return_sequences") == 2
+    assert result.get("top_p") == 2
+    assert result.get("top_k") == 0
+    assert result.get("temperature") == 0.5
+
+
+def test_transformers_legacy_params_adapter_model_specific_params():
+    result = transformers_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {"length_penalty": 1.0}
+    )
+    assert result.get("length_penalty") == 1
+    assert not result.get("foo")
+
+
+def test_transformers_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers` function is deprecated",
+    ):
+        model = models.transformers(
+            model_name="erwanf/gpt2-mini",
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                samples=1,
+                top_p=0.9,
+                top_k=10,
+                temperature=0.7
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!",
+        10,
+        "foo",
+        2,
+        length_penalty=0.5,
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_vision_legacy.py b/tests/v0_legacy/models/test_transformers_vision_legacy.py
new file mode 100644
index 000000000..bcaf4ea3e
--- /dev/null
+++ b/tests/v0_legacy/models/test_transformers_vision_legacy.py
@@ -0,0 +1,110 @@
+from io import BytesIO
+from urllib.request import urlopen
+
+import pytest
+from PIL import Image
+from transformers import (
+    AutoProcessor,
+    LlavaForConditionalGeneration,
+    LlamaTokenizer,
+    LlamaTokenizerFast,
+    LlavaProcessor,
+)
+
+from outlines import models, generate, samplers
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+
+def test_transformers_vision_legacy_init():
+    # Test with only model name and model class
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+        )
+    assert isinstance(model, models.TransformersVision)
+    assert isinstance(model.model, LlavaForConditionalGeneration)
+    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
+    assert isinstance(model.processor, LlavaProcessor)
+
+    # Test with model, tokenizer, and processor classes
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+            tokenizer_class=LlamaTokenizer,
+            processor_class=AutoProcessor,
+        )
+    assert isinstance(model, models.TransformersVision)
+    assert isinstance(model.model, LlavaForConditionalGeneration)
+    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizer)
+    assert isinstance(model.processor, LlavaProcessor)
+
+    # Test with model, tokenizer, processor kwargs, and device
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+            model_kwargs={"output_attentions": True},
+            processor_kwargs={"num_additional_image_tokens": 1},
+            device="cpu",
+        )
+    assert isinstance(model, models.TransformersVision)
+    assert isinstance(model.model, LlavaForConditionalGeneration)
+    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
+    assert isinstance(model.processor, LlavaProcessor)
+    assert model.model.config.output_attentions
+    assert model.processor.num_additional_image_tokens == 1
+    assert model.model.device.type == "cpu"
+
+
+def test_transformers_vision_legacy_call_generation():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `transformers_vision` function is deprecated",
+    ):
+        model = models.transformers_vision(
+            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=LlavaForConditionalGeneration,
+        )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                samples=1,
+                top_p=0.9,
+                top_k=10,
+                temperature=0.7
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!<image>",
+        img_from_url(
+            "https://upload.wikimedia.org/wikipedia/commons"
+            + "/2/25/Siam_lilacpoint.jpg"
+        ),
+        10,
+        "foo",
+        2,
+        length_penalty=0.5,
+    )
+    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_vllm_legacy.py b/tests/v0_legacy/models/test_vllm_legacy.py
new file mode 100644
index 000000000..97eed996c
--- /dev/null
+++ b/tests/v0_legacy/models/test_vllm_legacy.py
@@ -0,0 +1,181 @@
+import warnings
+from dataclasses import asdict
+
+import pytest
+
+try:
+    from vllm import LLM
+    HAS_VLLM = True
+except ImportError:
+    HAS_VLLM = False
+    warnings.warn("Skipping vLLM legacy tests because vLLM is not available.")
+
+from outlines import generate, models, samplers
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.models.vllm import vllm_params_adapter
+
+pytestmark = pytest.mark.skipif(
+    not HAS_VLLM,
+    reason="vLLM models can only be run on GPU."
+)
+
+
+@pytest.fixture
+def model():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `vllm` function is deprecated",
+    ):
+        model_instance = models.vllm(
+            "erwanf/gpt2-mini",
+            seed=1,
+        )
+    return model_instance
+
+
+def test_vllm_legacy_init(model):
+    assert isinstance(model, models.VLLM)
+    assert isinstance(model.model, LLM)
+
+
+def test_vllm_legacy_params_adapter_generation_params():
+    # default
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].get("stop") is None
+    assert result["sampling_params"].get("seed") is None
+    assert result["sampling_params"].get("max_tokens") is None
+
+    # max_tokens
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": 2},
+        {}
+    )
+    assert result["sampling_params"].max_tokens == 2
+
+    # seed
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": 2, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].seed == 2
+
+    # stop_at str
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": "foo", "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].stop_strings == ["foo"]
+
+    # stop_at list
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": ["foo", "bar"], "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].stop_strings == ["foo", "bar"]
+
+
+def test_vllm_legacy_params_adapter_sampling_params():
+    # default
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].n == 1
+    assert result["sampling_params"].top_p == 1.0
+    assert result["sampling_params"].top_k == -1
+    assert result["sampling_params"].temperature == 1.0
+
+    # beam search
+    result = vllm_params_adapter(
+        asdict(samplers.beam_search().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].use_beam_search is True
+
+    # num_samples
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial(samples=2).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].n == 2
+    assert result["sampling_params"].best_of == 2
+
+    # top_p
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial(top_p=0.5).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].top_p == 0.5
+
+    # top_k
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial(top_k=2).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].top_k == 2
+
+    # temperature
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial(temperature=0.5).sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {}
+    )
+    assert result["sampling_params"].temperature == 0.5
+
+
+def test_vllm_legacy_params_adapter_model_specific_params():
+    result = vllm_params_adapter(
+        asdict(samplers.multinomial().sampling_params),
+        {"stop_at": None, "seed": None, "max_tokens": None},
+        {"use_tqdm": False}
+    )
+    assert not result.get("use_tqdm")
+
+
+def test_vllm_legacy_call_generation(model):
+    # default
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial()
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator("Hello, world!")
+    assert isinstance(result, str)
+
+    # with args
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `text` function is deprecated",
+    ):
+        generator = generate.text(
+            model,
+            samplers.multinomial(
+                samples=2, top_p=0.5, top_k=2, temperature=0.5
+            )
+        )
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator(
+        "Hello, world!", 10, stop_at="foo", seed=1, use_tqdm=False
+    )
+    assert isinstance(result, list)
+    for completion in result:
+        assert isinstance(completion, str)
diff --git a/tests/v0_legacy/test_base_legacy.py b/tests/v0_legacy/test_base_legacy.py
new file mode 100644
index 000000000..f445b4f5d
--- /dev/null
+++ b/tests/v0_legacy/test_base_legacy.py
@@ -0,0 +1,275 @@
+import numpy as np
+import pytest
+from numpy.testing import assert_array_equal
+
+from outlines.v0_legacy.base import vectorize
+
+
+def test_vectorize_docstring():
+    def test(x):
+        """This is a test docstring"""
+        return x
+
+    fn = vectorize(test)
+    assert fn.__doc__ == "This is a test docstring"
+    assert fn.__name__ == "test"
+
+    async def test(x):
+        """This is a test docstring"""
+        return x
+
+    fn = vectorize(test)
+    assert fn.__doc__ == "This is a test docstring"
+    assert fn.__name__ == "test"
+
+
+def test_vectorize_thunk():
+    def thunk():
+        """A thunk"""
+        return 1
+
+    fn = vectorize(thunk)
+    assert fn() == 1
+
+    async def thunk():
+        """A thunk"""
+        return 1
+
+    fn = vectorize(thunk)
+    assert fn() == 1
+
+
+def test_vectorize_scalar_simple():
+    def passthrough(x):
+        """A passthrough function."""
+        return x
+
+    fn = vectorize(passthrough)
+
+    out_vector = fn(["one", "two", "three"])
+    assert_array_equal(out_vector, ["one", "two", "three"])
+
+    out_array = fn([["one", "two"], ["three", "four"]])
+    assert_array_equal(out_array, [["one", "two"], ["three", "four"]])
+
+    async def passthrough(x):
+        """A passthrough function."""
+        return x
+
+    fn = vectorize(passthrough)
+    assert fn.__doc__ == "A passthrough function."
+
+    out_vector = fn(["one", "two", "three"])
+    assert_array_equal(out_vector, ["one", "two", "three"])
+
+    out_array = fn([["one", "two"], ["three", "four"]])
+    assert_array_equal(out_array, [["one", "two"], ["three", "four"]])
+
+
+def test_vectorize_scalar_multiple_outputs():
+    def passthrough_multiple_outputs(x):
+        return x, x
+
+    fn = vectorize(passthrough_multiple_outputs)
+    out = fn(["one", "two", "three"])
+    assert len(out) == 2
+    assert_array_equal(out[0], ["one", "two", "three"])
+    assert_array_equal(out[1], ["one", "two", "three"])
+
+    async def passthrough_multiple_outputs(x):
+        return x, x
+
+    fn = vectorize(passthrough_multiple_outputs)
+    out = fn(["one", "two", "three"])
+    assert len(out) == 2
+    assert_array_equal(out[0], ["one", "two", "three"])
+    assert_array_equal(out[1], ["one", "two", "three"])
+
+
+def test_vectorize_scalar_args():
+    def passthrough_args(*args):
+        """A passthrough function."""
+        result = ""
+        for arg in args:
+            result += arg
+        return result
+
+    fn = vectorize(passthrough_args)
+
+    out_array = fn(["one", "two"], ["1", "2"])
+    assert_array_equal(out_array, ["one1", "two2"])
+
+    # Broadcasting
+    out_array = fn([["one", "two"], ["three", "four"]], ["1", "2"])
+    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
+
+    async def passthrough_args(*args):
+        """A passthrough function."""
+        result = ""
+        for arg in args:
+            result += arg
+        return result
+
+    fn = vectorize(passthrough_args)
+
+    out_array = fn(["one", "two"], ["1", "2"])
+    assert_array_equal(out_array, ["one1", "two2"])
+
+    # Broadcasting
+    out_array = fn([["one", "two"], ["three", "four"]], ["1", "2"])
+    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
+
+
+def test_vectorize_scalar_kwargs():
+    def passthrough_kwargs(**kwargs):
+        """A passthrough function."""
+        result = ""
+        for _, value in kwargs.items():
+            result += value
+        return result
+
+    fn = vectorize(passthrough_kwargs)
+
+    out_array = fn(first=["one", "two"], second=["1", "2"])
+    assert_array_equal(out_array, ["one1", "two2"])
+
+    # Broadcasting
+    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
+    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
+
+    async def passthrough_kwargs(**kwargs):
+        """A passthrough function."""
+        result = ""
+        for _, value in kwargs.items():
+            result += value
+        return result
+
+    fn = vectorize(passthrough_kwargs)
+
+    out_array = fn(first=["one", "two"], second=["1", "2"])
+    assert_array_equal(out_array, ["one1", "two2"])
+
+    # Broadcasting
+    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
+    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
+
+
+def test_signature_invalid():
+    with pytest.raises(ValueError):
+        vectorize(lambda x: x, "(m,)->()")
+
+    with pytest.raises(TypeError, match="wrong number of positional"):
+        fn = vectorize(lambda x, y: x, "()->()")
+        fn(1, 2)
+
+    with pytest.raises(ValueError, match="wrong number of outputs"):
+
+        def test_multioutput(x, y):
+            return x, y
+
+        fn = vectorize(test_multioutput, "(),()->()")
+        fn(1, 2)
+
+
+def test_vectorize_simple():
+    def test(x):
+        return x
+
+    fn = vectorize(test, "(m)->(m)")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
+
+    fn = vectorize(test, "()->()")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
+
+
+def test_vectorize_kwargs():
+    def passthrough_kwargs(**kwargs):
+        """A passthrough function."""
+        result = ""
+        for _, value in kwargs.items():
+            result += value
+        return result
+
+    fn = vectorize(passthrough_kwargs, "(),()->()")
+
+    out_array = fn(first=["one", "two"], second=["1", "2"])
+    assert_array_equal(out_array, ["one1", "two2"])
+
+    # Broadcasting
+    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
+    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
+
+    async def passthrough_kwargs(**kwargs):
+        """A passthrough function."""
+        result = ""
+        for _, value in kwargs.items():
+            result += value
+        return result
+
+    fn = vectorize(passthrough_kwargs, "(),()->()")
+
+    out_array = fn(first=["one", "two"], second=["1", "2"])
+    assert_array_equal(out_array, ["one1", "two2"])
+
+    # Broadcasting
+    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
+    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
+
+
+def test_vectorize_reduce():
+    def test(x):
+        return x[0]
+
+    fn = vectorize(test, "(m)->()")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, ["one", "four"])
+
+    fn = vectorize(test, "()->()")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, [["o", "t", "t"], ["f", "f", "s"]])
+
+
+def test_vectorize_expand():
+    def test(x):
+        return np.array([x for i in range(3)])
+
+    fn = vectorize(test, "()->(s)")
+    out = fn(["one", "two"])
+    assert_array_equal(out, [["one", "one", "one"], ["two", "two", "two"]])
+
+
+def test_vectorize_coroutine_simple():
+    async def test(x):
+        return x
+
+    fn = vectorize(test, "(m)->(m)")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
+
+    fn = vectorize(test, "()->()")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
+
+
+def test_vectorize_coroutine_reduce():
+    async def test(x):
+        return x[0]
+
+    fn = vectorize(test, "(m)->()")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, ["one", "four"])
+
+    fn = vectorize(test, "()->()")
+    out = fn([["one", "two", "three"], ["four", "five", "six"]])
+    assert_array_equal(out, [["o", "t", "t"], ["f", "f", "s"]])
+
+
+def test_vectorize_coroutine_expand():
+    async def test(x):
+        return np.array([x for i in range(3)])
+
+    fn = vectorize(test, "()->(s)")
+    out = fn(["one", "two"])
+    assert_array_equal(out, [["one", "one", "one"], ["two", "two", "two"]])
diff --git a/tests/v0_legacy/test_function_legacy.py b/tests/v0_legacy/test_function_legacy.py
new file mode 100644
index 000000000..17a4e70ae
--- /dev/null
+++ b/tests/v0_legacy/test_function_legacy.py
@@ -0,0 +1,151 @@
+import pytest
+import responses
+from pydantic import BaseModel
+from requests.exceptions import HTTPError
+
+import outlines
+from outlines.v0_legacy.function import (
+    Function,
+    download_from_github,
+    extract_function_from_file,
+)
+
+
+def test_function_basic():
+    with pytest.warns(DeprecationWarning, match="The @prompt decorator"):
+        with pytest.warns(
+            DeprecationWarning, match="The `Function` class is deprecated"
+        ):
+            @outlines.prompt
+            def test_template(text: str):
+                """{{ text }}"""
+
+            class Foo(BaseModel):
+                id: int
+
+            fn = Function(
+                test_template,
+                Foo,
+                "hf-internal-testing/tiny-random-GPTJForCausalLM",
+            )
+
+            assert fn.generator is None
+
+            result = fn("test")
+            assert isinstance(result, BaseModel)
+
+
+def test_download_from_github_invalid():
+    with pytest.raises(ValueError, match="Please provide"):
+        download_from_github("outlines/program")
+
+    with pytest.raises(ValueError, match="Do not append"):
+        download_from_github("dottxt-ai/outlines/program.py")
+
+
+@responses.activate
+def test_download_from_github_success():
+    responses.add(
+        responses.GET,
+        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
+        body="import outlines\n",
+        status=200,
+    )
+
+    file = download_from_github("dottxt-ai/outlines/program")
+    assert file == "import outlines\n"
+
+    responses.add(
+        responses.GET,
+        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/foo/bar/program.py",
+        body="import outlines\n",
+        status=200,
+    )
+
+    file = download_from_github("dottxt-ai/outlines/foo/bar/program")
+    assert file == "import outlines\n"
+
+
+@responses.activate
+def test_download_from_github_error():
+    responses.add(
+        responses.GET,
+        "https://raw.githubusercontent.com/foo/bar/main/program.py",
+        json={"error": "not found"},
+        status=404,
+    )
+
+    with pytest.raises(ValueError, match="Program could not be found at"):
+        download_from_github("foo/bar/program")
+
+    responses.add(
+        responses.GET,
+        "https://raw.githubusercontent.com/foo/bar/main/program.py",
+        json={"error": "Internal Server Error"},
+        status=500,
+    )
+
+    with pytest.raises(HTTPError, match="500 Server Error"):
+        download_from_github("foo/bar/program")
+
+
+def test_extract_function_from_file():
+    content = """
+import outlines
+from pydantic import BaseModel
+
+model = "gpt2"
+
+
+@outlines.prompt
+def prompt():
+    '''Hello'''
+
+
+class User(BaseModel):
+    id: int
+    name: str
+
+
+function = outlines.Function(
+    prompt,
+    User,
+    "gpt2",
+)
+    """
+    with pytest.raises(
+        DeprecationWarning,
+        match="The @prompt decorator is deprecated",
+    ):
+        fn = extract_function_from_file(content, "function")
+        assert (
+            str(type(fn)) == "<class 'outlines.function.Function'>"
+        )  # because imported via `exec`
+
+
+def test_extract_function_from_file_no_function():
+    content = """
+import outlines
+from pydantic import BaseModel
+
+@outlines.prompt
+def prompt():
+    '''Hello'''
+
+
+class User(BaseModel):
+    id: int
+    name: str
+
+program = outlines.Function(
+    prompt,
+    User,
+    "gpt2",
+)
+    """
+    with pytest.raises(
+        DeprecationWarning,
+        match="The @prompt decorator is deprecated",
+    ):
+        with pytest.raises(AttributeError, match="Could not find"):
+            extract_function_from_file(content, "function")
diff --git a/tests/v0_legacy/test_samplers_legacy.py b/tests/v0_legacy/test_samplers_legacy.py
new file mode 100644
index 000000000..b8470f77f
--- /dev/null
+++ b/tests/v0_legacy/test_samplers_legacy.py
@@ -0,0 +1,277 @@
+import math
+
+import pytest
+import torch
+
+from outlines.v0_legacy.samplers import (
+    BeamSearchSampler,
+    GreedySampler,
+    MultinomialSampler,
+    beam_search,
+    greedy,
+    keep_top_k_logits,
+    keep_top_p_logits,
+    multinomial,
+    rescale_logits,
+)
+
+
+def compute_logprobs(logits):
+    return torch.nn.functional.log_softmax(logits, dim=-1)
+
+
+def test_aliases():
+    assert greedy == GreedySampler
+    assert multinomial == MultinomialSampler
+    assert beam_search == BeamSearchSampler
+
+
+def test_greedy():
+    sampler = GreedySampler()
+    logits = torch.tensor([[1.0, 2.0, 5.0]])
+    weights = torch.tensor([0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[2]]))
+    assert ancestors.equal(torch.tensor([0]))
+    assert weights.equal(logprobs[..., 2])
+
+    sampler = GreedySampler()
+    logits = torch.tensor([[10.0, 0.0, 3.0], [-math.inf, 2.0, 5.0]])
+    weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [2]]))
+    assert ancestors.equal(torch.tensor([0, 1]))
+    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
+
+    params = sampler.sampling_params
+    assert params.sampler == "greedy"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 0.0
+
+
+def test_multinomial():
+    rng = torch.Generator()
+    rng.manual_seed(239)
+
+    sampler = MultinomialSampler()
+    logits = torch.tensor([[1.0, 4.0, 5.0]])
+    weights = torch.tensor([0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[2]]))
+    assert ancestors.equal(torch.tensor([0]))
+    assert weights.equal(torch.tensor([logprobs[:, 2]]))
+
+    sampler = MultinomialSampler()
+    logits = torch.tensor([[10.0, 0.0, 9.0], [-math.inf, 4.0, 5.0]])
+    weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [2]]))
+    assert ancestors.equal(torch.tensor([0, 1]))
+    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
+
+    sampler = MultinomialSampler(samples=5, top_k=10, top_p=0.9, temperature=0.8)
+    params = sampler.sampling_params
+    assert params.sampler == "multinomial"
+    assert params.num_samples == 5
+    assert params.top_p == 0.9
+    assert params.top_k == 10
+    assert params.temperature == 0.8
+
+
+def test_multinomial_init():
+    sampler = MultinomialSampler()
+    assert sampler.logits_processors == []
+
+    sampler = MultinomialSampler(3)
+    assert sampler.logits_processors == []
+
+    sampler = MultinomialSampler(top_k=1)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(top_p=0.95)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(top_k=1, top_p=0.95)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(temperature=1.0)
+    assert len(sampler.logits_processors) == 1
+
+    sampler = MultinomialSampler(top_k=1, temperature=1.0)
+    assert len(sampler.logits_processors) == 2
+
+    sampler = MultinomialSampler(top_p=0.95, temperature=1.0)
+    assert len(sampler.logits_processors) == 2
+
+
+def test_top_k():
+    with pytest.raises(ValueError, match="`k` must be a strictly"):
+        keep_top_k_logits(-1)
+
+    with pytest.raises(ValueError, match="`k` must be a strictly"):
+        keep_top_k_logits(0.1)
+
+    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0]])
+
+    logits_processor = keep_top_k_logits(1)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
+
+    logits_processor = keep_top_k_logits(10)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[1.0, 2.0, 3.0, 4.0]]))
+
+    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
+    logits_processor = keep_top_k_logits(2)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [[-math.inf, -math.inf, 3.0, 4.0], [-math.inf, -math.inf, 7.0, 8.0]]
+        )
+    )
+
+
+def test_top_p():
+    with pytest.raises(ValueError, match="`p` must be a floating point"):
+        keep_top_p_logits(-0.1)
+
+    with pytest.raises(ValueError, match="`p` must be a floating point"):
+        keep_top_p_logits(0.0)
+
+    with pytest.raises(ValueError, match="`p` must be a floating point"):
+        keep_top_p_logits(1.1)
+
+    logits = torch.tensor([[1.0, 1.01, 1.02, 4.0]])
+
+    logits_processor = keep_top_p_logits(0.1)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
+
+    logits_processor = keep_top_p_logits(0.95)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[-math.inf, 1.01, 1.02, 4.0]]))
+
+    logits_processor = keep_top_p_logits(1.0)
+    result = logits_processor(logits)
+    assert result.equal(torch.tensor([[1.0, 1.01, 1.02, 4.0]]))
+
+    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
+
+    logits_processor = keep_top_p_logits(0.1)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [
+                [-math.inf, -math.inf, -math.inf, 4.0],
+                [-math.inf, -math.inf, -math.inf, 8.0],
+            ]
+        )
+    )
+
+    logits_processor = keep_top_p_logits(0.95)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [
+                [-math.inf, 2.0, 3.0, 4.0],
+                [-math.inf, 6.0, 7.0, 8.0],
+            ]
+        )
+    )
+
+    logits_processor = keep_top_p_logits(1.0)
+    result = logits_processor(logits)
+    assert result.equal(
+        torch.tensor(
+            [
+                [1.0, 2.0, 3.0, 4.0],
+                [5.0, 6.0, 7.0, 8.0],
+            ]
+        )
+    )
+
+
+def test_rescale():
+    with pytest.raises(ValueError, match="`temperature` must"):
+        rescale_logits(1)
+
+    with pytest.raises(ValueError, match="`temperature` must"):
+        rescale_logits(-0.1)
+
+    with pytest.raises(ValueError, match="Please use the greedy sampler"):
+        rescale_logits(0.0)
+
+
+def test_beam_search():
+    # Two beams, single sequence
+    sampler = BeamSearchSampler(2)
+    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
+    init_weights = torch.tensor([0, 1.0])
+    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [1]]))
+    assert ancestors.equal(torch.tensor([1, 0]))
+    assert weights.equal(
+        torch.tensor([init_weights[1] + logprobs[1][0], logprobs[0][1]])
+    )
+
+    # Make sure that initial samples are different
+    sampler = BeamSearchSampler(2)
+    logits = torch.tensor([[0.0, 1.0], [0.0, 1.0]])
+    init_weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[1], [0]]))
+    assert ancestors.equal(torch.tensor([0, 0]))
+    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[0][0]]))
+
+    # One beam, batch of two sequences. Reduces to Greedy Search.
+    sampler = BeamSearchSampler(1)
+    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
+    weights = torch.tensor([0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[1], [0]]))
+    assert ancestors.equal(torch.tensor([0, 1]))
+    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[1][0]]))
+
+    # Two beams, batch of two sequences
+    sampler = BeamSearchSampler(2)
+    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0], [3.0, 2.0], [0.0, 1.0]])
+    init_weights = torch.tensor([0, 0, 2.0, 0])
+    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
+
+    logprobs = compute_logprobs(logits)
+    assert next_token_ids.equal(torch.tensor([[0], [1], [0], [1]]))
+    assert ancestors.equal(torch.tensor([1, 0, 2, 2]))
+    assert weights.equal(
+        torch.tensor(
+            [
+                logprobs[1][0],
+                logprobs[0][1],
+                init_weights[2] + logprobs[2][0],
+                init_weights[2] + logprobs[2][1],
+            ]
+        )
+    )
+
+    sampler = BeamSearchSampler(beams=3)
+    params = sampler.sampling_params
+    assert params.sampler == "beam_search"
+    assert params.num_samples == 3
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 1.0

From 2985a595c2977f726a67eda76d776acab06f9de1 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 8 Apr 2025 10:29:59 +0200
Subject: [PATCH 305/505] Remove the call methods from the sampler objects and
 simplify tests

---
 outlines/v0_legacy/samplers.py          | 267 ++----------------------
 tests/v0_legacy/test_samplers_legacy.py | 247 ++--------------------
 2 files changed, 31 insertions(+), 483 deletions(-)

diff --git a/outlines/v0_legacy/samplers.py b/outlines/v0_legacy/samplers.py
index 2ae22161d..9f1ff5cc6 100644
--- a/outlines/v0_legacy/samplers.py
+++ b/outlines/v0_legacy/samplers.py
@@ -1,21 +1,5 @@
-
-import math
 from dataclasses import dataclass
-from typing import TYPE_CHECKING, Callable, Optional, Protocol, Tuple
-
-if TYPE_CHECKING:
-    import torch
-
-
-class Sampler(Protocol):
-    samples: int
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        rng: "torch.Generator",
-    ) -> "torch.DoubleTensor": ...
+from typing import Optional, Protocol
 
 
 @dataclass(frozen=True)
@@ -28,6 +12,21 @@ class SamplingParameters:
     temperature: Optional[float] = None
 
 
+class Sampler(Protocol):
+    """
+    Protocol for all samplers.
+
+    As outlines now delegates the handling of the generation loop
+    to the models, the samplers are only used to define the sampling
+    parameters that are passed to the model.
+    """
+
+    samples: int
+
+    def sampling_params(self) -> SamplingParameters:
+        ...
+
+
 class GreedySampler:
     """Greedy Sampling algorithm.
 
@@ -48,45 +47,6 @@ class GreedySampler:
     def __init__(self):
         self.samples = 1
 
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        _,
-    ) -> "torch.DoubleTensor":
-        """Call the greedy sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
-        next_token_ids = torch.argmax(logprobs, dim=-1, keepdim=True)
-
-        ancestors = torch.arange(
-            next_token_logits.shape[0], device=next_token_logits.device
-        )
-        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
-
-        return next_token_ids, ancestors, weights
-
     @property
     def sampling_params(self):
         return SamplingParameters("greedy", self.samples, None, None, 0.0)
@@ -123,59 +83,6 @@ def __init__(
         self.top_p = top_p
         self.temperature = temperature
 
-        self.logits_processors = []
-        if top_k is not None:
-            self.logits_processors.append(keep_top_k_logits(top_k))
-        elif top_p is not None:
-            self.logits_processors.append(keep_top_p_logits(top_p))
-
-        if temperature is not None:
-            self.logits_processors.append(rescale_logits(temperature))
-
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        rng: "torch.Generator",
-    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
-        """Call the multinomial sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        altered_next_token_logits = next_token_logits
-        for logit_processor in self.logits_processors:
-            altered_next_token_logits = logit_processor(next_token_logits)
-
-        probs = torch.nn.functional.softmax(altered_next_token_logits, dim=-1)
-        next_token_ids = torch.multinomial(probs, num_samples=1, generator=rng)
-
-        logprobs = torch.nn.functional.log_softmax(altered_next_token_logits, dim=-1)
-        ancestors = torch.arange(
-            altered_next_token_logits.shape[0], device=next_token_logits.device
-        )
-        weights = sequence_weights + torch.gather(logprobs, 1, next_token_ids).squeeze()
-
-        return next_token_ids, ancestors, weights
-
     @property
     def sampling_params(self):
         return SamplingParameters(
@@ -190,85 +97,6 @@ def sampling_params(self):
 multinomial = MultinomialSampler
 
 
-def keep_top_k_logits(k: int) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that masks logits values smaller than the top `k` ones.
-
-    Parameters
-    ----------
-    k
-        The ranking below which logit values are replaced by `-math.inf`.
-
-    """
-    import torch
-
-    if not isinstance(k, int) or k < 1:
-        raise ValueError(f"`k` must be a strictly positive integers, got {k} instead.")
-
-    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
-        num_to_keep = min(k, logits.size(-1))
-        mask_idx = logits < torch.topk(logits, num_to_keep)[0][..., -1, None]
-        return logits.masked_fill(mask_idx, -math.inf)
-
-    return logits_processor
-
-
-def keep_top_p_logits(p: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that masks the lowest probability tokens whose
-    cumulative probability is below a certain threshold.
-
-    Parameters
-    ----------
-    p
-        The value of the threshold. We keep the highest probability tokens whose
-        cumulative distribution is greater than or equal to `p` and mask the
-        others. Its value must be between 0 (excluded) and 1 (included).
-
-    """
-    import torch
-
-    if p <= 0.0 or p > 1.0:
-        raise ValueError(
-            f"`p` must be a floating point number between 0 (excluded) and 1 (included), got {p} instead."
-        )
-
-    def logits_processor(logits: torch.Tensor) -> torch.Tensor:
-        sorted_logits, sorted_idx = torch.sort(logits, descending=False)
-        cumulative_probabilties = torch.nn.functional.softmax(
-            sorted_logits, dim=-1
-        ).cumsum(dim=-1)
-
-        sorted_masked_idx = cumulative_probabilties <= (1 - p)
-        mask_idx = torch.scatter(sorted_masked_idx, 1, sorted_idx, sorted_masked_idx)
-        return logits.masked_fill(mask_idx, -math.inf)
-
-    return logits_processor
-
-
-def rescale_logits(temperature: float) -> Callable[["torch.Tensor"], "torch.Tensor"]:
-    """Build a function that rescales the token probabilities exponentially.
-
-    Parameters
-    ----------
-    temperature
-        The value by which we rescale the logits.
-
-    """
-
-    if not isinstance(temperature, float) or temperature < 0.0:
-        raise ValueError(
-            f"`temperature` must be a strictly positive floating point number, got {temperature} instead."
-        )
-    elif temperature == 0.0:
-        raise ValueError(
-            "Please use the greedy sampler instead of setting the temperature to 0."
-        )
-
-    def logits_processor(logits: "torch.Tensor") -> "torch.Tensor":
-        return logits / temperature
-
-    return logits_processor
-
-
 class BeamSearchSampler:
     """Beam Search sampling algorithm.
 
@@ -282,69 +110,6 @@ class BeamSearchSampler:
     def __init__(self, beams: int = 1):
         self.samples = beams
 
-    def __call__(
-        self,
-        next_token_logits: "torch.DoubleTensor",
-        sequence_weights: "torch.DoubleTensor",
-        _,
-    ) -> Tuple["torch.DoubleTensor", "torch.DoubleTensor", "torch.DoubleTensor"]:
-        """Call the beam search sampler.
-
-        Parameters
-        ----------
-        next_token_logits
-            A tensor of shape ``(n_seqs, vocab_size,)`` that represents the
-            probability distribution of the next token over the vocabulary.
-        sequence_weights
-            A tensor of shape ``(n_seqs,)`` that represents the cumulative
-            weight of each sequence.
-        rng
-            A random number generator.
-
-        Returns
-        -------
-        A tuple with an array that contains the ids of the sampled tokens of
-        shape ``(n_seqs, 1)``, an array that contains the ancestors of each
-        sampled id of shape ``(n_seqs,)`` and an array that contains the updated
-        cumulative weights of each sequence of shape ``(n_seqs,)``.
-
-        """
-        import torch
-
-        logprobs = torch.nn.functional.log_softmax(next_token_logits, dim=-1)
-        weights = logprobs + sequence_weights.unsqueeze(1).expand_as(next_token_logits)
-
-        # Flatten scores to (n_batch, n_samples * vocab_size)
-        # and find the top-k weights for each batch.
-        batch_size = next_token_logits.shape[0] // self.samples
-        vocab_size = next_token_logits.shape[-1]
-        weights = weights.view(batch_size, self.samples * vocab_size)
-
-        # If the weights are all equal to 0 we are at the beginning of the search
-        # and thus only need to sample from one set of token logits for each
-        # batch.
-        if torch.all(sequence_weights == 0):
-            weights = weights[:, :vocab_size]
-
-        weights, indices = torch.topk(
-            weights, self.samples, dim=1, largest=True, sorted=True
-        )
-
-        ancestors = torch.div(indices, vocab_size, rounding_mode="floor")
-        next_token_ids = indices % vocab_size
-
-        # Re-shape the weights, next_token_ids and ancestors to (n_batch * n_samples, 1)
-        first_batch_idx = torch.arange(
-            0, batch_size * self.samples, self.samples, device=next_token_logits.device
-        ).unsqueeze(1)
-        ancestors = ancestors + first_batch_idx
-
-        ancestors = ancestors.view(self.samples * batch_size)
-        weights = weights.view(self.samples * batch_size)
-        next_token_ids = next_token_ids.view(self.samples * batch_size, 1)
-
-        return next_token_ids, ancestors, weights
-
     @property
     def sampling_params(self):
         return SamplingParameters("beam_search", self.samples, None, None, 1.0)
diff --git a/tests/v0_legacy/test_samplers_legacy.py b/tests/v0_legacy/test_samplers_legacy.py
index b8470f77f..a4d750116 100644
--- a/tests/v0_legacy/test_samplers_legacy.py
+++ b/tests/v0_legacy/test_samplers_legacy.py
@@ -1,25 +1,13 @@
-import math
-
-import pytest
-import torch
-
 from outlines.v0_legacy.samplers import (
     BeamSearchSampler,
     GreedySampler,
     MultinomialSampler,
     beam_search,
     greedy,
-    keep_top_k_logits,
-    keep_top_p_logits,
     multinomial,
-    rescale_logits,
 )
 
 
-def compute_logprobs(logits):
-    return torch.nn.functional.log_softmax(logits, dim=-1)
-
-
 def test_aliases():
     assert greedy == GreedySampler
     assert multinomial == MultinomialSampler
@@ -28,25 +16,6 @@ def test_aliases():
 
 def test_greedy():
     sampler = GreedySampler()
-    logits = torch.tensor([[1.0, 2.0, 5.0]])
-    weights = torch.tensor([0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[2]]))
-    assert ancestors.equal(torch.tensor([0]))
-    assert weights.equal(logprobs[..., 2])
-
-    sampler = GreedySampler()
-    logits = torch.tensor([[10.0, 0.0, 3.0], [-math.inf, 2.0, 5.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [2]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
-
     params = sampler.sampling_params
     assert params.sampler == "greedy"
     assert params.num_samples == 1
@@ -56,30 +25,15 @@ def test_greedy():
 
 
 def test_multinomial():
-    rng = torch.Generator()
-    rng.manual_seed(239)
-
-    sampler = MultinomialSampler()
-    logits = torch.tensor([[1.0, 4.0, 5.0]])
-    weights = torch.tensor([0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[2]]))
-    assert ancestors.equal(torch.tensor([0]))
-    assert weights.equal(torch.tensor([logprobs[:, 2]]))
-
     sampler = MultinomialSampler()
-    logits = torch.tensor([[10.0, 0.0, 9.0], [-math.inf, 4.0, 5.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, rng)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [2]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0, 0], logprobs[1, 2]]))
+    params = sampler.sampling_params
+    assert params.sampler == "multinomial"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature is None
 
-    sampler = MultinomialSampler(samples=5, top_k=10, top_p=0.9, temperature=0.8)
+    sampler = multinomial(samples=5, top_k=10, top_p=0.9, temperature=0.8)
     params = sampler.sampling_params
     assert params.sampler == "multinomial"
     assert params.num_samples == 5
@@ -88,187 +42,16 @@ def test_multinomial():
     assert params.temperature == 0.8
 
 
-def test_multinomial_init():
-    sampler = MultinomialSampler()
-    assert sampler.logits_processors == []
-
-    sampler = MultinomialSampler(3)
-    assert sampler.logits_processors == []
-
-    sampler = MultinomialSampler(top_k=1)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_p=0.95)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_k=1, top_p=0.95)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(temperature=1.0)
-    assert len(sampler.logits_processors) == 1
-
-    sampler = MultinomialSampler(top_k=1, temperature=1.0)
-    assert len(sampler.logits_processors) == 2
-
-    sampler = MultinomialSampler(top_p=0.95, temperature=1.0)
-    assert len(sampler.logits_processors) == 2
-
-
-def test_top_k():
-    with pytest.raises(ValueError, match="`k` must be a strictly"):
-        keep_top_k_logits(-1)
-
-    with pytest.raises(ValueError, match="`k` must be a strictly"):
-        keep_top_k_logits(0.1)
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0]])
-
-    logits_processor = keep_top_k_logits(1)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
-
-    logits_processor = keep_top_k_logits(10)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[1.0, 2.0, 3.0, 4.0]]))
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
-    logits_processor = keep_top_k_logits(2)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [[-math.inf, -math.inf, 3.0, 4.0], [-math.inf, -math.inf, 7.0, 8.0]]
-        )
-    )
-
-
-def test_top_p():
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(-0.1)
-
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(0.0)
-
-    with pytest.raises(ValueError, match="`p` must be a floating point"):
-        keep_top_p_logits(1.1)
-
-    logits = torch.tensor([[1.0, 1.01, 1.02, 4.0]])
-
-    logits_processor = keep_top_p_logits(0.1)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, -math.inf, -math.inf, 4.0]]))
-
-    logits_processor = keep_top_p_logits(0.95)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[-math.inf, 1.01, 1.02, 4.0]]))
-
-    logits_processor = keep_top_p_logits(1.0)
-    result = logits_processor(logits)
-    assert result.equal(torch.tensor([[1.0, 1.01, 1.02, 4.0]]))
-
-    logits = torch.tensor([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]])
-
-    logits_processor = keep_top_p_logits(0.1)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [-math.inf, -math.inf, -math.inf, 4.0],
-                [-math.inf, -math.inf, -math.inf, 8.0],
-            ]
-        )
-    )
-
-    logits_processor = keep_top_p_logits(0.95)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [-math.inf, 2.0, 3.0, 4.0],
-                [-math.inf, 6.0, 7.0, 8.0],
-            ]
-        )
-    )
-
-    logits_processor = keep_top_p_logits(1.0)
-    result = logits_processor(logits)
-    assert result.equal(
-        torch.tensor(
-            [
-                [1.0, 2.0, 3.0, 4.0],
-                [5.0, 6.0, 7.0, 8.0],
-            ]
-        )
-    )
-
-
-def test_rescale():
-    with pytest.raises(ValueError, match="`temperature` must"):
-        rescale_logits(1)
-
-    with pytest.raises(ValueError, match="`temperature` must"):
-        rescale_logits(-0.1)
-
-    with pytest.raises(ValueError, match="Please use the greedy sampler"):
-        rescale_logits(0.0)
-
-
 def test_beam_search():
-    # Two beams, single sequence
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
-    init_weights = torch.tensor([0, 1.0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [1]]))
-    assert ancestors.equal(torch.tensor([1, 0]))
-    assert weights.equal(
-        torch.tensor([init_weights[1] + logprobs[1][0], logprobs[0][1]])
-    )
-
-    # Make sure that initial samples are different
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [0.0, 1.0]])
-    init_weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[1], [0]]))
-    assert ancestors.equal(torch.tensor([0, 0]))
-    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[0][0]]))
-
-    # One beam, batch of two sequences. Reduces to Greedy Search.
-    sampler = BeamSearchSampler(1)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0]])
-    weights = torch.tensor([0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[1], [0]]))
-    assert ancestors.equal(torch.tensor([0, 1]))
-    assert weights.equal(torch.tensor([logprobs[0][1], logprobs[1][0]]))
-
-    # Two beams, batch of two sequences
-    sampler = BeamSearchSampler(2)
-    logits = torch.tensor([[0.0, 1.0], [2.0, 0.0], [3.0, 2.0], [0.0, 1.0]])
-    init_weights = torch.tensor([0, 0, 2.0, 0])
-    next_token_ids, ancestors, weights = sampler(logits, init_weights, None)
-
-    logprobs = compute_logprobs(logits)
-    assert next_token_ids.equal(torch.tensor([[0], [1], [0], [1]]))
-    assert ancestors.equal(torch.tensor([1, 0, 2, 2]))
-    assert weights.equal(
-        torch.tensor(
-            [
-                logprobs[1][0],
-                logprobs[0][1],
-                init_weights[2] + logprobs[2][0],
-                init_weights[2] + logprobs[2][1],
-            ]
-        )
-    )
+    sampler = BeamSearchSampler()
+    params = sampler.sampling_params
+    assert params.sampler == "beam_search"
+    assert params.num_samples == 1
+    assert params.top_p is None
+    assert params.top_k is None
+    assert params.temperature == 1.0
 
-    sampler = BeamSearchSampler(beams=3)
+    sampler = beam_search(beams=3)
     params = sampler.sampling_params
     assert params.sampler == "beam_search"
     assert params.num_samples == 3

From 41aedb79057ad5f07021c260f0a99ae98ca7eb61 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 8 Apr 2025 11:23:24 +0200
Subject: [PATCH 306/505] Modify warning messages for v0 legacy functions

---
 outlines/v0_legacy/generate/api.py    | 33 +++++++++++++--------------
 outlines/v0_legacy/generate/cfg.py    |  5 ++--
 outlines/v0_legacy/generate/choice.py | 27 ++++++++++++----------
 outlines/v0_legacy/generate/format.py |  5 ++--
 outlines/v0_legacy/generate/fsm.py    |  5 ++--
 outlines/v0_legacy/generate/json.py   | 13 ++++++-----
 outlines/v0_legacy/generate/regex.py  |  5 ++--
 outlines/v0_legacy/generate/text.py   | 12 ++++++----
 8 files changed, 57 insertions(+), 48 deletions(-)

diff --git a/outlines/v0_legacy/generate/api.py b/outlines/v0_legacy/generate/api.py
index c5be8592a..14165a213 100644
--- a/outlines/v0_legacy/generate/api.py
+++ b/outlines/v0_legacy/generate/api.py
@@ -38,7 +38,7 @@ class GeneratorV0Adapter:
     """
 
     def __init__(self, model, output_type, sampler):
-        if not isinstance(model, V0GeneratorModel):  # type: ignore[misc]
+        if not isinstance(model, V0GeneratorModel):  # type: ignore
             raise ValueError(
                 "You can only use the v0 API with models that were already"
                 + f"available in v0. Got {type(model)}."
@@ -64,19 +64,14 @@ def create_inference_params(
 
         if isinstance(self.model, LlamaCpp):
             adapter_function = llamacpp_params_adapter
-        elif isinstance(self.model, VLLM):
+        elif isinstance(self.model, VLLM): # pragma: no cover
             adapter_function = vllm_params_adapter
         elif isinstance(self.model, Transformers):
             adapter_function = transformers_params_adapter
-        elif isinstance(self.model, ExLlamaV2Model):
+        elif isinstance(self.model, ExLlamaV2Model): # pragma: no cover
             adapter_function = exllamav2_params_adapter
-        elif isinstance(self.model, MLXLM):
+        elif isinstance(self.model, MLXLM): # pragma: no cover
             adapter_function = mlxlm_params_adapter
-        else:
-            raise ValueError(
-                "You can only use the v0 API with models that were already"
-                + f"available in v0. Got {type(self.model)}."
-            )
 
         inference_params = adapter_function(
             self.sampling_params,
@@ -87,7 +82,7 @@ def create_inference_params(
 
     def __call__(
         self,
-        prompts: Union[str, List[str], dict],
+        prompts: Union[str, List[str], Dict[str, Any]],
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
@@ -95,7 +90,9 @@ def __call__(
     ):
         result = self.generator(
             prompts,
-            **self.create_inference_params(max_tokens, stop_at, seed, **model_specific_params)
+            **self.create_inference_params(
+                max_tokens, stop_at, seed, **model_specific_params
+            )
         )
         if isinstance(result, list):
             return [self.format_sequence(r) for r in result]
@@ -104,7 +101,7 @@ def __call__(
 
     def stream(
         self,
-        prompts: Union[str, List[str], dict],
+        prompts: Union[str, List[str], Dict[str, Any]],
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
@@ -112,7 +109,9 @@ def stream(
     ):
         return self.generator.stream(
             prompts,
-            **self.create_inference_params(max_tokens, stop_at, seed, **model_specific_params)
+            **self.create_inference_params(
+                max_tokens, stop_at, seed, **model_specific_params
+            )
         )
 
 
@@ -139,10 +138,10 @@ def merge_prompts_and_media(
             "images": media,
         }
 
-    def __call__( # type: ignore[override]
+    def __call__(  # type: ignore
         self,
         prompts: Union[str, List[str]],
-        media: Union[Any, List[Any]],
+        media: Union[str, List[str]],
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
@@ -157,10 +156,10 @@ def __call__( # type: ignore[override]
             **model_specific_params
         )
 
-    def stream( # type: ignore[override]
+    def stream(  # type: ignore
         self,
         prompts: Union[str, List[str]],
-        media: Union[Any, List[Any]],
+        media: Union[str, List[str]],
         max_tokens: Optional[int] = None,
         stop_at: Optional[Union[str, List[str]]] = None,
         seed: Optional[int] = None,
diff --git a/outlines/v0_legacy/generate/cfg.py b/outlines/v0_legacy/generate/cfg.py
index 9f59d08f2..db206996d 100644
--- a/outlines/v0_legacy/generate/cfg.py
+++ b/outlines/v0_legacy/generate/cfg.py
@@ -20,7 +20,8 @@ def cfg(
     """Generate text in the language of a Context-Free Grammar
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Instead, use the `Generator` object instead as such:
+    Support for it will be removed in v1.5.0.
+    Use the `Generator` object instead:
 
     ```python
     from outlines import Generator
@@ -51,7 +52,7 @@ def cfg(
     warnings.warn("""
         The `cfg` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, use the `Generator` object instead as such:
+        Use the `Generator` object instead:
         ```python
         from outlines import Generator
         from outlines.types import CFG
diff --git a/outlines/v0_legacy/generate/choice.py b/outlines/v0_legacy/generate/choice.py
index 22cb68f91..cee339bac 100644
--- a/outlines/v0_legacy/generate/choice.py
+++ b/outlines/v0_legacy/generate/choice.py
@@ -5,11 +5,11 @@
 import json as pyjson
 import warnings
 from enum import Enum
-from typing import List, Union
+from typing import Callable, List, Union
 
 from outlines.models.openai import OpenAI
 from outlines.types.utils import get_schema_from_enum
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
+from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
 from outlines.v0_legacy.generate.json import json
 from outlines.v0_legacy.generate.regex import regex
 from outlines.v0_legacy.samplers import Sampler, multinomial
@@ -20,11 +20,12 @@ def choice(
     model,
     choices: Union[List[str], type[Enum]],
     sampler: Sampler = multinomial(),
-) -> GeneratorV0Adapter:
+) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter, Callable]:
     """Generate a choice from a list of options.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Instead, use the `Generator` object instead as such:
+    Support for it will be removed in v1.5.0.
+    Use the `Generator` object instead:
 
     ```python
     from typing import Literal
@@ -53,7 +54,7 @@ def choice(
     warnings.warn("""
         The `choice` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, use the `Generator` object instead as such:
+        Use the `Generator` object instead:
         ```python
         from outlines import Generator
         from typing import Literal
@@ -67,7 +68,12 @@ def choice(
     )
 
     if isinstance(model, OpenAI):
-        return choice_openai(model, choices, sampler)
+        if isinstance(choices, list):
+            return choice_openai(model, choices, sampler)
+        raise ValueError(
+            "The `choice` function with OpenAI only supports a list of "
+            + "strings as choices."
+        )
 
     if isinstance(choices, type(Enum)):
         regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
@@ -93,12 +99,9 @@ def choice(
 
 def choice_openai(
     model,
-    choices: Union[List[str], type[Enum]],
+    choices: List[str],
     sampler: Sampler = multinomial(),
-):
-
-    if isinstance(choices, type(Enum)):
-        choices = [choice.value for choice in choices]
+) -> Callable:
 
     choices_schema = pyjson.dumps(
         {
@@ -117,7 +120,7 @@ def choice_openai(
         )
         generator = json(model, choices_schema, sampler)
 
-    def generate_choice(*args, **kwargs):
+    def generate_choice(*args, **kwargs): # pragma: no cover
         return generator(*args, **kwargs)["result"]
 
     return generate_choice
diff --git a/outlines/v0_legacy/generate/format.py b/outlines/v0_legacy/generate/format.py
index b1b49eec4..459eebff2 100644
--- a/outlines/v0_legacy/generate/format.py
+++ b/outlines/v0_legacy/generate/format.py
@@ -18,7 +18,8 @@ def format(
     """Generate structured data that can be parsed as a Python type.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Instead, use the `Generator` object instead as such for instance:
+    Support for it will be removed in v1.5.0.
+    Use the `Generator` object instead::
 
     ```python
     from outlines import Generator
@@ -47,7 +48,7 @@ def format(
     warnings.warn("""
         The `format` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, use the `Generator` object instead as such:
+        Use the `Generator` object instead:
         ```python
         from outlines import Generator
         generator = Generator(model, int)
diff --git a/outlines/v0_legacy/generate/fsm.py b/outlines/v0_legacy/generate/fsm.py
index faff9962f..60e130169 100644
--- a/outlines/v0_legacy/generate/fsm.py
+++ b/outlines/v0_legacy/generate/fsm.py
@@ -21,7 +21,8 @@ def fsm(
     """Generate text that follows a Finite State Machine.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Instead, use the `Generator` object instead as such:
+    Support for it will be removed in v1.5.0.
+    Use the `Generator` object instead:
 
     ```python
     import interegular
@@ -52,7 +53,7 @@ def fsm(
     warnings.warn("""
         The `fsm` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, use the `Generator` object instead as such:
+        Use the `Generator` object instead:
         ```python
         from outlines import Generator
         from outlines.types import FSM
diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
index 1e3c9f5a0..4a0b15ba3 100644
--- a/outlines/v0_legacy/generate/json.py
+++ b/outlines/v0_legacy/generate/json.py
@@ -23,11 +23,12 @@ def json(
     schema_object: Union[str, object, Callable],
     sampler: Sampler = multinomial(),
     whitespace_pattern: Optional[str] = None,
-) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter]:
+) -> GeneratorV0Adapter:
     """Generate structured text that follows a JSON Schema.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Instead, use the `Generator` object instead as such:
+    Support for it will be removed in v1.5.0.
+    Use the `Generator` object instead:
 
     ```python
     from outlines import Generator
@@ -62,7 +63,7 @@ def json(
     warnings.warn("""
         The `json` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, use the `Generator` object instead as such:
+        Use the `Generator` object instead:
         ```python
         from outlines import Generator
         from outlines.types import JsonSchema
@@ -82,7 +83,7 @@ def json(
     if is_callable(schema_object):
         json_schema = str(get_schema_from_signature(schema_object))  # type: ignore
     else:
-        json_schema = JsonSchema(schema_object, whitespace_pattern) # type: ignore
+        json_schema = JsonSchema(schema_object, whitespace_pattern)
 
     if isinstance(model, TransformersVision):
         generator = GeneratorVisionV0Adapter(model, json_schema, sampler)
@@ -124,8 +125,8 @@ def json_openai(
     else:
         raise ValueError(
             f"Cannot parse schema {schema_object}. The schema must be either "
-            + "a Pydantic object, a function or a string that contains the JSON "
-            + "Schema specification"
+            + "a Pydantic object or a string that contains the JSON Schema "
+            + "specification"
         )
 
     # create copied, patched model with normalized json schema set
diff --git a/outlines/v0_legacy/generate/regex.py b/outlines/v0_legacy/generate/regex.py
index 7d8df6105..af4beb260 100644
--- a/outlines/v0_legacy/generate/regex.py
+++ b/outlines/v0_legacy/generate/regex.py
@@ -14,7 +14,8 @@ def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
     """Generate structured text in the language of a regular expression.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Instead, use the `Generator` object instead as such:
+    Support for it will be removed in v1.5.0.
+    Use the `Generator` object instead:
 
     ```python
     from outlines import Generator
@@ -43,7 +44,7 @@ def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
     warnings.warn("""
         The `regex` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, use the `Generator` object instead as such:
+        Use the `Generator` object instead:
         ```python
         from outlines import Generator
         from outlines.types import Regex
diff --git a/outlines/v0_legacy/generate/text.py b/outlines/v0_legacy/generate/text.py
index bcc6fe081..b71e00725 100644
--- a/outlines/v0_legacy/generate/text.py
+++ b/outlines/v0_legacy/generate/text.py
@@ -3,6 +3,7 @@
 
 import warnings
 from typing import Union
+
 from outlines.models.openai import OpenAI
 from outlines.v0_legacy.models.transformers_vision import TransformersVision
 from outlines.v0_legacy.generate.api import (
@@ -12,13 +13,14 @@
 from outlines.v0_legacy.samplers import Sampler, multinomial
 
 
-def text(
-        model, sampler: Sampler = multinomial()
-    ) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter, OpenAI]:
+def text(model, sampler: Sampler = multinomial()) -> Union[
+    GeneratorV0Adapter, GeneratorVisionV0Adapter, OpenAI
+]:
     """Generate unconstrained text.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Instead, use the `Generator` object instead as such:
+    Support for it will be removed in v1.5.0.
+    Use the `Generator` object instead:
 
     ```python
     from outlines import Generator
@@ -43,7 +45,7 @@ def text(
     warnings.warn("""
         The `text` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, use the `Generator` object instead as such:
+        Use the `Generator` object instead:
         ```python
         from outlines import Generator
         generator = Generator(model)

From 2ad12c4b5363e0f1e8c1e4c328341fa5a3bae52e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 8 Apr 2025 17:02:23 +0200
Subject: [PATCH 307/505] Add more tests to increase test coverage

---
 outlines/__init__.py                          |  22 +--
 outlines/caching.py                           |   8 +-
 outlines/models/llamacpp.py                   |  14 +-
 outlines/models/openai.py                     |  22 +--
 outlines/models/transformers.py               |  12 +-
 outlines/processors/guide.py                  |   4 -
 outlines/templates.py                         |  22 +--
 outlines/types/dsl.py                         |   7 +-
 outlines/types/utils.py                       |   4 +-
 outlines/v0_legacy/base.py                    |   4 +-
 outlines/v0_legacy/function.py                |  89 ++++++-----
 outlines/v0_legacy/generate/json.py           |   5 +-
 outlines/v0_legacy/models/exllamav2.py        |   4 +-
 outlines/v0_legacy/models/llamacpp.py         |   2 +-
 outlines/v0_legacy/models/transformers.py     |   4 +-
 .../v0_legacy/models/transformers_vision.py   |  11 +-
 pyproject.toml                                |  20 ++-
 tests/models/test_llamacpp.py                 |  40 +++--
 tests/models/test_llamacpp_tokenizer.py       | 133 ++++++++++++++++
 tests/models/test_llamacpp_type_adapter.py    |  41 +++++
 tests/models/test_transformers.py             |  27 +++-
 ...st_transformers_multimodal_type_adapter.py |  48 ++++++
 tests/models/test_transformers_tokenizer.py   | 142 ++++++++++++++++++
 .../models/test_transformers_type_adapter.py  |  50 ++++++
 tests/processors/test_guide.py                |  15 ++
 tests/processors/test_structured.py           |   1 +
 tests/test_applications.py                    |   9 ++
 tests/test_cache.py                           |   9 ++
 tests/test_templates.py                       | 122 ++++++++++++++-
 tests/types/test_dsl.py                       | 137 +++++++++++++++--
 tests/types/test_types_utils.py               |   7 +
 tests/v0_legacy/generate/test_api_legacy.py   |  79 +++++++---
 .../v0_legacy/generate/test_choice_legacy.py  |   7 +
 tests/v0_legacy/generate/test_json_legacy.py  |   9 ++
 .../v0_legacy/models/test_exllamav2_legacy.py |  14 +-
 .../v0_legacy/models/test_llamacpp_legacy.py  |   8 +-
 .../models/test_transformers_legacy.py        |   2 +-
 tests/v0_legacy/test_function_legacy.py       |  41 +++++
 38 files changed, 1011 insertions(+), 184 deletions(-)
 create mode 100644 tests/models/test_llamacpp_tokenizer.py
 create mode 100644 tests/models/test_llamacpp_type_adapter.py
 create mode 100644 tests/models/test_transformers_multimodal_type_adapter.py
 create mode 100644 tests/models/test_transformers_tokenizer.py
 create mode 100644 tests/models/test_transformers_type_adapter.py

diff --git a/outlines/__init__.py b/outlines/__init__.py
index 14c90d090..8234b0c14 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -9,10 +9,6 @@
 from outlines.applications import Application
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.generator import Generator
-from outlines.templates import Template, prompt
-from outlines.types import regex, json_schema, cfg
-from outlines.templates import Vision
-
 from outlines.models import (
     from_dottxt,
     from_openai,
@@ -24,6 +20,17 @@
     from_mlxlm,
     from_vllm,
 )
+from outlines.templates import Template, prompt
+from outlines.types import regex, json_schema, cfg
+from outlines.templates import Vision
+
+from .v0_legacy import (
+    generate,
+    samplers,
+    models as legacy_models,
+    function,
+)
+from .v0_legacy.function import Function
 
 model_list = [
     "from_anthropic",
@@ -54,13 +61,6 @@
 
 # v0 legacy
 
-# flake8: noqa: E402
-from .v0_legacy import generate
-from .v0_legacy import samplers
-from .v0_legacy import models as legacy_models
-from .v0_legacy import function
-from .v0_legacy.function import Function
-
 generate_module = ModuleType("generate")
 generate_module.__dict__.update(generate.__dict__)
 
diff --git a/outlines/caching.py b/outlines/caching.py
index 6882bac6b..5e9d5b679 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -12,7 +12,7 @@
 _caching_enabled = True
 
 
-class CloudpickleDisk(Disk):
+class CloudpickleDisk(Disk): # pragma: no cover
     def __init__(self, directory, compress_level=1, **kwargs):
         self.compress_level = compress_level
         super().__init__(directory, **kwargs)
@@ -57,11 +57,11 @@ def get_cache():
     if outlines_cache_dir:
         # OUTLINES_CACHE_DIR takes precendence
         cache_dir = outlines_cache_dir
-    elif xdg_cache_home:
+    elif xdg_cache_home:  # pragma: no cover
         cache_dir = os.path.join(xdg_cache_home, ".cache", "outlines")
     elif home_dir != "/":
         cache_dir = os.path.join(home_dir, ".cache", "outlines")
-    else:
+    else:  # pragma: no cover
         # home_dir may be / inside a docker container without existing user
         tempdir = tempfile.gettempdir()
         cache_dir = os.path.join(tempdir, ".cache", "outlines")
@@ -108,7 +108,7 @@ def decorator(cached_function: Callable):
 
         base = (full_name(cached_function),)
 
-        if asyncio.iscoroutinefunction(cached_function):
+        if asyncio.iscoroutinefunction(cached_function):  # pragma: no cover
 
             async def wrapper(*args, **kwargs):
                 if not _caching_enabled:
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 9412665e4..be501ae2c 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,4 +1,3 @@
-import pickle
 import warnings
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Dict, Iterator, List, Optional, Set, Tuple, Union
@@ -66,7 +65,7 @@ def convert_token_to_string(self, token: str) -> str:
             from transformers.file_utils import SPIECE_UNDERLINE
 
             token_str = self._hf_tokenizer.convert_tokens_to_string([token])
-            if token.startswith(SPIECE_UNDERLINE) or token == "<0x20>":
+            if token.startswith(SPIECE_UNDERLINE) or token == "<0x20>": # pragma: no cover
                 token_str = " " + token_str
             return token_str
         else:
@@ -78,8 +77,15 @@ def __eq__(self, other):
         return self.__getstate__() == other.__getstate__()
 
     def __hash__(self):
+        # We create a custom hash as pickle.dumps(self) is not stable
         if self._hash is None:
-            self._hash = hash(pickle.dumps(self))
+            self._hash = hash((
+                tuple(sorted(self.vocabulary.items())),
+                self.eos_token_id,
+                self.eos_token,
+                self.pad_token_id,
+                tuple(sorted(self.special_tokens)),
+            ))
         return self._hash
 
     def __getstate__(self):
@@ -224,7 +230,7 @@ def token_generator() -> Iterator[str]:
 
         return token_generator()
 
-    def load_lora(self, adapter_path: str):
+    def load_lora(self, adapter_path: str):  # pragma: no cover
         warnings.warn("""
             The `load_lora` method is deprecated starting from v1.0.0.
             Support for it will be removed in v1.5.0.
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 84c219aeb..2f9408e45 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -4,15 +4,7 @@
 import functools
 import json
 import warnings
-from typing import (
-    Any,
-    Callable,
-    Dict,
-    List,
-    Optional,
-    TYPE_CHECKING,
-    Union
-)
+from typing import Callable, Dict, List, Optional, Union, TYPE_CHECKING
 from dataclasses import asdict, dataclass, field, replace
 
 from pydantic import BaseModel, TypeAdapter
@@ -182,7 +174,7 @@ class OpenAI(Model):
     def __init__(
         self,
         client: Union["OpenAIClient", "AzureOpenAIClient"],
-        model_name: Any,
+        model_name: Union[str, "OpenAIConfig"],
         **kwargs
     ):
         # legacy mode
@@ -286,7 +278,7 @@ def stream(self, *args, **kwargs):
 
     def new_with_replacements(self, **kwargs):
         if hasattr(self, "legacy_instance"):
-            return self.legacy_instance.new_with_replacements(**kwargs)
+            return self.legacy_instance.new_with_replacements(self, **kwargs)
         raise NotImplementedError("This method is only available in legacy mode")
 
     def __str__(self):
@@ -459,9 +451,11 @@ def stream(self, *args, **kwargs):
             "Streaming is currently not supported for the OpenAI API"
         )
 
-    def new_with_replacements(self, **kwargs):
-        new_instance = copy.copy(self)
-        new_instance.config = replace(new_instance.config, **kwargs)
+    def new_with_replacements(self, model, **kwargs):
+        new_instance = copy.copy(model)
+        new_instance.legacy_instance.config = replace(
+            new_instance.legacy_instance.config, **kwargs
+        )
         return new_instance
 
     def __str__(self):
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 8a48961cc..4bece9546 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -20,28 +20,28 @@ def get_llama_tokenizer_types():
     """
     try:
         from transformers.models.llama import LlamaTokenizer
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class LlamaTokenizer:  # type: ignore
             pass
 
     try:
         from transformers.models.llama import LlamaTokenizerFast
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class LlamaTokenizerFast:  # type: ignore
             pass
 
     try:
         from transformers.models.code_llama import CodeLlamaTokenizer
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class CodeLlamaTokenizer:  # type: ignore
             pass
 
     try:
         from transformers.models.code_llama import CodeLlamaTokenizerFast
-    except ImportError:
+    except ImportError:  # pragma: no cover
 
         class CodeLlamaTokenizerFast:  # type: ignore
             pass
@@ -230,6 +230,8 @@ def generate(self, model_input, output_type, **inference_kwargs):
         prompts, inputs = self._prepare_model_inputs(model_input, output_type)
         logits_processor = self.type_adapter.format_output_type(output_type)
 
+        print(inference_kwargs)
+
         generated_ids = self._generate_output_seq(
             prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )
@@ -280,7 +282,7 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
                 self.tokenizer.decode(generated_ids[i])
                 for i in range(len(generated_ids))
             ]
-        else:
+        else:  # pragma: no cover
             raise TypeError(
                 f"Generated outputs aren't 1D, 2D or 3D, but instead are {generated_ids.shape}"
             )
diff --git a/outlines/processors/guide.py b/outlines/processors/guide.py
index d46228fe9..983050c09 100644
--- a/outlines/processors/guide.py
+++ b/outlines/processors/guide.py
@@ -176,10 +176,6 @@ def iter_valid_token_ids(
         int
             Valid token ids.
         """
-        if state.parser_state is None:
-            yield self.eos_token_id
-            return
-
         for token_id in candidate_token_ids:
             if token_id == self.eos_token_id:
                 if self.can_terminate_state(state):
diff --git a/outlines/templates.py b/outlines/templates.py
index 325bcfe04..a1df14aa8 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -27,17 +27,14 @@ class Vision:
 
     def __post_init__(self):
         image = self.image
+
+        if not image.format:
+            raise TypeError("Could not read the format of the image passed to the model.")
+
         buffer = BytesIO()
         image.save(buffer, format=image.format)
         self.image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
-
-        image_format = image.format
-        if image_format is not None:
-            self.image_format = f"image/{image_format.lower()}"
-        else:
-            raise TypeError(
-                "Could not read the format of the image passed to the model."
-            )
+        self.image_format = f"image/{image.format.lower()}"
 
 
 @dataclass
@@ -230,7 +227,7 @@ def create_jinja_env(
     - `source`: get a function's source code
     - `signature`: get a function's signature
     - `args`: get a function's arguments
-    - `schema`: isplay a JSON Schema
+    - `schema`: display a JSON Schema
 
     Users may pass additional filters, and/or override existing ones.
 
@@ -314,7 +311,7 @@ def get_fn_source(fn: Callable):
     re_search = re.search(re.compile(r"(\bdef\b.*)", re.DOTALL), source)
     if re_search is not None:
         source = re_search.group(0)
-    else:
+    else:  # pragma: no cover
         raise TypeError("Could not read the function's source code")
 
     return source
@@ -327,7 +324,7 @@ def get_fn_signature(fn: Callable):
 
     source = textwrap.dedent(inspect.getsource(fn))
     re_search = re.search(re.compile(r"\(([^)]+)\)"), source)
-    if re_search is None:
+    if re_search is None:  # pragma: no cover
         signature = ""
     else:
         signature = re_search.group(1)
@@ -351,9 +348,6 @@ def get_schema_dict(model: Dict):
 @get_schema.register(type(pydantic.BaseModel))
 def get_schema_pydantic(model: Type[pydantic.BaseModel]):
     """Return the schema of a Pydantic model."""
-    if not isinstance(model, type(pydantic.BaseModel)):
-        raise TypeError("The `schema` filter only applies to Pydantic models.")
-
     if hasattr(model, "model_json_schema"):
         def_key = "$defs"
         raw_schema = model.model_json_schema()
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 6b16c5897..66a0d99df 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -51,12 +51,11 @@
     is_typing_list,
     is_typing_tuple,
     is_typing_dict,
-    is_interegular_fsm,
 )
 
-if sys.version_info >= (3, 12):
+if sys.version_info >= (3, 12):  # pragma: no cover
     from typing import _TypedDictMeta  # type: ignore
-else:
+else:  # pragma: no cover
     from typing_extensions import _TypedDictMeta  # type: ignore
 
 
@@ -299,7 +298,7 @@ def __eq__(self, other):
             self_dict = json.loads(self.schema)
             other_dict = json.loads(other.schema)
             return self_dict == other_dict
-        except json.JSONDecodeError:
+        except json.JSONDecodeError:  # pragma: no cover
             return self.schema == other.schema
 
     @classmethod
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
index e1be024ba..f46b516cc 100644
--- a/outlines/types/utils.py
+++ b/outlines/types/utils.py
@@ -22,9 +22,9 @@
 from genson import SchemaBuilder
 from pydantic import BaseModel, create_model
 
-if sys.version_info >= (3, 12):
+if sys.version_info >= (3, 12): # pragma: no cover
     from typing import _TypedDictMeta  # type: ignore
-else:
+else: # pragma: no cover
     from typing_extensions import _TypedDictMeta  # type: ignore
 
 
diff --git a/outlines/v0_legacy/base.py b/outlines/v0_legacy/base.py
index 8d767391f..60bf655a1 100644
--- a/outlines/v0_legacy/base.py
+++ b/outlines/v0_legacy/base.py
@@ -7,7 +7,7 @@
 
 # Allow nested loops for running in notebook. We don't enable it globally as it
 # may interfere with other libraries that use asyncio.
-if hasattr(builtins, "__IPYTHON__"):
+if hasattr(builtins, "__IPYTHON__"): # pragma: no cover
     try:
         import nest_asyncio
 
@@ -45,7 +45,7 @@ def __init__(self, func: Callable, signature: Optional[str] = None):
                 _parse_input_dimensions,
                 _update_dim_sizes,
             )
-        else:
+        else: # pragma: no cover
             from numpy.lib.function_base import (
                 _calculate_shapes,
                 _parse_gufunc_signature,
diff --git a/outlines/v0_legacy/function.py b/outlines/v0_legacy/function.py
index e0e10c3a0..d15ec702f 100644
--- a/outlines/v0_legacy/function.py
+++ b/outlines/v0_legacy/function.py
@@ -9,10 +9,50 @@
 from outlines.v0_legacy.generate.api import GeneratorV0Adapter
 from outlines.v0_legacy.models import transformers
 
-if TYPE_CHECKING:
+if TYPE_CHECKING: # pragma: no cover
     from outlines.templates import Template
 
 
+def function_warning():
+    warnings.warn("""
+        The `Function` class is deprecated starting from v1.0.0.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Instead, you should use the `outlines.Application` class that
+        implements a similar functionality.
+        An `Application` is initialized with a prompt template and an
+        output type. It can then be called with a model and a prompt
+        to generate a response that follows the format specified by the
+        output type.
+        For example:
+        ```python
+        from pydantic import BaseModel
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+        from outlines import models, Application
+        from outlines.types import JsonType
+        from outlines.templates import Template
+
+        class OutputModel(BaseModel):
+            result: int
+
+        model = models.from_transformers(
+            AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+            AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+        )
+
+        template_string = "What is 2 times {{ num }}?"
+        template = Template.from_string(template_string)
+
+        application = Application(template, JsonType(OutputModel))
+
+        result = application(model, num=3)
+        print(result)  # Expected output: { "result" : 6 }
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2
+    )
+
+
 @dataclass
 class Function:
     """Represents an Outlines function.
@@ -23,19 +63,18 @@ class Function:
     prompt template.
 
     """
-
     prompt_template: "Template"
     schema: Union[str, Callable, object]
     model_name: str
     generator: Optional["GeneratorV0Adapter"] = None
 
     def __post_init__(self):
-        _warn()
+        function_warning()
 
     @classmethod
     def from_github(cls, program_path: str, function_name: str = "fn"):
         """Load a function stored on GitHub"""
-        _warn()
+        function_warning()
         program_content = download_from_github(program_path)
         function = extract_function_from_file(program_content, function_name)
 
@@ -75,46 +114,6 @@ def __call__(self, *args, **kwargs):
             return self.generator(prompt)
 
 
-def _warn():
-    warnings.warn("""
-        The `Function` class is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
-        Instead, you should use the `outlines.Application` class that
-        implements a similar functionality.
-        An `Application` is initialized with a prompt template and an
-        output type. It can then be called with a model and a prompt
-        to generate a response that follows the format specified by the
-        output type.
-        For example:
-        ```python
-        from pydantic import BaseModel
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-        from outlines import models, Application
-        from outlines.types import JsonType
-        from outlines.templates import Template
-
-        class OutputModel(BaseModel):
-            result: int
-
-        model = models.from_transformers(
-            AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-            AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-        )
-
-        template_string = "What is 2 times {{ num }}?"
-        template = Template.from_string(template_string)
-
-        application = Application(template, JsonType(OutputModel))
-
-        result = application(model, num=3)
-        print(result)  # Expected output: { "result" : 6 }
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2
-    )
-
-
 def download_from_github(short_path: str):
     """Download the file in which the function is stored on GitHub."""
     GITHUB_BASE_URL = "https://raw.githubusercontent.com"
@@ -150,7 +149,7 @@ def extract_function_from_file(content: str, function_name: str) -> Tuple[Callab
     spec = importlib.util.spec_from_loader(
         "outlines_function", loader=None, origin="github"
     )
-    if spec is not None:
+    if spec is not None:  # pragma: no cover
         module = importlib.util.module_from_spec(spec)
         exec(content, module.__dict__)
 
diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
index 4a0b15ba3..3c71ed8d0 100644
--- a/outlines/v0_legacy/generate/json.py
+++ b/outlines/v0_legacy/generate/json.py
@@ -78,10 +78,10 @@ def json(
     )
 
     if isinstance(model, OpenAI):
-        return json_openai(model, schema_object, sampler, whitespace_pattern)
+        return json_openai(model, schema_object, sampler)
 
     if is_callable(schema_object):
-        json_schema = str(get_schema_from_signature(schema_object))  # type: ignore
+        json_schema = JsonSchema(get_schema_from_signature(schema_object))  # type: ignore
     else:
         json_schema = JsonSchema(schema_object, whitespace_pattern)
 
@@ -106,7 +106,6 @@ def json_openai(
     model,
     schema_object: Union[str, object, Callable],
     sampler: Sampler = multinomial(),
-    whitespace_pattern: Optional[str] = None,
 ):
     if not isinstance(sampler, multinomial):
         raise NotImplementedError(
diff --git a/outlines/v0_legacy/models/exllamav2.py b/outlines/v0_legacy/models/exllamav2.py
index f208fc555..4b4039acd 100644
--- a/outlines/v0_legacy/models/exllamav2.py
+++ b/outlines/v0_legacy/models/exllamav2.py
@@ -15,7 +15,7 @@
 
 if TYPE_CHECKING:
     import torch.LongTensor
-    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler
+    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler  # type: ignore
 
 
 def exllamav2_params_adapter(sampling_params: dict, generation_params: dict, kwargs: dict) -> dict:
@@ -332,7 +332,7 @@ def exl2(
 
     """
     try:
-        from exllamav2 import (
+        from exllamav2 import (  # type: ignore
             ExLlamaV2,
             ExLlamaV2Cache,
             ExLlamaV2Cache_Q4,
diff --git a/outlines/v0_legacy/models/llamacpp.py b/outlines/v0_legacy/models/llamacpp.py
index 86233a63b..cfdad98e0 100644
--- a/outlines/v0_legacy/models/llamacpp.py
+++ b/outlines/v0_legacy/models/llamacpp.py
@@ -78,7 +78,7 @@ def llamacpp_params_adapter(
         kwargs["repeat_penalty"] = 1.0
 
     # The choice to stream or not should happen via the high-level API
-    kwargs["stream"] = False
+    kwargs.pop("stream", None)
 
     return kwargs
 
diff --git a/outlines/v0_legacy/models/transformers.py b/outlines/v0_legacy/models/transformers.py
index b275eca12..379d64f2a 100644
--- a/outlines/v0_legacy/models/transformers.py
+++ b/outlines/v0_legacy/models/transformers.py
@@ -127,7 +127,7 @@ def transformers(
     if model_class is None or tokenizer_class is None:
         try:
             from transformers import AutoModelForCausalLM, AutoTokenizer
-        except ImportError:
+        except ImportError:  # pragma: no cover
             raise ImportError(
                 "The `transformers` library needs to be installed in order to use `transformers` models."
             )
@@ -176,7 +176,7 @@ def mamba(
     try:
         from transformers import MambaForCausalLM
 
-    except ImportError:
+    except ImportError:  # pragma: no cover
         raise ImportError(
             "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
         )
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
index 96edf313e..4d46a0618 100644
--- a/outlines/v0_legacy/models/transformers_vision.py
+++ b/outlines/v0_legacy/models/transformers_vision.py
@@ -88,7 +88,7 @@ def transformers_vision(
     if processor_class is None or tokenizer_class is None:
         try:
             from transformers import AutoProcessor, AutoTokenizer
-        except ImportError:
+        except ImportError:  # pragma: no cover
             raise ImportError(
                 "The `transformers` library needs to be installed in order to use `transformers` models."
             )
@@ -105,13 +105,6 @@ def transformers_vision(
     processor_kwargs.setdefault("padding_side", "left")
     processor_kwargs.setdefault("pad_token", "[PAD]")
     processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-    if tokenizer_class is None:
-        if getattr(processor, "tokenizer", None):
-            tokenizer = processor.tokenizer
-        else:
-            tokenizer = AutoTokenizer.from_pretrained(model_name, **processor_kwargs)
-    else:
-        tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
+    tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
 
     return TransformersVision(model, tokenizer, processor)
diff --git a/pyproject.toml b/pyproject.toml
index 6956d4b96..352fd5a91 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -165,27 +165,37 @@ module = [
 ignore_missing_imports = true
 
 [tool.coverage.run]
+# we omit the files that require a GPU or Apple Silicon
+# as well as the models that make API calls
 omit = [
     "outlines/_version.py",
-    "tests/*",
+    "outlines/models/anthropic.py",
+    "outlines/models/dottxt.py",
+    "outlines/models/gemini.py",
     "outlines/models/mlxlm.py",
+    "outlines/models/openai.py",
     "outlines/models/vllm.py",
     "outlines/processors/tensor_adapters/mlx.py",
     "outlines/serve/serve.py",
+    "outlines/v0_legacy/models/anthropic.py",
+    "outlines/v0_legacy/models/dottxt.py",
+    "outlines/v0_legacy/models/exllamav2.py",
+    "outlines/v0_legacy/models/gemini.py",
+    "outlines/v0_legacy/models/mlxlm.py",
+    "outlines/v0_legacy/models/openai.py",
+    "outlines/v0_legacy/models/vllm.py",
+    "tests/*",
 ]
 branch = true
 relative_files = true
 
 [tool.coverage.report]
-omit = [
-    "tests/*",
-]
+show_missing = true
 exclude_lines = [
     "pragma: no cover",
     "if TYPE_CHECKING:",
     "\\.\\.\\.",
 ]
-show_missing = true
 
 [tool.diff_cover]
 compare_branch = "origin/main"
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index af6802679..6a77528f6 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -12,7 +12,7 @@
     from_llamacpp
 )
 from outlines.processors import RegexLogitsProcessor
-from outlines.types.dsl import Regex
+from outlines.types.dsl import Regex, CFG
 
 
 def test_load_model():
@@ -70,24 +70,24 @@ class Foo(Enum):
     assert result == "Foo" or result == "Bar"
 
 
+def test_llamacpp_cfg(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="CFG generation is not supported for LlamaCpp"
+    ):
+        model("Respond with one word. Not more.", CFG('start: "a"'))
+
+
 def test_llamacpp_text_stop(model):
     result = model.generate("Write the letter a.", None, stop="a", max_tokens=100)
     assert "a" not in result
 
 
-def test_llamacpp_stream_text_stop(model):
-    generator = model.stream("Write the letter a.", None, stop="a", max_tokens=100)
-
-    result = next(generator)
-    assert isinstance(result, str)
-    assert result != "a"
-
-
 def test_llamacpp_stream_simple(model):
     generator = model.stream("Respond with one word. Not more.", None)
 
-    x = next(generator)
-    assert isinstance(x, str)
+    for x in generator:
+        assert isinstance(x, str)
 
 
 def test_llamacpp_stream_regex(model):
@@ -107,6 +107,14 @@ class Foo(BaseModel):
     assert x == "{"
 
 
+def test_llamacpp_stream_cfg(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="CFG generation is not supported for LlamaCpp"
+    ):
+        model.stream("Respond with one word. Not more.", CFG('start: "a"'))
+
+
 def test_llamacpp_stream_choice(model):
     class Foo(Enum):
         bar = "Bar"
@@ -115,4 +123,12 @@ class Foo(Enum):
     generator = model.stream("foo?", Foo)
 
     x = next(generator)
-    assert isinstance(x, str)
+    assert x[0] in ("B", "F")
+
+
+def test_llamacpp_stream_text_stop(model):
+    generator = model.stream("Write the letter a.", None, stop="a", max_tokens=100)
+
+    result = next(generator)
+    assert isinstance(result, str)
+    assert result != "a"
diff --git a/tests/models/test_llamacpp_tokenizer.py b/tests/models/test_llamacpp_tokenizer.py
new file mode 100644
index 000000000..dbf2507f3
--- /dev/null
+++ b/tests/models/test_llamacpp_tokenizer.py
@@ -0,0 +1,133 @@
+import pytest
+import sys
+from unittest.mock import patch
+
+import llama_cpp
+import transformers
+
+from outlines.models.llamacpp import LlamaCppTokenizer
+
+
+@pytest.fixture
+def model():
+    model = llama_cpp.Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+    setattr(
+        model.tokenizer_,
+        "hf_tokenizer",
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return model
+
+
+@pytest.fixture
+def model_no_hf_tokenizer():
+    model = llama_cpp.Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+    del model.tokenizer_
+    return model
+
+
+@pytest.fixture
+def different_model():
+    model = llama_cpp.Llama.from_pretrained(
+        "TheBloke/phi-2-GGUF",
+        "phi-2.Q4_K_M.gguf",
+    )
+    return model
+
+
+@pytest.fixture
+def tokenizer(model):
+    return LlamaCppTokenizer(model)
+
+
+@pytest.fixture
+def another_tokenizer(model):
+    return LlamaCppTokenizer(model)
+
+
+@pytest.fixture
+def tokenizer_no_hf_tokenizer(model_no_hf_tokenizer):
+    return LlamaCppTokenizer(model_no_hf_tokenizer)
+
+
+@pytest.fixture
+def different_tokenizer(different_model):
+    return LlamaCppTokenizer(different_model)
+
+
+def test_llama_cpp_tokenizer_init(tokenizer, tokenizer_no_hf_tokenizer):
+    # regular case
+    assert tokenizer.eos_token_id is not None
+    assert tokenizer.pad_token_id is not None
+    assert isinstance(tokenizer.vocabulary, dict)
+
+    # tokenizer with no hf_tokenizer
+    assert tokenizer_no_hf_tokenizer.eos_token_id is not None
+    assert tokenizer_no_hf_tokenizer.pad_token_id is not None
+    assert isinstance(tokenizer_no_hf_tokenizer.vocabulary, dict)
+
+
+def test_llama_cpp_tokenizer_encode(tokenizer):
+    # batch case
+    with pytest.raises(NotImplementedError):
+        token_ids, attention_mask = tokenizer.encode(["foo", "bar"])
+
+    # regular case
+    token_ids, attention_mask = tokenizer.encode("Hello, world!")
+    assert token_ids is not None
+    assert attention_mask is not None
+    assert len(token_ids) == len(attention_mask)
+
+
+def test_llama_cpp_tokenizer_decode(tokenizer):
+    token_ids, _ = tokenizer.encode("Hello, world!")
+    decoded_text = tokenizer.decode(token_ids)
+    assert isinstance(decoded_text, list)
+    assert "".join(decoded_text).strip() == "Hello, world!"
+
+
+def test_llama_cpp_tokenizer_convert_token_to_string(
+    tokenizer,
+    tokenizer_no_hf_tokenizer
+):
+    # with self._hf_tokenizer
+    token_str = tokenizer.convert_token_to_string("<0x20>")
+    assert isinstance(token_str, str)
+
+    # without self._hf_tokenizer
+    token_str = tokenizer_no_hf_tokenizer.convert_token_to_string("<0x20>")
+    assert isinstance(token_str, str)
+
+
+def test_llama_cpp_tokenizer_eq(tokenizer, another_tokenizer, different_tokenizer):
+    assert not tokenizer == 1
+    assert tokenizer == another_tokenizer
+    assert tokenizer != different_tokenizer
+
+
+def test_llama_cpp_tokenizer_hash(tokenizer, another_tokenizer, different_tokenizer):
+    assert isinstance(hash(tokenizer), int)
+    assert hash(tokenizer) == hash(another_tokenizer)
+    assert hash(tokenizer) != hash(different_tokenizer)
+
+
+def test_llama_cpp_tokenizer_getstate(tokenizer):
+    state = tokenizer.__getstate__()
+    assert isinstance(state, tuple)
+    assert len(state) == 5
+    assert isinstance(state[0], dict)
+    assert isinstance(state[1], int)
+    assert isinstance(state[2], str)
+    assert isinstance(state[3], int)
+    assert isinstance(state[4], list)
+
+
+def test_llama_cpp_tokenizer_setstate(tokenizer):
+    with pytest.raises(NotImplementedError):
+        tokenizer.__setstate__(None)
diff --git a/tests/models/test_llamacpp_type_adapter.py b/tests/models/test_llamacpp_type_adapter.py
new file mode 100644
index 000000000..e49586d5d
--- /dev/null
+++ b/tests/models/test_llamacpp_type_adapter.py
@@ -0,0 +1,41 @@
+import pytest
+
+import llama_cpp
+from llama_cpp import LogitsProcessorList
+
+from outlines.models.llamacpp import LlamaCppTypeAdapter, LlamaCppTokenizer
+from outlines.processors.structured import RegexLogitsProcessor
+
+
+@pytest.fixture
+def adapter():
+    return LlamaCppTypeAdapter()
+
+
+@pytest.fixture
+def logits_processor():
+    model = llama_cpp.Llama.from_pretrained(
+        "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+    return RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=LlamaCppTokenizer(model),
+        tensor_library_name="numpy",
+    )
+
+
+def test_llamacpp_type_adapter_format_input(adapter):
+    # anything else than a string
+    with pytest.raises(NotImplementedError):
+        adapter.format_input(["Hello, world!"])
+
+    # string case
+    assert adapter.format_input("Hello, world!") == "Hello, world!"
+
+
+def test_llamacpp_type_adapter_format_output_type(adapter, logits_processor):
+    formatted = adapter.format_output_type(logits_processor)
+    assert isinstance(formatted, LogitsProcessorList)
+    assert len(formatted) == 1
+    assert isinstance(formatted[0], RegexLogitsProcessor)
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 49e88f538..96d8d10f2 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -1,8 +1,8 @@
 import re
 from enum import Enum
 
-import pytest
 from pydantic import BaseModel
+import pytest
 import transformers
 
 import outlines
@@ -13,9 +13,18 @@
 )
 from outlines.types import Regex
 
+
 TEST_MODEL = "erwanf/gpt2-mini"
-TEST_MODEL_SEQ2SEQ = "hf-internal-testing/tiny-random-t5"
 TEST_MODEL_MAMBA = "hf-internal-testing/tiny-random-MambaForCausalLM"
+TEST_MODEL_BART = "sshleifer/bart-tiny-random"
+
+
+def test_transformers_instantiate_invalid():
+    with pytest.raises(ValueError):
+        outlines.from_transformers(
+            transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
+            int,
+        )
 
 
 def test_transformers_instantiate_simple():
@@ -79,15 +88,27 @@ def model():
     return model
 
 
+@pytest.fixture
+def model_bart():
+    model = outlines.from_transformers(
+        transformers.BartForConditionalGeneration.from_pretrained(TEST_MODEL_BART),
+        transformers.BartTokenizer.from_pretrained(TEST_MODEL_BART),
+    )
+    return model
+
+
 def test_transformers_simple(model):
     result = model.generate("Respond with one word. Not more.", None)
     assert isinstance(result, str)
 
 
-def test_transformers_call(model):
+def test_transformers_call(model, model_bart):
     result = model("Respond with one word. Not more.")
     assert isinstance(result, str)
 
+    result = model_bart("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
 
 def test_transformers_inference_kwargs(model):
     result = model("Respond with one word. Not more.", max_new_tokens=100)
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
new file mode 100644
index 000000000..28569d25f
--- /dev/null
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -0,0 +1,48 @@
+import pytest
+
+import outlines
+import transformers
+from transformers import LogitsProcessorList
+
+from outlines.models.transformers import TransformersMultiModalTypeAdapter
+from outlines.processors.structured import RegexLogitsProcessor
+
+
+@pytest.fixture
+def adapter():
+    return TransformersMultiModalTypeAdapter()
+
+
+@pytest.fixture
+def logits_processor():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=model.tokenizer,
+        tensor_library_name="torch",
+    )
+
+
+def test_transformers_multimodal_type_adapter_format_input(adapter):
+    with pytest.raises(NotImplementedError):
+        adapter.format_input("hello")
+
+    with pytest.raises(ValueError):
+        adapter.format_input({"foo": "bar"})
+
+    assert adapter.format_input({"text": "foo"}) == {"text": "foo"}
+
+
+def test_transformers_multimodal_type_adapter_format_output_type(
+    adapter, logits_processor
+):
+    formatted = adapter.format_output_type(logits_processor)
+    assert isinstance(formatted, LogitsProcessorList)
+    assert len(formatted) == 1
+    assert isinstance(formatted[0], RegexLogitsProcessor)
+
+    formatted = adapter.format_output_type(None)
+    assert formatted is None
diff --git a/tests/models/test_transformers_tokenizer.py b/tests/models/test_transformers_tokenizer.py
new file mode 100644
index 000000000..bf3452302
--- /dev/null
+++ b/tests/models/test_transformers_tokenizer.py
@@ -0,0 +1,142 @@
+import pytest
+
+import transformers
+
+from outlines.models.transformers import (
+    get_llama_tokenizer_types,
+    TransformerTokenizer,
+)
+
+
+TEST_MODEL = "erwanf/gpt2-mini"
+TEST_MODEL_SEQ2SEQ = "hf-internal-testing/tiny-random-t5"
+
+
+@pytest.fixture
+def tokenizer():
+    return transformers.AutoTokenizer.from_pretrained(TEST_MODEL)
+
+
+@pytest.fixture
+def tokenizer_no_pad_token_id(tokenizer):
+    tokenizer.pad_token_id = None
+    return tokenizer
+
+
+@pytest.fixture
+def tokenizer_seq2seq():
+    return transformers.AutoTokenizer.from_pretrained(TEST_MODEL_SEQ2SEQ)
+
+
+@pytest.fixture
+def transformer_tokenizer(tokenizer):
+    return TransformerTokenizer(tokenizer)
+
+
+@pytest.fixture
+def another_transformer_tokenizer(tokenizer):
+    return TransformerTokenizer(tokenizer)
+
+
+@pytest.fixture
+def transformer_tokenizer_seq2seq(tokenizer_seq2seq):
+    return TransformerTokenizer(tokenizer_seq2seq)
+
+
+def test_get_llama_tokenizer_types():
+    tokenizer_types = get_llama_tokenizer_types()
+    assert tokenizer_types[0] is transformers.models.llama.LlamaTokenizer
+    assert tokenizer_types[1] is transformers.models.llama.LlamaTokenizerFast
+    assert tokenizer_types[2] is transformers.models.code_llama.CodeLlamaTokenizer
+    assert tokenizer_types[3] is transformers.models.code_llama.CodeLlamaTokenizerFast
+
+
+def test_transformer_tokenizer_init(
+    tokenizer,
+    tokenizer_no_pad_token_id
+):
+    # tokenizer with a pad_token_id
+    transformer_tokenizer = TransformerTokenizer(tokenizer)
+    assert transformer_tokenizer.tokenizer == tokenizer
+    assert transformer_tokenizer.eos_token_id == tokenizer.eos_token_id
+    assert transformer_tokenizer.pad_token_id == tokenizer.pad_token_id
+    assert transformer_tokenizer.special_tokens == set(tokenizer.all_special_tokens)
+    assert transformer_tokenizer.vocabulary == tokenizer.get_vocab()
+
+    # tokenizer with no pad_token_id
+    transformer_tokenizer_no_pad_token_id = TransformerTokenizer(tokenizer_no_pad_token_id)
+    assert transformer_tokenizer_no_pad_token_id.tokenizer == tokenizer_no_pad_token_id
+    assert transformer_tokenizer_no_pad_token_id.eos_token_id == tokenizer_no_pad_token_id.eos_token_id
+    assert transformer_tokenizer_no_pad_token_id.pad_token_id == tokenizer_no_pad_token_id.eos_token_id
+    assert transformer_tokenizer_no_pad_token_id.special_tokens == set(tokenizer_no_pad_token_id.all_special_tokens)
+    assert transformer_tokenizer_no_pad_token_id.vocabulary == tokenizer_no_pad_token_id.get_vocab()
+
+
+def test_transformer_tokenizer_encode(transformer_tokenizer):
+    input_ids, attention_mask = transformer_tokenizer.encode("Hello, world!")
+    assert input_ids is not None
+    assert attention_mask is not None
+    assert input_ids.shape == attention_mask.shape
+
+
+def test_transformer_tokenizer_decode(transformer_tokenizer):
+    input_ids, _ = transformer_tokenizer.encode("Hello, world!")
+    decoded_text = transformer_tokenizer.decode(input_ids)
+    assert isinstance(decoded_text, list)
+    assert "Hello, world!" in decoded_text[0]
+
+
+def test_transformer_tokenizer_convert_token_to_string(transformer_tokenizer):
+    # regular
+    transformer_tokenizer.is_llama = False
+    token = transformer_tokenizer.tokenizer.tokenize("Hello")[0]
+    string = transformer_tokenizer.convert_token_to_string(token)
+    assert isinstance(string, str)
+    assert "Hello" in string
+
+    # is_llama + <0x20>
+    transformer_tokenizer.is_llama = True
+    string = transformer_tokenizer.convert_token_to_string("<0x20>")
+    assert isinstance(string, str)
+    assert " " in string
+
+
+def test_transformer_tokenizer_eq(
+    transformer_tokenizer,
+    another_transformer_tokenizer,
+    transformer_tokenizer_seq2seq,
+):
+    # different types of object
+    assert transformer_tokenizer.__eq__(1) == NotImplemented
+
+    # regular case
+    assert transformer_tokenizer == another_transformer_tokenizer
+    assert transformer_tokenizer != transformer_tokenizer_seq2seq
+
+    # with model name and kwargs attributes
+    transformer_tokenizer.model_name = "foo"
+    transformer_tokenizer.kwargs = {"foo": "bar"}
+    another_transformer_tokenizer.model_name = "foo"
+    another_transformer_tokenizer.kwargs = {"foo": "bar"}
+    assert transformer_tokenizer == another_transformer_tokenizer
+
+
+def test_transformer_tokenizer_hash(
+    transformer_tokenizer,
+    another_transformer_tokenizer,
+    transformer_tokenizer_seq2seq,
+):
+    assert isinstance(hash(transformer_tokenizer), int)
+    assert hash(transformer_tokenizer) == hash(another_transformer_tokenizer)
+    assert hash(transformer_tokenizer) != hash(transformer_tokenizer_seq2seq)
+
+
+def test_transformer_tokenizer_getstate_setstate(
+    transformer_tokenizer,
+    another_transformer_tokenizer,
+):
+    state = transformer_tokenizer.__getstate__()
+    assert "tokenizer" in state
+
+    another_transformer_tokenizer.__setstate__(state)
+    assert another_transformer_tokenizer == transformer_tokenizer
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
new file mode 100644
index 000000000..44f6ec4e6
--- /dev/null
+++ b/tests/models/test_transformers_type_adapter.py
@@ -0,0 +1,50 @@
+import pytest
+
+import transformers
+from transformers import LogitsProcessorList
+
+import outlines
+from outlines.models.transformers import TransformersTypeAdapter
+from outlines.processors.structured import RegexLogitsProcessor
+
+
+@pytest.fixture
+def adapter():
+    return TransformersTypeAdapter()
+
+
+@pytest.fixture
+def logits_processor():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=model.tokenizer,
+        tensor_library_name="torch",
+    )
+
+
+def test_transformers_type_adapter_format_input(adapter):
+    with pytest.raises(NotImplementedError):
+        adapter.format_input(int)
+
+    assert adapter.format_input("Hello, world!") == "Hello, world!"
+
+    assert adapter.format_input(["Hello, world!", "Hello, world!"]) == [
+        "Hello, world!",
+        "Hello, world!",
+    ]
+
+
+def test_transformers_type_adapter_format_output_type(
+    adapter, logits_processor
+):
+    formatted = adapter.format_output_type(logits_processor)
+    assert isinstance(formatted, LogitsProcessorList)
+    assert len(formatted) == 1
+    assert isinstance(formatted[0], RegexLogitsProcessor)
+
+    formatted = adapter.format_output_type(None)
+    assert formatted is None
diff --git a/tests/processors/test_guide.py b/tests/processors/test_guide.py
index 2d46a2d2d..53503c63e 100644
--- a/tests/processors/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -47,6 +47,8 @@ class MockTokenizer:
     assert fsm.is_final_state(fsm.start_state) is False
     assert fsm.is_final_state(fsm.final_state) is True
 
+    assert isinstance(fsm.copy(), StopAtEOSGuide)
+
 
 def test_regex_vocabulary_error():
     class MockTokenizer:
@@ -295,6 +297,8 @@ def decode(self, token_ids):
     assert guide.must_terminate_state(state)
     assert guide.can_terminate_state(state)
 
+    assert isinstance(guide.copy(), CFGGuide)
+
 
 def test_cfg_early_termination():
     class MockTokenizer:
@@ -522,3 +526,14 @@ def decode(self, token_ids):
     state = guide.get_next_state(state, token_id=4)
     assert guide.must_terminate_state(state)
     assert guide.can_terminate_state(state)
+    assert guide.is_final_state(state)
+
+    cfg_str = """
+        start: CHAR
+        CHAR: /./
+    """
+
+    guide = CFGGuide(cfg_str, tokenizer)
+
+    state = guide.get_next_state(guide.initial_state, token_id=1)
+    assert guide.get_next_state(state, token_id=4).parser_state is None
diff --git a/tests/processors/test_structured.py b/tests/processors/test_structured.py
index 00007b3d3..54afae2b2 100644
--- a/tests/processors/test_structured.py
+++ b/tests/processors/test_structured.py
@@ -132,3 +132,4 @@ def test_structured_cfg_logits_processor_call(tokenizer):
     input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
     output = processor(input_ids, logits)
     assert output.shape == (1, len(vocab_token_ids))
+    processor(input_ids, logits)
diff --git a/tests/test_applications.py b/tests/test_applications.py
index 09eb5bd97..fda179195 100644
--- a/tests/test_applications.py
+++ b/tests/test_applications.py
@@ -47,6 +47,15 @@ def test_application_initialization():
     assert application.generator is None
 
 
+def test_application_generator_no_model():
+    template = Template.from_string("Test {{ value }}")
+    output_type = None
+    application = Application(template, output_type)
+
+    with pytest.raises(ValueError):
+        application(None)
+
+
 def test_application_template_call(model):
     template = Template.from_string("Test {{ value }}")
     output_type = None
diff --git a/tests/test_cache.py b/tests/test_cache.py
index 9987747b0..149d856c0 100644
--- a/tests/test_cache.py
+++ b/tests/test_cache.py
@@ -5,6 +5,15 @@
 
 import diskcache
 import pytest
+from diskcache import Cache, UNKNOWN
+from outlines.caching import CloudpickleDisk
+
+
+@pytest.fixture
+def temp_dir():
+    """Create a temporary directory for testing."""
+    directory = tempfile.mkdtemp()
+    yield directory
 
 
 @pytest.fixture
diff --git a/tests/test_templates.py b/tests/test_templates.py
index 2fda3caa3..cf9d0ed5f 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -1,12 +1,75 @@
+import base64
+import json
 import os
 import tempfile
 from typing import Dict, List
 
 import pytest
+from PIL import Image
+from io import BytesIO
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.templates import Template, build_template_from_string
+from outlines.templates import (
+    Template,
+    build_template_from_string,
+    Vision,
+    get_fn_name,
+    get_fn_args,
+    get_fn_description,
+    get_fn_source,
+    get_fn_signature,
+    get_schema,
+)
+
+
+def sample_function(x, y=2):
+    """This is a sample function."""
+    return x + y
+
+def function_with_annotations(x: int, y: str) -> str:
+    """Function with annotations."""
+    return f"{x} {y}"
+
+def function_with_no_docstring(x, y):
+    return x * y
+
+class CallableClass:
+    def __call__(self):
+        pass
+
+class PydanticClass(BaseModel):
+    foo: str
+
+
+def test_vision_initialization():
+    # Create a simple image for testing
+    image = Image.new("RGB", (10, 10), color="red")
+    image.format = "PNG"
+
+    # Initialize the Vision object
+    vision = Vision(prompt="Test prompt", image=image)
+
+    # Check that the prompt is set correctly
+    assert vision.prompt == "Test prompt"
+
+    # Check that the image is encoded correctly
+    buffer = BytesIO()
+    image.save(buffer, format=image.format)
+    expected_image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
+    assert vision.image_str == expected_image_str
+
+    # Check that the image format is set correctly
+    assert vision.image_format == "image/png"
+
+
+def test_vision_invalid_image_format():
+    # Create an image without a format
+    image = Image.new("RGB", (10, 10), color="blue")
+
+    # Expect a TypeError when the image format is not set
+    with pytest.raises(TypeError, match="Could not read the format"):
+        Vision(prompt="Test prompt", image=image)
 
 
 def render(content: str, **kwargs):
@@ -441,3 +504,60 @@ def test_template_from_str_with_extra_linebreaks():
     """
     template = build_template_from_string(content)
     assert template.render(name="World") == "Hello, World!\n"
+
+
+def test_get_fn_name():
+    with pytest.raises(TypeError):
+        get_fn_name(1)
+    assert get_fn_name(sample_function) == "sample_function"
+    assert get_fn_name(function_with_annotations) == "function_with_annotations"
+    no_name_func = lambda x: x
+    assert get_fn_name(no_name_func) == "<lambda>"
+    assert get_fn_name(CallableClass()) == "CallableClass"
+
+
+def test_get_fn_args():
+    with pytest.raises(TypeError):
+        get_fn_args(1)
+    assert get_fn_args(sample_function) == "x, y=2"
+    assert get_fn_args(function_with_annotations) == "x: int, y: str"
+
+
+def test_get_fn_description():
+    with pytest.raises(TypeError):
+        get_fn_description(1)
+    assert get_fn_description(sample_function) == "This is a sample function."
+    assert get_fn_description(function_with_annotations) == "Function with annotations."
+    assert get_fn_description(function_with_no_docstring) == ""
+
+
+def test_get_fn_source():
+    with pytest.raises(TypeError, match="The `source` filter only applies to callables."):
+        get_fn_source(1)
+    source = (
+        'def sample_function(x, y=2):\n'
+        '    """This is a sample function."""\n'
+        '    return x + y'
+    )
+    assert get_fn_source(sample_function).strip() == source
+
+
+def test_get_fn_signature():
+    with pytest.raises(TypeError, match="The `source` filter only applies to callables."):
+        get_fn_signature(1)
+    sample_function_signature = "x, y=2"
+    assert get_fn_signature(sample_function) == sample_function_signature
+    function_with_annotations_signature = "x: int, y: str"
+    assert get_fn_signature(function_with_annotations) == function_with_annotations_signature
+
+
+def test_get_schema():
+    with pytest.raises(NotImplementedError):
+        get_schema(1)
+
+    dict_schema = {"foo": "bar"}
+    dict_schema_output = get_schema(dict_schema)
+    assert dict_schema_output == '{\n  "foo": "bar"\n}'
+
+    pydantic_schema_output = get_schema(PydanticClass)
+    assert pydantic_schema_output == '{\n  "foo": "<foo>"\n}'
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index e92755f3f..d5d53997b 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -1,6 +1,5 @@
 import datetime
 import json
-import os
 import sys
 import tempfile
 from dataclasses import dataclass
@@ -37,6 +36,7 @@
     Term,
     either,
     CFG,
+    FSM,
     _handle_dict,
     _handle_list,
     _handle_literal,
@@ -52,7 +52,6 @@
     exactly,
     regex,
     python_types_to_terms,
-    cfg,
 )
 
 if sys.version_info >= (3, 12):
@@ -65,26 +64,32 @@ def test_dsl_init():
     string = String("test")
     assert string.value == "test"
     assert repr(string) == "String(value='test')"
+    assert string.display_ascii_tree() == "└── String('test')\n"
 
     regex = Regex("[0-9]")
     assert regex.pattern == "[0-9]"
     assert repr(regex) == "Regex(pattern='[0-9]')"
+    assert regex.display_ascii_tree() == "└── Regex('[0-9]')\n"
 
     schema = JsonSchema('{ "type": "string" }')
     assert schema.schema == '{ "type": "string" }'
     assert repr(schema) == 'JsonSchema(schema=\'{ "type": "string" }\')'
+    assert schema.display_ascii_tree() == "└── JsonSchema('{ \"type\": \"string\" }')\n"
 
     kleene_star = KleeneStar(string)
     assert kleene_star.term == string
     assert repr(kleene_star) == "KleeneStar(term=String(value='test'))"
+    assert kleene_star.display_ascii_tree() == "└── KleeneStar(*)\n    └── String('test')\n"
 
     kleene_plus = KleenePlus(string)
     assert kleene_plus.term == string
     assert repr(kleene_plus) == "KleenePlus(term=String(value='test'))"
+    assert kleene_plus.display_ascii_tree() == "└── KleenePlus(+)\n    └── String('test')\n"
 
     optional = Optional(string)
     assert optional.term == string
     assert repr(optional) == "Optional(term=String(value='test'))"
+    assert optional.display_ascii_tree() == "└── Optional(?)\n    └── String('test')\n"
 
     alternatives = Alternatives([string, regex])
     assert alternatives.terms[0] == string
@@ -93,6 +98,7 @@ def test_dsl_init():
         repr(alternatives)
         == "Alternatives(terms=[String(value='test'), Regex(pattern='[0-9]')])"
     )
+    assert alternatives.display_ascii_tree() == "└── Alternatives(|)\n    ├── String('test')\n    └── Regex('[0-9]')\n"
 
     sequence = Sequence([string, regex])
     assert sequence.terms[0] == string
@@ -101,21 +107,25 @@ def test_dsl_init():
         repr(sequence)
         == "Sequence(terms=[String(value='test'), Regex(pattern='[0-9]')])"
     )
+    assert sequence.display_ascii_tree() == "└── Sequence\n    ├── String('test')\n    └── Regex('[0-9]')\n"
 
     exact = QuantifyExact(string, 3)
     assert exact.term == string
     assert exact.count == 3
     assert repr(exact) == "QuantifyExact(term=String(value='test'), count=3)"
+    assert exact.display_ascii_tree() == "└── Quantify({3})\n    └── String('test')\n"
 
     minimum = QuantifyMinimum(string, 3)
     assert minimum.term == string
     assert minimum.min_count == 3
     assert repr(minimum) == "QuantifyMinimum(term=String(value='test'), min_count=3)"
+    assert minimum.display_ascii_tree() == "└── Quantify({3,})\n    └── String('test')\n"
 
     maximum = QuantifyMaximum(string, 3)
     assert maximum.term == string
     assert maximum.max_count == 3
     assert repr(maximum) == "QuantifyMaximum(term=String(value='test'), max_count=3)"
+    assert maximum.display_ascii_tree() == "└── Quantify({,3})\n    └── String('test')\n"
 
     between = QuantifyBetween(string, 1, 3)
     assert between.term == string
@@ -125,6 +135,7 @@ def test_dsl_init():
         repr(between)
         == "QuantifyBetween(term=String(value='test'), min_count=1, max_count=3)"
     )
+    assert between.display_ascii_tree() == "└── Quantify({1,3})\n    └── String('test')\n"
 
     with pytest.raises(
         ValueError, match="`max_count` must be greater than `min_count`"
@@ -132,6 +143,44 @@ def test_dsl_init():
         QuantifyBetween(string, 3, 1)
 
 
+def test_dsl_term_methods():
+    a = String("a")
+    b = Regex("[0-9]")
+    c = "c"
+
+    assert a + b == Sequence([a, b])
+    assert a + c == Sequence([a, String(c)])
+    assert a.__radd__(b) == Sequence([b, a])
+    assert a.__radd__(c) == Sequence([String(c), a])
+
+    assert a | b == Alternatives([a, b])
+    assert a | c == Alternatives([a, String(c)])
+    assert a.__ror__(b) == Alternatives([b, a])
+    assert a.__ror__(c) == Alternatives([String(c), a])
+
+    core_schema = a.__get_pydantic_core_schema__("", "")
+    validator = a.__get_validator__(core_schema)
+    assert validator("a") == "a"
+    with pytest.raises(
+        ValueError,
+        match="Input should be in the language of the regular expression",
+    ):
+        validator("b")
+
+    assert a.__get_pydantic_json_schema__("", "") == {"type": "string", "pattern": "a"}
+
+    assert a.matches("a")
+    assert not a.matches("b")
+
+    assert a.display_ascii_tree() == "└── String('a')\n"
+
+    with pytest.raises(NotImplementedError):
+        Term()._display_node()
+
+    assert a._display_children("") == ""
+
+    assert a.__str__() == "└── String('a')\n"
+
 def test_dsl_sequence():
     a = String("a")
     b = String("b")
@@ -177,7 +226,7 @@ def test_dsl_alternatives():
 def test_dsl_optional():
     a = String("a")
 
-    opt = optional(a)
+    opt = a.optional()
     assert isinstance(opt, Optional)
 
     opt = optional("a")
@@ -191,7 +240,7 @@ def test_dsl_optional():
 def test_dsl_exactly():
     a = String("a")
 
-    rep = exactly(2, a)
+    rep = a.exactly(2)
     assert isinstance(rep, QuantifyExact)
     assert rep.count == 2
 
@@ -206,7 +255,7 @@ def test_dsl_exactly():
 def test_dsl_at_least():
     a = String("a")
 
-    rep = at_least(2, a)
+    rep = a.at_least(2)
     assert isinstance(rep, QuantifyMinimum)
     assert rep.min_count == 2
 
@@ -221,7 +270,7 @@ def test_dsl_at_least():
 def test_dsl_at_most():
     a = String("a")
 
-    rep = at_most(2, a)
+    rep = a.at_most(2)
     assert isinstance(rep, QuantifyMaximum)
     assert rep.max_count == 2
 
@@ -236,7 +285,7 @@ def test_dsl_at_most():
 def test_between():
     a = String("a")
 
-    rep = between(1, 2, a)
+    rep = a.between(1, 2)
     assert isinstance(rep, QuantifyBetween)
     assert rep.min_count == 1
     assert rep.max_count == 2
@@ -252,7 +301,7 @@ def test_between():
 def test_dsl_zero_or_more():
     a = String("a")
 
-    rep = zero_or_more(a)
+    rep = a.zero_or_more()
     assert isinstance(rep, KleeneStar)
 
     rep = zero_or_more("a")
@@ -266,7 +315,7 @@ def test_dsl_zero_or_more():
 def test_dsl_one_or_more():
     a = String("a")
 
-    rep = one_or_more(a)
+    rep = a.one_or_more()
     assert isinstance(rep, KleenePlus)
 
     rep = one_or_more("a")
@@ -332,6 +381,76 @@ def test_dsl_display():
     )
 
 
+def test_cfg():
+    cfg_string = """
+?start: expr
+?expr: NUMBER
+"""
+    cfg = types.cfg(cfg_string)
+    assert isinstance(cfg, CFG)
+    assert cfg.definition.strip() == "?start: expr\n?expr: NUMBER"
+    assert cfg._display_node() == "CFG('\n?start: expr\n?expr: NUMBER\n')"
+    assert cfg.__repr__() == "CFG(definition='\n?start: expr\n?expr: NUMBER\n')"
+    assert cfg == types.cfg(cfg_string)
+    assert not cfg == "a"
+
+
+def test_fsm():
+    interegular_fsm = interegular.parse_pattern(r"a").to_fsm()
+    fsm = types.fsm(interegular_fsm)
+    assert isinstance(fsm, FSM)
+    assert fsm.fsm == interegular_fsm
+    assert fsm._display_node() == f"FSM({fsm.fsm.__repr__()})"
+    assert fsm.__repr__() == f"FSM(fsm={fsm.fsm.__repr__()})"
+
+
+def test_json_schema():
+    # init dict
+    schema = types.json_schema({"type": "string"})
+    assert schema.schema == '{"type": "string"}'
+
+    # init str
+    schema = types.json_schema('{"type": "string"}')
+    assert schema.schema == '{"type": "string"}'
+
+    # init Pydantic model
+    class PydanticModel(BaseModel):
+        foo: str
+    schema = types.json_schema(PydanticModel)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "PydanticModel", "type": "object"}'
+
+    # init TypedDict
+    class SomeTypedDict(TypedDict):
+        foo: str
+    schema = types.json_schema(SomeTypedDict)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "SomeTypedDict", "type": "object"}'
+
+    # init dataclass
+    @dataclass
+    class DataClass:
+        foo: str
+    schema = types.json_schema(DataClass)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "DataClass", "type": "object"}'
+
+    # init SchemaBuilder
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    schema = types.json_schema(builder)
+    assert schema.schema == '{"$schema": "http://json-schema.org/schema#", "type": "object"}'
+
+    # init unsupported type
+    with pytest.raises(ValueError, match="Cannot parse schema"):
+        types.json_schema(1)
+
+    # other methods
+    schema = types.json_schema('{"type": "string"}')
+    schema.__post_init__()
+    assert schema._display_node() == "JsonSchema('{\"type\": \"string\"}')"
+    assert schema.__repr__() == "JsonSchema(schema='{\"type\": \"string\"}')"
+    assert schema == types.json_schema('{"type": "string"}')
+    assert not schema == "a"
+
+
 def test_dsl_cfg_from_file():
     grammar_content = """
     ?start: expression
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
index f32aae924..4d73312d6 100644
--- a/tests/types/test_types_utils.py
+++ b/tests/types/test_types_utils.py
@@ -36,6 +36,7 @@
     is_genson_schema_builder,
     is_int,
     is_int_instance,
+    is_interegular_fsm,
     is_literal,
     is_native_dict,
     is_pydantic_model,
@@ -358,6 +359,12 @@ def test_is_callable(sample_function, sample_class, sample_dataclass, sample_typ
     assert not is_callable(sample_pydantic_model)
 
 
+def test_is_interegular_fsm():
+    fsm = interegular.parse_pattern(r"[0-9]+").to_fsm()
+    assert is_interegular_fsm(fsm)
+    assert not is_interegular_fsm(dict)
+
+
 # Type conversion
 
 
diff --git a/tests/v0_legacy/generate/test_api_legacy.py b/tests/v0_legacy/generate/test_api_legacy.py
index b139f20ea..75cf72ff0 100644
--- a/tests/v0_legacy/generate/test_api_legacy.py
+++ b/tests/v0_legacy/generate/test_api_legacy.py
@@ -1,6 +1,7 @@
 from io import BytesIO
-from urllib.request import urlopen
 from dataclasses import asdict
+from typing import Generator
+from urllib.request import urlopen
 
 import pytest
 import transformers
@@ -9,7 +10,7 @@
 
 from outlines import models, samplers
 from outlines.generator import SteerableGenerator
-from outlines.models import TransformersVision  # type: ignore
+from outlines.v0_legacy.models.transformers_vision import TransformersVision
 from outlines.processors import RegexLogitsProcessor
 from outlines.v0_legacy.generate.api import (
     GeneratorV0Adapter,
@@ -40,8 +41,9 @@ def transformers_vision_model():
         match="The `transformers_vision` function is deprecated",
     ):
         return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            transformers.AutoModelForCausalLM
+            "trl-internal-testing/tiny-LlavaForConditionalGeneration",
+            model_class=transformers.LlavaForConditionalGeneration,
+            processor_class=transformers.AutoProcessor,
         )
 
 
@@ -80,29 +82,29 @@ def test_generator_v0_adapter_init(transformers_model):
         match="You can only use the v0 API with models that were already"
         + "available in v0. Got <class 'outlines.models.anthropic.Anthropic'>.",
     ):
-        adapter = GeneratorV0Adapter(model, None, samplers.greedy())
+        GeneratorV0Adapter(model, None, samplers.greedy())
 
     # unknown output type
     with pytest.raises(
         TypeError,
         match="is currently not supported",
     ):
-        adapter = GeneratorV0Adapter(
+        GeneratorV0Adapter(
             transformers_model, type, samplers.greedy()
         )
 
     # valid initialization
-    adapter = GeneratorV0Adapter(
+    generator_v0_adapter = GeneratorV0Adapter(
         transformers_model,
         int,
         samplers.greedy(),
     )
-    assert isinstance(adapter, GeneratorV0Adapter)
-    assert adapter.model == transformers_model
-    assert adapter.sampling_params == asdict(samplers.greedy().sampling_params)
-    assert isinstance(adapter.generator, SteerableGenerator)
-    assert adapter.generator.model == transformers_model
-    assert isinstance(adapter.generator.logits_processor, RegexLogitsProcessor)
+    assert isinstance(generator_v0_adapter, GeneratorV0Adapter)
+    assert generator_v0_adapter.model == transformers_model
+    assert generator_v0_adapter.sampling_params == asdict(samplers.greedy().sampling_params)
+    assert isinstance(generator_v0_adapter.generator, SteerableGenerator)
+    assert generator_v0_adapter.generator.model == transformers_model
+    assert isinstance(generator_v0_adapter.generator.logits_processor, RegexLogitsProcessor)
 
 
 def test_generator_v0_adapter_create_inference_params(
@@ -119,21 +121,27 @@ def test_generator_v0_adapter_create_inference_params(
         "foo": "bar",
     }
 
-    generator = create_generator_v0_adapter(llama_cpp_model)
-    result = generator.create_inference_params(**args, **kwargs)
+    result = (
+        create_generator_v0_adapter(llama_cpp_model)
+        .create_inference_params(**args, **kwargs)
+    )
     assert isinstance(result, dict)
 
-    generator = create_generator_v0_adapter(transformers_vision_model)
-    result = generator.create_inference_params(**args, **kwargs)
+    result = (
+        create_generator_v0_adapter(transformers_vision_model)
+        .create_inference_params(**args, **kwargs)
+    )
     assert isinstance(result, dict)
 
-    generator = create_generator_v0_adapter(transformers_model)
-    result = generator.create_inference_params(**args, **kwargs)
+    result = (
+        create_generator_v0_adapter(transformers_model)
+        .create_inference_params(**args, **kwargs)
+    )
     assert isinstance(result, dict)
 
 
 def test_generator_vision_v0_adapter_merge_prompts_and_media(transformers_vision_model):
-    generator = create_generator_v0_adapter(transformers_vision_model)
+    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
     prompts = ["Hello, world!", "Goodbye, world!"]
     media = [
         img_from_url(
@@ -143,7 +151,36 @@ def test_generator_vision_v0_adapter_merge_prompts_and_media(transformers_vision
             "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
         )
     ]
-    result = generator.merge_prompts_and_media(prompts, media)
+    result = generator_v0_adapter.merge_prompts_and_media(prompts, media)
     assert isinstance(result, dict)
     assert result["text"] == prompts
     assert result["images"] == media
+
+
+def test_generator_v0_adapter_call_stream(llama_cpp_model, transformers_model):
+    generator_v0_adapter = create_generator_v0_adapter(transformers_model)
+    result = generator_v0_adapter("Hello, world!", max_tokens=10)
+    assert isinstance(result, str)
+    result = generator_v0_adapter(["Hello, world!", "Goodbye, world!"], max_tokens=10)
+    assert isinstance(result, list)
+    for r in result:
+        assert isinstance(r, str)
+
+    generator_v0_adapter = create_generator_v0_adapter(llama_cpp_model)
+    gen = generator_v0_adapter.stream("Hello, world!", max_tokens=10)
+    assert isinstance(gen, Generator)
+
+
+def test_generator_v0_adapter_vision_call_stream(transformers_vision_model):
+    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
+    media = img_from_url(
+        "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+    )
+    result = generator_v0_adapter("Hello, world!<image>", media, max_tokens=10)
+    assert isinstance(result, str)
+    with pytest.raises(
+        NotImplementedError,
+        match="Streaming is not implemented for Transformers models",
+    ):
+        gen = generator_v0_adapter.stream("Hello, world!<image>", media, max_tokens=10)
+        assert isinstance(gen, Generator)
diff --git a/tests/v0_legacy/generate/test_choice_legacy.py b/tests/v0_legacy/generate/test_choice_legacy.py
index fafe8bbce..fbc4650e2 100644
--- a/tests/v0_legacy/generate/test_choice_legacy.py
+++ b/tests/v0_legacy/generate/test_choice_legacy.py
@@ -86,5 +86,12 @@ def test_choice_legacy_openai(openai_model):
         DeprecationWarning,
         match="The `choice` function is deprecated",
     ):
+        with pytest.raises(
+            ValueError,
+            match="The `choice` function with OpenAI only supports a list of "
+            + "strings as choices.",
+        ):
+            choice(openai_model, EnumChoice)
+
         generator = choice(openai_model, ["foo", "bar"])
         assert isinstance(generator, Callable)
diff --git a/tests/v0_legacy/generate/test_json_legacy.py b/tests/v0_legacy/generate/test_json_legacy.py
index fa4ac28ff..d09d0347b 100644
--- a/tests/v0_legacy/generate/test_json_legacy.py
+++ b/tests/v0_legacy/generate/test_json_legacy.py
@@ -189,3 +189,12 @@ def test_json_legacy_openai_json_string(openai_model):
         generator.format_sequence('{"key": "value"}')
         == pyjson.loads('{"key": "value"}')
     )
+
+
+def test_json_legacy_openai_not_supported(openai_model):
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `json` function is deprecated",
+    ):
+        with pytest.raises(ValueError, match="Cannot parse schema"):
+            json(openai_model, int)
diff --git a/tests/v0_legacy/models/test_exllamav2_legacy.py b/tests/v0_legacy/models/test_exllamav2_legacy.py
index 34afc7e9c..06fa9e17e 100644
--- a/tests/v0_legacy/models/test_exllamav2_legacy.py
+++ b/tests/v0_legacy/models/test_exllamav2_legacy.py
@@ -3,7 +3,7 @@
 import pytest
 
 try:
-    import exllamav2  # noqa: F401
+    import exllamav2  # type: ignore
     HAS_EXV2 = True
 except ImportError:
     HAS_EXV2 = False
@@ -12,7 +12,6 @@
     )
 
 from outlines import models, samplers, generate
-from outlines.v0_legacy.models.exllamav2 import exllamav2_params_adapter
 from outlines.v0_legacy.generate.api import GeneratorV0Adapter
 
 pytestmark = pytest.mark.skipif(
@@ -59,10 +58,19 @@ def test_exllamav2_legacy_call(model):
     )
     assert isinstance(generator, GeneratorV0Adapter)
 
-    result = generator(["Write a short story.", "Write a poem."], 10, "e", 2)
+    result = generator(["Write a short story.", "Write a poem."], 2, "e", 2)
     assert isinstance(result, list)
     for sequence in result:
         assert isinstance(sequence, list)
         for s in sequence:
             assert isinstance(s, str)
             assert len(s) < 20
+
+    # stream
+    generator = generate.text(model)
+    assert isinstance(generator, GeneratorV0Adapter)
+
+    result = generator.stream("Respond with a single word.", 2)
+    assert isinstance(result, GeneratorV0Adapter)
+    for r in result:
+        assert isinstance(r, str)
diff --git a/tests/v0_legacy/models/test_llamacpp_legacy.py b/tests/v0_legacy/models/test_llamacpp_legacy.py
index 318df4bef..aead5bd86 100644
--- a/tests/v0_legacy/models/test_llamacpp_legacy.py
+++ b/tests/v0_legacy/models/test_llamacpp_legacy.py
@@ -3,7 +3,7 @@
 
 import pytest
 from llama_cpp import Llama
-from llama_cpp.llama_tokenizer import LlamaHFTokenizer, LlamaTokenizer
+from llama_cpp.llama_tokenizer import LlamaHFTokenizer, LlamaTokenizer # type: ignore
 
 from outlines import models, samplers, generate
 from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
@@ -36,10 +36,12 @@ def test_llamacpp_legacy_init():
             repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
             filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
             tokenizer=LlamaHFTokenizer.from_pretrained("distilbert-base-uncased"),
+            n_ctx=0,
+            verbose=True,
         )
     assert isinstance(model, models.LlamaCpp)
     assert isinstance(model.model, Llama)
-    assert not model.model.verbose
+    assert model.model.verbose
     assert isinstance(model.model.tokenizer(), LlamaTokenizer)
 
 
@@ -119,7 +121,7 @@ def test_llamacpp_legacy_params_adapter_model_specific_params():
     )
     assert result.get("min_p") == 0.5
     assert result.get("repeat_penalty") == 1.5
-    assert not result.get("stream")
+    assert not hasattr(result, "stream")
 
 
 def test_llamacpp_legacy_call_generation():
diff --git a/tests/v0_legacy/models/test_transformers_legacy.py b/tests/v0_legacy/models/test_transformers_legacy.py
index 91b25a571..2c34c2517 100644
--- a/tests/v0_legacy/models/test_transformers_legacy.py
+++ b/tests/v0_legacy/models/test_transformers_legacy.py
@@ -33,7 +33,7 @@ def test_transformers_legacy_init():
         match="The `transformers` function is deprecated",
     ):
         model = models.transformers(
-            model_name="prajjwal1/bert-tiny",
+            model_name="bert-base-uncased",
             model_class=BertForSequenceClassification,
             tokenizer_class=BertTokenizer,
         )
diff --git a/tests/v0_legacy/test_function_legacy.py b/tests/v0_legacy/test_function_legacy.py
index 17a4e70ae..24799a880 100644
--- a/tests/v0_legacy/test_function_legacy.py
+++ b/tests/v0_legacy/test_function_legacy.py
@@ -33,6 +33,47 @@ class Foo(BaseModel):
 
             result = fn("test")
             assert isinstance(result, BaseModel)
+            assert fn.generator is not None
+            fn("test")
+
+
+@responses.activate
+def test_function_from_github():
+    content = """
+import outlines
+from pydantic import BaseModel
+
+model = "gpt2"
+
+
+prompt = outlines.Template.from_string("{{ text }}")
+
+
+class User(BaseModel):
+    id: int
+    name: str
+
+
+fn = outlines.Function(
+    prompt,
+    User,
+    "gpt2",
+)
+"""
+    responses.add(
+        responses.GET,
+        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
+        body=content,
+        status=200,
+    )
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `Function` class is deprecated",
+    ):
+        fn = Function.from_github("dottxt-ai/outlines/program", "fn")
+        assert (
+            str(type(fn)) == "<class 'outlines.v0_legacy.function.Function'>"
+        )
 
 
 def test_download_from_github_invalid():

From 58dbe5d687b0c2e506770bcf3f3477e46ceed33b Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Apr 2025 09:54:08 +0200
Subject: [PATCH 308/505] Update the exmaples scripts to use the v1 interface

---
 examples/babyagi.py                       | 11 ++++--
 examples/beam-cloud/app.py                | 11 +++++-
 examples/bentoml/service.py               |  2 +-
 examples/cerebrium/main.py                |  4 +-
 examples/cfg.py                           |  9 +++--
 examples/dating_profile.py                | 15 +------
 examples/llamacpp_processor.py            |  4 +-
 examples/math_generate_code.py            |  2 +-
 examples/meta_prompting.py                | 24 ++++++------
 examples/modal_example.py                 |  2 +-
 examples/parsing.py                       |  1 -
 examples/pick_odd_one_out.py              | 40 ++++++++++---------
 examples/react.py                         | 48 +++++++++++++++++------
 examples/sampling.ipynb                   | 48 +++++++++++------------
 examples/self_consistency.py              | 17 ++------
 examples/simulation_based_inference.ipynb | 18 ++++-----
 examples/transformers_integration.py      | 25 ------------
 examples/vllm_integration.py              |  5 ++-
 18 files changed, 143 insertions(+), 143 deletions(-)
 delete mode 100644 examples/transformers_integration.py

diff --git a/examples/babyagi.py b/examples/babyagi.py
index 3644ecf51..4a817906d 100644
--- a/examples/babyagi.py
+++ b/examples/babyagi.py
@@ -77,11 +77,14 @@ def one_cycle(objective: str, task_list, next_task_id: int):
 
     task = task_list.popleft()
 
-    prompt = perform_task_ppt(objective, task)
+    prompt = perform_task_ppt(objective=objective, task=task)
     result = complete(prompt)
 
     prompt = create_tasks_ppt(
-        objective, first_task["task_name"], result, [first_task["task_name"]]
+        objective=objective,
+        task=first_task["task_name"],
+        result=result,
+        previous_tasks=[first_task["task_name"]],
     )
     new_tasks = complete(prompt)
 
@@ -92,7 +95,9 @@ def one_cycle(objective: str, task_list, next_task_id: int):
         task_list.append({"task_id": next_task_id, "task_name": task})
 
     prompt = prioritize_tasks_ppt(
-        objective, [task["task_name"] for task in task_list], next_task_id
+        objective=objective,
+        tasks=[task["task_name"] for task in task_list],
+        next_task_id=next_task_id,
     )
     prioritized_tasks = complete(prompt)
 
diff --git a/examples/beam-cloud/app.py b/examples/beam-cloud/app.py
index fb6c2cb2b..781bc1d3d 100644
--- a/examples/beam-cloud/app.py
+++ b/examples/beam-cloud/app.py
@@ -1,14 +1,21 @@
+from typing import Literal
+
 from beam import Image, endpoint, env
 
+
 if env.is_remote():
     import outlines
 
 
 # Pre-load models when the container first starts
 def load_models():
+    from transformers import AutoModelForCausalLM, AutoTokenizer
     import outlines
 
-    model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+    model = outlines.models.from_transformers(
+        AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+        AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    )
     return model
 
 
@@ -34,6 +41,6 @@ def predict(context, **inputs):
     # Unpack cached model from context
     model = context.on_start_value
     # Inference
-    generator = outlines.generate.choice(model, ["Positive", "Negative"])
+    generator = outlines.Generator(model, Literal["Positive", "Negative"])
     answer = generator(prompt)
     return {"answer": answer}
diff --git a/examples/bentoml/service.py b/examples/bentoml/service.py
index 884904a73..370bac901 100644
--- a/examples/bentoml/service.py
+++ b/examples/bentoml/service.py
@@ -76,7 +76,7 @@ async def generate(
     ) -> t.Dict[str, t.Any]:
         import outlines
 
-        generator = outlines.generate.json(self.model, json_schema)
+        generator = outlines.Generator(self.model, outlines.json_schema(json_schema))
         character = generator(prompt)
 
         return character
diff --git a/examples/cerebrium/main.py b/examples/cerebrium/main.py
index bce3f441c..4e1987197 100644
--- a/examples/cerebrium/main.py
+++ b/examples/cerebrium/main.py
@@ -1,6 +1,8 @@
-import outlines
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
+import outlines
+
+
 model = outlines.from_transformers(
     AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
     AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2"),
diff --git a/examples/cfg.py b/examples/cfg.py
index 22fdf7ed0..bd3b58a2e 100644
--- a/examples/cfg.py
+++ b/examples/cfg.py
@@ -1,7 +1,8 @@
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
 import outlines
-from outlines.types import Cfg
+from outlines.types import CFG
+
 
 nlamb_grammar = r"""
     start: sentence
@@ -85,12 +86,12 @@
 
 batch_size = 10
 for grammar in [nlamb_grammar, calc_grammar, dyck_grammar, json_grammar]:
-    generator = outlines.Generator(model, Cfg(grammar))
+    generator = outlines.Generator(model, CFG(grammar))
     sequences = generator([" "] * batch_size, max_tokens=model.model.config.n_positions)
     for seq in sequences:
         try:
-            parse = generator.fsm.parser.parse(seq)
+            parse = generator.logits_processor.guide.parser.parse(seq)
             assert parse is not None
             print("SUCCESS", seq)
-        except Exception:  # will also fail if goes over max_tokens / context window
+        except Exception:  # will also fail if goes over max_new_tokens / context window
             print("FAILURE", seq)
diff --git a/examples/dating_profile.py b/examples/dating_profile.py
index 5576ca50a..ff488698e 100644
--- a/examples/dating_profile.py
+++ b/examples/dating_profile.py
@@ -87,19 +87,8 @@ class Example:
 # https://huggingface.co/mosaicml/mpt-7b-8k-instruct
 # Motivation: Reasonably large model that fits on a single GPU and has been fine-tuned for a larger context window
 model_name = "mosaicml/mpt-7b-8k-instruct"
-config = transformers.AutoConfig.from_pretrained(
-    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
-)
-config.init_device = "meta"
 model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(
-        model_name,
-        device="cuda",
-        config=config,
-        trust_remote_code=True,
-        torch_dtype=torch.bfloat16,
-        device_map={"": 0},
-    ),
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
     transformers.AutoTokenizer.from_pretrained(model_name),
 )
 
@@ -107,7 +96,7 @@ class Example:
 
 dating_profile_prompt = Template.from_file("prompts/dating_profile.txt")
 prompt = dating_profile_prompt(description=new_description, examples=samples)
-profile = model(prompt, outlines.json_schema(DatingProfile))  # type: ignore
+profile = model(prompt, outlines.json_schema(DatingProfile), max_tokens=500)  # type: ignore
 print(profile)
 
 # Sample generated profiles
diff --git a/examples/llamacpp_processor.py b/examples/llamacpp_processor.py
index 909df38fc..a0bf4aa3f 100644
--- a/examples/llamacpp_processor.py
+++ b/examples/llamacpp_processor.py
@@ -3,7 +3,7 @@
 from llama_cpp import Llama, LogitsProcessorList
 from pydantic import BaseModel, constr
 
-from outlines.generate.processors import JSONLogitsProcessor
+from outlines.processors import JSONLogitsProcessor
 from outlines.models.llamacpp import LlamaCppTokenizer
 
 
@@ -36,7 +36,7 @@ class Character(BaseModel):
 
     prompt = "Instruct: You are a leading role play gamer. You have seen thousands of different characters and their attributes.\nPlease return a JSON object with common attributes of an RPG character. Give me a character description\nOutput:"
 
-    logits_processor = JSONLogitsProcessor(Character, tokenizer)
+    logits_processor = JSONLogitsProcessor(Character, tokenizer, tensor_library_name="numpy")
 
     json_str = llama.create_completion(
         prompt,
diff --git a/examples/math_generate_code.py b/examples/math_generate_code.py
index 9dd21e95d..a5a53f084 100644
--- a/examples/math_generate_code.py
+++ b/examples/math_generate_code.py
@@ -37,7 +37,7 @@ def execute_code(code):
     return result
 
 
-prompt = answer_with_code_prompt(question, examples)
+prompt = answer_with_code_prompt(question=question, examples=examples)
 model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
 answer = model(prompt)
 result = execute_code(answer)
diff --git a/examples/meta_prompting.py b/examples/meta_prompting.py
index 510b4befe..ed4b7455a 100644
--- a/examples/meta_prompting.py
+++ b/examples/meta_prompting.py
@@ -30,13 +30,13 @@ def split_into_steps(question, model_name: str):
 
     model = outlines.from_openai(client, model_name)
 
-    prompt = solve(question)
-    answer = model(prompt, 500)
+    prompt = solve(question=question)
+    answer = model(prompt, max_tokens=500)
     prompt += (
         answer
         + "\n what is the only option that displays the same type of relationship as : :?"
     )
-    answer = model(prompt, 500)
+    answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
     return completed
@@ -54,9 +54,9 @@ def fill_in_the_blanks(question, model_name: str):
 
     model = outlines.from_openai(client, model_name)
 
-    prompt = determine_goal(question)
-    answer = model(prompt, stop_at=["."])
-    prompt = solve(prompt + answer)
+    prompt = determine_goal(question=question)
+    answer = model(prompt, stop=["."])
+    prompt = solve(memory=prompt + answer)
     answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
@@ -93,9 +93,9 @@ def ask_an_expert(question, model_name: str):
 
     model = outlines.from_openai(client, model_name)
 
-    prompt = find_expert(question)
-    expert = model(prompt, stop_at=['"'])
-    prompt = get_answer(question, expert, prompt + expert)
+    prompt = find_expert(question=question)
+    expert = model(prompt, stop=['"'])
+    prompt = get_answer(question=question, expert=expert, memory=prompt+expert)
     answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
@@ -120,9 +120,9 @@ def ask_an_expert_simple(question, model_name: str):
 
     model = outlines.from_openai(client, model_name)
 
-    prompt = find_expert(question)
-    expert = model(prompt, stop_at=["\n", "."])
-    prompt = get_answer(expert, prompt + expert)
+    prompt = find_expert(question=question)
+    expert = model(prompt, stop=["\n", "."])
+    prompt = get_answer(expert=expert, memory=prompt+expert)
     answer = model(prompt, max_tokens=500)
     completed = prompt + answer
 
diff --git a/examples/modal_example.py b/examples/modal_example.py
index 3ec6c2c6f..fe0e6bc67 100644
--- a/examples/modal_example.py
+++ b/examples/modal_example.py
@@ -60,7 +60,7 @@ def import_model():
 }"""
 
 
-@app.function(image=outlines_image, gpu=modal.gpu.A100(memory=80))
+@app.function(image=outlines_image, gpu="A100-40GB")
 def generate(
     prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
 ):
diff --git a/examples/parsing.py b/examples/parsing.py
index 9e08eae14..c8aac5a2e 100644
--- a/examples/parsing.py
+++ b/examples/parsing.py
@@ -36,7 +36,6 @@
     start="file_input",
 )
 
-
 class ParserLogitsProcessor(LogitsProcessor):
     """Bias invalid token scores according to a running parse state."""
 
diff --git a/examples/pick_odd_one_out.py b/examples/pick_odd_one_out.py
index 55127c5f7..36ec9dfe7 100644
--- a/examples/pick_odd_one_out.py
+++ b/examples/pick_odd_one_out.py
@@ -10,35 +10,37 @@
 
 """
 
+import json
+
 import openai
 
 import outlines
+from outlines import Generator
+from outlines.types import JsonSchema
 
 
-build_ooo_prompt = outlines.Template.from_string(
-    """
-    Pick the odd word out: skirt, dress, pen, jacket.
-    skirt is clothing, dress is clothing, pen is an object, jacket is clothing.
-    So the odd one is pen.
-
-    Pick the odd word out: Spain, France, German, England, Singapore.
-    Spain is a country, France is a country, German is a language, ...
-    So the odd one is German.
-
-    Pick the odd word out: {{ options | join(", ") }}.
-
-    """
-)
+build_ooo_prompt = outlines.Template.from_file("prompts/pick_odd_one_out.txt")
 
 options = ["sea", "mountains", "plains", "sock"]
+options_schema = JsonSchema({
+    "type": "object",
+    "properties": {
+        "result": {
+            "type": "string",
+            "enum": options
+        }
+    },
+    "required": ["result"]
+})
 
 model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
-gen_text = outlines.generate.text(model)
-gen_choice = outlines.generate.choice(model, options)
+gen_text = Generator(model)
+gen_choice = Generator(model, options_schema)
 
-prompt = build_ooo_prompt(options)
-reasoning = gen_text(prompt, stop_at=["Pick the odd word", "So the odd one"])
+prompt = build_ooo_prompt(options=options)
+reasoning = gen_text(prompt, stop=["Pick the odd word", "So the odd one"])
 prompt += reasoning
-result = gen_choice(prompt)
+raw_result = gen_choice(prompt)
+result = json.loads(raw_result)["result"]
 prompt += result
 print(result)
diff --git a/examples/react.py b/examples/react.py
index 472875cc1..b7c807726 100644
--- a/examples/react.py
+++ b/examples/react.py
@@ -11,12 +11,14 @@
 
 """
 
-import requests  # type: ignore
+import json
 
+import requests  # type: ignore
 from openai import OpenAI
 
 import outlines
-from outlines import Generator, Choice, Template
+from outlines import Generator, Template
+from outlines.types import JsonSchema
 
 
 build_reAct_prompt = Template.from_string(
@@ -47,32 +49,56 @@ def search_wikipedia(query: str):
     return ".".join(list(page.values())[0]["extract"].split(".")[:2])
 
 
-prompt = build_reAct_prompt("Where is Apple Computers headquarted? ")
+prompt = build_reAct_prompt(question="Where is Apple Computers headquarted? ")
 model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
 
-mode_generator = Generator(model, Choice(["Tho", "Act"]))
-action_generator = Generator(model, Choice(["Search", "Finish"]))
+# Define JSON schemas for mode and action
+mode_schema = JsonSchema({
+    "type": "object",
+    "properties": {
+        "result": {
+            "type": "string",
+            "enum": ["Tho", "Act"]
+        }
+    },
+    "required": ["result"]
+})
+action_schema = JsonSchema({
+    "type": "object",
+    "properties": {
+        "result": {
+            "type": "string",
+            "enum": ["Search", "Finish"]
+        }
+    },
+    "required": ["result"]
+})
+
+mode_generator = Generator(model, mode_schema)
+action_generator = Generator(model, action_schema)
 text_generator = Generator(model)
 
 for i in range(1, 10):
-    mode = mode_generator(prompt, max_tokens=128)
-    prompt = add_mode(i, mode, "", prompt)
+    mode_output = mode_generator(prompt, max_tokens=128)
+    mode = json.loads(mode_output)["result"]  # Extract the result from the JSON output
+    prompt = add_mode(i=i, mode=mode, result="", prompt=prompt)
 
     if mode == "Tho":
-        thought = text_generator(prompt, stop_at="\n", max_tokens=128)
+        thought = text_generator(prompt, stop="\n", max_tokens=128)
         prompt += f"{thought}"
     elif mode == "Act":
-        action = action_generator(prompt, max_tokens=128)
+        action_output = action_generator(prompt, max_tokens=128)
+        action = json.loads(action_output)["result"]  # Extract the result from the JSON output
         prompt += f"{action} '"
 
-        subject = text_generator(prompt, stop_at=["'"], max_tokens=128)
+        subject = text_generator(prompt, stop=["'"], max_tokens=128)
         # Apple Computers headquartered
         subject = " ".join(subject.split()[:2])
         prompt += f"{subject}'"
 
         if action == "Search":
             result = search_wikipedia(subject)
-            prompt = add_mode(i, "Obs", result, prompt)
+            prompt = add_mode(i=i, mode="Obs", result=result, prompt=prompt)
         else:
             break
 
diff --git a/examples/sampling.ipynb b/examples/sampling.ipynb
index d6b50e33d..75561451c 100644
--- a/examples/sampling.ipynb
+++ b/examples/sampling.ipynb
@@ -7,14 +7,15 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "#OK\n",
     "import functools as ft\n",
     "import re\n",
     "\n",
     "import numpy as np\n",
     "import matplotlib.pylab as plt\n",
+    "import openai\n",
     "\n",
-    "import outlines.models as models\n",
-    "import outlines.text as text"
+    "import outlines"
    ]
   },
   {
@@ -40,7 +41,7 @@
    "id": "2a3514d6-d5d7-46e9-9b69-1251d337e094",
    "metadata": {},
    "source": [
-    "In this example we will look at completion results for questions from the GSM8K dataset, using few-shots prompts with 5 examples. We first use `outlines.text.prompt` to build the few-shot prompt. `outlines.text.prompt` is a decorator around \"prompt functions\" which contain the prompt template in its docstring. Outlines uses the Jinja2 templating engine to render the prompt when the function is called with the variables' values; it thus allows you to build complex prompts very easily."
+    "In this example we will look at completion results for questions similar to those in the GSM8K dataset, using few-shots prompts with 5 examples. We first use `outlines.Template` to build the few-shot prompt. Outlines uses the Jinja2 templating engine to render the object when the function is called with the variables' values; it thus allows you to build complex prompts very easily."
    ]
   },
   {
@@ -86,8 +87,7 @@
     "]\n",
     "\n",
     "\n",
-    "@text.prompt\n",
-    "def few_shot_prompt(question, examples):\n",
+    "few_shot_prompt = outlines.Template.from_string(\n",
     "    \"\"\"\n",
     "    {% for example in examples %}\n",
     "    Q: {{ example.question }}\n",
@@ -96,9 +96,9 @@
     "    Q: {{ question }}\n",
     "    A:\n",
     "    \"\"\"\n",
+    ")\n",
     "\n",
-    "\n",
-    "# Prompt functions can be partially evaluated like any other function\n",
+    "# Template instances can be partially evaluated because they are callable objects\n",
     "gsm8k_prompt = ft.partial(few_shot_prompt, examples=examples)"
    ]
   },
@@ -107,7 +107,7 @@
    "id": "1eae0ec8-89f0-43fc-b055-6fcd64cbc03b",
    "metadata": {},
    "source": [
-    "## When `text-davinci-003` is uncertain"
+    "## When `gpt-4o-mini` is uncertain"
    ]
   },
   {
@@ -115,7 +115,7 @@
    "id": "a273ed78-e813-467e-85f3-16d7f283ba87",
    "metadata": {},
    "source": [
-    "Let us now sample 20 completions with the `text-davinci-003` model. Outlines is sampling first, and allows to draw several samples with both OpenAI and `transformers` models easily:"
+    "Let us now sample 20 completions with the `gpt-4o-mini` model. Outlines is sampling first, and allows to draw several samples with both OpenAI and `transformers` models easily:"
    ]
   },
   {
@@ -125,11 +125,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model = models.text_completion.openai(\"text-davinci-003\", max_tokens=128)\n",
+    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4o\")\n",
     "\n",
-    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
-    "prompt = gsm8k_prompt(question)\n",
-    "answers = model(prompt, samples=20)"
+    "question = \"When I was 6, my sister was half the age of my brother. When I was 14, my sister was 3 years younger than my brother. Now I'm 70, how old is my sister now?\"\n",
+    "prompt = gsm8k_prompt(question=question)\n",
+    "answers = model(prompt, n=20, max_tokens=512)"
    ]
   },
   {
@@ -201,7 +201,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqsAAAHgCAYAAACCbCTDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAAsTAAALEwEAmpwYAAAVwklEQVR4nO3df7DldX3f8dcbFtmVLb9cDDgLrB0DjNWUlWQzWBaDtEJg0BBiJk5VsKLFKRSsU7u2M8wKZouDjsg4Y0chBn9MTfgh3QkapQiCHSJWFigBAlMgggUkJlVJDXXh0z/2ILuwF9LZe+95372Px8yZvff7PeznvcB893m/53vOt8YYAQCAjnaZ9gAAADATsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtLXkRfb7XCsAAOZDbW+jM6sAALQlVgEAaEusAgvGddddl5NPPjn7779/dt9997ziFa/Icccdl69+9avTHg3YiTn2TNeLXbMK0MIHP/jBXHjhhVm5cmXe/OY3Z8WKFXn88cfzve99LzfccENOOOGEaY8I7IQce6avxnjB91B5gxUwdZ/97Gfz3ve+N6eeemo+85nP5CUveck2+3/+859nt912m9J0wM7KsWfebfcNVmIVaO3JJ5/MgQcemGXLluW+++573l8WAHPBsWcqthurLgMAWrv22mvz+OOP55xzzskuu+ySa665JnfeeWeWLl2aNWvW5Mgjj5z2iMBOyLGnD7EKtPbd7343SbJ06dKsXr06d9555zb7jz766FxxxRXZb7/9pjEesJNy7OnDpwEArf3whz9Mklx44YWpqtx000356U9/mjvuuCNvetObcuONN+atb33rlKcEdjaOPX2IVaC1p59+OkmyZMmSbNy4MUcddVSWL1+e1772tfnKV76SlStX5lvf+lZuvvnmKU8K7Ewce/oQq0Bre++9d5Jk9erVWbVq1Tb7XvrSl+a4445Lktxyyy3zPBmwM3Ps6UOsAq0deuihSZ79i+O59tlnnyTJz372s/kaCVgEHHv6EKtAa8cee2yqKnfdddcvXpbb2jNvenjlK18536MBOzHHnj7EKtDawQcfnJNOOinf//7388lPfnKbfd/4xjfy9a9/PXvvvXeOP/74KU0I7Iwce/pwUwCgvYcffjivf/3r89BDD+XYY4/N6tWr88ADD+Tqq69OVeXLX/5yTjnllGmPCexkHHvmnTtYAQvX448/nvPOOy8bN27MI488kj333DNr167Nhz70oaxZs2ba4wE7KceeeSVWAQBoa7ux6ppVAADaEqsAALQlVgEAaGvJtAcAeDGr1l0z474HLzhxHicBFhPHnh6cWQUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCAAvaF7/4xVRVqiqXXHLJtMdhlolVAGDBeuihh3LmmWdm+fLl0x6FOSJWAYAFaYyRd73rXXnZy16WM844Y9rjMEfEKgCwIF188cX55je/mc997nPZY489pj0Oc0SsAgALzt13351169bl7LPPztFHHz3tcZhDYhUAWFA2b96cd7zjHTnooIOyYcOGaY/DHFsy7QEAAP5/nHfeedm0aVO+/e1vZ9myZdMehznmzCoAsGB85zvfyYYNG/KBD3wgRx555LTHYR6IVQBgQdi8eXPe+c535pBDDsn5558/7XGYJ2IVAFgQnnjiidx77725++67s3Tp0l/cCKCq8uEPfzhJ8p73vCdVlXPOOWe6wzJrXLMKACwIu+++e9797ndvd9+tt96aTZs25aijjsqhhx7qEoGdiFgFABaEZcuWzXg71fXr12fTpk059dRTc/rpp8/zZMwllwEAANCWWAUAoC2xCgAseOvXr88YwyUAOyGxCgBAW2IVAIC2xCoAAG356CoAYMFYte6aF9z/4AUnztMkzBdnVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAAIvEj370o1xyySU5+eST86pXvSrLli3LXnvtlaOOOiqXXnppnn766WmP+DxLpj0AAADz4/LLL8/73ve+HHDAATnmmGNy0EEH5bHHHstVV12V008/PV/72tdy+eWXp6qmPeoviFUAgEXikEMOycaNG3PiiSdml12efYF9w4YNWbNmTa688spcddVVOeWUU6Y45bZcBgAAsEi88Y1vzEknnbRNqCbJ/vvvnzPOOCNJcsMNN0xhspmJVQAAsttuuyVJlizp9cK7WAUAWOQ2b96cz3/+80mS448/fsrTbEusAgAscuvWrcudd96ZE044Iccdd9y0x9mGWAUAWMQuvvjifPzjH89hhx2WL3zhC9Me53nEKgDAIvWpT30qZ599dl796lfn+uuvz7777jvtkZ5HrAIALEIXXXRRzjrrrLzmNa/J9ddfn/3333/aI22XWAUAWGQ++tGP5v3vf38OP/zwXH/99Xn5y18+7ZFmJFYBABaR888/P+vWrcsRRxyR6667LitWrJj2SC+o1wdpAQAwZy677LKce+652XXXXbN27dpcfPHFz3vOqlWrctppp83/cDMQqwAAi8QDDzyQJHnqqady0UUXbfc5b3jDG1rFqssAAAAWifXr12eM8YIPt1sFAIC/J7EKAEBbYhUAgLa8wQoAYBFZte6aGfc9eMGJ8zjJ348zqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoq2WsXnHFFTnrrLOydu3a7LnnnqmqvP3tb5/2WAAAc0L7zGzJtAfYno985CO5/fbbs3z58qxcuTL33HPPtEcCAJgz2mdmLc+sfuITn8i9996bn/zkJ/n0pz897XEAAOaU9plZyzOrxxxzzLRHAACYN9pnZi3PrAIAQCJWAQBoTKwCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2mp5U4Crr746V199dZLk0UcfTZLcfPPNOe2005IkK1asyMc+9rEpTQcAMLu0z8xaxuptt92Wyy67bJtt999/f+6///4kycEHH7xo/4MBADsf7TOzlpcBrF+/PmOMGR8PPvjgtEcEAJg12mdmLWMVAAASsQoAQGNiFQCAtlq+wSpJVq27ZsZ9D15w4jxOAgAwt3TPzJxZBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2hKrAAC0JVYBAGhLrAIA0JZYBQCgLbEKAEBbYhUAgLbEKgAAbYlVAADaEqsAALQlVgEAaEusAgDQllgFAKAtsQoAQFtiFQCAtsQqAABtiVUAANoSqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG3VGGPmnVV/mmTF/I0zoxVJ/mraQwBtOCYAz5jP44G15tZfjTGOf+7GF4zVLqrqv48xfnXacwA9OCYAz5jP44G1psNlAAAAtCVWAQBoa6HE6memPQDQimMC8Iz5PB5YawoWxDWrAAAsTgvlzCoAAItQ+1itqr2r6oqquqeq7q6qI6c9EzA/qmppVd1SVbdX1Z9X1Ycn2/+wqh6oqtsmj8OnPCowD16oCarqA1U1qmpWPnJze2tV1R9tddx5sKpum4V1Dt3q97ytqn5SVedU1b5VdW1V3Tf5dZ85XOv8qrpjsu0bVfWKHV1rNrW/DKCqLkty0xjjkqp6SZKXjjH+95THAuZBVVWSPcYYT1TVbkm+neTsJGck+ZMxxhVTHRCYVzM1QVUdmOSSJIclOWKMscOfGfpi/VFVH0/y4zHGeTu61la/565JfpDk15P8qyR/Pca4oKrWJdlnjPHv5mitvxlj/GSy/V8nefUY44zZWmtHtT6zWlV7JTk6yaVJMsb4v0IVFo+xxROTb3ebPHr/hA3MiRdpgk8k+WBm6fjwYv0x+UH6d5P859lYbyvHJvmfY4y/TPKWJJdNtl+W5Lfmaq1nQnVijzQ7zraO1SSvTPJ4ks9V1aaquqSq9pj2UMD8qapdJy+1/TDJtWOM70x2/f7kZatPVNXu05sQmCfbbYKqekuSH4wxbp/rtbbavzbJY2OM+2ZxzST5vTwbwL80xnhk8vWjSX5pDtdKVf1+VT2U5J8nOXeW19oh3WN1SZLXJfn0GGN1kr9Nsm66IwHzaYzx1Bjj8CQrk6ypqtck+VC2vNz3a0n2TTJrL40BbW2vCdYn+feZ/bh6sf54W2b5rOrkUoM3J7n8ufvGlms2Z+1s5/bWGmP8hzHGgUm+lOTM2VprNnSP1YeTPLzVmZQrsuV/HmCRmbwEd32S48cYj0wuEXgyyeeSrJnqcMB8mKkJXpnk9qp6MFt+qL21qvafo7VSVUuS/HaSP9rBNZ7rN5PcOsZ4bPL9Y1V1wGTNA7Ll1aW5WmtrX0pyyiyutcNax+oY49EkD1XVoZNNxya5a4ojAfOoqvarqr0nXy9L8s+S3LPVAbyy5TquO6c1IzA/ZmiCW8cYLx9jrBpjrMqWyHzd5LmzvdYz/fFPk9wzxnh4R9bYjueerd2Y5NTJ16cm+S9ztVZV/fJW+96S5J5ZXGuHLYRPAzg8W97h95Ik9yd51xjjb6Y6FDAvqupXsuWNBbtmyw/XfzzGOK+qvplkvySV5LYkZ2z1RixgJ/ViTTA5u/qrs/RpANtdq6r+MMmfjTH+046usdVaeyT5fpJ/OMb48WTby5L8cZKDkvxlkt8dY/z1HK11ZZJDkzw9WeuMMcYPdnSt2dI+VgEAWLxaXwYAAMDiJlYBAGhLrAIA0JZYBQCgLbEKAEBbYhXgOarqt6pqVNVh054FYLETqwDP97Yk3578OhWTu+QALHpiFWArVbU8yVFJ3p3k9ybbfqOqbqiqK6rqnqr60uTuWamqC6rqrqq6o6o+VlW7VtUDtcXeVfVUVR09ee6NVfXLVbVHVf1BVd1SVZuq6i2T/adV1cbJTQ+um86/AYBe/OQOsK23JPnTMca9VfWjqjpisn11kn+U5H8l+W9J/klV3Z3k5CSHjTFGVe09xniqqv4iyauz5Z7ltyZZW1XfSXLgGOO+qtqQ5JtjjH8xuZ3sLVX1XyfrvC7Jr8zGnWoAdgbOrAJs621Jvjz5+st59lKAW8YYD48xns6WW7yuSvLjJH+X5NKq+u0k/2fy3JuSHD15/MdsOVP7a0m+O9n/piTrquq2JDckWZott1RMkmuFKsCznFkFmKiqfZO8Mclrq2ok2TXJSHJNkie3eupTSZaMMTZX1Zokxyb5nSRnTv75G5O8L8krkpyb5N8m+Y1sidgkqSSnjDH+4jnr/3qSv52TPxzAAuXMKsCzfifJF8YYB48xVo0xDkzyQJK123vy5PrWvcYYX03y/iT/eLLrliSvT/L0GOPvsuVM7L/MlohNkq8nOWur615Xz9GfB2DBE6sAz3pbkq88Z9uVmflTAf5Bkj+pqjuy5dMD/k2SjDGeTPJQkj+bPO+myXP/x+T785PsluSOqvrzyfcAbEeNMaY9AwAAbJczqwAAtCVWAQBoS6wCANCWWAUAoC2xCgBAW2IVAIC2xCoAAG2JVQAA2vp/jdj4sUZoV2sAAAAASUVORK5CYII=",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
@@ -222,7 +222,7 @@
    "id": "661a1135-ac2d-4a49-a786-d04a7ba68b48",
    "metadata": {},
    "source": [
-    "We see that there is an important variabilty in the answers given by `text-davinci-003`. Depending on the number of samples taken, even self-consistency sampling may lead to the wrong result here."
+    "We see that there is an important variabilty in the answers given by `gpt-4o-mini`. Depending on the number of samples taken, even self-consistency sampling may lead to the wrong result here."
    ]
   },
   {
@@ -248,7 +248,7 @@
    "id": "0b15b230-b667-4c9c-8a5d-366dd61de9b7",
    "metadata": {},
    "source": [
-    "## `text-davinci-003` on an easier question"
+    "## `gpt-4o-mini` on an easier question"
    ]
   },
   {
@@ -266,9 +266,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "question = \"John buys 2 pairs of shoes for each of his 3 children. They cost $60 each. How much did he pay?\"\n",
+    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
     "prompt = gsm8k_prompt(question)\n",
-    "answers = model(question, samples=20)"
+    "answers = model(question, samples=20, max_tokens=512)"
    ]
   },
   {
@@ -286,7 +286,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
@@ -316,7 +316,7 @@
    "id": "22f31872-aab7-4a68-b9f2-d335a4f1a875",
    "metadata": {},
    "source": [
-    "## How `gpt-4` compares to `text-davinci-003`\n",
+    "## How `gpt-4` compares to `gpt-4o-mini`\n",
     "\n",
     "Let us now look at how GPT4 fares on the original question:"
    ]
@@ -328,11 +328,11 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "model = models.text_completion.openai(\"gpt-4\", max_tokens=128)\n",
+    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4\")\n",
     "\n",
-    "question = \"When I was 6 my sister was half my age. Now I’m 70 how old is my sister?\"\n",
+    "question = \"When I was 6, my sister was half the age of my brother. When I was 14, my sister was 3 years younger than my brother. Now I'm 70, how old is my sister now?\"\n",
     "prompt = gsm8k_prompt(question)\n",
-    "answers = model(prompt, samples=20)"
+    "answers = model(prompt, samples=20, max_tokens=512)"
    ]
   },
   {
@@ -350,7 +350,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
diff --git a/examples/self_consistency.py b/examples/self_consistency.py
index 1b578a4d3..061ff594c 100644
--- a/examples/self_consistency.py
+++ b/examples/self_consistency.py
@@ -44,21 +44,12 @@
 question = "When I was 6 my sister was half my age. Now I’m 70 how old is my sister?"
 
 
-few_shots = Template.from_string(
-    """
-    {% for example in examples %}
-    Q: {{ example.question }}
-    A: {{ example.answer }}
-    {% endfor %}
-    Q: {{ question }}
-    A:
-    """
-)
+few_shots = Template.from_file("prompts/self_consistency.txt")
 
 model = outlines.from_openai(openai.OpenAI(), "gpt-4o-mini")
 generator = outlines.Generator(model)
-prompt = few_shots(question, examples)
-answers = generator(prompt, samples=10)
+prompt = few_shots(question=question, examples=examples)
+answers = generator(prompt, n=10)
 
 digits = []
 for answer in answers:
@@ -71,7 +62,7 @@
         print(f"Could not parse the completion: '{answer}'")
 
 unique_digits, counts = np.unique(digits, return_counts=True)
-results = {d: c for d, c in zip(unique_digits, counts)}
+results = {int(d): int(c) for d, c in zip(unique_digits, counts)}
 print(results)
 
 max_count = max(results.values())
diff --git a/examples/simulation_based_inference.ipynb b/examples/simulation_based_inference.ipynb
index 6ad52e58e..e1f888324 100644
--- a/examples/simulation_based_inference.ipynb
+++ b/examples/simulation_based_inference.ipynb
@@ -20,8 +20,9 @@
     "import requests\n",
     "import re\n",
     "\n",
-    "import outlines.models as models\n",
-    "import outlines.text as text\n",
+    "import openai\n",
+    "\n",
+    "import outlines\n",
     "\n",
     "random.seed(0)"
    ]
@@ -49,7 +50,7 @@
    "id": "aabb4db6-fd94-4c42-ab7f-97c3de45b2cc",
    "metadata": {},
    "source": [
-    "In this example we will use GPT 3.5 to solve problems from the GSM-8K dataset. The state-of-the-art performance on this dataset is obtained using few-shot prompting with 5 examples. However, it is not clear how one should select these examples. Here, we will use **simulation-based inference** to try to infer which examples we should be using to get the best out of the model's abilities to solve the problem.\n",
+    "In this example we will use GPT 4 mini to solve problems from the GSM-8K dataset. The state-of-the-art performance on this dataset is obtained using few-shot prompting with 5 examples. However, it is not clear how one should select these examples. Here, we will use **simulation-based inference** to try to infer which examples we should be using to get the best out of the model's abilities to solve the problem.\n",
     "\n",
     "Let's start with downloading the dataset:"
    ]
@@ -112,8 +113,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "@text.prompt\n",
-    "def few_shots(question, examples):\n",
+    "few_shots = outlines.Template.from_string(\n",
     "    \"\"\"\n",
     "    {% for example in examples %}\n",
     "    Q: {{ example.question }}\n",
@@ -122,16 +122,16 @@
     "    Q: {{ question }}\n",
     "    A:\n",
     "    \"\"\"\n",
+    ")\n",
     "\n",
-    "\n",
-    "model = models.text_completion.openai(\"text-davinci-003\", max_tokens=128)\n",
+    "model = outlines.from_openai(openai.OpenAI(), \"gpt-4o-mini\")\n",
     "\n",
     "\n",
     "# TODO: This could largely benefit from vectorization in #52\n",
     "def one_train_example(problem, example_set):\n",
     "    example_ids = random.choices(range(0, len(example_set)), k=5)\n",
     "    examples = [example_set[i] for i in example_ids]\n",
-    "    prompt = few_shots(problem[\"question\"], examples)\n",
+    "    prompt = few_shots(question=problem[\"question\"], examples=examples)\n",
     "    answers_raw = model(prompt, samples=20)\n",
     "\n",
     "    samples = []\n",
@@ -203,7 +203,7 @@
     },
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 864x576 with 1 Axes>"
       ]
diff --git a/examples/transformers_integration.py b/examples/transformers_integration.py
deleted file mode 100644
index 16e2b8ec3..000000000
--- a/examples/transformers_integration.py
+++ /dev/null
@@ -1,25 +0,0 @@
-"""Example of integrating `outlines` with `transformers`."""
-
-from pydantic import BaseModel
-from transformers import pipeline
-
-from outlines.integrations.transformers import JSONPrefixAllowedTokens
-
-
-class Person(BaseModel):
-    first_name: str
-    surname: str
-
-
-pipe = pipeline("text-generation", model="mistralai/Mistral-7B-v0.1")
-prefix_allowed_tokens_fn = JSONPrefixAllowedTokens(
-    schema=Person, tokenizer_or_pipe=pipe, whitespace_pattern=r" ?"
-)
-results = pipe(
-    ["He is Tom Jones", "She saw Linda Smith"],
-    return_full_text=False,
-    do_sample=False,
-    max_new_tokens=50,
-    prefix_allowed_tokens_fn=prefix_allowed_tokens_fn,
-)
-print(results)
diff --git a/examples/vllm_integration.py b/examples/vllm_integration.py
index ee1b24a33..dfe873517 100644
--- a/examples/vllm_integration.py
+++ b/examples/vllm_integration.py
@@ -17,7 +17,10 @@ class Person(BaseModel):
 llm = vllm.LLM(model=MODEL_ID, max_model_len=512)
 tokenizer = adapt_tokenizer(AutoTokenizer.from_pretrained(MODEL_ID))
 logits_processor = JSONLogitsProcessor(
-    schema=Person, tokenizer=tokenizer, whitespace_pattern=r" ?"
+    schema=Person,
+    tokenizer=tokenizer,
+    tensor_library_name="torch",
+    whitespace_pattern=r" ?"
 )
 result = llm.generate(
     ["He is Tom Jones", "She saw Linda Smith"],

From 5b35e061d6fb113c68a1b41be2b0073617586be0 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Apr 2025 09:54:57 +0200
Subject: [PATCH 309/505] Add examples' prompt files to load the prompt
 templates from file

---
 examples/prompts/babyagi_create_task.txt     |  9 +++++++++
 examples/prompts/babyagi_perform_task.txt    |  5 +++++
 examples/prompts/babyagi_prioritize_task.txt |  7 +++++++
 examples/prompts/dating_profile.txt          | 11 +++++++++++
 examples/prompts/pick_odd_one_out.txt        |  9 +++++++++
 examples/prompts/self_consistency.txt        |  6 ++++++
 6 files changed, 47 insertions(+)
 create mode 100644 examples/prompts/babyagi_create_task.txt
 create mode 100644 examples/prompts/babyagi_perform_task.txt
 create mode 100644 examples/prompts/babyagi_prioritize_task.txt
 create mode 100644 examples/prompts/dating_profile.txt
 create mode 100644 examples/prompts/pick_odd_one_out.txt
 create mode 100644 examples/prompts/self_consistency.txt

diff --git a/examples/prompts/babyagi_create_task.txt b/examples/prompts/babyagi_create_task.txt
new file mode 100644
index 000000000..434063aad
--- /dev/null
+++ b/examples/prompts/babyagi_create_task.txt
@@ -0,0 +1,9 @@
+Objective: {{ objective }}
+Current Task: {{ task }}
+Result: {{ result }}
+Previous Tasks: {{ previous_tasks }}
+
+Based on the result, create a list of new tasks that will help achieve the objective.
+Please provide the tasks in the following format:
+1. [Task description]
+2. [Task description]
diff --git a/examples/prompts/babyagi_perform_task.txt b/examples/prompts/babyagi_perform_task.txt
new file mode 100644
index 000000000..bd8098924
--- /dev/null
+++ b/examples/prompts/babyagi_perform_task.txt
@@ -0,0 +1,5 @@
+Objective: {{ objective }}
+Task: {{ task }}
+
+Please perform the task and provide a concise result in the following format:
+Result: [Your concise result here]
diff --git a/examples/prompts/babyagi_prioritize_task.txt b/examples/prompts/babyagi_prioritize_task.txt
new file mode 100644
index 000000000..f42c8b05f
--- /dev/null
+++ b/examples/prompts/babyagi_prioritize_task.txt
@@ -0,0 +1,7 @@
+Tasks: {{ tasks }}
+Next Task ID: {{ next_task_id }}
+
+Please prioritize the tasks based on their importance and urgency to achieve the objective.
+Provide the prioritized tasks in the following format:
+1. [Task ID]. [Task description]
+2. [Task ID]. [Task description]
diff --git a/examples/prompts/dating_profile.txt b/examples/prompts/dating_profile.txt
new file mode 100644
index 000000000..129be4582
--- /dev/null
+++ b/examples/prompts/dating_profile.txt
@@ -0,0 +1,11 @@
+You are a world-renowned matchmaker who understands the modern dating market. Your job is to generate dating app profiles for male clients interested in women based on a provided description. The profiles should be authentic, show off their strengths, and maximize their likelihood of getting matches on dating apps.
+Here are some examples of past clients that you have successfully created profiles for:
+{% for example in examples %}
+Description:
+{{ example.description }}
+Profile:
+{{ example.profile }}
+{% endfor %}
+Here is the new client who you need to create a profile for:
+Description: {{ description }}
+Profile:
diff --git a/examples/prompts/pick_odd_one_out.txt b/examples/prompts/pick_odd_one_out.txt
new file mode 100644
index 000000000..4a00ceb00
--- /dev/null
+++ b/examples/prompts/pick_odd_one_out.txt
@@ -0,0 +1,9 @@
+Pick the odd word out: skirt, dress, pen, jacket.
+skirt is clothing, dress is clothing, pen is an object, jacket is clothing.
+So the odd one is pen.
+
+Pick the odd word out: Spain, France, German, England, Singapore.
+Spain is a country, France is a country, German is a language, ...
+So the odd one is German.
+
+Pick the odd word out: {{ options | join(", ") }}.
diff --git a/examples/prompts/self_consistency.txt b/examples/prompts/self_consistency.txt
new file mode 100644
index 000000000..29f1f8a74
--- /dev/null
+++ b/examples/prompts/self_consistency.txt
@@ -0,0 +1,6 @@
+{% for example in examples %}
+Q: {{ example.question }}
+A: {{ example.answer }}
+{% endfor %}
+Q: {{ question }}
+A:

From 0beeb6f9a12280b3ae689be34312daefa7c518b1 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 11 Apr 2025 15:40:22 +0200
Subject: [PATCH 310/505] Update the cookbooks to use v1 interface

---
 docs/cookbook/atomic_caption.md               |  52 ++++----
 docs/cookbook/chain_of_density.md             |  51 ++-----
 docs/cookbook/chain_of_thought.md             |  87 ++++++------
 docs/cookbook/classification.md               |  61 +++------
 docs/cookbook/dating_profiles.md              |  30 +++--
 docs/cookbook/deploy-using-bentoml.md         |  22 ++-
 docs/cookbook/deploy-using-cerebrium.md       |  11 +-
 docs/cookbook/deploy-using-modal.md           |  16 ++-
 docs/cookbook/earnings-reports.md             |  24 ++--
 docs/cookbook/extract_event_details.md        |   2 +-
 docs/cookbook/extract_event_details.py        |  31 +++--
 docs/cookbook/extraction.md                   |  16 ++-
 docs/cookbook/index.md                        |   5 +
 docs/cookbook/knowledge_graph_extraction.md   | 102 +++++++-------
 docs/cookbook/models_playing_chess.md         |  28 ++--
 .../prompt_templates/chain_of_density.txt     |  27 ++++
 .../prompt_templates/classification.txt       |  18 +++
 .../cookbook/prompt_templates/react_agent.txt |  27 ++++
 .../simtom_prospective_taking.txt             |  10 ++
 .../prompt_templates/simtom_simulation.txt    |   7 +
 docs/cookbook/qa-with-citations.md            | 125 +++++++++---------
 docs/cookbook/react_agent.md                  |  90 ++++---------
 docs/cookbook/read-pdfs.md                    |  79 +++++------
 docs/cookbook/receipt-digitization.md         |  99 +++++++-------
 docs/cookbook/simtom.md                       |  52 +++-----
 .../structured_generation_workflow.md         |  62 ++++-----
 26 files changed, 565 insertions(+), 569 deletions(-)
 create mode 100644 docs/cookbook/prompt_templates/chain_of_density.txt
 create mode 100644 docs/cookbook/prompt_templates/classification.txt
 create mode 100644 docs/cookbook/prompt_templates/react_agent.txt
 create mode 100644 docs/cookbook/prompt_templates/simtom_prospective_taking.txt
 create mode 100644 docs/cookbook/prompt_templates/simtom_simulation.txt

diff --git a/docs/cookbook/atomic_caption.md b/docs/cookbook/atomic_caption.md
index a2b43e5e2..9443f87a7 100644
--- a/docs/cookbook/atomic_caption.md
+++ b/docs/cookbook/atomic_caption.md
@@ -1,5 +1,5 @@
 # Vision-Language Models with Outlines
-This guide demonstrates how to use Outlines with vision-language models, leveraging the new transformers_vision module. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
+This guide demonstrates how to use Outlines with vision-language models, leveraging the outlines `TransformersMultiModal` model. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
 
 We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
 
@@ -11,17 +11,21 @@ pip install outlines transformers torch
 ```
 
 ### Initializing the Model
-We'll use the transformers_vision function to initialize our vision-language model. This function is specifically designed to handle models that can process both text and image inputs. Today we'll be using the Pixtral model with the llama tokenizer. (Currently the mistral tokenizer is pending support).
+We'll use the `outlines.from_transformers` function to initialize our vision-language model. For this function to return a vision multi-modal model we need to pass in a transformers model and a transformers processor that can handle both text and image inputs. Today we'll be using the Pixtral model with the AutoProcessor.
 
 ```python
+import outlines
 import torch
 from transformers import (
-    LlavaForConditionalGeneration,
+    AutoProcessor,
+    LlavaForConditionalGeneration
 )
+
 model_name="mistral-community/pixtral-12b" # original magnet model is able to be loaded without issue
 model_class=LlavaForConditionalGeneration
+processor_class=AutoProcessor
 
-def get_vision_model(model_name: str, model_class: VisionModel):
+def get_vision_model(model_name: str, model_class, processor_class):
     model_kwargs = {
         "torch_dtype": torch.bfloat16,
         "attn_implementation": "flash_attention_2",
@@ -31,28 +35,25 @@ def get_vision_model(model_name: str, model_class: VisionModel):
         "device": "cuda",
     }
 
-    model = outlines.models.transformers_vision(
-        model.model_name,
-        model_class=model.model_class,
-        model_kwargs=model_kwargs,
-        processor_kwargs=processor_kwargs,
+    model = outlines.from_transformers(
+        model_class.from_pretrained(model_name, **model_kwargs),
+        processor_class.from_pretrained(model_name, **processor_kwargs),
     )
     return model
-model = get_vision_model(model_name, model_class)
+model = get_vision_model(model_name, model_class, processor_class)
 ```
 
 ### Defining the Schema
-Next, we'll define a schema for the output we expect from our vision-language model. This schema will help structure the model's responses.
-
+Next, we'll define a schema for the output we expect from our vision multi-modal model. This schema will help structure the model's responses. We use the `outlines.Generator` object to create a generator for our schema that will then be called with our prompt and images.
 
 ```python
+from enum import Enum
 from pydantic import BaseModel, Field, confloat, constr
 from pydantic.types import StringConstraints, PositiveFloat
 from typing import List
 from typing_extensions import Annotated
 
-from enum import StrEnum
-class TagType(StrEnum):
+class TagType(Enum):
     ENTITY = "Entity"
     RELATIONSHIP = "Relationship"
     STYLE = "Style"
@@ -86,7 +87,7 @@ class ImageData(BaseModel):
     short_caption: Annotated[str, StringConstraints(min_length=10, max_length=150)]
     dense_caption: Annotated[str, StringConstraints(min_length=100, max_length=2048)]
 
-image_data_generator = outlines.generate.json(model, ImageData)
+image_data_generator = outlines.Generator(model, ImageData)
 ```
 
 This schema defines the structure for image tags, including categories like Entity, Relationship, Style, etc., as well as short and dense captions.
@@ -95,7 +96,6 @@ This schema defines the structure for image tags, including categories like Enti
 
 We'll create a prompt that instructs the model on how to analyze the image and generate the structured output:
 
-
 ```python
 pixtral_instruction = """
 <s>[INST]
@@ -131,30 +131,34 @@ pixtral_instruction = """
 </Examples>
 </TagCategories>
 <ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
-<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG][/INST]
+<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG]<image>[/INST]
 """.strip()
 ```
 
-This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required.
+This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required. It's essential to include the <image> tag in the prompt at the location where the image will be inserted.
 
 ### Generating Structured Output
 Now we can use our model to generate structured output based on an input image:
 
 ```python
+from io import BytesIO
+from urllib.request import urlopen
+from PIL import Image
+
 def img_from_url(url):
     img_byte_stream = BytesIO(urlopen(url).read())
     return Image.open(img_byte_stream).convert("RGB")
 
 image_url="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
 image= img_from_url(image_url)
-result = image_data_generator(
-    pixtral_instruction,
-    [image]
-)
+result = image_data_generator({
+    "text": pixtral_instruction,
+    "images": image
+})
 print(result)
 ```
 
-This code loads an image from a URL, passes it to our vision-language model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
+This code loads an image from a URL, passes it to our vision multi-modal model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
 
 ```json
 {'tags_list': [{'tag': 'astronaut',
@@ -184,6 +188,6 @@ This code loads an image from a URL, passes it to our vision-language model alon
 ```
 
 ## Conclusion
-The transformers_vision module in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
+The `TransformersMultiModal` model in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
 
 By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
diff --git a/docs/cookbook/chain_of_density.md b/docs/cookbook/chain_of_density.md
index 10bda9771..8a822fc2e 100644
--- a/docs/cookbook/chain_of_density.md
+++ b/docs/cookbook/chain_of_density.md
@@ -26,42 +26,12 @@ The prompt also asks the model to return a list of JSON objects that contain the
 
 ![Figure 2 in the paper](./images/chain_of_density.png)
 
-We can now implement the prompt provided in the paper:
+We can now implement the prompt provided in the paper. We stored the prompt template in a text file, and we can load it using the `Template` class:
 
 ```python
 from outlines import Template
 
-
-chain_of_density = Template.from_string(
-    """Article: {{ article }}
-
-    You will generate increasingly concise, entity-dense summaries of the above Article.
-
-    Repeat the following 2 steps 5 times.
-
-    Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
-    Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
-
-    A Missing Entity is:
-    - Relevant: to the main story.
-    - Specific: descriptive yet concise (5 words or fewer).
-    - Novel: not in the previous summary.
-    - Faithful: present in the Article.
-    - Anywhere: located anywhere in the Article.
-
-    Guidelines:
-    - The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
-    - Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
-    - Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
-    - The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
-    - Missing entities can appear anywhere in the new summary.
-    - Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
-
-    Remember, use the exact same number of words for each summary.
-
-    Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
-    """
-)
+chain_of_density = Template.from_file("prompt_templates/chain_of_density.txt")
 ```
 
 ??? Note
@@ -84,19 +54,26 @@ class Summaries(BaseModel):
     summaries: conlist(Summary, max_length=5, min_length=5)
 ```
 
-We now generate the prompt by passing the article we want to summarize to the template. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use JSON-structured generation to generate the summaries:
+We now generate the prompt by passing the article we want to summarize to the prompt template previously loaded. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use the `Summaries` schema to generate the summaries with structured generation:
 
 ```python
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ")
+import outlines
+import transformers
 
-prompt = chain_of_density(article)
-result = outlines.generate.json(model, Summaries)(prompt)
+MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
+
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+)
+prompt = chain_of_density(article=article)
+result = model(prompt, Summaries, max_new_tokens=2000)
 ```
 
 We can now check the results:
 
 ```python
-print(result.model_dump())
+print(result)
 # {'summaries': [
 #     {
 #       'missing_entities': 'English mathematician, cryptanalyst, philosopher',
diff --git a/docs/cookbook/chain_of_thought.md b/docs/cookbook/chain_of_thought.md
index 5ac18af25..42d6c8550 100644
--- a/docs/cookbook/chain_of_thought.md
+++ b/docs/cookbook/chain_of_thought.md
@@ -11,20 +11,23 @@ We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-
 pip install llama-cpp-python
 ```
 
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
+
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -38,20 +41,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Chain of thought
@@ -77,30 +69,27 @@ class Reasoning(BaseModel):
 json_schema = Reasoning.model_json_schema()
 ```
 
-We could generate a response using the json schema but for a change we will use the regex:
-
-```python
-from outlines.fsm.json_schema import convert_json_schema_to_str
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-schema_str = convert_json_schema_to_str(json_schema=json_schema)
-regex_str = build_regex_from_schema(schema_str)
-```
-
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(user_prompt):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON "
-        f"Here's the json schema you must adhere to:\n<schema>\n{json_schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + user_prompt
-        + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+generate_hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ json_schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    {{ user_prompt }}
+    <|im_end|>
+    <|im_start|>assistant
+    <schema>
+    """
+)
 ```
 
 For a given user prompt:
@@ -109,11 +98,11 @@ For a given user prompt:
 user_prompt = "9.11 and 9.9 -- which is bigger?"
 ```
 
-we can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+We can use `outlines.Generator` with the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-generator = generate.regex(model, regex_str)
-prompt = generate_hermes_prompt(user_prompt)
+generator = outlines.Generator(model, regex_str)
+prompt = generate_hermes_prompt(json_schema=json_schema, user_prompt=user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
diff --git a/docs/cookbook/classification.md b/docs/cookbook/classification.md
index 023e4543a..18a933b47 100644
--- a/docs/cookbook/classification.md
+++ b/docs/cookbook/classification.md
@@ -3,52 +3,40 @@
 Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
 
 
-This tutorial shows how one can implement multi-label classification using Outlines. We will use two functionalities of the library: `generate.choice` and `generate.json`.
+This tutorial shows how one can implement multi-label classification using Outlines.
 
 As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
 
 ```python
 import outlines
+import transformers
 
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
+
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 ```
 
-We will use the following prompt template:
+We will use a prompt template stored in a text file:
 
 ```python
 from outlines import Template
 
-
-customer_support = Template.from_string(
-    """You are an experienced customer success manager.
-
-    Given a request from a client, you need to determine when the
-    request is urgent using the label "URGENT" or when it can wait
-    a little with the label "STANDARD".
-
-    # Examples
-
-    Request: "How are you?"
-    Label: STANDARD
-
-    Request: "I need this fixed immediately!"
-    Label: URGENT
-
-    # TASK
-
-    Request: {{ request }}
-    Label: """
-)
+customer_support = Template.from_file("prompt_templates/classification.txt")
 ```
 
 ## Choosing between multiple choices
 
-Outlines provides a shortcut to do multi-label classification, using the `outlines.generate.choice` function to initialize a generator. Outlines uses multinomial sampling by default, here we will use the greedy sampler to get the label with the highest probability:
+Outlines provides a convenient way to do multi-label classification, passing a Literal type hint to the `outlines.Generator` object:
 
 ```python
-from outlines.samplers import greedy
+from typing import Literal
+import outlines
+
+generator = outlines.Generator(model, Literal["URGENT", "STANDARD"])
 
-generator = outlines.generate.choice(model, ["URGENT", "STANDARD"], sampler=greedy())
 ```
 Outlines supports batched requests, so we will pass two requests to the model:
 
@@ -58,10 +46,10 @@ requests = [
     "Just wanted to say hi"
 ]
 
-prompts = [customer_support(request) for request in requests]
+prompts = [customer_support(request=request) for request in requests]
 ```
 
-We can now asks the model to classify the requests:
+We can now ask the model to classify the requests:
 
 ```python
 labels = generator(prompts)
@@ -69,15 +57,6 @@ print(labels)
 # ['URGENT', 'STANDARD']
 ```
 
-Now, you might be in a hurry and don't want to wait until the model finishes completion. After all, you only need to see the first letter of the response to know whether the request is urgent or standard. You can instead stream the response:
-
-```python
-tokens = generator.stream(prompts)
-labels = ["URGENT" if "U" in token else "STANDARD" for token in next(tokens)]
-print(labels)
-# ['URGENT', 'STANDARD']
-```
-
 ## Using JSON-structured generation
 
 Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
@@ -96,11 +75,11 @@ class Classification(BaseModel):
     label: Label
 ```
 
-and we can use `generate.json` by passing this Pydantic model we just defined, and call the generator:
+We can then create a generator with the Pydantic model we just defined and call it:
 
 ```python
-generator = outlines.generate.json(model, Classification, sampler=greedy())
+generator = outlines.Generator(model, Classification)
 labels = generator(prompts)
 print(labels)
-# [Classification(label=<Label.urgent: 'URGENT'>), Classification(label=<Label.standard: 'STANDARD'>)]
+# ['{"label":"URGENT"}', '{ "label": "STANDARD" }']
 ```
diff --git a/docs/cookbook/dating_profiles.md b/docs/cookbook/dating_profiles.md
index 1cbea556c..5852e6df9 100644
--- a/docs/cookbook/dating_profiles.md
+++ b/docs/cookbook/dating_profiles.md
@@ -3,6 +3,7 @@
 In this example we will see how we can use Outlines to generate synthetic data for a dating application. This example was originally contributed by [Vibhor Kumar](https://github.com/veezbo).
 
 ```python
+import json
 from dataclasses import dataclass
 from enum import Enum
 
@@ -132,20 +133,21 @@ samples: list[Example] = [
 We will use Mosaic's MPT-7B model (requires 13GB of GPU memory) which can fit on a single GPU with a reasonable context window. We initialize it with Outlines:
 
 ```python
+MODEL_NAME = "mosaicml/mpt-7b-8k-instruct"
+
 config = transformers.AutoConfig.from_pretrained(
-    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
+    MODEL_NAME, trust_remote_code=True
 )
 config.init_device = "meta"
-model = outlines.models.transformers(
-    model_name="mosaicml/mpt-7b-8k-instruct",
-    device="cuda",
-    model_kwargs={
-        "config": config,
-        "trust_remote_code": True,
-        "torch_dtype": torch.bfloat16,
-        "device_map": {"": 0},
-    },
-)
+model_kwargs = {
+    "config": config,
+    "trust_remote_code": True,
+    "torch_dtype": torch.bfloat16,
+    "device_map": "cuda",
+}
+tf_model = transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME, **model_kwargs)
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+model = outlines.from_transformers(tf_model, tokenizer=tf_tokenizer)
 ```
 
 ## JSON-structured generation of profiles
@@ -162,9 +164,9 @@ every few  months. I also like weddings because I get to wear those suits, and
 it's  a good excuse for a date. I watch the latest series because I'm paying,
 with my hard-earned money, for every streaming service."""
 
-prompt = dating_profile_prompt(new_description, samples)
-profile = outlines.generate.json(model, DatingProfile)(prompt)
-parsed_profile = DatingProfile.model_validate_json(profile)
+prompt = dating_profile_prompt(description=new_description, examples=samples)
+profile = model(prompt, DatingProfile)
+parsed_profile = DatingProfile.model_validate_json(json.loads(profile))
 ```
 
 ## Results
diff --git a/docs/cookbook/deploy-using-bentoml.md b/docs/cookbook/deploy-using-bentoml.md
index 2926df0f6..5d000cb52 100644
--- a/docs/cookbook/deploy-using-bentoml.md
+++ b/docs/cookbook/deploy-using-bentoml.md
@@ -119,15 +119,22 @@ class Outlines:
     bento_model_ref = bentoml.models.get(BENTO_MODEL_TAG)
 
     def __init__(self) -> None:
-
         import outlines
         import torch
-        self.model = outlines.models.transformers(
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+
+        # Load tokenizer and model from the BentoML model reference path
+        hf_tokenizer = AutoTokenizer.from_pretrained(self.bento_model_ref.path)
+        hf_model = AutoModelForCausalLM.from_pretrained(
             self.bento_model_ref.path,
-            device="cuda",
-            model_kwargs={"torch_dtype": torch.float16},
+            torch_dtype=torch.float16,
+            low_cpu_mem_usage=True,
+            device_map="cuda"
         )
 
+        # Then use the loaded model with Outlines
+        self.model = outlines.from_transformers(hf_model, hf_tokenizer)
+
     ...
 ```
 
@@ -142,13 +149,14 @@ We then need to define an HTTP endpoint using `@bentoml.api` to decorate the met
         prompt: str = "Give me a character description.",
         json_schema: t.Optional[str] = DEFAULT_SCHEMA,
     ) -> t.Dict[str, t.Any]:
-
+        import json
         import outlines
+        from outlines.types import JsonSchema
 
-        generator = outlines.generate.json(self.model, json_schema)
+        generator = outlines.Generator(self.model, JsonSchema(json_schema))
         character = generator(prompt)
 
-        return character
+        return json.loads(character)
 ```
 
 Here `@bentoml.api` decorator defines `generate` as an HTTP endpoint that accepts a JSON request body with two fields: `prompt` and `json_schema` (optional, which allows HTTP clients to provide their own JSON schema). The type hints in the function signature will be used to validate incoming JSON requests. You can define as many HTTP endpoints as you want by using `@bentoml.api` to decorate other methods of `Outlines` class.
diff --git a/docs/cookbook/deploy-using-cerebrium.md b/docs/cookbook/deploy-using-cerebrium.md
index 6fbaab390..bf010d583 100644
--- a/docs/cookbook/deploy-using-cerebrium.md
+++ b/docs/cookbook/deploy-using-cerebrium.md
@@ -36,7 +36,7 @@ provider = "aws"
 region = "us-east-1"
 
 [cerebrium.dependencies.pip]
-outline = "==0.0.37"
+outline = "==1.0.0"
 transformers = "==4.38.2"
 datasets = "==2.18.0"
 accelerate = "==0.27.2"
@@ -48,9 +48,14 @@ Running code in Cerebrium is like writing normal python with no special syntax.
 
 ```python
 import outlines
+import transformers
+from outlines.types import JsonSchema
 
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
 schema = """{
     "title": "Character",
@@ -89,7 +94,7 @@ schema = """{
     }
 }"""
 
-generator = outlines.generate.json(model, schema)
+generator = outlines.Generator(model, JsonSchema(schema))
 ```
 
 On first deploy, it will download the model and store it on disk therefore for subsequent calls it will load the model from disk.
diff --git a/docs/cookbook/deploy-using-modal.md b/docs/cookbook/deploy-using-modal.md
index 15e200cb5..23a846a6d 100644
--- a/docs/cookbook/deploy-using-modal.md
+++ b/docs/cookbook/deploy-using-modal.md
@@ -68,7 +68,12 @@ When running longer Modal apps, it's recommended to download your language model
 # downloading models, setting up environment variables, etc.
 def import_model():
     import outlines
-    outlines.models.transformers(language_model)
+    import transformers
+
+    outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(language_model),
+        transformers.AutoTokenizer.from_pretrained(language_model)
+    )
 
 # This line tells the container to run the import_model function when it starts.
 outlines_image = outlines_image.run_function(import_model)
@@ -134,19 +139,22 @@ def generate(
     # so we need to import the necessary libraries here. You should
     # do this with any other libraries you might need.
     import outlines
+    import transformers
+    from outlines.types import JsonSchema
 
     # Load the model into memory. The import_model function above
     # should have already downloaded the model, so this call
     # only loads the model into GPU memory.
-    model = outlines.models.transformers(
-        language_model, device="cuda"
+    outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(language_model, device_map="cuda"),
+        transformers.AutoTokenizer.from_pretrained(language_model)
     )
 
     # Generate a character description based on the prompt.
     # We use the .json generation method -- we provide the
     # - model: the model we loaded above
     # - schema: the JSON schema we defined above
-    generator = outlines.generate.json(model, schema)
+    generator = outlines.Generator(model, JsonSchema(schema))
 
     # Make sure you wrap your prompt in instruction tags ([INST] and [/INST])
     # to indicate that the prompt is an instruction. Instruction tags can vary
diff --git a/docs/cookbook/earnings-reports.md b/docs/cookbook/earnings-reports.md
index 26c2a9ccf..cca1e9513 100644
--- a/docs/cookbook/earnings-reports.md
+++ b/docs/cookbook/earnings-reports.md
@@ -26,16 +26,14 @@ Choose your language model. We'll use Phi-3 mini, which is small enough to run o
 ```python
 import outlines
 import torch
+import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.models.transformers(
-    model_name,
-    device='auto',
-    model_kwargs={
-        # To reduce memory usage, we'll use bfloat16
-        "torch_dtype": torch.bfloat16,
-    },
+tf_model = transformers.AutoModelForCausalLM.from_pretrained(
+    model_name, device_map="cuda", torch_dtype=torch.bfloat16
 )
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
+model = outlines.from_transformers(tf_model, tf_tokenizer)
 ```
 
 ## Set up the data
@@ -156,7 +154,7 @@ year,revenue,operating_income,net_income,basic_earnings_per_share(
 \d{4},((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),(-?\d+(?:\.\d{1,2})?)){,3}
 ```
 
-Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.generate.regex` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
+Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.Generator` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
 
 ## Prompting the model
 
@@ -234,12 +232,12 @@ def extract_financial_data_prompt(columns_to_extract, income_statement):
 
 Now that we have our prompt and regular expression, we can run the model.
 
-Construct our regex extractor function. We'll use a greedy sampler, which samples the most likely next token at each step. It's a simple sampler that is more reproducible than multinomial sampling.
+Construct our regex extractor function.
 
 ```python
-csv_extractor = outlines.generate.regex(
-    model, csv_regex, sampler=outlines.samplers.greedy()
-)
+from outlines.types import Regex
+
+csv_extractor = outlines.Generator(model, Regex(csv_regex))
 ```
 
 Provide the prompt to the model and run it:
@@ -247,7 +245,7 @@ Provide the prompt to the model and run it:
 ```python
 csv_data = csv_extractor(
     extract_financial_data_prompt(columns_to_extract, income_statement),
-    max_tokens=1024,
+    max_new_tokens=1024,
 )
 
 print(csv_data)
diff --git a/docs/cookbook/extract_event_details.md b/docs/cookbook/extract_event_details.md
index 0f87d9586..063433ad2 100644
--- a/docs/cookbook/extract_event_details.md
+++ b/docs/cookbook/extract_event_details.md
@@ -8,7 +8,7 @@ place, 3 avenue des tanneurs, one hour will be enough see you 😘
 ```
 
 Let see how to extract the event details from the message with the MLX
-library dedicated to  Apple Silicon processor  (M series).
+library dedicated to Apple Silicon processor (M series).
 
 ```python
 --8<-- "docs/cookbook/extract_event_details.py"
diff --git a/docs/cookbook/extract_event_details.py b/docs/cookbook/extract_event_details.py
index a4c7f9188..cd90bc439 100644
--- a/docs/cookbook/extract_event_details.py
+++ b/docs/cookbook/extract_event_details.py
@@ -4,7 +4,7 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines import Generator
+from outlines import Generator, Template
 
 
 # Load the model
@@ -19,31 +19,36 @@ class Event(BaseModel):
         default=None, description="date of the event if available in iso format"
     )
 
+# Load the prompt template from a string
+prompt_template = Template.from_string(
+    """
+    Today's date and time are {{ now }}
+    Given a user message, extract information of the event like date and time in iso format, location and title.
+    If the given date is relative, think step by step to find the right date.
+    Here is the message:
+    {{ message }}
+    """
+)
 
 # Get the current date and time
 now = datetime.now().strftime("%A %d %B %Y and it's %H:%M")
 
-# Define the prompt
-prompt = f"""
-Today's date and time are {now}
-Given a user message, extract information of the event like date and time in iso format, location and title.
-If the given date is relative, think step by step to find the right date.
-Here is the message:
-"""
-
 # Sample message
-message = """Hello Kitty, my grandmother will be here , I think it's better to postpone our
+message = """Hello Kitty, my grandmother will be here, I think it's better to postpone our
 appointment to review math lessons to next Friday at 2pm at the same place, 3 avenue des tanneurs, I think that one hour will be enough
 see you 😘 """
 
 # Create the generator
 generator = Generator(model, Event)
 
+# Create the prompt
+prompt = prompt_template(now=now, message=message)
+
 # Extract the event information
-event = generator(prompt + message)
+event = generator(prompt)
 
 # Print the current date and time
 print(f"Today: {now}")
 
-# Print the extracted event information in JSON format
-print(event.json())
+# Print the extracted event information
+print(event)
diff --git a/docs/cookbook/extraction.md b/docs/cookbook/extraction.md
index dfcca0cfd..cfb39ca5c 100644
--- a/docs/cookbook/extraction.md
+++ b/docs/cookbook/extraction.md
@@ -7,9 +7,14 @@ Getting LLMs to output the extracted entities in a structured format can be chal
 As always, we start with initializing the model. We will be using a quantized version of Mistal-7B-v0.1 (we're GPU poor):
 
 ```python
+import transformers
 import outlines
 
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name, device_map="cuda"),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
 ```
 
 And we will be using the following prompt template:
@@ -17,7 +22,6 @@ And we will be using the following prompt template:
 ```python
 from outlines import Template
 
-
 take_order = Template.from_string(
     """You are the owner of a pizza parlor. Customers \
     send you orders from which you need to extract:
@@ -70,14 +74,14 @@ orders = [
     "Hi! I would like to order two pepperonni pizzas and would like them in 30mins.",
     "Is it possible to get 12 margheritas?"
 ]
-prompts = [take_order(order) for order in orders]
+prompts = [take_order(order=order) for order in orders]
 
-generator = outlines.generate.json(model, Order)
+generator = outlines.Generator(model, Order)
 
 results = generator(prompts)
 print(results)
-# [Order(pizza=<Pizza.pepperonni: 'Pepperoni'>, number=2),
-#  Order(pizza=<Pizza.margherita: 'Margherita'>, number=12)]
+# ['{"pizza": "Pepperoni", "number": 2}',
+# '{"pizza": "Margherita", "number": 12}']
 ```
 
 There are several ways you could improve this example:
diff --git a/docs/cookbook/index.md b/docs/cookbook/index.md
index c36b98969..931262053 100644
--- a/docs/cookbook/index.md
+++ b/docs/cookbook/index.md
@@ -16,3 +16,8 @@ This part of the documentation provides a few cookbooks that you can browse to g
 - [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
 - [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
 - [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
+- [Extract Event Details](extract_event_details.md): Use Outlines with MLX to extract key information from a text message related to an event.
+- [Generating Synthetic Phone Numbers](structured_generation_workflow.md): Use Outlines to create phone numbers that are valid for Washington State.
+- [Deploying Outlines using BentoML](deploy-using-bentoml.md): Deploy an Outlines model using BentoML.
+- [Deploying Outlines using Cerebras](deploy-using-cerebras.md): Deploy an Outlines model using Cerebras.
+- [Deploying Outlines using Modal](deploy-using-modal.md): Deploy an Outlines model using Modal.
diff --git a/docs/cookbook/knowledge_graph_extraction.md b/docs/cookbook/knowledge_graph_extraction.md
index 6a6877756..f4b21c84c 100644
--- a/docs/cookbook/knowledge_graph_extraction.md
+++ b/docs/cookbook/knowledge_graph_extraction.md
@@ -8,20 +8,23 @@ We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama
 pip install llama-cpp-python
 ```
 
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
+
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -35,20 +38,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Knowledge Graph Extraction
@@ -92,17 +84,24 @@ schema = KnowledgeGraph.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(user_prompt):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + user_prompt
-        + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+generate_hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    {{ user_prompt }}
+    <|im_end|>
+    <|im_start|>assistant
+    <schema>
+    """
+)
 ```
 
 For a given user prompt, for example:
@@ -111,27 +110,26 @@ For a given user prompt, for example:
 user_prompt = "Alice loves Bob and she hates Charlie."
 ```
 
-We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-from outlines import generate, models
+from outlines import Generator
 
-model = models.LlamaCpp(llm)
-generator = generate.json(model, KnowledgeGraph)
-prompt = generate_hermes_prompt(user_prompt)
+generator = Generator(model, KnowledgeGraph)
+prompt = generate_hermes_prompt(schema=schema, user_prompt=user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
 We obtain the nodes and edges of the knowledge graph:
 
 ```python
-print(response.nodes)
-print(response.edges)
-# [Node(id=1, label='Alice', property='Person'),
-# Node(id=2, label='Bob', property='Person'),
-# Node(id=3, label='Charlie', property='Person')]
-# [Edge(source=1, target=2, label='love', property='Relationship'),
-# Edge(source=1, target=3, label='hate', property='Relationship')]
+print(response)
+# {"nodes":[{"id":1,"label":"Alice","property":"loves,hates"},
+# {"id":2,"label":"Bob","property":"loved_by"},
+# {"id":3,"label":"Charlie","property":"hated_by"}],
+# "edges":[{"source":1,"target":2,"label":"loves","property":"love"},
+# {"source":1,"target":3,"label":"hates","property":"hate"}]}
+
 ```
 
 ## (Optional) Visualizing the Knowledge Graph
@@ -142,10 +140,10 @@ We can use the [Graphviz library](https://graphviz.readthedocs.io/en/stable/) to
 from graphviz import Digraph
 
 dot = Digraph()
-for node in response.nodes:
-    dot.node(str(node.id), node.label, shape='circle', width='1', height='1')
-for edge in response.edges:
-    dot.edge(str(edge.source), str(edge.target), label=edge.label)
+for node in response["nodes"]:
+    dot.node(str(node["id"]), node["label"], shape='circle', width='1', height='1')
+for edge in response["edges"]:
+    dot.edge(str(edge["source"]), str(edge["target"]), label=edge["label"])
 
 dot.render('knowledge-graph.gv', view=True)
 ```
diff --git a/docs/cookbook/models_playing_chess.md b/docs/cookbook/models_playing_chess.md
index a0df2eb2d..b8d89958f 100644
--- a/docs/cookbook/models_playing_chess.md
+++ b/docs/cookbook/models_playing_chess.md
@@ -1,6 +1,6 @@
 # Large language models playing chess
 
-In this example we will make a Phi-2 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
+In this example we will make a Phi-3 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
 
 ## The chessboard
 
@@ -18,29 +18,33 @@ board = chess.Board("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
 
 ## The opponents
 
-Phi-2 will be playing against itself:
+Phi-3 will be playing against itself:
 
 ```python
-from outlines import models
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
+import transformers
+import outlines
+
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
 ```
 
 ## A little help for the language model
 
-To make sure Phi-2 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
+To make sure Phi-3 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
 
 ```python
 import re
+from outlines.types.dsl import either, String
 
 def legal_moves_regex(board):
     """Build a regex that only matches valid moves."""
     legal_moves = list(board.legal_moves)
     legal_modes_str = [board.san(move) for move in legal_moves]
     legal_modes_str = [re.sub(r"[+#]", "", move) for move in legal_modes_str]
-    regex_pattern = "|".join(re.escape(move) for move in legal_modes_str)
-    regex_pattern = f"{regex_pattern}"
+    regex_pattern = either(*[String(move) for move in legal_modes_str])
     return regex_pattern
 ```
 
@@ -58,13 +62,11 @@ We update the prompt at each step so it reflects the state of the board after th
 ## Let's play
 
 ```python
-from outlines import generate
-
 board_state = " "
 turn_number = 0
 while not board.is_game_over():
     regex_pattern = legal_moves_regex(board)
-    structured = generate.regex(model, regex_pattern)(prompt + board_state)
+    structured = model(prompt + board_state, regex_pattern)
     move = board.parse_san(structured)
 
     if turn_number % 2 == 0 :  # It's White's turn
@@ -79,7 +81,7 @@ while not board.is_game_over():
     print(board_state)
 ```
 
-Interestingly enough, Phi-2 hates capturing.
+Interestingly enough, Phi-3 hates capturing.
 
 ```pgn
  e4 e5 1.Nf3 Ne7 3.b4 Nf5 5.Nc3 Ne7 7.Bb5 a6 9.Na4 b6 11.c3 Nec6 13.c4 a5 15.d4 Qg5 17.Nd2 Bb7 19.dxe5
diff --git a/docs/cookbook/prompt_templates/chain_of_density.txt b/docs/cookbook/prompt_templates/chain_of_density.txt
new file mode 100644
index 000000000..c4f9f55fa
--- /dev/null
+++ b/docs/cookbook/prompt_templates/chain_of_density.txt
@@ -0,0 +1,27 @@
+Article: {{ article }}
+
+You will generate increasingly concise, entity-dense summaries of the above Article.
+
+Repeat the following 2 steps 5 times.
+
+Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
+Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
+
+A Missing Entity is:
+- Relevant: to the main story.
+- Specific: descriptive yet concise (5 words or fewer).
+- Novel: not in the previous summary.
+- Faithful: present in the Article.
+- Anywhere: located anywhere in the Article.
+
+Guidelines:
+- The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
+- Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
+- Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
+- The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
+- Missing entities can appear anywhere in the new summary.
+- Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
+
+Remember, use the exact same number of words for each summary.
+
+Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
diff --git a/docs/cookbook/prompt_templates/classification.txt b/docs/cookbook/prompt_templates/classification.txt
new file mode 100644
index 000000000..e260a6771
--- /dev/null
+++ b/docs/cookbook/prompt_templates/classification.txt
@@ -0,0 +1,18 @@
+You are an experienced customer success manager.
+
+Given a request from a client, you need to determine when the
+request is urgent using the label "URGENT" or when it can wait
+a little with the label "STANDARD".
+
+# Examples
+
+Request: "How are you?"
+Label: STANDARD
+
+Request: "I need this fixed immediately!"
+Label: URGENT
+
+# TASK
+
+Request: {{ request }}
+Label:
diff --git a/docs/cookbook/prompt_templates/react_agent.txt b/docs/cookbook/prompt_templates/react_agent.txt
new file mode 100644
index 000000000..09583f42f
--- /dev/null
+++ b/docs/cookbook/prompt_templates/react_agent.txt
@@ -0,0 +1,27 @@
+<|im_start|>system
+You are a world class AI model who answers questions in JSON with correct Pydantic schema.
+Here's the json schema you must adhere to:
+<schema>
+{{ schema }}
+</schema>
+Today is {{ today }}
+You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation.
+At the end of the loop you output a Final Answer.
+Use Scratchpad to store the information from the Observation useful to answer the question
+Use Thought to describe your thoughts about the question you have been asked and reflect carefully about the Observation if it exists.
+Use Action to run one of the actions available to you.
+Use Action Input to input the arguments of the selected action - then return PAUSE.
+Observation will be the result of running those actions.
+Your available actions are:
+calculate:
+e.g. calulate: 4**2 / 3
+Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary
+wikipedia:
+e.g. wikipedia: Django
+Returns a summary from searching Wikipedia
+DO NOT TRY TO GUESS THE ANSWER. Begin!
+<|im_end|>
+<|im_start|>user
+{{ question }}
+<|im_end|>
+<|im_start|>assistant
diff --git a/docs/cookbook/prompt_templates/simtom_prospective_taking.txt b/docs/cookbook/prompt_templates/simtom_prospective_taking.txt
new file mode 100644
index 000000000..8e937ce60
--- /dev/null
+++ b/docs/cookbook/prompt_templates/simtom_prospective_taking.txt
@@ -0,0 +1,10 @@
+<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
+Your job is to output only the events that the specified character, {{character}}, knows about.
+
+Here are a few rules:
+1. A character knows about all events that they do.
+2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
+3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
+
+Story: {{story}}
+What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]
diff --git a/docs/cookbook/prompt_templates/simtom_simulation.txt b/docs/cookbook/prompt_templates/simtom_simulation.txt
new file mode 100644
index 000000000..ede934783
--- /dev/null
+++ b/docs/cookbook/prompt_templates/simtom_simulation.txt
@@ -0,0 +1,7 @@
+<s>[INST] {% for event in events %}
+{{event}}
+{% endfor %}
+You are {{name}}.
+Based on the above information, answer the following question:
+{{question}}
+You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]
diff --git a/docs/cookbook/qa-with-citations.md b/docs/cookbook/qa-with-citations.md
index 79a2214c3..213942d02 100644
--- a/docs/cookbook/qa-with-citations.md
+++ b/docs/cookbook/qa-with-citations.md
@@ -11,17 +11,19 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -35,20 +37,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Generate Synthetic Data
@@ -74,26 +65,28 @@ class Users(BaseModel):
     users: List[UserDetail]
 ```
 
-We can use a `generate.json` by passing this Pydantic class we just defined, and call the generator:
+We can use a `outlines.Generator` by passing this Pydantic class we just defined, and call the generator:
 
 ```python
-model = models.LlamaCpp(llm)
-generator = generate.json(model, Users)
+import json
+
+generator = outlines.Generator(model, Users)
 response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
-print(response.users)
-# [UserDetail(id=1, first_name='John', last_name='Doe', age=25),
-# UserDetail(id=2, first_name='Jane', last_name='Doe', age=30),
-# UserDetail(id=3, first_name='Bob', last_name='Smith', age=40),
-# UserDetail(id=4, first_name='Alice', last_name='Smith', age=35),
-# UserDetail(id=5, first_name='John', last_name='Smith', age=20)]
+response = json.loads(response)
+print(response['users'])
+# [{'id': 1, 'first_name': 'John', 'last_name': 'Doe', 'age': 25},
+# {'id': 2, 'first_name': 'Jane', 'last_name': 'Doe', 'age': 30},
+# {'id': 3, 'first_name': 'Bob', 'last_name': 'Smith', 'age': 40},
+# {'id': 4, 'first_name': 'Alice', 'last_name': 'Smith', 'age': 35},
+# {'id': 5, 'first_name': 'John', 'last_name': 'Smith', 'age': 20}]
 ```
 
 ```python
-for user in response.users:
-    print(user.first_name)
-    print(user.last_name)
-    print(user.age)
-    print(#####)
+for user in response['users']:
+    print(user['first_name'])
+    print(user['last_name'])
+    print(user['age'])
+    print("#####")
 # John
 # Doe
 # 25
@@ -135,23 +128,30 @@ schema = QuestionAnswer.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(question, context, schema=schema):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON with correct and exact citations "
-        "extracted from the `Context`. "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + "`Context`: "
-        + context
-        + "\n`Question`: "
-        + question + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON with correct and exact citations
+    extracted from the `Context`.
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    `Context`:
+    {{ context }}
+    `Question`:
+    {{ question }}
+    <|im_end|>
+    <|im_start|>assistant
+    """
+)
 ```
 
-We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
+We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
 
 ```python
 question = "What did the author do during college?"
@@ -161,11 +161,11 @@ I went to an arts high school but in university I studied Computational Mathemat
 As part of coop I worked at many companies including Stitchfix, Facebook.
 I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.
 """
-generator = generate.json(model, QuestionAnswer)
-prompt = generate_hermes_prompt(question, context)
+generator = outlines.Generator(model, QuestionAnswer)
+prompt = hermes_prompt(question=question, context=context, schema=schema)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 print(response)
-# QuestionAnswer(question='What did the author do during college?', answer='The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
+# {"question": "What did the author do during college?", "answer": "The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.", "citations": ["I went to an arts high school but in university I studied Computational Mathematics and physics.", "I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years."]}
 ```
 
 We can do the same for a list of question-context pairs:
@@ -219,12 +219,13 @@ for question, context in [
     (question4, context4),
     (question5, context5),
 ]:
-    final_prompt = my_final_prompt(question, context)
-    generator = generate.json(model, QuestionAnswer)
-    response = generator(final_prompt, max_tokens=1024, temperature=0, seed=42)
-    display(question)
-    display(response.answer)
-    display(response.citations)
+    prompt = hermes_prompt(question=question, context=context, schema=schema)
+    generator = outlines.Generator(model, QuestionAnswer)
+    response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+    response = json.loads(response)
+    print(question)
+    print(response['answer'])
+    print(response['citations'])
     print("\n\n")
 
 # 'Where was John born?'
diff --git a/docs/cookbook/react_agent.md b/docs/cookbook/react_agent.md
index 15b6ed369..bd3d51d6c 100644
--- a/docs/cookbook/react_agent.md
+++ b/docs/cookbook/react_agent.md
@@ -15,17 +15,19 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -39,20 +41,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Build a ReAct agent
@@ -115,51 +106,16 @@ from typing import Union
 
 class Decision(BaseModel):
     Decision: Union[Reason_and_Act, Final_Answer]
-```
-
-We could generate a response using the json schema but we will use the regex and check that everything is working as expected:
-
-```python
-from outlines.fsm.json_schema import convert_json_schema_to_str
-from outlines_core.fsm.json_schema import build_regex_from_schema
 
 json_schema = Decision.model_json_schema()
-schema_str = convert_json_schema_to_str(json_schema=json_schema)
-regex_str = build_regex_from_schema(schema_str)
-print(regex_str)
-# '\\{[ ]?"Decision"[ ]?:[ ]?(\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Thought"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Action"[ ]?:[ ]?("wikipedia"|"calculate")[ ]?,[ ]?"Action_Input"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\}|\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Final_Answer"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\})[ ]?\\}'
 ```
 
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic:
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic. We can load a template from a file for that:
 
 ```python
-import datetime
-
-def generate_hermes_prompt(question, schema=""):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON with correct Pydantic schema. "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema>\n"
-        "Today is " + datetime.datetime.today().strftime('%Y-%m-%d') + ".\n" +
-        "You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation. "
-        "At the end of the loop you output a Final Answer. "
-        "Use Scratchpad to store the information from the Observation useful to answer the question "
-        "Use Thought to describe your thoughts about the question you have been asked "
-        "and reflect carefully about the Observation if it exists. "
-        "Use Action to run one of the actions available to you. "
-        "Use Action Input to input the arguments of the selected action - then return PAUSE. "
-        "Observation will be the result of running those actions. "
-        "Your available actions are:\n"
-        "calculate:\n"
-        "e.g. calulate: 4**2 / 3\n"
-        "Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary\n"
-        "wikipedia:\n"
-        "e.g. wikipedia: Django\n"
-        "Returns a summary from searching Wikipedia\n"
-        "DO NOT TRY TO GUESS THE ANSWER. Begin! <|im_end|>"
-        "\n<|im_start|>user\n" + question + "<|im_end|>"
-        "\n<|im_start|>assistant\n"
-    )
+from outlines import Template
+
+hermes_prompt = Template.from_file("prompt_templates/react_agent.txt")
 ```
 
 We define a ChatBot class
@@ -175,7 +131,7 @@ class ChatBot:
         return result
 
     def execute(self):
-        generator = generate.regex(model, regex_str)
+        generator = outlines.Generator(model, Decision)
         result = generator(self.prompt, max_tokens=1024, temperature=0, seed=42)
         return result
 ```
@@ -194,7 +150,11 @@ def query(question, max_turns=5):
     previous_actions = []
     while i < max_turns:
         i += 1
-        prompt = generate_hermes_prompt(question=question, schema=Decision.model_json_schema())
+        prompt = generate_hermes_prompt(
+            question=question,
+            schema=Decision.model_json_schema(),
+            today=datetime.datetime.today().strftime('%Y-%m-%d')
+        )
         bot = ChatBot(prompt=prompt)
         result = bot(next_prompt)
         json_result = json.loads(result)['Decision']
diff --git a/docs/cookbook/read-pdfs.md b/docs/cookbook/read-pdfs.md
index dbe4ccb02..1854aba42 100644
--- a/docs/cookbook/read-pdfs.md
+++ b/docs/cookbook/read-pdfs.md
@@ -46,17 +46,19 @@ We've tested this example with [Pixtral 12b](https://huggingface.co/mistral-comm
 To use Pixtral:
 
 ```python
-from transformers import LlavaForConditionalGeneration
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
+processor_class = LlavaProcessor
 ```
 
 To use Qwen-2-VL:
 
 ```python
-from transformers import Qwen2VLForConditionalGeneration
+from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
+processor_class = AutoProcessor
 ```
 
 You can load your model into memory with:
@@ -64,17 +66,12 @@ You can load your model into memory with:
 ```python
 # This loads the model into memory. On your first run,
 # it will have to download the model, which might take a while.
-model = outlines.models.transformers_vision(
-    model_name,
-    model_class=model_class,
-    model_kwargs={
-        "device_map": "auto",
-        "torch_dtype": torch.bfloat16,
-    },
-    processor_kwargs={
-        "device": "auto",
-    },
-)
+model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
+processor_kwargs={"device_map": "cpu"}
+tf_model = model_class.from_pretrained(model_name, **model_kwargs)
+tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+model = outlines.from_transformers(tf_model, tf_processor)
 ```
 
 ## Convert the PDF to images
@@ -159,7 +156,7 @@ The structured output you can extract is exactly the same as everywhere else in
 
 Suppose you wished to go through each page of the PDF, and extract the page description, key takeaways, and page number.
 
-You can do this by defining a JSON schema, and then using `outlines.generate.json` to extract the data.
+You can do this by defining a JSON schema, and then using `outlines.Generator` to extract the data.
 
 First, define the structure you want to extract:
 
@@ -170,7 +167,7 @@ class PageSummary(BaseModel):
     page_number: int
 ```
 
-Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers `AutoProcessor` to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
+Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers processor to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
 
 Images are denoted with `type: "image"`, and text is denoted with `type: "text"`.
 
@@ -191,16 +188,15 @@ messages = [
                 """
             },
 
-            # Don't need to pass in an image, since we do this
-            # when we call the generator function down below.
+            # This a placeholder, the actual image is passed in when
+            # we call the generator function down below.
             {"type": "image", "image": ""},
         ],
     }
 ]
 
 # Convert the messages to the final prompt
-processor = AutoProcessor.from_pretrained(model_name)
-instruction = processor.apply_chat_template(
+prompt = tf_processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -209,10 +205,10 @@ Now we iterate through each image, and extract the structured information:
 
 ```python
 # Page summarizer function
-page_summary_generator = outlines.generate.json(model, PageSummary)
+page_summary_generator = outlines.Generator(model, PageSummary)
 
 for image in images:
-    result = page_summary_generator(instruction, [image])
+    result = page_summary_generator({"text": prompt, "images": image})
     print(result)
 ```
 
@@ -225,19 +221,21 @@ arXiv identifiers are optionally followed by a version number, i.e. `arXiv:YYMM.
 We can use a regular expression to define this patter:
 
 ```python
-paper_regex = r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?'
+from outlines.types import Regex
+
+paper_regex = Regex(r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?')
 ```
 
 We can build an extractor function from the regex:
 
 ```python
-id_extractor = outlines.generate.regex(model, paper_regex)
+id_extractor = outlines.Generator(model, paper_regex)
 ```
 
 Now, we can extract the arxiv paper identifier from the first image:
 
 ```python
-arxiv_instruction = processor.apply_chat_template(
+arxiv_instruction = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -256,7 +254,7 @@ arxiv_instruction = processor.apply_chat_template(
 )
 
 # Extract the arxiv paper identifier
-paper_id = id_extractor(arxiv_instruction, [images[0]])
+paper_id = id_extractor({"text": arxiv_instruction, "images": images[0]})
 ```
 
 As of the time of this writing, the arxiv paper identifier is
@@ -269,7 +267,9 @@ Your version number may be different, but the part before `vX` should match.
 
 ### Categorize the paper into one of several categories
 
-`outlines.generate.choice` allows the model to select one of several options. Suppose we wanted to categorize the paper into being about "language models", "economics", "structured generation", or "other".
+`outlines.Generator` also allows the model to select one of several options by providing a Literal type hint with the categories.
+
+Suppose we wanted to categorize the paper into being about "language models", "cell biology", or "other". We would then define the output type as `Literal["llms", "cell biology", "other"]`.
 
 Let's define a few categories we might be interested in:
 
@@ -284,7 +284,7 @@ categories = [
 Now we can construct the prompt:
 
 ```python
-categorization_instruction = processor.apply_chat_template(
+categorization_instruction = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -310,14 +310,13 @@ categorization_instruction = processor.apply_chat_template(
 Now we can show the model the first page and extract the category:
 
 ```python
+from typing import Literal
+
 # Build the choice extractor
-categorizer = outlines.generate.choice(
-    model,
-    categories
-)
+categorizer = outlines.Generator(model, Literal["llms", "cell biology", "other"])
 
 # Categorize the paper
-category = categorizer(categorization_instruction, [images[0]])
+category = categorizer({"text": categorization_instruction, "images": images[0]})
 print(category)
 ```
 
@@ -332,12 +331,12 @@ llms
 You can provide multiple images to the model by
 
 1. Adding additional image messages
-2. Providing a list of images to the `generate` function
+2. Providing a list of images to the generator
 
 For example, to have two images, you can do:
 
 ```python
-two_image_prompt = processor.apply_chat_template(
+two_image_prompt = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -355,17 +354,9 @@ two_image_prompt = processor.apply_chat_template(
 )
 
 # Pass two images to the model
-generator = outlines.generate.choice(
-    model,
-    ["hot dog", "not hot dog"]
-)
+generator = outlines.Generator(model, Literal["hot dog", "not hot dog"])
 
-result = generator(
-    two_image_prompt,
-
-    # Pass two images to the model
-    [images[0], images[1]]
-)
+result = generator({"text": two_image_prompt, "images": [images[0], images[1]]})
 print(result)
 ```
 
diff --git a/docs/cookbook/receipt-digitization.md b/docs/cookbook/receipt-digitization.md
index 0fa256db7..8ecb07969 100644
--- a/docs/cookbook/receipt-digitization.md
+++ b/docs/cookbook/receipt-digitization.md
@@ -38,18 +38,20 @@ If you want to use Qwen-2-VL, you can do the following:
 
 ```python
 # To use Qwen-2-VL:
-from transformers import Qwen2VLForConditionalGeneration
+from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
+processor_class = AutoProcessor
 ```
 
 If you want to use Pixtral, you can do the following:
 
 ```python
 # To use Pixtral:
-from transformers import LlavaForConditionalGeneration
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
+processor_class = LlavaProcessor
 ```
 
 ## Load the model
@@ -57,17 +59,12 @@ model_class=LlavaForConditionalGeneration
 Load the model into memory:
 
 ```python
-model = outlines.models.transformers_vision(
-    model_name,
-    model_class=model_class,
-    model_kwargs={
-        "device_map": "auto",
-        "torch_dtype": torch.bfloat16,
-    },
-    processor_kwargs={
-        "device": "cuda", # set to "cpu" if you don't have a GPU
-    },
-)
+model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
+processor_kwargs={"device_map": "cuda"}
+tf_model = model_class.from_pretrained(model_name, **model_kwargs)
+tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+model = outlines.from_transformers(tf_model, tf_processor)
 ```
 
 ## Image processing
@@ -166,7 +163,7 @@ class ReceiptSummary(BaseModel):
 
 ## Prepare the prompt
 
-We'll use the `AutoProcessor` to convert the image and the text prompt into a format that the model can understand. Practically,
+We'll use the `tf_processor` to convert the image and the text prompt into a format that the model can understand. Practically,
 this is the code that adds user, system, assistant, and image tokens to the prompt.
 
 ```python
@@ -194,8 +191,7 @@ messages = [
 ]
 
 # Convert the messages to the final prompt
-processor = AutoProcessor.from_pretrained(model_name)
-prompt = processor.apply_chat_template(
+prompt = tf_processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -222,17 +218,13 @@ Return the information in the following JSON schema:
 
 ```python
 # Prepare a function to process receipts
-receipt_summary_generator = outlines.generate.json(
-    model,
-    ReceiptSummary,
-
-    # Greedy sampling is a good idea for numeric
-    # data extraction -- no randomness.
-    sampler=outlines.samplers.greedy()
-)
+receipt_summary_generator = outlines.Generator(model, ReceiptSummary)
 
 # Generate the receipt summary
-result = receipt_summary_generator(prompt, [image])
+result = receipt_summary_generator(
+    {"text": prompt, "images": image},
+    max_new_tokens=1024
+)
 print(result)
 ```
 
@@ -241,32 +233,31 @@ print(result)
 The output should look like this:
 
 ```
-ReceiptSummary(
-    store_name="Trader Joe's",
-    store_address='401 Bay Street, San Francisco, CA 94133',
-    store_number=0,
-    items=[
-        Item(name='BANANA EACH', quantity=7, price_per_unit=0.23, total_price=1.61),
-        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CARAMEL CASHEW', quantity=2, price_per_unit=2.29, total_price=4.58),
-        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='SPINDRIFT ORANGE MANGO 8', quantity=1, price_per_unit=7.49, total_price=7.49),
-        Item(name='Bottle Deposit', quantity=8, price_per_unit=0.05, total_price=0.4),
-        Item(name='MILK ORGANIC GALLON WHOL', quantity=1, price_per_unit=6.79, total_price=6.79),
-        Item(name='CLASSIC GREEK SALAD', quantity=1, price_per_unit=3.49, total_price=3.49),
-        Item(name='COBB SALAD', quantity=1, price_per_unit=5.99, total_price=5.99),
-        Item(name='PEPPER BELL RED XL EACH', quantity=1, price_per_unit=1.29, total_price=1.29),
-        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25),
-        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25)
-    ],
-    tax=0.68,
-    total=41.98,
-    date='2023-11-04',
-    payment_method='debit',
-
-)
+{
+  "store_name": "Trader Joe's",
+  "store_address": "401 Bay Street, San Francisco, CA 94133",
+  "store_number": 0,
+  "items": [
+    {"name": "BANANA EACH", "quantity": 7, "price_per_unit": 0.23, "total_price": 1.61},
+    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CARAMEL CASHEW", "quantity": 2, "price_per_unit": 2.29, "total_price": 4.58},
+    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "SPINDRIFT ORANGE MANGO 8", "quantity": 1, "price_per_unit": 7.49, "total_price": 7.49},
+    {"name": "Bottle Deposit", "quantity": 8, "price_per_unit": 0.05, "total_price": 0.4},
+    {"name": "MILK ORGANIC GALLON WHOL", "quantity": 1,"price_per_unit": 6.79,"total_price": 6.79},
+    {"name": "CLASSIC GREEK SALAD", "quantity": 1, "price_per_unit": 3.49, "total_price": 3.49},
+    {"name": "COBB SALAD", "quantity": 1, "price_per_unit": 5.99, "total_price": 5.99},
+    {"name": "PEPPER BELL RED XL EACH", "quantity": 1, "price_per_unit": 1.29, "total_price": 1.29},
+    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
+    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
+  ],
+  "tax": 0.68,
+  "total": 41.98,
+  "date": "2023-11-04",
+  "payment_method": "debit"
+}
 ```
 
 Voila! You've successfully extracted information from a receipt using an LLM.
@@ -284,13 +275,13 @@ class ReceiptSummary(BaseModel):
 which gives you a result like
 
 ```
-ReceiptSummary(
+{
     ...
-    roast="You must be a fan of Trader Joe's because you bought enough
+    "roast": "You must be a fan of Trader Joe's because you bought enough
     items to fill a small grocery bag and still had to pay for a bag fee.
     Maybe you should start using reusable bags to save some money and the
     environment."
-)
+}
 ```
 
 Qwen is not particularly funny, but worth a shot.
diff --git a/docs/cookbook/simtom.md b/docs/cookbook/simtom.md
index a80ebc24f..53820258e 100644
--- a/docs/cookbook/simtom.md
+++ b/docs/cookbook/simtom.md
@@ -17,7 +17,7 @@ SimToM calls an LLM with two consecutive prompts:
 
 To implement SimToM with Outlines, we will need to:
 
-1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
+1. Write the prompts with [prompt templates](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
 2. Define the JSON object each prompt will return using Pydantic.
 3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/latest/reference/models/transformers/).
 
@@ -30,29 +30,8 @@ The authors have shared their code, prompts and data in [this GitHub repository]
 ```python
 from outlines import Template
 
-
-perspective_taking = Template.from_string(
-    """<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
-    Your job is to output only the events that the specified character, {{character}}, knows about.
-
-    Here are a few rules:
-    1. A character knows about all events that they do.
-    2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
-    3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
-
-    Story: {{story}}
-    What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]""" # noqa
-)
-
-simulation = Template.from_string(
-    """<s>[INST] {% for event in events %}
-    {{event}}
-    {% endfor %}
-    You are {{name}}.
-    Based on the above information, answer the following question:
-    {{question}}
-    You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]""" # noqa
-)
+perspective_taking = Template.from_file("prompt_templates/simtom_prospective_taking.txt")
+simulation = Template.from_file("prompt_templates/simtom_simulation.txt")
 ```
 
 ### JSON Structured Generation
@@ -65,13 +44,11 @@ We will need two Pydantic models for SimToM, one for each prompt:
 from pydantic import BaseModel, Field
 from typing import List
 
-
 class PerspectiveTaking(BaseModel):
     """This is for the first prompt."""
     character: str = Field(description="The character we extract the events for.")
     events: List[str] = Field(description="All events that the character knows about.")
 
-
 class Simulation(BaseModel):
     """This is for the second prompt."""
     answer: str
@@ -97,30 +74,37 @@ character = "Aria"
 We load `Mistral-7B-Instruct-v0.3`, create the prompt using the template we defined earlier, and generate a structured response. As a reminder, the goal of the first call is to get all the events a character, `Aria`, knows about.
 
 ```python
+import transformers
+import outlines
 # Load an LLM from Hugging Face
 MODEL_NAME = "mistral-community/Mistral-7B-Instruct-v0.3"
-model = outlines.models.transformers(MODEL_NAME, device="cuda")
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME),
+)
 
 perspective_prompt = perspective_taking(story=story, character=character)
 
 # Call Mistral 7B with the first prompt
-generator = outlines.generate.json(model, PerspectiveTaking)
-perspective = generator(perspective_prompt)
+generator = outlines.Generator(model, PerspectiveTaking)
+perspective = generator(perspective_prompt, max_new_tokens=1024)
 
-print(perspective.model_dump())
+print(perspective)
 # {'character': 'Aria', 'events': ['1 Aria entered the front_yard.', '3 The grapefruit is in the green_bucket.', '4 Aria moved the grapefruit to the blue_container.']}
 ```
 
 Not bad! We will now generate the second prompt with those events.
 
 ```python
-sim_prompt = simulation(events=perspective.events, name=character, question=question)
+import json
+
+sim_prompt = simulation(events=json.loads(perspective)["events"], name=character, question=question)
 
 # Call Mistral 7B with the second prompt
-generator = outlines.generate.json(model, Simulation)
-result = generator(sim_prompt)
+generator = outlines.Generator(model, Simulation)
+result = generator(sim_prompt, max_new_tokens=1024)
 
-print(result.model_dump())
+print(result)
 # {'answer': 'green_bucket'}
 ```
 
diff --git a/docs/cookbook/structured_generation_workflow.md b/docs/cookbook/structured_generation_workflow.md
index b9d7c28a4..524283f84 100644
--- a/docs/cookbook/structured_generation_workflow.md
+++ b/docs/cookbook/structured_generation_workflow.md
@@ -2,8 +2,7 @@
 
 This is a condensed version of [Coding for Structured Generation with LLMs](https://blog.dottxt.co/coding-for-structured-generation.html).
 
-For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this
-with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
+For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
 
 ## Unstructured approach
 
@@ -11,36 +10,30 @@ Before diving into how to use structure generation for this task let's start wit
 
 ```python
 import outlines
+import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.models.transformers(model_name)
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name)
+)
 ```
 
 Next we need a prompt for this model. Since we're focusing on structured generation, we won't be engaging in any form of "prompt hacking" and will be leaving this prompt untouched for the rest of this example.
 
 ```python
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-
-messages_phone = [
-            {"role": "user", "content": """
-            Please generate a realistic phone number for Washington State in the following format
-
-            (555) 555-5555
-
-            """}
-]
-
-# This allows us to properly format our prompt for
-# Phi-3 Mini's 'Instruct' interface.
-prompt_phone = tokenizer.apply_chat_template(messages_phone, tokenize=False)
+prompt_phone = """
+    Please generate a realistic phone number for Washington State in the following format
+    (555) 555-5555
+"""
 ```
 
 With our prompt ready we can now generate 10 example phone numbers
 
 ```python
-phone_generator_unstruct = outlines.generate.text(model)
-for _ in range(10):
-    print(phone_generator_unstruct(prompt_phone,max_tokens=12))
+phone_generator_unstruct = outlines.Generator(model)
+for _ in range(3):
+    print(phone_generator_unstruct(prompt_phone, max_new_tokens=12))
 ```
 
 > I'd be happy to help you generate a realistic phone\
@@ -81,7 +74,9 @@ For a simple example like this, we'll just be using a single phone number, for m
 The next step in the process is for use to define a simple regex that we feel correctly models our real data.
 
 ```python
-phone_regex_1 = r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}'
+from outlines.types import Regex
+
+phone_regex_1 = Regex(r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}')
 ```
 
 Next we need to validate this regex against our real data.
@@ -94,8 +89,8 @@ We'll start with a simple method of validation: just checking that our regex mat
 
 ```
 import re
-re.match(phone_regex_1, phone_number)
 
+re.match(phone_regex_1.pattern, phone_number)
 # <re.Match object; span=(0, 14), match='(206) 386-4636'>
 
 ```
@@ -107,8 +102,9 @@ Now that we have a match, we can move on to generating structured output!
 We're ready to see if structured generation can make an improvement over our initial unstructured approach:
 
 ```python
-phone_generator_v1 = outlines.generate.regex(model, phone_regex_1)
-for _ in range(10):
+phone_generator_v1 = outlines.Generator(model, phone_regex_1)
+
+for _ in range(3):
     print(phone_generator_v1(prompt_phone))
 ```
 > (206) 555-1234\
@@ -135,21 +131,21 @@ We've walked through the loop once, so we can go quickly now through each iterat
 We start by improving our structure:
 
 ```python
-phone_regex_2 = r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}'
+phone_regex_2 = Regex(r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}')
 ```
 
 Before rushing to another round of generation, let's validate this new regex. We'll add just a bit more sophistication over our last check:
 
 ```python
-re.match(phone_regex_2, phone_number)[0] == phone_number
+re.match(phone_regex_2.pattern, phone_number)[0] == phone_number
 # True
 ```
 Now that we've validated, let's generate with this new regex!
 
 ```python
-phone_generator_v2 = outlines.generate.regex(model,
-                                             phone_regex_2)
-for _ in range(10):
+phone_generator_v2 = outlines.Generator(model, phone_regex_2)
+
+for _ in range(3):
     print(phone_generator_v2(prompt_phone))
 ```
 
@@ -192,10 +188,10 @@ This prints out:
 Ah! We were missing the last digit, let's fix that and regenerate:
 
 ```python
-phone_regex_3_fixed = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]'
-phone_generator_v3 = outlines.generate.regex(model,
-                                             phone_regex_3_fixed)
-for _ in range(10):
+phone_regex_3_fixed = Regex(r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]')
+phone_generator_v3 = outlines.Generator(model, phone_regex_3_fixed)
+
+for _ in range(3):
     print(phone_generator_v3(prompt_phone))
 ```
 

From 7669cc2eccf25e3a453f66bade2baef6f697c5d0 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 14 Apr 2025 09:25:29 +0200
Subject: [PATCH 311/505] Replace parse_raw by model_validate_json

---
 outlines/v0_legacy/generate/json.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
index 3c71ed8d0..0e6ed8d46 100644
--- a/outlines/v0_legacy/generate/json.py
+++ b/outlines/v0_legacy/generate/json.py
@@ -94,7 +94,7 @@ def json(
         setattr(
             generator,
             'format_sequence',
-            lambda x: schema_object.parse_raw(x)
+            lambda x: schema_object.model_validate_json(x)
         )
     else:
         setattr(generator, 'format_sequence', lambda x: pyjson.loads(x))
@@ -117,7 +117,7 @@ def json_openai(
         schema = schema_object.model_json_schema()
         schema["additionalProperties"] = False
         schema = pyjson.dumps(schema)
-        format_sequence = lambda x: schema_object.parse_raw(x)
+        format_sequence = lambda x: schema_object.model_validate_json(x)
     elif isinstance(schema_object, str):
         schema = schema_object
         format_sequence = lambda x: pyjson.loads(x)

From d43129f1bd8a4b55e69ac51dad4f157ea93e8afe Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 14 Apr 2025 16:15:17 +0200
Subject: [PATCH 312/505] Fix errors in the mlx tensor adapter

---
 outlines/processors/tensor_adapters/base.py | 24 +++++++++++++++++----
 outlines/processors/tensor_adapters/mlx.py  | 12 +++++++++--
 2 files changed, 30 insertions(+), 6 deletions(-)

diff --git a/outlines/processors/tensor_adapters/base.py b/outlines/processors/tensor_adapters/base.py
index 4ea7b4459..fedcb6924 100644
--- a/outlines/processors/tensor_adapters/base.py
+++ b/outlines/processors/tensor_adapters/base.py
@@ -1,4 +1,8 @@
 from abc import ABC, abstractmethod
+from typing import Any, Union, TYPE_CHECKING
+
+if TYPE_CHECKING:
+    import torch
 
 
 class TensorAdapter(ABC):
@@ -8,6 +12,11 @@ class TensorAdapter(ABC):
     manipulate tensors in different libraries. Concrete implementations of
     this class should provide specific implementations for each method as
     well as providing a library_name attribute.
+
+    TODO: Update the version of outlines-core used to receive plain arrays
+    instead of torch tensors. In the meantime, implementations of this class
+    must make sure that their `full_like` and `concatenate` methods can
+    handle torch tensors.
     """
     library_name: str
 
@@ -37,13 +46,20 @@ def to_scalar(self, tensor):
         ...
 
     @abstractmethod
-    def full_like(self, tensor, fill_value):
-        """Create a tensor with the same shape as the input tensor filled with a scalar value"""
+    def full_like(self, tensor: "torch.Tensor", fill_value):
+        """Create a tensor with the same shape as the input tensor filled
+        with a scalar value.
+        ATTENTION: This method receives a torch tensor regardless of the
+        library used.
+        """
         ...
 
     @abstractmethod
-    def concatenate(self, tensors):
-        """Concatenate a list of tensors along a specified dimension"""
+    def concatenate(self, tensors: list[Union["torch.Tensor", Any]]):
+        """Concatenate a list of tensors along a specified dimension.
+        ATTENTION: This method can either receive a list of torch tensors or
+        a list of tensors from the library used.
+        """
         ...
 
     @abstractmethod
diff --git a/outlines/processors/tensor_adapters/mlx.py b/outlines/processors/tensor_adapters/mlx.py
index 47811e183..0e9de15ef 100644
--- a/outlines/processors/tensor_adapters/mlx.py
+++ b/outlines/processors/tensor_adapters/mlx.py
@@ -27,10 +27,18 @@ def to_scalar(self, tensor):
         return tensor.item()
 
     def full_like(self, tensor, fill_value):
-        return self.mlx.full(tensor.shape, fill_value, dtype=tensor.dtype)
+        # Compatible with receiving a torch tensor
+        return self.mlx.full(tensor.shape, fill_value)
 
     def concatenate(self, tensors):
-        return self.mlx.concat(tensors, axis=0)
+        # Can handle both torch and mlx tensors
+        return self.mlx.concatenate(
+            [
+                self.mlx.array(t) if not isinstance(t, self.mlx.array) else t
+                for t in tensors
+            ],
+            axis=0
+        )
 
     def get_device(self, tensor):
         return None

From 3a3c28609c75135caf1ab41ae73b5e0d11bd6dfb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 12 Apr 2025 16:52:41 +0200
Subject: [PATCH 313/505] Update the README

---
 README.md                           | 837 ++++++++++++++++------------
 docs/assets/images/install.png      | Bin 0 -> 12669 bytes
 docs/assets/images/readme-dark.png  | Bin 0 -> 25637 bytes
 docs/assets/images/readme-light.png | Bin 0 -> 22109 bytes
 docs/logos/cohere.png               | Bin 0 -> 113536 bytes
 docs/logos/vllm.png                 | Bin 0 -> 89976 bytes
 6 files changed, 488 insertions(+), 349 deletions(-)
 create mode 100644 docs/assets/images/install.png
 create mode 100644 docs/assets/images/readme-dark.png
 create mode 100644 docs/assets/images/readme-light.png
 create mode 100644 docs/logos/cohere.png
 create mode 100644 docs/logos/vllm.png

diff --git a/README.md b/README.md
index 0e84ecc4e..af7380b0c 100644
--- a/README.md
+++ b/README.md
@@ -4,523 +4,662 @@
 <img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=500></img>
 
 
- 🗒️ *Make LLMs speak the language of every application.* 🗒️
+ 🗒️ *Structured outputs for LLMs* 🗒️
 
-Made with ❤👷️ by the team at [.txt](https://dottxt.co).
 
-[![Documentation][documentation-badge]][documentation]
-[![Contributors][contributors-badge]][contributors]
-[![Downloads][downloads-badge]][pypistats]
-[![Discord][discord-badge]][discord]
+Made with ❤👷️ by the team at [.txt](https://dottxt.co)
+<br>Trusted by NVIDIA, Cohere, HuggingFace, vLLM, etc.
 
-[Youtube channel][youtube-dottxt] | [.txt blog][blog-dottxt] | [Twitter][dottxt-twitter]
+<!-- Project Badges -->
+[![PyPI Version][pypi-version-badge]][pypi]
+[![Downloads][downloads-badge]][pypistats]
+[![Stars][stars-badge]][stars]
 
+<!-- Community Badges -->
+[![Discord][discord-badge]][discord]
+[![Blog][dottxt-blog-badge]][dottxt-blog]
+[![Twitter][twitter-badge]][twitter]
 
 </div>
 
+## Table of Contents
 
-``` bash
-pip install outlines
-```
+- [Why Outlines?](#why-outlines)
+- [Quickstart](#quickstart)
+- [Real-World Examples](#real-world-examples)
+  - [🙋‍♂️ Customer Support Triage](#customer-support-triage)
+  - [📦 E-commerce Product Categorization](#e-commerce-product-categorization)
+  - [💊 Pull Medication Data from Unstructured Notes](#pull-medication-data-from-unstructured-notes)
+  - [📊 Parse Event Details with Incomplete Data](#parse-event-details-with-incomplete-data)
+  - [🗂️ Categorize Documents into Predefined Types](#categorize-documents-into-predefined-types)
+  - [📅 Schedule a Meeting with Function Calling](#schedule-a-meeting-with-function-calling)
+  - [📝 Dynamically Generate Prompts with Re-usable Templates](#dynamically-generate-prompts-with-re-usable-templates)
+- [They Use Outlines](#they-use-outlines)
+- [Model Integrations](#model-integrations)
+- [Core Features](#core-features)
+- [Other Features](#other-features)
+- [About .txt](#about-txt)
+- [Community](#community)
 
-First time here? Go to our [setup guide](https://dottxt-ai.github.io/outlines/latest/welcome/)/
+<div align="center"><img src="./docs/assets/images/install.png" width=300></img></div>
 
-Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales) with us.
+## Why Outlines?
 
-## Features
+LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
 
-- 🤖 [Multiple model integrations](https://dottxt-ai.github.io/outlines/latest/installation): OpenAI, transformers, llama.cpp, mamba
-- 🔥 Fast [JSON generation](#efficient-json-generation-following-a-pydantic-model) following a JSON schema or a Pydantic model
-- 🚄 [Multiple choices](#multiple-choices), [type constraints](#type-constraint) and dynamic stopping
-- 📝 Generate text that follows a [regex](#efficient-regex-structured-generation) or a [context-free grammar](#using-context-free-grammars-to-guide-generation)
-- 🖍️ Simple and powerful prompting primitives based on the [Jinja templating engine](https://jinja.palletsprojects.com/)
-- 🚀 [Serve with vLLM](https://dottxt-ai.github.io/outlines/latest/reference/serve/vllm), with official Docker image, [`outlinesdev/outlines`](https://hub.docker.com/r/outlinesdev/outlines)!
+Outlines guarantees structured outputs during generation — directly from any LLM.
 
+- **Works with any model** - Same code runs across OpenAI, Ollama, vLLM, and more
+- **Simple integration** - Just pass your desired output type: `model(prompt, output_type)`
+- **Guaranteed valid structure** - No more parsing headaches or broken JSON
+- **Provider independence** - Switch models without changing code
 
-Outlines  has new releases and features coming every week. Make sure to ⭐ star and 👀 watch this repository, follow [@dottxtai][dottxt-twitter] to stay up to date!
 
-## Why should I use structured generation?
+### The Outlines Philosophy
 
-* It doesn't add any overhead during inference (cost-free)
-* It allows Open Source models to beat closed source models ([Mistral](https://x.com/dottxtai/status/1797692104023363765), [GPT-4](https://x.com/dottxtai/status/1798443290913853770))
-* [It speeds up inference](http://blog.dottxt.co/coalescence.html)
-* [It improves the performance of base models (GSM8K)](http://blog.dottxt.co/performance-gsm8k.html)
-* [It improves the performance of finetuned models (CoNNL)](https://predibase.com/blog/lorax-outlines-better-json-extraction-with-structured-generation-and-lora)
-* [It improves model efficiency (less examples needed)](https://huggingface.co/blog/evaluation-structured-outputs)
+Outlines follows a simple pattern that mirrors Python's own type system. Simply specify the desired output type, and Outlines will ensure your data matches that structure exactly:
 
-## .txt company
+- For a yes/no response, use `Literal["Yes", "No"]`
+- For numerical values, use `int`
+- For complex objects, define a structure with a Pydantic model
 
-<div align="center">
-<img src="./docs/assets/images/dottxt-light.svg#gh-light-mode-only" alt="dottxt logo" width=100></img>
-<img src="./docs/assets/images/dottxt-dark.svg#gh-dark-mode-only" alt="dottxt logo" width=100></img>
-</div>
+## Quickstart
 
-We started a company to keep pushing the boundaries of structured generation. Learn more about [.txt](https://dottxt.ai), and  [license our commercial product](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF) if you need SOTA performance and quality ✨
+Getting started with outlines is simple:
 
-## Structured generation
+### 1. Install outlines
 
-The first step towards reliability of systems that include large language models
-is to ensure that there is a well-defined interface between their output and
-user-defined code. **Outlines** provides ways to control the generation of
-language models to make their output more predictable.
-
-The following methods of structured generation are supported:
-
-- [Multiple choices](#multiple-choices)
-- [Type constraints](#type-constraint)
-- [Efficient regex-structured generation](#efficient-regex-structured-generation)
-- [Efficient JSON generation following a Pydantic model](#efficient-json-generation-following-a-pydantic-model)
-- [Using context-free grammars to guide generation](#using-context-free-grammars-to-guide-generation)
-- [Open functions](#open-functions)
+``` shell
+pip install outlines
+```
 
-### Chat template tokens
+### 2. Connect to your preferred model
 
-Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself. Chat template tokens are not needed for base models.
+``` python
+import outlines
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-Please see [the documentation](https://dottxt-ai.github.io/outlines/latest/reference/chat_templating) on chat templating for more.
 
-### Multiple choices
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
+```
 
-You can reduce the completion to a choice between multiple possibilities:
+### 3. Start with simple structured outputs
 
 ``` python
 from typing import Literal
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from pydantic import BaseModel
 
-import outlines
 
-model_name = "HuggingFaceTB/SmolLM2-360M-Instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
+# Simple classification
+sentiment = model(
+    "Analyze: 'This product completely changed my life!'",
+    Literal["Positive", "Negative", "Neutral"]
 )
+print(sentiment)  # "Positive"
 
-# You must apply the chat template tokens to the prompt!
-# See below for an example.
-prompt = """
-<|im_start|>system
-You extract information from text.
-<|im_end|>
-
-<|im_start|>user
-What food does the following text describe?
-
-Text: I really really really want pizza.
-<|im_end|>
-<|im_start|>assistant
-"""
-
-answer = model(prompt, Literal["Pizza", "Pasta", "Salad", "Dessert"])
-# Likely answer: Pizza
+# Extract specific types
+temperature = model("What's the boiling point of water in Celsius?", int)
+print(temperature)  # 100
 ```
 
-You can also pass in choices with an `Enum`:
+### 4. Create complex structures
 
-````python
+``` python
+from pydantic import BaseModel
 from enum import Enum
 
-class Food(str, Enum):
-    pizza = "Pizza"
-    pasta = "Pasta"
-    salad = "Salad"
-    dessert = "Dessert"
+class Rating(Enum):
+    poor = 1
+    fair = 2
+    good = 3
+    excellent = 4
+
+class ProductReview(BaseModel):
+    rating: Rating
+    pros: list[str]
+    cons: list[str]
+    summary: str
+
+review = model(
+    "Review: The XPS 13 has great battery life and a stunning display, but it runs hot and the webcam is poor quality.",
+    ProductReview,
+    max_new_tokens=200,
+)
 
-answer = model(prompt, Food)
-# Likely answer: Pizza
-````
+review = ProductReview.model_validate_json(review)
+print(f"Rating: {review.rating.name}")  # "Rating: good"
+print(f"Pros: {review.pros}")           # "Pros: ['great battery life', 'stunning display']"
+print(f"Summary: {review.summary}")     # "Summary: Good laptop with great display but thermal issues"t(result)
+```
 
-### Type constraints
+## Real-world examples
 
-You can instruct the model to only return integers or floats:
+Here are production-ready examples showing how Outlines solves common problems:
 
+<details id="customer-support-triage"><summary><b>🙋‍♂️ Customer Support Triage</b>
+<br>This example shows how to convert a free-form customer email into a structured service ticket. By parsing attributes like priority, category, and escalation flags, the code enables automated routing and handling of support issues.
+</summary>
 
 ``` python
-from transformers import AutoModelForCausalLM, AutoTokenizer
 import outlines
+from enum import Enum
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from typing import List
 
-model_name = "WizardLM/WizardMath-7B-V1.1"
+
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
 model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
 )
 
-prompt = "<s>result of 9 + 9 = 18</s><s>result of 1 + 2 = "
-answer = outlines.generate.format(model, int)(prompt)
-print(answer)
-# 3
 
-prompt = "sqrt(2)="
-answer = model(prompt, outlines.types.number, max_tokens=10)
-print(answer)
-# 1.41421356
-```
+def alert_manager(ticket):
+    print("Alert!", ticket)
 
-### Efficient regex-structured generation
 
-Outlines also comes with fast regex-structured generation. In fact, the `choice` and
-`format` functions above all use regex-structured generation under the
-hood:
+class TicketPriority(str, Enum):
+    low = "low"
+    medium = "medium"
+    high = "high"
+    urgent = "urgent"
 
-``` python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-import outlines
-from outlines import regex
+class ServiceTicket(BaseModel):
+    priority: TicketPriority
+    category: str
+    requires_manager: bool
+    summary: str
+    action_items: List[str]
 
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
-)
 
-prompt = """
-<|im_start|>system You are a helpful assistant.
-<|im_end|>
+customer_email = """
+Subject: URGENT - Cannot access my account after payment
 
+I paid for the premium plan 3 hours ago and still can't access any features.
+I've tried logging out and back in multiple times. This is unacceptable as I
+have a client presentation in an hour and need the analytics dashboard.
+Please fix this immediately or refund my payment.
+"""
+
+prompt = f"""
 <|im_start|>user
-What is an IP address of the Google DNS servers?
+Analyze this customer email:
+
+{customer_email}
 <|im_end|>
 <|im_start|>assistant
-The IP address of a Google DNS server is
-
 """
 
-unstructured = model(prompt, max_tokens=10)
-
-structured = model(
+ticket = model(
     prompt,
-    outlines.Regex(r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"),
-    max_tokens=30
+    ServiceTicket,
+    max_new_tokens=500
 )
 
-print(unstructured)
-# 8.8.8.8
-#
-# <|im_end|>
-
-print(structured)
-# 8.8.8.8
+# Use structured data to route the ticket
+ticket = ServiceTicket.model_validate_json(ticket)
+if ticket.priority == "urgent" or ticket.requires_manager:
+    alert_manager(ticket)
 ```
+</details>
 
-Unlike other libraries, regex-structured generation in Outlines is almost as fast
-as non-structured generation.
+<details id="e-commerce-product-categorization"><summary><b>📦 E-commerce product categorization</b>
+<br>This use case demonstrates how outlines can transform product descriptions into structured categorization data (e.g., main category, sub-category, and attributes) to streamline tasks such as inventory management. Each product description is processed automatically, reducing manual categorization overhead.
+</summary>
 
-### Efficient JSON generation following a Pydantic model
+```python
+import outlines
+from pydantic import BaseModel
+from transformers import AutoTokenizer, AutoModelForCausalLM
+from typing import List, Optional
 
-Outlines users can guide the generation process so the output is *guaranteed* to follow a [JSON schema](https://json-schema.org/) or [Pydantic model](https://docs.pydantic.dev/latest/):
 
-```python
-from enum import Enum
-from pydantic import BaseModel, constr
-from transformers import AutoModelForCausalLM, AutoTokenizer
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 
-import outlines
 
-class Weapon(str, Enum):
-    sword = "sword"
-    axe = "axe"
-    mace = "mace"
-    spear = "spear"
-    bow = "bow"
-    crossbow = "crossbow"
+def update_inventory(product, category, sub_category):
+    print(f"Updated {product.split(',')[0]} in category {category}/{sub_category}")
 
 
-class Armor(str, Enum):
-    leather = "leather"
-    chainmail = "chainmail"
-    plate = "plate"
+class ProductCategory(BaseModel):
+    main_category: str
+    sub_category: str
+    attributes: List[str]
+    brand_match: Optional[str]
 
+# Process product descriptions in batches
+product_descriptions = [
+    "Apple iPhone 15 Pro Max 256GB Titanium, 6.7-inch Super Retina XDR display with ProMotion",
+    "Organic Cotton T-Shirt, Men's Medium, Navy Blue, 100% Sustainable Materials",
+    "KitchenAid Stand Mixer, 5 Quart, Red, 10-Speed Settings with Dough Hook Attachment"
+]
 
-class Character(BaseModel):
-    name: constr(max_length=10)
-    age: int
-    armor: Armor
-    weapon: Weapon
-    strength: int
+template = outlines.Template.from_string("""
+<|im_start|>user
+Categorize this product:
 
+{{ description }}
+<|im_end|>
+<|im_start|>assistant
+""")
 
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
+# Get structured categorization for all products
+categories = model(
+    [template(description=desc) for desc in product_descriptions],
+    ProductCategory,
+    max_new_tokens=200
 )
 
-# Construct structured sequence generator
-generator = outlines.Generator(model, Character)
+# Use categorization for inventory management
+categories = [
+    ProductCategory.model_validate_json(category) for category in categories
+]
+for product, category in zip(product_descriptions, categories):
+    update_inventory(product, category.main_category, category.sub_category)
+```
+</details>
 
-# Draw a sample
-seed = 789001
-character = generator("Give me a character description", seed=seed)
+<details id="parse-event-details-with-incomplete-data"><summary><b>📊 Parse event details with incomplete data</b>
+<br>This example uses outlines to parse event descriptions into structured information (like event name, date, location, type, and topics), even handling cases where the data is incomplete. It leverages union types to return either structured event data or a fallback “I don’t know” answer, ensuring robust extraction in varying scenarios.
+</summary>
 
-print(repr(character))
-# Character(name='Anderson', age=28, armor=<Armor.chainmail: 'chainmail'>, weapon=<Weapon.sword: 'sword'>, strength=8)
+```python
+import outlines
+from typing import Union, List, Literal
+from pydantic import BaseModel
+from enum import Enum
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-prompt = "Give me an interesting character description"
-character = model(prompt, Character, seed=seed)
 
-print(repr(character))
-# Character(name='Vivian Thr', age=44, armor=<Armor.plate: 'plate'>, weapon=<Weapon.crossbow: 'crossbow'>, strength=125)
-```
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 
-The method works with union types, optional types, arrays, nested schemas, etc. Some field constraints are [not supported yet](https://github.com/dottxt-ai/outlines/issues/215), but everything else should work.
+class EventType(str, Enum):
+    conference = "conference"
+    webinar = "webinar"
+    workshop = "workshop"
+    meetup = "meetup"
+    other = "other"
+
+
+class EventInfo(BaseModel):
+    """Structured information about a tech event"""
+    name: str
+    date: str
+    location: str
+    event_type: EventType
+    topics: List[str]
+    registration_required: bool
+
+# Create a union type that can either be a structured EventInfo or "I don't know"
+EventResponse = Union[EventInfo, Literal["I don't know"]]
+
+# Sample event descriptions
+event_descriptions = [
+    # Complete information
+    """
+    Join us for DevCon 2023, the premier developer conference happening on November 15-17, 2023
+    at the San Francisco Convention Center. Topics include AI/ML, cloud infrastructure, and web3.
+    Registration is required.
+    """,
+
+    # Insufficient information
+    """
+    Tech event next week. More details coming soon!
+    """
+]
 
-### Efficient JSON generation following a JSON Schema
+# Process events
+results = []
+for description in event_descriptions:
+    prompt = f"""
+<|im_start>system
+You are a helpful assistant
+<|im_end|>
+<|im_start>user
+Extract structured information about this tech event:
 
-Sometimes you just want to be able to pass a JSON Schema instead of a Pydantic model. We've got you covered:
+{description}
 
-``` python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+If there is enough information, return a JSON object with the following fields:
+
+- name: The name of the event
+- date: The date where the event is taking place
+- location: Where the event is taking place
+- event_type: either 'conference', 'webinar', 'workshop', 'meetup' or 'other'
+- topics: a list of topics of the conference
+- registration_required: a boolean that indicates whether registration is required
 
+If the information available does not allow you to fill this JSON, and only then, answer 'I don't know'.
+<|im_end|>
+<|im_start|>assistant
+"""
+    # Union type allows the model to return structured data or "I don't know"
+    result = model(prompt, EventResponse, max_new_tokens=200)
+    results.append(result)
+
+# Display results
+for i, result in enumerate(results):
+    print(f"Event {i+1}:")
+    if isinstance(result, str):
+        print(f"  {result}")
+    else:
+        # It's an EventInfo object
+        print(f"  Name: {result.name}")
+        print(f"  Type: {result.event_type}")
+        print(f"  Date: {result.date}")
+        print(f"  Topics: {', '.join(result.topics)}")
+    print()
+
+# Use structured data in downstream processing
+structured_count = sum(1 for r in results if isinstance(r, EventInfo))
+print(f"Successfully extracted data for {structured_count} of {len(results)} events")
+```
+</details>
+
+<details id="categorize-documents-into-predefined-types"><summary><b>🗂️ Categorize documents into predefined types</b>
+<br>In this case, outlines classifies documents into predefined categories (e.g., “Financial Report,” “Legal Contract”) using a literal type specification. The resulting classifications are displayed in both a table format and through a category distribution summary, illustrating how structured outputs can simplify content management.
+</summary>
+
+```python
 import outlines
+from typing import Literal, List
+import pandas as pd
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-schema = '''{
-    "title": "Character",
-    "type": "object",
-    "properties": {
-        "name": {
-            "title": "Name",
-            "maxLength": 10,
-            "type": "string"
-        },
-        "age": {
-            "title": "Age",
-            "type": "integer"
-        },
-        "armor": {"$ref": "#/definitions/Armor"},
-        "weapon": {"$ref": "#/definitions/Weapon"},
-        "strength": {
-            "title": "Strength",
-            "type": "integer"
-        }
-    },
-    "required": ["name", "age", "armor", "weapon", "strength"],
-    "definitions": {
-        "Armor": {
-            "title": "Armor",
-            "description": "An enumeration.",
-            "enum": ["leather", "chainmail", "plate"],
-            "type": "string"
-        },
-        "Weapon": {
-            "title": "Weapon",
-            "description": "An enumeration.",
-            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
-            "type": "string"
-        }
-    }
-}'''
 
-model_name = "microsoft/Phi-3-mini-4k-instruct"
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
 model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
 )
 
-prompt = "Give me a character description"
-character = model(prompt, outlines.json_schema(schema))
-```
 
-### Using context-free grammars to guide generation
+# Define classification categories using Literal
+DocumentCategory = Literal[
+    "Financial Report",
+    "Legal Contract",
+    "Technical Documentation",
+    "Marketing Material",
+    "Personal Correspondence"
+]
 
-Formal grammars rule the world, and Outlines makes them rule LLMs too. You can pass any context-free grammar in the EBNF format and Outlines will generate an output that is valid to this grammar:
+# Sample documents to classify
+documents = [
+    "Q3 Financial Summary: Revenue increased by 15% year-over-year to $12.4M. EBITDA margin improved to 23% compared to 19% in Q3 last year. Operating expenses...",
 
-``` python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+    "This agreement is made between Party A and Party B, hereinafter referred to as 'the Parties', on this day of...",
 
-import outlines
+    "The API accepts POST requests with JSON payloads. Required parameters include 'user_id' and 'transaction_type'. The endpoint returns a 200 status code on success."
+]
 
-arithmetic_grammar = """
-    ?start: expression
+template = outlines.Template.from_string("""
+<|im_start|>user
+Classify the following document into exactly one category among the following categories:
+- Financial Report
+- Legal Contract
+- Technical Documentation
+- Marketing Material
+- Personal Correspondence
+
+Document:
+{{ document }}
+<|im_end|>
+<|im_start|>assistant
+""")
 
-    ?expression: term (("+" | "-") term)*
+# Classify documents
+def classify_documents(texts: List[str]) -> List[DocumentCategory]:
+    results = []
 
-    ?term: factor (("*" | "/") factor)*
+    for text in texts:
+        prompt = template(document=text)
+        # The model must return one of the predefined categories
+        category = model(prompt, DocumentCategory, max_new_tokens=200)
+        results.append(category)
 
-    ?factor: NUMBER
-           | "-" factor
-           | "(" expression ")"
+    return results
 
-    %import common.NUMBER
-"""
+# Perform classification
+classifications = classify_documents(documents)
 
-model_name = "WizardLM/WizardMath-7B-V1.1"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
-)
+# Create a simple results table
+results_df = pd.DataFrame({
+    "Document": [doc[:50] + "..." for doc in documents],
+    "Classification": classifications
+})
 
-prompt = "Alice had 4 apples and Bob ate 2. Write an expression for Alice's apples:"
-sequence = model(prompt, outlines.types.Cfg(arithmetic_grammar))
+print(results_df)
 
-print(sequence)
-# (8-2)
+# Count documents by category
+category_counts = pd.Series(classifications).value_counts()
+print("\nCategory Distribution:")
+print(category_counts)
 ```
+</details>
 
-This was a very simple grammar, and you can use `outlines.generate.cfg` to generate syntactically valid Python, SQL, and much more than this. Any kind of structured text, really. All you have to do is search for "X EBNF grammar" on the web, and take a look at the [Outlines `grammars` module](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
-
-### Open functions
-
-Outlines can infer the structure of the output from the signature of a function. The result is a dictionary, and can be passed directly to the function using the usual dictionary expansion syntax `**`:
+<details>
+<summary id="schedule-a-meeting-with-function-calling"><b>📅 Schedule a meeting from requests with Function Calling</b>
+<br>This example demonstrates how outlines can interpret a natural language meeting request and translate it into a structured format matching a predefined function’s parameters. Once the meeting details are extracted (e.g., title, date, duration, attendees), they are used to automatically schedule the meeting.
+</summary>
 
 ```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
 import outlines
+import json
+from typing import List, Optional
+from datetime import date
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
 
-def add(a: int, b: int):
-    return a + b
-
-model_name = "WizardLM/WizardMath-7B-V1.1"
+MODEL_NAME = "microsoft/phi-4"
 model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
-)
-result = model(
-    "Return json with two integers named a and b respectively. a is odd and b even.",
-    add
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
 )
 
-print(add(**result))
-# 3
-```
 
-A great advantage of passing functions directly to specify the structure is that the structure of the LLM will change with the function's definition. No need to change the code at several places!
+# Define a function with typed parameters
+def schedule_meeting(
+    title: str,
+    date: date,
+    duration_minutes: int,
+    attendees: List[str],
+    location: Optional[str] = None,
+    agenda_items: Optional[List[str]] = None
+):
+    """Schedule a meeting with the specified details"""
+    # In a real app, this would create the meeting
+    meeting = {
+        "title": title,
+        "date": date,
+        "duration_minutes": duration_minutes,
+        "attendees": attendees,
+        "location": location,
+        "agenda_items": agenda_items
+    }
+    return f"Meeting '{title}' scheduled for {date} with {len(attendees)} attendees"
 
-You can also embed various functions into an enum to generate params:
+# Natural language request
+user_request = """
+I need to set up a product roadmap review with the engineering team for next
+Tuesday at 2pm. It should last 90 minutes. Please invite john@example.com,
+sarah@example.com, and the product team at product@example.com.
+"""
 
-```python
-from enum import Enum
-from functools import partial
+# Outlines automatically infers the required structure from the function signature
+prompt = f"""
+<|im_start|>user
+Extract the meeting details from this request:
 
-from transformers import AutoModelForCausalLM, AutoTokenizer
+{user_request}
+<|im_end|>
+<|im_start|>assistant
+"""
+meeting_params = model(prompt, schedule_meeting, max_new_tokens=200)
 
-import outlines
+# The result is a dictionary matching the function parameters
+meeting_params = json.loads(meeting_params)
+print(meeting_params)
 
+# Call the function with the extracted parameters
+result = schedule_meeting(**meeting_params)
+print(result)
+# "Meeting 'Product Roadmap Review' scheduled for 2023-10-17 with 3 attendees"
+```
+</details>
 
-def add(a: int, b: int) -> int:
-    return a + b
+<details>
+<summary id="dynamically-generate-prompts-with-re-usable-templates"><b>📝 Dynamically generate prompts with re-usable templates</b>
+<br>Using Jinja-based templates, this example shows how to generate dynamic prompts for tasks like sentiment analysis. It illustrates how to easily re-use and customize prompts—including few-shot learning strategies—for different content types while ensuring the outputs remain structured.
+</summary>
 
-def mul(c: float, d: float) -> float:
-    return c * d
+```python
+import outlines
+from typing import List, Literal
+from transformers import AutoTokenizer, AutoModelForCausalLM
 
-class Operation(Enum):
-    add = partial(add)
-    mul = partial(mul)
 
-model_name = "WizardLM/WizardMath-7B-V1.1"
+MODEL_NAME = "microsoft/phi-4"
 model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
+    AutoModelForCausalLM.from_pretrained(MODEL_NAME, device_map="auto"),
+    AutoTokenizer.from_pretrained(MODEL_NAME)
 )
-generator = outlines.Generator(model, Operation)
-result = generator("Return json with two float named c and d respectively. c is negative and d greater than 1.0.")
-#print(result)
-# {'c': -3.14, 'd': 1.5}
-```
 
-## Prompting
 
-Building prompts can get messy. **Outlines** makes it easier to write and manage
-prompts by encapsulating templates inside "template functions". Template
-functions use the Jinja2 templating engine to help build complex prompts in a
-concise manner.
+# 1. Create a reusable template with Jinja syntax
+sentiment_template = outlines.Template.from_string("""
+<|im_start>user
+Analyze the sentiment of the following {{ content_type }}:
 
-Template functions are created by loading a Jinja2 template from a text file.
-Assume you have the following prompt template defined in `prompt.txt`:
+{{ text }}
 
-``` text
-You are a sentiment-labelling assistant.
-
-{% for example in examples %}
-{{ example[0] }} // {{ example[1] }}
-{% endfor %}
-{{ to_label }} //
-```
+Provide your analysis as either "Positive", "Negative", or "Neutral".
+<|im_end>
+<|im_start>assistant
+"""
 
-You can then load it and call it with:
+# 2. Generate prompts with different parameters
+review = "This restaurant exceeded all my expectations. Fantastic service!"
+prompt = sentiment_template(content_type="review", text=review)
 
-``` python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+# 3. Use the templated prompt with structured generation
+result = model(prompt, Literal["Positive", "Negative", "Neutral"])
+print(result)  # "Positive"
 
-import outlines
+# Templates can also be loaded from files
+example_template = outlines.Template.from_file("templates/few_shot.txt")
 
+# Use with examples for few-shot learning
 examples = [
-    ("The food was disgusting", "Negative"),
-    ("We had a fantastic night", "Positive"),
-    ("Recommended", "Positive"),
-    ("The waiter was rude", "Negative")
+    ("The food was cold", "Negative"),
+    ("The staff was friendly", "Positive")
 ]
-
-<<<<<<< HEAD
-labelling = outlines.Template.from_file("prompt.txt")
-prompt = labelling("Just awesome", examples)
+few_shot_prompt = example_template(examples=examples, query="Service was slow")
+print(few_shot_prompt)
 ```
+</details>
 
-<<<<<<< HEAD
-This helps:
+## They use outlines
 
-- Keep content separate from the code
-- Design "white space perfect" prompts
+<div align="center">
+<img src="./docs/assets/images/readme-light.png#gh-light-mode-only" alt="Users Logo"></img>
+<img src="./docs/assets/images/readme-dark.png#gh-dark-mode-only" alt="Users Logo"></img>
+</div>
 
-It is more maintainable and means prompts can be versioned separately from the code.
-=======
-You can also load a template from a text file. Assume you have the following prompt template defined in `prompt.txt`:
+## Model Integrations
 
-``` text
-You are a sentiment-labelling assistant.
+| Model type | Description | Documentation |
+|---------|-------------|:-------------:|
+| **Server Support** | vLLM and Ollama | [Server Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
+| **Local Model Support** | transformers and llama.cpp | [Model Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
+| **API Support** | OpenAI and Gemini | [API Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
 
-{% for example in examples %}
-{{ example[0] }} // {{ example[1] }}
-{% endfor %}
-{{ to_label }} //
-```
+## Core Features
 
-You can load it with:
+| Feature | Description | Documentation |
+|---------|-------------|:-------------:|
+| **Multiple Choices** | Constrain outputs to predefined options | [Multiple Choices →](https://dottxt-ai.github.io/outlines/latest/guides/choice_generation) |
+| **️Function Calls** | Infer structure from function signatures | [Function Guide →](https://dottxt-ai.github.io/outlines/latest/guides/function_generation) |
+| **JSON/Pydantic** | Generate outputs matching JSON schemas | [JSON Guide →](https://dottxt-ai.github.io/outlines/latest/guides/json_generation) |
+| **Regular Expressions** | Generate text following a regex pattern | [Regex Guide →](https://dottxt-ai.github.io/outlines/latest/guides/regex_generation) |
+| **Grammars** | Enforce complex output structures | [Grammar Guide →](https://dottxt-ai.github.io/outlines/latest/guides/cfg_generation) |
 
-``` python
-import outlines
+## Other Features
 
-=======
->>>>>>> 732de37 (Add deprecation warning for the `prompt` decorator)
-labelling = outlines.Template.from_file("prompt.txt")
-prompt = labelling("Just awesome", examples)
-```
->>>>>>> 8b15cf3 (Rename `Prompt` to `Template`)
+| Feature | Description | Documentation |
+|---------|-------------|:-------------:|
+| **Prompt templates** | Separate complex prompts from code | |
+| **Custome types** | Intuitive interface to build complex types | |
+| **Applications** | Encapsulate templates and types into functions | |
+
+## About .txt
+
+<div align="center">
+<img src="./docs/assets/images/dottxt.png" alt=".txt Logo" width=100></img>
+</div>
+
+Outlines is developed and maintained by [.txt](https://dottxt.co), a company dedicated to making LLMs more reliable for production applications.
+
+Our focus is on advancing structured generation technology through:
 
-This helps:
+- 🧪 **Cutting-edge Research**: We publish our findings on [structured generation](http://blog.dottxt.co/performance-gsm8k.html)
+- 🚀 **Enterprise-grade solutions**: You can license [our enterprise-grade libraries](https://docs.dottxt.co).
+- 🧩 **Open Source Collaboration**: We believe in building in public and contributing to the community
 
-- Keep content separate from the code
-- Design "white space perfect" prompts
+Follow us on [Twitter](https://twitter.com/dottxtai) or check out our [blog](https://blog.dottxt.co/) to stay updated on our latest work in making LLMs more reliable.
 
-It is more maintainable and means prompts can be versioned separately from the code.
+## Community
 
-## Join us
+<div align="center" style="margin-bottom: 1em;">
+
+[![Contributors][contributors-badge]][contributors]
+[![Stars][stars-badge]][stars]
+[![Downloads][downloads-badge]][pypistats]
+[![Discord badge][discord-badge]][discord]
+
+</div>
 
 - 💡 **Have an idea?** Come chat with us on [Discord][discord]
-- 🔨 **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/latest/community/contribute/).
 - 🐞 **Found a bug?** Open an [issue](https://github.com/dottxt-ai/outlines/issues)
+- 🧩  **Want to contribute?** Consult our [contribution guide](https://dottxt-ai.github.io/outlines/latest/community/contribute/).
 
 
 ## Cite Outlines
 
 ```
 @article{willard2023efficient,
-  title={Efficient Guided Generation for LLMs},
+  title={Efficient Guided Generation for Large Language Models},
   author={Willard, Brandon T and Louf, R{\'e}mi},
   journal={arXiv preprint arXiv:2307.09702},
   year={2023}
 }
 ```
 
-[documentation]: https://dottxt-ai.github.io/outlines/latest/welcome/
-[documentation-badge]: https://img.shields.io/readthedocs/outlines
 [contributors]: https://github.com/dottxt-ai/outlines/graphs/contributors
 [contributors-badge]: https://img.shields.io/github/contributors/dottxt-ai/outlines?style=flat-square&logo=github&logoColor=white&color=ECEFF4
+[dottxt-blog]: https://blog.dottxt.co/
+[dottxt-blog-badge]: https://img.shields.io/badge/dottxt%20blog-a6b4a3
 [dottxt-twitter]: https://twitter.com/dottxtai
+[dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
 [discord]: https://discord.gg/R9DSu34mGd
-[discord-badge]: https://img.shields.io/discord/1182316225284554793?color=81A1C1&logo=discord&logoColor=white&style=flat-square
-[downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=89AC6B&logo=python&logoColor=white&style=flat-square
+[discord-badge]: https://img.shields.io/discord/1182316225284554793?color=ddb8ca&logo=discord&logoColor=white&style=flat-square
+[downloads-badge]: https://img.shields.io/pypi/dm/outlines?color=A6B4A3&logo=python&logoColor=white&style=flat-square
 [pypistats]: https://pypistats.org/packages/outlines
-[dottxt-twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=social
-[youtube-dottxt]: https://www.youtube.com/@dottxt-ai
-[blog-dottxt]: https://blog.dottxt.co/
+[pypi-version-badge]: https://img.shields.io/pypi/v/outlines?style=flat-square&logoColor=white&color=ddb8ca
+[pypi]: https://pypi.org/project/outlines/
+[stars]: https://github.com/dottxt-ai/outlines/stargazers
+[stars-badge]: https://img.shields.io/github/stars/dottxt-ai/outlines?style=flat-square&logo=github&color=BD932F&logoColor=white
+[twitter-badge]: https://img.shields.io/twitter/follow/dottxtai?style=flat-square&logo=x&logoColor=white&color=bd932f
+[twitter]: https://x.com/dottxtai
diff --git a/docs/assets/images/install.png b/docs/assets/images/install.png
new file mode 100644
index 0000000000000000000000000000000000000000..2b89b72680c6dff599fc96dc5a2cbfe014a48903
GIT binary patch
literal 12669
zcmeHu^;?u(*Y-8EAYmenq=<mh4GKs%(y4&bokI*N;H}an4bl=r4-JZdbTc$acg!%r
zFvK_NgZFnF??3P!-}msthHLL@*V=2HYwhb?{G*{JPfAQp3;+PBqJpdz0N_Bde;*JL
zU`vvp2{YJVcy3P=b%?M(engh<u+JbjE%_%vNgwSx05AfIvX6DVO}A$V{dBCFF81eC
zsBPZ99l%?^_nPr_K7{it@rEHFrvVSCgS;o*ZN$u7%wmiBqo!<qpsein_mC3ODMNr&
zP~#?Enbys_l@vn?iHR3TG2yLjm1vI8o+ZD&&1GdX>y^aShIBr@IiE?FD}*>ij1b^g
z0m;HM4i`02yA8slkOqDg(nJtryuZz@ctj8~J7+b@%XUV<mGnP-0UWX$jDTp-R}quT
z&Jf^-z@LG!qYJA*fVEPBRrhOuMg~AQjPT#H{wn4lf^afgsZ)mKeou=nuoL6{KR&S$
z&V0sWABU#n=V-}L1J}CC?Mmz5Z*}QtUV-Gg&PXUW@u^z)EO?MoChQVj({t6qzjyUg
z{LU#5sFq)e+4pVC0DHN3W*4e@-ExYSLq--<YS!HF@w?94tp!&T>-)S+LtT4*m*)Zm
z5#S@*R*j5Jb_^>mS`pFC#ox#v@)38U5{;vdUnbNzk{q@9(En8n=`|e5{oVqVf^(!0
z8r_yDHT7{g1j(|ER#rfsQU2til}LMYd6Kf+8FvS?S5Eq!eQ{ceA-JI*Q=W(T??yCi
z#Qs$UMv7`kzmUiFOU~PGPfjCoGq{iU*t%j?f<GZCzX(RiU#h@o&dA-8_-+}KJJ($S
z+K28-9^C1hnQLRIL;ulE?2F&7?bZ6hp+;2;KQ+>5u&)WbFFmiI_0hV=^X<i7Rl~Et
z?u<?|thWu``7yM_d;kcuIK>!<4mEZ{TX6q<YwN~d=So#N*Y340lD9Vj4<TEp&rpbJ
zn#|pmvdhDPq~bJnRTb)GzeNz=#h2#tp1<N_&1_5duW<FR>DT-EC?s-c6W#z`<RljS
zHLyJ%tkUb@UJCaAOQr81XkkfEn8S<Wz9QM_!C;S?F^x)>;GS5!R)7%Ljf=G07=>x)
zxc8MrPhZu(TWqHI+MZJ-I&3N?r)#9bAVM}>$D`!BdboT;G5J`3qtxT;hinQ%)htmP
z)+J^Y3wsNH&gpwKWD#c`A{+IO4b$0-X@P=5Y)>qUO`Lgjh6tIrxR$XdD@U5G!LCR5
zuocth7jh?DvNf%{328bH&P3@Vm7HE%X4kO@ESrD258YE#tW}Smp`R#ij1xyH<`!W@
zmP>U@zZm7jM%+tK((ada`dp(OX43Uos^FnT=`{1`j;*2~N~u>pE}N?}v?`$C&VLS^
zpL_lj@M-a|GnI;aJ#H;pkjf@HIgQQH>#R}76sqo-5Z~Dp&5O>QWcP|YwuHiG>5Eg`
zBYN&=-;<cqQ}Wzov4OX9IqF58p#942()Eas^g1KNVysqZCdy$4aEdYS<WLlmQ?9&H
zKTp(`@X_d4tJfx;?`x);JyXZ-53Zt=o>>Iz+=^2k3YS$3WR;n2I~?OFidj(nBQjx~
zmTMpl@Y(sNpE3M5PU@;VuND-*b~xjCZxCD<E|eDw{c3ccbyQSWj&`I|v62z4zNw>Z
zV-R=hU{m7I7&N&F3$%}7$;(r6%{!p5W!I{0)VZ(Ror97pUn9M$7*5Fi8v)GIr9Y5E
zsL#)ne#Uu<IIu`mJ=2fQSthiE&-x=eu6vEhWP~&~8`mgL!c?O^R_^(BDs_pMyQv4C
zuIG4ak_S<rzh~Qu8JT{ds6y@}467TtO0&xyUG#E_0;jbO$LIPVVGrJVH3-rO^{1@7
zEyByrxwW^tI!19kJj;|=ZuVJ_wy3oU2_3&{BE}L*cYcZ*rl_gP(8}f7xKs7hobHe-
zr9sco!<NqT7ztOGnGf2=uwtTUL*8Uz=yUo182Xdj*zxxQRgX8*6R8oUP7mei$V|<Q
zxi1`_nDd8UP&FH!c%eeXw`jJiJcPZ^bYZ6*$lAgDw3}wSEJMt9mn`PYc}iNpmUl8I
zeyI4YB}M%B59$T&;ThTkBbjqjRi2%isguF%H<k?M9OqaACw4!0-_s8bxbQxwU!US?
zuvj-|uX5=Ip*J}lnCXefGUpcd+oEUr*uMniNbSV+tnyJ2hkv;reD}B1DUzf=kU|3a
z8UtmfvXwDC1Gl;O+hG+oI=b7OaZ<A5B*$}$p4|hIlIxLlvklGWs0a5Betg+AMg*F^
zE24D1$YydusYWkV7lcK<Jvi!;SYDXYXDv?uO!<&SnEvB$^q0PgZ)gdOsHUa&Bb4B3
zbgvyz*Gn=TuT^+}gFf=OhBg{hIwDS|V;zPXm2C%#O}~|-PcT^{`<GoiU-$iVy6qt7
zGwVOa-6>q|@T;{r-SL_>Vwk=&^@XqX!~;>|r)i;o=8?tW;srMLb}xEey9eo*H}Of>
zFumb3h>Q0P7UZ}LwC6K+A%WWy-+dP2iTSRk+&Fw#N_G)%YRpt1QO;af^ID?TPdhW4
zr_=n-_-zy7KM_ZUOVDj9pd<TuX{EY&>76||ges8ZK{V!&Hd5&Rw<%iR1IYE-^`h(R
zLy`>+&#3Lj(!F%MrUR{iByaLkgI_fUN;x;w4tMy(MGp;r%$jo6*&65RRF26RO<E)4
znbLUnUiSC-X2=m&{s6q?zNnXUswSk2{lw2`H}u^-zwTqMS0D~%YyiB`rA)QfBOse(
ztDQu<*0Q{OC3<i6XK5Jz20%}+T=q*pzn6D+_|B=~MS<?kc%=mzZpCg`o*#_@$MxTF
zLC|ehY2a#>$Q6i12EKSOa3_(W|LO0^8MS`xkqXC}NO(wyp#P7b&0r<LuIL97#qFTV
zB_XcWkUuic{w3pb(yLP~vM25H=syrqY58k_s;}cjqUBu9<3x3=&D$bEzYQXfq;eei
zQ9REy^&-;ftH@ba5c2o<uR3wK1OeNuGw1|<PN4M{cZV3NIKk;UsQwsbLU*&gH;r(h
z)%nX)$Mv~U&Kc29sfAh2_8j@~D5Dt!L%;kOW)-TmZzabhah;`B+r8SZVA}NlZ=>0s
zNPk59_GZxU<wBV84Bdeq7mQ`ll_2>QOGRJX!gth_m8X!?N|-|Kr#%>)X7)U=iZe21
zCkm~%6<I<T&C1MG4#r7{^|7>joTFkiqInqqR5!jkZB`}XrrIoLv1nEN_k$mT#U<Ru
zvA9!5%7@G%3*6yqXH2#lX@3+R@&+F}>xk!Q=6tiYcaDi#a_KMZG65fTGAK`5%$;+M
z3*^FrG2OxS+5S_pzZ+A*_m|Xpsz@-0PG2|jWWLJXI8?NC)E&R8(Q1a?Ve0g)CWQsM
z?S+!&h5y<ag%}Q(Byb`#&%oyw9iMg^;ap0FiI0o#G*dR#Bo93Amc};qO>wQ;DBdL}
z+$?qa1ILadt?A7H)Hp{GT+vGhBaxwRJjP+=mBxk_%pdkB#dg8AuYZj`?+lX1lXiGP
zN7lsG%G!9{WQdJf)m8P5d;KOu7s}VaeClnJKdTi&rb>P1{=S^UxX?H%xAiMu`%R_(
z!?@@fQN_tBmk(37vI8X3w<cHPzMVZoEcRI@qMv$>g!sQ1RS$imeCdng?ki4kEV$Nh
z`q%Vt#(nulTyXz>LN?t$(dSA8txaSX89xtx2V&2aTONV-G@O#DuBUynz8zZAs1+4?
zFOfrW=>E_7luBIsh?{ykitjrf<nGliuk)4_!+FN7G8%*ar|+hC@IQRFLt_9|>Wx9m
z%w0J{BYcC&_jYt%Bp^0ZnusI3yCt-?OX8Q(wOIt_@Y=Yddg(qE!$_+B0CR>FkuW!K
zlB_E9VaLu@IRP)d^1Ld`p;>+DWrVVzuk4qeK%v#Be}k8?0@K3#nC_YK;)jZYj_UFE
zwfn6luKR0^)+&jMupQ*k6{Xkra9si4V6icZR`B41Lk(QKUH<r%=SD4Mpoi7T!IOWm
z8w=ovS15q(iX6+!D3g&5p9u13;(-qF<z1ju9Rj>c=VJZiWU*ZyaYzFnJFPRd{zk6;
zIC#<+dk(70F7FgUIOZyXBymsv_#P167WN7^`BJ&_vWv6|F$M6V+#vtve?k~W^K<|-
zAZeuZ2T`#7o@+sXJ2Aaaa{k8T80_8ca6*gsvI|8F4qnSA``h<}f7>dy7-yFTD*NN3
zFC*`uPd6EXcFPJ4j?4YENMi4U{?)O63=P}mdo}=i$d(+e^k0!&O7XuU`QN1_AvVd8
zk@sG7n6fOkhsC|`A~tZoz(wqLFoYTyA|<&bBusKuliAo)9?E)djdSz8?2AtiI4SVq
zy!ExvZi<DKwJQjhkm+8|y1K(~ZrHR-$F}Q;ru?YSP<t4S)!~+imX6Mdd60kj2D8_+
zXT9aBCERB^*{P9LUHrU8M-QB2%0W*xj8kk_75(Ll?Y{RHn&Lb<_|pXL)LsE_D6Px)
zYY6nq^zH^UGCMoFa<Fa>z@vC2g0;+tQ@oEJJr3|%j^jPqswgH$$X}d$`5ZMKeP2Ze
z8$3r!aB*@fJ2syQu1~?BH+Q&V4cx4|z2*Yoez%B`aWUG_R>!43d94p#p?|#H*E98b
z)?kW}xdGZUBaM%qRC;P6Xex@@VY9+0FU`m)TOO%5kJt=m>vFhmuu>|(1N@hy6%!j^
z$M_uMl>$BS5$Sn0d!iSoB-z?|;k<$8IGcoeFGA@+rFu|P*VT_yW8~pZ&1bsSy-8e{
z!)?2w`=@juJi|yVHuSVt;>_;`uxF``4*)gJKbgzzClJD<kYZJ!;mi!=eE0ypQ@`69
zYtxiRO3km3JLeaZotx7c&8ocAMrkJb091NN_0rn9YdcrIbna1yf;JeiOBl?0Nsfkz
z?Z&30(EgK0FO6fqw`m4G%T+-)UZoLK2{`vjp7wzEu;ELKm4ul}V5dTw1E(uAq>u{g
z?HtD@FF`04!90m50A`I*^8ON&`uBNth5@H8F|l!RavFw6iP2@(MEZZ!hRR+La&U9U
zt0oJ!Eju%VO3@{EFF0=ii_E|V|35o3^@_T>b`!O3dG=F>%Ess34(FU{)Ql9(*z-Gf
z5VJK^h5+NI33#A~k~ro_Gma4~Fee#d>!!I{Hsw{Z2rbZ6O2PzO2U?rFy_s-`Xmx*i
zewf(TqOs9i^@}~u@<CPiD`Y~$T~F5Y6nAGES-H4Isc{T2%Z5H@VQ{OX9SM}+Yi`~e
zM}!Bj>r#Qh`Rs@sqm8v=Qb{$P=m8#jtD@iXCo{?NQjB_hOlazZnL5vkuIUD=3O__A
zI|3_tMIWD92TB=7N25fSj;n*U)<N>xltBlZbLx)uC|yNG8&-13V@Q$dSz?N77g1+-
zH=m61OwQ-e-+BTu?I|5%h&D<)nv@R-&u`<UxJ<6|+8;kGG;t?Oo!XA>>QX|tt7w?P
zJd2BZ`V(gAJ>mI<eK|>@R`1u4$*ybnJHR`&@?%K|TLwDO%7nSo6(L^lJqqnmvgoKN
z?q}Ao0y_@35v>tYKj1WihJ_8u<>jA5AZ1lVECK@05KqHmGiB6bxlQU7bT`M!Va`zd
ziSlQX=W8#}d#&WfcEyL+jsxb(vRZm3>brYpFtzQiI8sJky*DGlyUBJXM?6Ix*)J`B
z6c6Whw=5C}E+Z=GW14x*=>66*37{3utDz@z6~(=1LV8sbyz9M34Rxfn!qRFID4k8@
zlf3j-G;2G!W=_Y3@lgG{&8J#k`x@qrZsl%E4amMY5H8{SfV6c-0^ggI2|rFyW}Qj>
z&5`PQ{`YfoiX0sg_&;#zSm$ein!t;I%@2Mf`NJySiY3G2Cm?>f-#e$-@|>?cZ(F27
zZU91wB5zJWLiWeboLeK3lIR0%pQWFnz2CRas*NKDi6QIjHi&2g`@t-Eaw>64=hV6{
zy+~|ir)TD0HR}^B`Y2%NyYljbiloVHPeKW}*Q~!DJ@}xE+H7Qy2qJp%$jLIz7(7!I
zof3QV0OjFvM9ARB7$I?}pEYHLjW5^%dY`2z*X<>{Vtf%4-DF9jZgKf!)gieQj=Aci
zK;Ywn1oZK?7BWo=`H}B+)2gzvlI38GdKY4O7miuflxogZO=UZa*k(A8FmecJk+7Xa
zrNZVc0<yT~m~K+<;)IhxsCgSqp(qmYa#x)7etSf;y>A&oqFL0Na0W|%#$tntr#sCT
z=TBSVPHG}@j>l)D3$C@vm82@><QZ_VbNwd?uxB@BH;_?`*i*qXnU=~Ep{NT@8PkM4
z^4Y0<>0lO_dE*;Ty?r#wPIQD`dRuiMGASy`db)|fG2iSx+^^tJNaC$bC|TV1@3Iv`
zWzs7>X(sGC^&e3p*0)gE-P{$2+no2tIh@XrT!nB|axP5GVtCBg4)S<%gx=T{NY$L*
z&#NFpf))wzp?>2Ve0+SZN~nTiBA04iqMW2m&-ptCw@E_>#KH%jP7)sRIrC(!P_e*q
zC@z%v>?dxx#8{n<d)4N~XkpQ|+9b@$3DsmLJ+gRqG$+J~zA(*#OAG<}n3In=&su!L
zY{{Dl@-@H!vrj)eJ9~GwVocG@je}6=;YQ%WzB~e{J|&w+ey~2%eKM_4=pP)+>$M|p
za5SnXa1Z4KH+Bo7a=2r27HDGAyw$j@Wn4aYa_X2cr52JVtmBK?;8K6!`x>~j%MpAs
z#Ww!~(AU(o3D}pqzuO4g8sL$kV;dsBMK?Rz%wt@mq5H|=UC1Ho0}EjL!8W3G=4jgF
zMSo+s^|8xI@vt1sru01Eh;|BQ`42x}Xn|VRTS8Ns3XjO-nWgEO_{62Sw6C*0c-DPG
z{Q>XKo%9912oysQjmBo^nZ(fyJI$C=zFXNQ^sH#i!E$(|mz0b(^ag-xbH}eFL!a-p
z=K6NSOCO%CB+DJW7`A9NsIWG)5Ey?9*q;OWU(XMQ)nW=uY9}`)M%U?(EB(+icxp9<
z3#J0|g;sUgl%VS!-)~6~K?I<CEq9lk=a8ajR$Z|L)-{5?CqL_74KfXCQe9+j=YsY^
zXdEAEKju{-pLaEQb*Nt~tl_fuC@5Kf)AQL0l|lWHZlE#!M9zna&xK0W7bkCqs+}~6
zpE9c9(^3IaLK1$+)m~V&_!aT+N3F83Pa<M<jon|K6&9Pegseo!UI$RP2g`b<@vQw^
z1%t`i43L*;KDov=f{AJh&dIa6f%+HYwV5v2AdOYI84q}oC^(5O@O0l5sVZdXmO1OU
zNQ$)c(+qPP<+^u@ArH7XBj>eSd0^+LTO(fsDlLN#qm?L4y_<|(J6Wv@(_L#<P5B}Q
z@Q88DO@j0d@U9;(ek1lmu5_-ZLkl#2PTtiFC1&9BGT!eeWB~}sAwL^dt5kD8iPW<v
z&5HP+r---}ol)2^Oe%Y$M)i!`Dm{`u?j+n1W_ajKE|bc6ww4t!CKg&zb|ere-?IM8
zq9^D3B{4+}r>-;gvTFnF#jxq;q#xD@!yUs7=LlmjKL3NA2BS4p?m^_;zU`&pSZ{3S
z(&jhJXR#`~qJTU9{?=|cvo-Tpi`lv_y=W~)ts=4PaDMdY4&C*1QI9XE;EYD#g!OEh
z(k$dGt&$V4)xf&Yo*NCG<=ztX274_y70&a_IOY;KW!HhKgjXT8!@-A$b*3}}uikhG
z@Fw@pkiR%8{#6~(&t#{jvfJ;_yYC(YSLuOPpZ?9&uOF$r6zV;<mr}iYQcU@zpiON;
zlVGnMkrzca9LvkMtSx#Q1tcjC4m}LjR{K)rR<-6FZhLNzmoabklcv3cXyemT0v?**
za92pgr7l|Qn#FN-(Hjlb^}9_*g7*1hEL34YjUAjj(I#MG3^vZ<9264Z(<^wiGUKq4
zH{pR*4UZbP^`RWZh<5)-Ewms&V{?Y!V2;86AnJkso^;q~*vaF-LxY!{xAcrHdO;)f
z6MhwwdaDOpd`EqwKXpmFR3(_MKt%D)-x{=u4;3kmJZmZTV0C@+BsHOgUq*>s;y6;^
z<%98S{W2~8<kC%g?yx8py9epav?&Iy7BvoGJh#dn(JNBVSpj<ggTb+FQV;qJ?tDY%
zi`TwFw(!}2M#2iRU8kw)-f2ScSmrDnj`<nr;q=kzKkRAugl2_%ljh(fGJ!g?&E`QU
z^iQculM;bpkEdbOR%bkP3!~_1BSELCCr#W-g*^I#C;=Z@9JOoC#=uA$gJ_vP%%Z>_
zZ+2GoO^5%AOC2N2_%%iH46>lL1XX{eHcseMRI3WZ%T=Cf1swCIPjlB*{U;vi8fZBI
z2NzmU9^gwM0d#+{qw~|F7e)BY0=OaYo^Fh$DcGT4g8JpNRD>N&8EYIb`_ruZ22o?i
zrR`*-5E2H{SJeeX0aMN1WDtYx$u25AwaBsBi7?Givl>vp)1hR<_684GZ`>&G$qOwk
z@>%c5x{-3P1%$hFyW+>=RnHCaa`~Q=StULg>@YL1x$WSDoc}|MyJ=qE?g#_DF9opq
z$DSztmwrU?qU$<IORAwWlA;XtO$=M2JVzUa(CE7|AtN^0{(W^Gn@ep8rm?%dv|qg&
zq`+BDlJpSio7gKm=$$UW->8|TddFr5@Sbvj{nMvtzYfwsSFx8u*UP9RpVw49*ACr2
zAZZPJwdOlJJ*5YxE-aLI%!6^p$q1>k$<=#anh4G3tY{XLXksWjZ87xLHV=tAsKO0}
z_%AR6=aV>^q!4vU9{l9M_uu3vzcTRyYC>Q8nj<p`K+Y>YiRB)gpwgz0Jsun^rn4m-
zVYJnyn4w8WZu6f@HE7IMiEV+AG%Qw0gTc7ubX|e-UkQ5yAa+DSnF|PS`Cs2>LU{3t
zVq%-mPc}0spRh<kROxZfA5+SYNCk2j>b-J11jf9`3=9lhhZNZjXp|X7^CARp7MhAm
zDZ{p}c)rDY01=_&TTO)E_qzHwy6l`<ia{PdYZbj3sO5!6cgt9hfTzBytajw{-$`<i
znvL1h_CGK?pIOW8>}E*)V|rF*nAwdP<>`m>zIFr~G0+XAx?r}4*u?5PWm-h%^eAb!
z#$B9gKIa|)8{pOGD~rpokvy@8@*!%P$M6!c@&o}}3Xp&MBrV_|(;EP;(frB<NE3AU
zG7%`yHc5G~o|o=Wsn=Yvz1y5#8^D%sSv|DzTo>5fAFV^>s_xjhD?@Xe$RcDiSg&A3
zVY?_`bxmO#EsYX5AsT*x!py0c>26ByAKxH^i97<&<BQ=Of$TS!C``hNK1g@<m=J7<
z9w^%{e7}O2nmgBh-}C%q{p$3L`QE6n>=4bs2y$Rsj5}3WvrlA|b{MJ0N}e26=CS$C
zy#4%Po|mV=bKk3G{4_d3mA-~=X4VE@8Z<VYw3??GP@NYf^5|j)e2ra}cZ<;PPilg#
zzjl{fk&=TjNBYBQ-wi5W8elj>T=BRw3Y>0iX>^IEM5ywb?c{xw(yJpEeo7V_+X?<)
z7WrD05Ilkt5nHLt>$)H}RI9Ge!7<IgE8Zka?1eXH$Ufq>HeMDZaG^u_=5(MXBELnB
zakJUbiR<2;lk#T?6+IIVXHI5^=0t06eZ4}sF!>EjXtMZ3T~=^Y9_}ulR()R;ZpNWt
zNQImte+jAG+SJu3JN1IJ;Vpx(iyznkl%b|Z6*e{K;PoVNacaa)I#7D!Ji^zpzE`sF
z1h+1eAzYO!q2unr5Z*6j&Kk(I#~qFxh54?2NOo)(OzJp{2yA$@<GugnO;Fa#(*e~f
zf)zO~xIOOTzRr~A_PQWDrHOlxvEP21mZpJ)Llkm7uaMfWQK9*2k}t0>STHsHr9-vD
zp`#zXi2mwrzynoH0KHnVtx_7y5d<v+?ZH-4U(WoK{rL;;nvd7+n5v@y39+ch)~jus
z<Kq_*-BNP_Vy`4BQ=RzXZC9*u*K<!J^ZS{cL1ERwj*6D65s=(!`4<t&4~Orm-Epfs
zpn0StZhb6%(I&LiAz(IIrX*wpqy{Xv4IRD<FZaTz`5gVTszUL=w&<Ac!URS?#P!ad
zlc^DZzQM<hW{x0ttV`$2g5TmBtLs<%Rih6HD~ZoGPToA&8}Ju<`6;7-MoH*4fQk=}
z=~Z%_vC|Qf9y975g%K5%#e8h9%1k%V&QmM%UFLsld&l8b2X3);_O6dZjE(hZa}$sI
z*Mz>d!#P1{nH$;i?%2EbW$PChxBe=@n6W6_#vB$b(LQq18OuQ_ehU<;KPOaMBA+HQ
zY#m!*PmhN8%zBXaY7Uw^hfah`v$JeAIUn_=8okgSo}@G~2S%FL6Iy1h)oXW~&UOZ5
zio-H`S1KEQQRQ8AM`_^At+!4ca`q_e**&l#ZzH{OLeu?zaiq!{<Tle8?||C?&v&wI
zu}{wure4po?QGbUL~HtFkvWbS^_w~d-45E$l(>qG6Ych@9YpQ3SN-3l!Pb|-o4rC-
z`@`y^`VTU|3G4T*W(b}#K&M(B1Z>zCoJ`t1hLWZ#i`ooXx$e}i>RYW>KVnD(_l59M
zNjNJrHO6$=4}VxQOEH(n$QP;zY>4DHUu6K7ti6A}eP*;j&8bCAvBS?o_JfA|jha|n
z(nalhH>w;F49~7Bzi&8N+SNFJI@gtRgH2zXh<b|1P%nAI?@=R9X>;m<v6`4etg(Y`
ziY?Dx`Eqwreef}5%qy`7Ix)3&Q};|1br^n2OWGmsEyAU9G)WOAAk<fOagwO$Lt6hE
za*0>l9|r~msKbsL9MRce&<4hKI9FAdsK`Ii3bofy(%0Kt^$|x8WlNqO5Wn%Q=47KV
zy`!!)4^~^KP<lozBe1$pR4EIRCy;hfn9<vHsMy~;nhSi1x|8bCo;XGrLN98ZVbsF!
z4kPqWAzNPyM^Q6XzEqHsh8LlyPfCZP=AR%p-H<C>p}zQbgwkBCb!UlJ=<&r!ing`-
z1l@XFr&bJ61C7-O;VhFF^P|kK+O0)o|E%|p%=ybZ1U!U!AoXF=nB|0R0bM5f#ct^%
zk%r-|(Xr8kv?77R!vp%>^PjMrZs4YvUgNY(_7rpf+Kuguo%$n5z65q}o~V+vv;ZWR
z>lWmEOk;a0H^H$W(QahmwrJhFER19jImQ#yui`ucKG#j8|7z@}%+WX#TXOx7;p9%E
z+#Fv&zj&Y`d){U|<%wJ#eAo`!=(uv5p0*vz|M+D^ZK2{k9fd6nI?5?__@vhR)SJ_L
zW8k1G4y~(RYHBn|lW&AK3M(JovCAn#X!p|5BDfo+MU&!MeAkRyifNkSB&IwSl5DcL
z<&P?>lLnDr?g_yx`9b@6VS{GH1@MDj>la=Z7(<x5p~VhYA?)h#(fIm`Y2{3vN7Y8Y
z1eiPt(63i@=T0oKgm22uE!HKO$P*q24^EdYELpH8B9wB%uCRvTOH0gY_s~{tLYrsn
z$ZNUxe6b&sicL#HZVOF#IM52M2_pj%9eV>u{nh6l?fNv5el}gKJGS{y`xJLqjvlKr
z=L1cLlXgXS-6Y&!oi$o@7d^}$7H}VLG=lB8p3X1I1rChDz!2`6!#_8_XXBxl7S+S%
zl4L-dl&{zPXE6us4qO|)XDIkG)r+FlRxC3p@H<X^r_D~V>Gx;`TymIc29Qe}M52NR
zV(DpS;sQ>%gc8Fn#YQ=lLz7l#g^mW;`hq~t*pR!t!aU)Y?uMBAD~0qzv+JRYrA;YM
zKStWIo7tV!k%=F?sXWx>89e6QBa}`Z9vR3am^Atvl$R;fZ*jb5F&5H$l7sbmnJwTR
zTAs!^=*7V5RT|H7l{7z-#{R($8|YkUE+N!wcXlk~E9o}pDvusmh}~3lu;-C@eIMzr
zM^vZru+bLz^{;)3yHet)_JYtIE1KgP=g@faj;d_vV3QrCh{pncbfSrymG%54xaa*&
z|C}K=jpXUbq*i;l4CR|?%h%riyXqI4>*REFv!o4EZ!erCwR4A=+aqZ7cAFxpXAey4
zC{6gs-tOa)_JKxt1u|X%BNhWE;eIy|4E*U*VP|U}uyAZ)*MJUkX<FwPEqD|ZdiZXm
z7>NVEm$jb;hYy&SOnqG&6C<TNK^J<RcyGB>Sh%;`mc(9>IU8t1-8a57M~G`J9tL_E
zzli-ux`7xQl*A{e#rPn16ZnzJ2SpYA@!Y8qrH;>X8$D0j&cD>=IpiZ<58%1nx@&V#
zvoP|Z5NupYeFmWu_b$E4CzP}MY9(Qvu$*Avs3&xm#;K7C@hwP!=gGE*D0=BeZ^E~>
zu_d1i3q4?DFUKv9-<#}?aLWd0lpaUopw%qYJ3et1w4FGxHx?_?y){~xhyF(I)J3b$
zyacBN&a7-B83Ol(k*)K>75S@XQdjx&?IO|-)olCQN-bKG+!O9_>GjKn26;rt%s93O
z@(TbPivld_HI}b$X`xQ`7|Q5@sx-$6;f;5*64`Sc?1mVejnXvwy0sUrp@qwsQ8Pmc
zh$jPeyqXY+2m*m9svkNgr=uTf6R%S@-=2CxFs3-QPwV!Q*g>8ypvt#H&>?<#7aAGI
zP*fS9q9)LqRHsppDaUxuTP5B&v-k>$&f2E&#xCON=Hx)#M=HL_Q@8L}<aHONe87g?
zT2)cHzmK(3%L_tUqs7Uw>f?cfN^T(sWY#)Wkp!uH=xoF~_+ttpSzATjjG@8>+#$2;
zc2;(ndgZ{W2^~_z3yMsWs-H87W$4EVzY0-BM^<NP;(`J4j`IM|6kIE&-w7oU=~EI4
zfcyo`b%2~%kmr)ydyM5JL-!XL0s5!*rkCU-5oEF5gk`Zu)-e_L2gSo(qW{o*&gFmv
zB(w~*4if)3-f+n=4&W+~Gw$MCQi1as5N2P%#{Tjd2nxiKp|#Y1`NmM}Bi-113il=9
zC=D!8UnZI2(XvSc?)q=s{$^JKph&w*zLUcF6PC`jAnjZK&!zw@MOt#nk4m>#U>`#T
zg!4XJa&Fj(3I03re?|Qt6`^G_w*zST??J<a2K5*hlsrFi{;MnB^C3{z?TuR;nL2&9
zy56A5a6^|It|z#YqkP2a|J5-9P`bS-W&%N<8B6O6(2Ba29UI3y@wrC-SLqndV<6o}
zBa_cKxk+*kB5~&-naof(v@v-wf=T?}T<sQmP<pr5S=}>Eda_@yWh4@g&m`YdPokv4
z=CkhW|7$?`Bun*$%Ltw({U(vvXc;ny;c?4d=Dc^^9)b6GzW-lov!5T#BUMc>6Yude
z<EeFlEjSMk^WEMfGj5le+Z!YZ?|+B<f$z7FyvX5pwi-g(t^9<Gs|d*FaXol<EwS^~
zKY?s`AxeFVf9d`oc>Jf&a&<1y$)~3a&Pv2$z{Up~b@S(6c8U3Di#XLk-MFJhvPXM4
zdd4~qdI8B*!tKRHoVVJ4!Fvr~I`D4Y!|yRmV#7D%*-hcjF6eOcinUxAOT?c%E3AQU
zQ0_fT+CB=n2HTZM|ABaWA3l}IeVh-S&c~BZ`kO*6%@~0F_OU;#J96{#NZ!52(uwy!
z-u#^+^9$>~M_?-gCf_Ih1qsb-jKI_2s3(_D5Oj>KXjeDV)V^FZ<9#eOm;X?U=n|AD
zq<(RAHk`KaFR?^c4$I5kdU~1p2f|yB$5u!=$f^DnR6dqD%=)Hq_isMNDYjxUuQ;FM
z@36P9IUG87IPm_io&j5NT6s(0Zzla`CTt=IF>~<WME!kcY(;wY=iI-;>R@G{X^~Cw
hclH0fg>o90?Hy*n8bj!a{r?F-QBF;^<cay~{{vwg+7|!-

literal 0
HcmV?d00001

diff --git a/docs/assets/images/readme-dark.png b/docs/assets/images/readme-dark.png
new file mode 100644
index 0000000000000000000000000000000000000000..57951ceb461b031f6e1981fb2c7280b06bd80acd
GIT binary patch
literal 25637
zcmbrlRZyf&6E%np?%KEwKDavr46cK_ySuwX<36~<;O?%2ySux)Zol#UyH^`~vl-D(
zcXUKoWK~z4%&a^oLP1U(2_6p~3=9lOQbP1M7#IXT=yNbEH0a44egXk{gR_^=Z~_BE
z?*H!vK5SQN4Ehq+`Iox0lAWoun}MSVn46m$gN3b?laYbF34@)ZS;i$F9vB!2n55`W
zW%tbUEH`h0h1Twzr*Uf~w|KAqjh5}e6F~ZAqL2isHu~&cyX2)T1PxS8rZx#2{rWoF
zZwX<9rT~9P((m9Hp3M3|);$|7V>&it<J{eXlb4oiwsrItCKf~jeXAR+tY&JwW?grx
z%tNUE$EI~lh2=^9XSr4w;nDv8Ggk^!eDMF-g%k+~#Q$tHS;!jdf7U=3unhY@GgpJC
zMf=})(^at7CBpJ#B)J*a4a|oqA=#orAQRt%_)<KUgo1L$_~S2#>52Z|$b_=nVsQU2
zBS0CF{~v#J;0DD0`#H$k(f|Lu_WwDsT;zV-n>OhYCaOI(tUf${gulOSpUzMkBNoWp
ztRSE{6Qv(qHow;+_~1Z=M*S_v=rPKiqZv5(H!&f^wEvdHnIK^Kcqlx4-!ed3QBg@L
zk&4OE%Bnt=j*f0(z16Ac@88JL(NX@zvzxu)BmPu!(ZELjTz#1|>4=PN!2lIIo!-p<
zXP9hQwx}p=gLpdH)6-KZ{oR3xNh*Pxh<pK`S5c!{buPSU2n9X8tCHVk&%9S1bM8l3
zZj?9-+9M4XbJX<o>_V#H`;oE0Yt;CjQ?zMP0xz6jzG!+q_Vm*w)Z-(9GCCX#sQ84E
zYW9uLXA3*1;nB9jRaoR|#>NA`bab)?_8ihV?Kb@bAYtSpM9GG^Vpy>ORY`bYm{Af?
zu<_V*{xIJID2>KdD<3S4#_Nn(UM8z_+BZASs@hW~a|8t&8E=@a7Jp!b$M`?)F1N$J
z{L&6F#qI<14>#g?cUXPP5H&U;;ow08CpaIbA3?jWH8V|nNm%y#=#xDjIbtvxPc@y)
zKB2Uv6$#0fu>wH-<90fP^1ENG55qFLTyIrYQIQ0BIiAhmZaf6URhcorCs=Zk$BExr
z4%xA1?m^yq(!1xoDswqo{J5~vds~IkYO<awDk?JV4g6Nz*2aTM2yPSOFVhtkYuJKj
zpnPd%_OLN{6lMecZ2F&DAql9IUMm7D=AVIp9(|zrb+QFvKIVasm*QqK=O8~F^@+yJ
ze(*ZA_$*G(E+k!VM{sGkSYP{*pwkoUjj;S*3^MvkfQyUE?I5Es^L>Q8x1pGW!_9$I
z8N#o^Vc#c1Huval&Oo`9;oU&^04*q6psII>x6wG=jr)55g>(wveG#LNg_VUxSy@E|
zTR;(cZe`jl)g)V=yg!YX*9U$9hokTmPs3s&MD5HaEUc(BK_~&?FKwn(k~bMB%Za<n
z4UR2J2m4`jUu)DPp%1y7hbu~Eo|c&nDhPg<$AYVoyrie{!YEUkV$l|q_fwopFK^jA
zDkXCk4T^<@g#<bl)2VD7$HR0pUt07C?{9$NO-$iA?Bz!uYadA0MWGETbM;hgd^jdu
zO>3@PZToyw+K_}i1&+RHql9TI+WTyH+kaifEp*9J{4EZ9LtmZy3DTP=qT)R}ED!RT
z=D`^ObC2v|aO><Y--@cbP416pK0p(hgk=af*oGQB5L^&741BCi&aF4f;)jMxf*sQf
z;ofq+QHl_jubd+#yT@k-2JDHWMT)`mXT|i9+#>vrx6;jIEh(E$qyh6_Nc@I?T?+b-
zTQA$rBBunu-WkJ$^X$SUaysap(Jl1?8e@Cgs0@zURTT&r<S=PtSrK{E0JjFz#5Uig
zkcZNe7GLuo&X&I9rDxD!WBeEG$N6Ox>>=$}0)~(o<?Qc**#SnE#H^;{CI78{OE};)
z#-Zh%>W@TiZYTK}Ij5n$&fu*Y;$4;7v@QvU;|X&OV}FS}89@}bCvAvfr`V+5%cnRF
zjwczRgsOSr9?L#!&I|BuOz-n{R^t1u2pTu7-TkDS!3(keK~7#*R`k#O+#_bDQ>lsT
zF{~K@dhzexTisSSyMuzSeK=hT_@EO8$$Wr<*uUe|b7Wz{JRS9BccdiJWC(m&Ph&Rn
zy=uC`gk;f_tNWP;Z6x%435#?;%+WdFZe3zBvw|U+oMAop!(KZzo=Pt7DAzY6TPhE?
zl&L!B{U3_Yeh3VaPsenkg*=M}C2oAYL%Q{Hje%&6fX}5f!O;)L4T_9UB)ct$vz$8B
z=%AkwAK>lZ&F9*znF<WnGHNbl2m3Ubb{+c%tb%8E($Ua{9lAI?Zug~gye?bIOXPDt
zQFE;<Ez|hDIEDa;zByUOr0~?3!|TH&MXf|(QxGskIFw235c)umKS3K@+xqd=)}d(8
z*lz^Oo|a-Vt|}~sv#GZqgHBM+*oFs11pIfwm(~;Jk|VIOS+`|T44$7|FY=h4EEfYr
z!bTvsFROhY%FFw9u48`{vhmhEVyQp-EMK#|@uwyW4TL1XVuq05)y0a$$R<xQpq}_O
zh8Cj0q3qI_ORVM;Hb3<jy7sqsqi=rk17F!>x`-lS);P&oNU)<en$2hOQ5bbQp7E7R
z<SM^8i`k<pr(HG?d{KE8ZR1gR6NVF>KUM*TQU=lDO^k+PbdUm;&BZk}!T#7lBs>90
zF?vZ?mICkES(P@cIrrpvi4=cY<#_{V#XcQJ4{+o?K0k*o;rTN-D9<>`s;W{f=8DUB
zFIx(B3lrsD%l5cZ&BWPlLPM4`QX@9Se}op~^1)*<Q%7kU%=|4VX^y4|4;g8P9#E>G
zL*~uMWvDIrCQ7x_xz#huZ9m?~^x54L0T#Z;MV!Isaa+6AZ1**;VV3paXk$}ZR#wKm
zHyq0e!h-<mB7BE&e$1$A6|^pxB;r%wp(GA8MZ7pa(`*FQF;~)$TB$Y?zLQ4b8kJov
z?r&Y+KXNR)693s!tz1Czz+ip6F;$%g3z?s1#Dxx1d7m5Y2+J2%ju>ZzhtJQ?{}-?3
zJ3Q{hG95>0zD6+c%87!z2_Ub&T(5)+s1f@;`wGFC@;UEmZL_A$k!Dcu=|h-(^bP$0
z3dlR1=e&&wx?jle58o#qo9oDgMMKq;mES*B%yK@I|BDY<ylR$~mW6F?Z7)RBxnnE_
z3_CPWx4!}o44=`$NZE2%LLlDh?xG--i6`YCS*W3gP1|UHO2Qt{k_`&Sf=?fnGWq6*
z%U%sEckuN!u`|2c!Y5iQx1fno@1n=Fgu{|SJAAu<B*SEo%h!0PE?Vv)N)g!8e{>mM
z9Hco}2r7^ltSv9DZ=%35IKcAxoRt?)IgV%m{|)l&AD3bVr=6-i|HJ&f@5d`Y4H-Gu
z*<be?y6d5bxW8LYC`F1~f2k0w{CDl02*kX0Xswt}wBQYJ{`}rPduNu#4>9q>6s8Cr
z-Gf3C{f>zo#@KZ`N|yzKEy1r%ud5voBI%SLd4K=@^*JI?KsDw#ok4jn-MhirmJ0AW
zlqW{aA?QffHIRqYyI3e!BZm@b4KikIWPpN#`MHDbbx%Zzn1O24<DM$ekB<4|YOigu
z=k@2UT5gA;5eYS<fujHC0IBAo68AvSm-=At4Gna(BivJX!C?OaG9J<_OLjrlm-h*n
zBKdY3@|YeopR*6_)|u}^W?wCJB4b(v20UMEWv~Auzu$Jb6VaCPaVr6BYj19P!(vH}
z@9TJ7UELd)R+`lly!$Kqa#yNoraCONin^O5=?rw!BL3nPI$!6}S-+qG*9!w|vZ;y!
zZSw1aEa%33B`8hWH%Zk=nNTKe(RIe1{nly=N!fVd8$~d26{8c-*CF6}e|(aD6qJ0Y
zkJ>~176oB~X)CJ8J<4Q22~5)z(;!9>mJMflL{`TYTFG>?X-);&4J?1*{${gXs#u9P
zydWF2iji+98J>n?z%WZ;vX>&?0Z?OVdI89}OT3%##DsWR9vk)`@*GfMsh1SDV7gR6
zbAzX^mvamrB4Baz%g6~U5wzT6yX6c8H<_-k_^1R_{)D@y!J>`S-;nEX3VoNXe6O$Z
z`j+`ZjrjmWf(h^QvT9YW!p_eA)#-R3g))7Tqz(%;j3%pTybiRt_B!|T)*v7C{q>pt
z=B>YCj>Ub0MYb($fs$IwN;s5dSxP`ws*YivL14)o?qCfJRkhZ<yzXJA;j8BMkrg#)
zfGA)VOZ+-gSS;GL^h9ytKFmbLm}KIr{XRO=RAn|KJge*1*HZ`H({ob*V6m{yD{<I2
zFelg-AB*YPq@}kqFP+HXBwhj`Ih8*Pn1DM$_6wIg(u>xRu)Q+l8N*4%x%qs>YSp8C
zWfCHL=9_$;wee$_D4D~VxWH}v*`Ggu-q$q=pq8S^V~j3$++l24>^;4Rj{X{GaW!Ab
ziSAb@7KoXA6SpsS6><y-&G$wn_{sz9O-Z1ir4KxLPjlET0D73I72~M$earh(9*B53
zDI++|CqEtG^5QC~SCy|6o7o2!T_U5)cjp%ZZ*>r}8<ha(I->AR71u9Ej5`M~NJ@g3
zhNs`)n+wtKg)D!G3%#Mn{KDmPkO9rqm<+R0?S|E#oWYfxoK3Y#yKJ#SgY|D+og)!$
zUhO5E`zy<}7N+clhB2g2mK+h^dVvHQLVsjs=#X)Ja)EF3^;7*$X17TCMG01fXxGEE
z2udX=ZoV=~EbW2?@(_rl4hsm`2@YN***l28_foEIPrLuX_9A_H5wbr2hW77-nvB|k
zh8S#)(IK5l$zB={RZoJ$Vx4JwnHh;!XR!m#@Wo1U@rO-*mnwM~EOg$M`XH5UB{@3e
z)))bdP@FN4@JuJAb910BRhy%`<eHDs=0;*^pNZzQsn@9+2m6bBaGWj*mz_KUWAQoS
z61Qdr4Kag902Y=X5twktT{jTAU)=opJS@n7&*y#(NoGjbZ1bGjXx@N{_c5lgKNz3b
zjOJosO{`vv<a*GN+NN^-<b;Itu0@RpO9INpho|_T*6Wl&=tjKELx7I?$>TUpL@eCW
zLLRd~zR%2*0_AQno%_>+N|J;_Gwwk;(jRsZE7Ag|U8Zvxl~yj`MiIKfd_d?gs9?q0
zx#;d+9SkkrKnhnqcGeDMCJG<gCevIQSF@=I6NV%^R)9R{#~emhl=-R|E5-_KCW^Y=
zl$K10CNj^0yAq0;nsm8driY91q#Q`sTTFW3Fh79t7`KrrYzfyDfyVU26XzJpR6eF5
zfJ&$}2h|mD+n~qCZ&7QojmN?3@!?xmIViv^<ZIHUOIQ}3I9!1zzN!sQ(?LtnsVe+e
z%e4JxH>ol$8G2P|X^YophewIF!yr~UDC!tLp07-+=+MaDr|;muVteAn%-A{YEV`k%
z30_=q{4+6p$)>5sg{X}W149mvbtDwarXGKnNh1<LKk=6jy7^`V#T<FS$kXv2J~dRN
zKl>>kw7d;SL`IY?iH!s#IaY)$Oo+O|<|@KBFo(}_B^YndFzHvAuHM3ab*a_As`Ngw
z8Dc^c7hw7J)$9&_$KO)Re0hO2GNVK*ni*OanrCQ3cK!|-OXRWr6Z}n_i_BXg-Cx0f
zbyc&hvQnkG>-AJ6X3rbbe}Ln8i6U+}yB<xvYI=l(j9_7|E47k4&}dDo?=FX3nICMI
zj4Zi8Ec$g~Je_sQ4^%WH_*Yd`O#&}}M@5UpU^93Bc;*w;Dc?TqUkUS<2~D?QFM+HT
ztVLT#gj_L1Hyr%w(T$3ugR$Gc<yr<<4}bLOND}i3v<sHK`jzI@d3`<Mu-Kg?>_r2-
zVIG13$||WqXUag81cgV5%_!KJS1f(U2(%;cE|{iy+k1naVl|)2b%`8BL(=xN36AR8
z`#mFQtfUu>MI9c?A%LXCf4D_-l-}M~xXZBWH0!rOKA646jYgqN#9lf*cMMN}rnZ}s
zP8G{{I})CspPvEV`+U7_mS_9xQw(x6);cf#CsgQTaHM*dx%xxg^`w0L6n6~k9*p2~
zuiwy~18C(g>JWgZ?FVk*?N!rv%QJVZYTZwpUI-{=#7dV&1k>Ho#m+)QYwpom1>&)r
zMw<a9<Rua<j);3x_FQ+;s5H{~GaR-eh=!Q|hEN&#Njg5w;Tf~#eojPOk|@C`zhFK%
z9}BJ*eQ|~P77%n4Lv?rkmY_(I%~26994WX$P3=u243(X^Cn8Hzm>sgD)^d;y`@{8$
z7?zhP!uN28diBZst?Qpfg6VZ4a<u?~S^paYjy6VqMNW?G>mpw{gX`VQrSkn%l;dh?
zo6%G8oL-Ws=-=}4`VTqN@w6ReVu5!#78T6Y_Z%Cz9BGpXEPS1B6ioTeHO{PTrg%fR
z?t46WTV)C!d|<We=b<lmM}pkm&(}rC$;m8UkJfG4l6MxAgBU(`Q1F%;)95E2%QuDs
zX3q1tTudp2YRi!B(97hIm0FR}hthUD$yj0L+~e=o;SOfS_5{(7jg@+N!3xU;!+)6W
zt|V$NZGvWabmaNy*#>f9F=#muvn<n?3ep1h*k_0&6(k2>Gq&6$*$S`GCp3)yd6uqa
z8XlaH>V{jq5+e=kq8bhSlN2@}W-J2_%DIh5v1E8pk-o|!)%6wHqwtyiMDH0kH`p-D
z@u){k)y63p*2m8v4+RAh#F2ufZRB=bSZXGVa_DsAD#7<16h=umyAYkLmC@W*O!11h
z!ty*QvByw~j1lawTa4@VqwQU_UU~>>A{OJ3ScdlVI%(Di&NjoE(*HKeXA1aIlqBKI
zPc1PX@eXKbB~B?h{vW8N1PtE0hp35@^0@@WK}aL>u$l<vTR{6QH}6Gy>@^o=wW@GZ
z0;e?J^j%iB(3btTzH3}0Q7?`afRfQ4cFD)MmtUQ1Gzqc-#2w%$TY#h9D;{f2^xo%D
zO?ftbMB1d?Bs#vymo_rYF&abL*~+sSYptSmC!GXY>3l%Wdhw5ok6Kq6+E+DfPk5b>
zp!4$09f_*9{c^nzO9w^8nP`8Th0ocqAZ499bMfA5-K+w2H7`^L1NABInVNd&%~tle
z%-^T$?V#ej1ISB}l{LGgWTRG`p0xcT4~nZ?-qsAr_L|~^c7{&UFMgL754JQMKbp)J
zXH*QTS_bpBM!2NXWhAJo=(t7=UhWhRH{$f4D3OL#Pa6gs2r_G!K7rIZZ!W><dQvk^
z>I7Jiuwgw8NK!p2zk;1{pV5HFrcv;;8*Z|7E-b*sZ8)nf8HcF_8f*bP;S)d$N9F+7
zt~q7q#Z8`-NU$_3cx9X-+YrFjmTf#EtI$=$p}xH+sw(;-V6QDWKVDeAz)D|T3lti{
zfp6c5V#kIBKW`W#M-+NERM)PZKev0aIBXi|<_aW^H}ZcC?dHI`&>^WeD`{w~AoYi#
zxSgGz78Vzav0xguU4)$0A2f>z<HPgEikNt0(io4PHnU3UYg>@_yfPaNbu=$A6Eerk
zg;0yqtjF~`q=s-`X+bx0MJ(bmrXycMp}JqapvG=xlPQ{6W5fEk@s{T|boYmK;~cEp
z6;oHr<oZNcv+HW_6cIF4Q6Gl9?PObFyziDj@V)5=N>|ex4wYa*QnqpJ$SBY@t^BEN
zf(r^`{-c~YRp0&Ke!3*!%NKlwKo%#3=Y7Gt+wvVphj0)H;c=?8rc)Eq5HXr}dnaSp
zS`UBF9BV$33pSE=6rMk%D(KoYucU_2kueT*(d9Ts-<%{~F!Ke`V-?I<e#c22RX6;r
zrJSV&LRhZ8wa?0*)65J9t!SS4Y29h+8}6AHk8J^ok_~5!{aGmX<Z-D{G<A&9xuw6~
zxs?njI)1vbk_h#w{i~YxUx9KsEDFWZ=4^Oo4f0-?C__G9tw?1fCtwI*ZFj$PkD#h!
z!eo3LN^TjZwhG5Pj4*?c=qrAVxI2FPMG$YXlk!}<Nb-^ek-I4ZHwR6?seQ~u(D{U2
z#^fm3?~kRHQh!i0(cSag6s9kWv68J#<!|K?rrZ7RC4MV|NZb>IXRP^y09U(NzIQ<y
zS39_pAPq=Ok%oqrX{BS9@DW|vzggf$h@dLjxB^c_MI{Fv?W{o0f$`k4+oL8tdhVE^
zB)(i<-{Z5c;nvs4wJQVZ9dsA`Ca0m1o>*(^k7ZBDH|P3m!R{q1etJlFYEK~!xuTl6
zjPFc%#R`N7bhYuwnM@4)p>Rr!_Dt^2G(j_gEP>>KihY+8%!ga(R14ci*RyM@!g`65
za1JL(p}X19&v0khittHX6&irusU`E29};nv_cp4Mo?4YiT`;Rlh0YivF_PCe8JIre
zyh7oJDnFcj*R~aa1}|2jg!tANZ!+UyGtfqCiKaYBC^fQMbvw9mjmGl}mZ#07m-P}&
z;5+J|8iOuV`7~a5%D~PTh?lM&ohUD`5-nDs2W_pezt((cH{HJc$j=RxK(iB(8u{|6
z)A@3G;qyGh&+lPl(I9)*|2zBoNiC}Nle6<>YVFl=kJ#&GmbbP)3h7Fxf*PLv>Zcc3
zl<jD-dBjLk_xk~3iD?DCj;uMDeY#|%z-<LS#v%HUy<-?5BRz-ce!5Te$?oK^HcMH)
zz?vFYg`e7-A20|l!$r7Qv=zLXRA%F`n9s^T7r?{bm#+6{Zz#ZHGW9bg?pA;19Tvei
zwMTxHmoi>vD?vsGTBAHD`hzLy;il!j4c6^TEwGqywhz&wxtieeI`2w&s+R_6$*^qF
zklWa=)Gb--6OCkMb2)7O))Sg<zq3I4Du1pUJ%^l1a?ohQbUogE#HXlr>~S8Qxp-4&
zza#8%T2>;5g{I|iePLANrE#ZE(M0e<s?a{SbNW@jpOyRq;kp_4TY#k?=1RL3a*3O7
zrM6zzVPcBTW3SVE^DdT8tFM)c&Hh7WEj$=`2S7GmtzGYEbq81L$w1&$kmp_l_E$WX
zC`bA``qbjW!aq<kOzCtueqz{#sSvH@dzOmWA}4biE~$vX=@_W!93H{33lPKN1fK^I
z@q2!B4Y@FmiM<A1QTS}8Rila9ixy=hu|)+&l13v6@Lc?ncvC?P)yeSombags&}QB1
zY}qY>;eQYI5sq`lQA?cr#vwAqVW4Cx{W{3~afg;9&4ILp4X%Wkb$)w;Pv}W}umWMA
z-@Xu~5A4tH4)u7G6a;0#g2DK|Ghe>?fZQwnXh8t{sM-7}BO#%;y!}^56k0^wa<M}L
z3yYJIdd4m{wa@MBUmVhVB$6I|Prj`s(_k4wkJt4@@1NCSIK=x_Euh`>go>xH*3jLl
zJw8SJm+Kz#9>_i5m=sAIX5u%(U|DsR{S>A5p0CD7l-lZa1WHTJt8ij`L7Pf)L;cu-
z0Ff~LmZzyOK8^*(Io*}KHeRd(O8qfk{30?AX>uGMsa#B%`rkZ)QEsL)Kay6X)6vj(
zE1XtW?L`|>QNz%J;@0rd0lj^7R!3mbgjwE3j+_OM5twn}Fn^D_D3`CmAj88r9#*Y1
zFv<_)2uCs3`l<XD`SeA2kiIoM$1ccXOH`gAjAk%&BrL->CPJqbB;wZk7X2vlCax)d
zFBiUJ$GGu!i~vKld{a2_vC=aQO0iPgs=NgB%Y$vT<5<%~s?cBi)|fZ7oBUGOJN+1q
z)IRbU`|Ub#4hEZz-UwOke$r$#5$B_6LD5Rz4)#ZSA@U^qYDmd*?F8R}&3+iS88fw6
zmVQjcW44nLQgcuyndnfz0hj!3sW{GmX`ud=4Nd(-)GVo<K8bQL#lrK76tC`$bK5iz
z<2=NFPfFa>;t3VTE!N3ZFqrF0nbAfd7bNS){>BO$9{Zcf%tsQ9-bI0w>!Kcw;W>r!
zcY-jtuT%VQe7w=WOJsZwW)Rgg`z^ZGs*(61_1$;|ui^<@Rh{OVYKnK>vT}p5rp#Dz
zz0$>?z(bz3QFZ0>+S3m&ZmWM>R+6nrLDocYd>YoX@2toGFXHeV7jq*uk-Zq2!Zv^v
zrAFWhoGzC}_6CanR(A%Qsn_$w3Q9O!OV$|+v9U8I9x|*&p+blI2NkKRrsmf9?EtpU
z_P=Hsbj(@lUeO*q-}cH!^qHw)|Endd;8o}RK+}OD5S(84a-v&NGNya{SDB1Ln*Ox+
zwOm4DHBR0fZb`P?1>j?-7~s$mwQV?%ckpeUWS=59T+%FMIo{;sH^CR&&byVcOgo$8
zs61E~7RKmu6TMXJ;RNiRopeI!(`|5QI4VP^(-&ReKs~y?H<06G_Ruv6A1!H*Bt|%i
z#G*eM{LL2vOoEM#&|RUYYr1DJi(M7+94`(IQT99fZ)Ig?+gB|05Btw;T^M^Wd`K7B
zngXIfw`ZyB16s`;i=0HSEZR%+Z8m3*uD~Q1UVMEuNcXqe!&dlFBW}9j0Cp6!ROuy`
za)GaxDRC#SV&_49MT&#&U&0v({IQS*k!j%05=mYWXRH;};Mb=k@#Jn-cGk@p_$khV
z6NT6J8<Ov$Y#ZA@hAHz9jpPGHXYlaF4Vvd94SG9~j*9_eYpXaz{)#km?X#q+2G$3?
zs_F=T`UMarA}xsJSRUuOMxQkhf{`?_yyb5`17ZDInI#RIC+UJ{AqJBpW3}-OtX1F(
zW(m!a76Zwkfqhnv?gzcpUtPSSdoi#>g>MaU7(wXj+~$AF1+R1cI}HZXV!PeHjNUr?
zX}*(@owPuJ$?@HYF1c*bg5h!WJpOvYzo%Y(n&cy1v<;A5GdN>VS3}TpHPs_pFLA-H
z>;33}L|KIqd2O;?>-EtZH*q7cSb=Edy~td$4j3Bfa+|InJgT{bbHvU|D6eWeZ4o6S
zMZo@HelQ*@G^v&1uUs-EGefxYyQ_j(N#@C)=5j;X8tg<9(`xHpqfST&F2?FxST8Ls
zRZ8w@B&@Ayq6JQM^i&RJ(B1#d1t>1+)AD1bRwJd&O0dj*s`*akyZ%f0&}ic;r2R$3
zYc^Dy&JhRzqG_wmI*>3#9g07n*}(0#v-)UMSo{(_=2pJBAr|94ZD~Q$o!}02{5Fi{
zUHN=8rStjm_MK5MxAh7y7K>3#NIS`B*m%|~gD$G!P{)}c{UKVXT1fFNaoUi<dne%K
z4xT@LG0q`R+v}ncvK~vMxkDe4gcUP{v?FJ{u(F+GW!JE~931alVDitVM9VMik$;JQ
zT0YJiwOx7{ULS6`Jk=9wMUXr5-*D&fXc%XJxC#6C#9O`eUm9aIzAt@-_Ij~KVu!GB
zdRhVTZ-h&D99PS3-9xhyz7dav(zzDWuFjy8n&w2WsJOWJESXWhfck3!O@HIK#Xb6V
zanMSr85<uAdW)0Kt9iI8+CU+J|Eqy296OW%-)wiq;LKG*&$aQ!{ZgK$i5?{Y&ABZ@
zYII4aRe|+{FGS+ZQY$S}4`>ndg3A01-O&b!-2(r5l?_xT%i1CH<xv#{u#yjy*dot3
znUkBZeV4OJY>Wit`dS5hEFstl5Q$q61E(HQyoo-4Cct#V;9KSc2>a%6-K3;(6rKtU
z-X<YY;74kHb32ZYJx%!KPh)VS;^Igh-uQz<7AMev*Sx&E_nN##gfW?-G)tyXdKnNe
z3JxAItS9qlgrdmY>{ZV@uI`I<d7j7-eat{#ets7BOZi-^&F%eLTXeWW!f)Np+{hHD
zj9=?gm3Eld<P@peS<6|KW6N{`y0hq+8(({$X|CXthILe67KG3<ng|XnP&-!~d8`b#
zE*3ESqH4C4d4a?NP`b_EP^PdO0xdA()sWem@S~#d`8$}|E}8f~tb%gvorHOi!o?yC
z4v%t&dm4BlAcsg7jNQ*wV_#RP;lEKTt<3FZR`U?}ga-Ylmpksv+)B9f8_<*l1FBl<
zYL9nrJ(j+-=XM@0BLGop1LmqCuXpg-W_00_EL|&&D>3rbzg*!^yqC-_@Y$ALMxaV9
za~kb%iJoJI-$5nS8H?f*_s^0IMGG)q4r`iw%or%dG+sBa8HcH@yJOuqQfH33!OPiR
z;Ew%s+If_g-W0KRT~3bzn%0D!-9Fz}Yn1CvCtWkEg>hg;S3|OT;Y+Y1q0=%u2Us6z
zB4@I}htj(J0taQR)OEd`^-duiy<yzrlBliT6<qH31i_WcYExpmBFe|p1?gyFK>@9<
z4!R*5p`#{SK-iqc<6AQw2w7gml)^{q$;D&VV^Jn&qX)%wxv1%MZ#*s(Yde(){xL{F
zYt6oAR);u{!}tEIIVsjkdE@iV5U*pF;W}MzK>>G2$h>2pa7Z1pFF+=%(#foQYaleK
zXTVpKkRXleLkg3a$Ex4vL?dd+%U(UeGy2WpDncOIdQXUZSg?Q=^QKO)C0}G_0jUrT
zOC6@bHJ(snI9^$0v}Q?c^nMs<2>rI~KO;~k8cFeW-*U9}&>>I_TU`dL5J#6%=v5X|
zY4xa)K?w$Sm+5G{9`!W!b>#+18PBtH>i12hfCSv=OD@CUxk@hw|9aaLAH`pEce#@S
zXAhN?!5J$5?p<1OGFvJP`^cEd+3mY+tqt5ec+ni=L<<=+E+|MFasZt58lL&aAKmIq
zyhID1vfMUTc;h8tm#uRc#yrN*k0;a9^ATBjPt_BVr+;@*Y^YlsGXeGrqK6UuBx)Zk
zC00|Xwdwlx0em!*<5M9Dc`~&O_^@rJc{@AA&uq2`W@9>lnHIEICBcmSL{4!4+lv2Q
zfE1@EYZIR*k1vJ|A|;mL(JDIjyWGthBc=OC{P!~G-f;eG;hiBwDqphkgXX?%NeAGK
zWXY_$jy|S&_!aPa|Khqihcf)Cj-4OTchk%EbBRjns=}gVXJyN)gG-z<Rq(2K;0!NN
zXr@zNA5Gz5_v6ltfMyYxV$4FxH-$0m5CcI{JMn33pX@?-9vg{%tYW%08MTb??4Ql?
z{Fjxy$6r^0Kvv5fN-8S%3Ov$p@D@_R`01^6DR#LBM7IGm4is&-T!;4mec+ZSoHhAw
z!Bg|HIuH~N%EMHuytiN%zoi55iMq-EQQ1Y#8rd6h_zIKeBF+gU<@;)-sU5llE{w)J
zEEl^d*1e(+cdhc3V<ZC2Wk$C30c@?=W5Zz@olN5yC_lI28H$nxYzMYWa&4nTE~Iiy
z>?qasl)|?54eMbDdr0yfgh@jQ%}U!?q^b-4SXVOVRFK3m0k$;d7Q1nG{^@;R>Ef}v
z8Yz3PZBE>3Y8UT;3rTmZwVxG7(zGII%$aBd$4{*?es)NN=Z~K3TO-->J^2*R`Ef6b
zoD9_H1`(m0eUM;kqL)t9j!=-f-`qHjk2T==rG)VR$fuEmE!75!WH|rJQtem@FQ_|l
z1Cz~w5(JN#sQW=rPYBk*ZL``SW7JgD?wHy9qZh-+7V7fHy)p}d7vo1jEuq5?@R+8-
zj;0tqc1s7VCb$9U){rJlmkjnhJ6ZdGE{cYoyWxF{<xJ27{Si6JZ1db5cZHoK8O~)D
zm2^7+R*9Lf#0En-7WU+l4dy^9eQz?8KS#vV&)KAy4zq9H7!Oti=qQz_f0+CV1}Fod
zfe23T!}`bWL6Pi+-q#I3aT6=`EgRx05N=M%uV4E?od`-D27Zi!TE^L*)uYo!abqWX
z3WzzE1_K;9b7Go}&@L)X6`Esl|Mu+r!xsI^qpts=J$$5IzDbu{Y_4j~ekjzrZIX)t
zVIw3^!0@y%!`Ah%rKClBxP|4D7GeJ)R*_R9`o3NcMu|9<kS?l$X)j%5;D5JVqjlEl
z%!`b?n!VY7nkf?q$wu(Et5@LJkK!t%{`qSJ=1kZ^1qUc(4d#Kxh-j^C>HaL16RS&H
zRI0Y#^=Fa;38#qbNKj-zO1R_EzOHpS7wwNLAp2D#k~7DDj}eLa@EP^@Iw;e&|A{cf
z`kf!?SZ2X-7D*lq=HOYBPJiA3IdLk2N&h#7n3e%KpuspQLNhAFRpZVx-0bqPRL7v;
zr_83;Xq;C;eAO-~@{HreJqi}{1$vHsZ5TT7l(gA^F>NYQ-r4bvVZ%<);FaTiEd~OA
zGP3wK@oMRrp650r43I_eH}_dZ{pdfaIf|yE?qleQ=f}uBSYOW>RvL**bVDC-*brfU
zFl-Pf?s})6^@dqT%|^L0(v9*g6d-q1isH5VbI3~QzYbKF4E}P9=-o(cuyi*u#cwq6
z0ZoEv?4*p{hH~vuPqF7G;)L&eah+uNIjaYe!+yJW)`GD!RH6U@_Rz;fAs*UwY6-I%
z061si2{Y5Ue{Iw7sA8gUu;)PU{GL~nqlKwwp2CpfS&fSe(Cq8D_I&U=kI4q*39bHq
zRMfba(n+=xJe4)T&om7~@Kk}L6wYMUqJ9Ya)_7MEs_oX%L5PE1OSXGP9F2~S=3bcx
zg1AMk;jSO&^L->zSiZPY(n}Fa|NZ$UDLI*4+`U+H;D|tyAwFgxUS@{W>d!+x4q?)7
z3b46iS-O#V_@8EVwrR)=hu`#p7L9|fsQ6xSXas*;q7YGhw{`;Uux6~(`E0P~0cfHg
z1WxhtUk=QlH`1}PS1bCmUg^y^7j4ls*J!bvL~9fN4w2=)bOUZ7l5B0yG`$Y&zYb_d
zY1!qtV(=lXZX`&$Bva8jsg2gg@p`0NtaUD~7xyCFvMop!O09FvvtLM+{FIBz6IU@+
zDHvvmLT9Lg(>x=J9VPlW9OnwlfrnQC+GC+7tDK;&&Ib;*<9KU{iwVM%9HO98QrBMV
z+h<z=h?m3&WCqbmh>Z?LlhI>`X&4x6A7f9j)5$LGo2xXi3@%JlF{s!S6=$3H(=cdu
z7AO=EnfB6LX<Hc+s{4Rf@m(<hqqdyx8BI=G2w%++3xYRsv0T6++OXpwQ&R<;{sD+T
zrP}D({+KZvA5+QHW`R3iwy6CaO&wG+pcqrTgGIze+V2l1>wP<HpzB3Feilm+an-=M
z8r9gS57-j1%B#RpMwZHVfncF%p8c7Pr!_JL_z46zg5=~cU>@BbCgkYGH`>{YuJD|o
zQe*@auvBiFt3o$*wD$IfBN<}BDb5ZX&88y{XC5f<P{)MF7C;FuG!_i41rBAkYt)<5
zh)~7S-7ByD!6QFB>@8c)IT}x&JJ8yQgiY0P#RZEH72dsZU(>zN>aEy%G+Y=$ovGQ`
zRXT0G?bx`#aydR2?S15b8FQjoe>29_a^r4^*}2hLoNLb~{47kUpI)Y_+ZDM`R4?_)
zjg0iG-1}#RV6kznGOWQamOK=(ngW2q8paBb{xF23KjVPAGacrB7=@h^*e@99=WJRG
zHMb{+lj~SF=sQ8f7BDD%p<=sa(|^AYA#UI>-`cjym}LY2VMhUcl7eD+_SekR;KN9@
zUY&;@M}0qh82%L>w38m}7uK4x$j*LfQ)GD=&RHqsF?|dJD6`3d`~sm`o96Vm+4Kr2
zea<@>4^1Zhvj<7{KT4M3z(cY}4Gb`f(o24Q?H_>lNzL@#i3CeZG~z$wx|PhGrqcAh
zwWBtT9NSE6d3dL6hJ)H8X|AsUXaSE!+fWp)7TbQ_Q-VVQrM~qN>6NZ_?jcdDJmx%q
zS5$o8sHtBe_0r*2YnoR#?j2tAbXh%Z2HW>BuhUN2)Ri$y{-EAHOyZN>-<j+CYNa23
z8V*A^1^2kwLWve>rb6Ddp#$7|2z}#IZ55syCMyU6XQ`G=Hjh6MrJyLFFN_`4tsNM}
z+qc6c*r-ok%{i)MY7HXBV4@Pv)rp1AhMegULmC4bT3SaQV^vDyil-F9W~jo_+*9NS
zNq&|j0FwKj*yK3rCKpnEc-<haOQhX)b%GF6#2~8YiznZ2CRvA~(qN}A1eDSxuv;z8
z6FbSY`J0!r`NeGsh{X~W2{iI%g#6^A;d{4+yVIjKeQn{w$(-+@N&G_{aq;n-&G^Hc
zv2RP4_As3@vTm0GrwH+2@WQzC)+GjR0lb;I9TnH;6fA$nyjjZ_-eLna&M5b9K};$#
z@<PlWpA*yCH^TxfKw^y7w1Cyt?7Ca@R6|rBfm>N#sTz6#I~n}c-10QcM9qn0C}GL<
zp0KyoH+;3i=cDQb(MWA!=Ag%3dH8=~Ic<bQ?ah>BTYn`v{3SzYC{5@A`Dj^ac<J8;
zBw^ZVphEB;N&!C*b52<<2blQ9*mlKs!#Ug|8VYA}or@<rbr9N?E{1oEfCeXlskpp*
zT(qw^d_)y4uwS!jYkI9;)qbo2-9QjEI|I}PSJKk5(skbnX^UcLqoyG<$ft)g{_DfA
zD2^m?^9kdnN-68n7iBl>escjJU1IXIPf!d}x@UM&gm+z}Q?)~^HCA!2h9FOkM3?oz
z9}WV0gt~UQtIO*{ts03$nYwGbvYKO%U--)vtF{RoAtdxUmV9E;&D9eppgS}d!yE?3
zjwax)d(>}28W`$Y)?yLoi6LJh>3G+V{9_gpVJ4RD-Xtz*mt)<F55D$S)SgUW5Ykm?
zcDrVu7+LpkD`!Nwzr*YdzV9m&v_wM8ZjCKq#&~rx${*gs(4?<t<`<%%z@UEt`Ta2h
zx}>2*z9_1m5F1|5Y#=sdf<wl6IyC+?50K-B&NaL^A#c2IK`53d&zm94wUx=f+0@q8
zF*MjE-FYw%<Vjle`Jm&;Y>_|>E<k1`_r3SuBN~39*!Sl*y1_TX#M|IMDrlCfzmQJ=
ziQ<GcvV(c?vlCgg;KGBbwpBc<^YaZfZtgBF?DeJ-mN~7?!paW+s^5zT8Vx%}k@>6a
z#l-1%_Qw9$4-hD0OND|CR}}xJ`-Hr)XVm#B7~{L--76o!MJla8UJa43M4?EQg9C#2
zr@j3gLt^knME*sH;1yuD!-+Xi0sw0uq3$pvGUF#2OuB~&In;hnxFFQhf=uV1<T7<E
z173IdQRYm0-c<<*nC8M%?fm3u=d?4~hf`9>Dx?ewI2r`)4?M5)0@KBcpteRG7B5Nl
z)MPmpG9#wqShPsdD6`<onE2%O!5=&~_(3tpyi3lCjP?4x5Fi$1US8f$t4cw80y8EG
zk$<YIkk<_^MGX${=l@-j?EMTnYR=Jd7}_6#MB%|i*Zp)<WSJ*f`K#}FwpdBLk9_5t
z((qTs!6X0HG;WRezU9vx@2E_c1&v?L4IYTIFY-N0GSDS`06)YnTk7{+Lvc$plX0No
zeiAyZxUho>#>{VgMA~4qJ)_$ni0~Bf_V#2%Gw3<$-|hj%rN^ncTd-jQB9W!XRwS$b
ziqJEU&phdy-`e<vdm)f}nov0Etj0LRWO78OdOr}@5LUjeW3k-Un-M>c!N3q?_viC2
zA`PTXMP4h8O4;Z<jz9Dk&gFsYArAJexp~*X?I;#}2+F)3)}*f8>NvB3xTGa{dEk`~
z@WT#rWXceOL3zH>gxgkL4$;B6O+>D2npR=u)||_m6bjz{Vm4j|3Pr<7GjO_EZ8?t*
zVLvr$4;yI0;vfH9>svvd$B4|bw!|pl1#}fnr&(9kf-A(QW~qF=-%Z(Bwzgk}($c;Z
zZ6!VHifeERaNxigh=~&*=m#Pp%Tgx(J3?_?6S8hVgC^dCd&n^OXR3bd%x^1-#^6Xz
z8`jm2WagtIF)x1edy#NAlr5f2R&fe*5KGeVAX2?nq)<X3CbjC@X`TzbQ3E|gHbYr_
zr6rz%LWgL=Q?_4LLN*qNMU_1}noa0(tQO7OjPh7;hs<K5viWlIde=C!1e@u~v_%T}
z!Q>6+(ADr9gQ5!W_%UFI?70obe#47IRwvFm3zAaEQDfHi#_&_JMIreU%GMa+vmF>m
zd=D5Z#3JF3)Hk2v=Pi20+s($ZMdPf_%YgNM?bzt}v`zBac;fB(u8$D6=H*WPDAnEF
z2c4?AaKYcg#!8iytu*F%KGajp(?6uMoiW}lEb;pIVQb?bp7CXK!cFm}Zq<1hCirgf
z5sHL=yimU6Yq~smqJK8@CM$!n14eki8X6flm8%Yc=HW0~D2W8q?nA<{8tlwUZZzA-
zXiYJtoGAbI-*ij=wkX!od$V}Uh{)TCE?wF=5B*Rt_Wdohf3**jq1KSuZ{@ig64Ybh
zq}az}R>$2G4mB-f(bw|MJrNJ;K|`h2FMYa^9PgQK&zG*3F=W%VX5Y3pVcerHa6OpU
zre5bz{BfTsJ3X(U^`R0<Un_|c`GSdiv47hh+2@;`Z<V{d&WOLRPOf|RCh)D78*{S2
z*YGP*ap#8N;r04qcrWhKt+6Q-h(db#&wmkd`Fdke0qu5++zh?WUE%E;Lz#Qvx98w@
zh3vK&kg$53Ae!MCjt_;e;B)9ESF*keIcj`ytSg(VNNXZXei$BMBCNQW9JnrT)?Aqk
zE={M6`*k{cKB`UKZ9)M!tkoa6(c*27r1Fd^@n-{cbg7cqjjMQJE`$Se2ffz}SOy!}
z#8~Jz=HRNJ2dK52*8xDgGdFW1NI6#Ud}v}DwpDXr6Vl=}xL)1v_oWbw=@;{LmElM=
z?3>f|?%MtI)Rhiv_M5h#F1V3x$ZEsqyl~cokT>FT@10NaZWkuuy^F66E}I!bD`Mpq
ztpI!r{Ji)eV6S2|eV31p%i3kLPia|MJV+z~La2BQdA=DMT3S50!{Af5)84BZnr~N2
zn&rK-6N|g?864%&jd4F#ppcp>!q8hDM=z3aIAbxk9U;4Cn`}zJ$OGg{)?mDc)q8g!
zSdrGtNt>x#eqV3UH65u@Jg{}Nuhdw$)lY-9r<_Nwy)07bKe*;xz`PKGN{-H$3GObu
zhx<8s`5Qj@-J6#A#YGM+@iu&w@?FE<f~UC!F6zwHa1-<gC}lK8Abu;WqotRb4*{i)
z>sTaw@}zk!ct#UJE5O>aT%t)+%+)WU+f16YPTbK*8A3CN8Puz^TxtAg8W!<8>FMNQ
zdin3(QRIl?(6nB#V{jaEK2#!D{ah|bc)UrBBKuUVmvL*l2{P?8hINhE-;R5Dla|#U
zE;{$MD@HRQTChfxU)v>@+tMWpG5^`|en9i7{eS$v-&$HA&5m}q_wB!IvrLDfE1e~N
zPTu^dTGSKph|VO^vhdzq`0x__e%mw6k?FMHKhbZDP=^-kU09U&kmAg_&BL;KRR<Yg
zYhj{NEh9ag<JoS0IqXdR=PR5Y5G^A8N0>igx}Nj0#xwt504)6%*4UB$`D`h?;-%Oy
zKT~c&;W+9G;U<aMRE)lMs=L2tes7sb!37DS2te|b?i6}0*SnLFoX<oM|9!kJ`OU8R
zCsy)%pPiI<Gk@iOJ#^$=$&1FX6qz4(5T$F4R!gKB%#gJ+#pfOZCJ;X<nnNskqS>2U
zs?Qym<FLCM&wXamBp&HkPTaaPlA8L=CnjYZ*K|GgKZ7n@fSh((?DL9~oHTSF9e_w4
zZx_x(zUqqmlMpv%r&S#87RR)UD4}|ly%a558o{^odSdK&3@|1EYfo&BZdf_Q4DR*7
zuA;>QRD_=%h^p2hS=<3>^dVx)wTGeZwjE(+I;=r9YHt`r!>teHjq5Mvo7!`^kCvvc
zD@x|Z)X=8j)!KE5bd2Vtu&$gQ=jP*jTOu*hjQ^s4TEtA%h>9>3qxYT2lAF{xJY`=o
zs-W1dgYi@LZ6in`#7eN?Wi{&e1}2m#mH2|LrKCX8Cwfpm!m=@0WN5_c3+1TgMM##k
z{XO^CA9BqfeC9jo!q>l~L^i`!;>Z<*fmsbIBsLeQP!zN_|G4ZMqaX#g>e9?F4W4d0
z(i(=3z^2ULqyNazxH!}e&Oq|u*nK%)qyt{M3fBJ$PkrO!Uaz@C12`MxQ6)$Z{hcY1
zCw6&UG5cbnV{`3<L0|~8u{-yJX%*Ij10Ew?*-o~J-BL#b#GYyuw~5dFCtN0~7dV8z
zZmktgBr!6$d>6$s1s?+`W`f$ku2K!xlh<(FoV}-(I4&ob7%0#HIu=SEg3I6;r;qDH
zDonKE<~_QedB&T=V~;SorGlBayZ~n+|KR=|yZ^)l(sH~P4PH1u)N{c$Duc!FtTrTk
z<z@crNFiq+|4_`{6Xq^-M}i#?Yfp?AS#Dgpe#!|^JSCxztR(cUu`Eh9*5N0cw|SUp
zW8J86WzaSmG?sFc@v9zmvjqe_enB~EhY{2y1IE9Bz+8*}Btzsc0zb~|Wg<{zFJ7F8
zd)OgU8q9nhOg<Y4$qEHTFosGx#9SGLjiJc6-6I8qPC`h(#R#JPGiTb{#^A~(Y$EB}
zGB+G#XA@easICCdu4uj<_C#*1?~ZOd-|v9*=penDoyRcK0SO*@Z(fDVl1tDuf22hR
z1z5z$r)PkBkmT0SNgIm3u9n`ZcnJkJg9X*Rk+y9Qi&|f<4-0B`<sg(?sW(&7(b?#<
zv;$otjcw8TK{;A^o$K<j=ob7y#TN(TU@fvJk)IGmZ<6ig@M={vuZ-9wk;R4ic;7^e
zcV@0fwo6)}Q6$S0Gz3eHAsQ$PdSmM7_<nWhX5@pxShY1;Ft)^cwb97tVWS0Sa&uwz
zg&)6mT=8?0jI+bz?Nz+k+i5#rSpFLcM!BM{?&p?q*K@URxr;EQ{+;cyPLkgevBSkL
z?}Mhpqa-0kt}lmb!?Vbdnj#`koaUMou)V!UDux<n#0(vCg(eqW84*SlXsSH*tFQTu
zx;4l@Yn?daw^o5%qisQsl`Ngo*0gpELjZwg3-bxLal1gA?jPR9%v{W(g<Chn#C{%p
zL|@&`HRexfxZI`c|E64&ZrdAbU5xs)rG!I~TmLB-(uG0bi^UMM-YYzT^u)-0k7_qI
zp>XGW^J_~l%eOC!F}(9fPaG_JW+y3DVkvRIz0X=|y^n$&ltmwP=F9vffzrc>hqKH<
zFBh%DH)qW$cLk{_MBZ4Xo!5^S#P7)QQ%!?it7s0>PY@Ug+!5O4LmO^e8)~9*P`M8L
z6~0`4_P@sALtcX$jrTW$z3@`(f@*z6H0gWbBeG;UHeB@X8ES2##y{~AIYf$%1lBJU
zA;l@qr+O2v5gU<Sz;k@09tlbYhEwiq@-41)rPwPr$}A+#)4O#rn@&2in;tk>uB@IM
zGkEyU3&23f1%(1VA}1(1^+Z)Q8W9w`_%@%0VVfHcY(*SQ<LIc@;tL{olUf9pla@q2
z@m(j|^Q%VDT>dtuS(C7FdFWG=cRMn3=D3dghKNb`#q3(miPhiI;pq0d33|Njj2p}l
z62EW4X6cx#2g60#YB@>%Gy~zN(ew3A0k~G1(;AsI+$LLXoPn_&K;8DtlwiY7o_RX1
zMirQ{G~Ubq2f@n%l7`VwXnw;<@65GH9{=s39dVVd`V7sD5gS29cl&fY+bhLnOZlk2
zh(XUBGwi~QwJXO31?kL!o~?Dh>(9r6L#AZ*5lC*UqnZ`r+)eFdvaSbkSZkQ?J`uu^
zY8#^?F!YH=1nHjbH5bQ@D+IkVtey_e=EUySvCLR<ji-cVDk*lk7)E@&kCeLiQL#6E
zHCKPc^r(dK1wfEUso;1b;}Q(hj0sHB2h!?@->&2RBE?T+Zu`}<A0wvTN&e*w%ge+x
z%%ZSXbM$Yv`^T0|!nOX~WfOndDSA!kubN^1TW@K^o+?TS3%c)~SXy+y7#bgHY{l!R
z6%6T#_g;lHqH58ozWAR#ie{r2+m7??i!X=HSL%y+@i<%TeG1IK4x83Z#D}bm2SR_-
z{-=1q_~$0i^@n#VbI~=O%dl(QjiLKVpTpy<q?{^BB7;GlaA3gAx6NTaAgmDu)XgP}
zk27Xj(Qf@~-C)y^k|$t`pwn2$XL_~oM313$W-#&S-K(f!%_;z9d1ey(BUzoAks8xK
zx;<pbcum{K2eQ2Bgs)s$5#ZTaW9hzXTHxPkifFkbu%dZ&(Amt3d<g!3a{=;e8gK8d
zNK<+q<yN$F1ix+%MRkm1=E!6D&_$CHn{2l)DN0_qxj*IaA!00|d<etiSbz)`5bkdK
zBk^a*W_tf=^_NkI30c!J<gH@wR<$SH7;}8K9k7`veF*Xv0PLH7n4ZXB<*JpM-%ehM
zvG%N^!x5p?b=#ko5!5v3hPR0(R0CH%)iaEC;_f%P!B%0%Ec5b0;sEN|OH|)%^cOg;
z*Lrg5T*D}L4$N{9t445OX^2x6EO)bg9ECJ|W0P!<1j^l7R-WeDmxoFR9?|ipljb{`
zzy0>TW8NSGI4a()&3TqBYNAJoX44gEIqtlQ<=Ir=)1>|Tvb!4ZFDSkB{o*evPc6iN
zl~zR$&l6FHQiNj6H3C3WkuX(D0j3>r=Wkq0y^|fVuYCTm+Rpkb>MmO2Gjz%zHA91>
zG{^wbASogs-3`(qAq~=9(o!NKoi8Xg2!b>y-67pw0{6J?TK7M={KOhqi{YHJ_jf<f
zKKnBV8B2!gM=Rr_Zx<&=)fG%&%$<+u;nW_HP5CmU3`tyF`O``=$*Qa$_Di2$B&2L>
zo+URlqkBsXr(kP*DRnPR-ae^Vr2N$2Yv)J7*+M^-SQ@@fY%WWWBV8(vRmGV{*ULcL
zbujTa;m-51Z>_Z;v{=E7F)5U<oO(;-XRQj$%Um%seYK)S^!$LBl0NU6%Bt+~pf}nO
z#i$JrwkI7Un}2F3C|$Mp(;%1M1ImN*E6Q{f6;o}hHn&b6Y9h-zA0N#uS=w&%2Bfie
zBaCQi^?r<Ek<o~=*B}KFNf&QUCb+|^47~dg42HT!Pq=y0-UVYf2aHSk7s_>7qKmca
z38ns0qCz@ER+W@MP!%eR9f^c=nv7;OPQ1}_%NfH`86uCxjwn%>oJbsocXx%LqdU^b
zG{$!;E_E9xn&?g<W6Gb2<-zst>&qG4+xd5G@ie>sB9m{C&fMljhsam>K_fcX&FXls
z@EJv;e#+n!Q&ezAAI1?pGszVb6nxLb#PpXOf(fs_PnY`Lv4oBM%GS3Nv_6DV`=pLE
zwR1a|4e)Q~tMHJKIY6u6Kg2Zj`a3m+%I}-tCTVj*=pW1#KUjB<vF*z6*HZTPo3*R=
zD`Gv>e?2pgrz%LyD{od9WFbbOefP$jlo||a?M*LrRLpasd?(TIycCFKG$sIwfLlP|
z<nJ9l4#nhH^LO)k(^31j36iRUH(p-tl5_4xiC?NhVO`zb=7U*+=rbm1R!xT7Kz<aK
zmg2p5@#4^@^^nNAZ*Wk~$LE_fL7dmZTkNy5Gw6-|4Y7<%C!r8MOR-Q>B42n`7G-K`
zD#`)*uLThq6~&j+*wR96;&+4zq37m)z@bwa`XxcNu$J><T^-5b;2;oP(iRqvo{EYV
zoU2>f^(Hebs;VMmXQ!qj8y5paH_tXRk&cw>1DPamDo(w{rW))i47qjMH$*5hpE{9h
zKV^AlQR48|bC3{nN4pylaj%&z`}4|v90K7@-2Hk|2O%hZwdX<~3iB7eqA>B}lTBuF
zDt6OGGil!sgxu@g7@V<5S&@u5ScuQ7``-U_d^p@!V!QIXR?yKe_+{j>Q`BiFEd@Th
zN}B{`s|q#x8=tsatt>;*<S2$mW=r_)_7t-3^D;&aCGMmPQcnXW66(+`%ugqk3izFB
zdR3oH{DDB^wAAMxug3Nh6qgC8`0QsJd)Mb-v@PHXJ3%2`U0r|rYyYOs?y<<p$!)CJ
zNDCJ{ym~2zynTuGR^{Fo+iluj5#1S&+I2TMdO%kQLY)Js57k`!6y;NJAt&lwFgFfu
z{j@^|N$(H)^-IHHrXDvN=@>G*_i8Ic$eETl;L<EZg@v~p%q{Sho3zuul)QcO^Tfu=
z$_#YBs&#$n%(b+RmfjH+78PM2vk9q$oSw<chdZ~$@;S^k5&zpYxP_v!^7p!7iI^r~
zi$uMW2pui0oTnamz8CvHPKG5D08K#5)I<hnzx_Ud)3{}d)f{dy)#ywIq~Pxj@sFu1
z%Mo<l(PSKc7qWrIT3W8RXOk8iLm&KeHw@SAz7{;ycm|OPjDeZI<Fz^qJ3MmWFH3>R
z)hxf%*PmG%F=RVCK)+UQ;<GkYerU?rGNP3d)8llwUxG2@vp^-6H*JPIQh8rW-sbGV
zLBztm<iRDc2W>289Fjp}bbPYYQ6s;=Blod@CbEtVPBPn0R$_qF=7hu}RBqaNq#G-@
zu4`KU!Q`Wn4sOGLgomXVHRPtp!u!TsVFkUP3ctNam7NB7|CnvcwyuveIoPU`s;jhj
z*z=XBk;ch9o@~DpT)w+GaGvn_<hSuSYWr*Pf?RY89eO8S#G4kOEUq<CX<QVX%CiN_
z1=_3*A()D-61tz=^EC7r*SZ93gAlJpe~J{A7spZ2D=X$8*gC)BGF&bb1x{^Tog5#V
zA}=JmN=pktlr($p;jxp~UtRL7_|)fJHCsZsv1;q;Tmjz1e6h`cs>#(NXzf{GyqZO5
zmVu`ylAOnwntr{0mHkuG$&l!c_~h5V?bjq9g(<UR2+MlH!QUtpPKqqx5B=5_iGmAQ
z7V2`QEtr%vPdA;|QZ*`Ouz96e#-UpmA7g6yx>wn6d@!2zKyVb(a>%GYY_#5_qsc}n
zMiwNhbkMPtb-m*`gJuz6d>%$~lZ1XG5TlZ((~<G%`^*}uILH!eVv{p2JlrK5Fn4PC
zabz<scr}i|L@3)&I{yfno5?R2W%E@rbrJoZl5wA9zPc+nDaq_|K2XOa97{0i9&WeW
zeWhz6yBb1D)p=zx(Zj{5TMYq&c#pedm&vPuf4aIkin6P4-x&I_Ir{Nk?yx2QGPWd+
zArBzRxPbmIbhv?s7zM{?7i0Sr85j4%|6)FZHP`#QsS)CFss`ZIV7a=w`VZ~&aa47U
zeNnH6g+}f29l@=llwI454U3sOkOJZTg`!D6rKERn56INtIKkR07ajLeaE!993h4FB
zLnnl=E7s+I<M3gTK60O=1N<`5&EvzzEMpB9AGl{|nO+DzGP8BBHfYivk?a}d##}p-
zYR<l1-+~Xzg}29G@_M#oqV4lAgr=sa@8J6q2INva1lbuUCBk}I>NG^w_qw2|LVM@V
zCFcOF!7bN7Y3cRd_+8cguk%NtOVJ#aC>pOhs0@x6Em3q#4Ej)JVq(zSlXXlQAqEDh
zj-jhu(0D;p)6_;G;z2`0LrF;q3TS!)K1%(ZaL*T^usPQr8r}aKrmK5WnQ(;`{pyyp
za#UF#(2Ci@UV@crS`McgY9>nW*9iZkN<&MUj_oiD!J}8K*g7G?dUMvU^XJf8Ot8&a
z!yQ^`@T9p#42;fWJPvFQYu05+*wDu5_Exj4RxXrf80woB@X0dN$lI!p3!pJjn2yb4
z?kwo|Qdm2UA-Be8``vbFWeoMW2tr{yULi)P`!cKJ*P@=5*7&xIupd?1P8uO3-}-u^
zLtlv7pKKb!B4?|N)oox^G$UpjyVpN3Kz%0@hOg;x`on!#JmNivj`?)0m2R7#h<3Sd
zFz|3zGHr4@rt4bUeOs-TK1<%RJ0Ph*k^)Z>OFd*x3M{0hr6ox2JuIZ3r=7aDw>Nw*
zx^O8@XaRoSP*NIwv)@bvl4a-Emsfmd-!Uw0Y*r@>oPUgdQ~)%Aj`>GQ4*^BPq{FV^
z+<konCR1%Y%4lh2g{b@tQfh+uxh|@$ao%&bx|<4@C7v=W-6eUeeXXoZfFa8!Cy<Ta
z<B`Z+6bY+Dn6||u?%!%eXplzI)Nxm@sK6NxXy8q|@Oz5J!qA@A81_7pq{*vs`=ojW
zo0|{X7&!(u`xtQEEE7b4p<6U=Zf;dV<6uoIG5L=z3mH-9keb&hxiTLwt#OO?``ZHp
zh*VjFYg?-y==SIeWl+meD8pYF3%0#72d*Fqt7?8z6C7Oqmv(jKTkzT8a+v={wli7P
zlg#{lWMl*eW94&Qd8U#hOdJZM7WE`pTU*1z$N5CMcz<_c)}W70TeQ}nE~`%3Neu!o
ze=U~?n+x_zos=yaO)pK|z#s+;#ajS8%jI!DkH0oyTt9a+!K-2w#Bl~o%D-dhjuM}#
zwZfOUTte5>)}}0Hr!aa$oW`l=X6%pZ%2c3t%EHv={WgTbaqbep*nfPKL>+jtK1di_
zBM)g|?VGMeeXx9=|BX!8upeFWHpVTJ%}ZxM1(W|F+&tvWiw-Rnf_vw$E2BxicMLa@
zK)Sn3cB}QmJ74+IfLjyKUN#P>{qC02vl1Y<)p{~#6EyU65yMK4n}>_b3Id~}8-|!o
zkHE`49+h&f`K}16u)NX@kqth`n1Y0q*$sX={y|}~1kM-k<<~)LS=9%_$nD+TE<nU0
zz<-STEyenRU%zSBQ`F7fU7B66={1Pb(RuOE_;~bCzVPmDd9gzj`k{}nBn*e4dfFjF
zGVm^u;=aEA$)B|~L?wWmJh%rF=Oe#<tvt^hH_)7@FqpLItAt3I4xz1+7RQ(DFvPwY
zynSKq{{;K={>He`$q0)fbc4T<8afld<F&q3G9A?t5f$Y!iPAa4SUYQBZYlijPmZgI
zkTX3FT%*IfLogXLUk2RGzO%|Dh$w1WsNI1O$vf_$sXi$c)=*HJ?dFnP-J5GD`kMYk
zlgH(naM3@B>9v!OkB_c#CAV<p7cb(U2}G7U-z6C;<4;q4x__+*X3;$8DRB94A<Bxp
zUIiuuAq*`P2Fl+{(yMqHQM83!F7lXV!dzHLP7bw7CjSq$C1rKemaFw_Vx0?pJo^_f
z;><)DFn$*kR%l@|U|d~YwFT+L(TIBHH#CrymYVmcaY;On2Q(wg-iMaf)(>6}91t)m
zDzRe82c6UV<~b8I(~r0)si>%2_^)11Ty}bMcv&%ZxukauJTd(RZ;SNw@m}v)ewyEb
zXblK;2#QZq9Qr(1QP}c8_$`*LyZeQ5ADh$kS6gZ?uDCeDSc=hk6B)%{JOM}e*%H>P
zezk7$l#fSku*~P1cF6`X66Y2JmgyfAQ{h@8LK6egaI!tLA_-~q3o6b%R{>6mNo95e
zBo%@zrP?=J8l8OL%?lK@ph;O-F+ZRsQ2)vs3zJzo^KsoCV_!r%wjmG*S)ZlVl@*tr
zipJ|9IUH#(FCpW%Yh)(*dp3Fg;VG<fYozsP_8a{XDxY=s13$ZTju5{QkeyoKl*dDJ
zh&#^+3kwUEM82~cd^*(R{yBo8@mGE*Prm*@NEBM_cYF`Q(66t`<kL<5Mrh&PRbMt=
zeq$Wl7#JUPsZGl7?+1Im`!~Rn3FF-BXgIcplHVw5+>RNdrlIjL^K0%f!N|9QM7D`Z
z9QVGmaB%m-yVclRtQ&jLM6+q$sRbB=&>mx-txt+dN>G#l@^l4#N&DZvVN|G6I@O=-
zaCqCLg-=aQN%Q;-g>f1-!9+wv!YSz$;;H!}SrdBRRTwm48JYwHNUW@OlJe1+wYg%w
z?k0N;SH+%JQPicoBu4GjJg>TFH}GcX<b2YFTVG#)nswQ2WT!Hms<rg&5q_jJuX=7A
zpTPTchJ-ps%lX)46x?NE*uuII<31=-kP02yeET>l9A5^<oC`zAwPnX~q6r0mpojT5
zTTA%N{2k<VtKM8}O?MoX(6<V0j{d>H@&BUo32DS9;wIY8qwHF`OAhN(FBBpPCF_Mw
z@6M+i>~*9Nhv0I=YVvb#C<6v4=I~r}*rG!|h<eF0ODU6J0JzZ^F23X3{{Fu7a4;+c
zJdCh(?U2q3)Gl<uP8}Jy`g=Ig+uQLy3?JJU@#y=;nb{VGxG<aYJ4Yn(C$^baxB^uJ
zS9*41<5U&|pn+iuF>TPFe$LbSCX4T!5Q3t{9%vf&mP$|Wu^FpJtO`KxSn`n3lXHg$
z_75LqJ&mh@Ta(D<2sV`zbWNGYK}+1R<~?+6E7n-S(Z)9;uyy;wR%IbP``S>Jg%g_W
zlTG2%x<K?VKEsnwesm$Yu4|k)9|vPB`Hp7bVRoLX8$n!rj{o$U5!Mr%TSS5u4{J%8
zfb~hH3d_IJ-v;wAFr})r^!V&))RS1gsXfoj8|8+ep^oUd)KpSVqi<uT#IGG4<y>8B
z3Sg9<NXf{^!jCsYVHtzga{)Jd8(|!4-p@sIa@m?SBB`Hj->lrQ&|pqQTDuXTkK1*>
zd&F0Lzc{^9)xNNI<n6=7DG*4ga#;4ajNiEJMv(^ODuT-QiVS5^6^qX+o*}Y~2y8?V
zGxLvU<_R2)2)I6=GUHL$7Y?jTv4*IxMDO`>wjh1mT7L5?%j&U0B2Evh58vLBgMp`u
z#TlWfsHpFS#AuI4Kg;R~&Dgt$1gVc^D0sF|@v)HM$Uc7?7&--Kq)otX_Budv&cyP8
z{~n&qWCqHPaE%3h^5}c2z@q$A^B>V<Ks@meKq0C*!hD-275WW$tKA6y`BV+Dv$0oG
zH5S;O2MclY&0hZ`(NvqisK%>T=;y8LS@ugV!bA8SHD|hNRq~<qd8<hkHZr%G%%?=^
zIVzW?mw_Vf10^)Q>`W{y{Mm!ukfqj(g&U%7c>C}e8s`F*Wreg<5?&SBSUoXy<=Wtc
zeshSI2n}~ERBI3iCTL_FT}mDu%@VR2N7_S$tD3_gJdUB+kQfyGAr@*P6ftCqn4s?1
ziOP|{_tTSA8%Jwu4c_T)dLI1g9%IF)?7Od6?mRvPw-~)r@`0C{ni?vUiW$dWS4dMZ
z+Q`_~kc|m#-+%!bj>CZA+;&x5g-SpFtCA~*0OuDqu4Ce;t7)azy+7aFDG_06lB!|o
z*30buCrJTKLlwLZfJaZ+`opIkx*Z$m$z29OU0(xIC#Ojp_1m{^fA;rx)>`!^rl)Uk
zT(6m^pMv>w*O8r(a>k8l*xQvli&3iB5`V17LF2r;$MV%t)37heIYWVTUpf#L<@xz2
zibe*dhs}V<NSU3LwGTzn>GwV`Y(+F}S4pWTSEsm?b@U3fHVDnkwNy{I96dJtdD&w}
z@ZKp?$P-`TFJQ%ejwmR8U(oX4vNTtRi%|PqE*ffmN8WV)h+g8k7vA~aLaXoOE9J~_
zJFBCkBNcah+z?p{;wPRK7~A%|Rg=v^a~tF!Zg{(0mQYYl9i1C;O=Wga9Ryr7FPZ_p
z5`OzKPhxLpiBc83_>hDoDK#~;yqpt64k-}sPZk1h9944QQ!2Y%Q6y{}_!=W@7@&dz
zwNNL}GH6+qSLtsr+W#|svEbBlBrhwcE`yVtoXi_|=Le2Su7LTft-GkF_Sk9rYBd9y
z_ijU5ogh4;bfq|dt7d~~pdXd?k|%>U_~(y~f`r*brl`SO@teW(^YgP24D14_^QTMK
zk5Ie!gt`wy@_v>b&QP&Z26qbnU7SH054!V`hdCkIzU$tRM=!6OqbX3;hghP&jM37}
z|43!~hCjz{<$GyMZvKBCLI*qxfXaS>N|3-_ze|dVr-fS(_QcZ|v^vdy)6X6(o^e33
z7_y<$H`cs^UEOaBz132N^Gc2}EO#kuS@wK|F8w^+`rT<vu@U0HM@I;^u(h+3QBuNn
zYCY3llu=>-u{~afe(rzJdO@%@()lBXr_HuxZ+Vej+r%`6FPy^I8>+g2Lhoby)O343
zOQ?<A4F?d9SimNT5R2yvFDhdCS9$%n@#ha{fMQNHd-32f{1eeftq1-@R#sL`Kn)!2
zHy^C5rJG2wmJ%#9YzS~(O}>vzV$@BHtu?@MtEoE}7Cu}{S^OIu6st(k6~tglZ#m9<
zVV#<Asd-t4>g(@Eyg5TFevy5?QXm_#-IG9fPa%Ud{JTCLoBIac<)|tsg$KPS4}B&c
z^B|D1Z0>u(C%=b~WZY^c0`e%-+;2j%XPEG?ayK2N&k-?krd>8r_dCd&iKC}vMm7*<
z4*H#*YV<TdDF-~$VGMcEpjy^oJEL}%_pa)1ir-=X*PVqZw?-2f-w(#cTm8~P8>WMK
z##XfP^hW3Uzh?^)0W<SbfB&Ei9`)x1UT=MFEH`ewwvYKyxLrXho>rU`SUqfJOOUJm
z^wT2jCoaTg4DkpGmx*ZY5b)QeVEP_}(`65laUjSRTJT11oO6pY%f8z)^j%&2&(Yu+
z`G{#H_6rwV8QHSLYoMT(uQu|CT#J{OSXr@8{mo*RgRyB*0NYA(EZhxeQOO9Bjd)<T
zt83wP1dyf6w-<{F=H`t56djl<q6hL8Ai#7R?8yFeoJF!pld^9cIn6blShTp_Y9&#w
zFo+#*2Z=s5t_?8nNPih1@II(~Kk$G!m@lLueJ;Dky@(a<QiD@TU7fgpcsSzsmgW5X
zd^woF_;=<jfL!4*tTydfC68Itx79Z8hv$CbzopGzmgN>}n<s3dx`E@&eOuPTF~u39
z&RFyQyN0@m@g7d=NUn!mUOMHQlb}twR-x6iW4NFjDq_4Eyu~NzeNe=WMlA{DPdhod
zHI4OXGOYYe`}R-LXXAjYHEyuNGkczbe0=I;+Hul}vs2QZtd1BR_t2n6dwl8!A9>=X
zBuT};$MF*Y-P#@~xxKR^`wdpVn~)%JU1Cy0JP<aLZXms=EJ6Cej!{+1F<!uNN?!iQ
zPF>N`_SLG^bA6LNSR62CcV`=iCsUZG(o0KAs;YV-=mOwsj%R?W_9GORBr!QTWPRNl
zf&%%(<?_(%iJ%~6lf8WM9&6drPM|#fyU4`*Zi^~%0w0Mzo(fE}4b!KbuEC?sU`>h7
zvM6s+kZO?Tt3l5400`64&Sf=FOnd`4NN4N8ERWw?zd!5ztZf-~A?1Uz>!9nj;{3@&
zOWQ@O4eX<e3&ZbcTdIu+vN|9FSOdT1_HAcWA(zx&*f24HJ2kH>v&;B%sQ~8cKQERF
z>99#CX%l+t);ftKTfQ&5h>@$<4rK7XI~x-xF0_ECiQZwDJ7ETW+^bRFF&ByHvHuxc
zL&WHMhd)u<0e@G>CaaHXShv2Ly)`0*0Cg4Oe{+u6q<Y!{a)Q5gf$he5l}>?&<9Pbh
zi##1(xW(q?CaO1u)fHszf6AE7|5RP7QQy&F&v+~0bLw?g<<`azcqb5jOP&{t9vOV$
zd$Wxe|B3-$S~|LfzkmH&4!`4ry0`-b{Na>Ef*F8fDL#LW&sK{3CFZ&W&oM^adkn7S
zFGsPz#Y%9xYb>DFRa47H)Aj$;yN!)73q$9&r;*0#fO2$tyb2Ixpp@p8EzHjka-6D$
zLdN>~ltBTWoSBK6-v6VrX(jps<dk;ji~V^WFdlkWHtlf7z)*A{yFiU?Y1rSM@%^TB
z>;Qep5A8`Hv|a~InH7blHmw(p**&!WL7k=%s%^0qsF$|5wZAW6hs;i4`(%N`Py<mB
z*QRVNE!`_EMfTUunKsTkGlp7*sBvSk+~FS%dM=Gw=ii~DeuqOV4m3E45^02}6my7}
zcp%+ISQ!osMCw`%y?dU{)9r!xb@r2$x7`lNGD1Q^P62_GEBv^S5HyqON>j)mq^GIZ
zuWnV$_AhVr6cj9ej#^_PYUUx;lB}$(jMi2m_?nJcD#byxv=HZ_iD|^QZrCSMi|*JX
zLCW9cN}qot+=Aoc;=<;gh$vyb@OS-d+d2-m7Tb*g$8yC<Tu|9_FQ$*FDW|@4<n4EI
zLBUT6m{XwEI3Ls!Rj{|`VEd$~s2J31$v-v#wjAz)S2gW`Jbgdf4c0<lK>@q?WnMf0
zN_OsfK4M|PN|3x0j;E6p>Q^?*<tLZBcVp&PAgZWNG0+x6)@{j)U2p&AGVpn~*O^C0
z8VTiOrQV=I<HEQ+8#qJo#Eo2{hr?t*G*_FU13km^_3PK7qN32J)DIZ@`H#HCG}TE9
zOCL{z?AG<K+<)QVuxqEhMS8knclZ=KDHK$Pw^@!gHyG4Nqed2#1vz1B%SgyYV^YEA
zOdnDQmz(bV0B}yM-hD%r2%@H}Co7AdoD#>yr>VsZskZ7*t1kmKs&g_ECKjykgV=h}
z9(<`3DrN)y^!c<&Bz~BI<-LLI3QJg4iX<(lTiOA|Ykbn!*KtyoNPJ9{-HIh8N{^Gg
z;a|ec6i|puNU&jMW-hO;P9!5G<>m6Urbox|VN;5p-S+=>I(jWA>pUusLmuzGvPkU#
zQ4_Z6OC?N#uZECOb(F(Kg^@Xz!>fjwI!#GTxO{wkF3t1d+zlIyEHa1vBCihH*Iz1V
zIBW_@ihrmLkxb7;t>*ckX{40-8Z~Xaxr}n23|U!(4&YpTl~q%PU!|~CE<3viAUEdj
z!nQ!&!VvmrV}k@zQeIvJP}g0{F0jVg7Y$*C>ZJUjEIL_Dkd)vM7r%X4HvTPHBWwHu
zM^;vAyb7dOIsoB4H+DR#tf&ATFQ1VJBgr1EGVN*5QCZH5qZHg03QcTVln)OFhLnx!
zXLlPgk39sm$(=z;l#V~B$xE4(RFavj8}u^=tu6HQ^z6X1hJ%wc6bFZfHqh^)>qD|g
z0dIHL_swSf_|AR#;#R8I8V+|iH;L#jIOy)ca2Ro+iK6TXA(dW~79#2=f{Mug<qHx$
zs}j2%)Lj<KZb?bUP&B3YTio3ZxP_OTovPKapJ*ogAFTH^KSFEt=$R1cxR6~cA{<qc
zMA_Jyhl+1D%6(ZhBP`k;;V^{egUte_b4p<$BfrfM^@?Y+67U+5tTI8X795i3?`4)O
zOiUyHyh9OSOT?I)o0BUhKS|ffiY;g^oUi2!ocs0bc*V$TzQOkf6u!W-l4rL{X5xd4
z+?^?;Sry)v4IV|Y=E6@t(%o+U2Xucn+ox$9T7svrZ<<;a$KGOj!XJi+l)+cP-ASGh
zu}lU_XuSv)F=C_~A-?g@zMuCCzz6E9r+T68t*J&y_NW-<z&fAjh$OuYhJn6wBrS0O
zk%Sclf*$|xGXlt{LV`cmxm`Z%$rP}O{Jq8G)R+wk>R&5_n2eQ`l_|No@k5)e7&tg$
zB(66<A^@5A10@}uq)~EFFG~vhJ6N(gf948Q7n;$5zApyRx6-?lS5IB7sHnh}yao3$
zD*?Uh`p+NBO$u;6E75|qT}dFy&)?4e>&IQT<?&g_tC{_A5#9uivh0-HT!}^Ny1_Ii
z_H2zTJo<PIJL_0th;es@f!25(#7>-c_;o^$du>C*Ss6$oms3Ez%@#)CLpE#bc_9N}
zYI{2k>`fUYF)!~RW^m+FPmG70sZ=CkFue~>jcYuy${R?KvG*p02c)F3k_D2OmWDNW
zT<{eMF$L2szW_a01sD&4Uk$m>!1!z*aE{i+-PZ{qysaqk`_>O4I&Byt6A}&@A==<j
z`k1Ria@7^7A7IZp31LF28XI%GLAr2rbK88cWBhvP=g&l+$`>Nuhxv`QwZABSe!|5M
ziotixMoI?Z6WohzZ`(+9S>s&G{NLLv^k5thQG`m>c|w8~RtlnsP&Q{$plkfn1+<d^
z1%7g{^4{2B>~<aoLQ>!Fox}nI6KKrC6M28d<g0Clg)f9zH0UEx6`Wb95kqFk@F6jp
zEU#K9h>5}o7}W;UXIUDD0XSWUCQ}Xm&!0c<jYSBI%+EXOj&VGGtY~e`!lYylyoI*?
z{yTi{1dk9xPw1r0B&7AbVs<w9DW#q2H%Av?>hQPk)6y7)gwmm{AM}0gtS;A?pM{%N
z<J2&~o)Ix1hXiEw#7=K)Cy0U!r0O#O*D}TwUl>C0n@}2pG4R4UY4o7{M<7HD>luj{
zNZxGh>}J5L92q7I8crT{r|a@zfIpORz`+=mjf^{i4`$`x(Ec;$e$1IfLd3cmB_$WD
zgLW{nc>l^7N;c-|GqQ1V5{R${;YS}IyRAs__z%)BaAPe0->L=|H2PgS5|t#aHkx!v
zcr}P@_2BcjnfVS6cj?nd!YLfSx&=$oM?NVa-&AW&E|jho38OUG#FwrY_IhfJi-87y
q*hK&Dx&{mf15Ntb|KE34_ta>A(zx5)WWh&#K$PTEpH<43hWrn64nkW1

literal 0
HcmV?d00001

diff --git a/docs/assets/images/readme-light.png b/docs/assets/images/readme-light.png
new file mode 100644
index 0000000000000000000000000000000000000000..bdce8a8feb5d10e31a45604e7438141b7c51750f
GIT binary patch
literal 22109
zcmc$`gL7o<^FAEgwl~>W8{65~HaE6y+sVeZZES4YHYZMg^L)Pb{ta(e&CIDYRa5sl
zr|<5oufBRB6yzij;qc%<KtK?GONuIifPh{Aj{{(!fbY8(iPXRkSbIrLClC;%!T(;M
zBX(sbz>~PnVj9lMc4p3QhK{BnZf<T27PeMS#)kH$40ev@S=YRHARr_lzeR;q-2Yx?
zdw8oHJ$>(-Y-fz+fEdD3SmU9XD)#%78(@OATEe?jH#b*r>42uK+a<Q_bofhG0<>*x
znrl*)X;d|J`YOr<`aAX_#}frfP=uE=OpbGQh&hrbGx=D5x81DJGNxyoY<ropwCA4u
zJ>fkGuh`%HpUt{4C))T2{m(=;IE}9U@9c=f(0c!WmLGQz-u!>x5|5x-LH}oA{ITTA
zNdGe*0*NX@g8%G|GggC#^FMoZAgFh>|Ibbj{(t^Ka3s`(-CiMKeR!YC-C1H3RoNEA
z|JJQ5BpN*GrSbds?<`>*6%`CuL50?(c3kd9p&|&8sJj2H{}el1a`ltu=QuuxZ8X|R
zWMpLB$#RYE#AF65F1xkHBBI^>;TZjl7)-m9ld~kAGLb>}$dpmA+fI8r)&KoKV=$a^
z5vK&4qpWNUh>rVSNKFQ-xmgL&N{fozd5Mrgnwy(tV`5^SD%ERiy!Iog@Hnhj`=W7K
zN5{vN^2A1sBitIHZ#L-&J+noKDMf;FBOLjdel8sLLC3yWXa4UQ_{6%C5n`NXX5!s>
zcHe(I%*c;UO=a)7c9v$Ac*KVjkdTp!QefJH*TgSCsWijhf_2hOyh*8Yu(7lEsH>|#
zPw)SW5_sJWLI%4V>MSdhB%E_Sr7xiyJOJG3s!BoJgDA<;`<#Z)xK*CFfRZmqsU3eK
z9gdo8%mr2HvXnj!o8Zr_prB^l4YDpY;K|B2*jU7cQs$<pEi3ezP2u@9*qlkg1j9cz
zm`<c~IhClWsNlr$zn#_%3=FIh#hv_W-S9zUB6F8<Q&1HxVnKoFI)8pwzuIhfS{ev}
zm+t+zm!29QpKrEWv2<{#CtpY6i%m8la0S_cumwee@v#Af)AR_~Vpn)BfuLOkV*4v6
z+DIVZFD7yo@5r>Ejvc<-{N>rahdvI>p%~rEg9f!2YpH~X@Sj1^W1BS>XDhDuydR?i
zA7igzM13qG@BZ&Lwx5+`u^dlkICX4X_4NIk$`JI-qEZ(4wdCImS|zRkIS6S6*T&Mn
z1lXJUReo-Th=>?xw_3M%R$1HAK|@O`9fpWs5?gvTt1R=4Us`6*!$iM)u_CiNJttzx
zj9?r|zU-6ssZ<r${ztFHqTMg=^f@3X$wuBTPl0Ou69Uh#@p>#JWvqoN142wo^d|#D
zszQSAZUA&BvG;9%TIbfav>1iJF|1`z;MO-X-~+pYzk!$|IcNu74gC*@0t)yLWC`@<
zT09q$BjwMaQaF(~|9lDm47{1iaiG_QwOX$^K6{*Gw=SXDgk8_m&2xZD4B;EYd|13;
zn{<HjP|5w}oJYOrnehFwtgp>vm%elK#&+v7u!0NC5+gW=4?Wc;8x>h)f}aGTCA^(a
z_W9iV&1bjS<`5;b&=ox@DW!~P?FZY!5O&2Vyw#FZKO_^dH$nsH6vAvS1Q6BU6;W6c
zpJ*W30D6NRHl0Qtmr4_1zw0SX82{uP<%indAz+v&eebAMxdn`u#bJLa>Nzc|K!n(I
zxk{74PVbm;=;#g7Re1I*7`p9sRi1;6)3X2U;cF*pCWUGvPZTASx5c0z4E%EPdV-;o
z&gbjnRp7~;7e0WL9=Z&&bK8IB3qkuEv4@Y(PNff;XBm!XdEgD7^w~cBg)i!Ehg}jE
zt!cwJ^VkiUbJzd%V!680ZL7mY8jC@vt+t>*7__pjt?l1_4##Z10PCkBl46t*cuz6m
zWPN(3GO~Yo7X*0O&&UJKXMtFOn`mlF9C>hP%8gWh!D3K-udDha{3IJmJG2gV!gJZ)
zm;I=blf%OgJ<*aD1T_KgC;3%(=u17jDAzDX2%JztX$Wn8`U$<>>q&Y>Qy2pGe3mC}
zQgU9j+KQ~_s|~j!SgOBuj;nQssRyP!qAZ`x!<PQ|_2i*u=CPkg{lzk4H6y>Kn4yW>
zciBf8T_xd0^xc=;mYBX|sCxUr(@C+qJT5)V37&@0XMF!!8+%6l%(n%U*xkGHs-x?C
z&H57_`<|rLwPv?RSqOIEv+%|r2dH2dwe^__Rz((o!e@f*K8T2*FS4x*#C)xiT~`Lj
zU1RaTv&wvc5_p0a9j)1kn{4I@1${kjd)@H8oLAQBaab)^`S`0UDWPFl_TM!;OP747
z7^dq~4yX!Bna8=|j}J{k`O72m-uBtd&hT9%Vxl)#fjt!>v?2|%^jMH;lC4t&(iBCp
zNQR(RM5EyR#^!ILU@_pkIAGwsFx!0tMk9$e(@3_nC5havENc<Z^wg)b{e7bhX5-Mi
zN!FQYVm8agATmemiL#2g-KomUr&D+C6J|pNZUm_ynK=5@Eg?{S^vj2VZO<~BX_7cG
z%Nj@b2>V&&{y)0Sic4{T%QdIFzg$lIFg2!(+AT6yX>WJ3%lXPO9{!}Z9O@)G;3h6l
zyvk1J?G{SZeck9(av!*lg`zo$Xv~0YCbPe1jj{cU=nAodIJ697N)(tP=8SyRkN?UJ
zPzrrtXAw53xj>mI;RB2@xPBXO$0r51Ht0KTLYi+wI+`|Ioem-(m%UQhTQ(q`Q46#5
zm;z(dFfc_N7N_ZPGxPAgJD}nst!8Wx(}lSCAqIrvP=Piq7`RiKl(vMndJPGZEZKZV
zk2Z8)Z@DNaT6yi|yFbI&eGv4xPwo#v2mk%uA4(c_czoQ`<$AR~5zBR)eq^;$(|kOW
zJ99Xe6ePhN$mb8MM(E^qSz)}zx3hTpw(F-4_f;zIwH!FQe<Ci|fN-EA+6&48I~vWW
zA+PJGrJ1ZTQUJkxu8)&#4*6&@`+y;r=zbzg1U}u!e?8FX7ja$-jM9FBqSJn#bLKgc
zErfjj{@Idhii)I|n3%)bcHJ&|;^M=^ux6eswgKwB@I1@IoLUBw+|iSvO%Nx<T9#3+
z9?kT-32m%<r=7{LX;PBM?tLc@x^B~4Cz}hivnVjSBg3edpd+Zy2S`Ou)Kbi@k1Y~y
z9^-0Aqqk7r{5y5!%B9yK{4~oPyd17|J4fG8sXAl>3F~^O5U0r+VM<D5;#-P^Bjn3&
zm<*ueZvT$;3<M*d98PjQkjUhJY{dEwO@sMa`uZ{CkdZ0r=+J=x?-&~m4NY|Gj_>>Z
z;^Ja4nGx+xZ1e3nq(U$G8^l_<0@J8ooE&B9z%-tzL^Ddbq$4_bwte|{{j=mKzfEO*
z{oOM0=cz=m=WRdq=fla28-$lXs*=m4nJ1?Z`yj&Y39h~YjGgRX&b##82Ok=Qp4(SN
zV?^fO(okRQ6lTfJolEWjICwxH27?|Ox5ScT&>nhm!x-EpT3Y+VG*_|LK{Y-mW;vuc
z=!fH8<7@9AeOkEV4<Cu9cfJvwU$uQG5B0!3G=oz=b<_#t{&OwVIYdLa+i)<H7zk80
zL04^0Py9k$1OogyJ`B=ejiLJEqltb}tb^l#kvE2HI-5@D<<n5e)MDMQfeTMz+utk4
zv*7gIy7R8=xb67VWb?RvH53$()j1pt8?d6-Lv-r1e%k$YTm_|rM7oN3kRP+?Ls3{_
zI%$-5UtZ}=7Z|ooq*ZB4p8ZTo_}PiN=3iV((q#(eoTL@wWyfdbc2$t_S{bq3u>_+P
zF^&yaOVU9$F~cKfp1d@|?Gs;_8i%v@+d;RGMxX@$H`ta*pZoHNuKhCi?y4&^x9I(j
zOB{V}jx~9Qpfro7LfMYzMKwGt3(GaMPTuWa_AOxGRJ7)=3Qn38Rl<`hjb)0B>YN^6
zmwNmXQ&?PRLK%qGIAn79KVwI$bvXmKZa6Ou<xn<~`13hxoz`A-|8jT1sq($%1tDpn
zd-InY`5Z_?wttDY3yb-|9S-BF!FQ3}#^Zgdr%p2LU|K0aKWjq!%fiLIe7?1IPXFA%
zv&$oCIDP#PS&G-j<Wofx7$HV6^7()?*DpF99BV1;orf7*7HIKy1IC4)`R`>bHM&+n
ze;b`(?Ddv%bp<H<6H<NMy1IKgy!2Rr-aC>A@BmUPd&GUnz3a8_-Ued`f8icig+3~R
zlv(DLlpK4F5V~!m3-#-@SZ6GKWRkCb`k%=&MN$%7OD@&t?7D^7P%;+JW##1X7v<&U
z{fJ(e;zQOgG+D!#*~f(3v9R2y$NVHVf@v0nv=1|7Ha`2?z-pJJ+(DP4;0A`yDY?5s
z)Y{efN31`4rDi$~_gYQGE=&FG`r2`2Ep&A0r|MQ0z;~odvrz}j(;Rq3Gi_pW(k+PD
zq(2w`{rjh+tV}M?_p}&XRaF%R3->hg+_$kJ2GiX4qVsd~j93^k{yHE?LQrkUcD5p>
z&NFtuQMxB8JR>{X=O&ZQDz>7mj0w0Wi-1X6Ui23scj(c{rYP*wEQWvwbxhg0pKu6X
zVx5`K)I(uurslFu*(Bw%iBP%Xi!=#|^fjlkHg4Zlw?P<9JB&JRs!pA~YQGKHmjFYC
zt{yhmb?wgQIQ)n0-2-H88L6MRi#TICD~Sa`#*;`uKJY0CkOS`f3xNkgl6AXdz*rr9
zQPas_^>{iL7a7inyYq(7<>BJw6k{d>4i~mQX7Y)cMlC~7<6Y0=P1t!4zIXIGUjChW
zjHI6-5{BgS*z<nP0~n{Rty*n!C`?XHR?yMmCF#ECYb2jWtq=XNfduYESeoXyy7dr#
z|CCc#N-i^Hn~;}ArZ1hYq2FljoS9K9z%J_>F`kDbWGZbC6RFeaJl*ol<2a2rbXz?-
zfvCZvBKx2$<OcEtRuGREZS#dI_wmfFll-IX0)M53=4I`wG^GdIYS!jKYNy+5>2!^9
z6H-Squ>kE8q4f5*o8m>DN!F~8)>i>74b9NS(>O=ZU5tLzs3C|-5A=8Y5CbXRQ1Tb;
zq7VWq5&@k4C8bQATNhoIOS9t#@jpG4BxVU>pT}<HQaQfwerV#R#!09@DXgeSq9(r^
zuIlP)e&uN!bsOMP+AASvEHd<a?v_}HUs+2@avrarz*A;YK9@Pd)i3*|;Gsr#Wdg!>
zW<EDpmfLH5JwHg?qz)7#C;#ngB!eqBGCwwHXt&n*-SzzyG(MnUdqfC;6ePV1X+fI{
z(XMFyz8ZVQbDEfpV<*^~UT;2tLVbmFB@f*$(67}b_L^H8i7KuI#(vPgunQt>{^n1^
z-x9(r<@&&?jw=Wn*%y!LPgS6KTu6nyus8?&Q{LBV4lh*dfO%&!p8n{k>~63CEqX7u
zx7@|Wg`fMPqQs{S5Wg+2H(OPk^%E%=O*PC}1Vs0#_2({@tzYL0RMz4-Z>s~5#NYWP
zCmFwVa*XZk-<|uUeLf4l6eCwVtUz?UC#V5sT<f_EpBll%Znfw=xGSeR7R4p`O0bu>
zah%*Ux9lNq9-C<vu5&DPL?;H98P{xKoN<w=+_MVZ1=-d7$7P#d<2Zw4#hBwgme&!6
z<~eR^GS?Nj6pn3XKW!E2;&-zuJFwTMW?JzSyiq<F@I%{$MoB@&EA$Q9Hy$bR2H8^(
z`n?-q(L~L_Fpc}`&+HA1T;xs{jZRtrb|#xbFy7^rE;JG*{~F|W$uSk?c83;@F9AX(
z1c`-(MJw;u%K@W&7Q4;H$Lj;M8K?JG3zUcLP&Lf&ry5MwY+JSRrr=uXKws9v1)syV
z&+yR+Q5d#Iqn~bb=UvhDny;+^ABg)fZ1)wQRE!bii$8uC(P?SgU$#|!lp-W>1muE1
z)(^V||7e`JNS&niD&TLvgAlTX0Q|!_7H*rLoQJN}NXR05m#J`@-1t&gY{v=LI&lRF
zwxiex1`U`^w?<zV9-4-MXFtC6>~=gaFJ-_=xC62F!ealV|H#Y7W#ZN3A@f3Qg@ry2
zO%t}M2wm|rNK=LsrEr@s`rQi_&Di@f58SWD($vd`I9oh8=Wjugl`Lzn&e*>djBxv!
z{+^;0=08Si=(iT+flPSAwS{W%MM6U6I6+1?+GT5N;aw1@^JNK?0Mb8t!K^T+$y`Ud
z^8T=m>#4F82bC8+Pj9MvZTz)q{W)A7IAnQ2-WQ1R+w0@JbY!IJp`yZbqs!*|7s3Hq
z0Q<AejU9p0To>5B-9M*O1gYHZ7R<>nsIv0%gwkAJPhlykad}Xn?{fHxYb0|_AsRzb
z4K5A=V=Hsepig~vZNE8k6Yq&@=gbbU3ZK)fRR;!DikDZ!pVL1pv1#p{8m3N>lvDSF
zxcCiMCT!322^-PO<D#AlQVL5op(=&UTA8hj<?$KXULWZdvqP}x&>z+@Uw2$P-QkaT
zM8lf`$-Jv>rFTWy(&rZ!o$JD1v&Aa!)#-IK?H2}?UX!Gxing9xoTryt|0S=5e^E`x
zd(^0W2Sur!{6iGm=JvW+fDwSQh~zNy69F?D=-CA)N?=U15#}Z9KKFSC;2Mv_b_W=E
z>^|EyUih{WfJEcZe==zLWE*CnVeaqT!uPjQ{ahW6Cl&cJ2X04kAWvhx*``=0Fb(=+
z{d21mu;_$9VTL;{n322$QdeA8$8@~o9rv|Fr=@7maw?S^bP$93;)nR8^YPy9RY9)*
ze%7#;anSH^!z@Y^gUgNG5_S5_s=T?7HckCG1tcZxHu*cPH+0RfHY3jMRpl^Wnl<j_
zwwPa^w{JIBT}y9Ub2?l)H-4NHI6!DW(lt+ZN-axmK7IZuP2r;pbup&eSVz_n^(;l$
z@r}5ZYmogDmDl%Q_CH^$q-x)~eg#>y4T-x!BIf0VXL(mnTeoF$tqYZv@=m$krl&Wu
z&ODx^%5mRs%X8c6vdu=2_KNep)~%#qfbdmFP5ZR~=1#huPK05FDjH=``4C>W#pt-l
zcanS^;capEF!;dIz&oMZPSEN8o0FyzT-I?habf&R>(>LV-QvYwQMqIL4g#;+(mg8q
zdgI5j@;J@EL*ze?1^kgPj*2dWWo@|akrg3UG5!a-;3OeqEe3f8qKnYi(q%M2v}<=)
zF%pgI^rK&L^}SV4+2+C^4)*n3Q9>(D8YhF$SK#&V8*DKRGM6<BiLirnpwo4^<{d4u
z?Q=<aU9&cn2-8Yc(WC2%)Oy;Z@o{3bv{r+eR19%O0K-g-ko@PkzE4pOCSBq@Y^;md
zrXRH?5C63|F!O0Nk$t00Bnptv<%R{qd9DyTwf5rQ5Jcq?@(ES-37l<J>wJd0!-gPb
zo2*#H(Z{7I&td=3w5kTe3L=lg1jA56I*>eH9_z?c0T!0diM^Y1?w~L=lZOYMu!kV+
z`<JVo(A(=1zKi@7b{t-SFuVGfE$2=zS6pK6;XPoad<#A$W|ZYNy}I;vCF5-`o1Qq(
zB2(WFi2srQyF8*Wei<lxm4cd;&0ei7+viMoW6xg*wvf)aD6EE68Nf<6CK4_ob;b5R
zXlS}ZZB2)d_eSFCs@FEi!v!%$Ad1bn+vs0#^1AA4uhNThm+vwe@6pnuQOr4~fzwuG
zAehUU<K;dr638jYdQuXv=;3#wjffYJof{pF!Ovi~S@$3_s=e{K_uEAtYnO58pRH+*
z`u@`x_7Ej;|85$j`7DHY<DEJnuwr&MsQu9PPR#5iz%zC1$btQBNzp36pT4p!eXup7
ztnsBqDWJ(6s7*v9oL4C;g&5Fff1}0khk{ob_>8tvrdY`E^y-cFSOcx|^Y7Q0Zr1pU
zWc)*lnHGkbC+If5fbYk&`+=uQRa%Z>i?-PJ;&8kQ_<>h~JUcc?#KNVJg!$$68a=7X
zV4KAqJULYezuVKo0d%p#H+C|9MCXo<-V;X~V-$yCo99wnLTVjBI>yD*Czi|dwHHWj
zZF!IUvMM%y;Csak?}LoGrRDiQfBN}c3HK%PhJjXAm-ucr%9U-tzxRA|emS7<MBe;i
zexIdQN&X~={Cne_p^Yx)k8E@yepO@pcC?+SJ;bRNp&lzee78sF^u(bXS6_hklbj1s
zjTUR%A3uJ8Gt5;9c&<cLKqA>ox`LPdQIVdM^yH;qfcdqBJakH7M_#bgWBZz}m@xC+
z<jYk@KTOJyR-EQG@pFlQ)?k}J2Ceh!-v@KeV#x_gyZ_4%)$NC<*DIdhJJm@jqLu2O
zKkduExpr@D*Vs`|3nIrjov}W1pX7T`Dv<=|EdXCdKmv>1dX?lE$jO+XW+3zd(()J%
zXI6h*`Yh@0>ereySlMk<vwYuMBguZ%SZzGmnJ9HsUac$mX=lg=)kdu_K{&DwAb)-2
z&yHQ*ibY{fJ$iPtI33G?Lg^LY8nhjKW2i4)=k`QWFtyXQnMScMSLS06y|s9Bjw`<E
zcCBEBYTD9DdHV^t$v2mM@gFk}n5tLdtaT6Ersim>>)QhcQjB8}USgX292`byh%*`Z
zE~XdQNB%nNJZ!Fl(<P{K3KK#{NRc?7at7MEAKaOzQEtS#hr|3AJy7$2m1?#S6ZK<d
zZx}LB;`UMVd8>eJuh~m1A45wsd5ewSK@y-=QP=$TY}r;+G|cR$xFliK*0+4wYsHbF
zoZR$Ue*SHZei<@rXP4Y5@%B?PG){rN25|G~Aa^mNz%*&*w5zl8X4=jWEAq$k2)>ND
z(83kaBzW9yBj2k=lRGLg`_0i8LJXpHs}E@*a`(g}+-Q_ccOE6LcGuMECit~YAap@^
zO7f_0LR_X56c^N!Q6Lc(f~9%s=`#pu>ga4X=vo139dRkC$gSZ8edJ==0qB#K1J!qM
z1y%SIwp~QV<>lGc9@}!^?%IpidQZENruu3813bi-mMzt0z<anvg|_5#lb$5~>b>*C
z&B~Eu_I-1$&(y)w`Y^(%0;fUSA@XN`C!mNa3WCiN?+{5L{#ZnHhx0IA#Bu{~gBEJb
zaUA=-4qtcZ9qI(?vVhX?!}mJJQnZ$8*w>WpkSZ#N*vHxJ63q}oYO@L>AdTt6$u3tH
z>jyg-S(JKAj&XQb9#c)WboZ&NQ*=H7<!Ry2Hvtr7>TqruSJOW|8LsQ6UcRNRo~v$g
z1<}A_F~6rueMGE;GZ2+7vAQ@`gS4H=KZE2pe}bz!^Nffu%R_B2TVB{$U4sR_>Zgj@
zQWV>-;-W^{x&Ut4k$tPEs0RQQ5sQ#dyMO%eKJFW32t1s89}Z_uq$g=}9Gmuo5dNkU
zsfe5jw)wg4D-k-iIho4XI$<`f5J=seUaLyV9Xi2&LteVU(lz?o#F&xWO1c2~-CRZJ
zvJU2_J@+Gl&WMoH+{N46ZWC9J#Tdta7G~@kx|i|_ep!6eUs+KFn`vBTvuC}Zo@Y7Q
zC#k91?4_4vV&-I38~TwY={OfjHX;HhQD?_>F319n)N7$lVnu@jF6WEnu&Ef^L=L@U
zayC5m*6JzQ)L2rcY%kV^ybr<aWncp**-C>xQEvC$3(w3fD>Kv*%d8Zg7fySuakb<;
zgZnQQh#$vco$VLny3x@n-{h1Y)Hp=<e}V8y33T<QupkB_NC<TguyrvvoZ6uvuBW$M
z&!1acOs0dqBbQq$;J#83p<f%jgvOevhkikIVkJn2g(k2@_rqX}FQJ$>_irs|E0k|R
z9oIRGX&~*=_Yfv6<>Mx$bn&pD@88#Bz(m?4L)<#Q>%Yj&%lfrfzX3gYyNeE4GMz>f
zxw9j&FmD|8e|p@dm(7zL6LP;0_=tb0rTYo)G1U5|ebQ&nB0^~Do!ra%KII=Vr|G@n
z_0!qs&QO(?msfi!x{1@ZH!qcP8weRA)T&wBDVP%+ntoNODe@kLd)DyPCIYYZ^^Bv^
z-SszaZ%36^@d8z}t;y8Z#>=N_XS$aRAr;YjAHBe~Y~)5xGQu>R{Z<HSXY9VmY){GQ
zm_)ov>ps%&RI5!Yr6<J+k+>13qMaM;Omw~XTN=sd(}V4&M=G2ju$x;8bjxh-fGew&
z{<EdFk7_4r>7`vb61~6Y_KSp=NZ(hRgqRRjb^)R(6p(yVB7*`_%SbodIfx*0-(}?z
zTioUwpJHNoPwFi%uYrR~6j!Rr(iV-VrKyFJnvPm$x_9CjnB^`Tmkqt@?-_xP%9qdV
zOd5@5G5RF76Ut#&6_D*OPa`>8@7=xC=h#~t!tid^>rJq&8+?c$h=Qh~TX`+QyP-qR
z9a-q}q6T}ADRjJkjI7ADC!sf&{gQoB`^n(moAwSrsAkGeW>#)YKFY#>hu~6)Y@w{5
z{-7`^4yL`me#*aCR{rqqolIv2$7#D*o*n7FC>c_v$4^-w1=1p94Q4=I`Ko5*@lD<D
z?6Scx(%x<pv|@!$hncb>dw+hR)lOmJ4U3suE5csfCg^18LOmvXQ>?MY@a#FZV%ztA
ze$6yjwCAM|QKAjIW>mjwfcvs%#fGF-3rd1ML;KB9g28(S*YrvHIQ6U6R6`jemBWn$
zKfFnI%J!R*#NI7+g5w-`3p^;+hMiM)J)hbf4n##S`9Yes2PdLB5F~BLrL`d%j45(G
zbmF(q6SAQG`dPBIes={5WjJ&#>+MdWTWa1fzzxSlJ0MR<Lb^pirzXw0t_p0S16pgm
z1pau4kU91>ujpl=RhccjOuU2otai#>db*6XSVR8q<5gAt_YEJfet=*frQxU5W@N46
z`ZVuh%4lFuiNXhujs*Y5;IkkFvUOyfE^g4P?1M>x9GX5u9kBB1%ACqnqs{hKR;7FN
zjhdj1x?VAKg#DO0nfX`Q=toiW$W`M+iy_+=OtF42KaX_Ztv~@YUfIn_8A%)-jj^7e
z^($uOn<_M1!~drRh~OZkm|y7pxSvo)7_76RT=X{dS`kvxvGK@Y7ilz<#A=h#PGMCt
zT%2XVRQUZfRM5A^rq*H=fIQaVw@yairCq1ux9#|inIN=KDo>11=X#}O01`HEwy^Eu
zJa(a1(4Cr|KtaLhGj@5;JyFMi5nchmJ*kk;$67DyHg-9mwlsSn$RFvTLt`KF!y4`I
zM{*~Ip70CG3x)|_9fC9SXnuy%cnM=h=aszM<PMf<lkwfd!2?zeflplVTLsKW(~~a-
zyRrScjnla8&&+Q_eCoVA%`UQUcQ)l^X2FT%p4ZWbB@%gH0_P6|i>u6p?)q9~#qI{Y
zux=#TD5wLu1l!kdP4jHFzXrdDXVuaTqFlVf1#Q`*@j(yn4OAEmFFJg1TPIrFbaHI`
zY+c`zk}%X?{5zCjGj>OTP9y&6%@6GSh!f{lKZ5wocJhBPQ^x1L*OM}Fm4X_085$af
zxz0FuX9`la`?#*PSg9ZSP1Hq24AIyt{ehjR^*r-$ck}t2sCz4uHc#_WYH)p-Yevk`
zC;vz`09epc(=5OJa2nUc5ItsWq#{i^DHIc}6$@(sk}39z&+<#Nxv&TpCpHieL@?o-
z_9K@zx)m;+e_ZsSQafP|rBtAu50KMLu^<t?J;qwdBOnBc9V<0I#t-D5-@8vLq=15i
zsFSqR)KuwPA|>JkiSx6eYO$iJ=y-lv^8yfC<xG6^HRxBs&j9g$Q|Mb37M4)xFk4{7
zDq6&7*LuSqD*c6Wo5cb1<*JQK1@`3RSkM3g?0F;_LV@TN1$73atwRD)sp+@sqYg>V
zEaA1SrH;?jc)|gh{?E+zO4qY4VB5`zsOI!QaME)t@;vE*?e@?caR}QoA*6C-TBE6A
z7g?ibM&-%xk9=9kGactu3zuoyj)RtZu$IZ(_}qy>JG(!FFqCgtje4&8IjezY!<8hM
zUNVoa@As%n+~@l{xKLRT1ZzJiaOGO9if?3J6gsfo&5y{UeK3m8C6+?kPIj^j1PFD8
zgHU<RFO&YJt{&NvyQwBwsG?}?dRzTK&V*vJ5vjphZp{ilm)I#5XIUU1pE*><DdKh9
zg%Qj9eOXy*2f-)bBeNa#aG)}WkZt|IFhI)@%3joKMYoBYQ3UMRkmfJk>j_MN$Z6j@
zwqYqvt18DC@eNNQ9nD%b`F|V|-QhRfQj&4=9>J)_=nUec%;k@$0J#v7)(U|m<cl0{
z$G#;AWCqO{U>pS|G27P8=J_yzf)^!h>1d=s2pA$HvzMCMzA!I+HnlGJp$V{nlg?g)
zbVs{BJH1@%8!bU3bUCTTr@+rj#KBVqBNhswD$vyFjkny39AefLw2NFA;{vFk#&iOV
z^m>~1%Ho!5TVkwiMi*)4y5@D58d`yyioM={#QP4~jpxP#i36`u611f2fjAEbmIZVY
z*0S~X;<R9u2=&@YGoaI#$4x>^eftC$x<)D-3t^l<)O!)1T!J`=z06MTfGw?%$jw}9
zZBJwm^|`Da-ao?vO=W-jD3{7<;IZg-Cm635(NQgQV;jhnI6Tifdycc6Ezc3<c$)(H
zHTgPIdK=ac1#U_bve0*jM7&IEekkTfL4Ew%2@|8Z^wrVPkx^7sv|F8XOV`z-q7~l^
zbhKib;`~4@p^lauYTgWGVQvK7JpgJ!TC}9X5-{qa+FZ7mpQ6;_vdUAxVI#H<dA0&K
zth!&#8GZA2D>mfleK>}3A{C+<%QTxB3<&JuU@8FJ1umq5XD?eKy~0PeMDh!t-|tV^
z4qj2ig(lT)miXdRm<hB2Nur6*xX+k+ejhu2+}9tCauS?8Yb}o$aH*tptO(t&OJQlf
zY1Uq7Qzf!lBvrSvH#4GE_z02TiPARRLY|S1%Ms1q=F|M*2xmEe(ykrvUZU1`A1o;m
z`AMk;_uGq$i;5q|Jsl3G%KmvUA&6wb7lQ1Kd4p&4?CNTdk44-UBTZZO8UI|pSR$k>
z7j7^C9M}HGL$9UoxMk6#iG_w|Ey&5ugV@n&alI+3eiKTcT!*J%ZtI?7=&|f3bYzxb
zK2Pr~nF&2<eC!valH+A<Vi&xofkImZK7hV)Q<t-yyn1V11k_x|+@Ql;swM1n!fZaA
z7tR3wRa3#{*VUn@+PIYJqBhr7=U{hG_M$8{c;?;?MQPIc(Fb{qEIWLavk_K25RR+!
zZ1OhHWMSHw?ZR8&?LWYS(t3C7kpDTPhMZh32>0#X!GE(<RGjOZVt_A3{vWtT3hh&E
z9=uqJH>ro}JR1o#y-%vgePtxw1U($VNrij!8DwJAX=>R0!Te}gnUI#6qqy9x6%<)q
zx8~ofdZ^i(&WPygG%>BAI=Q|tF>_pM-f5bur9-!$1AcfJmwz2wcU)`LC86(l{o5?b
zkMD{@_$=s@8@1hLtMFT|24NRt3+caL0S6Bcls{EB3vR9L1?{J7T|(hnV3ia7!zZhH
z(hK^pL<g3*4v&XW$3)5_4$AnWV|4|tO69_;)ngsV${x15lotYWF&1WKnIR)v@V!Q=
zKMa~Zmur48X|+0Orm7w|wgV<^)L6PNG_4#!^`#e|O4i0q1Ax64b#5n}`G0tqp1XZX
zEA+$<nJH*L#gEX!r&HB_U~=lxXHZeTW7X2F8ua5sbs=A>9Z93gLOTS!DO!jYVXcxE
zY^WZgiwZbKbS15u*f7G$;_y@(x9I(^X;VI8G|I(g%O#xxwl<SE+%c|=%-4K!Dm~3N
zV13^!4t8<liW$N!^hI!cakK4tzXRS=;w1r0d{96TTUb(JW~Zl@1Bp{4emQJ?`8t0~
zl%~Yg6;ZK%iuMT#zw8q~+<`oA-<f<zo`u8F5J9Sm1*4*d@7=uhAB!XV)ndN*I{4x4
zDZ)w*DXo2jnL{9^rEUJ^-SgQ{%n+|~<GG_2D>`cHXYxGe42hF<4e9O$lmV$`bG68a
zyTohHtao%5i89m*I6i)+=5qw($XVk<8paQ-K=#!5Yb<ZdNcZ%PJa)`)1r_y_OMy3+
zn3^5``t-dn5%1VTq}R&k&;dVofGN$=bi`wVQ+4Y73+9&8O5|xd;{cB(GjjBf{Rg^K
z64tysayzA?8*l@Zmz;Ti{{x+FvDu&j`obsKJ}G%Yjv`|?Tc07ixA%+bhL3mV$AI|;
zuXjteRVR6={9}ti8SM+18LY)%zxC8yWmOfzMiu`qhn7S<-OV~$PfnTt>EWd(0;uZM
zq9@xUGtj&DAnSbBSYD4e)(VkSQz{jvMfaccp%j!q#2z|YnrRYg)eg-%s~4$5si`8E
z;3r&yS6AEuH~9n~T1vP^blsPQleJIXLw#>Z7o+}(_#$pTEu+u)EUyg;bQ!;&`fEtQ
z1PJ#N6cm(KoU59aPB01B{jprPKP+%K>NfN&7y~K-YgZ~cww#rZE^2%A5D|V@dB2d@
z4ZYPH&7^f|f|wrfA@Vj<JpUy@eOWi;i-#q;`1B+{_R)k!yEpt*qut6NQ}s-h50!YV
zOJ(fmV!l$Dj~wX||N6kjM7)ifAYG%y#Xd9S*HZQu-1Arbg7V7iTqyY_?h$68ZVB7(
zJ;_;g-XdAdB%QwE?B5h_rx+JRx;>9W@H6h?N(ZDN=!wQ5J+x$QFDw%L^<O>j+<)^a
z$nLkVwB@jvdn!*eV7LG)LxNd6S@3}{8kPzBEFRy}^SWU2@NcEm>T%a;e7#iiIaNR%
z1*ShCk87E!rha+}SMsGW)20ENU$89bt+uOoM;Bg2H^4XzRI{9kuJNiy$AQy1Fs<H0
z)|xtw&S;et<CufR!+X+Gny_#g8q-DL0<yQb4S|#g^S&X!XHJg=CUKzyt*FBE`|G*a
zYiR9shGM-l=ZU1;($TuW_6{iDQC#v&n^-!l3HDK|3i>s};0EGHN1gC)6-UUdvdE`P
z&nLHM9{?_fLqp?Wo}b3g$?tLqW2aVVahOR%PQL^$dT0<L(x6Idcy1q(iX{l%$gI0_
z2$Up!qH*56JWOX+LrMQ>ozRS%4Zp7=-F#g){@z~A0O#0zGmV`%c)Rm*v+hP4U63<H
zxUdf)run%2<dGJrr|aIUvyu?l;#zn|^6l0C@cj-eu^@^uXEKQmIhO@Q{U<3|RW*)s
zo)E%vw=#62KQ!x0@1KKI@%H`mr-5qhET_(MFSO6?L{<4dZL`g_7>tU#ix#Eh*^#>r
z`K2J{+4=78`RbDn9{B9+AFoM^>(liUQ23ThRhkRo_tW1K?_0{qHR0<;?P0CFNDyGT
zRr<aB@i{~Zs=d9;K(;Ig-<MWcm?rDW$*D^tf_nAW;)1B2B|6z~Wk^U!b`C1QqaLZ4
zOSA`GW9>s^Z2hjBtvM!RKT&`V*x=z&AfJUvhbb2X%%YlbL}bul;bIj*JsWLY>L--+
zj9<vp=_6M0?|n4JDa_Gg9)sI3%DEyzZeu*kpUZm{G-3|(xA|NkNSh>V9DHH%p|!01
z4(;#dZ?oEfnqXnze9Zii3)@Z-s@p7^IY9@>oq66#PCnLOs4LX|hiCBK^FpO$;Fe84
zjvlr_<R*e{y#~)lYBdZ7RD>~$^cnkII>nVAjHNZcxI49(+k%)=f2dro_kTk&VBdl(
z2MA~Y7_Tt{iop`B(2_>pfh2?|dDAe6^%lJ9(_U-jr3wxYj?xwY;;>OyVFkx8leq5B
zHuAD{6}+Q1ZHh<twp(8gNOPA<0_c_pEm7!38`UmHjdZYEHK?4N%AvCsm9~-ieC-f7
zmKbaAJ=nGwEj$!F0Up5U*6Z4R{4;cg$j^+={thUdV6Y)baA`2LZ&|gupqzW%G2t4R
zNF9fZE;(|gs*7#h^WDtpu)(J8fDu8uj;T$cVMVzpe7;!+JXom5eD&(JiOQ1#xoLXh
zejx5~2TEa`l$4aTo~PnCQG48*oXo4Of@Lw|1xCMCi0jj&0=f|u)RCZL{8g}R(;OFm
zc)j#jdn|foxe0;xDsU=m_fmw>{spf>X`8dW`eKzfj2sXK=ePD{V4d1eOH(rn_FUSg
zJX+A5#R>mfQuo?B$YwH3Z>W|kx4PVnjgMxQI!;f00IF1T9@njEtGQnflXj5KEC{`Q
zzLhRTJy!-+ygu=!Al}v8ntqs3eCkyZLQzg>u2yFG{iLR5*86zr7M2D>$FL$G8&qr%
z+#)N;L&O>;8lqvPKTx)7@ooC*Invbv|G2r*k6}Z6Ub1bQ)FV3ETFdc|xDR{2*<?3f
zXDjp*JTbb6$U6crD=R~(U7^(m%3_oh6mEkJaYE<|#j}M?uKKXV%?88i%4~ugXf~pd
zln#RA-I}`%3z<RmFGGVQ{7Gf-Hb%d7I<oL@@Iy`eRV}@i5_Br<BXpfOqYeAAv2-j2
z+siS0?2ysngr!O<XtG>Ku+B+*;rZ8ol!=>obmkkxcT@H@Lb(p5lgILE#xH}!OdN3=
z0+TBvu+XbNd|Z(41!i~3bo#<~Z8c7c;m({#?1uSGiXxOwkfcU@5C8@imTUQbK2RP0
z83<HeRl@KOew}$hyzT<#+i~bv5LYHCxL9}mzNc+67I3ys)+1AslO>rTdMuYJMD8CR
z)KRQmE-HiR<p-~ycBLTwhTwu=5M~-ya-d7sNDpUd2L`tiw2~9REQNf+%T_qnt#&<J
z*<Z>fmW}Z+tK3P_>{doxt+zzsUE)dm7?tVI{n~iA!MnPK3-YPG+sU_Vdeb_X6p;lO
zcrYmutZcHmcxhWSZr*XWyTBYv2iu(bvtz>D=NA_I8iDI+I$J3H_5?EHK({2T==g?p
z1?#%N;9&h74;B0SIpF{hq-)e9jTr=#+ppHHee51DX3`wMbq84qg;H&ms`h+-WfvbU
z)G)NsjHN;(<SD2F8(iOCH{WJ4cx>kmfx_KaT<%xzv$GK_ZzY0z3NLNEo;>J%w=$-e
zC_MFtl~Swd4Y6H~P~s-7FK<9*&HQ+jGCC}*kE)z7*BF**Z2}crZD4fN#;oG90#W^Y
z69GqH2<xI2uzGQoP}IKHm!H5_V|+uk_S*3YUbjcdc?UWwfGEgIG!tXj^Lp&XYCi34
zb}$?RytsyMn}dSqZY%Nx?Us&qJw&-r%dj5dG0n4&{rv>Gw`qGRDKwS{@0*&2sApBY
zqY6jb(OXfSM&r|zo2~k*jj<N8<j6?4a$s-LY%24~3AO&2;L00wdD2e5Om(pPPaXO=
zQaBpBHi87(O(1%}T+l${KYVb-6Bco{{(@}446{r8T<AffVSl}g{ij@;t9g?GxQbWG
za6A}giJ^HlbZbQPc*Hh~5ez{*{?<t6`o<TYln-v#qwH!k!_QH1pjy*L;C*d|r3a|j
zFVw6PGZEM*WdUY~#T_fJnh=N@b@SFLbvMl_=vZ_5x9EyH3!rF#8iCX-tsZock%8zp
zi9M`6B@;M{GKoRy1-m!gioOEu8;ucNq3w;p%jd(^>Ia_OS$45SipDpJiJtyeDHR#=
z$Mdx;Bk|Y607J9+)o*s&&7UB>uTysOz-DP22#ovQ(?p@j_TA47XfBIAqtv~C4}w)J
zC6_tlP@AOzoAtHb{B0MHW<pz}e;`PA>jGE`*P(FO5Q7J}SO`xW`<sAT9qL-x9b`Vv
z0oyqEO>y^>mOvo@d!xcfs*+Z*^APp@O8-R~{LTwwctm*g_c2GW$Mk%$G_j2w`^F8g
z)hXZ2+iV!FZC`Q@uQ{(yflLDsy2+<A?M*P`mY&E+F-XhNc2cdeb2|IuGw^~u$|X0h
z<8nb9TC2=Z`po9GT?g_=%D*Ct`CX~ek0Q4Eu`k*bbtAFV^PHUd{2F}p`tt7%p{EWH
z$H2yh7X+#n{<`l+u!;`pfid`DR*>QC9+_^Csa*w~huVAGz3xr$F&X}8Q|BgwufV&c
zV9h$MyQ`bDude~DA2jY?Ub6Zt$^zug%kOB|wr`uy4KwP+Ti`s^D!SPTBWb+bM+&G>
zd>K$je>cZjJZ@Os_Rm<{gt`9Vow=q?NKJJOojLJBxP)E!!hyS~Zni>~TLzTa@C^mi
zF_woyPK`C_{(}wTp@2f&3x$tT{Okf3Cj5;z!1(9Z=BNYS++``a^?Ez`ig{rh;;l^q
zI>D~{yYmYA$3~<9l)I<r+hQu!vo#FyFA7~;zEP~ph{4@>-VJgjQ~jXsIBpD79G)UJ
zf^8g>57bY#Mx}ue&4Bre>*V*inHHMa_$+tV3c^R&VrrDWoi6ir6t7;FQycvNdN1%x
zD;Vta{xhZ@Gjf6x6AdC4LGy#in-6jm<bzBlFQ2}^w!~T1UM_@TTZ|WmdR$B4lCP4{
zFAU(MlV@(Cw6do6mvkA{T=&f;lc@f2D17bpKz-t|yoX*2c&EeXkTK*Jm1dgSQXSb(
zviWl(Ez}4@v!Q*bjuEV4*g>tx3=joHe2gKr`AvG<?XrRE^*@+In2LsN%g~({o@&)@
zOQAPuSeHyb6e%uHY7ef}30y6oLKv$p*ff^fJ53`^0$~IBuBa&rGUOk88r2!cWQoiU
z`=SdM%`4qm&+RvU1AM@Zce}GZXfEsQn2}l1Ucz$%9OlY*qt@K*ztE1FHI3thTjFWk
z8@S+h8hv-O?F8lQ8L^fmfM*wI<03sm*%-WeZoL-5^Ok&x@;c9}Vb<b>E*A4zJ6Zzh
zT`1f)kT!d$+DbH|F2PXnECFOh*3^Fddh-&m^YM*ou(0=RQopiMYEj<^%s6l1J5-Q*
zep_2xKhElB-u8>v%2mQmQ04#Qcv^T490M8!<LLl(^|$mJvd49MPImJd8jKDYl38Dn
zi2`Ts0=WSuGRE&ARI9M<;(l*}qf;V*+H-mxw|)+8md9=$>4$(sqE=GoHcLSJjLdL|
z2yH6p;>wEO{Jp^<bv#x>(3;YVbD9Ds`QD_B@E#KnEUo)6;gajTCW>-i9#`0?UB5Zz
z>?{VuP(sEY=!710YX>fJ>mq-gKodb(-)Z>Yk~i_A1-SPPAula(I4O&i+<QE)VN#>s
z9q_Oas!7OlOPE^^NO~gf)!dWOWiA^Ml$X(ls?_<KK3`2=j8ny%_1eGl=qG0|z@*G1
z*13qriuRua#a%FdAdR+nyZX*aXu9{$0zz>Isvi$bq*O1WO>}wM^sb)Ye*jheZK>Fn
ztthG+qy}q0L^#y76ir1SJ0mss>Xs+ZuigjXLm}Wx{IlZ$cXIj5D;09o3_rtRq*q>o
z=~L7=9E($?QlzRGH%$roIC;JG51t;z7_?fP5vAatQ{7Z(flm^YVNk9JKZB642}xr-
zRqCZ{03nOuSC;VBa;ikAA%X7{+e%X~IxwoX?Obic6_<8dB@Bi1l^#A$da?m$`ewCt
z8c7fzueZCyiw+8=thyY`&rnqPdfl#0O@G~8iIA)Am1NmU$;--q;(~w@S8)jdzAPbW
zED`x>!p3oy^q`B^*m{C=E5|D?T`oxqp_|R3py^h%FcWZdOzlA7P2IA`TDr4q;fhyf
zxdDqI`o;=c&QLGQG!_;^hQGX~gj|=Pepa(61Y=;!;Gx=(|4e!E<);hNZBD3N`;|OM
zi``nKjUVdJXPTx9ps#DZ<N5r8!s9#9ow*Efat^1~bO8a%xMONvo^^mTi;qY|50)f6
zpz7Rl_|wGLH~$b4Zf3uGPY#ki9Dwxh$zP{wy1r{^%?_{%GVZ8Marf}`lU6GazZU5`
z@!-aVt(Lsv_MnUf*8!Z>2#Z%zF$tx~TVu%1L=8*PJQMSne!%%Otm`Yksd+F=Nr$1g
zZIi?$vrVn$Pzz(airPU?wbvBZ1~L(9jy=?_&Hck#4G>7WieYL;T#mGR@mbr&q??(T
zc(XR4ulV9a*_Cu|KShUe;f&r0CG9630F>jm!^5k9B1*03M7Qk1-kl+WH#$eL0fu5n
zQyp5Bq!#&h=mf-8UtDtfb~jkFKyVWi6N0t-f!Oe(WzsWy_nuEy3kZa%Ke`iJCz$N1
z@TTvV&3hz73q(AeThk!Oh+%P)q2=Uz@u*nPhvHtz{cQH~Lt3{0AK_*W6gGaW_b!DN
z#*fYZs;f~N69aQc;zaX}6l@mgt)7_{KS^$aU%cKv>q=1#6(IN%vH*{(wGdLkiPJ3%
zXIJ0XBpLdNYuEg}{=zS0pu{U_3l(?oopI~21|0UXfsMYDpY!Cx6^4MjJ(h~>wQuG$
zee@laQ32c<oFAm6%u?kC4JtJRL%2gMh<#^8izbYSs2M2Lu>08$RUchZrOd18aCaZZ
zTM1PgOa7VPeZOsbXua~Mp;bUl`$`2hp{;qdMzM0(k;~6hU27u32|#p!+9DQJpS<n|
zieP;mCPbWsK3Y8xQM(hgp$?HaA{e(;XZ3ZGycnoCzNm@@gZ{wOkx)a7Cf+~MqnXok
zHc!DpUIh|IeE`c<Wvi%1fRC$&_h$Su-vYxRR=*bUUN?39;30LrN@K>a)-CAwr0i-N
zHq{XN6ef=%qo(>cD{AX<gdP54_Nxam&r2cE%Sfw)Xd;tbMi#5CVQ6idkWk9PqnCVK
zzzG`r7|Q#@27ZGH%KdAk<TY><EdFg-+wS%MwC);Fk&%FnNHYVk70}-LHd@sw3fT9`
z&UIkhf0}aZ5EQ=Co?unKMqca?br-n;4{y8-5(}Na1g;8veIwVET-5Z1`quH`n{VnQ
zuIl>(^A?+=P5#g%ukl?AbCrVZ8|+$bX`b_78nOrd;8oq<^&j2~wY@BU?FvwxTLJ80
zPA_#rIJ*Y61|gkvJV-cP9Bhc>di#x~Is~GUW43<p6;t2aksJOBGt~Y^%2yDlYT8XN
zn*Wx*-!QmD&T{3lhezz5b*Yw4-k5b^#Ss^S(Gi<?+;S1+r#GiX^6!FUO?5$1&lSPc
z#!NkiumqWG%B7opaD|YxVW<quR26jGYBF^>W39ZUOd=L?<9f9Dv#G1h5IDYT0xThb
zP>G!E-Hp>5*izyZ_+Z~Xo+fAk%F5SkSeK1uTK;+JTUeaE0ecaE&{k=(!Cf#JFkkR4
z2C@ZVy$#4NqLcK6b>4DG27Bx1G<NXstHk1L5r5d#qyRUu!wtUd25PdPKfRYGShE~j
zw2=g<FIJB3hsu^r412DRXw<fDf2)eNNm(}WU?8-pk|7@kpn(N&6kntON;SzD5l{Z5
zzs^SCq8X!?wCM~^W)j>Wd1AjXe4=Op946{S7<og1Rg4=jrepC`$f{i7?L#9q4<6J9
zcTj_5xk~(^L><i~G^)AwylP(QN}`fS@~JVmz8Y~hz4A;-8?5Ik1aKG5<KbrRLu;@)
z9Z%v^A92e!ELR~z<)9{jfv7h7mmWlCFHE<afu1qv=$BV)=`G$+wD*xecKw+Kjz1Y}
zx`|S}LjGzJX#w1_nOYwzhvGy?+DpR_m0yS=k6$aRs1$+pH_vx_J@>5N?}wusj5ghZ
z1D4^5li_(PBHC?eIlw77YM=4II;{#vTv8I&_9C?OdVh;3UnX(9z|yuEMf!Nl&*J#=
zxj>0OaFrgeS*<0!(0j$bhVg{2r}00+;P}c+3<8+}4z*gntDUxbwtPFk?VcrJ?ByDC
zO(SP5*JmJKk~j&`=&-8i8gE{|5=fpwM`pavIde7?+y2#fQ6e@MY?FojXGl*vE)4#q
zP_oJ0ImC?d(~DWpGxyY`5|+YVTiBaIyuCz^ibjSYel#=v6SJKD!^1Gnh42DKXoj1O
zO%=4~?YvSJqSp?QC+|GzH0T4g)9m@()Rp*LD>u|fL^qTWDTf<w_yo#`($NL76?bc~
zkYAjQ?fvz3WX+-;eZ0eR<b@RQP8<Gnw?HsP{?!q=OWjNY+elUo-~+ZDiM0V9(02lo
zpVJ()5YEU80)jdJ-~R&0HoEC9eMx4C)~G{2tGkr~xeq?za4r~!*hezRGq^sA0PemH
zDy*mBMgWI#v;%1AFt>*2<`7a+8kn8S;tSHP&t-5quK}Gs#;w!t0p~$u{op=dKdETi
zm^E0<zpqUZ<_P*rSaEcLr?@BCmeaZY@85@4qWX1hZJd{O0N`9Wf9P&e8}M%o5UqHp
z#x#?}SFPqsEWD<PZ!p-e{j=Hj1`e4?ISS34V9Bg(IiQUEptH!1*Y429JA|16Q8ERO
zTQ8q`9m4)m@r3ZXw3MBktm5vWle3IwSsJ`Y(YF96nhX2&8j<#c`?Jo1-S|v#iGDCY
z!65{QT&AXPP|2jD$dhyXZ4+_NV5xrYgy$NmexkOlZpv8qPlNS9<yC<0W^kV5;-9=0
zhG)BUNvV>rymUMu)~YZ0RA?1f996sll_}tpAVe`vlaOR+0%OI%VMXWkZ}7{ZCcl3;
z0;SoKqG0huJF>stGIRW0+uf}7g*(=?YH3B`0!I!krf4b|k%@UZyW6jfBk6^eCDyac
zcqxDVT7xwsz7E$%C^APt@*`kWA$WEEe{Ec4R8-LxCR7?Eq(Malq*D=wl8}-Cln&`q
zT1r}^VUQFA2a%GYq=sf_q=wER9ZFzGXNC}Xm$%mY_x|0x)>-G=yZ624eEZw`+xuj>
z)iu4pBQUsgcPF5(*Z1+iv}9ZXj|5<2u6nnreW%k(=P&uxPuFfgn5^;bTq$Q3M&mgS
zE99{3$_ZO$?jBZApD85>4xWLb^)p>7uz6C3i7#0k{eM4QDJbvWKAy>L-BRD6JD2Mw
z567Q8F*jffyGxjD(*h!u7VkagC~6$kE5q(T&#upwgS4s&r^$bEVwBR;?3Q}NmHX6g
ztidbI4U5>LkHOk~7h$nga4&_aItUm^eGEiK&0c&hPN?lc+0SxfGBkheuIiX5adn{{
zIqV|*z#n-nW+I_89M)itw~KF90C(o*+D7+9+eb~)lOPIr<BjWe-e47;1sYdcG{gPg
zVn*JLHc#KkV0PWBzA4vWNkwK_<Hnfb&*|hOK5l3goFMkJ8+igTkO<%%#f-ag6h4i`
zjCfi?26Xqhi*2G8l}m0xT`sRkx{>}@ZZrW{i-L<_v>e2~XQ8^NGnCF%ljgW%vb2{Q
z^z1h4Ex$(d?c-C^(BjjY4pLgVTtuV3$t#pLDk4XG+_IE2?bf3`s2@de$PX&oH%k36
zH5`In^m{%tXS>}$p*%egDK%@}B!D^WL#b&kRGgib1{j+XhtfHxhq|iEhC7gE07G-5
z^X5XH?UK*4o$z{|&pqC{%_>~<LpHIZii2$c*-JcVn?TD|-KlEFep_J2rp~4Iv+6!0
zZA=}8atk+|Hy8sUEz_GC9u+4$3rs-sjE8uBUgBQbZ^C;>xy78em;yC=m)-8EKCAty
zy!HQv%gpNc5;yo*>>%N>pCf51-Z04X*Vp6_)5gAZn?MrfiVFH#C-3Q7(sLqp@L{|V
z$FIME`tx$%K_nFT(n1gcCSzo82`My71?7c}w%NKcpe>S5W7`NxpxeW5ug1<ji*uhR
zNyfEki@Tl*e_TYea58SnS($SL2V>o_^n>q@7GfE8TDy@rh_>V@Q&pst*W!EXs`whv
z9rV+!MSc$1-yj7(<S~Ye>+l9XFQyM{&6G~|urb!brcC|d&XZ$QF|PaleWRg1B{tjE
zX_^xkfK*KtpCgrVH!(5!l9KXW;qov^;36b+2RHEGbNfMzl~H?jP0sQ%b)f<4h8T{~
zCvi1Vkw7I4GsSqPZeUL)en2KZPG;61KMfP#R{v^JEqW}FIQR9O;mv*5`dK|!KuX)q
z?nyL6nw>q>c8rt-fB2e?0P3BD`1$qwY9K{I`=%48tOr*pS^P@F_6{Tzxj3bnQBKku
z|5FG}LNsG3Dy_qs9cDr*woc0<L>o2g`$gOGZ?ahAqm}TnFYsRJ)YZjD5n2|^->TZ7
zM9!KdG+f*dx%KFt*`Z0);OjlOd2%@Sjt?Gsd599wYWBk(X&V08m#C4Lo?2R#^DA#%
zUVe%=mUT=^_fzZbJkp8F>8$M@asc(rKQu0@nWJi9^9V|g8QQ{o>yrNCQ#O)iqc{qN
z6d#?2L@s=c30{jb#67e6fj14kd){!|?j5KYF@4x~D|wep`|eb|ksr~Oj2oQcHthw^
z`}YHy4zlj69>#u{`8w)Q8)M0vhW`j0NZu*>$enC894tLAYMR~GQB-VAd)fOUTYhqG
zergHhv0@v&VYu3rPkoo&$4Q<%FqD@&NdqF9yX$D-&o%)q(S=Lg+z;kv2T?nV>@?mJ
zEoN*xfRRNwZtd;{d94qm3jr$*4h+=7ajCi+AqmXg&A49g!D;5*N*oKgN+&g&GF_Rc
zmW_ej@7KppiC7~g%gvoksl`K{Ts{c@n5GI3P<HyyB_2logXZfaKKe8oUFtISVJyW5
zUK+?mx;Zz<DO(%dA`=-Fl1R@Ifa^5aWC-$=r`z7%mSjzI(a?yE>auPH4(4S*D{vxr
zl3e+V#bX`6>J3)Xkn|%kNDYFY$IJz`ZaS$oP~3~-ko$RY&`!i#{LHbkM8cAk21QzX
z)S*Xjf3iZ#zC(!w<-WQ(_@wiWh~k=ECF6`wC_Loi%qLmBTR>RYkhww7y8mIWpO8$*
zu&glOK7j`SwwKtqTOZhe_gx%d8o<qzS8_{JT@@-lQ?Wgb<{a?`Wo@45261M7_9<g}
z$srUoBO2>77i6G9-jMw6hBov{@>g3G;*)&u{1qu>__=wD3$hP#^u*R(t7F!Lb?>bA
zk7WB5XOE2-R(Tb$Gte^Q;9X$gdGJZ|V{Xc01oi7BvUoszX?hS=;V53<cQ^bsM;IpB
z>Z-4IXV;6wcWvPL{9xK+IU*Ddqv8)@x;8k6cxoKlqAUJZ0YJQ&f!i;aAEBtT&d*n3
zM(2~4&49O$r}scp|2Wn9lv)}N-Mlzj|NOemxhsl_16W^6Qxh;c|E#N71l@{7eKRD#
zgDQ<2ndb53Y2o6J?UBvlY@>rB+2Tzlc}Y|lUaYSB+}pr`!bo?j+Ln3by?Os6&q%%e
zLfW-W(idW1F`p!>FUbul*{f~Z;;NV=1w4%>rvl2_)<+KQ1Z%1kZ1nrLS6dD{5xI4T
zPiQdBoNK*ewZ#PL{EMYVlMC`~cLtC2jEst3ym&$Gzqq*gyQyM&auT!vNTS4ndh2Kj
zpPl>q&fUCla(RDFU$?M)1dj)#%%%D#>U`1WSGc-*?8lEDztRX*LAo^;B`B%y&=G;$
zMj~NTD%ZQ6vBab4AE34uarH4|i%j51jS1@k_EbK6jV#IEB(*7Reuyt~wq~cMr!lGQ
zmnvQr(_ur&V|{b5j-{RG4P*-`^6FMj;0^(J9`WADXr{;!8nA(rdNa))ufk=jr?;A(
zp6)EjIMLK3gB@%){ovV&%+G?DrxWaRW}9ofSx#npD4mkA6Vf_rzjJyPR|mtz{VH9H
z!?I)wna^xMd)PAJ>8E@rX|Gg@2uJAHZ-ME}lU;v^ZKj}<C=)Io2%%YewYhs=ml004
z9WeSWpMk&OAst)7P(L}JEBW2v$=1}D^$b2aSjK}nv7f~RTlc|n8N!#*HP6|SA4!s{
z2M4$IqCJ6h!U2J)si|?TlitO~aCT^$|KV?Jvie$6n*>+-&y~v6q_&&SLkOF5BR81x
zyko{UoM~j_^1Q?9K?%Qd6^CE9KLjiKljq2FzPO#cOsR0nxEpYPwg|$Rm(S~+*B*U2
zI+8yAM|5+r7b&42)gF)2n<6fG<$^t0AFQvft1C?HFJOxAX{oLjPgKh{1solEun3Rp
zii_Yl>Z)mrKN=xG$XUt=$J%>%wKjcvsCF))2k^rPI9{%OC=8t){VN5OjyQ6mHOUOH
z5aa?vRMVB2HFwFQB3O}{XL+{R&T!=K_p~gd6S^vzs%zr3jpi!7aM(!G<{t5)@rs_@
zs7TeNLg>?!)Qli+4fj))Khg)&y89gN3`;_Xal#G$Q**<`s0;#0&ZNNOjp1V;0B-@L
zup)G^{#u`fUF8Ad$nf!#Cs(mMP3~6jTiLj8jE!_w_pNR)Hf&HCpK$Og(B$aVejif$
z=9Nc_T4zhtO^W(19-}~G)JB?4L2kvgamkO2jtnK4dE7p5@$;`<CN7IT&u~f^gTdf|
z{M|*aX|2<kmaOlt$AjdUYfAxXK@nh66l6S(oae`idhm?Z`?g-36nx_W+jf4BqyMo@
zrnXu@MZ(g5g@_UCXHLN+Bl2Jo4hymB4D|CzH$5Bz_T}ZW?8`}8lE%RF^o6;%B%)7p
zrQ_6De#S417u+RcI$La1v7_p^C<}agde2J~uP~cn8ou+Y&YdDPGhzK($h4<Hu*5Oc
zMmsrmx41X@hGUSUH|_07s8>2nKBZBO>OdrO{!gD+q)}TwjPNT9`7gQp3T12sQNm=n
z2Qdvu;moOnHOFrdr@${xI9+w~W?Q}O26(@syqqIqo4ptk69cOT+DbXQxTMGh;-oDs
zEXK#iRvRh=T3cI@wayc3=|+}~*U|oRf(qBn?FBNL$p~EQr?W=|OX2y554N%|W=*3T
zf<3>z<Izs)TFAEEV5*K(Se%<XwUhZO&rBW(dzPmgiIz?kT{ao<Srj&q%g`xKYN$W~
z|E%X`7ze#24ZA-TF6b_G+f9yh7B7g%{hCYj5a{8T5mWLpg{b~b=D-L&;M_nhEnWTv
zv~&mnjLgG<G^|LLwhNG`F)}wV8*`eBSIcj5p5VD^w2gd!!(RdP28)bq-fLQqknJrz
zUaD0W0SjK<UWwk!4Gqy<!>46^vi1<Q2R_m}+Fc2kH)7Y=GF{DVd(<t$dBd()dVGZY
z7$o~Q0%2#}y1enfrUEL?rDy}#C02hBLpwkAnTchs?SzRWb}p>~5W>$JU+x|zPn;t)
zpcj&`tvTK5;bQ$;rX@ZTTaO5CSXTvOovrpBB|gzqIZ0j#KL21)Q!gSzWuY1!4}-<1
zb0yznjCTe$2&SaI!h!;oBo5h)+S*ztJ_fH0Au}BP$nm8+a9lXP4mn+~1IUQBhwRj^
z6|eToq;((9v}`QeKnk1lHf<mL=A~Bj6Zi{fJM@hkns@s{`6JW);~1G>*?5L#VVn1y
z{pZ%n`FR)t1ih7~9Q^~vD*YS}s_AWUo+z3TZv8tRN)g@lnZRMe<h7(oW#8nMd>e%h
zcVaN~O?RVHoJN7bS7(*R2Jze%bGhg~;A}8U>0dkcTkz}?SqeGFEa~!1er~z`Wl~b;
zgaia8qM~k9=JL8mzW*ql>i@SKO@UCe&QA+y{Nc3=e|xHJT!s9ZWnM*Xqvb#`@u!Hy
zpR~a}?S5%xO0%#p`-J>E>3<BfV7L99$aeco@iK2wjMPZC-K6_X!^F<%<eI(48DsAW
zL7k6o&)0PBfww)V5uK@lCRYogOyfBeh6YavHrwzXgGQ777fg&^|3L2|-s-<lsNKTN
zDHYoAa<g=(8EA0RspJQn)qxZ>wY2I=O9u%XFmP1LpLb_^BgdSmCZC;ujozD=fUN!x
zld7GPRpOOy7452ql3YMkL49%YX1d=}z-Akpdy8i-TG|vKZ2Il=OG`^jDv&RD4|~ow
zvfY;uZ|ACa0K)$SdvtUptGZS|sB@!e=yBy(K6JCtd?vWHt`3MFTk%UX*RJL4^OR&D
zch=KO2JjbEbXRb!-I&uWpt*rmvijddPySo~I_MM@pg|X1Q_E1lle@f;<4V-JN?8((
zA|_r#DLK4vnL*w<VmvoMa$9(oCeix`5vNPKVWKBi1!6)^Ln%SjE%e-z%8?xss}1;H
z<?nICD^aUDYAS1ww6n(FLJ2R)C*YGz7`{=0e_5~)ducQiCGBd3nLJ4lt#O=(Fw8Nm
zK#%&tTV;RxnZ+T$CpV`rge4IHyNXg3ei8hGs>Y<+&-r>L#@2nM$756I?3nx$rK_<Q
zEHxqeX0lESf0s#ARMffY07*J`$t%tI`T2s<QY~>7GypODp}c=;oeh9^<_`!v`Lc!)
z0A-DU$qH9JP9z}coDn_&tr8b4KR<lwKH}<fLcp+Wc8}s4S~{4B6zJShb`4ba4R9FZ
z16OUH)T6ajaK`exi5GP*i?kc1*r`Pv*2W}7fB#F~Y1;So@oD|4Ss2m+fKWXA?cDe}
zili5PSG(Z+VX8@<yehD@)GZvh+>Nmo8GX8@k>7pXf;&CleBHtP;-?z_bb9-pwW#ke
zVKPqRI%@t#pUxjjm%#$q++%(fQgS|aC5BV;IApc<8l|3FZ4p?G)NZP%Fs5XObE=3G
zv7-9gmA*-Rd*}$K%Dwf$)aX*(Aj!@Ft7OsS3)@Co&#t|8eB8llQw(#vlbn^Q(E}X{
zHYZ&uf}wx%&L3%XyzWEt&U@%HTg;?qWH6S%V05BT%w)++`ze^gx{-y?RKPVL`jC>6
z(#Rz)z(D@>4tXT9y)58)nA^y)3}{4lw`p$)3WZ{TTGsP9fP`wW1^Thc%@J7u7Q9x|
za=HKs9P%PsH8ZnpAh@NA?;+x41V9z?K@v#6&aOWN42FVEf9vwSLQ5mx^?CA#X>k26
zarwL%;0oR3I0Wr00uEY&!{J|1K+Px|5Ri6Zu~@k=r?g9=ui89!M!$UdBF`qV>W0lO
zGERn5_QrJ2ANAjp@cfdP^$ue&wS4tBC1c`)$h_ifB34CsJymb&FzFKshoRd9%g0fv
ztAFn4moS7G8W@Z%E4w;7_tPpO!eYR_>y;{8_YeKe30PTKxz!s0g!9V4z;kQ#{`ax1
zZ>SpfzCUaVs6aP2x8C~7k`fK{VFS^L$&!`oyf~=L*i10vC68GxVURo9Jrw%ZgC&CD
z;bDCXi$&h)e!CPF32nBpgnNTPyO}eUDg+Z%QNOCKZT|H1H0}-!OQ#`SY&QX0P1i|V
zFw1IDQ<JIn%EwMmb}R@SWQ|L5L>U?yPV<feHsGd<m1iqgH;anD^CrGWU-us)<kD9}
z^gC(qf0vvi+N%k~ySccycwK%K`T(>)G&M8R%221-%;WhaTXoR276#yNN|S@``f`MY
zRa1I6ty9?9*%{$tb(y-ETHN)d;tqeA-0nSZSuF@74t_JBl_AI?sslVLfH_wR2yiqH
z94n+)33r=)byeEyAK5DBS^&;IxsL&M>EDOa7rfM$jUh3z$1Jna<@S@tRd8cL^fS1}
zxmib0^2cy4YPC=Fu^&$~jO8P#AK+=w=L+Yj;%e6;3^{9YsEO{5oEdrS^;3A69;&i?
zMYb#l-pY}f=3<ttItbl!Hw$ep6E*9l6PRX<@3E=U_Tce)Ky!U%EcoA*7~}0bZF}ag
y?=#3C*Yg0EJko|`Lf7N7R{l5W!~Yp{MN^$-k2}tOGXk`RCD2gSQK@)l_2EB5eKRfq

literal 0
HcmV?d00001

diff --git a/docs/logos/cohere.png b/docs/logos/cohere.png
new file mode 100644
index 0000000000000000000000000000000000000000..85f6e6d9904b0d49302a2d933181279e974a7c86
GIT binary patch
literal 113536
zcmeEvcRbbmAAg-FBXP_^=qR#M*UCO+Mksqz%FLEMI@hfZWfdvAtn7q{aJo?<dy|p9
zPxj{b{;2M~<#YP|9>2f8e_Z!Gbl#uW`~4cv*L;5jTvAn_AY&xkvu6**1w~o)J$v>i
zLI08L2fumH;CUAOZ?Chu!udVfb<E>?_OR`_AbU>J-FT{p{Be(t9{-lx^b<qRM217O
zWOCukwvF0Bw&onfiEZbjI-604o;-O)LgwyoMkkt(KqQGpO<cZAOH?VGmm2)pTrGNE
zuey@C+O5Dx<-~P+_imqLneEtfo1^wx&fvnI$w{?SPd2;rmbZN7Bs@s2Wx18&EwV<w
zd@S_hxXnsLLdAx@wP!B^<^vh6B5|8`6#PH#tV5F|Ng&{setRnBJTZ~_?Ky5$0tCRf
z*h1&_|85I(REl)f8E0Ylx3lgRx|)gL2}r0;YNFY+PUZ~29PbBN1eZPG{h<>s?Zs$5
zeG^DSm<UPI8^ZhB^FJm4{rVph{>OxG_Vzzb_@5^HPZR#}7ymPd|Cz)8%;A6L0IcAD
zO!)t10_HG_-`bHhr@zM)4g|+MLaC~qaqyIt@>y+1j{j9vOXH-(glc+kIYJ?N)hxZ@
zeO2x;ryHNRgp;b6*dMi5y!Z0!U7s9Yv@#1U+Q!G;kNuA*v`B_ll#<TUI+n-RY)g*H
zAjdBSpAC0n4y=~kHVAfkPCXuHlIpEc=jfHJF8bJ+Ia6@{t@(*V!t6c13Bt>M+KoYc
zJE@0hB?6H~r=h2~BWF*A{=Wli>U$5beo9uF58EeN_X%kspCrIu$<98iAfI&iL(dfr
z<3lm3)3G02l>Su>GzL*~uKlf-?}%SWco~)Gp^+F$ZB!2n*H&8ReYIpI@AS(VewjF_
za1IDnrD&-X7r!Jk20-Uo(=nYK>UiQWT4yIeKNZttrct4XoTWfIY6owOxu|_TZaUG&
zY5K>%$z0Fzudl0SDY=H#L^OsqxaLwdg4Osa-ls}A*lud{v^+P4V9Rt9WO$Q)2gJT*
zM<#Sd?$)E@%2xevEQRY;!s4j7SQtcOkBUkQt9!F^|NC$vl7&!Qs&j?YrNy+ECQkOz
z5KZ;@w%g-O(#<jgmJ!f~yN<JkTGJ%`vBSOuVvL}mW2K0!l&4+N$_VXjZO!%(<fwcF
zyZ&=Cu74~CeDJ*7Zz{W{m{cT$>eClxEM!?z`s}G(x)ou}EME^j>%|{?F)h_GnW!CY
z%b@NM6Ls~iu65Q>8cm`d9ge9BsH*wWJfcOHzr(k0;Jg>H$@7(xYi^q%ZSISCMOGSh
zp>THey6kqa4aBl=WsqfNuM6`*JE3KJvH4mj#&xA3Z2p<LsK^XyOkK%xvud9v&QXc_
zKWSvbMTDDtMPG_y5cbXK7x@ci(H6_TWf%9DpFrP2my?hrtqpV3o`a-cNj}wUbYyMp
zKz-W5gNC8bWo+t(RHAhu*blURMKnLQxWlP_I}CGt6%rp3zMQIvw{MwXWtV{-(ZkOc
zI<Y0*bPU>Jra!KY$4w%$iFsowe4f}t)IHGUoUBA;WS7C+8i(Wm<tK}N|8`2%dZbJ`
z?d+DnGUsQf?x`uG28d=t2@#Qc&rR0iW01Dpr0x*b^HHkvIG&o%hq{9rh`DK9nY**6
z9}4`!9OL3#=kb(!d_-KBxyx;`0KYjGrOdq}2^(rE#jJpOb4W?X<S<9YU2m)nq($6|
zs`;X!+KsAokd=7DV9_Y>-_4OQ>-1-1H_EUs?;1xMnWP;ygSID&#CahmV_uPyBz38#
zvq1}7qH%Kd9wYA#%!tk9V+E<Xq=itn)2~<-cX;#9CXZ4p&W1F8a;{aH%3>D%lEajj
z=jJ+g?E`e!7GS{YCgop9sn{Cg5e(DCQy=$PeH~D&wG|y0pdJ6XDVmn{lO#-ofOrFa
z8b66J_k`MQ;0hMivU205A!1+I5C|?ce@M+e^-XKY3vYC6+|GM7q}26@-Gmug)3<K|
zDJdB~0%+~<W@m$@r9@IJTy~!t-^^Bh+;(c_vo9X5@)Q~{1m|0bzJ*Tjvulw6-#m%I
z$?52nJN7G<@uDQz)xF(hV>JKF4f+N>KeiZ4w{lh1Q%AN%*k|URU#@*$S#li<UJ!a#
zBzRW&OtT<()|fJeuXQ58KXqt&pQt3ef&@D%t|ii+F%a7Nk4F&|tFtL=7$}X`tIOJ6
z@thgKRvC3ZPXYhj0HNNgGUhBKVBjZ&Oo?k8Q`+2prNE=216Ih~v`;MNoc|sW$jlC@
z7+)KfQ=g_Me%j|9Ur$qn6!e&B@EE)DW1G-pjw41*e8w+uvi7)KsLN$npTaVS@O&J5
zfBdHh8PiqG24D?qMBC|e%D*I*irp!L1V;*rIh^6KJ4W?^c1L07%l2>&3}8F3Kc9(Y
z;W;j==8L9MZP(umLGj~_NAgv`2LA>sf4h<UBBaukv>j4XzH?tN;^q8+(WxC_sJax{
zAIko1J8PFrO#&AlX;88}=_zr;f02r*R|Jf+)I+ni{_%tY;$cG|&WYq^JMO8)2-ZR&
zP}zqpoler`C($Q-AU-6o<b_I>ww&@<R`B$up9b|PQx(7#q1n!z&3vHNJ@oXDZPEOk
zQn1CSqFM-6CFvgl?R-FgB)0n~y8Q*G!|k<HU+a@LME{->o$wcU2(@u>7$A7*FWjn*
ziD%OV<>>cUS07$N)?=}?axsY<{}nb+3M&k=NdRqLZIz#Ug7EfMT%6oIbf$CQOo&Ix
z(3!p#APmx5X+&Zk#6NM8;6X_~r2P3A;MC}?Yr)mg!Lq)Sd<ARe1uvWRA?1q#%GYjp
zu^c>Q41o#zIFaWRh!0`bFlq><|K&W=B%elN{OldgeNChfn(YUxB`%@ahE*V=!8sDY
zdgD-Yhr-1BQ?Ta85|n}vN>y}*gZ$@b?7_5>c%M|*=o%0;*y44?+hwQRl7vXh14R+(
z4U7X{Zy-ht4i4_y_8V?UPNObC&CwypTTWRjlS2pnSLqUX5%G1%a>t=y_Zq&%HJ$1|
z@<mCh#!SG2wtKTag73cI#(c?mA9q-JRRL8=34E8fRt3j@%m{j1f)S!{6x-Jg+<}2e
z{miC!Nnkp_D2yrqGcmpy3Gj87r?}vw;=)HId`xwQs^{6~G)i8`@d4BLkLF=sk~6C!
z+ea`989|dHQkPv1&eO0pcmYxOmIxdLx|3jwc&B5$mG!!Fqmg!1$>c$fZ6?_-&p;77
z$$rvtnr~mK|Lou#b|4Kh09P`As%dHT5OkqU*H>0P-rbLnQ4}pbt!~({kIyFTCuAU@
zI{OND*GVbIC??M902{RX5+LAidqqFMJ<Er<S4OQ*6C>QO1iHJ*#<Zloto|=cp#x3w
zJJ@Ir&;>~@t}fck9RRD&z5+Gh8gN`REBTEh<wGqL{+tq=a=zze*V?35S}lJT%Pn-p
z-Ka3X3AF9j=4nWpwh)$-AU7dqQy>gD75KAdhl!CM5-3$mnF}X=k|{VZ3sMp37w7D$
zXJ$V0PO|BxFew07a0H^r8?p$Z>SN!0{a8ceXfiE77VBJdrhw}w+Xc@!z)*$HJgDjG
zx_#4`H5DKPW|1B^yJd>A@!*6uw1>;Arvh@rO&Gc2XpqtEx^=F<YDGCw6mq<{T77<W
zxxQ}O%rD5C5%B*4FgWGHWF7GB&uXUEtf#EPqWM3Dr6a%8JLl|pm;ao`{UnB6bf7lk
z?OPpw!}U5pIXSQoRq_?`PpbFCL%!`JDpN{)c;$(r;}=Hd&R?tc;cEStf>(NKqxWyE
zmr47!znu%RzltDg1F=tm@mw7gw2*(Z`nU%sRH8;S$PoK089Q+BlWOMBR3Wz?p^Q7v
zTGT|FEzfYE!2W6Z*h0DWb$!6c%ya!|gW@cXE~a0jVg}@pvoD0gU+#p2|2Ykt123lA
zsZKigtUx&>ge`#}*riL$5C$K$xlVhxcF>q{l}o4_(jlX@_1+8r!q8i|Kq9A#Zl3Wm
zW%?BRjZcBXx{z4tK{U(vrye*t21WczHP~2!YPN|vjS;zZ{}i=>?Jp1M#~>ipmYz%A
z=rWjC%;^Vl5L%pg<GO{oP%8*8aT7#M4J{o4iBC+T)DLE5|7G-xYJwPrbVEaR_4>W$
zd9N<*%f}EzN+3+*-lW+BR(q1hh)9*qw!D?8zlV*Kt$5g$K{@*0_;JVHgZV-!1vRz;
z_BD%qK97aO2wEmtS`EHnv-u}aOGx>iXO!3&6rHP=tw_}WY$pmB^j|)r6CNT&N8Jss
zzw57|&-NKDcbzk}K(aNM05jPtR%-?pQ~8g_U3eCQ-;!-%FD}x8T>5_i2RyY*o9)Bg
za=53KS<03t6g=n#-9lRnSua7BMvSpFGfNdz3${Ux+MovR)4$UHPhJDI4Zc722WFx0
z@`ZXU$j23Do|y);hru>Zm9tHQ9hD%u<Gm`a?o#q?x+E<OV9#c`aW?KcPQMM&Bh<iR
z>QTe4Qa60>yBpxJFQ`GtuN0mM$%h>AZct5{oziaOb@<>1RAm6)je_uR^#46jgA8Kz
zVT7i~+pY1}<ugxH3Y2c)#7Ie!+DLmrM7t1=^B5dtVik5@iK1P#@Ij?-&$P?_%+*`C
zTw<pRR*IAn<z8u@chmR5HolsIsLvj<OB)77cAC}<rxROVL++ouYSMmTCE$f9#;;a%
zyS?(s>ZDh;GX+3zB=95(kB^=QxdGWhHSKa9R4)~>uPH|$>fW#We{mH<i?>ZlW|ewj
zfw$0k^Wi+utYbj4DutLH01PKTINjx8m#*mx!nwwou8Cc;^fc@b4xi_tHy`oiL^*}j
zKLdvOZ$cI>vy)45&Dm&F%vYbu{T1-`A~NWz4mh#I=*D=or|KBIl>+&v0st#%H!Erc
z5gF~-?FnTyvDvYa$W-&KaYe_g7zHliTz*!~F-{EsaJb(MM^9fhX^QLg2qJw3D#Zdj
z^8<Sk*Qq;7Oa9m>xAX8Fa=$4nk$y2TdKW}A$YZ2?^7B{kF4xxeHolKi0!kzU^ogaI
zH;bCBK@aogW3GklP`g5fhM3pO=AWzr+Kdm^*Ku>}EgWW?$!@%6{Gi+|oDA@O5>!B0
zCD5jp?RCx?$#d_qE2qx>B7CsemQJ^t%eY`=h_9Tw#7(4IU+)Zps8}B`p|w3x6hS1$
z_2Q<})F-97oB&#2-%6bIu8>dtX%iqp<QmxAw~r~#XZ|LB#qi09Fc0Pr5D8R!uY<t3
zKLf!|lbG{~s}zM2HiMF|pC2%W#Z4n;WrFlKjgE@e9s|gl0SwY;^{6TU7xIH@_kHb?
z7E-V)DT_0et3Oi)0Qi7x<86wHNwIk!D_&f)X*EbCffg`N-7xPMDV2KyuBzpFU+6pz
zFJGZ5_UmHcwoiSXRjSBaYM<~cpNX{u&pQuA&M&x?ZlMdwZureC)!sc6KUYmVD**BU
zgY_1MpZ5+R<1fTrjbH}ePc1ooU#8?a8kPa|vhu6@sM#FiabH_H-0N1`H7Ytae`47V
z4~jKFKGHi<?=#N1Jl$Q`5_@za5Q_9jlC%@efwPc2aXQ~-^mUs0YALp(S0l*g(J!21
zyJU=c-;cXv(c|4CE~R6Pct|tG56sW6C)J9Cswf`!$a=AYg7X7^3U-?j$<+UA&;l?b
z95>2Sd$`-r$0s+nxQ%x&;sbC{9*4odB;y(?#<Ihkauh1e)($M6xc*}nzzHq0%Mh!N
zIUJWW%)6Aq@3+V#4RGNK*aoFZTpKCXiYLZaCn&aAA}Vs5vU>R|i~>+%!Uxtg>`m6*
zmU=C-%!CA7JeS}N*f`hoI)D^xq{k!<pR{s2d>|Ed)y(9O)yZEJ@D@4%pje+)Q)3MS
zaFZ7Uj@(yGumDa{((5K8P(Ue6SU@AIWy%5**T4<#-CE$o=Wtq1!se;=c1BrTlLOHj
zh|gQ^Mp_?7Tt{{YpK)z#4DX9Uo#LtaH1bpWwxbKW^i|T`&uh2I$U&$Bp@+CzvIc@^
zhV!(b=-LMD>>AdE`|z(D-h)vm`y_0=Xc6h;BttWH4&s(^)FaYxJ=0Z<%e5`u2C~IT
zpAP+;cJ7cXGDtuwy9EiG_s!!a0Hf`|Qn$X4nH_mo4UNq^RoJg-hFk39jz1msYdr84
zX`(9&w<r4+F90^U<bVLt3oma0;xe9ojmvR}UzHo=Cq!G4ax6yAYef_REX4w$%)DVR
zLb5sJ;2LBXUvu&KadWHspa0dS+0gyRhz_OFme8k6;!|XAp_hSMD#ok2fRC~_VwuGS
z<a})ciw=_PqPPbWK{lkwZ;)AAYL`5j#g0x807UD~ir)rt578aYbd9h9w;j){x63mF
zs&<%|c_xPPRdEwq0B2#qH6MlcTy(&Qij4}n<5MZ@RuY-5U`Y#+CUtnVM-CjYOKUHG
znIa|({24u9u)$%vV`$SXP#H)US3l89j||bZ;amSDeuj?nnLTvN)ym51!vPEX4F(jz
zGQeYNpUqg+gat(R`}e9=akTc>s(LQT{d;!n1cIo^f@h!O$1RWx8KWIfkW%SB1<P+=
zdIWgyi+LK*IL(zT32LG185uv9Np=Jcfc8Lin#V`T5nMTc3%3T4r0qjb8i>|TO9(C9
zsaoY!qTIV%R0$dgx1?~_Qy1$?b2JW!-Q9!PhCl#h{VtlOmIHww6vWoDX&uigP`Vz5
zkr(z|*nHjlPa7tG12oKogykt7gia%t=Xmcr5^jMS7*dxx9QP}9p-m-;gOg|n$(tA5
zax7@-I4%hTMt9nMenZrD?pv+N{ezirbL5!80dIv}@gxbea>5ZV6~UBI^I0I|BVesX
zOzioDd_<Y7iB(Q-|4*t9d=AErp(?C-o=L#jetJoMQ6Yg90R5)JJ#)5(WN1yK@%>l3
zA-aq^RAX(na{RRj>KHmf7f{J9OQ$)Rg{!)@Sm<1FPnq1*VYeLPkX2m9xo3^GhvTet
z6Nx+_IPs56xsK@JePQdp;8^>1Asf50WH+N>|LZmZe_7bRN;4L$sa*yX4+Dy4veT(g
z0yNH^02P7)tcx|*-h;cM5VnS95Fd-2s{6DpD?1d&B0Eo143WJZ`RoLu^yKM3MAtGz
zoBSp#dlz@=FhA1;nj_lV@;Ggq1PKBt9#GNFNwp?Y;&k}3ay^8RIrEJqDk$8}ec$fs
zqJ=c{R+hZ&R+QAyI$feRIgn(XnmTl%ffe!YVJkim+lh6T^}BeDY!{REB9=*#!i`H$
zOBP0hNNE}gVx|w5=@8dx6lFA8<5fg`&!F~K3%~K$ZDmTR&^e)?@zp-=kNNM9UwRHu
z?2>UMDTsH!SBqnUv9X&Xu>rf~N#Y%g3jw$*-4E=e3uu7Nd|os<jY!_rX!b8dJx?`R
z&+}>fMSvgtnL)SEPe7Hyu=9DV&2{<H;+O&uV6|~uN@E_798{b3_8h8?m#=uI^vBCH
zySGd7nbSc~Zc6`oYhUnRz`p||)E+fwz^Ienzj(jwkJN<;?DKGIhj+VmoN7f4liYGY
zmb|^aj~w7%8*ph^(VgcK1VMH%$XS7VDjh_dn0%xw;g{V)#>QgUf2q#n!@#EvDNxGG
z=Kw;jt+7TVBM(0*o?>!hDLRIVQ;&K|0f(aQ2+^_!PB%=7%C_DP@-40tk!+-bYK9E}
z*h!*SDpTIp0$L-e0YRva`)@k(J+cEVeydbbxo)#qZXt6ER}Hw#<NPuZZAEk^Ro?Fp
zr?o}w;ghuu`#~oDD=0J+<H`kjlo?n$@|N?m><j$St6=qTi8Bq&*%aK;boT*D$+d8i
zWbEj%Z+(%U>6^+SSZ%c*-8nSXxn4<^^$H8(0wVPjW!{N2MQZ)eMkVII>{AH-B$tGQ
z2?79*jyp7)MGE`jr>3SFAGzvXOt=Qn?152{BdNqC(`)T@_Ey^5A}QFs@!f|87*vc7
zUP-Lf-@L2eXLSHG2f(-5xt-(0*P94-jn*1xffV-20CyOEQR$YVRc8Y$$Jh9C%KnTc
zS~CE4+qY~ZhM?qzaC<&%K7;-5_<g;kyR>o-MwlcC%c=`htM`t&64jXaUO!}yn~pUb
z;se9sC+T%V%q79)O>ovnq)%N{Ya8^0iC(|U><oq)Y*)$Qqx^9Yv_6_uR_!_K{&z3-
za0-lPzlA=AC=}7?^)&U|@?%gpx{Q$I22GpBd(FZaZQ@hR<x`ja{r9bd&WiT3zS=Gs
zgXC5j5znGni+A5%peQWNueplw;RO#EtPx_wtbaE}tC@RMA<3}R6^Ix?+PpTi6Dw9(
z9Xj@J_N|+Ndlit2zmc*LhdpbSRWS8a+5S8GOotJo1Jm~k*4$->BJeLCp#6X`ugx2)
z$bsk~OCvE-E6Pe(JqG*Nui6QDkvj74pj9%@DISJ)gTk6gDHj<~sFcx2)Jcf}!NERT
zo^-0+vf7KtA{%PCZ^3J5;%A@1l*QSg2Gq;9SS63DNXPVA`r6BwL7iK$szrx5pdq$q
zOWrKvo5EAvNQAjb1*#|eirGnNPln#%l=l-?_q>d;IJw&`0B%nr3gcWI#-Bm_vh&F}
zO5|w*6l~qi5&<B1r(w5pcPUAcMiEXo$D}aKjy+eid~Yt_OkoEdK9ZzB5myA}Z(`S}
zj@<G7G=D#OE_rh4PkVnh#cyKI)Mp#rFuNx(UtT_e5T+ix4BT0Wam*`H0Et_zk#R0f
zVJi3l+V1DDr(Q%TM5T#(o37n4C@#~Wptei|NFVfA2|1Sky-gO(yXhh8?Uh>Z$^ysl
z4EmMSM5g`xE4cVK^V`|Qe+ChGUIb@1V@NSVER!0ZbzJTAZ`{aGoRx(IgoPMf*?Uog
zG7(U5UosqJ<aj{_Dm2^;H;%pN1$F5l$uq$t%W@To?IjGOyImtF=!ju}NEs|9_)U83
zN=OzSf&CZGC>Ii4y*BaEzQ&`UpMBsUGaTa<UZLy8=PIAPe%%{)ewip3l-tD0!sv*K
z<@<Y2Po!Yob=w7#fv?^zvNn{1!oehadqa`sfCAdwNOU0(S@Gtb1%-ZNOqfwCx$}vw
z^M{8oz;Jv_9AhBjGox$(fVQy3?cotDt`uqjV$hVp%S|x{XnLs0#(ds4ZT)NY9P$C6
zDycry&et+x(l#JmL%MJu9M&RRDe*d8k;r}ht!pi$*M&}t>=qM%o}sGRdrSf%yy2$W
z?2kayqV>13h!dXVpvrwaHLjkM_a|b-n}6z_e>*lti#%!NgSq}%+di))U&mIV8^=K`
z2MTeehi#FdU%uoMQMHKeko~p9ytCT}BuS6$F1F|Jf*67~4D3}l(zc}+dARJQ7-OyO
zp~Q_{ehqL8I0Vtr0I1ldEd0uG-4N)~D8r36y)=iTx;!paM=*i>6n`_v3{EgzY+ozi
zVkr8&cWNX|)?DWjcz|;=a-@i2a|i)-ijHQdI(z`O_Tp}#f^KO6u1r2_OA;BToMpOF
z2BM|KS}n$;cH&pl>4$_gHm)a~&26`6l-|u>5L;1IZD(DM-Y#&}RQCrZ6%e|t?~hSq
zL%mX+{;d8n!UyU22UINDtrDMk9CQSlZXd9AE~GxzFbQ~TTz+nI`3<h^W9~dpqqm8y
zSi$JsGcZ62sM4==Fgy}Aaex}})2A|7FscA&#nPsE5tLMiy9(w%Yo@)CM%CT=xwrj&
z8yGY4p+4)Dk({S836a2wmwJ17;;kTfR!bjRD8ssIs!=}kfZMd#u@|M6Gx}Nr{A11~
zkO06zrdv~=4F=T=x)8N-OPRyNX}b>!Z1p_G&$4AgP4|m_4T(!EP>9KFS((jKUt4C6
zhUhXLmZ0J)%k8reVuj@}jbfNd;iYIly>;$%*#u|6s5(9I0Q64ZC&-_Z*YA%BLs&zn
zvb@}HIplH$YTNM5<i+5MA@OXI30|V#X=yX51rOto6eRNsL_dUKGngLO*aR_)$!T$M
zul(r<y5pA;z}C@*$pB(O&!=02Rm!{C3pHQ`Q-1;ie?bpQFT)@QGci)8@FI1A3V&9P
zqY<j6>r2!AcUgVrda*7S#_x!nhC8+#$>)f9oaj9P{F)20vi10-zW{9#Eg^`um)rG~
z!9tKYWZ#1G;vjM%dy@mkfOME9x#WJ_UWha@bx&X83>%zyDgGF0I{m7Ym?iR|Dt_I}
z+U+arJQ<1!n}E|x(dml7bAS0bVZ%N!d=^HHLCZK4_ZAr6=(j@IC7<i0`6}S3!b-P6
zWQ+T%xp<fTuAji$L|CKO#W3S7KefbbO&v?8(FKh`>*+h(R4a@iA<xYXpA%<)_U9d#
zz3xTaCf9t{)u9p3zrlQfO4lDOq;P*l4HYGa#vFndbdb4E4(&c8_>I`{uYTMns%l-*
zBFo8#Vn+`n_<&mIbL3$_z5bPvkgMvvrSW}i^Wj~EIOz1CRl?O|&)YI$-1Jc(RxW`K
z7y*xd8?`$XChMV@g*IwiMK*@`r?H`b_&mwxaD0E02h@$tJ!B#N05qtH)IXef32MsZ
zF@B55oU!@&R;v#$_x=6L9Kv^-nml<44D*;xx|b!AN`TC)dXVkKehly@lq|jrTmati
zdNM~0V3H&dK%c@)gCcN>E1kw&G&Hj_{XjD{93k_t71Zywl9_uO{@LY!KDZ>}`b4a6
zn$)H-*{2Lik|rr_?YUp>3Dlu6QX#258;YpmRKl_xtvddJA$7F3W2>5*TDKHPXVKO5
zk*?`V^@OFM*(ub9faZXksmm@M+>0>ccJRu3xv&Dcom}PhW^Gc?P4gt>@2Rq(*_qQi
zF?Urq=EW-|SGcGEoxp~nE$@7sz3p<YHR43WOJFqD|6r=-{hn<zuYeh%blBn_knJRK
za0OZ{$Fb)#ClnMtIuG1sppEb~aiVVlnUy~)7CYa4S&af_<Dh}b9cor~Mk9yb^UeT~
z_DQx-{bS+RiSBS@_&q|!$^8kd#*)XsA*#%#b%7dC2OrPe_trVMuh$k^k+bB}S?P?h
zN=(L3G@Yv)T^}DCE2+P7;Sk$bDiBT7`-7UIJV!=-HPUZCybhI~{84mSx{CLlVFl{D
z{W<vp85NYdb~~}?-p8sChwD6m-E1t?iXpG9?|C!Vr8-sSk%Nd2U|`FGfrv$Rr(K&X
zcy`4NIco_`x9m3dvFjjUO9cU2d1sBlSZN{Tczpbd`4ztvq*f~PmJ|ZX$;Hm=!7|t7
zA~ay6-?ecz{=vlI*ZwTfy|LpY%rd5yFFkDoi^$-ueNCS2gEEuAV*7Hf#Qt$Ea9lN`
zNOSRHQ1bVzYSm55@sR-BZrK{R_5woT`Od*70HIS*cYUln*Ea2meIR!4Y4{V0V3^0R
z-z#|CzSwOzGVTkAE%${T5<l+nrDE8EGG(Msu44SCSkoE!)z5KrsY$~8wp;Igz|5J;
zg6Z7|W1xU*oO?|5`ZqH8CC8JJRaoz1odaOF3$(YKc&=X9dcM4=ParD08`yY{EBl%r
z1LjM>L^pfoa!$>cLoj6lki=LJAF|W$@_wau;U<0QR4Qpw^%mk2>pKlI{=gW&1H`Zj
zSd9v0H-4>~QJ%s(ALaaG1XNc6^VBG@FO1vj3`#iDj42ozLfy3<ol5Mjl-@jWLRyVY
zYwp0Mk=ergCQcLY-_qFK!{${_nf^Nau=5Nh?$9QFP?PjcOKN5+c?X=+A%F2~8=|JK
zLt|d-O(%?+!{p&5sRv0i(=APnMloV-twwwUSQi6opmmt;^se6V4k?s3KUWT@+#=Fc
zFNmkw5Xq;!k9*|YyUdCSU+xnsd<LH!T<A9`f04gB)opHp*H4{4YCyf$@0iJhd%`1<
zh;EMeb=D~xVc4_&FqH;~L!UQuX0{d0inloEe+)9Xwv&P;F9Ygo(IjItY|;xRFW5eC
zG72nmX0^}O@!bVz$Gm)_8hbd3aXlF{Z?C{=$}L<wsYKr8jPiTrm}fvw<UXc%YkePs
zDds|P8?`<*hvy^(LR6~bahgdBMb_9;X<)&d9yuw+@wlZ?W5=#9?-ZmNks%1!m=iR=
z=&K@JW(VS4Z+er^LXGJ2TD95KhodrkLs6$D(^CXt*J12vs%B^2U)&6|vr(4b1nVY3
zyaX%#LDHn{HzX}}WO4l#>|W9RMTpfgQ;9G8+S{aWevY6A(IALQP>+X8*<akWUbqku
zV;6+2Ep0APh4q*(d=a8^(F`{8{D$R+Vm{{pW$-QSd{BS0cV*?h#sSJr6<LYoAzCm#
z2dCjY2)E?7Z}S+SX^K}bKk6es&`>EJRja};C*i(UjNxUK_$NM&2*K49z!)uU6*5OJ
zv(@BvFyGl?O2{j?r(I{Yx~8+(WEij<-t%??7D05Aanrf^t7n2bJH>Tg_X#6RI01B>
zTvPf{o_n+<g{qs%czY0@)v9D*rsDm=XC^U^q`dEyGK1prJDeHL$jf{^{rlmq*^5}J
z?Rjgxt^I}8YEWz9o}XUsoO(xNCQZv#o?QR_4|rg1lS^p2|7EkiouPf$;qDTY6yUS-
zWA+*C6b{l>8?k58Vh}0tB%_I)-Xx#V@bit%;MP$GID3J=Ea!bMm<L$MHCRfemGKdZ
zyaKBhFUk-mTXX$i1nf(fN==OWgkyRsA<-69KBzza$C<V2m?r*+n71pi42I$w>Cca!
zc-gFzXXt+Gsl2HlIoc((b~Nq5A=wJk_#ks;c=v@t8(X%F6y7|?u0Q4ob{t}2`lOqc
zg;n|(W$$VsSJT590=P#)%XR69V&9wX;@2Sz9(TGnp(*Zd@tJPYyQ%5rNKxN(`vZb7
zy(0iB#pl>QOakW@=_<qHIc5Z?;(LwK#92GFN>A%tT~%K6c$e%;dIwI06ZRqE%O}JA
zd_O@m#)=2(oMhw4R%$k+N;`a6Cg04?2BjYA^)LV>ed&6uU87YynC}vVuIuZ2{KU_q
zeIYR{4j91V=Ljoz_m^v3Sa(rUST<3OJ;qT4>HFlHrHVes18nkfOWu@hkSmerKm0tY
zJOHcYr)Dh%yVNBDo%f)*bKPQO>vcwP)5UsFb@d!c7cBN!UwZp4L~6JVdp2_aE@$*@
ztJ-X-?v(>Bn~~yP>CJ1P)6&W>qW7mL0Djle7!$2OFaI(+{armUGXR~^06Ik%=bSmZ
z_$N?e2D=L(XTFUV=xw>QupNTkER`-R#_$3I|ICx_a;a4Dg70bgi6#AhwI8zp6Fz`d
zPq7{9&%%OW8VR5(luc8`<?H2Gy!Fti{qdb*5G}MW>Lk5|!gKkg_ORVC81x&=O@@_7
z`7B>;*`b?w*?p2AyqQRw<TlBZq0y0$N=_`4s~{bY?yZ2e08p}wnRnx_`-Gn?t8;0o
z1*T(HcC@q-cvMh0V}vb+gGM?vCfv9DjO4PJQB#nxr!3efJA>AHVFiv4+0|h=e1lWw
z@OI8XzOfE6aA_x3*Ck7a<>v*bfEL~Cylpwkg5evCj45{f`j;AL4n2YQbD)GQ$Zq5b
zZsCmfv_{w!Llf7vg&G6B0Mh)Jk<z+X5sC0wgx3#%Sr&etX9p4O4jSWx;(#UAK3e-#
zFNM|JLjXITpogo!X6fc=sf(=+A}zZ#%1PMuyT|a6K(Uyp?$v;&C_)s)g_k%j>1Sr|
zl2P+xICL8X9~|j9AGv%FkrPneQd~FaXgZgaDO0%K1`5cNeU7H)M+LPTb#)(Xv;u_U
ziQ)ITVm0sg+SYX`I~%X*KY0qIEXzC%I3K9N_!|41diOlZjt2EoDRO1&KF%&L5h-pR
z_JsO%8hxE6bl>ZOcsbar1T=JW$);sOL$Wp3y#0GO(<cGsIyV6n#VW(+`9z*d{lP$8
zmDFx#cHRWFxh(P$89e`KK|LSGz!h8A+A_#@y$})MWft2DpG}sHN5y860An6@)vbwP
z?xniM`3mYt9YLlJ5<1>L6({`7?7{Dqwagq)h|Juw3tVx81{+~)t050j2-=_Ip0B)u
z+#nrCn%`wsHl2)!jMuRewhzMYJx^d^3Bq4lIpmFpD}L0=4H3)mZ4h?z0NvrZ%wAfg
zn#?aZ3Vx86JQdnZDPlg-+cl^Gd=tQTOp%gw5w1#eaV}=i2pTwqiB1yuiMH}DXW%F9
zT{w6d3m8MCEyt?#V=O=~9$cni2ZO<I>bA)iTHf%yHl79fnl<H&%sL27rbV=7&m!0#
znt`cf6H54`lW0wmO{7rB+)j5aG2ng^G+57cCPJ(XSE(VcCLP5SAAr>eF7G0Q9!D@S
zeraa}w#g-%bL7RLe%Q&vcXv|X_Z}AAK1s-r2w<35Z+vRupU6wQ-^>9hMzckTmVqud
zzjRL1jb29Nz7Hn0@VO1i!eexq&3Ik2{gcgm{0q8$UDLM!oAy-&7b%#!zx33Ql?VcD
zbvR)zq@q<<(<{eUB{xN5)7ntLpzg($^3Z=PeLv4}_rDFd3@EHKXq**;4(CLijp4i7
zpdv-a>55xfS;1KxIOM~tE)hn+buAR&MwtrM*w;PJzsphrBLdvQ(fUZBs%2;8He7kU
zGGTK*Fw{=>T(i!8=PF6Qmn{RhX~#{X-RVVCSH9NtP*uNcAq45Tp9c1jRAYX?3xx22
zc?jQaHS52%fuGK@)a!l;O|@mHE0q}ZTJ9sjpCokP8;hsHep+h6Ipg1N(nv^bEeKq7
z7@SK@RDDJOR+Et{7ZbJpysc7%#sUF7o~th{g&>&fyN|We&TD5Ou+uA;<c*H^QybEY
zUy1bX^MDv@6RLaIUo2K14apKbuzd_Ae3}3ZIFSy;aMD|h<nXt*r1+fN3zEOfSZFdU
zL_KhnUfu9A0g_-$s;>crJ&tZ7X(iv#!{|ZY_sU`!ABy|Lk<Q2PS^)0Mu-dgu^zFS3
zFc=I9Kb?k-2yO_{+45(y5nyQ`I!{keY!eq(YBCjZXHhQ`98$NhNvluz%u;M^xU_EZ
zTMC%D6v)C&*h8ac!OYquzVf^&@Q%Yr^R-OfofhgrK5fgF4j)yLEIh;U1EAAcygKM<
zR}QFcrX*(-+3MVUt`{J0wfEq!w+6nW&@#hH257UCDw6O5^?u|33S_szoAua@g;#W%
z(AYW*PdxSd>p+)KHMplFx>Vw?-fng<Y7t^Ko&iQzeY}!3M3V3Bfq}T4v(Y^ta{03L
z<Td3nkTuTj??M5eQF=_yu<g;+Kv-bBh4v#!@>VhP+$vl%JFJljxhQ%`FFgI%@spGo
z_zi;D`>mwQ(d*~LfKvA2qKLs1-J;@|;*s-Y1Zb3#dKZD(yh1s}^|<dtCyvs4<Adjz
z;o%w#`^{L#<PHA?#+@&ueFhPiQ{Q$}D(e$wAGct^SA+?fkPVs7zp0$GU$HbfWbvF1
zFkSoN^)LH~Pjv}-zD1wd<U0b#a<Gf92?#F9k~5Mb%Fln$OJR-6+_I9rpOc2Ig^?3j
z)X)wPObE38dDcA@-yPLKv)wdAkq4hR{m0Etks*n%l*r4+2>3A~mWyn1aR$$8O;mpn
zOa}+dEE83b@M|T$v3oFdCk%N>;#~|lX{L8aXfzkNB4ha)gj(suI%}!RLRBjJRkNCC
zTQ~;@V&EoHZ;!8EKNcFV_wq;DIqdV(0?L0TpBD-%d;;As)G!fRsZN$u#ZF$ZD$jLS
zWT6KbDdp1>nCbe`f&6!OL$;h{V;Y*|@4;3SpzDv)UC&#$6F%*^9&f~^wZzacumQ?!
z(yXG}SMPQ6!sTILuYZsyEtCX}zh;Pdk}#D=4pry$AN#p`Z~S#wB7bB2d9!|pIZOw4
z$DZ6x61BHa9UKN1LL!!D4C1`qtrs}iQ@0!E<eA{rEnx~_0w9Pp^61k~^MIK5I$xcX
zpx)*OT1Qucd)$j?%;b&rF|F^de|8c~Go{BY_s+G*&?J_qD+%z~=Myj<00cwc7tGAJ
z_^*U(fKq3jm%YaX=rx^}Jy`gr5xUI>XG1S|LkCua#@7r?e=M?^2urIe;~T|L7Uq8h
z_mN<iIY?ueXQP0Jjkr|J1-#v8X`$dEh&{zn90pz2&cMPCH2C}7VbO&<fjGC?1J4uZ
z3EiTPsH*9^X6H~D%?q!c(B06g-10LT<%chd0_7ZjvrMNZ`Nw4+71dd1o?g?POV}{S
z1DEBbg{&7p{LS@^k7du&+;*lxto*l`krtVj{&Mi=L-)TWo=d6su^KG%S%1V9p#~UQ
zCT=oYA@eY?UM;Nm%>V^G%oyB`CEca6OxYOP*F3+?r2cIxY@SX{LPD>U4^?M0yPq6h
zpF!V>SUv2N<fZ6!XT@QVBt_usbOdL?_@<Xrg-Nz$E}@;85`)?rIlaxiU(uh*j~1Q)
zg{eQ4Z!~yNO|y@MtE}8;v1PYUUh$ji@p2#5G2-8<?|R~@kI|sDztvBP)T9b-*jFnO
zA%i3`c)2gx!qSxt%f@DWh#RGDd1TmH?%btwvS?6~J<Jt5b@8t9TclX$lfQ9axQ98;
zTb(1z&rgn3HP7i2kV}ZP;`}&uysxIlZ{5hIv@fTCg(coMu4hL^80<!%uiK0ZbuG>5
zQ0v}U)|a!6u>tXxa^QZo{@slZ=u#pa)R>MVdVC8-3Kr^p^ceQ;ODq07MCEl~Mde1F
zXt_Jqpo>7fOu|exwp>zrhW0zTKHtIGXoJcVUWYcsCO8{lNdcrdMsC=PUz!mflC;&?
z3#gG9HtEYVPv)zAL~g%)WFfZelF&ET$;cimV(hyZChj(4c;lj(oIUO7&^anQj!PE{
z`0q1^R1nDKh_q!)L+5<l)ard)g_d4fcn}{<s`e7GiMuh?@}W<>aLLcEmeQ3#)FP<B
zMj4E+zY%Y*dPRkj<glX0fj)+lVuLvWiuLCR5>#dzT`0Q!q_eZNtFLSrY!g^<rbjxB
znuOzEgj?@d6XdLP+lIpdcs)m(5vdQQ*1C=MHa-43u=k`X+wif^Cr`H>*$!eJuZ3@J
zgG+eUWJ!vCf_<@@HLViwdWV4maQDS#DnGmW(&uv0Szn_=ZYSZvL-0a>#HJmdScN+?
z!$i?z-vQsFLN}Y#MkVItz$Fk^A(|Ett5-yil$2UzE{QJjN8U0~jF^;aJE78ORgOY!
zUHc8Oi)`hIeX(BW-zkZdWn(m(<8dRkSpyFwBjqcMqE(;5POpF&`T*SedQb1W_Mzi<
z!{u@4bU~e5sdMsbw(*bk9pA3S%xJ>HsbvB#b*-m&lc}SmF@hqA(PtAWm%8~1+tK2%
ze-R!oOs9#;+VD%hPSe58>PLuxvN(`WhJWc$2@xBqqEPUYL<Q@nk^~dTS4Kz=0(y$`
z)*>aj7)^hC`QxaSo~qTG6dafS^Rkxz3Zm^?|DDRonKu@bw{^1KH3l1U0+YD;C;z)#
zu+uMC_e)1!RuPCDYmtX9_<NZMmTr>QT4knTG-dj59cdw(m7lGB-aK7wg{@XWg?r24
zg-|1bJXZY=B*DgJfS%BP{u;;1YwP@|O4f)IA@Fl5Yz@aTE?3LN*XuZh>^yb$rQIpJ
z^{lx}5H-`=ux1TgF9J6Z$Sy9Hl+LWlJFMxG9ahXrcD+z1=4(BG^pEM&d2seBtbW}>
zzaw%Qvfy<Y?QMC{6_jnm){2|t=jgT%q*b#wy%_;7xrBoVu5DjG&zYPWD!Dra$kU<v
zOruRPPE!IQk^@y4_@!EGk(t$bQUZ1xL=8svi%$1=c_b={yvf99zGK0SwN%ik*V)_a
z+TJ8UQx=4i*U5W&?<vt<2-l(qG_Uu+v^BZ-3b=^4ORo&)2plkx$haB5w3To<h{?DO
zoqmq8|1Xv3nIgA$iwCxxeG~+rze<NUzDs{#m{(cpi=FFv^>`h%FXPbGbdEKzZk}oN
z^&E27=yduB+cw#YsHIj6tB=LkNv7##MfX57m!-R`%K7Cao}G)(BB0MsWHEjCCDl_F
zf@#_^fRvD?a*;*u_!((xrr<Xy)MpP;yuxu<+Y`LNCiljotlP`wp_oW%#ur%`V|1gH
zG`I`e5L#?a0Q$g){nSMLg7523$m>kmw=gXZlFB9Bx~ZEp%(EHE6ekI!dV-2<o8`W4
zOXn!KKd+?kSEq1Cd{Ddi*~;PWMi%z4721f7fb=#H@zZWz%Y1$pN$;L8WfS84NR1ch
zda|-eyP9<(_$2&P1Bsw=KMz36)F`djowfi~9Jb|BC7F=Z`!%)T#v*~Wryp~;Uf#X_
z#acGP=`doy`nyZKvn-V<DheO&Oy=CjvMV!dKrb|b(G|-_xH8|u(LMu_u{w$*Yv0bq
zhl8mp;pW!H8j|oUo<Bm*xOrt+%RO3h0PuB4AqQb|Pj%fE++IXnmV*xiF(V`^w>6z-
z*50S=R~IV#x^CO$GIXYIvwrcY(O^_M?2OSZ^deFAq??ztW@!g?jwVo!WG6aZ5i!v-
zVd}lwnCB=0GR`9^>%lJtTWJrQ7Yc=@&u%h2oEDJ_Ol>cEW(~^%wo57)djKM=Y343z
zIw(`N0_uXmOfIo$9@M}ox&WSEUi$GPz!6mFp9E6+5Uru3u`eyE43n$TP5Cjas3>Kz
z1@7tqh)k@>y$kxLdkVVfl1zeu_BrKqFW5F(ZClC4=sm4^M&LkZL`}(?GgFd`8aj=q
zL(|9EPl`=bDpdGe>#Ok*5Ya0f18`@h=|<<)hCx)pr;vf_lbX7qRx~-V<rArOJexq}
zr9lw9GikDZVWv61OJ&A}P3v7Zzu2^eTm_T0uF44l#O~r!0WtoT#6Wsw2owX=E6cbz
zx=R($80U5j5Ioe}hxMM<2ZU3ypKaaI?CHgA3fqd!J*H?kIf_(h%ZipFAS(vUo07g*
zrYvP9$wPeo4C;4<(_0u4V`p6h1r5>~{w6@KnW(oNzr@m+-?VeAyl&R%{B2=S8mzTW
zBY<a%)MG#?0z9>vMwWn{TQ5lks;&zl9pj@~)$hWWL9CC5veyHzA|{lcRDBFuv>+QS
zK3x%SO(E?4T0TN^cRat7c!-e=t$f(Jh!>gERY~<HKdjg~cKyEVdp`lWwY!X3>bD6{
z14B%2T(1-_$1rMZj0^9O6AV=KQkIp-B}_}7Y)EIJ#5uM7Wt92>I-pPW?-b=MHVlyC
zkJmJt;BF)Ue3^hqIRH{~Vy?~O{QzNvJdPx$21}nUXo`z!xePb|3@zggox?Auj&2p#
z(cLV0eG8+hVqDRoxxOm;fyYK&0e*gr>Ul0Yc;o)<na|PIhL;Nrfc~})U7*#NCTIV^
z{+m^WIL-3z$+nVbt|HK}-g=##t)LMx_86tmk*j#m*m9Xb`fYvcwBl-RN$TjS=N>AH
zcEJ9%{6`%OEe3Q9!upK-xw&Cb6G#ockEn@_MXWq|`7OZZ*>Vu}TYseutXrvJZz8w8
zBu5NyEta=%ij1GvpN&tOSfaJ)D3T`T_XmDlk_#JHSra74eqM)n12^4z^A`W!=NI^X
zTSykipTzCk$SJY-SK`u4X>%(w_vwIE1%c9uImwW3VY{`w7GF6hINfZa!$wO5`+m{#
zoB<(1Yeq=Cv-+ktuQ^Xxd!j)=Jb*X7VSUxPZncsw=a|%g1I&ND3P8k%01+9!AtK`&
zOrz?$57z)v==rnV_kMSDT2Q9H55%*iwa?Rsmx(scr~hsL<u4l5ju*?>B3ql!6S)m#
z;HD!1G*~`B2hXnwlxjq2>gjBMEK@4SG&XYttND9hPF5y?yAuQm)>2o^z6W-uSIeE1
zWumCHLUCWUKut~5So;gvGxz`#M2hOck@fWp#vLG1%mI-2h4YdNNAHm^R*l~-s;hqz
z|HRAoo%K#0ayuRbC#);$m)$ZQh#Lt+3s(@L$r|)27WFo|)9>4X?;4{y94$5;>BO?C
zC^vWyc1&h3f`YE<wNUZtJX<%JNs(8tFq(2XzAjE1h3ZN<IhNkJ1ZX`)__qy6y2h6b
zHr;8nzhS#CEzsDf@ySXmb|XCSJwr4+<V10W0;pG%9WrmTx3d9jdN(aCHXhK9m(d0B
zLvNh<R~i1X8bOWzIp0FJLUV0<SFnVvcYBtG#*OjwF~5>O8|4V(5FF$~E2e%a%MBd8
z%8ju<W&vmf3raAW#P(Z7X|oUYMQw9TU7)vh!K5z<orqCTI(QhZ((q|%zdGHU<;Jvt
z$fOt(&9`GH0+N16tiIO0wJ?Pjev**ZLy^SoJ`aFV5;=aE?*6Gi2pmqIExs3DSNg1x
zh=R|aO{?Hu-<#K3hA&6EuxHb)8;1#mz2@Xe%xX{K$76eL+%|tD5}Mvo%8)8Kv-iwl
z@64;YmGG<fBEr-?S2h-=)-MaH0h7GyW_+<Ao<+1*lc1a3gV8no?AK0dSK6O(G72*O
zTyg#sjx7$St=Pay`bc;wlT3l2b&FX=rVoSBtUOgVb{`+8XhwIB?-V=}s8)1A@ELzN
ziVGs(1tJE#Y(*pC?+Y5Seel?P(#xe{VQK9LRHDZl1VP@i8(z0j_h4r{09`v;kX}TA
z^nz7E0AMY*3tld@CjdK6CP%;PeJ9@)!rP54VqRbrjKb{wXN|=`CU)b^Zx($%*`PG8
z<Em1>#S?4}GD@wn-g|W`UZ|!-ZdyG8WdJsZMBG)oo_T+P?y~!dPk?)*D_(vk;Co$G
zEaiNd93gQYAX-`~DNVd2r~{;lax=NXSnhR0F;bggc!uszLBx-_c~PR93_(^M_<3_|
zs6tlB!$58z@B)0Vn#hj(1&BE%kr$jHt~o7Nwp1Jmn~x_P1ETv6Ah=>X<KuitAB;c@
z7@D~U1r?83bcV_*2*mfSBtzk<dXM8ILw?W!)rA0Tk>782R>uFr@5T1$sa~hI7NI*l
zAag7CaLy>YR$qp-)H;@Z9}Z4{x%YD*)`8aula1m73Wr%|m0Zj%57p07(_%}<5?V8Y
z;F9_rKXde6awjC8qV3#|v)d0fy7UQV5_>QPBuRcVZWew=mGdCxIc$UEh+mb<HXE&p
zgvXs*xLRT-i5%%>FS<wOAdu7Q-}=AhMtc#3REnHSC3*OYrW;VX^0Qx~mERwz8J*wk
zESmJ7{=MZ6pP69gTtNe%K8R%#m#)`{GXxSaK!kS1wY*z=KRF=rpd2$OKD|zflDa(i
zcmO{80;eZU!Y+Su^9qp%nR)t<y{U`!04djKVWdQgJ0UJ>0~ZzvUh@!#xdWoTkA3LI
zBij$L3xN+l&_ZukgIo4li7nZEW5dOsi4h3qgx@{@x~%L?%}Oj9&`bW{;N5`-!-Vp^
zp;qWd{9{e(raf}DT=_R^$T2Qf<>FqHyhMjUNExwlls0xr2Q^@KB7_U}Y9}CF^!j^{
z@q!AJJ6D<hi74BD#{)LRD)HvbXP(CIi?Rh~aoR&6kABNGu7R-mp5FC_Z)*;Jt5&Q3
zPCDeg;H9mbSq3>W-^FtyeRCt`UHuVdUo`?U#bpsmf!x>^Lyp0d9;Y3q19Y~JDQuW~
z&?w0hz>h<y&_UL90m*IO?!lEZAk+X>nMN!8eeB)Hv(J8g#o!MS_iKauK@Ueu@<tPm
zeU$_nM_qpVYWik;*kBytvIkXB1ezzc1RwRJ^dJRTGgpq$bm>}W|GVMEhT91Oog!}&
zHovmI*mXTNfOMuEmz$QqhJO6eV<yAsot89P0iTHPMI2!Gx}5mtzCWZ13wrnCgdM#Z
zhwdpcs__z1xl4Qfq^sYZeGE8op|7%|rQ%xMYl9~DtAwn_pN#j^O!?%2LgTMm642h*
zw3?PrgWAs^?P|U9ZwOIAg!0Z7L4-`?&xmdnD~`MUZh;N7AfpIk*#oJd+`5%l%vR5S
zbM``NaCV8nZfc;@<J~7yy1Y;y{C`J)2O^LNa;gp{P<q&@iV%zi1a<n^H|^>v`J%kG
z3%ft_Cn>F*4pfN_4yI`LOZ}@+mY_iBs<=vBu6QmTlps390hsl-+iZh|S}|VHX6OO6
zJPrI^G9nUY%>*Ir)%Hid2o?y|J@%Pt&}@s3ZrJ~(aAVA7(u%L!bdVi-ud(OsbEv!U
z^$M*9O*Cq>FyeF+{KXyMb^y2y#N4*Sf3zg|{0NALv`_ww&}>QA{G`%h75FM86&t{+
zaCHk`PuKmBfXGTQKLSz01s`YPIf{Gjn_uIy7>rvD5w(;jpgIS&{#`$*t|NYCfeZI<
zp6v>Qheoui9(YySvsF9%l^Xq_!S|#<!M|1m2QG~2I&q(a8XkkGif*HQ&y|ZWEAbN9
zL9yju_{6dM{IW@dzdM-#=3K_j?VeG)by|9e`Cj{g)33y?j1vwmqd32If@Gj?q_xfc
zHNy8;fnl!DO4(SxJr>K})_+m~4(&lShoK`M3O6o3E$Q_>P~^pnGmxVwnHW?RpY_Y%
z&y>v#D%G%#0`#r2h?Xel_>JCNpGmYNVOGMSCoY}0*{HwAsj_insy^Om(CnxJ0qA?g
z89L|;;#8kUg;nh;ps2vzSsS7%HB%J6pZvtV6YvxLz-uJqctDm!{(?$35ES{;!`r?_
zKNFa>L3(^(WEtP}X0h*>`QirV9OFh#PjmM&uh&!JAgHa^S-=qM%85!*rdNiarw2nS
z`!KX76Kc!8jHH-k5UufPSqGIdP3HTFD{M?N9tHP0;q~N&H-<R}l)~c*H*RYSY5|d7
z6Ok5cm69BG*ZX)ifIv-*sEj|<*~aCHdAHqhqDKJJ3N+J%*0RSg#?XbSM6_3R!(ZxR
z1Bc@Wp{^I7R|IhoXku`8E`G1dM)E#73%^5jzdR<qDZWmv-tPT5Q&AxDnnrzpdj;`n
zzYECJ%(s_=;SxVclr{7n<;Q$S2J*bNbEj_83+h9sYX|WWJ_u{V^&4H;i&+Lq<K0BI
z7f^UT9GEq?{H;mWz|CsrMTnC(Ftn>y^F6N%cjU$ZosI_J<UVQVjcp~ul5^NBU|QN5
z`%ve*TyT<v7vuoeudilo_%vFXAdzl*!)7-KR56HvUJgucx!}2K42EnFeLIS&e3GNP
zT|SVNuqh;rBSfXsr3IFIQv}A!{{RP!c%9Xi{;at!<JIdRLg8#R(J4@acQff!&wxt!
z<Gld)#<OLi7Z~rhRY7C6)P<(o<DCMGlzYJ+dI0u)wYXK!lC3~2+Z8?w!bFhsj<iWi
zpUs=D;pux@odSa9h$&*|$N37_aw+;9cnd`5>dfas#=3;9D?sOua`eB5WKyIsXz_*H
z6~N%L2!F+p-kw;W`Y%&g_#j7F3jHv`!t0)8K6=X=pkZDR$0gsBK=Wj-F!yqRwn|<;
z4LpD{$DylNMdt;ab0pswWpYqmVbgl41H9NvawL4MSOvV6J@#$YfRP(0!E~#fXz+3e
zbl&ALSF1wg@(J9JV@^TWtuB%aIG6i8?$%Z9mxqoaN%Xt3JjW%p7H`iu7y6`wo-NBu
z07;?eDBZ;OrM;ue-~H0G9W1=V_971?jaNz5JO0E4L1^nm0zobnJj^zrF)>YC<jr}b
zLNAJ|GmZmXP<p`wdVrv1(*_?P;X;YZI>3~MP*Uj)ng5q#JO&5v1P8ab1kF+%Jzy;n
zn8=b+&_t>9CSWKsv@6=6FcEfa#)1bS0u&`;>e%lT>miZ4Se~@x`U4KFwwtz~p5FG}
z<6|no)F;(<;T7w&A!bs(segt+#eTcZp|F5yZ+R$p4-<fZo(K}yF+Yqvc%4#vts97r
zkk=y&oxuWq=ecg_1^`hED0r8`hzgyrDy28C+rBC@k^cHUkcZ7d-{*s%`F3I`9=@7|
zdqM0p^yK!8p+(1fR|swyoJMJh9{g}UHn7`p!EX}wBCJlk<2D{m$3unb_6vZ+vHkXw
z9nYFs{wkl=dTBsFlZEINa}%X|#M9nWXz^g6u;|p66eQ?Fn8@BZNf2xkQV{D0RwGaK
z{;}K`)uxeg98$8^SbQZw&Qhe~HNrj!+zZUj0tvSF(Ts7;Q$YWndK5q^2j18?_Mnm)
z-r9&JxW1Bu(sc7;W{TDmz>8`ZCgg$r>N*M)M+rOE=iqg3zJ<#bYpqSs??d}5iTGaB
zxt0r$O?xn&qz6$0TfRrNa+LX;nSfpPsDf;s9p#%ucnu?IQq^dk+i09-qEdReA>^$X
z1iP&KB;__N3G-nQdH8<PTc1-=y<4|I=O$@tXNOOpHE&tVZpac|Cxom*+Ut!i#c4kj
zv(}W|p{C{11qNh#e#44@e+pcN|1J{(o8*%FoJja+`Xt}GP0s?UBcOz%^CsQHX@1#2
zuLy#U|B??Tr^i8M<5Qe5Uc5Eu*|}Cod0owbD^A0%+$JpK^i!b&BJ0hiewm&Ws~J!+
z@}(Eg_w<IKAAbI7XM~4eUN+m6Vmt%N(_TjNaC9@s1rGj(A1}~K<6q{R^YA&Kp3ie?
z2kLpn2Jr!kA-{3K#owyZ8B8W~Vo>TiMgKiACm2m?YM7eXoY|hr&T=+MH}_7Pg!)r2
zqj|-q`H-^}gn7YzG8DDh>g~4l8WgXcYIivJ6mYN&yRt}lRXXQa6A@lxw?0V&{x|8K
zjVH3BfkA(?%YPi@?lZJuKsf5vyvR1}UcGS<KLvvS!jv6d>;NghPS_wN;UMUFI64mR
zHX59xdF_e~RLpxhc{KV)IdAIOQo=elK{90%lyqQ9%wp-mgzTgInH^$ZKs}FPINVtM
zr7k?c-9pQO!MD7T66g9HMZ+^pJB++{N^H6>1gk?h3<ZDnmU(tK-@T=NvCytG8{sH-
zBNXDvDX&a{Ehf>X=-(*GpXfTZvHDM?iDZ|r-+(g4k258YnWaV5#c~KA?<ILsM`Qd<
ztme|=IUx<mJeQ@~thSYc-R%h*^)P)$$zy}%)&5Vf_#?$rTz+ILabjEjElOU5C2%qq
ztu)Iy-&`uq?<m{IxTDAiPP<MF;$scgrF951Q;*om`^EPBhv4@kwf-;V!|y@I>&4~9
zE`_ihnf;B9RdYGWFy8-c?_0AP&u<}-M&b=Z-}D>&SLR2w9Y4lxUcYzZ*FTZ+y`CR_
znQhw<gk=oZFLKaeq9KR=Rq{LZ6D|={qa6CBTSc11PobRjP#@3rzk-ZmMHqDRUt)`F
zND~8K9ra$sFjbKhg}iU0UwRI6<gv=9#K7O$=8u>FiscFw^*O6=4-u$_D}nco)f<h~
zZEBtPo9qIv4-9C6%E@2}s!3P0`2vAkWs_Xs4FFMKNa0o7kr)-zf3joWOo1{!g(zMB
zL}miFf?BD+#!qCl`CM4<gG|Aj3{2Xogbj=n&i6ADe?(KdU__7O1q!%Py8px1b;ncr
z{qee8GAf(CN^VxRmSp7?GDG&JGO}fF-G)TjBV;5aWQ6Q-E6P^(CNq0wuitsDk?QvQ
z=lkmWa-Zj%^V#G5InTS)so2;Hn{%E&A2bV=PhpRE<j$ccH_R=HZikgKTD}H6$EW@H
zLvM{pp1}QC+9P{<>;;AviM|wiBg<0ngj{!%$U1v=)F<E!_}ELD)dX6?9OQjm7RM&c
zq0e_<iA~;w@7Y>)b4Pp^Ont)Q)<o0dYXBiv`goV_A0Z|+A%LuM_eQ!-QPYlJh6#g2
ziJEo9@)8@<nN#E}?hcm=+y4QrU}c|qnnLNzx2T|z+ave=SS;BLPh8b9?k_OMvvxh~
zr7RRTS)SXL?$%b3Lp4scS{50Nh;A@wZ5s#c5*c=G1D{T$^*Ud6R+YkDcW9=p;1zPd
zWiWNL+5EI1Y8sOBpN!;@vuyMY=NO-#`RjK4?bp1GMS2PF1MF)T>#18NFcT;{Pw0{o
zdix`rXdOO(MM~$p)ISLP80oW>cWcX*r~@DjQjEWe;e*sosfS0-rh}ip|NGDl=Krv#
zJ#QbNd+xqJ)gWZw`9<ZHg$AiA4;0Yytab`HyuIKi<71;PG4S*s0-ti)Zy<Y9+QNxD
zC+7@m0TevcWj8u`oozBVs)G7g{(aITYT(Yt)yHA^y{G1Ka_Hp<fb`01eo7U7X4g>~
zN~itrvMl;6j=amV2fy~Ef|2LWi$|?BoJp7=%cW-CPs!_5IU7hHKmYG4@bM@#j|Q)x
zZfnhi{-@3c8F&={^?}dbExU#9g1`P*cSU{2J0dUX`vUaUwk&h92vJrLU+1;^Z^=4+
z)S_CNzHpRlbi)%Xtx&h{2F2@je$+*AKP#TN3MOcPnwLa^)Sc1i%s)VcC!^4vph=eW
zB=>PMzz2eT_B&HgW|s!hP2KGGjQxkoRKc&<jIdF;JAUz@aaPIcVudwBQQNF5MdYH3
z2>HK@_v{)7Ehpi@!;_<nX{w9zgm{Vov6oy%rDD^eLD&EE|2!z$*XelnxK>-{MsH|t
zO)ESZiUa!B*!ai{MF&TM_cs}WS8?igu>ilBiVPdn2FYv%2-Yp6bo!-W3s4(nR?tW0
zh5-K%Xde6%YRP^RA?nuQVn4+pwcce2H-`AxKhB<TYVAcus(r}*#eZKA4x)lkfXk)j
z_^uF5{3agA(9`q<fJ<1*wpx{bO7HcLe##TligT^eqc=oEt~4cX@}deLDOfk~mg(F=
z2Cw^+(PIHK+(%F1O*}1seX6CmV1Z8(9?lMSIoHXR(6RL&@08j<n?z_1fO4)s;&b`-
zM8CV+7oI-iy9JP!)&W=)Oatei9VtsPDCqHjdELIYMjO}T016LeGNlgyo~M?jpXcA5
zxn&fZ2js)ApnMoc4!(rL2JJNT>?PjK*q04b|L4IpbTh~YZ&6u<tucd}BHa+sgROGX
zIf-=T`TuUwiD=R~&QxsHEJ7cY=z>zGL9W|xFK7>nZ+=M`Tw=i5*?Uga7L9waVLha$
z@C4Vv`{=7sK(NVvYBeZu`h2Fv?7zE}Hz+Ge?x_Qq*`@!k+QV`oKm}^)rcK8d1&Po<
zA*nI<z?GhcTs1YtukMvoFl8J8=1N~TwKY-Huz!PO<<_`ui8c1}Ti+%4PBWPn03dY0
z>0n%OFHE;w;~WfdV!g0j&<{%GxcJ*C7tfoFWVkXA6;OkFuc2zwQA$gau3bnL`HyfX
z_lzI?)~9Rsbwv0AWH2en9Md&9xhdN)xpB}<tJWW>`v+EMi&FfaoqNI3IEg>Ffg<A~
zI9@z&>bS}sg)<vAb@V!#t;EOvflNCD@*(%BHig^s!w10?At50Ad~cItyOH@LJKfjw
z`8f5*@4u<tu5y3^U5VwuQ0vP)>yf(h>lc`1!7vrBJR4HOm$6*YS$5E2#_8pxG<J9A
zxn$AhuyDt&#C8NlOSCKm)&;rZd+CgYRhMn2)!`I5t%>#Fi>iIXC#Za#%IZMl)hGis
z20tbRS_&D<E#|XrgUjAll3Z&4CNatAp;q&}%`)G}jVTi&Kr1UitKiPm(W=-+Hy4qj
zJ2!}Iu|v?q+5K*Ri7b8ZFbBSPw|FTGNZ%)5<ETCQsV8Qu6?`|sHu4u%xGz_cV5Jv=
z_?_**kc#YNSH3j~7@>c{rHim#_Ato-tz6pW>A;>_wa09*->pT>;@8^+x@86;gV$SB
z!qKcCZ%u`b8Jnrue6|;jdk4)R{bSAaQF~{1H0l8AJ1}xbTdkGsPeh*FDhz7KQduC5
zz*@J8c?EQ(7L@tiJD`~lRV6?mr6HQBVSv0S4VHJ>u(St7Hms}}403b6<<wH@zCe!m
zHULyPSwEzG&bbpgOD7ht^SpQxD?kR700voQBLIRcK51itDFf<bnY<E+yKriGP^|a+
zG4B)U|25d<U@t^B;-u9-bolr&q{k;g_=hZ*S~?P^tl?0xMin_*FB<1e<fze!)z8M<
zf~_Pgn+qC)K;j9tvZfXcO6$RA4cLUTe>{t(@|V$GIi`#i$_%RbqYW+M0>-VpS)muT
zuDQcZfX{lb+#t+n2vC3JEb+C%9d(zma^!9}`V1A-lF-;g4X$P5Mz3)x@P!Oe<y3IT
zG4!?j+{w}~K@g&jb$kVtK=@OmHQ~!dcw}+GDDf(pstMrCdrG02`Oi@<bG{?x<>@rT
zmqM^JRSm|Q(+w_!P=IDQbJfb<Gs+zUk6U$p%9IDd-mPaRmo!`GXt6&GNkl8Ko*ObE
zgC3-BZs>MHd)^=jp^f>GohwpC7W$9t=kYE|BfdZwQ5EjD?+d`UCV;grVB@uk0ZC4g
zr%lVN(x2FGzMy8atVIkdfRR3}uc`;j@q@&H<rN8i*Tf78C^MdI`AyTVSbK&jCj3E9
zu0&5U&_$BvNy<%wS3$0NxHJE`>FzYfQ%_}_0dK4~5ulXFE9TdHFW>IRxNGHEVhu7x
zhBTm*{)6D5c~IGN5Lv&iHdlc)4+Ni51PD37xjB=l0>~;Z)b+anuK{KB->7cdeBL@o
zCzu*LTaML=gTQ!kq{6-EYvfwWEpuZo82Anjm{$>~=I&1VksVDzVxih7OoKJ1LCxX2
z@F#u%<^N7DcAbmhlNifUAj%N!{n*wdQ5rN~A;wCrV~SN}PnLkh!;}0Bz}d^vZW`Eu
zFS&9<I;3FFz0|XZqcrCBeTU^%Z(s$gSp0gp?@l`RPg&c$6uw-Eqk|A~pcCyzbSHAL
z(7G&LIq$&_th%6@hOA>LHU)r#8qbPqtqIWGYY_t_Z`ZCTR>qJ~r0o)-sWP~PUGrCP
zN@!9qr@EV4q^;j%WZ1uI`k?31zgJC0Ga8+pYg^$Mad%51bP>CcbOgail{HnhjMYa4
z*~`Y$J^WfUBqu_O+hABprYMT|()C_!B67X0eYzxv)qe=AJ`A1N9gH3p^V6MDpQY}{
zs!4siu!kq_4PVP$IxhgnQ-n}s9ZOVqBA=Gl_-|$W3kr_rz)6!PNd!2sX<=b7lLPp}
z>h|Q0!sM5slOrlc9?MwCo0kMWb_&KV&RTC$y#tpc69FOi>Kzb{wwlwrzR00{^Pii%
zkCgG(b{!n*b=$0q<fj7PSCXRz4^wyPgE~4Nom(2Nk|MUoFXTD;y;=}55xou0_QLp0
zOmdKkqSl-S2<7K24Hy1}$q#4AuqHOLvbSF}sePNyjar4~FZz7ED;E{xF6aJhH8t3p
zbsd-5^2*jiKwFXH*SBLGq<5hb-_fbH(-AVOl}Arp=S;s!L4mEJan}bulzLU*7J#i>
zvt~7sjL8QK<j!9M8m=eOmCj__g!P0yuuidtzDg*#RxUO%7Q2sW>ln{be&(BO9ro-^
zk)Ywmw%nT}WZmo&LSP76v!m2OoD5;}ZbSiWk(*VVjrNVPH`ojcauPoGWgZb-kfFa;
z+L{)A4S;RrL9F8^(!9JkOsFC%l0yA79$OtO<}{2~)M=rI-*rfKvFKiXiHD7HAcU!e
z##<yn5$y?Gf7?H%TcAlP?s%Iq2<)7qCF}%)9|J|gF-sh&>=F#vAX+6>JQho>2=P@B
zb#lZ^z<6+!zt?Fq7&KNR4xnf;YjTrEKGpHA;5Sy3punbE$!9EWeSSI<cDK7HYuh{p
zpT>eZ#mxa)Ph4dk#_+`^yPXqL=x=}GJ2@?m?fGy{zk!EI@7){Mf8uPl+YoHv0{|hQ
zbgj0?cD~j6HU!HD0^xW;nusBn&+9c^Qrq~&mhIk~%6Qwmq{Vq_AC<J0%|)#3bAUE&
z*s6ZGp+KmSv&9cxWP$-E)%IM!*D-L4_~C~y#>!GO*fv{r#?n>to#ST7mrYUA9sy<@
z0bsb91HpE~%<qV+dDWGL($C2rHRA8YMrCAQ2#l6m;-LZ;=sw|KK5@DYbzxnEX0*X~
zf0X`#Ntbg&nJCmH)`O}zsK>4hNQ&P_szD~uicGlX-1czar7{jPY;_Ri08YbUd3BdL
z41pa84#4r)ffX}u*BjEAL7x#?J!Y@kwfx+(O?5My6kGHpXO41F0-w-<W{avK6~GOO
zJN}vXnJGxdB1u-xJ$d)|uMhQN;?1_6s?o4Q0|PE$pe(3QV9<^4WHD(vN5bc5GT!2Z
zG7?d2se4Qj?C@rmODPy~t>dx~sl+3G142{9U~`lI)V|2Lt`}RIY>kAxD@(*~MNiXh
z4Sjyzy53%wJ5_WFx^2s2XN+{}LdSol459@V|EOOhE5T=dOON5MxsUgAP`M=NeFGiK
z7+lEC<dt<Mp9Pb-tfB(BpDC2EZM$y<hf|70&>q0nhG%j!p=+#;L4m|T=sBm3vRNZP
zuH<{7&Dd;>YCih5$a?MP^4E23`L^TSY`vc)f&85W7@V=M*OfNuKq0A>zMoTx^E5w(
zu76m%Pm~q&4_hiOfXS3eg-<>05TXMEx~%Oj-!<g_7)=dfnt5I^>brI!%R;+jqy&^_
z(a?aZ$AD^wmQMt=P*QN?VgoMuvt5wtzUnLLJ#kwoT`{NDi3t1JI_+Z0!+g+48)lGf
z#W}AROwyjp`o79W-~S_AMF|^{8TAl9&S4rcmjzt}K?$W&Eg&&8E*@;1>56UaAtjO3
z{6tTKO{&577zXP{x>i3yH#*q*(t&lK0}zoh8Uhi?h)DCqBzs|#4iU`=g(8<;8m={O
zL5r>|BIrYhRZ!W)Lqm1w$d^dF{;j)U?C=8Jb>$W2PHYiDMih)CfhQX%1kp?Do}uyo
zKDC2|@}OFlaF58hFb6{sH)()4z>+z|+yvhCwZBHa`iu6Q!kgv6jL4YtK)8bQlRpvi
zfpX8YmklR~ET~ej6*ic|&^&bIqZ^}9?n^3}-T+sCiU3ZRdr)+JuIY}ugS)wFw|h?d
zX{xlpM{kkcIOd)w)6HDZY_zAfx2KG2D}w|+8RWA{8~mD955LJcJ9}L=sw8y5RviH!
z`?4}UVRUqJ1{%O)*0~IQz|GPUn0egdXYcCD0<8wKPAIA=VX>IDq|16D>oNdOlFJv|
z06vvw0GtZ<K$L~4<9b<HB(8mhj<@IQ6N(zx9u5X|BNY)zL(Z<H5AL)PJVdNo*BxMR
zYET|)&Pd_nBh#Rt+cLs_3xMKHT^*72Nq`77g9ZRol2wrgV1Jl4$SsRoJD_}?w>bq^
zq$HUs!_{w&b|NZj&9wPJHRw11K?VDs&^ozRbI14c+*qM2J>67O#@HAZgZbtGh~nz%
zHC}{usy)s)4}PJW1eh!poczrm*vn|TIfZ{XQzfcp5s{bQfYnQPB3S`KV0jiIa*6jg
z5K`<=N(z<~D8>|~BhGsXOxx8x_0-I<gf2JGRRt@43V7RgMKsY>RHQ2&`+Srf6)OFD
z9oJ-i&4!s4!6;;-JE2QzeykmjG~TkUnsq;+u49IUuj7m};NFwBv?W%(I@4|M?o2y9
zwuI)KT7bnePA!+R>iLm0KqQlzoyc&n><?oF5nRq5l|nz#YTD)gQ;i>Vh!dvt7l26x
zVe^qV>UAd2Weg;_SBgva2FrbLpkk2bOP=oQ%2SkGx}epoIr(V<V3|;vP*Hfnpar4w
z?LT*I5k8ZnK34F*ttBb}pFP1d*XJ3C>JQiPPR}I-w?ZRbxuAAuJQH!;d9G3@ii(QM
z+tVNS0@n3kz<9-a&bRr=lrB&7xF4~;;|%H*F;#%#N+rdo^}vvHZlY(PM@BFQ7F&N&
z9b8vdo9a)sP~W{cdtI0r+B$@&=U!I>m-|Q$zbrgzD2L@6D`+nqE1$>y;-$GJ28+UM
zJ!Rmd+;~k;@phQX<&D$cR4eckK2KI#gs>G4`wU>Tl@A@hgW}6EPKXbdYhc&f;$UpL
z_XVBcA*`9b6zQ^Q-O4l2Im#6k(QY81g--h&Ja^Rnz~M2fKwlc~rjEzhK2d?{q*#{p
za`_B(3SL6A&~*flRnI3+wu0GnDfzgymYkT|SQJp1bLKhuD8(1e{Amm5^BML-ovOf|
zqM*qt?xiV<mZBHb?lOZvsj=RP;-b?x=IO8iyy@ock(Sg}ODG)$kV#O2>q=8*^X>lm
z=8jL7<<C~|Rj5=i7A&Y;-4)njH=Ko4)uQi=8>!Q)!*ih;m}nmok~yXluCsjYZc>u7
ziVRpbPtBSkM7O2D{yA&|z)y_cpgO>hR4ITaE8R*RPIw*Y%g-7A;Tm_N6qec4?T_F+
z{IJ=`1x<J<mv{$b_@n^{*Ymj+4o0QF&@~*&z!G2Hy^r*%7!flWEZ+dI#iE0*N+7{N
z7XZVRdcu{L=u~95gI(3ibXAxPD;A9wCarkKab4Z;xX97UMyL{Y`O$q;&>5=>sKZxQ
zzS-w}>i?WBT1JEI{-`&vkX%+y_uLH@>A5EopkB5l)WPUTQ7hF2P4#K|@(Zn8s6Z@E
z?TD7H^Pr8#)uyjnX2A2H7{O*ao+hc$QUsC_c?Ah|y>03E+rl=<Bg%bR5Rv(iBD|kS
zBnK<9LckP%Z}ssPqJ}OiBGek=){$7MO}dw1qgtsaxz2l1`y^;V3Xlf~`-Dxkabfog
z4{Y~Cpu!2jR9D>9m-!}A7OxU0nn9R^WQ3A>N3#^#)t%vM`+|#fZnqa?t+C8)qdt*~
z&-S!zI{060y2Yms%o<7w^a#}%+xSfRL0EZKv?%+Smd)R$@=YT6d}Zs&m74DEeDAS#
zIW|b^q1mNVOL?puUN?MK{M#%C^-12{Q#zJ`<tjh$+2&=@Cdr|<ka;Ztb|#h3+4dMs
zq-iN*yo0eVsnp-2qjH9*6I<66WfH+q(S}^4^BAaLp&&VGp1l5~EAhok2f4WQ@24$H
z__4@QHM#h-@ep*(^X)FsF$2Jw!Gm;FC-1y%+yrMCI&T>C{I6&dARAfi$`HjFLKoKj
z!!Id9*Nr^~wR+<+Zd&ci+s3yQ2G?^wb(R}rTavWx6mR4$;`i(qplT@i)-bMoJh(j@
zhmCJCU8JnLDp->Gug6dw_=7LJ#SN)Nnx_mX==Lm{RH_ibR+#&&C&#-@V?7{oh5wyP
zKvRA~23-H`Jcj5PNFW{6`<`>&J99f6bW_<qYA^pr*~*<j9eVi2{s?%*#1;F_MB_$D
z%r)D2!9d}}-sz#<RW4Rrc{PR>p|JsIk8%3ZH*_9O{dxN#vn+J;L@h+7&w@npL!T~1
zF5mG!acR9(?o!s@SZJgIPQgRv@p#;uH;Ubg@rmyygP2vp!tHUfchb{VvAWXtABRHz
zF1=q=&~=Y{;iJ5owaJ?mZNh|V0oLJe4tO?e+P!^}c6D&E_&VhPdaLoP`r>WEzkOaf
zQwEpSy0wY>W|o%5NJ&%n+$yBhXd;kj6nW?qf~rP{r$c`xmHKX{>X#aPB}Wy5?<bG=
zZ*cWObDI(J>aq|o^UMw&VoAT#4*O!ubgIp3{hK~l_P*3)$L+^V!CFNJdCZUV&eaQ9
z);Wf@7Qe-y-7{=}IgZy@nPY<OOsmWNy>C2`viJT23nv!2q83o8{3SBm_k47LdQMe@
zsy`A;nN^67ex%^Y`TGXi7*1Gg+{yIXHH+oy%@%vtEYA3!(5X=0fg3v7n|+R-Qejfi
zf8fJ&mbZUj0S8J{CBfF4<F0Gz?zPZB)8lQ(&CVpw=874?v#5?M=ic|jL?HL@JKbuD
zp}5&g#aA;>JJ-Z+am#R82A%bG+GWeV{X_YoT`WM!5?MeDqT@d^tplm*f_!w88Ub%E
zjK<P1d}z58AYx|b+>z8KYQ%3QJ0g7{angq5UXtg1LQV!<n@8}ncqQZ02Pr7nTaHqF
zHBPY=W~`p?>UIB=JyPRt{FtHMipwG`CA4_Vb;~Wfe%xLK#Z|vyDkWy3PIvzR3~x{v
z$j%jVg(BSO0Lj_!3N#1d#K;5t|I@%Q!+$~R$3}3dO>VA*tirv5?FYrcgKj4k3MG{W
zvW*N%*xw%@7yIo&%zFaZ`<$+!>ZVUDbf&b{>-0tsCff)@KVyD-;84=X)r8v%V+pG|
zQhetr^jV&Q4I!Xi{rXSbRkTfMp<8B7YL-?&;;ZvCB+%{%*nrW^C`nQ+PDt7e<m`Uo
zl^G<0WpR!SNy^Q<Lfdk)$}|an=wztdgaBBP<8Xc{3#|KOIY{w5QK<611-F;>b3mbF
zksN`YTcg@tv&jBVd18nrF$^%!X#W^hbi*V(k-NZIV=nv~-hN_j(@y*?8f|~=J?kBz
zA0;INPs?)RjzeccGQf|Ei}oKUC60|}eB)-bK_77J2J!^!uhZ`RtCGU9m~kElby!F%
zD{mmTl0)(WX7|kaXZ^eyjLHP0Z11>cNqgVcez^SC=I<lrakN9%%Jb()sCum@Yu&|-
zC7ez}bQMtmOuJam7?rd+uONS@m(f;4fY6T`%L!ud3yu?TO}KyJ)ah+sQ6@dnfDio&
zu?h$f>wf*6q-vw$mPDxeGo`y9G4imtb<B32lF!rzMD}*K@N)StFh7P^O3Zc;SN<Je
zn(CwXfbt6gK2Jr}uVH(XEaR}&=Ii%QZWP?);JDR4b`wKkL<m^<wxj8{V_L3Gu94Sl
zuk$mG{voOmf8+2SRPL^RFK8LcWk~(c74wUJE;N#L)9`pDuF~^`Zi8emz021y^!R<#
z5~OFlIa^EZb8R=zOT7Q_0NS-5JJ@w4UwuV(Sl7kLGNhZc9*hKzg<v7~rAib}xs*EJ
z`*Nr<`T7YAyBG-0`bqXDU5`_+RoKu9Fs3JcavTf0Na301-{YV*ENd5Ne3@Pm(_nVp
za2o8IW<)a}<jalpbb>>+p?cPrM6pOY<5#w<l!u!$VkRA-781(b9hhUqT?M?RmdSra
zV-h}QVLw_%dR^B9+STuyN7TVA@-EivkEEN7FKfN@_QO~JL^q&zgX<=ag--Y7$>U+^
zp=DXhmS&g!eEI+ij$d@!AT%iSFjcC0-p^+ELM#DB)+KR{lg9n*^HwV?c(+Lqc15Q~
zu$}PtrJ<d8CZP$ZR!24HoklE*rWSebbfDU{DYX9hz#*1XUJnRgem$!XK36?q6lTY^
zY+(AkeZU;GOrh^)XuXpl|6IqD&d#-dHB{5iJ8S{>1ZYG_-WaEC=?t&lx_Rji85SB`
zfprJCb#%AnO9y;8b4dbYV~CHx1QzOp<Glr9oCN7G`Ps23eP2mxp)y7=sQlME#J$ky
zJ|jowt<i#<p_D**uT$GQC<+9pi1LlaSLAMvHfRWlHLJLeMFO(ntD5S1|E(=ejq^u&
zDqn2amP|QSu>Z#k`U_i3a?Pr}W(lmie=+a?iX6X)v%Sdl68~DyqK3BkO^f7hwVH7U
zQ)shi$(3RX?sc3yUt0PmBJ)3pIloz0gd8<$q}JwMU8Z#_hx&?>5nTq`_K&_1>mn$g
zTZ=1sjU^bhif@i81}Cup^?aN}A4!<O!t>S1_v0h!(|ZmAG1Cgn-K;6OR^OuwGTl58
z<Gt)gdf#{qh5sTRGR6`%&bSd%m}+hGJx5o6`v?6jK>njfDA$^rZ`k(wPxRi=dGy!n
zi2L`mKOS29)*(t`g5Hr%R4$b+qCD^>DT+lQL9wVnrTvc8b2f<s*p$bEzR#(8s8v<%
zY~k9f&l!&_@}+I!`QZZ68(kjqw0mnDWtW`aOa9HF?Ajm71OB&1rq7W_OFhAIYOIAK
z9r<X7wuMZf_yW*9$w*l9jfK*npG!FEQTp40wO@%)0{CqYGiN@Q+^o+{jV>(_lJ(9!
zvVF=V5?w@a_-h}7Z<UGBRwBe6r6FZm$7|{7DE_2^5v8R>?u(0sDXH^y7lprW`*46+
zISW6YhFBufh)I4ob^Zdsa~t8i-&v251V38TWI&qIXM@=dxs`+Z6x-AXtUD5<*|$8L
z<UAzM#%`1z@Z|L0567wX;aEqL!LK}-ej+)42WbyIjJk#geC#7ulVj>J$w`}ROj_rE
zoX9`;x6&NXxi2(mo~2W-lyjx4r-x@^NOhOMo7e86$caSt;K>nVhfAc!Dk|V~cz>Q7
zk3kb{P8d(v^4_+~e<rpG6IdW)1|cWy28S<aA#DemsMRuXNwA1q3@mNsYFMqdwd<GH
zINeXmcX-pc1#EWzEzcmN?R<8=)s}qRHGntz^Z5Y8j931!fmOS=TXV@=IRkm@q3zAx
zBLzB>*eH3mrNhJHX0g7YuauK4!I(VBU#}qGn#r!=8tXJ5Y_Yd1QCgmC+oe#C3^;qe
z=pfuUd|dWkYeNMNJ@H>hFB1cSk!IGh3gw=!Jv~w0vX%imf-4PqEP>Xx++_2&r4k7W
zSk*?(6s0kACB=DmPKVt2&f+%Z4xmao(MYdE`T!X1-ODS2`ubrgm8IuiddR=sB{B%U
z%4kA(Xsxn9wEC2f{MpPMyMlIy2q@wC772Ww(8g?G9H*}-^1)yJ7MTQs6ce{gp@n^r
z*<!XmzPJ-V?(U)iQXQ>;WO|wa%fGFb-a%*ODJqpgRxX|9p3Uq2JcgJ6577De17NsM
z$J~>eh5PW|>9}>Pj+UjzXJS`1f4z!?a`}C}TgUsn`2svc2I?Ie7d?j@k9P&3B2Tlc
zMkcN373PC~DQ*lBwrP!HcW)`SZV0F2{;p=npn>cW9I2W6M0NM>-3kGX7U$?evL~g~
zY=2ta`$%cH6Qd3}ZWv(;hx($W^m#I}9U0m5f;RA66;JxgMiBjjvt5WJc1jff-_fx!
z#d2|2ZSBvc`W2<MDyk<1cuPBEMrK2y=D2|P{4to)jfGzB(Zh(pIY5Ww#*s^?Q>~=A
zW$Em!Lw7=+0J8-YIS7(LHt;2tFMZ0p^Qk}8AJU78B*>G_^)RPOGa1jjlc-{SaEEqf
z${^@4_2rEG;ApGES0CB(i`bZdJtb@7Y_;80yQd~%r*|}KnGo{S<nPGdm?v+kNNiCS
z_Q9|38iQS*2mI7VCSry+MMDL2C3lvLgAx&^I=*u^OiWCwotC->v@6drC6J5mKDE1d
z@$CD8%!FJMCl*{fS69Z@$h3m3cD6T@1K8$p_BPARhf!NGLzwf~kni}<|1ft*C%E#C
zk^QNdz_KeXk&uJ;ZX*B=-n})Bwk+qshd-r7Jof%L2a(O=YA&tTg7sMj<v6mPr;lg@
zm|tIQ9z83JgX8ML>+xy-i!7}E6{tVg_QE;Evm`Z{+Be2cL(?fgfwbKroWlu-aFp*T
z--K<>J`}9jP?y(#>rH%W_td*S(SaaOFZT(hc4nIGiloG;dg{0x@}jOm8BzAxaVpK2
zgd{f)DQkifk`sT4D>@pd$FH6H)kkkf4q76do&8S$QBBgi*&mM4mR`oBOLJH8J^$SN
zTO5}xy--|>j^LsYu9m|R4%>V80>&o!$DpYy9*V5)vTd(v+7^j~Gx=@TmOpbjL@kV|
z`#XwrRJ$qvLF-)nibM?m_6~%BtLL`ipL+HJm>bu?xE9ZHx??+=+dIQO3Wvt-^aUi<
zoO(&#snhw}{PFYwd;jqgM`8{k*R5Krn4mD1K`yoT(2i*M$sBA|Fn3EjoX-aIbj68}
z<nFFutWW|6ZE>@OEeScPM%$Z}%d>PsVw%N>_|l(=Fv0-F+0B^i-NHHb!&A+se<$rh
z9sycf-sk6e`7)~VhV7TbnT3qAE<`>)DSt>%15=-qRKoa!XIgo~t3G0l%F!h|+mjPG
z==Q0^p=Cl*1eH8&)XTHv&e9)ZipjY4#d*k=@F;&58Bx*p2TFvYCDe(1dOMi}$^n>b
zuaSIzC&L{PEM#=GGz^_v`eP5o;{IsXe4Pf@<#*Lh4WtQgc$9X25mN>^T$cl;I^Y<I
zvb+eq>Qe#PhTrXO?L>0FG?9)gqdQJCACYI-9ji|yc7<ctvLgJ!J4hYh2CZhH2Zsv+
z59LyL{>2KugN>AdoNQe6H3RQmlK5Q-=rKnm<^?L?&rb>)0+CMs=51?N?%wgtQ-4YU
z2}NFKmwS%8`C#qj<Yd050R!dEPvWG338^|J@Z8q-?p<bDg_c5@xu-@GvWCCAeUu{J
z&%CDLt7pA|2=*84lq--E5EG{=@C3+F$)9~`?^c-w80Eg|su$d0(H<K_6wpz&^Z6~k
z_Kx;zW?W(y=#%y+^Atrcd({}8Og)|u(ca%gp1}P3_K^WN1zH7n%``@iN^nZ>>}*UN
z65+v@j5qAA#uPYq#9Ij&QgIHE{=V&l@K~m=XBtV>EZvp>(M35<I_@ruvIpTx`Mf_X
z%EQB>rbW8m_<~Y=^5tD`wrhX;*=HUU>hxG#GIZ4#k-Eou&2D%7bg=%<C|Bc%)|V?!
zT?Ha&mIL+H;wVx5f1Ldrm1VbyLFi8(H}gu!_P(Zl@o4uaXdNhAZ}gW$4J5RxJVq%?
zRmG9sZ`l3wjxo7O@xy-Q2FgZ4rE_lTjRynOE`82IInqOlfz>iRV1jEew&Ofy5&0wF
zLtQ2;`q8kidC7jXcrv$$c!W-GPepg4tz?|!6XO;EqfLtRa(@`lm`AY17?vWvnw0#O
zU^bP-kh6Fi+mZ)Zi$ejL9dg=~z6I#G)z$7QF{9$RPp_kbUH;hrQ<94O?2Bg$<5J!a
zOqFHF?gg_m#~|M$AkTF+KY!{g+Dw<3n{VOxfj@+wT)vN@GQHMm_<=JOSEqW#bJY{O
z7Tz-)FyT!ko}(n4T;pZ4<OqAo{w*>_-h+SSJEZKo1v;;4+vneJiUK?QVMLmHhm8(m
z2i@i)55rMH_=-HTYrR%mLihhzAK`ZxbtN!YD!j2dr-+|u{2hFk{=_DL{_n0_`8c8x
z15$maFqzt=g11k8_W^#C2c264UbdRpt*qwMg*G&VNA1!FT?h1MQrfdLniVCnuuvH6
zJNG_-COEl2=#N^*aZqBh$7y)Ekrgb}+1Y9RN&Q*Gu8l&$p-M~ND|sVuu|0RwbHeJ0
zgA&u!J(69!#(HNtKT8x(*G7S~n&oWxIRB{|4k}}NanUEB4BYHj%w>|1H|DcT?U(o&
z^#eNmwRd;Ab0lF(hqCzU^A;T+Ay@2CM8LCS&JVy?z#D>NKP8OyfxKnBgelh>*Pcpc
zPgb^H#Gr^c(>nLnw5}k>ho7{=Uw!xe$Pa)L6wg8)2LgwrSt_13%wY3sC$HH34Z06U
z)2T9!E%TJ7-E_ac#^F+!{V!G>0f|>XFH-jmh@5Rr7$c65?{aC|2!RkrvqRTB)P`#+
zk5w?$$ZMBN<RIlx{BEkodm8qIN4A@d3U{f(M*JG<%|p9VwjD?GP2#Ld_n5!ZDlm{3
zQTN-mBM0GgXF#&<9`ZIw7ZpAn_@&UWMZ(Gcfr<R@r4sQJwDtLj){!grkRdS=jp5nd
zpPV?5qQ_18lOp(v(QZ{Y2|k%i6rM>wuK#NZ@Gn#oL0h_}o6*pG=jNT}WTp0|^KvJ5
zcgPHd{d9;dQ7d?Ic%oaAK%SC36L0EWOdW5}2*-(|7ed0+t*=^shY}Z=o7Owk67UN~
zNSEc^F8RJjDm7-;zi_#AGAr=E-S=ZCApQ;=XjPl~rbhFf@%IK5g<pZ#ZgIFS;U3i+
z<K(S02_a8sqPgFhk56S8Hf6aa)tbYh!&ynz6Y8|fSfm~()`%|N#iO)EjeTBqJnwpL
zjJNW4HHwq#vkgqcXD_;IJMMKxp)I;<@99AxI-*nJc;shL;Ryx$oDr0XHM=f;=K!Sa
zGbeb~+BL$WOn9{I*COzzBz6q|5@837&~1A?xK9AmR``>6{m(9=gk<I+I>KEBVF>6)
zhjI(;4Z2rqTH8-iD!Er_0)Ko5CB7*Gbf)AuJ{D00Dx62A{73zmhv|1i-EArL6FC@~
zBN<K=6wL9XU~0Jc%scwXT^6GlKy_QP!Kfi#55`@4#@D|7ZQ#$TfXBLbesgGu6&PnD
z4U<;OLWxAAu@QctPMF6Y9nJ$fFb<qFA~}c}Yn0`cyw%2Gw1ii78ut4+7*L@%a?WOS
zy=+cpi;|W<SFoo;Mo`ryBrw77=B^ZpHh9LGRxs@@fl(P`CeGA_75yO!i)>K#Og_`D
zWVFXB1`VJ~nf0M|jKpt@-$s}lcuQq)YwXzw%zGq=+osYOulg0{r7VOT4wUme_UrYe
zAfJH234jeyT%xJmEIuHW#qj&?r4P1}8Q3{GidqgOKbF$DVIZNtXT(N<3<vjQB|4&T
zk|PqeSeL8`ijG$QE`|4~egsWqPUZM$3}`=E|FyEa7AQQ(9)>+8%NQG8H5uf~ved{p
z|J&NcLI*>&1IDXfKHy#A)Xw2{J1YKc*N7l?U9l4!m;P%5X_G)sNwZetVe1pEzn>nT
zB2R4CdeMb3S8u2)xEOy@oMVr}s9eB+@;fLb3Wu7S`h0!JLKpMs08WqB(5|h=h%<_r
z;Afu6J>o&mDdXay56rj>zvK-09jFpT&kGu2IWsdewds=Ao&QPBp50?HI55%}t|?pT
zl9sulVwc;yXl3lAK)$EveP&SIR2Lz~fI_`etS@;T8fIhVpQDGV>F@n2rVWT;DrLvZ
zv~rqU@wwZFb%lN%;w}_a59j-_zJiHeycp!m(phpX)0Zim?)|Q?L+?RE!qQ(%jG?Wp
zm2=Y^&PT$tf2F(*XMjNzpv5Csi&=czQPtD)EmvZ7jn1Oh?iatHgOuQ$@-vbVUC9wm
zp-daAT<YKFdi_p<cU1C9g_nuz%DZlu551mHu9Wq>X&U=GKio+(Oa#BKzxU>ALGH7b
z=rrI?Iw%;^uErdI9THCRx>-;gKR}ovAocs>j+8UCR)w1RB`F1+7S6o&-dQL2WCr;W
z(2{ZeeiD`0K@a(FS7;mdBo*xbzEuDpM^jfCW2C*fmGnN+9EbACS8whS!5qnTnS;fo
zL-v<mQ7-rbTt-*>E?Ly}`}7W*+-$kJ8>$kEZ-=CUwe$>VIo4}Wu9EL@<siI)t?zud
zQCj{`y@f`OadWb%6U$wnJ@NIk0OVSs)E9(%FrPge#7Y}4BSCwrpR~(k?<bHIJR;gh
z5k~1Tmp`r^Ds&|XC-3l{OdLQ(0u$B|M(j?Dq9bUEQqpMe<afdmPNIv=+F!LDrO5lD
zWzEKx%c1?=Qc2e7zkR!qI|6o5E@wal$32afGbUEm-1E|q1V1FREY?p@Rx_7T0X2nV
zQwC|jWCp4G1DJ6S^DmKj<kB$(uWrR%|H&Ra9{0fP<H!9{13G@8LX;JpIXsgEm~w)`
z%h+Fk_Xi3{5%wB^2itN-b#!!O_PC1WN*beouYw+bfVQjb7|M6-_Exbu`rXa^7YmRG
z3$Ts3>y_8=$VJegfScwwVU42}DY{>S5l>;I{wqZ#n^BL)JKZ<cNeQ<*j2K2D+`tR^
z?F|@YUO`pF#{qRWZrQT#H)7~C{GZ|L*R7*-4Q9P`CObGZ_ockZStyXI504~19)eU-
z*s#B``XPhzyTLw5AQle-Jz8Oja22(icFh<os>oaQ%S13p0wW&l6L?+%IP)mGocwzK
zvMu{Sx3-@<rBaCJed&Zh!$u7PYc(1>?orA8A|YPc)%(3oKA%|pNl;sKa?P6d1^HTR
zEC+rgVFN>7gw_X_>P78u-RNt$1iM%Ox6W!KV;Pmb&7zh;fyQg=zXrr(;1(AW^Gu8g
zQWW-uB`9)&w&$-dx@&2Cp<!W{{grg19svcGthg#q?+W9-VcX1^8e}TH+R=D?pTNcu
zwq*ryRTx;$Pn(qvJ&wzbr~Re8bCB|GJu`ZtiyEkVVW;|HzLVj(tmTd5eM$^M`_Kuy
zN-iOUMteJ`L`zBCbNXc;7<2#~=jva0;8R_Gb*rkUJ(J{6>D|Z|Gs!#E`fYyu!#Ioe
zGE**u6uLm64VhYi{4Z)If&4IdRqVih20ZEH^xjmLAg;iP*AhR35AV?ksP^eTtpwev
z7b<I=DruVS5k7U<zp6;$DZrd$M+p>@o%#CptDj?<=^8gDzu>(g=6ywBq!m0Te{CVE
zQnSGPTjrBM1@HYnzbFf;^AW>Ip-$YA-~|&2RgSS=@`50?M?IgqvXn~V+G++hG6%Fh
z_IXykBcL$aq+6bM3UwT$sjZRb=B?sYWe84^eL5kYj7*oB9RA)lRnArrpdPS4@z^c2
zQNwRDTWZc{6Fr`y3AkQ<_KN?$7kH-Lucjp=ui;9^OVfOBIhyj0<C^6@8qG0-Al>%`
z`(tDioC!qPBv+63a2jpB(WfWc<M{y?KG}~8jkQb>xe4zF(jM8q-d|4p1WI1qrq?Ku
zhz?j_L_#sYU%KB#mJSJtz32zHq?4fB7sfHOS%T9&nOl7F3$ETi6BTnDsJ7Runp1c|
zs@t9|>f7Y`=HSZ&o%{6L4~}}PNi!VY`aW*{m_*bgd7FJrvlu_{3WB8S4m=Slv~hSs
zt<_4+-Ez94`_Ax=#Nl&1&Q-Pc2WjKxo35tvWWU@yxC;WPg%|oX@j=vWf~Jctt{-YS
zL&?dlbr$zI8AJk%H!!^-Y&g|Kt0X{Hxcn#i(_guyoE0$Aj6yINLe2#>Hk?Eoc$i0$
zT4}wzeP5#obs4WM%b*~$RLFS+Q+9Y#ruR!Gk%%gAj93>wJ>i)9ux8NW3lh-eY|Vq2
zf7)OBD3e6TYrq=U(^~5E-x0{D?qeq%sGqX-9~KdS6R^lRw|rTem^{kQAoR*_E#uNY
zTHK_xG&0e(3toCcSkM_NL^bqQZ=VfAq(P?E$!hRALKb%P=+UrLdMAU3@f1h(fc?3b
z=R>s5v`4_<`NCuFTaxdyAKk;daed!j`dPtV($0vP!{^w6g79?Lqy;NRjl`=R+h=Q-
z@dfC!*s?a-aa?#5gM1#d|I%>@NXLJYCD6hi62IsYu$Aw-ni%JL_}+=oU$z86Y>Rq3
z7<sDW!Fwq_i{QbVMABFGZ6N@(Tv?6INYcVRSS+u73M$Uh*JoqAK|L(r^~=ZKM;hU1
z$~wldJ&<mv&}!{oUu*EVh3Q)E?6ZLbyrp|Y)*0TT+~?`xF?WvMDa~(O+H+rY;u(f4
zm7=-&W<^ViZ@RBh?i|11Zb=u!2MkaT<hF0fZM-8oI~8@0QquhmAUR!7$jq-lN}*Yt
zmO5aM)ARhd;8qusZ(RJ(LwIpFFs*w6mn6-BhL*_PqAv&o{HZD=3d)6RhYDZOxhune
ziIsRhFm*vf2b#omRp3nAE;iCDp8jT5H`;^8^Y^(Q6pT=SYl$thiapPMHZv8c$NRS}
zAc(*g=I&RJ;>C%^B;;>uOs&3(lUC<z-0e|;;PpI+e<O#(H8$6C^iz~mtFANscXc0;
zkh+QB9d?psICzK|9Z+3MY^it<!BBkozkNLOLd+Zvr7Tz}vQ4d~<@)Gm`di}z2LKS1
z4k(6*AVy(lE=sraYbdvHj?xD71iSr`oOse1PU}LK8k6=D%E0hzB$H2g?~m73KsXaj
zE2exE;R+SFo9`@bMdZ7ykZ!R{^pb&3$?j}9h8Wh_wc%8)H*CkHz9@J4lJ1S&KS3^v
z`T2p;r<&CEcA|)Lu;%BswEpc2MIX7&iAnC28e_;|r6D$>5&Ffg7$}Pr$$i0n#u*ut
z0q8+6UOk2vcWtMyo+s1Lq)49dE4@R{;AmP@y1bK~FVG0>y|9j>w%ZiOY%`V}2xnEI
z@Fn<d;D!VyO|46j4HYj$&v5SPJMtTR{j}OyAqU~sx1&GWl(Ws+e?zEA0!2-ZTQ~JV
z&h2(bEdp%x<~sHP*>pW&f&tx#`r2{rqF5786U!5OuLqPZQw8$<jL7!~o`{leRS*kB
z`jpz@9SNPlG9klQC%M5rBMLg)o^g(@?UO3$7F#POk9&Q#flWIUM<m16uisOY_0Ys)
zK*kt2ZAhH@p)N5kgQVGqh?@vUM#rIKYwE|%$Jd9_x<4ub&LfN9Q){!s=O}`e+oP5A
zb%!Xzo~aC9Q^I|v5?(}iMA0a%ipE_@?t7Z#h2GiHeCKwVPW71t=jDR1;+z4YTT}15
z&^p!U)0Irq9VTZiCTrPb?Y%01m;$}JNd|)7_hhGufE#?#q_~w!A7>)X$9t>!&BHyR
z0@5t4(PieN@0|X_ymV6E5*!HihCsmPTj2CDrFV%>z8|k#I@0w;yrc2ZiTN+nqNix8
zdh|1C+-1|wou6bms(E3i^yvCh#kvEyAc*0_GJMRZwBQLpt}5rzb?arN{4E)SQX;`+
zt>b#`?h}rOuh545gkz#>anNf_YNFwNOv3yesWhLR#qm?^{G@U?97G4Z9FjYLa+Bbx
zc6D3oAF`L6gH3!000%>J5_Z{tZ(Q@itNvX0an|5cUH9QlKJB?#&gVQvzFOMs`Ba-9
z;(GHZ94n)P*BUc*O^7`-K4xZeh(#A}y{=4^*@hp=gS3`4{)#ib-+5GJt?d(Bb14Rz
zz|66xy;3bH6BN|0yR9gVck<~-s`7#u3Hb=8rM|~WJZ$K5*x>=;cGWOiJ@Y$5ZZXn_
z#V$K_#TGM<vIIt1w;Ej7xOCB(X`qC$X2FbiHR8$o74ydx7-g1o0{%8$=nyIR;5#NK
ziP~6Pj>>lzlV?&IV%gL8G7Vi|j-|Rku9#@3tUmx=d+n@*FR}^e(#x%%S3~fb)@S)g
zeF(GyYiu6HEr|-+d(AC~chHSW1PUH^%3}${zjVrWs_i+;`50D+o@fM!e~*$rt0qAr
zgkao_<3fF?4#wy8^<Hm(rO9%~5B8cG5ISk8%o)A;h?m6*Gooy>+$1b~qF+q*@GDbA
z_MHO@U%oGiNE9ZvGxl_ET(vhNc%Xad$7YOP1XJC~HO(S^e@w>#K?@n(R~6KFh_ehR
z2tOjIFBtmbXZvXm^bL%3V*c}(6rAyKfW8)7p0$tTlxf#|?h|!R{;Qd<*M*+Q9@^rS
zBP|sQULWaatmLUOeVu=a_2+Z?>NPjlA1XwXH{5KgF&=eY?Y{f>d3L~^N}S~Iv9TL=
zsd0i*r#ueqE`&_RS-z;4>>E9YH+WNbz&EO$O=XmO_-b&aZ{S<Wh6(}m5AsimXT<bB
zTzAa-Tr(YJ*@NnS`bHEVGMlJeP=+dh2&}n3@f+TGvNh&Tp-%loF*@SYJK7Jy*7985
zvE><K(w5{KN^s(e!28u^11lQ7+2|DEEaSjd@k=wVvg;)ZQO^{zrAtlw$P^32hR0#^
z&YM$3v%W8C6!ehbF(l$0Y`qNePE6_?yq`txLh&X8(Z%TL5U-IH|6OeBZH7LlkF}IP
ztJC`nU){Db7ZF9kU?yCSF1}T!qQPM*Y}v+PxK}nVGQqxi`bncy7&SQNbqYGWeQ7Mw
ztBo>Qn&(p*AaI0u4p2e8Xs$_O&v;Cwlx=(iRb9%&LRL-sXfMo-W%+8D5HS?q+uY)4
zc#%E;JQ>TB@qIm!C{tQ)d~Qsr*lGIB8+``ZWOz=sZElz{TT?e#$+@X1#rKI2!jAY3
z5chLe1Y+s^Y@mAjnSxL)j+L@F`(MeKoGwc2Mh-fwSj8SHV4;?7I|4LiKH62NN5G)W
zbIw=2?~KzM-zpo+8a5K4c33ClhWu164yxC#$?@<eQ@bsuU|FUDY`IIl&mdl$a0~z?
zJO139-&~fNa}UgR=96qB-3{I<ypidZ7B=k<rgG|ZhV?}$uO2{&5OncOmyiruloWpM
zmJJ#ekqC6q5T*0w<A~3ZEa>jC7$2|iahu2`a^v7AABE5>BA~vg*d*%-dSe;FF$pV!
zvuQOh`7#fF;c5G5a889}$HdZQSvG%+PrnT%^+l0rkNMp7NZ8h!8eF&MN^C|9@^}o2
zx|2S#vXvG>W8;VI7_;f2!Wv2y3hc|zk$`<~vdeV~q+Xp06?Y<H-IeJBuq|2iptn`S
z(0E5k`ez()yya=|;i<Xxo~!P?U)VFq7NXO?cM>te<<ImviJ~?%(>kj%+bP60=dW_b
z)!qLHUWCL4Mj@CuI+X;edl&Zp=+W0s7dqV;vi=Kr@GAAA*6vbBQuz8Sl*-(KVXjVL
z$;L=g05iUr<YD+6^UPOKjWZ6%-5T_1Loc58mC7QANsu_)?OSkr)joHp?nEIgOZ*Qg
z4iZrU<ihwzM1}$?dDM0~7o9dtrOY4S58H`N#6!>#RvdOXge{;L%c<6U7kH^p6b^kK
z4bR}-T4o8OW+=AE8^0$6I#bIJBlr%rHt^`>w$_|16cp`Ff}94qC|Kj>i*~>Q0wz$i
zRTQ%8RT*4#m3U2<Lqxcjk;PM~mJSrY$_(~aexF-<bM-O<<nr)L^JYeuA`Gdott3kG
zF|g)WppegSa6NA3Mwle6y0kx<e&){jSqlIyz+tX~h4g$LAb@c@i6XlAay?2~ZCF0=
z3hztXyghu8=L_`sd6VFF{ci7&qeaS|w1AxGW2nQN9die(R}yO`6asn(K~I8Q8e!PL
zHE29OK3%%fW<guSW<_|t7O=<92563i(5=<ez(qlcnKcpRi;I`^o0E2(1U{){t9l@Z
zTv}{uVR57iF5?f=qz7Qm{*va5X)ybG&|q|c>AGc0zcv7p1t8N4m8*&-IdCG2S}epp
zx<I*r1wwuFe@f(oNl?5IOv%YixA3(nTwB`LDfAOZnN7L5em1QO&#Ald)Kr`Wyek_m
zOeLU2rapYJQkKGf#{8>W*L~zcoSuND?^E7PZm#QduqjODhrA8-Fou~~fREYI(w!Ss
zIEMEzlaL|OV=ogYzJ*#iU#sM<&lc_BHN!%36jU1pE<2BJ4u4P7w+o3|qPS<0bL{}?
z75=)+oC!scmgCR-H6gxi#ykjuEnWg_*kJUzqa@<R$mY>TpRBJV5`5=>HJy=)a3=<i
zdql%&T1BQ>ir4%>7f|v{Mv}gQ<JOOF(Pu}q&POt{Fl$lUT|WSB_I%5|lv0&d$Z=&|
z!`b4*XTB~-uo$3@V!uz&H4y#hU;=OYCD(5?ZirC}xxENcl3^yVayVE8u960U$4T{d
zHGmQGLqE*Qb(3otxp76S;RT*3JwTKnk&XG%ZIOvi=UXj0G<~v19@E%mfTmZ@*ZETd
zgW?Sfq`tzJ5NkTPq{XKp-_F_dS$|g<0ekmVn&-Z=e=Y__)*W?pMChIGd5Y+BD{7oH
zBwN=n0;#qqw=@f}u~p-=AM)RF`AT>fs&yGbU6q!=L>?Fm)eDzZ7BTp(<qHzDGQ2no
zJ0U)RFxMBh2e;5__jH%6=YKT&1e$Zihf!<8-8{F;PE1EJHL3+{(Yqr#k0oBe4xi)n
z3tYeCXxUud8Kt`TDuhUXzT^-g<e5o;d{KP=bQ-`B$+fq`;$`+qAJ%pA3Gi3UiHwLX
zN7K+7nQ7(K7Sh5xe02BG{Dh`27i!|pW@Oda7DzjPhYaVsM4xDA8qvqDb^PATKS&fI
z3waT$2!3K|1Z4_4Cf(**Tf>4*B5CTvEbo=IJ!<I?ESs3y-C)Y7Zqe+WWd<ztGe;$=
zM<sV%;%oeoV9h`iVScoO{yw@6XM)PWZS-dP^T`ZKLVGts6)1+DB?l*{x2ER?CqQCq
zJW)M@CB}2p;vHEBG%|Lx9R?Znsf$Eh^Ia9=g`J(bXVOkRCPwPY!ItWstXw`<m9RIC
zFch1}@mQY&r$(6FTj?Sp7^da5o~Xuv6C^;S9-K`U4@y@@G%)$p<ERy?P<J)?l?*Dq
zUB@LJj!c*4f%9tRfQEHPT<~k?=Q8M_obL}f6lzU9;!x8b{Y&b&VinCmP(xwG0hCWD
zHEn7f4-4Qypj$ukgKhyqL4k1qZaApbcj$TIGY-O!wAIaJT&(kT)ep95R(7*bBD6Nk
zbeaNGyk-+a1{@yfT}SxBA`C2CRZ5@eoXz5r;5&;k^<b4PxOB?UW+VQEFC08!!cbmO
zrUvBrrL6A+9zu?M+u<9!{Y(^^R1Fe@Ofx2CqGRY&XUeKB?Bp|lF2@=RJke8Nw6bk8
zKY}yNH8Lm#G2?d9-A9hWCrBi+s<hoV3g|U_CZx0=oD`A{GAjMYS%70HCw^o8cFnWK
zjm}8^+~OUg9bdwng!T?dt8ubyFZK>K<1iuxFg8>Gp+NJT;*iOm^R>f=9x@k~4x;b?
zi<2lUM;$A<Mp?!S-;Sj`DTz|Y^8$;o1J_tS*M<iQW%1;1jNYy(-d%lF*`rc*a$>i3
z=17I7UvT3U*JT7x=Ii2O!O$fdRWMt1n=xbdqhZg90@wDqdI@T>l$l4OV3>!nh-F2;
zYN4JFlMRgOkAga5C8!DnKeLVl7(~z<%gM=kYhU^;I&j=MV3BQZ(R8;G=}3`b$XF_=
z7TpW!V-aWVZJ>|*2Q6lG2Y#g58t{FICtJDF-bQm8MIs)7whJ3xU|S!3H$t=Q(oxk2
z98y0c$hEacMAd;s0^mZ>6c0Ar3fJoe$hNY6lHH!qW&yHpkZth8dfKPnQh5%J)2bL{
zSu@yJv=CF>qS9Mrlo)78z3L3!@(SszH^0iUM24$yPC$^3-j5XcDi+?m#(-(Jw!7vn
z52wTJiWEvRGYa4F0r7)0{)lj&k#*^L1!cFP^t3IaO>)r6%%?!b;R)D$XyoQ?KXSpu
zPcbS;`zKKECW4>hS4w-=zZ^-BUppdq5>){;A}78P9{~qqKAV)ZX(5(Zaw{Q=agyZA
zF}>}86%ht=%Zkv+$+0c34{Y~U1|Ir2Dy>~Z%x7-q+1H7afj*~diZvnsStg2(m&&5t
z<bRly_wqO|$uCcWjDf?x+=#fu*Ci+*mO0Nbsqg^LbQiJQ;LTZEuWudN)9F@Ch*4f5
z&7y^`2%Z_`Lt97EyM`UX)`ReVR@mfcx(aRcnlUm7JWc}aF2%QprvXxrL3qcMeA7AG
z$TL{<<jQ}El7x`#f|B}M`}TV6dgW|EFh)T2ut-GY018IXW%<zXn7g@2@tRMoiMt4C
zDV03ZA_Tas3IbD}M$P)gjYl68_3Ax!@ffu4M?YfR^kf0h)tkg`Qd2p`JMEjoIM+hr
zKE?VyJ+jRR2I&}`hUarL4N!fawB<P;AQXJthdh|$s`x1z@cGy+d`u%OfO#fl(0cU~
ziICvpiM8UOvIjQ69shusG1cw|m_f=a*QmP0lG|4pAS-=W)Ofq0A}5VPsm#{a8(3RA
za_F3EZzlxL2wWZy8vv$CIyy-4RCgQ$RJ@gBo$;tKWYIV`{==eVq&9%*mNoF}0Jxsi
zbdnqN*$~DMc76*nIvz0QU?$!q^F#o1hHq3ew&ts9TW>}zRE69((21{?fk9x|PFS74
zI)C=L8g%h=Ik_SPkHck~;SW{Uc66Pmpfz5Rv&-PI76dYR_BasRhu`{vVjyD<z?u!E
z`OB=`qVnqi>=9;)DK`L7VGss5n;NRar)r=H#uB}tSfH;*#B@K2*3%f|Pw>8KH<a$U
zbJ^q`7kH(g73z!T;^fl{UwU3z9UcRg<RGvl6vUzNz`LS8;GOIhuuXs28OpHc1&YGf
zTswFI0Osm_u{M-1Z4R(JY&x<SFc(?|Zl3qy!V@#s#eqyq4?S93KyHvDro+a`+gg3D
z*iRG`rBwlAaQ{HMXa9dtUm^C)*-`Jn>xxAsI<F&Y2b~jQ?XNIkf)r*@eyZO3q!$Ve
z1!mqY%p?Jktg`N>71@b_FfXpKb?<;hUsE~EbOS!IWov`9Oz-8f=r3m5RV<_keEE^0
z$gu~|)E~MNzv*JLyXstezA5#!ZIOz*hRzn9vR>0ud22E5!a@<r0)&2=IpbbSNGZ{w
z$U6;Z7@pt9jLpT|g{j%sro6!`SrRI7LDPEWu$?%4A9)+liE;5R3usYGkYXU<-I*g*
z6IlUvvhqATsFt$`-8s>;zEueDK>`($KsDs!6v65h4kHg~_a_)6B61Y2>C_z;Ju&NK
zzIc`6@InwLK0n|?7i7^l;FZ3>=97ZRCrFT8WzjLPgs$;w6058lM*ZUO^WqBi3({gs
zttyij7kL7d{jLBpaX=lV610CYVx@C#URLUo(+ry>2D1-%8x_2+!&Gz9KD83pjq?lS
zfPmizHShBBazGbAHFfm|V`5YMq6Q<9t81QdbhtoBfGsSfo5{=9=}fQThQKGbD&~FE
zP)d&0;;;|isQt0zcFo{|X7UX?jSfhko)YKP>qSSLwVn`udRG`96mcP>WbDl>z(j=H
zaR9S2`ch$&HuyZCiZF{X1ydG5oMlIacDqXG7Nn(^IWA==_k*sn_%n1u^p%AuQ>NvB
z2-k0Mu4D?&=%9`A!86V6K4_%5r-bp^IrFhX39|>7O*YPZ&;=-lCnBP2)tGI6m_Udv
zmurr%)bk|4IgBq?ezEouiryZ<fNU;jp$}1+rykWD5#PKgp<jWq`ekb9=zH}pcKxkH
zd2DqY6(ENy2bFB4lZJK>3%yt^H3yo=tO%|cCg4gGQrTKx!ZdZn^Y07k5u}X<>tvOr
z&c_Gc0e4TrHyLIsWVKIrUx*#KfLb>V_1SJjBh+A1CCuaWb&!pkoiEj+;V>hvfLf(R
zR+-SmGBL+a6{_>Z(4yl7B~dmSBh2;#==vI@P)#9OGvT(WY%OQvsRRKFm!Z&Y*b{B)
z5q5>mDEE7M1p$U@B$N)#U0aSHTfcC5T3(L@Qu`}~1_RjyiZ`NHg~;L2JfqSQYvk4z
z+if^7L``5HQEGQjOzidvd7#6%9G1OBYuAH{-_<7>97-Q*#DA!L{{&_~{hTvw7lcec
zIu1lXb0z_pJER+s+o9ILX2?CL;z4@Xpk-L8Nac-wgz1%dKD#fG!uTGCGAGqsYgSV4
z0W{Pr$!WPIn=_%JoWPYoN*K_iFo%R=`i{);XuAt0j(S^%I4@6LtPVbp=M2`c0U?#>
z2)p?I*!t?QD7Wry89@h7Iuwr5oq~YU(%p#CA|a)Klz_lkAW9A0iZn<!0wV?uN=U;<
zibzN|-+pFL;r)I8oXhJl^URL5@3q#w?zJC?3#2f4HQlE9>y5*LS`6_!nW?7K)8?2)
zM?@uj*N|q@)W0niQtr~ke3=B!X&j`^=(x}(5P<43TLqPH)nGSW4*mK|lG(ETA1%Nm
zO}hJ|IK-cs`7V)4gEAo~AOyBT<v}3m53x^lcj(wHVHHvy<kA>u#&Ez|a3Hj$D4(X_
zHthRWR74_H$O_!Hf%!a3=;Ec~`d#juPUVvUjTQLKn7k#3J}hFcL8N`Z%zC<iUcNt+
z2OwoC_!&ko{*Mp>s&HVcdEA?MlXP5GMC+J{fxQ|<XKaYPRO75}-)JH~mUI?)oMjlT
z%{>Xu$U2ARZkG9rTPR%YLWlWFF%85icK)}K3xCQ8QaK835bTl`4M22SU;>D&M)oBe
zByZ<y9{k>sA$notyNo%BzmO;{<d{RTyoqCy$3oWyV~~4illAu$XN!wjud+>vDt0;l
zg&-fo%2q19bpziCz~|TUjJ+rU@_#u<h*#Pf;0mo5=p>}gyukFJq>Y9}ieBJ79K+!9
zQy`>bUGHnSf2ef$lmyxWe7>$~QB<8qJRuxDh~U|=YPl>sZ5HN`@}w^v(@-Ey3iP+8
zT{p>XLTYLQhl+u!ClacMJSX;_hkVr)0?;*f@VG{^;=%`xx_g`<HcAkHq@zL@B3tkN
zygOGZ6a|X}F^CD7=bdkYY}0MlvDv0N)-iX-86hwQ_`!&_o7*TZxVRI%XG+sDKMWFt
z-;lVnV1<zfI+Pq30(^caSd?SJ&V7IbsUXiue+%>7<Z#!LxRK$tx{ZSCs;Y2&@bM-0
zLz;6x=zybpfc3&7GVp%^I?7Xc>fb$B`B4yBn;qzlnTRx)2>zD=i1kxS#e}?kbJ@g`
znKcfEWMUY3PT+GCYfLof>cNc~%dNdj&1z}@Rl`xA2aG~pi0e5^td(n`Fj)vV67=+f
z?X-J_3&W2MiGCsgg@A+_dgEqPxWGhTrNP9W>Bfzc^L(MGO5uFI_#gZ1^TMkjZZJia
zeTKPl##9yAH+Qu^UsSyY$dhPPMzs3W-W$av>x}bAePa7dcMoIN0@!}mqhej)*AEm|
zs<~ePbUD0c0S<9!%tq!eD?Q4H(<f8nX*MrDx{Vx(h7@$t+PJ1}TW<w;_#(!NtKB|v
z$qz)S1Ka-wBQ{wYzER=ytW}E7cJW3GgE+KYp#7q}^;fQ9ipY8pb4(g7%ogOhQJ-})
z1;JmnWU&29`wn9q0=XiyACR1?=K=sf2+RRroUv~i!O(*<(9#Qaom_d!#sy490m?vA
z-?H7zy3ed0MG?u0UvubwuHbm^T+qJxU^;x}*UD$~!lWHIB{=|-vy#tLL1;#S6~S1P
zYOHFpwXQmw&c{5G&13|@)phfz8qt<VP6MD5S(9kpIv?b3s|4as-*vrDnXcnCcSgHv
zd9h%nW+eA;{8-QAeUH80I6}L4F%5ha2ZVLvr!lq>PI}I>EamBSo8oN0U0hVrt~aNE
zqAUsILH+BeyHq0Fo?LVP1jSr@gWnv$!F*1F;T)|SmbmWPAukJ%rn5t<Pao!v@U1#n
z`D#Yzr#;B<BlKuI0XSUrlV=>>HiAq)t!^b*vZ>HSeaY#8P$xO^-F<VoKBz3$7`!!%
zdJ3h$=U$>FV#G~<U-#Q`jcX6P{D6PM1v(GFlrNFp7#QmtbH#AY3USW?5*`3vlOhx&
zo|opvwae&BnVeoLzVIRYg9T}P82A=y2(*MOo5c>8R2kpxl~BF=;0!wyhp=P$2lKe|
z*6UZ5{lp@L4)uMASbPWPn;$;L-CleSUbX|at3U=DK9U5!K{=mwn`j_i4_!ulkMxu4
zGIjoS4LbM1Af>)uHDEIBYv5rt!<22Sp$*C$HU6#huL+Zr$PBN1u+(M7docC{4+F;}
zf#uTIGZz1QI9hr0YlyNahMxA339^C01Znzpn0|1kS~*G+X>Ts=Ci9C>8?4nhFaJ`l
zW_{!qpA7K=1p1d2o;HEtBb~bcviDp~an}sXL2;UMx~AMAX{^FqSghv$Ib5)@&Ul#_
zI5>bs-~xVCH3(q=zEC6ZZIVlO7H)kC7<Nmo(m=X;8W%@^!oLMqp~<oka>2E8_^Pmh
zubH>EH*dKk97HDJWS;0$9y+*85|eFpT8q+Cro$fgbZXBItITDB`Yt+Le3E`jU}80?
zAB8ZxRz!)Z3_YnD{!Qgorq-yO6XJOO5%8-J`}mf2lS0qD+zxUu+V~1YU@f1aba^!O
z@6smOx(;0m6cs-I_729z{6XKuHp{{ESHVUz<H8F!fVyp&5g-lUOKuFtg9$Ow(6nq2
zc#WnIzEdcsiLhgyr;XW#{4f%}`+m~i*zm%W^<2{WA@{Ky{oG;}<%SGkx_1yeHWF7Q
zZ(Er~&)x@`gCE6_m=H^vehiio5>yv%^WXm4+KMS}+i$Wb5W9d_YYAn_#rp9Z=*KqS
z=qM)NeU{PZD5R+xrJE3y183|z;jTv4$bzrU=U)c)gQ6m5{gkgmpg(GQgp%@${8-Ta
z;06ZKNzPTHbXmo$8rq78&pyOo0ArTHUm_ooX)eap{G}a$vVuk-CvWC4(jY-^z8KNB
zJAYwF1{eI82bGfq!%B_G`;ldIj6F5qLmB}HdA}UeQVo3ZzZ{e&2x{r(usnYlE6)9g
z?jTI#QPZ`lW+XWccFXw+ivR|hB8^{dzqAdw@hw+YH>%l=IpxN(7GygN@J)N9W0-Ai
z^MW?(eVaZvkpo{ip)=s<p(5s4YR9BJCv?s0-OI|9vW8C*u#<wcA=CRMil0No7E8Oi
zE4p7)fkg-cyOzV==78C?u*;ZkApMZ?=GOrcEVQ1V#&DGbSl+TdgsNO9wxo-ZtNv-h
z3vs8nR(0DnBRAn6_0ZYTnVe>{5w^l~=j49dp<^#H6=BZRILDIe+?uA^X2+(%`1zC`
zII}8nb;TVK!Wc_rp@<c*Qu)N?Z5a|dfJ^9pCX2PnV>haAFKWrSUsMP-WLe#4HIiTG
z{p6ju+Z=)K*SPqJOm^5;ihj@Uf{RhV{v%$y8Mhu}t)hcX<H7AgnNhlXU&sF2nENP_
zP7fmm>L$wHCmFFJ#4eu3m5H8*J1ztDgcKUC1Go_(k*;dH315Ld5;#Quy0%d$=IS<9
zV!8DUnE@sC(r8WC==l53UybJ4TtigZRL1*8#4lim4v#cN^F+k6wf$F-x2mn=FYSf5
z`P|v7GkN-OW>6ct1IMBm_k`rN*K|`SQ!)fDu-N+G-!2jYVuya-S<(op|KcSa@&g{f
z+_SD94PqM{8hSwN)1&c19G+&3-d9Z?6N{KXc$DUcyv#fGUk|o%jvfh{-rj#w5W3T+
z^o)rD#sQIXk7P4|9|GRe2iFc5<yz#%#ka<uk!>LaY52T*Nof)j-H+1XXAEB{7AvZK
zXN?iAI-LL`7Ix#gxt+#Lw`w;GupESkXn>9R&CS-aHy@`hGu~|ngcS4#dVUI=lJyY*
z37d=Cy|Uf`E@B%1#=n3-|LjPJ7#2U-<580twu>F?J_W->_*08zD5B_{_s9#}vm^3^
z4GQoxswk0Ghmx9PxTU5SI&`|Ys&V0_OY7|qhS5+j8G@0}wS837B{IOK()O7p`HOcF
zL)b0#2d^W3S5wa?HU+9pyMu)>766g^tsp#8`}7n$KsjAVb8UI!n|D?f(HXUXXBRIz
zP`xu)J~>6h)qVG?nnn&YLFMkH2npUTau<n<hD$eC&m{jcEOxz5wgW|1T`WtePe#ON
zT|^a^JC|msT$3amG1Ee_gP}GIH;x_zZ%9Ubba$1Z4+1fIswbx<hNX}ORCDiHI`5d^
zrk8~G;~uu})(Xt?4Io&9-E6tXjG`2Om2yvcsaS_9@6*O4yACihaaTM5I$MOnuisFE
z7tlhV74e*2nkoNQLwqK6l~j3GR8*Ty4WraTL=wI`r^H4lbbt6EhsP^Pm^>+(txNuv
zH90U$aB5R|v_`Zl{7CxyjbAw<9P+sS<ElX1>P4~To!Z|{z9T|!Vv6T}Cb6`4(nxlo
zTfkf6s^4JiBOA=62c{bqmmtTM3Zw1tv!KJ9G7Fncao^nugX(^m_(2bGF@grNZtiTn
z*{8rYfRt;Bx#<x2KjKMxTJK%6x=KsS`Ja7$(E^Rl;H6TUO`_cTsM&kW^Q6CNY(q9q
zD<`n{X%HlbHBhU2aYULDq{X;o#3(eG8IDwsM_p$63SF4e&+pxe-vuP$*`#N&6C+Uy
z^-p-{HvZtYUv%>4%kt=h+l!{CiM-a;(6LTt2?P&k183NC3X)P`auO0PExAwRsTQyC
zA8}<hX$#m!P-&+!wx_k15>|6@S8ppKrTb0>H5BJ%u_tp{8s@MUvP3+h4@w-=iEmpO
zEBwhw5zl&mvvzT3-p%<f@8XX4!p;h9!Hea~%4>Kxl=R0EpJNx~qanJCXL0{@x`R!W
zf>|O`qAbg$B1wIpvDqVPs6(JutLw#nF^a~>qV~!+b}K%cIjjBC%EV1;r<)h=mo%!^
zy*Hs%D}S#>JLT1ub~;1N`_}@U&(vwnNQqOYA(X#F%FbQur>N<;?;jV;lNxWK^8$!b
zFgsDW2dj;9@Or;q|CtsX#fvypIuqLC7z+WbmsN>+y)xa--g_^3Dm^?Y<HlZ|c9f~P
zHDBZwH-}K`hzD}OHOfb&#X+U{-Tu9U0sRz#l<{H?m)_RP@-#cGb8D*yDcPnAy3DyT
z_CcTbcXT@?*8C1Oqz)E$#MDxkWRVy5TFr=6mornQbefH6Exp>NLNg=1@+Q+28>5vp
z_dHZ3cvl)fM=P^vIx1~7JwvKL@Ocp|Dd}dp{VXNlxXPAnz>^Y;QyOgxdrG;nVE=~6
zCb~`Xj!zU{%Af5%N6v3lWfgE*Lwu?9WW@>Kg{COY8T3sxHLz|U*gDpG!Ec~{=GNXF
z`@Xgq9@P={OC3j%<<I3IJsRkM?(()P*7Y4G7jsQ;S(!k`6qU~J8eN)0Ac4#pT2XP?
zSB0zjCbf458`bu;m1|C7qA%z#4zlKNf9l?AO)P2jN&^mC9LAm2GSS3wxD%M|0RO=u
z$TPhz-k|FnV%dmyE>!98&m@KZs`W9uJnaLY%WxXOttOzKgMl(Gm+^l(%2A%U1geW;
z+SzQH(XP+b3NChviT~BOi3gxQTwfLu;kmO*DFe3qDwspW%7o+L3SdYP0GHC>CccG{
zK1Ib&;gtxMg!4+=7FcWXmtMsaR`T*@O{lxNt$)WKyo%$7Qg26c^Z(2tf}%2eE@L~%
zro*?Ot;1<HI6^$64-4hs@iS!o4%Ul~cJ`G-AUTeWJwcfLyz2L#uKhp1bQbU1dY<>J
zlF858ch|KeOs{`ne|f0HK}^D5#;)#XIBdsHLRWT7k%JADRMo(u=HCs?DnvQ928_kJ
zByntcGxr*J{@3_d`Orzh9yide?s(Qf2CU8y8%oA*=bh&tS<zuQ11_BQn7o*{^)Zr7
z^3HMzOKob)#Q8skD7d&0%1t3dVvp|&@K>AQJO_^jvHh9kp;6_qJIxipHt0=vaKK-~
zQRd4UzoYQiV1}s7Im#TFk|FEmPTRb4IP*W2gv(U;vrAw1C*9cEL+jgP(V79L$zKAU
z1=dhd*Kvnk?d}_Gx1zz%mxG^o913JR9QdCl;N*6)&#yFkSSvZ55vt^6`?S#+llMYD
zp~Z{i(H(G8CF(`Q{c&!T3};7$>Yob;u!cbIgpgVMqAegM+f{`82@6#f|6?cS_Au`B
zO=Eg~A_y;}$N!3Du|EvnytMmkT+F7uUto>SD!ZFytiXru{vkmDepwJlq$DDKkS0Ix
zChQawNhL4<6knj{pQ&LMK-w7h`b@w~wa>5I*ZbOCtCg#%7@8US<B|c6hg5b|?Qi=)
zF!vHL_utuvw)^KWXT_oh^UHJ<-Qyo58r!INnS46s%zt=9=}DeRH=Dd|pP$Wb2$Utj
zZGcWvhe_C<iY#^!1A|d-tH$rTt%=G`(}g?q^;5k0+uUF1h|s?8?V;=aVWWJHU6USk
z<WcD~SJ}s7V(P#9NJ)L8#Nh|e8*7``*>lQnlypre{;VLq8iK8TH%Z+wSmfFHP(8Sw
zng3J0wf4B<m#lnLLEBJ|UA6xlot@eA%T>7`Q2$6b6FIeKgNkUi?X_)1a;Zlmq=BJe
zAvcFv0O9e+0KmTxIRtqf#GP(@NlYczmBsYju6+3Ocf=jILu6)1OX=e8k~<$+SYT1m
zLAq+Bad^%S7t?}6tBLrbR&l7ZD04I($KQ^uNCR_O6|ae52QU2xp;MP}Q(1r&8E!zO
z)guFV#R{|cT6A>cO8UO6%QCMML2~A=tTdHtDYGZgRIGyOp@%D1*nfy-{KByKM7n!_
z&nd<xiYT3`D6gNX?^O}i?Gj^2w=e=K{69D`g%_0F)Aas+XVRYZl;R~E#1g=yrygwB
zO3&Yc*l{`yExYlR>Wpv*B;D;A6*~LjZx0Q<1}uLlGq~-F0|f6cba|@(k0-=FhaTKv
z&to=bY`~c`pFyQ8NTwAR^mia@cxqDlev?ISrd`_~6mNfs3;h4XqjWpo!n@55TiOaW
ziEGw2N?wlbCrSTqn9?Kv*&+4N%kTQC^CF*AyULlX`9GydIeBbN=@HO4HkKnZfuE8^
znj)oP|Jvw;DEYh7)g5O;doSaDAk9hzM!{(zh`j>p@9{_nl1%l*?TVN@*JsV6yd3(i
z;%)zS{R^AGz^eQedT=DDbd)QtfE48qDuFa074O@0fSEXB66aQb3Mwh-&Y70K`-Q--
zwD-O5y`6)yAU%yJCM*o&6yv}IA%E8C1nd@kxIEQUD<@5mC5hhF>s+=HHtxe5?K23T
z>s<*6RB3)s&cK|hz{#8$N<Su4XVySx<R|<zd)IBcGPzxu*2hYOQSf150~Zjvmsz|Y
z>XzS9$DJTQk8lPiMMF(a{1<jH1Gz=`@}W_k_SX3YQIm|31`Sm}${gAX;w@YpUEFus
zhIT>zG+=%U!-6K>9_eBPTSdl%f%SBqa#R{s9gmqsep2e;{78&&YHsW9krzlzCO*6v
z>?j|^avdCPi-P3we=7j!7=4_@WxZS9wr{ITt~+jxlJs7j{I~x7@&0>#_cr<`eoTZ`
zX#0=^Mu1JcGQw)|XN=H30#Y-VxAtBD2d!eqW@~4_Ak#N1j=$cC*$~CoVM*J#zeHM3
z5*Q4=-*3z!f9(7Du==gt?IaPYo1%TjHZ7~EmFsDL?_cppb>r>Q2@ZTNuy_Qp_=eJ_
zDc0G4`vmNvh@9Q{oiFj2j-h3SE^~XJ$u)w1Yb>6EqT<_1h^M~1e~uM*oE{(pQH9dc
z&OVaBDCt|DYInAOG_ke+!c|@?;#B+-cLLO!C-r2Xa%;ao<Uu!@9OFraAOSG{)RXHl
zH9B-Oas(91pFXgdF%O85;Y-{8+d~_|%DZFc?n}1@o~bX<lUM;e)~=H%{%faTF93mf
zosa9Hw#oom1Rd22naICbLBd`D4q;sRYSENLWH(7xK3Kh<UV_Ijz1U1Z{oGwLj@eN1
zqCL{_1Q$>7Z{@wyA?$OC1vixw%sWQ}v`-ybcyNW@tyu~09A3BZw4xN!Pm8UL-VuP<
zIXs&=oM;xV3nhL-t@TE|*=J;5slj$XXR3w$_cw~eAfJOOCeD28`_pFFEhlZ;QKEC~
z?=wI^d>$5RWV`w*1J8RDaR%ytV4y6cWrC{me~+@92D*G|&h$=tkA<L^QP;~RO8UPp
z(qH-tEWc(hQ}bT({ptMB0P-MaxPy`Z)1LBRPYi4MXSgJ>g(!31JK@SR%EI>>0+ujL
z!68^^k~?jAB$Na%!tfC6;1~vVIfi4x^*>)`WYYR^K~%*$MmhdM+!x%xet`tk@v0<;
zdXo3`EMerAe}nLd?11#dpRZvSvxP&e(sS=(UApV@xKZ7X`&W-#LyNE>>>m;wcs3V)
zjh+B}In_zk@?S3%urBl=1S@9pB8}Lj*f=-m80r744PX`nOxAx+JA9~h=8y~Bgb-kH
zqzj`r_%nnYX00DQbw(MK$o0G~rO*Esm3}GGzsh$jvvEQ4S5rQdBK<XbOJIBqV?+Z-
zY(Sd+4x+a;z`7=rH)1#I)nl@M{dC}1FN}5%k6u}i+TN9+zX87d-t_GIV_&vJ$lUua
z6DLXPqEP(r%rPsX+D5<jd~-5!q3OPC2hH34_T;}s1@n$4pokt@VSVsZ6=`bPO@HR7
z#Vq1&KG5-Dd<0?c>Ke(v5rH|&ke+KJUc9@^rz9#IsW$oVq(Fuy)oj^aposVXKcVKh
z`E`SHrbWJ+!v|aKx-n_gxJRN`U?>jisCQi5-!^TXNZ{%KDes8LTs(SQ;0osaQa(;9
z`}cTjIb*f3Ei>2wpttZ;t}j1Mpt`(Iz>X_J1_s{c7zW9|{|@jGxW;~~n)+MRPp^?x
z2Ibv4jzA*@+&??*kE+Bo_Q<^~6mfI2>+FPYz-iK~!V~pHIsE+pQUwSqN&!^R`PS6<
z>38vdRna&a;p}UotYU<0Xf;0IBDA2;!-f3l`ao+w9PaHjJ)NB($=BB`bhhDd+yNH0
zg&ECJ`*v%1)h}}PO2{>H9k6SxpBh^MQLTp~wKLy+*(%dSNS{x;*)&z}=O-7xB0aBg
zOEKj^<@}q-&d7lfq<f_O={Rd8Nt5Di<mY`FZ+G$fi{YzjlbR~)PmB{R2ia{O4~r<R
z$Et;@c*!j*ruC#)r3dI02ShGQ<>+m^5|&tfW#Te2{rk+sA^IX&L8SCuBtykFU9E=Y
zt=PkVIgJI+68g?w<ki!OE5)a8k04#bNx+&vw`T<6Pd!}}b?1=$8@FJU>}xDSO637M
zgX;S|^{>3*()_|!Q|+udC-S6JtIfI=r?zX=HdGAxR@QWKvSuYMP4@h_C&WVzNv}Hy
zH<P&xDPh6h32U3v7b1^d5~_9sRUgwVTz9_#lucn)V8``Yl|n$*!X5n0)GfZgta_Q~
zL8EDJ)$9ReXLM<18ns-m`apNc!?ND$gDrRY1D@bLE01Q^%<>-^bMv&EJC}*DxbR8=
zmK$@nHY2haCt^Kc8~5{(h)0VOM<i)*IC|ej|Jfy6#0;48Cq+Q8S@56<jW)DRW_yvf
zDt8NzyHUt_t}UV>UehWUTILakSrO!cg34Qc+f{cXuilX;+UCPNInw;a=lMTc00Z;h
z^xO=!fkOM3G!139DT$=+f96X9*D?R_gQ|Aizhm{`wG1XbMqptq<MABNrFZe(*^lZu
z<`0>@88pk*Q()<A>RIjSygP0ngnKb!$1XjOe&bs3=BuD|Bgd(huXbl_4%IdjlsCF{
zhjaDraj}f~y<1%RcdG}eNRuor+7t&XetJX5lgbX9S5SolD0qeOIXGU=aSgM)?N_U?
ze9|0W4K-qA$fCKYlHJ88G+tk5WiWE_hRQew2AP%2-Rd?OVtBt3=Q022%Eky7FivMl
z4IwBvuqyT8ku(pON3Tp$98h?hz(%j$W8-j(Z6k^rdW{STyX;m&^!>XQu5k03Y@9D=
ziP7qp+ZFv*V*}?@uDD=+;-fBDw2gVK`dgD@YNC?KXKnu`B_JWJ#7jQM6a32q^=M!>
zamhL2w|N!#wJ25OeZaan!0I+@#ac5jFlb)j-TICs;vV|`A4Rf%C{Lzq$)ptW-fsAs
z&rZ>j3=(@keGaT*J-@Rjd({n;0_;w^9El*btPv@ft!Rlv;$pUpLuAnpphOf07-Kx4
z_}WTJa+Y6&idRJ2DErnmA!2#|LHx=Cx#((Tx9vKU9uM95NBo$}9_W9<a#ubH$1H0v
zUh_SITrIrl(P-Oy#@g%muaY2UK#uyU5g45vBYLLnvTe^8=Rm@gSBaXN5PriK6rRtU
zw!3=G`?_v#MEa4rzSN%P;~2+)OJC^K{N=^v_EW}Q_9vVJ|2lsFYJso>72)liy`lAY
zY#*I0gEB#w<2lOpmISVWnrYAN{<ZDVv?^RR%4+NG#DGtMhcHvEHnr}%=`j|p4Odfu
z#)|k!aE6!Gt2WMrrTcDgQrqPt6JTUTZ)*d7<XK=k>)j_{ZR9tX14E>L!0J3ld1>DI
z#k7sF42UtS6H5qgF|sV#D33*LmC?kxyydB}kWK3%r-E?&)i2oNt?wmjb;ddm2Z(2+
zk34^z9WBb#oxC4&<@y_4Pk@^ZHRNOrB(^w<SDcBP$Cy?;RJ~NwT1g~DVGFz^_LJM?
zI_o@h3$#Ype`7)gdIp$?+lb!5_)TxWo68YoI&twwf}==Pv~$dSU!r4Wh4E<@B3LI#
zZi70(5mGZmiA^uErjN3%Y$%Jmbe(oN0gIsC3~8%>-9=<pk<2G0yftwJ)89+4hZytm
z-Rsblyji{OSC>A=@Y?#{4ToSW7zit8r4oj^c6WQj_&8VxK$Jbnn2nGw!+SRo8Q1lB
zg2nqlSBg5w4E}@zjode}=6YLm&nlKw8DsAJ+6W`>KAWGvvaNb9l1n_^ID#9}=4z;e
zOA(H{oK$c<9pF}3+^*6h&JlRp%oQO!V_e^*>7C)y;i!WVt5&N_pZq4V^my}I+Iip@
z!_ENKXUSQhUZ(4%<>d^PzrOMWjFQsxN)*k&y@je0x~v}*M4F|*V&u%Erby4F`16T>
zKXEfhKjXSGAMKct_*3aG;3k-Nnz|Xrw$!FYyg$;OVEEQeF`Ss8_}%Rn#&vL-BfCk$
zo-1v}njl$}$L4YlrU~NJPmN#^%Q?{|bR*XiE{Z8Hur>d3<7WT~R*C(T@6)O6f)Gc=
zAjrxukybKopxfGYg-uvy=PzZ;{o#J0FyIjx{{U@5@mZ-eoD{b#m;j~;8z__FwlJYz
z7!m0!(?uq4+j~9Gn4-XCWN=DRGmDC<vrx_7Hen_s$0%mPQ1PSQv?|xDuPyJ9=_TYG
zu@SwSsD=LGiIeK5liktLoW$0~FbJN}$m~E=Ij*sQTbq=;HC_1XASO5mb^eTdt@EwX
zb2Y5%CXgl>E+>sHr~LF-hsA`=b_V{D<!JC6PUW;UKJ9SZBTbb=QVB$X*GvgtlVpW}
zg}zmg8LeeaH*L10*O<r=PF^XS(JCadR`AVj>P@(+ra4$v`ctr6?||v5fj(yS;A=-)
zujxtf5WyS;L-+3PEPT+0tA73g2k~u(NjL?f7=N=NTBOM}UX8~}Eb=i)U<Il_CfiTD
zD?NJ7jCL`n6xzWC(i=&1+oi3Nlp2i4BXKW~KYKjPt&-RqwJc4yDgz6luNr?9j!Taw
zoJpZC>9+Ech44pO53i25gwGHH5gutMF`{?HI(*UGyV6}nuWeZV-Bfb|=1j?7;UQcm
z;=`A}_e1|7ScYB^8%CHu^-@T*pe3M#a&sgT5%7=tD7TW~dUMs&`9b#bn75*ok$|Fy
zy+N;fjp_29{iPJ(YW{)Wk_civ`ocls1|svqmGz|Cn@>?`0HM_5!%mXGK_HV@`bLLo
zrpYxrZKOd>nG@H7p-^F2m0M(~pa>&5m45XM;L_`>tTm=x7lmHFzjM@*H(&U-Qj6Ho
z*5F)w^&h~cme)<^;keA`7n-9t_a_^t)}570U_lBLK;y9lF6CsKJ70NaK_2RXfa}RX
z!ERx;>l;f!E@w%bBT<My>Its!-J-71)8)0h8{}RIF(jVxCkQztiQpgQpb_W(R2XCq
z-V@?sEz3ekQ{a75HOx0YaETBP!fS#qbJH_2FZ9?8w@zPFiPxSVXggZ|K+I4(ZEz(F
z(MtJ~!y-U^MCj(VQN>rEVWYjvWR9zN0<9|XIulo30{OuKd8MXK!PNH3E6jR|SD?%n
zztpt-qTRiJCgtcIwJ<8B0fxByw7wUeWK!j&0S4jnJ`aTA{ThwLboDwdPM*k@K&yTG
zk#r%<G%_>VP;+eNaj+6USW6I&J@s*}OXFfmx1&H59o;&9UK~1ymJDM4^J8bNjZMW%
zg@94`@Pcj*|IP%1%mH0dmc`AKdxYh>AIHh55zK_U*YguWA<76z`UC1%+|`GcIJ&vf
zcj00^+xOc?I1OnE+%`$1t$n39fp+t`j1%=+ge#^>Pf^$L-o07lj^jg?ipgQ9o1OH$
z(vQ`r>(2L{!mu{1T>P<ozTeVT*u04rk3<Om(n5H<#5765VkJxIv^iI!ZvKY2=9HQ=
z<(GIq+Nv+oRwljNUV{5H5tjI|`F&Ytma3=kYFu|7WQ4S7#&<zvpvtnNguH+m`k?`X
zBh<Dwkg>j+nHrxmTn1IMjQlc~mfpd9P3OXC){3Mgc3HE)!ODQ;7qqCRl%*StT<Cs=
zkznn!YBf<e*3~v<haEW%Ku3uuoUX&<-67$zelpxqdMO^H>pABo9+P_B#WCxz4%Lxd
zXFDqij04UrkK*llxg+UlB6Ut#<+c|XsSEDsIsrTN#igmcqAXXB+&>C}3<tcr*yaq)
zP)D&KeZXYueY$h?9Ach&DR-uK)<Cy9EwhdjxHtN+D>z&&wXI&;2RzQb9Kc+QD9z7P
z4&|m}Ua<tW4Lr&PPeI6Vkga5CY7VI?z49F3R(v4MBE1$4&f=Y|S81B9JvR^&R{?4>
zvX~#FgBL86ul`Jt;}pIHG8lvnyzu2X)6v~F#@n5XbC}qf7z{Cu0{fzW-atd%d>A(R
z)vK^gifQ(TbN$E1kVC}T=xkN!tQc-lL~6%I0}EioiPm<mR{wXsB&bO^AR=AA04|(T
z82-`#5G9`Jo<s-%!p#92&F+^{z8V?p0(9-p;LCgnsisx@xrKUh)3Q5&)N!T;xwV<y
za_Hw;w!I@RT&WP%9_O$fh`W)w?G8AIK@I+kiU3f+@KW<!!!2iB5*yEh;hn(gTg$2V
z<gPtt4jY8tNWsXBGEl$-TbOmj{0HSu5W8grYA*MNJNi7uUjtYo^ZMPBAMDTPFrt{P
zQ(?K7s}9Qmn|TIT@^;lep~aWS1Yoebh~C;yKsFs2`|ZuK3f*oksuPgJuU{RZ%3J8s
zNh{P!-*~d~&onxNEE~u)wepGPW)C_o1;2pH5M;>NCj$(W_J4nI3IEe-g_$uY8_Gag
zzeD&gTS-{ZP;eL96|kieH#io<=gz)=3cV*o&IFyZIul1YU))+QR}0?)(&TL@a9%Ol
zc34gL0vyx7r7PF<FF+1z6sAdb$Z%$hXfEc!JsWokEx+kU-8uu}^E85CrrG~m*ZkPB
zPpksb?jYfr=@)c@vu1WFPjHW7jTRwJok|YX&#wv1GMru>3gdA2=6JNM{_;D^06EMp
zU-U$<>6AX%l474Sz5jS4h``-dufB-+B(@o-jBikVtrqP%0^wR3(Bl&0FG;)@yZm|8
zB3oY!XVGD+=U|sGPP2~}Cjw0BJq&0ZjDdAXzCQ)pbr75cU~Kn<W^0f?bFV~e9~~XZ
zs}p7{l<0I^s^?Gj%&BD(e-Lawc-v01=7)5Zb2S1*Dd4xzhISHr7m9w!;Nz0OubZnc
zN(4l4uSJ-jBs<P$aiWWj;!DmJ<o?upCa(exYFhq_!q)_zimg=wJj6nDK%@#TbFbHG
z=`3P7DWz%YtpX%{@_46KWbdnOX9I5~8u;NKQ2;)y17r2Rty}&*Ae=zGKL>&n8llG^
zW_Mh1@9WuCWv&$o+4}*C<IW^YPorsXPzA8Mf>X*55iRla`#l17ef(tM$EZdbQvTUz
zpy>NUYdMUG8=(HJk<+%Iee#9jO!FQg(*3!XLn<&}0ld2G>T1FH)Pg<9xHVhQp*Tf$
zqwH5*jGD6JO_rOtt2!qCJ7O#(fico5qId2J*K(S^y`u|NG^eyCUw*{*w6Ql`F8@fG
z2}n`7n`qy-lv^Vw$6mVN0~Q%ncD}j4O%fHbpVnl#pjAVm_K&;=v-pCiwr#B|yjS<-
zKC)mNoVA>L9UtL@n~$PdduPp<FHn7_kpp|nOc)z|!{gwAGAs1n1BZI3CgZy~*HOVS
z>MK6_m%{>!`2e<%JXlcmFm#NiKr}c-6DkQ2<9_BqU*Mrb&gv-Ze%>Mog+k%i1iP6E
ziW((WTzyYz!i>QQjt2f&tD{Y{K}G&pybU)%<u_T<Y7mLgHQO<uWPq8sy-f7yi}!mz
z+!18#1cUroC(jaF)x#dI?K&K&e7ocWd%FAOG|LWw%27k~N4dizRfMbi_8q-POD-%f
z-Dv~C)TiA?*&<hQ%!Z3oRJIM23sQ!a!8=%Ba9pmX^Q~)UBo3mWze-vi@00;WXgM;!
zz!v`OnCoQ58(w=ly?)uRNeWqE5ch>UjdP+U`x42L{M!}gdgJ_?m$GE>z(%6CaO<+q
zcWV^o7l@yMu8UnV0Nm=g`I+o$eWbhGF*=YW)*ls7K43OjTKT(^+%v|&0wOt{#i*n=
z6K<w{<E>huJQ<NWi3`NpA*e;>gRrSncb{umB7s?@<9^HAT=Zj)`sv4FyP?vKHa=!V
zVDm7nKkL4_?Uh3rP25Mg=g_<J@rF|Cf@C3@YPLZT7-&-SOZ9@_;S6Y+U^YNU?=HQ(
zpkFOtJUz+z<?pT6(B^{_#SiSz!E13FXO$i(f!;gSY$lAeD9*r?UdPFyTP}MQA}In>
z{cm<u?@-BK5{(15S1iMxPJObU^xBxz90wCQVn)&eE8V~a@O^+mA=4?6u*nyOArU(V
z5e}!NF`r8&+pM-0_bqmNXv1_7xR(DA4a=_yC%;7{0-@9qh(&P1^XyxMe_06>G;k~6
zLh1o@D}$jMu8(%BOqExTU@)X=k$!?}{8->2%4N@<DKb6_=uNIcy!D8HiQxOb2gj=^
zAk641id$c;U{EOsNu}kO-}sRC%&uSbuhoyTC!oMxJjQ?SR`0;`^0Y;7@X{M;AH2Ip
zb#+&#ZcB-%2#yaSt&duymMkIxFioUFiJd=AfK&v}312R&&fg>HifV$&L*(eqy;(Jg
z_+w;8`&O@=@wxP<`8fzWhDa*c#>W<sGy^*PvsUK+sC6h-iXwB`$_3R7AL?io97VJ-
zl@4hoGh?&?N=(Buox7KhLw7XdDm1IW8h?{JRdt%5YgzV4l#r+#RRi_7?ah2C=k9CD
zKh^(T0SZ2w^&Axg+-QKkK-}wVAfO=_r-5tp;4H@aMYes8)~?aBX9cX)34|y9W(>Co
zN=p3moQ`dTIih##HJ9-^0+(D1`&zI=_m{&~_&|4W+JyM3O^RZ!2vV5=058F;aFdQU
z&SE0qjG`znG<=8lh77kYzju{7(cC1)QBB4~l##>$QhQVh7+vi1T|KI#0R?rn7GbOI
zT}kG5x!Ed=R66>%td-v4y(?LpKo(@V5lutK_7d6VQ-xoifK(0AX3L=uJBJv$PC={^
z<@zFR`(xzs_g;Wks;Aco3_LT4ZL<eg6X{cf*vrl3)O912n2{9M)>dzFD<m}m)pn0o
zpJX@U;b{MSJvn+|j2frvK}*J%(x8NJIy**D4A};>J>jHIvp=RCwx2cOvfdeh2)|(D
zV;PF2j&Gf+zSe;QO^^`lqa3<ZpmwZ!r=BiCvImmhKlS--gbWFOs7-LhQTt2Bf?VBZ
zuxmG&xos%>4kwik<<!)YA)rrMDUq8DBW3S!t$`QN!Bug&U?Z`2`v92$f>;u$6t>A7
zVMgX82m0X-2L`S7E}qI-_O`@KVvW(IeSO<aSLgs<v|vN8O!wLrY~O2ITj-B^1Hv47
zl4Adz=5K=h>j5~MfaSp<7#^&V)U`=BQ;8=t7iT>7)JM3+)rD6R_AMWA>gtn$RPa<0
z*;0tEbxZPQ#-!pW>^1v{A8?;Gm%SJTHeO{?ptjFZT^zdA@+`azX`1}<^O4(Bft7{f
z8E-$F{`yQJuT_O%hS>;WG^6?=y6xD}gc^hT+2Qw^UGJQ|&5zMZp?a+)_PG4P*PL)q
zt$fM_L~NoQGv-FKL(Bn<{*^nI+UJA&SVtI?_iie&%$DCC?k!oNz5FRZywnd$e4z6Z
z;=jW<cfLuUjbE8GumJUQ^mS<EfYonDCBdWc6_|w*0`af}X}iWRd?@xmK~wOA1wE5e
znpytRN3koh5E%|#g{V~R@7LvWlA6^7u@CRjfs$UXHWfVOZu_{$14lu^AN39Qr=o(H
zaOUmKov+HzAlw<g?_ji@O8ko_*jF)s&%y|40c~F--f&Hx4Z2%9U)miMb<rlthPB;G
ze3M@4dSYRJ-`D}o+JYDDYnoB(tlVa#v&aWL|0cWKbrf~^!kb7xgD%N%o@-F`2T+gE
zb2BMd$z1w+K=U-e)GkLv2?^1RiqbKY?j&1!o#Ei68FBT>Z)&s|(Wiru+XGA41uvJs
zE*8BEnp%Ti>^08egqYfvBJZ>-m7y0$2ys12P+>WUO50xQ3O{y5AfOBUqk!0_c6TTC
zBP2K4UA*FcHf1|bn~7;`Vrcc#f*l7Gk5Tg{-5c3UdEu?T&7ftEC^NCo3~7tk4@<mf
zYdqI#E{Q#E*oAy4iwz<!Ha-?7a|T9DT8~QmwDD{AlzPz|bGx#k$2N4VE|LByHTKg~
ztJXS+pvSw;$G2Y1Mf-o=*4jzcUwW;41uEI2KZ07b-n&JS>+?SEm1}bDPI@WtG#(l>
zXiOf=%k-1)vg@Ws*unE=YiA0P>Au|0{W=OxT@gI$@m?9jc!0e1m2rTxMl86OP^itD
zp;r(W!>(o=gw1L}m;6Ct^3o*`lF+<hjg+P*K{Ryjc<7Bp)JI1WF_U3Cd;dHX;Y`q9
z7U%Hc(u7ZJBO_GaPo<7Y>zBM|nq=KKMn9{52=D;A*ly+@*yI0DmdDLu1W7kvjH?|~
zKRQHe>;u^gP+5-aCh=4v`&I86T@00g#mH2jEq2gHQI{Dr_0@dGoTtFo{&Th8x)0u-
z6MjPm)2BSwMm=WN<)`Py{#VO|c`U+(pt7T>H1i-mQPJi44e+prYT_<&S%9ngVXHXU
zw~ShFX(5La-exXl^b3PG7*kSJvPg;FR@_QLx)&_5ha4#hgO%ZA{gW5;pCHPw0!Jj?
z3<oD!6+&vhtvEOak3O2B8RF^(s!!j2@lcn5Q40sSdFO03LQbd4ibqlP1Q@!~0~0TZ
zdDOqqQurq;C|5iL{-G73PRtkN4AFOk`?bFfiBQZ^Jy8TgOh|W?xJ=eM%vmFNtTfHZ
zgkJsxRIv*!i5=Nxpm?o6M{B-cf{S%e{+P>A>78NqUtY#38unSylJ_Mxz}2L&THm!f
zKaQ<^a?d1#KO2l)#}jxy{1J|m8&_~f)vhuYg>eQj+C-|O#AYf~+^q(PCg~wy0!7DI
z6Z?^ejdEO<ZAr6&rC)peFplFXXm<s9^gr+ikppR#X0J%cRgVe_D~*{RLy*kpq)V0@
zXn;xr9ZL!AEzBlExYg`(l1`%<wnyqIiFm)ew7^=g+jS=G_}F6ceHbIu%_vGVdcr@-
z1chEBfsHyha9L&3sSL6NN<YDXZ&ndD2KjaccY+k;eNGR4pSoYm((jLZl=j2HW#vXy
z%qQ@hp!tf!2zDPKNjfp)Tsoa{3-fC}>}$C}nt=(DoD(%y&cU8TQh*EhigU?yV^fd9
zDWGub9L>ZF5})dcFAC|reOEFC?)`kJclai0J!RCGRTX0Ky6w5!Bydx7$nax95zT>B
zF=f89lH7k@7zRuZTG{6Wl_@cx_$BbQ?*X40z?EC0sunsA4g<o&*JuGwqyIw=t;8^I
z_0RQrpYTS~z;tN_sdqBVp^e4=r5|6ldi*8Fo$<|EfPQN<3RYq>>nYaCPY(RZ0vf3S
z!q4NL5;8pSGIIa)Y*5}@09HoL&dpP2=w_(9!_0i>&JIs5C!hJWw@{hTzQ+{XrPqc~
zcE{45SWmtr&(qkvwpfVe)%YcGcAiabhtQ6dfLe$+NR&TFUD~?!+d&cPlf_OGdLVic
z&)D!wqWZ=-86BAC@2b|O%fCCw1&||*14QBY>ROof7$mF&^r)ZW{NpfCb0Y+T&Wh^Y
zmY<r|M=J)9vi30SP3S=X(cVd>_*b+iyv|;eii*63RRJRAsrVm$tlr@pTiEc#LBI)z
zk#xyLZ+w0Y>Akz$AQTAEq}W+NX6#Vle~+Vg(F^@(WWTbOZ`mgn!j*S1{Ta(0(YeeQ
z=!PCqV2mp38R~oV+<~*usY@pjB)9~%pV9?*;0%Q4kIwf5!j2$zPF8(>wm|PbYY8BJ
zWO7`}6L-(xnDq>@Y!Lcm0|;{$m~V{r%ersX*Cj7k-S?zLEa5?-bc6J-WyqfmnyNC#
zRuPm5`z(uRHzN1+i1MC8C{@UIFN@DEiESe{qA&3?W;#)%OZHF<9d`XxjP6M?V8#a@
zCj;;(p0gjBEwBY4JmGIr+x;&vI?j06IQ{crW1^g=dNT0=`?EK>U1eNyILuc*F!$F}
z+mbiyR<7V93ULWq14j8OqHd)hyS5x`j3bVp2kX#glAWfN=@t}KjYX2DFrbroM5s$5
z_5Nr9RJ|nlv15reK|M@fJG1LO&^aZI3nO8ld?AbM!jb!*H{NK8;cT=D&Ii+Fo19>c
zQ3XD;iutNdF0Ag(K8uSO;h~U*lX<ByZ-IVpf6#<dH03sF)^l)7R-K!#_gG$$Jxbb~
zK)k};B?g?R6fWg+esUnM5MmHEr3v--DD_dw;0v9(nS8iCWU#Rc;{)vo&lCev_#c&W
zQ>h!$^gliRhJVHb{i}@ZT<>sO!|Hq)P3Q13h}<|snKFS>qAtCht&z*Dj)SEF5kZGW
z{%M9E@Lq6=Xlg#By)&i4rDH4*N#GOzn!$vj?k%M!ER**EmX3eWCWUyk=V>1}U9JRR
zvA0kN$2XZG<;6MhoYAt}pBW*0zp`czDT9B<?-e(A1?87Wbvl#i4_DHxfo0n%(2czx
z9f*&4$5VIK7W0k}k}MRR{^IHWnN!bBwkvYumU9$VS8TF~Bo()xe(>*E(7(XVjnfx?
ziY%6PUfprfxIXV$X2n2Ru#@+I)~rWVbHSib9FBuI4oCp1Z`aQJ7tyv{8d427nr(yF
zC|@QH$~E4}<1V;`2i^6xePH$Mlw{0G4s>3;z7?fRVtKS7=6SN6VhW6i1%A`KXllDg
zmPkn$w;b^N#D0p)9<Ptf1_4ivBRG{X*frlT2Uu!7*G1mS$<cx3v&xHSdCwuRLt}TU
zoZ9wF{5b9!k^xmeiMVds{jzadKBBOLpbK;X<PQnX&sL%TJSuB)^s7t0TI*v1`EwZD
zVkR86uG&9|JR`kv2cKeDwnEAMPN_}=rnLWe;j;3j_nLmRPP_hk;OCYp@YWdYLJLbA
z`u^^_Z`Q%v{S`ka4rQqf_<4oKenv`&rbi%l2T4ao{3_j#<JJau(0#=cp-$|u!f_D}
zRldXe$lhv!DatA<%5e$VV(gua0qvUy@_sQ=6OQvY5i)g~y2=4^|ND+t1Sfv%Y*$A4
zYGcmjkV{rjY+?N+?ZQkFPYQB%Hf=2fBc}xS+6?z)U$)AJbPIB1<bt<gPe)(Bv2?%P
zV5j@{3QP>-$G-AE5RPyhzW0c4#V{DaK<_3jReY4X;=df^-RCZ+vHDbIY1FR3+{G(A
zIO^HwuHn>)az2Vtusn))S=lF~6mv3mBWOBg>~wsZeFwy54)g@-YSoIRPL)FDlgeQ3
zRqwXpCqZ*dIgSGHesoPSKB_JyvTpUL4ieIzp*q7k$t}x5Z!zdQ{*G;13E84<%Hv{n
z%fggONIY=*!c^AoZ+?5|T>s<F8rS3tsj$^&chXr8oAeK%Q_Bd)H$=~Ik4-S?X^-V9
z*>=81=|%396t+^yGR%sDtZ7|-ZdObp=Xdi3K0q~W%UY(Ve9onG{2%WCU%H=`^V<0N
zdBFF*rdbhusNG6DxobNbYbLk`J_=rJ8mZFQ-<6vmY;)C6UinoRU7hywo7Bj-yyGa&
z4g+lgsx!%AI|#rR-}V7U^?Bx_t{>@A#}*;85y4?IM>qxlPMO<A`*TZonMMWwf{>tg
zeVt=Lx*apxx>2Y?$mfm1#Cfr6z!#7|Q60ChD|Dkv-xWdrME0k2V1BxY^hdez!FYe^
zYo$qGoU(oKzmmcUHK20*Ut5=^r@-keR`d3{#qQj6NfJD4;{G^?^61rrIb<#I8SF<H
z`NEcE9hy%_5no!SK3|JzZOiTY(NDMZJ>|n2kB7|1$Yw2EHK~r$bYfx_qN|_TQ1_lR
z&O~lzf{LYFoFC6`k`tKY@lhVl+PYTH!E5z{^a2!Grn4U0aV3j5^8O5Pq=DK@xBJam
zzCvCuytoi8BT!oWbozDz%y)gDyY$23`UvjY_JBjgLo~rRZ${H@%2<w*U0w$2xihs*
znFD*s428Ngct>Y^0zpG~qnX&j@XKZ{O{4j)JMaL5M0shFKQHjKTH0ya`FUY@-;dYP
zq`kIgHCYD;keJ-if`mVsal75uo-!U@J8-zcB-%JJNh67ln?7<AYy!eSf?~R}sp_pb
z_g-Me#^GhkOKcZbDl@uz*lSygEWODxI0M@bWL;~1quAulfPc?%W4Qra@U^<{J#iYa
z<nc3*_+)0{0bsbMyp$C`U_1CgC3ASpLE_{A<&ZB0p$xD)N&V&4$8cHr@%<$Nx@iE4
zwG&z_92bBZWGaL0OShPzBzw03>k?-H7OY5E5Jud62i5F&toEz81Y6?;EVTk&&C35?
zx&}pFPq?)Bh8}2?e}C;nj{G$XEO=8KWQGOF2ATS69<4oa=*BDX!6Is>BwLBSga3At
zDe@rSzAM~ZklXC^?06b-iRzFcJ7@4xUx-SlsJdf0>NnuWRD6B6G$xs`^LRw%sYR>n
zJ;Y)lN#m;0Ag}W$cR>E{yR&h*4>K=+CbD^K3!Atqmg_hmI;=I2v4&L``*}Br>Q*#>
zKA<4)sGn{IPaB>7wxL1uqz_ZWDT#_Je>M?uQ_}10O(pro4$kC$Owx)7BDOdTN8jV8
z#dCX1#iK1q*^8k<gp{(oh9x1%G18m_eiP;9ks(p*(L~z3?gy>e<=rIzls>OCg}t4_
z;5<wTr(0!iVMg;h%kpxo*>=Z^#l-SkmV<O^@%sDSSC72x1>@(`%v(d&#4dUc-lhEt
zhnSO1553+ez@?ZWNA-GNpTVY7^vpQjLvL#iw!;|1S>nO?ma>s%7u#3Uj-FNwG8c;o
zThaQS3odQe0CmNIjRTMQhd>Nq{P3_Q4=%u?Gs-_$dP1vJe+DXYmhsl+oKb`-sz-et
z>OMR+(y=8Y<|IWO2zi$lv+VEzTGCXRnv@~V9I9z=;OxAb+TJ%S{0*mi2ho$}5eMR%
zuduf`{lz6lUi0MwAuC$VAkSdAk2VZZf%c>B%Af?RIvwM%qnwQWAU!F}<3|2L)bl$k
z%|UQbmoyWem8HCzcszy#_q^~HvPP(jTq8Ri);Rvd(RZTr`clk&3zT%9PaL8gllhEc
z{$dIvdCaT#aBRcixw!Sy%H5Cf<w=eosyf)4`jI!c27fmW-SOf2{N_ARDh^Mi83H-m
z1Ek+woq(mo_Zb$EFYT<&+_bIvRm@YAtMEDqG`~>~8Mhd?!UK+0mw-v_s+*Ti-Y2^8
zWz$}%0TR8h@9BvrVM_$^%3}IH=T2f_EHDvEC@Sy@BkVbW?the}+f?Z6dx*(90_@jd
z)4o2s-V-gj`XQG3R661B%9o$6|L^g#U^e&ClSGySF@Z%-5GN(YjoLiVCz-1s2tv@u
zR7Akl*}!9W1=R;W6Z+#rbXfm}?pIexJjvO6R`qkINLJQm?40C`P`staV{C6m5a70D
zkSfHeik2C!#Qsm{$Ips3>$IdD3zhE&$D*MwB`5=U`w@Y;ydNY)*q9U4)HUzOo&7d+
z(D=jP^)(i}Z*FfB_J8D67j-s=L?wPW&Yyz%AjQ&{b#d^{cQUAIQ@EIEF^lEvENjy^
zC}|tsV-5DBJPPP#b~<0EUhvXWFfXj|d=z(z^S)^2ae#u0hoXKEttEI2&m=w+enSEE
z-8d~~q&B}Rv3i~+e^|NzTlZ1XyhNe&G^l>35Wba*kn#OdgRFf+W^*5t%pbx9_-WJ%
z;Y+WaE&p{1f-6tC09iL&lYK+L%xAfS)3M5cOiC=rqUApf=7K9^yJzxh4tD$3Qnha4
z-mTr-sObrIs4~;Z;Xdv1xA7B$R-oL8U%C5a+F_!1XpM~6Kp*Ule$|I_(=Vp|6z=Xi
zt@GzUroc@6D``J{@UOd7QK52d0^njgsN2QBx;0lQ^y;yb3>K1XQx0@h!(*I4RQS|e
zgi?7LFz(uJOdHDa`r$bLVMWx^@9=mFjSp<bO~T#DNr9H)UX;(l^LMRc|MRp6YNy$;
z_>x&&RzET}J*b+WZh2CoPt1_2^<(svFFVl1t02)`g86m53?Ip}pNe?T#$w(JYVSrB
zDVaZgb^qeWU;FNl$EZ=@cN#!YK)EoY#~kNT{u1O=BmZ(P&@R{wn_E~^m(dpu(FDXh
zyN&gQ3GevKqvr%Xk&^-4wU|(1yFcQ=;?(NON|-HhP2u~w#XDDJGx+2fxXxp9QqYip
zv2|_nypv?A{>@imeEBh@-cgML$Jjqir<0L;-q>$cxBC!W?U0)YN(0;ksBq|qf4GVb
zall(>EB*StPSr)GwYZLegbRa&|Ir;g!;lmtO(#0e`|gLIgKP}%zd>@&$@|}dg7H=4
zS>`leA?(Pq3jIgys_OdmJmHOs(&kQEP8C%i3iO<$*gqel0-~R=a=^LW@$6XO!>@r<
zmJF0%Ukr`*kA7V~88=#3yZHN$6BG}SZV4@8IXU*3Z2;1}Dd~s~5MaXie+$<@!e7#n
zQT{BMzbkA!4xC@IrMRW1JomIzq?DxfIk5*xO9E(t08hOB^Q@)L9pE`W+&Af{cm$uf
zXgY=k%2b;cKx-?}e1gHlhJkW<K9VDW&u%s%H(jmK`A@L5tB=w-ZL|?pkINhfO3Jd-
zO>ep4<{>@B5o>$Gjvi|Tm~!4iKtP#0&0?Qp=_%W<biJSXv3+TndvQ*{gvgfoZr(+I
z5Tfbmry&Fldqo8y@kSHpfSM3;uYdound?8duI3#-mWj9h92nV;Pmx^F9&L-!=cVq{
z7{5xyP5~u|_j6STM2R@&NR~qPPY-9qyPEntL&yKME@DOq8rVVsB}Dy<Jb+mfR2V5I
zIv<BV#51;OeT;MuasKnNLy)-iZGk1L8hmE%RR(Gkbl^qK#$_`C#K*=4R0eqEzrfoP
z5SjDkq?w03##XA67gjW*2BxgU!X6*a#GDIuz0}x@>2foK#8MHfgO+6($CD-ShC1!`
zc5bz?i-@%6FPdaHKN~D0@qkNb&yDdxS6kxX!pPVMewUZtvyy%usB;ZV1)G_@C;jjE
z>Vt|pj-a=quF(fKa-XC=P9SE;c~n^@K8VdIv#m~nDkZcUKnqmg;c{|ONn^}X#H7~e
z$D0Lr+hWQ)&~2vwLJA-<@0tZn1N8^Bm|uEeH@yx<2L#roD!VNVLk+PXe8R?$jK5uL
zurK<~oS74dKkxlcu=did>YRc~wtx2$c&}o(ab$Cbly-E)m9#uyuq`d;Ki?W%`+A{t
zwn$1j{-Oe-4)&+R;0SQeJy=t1AE}x?EbE)E@!ix`7GsV@{5&!^fLBd0vv$kwgL5vZ
zH0#?8lphi{I--jTUXH7h-S(`|!vs8VL$oi*bGUyzd6G~neQQhN9x`*^zU#E}v5A6b
zrQ_zeT6X&w?WxRajDuQ!1d4D?e=+vpc3drQkI&>`^i&RGNx27!iXEkbvcDJ^b9Ten
zkIfQn6(#>3XLqzD{}<QiA)jP8t@{=UHLpl#_Y_!Me<6yUfSfF9&v5Si6;_aqUnK=}
zgJ(_JRq_Wh9{=;K9L7RP5IE^sk;#?=GOx$o`RPHSM=5tiw#054Gd6Pvs?KR|!wr<g
zCC8W_h85F6HNEOzYo9BC`$Uee7D*i{0#v>XH!nrRggw=3TNoBIT8Q}ASr9l-tAioj
z=v83r_^oRu6zLx=FQlM;PRY`_Rj4*@T-H2VWCrWjqw>n~4c}nBV`qT9P)d~%beP^~
zKJzHQ$hB2rq4%8gaFUuR*BZBt&R|m~vU{q)$#1F)*4h|#{@yr-ilBvs3!OYz0g0J<
zh4H#mIDlVPNWoupn2otCr4W^cEr9}&D-mEAL*LLO4|h^kZrc?MdF(B{LAs-k2?s$f
z)f2%46cp|6Gu3WLQ|Vl)a$3A%mEq9cA48vEq4S*PaBv2w-IZ?=PVTORVY>j%e11DA
zOUWF0a~^cs>3wiS#PX}&ey9D;rhZ)pde}Xn#K>vm4OPBihc#C4vcl6mef9P|s$mBQ
z;EIYDGc~I>kG|s=Is*+rO&uy!A+jv!f3gOM2M{JrauYyG=O$3=n7X)fh>P$@Bh={a
zsyh3YhnzFL3WrRFonTV-(TxP>cZdaBRBm>!{qTVjQcF0Gw-7IdY}GXX8g0?EfF16<
zLjbCeyU(x7UA*%Rf?&7@ykqx;&HP$y+AKnW`IveWght)C!?l@JwMFESg}}=_b)zF!
znYi%Qrg;~MU)!d7I07vW1RGFYmiz*$OFg?2GuIW(*OtAB+%!jYl5EwmT=3sBAN$V#
zz#fp?Uz<}t;7?J2VQRF~BuyppjK#QG27ea(!}=xbq71qFL{4ax;+^si5N*#B8{|<s
zjzr8#C<m*Ix8FB$g+k<@Sh_M}hG_Nz+X4gES;cp!eq${vFc;MuJtxJGKZWed4de6G
z>>7&bib+1UdO**5x6QKq-t%jrH}rww{+^p0T9e;}J;(eYi3ftM>^DovFQlq)W9=Q|
zod+?Q*E?x^<+=#R2g6PO=Sg_it%_$KTl8~vMFBC^_BQjPL)A61gwen;ChUkTxV4tI
z%3aYp=R{t*mG!nhGo9ctqMSoU9xu9Cu%f$60M1HY%B9>A<xXStd5R6|4;UK$A6r))
z)`a)<bq*;NCLkz1RJueRB}}DTLXb8PK?Ugsrvejc=~hY_MY^U^qJ*T<lG4)6-a9sR
z{J!shK9A$>oO4fp&N+8<gzgIx0?lr*_24|4bdQfWcdv%|Z&PcX0t^nQR-dj~dHuuR
zTl@U{Y(@u2FW952n0``#3Bx_5J_uQj8r!^k26xbj#ae4c@rw~-Tto|Vz{G3x*j}m@
zV991YOM@{2cie`ZC${sHSF2`Se9Z~gn{*hG+9q7IoO%DgP;|&kS`gf*l!nziRog=T
zKJBv?GEtV4lFPsPwE}0vZBu&eqSyL>^M`Lv{&rcKI$sW+!fxl#M}5bO7OX7KzUXRq
za<!$XPoB*AvRnREK*vj6vZr-F<_}=?w9O_7X4__|ptD&H-yhyk@b-{yiEfrK2n<=C
zMb~j&#M=NlP5$e#%~Lej#TOkY@y(bJ!7KYiIe0zeFEj|fPE~Y{GQ%`)eB<fH>j=RH
z2RWG$8vIBHYLq->o7&*q@1;??AghrN1Fzzu9b}3R;@?>Q<GN8$O8<zv<|nG$jh>J^
z$pRC{CA=q1*nnynw9PDBYSQ{+N_Gd=1g1}}+c()!gCTQJ)yEpTj+B@TUH+fAX$qB7
zmYR4>5>S2YEcNA@D?CqFA<}uNipe}OLgFF#+fc}VjQwhR$oDJ*n{6lM#;}~C2%J6+
zvrk_SK$H|oo0`&eFT(<Hh1A?FFo0>d3;7%oxJAN3fPulltrCY|K`QS&6~__o1a;G<
zMK|Lci?m*$fkC);hVcatE_hNK;r;Bl0M_bJaG}oW13zplQ<6Ct;u{E;X~1U2*&ABL
z8yY|zunI8nmu9P-_fo*SGQ{UR1__*GYVn1{*j=hSgr|4>d7nEq#bySqFq&mDc2i2?
z5w`=7pJY)cVCa_9mj->h<qX)dFuv7x13z={k1`o8jM@Bs@Zf?-%;ZRPjJI~pTqk{D
zg=+3J{6uUC;j4dFL)LEOdN~Pwq{n4tk@k<)6&3eCw!;QmM)*D7bK*a7qj(bnb>TG<
z`mT5MMRR>!{>;*Rq~QuTzO9^Bb^#<nu4PDe6MLCIx<Jsg*Ste*PsSgVA<B*pM}l^E
z&0L3sffxJ6W!MYVKU1S1*|k79tSwR19Q1-yD8I<N9^VCJ3c0sWRKWEg7vlf?KDDK{
z>-P4}6mE(-VI_rgV|Z;YyKPxoZIkU7A?HHrQ@D5A?#ySJ9k?r94XBC}%wXmGAJzIr
zQ8M!mylY_b?IHEvCEBNhPNCb_Y5<w+SsubQNg9|w9osZHpi%<*7{lJ@Jm7uLPVaYM
z)ZdF9@3@`q6|(MaS&+f+dX*iqt0=f2=wfg(aQF`-`X&|0v*K!CJEm@V67{2nAB@3N
zpPAJ2|6h6W4GjWdBFNA7X^9!AM_w_mG)I^yi3?o`!9DuBClUjiNAY0NP8mb4mDFlb
zp`2FvQAM8TfB0J<UpGD5TY_Cr7|XB(SD5Kzd-AUHL`8hcrtg<WwDJU{LFSOC3>M7o
zI(3O3UP|Q8FS<Y55pqKWaeH5p1XSJy1<E8VfA<#*py*w}{wQB^E^o;+-kugyq~I?=
z)<zVlQ=a%Xa5ufskUIR@7rv(v2{eaE!>%b6yiACHq99{Y(0(^PLXinc^6qNZ<D`b9
z95#w876t0E7|5(xI&ErXhK5q?9qDSu$~7lcTGPqZ-#eab64*t_`yO3eKPrq>yG|-C
zMwPnM(ZJ6R=!TZuI?jfCd;p9^@;$n3=`eRN!x^GmNUt_4%e0+q??pz9|6eHXA4z?X
z!CZu;6Ym;Xr;;_hPUzZKgLD&SvC4O+ZE8>NodP|^8Qhj(u{Ed8JQs!aD1r2^ZVSq%
zNHbgM_U}8(^Z=xE;En2$q})(m+WcUB_P+eQvGK3I@ZkxceqCwE-WQshLd?^yPZS=?
z%KhyJEt8;c4x%-<ZvW#9{=ac#gU@k=Vyaw|9c4lw0d6NKGeX@Vwszw~^XGnhC4^Q)
zM(v$6ln)UTw~#j<x!EEnq>Zw2`)SQFym0kl=E=Py{xEsu%24@=+-$>T+Us^^kQ17`
zc8pUw?yGf2g+ua~EVY{OETAq^8pi6!kC67u?2M7y_fk^Agi9+7)M)-3gA*F~?TYPV
z8BZ?xd24@Q8P%|{wcA9%+gKymGSPb!J;vBVyK>9DT$>WTEW~u<nBy?D^b;4P9s7rO
zKR9Sa<9Q>T=)?8939#70zS=$|1n4@mjvgKV`oXNjI*j-{O;5R=stKbnrNc8r&@y_@
z$@<q$Cy@Ei|I7jm6+E3zI=7$WYeQLpW2vh%>EYhm8`6hLOBb|Y8)1vM6c8`EEvRi=
zq@%sD8d#jND?gDKZ0Bhr+@^4uAbvcFS+wgdfWm5>u1A$YkgBVhK~Yb+!i(cQW%?>s
zBQAGbe~D*^p%mk*nPu10phfoE1Zc;{7aYSYqWe@M2?7zz&9lyBHb5!zh&%!E15ZGT
z?6m$$4WgG!b0$cy5n2_mA77p`lWDMX;#U#902<x2lbgk|k2`-E(^o*t0#XXne1Vqn
ztdnxIs8tJTy59b9+fN?PK(;npju`txuign=*Z#^~@~?&T1JN7_1J!;!Fyafvyn))*
zOd!r%a{6ahR~6IYj!TCl%y-$~#u~q)h6x|$rBj%U1_q#feeblHhk`Szu$5gM?dSV$
zRE|qX*Scey+4>pP@25HZDHO{LzJ8(R=DT8=+uB6L>9Hlo>bn`g9iA3gRZZ$OUbm9z
z%djS)gX7JOppe%n+$#p&Y!t6ttHMw%7X|aQ-e$68?~L1Ul<7+y60YUIfBL$>Z$w6?
z%VJ|*`41}87r~NUsp+$;gRMdOmc(}c7zu+ermoa=+<p-(<AXf`IQ`AZo>;l?EUE!(
zJcNEtU^VN60%b;Fuiy=%!jUNoLzkZ)uby>Xc;2aF6T6RBKX@<)w#lb1Oz;4I<5N8a
zx?g9G!OmF6d%BJ;H{VJR3&ej~uMh(O-IS%>?X0ZaWQ-rlxMOo-f>xPxEr}t)yAasD
zLOzOq0dB&amZu}icY#@xWcR@6__DQZ${_zEnQw$od6KErB)7T5Q>`V3o#LhWRWxC8
zB1FLO$HU_N>h>7ob=|tdGxc$4-NjIEZB(KZ0#S7dZp}x0)Tpy!&o`H=(n=X`&!I|J
zhh4-LyTbb&U+iO)f-`s|V5O;$o?f7vf%F{^+sMTDS2~!DEsDcW4zF&@vTT)7g^t*}
zCQ97z;Hb*-69b{P>y4#!lch357KOS(F_Vq3G@yunpmOzw0^R!0rLF*kY_mL_=<9oU
z6F_=F1ypmIZ*OTSg@=;?H#GkO&6z!#D}n!S0%aOVR9JB*LUl*&;LCuArzta@NO=s1
zdl-(|t6m;?E=wRvcWAR?im7)nG?QN*XjuguzSX=umL#T3OcI=Obo<@rc*YAqCY!!@
zfT-o#r61;TCunF`x$a56{UdhDKHGG+R7&hJ<BmYqKpS+mh(1>xCjIVxyV^d3LVhE$
zZEWe!x0uS2Z4p!3;7g&{>~V1%uMHVqAqn6-fw$yVYg8OdjN{C%O!b#cCubCs7?3ZO
z%ropoDalA0Vph4lk<A<2j2G2+9JoIQDk^wQ3jDe{F>Wsbf5<~j-stko``c^1TSnI_
zKT+DZV7I@$9Y<p&w-q_}>jfXAGR$jyR8<SB9Z@k7+<ql9DK9RTagaGAQNPvs7J-+Y
z6Adl~J#Xs;4Q7j{k53sn{2bmi7Icby)B)QcrJiF@#!?Gm{94%Sp(VFrxZ_&lVnys=
zHkw2iVZfGJoHkpY*gOAjirIK5lXBdv3K1gZy#7*czlq*}9-Db_Tiv27S5SGtron<;
zr&YJPDcV}L#)$HiIv@VKlLjH&SS=6}%s2PAo#_|D&V+@U*t<FN?BiDB6v{REF7{e}
z_aYmKTkunxMX~fjgzNokP0T$wfzzinB~pdr^~Z868jCqw`^CKOrYvtg+b=V55S(h=
zlqoZYD-}n7OlonYs-F;wQNUjl_>t_uiuMqT-b7EQVE&G(9L#_bWuA(dag)zJGveXM
z*R$@h43AEO<GjUh_NA`Lba}OTKGX9d1SPvnF?xYYO=4iR^m38LRuL|ckigzXAKmBu
zwdIBuU6WUX8&$MIE(7ga4|Z|v2@2pIDo5i@?XW=b$ou!nORrFC--fn8f-!;R(%f|5
zBi1T54La0v_e$NUR^~ZXopIm`hm^;EL^FpB&p_m82u__L1=(Dk;5f7Kl0c5p3wrtH
z^aA#MUb+$d&SvGcC2wC6H#J~~ml{5Zx=erVh1VBQFsPo!-<WAIM9f-QVs*Qj&ESoP
z0dO@RNo{K^9_$zMImkp8W1+70-59?kx7c+szF(W7l{tpLk@%lIju(ND9!gp%r+JCA
z_sR+@Er^d&EHJ7wSBcE|Af%JJfjhqRX8Y*<EMZlLG#qON<Kth=;lm&H9aOED)BF;5
zP_IDB32B;;o#M!#!^D#20S-R#?JHpi8`-*yuM8(NN!%Oz;x_bU`m!E>lOcqv!#u&c
z3OW7KVy!3q2IWkx-F|M;E^yr2&)6vlcUy87;;gh`fJLT=H;DZM=c<><o&1vLHu|5`
z9{0s73PT`fC8JZ!Q8dSXqZ)Vl#T>%|GXHy&nELj2pQbhJnx%?XJU38))reBzODCt>
zu6(%loFIxL@=1~U@fSW>q<&fBOF4IRpUu7r#8j1MSANM56djl7jDB<KB~Zs1W}uEQ
zInsG4PAD@GE=><YvlXN^VPyL$zh{!>+e1M;g$*`hdrcCDV6z6^^cRbdXqz7gmtSd?
z5WlgZu<(J1sR&F>&GX5MVr(xs<u?16$Ug1yBR9MTG6v6ah*Tynr#xfr(RDJ!&5ekR
zdyvVHjKVFBtM}7G=7Sec5~nZ7Lg;Shz?(}s`W!m9PvI!IZDr+tO~&&okua3m#>`V@
z-3BgXJ8rHY<4z>LWHN0<wz2J=mDu#=>^2ph=-Mlprh?J2?SerHdE9z@HU!JONuIJE
z@V;`X{*z^Tkat+ie#<@+JUp|R)<Y5=lL7yP6DGTzT^b#%faJxQV+6(JL(vvkfIYpN
zh3)lM8qN1<tj_TUu{6Yk+{$Dn5gjv->_4fzW7WKUS*9Y%^lc5n+I-FiH>>9K=P~F@
z1Z@9@2?e&)fluAm;&Zp@PGy`+#?ipNH1l;%L0BtaZ<ugrpN=;y_yETciX*~FVdKcY
zyHtW~Y0gJ!A(+?3c$&1sZ{Rd5)WvbLGNkP#<v*oxhFB}-G~4-ftEF4Hv1<`u2AEhT
z>tMX#7W70a)tCGGCy>mi!WMoJ=|T*0GpnO#=l!d~_J*LDrTtgw0L{%rSWah&s1=?C
ziv-`@fYnyZ)L{AkUBgdu!x;Ve`eS+VYE@VNM~cAi#Re*G+RaXOMy0uG;Y6ScM4%Hg
z>_y8LAxd!8zoENl3dN}yjV0pCX1#Y{Z+)Qq)FI5+)V%UqujVE<k2vTCe_*+YFqdM5
zQ4tYwlYDf&`+c%q8_p_7z3vP;!G{%2?Dwu#D5iziM+16T!`%F)(ux5$z2oD{4zS7Z
zMAAsW&}(KbC4ah~u2o$5I{;lmO1{rfxr#9|f{Bi85gAnBM8|y9g%4K|sg@*aS}SYh
zg_h5R$8U+0?X_;|%0l8q8iRozTV~Oa-YU6&_`1nY9dbQ|9KSqlq@7)z?yuEE9ofAv
z)qtaxb2@rF){c`1cyxeLvJ9OM<=as{MWm`IVR}&w>yEC#(#x;vaZD|Zj;jdho&IAb
zKPmd+^y8C#dTlGp9eOEE&}dTs;wO{jM0%#-ifZX#EKKCup`~L1O*qq_`|LXTq%gb-
zJYv9BUj;;4liwR?S8e6CAOhR?F!I3EkytKaCTu&hmBKEr8xwElpujD=8Q|g$8h^#$
zSNJMBrm|0Z_AbnRe<*0racgItA}Ws9VCkq8Tns$gJLa`bLmLAD`ktR&2$*?_#(PgW
zwqSi`Y`{gCH6VqTFq<iA14}|iNnm8>gu=%1W?Dw|FQ*$To18<e5}W2fB9%^Pdfe20
zw)ah8`rw`Pl9uA)%5`<_#ZSJyItXGGsDGtnWTtj%6k`LoRHW}`cA6Zd%-D6!EsOE)
z$Rwx38&xVjc#n6E#<6!PQZ7ZFsyNGPpN_T4qXsq^IlEIIW^~ubD6QN8(rih};x}h6
z)W5v8Cv)Tm315oOmQQ~?<$Mrr5aa*C6-hx5-A={})r=)UNl&@V*e{TE>U6vh)+b5G
z+WV>`I7oaxc{BhyT!Jlx{dDXMAeJL^lPf=c;7!x=+XT%+)PUFnQ<h5l3ORBuMtGGv
z3b^`Nb-TFISy5&;_5b<v-+rLFwql*;zHk71^k7nMT8ejv<#OMdy%^DnJ9AFL(Mez?
zkhUt+s$`~&IMq(FhH$5I<_t=1R8`Sw-(v%`9=d7da*Z-0yth<wPi-^dU@*TgGOoAu
zH>X%_-qHWaP!LtPMYY}Vyr3Q{Up<u}!T`wlP?B4c>D}QsNvO}lVK_3h9D6D0WopL%
z%LPYpwWhVo&Lc0&jB^U$?=Mf~CKY*iP_!E2T{j&K?(d&EIyEQKaBtig`DE6;PmDXq
z*Ly)iSK~>iW&Q{-@wC9`ni*vV-CIKG#B&D6#6(>01?+ZQd#ulYbm=9g6o4Ic9+|0M
z4}c3lj4lst11QfmJG@b8E8}!={zrf5`iqc7lOz8T0zE3uGo#97QE*FBIRr(ar*JgP
zX`_}uCBk|fpAfx?0WH=?Cd-XRFM?tkHs1YL2JFF_o<p0eS|&5{{1*=M0)q7h&2)={
z^)E(K;#J9ZPnFt0k*O%Q*!QS>icmQcOv;101R!V3oxb8qM=k?;$r`>`++;6g;y1@@
z@`XK3&eeT?BJf6mK-Kxa+<m(67&ArEHrUPhQo49koae|@Z~{Z)mtWw7BYg7s6J+;J
z$nSnoKHiKN@MvoKG@Hq|&-Z8Rv(?|9ioIW5I3-STL>8RE^;OyW7JH#RFDL#C9o!k%
zcevCHKAih$KAU6z0fubYE0w}41L1>Cc2PKnJB+UgQf8cC6T7mfk30fEGTWqcMH6`e
z3R?gDdu>w_>sb%=e%r5D{Yj5CADkc!=RY|X1dH>99Ks875qGZFFtAB2m`&qviD?BQ
z4}B2AOIwl#5+YM?CnH7_xHhuZ`8TP!KqiAv>x3XUqfDOfFZ}%>Y-GZB@P_<8&!iWE
z3{>;5dpP1ZS9bn5_-si2pr;nGgN5P<ke@H27l#R7+HMvnK&^P^glU8>-T)xWeBMm1
zK{C#Oy*HVc-|us;fGPLoN(p-7NX?2n>+ufzVGufUl@#C3mYD5Gx_$?X)8NZEZdUK4
z;ru)`KG0yupx_#U2Kp~xGaD>=|0@F68-(R5BQJ{aCo|Q>E8ABK0d_>>-uJ#A1Q!?B
zM$muxM`5Hv6(XNKl)%dY(6v+SRX}>bYw@;Bg7XEtSM8mer7vo3mj86>3fbooBlv?y
zE&!}#P=ECjy&MntCe7X6i)#SO`5)pvI|IE*SG(%|p9{!ECZOxBMYC2>V*_@c?|}f4
zSE^N={(A-04|>h+d;OZDs<;5(UR(IRu}_i!iFLg{Z$IG{w;q~YT^Xm`9B7O&DUrt>
z9*o2VUMe^~H)X(U3CHVRV?8wb^EdnM@ck@tJ#wkM=GJh$Mf^G&&<Dt+x({6cdmD0(
zcLymmf@Er10@KQK@v1;LC7%|>^!EPM<SxC&8|Hq%m+54$fuiiY+n;dD4x~h}@{}^8
zRywGm<)o^U(VnsWA;Od!!d=_mZJk-^#%&(Y9Raekw*#L51M}<z2E2iqb9HcxI|@{Y
zr#p|`E1Oo@XA;UVJ;uoa&WmTOi^2+;X@NV;)SCxe9QkY1cn|4+1=u?66wGec+oTC*
z&GH`-$RTzsayeDc>;h>u_L7!iCy-uVCgkQEI5j`$Dl^mX(x0p~Wh5RmSGM0MZj#s<
zY781mY~@cdai!q$$^=}Y@t-Sj+byiGk9jWmA0a*xMLjo`noZ%?VioTb!@3CrM?|qI
z@ZW1-fki$ne=tkczk@24dxn)`X`=Z>EDe;z@shDOg~fd%&qUNqNVLRAw#~{-=kF3F
zVg5)U4d^dyx>@lhuZJ=*r;H%+`)*`N1HwK0begfS{u_55j+uex|1yKcdUt%3-k!IP
za2@@1{FlnU8}X%8uyORznuUmu;SlHVy#RLmA8~GCElQobKN@HTnw<Q92%w?K@NB#B
z($2mS_(+}n;)4IRaL6*Y{avrIqy&7P;Xi**4MBh;H2E`_E5Rv}IT80qOon1JI!h;h
z_f9!$n9h4s1}RrMOYg6sxxz4sAa9axdaa@UZd>St<`)FxucMS1s6`&NJw6-R9$mF`
zaPRC!8&COleH6amRU0YSVQjF^(~5>68`HZ1hDj1<ONlF2c0h-R5oQxaR3gM6cQq|%
zOR=pz7WRy|22U1j;p+dO3)B!&D?j^UU&L8!0hv?m*ZEA@3vZQuFa+pHo=lAgsg{ur
z{(F<45EF%*=AZ~Rb2E<SLVAHOu%OJD$A<|x`d+Dv%z)zKsBUJ%M;1+-cv$`W;%@kp
zna~;KJN?PiRoUvR)0pF5`gEx$_mX{`jjQHLU)DUo&+s)4pyCjn2z4X9$EV@h;3&)R
z$2(OCvF*C@r%mYkt@T#$-{=CB3ek`Ejo`Fe33YVh-*p|`mhX)HDnoo2S0Yl7S#)<)
zwcf;_YV<fZ38!{aG^Ju9^sp+PGSm4#&g?bD4D<rNQg2zT9D?*y(zBkb+_n)SK8j7q
z6183GdVA_avf3*zFAz#LN4^*Hl#1E?W|el=u^*Hnv?0#WG;HLx2VPR=zA>d#JDY%I
zb$WiG*t1;o>dAe`-Wcuo8`d2~VIpN6(p)%YY>@%gyNAqr=C3D}F9=o1Jwm6hb<@IL
z0aZEcou&1RHF;)<xZi|=QKT44+U`X3$jdQYo|Xcc*b1zGHPzN`q?}_>YwMaaK@F3_
zAl3i|qg0NrxbkYEhjRWaOg*nvL5>)hLzY5q-92{3Kf9$Xlp20Wt-J($_Y;qcK*w@m
z3o}GImVgQDm^$wZ3i_+$Ah21#2s4%?AX0?tUhXuZ0}%A4yy$PHKb|9!7sRxyUUW*v
z37;NikSy|Q-ft;QWqAepSuR2EC9Jvog2Mk6lvwY>G}$Je*FX5~j{Tx^()ia^z$M+x
z(~#okI7YE%BFN<xc?x}c*Bs(lPQ08iZOP&qy?T^rFqWBWU&;lXRnRo!!ObB9eTAa~
z&NrBC?e1QL>whU56~cc`gBf*RIKHTQhq?3Qn~Im9p)eSTE)R2pd?xm38kz^CbMsQD
zoa*0#uGMOGsn`x9Y?6uy_t1lgsVPf{v!1#x;bCQx)SH0aLN7DO?bob3s1COW9_==f
z$4avt4}libDWTuCUnX)ysu{$#tDcN~6jR^x51H`bMQ)3&1m3K3YS9x>pgft18xA*1
zxV*p6RRg^7_P^x6kni20nd)ON0J=-sqoeeNM?V`d?#Wa`6y;GG(E~cfk@t@rnyq~F
z(<Xy{p7qnW0Ah!W`bNHP<B?pz`p6tqS%(CR-L0DA7+~14KDTB<dzTKj-i(-wGu|pL
z*1i*?8MLtQBDZQQWNUtT1s%n;!uFB44GbO&2?8k=Rxh`qO#qiCb1Pkyd|50x@k_4&
zEA9JTlb7IdL?I^XIn7jboPno7&TJ?XC-dY_Sl^R_5B!l)<+WHQB~DczTxx8;_z3~r
zma&X;>?bWTylrHj{r#h3YTLQfhi6F=oWjU4?EN1}$+F4zR8<MPsNGm>ER{T{95m)~
zbj;qW*<B1ihX}hHL#Q%jkC*TjiU)CPCfT&SbDG#L&{DEyb|PIOX3mrWLh8CvqgAA`
z(<|gUDeR8Rmt$FrZYRW-p3{6FI&+zn018@?r#QRy47fNJ`tYosw<{lK4Yu37Qkou|
zK+I$+byOC@az*nrovkd$<#=x~&ZJg2j9bI&X-Ioz9SD+o1QBaimk=-T{&Ha)1&}F5
z_^{ONW>a@!%Q>RWmfB%hZDCuFjOCKj#O3CTtYt0$a!poPm?i?Z6;N>|5sovZ_dq|w
zxC&W(2}=)~q>uD!-yg>GULykwX$m)^*Q%|aPcZYo0#5ZHD?jKZ^Jw;5Cu)(<Dr}qt
z+Uj)lop;X{`l$PY4;o>!V}tfjiTQ*^oDCMtLZof0jiufUUkw9iB?`1yxf`c!*YBr|
zH5Ab1j;C$yW38{8pqKjwqJKY>m`Xlsl7UFuePCf8&y$6t%tXNY(#ZP;(3T_an5fLm
zJ&j!cGYcRs9`E$|!=Ub{ItnV_)jl@e)o;Iy;23Uz)E@UHrd9`5GKourws2`sD$k+l
zkydayB-?jJf0$$SQVu2Z|ND{;<n9uM;!cc)F5$Wyt^K9He=01XJ3Agl5+4Y);tjT~
zaM>B}S^BPR0^;Ul>$u{b#`uwO11kaIvkb%a<5!;Y<}F&{S~JD6N_qOKNAV8y&v`>d
zkkUvm$4?O%LGLoE0PmLvMG@vzw6J>G^9g!*mxCN6X){UfnoD1)%zokl`|Xyp`SA@I
zv2XN5vbA(GYSrI7NriL$RdygM-Im!_A?nK+vCS*xTnea))X5iwh}p*67xcUYpZga1
z#E%HFLq5h_M(Kil$?lrJ?;bH?U_O8ko?x)AU$_topk*J=NS!<{(7U+82~XS7y|!<r
z#lH|a5b0W<P&Qd27*zO?;V784X{~QAU^F?&U0s&6P(wg1H~CwqFA2nB1F2Tavt4&+
ze$B;sS*6(je&MrC9N`Y2Iw>ovL9Kgp#Y~&)1&%Tkp2SNcG&FC^i4O~2TmZ^@AJ-Y(
zB{#JB7CB#j7h9FLocCIZ*n2aWpqM64GM;)lf64^45Zebjnbrx#R@<K&HWr`M(GYs#
zztWLPgNnUQgQ<yUI#2ln+wwA#=@E*W$|TPCO&>saZ|?-WtWSHGfw)M+cphZyo9qn*
z)sKo5oJ!vkQOEnp%+f(^(ps_eFuPE<v815<fH=Cxa^GQwU_@j7;k^!HF@dF+rD|O2
zzpdENG`WG+=Ka^ruOdt72W^rY9^9!7NRn{@J!hE&G!@nR=flKYz9~W)mt!%)oXsp)
z)=rjo`@NK92$!-{NW)ol|II;oz#8&YaxPVOM{4>V(H<Ipu@u8vZnY2#`zg<NOk3ZE
z$TMZ~1`q05t6a(`TZ*1p1sN5^_K$`Fk#czL4U#~vQ+TCpr9y=zh7`>&yfEClDfa9X
zD7Ku&(dksL!uMh4ZxWD!>0OBK0j<gpD)yp4b3c&>{Pt=wF!#J<yZw!f<DM$MlsIxV
zDUv0udC+4;NPqIJE2`9Jsij4%(qbiCo7j_JzLRg)1DrHSwO7-6<}4zAkX>iA1)1Q{
z^$ETrb$!(T9RX9!I#I$6tlY?fj>P9|KENy=___DDp0eCGzu-qUN?TodQ)||V_gtWG
z1ox0cN(A}EuOl0^>4O*MidGFi&1BHy`4zf@Vqygu0~M>S?(B8$)`x=I1~S|0JYP!i
zH3A`s*c>mp&>uQqyc?I17u4Q*cgG9ohp^{C^JD@(e3>MOn=t&7Z;L{tUZ3#ht)3g_
zg09ZRS*M2GmH%d-PIM7uVb|E?E6lDYuiiKNDqx3CZbkEP$*qk{oK*f{3PPkyrfb-c
zwWv>!W_U=}AIbY@%$t%b`d2Mygj<UX@p=9^n3@>U?iTUYA%y9=NE(hdul2q@@B<aw
zF;D7ggYYg(F9RdvsV8pTQk}hA=9UbmQs~DyUSBMOiW5gx@Px?5JCJw_n<?^6m;yC9
zJknE~xiJ%<?*Dun;lcZcnvk;bp*YqCxgVB~cHQz0y~_>pd@!gP7DsIRz$0x_0+`9|
zhJP_I{11cDi5{Cfw{FcwQnF}C;yEytg-rT^ZqrNNvMKPy^|(M3_OoP&0ILi2LA>k<
z-}V4?AM|Efqu^t=>cgyXVc_YBxvM*UaiwSBfzwyH&k%0v7WwuyONRuLY=t&Hc}&o(
zL107yk!krSBLkP)js|@04j7rayXy%w_qaP8q}!qB_0;1sbOfMCF9PZ5X*s)*Y-VIr
zq5-j+tzEJm&GDu<a|BNgzB}s2@w-s;kxY)~>W&qU+G#~O90ts@VYb}V6#WBzkAK83
znGgm}LR}|E?RV!Y&Com}uSm+ub1$7)tVv(r@4z3D*rw<{yr)tXa}@OyaCl~&Y4k@x
zP+*uK|7C3$J`+ZzQucxFngst&$xB(f-?z8?HqY4-0q#P#P}Xa0sLwAEpgRJaXsyB_
z?j$Z~3BQoyOJi?JHgc;{-$Jd=!1ox@^T7-Pb)AT`8Tj^uV*RZLi3h`5BDphUAe&w4
zeVVkILVQ{)G{8~bu7SN&FJna#{n1*=$8}Zkw>b0&40~9CVUJ=er>|IJcT}ZB-X*cr
z=e&N2+E<8G^@!?Fmq%qG2rpw%*bDCiPth5JDEwP}dGMf}-S}&?vxVZYAHFR>#&kVt
zt)|1ZevoA~>QE{*;8kfs<qDD06*`TI8(Cqz*zIYjF#DeSInXgh79IpbO@363_!v{{
zC9UaDYy2_4=JZ7owsBg1`NT&dP4gjJYpKSw9jf&nF8L3&1!Hli8S^33^JKH*1r-g2
z3G$aXQ6dR?lPGOw6t9y}@lE(zZg#WG*?oH72x5W3#q*ZcMelm8bZ|W%nY`Fo5iARg
zGmPuq8@y6r*h$)QJ9_#bYw-ijV6TbR8PmA0I@e41TTqvSA<v+vfTrihxWIlI{vbkm
z`R?mIX`rsoCohV$tzMdc4bB2(YZLeI6H8UIQ^b<zGh$0$Z$BDqSzWfws22Eg?<k3u
z=55=Q_4ebkCkS|=Xc(r(Y5rxiTet2{ZRH`q7gg?q7q-?sxZzzo_s_KxE3By^g!JrU
z%!qoA`^<2tn#6fWoX)x~Ryb^yG)%_)JF_GJxqAm#Q=(}m@USgdS((jROEq&IV)FW`
z9uG2)wcFrh3EhlWOH*`<%(FpNjlXyX-wn3<4KPUS_<4evFC5dCTgy2bH>jK0`jJQ{
zkftk;DhbOKpYS|T<-g->Zkj~Mq|k@+6_pUgj-^vrcm}2fJlQ&F8$#Oqnro-wwZffo
zt=dF>g1RIJ*|;#i5_`Yp_SK<4;0BdeHyR4st7*B7G~az6CVV3)i7gX4d6hN1$jFvA
zc#7idtf8}m+Ct-$+P!nc+_@({o)26^>4gFiG&IW3j?22=Ij>>key!qZa=b^$KZ2r4
zxjwP*MRfiiIZ1BMM(Sr{@qCSszu?5q+4KiQtc}$QxCql9^hOL4IQJ#xLP#0OlV9NL
z$vQ)}#CUx-0S3EgZyBn=;HHMmcAwsn`wB!P)!i^1tf|@k{Ib~<;}t0K<VWOLfJmKB
zzY^cr6b*{+L2hlXn*^WwiINHJT9zc@wEk2r%m~crMFpbOeoO=>{z-*#vEJ!)u;zOJ
z06$EV0buY6P33Dm-=ihn>OR3$-LwBqNX+RCy#Uw`Tgn1;$+l`_o`F;QH<RaRx4`<E
zG@=%R=@AkueZ~yOTesWtqE3FQH~HLvYnJh1;=yo_rVNwTV=g#B7nNZOU|^zhDKwSI
zA?*~%YPaX*Q$(_a!K2MY+WPwtL=3CuvD9%<02%=x|3dAIC-GmJqNSgsO?dy!F7(o`
zdv{D#(6_FWeKmU2^mM5vrp3Rd<}AL~xu+ne{?Muuq=9cbTdD?CZ=Wx?rppbEQ^H&`
zYQgKrxy;>7f_{8CzHa4igEeYD6Un64#qX;uC{%3slSZues**H%Ux52v3E^(ewazp;
z;0<~Ld?^vhI!OgCy;^~O^L;D5As-n}{4$ckI~m^-4fD91Pr-Z}WEe4CYK@aE_13mD
zM2e!J14bLh-1?k{4%S;s?3U}qe=~dv%m2#YKdq5D{$p}YPD!0bNko-!!Qqsn>xG~&
zhzO^DRFDn%3r=^-)=)({94;jWVF00GwM!QtWL8E~pvC|$#IKu;oB+<71hzm*;&++?
zug{OTP|tVd)0YHF4>mhffZ@ME74C<uEt9H*#1MBi$Z*_P^K`&x(o}^~CA(-dQJ^=U
zY*-W`w-?Q^3QVGpc94F&1YmqwdFU#byUH=C?ch$tm>VGaW+NOBSa}&HovP*IM7$%1
zu;eLm3GY3ZMz^A<^*~hUQ0kzt=R}wb1fb>Vxk9-Jj$n;qbh!<TV@@ae7b8I`c~i#o
zhcj;!5iWoVrNj+4?`VHEr}laVh~uc4m!AfS)jPENO=-&v0xoz2$)>K7#%wLUF7a&p
z<ZaZPY<Iu(>e@KRty}&l{fQDu<Z}|+x%hWCmuz$+oT1<30SJ6ym{jlKR)x!mc6L*D
zZc*SCfjA^o08H@QKCF4;TN?oU_-CFonTH9KV2FJb`n!>JM*_(cj3rQ*>&d#o<sgN;
zAtRZ|W>Z7Ru6OeuvvyAu{JsRyoh!5YBr~^Rc_op8$PYB0&7a3qiOz-Ho!xv#5?QPV
zxU6*zqD%eU6dfs7Bg<Ph^JlD26crbwDzUmwwfccPGS?PN!%)E%8rn$cqyG5;)LS}*
zkfJ<ml?0KZ1Bd5vL3EHyUYDyyjg^Mm!@{22{wzLsvf>CH*|CtHAiqQAo`(ZJ!qCes
zJjC2_J;yj8WL;Z7F}_4Zp3aLR*&7BQ;p5Z?!(%2gt3k$JF55-yef1Gdwy}v+FgzyN
zaKbWDolRa1WNa<+HHR7_jc*Y#Tc^q<EBVkPc~y7hVzCP;2M|9)_bc6|0+H%d+}p|B
ztz#GPKFF8$m#oxcf<=97b*(Rk3i$OHFVE&l)jPy4MdKc*9o|^yUaD$6>CGDsSj|Y7
zUi~!A-%_fa{R6v#*3%6M@yF{(BVRsep}2?{3la#=(BoUD(`NIo1oL+PE$Q<m0zAa(
zYpMe&uIYKsnv>{)oxZM9YX%tvgzCHsx(cel3-AP^S0VoRMzs=8a6*KB`HsTV(5eBU
zTTZ={g4jV5h2@x=%WeBmpD!mGHpW((v?lns5-6c_pmmgag*B)z&QnunbDARH4(*<d
zG~`f%>DMQs&K-_)afT#%?f{j+M$8My)m-xJ*5M?QBN?A8o7dcZgy({nq-$^Tn&8-b
zkw0EFH13Omd2Rrc)|uzLXZL&ou2_r*aa(!iu*kJ)5?oP~7FM)Yp6g0{-Fg8?zvVY8
z0iUVz(tOIH$8bH!w=J>>@nQ(gIIXvD{qb`@jU27hJ9osf%(`Km`6<cA*C$qoL!6#h
zP{Y<v>Kyj?A5X=-*xQ9jOStNdVZ&82uf#vbY{Z%G$Sk`ivfPB`?0p)uPBFTz4$7>=
zlO_jV3hE}6-@mCmD(Y3b7RO061&KKTt+J_s(MWHORJ&mFjwKa^LO+7{30R->#qE}(
zm%md4-o+2nYS^Rb!RnBzEETtj4b(kS+h*X(@qg00Ef_pf3%bvUz{*39QDDyNsK2AV
zx1HUH9mir~fB>p_Pb#K-Bvp*^1G57by{w+*eCdc~KN5gMwfKcO=9D~su}^Y2NI&(Q
z!SfrzwA^_`VOvlub#TG269f9<#MU2%OKX2y@W+ov`EmVD1JNxX{oxKz5hlIZV-xM+
zRNY(1JuL4RuPy$MMko(Y@CY(HFVSeL+ppO?`w1Dq|3}$3nK=;Vj&d@|gASd9ew3BG
zDV!ytdD`8d3KdQ^(_gsW7PchITRKgcsl-e0I%lw8CL+zasO*##jc5gM$v*>>T;PaA
z2Td94+_0bCc&{)<vmV(oC7JF1_KZL%aLJF1WS9(6RNfKAaSa!gU`}OwJgfV#a^SAx
z80Uqd27tB5?LwbY2FHrb2chea=00u2%Ln5{KDs@PtDs+9HoX7=6IOzLwAOn6gXHD&
z{b>jYOb32~UI*==!Yd``48Is45?I7c41KcC;nVEnl<pVw<G0ets@Dr^g7lvKBK3;w
z*exHNokG8%V-Gq)bSc_~Y_*Bqh)&PaVXaenNZ?dST>UT00H-P_Q*-F>H3L<=u!pLv
zBr?MtXMQw3K}tix)(KWlZIbF=6A6!^^i?5u8=LN1U;25|JmC??8OW#OplcDajNZhp
zVtKJiSj)^762`IUf30|)2e8~zlLNz4s4CoY2CnW{PK|;S(MK)%4B0rTIE_%RqB@4U
zi5n-<a)VYbinUZaEwA=d!v<WTyZSh`k+me6_HC=pb$;p=73`F_G5*o8HaFkC>k>rz
zQYXtMpI%iuKL2iHaeySIJ@hp)NcasG=+O9}mtSA*)WyTF{y9iNmhk3zcjpHah;<Bf
zyoZvt0hWF=8}*ThIR7k6cbA;_&zic??m9Y5UovhsD}JX{d}93M3i^5F)%$N?d!B`X
z6nJ1Y3Gtv)zSo*W-BG2Kxh{_q^p>Lck9?qbC{t+FO{z0i=XhG~6%FH>S!)kBWB3r}
z19UL;>HF7e_xw6WlyZ5|UfioiqAnUU^OP>CSIq00+GOg}+1KLutY%M}N5IJWz`Ix1
zwhC>cyW?sBaXbUDM^0ap+VXF|b!+(0A?{o*JbBV6FA&6^k06Sk`*I684Qy8~!X_z0
zhK)bXWDx+Uy%WP}OU@o+veDd`+3MH+Jb6Ti0wngDl3HF@z;vy&TusYzdJH}`n);L2
zR-B&wRdLgw#9!((DE4XxRYFW8QXG4TjYJh?++3-Jcpx>`Xc=gp`3{(G_*m*@un4|5
zW>w<Gho!n3d&@Tx%?IwXChRQ%QdL>m2*_{O{4Sd!!qjk?8XlS`Qxb_zjRX2T=3Y#f
z4z@l@(V`ZAx1>hER>K&G>J>KTLfGHndS3*64Z+@YW_BQ&bxpbg!M@wA`|O+kFhloM
zw7({<NfIj6r*wM)EK<p&NE?EwD*S~+QT8#AcCow^aL!6By)@4H>}Js5L%|A8c*41v
zs!bx;L&Fjk3u;kTX^W+|FH#e(tTzIdb;h|Pr4&GaNea$V0*YQjQjR?nyf~jW>*g7I
z7<H2p7uu1{la!z4vRH{Od5o@sTiWSUT1IgM=t5?J;z*~E-sq!bsn}8*y5t_YJTsx<
z-%Z5u3%{NFZa<LdMMr_UhBG+a2;#NnF2dd+E)_%Bl8-ic6}%E&8MRG~D4GHS)baY+
z8!d=HG|P|nL<L$u*t4em<EUAX<(i+N?`H>CY)g?<NiYAjmDIYC=Gv_Mr3*y)$D1@{
z;JU3Espl|gxO*kX(8OeBgJaD+N_EDY3q;Ii{fx5sM(~QuK#%gUF`eCMC$ZjXEIIO^
z`4$BrvmN%0`>uEf(V1M2k!qFB9h>}xL9}cy$e>jql34mnQ@c3cW!F>*6b3y?oSI?U
zpte@2%x*;jv-w)k!@K}CtwO-JzA-*EyB_~H8ARS7aRftJ{ANc5oeF`E&fKYT5;aPQ
zmlDz3+z~!2geOh!39{<U*<&hcpj5a??#s+B^azzDusuw*=6YY673Re3EOL~X!XRXz
zm(f;^_0H(?UbTus$yaBXIf4GzxXxbv8WSr*>P7Y8P+i=I^*w(wbPU=y5Be{<aJU9A
z7KsUyV9m{^aI;XNHWMm@qAzBtMv?L2Aa+nb=87G}X<3lZzRrLRVQx$f+ZDcDA$>Rv
zmkOXjb$FLa?3!}#fG5Xkh%D%N7_L7`ahCaJ`<ZCH1@oK4wpp+*-(^Y_H#`UsyY*uk
zsRlZ$cR+KJM{_%<@z^~f{B7StEd%h=W6m$HT*34)E=M0Pd0sIN7PD#&^l}lGN(&^8
z-He6ibkb?hMsQ8oxx~ll1_Oi?udOE|A)g0pL{+pVebV6qrK1PvA+o@K6^h2%S<zic
zG$it_M+}EDle-2MgK3&2e@393+8$$g!0lWo*9)Hpn*_-rZR)yaEOvYD1f@essFAK4
z$sdFqqAgh;t{AV+EuVV{+Za7qbS>uZZ6840qHwpHt?60eOmO&mMR-I}c4!Nn>ki*l
zW<#VW(uVAwBE=gRHH>@r!o&HvnkUGUU`202T_vOp_BQQ^iy09DkOX>NsqQ^sT-B>P
z($bHb;R(nZsDr<>!4c_T4=zDJqQVv*mQmy>-=m70#1CcV38ADYfnw)9M@O!eX4?{R
zO{>i&@7n28H=4D#qJ~?^UHq_)C0XneVL9$HOr!9};q2`$1eT%}l(~Rfy9>ZswD*HB
zAYIt50sy(<m$G3Y6a<$U4Z-|q*+CcecFOk0zshw)$5ho-Mcir4?CNaBNt6^Az$sfz
zw@p}|kP6cz>ae4RDGWW9dJNP&zDINR>Gh#QVWRIJdxxplt&xL0w%hPpdXBdER{VVk
z(>fafcw5BBj<FPuOK2wK{rKrl)-w>X>Ryd0zJ04HW2GSQ(t0~=n_WX37)aZtA47-0
zki&#bpdIO;BI%w2Y6zi>T{XnL);dX)&m2RYp>%icO%M9+wN~!LvG_F7{qyC$kL?91
zXd_#&mXf-DwZ)?K7wYSB5MPjlES8F<Ac2QH3*Wx#)3SHJ{vpWCN`UA$3ZMG^hK{ae
z>SOInrSYz|Om@yMbltx=Amj)J)WX4l+8|^3hogqvfJUWD=2#z)Bbm$9NNoP;OKO;+
zn3im&c1hFBjct=K+_D_`cW|a0%bhx_Ts6c-<g~r%{%n_cALU{|ox!0KFiWPOu^-XZ
zezes>fd#R9%k6b4Z1s65%<<BU1sMSbH^E}3LvOdod09_q@A~tEP8iPKrvNU}9jm#S
zNM)@uV7PNrRnGxXnOXgtXZ%Z($P|pmP50VI{M?qtVLX0kM#jT|8Y45b`2gl(RI+!9
z(V~`OYjH&QQIZg35PhgpYs#F}N%iJ&@++ovz-?z?7=5$AN?vKmNSsYnr$6pKh0-8m
zm(I_A-+3+^Kk%K79yDwc=|iw3!?#ZnS~|G=jE-bAwZm<669CuQexIT(B!SV@E%irA
zIt&y&gjX$<dIix^87`h7-E-Z@8&KOB6$x>J#^btb5MCH?Va{GD&V5RAWyCJ*9#k{M
zJjYsPZg0%;h#Uw5Lb>6Yz`?foqn)%Z!7sJ!g|!CCh4?KguFBQIVho#x8H!h6bJ3%F
zrgyDW!6uKV#p2>W{po}$RGR=`KqtP{!1n8(SpXN6D?_+usr&JW(;$B*rPm~-*V@G&
zZb?Sim(3FP=O*iWi!pTjnISgZA4ePyDlop?gQ6Qz@A+3K619<#FSsWi&VDapX#w7m
z&30EshDkX0@Hno(01u##I;C}i=1`l=U3M_737~RBI@;tC7}k^^QX>wUlFjkL;<(Lx
zFm_;bhy46mAtbo>gTvQz><7dvxos_V9||ntXas#rCbaS90%+}I)mzyf3M#~;n_$Ds
z=c>ob69dsbdRH0#!AVGA5&#^!rgybS&BffRd+7(p;ep?^7QIdfUTAoSUViiTR8Kzc
z;x&X(?v7xhFzMFSlfIZxTpq<a)&N>A68w`-t`eF0k*h;pOq1q_^Jk@n{qMR`GXNB#
z<ISjY528ga@f0`nw)_$NI~vFhsHVH<!};}(ev6fHH6FBJKE&#4$j<J2sTiJrdow^4
zD;pM;H^+P2u3!7414#u#vplsTbi5#RdFrIR&(X_B!RI~7)D4rbEb6UHY>J>f@|pa&
zi3=Zvh1Ztq7DRQV8MQ*Sv9-#s;>plP1=B*Dp3^b4Z-Iv%Zo@Wi-WIQVi&M%Zf5PIa
z+R5ZEDqwLqvU+mSVcJT3lLPj%dSWwx=xGID)T3egj2Bg{jD~IM$-;2$qh)o`n1eus
zKVg%Ja_vUC_}FTQ+Ft1BS<?DaIVK4z0Q|P~-F_X|<gfjT=}oIJ8M>MF+5E)0SPkgm
z(;&LCeYfF?Ew<`AAy#e4boavE0zsG|q0-GB)k3Q$%>(b-eTh$VJ{g3r`C2-n&^6Qw
z;g;GgwD$b@Eofl%*AAM@AGBLKL@QSh+-)joXAKDRbUlz7j{l3D<IXyk4#xd?x2CAy
z#Nd#ycWC>Yo#Wue#y_?L7V=ZpJ9;?gw%^d0*e(}?pI7Aqu^1FZY;&Sv%?0i0a!QFz
zzBiiomJCv9*A*-sSZ*|qUZG)l4XWthk+21S_M#5Izp)x4O5uKYFlWnisw0PLdtCIG
z^<8i^QirlmG&BVmz!x{N1XA!H{#A^Lv9|?5YyixR1R*<d^ExZ?Ipr4EdXC|g%%;06
ze~3H?A%T@_VtO2;JsP8>ZLeNlL|-=j1&+PSnRoY-q{PmE!sO@9xaPesi-r=4DYh*z
zkoD^I#(El8T1aBm@6Ddwf!$JcNj=x*G@%2C(~Xai7hvV_Rl1e<+!wlpZ}*_bz$~$Y
z9-K4+9$c)(N-yTY4nG@VJ8*K35Ft~PQ|k5lAXY}9cZ;1gQD7a%PAr{D?pZab_PmSg
z-;zcMf&<9ylh)c-3WCb-LT3w4Yj7R(`o`B`7bc-%)eI(*2lVH-Q})t63bH=x;yG!t
z*|Je(x~N(*sE<Hve_{R^+jX0-S0lT@PZr~akn{v%GM%GHC9z@lQdjrUMGlx=5xDnv
zpAAbmP+*#DC1j*F44-E-Ks8<O_Ec?!hOBf)AxUr#%nVR-8*8Jk`Qw8v?qjCf5K!EC
zI(pn*%)Ca@8j1V{KrFviI9<cbSVn2O5cLcUm)xEkZX-`I$|x+qR`wl2!xe=bh~ONb
z^XxZ~y;IHc-qf1oFBXI3hE_3WtTk+=jpTC4>a&v{Q3Re{p>?F=taY^Ky=+BR*7WG6
zJ;dFcdVK;+^x3nAl3fPRqG0=sB^L!kq>sbWubVP);G7?xZH5Sp(=GU=G|#Tl>LXdN
zq@aypol7SolIKP<02mczT#Y{8da?JG!x-IndeY$z`^)CJ6&wm`NzRZ+lSexqU8*zH
z=iqZD752yf6?9y27$;jnX)D}x)d_)UB$R>;ocdV)jpZ|Ns<O;xcxJP?08Z>&!N@=p
zFu`HXtpUGA2KMeVjk<T&QIZP_v^HyvN6e`IDd-OUy4BoRVzK{}bhck7SQ5!>HaKAw
zy1hM-_rk4z{WDry&`=bGlRVg&=B^faIJnd{b>z-*NG%O;ms02CcQs_Pd5y)N@7>`Z
z`~HSQT?&<)9SCNp$%dC?$RrHVcoNSz9+@u&EzSy~&(*bWo)E!F3&~C*1seB>{CCd1
zBGXne&8;jEmiKh4tvmPkcEGF@Mc;8TX|U!sHLq*X-~rJA+l85%l=P-mlI067V33fW
z&5vv^LYzU-BINvh)0V^&bK3OovTNMI4W5r-x5`tDwf@b^fX_0##;C#;o`L1HpsO<(
zbf`9v6@{-(HquD~EKj!P#+VX6KTd&Z_1RL*!TS$_-1c%m$+R_aYx1CxtjjAJ6Y`H?
z(1exm6aDrj=ti~Kl$LVUC>z)}1)gY$3L{hxdbU&mt)MDMH{K)i8Y;|_^Wk7+&g=G*
zFOWY;z{q5E0J@%D>dj-^|KnBYFqon+j&Yv(96ML-Bs8XTMne?@Tg$5Z?CFhpqC4M3
z^H*<&+<q5)&>NRK;@)8e>#HJ-$FHkFbZ9{jv75DUHSG|qQV6SY?Z(%?7LEk9UzKZ8
zB&$thM21(g%wx<VuB|pPw#);P{^)v{W7R{};eojTdIz9Nna%WWqEfj)sf+(nf3+lP
zNTfCP<E1X<w{sk@w^h`rXMgpbO$fmL-Sg^`A<{phwzrcEGSrj&z@_WBcRpzeGfy{R
z;I4d2c=rbNk$iMsX4E&mOQOvC#aU{UFx~ps>D=<QoZ{78F*wWlG-w?j`NZ85E9ySy
zVvKy;`;G6l3nzc1KsSM>>oigCjNl56Wo!v*Fo<yO6W&v2>X6|`fXSJ>2CAwDA?({p
zvHYw*M?Y;wh|u83Ux~=g2TUbnAu9!5%e*^T8OKozI4rFiA^5chU~`eM$&=pXr}qXD
zX?59ZAM^}}RhJw*!yz;Po?F6A%DVHxnJ*0--z$Ijyc_`Y%icw<N`uEq0O;Ua5RJw>
z0$GgDbWZOY*{Etd>&fXZ_w$<KB=||05)agN9f-oWt^fry04lXxLpzxavuU}-8BTb6
z2S^qZO0vs+jA@}{^T%4b#b*T5Y2=V2E-4h1;|Xqwc6qcsJS6z3YF^>#y}fRlEZur6
zz==!W{2YZNL6w<X@#rTzP35{TTGa2~I~?HW>Qm5EETyXvCZa#T(pYgXF}<VV0xfv$
z5aZh_1*18PUIMnet$(l=K&lm!*aG_f>@4C690?wvb#pRa`xWLI8NvBB<)6M*nXpZ4
zuO5v}PiXQ!fZ7Gtw9>6rNJlaHBiba8)~oE}bS}=1Ay=SQ=sveQJqv*?+Scc3U~j50
z1My4hYTT6)@GfmHeHsEB1qI$<s1KN@pI_4*CWtpK<*E@Ov(W(fLy=sK%lSP^L5^<Q
z0U*1Y^YUY)3m70eaTi|S%kgngtZ~af^iTV!PBAG`YrD=zwGe@uQadB0a?neRDXqxZ
zmktE>S#-r@J}EO!lTj3uqvoky3H@{|t@ogHa=;t?&Jz1*@~eKl_UNAb2mEs7V48q}
zfcHnXlqFC@P$irTej<OW7!5$MF7$5^{rcITO~AFIu*Vg^X>)_VfH(&zu<}W?b1dti
zZam@wVY~sEbPIx5ZmY52v@XLA_%)dUMsdJt-%B!kz$l_h5H5|{+hg^ANhlv}V!mfX
z>tun>EKG7Q*23O88s5Fg`6nqtl7r2OpU461cda5x(<-Qi>}d^G;5XW&Z(bm6Fi5bA
z)GL^uhqryo6nnLYZ~--Go7TF&X06GNOe(`cEvt$&F2hE!x4C6nG5&vJ_<4}K#Z<+5
z-P{q=G7(h0dzn1U5agJnRC;6U03L}_n4E_X*Jnwl|G?b~+dzbV{KO+*hMN#wBO&Zk
z<@4nMbcxnV`UUpCq@qYxXUf0mCb&&es`cv*`M-v}i33_VmhfKx(3ss_7N$xtq0v{Z
z|1x@SWd-0RwQj(=wJbmhC<Q3hckhsCCXLi|VouCZ{c}5P8nh_djm~LCtvntj7O_7m
zrlX=cNHIY|$m)-F3%7^vCD<S#jjW?zON_uFLu?Reo$^p4B|S^1>w**a@Z89SQj5LX
z5GMmUu1CMiUnLk=cdYsgJ?(<l;dbP+6RRaZO^nyc%Z#qc*My8I@^j;u>&rz32FWsQ
zl)N8l8hi~rtTTmS>2%pqn#zIH5chB2oIN2be=sbDObo(KP}&6-(2o^n-G!2YnHcXd
zDj5qXfMC>KNG_n-peMp`Bv`2rH3Fx<F!`j|lLP?$Bm9p#y<bebRk~Zq^tn}>QNgcg
z0PzbUQG#_fF`Qjs^wI9n%&YSUA<Dpf^dX@fwXg<r>FLkH-z3&0NcSd^!LNzBi?5Dx
zrmQB^+SWzJ96)(e+6!*C1<B51d^v`Fw~P!f?KxMReF)PxETCe$`xn3Z#=CqBxGiuK
z{8l`{GT$<xBqf&KD=g%W_1Ww-pefjbhYQ|1&eF8cSFXfBQedb9C<r+LSq88DT_*86
zLzs&Ij^R0pX3Fd|<I9OKFatoYtp>EqR$ZLs(AbZF7IYZaRx<6v86XB8_D2pW7t~dH
z2Uj<gRBn?&pqm2t9zuM(8@#?{xw7QRU#~QB`jMU5<PT66RoTV)T;xTa!$JN(!|zC@
zJWdUx!daFQ$0LX3wwG-i<#__Cw=R=YGd#zPS(mx@m&FF|E}^4X6c4EWRRbk~;0e2$
zmlw~cNf|9(uHn>B1h%ClHNAc$yjV^8lI>#Wv5}=wFcy>a07B}eVv_CRH}B8W`{}c?
z>3c9P={?v8tCr}YE&Jn{$ih^HK21h!FRx6fcvRjgtvw@j5d_WKOC6oe=Mm-!$6@ms
zbI;C#rXtdUM2dO3OR5%}H@iiyW|{*0FuxSXSl@FrhukR{yT5nznsOOFF=&L%>B(v@
z>fm+N+aU8Vz|~pRp5Av`%v&U-&VsP<admG^%8&h#*~}Mp)!9ogYSA1d2O(y(zl=-P
zTYrsJ2coyagsZ>)mE|G&5_VZ}50G_#Iy9eh@E{}$OdjZ8cm<J#H5jC`(un~FK6fol
z#TM8<Xw%xd^SRy!rsX`XTW9Nem4>ZV7o~3E?D5dc8tV&&gFX1O8F1Eu%Xg4JNTezp
zv8YKJ^B`={J1XGxY;eoUk&!Gf#g@d{sRE2$b!E02qWmA=CM#)GhuRvjCT?picg=4k
zP;Ps69|?X;ayj!l`9VI*qo)t9w9R&D_GZ>P&O2r)9KZh}N0XOM@$u=_>+)I1X0-nO
zOBz(CKfEs*d)=jv(7XITC;XZ7cRJZCqFiqtK7ojRbuFfOm~;Ki=daltUB#ZIxrV3S
zRjpOd4~lr5Ub@qzf-QWvZM1fgKS~lxww#Aa7uaaKNy=o8>CA9(#TEtseXYav3Di<`
zK*)T{zuWmm75Q<8BPbTo)?Oj&xll>-W+LA6ijO#}g01_!l>-e<5GkXX?iOoHjXp<{
zaj{M4Bm|LDHXC0!T6}iof%)I4xa^JTR?1j17mV8osZ!uQeGoFpY05r!=yf<`rH@2J
z#1Jd_`KA614Q_DcdG#R=vma&j2*C)jk@#NKw0){qKxNUVXr=A%lH5=|W!PwPYJuCp
z^9iOko$G+9bl`=(npLMNsoyB8N${OyF;LnWHm>MX#$A?aHA$mE<GaG@;=v#{T#dKI
zmA}x|3tk1hkvRHlYyC~+U*%p-#N1QbN&e9rxXWSvt=sF<%P5Q&*qL2iXmv%E-PsLS
zH2%K%ngXz9sswGE{_?KSOZ1FxdDsCI>cskTqru%{4nEkkpR-Xu8h_xT)?28UzHlp#
zpM6VrjR%KgGGPq3XD?7ES4`=1UQGO7d)FD(bhh+0h7~Lzc(H&2t^_tB1VNNetQfjM
zK@!?tL=X^^D&6R+YehpSQbJT(L_m;834!dYumR~sP$0N~fly5h5L({z53tX3@BRLM
zk}vQie@@PsGxIyYnKLt~Mj#4z_(_ZdgVk<<4W$+PjVr3po#6oH;T|C2e~u9{w}(1u
z$*W{U*Nlo)4?@<tV-!{8#)pOh(-OL(d^0&q%r)i>uvI3>kJteZ&@MWn4vg#Pl%r5#
ziMxda_g-DGz&GdM^?L9zQ!a!Dd8PX`k=xh(lk4hz8*SZ~N*^9RJMR4)*pvuVi4S@F
zN`Z4*CA_jbOE;gn=n2iR2sV@1DfAT1iHwSB4WYgo16#XFzIJBVjxXGt)X+0Ew_TNG
z_PKsOV(5GKy$gQizjqa=quu2IZC6{{@wj1I?)J)I$JTt`__?hOHv4vi1u3Y%Leh+?
z_gC}y4=mN7%yi;x71ngq>WZT^nir=zT!+m8qOZyy6rRvt2jC@bt})}F%~`AS2qySb
zBB(9jG4rd`Iw=Y`KH$?X{Q)|R*OdFtMbT_!0n~RWLcM?9LeTihdIcc}cVu(%hBPhx
zZ#^FV$;te*qu0>(fUTw&?krbxp*&0rJw9R9AB`~yX+;GA_<+yhC{?umc+uUe0u^I$
zi=zJON=G6hi*%0|{8K^YpR2>h$mR-mzirD#rG<39t?ouW9#TcSi@?Cs?8bOa*<4Go
z>w2awpNQcvj$C8HH<I>iCB4KS_#?*<V$vqEU6q0LuPf36PY^T4SU0u)*`+Q^LPQuf
zJ?VpKSk3ww-dfaNG(`e{ujY<t8wR%L-s5DX5)%f@SW;jj-uKW9RMx`D^7FlJC0UP^
z;F6oNnH+Z1{M*opu__b`wK@A*=jw!Oq+U)Ywy;euekl_5m7Ryls%TTexbbzPQ!#R2
z+{H-}&kG|xnbT7YslFWi5Tcj8zrLX7pp=&p5zN{CXGjukWY$ZYz$p^^vY1ssILSPW
z<r57!O40J*ZtEv|DEiDP5X5cCy%)Q9g*7~c6e>T0po0RNY-;TFBIzkkC^s+p0POuY
zL6Kc1m~Hefu;OAg3#hBne99eItdU_9Gk_d5ZQI-S?y`=oO0y!cs6KsqRuh~{l$bol
z4VZEAy8<b!DFN$>*K#t_sXAs17!2XUMrWhI@hREcCSOCSPQ=W%v69x+V8xtkKiik;
z)#RejXGF7)gZ-}21R+HKqw7n~Qnb=bwL_XdSImO@c0%YzlAmpzWj(_8a0xv(6o_$_
z%&bYklU*&!br5lXFK2Z@>q5eO!xARKZE;sWxWorm9Fy8I4CVS(DL!E`{sl?)s0C*n
zWAY|xrUEK?GWl1TX?o>L=gJ9@V9Lf3%#0!n%cvatdMsz$a4OCciSjwFvS-yO#z7Ys
zJlE0&r_Uc4&xP!ED^$2}n%H<TA8iu0^mu1uhMLeUUy};Xts6)?ZfZSZ)XgAyhz5LB
z<_!E$LHAr`dqg-!dgz}Szp91_`bfNr0KfUxtfpPzUndq#-Uk6Fb_DE(@w&tBx&*gX
zSe=pb(RQlnAvorlXC%ig)T`~aT*HKCoF{A`5I^nCe_?bXxPDk3LvDE&1$GEfF(6L`
zO}oIMse+cE3!0|W8V?$gyMXFIZ9B3Sv?dP&fc+zgSlNM}6?`5XM4<btC3_yVEciQ>
zPpFBLCZ^aDhw?uKhFUl!o>x!?9|URG8iF#)j@Z;OI25#Pcl2(k7S=c!kcS&m!f2S8
zik*4mRqqM*w@y`lKX{x{4X;Z982fCgZa)+(i9lT*NI@y<p1G}6$Z8TSg&tl#w&srZ
zD`pLLsQagW>(s$kL_a<UvaYj%2jNp^u7a-K(2n5OR!!D<E~dEo1J$<ZCzApl%FmDA
zBKYL*hja(|>WbhJ3dOpsDzlSwT4^-}t>gp+Z_#IZrvB#ZZ4I+0@fIV%*9rj87%?!G
zn7E;xZ);|d-k3@>&d-wX0N$%eVoAp+th)GR_hoaDV+$ll#`YcaeOTK4s#-@&L0{j4
zWkH|^;7di>OVd3kbj?1CjktL`FyiO~YI{<(?2<lI>$6(3Ii$EXgtc39oz5i|(u0|}
zJjm<A$d+AfNYtg%UjD**8XrgiTDrVpo8@y^E(oWCUM(%7iVCMeL(sY7Kn-kvLR{cW
z084AaaJw8Wce58t<m6PE0|{}Sxs@-ROF}*BBVT~d$OQ6+sEF=MaSDbilQz72CX6o6
z)jgxiUR7anU2box{(`CI=gw!VUB^pp@rH^SQnRV`q>b&mO~xuRjG0za<-hCRN&-Ej
zY^coMkDsYHqFHzt+POWK0XJ~P#_M6EN8d+Jtf-iz6@j__B%QuJJ*w{tKD*!F5zGwA
zOG%V=$fwE!lM~@AoYLDpat_`345;%i!W=8O?pffs3foOKUJ!5A4uP>3eW+oo1|eK*
z*=i4M?LT0886%-}8n!t7f#+MJiU)x9ZP|!ahQrk`6fNUR5z-2-Pq=wmYInMd;F1fs
z(fND9Ir$L=p*Nv#V^2v0=({Dz$HZjMl+stKlT@^S1tr@&Zue)e1<X!luar^85eQ6B
zVBbu4Nm><ylYBBMzW7gdq}t9;-{3!;rVJ{#YTm2o(L@J-7lZ&aM%%-=r^fwE$KE7$
z;%fSI1hUpF#LD$N+8qU-D9m!6W;U<eZXy^PD4l#i%79v?%>!Vgv4fRVUj@KISPPz*
zZIhqvT_J!}-lv_?p5@s+mk<S+7sUXU>JT;bXw!J5HtU&cTS~muY6c`TcI%G$$RWC6
zdW)Iyx3qz(I*6e$fN#aUS!^U3<w((bGE+hd#Q2ccLRlK~>UL(2{JkzZ6-|F#jO??A
zR6)BN$a)Fu;kL)j?AxJ{`0??T5^VPj2~#^?&h>JRlHg`^Oe@rlX$L5WZP%YNws4P?
z;QlS-WDbG}&CG0S0>_m%Oz@_?@usKnlg3Z^<yGM6VMl`E#~#yCdwMK!5V`p22(VJ|
ziq686CsjFVWy8a^v=<3FfIU6{TF1U`(fmfL{RIEQdHsM>NR&VRaJ1x>WOqW!*>cfe
zY*au*+e0bS?kmt3&Fcu(zHcg3X!jY>{3|A9jq>7=A?f|AtbjCE(Y?_t1`frqvY|d4
z`vkq#Dj4xj^2{+dFo3lL<t?-Hb{@AlYTWFxSVNY!ca=oc=V9x8v4K}brD8jWR<VvS
zimliiv`s0<TKPPvY9p)#0#>-TC$CEx?A&#J)uPEyuBq)e=Jytn;F4A;=!V3A$vijv
z#KOp#%-%2V(4v#&2)3@F{HJYU$|b6-JmM}L#2)9B@I#3+#WmG1_)A()+}!#4d%$eb
z84&01X1=>rie^lGjNL~Mi_0YGe(pU3@&sc_(BaYKGT+R~%Wr=Xmm7N*GPC8~qORVO
zj=DrlDlN?d5djE3&;b%$P-^LzQF8ffZDeHOXYlH)Fre6xoWZm3h<yN&B5_#P^g)AV
zT(HVw&bMk7YG`!!87J)N5r-QX^0Q-EVM>g=iI_{&o%2=CjxiX$msZB#2oKG<a11&0
zH$*Z0k0{Fn(o+O>^4{${oM>Zv7T?CUbm54F&9LZ@0${y+;RP(a@aS^yaoTUt1l=!7
z9CC@M9cb-cyp?Gjds+VA3PMjG=}*Okd)=z9Gw@OlnKw1&WI;020T_aO9(I7h08DFp
zS~?bG#rvHRTc-hFs|3yRB08VZaw;uoF}MjO)B@=9a}F8Bq_>u5sx?iFyw_Suvz<H9
zuR#?QyU;^L-{KQZFW{?xON7$PJRp;omi&GPz1zS8H^IJggqX!%Qk^D5%i)0s8DmEo
zd>CAh01AHi?J^-Wv*@qU!^alHSLo>3X0C?IPw{Tziv1@3e9k@rWsWa`)3u{8K0eZ2
zf-B51(MkV;z%=eW2Y)a1J%{-wbAOX2JuzXk;>r2f*V`evHv1B+^WK=6rAH9npD7&b
zUG8t6^0QqbRnPY;*@(~0RwY6M&x4RQtvZmrSNlCIWaRn)(JsyB#+o$>WX$m+XVNlo
zZl>=~JKI78J=ciXc-IQ0QTLbp=U>H4fUqdrRTfy^R&QrurYEX5av6)Gcr7c9o7C7o
zs-s%^(Ch6DT|JJhn;_qU83#OT-j^ekCJC+tUpb>CEtQxw9WMV(3iyYPJ}&7`Th0#s
zy?Lf2Kg|?cgSH+lRf{oj#WOhgil#81is^S)Hhp(Fi8on;yq0vQD?H+JkhT=L3xYf(
z7(PWVG)XS`DxqTbdj@7?zZI|ej#=*=xa8Ok=0jCq9mC4(mr;|w^HXU_Vtzni0C&$b
zZ1KF;Ou=hQ-d$DmF~JSjmDX?vttqL4Z}kMcb#G$%g%=)t_|#o;jR&dRGspw7Xf=Q<
z{A}0cb~(y0{4BP~=alLJ$xU~e(<ELE>C557du+bo2r^$BAhpV8Zq5U78hzn7(QsNO
zzNW9WTV@5vWZ*u$_gKCcjbxTe_paF;y~XJ`c<VlBvgFhAo3u)^6<Ey)^Q>UECnUBd
z_m2iEq&&evvgb+4;T)qVCnsTF0PWdhIq|M1!4w?>!oioonX%&Mimx?8FQO@nuo?BC
zCd<+)kLhsp*q4hTP*DYQ26Rsn{tg}YhP-FTg4||al1u_PfoJMRUw}Tyunfxvd>t<&
zwx`~*cR3$H$*MoQ18+-3fY5TA(ikmfrQB?ofMyBw$``J!%VETJI?0GG$oK~X5(`%V
zbyyvL6-l0O4+IXV5M|v^A8&jm(Esv2fPT5c+wF6m#s6MRSl8?zZl*fyL1-%Ii+JbF
z4NwceXiaXeJ$roFmKT4yjJNJ<Lkf5Rk^5TXB+f&`_c&O9bxAH?@;d>|aQ`EyZJhjs
z<Tiez>}NUeTWD6r{jZli-&e$XpHxALm6F-F(GM3gZ|^%ob|QvjvL_egU03e4$DT9i
zuv{Z?Xd^nR5pu7aq2P6A+_@t#GaNVp&3wn)KE%v$j-8WLnPS7M$T;?69NX2wS%?rT
zE1JTVFr81ND;G!BK5UCOoc3boOh+mU#S{#^7j98tTSvbwW9-c+9rS&799&;fo&u=Y
zy!-w+cmxu#+vLRg&;<tz*TnPwsL9angj*XH2D;?-)qkwcQ>8uK`F=T1Ye23mcw~2+
zM#%RMIKCq0;jWLrDlK1Fty8xvCE@yB(<fp$pdPc-v!AO0st*Z^;GFp+b1-g;_iG@f
zCFDq_E>UOt&K;Xn&`h#_JzVJ&Zz7mC0THL>Wk9;2`s!jwCr4f2i41R45p*sWqK>7a
z8KtVRE|!d_#y>p@x=Xzn;#@;;DFiwqPaTus8h9BWTt9yG3f3)+r0(%^Gmd~WBV+qg
zN6fzU>F{!9Xm4`<EmlZCXH03l#=W<V=x%mCUvmaWmI$q`lyLFYM2WNen8wq<vxeo|
z(a-+EVO?`_Ieq9RRNlbM1SgucLjDb@5_5{H_9T*;E<p_;lm=aFa$aA_a-lk<esaNs
zEg4x`YPYPM3w8us6n*^kvA2SoNKaQ{j{CA4*X_0Ap=gPD6|@Bbi=#xl=pF&pwGl8N
zvY2w-y>_d2D%!<(QUW>jT@~dGI0O#WY|UHBe2J>>TnZWeyl)vJ)a(FuyP0O&->zA@
z(%O}=&davH<>B&2GyoN+<LWv*%N9D)UdL51c0SF7&Qm%7k578Gmn6Si0lqo@`!4+`
zoR5iS*22*ytx#D)w0y464l6P?m|gS~(kFU9)-qN?P_r>0;Nu_b#~Qv)`e0c%uw1N8
zFqNOhS(qmKm+$LcXL^vcU>=ia8~Hl~h*ba(xv8h6EGXrGV_Y=Pkv2oI_*mZkuWAW?
z*Wd-#n#!jtlz*xJ$TmPT&Ms%fnaFpQaXaHJMk-aPP1bciWY>x1!r51GFoe*#v@_2V
z>Dc5H{5@oTwoDvwW9oy5NR8AmXl+sD2K?(VD<6)d{b*tK;FBoH?{@u@_Vs7kOXs`X
z5{4YFM&4BVlfInya$m}?>U*Mjd&?JlIwB`A#Vc#@GGK8KzTTCXz%-a!@V-24X_0Bb
zMU6RT>&J~koF3{ku+AiANXn`JZ6Yc$zu-OVh#f6d9V*92C-*3(o4UhfR0TJvL)=Z{
zRTa*aU&w1m3$*|i0aEc2y_=feaQjlYMMb*3FN=yHC%<ExDH<L41AJ(+0%1MT+08;Y
z-$IgUa+w{rMr6IoPX8JF<8aX0*geDiwRdePZgB2j7uSES`ohT`Od=Yqvkoj0T^dhH
z^$>|No@4VzdyD69c)q>Wax!b{O0%=sr?3oMIOj2FUyDk>)+E|D*#xRC7g}q_5F!G6
z%J;BBvc6V~te=>z9cVI*shw>Kfk6|tx&>(R`CP;JHH7u@Al8&l!jTisk8x15+h5CA
z97;yz?zCo^6XzapN{~sq*V%h8c?(V)v?GK{`Bemu+WX`;Y0*q|k+#-4442I`F+Ih|
z?TGZcKZ+eCOSF&iFD%uo_70qbq>~H>aEC#MfT-@gni!6D8^xMF7jW!(LF{PDq-dje
zd3aU3B?B5LW^s<cNH0Y0RSbQTo3>IS$dr*3M`Yeg8%?u%aW+V#89rX*L_Bt6nJTnL
zArT+dhbn>%KQ<X~T0|D&gzq^h8WHQIZb0bV(EeCx{*{nMO!3O>^vWc~<M9qf2MD&~
z!uk^<u1UoM9z%auFBoovCM#xrHC3W^>)V%};C(nG|AwYSqL*rTSN~n;LYhF39ai(S
zr}mE05g(IUT$Iy;mTcK3+T-#N4Os(jVm31$omFj~gbC7gl~P8GG7?+nN=QYe!6RfM
z=}XL{9z|GN1^VtI1zV|tt(cix#eQ$!tfkqle3wnvMnVeCA6FFOa7$;p!#;;J=oN1q
zh*A=WeJY@645JZ08)~X=ae1O&RW;s}UvffsBEauFq-kjI?R?v?18Tv5ecA7M#fh}b
z;tvLo#@qAWsj6F_kPb6VMhk<nA_P^G=FiEfze`h89+RqGy<@wo4TXJ84bvlzp48Pq
zs0aJl<RWZO1^#uzWK`V0nvw&XK?r-gTB-44PUr-k3gNq;hx0zR_GRKbRRs_d?)d$H
zB#2mEmEdaFn*7ohr;qfU>3Zg4gXO3W9nT=`C>(P&6Xmz}kZih1{$cjF0zIS^7%elg
zpM1?wD`|B#NbTu_Gr%)5d>w40KIVi)y!Ht^ij7fY9T_n08-O-xYCs}1X>eojpf20V
zcu3dN@leLAt2LU4tUs4I^a@qp;!Q&{`oGs^CpuA|B;d%8h_&jGF*X4q-!K}E6PDa+
zwIM!I5QJHDK53MJgxrdYj)tohuG%()REHShH`=DA7wL1qVe7wPOte@Yl>)Z@uO=X_
zyk<kYCZ^m#?{E?^l}_t(zI7{r9Mt$+6td>P8R*<3M1!ooPv+Vvbu?u+Ai$L##JR3a
zej9Zt$9~iU8|YWe;T-$a5?GCo(q$#Z2U@l=5VWWRwE6d&5>O<|PId)5Ct@|<tXZ3&
z=m46{{pB5PHQw$)<aHI9$s~%L9TIBw_7~_wP;!%*WdE8*iooZ4S`v*ca(Zw&5u<lD
z#tEZPurhIp=qD)0q_5lO<TYl@hW3p%^Q`;Q4OZ&9tqu1xcO}>;C8U@mgjChIQ-QNF
zN@g8#i+Ezu47W-H5tf~56dw_3*q!44=7M*yW%<>&Cic5q7sd7hR;niH=NeLbA7>tC
ztsY0EHNMO;-cM6<ts|RlR3ZmU+j6@^%JclE%HVnL7feTD&mtpZh}=Hm*f!D=@6Vtc
zk?7Jr1{EQle|yxra+X6A$p<0Scpjp?7ItY{7vCgLT7T?|fwp-z{-IS)&~YD=pGTC)
zq_@Ofi7CgxB>~cNEY(+;m~z9urf=UX8_PNZrY7bgBQzk!&$3vTY81~y`%Kufx(?r7
zbnsbpz)o1Rrd2-^D>O$I@D{?86}<tO_SK2$LR{!2I=ylvAZ)C`)AgS$X%GX+i1?ho
zUMAM}`;-n&rpZp)w&4p>U?fU4E>UZ<WR;{emgr1Rw;<ZEqa`?Ay4%MhH%?@=P4<wH
zS?$vP5wNCpNV`evx-K?<dcmbo_46p%?c=48rdSP8ABGVIe)nFYC5~r>g5m~x>4VOv
z<KX(O`Hhp8Nz~YG?S<vg)dN7deBSK<3~wbN?|=SSjluqt<M2NZHcJAzp1gkq+%ZFl
z@jvea8W9Oj-A_jNm9L=_pK$;6;E##?x9EOI<cCClSj~Uvzz-bxZ;1UNkslKIar*wJ
z4*cLFz*+kbD)vJnKP0l8N%4clulgPTZ+^|Jh@8IWRq`#5xi#PibHe6$!BO|?{|Dl<
Bsw4ma

literal 0
HcmV?d00001

diff --git a/docs/logos/vllm.png b/docs/logos/vllm.png
new file mode 100644
index 0000000000000000000000000000000000000000..1ead9972879c29e17160f1d3ead150482467b1e8
GIT binary patch
literal 89976
zcmeFaWn5Hi_dg6HD4_y^q9QP&gpx}4C?e9`Eh^I8ISL|@2GU_rA|c(3prTSrcS%YO
z-SAv{XwUhb`~UKJ;kiHOb8q(SeZ}haUF+J%S58Lk7{OTrEG(>J;<s)nU||vFV__ZA
z!aEG^)UozI!NS6(Hxdz%GZ7IJu`so;R<zVNxO?r6#qGNaV%M+m^73F|F+4TU(^I&`
z#PC)Bgr45l8pcZm)(#4fANwoldDNo5z(2uj;a}m=VOm<fr-^$juyEMUo^&nxg2H>v
zI`*Kt{6!^UzJh|^bL1Q~yjF&V;j+G-UNu%`NR=c3Vw?)DdGQiOwGP&8a%@VxI~Ld<
z8qSZC>Twb;P-a8~!wuo9hk|QDaqh7r$NMFHGH?;5NhNUjNZ7O37w@oUN*=w6kN2kY
zb%*0gC@@MOvTfY2-Ic%ReAD;A1-7>eCk1Ao&D)asANBV?a~jciTz;LG_*{(tZA0SQ
zc%H<O7e1nT%U&RngGAS>--aUfk;e}m+Enrii#UJip{LNHic;Qh@uGL-3{Ho8+Mnn@
zgmvCkLN!)GxCc<~#$7e>d(zTam%%k277Y6Y)**0(4gQDaV}yl+xyHhx1OJbO1&f32
z+#$@5JG66sNDKPWSYt>HJc4Vaq-LuoEyZVGVaB2lQJ=-p%o2J4OTdv2T$<gr)rUKp
znVQ@1ISSHXp5O!5&}~*4IOY*s6G0j^X*sxvh4o!H7YiE;8;uYF91a(-zH^UH;l|DF
z*TG+cH1}<7E%{hk9UL539N1YbtPNQKt9e=3uCiXe$_$=hwsA7I)pukzx4E!2$j&%7
z?%Ehw8(G>KS(wA2arJLo*x3rw&_EOI{<F1CTcdk>Gnw0LZwqXY75ats3JV+SzrA+X
z(dhqp4f<v4HO#)YrW1e$<CC>Ex(gNzElcQ%0A_*%_x6@20FA;YXXJR-RPBb5*<EuR
z=pA-8HZ}p)fBorSUsd0Gm51xVoBMv*ep7%IBGf*@ZE*#27qCT$K!A0Z#X<x<eZLH`
zun<_{H?Ao;VvlzbW`{Jn!O=FyUz2=*NxyG@b>(BD|1mCcvFv-I?*ghc*p=kHE=F@j
zzioIsWtAh%ZkCwN;dbY$Swn(&Km?a$)nKV3&IPg?@`p)~^?EOED6w6W>xKv6Eq1mI
zKQfV@8gm`)oE2P{djGj_(Y42YZL`d6DsOqR%`jHs+e0jD7!DpW9O3!*;x7^YO2S_)
z_^SnfwcxK7{MCZLTJTp3{%XNrE%>Vif3@JR7W~zMzgqBD3;t@sUoH5n1%I{RuNM5(
zg1=huR}21X!Cx)-|CtsHNX12EIeJSTtY^L^S1WOz5C8QgAd=%=b=cI@#6<LI1l^0O
z!K{jw>AoKIzSi)*u`_+G6y=MCULFP&vmg4F!_f2D==6m}PYE6!eBD*5KCKC-`4_`E
zQ>(E)iX}~Q31>?e8cUmNz0(EhgIY=#sD<5UR%48F+FdfGCbVQzCoIQrOsq=8E10CD
z*Tlrcl;+C398Fgo`xP=J($+1Pm|mt)q8QT>xR_)2eR^ZK<RgQvda91)N8BFAhRAN)
zq*D)%K*iYs5tOT8X;r1{I|h7W_>X*4q21V*7^k6d!&n}x{wPZi#Z~8p)(-}iJ}yyK
zujaGLnQJJ@r%P9h(Q9*x8}nUb-qr@?Wew$mO=0b2Xt$nqouyLyx)z5`^3nxC<KdOE
zsm-(B=X+M?Jtk743Gc2>wi)-deD@NYph}%`XNz(f)OoLxZNGMBxzyz&@suyR*nU%$
zU#BRP(C2cctkIkFAj?vfQv0$c2^acRhNj7rBGDW6tCJ)m{BHBmGAf=An^YPnu1ub<
zSakcaNKNRo+CJ(*HA%Yqy*aR@)Tw5=)L<P~hJPZvTcbZGJygs3+auzRX}NBwG+l7Q
zaiPtyg^JI1BvbK0ns(Jn+Hl{PFFM_xa<XVBwwP;SeSLC(vpBu;+muPm!bZWX%@wVb
z9%1e|XeZz2^PIltl%<bfnrLRpTc4QMZZR2_iJ9_CR_T3F5SK1>KhL8(IV5?oezCY&
zcBz!?d(fTI(V~r(6~4QrqrxgCc^0>Md3&Oqi%%)^yb(@&F1q@?QD&}XG}dCJ&PUPB
zraHOO#6~Jr#w=#kLn=mK`J4E2I?ANaz1Oy-@3DH^!s}cN7x$IegGrIhnz_Zpzo<8t
z-bHyuhO<>%_hP)x7Icpej)a~=1;d7l&O03zG%BTd@dY2qz#SEMwUg>2m*UOPHZ0tQ
zC#Xc>_0b@2`pu>9*gfDC_z{>`5LF5K5cLx3t`8Bl5NCTk#an@^mn}QY^0X<hyRg0K
z!}r@ZRzI#HTS`RA?h1}Q{=<AFX~M^cX5H*m6_A);2Bc7~p;_3ZERY;tJcVb|?e!@{
z*)$oUs@l@i#MNeIl*o4h;fZ(%^CM^7NNHg*x5yA}0#dr&PLO>Q1zT+98S_3k4~Hn+
z)i{;%<%wI5f;O1xlM)D^@!ww5E0sKKQ7So<a0>igPljg`gWF(q96{-pu3P3l{#;hb
zNs>~Tj22Az4%R2lYcW1K853em%$KeimzY@Db7f3LackYg46i+LCq>UG^OEc=1zHnA
zANO*|_pow0`YT{QVz{LO%5_)_^U7@!w&6jOw^IefF&z~T<QE5*&vN}d@mNJ~v_8J9
z%w=)%{%j}@v`Wl4rMOI%b&E^BF$eZVLyA|l>U3T(^GYa0u-jy8J)e1gThoDwqhR)y
zbuIbn+C>ZFi80Q<S6oGzHK0fEEWVuSX=_`ixH80IxIH`;@;07Yua3Xf3*4O<G4tG@
zkz)yEwhX%SK3?1t3mJx6?QkWiv=fk<I1~{!TdWhAbv~BcqB~l@FUhK^EgwG=VG0Wp
zS{bR#HS0K~***g(x3yq~U{8uE*s%nq1Dzjm0;DyWIv4M^3`Uwfx=7<d@bDX^-oj|C
zR@YE0>5~wTn0gf@6{tLMy3#dres@l}4)d$hCcT6Br$Iy9?PcSvso9~M`<6X=R|ber
z(_#~Lis7qd9gG(AIB=NUPajh@I{|T~8K>Gooq}W<dJV-u;HcQb&>wiSp>Ok2e1_#*
zT83w1Yrx?QE_~wqYI+R`(pja`0TyaRjTv|NjwxL}IV;IG=e)hUt8~4>*=Hz-`}S4<
z$pG>dj;rgs9A7OY5eRuOo#+{HC(Lw~u%Y5UuS_TR&reQTv}GCAd3?KP`<ey;zlTH*
zMM{W?Rg&N43g+M81>y=0lO^jHp6?G1a-Rf`|8qg;zjDWumo@g6TmL8D`pyEIscb(N
zp-S0CoF`1^@vORyfe22MS{Ca-YoJM6BR&>E`Sn}*XEE(*5?E*)9A@yv--<O>kE!A{
z3H3xIYKtYbKyB#usSVq3d~%Ok&*n<{CZAnTshe=U8ly2Oc;5+@!`)@;;NX$$#BA)l
zyLf~%OovRiPN(i5f8xxM?UhGcw8e2TVj0YiJ*!=NSSlZSDp(B_6&l7e|D-aIlF`o2
zt}DbD4ud!+j_fLOC>z^Yo$_EO-VtYPm?$y&wIj(*`QrzsAb#!{{ng;CSpkp$xDFV;
z*nrgry!44queNIXagWPXNvOB?q@s1ygCF?B48f?i<=HYvKt<SHGQ-P*=*R3f>R?|O
zqUkW}Ay-ob1$Jq;bzhwv4HIh?yX6W^{NpH?gPkk~W1XFdfx_#*IP5!(?yHCo3=CvF
zYTd36X3v(=+uT@JG`1)-QupDb+2IMC?K|vLe$ds7p#x0uM6lz&YkKFhj1w0k@%i;r
z9P@X8Rd8l7(^V(%FLw?^t^^<e((6NWCHH^6HVHb%I};nx5TOG7BBl538SN&$mftgF
zl^y#G<nu6YF+<G9Ya}Ojgl>ylr=?J0jmPPJVJs1uILcM^?avp4&@)GA^=c_3=&6Ux
z)4s)Ln9S%7T+om|0hhah4ApTSJu*}m>tR4*^PF^NJVY_rO(p-rRr=ixf(zt1g4S?O
zpNDjJAl@Zuf(f|82{0U2#J16qIWozq@8ca$4czgMf^^>TOJFz+8uSklqV?|X5*N+-
zP#LhP?GdvjfbmxmAM@FjG|UCZwYa>zmJ(OebDS7Pw4Gw>x$EM3Lk!|$4Jq3&XWiAB
zm-Z9w${GdMgVYIU9+u)0KURcSUogze%iGk5Y9nVK1JRWA753*Rx3?1~gw6V-WjLN=
zXM0;y-qJ)7p2NQhFK(@5Mn+x+`?w{Kvg5}tU1q_K!2GET<apyW2kiqO0J_I*ro9q4
z5=g+_{pEW5e5f~?COYzhGO}+&6xiC;%mqE9i~ARew|lps-{R=>rXR*0f~fZR8p@Tt
zf}!duv<72fS{|&O6WCmvjbEKA+jJm&7RZ4|{1u6CRIC~8tKp3Oel6+x#a$lA<pIBC
z_dI#`#@0nWDITj<L8xqyE=Eaz&;arBi?b7fNR~_(HmgAWAmQ+j?(Pj=wuE01?MYw2
zn%EwrTFN%&iA=kT+j<C&Ok0I*5d}zR4{YO2<kU2&-~n>opp#2h8lNO@3zG~5k6s}N
zo4S3@ju&VcTm~@z!<Ti?G02<>?0(CWkPuU_i`NK8_L?xK*&#O}zYo6qEDAfYkQR?#
zbM2B1bAj-~*;19}QvnwdicnB+2~6^=FLLTxL9!?iKMve#C!^m%GClg_i)DC(vUR7U
zn{y<<4^#L8>vIh)LJO@1hSLl&2ws6Fp&|Y|!_N~s$``w{bME|okT=^#75*c`Q`Gbr
z>Af^$O(b_{bGq&(VK|`_0tb%_NQiGjeOsH8q5g?yc)LVFVuNi^V|_jkv2{nPfzB#p
z`%!1|UG0Z#1M(i&K}2#g^VTzQO|aRanpG_|Prlry_1{5QNFE&f$<Ad1A^|iC*Xj<V
z<jz<~H4sAt*$oUz@2n5QYNwNEk{l;WUMtA3Ap}7j=$!om@@z-}z~>nO=ZBw4t$uy^
z5Xog0!THt52N<JZHF&X8f`G@yn#X&U6yFXxjO1-moC@rqu7bgL?-oV~qR8<0+Y3Ww
z(GA2pb6NK<ZYzf*i1wtqT|=v-q3yiAq*p7c=wZ7W!|yOTp@05&*v|>kF~~eyqI*C3
zO%_d+I(&Ya=elbT`$>V7cQs%!*xBPQB@bSl7Dd{PHAm;%t$G<t^_hi3+7sag^E<`7
z5@}EwC>rJ@aLI?etS#w6Oy5h$z<`~wc>%Q$>E!`3vVW+n486F4M~^=VhBg~F3W$~P
z+&CC|<mzSMLPk&=7v6g;hk6t+j`vS;F5lBLeu#yv@2l{s)6KVh#a+9kzRT6e*HJaL
zWED)}2iDH;1d*d*$Ey3{;|oj>fAq56PDD=35a!97u6Jl9!d4|+XZ?#=N3OI-jlL|z
zkXw4oGlL$ky(r`jJLNqO63h*()Ues7Pk$t*c#X`E<~>o%i`cTY+(5m%Z%P$ITbHv%
zq7Oe(aB*1;tzem)6!#Nqow)#fQ6HbRjG^H$YoIlE#GVZgJ_BMcc6Kt@`7OhNkpe)F
z3c?o^21^PO7%@h=5{6C92)e(9-a|eGa&^-@q~#f-%6rGz+Gu{(DwDN-Is}{oMkKV<
zYw^fLoa+I}UL=Y<1%lIV*d6VCDsgZ!jJD`uVW+no_f&W*A}A;r4{Su&UGhbmfOlGW
z*(z`?;cjA~WBoxI9Vy*AO9TSP*h@q=x-u4hfXR4`jJ-OgI#&iEIfDlQCJO{cgl!2O
z5*zm=v*yAD)?#!y|6zeqIAHxF616h-^*Zym%)4O<GUTGD#3Hi07YDafGLFl=q7QE3
zHtmrNv-Z@J40*Xlx99F97T@-N!vw6j1keS&{2B-RA4-6hh!tSS=PLCYFE}m_hDs8i
zVp^(Ca(D|2J^?-PMv!pAn>W8~yd|Zs?^%jRBpE2T)9}Py_1N3!0m>mRz!tk4dIpai
zKbN2Ml&g^z0T+TS)aKlyV-&UitTtzMVjR3rEWr7U3uDh5SUpk;-%C5PvufD*6AnhQ
z#6fT?-q;9<08)SeSPdTIZ?}c42X2`Gt17Fa36%RbSyBy&wzjszr?8@^y%jpjch(&w
z{(+@#{dD>+Qq=F1cAjBf%(pk40U#b?c#K+@=wS6&>hq~%zQtF+t5tBhP{^NIU6MMe
zxb-KKlr9h2UifM_)npu^DdZ#5h=X?nD3>G`dZ_4*xiJXR;nKkAM69Gvt@eI=7!PcP
zTf)5H*i$TIF}_z<M^#r>moC9(Wu4|oznNUzIjUn^ulTlYLmV-PZOCt*rrk-y_ttYz
zP}RxRGW0%<XIM4`R;yDpOq1MFhr6~{bcC{KuUxp~^U7t<7$oDX^~Rhn3EZO<L>a_7
ztWVT(ab`X%88a6zDO=B=Das4O24CRJd`GP{uP|vaGi&9egKQ!_Q6PcwMfHF7+gun7
z)ioE_<=a!#U{PQm46T$t-8e|Bf5?qD*~stTE3F&rtMsR?@`{rH&xU$>LCoFq=WEdx
zuFl}Ydv6S=ff(!@Q4ENm_YnaufwX?1&K+hOuh7>V_v$wg+WQ|o$LwL2v)V8YW$=1%
zDTr9tuiSUS0k)q>aAo98S=UUU-7QZ93(iX>^%;7f<CNb;-#^%kT-Y?A@c89>`vWum
z%T7>OFiJ|w*zt)8N@3I`AE0V#aSNlzqoboY1#iQWcZHpJgfs)yMFiV#@IlNG>glrV
z8W4IPng$;3xy%wCsR1cnurLspjW6#nkyv!(rp}=&h&G)S&o%&W?oFarWItgs(VmmS
zwtVZYJJLAkcZ2qCL&fd04-9dh!eL~Va`FF*?ro>SKz1GLt=-uZJi#C;=7~tbF>ZN?
z9zDL9qP#x37v~~gLioVU$m4*6%2a$42x>a-z$hNhxGM+jqA0ZP^T60=eq`&RwNZo;
z03)EV^^LM+Qx<Mxn(<=PZgUq##b{tKoJ=^n)gWlSyg5}fi}=auJs!V>tio>*IM$xD
z?pdL|$A}-x$CZ?<r&pxa7#LJ**W%!<`5^L-M>RJ$FSKl0eR9})ssB_4N^o*I^k96r
zUxFaMUc&d1?Lj-6Ni3+3+Jl_jfZAhix=r0z>>M!V7tjOGWj9(Z1$iz+ch7?mZkvs8
z7ubg0vyv+$BnSC=S_#$fbIm9Ul2RlLKWvJCBTQjiHy)kLU!5!}J8f@4#cp~H7mt|X
zEn;as-AmRcza;pZ_&al?wPWK>o(hx_8&IFMH%?+NAl;hXF$WcP#h{pN&iF(fM3*>E
z;7nl0-@SnT0N}?l3a+U68VmQ?b@_{oAP~@@M|axPX%{%p|K={8^g6UhAR353{U1Yj
zEY8;U24vx;y}xvXJSymJL%c{*1o;zvSU>b$uqF^sbDy+PWL-4>Vp=;(t+a*OLjde)
zz}LD%iwX-T@pRcK6MOggm92p~W}WhR9+MaCO}+QP4nJf_90#6?(Fx90v#Z8hf6W7y
z6rSNQ)h+pwRoC4LCs|@I_(Ny|C2jc}&vuab`%J?&qwc36%oc|h78Y1r>}|}=t28bk
z;8ieSb7O_w*BoDpkWuW8fK&o8rmVMi;=wh-KX{TClKxOmQ8vf##JJEWu(S?*^dBA-
z&Q9`PtH62&@*j0l2oSgEDd7g5<Gt!M-T%2kEQd5;&meqF2F`?+l1G2<)<0Jc?jWpi
zoSmH|fR)+2L%!JWpDYE^W6P<M#SJI50U;%lVHyCtY306TEVl~1y0s6qGxp{s=Mvuw
z2RE@RwmW3;_G~fHd=>RZ2!tK>)1_{!2}2BbUY&pkBjCkP?<xRT)aQhf@*doYAOqrl
z;xn7%vmFJ%j6`q|gfxhUmmo-u7>K?rH$wpF*3^inxPjEdI>sg@MIhc2rMg={!Vm(P
z+;4&|3vv7Qt4EB@dg3VT*PpzUJBI`%2NoHwt6S=G2n2hF?X6y57BC<UXL2nW2OgDt
zklB3t1f1<E2rpk+0a}%DX4XsW!KIHCf#=|<)NcMyio9i#J0A2x<cHf~+(0Tpjo-j8
zxgHiF1+axQv0B!>8nM##2@CF5w9;hqsLf+fc-~oWhr`c65E*krn)0B{3V!BE&ZF?4
zg&E>XHlRz3f_e1zn$D1^6+2Gn=)YPey359K9(X-ZL~DtgDF7S|Dehla-Wy3Cj7{|q
zn*URAW@s!&e!QxGfJ?~qpr*>x_c;m=m&?4<@fWGAz!iJ1&nNf5B;r&cdOwesd^+ge
z5CX6!<D7G{xmPhMni@c4KLi#RLl4Cz!Ss=<%=&r5)C|V%Gc?4Jh$Xi)fWGxmX?E`+
zvGCv<fX)|<fA47X)-sVVaLUU$7iTWDjlF~zr$Mh*Y7(6j1N{ou2BxHNLw(9^s=FjU
z<McvZCRO4RkfU%E@D1r2fb3pRLsXvZN*tmS0DM6%Yj<V;l;?fo1pRd8WSFqWhFE!8
z9Vz?Sm<%xFE5wo@i-Wy=_m%SgC;Nm$2P}HT0|)Eis&{R|&IRHWu8w1$cReVM<;dQ<
z!gYlN&Ss4y*LAh}y}sBZe0PBJg}*ZJi64eoEVlWFN(Lq>_Fdn;*b9=tU@Ihg<q?3{
z43v_UirHTL2lg2oTMF!{JOB8GtGzqTDUd>@pTvv(g8mFZqZX=#s9m~XA+O;SGYH}N
zSNx}pLnJptT6nB>#GhnTS-aV8H4=-3yaQWZm?&a%Jy8)vpOOj{EVenv1LgHLop%rB
z_5YlKmt<soe61nFZ<WIOL>e1DF^?#m$$T^AjaX2O^(SD(cWHZ+9$=f6*K7aV?CII!
ze1FsMUd4|t0|d+U7lH(4e!d>0rG|-;<5@^K{Q_cs@1&sY@*WrO=K`UwIn_k9yLimS
zX>EXSoLG89PFg@~xsB4~o71DXduX>n8&HzG6$cU?iwai{=fLZyE+a>9PqFH(eOq5$
z6q~}*8vshtw2uB-@M1eWBwzv&3i61m@xD#%?rz_O+)-pDwMX%c#pZ%VmfzC=COk4-
zc=7qWt{~Ah`&_5=Xz%?9RxsqObk42+6lvsBJjW9IixS_9X^8EuHY+pkFJ!A8UwVXv
ztR@hc`yo;`w>Dd6$R&i_Bg)rI5DngOvi)@PKSu_Xk|B7nnKxNYAL}}^jsTHqF2eD`
z_tA-o36Bs=bhsmD@FF15h7P7T{uuG_2}@St;O}gemqR3+!f0e5_9AjY?uP%RVyE5$
z1YZPMU^DE_%M>%e8GkM?I`|h!YrfSl=AM=2ytSAw>;o9ngJInLOwUaz?lVd@dt}-Z
zqJ3mW=w;(HzgK7P3pb1@xR>4D0;*L3_+H#`{`f_C9v?oaGzR{^DdoXz3YrzDX>M+=
zN3NgCmm}hiKM4fp8x>X40hz|{V_%fnrDz;JAPG0N{I<{Ap;VkQOsCj9B-<TYHpu^G
zwIfzmK=}*{sRE*6jb!cjrBu!%uLyH3dfc5pWIj8FTqE=vb1ulrnhjhtq`|<iTPAIk
z0JxX5vhIWC8Vg_-b8~Y`6t@VY>@W%2C_wM(=UcXFJNM%O+QXa!&}n-s*1SwzH(;%@
zQGt~F-8t5S-?@pjzpL+PIiU9=Qgr|1|I|vbc7J>(-O@EP*3zZ^_%no(JUFjUCYAKP
zEh<`FXs3`9*vF##QoyW!S!;y0jN#}X_Ia1D%;Y>E3`++4mMO~5i_-DIISfJ?9;dNq
zw36<osOQI_n%edd|3?bGZ`$o=GtbLuO1S`!n0NQWS$X2eno#Z2W43F`o!cCzjT`dx
zk0^fbEG!NSu#jAjIDU5`%*8YwD95o{UOa7XzKxtfVc$1&!ESn!Z%>r&(dSM6@<9U{
z_X4xkc<ToU-6Fprt=o=s^#XDCee20`Z^E9~aL=p6Uq6%UMj;uzb|(o|#rDAeI`-8!
za%Uks``mp5bAe0F12A_<Rb4l`k6ETnC7_nH_#2qk4LO6<Bmg*GfC0E`>EXj8L|pYZ
zUi}6&lK6Os;$M{B-^H))k~#Hc2C5{d+j%bzI5^4ngGfG{do>X`se%1JPr8l9he`lo
z#e-bC{{&qy39`GpJBy!>&z<u^-yIr%O^cHG`*bi8oE`Qe{7sHz?|ciu97-D}e&FwO
zQ9IA=-|c5|$1%;Gl`2&&tIw1Dh7kf4-h||}c!XVblHCs?K%xGmV=wZE`@A=~&Ml1+
z%lYnw+&tw(>Bt4f0|*r}w7d9DsDQtJbYKZ6Uv*Z|!vUB)h|>gx?pPoksdo6*TjTJp
zvk?SvhzNj&k&Kl&Rx#(T<&lI2X7mCuLQ?W2)AB5yV(3+*5^lNOnm<S%WaH1?_`*j7
z%nFlN_9X*-{M{rOZU};sx;9UIsMfGc@qb=C#h-(iQ<W$a?aYx`JSLF`6sJP}!^GRd
z%K%nFh`>x*(xPYQ8e}zHY!s|!2EN6QHYD~?#w>8*P&nMBu=^yqk-3eR$q+2b=2CCf
zGY@Wh|GZzb5d~U8S6mjxgYNNteetj($3)c#WFuvv)~5^Ui31RhBHzMlYisMoo}=+j
z9E;7JF6wihsJV3^L&0ljFM9U@4(qdEg$dsNscO4-uEN>O+OumfDJCef6HD9^0l-cn
zGgF};e-<`@KLUwUx?E$SQv?A2EP@vbBFV+0(iWGsJ=W(>t#Q9i+Rj}7Kgxh3Uq4a0
zPwO$;^M+tHq4vCtH;kBuhY@nNo_Wb7Og1ih60ZnWCSX__)x%`$Uh~xigfsDjRiQo2
z{<+!LJW;){zJQB^fQ9@=SV?-q!`Tft-1240$aIIU?wY65$Dqo-h;rn9^6gok9TK0`
zL5<~9P?Ws;G&kBX1=YfI?W&)v6586^Z^)~;x(uZRIp2f|srKWo(*`LjsfBUjW<v$3
zHnqHowzD<dn^OYAaQhu0L7L((JrvLVBEoTSRy`i<#hD6xUQCcHO3c=CQlD&aYY8{6
z!N-rg-dPxLi#_X|K-e(AUMHYPKxnZk2Gx6jfo`<7fEp(#`%J#jNU}zqVZT>R*fQ)k
zDk>KhQ&AVP+Sh+HQEH26$ip~miAmT3%73_)KZQ6h2d1jyG-yaK9nsh>W7DkJ`gwix
zk?6C;VJ6eQ6JNbiY1&0B8o8!zK|?gHr*L@9JM-HoCpNQiwMTdR3g9$=z8B7#CLd%h
zC><H|n|zKa!Xz4SD&Qvt444%Gfj+UYS@nNfXE&@}lNdiPGPwDPR91=0v=#mF)ps6%
zS6xM>yU(Y@@&o6Q`5pJY-THh>k4Lk*;RJ)8vBGYl0u)JMi)7Pw|0{y6Aebx7{dA1G
z6V0!GZP0|s#r+8avg8#My)yiR!4?4Pl<SQT)#0-M+9TYg$L^gh1Xx@j#cShk<>{s4
z&_xyJFS5tQ!@wZ9$(^O&)sBC^fcwFIRR($JfAK$~TXox1-S^UXultvwnn=Stv{Y8a
z3=dGi3vOIwvSz)0S^8U_+muVso8c&%+UD)r?sg=U2(m<Xf=6JL|CHE$ci_}8&@|DI
z9_79=x{Sva&oEWGI>`@Ga|4CnI(aw*v=G!{Q&?1Vq&>A*&MA;19Aw{xgD*9OG+06>
znuQdyNnQ}f43dy3?idF-N5EA=;ZG;_z|$=;EoK`q#F*mSKzcv3eYd`dh`SF<>k0@M
z@xmv~)pqEz8>29ZZ_Q~*)8^N&ddV!TKF)MkERGX4)||RJuU%xHlaO_8&bmoiRZ!4r
zX24<qu5^A^bOz)ASI-@f2tKf3(0AcS-B0Nf4x&#u9grlcH_WOSRb!(;Vmll;zG*m8
zGW~@@SgS*{)xZ)H-5oE=7dadC^$e_nz^msKcQ&ZXf6KtVa2frP$Q><`Glqs=BC>Z4
z?mxblusG&;Mz*TDIQ|g=|AWE2BuNugt~H!30+G)x74dDF-YE|YE_T`o$4CtG(aI4c
z#i*PHL^^WK7;kX}BwhiwlN_r3RGzgEC)_7U0G@NxBNscc(*sTp^DO)qET`)suz<_2
z{C&cW9YHEjW*?I_(XS)JNgPf3H5swg#hVHeomt*v9!ljgf-d%EiwnP#IMU>Ij1e~M
z8Vcqhe0Xne%$5*+xW$ikm(IV0LuLU7dp4SSt2aQo_q893^V}20v{nj}jSJsgK+BP0
zjK{{ZL*U!@An+*G=<V5D@A3FOT}dtct4Zva+d{2F=`6Z2S&{m&s-7)7k%h#}y|+JU
zo429IU{WAVHdp1!{}iT!q(!`im52;#cZfRHM)BHPQNu<*`I5#@PENAAh;$O)k|8;b
zfUngmL2V+#Sc$I%A+zhjqvhc82>bL*p`euZzA{$p$FBcbBpT4{eJ07GPGwIf6fwng
zlY1pzGIV1EP>_YHOD9_<P7S7+XC9pkkX0MM5ak(D_az^Xr6o`jDP?maG~Sh2fN6MA
zc8U$XCHvF7{=v;VO7L+y*W19tQ`g*g496QQ>RC4j<*p8P5`Q?}03vR|_eh_^P}(`K
z_DtdmyM@q-`31)+zQ$b?8j%gcSuy^(69-Cb*f4o~kS!=xFhi054Mx3EKl>3{8BatY
zu2xYbk0U_6$D*#@mB{gxK9!aX#h4`!Dv)96EB+Y8fCoa`U;V(RZVt=e=L?mVi4iE6
z>s~BfoYXqZdVfcRke!e}n|vgNIv_;E1)$LJ)dTA-@Mp+dApClIG@An>&;X@!7E@pu
zzq~dBSyG#ecMX_r2EIP6e=X{_!PUGrC>=+IL_Z2H6SnHVA*f8Cz0j;_v3?`JmWt8#
z*|Xhf@2kHjIsVmumWmIo0vIObO`HVTSoaoOa;+b?NWIfno6_CwzCcBER@c`jF8=UD
zO?DM|SVAVg4MFPt&Uvz6_+|I{q{F5NrFg(dfVf>Yx%5vo-2-cmZxUN5<o{XFKPa>y
zm|7;m8-o$mMR1NaEc@oqKc(Waj3YZWxgk2{(tH8Ah6iO<iaD|Z=%pA`bvXMb3zM~n
z?_R$xQU#Rlx-T28|7qis@#hv6+LswP-6%0>cM2eI>a%7dNX+hN_X<uq)fv)BN4%J4
zQw*^9_yAYf0kk|!$#UWWvyg;>DsQ#1!kZOmj-?OkONX(&B0H78sb9t4e11nx5si>M
z39dc+L!SPiE@BARdCb=m^t7#3+^$Y<RHV<%EOMA)N!2QtA-Y|qS&DO$2L1IyP-iai
zGEd0@URV2{VIO)VTiL2=m}#e-H)g77wJS(?0EiFL3y%LEm`E6od4Auc9^SF{S+^w$
zCbjta`8_VU!m1aWwB>POxTo;SNDXs=O?O8szGut<v1f>R;b=m~?{{zapk^c;)C7QY
zL;6p`9^7A?4l*OSek^DHbwI}2()2G~yrSc*l={;Qeu7X(6i^M#{Ii22lFn>MA8Pld
z;j-E2siXFMg}?f0u1i+T+Zx*XCWcwIGM7c&UM$=RJMY{?QA_=c;8qZ-t+5sVM9^Li
zc|!i^@B=G<AV6nMg321j>77nIL|i^hzojoI8hG>1(i!;S-qG(i+17f-s_8i5@nHqz
zcx8l%cDA;zEQ`JzmyqrAHUKE0Hc_yN*sIh2zb4eor$(gfdv-nd^TvSqS^&^W>R0~O
zeL}12bmFuG)XF;5Q|vU``s({9+sSs5bZy<DK)Pz)pLYEEy6<*TDKf}+?^LS(Fn#+t
zKwJbemr0DfJ|{GI(f-N#St5QF53X@7vy|2IAyG_)h$YIT?3|n~^*I5t2;GT7UXbO)
zrfZ9dX^8Re$PwtmBeSIMk&-ssoqyjPTcOI9m%E2UCmXXCV3Jla)K-o2;v=FwuqmXH
z%4J;S%u(rEI2v^5YW6y<GwwumIqxg{9XQ-p<`V1pr;6#XQYD~wb^ZTB2T_ZlUCcOq
ztMi1R&~rI$<N}j0w8~pFavg3<Ly;C6Z5Qnf-jM=m?|SRUmFzI<myAevBsxp?c-bL`
zdEvzzr3erUADQTmsg=E+pPu!Sk<z(j+q;}0MaRDOKla#icOn0N0orC26QtB_KA%(_
zH`DR^Y+?jXkr=wJkLi|ua18(U7=?y`8hP=+$0O9P9cbEL3?ylxtfhm@7V6|;xtY6e
z`RhB7qp4G#vHD*`t$^v1_Zzncw6aWA+yd=5i)k6p=fm`$@Jh#7!e)o-Hm}Rd%J$7j
zMEv;H)sZ4Tma<nMj5`CtY9jI!2et4Y0u!u`lw>Km(*~l7pnF`Xqd2+l>nmA#rfoB&
zN4;opV9caE)&pM$K+$!ogn5kZ7iGz6*Obt?v{0gZ0#bdu8@K@n7%a|1Ceb~~fhGo0
z7_V@t(j4vq-DfBTj9W5b62UNz1;DVvB=)o$p9h#f$<M~POfE~be(vIX!1GA+<+V7F
z+Oq~x@_*@ni%QMwUa{$OHsW^?CfVIiGGYD|X*WEaeL}dCaUa;E2;SUrEAvl@EJfp)
zflzWSrggsjL5ga+&?0E?FS=UZ7=P{+3VG-V9n_d{h$KNs?>hwW2l-HTk{@%-+;}Km
zVBXS`dUkhaQP9m(!f@dd(ZRj`V<D$OAR@6gGv1bEL6PlA8lU0flAM~Fn(J32dwq6R
z&JhWu1nsam)uX!vG_aRBuXAR)JyF{~yV&~81>!HLc?aA3P{$+WN5JU|*kaVNFw6fH
z$A+V1*FMR1T7ASNznF4PJoN~bV=ptm{kWs*#Zd`%1tgj-&<bEuWt2uM>5{9U!fiA9
zE+wsS*pB|(UZw+TZ@80}K5o^U_Nl`CXn=~&a|btDjbRGv3Mbfl;&IY7!2&txdUfKj
zloguz1Xyi3X**T_BuGAux{IBl;_KtXg5^r8X<Exq(KiF}*r|C2Gh9~b%jc(CQHGRI
z&1}nG{&2?jU^6NypgOoyfYMD4Qv6iJKtwY%dZkrWw|t*t-pRfhmlxjEX<2bBHk`vq
zfjyE*OS3AJHMS6RCFPN`T)`*7gZ0%=7<6=W2owHtfelAH=!_PQrLFqVxp2%Eb`7+0
z?CMSMAy7#j&<pnZZ^sGjA_-m**;FjA&5&wDgT@t2C#V+;G_L4gzGt8LL|botWu99I
zAl?QB4kLA&36bV2Qu!>UjhmfY85L%`>cpURfbb3c@c$5``{4+H7?6TY*FxRHX?z4^
zoi24h--xkR8;swkV+#Yd&yxC|Kdn~vh7I_IK-RCI*1IwMVK92owIcG^clAJ7({17X
zk08iVpMlePD*_?}wM*tt&6tkr0Uc0@>fr^a4dM%A?~p;(;1-bLiP<-;-U;~GT6&<z
zb*d{vzryRKL)6dX9z%xiLot@rA6WMtGD5&_L*NCK>%NYjJz)b!1vU+*gir(TY9CJb
zbSa0bFkU>515@GXI8BcIg`-2SknJWs^&;v+Ky;l|P3Tl?{^vl+6P5P2`F#y2D(xAp
z7_Mu>Z;(EwalSz+N_PI;*`&D4ckHvm51%r7Wji=<^4nC&)>XXea_O<DUEQe7aIl*S
zXWyX*8QzORn#v55X>ymjk)J2E4HN22(n}Vq=f_&6qQf7MJu^W9we14dk+p`i-2U_8
zL<ctkVP{!k*DDG4Hd^64#_+h(T@85#P9C^q4vX3E+KPX__<;+PmX@vq-T&4)<hIe_
zEFg=g@D`mE{w8&e;9CwV^KMnQO(mtnxnL8oi6@(_l|uF+%MDDq`~$l^!`LuiV1@**
zD4vIU)c@TnZ)sH0O9W8(OXE6g;VaHTH62%Dxy;(>%WGR6wm-27M4ZI|!R=hp?GIs(
zl!^o@L0)!~nVDI4d<J=RS0@p1&tV(!-Qqkof6RoO!+Rn8fKRaP`&Kk>D}$O(voY6z
z`~spBcL6jrR?6^=z7<_oxnHyr!l-()Ma!mUu_Ms>B_b17B~_!zwc;4}x#)4Gl+Z4t
z#OMW-C-xzor0&Nzjnfg1vmvWt6Q56PE9V^zU;?FN)BFEe!%mpJZ9jsYkqWBHa^0WZ
zs>7~$G%K$g#`k=CPVYfm9G>Q(@>^bP1E)Na!@Ur+p{|`Qeg1^v?Ah1gz@(1a#&X{C
zh7i8S5I5D!VtZM_MW}5KM>eqWzo$jSncyDh8;NrX5?p8v(dL&ZbkQxeHL`7x=5}N5
zQvNEG4l`+fe?|Z((#zuJGJ`Jq7{`8g5uf93)|J0X_cDXvi$GPLb5L0C1(|zn*s+n|
z4)Ps=m$ttjunBp7F8(VZh2LZu@9(Ckt2To9l7@i>k-X+K?a2IM<EL^<PUPC)C~53S
ze^-Zze8iPq%X*a#$m<t6CEh<I2-SgGE>cvOM{TzSRNaw8&9i9ECZB~wEh!BJGV005
z9+177B~2!co}B=d8UM-rUxBO%j3#YE+EB~$dV*B=%;sR1zUF%6Ih*SG^@R?L?%b*M
zJs=SdRrl%b?hX9W$^8#HE0f?^kae5|k)4+EVaaJYDCEqSd1-gLB;?~H4@>p_6r*1E
z%lIm2s618JIg_aD73=U>TZ~811c_#E)=1NkeK5VXW8@9IrFaIC1~_z#@;_B39wjeE
z$+;Z&@OUOVD?r>49I*WjGWu@WT#|k=<R?5Y?{YypnW&}bpDrr>x^dJ0^an0ijn8JB
zG8O2(k=tnjb&JAqc7vpUpAvKkmZyD*ET8=1@qF+sqkP9(@#icAgg!;8uUXxQM<kE{
z1lpto$AxIdZg<}hnkt?ZqI9o4Sy->tvp3ge+FtTy3Gv^%HI;2qx%D35=r~Mu>$aiT
zZRlyaOV}+%JI~E9S@!$VZ<xorEqr6YW;y{P%Bc;IVMb9<FnA)4;g(0Tou!isL+f%%
zotGD(2IW@GK%V}`;l2_QdxbHac4D+;-m!Cu|4C8M1TzGSH@-{U&6exlnQtZhuG!Iu
za`-!Adhy$0PoC7}D*jcNtibdqUvNnAe$TjSdno}pP`0{^o;8zPea(16IVZN`<Tn-i
z-7WiqmQX<}MVj;fi<coNP}^Q+J#^&w`zCbq>BUz*RKHh2jVSizlRLEG{NP-lUm=JK
zwIYhj9pusA1lSuQw-FK5iT><;y&xwp0bbH1iYn~R|L+%agvgAl>EFOP>+1M@Xy6z+
zsj4--ni1R@@sLAWL&n+-R6B20NAo+BMZKWVytkzEB;7mKgu>tu?#1hIYD~?vG$i6X
zVq(nHOC8A>gH^GIodRGs05+%{oFJ%lR_NEqa`O7HvVA?<m^YD1_%Gf?PgDvmd`Jgw
zEWN!V!MGUUQ4VMT3JJY<_m(!&2|43dm8_Vkb8e&7HFe?=)cf@FQexE1*utkLdxqy<
zZ4sgz_dC&JbEn(0Bp!lRV<PO-%uL4JS*0%JiZ7{<+&JIUwaGBFfset`9Mog==P}NJ
z6Sus7A>I&(+I!=`@Z1mI2_Mr-g#k0&RaVfz<#eXj0T_)V)iHvr=%b;nuj&D!KU=52
zxG+y?tQ4*{#g^p%MrTLD3P5o{Fi$bvazE>{b>r6JtS8yy4N8ULm#RaE8tXK(K1{9;
zHXk|eYi{nrO0#SBcE%Yc;Q@GZSv@%BId@U(p_E10`oa*c<jZsN4__S<*Gq&!&Z-XN
zLYWOVPW(KUy{HjTL|GdyZc5eJnHp&db*}|DDeaw#01m0}MqF`*TleewD?Q=zTDGHI
z-YzO>td!YiaSs>$Uw@(cY12`<$o+@sLck?4VL`0bmou}ON{aC&mrnT&YIkmSuuO)$
zY|=Hv0T!Kq07S0*DePuht@j#}<U^01X46wXWumxunr{KB9B-UU!5n_s*(&8J2H_X3
zrsAVX)LN=qjz1~AU0xRJ7<YeoF5=T%Lr85=<OI2);2TJ^#=O<2c+_dkacP7ouGZA$
z@5$_#?Gk&Nm?X!?53<kCQ>En1TSHz(pHVeUAUpN+FrJc#9so#sG|SxGI3G!@odyxG
zU#-yUxCO0{JpQB<%B?y&nP(gHACN(neX`*mB-jCRAjw6g;eIFVh*j6&TchVTcbn-k
zzudXG_M{%W=!uZ+R6VP`UF|NwwL>}tF0RDM@e~p5iJTUlplI8AaYFt^2E&h<xu(8p
zr7OQBtyOZZN$H3uO}Jztxh5p>B~$K=M%gv1vsch8-sO};?L%7ap*F`ZoW=tscXeV0
zR#Ojs(a{Xu?!x(})VjKk=;XV0Q>H0t=6$-)SBdWDmq5GUQQDmgBXWkRej)IE3E9Mw
zz_i-Um9cbT0C6`Rv7hHzq0eAMQdksEOM=FO)}~P{%BiLlRSxI5k@Uv=k%*Ej$zSX!
zcF&W?T?c+r6;FYCzs0$gf|_ZO^jR}E?~Cz>Qg<Bh3mxn3h~Ycl#s8$-=2>V5DdVH<
zfM_>-2)>7W2tt$?0}!I*aJ2krDzNCxzY@V`_ghq3FmMBZTncX5{nG{jkiy+<1p-vy
zGYk!r?ciyhuLcvwyPWzO2$I9LaLJJUTaFWfQE<IjC|fJ-A1{lPcU&nQe*K0&GrqQH
zA&?4>Kg_JF%@Z8!`aKU$ImcvK_LV0@M>{~E-2P3$e_TTbcIJ-gfzNRCQQev@o~_OD
zb2-`s&|FyyGc(Flq(dCYCxo32J#N3CMt~=akzbA%*^Qlw233<-NrOq~eEs&A86{wq
z9VSZ=`{&>#R#UxvELtF)XCgh3OeZWW&e8f6Y0zULN)3n4flB&`Z{79fR~JEioUIS)
zXbqr{m}g#jx`#9=1o1)5mgud@(icp3=R5rTZsmY>a9|MAl_f$XIG3<kGUY~Lt>#=L
zsZ(Tc#sS8N%_%}6U?++37B5mjel(K6HpG`jXys>+5a<0Lak$|OH}~+2f>UTq?v{ph
z$9*K3xtNilbk?@(g7@aekg_*zi?E7-q0K9sro~Ar(bc2x%RDs4lv7lWIRc!0V=y@}
zMxzMy`%RTOtwjK=*3R0?OmOud%`w48ku&sd0o1}9tk0fGTW5hj_%BNty^faF>@U=t
zaX?Y+sUj6<-|tpWNnxM&idB?f%Jvl|^l-Q>`MlP@^<Lz~fgkrChzp)^f22_r()jHu
z`4$)Dl(EamltVi7_Ompp6p??vdvl0BZ=-UN?vzgCish&xMeexH@hM6bay!4*-TJC&
z4#EHdCJhB2!Y~Heq;AKn7D@Z1?)`0#+m4O!oL!?lDRdO=_B+3k{(*{{c#wVk;OFyG
z5OO_)$V91d>obAEc;~<dSSRv&+-Dv4#PsU}2zLt;<=ndkN9fVR`mkkVi%y*?-3++o
z-9q0<y*?Nbz>oj?)2oL^EPNqB(*)J~kV<azP9A$h?YZ%`+0tlA2Y*B=PI;}eUC!AK
z%t3fWoCa`sP0zTFL-ZI>hHIH{#{$9hc;2$op}7cB$TH$dNzNQ@opZ=q0xy>rVx7P%
zi|}^{nM{-?N@<;Tmuce_G$Bo)+Li!>3}neiKQZoOA#$9+>&Rol;S#wO58ztVJic|-
zFRHvxH%)cREk2P08t$5mtd`JyRUHJD&^H_Q^8y}x6G`N;R8iIi+;DxEbCdP)*adQ^
zg+w>K-ShDGctrjY4Q~B%oLHLWp>cr0_oehK*E)2+tRjE-de?bgg(4xkL}mGXLl4Nq
z1sX>e2A<V>Z*rvMFTCy`*vuW6|IT=Hq=2{mP0P#t!1Sa1xXbyIJha2RBW;;VWFx+p
z3sourWx9hDLLEa$V+5RL^1yk8?)2{!XIaEx>A;MC%<Fa@bFopfj*NdK8@lB0l4*8=
z7kv0)YlFd8LC&PerC+~?A^~p_2-DYhsmMwXsa+qH_U7y>lO!)u`CV!t6<=y$-cH=1
zEFw~|`ROqJek1}j;9I3$=yc6D017HbxY^aagFdd;%d^9q(sno5C8%%{WTIF6uD=Sd
ztnX2o)%T&yHTv|JguTbU9H~U$#eKOk{;HMuPRMo{YIpU%X15D4L8+TDzL)pJe0s48
z-(aW>L6wKCw*K!2@$yq?R7C~Joc52w@%cKxV<!-C-(iD6scAF8kG3jx%H)<MDSQRk
z5C!3hutSHBsx{Tu&#Xqh?3v62%_Vc|=vDD_dv4-eanVp800kdNIU738#4Klk4A=eq
zd*AMe<?APRp{-y>KsOuCt|$BG1}B6KdP_>`vQpYich=AL6s1|tR<MdHcU(2?`4V8|
z6$wrv=3Zs_>U2=FK)QbfM}qQks>I0Grx(h8g0Hl^^ABmhc8W#obeBz?V63dk#*9oF
z3Y4teGxSB)=%wlM9(=!R51P|%0-&Q)Ib=*;c2A<7od;yQI&|&S-YsxA>?Cl-x&`J9
zF6sH{OG^%hWWekd^t(5?78a@Y1jzSHnt6;{e4WnI&x<ooI5|eQBOkj?LCy%7k?%Ij
z3tB)7;#X@lJV@zf)><sflEj*SE67gZsHADNL6r&NbnV06&N=}ALHF3A0XHpU3U{Ve
z5D)S>4egF@MnETB8ce^1>?s+tlfbl7jqm>Vhu$uwf+c-Fd1XV6GdK25qtF}EMKtpx
zKMii1(~I}*-xGwBq~Y8QI6EH?ke8u%LxHDOmwpAc2(O_l$ZfNu9brW$pc7A<Q@@*1
z9PId1D;L4Z`;kUP*>$_LCn`)k8jdgr>#3_zV%({!LUP7Ike&PTQH|@Jv3cES4abF*
z%GYnW^GcA;p)d126irMwfdn~naNWW}w&Ud=QJ`>m1Jy$ib|B_^H)h5GZS`vbLQ%2d
zH8@gOTg5A4x(wysqTocn`3EEn59Glu^6B6^N9!U|YUVO|+1YCh<K0vq4P4wbZ44aM
zM%rAL<#*y@y+xeP@oquk)cyMZgBI4OUlZ3xND%6=+FoAOKowZqnh`Jg$jBklPX1DR
zV$YZSr4W$1X;F|52<&Wjhvab&VVp}$MN3C3&VH%~ZcM1*#}8uas9%o(wll8-j&rv0
z(#bh{<3P}aER#-g{`HdhqDV_>E<RB2+@J-h%x;hc1)ofoJvjk8$RK8}q(X#RYqw)d
zmgeR!3Ppl+Z9CmjQj^;)FI4G2udi4<cIL=lXW|uTTo+ybXl{4)D(2fo`<+YN0?Y+O
zrUvvNGJV}{_-y=WT7Lf%zX~_jG16y((!RAU#qNdDQ9RSCzA~9lzD2uC7KD1cd`L;*
zxg+Ld9h?8=cK<H9x@mf;gDv{d$M%!K_yRH2k2!bra=k+-`<slRXGO2ev_6z_IgJ}P
z4kawf7EZKf4f|{7I2;ISAqfl?Mb1R7gD*v`e`9J^`4$14;l~Cc8ejY~x8IDqFmPtz
z>ls<$b=x{u<&s%r3JQuU(7I_eydSF%hMNAet^~c?XG7ufklhWc)Vbuke5B>1hKHR?
zAN9-PnCk0~Kt<)(1_w7c>tjEw7~nmBkByBPbLLuMwus2V0b)mMb`GOqXqBy4^;aOj
z`JiUdin$mPoQ{a^(+T4w;4fbiQi)qD1Rd(SXDPy-djg%l2hMrF9DJ*Fceh_;KnW#G
zWy3f9eJ>_N#O1>JUP<M7OXlgIvm*V*Hq7itL+jF$%!JGWx=Tt=9~&)7P7R(Z5Y?~D
zTvA3`mGJ!In9QwkL_WR3Ibru=?M{Osv&qhaRUlbQtzm45_{Gw_Ksg0hkVmLaujP{R
z3j6?NrV((gVAJ}zr0QOaW-X+vZC;G~+Fg<D1TP#YG-jMft97d{y<h7K_<2&6<&s^x
z+0E%UL>9&Y?H?BB>wdYM_{TZsuL&rCXgD{l_H1snQo=IT-B#aSKIQjQSykK4C|`I;
zvAJE0?lRsBUDoIHvbip+k?D^b09(_5lPMed{8Sq`!|pY%l6ZB;u&v^*DKxq>KHYDn
zMzRsh*Q#eY<B7Cg4W~n<q*%2Uze>s*B=V<NR}6t<d{WlH;MXy$EB`!~l6e`F`Fb$m
zoyu5>k~ZYX&qFT#vBZ30>^LEldgg$3UYC6bf0nM1t9MtSo!`wMs)=DKfe^gyC1D|t
zLOC{4TxtIS;BOdjkdwyI|77)v?0VN+dGsRF<zvjW++ET4xQdM`WP&U_EJ4QQm%G{B
zg7b_{Zo}h>5%0GxgiJ{T0M+%cSGF80CFv6Ch!ER!??WOM`}fMFuR+d#0#Zu{cZ+W=
z#Zxjlpg6y3=d|eI@p|O#a~WrA00!!ycAGOj5nCVEU`p9LUx|f`z=Pbuo!<{XA1UQ5
zLH~$Kl<*;|&&?D}njg*Ogu)`8GAAEO`_`d+_@7Y`HY}R#)@SVyAoK~~-xfY~1M`S3
z$6o<rX=)$J13L@Cr)$aE+uQlZp~Ujltl*u*yKZi7bH1{!{m=V`r7pWu-4oNH-DP6V
z2~e|CnZaZ)+x7#V1xrnhcQX{yC>~8w1Rq#&es4Q4blq7}&=O%{I@9;+JL~O-IQy*`
z#-CclEz9xg*X}`};kEc^R*+}}`~=OstiN4}G=@a=zQ4F6+?{QtsPo+{s}@pUtJaeB
zk6aI*+0|cPyHn!>J{GfOEs`Ms&rXp}e}~OrUC%BP3u!o^$8o*!drEWB_1Qj863riX
zEeRxttX)b61`T&IARxeIqp@6dcWc`h$Tb450IX&okPq;d!6s5%hpNM=wR4P{&<mn|
z-`}8s|6Ihg@LI|eXfDl)0G+(TSuU%!cdk0UXL+uCY3GF8k88l9@#E1QFqixUAYNOa
za?eHkSx8+fc5ZTupElu|JpHtLYSGM}m2UjpoP;yu3*P;P8QNuBCQLtZ&;v9o>D}e8
zQL_MGTu?d9VyxkNra>jS-rNrJ)!*QQUzb}AoQqa~0A{R2gRiJfjHV)n_IAey9qF#D
z$=+vTJy8^pjh%90B`aNwKI7fOTADH!a&jySmpy+mA<3gI)k?I6fd9zJ{s&R%Nr(@x
zO9m6!7ZEe;EtkO;B`83a5*;blQdC=D1bt{ngCKt>N4Bi7G~>>VV-T7o3H~dB-@ein
zr~~wi)6mCz(uJN|B)mx?&b7x5aJT1O(~>yFA~0)ZY@HA7N^Z4(`nn4{$4TOLj!!D+
zMDGN^Tb>5(8DIH4{YB}_!!d$X@-CH?jb|-cYIn&vxy>FkgQHXCWl+5PXT0_`90diF
z;PX6XQOaE(tX0!>D4}#kctnlyrxS5L)LC0(J6bRB;;iN`m!8#b>AteD+lQW$2fyQ{
zqYaY1xl6@U`7X!yOEQR$0Ri_MfjeIj+HY7jJ4;`p;t_YwmwVl<t&Xy$!T&f%kEq3g
zaLhM10}9U@-gU_D1SqPzeN6=CL%vSAht(jh%jlv9jW8DB%oz{811R|yW&w!V8q_wf
zmsh^)%JD8jhbBMCKE>I)MVyp_a7<Dt8<?;Ede8;f$!H6^Z^c6TPedLab>f7=Z*gM0
zP3jif^F+LujUm&>EK9~DwB0a-M+Opt0tp7$4CJns_XA-YXCWxTl!d%`7IO9i8W_5m
z9Du@{>+}df;w2J+!qrHaL0~vB^(7ifSl_W@lLq&Ut~XuZ;)pF&d6Q`{+D#llkk|~c
zI<J@Mm)u7B`Tf#e4W0FyFR+RRsHZz=x>W!Y>2y8mDD7(M0pZNVhigT~<GWu>+>PF%
zVe$4~XfHFF9z|<`!`Q6y;<Rkx0Eu6-%O`3PMkb3VgHPMo$qDP1dr7_%Fum<#Vq~-&
zX}M{qc$?Ie|AW#-r+Lvu!(9r#gA!BDoPd4hB*bNa33RBYGL~!}a1u2gT}wj=&8-E%
zD72jXa1FL&lXeA$_aq*nqNG0Zw;`wxT3`NAnyCvQfDx$A$awX&=w0>RZ}hQiKPrfX
z1>jYjxp}$z6^}%r<9{BxAiv)=1jzT$w?cBxC}X+_w<WNM7QOPN{-Yu_Q|DspsJG9s
zfHL3P;JlCti{Fqw*c~M@qJHvk>hJ-y0l005(?@4BRN_or798*2h|U<jF=dZS`28)y
zso^P_542Q7X7K|gk^0eU`gBq7otdUMhCKr-e+OBlioLr$aPa_PVq(-<z5JP?R4@l!
z{T`$UT;wxV^>V8Ij~2RG=@5>++3uU`i@II+f}i7@xJXQe*G$VmnY<-*EF?haPN+m<
zqAq0D3A~nJOdy`489fj;4e~zn8ql@m$IGDXTSN?fE~kf7=g^Ufe%rL;L{tJiD`+3#
zwdYzVQ`PpC$Pavd>7mE_d-1=9BGi%uN1AcHtIFKW`UZ`NPRaIU(CW@ly>FkHoanl}
zk#m}=R!7MlQIxe1_|J0Nns_hj5K-|eNCPyN6EBpuhoeA^&Hcn=(|l_iZ)GMcvAxBn
z^ar(S_g521Y~$VT-$37*zRgp6HT>OT$k#MP^)v7BQ)xSa5+uem%*0eLIlJdg`+CZ*
z#G~6!E`|twNSHWQW#iq{DZKIAVs&nm{gNuf{senlr4-<%C*(|lK-<zs4kxsM51Oib
z05W@A<;G>43xWYuy%Xvh;adngV!b8`zC!qhy_V;fg^g73YJEi(p-F-D=4GsHi}?bC
zYy)=t%qK`G%WL{{h)UE}njP+uHEUU`j;5=;uTx&0xSU)3(4{k$O3+z=g#6b(QYudr
zHBR?z%wi{cW3kMGG$<FhoTa;rU~#H2hjP%g8EXIC$YR!X&A#WPBfK#kq}$NinlE5t
z^L_`k0%W0t4wLQATtJszJi?K!dHvhx_`3Cm>hcxc+DmH<&sr4KZ-&x$rMx0OHd(@J
z9`zZN!{?hl0Y*ehOKe|dLv&G|o#qTA8Y2XtY?&t4M2`{OB!5J{j73TB141Fy@b1ix
zDuBN>+H9;X58unX4i4NBPj(zHTxd7Zv=@k}oB|(grNp3}vuZ^VyVe44oT0aGGLU9p
z=^X)A1W=%O1$$Lq?%G^Jd3$WP*3!i55TS4HQt^fK5GG2hEh?WVhTW+s&j0DE;Bvf3
zg~7si+Okmd>L-!oyl>Ttj@Dn9X;Rf)@0bxRvAx4s)dkL3EqIuK7`LcHnywwH^EiO6
z57v9H!x+AQs+;hy_P)AfiIT}cbQID+XSAmoIQ=-~A{i^HDxosVO;P<O!lgnH*?LBe
z>}dhB@{OI1XRCfRHgd3oe15-zV+KTN;*_&YMt%vPSj@>gH&@evGg-E|flWxxA~0`8
zf9`7(BHyRO;df)QM|43spS;}SWY-&T<Z5`+_H{Ddu2IDK7wm6_k(la9W66T^=!oC&
zDczNB|MD855~Z`jx&eWzr53*)wcZ3=)jmye@gF2Z^qRrwqb$oU0Hm~#`|EhIaxYX5
z&MroN1{oiV2@KTbHab(%OCxeJ`IPG*gS*u2GPSOhc-G{_B_#S@(o&sTjne1cisiT#
zN*K>!vY%n!{x6YYK%?Bb7{*CPp^a1%g2nDxt^bd%_l~FffB(l{i83<EDp3>-L@9)D
zN~L6HCM!us_MV+el8lV3>`fUVdn6LF$=-W!$M(CP=O{kkcfbC6-P~S|=XpFIkLz(=
z_v?PYU%Ds5(Cuk=JEK~LpL(jP4A50`&hXFx#DI<_@^!k+eBtaWF2ISA0MP$H&I!8)
z;q!QXg_UcNj|(dnju<Rix2Rf?A}UXhS_S#UHZ{f#XqO=Ela4JVwW8#d6we0_9ze5R
zC_5&kg%+Lrck^`v?_#~<&K|}c*b3ks7mMjbE`_D3W6rB)p{xuK&c%Ly@p!PKQ`ix+
z{QC9l?K`)>e%YG`A}`vM*(>2z*lAIZjgjjw4-7qB<5$y5Q+fye2IB<YS&L4N3m1^m
zzP?s?JD;0PvqWeyg1jJOduv@d;6skSPH5m-PFGEk^WUf`CtWN|!}=zHZO_I_WJ@(v
z)g86)G$>fsSEQz?wq@tHG>ga`(0P`D+@s!WOB1{T|J-;S5J0Mvpqp#9%CY~WwTK3Y
zP7YpxqK}ZijS{R^o)YF|Z7VLbPCUGYK)`X6-Q9cwpZ}Z!{4Z>9sauXHXz$sCu;ZRj
znpF8X-zzdx)vRc8iAv`&s4z&7zW@;B3PUIOQLUNE`*5{(;Fo;o2$PxRp+1OwAfrCg
zkPenvLSAigGXBE_OBSEa0Uhv9Zvl8CU47J0E*snkpMr3dwxi5}mx8i=5y&NiCJ<U2
zD1+4G8~MlMqZEL(`OQksLi*+3HKVwLMhKGu%;*2Y4-|~E2hl)k$K<WIl}HuxWv0$o
zuHU0gZo3mJizl*f_-@VIQ<c|2%X+pjQ&wx-1?5N3xj%s`yAOHN@H6nF+?u?&jXau-
z0_$G?cXZt2KRmL2DY^v%34ar@=u<3Atg*}-;$;_y-e!Y=V%ygE56_d^x<;0IMT(NA
zCQN1Hxc+RqXT;fUU1vOfb{CS6`8wmZ#nGiYx0sZz{%N#6Qbshl{AOWFG8n3Zc%mYm
zqE_H|{9i{mToq5}qsb3{;EcB=x8@&!Zr6FFlbEvbz=>wU5n(xWTC}LHYo~qn_`O(&
zDKNUfVx%D2;kt+EbP!6dg^N{se|gc2Ub5)a)PnUFEHSkkf3Atga}(z;FcawVVtOco
zwZnuQrak_V8!^^#pqWHoU*iR`vP(Pl$AR#P<h<!u&{jR^rS;;MuUdD31Jo|GI<eth
z?2kc=6Sm+QnCBY*w`M~<A;{JK%C&nA03myTo(+%E?0Y*pAY71OJcK~>O*G!#r4tik
z_1&91KeScA6H3s(oB$^@-LtY_Ff0ULB{KuK10=vFY!@YVzOha>bTKhAlRF3+ZbdE)
zqC=@0;(qz|tMmJx)P_1($G#nZe9490^v>L9lxJbCPc`>neGa3S%|3x_9mlu3*gc9-
z5Gi3ctCH@W-PDuRT=h(56Hm`B;5W-sA9x)vDw0tBt)%nei4(ar9fwT4Q7;H81zxdS
z6->ZxTJ&2&h-<H8aCDmRMBtfYR+02IGLbB(<XcJbe>G}=s0p1HSFP+ILh3-_9vC%X
zDLTKd4duL1$>#6sRch(KtM?!JE5N5l9p#8N3L7~gG9@xftG!tmUu!tX0UN6mqeEqA
zGaB<$gZedhX)s>3#OE{`z}i>CS{e84{u1uD(aDFOu7|9uUVb8Pdjb0BkkCrWYo}6?
z&jZ{cifCG8-roD^Fb;Zeh9kqXpE^bGpF3c>c=_i_DV>5%+<t9!PLJD=)ZY6vG_Ln}
zbEg<hpqRZU!y0?nd-u<MIWMi&T4;HHtd4DlO=x~9!TI2{vXX&iOWvAUS7!#3WnFA+
ztj)hF1blvWW9VLVJ8aCAr%W1u|HvHxed9R9OyjDKwoYtkYWC)gipl_Q`%m$2&!`%p
zQS)+knC9rSqkOv+3J3;M_FPu@dZ(Z{l3>&R(wU1+cKeSPKu3{K!ffehDZg%&bHAD0
zJIULzSY>*Ggv;#hS4-e6G|uz#<L)u~lL?OcqLfA-93oNUoRk}&=YEj(=$84!=PXBF
zQ5K!pCS#{GleNV$^`GL`^JIPk{r*@843ZG;a_6|R+Yj<0L0UwYs^9K@x)(iIwu;%Q
zVxF3`x5QG17EPy(eh((8sJP%6NPjV$J26E+Z+_Nmsb$Es%21%Pjg{D1;Lh39@A_AV
z{kmNxxsj9_SFOHEo=`nocLjccnj(0(peKCL^=EJDcd|T3XsA7#x#9IjTtB!J8mHS_
z>>mshpV8c&i7)=7oL>0FG=}{AxjY0eFCr<*D(aRIncID*^9ZZ4`^m$C-5=>Or|QY;
z7Ehl2y47Np%*GzFCO!9&R7KDCjb_iOIwVy$2C<&N*adfctda8WZ^C|*mO-cXT1<du
zNm#vJ@^Wh%Xo@3|7@TyB{X#m6GcWVvN9b9SUslmLNK6vueJlhLJEs=(nHzs%FSLUd
zpoH#}EZBnq$cZvuD;_n&l$4M=+3RaI%ah-9Dtw}4@q=4z9X}|Tr}O~C)&<vpoQ{2X
zc`U}@^Co$9&`ZQJ$@AJ3GPjs*j!v`55SIt((gTP(xVwt8=%cf&W!~%bS|K}IfM@5s
zrtciw{bHtwVuS3Avg_`T^q8c(14q7%kT`929dllyt-Y{0r9Ywf-uY6;k<ktZ{S48R
z*3XNy<j1&5Co>7gcFWwIEgX)A#1RQiTB|&tawQ>X>eoYSZ!QL~aY{yt>BGDa9hRDl
zBvGu~QV~*NLiNfzAxnzfQaABjX$f0i<d&X2e|;x>pr`>NPr>`b^f#Cg4&G(FmNB~A
zTBRbq`CYjzvnaP<F}*BhBw@|G*|zuj@77n)O>jB^mlcX1zjOcH@@o}XqcuyZ;w3}O
zMh318z)idIQ&ea^?E7odB%2d=r8F9%?`|KN7FoVy8){Q)1Q&hW1*Ozt!}UJri*S_)
zfD`M=qi%J|@!fl?PHi7=UI!b?p7uJLhoEGC%$klGoBLW@wR7{f<vsH@vlPGFthVBc
zAd)M*Wt_(3Cj$l2hh2B=4|nq6pLwYBod0PKbT5J9@#bB%@hO84DZwSVO>o89PN@0X
zk~~3BiE;ZrJ>_q)d8guJ1k^M2NVCg-LtJ|okzwRR*t9wvk9RN|B;WDF8#+kM`I$9j
zHi^$!TevX!;${VlOphFm_xEVbvrY<?rMd_a+8_UDfm6f>U|Ew<kW$KjD*9we1?*ex
zY~HpDNlP&W$>90wz?e^@d8d6M{^nNPN7KrUZ!ZrDtGX7-Hi;T`2lz(vy=UFo6pZrl
zU~zi6%)3s<@T`4(GF?eYyeCa6RLgX>w}MKZkrz;rN=jOCxl*>K*T%-Di865*IJR;d
z+<)k`u)9V7><2$eLX4Jzb5|&5%0mFx#@>sb7(2FgJfwQOK~;XW+r)Z*3Cmb=Bv)HY
zON%IYy%b*DaC-L5O8d^z-}Tv%OV?NKhBoQDXzt_pX7cB|`}2*|RfK1zon!WQ2Bt(C
zTCC|!_s|JQ181OC3sCVpuBg>-dSbGEXRu#$G~@Y%YTVcr+()^>@Q-Y#g0!?tCN?Y=
zB8Vu@Y$$@*j!D;MNHDBL&2GvIWFzz+tXxbtPp<wHRns#xbbLdAUM7AOru*A8llI=u
z$@0<#Mvb<uE%h#Y03c-m?Kfk8v30KjTZJG?NR>Zd`jf%7ZBi)8tnqd0x03SV?j!$5
z+3J??ayrx_jCu<1LA%*PtEDm4?iDmjxrM+@JuW8i?WDFa7zMZyJGV-C{!;4`NO%f7
z-LPL{(^EO?IR!prR(~hNXz;@6>><4-u_wOW(s-7TG4e^&&8P5E)NKE!^0z0u^0_(v
zbkb{b<Ecm|vMzniMoQzMf5tP_Qxr7WPB{SZsy0}C;4c_YOmZ|T9~k^K3Y7&Wlw1FI
zMMXutWgx=;VMF*o0{Z^w_q%^SS`otQVok`e-5;6f0OZiL{Q*M$S{@}yRopc>1KqC9
zArZQ8V!ZEkytf7O$_$H~21JF|WQ?@qOYD#CwsUq0=+w(-xcx@}cX{VxX<3sdOw<WU
z#!K1@4;CYY8Rq0K5Fl{KclYB1ye;wC9?f)iAZ(?5-PNIGRI0#sa^O!Q1o(#!)Camj
z7m)E5JLf*$8|f{NOzV(r&sHZmx1`uJu9eW$o=#*|577(?exP-bIPc{q57M2dB6{WG
zI*pEgS{RKDA0d>0x_t~*z#a>r(f75h=>d`?E+Qf#dtdyfTA$jJE4mHQRV6(Zu6NFr
z-~IF;Qx63W7n&YKTS{wh4xi@92gAU;Ezg7Yd)kR87yN6TQxyIhyD3~74Iw$&_S(vg
zG0q}^sMM|gJh2{|XR2Q(s`V|ZH+f4xbjJa63eH2*O~wtU2Z<HM2XLu_Jb=+jr@lrY
z$2B|_FWH8BNUE$C!LAp<3s%kd?PL9bK~SudDSdfdahO~DTitaHXr(&AU@kqH>l%=s
zJ00$h?x*A`p!;*;@d?dk5g58ech@)q2(x{fdRgVP{z7x#C)_3+EmZo1X^KK)#+$wr
zpt5FNvqjc_7*?ob({!VrnwI|!B+Y+$nu<<?=I%=8s3HMzSZgp2Yvh>b4$9T&5txkL
z4xu9ri`VR#^RS}wgc{HlVZbLGCKSeO(%-8#xVuBT7!`wLDNj~$4=?`+2gM2gCgGwl
z=EJpyte-Yof=T>^F?tCITe{nj_zK{?bdR=Jj!uVd3vDQNR!3afy=eaKl;_3!4$>R9
zfDc5`7d%z%TsIg@Q)D$%>F4q!_a{EB70h*VV4aK%>phg@{{Xr->MrX^!AZQ4?|0Yk
zu@vJIWszlKg}*Kbu*YTOx+|W}RW`UdTIg$52&<`NPglgqoc6aTyaGKCns(@3S7C!)
zAao|4jzsxZ2bvR6{499R+hNhq(rep7bVUOGgtXnhmi+eG>;0DrTUH(^U9{8er~GX4
zBAD06qlB!#=(7o34!uyP(KOAZ)BCD^t8=zMvv!jNu7A&zaqC`|pE)9JCVN4&y97X4
zoiAA)nh-n9k(0O5WR-HEbG5Fj-s?)sX^aQS6^y5P?+IIn)B!>LJsTN!dPlj?vi?gS
zy*<bUCQ~OwL85V<W$ZV5wQ~9sh9Y3%?$`f-k~|0`Hy7b$kS4E~*InH1t3J1VwA;5K
zIXIQiA?MsK?+%p*5zmO$hjI68jL}+eE;TSIb}QQV&csrm$X#!1V9Ie~4cIzAWlT_z
zM;&>+AS<JdBcJHeCd|Mo%B^NRyBnSUgoZ>`3YuSr4B(_YlEnX%_+SU*|3DqRBH*fM
zGF`G+y5>rGR!z2yiBh1+i%BF>%Y$!N$5kRV*VP7JM%5_l4+6mjwvcwiv2xdUH+VN|
z-2+zA;CYR?Sfd*DYYr3lV@{I@t*m`U@vTiikNq8=oILdcnj4PCP6XJl)baSF@BS`s
z-=Gso=z$DMNyazmmFOA#bstj5URF_2af!JwlQE2FC{ALEw%3dHHwHPeEC8guXmkMb
zC0myDCPfZ{N7MZy4L9p?pNb+wxQu<#Y=Jwm6$NHG`eHHPS_B_t&mX5G=By1;zh<%U
z+rH^@r7Fo47lLUMC%WUu<4Vh}N`dB~>`}d-*A>Z+Sg`|;jgzey`j6n8<YmuN@UXNZ
z35wqMkrdMss0rPHcRlCBN*-L*S28_i0mO$wcK@;&=|xbNMK;2GIROFsWvsF6&U*-<
z0u76ganPUjgovF8d%n(k!Uw27c9bM3k`D3ajuXStPgqm*NG3b7nJj<KhA$%7c_z&)
z-=WOVeFJ-1o>0+b64<wQ2KNIsW7G)I!Rf8X8Rml>S=h;GA2LbR{`2>&!BcOh5L)D9
zV#@8R-qZ0iQL(S9MbxIeYZ}`U!aBx2^8TG^bmTO=;774HPVK%VODV={sbdb9bmtj+
zGd;~}YK*8;&Xa-Zr8tYP_y?BT=}AYMmwC1{P1f%r2yEErO8Nh(ZGL(b!?ybEew?>v
zdhpWyT<|tL$8Nf<CLYk3cOKgnS)q#Pnq=9mhE4{h`w79d?U<q?)PHs_EE7TouLvXw
z-2D+!b+&d-eWSgnzA^po#zdKw-&S9yMacN=sCSSGXZ}!Vt*e_|_-7u#O$~SLWydSP
zjt(03di@QY-q!q#XEM%{=EPnq2g}t-D?~3C_M?+6WC3J7Hg-ybTe30h%_n!VO(-p|
zkwD9Bq)l;mNY@$sP-TUq7)XNM;Bs&^rK-P$hg@IF^e0MwA)PkhEcH<mkIM$-W?+vf
zDi0`d?Gh|F2}Se+#)k`vnHd=mL&r}gdTlhs+>1*6Z~eIZQC5EuEH=`SZ)3)x-%~gW
zXTaCWXxnYdb2X^gkXWfIGx&MX&c0|v&<CDj8S1~UkdH<wHDjxU-alp0lE^pG;rIY;
zHli{k>1Ynw6KQt-e{>M>I~W%=0R#x{Y9`W`QZQuD2+XG(8ocy_Odo>PZ#YO-Ee(Q9
zrxPDa@eTQ|f}^2hY|>f_=kDU<BW9Q_fr@{%C#W+3k_Jvoaa;%ua(zTquZ?Gi>7rVi
zRx?Q7o_%0@ZgNYzDu&1Jf6)bIk?_B{B-fX_hBRPm`REEbD5Ws-em8%@1IVBe_?q#A
zR487YSX*1eW|L$X`Fqrp3VX>L_;6<y9NN99qOn8W);Vo2lo8x1V8!A@1B>I$6T-ko
zVZOn5?V|pA7qRP4*(2C>0b1e18OvHJeIzp9@qDhYW;@L48u*%ETAd7I?j7(T(v&za
zd;6Th^nZyr(vPsOO9pZcB~g!aG<QMkfru0I2twp>xr7zFr}s`g>}jHqqq@7%;HhVE
zV~T~a_Pq$f_mAe;>i`-0x&egE8dqfZbm2jAX1DbiOBS1g<V-)z2iJ7<BEn{nw>{`r
zrO&(_D>;SC>RbcZVx&>^>doZb!lmCT!dU<TK5WiUdW864NVOF>DDr#kK2572l$%yN
zUl5l)a#LID=`Q93oza*rC24YzD(}kc`O$g}1I-bC<i7r!WA6Ni-f8K?>ePx8nfa-f
zz-bNBefso{WhBv05R7iQr7l=w@T0xQsa|M$BAJg)0!QSHij!u)F0qkc?R}mDOXb$J
zlGC-S{CUFu{)InRSR0}~7BEXV^5;fGe}pWRNRXEO0z+W>7r&0{=*@MbH@Pk4LiYlw
zPrO-Ue@@m;ynD0nBsWbDG=6$9d4}50SGehgh&Ggm2|GDrVAO%Rc*BCYts^okDF;}I
zUuu}eo51T<h+#T`#z#nJeh)UPuy1?z&cv^Zzt=S*1}~Ub^k6|Y7v|7^kMM)mvwqyX
zDFe{)zp-pa6Fj5oen-vzBlk3J_^UgvIyp;-7+CdG<ecwt?&@%AsTdB_vxG#!$hy|9
z+e{?(t?Z%Q6ZHz|@TU^J$FlyhOAm$IwKR7A^rLx~tnBP4{pHbE;u46Z@_Cb%lH7_d
z@r2r~1P-Yq#Q@;W4`;XpGKq0tD31o#eE%?_zxZ0x%3Xh4=LDoEg)n61=rPLSMGkyW
z4ZBj7aICoJT@6Y44oYoElplWKR=mZw>n{nt1VB%AdI`94Lr$#Mm8FbRE|GKd@%?a7
z5vKO^+TrZ}uN!MEMUGfL$=|5-C)_*+*3;KAggAaCunj2C(3+cXNTtlNIKuGpgN6dH
zcBO9>e=ECB-;8Zb-dj-yi>6X>nz%h=1EM3NAHp9-?-bk_OtbBiKAvyC(qIyj*p^tE
zn=$98-2^Ab$Mb0Km7UPE`)b&Cz|u1}H(o_9QJ;{PF3*2Fa$4rRUcB2lUF2FkQ>j%_
z3W>UC-vmX`)J)x1p+q;j8+mKO^qkEtE;BvZ-;N(3bTDVTH56Q6(Pz9OV1LqiprfY9
zg#9C1m}mF9b&+E+6x+9ZhIouR*~vCtOu9F6IW{(jTbO3u7nnSSOQ}U3uGW4X7;@ZN
z9C1$(`P^~RZooiSCGT09fNPs&WsfRb9j0qry~(a9{rkP*u;e3z+vLlu)9*^It2P%o
zrZX9%8ehk6e9!(Y+_z@Z26@ruBb7@vEce2d7TZWeJ{}IUsmJN(rAeTdEt!<pZ4o4i
zIH5G%F6K_@noc&kBsb%T$<uK^dg;;q)|h-9@u$gBJ)m`ZiKs}(55DR!Bd!0|uMQ^a
zhs{Q6x>8<Na~(U;Ek3W<7pgazd@7J#O{V#Xadpy%43nJ%?j~{-j*h*@;R!1~D2D+7
zTWYletF>jTE0yX-Dp^Wfw(Djc4-B0QbdqSZgo{2`*YsD8SkmOBYjv#YPIP_^nD5D>
zGg=Q|C^9ta5%>xNVe!SB-J6e{E04<W&K;^NODURiIKf9FvsMG3<pV3rl?&y7Dw_?d
zfVVcvlHluRyKfh<Jg#QjG7o-^<AeVUdhXO%_UoyVah?VEgQyddbaJ{ARWFUDn%qHB
zxv*I)vx^Jy5ezA>HRLff+gme5Y#l?JQf6jmi+vzs>CYF4FRB92IQL`gf4ChBP6nUe
z-2Kj8f#YD}qXmVsTU7f}VlE;z3XQd?@^?OF!crn^)zA%`GM9lhpdsG|u(0CiO;+p#
zn0BaJ^?k~5lfS&4p-+R}IT4!q%pJEB!gB?tv$cv4eW)@z*|ffFeyEB(dv!S2#<`6Q
znwSpOTBWZaiszrwKSR+i-rr-v-N?J+k_AHAA}SVLo<ZDMXs3uzT6aua4;6Kabf*B}
z^w0OjFIejRP34kh1(@nm)_*@mrF?$&^$@w10{PL4y-WW>_nXv;f7ee1yfjYHX#_vE
z*-Xn<!~D@o?6-mU+dNYx;4?xO8B$jJk#AuH#E`l==9b@q5|J{R(I~mzGN};1hsRO@
z<WNE71*YTJSY8*4YCP!Vph9T<8C$aa%tu+zHf#qD#h?Q@<<I~<_dZ`roBA$NtNwcX
z8l#8b{}T7^xD*R?&v|SkO1;MS2BhivK-D8t1&32jkdjl2!0cPgpCH(@Q(*O;>Sr#8
z?yyX@e{QUsONA_ET0lpa=MVLaA|C0ia-(>eaX8y2RJMu2q|k~|UbYpDv;f|*!`2qh
z77MivqjVng<R!mDIPaBHEh#;V$yKQTXNoBj37ct-ay|Z;3#8S}jP`-pXZiCUgJ9{j
znvyZ3I%RJJW@ts~MM93J-StM~r{+rHG;Skyp_oL$<hk7xWl$OF)N8ec?Z{;aD653%
zHrx-te)%Pct2N{tA<TQnJDAwZO0z!rrGJH$Oefx#c(UVYQL&)!@?)_oA09Lr$tzys
zwbnVZBwg%1s0%s6y^ePkI6}C&d+{-yk8@D~sai>l92@6m?-s&JN44Xl-+nkUJ;(c+
zcb|PkWJqjAI#AzbS5!OBvd5WkMgR3y@REe2s*%s=3j%th=}BH4e_A#zGT9$G;E!Sz
zQ_p+yECX^5PnYApjy<1d-w`zC61YXx%4Mr`X^U&z36qd1I^mq1NHYbw(PxwEHD{&5
zH}X*PU%YYPLHs8GJGBLuuxGl0+PM1ObZ0bQ`fTn4KB&8@8Xt;qoAzXzXkD=wZ&H`n
z`gFzPaQA<+evrLKUs`%Q+?i$6b1q+Z(1^7?p@uMniNf)9M|>3N$sGd$aPm_!_6^HX
zz|Q-7d?$Q+;cPekw85tu>APYT0jP=Bzn_8?Xd_HcPAGvv24WrCm=Y3BUoHj1%&KE6
zzvg9o^vLz{&4~bYPf>xf80)o-@_q(?@|N9euAL1*_>Hdc@O<UHmSuwAvSVP+v_4Fu
zLF??|u3!rujjJx$YgQ-3;`c{@!V7$pBB}S_*!vL7UE@eJE}7wHmF6{k7_r_vAXHy(
z3Brv#MaABJ_3xAVTR7s1q^T!Apv<;O)28f}#w|qt!7cp{(T}Od(S`x@hBuCDmGw-4
zVOQ?c_6*(a2W3q6FEv3G>9FO-Ss3v4VIEfnq@TBumVv#ZN5r3r8Ax$$354PG8dZ86
z66Z=g>a<A2+g>I(*T{SwkDAu{k<HwF>KQqgz$5W)V)dy%fKv@XkD6~n$%4H2CG{bN
z*!D+hqbKfYpTlA0L^QBDX9w@#23R$rG(dK)P8>|%q%mktiWmc>AzfJS<WG4V9pIId
z1$rRWCX3V2aU=JkkMSN15`}|9-)ZkQJ{Se?PE0@1;gH8hLq5v*q~nTDkDC9AGXt&m
z<x>=fUM?7&l|~tKufldITg-w$yQF;~{%dZ-ceYo2+u8Z^KTHFp^PMN_*dQ({0LJ(g
zue(lO#VHEOn^8+Z?P%BOTCVn3UV5JwpL)@%4|v2UD^>d#*sWpj)JyPoEs`@Bd}x~m
zSzp*12(B<S2IwjFPZ1GJbYS@G40DVfw=x>VkM8DqVo*7$-Iahr0LYnQB{_6;IC3H~
zKI1E>6g@yeMRdq!IrHbtZ03Sg+0XBQ-!4ox#?^BED<3C}eh{92F$<LR=|j%BTNxKF
z!bz!hkzhKI!Qi)<J|v+4sD;0FQNs$cFo=aY=)ku5^xXSXmQ20w`|_T<3OwA}zR2<8
zj|30I2OF}!F3Jk0k*9_6@SPty%#Yh|oV~0PUP4G=?N(euU7t3=TL3=Hg!vNTIZhjR
z7;t&(+WPvM?0<>{!pD*twSpM?K?JA`$SGlcSO*tf)I>_rRxtH6k3|<^*C^VBNLN9w
z=}bvh+Y<@lR8<4BO#FR(w<!jCKz8Gq|2{}#FBLEr*eAMe`sAm+!x8N0VLxi?alaCE
zxv!!H+7jV*4wu@Kj5hKZ0YA#Vs-KzUg8GQEW>lvALs7hw2QjBzM|ul;eu0QJM34);
zM&X6pe(R1@vsK=bnhJ-3)9J?!>!k!kGd8B}Q?iyk15`0pF61X2oys4$?$rT#Ba(XA
z`|7w)JAl_0YaY=NTHaPxaA5g0v$O<y);OZCLc9K0Gkv4d80Aw#J$Ha<{b-Yq->+)%
zSd>4jh5c5zbH9^p)e!p9j-RmVG^rA3<dJ%&=D0dEiny(ch3a(|{zk$JL&TOymhgv`
zlyc9L>fHqe0WE9Ie`7WZTxl-dJL&BeNx$=fSJC@UZVYiV{ot53k*=Y<lU1BLq9$Rr
z@5_uWwzgu3hGT1~`Bwf2Xt`#k8w9=BB|ilrzubEP0J2@F*laAyCc;oDb^Sc!9MYjZ
z=~8{$woLR1=ObYJ6?NxY=r0pbXxy4{+Foxd0(-y0&o8TA?N*?S(wh6ot=JAF3Shr9
zSezJ}<<>dOrqTX>WT16~QJMiA=P!EJVBR2tzxDhHZR^%=g44A?p*1mj`0gLQyS5|N
zEs*_IK_kMzm-6ywU?IMn23HFbtr9LP{>g5kNyMD(=`iuc&FdqO`<|_;?c%SOqr{^=
zOqf!X`_?>~a%c5Xl~CwNt>TM?VNpc`%BTFclZ&$!8@~%W=vIOupb}l%66-Td5L5ah
zPdnQ6nSyjOlf|%80x%Rd5JyPmqt4GA6Z=0-8L%bp0D{Gt!_D5gDeNl78c`t*7lAPF
zC%b?H<{xbVmQI`2-w${OkO?l<T@a-=DgDK9{rApzTI`l!K-4xO4`1NGBvXR%aiQGI
z^<pidkj_KeU~jJj7Ap5?*xiLX+m_phM0$(&O+;losFu9^<@Vi)Ms#F=B|;2$Y%KQ`
zrAIp)7{WmdaM$XGQN=CuWgfGDdQ{4xT%cRV;PTIkQRofK2pM(fr`bd(K@;UfW1Q4>
zdVW)fLI?!t)koP*ba`%fR2;5iME412ZQto5`Lln~HHdrbG45yBO^3Qc5ahwYdm;LX
zoBOL8&kVxJ*Tu6VQ#l34?N^_+15cvd3FtqG_D|!z!6WKzS)?7Kk&X%0B#P@yudRO?
zkvy&rj!#7?W#{n$z@P|0!>7v`WI1dG*BZPk`Yq4tL90g{k>-$zE?}k*#~n}m!PAO^
zDW=rByY5cxbC<k|kF?6TfGbIqpXRn6t|i=<@PqC!3_3=;sI!JDq^8I|8aaF3b*jUI
zU>YiH`zz;HD|d!D?w}vY7Ykp*gK0t=ETE({S~@7%_D1s{rd#tY=JNdymM)E?ZYy?D
zL08697z_ix9YEaANR05>0=_l)?v8<kG!NR?rtQqQ18aGq&iq&F`96aUp2EE85+Erb
zIWybzWz>!6-d%no%A>9NW=&0I`VSRSNa?Q{Kl^77yA3o3)+!JSwmsWEr<(df-Y)O&
zOwb#scHac?D9pB=>#vN6$%u;(ak(8AV%csa8&sHa`Ha_6n82}Mj-j2<t$NOD%;}v)
zL~t^0yFDVP%*)0cweQrS3=S^^?z9&w`+b}KH}3{b20iAhR(Qzeq;L52>58`XbLF?B
zRC}Ws_P)Cl`rM9=`lAn`6Vw3^F<Apz?BmeHIGGKa5y~&hs#MN)wOwby0BhmX@0Rzu
zh*^)1@l4Y&Bg1t!&m}W~?mN3vO^(CDWuXUfw%vD<P}@;Xe0+n2=fU#{PN*=EK{q|a
z{CuLY!wegq0EF7N+r1u;G)VO*tCSNC-eSBK{i*8S=W(R94jQlQ?jfsbdEEFQ{!<i=
zI@Senk75{feOg+9nCs-?+#8r5hxjDhZ-61jcxX#bmx9$wQx(>_r~dyqkiifHxTB<g
zT>FV@{mRL~)!E&CarS6lwxXQ84!+*$*`NLW%3t{s+#PIhU%2(14iH+PzYHcUHZeDT
z1v+-~BbAX59CV{hVO&sE!k&mmVp03)-k_EfESFT7kYQ6mztK`O4@nRjsR_yJ0cMW9
zT<lxovR7e(OFm~La<A%JSj2|PMBP>u(kT9~<CpmXDO@RreKV@(0Q#rxAICYFG|6gs
zYH<b~{Q<Ub7tVfb#b;$q)llxDWvg@N;9^geDsq>ZF-)5X2_rIcY!A>p5VH$n{y7ik
zsBrZn$10$Ae>~1#$Lt?b_UKa3Ma>dz&Fj{&mL^VMb|z#FS_JZ+L(ztc<}2Hb{OgnH
zMPIv^u!t2l)B-z_0%{COuEPzukM*MgzngBYx-=)XF-I6w2}xLqh>xQb2gkHx!C`mb
z+i+!-&vbVKr`zeC*0ViMEq}J(C_mcMX=6Zr5=fgZq+ahsVLr}p5pV@r4t?i)AFfM7
z$b|j1V?w(UPHzYsYbgyDnp>^a?ZRHbDg#imK>y*+jL{=Zbw%qsQ~jZD0=@9jHAJ7<
zC<^RW+iNnLS%tBOz-_7{Yo;0@_dQlG7WA5|O(VuubZ;Nd<P+~)!j>f9u+-}eQ$ozt
z@{SK)F$Q0SdT`t0tf}Of9x7YYtl!$l7g8hqfq{p>wFc%BaBj>!=Okpl8D3Z@bU*Cq
zdkO{Jor8Lw$~axX+2@QlHul!3p>xUYHST$SD|bjPm;+bCJ0FU*Qm91FvYl>YmI#kX
zrkd(9x^M6R85TwIPX?kX3xyETS95rg_PLO`Dq2{@wgwv&uEqmjg1#y6>s|fJ0MhkP
zyOl-9$S>JT3U4mjRT|C5uN5D_5mj*M4>AZe%k3@&?(C<)0TsT?^!Q}a+NA=P%8#fs
z5Ks0A=3H5mJ>zy?t_**9riacViL-3UYUE(a@lfO%^+dIPE2sRqDDE$gEFic=g2eDu
zBi7$X_H1vEfA0CJhb7R3a_INHFR6^nDnX1;m6xq9*Jqg2-IXYx^tuNeTtyWfdS8*Y
zepboC>ZTXwgRdR=5M+*f8Q2H32A3u)P6FKZXoAkoB{vx384ak_$rm3w{4)~O*jOdz
zuFNhj;Rmq<Y(A6`+?V<vgcvV_eQwfuPaP{s>{WoKe{5?y^6;0&DPS#IjlZR4TfS#@
zE8=ySAMM1*X07Yl(3M}|*I&{Mc15^NV9fH|&=!BB;XiW$q4?u-_8*$@VsC(O5m#sM
zSGyF&{lMK|S7mTlN)x(8bT9J^-Vh>zU!3k*ugN|-LNo{j?08JH5IN=N+wg@3AA-%|
z2<Lx8*jo@S3G@Ko9xLFN7X7Xm&e27jnMZ;Dy4{jNLodW_KAWN2csq`JTWm<R@D>$F
zm~O6g;7y7}>@fjE(IU4yL3}Rmecz=-4(Y|&)}lKP)aGHnY73qe2U}>?M0yRyo5)a7
z=cjkVs?(?-<g3yBW5rO6S$LWeMkc^=3Hx;cb&xphlGg32_Dp>j^YFWmZir0xsHQ)W
zcr>3Ne+u#Rz<NHaeg47AfxZfA4J!Sdey@)^9RO3NUjN94;_0Vwca{w;rdADjgDLoU
zV-$qEoD-oBWpghwv{imlKu&5@_GbL|?-fav<mdGLqHulzJ7EYHJsOy1VK#=@a-Yk|
zLMDc`-yBsCL_~SP6PR4j|N6*nGR((y326Lq3bgR6e6E2$3+TqC1E1(W*zJjWA=2N=
zR};i=!*2wzYo%3pDXa%HVsJV?7KiKjiwaWu`@K5Caxwq>mW>vm3!L}l8oU$UK(R75
z;KOz*0h^lM`2y97w~nYGa=@hT0zE=ZZ(QFBzvM;FXOK&;3+FY8s<k7AWM(kK^n`Ar
z6{sTAALKS26nxEV@`==FI9pJ{?Cc(^U>n5pF6DBHA+qm(#-OwY6amhu3RE<2$5&IM
zb}+8~k)eH|2Wd?M&{U9+Fy@pL;y*{oEYHYJXF|@L;B%T2`PKLUc(Sr{7P!R5^gNbE
z>IiA1rp`n_o6}OWViYXSf{|9q5i|Y`k5;UlMNIq#xA-99V<2aJSv2DH`J;PWwa}Kt
z*4Z7B=QOdw1ngrhz!4{><3zdRC6df|G`L%yP<fbrv1%#d8HV7%V&@Uk$pV(-C{h>@
zsZRY5QxCZ1zo3*>@>XAvaI838=_f#Z<ug-_^VRJiPl1Q<>^BoA7=CTGdwnMpFhGik
zC%+wUKmK!zz-@%G!#Cab3|MA##i{p{k37{|o8N9~zLl1ADqSx90b%sr@ce=P7iIMU
zSBBVI#z`i7n&<?c0=LiUwTpL5E9!sv=n#3jsjh;L2uC-V1X`C8g6G3kWU|kek+@Q|
zl*@WGDdidQm1otIcTsp)P$pYFmwR8r9W*K!G^`I=o;?2}yg+|M%nW>76>!YK)-$tE
z-|hEYZ@{J|sDAv{9QGIH?QHvFS(J%LXNs73BEYIa>K*A3FNJ+4sY8DU`e2*4<~PB?
z;Pit>*c+=FC{|O!f=2yDVU32O&K7>3!5DV?o1vBOOESc#R7!K7O$xJruEw_)_mPHD
z_uZN_x*P_W4JTv0@I$`&Xw^|Cs#&vjPpZN{SpvoAyXVfALu49pK623y!xOgs9gS!)
zzEY3Y8nf0*IfiH36r;$W%^oAkt240<E-svcSOZ+k@~>s!ItG4Ro%`o*ARmXm25N&L
z%@55ux@xXc@5_@Ly|o9(Pjr?=MHj6Do%9ye&E;WdGri~IoDtmZOZ2c}bv}|;HqQ>b
zyCY>dc7NfzS(Bm9gS+xZ;<#~FR!s3pu##V{Lf}_?E{@sK{t_K7wa9tP`v7JlzCg}A
zc0}YUAIOezJys3AF1bs>`}<{EU5lY$P0K}cNddm0Ix;-XeXB*zb|zg`S3kZn^5;;s
z9qi08b2)P=knT)ct<d_-H~=7x^a5T)U?UhV8jZe!s~`l`xZ|38J$x*HbK6Spt^LL`
znltxF0}m6D-^jh7L>KZNtfd=RAhhw6E_VR>=H&S=O$$nc7B4Q$aSc%?S(*Iqdxi&n
zz|k$xvcBtCc97j7!vsxWO%B~|d7_cr!wLEj9fz3Db4uFm<Q%YQPvhXBrd}PvwP3!S
zkyf;79iD#axp@OjkI2!!B$IignZAmQn;f*QMNayYM?p{U>!0ild`oCfs26D?kT_}H
z)u~xaJV;;Hp4lvTPmwimR3T}HJxmKasjzTU?_)9eXGK9$yP5q%_h=}v3rlJ#^>g@{
z&jjEW3#K5bzx~3WK8}mGQxvIgb^>0nDB>fFwK!=dyJh41pT8ChL8-2cx@)>0*IGEi
zZ$6q1iBJQ(w3yj^W~Ja{le}V5nmZPd4LJgwNc`acS5Uy!>RjBKrd4#5N4B+0*(HnU
z8W<}TjMRp0r#<UA+cyL9{Vl|F?(UtKSiV=hx(`n{imItLG9TI<Q2HDdhag#0!8khs
z@J|5Gx#I^I)?*ChqKNUvXq8ewO;?Sj+#parpK7#CD&zUOb)<XYyGXEvYM@Kup6#)-
z+dywjlU0O(&j-J#k4-CHMFj5NUnb@eu500NkiKNsW@1B^2ow#AbC41zjV0e$-rfE4
zh$PN&L~RKBqh3<9pvUHXjHXEMDKb2KJrKt^E&{E@-1=GiMcW2~*H|W1$;<Mva$r*J
za8LhkduU%sG8#lm(*bu&2XRuQ4O(A7dhW%@zKgp192z3W9kT(3W{#F^qz6k`&o1m1
z;Fc)q-F+m&|LYI{(?h$<3f6nICh0=l$K-CkB`Aq2JF(JVNkz-Jl!zr4dNejQncw>d
z2#(6de<a+%p*j?Owgc#96LX-9wvn;rN>+^p<f^B|mKJky>p9}rt0dt3_#HjZKG%}q
z%Y|O=3?A>H0HSn|SN1N52m6eST6hvKg$-#j@2V3_8_v#_HjKWDEH@tu&Srhm2~wJ=
zW~=*?ZxAyqzJF;L|3h06`dXP70mb&_N*+V~Dp?NT2pF5K)e&K(d(PWo1?&I$Onf4D
zc6wrWGE-Igiy~VeU69GQTVMCRJpMK9f=addjY}F0y|29n&+yuKliwgB^5UfJr^sIY
zS$65bI}xM48&6N}yJ3+3h4kq0Nx#R+9Nw1AmcyYeR23hVVWdD{qoeH!;r5UP;hFXA
z?e4kI=0K~hh1FW8EelMKlau%P=tt8eF#2o)?!|lo>Iy1eq7-W?tErtT86v=&=cLo5
znpX(1;Sz=ukF~mC%sEsygA5dz?C{!XvY(j^HMb?BmaAz3#=5&{92w>s=iJ=agG1cr
ze5l_Dn8<rqGX|$CvQlGEUvkW$trQ*(DgM%h{0UVD!y04^s5jg|cgKVC6wqmDX-1*o
zu-c*L&Z+lsBTva3C|mi$Tig8hv?)a3>RB+iv0XqUqNDfI!^0zRdWy3eQYaSV=O}wm
z;rC%o2cvucxZHz(Lw<JfTWv2Kns7%gJbyN8n<?=&&o7oDDz0<k#lJ|%2Pm7hNR)ED
zM;I^iT%mq9JuzV?b@3k>=>O1sgr#;D*pBlUSbw9JJZ~`SU$_`EkY(7l7R~%$;nHWL
zf~(VvVW2k(S_rJQA6@|$ex4ZC^A!{UFHsv$0zLR%k>GkNSUJxDOee=@fY;$H6?fu`
zm7mCqb69!T^RYuY5{4bVf|-u>bCl$~kC?Rvl{K_GzX>kt)JNV@Coj7iN`juvV=$G8
zJ%#*scrUCZffL?21j8hN{ykgWBb}V40Z-vAGq4Tw9?{o03^PJbBE~1%b>ke<bM_%~
z`7?yk{_EeLJ$ojz@#G!t52MQX;X);XVP%brwPgvoP7P$9N$}n?#awat1ObIo_DZHx
ziksa&ng^~4l-b(%hUS}CSR&+z7A+<{X_lr*hUCU(W{QNV2CbI~`f0~4b&B5JWHlyd
zK5Xd&cV@T{5Es@u`_muJjK`Xtg3aW87@B`Sx;~lFzTrJ21y{qCwwZ9cwT}IAdYIr4
zXfpU>{rq}tCqCEaK(55E!ly$w@&Mq;;rjaendnQrSdZbIME)W#@SH}5S#YW4-~$vb
z<{@WFzbt#I?kQXCw~amzi1ANkF?JTfzs=3dCuu`on8)nN`gT4BW@O)&<Qw{t?UT1I
zh0g4P30mT`7z}<1o#oZ(?(SuTJWxP-{NnfhWnCR%6DM?E(@Kx;Mlv#r)Nh;r;ix(A
zi82|p=Bk~7<b)ZrjE-}fvmT$DPe0&`%lUFPHIlx&x4WZb9X%<3W`Li-zSK0tSQ1%j
zv^bc<YU}0T1TPB|#?`?GL+`7wQ5ey8+d+c8ZK<zQJ!}xRF?hDNtI%Qnt-|B&fs@-S
zA5$Er^hq9EQVbe-)AKy-wB!CC{*51_I``DR@v6lKFzugjViOqj4MJbB^64=;zJ)Ae
znNO#PqEQ+X9;2>vPFoZK@*3vGm`~+8WXl+g6YtS~i0_{lU&B{OQQPc>cVY8R`rZAr
zXU|4Ct@=ByCOuI+&MST1Ves@x8}$vV^f<B26%))dZK+X7fTRP=d843ss;|#RaD)C&
zONHtNJl75OooWt_!&lEw>+A7!s{!>nX|89a{fnnmSSQB$>+d=rR_e?<OE=zqGSR9W
zy0<6D2>2w!f8x!Nd{jHiI!}q3f_HK54HPNYX&waPqW5+53dX$xtjvopeWT*Kyj0EG
zV?PAbko<Wr_aP&6?y9ep&?<17d9c6REIlh^(sMs_-dE}|lOo@qL{B;xISYKQxMaim
zh*IWwXP4?}ot1$48=rF@`A`ImC7X3P)K_^2bbM?(eDrt_@tKbL`nlwlvRluBUdZC@
zc?~<)=P;Y`+E`qXg|)SH;rGF<4VXvCyao?Y-^9D$e&zegt#ya><)Lmaxn3xd)erZa
zB()?Wav#`hgfn(H^_!reJ`=8?;l+xcIz5rnSSBcc?^K{YF?+VzJ&TKBF$sw>cx+WZ
z-y(oUDRkt}|MD&1Z0^H@=x_fNyd`FgiluW-_D7xE98D>Am!#o|s!K>n@Pi#z9@hRY
zE&a+cXxbS*s}-!hRIXG`-rf6zuLul`$&b6@zMeBxU0H!wRq`keN#{Dsd|sq)xbs3y
zrTN&A_}ijn(`2!Tws}s&>v<<(Dd$@S)BEpk;DVth5HiX34`^a5)es<LR^J=<nb3>X
zljvO?3EO`C?RtU0jdn;fP{?qMO^ozI!EykthONXUUk7V7UAFvC<C>RM#4+7nc=jG2
zEwiF<7Ri^ozbdUO_h7F*si4DgB?eI%Ph;Kkgkb=Yghu10-w5wB<NYHyG**kJ<Ks?e
z`@_5`0{WE4LyyX+>&4M4+msko5Rria@@#jUVVALfxQ^qIhEw3}-@ji4<hu)FH>d<^
zPYrNPr?~h0YN)f?4{{VOI1H;lwHL%MEW~lD7hbcS_X*pu3go=G51)+Ujgwlor_FO0
zPF${s?k8S}iO<$gF-86iBL&sD&-_+a<=&T_1dg7A_`pX}%6sl!mc-54;00I7bH2q&
ze4YJ8Fz<@N`G(A2N6(#j23IR9et%6{%J|sW-CG>Gl8+ueIt8=Sgf~^(f6Y~?Z_RJF
z7CP9ScKJ1^G-oy|ar*RWfy;sU(R<R)0C`w3WgGe(%#Vg$6?1|ZKZsFUuOvYExar_Q
zmBbL`dk+dfQ!DXN8U8w9Rn>4lDYNPk5BBl*hFG)`V|S+`hQ<Hkyabd9h?%ex4X@xB
zsp#v|PjbJN4;1jz2hO|!iquWS>0f1DHD-MG+)ydPbMz+{RAQeEC4qNYwKMO-r>&5n
z7X*hHrPGVaze<D^>Kj$};gOuHQ%k%ldT5eNp%5Gy@?QMIq6uEPQn(zed<r;f`dgWs
zU$GwCAhp7DS&6EzRw}Bszo{%-YfdQ~J}bO>_w?(^#iIzdbDfss;fi#uo7Z0F8y9K_
z`l(m`^umiMj`D976|B>$BW3Il!cM;W-(Iwr>3rN}8qTFYddfOZxZ!qT`uAe>*=v17
z-K_6_5EvLrHV>19)z-85iS=&B+yDN|3XAtx2}*R2GP8oD{nh7@9j8{z&AH0$_3aw|
z2Ph~fY(_%j_KA?b-rlI)-tbs)R8u64a3OWAohgK@yW6_;`3w4^^z`pI3Kko9srGtX
zQk;u^PHRPyGx-)5fz2q-`6-O`3-NN*<3FK(qyFk?5oB~<lwZ=b=NBv4us*G8X~FP*
z`(K?s+~GmY7aPkqVFsA3<A#;blb^MzP>PFoAi55t0I6Hxs)h*_5E2p!V=^_F*7jGm
zTqrIs{_v10>aRT`d<fZjjhE|8%GY-EsMnC&ThwhO6t#(W|Ni|dCu^t85rOohcv?dh
z8WD5l!rNgQ|Ld&=J6?k7=<(z0Kfpz^;Ir%f!6%~#?Lg3EzpsZaaS-3kkK;xgVM`x_
z<IHSrZ6isK^K9G>)srPb?<eTrZkZb5rzqV~c%d}G5=pK#MiCcm1A>bo3GgV>$u~6A
z0U|)JGz{u(vZ(3abm8y6gtyI7*C@n5ikT)gMTUqL{^QwD%LIvSoJ>X3rCl9tbFT+z
zhqFq4$S;R&uZ9(5b(&ViEWNWvdUlI2Q7t_$)NH}L<`%8$;!<~ax81zleM$W}Gl4jg
zB=(&ng-XIVFgEhbu)&@V6d69v$GNql>g&hE7`?KKSB6LZT+cG0<w!m~ym-u6C?F*E
z{U0xY<{TKl^<$;%`<NW<DZc+bp*INW;r)dPtPc9b5z9<obK8h>T8qP!F$8staQzY&
z9r1KrjG76q;<I@-H97eXiFXq=pClyqTm$s%@G`%hD8!|W_)iYkl{wJRyAU>=qbQ=N
z&(SqAmvrwdA!bcAeafY@+S{wqr(Q>Viii<1_>I?I4f}f5u)=G?L0N#a{i_oHQsGxH
zUYp>7=wx%oTjFX3SRD$G^czl&)Udb%-XA)bvJE5YhD{|;XaWDEa}J#ByF$y`n|6D!
z&x^m|@noNz?86ZeP>hV>Zg&6u)Wq9|7L>{8FDJ);LC=w2EF;qQL~(O-bK9vef|!`1
zNajcRI*s|1A9QZfryw_@*3E1NHWK;~N%k<c-R6RVL+7_m5$F%wvlv-$M}=m)t#8wA
zZ>u*tx?fFRd=I_p&Bn&Et&8O=loz5#U>+>zn;!ALZl%kizCU*RC|+FXpsjDdUp~g{
zsBD<e&8Blq->ca$B8t_}!waNoHJ2L7*X5jhY)cv(vhiZNpuQw@GeUQ{l%cag{#5kf
z-oya+I#5;k{qE&<yfC%B{ewh6pKIaPQp!mah;d<R_lRKBB#?ZBS1YviU8~K0ftubz
zy|Z95&n8*s8U7P|doKTxKUQzOMX=v6;(rns_SD{=BimCSrA)z5RhY6hgFBao0NsZF
zqq1*$3jVc>$z2JK)PiH(d*bJxW6cL7#j3U!tT?a7#}d6DohUSfxT~+5b*(3x&CVP~
z$k&>ia5|qU0#<BFM}eKyId)7rV)i;xhPj#PS+Oa`w#&6Bb|T7dn+n{yUw$cvU&*yi
z6rsiN<?*3W8;@>bR&FlcWhVS}=AuIA$En#`kENw0)sz3ue#-FX4h&*b#I&-;C#Jca
z4rBL3dw`=P6?$P6vY#}>uHjePqqoMB2~rN@BFIQ8#kbN)bbGGJ70!LfI|V6m_Z2!9
z-`pg{t@<tjZEFrM|8Vy~5|8kW=O}JEhC^1yGvniHwZ*7CL(O+q7GnI(USK2IPt495
zpEA?YZ5SIH8-0D;(7MM!wNrR|wRU?6%*uRYz65zsKm$yjzR2cL?)jI;_u6*L5x4e+
zR%*ZB>6jNGv1^(|XUlNyvL6#nT<a7%ESMJ*YW~>!20CWbF0!zuH*V5U1(dWkRI!{f
zG(<4fl^~4yf>_xV>^2J5g$h$5iW^Htjur>D`7nWr%I7DEgZl^Ef0BPz&09X{UCX?A
ztn_%RRQuOw9ceC==H(s?r%$fcxg=k0N;@UE9DP<$S8Ai@>|21(_pI5QQ?O(tf0m5k
zfT#V|whE>~l35ys?%+{$N`J1+W<!RjCBkiC<RrRrDuUKxq?gsVxE9Qg=H!q?L!Sa{
zm|D|s2OH*Mgh6Gr8+~<k(g(OCc*-DyzQx;ZJ<!9z>V~l6orEr3GKm*)wl#oEx1|&#
zf}u?uwWpdmVJY<cud)hV3RdHi@48E|U!QRuNi+m#!&dFuQf?C92XRt3{W9oTaqz1j
zu6_KPo#dV0*_UIOr6mXUf7VH;7f5Cp+yap@5dZ$qUrk|aO>v<@+}xS{#jgg|E8LQc
zw$}Vlt|%QY5*iNBibQ<+)=vFS9_}YDH5yIzPs*;^$plfo{{au2;d$Q<zM(OVHxw1R
zotzUkbO=u-;pqYL88^pg6CRQBpSyDiGs+~|{K+TDLcf5;sldyBW<bpB#$Z2Im>3s#
zwvW9**daehaIN>CuyvjO1*q7zM`{GR4m<=SpPABN2TZgR^&Z2B=cp)1_l+{*al~fS
zH`wr;tuNTJ4AC<^_j4kreL|E7(NPEDB9oVU<Gy;7bB?~U`bK+0{Qo#Z<MzES%{l)n
zVdzu46@?7bH+pxJgEwH6<#eY_skjh4h%?fOV{7zQk+ne-{PmYZ=iW_t2leFdxpkP?
zG+x?$*?NL&5av*v3!7U%XhBnPOXe+O9SXfd;M%Zqb8-9tnQlBi@jFrpf~4{3X-CnE
zcTm#=tbSs%Arv{H@lnMqI%#%{7TJ8xkWJi$VBdUp##S*bI34z=S2PqO-fP2#Us0X$
z5xGyQgZ#}QeAy-b@+^X$4Px%;xFF7i?L@!;*lt?mW_iVeOO9kyl^gx6d!SPY4*A(C
zw81_*8-G86SWJCFO<%Lo)!9qxl#Iq+#l2H1iEd=JrM&njSRj7cH=Xv}{F;(7W9I45
zbb8OX{sYrDoKJ<ftb!=O@)`nFbHA*?(8uKx;K{4n4!HBu6e?FjL@}>m@y};bWn<n$
z55a!`xtASVW;M=L$dp}LXo_^^mx+zo>5H6b*Mm`%SreBpbAHbc>ugqY{jfYk^ns6y
z%k=*TBJ(BuWp3grU?MEMCBSklhSwb^5~2t(aC-=94O3em1DPj^Fbzexv^LO1<w}Sy
zVYFEBS+f{DKIXsI9z4#Al~`P6YICHnOi(H*;lpi`PDdwghp6%^YK8beUZ08jTC8tB
zN5ZfEX1gOS<sZ`2_`Bf`?)!$z5x;YHcNl=Xru#$XpzMbx;O<h%v-<gJH@BOECtf6i
zd}34tlG=h8dq*rOrAhyc1v%#p75}XV&oXgn$r%z~QT6%cGg6w?c!88}1fKaC=B2DP
z{9@tg`#_9@fQ6;S#e4xI1faje$)zC5cOOiR<Hy}kT&IEE78v7=b7VW`i`ozn(*_e6
zPIkHax1Uvdzf^JL*%%ue>v!$Hy4M5zAtA(aH*L8c0MiAfyPDt=@;#}?%L4#wMs7T}
z=}x+(*fYmnqh&R;ZCC?Z`pew5FZ??Y5fgZT1f@0v#BzPX8s~qhihZUWotknFUHCY%
zaeLy)6-r3OaeHzfo-PzMf}h!NZC^$&nSH9AmvfC941<cLzKX_8w{~Upk$&8d-m2Mz
zG{hl7j1@AbTpDLzVpIFr3FYfz^4R_)6;i?!yQP@J2U8(ehX#pSuj#<Z+7@Vr3-gQd
z#Zs>K_PcRzjl*!c=>#S@9L9_b3w*Tj2qPq)u3?|5P7fVsVWMvO<79@d+QMO&V6`$_
zNzxD$oZrBdnt1J)cr*Ze`_*R2xc+w}3&owZVg%J8X8>VbcXn7mw-{GuO-;?ysrF3H
zdt3+2DDS+4(ZrelY@{Si2XTwQXo84t3h{S9pjFCqDb`~G^NN0N34f_-44)Z%hEYff
zT@S1|=UuEsdc1SMb9l4-R3MCc@`28!?vYQL$VCVCHOW>956CKZPU=D8Qd0WmAF7?|
zR=rw<XJ$#zI)W3*18%gOg`2cJd#5cS`DL5x<dUKw_N_d9`<!E1LX-O`55aB6)C=<h
zRy-}q+RC0}5(Fbbp`lxn|HiC+Nswt}PQ3j5LG$b}EqZ7e@q=LIZ+h@$o-K6_?<DES
zTW1Wx<LQVtQ+<q!1YDeu9S{3U?<1wiHq+0r#S|Vm^!QCDG5c5-C8?!^%XmlHqq4Uy
zH?Hpyo9?DDd8mA}4cB!4R&2(|`J&&%A03w4Fl87N_v+INd2N3-89q?ip{SSR^c=+g
z2n1h_zsnq~mrg#(!VBb84sRNqyMJk8rF60C{`)t~vTPR>$gGTJCMP#kLfmBYUfpGh
z^^upIA4YI>v;vU(GG#j+B3t-Rtd_jl3%moTDr`Cxd?f^LF8D;~#EGZ;Ew^#@C)lv;
zeFQezd7A~?Nyj5ZPI?Q$x%;k=!AgHt=-_?k!NEUxo&3m3PH}79<GRSKEX)sVm8W;u
zm^_+><r4O9S^&QNDJ6UZo8SNmpx^x$E;}NTPCTMdaTVsFo#2NnzED;qo&wJqaG%>W
zk$2^oXzuUaiN*Plc!bi1`CvccKgg)fO~NF~)9?uCSEJTRq+^au(44!~JR1M4>)Qe_
z+bmW7@zNQ4@c6q%3>-MKdtdZ`K|`god}V?tGEkcHEoMPsoUes1fo9fBpt<VwpZX8A
z23NV%HeCfSyMBz6-{PEMgCk6{n|5)vT{9mxuHi2@jnAXIWUNGI@0wa|^h;(}+im5@
z?Nsc4?jdC9Pov>qHH;2g755O66=T3dM%9KQg+|z?0lB!n|En3K%w~nfmAGAamt?d*
zbdfkf_&j>fbu=h?n8%0*sVN*Lpo=y0>o)h7-EwIs?WbV8$cw(mS+LkFC@5$qaMwH;
zAH)u`^V#3Zw;oc*9Nmp)_;jb>Vi7q?idBu^LtOwD9r?EH*8z{iKZD`$Pqk;T@Lx8v
zNQg@VoFSq2IW!7#X$ky@kM#~XOeig{l*wXH<Hlylkd`BeKh6H0rDU=z$oFBi0gBFG
z3+ez7k#E&d9tUJHMrW@S=u*;LU2fH4Lt6c$L~)w7d!9y?wYRLxm*ZUqZr_LEi^UB2
zFu2vOL|to+rol@mnskj!v-?_0)2CEVz~9bV|7EM|`fI3ZH_!b$xE~<s=N*wpvP^Gi
ziyeBxIkZX)&iK=rb7qW%A=6+WevsIDU>;3ZxP!XCg)-gly`nmTm;(Upj7L(l5|x!@
z%zyvz9l73hD^Xg7#cnZk(gvZ`w-E{Pf*UDmMrFKTFtG(RtN%j^eZbgjs6yi96YvO3
zDY8PXx^0BT9~qjRIeU?xpH6V#SZy{(&Y@qOc)%W4d2T?wvk<6If)W+h`G*DV*j4xs
z{eft-zne>sG6tp5_@Po?*6THuEIYL2j{P#!(CAhDHv+}Q4hWP363O3g*BDdFh9#k(
zjIABuChuVmnJDhZ9VToL{pk*OdXBP~Gj<cA7fINamMV+HoIY?dUlc~`Wwu`yk`Z)t
zjJhf?gX{qfSJ(sU#E8LJ_!IQXNV;OB{_i8$Ew4-s)$DlXj%b`pMWgI3wYo4c8@|%2
zL|knqCnpz0THqaT@283Fd4PCP2zRzkA9s?SCKhS_{!&#{sPYGLB`uGj-w{r>A7>Dl
zGrsNZ_v92_0YhWGcXz?J9Po@XSO3Rhhp_jKq%<U-4Z%gO|EThEbjL&TE4u-^u<6K_
zQI_2qi6|L3<ecM`NjRN5GxNoa`Jk+`S2A@QS-d(pa<e*BY$S8zwLxF#H&k(M(Y0#M
z&6c-pTO#g))oQee#_+8k^T(%f#Q(Z$5bB28=&)`q4J^RWOkVPmFvUY?dpd}B#BBMA
zGyai*$AMh@?Zqfmv>PuE9Mawk`I&OS_N$>)vcYaYurJQdj(c`%u<<Ecg0eo-{Y0bn
z>XD;vyxA^Wix^B1FV+9)dtiSDzI#4lsnK$45$J~$2V+iH&_BQ?Qo4{xO+<Lw`u>ML
z1}{~Q(?9Od<QGPo*?aHfc3_P<Q%Gu5Bn?Ywh)Q`G4o<hqs&+??iTlW)>%SSFPK=0%
zFnXMEO_%ac`g1!*T{VUrWZR>npg%#!qKlg|gbE+RhBs0$qJANjte<1Ae`a!G!Yw&>
z{Qu+ZzvH?7zyEQ(9uX2rRw`7sj7TWs(Lj+AB4qT;mQ6N2(v}e+WMn3@kR27-p+ebv
zZ$jqx+#f@p)${#%{eI8C9+#eVzu#}Sv(LGnb6$2{uf1-BgOK6QIEu2oVeC$fYsMgg
z^0T+=c6x;>VKVWlb90NSC(U-yswrdcNVvulODvyhoP1EgAVIam_Fub1a3@Dc67}#+
zU=us}pf7^;$i$bw*R9DFR<3dsyOeI8k3l6`2t0o;2iIi*#rzn=k0=$k{|A+0`Mvn~
zt4f7uWSCdatrzCuzlD$d1DAv@`m=80S)5T%-EEy7C;aI_oo$n*u}%({{?b{0^fluA
z{CuPQoZ`Y)v5dU^C=bg-H%uI1-W+sYQB0WX9mMGi%(ES!-@n}9wbpDj%g!>584n_c
z{aO^*tgf1;=Q;@lsXKU;lhg`D)rO33B7&>D6BzBEluT~qdsrS`nH47|-G78i#N!*S
zqmAd6cQEIB6ooYTEQB1<w3Tfd^_SY@!=~=Zcvs}mWP~f)K`0gNy&Q{n3wOZG9MFF0
zrOAJKt(FGLQ51Ji=JIR58FQ_zdH;=Q$B{0ka~OKPVYq8fe<<=`=(Lk?6ii1UeH7aF
z?%kW!((N8GGdK5!?IP=^qqH=0<29^TQENIH#Sucd%fsv^gLUDu-bplAb>}aP3FVsr
z6~7Xg($A}hx<|b@e5OH_iE>g$@@o<k&;63!*kp{{ixvx;l26M1y*OHL6mpOsGu((k
zku<02Ec-d9rkglOjGHPgpKtrxTOggRy=h`8X^j$Snx8+#@KY_GCYIc~)Ldv0c}6q>
zDniE<)FQ(lo@ITkg?Tk&Adr)mmgYEOKx6Qi+MvH3hMX_AYXO2KTV2DWRRv2U$x)M$
z#z&j9f_{A=BN9a(7A9=ZsL=!$LM%}1;2jju`q*1B7n>cbfhrFnbbHUcw!FM76&&*<
z9&d`V${0^JmfcwILF0-|S>m41wm6DCqA*`DCtWLhq?7S@GzMoVuz#8J5pdTXO|v^V
zw`~sLu8wTee~jYUeIrWi<vlXuTS7}q2Wsx3wUZcmncEIrxM5(>=L<E*_vEEkFO%q)
zo=a{4n{H=E2M?r%T@=aT?LX2iPjJ8!S@wcWH)yL3qreP`4~jL<&&?Se*3@KD(FX)B
zzi&Z~2BmYo*RKm8rFtq|Qahn(y&unQT8F_b6`3#ANLe6ypxz0?EWse*(A@FS&%AVn
z^bw_^bcDq_iOSw-Z;5W5sh?LblHu=QbE@FOjt!53`FIJW^q;|wV%Ny2+Z=A(jpyZj
z(I<#!npn9bs$Sz?tHKY27zsId3;Su&3@BIiR9Eez+?NcQh(CI2xu0hmRZ;B)w<_@6
zyZnNJ&fR~{h_aw;gzh8Tg>*kRG+gw5ID2UTsY27Gcdxl!if7#9$!M{`3zOq&@g}W?
zTR_luLQ$v9n9Wuo9yMKlF2|zfie~udxws%|U^X>@z_)MT){*>W&I}@MB`_VyN&+Cn
z6Wq#`d7M9>w>Dk;N0L&}LHaVzO%oO|XCWi4bh|wcM}d;(;3`W!d076LPe7QKce#7j
zv`)q2{0{qUCE3I7uD7@zJ$lr3068kU9c5E4090?nLQqHzq@l{d!^5V8J5sbdBBeGQ
zQ}m^t%`OjfX-IKSRM@z}DQ=aG_y+hR7Ny%(JM{iMPQdyu0>g`^nXQ@Uk9H5G6iFHg
zuaiso?a=tx-r9;vFD~PscRRRNCuGX>Ns)j8`o;GZi6zi7@YVxbF|OAcJN!)M)=eKW
zRdQ{CpBJPzF<F)hNVU3IxHnEKkOI#$u~I|rpq1lg`Z-Ox&*wvt2t^<^GBVN;x#Pt=
zG)!D0-i~Fh3Zx6RL*ueL^28~aBT}e$j%Yn?`(dI!br#E=7d0W(M{cXF|7nLe45ZL3
zH0lnI5lhw<^V<#B-lMEc8^?d@4Weh)K#lKZKc)kRPk?qY3?l8<z6?}lGK$7Mx8)R|
zbF|*IIy^eMY{vQVkLBBN25DLZ6)ovn#g%y5DvOqulg)u&C;wRf)O+%ELAGIt%GtrM
zz5lH`s4^(<7*0OwO&~BA*81v%Cw(FDU3H!1H_p5MxtT@&<a4Ax@6U1Hm<DDkW%;<L
zF*|nJPx#TzkC6xIX+nxS7^_!nl%iFVa!CZGP8(x8zPE!-7tA=Pf{!mosPBuoO~4F-
zUSWrbvAAHeE3mw$xy<~Bbv663%@YvwXs3ECEko4t;PnDj{9}%>vCFmtQHBt|(o1*w
zMkz|!<*Pg~Yt-~Bc}d)@)MH^~wT@ni@iSJDDEYL>JHQJ)=c5v67IFGO#__C6QcF(5
z?cFWn1_bwTk4r&GYZE6JTmVjZ#QPK`Dk5IVMIt)!iIyCicy$LO4ShGe;g&HY{#kX>
zM?oL<wYj>wuD_zqd-`W%<4wTV_ptAF1#hA_yq|IA*>XRWxqG|1x>m7epkL}XOA+4J
zq#6DI$9Ye^z>Tmgp$@x`gief6FCw2*^s(1IDjDih6KDn<K1l@JCwe*BWhw_>3Cx3_
zsk;9)H+EMV<-x8nM^c(yyG5TNtJOL)!J+tYVBYC83f9myWJaxyHM2tmg|nnM=6w!4
zCw4`R4I6xb>j&ySB&f#t0MxN0mO4mgHN#&r9u`30K|^7b6mMd$@JZ1i-V^37Y$yFp
zCF6y?s4AG3r1BR18K3wV;GlA}>CA@pOCohvG{z!YU7zLoSEV2yvvxkVT$l-S2og5I
zD+Ouy$<vK+Dx&Ks%7e>g=K~Ua(2|9GNpO&V`E#P<7NhyfuBm4#oyO=9w?@y+`CIYH
zlPA}YdtCgaI-GHufDzb4NVudf7^>nrJ2f@+Zr2juQ4APkPLD6|c&Oz|a3u&FZ$LeC
z`@$8z<QVLG(H@}j>gu!xQUj)IgF0)p6-G=`kYc@PSABiG&)-8O(iIhOABcCsV6}N>
z&!0E>sc9igKyJBr|NcUu2G$FkQOgscVC&igxN&29`{Y5y-BOKpUp|mxdhO^Ilh4in
z<3;Q=D*grI0bz!YuD1r20j-Qu1-mzqjG!<2w6R64Gi~CBI;D-G^TE;N^vHbY?Ivfq
zolgGYugef*2V3%y?m|BHkD~8GP(W|f_(f;-Oke|<wqx8$S(qf1;b!glwNaOX&Bk%%
zeVqYkg$?Q+6{%z%rHx`Y^TTyd9$KcY2$Q~s5u+OAE_oe|m;uBEW95f9WreYV#R~<w
z$~{P!WGwj9B~8U+=}VNy0!%4bKdd3mh3i+q>Sl}@l78L5kFp4a1Jh>?iw55=NnN<a
zC;yMGosq15zp_)Oi?J~P6jUE$``0$vZ-jH><hugo`SRZ6<cR2;r{xR})rDR6*PTwS
z<cN&Wn6wAAP;Yh3(|Ug$KMIXy9N0^4c}JxmPSO@b;iZwJrn_ILC&w#<$2P~YF+0bp
z>Zgd)UO3`0Js$X1QVXGu*Sj!@%S2z{Tj%WPTN^LmKZTYjLVEbdlhsp~%;edH&eIXE
zw0-J%S&0qffUEouK|V$`i$Mjimnb0xxZG_a=o0s&q0sm<Q|>i+hx2!5LtFxhPqcD3
z>|QPK$~*#ov+Q3#frxa!ihhFKBCW=i>CpoR4%`!;csH;k1yG*S(~N3iC%qdsu?pu)
zkZx%mX9ryG+D^oodna9ewQ9k_FEYzrgOXfT<pYuWP6g@lZckHp9*=2bs}JnL&~%th
zdV9St#gGP@gF}Ytr8y4kGJnUXC@AO{_U_H=ha_Ne@;lfE&SKs3fJ0GY`hn7lJxk41
zoJ8XY_LaBG$g$mC5pM*vc(MLl<%pEkkHUe&*3l*7t=#6jR$v(@6E9MenLFe6^DWe5
z#{Hb@xBcSbUn;ZayncfB@+A$OAMY0@#UePIJW$B~<LYa(tF!yczHj~~zXfUMo|n}3
z>aGsZB(TNbB2`On{JiMis6hu9o+d85O@$ojX~d@9+jCz}e6FGH-iK){0H99>{23AX
zBr4=+q@=k6#5eb|D|=!6iyMf4k$zC^+%_Ub(eB$-$V#N=zs2RG`_)0seE2Mi&#=WI
zj?J+&x#zZ8*5$!3a+GO#ZYZMM4=l%)6}fs1!t{dOyAV}Rko8SJsA!%#$5tS2Bh5~l
zdMXU0{MT+=XL?K4;{Sc;>1cn{ye+NmcUt@i3vsOyerZ3tX<WLS#oq;jbJgOrc?wgJ
zBr&QtCMqqQcbMNSshUxvSvuASX>7;QPfc5{pP_mhi(XleaAmPOwg;=fz9T?`j<8C6
zpNPpLblatsFEC~PVL^^hxA+(*EJu5Ug|2?sPB<k_+_#fC+2ow^zl&#oEe?p#M^kF;
zGmuilETu=OBM~JX1CJ8ZO3A^uqr<~LJMTGtJF}UNgj@K^MOEBjwGJH_db)$dipdvQ
zCZ4P(zH(S(H}UBu+mv*_S<(%9R_fWKa-1-LhYy@T(qCz43$B#K%G(igJI!Y?jO+p@
zj?wN<E-qeo{SRTQIo}1VWtRqX16>UjImps9yY3(My$_Mwt(<1JTj<T4UULRW?A#@W
zUp{_L5lGxOw(po-(y898>rZtGV!D~Hd#|ju2~LkGWMvpzsZDf9nfFZZW4pntm1R<%
za*C3glw9q@K`a9OCVBvwL0Eiy2XcB0BeIGE+J#tHsT}D#3m`YsNl_-I8&!e`!o;`5
z$4j4XAo13QoD6`{|I}7^7AyH%QCvvLPulz9O*M~vp%d2@-lAw;mgW<y2Y-mFK@hb~
zi_$FoSDI7D&Z2&ht064~BZGF>Vv_;1wDBD#XH&pE9B>QJ8s$+3o^=@5UlQG1>#O##
zLva1##Ze6r3>|r{-}is4zu*<cFb#}XJ<%+UFAJb4!1grXWR1Iu%287DEV}Bomoo%_
z2&(V$U0hwQusK1T+StVK=x9DnZeD2`Aj=B#(gGvwFJqvmsg0r4^5pfK|56bq8KCYc
zwhQAY_19}5nW#~nuisxbnk^^r230@QD$Y0G4Y`R%z!t(Zg2JNUhSJTiBD`EVsswQl
z2Od9uyrg)e=^D}Y?b{(Cv;OHCr4!-rU`1Jh29hBUa_{2PzhOvuXV#~dN*ar>vzH=6
z-J`$URLn6l{$jr*izKz9=+gO>&Itek{w3><*4IT%OXOk`5g2ERo<6-DS8OZU)>{B5
z_(e~$YhykLLg*|sup1pqI6izcNh|4yP4|$uGZt`ZAf9k}v0%Dv+*xqNKIVOMag6xh
zV?!e&E~=31#4r*iBqk;c-zIa0f@#kiR~A3PX=Lp3=ObX*J4c{8KRAFu;#j%Bvd6v6
zn>~V{KMLPo;Ty~=4gT8H-hKd(VH07&7mV=66?Y^1Ss?koO~4Ib>NQZ+cUl}f5aC_t
zbnSR%hyxwbThGq9<k$X*`vdn}7CT1!Onkz2?)zV0eTan~;H~fMlzIhlfVvvj=v%w2
zVR@B@m5Ip^39+%VBqn%=TJj~Z@7$JxIJRU-MeO@%RRunVdm32s_FCiuz1@jZ0w6p+
z`yNDuAD1<5WZ+A=i5ibL%K~>NV+P1u@K+26C0o~*8p`;dSDM6pd3^!q;OiUv*i22K
z0Eg88oAGq_d&>E*uT$3bRdf@JTTxo|Zj@Mo$B!dCj%zoBabw_U4C}?p8eVq!x{$Pb
zqb|~<OrljaJhD`UMSv0NZYJa%$?I|}r|Z{f;&^Z^^c!^;?2kKC2$18UNS<c_RoU_9
z6|mlFA+5W+ds6tS$r)^#1~2$5oZ#sB6S{2tX@UJSU#ontnthWRG@JYmfMPGtzwx03
z82wI42FHg>;n>R$<Auyz<{ckwADSu5%aH%TU4fmKfxxj#E(}idq~-QBgTt-*SSdK*
z{&)<4e~Zp7%pXY&yYkD~=#z&;jazD>8P1V@P7KJExfzkvMW{LnE9`44$)jT6xhIl~
zV0zf-%eXhC9q^xj6MdIRMn=X`3`tZ}DOV+(`vEiH+uoQpzD!RzwD<fbK~+8nKGiR%
z?N3_JZpG%|XAMEU8@t6U-DVv)ZhV|5zcn%T@!O(z29U>mEes_dv=0i}bAlMp;8u!Z
zYM{Dlcw?9qYvSx^pey9?Q)`qR&C_AD2ZA%PPB@$=l}#?bkXIKm+%dtVHg0g}*O94y
zQ#k%9K<mn-L5q1E*VtZmrcc_JMoi<d#?{5*h{yA#s2+N+gneS^uQt^5rXMd9u;q@E
zNV%h#j(3Dh$+_yQP5FU^y#R)1Jxo=c&~-qpvOb((UqV6IacNTx<F93w9uT2xcX=5O
z?hzTZ%*%P5MogDbkpaPr;8S-;$AxuV#f28eB4?fcz8|p^{>Xxq&pi4bco}j)2}M*O
zS<s%ZY{8#zBQ-xkijyk&qMz|;F!(phEYJ}|p7l)*mv4~<9BL6KPy5P=a*h~{W1e(E
z2AcZUISbzvfs5=KZ2n|p;5W|i#u~E}a%9SL)QSEls`Ca#zD9%41tIm8`eggJG3Q5{
z+aWRjaK?iN30C|9;=fiJ%3u22Xy!2FTif_pjg2i-jcn(Qac{c0v0eRH{2g6u%1G?;
z%Mfnda2N!otewn`HpL0fLVefzfQwhvuV;%eH@?`!i<l$i5S{cl2I4>5`Z7q-F8N;I
zoQ28zk#6gs7`^V$0xU!fnx0}~dk;*L!%AipSAHleQnmK0@NtBh<B;?hFVDeU$ak2$
z0fZAGUtNPJ8{?VLUd0JomV0!Uv&}BqL)O1&p<Y(v*Bm~u^-TaZ?9j5x#M@9_r03(#
zH?e1CZt{j_D63Uewt?MFycPQcx`?oKON=@h#Nynhvj{67O@EHi^oLJTpUXzK-T$W!
zfi*^ru<p449UN;MuP!bo7~c{L3q$WF>QQ2tpUgjWBnuC_#6&{}@U64>)*4I-a_5af
z@{*H_16NH~-VU)<8XxjHM*fKMj^5}4J*p*-FmwNX1PR_m<U?dB(9O~g{j$>Bc?nuj
zSf_(klFpu&R5SiHM*TNChIgBUBxhqzzEN8i()F4P=M?yy!!a`XjIuVv0xF?_&sp#s
zvH7*PP>uSxx1fYcODVW+SbWRqP8GgQ$K40ZOIKRI4Lm+goBGX&97bkezq+v#;NR7<
z=piI9Ufrv#x4Ne_7D@;rgK1XQ#{H-DQ}%b{bTl^p7+91(C^BvLs^=lWBmJe4D!kEc
zNh?UFhclOPhr<!AEKZjJp{Soy0}}pQH+T?!)T2Rl?YnP^pUDjP5UA1F>c}P4j0~P{
zf0I4){@5T#P9L%Q&O4o3j~+WV_n=QG=oAz2%jqcbrg(NNXIn?#9Bs!Jonf?o?m)k@
zLu0TyCYP(s3+6tqqP1qt9dwa&|Ggg9&{i#p^q-qQ0U~Y+B;v;JZ0S1<o~jz(HCOJ;
zhk+0E)UFp;9MXzvTImZaL~FM_S$*+tYWqT&H~qT6&+CTJSh*Ec$bLLtvLY2!$VRU2
zCi*aOk3xGoe@L~@e!;+N6qJ-QNQ1|ufuQsGJr(IUJ)8R-E%0-PNo!6)Yy#OXJo6i9
zM7z~ktxDg>3~}}*mYHZ%8NWXLETP)J1VKn>xMpDH$89)X6z%1ffB6^j-f@_!vmpF3
zOx!o#p*nSSM^RsZ{99<TaGbnK>57k9aAbQx1`^It^Z=*G0parGuF|)LB7+~{*e*h@
zsShMELL3(Ip9XreIY*&v+jXJ6y}d(ikJ?`fIcOF!jFa6$Hq^8$+5D$!05Q>E2iPwS
zwr6*TH<=0+Ys6&js_Z$1(@iXXC@<OdyA%_{n%`7hRK!FB6&4hWd%9%ZWV`gCatK$x
z*y~g5G^I&^{BNcpnneLT#X@Esm*MzMMODSrnto(5#{^`%b|Ht@{CEG|a7rY|k3`^h
z(|fII+z#$A34=_dMQ0XU{tMGR45G8B2#cVh`xuFLC+PUDIN^%psdNCxY3#wGnlP-8
z)D2J;!R>rz@js7INm46XnW@%m`Tjk$a%C{kURG8%3x?u^)G*W-?;qUkpG0dytyV<7
zDHlgRG&gZo`AU;soZ!*(@zPxt%$5rB+%xB5OCILBx@vdLlC-q6So-QQv_ySe>3_0s
zE5G=F5f?4Cyp;*~<l0-k*y%`pa?rq|QJwTOg$(KuE5>b=D`_eH5}j^QKWmITgjJW_
z%55Y!Yl<+WwvuUp8&LqtS_RAg@@4_~*UabUioPVBmp#Nq0}&~4CiHhXTm2Owuod-y
zyvAOjc&3U+k47Z*oOp;j8&$LTm!j1>`8hdqZQ?`&>{RQ;n$MWe*~RY0N8<q2<CiAZ
zw(iH@%EdVE0^@hx;2BTO$?tNk>4gpf^<WM!Djgl2PEssFq92m~jS~LB$>CTSkrpp(
z_%XAY#oGpFDux3GR!3lHSf<3T?5`>xAnTixdihn4d3Nc>24<{PP=UgVf?Nk4+>8DP
z+vgRM;-UHv*;COV)U=HMr|H^;FRG!>;#3aSH@}k7YAfJ>B5w%#w)O95G!}nKOs=+V
zKhmXSbK<qB2OW}!xv(%VH)|}oZS#eX0OYX9<D@bE)O+a|#Fs=M4?49ZeE5fijon0n
zg}mi<3(JpJvZ0k&VvV!6c(zB`GWg2DKjoZwH{@9MNIHxpC4S6#FuOVk6Hz-HU`~nq
z@y<&xe@!|?=9_?+-@BxS{}cgZFyw<5f|vJv^i;{y&ktJ`q!rzTp@rZx)78aAk^$;i
z`jM#VeO&h6!V;f<L1dR@70G+)9o~z4Ir<t*ux0Eup+U{k-VawF^Mg5ALY+3BJ=W}^
zmfJ$gvK130l>P<cDcSB=hrfsD3{dkN`E&7SjipNaa;zTMuJ@E3teUUiPh|GJtE01X
zo!O2-1%Rxnv><HR9Y#nt17m;)n~ew$?jio$uwB$kddt~*)fU$<9Fo%yf)E54^QU(z
zTo`m3@yo`SpkZC^5O&jj<%WSyTJfcLZpqsNZnC4imgA}E+V_#u4@Lq<m`Wbx(O>?W
zDT3|(wOa<j=#AUA_W!XG0=W!#a9?I6kSweFlA!qo85p{@T$}sR-Dhtt=Vn5gq=9*<
z{ULk;|Jev2MD;D2SI(-rW9XYJ^YsRUgi6FMHV~MEq_yKmeyU4e7`(%=nQsQTyCSQv
zl$AEKWx%LyM?V-j<6s%4bYf{JHWp^T%`$(6PHbBaKp0O1Z){om-F@Oh4sA#tpnM{y
zT_f*nMAPjlA;ED-=|45KvYITv^;ZZ#)wxu3Tgg8fZDp$#NeUUCN}IqdF1NAh4u9t0
zJ9!MJ!f|2mF$8NV<IRXqELS{(7$8*p^<9oaQ~BXb6wF_5iXqLp6nFKOlhl)Q249%O
z$7z?i+Kya~T9m%>{{+7gQBsRSJk%iy@bLO;nrlWsXZjm5BY&h6f3l^p-9)EoCy;R7
zj<96Bt#f(;AL55Ht<8yVy09mbt>cNffQO#pvHe^zFHqIn=g&$U2-2+f6QN!;0K9!6
zxkuB*zvd)ZOfiePonAKOv%6R$WMc>X%zuXGCLHi#tP_Q1PYo82_A1@90qT7bINz`0
zKDdv9ua!UJ^BZA2kP)$S|02G0Shb-yCqFxCfK5(3=hgE{eM-=B|CRG$*@Hsdm(l8e
zTXx$B-b!#kWJ7dahNEiedKotJ$Y&(xWz>*X#M_i=h&!#soXyw^e{@s@C7qW!ltgn`
zlhVq1a&mJ03U^=#u7T$_2hNV}{>exs*kU^%WG8vlkfU_pk9D!Me$112wga1qO@uUK
zQxl-j4-diFE~+nSIDBNDy_|1hXJ4DnG);@~B0eD(Wb(aa!i!nF)Nuh8-L>OVnsM?l
z(a=8i-gU%2M~i`SskITt@8pE$9H?jXwLo)Xjk=_$D1*+8?}q8s*SAt{CUQOMz@0SR
zX}q{$wh<ZS=St1^n0ICww>EpM-@NA@YK6t&kk9DEj0_w`5}XdVb4cAszp5&=i$pm;
z`!y;3S0W4MyHxeh_%)?!FSPS}OnY<0^JKQX&KrvHmV{NEj#d^8xbhf|#4P@N;L3c-
zW_pcJ=~Bahk{(Mh*Z1D_?}tUCXe11m)94gUUe=JM+()oxr8>_#=pNcUL%c(!s29g=
z#jjyC66P3$-|;q=`mo)EwJ?;y{O-%6WXEKngSfWMODykbY_`Xfg`!FK)TpnI{-t7I
zd@B7_FLpt~&KDSQhBPiC293_P*2y>iSfA9}_}51%oGT6%VX;uQHuQ0x<bPYMO5g<|
zq8^dvZ`HH{ws;ZZybHYi@YSc6cwQc0K`@b&-KT}PsT|9>eU#Mno;A$Hx1%tT^TQ@_
zR#w&t&}Ds{40#C)1+stWDs3M&g;NXT+Rfi&OmB7we<gXMhY`Oh;u{g3hntWhWXr?G
zh{>LlZ_c(mM2(lWfwN)7Pb_I34V|V@J(k>G^2#9;-^`ncna308*N<pv$VTOmw&0r1
zZld2Hk8W~l3>;Ttex%r%ml!D&p-KbT#3cPBaB|mq{)&VB^RY(20p<xf)Fw0`AT=4+
z!5*QG%W+}e>Z7R(3dOl5!pV-7sMs@s@Al9gxjZ$FB%6HbUmW*1cNpskV7@{pP`u@N
zRxI@uV91I&9^Sv{7??4n@2)*N5Wm&D-T>oXE<VX0CNBN>9zUC=8+fm+_Imc00>qx4
zY)AD^xMTY_Sr>LvP)IOf3)NqtP#q8!?~l#U9lHpv6TkAcSU(v}P6VZB;)(?P1%BoS
zt2vkccqcwUxJ`8TUR_g}#KC2{3$CRR3$h;cmRs%B6Cs^dkOMvLg~@8%!>@xIrGqyX
zc>Qr<gd_ROZ4Gx($vYB&&!~JxREU;(%q!=WuC<{tiFX5SJ`rC`zEdL^(hIpznK<+A
zZyn-3RP_eYyB~gpPezXtt+ZK2WQP)*{yfM@OKi0GqN6W@>medHbF}vf)9g8N=G^hO
z!y#FKNH$cAX7r)Fb22w4C#N$w?=vfIbsym|o8@hTwa(=sStmV~VZ2LGi4;~=7>@~h
z0nYaobRl0nTz*XbEs@O9yo`H9kn+XWyZ7!n-T{xIo^@-x--wsPt~f~%WBAuBTmV8`
z>eHFTBgFzB`zEi_+4kFOAxMC>!T#6humA_hNCtvgGu86B+h_aLZ^EiVbbDnIGOMAe
zC^>nwHeRKYyk5+mk`tc>-uUi6f1E?07ky*3_(L0<gD7ApuR2Skw&#h}lvN-E=G6XA
z*(KkUoxGNz_|IYM6D|5*^2S02)#goM(Fp+4Y_K+j82ECZyO>1_by4A4p@>UFm^Zmx
z%a6SZyy9VVu9wt*)IJD&Ly+jFnQ;Gh>~?tQt+E>UR{Qd2;_v#Z_v%CP)7;I@-UVp{
z=d|wYQbP**mn1#LUX@Rf!AO^z=n-TIcYDA&Ec^g^vs{{q)c%!1{`EezIb6JDY7V}w
z1hW&Nba&aZ^T^408NUK!*ZEQ9Q?o?s&?-YR2~B_U{}CqrPEQ>qJzh4zio43_7r>mx
zuKk|5V)-u^uYZadXEc<O!q7s{y`|swr#Hg<VVw0%P3opSQ}`-N5|X1X6Fi5_+5tvB
ztQlpX`n&cE1(d3Uojm(h+<REg!$k#tjaqNZu}CX4T8m{kn2ypJhgf$_;@!-zSogQD
zCCD)}h0fd7w%6>%{eM`T3CkTmNMEAHX<&)d0ak=r0QFbkP2r=hDGkYuW|Pn^J!Ir#
z)iy?+!VvbTMJ}`XB6pA^(|V5QJf|!f)wv`9L38Qgw%Vc3K*f)-%2eR*Tz2j>B-mNM
z))ML~T8PeV4rug@96l7FB$?rpJ545;#dE@=ryCE3gnmr#;qmbzzW-6FTY^#_#SM`H
zC7^78d0t+$R4bh(<(+NSqLL@>?OdJbfE{{M2>|7{e+N1K;voFxDdzyoI(!ksci?fp
zMw|ALopb;*B&6>KQ>5Mq23znDFP((h6fO&Q5>k%n<lAc_Ib2?<7jN__ag@cx!QBMt
zZk@#<%aS1y(_*t=`t71v0H2X-Y?Q=`&nU$yfJ|qlR&>vT=k1kamBeJTNGkh~Pg#Cp
z9!lM#a`7thL+T?O{+D0~jUG%<%XgYV3VE0?c$wLTb;Gh2tbr*_H_h`4tqQ{TfV0$_
zfs|)!wYlM$!V<BqpHwW{cT3o`ej4lJ`O<`KV-5tA>&oPR6uN%XwistfPD%1ha^UN>
z(va1c*jIP6HAk(<@UW^TI~y|VMAQL*-Jbse5Fn@ug37BwX7e~ti{bGoocW+Y9i+C&
zIYfz-@9FikJ<@(dZ|=2NBzqwm<?Xs~)=THh^M{U*z%-Hlk4*aS-~1E#W+^Ew>JK6M
zA&B?ZSua~J4~*!KWIoaV2X#%@9%(Rhb4^lv$sQc{fNVbtRk2OE-SieeF}Go+iv1b4
z!6#rb1(T4k0T%k-XyMJ-aYW!CeyW|YPxl+ceF$bJf)$&P&R@U`^~v5y#NwF%OR|%_
zzP;TYsd?h0^!JMay6Q8bY%&kH<_&ZYs^Jxd6ciGx>a<=G>UYY(LqyZtRIT!hXI9Yn
zKM*6$VkY(iE0p(h!&}CQja2Y)?ytwgIiie$q2~&j9%r{_%FfkN7v|653Bm;FCj>B0
zaKvB8B7zK)h}Kgg?IOO2W1LiNTP^l9%!Y}0J<jCIxUTQ`wf!Pl(DVlw3YPSD1p99k
z*@~KllnYdYWC%>KazX!HZ<;&#!!TkwVcAcItu<&UI$!0ri1{6S7G{TeWeaF44UJ~K
z{%5D0iFO<mm`zALP4n^`awnWfmFRhh9wF%X_hmx!551#Z=93-xSRjB77<QP}X>4Ur
zD=cDfS@>2iak`5h#dD9{bGrZPIuirKV$IcE_zvDbzWZPLw_X_((!LFah)vKZj_$Qy
zKHDn)CFt?>&#d(g>JNe`#wZhxiHo~A0;M7PSJOTjfiqg@Ll9v-$Oy;IaiE6qLVip8
zlYhMdyr+DmOK)MR0-rjncnabn%zrs)pyI?f(BQ2JwL#S*;_arykZODOu;s)0p(XVs
zwHdK3#7>f&o4ShsbuFyC0<%l%pT)2ofV|OvJK_-#!U337C<kn1e`N6M*RPMI+QNLV
z7(7<Q;<#S{j=Oj#HiFm#1K7@thn`alz<x9EYMSG0i5)&({KalKjT<KeV*d5WowInM
zC`6jZCmAJYm1b3x)JJodpQqW)D;f*CI_E$|LwxVE1n9TqLG0`QO$6+FD3G3yt53EQ
z>VPY9`gJcItd<)c-^!iVcav|032n{KzEcg2KlQ!qy0AS^2SEk@k1FpjsE-Z-xovot
zlE{M%!0u^)o$WFCsR2HXom-5pWi`j~t47!!brsWjd-?d!B)#iylqF*qeA!v3)*mIa
z3#4*kMGP$u%3G_-0t*Z{rwqB_Q*NqwZMt#PB%}{JTmJA-ywHz=7gCph3zJ%w8M#k9
zm*b&&TM;h<dvpmlbo}Av<u$nrOA~pB>I_V?`>-nM2ZXPUJcGDSYCJvS&@7gPy`2tR
zmr1zr5?@IU9V;+l3f~~*uZp-L+TRow)INnbb&BQ@SCrK7+)cQp31b1*UH>n2^~<(`
zuK;r*I1}Fyh(`kMnMa0f^4r!dHkV7u1A<EPM0EC(K1f}ZO;Q-@Nc#VyHvfo}A1^Cy
ztTCsghUmWI6|`K44V~(&A$<~72JO2A<suqN*P9kU*1pBL-_R%S`H^|sh~1Xg#ckXK
zy+crXgwNcd0MWlusW;K^xheSr7T>E+6zrXB+jn4A)jcP(;3jW{+jSE`(P{usQy|P5
zuxDz(`@~qUNNO&L%Dr$^c(6ZI!*mWTra3x9V<${hLP3nJc)Q^5-=M;dbcgcZ$1g6x
ztdKqN6}1WBn@H7eZsmwE&apf$WT-@liL$fZrChTKc(>o51@qo>Zx&tItA`%O^OnzU
zfI3~<?}M#Rz%id*0twf2W^|c)^Pnnu--+mVwrm+$o=RVvHOU&Y91IK$jQ>}HKt;O4
zNfj40gPLK&^}tkydagX(#$@sW!HVyTdXZ)}6urGV2-@b`TMG&b)&*hx>?YXlzx2~4
zviKhaB`cJ5Sp`?nCwk3`ZDBMnxOLY^&uqv<3Pw9s4mH1-F!@;VvBn$06I_;4QmQM*
z0%s~>z=74gA}og7Ldg{ny#a3a$iEw06%K%cjs%}Qjbj3|E|RKStMc%Sv6Y@RK6*x-
zxJABkr~=deZq9L{?dOO8QGFypp1*@e-3yykfKyWP0>1XS4&$^*wia*u@!duFyRYp%
z9EFladV|Irm2G;<VXAl}Vn_pM?t69+M^&63FviaCXQ?Lz4@8EO5sT@64WUKIol61O
z4Q`um4E(-*ZQ8t+k~rlNU5_1QLl-qIdq>7ZcN!ecG(NMADVb0C&^4l#aeHE7B4_*G
z!JIgy5q4OgU&HQ-58ls#CIo)zz1jl9_!E7nzij=HxkzI9Qe65iWOfMivkQ8<f^vQ!
zcq>Gj(7Q(eCGgb+KpJEKF$nm#IAqXQQ1u}$+*kfm9XR4Zvq^aSuVxb}lbh(IEH@SU
zg%@A9RC(Rab(c3>J+*7`Zpl^%lfE>3UXvgZw;kf|rI=V?TDlLR%bybQ`=4kXgw!n0
zlUl+EXnaOGbT9{$x_#@wi2R_~qc3(yD>7|1afdoSGLkCaDYqHRiTz%KOVMGCrTxgX
zuv<G8^K|O(Xkd{R@EMTbI1c%Z&ybGBeiTd;Ml9cc-PL<`qS8ZSFwNr|^^Z?XJtmKd
zkp`_9q;W$M0R!zIYH4ld;W~F_2lk&vCxL-g=oiK!e&f!8<&gG1Nc-FM!E})0xrs(x
ztav%6QHT~_3Ml5&swMflZz$d!Z9t*7X5eZ+@<~FNXa`qH$pw;y`lvhP^d1t}Zre~O
zrm}vi=D_X=e1T-6Wbj)W@DzfO(6c$-{%Ra<U<7HFI&eekkK+W4MOO$azCUrKXHDMc
zr~A>jc{8_u`gDznlVdw_U|ev}<-H6H-{}6@;rfSMM&}`S*W5frxI0ba9UW!v>fu_u
z)RP<Ev`(bz>X;Yi_K>9<QQGc3F*KB$2E!e1{jFH@r-l#X2KuWyQxN(ep6YZN>$HVD
zn<mMW6=S4g$eUfOn9SZ9cZiuO`Kp&IV3bI_ugBf^?ZSU&T;MXa7SVMgxoWuzeCrKo
z3fj3$zVqU@o_)%T5!$hhuJa=LENA-E^~*s1-mI;yts})$e<qBXi%O*$`*>)T`fKFG
z2;k8UNPy2sAv-}>f7IfOwHq7`k_6K<_`Kur{B?+ICB=JrYD9!4X&|>rC3fUM)Kd@V
z*pGsWYed0w6A!lMC!D!6ei_zs9sVm)g>X{}bTEgivg9s4zyBK9zH&V!@CZ(}Fwe-J
zD7k((X{Y;-YE`m&@_6r_25E)8ZP8d99V8xa+<03AiO1kzZ(0r9B%9v_knkepsQwP1
zv@+k`n!&%#_MkGikD>#g<m`g&#S7-%qBCh0sS)+pzf&XdK=!sTUuIwZj|w;#QNYwj
zfrJZtj~k$6t!8;h&gVtn@`7SiLGkDNj~_jj#i*$H$lQB>pe+REwQ~9WNPe&dQx5sr
z83yWI$Uld~%9Tmm+!t~LN0abO>`~ASFv)|j<zX^z%q&u*Gqul#j;SjV$Y=|($U_jq
zR+HS2)d4kM<`^OuKgG9L&-u4eDt}znv9mzry8?`T>u($<kJ{=iT)jEj{(XVZG1o1H
zQ~I-t_0D+VmV_h=UarHu<V5EW^kLq=iV%IZ@1%pdyh(T#W&J$!bK-xZPmL`?W47q=
z&+Hf<8q3`p3M$_?1M<zfSR6)3*vcM8(s9WhVtd%uZ<}Z5%2%mb%jY=oX{>OfZJ9!U
zvPPnKcG2B9*Z7#-0Ab>6v<QmrWLXhU7VCL-B}w`Nqc3*{k@IF;sqM&-6chWId`HIr
z%0c4ylf^YwmNbGk=Mf(ookW!47$0U?3HBrUE2^nbMX9*&H-(L1M+6^ZLugVF*om2=
zSx7olE-CQnElHbkn}z4-#jKUD-z~K-7FVFE8yjop1_lS$IfeeOXj%ZWWE3AY`El9=
z{fW4cv+5_&UM8IH0k~cF&e_a`5dP6mLTgQd4ZBW$$hFIP{q-qBvb0E@PSu44SUad>
zfH>OGoT!d%TvZovYXh~&E>CI=bN*o!u-t)pOf_Y+aPC6jDZITjq&}x)!z^V#rWeFD
zSk&h-V?Vy?nv|$0Mr5gd*_(UsdiI_*qQ$6>HiZ`{cWV;dhRwqATWlqBhUHpyxe4o4
zLhy)*AsZ_m-KJ<`bu{VfwVAmOiGjsUht6KCBo3xm7a|@Q7?6Cn7iL!dm(TU<`TEg7
zhh0y!jd54D1v5|7v*;_|OQNx6=-SZlo$n;(GM*Gt9g_2?mHu*Cn&I}kPoLJ(Yrqqp
z;qvdZyDWn<6E58PMWbAaACXhAlmpH8EBQVr1XP5heSpE%?xGVecu$YiV8@u=&wdg?
zFBuV*38SITMa%K>{G@ctSHwo;>1Xd5@fou_Pj$Wf5{lHwVRF|3`r)x~TO`oO7@?X9
zj0Dp*$5r}JKNFqrbT}-abQeFsW=^lfJ7#U)C>~ABlEHy+S>>5>PaCBma^2y+q@l=L
ziJV1zM4!L3_C88ihOuI+mQI0AnckLags=RU3i$0xhrA%S=#Fk=#ou(a28tY3>IN2>
zC3B9mo)=uN814&>={xM#L4HI-Lug+_RMh$vqrXnM#S>tIWEc#%-ICU}2M7$(LVtf!
z)W&oLw&^u^_&k14BdS;6%)}q*x3)x19~Q!M(0cj37W@0QwzeokgoXd#<Y&l*uiwVe
ze30<<LYNTCr6eWk^J=3h>Nc^(Kih(GXhcqiI`Cw8j@YV2<Oej?EWLmKJ}-r(AEp*i
ztp!L2G{5S1yV9|(5-(?r#3_zGj0f6tIvh&bkUHiE`iubzPs&HXe@;czO_73G5e^w_
z`3+A0jon&Sw9MVsH2!{Ku2pqr1WEU{;>B*&YwKANmyh<-h-%eJ78R)^wK12!qIlo$
z_ON2p!l{jG`}dCuq+BEx#qxn5x^XCStCe`AlVz(TcF&h8h7F>OiGi6wzZgktH!qC0
zF5F2anfPy#+a98#bbTGd!(n%o0Y|YCYtS-`X~i7`4QD}CI&I#}5#qZ#`{ow;{<#|Y
zhJ~nNwJxIgrg$`U`x}4vTg%_Reap!oJLmBK_BFO(B$4z4$K_fAAGq*lOnuZ_@j*rO
zFin;XcB^JbeV3PTs5)qEU?~x{H{FoU)yULz__VseEebq9WiA7Ce6_C?Xwxbo{#V<!
z1+yE<cJ4$Rwf=bfZ@VKdF;}V0ex<|Kh*`Q@N$?|w65Hev^R<~zbf5{0wks$otZMy_
z%={cl(H?U-KoHmy^e_0<hOFp@XciQ=Qc5gcIo4>(Hr>DbvhL;T3pZR`mgEtk@6%(c
zT&MEE4Y0E31r-yfjP1A)I0PzYcPQtF-@910b3k&lE9D1)sfM&zJjTPDIVe-xzBruu
zV8t?FPwMHw=klRNCNaA-L^_(fIcD;^BbQM7VOd}C;Qvc$3J_sLeaHJ5Zx!(2Z)k<g
z>Ee@Ov7?9ZWJJqjMyiV9#miRvgevQDZeP4?hSB~oFhBjOi6oNk{7mZ9$0?Vj=d4Q6
zZHRSQSgA%LRm0sG9YKZ({rI}dzj+UXB6rFxC<){d#=Rr@DavLgzdFdPyhEzzyMz5L
z*Kbe7Z_KET^B#`O>|VI2pO+u@eF=0x`v0l}kQU7|{muCLHd@~b<y}7U^ZI_-eXhw8
zF4;{@G2zUv9&;ZpO1<_V2H?gUtIoISSR?s|NXMc_rP@Qvcke@H4p#AYE!@Z()FW&u
z!lXgUXcA9AM2Ej42L|(F7L^=H%*~-#SgdYn-xqFikNikziJGBMzUd7Kp;1@}=P3d<
zAG4c^)U)yZy)zryy&Y_F{;Vn%W-cRJ5+-uO&JaI0Vna!nbU_;)^Rw(of(4SGeUVnU
zo+GnJvyi*zk*p$HasBziZ?<Gd_*102J5N_@?=RnVcU{MHZyb*2{!I*=ZV@<(nd;A}
zqAF;{Z5jiECjv%v(1?2{K4;HTA6n9=+D7kJl%A;-OMjrFVxN?A{`{F(5!=H|PfBbZ
zgfo*vLqnB}|3@{QkB}$|wT+<D$8@5#iUy^8r?@68*gUu%TDtb!d2^aLNDZhT9Qw1T
zr{@gTWD@@V-%f!u4|ic+({g=c$ri!|xzgX)TLAvrR^CytKsTc|+hBZr-GjGsUDU@~
z6t&DQ5>qzplk&-Yc6RnH`{Veyzl<^3xPAN&fDKOXfAo%u@Cu@Qs&B<J6P+H@@9s%P
zur{2fukCv?Qu$?C;-bMlS4$3)if!hDbMtcXGohls=4Y5*PLR#1l<=2Hm_(j%;9jcI
zc>B7R?|eWD=uo4gyt(Ng%O^b(LPncMrhk<Gh`LzhvFfO}YNM6cp|sj!c{H`^#>Yvw
zXY;RSwcrn%QpP_%U0RKJ<@4i9!sqht8^V(=*Ik}pT6NJ(y)kZX4o|r8d8u37YSxUI
z;;2a0fq!y*dd<e?NAyxPm>sV%|2RI{eELF>pM}Ts!nOV9(yjKHOBCCSpKMK%N?Qor
z9Z1()o}cPhTU(oIGk!f&GV<e_@a({kw_DTPukhrn(?u4vE?JGmMlAhUjF9>wdAe0&
zu&_f`f}Phqx7gwlMRV&*J-_tzo$2InRz{j{tX)`6FCRaWEM-X>7~I%<<BeTU=>w~K
z&aP}eHq|uW!jfywZPf-p$lTm{b>Z4#j?<5Y3#&FeE)^GRYp!aMCQGf4n15-E()=-M
z(cJ8{)T?Zj?6K;xx^VVP4aK!qG2g6p>tw&-5r*)u#ailhiAUMLcB-dK`et>1TvRT1
zflZtRjS@NE$JNyrQr0hAOL1(DUol~`{*l+njo$ljGYkImbr69Ni;w;IsuGPNi?3d>
z&BXJC)m|R{wWvBDWm}t**C?1tl{yxd!#xXdCOtnREOc>hZq5oT3ei(2<GM&6$;EfO
zCTfIMt8c93w=2^6kx>nWmyC>zv_GF-Z_lATcEx4usLpT{Y71%K=@OR^Z&LyM2r_+p
z;zB*Kd^_VloM}ey0@FNlFd>1!dLO8P^CCOR+|zd@uY3G(?S9|iEN+*kmuj9viptSw
z7X0!wJbc;ZmrX0#LiGN@M2Fvjo0)vbL-cduzPp!3@nFMW0;Wu~mH7qDRl0t>+x0}3
zMWW1oroM`whN7pNp4~IoM65E=5p~bFDK_sxZOV2%%_~>Lqi8n-NPrZhl{&Ec?G$<x
zn^m|A4)s3n;!gyxV#7VWLp6nwJD6}0x1j{oD5n;KXWgLJhZoO!m~w=SIg}bFEI1S|
zh!F=TCx<@p^YdGG*<XI%D0&OtO2L^vzPH%YK7{K!+IgwOMMTXbf?$~mcLi0ru5Cf5
zFT}DE{^TM}@!F4lLECsH`5sCKunKV~ainv}NzM!vflA}eYyg$UZ}dy0?L5a)`0<Cw
z*v%9N>=dC7donsG>U8TiPJ;h|i6LD?$8!pCVY`*!O3_r8#pg@t-}{Z}6d#WMv0^so
zZmhc(gJJAZhA)3UmNQZ9@v_slns?VO!<#4A>SZw*nx%SdL$T=za9j0CxQjfFNp-Y1
zjdW#0P@m!$q3}>RU5LNox3{Ckh}@!8e4T^Ya2%~n4Mslw!hZ4j*~9*p&xAk9My9Sh
zsXv;T+0iEM$rYAjFfr{ez7k&1L9C}M6N<mIZFCIeIa?i*a4QG|thbJ~msC37gAf||
z5?7M4Fs1K!HS9_=6=ULO{gMzJ&2TI0{7hQk|MB=txC^u7f%b5XYrTwbOB(Jp)||Sa
zsOYgN=GDK6I=(_drH!2hCvXhjQ$OVJ2W$#ALn~D70YUMnflK;;XAg;bN06OgFh}k^
zsoKKH$6qR+KO59{Izw|g&B*iKH4l%qUH3%K>JS*glIlgH8Lu8mr}8^oSE#@(VE*0D
zj-A0b8x!~fs)@+hnr904y9qCJIjF4TIZA)#B)w}xlR%5p{Pj7JPln9(5|5q_3z4Aq
zT+1FB_$eQHSj=s4YR@KHwT*^UuP<e1#z`IQ?O>jhUY@-BK1~(Rx*Mhi{TvKC|5tlC
zLV>|rfko1$r)uu=D@Wg#<%f;UtoAm(+UK}VjD*}DU)M>xbNcrBtoh08BD#CXe`mZ#
z1maGkn^D)gn;Z`Jo0hm}MCJu^!Ll{rQU~$pO@+Rd*irPG72QKVkcgPNj1}C;F`swT
z6nC$!GfG$T#BQfj{Q0oh{e<)63!hde5)SLEF!#dsv9U2(5|Kj^1X{Q3Ax(EEVs2#$
z!nR*ATm=e}Dn&lW-?=l@RL3bMqjwDW7d$6-x_E4kOMoD{XiuW|!;Rs+_a@fz*0MU9
z)fc3|b8c>CIwtuURozXrkfQyqYmGzmC~lh$T$dLeX?jBu^GOr?Rpc8=N<b7gMSL3`
zT(mVxQkkpmi5wQ01+SlaS+Z6jxfXvBLZu}7*kU`I)yDm8hwC{7L~Phj9COW^{+zb3
zi%Hzcr5Qy}P*NXgU+m0rEyA7#g8|xYf1iTeVue?W2!Eh$zz^h~XsaSjv80Mm%+;V|
z@|30I>M^=@!Z|;<eERslw9##JCOALf=Ukpt*YjaQbA>;bKkI{qYeGTT`4lBD$S+UI
zxNrACzz+YCZUHyY$vZcNp7k1-vY_Gx{cj3uDtae}FZK?TL~}$bbbOB;IHn4_2|!<J
zg6(!cj&(TcNcDQo$v7?Eh5Ru_h5UANfnt`$>|XsM^F4A11l*Pr;*skrDb5kjusvqu
zCb2)NkkW(LfMzi6gU}C%dlpl}_mF6Kup1K<M0R;h|6{+%ALUmlQ~TYZ-&Az`Sr^ms
zTu)x5YscyEhd}GFtOSMX=#ey6ui~N{X{roZim+i~c3vzQG&oaA@Iw8WxAH%n5u%qN
z6ac|!zW>6p#yJ)z)|x}<@^wmxJ|^e~kA&F;p*wpnKYP2M_kyA1sU~_lx=*xx&SZEN
zy+Z}6>@@2kGVEv6gMn>E$!?yGg+Gxt@QbN8em^`I#w_BVoV%)RyE-FsRNJi3?tN-A
z6Z`n`(Ovl^4I^d=Wu>L1+q&a&S%s7dWYQ-^g@=a+9kfrzZ9WiC24O(wIpDhj391A#
zKXKsy-g%~uO)et;+LGq7Q*Zg2Z0^!~AGR}j)Ra#;J>&!8Y$4HBo}>5Uen~Q_n%q2H
z(!|!v%xr{Q;o%hmeRk@wxYwSXJH>$eh@s4zKho+-!Zsn?addow3^$;@@S29e*-2>8
z#{A<p*8PdIFJHZRbE8_X-*DMo_pzKr8f$*455+`aU|{DF;Xr(z6LW?9(H-qH5)*6e
zbw!ha=44_q+tlZa<5gu_V9_TlWYF7!`-@jljX@-Dso5Gr#f?@M<%OhjHLH6o9)WF?
z9cYRjp044j2#t@ACwFE@KT0)}iT*q|FO2P2cBi(Ev5?JZD)7xE0{LT+f;ajOz7K0Y
z8j7nf{bEpb10sM+JFD?6BP{hS{*gVB9zzp3!dDoZ+MSHcJ6&$fx}QvUz7WlqaYZM!
z(_+aEZt{!*IW2)KGY6F`%LN|cPBaHTM6-sj4_=KC(?emQYtIo0{GlX(Gh6Ja@dYI+
zn1;3%bjEYPqK)TqC9*q#MeIyq^&%3F?v;N_<ja?0NNGYy0N~$t2F!vexDr&W4%p&|
z7lh2e^1K9w0x>;A4z5H+CXWbSh)VcSHg0;Uud`<6bslbaHf<$?O`r|qN)ZPP7M2*<
zFw?no?XUS=y_O#zmVGNE=e!r+>IA9csWQdM@?fhLLolzH&k$_NiS$zw_=s=|q;pqF
zNVp|98d>nVhh-h*y}p`|e)mgO<frO$#qJ67XU+$NG%(0^5LxQv()bv!hg6Haw{wxO
zx+X%v50+43s(j0tun8U*O?G+mp=h%0A6c56$}Fzo6<bJRdGLpW2%5BXp<Tn9{+`Ff
zW%bt_Q9cvxE?tzE1E?JSl!;v1-Vd;7#;?t(krq`(Lz;MSNly7_qB_h%d4gtfJ6N~;
z%HmVzk2D2r{jcIVh(hN1R|5`&ZJQfsq~g4HbC+7x(YlRE-EAz^UEfxB-SPJ8R2Fzj
zZc`LKb+BWu@=N9Dh~Fw}hH96IgT-L{J|-mv1^1tKz4-`_n&Id=+qzQwU^k*{(34<`
zibNU@;YEQOGnE(Gtil#*Zi-KJr6L%U@mRtd9OcUx>o9de=dQGzT_@jV(#JGEb8-F_
zTkok$oWK*_@OjK-V9IyD*vy4Z{`(ZM+c}99F#!~EXG6ts#`TXb@bX_ggS?#v>7hEU
zl-@^hX;O*leBFD@5BBhD%l^ojYZx?>YUyOklB@zNJ2&q%TDaaWM5fo|B(O4BH4&di
zNX^-@>3`5mOZcFVJSSQSL{RdbS@6H#1;AE1D1}uEewiU8uf`4`@@HexpXkWN_$@xL
zu55uaMd*OfGbB}DS0f}ORQ&i?W(wlzsioGt#rtc#4k8ze_Jn&}{n7Xb3@~~p5|dpG
z?k9|MH#w+WG>VbhRmP!xdoGppd_dkkrAEt9Z4yQ%XG|;oYmr<an8dM}2XZ&j#;Ehg
z-B{2di!3}QM`~|V!8~@!D2o-G>yuGwSaWEI34s;y>>o{pyiz|Pe&dyFSk<zq;3Zim
zp;$im9-Mi8-Rr_93=QcnKW)2DmzZv_`)hD;aAe&M6+)r5!VA54D{<l4tibM%SW>yu
zW_0`r(&)wN$eZYQ#P0`hbe!Z|+%Jz;VJ1qD=%0Ufmf&aY;LTX~VP5s@e(|E1!9IW^
zDrxn@sfDEChTY)nMs?^WsY&kX7)$k0RF}bn9-dq&dGLnK6Lk)8BiN9XoIy<dA1S$^
z5HLln0gZh4^XlIw`mEZ<BVXQmwBq2sFNHmUiLnd2UMso#1cTYr8n~I5nCQvzt7gT*
zf$DZxFi^bv4=#Hgjojti1~Yua&o9PQW&|FQ7iy#?R3WnHM5pfhd7hMuNze8j(KOw!
zrRB6}|GL<i*X)Hp>b(8(>3OmVJ3G7e9Q_CKxNTfioEhlFfI%nBmq14B_vS)OIeYfk
zO9)~xgbW9>ByJ>_L7KBj@H$daRY$=ckyt0~IlB^n>PJG<#XXn9iuTtke$}EpDQ>K%
z1rcng;5kC|0QltU2R$~~4@YsMk?ZLpOxL?C-)XYcVK}^d3qc9seOqeC2<ygF1{2_f
zEb{uDqf*l*_kWtHU*WG}7SnNc<!@<_IZE6jhNS)7YVL)XJ@eGerXvlPc6-qob`HFC
zB3mAT<HwD>T*EW3{g+>cX#lT!)ublyB0)@NkRkbz#^v7#;=YaUelKUJ=uCnmMRe;b
zlIGrvmda<ii0KC!Ru~%_ugm|^DQI6J=rwWjy?YJUbsEi&&@jUcuckQt0GHJtY95Mz
z8!u5Irx_-e;8!clU{=%Dzg;^x$-f}9P+Qy5%H+-$He$@&!lZ+FHASCj)gDsxZWY!J
zNs8rbGUC&+$}(FyT7qqw`VnI^1IDOIWKi%=V}z(NU2ijjU02?MnJ(*fm(W?Qw3|NJ
ze0;|iwmWZ0+XRs~mQ~Y2ifOIXbe?Z9Mg0lha1(9k1O>hIpHXbH%UyOzK`<8i5_}+>
z^nrNXfjVB1fGXE+y!4|diRdQZu2*IhExzl>&Mk@$ObmE1t|5L1f{2EZk>p)rZQ-!Y
zL^<$K2;o_Iw4{3V-gYNNx8W)?64W5!q$)xi@4)}SP(u|=x2H0MpxvT_N>(rh9jz}5
z!(YQ!gpZ8SUFn*?Gv@UjAyY=z&d-PI)mX@K67K9ksLL{m>UCz;*ZpW*R9$_6fdp7)
zo)!G|o3&xhTZ5H<S6{~v^=0&ujez(bx}ZP5$Wgq{H)N&glbNBcUXpafwL0VS2b_r|
z2Jv+5kP;C3`QhF!sm=OX3M036*WfJn1kr=Q9Fc)$f25O@ksiis;m+3t0k>xra33O>
z9Di|o-#evZ@x2z;-|Xg$dxFXt*Y^4NUW7-`a@QRx$G4An#}br<zbr7tfQ|Yi?DSR_
zq&$-SVjeGEf@I&SFC}~r#yd3iYM>w!j$!WvYy1W@BxkNr%N94vKj6>}8!@zN;eLvw
zafD*OfB){cDUH*oT5`A8?UVOKtj)GUZ1#_@IDN87lSJDiH)mPJMo31$%`MeW>y4zy
zj1HC8L5sQ8i?(=Kh_JR?^;(YFR6giaW;sv2Z_QSe|1P#@Q-tXHCROu4I6)9_p}HS6
zLxk|UH+mc9PQirdQMnkU8_MyfL;ba)OA|bS3e~hvwDm1}<m|1D4)5vidgqlW2;eu)
zH)xfyIRhAC51V*PZLV-x-jJxPacQ<wp#VYE7XUK(Mx;6)2uvbqMZ)Q5kp0@fqW-NY
z2`YxDb?x<hlG?LMQBe;9SSO@aTR5XjZ1*MRXmAJB-P=`VPJBU64tD}H8PWSu1`e=_
zk%(2qyoPYKYHcTBq<y^)+KJbYUqj^P2wq{_**z8WVzYFu6VWNTO~hvit!fmEcUR74
zwjY1)fq_zglVB!9P|Mdh=_W<u0M==b>xzRPGZ03Ic7$_vXqw^Y`x1@>K8QLJcsL9o
zh&Jqf4k@#QO07+vNU=S$sM<~BD1xdbY7u+q4q2hPutrBh!V?@RMQdivdBL-Y=nQFr
z-!J@pJ3)_b(5#;hWV3CS-<6;tT2b;GOu1CDmBkhX?|n-eeO?XyY^*$>A;D^`&{Z<m
zZ_?RkL|5Jdo6kQP5pxj?QQu|YJpz&dT?eodiP*2$p&OfxN%S*>_v|_h31|5hSki*o
z!`z`-raU}6o7uhITcoa$-i1QgZ?EEo3vPnGy_*72mPFL~fNxVFU-FJ(SFsL7fSc$T
zto4Tn{sh^X{Q6U2Awi3|Qss^>-fMnO_rvqey_XGvgU0hJ3K0RMZErJzLx-UXM~`Ja
zOv&P6F2BnYEC;|ATJUP^%r>%<k|0tCYP^Nhw;g%Dae;W=F-6i`wQBMN7r1^O#>;V~
zG_gmtWhg6~L+{PIBGt^-rR5i?voqhVCWKoq8EIbV!@Dgssw^wQjQiFeie<k>&Vtzv
z26@}v0{lxbl*DG)TJaya-xH2=2N@>l|Am_X*+j1s^)$CTie;@uD!PakOEMK@Nq#?1
z`-+WPRIr#y!KYcvX#ZN_&*7=9Kko#e+Cg~CPh`W_x}Cvo<VE#GaG9=WcsCOa%W`tO
zV1$zr^=1g4Q$)7^+3JH%3Ss^?+RK%*idI(5FCW=dlP8CoDk&+gZ_3J}*NCq#R<G~h
zi*-Hh`9bk@D3jx#gVzplTouH0j0}emwmL&;bNaz@TB*&9?+tS)Loe3?wfILTwJglN
zmP6zo95%)AYP2${oBy+u)Ei|X?!Cw#eN!lL$q9e>heFzlia-+Adw5@MyjdGX(zy~?
z%*<4n;qbgHB2hQe;^NB}HK<<5cLqbglStNg|4&6kT-kP?d!NdhBo_$MCrtyPdw8Z9
z!GW3Kf&=O6rZ^waCnHz~!u81=hFddD(2614TE~vkY4_|GbVY2*jhW_q8tza;nHz}(
zWv?{3n_pyGE|z<{ThUXSKlw49X)!&dw<)ggDXWxW$38;e5wb25d>bK5%kHUW>z0eu
z_`&~APO-T1oZm~+M{wv749yv-`9aV9+>%W%XMp9kXnA<@N@m_`C6quVM;76Dyq=Lc
zQz9ZIdoA)2cJ*&Ql35*0B*11jLa^vLM10q{wh;a#!x3Zs@N7Wo`D)d+C>Q4&lLuaC
zI)N)@X5<S~E%iSA+HJr<v31{?%>U0XsE;V=7V%u=kKz3JrCsZM0H;e8*+fA@E*+84
zN0He0@71AnKqzPSUw*s(F(P)EYfoixl9dn6ibGkV;0CJ^en4;#`D-Gm4q2(qlSdjc
zIg?Elp^cS8pN$N24mFdWh2UJK`)8oTve#4yD$M}Yow6Wi>c>HB(t_b481*+&HvfM`
zTSGY|JyMHHd#*;!OpcEYR)!cm>drhZ2?oP%G3t?dGBYzX%1nKL;7jR$B?@4*i5Y*P
z5H^h%A2a~B3jUJ!1{r{S<}xMKE`o>?&9e&}m0VRC5iT!JA8&OYS-G$p5JhB3b(xYQ
z*SI|)A%Ust;YC6`$3Tmj3NbBYKFJ+u_zdB?3}axr$YC`?08C&QQmR}?vC}GTOrgX?
zlB#+3J<mK(vo6>2t#r{TOVc~XSohVtB92_*>v1QtcaKNC4G7Y-0Km&n1eElLCHnJE
z$`J2W)Q*dgm@cU&=KI9EvMlO*{QY^Ih9vJV4KlUWwa=C06iEQ+h{YVa9TyLIuff*#
zKRlkO>X3kL@Y!E*j(ShbNyU{qdE>R#T`SjzhG&(E#^&bNy!|Y9$$|+dk&=SV=U6-O
z9KDz`JwDa%SFEihaX!E)XSw_TYwz4+n!3UOZV)GP3uGA+){1UIhS>(JQwl=?2kK;O
z3_(F4B817K;IQDMl+ub*nM!$thXxAFDKrDy@+h=CYS9S<gKcbw1zNdlZ~_G?6|Z2s
z+wQnoYQUIff1mr;&3ALp`}dviob#P?f;V=?ewR==_&y%5Yz;}EMT?`9Omgc;1Cl|?
zf3(vS&x%2QCQY~K=n3wZrJOFA5#M*uQL$g#6f!%i@!%ki3W~UjT#KII3q0F;+i(du
z9veN8pOu=%B!fk8cR3Y}*$Uq9V{+onF^*rLDniLrR5A%1<XTAdQ4wt;-?M6zhiDGx
zb>P(_dok)l`ygYJ5*O)veZAtEH1Svf#yKu0R`htrrmqVYiapEu^f3x<il6Dr*`V1{
zz;2oW$MlSl1{p0)r+s2(>d@afIx>4+Fqlr1tW6k>VD#<(I3=E%dQaObo7Uy&V-9;3
z)ZW*B_(KC-h$J5solEtvV+(v+1!Y-{2fC>a)w|UZk#9=xa*&nGc{~F-JH)xfuST`|
zTAWN=0Sm|T>Zqfub>)nol=X6o=&o2t1*KNSpP;s*+#3{j6s7A=gd)D%PBzN)cW&dx
zK$$4Noyr4?^le!=F<37hXGrft97`f>1&LTXhrm3BQ+7m%Qb0ZfF;3glF#jhl`xt!`
zo}GHcZt6a<U|{uRg@dtjasfDA?!R=!8^dahwmj1}<MnjkTYF_wXsSA?c0Jyw{wf(w
zos>=~xV?FODV=3{G_>64*MSQcuhp=n7aIs_x;~pUMHW`mxvcU7hh1}6_ue8*O1Ef9
z&(NAltU)1v+?$$GsV0K+&2FL1R7^kiB#AnsaofgTcQUkZ_IYpO?DB`_jnR)TC`VPl
zzp!A=yEdYAX}YkI#d-Yh&j9(jt#eE%Z9Ml;7g_+A>d$e^EHe3(OjK{6Y^$hB#)cor
zWHRH*)wW$M6D@z)2xh6B3?K&Xk1+`C(Y_x?UhKvMllOw{-U@Ep9=xeI$-}#<StULF
zaM<HctVE?+A7Z4YkjXmQTtulw16XX-r4iWg0p{06Nj=L<Vc=RJYCzCrH=JR%DDSd?
zwdT2#<yS-kL8oi%_w9e%erXV|iY9bZt}ft^^FMw4^T=~mHmc*wA&){4Uti}^Fd7Or
zBJ{qOmwntVom~1@GVdNcFaQqh(qDLK-Ez5>u}Idq7HK^)%npdaBJ#?02}+c*zL-0g
zIW7@xp7}l_<MR&MvkXKpKSHq9cFjqKkDM|@6Pk-xnLCz$ZS+?b&Gt^Kgqb(aFNXh1
z&-lcO75Ma^9bfMKpwS_>F(~U8(2g6<)Yxfjulv;6QwAPB4U=?UUDV~~;<O~$EGY*4
znf0PL$HAtutfZ%>M}Uf)o$LeR5)&UfpV_^`?1}K&#5w_EZ1m-|i0pT4*usO3c)X!I
zR#PADo(qn|w!2-{l=dIeWv#IyV+A#fDYMqQ6}j|Jj}`jUb>PMaO#8LL`O_ogRVPBu
zhMT;0d9MmIk0@T%0L+bK=*r1=Rr5CN#lh|NajSUwW0H3d7U{$Lbyt-2&CCZ4M8U;F
zZ5Hir{&{)b2=~PO)f>#U#5hsY4SZ)ALU5<Ec=Di~Q|ol;Y-Br0eH#0O=eJHG0{#<)
zw(KO)i*cYOcjoGVlGktmKP_*9Py&&GTtEsS1&{(r0i*y@04abJKnfrQkOD{nqySO?
pDS#9}3LpiL0!YFCrl8h-!6cShZH@{^lCJ<y(9Vz@+<>@Ke*;LjI~f1~

literal 0
HcmV?d00001


From e496f8615a0ad9bff02d13cadf6b3fa3b2b77856 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 17 Apr 2025 10:14:33 +0200
Subject: [PATCH 314/505] Update README.md

Co-authored-by: Cameron Pfiffer <cameron@pfiffer.org>
---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index af7380b0c..52fed3bfa 100644
--- a/README.md
+++ b/README.md
@@ -61,7 +61,7 @@ Outlines follows a simple pattern that mirrors Python's own type system. Simply
 
 - For a yes/no response, use `Literal["Yes", "No"]`
 - For numerical values, use `int`
-- For complex objects, define a structure with a Pydantic model
+- For complex objects, define a structure with a [Pydantic model](https://docs.pydantic.dev/latest/)
 
 ## Quickstart
 

From 35abef0fc61501bd12e5294244c2c88e8761d0fd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 12 Apr 2025 16:52:41 +0200
Subject: [PATCH 315/505] Update the README

---
 README.md                             |   9 ++++++---
 docs/assets/images/readme-dark.png    | Bin 25637 -> 38912 bytes
 docs/assets/images/readme-light.png   | Bin 22109 -> 38943 bytes
 docs/assets/images/use_philosophy.png | Bin 0 -> 16463 bytes
 4 files changed, 6 insertions(+), 3 deletions(-)
 create mode 100644 docs/assets/images/use_philosophy.png

diff --git a/README.md b/README.md
index 52fed3bfa..7aac9bd9a 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,7 @@
 <div align="center" style="margin-bottom: 1em;">
 
-<img src="./docs/assets/images/logo-light-mode.svg#gh-light-mode-only" alt="Outlines Logo" width=500></img>
-<img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=500></img>
+<img src="./docs/assets/images/logo-light-mode.svg#gh-light-mode-only" alt="Outlines Logo" width=300></img>
+<img src="./docs/assets/images/logo-dark-mode.svg#gh-dark-mode-only" alt="Outlines Logo" width=300></img>
 
 
  🗒️ *Structured outputs for LLMs* 🗒️
@@ -57,6 +57,8 @@ Outlines guarantees structured outputs during generation — directly from any L
 
 ### The Outlines Philosophy
 
+<div align="center"><img src="./docs/assets/images/use_philosophy.png" width=300></img></div>
+
 Outlines follows a simple pattern that mirrors Python's own type system. Simply specify the desired output type, and Outlines will ensure your data matches that structure exactly:
 
 - For a yes/no response, use `Literal["Yes", "No"]`
@@ -607,7 +609,8 @@ print(few_shot_prompt)
 ## About .txt
 
 <div align="center">
-<img src="./docs/assets/images/dottxt.png" alt=".txt Logo" width=100></img>
+<img src="./docs/assets/images/dottxt-light.svg#gh-light-mode-only" alt="dottxt logo" width=100></img>
+<img src="./docs/assets/images/dottxt-dark.svg#gh-dark-mode-only" alt="dottxt logo" width=100></img>
 </div>
 
 Outlines is developed and maintained by [.txt](https://dottxt.co), a company dedicated to making LLMs more reliable for production applications.
diff --git a/docs/assets/images/readme-dark.png b/docs/assets/images/readme-dark.png
index 57951ceb461b031f6e1981fb2c7280b06bd80acd..d4526256d462dde768cbcbde886afe07bea2769a 100644
GIT binary patch
literal 38912
zcmdpdV|Qg;&~4aJ$4)x7ZT9KdIk7vo?T&359d~Towr$&d&vVCp?~k}^jLosX>{?Z|
zs^*+4OhHZ@5grd73=9lWQbGg>1_luiy7!0u20B0VP|t%daCQ<Jj$mL&eg7TcL$)PG
zpqIE#qUugcwkA%l`VPimuCA_(<~Ej&hWd8KjJ6J@8JGNcU|_^xk|IBq-7?R!T-`8L
zUWbHC4;R<(rT&1y5fh7wKw*j~r{_|a8(dX-+Z$t68)zf^wmn;|uKFm>uMebj;RO~@
z-jf#`Q8rCwZV{yfl9Pj@1K{u6v-y!*91k<OQanvN@3Na<X(+z^zu}8nSc=H}zr!pI
z2Qb_He@CwF|0fT*ai}ohrw@+T+K`|Tfc~esxw*Qiai@y7AW_yj!0+e3Jnj1=-zuPz
z(!bE|D10u$2I}*bI@2Vkqk8kX{T>eP=HQzH4rXR<Mp%-54yG$YZOv-isinEe<(2tI
zc2|)q-Q?xW+0^y*EDdos!B(AHjJhz=oRsrayN3jbV!;+>%6-`u{eY?(YisJ@SaxPJ
zVn8VE)?bGkbw)f#M@PE;zjG)279_Y>SgI`M+S-|cWxKW7>JWn|%iZ0%GtI9-<KW5F
zsmgijQ)^#$zUUo%CT+>|8oTBeZRvC@EINHPuQ>~-H$iV~Og!n125S^u_c7A>REHQA
zCGhnzY;_L*_&oKpeMb&AI-S7>p5#V4z1^d%J-wL&jUdx98O0!@iyHUT^I)ZlkP{Hn
zSEG}KjlnfE_>&Qa9A=(J>8b*quQof1H*Bsdsinfp$}ax65d#%5_;Mk5w`ZI?poZt1
zy?u#MmQSAEnoLYhtB8bryecP55>PNwlY)#Xeukm+UMu6_;QT<R99Q4sBd0yFPvE#G
z4OG}@(W-Sgl+`8=R8Pvz{xncpTUv*eKF7m0PmynkQL4gaBOZVB@v!&Nwckb-g_TWt
zHK94+8sKTePSN}Ocg?9)3R{Yg<o*JPz0u}c&7Tx+rvJc+JYqfWB*Q#MH4rRB{9+_&
zKa<nu?TTe;>I3Z5HOi11fj0<O-rvkGU&aMi-g6CXS;LNeW2KNa0vnOF9;yLIaZdc8
z@#JhEnD_zS5HW<U>qb)R`}KLGNuPX&A&N^Lc#M>aSDg*z^7iq$<)ot*pOBCs==1(8
zU?-uR>`7ZJn?V^>VGDAzVCKs(J>HO^Z1xGGh-LtFc+2}kK1Ut($}@C&5u$dz+01QG
zeN78$dTji3al*k;S4_0bBrR;U#lH|6NvziThV5z^?hW2EHUY1<y1OTXl+1u>u2bNO
zr-kWR?`BrHzHS#eC!o;KKdnRolO%V#w9ce0Ba6p}k)JHGDmm%+pQ{@a)9C0kTCU;I
zq!7QI@p-I}tLF=I$8n{9dDfSCcvlAr7Z)fT!X1!}oWW0=cn1>X1#w*_d5*NLI5U{k
zRGg`Q3nWWJ5=@vlif`W{5%3<<sg|qnJ-u}e4-Y?|r(0I8q-12!?d|UF2g0HE8^#Dw
zBEw7Rhf0F}{_nx=e9oF!zpFQS*@rn7P521#qnXMQ5je{KZ=to4=G@YPbgf`Lzsw<U
z{=gGjgYQUHa8Xdqdp9~FKmKX|Tb4g$+uF$RZB6TVzyiJ9GG_3-%k&rdTes^WHllh{
zpQDn!>fr5ge`3Zt*3Am$FS5ky*x#RZMbd{=)D%@BeFrl%jC1a<3H+BITJ;7j5l#7P
zltY?MPa)B;Ar<|FdpD8=-7fE2am2ymwSG8+Ba8An5Vsk&ohR9;sctD{Wolrk9CF)q
z!T!z)^>CTE3XL%9Fx`6J!I5(VnEhxltQ3FSLgo>WypxBXBF1g_4_OXYnXk6b&d#dA
zAmZHzfZU&khKBdY>LX`k8{+rxrv_H1v&+NHmrohzajrCFErSXS8(QO=Rh42EeqYk^
z@?UE=jWwlVW(oXuZbC!0doUq)uddM9&Yn_hd`}m2kx~*TG|KK9Nk_q7XNedY%Qag=
z@0f-4Q5-BKrm)tAG9ui{jh7EMbHS|iCPBt$W4fbj881^H`@m(P%i2br6XAW1iE(wr
zSA4(3^e+R0RfaAI9JeB3@>LQJ;bu!<jM9JeAri86a`#y|eiRR))!#u%8vn@FebH#+
z<Q(ucbR$RnZjq-goE$Mc3gPhi{!%O=Dk@M=cB)>MG6SF3^Eps$3vTp<f)NX>>LNL0
z4#fwvo=U0j0?U_2g>hX1A1^D5jtULMX|vgAO=3(rD`lJFcbx2TlM>+INE#yaEPuaZ
zfii|fL8mOaIQNsNn<6NpvxuO{GP$J#cTA;Tf1hlcENUFcb!YL|9onhU36Y637y2D<
z(I~dSv6xUt=!B?aI&xE6Rf-Cr27Y7O4DTkiP^;F2;*6*+O+WGLF?GV<s!QN#kq?jJ
zRHXC*m-I>u>*@HloN;#aH)vp;l{X%uc(9XS=;+ZoSK18SxV6|bhOr}PyC_n|cgS7y
z#IPwkEWMd;iD@%SFK~er^hIw0d@TY*`O0OHv=#z(HmaCda?CDk)J;I2_0f~_vO_q8
zQ)E-@tNx`<ch*6pc*6xdVLy+k1|O9qO@o+sPUM?ZA2WY^MZ&4>aG`Vxtz5B^*Za-T
zMVr&fZ8OOK&}8D_NFl442pWCCJUd~0&G&GzH#OoN3UkLEP&N;$`+%c4ptYkol2E|I
z{QyU*kI(%cmP91v-yk_1Ff$<5l$Zxv0vrIwmXzp)APmz^T!&H)phc)Ty<^?VO4yhW
z#$`wRE8b-gh3kBCK`DN(ykKSP$bN)D@0&DR;aISAcJb1OEcCvz%*n~Y5=J1uSa{o&
zpDFiIsEvjK2p#U1({r&j#U;Wv-<@aVl(R4;DW?9jMz@l*XWA!+obmL0>>D4|nPp8K
z{@b)M#x7(Ox;TWC!U+4@@F;RM)7Df|7xz<%4-Ye!tLfrMhn$CzvhUCyq~|iKu%D8l
zZDr2WY&h<6-{f{S3B86|^4*@t62^9>LYu>$64$_9HFS>olBtKA>(3juHmk;vgJ8tY
zl72ZRd2Xs*BD$tlQ`;5n<_2~uvf`Rex?d8UD8D2crD)^LM>}GRvpvdQPKt9n*9>a-
zX#-&^a88)~CTx2Tzy7e@ach+V*7T<>;|n(I6lK@Ue7bxW!!hXyv$^dJ^j}<_j&gi;
zFY!1!pE;&aRu#j{ivD3V14!k;^#yM=7*PBWMt+DcWv|?vv}4LzNB7+1eOcS4w>@xw
zSxB3$8<q{s+?2>C#(!*mWC@RS3Y?sr5NN&inaRj3*>Q-xm3qb(J`wSK6k}g()v&h8
z9&N_$(0jkPKAR;-R=@qpc{hkU?*;qU)Ka5?;X-g-b#Li~Z-b3Y)l{SY%L>9*I35La
z{r$r_VNFBfv`b0cMwMfL6q9kLKwT#@>!lP0Akui^)X_(2BOl4~#QnET;9Xepk01>v
zTE2`|YyEQ*$Jz%v)a!>YAKy91f*#a&$9@T`8ZIjpq$iPtB*_YGy-!_yO<&LAD+s`y
zr8dS$T}%Td5&crt*A=v*ae*|xZ2G5aLh0`$E5e}T6lI|!cFXtcL(<;fzD`+%K!Jr;
z;TyhQFyfSd(>5#{1SeaddbbJ0Ke%z)5_oEuha4z&0tnt`_Umum7ceQv@8UIds9a#L
z0`>|DgaE0QsS~G7mqiWJ!l8kIlVI0cllZD1OK|TAsM?VAC6#E1KcyhTx9k2WQ(b-T
zuYJf`$NbzQ1}Y`d^;P1T!bM1PX!L#FJ_snJPkh?iik+^)fNa0S6&xii3k1e@$TVeA
zI_p5c^bddP2guTTeQBLQ0%9>y%lLg)R(596k@VgbcAY)$Z3AP+;==zb;m_@*<e>Mw
zCl!{D*~O$#f4n)>@^hcwE6AMYN#FW|UT+?H%Te!Uv|81q?u)zXU~DL==<DhJS?S_U
z;nC47%P;GM<*U2I!QXJ!%g#Zbjo3mDzm2XuxfrL^Z9ecX=^s`|!qi%)*Hn7EOru&G
za*?X;$Kd6#Qteq1;?q;Z_}II3R?ZpbUiCk5)`uPG>h^XN6<66|FP%k@R0Ed;-b!@K
z1W!aT5i6=l!kXlh*Dp}ez<Lm$P;FyCg&6m6aLFl?Xt>k0IneLd8jn>X!oj)2aUTaI
zR___AvtVk!S>#Mx_)GAKEECVeBf^#5ear2K9I&B&{G&VM+W0|@^Gqt@^v_LFiU!iJ
zTbilkezwo?aQp_(;1*@Wwv&7OXr@SNM4&$3qxykZTydX)PcZ$?TM6I5&Z@=WH5m2r
z*d^v)W3}Hy)G+l%=O6zqU6G{+OKe8%=NT5x8V!_9#zL*;$wS5lD$YjQ+8J)*hBky+
zm(IoM7P&~*^6d4us8IH`X0RYdh)ES$a_~^8CLNVp(?j=KO3|9KTc)RCxR5e7oFs$-
zqaV|GEY@7y9u{bEhJ%qcn~qZgQ|S|jYjodu*|v|g^uO}csen-Z+{BBh7%@Ze|KN53
z0Yd3WK{mq-LBOjW!>7rR@=m+aU(=@W)Fy!_y;huGcY0s1YF#^hA$rbt6I?{wze%GL
zZvz?76CEB>Nr2%^ewDJ;(rm;(Pj}Du{|(7(zI3%PWrya3EDi>yoqb#GTN64lTl?_x
zGgFYMqVi;b%8}G^&X>i{bNJ_Bz%M_!ATXYE8NP+@Km@AIIVdO57g2&DoO?KuQN-*b
zPr(J*0LkD8f1pcC4dUNqzMa|Cp=by9`_L~kKc6XM$Km4Qjz;3L^@bo3yq1=g_0fL2
ze1`SuK=7gnkv9MmlYHOp^Z&AuQKq0)HY89u3^+%lc2bFQo9C~7MO|it3HG6flSYO4
zS|ss)z1MhrSklwuw^?syXlZGQ|4ljwM?ooCrod~1>sTesKT=ciw2~k98xU@qQ1!za
zBHP@;^ZV%#rhLV@anxxlUb5XSH%a0*svt?7zKl9TnuDAOY$h9(9prpVHA^G&-${uM
z`TjU!5v2omRww|OaGN>zG>lhER5$`5u+7RPeyZV)z0h?oH78}+^`LDr@)~i^?EHlH
zd+L_D&ND%7`F?X+R|s^!<JU>Ya_+dsMoY9f+l~iS?X6C4Qv?z@Na4F)TXg-gj1c);
zSTl4a>t4J@=M4pI!DS_k_0P)SdJ&Sz7vXp({R4a{8#%d2QILOV5MW_Zf=pMIL+#x*
zCGey@H`y4y6#Jf)oaFA!KO?N0@zTYU^TWO5r{m7-d=Bu(%?2UI3RxvuxW{;W|9eoH
z(L@V6&A~rVZu#O?z2k%lM9G&0I_Q@Cs-YN2zo6iQFo?(V*hkj_cXUQxHK+7>{yUCM
zq+e@v_*_LlIr&~M=$Lg4a2N&5{5s{QhDsqTy&I#-gBr1bI`gvznpWl8q5Hc?$N+ay
z{k?uSC0YTY=*|bqGhNc|2}Ytz5kv#$pRxTV+4aj8!$`?N_PA=llk>_E^gcu&@c`eR
z2ngCEIU4Gt3a$P%RI=UZ=$O|H-G+cHKZy!|pF?uutruh$55ywmw)=gb%3`Rhf|XFg
zV^aHN-IR8h^(F{G$=<-g@CPl=0+LlQvk8P5u#{v(5NRf&%v3G~n0W4205}NlE-5F4
zx{B-;GLBmXZFYKtoGHWV*-3ZXoN$H6jn1Y8k#2IX;HMu?G9>`4e>)~)H<4Oht63i^
zGE?fFjMkNMTGD59uAgtaN#jU4fPQq)B@NH|<kgzJTmWWYQ&daAu5t_gDH`)#9wTPU
zHV2Gwl%9kQ)I=FcbaQGiG~n=4L)hp^pApOTabn@FtQUX8Z;09JRCcTdLsC7rouFYs
z@7M84l(@n+5^D?$6}prn*kcsTNHr@Pkr3f#L>EjbQDYup2<8)lwkHK-Ha09YiHNe7
z*ZWhQ=@g-$%lR@QK9>y!$SOcs5?)>R9p69_(wJns($X)?($b?b^SNmE7L<Ts#~nz=
z{QR}242(7RX`!Hzbu(;o`A0mkm=Pn*MRiMdOx1|bdr)0NN-sh(U#T!2_Mo`cI5ibT
z#t;}neScxl#zR@NhyP#Q{juOK$4$cHoEI7!FCM!tdvT)}Bw<qO-;{#~>fDS3;C8wT
zkF`&ui7hSCpJO(9|IG!D=BE!%;m|%osLALM3nu!1L0sV1cK(bC9VK-*p3ZBmo8&!H
zA`uY+8+GYxzw2)Kd#b_7<`f3)$;QZF2F!J&4@~;jk&yJTJ(vQOjy>rW6=1+h$-;88
zrcq~-<v1tHbx{a}kV{zEl5lD@7&nS-Vs9W|h|D8sq9=m!T9^;UQ8u10V9H1LjC@sU
z2=3=P<OL#ckF5Mjbhvnbxr<RaDmO-rOJ)5n!Sq(Bs874~aaz!P-;Wjt<_#=wzm)}l
z(!uH*{>+O`(%L`<SNj4PAaxo@5}2CKSQ*Eg5%(UEE>b{@oEjMt%_cVbhDS^KkBmgI
zw@XKmwWuKMO^GW^a|~RTh~|iz<R0~Upr*d(5A8=Ss7y|RA@aHwRZ*FO?9M{Gn-tO(
z30~XY4?JWbr9&rz-`L&P9}&v<LKyo}A;Abjk?in%`h34^yK*ryGNOhNNbpz15L!Mt
zMVP|$r}_^d2^DsiC@9KiehNzlhya>?Mlq0rd%0)Cfpc~D4(gt{5x2TD3^rcOlkv7X
z2KxWt02K!=2hS^wTSE_Iq<GF3htEzv4lr~n-{JS}h@_5Fe;0|1pH;hDtOkjYy9f%9
zc8={>&sppQUQ2A|P{tS2Z+8J%^OH$@pqSo9-FAbdEgmrl4fFq`8GE*v-xAuNjmOmn
z6Z5-yRMSII%E+R-O1OcO_0|7Wj*g(mjvGs1>YB0Q-1-i+hF28j7k{B5ABp6bh=Zs8
zbMG151pQtj)=H4LtY6l~OA%$Zk`(0MGheHgRo*u%ra2uPTwH2a>&>YCOC7hs6NcnV
zF1;U=<X5BU`<kxjam7Ve8qQePN=jM|^b#EvzQe{x$I|0vkqd>)l3`_MCF#-d)WG~>
zw4mQfT+m^$Fux)TxH*y4-fo{=&b|LtSq=Y4u5P}I31Sf6{k=l;N7Ev^;YT2fl-Jw7
z&|oH}4^U3AyU9I<ONeE)f6Wo0#lTD?E5((7I*#{kM!n3FN2z8O4pRk+%ex%#`)4kg
zAo&YS#!xD%TdY+wIMxXvqtb;)-n3ZBvgHf6xp4Bl9dT;0r^27j@H7dG`+z$wRC0Js
zM9IAQIE5M?miVMu@zO-?yj3``XkNiqZ@xA4*2?e2(o}^}6moODE>9O$5gHvE3}kDC
zk+iERaV`Z<jVgNDK_~{d5TBLZb+wH)nEo7qqrnrh{no!}j;8?BpMTg2r1VPXbo@M3
z(9=A?Qke5{Q8E9>3G0#Igirnn`0C=Imvx1bE-DH$g*SA&kD4p!DzcvwA6_ah3Ug*{
z9&l3jRzyT9cF-=EJHaqFHy8K?0AwQ*&g-u@$9A=zO0pnHM4PvK@?>aC3xIPTDa-b1
z@Q9+=#~1XegXfl=K6HJ)p4Ean?K?7mt@V5$rEA2%*$mL{uPwBx9S+)ulxA3L=HsB$
zQZVuxLwGm}GJ6u@8-k2?Fe3hAm1qQ(fL;zW8cAUDPE>DR)bCY&x7tPGLMJRNEJh*1
zGY(3ilt81-?EqlsDMMT|XzFDnjhZ5{*?Dy^RlUBbe9>`hX>=HNp$9___$F7N&6#a1
z^a-$+gbUA)U8eMVI9o&?UjHQ$Q&}&65=aV_UWAT{I+;omF#Pq*=0iu_y}LHjnZ-y*
z(LT5~ik*(|VT69h`cXp3r{{Ak*u|hHX>uq>uZn=Wv%{^!xaoG}&Fg$3^Ry{|3dl9b
z!-@!!BjB^E9O?HY^6oWUX$y-|;HDaq@jbXkoYv!N!JXca>#B5VJbyqtQRvSy{A+-O
zVKJ_!!FymPLSDaHe({zZuD&qECo#C_r0tL4fAlNd?OH-(A)yodVqCS7SY5)4I0@=G
zZuCn=lYTV~avf|PntZV}bernrs5b4du}g*ToWYFZ4VxOQT;<qi@N{@xqv*l+iwMTn
z;n(B8F`v|d#|Yg-bNa>&p3(&bt^Ez!s2aFULe!LyNxOgO<CtA|0@ddC7$=8^ahUac
z1IH5+F*@4c5E<s_R$jtcs7}Jwkte$TDzljmP1-oh#m_~1$7~jDClq^7WqQ}T|C^9;
zQz{qot%!O`;Z<+8%hSy195STzouSMd67|GfDCZ3JJ~-|KK6*QIcy#j#2?%`n4{D#e
z#V5w!4jL{EY_-}s;~IVoyj?-2xIg}VumK(9G%w3xV%??BxVO)CuKoI?=JwWs{IdVi
zgoLjI+}$sC6t8`JGES8Vc{{f0G$cj8YEF8}?n>VJw6k}$k%($aZmMTxY^Q}-`Q%N0
zzeacL{45{rGMrzQcE4-DK8ba5!yE9wAHYaStp0t8eIK?lSQI10BVO8?$D{Ui9k3v7
zP#*?DR#rN<LNtPGBvq3?cO1Oi3hBj6ObB(AD!*wOpO4MI7*wAc%3716g-OS0vluqF
zS23dU@aTL`&!qgEveKuDH}(7ZF3NOycQoS7CN@m<q3?us4W^(OAsY}Bwsj@Xm&&0H
zIh9T2%Cl_+*Y$;oFXt4<U2h%-!Z`T}W8Aj=U_MnyNRi*a{UfUIi0?y_(KJUSMf0!8
z(%CJunR*M4m9F~M(ThqrQ)4)k5)l!Rbn7Nr^7`38?i3WlEV5sd>v)@7V=rsnU$Z+3
zvvh=!ILrlFOkUA84-*~nv_v4>9Jxq1h$<>FqUQR_&i%nOpusy|O9RW_!LAB1?LSK!
zs5<|-{FF?SyV4b?PPI%1+u6}~I$JFiLxtOR0l9co&&2`@9Yn>T;%{__Vy$A&rUW}~
zfJH;8&RS#OvgM_5{uys;B|2_KU?ogbhoyn?mjV(7^s|8m3kWlLG5?!UjS|GrCXf-W
zDgvOxh*z5J7b&eyU9HW6c5mHoKOs9fYM5Z*(Yeoxr<@Jkrvl&O{zd{el|!=u%2+4i
zL7T!+EOEjx&_1cP3Ym&){BvTX5IeuQYn`n)2$0Kc&8O1Xit7dp(jbJp5yvX4=!?3`
zzlo~#WdPWjddsIt%EDb0_bdK9%+a-zg1|!UHkS*xm?1I;M@LZ$3yX;LhGVafWh7!e
zj1Ut$SegeyS?j(UBb}np@+GEBS!;3c>SfX?0m3s-jsnXyA62xrZNw-z{kk1`s1%U2
zrv#M4^N505ne_azQ4!V1o_7~bzRmB5<ji7thCpjt3z$+fgu(*@l<ELi$~oz=5Wr<;
zD=X4m$#mMObe?06yRlm{Vt#FuY$%UgTyOq0Zy1rCc0#OcIoy|5QMqyBS0AU`B~#%f
zw`jL?t^rh?a@|?R*LeiT>g2NIW)R3GSgjr`)WJXCHZW>KR)|Ry*`Ottn(q>9Q<M}|
zHxrAeWWJ5VYWx*vaxR6<N~hEnt3|`n9s6OkEl%SuFSWU-EZ1dQ!+nZVUK}!-MS?Y%
z5T9a!i^kxDHBUil`aNC_fXncGY9E#6sVK?R+>@Vhrqlbac>A~3sI$EA$YL24g?K>W
z4AFB&@Uy~CCkYkf=10jwhmP^s*w|t!xlA)x-mLsW6MIX+4=jr|B%F8S0zeyxA5(X@
zsCS0Bm43EWT2mwnv><fZS40%u9np<QWVJn+z5hE(WHu>6K_wpk`7r@D;Q5GAzN+9c
zSSzGWCyxBx_%;xSUvOk>YhR7WwBg=<vV^&46K|;SSMQ9|^@~>^H(+?ZB|^_v?`ImF
zdKH&Y7xA%uWEf?x`T6kjNEyeOdW(nbfS2k&Xcd*~SMroh%t4jCd^;*QB#A7mQWDS7
zbuY*7Tv5>gDTP4)3A}QqF1Hib&JQ2PEibN+_V?T@n<o5(k-)VJ{hzP9w3cb(lukP0
zu+VKPJ49;zc54+i?>1Gm{=O^w20f#|Z;#b%pc>7?xpvg`5HphgKiWhXHczbaB$YHi
z;f&sCi#Bi{(osZtg%C@kxinf|6kxi~Mz$ZEjwGi+{H?jYJn0%4JfswsWEp<lKO+#A
zY(F-Q>>!Fwe6$N;<U2?AB#pVjT4K;{9*s~$dV&Wni*?EO@gFo_G*C(?AwxNmu;QkZ
z^U(Q=>I1j|1xkvz0sY7PiZDra^(l-xv>>j}A-Jpz8&aVbk&|ETA@I5S_fq<z%#fL)
zD7B^T^GEh5g&(i3=S4$Pk|#w*Z+|dxAPT@vMQLrOHFl`oP)e!JE8>c$Drc3AKApty
z>qQi10c*9TW@$$I1xo4>jNW|2**FM=wA1^x+%M}K+PfkCLosaR2yEG^*3Rf_nmj{e
zCqplD-CL5~@SAFf^Annhj}+5Ei-vo9jw%!y003Z4lWD(s^o;B@g6Mk3yFZXgzeOSP
zzJ<<1f=Nqj<Mv!O+x$94&rbdPk<nxY*=@zDNJw$N(`z8;(e}rt4HGR$%3s0WYTEsy
zkU#s499cF+f~H0@AdV*@nSL*BzE~l!x1=WJ_UN#JIy`)&VAr9gdFRTt*61WJO0I51
zL_z`<C<QOY4&37DuKSueP<CCISh$o_C2VKKYEm-=5gTOk$e3{?7UF+Op7%Q?X=wPS
zPEVHOawBvUZ$`W=J#)0$V>9c8Mb1ooQ2g^6kkFo;M~vsh7c<iYC}h_of8+Z1b>*Vv
zB=Rfc$X0>omXB?4bhYtUYwjJCu~jn&^ZGEe$`RC_6p7ay+cBeg{_JtjUtBIIo=kc1
zX-2VI;5-y5s^ywJaSCVDX{DZ-nL$>%)Y0PMpw1tyNzL`%TST+ON=@u*Y+gxWT;}Vl
zsq&lpUoF5cN85(rNe>K>lk%ffW1{+qoLZD1w1l|@-}iK`CjJ&ds;!mU_IUlQ*DNZL
zy{$$8p0l{t;L3!};|LB?MQgtAcx=q)gIYr(VPPpELz*(XDl`9522f4B?Idzy19P~V
zA^tVB5xMXDdmqB8r1=MYs;E$CQF66QVGG=7^CCvF=8*R^gJ)rx5fOvB!XC-ubr@M@
z0du;D2P`qt$fUbb>3QD@>hMxQqD-?FBocR@%r+|`?(h$UOHDg+h}GDWk%*YUAXveY
zOyB>WpSivZnXUPSX2{h3XXdY1_5D!JGzX%nVt|W?80-G|&!lqflCp52+@0IgOjQGu
z{H>l`3plO}7xyfA@A$8ETi;gO=hLBhJyL*j@FRk3l!R2c7=^Q&kocPxW2ixI?6q4d
zYhmM4%K*X%tPmlbq;pgp?-V~9a8X^sc@*M?+rO!#Ma&S1_6QVuIb{{>@`z5%H1uA-
z^SB=*AeB<!@uwRW!O=3bO8ErmNSnb*1W2Et{+qbHn@|u%^Qff4(aez8h3Dd<^AK1d
zp`i7ywtE~3k1DeIFMLD=c}yzH4s9?B-&M)D%1GK?2DYDBb6M5PQGDjC(rN~%(c*X1
z(W!|>oIz4Xan5aJsa7wVn9C$cct&8EkCwnE+925}E?yD3r_tURnO7WmAI?af#dcd?
zT?%%m#-m5Q-gOt&TRuMjPU<@`(jpux`cnnUqz*Ts%%Ewlc|U(<Tpl@j>E6S;e_(L?
z0lJGa9OeD}?CSk!!j(4OR%&c?8X5kV6Z3*Y7!Wr22thQLFtTOdnIMELP?)&;6*1#@
z?jO~sFxbKGA6*kB0|q=8Bj~WtiWtycLVxrU8ccp1Vp}2~U5zzYG1J|BFnIDWeb)CJ
z5W1B8)hXX#_5JPob{xfr%%chAM!HI_hntk@<t@wEWLBg@tpne#wS+@rQAx(dqvIrc
zEg>sV5%fC;BU8XI_5Gr1m7GL->E(2@khHdu<*7K#EhR>DelvQQjy$D<8ECS=IkeAi
z|I!0E_$J<c_7ML3{M<%^jboXUE2Wa5U|D_bCJ>9J`fk1#G?njmo`tiSN$~UV(-1$q
z_~g1zlW*#sDSgOf;QBHXw)Pf36>?FqwrqV=udv!xcc6+h<6`n>z_&EbbTy#Vq=f~(
zCO!bQK{rH`m1{!1LuhIw1EdI;Ar>ji*R=YD=S@<&H}MHcv%+FD96K(iSA`l2yqc{`
zco;ZFIVf%E;uG?=w8`uA$QWAr8<2DsERM+6CLe0v6Am<W$@AhU(kV9ecsy;D6zA7o
z^9lzOMOCi|d6)HVGosJ&e&HxIHVzDdX&UI*%)Op<mP{K9TJk)8kzR6QmlX#CROxt0
z3eQnUQB2iM9U1QI5*=^&qNGIQkL9w#pus}XMz2{n{vySsW?=ZG2&J1tBn!jTz{*}|
znZj_9prnc*YW?+bH8~ZCvd&BSdXFH7%yBVsXUXK7#nh#)T02+(Yi*Vj`&A0~kt{Sz
z3m=b=l~UNdF;A#1B<SoHn#~DRJ;CI4Y+mPd_ghw0(h;N4U~>#yUj(a*XYk1`4sM=!
zn-gJzHh~=B$fEv9+TuJ&tzl{MI3HvX`3Jd#Bzumyx#DBlFh<45qU&kXlf|U&#a6i5
zapH1CMcz!Ui@`cYe5Z%Y;Kg2Np?S2)RNSHO*x(XJZGJnL`(=RksRL_>Xzd^G#8xIj
zj}R4x!^_Kqj9^0sv?2*FbmcF#@~A1vc(Ig}56LtZ!@=sF0GKJE<0LDZ9bhqg19%0Z
z&Qc?hyYiTk3gy&5pku~F<lj>3(ughryF<zF6<tS;c*D@GmK<RXB;pjA49*JCMI?sN
z&4%m9xweWdFZXftDWABm?WcxKvyi?ct$MKy*Q+gqd?z@$CPLo&7;iPJ=FyV5Uuhrz
z>V|^#HaI4asGi3eL=@)1zw>{vzxxP%HM_!df$&726a-9DC#2>dIwpL1b@jq#Y4Fxc
z=%#Y-20d{BU_Pv$0+f&M_O=&Z-=qLc!~hVLOXEe8KCpz89#iG#F+|Q(2y=>QGND|C
z*<ABbWlZ16touQRKT+_LW3_Qi3W#WZ?_7x*tI%t*>2kH{W<swLVsa2bgW%RH7nF~9
zl7RR@5Bu508@#M@{gjRb%iUgU7K{4Ch|++KA_QNs9}I&Po|mMT(JmYb^w%M^z=(sf
z_Aok>c=>5P1cD+E3MNx<IBqQp-=$SMD>kb1y2fab^_Fibhoy`$e<c5Ay70K3&IpN+
zfb~GxdJ**hD&#6ZtSRd2U>YK#1R-NSJ#1eQl-BN$m%i3yWuxomh5ow}m|QW_)6-||
zx>eLqQj5NMd%tD_;an`}K8zLgYF%A~3|*D*XVll(tsg$y4HVbBb60-L@}Fpv;B`@I
zxNC{f7)PuxGMdU~`>J^rgpqgyY#8assW<9OW7evmU7w*)h|JO6Zr@8_bXXchIV8KQ
zq{-2`gcNp+l`(>93>k^NZ_vIW9miCjVUo`6-NdI5-D`QjJe-GzfdWp^pKxTgTJ@IS
ztv$pmjD*A`+v6QI+xv!NRGV?!#Bsp8vbqFTmXEy@1;P)k^zYbe(+p&5!FpVp4$koV
zgZ$HE|8cJG%u$Mj=!#Zqn#MfOqQ}Ry(hmz8`#-|M8(QAkswCv#1%q9Ft9x!x=*WZQ
zs@@t0I=Mxkt5^}+mBGv8!pFQ-n~hSb;6?N(t_VG@!WEZHa8r{wBst|5&xuYfs=Z6X
z|Hkt97ZW|nI&`b5SFF&%roF2>(#7Pt4f4IC%2zCmS5{lu+PB_0AyS$?mp+5&epro{
zf4KjYayuB4IK7~*&w$=({ADUl&^S@Xyllj(!we$vLPwn`GVUH54O}P;+)iR>Ah1TT
zPSbcblPfXVIzIQZ>YtbO@T?mH2~u9;Hp?5p1ZC%v31c4cB9Ly9?fyK&Dgo5n5+@}k
zB_A7S4w1n0KGXqH`9W_?5ylIuF`rj4w=Ujyb$4GDSu#*r=*Gszm}^U5YI^RtTsppp
zoc}Oo=Rl3Y`P69dg$vtD-q7<}($d{WJ9H0Cps-Y!B&M^N9UlrTYGeexp5+L#(9KY=
zY+z83!<EqGc0h~}Gz}aSI~J2paAJIp4-ZCAu{j=W^;)@`7w@NLdcpywxt_c?pXnai
z6#@75QoW_`$jzhM=f8JbhX@FeUW}|@6M<MzjGzQoB9g^W<>C1Bt{#ts=MJI;QqlCA
zu2xEn*1s-p?^OD4NRA7H`<HK>*fogk6{bdn2}K1FYAEdhf*mLbT{jc*1WP_a3R&%_
zeQ681@2uQ`C7DKdATF&58}J>B*b=VU5h;qD21TW7d03C@M~jR?sn$-+#lzBjSGmwF
z-rJqb`KXPsjEQzbG!kwOG_-MS?A?@*Z}XC;1R~ebg>@cHTz&qT=ACd=i?|luZVprV
z_$UL`5Qpk*%}}(@c_3UVR-PuFfFSS`FB{UAm?8tzJOC}*HV@_g)4pva&<B9kB0y57
zw+~fhP-buNk>97tGOiA#h1pTJ-#ZJgJ-u#f*11R*3k#=d&Lzo$sjB3jpnz7G#wzgF
z$CHV@7$~hXi}c{7v%l5`q#29vgQi3iyL(TY`x{)k(82=8x1gX<=uwMkMmO=xhgzb=
zj-QhK;-d4_l@W2{PSN9y=KJ(PC#=HG*&1^H6>@#Fx|POnZZPn}$CsIF*`nb$3!IdN
zx-<kJw$pMbT30Z~rBXxXNW`P1gIG=bYw|sx;KE?yUZzntT;|-({Lts>M7OKJ))XJj
z{?|i`<^r^uirbIjD#PJlrFZ>1o>7F^Fpo>jf&h-Yt8e&}Zg?Xb&WLf>{gH9+{~C$w
zK^?fDKTDiA@cR`c(UNk`tw^C#u^buASNT=><1{X&!#)xxBNIl7UvXJqTl4vGr=U1r
zCCSku_v8`ss?K4aT#9i)$7!eBiOtW?^wGjUj9_MPYij^AlgP*mLrF0w_1X6#&Rw_b
zHO>o-n5wj7;2Gbm(Snt{*}3t15Z`o%T<#$)6@%aO^FAkpO*oPg!wUr-1!EQh4DaQS
zS*rcgW{ISQHU1L#f-1(nbbVA}|As<Ll_$$2n~Z_u_Og`Q;D*RlZgjst{RTDjcy~6O
z8UNhqFiEO4Dli+!7?QUq<12q{UAmZ1xVa5z59*fYLDt=~>PFa+c$W2ex$9JsYW~~f
zRgZ0_^xtrWvS|XvunCe=@Y3GKcyfq=Fr<VNZuy|XZza}2M0n&ZH_3Yh#hTl?3eOJt
zuj<N3xL>_|u5$H0_7l|-;;&`E-g9`$1Xj*gouQtjB})~+ZRwqHCx%Rgp+vj%&de#(
zBo18`U}qC=3T55HH=Vo6uV}&9@D8$R;s*&)sieq%$*#B{Vu^zP*T$C~FE27E*V(4B
z0Cay<vrQ2pvb7xmt+n2^zj6v3R9O{Viy9gAZ3)lE0w6`Jsjwa3>5!UY5axYsD571o
zqmBlNUbMmM|8<J_z@77=)O-5+OuJMhZ%B&5l3J?>d3qkHzRwx4X5nLF7nf@`NP)3G
zU04Sgh9-XLz6|1vYrU3>I=-P>J{=I4R_)IiAX+=>2Gl)9``<oG<PN|~){PD`hrHU!
z>Hi7(Cygt_9T_nNUp&u{q;VH9O!GZf3Fo(24$3=tn4}6CI0VpH<q^$h8q$gI_g?ne
zUbDRsIi*UENnL(f!x44cItSGm7$)u(xGZZ)aTcApSnT@Rs0nU2jk6a4#$^X%GTkJi
zSZA6tdS3ME-qr*qI;Z7CMnZZ~9yNptI0(sD-HBUg=Zv@B3AqFp6#m%^M%qEzV<G13
z?sVBkZ=$Rnc+!)@*_zswpQ!lp>m<vnZ!f`W<!qo}mSNl^UDv%sm4Y1nM;Eg+eDWf%
zF!RmlOTPR=N-=Um4kU;54mG?N`-$fa4nr>zHM7Matj6BJSr3t@wvix0?7B~&yOd(H
z)gbkY(+dJy@R;aF#0eF76GTO=lejD_$*9&VI10$TC1rT@FiN$EqmMuE%4vH(=Nc5Z
zp<Ek!;V2mGzA<6-$+_xtsx0Vzw7R>!KJr)oI3%Z)`RfP*q<I$<y>?)=?Zj})AU613
zo1bn0gq?yp*Rf^()xOBcHK<A1+8dp@7fCm-Bu&kaL1nIYE565qq=9B{pkBG){(L3j
zaPve=(cW(-N(f8h*}VAhxw`a%JG=ay?K-m<*QHI7{&7!P$bkWk-=^s;Wv@opkT|H;
z{ya_JYVGK6i40YL*)}p1U%R%W50n6}I!|#-0L9s;8NMMUt8x@XXoDrirPO35MqaH&
zONO`!!FIqsX?b#PirrP`$N6Sm%KJOS?d>fM1L9(cvPzO10WVljYkXm0VPbO9jzimv
z<Cv<?hc~pw3XcR_m5)~USBF-pXJO9I@F3VcyIY$eHN6SrLveQr|9q#d&JQmkSMZD`
z8^f{23#;{eDtyilNmUml&Wxx(MR6^T*lB#O_QYH%{=EQ~tCg`}S>IaEDP6!h7Hge1
z|GNiLRae~9%0LJw`v%ZIwkN}1%qUfdqBpNrzc_`DF<cH;fH*N!0W(JEIr%@RU#Jct
z%}w%qYhsAU>8#Y~m$Jz!d#QXjmCa9ML<J?RAPvP-=QUb;lLyj0zz!#QvDAp<!lWd-
zRw|phANSRZ1x4U#MBo4x`jUWs_PW~K4JgOb69{68Q#T?R!H+bb^dViXI<6Q4+rDo~
z>Cg(=yQf9|Mq|H1!yXF1eIAYxe5HniHQ73P*LGGYN#f19)yZDfL#|7}LEnCq4tGBo
z``t2K@C?SZnGn%e#AYCj`=qP*r!s=<GIL3E%7|&~vkpJwAeK`qbSCu~zz+m8|1!N@
zSP+bFSk(#1t;lZ3ZX!)hEr4n=L_<T9d-CX58*6IHup}Uc)}TgBw+dQn+>`0Vf?cN;
z`f7jdXiyzyU|YP<`PekKbq!5;Vt>>W$ERe~2b57b?RVjIaF$)V?v1TM06!n+<b@XM
ziqz-+BAnd6QlK-1nH(OyoKHO5&52>BAR|(-Di4^Tw&&>fKlki3^YcPa3TN~q(NuZ7
zv(EY3ouX8Ng#ixqQG*Wyn;u(21y%<6u6)O%mUIV{*m9_ty1-~C%F6Po;pK`GRQ@fo
zdX2m78W>)`e5hAoopa(#N?xRW_OlcET2Nk8Kxk05C{}pT5$3Dol2&;6bVrT-y?f25
zYd#n}qh_SpiKyz#D<4;>SxmNGdp>A7_};dSCU>z0BdTcEqAfua79LrkHbwYeqfgfL
z^~!dXS?*Yf)=QIbp{z}V7AA2V8(%8||8WFTBL0O5taY=-HBmvaP{R@8&Rptb$W^1l
z-}*S*kB-PUBeKk&S8C|f`*$bN+tZIPse|o5M|y{BEBjv!m5gT?x4&RFNn_G8`*s+P
z7m-24;ZM5FI8^n$N>aq<CP<;UC<2CpUH~4IM^A;Pqj53jKn9;y+B7mT=36dI{1lHL
zKWA82Sh4bX?_ou`2lGAbLPDZuCt;Fqm%bAVvQCbCjGSQ67Ai*EK?!__JN?-&V^lah
zRVFUiuZqym4S{wJij2g@1s6igI<1PmPGeQ*MybJX>;An_m|Y;Df2W^tu7)QWIOz{b
z9^kxUN6+>!tTsTca`o=zT~&*(mND6ODf{NR#5?Aeo6Ix%xGMD`YO`%2r3-rF%Hqb@
z7c9R|`CK&!#EP<7{zUcrQ{cF^{UAs-7K0LtQh56?B{zAB%X(QUoqxreCX!0jvZZqI
zSOQUoWRFT~JLQqXu+}(NsiOs|w+2Qw+Tzwr$u{Ip0ajMut^ZQM*~eurUSY@y6msdb
zIZ<*_2G7m{u?iUa&u9uA^P!8IqDf&PcFAO$Z@KQ)<nPp5?Ksoc7UMfy+0xd&i1{qJ
z;!sgDXEwt7c7L7XVR8JKuo$$c*>-4A?{vTnf`(S7&g(8k1^#l<Vc`XuoxM3X<hV~g
zr-pnEUsqEtO<nsS2}&^K>y*QZb`O{a-JD~)XU=Z|Vvtlpzk57gHW05_7z>rk0_>D#
zz9}#vS1MveF+}3CO$3=1T39xFXtOL#MQN@d^C1-V4%SUH#VLaY@M*fr?!HWAKI%1N
zZ45P5urGNkQ5Kgg*;5~j8tH!OrJ?4ah?$K~@-&V^!<52|HT)6r*n~qnB(7Ppc!>!E
z&!tRo%<Uf=S<>T92NkFm_3wwT91W2`y698%kbKuVb89LW<<WGZH{y<^{$u)p(KE;Z
zBnz7{y$_gb$_XR$9k+R-JCGA$Nw)1f;<Kb}_Us7cj3fh?xH*$@mKI9s;cGeSZCApm
zS0Js+!Lk8wrZ4~EjZp<S@wERQ8Mv~JZN~awWRc{mNJeOc0#Dag-Y5py8y~D${8AU)
ze&Qsj4G!u^`#~XNjNS)w{6<j7R{9OEGr3*giwOX}Yp7w*7sL7m$9gM8{myi6>w`xc
zo9Xv#Aq({Y-HZL@8HVjg>ia~__QoBe$D*s-qnCJ0Y6obEBHON9>$%?Zz=mvH3qxD0
z_FUS#wj~E_$@rV&BRyTw>)Z=~)Fo6rtSRFUj?`;)U<_1<AHsO7T!<7*QOr+Mnd>v7
zAUqZJ-%klFsST~n1^0`Xvkjubz|wU##`Jp?WBRS#!+RT-U9x}-UV_~Tk)YDZbE5FI
zVlt@vQ=<yB)1|T|-*^3WeV17T?u+UG2vK;!^j6{T*@8fuaxb$%yB3F2Z5HM97bJIe
z&jB2`92wniAwmi4DbfIEAxm;~P78!<7Vyn%7LX=NPJb4=*CzTmE4&hqgZHoiY-JXY
z_NQH-eM7#)qr^`d6tyj$bUL6ErA(^AO<P%67M%~xCeXS10~YRD-({LCAdD^@$ppP$
z^v}a`qY(dLs<Ih2x=j7eYd}6LE&H<)_QF<-)KyA8^$1v!hb@;@Rb!$-14^wah#!F%
z!3uBZ-`l%N;b<U{otVmKz`cn(o#2$Nko_5rAHL$EeJbuzyF^k*KRPz4>&(As$zFhF
zQb%1<_Ittw87Tt*;-%lwqJ|ZM8d*%jrJOH2xzk}o049ww<?nyK#xIjawVpiqe%M)(
zhm{qY4N3CR9kyN*-;Dj9xn_o<6c*hUL!*R7P))8iE{?H*!MNpeLlK7ZXC=q$=^sXc
zLKnSQ+C{*`PxJjoy|({Xdlp`pOr2E%l2<w@rE_#>2m3oUTVK)dLWwXDs?;8Q;$s?q
zd3A}B$EQRR#b1Fhk8}u+hgd<!cWTKFAJ-B#KQKqyx1{a00_aRJp(G+XXTHe!a5V(S
zDStkZ)0rdy2Qp^Mg7)uBrJnjC1_nOCx)ha~5F`sQ#<N-#G6lR-%V%5kU(H%Rdx!>3
z(leg?5G+zZ>2+>>AzmS@{jzz<sdG=`NiiBr;k$&qJI6~VI+X)~$&&|%u-OICa$HI0
zA0qa4OLGCnJE%j8p(sjNcG&)D^P3!{qDmV|MbuK)Ujo|@PkvG`A__Q!VX)%Y;wj>c
z=<}J?8o9r`rfn2m!Pnu-L2)~qmE>!w=WVG4J40_s5j5@l&E6U`mHx5B$;6Ocrth|!
zlSQ0GJl8R}XOERWU0Zwv24{?Gi_4Mimm1=0d9K+r%V)=%XrLAh%xAS)|5wC!MAZB)
z9)6Op99kowWSW>;Xz!`DO)tSu)CRgCarMwBCSUp;y_CUi1PQ58C!ekJ!_F}SQc%2;
z>SpK&KSuWn#9stT@hLtrIC85lGCr}Xs!J%IGF1Rj?lBI@bn7C*ok9Afsv&P0X$vU!
zdhm^NI8>yhV!;te@U_QR90BIUdX>WR4OEQY8dO@HH4V3qHV7L+OlV;_{cg0hR0kTM
zmQG>^-Cq?1$v;$+32&G3=-AZN7@?*55-_E8W~TsqA7_@A`4v_*;z1d?J>G5e(F9If
z67m6g7(<LB6(LpprVg5MMS>O?Y<}Y0CB@8IqGroMiI<6)O2sUiJ5eqSU2!|@Y?}=K
zfR6?`G;1&n(J+|qELlvdbd3+og>PE%q2=RxHEX~MlE^oO+Afq{LB~UxA<%9y4Tng2
zOLhow>$nl4S!Uk#ABZo&aPyH3=`D3uYsrr@<pzMynn^k?5o+^@Ns$i$LNs`s_h4pe
zdU@36^wo$)%siw7%^bdG#UL>a%}92l)#fx61^EDWtdYr^Ig#n00a||`3=<?z5S|8f
zVI)3Aa9Q`O4D!*veJ=dZ3^Vkxre;XXMo+?vsF%{$<=;P<g?W5*BFIEP$k3l8qgZ1;
z&xazyr`nxm=BIshtc;FeLhY!_*H2CQ&GWn(>r*#JI2$h>?17#L1ID~zER^fI1SI>_
zam75r7lPY!WT5S7n-+%Tba&&hEkEa4{-kS>Ksd$m3d!jZBdsANNbnd2dg+B!LRrc#
z@_~K$LFbPpvfrp1QLz?m@BKxI1K&0V)$)<359o19*x{ix2>nhNi$BjW<2dZ61F*aT
zZ&eZ8IkHfQKDvS>D^7LFU$es!6BA2rk+>#WnBAErMKY^7lw;yq1Kl*r{5JW7UM@L)
zi4aa^$<uE+=EiEsng=Wpu1w<%ii^5`NXX7<8s*D^nAmO;)#RKtuD~^>NYXCyXM4hB
zGN=&Qyb@(Y-%g@BNfd1T-MxW<d!do&xe9a(wdJ&nqk#H+jb2L+>y&xT4_Ez^c?FcN
z%9X>`x4A+J%_%w6MPoOKP}aD#3)3fXp}%s-#uF}FT2HUcrT!pCR+sTV&|25#;*YN}
z4brcp<-_Q=t=J^fn5_G|A5L>(z|DJyGC%)V9f0x9PobmDW!Rjl^RdlFW`ui98k%yM
zGld)8i?eOj5$zI`a1_o-egA963jA-ucH-Z4BdFAWeTk-G!-(R99^9qUK@$fDE>OtC
zp;c2qD!@rZrZcvgDm;Wa0{8Vx7SM*@RmM(&f`U{=7R`g(RzpzQWA^8ZIRCzotmpcc
z9m{OpoIGh|tfn7Mjxvn0-6(n_xKm&zJUo-H8dZTWLt6bG8cwxR`ve=HEnw+`{nn>8
zq?3tZY6430Gdr}cg!c-SW7o_!cHDQ#DQK8wyVW$~I=WIrZfv#_bmKQ|5T!3KxD<-k
z@m%}T9kw!%{5}&|-*KlusJr;;Sdm01irN!#tkdhAb7KBb@_Z@7O!=lw1*i=G599W+
zV8{q$Xq(%4J&~UO>occ(scAG*cJNhWI}D%T$jC`jMI80gxwSCIf5rp*Vqs}sOdQhP
zRC*y{1`Ih0yi72<Ai5xAB(eEG5?E$papAP@@B_@b70l?-{Nz7kN{FF>5?CkkL#EM*
zLW@VXGCH(?F|T)>Zg1GtMbqfTz{FYlK-fi*av8FhIZh+>mCmX0?M)j4vKVW)&Un@!
zqNSCvh<O}42Zjbxe82+$YGIQtVZ<o-_rHG7Hfob{Qj+Z_Uo+M;pr061#w@@NG6$A}
znQgDnT`QV<73_~zan*XqYlqU-1q#(A@1|u;*V|cY$($I`BLAK`I`tTZLM^h+lCOGg
z!GT9nVWNkK*U{Tim7Om8<G9^H!wU5Gv!m{l#Sxl$<_xHo0dSMA)BVznJsLX;`nSK|
z^Fjq+ZW|)t9bHqJBtP{tzTZco4%ueBx&f)3Z>PVUOiT&Es7EI=-@66~CiGn%E?dT=
z2bS6TKyCbPK)nIr7sliMtG&CkOxbo{(15m+7|z=rB)zWEqh?QM<i>RP*w_4E9ofDK
zM>j5Tas5LK%C`6mrTX*JiAC;5DU*r<^&of$`FE&qU;=+Y0_$5?H*zni@2P?Sq^g)u
zFK$_KSy^=Xl$T|SZFi9T>!-i}5x0QO=UWHP=ZDRAY*`!e({db_BX<5iz9xx-uNA(l
zXX>&JWFgbWVlIh1ioi|L9te15AG3=8)dIMn4(Kv2qYP|TlIkjwJb7F-bUx~$h|(`J
z5@7rc|Dzm7NDNo-)M;%p`8HHw!$~d-A>uBWBkXcGKEASXe*W}+I;I!scw$K4e%MD%
z==+94vG(?Uk|9&{3qy#nhWq&;A3*c6?KZUsmHT~!FPsska6FxBry!Ky%V&l5>+E(K
zp04Sa;n})lDsx&!*{=P(vIbckkkR_WvVoJL<t^TK6ZoQIUO4zb26(n`r&$T2J$cb3
z^_TUk&P}}}sDRyWxI!3Vm{~N&E?dGpIIk)lpr7_#rO*<Sf1{&tBCp?%a!pZ=t8MI0
zNCYa$oFWv;eQdN?j*cZ;mu`;>f?I7K?KbJiL0G}JA-SnHRK)9MmG_ILzid?%7S2|5
zsdz_sC5qOje?`W-OO6c$B@uqY@ki(6dV#|MZa4O%)!zb1v)A<c+pDS)7|ac3{Tk3Z
z=l=&nLA|~`eDTs{(q9quLvsCWVd@&nMG__gmjUJzHgN6uVYG8G`P`57)-kA68aY0A
zRKc88K`dAk$VEddxn)!--8&v$GGuzLnVCyMn6oknx3~74b-D$TRxqj>zUtc3`n+Zy
zkLSpHJst-vn*M@nR&EC(j`v)SJt9{iKkz8}yXi9*q#y6uv}GRozbzS?J~`#1mbOYZ
z)o#=h6nRYC+HNC0QUjJ9i|e&Et}^Yx4;B=kcHb2d@wZ<R#!Isc5CVqwbl^?*Q0~-u
zW=5_wn}oP=WXS31dEQ^cV!Ga4oW1`804wfgJEh5~f#XQB^>U?7E*hm$@-;4cOB1~@
z6CxC&^`Vqhprc}BX!jQu_DYpfBnn3TKZg+L5}Zgfq(*vFD3$pIoKj^a9Vjex7Dtv#
zxJ`PIQp~S@92S#!L6_)&xx@Pf*ni$!!OW#4T9;^xDI`F_Y8G^EZ(+ud#f%sbz=#2X
z%vn>WuHWOnvp6rcCL8lFDI@`=j+83!<P$%M1ybW9M1!`7)tn)9euxj{;N-<(7AAfO
zFtefR_+}$P3P_RgD}tq7oyD(QERU!M_M|1Ep9#h#gmRPgSJvnxbtwbdVF{_fRcjsN
zT11FOW`6C*O4S4l)&#O}Z6K%ibaC0R3Wg7G5+3Y-O;YYuShLN+{9glE{f7hH&osk>
z_15vst9t#@T0-kRPD@Mk2{b*iN&b34Ue~dgO_T8Y@t(^uLITMM`IZi!rF}l{2+M?>
z#Fq`bOfR$#)o(vmbtQ!m;<opTN3{Bdq>d|RD^kk6OP4N9YxV0L1M>1xvx3?uocVv>
z<p2BZegT(71e@B21}dD<+d)C8#+LoAzbisqdeb?<7mOGX2%x0p<W2^|gOQz`T@fCa
zuw<`PujnZbUn0G2s^FZ{16jGt%9vpnFm=k~{IYU2PNxfp!%j$WFoA&qXqraafpngI
z?j`<q{q>AGXE=&Up}4r1(z0@p5}VCNubw^V)2jy;j=q{*mE~M}hMVu^l<=_DL|tG%
zc#!Or-lCFrfwgt0h5pFc^CIlsJ6P!uW#*M{3$qU7dOriseCVo(kxq{!qrgj1nO^ka
z!lE^6cUAnSI4iAQzu~{EkgVpp3*@S#<zirko5vM}NmAnX-4W0U!DMtMS9&OUVu@HU
zVVe-l(Is&Lf(4i*s7NYNC=`-xBnNbbJ!XY%htq||mnWBsKAOrk2(DA0{|OzSb^@X6
zdV$Gg>WbaR9qer#A(;^CW!wIS`BzfHhFx|x?6PB0AR*36Vh0bAAsRwJQJIO9EDL+n
zEqI&gM3~^)60xQ5`iNV5R9y?D{{8pgw>e>wl2!3ed~D#gz&XczF0F+GRUlVOcthAK
z=ci0xcFf(de|C&m$x}b>+)f^oSjjQ>e-W!#?9C|+cT{+b)6gmlky5tlZZ^J@=cZ%>
z6NC_=N3U)rPyQ_;xGF|c8`{J1)R}!95y|O>hp5u6Q|{91c?5+nzuti7ik_y!mq@V@
zW^5M0dvCtVTkn2Ar;Z)C>2KFJI)34jpO`q|R=Os{H##OGGm8QJ`_ivZZ%TH)N2o<H
zY{-YnAG#Me{2=Sp3KfulFwaZ(4p#C@G|uef=r}0R{*U+P7nRtp3Vpj+$t_a(VO8n<
zh1sdI{GInNFC+n`j+8FopN}jSA6SgqjuYUJS$d%oE&wjjBwb&;NVa%Owg(pqn?o=h
z6Qh93k@RS-1d%^8QdU`g9qmD})+UgsTB{TpAJpB=+|`F(DQY_GO*cg2kzS)!@c7)B
z0y@^bI8w@jdGqFZ0|EjF4i1)m`}S=UYZ<w0nc8*S=v$-|>-qC_aw-*q**d(FHfzDA
z6R|k_pNS2-#0%}i<oIKe!5x@5bc9gMtzHM!mHj(aVdQ@bAsG?Z^=yZYDVm1M>(!=6
zDYr+&CFFZG8EiHSqQlLmtA+<he6g&w{Ez*vpZ8^Y&M7F<(u*s#d+MIYtXIz2B9zkW
zkRjH?_~O+8hgFbWs1hCB4v(kla3;6gjmdQAx7zFVVm6xz4Gp142P}dzmyQj=;mmKH
z^dhC4S$CUa@=lrcb7|LA+bRcU<a<{I*p$#N(U!CByfi$tuuP+S2RN9gqDv(&JT@U-
zQFz{NQTkQ7^u@(Fk&~s|;5{iM>rmxte(=eK;$NDHi5P;i>5elkK+<wyuJkfUN=6De
zH^9s}QWA9HHlk}AXquL0Hk(11AgBwJCL22>;8G!=W!oJNALv9;@avLpDMk#cY@HLk
zd@P+Ii8U5UUDwkTMUgg}t>!OM2ytSU0IFjD=kwFz$6a=VFiL+lTCS@Nk&NIYsi>c(
ze749~D1CmMXY?0f;&U4gZ2b7tPr5|Qdt2gsWo2cJDgcW;=oqy{+zg;=nzsB{T(32W
zjO-rLJ;5^f;cFs8JzmLkU*-)AkBk4i&_!}ik(v-6ZRMKbA-wi|;r{GG_57l&l+?z$
zbLHmdmWRj1f4xQb-qD;Q<MVVEJ3}=>gB6DM3m`eegV}7xY%#uHG|lKyttdv)r`zqu
zW;1R@UauEbJ#?g;*=)vDb&Qm7)`&oUOmmZ*;$}iw08!0ICM}XWJEeDjNlx1Gy4wrV
z)6;>og%FCAQWrvqnH63BdFSYer}}g;=9pdRl_**$D0A)~6VbnKyp6Z!6!qCkK`QY2
zNg?Tn|BRA(z`v7{#D6Q!V3-hGjbs#po-K0)Ato&$UeP%Ph0`RQE?@|TdmJ!ap(i;y
z%x1H6yWPpvA6UYA*9u7hrw8n2$SH%uN^q-&g)-0zk%15%aM&3wQK1^c`a4<qd+UT!
z_2}fGSC^ycU{>Oi<d;LaTcGM-HAA=qVnU$8TR+(*w0F(#nx-ZDo~kW~$ZAqQpSLW2
z?C5a<F_WhI*`J8G1^lE--m$dPe@cFRL7Pf^`m?Zlk9a;hoZIbI*{Xj-La|)Xn&$_Q
znj3fCPNbCUmn~b?ShJ3{gVocid&iJan?+!=2)aaD<If))^xe`;<yvvMj);NeOb=NF
z>XXIUDX9Q;MWnB&l*o5iNcSB>X<A6GlO7(rJd(&zGj@wGgrvztfW!D6c>Dbi`QN;S
zm=pz@&Bo4Md)Tpg4Hk=qtn3^%ZP`j<*IJv^^iMx$)8^k%HH~*(e;J3vj;5+uEfzE>
z>D1nW5JoTTf-;?`W)$FAB|T{HzLd48<U$BDpi3$BI@^`voV28+n{BtB(c2LR;Py&!
z)v(a77yec*x^=K(HVOK6v9Wc(vm2arZ6A`UQ>6~9tonIm@{b}8jWZAo5`tb563hx=
z4=xn8$lvf9<Aq3Cj>%;vSaTAELX4u)359rFr@Ju3XdeYatKa*R9_!XYRaH?|R<=DP
zq$<%Sau6J-Q)|B13E8DkS;rEmRc$YJ8iKNGB-nkdmR=1iJcinLSQ_lVWmGAvwmKSJ
z4A_8kB<vATH@f{4jADO>stA1RrHynbbgRbm5Eygv?VIh<Af*6GUG@LmTwciXLO~^H
zKFKzbgK7`=c(wyncea7HW09;Tb=LfE<HwFhNbxD+__e(|OfrznRHXcP-<R`G77#zq
zJowy(0~@Cgp5CQB_cT5F$Qb?DcU@xpk7;#Jn5SSbx~J?<QT7L*F#EvnwL20wca624
z;_^tgrnq(sL6=9bi>eX*nFT6RS2u>mcAey~iNV=Hj$7rnZJzMBt~ZCq#7_ZwRgW#s
zOxqY9({*)<q7F-FBI#zV;$Ys)67r=+!XP6KUQrZmHbaC41_tuOS2Gzmej?XheFb*A
z4U5G>MrIZnnOV$Tu!!rfz5;+1tA1t6)@@7}e>2mbeI6+#4!Z*om`o;g9p3n+fYXXC
zTp}zaHX|X(gJhNNuFw_FM|HP{#U%U#sb3ru6LWEPc6LR0Y{G3F61qN_zJFhjjtPm|
zsVXBl>G~+v?Q#lDCvC=(;x1?O31I2wa+fZ}=WtR;9v07EC^LcYs(t{NyhQAGn24`D
z6ZhmLLPMaSW5ZJweWw}e+Q^3Wo=?*HEL^y7`-BM-yg~>IQK#46S=fY&xWWTFbD&3i
zLsY6d94Laqih6Ck3A93&2;+ZXa1R&fpH|6F>l<p7izGZEp*=<l|5yb)KsVqm9jw5w
z0-hI;;dAW3Zf-6b;!`8vr9xUEl=|dXggPKT+^{&hRoImao<q1#jD59#p#gCa*s11W
zadGixT{nbgSy`FbzJ0r2m(Yn$>a6+SCXBr_ACy^~xV2V25>_b44=T;-{OnH``)Q1D
zlH<k3-O6*(VfuKxwRJ4pp&bUDcPzdvs@o#+Cy9@gQmPRViRZrgL%}#HaTPnnEcece
zu9*eu!pS#8bq26o1d7Rgw!<#e2^WXm_}2dl(;{QLd^Wgy!28i*W<R+O^2H^(d)Qq~
zB;E8D7J3T{I`D3ulMXQ!gb;YET-LYWG7bQz%Y_hvYp*gy<Dwsz(zj19R<2ruuIng@
zLRNMTJ9q6SF(H8;zBa-c%@#8jiy5H^bX^kQ7M!m%KT2nf?x7rgdTB;Ry`-YJD5dzy
zGsA{`vZUn2@b(Gs59#B0KQ7Xg8X4QAPk5*$zIxoUEfsRXZ)FopauPn-q^4ZIagVFL
zW>yxIW~c0dlR~nk5LCO%TBX3>Nt%O{lBA?0z_6UDsw%F!>Z*${w3>zJvp_L}ei)UT
zR2`D@pj$h`qURe&2!`O?m1B$vHU11Hk1OQ&WE*+KMi8~Y5Rl{QNUsE7yo65#Osmst
zbC`sg5*|^Y`c@z4(3yIB0ZLsDMx{rxJJ)bg`%>bB9UkZw1t|p*PZO!&zQok8Jkq{>
z`<uV|>Z`ps-+c1{pOIC|q_#uKtOdU$UO(nc-J*Sf6c-)u`P3m(B0fi#)6>41zw5As
zdJ@XU(3Q~=(NjT*M`-ON1c9g&$U>>hG)$&lWkuf1kyA&u+3qDL0u!IxaA4yngForg
zj=Njq{ANecF=~sr1u12j$z=NNSX}Q|P*}LH1bAC??`{jL^!6!?4BjsD%f`)LSGMAs
zvqOZUfWs<Y>fFx!(kb2Sbnj?=M3?fSyGDoG7yMRUQzNcAUa&YPb?zTA@mtfSHlSlO
z>NX0DE-$94yRGW2|50<h>PMvP*twfCh7MuIo-f$2X$wP78_ZQ>Ml)yLLWZ9|ikF}J
z7Z;p25~)ipRq>WOU^|Q?Q7@7orQ0JE(~AJ2Uq5Sg*0#$fBTnt%pu#05e)?nah^vPO
zTXv?o1EesG3$!WSRZWspq_TKp*@q=LX&;BRSNbaooAWbMcGkVdU#_#wN&d8mq@*OV
zefxG%R#qmaOqt?HO--E{85#Mh5W?sYZ4HB{)BGPDVuw>YLcdsu@UbL%HQ1HyUqn`;
zOW2=ZlldGGqVdXuxda6mBJXw|5n0Wh66WYeE2MhU0bbJ~9<Z7v&)t(xmzcUPcpdiU
z<5xu1o%hZhaQTSHP&@RAfx#UivJnq&hx+=;YPH@tY}l~%n>TO1)a7!Cyu3V-l$0dE
zh%IV-5+^46zMgX+GAQRVkaq()$8joNAy|qOH#*JU9;vhDKA!sJyq(Pvk+ZL#ZQAh4
z%Ap%yTlLb$uvJ?TV!t9-hR~;jctIeZ5%3&{_l2S?7E<mF4m0oC{QBy*H@?35+>P&V
zY}rh+zYyzp%jb`_+l|ZRYVf%cB1uuqwpOnnse7Gx3U&jmTfF$sXYmfL@bpV)Z{2iR
ztjw~R&6oYQ%el|vm1GsDj5#yJ92so744`*soAr{@f=@A<1i=oaLqu%+sDz%~ZtdNt
z+eR?jE=q-|bM@xG)rz3Tn51d-D`7|^PN$QqbMw>B%UQO31wny<%>M2N0IXIkGpE1D
zV-MfY%@gin#j0O1nG`H$v*Fo<`UZt2SgCs`lJbMXjC}_H!U`&$7&)xh{E!Y^dq#$u
zZw46NFTj-8!P=*Nn3=F(Q(Ro6Y1kEK1+#Xilizl`&^1vpa#+tj7Yzy6qU++}2G6<q
z(>cjWPE4wW%=YcuMP6Q>xbemttvBC%^Oq)*>1qI^1_%BJYaawxBc(he1Hr5q@r8*|
z5a-jP+4pPRVX3h`)}ZQ;UK_LCJKn>)kL7SyxW>drWPt#p8hKtpz%+->D-Y#zRzGL`
zomr(4w~s&!wn0!6?v8drwXS-T0#S{WDoXhpe!U0we+ingiWWkIgoTBDv1ZMh(Rq1!
z%J%KsMN(3dXoDVZ05-h0!J9H`{(DxGejuI$3XkFCvjx1Yqnwg5YyO2PU(Efgxi0K2
zuPzVU@Y?DL-G;@jQ53O8z@q~Cil&o@ECLdxl=leaQsH!N+4TCViECb2b4-eC|CzY^
zznfF@ikaTh=e_3Xb9$RB0UcWUyuP^ewj(t!z>?O!*0Dy2bIm1L2ev2*dvCiqEaItg
zQIpMP(QWa@a{6?!(LUT15*2Db3!tb>qkn>ptU{IEoo(2yrXL=-JmQnyU2J`XIL!i3
zoSFK+?b5quPcwRuNI(}=ZW<}keOe?cih{eUB+%~Uy<B?ng>>%RiN_wfpJl%oosrl6
zwt>{N1Kd3J2JRStGppCEtA1igB#I*NXwAFlRw7xfyUQ$^e71U<-9qd=V<IjcnrPe6
zy@NF|uUN&aNxH{bDXGv%h_Q<HVdfG2;%#h7c9NXoX5#4Zhab2yGBG09gcN)sgb*<?
zF@ZJDt68%-$w^LNs$AM4FE3Bz<>e_4KKS4>CX?x60Ll-*+%w>(D`CeQ^|Q9?pQQfd
z4xtbg=znam9s0%?f~<3R{e6X%^$L6Uj`#5Tv`oyyDh}_{CV)Xs`X^;^Mqi)Ru#}QK
zpC&W0A<!!tdPg_3s8$ndH)1(`PUd%P`a;<CE-e2W{4fxTcU14O1_uXEpEGAp{Qmv>
zYs6$*v`SX9d)A!%lv#8CVOI5C5*`Eopk>?JR^yWJlMwP2v!eIhKWpA&X|v{RZ>|e_
z-ODTcY<zvy%epY{R*)Y`<PeVZ9xf3;A0+Qt6>a^-S5{m^8-5Ru2kZ987u;T)O^<9o
z_NZ<ng=pd8wtjGB8U3|YRsB=6MXV;gL;NWbv8g+PJ9JG<&hS2+lI0;b!px)BM)K39
z3VL_8(W{dUt*S+?UX{DBFVjO-q003mLabe)Em%xg;^ML_U}Sa+T|WA)=4^K7Y$2+u
zTGaVumz9^3pI<;$b`DZXW_<Pqy?gZ_EHsocqeqdFdVqNg7W2op9o+NKV=P?!6I*`&
zgL6lmjizbnx{j0*i>YDnh{jl;yQmP_^SMTWb^xqalS2_O{^BsDuuLO3z{JKqPP%ol
zvM<X^QJKcTu6E2Ofs}A&?*LBkZLBrd?{O{;P&~gy#wL8KTLNCIKIcgxd6G7b5Tc$-
zd%=PQ-K|#ZBLJx7u;dcR-_Z1lO8mGkx<nY3M_=MXWKA%9ba4GKQW_c1b>|qY!=3r?
zY$jY(3hfWEvJKbZ+;Vs(3m(peC_iJ1ow8m#LyM`t1=ZJ@3Sc!s*Qkc^yT#3L*_7;t
zMHhmz#seJMrAwFRs$JaBwn_mtM)KDS@=|BbdpR|$VzAJ~U|=e+0w_7cQ|cfN0<)!L
zf~tycsk7!?xc~F{pC!*+&`A1yeU>LJw{LlU<rSM=U%l93RyGUC-!XEvS_}bwg`&*g
z^xDc-PFg3M&)xssoSI$CjFvp7*&1++>PK!tN|~FNm$&s;T<=(+>x$*?*G6_9*2DI@
z!=_C9eZO&kt90qaM_Y&pHX(!&8I+viA=sgCFi*whku3kcf+0Qa{IsbYi&^0D1lI&b
z7UZO^Nk^NpLO*oXErF4E;#H+_PposlEhs4D+N;J8-=#ApC8Y!f1=6=qFG|bGP!xr!
zPd-k1MkZH|8O>`i{+EobES{Y<g)@ephR5qg)3lmpzupXsWQNpOp?mfg*~H6rx7*CZ
zoRP1x<A9sQI4fKBxk%3R5Ep4CyHI7`+A{ja8=kGWNDE;>3MCa9xy33O`D%|Y9fDFP
z-4J=x(4Mspkdw(aM~T##A{2s;_dHq*g*Md&uR6gdFE3BHTrLqC8+#96D7&pMK=IB)
zpP-teN+Urfw9faaB8>7sJ96q9?gFg!<ygBQBb&RLoKk2X)Uaz#oEOFx89!wv$TdTS
zVys#NbZDnSSTmNIS*7(CfRO?9ujQ~9kM%lpj*m24_jRfC=;W3+Z!uY!2^*h;;YRfc
zlgTt@&YU@2uD|~J)GDha)22<MEoO52B9|>wf#0is{+TfTsxaN-O;jZPq$JU3mntJc
z1VeT2W$6M+B_bO^DpKw?>HMKPDh?b;%fiiXu8xzMep^VG`F35BcntEH%t#)xDcXR|
zZ>+rGln2hvYwZjE8CbX1{GaGBb-dkj_=WE4meK;a^wKwM(O2Kr>a9FAXA$1wG=N{W
z)oGRYXTanZyGzP7&z+Zs+cjODJ7-N907c~*B^5?d@XvlKX6$*P{NFbPoY6ag>_U|l
zTPrwsP#{Y;S1`1fgLONd=(<+#{ZXZ9@;u#j*+5~53U8zUxDM!wVKMZ*6UKMx%(SPT
zsChhP@}od?%%j4C_uW;quU%9mJ$rPkJ+G#tX<Ch#lqSg4bOMDS&`)-+4$RUkrB}%N
zWlBoD`;W~6o!eXZd2<EVjR@i8Zwj!Q1z~|E7OgMmj!|K(+E&Twy&MEN6gKQ}(Lce?
zuiGnm=B9QD(cxx1s^q=7p5Lp_c~VFoB`;hezc}7=_$yP_8tHbsh1qN#W$cl#=fnDg
z!yFJ9AgS~U3Mv}51l(T$y<?0b<<$gOAua?m>KLVb?+}rJ#(u5J@!`=RzZ}{hVto;;
z89#?hu~Smvsl9Uupq<ZkWX-?ExOPDe<2k*WWN*G`@LbG_@q7ZUMoFRkI$?YJzJN1c
z2Xi2RDK<9t5|_*MUX@j{%?rrJNSXe9F|e`fr`4D}d$uVtJMJtbcOm7KAR39SXjRUY
zy7+11t1B-#_2IL#j>Gla6!$*++riB<27lZs>gelA2Aw&ypULWI^){aFoUI@=bY1-1
z+SfYP6c+9)x#avlOU~#MaG5~h@k(?F{Sxdf*=XF$N-K2i7KOVm3+L<QrF3j(A-hQB
z)E*9A{wANx&Im?Pz+x$@cY`j=-kTc{7ys}qy>jNgrl5NAZKL$CNO2PqqR_XagMv~O
zkj=U)+HAHaJKLDdCR9yBRaJBeXPy<vf^-jQnI5jFv@=N5_niH#yGWJVjFPOB#lGiK
z6r)RbNT9;518(9X%_L`d_;FJ?H;xQpKvz54Qr)PU@!Fl$*H|m3=c}C3&G;VKu*c~w
zEq2$GK|JZ&K1rL!H!kn#r=JcILUaZUH*95{??JO7NsNMsKtaba=pSdqH=62v1XLZ8
za~lSr#D>&&Usw7p;?0U7*6Na3dgzSmBSA`8H<<@8y6@P1)`KpujxLSY#%h92VfFXx
zx+FQTezJ}tjJVJNRj*5Apkc+Vdv47MCAH<!6-DXecDqHLx{@vOH2-Ssczb!!=GRtF
z=$aYx8@gPMl;eQoAtI|8AcXjS^$V;2Jf4<TWK%Lnoxu%Y3yV4LSk#WIYKtgU)%~X^
zMDEMlU9LB_?su^-%fsfqE(%Ij1P~gipi8h?1iRDSv<o$JR^I>?tuLoroR#cC6}P9h
zv5mvw&;UAh>f{KEiN7N(rt2LJ;r)JxRKN3eaWB)|jK40N$8K%MrKbngL_X=wxF=d|
zR`e!%>IxySn9YXZlyJ^~0LGpd${GCv$d&cKuQH^@eBFIOTJ^t&$HiYA9+U7;`}PSj
zHBU^U0ywM+>-V@A*ws#9na069m9zQ=pzE-8zl)GS6D1WI9ixozzKnb?^VgN*@<^6!
zsQ4y7|6on2<C7N2leBrN-P)ZycXm+oH|_HQmOgiAMHumo5dn}|Xmqjxqrms>TqAL%
z+7&DW#77vVbO3V8>$|`KVuK;Cyw)n|)gYtTPjbsqkX<HmdTW)hs1kp^<d)Sw`A$A!
zS(mUUPp86DTNEwW1__Z3yWtqruZ&vfWmOdAc#562&+OTzO&J|-MwpX8b`~dA0ov0Y
zcHT)!O1fq0)Tyo5Ilk>P;n_`Tn`aE3e)v|&?)4rimSa#qQtG&}w<4uHxMIbMy{&$|
z<3Mr7{w3XebzA+!4N=1qVytZ4@8<kd%ye#NA*V>CeVCbVSC;aR8>1MMXeT1XM6XUZ
zb{{mvC4fgoOHGZ`0CcnEt9vd9yIcaF{aA8^=5>w#UKc~V33JB|RkbC@C=Ql<Q9`8&
z;!m|<w<(mCm9cI64yJ$lIo-P@q9_WA5Lj(Cq^{#~yNQg9prD`-o6W|mU)Q3lsxhwW
zM+iY-Q4w?JFJfOxDyI)Q6-9v$<`rSvs}m<ooLQ$Zxf{vHT7~98-lBNLJ=Y8m{dbpW
z3$uPMd89*!SX4(o001BWNkl<Z4t+8*GAf-O+yJU3sdVe~>|{gN;ljbj?fl10<y5*Q
zL3RbNDhaeJlvV0v6{_5JaTqgxELJn~J^!lvnx7PsCuy_PEkF*iZr!@F5hF%;gb-F7
zG2n=<O%Ty^NH3Oj4zK@hZ81Ujc7`}SSY!w-r?*N>wyzaoK{g|cJHNsZo%t2@v%1YD
z!v%h@j>Ryy9Kr)6A@-wJ2u`mgv!woc+6Ogk-&p3-$*YQc47JzK;MS#NZyuDnYpbNk
zg+gqoe?v%iiHb|Dy+@er5L(-AUf1<PRaHfGJ-no(B++K#B<~s%Kv>W)6wXB#13`8M
z90;K!$VJNCpsYn`Kl0x%Qd|1+>)u%2(W=LOEEIWx^j+kfFvzi^!_Hp<e0scRa-y>-
zndjSuaKzniE`9I~Tg;%_&ur;=`J$F(p;qJo{M05sr$<0!$L<|<O-p|D`+{Ku66`Eo
zUruC*iJqNoY)f&G5Mv=Z!_DH26<j(rh<@=l!h%hBygHrQn|W_R1?5_}sa>b64}xR5
zJwCL%?UErq4a322JDry<Tiv<X6q9<>$6Do2lgvS8JAxj<OpmJWHy@=s>6B>Y)i>T{
z?TQ~+y=EO-w{6Gea--`yLI{KqC_->BBa;=Yer41Z*D>zK>#$j^M#&%vs-|I56vD&9
zxckoA@p?VnanD2Cc$UDOuXCHJVReqLG$+~U|0>E#`K5c$#OFs1H9Tc!_6g|n!wT=9
zu$Y8S;b9>cy}QsufLUWef{h@%LXbltA=XM>iORNAH(g^blvV141S)J#b+dGH1)bVi
z_+fP!8}>L4YPuKz)FdFB6p|<TQ{Z;H#hY)w>6tWX(taUCFA&iERwLo2nw&DD3rv@Y
z+SUWrv=1`k0{7)t^{w{6?mXz(u2xKT41*%4AxcvVjabRLF74=GLtr}V6b?@*fIcx2
zi|+*w16`Nw%csU&4Tb>rwMG^@b@mr1h7D2FFHIz+WM7^}iOcxRn@x}yX%u4jMM}xR
zVx63_`uB(UYrs<bd|fVA&5fkZ#7R!Nr9CF?A;N>l12IUd41smt9!Yl)!$97SR9rkU
zZW&U%OhET#w19PP)9b5FwMg^#5;`^7ko`nJA`~fK-u%YOg(s754$XvTH>LePWAG;(
zqMG0BMpd0Pe{_NEP9Lh6?IDL6-x9cTeg~<e>*7z*7SSKN>U}Og);jjVD<cRAFmd7F
zAa<v_nY*TptRj^OmxM8}tDUqQ51I~9AtquY%<RqZ&~xZG?wW8LQBje~`i+}!9eve}
zv3oP6X-~SxxP@kUR!g}|1%3&O?fT@WTICCOn(C*!=vC)juyzHbMqffq`v^jV9GFZB
zss^eCQYe5REu)Z;0|Q9O$>XP=mJsAHVYQl3guw0bqH7wOt}*uqaM%=L0}JUG6+(GM
z!`sj=x|_Aqv#=yP^;v*|Qf<ZR?Ukp;M_bsi$BAy@_WLK^G2!VaCy}3Dz}j`|dFPFP
zai&LSP**#OfM2#&GHOT=sW~2IuP7lsUnM-q#7!535)o`dN{9$HCCpu0_CjRG?yC#Z
zlhfg3jO0n$IMs<Z0E-td?x`qBA^@rra(-=iKtTm;&uZA48Vn)XC)#lJ0#vwRPhM^I
zv!%-F2!@DEErguXS|w2J0<U{l@o0kNJZcJ;9WGMBz5;N1>kH8?5%pcx1r-{}c}k6I
zy{@%VO3A(gNpY2k46qo_)1Qc}^l0qJk>r#$><3n`8Qir6-2(#yFICmUY`u8ANJ>)p
z$4&R5s5^ltfmoUskY)!)A><OAd%okbab1pP8e3lbb%=l;flf^hJ`oTiO4r5Y<2{YG
zz=mX=bE~5Hv6At%zuF@^p4aj-YDiVtAL;eFmLH4j9S@4KQ*Sz$<LUeIHw8<w3sph_
z6^8V5aQ~GN+<r+IU;k3d_B1y`dOAqU@o>T5ApS^k(`(p7rc8c}>u<iDoZLJH^zVxh
zoPT<6hw}Va`O9ATx?pON+kI8_bxN{RUfQC2XU+5JdQ)Di5nxxi={y^k4D`?=B8zab
zk7&zYVyt_KRQ3=qcVTyJ=V01it{xsjzb*=qru`%Yr_#x>mkzdMx&)@sIp_eLgEEM<
zrxUImAU?*-!r#iV`sC*9k}BWo&OKQ5aZ(yVlSSE4Q>HH}y8oF^@_rrC(*aW4e8uIX
z&^3)m9)FUUm}p*o<5OlV6I^qS@x7)9*pu#I{caaMI@);Vrgog&KahxE6JAyF@uH#w
zbAK(lyeKO*zFI`WNg;WXKM8yH?iJm;cYlED(mz|0AZL~T!S1R!$drPHi8W>u^lWb=
zn*fx%jaG!!GO&Gc{aC_-C8A!LogiDoI*t_{L2{nftjO!SNML_~q}W+s@U#npaGw~>
zgT=LD%u4+^q!sBDRaUvX9Y*}8Z#<-?OAZui?8s55^!UFo6z_nwkJMhqY`*-*AAjst
zw~WaN=`EW)A(+ZN-y-Fk2yvuAs|s)hX0evZV=p|C6Kr^ORaYJTe^L_O^ym`@0e5YD
zd3j?Em`+l~gl9LU<rm9Onmaze&hLuZ(dzXhrLN-6`2!)u)|D$)W;cCA+a*6Ub?5eN
zdq=%BxA2|pLY4dyjZYRAV>1gLy*85eVP+~^I$l-s<?>R_x%}@u`p|t?EM@`&1KG0m
z58nUq6F?U2&+;OKTw0Ws^3T%r^qQ5llu`y;)H@dIuC;6R+62r|s-4aMD`)BKQZ{^F
z&b9>=9Q?^i%6umU3tg1`?51+L8}k~KA#MvF&o3e^*@Jhzm&`>@vKKfxu+T~RVi$Qo
zx+q@i!uhiY_fijHg_lm7Bopd{r)NrykF<(Hk0Qntr==Cu&ToUaFe~*fp~wX}MJgb-
zj-PnHzHa?SI>g43*foKEeS7ovyPq+0g_)u<jUFAWY)^A>&*kBq*3-evrNwMYcB1K$
z>5EHt{ql$N^up}4dA?y0Co?Ti(q>6YN>a9O-!As=->+D$R`crBs}BkxBGKINMGScA
z#4B1%#*fQezrLbbfevAYIP<kn9VmhvqYfEyp|!5$YH{h+4B^--y8f{e*q3Lx!Rrzp
zpwlg?`HQ!@*RLkXX2fp#HX+%UFDa}Pm=z;&Cag+m=9cTE6&lx!41|P;hQ}6EXdEmS
zc(wY8JPwQDhpAg8Nio9>=|=o1;BvX1>fgWrOQE5mde^R9^{>DFTIWP^aVL!rqtdeg
zK}*J;IT7+^{yXd2<{q$P_71Bnv+x(-j24{7F+~1-@Iz<6a=d48qVs;zX`M$8lDlnY
z*=VffqAhl2EDxljq9WkeUw?HDpYTF(k6{nwDrV~u#aA{LPvKs&7d(KJ@|7Qd{PFLt
zd94!#A%ywVQ+lla$KRrcR=RcGo?FDFLxSiVZ==$sGkawPH{AXVV{f>&=9uE*5@vq+
zH4)+A^zGA&J<0pn`o}iDnEl<)dAY@Rm*noN*B&i6rdxYE`l>t40p0tF!*5^jLSjZ7
zQJi6+)cHTH+*vH-#f6#sSNhH?G|2wi)mMzZZp7KcOzongC@U}LV0s2uT|S!V=-R#p
zsc8q9IPq4l8B#$+h?z>4&RcT|xq5gAJvv%hw7y)My}azyqMZG|H<$z`Es`f`qg0pv
zsm|D*K7D#;Aw(o#SkCH-NUI6@#~N)7Pw8kB84t7?`&1p$3Snn9RQQzj=^Spjs!fU^
z=&I#WkPQaL!N3kiOF~}(Dlodp1pCw|E~-?>DQjNw$fC-IBGO?o5`+A8vP+B-4C-La
zDXfmQGN(+S2qV6;Yh?XnU0$8txf=TmOf@3XZZX{Er*tr0uK{sJEzN3l74p~DJT#ll
zr=GCx?WF6iI6Ysru!sb3AjJ%x7}xCL(Ulnm57TBvBnTmHJl=C?3rw84;Xq!gJggqv
z9ea&Av$aJe)V4*`H2#!jfBYe(RQG4OlU8l3B+#yK@8#iqzq*vtN}XBDE1B@%^Yug|
zKv-BP_uqXdeR}oeU;llHO`Csb=k8?gzGM7(XAd9pTWD<Jkm|7!5s9H@wD(jYCw`<?
zWF$AEAGa^|BNUOP=)9>_c#F{QmAd?UNQZ=8b?4PBzQgior#^XOVPUap-h#z^_{nED
z>~`+HV|+ak2@v0<6Q6uCi|;pukz1_t?W$6)JS&(U9jz3WY5csU;+w`qB($Amo3X&?
zW{#DB3n&0mfDOPV{P)hi3b@?Q_H2wes;0OXIL*)Y_keXk0Pv=t{aL`2I*)GvE^gvl
zI^f1=*OrNqG4&fzcBq36aqbU*HBB4~5DvU}i2WMy5FC9SwZp+TV>>i7w5CoLUY|bH
zR&#AUp&|^+Tx6hOF-$KutdtcVLrg}87?#KCPBHC*jLc`RUSBW@VF=X5P=q1+cIVa?
zsOiND;SL?EZ=0DqXgZ|%G#Ij(4VU{N4#VPSK0VqAOiKOZB#=|;C8NZQE{z>llVO>S
z3~cy3?dJNmGreVXc8ln`mrF`Y5>uy6JuxhjF6%QOFKVT;2O^otq+3qo#ZPk&HE`pr
zD`QQHc%m&kf20(rZF_B9biapA&uMz}Uz;5Vc)ooww^=PTtKFm}AncgMM*@{ns;iI1
z^-dHDvr-qtc1f75>GHV|0|Lc;V<MQf+{NQhzs?2cj_`X-dPXMy`tOU(`0#Clf`TY0
zDCBRq-pNx>JQ|krz@u*gLjka;{RVdrxN>WXtEWrlns;Tz|L!pbMt2jfE3L|qysnmO
zIYLfQq_}lxAN%=hce=g@b^+)%{jNtJy07<)nP2ebrymg+8G+mF;fbfG;qiF5<f022
zoHH&qhR;5o#rO%gaZW#-ZgEypvORpVr1+PTLi3%6nonC1l7leB<dKpNyaBw1A;zjH
z2FCjtuUCyd)J;<Xo(5ukwrkXiP-`RS-$i{PIi!(sP2mRS1OLVl%l>HW+&|MiAvq41
z(8QSKz%lLcBZOeuv}rYubzQGbl@9jFhp6xv?&u&N3t_ke!t9V!4w+S*X|hX=M4awX
z606yeR3@c<VoP%|DbTGQ?8vT-#nh!Bvs4m)h)})U^7^rx)uPm&P;3g{SV&cq?9G>y
zxh*y4jtMb}uKTuh1f#B|Zy)lYgqmNaG?w7iC4f#C9v?Rdg~wa<8iG!k)c?}d=R=5l
z+ge;?L!{!7?qR@pO%DBwGj__Rv_EDJ{wOw_`<pY~WU)2;R&QaX=E0r29Vz9eWy_Wo
zH+{sPi>$Pi7s6r^(pLT9d^IE_DCR#eyvJE*4D~y<veL=Jlm5XgFaC$1pdfx)_6zTQ
zIGqW%-NflbPoqbV?gPSNx^ykcIglC>8z0=cy@i2^_TJN%6y4$0l?&g|Dt^4n6d2v3
zmBeHTI&Y}uS{Cwl(A$kZBjo((a1-lxIs=6eA~>pRcf~AZaBwiAM_tUl4?o86v(Dgw
zdnWSYfBwm>cic<&ZiyszO=xiLc2SXhHe)8^C*010b=zHkq<G#gw8<$lFZqy$qQ8o3
zyCd*C@GJF&-%7xrH9-NIvSge}LlFtE8pEP_EU*DrV^}7qG%@x_xuY3l4!K%sEw*pp
zUL#&MY}k-WwevZ!6GWa`J5Wj^`?_wsziNblm|(;8U7cW4>4BX&qCq0gk>IbhOhJV>
zWGjO_AB$>-(E6Qqjuc%=cI85u+xXmvIiO!`eG6<|ik*^OIkiRB!FX+>YF}Gj*Y}-}
zu5Cp;OLgK^t0Knn#7+Gh+q*nzxm|(VT7MqL0%gGQo<Un-%Rc2rk4ojC#&|p)G3K#(
zQ6@`3zt*0QJ8vgab0ekrv9+&tqEeil@>O<;^7AWiOm9>~>bm6NN&n!{hwi6yr;e;y
zx1L#F&Su63@9@Li`KA3%J!|>0l`9_#3o^ekXi)dfCMmn+mud{}7r^koffJ;nJ?KKN
zep{<Z-zvR+uW>ZwOPyD=a!(f8LV>)jR|k92n6rW@F4xhe>@a9Rx8GtSmCu6B-Z#hJ
zdEc8~eErS-|9k%5D2l?rpMM#%*~}|1{)fp=B~f1C=lc){4+~|+^y!pn(fbRtQYQKK
z>h{N9MMzdN7?_7)DFje~IgO0Hq=|D~*2wnbv0^s?&(b`1{NXSZ7}$()BY_?*eC@;H
zzWeSg(sjMcDl4G-ZS@Cex{;;5BO3}U8}`e#n2Z?0-qD8T(xX9zyJeh_k%9GF4eC-*
z=Jr3<={2&-t1&Bw?%91fIK2X=SKzQfk9LNY)7N4tffAP?IFs`lis{60pw#NKva+&e
zEi<_mWb%!@K{oaIZ7FQPeU0r54L1#wf)2-MF0D=1<2`}4!p&1R94IQ|<EF+1#U>6H
zCbO-naYsv)TN@$dpRT|5;V}4=0V7Ts*!X(rh5vhvbB3SAnWqn7@4o%K^7`An`O1r=
zrlqlH$<lAK^NMdMu`$DBB5U&XQ3Foz9q@X;cpIDdy10Hsh=>X?-*Lmp&?{6$yYWM<
zV)ts@)9AGhmjhDcWv#3tUzgX8IV-qV*Es8Se>*=UXj8J2zVSBtC)l6-$JlnKIcy^L
z;K75=qU@B1ckkW%_;)|d<-WV_z-qPf&ifzJzFibgJo*riJ@Is-RW?FHLKr>jqCO$*
zx+XSfZhti)35F%HZY!(Cgx^IN0<AIp2}%7B=Q|QAFx>COKqXCQ!~PThF`Gw<KWpTA
z8pcD4M#dgfq?ENTo78U_4iFe_?tS^NHB*vNQeU)zAtHN6L!TH3akQMJEzBoes?zKK
zz4Yzl9b)%yEj$tetj2Te8{4qOqgRt;l}ff{!k%0~nah8b#nf*>)><f4RXqXSWTwAy
z=+ABoBd=uAE&kg@N^wDh_MB}Mg?7h#3T=svY5H?s73ZPGSR4_p-xd+0zqXXJSXI>x
z$KrY?DyphAnho>A-1(H3S8&tct|z~sfJskI<;_=LARxd&?_NE5{@H1_-+t3|Ib%m&
zSW@Y+T$fp>5+7p`F%f2TT~bn^bMC-EE;=>PG-^o5Gg2{K_=Q&a>;H7;QJ1m!U3z&{
zE6*vG^8CR)>}QWUEx6mzo(>{IOca%Cv=1{A6K)o&2I+a;Q~IAeZ2gV5+);7s*c)e$
zz0pV}n)Jv6>`zVQhq((l?bK5^eaNZ2`R)geUbCd6RJWlwm4gAaofyf+_yQw7ueu+w
zz<5qwfop-zjU0C$@Oss8KVt|rAO9m!81aQ`{T$x`IK|KQ^}xZyc>+x^4)|Zy?*fPf
z&H)|<&h|6jgb}xSMGIUz81R>=`9{_GukqP_EASN1a!ar#m_2(oBS(&`d8{ak;jd6D
zAg4||xxk2347P*QW4O2rDg>D&kW~U9c1d(Fgw_#`fps#KTY`qMivGu`J|1uX{jZM<
zLBlxFa<@))nV{Gykh1EpEYN8G7-+3+7g=30sWchNQy|z_s0IfI4*^cBE*YXpZ$e8G
zjYS`X5!O-<DdkYVS@vxay?MzaCo#BX%9f0svrc`#U8sD>-+q_Nr6?A=|7?#IM0c0s
zE!>X~V&$@B%Z_PM<cY?{En61t+LL@&uO9XLoNw8>jVGV^7fa@ShtuWazK169+`p#M
zzMZjl5<)O>!mZWAt;u`$Ss#C5GJpKBg`OR4^o_Uk@uDKSb+D3Cq|&d8?V{v#_Xo)Z
z+UO-t&zo6NA9t%M0LNjjDJ8H-cQIdgZIy6EWQggG^G^wwv^Ud()gt(AWf{Z!2Oy>7
z_x&z3sqoPK_jGvTu}3;stqmi4p8MBS?zrb6Vq>BidCqVyzjiDG`t@bR*~99MOFM9o
zx$_q;%FRt_R+wH}w@9uA-p9y5ejg)S`Yqs5pcjo>2fG8^s(!CdSn<cGMz)W}zrCh4
z-P#h!ue!!aps|{6mtqLcqvb{nF<F<=s_TE|XMB5FM!+_kY15|FEcE{U?|%mfA&mZg
zo>Hh=hQVYqEP^Is2&wo8=pR?x-mt_a*pnyNl_MKiC|e56XH6;^Q5U|}NA0LP@-;Fa
zU6Pz9*_oqIRB4FFhydspXSn!7>_)OskgZ`qZ+C4*xyfXzm$iM|Gq(SJy#r|Vn6m~*
z={M*~isKvd(UJ&p0@Vj?BWz32Ui51C4MyvD%7hrjZ13CJb8zKuL+W0nlqX&o$}z_2
z)~8Oq_o0JX*|qv8CnuNj6YtxZlbL<i4L9AM_P`^L^T53m>Di-OgX6!SGmkav*3-RP
zBHw)VHItushItz#S%qGTD|Aw`y+ntbnX|T>znvR)aeJF-dT~y~_^nc%^rBj>X2|Bp
z-B%)cU#nog?)uuSRSdIQ#I?6v5ISZ4x^g<Tw@_53vGbst=rA+ub~%ZPO=R*vKjxVy
zAIG1F1TdLQy!pxty!guNjK2EDt(oZu```1xqrascsI`A(XXkM9gu61!$}5}Q?z626
z$wT1;ZU>4Q**~bx?~Mt-#w;8D+}aXoc7fufL<c4ToBixF11+)2_2;Udjp1%zh#_iy
zIW9>z02?-J@KWviw?~1cjs{jWBNMtVR+Awp`^MI~oy**UgRSJLE%%9uY|PT=yS==X
z47?d7)O5@Vu^TS$#HjkQoM4nls_SkSCJ3r$=y#jV<{E*Kl$3OQ&2MH)M`~lBj~$W*
zu|%n#EVqMb_5R7n3MnZ(-g9VcTsLJ)Mp>nN-`~FA-or);)1m5lwG7qqk*ccxWfN_x
zDbLxTvvu2!5ocd8dg=WSPtvD7`#htryk1K=aPZZ_+_ZJO_av{r=z{YYewP1|uZxx}
z<>{pVuxIZ+OeTdp$KS$Nvlo+RAC0P;sdVe8n#5)ntlQ<}fvX~~Oz2+xfk1wfjrPGy
zT3Oag-QDDhY@bwlPAl{LF1=p~!E~=F@S5=#g}tEZkecHq*r8Bdp%LHS!pBP`{my-a
zUw)j==|h^1Yzqtw;LTTFz-G1DHH*y5%*uWGyi3NE|KsU@GWB22alz;-H*Wo7``HB<
zdz<ZD{WA+mFlsK&Yh+*D0Qg1VH^c5!Xb>0V0Q}Yc`W@JJxDz}IGz=@OKL$0jy}1ma
zx=vPoKB@ZO1AOgg|M}4H3$-=y4HEwT`|rI{%2dEe2I=f$98D`UYDW6Dysj?tU7fvL
zZKZ5UyjsKF-XV>|*9Thb7Y(oU81ChkCL%x(7iz?U_H5tK<t;HvA?@>7FT{)hdu=gW
zuh+W=z;3sn7}hGY9>Fn<kE}w{poL&*euRnU_495MNMy$Go<mz?*8%a8s)+`%X10K+
z)~_EaC2pg?cJ}h+%eS}s_1XZ%*{Lb{1x4?D{q46I%N9=862kKt5O2FXET+qp$&Woe
z_WEnC^n2W|Yky<T{6)O`#{bcyoAEWbYj-kX;bA=Y>@?=iUBVe7Z{WklP72C2wxzg;
z3N<mjU%;(#k>;(pUmSLi8(#ZDcP@BOtDw|RX=@djtGjqjuShP?^c&6_6fkP!z`(hQ
zvDOcU_Hu}IyPP<zf}giku)iRLQP)1q4|9HE;_bJRo14evr>1enh>N)3(kuCF)>n-t
z)5Jta^XW(L^oR(I{3b5aH0G3euYJqvFSXA;`_P@9nX|6Mp=2J;Yx-vwlF)d4#k#@q
zl^9~u7ZVN30G;b|Y&FNq8d;ad>OgjBWP5Y*XN^pzehipD_qIaQH78q?cJ10p0JGUF
zJswZ(ihFQUeXAruRw-=Hg2GDKU`gn(7;O+kTcN>3R%v|!>#*q9EdD=xCZ*vVGD}*1
zPC+(9IQxqQEph7X$cBT(^|RX*3!L`fx+C>Hd-l}pl5yN!+s6yP?`HOi)GYF-Q84X1
z-g9V+jCp)hW`#?Cr@;=Bxw%a?ju^Ev4{vD(z*67-$bVHtLV{fpe0<aSp>cO#7Gdfa
zZ^tCW3zuAY!BdYt?7#G<S5=wx<TE^Q&z*EjtgXMeJ9#go#$3yeUAyVju>&vsCyB4W
zTh6@?JVCn-J(%@#32v`sSl@u)Gy6EOn#H)1?9|J);=SitwXAHd?y0#&q)L_N)lz+-
z?s`X2%LhwAuU?&OAs3z+XbyA;%Wu1!q~}Xc9eNHI-uNFjty#eX_fEv&aPa;|GnoGA
z=ZwAKI%a(E7MEUpA&*a)#y_6^XQQt}|Gs^gI%V>?xuxp%$PiQJ_)Ef+2d|8<#Dpt%
zhq=Xdhx3~L6@+9~Binr~1<iGDH}>H_9%X+TE0`X76z6mRH~HB<gZf{K>wrCe#@)s-
zsNYvN#IV{LmY$ye0fwa(g1f@VFTvzvX>+NtFJG`N6LQOBtHmHzQ|W2wR@ZdFzI;h)
zq2xf3<UkP|D3YWUNRsow>66i2SP8|4O#VE2bScR%mu%0L?8#FaWRD9I40|8i)f%c*
zRb6z&6<4I1&1M-K94!0x?R)ZzstyAlyQl(j=vsnp2~~Li(iy);LuQV6Nte{D2u+S~
z>>+Wj9`@kQ*$Pre*Yy)CA?`RLQ&L<~%Kp<A6-}J{e$J+>0<XR&-JL(;oMG0+Lf7B>
za5^O=rKF^$v3t*60K8r=&%N|2=L|oK9^Gr>DMLd;xciRrEL%LEm)@SkS(n^Tsbv5k
zEOMdivQKz)m%o9wp-jq)KGQ0Hd{rxFrdG+zTKS$Vsa}E9U-Fs)h6ur>E4Ehh{dz0i
z2VTs;i>9!0#jkw%*~eUQ*`-)47V`58c;cyPy!GCPO#1tS4D8>Ro;|wr_lNG|%TGV#
zqZu>#dHG7e&k5l2(HA47GBh&8^xLyD^9!ean)~0ZLTy-aR?6px^P0BxtnKFDzv^Z~
z=3^w@`1%|##t_^(hWPX+BzIK(9BUd|2po+&?HXv%g+l{A_1PwYnHVL9>Jp35cCuB4
zb@l4{`s=S3U|1wajvTpT@7}#1*=)9n0Cc((F8K+TkA(`Ka1WOXX+=g%WO#s)jVtMQ
z000+VNkl<ZsT^YS-^cn$@o2C&zhTWvk0vN=pf5xtzZ;|!K(FYQDNJ6C1j^l#!U`y;
z6pe~%dlU@+2s)3cyVaSTocy%eY?d~g?WBNg1Yv}`L^nC`ghv((i5t<FcjqKcaP|Y6
z^3uK={6?USrvU6R16%f9(+E}Zky6+7<*k0PHbhZYN-FTN=+|%N>=mW|K*(UH)Bj>U
zBQuNVUw(}b-+7%gPand|ufB=RX2WbY<954w?BV+{n`)zJo|^V=?!9XQ9pYl?(Jhf4
z-4c1|zPs?MDtq?qHSO57`}364G%~ZYDJm}3D=I3{b)5^Xmc*dIpgHZLB1OlJadhk2
zmA-v?(Xm5Zz4@f4XK)}rgHs0&A}T79e?2pW<ttY6#@p{x*P=S&>|xwF{uVy?cn0T<
z^jG8stQIQ*tnbV%`a?>6a&dOr{v&xk{|Z8~xhAJ#U@1n{b6whD)Moq*=#PK5abu#X
zwZC9HHL|_Dk?qaFU&74K_%CZT6G)<=@VS#?WTEZbx3A30%abmbORB0W-+c4UDGxvV
z@F0`PWSGvPhx7mKop*dw<^IRt&q>lGEv2l=6rqS9vMb_cPf-*F#j2oQC#WdH1u`r+
zKtz@ba(`UB0!0KBMN|-FiR_J_MW7TYvy|>Ow8?pXf1IW*Y0}b2OOfaOdgb+MPR@Dq
zO`7C<&UeyhFYJtj=vm;|AXb-<kZ#<}no)RDArOMCh0nDvilj`4J<nyCjsr$ppg+$*
z?TSWg9)EG^hImZMG;GXr`97No4eC7#t=<EB_56>>#Kgp*-MV#)4h#&GHk(blQw)R<
zKJ#))2a$~NoiE*Jixl3gv;@15w56SASQA0FfDu7JI!F@)LlvpgL5fBxqSB;;bm;-<
zy-Jl5gFpyXFo1vr5b2$OAXS>wP(m*u^j_|IzwgKWfA`<)&dff~o;lB%GkeZ?`SncH
z2|=lQ_u=d4Ku8kyy+#E|ni;{v$>K|l#L>$q8!{UBr|K+CfN`<rxsEX4aGz@yw8*5X
zF=h-pg!GAP%Cj@FWy1MH@|UV$uEV(yR)0@%9pkP9zYh#bWms5Oj-}gZ;e@kak($R>
z#=YNLnton}*}$R4qAhSfVgDETs7AuETPWe~iZtjNpFB9T{qw-H`FfX6|GQ}?aQnv|
zT+)zGF!1^i>?|elc`hI$nj<9Y9N!RKUhm}&Wj&Y?_iSl^4vx3%^p*|fn`ZlNsOf6s
z^(;-vb8*=Dmlw6u1S$CrQWi^2{^&X6Nt7>M^ArKSgj&QS4v>$vx-?mZSwKWE7)(q;
zLU+np%0Vh)`KQCkFq5F+`<9XQzFrQ>Zv-`((dWl)uQfmXR%J)Nk@kPZ{xKTkR(|~x
zbZfAHMQ=PZ796X6$Afn)E!#u1dHyOd&0IQ8bpaY6H;6BcZ|ptz9zHt|cMM0Gf9fb?
z$=z4dK%?k)3>xR1(|7neCB-8Yh>md)R2~Y-NO69ChqZp`xyJkV??0(AF0He~b$6@z
zb?%p&9?>()Zsy0cZ!U)EDOQ{7sdZeVeo`7detm{}J?aJFBz4w%89Zg<t<Hvu_aJb}
zPv*%)dw6jzA8p-?zS;=z0diEAl`dcDr!KrdS!uu<wOEQ<t5UbvUq{9@7Z?maUZ0i3
z-)Rg+^X6U4#k?3Hy-m@FajBX1Jv(jdJHH^`#!W*%L|aePLr4TdjL-!_)9xaT^aDAo
zYFpX^zq)Rg&&P>)i~~V0617E&CF55;U&1u^aoE!iw%B)r_;Z4_KPXBd23=6Jw7@oD
z!_)`UdDsqs8+f=y55FRwC$e4}VT-tOt}(}W;~h1|XuPNfI2+%a#hfke3f>bfZ-~I{
zsmRU~b3ji`vW_KY(O8dNA>O99$_V$rcNYuq!JjAE)5_NQ`M-)^Na2ul96jyNke*X`
z*OR;}l>s5xCNJ1K%*BK({M~(TLB*DNlV9;Rjz=Z<s~zeKZ@k+_-hSUg?x@yYcLw2&
zcJ@c@XcoTv3RG-%AORLrDr;IxLJt~4fv~2#$qhHQ#@qgmTh5U%l`VHmJ^a%~3N?Ba
zb*y_I#thnUhCllF2%5KO|FmlUSIDyub_uJ;PXJ5W1?AxQQ}4Cu#Tch7ozeeqv57pM
zRkOfm>Z%+lh0-cKP(W7qZ`x)ARdx3rNbSr!TD?4(uO?0}75E4(m$kDoDhp{dMNQ2K
z#R$^*Omo0X;wdQ}eC5uL!7$vND%ZZ|%MSYPzT86_gvVnR#&ShIQkK9!XJw7cS}%(@
z8CH4U%NSoC^Zo_M!etaZ%XTUUs3RTCCxedK<6|ZJu^+@}GykyBQ3+nn^E7HcO{7<}
z%f<PS>WRUoJ^>L4iN|;lt>CE@`{d+gJhOP>v`I~#8AdLx-_}^d32caw3v3DGp`|`3
zu!FkPn)(fVteItdu}OtF+1qrDU32mU9bYs<AG>#;l9L8kL~MS2pKR&sV8?KK1*^$<
z*i0_ho*01ltgw&dowKu>^_geoT!q|Hwa%&#Am%qgfsxfA`_}yg^KO9b_htR)o(00c
z`8-f<?2d^crDbEwWW}0QJ0q-rC-IcI%r^Qt;HIkh9Me-*3CrJ-^v{MCpcl4o$utrq
z29qP29pQVqz2Vjj<j*RfRBHbTj9-8WhJ1Bv&~*vm2;Xe29&cOEcX_7X?Owl1#-OzR
z@C)bYHu2%v(?e90^pE>SRMFN007w36pc}^ZS%@*trE>D%(<d4Q2?>dDR->qg)gc8E
zkx1@X!Cop?>PTu*%~{#Y8s*kjoyyf8TwJHBV>#xs#j_QBL4#L0Q8t+qv^>a4BL9SC
zn470UD{<j|INp)GQ@Nu{wUAIe3;`NLyO<m;?p^nty;t|}G5`JAr25g;A&i3Mz|>{i
z%3b~TAa!u0{({itaaT?<dHu;;z4f?DyUE=mitU-6Py+HM|09dVl~RhJp|$fS>n>sc
z!>y;K>5?)#<Deem6zkzpXxibz(o7Y*e>rtr^I233i9Z*O1SE5OuvvXsJ^r3#Cd_&+
z7m_h8>n)$i^Af(Rv1C7xnXp<V{YxhGa%Ut4DPF;0w%rY*OCmmv3I`7py=rSXF6Bl~
zjdgGK0`X_<_EJ}#1J<(>2_ct)_|7y9kGgNMFuL+BF_aNOje+|vw$uOl^N3q(fR*Cq
z<8vZ_Ojlo40k~S+>Oj_XA+Vy$07*(8KDXDZYI1A1<dSKiO$?$g)PIV_f4Qcl6ztPF
zXSY9P)V+9|uF~57;mhB*<xYZZMieM=W%lI&SxWF-aDwprO9d5kHS^ry+D|DdRrbf*
zm^a&XtJ!Ek8Ed9-^}rSi1OoXR8m4VC!rryBuB-eYZPQX0@=_JF+qe$&{vnM(VBPnK
zh{;b;w^Dzy4F@l$NMUn24lET?0kH6?a>>}9(F-Mb^42+k3Rrx3=64m#o%3<h|MGCW
z#;<&iUfeA3YLd^4T}(~Fbt)~G`mq0SPc}>AR*P78(dS?EI?eJNug=rlmxTu`=bAm3
zn($4mZ^w;lg1PADVu|Ka>==Z&&kd^Rn?NWk{AH>bx2o<HOHgRV*XaEzR@c4pTy$NK
z!1+-#+zsD6_$9Z@ZAXhjlgbwcEzm&fD1TL~zD84v*}2>~KE!niJTtdxH-P~;KMbxd
z(tdBi7YI8#O%5Sgo&^s#`5vWcLO37Db&(EgeP<#0PUVG+ZBowQa%nd&tkGyLQu{;r
zrPPgETJ2yl63s_svda1zg>SYdE4wC<FFuz&)csa&bGR|~wph1FtktSw1{BeKe&NLU
zOjoa$<#uuk>F)G>OTNrG^4~+B{E2=9t`pw~8_RVkM^_rGUJ))>@-Ih5L`3BB^YfE@
z2oFb*lai9sf{D^!@rutU>DbD#C5{3oARa|UMZ1)&Qh|fcQa;4--YWN2o|e3L?@5+<
z!ZLshjhER;xw0XPa`(EAyrJuOJH_NKF2Ffk;YeZj6~CsfjZNQlt=nSZ<@vEvzC-#?
z<I;c9fBN)keZTEzv*$J}+OXCRViT4jSMiwP2kvelQx2x;XO&NJS5vY<m(Db^8KWuM
zM44^eW?(-qk;m!32tgf=1kqoYewmvHriQ#aFA`|mUF;GMD||RgagLpSrxYx+S7iy*
zfxi-CK_J*3vv*C?^A0sR+1-_NCV0|y1evH17a0+fm5oeGGYmFuPWnBfXuGvWE8#Fh
z8anL)d13J-3($JEfK7d5exb)R2v_%s>cI{6&0`wqbNSYDS%yV|++4l2iDn^6m*T<E
z7mF;4Vo67D1H>T1zV$x1_yPxtkawhS<-9F`A2tUY=kMDKN~cx%az17Wy?eAG&PWf4
zoi2KaYu6lJj;9lDde|sonC2lM>=ou>Rvi+Cm+!mKZC@K-pTwMLP>48Ya!4`2FHV!(
zudgmSd3kR|qe1_QZOE`0J31CSPySdIsYjKiD}@ACNkOKb1KCK*OA$>XROBSO5$Qh6
zH5Zz&tXqk`-5SIOskFDNiCFy-Qoj)PE&+`BHgq&Kwf#szzxV4Acg;TD&-}-(aDlQD
z<#Df}YRGi6wzf8q_YfX6e}$SV0i#zFSeX~L`$j%x5u%($jap3yXyQ`$;4@aw)e)&#
zSxxdF&_W!OnXly`z{AZvB}Pt@|MP?QqbqweV?4Q-KP=g)&kEF01fctsamW|r%y7{;
zp?JJ8LRCFeX9F5dy{`40jOpV`>;7~}jYgk#t>mXk8Xe#7B(O>cInK|A<@8;7j`n^a
z(ZdZjk-y<fSUp&8O#aq^v?08St2_4Do<&nikJ|Q}<K0)Af?lW(unJp>A!@%m+3A;F
zrXZJW&Su~S;k1|IPZbfD{0{5u7v)wwA<sFw8A<><YP<In>bHJshmdoSHy1YQSy~sD
zx#JOWMXjKQVRhWUn1_o9URs`hcNT8Uau*ZMMb2^PRpGK;QW}^SIYKnKqmK97F80~w
z6OISl^YuNlx&pk4S*&OM02*qEPp?o-cX_!eV6$|o$;gHPD6wf+SeS5JXP${5<7H7{
zwwzH*@a6hgv7S>6Pz`A3=g*%N8igL@THW$s%a~vDqTfa-z$Ps0-Fwn2x`s@5AUZt}
z|N6K#9qMNQjsm38P_X{DO_IxFLPA1!SMLUeaYbNq*QuZU4Q5ULFo2+U&tqkYvC;k!
z$;!mYOnIS>2Rb|S6Q?o~wYdve*rCZvJB}=XoSRVPIObmrpp2Cun0m=x$Ex}*?}lF6
z%JqK=`#2fL2(#`=O!?I-@PU)gqJ6a_ED0r{2;KHi>F(wf;mx4NB_m&yD$S}^qzVxU
z-}ulT?WF~RL~@eUxti-^*VCX!E^~%>z(-;ETi4ap)ogU8AnCN^RYh8E*xfUrGM<nV
z0jS<k9s)Ks+AxmkOE|;kNd_FP4gBL<wph|*8a#V2D6#}#v-3ON;d6}d7i#+1ob>aF
zEKCCpeL%8oPCuWdPyz|Q1<EiNr0g5FLJ{oM5qX61znYVtyZj>x+TQt-T${uMBgp%n
zO(O5h;oIMP#1zJ&%r;H`b+VSvI@BkS?G3dxw}qu0c7%Dhx_zb-#C*xG5PQpIz~}nr
zd<yvQuJY}u&<v{Y`xVf#Dh~A8O@~`nb-qV3veO%G>+IuvVVX_OO6ipvMTcML!ArrY
zt^`&-AaG9vySNCDS&QGxuwmOH$=laGZx@d6d%k6|@V)8vmI>gkKUD+hmBt3&4J)d~
z94Vy{uU<#e=si#Y0r*hK$BzSo)YW@x#BKWw&wlgeATwU^7CEGQ1{~jZ!s4o_<Nq|G
zHrs)Odj<i0iDL7Xt7qN~O@##xln-VjEviS><+nWLd}P@YA$*$IGco{?Z4EePtd)5E
z*=6usVFIuh@b?E{L@JpFgWfS;d5Us+D8$&p0e|<oJyZ<jsO^?Gb-EBm3POm$%xvGJ
zlKr5iaFVTktKCv^yHhd&a8AL+S$KNm{Na7T;n<?teZqnJo7(@oyPI25l7XyI%t}JC
z*kJY2r;9E`I4rfs#|Ax>^X)TzyT)Mmd;xEbn9U!L#;ZLG0w77l*YCvMHb{avbmD7Q
zZE2A7${<3y29>2IpVJMjg4+E=7d8IVl?INJz+JdqHEw3#{AIO=ZQw<P4wTbW2>gTp
zPiybd+tysp#8RrC`#*v>v?jJD&L1!r_Yd-aL?u=e%<{Gty6-yUZChU}e<%L+`<(Do
z{E&cHY94WVSyhcyOyeGN5QhT4%axtz>^}DmeXGV5olc<#|HMepNTx2@kzV+Xh^tXB
z|Bkrbz;`-!b~6zfnOg5BPo8*HPMW?TrA0cVcTQkZK>6};1j0{VQBl!<^qV%xOs=IX
z;jqv^;AB>M`a^=oShJe>EJ-K*<-esS(p0x^7u!`B76a8KY2vM@`13v2>a&xwvhLyE
z%V=aO2A!$t>W;#Is$sING;yDp^hI9DhYuZG7Snw?Wuet(6)Xw0O739_>mbJJ@v6Mu
zU){;#aEbY(6AXA-$ak^auY341s3F(56shOAwdvoqs=O5}#gyEwG4-@SS$unU)$xIg
zl0(P@PHK2~_|Gd|LY%j4h#d57z|r3J#E|ZHdrzW(t)kVmouiC8y?LcF<7cDoB8W#2
zy5$~XP`=Fqwv3Fjv>mb_<YxMQr+33k(EK#cy1#x--^Qk3P_tNf`#f`H!yuwWS?%GS
z;KkQkgiA~*^BDRyGp>*;PG(N=>!86#)f$<b@0q^&eynVF7VeAN2o4_1R3DUVT(+vO
zse|;2u;17V;JAS@zJbVrb2-vk7=<4nz3t?=2{N++Crjhcc`B{XTuX@s@0SUn8PGuN
z#HJ)D$3d|c-6jBmg54zF<T~Q~(`}o3*=_LNPH`)c3~7i?zR}1p3CD?0IcZbqm?MvX
z0BgzE&YP^4YG2fsdqTB|gXiZ#?HhmFTq<Fy*#o&~)d^tZzco{(x;@i$zJ@=)GF=>A
zc7Kz)UKkPaf!|;!e<d#hiA`@4U!o*d=3&l$A{xQNoHyL0a@C^X7AJl@r(dGCJ=bFH
zy)pV5F7s?@>5)2res|p>ODaiBZ0yiX^)gfAtn<gXxUc%9#%2)0_t9hf0V?~`tJM{K
zjKe4B+ORO2f~Eu*)uOqf(GS)GI|NX-XIa$Y_-7S^XP@;Db4-UFbpf67`#a3=q_xaB
z{vRU@X{1Q43lCbbeVbmp%jheA@-S!T3S-w|y|2buj(R13zthm^$Ub=R$FXuU;>8HS
zwXl(66B81`z|t~3w?}`A^J!$s(l2_}=fT{|eq(2{X_$O|2ww3DfQo8)`PFsXJ3+R`
zZZ2-?33)`uZXJjkdCeL~Gm)#Ml;6~He|2E=rANpe75OQmCAIj`l$T^en@=Rb8l5Yd
zs!P~jcEu^hpl3H<_rPhemY-5Yj?c`x!7)OFvL-c?EmRydW=G_qcRq&PS0S&Lvvrgu
zcC#g`(BNOz^LbkM?>qCg*ng0>B=f`4?Gf2p`bMEJg#G6DZ0WX)ITUsxc@EWUIjz^f
zqw19FMy=z0Bc1abm3>r1sspJT0l!pkEJwvJ2&en&*z4M|th#DbKX{#Z{ThY8OicZg
zm=jivB=;I2?BN1)caYo}V<m;*Sl}A20zT*-33e0u8i52agxSTg2s64xF0fQLXA+i*
zwwbVHggR-;Fm|?vF*<S^=jLfFh4jQnyWnc8Y{8e*0VEk8L?Wn3MUeR^;9RW>h2&4V
z4}@>j)b>G>=|nPKzDD{@dFJDQN{_cYtD@<WZbt&WfJTefXULuZHiC13H~;V5$aax`
z<Nx#EB=Y}u2kZQ&2Y7)=9Z<Rd47~u8z4-5hk>(FTMgL|7+B@Vt+zkK!$+6cr^0}YK
UqCg!_2!QK_vc_{1*dqLY09x|HegFUf

literal 25637
zcmbrlRZyf&6E%np?%KEwKDavr46cK_ySuwX<36~<;O?%2ySux)Zol#UyH^`~vl-D(
zcXUKoWK~z4%&a^oLP1U(2_6p~3=9lOQbP1M7#IXT=yNbEH0a44egXk{gR_^=Z~_BE
z?*H!vK5SQN4Ehq+`Iox0lAWoun}MSVn46m$gN3b?laYbF34@)ZS;i$F9vB!2n55`W
zW%tbUEH`h0h1Twzr*Uf~w|KAqjh5}e6F~ZAqL2isHu~&cyX2)T1PxS8rZx#2{rWoF
zZwX<9rT~9P((m9Hp3M3|);$|7V>&it<J{eXlb4oiwsrItCKf~jeXAR+tY&JwW?grx
z%tNUE$EI~lh2=^9XSr4w;nDv8Ggk^!eDMF-g%k+~#Q$tHS;!jdf7U=3unhY@GgpJC
zMf=})(^at7CBpJ#B)J*a4a|oqA=#orAQRt%_)<KUgo1L$_~S2#>52Z|$b_=nVsQU2
zBS0CF{~v#J;0DD0`#H$k(f|Lu_WwDsT;zV-n>OhYCaOI(tUf${gulOSpUzMkBNoWp
ztRSE{6Qv(qHow;+_~1Z=M*S_v=rPKiqZv5(H!&f^wEvdHnIK^Kcqlx4-!ed3QBg@L
zk&4OE%Bnt=j*f0(z16Ac@88JL(NX@zvzxu)BmPu!(ZELjTz#1|>4=PN!2lIIo!-p<
zXP9hQwx}p=gLpdH)6-KZ{oR3xNh*Pxh<pK`S5c!{buPSU2n9X8tCHVk&%9S1bM8l3
zZj?9-+9M4XbJX<o>_V#H`;oE0Yt;CjQ?zMP0xz6jzG!+q_Vm*w)Z-(9GCCX#sQ84E
zYW9uLXA3*1;nB9jRaoR|#>NA`bab)?_8ihV?Kb@bAYtSpM9GG^Vpy>ORY`bYm{Af?
zu<_V*{xIJID2>KdD<3S4#_Nn(UM8z_+BZASs@hW~a|8t&8E=@a7Jp!b$M`?)F1N$J
z{L&6F#qI<14>#g?cUXPP5H&U;;ow08CpaIbA3?jWH8V|nNm%y#=#xDjIbtvxPc@y)
zKB2Uv6$#0fu>wH-<90fP^1ENG55qFLTyIrYQIQ0BIiAhmZaf6URhcorCs=Zk$BExr
z4%xA1?m^yq(!1xoDswqo{J5~vds~IkYO<awDk?JV4g6Nz*2aTM2yPSOFVhtkYuJKj
zpnPd%_OLN{6lMecZ2F&DAql9IUMm7D=AVIp9(|zrb+QFvKIVasm*QqK=O8~F^@+yJ
ze(*ZA_$*G(E+k!VM{sGkSYP{*pwkoUjj;S*3^MvkfQyUE?I5Es^L>Q8x1pGW!_9$I
z8N#o^Vc#c1Huval&Oo`9;oU&^04*q6psII>x6wG=jr)55g>(wveG#LNg_VUxSy@E|
zTR;(cZe`jl)g)V=yg!YX*9U$9hokTmPs3s&MD5HaEUc(BK_~&?FKwn(k~bMB%Za<n
z4UR2J2m4`jUu)DPp%1y7hbu~Eo|c&nDhPg<$AYVoyrie{!YEUkV$l|q_fwopFK^jA
zDkXCk4T^<@g#<bl)2VD7$HR0pUt07C?{9$NO-$iA?Bz!uYadA0MWGETbM;hgd^jdu
zO>3@PZToyw+K_}i1&+RHql9TI+WTyH+kaifEp*9J{4EZ9LtmZy3DTP=qT)R}ED!RT
z=D`^ObC2v|aO><Y--@cbP416pK0p(hgk=af*oGQB5L^&741BCi&aF4f;)jMxf*sQf
z;ofq+QHl_jubd+#yT@k-2JDHWMT)`mXT|i9+#>vrx6;jIEh(E$qyh6_Nc@I?T?+b-
zTQA$rBBunu-WkJ$^X$SUaysap(Jl1?8e@Cgs0@zURTT&r<S=PtSrK{E0JjFz#5Uig
zkcZNe7GLuo&X&I9rDxD!WBeEG$N6Ox>>=$}0)~(o<?Qc**#SnE#H^;{CI78{OE};)
z#-Zh%>W@TiZYTK}Ij5n$&fu*Y;$4;7v@QvU;|X&OV}FS}89@}bCvAvfr`V+5%cnRF
zjwczRgsOSr9?L#!&I|BuOz-n{R^t1u2pTu7-TkDS!3(keK~7#*R`k#O+#_bDQ>lsT
zF{~K@dhzexTisSSyMuzSeK=hT_@EO8$$Wr<*uUe|b7Wz{JRS9BccdiJWC(m&Ph&Rn
zy=uC`gk;f_tNWP;Z6x%435#?;%+WdFZe3zBvw|U+oMAop!(KZzo=Pt7DAzY6TPhE?
zl&L!B{U3_Yeh3VaPsenkg*=M}C2oAYL%Q{Hje%&6fX}5f!O;)L4T_9UB)ct$vz$8B
z=%AkwAK>lZ&F9*znF<WnGHNbl2m3Ubb{+c%tb%8E($Ua{9lAI?Zug~gye?bIOXPDt
zQFE;<Ez|hDIEDa;zByUOr0~?3!|TH&MXf|(QxGskIFw235c)umKS3K@+xqd=)}d(8
z*lz^Oo|a-Vt|}~sv#GZqgHBM+*oFs11pIfwm(~;Jk|VIOS+`|T44$7|FY=h4EEfYr
z!bTvsFROhY%FFw9u48`{vhmhEVyQp-EMK#|@uwyW4TL1XVuq05)y0a$$R<xQpq}_O
zh8Cj0q3qI_ORVM;Hb3<jy7sqsqi=rk17F!>x`-lS);P&oNU)<en$2hOQ5bbQp7E7R
z<SM^8i`k<pr(HG?d{KE8ZR1gR6NVF>KUM*TQU=lDO^k+PbdUm;&BZk}!T#7lBs>90
zF?vZ?mICkES(P@cIrrpvi4=cY<#_{V#XcQJ4{+o?K0k*o;rTN-D9<>`s;W{f=8DUB
zFIx(B3lrsD%l5cZ&BWPlLPM4`QX@9Se}op~^1)*<Q%7kU%=|4VX^y4|4;g8P9#E>G
zL*~uMWvDIrCQ7x_xz#huZ9m?~^x54L0T#Z;MV!Isaa+6AZ1**;VV3paXk$}ZR#wKm
zHyq0e!h-<mB7BE&e$1$A6|^pxB;r%wp(GA8MZ7pa(`*FQF;~)$TB$Y?zLQ4b8kJov
z?r&Y+KXNR)693s!tz1Czz+ip6F;$%g3z?s1#Dxx1d7m5Y2+J2%ju>ZzhtJQ?{}-?3
zJ3Q{hG95>0zD6+c%87!z2_Ub&T(5)+s1f@;`wGFC@;UEmZL_A$k!Dcu=|h-(^bP$0
z3dlR1=e&&wx?jle58o#qo9oDgMMKq;mES*B%yK@I|BDY<ylR$~mW6F?Z7)RBxnnE_
z3_CPWx4!}o44=`$NZE2%LLlDh?xG--i6`YCS*W3gP1|UHO2Qt{k_`&Sf=?fnGWq6*
z%U%sEckuN!u`|2c!Y5iQx1fno@1n=Fgu{|SJAAu<B*SEo%h!0PE?Vv)N)g!8e{>mM
z9Hco}2r7^ltSv9DZ=%35IKcAxoRt?)IgV%m{|)l&AD3bVr=6-i|HJ&f@5d`Y4H-Gu
z*<be?y6d5bxW8LYC`F1~f2k0w{CDl02*kX0Xswt}wBQYJ{`}rPduNu#4>9q>6s8Cr
z-Gf3C{f>zo#@KZ`N|yzKEy1r%ud5voBI%SLd4K=@^*JI?KsDw#ok4jn-MhirmJ0AW
zlqW{aA?QffHIRqYyI3e!BZm@b4KikIWPpN#`MHDbbx%Zzn1O24<DM$ekB<4|YOigu
z=k@2UT5gA;5eYS<fujHC0IBAo68AvSm-=At4Gna(BivJX!C?OaG9J<_OLjrlm-h*n
zBKdY3@|YeopR*6_)|u}^W?wCJB4b(v20UMEWv~Auzu$Jb6VaCPaVr6BYj19P!(vH}
z@9TJ7UELd)R+`lly!$Kqa#yNoraCONin^O5=?rw!BL3nPI$!6}S-+qG*9!w|vZ;y!
zZSw1aEa%33B`8hWH%Zk=nNTKe(RIe1{nly=N!fVd8$~d26{8c-*CF6}e|(aD6qJ0Y
zkJ>~176oB~X)CJ8J<4Q22~5)z(;!9>mJMflL{`TYTFG>?X-);&4J?1*{${gXs#u9P
zydWF2iji+98J>n?z%WZ;vX>&?0Z?OVdI89}OT3%##DsWR9vk)`@*GfMsh1SDV7gR6
zbAzX^mvamrB4Baz%g6~U5wzT6yX6c8H<_-k_^1R_{)D@y!J>`S-;nEX3VoNXe6O$Z
z`j+`ZjrjmWf(h^QvT9YW!p_eA)#-R3g))7Tqz(%;j3%pTybiRt_B!|T)*v7C{q>pt
z=B>YCj>Ub0MYb($fs$IwN;s5dSxP`ws*YivL14)o?qCfJRkhZ<yzXJA;j8BMkrg#)
zfGA)VOZ+-gSS;GL^h9ytKFmbLm}KIr{XRO=RAn|KJge*1*HZ`H({ob*V6m{yD{<I2
zFelg-AB*YPq@}kqFP+HXBwhj`Ih8*Pn1DM$_6wIg(u>xRu)Q+l8N*4%x%qs>YSp8C
zWfCHL=9_$;wee$_D4D~VxWH}v*`Ggu-q$q=pq8S^V~j3$++l24>^;4Rj{X{GaW!Ab
ziSAb@7KoXA6SpsS6><y-&G$wn_{sz9O-Z1ir4KxLPjlET0D73I72~M$earh(9*B53
zDI++|CqEtG^5QC~SCy|6o7o2!T_U5)cjp%ZZ*>r}8<ha(I->AR71u9Ej5`M~NJ@g3
zhNs`)n+wtKg)D!G3%#Mn{KDmPkO9rqm<+R0?S|E#oWYfxoK3Y#yKJ#SgY|D+og)!$
zUhO5E`zy<}7N+clhB2g2mK+h^dVvHQLVsjs=#X)Ja)EF3^;7*$X17TCMG01fXxGEE
z2udX=ZoV=~EbW2?@(_rl4hsm`2@YN***l28_foEIPrLuX_9A_H5wbr2hW77-nvB|k
zh8S#)(IK5l$zB={RZoJ$Vx4JwnHh;!XR!m#@Wo1U@rO-*mnwM~EOg$M`XH5UB{@3e
z)))bdP@FN4@JuJAb910BRhy%`<eHDs=0;*^pNZzQsn@9+2m6bBaGWj*mz_KUWAQoS
z61Qdr4Kag902Y=X5twktT{jTAU)=opJS@n7&*y#(NoGjbZ1bGjXx@N{_c5lgKNz3b
zjOJosO{`vv<a*GN+NN^-<b;Itu0@RpO9INpho|_T*6Wl&=tjKELx7I?$>TUpL@eCW
zLLRd~zR%2*0_AQno%_>+N|J;_Gwwk;(jRsZE7Ag|U8Zvxl~yj`MiIKfd_d?gs9?q0
zx#;d+9SkkrKnhnqcGeDMCJG<gCevIQSF@=I6NV%^R)9R{#~emhl=-R|E5-_KCW^Y=
zl$K10CNj^0yAq0;nsm8driY91q#Q`sTTFW3Fh79t7`KrrYzfyDfyVU26XzJpR6eF5
zfJ&$}2h|mD+n~qCZ&7QojmN?3@!?xmIViv^<ZIHUOIQ}3I9!1zzN!sQ(?LtnsVe+e
z%e4JxH>ol$8G2P|X^YophewIF!yr~UDC!tLp07-+=+MaDr|;muVteAn%-A{YEV`k%
z30_=q{4+6p$)>5sg{X}W149mvbtDwarXGKnNh1<LKk=6jy7^`V#T<FS$kXv2J~dRN
zKl>>kw7d;SL`IY?iH!s#IaY)$Oo+O|<|@KBFo(}_B^YndFzHvAuHM3ab*a_As`Ngw
z8Dc^c7hw7J)$9&_$KO)Re0hO2GNVK*ni*OanrCQ3cK!|-OXRWr6Z}n_i_BXg-Cx0f
zbyc&hvQnkG>-AJ6X3rbbe}Ln8i6U+}yB<xvYI=l(j9_7|E47k4&}dDo?=FX3nICMI
zj4Zi8Ec$g~Je_sQ4^%WH_*Yd`O#&}}M@5UpU^93Bc;*w;Dc?TqUkUS<2~D?QFM+HT
ztVLT#gj_L1Hyr%w(T$3ugR$Gc<yr<<4}bLOND}i3v<sHK`jzI@d3`<Mu-Kg?>_r2-
zVIG13$||WqXUag81cgV5%_!KJS1f(U2(%;cE|{iy+k1naVl|)2b%`8BL(=xN36AR8
z`#mFQtfUu>MI9c?A%LXCf4D_-l-}M~xXZBWH0!rOKA646jYgqN#9lf*cMMN}rnZ}s
zP8G{{I})CspPvEV`+U7_mS_9xQw(x6);cf#CsgQTaHM*dx%xxg^`w0L6n6~k9*p2~
zuiwy~18C(g>JWgZ?FVk*?N!rv%QJVZYTZwpUI-{=#7dV&1k>Ho#m+)QYwpom1>&)r
zMw<a9<Rua<j);3x_FQ+;s5H{~GaR-eh=!Q|hEN&#Njg5w;Tf~#eojPOk|@C`zhFK%
z9}BJ*eQ|~P77%n4Lv?rkmY_(I%~26994WX$P3=u243(X^Cn8Hzm>sgD)^d;y`@{8$
z7?zhP!uN28diBZst?Qpfg6VZ4a<u?~S^paYjy6VqMNW?G>mpw{gX`VQrSkn%l;dh?
zo6%G8oL-Ws=-=}4`VTqN@w6ReVu5!#78T6Y_Z%Cz9BGpXEPS1B6ioTeHO{PTrg%fR
z?t46WTV)C!d|<We=b<lmM}pkm&(}rC$;m8UkJfG4l6MxAgBU(`Q1F%;)95E2%QuDs
zX3q1tTudp2YRi!B(97hIm0FR}hthUD$yj0L+~e=o;SOfS_5{(7jg@+N!3xU;!+)6W
zt|V$NZGvWabmaNy*#>f9F=#muvn<n?3ep1h*k_0&6(k2>Gq&6$*$S`GCp3)yd6uqa
z8XlaH>V{jq5+e=kq8bhSlN2@}W-J2_%DIh5v1E8pk-o|!)%6wHqwtyiMDH0kH`p-D
z@u){k)y63p*2m8v4+RAh#F2ufZRB=bSZXGVa_DsAD#7<16h=umyAYkLmC@W*O!11h
z!ty*QvByw~j1lawTa4@VqwQU_UU~>>A{OJ3ScdlVI%(Di&NjoE(*HKeXA1aIlqBKI
zPc1PX@eXKbB~B?h{vW8N1PtE0hp35@^0@@WK}aL>u$l<vTR{6QH}6Gy>@^o=wW@GZ
z0;e?J^j%iB(3btTzH3}0Q7?`afRfQ4cFD)MmtUQ1Gzqc-#2w%$TY#h9D;{f2^xo%D
zO?ftbMB1d?Bs#vymo_rYF&abL*~+sSYptSmC!GXY>3l%Wdhw5ok6Kq6+E+DfPk5b>
zp!4$09f_*9{c^nzO9w^8nP`8Th0ocqAZ499bMfA5-K+w2H7`^L1NABInVNd&%~tle
z%-^T$?V#ej1ISB}l{LGgWTRG`p0xcT4~nZ?-qsAr_L|~^c7{&UFMgL754JQMKbp)J
zXH*QTS_bpBM!2NXWhAJo=(t7=UhWhRH{$f4D3OL#Pa6gs2r_G!K7rIZZ!W><dQvk^
z>I7Jiuwgw8NK!p2zk;1{pV5HFrcv;;8*Z|7E-b*sZ8)nf8HcF_8f*bP;S)d$N9F+7
zt~q7q#Z8`-NU$_3cx9X-+YrFjmTf#EtI$=$p}xH+sw(;-V6QDWKVDeAz)D|T3lti{
zfp6c5V#kIBKW`W#M-+NERM)PZKev0aIBXi|<_aW^H}ZcC?dHI`&>^WeD`{w~AoYi#
zxSgGz78Vzav0xguU4)$0A2f>z<HPgEikNt0(io4PHnU3UYg>@_yfPaNbu=$A6Eerk
zg;0yqtjF~`q=s-`X+bx0MJ(bmrXycMp}JqapvG=xlPQ{6W5fEk@s{T|boYmK;~cEp
z6;oHr<oZNcv+HW_6cIF4Q6Gl9?PObFyziDj@V)5=N>|ex4wYa*QnqpJ$SBY@t^BEN
zf(r^`{-c~YRp0&Ke!3*!%NKlwKo%#3=Y7Gt+wvVphj0)H;c=?8rc)Eq5HXr}dnaSp
zS`UBF9BV$33pSE=6rMk%D(KoYucU_2kueT*(d9Ts-<%{~F!Ke`V-?I<e#c22RX6;r
zrJSV&LRhZ8wa?0*)65J9t!SS4Y29h+8}6AHk8J^ok_~5!{aGmX<Z-D{G<A&9xuw6~
zxs?njI)1vbk_h#w{i~YxUx9KsEDFWZ=4^Oo4f0-?C__G9tw?1fCtwI*ZFj$PkD#h!
z!eo3LN^TjZwhG5Pj4*?c=qrAVxI2FPMG$YXlk!}<Nb-^ek-I4ZHwR6?seQ~u(D{U2
z#^fm3?~kRHQh!i0(cSag6s9kWv68J#<!|K?rrZ7RC4MV|NZb>IXRP^y09U(NzIQ<y
zS39_pAPq=Ok%oqrX{BS9@DW|vzggf$h@dLjxB^c_MI{Fv?W{o0f$`k4+oL8tdhVE^
zB)(i<-{Z5c;nvs4wJQVZ9dsA`Ca0m1o>*(^k7ZBDH|P3m!R{q1etJlFYEK~!xuTl6
zjPFc%#R`N7bhYuwnM@4)p>Rr!_Dt^2G(j_gEP>>KihY+8%!ga(R14ci*RyM@!g`65
za1JL(p}X19&v0khittHX6&irusU`E29};nv_cp4Mo?4YiT`;Rlh0YivF_PCe8JIre
zyh7oJDnFcj*R~aa1}|2jg!tANZ!+UyGtfqCiKaYBC^fQMbvw9mjmGl}mZ#07m-P}&
z;5+J|8iOuV`7~a5%D~PTh?lM&ohUD`5-nDs2W_pezt((cH{HJc$j=RxK(iB(8u{|6
z)A@3G;qyGh&+lPl(I9)*|2zBoNiC}Nle6<>YVFl=kJ#&GmbbP)3h7Fxf*PLv>Zcc3
zl<jD-dBjLk_xk~3iD?DCj;uMDeY#|%z-<LS#v%HUy<-?5BRz-ce!5Te$?oK^HcMH)
zz?vFYg`e7-A20|l!$r7Qv=zLXRA%F`n9s^T7r?{bm#+6{Zz#ZHGW9bg?pA;19Tvei
zwMTxHmoi>vD?vsGTBAHD`hzLy;il!j4c6^TEwGqywhz&wxtieeI`2w&s+R_6$*^qF
zklWa=)Gb--6OCkMb2)7O))Sg<zq3I4Du1pUJ%^l1a?ohQbUogE#HXlr>~S8Qxp-4&
zza#8%T2>;5g{I|iePLANrE#ZE(M0e<s?a{SbNW@jpOyRq;kp_4TY#k?=1RL3a*3O7
zrM6zzVPcBTW3SVE^DdT8tFM)c&Hh7WEj$=`2S7GmtzGYEbq81L$w1&$kmp_l_E$WX
zC`bA``qbjW!aq<kOzCtueqz{#sSvH@dzOmWA}4biE~$vX=@_W!93H{33lPKN1fK^I
z@q2!B4Y@FmiM<A1QTS}8Rila9ixy=hu|)+&l13v6@Lc?ncvC?P)yeSombags&}QB1
zY}qY>;eQYI5sq`lQA?cr#vwAqVW4Cx{W{3~afg;9&4ILp4X%Wkb$)w;Pv}W}umWMA
z-@Xu~5A4tH4)u7G6a;0#g2DK|Ghe>?fZQwnXh8t{sM-7}BO#%;y!}^56k0^wa<M}L
z3yYJIdd4m{wa@MBUmVhVB$6I|Prj`s(_k4wkJt4@@1NCSIK=x_Euh`>go>xH*3jLl
zJw8SJm+Kz#9>_i5m=sAIX5u%(U|DsR{S>A5p0CD7l-lZa1WHTJt8ij`L7Pf)L;cu-
z0Ff~LmZzyOK8^*(Io*}KHeRd(O8qfk{30?AX>uGMsa#B%`rkZ)QEsL)Kay6X)6vj(
zE1XtW?L`|>QNz%J;@0rd0lj^7R!3mbgjwE3j+_OM5twn}Fn^D_D3`CmAj88r9#*Y1
zFv<_)2uCs3`l<XD`SeA2kiIoM$1ccXOH`gAjAk%&BrL->CPJqbB;wZk7X2vlCax)d
zFBiUJ$GGu!i~vKld{a2_vC=aQO0iPgs=NgB%Y$vT<5<%~s?cBi)|fZ7oBUGOJN+1q
z)IRbU`|Ub#4hEZz-UwOke$r$#5$B_6LD5Rz4)#ZSA@U^qYDmd*?F8R}&3+iS88fw6
zmVQjcW44nLQgcuyndnfz0hj!3sW{GmX`ud=4Nd(-)GVo<K8bQL#lrK76tC`$bK5iz
z<2=NFPfFa>;t3VTE!N3ZFqrF0nbAfd7bNS){>BO$9{Zcf%tsQ9-bI0w>!Kcw;W>r!
zcY-jtuT%VQe7w=WOJsZwW)Rgg`z^ZGs*(61_1$;|ui^<@Rh{OVYKnK>vT}p5rp#Dz
zz0$>?z(bz3QFZ0>+S3m&ZmWM>R+6nrLDocYd>YoX@2toGFXHeV7jq*uk-Zq2!Zv^v
zrAFWhoGzC}_6CanR(A%Qsn_$w3Q9O!OV$|+v9U8I9x|*&p+blI2NkKRrsmf9?EtpU
z_P=Hsbj(@lUeO*q-}cH!^qHw)|Endd;8o}RK+}OD5S(84a-v&NGNya{SDB1Ln*Ox+
zwOm4DHBR0fZb`P?1>j?-7~s$mwQV?%ckpeUWS=59T+%FMIo{;sH^CR&&byVcOgo$8
zs61E~7RKmu6TMXJ;RNiRopeI!(`|5QI4VP^(-&ReKs~y?H<06G_Ruv6A1!H*Bt|%i
z#G*eM{LL2vOoEM#&|RUYYr1DJi(M7+94`(IQT99fZ)Ig?+gB|05Btw;T^M^Wd`K7B
zngXIfw`ZyB16s`;i=0HSEZR%+Z8m3*uD~Q1UVMEuNcXqe!&dlFBW}9j0Cp6!ROuy`
za)GaxDRC#SV&_49MT&#&U&0v({IQS*k!j%05=mYWXRH;};Mb=k@#Jn-cGk@p_$khV
z6NT6J8<Ov$Y#ZA@hAHz9jpPGHXYlaF4Vvd94SG9~j*9_eYpXaz{)#km?X#q+2G$3?
zs_F=T`UMarA}xsJSRUuOMxQkhf{`?_yyb5`17ZDInI#RIC+UJ{AqJBpW3}-OtX1F(
zW(m!a76Zwkfqhnv?gzcpUtPSSdoi#>g>MaU7(wXj+~$AF1+R1cI}HZXV!PeHjNUr?
zX}*(@owPuJ$?@HYF1c*bg5h!WJpOvYzo%Y(n&cy1v<;A5GdN>VS3}TpHPs_pFLA-H
z>;33}L|KIqd2O;?>-EtZH*q7cSb=Edy~td$4j3Bfa+|InJgT{bbHvU|D6eWeZ4o6S
zMZo@HelQ*@G^v&1uUs-EGefxYyQ_j(N#@C)=5j;X8tg<9(`xHpqfST&F2?FxST8Ls
zRZ8w@B&@Ayq6JQM^i&RJ(B1#d1t>1+)AD1bRwJd&O0dj*s`*akyZ%f0&}ic;r2R$3
zYc^Dy&JhRzqG_wmI*>3#9g07n*}(0#v-)UMSo{(_=2pJBAr|94ZD~Q$o!}02{5Fi{
zUHN=8rStjm_MK5MxAh7y7K>3#NIS`B*m%|~gD$G!P{)}c{UKVXT1fFNaoUi<dne%K
z4xT@LG0q`R+v}ncvK~vMxkDe4gcUP{v?FJ{u(F+GW!JE~931alVDitVM9VMik$;JQ
zT0YJiwOx7{ULS6`Jk=9wMUXr5-*D&fXc%XJxC#6C#9O`eUm9aIzAt@-_Ij~KVu!GB
zdRhVTZ-h&D99PS3-9xhyz7dav(zzDWuFjy8n&w2WsJOWJESXWhfck3!O@HIK#Xb6V
zanMSr85<uAdW)0Kt9iI8+CU+J|Eqy296OW%-)wiq;LKG*&$aQ!{ZgK$i5?{Y&ABZ@
zYII4aRe|+{FGS+ZQY$S}4`>ndg3A01-O&b!-2(r5l?_xT%i1CH<xv#{u#yjy*dot3
znUkBZeV4OJY>Wit`dS5hEFstl5Q$q61E(HQyoo-4Cct#V;9KSc2>a%6-K3;(6rKtU
z-X<YY;74kHb32ZYJx%!KPh)VS;^Igh-uQz<7AMev*Sx&E_nN##gfW?-G)tyXdKnNe
z3JxAItS9qlgrdmY>{ZV@uI`I<d7j7-eat{#ets7BOZi-^&F%eLTXeWW!f)Np+{hHD
zj9=?gm3Eld<P@peS<6|KW6N{`y0hq+8(({$X|CXthILe67KG3<ng|XnP&-!~d8`b#
zE*3ESqH4C4d4a?NP`b_EP^PdO0xdA()sWem@S~#d`8$}|E}8f~tb%gvorHOi!o?yC
z4v%t&dm4BlAcsg7jNQ*wV_#RP;lEKTt<3FZR`U?}ga-Ylmpksv+)B9f8_<*l1FBl<
zYL9nrJ(j+-=XM@0BLGop1LmqCuXpg-W_00_EL|&&D>3rbzg*!^yqC-_@Y$ALMxaV9
za~kb%iJoJI-$5nS8H?f*_s^0IMGG)q4r`iw%or%dG+sBa8HcH@yJOuqQfH33!OPiR
z;Ew%s+If_g-W0KRT~3bzn%0D!-9Fz}Yn1CvCtWkEg>hg;S3|OT;Y+Y1q0=%u2Us6z
zB4@I}htj(J0taQR)OEd`^-duiy<yzrlBliT6<qH31i_WcYExpmBFe|p1?gyFK>@9<
z4!R*5p`#{SK-iqc<6AQw2w7gml)^{q$;D&VV^Jn&qX)%wxv1%MZ#*s(Yde(){xL{F
zYt6oAR);u{!}tEIIVsjkdE@iV5U*pF;W}MzK>>G2$h>2pa7Z1pFF+=%(#foQYaleK
zXTVpKkRXleLkg3a$Ex4vL?dd+%U(UeGy2WpDncOIdQXUZSg?Q=^QKO)C0}G_0jUrT
zOC6@bHJ(snI9^$0v}Q?c^nMs<2>rI~KO;~k8cFeW-*U9}&>>I_TU`dL5J#6%=v5X|
zY4xa)K?w$Sm+5G{9`!W!b>#+18PBtH>i12hfCSv=OD@CUxk@hw|9aaLAH`pEce#@S
zXAhN?!5J$5?p<1OGFvJP`^cEd+3mY+tqt5ec+ni=L<<=+E+|MFasZt58lL&aAKmIq
zyhID1vfMUTc;h8tm#uRc#yrN*k0;a9^ATBjPt_BVr+;@*Y^YlsGXeGrqK6UuBx)Zk
zC00|Xwdwlx0em!*<5M9Dc`~&O_^@rJc{@AA&uq2`W@9>lnHIEICBcmSL{4!4+lv2Q
zfE1@EYZIR*k1vJ|A|;mL(JDIjyWGthBc=OC{P!~G-f;eG;hiBwDqphkgXX?%NeAGK
zWXY_$jy|S&_!aPa|Khqihcf)Cj-4OTchk%EbBRjns=}gVXJyN)gG-z<Rq(2K;0!NN
zXr@zNA5Gz5_v6ltfMyYxV$4FxH-$0m5CcI{JMn33pX@?-9vg{%tYW%08MTb??4Ql?
z{Fjxy$6r^0Kvv5fN-8S%3Ov$p@D@_R`01^6DR#LBM7IGm4is&-T!;4mec+ZSoHhAw
z!Bg|HIuH~N%EMHuytiN%zoi55iMq-EQQ1Y#8rd6h_zIKeBF+gU<@;)-sU5llE{w)J
zEEl^d*1e(+cdhc3V<ZC2Wk$C30c@?=W5Zz@olN5yC_lI28H$nxYzMYWa&4nTE~Iiy
z>?qasl)|?54eMbDdr0yfgh@jQ%}U!?q^b-4SXVOVRFK3m0k$;d7Q1nG{^@;R>Ef}v
z8Yz3PZBE>3Y8UT;3rTmZwVxG7(zGII%$aBd$4{*?es)NN=Z~K3TO-->J^2*R`Ef6b
zoD9_H1`(m0eUM;kqL)t9j!=-f-`qHjk2T==rG)VR$fuEmE!75!WH|rJQtem@FQ_|l
z1Cz~w5(JN#sQW=rPYBk*ZL``SW7JgD?wHy9qZh-+7V7fHy)p}d7vo1jEuq5?@R+8-
zj;0tqc1s7VCb$9U){rJlmkjnhJ6ZdGE{cYoyWxF{<xJ27{Si6JZ1db5cZHoK8O~)D
zm2^7+R*9Lf#0En-7WU+l4dy^9eQz?8KS#vV&)KAy4zq9H7!Oti=qQz_f0+CV1}Fod
zfe23T!}`bWL6Pi+-q#I3aT6=`EgRx05N=M%uV4E?od`-D27Zi!TE^L*)uYo!abqWX
z3WzzE1_K;9b7Go}&@L)X6`Esl|Mu+r!xsI^qpts=J$$5IzDbu{Y_4j~ekjzrZIX)t
zVIw3^!0@y%!`Ah%rKClBxP|4D7GeJ)R*_R9`o3NcMu|9<kS?l$X)j%5;D5JVqjlEl
z%!`b?n!VY7nkf?q$wu(Et5@LJkK!t%{`qSJ=1kZ^1qUc(4d#Kxh-j^C>HaL16RS&H
zRI0Y#^=Fa;38#qbNKj-zO1R_EzOHpS7wwNLAp2D#k~7DDj}eLa@EP^@Iw;e&|A{cf
z`kf!?SZ2X-7D*lq=HOYBPJiA3IdLk2N&h#7n3e%KpuspQLNhAFRpZVx-0bqPRL7v;
zr_83;Xq;C;eAO-~@{HreJqi}{1$vHsZ5TT7l(gA^F>NYQ-r4bvVZ%<);FaTiEd~OA
zGP3wK@oMRrp650r43I_eH}_dZ{pdfaIf|yE?qleQ=f}uBSYOW>RvL**bVDC-*brfU
zFl-Pf?s})6^@dqT%|^L0(v9*g6d-q1isH5VbI3~QzYbKF4E}P9=-o(cuyi*u#cwq6
z0ZoEv?4*p{hH~vuPqF7G;)L&eah+uNIjaYe!+yJW)`GD!RH6U@_Rz;fAs*UwY6-I%
z061si2{Y5Ue{Iw7sA8gUu;)PU{GL~nqlKwwp2CpfS&fSe(Cq8D_I&U=kI4q*39bHq
zRMfba(n+=xJe4)T&om7~@Kk}L6wYMUqJ9Ya)_7MEs_oX%L5PE1OSXGP9F2~S=3bcx
zg1AMk;jSO&^L->zSiZPY(n}Fa|NZ$UDLI*4+`U+H;D|tyAwFgxUS@{W>d!+x4q?)7
z3b46iS-O#V_@8EVwrR)=hu`#p7L9|fsQ6xSXas*;q7YGhw{`;Uux6~(`E0P~0cfHg
z1WxhtUk=QlH`1}PS1bCmUg^y^7j4ls*J!bvL~9fN4w2=)bOUZ7l5B0yG`$Y&zYb_d
zY1!qtV(=lXZX`&$Bva8jsg2gg@p`0NtaUD~7xyCFvMop!O09FvvtLM+{FIBz6IU@+
zDHvvmLT9Lg(>x=J9VPlW9OnwlfrnQC+GC+7tDK;&&Ib;*<9KU{iwVM%9HO98QrBMV
z+h<z=h?m3&WCqbmh>Z?LlhI>`X&4x6A7f9j)5$LGo2xXi3@%JlF{s!S6=$3H(=cdu
z7AO=EnfB6LX<Hc+s{4Rf@m(<hqqdyx8BI=G2w%++3xYRsv0T6++OXpwQ&R<;{sD+T
zrP}D({+KZvA5+QHW`R3iwy6CaO&wG+pcqrTgGIze+V2l1>wP<HpzB3Feilm+an-=M
z8r9gS57-j1%B#RpMwZHVfncF%p8c7Pr!_JL_z46zg5=~cU>@BbCgkYGH`>{YuJD|o
zQe*@auvBiFt3o$*wD$IfBN<}BDb5ZX&88y{XC5f<P{)MF7C;FuG!_i41rBAkYt)<5
zh)~7S-7ByD!6QFB>@8c)IT}x&JJ8yQgiY0P#RZEH72dsZU(>zN>aEy%G+Y=$ovGQ`
zRXT0G?bx`#aydR2?S15b8FQjoe>29_a^r4^*}2hLoNLb~{47kUpI)Y_+ZDM`R4?_)
zjg0iG-1}#RV6kznGOWQamOK=(ngW2q8paBb{xF23KjVPAGacrB7=@h^*e@99=WJRG
zHMb{+lj~SF=sQ8f7BDD%p<=sa(|^AYA#UI>-`cjym}LY2VMhUcl7eD+_SekR;KN9@
zUY&;@M}0qh82%L>w38m}7uK4x$j*LfQ)GD=&RHqsF?|dJD6`3d`~sm`o96Vm+4Kr2
zea<@>4^1Zhvj<7{KT4M3z(cY}4Gb`f(o24Q?H_>lNzL@#i3CeZG~z$wx|PhGrqcAh
zwWBtT9NSE6d3dL6hJ)H8X|AsUXaSE!+fWp)7TbQ_Q-VVQrM~qN>6NZ_?jcdDJmx%q
zS5$o8sHtBe_0r*2YnoR#?j2tAbXh%Z2HW>BuhUN2)Ri$y{-EAHOyZN>-<j+CYNa23
z8V*A^1^2kwLWve>rb6Ddp#$7|2z}#IZ55syCMyU6XQ`G=Hjh6MrJyLFFN_`4tsNM}
z+qc6c*r-ok%{i)MY7HXBV4@Pv)rp1AhMegULmC4bT3SaQV^vDyil-F9W~jo_+*9NS
zNq&|j0FwKj*yK3rCKpnEc-<haOQhX)b%GF6#2~8YiznZ2CRvA~(qN}A1eDSxuv;z8
z6FbSY`J0!r`NeGsh{X~W2{iI%g#6^A;d{4+yVIjKeQn{w$(-+@N&G_{aq;n-&G^Hc
zv2RP4_As3@vTm0GrwH+2@WQzC)+GjR0lb;I9TnH;6fA$nyjjZ_-eLna&M5b9K};$#
z@<PlWpA*yCH^TxfKw^y7w1Cyt?7Ca@R6|rBfm>N#sTz6#I~n}c-10QcM9qn0C}GL<
zp0KyoH+;3i=cDQb(MWA!=Ag%3dH8=~Ic<bQ?ah>BTYn`v{3SzYC{5@A`Dj^ac<J8;
zBw^ZVphEB;N&!C*b52<<2blQ9*mlKs!#Ug|8VYA}or@<rbr9N?E{1oEfCeXlskpp*
zT(qw^d_)y4uwS!jYkI9;)qbo2-9QjEI|I}PSJKk5(skbnX^UcLqoyG<$ft)g{_DfA
zD2^m?^9kdnN-68n7iBl>escjJU1IXIPf!d}x@UM&gm+z}Q?)~^HCA!2h9FOkM3?oz
z9}WV0gt~UQtIO*{ts03$nYwGbvYKO%U--)vtF{RoAtdxUmV9E;&D9eppgS}d!yE?3
zjwax)d(>}28W`$Y)?yLoi6LJh>3G+V{9_gpVJ4RD-Xtz*mt)<F55D$S)SgUW5Ykm?
zcDrVu7+LpkD`!Nwzr*YdzV9m&v_wM8ZjCKq#&~rx${*gs(4?<t<`<%%z@UEt`Ta2h
zx}>2*z9_1m5F1|5Y#=sdf<wl6IyC+?50K-B&NaL^A#c2IK`53d&zm94wUx=f+0@q8
zF*MjE-FYw%<Vjle`Jm&;Y>_|>E<k1`_r3SuBN~39*!Sl*y1_TX#M|IMDrlCfzmQJ=
ziQ<GcvV(c?vlCgg;KGBbwpBc<^YaZfZtgBF?DeJ-mN~7?!paW+s^5zT8Vx%}k@>6a
z#l-1%_Qw9$4-hD0OND|CR}}xJ`-Hr)XVm#B7~{L--76o!MJla8UJa43M4?EQg9C#2
zr@j3gLt^knME*sH;1yuD!-+Xi0sw0uq3$pvGUF#2OuB~&In;hnxFFQhf=uV1<T7<E
z173IdQRYm0-c<<*nC8M%?fm3u=d?4~hf`9>Dx?ewI2r`)4?M5)0@KBcpteRG7B5Nl
z)MPmpG9#wqShPsdD6`<onE2%O!5=&~_(3tpyi3lCjP?4x5Fi$1US8f$t4cw80y8EG
zk$<YIkk<_^MGX${=l@-j?EMTnYR=Jd7}_6#MB%|i*Zp)<WSJ*f`K#}FwpdBLk9_5t
z((qTs!6X0HG;WRezU9vx@2E_c1&v?L4IYTIFY-N0GSDS`06)YnTk7{+Lvc$plX0No
zeiAyZxUho>#>{VgMA~4qJ)_$ni0~Bf_V#2%Gw3<$-|hj%rN^ncTd-jQB9W!XRwS$b
ziqJEU&phdy-`e<vdm)f}nov0Etj0LRWO78OdOr}@5LUjeW3k-Un-M>c!N3q?_viC2
zA`PTXMP4h8O4;Z<jz9Dk&gFsYArAJexp~*X?I;#}2+F)3)}*f8>NvB3xTGa{dEk`~
z@WT#rWXceOL3zH>gxgkL4$;B6O+>D2npR=u)||_m6bjz{Vm4j|3Pr<7GjO_EZ8?t*
zVLvr$4;yI0;vfH9>svvd$B4|bw!|pl1#}fnr&(9kf-A(QW~qF=-%Z(Bwzgk}($c;Z
zZ6!VHifeERaNxigh=~&*=m#Pp%Tgx(J3?_?6S8hVgC^dCd&n^OXR3bd%x^1-#^6Xz
z8`jm2WagtIF)x1edy#NAlr5f2R&fe*5KGeVAX2?nq)<X3CbjC@X`TzbQ3E|gHbYr_
zr6rz%LWgL=Q?_4LLN*qNMU_1}noa0(tQO7OjPh7;hs<K5viWlIde=C!1e@u~v_%T}
z!Q>6+(ADr9gQ5!W_%UFI?70obe#47IRwvFm3zAaEQDfHi#_&_JMIreU%GMa+vmF>m
zd=D5Z#3JF3)Hk2v=Pi20+s($ZMdPf_%YgNM?bzt}v`zBac;fB(u8$D6=H*WPDAnEF
z2c4?AaKYcg#!8iytu*F%KGajp(?6uMoiW}lEb;pIVQb?bp7CXK!cFm}Zq<1hCirgf
z5sHL=yimU6Yq~smqJK8@CM$!n14eki8X6flm8%Yc=HW0~D2W8q?nA<{8tlwUZZzA-
zXiYJtoGAbI-*ij=wkX!od$V}Uh{)TCE?wF=5B*Rt_Wdohf3**jq1KSuZ{@ig64Ybh
zq}az}R>$2G4mB-f(bw|MJrNJ;K|`h2FMYa^9PgQK&zG*3F=W%VX5Y3pVcerHa6OpU
zre5bz{BfTsJ3X(U^`R0<Un_|c`GSdiv47hh+2@;`Z<V{d&WOLRPOf|RCh)D78*{S2
z*YGP*ap#8N;r04qcrWhKt+6Q-h(db#&wmkd`Fdke0qu5++zh?WUE%E;Lz#Qvx98w@
zh3vK&kg$53Ae!MCjt_;e;B)9ESF*keIcj`ytSg(VNNXZXei$BMBCNQW9JnrT)?Aqk
zE={M6`*k{cKB`UKZ9)M!tkoa6(c*27r1Fd^@n-{cbg7cqjjMQJE`$Se2ffz}SOy!}
z#8~Jz=HRNJ2dK52*8xDgGdFW1NI6#Ud}v}DwpDXr6Vl=}xL)1v_oWbw=@;{LmElM=
z?3>f|?%MtI)Rhiv_M5h#F1V3x$ZEsqyl~cokT>FT@10NaZWkuuy^F66E}I!bD`Mpq
ztpI!r{Ji)eV6S2|eV31p%i3kLPia|MJV+z~La2BQdA=DMT3S50!{Af5)84BZnr~N2
zn&rK-6N|g?864%&jd4F#ppcp>!q8hDM=z3aIAbxk9U;4Cn`}zJ$OGg{)?mDc)q8g!
zSdrGtNt>x#eqV3UH65u@Jg{}Nuhdw$)lY-9r<_Nwy)07bKe*;xz`PKGN{-H$3GObu
zhx<8s`5Qj@-J6#A#YGM+@iu&w@?FE<f~UC!F6zwHa1-<gC}lK8Abu;WqotRb4*{i)
z>sTaw@}zk!ct#UJE5O>aT%t)+%+)WU+f16YPTbK*8A3CN8Puz^TxtAg8W!<8>FMNQ
zdin3(QRIl?(6nB#V{jaEK2#!D{ah|bc)UrBBKuUVmvL*l2{P?8hINhE-;R5Dla|#U
zE;{$MD@HRQTChfxU)v>@+tMWpG5^`|en9i7{eS$v-&$HA&5m}q_wB!IvrLDfE1e~N
zPTu^dTGSKph|VO^vhdzq`0x__e%mw6k?FMHKhbZDP=^-kU09U&kmAg_&BL;KRR<Yg
zYhj{NEh9ag<JoS0IqXdR=PR5Y5G^A8N0>igx}Nj0#xwt504)6%*4UB$`D`h?;-%Oy
zKT~c&;W+9G;U<aMRE)lMs=L2tes7sb!37DS2te|b?i6}0*SnLFoX<oM|9!kJ`OU8R
zCsy)%pPiI<Gk@iOJ#^$=$&1FX6qz4(5T$F4R!gKB%#gJ+#pfOZCJ;X<nnNskqS>2U
zs?Qym<FLCM&wXamBp&HkPTaaPlA8L=CnjYZ*K|GgKZ7n@fSh((?DL9~oHTSF9e_w4
zZx_x(zUqqmlMpv%r&S#87RR)UD4}|ly%a558o{^odSdK&3@|1EYfo&BZdf_Q4DR*7
zuA;>QRD_=%h^p2hS=<3>^dVx)wTGeZwjE(+I;=r9YHt`r!>teHjq5Mvo7!`^kCvvc
zD@x|Z)X=8j)!KE5bd2Vtu&$gQ=jP*jTOu*hjQ^s4TEtA%h>9>3qxYT2lAF{xJY`=o
zs-W1dgYi@LZ6in`#7eN?Wi{&e1}2m#mH2|LrKCX8Cwfpm!m=@0WN5_c3+1TgMM##k
z{XO^CA9BqfeC9jo!q>l~L^i`!;>Z<*fmsbIBsLeQP!zN_|G4ZMqaX#g>e9?F4W4d0
z(i(=3z^2ULqyNazxH!}e&Oq|u*nK%)qyt{M3fBJ$PkrO!Uaz@C12`MxQ6)$Z{hcY1
zCw6&UG5cbnV{`3<L0|~8u{-yJX%*Ij10Ew?*-o~J-BL#b#GYyuw~5dFCtN0~7dV8z
zZmktgBr!6$d>6$s1s?+`W`f$ku2K!xlh<(FoV}-(I4&ob7%0#HIu=SEg3I6;r;qDH
zDonKE<~_QedB&T=V~;SorGlBayZ~n+|KR=|yZ^)l(sH~P4PH1u)N{c$Duc!FtTrTk
z<z@crNFiq+|4_`{6Xq^-M}i#?Yfp?AS#Dgpe#!|^JSCxztR(cUu`Eh9*5N0cw|SUp
zW8J86WzaSmG?sFc@v9zmvjqe_enB~EhY{2y1IE9Bz+8*}Btzsc0zb~|Wg<{zFJ7F8
zd)OgU8q9nhOg<Y4$qEHTFosGx#9SGLjiJc6-6I8qPC`h(#R#JPGiTb{#^A~(Y$EB}
zGB+G#XA@easICCdu4uj<_C#*1?~ZOd-|v9*=penDoyRcK0SO*@Z(fDVl1tDuf22hR
z1z5z$r)PkBkmT0SNgIm3u9n`ZcnJkJg9X*Rk+y9Qi&|f<4-0B`<sg(?sW(&7(b?#<
zv;$otjcw8TK{;A^o$K<j=ob7y#TN(TU@fvJk)IGmZ<6ig@M={vuZ-9wk;R4ic;7^e
zcV@0fwo6)}Q6$S0Gz3eHAsQ$PdSmM7_<nWhX5@pxShY1;Ft)^cwb97tVWS0Sa&uwz
zg&)6mT=8?0jI+bz?Nz+k+i5#rSpFLcM!BM{?&p?q*K@URxr;EQ{+;cyPLkgevBSkL
z?}Mhpqa-0kt}lmb!?Vbdnj#`koaUMou)V!UDux<n#0(vCg(eqW84*SlXsSH*tFQTu
zx;4l@Yn?daw^o5%qisQsl`Ngo*0gpELjZwg3-bxLal1gA?jPR9%v{W(g<Chn#C{%p
zL|@&`HRexfxZI`c|E64&ZrdAbU5xs)rG!I~TmLB-(uG0bi^UMM-YYzT^u)-0k7_qI
zp>XGW^J_~l%eOC!F}(9fPaG_JW+y3DVkvRIz0X=|y^n$&ltmwP=F9vffzrc>hqKH<
zFBh%DH)qW$cLk{_MBZ4Xo!5^S#P7)QQ%!?it7s0>PY@Ug+!5O4LmO^e8)~9*P`M8L
z6~0`4_P@sALtcX$jrTW$z3@`(f@*z6H0gWbBeG;UHeB@X8ES2##y{~AIYf$%1lBJU
zA;l@qr+O2v5gU<Sz;k@09tlbYhEwiq@-41)rPwPr$}A+#)4O#rn@&2in;tk>uB@IM
zGkEyU3&23f1%(1VA}1(1^+Z)Q8W9w`_%@%0VVfHcY(*SQ<LIc@;tL{olUf9pla@q2
z@m(j|^Q%VDT>dtuS(C7FdFWG=cRMn3=D3dghKNb`#q3(miPhiI;pq0d33|Njj2p}l
z62EW4X6cx#2g60#YB@>%Gy~zN(ew3A0k~G1(;AsI+$LLXoPn_&K;8DtlwiY7o_RX1
zMirQ{G~Ubq2f@n%l7`VwXnw;<@65GH9{=s39dVVd`V7sD5gS29cl&fY+bhLnOZlk2
zh(XUBGwi~QwJXO31?kL!o~?Dh>(9r6L#AZ*5lC*UqnZ`r+)eFdvaSbkSZkQ?J`uu^
zY8#^?F!YH=1nHjbH5bQ@D+IkVtey_e=EUySvCLR<ji-cVDk*lk7)E@&kCeLiQL#6E
zHCKPc^r(dK1wfEUso;1b;}Q(hj0sHB2h!?@->&2RBE?T+Zu`}<A0wvTN&e*w%ge+x
z%%ZSXbM$Yv`^T0|!nOX~WfOndDSA!kubN^1TW@K^o+?TS3%c)~SXy+y7#bgHY{l!R
z6%6T#_g;lHqH58ozWAR#ie{r2+m7??i!X=HSL%y+@i<%TeG1IK4x83Z#D}bm2SR_-
z{-=1q_~$0i^@n#VbI~=O%dl(QjiLKVpTpy<q?{^BB7;GlaA3gAx6NTaAgmDu)XgP}
zk27Xj(Qf@~-C)y^k|$t`pwn2$XL_~oM313$W-#&S-K(f!%_;z9d1ey(BUzoAks8xK
zx;<pbcum{K2eQ2Bgs)s$5#ZTaW9hzXTHxPkifFkbu%dZ&(Amt3d<g!3a{=;e8gK8d
zNK<+q<yN$F1ix+%MRkm1=E!6D&_$CHn{2l)DN0_qxj*IaA!00|d<etiSbz)`5bkdK
zBk^a*W_tf=^_NkI30c!J<gH@wR<$SH7;}8K9k7`veF*Xv0PLH7n4ZXB<*JpM-%ehM
zvG%N^!x5p?b=#ko5!5v3hPR0(R0CH%)iaEC;_f%P!B%0%Ec5b0;sEN|OH|)%^cOg;
z*Lrg5T*D}L4$N{9t445OX^2x6EO)bg9ECJ|W0P!<1j^l7R-WeDmxoFR9?|ipljb{`
zzy0>TW8NSGI4a()&3TqBYNAJoX44gEIqtlQ<=Ir=)1>|Tvb!4ZFDSkB{o*evPc6iN
zl~zR$&l6FHQiNj6H3C3WkuX(D0j3>r=Wkq0y^|fVuYCTm+Rpkb>MmO2Gjz%zHA91>
zG{^wbASogs-3`(qAq~=9(o!NKoi8Xg2!b>y-67pw0{6J?TK7M={KOhqi{YHJ_jf<f
zKKnBV8B2!gM=Rr_Zx<&=)fG%&%$<+u;nW_HP5CmU3`tyF`O``=$*Qa$_Di2$B&2L>
zo+URlqkBsXr(kP*DRnPR-ae^Vr2N$2Yv)J7*+M^-SQ@@fY%WWWBV8(vRmGV{*ULcL
zbujTa;m-51Z>_Z;v{=E7F)5U<oO(;-XRQj$%Um%seYK)S^!$LBl0NU6%Bt+~pf}nO
z#i$JrwkI7Un}2F3C|$Mp(;%1M1ImN*E6Q{f6;o}hHn&b6Y9h-zA0N#uS=w&%2Bfie
zBaCQi^?r<Ek<o~=*B}KFNf&QUCb+|^47~dg42HT!Pq=y0-UVYf2aHSk7s_>7qKmca
z38ns0qCz@ER+W@MP!%eR9f^c=nv7;OPQ1}_%NfH`86uCxjwn%>oJbsocXx%LqdU^b
zG{$!;E_E9xn&?g<W6Gb2<-zst>&qG4+xd5G@ie>sB9m{C&fMljhsam>K_fcX&FXls
z@EJv;e#+n!Q&ezAAI1?pGszVb6nxLb#PpXOf(fs_PnY`Lv4oBM%GS3Nv_6DV`=pLE
zwR1a|4e)Q~tMHJKIY6u6Kg2Zj`a3m+%I}-tCTVj*=pW1#KUjB<vF*z6*HZTPo3*R=
zD`Gv>e?2pgrz%LyD{od9WFbbOefP$jlo||a?M*LrRLpasd?(TIycCFKG$sIwfLlP|
z<nJ9l4#nhH^LO)k(^31j36iRUH(p-tl5_4xiC?NhVO`zb=7U*+=rbm1R!xT7Kz<aK
zmg2p5@#4^@^^nNAZ*Wk~$LE_fL7dmZTkNy5Gw6-|4Y7<%C!r8MOR-Q>B42n`7G-K`
zD#`)*uLThq6~&j+*wR96;&+4zq37m)z@bwa`XxcNu$J><T^-5b;2;oP(iRqvo{EYV
zoU2>f^(Hebs;VMmXQ!qj8y5paH_tXRk&cw>1DPamDo(w{rW))i47qjMH$*5hpE{9h
zKV^AlQR48|bC3{nN4pylaj%&z`}4|v90K7@-2Hk|2O%hZwdX<~3iB7eqA>B}lTBuF
zDt6OGGil!sgxu@g7@V<5S&@u5ScuQ7``-U_d^p@!V!QIXR?yKe_+{j>Q`BiFEd@Th
zN}B{`s|q#x8=tsatt>;*<S2$mW=r_)_7t-3^D;&aCGMmPQcnXW66(+`%ugqk3izFB
zdR3oH{DDB^wAAMxug3Nh6qgC8`0QsJd)Mb-v@PHXJ3%2`U0r|rYyYOs?y<<p$!)CJ
zNDCJ{ym~2zynTuGR^{Fo+iluj5#1S&+I2TMdO%kQLY)Js57k`!6y;NJAt&lwFgFfu
z{j@^|N$(H)^-IHHrXDvN=@>G*_i8Ic$eETl;L<EZg@v~p%q{Sho3zuul)QcO^Tfu=
z$_#YBs&#$n%(b+RmfjH+78PM2vk9q$oSw<chdZ~$@;S^k5&zpYxP_v!^7p!7iI^r~
zi$uMW2pui0oTnamz8CvHPKG5D08K#5)I<hnzx_Ud)3{}d)f{dy)#ywIq~Pxj@sFu1
z%Mo<l(PSKc7qWrIT3W8RXOk8iLm&KeHw@SAz7{;ycm|OPjDeZI<Fz^qJ3MmWFH3>R
z)hxf%*PmG%F=RVCK)+UQ;<GkYerU?rGNP3d)8llwUxG2@vp^-6H*JPIQh8rW-sbGV
zLBztm<iRDc2W>289Fjp}bbPYYQ6s;=Blod@CbEtVPBPn0R$_qF=7hu}RBqaNq#G-@
zu4`KU!Q`Wn4sOGLgomXVHRPtp!u!TsVFkUP3ctNam7NB7|CnvcwyuveIoPU`s;jhj
z*z=XBk;ch9o@~DpT)w+GaGvn_<hSuSYWr*Pf?RY89eO8S#G4kOEUq<CX<QVX%CiN_
z1=_3*A()D-61tz=^EC7r*SZ93gAlJpe~J{A7spZ2D=X$8*gC)BGF&bb1x{^Tog5#V
zA}=JmN=pktlr($p;jxp~UtRL7_|)fJHCsZsv1;q;Tmjz1e6h`cs>#(NXzf{GyqZO5
zmVu`ylAOnwntr{0mHkuG$&l!c_~h5V?bjq9g(<UR2+MlH!QUtpPKqqx5B=5_iGmAQ
z7V2`QEtr%vPdA;|QZ*`Ouz96e#-UpmA7g6yx>wn6d@!2zKyVb(a>%GYY_#5_qsc}n
zMiwNhbkMPtb-m*`gJuz6d>%$~lZ1XG5TlZ((~<G%`^*}uILH!eVv{p2JlrK5Fn4PC
zabz<scr}i|L@3)&I{yfno5?R2W%E@rbrJoZl5wA9zPc+nDaq_|K2XOa97{0i9&WeW
zeWhz6yBb1D)p=zx(Zj{5TMYq&c#pedm&vPuf4aIkin6P4-x&I_Ir{Nk?yx2QGPWd+
zArBzRxPbmIbhv?s7zM{?7i0Sr85j4%|6)FZHP`#QsS)CFss`ZIV7a=w`VZ~&aa47U
zeNnH6g+}f29l@=llwI454U3sOkOJZTg`!D6rKERn56INtIKkR07ajLeaE!993h4FB
zLnnl=E7s+I<M3gTK60O=1N<`5&EvzzEMpB9AGl{|nO+DzGP8BBHfYivk?a}d##}p-
zYR<l1-+~Xzg}29G@_M#oqV4lAgr=sa@8J6q2INva1lbuUCBk}I>NG^w_qw2|LVM@V
zCFcOF!7bN7Y3cRd_+8cguk%NtOVJ#aC>pOhs0@x6Em3q#4Ej)JVq(zSlXXlQAqEDh
zj-jhu(0D;p)6_;G;z2`0LrF;q3TS!)K1%(ZaL*T^usPQr8r}aKrmK5WnQ(;`{pyyp
za#UF#(2Ci@UV@crS`McgY9>nW*9iZkN<&MUj_oiD!J}8K*g7G?dUMvU^XJf8Ot8&a
z!yQ^`@T9p#42;fWJPvFQYu05+*wDu5_Exj4RxXrf80woB@X0dN$lI!p3!pJjn2yb4
z?kwo|Qdm2UA-Be8``vbFWeoMW2tr{yULi)P`!cKJ*P@=5*7&xIupd?1P8uO3-}-u^
zLtlv7pKKb!B4?|N)oox^G$UpjyVpN3Kz%0@hOg;x`on!#JmNivj`?)0m2R7#h<3Sd
zFz|3zGHr4@rt4bUeOs-TK1<%RJ0Ph*k^)Z>OFd*x3M{0hr6ox2JuIZ3r=7aDw>Nw*
zx^O8@XaRoSP*NIwv)@bvl4a-Emsfmd-!Uw0Y*r@>oPUgdQ~)%Aj`>GQ4*^BPq{FV^
z+<konCR1%Y%4lh2g{b@tQfh+uxh|@$ao%&bx|<4@C7v=W-6eUeeXXoZfFa8!Cy<Ta
z<B`Z+6bY+Dn6||u?%!%eXplzI)Nxm@sK6NxXy8q|@Oz5J!qA@A81_7pq{*vs`=ojW
zo0|{X7&!(u`xtQEEE7b4p<6U=Zf;dV<6uoIG5L=z3mH-9keb&hxiTLwt#OO?``ZHp
zh*VjFYg?-y==SIeWl+meD8pYF3%0#72d*Fqt7?8z6C7Oqmv(jKTkzT8a+v={wli7P
zlg#{lWMl*eW94&Qd8U#hOdJZM7WE`pTU*1z$N5CMcz<_c)}W70TeQ}nE~`%3Neu!o
ze=U~?n+x_zos=yaO)pK|z#s+;#ajS8%jI!DkH0oyTt9a+!K-2w#Bl~o%D-dhjuM}#
zwZfOUTte5>)}}0Hr!aa$oW`l=X6%pZ%2c3t%EHv={WgTbaqbep*nfPKL>+jtK1di_
zBM)g|?VGMeeXx9=|BX!8upeFWHpVTJ%}ZxM1(W|F+&tvWiw-Rnf_vw$E2BxicMLa@
zK)Sn3cB}QmJ74+IfLjyKUN#P>{qC02vl1Y<)p{~#6EyU65yMK4n}>_b3Id~}8-|!o
zkHE`49+h&f`K}16u)NX@kqth`n1Y0q*$sX={y|}~1kM-k<<~)LS=9%_$nD+TE<nU0
zz<-STEyenRU%zSBQ`F7fU7B66={1Pb(RuOE_;~bCzVPmDd9gzj`k{}nBn*e4dfFjF
zGVm^u;=aEA$)B|~L?wWmJh%rF=Oe#<tvt^hH_)7@FqpLItAt3I4xz1+7RQ(DFvPwY
zynSKq{{;K={>He`$q0)fbc4T<8afld<F&q3G9A?t5f$Y!iPAa4SUYQBZYlijPmZgI
zkTX3FT%*IfLogXLUk2RGzO%|Dh$w1WsNI1O$vf_$sXi$c)=*HJ?dFnP-J5GD`kMYk
zlgH(naM3@B>9v!OkB_c#CAV<p7cb(U2}G7U-z6C;<4;q4x__+*X3;$8DRB94A<Bxp
zUIiuuAq*`P2Fl+{(yMqHQM83!F7lXV!dzHLP7bw7CjSq$C1rKemaFw_Vx0?pJo^_f
z;><)DFn$*kR%l@|U|d~YwFT+L(TIBHH#CrymYVmcaY;On2Q(wg-iMaf)(>6}91t)m
zDzRe82c6UV<~b8I(~r0)si>%2_^)11Ty}bMcv&%ZxukauJTd(RZ;SNw@m}v)ewyEb
zXblK;2#QZq9Qr(1QP}c8_$`*LyZeQ5ADh$kS6gZ?uDCeDSc=hk6B)%{JOM}e*%H>P
zezk7$l#fSku*~P1cF6`X66Y2JmgyfAQ{h@8LK6egaI!tLA_-~q3o6b%R{>6mNo95e
zBo%@zrP?=J8l8OL%?lK@ph;O-F+ZRsQ2)vs3zJzo^KsoCV_!r%wjmG*S)ZlVl@*tr
zipJ|9IUH#(FCpW%Yh)(*dp3Fg;VG<fYozsP_8a{XDxY=s13$ZTju5{QkeyoKl*dDJ
zh&#^+3kwUEM82~cd^*(R{yBo8@mGE*Prm*@NEBM_cYF`Q(66t`<kL<5Mrh&PRbMt=
zeq$Wl7#JUPsZGl7?+1Im`!~Rn3FF-BXgIcplHVw5+>RNdrlIjL^K0%f!N|9QM7D`Z
z9QVGmaB%m-yVclRtQ&jLM6+q$sRbB=&>mx-txt+dN>G#l@^l4#N&DZvVN|G6I@O=-
zaCqCLg-=aQN%Q;-g>f1-!9+wv!YSz$;;H!}SrdBRRTwm48JYwHNUW@OlJe1+wYg%w
z?k0N;SH+%JQPicoBu4GjJg>TFH}GcX<b2YFTVG#)nswQ2WT!Hms<rg&5q_jJuX=7A
zpTPTchJ-ps%lX)46x?NE*uuII<31=-kP02yeET>l9A5^<oC`zAwPnX~q6r0mpojT5
zTTA%N{2k<VtKM8}O?MoX(6<V0j{d>H@&BUo32DS9;wIY8qwHF`OAhN(FBBpPCF_Mw
z@6M+i>~*9Nhv0I=YVvb#C<6v4=I~r}*rG!|h<eF0ODU6J0JzZ^F23X3{{Fu7a4;+c
zJdCh(?U2q3)Gl<uP8}Jy`g=Ig+uQLy3?JJU@#y=;nb{VGxG<aYJ4Yn(C$^baxB^uJ
zS9*41<5U&|pn+iuF>TPFe$LbSCX4T!5Q3t{9%vf&mP$|Wu^FpJtO`KxSn`n3lXHg$
z_75LqJ&mh@Ta(D<2sV`zbWNGYK}+1R<~?+6E7n-S(Z)9;uyy;wR%IbP``S>Jg%g_W
zlTG2%x<K?VKEsnwesm$Yu4|k)9|vPB`Hp7bVRoLX8$n!rj{o$U5!Mr%TSS5u4{J%8
zfb~hH3d_IJ-v;wAFr})r^!V&))RS1gsXfoj8|8+ep^oUd)KpSVqi<uT#IGG4<y>8B
z3Sg9<NXf{^!jCsYVHtzga{)Jd8(|!4-p@sIa@m?SBB`Hj->lrQ&|pqQTDuXTkK1*>
zd&F0Lzc{^9)xNNI<n6=7DG*4ga#;4ajNiEJMv(^ODuT-QiVS5^6^qX+o*}Y~2y8?V
zGxLvU<_R2)2)I6=GUHL$7Y?jTv4*IxMDO`>wjh1mT7L5?%j&U0B2Evh58vLBgMp`u
z#TlWfsHpFS#AuI4Kg;R~&Dgt$1gVc^D0sF|@v)HM$Uc7?7&--Kq)otX_Budv&cyP8
z{~n&qWCqHPaE%3h^5}c2z@q$A^B>V<Ks@meKq0C*!hD-275WW$tKA6y`BV+Dv$0oG
zH5S;O2MclY&0hZ`(NvqisK%>T=;y8LS@ugV!bA8SHD|hNRq~<qd8<hkHZr%G%%?=^
zIVzW?mw_Vf10^)Q>`W{y{Mm!ukfqj(g&U%7c>C}e8s`F*Wreg<5?&SBSUoXy<=Wtc
zeshSI2n}~ERBI3iCTL_FT}mDu%@VR2N7_S$tD3_gJdUB+kQfyGAr@*P6ftCqn4s?1
ziOP|{_tTSA8%Jwu4c_T)dLI1g9%IF)?7Od6?mRvPw-~)r@`0C{ni?vUiW$dWS4dMZ
z+Q`_~kc|m#-+%!bj>CZA+;&x5g-SpFtCA~*0OuDqu4Ce;t7)azy+7aFDG_06lB!|o
z*30buCrJTKLlwLZfJaZ+`opIkx*Z$m$z29OU0(xIC#Ojp_1m{^fA;rx)>`!^rl)Uk
zT(6m^pMv>w*O8r(a>k8l*xQvli&3iB5`V17LF2r;$MV%t)37heIYWVTUpf#L<@xz2
zibe*dhs}V<NSU3LwGTzn>GwV`Y(+F}S4pWTSEsm?b@U3fHVDnkwNy{I96dJtdD&w}
z@ZKp?$P-`TFJQ%ejwmR8U(oX4vNTtRi%|PqE*ffmN8WV)h+g8k7vA~aLaXoOE9J~_
zJFBCkBNcah+z?p{;wPRK7~A%|Rg=v^a~tF!Zg{(0mQYYl9i1C;O=Wga9Ryr7FPZ_p
z5`OzKPhxLpiBc83_>hDoDK#~;yqpt64k-}sPZk1h9944QQ!2Y%Q6y{}_!=W@7@&dz
zwNNL}GH6+qSLtsr+W#|svEbBlBrhwcE`yVtoXi_|=Le2Su7LTft-GkF_Sk9rYBd9y
z_ijU5ogh4;bfq|dt7d~~pdXd?k|%>U_~(y~f`r*brl`SO@teW(^YgP24D14_^QTMK
zk5Ie!gt`wy@_v>b&QP&Z26qbnU7SH054!V`hdCkIzU$tRM=!6OqbX3;hghP&jM37}
z|43!~hCjz{<$GyMZvKBCLI*qxfXaS>N|3-_ze|dVr-fS(_QcZ|v^vdy)6X6(o^e33
z7_y<$H`cs^UEOaBz132N^Gc2}EO#kuS@wK|F8w^+`rT<vu@U0HM@I;^u(h+3QBuNn
zYCY3llu=>-u{~afe(rzJdO@%@()lBXr_HuxZ+Vej+r%`6FPy^I8>+g2Lhoby)O343
zOQ?<A4F?d9SimNT5R2yvFDhdCS9$%n@#ha{fMQNHd-32f{1eeftq1-@R#sL`Kn)!2
zHy^C5rJG2wmJ%#9YzS~(O}>vzV$@BHtu?@MtEoE}7Cu}{S^OIu6st(k6~tglZ#m9<
zVV#<Asd-t4>g(@Eyg5TFevy5?QXm_#-IG9fPa%Ud{JTCLoBIac<)|tsg$KPS4}B&c
z^B|D1Z0>u(C%=b~WZY^c0`e%-+;2j%XPEG?ayK2N&k-?krd>8r_dCd&iKC}vMm7*<
z4*H#*YV<TdDF-~$VGMcEpjy^oJEL}%_pa)1ir-=X*PVqZw?-2f-w(#cTm8~P8>WMK
z##XfP^hW3Uzh?^)0W<SbfB&Ei9`)x1UT=MFEH`ewwvYKyxLrXho>rU`SUqfJOOUJm
z^wT2jCoaTg4DkpGmx*ZY5b)QeVEP_}(`65laUjSRTJT11oO6pY%f8z)^j%&2&(Yu+
z`G{#H_6rwV8QHSLYoMT(uQu|CT#J{OSXr@8{mo*RgRyB*0NYA(EZhxeQOO9Bjd)<T
zt83wP1dyf6w-<{F=H`t56djl<q6hL8Ai#7R?8yFeoJF!pld^9cIn6blShTp_Y9&#w
zFo+#*2Z=s5t_?8nNPih1@II(~Kk$G!m@lLueJ;Dky@(a<QiD@TU7fgpcsSzsmgW5X
zd^woF_;=<jfL!4*tTydfC68Itx79Z8hv$CbzopGzmgN>}n<s3dx`E@&eOuPTF~u39
z&RFyQyN0@m@g7d=NUn!mUOMHQlb}twR-x6iW4NFjDq_4Eyu~NzeNe=WMlA{DPdhod
zHI4OXGOYYe`}R-LXXAjYHEyuNGkczbe0=I;+Hul}vs2QZtd1BR_t2n6dwl8!A9>=X
zBuT};$MF*Y-P#@~xxKR^`wdpVn~)%JU1Cy0JP<aLZXms=EJ6Cej!{+1F<!uNN?!iQ
zPF>N`_SLG^bA6LNSR62CcV`=iCsUZG(o0KAs;YV-=mOwsj%R?W_9GORBr!QTWPRNl
zf&%%(<?_(%iJ%~6lf8WM9&6drPM|#fyU4`*Zi^~%0w0Mzo(fE}4b!KbuEC?sU`>h7
zvM6s+kZO?Tt3l5400`64&Sf=FOnd`4NN4N8ERWw?zd!5ztZf-~A?1Uz>!9nj;{3@&
zOWQ@O4eX<e3&ZbcTdIu+vN|9FSOdT1_HAcWA(zx&*f24HJ2kH>v&;B%sQ~8cKQERF
z>99#CX%l+t);ftKTfQ&5h>@$<4rK7XI~x-xF0_ECiQZwDJ7ETW+^bRFF&ByHvHuxc
zL&WHMhd)u<0e@G>CaaHXShv2Ly)`0*0Cg4Oe{+u6q<Y!{a)Q5gf$he5l}>?&<9Pbh
zi##1(xW(q?CaO1u)fHszf6AE7|5RP7QQy&F&v+~0bLw?g<<`azcqb5jOP&{t9vOV$
zd$Wxe|B3-$S~|LfzkmH&4!`4ry0`-b{Na>Ef*F8fDL#LW&sK{3CFZ&W&oM^adkn7S
zFGsPz#Y%9xYb>DFRa47H)Aj$;yN!)73q$9&r;*0#fO2$tyb2Ixpp@p8EzHjka-6D$
zLdN>~ltBTWoSBK6-v6VrX(jps<dk;ji~V^WFdlkWHtlf7z)*A{yFiU?Y1rSM@%^TB
z>;Qep5A8`Hv|a~InH7blHmw(p**&!WL7k=%s%^0qsF$|5wZAW6hs;i4`(%N`Py<mB
z*QRVNE!`_EMfTUunKsTkGlp7*sBvSk+~FS%dM=Gw=ii~DeuqOV4m3E45^02}6my7}
zcp%+ISQ!osMCw`%y?dU{)9r!xb@r2$x7`lNGD1Q^P62_GEBv^S5HyqON>j)mq^GIZ
zuWnV$_AhVr6cj9ej#^_PYUUx;lB}$(jMi2m_?nJcD#byxv=HZ_iD|^QZrCSMi|*JX
zLCW9cN}qot+=Aoc;=<;gh$vyb@OS-d+d2-m7Tb*g$8yC<Tu|9_FQ$*FDW|@4<n4EI
zLBUT6m{XwEI3Ls!Rj{|`VEd$~s2J31$v-v#wjAz)S2gW`Jbgdf4c0<lK>@q?WnMf0
zN_OsfK4M|PN|3x0j;E6p>Q^?*<tLZBcVp&PAgZWNG0+x6)@{j)U2p&AGVpn~*O^C0
z8VTiOrQV=I<HEQ+8#qJo#Eo2{hr?t*G*_FU13km^_3PK7qN32J)DIZ@`H#HCG}TE9
zOCL{z?AG<K+<)QVuxqEhMS8knclZ=KDHK$Pw^@!gHyG4Nqed2#1vz1B%SgyYV^YEA
zOdnDQmz(bV0B}yM-hD%r2%@H}Co7AdoD#>yr>VsZskZ7*t1kmKs&g_ECKjykgV=h}
z9(<`3DrN)y^!c<&Bz~BI<-LLI3QJg4iX<(lTiOA|Ykbn!*KtyoNPJ9{-HIh8N{^Gg
z;a|ec6i|puNU&jMW-hO;P9!5G<>m6Urbox|VN;5p-S+=>I(jWA>pUusLmuzGvPkU#
zQ4_Z6OC?N#uZECOb(F(Kg^@Xz!>fjwI!#GTxO{wkF3t1d+zlIyEHa1vBCihH*Iz1V
zIBW_@ihrmLkxb7;t>*ckX{40-8Z~Xaxr}n23|U!(4&YpTl~q%PU!|~CE<3viAUEdj
z!nQ!&!VvmrV}k@zQeIvJP}g0{F0jVg7Y$*C>ZJUjEIL_Dkd)vM7r%X4HvTPHBWwHu
zM^;vAyb7dOIsoB4H+DR#tf&ATFQ1VJBgr1EGVN*5QCZH5qZHg03QcTVln)OFhLnx!
zXLlPgk39sm$(=z;l#V~B$xE4(RFavj8}u^=tu6HQ^z6X1hJ%wc6bFZfHqh^)>qD|g
z0dIHL_swSf_|AR#;#R8I8V+|iH;L#jIOy)ca2Ro+iK6TXA(dW~79#2=f{Mug<qHx$
zs}j2%)Lj<KZb?bUP&B3YTio3ZxP_OTovPKapJ*ogAFTH^KSFEt=$R1cxR6~cA{<qc
zMA_Jyhl+1D%6(ZhBP`k;;V^{egUte_b4p<$BfrfM^@?Y+67U+5tTI8X795i3?`4)O
zOiUyHyh9OSOT?I)o0BUhKS|ffiY;g^oUi2!ocs0bc*V$TzQOkf6u!W-l4rL{X5xd4
z+?^?;Sry)v4IV|Y=E6@t(%o+U2Xucn+ox$9T7svrZ<<;a$KGOj!XJi+l)+cP-ASGh
zu}lU_XuSv)F=C_~A-?g@zMuCCzz6E9r+T68t*J&y_NW-<z&fAjh$OuYhJn6wBrS0O
zk%Sclf*$|xGXlt{LV`cmxm`Z%$rP}O{Jq8G)R+wk>R&5_n2eQ`l_|No@k5)e7&tg$
zB(66<A^@5A10@}uq)~EFFG~vhJ6N(gf948Q7n;$5zApyRx6-?lS5IB7sHnh}yao3$
zD*?Uh`p+NBO$u;6E75|qT}dFy&)?4e>&IQT<?&g_tC{_A5#9uivh0-HT!}^Ny1_Ii
z_H2zTJo<PIJL_0th;es@f!25(#7>-c_;o^$du>C*Ss6$oms3Ez%@#)CLpE#bc_9N}
zYI{2k>`fUYF)!~RW^m+FPmG70sZ=CkFue~>jcYuy${R?KvG*p02c)F3k_D2OmWDNW
zT<{eMF$L2szW_a01sD&4Uk$m>!1!z*aE{i+-PZ{qysaqk`_>O4I&Byt6A}&@A==<j
z`k1Ria@7^7A7IZp31LF28XI%GLAr2rbK88cWBhvP=g&l+$`>Nuhxv`QwZABSe!|5M
ziotixMoI?Z6WohzZ`(+9S>s&G{NLLv^k5thQG`m>c|w8~RtlnsP&Q{$plkfn1+<d^
z1%7g{^4{2B>~<aoLQ>!Fox}nI6KKrC6M28d<g0Clg)f9zH0UEx6`Wb95kqFk@F6jp
zEU#K9h>5}o7}W;UXIUDD0XSWUCQ}Xm&!0c<jYSBI%+EXOj&VGGtY~e`!lYylyoI*?
z{yTi{1dk9xPw1r0B&7AbVs<w9DW#q2H%Av?>hQPk)6y7)gwmm{AM}0gtS;A?pM{%N
z<J2&~o)Ix1hXiEw#7=K)Cy0U!r0O#O*D}TwUl>C0n@}2pG4R4UY4o7{M<7HD>luj{
zNZxGh>}J5L92q7I8crT{r|a@zfIpORz`+=mjf^{i4`$`x(Ec;$e$1IfLd3cmB_$WD
zgLW{nc>l^7N;c-|GqQ1V5{R${;YS}IyRAs__z%)BaAPe0->L=|H2PgS5|t#aHkx!v
zcr}P@_2BcjnfVS6cj?nd!YLfSx&=$oM?NVa-&AW&E|jho38OUG#FwrY_IhfJi-87y
q*hK&Dx&{mf15Ntb|KE34_ta>A(zx5)WWh&#K$PTEpH<43hWrn64nkW1

diff --git a/docs/assets/images/readme-light.png b/docs/assets/images/readme-light.png
index bdce8a8feb5d10e31a45604e7438141b7c51750f..f59fe8843787c25f5cd994e7170616561a481df2 100644
GIT binary patch
literal 38943
zcmdpd1CwN3ux;Dsv~5jK+n%<&+L*R&+qP}nn)bA9+qPeQH{N|e;_Zm4I;Y}9RMgJg
znJaUx3|Ej7M})(L0|5a+l#~!v0s#R_0A2^cKmp$$`=}RyAFy^38jc_!NCW@9Ku2uL
zjDTO_IEkq{DchPjx#~L@gSfi7GMd|1IvVQR88g~Cm}Xw_;(~w>gGh=BskmibWV?C+
z4nIe>r*T}3uj2$)q){{q{J<3g&^gM~n-?uA7h7ul03<PRoNE@<l}oC=i>ggF!F~ax
zd0~FPjL=p8>2kEcXB|I}$O@y`{6K|)J^u7~y*tV}&N}u)e$IL4+J1zgp}_ioD>QO$
z2}=EcSCc$FiMjs&U9+_Rzuh#ou>bAf6BUIaDN1U%`i3C^Df8WQ)7o@dRc+DypFL;_
zs`JUEr;i~lkV9PZoen~oNMjbkzsZA<$>s0fja-v(($v$1hmSNIE9yMr-1c~*wbfdm
zos~FScqCs8TQzT0!(k)2tbJ+gEHLmi)>?O_Ru`V@--A)UHiq~&Icbd&eZEv}>87Kk
z<nYhn7X?M}t~B?bzkVUh(L)q}yyF>wKaHDsrFrD^c<P;L4l&a0;#ms6zLv9}9TY`q
zN;0WT_jDxSOuet%@!;kysyD)=<pM1g`edJ@eZ?N3YW>8A9(Sa@SZYf{x32BoWNpKc
z64zA62_}jb``BNv%i6JN1G;nnI#^lTld;n7cpOz$R)(RoydQ)jH_r;7DK4ju;P`={
z>dBikS5Qz8U_M`xjD?Mz3Bt_GjP?{h?xC;2M0?)9x0Jfp-@g}T*O#5Uru>H3F6lhH
zSzFd3<9k@si{bM!e(3Ml@Dj;%bZKGv9>bBq?M(0o?l5Kko!hsi_-}7v73~AD@Q)HB
z5o7(Yr(whWP8ZoSC5+mQp3T&i@^x#D?A)6TzZ#??`Wl0c9cr`!fJ<tKdj06RM#t8D
z!-EAf8+KBCw&zoPh|=y>XZyiQ&%F2l>~bkdUb&kUJ-+9|2}zFoQMzt@eSM*`jsRf3
zDKQwu+yCXw5IF#1LuqU>U%@QeJcDj*d^~m2_xtmFxk|&36)PpBtf{V!QwgE>k)r$i
z?|sdzb~ti*N2BTIQ-8ZF^asT`Uvo1Xj&v$xv+zdr$wKq{1MGkyA^yx{jm;MNpGPdB
z&3Wg7EI6Ck(IY<p#aFTX5$oTRR1}Je`EI>!RD8PL3UDs3;B&b@3bPjYHh&vLMW84E
zEYW_Ei+FG;K=6Jm4!zysa@CZWX`u=#m0qz-aSyR|=h3Tl+RK~C822q<THp3C01v2r
zTGN3yJq-=2O{<o+_v2!VmTk9l)%amX8TA+hiRe;%i;$EC23+vb`{x%j|L4Pemqw$B
zyMcj$k)eDs1vSvikg!P{k}Y%IR*hR$l5jF6+3IfjSri6xTuc6hd=O3N3Vau|G}Kp1
zNS<P&&wzbaCTdq6t~L7)zoW5q>N&29h1)(%%xH8C55(yr?G*wd=tKtFR1ypU2<TW<
zRr<5f%_fS)w*~<HFM@&4C!f=d`rF=<lp2oo%6{v*M$RxA{ZVdYMWkTV&#bN+MR(o0
zmZONhy}b;fKo8q2dvA+ef*~P7pX!&{i2e`g*i*>z0ki}xtk6(1?Vt1Nf<4b$o}csc
z^R`NWM+*{Bi%8Qr%9n_I7mgPxVg<+$<Tzs1nZmvtm?=4wMM&1;KC21@sKD#C&8yyA
z=Oqm~0;JH3*}_m*Sx5?1!s6gyF?bk~!Gl%HVV*s1fr`~ad9e$kpSM?EAqiqEKX-E=
z-Q7cz-j@}_CMRU3q@1epw?<ZtXV2DbU98sGHKoi!21{KpneWJ#v3t9S>+i1;h&!c4
zthV+`Ei2?wRS0|)8W^{ycXE&nwBeY1mFgWF6)zzc7v!%pO1^2=%5_({G}I_#qWbjF
zUjh>j@wx^(FgA1(Wjn#ERt}8&Z_GlTo$qcN&h{49y=)<bqb_GhU8FRe!^Sk-UBvoY
znAZ2!)GvKwe$F37jw67J1A~z!M&DaH`oXic^~mk-ZtPzi9BRNIRwyKxl9WU(okEX;
zj)&*jiekVU-@mx1m3Jde{e?TGCUgh|vxvVp2q$VR)IS=>i$C{U<V<2lJ6e3`yu(M?
zz+g|-a;Zu~Q%mc8p-3{3pOKM~U0O<t{0|r~GH%|<0mCP&5DW?Br~P8txgNE*sOxYv
zpTcl-tsUSc*nCRh_Yna{S%~$+i~g+66#hxUl4*|W?kO`XN%zR9dH9*wF;X<we!zTz
zj_0_D)g~Ts@6WLLnmV`Y)JQpHoP_~@CgN9z=ukn0J@~ZdjPhJ8Q9Z(+gAt%Qm-`(h
zr=v{XN<yg^(fZxlQM{w!6);j2!bz$)MHFUkpOop6xUJerX_`8gUss#Oi)oZuEaVZ`
z0*sb>UH_RBO3UFf=MpdGskxpvK|C#sR}rG=BAWZl-AlEltG7CubIR;rpw)$cmVocq
z>Fw=pEF8&{#V^eag*!H@TyP!%h-dbgG8P4A;o%>W%XE3?#$cX%J|Z&n+5q5I&HRfe
zyH`?Dns~#P8s}N%Zm4M|7m4pv4k~yMiwh=4C0U;aFT_w1hW+OiS4klExsF)db!}tW
zNWYdM>HFbQ$5QWEa~0JIlaFuYk2bm$>m!?IAbuMFc|Xtd23~%$xlX8u-qB)k{iFtJ
zFy~qPB%!RDy)>Q9{N0JmO>xSmmW7JG*`0{!_ebJxUrs=J*;6tGAStU5ZeQUfxj3q&
zDDqTimawk^QWVrqXW{lP{XCx>ynY=nb(-RGT6QAO#LZ`7XD=4^$n%Mzh6^E*aLw+i
z7_`&1tN8G*j@ef1a;%{u&Q*}({bbZ|4+aMffZFvTIkZ)r`Ds|;YP$&$H&eden2pw~
zhmi!yR8GH9zf-1LiS(txd897fus^c`uc7DPZN1KJJe9?TqoJX}0sv%-?l0zh71zBI
z)!ky2Z8=#*sSw0D!#?yP&P^A!a|lUswbs?$)I>7%xJi2>0fQNGOd6}?WyNAvHH<wM
zHF{CfCTMf@CAR?#uLKtCL9cdx+4EA*qw3BMw1yJ~qvy?i4l@pwY->AOtx#DgMmj5R
z@rC>I?;;!z%dd~u#MqW-Ygu8BR&lEihhCwdl}^hwe8MnnB;D)K9N9#^eSfcz46@*o
z*`VlJUtaVxV;^}wf@IiXB%{Vje(e{ic!7qG*gDL=-ja(}T5X}nbxH+pW2Qj0s=60d
zZ&>Hh>^_ZkuWv=jU$2p|>H4~g3LNKEsbs%UCs%ud1E1dsxxKZbf%#GEt9q+Yy#7MT
zr|8?WVf~cD`76Xn$7)c#a!`!MtD~#LVXO7(GziSMmhD2>!B)bjylCuaN>#P@Yv^UZ
zK+5H+gG$8(1BrW3eUQ;eUVB&-vZ3$urfYBA`eiJ7O$X9GjUQX1{{A+Jil-U68Qco%
z%;JqDNk6WF)KBst1ek?S*Rt)|BxQ=gD{Lbj;l>Zj*y&@>i?8?V@A(QqWk5=rRr-}c
zFFm2&jcU*bW>^irrm?rDX`^3gYpAyESnJlWwY49Sg(DPP_bdjTIUmrOEW4EX7Ec3Q
zEiUiEV<{8Etdp78^y)WbvkkXy#QDn4c=A^^zQz(vp?A}efa8&he5NiX-_wNn+2v?F
zw8;(IO{-1xl_SJ7I#xjaYvA1+&GF!E@6aOKy35P^Ace(rHV5xp!>)Ih+Xp2DfCK@Z
z|E;6MMBHDXvx?dmM^HPJF`S!I+Vye`oH$4U{g+1+#@vVJ)Z(ajAc>@E#Z>6{hNpnu
zWp|I$^b+Kz=d&jF`O!kW-uKC2t|Q%j(Sduu-J74y)Alwy^Q<bs@}~9`V0`ZSS}McO
z@?@RKSm5c>aQj-Z*1)vv!B|DR@iiIWqni@P+YfZ|&v|W8oAnl1_t(QzWhg}4hk}ZV
zcUsKA#hU>c^=E7tcTN6$QsQAGW~8tNe`)ATlx!`f049iL(6U>RX5^22aH2$q45Pus
zfF_oVY$?Xo9`E;QD%p%AB6M`JktiJHTpk*(z9FMj$;7ORqju@PI)A^~!Kk3%i%T|=
zzSp4Fk@*)2_?T~VViWc|kvpYbDK<z{_&V5*skNS1o3)OmX^d-M?~iSH=Ga@<!qfR0
zW#+x4i2*1gjY`f#`y4dYzt};i%u44d*mMgkmh!6BZos#t<G>!&Q-85i$Q3qmivptC
zaRp(ZOp8y=#(GR=X=8=Vi|fVVT!{W*>7a{rl$@~p?Ny*$X*xDyB!Qbn{Fq^ojeUZW
zfg2O%p}2rVnW5uTly+Ff6b>;mwXY*xDsD?K0C-9!3n^8N#T5K$5AxpqMUq=Z|Ihch
zRF_9uf>58HWc3OaB#}~93HMMq2aAa4I|QXbdZ6smaVK*-&{Y1Qks53-I}QV~KIUdy
z98M2qtXn}9&u4RP{p7K+F3+im)hbyjWqDy?<EP8TiaV<CCe^`Y-Z>j~E?FECj8JV$
zOK~--0^Hr2wXK&|Kc=#i=<pUlRfl8m{PjX%!j%%Mr&6Unxh&k)J7ix*|0FzC)~x*4
z=@BaHeH$@*qxklf(&`O7MMbqN7bSgyPtQ;9UYhO9?+tmrR6w(}em%0{Hg*GlQn$fS
zge5-G2)u$?_%%}?MQ^}ID_L$|jgSQbxp=V1k~nxM5)QY~V8A-Em`&LOYob3|OBdw|
z3-a$?^3e=xkRX(BASS<N)MzkTtJ5FI0iAF@Uy9Gn9FM|bkt>V1+HFNzmad%tE`zx$
zu&+6XjS4bCA#&+_>Zf_N_4-KU!^E;4D3WYvUq8{?*zi7i^;K{(9hT?Qo6!+0Yk~Qr
zFOC*3jIg{vH)b2A^l580ms3F|fl4)8O)Lc$XAp-XJXq?SStg;OP2?b;O;6$u;kM%#
zkHdOGMWKV~23rOx>_<tWDx4ALgbDXqMBW%Z*&G&SmV?~xa5&a!#lB8dKAlLX3k6Rl
zVjbjVoC$|xuMdio|GlP2Hca?6jUB`XP}rg$ykZD|Oofwsi#}MOf~-IkY+S#TuhML;
zkalr#k&~2^G~4WO{twxFl_4=vcFJ%wDbCMrV+GV(TaVj^=@5Ty?w^HUS%1G0*H%Z@
zb2<$lG0y$k<JLn@_FhI<JUX*4nmVV`M)cFaS&ipDzU^}jLD|fCknna8ZgM~iT}iGk
zoZ%Dc6^pRR_KPGpfcpbGYUnQ*)_MF(8^DD;st#Hz6^3*T4nCLjFJ*{<Ah%I;Py(4O
z_&^&L8~(&~BC6FaoN{vLT=W1TbCSJUe{i^WpkckwN_C?GMj|#?u3nto%e#%!=_DP&
z+-F4*G-udFO8vVk&nnqyg9pr%^ImxQk>~#IuE4haFlGJz{$AuUXY_QwbeUj*d9>pw
z0YO$-01~RXnPG(4r992nDcfN-jNtpLY~XOfnbpD3F*$m6S>_^xnOv<=oalDC+4tJQ
z&#&MgQ_MsAdO}nbg7|ce%Lol0!LiLMdY+Mpbh$7=;#Rr~n&CoZ=(GJ@e<&smv;gTZ
zsu1L7n7Bs4tQ;0SUf*r62te?m1R>P$Qsh4o<0Z4x96sHn$+48@$*Gjj;@?aVu?nd_
zE$t`F6VNrNNQtaa;+6Vcp9d1Op0ML*U4FzNrQ2K6k>a6ShbGSDJAQnzV|r7(xFr>N
zI1W(iEc@QhNi%(7&PYi~4R$!6i`v-KpzNu40AdNeYqQg;;0*a;q(up{$i?oS>)Q<(
zLI><84pz^Hk4!&Y>gn1?I>kjqrRM*teSj-9`1A!piiy8{Sci|y)DjBNukk-6QG?|p
zy<cpQjxTwpImq!2ymO99&D&xOgq}kSSoi|H7{n%*tl&D6$|p0CirugQVYD`moM@EG
zBsNjYG&&Wzz2Ou=8?uT&;keZ--J?D!GjqMId-{?+^AcH5U6ipCvBx)y5{X-A1$>o>
z?E;o}tE{!w7nd``-0!S9Kz=GgrOJAx==@Fg{%CTXfrdu$Y^W$HA%+s@y10iAhCT3E
z3rj6cWWz_cQ5kbfztu0Mjbj-v-3KBM9Si5*eM;60e-l(EW&9b!lIg{Z6K^hmtf#Dz
zP92=U_&iN)xWGsXUQu*1a3=#rKtJ+T3q-Ly8m%F*_muKo&$w5Y&p>Va)`>R~b+nbi
zZzcM-Pu+)72Q^9yi-&Tdu#;TwPCArCEm*CvLidM-k`BD?<rKDocdosJOt!%t&&|^!
zr%hcdM|9@cta}7Kdziroi$SxfZPVuVaam8#sC7LT1VZBFjiYWixCFWatb<CVUue14
z##6hc&ecIHy=boUT5VXFI?WR)>pPgdCl{n{)#{l+ugB{>43+E)G_Ae8U090dyhgdx
zR|Q)*tY%sJ1sl{9->XTCh4Ad_zW#gj>r7Zr3(ufF_9V!r^=zsQR@{?tdkbp&cPLr{
zX_>Ce?J2Xpf84=!u8Q4IdcD`vXn6C=JVl|wN(0qj4!o``=Z47F?L}Q^UiY?RASfbp
zzz#Q)_bV5*$3`<9)RUJi58|Tk!Y;l)d|+-Eon)hu=OrU)-rZGLdToF>A$B|p2mHmW
z7#8?;i=mHX9h$Iu-7VtUOSGx~_zgn{E*IqkF%mcIO}c5*>UQ1z{3b3g-YfNaMnOju
zK=i%zK&%tYPowX@q^p2=l1#s%K|)irN=4Z{^pi=;ZWb*J&XvufUE5cq$!s=+*>IR%
zr^yV}ZxvOPT(H%WDJu$t!+B?6x=<3?Yc>_++uBkj2IBsU)lE>T>xXsFOhCvCf^u1i
zc|}M$HCnj@4aPuH1gm0ETt6!&Nf8E!5`H?=<9W+Xt~pbc_P6p>{#gyPcz;UZE}eFC
z(o2`_^_(H|_N*`r$=-aY0Ln31>GF$F&p?S2&c=LS0H8UF0AEb?tgld7h<v?Bh$x?6
z3A!)RC>&?UYAD3#hxk<jlnX?8sD`92r7SJ;DGd_+Ee0ru(a8P6dU4j;$mn02r!UF$
zO+vZo`RSS`Tuj|sx4JI7^NherQ&>r24mIkW_zeVJIrzyi3a=V!-b>z8`BE~i^)Er<
zA8-h4GKABk=(g0f$O1kw8%*h+)s;n{9~7oNX@Lj1e_vFm%<BaT>>F>P3EhZk$;kE*
ziLv}-zx@_t+4%ZPgtzxule~Wh`PGDjk@QywoRx1v-c$7l>je;k{{hym9-oJ)di<X;
zJXh2;<Ed89Lmb^yI&lFdA2>zHWAmCpAMGf6)iF1JP=6Ltu}HeBs+D_^CM__~4fXV%
zqvA)$`ZTLmL25FPPk&B2O6r7eu^zx=V4$*`4C%F?#_9P!E~;H_=^QtpJ1!?nVvF)%
zZoL)Y>Y!JRKi_wbqtj4Zy61-ie*L3m5rm~Imc#2t+na%lP(+m=M}m+eIbmdN8AXAd
z69mBp>-Br_|NO^9YhZ3q!z`SSjEhYh6XsYQFVuquY6WGbAk#2CP3b#i@iA0P3{f%Q
z4-<TX%X&*flSGMD2U>wr25Udl%(7jpJ!zMZ4n|2G6K2rVF#J~^{b9PHvHGW=W`IfO
z@z?=$XYsn*;e3Q`@As^h2let}5O#vAFKvifp&O*6wBFcjgwWLf@i;)-$sinQ;}5QO
zY9f^`2Hf*BubL_w5xZp<JxIg#IJzs08=P6F)X_^me-YxYZajakytdnJFbk#_4$9Vx
za~pT(ZEWaPjF^#O16u#LEj{J9+JG8|<nWhTkSAK$6xr{n7cC2mvm)#D7SObdZ|U_K
zm~tt`xTPhRCda8Wuf-q;Vgm~~&%Hg*p6#!FJP|$l@cojOhTTfp`fvBgEJw!Ii^gfq
zWZ#TA#}5tR`PjE>*E}ciUngQxudlN?=Vr>!C{Eiq`rh)n++42{HRjxXapYqpH!tK&
zKH9SeC$_LY4Ix~d#(#@3uv2f==%yQ;*Ig<qDil?cADUFKm^6k7Tn}%2YkMTNx4IQD
zAP0ViMMsm6@aO1l+GCd@;f)>D<{Sl5*i6-iYZo;PBG|nqxuslQiKz<F)b@OE6eYRw
z>N@ERm^7loKyW@rIl9@c@kQY|VuAVV<E}X2F*DD4wcJSeeA-Y%9!={@q{KSszDZ!d
z*M_~`-r=Gd0^VF&8XKv@N`zBYvV*sm$80$uN>q{$h?G!oVBdpsc*Iv6?gp)7^~FR{
zX;w%urZSTm8j%5_M@MjwLa236>PTM57^Fxq;8kF)lA|eEy!PhOfhMNorkKI=<V|%+
z5kvViyi5+adqc-39beKQV3>6`7C&)~>8QaxN}GI9HRM&n$U8D@*7?P6_U{pM6#)RM
zp3kRsWiS3*Y8o1Iv$-w8Q?A3kEf}BKB^M`<tc;QvGfT1<j7-;vbRlqGtx2XJl!&=E
zRpbmlmyP(QQVOr&-CX_w6atgt!fMY?cb*9Fs_os<GE&hbk=<INB*S^luVcGm>J!|L
zuNXbDco>qC#YX6X5G0d-G*t66e9@9>+GBP3fj9y-2pT%v^IR-}H5xh+VX@G8YnLYd
zFZL>)GINCsSVL+_C~e)8iBvkWsf@cV^TIF%1==fC`cxg0CWbqHUs+rp7~!z6<m7{i
zSxKT+RtUl{0e!^X9#47SqhzN^Mf$cRSw0qeRJLBalQ|8<V#y>fSkS8ge4|a5?W>Bt
z>P#|141aVbi7g&AguCHeK07ATem0aYl&Yfd3CF`O@a$vcd(qPJ`#~mP9^nAm$YlBy
zLUy#hvEeXJWA&~{W7XlN%4(J7hb%^d{oh>wgfJ?2wZPs?7Wu9$9&u5be$Nl>_A!K?
zKSu8cyr;1Q3*%bWFR@xT93tuX`M)&IdZDFlLo*0uV^88Sbcs62X8B+XtYnF7db-6G
zn)CHDfqmh}kNat=_RTAoO|*Y;8_*Zu)HbuYP_50N;t}D>z3WOE<Gt%^F-rv&Esilm
zygmKbptKuT5!W$8UL#=h81O9^5Jtxs+1LU#JC)6Zcktla_9j+|_Gs#KfnMfsu<a0i
zDAW-F<}F#PCaujMTKSgT9X0xCXti<lzdjGfy>}l-&Fg+c(WuH|fVzYd4E)sO%^g1#
zT_)q<3?jXG|6I95vdZ|wLb5bUoMXaY&7W<JaHfCU8(FcciL%o<$v=nA)`A5gU~Rt#
zlVvCTi%Nc@oVqG59?#zsV>G#OcwA4wJNe-^_t%tsFrTUa%j*K~t+RxFEJjz*4sM>B
zs3IIMpOS~EY#h9)@%VP9l~GziF6b*NBF4r=xZ!EHJL#Usro1k>bT_3b$;(dMR^5=#
zLTHn1*X{5&Jc$Qb-97lqlgYgQX?@&Igk6YB3~8R4LmNtIP8@bm2dvRRG~MgNynL=M
z08pHiloSMyxTYbUr@JgKfhjC!W68sfpyXt0i5^3K?mC{{hvFX?%iGiTQp1&A!O(F%
zn=^6)K&@H{NpiT2z@+1rvCXRJztlVS)^dDK3GjO_xkqnv(Z-*Z3qHil_&V@4gi>E!
zC*K8wR?Ec}|4Cl{=r+@wQmD+#%+Q+LvTkjb5yC{=ZZFK6&E$BitLaB-t`I!RlMPme
zz4m(excduHjM~g%d=;LJt@(t(hy(ce_O&<wFo(uEm5m>1+couFF000`^${4VnLApR
zn6<&q9B*UJ&E%!cOr=0~W%-{0ehS}h@uIS^$$ucfz@y$FKwhxzHjPmI3p*b_VnmC{
zawi=rSMnPbAkPZn{ebH^H8eU2zrFjWO*{dgh(kGzH%tTDmd&K{5aS+UIhOu@l44A~
z8Qz`p1)Cqldwn~vb)9qmv;h%rzIn`+6&GL4bJWI5bC0Ss54Z}KC6bR!fn)zH)X%Ir
z=Qun(e53=y=>MSTLNf&SQAtaRf<!NxqzteLu81N`-rS5Ro<Mh{)7(D6u%4~<i{0_x
zkNh(6&(?LlE-3SmpR#Mt>O3{dv8b@8o4Rr*3;TT&6EfI;mnOBolQ1f4&QePDpVA(t
znv}0>G_4F#e3qWxD3Y6@lmcXz?Kv+#^xq1zJo}2hUIA@!-cR$>H+>_y2}yB(hbOh#
zo`&^)nU77UyB#*SS@xeZu35Ek$CenKZxKf>FYM89*unTG?L223U_-8G;EPK2i}A+T
z2~8pPc17zK4&U8t`3`vdrulCZfq-UI)O*T<l1=f3nj`aX{OPPQU$NR1l=&BTK?|#-
zVOTI$k}#x?ZbW{T)QPC=5^OJ0u+lX0?dJZvL`xEK4x8PvnH{(F=?lCNAQv5<9FvD3
zaktGmsgY2XYO4w+c@^Tk47X}$pPR~R-*b@nvxYI569#Z7Kw*T<<m=31hmP)4S*6pF
z7y9+Xh;V1t(VNzUy&SwQ0p`Lqho|GZaLjv$z80ALPk^N2Uw`mxEa4y`qGkR?)*=Q`
z&X0AUEZb_bJ_-pZHpP&;PG|`KiZ^S`R7M>_;HfX*AKZjeZNWvgxv|xsB^5Zhcl(!j
z=&-quGKuOK)xWm;6|aN~A3NkF9(ZuTm=4}xl`HUg{*Zay#k#u-@?;7Eq@r@{r*_I%
z;OT_Dnx~ZU_f=@$ooq5dSv2eIhkqkp-V=YP>p_y@uR--Ks_9HsTrbC^;OYWqE^A0w
z3ehV{W;;!h*=^SP!RN1b`^=vebag?4?(xsZvDB>Y9JW-m-$j@Q^HG>EuBsvo2H}Fm
zh&u69R~655uqkzXXv%45y59EmXz1w=e6M}OlJY8OvsqZm%Bl=LxD4^E)KoA#H$(K(
zfMuEbb9(56G39cG>^o8Eprr(jrC`;ookp{*0y4CS2PY9om^59s&r@C*)%SKr?$!Tl
z0i3knj72<o9QNnOuWV1bUEN6Bx4oD(|Bj_(cuKDnT4vp<v}`PTZ7)EQGMeVNK?l3~
z$)O_Y(eB{Z3T*(5#!IJV0vEl=<!&&);J!EgJc-?nY_eSi>(EY#6Ja#(!vi@=?Zxcu
z|1ZMn;^A~|Lw9l-#mk}2UCs^{?)QdG>&tTQS8~m{N1MUC8kl9qqn>Oms3|l}4c8A1
zSXtqXel-NkXB}kOc}A_eKYe~C!5kyqc^A%~hvgF7GWKCZv%721s@9CvDc|OnJV=-%
zK?+VIt0M-C+gpngG_F!K9E6dtF{}nVf|=!&f#MI2zlX%jgAl?8D1B0Z4x2FhS@^ZZ
zGC;|%ex}<k=IG?^y48{NoYL;Q_<^qzJf@JpFUb0W8HAF>dqY0;!<m~-HDsto9~Xhk
z7$rnMj~Wh$rUpgD=NrXcJ4}M#A3VBsCpxD5`|`+(HFyXyC9vRbof~9O;Y@TOsA<Q<
zWMtY(x}=Y%fCx=EUm?2=GIUeUYoiBW-aLakLii_!&+jt|);Cy0(?+dlHOsDV-3BQq
z%@(hLyx>Z{UK(on1U~w+yTPcd-iS!0TDtOIVa_yB{G*URWgf1qYoE{9OGMyDv)T;`
z;=CA%dxA;WD6`&AINS9o9*FJ+1?P<r1IEel)rfZjfKcE8SX}0sPNgnJKf&g;VtlBk
zS^XPDkBtayYaz@eey*T2l^BhI>70+7n0^%$MVRe&PG$|5)rJ*)DABwq+`{7m$_cdm
zZ&LLy=TWQ{ki^vn#^GDgX!7f{4bKU)HLqu1J<`#hl7fzJ-h2y6<^e*bD<!vQkJACY
z`Qwop@$vSip7&BA0shLUPJbbmTL#GQDzq=NiSC=l14-7zkamiZz2-#fxjG3h-G!Tn
zf;K!suo$%4>U7+#K=)WWbmEl&TK*V;SheL0lT2_87L~EkwbJ#*C1%&UytFOX&+A_F
z+bZvwD+ad9ZA~>CQZ%kDPW+j)3Yd_k^Zq5l%T;YwdtzPpSI`Tk??6nktd(j?vl~B>
zyVF1ZRBZO{<s&m73Q~daoE8<NyL==@T_8ja6(H1NwK<&o{RWoVVzs+As)hm=jGGiZ
ze_CY!)-iG}rik(3Puv$*RD4HF^TMCEC?`2_!d`6Vx@SqmP^qz1+L+xa=fdZ^AM{ca
zLbF;oKK+r*rqus7kypQExyEndlY_J=u2g}5kpP~&@pU$8)t>)4oaJsnAAl@&fN6{c
zWu^U9%G_|52Ne`)`(<ZKIBL|mJuhb{Ls0rNybB_oT&GO}gp@s~SuNg#7L~7ridPs0
zSm#HDZ2YUN=@`>*T_3po1Gf0i<e!GR-UM^_Q@v?y$nkoxv0G3atb>xvGc2`}QY|j!
zW&U~y?d7J@O!td8CSdMHZCO7t%hiLyRaZku+llV|pmYc(Vk03BwFe!t0s>Dj2_b_L
z#7QUjVK0*DYj=w0n%@6r?O?<+(IG!f*<AcJ>oQ0Jx#KdYlJLc4f6W*9Q{IDDz|v9+
zDTJkbd&MxG=yV;==6&AgUHt6sz7hJjr?$)ONvPiP#@D~^<RY^28Nltq07sDpNbOX$
zovVDSp6z^+<um~r(dX%1261rwbw1QCu9)m14Aj8$oit;QpE)w$M5-QxvEa9l1P$yW
zKm#wN??QlI#>t5cmZU#Nnnd(qER36X4npHNzKTch=MM-7rtUkYEuZ^b&g>~l%uRM3
z+Z%o`2r$qf6o?O@qC~9VA|0~lHT&Bsht@u90hKM?Bh{u!PcUr({v05KZncJ)c0V{Z
z7cG4A2y8_N#Ny#^?yo-6nXn4ThdxPUa5MFR=CD#^YpvXa#fuj07^X&Fh@i#GBJD3_
z@y;6DuX`Rx#49zRg(V@IXC#cWOh<n-Ioak&uwn8;bE}Yk#FuR0y>@PYUU`v-a)W5;
z1_~DXO6de2q*PuZ4e~@9v>n&dSXH9OgY?Dy>1$=QabxWkdiliAb>qm{@Fi#CN@XH;
zyB;^Q$w-$2Skv`RMZkpK<K6E&{$bxp%OVqx=~}yfj!Z~Mc-l=)x+`j7&o9iEK<ZG@
z^fcadxpnirky^uNHU#2z=ht|WUcILAkAGtXJhUmcH#J>GDpd|10>?=a-Rq_*7&hPI
zTMaudFyhXPIGE^(xTTxFZ0hJdB$%KLp*uAvOyQ(pC1mEwQ4p}-s0K%v9R}@c+K!;6
z_#Ii5#<=^S`xH>jGdM3?Hm)*4?tg@YqoBYweZ<j;a)!`A>=xkm$JgkKK;Pep%)5C#
z<$~aLC4=zy_6j)(x2^n!5u(NrgbB0gdrwa;;hE-L9U6>h=DajGcr6`9kPSof)P^4m
zkT|%<41PskBd|D&UWM&uzlQZoip<O{1qMy3k$*~{wXK|jc9unTjVN7<{ixeAL}Xg$
znjS~~(_$v~(ZKu`QE&tKs{tkHrgIhjU_;`)#G4*XrB66dr1_9DL4v0I`S;tW4=N@m
zpfw>VD4L0lAPWl}3lKV*gBDG-En_wds;X3e_j1m#Xhwi**}Cd}SeR1zuNC`*mxU_-
z5xwYlV9A9ibh9Uf=Mnpdb?bU9Vz6-kr3*ZaCxw>4Ds#T27T!!B47JD2c?<iV8S|_E
zwGX}H_`V0jZM~r*n~L{@Vy;BSOQesWR13W>Z|{)Z^=-^1r5A{Iqg9~+kE<M5n0>j6
z-$y^lw$+V;rT~@R$IgH9n0wWQ-%&79f8+F-)y{8s?k~>`eRp41AI{e^r_yos<4=(l
zbS_P7^-<k(84eo7^c+4X-vi=H%nNAq2zRbDnBHV)ZeIr+tv8y&5%JiWZ~O0Ia7g&G
zTS0V|nGpiBJ6#$}^Zcw|-@%|taBOG_&v#u!7`OymFA-Hft}89R*m@JX5A(~Xy%k(G
zyRI#0i-TUMF-elm{6KZAu*rwZR-FB@emz;HFXji6`_<jTFz_UF9#8BC9N`uBMT$z%
z#ch0`OS$Xxp&;6;kOZY=P!WQD#ipe4c^SC0c(KpLnCpc@%nxHL8My0e3c}190a8qJ
z+$mK7i#HjE#@ix88N-1j?ZuD{NE1}!1uziEAN4OD)rkdg&46QQH(&b)A+l!2-k5$O
zv%?W(#X%psEi3oxSsUYjp+r6_-90BzoLin!zb^(XW-U^zy!n{!4B!ZY*t8Z$0`d*j
z`WT9NtFGne(oG*1trXw6$YNOg#t43F!3nmkw+H|+$+}-nOdm)FDTe-8>9F&e`x%h_
zt!l6c9J1uoc9r857IdzzKT2)~s9n+`%7(cy3M%*aPhCHAxDkNbv|VX_eOj!Lpuz>;
zK!Z_7?3Aa+?mc>Fe176OTzwIsAOXNnGB_2C@JdM)lI`3!^r*<i5r~GbuU^<f^l$d`
zzC~44O}+Su`u>Co68~$IpDr+s_5LdQLk$6@EeAaZK_-lq03j2IL&R9i<$H@OUnA&(
zM3dj1TU*inkZ5Zm{Mm3ZI1)BWL-T3S<>LtR_pgmhK!u0Ny`!@m$?MCHbmWi1;!mX;
zg?sD9snpK!i{yz8TiwCEOp(Kaa`(Lt9s$g!_oLgCAZX1d_?+Bi1~PWYZVvgP48jy*
zu2-s+xn1*~lBbi|!V!3q{^<6*V^#wEW0=qut-hyj=-L~0iUXqlwv-iByBkAq(7vsP
zTohqj!lms`$+i)bAG3|7)9`rMoA}3k6k2-B#;D=Gqx&C3AEgeH514pTm9cuYMP%*A
zZHBW66KW59&ZpD41dpRJXt#%K*e1nh8YMxt9B&e_YTFk%C0~vYg4oyeHAex;h3?)$
z6arTdjg{mtwf`wBQhuAou1eTR?ug?UB_kB9LXePku@)q~d&y*!u&Wrui}hVUqZ1HJ
zCa20C9OddwZ}xU)&x<2Ko0$<Zg0M;5^qd}QvCI(DwvM5|%L;TBp3!!K3xPQoJWkA#
zweQVE+H+?Q4iEm!^A-5;2MIBssWf%1>AFCZ<8f9G)Oj>elS(|`OfLTQx3nad?;R`a
zFGNv5Z^fT??aeZ82T{Fa>r$t14XU^r<X-#l>L%_%NHT*OK(>Fea99rTU(B;jhSb<g
z-=A;p0*@9BmQ@axOKh8M4rlG_Z}S@vRNi#qo0lxEg2fsNSYO8qde9o)zj?h=Qz@-t
zYtH}Gc7HhpzrkhD<3{XB2J4IL)z?p*a@Ok%k3(gl*`J5mMlSschU%P|E@*=P_LM4x
zUd}K!-=mC(@h<t81+yY~HgbLR)pct<hg`afYuC}pB+34<B^M2B4hZn3QR=9PMGh7o
zX~dq}Gb3KTS`4slZN?c1Sma3>iCOMRsF-xNdK16WLHrXWoX-U8ABkrKMcJg149Uo1
z(raDsexS~7c#AoF<Xy!iVtw!QVJCInCud&l&PL&(MJohB<*gRg+9zD#;uz^4qx{l3
zj$pQzpt0k`H`6=_bp|NjdXO!TgCR~1i9l_534xXB5{P__1~AD8ET(m(UQpdDQE|v1
zl~S)-v8xmthhrf?*>hzxT!H&fA5-ut?muNFaJoCNUW;BgGKGms>_XnGU$*F`yV!4X
z)tjS>5BdB+NZ-r7rv=i2L&2qD$RD4Wl0X$+@C_|}DTUM7meFnsy08R)JVRtLdFv2p
zfPkz8C6Z|_Q!AVsW&g?6_OSi6{LfOnpBEtSrV|;&*I`X$ra3?I|F$|Gn}wr66bAQG
z|Iw+ewmN1-whR>}$v4+*`_3o@p<QMpM6i$VjT2P&!c+|!Dl`um0cte{Tp|9r^?uqC
zvg^$CYJp@DV!6!9CGu-j>qT3uZ+p9Qol25)6kUC5EpxGnY6=%#l8{-Iju5r*pak_h
zl$pPphzB&5qcNl{-?WyPlQFgJfK}VekM47HF~mqS&mhH@?Bw?^A1xAknY|yIB6~d>
zN&+Ie?yN&l+9L{9r@7beB?e-9jrf$tY_S=&-PFa3N8slaEQWLT&Jv$>5YO*Efvu5v
z{L%)KPGn`_;PpRuY3e<?*L}w5qN`$1eh-`*J-_9i^O7Kh6BMYsxTAEiSFBsGl%s+`
zIyVX3sT2S@(fU7tBH&R?<Xrk%{dpt-4dJ2S09EizSLl@{DsAHZ`}bL=?xE)H!omVg
zjOG$^(qi(l!@P>6ox)m6^|lVIW|!Ps9~z-F8xV#frw9jXgF`H$(r4ES>>Mn|_j2#-
zF8<zxt#tMvKV2Mllh4+u`M1u3UkS~WaxF-kK)|l(lJf-7M)+gKE7#9w)36=<#c}@<
z6O=G)zFj-4B>Y|Y=EnP8bn61v;SiNY0Zs5GQe!nf^<r!kT^T0y1bJb~Ob2Gl*TSo<
z-Gk8}jw&`Yo;QT}HEeY)rVQD#7uXmt7hb5w)NcBL8%!0}5S;i&v!KD`+d2g2Nr`gM
z?y8*d>()N*(s|c3lhS>6i&XCPu(aW~eqRK&(q5FvllJeuv3eUBSdsUa9~<kj=5P)<
z@+P}u-Z=|;y4s7ZcYsS+_<s`DRBNd21?p57>}S6KFUI@b%s^@EwI5T@ivi#>WS?_Q
z!#|nobw0kUbKWjnq0;oTE^{Qouw80db3#YpNc(FiO}}FD5xwFRDOzw9Yr+q2&0bv#
z5EXF18STGG;Puj+8AtMJUz+=+;W9;l8%#q1G&}jQ8RsQj2C3DpxdxwXXo))o=$VgU
zC@s5Ha6mz~PYU_sue{d4>h=54?fOJKUw1sYQusJdd>5d;_hhvjPd7J}>g$2NxY+mw
zTwkb+ns#W2!p(y3<FZ&y0i)0Ly!)e|Kb(LkrX%yo_CpcF+w>;p^ymWKV5P|&-ljD(
z_j6B;x<Z&3We_G9C4!idJ%)hCmHE89x6b6!gLtD!d!Ch7m+9~jQ&e1Bzc56n2c_i^
zsxT;8re6MJi4?U2P3mIdQdoqxSV?M%5MwBkDB1Cn%U;SyKW?3p@~*eG;&gshu6>=D
zPiViQ(ctxR`w5G!=ZGpQE{p$nNQpPCk4c4>bX&L=deBaNIsD)^E)Xx?lAeyHGu1$(
zkm7Bskav*2ocK=<mB(kKw+2{)G3KT-^xF22xyH}%g64o-@y>XVqHV>(*<C(F8i~O8
za?g=xrd~~=f2;n%wL_k{BU6SJmKp_0Y;kMUr*rs1m3m2~M>%AnNAM34elQgM4(%tM
zERoFVDGb{F;!4gI!)?{k9i{2+Cmeh+OQahSwfc5fA8eDc=_sq%wFM;eKHUFeziO&)
zmDxXvtZXxLV(D8?*krJV4yz9)sy>G_IP?(8!?oH?8aOyO$oJ{>b@qOIx{cI;iBH9D
z<}_Mb4|1sVI<E#_gmU66k<n`RnLHOcc9J?+1L~v*)ikH=M+|bj$4SRr=*UGvu<FhO
z`uGTdJuf%k^2~q+l-?fBssWBmI0a4Xn;B;stF_A!?1WSu)?ZgCK*6c*Gj0-uO1ilS
zv0q~F)F_zC4@#uWomN-*!|LyMXn#S!yNC$8MEiNsdDgA(8}y!UeTG8uoMGo>ng0F9
z0$R8PyGml6Wq0CdW+iR7Ib{1uuzuP&5|v-U5hZXY{I8}t@Av5BR1|0k!7vbD60qeo
zqZbVHdjqljMynj#4&UCc`P6WPvCuRWz&Qa<0wf>x#uRafYcg{;R`4LXWfL6tMTsKe
zkAFTj%N5-wLt~TtzjF=>eARe-y=$|83dF>eyjh<4sgdorSD(z>cS;HpP+34m(6ZbQ
z64;xUR|mip7&Aelc;<EDPXEE=hPM6D!|D4507zJ1a=GuN{uRiKzQ#ZR3b;%SHujJI
zSy$JW+UPv>1{<_LpT;}%9yFUfig76GBmMmFJN66BN&9QaKrtW?&tT~f?Qj&?OX7~3
zhHtJdE)3HLP}33f_9+_JT{#YC)Y^2idk_jL8bIoUp=p`&p4zQyBDS5N{^Q+~t?!MD
z!p$Er_&fb$|4VqR67cpLE|cd%Fet2OA4JYt43(2DytD>k6G7-uQ}T=&^iS(L^b&IQ
zIhA`>2l<fdP8vPMvd}6nTzRUM@A>Ck3wsfRdb;hGy>QkqN#)qsZ3!{QXug-S+?ZNB
zHYF)nW8*^(Xd-W~f|K7)$4PbD$6dR#6K>UMk#ey@mFMk=XtfFWrFWN<OqH!KZU-R?
zO+%KvLBdDpFvYgXN$E6`zg)U}5`nT#2$3P)hxbSr2#P)h&W-8{6okIHM}8k)chcO+
zZ;o;7Zb^@>2+@K{Fj-6pM9C@<eF<NxYo@BL=_`*9CA3Qwx04Qx7W$mg-y-ZAMu^IL
zx03vC!;|332^w7{Y_T%$INdDPK&3bd%LTZCsGCmo`Bb#&A{J2~>4Y+gEX~JWpsXjD
z0I!4LS>7RDch`eGq#?9;{Id?jCQoqfZkn>T%%>-FtE`M5pdS(a-L;vG#)Ox#hkPVK
z<M?hQi+2i?)SFp+G3J1|f2-4b#Aw|2@E}~}k4xfgKut(r#rsiYrfF>I<d`3F6bgO9
z&HL)AgLBcRGIp4rRA26{-*{ItY~O*#{-J3j^6s(MlA8vIwzqALn<VV>@&qS-<~q7F
z;b=%N0I=D*<Z&1)3r%FDb9tbC1KIBy>0ywCvdLOIXP!DDBf6A`W*djL9WX#W49g~0
zX8sUWU>WBW7hsZ#$QwO3Mdur|pa?=8F-M012b(*+JR|S$_Ha5<n08S&*%-?XwiF_B
zKc{2s-PBxdZBA$7X7|`}{mu4pIbl>|W2r^Cm2oabXg^(6$QO(AfI__vb2ABKv@~hu
z_`aA$RXqYi^rfU|5j~98eypv}DmZ(OmZ>{@>54Wf@20?`-$2O#=SEiPg61IWk`3bK
zCbU)tD(?<_rR+H|FevC3E|l`XGEg9b2KV9P<Es*L@*1ARiJFaR`Fkbl;n?QL|Af5r
zFn%_Ku*hgdV$yzMW<MXRpe+pakrc||lm1m9CZ_;^2C}Vxx_u;k5Pe>3>_96%tJDS&
z(2$VBM{8=6{8P<zIe-BLq<tNzJf=k8I_~eW<p=?15(?BpqOtjDj{{I=Pb&Um$8lN(
z#8YueMgUfI(2Arh<Seq5if>=#I7wiYpo-g+u&batShwl^oGrK;EKa~SNcKNgS-px*
z@$N$9RBFA|`FPPMv;rAm1O2-dp|Mm8)DL+@x2IdjgVuPc19}W^*t=hLN1!6B31S_?
zi?P{ck(z{yq}#;m!53}94ekS0Vv7ClHU6|WY!;!VFm(h)FnY=)bZlAZIKTP{yUVZ~
z5MnvFDTJ$;=!2XY=zWlvoHn;ET6VVPZj=<0s@|5JKgo+xgz;^w^<{=WRGE`I)@z4I
zNaMZo;17?0Ldlb$2T^ttYyQ}xVHU~dwHE5dS?j&>2%A{0mw&k-y>0|^FklA+m+di>
zXpQVB34_#dbnNbm$#vp(1B?Qn9V_Xb6(OUB1~~2sk#`n>04*<>AB3Cc5=<kN*9^|e
zO^Zdojh(}g5rMFT&?K?s5}e_4OTe7$PZTue_)SeVgmFO4YzAFX&HI<iqHYF<{^8Se
zSORy6yGlux|9eaxS)~SG@N^ma85?<XuS$!UfX+H%C?KD$v(NT=|2@WeZ-gzExwBQ7
z;BGBMLH5SIs)2t{RO15W)Og?bi-YVv^i#7huFArz2fe_elS^Jc?l7d!T`-6iE-qjh
zNX&oorZiIi!w&jGU&qm}Zz>hsZV|>pp-oT0C1FD1F4J%QBg$ztESF7Jmq=1{>pos0
zhg<$5nfToA6W8a8CB`qP6Bwb}mcssJ$q(8=`UpDg?h>x+merEiFg3vj3k$=b>PBvK
zV&gu}^_LK6#=46MYOAbdF6)XK@hQgCZPwR_XtkHN$5m}^-YCJu17~%dHnus071wC<
zx~ebD#FGL)oPG2!lkv7+c7m|uL6l2swe_@z^u;n_pq|p2>PPn6|Ec|`$!9?*+^nu;
z2tompn%6QX3)b~0V|s97{iw;-o`DNYuWH(;20&{eW%idPYJ0WLN_#a4>x+fm%SThL
zl{a+%yb<w_!rxT{<q0qoMMs6*_03ien(7vHEJSl&;B&I4)-K`X(&19Lv>wO!qo9JS
z`LL)d?;IsfLI)TjO1fY~)K5#G+23>bO~-@4PzyOpC}=dQP7V5jNwXX3wcQRBgWaQZ
zJ7=P&t~EZxp2Q>EAG|tzF7VptCryZ7sKHu`+}r~1wkE7ri>#bYT0VXqgU_mQy2<p{
znjbA9TBrbvF99%qZVVn^21Jm#Qoa1pltP*bgz_s^+8tiqKRjB6##zbgOo1)EqQp+_
z<N_rk4no;wgX^BeI-ZelleV{WY7VE8_mIYwzD%R5?V9@<qpQVVhWz2F`|)@Fb1jk|
zy2ip%E5_DHf%K~q=JjujuW#4>&b~bT(v<?CbR<Wu*|@f(SYx*5jiLa-|8QB6z_qTU
zDIqoHS9W0ZZTKuJ6@!G6Q0x0(=imxG@WzQEI1`5f_<^DNLx!6t1(-pABVIu$qC60k
zH)ExT=hBvkMLIw5*4{qxtzD$RX$CXDb<{t%liVI(l|8su+Qz5r_~o{=)rSrm4RKZ`
zr?whUD3VF0a52Ps>G34%X|<lA+P*ek27$9s;kiCqiH}=qY<f8}y#*|dhpEE9P{R&_
zT%<-nvUuzD_HQ)Mi#<(Rx|EksD0d~vLmwe>J0&<Bjv~e#?*>EDE`;r{q0!<4lxSXF
zFJQBnJ-ar%I2~bwfCQAX6Ni?+Ctm2GJO(4Ml#G7Up8o@w&pv~7<wA;CTd<}H07MTg
ztpPb;<hkz87eT38$`Na3g>_P5@*P95Ult>oM0&R2|NhCi+YM#xNo@f`O5wmux#OOS
z)<YaHXY4XisW*e~L^fM4xAg2<<0)!)wUo5+a_((XTqO*0*J=Q&g-jY^D&Sl&p*4Ev
zr}GG1o|2$mkHndPfoHXDULL4V`DUy#(ebg%==}431ZIu_$^BwYriTj?X!~>}<#u?T
ztNN9el2`B0h%ENkug2x8vn)f|gL9>8WI81P1neXU!zV?vr*C|86CWr;Ivy})>#e5V
zc)EZ+kFPR4a(B?3;<YCXh&9t)L3{eso6Mp&v7tK}XsffhaICZZcD%Kr9ly_&z57=k
zb1FN9ioHU?3irheiak@VA%8}M4JT>`frbmbcnaQQCoCA-t)nC%eOiaY5~(<^q|b>2
zm?9WEkiTFMcuznH(yFTHliwaEqD_W=hN7KWLq6gJidX8i(hLw~7drR=SSJ~*Tqe1l
z&kjJ$npF1e-Iv*-8rm~b>hK1A+GnVKhWNEp?$s06`aCozDLDu`hx2+ZT>!l|oTYZA
znd~Z=_oS7BqWoGy3@DKc%&a|uj-xE0S<_MP%$W=LAVO#9{Jd)FiSf;4o*7!#Q+3=b
zpM3#OxC|DOIumY+?bR4NV6K;d&8@DWU=T8%eP)wa#W?Z>0nU28CwQ4_Z;3m}xMPSm
zD2LF}M&2(=wNL&x96i?CqxRK4Ksncf4aHd;8px%ma)0@oNv^ZMW{)_81{J1f^Q2h0
zAcTbC`+Y21t;2M^(GDzxlZ@FiK%1JpP_$@6JY4WswI{yb_I)`g&D`yLskGw~M7!4#
z1eY}2mr;KN@dc|&gfm`e@t<bW*IW!<a((R}1rKxx73s9{r8#5O7O$Q5N~1yTDT#@6
zvW54f2+OO?5o9#}9SH3BM^^g3q{!CgeYs=TUkOrA*E}rukD@ME1noV;JaEp0zMtQ#
zJl8kW6`#?*Qxi`QThy?C2*kdK0K|(=1CiQGU0)A|W!?6OfJa&bv37P*1r<RBjQ9f9
z;%7OyxUI%fll6($PiFH@Nf?U!^k?@c|7<?Tsr1RI^k=W}_h^%Uy*xK3p7zyCwjf3`
z9mK|e!KbgEiQM$GG5=Q!fO$TAMI&7y^txj=y$|D$d?2GU3vK-HaT_rq$&yM~j^{^)
z$Puev-;<{;$A@p2DYA!btRcF2{fyuI77tR)RlhBxwKov~qj;KNwYv-eU;tiLbxz1g
zRgCvG9J+iYb5-yfQ@AuTuNc4Fi_fnTHYPeOoWZzkBY9om4UHubV-GlJ5C_!lZI((t
zfbn|5>l%6AG+ENLdFpQu66FL<Y2HHYGSevCZKB?p>1m0@6c<U*_EN_R;KH}CWx8<B
z-;VlbOrc%1-jOsxY?Ae~wxY^&5Bqkj*EB4MJw3>X&90RrYH&H^U#V_~Cf4<Nc`PGJ
zz5EAhjt1%!VgAo>7cdRz)^&#BD*ZSqlyb{L@zXPFRViAPo^+yN=V@{l126h?cW60k
zSCdTlr$Y5FcG7g)<50yj6^+L3=meht*boCaH$5HTv?#Sd;{hk}HE<h$7HIa12Jx~v
z-c`{xQch{o=kkblmK(yH|9bd805w6%z88J|M)|O93L%6LCi~zvA4uwW^49|F`t7&h
z_A7MiNQ0CseA}~qkIh|OCC@!E1e-;$d6$QbTrVABO>Et5luZ0`ZW&V-If)KeOB`k{
ziB9QSr$(V^+U!m8;FmZrlFzg%c1S<LH=G~vAnT4xjAUz_z(<k<iloj@x|_wi7Z(!O
zK|ZuLN*_{TcJ2G<=&lh0ePn#3>G=FoU)AVIg{TTx%&;TJM^dzjAO5JM#Hn$|MUfPg
z`T22K<@e<|8FOm;I5us!2Z)!>mW3lnib1Y3c?QXUX#D-Cz`a%*|9E1y_(}-Iy*h-Y
zvf1jJEkL46!Ez)s{4nLsp>jvdC)H!vH$v?a!4TjdDqIej+Qm)`9yD@feQEWTD_0u#
zEm$#y1U7D_j_JX|W|dUcc!V__W-Sfj=RZO?`y>~`&U12VFE>gH+c-^!Wt;3wnHR>Q
zwRZHNUzyF)MtmT6AX8|-uWI-1-A9eaT-KLAl&0Qv*^2_+<*y|LR0?u}s+!*2@x}MX
zy3%Cv)pqO)Zh3C)hP9uc{9Q_{7}j#@*1_ffoD`#&t!*A3>2v1bugXW)^`qJvS&OpV
z^l#!*x+M5D`T8{j!)=d`&&OgCbWgG3@k##JT)hzRoqfFHyuV)<*}cT6bsssgFa)>|
zpuAxBj_9PcO(jzA9WD0X);2+NQ5i1TQ|>42gub{2oWl6u7ooa6SZy}U;bE8@c6^#f
z-kxkicjr+S6-LLN-BA^Vs<Lu?&Pp^1(rTtl&mLTK!MQy2_%nR1Rdalr6<dbREy{sK
zJgn1wx~iJ$l7XdF8-3H@$lK04IfPluD*0|<Mb=lpRV=kz#NhKzawLX_sQk3dkW9+R
z^YYb#if^kcY_~V`XtWc_Ak<W~dGX2FVyYDU4>-MvW@ZGpNVw(kUnqPWF<a^^GApdX
zs5}%Kl+@xCP%L3Luw4k&3BfX(n`I448txl}c=8v{KyXu91UE>CInZl8Rw0Do`RAXn
zd%u4DdL5_$!m&mNHukp-8B2-vQ(PV>;H69WZMB2nRy&A~(inI`HT`<HNlW$-9c~yq
zN-9)#W}8{P-NvG|b_&ay*SJiMZ6o(f@ORSSJCjmYFJHd=(5kK<Bx1wz#+Oti-XsNQ
z9qzdt0_+sX4^_YXZ2Od1g)KL4$G&1+rs>7_NbOp)sVQZ?)<<a1{(}_N+~#9>^48V{
zb<b(3?a_jQ?DTiz;!?l+WkuDHxCqlnm!2MKIX>0Ks_oTUs9o81Tt~}o-BT<G0Wsk!
zabXP#Rsa(D-D>IUeZJWLioykvk{gCb;EK79o#`38GU~Zz+e#_9;F7DD_{OXB>3v+Y
z?Fx&Ex%|?LhzJXzn`s_j7nB(p6#I!iQX?Pz&$2x`>owh^=+QCW@Rs%HV4+jIxpP#g
zD%YgDGY9r{BwEZ0Zm-Vgzm&f(b?v?4+^kK({k9XyX7I{Txe+*{CTsg8pnEe-;{*&B
z3d3tm(+U5*h{F44VTrLLHWuOwPSV+5j9egWRc_7`3ReMF0ddE8QdwIdALxK1De$Pb
zZM;>zQoXF(+593@P^vQZw@{}37TRo^me4!3jgl+9K?K#F*zjHHci(;IcQ_n`g@wsJ
zefmftL?dtbk%Sc=uJEPZbj4kuF9D(s_k7v_I<QzE-}U)JX6~Lc<p3Tr?Pv+NJhyzq
z+6gCrmlAVmtdgoN?2y&huCXpHI27ZxC4>;Bi1@T~6%}`BQP!fuLiu@O=Ps`upK6O&
z6sU0N>O<GW+*wrNXWAks4y(fMLLZq0zE2zWqaeRpF1^p5k9EJj7>S`OE4Nqk?46f0
zV&v$S=8MDOAS7h()<u`gh0SWEZ=b$oFP}rGJTT|V3h6}{Y8Lf#Kz=rPUik$>`iF}7
zYpP^Kh$3zq8kzd#-15}Ia*dO^nyIRmXj1>RG&gHgL{irQ3VL5%^Ditf*u4YViDZj-
zb%^{H#CK2p!j(uKM(jl_yaqKlrRPA$963@b?z1>i)3^}9PyqvlviAg?cI+LxuJ2J*
z^%QLJU~+)J#t`h+3?FhU%K)9;r<$+lwrw$r)BCiMVbBvqB*W6`vbc3!->oRhk;@vJ
zLB^yRTe{qM`9F0bCyQfe-E#upXi|Q#`?ynA$>*MHk9}-oY{*haB}D4iw1%#vlzZ1)
zP!EBAq||+Q3N`?&nLBrGL0ga4mJmXSl+-RWE;}vs;%cvC*0R(Ogb@FVOiaxxEceI9
zN1C|e%rI7LuhxHFUiq5er+oU(^wM*rBKK6}WzKHcS8;CErs$-!WqYJ{S_jdxQ8-^}
ztd>5!NTglC>j$gVvQM%anznDo3|-f;+iivoDOur*oi4(q;4;O&PbP8=SSEcQO*ID)
z$&$RxNfC)@=^y=6dXg&jg;FQ-<;nh!@4GU(r?c8vQnHJEAjphLN_*~<Zno!pcd{~L
zsk1sFsq3Kf+?~shl?AnpA?2$>m<~*T>SvLv32qXCtAT#BLc-EBb47wC>5tCo63*fb
zPD97hR?;zlMfm;xbX8T2WSh`bD66r~OTg`i5DWM;D0dq~N>w1#0-^iw+5m&(MTY5=
zIol?mEjnCdK(B)!k`k!$NGd%7O&Zl~B5g)F$lzR>@a}cLL`~Du0jyT*QAZ}zC(ryg
z^`=WZfq0#xJJE29lzglR^^5dL-ycYc)OH*oZXU5>{kjSLrYFbHNME=$amfQyJ0$hH
zyjxq20pKgmM01xRg_wV|XYq$6r}iA!*KzTX{)U%lPm%AQyvS}tvWxw`!g7s-C=)T^
zDj)t>`gUnf##1$?ZELPSUCQx`bkAuws4d4Lexz4%)uj<6Mwv*8F)?|51*&SoW;322
zD^{&x?fMPG#l~PVnOMDMJ&!;90A{m^@`?%yi;76?T$k^ItGb%qyR)gNs9-?<epszm
z+-^5kiv?Xu?z}t-uU|50UOAgL`RFTJPUKuKeUwRkT1C#*M!MJKc^Q8If7HH@OzeEg
ztYwv3{&7i^8K9_KlhGmSxUx!p!0kgL39%`7{nGx;@8!1u?L=~a@$yA76T{X0(i1a8
z7&cCqk^u<%D9}}v#67%JI3v!$^E}As7ofT-uxp`23qOg9QQtca!&TlJBfC1{uwBq!
z1a7x`i`7~~3Zsv&Cz1eFUU0f0yS#CtiOmchA|Pu20-G&n$u;LV`Q*pF|7jIyl+h<>
zIhp+0!IhZQ16=n@T##KRar*<eDvE%V2yg`v$*6h(0KebA`M?G}GLSWS)|k{AFEarz
zgFJf3q{M^5&qz5g#2+$m!;~pUooMv2f^}Jpii_atwv3gG>T<#$p|q5ackrk!izqei
z*R~$3?O`d(FZFw*#Ol{2MdccQzc8|ES(R4ja9W??K??arbV}EO`2DgZKWi~SSYo$$
z6WUFsc^Plj_7<Z0CNI~;t5;x;akP-9myF6Eo2!WkQRv^pjv_!+6&!Y>&d68ad`I_g
zT^Tw0HBLOCFT3|-^UM<u<MDdQ$lS$`KmEdhQ~CjrmtVjgci%@!auUBSTEf<~%Wyc1
zeoQ74x(;72C}($u4_mn;RqWfdX}<13QS^5k9*am$8!DuTD9cTs0uY+eCBdX{Ygt~_
z>s@+uTLGK_F!)4=IQ2ODsUs&8vNOla30-XD6#GGnop9{9cn%9*y+S%M0+j25l%M!T
z<eQzuyMBSk0T+M`)MHnc>Zm|rVd1)nh#FEC(;w2muKzH*tofuHw-0s{K|&aGY7kym
z>4AbusPY)B&u)geP>8HAP;~Rq3cmX-jPlCH&w~(PxP+@Dd@JDNnsv$ne56BZ&9OC2
zOer!<=jI_5jXx-@hN4Qtq9}k6E5wJ^TU8~nyHt|z6f~8=tV`HkM!;vbSO`n2ds|*!
zz8=8i@f?-*aw(-0LcHGj#!D*&jDaIPWJG3ym;mNaGrpXbdAMiO4j4XS)!KCv`b}#~
ztE6fPKQv_#^VR^k6h&EbD8_3`RAg^0icU<s{rN8nNBRBIom=XC?zJh!=Uy`~?B4!8
zjCI-TlSrTHRIK`tQ;rK68J*Jkywa$!<w;igV_c*;9;hRd+1c5y$iyxmF4R3w4^{Rq
zwRWc(LV1=>q12gOUP)4{+0ZVoaVtLa_#;F`Me@U}*<5h$AZ{FH{B#yC`2(NN$Lu-t
zc>dWZ0hlv)KK)KQkx|b-!=KAnAcVkbv7#snilU(Fuw$!-huxw3T#zD5YV4H$KTC6W
z1^k`Et;$JBF{YOyl2V<~HZeab(lk3Y!4wI+?$cyltaaa18?Q_*T3+RmZ@&9|$>VX6
zrntNk-?XyW&<~)UNVek`L__%>kH;hYe!p0-U_nN=Zr!|qg@mBw-F?a2&UYF_taB9X
zDTCY~BIof#Sv5pEAT0(IL1d`T!`GMc+{b~5^?iU>b*N4I`G<u00@8u|B=iE1aE64h
z1dI^`2-rPxLkVGy26L8Dva?uF8YBSV_Cs+MBu5yQ#$6?ng3A3PfN`fogBT%?$5X27
zx_~3UvQ2ZSneY2uZn|{4M0^UQ9PT*<!7E^n6mmQk*X)c>=O0iE!^)3VIy@DgP?Nn{
zi`F|=cDujZ0mp5(BM_^ylo#TSSjmI28a8<^3Ncc-ZTv&$97MnS1gUxP6l_9DxpK~&
zIpr;%apa>UFJm$U)N1-RIw`HPL#+AnzG+rs!c`PiPL2sPeR=DU2+Ou?&&T$v>aQRE
zTXg?zd%TUjEq=+{^L5XIgB5FtwFRqW1Pb9oVUFfQRg6R$pVx34L`6jc;H<1fQIt9Y
z`1|6e48L(06TbY4g2E!=<6=3re?Ojiel*chk=%dpKLHSmz+y3CE_Eq!Aofek>}TDL
z)OjU%tq!r{er#B*^5GAq<5q2VeSP-@VJDVWYTKidQXlCMXC4Fw`5yN}$=1qp^ZDh4
zTb#gG!Tq-r$#xu55JCtkB^x(x6i%m80IXTFW=M2&^n27=pJGn|OE`F&u<Dr<hz~P}
zVTp?XiO>S4VeND7<+)zsGIpU|4Dxu{S(U8bZe{xKM%+$cV4O}sVRbX`vJMLb3~Ml)
z0B)2J35*a1kG<|(CzqU26*y*kF^x!ML*$ejb9M$<5fx#$?5(O{U2Td`CkGbR=-j#U
z-1PMH58is~t!KaZ;*09S!a^}}<Vf+{bI-w1R@XXx(vNex+;rtBQuxOr#V`)P&Sa*P
ze1Tc{H1ms@nfpV~SAMiILRPs3D++@^IR#`lBw;o)Cj3@eA@Y4I$L19y<z}IX<%%wU
z)f|pxr~LDjqefi*NWyJKtmM?Tuu5M2)PQcP$)4KU{RYERxEZN=kWwBqvOl#5Pf0t@
zuJE2;XQ5vapZvU{@~*yV*68$HF9UkoQdNbncB{h4-R(UcRxu(i(Za?|PnYlx9b(MY
zcAsX}t}n|=A77cDo*$LeW!!JN=aC^TSS4$r)Egr4`vVhcq%Nthu77@P+_V{o!$E9p
z3^RV1#a*}GO3xnM`Elx3{Oh5|m@(^TrhWAVAt4UTCKF(&R-5;uv~QNwSg-rmloe!t
z4-lEsWr#^=TG5`Yxn@;4OI6_d!Qq{@?(sa<ue+V)TdF&RJJb$6QmpEpLLZs=KJrWb
zuPY}1EnRway{fcA`?EAJb40*3@AGWh(T*b=wIp)n$dRJ3uu!<&Zn1jx>f2&sV!jhX
z7*;2#L)9LLI=<;nQX&oOU6&Z>lV~KNG{oLWWG#xT0*Kwyx0G=4=~aXQuLY6FT66>M
zl1($2izWO`8l?2`n@a*eEOALnTrICE&LCry)nr^>ml)%klbea<v?qcyy{@#R5JI(D
zt@l0p=%dr0e){Q<TB~GD>%&dN_8aMwzR%C>G3zF&D3=0j55p9Ez-&;i##+@YbMg=W
z$(%4FV;@PRSo!wifvetHI6+ibXDW)A4&ot@=K#q<5Qgz{b0B4efG$FaOOZUU3$akC
zc5hoXcJb&x-dd2_mKS)eVqF%ad|K;QyRj$s9?T($k2C^jbX`Ac>uWa$si`IXy4wf7
z_K$?;hMweDy*=AQW}c6jaFxEP)=0ZmxdXsv5ez;tM0kCY&Ix9#s+qpN@C@6kGkV+J
z6v9v=_oKc#SNBxxYekv5YP?dS2Nk@KNIV{6EiSL9ATPgwRjb$1AvuZffA|T2>gsBG
z_v*<nKTO4Lv+?=F$ta47qN+#<iUQhxY}se%)ga`vwQGu5$cz7db=;h^?%k#*MVqd3
zdnGa9DkpWd0g{wBlRY-VbkD`7hO%I-i*0+n^z3MTb;R)a_wE=Pbs>r<2_B=JNVemc
zK`n`_wMH&nxbVV|kdXHP6(nr^2)^lRyz&<}-HEk9{Jh-^J>m^3Wm7P#&@<j3WcHw}
zx$JsRk*Z2w_(u`rt}P?MbO1&>v*e*+r96K}abuODoFI>s!(z;*=RVBa)?m(-X7UCW
z+-Q7X-*$&>pHORbR8@8GuwlbKj*pL5YORuBkoo<;z{3mqoYXTX&iE-NEM<TY3=@*Y
zw7qpT8_BDR5PdQy|9El6q#4uq)rGzC)n6l4ja~SURbv;g6jJ_%5Vv3$uJ;E*B2z&=
zZ4su8E8kl9_PYNriQn(5_zPqBh*fKg%0;W!;5J)A&u#5Kn?g;ZO{tV}<)IkwP#`R^
zTYPj<>c5<MI|^3s^n7Yk1vj1>VZVNGMD(~HN;sj5jjl-+H7-K61C&>4M2D$t$aIq&
zW9Fo;_VfQXD7<T!L$#BdY5{QOZZE8o@|9V7%eSYFQT|X7#`7q6Kd7oIo*H-eZwnXm
zkK1qMv;n8^{@6EIvu*>GRaJaG@hd+3_%m!)E2j=Ph0xGYq|}YRq{O5OyunGK&0vG{
zu|;ZgOY^d3)gH5%mGEAjtm4j#Bd?5)GKng;PI|7Fv_vx*HDtSI2a7d2OeLq-N0nRW
z;p<{zIwzP}w6Ur{@t8i2Nk~02I=S=k+CJNfWGh2L2C)68eM$DYzW0%YS^~3i<3>?f
zSSUu19vv1H74@MIH35EqJP3>LhRV#ncjOMrin5uYOKj7&wv-46w;ANETWp|aShdf{
z2Cso4;J8c6VN@=hRR!jKSP0A#23EqT94r@CG!Cq;^5|6i>W?bIn1{n+%qyh{ZXqzN
zkinK#kALqIa;C$gJ7M{gb%#tQ)79U6^UV#lR>_egM~b5?TrvPFKK#&^{^bwfWE?l(
z9HEPoLA(NNq4h{*Ht-3^i?LR9%b5JblRLkhxqV+<*el*%(0A3?h3^QPeJ6;IK%PwN
ztl<tJ#QlD=zIN5yi-+xJ_|;z;Ycu6&zs3QkQZ`nV?tk}3W>m|dgPm7P&^;CSO45;1
zE}TFAz+!NZM3_}y_p|zjyqnmm%hw*S{^#l)#-*KiQV5ZuDxKoZB*mD>F7o5`Nvsxy
z(n<|ywa)Y<72JG&1UquPm{pXxs*t)xGQ!T7g}P^FP7Ad~6pT<MUDAj|ZrHe)&6~Hf
z^p9nfl$J4a^sDH)q{>x|#bU;0vvSux574t`ceZTZ&b2q*N^x-sH{Wm_KA#U=(~wf4
zDgu8C%Qb)x$#<IDFLWMmxW-s4SoRe8=n!kB!lmQaB<s>Wq$XI%DfTmWRTXFSHZsBE
zqfA^gFbq`)JU+?V9qvWC#s5gZ<Ltlo>|h^S+h==>WGmyfA+o8GitT7))LJ9KNCFu&
zXwW~YO`^#71~$F3Zyn`3jqK}&F(?r>NC=0#pqR=A8QR#)5M?)FBx_ONG!kzbBGCu`
zD1w_6*l2@vD-@cbs)i4Q=n&_Jt{&*)2K!zTf%1bAXQCW|1ToSMSwZI{g&QQZAvE2X
z>%L(9>yR)II$aKc!{HdwrAwD7g@uK_jT<)t$eIv?qn1(D2z##u)}DLrxg&NZho)(|
zJV(H(2#yC*fq2-f{7V^#T_86i#Gg8Hk)>4Muzl97{YrS*G;)*09Z`B2$a@rF8Vs1)
z&{&#?l#^GDU36ytJI*IAzW?I(<eU~Te8j4?>nHY`o)pzOv67vV`}Q<hg3E8UKHLQx
zkh%{Y=C}1&hZ34k&p)G=oy$)T8@6Vrdsu#{!SNTacOe8=OoAnwT%6g*&L^|W=on{W
zcY%+LTpt#bAiva)%cEmb1)Aor+oWY=WVoY~QXlwAum0g7HMCi`i^SL-xIH!Nc7uj}
z`}86^eG@_m6h&dh;`s<6m^tfbA|k@szJ3K#N`w%cd(Oase#<K=&~zPL>gc+}ybpW5
z*}8`U^tVg%)7LdTZZRoJQnZP0f3M(zej&_Q>ZIBuxu9PN-z};jK2l{*p^xZrm6&jq
z`D?1U@~m)n75KRDl(1_~=wiL5OOl0GzbeYE?X#Uow&R#Vtu?Z|yj-zbtv47)C9EA4
z*iaQ<GfUh)(J0u&<A=<Wx@4I;q(ne@b-h@(5RF~Ou_1v(ve*Sl5r$PW7!Dl<xQy3<
z!0m^Us=(DGHYBi(NipVG%jq1Jreh&9OLRP1W6PGR7;~@oHPF|Tu;#@&A}NIE`rdo*
z4Y>H?i<id5#mO2nf7Fu5V9cLC-%q{8{$nA8=<~O~nf+lNtJ^7K6_YPaQ$!emq6xn=
ziwdh>DfC*Mm6@pvyz)7DK-Ff}j#-+btNI->Qr-o0YVm~DgM#G21R0*T?!Uj>bi$)U
zs#-qfuZ^`CjEW888nfx(t;6=H<n#Ag)8!!Gsfmx2ns!Vj+Ef(o*||MCYUgpCtX(hY
z7s9u{8=L=r-EGWWZTys%Z>{FRt7FK_^HEu?Q&g_8WRr`hZj5E-(n{{RB8uGNN)#nf
zsJA3Hb5>+h+SkA9o*M=!frTonkgU+X_>h>)0--crHda;Z_nx&;Kqiw3m#doSs3@9?
z3=%@%_iOn58d8EU62=u&5rN=R)kb*GPcK_){HnV%&00vKLf>W;-J}ZqI;e`EuuQ}4
z(<!gg&@{=gK@t44tdgV{KW0_XG2Tp=T_H2i%aHydb!8>DWqazX0ksp!b{zE(LWt*|
zf4<Hd`N9h?gb5)G6>V28l&o*q7<wgw-7L_hkr<Ox5m;ZkxXQ3FCWqG_Q-on%+?-dl
z?Udlt17iy#?BFoh3wldwXjFEMZT*YRE|bUx+tONhc0>3z9a$r~Ed-=RH!j(gU9M5$
z3bd?QO@`$%Cd9aImk&1O)f3X<mEbM}TO0tPp`oX{-R?gM3kwBx>N>|JYA()O^V&x<
z?mk1T9{c;*pxh%Bu0%GOmvx9Bukf23lYe^Wr@ysZCijjSUCI5=;BJC5m@J`fuOBHT
z?!2`~Df8#gpTDW)6OJB8DP@<Q-9H(8Qpo61r$%A9pK6azh)p3e+Mqj{24$5R7Yzty
z(!6s1+*(a!s7hprN_MdivnmYhrNvq2$F2>L3Qg?x!1ucQ++M<*5HIS<+ZURPu4mfl
zoM7h6STi%1I_n-2Ex~TL1rW@pU{X!!x`yBHN0;#UO|g_$>C9VG#cwOTT;3p~+z(99
zT)J24ca;}ztP-NXS;`f4>I{k^`~Xp*3V&>NadI~svzApdX<j+^UlonjEXd6FGO(`$
zmq({xcf-wHP^Pgy!^3IE*_pS-wY4lS>$kc-TfWS8wBzVNZKn2#C!SdEc61jKU?8)b
zWL;v6A`Cb7@ks`8YPhgxml={c4H057q=9PDl;lpZ#jnH#m7dmRR!(>0Ozo5=ERChr
z_2VXl8Dz5|@|_wv<qfyU#{Byx1-kbg=0KOZ)WKa}WlK?%B#*}<f-<<<v9B13vU2RA
zzpWg*XcegQCEP&M$^C~Gg3A-Mus1jgv+)l#HtrEG`Zcj%8Qj5N>abzMRMliXyA`L^
zB>b*Id=+~D<{pc?z7n_hgDDFt%3>o_wr6|E%r`hxRH#CgTPG?^W!@UYI(fwzVeBsO
z(WkSObs28dnl(?;(Q+IP4InZx?fU+w^eE_ma*qN<`p@@Dao1;hRY0_4ID)KvA3L(W
zI9)n6i{ZX5Dk{R~+q=S+K(yFq^VV%hDe?RLC<2ymbu(>I1)1AD^c8{jz?phAd!+WZ
zP(?}aKHavTeMV(@baL8DwV!+a(g%i>b5oXwKB-n*9-RR_?bxjf*+o9K?Di5Fs!-w5
ziH$HJC2Zd1!KA?I9c~t_t9n(+dT&d+MY0`7JsJv-19-jO`n%Ge2tuiouc6v6aCs#T
zOJI^phe#vFFukZg;U=pDR3V88sc)MSVUR<gW{8r<gao>{qZ~#hnaT!6oQx7dpTxkp
z%L9cjC7C6XhR>leYhb)4fTru@Rv2w8YGa~VSP8pI>q(@|3|(UylT84ZSI6zEzeu5i
zJ)vO>pWp94ddbGfyB)Zz`XKc*z$jlNlrnUgL8WgK?@g)Pdj|)ORqrqDMx=k7K%BDz
z0&D;PAOJ~3K~#s4N!?aM9T0>N_pE+v(c+W;J#cc%t^c~X@#&T8)=%s=Z9iir1MV69
z_y_tc7Dt<x%k|{11F56y90O$$8>1{Mt28<}^}q3v>ZsutL^5x66}>uH>62<>d$yO9
zSQFD1J2|<VjkH7yeN(M;iZ@eOu5oHl8=H4|NYC{VS?=#8gjnDIlpgOrd3|g&z{``1
z-e0+TTdYj%{O<R9_5If<#=3ix>Yyw`r#wS4BfEl*9V|?r@go-vxq!dlbuU5)ygn~T
zu-dF3!Rz&6ci367^iRfrI)P1_w<3g*=~n9+H>oD1l!S+e5gs1S(m$7zXc8>_qmtW;
z?AWjd{E9Y6A9Hoj>M}b|C8JD#>c-fv4y(eKbIYHKO-}uy*eWs-brHRGryG+BX^9q+
zV@#M-7|_#>rb|AbQ%-!OiXsFSlRy!$Y)dt1iDn!&h4*KcZZEL~h`zTI$#xtYsFnQj
zcs$~hPd>>XJ9ex~2;sn<05K;+QH{cdl(4-}&^@j(X$rAG??fX;vNqN+qu97jV`}0L
zRRuam8Y1964R#bl_c#J{c{)YayTHLHv$(rdr*rhddjNKog4^GK)Dc*^5kS)=+Y1Fg
z&6s<~s6Y~1ROu(Pr2gKI34zYhM(k)XTwa}Rg@MQ7P6KOn-CMWYozd{)8iDd7ixCB5
z<lPS3mBWzS2n+=50e*ieK_Ce;yLs~FB}gWtt6%Z@gjS0gTs?NlwV?kG#QsY695P6x
zf?({*w-zoq<=<!L9`5-Z$=I+<ycicQuQI6z_FT0`rk&q%tAmF-cQt_2b^kGyXj6-*
zq_mNp6D&_y%z{#<Mtqcs@v};?m;^pea$&zvhW3~Iw5*bQuZX5&yg`r>qD(B<62hsy
zhx6#rj;5JE{=D*=snhS)b?V8zFf#4SLd}oOt@NChDom&L6NZN?QYb_M!UGnw;G3z_
zxoOySTz1J&>~<SMC@8{+pHzf`Pz34enar8<Gc$jf#j{U6j;bn{Os1NuI6nNE;m!u|
z`~5uf#IuaJISQ>pVhgfq7fGEjbXS?Csl#N(4tK}2E&(n4Qua9|xyy{Z@4Bs<NmW@r
z?OPH;(=n-nq-ZlCHiab{U7X&_PJXGM^_gz6i~PJdqlCB!6Ma%`lvQe+(aYX_?&_*d
zVTs*NcjoOVfOaC;j$;r51`MzPZj92^Mw!2&22yEW4L0W+1S>u?@MKjnVj4p#j8cda
z$S5&5Mzql#F(HQAyV7F_eQ(Y;^d@WLC2L*a`3*wo3Mv&G7M-}z11FP(m6F1sOznj5
zKtk;CYiui2aM#EJg<0$84N&6J$t*D@Duh9b6T)Z$l6<Gmo-zeJNTwvdX<-7l+bwG2
zB=`5y4jH}~rn1s>7l{7=$t|=N7%t}X5uA^zy~vZpM^hT|5g&edP!(ZTja_^n$T1up
zYjB)U#Ph)2hkGJNGHR@n2b{sJm~CyZA1QU*`RkBUZe6fo!LBwR@9<!LcA-Dn+fDv`
z$SI-b+U;R>v7Z@BoXlBSiNmJQG0seBrN&8JZTS3>UpGhb`RAYU`NXg2)2k<g&puQ6
zcG~nYHj6Ngo>ZtSiYQOd_Wi5IrR#}F>2}>j&9yvU4Nd6Kpp2}~CpqmnJJDf<WUu;w
zLbJeJW7+fhB_5wd*TLhHOkGsL^S8t^dqo*DCOk`rI8)7@zjhfl;x$bw*enX8Zi~ij
z5>$Eg086AB$%k5{UM00#$_vtW091PQr(T*|)IB;(O_Lx5%17hgej}=Pub#a4%76Lj
zgK>;~Zydj_n#OHIjlFDoo{wIgEbPjc{P2g9E6xa`Z>p_si@jx+hpF40^QGj&%2-=j
z?f6H-BG!&}>=&-R_S*jwLKr5qtZA_2{l<-SX}GRSU6M{wqLH9@Vz}Y%O)oOa2&5Mq
zBr~EWsJUaLK_J1%YECbb9G6gkYj=o*a<^fPtVLF-f=Tfc<=FrGuEZtDDh(ubc5`Ew
z_0A$uM<N9VS*}G{wMIs<f<)92=7=WBepUN4vP!_|X*vRY<6Dq)E_A#Mfbj6}$Im<O
zyorDQ`DdL!|A>?TdSZA7gn0sRpbzXMkna-hyq>3Syp@+HXC6o=E5<ImRe__0L>iFX
zy7HYRubp!LnL7{nRE}h9%4SqtI9Hq0y-Gs?AZ7gqZRoB44M`bvZ<|ZU`O0^rxrzYh
zH;i#RmQb3Tu>e>QncV5a%4&Vu4TB>(6jx~Em-zU<%c5AD?&gbM%eeJ|$hvAb3w9*&
z#e`3Y2oGn<chh;~!GDppYc~?%ijP!77p(FAyDUHBeJSfhBa5?l?2buDz2-wvHTO|9
z)D}^bauQMwW4<lMViEZKl4_5{?~{aC6|6cqJvzDsMFYQ->@M`-^5`tzSdGIXuqpzt
zl=yTII-pBT2uy%qgHS0rw}DGr2R_wYWTQV-mY4NI-S?24xmpO(Eg~`X@te<!cx7Xz
zC*zx`Ga_Gl;aQ<53Omv>c<F`zF#6RHHm>}F)C4n0(I$RgUdeGCtvq>QEPD$5tlVBr
z{~mVM>~w1%&nkT=XII9%!E5``vye4|{i&pJ#68&B(JH9bj|7N|i&N}&`>lWx7rFfL
zy{FDqxCI;YATi8HG->Dpj<myZ39z-$xPx~VL9YaZz=c{NHWZ3#!slE*LBq(!T5<5r
zdG+<Iq=cQtCVDD5VYYp@Ojfudqd4$7RK@7O;e4f(xEsVO#)TTzQGjZnPI{4HT{NqP
z1=SHmo;6*PQ=wDfY;1*Wh}!S9<E0ooW|8{ROD_!?IB;N{l*kcxt37`6NhtCMs410s
zFp%PG5KDOC#w&Pr@{0XAb=}xSrwJIx(YN;6kovvABZqq;M>4K|V%6FMh?OL=@5QGo
zW@~86Ee{U0Wf3))+ty<pZj|NhS`nL^I>c%g%dQz19=bhS^3ji_+&(mtC$5jBs9ZzS
zCBLP2X5xhZ5gB1@%AbAW5&m}b?I?=E4cA>AUsY8_?4A|33!(1>W;Gn6I6reqbYj{a
zAIPeS|2BmxHUzh-4%|*jwUpR}KnJV{gexe!Sv6#X7S43h8-<&daFm*@c<t)hXb~2T
z4^GisY}CEu%k#5d4{jcwm^Q#_F<rT<sxw0_jP0{-!v=1>^Byi5ay|pkH1;jeKKUqb
zzU^Sz{Bc~^FQl$q;=DCgWajy}^}+~t=6d<x&t)TWveKIrWIoy!vqnfsO#DA7L=z^b
zCfX(^>p%sNju3z7x;`a0H}|)OZBpanl6+?K?j}y3mz|R{q=o*c1D6FI-kzP4)2rt5
zpMd~*2MM!ta&j(h(1(Y;3mw<WkCXrz5>`klzt79d`@K1$oRXZp4d@bZP>6Ba+1dB)
z?Ro%GQc~6-W$$J_-jJP>b5cuvv?Xe!f9f)}Cr+H$Q3zpNYv~54*iqkD5k@SaMK!9<
z6jvG6L@ABLnPL}os!6)3g~I~9;$eG{VO`8BF~}uAY7~T8j6R|rjqiH9*{F)Mqo|$)
zN(noP1bq^9EX~z0YygknAg~RIoe)N1j4g-=3L)qbXT*2f%?8n}MMg1b(y+XCk851n
zro^SOt5o3A>VJE7v*E@Mu^98nDy_ecvdvJw6(WuUAUr&L5HQE-bRKz0H&5Kq3q_p`
zqBTnZCLxrccyjo8yfSfp%gtfI`wK$DF?}MAgpBM1fRNm<<DH+L?$#~`v{!7&P+o`+
z*RN64eOE=Xghn6AvWOnP$MgH481HbSC?|7sQl~Cs&N)70#2LLEm`#FrzAxdytD`9@
z*VtU%hl!t$BP67L&A;xNE4lXSE73GRpH7&>AIp|=)AiS#?DPA7OiJrHA<v}$Q*Phx
z0ps}-mE39fZYdx8P^)_5UNr<W0viI`KG?WI5L~1*_6qImB05udvru|xAefw-oNNc`
zq(q{VyFPX2tv8K2@%TPUS$R3#yLGh=IsaTtrpC1I;Ro*J!w?5QPJ4r4gTk4&ri%1j
zFLzuNg%B|7&&rJ4%ubDnYAyb1@;?ejP1=4)d9NbGf)2^a%Q___bggZWSy-40tivF4
z4PS%O($bpn0~+YxzrPt65`3P7Z-d4U_*^vp8w8IV{4!z0{&fJ(7VwZF#De7H<h3a&
zDJM5MIivV|a6c{nR+E#H`y=J?dzs6L9XfP4tVA-HJIBPtbfxy%g3BR>+YHg}aGN3b
z(>Dov#MN8vynci9Z7($9C~ILe8}+L@L>iXJT8J9?piWVN#~UKjZit%)L_O<*^hR~>
zn?_~{_#1ejgfRMxXoib~T8;U{g$9lpW>x4AX(SUhB#~af#<oI@okc3X8d364t6{P1
zn^a#3CoUAigRaq;S@&Oy#Zq?<9Z?cMPY#Pfs8ca2(X=*VK~ClAzsI!P6vF)Ghk)ib
zWk&;&OZ}nkmAhKN-$txhU+fg$?aw)iYHf4<NU8hr6l_3Bx&HUxe=ltLjAH{X>3i#w
z*`*t++&cYx*g30@gB9D{Y_IOe=O2#`Ad&TuJoxBSL_~!1{@DL=;rZvXWy?0+cxCjh
z(bf7ZwXLEPJM}Ut>Oyp)wn^WM<F%>-t(Upq_(gZ~v+kZJkiS^X%3POL{~L-(P8)jH
zZNo?B7ZfO7pN~OjpG9F|F~e`ao6@qfCi}bl&fB>B<`GPs=On_RaQo0m1TbT9g=bZ|
z`yMIhH+yM+H6qyv5(a3hx-2m<vEE`Q#FQqQTHJ1Tv%ALJ+}y!5VQElxy{W!hA5tEV
zlz(*S(4o~b|AdtHv^cD;-`iT_9SDLW8x%#U{~MTtl1SYC#%UlCc7t4XiK$PPDXWGJ
zxrW8Eu47@4styN`U>og#4ndjKPPal?_1<HYN?nFqyCFJ81}4w!D>_CQ^Ka-1FRs$q
zkS8dwHZI&^GU6tCB{Ytml(KOha&u^dkEQ*{lmN243LMvl19U_8H@j1>#p-t!#tA7O
zYQuRP5||X_yu&?*qXC<98KvnhatZ-(^+VGmRg0s|H{{;pZRp-Aq!7ncHJg@DzI%7c
z#$DQZFHI_1vUHORO_x+S&f$OK-p6Wf+5qs@yW=?d#1j~H?N#WyWX|0AAOxokIEAy%
zIO9$s4Dm)yQ9ImtZaBR<Tc1HHD>h00YvZ*ld~KX>_>1o5d)>WI(dDx#vF88&^U^3G
z6<L=&s7lVe?2@6PxVVJe+<cM}6Z!k_8~D%DkMq#u&!Fi|KYE7WIE>+Uk78Y>pKW`*
zyfwAB==-G=mzC#a%x?088vNynWDTNKRh?A(M$@z=$)BPq!EWn@xT=X`S-bP{_Q~zl
zO0Y;>pOzXIchI$36B83d5aPz5!^JeFVmGIxq#OXVX&p!@>nva%k4Ly%uKJyJ%*miO
zAVhAT;o{YsxXEKe;P~VkVRB*ApH44^?DE!G5fZ~1Ydz*x?7i>2pu#Cpj^;9~_aD2<
zbkd7eG+jZd5h6c6scDj`OCxjL)8M&iazJ$bvrpIcLciZH+-~=gLnfcOr61^b9)gi>
z;ql>T?X7{-mD{M*s62`xkpm9*9F7KD`{>Gb#T9buzO=O^b~{T^&21j;P#Yhq`~~78
zgW=5EQLtrm#NdyADx0FmUBdfg-^AR6ix*(l&vVGl&Eu}yZ=u@l=8n7XWAHg=bKn2n
z#k3hevNJt>dsJfT=%}Q$zoTh-W}c6GE{_)RQOY=}F?oaZO&zCI;b|i>`Gf9ehVI^^
z3tqB2lri^S5pByU_Jc0`5s7IxMkS_>*M&aiwKw0cyymLQx#Xe?xaa<d*|K#zy?gcK
zic2rzmDk5K+yAv!UBTTCz0R9cOV_Q*)OwfYWwod`dMp#kFA?H%g!o*-1OXEzEWn5_
zYz!$+OG!x?2#}YTmkw-hqRpj^j@5s5MH9zb#TqGLlu(qDYhGsxDgO>kqiLCyFrV4{
z!oiNID9TL)Sy%;p1S|<^>Huy!*zpb)wXu0KX3W?G7^Xx^By=4f*s!<?Hs(QTb>nZZ
zR+zkRV!i7)r(96wZJiaNV^pBKxzeNJ^*6mW^!Nis&1<E4S_73HNnV9<??>99Pm+=S
z9W0nGG2$(Qt+N2#Zws(Uy4~)TCX?yN%COP(CqW#FIB*lHe6rb3N1`D0TMofI4h6a`
z7`fo6kF7rx$jCLlsA;q)6J-gHKO|P(8vRIDNlD4lLowcw1}Rtgx_9e(+?%hCLQ$Hg
zJ{vb};mfbT;iVTwpzAshKJq00`oFvA+9j16ZoO;TsMlT_u%RUPtS-ss_y?|v{wzAo
z{ANtJiq9{(^}@&`O#G3WU_N6jzGug3m9nxG$)pZ^rMdV?ch5{sH0Oj`lo?l@6@Hr8
zB#4hx3AKyI9=Ixc(#hRy$z?fND}P_MDE0AYo_}&@dM5w-;BB6M?gg@T?dF;*FURBc
z^21Nf8yPRVcqmt1dPy&#w%|3Z!7)oDHC?|uJ3ITf?Ck8@a&mJ1z9%Q=mYkfNb2VM>
zh!B4^b&ino+WI%*n<kocPE1Tp3qF{foZJf|W-$O#h_4Ug5*lLno;`ck)V!AN$<3XR
zos)BoghAAjra;`>y?gg&g7)?e0vfrs{kr}?g!rUUV<``tdpp>*{mLt^xHL_x&!lZi
z;lk}Rgv2-GLV;5f%;jumBX+V!ykP}2wMI!Az-EH5AeV7f)BBnL5(!3}VsL`Yfx@f`
zq8-pP-blQ$1@-IKB?V5&#$3tHqCiVzEo5Ldxa{ogg{?EWR>BiECL<dqzO@}B7x2`L
zsZD*fENoFXgy?n1=F)}`5SQBd=ugr-6u9)!<r_*W*-NaXKnNielXY-g&qeoE;Vsz?
zuz1Ca6;0H3{c9j4IVq_*p_iiKVn&P{&AV^DhRI~&xlylh!Fhujc;@MZhK7>Xxl=zO
zl?NiLc_}@|*RykinXAqUJN~#%R+ewACN0sz;1e9y2d{}q>Yi-P&yw1mZ}=;H`EtK1
zXEq5BlJQy<3#IpyD7zSS%LS2x&N;zhKIeD`OE$Xb(ZS03Cx`UtlWJvsrZ*xwG3`c0
zGe2gx+0KoPi6%Naig(An#=jqXn&OfY{`1shd^Ke%+jcas=GCcV2Q$(fw=eTMW{G5T
z$ji$s@cDdKAw;8{qJ;jn|JR#|ZB$jY(Zfjyab=^1Qs(65=B_%(>FyOdIXMdv;@?et
zWUsEQJo7;NO-xMe3j`+a2=QlLUS7J-=bHiqXKkL`F(KiE10An5(YtqVMNyPK0BF^a
z5ya&q?S}O&(q5a1ELfXuxUqw4Vn^5viJjIWa8RGvQwxv3=`E`_Xz>YcqIBVb!depx
z86HGxoo-2bkz{SQAgfg1_KUjjD>%7l=OobUNpO68eBaiaU@JgTFXGT<N86Cf#Z6Cz
z(>VqdG5{n*X1b1zYdI9i%;iPDMsrL;TzvnOSjB8Rv90IfDcX$md(d_Ln5<^Auh_D6
z`xdXa@#fv@^>W|8ALr#4pCc+NlK*`=fr#)hhFxn!tVKsfvSh*R&bzj4xMjtXc@N!n
z`>kjE@Mjh4Gu+rMf)l&iShlsA4sm7_0au<G_P)&`oOwdK{7tR0e2YBr6r2*NGsa)3
zufsb+NNvnz1H)&AI}~$bl!+ZVUOL8`v6=;2cYDY!_7fi$f9Bg`UY)e;_g|k_zhd#w
z^UoawKyp$NFO3|*y$?Q$-|uJ4>o4)xQ~$wP*|-9%uInsVxH#9D6SGgcv41%t2~bc_
zkcX7>n)s+=?SFD}b2kE+jao?Al$E0~krd+FL!n!llauo~4O0Fapx=S^sj6yINjCoj
zC@3hXM2H!|A8JjXb4X#eTIC~;JQ6L0hy|cB3;YcXwic6NflH1s{yibNKIX8v3O45p
zHswhw+->)FYs!7x6oP*``!1Kd|EQ?e*_fxZr9e>ZG6qTrgA<bjzb|tGRVtqos`k{q
z4G9V9)^-y-Ae?i^M-<}hrVcBTK8JcL?Km=U@gu7?Rk&ocb;_f=^dBS?Q;Xu|2a6iF
zHoC6=($-@g$;izs`0uN4yxr*dV^2S){W^ca)M+z*kU#%2mkpb?@W_LK>UgTEl981~
zT54xrdub#Oj~Gw6mdv(2Ug9E5%v|bZPoWPXVAT${`MR^izf^U_TrKo7-q$L({;KbP
z)Z|v_=XI^pxl8zmNll+T@c6LVx=wtFQ={6e^TEtgIwzQ^a_eN|`Pkqbz;APY6vJ;E
zMvv}YiH(U4=&#ptJ-GMoyLj-?r+DSRWBiMjEdBg}ho5i<Q<c}pyu;dc8=jPM#XeQ=
z`b!eY8mw;OSYW0>)9&p-4c*3RX=#y2d0L}Zt>@O>E2NY@q+HwRICMUc{<?SXZbi!A
zcrLfk=liO`n2p@q0*3ER9;vM$rKC@vK6Q6NSXfxSOWzcfIO5lh`q;IwsD@Nd-y|d1
zMgUb_!PWvnVP)Hp$>7ZA=G^Aypl3@{?&a1%fm3H|p@Pe+AgD<?N;1BehI8<1MgmeD
z1eod#enN<Y3Z-auNO|Zjq`H25lgEWTpnLI%guoqU3Ogn%qO=t<i`3E0WpJz3kV6n3
zDe>g3MM_yDn|N0Il~9_OK4Z+g?>+U%lOy!aTeneET*Aw*k72^3uO2Jfv-8^jzV+^e
zFD8G@>o2{~C~OK~>7UEF;<}sIu``{4XP(Z&c|Xzh^t<`+XAh}~W~77=n?iPxpXe~v
za@iRnQ#2_ObdPe{G_Csk@mdvC`#>jk;3wV9yISQorxa)U{bF&y9*)_UoEDlqV~LZr
zBnzcZjZ=@avnk8NH_Jl#{NwR_JnlVWWB0B`HfZ3P3_ANPKAkZ6hvMw?+osQ)b<e|(
zKcnx-&Y`fdh^PNEO1|>O*b$|<=@a(#d;QB2NeOvP91i&9$flEJtS*;pQ0*J9*Lz9u
zc|cK4PR^kVXjYWsCXR(4NLN)=Rl|Vb)ZnQF1qF5X@7&znIT#t-4H4>dyKgv<F<S#+
zVPO*Bk3aq>p*Eo^q$7ki2x--ILnG1`<Y}^+4Edkli3X8U6#^>6v=*dnEawyK>aIm_
z$;R^Py;tZtP`LdHs$yJspQIWYq#)0wSA*?^G9cSMyc^i-#R0rtZ+`1f?x4_89qrbK
z5OLgnx4H3grOhjr95P5s?78WON-FP|z>xb^Y;ji0uY!+Di4Z1BtI4%B2d%mkZ+Rxb
zZ}aCjuLby5K}mMTE1!Hm@w_t!58c(h?<u_Y<~x3*yeTrd^C^A%95>{Hv2S2DH;&$@
zsBkjooe#L?%FF4JmI{E`Wa8iV-NU4>XLIHSx3FQiU~Re^yH)W0QYTlP6=v_(-8Nlu
z=wo_zw4St9df$CTb9%Pz#noLRb>7x0nJL|qQsT_H3DM&7xCr&fyDpCII&V!CiUNMW
zWb-Z$`BhOIclN`~{Qf5{8gf2ayLR*CSKlyw<}6&U=0jm`y8b#Ie&C)fqY_i^5Rx-K
z|8jEm2?GYw^Mq4<<HrB*{?eTEQTy|q{$+_I;BKiwZf@?%CJCpmH*{||lKqkJ-61W<
zsOybm!F2tAB7B?DjucY<ui^WedZS-9K3N=AO-X>a-+tSrY1%sDoq*mC1~#bj8c8-q
zm9o*s&|xu%RKE_bmm%E{MOA?$)~rhGmZtB)5KEv%(5o34)UAz9QO5OliD^t;rG&yt
z$;Mp6Qri%H9}ak1T3Y(2$z+l?o9#%++&&x@QiYJxV?BOwSMTSC?UTE!9rf5<tdG+4
zK*bQn7T(2VY5R~6PyTw)HFRCHDC6>1fDrx7LNNBy)520l-yW|$d~LM2eQ5N|PRSiV
znfS>^N#Ws5ORT;0`kQp>kiu=Z+*p_OJ!kF$rhfkeF)`7+_@Ad)wrC*{J@25z6NAgG
zlbPq^#4fgobB+(W{a;r`bygG+Uo5mU-_|NOexp|hNCZh>vF_z%-C4Q=-*AEa_Pmor
z2HkQ&M89sy=6+VQAg9=ml!C8UCh_!$SD80w9(Ub-3(m?)o_T&0FTXy9rGGBx>E~W#
zz`*lay=Gmr-^0_7KO(NU?BWkj>SmpL=S5N0*Z&c(-E~Q%5*==OJvzBdlVLi|;jc_2
zTYTV2DW#C|+eRM<ahdUtP?7SY;Exq44~bj50UBq7D#|{VPN_%7jvc!r#2G<{v$ON^
z76i8yit=gjF$n{bl9CQv<&Ihs<8ry`zyIF<f|Lsb8~Sy~C=mvstZG_UvbCY<kW(Ib
z%wdtZyb_OJ;?*QRP2%-Ss(q45FWAk26Wdc>U$?t0p-Y2kZ^)HomWW1K?Hw<Jp8u$C
zuIrmGzx?twweK5-8ngq1l*de#tPo=Lkyl9l$j7<&tle7eVPeB$_Jre)M>L~D(~77m
ziztQAo5~CL3u2`%rTleMmV3+MjV>iFQq^bv>HPhMVb`9}A*E?Y#<~p~`S{Z>xct(K
z1EhoI%v-=)?|i_9O@>-vbaWICKlm?JF8!VH<3DE?=doz3mGVlB%sd|^RSYf9$y!}%
zqyK#Ad1bWU>02v(<VcONS|yWo*G#wKJ)tBg^Ghkkg_<tOFY~iA$H!+2Be~@2e=_^0
zpP2UTciix|YcQKld^7cXZn*VM9(mwjjC<!z-W~HAD;Lk_^i%tD{VjJ;RNUBM37{x~
zM<2T1vSz1yK(Ef0*&DMw%CarhWmRr{mV#ciKi}xFzI#4sNFk;O2(C$;mXwr~nv|5(
zk;ZP|vh2LPLsrK$IXSt%5TenqUr4!6*;WmqY1;oaE)p;HBqt_5nv&Fn#<>vH1ZBo5
zhHTLN`#bhQKmsBzE>1d~PGK^c<Y%9K_W6DH-S?mnq65MN=S_tLH^A;61JA4#DBn|N
zxOQVh8e8RB3t6R&GqKC770Rpkc2?k@!|R7#C6KmH75Vlb)jl({abW-e7ko)XK~zam
zrIC2#)0=kfcfJ}1O$1SIEiElAeYvEhM30DwkhLKO?Wl)D?%DV1xkom1Sai7OaI~Vx
z=^yQ|nQl@9HULL*(-m!78}7U{NGUU-qoeoxnQ$zk#-sMA5aO|&`Kdz|ZE$^7RV_Ql
z$Hg|;Ue|RVeeyrt{`cX8hK6w8Lyz;sBM(wiQp&{1Q<(GPRF3acZ$DYJW*up%ofvfX
zSqwV+EOcFG<K`_a{c}0jDC;A3q-X5T$jsbP;wjxzjK1h2t?FC>=&%LIR)y{DlTSOR
z)8Mnu>>ZlclT%JUi6>q?mE@$vMsq7EEhR2ChV>gZlAe)CbhJ_Pz;3tk!8>nq!6jGm
z?g!&}{@JEwZhLg=ilUlRU!GhPT2$sA?NEI$=VoW`_r|xKNVb67+}y><Nl7I@K>CxS
zD3?ho8)-@+#PpgUYMUb^CB-3Nd{duEA(rh=UqT2|a#GT5jXRR3Aw=V3qUJlgwR`vO
zk8ZCiYJVv3+;h*#kt0V^SXe0CZnu2xwbv?#4I6fELPEkcK*byi7km#J--1>D-`<^v
zMOE%`0RO&c=FG5)h@)&O;SvgoVrni4u9=!k?t2KiW~sShg=l74wtKH-R`;sc)Vdmq
zWs>Fv4b8o9R}>evVP9vMbIyDJI1CKKfG~ohzMtpeQ4i;wc^P5m%e(yE2tY6^cVG~K
zlwA44P9^W|;%ry3zM?fG3f#)7SkT*A=>_r%?&i)FSAbhXXQ0R`l84O;iz~k&?+66E
z3$&XHkh0&%i;IhY>eHvsF5|GaMx!wX$%h;$Y4Jk{ZxWm;mU<x^iMLeF^pQiL<BQqw
zsI{IA{wpCVveUOJrCfK{#<k9?{y@l2fsl=TbyOAM)Bj!?=@5`k=@jA8p-4B<DJmk3
z)TQK-B7M0CNav+X!Hbl1H%NE4pme?a{r-OEy#Ky?_K!Vtc4nV_&a*qu%zS1h>j4nq
zeK>z9znIMUkYju}(CK$ss`y4xe-@`Ueokv5Z{xPh-LTjE&j30(=WA+OQP+-mYO7yg
zJ+oFvR%o!aQ*{KMFV?4p;_F)}X?PgyR0jtuO0Ufg{gWn9D^?CvfYj8~fY@g|T}1XO
zj0qPPH|xdFkWMM@PzE&?q~%_a>iSHX?d}G9051SL&x$cN+fqjOi71VWKmZ-?$H>r{
z*WHW{huNeZg%@LuU)skv`k&IW>!LFgO-R$<)DY(uX0GQPyw5?Z0X)v0N;{<Lr))cz
zG{2n}e%8*mx7$LwWQcx^F_!PBs;b)Y(^n!RO84+m!rt&c1+5es3xe}vv9(>vm>18S
z936}Fit`~52=2F<noSH2m;L9EizfW&M!fI%5VV>vBSF*#V_!?+dOb4I#J#6vr^G(8
zFXzZ#B)j%%Yh1*sxlu|x{9*fvFPt?GP#MU-kV;)WdL1k^lRrk<TO5Q!VX_HAnq80t
z3AtR1x7&Gfg?4Y>s-2ykA*#jQFHbg3vyBk!GNHJd-6t6jKTo(}NNDNW^S3Iie!tIU
zob}7<GI-1xSC&f|rY1#nfy3x|=hIaqmwU196lHAj!(~aJ5B*_lsG_HdDe&^tJ6_`~
zXYu2*cxX!l&PqR3JJefmd>Z@U)<0)_hpg^h;7&buSeTE)CC|I@ahoG+D_`lu3w-G%
zZ_+OyLj&u1u-*ejJbDKDk;YU`|4e7hr8?Pn3nd)$=4oHTdyvSQ)2JB<3FB{vPxuH4
zy1N~WY<AjUJ^`(I`Go}#fJ)E@Q=%Sr@qWCeFTkk@(@zogV;+^KsofpS;ew(Fa}Sse
z2vrF>RWT%Cky5ZPni*}*p18T}*VfXiIAt=%D0BDrwrYc9BWHu-3yXG>0mVBL@}SZf
z_jx{o(qR3_jf@X{6_{T!s>EeptY<janKtJ-r<|&(0n1xk=SScyys1Z{{gv<%jH>7S
zFl59b4DE1&Jvi?YhHw7|6`j}da${;|jfSo#BR^vFEN2FI0qg`nJnp0|OR6Y$n!zbE
zVpv^;e8W`GO`3eGg?us&<rYRRF_7f_Z|wk58|7aSLcKhK0c{}X5;37Ef6}8gQLph!
zfoQsv%RmhS8v<2i(ZxvUYbx)=2-5;S6JovEiT9jb@D!t!aQ)#&R8@KrjvC+bs<3GA
z8tRInBRJUKXSa2EeVfXpDhPD!nwI=#+2Usy?0ahN<)LSZtB{5}mpRL!x*CrA^V1*U
z39qr<^@*aO;R($wd|p;hOW84~+?bBJxd(cPX`d|FO?;)i;HXSx`s8+~AO!c!MZ#_V
z-M`%#hDsdm7eBUsA~TaS6W~duQ>!79yraK#iIWlRHoC^px6QQCR>U_rW@$BQKEcdE
zu2|r@vDweJS84Rb&5-i!sQKe9?7t0ShouNYN%ITstKCm=Q^Y*UB|ZOjBS&UF&DJvT
zJd6w1(((a|`(JEYrICHSW4$`XZ~H6rbER)`@7-)Z{{Ft>@yc!zUgj}vI5CWDP<ms^
zF~gE%$xF7tsduQiHI~VHZQknm-;X5{Ce_}aluW<#7bXhUyw%JmMBoS9($G-9k#pm(
zqCUJ}sb+6c_HKeyp7BYex6N)8eRp|r@8H71f*C1u&%V%0t|SPUJXum>*S)YH6AhI1
zGqC5Up-^^dQr?H=#i~+eO56pZq~v5|a6~=ApmIuIfwZWl<vz2gr{{v0LDIj%fX6S{
z^7Rt^_6>xz3Pa@vA6@rIMG{Zt<;{5{f}V8nhhUii3bn!Fg+e(5mYb21%24(^N!w&k
zLQ?ikrUu%kd`vWOiB8ss90tID>%%wO0j18GY^0u(q2X^4CZ?<d>0x!ABsIO#$%)a?
zgsZEoF~lNR1z(IS61z3uFUvRzuNJNOdgO$8<=>Wa9h`L)UWYOqrMQ$3y|VeS>XuEl
zjr(eiu@a~fI&M7EQ4N)S%7`3OSB|BJ(a8LZLn5pYVq}WVF)cm?<J5Fw%@TW)@xnGy
zn`EQ;;m*bv4ezFY;|h}S@-Ch%&c==AnHh<C*<(pb&AAh}Yh_48*<nh`=jdb`4Ie_?
zb_q9!Gr~3pr4x|pdXl*>@Zys)Vzd};b2Nv4^^V(BIdzlxyg9)FP8z#yp5fSZ$2E-J
zJ+k_K3ZpGItK|zwzY7Q%sq|*13U3NwFl)vxpxl!+iQ^zx@8eKji?uxr=13-|px9DZ
zf__QzWNWC)l%irUh3oP7pc{^`g>w2R@T<|~t09U(6v%i5@K0CajvXOEoT9)R9wT@7
zkQE}2$Uk@xYv!~<@eLTL^if(6IzyHjS7QMAp`lnn5HsHE=_(uip`oE?B#hD*h!=UV
zQV#+pKxZ0bacv_m*MQmW$4vsc{+PNk0hIAiD!zv)_8-Hfb1d3A%<v2>Edk65e<GDy
z17a;30yK7C)H_o`T*t4*(lL6bMdLeS&P&a=YgJWM8nveU+(~NhU0ek1MQORYxKv7*
zp3QRj>`>#Ghjm(es*vd36x3+*;NQo<X{ktm^|=H^e3B1+o{Ym!;=AfV`a3+lefYIi
z#_0yY;`!HwK1XwJA6|r|n~V)OS~?IRqc<}Ep^5~D{&Lh%`pi|F_LKz77rh3mV0sDK
zh`}X)6I#Ig22JPN#E36We(dm%1uBUYr1RVTXA=GBDMpYq=kaBxyA4tS^V)U+wWmIO
zeRSBAA|1#FEUy;urica-%3kd=ff#p#9oKX8sG#j`>9^660&%cl=kpd>xT}-t+1DoD
z4XN!De>sh;CnTWaN|CnVYs@-n8}l|)No$_FDbV9GPESDSsZ*TwS23Gdoc!^6si0?P
zXJ@oSr^-OC(1Q=89X{p=FdAoZQvB4mFv#C=1H~?RIVsD*u@N2-(Y=)YeZGQ`(x-EH
z*hLle$#%wQ?%}auA@$zVKK7KXM+tCpMqFKAA03~Z%u+vkB+=8^`6gDcc;Dm@r%rR@
z*VYsLMGT<>5@rsn5k^drA>fME(7^o<MkX!W((rd7NkEq}E;6_1B^M>FqRKO%E|io5
zfa{sEtLjy-tL^|IuuLWpH=6Sv&*L5?J@xm*=vPd}iCFy894`nhF7S0Ls89y}cmdLU
z;yZ&Jn`moma{)12Ta{Y_sd*bP%T-wB?DVvXmX;PGFOThMxXRjJt8X}y2``tHx)6~q
z!3Jf2O_hn)LJpEDK0C?%*vlBt>T<?nAz;$+dr#8283b#}+?M(3U^Wq+dK_@IpI&aj
zg9$;VB^%8>*PS=C6aP5yrqDd@{q`)_$FQZ9ZnH(!?^kqu@u%x02s=lCJZgO9B2xe2
zV8q7prW?f^y1ZqT=XPp3(;vk38T#SS#bak4U+!QAx7IW!zW7AGHz=^_xT{UbTpJDl
zvZ%fCZr)9UB=rG+_Sq3p4B=dB>r|h<$vW`yoTUWhrv!%EhK7i7T7yW8D&>31ri49L
zp?sEqaC|KBy~T|*GN8l^Qs28j<%Cq9Z{6ahN}RWW&N{m{92^rSPg9j;RotbS?1g9q
zQkb+gK|i*X>Kmc`N#3O7{)MPD9)>9oFRwceHa51RF)vs1#U?|Z4-YyO6-d)H<G8u`
zleM@;BJ@l*bl6#*2RF8O`MSHGSg4gQl4U05;|H4zN9*ONU;x_M+DTksTAOrNamk^{
z;T^rW9%bi)T6_?HVE`~1$BPoXrG=}O8>$2Hl~WV|I^bID^!$8@S5?oz!^5LN%xPAm
zuDbe><~j`Z#z*<r8usP0<_2k$@D_P|GV{GfB{tz@kTKu6WC?9gLld+XhgOaNS{I6K
zfEo{AASbU8m6V+Q<z`;cs*YH{jJ6xjT-e#%JY!&aUU5C1T_DZ31mlnQBT|vWUbn>l
zjDW97!+b1r5G^=qT|pU~%iBbv{~-JGx-kjPA+6TX^rWQPH#0SFuFv*o^atGNpPng#
zdSZ8dogL=qr!GIU`qh_$RXw2=U<m&CF@r)r-%pEkW%9MBoA3+r>{qoBzMp|Dd{5(`
zD<O*&;SV74z@qAVd}6UtjXpUA(73#RczW{t$h*P>dQ|&ckc2LHV1Tc~3jK+jqC4qK
zYjb_QZL!(WVUgFU621m)6!l?#p^ev7>Gv^Ik!3i>X@50%C_{2}d4*>Cv~zp=USnrV
zz@q+h=0Zd6+y{$4z0~&|`Ny`_4i1=#cwN!-v9{gX)>q$)KgG_5*+1EQbq^0okzvCM
z!6lZTUcE%S9*ew)&PZqQ`xp0eMu6CABtKL#fa&q+VQrSpp}uNL2Uv=@=ko7Xq+uxs
z3<i6S8YgSA)b@~M;8Z0iQ%V5jmzo1!?)@$&0GAMBv0<qdACY5iwBFbFuU~)W2V9?B
zOT^|q2n)stDiE?KOz0AGo~h}S5fXBZQ7fR0{aIYcj2G<6CntB0N=ZrS85tR&w50<%
z3L#^oqtzZZF}<!{xBe_9lE+d&<-~EGWBaUAVu^my^#Drw3eNmlHa)}py&A7tzW3Ip
z44(k&s}xS6<G=b6Z$QV8%beHFz|MH%i&&ZmwIyW+an*byaIVYqkdH36j1A}e?b=d5
zRtF1?eibA<`NA(S_t5b+yprs?TlnI#PHCGrab=U|+%7tc{4;0{jg};dw;46Z;oB;|
z8X|->9f7+yGN`rC&@Bv0Q#6L<GpkmXM8xlM)78G(y^Ghk;cb|sxZ*UwDtnv!>R*I)
zmIqoI$G>K@#gZ?8>ycuy;$aArB!T_TEvaK(jGL7kp6l==G1%1cG=&$jtD-=tTl0T<
zn#UFuaW&aPF@d$LLM(;Fy}4;wdEz8tVPOaup!){P{1{2<E_Z-nSN&;eZcYzI=BYni
z{8@SSVRK-V_E|YB$e?u6Aa7KqkoVi(f@>W|f8kGg@>obwY5BHuo8jogesR>x^WZ00
zOeXT&m!#RJ=o{KzW@O`po8>t=Vu3uo*aZ*v#oHMn+trlXy=SW&yDegW^G<qW5@TjY
zV!Da*0p`^yJW(GrbjTNzzp3euWyjpb+H@9P>Z#=2)e|ad!Q`xlOjk@eX1MiyQF68O
zU<0sjPt5YovfJCL_b;dS(M&{V|ChCOrikT<Y{&zGC?SstER9Sk6<Y)mWVn>{v_H_l
z0cQ4gzIx5-VbuD!n~Bf~H99NmVda2866=&}CABVkR#sL#z~CVlq@HtxhKECyRY5_a
z<j)^hCK8f3I|H7VW|>=N+(~_JLrIN*q0soTU?hh=p~!RHMh?}nPEda&f*m?80>D`l
z=RZ9xp4{a*FBBU(`cCN!(%;G1kL6o}+GGKo>hLx6m(cjaP*SNxGaWU1#sX0bM7~ro
z^Zdx(sSy~MTeG#b9Ue>*7sTY+BP%Zp=(>uJ(-6$BFG#$F>@Ww-9@-CKv0xtG{ib-@
z7@K~zzb%C&N-w`K6;xbr_MDYXYUs%A#6~>SX15{ykYAEJ(%tk&g+&+PqZ`GwR?O@6
z@peoY>=JrCwWOjs{>u8kx$4J{XZNjaLZE2@IslA)Z$H2>t9VxV|Lrm00q9g8g1wgc
zP|u@EEau^F`%heHKr^5t{ZUR;weB;p;O}rEs+!kbaN+zf+?KnzIBjC&K&3b&Omwm}
znEmf_A!2XLH!?`Tn%e;5p&v%fxiBLppvDcR&Bpi5&82~uON!jw+<&R5sReR6g)3x$
zqoX6c`pUrfjOOO%iRrq&S-Zuk_p^m}x7U9`3QtI@JXo@@R9iy>iN~@ri`A1e`fsUl
zr8>WS!UK<vD~-V9$d85JL^Owva=oRf3BhZ>jMg=7>U`Ck^`6;IR!%ZR&mT*t!7u9$
zAk|Uqb$OsAQGPpLac*H2Ym>>|F|0kw{4JY=#M#*IV+(wt!RvT;w467SXdBS%J&Tg4
zuZMa-sTO{E30OGYE@horcZ{$sLy3c#m%A7Ivd4FX-*@Ohk(6R6<m%g1{tgB?_;jAz
ztNjGDK4|YG<mO+na2{MuMw}@nTPq-7xhc46*=5P(Ky(ff*B-v8r30ZWeYfs-e6bE!
zk?=$+j&ioYd?`=5$*bZ3)ABD*1n5PGlDZr;=4>44m~mg&&cz&v+}U$WW4ISRby&)~
zeO2vat`0i?Nj(}WuSwnM5v0+zbApR5iCTf=b78LQysuypCAi@j#n~R%`}<;^Ul)r7
z-mi!ZP=t^w(!P^6b5GBuUO`t_4yNILw_oVn;;!5z9!&FV8=QO?V?o<GItr;169Zp2
zmn%ob2n29R=cDYFo!uWeKVDR9i%FZI7PN>2w**oO621oc{u-7ilfJd>lmv(Yihd8>
z*y9vf5mq0GFx~JIqLj^OmO+p0S(=f+ixpg2GBVxr%dp7<rE;w_=fy@<vswfS|0}L?
zcA4FT{~enSlacLG#KTsxk)5p|L?^&UJy@1b#k{9ueuFK$Xp(1I{xb5KeaKKLR0rea
zF`ntf?|@S7RGk|92Rl)*AF?WqT(k)PLRmt&%%$hhdL=FbkCwa0DiS(WiHfl3)S%co
z2o?Fy&=K=U1R=J92q%kfX;Z4<cugP)Edhcsq|c8hCQCeAJ)6(ku+!f%=NSEzz@X%P
z8f+UJZS(;~AFgtuKGQvQSXJauHVtmPDsTW~*Q{(P0w3Yaa?E{4)mKYN`zYAGmQ!>M
zTi6h$f>3xc4w-z2sb!8fH&)PN3S!~-Z2`I@kL-p0Y|}-;Oq&!=eDKM(U7~0R*OV<j
zk|ZTbm?VzlygX*HX{@`B(x-GhQ}X>(p2H~evCf!Rd<W{6TZ~%lTZLS4X6~W2_T9>!
z3Vye&j+qjl@60&WPghhZ@W1EOp05I!cs0x{sq*N@vl5GM+NTKHC#a~{A8j&5^R_??
zPzKpMo?Q;R>k$Lf6}yXz>-|18A#CkJg!&Gk#-Rzf0nakMID%CXLXg-lbovM!<$tfD
zU;gh6ZNcpSr4iab1V#UsiKeuQx%WRDgzd&**sA}J*2Tl8^8W=H+9)9YBb+9@`oaHM
a<PDSNm$7e|PI{66c&I9AJ+Dx-4*nnRE!XY<

literal 22109
zcmc$`gL7o<^FAEgwl~>W8{65~HaE6y+sVeZZES4YHYZMg^L)Pb{ta(e&CIDYRa5sl
zr|<5oufBRB6yzij;qc%<KtK?GONuIifPh{Aj{{(!fbY8(iPXRkSbIrLClC;%!T(;M
zBX(sbz>~PnVj9lMc4p3QhK{BnZf<T27PeMS#)kH$40ev@S=YRHARr_lzeR;q-2Yx?
zdw8oHJ$>(-Y-fz+fEdD3SmU9XD)#%78(@OATEe?jH#b*r>42uK+a<Q_bofhG0<>*x
znrl*)X;d|J`YOr<`aAX_#}frfP=uE=OpbGQh&hrbGx=D5x81DJGNxyoY<ropwCA4u
zJ>fkGuh`%HpUt{4C))T2{m(=;IE}9U@9c=f(0c!WmLGQz-u!>x5|5x-LH}oA{ITTA
zNdGe*0*NX@g8%G|GggC#^FMoZAgFh>|Ibbj{(t^Ka3s`(-CiMKeR!YC-C1H3RoNEA
z|JJQ5BpN*GrSbds?<`>*6%`CuL50?(c3kd9p&|&8sJj2H{}el1a`ltu=QuuxZ8X|R
zWMpLB$#RYE#AF65F1xkHBBI^>;TZjl7)-m9ld~kAGLb>}$dpmA+fI8r)&KoKV=$a^
z5vK&4qpWNUh>rVSNKFQ-xmgL&N{fozd5Mrgnwy(tV`5^SD%ERiy!Iog@Hnhj`=W7K
zN5{vN^2A1sBitIHZ#L-&J+noKDMf;FBOLjdel8sLLC3yWXa4UQ_{6%C5n`NXX5!s>
zcHe(I%*c;UO=a)7c9v$Ac*KVjkdTp!QefJH*TgSCsWijhf_2hOyh*8Yu(7lEsH>|#
zPw)SW5_sJWLI%4V>MSdhB%E_Sr7xiyJOJG3s!BoJgDA<;`<#Z)xK*CFfRZmqsU3eK
z9gdo8%mr2HvXnj!o8Zr_prB^l4YDpY;K|B2*jU7cQs$<pEi3ezP2u@9*qlkg1j9cz
zm`<c~IhClWsNlr$zn#_%3=FIh#hv_W-S9zUB6F8<Q&1HxVnKoFI)8pwzuIhfS{ev}
zm+t+zm!29QpKrEWv2<{#CtpY6i%m8la0S_cumwee@v#Af)AR_~Vpn)BfuLOkV*4v6
z+DIVZFD7yo@5r>Ejvc<-{N>rahdvI>p%~rEg9f!2YpH~X@Sj1^W1BS>XDhDuydR?i
zA7igzM13qG@BZ&Lwx5+`u^dlkICX4X_4NIk$`JI-qEZ(4wdCImS|zRkIS6S6*T&Mn
z1lXJUReo-Th=>?xw_3M%R$1HAK|@O`9fpWs5?gvTt1R=4Us`6*!$iM)u_CiNJttzx
zj9?r|zU-6ssZ<r${ztFHqTMg=^f@3X$wuBTPl0Ou69Uh#@p>#JWvqoN142wo^d|#D
zszQSAZUA&BvG;9%TIbfav>1iJF|1`z;MO-X-~+pYzk!$|IcNu74gC*@0t)yLWC`@<
zT09q$BjwMaQaF(~|9lDm47{1iaiG_QwOX$^K6{*Gw=SXDgk8_m&2xZD4B;EYd|13;
zn{<HjP|5w}oJYOrnehFwtgp>vm%elK#&+v7u!0NC5+gW=4?Wc;8x>h)f}aGTCA^(a
z_W9iV&1bjS<`5;b&=ox@DW!~P?FZY!5O&2Vyw#FZKO_^dH$nsH6vAvS1Q6BU6;W6c
zpJ*W30D6NRHl0Qtmr4_1zw0SX82{uP<%indAz+v&eebAMxdn`u#bJLa>Nzc|K!n(I
zxk{74PVbm;=;#g7Re1I*7`p9sRi1;6)3X2U;cF*pCWUGvPZTASx5c0z4E%EPdV-;o
z&gbjnRp7~;7e0WL9=Z&&bK8IB3qkuEv4@Y(PNff;XBm!XdEgD7^w~cBg)i!Ehg}jE
zt!cwJ^VkiUbJzd%V!680ZL7mY8jC@vt+t>*7__pjt?l1_4##Z10PCkBl46t*cuz6m
zWPN(3GO~Yo7X*0O&&UJKXMtFOn`mlF9C>hP%8gWh!D3K-udDha{3IJmJG2gV!gJZ)
zm;I=blf%OgJ<*aD1T_KgC;3%(=u17jDAzDX2%JztX$Wn8`U$<>>q&Y>Qy2pGe3mC}
zQgU9j+KQ~_s|~j!SgOBuj;nQssRyP!qAZ`x!<PQ|_2i*u=CPkg{lzk4H6y>Kn4yW>
zciBf8T_xd0^xc=;mYBX|sCxUr(@C+qJT5)V37&@0XMF!!8+%6l%(n%U*xkGHs-x?C
z&H57_`<|rLwPv?RSqOIEv+%|r2dH2dwe^__Rz((o!e@f*K8T2*FS4x*#C)xiT~`Lj
zU1RaTv&wvc5_p0a9j)1kn{4I@1${kjd)@H8oLAQBaab)^`S`0UDWPFl_TM!;OP747
z7^dq~4yX!Bna8=|j}J{k`O72m-uBtd&hT9%Vxl)#fjt!>v?2|%^jMH;lC4t&(iBCp
zNQR(RM5EyR#^!ILU@_pkIAGwsFx!0tMk9$e(@3_nC5havENc<Z^wg)b{e7bhX5-Mi
zN!FQYVm8agATmemiL#2g-KomUr&D+C6J|pNZUm_ynK=5@Eg?{S^vj2VZO<~BX_7cG
z%Nj@b2>V&&{y)0Sic4{T%QdIFzg$lIFg2!(+AT6yX>WJ3%lXPO9{!}Z9O@)G;3h6l
zyvk1J?G{SZeck9(av!*lg`zo$Xv~0YCbPe1jj{cU=nAodIJ697N)(tP=8SyRkN?UJ
zPzrrtXAw53xj>mI;RB2@xPBXO$0r51Ht0KTLYi+wI+`|Ioem-(m%UQhTQ(q`Q46#5
zm;z(dFfc_N7N_ZPGxPAgJD}nst!8Wx(}lSCAqIrvP=Piq7`RiKl(vMndJPGZEZKZV
zk2Z8)Z@DNaT6yi|yFbI&eGv4xPwo#v2mk%uA4(c_czoQ`<$AR~5zBR)eq^;$(|kOW
zJ99Xe6ePhN$mb8MM(E^qSz)}zx3hTpw(F-4_f;zIwH!FQe<Ci|fN-EA+6&48I~vWW
zA+PJGrJ1ZTQUJkxu8)&#4*6&@`+y;r=zbzg1U}u!e?8FX7ja$-jM9FBqSJn#bLKgc
zErfjj{@Idhii)I|n3%)bcHJ&|;^M=^ux6eswgKwB@I1@IoLUBw+|iSvO%Nx<T9#3+
z9?kT-32m%<r=7{LX;PBM?tLc@x^B~4Cz}hivnVjSBg3edpd+Zy2S`Ou)Kbi@k1Y~y
z9^-0Aqqk7r{5y5!%B9yK{4~oPyd17|J4fG8sXAl>3F~^O5U0r+VM<D5;#-P^Bjn3&
zm<*ueZvT$;3<M*d98PjQkjUhJY{dEwO@sMa`uZ{CkdZ0r=+J=x?-&~m4NY|Gj_>>Z
z;^Ja4nGx+xZ1e3nq(U$G8^l_<0@J8ooE&B9z%-tzL^Ddbq$4_bwte|{{j=mKzfEO*
z{oOM0=cz=m=WRdq=fla28-$lXs*=m4nJ1?Z`yj&Y39h~YjGgRX&b##82Ok=Qp4(SN
zV?^fO(okRQ6lTfJolEWjICwxH27?|Ox5ScT&>nhm!x-EpT3Y+VG*_|LK{Y-mW;vuc
z=!fH8<7@9AeOkEV4<Cu9cfJvwU$uQG5B0!3G=oz=b<_#t{&OwVIYdLa+i)<H7zk80
zL04^0Py9k$1OogyJ`B=ejiLJEqltb}tb^l#kvE2HI-5@D<<n5e)MDMQfeTMz+utk4
zv*7gIy7R8=xb67VWb?RvH53$()j1pt8?d6-Lv-r1e%k$YTm_|rM7oN3kRP+?Ls3{_
zI%$-5UtZ}=7Z|ooq*ZB4p8ZTo_}PiN=3iV((q#(eoTL@wWyfdbc2$t_S{bq3u>_+P
zF^&yaOVU9$F~cKfp1d@|?Gs;_8i%v@+d;RGMxX@$H`ta*pZoHNuKhCi?y4&^x9I(j
zOB{V}jx~9Qpfro7LfMYzMKwGt3(GaMPTuWa_AOxGRJ7)=3Qn38Rl<`hjb)0B>YN^6
zmwNmXQ&?PRLK%qGIAn79KVwI$bvXmKZa6Ou<xn<~`13hxoz`A-|8jT1sq($%1tDpn
zd-InY`5Z_?wttDY3yb-|9S-BF!FQ3}#^Zgdr%p2LU|K0aKWjq!%fiLIe7?1IPXFA%
zv&$oCIDP#PS&G-j<Wofx7$HV6^7()?*DpF99BV1;orf7*7HIKy1IC4)`R`>bHM&+n
ze;b`(?Ddv%bp<H<6H<NMy1IKgy!2Rr-aC>A@BmUPd&GUnz3a8_-Ued`f8icig+3~R
zlv(DLlpK4F5V~!m3-#-@SZ6GKWRkCb`k%=&MN$%7OD@&t?7D^7P%;+JW##1X7v<&U
z{fJ(e;zQOgG+D!#*~f(3v9R2y$NVHVf@v0nv=1|7Ha`2?z-pJJ+(DP4;0A`yDY?5s
z)Y{efN31`4rDi$~_gYQGE=&FG`r2`2Ep&A0r|MQ0z;~odvrz}j(;Rq3Gi_pW(k+PD
zq(2w`{rjh+tV}M?_p}&XRaF%R3->hg+_$kJ2GiX4qVsd~j93^k{yHE?LQrkUcD5p>
z&NFtuQMxB8JR>{X=O&ZQDz>7mj0w0Wi-1X6Ui23scj(c{rYP*wEQWvwbxhg0pKu6X
zVx5`K)I(uurslFu*(Bw%iBP%Xi!=#|^fjlkHg4Zlw?P<9JB&JRs!pA~YQGKHmjFYC
zt{yhmb?wgQIQ)n0-2-H88L6MRi#TICD~Sa`#*;`uKJY0CkOS`f3xNkgl6AXdz*rr9
zQPas_^>{iL7a7inyYq(7<>BJw6k{d>4i~mQX7Y)cMlC~7<6Y0=P1t!4zIXIGUjChW
zjHI6-5{BgS*z<nP0~n{Rty*n!C`?XHR?yMmCF#ECYb2jWtq=XNfduYESeoXyy7dr#
z|CCc#N-i^Hn~;}ArZ1hYq2FljoS9K9z%J_>F`kDbWGZbC6RFeaJl*ol<2a2rbXz?-
zfvCZvBKx2$<OcEtRuGREZS#dI_wmfFll-IX0)M53=4I`wG^GdIYS!jKYNy+5>2!^9
z6H-Squ>kE8q4f5*o8m>DN!F~8)>i>74b9NS(>O=ZU5tLzs3C|-5A=8Y5CbXRQ1Tb;
zq7VWq5&@k4C8bQATNhoIOS9t#@jpG4BxVU>pT}<HQaQfwerV#R#!09@DXgeSq9(r^
zuIlP)e&uN!bsOMP+AASvEHd<a?v_}HUs+2@avrarz*A;YK9@Pd)i3*|;Gsr#Wdg!>
zW<EDpmfLH5JwHg?qz)7#C;#ngB!eqBGCwwHXt&n*-SzzyG(MnUdqfC;6ePV1X+fI{
z(XMFyz8ZVQbDEfpV<*^~UT;2tLVbmFB@f*$(67}b_L^H8i7KuI#(vPgunQt>{^n1^
z-x9(r<@&&?jw=Wn*%y!LPgS6KTu6nyus8?&Q{LBV4lh*dfO%&!p8n{k>~63CEqX7u
zx7@|Wg`fMPqQs{S5Wg+2H(OPk^%E%=O*PC}1Vs0#_2({@tzYL0RMz4-Z>s~5#NYWP
zCmFwVa*XZk-<|uUeLf4l6eCwVtUz?UC#V5sT<f_EpBll%Znfw=xGSeR7R4p`O0bu>
zah%*Ux9lNq9-C<vu5&DPL?;H98P{xKoN<w=+_MVZ1=-d7$7P#d<2Zw4#hBwgme&!6
z<~eR^GS?Nj6pn3XKW!E2;&-zuJFwTMW?JzSyiq<F@I%{$MoB@&EA$Q9Hy$bR2H8^(
z`n?-q(L~L_Fpc}`&+HA1T;xs{jZRtrb|#xbFy7^rE;JG*{~F|W$uSk?c83;@F9AX(
z1c`-(MJw;u%K@W&7Q4;H$Lj;M8K?JG3zUcLP&Lf&ry5MwY+JSRrr=uXKws9v1)syV
z&+yR+Q5d#Iqn~bb=UvhDny;+^ABg)fZ1)wQRE!bii$8uC(P?SgU$#|!lp-W>1muE1
z)(^V||7e`JNS&niD&TLvgAlTX0Q|!_7H*rLoQJN}NXR05m#J`@-1t&gY{v=LI&lRF
zwxiex1`U`^w?<zV9-4-MXFtC6>~=gaFJ-_=xC62F!ealV|H#Y7W#ZN3A@f3Qg@ry2
zO%t}M2wm|rNK=LsrEr@s`rQi_&Di@f58SWD($vd`I9oh8=Wjugl`Lzn&e*>djBxv!
z{+^;0=08Si=(iT+flPSAwS{W%MM6U6I6+1?+GT5N;aw1@^JNK?0Mb8t!K^T+$y`Ud
z^8T=m>#4F82bC8+Pj9MvZTz)q{W)A7IAnQ2-WQ1R+w0@JbY!IJp`yZbqs!*|7s3Hq
z0Q<AejU9p0To>5B-9M*O1gYHZ7R<>nsIv0%gwkAJPhlykad}Xn?{fHxYb0|_AsRzb
z4K5A=V=Hsepig~vZNE8k6Yq&@=gbbU3ZK)fRR;!DikDZ!pVL1pv1#p{8m3N>lvDSF
zxcCiMCT!322^-PO<D#AlQVL5op(=&UTA8hj<?$KXULWZdvqP}x&>z+@Uw2$P-QkaT
zM8lf`$-Jv>rFTWy(&rZ!o$JD1v&Aa!)#-IK?H2}?UX!Gxing9xoTryt|0S=5e^E`x
zd(^0W2Sur!{6iGm=JvW+fDwSQh~zNy69F?D=-CA)N?=U15#}Z9KKFSC;2Mv_b_W=E
z>^|EyUih{WfJEcZe==zLWE*CnVeaqT!uPjQ{ahW6Cl&cJ2X04kAWvhx*``=0Fb(=+
z{d21mu;_$9VTL;{n322$QdeA8$8@~o9rv|Fr=@7maw?S^bP$93;)nR8^YPy9RY9)*
ze%7#;anSH^!z@Y^gUgNG5_S5_s=T?7HckCG1tcZxHu*cPH+0RfHY3jMRpl^Wnl<j_
zwwPa^w{JIBT}y9Ub2?l)H-4NHI6!DW(lt+ZN-axmK7IZuP2r;pbup&eSVz_n^(;l$
z@r}5ZYmogDmDl%Q_CH^$q-x)~eg#>y4T-x!BIf0VXL(mnTeoF$tqYZv@=m$krl&Wu
z&ODx^%5mRs%X8c6vdu=2_KNep)~%#qfbdmFP5ZR~=1#huPK05FDjH=``4C>W#pt-l
zcanS^;capEF!;dIz&oMZPSEN8o0FyzT-I?habf&R>(>LV-QvYwQMqIL4g#;+(mg8q
zdgI5j@;J@EL*ze?1^kgPj*2dWWo@|akrg3UG5!a-;3OeqEe3f8qKnYi(q%M2v}<=)
zF%pgI^rK&L^}SV4+2+C^4)*n3Q9>(D8YhF$SK#&V8*DKRGM6<BiLirnpwo4^<{d4u
z?Q=<aU9&cn2-8Yc(WC2%)Oy;Z@o{3bv{r+eR19%O0K-g-ko@PkzE4pOCSBq@Y^;md
zrXRH?5C63|F!O0Nk$t00Bnptv<%R{qd9DyTwf5rQ5Jcq?@(ES-37l<J>wJd0!-gPb
zo2*#H(Z{7I&td=3w5kTe3L=lg1jA56I*>eH9_z?c0T!0diM^Y1?w~L=lZOYMu!kV+
z`<JVo(A(=1zKi@7b{t-SFuVGfE$2=zS6pK6;XPoad<#A$W|ZYNy}I;vCF5-`o1Qq(
zB2(WFi2srQyF8*Wei<lxm4cd;&0ei7+viMoW6xg*wvf)aD6EE68Nf<6CK4_ob;b5R
zXlS}ZZB2)d_eSFCs@FEi!v!%$Ad1bn+vs0#^1AA4uhNThm+vwe@6pnuQOr4~fzwuG
zAehUU<K;dr638jYdQuXv=;3#wjffYJof{pF!Ovi~S@$3_s=e{K_uEAtYnO58pRH+*
z`u@`x_7Ej;|85$j`7DHY<DEJnuwr&MsQu9PPR#5iz%zC1$btQBNzp36pT4p!eXup7
ztnsBqDWJ(6s7*v9oL4C;g&5Fff1}0khk{ob_>8tvrdY`E^y-cFSOcx|^Y7Q0Zr1pU
zWc)*lnHGkbC+If5fbYk&`+=uQRa%Z>i?-PJ;&8kQ_<>h~JUcc?#KNVJg!$$68a=7X
zV4KAqJULYezuVKo0d%p#H+C|9MCXo<-V;X~V-$yCo99wnLTVjBI>yD*Czi|dwHHWj
zZF!IUvMM%y;Csak?}LoGrRDiQfBN}c3HK%PhJjXAm-ucr%9U-tzxRA|emS7<MBe;i
zexIdQN&X~={Cne_p^Yx)k8E@yepO@pcC?+SJ;bRNp&lzee78sF^u(bXS6_hklbj1s
zjTUR%A3uJ8Gt5;9c&<cLKqA>ox`LPdQIVdM^yH;qfcdqBJakH7M_#bgWBZz}m@xC+
z<jYk@KTOJyR-EQG@pFlQ)?k}J2Ceh!-v@KeV#x_gyZ_4%)$NC<*DIdhJJm@jqLu2O
zKkduExpr@D*Vs`|3nIrjov}W1pX7T`Dv<=|EdXCdKmv>1dX?lE$jO+XW+3zd(()J%
zXI6h*`Yh@0>ereySlMk<vwYuMBguZ%SZzGmnJ9HsUac$mX=lg=)kdu_K{&DwAb)-2
z&yHQ*ibY{fJ$iPtI33G?Lg^LY8nhjKW2i4)=k`QWFtyXQnMScMSLS06y|s9Bjw`<E
zcCBEBYTD9DdHV^t$v2mM@gFk}n5tLdtaT6Ersim>>)QhcQjB8}USgX292`byh%*`Z
zE~XdQNB%nNJZ!Fl(<P{K3KK#{NRc?7at7MEAKaOzQEtS#hr|3AJy7$2m1?#S6ZK<d
zZx}LB;`UMVd8>eJuh~m1A45wsd5ewSK@y-=QP=$TY}r;+G|cR$xFliK*0+4wYsHbF
zoZR$Ue*SHZei<@rXP4Y5@%B?PG){rN25|G~Aa^mNz%*&*w5zl8X4=jWEAq$k2)>ND
z(83kaBzW9yBj2k=lRGLg`_0i8LJXpHs}E@*a`(g}+-Q_ccOE6LcGuMECit~YAap@^
zO7f_0LR_X56c^N!Q6Lc(f~9%s=`#pu>ga4X=vo139dRkC$gSZ8edJ==0qB#K1J!qM
z1y%SIwp~QV<>lGc9@}!^?%IpidQZENruu3813bi-mMzt0z<anvg|_5#lb$5~>b>*C
z&B~Eu_I-1$&(y)w`Y^(%0;fUSA@XN`C!mNa3WCiN?+{5L{#ZnHhx0IA#Bu{~gBEJb
zaUA=-4qtcZ9qI(?vVhX?!}mJJQnZ$8*w>WpkSZ#N*vHxJ63q}oYO@L>AdTt6$u3tH
z>jyg-S(JKAj&XQb9#c)WboZ&NQ*=H7<!Ry2Hvtr7>TqruSJOW|8LsQ6UcRNRo~v$g
z1<}A_F~6rueMGE;GZ2+7vAQ@`gS4H=KZE2pe}bz!^Nffu%R_B2TVB{$U4sR_>Zgj@
zQWV>-;-W^{x&Ut4k$tPEs0RQQ5sQ#dyMO%eKJFW32t1s89}Z_uq$g=}9Gmuo5dNkU
zsfe5jw)wg4D-k-iIho4XI$<`f5J=seUaLyV9Xi2&LteVU(lz?o#F&xWO1c2~-CRZJ
zvJU2_J@+Gl&WMoH+{N46ZWC9J#Tdta7G~@kx|i|_ep!6eUs+KFn`vBTvuC}Zo@Y7Q
zC#k91?4_4vV&-I38~TwY={OfjHX;HhQD?_>F319n)N7$lVnu@jF6WEnu&Ef^L=L@U
zayC5m*6JzQ)L2rcY%kV^ybr<aWncp**-C>xQEvC$3(w3fD>Kv*%d8Zg7fySuakb<;
zgZnQQh#$vco$VLny3x@n-{h1Y)Hp=<e}V8y33T<QupkB_NC<TguyrvvoZ6uvuBW$M
z&!1acOs0dqBbQq$;J#83p<f%jgvOevhkikIVkJn2g(k2@_rqX}FQJ$>_irs|E0k|R
z9oIRGX&~*=_Yfv6<>Mx$bn&pD@88#Bz(m?4L)<#Q>%Yj&%lfrfzX3gYyNeE4GMz>f
zxw9j&FmD|8e|p@dm(7zL6LP;0_=tb0rTYo)G1U5|ebQ&nB0^~Do!ra%KII=Vr|G@n
z_0!qs&QO(?msfi!x{1@ZH!qcP8weRA)T&wBDVP%+ntoNODe@kLd)DyPCIYYZ^^Bv^
z-SszaZ%36^@d8z}t;y8Z#>=N_XS$aRAr;YjAHBe~Y~)5xGQu>R{Z<HSXY9VmY){GQ
zm_)ov>ps%&RI5!Yr6<J+k+>13qMaM;Omw~XTN=sd(}V4&M=G2ju$x;8bjxh-fGew&
z{<EdFk7_4r>7`vb61~6Y_KSp=NZ(hRgqRRjb^)R(6p(yVB7*`_%SbodIfx*0-(}?z
zTioUwpJHNoPwFi%uYrR~6j!Rr(iV-VrKyFJnvPm$x_9CjnB^`Tmkqt@?-_xP%9qdV
zOd5@5G5RF76Ut#&6_D*OPa`>8@7=xC=h#~t!tid^>rJq&8+?c$h=Qh~TX`+QyP-qR
z9a-q}q6T}ADRjJkjI7ADC!sf&{gQoB`^n(moAwSrsAkGeW>#)YKFY#>hu~6)Y@w{5
z{-7`^4yL`me#*aCR{rqqolIv2$7#D*o*n7FC>c_v$4^-w1=1p94Q4=I`Ko5*@lD<D
z?6Scx(%x<pv|@!$hncb>dw+hR)lOmJ4U3suE5csfCg^18LOmvXQ>?MY@a#FZV%ztA
ze$6yjwCAM|QKAjIW>mjwfcvs%#fGF-3rd1ML;KB9g28(S*YrvHIQ6U6R6`jemBWn$
zKfFnI%J!R*#NI7+g5w-`3p^;+hMiM)J)hbf4n##S`9Yes2PdLB5F~BLrL`d%j45(G
zbmF(q6SAQG`dPBIes={5WjJ&#>+MdWTWa1fzzxSlJ0MR<Lb^pirzXw0t_p0S16pgm
z1pau4kU91>ujpl=RhccjOuU2otai#>db*6XSVR8q<5gAt_YEJfet=*frQxU5W@N46
z`ZVuh%4lFuiNXhujs*Y5;IkkFvUOyfE^g4P?1M>x9GX5u9kBB1%ACqnqs{hKR;7FN
zjhdj1x?VAKg#DO0nfX`Q=toiW$W`M+iy_+=OtF42KaX_Ztv~@YUfIn_8A%)-jj^7e
z^($uOn<_M1!~drRh~OZkm|y7pxSvo)7_76RT=X{dS`kvxvGK@Y7ilz<#A=h#PGMCt
zT%2XVRQUZfRM5A^rq*H=fIQaVw@yairCq1ux9#|inIN=KDo>11=X#}O01`HEwy^Eu
zJa(a1(4Cr|KtaLhGj@5;JyFMi5nchmJ*kk;$67DyHg-9mwlsSn$RFvTLt`KF!y4`I
zM{*~Ip70CG3x)|_9fC9SXnuy%cnM=h=aszM<PMf<lkwfd!2?zeflplVTLsKW(~~a-
zyRrScjnla8&&+Q_eCoVA%`UQUcQ)l^X2FT%p4ZWbB@%gH0_P6|i>u6p?)q9~#qI{Y
zux=#TD5wLu1l!kdP4jHFzXrdDXVuaTqFlVf1#Q`*@j(yn4OAEmFFJg1TPIrFbaHI`
zY+c`zk}%X?{5zCjGj>OTP9y&6%@6GSh!f{lKZ5wocJhBPQ^x1L*OM}Fm4X_085$af
zxz0FuX9`la`?#*PSg9ZSP1Hq24AIyt{ehjR^*r-$ck}t2sCz4uHc#_WYH)p-Yevk`
zC;vz`09epc(=5OJa2nUc5ItsWq#{i^DHIc}6$@(sk}39z&+<#Nxv&TpCpHieL@?o-
z_9K@zx)m;+e_ZsSQafP|rBtAu50KMLu^<t?J;qwdBOnBc9V<0I#t-D5-@8vLq=15i
zsFSqR)KuwPA|>JkiSx6eYO$iJ=y-lv^8yfC<xG6^HRxBs&j9g$Q|Mb37M4)xFk4{7
zDq6&7*LuSqD*c6Wo5cb1<*JQK1@`3RSkM3g?0F;_LV@TN1$73atwRD)sp+@sqYg>V
zEaA1SrH;?jc)|gh{?E+zO4qY4VB5`zsOI!QaME)t@;vE*?e@?caR}QoA*6C-TBE6A
z7g?ibM&-%xk9=9kGactu3zuoyj)RtZu$IZ(_}qy>JG(!FFqCgtje4&8IjezY!<8hM
zUNVoa@As%n+~@l{xKLRT1ZzJiaOGO9if?3J6gsfo&5y{UeK3m8C6+?kPIj^j1PFD8
zgHU<RFO&YJt{&NvyQwBwsG?}?dRzTK&V*vJ5vjphZp{ilm)I#5XIUU1pE*><DdKh9
zg%Qj9eOXy*2f-)bBeNa#aG)}WkZt|IFhI)@%3joKMYoBYQ3UMRkmfJk>j_MN$Z6j@
zwqYqvt18DC@eNNQ9nD%b`F|V|-QhRfQj&4=9>J)_=nUec%;k@$0J#v7)(U|m<cl0{
z$G#;AWCqO{U>pS|G27P8=J_yzf)^!h>1d=s2pA$HvzMCMzA!I+HnlGJp$V{nlg?g)
zbVs{BJH1@%8!bU3bUCTTr@+rj#KBVqBNhswD$vyFjkny39AefLw2NFA;{vFk#&iOV
z^m>~1%Ho!5TVkwiMi*)4y5@D58d`yyioM={#QP4~jpxP#i36`u611f2fjAEbmIZVY
z*0S~X;<R9u2=&@YGoaI#$4x>^eftC$x<)D-3t^l<)O!)1T!J`=z06MTfGw?%$jw}9
zZBJwm^|`Da-ao?vO=W-jD3{7<;IZg-Cm635(NQgQV;jhnI6Tifdycc6Ezc3<c$)(H
zHTgPIdK=ac1#U_bve0*jM7&IEekkTfL4Ew%2@|8Z^wrVPkx^7sv|F8XOV`z-q7~l^
zbhKib;`~4@p^lauYTgWGVQvK7JpgJ!TC}9X5-{qa+FZ7mpQ6;_vdUAxVI#H<dA0&K
zth!&#8GZA2D>mfleK>}3A{C+<%QTxB3<&JuU@8FJ1umq5XD?eKy~0PeMDh!t-|tV^
z4qj2ig(lT)miXdRm<hB2Nur6*xX+k+ejhu2+}9tCauS?8Yb}o$aH*tptO(t&OJQlf
zY1Uq7Qzf!lBvrSvH#4GE_z02TiPARRLY|S1%Ms1q=F|M*2xmEe(ykrvUZU1`A1o;m
z`AMk;_uGq$i;5q|Jsl3G%KmvUA&6wb7lQ1Kd4p&4?CNTdk44-UBTZZO8UI|pSR$k>
z7j7^C9M}HGL$9UoxMk6#iG_w|Ey&5ugV@n&alI+3eiKTcT!*J%ZtI?7=&|f3bYzxb
zK2Pr~nF&2<eC!valH+A<Vi&xofkImZK7hV)Q<t-yyn1V11k_x|+@Ql;swM1n!fZaA
z7tR3wRa3#{*VUn@+PIYJqBhr7=U{hG_M$8{c;?;?MQPIc(Fb{qEIWLavk_K25RR+!
zZ1OhHWMSHw?ZR8&?LWYS(t3C7kpDTPhMZh32>0#X!GE(<RGjOZVt_A3{vWtT3hh&E
z9=uqJH>ro}JR1o#y-%vgePtxw1U($VNrij!8DwJAX=>R0!Te}gnUI#6qqy9x6%<)q
zx8~ofdZ^i(&WPygG%>BAI=Q|tF>_pM-f5bur9-!$1AcfJmwz2wcU)`LC86(l{o5?b
zkMD{@_$=s@8@1hLtMFT|24NRt3+caL0S6Bcls{EB3vR9L1?{J7T|(hnV3ia7!zZhH
z(hK^pL<g3*4v&XW$3)5_4$AnWV|4|tO69_;)ngsV${x15lotYWF&1WKnIR)v@V!Q=
zKMa~Zmur48X|+0Orm7w|wgV<^)L6PNG_4#!^`#e|O4i0q1Ax64b#5n}`G0tqp1XZX
zEA+$<nJH*L#gEX!r&HB_U~=lxXHZeTW7X2F8ua5sbs=A>9Z93gLOTS!DO!jYVXcxE
zY^WZgiwZbKbS15u*f7G$;_y@(x9I(^X;VI8G|I(g%O#xxwl<SE+%c|=%-4K!Dm~3N
zV13^!4t8<liW$N!^hI!cakK4tzXRS=;w1r0d{96TTUb(JW~Zl@1Bp{4emQJ?`8t0~
zl%~Yg6;ZK%iuMT#zw8q~+<`oA-<f<zo`u8F5J9Sm1*4*d@7=uhAB!XV)ndN*I{4x4
zDZ)w*DXo2jnL{9^rEUJ^-SgQ{%n+|~<GG_2D>`cHXYxGe42hF<4e9O$lmV$`bG68a
zyTohHtao%5i89m*I6i)+=5qw($XVk<8paQ-K=#!5Yb<ZdNcZ%PJa)`)1r_y_OMy3+
zn3^5``t-dn5%1VTq}R&k&;dVofGN$=bi`wVQ+4Y73+9&8O5|xd;{cB(GjjBf{Rg^K
z64tysayzA?8*l@Zmz;Ti{{x+FvDu&j`obsKJ}G%Yjv`|?Tc07ixA%+bhL3mV$AI|;
zuXjteRVR6={9}ti8SM+18LY)%zxC8yWmOfzMiu`qhn7S<-OV~$PfnTt>EWd(0;uZM
zq9@xUGtj&DAnSbBSYD4e)(VkSQz{jvMfaccp%j!q#2z|YnrRYg)eg-%s~4$5si`8E
z;3r&yS6AEuH~9n~T1vP^blsPQleJIXLw#>Z7o+}(_#$pTEu+u)EUyg;bQ!;&`fEtQ
z1PJ#N6cm(KoU59aPB01B{jprPKP+%K>NfN&7y~K-YgZ~cww#rZE^2%A5D|V@dB2d@
z4ZYPH&7^f|f|wrfA@Vj<JpUy@eOWi;i-#q;`1B+{_R)k!yEpt*qut6NQ}s-h50!YV
zOJ(fmV!l$Dj~wX||N6kjM7)ifAYG%y#Xd9S*HZQu-1Arbg7V7iTqyY_?h$68ZVB7(
zJ;_;g-XdAdB%QwE?B5h_rx+JRx;>9W@H6h?N(ZDN=!wQ5J+x$QFDw%L^<O>j+<)^a
z$nLkVwB@jvdn!*eV7LG)LxNd6S@3}{8kPzBEFRy}^SWU2@NcEm>T%a;e7#iiIaNR%
z1*ShCk87E!rha+}SMsGW)20ENU$89bt+uOoM;Bg2H^4XzRI{9kuJNiy$AQy1Fs<H0
z)|xtw&S;et<CufR!+X+Gny_#g8q-DL0<yQb4S|#g^S&X!XHJg=CUKzyt*FBE`|G*a
zYiR9shGM-l=ZU1;($TuW_6{iDQC#v&n^-!l3HDK|3i>s};0EGHN1gC)6-UUdvdE`P
z&nLHM9{?_fLqp?Wo}b3g$?tLqW2aVVahOR%PQL^$dT0<L(x6Idcy1q(iX{l%$gI0_
z2$Up!qH*56JWOX+LrMQ>ozRS%4Zp7=-F#g){@z~A0O#0zGmV`%c)Rm*v+hP4U63<H
zxUdf)run%2<dGJrr|aIUvyu?l;#zn|^6l0C@cj-eu^@^uXEKQmIhO@Q{U<3|RW*)s
zo)E%vw=#62KQ!x0@1KKI@%H`mr-5qhET_(MFSO6?L{<4dZL`g_7>tU#ix#Eh*^#>r
z`K2J{+4=78`RbDn9{B9+AFoM^>(liUQ23ThRhkRo_tW1K?_0{qHR0<;?P0CFNDyGT
zRr<aB@i{~Zs=d9;K(;Ig-<MWcm?rDW$*D^tf_nAW;)1B2B|6z~Wk^U!b`C1QqaLZ4
zOSA`GW9>s^Z2hjBtvM!RKT&`V*x=z&AfJUvhbb2X%%YlbL}bul;bIj*JsWLY>L--+
zj9<vp=_6M0?|n4JDa_Gg9)sI3%DEyzZeu*kpUZm{G-3|(xA|NkNSh>V9DHH%p|!01
z4(;#dZ?oEfnqXnze9Zii3)@Z-s@p7^IY9@>oq66#PCnLOs4LX|hiCBK^FpO$;Fe84
zjvlr_<R*e{y#~)lYBdZ7RD>~$^cnkII>nVAjHNZcxI49(+k%)=f2dro_kTk&VBdl(
z2MA~Y7_Tt{iop`B(2_>pfh2?|dDAe6^%lJ9(_U-jr3wxYj?xwY;;>OyVFkx8leq5B
zHuAD{6}+Q1ZHh<twp(8gNOPA<0_c_pEm7!38`UmHjdZYEHK?4N%AvCsm9~-ieC-f7
zmKbaAJ=nGwEj$!F0Up5U*6Z4R{4;cg$j^+={thUdV6Y)baA`2LZ&|gupqzW%G2t4R
zNF9fZE;(|gs*7#h^WDtpu)(J8fDu8uj;T$cVMVzpe7;!+JXom5eD&(JiOQ1#xoLXh
zejx5~2TEa`l$4aTo~PnCQG48*oXo4Of@Lw|1xCMCi0jj&0=f|u)RCZL{8g}R(;OFm
zc)j#jdn|foxe0;xDsU=m_fmw>{spf>X`8dW`eKzfj2sXK=ePD{V4d1eOH(rn_FUSg
zJX+A5#R>mfQuo?B$YwH3Z>W|kx4PVnjgMxQI!;f00IF1T9@njEtGQnflXj5KEC{`Q
zzLhRTJy!-+ygu=!Al}v8ntqs3eCkyZLQzg>u2yFG{iLR5*86zr7M2D>$FL$G8&qr%
z+#)N;L&O>;8lqvPKTx)7@ooC*Invbv|G2r*k6}Z6Ub1bQ)FV3ETFdc|xDR{2*<?3f
zXDjp*JTbb6$U6crD=R~(U7^(m%3_oh6mEkJaYE<|#j}M?uKKXV%?88i%4~ugXf~pd
zln#RA-I}`%3z<RmFGGVQ{7Gf-Hb%d7I<oL@@Iy`eRV}@i5_Br<BXpfOqYeAAv2-j2
z+siS0?2ysngr!O<XtG>Ku+B+*;rZ8ol!=>obmkkxcT@H@Lb(p5lgILE#xH}!OdN3=
z0+TBvu+XbNd|Z(41!i~3bo#<~Z8c7c;m({#?1uSGiXxOwkfcU@5C8@imTUQbK2RP0
z83<HeRl@KOew}$hyzT<#+i~bv5LYHCxL9}mzNc+67I3ys)+1AslO>rTdMuYJMD8CR
z)KRQmE-HiR<p-~ycBLTwhTwu=5M~-ya-d7sNDpUd2L`tiw2~9REQNf+%T_qnt#&<J
z*<Z>fmW}Z+tK3P_>{doxt+zzsUE)dm7?tVI{n~iA!MnPK3-YPG+sU_Vdeb_X6p;lO
zcrYmutZcHmcxhWSZr*XWyTBYv2iu(bvtz>D=NA_I8iDI+I$J3H_5?EHK({2T==g?p
z1?#%N;9&h74;B0SIpF{hq-)e9jTr=#+ppHHee51DX3`wMbq84qg;H&ms`h+-WfvbU
z)G)NsjHN;(<SD2F8(iOCH{WJ4cx>kmfx_KaT<%xzv$GK_ZzY0z3NLNEo;>J%w=$-e
zC_MFtl~Swd4Y6H~P~s-7FK<9*&HQ+jGCC}*kE)z7*BF**Z2}crZD4fN#;oG90#W^Y
z69GqH2<xI2uzGQoP}IKHm!H5_V|+uk_S*3YUbjcdc?UWwfGEgIG!tXj^Lp&XYCi34
zb}$?RytsyMn}dSqZY%Nx?Us&qJw&-r%dj5dG0n4&{rv>Gw`qGRDKwS{@0*&2sApBY
zqY6jb(OXfSM&r|zo2~k*jj<N8<j6?4a$s-LY%24~3AO&2;L00wdD2e5Om(pPPaXO=
zQaBpBHi87(O(1%}T+l${KYVb-6Bco{{(@}446{r8T<AffVSl}g{ij@;t9g?GxQbWG
za6A}giJ^HlbZbQPc*Hh~5ez{*{?<t6`o<TYln-v#qwH!k!_QH1pjy*L;C*d|r3a|j
zFVw6PGZEM*WdUY~#T_fJnh=N@b@SFLbvMl_=vZ_5x9EyH3!rF#8iCX-tsZock%8zp
zi9M`6B@;M{GKoRy1-m!gioOEu8;ucNq3w;p%jd(^>Ia_OS$45SipDpJiJtyeDHR#=
z$Mdx;Bk|Y607J9+)o*s&&7UB>uTysOz-DP22#ovQ(?p@j_TA47XfBIAqtv~C4}w)J
zC6_tlP@AOzoAtHb{B0MHW<pz}e;`PA>jGE`*P(FO5Q7J}SO`xW`<sAT9qL-x9b`Vv
z0oyqEO>y^>mOvo@d!xcfs*+Z*^APp@O8-R~{LTwwctm*g_c2GW$Mk%$G_j2w`^F8g
z)hXZ2+iV!FZC`Q@uQ{(yflLDsy2+<A?M*P`mY&E+F-XhNc2cdeb2|IuGw^~u$|X0h
z<8nb9TC2=Z`po9GT?g_=%D*Ct`CX~ek0Q4Eu`k*bbtAFV^PHUd{2F}p`tt7%p{EWH
z$H2yh7X+#n{<`l+u!;`pfid`DR*>QC9+_^Csa*w~huVAGz3xr$F&X}8Q|BgwufV&c
zV9h$MyQ`bDude~DA2jY?Ub6Zt$^zug%kOB|wr`uy4KwP+Ti`s^D!SPTBWb+bM+&G>
zd>K$je>cZjJZ@Os_Rm<{gt`9Vow=q?NKJJOojLJBxP)E!!hyS~Zni>~TLzTa@C^mi
zF_woyPK`C_{(}wTp@2f&3x$tT{Okf3Cj5;z!1(9Z=BNYS++``a^?Ez`ig{rh;;l^q
zI>D~{yYmYA$3~<9l)I<r+hQu!vo#FyFA7~;zEP~ph{4@>-VJgjQ~jXsIBpD79G)UJ
zf^8g>57bY#Mx}ue&4Bre>*V*inHHMa_$+tV3c^R&VrrDWoi6ir6t7;FQycvNdN1%x
zD;Vta{xhZ@Gjf6x6AdC4LGy#in-6jm<bzBlFQ2}^w!~T1UM_@TTZ|WmdR$B4lCP4{
zFAU(MlV@(Cw6do6mvkA{T=&f;lc@f2D17bpKz-t|yoX*2c&EeXkTK*Jm1dgSQXSb(
zviWl(Ez}4@v!Q*bjuEV4*g>tx3=joHe2gKr`AvG<?XrRE^*@+In2LsN%g~({o@&)@
zOQAPuSeHyb6e%uHY7ef}30y6oLKv$p*ff^fJ53`^0$~IBuBa&rGUOk88r2!cWQoiU
z`=SdM%`4qm&+RvU1AM@Zce}GZXfEsQn2}l1Ucz$%9OlY*qt@K*ztE1FHI3thTjFWk
z8@S+h8hv-O?F8lQ8L^fmfM*wI<03sm*%-WeZoL-5^Ok&x@;c9}Vb<b>E*A4zJ6Zzh
zT`1f)kT!d$+DbH|F2PXnECFOh*3^Fddh-&m^YM*ou(0=RQopiMYEj<^%s6l1J5-Q*
zep_2xKhElB-u8>v%2mQmQ04#Qcv^T490M8!<LLl(^|$mJvd49MPImJd8jKDYl38Dn
zi2`Ts0=WSuGRE&ARI9M<;(l*}qf;V*+H-mxw|)+8md9=$>4$(sqE=GoHcLSJjLdL|
z2yH6p;>wEO{Jp^<bv#x>(3;YVbD9Ds`QD_B@E#KnEUo)6;gajTCW>-i9#`0?UB5Zz
z>?{VuP(sEY=!710YX>fJ>mq-gKodb(-)Z>Yk~i_A1-SPPAula(I4O&i+<QE)VN#>s
z9q_Oas!7OlOPE^^NO~gf)!dWOWiA^Ml$X(ls?_<KK3`2=j8ny%_1eGl=qG0|z@*G1
z*13qriuRua#a%FdAdR+nyZX*aXu9{$0zz>Isvi$bq*O1WO>}wM^sb)Ye*jheZK>Fn
ztthG+qy}q0L^#y76ir1SJ0mss>Xs+ZuigjXLm}Wx{IlZ$cXIj5D;09o3_rtRq*q>o
z=~L7=9E($?QlzRGH%$roIC;JG51t;z7_?fP5vAatQ{7Z(flm^YVNk9JKZB642}xr-
zRqCZ{03nOuSC;VBa;ikAA%X7{+e%X~IxwoX?Obic6_<8dB@Bi1l^#A$da?m$`ewCt
z8c7fzueZCyiw+8=thyY`&rnqPdfl#0O@G~8iIA)Am1NmU$;--q;(~w@S8)jdzAPbW
zED`x>!p3oy^q`B^*m{C=E5|D?T`oxqp_|R3py^h%FcWZdOzlA7P2IA`TDr4q;fhyf
zxdDqI`o;=c&QLGQG!_;^hQGX~gj|=Pepa(61Y=;!;Gx=(|4e!E<);hNZBD3N`;|OM
zi``nKjUVdJXPTx9ps#DZ<N5r8!s9#9ow*Efat^1~bO8a%xMONvo^^mTi;qY|50)f6
zpz7Rl_|wGLH~$b4Zf3uGPY#ki9Dwxh$zP{wy1r{^%?_{%GVZ8Marf}`lU6GazZU5`
z@!-aVt(Lsv_MnUf*8!Z>2#Z%zF$tx~TVu%1L=8*PJQMSne!%%Otm`Yksd+F=Nr$1g
zZIi?$vrVn$Pzz(airPU?wbvBZ1~L(9jy=?_&Hck#4G>7WieYL;T#mGR@mbr&q??(T
zc(XR4ulV9a*_Cu|KShUe;f&r0CG9630F>jm!^5k9B1*03M7Qk1-kl+WH#$eL0fu5n
zQyp5Bq!#&h=mf-8UtDtfb~jkFKyVWi6N0t-f!Oe(WzsWy_nuEy3kZa%Ke`iJCz$N1
z@TTvV&3hz73q(AeThk!Oh+%P)q2=Uz@u*nPhvHtz{cQH~Lt3{0AK_*W6gGaW_b!DN
z#*fYZs;f~N69aQc;zaX}6l@mgt)7_{KS^$aU%cKv>q=1#6(IN%vH*{(wGdLkiPJ3%
zXIJ0XBpLdNYuEg}{=zS0pu{U_3l(?oopI~21|0UXfsMYDpY!Cx6^4MjJ(h~>wQuG$
zee@laQ32c<oFAm6%u?kC4JtJRL%2gMh<#^8izbYSs2M2Lu>08$RUchZrOd18aCaZZ
zTM1PgOa7VPeZOsbXua~Mp;bUl`$`2hp{;qdMzM0(k;~6hU27u32|#p!+9DQJpS<n|
zieP;mCPbWsK3Y8xQM(hgp$?HaA{e(;XZ3ZGycnoCzNm@@gZ{wOkx)a7Cf+~MqnXok
zHc!DpUIh|IeE`c<Wvi%1fRC$&_h$Su-vYxRR=*bUUN?39;30LrN@K>a)-CAwr0i-N
zHq{XN6ef=%qo(>cD{AX<gdP54_Nxam&r2cE%Sfw)Xd;tbMi#5CVQ6idkWk9PqnCVK
zzzG`r7|Q#@27ZGH%KdAk<TY><EdFg-+wS%MwC);Fk&%FnNHYVk70}-LHd@sw3fT9`
z&UIkhf0}aZ5EQ=Co?unKMqca?br-n;4{y8-5(}Na1g;8veIwVET-5Z1`quH`n{VnQ
zuIl>(^A?+=P5#g%ukl?AbCrVZ8|+$bX`b_78nOrd;8oq<^&j2~wY@BU?FvwxTLJ80
zPA_#rIJ*Y61|gkvJV-cP9Bhc>di#x~Is~GUW43<p6;t2aksJOBGt~Y^%2yDlYT8XN
zn*Wx*-!QmD&T{3lhezz5b*Yw4-k5b^#Ss^S(Gi<?+;S1+r#GiX^6!FUO?5$1&lSPc
z#!NkiumqWG%B7opaD|YxVW<quR26jGYBF^>W39ZUOd=L?<9f9Dv#G1h5IDYT0xThb
zP>G!E-Hp>5*izyZ_+Z~Xo+fAk%F5SkSeK1uTK;+JTUeaE0ecaE&{k=(!Cf#JFkkR4
z2C@ZVy$#4NqLcK6b>4DG27Bx1G<NXstHk1L5r5d#qyRUu!wtUd25PdPKfRYGShE~j
zw2=g<FIJB3hsu^r412DRXw<fDf2)eNNm(}WU?8-pk|7@kpn(N&6kntON;SzD5l{Z5
zzs^SCq8X!?wCM~^W)j>Wd1AjXe4=Op946{S7<og1Rg4=jrepC`$f{i7?L#9q4<6J9
zcTj_5xk~(^L><i~G^)AwylP(QN}`fS@~JVmz8Y~hz4A;-8?5Ik1aKG5<KbrRLu;@)
z9Z%v^A92e!ELR~z<)9{jfv7h7mmWlCFHE<afu1qv=$BV)=`G$+wD*xecKw+Kjz1Y}
zx`|S}LjGzJX#w1_nOYwzhvGy?+DpR_m0yS=k6$aRs1$+pH_vx_J@>5N?}wusj5ghZ
z1D4^5li_(PBHC?eIlw77YM=4II;{#vTv8I&_9C?OdVh;3UnX(9z|yuEMf!Nl&*J#=
zxj>0OaFrgeS*<0!(0j$bhVg{2r}00+;P}c+3<8+}4z*gntDUxbwtPFk?VcrJ?ByDC
zO(SP5*JmJKk~j&`=&-8i8gE{|5=fpwM`pavIde7?+y2#fQ6e@MY?FojXGl*vE)4#q
zP_oJ0ImC?d(~DWpGxyY`5|+YVTiBaIyuCz^ibjSYel#=v6SJKD!^1Gnh42DKXoj1O
zO%=4~?YvSJqSp?QC+|GzH0T4g)9m@()Rp*LD>u|fL^qTWDTf<w_yo#`($NL76?bc~
zkYAjQ?fvz3WX+-;eZ0eR<b@RQP8<Gnw?HsP{?!q=OWjNY+elUo-~+ZDiM0V9(02lo
zpVJ()5YEU80)jdJ-~R&0HoEC9eMx4C)~G{2tGkr~xeq?za4r~!*hezRGq^sA0PemH
zDy*mBMgWI#v;%1AFt>*2<`7a+8kn8S;tSHP&t-5quK}Gs#;w!t0p~$u{op=dKdETi
zm^E0<zpqUZ<_P*rSaEcLr?@BCmeaZY@85@4qWX1hZJd{O0N`9Wf9P&e8}M%o5UqHp
z#x#?}SFPqsEWD<PZ!p-e{j=Hj1`e4?ISS34V9Bg(IiQUEptH!1*Y429JA|16Q8ERO
zTQ8q`9m4)m@r3ZXw3MBktm5vWle3IwSsJ`Y(YF96nhX2&8j<#c`?Jo1-S|v#iGDCY
z!65{QT&AXPP|2jD$dhyXZ4+_NV5xrYgy$NmexkOlZpv8qPlNS9<yC<0W^kV5;-9=0
zhG)BUNvV>rymUMu)~YZ0RA?1f996sll_}tpAVe`vlaOR+0%OI%VMXWkZ}7{ZCcl3;
z0;SoKqG0huJF>stGIRW0+uf}7g*(=?YH3B`0!I!krf4b|k%@UZyW6jfBk6^eCDyac
zcqxDVT7xwsz7E$%C^APt@*`kWA$WEEe{Ec4R8-LxCR7?Eq(Malq*D=wl8}-Cln&`q
zT1r}^VUQFA2a%GYq=sf_q=wER9ZFzGXNC}Xm$%mY_x|0x)>-G=yZ624eEZw`+xuj>
z)iu4pBQUsgcPF5(*Z1+iv}9ZXj|5<2u6nnreW%k(=P&uxPuFfgn5^;bTq$Q3M&mgS
zE99{3$_ZO$?jBZApD85>4xWLb^)p>7uz6C3i7#0k{eM4QDJbvWKAy>L-BRD6JD2Mw
z567Q8F*jffyGxjD(*h!u7VkagC~6$kE5q(T&#upwgS4s&r^$bEVwBR;?3Q}NmHX6g
ztidbI4U5>LkHOk~7h$nga4&_aItUm^eGEiK&0c&hPN?lc+0SxfGBkheuIiX5adn{{
zIqV|*z#n-nW+I_89M)itw~KF90C(o*+D7+9+eb~)lOPIr<BjWe-e47;1sYdcG{gPg
zVn*JLHc#KkV0PWBzA4vWNkwK_<Hnfb&*|hOK5l3goFMkJ8+igTkO<%%#f-ag6h4i`
zjCfi?26Xqhi*2G8l}m0xT`sRkx{>}@ZZrW{i-L<_v>e2~XQ8^NGnCF%ljgW%vb2{Q
z^z1h4Ex$(d?c-C^(BjjY4pLgVTtuV3$t#pLDk4XG+_IE2?bf3`s2@de$PX&oH%k36
zH5`In^m{%tXS>}$p*%egDK%@}B!D^WL#b&kRGgib1{j+XhtfHxhq|iEhC7gE07G-5
z^X5XH?UK*4o$z{|&pqC{%_>~<LpHIZii2$c*-JcVn?TD|-KlEFep_J2rp~4Iv+6!0
zZA=}8atk+|Hy8sUEz_GC9u+4$3rs-sjE8uBUgBQbZ^C;>xy78em;yC=m)-8EKCAty
zy!HQv%gpNc5;yo*>>%N>pCf51-Z04X*Vp6_)5gAZn?MrfiVFH#C-3Q7(sLqp@L{|V
z$FIME`tx$%K_nFT(n1gcCSzo82`My71?7c}w%NKcpe>S5W7`NxpxeW5ug1<ji*uhR
zNyfEki@Tl*e_TYea58SnS($SL2V>o_^n>q@7GfE8TDy@rh_>V@Q&pst*W!EXs`whv
z9rV+!MSc$1-yj7(<S~Ye>+l9XFQyM{&6G~|urb!brcC|d&XZ$QF|PaleWRg1B{tjE
zX_^xkfK*KtpCgrVH!(5!l9KXW;qov^;36b+2RHEGbNfMzl~H?jP0sQ%b)f<4h8T{~
zCvi1Vkw7I4GsSqPZeUL)en2KZPG;61KMfP#R{v^JEqW}FIQR9O;mv*5`dK|!KuX)q
z?nyL6nw>q>c8rt-fB2e?0P3BD`1$qwY9K{I`=%48tOr*pS^P@F_6{Tzxj3bnQBKku
z|5FG}LNsG3Dy_qs9cDr*woc0<L>o2g`$gOGZ?ahAqm}TnFYsRJ)YZjD5n2|^->TZ7
zM9!KdG+f*dx%KFt*`Z0);OjlOd2%@Sjt?Gsd599wYWBk(X&V08m#C4Lo?2R#^DA#%
zUVe%=mUT=^_fzZbJkp8F>8$M@asc(rKQu0@nWJi9^9V|g8QQ{o>yrNCQ#O)iqc{qN
z6d#?2L@s=c30{jb#67e6fj14kd){!|?j5KYF@4x~D|wep`|eb|ksr~Oj2oQcHthw^
z`}YHy4zlj69>#u{`8w)Q8)M0vhW`j0NZu*>$enC894tLAYMR~GQB-VAd)fOUTYhqG
zergHhv0@v&VYu3rPkoo&$4Q<%FqD@&NdqF9yX$D-&o%)q(S=Lg+z;kv2T?nV>@?mJ
zEoN*xfRRNwZtd;{d94qm3jr$*4h+=7ajCi+AqmXg&A49g!D;5*N*oKgN+&g&GF_Rc
zmW_ej@7KppiC7~g%gvoksl`K{Ts{c@n5GI3P<HyyB_2logXZfaKKe8oUFtISVJyW5
zUK+?mx;Zz<DO(%dA`=-Fl1R@Ifa^5aWC-$=r`z7%mSjzI(a?yE>auPH4(4S*D{vxr
zl3e+V#bX`6>J3)Xkn|%kNDYFY$IJz`ZaS$oP~3~-ko$RY&`!i#{LHbkM8cAk21QzX
z)S*Xjf3iZ#zC(!w<-WQ(_@wiWh~k=ECF6`wC_Loi%qLmBTR>RYkhww7y8mIWpO8$*
zu&glOK7j`SwwKtqTOZhe_gx%d8o<qzS8_{JT@@-lQ?Wgb<{a?`Wo@45261M7_9<g}
z$srUoBO2>77i6G9-jMw6hBov{@>g3G;*)&u{1qu>__=wD3$hP#^u*R(t7F!Lb?>bA
zk7WB5XOE2-R(Tb$Gte^Q;9X$gdGJZ|V{Xc01oi7BvUoszX?hS=;V53<cQ^bsM;IpB
z>Z-4IXV;6wcWvPL{9xK+IU*Ddqv8)@x;8k6cxoKlqAUJZ0YJQ&f!i;aAEBtT&d*n3
zM(2~4&49O$r}scp|2Wn9lv)}N-Mlzj|NOemxhsl_16W^6Qxh;c|E#N71l@{7eKRD#
zgDQ<2ndb53Y2o6J?UBvlY@>rB+2Tzlc}Y|lUaYSB+}pr`!bo?j+Ln3by?Os6&q%%e
zLfW-W(idW1F`p!>FUbul*{f~Z;;NV=1w4%>rvl2_)<+KQ1Z%1kZ1nrLS6dD{5xI4T
zPiQdBoNK*ewZ#PL{EMYVlMC`~cLtC2jEst3ym&$Gzqq*gyQyM&auT!vNTS4ndh2Kj
zpPl>q&fUCla(RDFU$?M)1dj)#%%%D#>U`1WSGc-*?8lEDztRX*LAo^;B`B%y&=G;$
zMj~NTD%ZQ6vBab4AE34uarH4|i%j51jS1@k_EbK6jV#IEB(*7Reuyt~wq~cMr!lGQ
zmnvQr(_ur&V|{b5j-{RG4P*-`^6FMj;0^(J9`WADXr{;!8nA(rdNa))ufk=jr?;A(
zp6)EjIMLK3gB@%){ovV&%+G?DrxWaRW}9ofSx#npD4mkA6Vf_rzjJyPR|mtz{VH9H
z!?I)wna^xMd)PAJ>8E@rX|Gg@2uJAHZ-ME}lU;v^ZKj}<C=)Io2%%YewYhs=ml004
z9WeSWpMk&OAst)7P(L}JEBW2v$=1}D^$b2aSjK}nv7f~RTlc|n8N!#*HP6|SA4!s{
z2M4$IqCJ6h!U2J)si|?TlitO~aCT^$|KV?Jvie$6n*>+-&y~v6q_&&SLkOF5BR81x
zyko{UoM~j_^1Q?9K?%Qd6^CE9KLjiKljq2FzPO#cOsR0nxEpYPwg|$Rm(S~+*B*U2
zI+8yAM|5+r7b&42)gF)2n<6fG<$^t0AFQvft1C?HFJOxAX{oLjPgKh{1solEun3Rp
zii_Yl>Z)mrKN=xG$XUt=$J%>%wKjcvsCF))2k^rPI9{%OC=8t){VN5OjyQ6mHOUOH
z5aa?vRMVB2HFwFQB3O}{XL+{R&T!=K_p~gd6S^vzs%zr3jpi!7aM(!G<{t5)@rs_@
zs7TeNLg>?!)Qli+4fj))Khg)&y89gN3`;_Xal#G$Q**<`s0;#0&ZNNOjp1V;0B-@L
zup)G^{#u`fUF8Ad$nf!#Cs(mMP3~6jTiLj8jE!_w_pNR)Hf&HCpK$Og(B$aVejif$
z=9Nc_T4zhtO^W(19-}~G)JB?4L2kvgamkO2jtnK4dE7p5@$;`<CN7IT&u~f^gTdf|
z{M|*aX|2<kmaOlt$AjdUYfAxXK@nh66l6S(oae`idhm?Z`?g-36nx_W+jf4BqyMo@
zrnXu@MZ(g5g@_UCXHLN+Bl2Jo4hymB4D|CzH$5Bz_T}ZW?8`}8lE%RF^o6;%B%)7p
zrQ_6De#S417u+RcI$La1v7_p^C<}agde2J~uP~cn8ou+Y&YdDPGhzK($h4<Hu*5Oc
zMmsrmx41X@hGUSUH|_07s8>2nKBZBO>OdrO{!gD+q)}TwjPNT9`7gQp3T12sQNm=n
z2Qdvu;moOnHOFrdr@${xI9+w~W?Q}O26(@syqqIqo4ptk69cOT+DbXQxTMGh;-oDs
zEXK#iRvRh=T3cI@wayc3=|+}~*U|oRf(qBn?FBNL$p~EQr?W=|OX2y554N%|W=*3T
zf<3>z<Izs)TFAEEV5*K(Se%<XwUhZO&rBW(dzPmgiIz?kT{ao<Srj&q%g`xKYN$W~
z|E%X`7ze#24ZA-TF6b_G+f9yh7B7g%{hCYj5a{8T5mWLpg{b~b=D-L&;M_nhEnWTv
zv~&mnjLgG<G^|LLwhNG`F)}wV8*`eBSIcj5p5VD^w2gd!!(RdP28)bq-fLQqknJrz
zUaD0W0SjK<UWwk!4Gqy<!>46^vi1<Q2R_m}+Fc2kH)7Y=GF{DVd(<t$dBd()dVGZY
z7$o~Q0%2#}y1enfrUEL?rDy}#C02hBLpwkAnTchs?SzRWb}p>~5W>$JU+x|zPn;t)
zpcj&`tvTK5;bQ$;rX@ZTTaO5CSXTvOovrpBB|gzqIZ0j#KL21)Q!gSzWuY1!4}-<1
zb0yznjCTe$2&SaI!h!;oBo5h)+S*ztJ_fH0Au}BP$nm8+a9lXP4mn+~1IUQBhwRj^
z6|eToq;((9v}`QeKnk1lHf<mL=A~Bj6Zi{fJM@hkns@s{`6JW);~1G>*?5L#VVn1y
z{pZ%n`FR)t1ih7~9Q^~vD*YS}s_AWUo+z3TZv8tRN)g@lnZRMe<h7(oW#8nMd>e%h
zcVaN~O?RVHoJN7bS7(*R2Jze%bGhg~;A}8U>0dkcTkz}?SqeGFEa~!1er~z`Wl~b;
zgaia8qM~k9=JL8mzW*ql>i@SKO@UCe&QA+y{Nc3=e|xHJT!s9ZWnM*Xqvb#`@u!Hy
zpR~a}?S5%xO0%#p`-J>E>3<BfV7L99$aeco@iK2wjMPZC-K6_X!^F<%<eI(48DsAW
zL7k6o&)0PBfww)V5uK@lCRYogOyfBeh6YavHrwzXgGQ777fg&^|3L2|-s-<lsNKTN
zDHYoAa<g=(8EA0RspJQn)qxZ>wY2I=O9u%XFmP1LpLb_^BgdSmCZC;ujozD=fUN!x
zld7GPRpOOy7452ql3YMkL49%YX1d=}z-Akpdy8i-TG|vKZ2Il=OG`^jDv&RD4|~ow
zvfY;uZ|ACa0K)$SdvtUptGZS|sB@!e=yBy(K6JCtd?vWHt`3MFTk%UX*RJL4^OR&D
zch=KO2JjbEbXRb!-I&uWpt*rmvijddPySo~I_MM@pg|X1Q_E1lle@f;<4V-JN?8((
zA|_r#DLK4vnL*w<VmvoMa$9(oCeix`5vNPKVWKBi1!6)^Ln%SjE%e-z%8?xss}1;H
z<?nICD^aUDYAS1ww6n(FLJ2R)C*YGz7`{=0e_5~)ducQiCGBd3nLJ4lt#O=(Fw8Nm
zK#%&tTV;RxnZ+T$CpV`rge4IHyNXg3ei8hGs>Y<+&-r>L#@2nM$756I?3nx$rK_<Q
zEHxqeX0lESf0s#ARMffY07*J`$t%tI`T2s<QY~>7GypODp}c=;oeh9^<_`!v`Lc!)
z0A-DU$qH9JP9z}coDn_&tr8b4KR<lwKH}<fLcp+Wc8}s4S~{4B6zJShb`4ba4R9FZ
z16OUH)T6ajaK`exi5GP*i?kc1*r`Pv*2W}7fB#F~Y1;So@oD|4Ss2m+fKWXA?cDe}
zili5PSG(Z+VX8@<yehD@)GZvh+>Nmo8GX8@k>7pXf;&CleBHtP;-?z_bb9-pwW#ke
zVKPqRI%@t#pUxjjm%#$q++%(fQgS|aC5BV;IApc<8l|3FZ4p?G)NZP%Fs5XObE=3G
zv7-9gmA*-Rd*}$K%Dwf$)aX*(Aj!@Ft7OsS3)@Co&#t|8eB8llQw(#vlbn^Q(E}X{
zHYZ&uf}wx%&L3%XyzWEt&U@%HTg;?qWH6S%V05BT%w)++`ze^gx{-y?RKPVL`jC>6
z(#Rz)z(D@>4tXT9y)58)nA^y)3}{4lw`p$)3WZ{TTGsP9fP`wW1^Thc%@J7u7Q9x|
za=HKs9P%PsH8ZnpAh@NA?;+x41V9z?K@v#6&aOWN42FVEf9vwSLQ5mx^?CA#X>k26
zarwL%;0oR3I0Wr00uEY&!{J|1K+Px|5Ri6Zu~@k=r?g9=ui89!M!$UdBF`qV>W0lO
zGERn5_QrJ2ANAjp@cfdP^$ue&wS4tBC1c`)$h_ifB34CsJymb&FzFKshoRd9%g0fv
ztAFn4moS7G8W@Z%E4w;7_tPpO!eYR_>y;{8_YeKe30PTKxz!s0g!9V4z;kQ#{`ax1
zZ>SpfzCUaVs6aP2x8C~7k`fK{VFS^L$&!`oyf~=L*i10vC68GxVURo9Jrw%ZgC&CD
z;bDCXi$&h)e!CPF32nBpgnNTPyO}eUDg+Z%QNOCKZT|H1H0}-!OQ#`SY&QX0P1i|V
zFw1IDQ<JIn%EwMmb}R@SWQ|L5L>U?yPV<feHsGd<m1iqgH;anD^CrGWU-us)<kD9}
z^gC(qf0vvi+N%k~ySccycwK%K`T(>)G&M8R%221-%;WhaTXoR276#yNN|S@``f`MY
zRa1I6ty9?9*%{$tb(y-ETHN)d;tqeA-0nSZSuF@74t_JBl_AI?sslVLfH_wR2yiqH
z94n+)33r=)byeEyAK5DBS^&;IxsL&M>EDOa7rfM$jUh3z$1Jna<@S@tRd8cL^fS1}
zxmib0^2cy4YPC=Fu^&$~jO8P#AK+=w=L+Yj;%e6;3^{9YsEO{5oEdrS^;3A69;&i?
zMYb#l-pY}f=3<ttItbl!Hw$ep6E*9l6PRX<@3E=U_Tce)Ky!U%EcoA*7~}0bZF}ag
y?=#3C*Yg0EJko|`Lf7N7R{l5W!~Yp{MN^$-k2}tOGXk`RCD2gSQK@)l_2EB5eKRfq

diff --git a/docs/assets/images/use_philosophy.png b/docs/assets/images/use_philosophy.png
new file mode 100644
index 0000000000000000000000000000000000000000..23922d4e9476dcd0d3f457828807271ff58b1b01
GIT binary patch
literal 16463
zcmeHvXH=8V_U=m&P(VbQf{02JsY>q%0s;ckOXwX!@4bkEN>M;Sq<1MvAdt`rih_VN
zsR<<@ozQ#d#-B&e`G33j%UyS^d-yP{@aCP_v-h6;?3vjQubyfsQ(j`c1ONb~%Hv1c
z06-WH00dXbNbrAgiLu$k|04F1SJ5TI{{)lSyu^Pd@X}V62cSJSmH^-upz`RUuD{vp
zgrB#`Zsz_btd!jd&QrJlnyHo{mZ^v>x8|ujXGhd?8yYgSbb8W+>|RGl)&+Ta&s*vC
zCAq25R)HSl4Ib-F21anba%ey^mb)!hi8W{~)-G5%5bNcc@}kZ!bo;2g>F7I*<02^`
z*{yKkS0h(;^|6y)b_!L@KOcWJw}|DMiT^(5rnv<m;g=R>?)-iMAjCf=`tRfaQ?=Z`
z7~UF1sfMlfu%>JLqdtHzf&9xo9bq-#hWIr6p|(NLE6qfXn1i-gn)RxQvx;XP;xq)r
zSLA?Sjhvuff?a9r9To1xFEvjSb?jh9B4BX!1FZ`!O9RElH~EuU93pP^JyU;)6v%k<
z*PQMg1XwtmN}vosC@WpiUPHY=pPIC#Kfl<wR9*ehK)nd{dZ$~Q>hCbO2qV}I6;#z%
z!(hh3=Z3`g;Cw6MXv7I!GKcg$4Yrj!qlfF?@@rjkP@7`ehKHPARZ+8eMBx7AKrQs1
z40b>(C4GSUx8Qi`^Qa%I20E;6p?hnap5p*9=aU_1i42xB_RveM8~5jX5%csLyk#n?
z^`SFc;u2bHC#^ns#kd~53AuAla|YT_{i6yX!tkQQ`ylu3*WzJbULhm+VQZJ%7MaH9
z=hgjr|Lxr0i6>1Ota?^mp=<YmcxFivhy<5?T=S==qHdqTe?}!Bw&JKzA(|2pHSnVY
zT!fbg%X^YtC&NendOX<|eL13>vL_46U-i9L35Rco-}tu%%&2bxfrT7*|E{SkoCJgx
z894z=f23{U#;+OJAo1K^b^K(L17c}}+#>&2$zL%XSWNzq^S58Q3}Uiyk7x5P|1v&6
z03*8sG-=bkBl=^hmq`GDfJR><<)8UZ(Gd{l7&txi{MXtB2tBUe0@^NqsCoF01L2#-
zJ95B|s`}H*e`LF*K}g&p_HbL|&xFDa@B^EbT<nbaBik31aNruJ!z9U{fj3?Q1by7X
z55@n;79n==3NS<5J@$9t|1G6JM1UgpR}?DLOKCro_KtdeGd5nWz|<4F5qE};dKb1H
z4|q~8>9b33l@^BRn}!@}uy;k|3891UnwBCm7Iq%*s`$KdD(QwL*CTi5)cDj=-hHN3
zk$lUc?G@kUr{Ur6^?Q<yynV}rQyaPo*NyJ9CgKY2u<K!pidB?KIUY*ib8;#yFExJP
z1h%m@g*W=D+L)>bbG!TZs@BbXzIq)4k-rDlVdG|1c`$%$)eT72)X+O9P*;UGOT4=c
z<L^%JIUf6h2p<Y5EeQ6k({&Wr0qd}f3o0i5z?}x`F_LtF3pomvejrTJ$OYM5Gy_e#
z+S70Uj-xH;3&!fecg?*HSXGjZg;S#U_l_U>2p$!U%M+@41`7kjNFD8`o~p=^gV{~(
zY}(<kNJVax3q&f=Hi2T%ggw(!hn-(kF)``G-RxdYMNaVCXa^(OL&!AiKuGga*Rq0P
zd3=3_F;^2**QtK^v9VcH!JOS=ZBNf{HQPf&D!Uz*O883O4z`vgN(+J#)vQ3xh9R!d
z-zXrbM(aRF&}V!an!cJuK^YRV?P)SJvzB+Yq)*_W;YY!H1c`AO)(y(__Is?*(ug`a
z?&qk_N2N~g(z$&WpI9Ss4&sOo>#hAmP>{Vz8N^UXQzG${LvL>G$J&I6F_CG<rqyrL
zSaJ6KB+7Y#!rG3gES&5mCZ|y+%$96j5;yjj4+VV~U%B1)85=g~{-oa9nbid$DV&)9
zq`~wJt!`&nBen#sfPXcqCiCrgja4>bZur{#2c2Z?&~ightg|LrY{4^e(tlhw!j9gd
zpP;Wk<Q$_>01Ac9DqTA}ndB5EzIjiIOO`9e4tE*piY;E3;hB&uR6Abz_Bk6B%&7{V
zo2tAQus8?p9%go)XjYWvdH0=9JTZ~R=wLe)hJ2(jm2kBqNF;l0%C7lU-HqR?(Dj0`
zH*msF8OAzhtNO;oZ=zZK_DwOPJK)z{b66o_k@n5wbaERTti%$!!?eatXT9m%Fs>Y(
z$%6_gw)Udce7!Ekr9!05))dp=iY=OKTz;H(;5(?=VBnQr7;HlPuO)bg#-S2Fq0sYu
z(?p8pVXg;h96gx8#rG%nmHf+43*6D^s!zM^<Lqzsh3Q96)Ax11t=sv&tmhuB4^?yJ
zOF*7(rw_rL?Q>@vlN1cxt4clZ|F*~Y#p+SuK+<I=Jp-%1Rh3@6+u3VuXv{%Q-Zfjp
zzLpbdej4U_?`=mUu0Y{<-tlzX{QTe*qfsrp(b?(5qA`Kmni!X7XlcR1(g*Qwe_gsI
zj0BSt+k%fVL3%>6+HcRFQq55$kps;hW{yiNO+eI=)S7CwFZoyUQdjNwQP&;Xcz%gd
z9J|x&b~+orvzmsU)#j&8)l7YfZbF&3o`T<eV~816%p$)j&Vg={5ULwgE10Wdw^H3_
zPL0=e>f(t}RPOv1@H?74V(+^{lrwNTCSye6^9o8=&V`03Uz(NG98?^qbe=$+hlb=*
ztT9h~d0o_5W9wFUl1;V6x<-Or!@@zT>{aGwy5`U#4L+<Mn+vw5h)Im2-h}qIM^%wm
ze*^q|n6gjUWEBFJR<MRQj=e=T242u@IojI9D3}Z9>E3KC*FEx%>*l*2blIRo(A>7m
z9tUSVI-s-mE`C#1TA(hPsYnx<zxA}TAS{FLdf5YOcOSl2|BAZyG!6*_8}Ss`QloW{
z>fTp8KqlqtBb5><uj(h0<u7@F$xzEmDgS`FS;T;}m$f@Bok5qNxm_oy>khelanU=K
zZoT?B6W8td?47W|&LrE!BacKI-q`M4Yq?JL-^k~ng%AB?rAfdw0^=4g;+Dk|F{a<M
zbGTq^0~C<I=bLhHqv%oo2Rw4ve@*!t+ZB$8>e=*^-l%^Bm8&M?eE`5ID7so61zEex
z{9XjSM!!s4CP4oz-u2=Q`e&{kHPYFXzxvKWkmEh<o4$KL<UYQIoc6aY{L{Zf2O+Nr
zU|e{b3VrqfZd$bSZ75{=!?c(?{LwINY8Q%5wDs%GRTQdDPH7L4l$H6xUb6PI-N~kV
z&E~LMvLA7FC~LZz$;s|0_qxE(F>B7aH2&QyH)Tk5kgy0&P(%~rPWRizjhp3SE{7j>
z$Q+hMwYb*b@^-aT(nSQo8|5!A5%$UUZx|RdppU{nUHIk(;?~557}oo#rtYD+w362!
zIYMqo>9bQcjq6?N%4<kL-SQDfXz#SjcdR6`a*6jN?hdEusviCm70eUWJ`@LQMR+{;
ztrCb~jv}dCnE~$7nW=B4caIyt5p`6H4&zxoeb!rS3WsvzDM$x<e~f-j)=iAVR05fR
zOiQ?s58pBt*M7=%z8PBlvL*~%8azxFyoIDmM3-b1*eTKzf;AiuB=vdf+>rLiHTowN
z)A_8w;Kg4#=PRjP7}<#nQu6^EUf`D1ur@WJ$gb_N{O$xQRDy2JJ_lRe-Vhz3eQc-n
zVFx!4mGmFUW#oc=Ou>c?R^p$dg6Rok+$6Y`($W$qTTT7khaw)7+~ZI6U7x>t*64-R
zx%x|=5&C?6_Xu-_Dt$=LB_vfyia1o-Gk6?fmIC79-UhCwnVK37i`;j`*~h@aTmB_5
zbbUzrQL~ZXM!o@kM|h9k4=z!?El(Xb@sd|thr!z3aQ*nq@&sn(fw@<j+aDlfH6>d{
ze(p;?h1vrMr5{@l*my-15)+dYn#T<cIEOg73+g}0Yf}}g?tOQz4*?xUM!oqb@txw0
z(5C@3YmkmaW3fC5mK{FY35oYy1_szz5zRP6UXgxhX(@X?Qb}Z(y$f8-W(NsX51}j<
z-iRkxvvH`oR8p><P#(+H@dQToY)(%`TrW3hr)aRVNa6a{(T|Mxh$CS=889VJwh5n}
zanB$|S7q20SFjT(Ypw?wboBF9SB)!JcVoqazT@up@TrF5??lO+*w_@<8EN39mwx1>
z)=QgReO#0a8RdkdL{vmZUhCpyv%6T;yMzls$vfZ5Qhd{VG_+oKCC$lceQ=g}$2Vl$
z;}uwNR8X&s7@P+$c9rs*Z<W}na6L(=CUU$ZyigmwKpQRkd$qQV5#qC}zy{;H%zv`O
zE0XxMbj7*uUk*ujYf1&5E^?N=kpEXk0uWZx;M2vP6z+d~=vOY6MT^gE;l_7q|J%ub
zb@E?7`M<PwMDn;fiwnNJ$GN_Hm{j~YU{fXS_~cx5`s|e2w5%jaVsD9CXFWjff!COD
zV8`$aa+|sD`vp+s@NUrF3R>Z2z%2C=22s|y)yciJWS<=agZAwDh<ySPs;Jy?Tq@Bv
z)nk>(uIU{9pq7LwiW?jneB=!a?C8!4=qM@C)-ejuh)GUNnQshp@!1vUZLbU64m~~Y
zXhj7ZZf$MNq8GA~1H2ZNLVDxEpg4B=bxk1lg%j<_lR@D9hYxpZCOykc{7!w~UJHBy
zTOG2GAAZ`bU^}|Je0QYh8h6PO*Vf?}*k-1b=tr|c%D6w)e`%CmIi+{(_O{GKUf3d8
zl4XR*Kn%~p?h5uHhX%`C+rs3np<`TODV$@aOb2GEQx|C2z-MYRy3<u)TC4#-h#M&=
zn0jBf-Lz#zZW5vtpYl+rsACdQW;>7(mXX8~dtDs;z~P7PQotv7qtgy$Pxk>O*p<vA
z&(W%ry)m~`FtkqJ*BQ$fb~3=y8k&-wM7Ui<O@0MnEsE)WwMiEKiM@9nt-}{OrJRW7
zGq~eIAm!E<&;x7RTlJcoGcZf(l{vn%5-tAlwg|S+0fvz|ProuLtc)y|zMI|DlmcAi
zb$Yh|D#LyOt-qaQRaV(dV12Ov!k$J1@MV(H<-i8i6xiBz*7wRLQ=T8(Td`Mq7h9~U
zsj88ZJHVfuU|_~OL`Mfd*fp4KX`}4+Iegv}kD{y4DZaclpYGQ2J&i@hH1pUj_xS|}
z_6wXqV0j+ReW0jDCG~i%&VcW{XAU0Jdoy6`mFzdm(;3>ehX94ay=E9I&^98O;l~O@
z<<p*l$-;9XKM+29@z(`sOUt0|P9=z!Hrsd8LJyuER5U)gjXw;ySyuFjBAkeX;;s48
zDC*=gv&}py%Qui4id|TE`rMce2wYJ+Q1>7UpDbe_wAdcLNKoO}JI!En3w1EZ&!zx!
zFc02TOxqpeX(JR_A6d^Zj|2RB%5p3TuJQsHq8$tvnsD83oX@tnbFQQ~D-i)6sn=!)
z>VLi-i@6x?k$?F(jx5|#=f$o(p~b+HfWv^TZ>>~aQ=cMjzrKdeEfr^{z~S%;lK?F%
zvg!li+O?0)Ih&J1W!_P%x*fQFwM$fMsgUjY$Df*|du~ll<}0ZMrc<D6EHj-mc9Y-H
zM`eBK6M;2l`1R8+J=l<{BtVxexfE3A$l2r8l9{_wIulmMdJfdMRFp^jt_wZxNf$Gc
z*jeLE6JIUZIq(pNTQ@j^s^v&b{Iv|L!d>Xs!!I(Z81O2)ra}`Zkr1!UJEtbcLeE!6
zdZ~M2kq&JDvb!fG`+Vj0bwRsWXKb^=S<VEiJ>4C=obpl#_%QQDjkI%i{c<8ryJp96
znECXRo7V#@S(b(c4KOxE9h>>Ka4=5GsLmqwT}HXX7IY^%^Kzz_O@}P9GtX#@U65P7
zO=F2uLD?rYW1Oc13Qfo-)#!{$Cbbz@CX(PrdT`roxJJd+e}wJa?0YD&JX$wcO<U|7
zy`TNor)#gHIrBoC-4EZTeht_9uxAB}U{s)6ruW*gMpWF=QgAd9D$T|{zy9t-1O4Fa
z=h=@Y&1PC=?&P_nbtF`B+`{UKSv%5mrfG%^<%Q%qb4o7<RhlblF0-DJ`uU!jel2kl
zZ(M&~Zd!yL=vHP@FI~!^amA0*_+C*=8J{Q4hr(5whvCmZuw8R(irSvs3{@}KZ&dew
zzi|pG#MW+Vdp63=OP`l#NO{&i+~j?3;(q8i_iQ~_v$U(rcDjteE0;GtW2C^|w&R`T
zdNTA#S!~?8w-jc(B#F7iB2b20IAr*y=r_)rIGe;S;W0bdLG!?n)7*P`#2k)UP%+ZV
zrCph12ho2|ojKl9{dj3r$|C1U)3FS%4hSuF%x$#))jrPdRt;{@sA%Gx$Z@1DX2rD1
zIHhs-XvuT5_Uo&f?!E=vKtq<otwAoMolZ4R2V6>p8B0&h%KE35>!u|$XM9NoHkos@
zZ5}`S9;WU8^y;Y^im#5Pu)kpkqh`%aW}HXuJfmGfGjFNx#WG6mT3z?qML6AqFVCdG
zeIA~c8>95%Ke)W#EnG2a)+JfOCPNjX9z&1xmKOL&Yczd%&GUM<cI`7)Ge?@z&sO_T
zTp>q~p=N?y>_)HOFeiO(3bNTB;htUOYtU@X?gme^_)P{Bn|RF8dFCxX*H{VCMmkS0
zMqT4qp^ea|16HHpb0mZ;a&+v4p4#mq-Z#abuGsCekczk@vb#0kOD-(L4i}#7-&Bb1
zsAY0XK}smf3~H5iJ#g-ST4(0A__T|0l*pm4fJ<WWfKf#SX6}1y<F1UaE~yX8$o_TA
z&I1ACcqma%(#4u_4m4&JGiaG@dD!6AQIeM@7|N1=F<iv(U<4JYE8)9**K7W}PYUJ7
z(sIw;(>FqQiAh>?3iaT&-M6oMSn_jvUtqZqtu~uKqTf;3`t^1f{U`?E+3!ke5Irj?
zdO7tnV+P}pRzFw3{tG%vsI7TsXrx4!pKHAwX70u1rwCymxbxa&BZ;7($c;s+a2GGk
zV<VM9o0=|%jg%liS_j~Je;9Ut2UUZv72&eR)khYe>mL*4To6yMPRLB>l0FhT%fdJ>
zD^?*p;Np^BI%j3c!f!V-YE5pNk%xCkv<`$8?A;Lgelhfu@?6OSBj~ipx`fMDs!Qa=
zEpCj&JCgS;#)+@{4RrTllMvY<<q={#6!RjCQ~V!1QoUIod_N?pnEJX<_?W1WIl?jk
zixKXDn=l{mfOR>MY$HmUx2$jXpEVP+?1|b=eTms|>qIbWphA?0n|q*@K4(%!$8+|U
z{596SBaQ<_oo_;SSXKH34wt!M9qrknjQg)7pl9zltPBVt_ka(vh#uxXa~;zjrrGg#
zd)$4l@h)ogDhX#>tAchJ-b_M<_YfsPpSbxu6I{kLBIoy)A-IK!;%R=YcstZf7qseB
zTFO=f5?zwCE4Q&(Sj3nh;70>gwlS0LpKw#??=>n%at`?rpS_z62GF-Oq*SV<XQeAL
zHXpr*_K~fw%|*imD}4=bp41(i72ohZjZ7DH#7&Eo{J`1zt?)?iOlOTWMSEp3AC+z4
z_Oa44>MZI&#sgdU$w51n>W4GS00sM)5CN|u9O<KxtX13P<Vbd2UOoB{grct-ulwho
zuLe{DQv>?LV71}puR)Bbfs_>Cd%b%&WT`ew3im!(ZA8VifzRlOa*5}3H*62<=TtVr
z=>|#gH2uMLKt>VXfV`26Wc*yt-eX)B&wMXsMp>1OmD{wuTo`lp+6{pB%Y=Kc80U#m
z3kEq|u`;2x*od@daOr0nQ^45OBw&nPgpd96%S~#JcNR8D<;X*aRcq{)Zo0UXC--9#
z9di5PXOz`!(JRb9_+)k3o-A{xxePF;@?$||>sl;735jx&C5=#Hr2DiT^psWfZ+X9G
z$s&u9c6MS_0kQ+^emsjJLwB#ZFt``&E-zoCpJ7lA>Q!a}B4y1AOk2+A<;cl2xZj?w
zx3ukg8iGbvj-7hn$>g_-$;R`3<Ox3fs)T!^^y}MdcX|nX`q}-9CiaPn7rvg6@c19H
zm)zxb+;LeM3=4nKgfr`K7}4;B7(!*gVSUfci1Ei6G(k#6YCmNtvB`nFU`0fVH7N5F
zBTc!UU+;1<Wn^8}K-h{uXRPrA`pJ9~Zur|~6XtbU7!wAEEi`LY#@~1$i;VTdP7<Iy
z0<tUi)sVG~MOM^R+~xQk($H8pEk2ePa%Gpo=Qv|*&#>8kuJ9If^N234QKB$R2-Rby
zu^UE*(?M{be))x{D$>LEU|8yQ^=^;raQC)d7e-UoaU<qC;hbVk?^_WJZ=gJgZUME^
z_|Ac*@w{T~29B^Y+QUb&kSAzC9bIKNw8VrtJAR(-mORO-BwqadQuy8na<^tk^YKI3
zUikTb3KWFMeJBQZcbw4JV+^{>8$7qJvEj0e2W<|{;&1Ar6SqXM!Bt2Hy)MD9(=r~j
zMzPnh{05y!B8#}FBG@E|ABURslNc<hQAzHCqU89ljp#&bLxMM<zUyE)ip$awmhHs>
zr5Y?*YKRxkFQ5+%jW$<dt~mX*;fAIglR@(LI&loK>gwt#OCuBmYs1YN(mdG>I`qMn
zg%5GY*iV_W%VSDv`o8ss-~t~52D-sF#TO~U%c^%hdX$+ma~9Vpk#N3);Ftlc)Qm(n
z26N%Tr4KR%w*KeqU4_-PGJPZ@a>wV8xsCi012v<~8dGc&sqb5`U989FkK0&t)4B3Z
zhUJ$cKz*6|dQ-jlS^$is1tKwMv#TtqNkv`hy7)<_SX`QBIImi-;h90$mSyB^Hhhdj
z<#S=SL{-mk8oFsD54fBs-kAJRQ6^CV#zy_*3BkpY^d7gM9{SpnCk~&w^@f~V#!(MH
zbD$*e2^+Bw2<WJ3x>hGa?AR&2FZpm#9-Am?4i@^76H>GGKqYm|5sOHtbz+S^rQ{jl
zn{bcs$&~EMM}_JhIn``&tZD_~Hy@oHN!S$r4Dbfy`laepz_5VVW!t=zrOPrVOr!2N
zbg3NGx6<|jhVq9rh)SX&pA6mS@vSO9-|{j#*E<dL_03mrSdX?soj6mxiMx1_!cYHe
zM3}(J$nXeDBj;GE!>uC@mr<fY<Fyq#Bu5w%i#2ML@C}rRv}3km%|qViBVc&0cE(qk
z0x(5dn-b;vt}c$&WIq`oRY=dsAjxJOG`Rl*cn?L1tv8&@)SbY)3Tn6S?9&4P%j&OM
zfLg<>lNlD<nWzKaJjNS<pR2?@7F6Z3x4?FoK8~bmZOucYED!NL<Y<1GB)7+G<QCu*
z0x_*OINO|#8i0W9>tFEtqp!(;Av}F4pZCp=M8*$$$thk5V^$^981&3)f??Z38>B5f
zl(t^7!}op2eI<%1nJ09b{kF*#j<*jb!I>V>zOk9(nta18j@ATNa(`G0;J_Q4y|^$A
z%@!3{G2TC54_x~gb0h`{ALebg)>&i^G7bWq<AWw+2r3@6bMqe@dr9cmPQz~+)vXFY
z884oA8{)Gvs+9)otxY265~l{3WG>X>+;dcpcr?1cz3bk#@A{xA<GR6vtei;N4R8yy
zzO=ud6o9#PMeGKr0ANg>eJVX#7REs6W_WxS+jPcJ95cd5nbNrVdIf~);d*&tcborM
z<8)N(R1!WLXOLO>Ebg)kyVQDKM}6xcrN;W;&k^&>r}bB;>TaAKTjCnb1Wtm*-KWC&
zNVlKg7)0gY;8n=cyC8S`!sb)=Grf8Zm+XSbyaupe3Gay`c{qk&9PC<Cbb4ael%X$k
zu0Ek=@G9HCJu|T==}|ez=WE0I^9&1Wu|711cX`HPXb5*MCCRXPOg^mX;kgkh5f0{d
zZZXvYOOzHWnhZi2hLc|o_icUCxa@TO!Rpl#9xjj24(72Y$yw3+YVCK;g3qZkp%&=}
z@Yz+-4=W_`LV%qkcvds)h$c<I@)<2Xedo*@>D!XNYcIz#sjmSzP=^SZ+h``DM<i_q
z)6k}x9l@Ll9o#oF39aAlHXm)Nu%0LsG&j3_>`NE5YeYh^ng?+biMTtIwzxS&ipv~s
zo3ifPaP^#nux5@JsbQVLO9KM<lqk(*>>E!eF?=uGLTuxN&vSP*KeNlQVxoEaWk{UO
zmxD>qw0KP(Za%j@I_7Ht?@P==^~Jjp(V4O~HdJ*#Q>YX3cvHzy;~Z+<QWhUSO+%ys
zrI23kMh2oB^(Ze6dxsn{sg=VA%cWr3jp~?PBdwvkcdr0Ull>6{u*H_pw6ZV|AyJm#
zxH4$+hwv>7A17G$yx^f61x2_<{%3N`$19CxF_RS`%5QuELlS*nKC#GuE_~I{IzpJy
z5SDT(U165{G<u*S2aGT_8eV+=Rb9mHe(c5D5xxHD<MzD=nUc7$98b39Um$ZK$ae{b
zVaVX{PVcbgbsWw^hMgMQtFa%`%@J1|3xzLPm@u{0Eay{YthGeYaCV5DvfBxk^;A1G
zod#z4!hCPz-cubH+MzmE>U&!C8VpFxrA*tyb(dRrg%?61sle!lr<{H}*M*C;fa*Rc
z+qL^E!3J2VbS(!{a*>N1gVc_Iw;gbPV*kKy$HBfVr~PKo6JJs4$EI{*-x)|r<jA$N
z8|&KAPWry-f|{DW@_mT*J=;Dv!KT;bp4O66NQzjG7;p9nfk)48Mnrr+Lu2)rmZzOe
z+eMef9x|ENx|~`TzvwZZ)*zK*wmv8%Y1!poZm1xo7_bOvcUZ`l4Rk3KjOw3km5Ek=
zDz2x`zAl4$62E*+Q4*hh#O0rCAYhsIUwl%Ag&u0eeDqPy=Bhw010VbPC%5O<5WxG=
z7$oUgh0DRrtqydG%Z~8KP_N<XMhH4j)`}=dATRC0jT7l7v-iXIjLKvDM%hz4Uvfiz
z8Hvh|hp^8Wt`%{sb!G7VoU#;)AlTSILV9Y|(hqM0k2=xlc;Cpq(A?5S3X1p~I=kS#
zLOV~RQ>NWc;`TvMt#z^3K2y(`p}P;2L?v@zohIa%m?7rz*=Q~zt;NzS=<EeV5`Rlp
zTy>uB)6N%)6JB8OumE=4V(d&?=*N|CFzbd0eW$zQc0`T}C(qy|2Rewd(fUM}Cod-s
ze09d&Qx|ic=c)9ZyNNbT?0izTND6Z&&+!3iHZ2g?FOi70r`_c9&3s9e6EA#{#z!FN
zyJBd+l!~{TiSLq3%ccZgbLwXs(FICH60WXCLknH+xk57KcBXOMw)-hj$HgR?gojTX
zl(<pXXZ(tKts}l;*}>+~Xu*-LWp)wE!c8M2&v*5Dln0cr%8Eu^CeXB@3?Zd{q2)|L
z5_hx1I%7S<XJ~EEi;&3Tkc%>$pC^fxl#x^}eSEV-!~Y_}&uw9?ZtIk^O|mLZ6ehmt
z%^w|T3ceiPv&H;tcS0F1;{C<$9f{`CD*jaUyVt_Dd5<E6^@bO(*^`r>jr0?a9%LbH
z8;h2oGNo%Z>DU|ahrycdzXe(pO_WKi)OXM0H`})TV1CJ@Gr~75z_n2_>JvWZz`Y-Z
zKHg;-nct>)AltqnI7plU$}pPS2&PvX9d;7v?&qVkVDFUQsu{YK^CUw|>-D9W*;les
zx5tn!RQs#`ne7+I!ly~(R_QC(sE=(|H~_vrJmk1h66KuT88ou)CvKTSl;j;|DKY}9
zBh#{{1G2BWiDej2mnVqj(zTpvI6vdJjv5<DqRXI#9$&*-O!?MZMds4FC`J`1VUDa#
zRt(^PJ4HKN5LzrU`Vv~a+Yh!PY4KnL+}MGv)wc6_RPl28@~fJ`^=@bD&n_<$d=`$P
z3DJ-0S`g=VE{9W{^j!?Zo8T!}HtLYN=Kzj8Vm)7w9WcwPw4Qw*DubItS1H?H^QtDp
z-?UsK=vyLecF&7#nm*Pra2+*mA>fVa9=fo?QsWS#+dFcf^-Q=h?@&aE5zz5jO1&<)
z_O1-ptg7Mo?X=7M><+iVa{0U%GqoQ9?mL>Ofe9LSZ4@&n0+<PyHQ2aaTjSxPb!DjV
z(HoNpwO=NTFS_uSZ|W}U`S67sbJ8rXK(&4A#=7b4)*^Fgt0E&GG4lymo+(M&5@A-E
zb-JvJnB}L*_O;@QhVX?&`MzvXGcpMS&R(>9pE$FGh%uSOK^0pdL&bK$38V;%`g(oL
zMa^~9>=3cHH@?WT$kMoHu0tS@$Y+s$DA~P4xVFfX;u2#Qy(}L;uTe2)?tJ|k=&~Iw
zJ~!-SvwiMcLVmwVqL(zj;)#M2SpE<uvewgvMT<4Mb3_;#tkIipPNW#S+bxL<uw6O7
zwQ6jt*4f6pHXxjxC+(c~Qu?v82X0l3L6WM{-S<kR>^1E1n%`Vj(7H#xl)3+OK$<Xj
z<EO9vDd=<3vT1ED%O`}D9n`n&mhf7I+&~UQsS{SRTYXL=PE5(lYp`EQ49Ir%ObL3W
zNNzEjB-cDzDw*v}C7$uAz7*1hsfX4q#<L+^EyN&s-C8xdP|FD|Q*Wc!))FzoH$mEp
zA-%Y02Y~_w(#77OCG?RnhVU{7<<dy<G8A*_)N~<2G9p75Xg+YxA0m@SUw{%_TVi{w
z5ppb{@YFfDiuVn|X+9{9o@323E6$*y*Jwe=pY!cgg1$7i+}*$;mZ8ccnZxYc3jB3X
zl*~Mmb#YHj(kUF02q$jPw-haZc?c5!T<<Bm*h#^#v~dU2&hNj9Dk8KfNRo4Q&WWS0
z?qJ>aPH-F>%n^T+g?Vv`3p+%3<1l)|P$TZJz-3MCQfo$W<0yQ+*muAI%%wiGOR$;>
zscOrfbp9lWWaJ~#+3tQI5Tikrx7Yi{DIdfojG&0!28JNnA52K*pHcU;ex>)GX!0zq
z8c^QQe&CAw&|^ruYE^O0B<8`Mfs4H%tR1@9i+YQ6shXX6Ba<@X?Wq>WCw%1%E8Ny1
ze(`~TL^HpIi!3)-h@cPsP$4qWMZq3?(zLHVgRY6_Y8ib_2O?>c9TTM|5E!?$kQw%P
zIVKuqM$%UQ(yi97kuNlpyP<|TiOSpv?eHqof3V6Y-!!npyUx0eqNCrVVwi4sXQ206
zTU@27=ST6i!y`yLG%;7g=+@wO<LxT@*I`Q~Rd@)E5p^^iyzs2cJ32d{_uS2}35XyN
z$gs62m-K>0sexFB!ieMDCf#Vpy?;bPGNxo#LU*FM(r<YnfyF1T?#PrbCr>NYa+G-l
zBumEh%9y@d#hR{YH065bE5U^&`h2>t>~OzR#^T-4wl%(hffQdOW>2v6NLP!lzR-Iq
z2Q2TH{6_F;cHBTUvx-P}Ce7ZR-s#{rt$d}W?bC(C!U&nk-hKDZc5oj45;p2OkO@r%
zx%%yq`;z={679tFelR1$3^P{0$cec;44bfvhMe2u0F1@5Q&U=M)3$G9@`TG+&c&mf
zO;I?ETcA+gmFiW6k~!p@?J#awCVl?<W%2wjc^lYvv|7a41xIU=bcq$8;xHAAuVTHv
z)dr)bdL|gua(dLSTxgC8g}a^x1}tl8ih~K;X2=j%s=M<hHg;>srAPRQLz)G>hJB^p
zLmJ%6J$<Jg3_m*%<{%c)fguXYyY9_QNZRht{OM+7M~`yXK;^eLM{|!diL9uB)%u_s
zmXTK++?lI$C(2XueahbE)?i9r@y%71O}S(Wkc@qsVI}0qx!(z--<Hz<YJWOb0gj%p
z$5x-emkvTo?VoHzYnZ)t_U3CBAah|Y&tKJt%&JRV7K3Q5e_E3LS)Y~uqNwVEoITe^
zZXd3XG5j9-G=^JkILlo^q0x(vB44eNP}lZ(;YX%-kVlxHiN!&jeH^)H!foX73rOy=
zEYB^C`fVBeZrz1rhH3`)ijb_+kHc$L=>ql;SEQA^m!D;n4i@?I^NX6vwd!Ha>@|=R
z`yi&&zCHydIrh0c%;3#Y6d?#xcx$%g6Zyk2o@vE;CG#n5z6kP%(Z)_~c7&j^^0YBz
zQP*UE!XAj+?xm3fGRlR)^s*glP3WrANaB<{#_#754w^LP)-Z%7m7XL&jn?sN<dDGQ
z$$Nsn_YKQ?9>|SipNLT+E7;o4%CN|_&sreXRRq(J-2q17M`)Zl)m76+^`9aionJF_
zuX<~5*}FF}US!IpSpV2THIQlIwPU$Ai`486$DF#$g#{^EZ%T!9I%lf%$aE12QG{u4
z{1ekYgxC35J!1T@{<%7Y=hCQ&%;3mZS8vc$?*l)W`&pGKc)v9HNZcuN;c;@j8k}xc
z!H0Y-YT9qYr#N{1HFv>d5ZhS}Q<U2nW(UgJA7aXs#e4%;MPa_1roRANpFHp}q62M=
zrkGc@ruZFB%_VD>2OW~I^b^Z9=&utn`*fse^<3qr7ZQcYEb>%1k5&7uP4=MKzaV|8
zcro<2`XIIQ%I>8z1gkzlAJheW9STB@q@7qzyM1{+*(32h7g<0qqSB8Z(1G2!GgGwZ
z*A|UlsJtJK6gIo2_EH$6H)d=jI}our(-tQk(!Dz7C#I%7y0UF}QwzIoEM{Mx;!BNT
zu>igFTQ?4J9poPc_ePr4Z)@ruaxIYlL=)y*P(81AN2b=kh($<b^&%ssbR6&pEql#-
zu^0$0^5*DM)W?b!w}RrN$$1|01g@^fag<y@ilO&E1<E0L&F&U5y2E#?(9mncyXo)V
ztvIXsA_0e9SchOWxlWMuO8bg$QwqW&Q^I4usVlvqdD|ytRoAq>E2?GG4#3%m+?Gt`
z_Zq6YdN4D&ImGlu|8>Q#nZha-47+$ad~ZE|S)x=VeYEYFvBJaiiwqLrUJ*poN7>pp
zKSIJ?vhBz0Tx*aP82Tc_ke)_}F@ul6X~T~?Z1@0quqd4>k&8<jEL(a<$38iFB4}^%
zbdPxE{KIjVUp!11@WwCS+!4POvFpW251pAmJ&m3?Xh50WHpd;je~Mc}D|o-;<F~Zz
zGA4dTl|Jqv526gBV1lB4ytak<Zqlw{u~W9bltBWq%}4bdUDL9kLe=@;!XJC!a#gE@
z0-N#Id-C!=DR1_(_S`8%O3Hn-u7(mFY-fW-AKFDVp0zPO*lvCE0l}1462v9|=o<`m
zX02Z6@s|Qki5*nzo;y>F^>i!CN2NLt<#dzWl&Ft3fhIk~*c6ByPE-tcgay}De!41H
zNfu7(O=ZEYth}0a*=H1b2kDX(CfQIF%iCxM0aHiS$G<?=WiR%yZ@X-+j!JiwXF+z?
zm1U1|XBQSE-Kx#kU$uK{VOqPT=D7}Zghr%gSAKNa)~=0OguRWV0u>L?d)Kg7iXiu&
zSOcpj#|e&tKD-q#B;>ovcJWy`v6bvK$>mAZ@D@w3VV~jo&;s0Jul#MqkI&_+u9h<s
z&!~bX(FdYqoj+2P#X&%|z?l84gFOrA_~DIULvW6|bH&BWl~Uz%z4hAB8mPgvS%%=}
zRX?8%;rf;&QZ_ovGFa5?Di;w3LhnVTpWGGzFpLOf{>#HC3FrG^AS(`kuIoXNQ(pM}
z*x?H<>mfz0&Gs>7A@|E6QzTT3Y;UBtUd|%j_*rl_Kf_SauHG+>{+c|vbZ{8A+%L;B
zi!^;I$639dbQT%HvJmqgFpnX08uaPK(#x|QufCAxSxH5?Ok2;yu2h)M?p2Hm;J6p2
z#gCS82t}BX4X|U7jlbtL_nTgXP-!tK&&rnbwENf|FduObbG#Bl0NxvM7GQ(8!A;e@
zx|Q5T*&cep%O|AN{A8OybM@J9xfyHNFuH}kUPq-Iztu1_GMiaRNr2|se=i9Ak?Vyt
zWGGtwW(odjRI^gqs_v(bKl(-;I*HM6DB<7&3QFFx%)}6xJ+NBF>RF{i`xsi^Jwlm>
zaLUdl!lZyV4!NU_4>}nL(j=dhEssoMYnrDb<rtP>5kGwmCUXC2b|TyNBqr&LKV^T?
z0YngvC)fC?q<7xBs`&`;@rHN%^p1JnRwy#ziS)^}B{xIr*(A6^5MEzBT`G-odCwIx
zLq6lG*;kwi`Z{e!B;9`6EQs6i8bq|?0|{i^pdzT4#(yhl=f};sBbJNpJIX(mCMrJ<
zL%`Ka7A;JpR{O)0F1)(FRK}WLbGy?sRosRHMVESl3eu5cnZiJ**MN7QX)0gsNzf0y
zAd&K8bl4Ye0CQ}#QT5jOp)@S0cc4jN4~yRt?MZl5t{H-0z0Zc9p>Rb+1jk1tvQ$Jw
zIYvfr`lq;b3ajGds$D&*5ccbK6K^f{N4heHn~E+6xsZmO6G-g!$wbEq?z8OGd<(T%
z=}Y<A%2<@{lt9sljB99`j>63jgWQs>LQ>9`G{auwM6%UGpPngHUBchRkCq#o@Dzk!
zz;HtkpyMGU6Fr^+VQ1UsHmjJnh72K5b9bORli&F;X*iI0jo6A{qsnjiB_bC2uJ!D$
zsl9cgD@t#^I~`^XWDj&mQPv28xv~*;vdR|ZIlKhcprlN$S>gTTngg68P?ggAs%B#P
z#Q*X*98%JI^9GQ)dg#poXNZqymiX3+Il|v|64`G4{JK?xIyGNF_w>W@g$=xAGQH^H
zr5}Y=Qd=9WKVV_9X@d5yCTC(a0F%C-8dbP-(B9i<mQX|IpZ(_9hb%_*-j0UHX<1e2
zpB3Dm0I@E%VjzE~n#Q;pq{nk4L3muN<iPY0KAsPC@ipkKHJpb|xWb==^_?3H!~AN5
zG@KE2fnIo~M7kP0*}UDDfWKkqYTf?iae)b`!1-ZGvDH~0o@<o_jH)$UbfK7fdt1Tg
z)3X;_<6gaAZFyT#JV7<j0f&i6kw;(avofWrTcw->7{>Dbw#UP2vnH74L1MOWx|_}q
zUQ!akg^hbfcG}7aDw^UCkLUP9Z!1JoCikKa<jn%+lHd0rsYc$92%`XkibHkKDxC|s
z7Jo;mVqhxd9Vg>H0ySkAaOEJh=%Nbmp^8!COK4cH13BMbwCzgi+U}L+f1&sSvL_QW
z8~IBf8buCVeT0bM*s4d-*W(GZE;Bc45|=sB^tVY_$y|{>JVmSB;I1Iv5F90^dzOpc
zI|CvM15<4~%vT0Up6&Ub<W>UkLnW)P;*)gvTEiOGyJ(|drjZ3OoHy<V*P4XQP{RHD
z@1&tD2xiXi_&=zA4qW@x9*^cij>qv@xx^Te-jqz?^Ux0`v**di_x-)EwqgO_Lsr?<
zOSDk@41I`O+)-4GP)B?cwH+$?jK~6^p@eDH9l1Yvp|jP<2Cb?9I8RDUJFVeQAJ)3X
z{_<-cJK*1tH(+`3yQC6d1wD?h44O#yAo)`b`L+*VLH!&mbLlVJs+QpvP%k<|@w?Xe
zirlSSe5V&pLuhaQQ(28~-1+fA)vfLNAHI#8Jib$!`lGv2e|nuV!*hWqV+DT~jpA!;
zpVQw0IJ|eY{QvaY(u61AG+np<OSiemD+fGTT+ZM4(`$GTzP2++u;(x1=&Erz@V;+I
zdHOHwX`hf-Zr?8BFU^Oo8oxcvEqn9%Z>8gZ74lz&{MSSNKVC8s?@3xLTcZk#?R=o#
z<~;w7CRbi2%RM2Q8m+V*C{6Mlc9@QCTPAWh{S%)60t3QyA>gu?mr`On!x&gOHvKH>
zEPHG`tNkz66VJN5vr$CSGB#m?{x{2W=6~`xor#6MOV`?c%|O-F)&B==^o|P6cq27T
z@ei%>ii0bj<ocnY04K_q?KRq+Y)UHEaPV?7^+Y$gxp?A#e$_6bpdfmfFZ(ohtW1d(
zDAvr*GS>aD^IkKUsv$(d>~Ac;b)`uzk?r(8|JXi1z6z<tBM9-5FsG?wv0q1WsHo_q
z>|oZw75i@FZ@eUXbo(t^*yECt({;VFd6t;jd(@<IK~+a7a)%K}VwslrEsqcWMM?Vw
z1guk_`J{pcNt-UOgzA$+ELSP8i2C;JVk%`~Jw~lsE=e(XrdqF}j7=pg{Y~qC&{dxi
z2WuCksdD<WK^|Yl%ByIkHQAxdA^j3~9&s%4%71WwugKyd?4&$^^N(d(Xz)>AQ<vz)
zUtsvjiKn7w3$H)_L-+KT$HUJnr<^N)ute?z52u?|^O=7Di9i!iMU~c~q$T_VYA>_#
z*bzG1Nb?uNF8<;clt#$^g{V#Zu~u8}2S52gHj}!Ir=mU*eoFR7wp*ciG(9ld*7!4U
zIOi`M|44G_FGhXQ#8Xja!IKyN47~9ikAyTsa<~7;7IE;4ifZ32`Ipp?bp@YS<T2hQ
i{v+G}Q%X5Fzj6GX@NNZs9seHz02M`zM^Jf-Xa5V;XD-12

literal 0
HcmV?d00001


From 1208a2ed20bee18b82d88b7a4163868ea0aafd2d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Sat, 26 Apr 2025 13:52:56 +0200
Subject: [PATCH 316/505] Fix version mismatch numpy

---
 pyproject.toml | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 352fd5a91..70ebfe0ff 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -51,6 +51,7 @@ test = [
     "pytest-benchmark",
     "pytest-cov",
     "pytest-mock",
+    "pytest-asyncio",
     "coverage[toml]>=5.1",
     "diff-cover",
     "accelerate",
@@ -67,7 +68,7 @@ test = [
     "pillow",
     "jax",
     "flax",
-    "numpy",
+    "numpy>=2.0.0,<2.2.0",
     "torch",
     "tensorflow",
     "tf-keras",

From 9f6e917aae1964a8cbd92e4bb63290cdedb5523c Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Sat, 26 Apr 2025 21:45:41 +0200
Subject: [PATCH 317/505] Rename VLLM to VLLMOffline and create server-based
 VLLM/AsyncVLLM

---
 docs/reference/models/vllm.md                 |  65 ++--
 docs/reference/models/vllm_offline.md         |  68 ++++
 ...gration.py => vllm_offline_integration.py} |   2 +-
 outlines/__init__.py                          |   1 +
 outlines/generator.py                         |  42 ++-
 outlines/models/__init__.py                   |   9 +-
 outlines/models/base.py                       |  87 ++++-
 outlines/models/vllm.py                       | 337 +++++++++++-------
 outlines/models/vllm_offline.py               | 203 +++++++++++
 outlines/serve/serve.py                       |   2 +-
 outlines/v0_legacy/generate/api.py            |  10 +-
 outlines/v0_legacy/models/__init__.py         |   2 +-
 .../models/{vllm.py => vllm_offline.py}       |  17 +-
 pyproject.toml                                |   4 +-
 tests/models/test_vllm.py                     | 294 +++++++++------
 tests/models/test_vllm_offline.py             | 157 ++++++++
 .../models/test_vllm_offline_type_adapter.py  |  37 ++
 tests/models/test_vllm_type_adapter.py        | 121 +++++++
 tests/processors/test_guide.py                |   2 +-
 tests/test_generator.py                       | 149 ++++++--
 tests/test_utils/mock_openai_client.py        | 125 +++++++
 ..._legacy.py => test_vllm_offline_legacy.py} |  32 +-
 22 files changed, 1390 insertions(+), 376 deletions(-)
 create mode 100644 docs/reference/models/vllm_offline.md
 rename examples/{vllm_integration.py => vllm_offline_integration.py} (93%)
 create mode 100644 outlines/models/vllm_offline.py
 rename outlines/v0_legacy/models/{vllm.py => vllm_offline.py} (88%)
 create mode 100644 tests/models/test_vllm_offline.py
 create mode 100644 tests/models/test_vllm_offline_type_adapter.py
 create mode 100644 tests/models/test_vllm_type_adapter.py
 create mode 100644 tests/test_utils/mock_openai_client.py
 rename tests/v0_legacy/models/{test_vllm_legacy.py => test_vllm_offline_legacy.py} (86%)

diff --git a/docs/reference/models/vllm.md b/docs/reference/models/vllm.md
index 1b88b8d7e..fe4e55474 100644
--- a/docs/reference/models/vllm.md
+++ b/docs/reference/models/vllm.md
@@ -1,68 +1,51 @@
 # vLLM
 
+## Prerequisites
 
-!!! Note "Installation"
+The Outlines `VLLM` model is inteded to be used along with a vllm instance running on a separate server (can be local or remote). Make sure you have a vllm server running before using the `VLLM` model. As the vllm client relies on the `openai` python sdk, you need to have an `openai` package installed. If you instead want to use the vllm offline inference mode, please refer to the [VLLMOffline model documentation](./vllm_offline.md).
 
-    You need to install the `vllm` library to use the vLLM integration: `pip install vllm`. The default installation only works on machines with a GPU, follow the [installation section][vllm-install-cpu] for instructions to install vLLM for CPU or ROCm.
+## Initialize the model
 
-    Consult the [vLLM documentation][vllm-docs] for detailed informations about how to initialize OpenAI clients and the available options.
-
-## Load the model
-
-Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
+To load the model, you can use the `from_vllm` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Based on whether the `openai` instance is synchronous or asynchronous, you will receive a `VLLM` or `AsyncVLLM` model instance.
 
 ```python
+import openai
 import outlines
-from vllm import LLM
-
-model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
-```
-
-Models are loaded from the [HuggingFace hub](https://huggingface.co/).
 
+sync_openai_client = openai.OpenAI(base_url="...")
+async_openai_client = openai.AsyncOpenAI(base_url="...")
 
-!!! Warning "Device"
-
-    The default installation of vLLM only allows to load models on GPU. See the [installation instructions][vllm-install-cpu] to run models on CPU.
+sync_model = outlines.from_vllm(sync_openai_client)
+print(type(sync_model)) # <class 'outlines.models.vllm.VLLM'>
 
+async_model = outlines.from_vllm(async_openai_client)
+print(type(async_model)) # <class 'outlines.models.vllm.AsyncVLLM'>
+```
 
 ## Generate text
 
-To generate text, you can just call the model with a prompt as argument:
+To generate text, you can call the model with a prompt as argument and optionally an output type to rely on structured generation:
 
 ```python
-import outlines
-from vllm import LLM
-
-model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
-answer = model("Write a short story about a cat.")
-```
-
-You can also use structured generation with the `VLLM` model by providing an output type after the prompt:
-
-```python
-import outlines
-from vllm import LLM
 from pydantic import BaseModel
 
 class Character(BaseModel):
     name: str
 
-model = outlines.from_vllm(LLM("microsoft/Phi-3-mini-4k-instruct"))
-answer = model("Create a character.", output_type=Character)
+answer = sync_model("Create a character.", output_type=Character)
+answer = await async_model("Create a character.", output_type=Character)
 ```
 
-The VLLM model supports batch generation. To use it, you can pass a list of strings as prompt instead of a single string.
+The `VLLM` model supports also supports streaming.
 
-## Optional parameters
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `LLM.generate` method of the `vllm` library. An optional parameter of particular interest is `sampling_params`, which is an instance of `SamplingParams`. You can find more information about it in the [vLLM documentation][https://docs.vllm.ai/en/latest/api/inference_params.html].
+```python
+for chunk in sync_model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk)
+```
 
-!!! Warning
+## Optional parameters
 
-    Streaming is not available for the offline vLLM integration.
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `openai` client. An optional parameter of particular interest is `extra_body`, which is a dictionnary containing arguments that are specific to vLLM and are not part of the standard `openai` interface (see the [vLLM documentation][https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html] on the OpenAI-compatible server for more information).
 
-[vllm-docs]:https://docs.vllm.ai/en/latest/
-[vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
-[vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
-[rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
+[vllm-docs]: https://docs.vllm.ai/en/latest/
+[vllm-online-quickstart]: https://docs.vllm.ai/en/latest/getting_started/quickstart.html#quickstart-online
diff --git a/docs/reference/models/vllm_offline.md b/docs/reference/models/vllm_offline.md
new file mode 100644
index 000000000..eac185b6d
--- /dev/null
+++ b/docs/reference/models/vllm_offline.md
@@ -0,0 +1,68 @@
+# vLLM Offline Inference mode
+
+
+!!! Note "Installation"
+
+    You need to install the `vllm` library to use the vLLM integration: `pip install vllm`. The default installation only works on machines with a GPU, follow the [installation section][vllm-install-cpu] for instructions to install vLLM for CPU or ROCm.
+
+    Consult the [vLLM documentation][vllm-docs] for detailed informations about how to initialize OpenAI clients and the available options.
+
+## Load the model
+
+Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
+
+```python
+import outlines
+from vllm import LLM
+
+model = outlines.from_vllm_offline(LLM("microsoft/Phi-3-mini-4k-instruct"))
+```
+
+Models are loaded from the [HuggingFace hub](https://huggingface.co/).
+
+
+!!! Warning "Device"
+
+    The default installation of vLLM only allows to load models on GPU. See the [installation instructions][vllm-install-cpu] to run models on CPU.
+
+
+## Generate text
+
+To generate text, you can just call the model with a prompt as argument:
+
+```python
+import outlines
+from vllm import LLM
+
+model = outlines.from_vllm_offline(LLM("microsoft/Phi-3-mini-4k-instruct"))
+answer = model("Write a short story about a cat.")
+```
+
+You can also use structured generation with the `VLLM` model by providing an output type after the prompt:
+
+```python
+import outlines
+from vllm import LLM
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+
+model = outlines.from_vllm_offline(LLM("microsoft/Phi-3-mini-4k-instruct"))
+answer = model("Create a character.", output_type=Character)
+```
+
+The VLLM model supports batch generation. To use it, you can pass a list of strings as prompt instead of a single string.
+
+## Optional parameters
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `LLM.generate` method of the `vllm` library. An optional parameter of particular interest is `sampling_params`, which is an instance of `SamplingParams`. You can find more information about it in the [vLLM documentation][https://docs.vllm.ai/en/latest/api/inference_params.html].
+
+!!! Warning
+
+    Streaming is not available for the offline vLLM integration.
+
+[vllm-docs]:https://docs.vllm.ai/en/latest/
+[vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
+[vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
+[rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
diff --git a/examples/vllm_integration.py b/examples/vllm_offline_integration.py
similarity index 93%
rename from examples/vllm_integration.py
rename to examples/vllm_offline_integration.py
index dfe873517..e320c6184 100644
--- a/examples/vllm_integration.py
+++ b/examples/vllm_offline_integration.py
@@ -4,7 +4,7 @@
 from pydantic import BaseModel
 from transformers import AutoTokenizer
 
-from outlines.models.vllm import adapt_tokenizer
+from outlines.models.vllm_offline import adapt_tokenizer
 from outlines.processors import JSONLogitsProcessor
 
 
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 8234b0c14..47188c2b5 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -19,6 +19,7 @@
     from_llamacpp,
     from_mlxlm,
     from_vllm,
+    from_vllm_offline,
 )
 from outlines.templates import Template, prompt
 from outlines.types import regex, json_schema, cfg
diff --git a/outlines/generator.py b/outlines/generator.py
index 65842f4c9..badeccecb 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -1,6 +1,7 @@
 from typing import Any, Optional, Union
 
 from outlines.models import BlackBoxModel, SteerableModel
+from outlines.models.base import AsyncModel
 from outlines.processors import (
     CFGLogitsProcessor,
     GuideLogitsProcessor,
@@ -13,23 +14,14 @@
 
 
 class BlackBoxGenerator:
-    """Represents a generator for which we don't control constrained generation.
-
-    This type of generator only accepts an output type as an argument defining
-    constrained generation. This output type is not modified and thus only
-    passed through to the model.
-    """
+    """Synchronous generator for which we don't control constrained generation."""
     output_type: Optional[Any]
 
     def __init__(self, model, output_type: Optional[Any]):
         self.model = model
         self.output_type = output_type
 
-        if isinstance(self.output_type, CFG):
-            raise NotImplementedError(
-                "CFG generation is not supported for API-based models"
-            )
-        elif isinstance(self.output_type, FSM):
+        if isinstance(self.output_type, FSM):
             raise NotImplementedError(
                 "FSM generation is not supported for API-based models"
             )
@@ -41,6 +33,29 @@ def stream(self, prompt, **inference_kwargs):
         return self.model.generate_stream(prompt, self.output_type, **inference_kwargs)
 
 
+class AsyncBlackBoxGenerator:
+    """Asynchronous generator for which we don't control constrained generation."""
+    output_type: Optional[Any]
+
+    def __init__(self, model, output_type: Optional[Any]):
+        self.model = model
+        self.output_type = output_type
+
+        if isinstance(self.output_type, FSM):
+            raise NotImplementedError(
+                "FSM generation is not supported for API-based models"
+            )
+
+    async def __call__(self, prompt, **inference_kwargs):
+        return await self.model.generate(prompt, self.output_type, **inference_kwargs)
+
+    async def stream(self, prompt, **inference_kwargs):
+        async for chunk in self.model.generate_stream(  # pragma: no cover
+            prompt, self.output_type, **inference_kwargs
+        ):
+            yield chunk
+
+
 class SteerableGenerator:
     """Represents a generator for which we control constrained generation.
 
@@ -134,7 +149,10 @@ def Generator(
         if processor is not None:
             raise NotImplementedError("This model does not support logits processors")
         else:
-            return BlackBoxGenerator(model, output_type)
+            if isinstance(model, AsyncModel):
+                return AsyncBlackBoxGenerator(model, output_type)
+            else:
+                return BlackBoxGenerator(model, output_type)
     else:
         if processor is not None:
             return SteerableGenerator.from_processor(model, processor)
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 9bd06ead6..22e755b23 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -22,11 +22,12 @@
     TransformersMultiModal,
     from_transformers,
 )
-from .vllm import VLLM, from_vllm
+from .vllm_offline import VLLMOffline, from_vllm_offline
+from .vllm import AsyncVLLM, VLLM, from_vllm
 
 LogitsGenerator = Union[
-    Transformers, LlamaCpp, OpenAI, MLXLM, VLLM, Ollama
+    Transformers, LlamaCpp, OpenAI, MLXLM, VLLMOffline, Ollama
 ]
 
-SteerableModel = Union[LlamaCpp, Transformers, MLXLM, VLLM]
-BlackBoxModel = Union[OpenAI, Anthropic, Gemini, Ollama, Dottxt]
+SteerableModel = Union[LlamaCpp, Transformers, MLXLM, VLLMOffline]
+BlackBoxModel = Union[OpenAI, Anthropic, Gemini, Ollama, Dottxt, AsyncVLLM, VLLM]
diff --git a/outlines/models/base.py b/outlines/models/base.py
index 375f2549d..aa0f736c0 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -38,7 +38,7 @@ def format_output_type(self, output_type):
 
 
 class Model(ABC):
-    """Base class for all models.
+    """Base class for all synchronous models.
 
     This class defines a shared `__call__` method that can be used to call the
     model directly.
@@ -116,3 +116,88 @@ def generate_stream(self, model_input, output_type=None, **inference_kwargs):
 
         """
         ...
+
+
+class AsyncModel(ABC):
+    """Base class for all asynchronous models.
+
+    This class defines a shared `__call__` method that can be used to call the
+    model directly.
+    All models inheriting from this class must define a `type_adapter`
+    attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
+    attribute are used in the `generate` method to format the input and output
+    types received by the model.
+    Additionally, local models must define a `tensor_library_name` attribute.
+
+    """
+    type_adapter: ModelTypeAdapter
+    tensor_library_name: str
+
+    async def __call__(self, model_input, output_type=None, **inference_kwargs):
+        """Call the model.
+
+        Users can call the model directly, in which case we will create a
+        generator instance with the output type provided and call it.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        await generator("prompt")
+        ```
+        and
+        ```python
+        await model("prompt", Foo)
+        ```
+
+        """
+        from outlines import Generator
+
+        generator = Generator(self, output_type)
+        return await generator(model_input, **inference_kwargs)
+
+    async def stream(self, model_input, output_type=None, **inference_kwargs):
+        """Stream a response from the model.
+
+        Users can use the `stream` method from the model directly, in which
+        case we will create a generator instance with the output type provided
+        and then invoke its `stream` method.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        async for chunk in generator("prompt"):
+            print(chunk)
+        ```
+        and
+        ```python
+        async for chunk in model.stream("prompt", Foo):
+            print(chunk)
+        ```
+
+        """
+        from outlines import Generator
+
+        generator = Generator(self, output_type)
+
+        async for chunk in generator.stream(model_input, **inference_kwargs):  # pragma: no cover
+            yield chunk
+
+    @abstractmethod
+    async def generate(self, model_input, output_type=None, **inference_kwargs):
+        """Generate a response from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
+
+        """
+        ...
+
+    @abstractmethod
+    async def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+        """Generate a stream of responses from the model.
+
+        The output_type argument contains a logits processor for local models
+        while it contains a type (Json, Enum...) for the API-based models.
+        This method is not intended to be used directly by end users.
+
+        """
+        ...
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 4f928eac0..e199fa76c 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,21 +1,22 @@
-import warnings
-from functools import singledispatchmethod
-from typing import TYPE_CHECKING, List, Optional, Union
+import json
+from typing import TYPE_CHECKING, Union, Optional
 
-from outlines.models.base import Model, ModelTypeAdapter
-
-if TYPE_CHECKING:
-    from transformers import PreTrainedTokenizerBase
-    from vllm import LLM
+from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
+from outlines.models.openai import OpenAITypeAdapter
+from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
 
 
-__all__ = ["VLLM", "from_vllm"]
+if TYPE_CHECKING:
+    from openai import AsyncOpenAI, OpenAI
 
 
 class VLLMTypeAdapter(ModelTypeAdapter):
-    @singledispatchmethod
+
     def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
+        """Generate the prompt argument to pass to the client.
+
+        We rely on the OpenAITypeAdapter to format the input as the vLLM server
+        expects input in the same format as OpenAI.
 
         Argument
         --------
@@ -23,181 +24,255 @@ def format_input(self, model_input):
             The input passed by the user.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available."
-            "Please use a string or a list of strings."
-        )
-
-    @format_input.register(str)
-    def format_str_input(self, model_input):
-        return model_input
-
-    @format_input.register(list)
-    def format_list_input(self, model_input):
-        return model_input
+        return OpenAITypeAdapter().format_input(model_input)
 
     def format_output_type(self, output_type):
-        """Generate the logits processor argument to pass to the model.
+        """Generate the structured output argument to pass to the client.
 
         Argument
         --------
         output_type
-            The logits processor provided.
+            The structured output type provided.
 
         """
-        if output_type is not None:
-            return [output_type]
-        return []
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            return {"guided_grammar": term.definition}
+        elif isinstance(term, JsonSchema):
+            extra_body = {"guided_json": json.loads(term.schema)}
+            if term.whitespace_pattern:
+                extra_body["whitespace_pattern"] = term.whitespace_pattern
+            return extra_body
+        else:
+            return {"guided_regex": to_regex(term)}
 
 
 class VLLM(Model):
-    """Represents a `vllm` model."""
-    tensor_library_name = "torch"
+    """Represents a client to a `vllm` server."""
 
-    def __init__(self, model: "LLM"):
-        """Create a VLLM model instance.
+    def __init__(self, client, model_name: Optional[str] = None):
+        """Create a `VLLMServer` model instance.
 
         Parameters
         ----------
-        model
-            A `vllm.LLM` model instance.
+        client
+            An `OpenAI` client instance.
 
         """
-        self.model = model
-        self.tokenizer = self._get_tokenizer()
+        self.client = client
+        self.model_name = model_name
         self.type_adapter = VLLMTypeAdapter()
-        self.lora_request = None # v0 legacy, to be removed
-
-    def _get_tokenizer(self):
-        if hasattr(self.model, "get_tokenizer"):
-            tokenizer = self.model.get_tokenizer()
-        elif hasattr(self.model, "tokenizer"):
-            if hasattr(self.model.tokenizer, "tokenizer"):
-                tokenizer = self.model.tokenizer.tokenizer
-            else:
-                tokenizer = self.model.tokenizer
-        else:
-            raise ValueError(
-                "The provided LLM instance neither has a "
-                "`tokenizer` attribute or a `get_tokenizer` method."
-            )
-        return adapt_tokenizer(tokenizer=tokenizer)
 
     def generate(self, model_input, output_type, **inference_kwargs):
-        """Generate text using `vllm`.
+        """Generate text using the client.
 
         Arguments
         ---------
-        prompt
+        model_input
             The text prompt provided by the user.
-        logits_processor
-            The logits processor to use when generating text.
+        output_type
+            The structured output type provided.
         inference_kwargs
-            The inference kwargs that can be passed to the `LLM.generate` method
-            in the `vllm` library.
+            The additional arguments that will be passed on to the client.
 
         Returns
         -------
         The generated text.
 
         """
-        from vllm.sampling_params import SamplingParams
-
-        sampling_params = inference_kwargs.pop("sampling_params", None)
-        if sampling_params is None:
-            sampling_params = SamplingParams()
-        sampling_params.logits_processors = self.type_adapter.format_output_type(output_type)
-
-        results = self.model.generate(
-            self.type_adapter.format_input(model_input),
-            sampling_params=sampling_params,
-            lora_request=self.lora_request, # v0 legacy, to be removed
+        client_args = self._build_client_args(
+            model_input,
+            output_type,
             **inference_kwargs,
         )
-        results = [[sample.text for sample in batch.outputs] for batch in results]
 
-        batch_size = len(results)
-        sample_size = len(results[0])
+        response = self.client.chat.completions.create(**client_args)
 
-        if batch_size == 1 and sample_size == 1:
-            return results[0][0]
-        elif batch_size == 1:
-            return results[0]
-        elif sample_size == 1:
-            return [batch[0] for batch in results]
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The vLLM server refused to answer the request: "
+                    f"{message.refusal}"
+                )
 
-        return results
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
 
     def generate_stream(self, model_input, output_type, **inference_kwargs):
         """Return a text generator.
 
-        Streaming is not yet available for `vllm.LLM`.
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        A generator that yields the generated text chunks.
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = self.client.chat.completions.create(
+            **client_args, stream=True,
+        )
+
+        for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+        """Build the arguments to pass to the OpenAI client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        extra_body = inference_kwargs.pop("extra_body", {})
+        extra_body.update(output_type_args)
+        model_name = (
+            inference_kwargs.pop("model", self.model_name)
+            or self.model_name
+        )
+
+        client_args = {
+            **messages,
+            "model": model_name,
+            **inference_kwargs,
+        }
+        if extra_body:
+            client_args["extra_body"] = extra_body
+
+        return client_args
+
 
-        TODO: Implement the streaming functionality ourselves.
+class AsyncVLLM(AsyncModel):
+    """Represents an async client to a `vllm` server."""
+
+    def __init__(self, client, model_name: Optional[str] = None):
+        """Create an `AsyncVLLMServer` model instance.
+
+        Parameters
+        ----------
+        client
+            An `AsyncOpenAI` client instance.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = VLLMTypeAdapter()
+
+    async def generate(self, model_input, output_type, **inference_kwargs):
+        """Generate text using `vllm`.
+
+        Arguments
+        ---------
+        prompt
+            The text prompt provided by the user.
+        logits_processor
+            The logits processor to use when generating text.
+        inference_kwargs
+            The inference kwargs that can be passed to the `LLM.generate` method
+            in the `vllm` library.
+
+        Returns
+        -------
+        The generated text.
 
         """
-        raise NotImplementedError(
-            "Streaming is not available for the vLLM integration."
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
         )
 
-    def load_lora(self, adapter_path: Optional[str]):
-        warnings.warn("""
-            The `load_lora` method is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.5.0.
-            Please use the v1 of the `outlines` library by using the
-            `outlines.from_vllm` function to create a `VLLM` model
-            instance.
-            In the v1, you must pass the `lora_request` argument as
-            a keyword argument when calling the model or generator.
-            """)
-
-        from vllm.lora.request import LoRARequest
-
-        if adapter_path is None:
-            self.lora_request = None
-        else:
-            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
+        response = await self.client.chat.completions.create(**client_args)
 
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The vLLM server refused to answer the request: "
+                    f"{message.refusal}"
+                )
 
-def from_vllm(model: "LLM") -> VLLM:
-    return VLLM(model)
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
 
+    async def generate_stream(
+        self, model_input, output_type, **inference_kwargs
+    ):
+        """Return a text generator.
 
-def adapt_tokenizer(tokenizer: "PreTrainedTokenizerBase") -> "PreTrainedTokenizerBase":
-    """Adapt a tokenizer to use to compile the FSM.
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
 
-    The API of Outlines tokenizers is slightly different to that of `transformers`. In
-    addition we need to handle the missing spaces to Llama's tokenizer to be able to
-    compile FSMs for this model.
+        Returns
+        -------
+        A generator that yields the generated text chunks.
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
 
-    Parameters
-    ----------
-    tokenizer
-        The tokenizer of the model.
+        stream = await self.client.chat.completions.create(
+            **client_args,
+            stream=True,
+        )
 
-    Returns
-    -------
-    PreTrainedTokenizerBase
-        The adapted tokenizer.
-    """
-    from transformers import SPIECE_UNDERLINE
+        async for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+        """Build the arguments to pass to the OpenAI client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        extra_body = inference_kwargs.pop("extra_body", {})
+        extra_body.update(output_type_args)
+        model_name = (
+            inference_kwargs.pop("model", self.model_name)
+            or self.model_name
+        )
 
-    tokenizer.vocabulary = tokenizer.get_vocab()
-    tokenizer.special_tokens = set(tokenizer.all_special_tokens)
+        client_args = {
+            **messages,
+            "model": model_name,
+            **inference_kwargs,
+        }
+        if extra_body:
+            client_args["extra_body"] = extra_body
 
-    def convert_token_to_string(token: Union[str, bytes]) -> str:
-        string = tokenizer.convert_tokens_to_string([token])
+        return client_args
 
-        # A hack to handle missing spaces to HF's Llama tokenizers
-        if (
-            type(token) is str
-            and token.startswith(SPIECE_UNDERLINE)
-            or token == "<0x20>"
-        ):
-            return " " + string
 
-        return string
 
-    tokenizer.convert_token_to_string = convert_token_to_string
+def from_vllm(
+    client: Union["OpenAI", "AsyncOpenAI"],
+    model_name: Optional[str] = None,
+) -> Union[VLLM, AsyncVLLM]:
+    from openai import AsyncOpenAI, OpenAI
 
-    return tokenizer
+    if isinstance(client, OpenAI):
+        return VLLM(client, model_name)
+    elif isinstance(client, AsyncOpenAI):
+        return AsyncVLLM(client, model_name)
+    else:
+        raise ValueError(
+            f"Unsupported client type: {type(client)}.\n"
+            "Please provide an OpenAI or AsyncOpenAI instance."
+        )
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
new file mode 100644
index 000000000..808874c36
--- /dev/null
+++ b/outlines/models/vllm_offline.py
@@ -0,0 +1,203 @@
+import warnings
+from functools import singledispatchmethod
+from typing import TYPE_CHECKING, Optional, Union
+
+from outlines.models.base import Model, ModelTypeAdapter
+
+if TYPE_CHECKING:
+    from transformers import PreTrainedTokenizerBase
+    from vllm import LLM
+
+
+__all__ = ["VLLMOffline", "from_vllm_offline"]
+
+
+class VLLMOfflineTypeAdapter(ModelTypeAdapter):
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the model.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available."
+            "Please use a string or a list of strings."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input):
+        return model_input
+
+    @format_input.register(list)
+    def format_list_input(self, model_input):
+        return model_input
+
+    def format_output_type(self, output_type):
+        """Generate the logits processor argument to pass to the model.
+
+        Argument
+        --------
+        output_type
+            The logits processor provided.
+
+        """
+        if output_type is not None:
+            return [output_type]
+        return []
+
+
+class VLLMOffline(Model):
+    """Represents a `vllm` model."""
+    tensor_library_name = "torch"
+
+    def __init__(self, model: "LLM"):
+        """Create a VLLM model instance.
+
+        Parameters
+        ----------
+        model
+            A `vllm.LLM` model instance.
+
+        """
+        self.model = model
+        self.tokenizer = self._get_tokenizer()
+        self.type_adapter = VLLMOfflineTypeAdapter()
+        self.lora_request = None # v0 legacy, to be removed
+
+    def _get_tokenizer(self):
+        if hasattr(self.model, "get_tokenizer"):
+            tokenizer = self.model.get_tokenizer()
+        elif hasattr(self.model, "tokenizer"):
+            if hasattr(self.model.tokenizer, "tokenizer"):
+                tokenizer = self.model.tokenizer.tokenizer
+            else:
+                tokenizer = self.model.tokenizer
+        else:
+            raise ValueError(
+                "The provided LLM instance neither has a "
+                "`tokenizer` attribute or a `get_tokenizer` method."
+            )
+        return adapt_tokenizer(tokenizer=tokenizer)
+
+    def generate(self, model_input, output_type, **inference_kwargs):
+        """Generate text using `vllm`.
+
+        Arguments
+        ---------
+        prompt
+            The text prompt provided by the user.
+        logits_processor
+            The logits processor to use when generating text.
+        inference_kwargs
+            The inference kwargs that can be passed to the `LLM.generate` method
+            in the `vllm` library.
+
+        Returns
+        -------
+        The generated text.
+
+        """
+        from vllm.sampling_params import SamplingParams
+
+        sampling_params = inference_kwargs.pop("sampling_params", None)
+        if sampling_params is None:
+            sampling_params = SamplingParams()
+        sampling_params.logits_processors = self.type_adapter.format_output_type(output_type)
+
+        results = self.model.generate(
+            self.type_adapter.format_input(model_input),
+            sampling_params=sampling_params,
+            lora_request=self.lora_request, # v0 legacy, to be removed
+            **inference_kwargs,
+        )
+        results = [[sample.text for sample in batch.outputs] for batch in results]
+
+        batch_size = len(results)
+        sample_size = len(results[0])
+
+        if batch_size == 1 and sample_size == 1:
+            return results[0][0]
+        elif batch_size == 1:
+            return results[0]
+        elif sample_size == 1:
+            return [batch[0] for batch in results]
+
+        return results
+
+    def generate_stream(self, model_input, output_type, **inference_kwargs):
+        """Return a text generator.
+
+        Streaming is not yet available for `vllm.LLM`.
+
+        TODO: Implement the streaming functionality ourselves.
+
+        """
+        raise NotImplementedError(
+            "Streaming is not available for the vLLM integration."
+        )
+
+    def load_lora(self, adapter_path: Optional[str]):
+        warnings.warn("""
+            The `load_lora` method is deprecated starting from v1.0.0.
+            Support for it will be removed in v1.5.0.
+            Please use the v1 of the `outlines` library by using the
+            `outlines.from_vllm` function to create a `VLLM` model
+            instance.
+            In the v1, you must pass the `lora_request` argument as
+            a keyword argument when calling the model or generator.
+            """)
+
+        from vllm.lora.request import LoRARequest
+
+        if adapter_path is None:
+            self.lora_request = None
+        else:
+            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
+
+
+def from_vllm_offline(model: "LLM") -> VLLMOffline:
+    return VLLMOffline(model)
+
+
+def adapt_tokenizer(tokenizer: "PreTrainedTokenizerBase") -> "PreTrainedTokenizerBase":
+    """Adapt a tokenizer to use to compile the FSM.
+
+    The API of Outlines tokenizers is slightly different to that of `transformers`. In
+    addition we need to handle the missing spaces to Llama's tokenizer to be able to
+    compile FSMs for this model.
+
+    Parameters
+    ----------
+    tokenizer
+        The tokenizer of the model.
+
+    Returns
+    -------
+    PreTrainedTokenizerBase
+        The adapted tokenizer.
+    """
+    from transformers import SPIECE_UNDERLINE
+
+    tokenizer.vocabulary = tokenizer.get_vocab()
+    tokenizer.special_tokens = set(tokenizer.all_special_tokens)
+
+    def convert_token_to_string(token: Union[str, bytes]) -> str:
+        string = tokenizer.convert_tokens_to_string([token])
+
+        # A hack to handle missing spaces to HF's Llama tokenizers
+        if (
+            type(token) is str
+            and token.startswith(SPIECE_UNDERLINE)
+            or token == "<0x20>"
+        ):
+            return " " + string
+
+        return string
+
+    tokenizer.convert_token_to_string = convert_token_to_string
+
+    return tokenizer
diff --git a/outlines/serve/serve.py b/outlines/serve/serve.py
index 41fdf885a..84dc85761 100644
--- a/outlines/serve/serve.py
+++ b/outlines/serve/serve.py
@@ -35,7 +35,7 @@
 from vllm.sampling_params import SamplingParams
 from vllm.utils import random_uuid
 
-from outlines.models.vllm import adapt_tokenizer
+from outlines.models.vllm_offline import adapt_tokenizer
 from outlines.processors import JSONLogitsProcessor, RegexLogitsProcessor
 
 TIMEOUT_KEEP_ALIVE = 5  # seconds.
diff --git a/outlines/v0_legacy/generate/api.py b/outlines/v0_legacy/generate/api.py
index 14165a213..625bac105 100644
--- a/outlines/v0_legacy/generate/api.py
+++ b/outlines/v0_legacy/generate/api.py
@@ -9,13 +9,13 @@
 from outlines.v0_legacy.models.transformers import transformers_params_adapter
 from outlines.v0_legacy.models.transformers_vision import TransformersVision
 from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
-from outlines.v0_legacy.models.vllm import vllm_params_adapter
+from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
 from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
 from outlines.models import (
     LlamaCpp,
     MLXLM,
     Transformers,
-    VLLM
+    VLLMOffline,
 )
 
 V0GeneratorModel = Union[
@@ -24,7 +24,7 @@
     MLXLM,
     Transformers,
     TransformersVision,
-    VLLM,
+    VLLMOffline,
 ]
 
 
@@ -64,8 +64,8 @@ def create_inference_params(
 
         if isinstance(self.model, LlamaCpp):
             adapter_function = llamacpp_params_adapter
-        elif isinstance(self.model, VLLM): # pragma: no cover
-            adapter_function = vllm_params_adapter
+        elif isinstance(self.model, VLLMOffline): # pragma: no cover
+            adapter_function = vllm_offline_params_adapter
         elif isinstance(self.model, Transformers):
             adapter_function = transformers_params_adapter
         elif isinstance(self.model, ExLlamaV2Model): # pragma: no cover
diff --git a/outlines/v0_legacy/models/__init__.py b/outlines/v0_legacy/models/__init__.py
index 15d3a944e..1dc012c37 100644
--- a/outlines/v0_legacy/models/__init__.py
+++ b/outlines/v0_legacy/models/__init__.py
@@ -4,7 +4,7 @@
 from .openai import azure_openai, openai
 from .transformers import mamba, transformers
 from .transformers_vision import TransformersVision, transformers_vision
-from .vllm import vllm
+from .vllm_offline import vllm
 
 __all__ = [
     "ExLlamaV2Model",
diff --git a/outlines/v0_legacy/models/vllm.py b/outlines/v0_legacy/models/vllm_offline.py
similarity index 88%
rename from outlines/v0_legacy/models/vllm.py
rename to outlines/v0_legacy/models/vllm_offline.py
index 6b3309724..0928bfd03 100644
--- a/outlines/v0_legacy/models/vllm.py
+++ b/outlines/v0_legacy/models/vllm_offline.py
@@ -3,10 +3,10 @@
 
 import warnings
 
-from outlines.models import VLLM
+from outlines.models import VLLMOffline
 
 
-def vllm_params_adapter(
+def vllm_offline_params_adapter(
     sampling_params: dict, generation_params: dict, kwargs: dict
 ) -> dict:
     """Adapt the parameters of the legacy generator for the `vllm` model
@@ -67,7 +67,7 @@ def vllm_params_adapter(
 
 
 def vllm(model_name: str, **vllm_model_params):
-    """Load a vLLM model.
+    """Load an offline vLLM model.
 
     Parameters
     ---------
@@ -81,13 +81,14 @@ def vllm(model_name: str, **vllm_model_params):
     warnings.warn("""
         The `vllm` function is deprecated starting from v1.0.0.
         Do not use it. Support for it will be removed in v1.5.0.
-        Instead, you should instantiate a `VLLM` model with the
-        `outlines.from_vllm` function that takes a vLLM model name as argument.
+        Instead, you should instantiate a `VLLMOffline` model with the
+        `outlines.from_vllm_offline` function that takes a vLLM model name as
+        argument.
         For example:
         ```python
         from vllm import LLM
-        from outlines import from_vllm
-        model = from_vllm(LLM("<model_name>"))
+        from outlines import from_vllm_offline
+        model = from_vllm_offline(LLM("<model_name>"))
         ```
         """,
         DeprecationWarning,
@@ -98,4 +99,4 @@ def vllm(model_name: str, **vllm_model_params):
 
     model = LLM(model_name, **vllm_model_params)
 
-    return VLLM(model)
+    return VLLMOffline(model)
diff --git a/pyproject.toml b/pyproject.toml
index 70ebfe0ff..2fdcbf5f5 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -175,7 +175,7 @@ omit = [
     "outlines/models/gemini.py",
     "outlines/models/mlxlm.py",
     "outlines/models/openai.py",
-    "outlines/models/vllm.py",
+    "outlines/models/vllm_offline.py",
     "outlines/processors/tensor_adapters/mlx.py",
     "outlines/serve/serve.py",
     "outlines/v0_legacy/models/anthropic.py",
@@ -184,7 +184,7 @@ omit = [
     "outlines/v0_legacy/models/gemini.py",
     "outlines/v0_legacy/models/mlxlm.py",
     "outlines/v0_legacy/models/openai.py",
-    "outlines/v0_legacy/models/vllm.py",
+    "outlines/v0_legacy/models/vllm_offline.py",
     "tests/*",
 ]
 branch = true
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index 4efb93f4c..7e1f72696 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -1,157 +1,215 @@
+import os
 import re
-from enum import Enum
+import warnings
+from typing import AsyncGenerator, Generator
 
 import pytest
-from pydantic import BaseModel
-
-try:
-    from vllm import LLM, SamplingParams
-    HAS_VLLM = True
-except ImportError:
-    HAS_VLLM = False
-
-import outlines
-from outlines.models.vllm import (
-    VLLM,
-    VLLMTypeAdapter,
-    from_vllm
-)
-from outlines.types import Regex
+from openai import AsyncOpenAI, OpenAI
 
+from outlines.models.vllm import VLLM, AsyncVLLM, from_vllm
+from outlines.types.dsl import CFG, Regex, JsonSchema
+from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
 
-TEST_MODEL = "erwanf/gpt2-mini"
 
-pytestmark = pytest.mark.skipif(
-    not HAS_VLLM,
-    reason="vLLM models can only be run on GPU."
-)
+YES_NO_GRAMMAR = """
+?start: answer
+
+answer: "yes" | "no"
+"""
 
 
-def test_vllm_model_initialization():
-    model = from_vllm(LLM(TEST_MODEL))
+# If the VLLM_SERVER_URL environment variable is set, use the real vLLM server
+# Otherwise, use the mock server
+vllm_server_url = os.environ.get("VLLM_SERVER_URL")
+vllm_model_name = os.environ.get(
+    "VLLM_MODEL_NAME", "microsoft/Phi-3-mini-4k-instruct"
+)
+if vllm_server_url:
+    openai_client = OpenAI(base_url=vllm_server_url)
+    async_openai_client = AsyncOpenAI(base_url=vllm_server_url)
+else:
+    warnings.warn("No VLLM server URL provided, using mock server")
+    openai_client = MockOpenAIClient()
+    async_openai_client = MockAsyncOpenAIClient()
+
+mock_responses = [
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': vllm_model_name,
+        },
+        "foo"
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': vllm_model_name,
+            'stream': True
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'n': 2,
+            'model': vllm_model_name,
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+            'guided_json': {
+                'type': 'object',
+                'properties': {
+                    'bar': {'type': 'string'}
+                }
+            },
+            }
+        },
+        '{"foo": "bar"}'
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'guided_regex': '([0-9]{3})',
+            },
+        },
+        "123"
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'guided_grammar': YES_NO_GRAMMAR,
+            },
+        },
+        "yes"
+    ),
+]
+
+
+# If the VLLM_SERVER_URL environment variable is not set, add the mock
+# responses to the mock clients
+if not vllm_server_url:
+    async_openai_client.add_mock_responses(mock_responses)
+    openai_client.add_mock_responses(mock_responses)
+
+
+@pytest.fixture
+def sync_model():
+    return VLLM(openai_client, model_name=vllm_model_name)
+
+
+@pytest.fixture
+def async_model():
+    return AsyncVLLM(async_openai_client, model_name=vllm_model_name)
+
+
+def test_vllm_init():
+    model = from_vllm(OpenAI(base_url="http://localhost:11434"))
     assert isinstance(model, VLLM)
-    assert isinstance(model.model, LLM)
-    assert hasattr(model, "tokenizer")
-    assert isinstance(model.type_adapter, VLLMTypeAdapter)
-    assert model.tensor_library_name == "torch"
 
+    model = from_vllm(AsyncOpenAI(base_url="http://localhost:11434"))
+    assert isinstance(model, AsyncVLLM)
 
-@pytest.fixture(scope="session")
-def model(tmp_path_factory):
-    model = outlines.from_vllm(LLM(TEST_MODEL))
-    return model
+    with pytest.raises(ValueError, match="Unsupported client type"):
+        from_vllm("foo")
 
 
-def test_vllm_simple(model):
-    result = model.generate("Respond with one word. Not more.", None)
+def test_vllm_sync_simple_call(sync_model):
+    result = sync_model("Respond with a single word.",)
     assert isinstance(result, str)
 
 
-def test_vllm_call(model):
-    result = model("Respond with one word. Not more.")
-    assert isinstance(result, str)
+def test_vllm_sync_streaming(sync_model):
+    result = sync_model.stream("Respond with a single word.")
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
 
 
-def test_vllm_inference_kwargs(model):
-    result = model("Write a short story about a cat.", sampling_params=SamplingParams(max_tokens=2), use_tqdm=True)
-    assert isinstance(result, str)
-    assert len(result) <= 20
+def test_vllm_sync_multiple_samples(sync_model):
+    result = sync_model("Respond with a single word.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
 
 
-def test_vllm_invalid_inference_kwargs(model):
-    with pytest.raises(TypeError):
-        model("Respond with one word. Not more.", foo="bar")
+def test_vllm_sync_json(sync_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}}'
+    result = sync_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
 
 
-def test_vllm_regex(model):
-    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+def test_vllm_sync_regex(sync_model):
+    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
     assert isinstance(result, str)
-    assert re.match(r"[0-9]", result)
+    assert re.match(r"[0-9]{3}", result)
 
 
-def test_vllm_json(model):
-    class Character(BaseModel):
-        name: str
+def test_vllm_sync_cfg(sync_model):
+    result = sync_model("foo?", CFG(YES_NO_GRAMMAR), max_tokens=10)
+    assert isinstance(result, str)
+    assert result in ["yes", "no"]
 
-    result = model("Create a character with a name.", Character)
-    assert "name" in result
 
+@pytest.mark.asyncio
+async def test_vllm_async_simple_call(async_model):
+    result = await async_model("Respond with a single word.",)
+    assert isinstance(result, str)
 
-def test_vllm_choice(model):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
 
-    result = model("Cat or dog?", Foo)
-    assert result in ["cat", "dog"]
+@pytest.mark.asyncio
+async def test_vllm_async_streaming(async_model):
+    result = async_model.stream("Respond with a single word.")
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
 
 
-def test_vllm_batch_samples(model):
-    result = model(
-        "Respond with one word. Not more.",
-        sampling_params=SamplingParams(n=2)
-    )
+@pytest.mark.asyncio
+async def test_vllm_async_multiple_samples(async_model):
+    result = await async_model("Respond with a single word.", n=2)
     assert isinstance(result, list)
     assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
 
-    result = model(
-        ["Respond with one word. Not more.", "Respond with one word. Not more."]
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
 
-    result = model(
-        ["Respond with one word. Not more.", "Respond with one word. Not more."],
-        sampling_params=SamplingParams(n=2)
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
+@pytest.mark.asyncio
+async def test_vllm_async_json(async_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}}'
+    result = await async_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
 
 
-def test_vllm_batch_samples_constrained(model):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
+@pytest.mark.asyncio
+async def test_vllm_async_regex(async_model):
+    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
 
-    result = model(
-        "Cat or dog?",
-        Foo,
-        sampling_params=SamplingParams(n=2)
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert result[0] in ["cat", "dog"]
-    assert result[1] in ["cat", "dog"]
 
-    result = model(
-        ["Cat or dog?", "Cat or dog?"],
-        Foo,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert result[0] in ["cat", "dog"]
-    assert result[1] in ["cat", "dog"]
-
-    result = model(
-        ["Cat or dog?", "Cat or dog?"],
-        Foo,
-        sampling_params=SamplingParams(n=2)
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
-        assert item[0] in ["cat", "dog"]
-        assert item[1] in ["cat", "dog"]
-
-
-def test_vllm_streaming(model):
-    with pytest.raises(
-        NotImplementedError,
-        match="Streaming is not available for the vLLM integration."
-    ):
-        model.stream("Respond with one word. Not more.")
+@pytest.mark.asyncio
+async def test_vllm_async_cfg(async_model):
+    result = await async_model("foo?", CFG(YES_NO_GRAMMAR), max_tokens=10)
+    assert isinstance(result, str)
+    assert result in ["yes", "no"]
diff --git a/tests/models/test_vllm_offline.py b/tests/models/test_vllm_offline.py
new file mode 100644
index 000000000..21cbe8041
--- /dev/null
+++ b/tests/models/test_vllm_offline.py
@@ -0,0 +1,157 @@
+import re
+from enum import Enum
+
+import pytest
+from pydantic import BaseModel
+
+try:
+    from vllm import LLM, SamplingParams
+    HAS_VLLM = True
+except ImportError:
+    HAS_VLLM = False
+
+import outlines
+from outlines.models.vllm_offline import (
+    VLLMOffline,
+    VLLMOfflineTypeAdapter,
+    from_vllm_offline
+)
+from outlines.types import Regex
+
+
+TEST_MODEL = "erwanf/gpt2-mini"
+
+pytestmark = pytest.mark.skipif(
+    not HAS_VLLM,
+    reason="vLLM models can only be run on GPU."
+)
+
+
+def test_vllm_model_initialization():
+    model = from_vllm_offline(LLM(TEST_MODEL))
+    assert isinstance(model, VLLMOffline)
+    assert isinstance(model.model, LLM)
+    assert hasattr(model, "tokenizer")
+    assert isinstance(model.type_adapter, VLLMOfflineTypeAdapter)
+    assert model.tensor_library_name == "torch"
+
+
+@pytest.fixture(scope="session")
+def model(tmp_path_factory):
+    model = outlines.from_vllm_offline(LLM(TEST_MODEL))
+    return model
+
+
+def test_vllm_simple(model):
+    result = model.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+
+def test_vllm_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+def test_vllm_inference_kwargs(model):
+    result = model("Write a short story about a cat.", sampling_params=SamplingParams(max_tokens=2), use_tqdm=True)
+    assert isinstance(result, str)
+    assert len(result) <= 20
+
+
+def test_vllm_invalid_inference_kwargs(model):
+    with pytest.raises(TypeError):
+        model("Respond with one word. Not more.", foo="bar")
+
+
+def test_vllm_regex(model):
+    result = model("Give a number between 0 and 9.", Regex(r"[0-9]"))
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]", result)
+
+
+def test_vllm_json(model):
+    class Character(BaseModel):
+        name: str
+
+    result = model("Create a character with a name.", Character)
+    assert "name" in result
+
+
+def test_vllm_choice(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model("Cat or dog?", Foo)
+    assert result in ["cat", "dog"]
+
+
+def test_vllm_batch_samples(model):
+    result = model(
+        "Respond with one word. Not more.",
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."]
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
+    result = model(
+        ["Respond with one word. Not more.", "Respond with one word. Not more."],
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+
+
+def test_vllm_batch_samples_constrained(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model(
+        "Cat or dog?",
+        Foo,
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Foo,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert result[0] in ["cat", "dog"]
+    assert result[1] in ["cat", "dog"]
+
+    result = model(
+        ["Cat or dog?", "Cat or dog?"],
+        Foo,
+        sampling_params=SamplingParams(n=2)
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    for item in result:
+        assert isinstance(item, list)
+        assert len(item) == 2
+        assert item[0] in ["cat", "dog"]
+        assert item[1] in ["cat", "dog"]
+
+
+def test_vllm_streaming(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="Streaming is not available for the vLLM integration."
+    ):
+        model.stream("Respond with one word. Not more.")
diff --git a/tests/models/test_vllm_offline_type_adapter.py b/tests/models/test_vllm_offline_type_adapter.py
new file mode 100644
index 000000000..c1480dbed
--- /dev/null
+++ b/tests/models/test_vllm_offline_type_adapter.py
@@ -0,0 +1,37 @@
+import pytest
+
+import transformers
+
+import outlines
+from outlines.models.vllm_offline import VLLMOfflineTypeAdapter
+from outlines.processors.structured import RegexLogitsProcessor
+
+
+@pytest.fixture
+def type_adapter():
+    return VLLMOfflineTypeAdapter()
+
+
+@pytest.fixture
+def logits_processor():
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+    return RegexLogitsProcessor(
+        regex_string=r"[a-z]+",
+        tokenizer=model.tokenizer,
+        tensor_library_name="torch",
+    )
+
+
+def test_vllm_offline_type_adapter_format_input(type_adapter):
+    assert type_adapter.format_input("foo") == "foo"
+    assert type_adapter.format_input(["foo", "bar"]) == ["foo", "bar"]
+    with pytest.raises(NotImplementedError):
+        type_adapter.format_input({"foo": "bar"})
+
+
+def test_vllm_offline_type_adapter_format_output_type(type_adapter, logits_processor):
+    assert type_adapter.format_output_type(None) == []
+    assert type_adapter.format_output_type(logits_processor) == [logits_processor]
diff --git a/tests/models/test_vllm_type_adapter.py b/tests/models/test_vllm_type_adapter.py
new file mode 100644
index 000000000..46d021a8c
--- /dev/null
+++ b/tests/models/test_vllm_type_adapter.py
@@ -0,0 +1,121 @@
+import io
+import json
+import pytest
+from dataclasses import dataclass
+
+from PIL import Image
+from outlines.models.vllm import VLLMTypeAdapter
+from outlines.types import CFG, JsonSchema
+
+from outlines import Vision
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
+
+
+@pytest.fixture
+def type_adapter():
+    return VLLMTypeAdapter()
+
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
+
+@pytest.fixture
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+def test_vllm_type_adapter_input_text(type_adapter):
+    message = "prompt"
+    result = type_adapter.format_input(message)
+    assert isinstance(result, dict)
+    assert len(result["messages"]) == 1
+    assert result["messages"][0]["role"] == "user"
+    assert result["messages"][0]["content"] == message
+
+
+def test_vllm_type_adapter_input_vision(type_adapter, image):
+    input_message = Vision("hello", image)
+    result = type_adapter.format_input(input_message)
+    assert isinstance(result, dict)
+
+    messages = result["messages"]
+    assert len(messages) == 1
+
+    message = messages[0]
+    assert message["role"] == "user"
+    assert len(message["content"]) == 2
+    assert message["content"][0]["type"] == "text"
+    assert message["content"][0]["text"] == "hello"
+
+    assert message["content"][1]["type"] == "image_url"
+    assert (
+        message["content"][1]["image_url"]["url"]
+        == f"data:image/png;base64,{input_message.image_str}"
+    )
+
+
+def test_vllm_type_adapter_input_invalid(type_adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = type_adapter.format_input(prompt)
+
+
+def test_vllm_type_adapter_output_type(
+    type_adapter,
+    cfg_instance,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    assert type_adapter.format_output_type(cfg_instance) == {
+        "guided_grammar": CFG_STRING
+    }
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "guided_json": json.loads(JSON_SCHEMA_STRING)
+    }
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "guided_json": json.loads(JSON_SCHEMA_STRING),
+        "whitespace_pattern": "\n"
+    }
+    assert type_adapter.format_output_type(int) == {
+        "guided_regex": "([+-]?(0|[1-9][0-9]*))"
+    }
diff --git a/tests/processors/test_guide.py b/tests/processors/test_guide.py
index 53503c63e..3eaae2e32 100644
--- a/tests/processors/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -224,7 +224,7 @@ def test_regex_guide_caching():
             *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
         ))
     if HAS_VLLM:
-        models.append(outlines.from_vllm(vllm.LLM("TinyLlama/TinyLlama-1.1B-Chat-v1.0")))
+        models.append(outlines.from_vllm_offline(vllm.LLM("TinyLlama/TinyLlama-1.1B-Chat-v1.0")))
 
     for i, model in enumerate(models):
         # First call for each model should be a miss
diff --git a/tests/test_generator.py b/tests/test_generator.py
index eae1d1632..824463c93 100644
--- a/tests/test_generator.py
+++ b/tests/test_generator.py
@@ -1,19 +1,61 @@
-import os
 import pytest
-from typing import Literal, Generator as TypingGenerator
+from typing import AsyncGenerator, Generator as TypingGenerator, Literal
 
 import interegular
 import transformers
-from openai import OpenAI
 
 import outlines
-from outlines.generator import BlackBoxGenerator, SteerableGenerator, Generator
-from outlines.models import BlackBoxModel, SteerableModel
+from outlines.generator import (
+    BlackBoxGenerator,
+    SteerableGenerator,
+    Generator,
+    AsyncBlackBoxGenerator,
+)
+from outlines.models import AsyncVLLM, VLLM
 from outlines.processors import (
     OutlinesLogitsProcessor,
     RegexLogitsProcessor,
 )
 from outlines.types import CFG, FSM
+from tests.test_utils.mock_openai_client import (
+    MockAsyncOpenAIClient,
+    MockOpenAIClient,
+)
+
+
+MODEL_NAME = "microsoft/Phi-3-mini-4k-instruct"
+
+
+# We used the mocked vllm model to test the black box generator
+async_openai_client = MockAsyncOpenAIClient()
+openai_client = MockOpenAIClient()
+mock_responses = [
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Write a very short sentence'}
+            ],
+            'model': MODEL_NAME,
+            'max_tokens': 10,
+            'extra_body': {'guided_regex': '("[^"]*")'},
+        },
+        "Mock response"
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Write a very short sentence'}
+            ],
+            'model': MODEL_NAME,
+            'max_tokens': 10,
+            'extra_body': {'guided_regex': '("[^"]*")'},
+            'stream': True,
+        },
+        ["Mock", "response"]
+    ),
+]
+async_openai_client.add_mock_responses(mock_responses)
+openai_client.add_mock_responses(mock_responses)
 
 
 @pytest.fixture(scope="session")
@@ -39,14 +81,14 @@ def sample_processor():
     return processor
 
 
-@pytest.fixture(scope="session")
-def black_box_model():
-    api_key = os.getenv("OPENAI_API_KEY")
-    if not api_key:
-        api_key = "MOCK_API_KEY"
-    client = OpenAI(api_key=api_key)
-    model = outlines.from_openai(client, model_name="gpt-4o-mini")
-    return model
+@pytest.fixture(scope="module")
+def black_box_sync_model():
+    return VLLM(openai_client, MODEL_NAME)
+
+
+@pytest.fixture(scope="module")
+def black_box_async_model():
+    return AsyncVLLM(async_openai_client, MODEL_NAME)
 
 
 # SteerableGenerator
@@ -106,35 +148,62 @@ def test_steerable_generator_stream(steerable_model):
 # BlackBoxGenerator
 
 
-def test_black_box_generator_init_valid(black_box_model):
-    generator = BlackBoxGenerator(black_box_model, Literal["foo", "bar"])
-    assert generator.model == black_box_model
+def test_black_box_generator_init_valid(black_box_sync_model):
+    generator = BlackBoxGenerator(black_box_sync_model, Literal["foo", "bar"])
+    assert generator.model == black_box_sync_model
     assert generator.output_type == Literal["foo", "bar"]
 
 
-def test_black_box_generator_init_invalid(black_box_model):
+def test_black_box_generator_init_invalid(black_box_sync_model):
     with pytest.raises(NotImplementedError):
-        BlackBoxGenerator(black_box_model, CFG("foo"))
-
-    with pytest.raises(NotImplementedError):
-        BlackBoxGenerator(black_box_model, FSM("foo"))
+        BlackBoxGenerator(black_box_sync_model, FSM("foo"))
 
 
-@pytest.mark.api_call
-def test_black_box_generator_call(black_box_model):
-    generator = BlackBoxGenerator(black_box_model, dict)
-    result = generator("write a very short json object", max_tokens=10)
+def test_black_box_generator_call(black_box_sync_model):
+    generator = BlackBoxGenerator(black_box_sync_model, str)
+    result = generator("Write a very short sentence", max_tokens=10)
     assert isinstance(result, str)
 
 
-@pytest.mark.api_call
-def test_black_box_generator_stream(black_box_model):
-    generator = BlackBoxGenerator(black_box_model, dict)
-    result = generator.stream("write a very short json object", max_tokens=10)
+def test_black_box_generator_stream(black_box_sync_model):
+    generator = BlackBoxGenerator(black_box_sync_model, str)
+    result = generator.stream("Write a very short sentence", max_tokens=10)
     assert isinstance(result, TypingGenerator)
     assert isinstance(next(result), str)
 
 
+
+# AsyncBlackBoxGenerator
+
+
+def test_async_black_box_generator_init_valid(black_box_async_model):
+    generator = AsyncBlackBoxGenerator(black_box_async_model, Literal["foo", "bar"])
+    assert generator.model == black_box_async_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_async_black_box_generator_init_invalid(black_box_async_model):
+    with pytest.raises(NotImplementedError):
+        AsyncBlackBoxGenerator(black_box_async_model, FSM("foo"))
+
+
+@pytest.mark.asyncio
+async def test_async_black_box_generator_call(black_box_async_model):
+    generator = AsyncBlackBoxGenerator(black_box_async_model, str)
+    result = await generator("Write a very short sentence", max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_async_black_box_generator_stream(black_box_async_model):
+    generator = AsyncBlackBoxGenerator(black_box_async_model, str)
+    result = generator.stream("Write a very short sentence", max_tokens=10)
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
 # Generator
 
 
@@ -157,13 +226,25 @@ def test_generator_steerable_processor(steerable_model, sample_processor):
     assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
 
 
-def test_generator_black_box_output_type(black_box_model):
-    generator = Generator(black_box_model, Literal["foo", "bar"])
+def test_generator_black_box_sync_output_type(black_box_sync_model):
+    generator = Generator(black_box_sync_model, Literal["foo", "bar"])
     assert isinstance(generator, BlackBoxGenerator)
-    assert generator.model == black_box_model
+    assert generator.model == black_box_sync_model
+    assert generator.output_type == Literal["foo", "bar"]
+
+
+def test_generator_black_box_sync_processor(black_box_sync_model, sample_processor):
+    with pytest.raises(NotImplementedError):
+        Generator(black_box_sync_model, processor=sample_processor)
+
+
+def test_generator_black_box_async_output_type(black_box_async_model):
+    generator = Generator(black_box_async_model, Literal["foo", "bar"])
+    assert isinstance(generator, AsyncBlackBoxGenerator)
+    assert generator.model == black_box_async_model
     assert generator.output_type == Literal["foo", "bar"]
 
 
-def test_generator_black_box_processor(black_box_model, sample_processor):
+def test_generator_black_box_async_processor(black_box_async_model, sample_processor):
     with pytest.raises(NotImplementedError):
-        Generator(black_box_model, processor=sample_processor)
+        Generator(black_box_async_model, processor=sample_processor)
diff --git a/tests/test_utils/mock_openai_client.py b/tests/test_utils/mock_openai_client.py
new file mode 100644
index 000000000..635a14ae3
--- /dev/null
+++ b/tests/test_utils/mock_openai_client.py
@@ -0,0 +1,125 @@
+import hashlib
+import pickle
+from typing import List, Dict, Any, Optional
+from unittest.mock import MagicMock
+
+
+def hash_dict(d) -> str:
+    def make_hashable(obj):
+        if isinstance(obj, dict):
+            return frozenset((k, make_hashable(v)) for k, v in sorted(obj.items()))
+        if isinstance(obj, list):
+            return tuple(make_hashable(e) for e in obj)
+        return obj
+
+    hashable_obj = make_hashable(d)
+    pickled_obj = pickle.dumps(hashable_obj)
+    return hashlib.sha256(pickled_obj).hexdigest()
+
+
+class MockChoice:
+    def __init__(self, content: str, finish_reason: str = "stop", refusal: Optional[str] = None):
+        self.message = MagicMock()
+        self.message.content = content
+        self.message.refusal = refusal
+        self.finish_reason = finish_reason
+        self.delta = MagicMock()
+        self.delta.content = content
+
+
+class MockCompletionResponse:
+    def __init__(self, choices: List[MockChoice]):
+        self.choices = choices
+
+
+class MockStreamingChunk:
+    def __init__(self, content: Optional[str] = None):
+        self.choices = []
+        if content is not None:
+            choice = MagicMock()
+            delta = MagicMock()
+            delta.content = content
+            choice.delta = delta
+            self.choices = [choice]
+
+
+class MockOpenAIClient:
+    """Mock for OpenAI client that can be used to test vLLM integration"""
+
+    def __init__(self):
+        self.chat = MagicMock()
+        self.chat.completions = MagicMock()
+        self.chat.completions.create = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        def _create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return self._create_streaming_response(response)
+            else:
+                return self._create_standard_response(response)
+
+        self.chat.completions.create.side_effect = _create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+    def _create_standard_response(self, response):
+        if isinstance(response, str):
+            response = [response]
+        choices = [MockChoice(content=chunk) for chunk in response]
+        return MockCompletionResponse(choices=choices)
+
+    def _create_streaming_response(self, response):
+        chunks = [MockStreamingChunk(content=chunk) for chunk in response]
+        return iter(chunks)
+
+
+class MockAsyncOpenAIClient:
+    """Mock for AsyncOpenAI client that can be used to test AsyncVLLM integration"""
+
+    def __init__(self):
+        self.chat = MagicMock()
+        self.chat.completions = MagicMock()
+        self.chat.completions.create = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        async def _async_create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return self._create_async_streaming_response(response)
+            else:
+                return await self._create_async_standard_response(response)
+
+        self.chat.completions.create.side_effect = _async_create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+    async def _create_async_standard_response(self, response):
+        """Create an async standard (non-streaming) response"""
+        if isinstance(response, str):
+            response = [response]
+        choices = [MockChoice(content=chunk) for chunk in response]
+        return MockCompletionResponse(choices=choices)
+
+    async def _create_async_streaming_response(self, response):
+        """Create an async streaming response generator"""
+        chunks = [MockStreamingChunk(content=chunk) for chunk in response]
+
+        for chunk in chunks:
+            yield chunk
diff --git a/tests/v0_legacy/models/test_vllm_legacy.py b/tests/v0_legacy/models/test_vllm_offline_legacy.py
similarity index 86%
rename from tests/v0_legacy/models/test_vllm_legacy.py
rename to tests/v0_legacy/models/test_vllm_offline_legacy.py
index 97eed996c..318eeddb7 100644
--- a/tests/v0_legacy/models/test_vllm_legacy.py
+++ b/tests/v0_legacy/models/test_vllm_offline_legacy.py
@@ -12,7 +12,7 @@
 
 from outlines import generate, models, samplers
 from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-from outlines.v0_legacy.models.vllm import vllm_params_adapter
+from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
 
 pytestmark = pytest.mark.skipif(
     not HAS_VLLM,
@@ -40,17 +40,17 @@ def test_vllm_legacy_init(model):
 
 def test_vllm_legacy_params_adapter_generation_params():
     # default
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial().sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {}
     )
-    assert result["sampling_params"].get("stop") is None
-    assert result["sampling_params"].get("seed") is None
-    assert result["sampling_params"].get("max_tokens") is None
+    assert result["sampling_params"].stop_strings is None
+    assert result["sampling_params"].seed is None
+    assert result["sampling_params"].max_tokens is None
 
     # max_tokens
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial().sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": 2},
         {}
@@ -58,7 +58,7 @@ def test_vllm_legacy_params_adapter_generation_params():
     assert result["sampling_params"].max_tokens == 2
 
     # seed
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial().sampling_params),
         {"stop_at": None, "seed": 2, "max_tokens": None},
         {}
@@ -66,7 +66,7 @@ def test_vllm_legacy_params_adapter_generation_params():
     assert result["sampling_params"].seed == 2
 
     # stop_at str
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial().sampling_params),
         {"stop_at": "foo", "seed": None, "max_tokens": None},
         {}
@@ -74,7 +74,7 @@ def test_vllm_legacy_params_adapter_generation_params():
     assert result["sampling_params"].stop_strings == ["foo"]
 
     # stop_at list
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial().sampling_params),
         {"stop_at": ["foo", "bar"], "seed": None, "max_tokens": None},
         {}
@@ -84,7 +84,7 @@ def test_vllm_legacy_params_adapter_generation_params():
 
 def test_vllm_legacy_params_adapter_sampling_params():
     # default
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial().sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {}
@@ -95,7 +95,7 @@ def test_vllm_legacy_params_adapter_sampling_params():
     assert result["sampling_params"].temperature == 1.0
 
     # beam search
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.beam_search().sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {}
@@ -103,7 +103,7 @@ def test_vllm_legacy_params_adapter_sampling_params():
     assert result["sampling_params"].use_beam_search is True
 
     # num_samples
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial(samples=2).sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {}
@@ -112,7 +112,7 @@ def test_vllm_legacy_params_adapter_sampling_params():
     assert result["sampling_params"].best_of == 2
 
     # top_p
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial(top_p=0.5).sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {}
@@ -120,7 +120,7 @@ def test_vllm_legacy_params_adapter_sampling_params():
     assert result["sampling_params"].top_p == 0.5
 
     # top_k
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial(top_k=2).sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {}
@@ -128,7 +128,7 @@ def test_vllm_legacy_params_adapter_sampling_params():
     assert result["sampling_params"].top_k == 2
 
     # temperature
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial(temperature=0.5).sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {}
@@ -137,7 +137,7 @@ def test_vllm_legacy_params_adapter_sampling_params():
 
 
 def test_vllm_legacy_params_adapter_model_specific_params():
-    result = vllm_params_adapter(
+    result = vllm_offline_params_adapter(
         asdict(samplers.multinomial().sampling_params),
         {"stop_at": None, "seed": None, "max_tokens": None},
         {"use_tqdm": False}

From f17800d6822cdd1d2db2e7365e5a5f0ff0dd42ce Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 16 Apr 2025 10:16:45 +0200
Subject: [PATCH 318/505] Create the migration guide to v1

---
 docs/content/docs/guides/migration_guide.md | 244 ++++++++++++++++++++
 1 file changed, 244 insertions(+)
 create mode 100644 docs/content/docs/guides/migration_guide.md

diff --git a/docs/content/docs/guides/migration_guide.md b/docs/content/docs/guides/migration_guide.md
new file mode 100644
index 000000000..f931641f7
--- /dev/null
+++ b/docs/content/docs/guides/migration_guide.md
@@ -0,0 +1,244 @@
+# Outlines 1.0 migration guide
+
+Outlines 1.0 introduces some breaking changes that affect the way you use the library. You are likely concerned by all of the following sections, so please read this document carefully until the end.
+
+This guide will help you migrate your code to the new version.
+
+All previous functionality will be supported until Outlines version 1.5.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
+
+## Removed or modified features
+- [Generate functions](#generate-functions)
+- [Models](#models)
+- [Samplers](#samplers)
+- [Functions](#functions)
+- [Text generation return types](#text-generation-return-types)
+- [Inference arguments](#inference-arguments)
+
+### Generate functions
+
+The whole `generate` module has been removed. That includes the functions `generate.cfg`, `generate.choice`, `generate.format`,`generate.fsm`, `generate.json`, `generate.regex` and `generate.text`.
+
+You should replace these functions by the `Generator` object along with the right output type as an argument (on top of the model). The output type can either be a python type or be an object from the `outlines.types` module. You can find more information about the output types in the [output types](TODO: link to the documentatoin on output types) guide.
+
+Associated v1 output types for each deprecated function:
+- `generate.cfg` -> `outlines.types.CFG`
+- `generate.choice` -> `typing.Literal` or `typing.Union`
+- `generate.format` -> native python types (`str`, `int` etc.)
+- `generate.fsm` -> `outlines.types.FSM`
+- `generate.json` -> `pydantic.BaseModel`, `typing.TypedDict`, `dataclasses.dataclass`, `genson.schema.SchemaBuilder` or `outlines.types.JsonSchema`
+- `generate.regex` -> `outlines.types.Regex`
+- `generate.text` -> no output type (`None`)
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = ...
+generator = generate.choice(model, ["foo", "bar"])
+```
+
+You should now use:
+
+```python
+from typing import Literal
+from outlines import Generator
+
+model = ...
+generator = Generator(model, Literal["foo", "bar"])
+```
+
+### Models
+
+The model classes found in the `outlines.models` module are maintained but there are a few important changes to be aware of.
+
+The functions used to created a model have been replaced by equivalent functions named with a `from_` prefix. The function `outlines.models.transformers` has been replaced by `outlines.from_transformers` for instance. On top of this change of name, the arguments have been modified. You should refer to the [models documentation](TODO: link to the documentation on models) for more details, but the overall idea is that you now need to provide a model/client instance from the inference library the Outlines model is wrapping.
+
+For instance, instead of:
+
+```python
+from outlines import models
+
+model = models.llamacpp(
+    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+)
+```
+
+You should now do:
+
+```python
+from llama_cpp import Llama
+from outlines import from_llamacpp
+
+llamacpp_model = Llama.from_pretrained(
+    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+)
+model = from_llamacpp(llamacpp_model)
+```
+
+The `load_lora` methods that are present on the `VLLM` and `LlamaCpp` models have been removed. You should now handle lora loading through the `Llama` instance in the case of the `LlamaCpp` model or provide it as a keyword argument when calling the model in the case of the `VLLM` model.
+
+For instance, instead of:
+
+```python
+from outlines import from_vllm
+from vllm import LLM
+
+model = from_vllm(
+    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
+)
+model.load_lora("path/to/lora/file")
+
+response = model("foo")
+```
+
+You should now do:
+
+```python
+from outlines import from_vllm
+from vllm import LLM
+from vllm.lora.request import LoRARequest
+
+model = from_vllm(
+    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
+)
+lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
+
+response = model("foo", lora_request=lora_request)
+```
+
+The `ExLlamav2` model has been removed without replacement. This inference library is not fully compatible with Outlines, so we decided to remove it. You can still use it until final deprecation, but we recommend you to migrate to a different inference library right now.
+
+### Samplers
+
+The `outlines.samplers` module has been removed without replacement. You should now use the arguments of the inference library model to control the sampling. Depending on the model you use, this could be done at initialization or when calling the model to generate text (so when calling the outlines model or a generator).
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = <transformers_model>
+
+generator = generate.text(model, samplers.beam_search(2))
+response = generator("foo")
+```
+
+You should now do:
+
+```python
+from outlines import Generator
+
+model = <transformers_model>
+
+generator = Generator(model)
+response = generator("foo", num_beams=2)
+```
+
+### Functions
+
+The `outlines.function` module has been removed. It is replaced by the `outlines.applications` module. An `Application` serves a similar purpose as a `Function`: it encapsulates a prompt template and an output type. A difference is that can `Application` is not instantiated with a model name. Instead, you should provide a model instance along with the prompt when calling it.
+
+For instance, instead of:
+
+```python
+from outlines import Function
+
+prompt_template = ...
+output_type = ...
+
+fn = Function(
+    prompt_template,
+    output_type,
+    "hf-internal-testing/tiny-random-GPTJForCausalLM",
+)
+
+result = fn("foo")
+```
+
+You should now do:
+
+```python
+from outlines import Application
+
+prompt_template = ...
+output_type = ...
+
+application = Application(
+    prompt_template,
+    output_type,
+)
+
+model = ...
+
+result = application(model, "foo")
+```
+
+### Text generation return types
+
+In the previous version of Outlines, the return type of the generators depended on the output type provided. For instance, if you passed a Pydantic model to the `generate.json` function, the return type was a Pydantic model instance. In the v1, the return type of a generator is always a `str`, the raw text generated by the model. You are responsible for parsing the text into the desired format.
+
+For instance, instead of:
+
+```python
+from pydantic import BaseModel
+from outlines import generate
+
+class Foo(BaseModel):
+    bar: str
+
+model = ...
+
+generator = generate.json(model, Foo)
+
+result = generator("foo")
+print(result.bar)
+```
+
+You should now do:
+
+```python
+from pydantic import BaseModel
+from outlines import Generator
+
+class Foo(BaseModel):
+    bar: str
+
+model = ...
+
+generator = Generator(model, Foo)
+
+result = generator("foo")
+result = Foo.model_validate_json(result) # parse the text into the Pydantic model instance
+print(result.bar)
+```
+
+### Inference arguments
+
+In the previous version of Outlines, some of the inference arguments were standardized across the models and were provided as positional arguments to the generator or through the sampling params dictionary. Additionally, various default values were added by outlines to the inference library models. This is no longer the case. You should refer to the documentation of the inference library you use to find the right arguments for your use case and pass them as keyword arguments to the outlines generator when calling it.
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = <transformers_model>
+
+generator = generate.text(model)
+
+result = generator("foo", 256, ".", 10) # 256 tokens, stop at "." and seed 10
+```
+
+You should now do:
+
+```python
+from outlines import Generator
+
+model = <transformers_model>
+
+generator = Generator(model)
+
+result = generator("foo", max_new_tokens=256, stop_strings=".", seed=10)
+```

From b2d3e00bf7599977efc3dae418d9057b57dd3f3f Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 30 Apr 2025 10:29:37 +0200
Subject: [PATCH 319/505] Create the TGI model

---
 docs/reference/models/tgi.md           |  51 ++++-
 outlines/__init__.py                   |   2 +
 outlines/models/__init__.py            |  13 +-
 outlines/models/tgi.py                 | 250 +++++++++++++++++++++++++
 tests/models/test_tgi.py               | 180 ++++++++++++++++++
 tests/models/test_tgi_model_adapter.py |  93 +++++++++
 tests/test_utils/mock_openai_client.py |  15 +-
 tests/test_utils/mock_tgi_client.py    |  63 +++++++
 tests/test_utils/utils.py              |  23 +++
 9 files changed, 673 insertions(+), 17 deletions(-)
 create mode 100644 outlines/models/tgi.py
 create mode 100644 tests/models/test_tgi.py
 create mode 100644 tests/models/test_tgi_model_adapter.py
 create mode 100644 tests/test_utils/mock_tgi_client.py
 create mode 100644 tests/test_utils/utils.py

diff --git a/docs/reference/models/tgi.md b/docs/reference/models/tgi.md
index 3f0568103..ed1207449 100644
--- a/docs/reference/models/tgi.md
+++ b/docs/reference/models/tgi.md
@@ -1,3 +1,50 @@
-# Text-generation-inference (TGI)
+# TGI
 
-TGI uses Outlines to provide structured generation, see [their documentation](https://huggingface.co/docs/text-generation-inference/en/basic_tutorials/using_guidance).
+## Prerequisites
+
+The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed.
+
+## Initialize the model
+
+    To load the model, you can use the `from_tgi` function. The argument of the function is an instance of `InferenceClient` or `AsyncInferenceClient` from the `huggingface_hub` package. You will respectively receive a `TGI` or `AsyncTGI` model instance.
+
+```python
+import outlines
+from huggingface_hub import InferenceClient, AsyncInferenceClient
+
+sync_model = outlines.from_tgi(InferenceClient("http://localhost:8080"))
+print(type(sync_model))  # <class 'outlines.models.tgi.TGI'>
+
+async_model = outlines.from_tgi(AsyncInferenceClient("http://localhost:8080"))
+print(type(async_model))  # <class 'outlines.models.tgi.AsyncTGI'>
+```
+
+## Generate text
+
+To generate text, you can call the model with a prompt as argument and optionally an output type to rely on structured generation:
+
+```python
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+
+answer = sync_model("Create a character.", output_type=Character)
+answer = await async_model("Create a character.", output_type=Character)
+```
+
+The `TGI` model also supports streaming:
+
+```python
+for chunk in sync_model.stream("Write a short story about a cat.", max_new_tokens=100):
+    print(chunk)
+```
+
+## Optional parameters
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. These parameters will be passed directly to the TGI client. Common parameters include `max_new_tokens`, `stream`, `temperature`, `top_k`, `top_p`, and others as specified in the [TGI inference client documentation][tgi-inference-client-docs].
+
+
+[tgi-docs]: https://huggingface.co/docs/text-generation-inference/en/index
+[tgi-inference-client-docs]: https://huggingface.co/docs/huggingface_hub/package_reference/inference_client
+[tgi-structured-output-docs]: https://huggingface.co/docs/text-generation-inference/en/basic_tutorials/using_guidance
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 47188c2b5..cc9062227 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -18,6 +18,7 @@
     from_ollama,
     from_llamacpp,
     from_mlxlm,
+    from_tgi,
     from_vllm,
     from_vllm_offline,
 )
@@ -41,6 +42,7 @@
     "from_mlxlm",
     "from_ollama",
     "from_openai",
+    "from_tgi",
     "from_transformers",
     "from_vllm",
 ]
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 22e755b23..ee51e2305 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -16,6 +16,7 @@
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
+from .tgi import from_tgi, TGI, AsyncTGI
 from .transformers import (
     Transformers,
     TransformerTokenizer,
@@ -30,4 +31,14 @@
 ]
 
 SteerableModel = Union[LlamaCpp, Transformers, MLXLM, VLLMOffline]
-BlackBoxModel = Union[OpenAI, Anthropic, Gemini, Ollama, Dottxt, AsyncVLLM, VLLM]
+BlackBoxModel = Union[
+    Anthropic,
+    AsyncTGI,
+    AsyncVLLM,
+    Dottxt,
+    Gemini,
+    Ollama,
+    OpenAI,
+    TGI,
+    VLLM,
+]
diff --git a/outlines/models/tgi.py b/outlines/models/tgi.py
new file mode 100644
index 000000000..3bd2e6e33
--- /dev/null
+++ b/outlines/models/tgi.py
@@ -0,0 +1,250 @@
+import json
+from functools import singledispatchmethod
+from typing import TYPE_CHECKING, Union
+
+from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
+from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
+
+
+if TYPE_CHECKING:
+    from huggingface_hub import AsyncInferenceClient, InferenceClient
+
+
+class TGITypeAdapter(ModelTypeAdapter):
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the client.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        raise NotImplementedError(
+            f"The input type {input} is not available with TGI. "
+            + "Please provide a string."
+        )
+
+    @format_input.register(str)
+    def format_str_input(self, model_input: str):
+        return model_input
+
+    @format_input.register(list)
+    def format_list_input(self, model_input: list):
+        raise NotImplementedError("TGI does not support batch inference.")
+
+    def format_output_type(self, output_type):
+        """Generate the structured output argument to pass to the client.
+
+        Argument
+        --------
+        output_type
+            The structured output type provided.
+
+        """
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            raise NotImplementedError(
+                "TGI does not support CFG-based structured outputs."
+            )
+        elif isinstance(term, JsonSchema):
+            return {
+                "grammar": {
+                    "type": "json",
+                    "value": json.loads(term.schema),
+                }
+            }
+        else:
+            return {
+                "grammar": {
+                    "type": "regex",
+                    "value": to_regex(term),
+                }
+            }
+
+
+class TGI(Model):
+    """Represents a client to a huggingface TGI server."""
+
+    def __init__(self, client):
+        """Create a `TGI` model instance.
+
+        Parameters
+        ----------
+        client
+            A huggingface `InferenceClient` client instance.
+
+        """
+        self.client = client
+        self.type_adapter = TGITypeAdapter()
+
+    def generate(self, model_input, output_type, **inference_kwargs):
+        """Generate text using the client.
+
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        The generated text.
+
+        """
+        client_args = self._build_client_args(
+            model_input,
+            output_type,
+            **inference_kwargs,
+        )
+
+        return self.client.text_generation(**client_args)
+
+    def generate_stream(self, model_input, output_type, **inference_kwargs):
+        """Return a text generator.
+
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        A generator that yields the generated text chunks.
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = self.client.text_generation(
+            **client_args, stream=True,
+        )
+
+        for chunk in stream:  # pragma: no cover
+            yield chunk
+
+    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+        """Build the arguments to pass to the TGI client."""
+        prompt = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+
+
+        client_args = {
+            "prompt": prompt,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+class AsyncTGI(AsyncModel):
+    """Represents an async client to a huggingface TGI server."""
+
+    def __init__(self, client):
+        """Create an `AsyncTGI` model instance.
+
+        Parameters
+        ----------
+        client
+            A huggingface `AsyncInferenceClient` client instance.
+
+        """
+        self.client = client
+        self.type_adapter = TGITypeAdapter()
+
+    async def generate(self, model_input, output_type, **inference_kwargs):
+        """Generate text using TGI.
+
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        The generated text.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        response = await self.client.text_generation(**client_args)
+
+        return response
+
+    async def generate_stream(
+        self, model_input, output_type, **inference_kwargs
+    ):
+        """Return a text generator.
+
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        A generator that yields the generated text chunks.
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = await self.client.text_generation(
+            **client_args, stream=True
+        )
+
+        async for chunk in stream:  # pragma: no cover
+            yield chunk
+
+    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+        """Build the arguments to pass to the TGI client."""
+        prompt = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+
+        client_args = {
+            "prompt": prompt,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+def from_tgi(
+    client: Union["InferenceClient", "AsyncInferenceClient"],
+) -> Union[TGI, AsyncTGI]:
+    from huggingface_hub import AsyncInferenceClient, InferenceClient
+
+    if isinstance(client, InferenceClient):
+        return TGI(client)
+    elif isinstance(client, AsyncInferenceClient):
+        return AsyncTGI(client)
+    else:
+        raise ValueError(
+            f"Unsupported client type: {type(client)}.\n"
+            + "Please provide an HuggingFace InferenceClient "
+            + "or AsyncInferenceClient instance."
+        )
diff --git a/tests/models/test_tgi.py b/tests/models/test_tgi.py
new file mode 100644
index 000000000..52704f6c0
--- /dev/null
+++ b/tests/models/test_tgi.py
@@ -0,0 +1,180 @@
+import os
+import re
+import warnings
+from typing import AsyncGenerator, Generator
+
+import pytest
+from huggingface_hub import InferenceClient, AsyncInferenceClient
+
+from outlines.models.tgi import TGI, AsyncTGI, from_tgi
+from outlines.types.dsl import CFG, Regex, JsonSchema
+from tests.test_utils.mock_tgi_client import MockTGIInferenceClient, MockAsyncTGIInferenceClient
+
+
+YES_NO_GRAMMAR = """
+?start: answer
+
+answer: "yes" | "no"
+"""
+
+# If the TGI_SERVER_URL environment variable is set, use the real TGI server
+# Otherwise, use the mock server
+tgi_server_url = os.environ.get("TGI_SERVER_URL")
+if tgi_server_url:
+    tgi_client = InferenceClient(tgi_server_url)
+    async_tgi_client = AsyncInferenceClient(tgi_server_url)
+else:
+    warnings.warn("No TGI server URL provided, using mock server")
+    tgi_client = MockTGIInferenceClient()
+    async_tgi_client = MockAsyncTGIInferenceClient()
+
+mock_responses = [
+    (
+        {
+            'prompt': 'Respond with a single word.',
+            'max_new_tokens': 10,
+        },
+        "foo"
+    ),
+    (
+        {
+            'prompt': 'Respond with a single word.',
+            'max_new_tokens': 10,
+            'stream': True
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'prompt': 'foo?',
+            'max_new_tokens': 10,
+            'grammar': {
+                'type': 'json',
+                'value': {
+                    'type': 'object',
+                    'properties': {
+                        'bar': {'type': 'string'}
+                    },
+                    'required': ['bar']
+                }
+            }
+        },
+        '{"foo": "bar"}'
+    ),
+    (
+        {
+            'prompt': 'foo?',
+            'max_new_tokens': 10,
+            'grammar': {
+                'type': 'regex',
+                'value': '([0-9]{3})',
+            },
+        },
+        "123"
+    ),
+]
+
+# If the TGI_SERVER_URL environment variable is not set, add the mock
+# responses to the mock clients
+if not tgi_server_url:
+    async_tgi_client.add_mock_responses(mock_responses)
+    tgi_client.add_mock_responses(mock_responses)
+
+
+@pytest.fixture
+def sync_model():
+    return TGI(tgi_client)
+
+
+@pytest.fixture
+def async_model():
+    return AsyncTGI(async_tgi_client)
+
+
+def test_tgi_init():
+    model = from_tgi(
+        InferenceClient("http://localhost:11434"),
+    )
+    assert isinstance(model, TGI)
+
+    model = from_tgi(
+        AsyncInferenceClient("http://localhost:11434"),
+    )
+    assert isinstance(model, AsyncTGI)
+
+    with pytest.raises(ValueError, match="Unsupported client type"):
+        from_tgi("foo")
+
+
+def test_tgi_sync_simple_call(sync_model):
+    result = sync_model("Respond with a single word.", max_new_tokens=10)
+    assert isinstance(result, str)
+
+
+def test_tgi_sync_streaming(sync_model):
+    result = sync_model.stream(
+        "Respond with a single word.",
+        max_new_tokens=10,
+    )
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
+
+
+def test_tgi_sync_json(sync_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
+    result = sync_model("foo?", JsonSchema(json_string), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+def test_tgi_sync_regex(sync_model):
+    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+def test_tgi_sync_cfg(sync_model):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support CFG-based structured outputs",
+    ):
+        sync_model("foo?", CFG(YES_NO_GRAMMAR), max_new_tokens=10)
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_simple_call(async_model):
+    result = await async_model("Respond with a single word.", max_new_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_streaming(async_model):
+    result = async_model.stream("Respond with a single word.", max_new_tokens=10)
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_json(async_model):
+    json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
+    result = await async_model("foo?", JsonSchema(json_string), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_regex(async_model):
+    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_new_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+@pytest.mark.asyncio
+async def test_tgi_async_cfg(async_model):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support CFG-based structured outputs",
+    ):
+        await async_model("foo?", CFG(YES_NO_GRAMMAR), max_new_tokens=10)
diff --git a/tests/models/test_tgi_model_adapter.py b/tests/models/test_tgi_model_adapter.py
new file mode 100644
index 000000000..543ab75ff
--- /dev/null
+++ b/tests/models/test_tgi_model_adapter.py
@@ -0,0 +1,93 @@
+import json
+import pytest
+
+from outlines.models.tgi import TGITypeAdapter
+from outlines.types import CFG, JsonSchema
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
+
+
+@pytest.fixture
+def type_adapter():
+    return TGITypeAdapter()
+
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
+
+@pytest.fixture
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+
+def test_tgi_type_adapter_input_text(type_adapter):
+    message = "prompt"
+    assert message == type_adapter.format_input(message)
+
+
+def test_tgi_type_adapter_input_invalid(type_adapter):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support batch inference.",
+    ):
+        type_adapter.format_input(["foo", "bar"])
+    with pytest.raises(
+        NotImplementedError,
+        match="is not available with TGI",
+    ):
+        type_adapter.format_input({"foo": "bar"})
+
+
+def test_tgi_type_adapter_output_type(
+    type_adapter,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "grammar": {
+            "type": "json",
+            "value": json.loads(JSON_SCHEMA_STRING),
+        }
+    }
+    # whitespace_pattern is ignored
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "grammar": {
+            "type": "json",
+            "value": json.loads(JSON_SCHEMA_STRING),
+        }
+    }
+    assert type_adapter.format_output_type(int) == {
+        "grammar": {
+            "type": "regex",
+            "value": "([+-]?(0|[1-9][0-9]*))",
+        }
+    }
+
+
+def test_tgi_type_adapter_output_type_invalid(
+    type_adapter,
+    cfg_instance,
+):
+    with pytest.raises(
+        NotImplementedError,
+        match="TGI does not support CFG-based structured outputs.",
+    ):
+        type_adapter.format_output_type(cfg_instance)
diff --git a/tests/test_utils/mock_openai_client.py b/tests/test_utils/mock_openai_client.py
index 635a14ae3..fe1e2772f 100644
--- a/tests/test_utils/mock_openai_client.py
+++ b/tests/test_utils/mock_openai_client.py
@@ -1,20 +1,7 @@
-import hashlib
-import pickle
 from typing import List, Dict, Any, Optional
 from unittest.mock import MagicMock
 
-
-def hash_dict(d) -> str:
-    def make_hashable(obj):
-        if isinstance(obj, dict):
-            return frozenset((k, make_hashable(v)) for k, v in sorted(obj.items()))
-        if isinstance(obj, list):
-            return tuple(make_hashable(e) for e in obj)
-        return obj
-
-    hashable_obj = make_hashable(d)
-    pickled_obj = pickle.dumps(hashable_obj)
-    return hashlib.sha256(pickled_obj).hexdigest()
+from tests.test_utils.utils import hash_dict
 
 
 class MockChoice:
diff --git a/tests/test_utils/mock_tgi_client.py b/tests/test_utils/mock_tgi_client.py
new file mode 100644
index 000000000..fef054682
--- /dev/null
+++ b/tests/test_utils/mock_tgi_client.py
@@ -0,0 +1,63 @@
+from typing import Any, Dict
+from unittest.mock import MagicMock
+
+from tests.test_utils.utils import hash_dict
+
+
+class MockTGIInferenceClient:
+    """Mock for TGI `InferenceClient` that can be used to test the TGI model"""
+
+    def __init__(self):
+        self.text_generation = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        def _create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return iter(response)
+            else:
+                return response
+
+        self.text_generation.side_effect = _create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+
+class MockAsyncTGIInferenceClient:
+    """Mock for TGI `InferenceClient` that can be used to test the TGI model"""
+
+    def __init__(self):
+        self.text_generation = MagicMock()
+
+        # The method that will be called by the model when it makes a request
+        async def _async_create(**kwargs):
+            # Hash the arguments to create a unique key
+            request_key = hash_dict(kwargs)
+            response = self._mock_responses.get(request_key)
+            if not response:
+                raise ValueError(f"No response found for {kwargs}")
+            if kwargs.get("stream", False):
+                return self._create_async_streaming_response(response)
+            else:
+                return response
+
+        self.text_generation.side_effect = _async_create
+        self._mock_responses: Dict[str, Any] = {}
+
+    def add_mock_responses(self, mocks: list):
+        for kwargs, response in mocks:
+            request_key = hash_dict(kwargs)
+            self._mock_responses[request_key] = response
+
+    async def _create_async_streaming_response(self, response):
+        """Create an async streaming response generator"""
+        for chunk in response:
+            yield chunk
diff --git a/tests/test_utils/utils.py b/tests/test_utils/utils.py
new file mode 100644
index 000000000..d3a3fa969
--- /dev/null
+++ b/tests/test_utils/utils.py
@@ -0,0 +1,23 @@
+import hashlib
+import pickle
+import sys
+
+
+def hash_dict(d) -> str:
+    def make_hashable(obj):
+        if isinstance(obj, (bool, int, float, str, type(None))):
+            if isinstance(obj, str):
+                return sys.intern(obj)
+            return obj
+        if isinstance(obj, dict):
+            return tuple(sorted(
+                (sys.intern(k) if isinstance(k, str) else k, make_hashable(v))
+                for k, v in obj.items()
+            ))
+        if isinstance(obj, (list, tuple)):
+            return tuple(make_hashable(e) for e in obj)
+        return str(obj)
+
+    hashable_obj = make_hashable(d)
+    pickled_obj = pickle.dumps(hashable_obj, protocol=4)
+    return hashlib.sha256(pickled_obj).hexdigest()

From 1d2394bf0fae8faf1c566e7500d5fe0e87343eed Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 30 Apr 2025 15:53:42 +0200
Subject: [PATCH 320/505] Create the SgLang model

---
 docs/reference/models/sglang.md          |  51 +++++
 outlines/__init__.py                     |   1 +
 outlines/models/__init__.py              |   5 +-
 outlines/models/sglang.py                | 277 +++++++++++++++++++++++
 tests/models/test_sglang.py              | 233 +++++++++++++++++++
 tests/models/test_sglang_type_adapter.py | 145 ++++++++++++
 tests/test_utils/mock_openai_client.py   |   7 +-
 7 files changed, 717 insertions(+), 2 deletions(-)
 create mode 100644 docs/reference/models/sglang.md
 create mode 100644 outlines/models/sglang.py
 create mode 100644 tests/models/test_sglang.py
 create mode 100644 tests/models/test_sglang_type_adapter.py

diff --git a/docs/reference/models/sglang.md b/docs/reference/models/sglang.md
new file mode 100644
index 000000000..da3cfa89c
--- /dev/null
+++ b/docs/reference/models/sglang.md
@@ -0,0 +1,51 @@
+# SgLang
+
+## Prerequisites
+
+The Outlines `SgLang` model is inteded to be used along with a separate sglang server (running either locally or remotely). Make sure you have an sglang server running before using the `SgLang` model. As the sglang client relies on the `openai` python sdk, you need to have the `openai` package installed.
+
+## Initialize the model
+
+To load the model, you can use the `from_sglang` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Based on whether the `openai` instance is synchronous or asynchronous, you will receive a `SgLang` or `AsyncSgLang` model instance.
+
+```python
+import openai
+import outlines
+
+sync_openai_client = openai.OpenAI(base_url="...")
+async_openai_client = openai.AsyncOpenAI(base_url="...")
+
+sync_model = outlines.from_sglang(sync_openai_client, "qwen/qwen2.5-0.5b-instruct")
+print(type(sync_model)) # <class 'outlines.models.sglang.SgLang'>
+
+async_model = outlines.from_sglang(async_openai_client, "qwen/qwen2.5-0.5b-instruct")
+print(type(async_model)) # <class 'outlines.models.sglang.AsyncSgLang'>
+```
+
+## Generate text
+
+To generate text, you can call the model with a prompt as argument and optionally an output type to rely on structured generation:
+
+```python
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+
+answer = sync_model("Create a character.", output_type=Character)
+answer = await async_model("Create a character.", output_type=Character)
+```
+
+The `SgLang` model supports also supports streaming.
+
+```python
+for chunk in sync_model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk)
+```
+
+## Optional parameters
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `openai` client. An optional parameter of particular interest is `extra_body`, which is a dictionnary containing arguments that are specific to sglang and are not part of the standard `openai` interface (see the [sglang documentation][sglang-docs] on the OpenAI-compatible server for more information).
+
+[sglang-docs]: https://docs.sglang.ai/start/install.html
+[sglang-structured-output-quickstart]: https://docs.sglang.ai/backend/structured_outputs.html#JSON
diff --git a/outlines/__init__.py b/outlines/__init__.py
index cc9062227..1eebb8bf0 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -18,6 +18,7 @@
     from_ollama,
     from_llamacpp,
     from_mlxlm,
+    from_sglang,
     from_tgi,
     from_vllm,
     from_vllm_offline,
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index ee51e2305..503586de9 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -16,6 +16,7 @@
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
+from .sglang import from_sglang, SgLang, AsyncSgLang
 from .tgi import from_tgi, TGI, AsyncTGI
 from .transformers import (
     Transformers,
@@ -30,15 +31,17 @@
     Transformers, LlamaCpp, OpenAI, MLXLM, VLLMOffline, Ollama
 ]
 
-SteerableModel = Union[LlamaCpp, Transformers, MLXLM, VLLMOffline]
+SteerableModel = Union[LlamaCpp, MLXLM, Transformers, VLLMOffline]
 BlackBoxModel = Union[
     Anthropic,
     AsyncTGI,
+    AsyncSgLang,
     AsyncVLLM,
     Dottxt,
     Gemini,
     Ollama,
     OpenAI,
     TGI,
+    SgLang,
     VLLM,
 ]
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
new file mode 100644
index 000000000..8679cda66
--- /dev/null
+++ b/outlines/models/sglang.py
@@ -0,0 +1,277 @@
+import json
+import warnings
+from typing import TYPE_CHECKING, Union, Optional
+
+from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
+from outlines.models.openai import OpenAITypeAdapter
+from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
+
+
+if TYPE_CHECKING:
+    from openai import AsyncOpenAI, OpenAI
+
+
+class SgLangTypeAdapter(ModelTypeAdapter):
+
+    def format_input(self, model_input):
+        """Generate the prompt argument to pass to the client.
+
+        We rely on the OpenAITypeAdapter to format the input as the sglang
+        server expects input in the same format as OpenAI.
+
+        Argument
+        --------
+        model_input
+            The input passed by the user.
+
+        """
+        return OpenAITypeAdapter().format_input(model_input)
+
+    def format_output_type(self, output_type):
+        """Generate the structured output argument to pass to the client.
+
+        Argument
+        --------
+        output_type
+            The structured output type provided.
+
+        """
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            warnings.warn(
+                "SgLang grammar-based structured outputs expects an EBNF "
+                "grammar instead of a Lark grammar as is generally used in "
+                "Outlines. The grammar cannot be used as a structured output "
+                "type with an outlines backend, it is only compatible with "
+                "the sglang and llguidance backends."
+            )
+            return {"extra_body": {"ebnf": term.definition}}
+        elif isinstance(term, JsonSchema):
+            return OpenAITypeAdapter().format_json_output_type(
+                json.loads(term.schema)
+            )
+        else:
+            return {"extra_body": {"regex": to_regex(term)}}
+
+
+class SgLang(Model):
+    """Represents a client to a `sglang` server."""
+
+    def __init__(self, client, model_name: Optional[str] = None):
+        """Create a `SgLang` model instance.
+
+        Parameters
+        ----------
+        client
+            An `OpenAI` client instance.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = SgLangTypeAdapter()
+
+    def generate(self, model_input, output_type, **inference_kwargs):
+        """Generate text using the client.
+
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        The generated text.
+
+        """
+        client_args = self._build_client_args(
+            model_input,
+            output_type,
+            **inference_kwargs,
+        )
+
+        response = self.client.chat.completions.create(**client_args)
+
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The SgLang server refused to answer the request: "
+                    f"{message.refusal}"
+                )
+
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
+
+    def generate_stream(self, model_input, output_type, **inference_kwargs):
+        """Return a text generator.
+
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        A generator that yields the generated text chunks.
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = self.client.chat.completions.create(
+            **client_args, stream=True,
+        )
+
+        for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+        """Build the arguments to pass to the SgLang client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+        model_name = (
+            inference_kwargs.pop("model", self.model_name)
+            or self.model_name
+        )
+
+        client_args = {
+            **messages,
+            "model": model_name,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+class AsyncSgLang(AsyncModel):
+    """Represents an async client to a `sglang` server."""
+
+    def __init__(self, client, model_name: Optional[str] = None):
+        """Create an `AsyncSgLang` model instance.
+
+        Parameters
+        ----------
+        client
+            An `AsyncOpenAI` client instance.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = SgLangTypeAdapter()
+
+    async def generate(self, model_input, output_type, **inference_kwargs):
+        """Generate text using `sglang`.
+
+        Arguments
+        ---------
+        prompt
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The inference kwargs that can be passed to the sglang server
+
+        Returns
+        -------
+        The generated text.
+
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        response = await self.client.chat.completions.create(**client_args)
+
+        messages = [choice.message for choice in response.choices]
+        for message in messages:
+            if message.refusal is not None:  # pragma: no cover
+                raise ValueError(
+                    f"The sglang server refused to answer the request: "
+                    f"{message.refusal}"
+                )
+
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
+
+    async def generate_stream(
+        self, model_input, output_type, **inference_kwargs
+    ):
+        """Return a text generator.
+
+        Arguments
+        ---------
+        model_input
+            The text prompt provided by the user.
+        output_type
+            The structured output type provided.
+        inference_kwargs
+            The additional arguments that will be passed on to the client.
+
+        Returns
+        -------
+        A generator that yields the generated text chunks.
+        """
+        client_args = self._build_client_args(
+            model_input, output_type, **inference_kwargs,
+        )
+
+        stream = await self.client.chat.completions.create(
+            **client_args,
+            stream=True,
+        )
+
+        async for chunk in stream:  # pragma: no cover
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+        """Build the arguments to pass to the SgLang client."""
+        messages = self.type_adapter.format_input(model_input)
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        inference_kwargs.update(output_type_args)
+        model_name = (
+            inference_kwargs.pop("model", self.model_name)
+            or self.model_name
+        )
+
+        client_args = {
+            **messages,
+            "model": model_name,
+            **inference_kwargs,
+        }
+
+        return client_args
+
+
+def from_sglang(
+    client: Union["OpenAI", "AsyncOpenAI"],
+    model_name: Optional[str] = None,
+) -> Union[SgLang, AsyncSgLang]:
+    from openai import AsyncOpenAI, OpenAI
+
+    if isinstance(client, OpenAI):
+        return SgLang(client, model_name)
+    elif isinstance(client, AsyncOpenAI):
+        return AsyncSgLang(client, model_name)
+    else:
+        raise ValueError(
+            f"Unsupported client type: {type(client)}.\n"
+            "Please provide an OpenAI or AsyncOpenAI instance."
+        )
diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
new file mode 100644
index 000000000..2a594cc5a
--- /dev/null
+++ b/tests/models/test_sglang.py
@@ -0,0 +1,233 @@
+# ATTENTION: When running this test with an actual SgLang server, use the
+# llguidance backend (--grammar-backend llguidance)
+# The outlines backend does not support the EBNF grammar. The xgrammar
+# backend is slow and buggy.
+
+import os
+import re
+import warnings
+from typing import AsyncGenerator, Generator
+
+import pytest
+from openai import AsyncOpenAI, OpenAI
+
+from outlines.models.sglang import SgLang, AsyncSgLang, from_sglang
+from outlines.types.dsl import CFG, Regex, JsonSchema
+from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
+
+
+EBNF_YES_NO_GRAMMAR = """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+
+
+# If the SGLANG_SERVER_URL environment variable is set, use the real SgLang server
+# Otherwise, use the mock server
+sglang_server_url = os.environ.get("SGLANG_SERVER_URL")
+sglang_model_name = os.environ.get(
+    "SGLANG_MODEL_NAME", "qwen/qwen2.5-0.5b-instruct"
+)
+if sglang_server_url:
+    openai_client = OpenAI(base_url=sglang_server_url)
+    async_openai_client = AsyncOpenAI(base_url=sglang_server_url)
+else:
+    warnings.warn("No SgLang server URL provided, using mock server")
+    openai_client = MockOpenAIClient()
+    async_openai_client = MockAsyncOpenAIClient()
+
+mock_responses = [
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': sglang_model_name,
+        },
+        "foo"
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'model': sglang_model_name,
+            'stream': True
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': 'Respond with a single word.'}
+            ],
+            'n': 2,
+            'model': sglang_model_name,
+        },
+        ["foo", "bar"]
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+            'response_format': {
+                'type': 'json_schema',
+                'json_schema': {
+                    'name': 'default',
+                    'strict': True,
+                    'schema': {
+                        'type': 'object',
+                        'properties': {'bar': {'type': 'string'}},
+                        'additionalProperties': False
+                    }
+                }
+            }
+        },
+        '{"foo": "bar"}'
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'regex': '([0-9]{3})',
+            },
+        },
+        "123"
+    ),
+    (
+        {
+            'messages': [{'role': "user", 'content': 'foo?'}],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+            'extra_body': {
+                'ebnf': EBNF_YES_NO_GRAMMAR,
+            },
+        },
+        "yes"
+    ),
+]
+
+
+# If the SGLANG_SERVER_URL environment variable is not set, add the mock
+# responses to the mock clients
+if not sglang_server_url:
+    async_openai_client.add_mock_responses(mock_responses)
+    openai_client.add_mock_responses(mock_responses)
+
+
+@pytest.fixture
+def sync_model():
+    return SgLang(openai_client, model_name=sglang_model_name)
+
+
+@pytest.fixture
+def async_model():
+    return AsyncSgLang(async_openai_client, model_name=sglang_model_name)
+
+
+def test_sglang_init():
+    model = from_sglang(OpenAI(base_url="http://localhost:11434"))
+    assert isinstance(model, SgLang)
+
+    model = from_sglang(AsyncOpenAI(base_url="http://localhost:11434"))
+    assert isinstance(model, AsyncSgLang)
+
+    with pytest.raises(ValueError, match="Unsupported client type"):
+        from_sglang("foo")
+
+
+def test_sglang_sync_simple_call(sync_model):
+    result = sync_model("Respond with a single word.",)
+    assert isinstance(result, str)
+
+
+def test_sglang_sync_streaming(sync_model):
+    result = sync_model.stream("Respond with a single word.")
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
+
+
+def test_sglang_sync_multiple_samples(sync_model):
+    result = sync_model("Respond with a single word.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+def test_sglang_sync_json(sync_model):
+    json_string = (
+        '{"type": "object", "properties":'
+        + ' {"bar": {"type": "string"}}}'
+    )
+    result = sync_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+def test_sglang_sync_regex(sync_model):
+    result = sync_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+def test_sglang_sync_cfg(sync_model):
+    with pytest.warns(
+        UserWarning,
+        match="SgLang grammar-based structured outputs expects an EBNF"
+    ):
+        result = sync_model("foo?", CFG(EBNF_YES_NO_GRAMMAR), max_tokens=10)
+        assert isinstance(result, str)
+        assert result in ["yes", "no"]
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_simple_call(async_model):
+    result = await async_model("Respond with a single word.",)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_streaming(async_model):
+    result = async_model.stream("Respond with a single word.")
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_multiple_samples(async_model):
+    result = await async_model("Respond with a single word.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_json(async_model):
+    json_string = (
+        '{"type": "object", "properties":'
+        + ' {"bar": {"type": "string"}}}'
+    )
+    result = await async_model("foo?", JsonSchema(json_string), max_tokens=10)
+    assert isinstance(result, str)
+    assert "bar" in result
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_regex(async_model):
+    result = await async_model("foo?", Regex(r"[0-9]{3}"), max_tokens=10)
+    assert isinstance(result, str)
+    assert re.match(r"[0-9]{3}", result)
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_cfg(async_model):
+    result = await async_model("foo?", CFG(EBNF_YES_NO_GRAMMAR), max_tokens=10)
+    assert isinstance(result, str)
+    assert result in ["yes", "no"]
diff --git a/tests/models/test_sglang_type_adapter.py b/tests/models/test_sglang_type_adapter.py
new file mode 100644
index 000000000..b9630ba15
--- /dev/null
+++ b/tests/models/test_sglang_type_adapter.py
@@ -0,0 +1,145 @@
+import io
+import json
+import pytest
+from dataclasses import dataclass
+
+from PIL import Image
+from outlines.models.sglang import SgLangTypeAdapter
+from outlines.types import CFG, JsonSchema
+
+from outlines import Vision
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
+
+
+@pytest.fixture
+def type_adapter():
+    return SgLangTypeAdapter()
+
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
+
+@pytest.fixture
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
+def test_sglang_type_adapter_input_text(type_adapter):
+    message = "prompt"
+    result = type_adapter.format_input(message)
+    assert isinstance(result, dict)
+    assert len(result["messages"]) == 1
+    assert result["messages"][0]["role"] == "user"
+    assert result["messages"][0]["content"] == message
+
+
+def test_sglang_type_adapter_input_vision(type_adapter, image):
+    input_message = Vision("hello", image)
+    result = type_adapter.format_input(input_message)
+    assert isinstance(result, dict)
+
+    messages = result["messages"]
+    assert len(messages) == 1
+
+    message = messages[0]
+    assert message["role"] == "user"
+    assert len(message["content"]) == 2
+    assert message["content"][0]["type"] == "text"
+    assert message["content"][0]["text"] == "hello"
+
+    assert message["content"][1]["type"] == "image_url"
+    assert (
+        message["content"][1]["image_url"]["url"]
+        == f"data:image/png;base64,{input_message.image_str}"
+    )
+
+
+def test_sglang_type_adapter_input_invalid(type_adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    prompt = Audio(
+        "file",
+    )
+    with pytest.raises(TypeError, match="The input type"):
+        _ = type_adapter.format_input(prompt)
+
+
+def test_sglang_type_adapter_output_type(
+    type_adapter,
+    cfg_instance,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    with pytest.warns(
+        UserWarning,
+        match="SgLang grammar-based structured outputs expects an EBNF"
+    ):
+        assert type_adapter.format_output_type(cfg_instance) == {
+            "extra_body": {"ebnf": CFG_STRING}
+        }
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "response_format": {
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": {
+                    **json.loads(JSON_SCHEMA_STRING),
+                    "additionalProperties": False,
+                },
+            },
+        }
+    }
+    # whitespace pattern is ignored
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "response_format": {
+            "type": "json_schema",
+            "json_schema": {
+                "name": "default",
+                "strict": True,
+                "schema": {
+                    **json.loads(JSON_SCHEMA_STRING),
+                    "additionalProperties": False,
+                },
+            },
+        }
+    }
+    assert type_adapter.format_output_type(int) == {
+        "extra_body": {"regex": "([+-]?(0|[1-9][0-9]*))"}
+    }
diff --git a/tests/test_utils/mock_openai_client.py b/tests/test_utils/mock_openai_client.py
index fe1e2772f..5f466e30f 100644
--- a/tests/test_utils/mock_openai_client.py
+++ b/tests/test_utils/mock_openai_client.py
@@ -5,7 +5,12 @@
 
 
 class MockChoice:
-    def __init__(self, content: str, finish_reason: str = "stop", refusal: Optional[str] = None):
+    def __init__(
+        self,
+        content: str,
+        finish_reason: str = "stop",
+        refusal: Optional[str] = None
+    ):
         self.message = MagicMock()
         self.message.content = content
         self.message.refusal = refusal

From e31f0f5460cf6d7758ebaadfc77dbd863f9b1cc8 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Thu, 24 Apr 2025 14:21:19 -0700
Subject: [PATCH 321/505] Add commercial info to the v1 docs

---
 README.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 7aac9bd9a..0d85aebf1 100644
--- a/README.md
+++ b/README.md
@@ -6,7 +6,6 @@
 
  🗒️ *Structured outputs for LLMs* 🗒️
 
-
 Made with ❤👷️ by the team at [.txt](https://dottxt.co)
 <br>Trusted by NVIDIA, Cohere, HuggingFace, vLLM, etc.
 
@@ -22,6 +21,8 @@ Made with ❤👷️ by the team at [.txt](https://dottxt.co)
 
 </div>
 
+Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales).
+
 ## Table of Contents
 
 - [Why Outlines?](#why-outlines)

From ef12a82edefd32f8613bee4ecc4b7011d0e730c0 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 9 May 2025 15:19:54 +0200
Subject: [PATCH 322/505] Change the deprecation warning of v0 from v1.5.0 to
 v.1.1.0

---
 outlines/models/llamacpp.py                      | 2 +-
 outlines/models/openai.py                        | 2 +-
 outlines/models/vllm_offline.py                  | 2 +-
 outlines/v0_legacy/function.py                   | 2 +-
 outlines/v0_legacy/generate/cfg.py               | 4 ++--
 outlines/v0_legacy/generate/choice.py            | 4 ++--
 outlines/v0_legacy/generate/format.py            | 4 ++--
 outlines/v0_legacy/generate/fsm.py               | 4 ++--
 outlines/v0_legacy/generate/json.py              | 4 ++--
 outlines/v0_legacy/generate/regex.py             | 4 ++--
 outlines/v0_legacy/generate/text.py              | 4 ++--
 outlines/v0_legacy/models/exllamav2.py           | 2 +-
 outlines/v0_legacy/models/llamacpp.py            | 2 +-
 outlines/v0_legacy/models/mlxlm.py               | 2 +-
 outlines/v0_legacy/models/transformers.py        | 4 ++--
 outlines/v0_legacy/models/transformers_vision.py | 2 +-
 outlines/v0_legacy/models/vllm_offline.py        | 2 +-
 17 files changed, 25 insertions(+), 25 deletions(-)

diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index be501ae2c..93f8f5eb5 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -233,7 +233,7 @@ def token_generator() -> Iterator[str]:
     def load_lora(self, adapter_path: str):  # pragma: no cover
         warnings.warn("""
             The `load_lora` method is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.5.0.
+            Support for it will be removed in v1.1.0.
             """,
             DeprecationWarning,
             stacklevel=2,
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 2f9408e45..c4ff7335d 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -181,7 +181,7 @@ def __init__(
         if isinstance(model_name, OpenAIConfig) or kwargs.get("config"):
             warnings.warn("""
                 The `openai` function is deprecated starting from v1.0.0.
-                Do not use it. Support for it will be removed in v1.5.0.
+                Do not use it. Support for it will be removed in v1.1.0.
                 Instead, you should instantiate a `OpenAI` model with the
                 `outlines.from_openai` function that takes an openai library
                 client and a model name as arguments. Similarly, you cannot
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index 808874c36..4766139b1 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -143,7 +143,7 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
     def load_lora(self, adapter_path: Optional[str]):
         warnings.warn("""
             The `load_lora` method is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.5.0.
+            Support for it will be removed in v1.1.0.
             Please use the v1 of the `outlines` library by using the
             `outlines.from_vllm` function to create a `VLLM` model
             instance.
diff --git a/outlines/v0_legacy/function.py b/outlines/v0_legacy/function.py
index d15ec702f..465df53f9 100644
--- a/outlines/v0_legacy/function.py
+++ b/outlines/v0_legacy/function.py
@@ -16,7 +16,7 @@
 def function_warning():
     warnings.warn("""
         The `Function` class is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Instead, you should use the `outlines.Application` class that
         implements a similar functionality.
         An `Application` is initialized with a prompt template and an
diff --git a/outlines/v0_legacy/generate/cfg.py b/outlines/v0_legacy/generate/cfg.py
index db206996d..0f33b6efb 100644
--- a/outlines/v0_legacy/generate/cfg.py
+++ b/outlines/v0_legacy/generate/cfg.py
@@ -20,7 +20,7 @@ def cfg(
     """Generate text in the language of a Context-Free Grammar
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.5.0.
+    Support for it will be removed in v1.1.0.
     Use the `Generator` object instead:
 
     ```python
@@ -51,7 +51,7 @@ def cfg(
 
     warnings.warn("""
         The `cfg` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Use the `Generator` object instead:
         ```python
         from outlines import Generator
diff --git a/outlines/v0_legacy/generate/choice.py b/outlines/v0_legacy/generate/choice.py
index cee339bac..eff2fcffd 100644
--- a/outlines/v0_legacy/generate/choice.py
+++ b/outlines/v0_legacy/generate/choice.py
@@ -24,7 +24,7 @@ def choice(
     """Generate a choice from a list of options.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.5.0.
+    Support for it will be removed in v1.1.0.
     Use the `Generator` object instead:
 
     ```python
@@ -53,7 +53,7 @@ def choice(
     """
     warnings.warn("""
         The `choice` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Use the `Generator` object instead:
         ```python
         from outlines import Generator
diff --git a/outlines/v0_legacy/generate/format.py b/outlines/v0_legacy/generate/format.py
index 459eebff2..cd574ac4e 100644
--- a/outlines/v0_legacy/generate/format.py
+++ b/outlines/v0_legacy/generate/format.py
@@ -18,7 +18,7 @@ def format(
     """Generate structured data that can be parsed as a Python type.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.5.0.
+    Support for it will be removed in v1.1.0.
     Use the `Generator` object instead::
 
     ```python
@@ -47,7 +47,7 @@ def format(
     """
     warnings.warn("""
         The `format` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Use the `Generator` object instead:
         ```python
         from outlines import Generator
diff --git a/outlines/v0_legacy/generate/fsm.py b/outlines/v0_legacy/generate/fsm.py
index 60e130169..23a5477fe 100644
--- a/outlines/v0_legacy/generate/fsm.py
+++ b/outlines/v0_legacy/generate/fsm.py
@@ -21,7 +21,7 @@ def fsm(
     """Generate text that follows a Finite State Machine.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.5.0.
+    Support for it will be removed in v1.1.0.
     Use the `Generator` object instead:
 
     ```python
@@ -52,7 +52,7 @@ def fsm(
     """
     warnings.warn("""
         The `fsm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Use the `Generator` object instead:
         ```python
         from outlines import Generator
diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
index 0e6ed8d46..6260b6471 100644
--- a/outlines/v0_legacy/generate/json.py
+++ b/outlines/v0_legacy/generate/json.py
@@ -27,7 +27,7 @@ def json(
     """Generate structured text that follows a JSON Schema.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.5.0.
+    Support for it will be removed in v1.1.0.
     Use the `Generator` object instead:
 
     ```python
@@ -62,7 +62,7 @@ def json(
     """
     warnings.warn("""
         The `json` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Use the `Generator` object instead:
         ```python
         from outlines import Generator
diff --git a/outlines/v0_legacy/generate/regex.py b/outlines/v0_legacy/generate/regex.py
index af4beb260..261f72037 100644
--- a/outlines/v0_legacy/generate/regex.py
+++ b/outlines/v0_legacy/generate/regex.py
@@ -14,7 +14,7 @@ def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
     """Generate structured text in the language of a regular expression.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.5.0.
+    Support for it will be removed in v1.1.0.
     Use the `Generator` object instead:
 
     ```python
@@ -43,7 +43,7 @@ def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
     """
     warnings.warn("""
         The `regex` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Use the `Generator` object instead:
         ```python
         from outlines import Generator
diff --git a/outlines/v0_legacy/generate/text.py b/outlines/v0_legacy/generate/text.py
index b71e00725..148cf131d 100644
--- a/outlines/v0_legacy/generate/text.py
+++ b/outlines/v0_legacy/generate/text.py
@@ -19,7 +19,7 @@ def text(model, sampler: Sampler = multinomial()) -> Union[
     """Generate unconstrained text.
 
     This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.5.0.
+    Support for it will be removed in v1.1.0.
     Use the `Generator` object instead:
 
     ```python
@@ -44,7 +44,7 @@ def text(model, sampler: Sampler = multinomial()) -> Union[
     """
     warnings.warn("""
         The `text` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Use the `Generator` object instead:
         ```python
         from outlines import Generator
diff --git a/outlines/v0_legacy/models/exllamav2.py b/outlines/v0_legacy/models/exllamav2.py
index 4b4039acd..683512002 100644
--- a/outlines/v0_legacy/models/exllamav2.py
+++ b/outlines/v0_legacy/models/exllamav2.py
@@ -64,7 +64,7 @@ def __init__(
     ):
         warnings.warn("""
             The `ExLlamaV2Model` model is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.5.0.
+            Support for it will be removed in v1.1.0.
             As the `exllamav2` library is not compatible with some key features
             of `outlines`, we decided to remove support for it.
             """,
diff --git a/outlines/v0_legacy/models/llamacpp.py b/outlines/v0_legacy/models/llamacpp.py
index cfdad98e0..6c29630d5 100644
--- a/outlines/v0_legacy/models/llamacpp.py
+++ b/outlines/v0_legacy/models/llamacpp.py
@@ -108,7 +108,7 @@ def llamacpp(
 
     warnings.warn("""
         The `llamacpp` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Instead, you should instantiate a `LlamaCpp` model with the
         `outlines.from_llamacpp` function that takes a `llama_cpp.Llama`
         instance as argument. For example:
diff --git a/outlines/v0_legacy/models/mlxlm.py b/outlines/v0_legacy/models/mlxlm.py
index 76ba44465..a69594173 100644
--- a/outlines/v0_legacy/models/mlxlm.py
+++ b/outlines/v0_legacy/models/mlxlm.py
@@ -117,7 +117,7 @@ def mlxlm(
     """
     warnings.warn("""
         The `mlxlm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Instead, you should instantiate a `MLXLM` model with the
         `outlines.from_mlxlm` function that takes a mlxlm model
         and a tokenizer as arguments. For example:
diff --git a/outlines/v0_legacy/models/transformers.py b/outlines/v0_legacy/models/transformers.py
index 379d64f2a..d7cf135ce 100644
--- a/outlines/v0_legacy/models/transformers.py
+++ b/outlines/v0_legacy/models/transformers.py
@@ -107,7 +107,7 @@ def transformers(
     """
     warnings.warn("""
         The `transformers` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Instead, you should instantiate a `Transformers` model with the
         `outlines.from_transformers` function that takes a transformers model
         and a tokenizer as arguments. For example:
@@ -155,7 +155,7 @@ def mamba(
 ):
     warnings.warn("""
         The `mamba` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Instead, you should instantiate a `Transformers` model with the
         `outlines.from_transformers` function that takes a transformers model
         and a tokenizer as arguments. For Mamba, you would use a mamba model
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
index 4d46a0618..83008fcbe 100644
--- a/outlines/v0_legacy/models/transformers_vision.py
+++ b/outlines/v0_legacy/models/transformers_vision.py
@@ -68,7 +68,7 @@ def transformers_vision(
     """
     warnings.warn("""
         The `transformers_vision` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Instead, you should instantiate a `TransformersMultiModal` model with
         the `outlines.from_transformers` function that takes a transformers
         model and a processor as arguments. For example:
diff --git a/outlines/v0_legacy/models/vllm_offline.py b/outlines/v0_legacy/models/vllm_offline.py
index 0928bfd03..8efb7cfc0 100644
--- a/outlines/v0_legacy/models/vllm_offline.py
+++ b/outlines/v0_legacy/models/vllm_offline.py
@@ -80,7 +80,7 @@ def vllm(model_name: str, **vllm_model_params):
     """
     warnings.warn("""
         The `vllm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.5.0.
+        Do not use it. Support for it will be removed in v1.1.0.
         Instead, you should instantiate a `VLLMOffline` model with the
         `outlines.from_vllm_offline` function that takes a vLLM model name as
         argument.

From f5795f48a91695ca1ea0cbbe51e2d47de21279c1 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 9 May 2025 15:48:25 +0200
Subject: [PATCH 323/505] Remove the serve module

---
 Dockerfile                 |  23 ------
 outlines/serve/__init__.py |   0
 outlines/serve/serve.py    | 141 -------------------------------------
 pyproject.toml             |   9 ---
 4 files changed, 173 deletions(-)
 delete mode 100644 Dockerfile
 delete mode 100644 outlines/serve/__init__.py
 delete mode 100644 outlines/serve/serve.py

diff --git a/Dockerfile b/Dockerfile
deleted file mode 100644
index 7e6f2e68a..000000000
--- a/Dockerfile
+++ /dev/null
@@ -1,23 +0,0 @@
-### Build stage
-FROM python:3.10 AS builder
-
-WORKDIR /outlines
-
-RUN pip install --upgrade pip
-
-# Copy necessary build components
-COPY pyproject.toml .
-COPY outlines ./outlines
-
-# Install outlines and outlines[serve]
-# .git required by setuptools-scm
-RUN --mount=source=.git,target=.git,type=bind \
-    pip install --no-cache-dir .[serve]
-
-### Runtime stage
-FROM python:3.10
-WORKDIR /outlines
-COPY --from=builder /outlines /outlines
-
-# https://dottxt-ai.github.io/outlines/reference/vllm/
-ENTRYPOINT ["python3", "-m", "outlines.serve.serve"]
diff --git a/outlines/serve/__init__.py b/outlines/serve/__init__.py
deleted file mode 100644
index e69de29bb..000000000
diff --git a/outlines/serve/serve.py b/outlines/serve/serve.py
deleted file mode 100644
index 84dc85761..000000000
--- a/outlines/serve/serve.py
+++ /dev/null
@@ -1,141 +0,0 @@
-#  _______________________________
-# / Don't want to self-host?      \
-# \ Try .json at http://dottxt.co /
-#  -------------------------------
-#        \   ^__^
-#         \  (oo)\_______
-#            (__)\       )\/\
-#                ||----w |
-#                ||     ||
-#
-#
-# Copyright 2024- the Outlines developers
-# Copyright 2023 the vLLM developers
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-
-#     http://www.apache.org/licenses/LICENSE-2.0
-
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-import argparse
-import json
-from typing import AsyncGenerator
-
-import uvicorn
-from fastapi import FastAPI, Request
-from fastapi.responses import JSONResponse, Response, StreamingResponse
-from vllm.engine.arg_utils import AsyncEngineArgs
-from vllm.engine.async_llm_engine import AsyncLLMEngine
-from vllm.sampling_params import SamplingParams
-from vllm.utils import random_uuid
-
-from outlines.models.vllm_offline import adapt_tokenizer
-from outlines.processors import JSONLogitsProcessor, RegexLogitsProcessor
-
-TIMEOUT_KEEP_ALIVE = 5  # seconds.
-TIMEOUT_TO_PREVENT_DEADLOCK = 1  # seconds.
-app = FastAPI()
-engine = None
-tokenizer = None
-
-
-@app.get("/health")
-async def health() -> Response:
-    """Health check."""
-    return Response(status_code=200)
-
-
-@app.post("/generate")
-async def generate(request: Request) -> Response:
-    """Generate completion for the request.
-
-    The request should be a JSON object with the following fields:
-    - prompt: the prompt to use for the generation.
-    - schema: the JSON schema to use for the generation (if regex is not provided).
-    - regex: the regex to use for the generation (if schema is not provided).
-    - stream: whether to stream the results or not.
-    - other fields: the sampling parameters (See `SamplingParams` for details).
-    """
-    assert engine is not None
-
-    request_dict = await request.json()
-    prompt = request_dict.pop("prompt")
-    stream = request_dict.pop("stream", False)
-
-    json_schema = request_dict.pop("schema", None)
-    regex_string = request_dict.pop("regex", None)
-    tensor_library_name = request_dict.pop("tensor_library_name", "torch")
-    if json_schema is not None:
-        logits_processors = [JSONLogitsProcessor(json_schema, tokenizer, tensor_library_name)]
-    elif regex_string is not None:
-        logits_processors = [RegexLogitsProcessor(regex_string, tokenizer, tensor_library_name)]
-    else:
-        logits_processors = []
-
-    sampling_params = SamplingParams(
-        **request_dict,
-        logits_processors=logits_processors,  # type: ignore
-    )
-    request_id = random_uuid()
-
-    results_generator = engine.generate(prompt, sampling_params, request_id)  # type: ignore
-
-    # Streaming case
-    async def stream_results() -> AsyncGenerator[bytes, None]:
-        async for request_output in results_generator:
-            prompt = request_output.prompt
-            text_outputs = [prompt + output.text for output in request_output.outputs]
-            ret = {"text": text_outputs}
-            yield (json.dumps(ret) + "\0").encode("utf-8")
-
-    if stream:
-        return StreamingResponse(stream_results())
-
-    # Non-streaming case
-    final_output = None
-    async for request_output in results_generator:
-        if await request.is_disconnected():
-            # Abort the request if the client disconnects.
-            await engine.abort(request_id)  # type: ignore
-            return Response(status_code=499)
-        final_output = request_output
-
-    assert final_output is not None
-    prompt = final_output.prompt
-    text_outputs = [prompt + output.text for output in final_output.outputs]
-    ret = {"text": text_outputs}
-    return JSONResponse(ret)
-
-
-if __name__ == "__main__":
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--host", type=str, default=None)
-    parser.add_argument("--port", type=int, default=8000)
-    parser.add_argument("--ssl-keyfile", type=str, default=None)
-    parser.add_argument("--ssl-certfile", type=str, default=None)
-    parser = AsyncEngineArgs.add_cli_args(parser)
-    args = parser.parse_args()
-
-    # Adds the `engine_use_ray`,  `disable_log_requests` and `max_log_len`
-    # arguments
-    engine_args: AsyncEngineArgs = AsyncEngineArgs.from_cli_args(args)  # type: ignore
-
-    # Sets default for the model (`facebook/opt-125m`)
-    engine = AsyncLLMEngine.from_engine_args(engine_args)
-    tokenizer = adapt_tokenizer(tokenizer=engine.engine.tokenizer.tokenizer)
-
-    uvicorn.run(
-        app,
-        host=args.host,
-        port=args.port,
-        log_level="debug",
-        timeout_keep_alive=TIMEOUT_KEEP_ALIVE,
-        ssl_keyfile=args.ssl_keyfile,
-        ssl_certfile=args.ssl_certfile,
-    )
diff --git a/pyproject.toml b/pyproject.toml
index 2fdcbf5f5..4c3d6b287 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -77,12 +77,6 @@ test = [
     "sentencepiece",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
-serve = [
-    "vllm>=0.3.0",
-    "uvicorn",
-    "fastapi",
-    "pydantic>=2.0",
-]
 
 [project.urls]
 homepage = "https://github.com/dottxt-ai/outlines"
@@ -150,8 +144,6 @@ module = [
     "requests.*",
     "responses.*",
     "vllm.*",
-    "uvicorn.*",
-    "fastapi.*",
     "iso3166.*",
     "airportsdata.*",
     "outlines_core.*",
@@ -177,7 +169,6 @@ omit = [
     "outlines/models/openai.py",
     "outlines/models/vllm_offline.py",
     "outlines/processors/tensor_adapters/mlx.py",
-    "outlines/serve/serve.py",
     "outlines/v0_legacy/models/anthropic.py",
     "outlines/v0_legacy/models/dottxt.py",
     "outlines/v0_legacy/models/exllamav2.py",

From f26d8053b13381a8d05a2974ec32a5b4cb947c33 Mon Sep 17 00:00:00 2001
From: cpfiffer <cameron@pfiffer.org>
Date: Thu, 8 May 2025 09:35:38 -0700
Subject: [PATCH 324/505] parent 93d1e077201e46a46d18c881f0108427241afa27
 author cpfiffer <cameron@pfiffer.org> 1746722138 -0700 committer cpfiffer
 <cameron@pfiffer.org> 1747084888 -0700

Simplify doc themes

correct yaml whitespace

Create separate folder for v1 docs, docs_new

Create separate folder for v1 docs, docs_new

Update Installation Documentation with Modern Tooling

This update provides clearer, more comprehensive installation instructions for Outlines, highlighting modern dependency management tools like `uv` and expanding the list of supported model providers.

Update documentation styling with bold formatting

Remove extra whitespace and trailing comments in docs index

Add Application API documentation to docs

Add code example block to Application class docstring

Move models from reference/ to api/

Remove deprecated API documentation files and restructure the Getting Started section with comprehensive installation instructions and usage examples. Introduce new output types documentation for JSON Schema, Regular Expressions, and Context-Free Grammars.

Update index.md to enhance the introduction with a new title and logo images. Remove outdated installation and usage examples, while retaining essential information about the Outlines library and its capabilities.

Add user guide documentation for chat templating, regex DSL, inference backend selection, and unstructured text generation; update index and structured generation sections.

Note breaking changes in versioning guide

Simplify community page

Remove extra image

Overhaul v1 doc organization

Code style fixes

Rename "documentation" to more specific "user guide"

Update mkdocs.yml to enable git-committers and git-revision-date-localized plugins; fix formatting in application.md documentation.

Remove blog

move cookbook -> examples

add redirects

remove extra .py files from docs

Fix whitespace

Simplify doc themes

correct yaml whitespace

Create separate folder for v1 docs, docs_new

Update Installation Documentation with Modern Tooling

This update provides clearer, more comprehensive installation instructions for Outlines, highlighting modern dependency management tools like `uv` and expanding the list of supported model providers.

Update documentation styling with bold formatting

Remove extra whitespace and trailing comments in docs index

Add code example block to Application class docstring

Move models from reference/ to api/

Remove deprecated API documentation files and restructure the Getting Started section with comprehensive installation instructions and usage examples. Introduce new output types documentation for JSON Schema, Regular Expressions, and Context-Free Grammars.

Update index.md to enhance the introduction with a new title and logo images. Remove outdated installation and usage examples, while retaining essential information about the Outlines library and its capabilities.

Add user guide documentation for chat templating, regex DSL, inference backend selection, and unstructured text generation; update index and structured generation sections.

Simplify community page

Remove extra image

Overhaul v1 doc organization

Code style fixes

Rename "documentation" to more specific "user guide"

Update mkdocs.yml to enable git-committers and git-revision-date-localized plugins; fix formatting in application.md documentation.

Remove blog

move cookbook -> examples

add redirects

remove extra .py files from docs
---
 docs/api/core/application.md                  |  76 ++++
 docs_new/assets/images/dottxt-dark.svg        |   1 +
 docs_new/assets/images/dottxt-light.svg       |  32 ++
 docs_new/assets/images/dottxt.png             | Bin 0 -> 42419 bytes
 docs_new/assets/images/install.png            | Bin 0 -> 12669 bytes
 .../images/logits_processing_diagram.svg      | 157 ++++++++
 docs_new/assets/images/logo-dark-mode.png     | Bin 0 -> 7425 bytes
 docs_new/assets/images/logo-dark-mode.svg     | 207 ++++++++++
 docs_new/assets/images/logo-light-mode.png    | Bin 0 -> 9174 bytes
 docs_new/assets/images/logo-light-mode.svg    |   1 +
 docs_new/assets/images/logo-simple.png        | Bin 0 -> 2196 bytes
 docs_new/assets/images/logo-square.png        | Bin 0 -> 3323 bytes
 docs_new/assets/images/logo-square.svg        | 133 +++++++
 docs_new/assets/images/logo.png               | Bin 0 -> 6386 bytes
 docs_new/assets/images/logo.svg               | 134 +++++++
 docs_new/assets/images/normal_computing.jpg   | Bin 0 -> 64918 bytes
 docs_new/assets/images/readme-dark.png        | Bin 0 -> 38912 bytes
 docs_new/assets/images/readme-light.png       | Bin 0 -> 38943 bytes
 docs_new/assets/images/use_philosophy.png     | Bin 0 -> 16463 bytes
 docs_new/community/contribute.md              | 181 +++++++++
 docs_new/community/examples.md                |  19 +
 docs_new/community/feedback.md                |  81 ++++
 docs_new/community/index.md                   |   9 +
 docs_new/community/versioning.md              |  30 ++
 docs_new/core_concepts.md                     |  13 +
 docs_new/examples/atomic_caption.md           | 189 +++++++++
 docs_new/examples/chain_of_density.md         | 126 ++++++
 docs_new/examples/chain_of_thought.md         | 138 +++++++
 docs_new/examples/classification.md           | 106 +++++
 docs_new/examples/dating_profiles.md          | 230 +++++++++++
 docs_new/examples/deploy-using-bentoml.md     | 224 +++++++++++
 docs_new/examples/deploy-using-cerebrium.md   | 121 ++++++
 docs_new/examples/deploy-using-modal.md       | 198 +++++++++
 docs_new/examples/earnings-reports.md         | 278 +++++++++++++
 docs_new/examples/extract_event_details.md    |  34 ++
 docs_new/examples/extraction.md               |  90 +++++
 docs_new/examples/images/chain_of_density.png | Bin 0 -> 515603 bytes
 .../images/coding_structure_diagram.png       | Bin 0 -> 73586 bytes
 .../images/knowledge-graph-extraction.png     | Bin 0 -> 33127 bytes
 docs_new/examples/images/nvidia-income.png    | Bin 0 -> 185563 bytes
 docs_new/examples/images/simtom.png           | Bin 0 -> 109657 bytes
 .../examples/images/trader-joes-receipt.jpg   | Bin 0 -> 430638 bytes
 docs_new/examples/index.md                    |  18 +
 .../examples/knowledge_graph_extraction.md    | 155 ++++++++
 docs_new/examples/models_playing_chess.md     |  88 ++++
 docs_new/examples/qa-with-citations.md        | 255 ++++++++++++
 docs_new/examples/react_agent.md              | 273 +++++++++++++
 docs_new/examples/read-pdfs.md                | 376 ++++++++++++++++++
 docs_new/examples/receipt-digitization.md     | 296 ++++++++++++++
 docs_new/examples/simtom.md                   | 127 ++++++
 .../structured_generation_workflow.md         | 215 ++++++++++
 docs_new/getting_started/index.md             | 350 ++++++++++++++++
 docs_new/index.md                             | 147 +++++++
 docs_new/installation.md                      |  86 ++++
 docs_new/licence.md                           |  34 ++
 docs_new/logos/amazon.png                     | Bin 0 -> 19489 bytes
 docs_new/logos/apple.png                      | Bin 0 -> 21938 bytes
 docs_new/logos/best_buy.png                   | Bin 0 -> 22328 bytes
 docs_new/logos/canoe.png                      | Bin 0 -> 3719 bytes
 docs_new/logos/cisco.png                      | Bin 0 -> 5784 bytes
 docs_new/logos/cohere.png                     | Bin 0 -> 113536 bytes
 docs_new/logos/dassault_systems.png           | Bin 0 -> 21865 bytes
 docs_new/logos/databricks.png                 | Bin 0 -> 59004 bytes
 docs_new/logos/datadog.png                    | Bin 0 -> 32779 bytes
 docs_new/logos/dbt_labs.png                   | Bin 0 -> 13418 bytes
 docs_new/logos/gladia.jpg                     | Bin 0 -> 11198 bytes
 docs_new/logos/harvard.png                    | Bin 0 -> 31910 bytes
 docs_new/logos/hf.png                         | Bin 0 -> 91529 bytes
 docs_new/logos/johns_hopkins.png              | Bin 0 -> 76335 bytes
 docs_new/logos/meta.png                       | Bin 0 -> 20763 bytes
 docs_new/logos/mit.png                        | Bin 0 -> 1483 bytes
 docs_new/logos/mount_sinai.png                | Bin 0 -> 23749 bytes
 docs_new/logos/nvidia.png                     | Bin 0 -> 15146 bytes
 docs_new/logos/nyu.png                        | Bin 0 -> 30682 bytes
 docs_new/logos/safran.png                     | Bin 0 -> 24435 bytes
 docs_new/logos/salesforce.png                 | Bin 0 -> 20712 bytes
 docs_new/logos/shopify.png                    | Bin 0 -> 8003 bytes
 docs_new/logos/smithsonian.png                | Bin 0 -> 35574 bytes
 docs_new/logos/tinder.png                     | Bin 0 -> 28182 bytes
 docs_new/logos/upenn.png                      | Bin 0 -> 97063 bytes
 docs_new/logos/vllm.png                       | Bin 0 -> 89976 bytes
 docs_new/overrides/home.html                  | 155 ++++++++
 docs_new/overrides/main.html                  |   1 +
 docs_new/reference/advanced/guide.md          |   1 +
 docs_new/reference/advanced/parsing.md        |   1 +
 docs_new/reference/available_models.md        |   1 +
 docs_new/reference/core/generator.md          | 122 ++++++
 docs_new/reference/index.md                   |   3 +
 docs_new/reference/models/anthropic.md        |   0
 docs_new/reference/models/dottxt.md           |   0
 docs_new/reference/models/gemini.md           |   0
 docs_new/reference/models/index.md            |   1 +
 docs_new/reference/models/llamacpp.md         | 230 +++++++++++
 docs_new/reference/models/mlxlm.md            |  93 +++++
 docs_new/reference/models/models.md           |  56 +++
 docs_new/reference/models/ollama.md           |   0
 docs_new/reference/models/openai.md           | 204 ++++++++++
 docs_new/reference/models/sglang.md           |   0
 docs_new/reference/models/tgi.md              |   3 +
 docs_new/reference/models/transformers.md     | 148 +++++++
 docs_new/reference/models/vllm.md             | 235 +++++++++++
 docs_new/reference/models/vllm_offline.md     |   0
 docs_new/reference/types/cfg.md               | 113 ++++++
 docs_new/reference/types/index.md             |  36 ++
 docs_new/reference/types/json_schema.md       |  81 ++++
 docs_new/reference/types/regex.md             |  72 ++++
 docs_new/reference/utility/application.md     |  76 ++++
 docs_new/reference/utility/template.md        | 116 ++++++
 docs_new/stylesheets/extra.css                |  22 +
 docs_new/user_guide/chat_templating.md        |  38 ++
 docs_new/user_guide/index.md                  |   1 +
 docs_new/user_guide/migration.md              | 244 ++++++++++++
 docs_new/user_guide/prompting/applications.md |   0
 docs_new/user_guide/prompting/templates.md    |   0
 docs_new/user_guide/regex_dsl.md              | 344 ++++++++++++++++
 .../selecting_an_inference_backend.md         |   5 +
 .../user_guide/structured_generation/cfg.md   |   0
 .../user_guide/structured_generation/index.md |  13 +
 .../user_guide/structured_generation/json.md  |   0
 .../structured_generation/multiple_choice.md  |   0
 .../user_guide/structured_generation/regex.md |   0
 docs_new/user_guide/unstructured.md           |   1 +
 mkdocs.yml                                    | 367 +++++++++--------
 outlines/applications.py                      |   2 +
 124 files changed, 7858 insertions(+), 155 deletions(-)
 create mode 100644 docs/api/core/application.md
 create mode 100644 docs_new/assets/images/dottxt-dark.svg
 create mode 100644 docs_new/assets/images/dottxt-light.svg
 create mode 100644 docs_new/assets/images/dottxt.png
 create mode 100644 docs_new/assets/images/install.png
 create mode 100644 docs_new/assets/images/logits_processing_diagram.svg
 create mode 100644 docs_new/assets/images/logo-dark-mode.png
 create mode 100644 docs_new/assets/images/logo-dark-mode.svg
 create mode 100644 docs_new/assets/images/logo-light-mode.png
 create mode 100644 docs_new/assets/images/logo-light-mode.svg
 create mode 100644 docs_new/assets/images/logo-simple.png
 create mode 100644 docs_new/assets/images/logo-square.png
 create mode 100644 docs_new/assets/images/logo-square.svg
 create mode 100644 docs_new/assets/images/logo.png
 create mode 100644 docs_new/assets/images/logo.svg
 create mode 100644 docs_new/assets/images/normal_computing.jpg
 create mode 100644 docs_new/assets/images/readme-dark.png
 create mode 100644 docs_new/assets/images/readme-light.png
 create mode 100644 docs_new/assets/images/use_philosophy.png
 create mode 100644 docs_new/community/contribute.md
 create mode 100644 docs_new/community/examples.md
 create mode 100644 docs_new/community/feedback.md
 create mode 100644 docs_new/community/index.md
 create mode 100644 docs_new/community/versioning.md
 create mode 100644 docs_new/core_concepts.md
 create mode 100644 docs_new/examples/atomic_caption.md
 create mode 100644 docs_new/examples/chain_of_density.md
 create mode 100644 docs_new/examples/chain_of_thought.md
 create mode 100644 docs_new/examples/classification.md
 create mode 100644 docs_new/examples/dating_profiles.md
 create mode 100644 docs_new/examples/deploy-using-bentoml.md
 create mode 100644 docs_new/examples/deploy-using-cerebrium.md
 create mode 100644 docs_new/examples/deploy-using-modal.md
 create mode 100644 docs_new/examples/earnings-reports.md
 create mode 100644 docs_new/examples/extract_event_details.md
 create mode 100644 docs_new/examples/extraction.md
 create mode 100644 docs_new/examples/images/chain_of_density.png
 create mode 100644 docs_new/examples/images/coding_structure_diagram.png
 create mode 100644 docs_new/examples/images/knowledge-graph-extraction.png
 create mode 100644 docs_new/examples/images/nvidia-income.png
 create mode 100644 docs_new/examples/images/simtom.png
 create mode 100644 docs_new/examples/images/trader-joes-receipt.jpg
 create mode 100644 docs_new/examples/index.md
 create mode 100644 docs_new/examples/knowledge_graph_extraction.md
 create mode 100644 docs_new/examples/models_playing_chess.md
 create mode 100644 docs_new/examples/qa-with-citations.md
 create mode 100644 docs_new/examples/react_agent.md
 create mode 100644 docs_new/examples/read-pdfs.md
 create mode 100644 docs_new/examples/receipt-digitization.md
 create mode 100644 docs_new/examples/simtom.md
 create mode 100644 docs_new/examples/structured_generation_workflow.md
 create mode 100644 docs_new/getting_started/index.md
 create mode 100644 docs_new/index.md
 create mode 100644 docs_new/installation.md
 create mode 100644 docs_new/licence.md
 create mode 100644 docs_new/logos/amazon.png
 create mode 100644 docs_new/logos/apple.png
 create mode 100644 docs_new/logos/best_buy.png
 create mode 100644 docs_new/logos/canoe.png
 create mode 100644 docs_new/logos/cisco.png
 create mode 100644 docs_new/logos/cohere.png
 create mode 100644 docs_new/logos/dassault_systems.png
 create mode 100644 docs_new/logos/databricks.png
 create mode 100644 docs_new/logos/datadog.png
 create mode 100644 docs_new/logos/dbt_labs.png
 create mode 100644 docs_new/logos/gladia.jpg
 create mode 100644 docs_new/logos/harvard.png
 create mode 100644 docs_new/logos/hf.png
 create mode 100644 docs_new/logos/johns_hopkins.png
 create mode 100644 docs_new/logos/meta.png
 create mode 100644 docs_new/logos/mit.png
 create mode 100644 docs_new/logos/mount_sinai.png
 create mode 100644 docs_new/logos/nvidia.png
 create mode 100644 docs_new/logos/nyu.png
 create mode 100644 docs_new/logos/safran.png
 create mode 100644 docs_new/logos/salesforce.png
 create mode 100644 docs_new/logos/shopify.png
 create mode 100644 docs_new/logos/smithsonian.png
 create mode 100644 docs_new/logos/tinder.png
 create mode 100644 docs_new/logos/upenn.png
 create mode 100644 docs_new/logos/vllm.png
 create mode 100644 docs_new/overrides/home.html
 create mode 100644 docs_new/overrides/main.html
 create mode 100644 docs_new/reference/advanced/guide.md
 create mode 100644 docs_new/reference/advanced/parsing.md
 create mode 100644 docs_new/reference/available_models.md
 create mode 100644 docs_new/reference/core/generator.md
 create mode 100644 docs_new/reference/index.md
 create mode 100644 docs_new/reference/models/anthropic.md
 create mode 100644 docs_new/reference/models/dottxt.md
 create mode 100644 docs_new/reference/models/gemini.md
 create mode 100644 docs_new/reference/models/index.md
 create mode 100644 docs_new/reference/models/llamacpp.md
 create mode 100644 docs_new/reference/models/mlxlm.md
 create mode 100644 docs_new/reference/models/models.md
 create mode 100644 docs_new/reference/models/ollama.md
 create mode 100644 docs_new/reference/models/openai.md
 create mode 100644 docs_new/reference/models/sglang.md
 create mode 100644 docs_new/reference/models/tgi.md
 create mode 100644 docs_new/reference/models/transformers.md
 create mode 100644 docs_new/reference/models/vllm.md
 create mode 100644 docs_new/reference/models/vllm_offline.md
 create mode 100644 docs_new/reference/types/cfg.md
 create mode 100644 docs_new/reference/types/index.md
 create mode 100644 docs_new/reference/types/json_schema.md
 create mode 100644 docs_new/reference/types/regex.md
 create mode 100644 docs_new/reference/utility/application.md
 create mode 100644 docs_new/reference/utility/template.md
 create mode 100644 docs_new/stylesheets/extra.css
 create mode 100644 docs_new/user_guide/chat_templating.md
 create mode 100644 docs_new/user_guide/index.md
 create mode 100644 docs_new/user_guide/migration.md
 create mode 100644 docs_new/user_guide/prompting/applications.md
 create mode 100644 docs_new/user_guide/prompting/templates.md
 create mode 100644 docs_new/user_guide/regex_dsl.md
 create mode 100644 docs_new/user_guide/selecting_an_inference_backend.md
 create mode 100644 docs_new/user_guide/structured_generation/cfg.md
 create mode 100644 docs_new/user_guide/structured_generation/index.md
 create mode 100644 docs_new/user_guide/structured_generation/json.md
 create mode 100644 docs_new/user_guide/structured_generation/multiple_choice.md
 create mode 100644 docs_new/user_guide/structured_generation/regex.md
 create mode 100644 docs_new/user_guide/unstructured.md

diff --git a/docs/api/core/application.md b/docs/api/core/application.md
new file mode 100644
index 000000000..242caa135
--- /dev/null
+++ b/docs/api/core/application.md
@@ -0,0 +1,76 @@
+---
+title: Application API
+---
+
+# Application
+
+::: outlines.applications.Application
+
+The `Application` class enables you to encapsulate a prompt template and an output type into a reusable component.
+
+## Overview
+
+An `Application` combines a prompt template with an output type, creating a reusable component that can be applied to different models.
+
+## Basic Usage
+
+```python
+from outlines import Application, Template, from_transformers
+from typing import List
+
+# Create a template
+template = Template("""
+You are an expert classifier.
+Classify the following text into one of these categories: {categories}
+
+Text: {text}
+""")
+
+# Create an application with the template and output type
+classifier = Application(template, str)
+
+# Initialize a model
+model = from_transformers(...)
+
+# Use the application with specific parameters
+categories = ["Sports", "Politics", "Technology", "Entertainment"]
+result = classifier(model, categories=categories, text="Apple unveils new iPhone with AI capabilities")
+
+print(result)  # Expected output: "Technology"
+```
+
+## Structured Output
+
+```python
+from pydantic import BaseModel
+from outlines import Application, Template, from_transformers
+
+# Define a Pydantic model for structured output
+class MovieReview(BaseModel):
+    title: str
+    director: str
+    year: int
+    rating: float
+    review: str
+
+# Create a template
+template = Template("""
+Write a brief review of the movie {movie_title}.
+Include the director, year of release, and a rating out of 10.
+""")
+
+# Create an application with the template and output type
+movie_reviewer = Application(template, MovieReview)
+
+# Use the application with a model
+result = movie_reviewer(model, movie_title="The Matrix")
+
+# Parse the result into a Pydantic model
+review = MovieReview.model_validate_json(result)
+print(f"{review.title} ({review.year}) - {review.rating}/10")
+```
+
+## Parameters
+
+- `prompt_template`: A template that defines the prompt structure
+- `output_type`: The type of output to generate
diff --git a/docs_new/assets/images/dottxt-dark.svg b/docs_new/assets/images/dottxt-dark.svg
new file mode 100644
index 000000000..a95ff7d32
--- /dev/null
+++ b/docs_new/assets/images/dottxt-dark.svg
@@ -0,0 +1 @@
+<?xml version="1.0" encoding="UTF-8"?><svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 129.09 196.26"><defs><style>.cls-1{fill:#fff;}</style></defs><g id="Layer_1-2"><rect class="cls-1" x="39.12" y="3.91" width="3.91" height="3.91"/><polygon class="cls-1" points="46.94 7.82 50.86 7.82 54.77 7.82 54.77 3.91 50.86 3.91 50.86 0 46.94 0 43.03 0 43.03 3.91 46.94 3.91 46.94 7.82"/><polygon class="cls-1" points="58.68 7.82 62.59 7.82 66.5 7.82 66.5 3.91 62.59 3.91 62.59 0 58.68 0 54.77 0 54.77 3.91 58.68 3.91 58.68 7.82"/><polygon class="cls-1" points="70.42 7.82 74.33 7.82 78.24 7.82 78.24 3.91 74.33 3.91 74.33 0 70.42 0 66.5 0 66.5 3.91 70.42 3.91 70.42 7.82"/><polygon class="cls-1" points="82.15 7.82 86.06 7.82 86.06 11.74 89.98 11.74 93.89 11.74 97.8 11.74 97.8 7.82 101.71 7.82 101.71 3.91 97.8 3.91 97.8 0 93.89 0 89.98 0 89.98 3.91 93.89 3.91 93.89 7.82 89.98 7.82 89.98 3.91 86.06 3.91 86.06 0 82.15 0 78.24 0 78.24 3.91 82.15 3.91 82.15 7.82"/><polygon class="cls-1" points="105.62 7.82 109.53 7.82 113.45 7.82 113.45 11.74 117.36 11.74 121.27 11.74 121.27 7.82 117.36 7.82 117.36 3.91 113.45 3.91 109.53 3.91 109.53 0 105.62 0 101.71 0 101.71 3.91 105.62 3.91 105.62 7.82"/><rect class="cls-1" x="43.03" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="54.77" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="66.5" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="78.24" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="101.71" y="7.82" width="3.91" height="3.91"/><polygon class="cls-1" points="39.12 39.12 43.03 39.12 46.94 39.12 50.86 39.12 54.77 39.12 58.68 39.12 62.59 39.12 66.5 39.12 70.42 39.12 74.33 39.12 78.24 39.12 78.24 35.21 74.33 35.21 70.42 35.21 66.5 35.21 62.59 35.21 58.68 35.21 54.77 35.21 50.86 35.21 46.94 35.21 43.03 35.21 39.12 35.21 39.12 39.12"/><polygon class="cls-1" points="66.5 50.86 62.59 50.86 58.68 50.86 54.77 50.86 50.86 50.86 46.94 50.86 43.03 50.86 39.12 50.86 35.21 50.86 31.3 50.86 31.3 54.77 35.21 54.77 39.12 54.77 43.03 54.77 46.94 54.77 50.86 54.77 54.77 54.77 58.68 54.77 62.59 54.77 66.5 54.77 70.42 54.77 74.33 54.77 74.33 50.86 70.42 50.86 66.5 50.86"/><polygon class="cls-1" points="43.03 101.71 46.94 101.71 50.86 101.71 54.77 101.71 58.68 101.71 62.59 101.71 66.5 101.71 70.42 101.71 74.33 101.71 78.24 101.71 82.15 101.71 86.06 101.71 89.98 101.71 93.89 101.71 97.8 101.71 101.71 101.71 105.62 101.71 109.53 101.71 109.53 97.8 105.62 97.8 101.71 97.8 97.8 97.8 93.89 97.8 89.98 97.8 86.06 97.8 82.15 97.8 78.24 97.8 74.33 97.8 70.42 97.8 66.5 97.8 62.59 97.8 58.68 97.8 54.77 97.8 50.86 97.8 46.94 97.8 43.03 97.8 43.03 93.89 43.03 89.97 46.94 89.97 50.86 89.97 54.77 89.97 58.68 89.97 62.59 89.97 66.5 89.97 70.42 89.97 74.33 89.97 78.24 89.97 78.24 86.06 82.15 86.06 86.06 86.06 89.98 86.06 93.89 86.06 97.8 86.06 101.71 86.06 105.62 86.06 109.53 86.06 109.53 82.15 105.62 82.15 101.71 82.15 97.8 82.15 93.89 82.15 89.98 82.15 86.06 82.15 82.15 82.15 82.15 78.24 82.15 74.33 86.06 74.33 86.06 70.42 89.98 70.42 93.89 70.42 97.8 70.42 101.71 70.42 105.62 70.42 109.53 70.42 109.53 66.5 105.62 66.5 101.71 66.5 97.8 66.5 93.89 66.5 89.98 66.5 89.98 62.59 89.98 58.68 93.89 58.68 93.89 54.77 97.8 54.77 101.71 54.77 105.62 54.77 109.53 54.77 109.53 50.86 105.62 50.86 101.71 50.86 97.8 50.86 97.8 46.94 97.8 43.03 101.71 43.03 101.71 39.12 105.62 39.12 109.53 39.12 109.53 35.21 105.62 35.21 105.62 31.3 109.53 31.3 109.53 27.38 109.53 23.47 109.53 19.56 113.45 19.56 113.45 15.65 113.45 11.74 109.53 11.74 105.62 11.74 105.62 15.65 105.62 19.56 101.71 19.56 101.71 23.47 101.71 27.38 97.8 27.38 97.8 31.3 97.8 35.21 93.89 35.21 93.89 39.12 93.89 43.03 89.98 43.03 89.98 46.94 89.98 50.86 86.06 50.86 86.06 54.77 86.06 58.68 82.15 58.68 82.15 62.59 82.15 66.5 78.24 66.5 78.24 70.42 78.24 74.33 74.33 74.33 74.33 78.24 74.33 82.15 70.42 82.15 66.5 82.15 62.59 82.15 58.68 82.15 54.77 82.15 50.86 82.15 46.94 82.15 43.03 82.15 39.12 82.15 35.21 82.15 31.3 82.15 27.38 82.15 23.47 82.15 19.56 82.15 15.65 82.15 11.74 82.15 7.82 82.15 7.82 78.24 7.82 74.33 7.82 70.42 11.74 70.42 11.74 66.5 11.74 62.59 15.65 62.59 15.65 58.68 15.65 54.77 19.56 54.77 19.56 50.86 19.56 46.94 23.47 46.94 23.47 43.03 23.47 39.12 27.38 39.12 27.38 35.21 27.38 31.3 31.3 31.3 31.3 27.38 31.3 23.47 35.21 23.47 35.21 19.56 35.21 15.65 39.12 15.65 39.12 11.74 39.12 7.82 35.21 7.82 31.3 7.82 31.3 11.74 31.3 15.65 27.38 15.65 27.38 19.56 27.38 23.47 23.47 23.47 23.47 27.38 23.47 31.3 19.56 31.3 19.56 35.21 19.56 39.12 15.65 39.12 15.65 43.03 15.65 46.94 11.74 46.94 11.74 50.86 11.74 54.77 7.82 54.77 7.82 58.68 7.82 62.59 3.91 62.59 3.91 66.5 3.91 70.42 0 70.42 0 74.33 0 78.24 0 82.15 0 86.06 3.91 86.06 7.82 86.06 7.82 89.97 11.74 89.97 15.65 89.97 19.56 89.97 23.47 89.97 27.38 89.97 31.3 89.97 35.21 89.97 35.21 93.89 35.21 97.8 35.21 101.71 35.21 105.62 35.21 109.53 39.12 109.53 43.03 109.53 43.03 105.62 43.03 101.71"/><polygon class="cls-1" points="125.18 11.74 121.27 11.74 121.27 15.65 121.27 19.56 121.27 23.47 121.27 27.38 121.27 31.3 121.27 35.21 121.27 39.12 121.27 43.03 121.27 46.94 121.27 50.86 121.27 54.77 121.27 58.68 121.27 62.59 121.27 66.5 121.27 70.42 121.27 74.33 121.27 78.24 121.27 82.15 121.27 86.06 121.27 89.97 121.27 93.89 121.27 97.8 121.27 101.71 121.27 105.62 121.27 109.53 117.36 109.53 117.36 113.45 113.45 113.45 109.53 113.45 105.62 113.45 101.71 113.45 97.8 113.45 93.89 113.45 89.98 113.45 86.06 113.45 82.15 113.45 78.24 113.45 74.33 113.45 70.42 113.45 66.5 113.45 62.59 113.45 58.68 113.45 54.77 113.45 50.86 113.45 50.86 109.53 46.94 109.53 43.03 109.53 43.03 113.45 43.03 117.36 46.94 117.36 50.86 117.36 50.86 121.27 54.77 121.27 58.68 121.27 62.59 121.27 66.5 121.27 70.42 121.27 74.33 121.27 78.24 121.27 82.15 121.27 86.06 121.27 89.98 121.27 93.89 121.27 97.8 121.27 101.71 121.27 105.62 121.27 109.53 121.27 113.45 121.27 117.36 121.27 117.36 117.36 121.27 117.36 125.18 117.36 125.18 113.45 125.18 109.53 129.09 109.53 129.09 105.62 129.09 101.71 129.09 97.8 129.09 93.89 129.09 89.97 129.09 86.06 129.09 82.15 129.09 78.24 129.09 74.33 129.09 70.42 129.09 66.5 129.09 62.59 129.09 58.68 129.09 54.77 129.09 50.86 129.09 46.94 129.09 43.03 129.09 39.12 129.09 35.21 129.09 31.3 129.09 27.38 129.09 23.47 129.09 19.56 129.09 15.65 129.09 11.74 125.18 11.74"/><path class="cls-1" d="M.31,196.26v-7.35h7.35v7.35H.31Z"/><path class="cls-1" d="M19.53,196.26v-44.17H4.81v-7.37h36.79v7.37h-14.72v44.17h-7.35Z"/><path class="cls-1" d="M46.63,196.26v-11.02h3.67v-3.7h3.67v-3.67h3.67v-3.67h3.7v-7.37h-3.7v-3.67h-3.67v-3.67h-3.67v-3.7h-3.67v-11.05h7.35v7.37h3.67v3.67h3.7v3.7h3.67v3.67h3.67v-3.67h3.7v-3.7h3.67v-3.67h3.67v-7.37h7.37v11.05h-3.7v3.7h-3.67v3.67h-3.67v3.67h-3.67v7.37h3.67v3.67h3.67v3.67h3.67v3.7h3.7v11.02h-7.37v-7.35h-3.67v-3.67h-3.67v-3.7h-3.7v-3.67h-3.67v3.67h-3.67v3.7h-3.7v3.67h-3.67v7.35h-7.35Z"/><path class="cls-1" d="M106.9,196.26v-44.17h-14.72v-7.37h36.79v7.37h-14.72v44.17h-7.35Z"/></g></svg>
diff --git a/docs_new/assets/images/dottxt-light.svg b/docs_new/assets/images/dottxt-light.svg
new file mode 100644
index 000000000..567e239ee
--- /dev/null
+++ b/docs_new/assets/images/dottxt-light.svg
@@ -0,0 +1,32 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" version="1.1" viewBox="0 0 129.1 196.3">
+  <!-- Generator: Adobe Illustrator 29.0.1, SVG Export Plug-In . SVG Version: 2.1.0 Build 192)  -->
+  <defs>
+    <style>
+      .st0 {
+        fill: #1d1d1b;
+      }
+    </style>
+  </defs>
+  <g id="Layer_1-2">
+    <rect class="st0" x="39.1" y="3.9" width="3.9" height="3.9"/>
+    <path class="st0" d="M46.9,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M58.7,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M70.4,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M82.2,7.8h3.9v3.9h11.7v-3.9h3.9v-3.9h-3.9V0h-7.8v3.9h3.9v3.9h-3.9v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
+    <path class="st0" d="M105.6,7.8h7.8v3.9h7.8v-3.9h-3.9v-3.9h-7.8V0h-7.8v3.9h3.9v3.9Z"/>
+    <rect class="st0" x="43" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="54.8" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="66.5" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="78.2" y="7.8" width="3.9" height="3.9"/>
+    <rect class="st0" x="101.7" y="7.8" width="3.9" height="3.9"/>
+    <path class="st0" d="M39.1,39.1h39.1v-3.9h-39.1v3.9Z"/>
+    <path class="st0" d="M31.3,50.9v3.9h43v-3.9H31.3Z"/>
+    <path class="st0" d="M43,101.7h66.5v-3.9H43v-7.8h35.2v-3.9h31.3v-3.9h-27.4v-7.8h3.9v-3.9h23.5v-3.9h-19.5v-7.8h3.9v-3.9h15.6v-3.9h-11.7v-7.8h3.9v-3.9h7.8v-3.9h-3.9v-3.9h3.9v-11.7h3.9v-7.8h-7.8v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8H7.8v-11.7h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h-7.8v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8H0v15.6h7.8v3.9h27.4v19.6h7.8v-7.8Z"/>
+    <path class="st0" d="M125.2,11.7h-3.9v97.8h-3.9v3.9H50.9v-3.9h-7.8v7.8h7.8v3.9h66.5v-3.9h7.8v-7.8h3.9V11.7h-3.9Z"/>
+    <path class="st0" d="M.3,196.3v-7.4h7.3v7.4H.3Z"/>
+    <path class="st0" d="M19.5,196.3v-44.2H4.8v-7.4h36.8v7.4h-14.7v44.2h-7.4Z"/>
+    <path class="st0" d="M46.6,196.3v-11h3.7v-3.7h3.7v-3.7h3.7v-3.7h3.7v-7.4h-3.7v-3.7h-3.7v-3.7h-3.7v-3.7h-3.7v-11.1h7.3v7.4h3.7v3.7h3.7v3.7h3.7v3.7h3.7v-3.7h3.7v-3.7h3.7v-3.7h3.7v-7.4h7.4v11.1h-3.7v3.7h-3.7v3.7h-3.7v3.7h-3.7v7.4h3.7v3.7h3.7v3.7h3.7v3.7h3.7v11h-7.4v-7.4h-3.7v-3.7h-3.7v-3.7h-3.7v-3.7h-3.7v3.7h-3.7v3.7h-3.7v3.7h-3.7v7.4h-7.4Z"/>
+    <path class="st0" d="M106.9,196.3v-44.2h-14.7v-7.4h36.8v7.4h-14.7v44.2h-7.3Z"/>
+  </g>
+</svg>
diff --git a/docs_new/assets/images/dottxt.png b/docs_new/assets/images/dottxt.png
new file mode 100644
index 0000000000000000000000000000000000000000..1bdf8cb61ff415b10450a9f4a80bb235f048e694
GIT binary patch
literal 42419
zcmZU*c{o&W{5L-KY?Cd9jG{!v*ptyRg`t$}`^b{Wl5h}`C2OXUQbV>G(;{Tama!!J
zZYFDl%65=^5F*QS_kFJ4bA7Mt_eaguoS8G{ocq4t@7H#hU~Ht%&m+NuKp^-p8R(iK
z5G-`?r+ObJc&9wy_%H$yfw-i5!Tj#X&sy&;#aSCE`shls!07}Y+1u`S9x7hkY#Zm7
zQ;#%I@$N18-gfkj_9u4Rb)!aUmUICf>D+tKeF6m{hrCb9HJMDdZMVG@e$|%joh)js
zV|el<#o2gDK)7{TU-C)LQN6>PTtZ7WrFXOE)Hc2z8VcxD^L)24)@zV7UVigk&{?(8
z!7h!Vu<IKE*;;+%y5Fyq!sZOGZ~3Ju4tce72A-+&{j^?pcmL?$ZI02uilM{fJ~qX*
zy%UY#%}?w1C#5PjUVsAu;J{w(^_eZdDy0pd7AC3n6?pg4S+Au`-inQ1;B^eV_JCI|
zF|Vap;B^Q5dv?9JHX9~a)j^a6+)!j8+gM@5uf1QmA@cuwhrmS#*Te>1kkDD^OJpIq
zy#M(q!3{#gTc62D@PcIo;Q#k`+#qld|Gx46KlHy&_wV4^E2aPW%>VuU|99?x-~C^w
z_}`z}2Zsb-xW4to-XZW_WKOHn&S#~t^&ah|GtR9ud(SiO?!QsFan&cJ?~q%?2046_
zO!iu8eLwnF?fj}*(#?t>6u52X&-L9uNqc|(z2E=loJMtRc*fZHL7$rC$e-)?H_73_
zfuZNq$HuKbZVha>Rc?Igy&o7Dx~EWG%iB7%X6S$27BhFJZD(U5{P(|Kc&ehXcHR5l
z58>?LHNsrrz5gB@Qm-@{9e;mscQt#WbA3E)V{GQsv*wZYFt4RM7X~V>jo#kOdUJkf
zPJ3e9kc_DsJ-IiyRJGD^GN`WB=icPOfyE1?)i0%P8!PG7%7t^}nL0^6M!Rq?Th05Q
z2jQazcSrB<!@WUI>TqCvLTf+9u|gwPqbu%D;N*IbFIIm2H%7PiL4(|{3DNsm;4{x(
z)&}k)V(3rFOQV{!L{ib$>@ZeB)3>r|fd~={{O?KQDwUzRu_{loA3A&l+fnTHS#?Tc
zM2Ph|H{=jMK<CD?<Ktpu!O3zKQ%3*&02-Ev#k|v=3>iBYiJ|kJYm!7aV5d$U5@G~P
zSLkpP@ZWq5h?tJvTbj6kvm~<c0xZ<shn8UI{YVL8BZ%ab)o%(k$xBV9EPWAU1Rwss
zbCw-H8&f`gE`YKCrVgIkDSt0`pR)))-p>Hn8XdQxh%dRAKdu*?jf>NjnVJk^$qaG`
zZvDVUxQ=CL>?7Fn81fq+^#a5%mG0+&Wnwbd(UMUp4t|bb+EAKFh15k@h-Gf2=Kp@&
z5wf-JrsED>hCiDXp>0XX_U`^pX=SW>cA3wB?32o~Ws+#qV|~r?U=Hd~bbiI}f5g;t
zCh0gqU4*|>5#kZCaUE9d(Q9%;hc3dZBja1aU|K^I&wWt5$gc4`%=kJa9gPt@6=Xm)
zER$Lkn1_w=K?YjbL20m({#~yKO}F0HAOK3PZlkJTmCpOiYhhd7O$sc)kH7w)o~BVe
zzW&bwXp|&CV&ULm(Fe<F82z4OSXvGH()iv4)i5?-a5P*#&#);i7o2nJeH4fF)HmOF
zTu>X4<V2DC)W6A4sJG7U)KLq-D|`YgZCYJn;)8hhwMW^zb8@|WVk=F5K8w}5hekTL
zlEWG`C0%E?3%R6D(>KOL%_d`%?m2mcX0H{vy_&tZ6L&6Ud`Ky{`K0M+HOHf=45fQ8
z^>>{^_|DqgV%SpXMEZ+WvAxxBqooD2+jUyCwc(46hm5*6gARqC3A=ZFEzjyzWstkZ
zv$R#h-meLCW^m&{zsW;8G+Eq+Tf6574UIHf^26*r82KujB|Zha@_9L;rc2gFl-~y>
zG>+Zbsamri$rzYeP3LX)*%|clZmnl@xNO?jog(_r_PnatxOFXbWPP>6Z7}Jbi9>kf
zp<Qa-^|j@K>-ifSuavs4%NHz}Nx78}ly;w<_f2{mz!%#(<Pho|61r5qbS^+@>F)l8
z+tqu!e?EA$-bJ0?qHRuU>&g}^T?@Tm^K1V6U$xarv-Lh~ix|e<z-LK&V$*2jUP;!+
z*@?ZsG_~{KwgWOzyU$I|ACmIy72B2)CH4lLhH<lOgfP<%Dk0tuk-2?uzoVuLf(~DF
zC!olPOpY7O8aqm9L<?KybDSI6Zu;*toZFbXyUTG#J8zAqPw<W@3w9ysXj0HlmPTaH
z>0d%{Sy*|N5(jqiPrLWy)_d&O{ogvU+_7M3`Q5>}5xi;>mxU;+GyWLS7$#M0Vk^v0
z9IYD1UZ2biZXLvq8%FkQ?J5KY^FgBV`t?H(nbLfIEpW}j*X5|p*8WHSHxbwc6#m=&
z<sZ$>;DXFN$Hf^`ni{=MQk@~KkblOZ1C}}V#=c;I@k(rqiII4R$`lKEvS5yni&n8?
z%{-NT>eNyn3N0aZn;SAoEju&8ni=SSLr;3X)l-@7C_>TE`Svanxx-KXh+$mflp6}7
zEOcIAA+Y<2PI`QTbE=(}tvi`E-;N!s>|h3q@dig`DbwY-P9*szGF}A{pP8Ttv7ajT
z5kfRijChtxW^Rl=OR`hL^4g;*s}cC99#cx9-^j@F7;C2Ji<Drk#Wk>YwEjq<1!g=j
z_>x~;`LnY6Zn8^?oQ%uFA2Dl(@o_&rJ0&w4$IE#A(exWSjfyC=%pVhRg2<`<`<;pr
zBtbuMXOV@CBiWU0%wVGZC=v_VJowiMC6Cfpqvb{>EOBDf`FXP%4C(J>mRfX!7<c|>
zOzQXVvN-)%g|jM`*=Jk{hAKQcNJ)PvVaJ{yjxsRlh99TLZ)pzlz^Mn*xEZ$;dRXZ_
zU;vA?u-8;xvs~+B$Q}+OYhfR?2a6G3;bxCd6;yRgW>yvw%Oj1ndNdfpRGY=rB58aA
zrb$?Z=jjsR2*f=D%Qz*KPbgj_8~h<~V8x&7t_1v>6~(x?wwHEOumI;=KU4alO0d&y
z7RbR!XH=@R2JvO$6;9ti;s`~j<9J{*d^6ARrLJD6_f?A8pc>I=x5xG^hFLuQl0@Ps
z1cJW)fRB)4`un+aO0F{D3lemM`$kd5ohwlWQRf9|cgy`9YKKN}Z@k!eppl8^&LU;*
zVKVp&*7mh^0kBZ6T?(#Ma`O2!hkJ0N0wjKbxFJcnj&vYKISAMdGWGo9TeId>okKjQ
z;F`oHKKOA+#nxHH4qgMj965YKs^v+1To-RX=4_0oAw}o@$H;lrkv+?4j%0>h0d0ZW
zS(o9i!eeS`{ILQbe4$=f0Uwvqq__YxqU$_ZVq*{3y~PN&IH+KCZSo@T8B!7(8eY!}
zm0Np?({)!LSdfs%X^(sVU4`rs4>Td=4ZuKTVK5&XMhCU%F!HN00bijvc&SE|5nQj#
zf~}xbS-`417p#ixaB=ZIgAl%?LjM5mZfk{9wE!(%>#SQ!+iyz1r_(|lzKZoBq(UV*
z4mfKwR#P%_XM5Ic@_1vy#uGAQGSOlA;qQ#1d%u~B&5CM;)!$>itAzq<zUe7l?hT%7
zJ@<TkXm46BTs7ccTyo~>mRy#>(!%CJpDXAIr|pHkt?Op~yZtkGnd+kOg+aT%){TWj
z_mnke&u@3ReQw?O!FlF=>xP@p#~rw0<72ke`3;}FMP}OfGuMb8*D@#4#>evu=^mpz
z$5mnrxrUCm4h6_-y~1hs&m3O6Z0kOm*FE+&|ET(}*F|?%E)8Ec9KSVQ#zG>#xPyAY
zqMUape`B+5V%OWDFns$D{rs*=+wDu~8>6Rw^5l7&bK3nRWPQBle!lUKrsWbE$ECtE
zn00<pX;;QF=Y+y7#QqfN_92JeKa;7-23n#^o0${78{MiJcW0M&F>m<Z?K*rdHW++Z
zc&07{5zu+>i;GuLc-RX<?e?_Nwb>&+)r%Fh?1`>kQufx{Y}*FTrJu%r4<ku$_01xY
zi}&tq*Rev$52`Ep)hMf&?H|D`*uf4@UZ@?pUyu<ToU6#7iaBz@IZKsU;k*!$c3Fdy
z;xgYjKPyZ*a{B})m|B|3j7TkWZb&}&S_G-=)>B-o$huD?GC<u%92VF}ryl>saJOVD
zeeV!j3&1c|46Zp{yY@g`oPa>EG4z5OpDR0o6Tc`P;pB#LyZhVV$6FUm&ayG6UPg6p
z4RDTRx^)sy;M4>+jqOUs!kPwy>Zs-f6*RWbELFMS<F@qkaB;bE2PitXk<$?oJ*Qhl
z8Pb{ujxb81p5Sq<757dMZzI(94^RCMsFufz4U7<pJ!X~S3t-O2U%82kLtHyRzcFJf
zjs`auflLcRl`>rWuPZu2`w&-AXfwWd2($<J6bts^r_+4g5b?y|4X7T=3+JRaD>C$u
zIL={$GAJTfB_-+hEU?V?#1pD{7s(O~8?T~&_qz{feArWf5`I57L!pH}=N?MOajAbT
zh$P+m>cNV3{A|yLK6%nA62l|ML(viMiDpWRg>u|jkR<dET-eQ2!9KRhd>4T%8O*D6
z;f30|*|?#s<WKvcT%O92!UouQHA<XeL+oLz!-K-AtU)>HRR}=Mi&#vu_VO0jJZt9D
zUMD}uZ-*Ub?B3>KNNbgHGlIX(|AB0Aan|HVxVTe72Z+cj-^ZgjADTqeJOTL1+ijl8
zO6ZT=No7haNXz1V>wLJOw#^(>>{6)6-;fIkw8U1BKa^~;%0o5mJ^B0q?Ki`f3;Ti@
z9AquAugecgcJdXV>!wccgDC&$&A<W_6AN;(JbN2ckBH<U$iy&NVdI;cEOdFVx00SG
z_2c@{MiIzm_5)Nyd?bn$EkIYqwFXun@`vK-=J!5|U7$Q@lGjLImB>BCiS6e@vtYUX
z8dv{Oq+Et;Q{gc!?6H?_DPp<^bz$0J>LFh$ql33*N5O?1`-4?71|vA|mB`eS{^5Zh
z{e|ylN>AkH&~<NHH!x)P6~3=Q+!_SKFrsULKr7-eLC1=k<>PzvzMc|>0|n^Bi^F-7
zSCy3DPTK(J=bgwNtz=o6ORSAPm%Ta<&1KfyAEL02n-XLi6_gMjg$${%J81ObtlUWC
z+;y=$OyuLTW>_Frvz<xRydt9?Cpo#bd?g?uqz>yWawB@0L%R@gii3@?$IX8X63qwY
zDu7P(?e>5f0Uw;n&Co+LnpeSNtzxaZ-vBet7*Xg~)BxtSeq?aTqkHbT7(3OLA2ybq
z0`1{hp7}I~3Fh-i)}>&UzBrR9*w{Eb8k4-*jG*Tp%6ZO{nu`enDEs;?a|ZzydqvxS
zTojo{TaC4Gz<qG*&+F{U{9vn$To3LuC&$lb^FV?O9u=O+?pt{n?$=L`D@8QP_i$*P
zKy5grA0X;~q~$rD;dx-!XimUSAH)T-JfpF#*6F}P2gVs3lcK}#?DgH=TF_QlZCr-@
z!AO}>4UhIa#$q05TPQ*a@+B>jXlLBJf=Y#Y8LPxJ4%CPs0f7#^=SJ<!4JT*_T&k?-
zbs-_=VlfVt^YG)0&-<X{<+9T)wrKR0U92C}5|q!9DR#iPllSp|uI0ZjA|R>wpXp5L
z*}$_e)h>6B9IaoI1gOJ2m}R~_HxI+Y@_<ZBY?7v;nv*?;dgQMNt`V7n_sLlt*5+)k
zAq3X?2jg0Cd4{@cb9V`!iKJV%%)z{wxKx!TN;6L?UtxW@upQ0R6ZLa;*P`!pmvbP8
zb}&_D)1T4$8dCnT@K>F@pq<J*5fJObLhZ5iW1u9l<hMQ_a)my3ckmj!iLgLzzFBG5
zBUC3>J4Us;*2gp&TRjDVcENsA#U4J2S%6X076pBMUD+EXq#~NyEI;61LAu=nTU({H
zIE%b4YZ0QkZ&o28IiJm{?hj3*lSMTvd)0HV(@Cn6KL|nIp*5~8p67O#75^)qjFSqH
zKRdhOW3=9t(TK9$zA<yJ*(K@K?Dp*b^BLnqhlaO(0>^)5Ol-6rnbX>vn+Y#n-TiMy
zGdXRQOY`HWpBY8A|JiN*>h_UQo89ob|KN*OoFpOediL&M<-Fbx!T&TY2{~txY5`ze
zje1rRy2PNh0joIa^9#s3r-LtEh+E!5r&ssoGVId#3cJUYy_95U&_=euk+yZ`zV1{9
z%jp=a7W1dXRH0j##skX>AB&x=+{38b$XuhDfxodW`NYY4JAX#_L$6lldF>(##9fD<
z^5zWQJ)rOi6~dv~JifNwdvE1Zm&8>fz43IqU5)XJ+Nf!%rynaPdQ_@fI3}-jA7X<t
znS@xJn7F9t`3eBZjpWj2K}(U&3-OA?%zNC9hSVuK7$WEri(#Zlq@cNbuRcX*1_E_x
zra*tVcI<;dCvwstwUX}ujAI%B2(!BcsKskt3M+Q-1A-MhTa7`tr1h{#Go;5~3c(D4
znTNz4UakGkDQ!dVXJHH=zl&v4TZwD!!qcNo3o%S-=eVs+b~yfrCpYBYX!wjN7&F5G
zr-mscRpP{gPfN5~yFg6-t2RkI22;($&QRMhD|)rovM89?Gn&Os)2SBNM@dYj1tv>H
zA~(k@M=eo=8PxR@T%5=Ya+L|b@<EjE6er#VG6@Hbjx0#i`T2T&?F5ZYJH`+{`$cJB
z;U?A4yrobXdo=pA3uF+k8N<}$m>CJY8qqVV@2Y*2A^pfsj@WZjRTGo)gsC#tE{79a
zd?QT1AyQ(8kGs6q><R8tMU@V}*4wm0p-*(A>Rrcu@9+xcq?CGn4D@1&jUWJ^UI!)x
zMhzb~xc(t1;oDdVC$w*vL`0^wwlhnD_Jk=q)$TsK&n3}j$CO>5i1dW76)dno1d|ks
z!UTXWN=RMTJ{XCtfkUZ=O{SjtJ*$4u1%tXCCiUFQ={0d$BF`8aFIIm0M7UB)x+&%7
zMGkCY?X8BCl@Cs6^k*{0l(7Db0}UVMl?hfQ6kL;fjqln=klB>9AhO1e9A&5|8iIE5
z<D&g6JUs69R1(co$jOkdt!ZJ%o-j3Tk~g4Mw!@E~u*WQ3!iyciCyD7MGFcR^b%L%@
z@`9GwagMY%NF=onT#a_26uUpCtwwau-(WCU`KoO4LD5$jk1R|yIb0c=%oIEq9ssNQ
zjWRH%9wcPfb$(pyhwG_Cl6x491wF4W6pKOE_gy5%mk&G#<FMAU9Y8|zzwP#e43MT>
z*VV9Pu1_WCN4+|aGN`qzazvyfX>HpQ7Fs4)tj^U3V_hJL`giP*gH!d!L7fFSBt5J1
zLk*%^?TS%WQ*#cIpn7|)c;2+?v>3qkxRqrhGx8e-6w9WqqfmSNZFZW@i_addZ1Y26
z(HN7Gqz;&IEUz_OQRTAv<FHGym{0aMz_S^R!6{=ay1qG^JA6Ctu)z+&kJ#pkpS6iW
zln4HNYATmsimKteVS(N@ahguLkp$K7Sko(3bZ$*SvpoE@6pJ-ab&AF?Z9j3-T!#HX
zH*I)*f<b-Qdje|jFaI>n(Xuh78zBDnR`VpCgrTdL`9rY16eIX@C;Bah=9l@4_&E{-
z;79=EL_+@7?2kbD&Sy`-kFi)BF>0#lCdcIA!~qmKMDv^=jl%B|Gyg>6K~{>Ga?B(v
zoVx2|9Lz$c>!$U1gK4|%7{yG(>2E)u$6}klS-%6*sJVPK7gXP$UvnM5n#tQGCB7}4
z^#mfV8ur7)sNUNtnJdw+)Odajs|eB#&z~t#l+_oT)+`G{M`HY1#1_g{Sfee{w6KiS
zDrI`^bE8qwGB)}>sRWan>N^$z3DMVtiMOBAbnGuS`CP`*4risQ4ahbO@>5)hb#Avf
zkM7y}v}R?R3|m^*e)^yu_>0=vS>q@<m*o2J-w9|nh<8fqD%ewy%`_}<P<cQ+M&9{3
z5$WxB%Qiyr6^9U<y4i@u#&PI6i!3#Gf%Yb<ia>XG-JIqS)X}VHe~-{vx$J^Q_j!LR
zM(B+Gy5ixG_L+4*d?t1X1*f(}58W*nBW|p)jvU|4Ik3U?^$aM5Bk^74WD_w~6J^rx
z(|zk>im0GpT{J8<iSPncC$AGgRD5IR!#28hc7?8sF+ahOmcB)EJoDsLz606B`!y&N
zBRp?=<Z*ESD*eh>N={U&JoJ{^#8Mume_4}<VY4$E^69P>!8Ep&I^=rT=xt<={|Oaw
z^beNf4v9_jR5hNy0*iL0bk@_?L$d7|<qX*^$@#_CKdYySQMof$+)_AR_b_WJgQqzt
z^G~=3UO0;wf4ecjl>Y01zaC-q(*uqE@%TLuEyXox3l3y7TrtU;wNc}pRM~|R`pWj-
zjh5B(8&&-o*Z1CpzsT};Uw%3^L>;od*`s~_H}!YM#9EW^T-Z))_AV3x+Tz~Ikwcey
zH(mChOU>HQ3g4Z5tHtHm(-YhXTIj~w;@;xNS2Hzh8{y~A&XSQ~f=YyrN3-WP;4ZbS
z3rR(PUKDB37YeIQC$mC-T}rbWA{M4&gD8KqCx6;tt&APWgv2!MV(&?de(Q>m>_I2V
z+R1|li*nvM=4~2IS_EE^D3S@dJ3yx{d(7x&L#S*0->I2Lr!t-*C*}M0gM&=Y%Ek}E
z&lMh7(J-SDo};H6^z6(NW?wCz_&isAPHMCz-+?Pg>9dM*Kv`18PpzQl18)=e)Ur3b
z7#R_Yw=U`ufY3WVJv9M(1YlqIB6f!Ke7hhV!FxX}nSmu9;c5B4Bno(&=(KYrA~SF$
z564yU;5aVsk9kK}77JSXKSzb{^R0H2)rU2i0ib^+I@l})qJmc5urYvO*2#7iOQhY;
z=V4Gc=Pt$h_I5XIDelL`4K}oAj9#SZjNYLJWG8>WV6N{bem00woe|N~sgoe=4lt@#
z{u-SJieNQjMSC1R!;1b)(qoBch1`4VxEazrEh!&CQwlV*qp+=BUHPMSKVxk{7%x`k
z6TcLUz)m?AdB<@>pRSs*VDTz86|OfR_pA>RYz%3!z)kIR1lqLJ_M-z^=Irep7AGA_
zXCQx4p<1rmLy@=8z%-yKS6CKqir`yMe8N+7`lE5MkYY=gGPcaJfR)~MNpEyEnJKLi
z>ud+u?$pYN=MjdAQxgx&7?fYT{)`#fF)RV2Hd4V=KAf0eD?LDS`7+f8BLv8;6i#eQ
zw<{mvCbr&_h*U&eM?lG~CrwA1r1G))aR~bjdX0SqviRp`BL!}VNf`de5=(eNQwT`q
zhIF0$&~#mGw`wiyH`Z4%7-qu_GLwU2Hq9>{paG@20eM`U@};WVyil$<w3byG#iUNW
z<$-jw>{!tuYM@c`nUQgE>nV0^aO$1&qtC`ee7a#e>J{6VBy2GHMu@KaWOHRxml!p;
zE(zL=ytRYEO)czi)_@FJ;>F{}BuvL8%U(}>+^Me_%GjKUkN102A6+70`uK6&gq%Sy
zPyxzGJOB`^+mJ}7!xdC`nqcFIr=w;ONGIP%k;vm3PT@Y|d&(S>0EqQa4H;TgQQBRt
z1B+Z*+yt53Sn!x$cn&TAnZ6!?r$r)DMJs>4`~|b*d*&rr$V<p>SCw?g1AZdtb{ZFl
z%_QRDDvV9d$zFv<9N43+R}q<=aJ<0cQRvq6Y>g57Nj0o)^pEM>={?FLD{Wf!R|_)5
ztd0|kul@rbYGAM>3Z1u}!9swXgv5kEC5*^a#?}+RI799q)hq(IB`%O7Pv^R?zQA$G
zpD&k0hs++Df%%M2pEM^E(-Ys)*>i-2VB?^tHt>@K5lWt+_fChi$l$&+TXpJ1hA#=%
zxagg4NtpL!(+*z77R?40Qehrg$dSzC`T!nNP!lKcZUltkGt(IYw8M{1l~0}pk05B9
zg{JfC!<=NMi9^JUhq(%mpz2x&uRA_&O6Q~skMR2k{*c3;em43u9kO8riQ9*?NY+IY
zDiI4fO%Ckqz2TDiCsltfo8{fd%$mVE>;pGCawuW3xPv!Gck?KeV6B|&oKj3eYOHXe
z2<-Qy8Co7!0JxFzRKx4D_G;-7qt<)V<4qn5i9cXUS2zh%T$;~i0g>;wM*-l+{l=EF
zY1oZ_#NLkhBW>>CkO#%(>$02{_D&%kFtb34cV`<cQ2yp_wg`>VJpXiswM=D^bzqtu
zPOUk-4>BepZ&8mrlmb9=+KivT+VI*xV*c|BGgruw#9*c8rZjXhr3=I@2p*i>{1?eT
zA06*g?eeI}E2$iR{bm;au<IfNC9F@sKoD}O-W1SWJVyIhqS{*8<&|)rp}8xpt*m^{
z#(E==@&e=q6XgI4vTpCWTP!WZdT#?Dle06+o^CN5)<^H=#(A=cek};b&5kj0d*t0O
zJ-*2jYaOY9pv%`Jl@&M_e7;OFqG#Vlw^Yxs_;oW`%CQm*YLL_I;K-+k0OWYIbnPO!
z>+7?5Ri5^aFpEy!oF6|X155+jKj?nonEbR($!xcvxF86C;s>M1%*f34-s2pT217`L
zfVr-BTC~G_BV9~Aw1zr<_PFA~5Q2<?qH|z;{Z-FCm+4t7ZFFy?p!+mgqBnG0*p4C7
zDh5Qm(e*=VDu}T=E3Pe-w=qp=S|2z2vd@LgYEPs|XnytYWu{Rz44t<VH)`ZGX6=@0
z)XvUU?zZ^6+*td1$H}lKJ#)2XqgLtcZ2qr0?Xyc}TnK}%&GL^Mp>q>U8)vrMw`w-S
z@9%Ax`PT*e9|0<<&u-kFGM8%6nvqUXr)*14?&DXx3f<xyZ}S&$mDSUxTR8H4vm4hH
zi#&ZH&G3>eVT(13M^;B}z8q`kS3D!Sjrt2uhLgxO>=o5d7hKO{%d~PwoFuy{pWTPx
zGf&5pKdzjt8F}v;%wqfCp}LjcUuJ+x$1QfO`OYg)mWxj;WtmghJX-(+x&H^q@9sQM
zbf{t2uYPok;PsZR7H?Kax6&uLP_^!{7!TuCu$Ge5(C~tg3G+RW+lwsphcZo!glRgR
z?7BnFGjUU@ScZDlFXf)xzreOD!7^Cttw3$-zS0J>*7x;Ts1Hr6(hu;qlu`{|$0UE<
zkbrZtpxUdwm$>LKY58v3)41c8>#Tbd)T0Wr#3(uoul}7yjG}hGk&U79?#BEEAdgIe
z%8Yym!xIaD%rOc%PCgeqENf6ue{}{9fmh1nFp>1)_dT6VJ-NPSRX63hb%_pm>J&%`
zxIB|YyI*({gR$W#=7u8sE`^zO^lQHMgHA%IUUke(hoOsPlSO+j`cnnGw8VuFA|~&R
zX(STzkOzu-p*aaSGX*Vh#L9CZFr02n1@@?BP6M2J8zIKIR>XNTt9_McjMusq-2xU~
z%}6J&5id<=<o!OpTZv}Tus<i1H5vQf;i1Xn;ZTq<C;*zTQ-oh*?psXq?F}I1J{*;0
zs0>;g!=F3*7VG=P7rQ=TY1aJK)u`#R1d}M4J6GFdG75ksQ~drjMg*AAuAG_7wGs)y
zsC-xlT&D&nRk6mWTSV5<0SmS(Org1~JTysTX@J!=Q;^V`XG|@*!Z%Y7f6R%M*Bt&#
zY7wOAF8q7AK+yTD$=F9w?RIv{>x65#L8&Sj^?!DNR=}t|?0h@7B=Gl(1N5I9lW!FD
z;~?q4hsI3%M&NeXX9BwcGYzJUbtaR#?sWOnX&1t&hbC>ee_EbHw+E*I)1{0e$k5XY
z*L9D2|7pDTo1YLZqWct$y)wux4{)3;Hw{{1mqTdqg+toHv<Sbw+?x8%w6}SkZQm|r
z<o~`phk%}^`8WM15kgQ^49XcQm{o=5Vorvl$8I68y(VYx;4eIBE5*{~hm5-yr{Nmy
z5C4Yf1pb1bTPyH#8Xx>CwlR3%o+F+zgB6U}KE;mls$P_M`S-%ki!wxFMByxK<VQks
z6QY6=UXMmuzLes2;^S5RT?lmJL)!T?|Au&n;0rGuEi+9b{98*G5*tDR<SEf=-<HV7
zFe}X5<<08kR0DoZbG%`JQA?NV&-WVgr4unBm?nARCr>W=IZI(0g~z}I(6`veC|~9N
zUV!MN2nDs6<T^;(nDCxIka>{1o0adl#1bxuHR9K{&)#~4x6w^|d*#4$vpXAy2p{f9
zYdHLO<h-i4$wi{)_0<>sz<wFX|LpIuGkHF1e8_FD>DtTL?w-xX^S^cGv_3Aj9PC@Y
z|GjtD%cI9_{QlnrtQ+RupGLd-4x)ci_zpA8v+&ZRh3lIg{TqIhGup0(``+)84Xed|
zGp4y&=#g{E$8ufAUZx!}6@CMu+@7>fFINC$yHDM{=3%sU2OZ$%GGnW*>p>{D40KCN
zzKNW>E8iolv~At96_2cvdxko2I=U-oUE-Dv?5UEI)a4VV<^_v;FTpw3*Z%n4s(HAb
zHI?BqX`%j`8$u=B_`OuK;b5d?^YL8q&{VlAyx9s})IZ~?_EAMW`mIfb^cWwOk%^-u
z+76Scfr^|NS7S}UPro=p!p#Qgp6U?gq$F+)+AN0)!x5*BziRw2JL1erFu&px-3q#;
zzvvN$^yLNw%t-8h*?8KI&7{_Wjc`<lyr54x^0N$x@krdrDv)`<Ut@)9e#|Q}Ze2A^
zZdi)Ji3ODFcfX@~3XROSX)CF#T<&FN0`%<1sfHIbNH4FRWPyhh!-KtVKz!+EIO(|y
zksO?M6yBQiik64P0eU&FitQb5cYF>UoQcv7xPD3Nshaq%QACgDmzNPZ7Ocp(c@-9{
z`Q!4<&_O3Cd7ytEBvmglYQlzYbi4v8Mm5%4J6ignD*Z~9VkxHazcakb-eF2gr@vD>
z<6UEj7&zLI1?{(?!G>kG=Ku)j>6-sD;9K1I1lJV5ICDWCum97f8)g(2d<DwYkt7uf
z=O?Qq@ZB+agn~G_&&OMou4~?UfR>0U8_Bo{3+)@Q#fe33x4|g4rw1ZQJ6Xn}2LsF!
zZr7DX`6f0GENC+_1t>Zhb=@G;@!bMv%!(F1h1MaP$X_k!LlLHWwc431YnMj?m5H}G
zAp>{IU`!;kD)&kRh7Jqk;--=*j90$G8i>p{zt`4-St0gsDUyVy@v=ozpy*r8e+1aB
zg--?hPM5;MQ29P}#l`yN>gg?Uli&;BxjR!z?_W{Fa|K!#0<7hk9a4K7&s{-9poLoh
z4P;`IQU^fd$G%ulbQ(eiwm@MG&sndH!_9gbz7D)4O8;#ioHFzZ#ZmhwALs=<*B+Xr
z4r&50HYx-3j!i5p+HSulF_uWuc;2&{HQf8mSwvB8mc@~aE-XUVeOfJypG~`<?M1O9
zybmr-K*`R#SdF2@R4$hljnoJOneCt@5qZmaK^4e)*4{~P=tQhy5@77fCyLzo0^Eqm
zxog(7g}}@m`YQ+<pL{o81uW@Jh9Dd-s%;BDaWw$vk5MWgl3+ngoYRlsx45P6%C!<Y
zKg~hrW>C4E)bzQc5DjnvZ|n;i+scV@z`lc0okei*GCJfdk;TXHEGb#Gu#mTfDkwEM
z+_F%@cvTV*zt1*jfN;?AIWS5FZD;`6w8zY*Ih^nF!ZIeUd~jKVkKwZ~2w1gc$t!4?
zct>y7XrA$^d8X_f@ZgOO({%XF8X2#iR(_5^)~W$v2uEtpqv_C}0Wh|2_&Bh{dsTQ~
zzHe_MNozal(U?m&<Jy|bgXNq>e57;&(tL|Fo8==W@&Lm8>HCDXO6<xE29knMg0Bvl
z=?>_gk<_D_!?UXMvL4fLNZhhn3M~;E(ECfs_0#FR6yZF>V?G+Z1Q{ZbeUH8oQpFa#
zk99CF;J7-*rb92os%9D50Ka>?=ea~8($jzj*kh7<8L0n4CjO9nEyYxSHYR%d)FTjR
zs8<9z4@0>C$X5K*LtNZ_F|Tv{?{-|zsMJO+WAIur0g!><jBdas$hy#i9B-hrtFZt_
z@y5yG6K=E$yNMhPKPm-?qgG4l3Tp%ZF&;ulPz6wQEfNfN_0s*p8v42!DM?WO2yC0}
z*j_LZxS!@x4S%ELbjYPqzVGlos|EJQp){3iIDcPx+G+>yysg&DG78%mw-PGp*KRi!
zi@5ia0~I26Q<Z+${6w>S0sk-`EaU39X0M7>pVBBnxY(8=sD?{TR{oG^r@(``lAF)P
zB80-lftKem(mbGm+JD<&KFT-R1l&r5cNRNVT~d>co?G67j6^zZGlNDiQf+48$E?fa
zNxxEnMW^?*R1)p~*%F`^--Rw#zII=EoLJ1F0;eM@y8dDTmJYvEi$t0eZmvk>fpCM=
zC0=DQ_^~MxkQMvi1YTuj;@bfx^<st%3Hi45sCZ_^5IeS4{MDJ;a6o%3!NjomO7O)g
zV$=x?s2Iirvd*!*F_=vI=Rk_9T|DM2qD0KSOUU`IQ-y)>z8HHAnv4Cxo3DFJl6}h^
z@@YCt9#@Mx5mV|aJkF=j@DhZazPI&@i_#(*#EKjx(N7t&w|;u5&<(s#uhNMxatD(R
zs?e|8eNtlWvaRxNqdx$;l5l|wdwcV`=HzbIhdbXJvcLMaqV?03x37hV@JjvsGuX}H
z3-go3bvl!QFY_md{BaxZ`cnG)Pt@$$B{(m?@baUgKcV4D8nen-YvtELHf+fuEFh#u
z>2ItLNHX-*EraHIbGGJ3Rqxf_&1WF^Iu|^#p1o@}zwy_9|Cz_TJ9RUYhpuI7_U?V1
zHT6FiswVfY0(jEzW4^cX%h+0ei!oVX$TrSn8NQtwUmgVXUDdpRL5*at9S9h5HfO){
z)k!kIhrFOP*zHhzn@hqUg}GC^>2`>}1;jAP9SdJ8oaTz1O>2s7*US@&%U3UZ*F4<k
zev0o<Pv^UOIQ7h#k!wg3FWS(t&VF&HKs3CLfJ}HLyzNblr0Nx{15x@<cRk{Kb^EE_
zX8*7WC%rBB20Oh?P#%G8!CqNZM%p%9jW9Z)L?eCY@8o^Ys@^mfr(wW2cWptL1W7L8
zD)sSkM`z@4tp=G%;^?Q_MXjRn<BR^@cK?7RD?_FH*$G<WO7j6miBrwW{R6;@i06RS
zCuh*2M71mY0a{L1GHAWbd061-EE3XYrxQ~^9IWiys1sx|>?BG*`ohy_yf3masRh^i
zP@6N-#9G&<xS#<>sq_Qyujlj#zXv~L`TY+m!{mA3e*MdrQpW>qCkNV^suE3*5loei
zQX3LxUZ@a3Fi(8Fvrmp{ShRb!+8scDnlyttKaqgF7K@=BRF{R)Ze3+#$*e%DLgH09
z1>XYFmQ(e@oa{cl6af*+rwh2C!ywambfBB5_3h0xY%KOboFUzil0{o^k)^o!8Q78k
zVxc*4S@~qa<9MfghqDv#*UK2PwhqP4lLG~tr#!wl7q8|z+Vk4k&h&s`0A9Sg{#ugL
z4Ty+(1$;|sGYeuAv{<PBZxld{O!7HNHGHWDEJ|k8YS)&%o<v^`ZZ^Txa$dQv?CdpO
z!Nh^uh__Tg2i+(-7ZFb-2~)1Q%Gj4}_M;>f{WuY)jg_c#fsbiAHqQWiOTY=zip6F?
ztHN4MRh<MSk!=v)BX1DGCiuEomg`FVun@=#Oh?9In5VSWC^}!S<~=oVfds#Stc8b!
z!nJ>?+Jd9_3;c22E3A&zTG*EduBOsT2Mq^Y`|p->HXc9`j-7As|E#AGJh*qMsIvz)
zGC-qEo5DDRH>^<TNpnY4Y=377fY2v_9G`Km4X(#*0G{%*@dlL4k;MVoo;E8Sja1|D
zEbIA@M|u4@SM9P-O^Bm&ib3Q{RczDC&AYmJhWin6J@O_W-l);ft>@DFTV6>3<KYyD
z2zuuLrH1FdB{10|0%CwO{2}yht(aL0<(U|T#x|DK$!pB<`GBRM2jXr3iR9~?W)q>;
z(o6{W(n><1ZTkWanmXV>z;&*g7cvQzzDkUWHwAg%v~+HKaJ!Vw(u-UbZ1k4^%4i8H
z#zG|qH#@~Tzrs40-(x`t362Ayx8V}Chql&|&%{m_bea+x0wDKUwtUC0o$a?AhYg4g
z#V2Q8KRABs`U~0u$7CoceJOlrA?epv=ES8(MSFec$D+{Mb({X#hP_${!p$D1d!GV^
z(3w>$mqWH6746Q`P8NQU(Wv{GUv~o3pxv2?-O;yNAN|zc`=47|+Rpr4cYnDwJ2OjD
zdsZ!Lt>AaYwTXX5vUd2YrrE~{>O150jlUytQkSzeuP=j)Co`=#JnK;SpE!ND`@i);
zit8YLgXnhVh{pIbDM&Kx8{47bxj&C4{_xwvYJ>VxWHE{qknRKdAU6!*$Gwfb>QB`v
zQdaT5hZgn4qR$(~EnoQ%;8pYRx1iY;le(F_RqlNr+yBnN>j6;FO`^U3GFgs&8VRz~
zbZ`wj(Atteeg^NJ)}Gn5#SC^mW`rsDr5|U@ct$<t@QCgne;*<q_WY^*Yv652@}uu(
z^g}xaifU&x=6IJi4pI^y<|>NURc37-Dzdh#3wJ;JT+!mch{Ajf$VW5&quo6NjH47R
z3%%_`SFY>{9C>-xT${+2B+H<V$Ce@|Z!~tXz>l3Rs`!E3*?iF=ZrT0NJMR||O72B6
zNZp)Xy8?9fHTRrlk3lpDsk}@NS*v4#RU^)^qC*Rx&+;(_dJh6PHmKuOhAJ*yaUJpJ
zhXpPeIzeEjiqSxhJ*WD9N#RY1NX-N3Qs0EnPLj*D`4bC3hMyS#^1I&@upe8?3&=#w
z`^&&bT59kt{>=)B$8D4}o*pI_-zR=pb`KgnLesH`1}T=Wr)~lx9(y{5St7J42&?|h
zXF-2XaFk}0L_|jj$^7|m+_QUHg)YAztYtV5OihrdTytyTwot?3^|Vy6zc1RKB?Irz
z_Y4d6WqI#%t~i?Y=h_6ghA_Z_boeXXfa?j{>}Q7y7Sq6eJpluwiyrw3WJq=OcRJ+U
zR?lE5RqUj;A|O?Vi`RoWA>CUIe6Ya44Hvu>n`B>%wpo1U=Dif=+rirj$jP%%l@BHH
zYkyK47nG$QBt$>@<rydB9`{WU&MD4%LhF1{`+{nC*rydiI16&N_S-C!)p~cJcMs#&
zB>9Egy(7dk-xmeIpP!a^{;lJ{zQkN50`*+wVFM6u12mT-dRzh;=?73~Q-eoS<F*f6
z_#uP8#h1d4GVYeMjY$Jbu5^e8e$1P*oqXjY*}=f@3rPaDKA4?=;t=_04V)5Q5>8B`
zTKykKkin<LU}_n~{uw5ez6!S+kbI}0vq;NY!wrc2!)1z&KO{^^%m{8;kkxOUIzpi6
z#M}VM*!$0>H!o{pIf=vTng(#63zX}N2OjLSur>n?OfErzcuBO2zIrE<+6V+?uiHZO
z8yC-CBt`{FVv^i&`jWFxXfAf~ABdRYS=)ei?By%w<d<X``Q+*@3D^233kz%kef3C;
zF(HT8!%uiWYs3#d4<OH2e31vx^`y!2KDH>)BV~TjbX5z?Fo`f?P1g+XWiA3=GWkSe
zt0_?^MgffJErg8<u&Lv^_<K5daT?Bw_zQNiT^GsnA2>l0>0>EqTKG{Qf86p>H6w6l
z`$RsG`@VkvIsmYpe84~beY2YbJ6j2y_{Rr%KtsyIEX;}mN2;4Q;+L#aaC;sVulp(C
z`Ei2**@L#fwCH`D5QtFqr?>GA0g3roN-MDkjA<0B1~m7Uje|k;wz&zIkU)OTu}<Ea
z-~op0XNv}443Cw9MtmyzZVlQ!Nka@aICd)rLsJHgiXx<XLg~NtQ4tzq`XI3A#kCQ1
zS)6q#?N!7rz?Vu)fgSn10|~_ea$s~!8t;TLz6q<Ihz;Or5gc@RXmSRj%+tc$lB=va
zoU4(qL+)yxxB*!rFNqM<SW(ZIdcj(*+2YqPm<3#F+kO}2dySa$Tuir0bvZLj!<^h!
zk_Glc1bZ6>B%~ZD$e?wK2J3&n3K3q?iT77m)^ZOKmXz-s<QY;(SFtqR@7j;Ts<g20
zNk}IPCs@_7%#;}Q!3N0LCJB);_=K7M)G=KmYu6DSa@zJRADp^9FrzdgaiRdwsC_9>
zzAb<6fb`gl3$JPriesrN*rwkRu8;%*a|1F#j|sgIel4nW!l#(0#e~X0Kxxh(1O6BI
z#`BN!T5Uhw?LXJp#f%g?o=-wno$k#r!L8Y{{8R7w2%OTF69d{n2eecM|7fj3pgk8}
z=V20RZuE<G*y^j%>sb{%&>#@Nv;q?CT}>r5O5*&vcN0M=w9?D`5<^Vsv!gOCvHL8I
zU%hUHWeh7WS@Gxud;x;)MKs!2nKT|bv;?T%zD8jCxcfJ!sa=MgatAMgr7v_CKRXu5
z#W88}P8>LiUJqsh5@J2C;<<)vUB}?71F{rofTyK4oeTh(fTvB5e8GRsT}+h|x5{d8
ztPV6uw8!E=PHIihQ$0uP{rNSBO~t{yP6z#<Cmp;IN2@LM<Ko4DV$cY5n~&)<-ZL7m
zX!PilW^;nf%|YOJaclzKEz)LHbP0q{<MRNk5Ia$!IBMr5L?`mKqg8lDZ-oHW+J;Sn
z-tUn9OiZ~@voV-7H6<ZNw0j?7xY>5s{{312bMLUI<KTu!FkziB4FxaWtjUrJ-&@-I
z_{CDfZDMbiwwf+1uXZ1%##M;<1=jq-_5WYb?xiKYp4tA@DrMNanb*C3b7PH5v-P;z
z?0%m*GH|bdChLFxyA_tc*Zt26;F7w38h8Pkd+C`Q2ht~WZ}wQ-{+H<b82-s;&hU2a
zGb?)xJAI+(<NZ;q+vG{G9n5RwwcN^~0Y9j==CH%L?2+%(PttP?<H7xM#z%)L2d|sZ
zMfJ-~oey4)^HrL<HP(GiqKNhTK*-k;rzdH~-|FofT|6Wwzso#GHTINa-q|=rzb@bE
z<aqE(Tgm<1Sn3|iY}M+~_2J@^uiv`voE!N(f4AmtqfF_;Vb8eNIYesk+DQEi!GPCa
z?rNAIodQrBJ<OGjw92nh5A2Rms_%SfwSc-MQawQky1QzNobHllBQ=qc4`(FLzyG%f
z4+)oAvSkK6v33EVfIZ{0(;2Kyxl<r2$)o86*z0g@WlfP0tR*McYsH3MxT6X>mgQv@
zx_jFuNIM<pZa&3_6HB;UkUYn(y!|ec<Q-PZfo-_-N{Ry>E}j*I852+b^k`6Y!VKJS
zaiSzm1or4q;#9R@WIPw-8nzR%^8o}@bFQ95xnUK-rY06B;?nBXR=I0a7H@aH0dJXb
znsi->h^#&G4!G4gmMsKC6mDUMjnC_;1hLTL(Vnd#oRE~6C~mgo)c$WE7j_{1;9n3B
zc9s%gMN>r4Kn?Lt9tBrjCdJQ=T}VwLk{sU&!6@3|%H2b_pyeNWcmJY_?Dq@sGM-6)
zDmYVg{bZ5`UMw=73$ipSnT4MNYgk1D7dIZ$xdv8a!y6F9En#bd5f5fwfXG(^AXvXG
zg(tU{+aF{UuiW*C?9qPSq5|Tq+@HEo1d-iSGfKGDDXn9m+h<s@5L(Fpd<C4r7qAy@
z_~0>3i2G;h4XEwW(+(JQuozDy9hjhGDyUPQMjT`5b-pLcpgNG#vO>`CdtBW6%*j>|
zc`B(+n+7EA#_|vd2J3aGQP}7=!FCRbP%ClteCKf#dgPZ2C)CEDI7|Y|tL8}gcU5|X
z$`PtzqLazkQ2x+6Npy~7<`mra_!d8OMaLnYDgN;!$iyNpu@EMKTKKDET69Dn_a-up
z<1Po^p`sim8~p5nde3s8Guh6-8Hw@0y<G9o(B~R|1^RIfha8R!mnjoLYl&=O$2PQC
z0U_v=88AM$)1q*3_Nj6>t`CWTmAq!FIf>_9p#To2GP||34~jncm>t^`9U^0k^M^Vf
zbv=^Vikcd@rE(c$)S+3DDe6jOSSD>XHKW1^H11n*21I9-8YJ{CvgZSC+<Z(Hk3_5_
zX6IreF^Cr^jw<DWBoIg>jIx2?kRuSbQo6Z8Wl6gSp29q{Dda6(_u5DrisP+d{@_5K
zVOe6JJf7RH%LM++W@8q>9q&`Xa7{UXuvz|wL$xwi=_OYjhK8Jclwp#}kUe>&<eU{N
z0Zsp1Z={N~lztR%f@PG0h@FBV0zE(eD;9I)dh<i3-fry3rL%ytR;7f!YnE^0RzI%P
zHvAIJh88MR5hom#H=Rjw7BOEbVnxGG%*bMg3v3D>Z0kQ8;-$HajDN^8Wc1quEvP=s
z^T4ZVxZe030BLebg0Moq(!%mZ6&?@Wo>bgjFH1J;4UjLPYqw8I0W~i86M)Env@CIP
zzqKAjAU`;Q{@)klC`xB1Y#PE(uKFbrk;kcbn&jg{iQJIEY|tAZGrm(;0yyhPdcFsN
zA2LzL6}tUwC|F@t9|9_l3}OL+K9QXb+{V4*q@73j;{v9_d942eLpk%RecwdDJ`_i{
zvt(xx^`b(+wkPM1kf#b305Z{l=hPV(@hXFlpnCqc8pOR%Z(b7J$SSY$@NRGLn|%kI
zq4+j^B38fkJ^BDFu&#hYkJ`%jJ<B+b8j1+F1g`!EJrYT)hYSRz97)R#Uer1u4QOB(
zZkqO6AEx{Ep)wCN5F?NpW&Z(Ck@y}7eUU^<S5`L3C%a!fKoh8+1VOk%!_mxSsV6Vy
zg}Bdm^Ry&OQo$af^faS+cweJqK8Y0UrHVzkgQBF=W{^6@itRu7nV*m&ZTkvr(y>ZY
z#twdw%cbeuw^O5xK0w8I|Be2c&Ivi({cP&g!I*x6#lGe+;i8LD4Jqp<S+PsHu^A?O
z6rJve)2DKAanb21Sh~UU+gO@_LgxvT{_vOr#65n+UDVNJIQNXDKV+~&c|y}+mj)>I
zXAdiYuS&-l(miUeOUWFQ@||?^Sr}&w>gZfH7fYt>MH^|TJz9n~L0SuU2#(~uXY%Q^
zN13Avz0I1=HU=)Gf58->BBX5L&?~Y&4~o!l-tP(-s00}+|H^a}E;5X^uqcwoxn=$V
zIqpAMAeR06rt?h}QLUHkqK>(DWquz0ZXKc0+TX<#oIVJ$k|sa9FOu6HU2$ZIjov$q
zuViIVUBz-eo*f4_lF<sMW)!`l|D<*j0QZ=BRD-GIedWnA<MjZDg&q%Xf~{#hzpNvu
zHyaC49LDKc`#t<?p1L={29(ahlQBK=JwXbQNMCx|08=m^RD!1CPkF7zgL?iPRJLRr
zG<yE(LMqk_*9u|Wn_$(L31aR(jezxndg~&?;Bf<dTrN;3lPB|hc4rp3Q^a*fAAgT}
zA+DoT+?8?uZ#~<bR@|wxU&40cvNx}I^n7=Ik?HSV4l-*jkgEUg!udZBerM?Pmj0}B
z*aN|};8Oqb?QajjchYNq)=aE?>3uyTxwQi}rH*gZF2P@8wQjVC-S5)`yGMKc;bjXY
zu=`ymOFU>Sz>4n4!V{eq<vZ9fe*E*$$ZuWmP%OBtpI=IiTUi#l)RmEVz1#ql{#V}l
z67L!bjal_!!#QPAEP7#4O`FIyY61djQR3x|cT@D${Vp^%KT@g5RCmz3=-UzKj&3mt
zeFq4q(8z(-BYi!tC=g+L>DM=}Oe#()Sw1nBLep^28~C8~aT57a`$KD2*6#2A=HK)A
zMMnAW)x7Vf#&ey-Ckzur=v~$k!B?%d8Hxz{zTqhjPtbDi5$s<o05qA{4>H<hcoO_A
ze<5^?11nc_8iCGZq!d-*#G;=W#`zxYwdg$yq{pK_fB5o4xsP9NXClzXVwN9Ce=I!y
z{$t#1rQ+`!PaiRV!Gf0Y_)>7*js<O3TF!yJhWjzlrol?Tq9{2^h+@%ZhjXylM5fC9
z3<B}J8zr%W6HiI}wf=0D1AhGc+*$=lC>abje+41)>HU5?`yk@Ua1O{slWWo13*^gt
z<VSN@2xVvh9pfJ~(SbhTx8ECl%A_{iPf|4?wh^Sg=eo>xFXQ7R#IB&wpNW;W=fRE$
zL8WjvoLm}MvivCYLSf^~x97j}N+pUy6hnqrSdThKm(9&hb3BSbD#m7kj2M#rhptov
zCXcrd=Lk9l5o(TMbAm98$g8>wCKt)#hVOu@GiceG#LG}Q7g5TPb+NW*L7(KX0`)vT
zPf9IlZ*3CzM8(3O!;6{w0H!6Qe?X9aV{{<pqX&pHV~;Q0LO_DmH@}=V9ZFE1*xGSV
zmwGgC_V%C8G9NFcFxzXut_!deX$JwK<kf6eM&Bf00W<g#iL}oO+*;uLbAMprn9S*E
z1o1=e=iI<rQ!LCg@q==$4)a0C!*{jlx$!}~P<;Bs_j_J&{uyvnpNvFhp>}!Hzr7)9
zA67sbKzH5F6>?|!BjpDo$vz?p7Ys$pkdrUomaf_J5_109M)Z{K&1=y?{68}DB>n<n
zGU+-XXjf#ecQYe7%7oy@(Y`jCAk$6N5Txry;Q7Ipq#@@$044aK3-0EjNx;UinvKOo
zzNq1INmZst&w2of7-g^h3;4Mt3@ePXtFR&UsJagF)E`vA#`;QkofqEH9dlE_&z8d_
zbTQ@Td6V<l?pc6HFH&6Kk03ejhqQo&YX8Qe2uGjiUL@xpa;QZEQ_+D^0ElYmZLrPl
zM_>uR1TE2)e5FYq?$#|g0$Gigchgnva76k;qR$l2puf9MZ<xnnlI7d31hlJ5o|BqI
zP)cXe;%Hkos|RCXLlc{W7dbe3hkE2aqOTz>S&c`?`S}=c!JSw4u5n>O4*KTPl`vqy
zuz;M`XSN(*qErbaGGy~q`Te2sm_v?`;1OkUv_M^fINHxiMTZ>CYUTpD!=UQb7k&|?
z=`=2Z(D&5Z36M~HXsL|^q#3rd2GKO$Ua;ruvnj9lDIwZnO}5<3>*l8#bHjnjl}kt(
za5|&<q;Su+N8Vv03i)puOR#B1B9&tXQRpA1&N)H2#`djR!2Cz4tY}nVUMH_WM1CvC
z*4P_a=(m<dq#;17cIjv8o#-_VNEq(r$TR%iGj2{`)d&CrYkA&_x0p|X=V$8~uPS;A
z=lz-`(fJR%M8e%JzA%Yf&dtEZD)Z7hy^4U+e;5281`qxZN#`9%)&Ku-BV}j1#x0Se
zxK@-K*`ufrLe`bNskn9unYod@b-5)oo2-(Zd9OXvvM!?RkovuRe}8`ZB=?^4KIi><
zKVQ%1<3$v1^*N`kb%DA$^uf=U-d<Q06Md-1tXRfK@;KDRIG>|jBHV`-$KzNsQ^)#&
z%c!sr%YBAU!rtX4?u*i_1WC;kw?159|NA0JKgwp?3y;va$|W=UDy6Jy!}sDCDOZ?M
zr-e;6oVZWAZu~l0zl3?#X+YAJaxBC{g`Ci=4Hkcq#$v-%%czlK=&#>jKLa<Ny*xDb
zYo=Md@_aO{ckXy=s)Q*q6VJT44D~SYFPC|mg2r2jhGS?~h?-8_r6uFO)Pk6=^Y>^s
zF3=bd`Mg6sTJGbm&;9AX=3<y$B05}q*$YwW_e;m}UsKP-1~W&;XVT2(Tp)sL<>1Oq
z`pyrQp0NE2j-*i_TkjV^8Nl)w9!5k~8(DBr5&SBGLQg&j;h5hLsNDt>r082u2{>|4
zhh5UtG7mlz{yoo1fp6_eh{xbO4Ns|ZzQe7+7mIKPN6CD}6r)Oi*-J>6h-yWCQhAo!
ztISo(bducUt{1}WRtwF*iRV3iJTyx-?jxk!u7DJRkjuN5%WO#SeLwFSGE5~)85^a3
z^(pVaGKxl5Rvr6}qcRT;x#@3DPFTNfDBiL|ixpY1*|_i2%?>vHv}{Mp&XHkL<weRb
zVB(UP#5vk^4(IGtE6w19{ud0Mviu}H%NeIwtLQ8<J8&{cM0Rod(a+r>FJ|S`neltS
zPyZ8=UG)0<XK>i-?=Q8JuXer3s?XCmn==y?uP**uFZ>TI-x-4s&ne#X8Mu7ymvX?s
zi{E4|@VLw`ES#5FWVm)Pd1N4e<GUXIogrimnt15_{x5&}*<~IDdS&H&dU!c~{a4$`
z$iMYmQLY0buNATWjy`Qan|CcuT&_l+99Mip5!VxNcvExSW{x2Ti#QR3w~u^a>|L{}
z6-n!u<PWCsF+JZOlDGPd#*!f#C5rWbKV|)?%Kqx3*d4;~=UF++ZO?khTotUblx`-j
zG<6n}>SaLWreEPqId>ID@V2irJ%sHh^W;McN9?n9kvkR-!udkg4&F%p8+??1?wi)k
z_2ma$5mGWtxhuOL(P-@7UNC@f&*9_<{rt)7d-+A_LN?!w-a`WRdss4rp3qG^#E?_5
z-m^xXj`!S}=)*i^%atTtywV(E!JfU5;)#L2c}1C=r9Hat&L;TicIzw$c~6WyyTiIv
zU+V#KbwS7`NZ>Aut28922e5%6%8Mc9vuiZqvJ(hvdJ__S!jBOlNMYN5l>#|iYxv$+
z20b`iNu5oJ)h=wAn^XB25%+Hl3XB);O(WN%NYT3-uoZ8dfxYwGX=>zfct9ta4U;=~
zr$%1;zd9bsQPL)@4h5r@YbB}Cmwp<SO2D_^QgVTdmRV1h9Qe&X5X_vWHl5!}eom`$
z_)cfuLc@b`zJZrq?Lm`sVnG!jgZdQpE7aPh?;u6;79F;PGkek78Top6DF5|!azZ`J
z7vmQSthC2)%u=S{EMCa-`vXeA%N2@aSmoI<UUJhA12>t)+lc~Oa7cm|`7PK)o|k9r
z<UoK}aR#TY(oZ(It7h`-LT%}BbnP40W9SjXrhvQP|2B$%9)eE5C43dMF~0S}nVOP6
z%*bB#MZt@+eDul6T}fjm10p0j%N+>uGUSA-<!R>i1j435MFB!<M#fmK&WL(l$rmal
zl}FDX@axM-f+d-z)hL$P-QvRCCu*tRy+4xET4z|vYzuwxj^{_I;$?#CC@@=%BxNlo
zOz7&fCjRW=fs4+^H94UJkn!67>@8<#_NDW&qi`)O&Xzq9B2~D7=Hq;aU@s4+#c292
z)R}&J-qKb9@zWdwQM`<0Vy;eS?*lO9pIBLL27&w9Q+v0NSnJ;71U0_bDT}sbw)>X~
z$2q|)abtcK-0gqllZp2EmwAj*tRV2yS-71>iPRs4mxmtR{*QbP(UP*$uJUO0pn7~D
zWhS9F7|%Ur&L*hAWA85p@!f3`_vI|QF3T|}zrAzj$qL~)o*VEVJ5_w(M)B%j_AhNv
zS3C=$DpoBKZU`WABNw4vX;AgF-k!{6O$lYu<ZLNqsE*gSrOlIR6YrfQ3sQd}#tdIM
z4%r*tH*~~!uU>v+#_3C(Ln4H#;G0#)%|)@%^oqpi_L^5(t6No+8Y!}UnJOQR7to|-
zRi$zV_uXjL!F{^M0X7lZ_!TfnT~RZ^F-J!a8m4}|=2FyKAn$f~3RLhd-{N4DXABAT
z)Wu(=JkFb(urzA~&Mju_Z4y*TPkk6MmDyj4`H@5Qg2&L6&c)y^wz}q%2b(FqkF-|Z
z6Lf<au63bv%;uqZyBGi>Y;7H);dsvq<cEFF{08;q91e(KIEE=N&YR*OglnBog=8%P
z*U?BV2$+m$?Oxkb$iCK{fW+LCIY^x6CPeYKYazx>d^iWOb2dm8PvqF!52P=%EwPJ0
zw)IP=RJ$-iLf2ykcK?>T1ZZ3up=dgdy_{YYhNJYaqD$9{I8_zJ2SobYMG7hkiW<}#
zP*8SwI=V!l=mdTtkQS1$-Gd!orLoNV)<yE@v)IBYeU{+cJV;7E;}2JcJi3f9$GIK;
z^X8#My;3f*++8LMK%-pZ_TPXaou3CxE99m{5rV7Ja!E3`-O~!0mbC2iy^ONp;s~$;
zRRF&Ni9af-=mUaF%kn+A^~nJj2x_Q^<HX!INPr%^T%;$0eP`;CLnk?}{sTY8&_)up
z|Lu5H@94M9#okzT1|jgC_Eg=(t2HFXY)~0CoBum194rlsP$UbPlILT!T^A9l;(>Fs
z$-N8;taRW1r4{C?sbB2-1NTpqYdlL|kGPhr6Zt`9o<@1Xpa_-z=r+XTI36-e(PucO
zgQX-buK-0Fpe_Q9dr{)li#rWm1Wbk_&_%YN-M0K19&N->PQDT|F6Nj-`=gd-wqSz@
zamqe_P^m0@K;a+rn_$EU3ahXZHpj~BC{tLUsXKJ_1#noz@9N-f-$Z8=%2f7aa8&Jo
zjfC&usgll@z+S%?zCwl%Fr+?Wh>?D()5=2wRu^RH;KfoUB44^B);U<e*6Lv&iL#lt
zX56JrM?Ko){c%p$i|<`uHno_&=E4c(+Pt#j-)3)qnO)2?Iv=Z>?FykuHv(gEc!soj
z=3-C;gsxjo+G98T9kF3&$^ufOOteI~th5y|JmAzoe6En8?<wgc3JadNL*JRMkiLRr
zi&kYE6-NKl#j#Cfo&kYh=CkRCE5j<&Qr@xqG4;8<ctg!9iWt>)hYwdsxN#dSt5>t(
zYaZede{rlw<bViow|IY*)Yq5cr&G)2-S6s2J>5BYD%Mis{!T1pZTs{-aYwZ+mhZBa
z>=tV;?(yGWZH4lY>zMx(;xjh?{h2#I{Qp(K_#x>uM7;JsoPeUBO6%11pDVqW1=7=x
z4&m9G|Mq`(4Ik_uHM$2uV2}8p4&NMiEhA_5?!zrYq5R#w7HTcW<K?|La}Rm+=1^*W
zr_Q_CtFdny*U$l<?zb^%{q*a)>T~V$?z>&&40vm<BH!=0zSpd$9W7%5oDri*OTZ-6
z3{NkeP<Id$CZ?*?$^rJy-t&ZYmzQ~BgV5Vn);aCZ4K7%%<^1Jc9I`WhDA*;)xv3{f
z6#`_MJ%&Qu5ptH{Xq?uygs!r2WT$CYmMK_o06x;N$oq4~DD&AJ%1!r4KTi3`o(U)x
zZm?Ca@(8kV@olyyv^8+QlMK>-2$=-xI<}`9ZwpHKrc5oa-uvt3^~U9y!F*Ikp&66O
zyd&e|)0ZDQvqm^>6Jkt$&NE=$SeK-T_MfKvnP;Px*AQeD)@X}58uVP=OE<Nn)Bg;|
zu(gFp;<U8_f|lrfph$^6ek=jm*}D%(vC208{6H{8D6j9qF8QE=Ux*lzjWooo1wd#1
zG8G#8m~~Wa>e160IY4L7R2C54Ztu_DxQ3J)<L0t*$|(n9?dqFb`Bd1ba@A{}sgO1f
z#n1p}?q_(0WA^qe{Cp>Pzzb1viwL_*zz#Bwl1!A-+&K1v0yBpa{CpW2`T>g8SY2vM
zaFx$Gxwmzb1v91X6_=^bKDHGi#=Cj=f!yQ$0l|YbILN2W?wAv7aqTAS8Fxknre)61
z{()?QSf8!Slq>n4qNV6dQ*SvJ$}t{LBZKkOG!5hvS18eQp(f9G$%5NJYoOzm>Rmj_
zndA!*a|L|=1b|s~3^~8eN`CH;5AP)}@WSjXKJ^E;)`c~w(1$b2d2vD-hf^R0+EW1>
zsKy}^&ZYFJHdr>=J62ecU*#RY`_^%)si*MH-+F2Pl+e&&%nnNA30R=dp;UAPK@z{d
z2e$df2Tw>EjBu`=Bj?4Gzcm?Cj!dE*u?4Nb?G2=Y)leG<-Qd_Hq7M9AswQ5j!6Ye{
z5*hH&Q66n`x61dgM=XlM2ZE`mxi4dvG&CXS_qA3v7R8>T$=T-A&0~*cZSnu=P$3`l
z%^$}M2Wu{?Y?8)d7eD04wv%j+HHgd+ED-jw0}Wq$bAeI9+d@y4_I<^!d4eJfPG}ti
z@ZftZU9NpSM@E$EgXr9^o(whxM+fqdsAI)(o2Mm;aBpXe3=Wf_jq)}!SLfsw)J7IJ
z&wedUgHUr<o>*MHeAVCcR1Mm03I6|PKN<+|G}wvtHY_KU-z_pq#i?C8h8<|4AtNp)
z^AM<c16*=-j5+e)8U=47Fi97)3Q^o}@QytRMS#s)j8sE`QFP6}LjIX`t87mzyjRPu
z+n7AxO?J&LS_6>Sx9V7l_AfU}O$Dd~rL|IxOf*$FILLWHR%qPa9hzB3mh+7E&7gZr
z`x#>LG#hUCQ}Sl$b%+$GRKY@BFS?*fiKb9jj=VsEx8Qvtp%7--4A~Io?ikpQ0?ynH
ziKSi9=<C_{xQVAeW<&$p^yi~&*~RCNn~ATI>!wv=nVZvXR9Ku_UBgj4&#OMKB%ke;
zT$pVQM%I5TZbq?AgQT25q7>v{Ag8Hs0P@1OP?jh)!=)=l|22^(ncMFGeNlpTdBV=0
z($-lP6cNS2_9i%0FV)CJpjeIY6N|Lurpy9xHdHu*Ep{ib0K{wEF)$9uK74}^E~10(
z#qGN;eQy9a;Pwmg!eA>+Y8<RmX@4?z1GfBe2>UU|@(Tz@TtOp(#=DmrXfxTD7&=Lf
zk0F|^%BJV9a{t3~9O{CT9292%%An`qyFd1TIb>ij7LtcpyE`l7?pjqdf68ovZG+E6
z@coW^7<pe0i((+2LB5>dg>t^yv;>rP>81#<I`b!d%*?wiLh|nEAvu4wk2bPXdHWFc
z&jo^#f8R7%m_dNV!Y51rxnC0l_CvHqj6TcVx2FD-0aAw0kyilyc6h%N9wrJLs8HGQ
z8yQe8p*o9}*$QtPwu|U0Cijq)AFxk(V;-G=iuI#=OjnM~jbqDDPm|K9u(<#1q0<x0
zWKb4;$x-v!@O(g7lYpmo{7>5+29_@2;+rSGM-4S{ck^TLu|RxNt(gCE9ddN%N+GhY
zI0z)M_T##sH=jZ5`N3X(p?O)#g)-kA6gcI4i%67E3PCXk!t`89H$M0o+6X*$%v~O`
zu8AtmG=(H-v1b-suc(J3P`o6?K`HTTp57*@B#_R4IH-PR&J|uYC=p(BHDw?f<XU5(
zd8EULCa9TJAwcKrWaPy0#fjr%65fKfZ^V=remxaEbi^*{Yrxi-Xlakd*A*<_LjBxT
z2Shs8!K%z?ZVY$v4TytX7C1+nB=R}&0<d1*guLckQ7Da!>s4$=U)>Q*T<3ds<L>i^
zcq2=e(H5xzQP$}gPiz!|W=Yzz?57RKOXll^1lzQ(5-WaZ+Cs4AB1w*xz+*G$TLRnM
z^fxr&N*5Yz&o12n*(`6I;t6MM>uW*@Q7^f`!JkX&;}LkEBQ#fMHIyCxu5y_?0q=hf
zdND6<%4yXZAdwjawUMN#C(P#R9?9J5CAiZ1;>jFXFLFAy?IQY*n}eul>fAbK3fBR0
zN=^r;VGr)b-zQ2FJ6kqgaf-jc?g%FGDE<2XG9eDIqa#^gjwFFa{;4gy-R<*A`=pU9
zrNj8Rmy3Vi{jW~g{$DNeKfz#;=yJZXOsSG``$yBH{J)P2-;=U88&3y{DgNX;?I$3+
zSp0AHar?%DsePy#pJ(zMub%KYk+D3~^FeArq`)0vv+TA?7;XMaTp?Sg<op@9Dr%2W
zT$}k6O;~=pF92P4!5=D1k^|LyqH$0W=nq_X)0*+)Ocovl3)oGSWaim|;qI5~MdQxX
zdx1GlhhTi8aU;#Gm_<IQrtq2e<Lo=*^6U<YKh@$t@2NKtKO_f|#&M9Lc}|$PSGs;7
zPp)b8@Mj%HFRy|cZ_AVs@5Fpt$e9x-WSMFSJ``VAQub`H6IuJbK0!vDY?|$&A?GoK
zJ$z`sWQOaW@9+Ycm(ITlYXufhNnn($4xn-DLGvKfMukt(Vl6M9m~Fgv`bu74*Irgh
zvHAuE+ywb?pC_#O3%_w3+=a^Jp!o1r(tBu7vnDAFfUB;X$z{|!i!xllT2lR<p4GGz
zvEo!*d}RV_l^+xU+Cyx)_>NgXe|2fF6Pz#|Crg{GfD~dO@YCqlsX3H0D6H=+Y+yqc
zqty*&cIN<@fB0)v$ut$RJ}!ZjyRBwF1O5GsOY1W^WoLW4|BlUCylj<40X$HnNVVRB
z3X2aQ*<kw~L@aA1d&?5kbURkOW#B+?ia+ikiOgVlPZrv&CmjKK&`mm1W8<{O);y`4
zo;M!HOsB6&GYc=s{$f{#a*@-%Q8r0(IOgkycL)wxr}%|q<ls}Do-<$txbAvr3wa4L
z&GPRD{ba%P%1^uTB#k1yc^hFKWJX*X9Lrv_f?~apxH51(@t$6xStN=y6nC90Z7{?8
z{tFeBw!6ElT44FNe?NJGVGqpFmAv+UAuiax1Rki9zl?uqb@7_Yi083{3nmN>w~JV^
zUSWH)wGo6nk}!FR*~dobrSWRKN*87=h1cRNvPOu)M8C=~R8@W{m<40rUDWKVT?ZXc
zf68cuo>)<ev%oWWxC~tV_M9Agf55YcF4or?8+Bu#pA;W_l)LN>f}E$6sAV0bHWDF4
zo}_`!)d?HzM<DZREk3TW)laIc#q027{b2+Xj<v6on(;G|)a4I}e26w+uo|WI0-(+g
zQ<?~^*~wzql`RZ9$P+B#5V6}pX1<5Bl{oeV?9hhzfKo1kWL`MDPYYa=Kx-h@ey;U0
ze%$&M0sNDmj+Na)<r5K6An>=+<K>fQ@2aNaxXEY14;m4@2i@u(rE>tjVf>=vPnqbb
zwpwl?<XARVdmMd|$O1%^{I8VAA_a{`ZtbsE4{^8TAAm?;<OOBr(unv1M1)IkW<r?U
zvWOYufZ=w)p=@}cLZ4MOj|ypW(E(yo`jSSeAEe4cn8nAk!%^%=k}KZUMcxP>qx#ec
zw%WVUiO79(;d>)+J}cU!VIy=_X(I|3m)Oy_vl^Y78e|1#8GO9!<-xJr3rR&fm+tza
zgPFZsS#YuSS7o6G`&I{FU+0g3{epuQ)V4N@?Uf6}5H|rT^vc=nFcd*f^64_0RLEFt
zM}7C)^aPiwCCmJc<!u=$`gpz(IFb`0%sam-0BBMq25J*L%PWvR+&ltRN{muV>b-37
zw!&NKB8&&8Of7>s!i8!)Sp|UK<2hTLeHBUtQGO*PwECOsP!h^(!Md7V=(#(fR~xv1
zp<dz-D}ex@xVk(mY*b>*Lk|E(;)e7wg4{`78N&Q5ttDPqk+jTt8uCqZh_L_qR?wi%
zm-=*^1z&||$B}5hyn-G07?S;4S_+}6UG!om+A&Ly#VRgCS?h3349=R-=JqrxztzES
z(<1`Hr!3htFKHOST5EbgIlr+O=3WAi?#^{<Y<k!RqQQ>usWC2=DY8HwS^*Z&1n#pR
z-3Tq{|4$asq^j+(&&z6a(ac8KeJUlW-Oy<F0&{l8$rjCNl@C{_*P!#!el{EtE^03l
zC&7u|Y)u8NUt@_fe?huvQzPqkiC2pOCeK48C%$Ss-6`_H>V)%vh=H?z)FF9F8m(UA
zaD^biT?nj%0{3BhXeNvRiRQ9Flz!R{5=u?|JI2u0X_Nufi#xT~pFd4bhpgMbrB}#N
z19Qemq-W$RpwsqBH$p{DFh=X+<eN7_$u|5?^|QPHFN3!VEOfwXXXcnaA9}m)(m58h
zBo-3(=Q+Bez%E!6u{)AxsnkGD<HX0btYRSx;rl&yIf0A7^5aGoh1!881m;Q5lIrfb
zGq7S7Ota8u_Th>Rbv^`rFx5v({b3Q!EPk@~oTkdAI27twpI>4vX_DdwDK=1a@-!Yj
zjb)m!ke#GNC4SMjiBnj77~y%KtkPpGos;%?e`-Z~*c>diTmsjaFT6DJEh<{SupH|N
zOPIp}j0CuszEB^t-#9^L`9+tlpB0o0{4N8fW9UV}S73$T#^GXF2?)4lFUEn6hOK@M
z|NC}d`nd4KwlZi^otf;8N$S51&;-euFBpdMmJQvD2rbV<hY!DV#9Joburjvf=Ap~l
z{N`M~M;qo1QMIo~-zZZ*B$qtcZx>NM$PQ%s=!rcsPlahYt60`6l5z@|jLZC#!4~%-
zP?Rj?VlBYD?JMMGh$CtAjCr|D=L<l+<xYzokn^LNCHW}p#=`qvt%id7L@Q9I%BeRS
z6p~nFEitzK3X;C}-`^>ZT^pg8;rsmSc+aJpcOnl9f0*%`3K$)d(;5r**C*nb3me(v
zK;%DrDIZWSn43k5psQCO+%vtrLr`lM8{I5ecY$5;fua7XROZ<qzE_tdZ=quu56RMx
z@Q`DzR#)cV)IH@0MmL)<diK;qGBN2xZQJk6&+W;_lOB9i-EZ|-tr^cp<+BzmTkMz*
zdPU8~UR}dK3QQ@K8UFB9of>I|75NV^pA`!qAUv*bTJtHzbk<0c(S)~5+DFA<)~I~p
zKWen$j>3xMEm~iIe?G;4`7KFVWMu_9TWhvdSSyd4J_yJEHar<SnP<rb3mW(M=V}X{
z-T4s90KM_g&MZX#fphcg9WgN>9(0IYEBnG@=6#bJRf@{i%ag#j!e}a#_4Wc<`CA?x
z@r<lGo#qKj<OSs`miMF@oS=DbZ)kIC(mE^MmkL`z9chJia-YSK&ZL;Z;WH<G8#<Pf
zc99PZ%M+;@Tm+sPx;FxR@2P357HH~}h(7WhsP2sp-jjos6g>TR?OP@SXNsIJK0bk0
z3wvOhdFns!!b3%G@DxRzg5>P?Q+GMwY>^!_&;6FBp&&I&>K_G`aX1Zj837Yhz;ccC
z-2p<QrYb}>@+*W0wgIJdc(sz!Wq0V`Zy+y}T?g;X!^rpSc(si**dn)0pk+LB5+v7A
z?~R|w$W3pk11TC9XM01rHy$lirMhXg(-XWwcE2H!$2{x#<yKRnR}x|KRvv6+^3LFD
zcFEnZVI^zFeSmHJ$Imd5RIrx<U}$vX)gXDw3Cx!i&5b}=9_0Op;t&l#s)x;q2<6C6
zlSPoLi(u01x-<{&IO6?M3iJp&2R+72n3^dhmNIONA~k(F@I4I$ZLu3u1EAneWl#NV
z^U_=o{@pIqfP2yc#IWqD6L?#G(PfE#5SfDm`5V+{G?w>3324_LfwQ_}5gOPH*4>~*
zw^S>Dzba8?8kYeSnj;Ic0b@KNW^R82Bn`di#bdcMXXn-dW;%C#;rBi8VJvxflk>Ws
zA&?1QY))$l)NEDBeC>NAC-eT{y>@a_bqKuocXkZ{!SiGXf&gBoT0s1YJy+DL&ErU6
zDXtma^aGBtW}A1yQR#VOV1v8=Ep;0F4NfHA!v>y)W4alx@XebqK?<}AauclO&eP!J
zYDr*x`Nn|SN*45&A`m#}pPd7vzAi=zE4q18e}H}|E}~)0u{F|olcf8%RO5}FhZewa
zhlhJeXQqRedl{;|t^Gr!?8s~tU%+h^?{oPEkQzs_!QtzUE7}Hw*17@62hlm6vc!s-
z$FKz2HAVqN`n>)de|9AjwApv&U{>d}y7N<G5mxP9_Lxl)ZuXt1X0%%~7Zpa%jVyJ_
z@T|>Y+3sH)1r1Z8;ERtG_HeQ8E}HbZ!9H#qR0K}wQ#wFX38gA+qF5zg?Ip2hw?ghj
zY^a|^Q_p!38$D+Xstu2^q1NKeYQ21eK+|g}w2Ali3oKA|><)u9YPu!NsUIxRzcBE@
zUop!5zJ<tMygdnIu;)UpPxY=}yEr;nbkVf4lj`-ZQYC=dNKbXMS<4mIVL_=Cggz+C
z>8C(fv)?ahfbjVWKXN(nvNhI<Oa!K2AO}0SWlGM}vSrS*l|S1{Dn%ZB$qfyjj2SL<
zGTSOAfCr;ft~#OAnJ{da6L2MKUL>xxZBWtq%>Qf+eV`*Pp+aZx8P)dg@_AnRXD|Pz
zZ#-et!O?e4X55G&`P*a$%oN&UX{)@vu(#IKK#H(VdrqR1b_b$pc@`YxO(e)bTuJcb
zBva(i9-a3lO9>1u|D|x_&ovn$vu|6rM0|zi`~8ZRz2w+U8xt*Vu$y&_iAyN+VAQ1O
z&6QG0-EBSZYe%sxH9ZTxuXD~3{*g{TGbQw;(`KE8ppRTaIA9|XG#{P2R{>N_)shIM
zju0^QqJ1n_aH!$Gwj2Z&)zj6n1JJ^6O9?}COy5f;QRaWwM`B*8%D{4D;^|45&jJWJ
zy&n{1ba#$q3#qz1nh7&mk944(@n)Q~(>Rshlnu#?cUZGsD!ae8p<bJ*f#=!OOCAyq
z1!yl#m4HUNGM!r2Iv2|0(pOC-W94v09gqfi0;K%2_nT34<?&orjO7FEA}zZnN;^#`
z!eSz{1Q(72#l+Gfx0lw68uq$BitGgK`<pqSu0|A=>N}2quVSt3Ru~-laRy>#)jcjv
zyzZNYK(+}QcYpvN?J)QVQGZ?hDTfrzD=1<n2$=+11-ar4D(t}iHc)4mt3-Pk1h-!~
zQ$}oo*oyRIm4k0wmJbmpU5$5g)Mz})pqLPIg8{!m6+HiuAE>+7W9a-x9vYhCd6uY)
zw5G0W#<3R|*|)$VH0J{E$rv{ptVA!axa{423itcp9gwA+dc1kB_WWH3i_P>wx9QmO
z*(1vL;9gLiLPCZY;(zG2{=LN0hfV<l5PUt@N50rMbQ$=!I{f0+;2!n=@TZyoqkbQq
z!4(&o;uHZI{<yN<x#!3&uO)YxZBqU*xbBBGug&-0TL?k{$9M3B(&Me#y5`k(oNx0A
zMhmQS<Iz03pS(_v-XC)veiC76PKjBjez!XPj;SbccnovCkxM8qX0{8cY~S0Ju7{;O
z0xE&7!fC__gD3eFZP%61tsh@u5n3D9HjYYwz9r{pcP3*;;FKG7a%m5rPPlUws)1+p
z2O6_R*fmh@RWIGv1bBrUsq_R#fZ+n5M<+eFY?aV+y9CAe^IX^5>h*A*v(GN)ztM3c
z^R4%3#BQILTPzy;YIv=dx5R9wqvm+A@vzZL71x=#*u6urS#^IesVUsF!P_>hGS8w*
zT`P9qqT|<SGUSWgKf!B%IkjjUHbN^Q$k?)+275aE_OY4K!8+y0BHE4}M&(p}jv{0y
z^~KT+=ycb{Y#t-Ce9)l5PTm?iN2Kg(1P0GoLd2Mx=EQBN4N&lh{@>;+Jt)v+4j$oa
zG&E#|)K>*@r))|iKH<*fRq>J)%1OaxC2Gny-*E*+Y`=#q=gEHNMOAL^6F7B;`%z&_
z<`b>3eaNI9vfh`Tr=(oj`#>3`rEK2^25cJ^f?ii104dBO>^!nlC<XQTHW&<MxaPiq
zRpC7W{)&ON62ypWIoZla!J^r)VO9_!8w;@WP6jL$_NDkVBUw)?>9~9YY+&#3PD{rG
zTdYfW0d3mSMa6e=M4?61B)fyHced+Erhx=o=nlcWm&YEC0q=gTR41OGTLoA2w5@w5
zx!VKGLu$MJ(3@@aV#5CT+kivSszFr{;s~<ZD>@s$++K18J+@^%7PzZiM|SnQOk^S-
z{RG>sJ~)@^R!wa9wSt`;km1hqxtGKwAWp~LYS(U4viV;-xkT?}9W&E^g8w4R+o2-(
zM21lbQ|aH@!%)qqal8%3UJXiXMN*wShV4`i8DPYm<Fy<EtNQg425gHQ2j2F-4RGQy
zrfSNg?^$w_DKt}g&JZc(8~Sbvj^0NAM(?{4^gza0U}TH(INr*F3X{+Nu-#z7;LN>u
zYz-qO$W~(Bea9^xOA|u~31UWO^5nb+{JhqXWR@FcLEpwrzp;FsE9F667y_*#M0GOY
zkOKQ-rVdnYh9^3}M^me^<O_*q*-KiiLC|dX@be&TVjKWS+hq-5V9%)KAkbA*zic-^
zknhl_Rt<;}8WZJA*^s?beE{dLq?04HYSazgqOT~4XS^$o<=Jfw`trG|$L+{NydrP?
z8T)(srB>`>QEa=~w-?#@U_OOE0FC7R(~0wQ0t9gSN}&a`QnO{?C5>aI*?LZrKZp%O
zS!<q<Ud(Ig3w!M(_4EUIVpf+khyfq75jv(}N`o$+cu0liN;JO?RL|!4&iW`a?m=gK
zoGg5t7geh9#v(EXw$dw#n%vp%f%a^u<EuAMgS)M6V)->pjUMp!tuksP(4GXn)NtQF
zerg8nN;R#ysD$#=Op+7;a9>;Zlfkkv<x>EfbJ@)Lmrd1595zS-a-m7yz!mvClPX1x
z4!zC{8ro<kF1GyVP;hL&)J-sSv1w~|ZRJ82N^q8Lt8s8)h4?Q~IWj`D`F&KVbRz^2
zKhc`Z(=@s?HC(|)BLc=AL%QRVH0OE4(gwQ*N*oTw8nF)VYh(Nv$^XrIC@4)=+}EO)
zr4#;Kh3Y6`74YKI-(Ummg1S2%p*cdbmoNr0K1U}97%?Zy!JRu&qJl(DaqDvu1pLQE
zv>d&~O|E<MU>>HQt9820x_M0$SV$fPQ$vCO89uT){0|0(G_**%2QZ^RlJan*nN<!-
z_WAZ4h!q|;#SvO=X+=QZBbEJYzztX{RW`M(IIKR^Ck9;9@{X1)HT72+(J%)m|0&vD
z2EmrAk1T<kFL%@%(=4PyS8I&CYzp3@EL|Z_7#tVD^2`|e;ocm-vX=vE-B7k~L7b3<
z45%^=9=OP&)dtJ{#@4B4;$}rUbH8(*0=aIUDU09*pmA!k#Vk%7f!kJJtIOnb0gm~Y
z&n2zb+3Fy+Q&NH+V8IVARCVyqh(O~+*-SRtI|cHX+;B)ClU_H!Ir>rKqLk8rH&STq
z`@+x%;@FR7!KL^Hb(Pru!Q7Q5KKf!SekLWs+#AQbiT<bWgxQjrintrL0ydD)eM_j{
zr~y_khh6e=f`0K$@CdGT-;W*l#OnXmhHH$|0_o&T-@fc9RiEvZ(G_@zm_8H$^p!#+
zw^#ejYeg{2#;?tFDM#`??3LHT4{KV<c|_~89Pi%|-pbM^5~pg+wrHd<&uus^V(UW@
z;WcHp`$bYSLeB?~DL_IWUN8--2#%hJat4v<*xf~P8Z+Ao@JnQElnuLhW$l~=VrJlT
zs3SA<F68Pg3?PaKGODSQp?~1i(E|AP#PgYjDE&ZLK=bv-qz#CC=ubQf<`oGzCJA7p
z!^G!Ztv&iCm`{>Lz;$p4a*4EUIM6^&X*fUEeejmX2Waf=-}QdpID0zMEJV>m<R#(H
zIUCAapU8EGqgDYORRbQbyaHVy<~W7G&Fu8wfBLAztPH`|dbMp|k$u$un1>C}g5NAB
z7}Z?JA8-rI1&>PKp|oYH=L;ivUX`fnJH<7nfNEl{K?hRoS+TC83&Ip$<qhdAA-q~l
zLwY82`;T<GY*;nbA_|^5(aL>Q0>v-PEmJ=~3V`B^z4{LRpl9XHbk4E#^(UwQ{jMD}
za(KA-`dV^^*V{CQ#Xqm14gtV=;I9wR7}(c(oCLJlVQ}uJ25lsQ*Fa8tQvUAYuePNi
zq=zba`P;RWGIBNpPX}FUF8gvgVS8leXD;=Dy#31iKdA&6-!IvG|Gu8)&z28NId%7N
z^~<~FxAvW7e|pOF&fL({s1HL5Do09Ry`;c1I<E0z$sdzf^TaLoR`290CwZW0>y^o{
zc6fGk?y?y!tZ=yt>I?i<Js9+LsFxG1-I{p4KO0-}r7^Sbh{Sa<gRY|Ys_n}6cFXtI
zz~qlLyo1J7v8_RGUOnBcv=|_@{!hTqpO<7<Ya69)hiEHVFY<nec<&tH-p5;(r%$t&
zLs_1O`sRt*JLtHZ)li#t)t4FNG`Jsu66e(lIwdq;4eZ4*0n&4!+rHeQXa-_#y_>&V
z6LJ1S%Fz?F<ejfj7tEa7dWOiu4MelF(}9YHfP2*k*IL$h0BEgT^893-wIrhdUXBWB
z-DBa482tVGR&Z=nbi8t;#}l9qDh#}8^Xuesf>yFdG9VuBl=RDk`Tt;X)ao)tw(;(Q
ze2&<$hAKYeF8;giTqmUjtnM+!-^Xm70t=j^P}>jUu6p0lMa0R#&>VVz%i0(Y!iS`9
z<RNTw;5r+CwiogA92iz#g9^eMo$?+)u%}|F$bu)|KPQFIcT!>9{yIWmSM_%KV+wS>
z$XkUR3JmesoxkF^Soay}KQmOwWt9t-j(GwoV(xdg7!m}%8F-L_jMm973qHs!KdKJ9
zd816Hx`TNtEZO>rw)O}lzbBvck^H~<_`zs0e&AZ4&V>5sJTqAG7m*>%+Xw8p*qR#v
zxMe+;43Fe7u`>4%<{2$&@cCVZ@+ctG7~0#-U_MYkCJd<Vy!=d0k3<q>m%{8jC56+F
zy}w`t;-qLH<rgZ=)cW|S*7@$cBebE`X->PL1;?5DKT37XvtCpOas{_b&62zmgCz6?
zi~cBpA<|fgu(`q4b6q?nR@CoO2(wipw3emk=m-q-Q}YK+z$Gf<ZD8cclO@`pJz@<g
zN+%PA87d)3r!Vz%9A>Ws{{RDpcM%0zuI>^QlG=lp(0fS_16_Qt<{vki2?cJ}+w&O5
zOAL~o88NIC_>Ki#sCOgcq`a?P8J!O!`TtGo=}=oBBT{6g(fRpZN1jt45D*_uL7>c%
zliZ^EjhCz^Uwu)Na>9rSi)&8KAPAXwWfRm6?4Sc-U?BnQaO>oiqc9_*B_)!NCS3}h
z|11HVQ6XiN1Y4X4aq2A&AMQR7GnVcQMbdZCi#)+N)$}2r4xRY!+3)_CQ*sbrsA&&K
zC)hqMH^Ngd#I=ijj!W(#@xa`fFfh<7R|KBbT!*m^Cw6`ncZleEJ_2E*oedlwxl{Vy
zAFFS9B9LCgasbrzubhovN_PN76FA5IZnx--dFRL`z5<;W;+o}c-n-6q4lP>-&kdvm
z+Z-pCpwX9JT2N<?2OBM{GYes_qaT5pFj)t1s9S$7Do5T3URgXxpic6H&bjBVtu(rY
z(-zL`doMw&4dP%L6~2uxPT^)jozFpZFl<6didK!=hNIm`?HaRD>Mg`IHXD=6>>yk@
zENs*pSq2AVghqv3-{E}rm18ma?IF=_ez`jT`L+HwaoGQ4`j0yu9XOr=yqOi$qY0ZN
zir`D5!bQdz$=tyNQ5f0tq!bQ_k3_J}jIiK+TP_gH3!9W_&_wHY$mV$!<CytH2v2$#
zqC$!o=fiz~=Fnl74`xh^oE?_uc*j<<+Av+o6`J(HDhgW#6whmph5Cp6U;&FYx5g$^
z_}}PjD<V*<>PlnN4^?hKGh;{wI?)M}Ixr7V?&^jeAM~k%IDtn%_e6GA9j8j6f{~0(
z;f9Pwbbv}>gWcIdAZ6tp>M~(V5)WQl@vqX~|C9EN9$@I&@eSuRqn8(q0bV93k^aln
zu>Dy>fn?uQV?Ow)^S=M;{&4K$=$Uf^mP+V+<_1ah;Pg<yqltN%TYvk&2<c{Jtaakt
zj=tvWXI!}0RXu33Q-c>bE_Q$f6VlF^o}|Rw7ZNr~b&BoaBpjP&D?~Maser7<Y*x){
z1GQ8}fG{U>&AMiRBln(3+4ZnzG6LPTRtaEfJ*%uk%`z6jl?!Mv&KKA5o1gSxPbC6J
zSTZ(L+5hw$E^kf{;KS!FCaAK<={xH0kHQjfhkZ{CWdNSWv6YXzx)9hXHQ&AhL*snk
z0=jpEo)OXoN|09lier<dFy=vi{Gs>iRS%iM%YECAFi*o<<!WIWOu;>U<An8D(_99s
zN~aHR$K3@COdOP^SG2ki{b0yl{b|jOlL;@NAh_nhO-XguJW)-s_QIYcvepyxi*y9B
zALX(c#an`BSN$;&xY$EvjWH(SjE!8K0ti{h8Duma!$t0tQ*0Wh>Rw}_)Tv#Fm8FaF
zXaZ7Vi5Od&dWjgltziAt5h~rQJ3w~Q+f1AJ26zo)(a~ApX>;hqWF<)^sM30p9Rc-w
zve~5@!Y!XN!co!pON}IB!>sy<_AF~d0BLTDpvP)wzXUB=aqz5Gm(Tgj;L!{$xc=da
z>6_5%Nz%e4fJpF~fH<7YcBzj930pfQ*83`|+2ngCysxXks1_JX!#g!YK_h1vWGVx2
zlnY;kR>*>PlG?1WcmHia9=&%#>fP`Dd)ES{GCwErLjQkX;;#*$ygnTfb0qe{?`rq&
zjr9M6lwK=M*KXPQ#Xqi@K;w;(o6ejWe|<LY=P(!l7YP%X!@|e#NM}<1MdjDuIeT~h
z(HEz?I#KRqEh;RMZ)UE(<L}J8{H0$!bM~P@C-Hdn`oAqtELY4`vC(<27ct|bcAE13
zC5cWS7frpYGm5t#l%vF^Hr1vc&6$`_H~&2R6<g->=j765+ttW%8^>pd+Eg>icPy8l
zayHq_s_|~uYS!39Rx3@(rFQv#dZ5+db;6a>Cr!C-YwoUReC@*S#);C9w3&M=g6}!X
z1zGIsIGg3IJb3fmdp-r5eAyn#<Y|BPSlr6KPKsU_HTk#iA8fvTG-!kM3JOBAlINDP
z3WL8AL$LmU>(X=OE|Ws3qTaS+Xw7I&MJ<WvKR&*l^O|JOY1XcJcV8OCCBEQHGoT~!
zPRHYoPrrz{WGk0gPQ{kZ;$6U7nV#?>)fqlA?M$Sp6uY>Muqj@Rt91pe7*z?gGg#b$
z53J|NXNmT0aXfk^Y?Zc?><qD`ttF*u#-|wEJCaDY)>5V^+izGq?b9xJ{gLi^yqXy=
zOc7Jw15vgL>7>!>Ke}Nz1LEQbV|S>n)y3LkYQZ2sf>D@m{`$)hIQsUKc}Cq)bpL6d
zm+!qb=3ryW(i>4f*8QxOnbUFN0<XLY{ZsK+hJ^om|KykQDV)7mH`ize9~<?F>p;0J
zkx`&Bw<U^4gYCdcsp=q2k(rAq#pEtqrgN6#VI}Mu-31*q-1SiaQq_M$*KB3qT27kT
zH>mwYMG^myG?aA)OfK)QCUFY=IKKBuSoz7{9P_NqAA1d<2>a_3p3dE#biZ@U>|3*3
z!<}jdOVm=MCL_AY#2_>ye)!hDSf-kq8tzKS<U8C1C8hR%@{^$*P6v^e3aN8Wm3t}?
z+4h@GC4S-kqE%fHd;f$FIHlx|<Eq}j3Rk`ZBApYnIC@~>mQ*Nv%!AzNcx!vE=Ta0e
z=w7+x_IC4h*g|f<9VUR7KfO}P_>3~=)yLfu?M{YtaaNL~HRjS>a`Q*!4Q0mP6?mJV
ze>`V2xGd5#6B!e=(!~Q46GgY97J!#8B2dxXkNDafwj*8CA4RhF^eOqyW}g5GyT+y&
z)F?3@;Lb|>M~?Y78ovk${ZCE-fuTre@YkWFWbiDtTkEE&_fEUimY-%D@)VnvCF<p-
zUtW{P|HD!9r#qs-Lhw>_THJhNRJr!qrl(D*Od$@n;=K+liw54&e!XUvkVd)&Z|f~o
z1t-E`ZEald${NAL`FeDRnJ}A<=YKja9;vJmH5xaZDt3o9`)s30j+gDQHL$6USLCYX
zzhkCsdBrHsa7AZjtI0jqq!cvW0TykghPHlyk<@W<t1fH;b?5oj#aB4nk2`_B^XcWo
zY$|k_yFDZNm0qZiscSBQX2F5yPQL5xpYQR;Hr`iP*}$Z)e)PE7s95Ss$vP^$94pw2
zB4sUn8z<E;SK0scF;5aJVU6uyh)OB?X2#v_SogZ4!s+9q)XKpi{)3dRiqYMsP=qjF
zkNT#6dJ})!*{Y5O*Y}Z%r;d#F3AEx8R}}EBXKa+SxKqZ2#Kl?ed4KDXEaU1}E8u_c
zEw8QF#Qt$Tz3Dp}g@<)7SMs&$K8>>EX-r-bhQ#yAsKO^g#a@SIObm6e;=k@J(6enQ
z7u@e%#2;9>SOjfH4qCjuk6rlQ@t$}7LGu7#+d<wx(wO<b+2%U~rjIobKTXN)o!T>D
zp7X<ZeVq3B*-Ta-IcjeD9mKpl74-V^uG8W1=AT0Q81*lWI|sjFl>#l|2Q>yx-aCjN
zT#rSTAAZ^Tvv|*ZjnS4Yb?20;tzR<^=5Tlf>hr5&+P(yIdrt*cTQ}b`xqEWQD{_6T
z?%g&SXeUi#od7;7Gg$2`d4y%uuA^%Bu~A8rxJ%MH{rcf2-!DN+YhSfr$XDOB$X~g#
zne*jv<_lA!j~3I-(l=k7{>l5~kpGo#o545_q_E3ymCsScF=&g_bpiAmE+PDPC^|AK
z%>LqN+1D-k{<<hT$|^kz?so}&`cA?n%cWU@su|Kjd@MiDwO>Sh35V$nnA1U96_8Ir
zdj;N95HjWp;haA6iQ00njw1Vnpg#A5u=?Du3l(Z7e_v`z{3)v3)j%)-c%R|R|NY>D
zJeW%RR**ss!R;c+ACxk`mWvs2)m*GN%k)Q3QkC)&RW<EdeYtdCjgT~Cg8-gQCd9EP
z5^O2x)^FK)IpZj$Dv_4412#?xqW-!l_Sh}e*R)r0wZ+UBvoDd}S&9sjW!*5v5kBLC
zUqCXO5%RwjBPy^WF1Hk}!%F$=SVNSQ#*Yg<^UlHTlFD)R0n@kta!K953gsg24pO;s
zA59E9`jtj7gX`Mr{D4>|O9<yiz0>a@+axwJOT5-u18Wm&j(v5{5T~^_%$Pju?;CfZ
zDcPC=9>f(w+_hUc1P#sQg=R(psh<8QrW=}zQu3O?p-)W&Yq>&%H!a1rxI;30`cB|`
zw|>A(QmxZLSqMlKF~oa@6x5<FAfqUltnu3`a8gmW2^SJN-d(ADjGt8Q3@nV+24k}H
ztkM(!Ag}FT@f#JU*QJFawbBw2-I6!dO9Uar9b$KbvC^I{r~E8^g$sr0QLVF+L1O|K
z#eMRuAtmPAR#Xaa`HK#&D5#i3^fYd0X^KaWko;dUiEP4nsAjt7W#b?$i0l3TPpAW^
zP>!BP#1I)xuV>OI$pkZ8Yj9L-Fwc%Fe+$p}k&N)oXeA5!CcSC+arEWgK3t))(<Vz%
zDI4DWD6Mp{$&4DwprAj)d(l&6WkY&ToeuKQ*X13+&vx_PRpK_R;qwusT;)2$o9Q3d
zrDT2M5%$sXUY$ui{VVnx;cz-IxR5u)0@TSV(L=o+hIbA!mS4J5Vo%yVMqGxV0_oZ1
z&l6gwW4g-z?UxN}bVT*A^~0M@PcBY`2RXI<YhsmxLfGTFx{Bd-fu3CdIM(xtx_%tq
z*6{-(rvg*rgcu|*{#SPoCKi7~DWx7-J0Zi&nL7<r>kFi~i@8?KCJr^SmGu@9K3=p}
z=Q)MIGm5&-hF8j-b?ZF!*ufHP+pfccuWT>Mb&l+qPXCZX8{%4hG=IXR+0o!w=QnBb
z!9s^!EO!E{zotF-qAZsCJk%y7SKS*ERZdmkwzJSzZDqZST-9XpmW*2H9UHF@<OTy%
zxmf4d$pc+Y#q{-2+|(9ck=h=aeGW~^fpO%O5~(nfL*p23LwwWBH6nfQ2cxUYVY*1$
z)8*d0yt^Z8F_b)Eds0Z6G~;V-vCON%R{ogEfwLvw6P7GkCIh7y<5&q!hvc(()6c1q
zc9dhOx9PDwo1b()huhQR>Lq%p5HwZ`s(|^pWRXV4o>yvU68Hv(7lJQ+<&|XYc_$}C
zeY`SX8H%!vQO&MHl+`>5CR=Lg$ux9#?-8fAw&s%}QK@su(!K*YBlbXRmk%X~lO41#
z&E`|2TNzR+s{8JpTJ{m!6JsKq_(Y5~*F3ztAS3I@u{Dun=$Q1*;vATtlr(%tqvBD+
z_Y>*#?cH9dNK({(7Rk6gbvfxUPhMw?o2_3auOD6Y>X`OFdG+AX|B>WeCQ?!mlB%39
zCHdXki@YV2H@kU>&boj@JgD>nyO~w`+JIZ+qAnBHN=C({*HLKq@>~UVz1FZNRb9`v
zpY=(~ZY|hfPaG9?51BhSONjaL&P2#QUzc+ItvUm#W)~2@=ez|DNY1T^u*`c0LSmWz
zo#8%r?nUH|xb51$j4&gvH|Y2u7UT9c27;YkTLA9cgW>?$opW|{lG0yzhMEPIPe)c>
z?$nX7MLbj3h4{<+;clwF3o_)RmEwgpeQv&=qx+7jnJ7M4C#-AZj5*DoXXpAo%{Hpx
zfN<u1^(88MGwnU<{X;=PehIC}oltY$Q1u3G^=Ia+4zJVEgc4KNrl-F#KCq4lJh9^8
zr*+pQdMAie*L*TlBK|Xv5qQoZN6n~Tb~!@4;*8qMM+tkk;#ciG(@6pUdDp*O68)|i
zT_DACA4frS9cDkQg$!3WxgKV;`m~bA95h|8Bvn(LGp3w0TS~Vd=HV1gd~a*vVsbk~
zxOqvs(swJr(@Qhj(_Dl?e2F#|t7{cFRB;HNd_5gp-rn6)358yFZ-Eo`%*7PhGtZ>c
z8-r?>&fS)7;mVXSDs@PUIe$ZCM%#q-#xV!ochVQ{chfquKpJRN%S-bk1<U4Rkzh@M
z#d+qMpdmOpN1fkj_~DHWn)D}oM%!w|WV{xeHf<v$%F<gtp^>?u8@6Q6mz|e8C(H51
zZI2=&&~*vdN_$;l(=LJ}E~*nXlB1n@YU`?26H1l{uijL6$9s*(`YAaiNATS3()}k@
zdqvSpK^mW4Q}GnRzT(3sKZc+FI94uoOU|@d+E9Q{2(zk%hpsDZG^%@f7o-W>JfHg4
zy+)%eVs4Ag5d!k%mc1-KaAw8Q7uzddw*@b^uYSGepDXPzrndR*Hhy7xb+Vf8xQ;iE
z|6RCOI$rp|RoA<jl}k&rc1+cK^Yaagu0MvICoW5TXFpu!vI>%CRd%e8A)IgK*S&Ya
zeXXylPou}n&=KW9{*;q=b@+YE%y{2!y&dZ3<z$+~zH62LdP9w4e%jF~*G`S^S1g3b
zj41^uFJ@1}mR5OzD8)@C2SrV}EYWmQ-Nu>|53NuFIw7^A9atuobySpYl*`A5r8}~f
zxn`9pYH3|&*N=N0k7CEl*kpZ9!-SceHb>v@9uO<kQ+TQyey!PDmG0}i>d|;b?sNE1
zv9;ONxv>QMRly%=G;t|ks!lu-L1@0Gk-$b5Sk;v00~PhuAQT?*QT*AQpIsk#I<|&4
zz7?pW5?py}kN9dJm*_9ii#B)9bfoW#sGSVzt(4tL6SH?TU5M<n23zs=J#o$Tr|qz1
z*gGnBeO@Z{&owunyPwcO70>XHs|3-4NqFBghca(Yog8#~BYQ9%Lm}GuLPzy4%-48z
z8n}X#)B8u&gy-|3-V}jJEx!BFpX=sqJ_-7EdYs<p>n)eKl*BiAN1Tp5@I|=#iLr)t
zVpgJ_)=S(njLO`vyp1<6s<cb|o6d3~;Cyd{j}PU9n2g%SB^_0p^g%n5kGy?xOf=G$
z?b3fnQkDtpN*=s(x+(~MjZbkV!On5;sps)4<dg!w$t<gd@8O|dy$SSOh_OCW>4PV%
zGwRjH^-}BuzEGNdW}RuuwYEO~2hmx}(b9aF1jO$25HUnt?1Mqlw!IO)>d}wx8H5m@
z<I7Kzyk}G9?Y)rVCWbqCN%g;KN|t&o{ZhmBrR`5L3Y6b4TtE<%r<Zr9mJugcUV(K~
zHkHe&?`Hu#*T$R#JC}~M@{^hBcN>tUE{kRr#bUSZBo?%Pr#1Il(t8vrQ&DP02v=1G
zxTBJU1sByLo|%$D3bq6eFoD2muzox0$7$cDs5PG29Dy?DY^i-CioQf+S3LCON1*w;
z^X9t!MGE6$*2=r?&t_Z0&Y5e;nXQ@|YH=s?L37#Cl9>90l^|9)8LfT>XLXkSmTlrz
z6k`S}&Mw(YhhImUyGgdr_2Z&LT~MKUpIPgxSO2D4krDEi_RVgyoZg|pflRlZ33(&Y
zAF{TPdGA`o_Lw<L_^IT(_xNBs&g1>z&b6#{NG@`(F_smgWb|LavU6s}nyH8k)rnQo
z*aRd>&2*%De9@)Wf4Q4A?iIH*JTE$y=dOOWNQlXL@_3c@d&RlBgr$;!9gZ#D!Ibs6
z#Q!4t4Vt$<+id%(HAZXAbd$Qey?!dbS}$6RfP<kf5_1Q$1M4#KPZ}Y=K(}s{toqES
z&w;mHGpIW4TtUv@>ERFjG~b;LCQx&vYkvK<r2LxC4^-|M=B;Wd-wgZ@@U7G%j6{u@
zCPiiSmfizh_eU?~-I|j+i=7B&5?bR7od=b7@V2O0Z7sJkX%U<IegCTP6-ny?Vls`T
zPAAK|Ex0MqZKXL{s(p^5sJYV{{?~GO7kaOi*-;;JTR0Byavdx6X0!sQ`RCtbU8(;Y
zzO!Empr2qawj6bo<&W?*O1dqZEXsLZ+Hu?NzJ1!Wc>{ywM-|qCQ6afMIQAIN&c%Gf
zCU4V_#TK3fpQ2LeJ`fYIZ7t~Y!Q(F3U_WmH4)o{}NGR7`x|jHI9nVeTnr;mvni!|d
zEDCo$XELokV?RhI#Ga$U^U9advJl+7$`?F9j|}jyjuQ75;0Wg@vApDdqpmF3BMNhc
zdQ#27-xKoaM8mrhW^BoGtI`T}h04xFsroMAmhf72Q>w8^eiw&8GdEvj6ZV;~5#>~-
zi)PAE3j=!3EqJ;5=eArlx#!Kq%X)FtH;?|26oTu=2o?>?b`AtrON2C>%32fyB&ZNk
zjsl+F1|OX)5uu^`=YTjLz*yO3FdavQHMSIE_ptjQVdLfWh}q~~z^04WI80qox+va~
z769XGgC?EHm7HGA+wV);9yLyMj^!F~l~hrTpv$_$ICw;_7Kx3Xa)q8}?P*bNf<ahR
zwiNCeoqe9b$~<rg#d@j4Due#F1@Rzz?x*caaqDZ2RM^cV1gVdNrw1vvdq)uqI(+-2
z{TJ$H5AXGprL9%ebz4I%ju)h{9hcC0-TWU618}_%k4B3>hRNI##l@guC=>5nFQ}Vk
zR*38K=TXJtrX+9c4KCbi`tDaNa7ur%a#?rhw@Dd%4b9Il#6_(-&XVO!pNIz3SvrC)
zh*$g32b6l)z|7e?RhJ78_JSFXYLh3vdxWCPC2xr^#`FdA-Hu7)^pL!vBljuo<BG$s
zM&JuZjOLGs(rjDHn4|<<6~ANUpASHaPb7>3{_dXxT>2jnVh9T*-oIO`e3gC2W^#y)
zbI~BBE6N*jQB01bdnCW6Af$I>Ors`sQkoi)Vhd#osqt=a6Drm!C(I+f1FTHjMM}OF
zDWrguoDWB=UKe@v&l9~Px6fR-Hb@F(l@Tj_WoA$nAw4Q$msWb-PVKGkArHd^9&Z$H
z>j1q!OBdo9sitRAO5;;}Wg?xTAO%<QWpyz9?Zd7@NO%Hw(v*`dO%Y%E+Kh@>Tg1M9
z)g<q1AH9Xie*?l+0~u;Ur(XWhNMP+Lf0s5yh1)LeOH<IINHNu0_(*|&2s5$Mw6Axr
zR2~&e2?TS&HLIpG;^$ID#T*`rO+G{r&Z_UYe`KO@dCMQ3S#{=irwOl-yI@y(v%d1<
zNMmxs0}X3W>jItY9Jl)B7=>b_21n9mKWW0Op+~U5%wtir^h1J%QMA@(VLfIebPi@L
zrS{piFM}UP+neUQ>vV3(OM6(dO;t~AiMiJ#t~XY1{9j#H8P{anw+8~!B`qPs=#miG
z=#*xZ)aV95VWgsnFuGxcjP7nEMHGoqqgxszmArAc@?Jd8$M=2rar<q%uIsGhf1XEp
zGpo3M3x@<JI^iLgbAY4fZe4ww%iJOd?!QlMB%5DyEuO`j{CqSbzArbhBbi~A+pvP4
z`GWh4JIx)pVhw8qS9--O#{{Uu{^2__b;qz!j>UikSZtwU^&|CL&SWtF*3GT+>Ei={
zhR|bLNm(|DU7j<P*k&LdG8K~I@)<gJaajEB_~DDQPh4D)=40OBaQhOb2tU#2>4c}N
zLQAK-`yUved#jHWR)}D~#{;4XS&`2DM$uP*!ej$fswv=3Ia+=PTUt=A0A2(p9On^s
zs#)9nCLPWef5rY53ahHe4MzC2MgthyPP@2;a>Hf_l3%aKOHP)4V-yQBdztQtU@3PS
zzr_has0r>gd}L<=g`oGozOD}PBlGjvk{fKr43P$-%Hhf5_8|H{UQ+4<;b5B2TBf~y
zvps4xNVB?8z>=EouM63?&N(L=N&&=zvf-y<TUjy-9k><GNx%O?n?)3o<f;btZ;zyB
zmuY6UT|ZzP3I{c=3@lLde%Uu8VxrIDs?<J^0(N(;zXQYb6+ox)XlD}bC(joV&T)1R
zHFJYHc#RIPN%+Mwn{1|7u{Kfuc>^O8+k!R;^;&$46nktLN5N!RN<G>V4s$2m90>}|
zgj_*@;biO=EwI)&YiT}Z-5QlCET|Y<=VUQl-#v%IC$^l*?4}t^dr{BEHIW;nO`w;8
zZ8-9HNNw-2lKGR{@I&Mdi0g6}==b06SZDyD_aO`Si#!Ut9Jep4skRO2cPGgLopZr5
zB_%PVaTJV;1^Fe)85FTD<i#zJ4k5lr$cY65aP|iW+!*HyS?0#Szk}c!yBEa=1(+UI
z#~>vPN_r2j;2_+h-t$7!i_E0Dp)fNpJK5=rHmkdt`UsF#m0puxgt<(15Nbma>oPUD
zlse#B=y(Bk<ukQTc+UKf{)6rS_L=>=aM6V>1JjTwhKwS$QSG=`3Bp}r&#^`%CTsOg
zR^M@pH_D*j8L*<(B4vVUeH$X6&lc-rD(3h6dJYJMeg9su!fo_C_fkgtTa53t8f>+Q
zL=Gaj20#83ePl2em?A=x?g&_DbGVc$pbxD>iKS)hhlrI>Vo>8BStO~m&x3&i7&zmU
z0vuR>H{Qyu+amacl1s~&WfUyFMEEa_`3u^5j9UzzGk+%EREQ2GUqUd@r`Ml}lK6S}
zsuFHK-85>(x^0>EO@HU~;w2VTi<uH86XOo1H4pVO+U+?^#}@ebUWWBFUnZm{yD;@n
z*0X%sM=`{-iH(kJ4TvLWwL8*W&}L?2^*XLJ{DaQ>$Kvtk8QD6o@{Y0~8s9D1I-2QF
zhH;a(xPWwG<@MR9M%!wz2S<pN|K<^Zg0M$z-&Z3gR&lFWNIOT573z+2fMWSz{=b;1
zr7omK@aFlWiSgQ#{n^a-K~zl?QT;IPCg+4iFh;#kdG#{*M)1>pAZ3jb@urT;?@}7s
zvDcLM<N1AiXVwsgvjyV!%9kw(|EO+=UN0YrtuCFPJMAjYZS;A=(@<d?WYiu#ZD5ed
zN`h2t#PuP72v0WU+qhDO@{mX6CJRp~rHeWa<A6bd2RU(!>Dn*6;$#ds@p|<<6tg^9
zg!|0834d=b8Gwp*zKoDTnO`QpnPq{6f`Ky~WO4}Le8+QD17rd^ECY7~bwXJ1HK^_B
zCYY{cO_zp{5skR-5Xr8(aZj4Zp7dVUTwBWz54*O$#gSaM%G(I%t5}$t$B8WV{CC=)
zCB+4;VM&u_hecZnh_({^YwAVC?*<RBg+#hqQw*sI4~+`T&1TJ6iuZd4gPe8;cR7)?
zvyiX&I!mRuzd!+K)@8rw%;zX0)*V%HarWk&AixYaQKj3?EI74+0E7x-!~bXSYLA~D
zdtc2sj+Tu}A2_@mz4*%4ZQCV!7p(s4c&{Z@g!H=pK|2|>vTp*PN=xff6`Ol{<ga<=
zSDenR5u+Nl^o3T?txOq|2hwOEN5W@$e;~>?#bk*O*sqH|zTHA#UJ*GyUFe)3SMh-N
z)H^?E^P6urDX&g{U?2UH_x|Y!@<YZJv}k-Nl=8lpr(B+7NFu}>5JeaTPcg%7E{PB|
z(>_tuB{OQJ^tPOF9*n);W=);UiAghwEmh6Hkhf&pTzwNw1CfNk%jJ`tTo!&Sq+y*_
z0)b!#=8)SJ9*kHJaG553FhS;8)n$rdku;`D`6JX^o+<2VeX0QC48D|}iP3zx%>ZM{
zUicK3B7L@y^??d7svM0xfIuGKy`3TNk|ApP!aT9pwEn1-)e_DwsY?(3^1#A8o!rEl
zDa*c}O~YEV1tl}OG@Ef@JC!tZ^yhD3GvJLNbD`e(7w*7iu&VC)#CO!F7~VA4C&RaX
z&WkP3=3~SM4lhMtmNk4e$m+6<4Y3TiLQVI!5ZeJ%DPh*ScOqxFAHWktKk8uc;udy)
zW6@IvK3lwag$i>$!2ikMThN>2r0H?=O#2GWKq=>uTLoA%%0~hR%Pn%i+%U<yXMNQ%
z$F|~w5eD|e+kT<ix}08JeIHnl5BP*zG0wm(k{QLo41V6W(vIS;49A1+PqGy+r-|u*
z$RA6+RAY&?$lI0*8+~l&lwI&}DN6E{TE_=2UGZdOOYZi)Nv^T%-M_dKDyY4Nt~z#X
z$#lwiM{e*2NObMqskNS(E>Hb9OGWj=UgV|?piI40$d=HO_B+NX;h_UwYJ5(74jHy8
zrg_`F=POKYJqdaYS*J;|n2gT(F*%0#qu!Cygr0AiBb^9h^eK*ZolfplD>AC{_(*8V
z{vue{sHQzDVP|jG5@!vhl|V;}6z71x;*XDjx%N5m(-=5e(I>@v6EF~m7BU7h!~s<{
zXx1T<olNbL;|EVd{kqOGpKBm8L{_meGz^cV7$=P(?LL}RC!(vFtGTA;+4qypIGF=v
z;L_PJHZ-{RKaBuF?;0kE)$)zJYosE0jU87DC0=Pe{Fg7^*&9FUr_d{xgt2kY&e!^Q
z9VG=%)6#<qx``i9KTywNzNhS{C`kG;_|Wi;7aoLQYwOBmd+(u|_kTL4#y}X`@G<oR
zyVK`S=Nym)s=7Q2KW$Mmts^+6gsI8baDpyzjRAb-d#JO%Hw*@W?UDq>-39`8!my8v
zr0?zl3o&K;J3DCHE#w&k<UmyRs<c_-PC5<?T`Yav61ea;A_Cvsl0P^wv2uM?0;T@(
z63%x}d0jqFK%yD!n;!(F%9;4d+o!b<F=Q^T=mL0&Kl}#>5Eg~vb&RM98p^335;+;0
z8`$l{D`)eqJ-F7<XZljk1N%if+*oM!K)_WgK?*oN?d7$%e;P$aV)sy{hVL2UHjCUg
zIJ?R5GzuFZPFExbBL^}vUoPA}MiambT|?~D$5<rKNp68m0h9JHN%4%4*)&>UIjKv)
zLOfZf_UZggFi>?Sf8^qljP6SyKTuYp43e{ti<&1gA3&-c9JPGqds=SQ1UiU+hV!8Y
z5cf-U!lA7xvll7~cVQ}p!&;`+<HJ`l7zLNVl5eD|oE6H?a-h#1;Yo2HAIY~?zx27j
zrP_dqB)u>{w##M~`>3TDAe|re{<p$#DA|q@VxwuUXWY{%c;W-Q+)+|aDA(bjsH*+-
z`o~S75aW?$X*A3%`L0C<5~-YZN*L}YOjFxDUpC_MV4HBzJAV#)u?^%%()y@?I0a;E
zGl^J;9Pwg@z=fwFj`zja*GA%rSaRXC?9K<5)i*PMy#)#xB`v<ME_u^04mEZS^OU%A
z8Rtt1I76oodIl-ob`%|e2--ifD7a&vX9spA7Na%E$0t?ijNaj5PUo?>OnzzZww{o}
zFI@4?tHP1p$!SE^ymdPo)+nH9{BPq|9$mm!Q@RESU0MnNlAUZ!Dd6JQ{E}B+r!&<k
zdE=rU9+3BipBP*Y1iBR_@K>)z;;mHM5)oSK`b9HHb8#v)v)4vMFj@HY{$s2jmu_O+
z*e+Am720xA<%`gfKEJIuNs2JPd7P_DGFi}IDF+>55zJWTgdOU+lb4X0nxLiCxPXxq
zS$*6?>PM!|a1)_UVY)%od~(AQy5T&AyllG%5c@{ig#jTG!f8M&l*ml1_>RkIaNy<#
ze(yPWZ>Hw`%f02`22YY$3(FR63JnvhsXkXY*t}+Z`-|~59F^+rKVOW$Dq^sHz(81i
zr`P#m>}ip?!!pCk=rVKDLDN~$YJ~YxrRN)xhGhGZ#N-WE0%~d+QX@zxi-730a4TPg
zqM0l5Xa<w2FHf>N!=y0FH*#yI&IT+&q=+ms#Znuy+46vlrM9{dSCVS1+GBey@+Spx
zkw5MX5Nn~PD;osVcixJJTv{wIlLPL+!X#NPX>Ark?@Sa|Z=nDl5RjZE90=do@8w?0
zBBQS~RWS+i%#+yKh_RYak|j$Ww`0A|A``bHj&Ec-huZphV^7{y=wm>S?-eLoUZ(j0
zBni4<9V)e>I>2zhFYR`LC)~4WNMzvDY9=Dq?G>enX7B3*$sVxui78EYkhTvtRi_iO
zG?!;TN_NJh$Lk$&X-Fmv$2Qb7B#r;lK|_lUFxG%r{*I>U$~SS!St)iWEk3B@eRpT~
zQnSZHR@52w65#+js_44B+Cc&TcUBJn@JH+!5rvYvm+&A(LMz3sczsTZVn`UpWFBg6
zYI_Su8Px}72@3jav%;Z18Hq?ziJ-c;7lhB*zVvXk_j0Z8>3k$7%&#uy5dJns?12iW
zSjchXsqR5|iE0cW%Ko`og1%aQa?BKiqbBnF*pRf?E%;eJ?Df5w%8&@w-WR(Tp)3Iv
zhD<Vwly+MLG2355>*tw!t_IEH!#VKmFCIamHkd2k>!+1X4>N{+6Gn@wlrsVmgPNgZ
z1>R8fW_nTrz}38Xyi$Qp*VtuInlY*|-PEK&rM%bb${RbY_aCi#H7WC!7lI|s74h|b
z2!G-E)<ba@DI~aNDIl6Z2X4GCu9b1L%M=~W(^(n0A;xXuP}c~h-|vDdtFL^!ejV_%
z!jLC9B|JwKWfZfv?6ED27AwVQ>B^3j{NX?`7TamLRL_JY>J4Gnv7EeTdpWP;Jov1L
z(0%^K8bwd4`;G;=AsKW(bFMmeE5wM<tmXDef}uJLhMIp(FF;cfCXEJ&`4KI>OD1Nu
z=e=-0H4Ds?Fn!#!Zc#jlqmj5)hrP~0hv96211LodW-(Dyy_j0KH#C9LcGT=^9BVP~
zP1XF>y#EWlu5x8==o=&Q-qdb}T~O3uskcH-H}S}$BBAzl9}uw=JJLnY9wA?sn&RBX
zKhr%m_D65dl;{6opz4$nO<$}{F<#aabmyC*Fb67}vaMbU7XdQ1j0Fq1&QGCU@h>*$
z5~|ue-PzgRKTEy~E-GGC>r+R>CPN1{4I1x(au#{odG4{t8pf6~sFcl>bdDD<KY*K6
z%vrR1>G4ey?&H|ekZ46aOH-U>Wl>chAW&!w0;GrIO=^2%L+#w5Kq!}{kA)u9U>aK%
zg5?J_=khU_ElhVeFeRtP(mHu|NJdqNH?3#qGsoH9@-rvpZ|gUJRe%p-@YpOPy$pNN
zu}n0Hw$hyp4BU8<a`1RT@Z~V><=r=sZ4NI=6AHYJAZR+lmKb>Dr93(cQ=svTg4-Wj
zR^37dJqLCi1x^H~$fz~O-t6I2$H&KS-dr88)KqcEzy60Lg)ipV@>2+|72;+aZ*aPe
z$;iuHU~^r0D>vQwiR16d=0Kw1$CNR?LH@0}t#`rKrzOi*V{G=>Ua8~|zXX{h(D!zR
zF_{{_lz`GQ!Dip>e=6;owI%&#Fb4dmTDh7(2nu&vOOa<C?T;diH^{$rPXVI3j4hfB
z^LA5}y9cxj;0gY9@-jseDS?^GVdKLwvy@8EGpD8sbu&Wy2;FC+Oh&wog1ePrE0e#T
z>@H=!unFn8Q#@RJwSP7CYOaTC|2)j*(|5d;hM_y|x>k+Cijjv5d(#W1dRz;|4!_d^
z_I7x9<S=v(QWxyCR?@^Hb-;YabCb*@MaR$JN!56z_{F4a{4H(*Ycf}o3&+R1ICeOF
zu(9*F$=MzV;@DPzk=mTN(Bn}~jCR=#q^wlZem6sb_N)D{U&Dj*QE8*c>Dv)P3RVz!
zg_Eb9v8;Z|frRrhylo5}h%f0PXSDe${)C8H-U6oDl20afEO!kmmy@Pe;G?f}SY9)#
znsDU?Ynk*6w7b;q76*&&<L9Sdq46;SLe<DJXl$-Z;@A*IILOESDWEC#*H^Fbxg!!`
zKAHu_`O9@wh!q8URoWRHpdr?wq%oBZe!U{#zfTCbGCa4=V~q3AEhl=}(8FSKN|g-l
z5l!%wX^<bewHiCy!6b=jyN#=qP%+CRDTOV3)Y9Xj(dQjR)4$d#ff%)CT8jO4FrE`X
zO^Gq(0nW51<N7T(lB#u=HL|Fp52Q{aGn05gr&P|Mj1PL~+1&50AeDP>cb+1)-SexY
z!a}!(=KKPQk$|0dq;;Ob1(bbS^;(IS>La2Dn34itjI{#e;(!(2XZV4L5PXx7Vpz$F
z#yl|o;wE5(Et%(CZ{^-yv_lFQ|7k|nlFsNQd8N>jw~QA<KvFAznI=r^hD)B^<pVtR
zV4xKOOL0yE?uZsPeG$21lEQF_Kjuendt21#(|01c=?Xt%g}0OMwR<Imk4+CYtPHUh
zY;bxXV`y%<BH%XAiNW$8_6yvsqxcZ>o-xuTJLRRFJypWlRz*J$nwqkFkq!G>tR{`|
zXY$hL8LX+<H-@Kg-8C|rB};<g0!+_TQax<j&H;jIVp!bdXta!7Z31ulR#->GyjKPT
z_-=Y=8fiX!yjpuMz`~t>-R*93`p+Uu+ggSRDEYYFaeJ`-4E`ZCLE@idB*0yFykbYE
zA_Afa+;!2klNIvgc0Cm%swJc8OSMB4EryIVZpp2d%K6Q!hQ3)MKP!T)TpCJLrKWR>
z^t{Z5WCTY>OLq02<L#s|O-wTf5npdcz6rSKk7c~Eva$Kj6YT5ji@P{@MyD1~n<pRA
zxW9gUFf%iwKjCw6dV2b}LG$L|BopCiI5ufAx76l+bi5<*9^)nsg)<f-Tu!E0P2k^C
zJ@&)XClk&sm#fZ${Z@B9C4%yxQ}{f&xDtoT%MgEmRIt<cM|Fd#ppFu=9*60BUkWJS
z8lO?`?<Kc5yrgK5Sn~LB27gg|S%9o%HB_<4C^=MVPt>cexZ^w%R5@3m@}o{(!Tbpb
zLClVx=c;R(d`3PP{xb7?t;2f!JO~lD#=vD|72N6j+5=~t%P}Y6)Ai)y$s`XZVYifB
zS}}4BOP~Ah_jYAmys}3w5!HaDpB#iy2x2>Bjws%3mkRW5r+!*nb3m^z9FsBzKE43$
zQOiOah+7;SQOapXcDarha@+KJcLpt<w%s?ZmPHP|E}DXUx>sn_H!-9<d?YLW%hNda
zizI?-f)ljTm%$_TKpW>n-mo-TzS&aEcA)AdP!d{oT~qPz<n$xIQ%#Z=#qC-b=JNd^
zIb%}ne{U`h=0I%t6EZf%(X3y?vA(89=w$<CI#^#>bb+NsTyZ!p;TCi&aVVsYUZpIY
zKK*AgyKlCoc8`m@JDxZx{shT3;SedbK-0TDgg39ekq&L8wU`tj5LZDf&#F>QYs}hf
zF#pjKzR4}Ho}>!7GILtpzycRaBd!%m_(U^*kNX5|;=d2)u{IeNGG2(OpgBxp&*_Oq
zAI;EG^yB;zz7TTFNLVgjEq+f;0xL#>5M=D|;SFE+;SO%3W{PTL=-SR{8iO9&PHuMO
zK_F{1Z_%-lTyDi{$AJ$%x$(`v8cFwO<bhyp5S&XhH&E3`sD*MJd||-Wm_S<SmWu_G
z`E_q@E$Y$R>ZifDuY4P=?5k>JhHrlDQHOd(%}5W?9U}f@607JjMCDa_J=fm(j>@a@
zpViMoyTfp^;{*Jk?nf%zK@cnQhDT4o&^0tTY8-+b<)pA`mpKd=!>+kOrR@Iu$$Gg{
zTYiX0lwO))&O_-CPbi>nP;7(klHO%)`zZ@)u}__T+g6foD7pN6YUf-aBr1XD-ik_T
zXH&VIew)pqX)i$f($~#Zz<FP$)_AXWU7u}iY%r){%c|<EyE{mp6BPL;@p5L2pMH+v
zi;<@cNvu!-LFm+g!WeJQfR1pFnjl{V2KaqYhf=D+LMfI0_lN)IoB!`O98KQf3K)UE
fatrxO{cYRE@}5!tR=8Ib2zY6#>8aK#+lKuQwEX;Z

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/install.png b/docs_new/assets/images/install.png
new file mode 100644
index 0000000000000000000000000000000000000000..2b89b72680c6dff599fc96dc5a2cbfe014a48903
GIT binary patch
literal 12669
zcmeHu^;?u(*Y-8EAYmenq=<mh4GKs%(y4&bokI*N;H}an4bl=r4-JZdbTc$acg!%r
zFvK_NgZFnF??3P!-}msthHLL@*V=2HYwhb?{G*{JPfAQp3;+PBqJpdz0N_Bde;*JL
zU`vvp2{YJVcy3P=b%?M(engh<u+JbjE%_%vNgwSx05AfIvX6DVO}A$V{dBCFF81eC
zsBPZ99l%?^_nPr_K7{it@rEHFrvVSCgS;o*ZN$u7%wmiBqo!<qpsein_mC3ODMNr&
zP~#?Enbys_l@vn?iHR3TG2yLjm1vI8o+ZD&&1GdX>y^aShIBr@IiE?FD}*>ij1b^g
z0m;HM4i`02yA8slkOqDg(nJtryuZz@ctj8~J7+b@%XUV<mGnP-0UWX$jDTp-R}quT
z&Jf^-z@LG!qYJA*fVEPBRrhOuMg~AQjPT#H{wn4lf^afgsZ)mKeou=nuoL6{KR&S$
z&V0sWABU#n=V-}L1J}CC?Mmz5Z*}QtUV-Gg&PXUW@u^z)EO?MoChQVj({t6qzjyUg
z{LU#5sFq)e+4pVC0DHN3W*4e@-ExYSLq--<YS!HF@w?94tp!&T>-)S+LtT4*m*)Zm
z5#S@*R*j5Jb_^>mS`pFC#ox#v@)38U5{;vdUnbNzk{q@9(En8n=`|e5{oVqVf^(!0
z8r_yDHT7{g1j(|ER#rfsQU2til}LMYd6Kf+8FvS?S5Eq!eQ{ceA-JI*Q=W(T??yCi
z#Qs$UMv7`kzmUiFOU~PGPfjCoGq{iU*t%j?f<GZCzX(RiU#h@o&dA-8_-+}KJJ($S
z+K28-9^C1hnQLRIL;ulE?2F&7?bZ6hp+;2;KQ+>5u&)WbFFmiI_0hV=^X<i7Rl~Et
z?u<?|thWu``7yM_d;kcuIK>!<4mEZ{TX6q<YwN~d=So#N*Y340lD9Vj4<TEp&rpbJ
zn#|pmvdhDPq~bJnRTb)GzeNz=#h2#tp1<N_&1_5duW<FR>DT-EC?s-c6W#z`<RljS
zHLyJ%tkUb@UJCaAOQr81XkkfEn8S<Wz9QM_!C;S?F^x)>;GS5!R)7%Ljf=G07=>x)
zxc8MrPhZu(TWqHI+MZJ-I&3N?r)#9bAVM}>$D`!BdboT;G5J`3qtxT;hinQ%)htmP
z)+J^Y3wsNH&gpwKWD#c`A{+IO4b$0-X@P=5Y)>qUO`Lgjh6tIrxR$XdD@U5G!LCR5
zuocth7jh?DvNf%{328bH&P3@Vm7HE%X4kO@ESrD258YE#tW}Smp`R#ij1xyH<`!W@
zmP>U@zZm7jM%+tK((ada`dp(OX43Uos^FnT=`{1`j;*2~N~u>pE}N?}v?`$C&VLS^
zpL_lj@M-a|GnI;aJ#H;pkjf@HIgQQH>#R}76sqo-5Z~Dp&5O>QWcP|YwuHiG>5Eg`
zBYN&=-;<cqQ}Wzov4OX9IqF58p#942()Eas^g1KNVysqZCdy$4aEdYS<WLlmQ?9&H
zKTp(`@X_d4tJfx;?`x);JyXZ-53Zt=o>>Iz+=^2k3YS$3WR;n2I~?OFidj(nBQjx~
zmTMpl@Y(sNpE3M5PU@;VuND-*b~xjCZxCD<E|eDw{c3ccbyQSWj&`I|v62z4zNw>Z
zV-R=hU{m7I7&N&F3$%}7$;(r6%{!p5W!I{0)VZ(Ror97pUn9M$7*5Fi8v)GIr9Y5E
zsL#)ne#Uu<IIu`mJ=2fQSthiE&-x=eu6vEhWP~&~8`mgL!c?O^R_^(BDs_pMyQv4C
zuIG4ak_S<rzh~Qu8JT{ds6y@}467TtO0&xyUG#E_0;jbO$LIPVVGrJVH3-rO^{1@7
zEyByrxwW^tI!19kJj;|=ZuVJ_wy3oU2_3&{BE}L*cYcZ*rl_gP(8}f7xKs7hobHe-
zr9sco!<NqT7ztOGnGf2=uwtTUL*8Uz=yUo182Xdj*zxxQRgX8*6R8oUP7mei$V|<Q
zxi1`_nDd8UP&FH!c%eeXw`jJiJcPZ^bYZ6*$lAgDw3}wSEJMt9mn`PYc}iNpmUl8I
zeyI4YB}M%B59$T&;ThTkBbjqjRi2%isguF%H<k?M9OqaACw4!0-_s8bxbQxwU!US?
zuvj-|uX5=Ip*J}lnCXefGUpcd+oEUr*uMniNbSV+tnyJ2hkv;reD}B1DUzf=kU|3a
z8UtmfvXwDC1Gl;O+hG+oI=b7OaZ<A5B*$}$p4|hIlIxLlvklGWs0a5Betg+AMg*F^
zE24D1$YydusYWkV7lcK<Jvi!;SYDXYXDv?uO!<&SnEvB$^q0PgZ)gdOsHUa&Bb4B3
zbgvyz*Gn=TuT^+}gFf=OhBg{hIwDS|V;zPXm2C%#O}~|-PcT^{`<GoiU-$iVy6qt7
zGwVOa-6>q|@T;{r-SL_>Vwk=&^@XqX!~;>|r)i;o=8?tW;srMLb}xEey9eo*H}Of>
zFumb3h>Q0P7UZ}LwC6K+A%WWy-+dP2iTSRk+&Fw#N_G)%YRpt1QO;af^ID?TPdhW4
zr_=n-_-zy7KM_ZUOVDj9pd<TuX{EY&>76||ges8ZK{V!&Hd5&Rw<%iR1IYE-^`h(R
zLy`>+&#3Lj(!F%MrUR{iByaLkgI_fUN;x;w4tMy(MGp;r%$jo6*&65RRF26RO<E)4
znbLUnUiSC-X2=m&{s6q?zNnXUswSk2{lw2`H}u^-zwTqMS0D~%YyiB`rA)QfBOse(
ztDQu<*0Q{OC3<i6XK5Jz20%}+T=q*pzn6D+_|B=~MS<?kc%=mzZpCg`o*#_@$MxTF
zLC|ehY2a#>$Q6i12EKSOa3_(W|LO0^8MS`xkqXC}NO(wyp#P7b&0r<LuIL97#qFTV
zB_XcWkUuic{w3pb(yLP~vM25H=syrqY58k_s;}cjqUBu9<3x3=&D$bEzYQXfq;eei
zQ9REy^&-;ftH@ba5c2o<uR3wK1OeNuGw1|<PN4M{cZV3NIKk;UsQwsbLU*&gH;r(h
z)%nX)$Mv~U&Kc29sfAh2_8j@~D5Dt!L%;kOW)-TmZzabhah;`B+r8SZVA}NlZ=>0s
zNPk59_GZxU<wBV84Bdeq7mQ`ll_2>QOGRJX!gth_m8X!?N|-|Kr#%>)X7)U=iZe21
zCkm~%6<I<T&C1MG4#r7{^|7>joTFkiqInqqR5!jkZB`}XrrIoLv1nEN_k$mT#U<Ru
zvA9!5%7@G%3*6yqXH2#lX@3+R@&+F}>xk!Q=6tiYcaDi#a_KMZG65fTGAK`5%$;+M
z3*^FrG2OxS+5S_pzZ+A*_m|Xpsz@-0PG2|jWWLJXI8?NC)E&R8(Q1a?Ve0g)CWQsM
z?S+!&h5y<ag%}Q(Byb`#&%oyw9iMg^;ap0FiI0o#G*dR#Bo93Amc};qO>wQ;DBdL}
z+$?qa1ILadt?A7H)Hp{GT+vGhBaxwRJjP+=mBxk_%pdkB#dg8AuYZj`?+lX1lXiGP
zN7lsG%G!9{WQdJf)m8P5d;KOu7s}VaeClnJKdTi&rb>P1{=S^UxX?H%xAiMu`%R_(
z!?@@fQN_tBmk(37vI8X3w<cHPzMVZoEcRI@qMv$>g!sQ1RS$imeCdng?ki4kEV$Nh
z`q%Vt#(nulTyXz>LN?t$(dSA8txaSX89xtx2V&2aTONV-G@O#DuBUynz8zZAs1+4?
zFOfrW=>E_7luBIsh?{ykitjrf<nGliuk)4_!+FN7G8%*ar|+hC@IQRFLt_9|>Wx9m
z%w0J{BYcC&_jYt%Bp^0ZnusI3yCt-?OX8Q(wOIt_@Y=Yddg(qE!$_+B0CR>FkuW!K
zlB_E9VaLu@IRP)d^1Ld`p;>+DWrVVzuk4qeK%v#Be}k8?0@K3#nC_YK;)jZYj_UFE
zwfn6luKR0^)+&jMupQ*k6{Xkra9si4V6icZR`B41Lk(QKUH<r%=SD4Mpoi7T!IOWm
z8w=ovS15q(iX6+!D3g&5p9u13;(-qF<z1ju9Rj>c=VJZiWU*ZyaYzFnJFPRd{zk6;
zIC#<+dk(70F7FgUIOZyXBymsv_#P167WN7^`BJ&_vWv6|F$M6V+#vtve?k~W^K<|-
zAZeuZ2T`#7o@+sXJ2Aaaa{k8T80_8ca6*gsvI|8F4qnSA``h<}f7>dy7-yFTD*NN3
zFC*`uPd6EXcFPJ4j?4YENMi4U{?)O63=P}mdo}=i$d(+e^k0!&O7XuU`QN1_AvVd8
zk@sG7n6fOkhsC|`A~tZoz(wqLFoYTyA|<&bBusKuliAo)9?E)djdSz8?2AtiI4SVq
zy!ExvZi<DKwJQjhkm+8|y1K(~ZrHR-$F}Q;ru?YSP<t4S)!~+imX6Mdd60kj2D8_+
zXT9aBCERB^*{P9LUHrU8M-QB2%0W*xj8kk_75(Ll?Y{RHn&Lb<_|pXL)LsE_D6Px)
zYY6nq^zH^UGCMoFa<Fa>z@vC2g0;+tQ@oEJJr3|%j^jPqswgH$$X}d$`5ZMKeP2Ze
z8$3r!aB*@fJ2syQu1~?BH+Q&V4cx4|z2*Yoez%B`aWUG_R>!43d94p#p?|#H*E98b
z)?kW}xdGZUBaM%qRC;P6Xex@@VY9+0FU`m)TOO%5kJt=m>vFhmuu>|(1N@hy6%!j^
z$M_uMl>$BS5$Sn0d!iSoB-z?|;k<$8IGcoeFGA@+rFu|P*VT_yW8~pZ&1bsSy-8e{
z!)?2w`=@juJi|yVHuSVt;>_;`uxF``4*)gJKbgzzClJD<kYZJ!;mi!=eE0ypQ@`69
zYtxiRO3km3JLeaZotx7c&8ocAMrkJb091NN_0rn9YdcrIbna1yf;JeiOBl?0Nsfkz
z?Z&30(EgK0FO6fqw`m4G%T+-)UZoLK2{`vjp7wzEu;ELKm4ul}V5dTw1E(uAq>u{g
z?HtD@FF`04!90m50A`I*^8ON&`uBNth5@H8F|l!RavFw6iP2@(MEZZ!hRR+La&U9U
zt0oJ!Eju%VO3@{EFF0=ii_E|V|35o3^@_T>b`!O3dG=F>%Ess34(FU{)Ql9(*z-Gf
z5VJK^h5+NI33#A~k~ro_Gma4~Fee#d>!!I{Hsw{Z2rbZ6O2PzO2U?rFy_s-`Xmx*i
zewf(TqOs9i^@}~u@<CPiD`Y~$T~F5Y6nAGES-H4Isc{T2%Z5H@VQ{OX9SM}+Yi`~e
zM}!Bj>r#Qh`Rs@sqm8v=Qb{$P=m8#jtD@iXCo{?NQjB_hOlazZnL5vkuIUD=3O__A
zI|3_tMIWD92TB=7N25fSj;n*U)<N>xltBlZbLx)uC|yNG8&-13V@Q$dSz?N77g1+-
zH=m61OwQ-e-+BTu?I|5%h&D<)nv@R-&u`<UxJ<6|+8;kGG;t?Oo!XA>>QX|tt7w?P
zJd2BZ`V(gAJ>mI<eK|>@R`1u4$*ybnJHR`&@?%K|TLwDO%7nSo6(L^lJqqnmvgoKN
z?q}Ao0y_@35v>tYKj1WihJ_8u<>jA5AZ1lVECK@05KqHmGiB6bxlQU7bT`M!Va`zd
ziSlQX=W8#}d#&WfcEyL+jsxb(vRZm3>brYpFtzQiI8sJky*DGlyUBJXM?6Ix*)J`B
z6c6Whw=5C}E+Z=GW14x*=>66*37{3utDz@z6~(=1LV8sbyz9M34Rxfn!qRFID4k8@
zlf3j-G;2G!W=_Y3@lgG{&8J#k`x@qrZsl%E4amMY5H8{SfV6c-0^ggI2|rFyW}Qj>
z&5`PQ{`YfoiX0sg_&;#zSm$ein!t;I%@2Mf`NJySiY3G2Cm?>f-#e$-@|>?cZ(F27
zZU91wB5zJWLiWeboLeK3lIR0%pQWFnz2CRas*NKDi6QIjHi&2g`@t-Eaw>64=hV6{
zy+~|ir)TD0HR}^B`Y2%NyYljbiloVHPeKW}*Q~!DJ@}xE+H7Qy2qJp%$jLIz7(7!I
zof3QV0OjFvM9ARB7$I?}pEYHLjW5^%dY`2z*X<>{Vtf%4-DF9jZgKf!)gieQj=Aci
zK;Ywn1oZK?7BWo=`H}B+)2gzvlI38GdKY4O7miuflxogZO=UZa*k(A8FmecJk+7Xa
zrNZVc0<yT~m~K+<;)IhxsCgSqp(qmYa#x)7etSf;y>A&oqFL0Na0W|%#$tntr#sCT
z=TBSVPHG}@j>l)D3$C@vm82@><QZ_VbNwd?uxB@BH;_?`*i*qXnU=~Ep{NT@8PkM4
z^4Y0<>0lO_dE*;Ty?r#wPIQD`dRuiMGASy`db)|fG2iSx+^^tJNaC$bC|TV1@3Iv`
zWzs7>X(sGC^&e3p*0)gE-P{$2+no2tIh@XrT!nB|axP5GVtCBg4)S<%gx=T{NY$L*
z&#NFpf))wzp?>2Ve0+SZN~nTiBA04iqMW2m&-ptCw@E_>#KH%jP7)sRIrC(!P_e*q
zC@z%v>?dxx#8{n<d)4N~XkpQ|+9b@$3DsmLJ+gRqG$+J~zA(*#OAG<}n3In=&su!L
zY{{Dl@-@H!vrj)eJ9~GwVocG@je}6=;YQ%WzB~e{J|&w+ey~2%eKM_4=pP)+>$M|p
za5SnXa1Z4KH+Bo7a=2r27HDGAyw$j@Wn4aYa_X2cr52JVtmBK?;8K6!`x>~j%MpAs
z#Ww!~(AU(o3D}pqzuO4g8sL$kV;dsBMK?Rz%wt@mq5H|=UC1Ho0}EjL!8W3G=4jgF
zMSo+s^|8xI@vt1sru01Eh;|BQ`42x}Xn|VRTS8Ns3XjO-nWgEO_{62Sw6C*0c-DPG
z{Q>XKo%9912oysQjmBo^nZ(fyJI$C=zFXNQ^sH#i!E$(|mz0b(^ag-xbH}eFL!a-p
z=K6NSOCO%CB+DJW7`A9NsIWG)5Ey?9*q;OWU(XMQ)nW=uY9}`)M%U?(EB(+icxp9<
z3#J0|g;sUgl%VS!-)~6~K?I<CEq9lk=a8ajR$Z|L)-{5?CqL_74KfXCQe9+j=YsY^
zXdEAEKju{-pLaEQb*Nt~tl_fuC@5Kf)AQL0l|lWHZlE#!M9zna&xK0W7bkCqs+}~6
zpE9c9(^3IaLK1$+)m~V&_!aT+N3F83Pa<M<jon|K6&9Pegseo!UI$RP2g`b<@vQw^
z1%t`i43L*;KDov=f{AJh&dIa6f%+HYwV5v2AdOYI84q}oC^(5O@O0l5sVZdXmO1OU
zNQ$)c(+qPP<+^u@ArH7XBj>eSd0^+LTO(fsDlLN#qm?L4y_<|(J6Wv@(_L#<P5B}Q
z@Q88DO@j0d@U9;(ek1lmu5_-ZLkl#2PTtiFC1&9BGT!eeWB~}sAwL^dt5kD8iPW<v
z&5HP+r---}ol)2^Oe%Y$M)i!`Dm{`u?j+n1W_ajKE|bc6ww4t!CKg&zb|ere-?IM8
zq9^D3B{4+}r>-;gvTFnF#jxq;q#xD@!yUs7=LlmjKL3NA2BS4p?m^_;zU`&pSZ{3S
z(&jhJXR#`~qJTU9{?=|cvo-Tpi`lv_y=W~)ts=4PaDMdY4&C*1QI9XE;EYD#g!OEh
z(k$dGt&$V4)xf&Yo*NCG<=ztX274_y70&a_IOY;KW!HhKgjXT8!@-A$b*3}}uikhG
z@Fw@pkiR%8{#6~(&t#{jvfJ;_yYC(YSLuOPpZ?9&uOF$r6zV;<mr}iYQcU@zpiON;
zlVGnMkrzca9LvkMtSx#Q1tcjC4m}LjR{K)rR<-6FZhLNzmoabklcv3cXyemT0v?**
za92pgr7l|Qn#FN-(Hjlb^}9_*g7*1hEL34YjUAjj(I#MG3^vZ<9264Z(<^wiGUKq4
zH{pR*4UZbP^`RWZh<5)-Ewms&V{?Y!V2;86AnJkso^;q~*vaF-LxY!{xAcrHdO;)f
z6MhwwdaDOpd`EqwKXpmFR3(_MKt%D)-x{=u4;3kmJZmZTV0C@+BsHOgUq*>s;y6;^
z<%98S{W2~8<kC%g?yx8py9epav?&Iy7BvoGJh#dn(JNBVSpj<ggTb+FQV;qJ?tDY%
zi`TwFw(!}2M#2iRU8kw)-f2ScSmrDnj`<nr;q=kzKkRAugl2_%ljh(fGJ!g?&E`QU
z^iQculM;bpkEdbOR%bkP3!~_1BSELCCr#W-g*^I#C;=Z@9JOoC#=uA$gJ_vP%%Z>_
zZ+2GoO^5%AOC2N2_%%iH46>lL1XX{eHcseMRI3WZ%T=Cf1swCIPjlB*{U;vi8fZBI
z2NzmU9^gwM0d#+{qw~|F7e)BY0=OaYo^Fh$DcGT4g8JpNRD>N&8EYIb`_ruZ22o?i
zrR`*-5E2H{SJeeX0aMN1WDtYx$u25AwaBsBi7?Givl>vp)1hR<_684GZ`>&G$qOwk
z@>%c5x{-3P1%$hFyW+>=RnHCaa`~Q=StULg>@YL1x$WSDoc}|MyJ=qE?g#_DF9opq
z$DSztmwrU?qU$<IORAwWlA;XtO$=M2JVzUa(CE7|AtN^0{(W^Gn@ep8rm?%dv|qg&
zq`+BDlJpSio7gKm=$$UW->8|TddFr5@Sbvj{nMvtzYfwsSFx8u*UP9RpVw49*ACr2
zAZZPJwdOlJJ*5YxE-aLI%!6^p$q1>k$<=#anh4G3tY{XLXksWjZ87xLHV=tAsKO0}
z_%AR6=aV>^q!4vU9{l9M_uu3vzcTRyYC>Q8nj<p`K+Y>YiRB)gpwgz0Jsun^rn4m-
zVYJnyn4w8WZu6f@HE7IMiEV+AG%Qw0gTc7ubX|e-UkQ5yAa+DSnF|PS`Cs2>LU{3t
zVq%-mPc}0spRh<kROxZfA5+SYNCk2j>b-J11jf9`3=9lhhZNZjXp|X7^CARp7MhAm
zDZ{p}c)rDY01=_&TTO)E_qzHwy6l`<ia{PdYZbj3sO5!6cgt9hfTzBytajw{-$`<i
znvL1h_CGK?pIOW8>}E*)V|rF*nAwdP<>`m>zIFr~G0+XAx?r}4*u?5PWm-h%^eAb!
z#$B9gKIa|)8{pOGD~rpokvy@8@*!%P$M6!c@&o}}3Xp&MBrV_|(;EP;(frB<NE3AU
zG7%`yHc5G~o|o=Wsn=Yvz1y5#8^D%sSv|DzTo>5fAFV^>s_xjhD?@Xe$RcDiSg&A3
zVY?_`bxmO#EsYX5AsT*x!py0c>26ByAKxH^i97<&<BQ=Of$TS!C``hNK1g@<m=J7<
z9w^%{e7}O2nmgBh-}C%q{p$3L`QE6n>=4bs2y$Rsj5}3WvrlA|b{MJ0N}e26=CS$C
zy#4%Po|mV=bKk3G{4_d3mA-~=X4VE@8Z<VYw3??GP@NYf^5|j)e2ra}cZ<;PPilg#
zzjl{fk&=TjNBYBQ-wi5W8elj>T=BRw3Y>0iX>^IEM5ywb?c{xw(yJpEeo7V_+X?<)
z7WrD05Ilkt5nHLt>$)H}RI9Ge!7<IgE8Zka?1eXH$Ufq>HeMDZaG^u_=5(MXBELnB
zakJUbiR<2;lk#T?6+IIVXHI5^=0t06eZ4}sF!>EjXtMZ3T~=^Y9_}ulR()R;ZpNWt
zNQImte+jAG+SJu3JN1IJ;Vpx(iyznkl%b|Z6*e{K;PoVNacaa)I#7D!Ji^zpzE`sF
z1h+1eAzYO!q2unr5Z*6j&Kk(I#~qFxh54?2NOo)(OzJp{2yA$@<GugnO;Fa#(*e~f
zf)zO~xIOOTzRr~A_PQWDrHOlxvEP21mZpJ)Llkm7uaMfWQK9*2k}t0>STHsHr9-vD
zp`#zXi2mwrzynoH0KHnVtx_7y5d<v+?ZH-4U(WoK{rL;;nvd7+n5v@y39+ch)~jus
z<Kq_*-BNP_Vy`4BQ=RzXZC9*u*K<!J^ZS{cL1ERwj*6D65s=(!`4<t&4~Orm-Epfs
zpn0StZhb6%(I&LiAz(IIrX*wpqy{Xv4IRD<FZaTz`5gVTszUL=w&<Ac!URS?#P!ad
zlc^DZzQM<hW{x0ttV`$2g5TmBtLs<%Rih6HD~ZoGPToA&8}Ju<`6;7-MoH*4fQk=}
z=~Z%_vC|Qf9y975g%K5%#e8h9%1k%V&QmM%UFLsld&l8b2X3);_O6dZjE(hZa}$sI
z*Mz>d!#P1{nH$;i?%2EbW$PChxBe=@n6W6_#vB$b(LQq18OuQ_ehU<;KPOaMBA+HQ
zY#m!*PmhN8%zBXaY7Uw^hfah`v$JeAIUn_=8okgSo}@G~2S%FL6Iy1h)oXW~&UOZ5
zio-H`S1KEQQRQ8AM`_^At+!4ca`q_e**&l#ZzH{OLeu?zaiq!{<Tle8?||C?&v&wI
zu}{wure4po?QGbUL~HtFkvWbS^_w~d-45E$l(>qG6Ych@9YpQ3SN-3l!Pb|-o4rC-
z`@`y^`VTU|3G4T*W(b}#K&M(B1Z>zCoJ`t1hLWZ#i`ooXx$e}i>RYW>KVnD(_l59M
zNjNJrHO6$=4}VxQOEH(n$QP;zY>4DHUu6K7ti6A}eP*;j&8bCAvBS?o_JfA|jha|n
z(nalhH>w;F49~7Bzi&8N+SNFJI@gtRgH2zXh<b|1P%nAI?@=R9X>;m<v6`4etg(Y`
ziY?Dx`Eqwreef}5%qy`7Ix)3&Q};|1br^n2OWGmsEyAU9G)WOAAk<fOagwO$Lt6hE
za*0>l9|r~msKbsL9MRce&<4hKI9FAdsK`Ii3bofy(%0Kt^$|x8WlNqO5Wn%Q=47KV
zy`!!)4^~^KP<lozBe1$pR4EIRCy;hfn9<vHsMy~;nhSi1x|8bCo;XGrLN98ZVbsF!
z4kPqWAzNPyM^Q6XzEqHsh8LlyPfCZP=AR%p-H<C>p}zQbgwkBCb!UlJ=<&r!ing`-
z1l@XFr&bJ61C7-O;VhFF^P|kK+O0)o|E%|p%=ybZ1U!U!AoXF=nB|0R0bM5f#ct^%
zk%r-|(Xr8kv?77R!vp%>^PjMrZs4YvUgNY(_7rpf+Kuguo%$n5z65q}o~V+vv;ZWR
z>lWmEOk;a0H^H$W(QahmwrJhFER19jImQ#yui`ucKG#j8|7z@}%+WX#TXOx7;p9%E
z+#Fv&zj&Y`d){U|<%wJ#eAo`!=(uv5p0*vz|M+D^ZK2{k9fd6nI?5?__@vhR)SJ_L
zW8k1G4y~(RYHBn|lW&AK3M(JovCAn#X!p|5BDfo+MU&!MeAkRyifNkSB&IwSl5DcL
z<&P?>lLnDr?g_yx`9b@6VS{GH1@MDj>la=Z7(<x5p~VhYA?)h#(fIm`Y2{3vN7Y8Y
z1eiPt(63i@=T0oKgm22uE!HKO$P*q24^EdYELpH8B9wB%uCRvTOH0gY_s~{tLYrsn
z$ZNUxe6b&sicL#HZVOF#IM52M2_pj%9eV>u{nh6l?fNv5el}gKJGS{y`xJLqjvlKr
z=L1cLlXgXS-6Y&!oi$o@7d^}$7H}VLG=lB8p3X1I1rChDz!2`6!#_8_XXBxl7S+S%
zl4L-dl&{zPXE6us4qO|)XDIkG)r+FlRxC3p@H<X^r_D~V>Gx;`TymIc29Qe}M52NR
zV(DpS;sQ>%gc8Fn#YQ=lLz7l#g^mW;`hq~t*pR!t!aU)Y?uMBAD~0qzv+JRYrA;YM
zKStWIo7tV!k%=F?sXWx>89e6QBa}`Z9vR3am^Atvl$R;fZ*jb5F&5H$l7sbmnJwTR
zTAs!^=*7V5RT|H7l{7z-#{R($8|YkUE+N!wcXlk~E9o}pDvusmh}~3lu;-C@eIMzr
zM^vZru+bLz^{;)3yHet)_JYtIE1KgP=g@faj;d_vV3QrCh{pncbfSrymG%54xaa*&
z|C}K=jpXUbq*i;l4CR|?%h%riyXqI4>*REFv!o4EZ!erCwR4A=+aqZ7cAFxpXAey4
zC{6gs-tOa)_JKxt1u|X%BNhWE;eIy|4E*U*VP|U}uyAZ)*MJUkX<FwPEqD|ZdiZXm
z7>NVEm$jb;hYy&SOnqG&6C<TNK^J<RcyGB>Sh%;`mc(9>IU8t1-8a57M~G`J9tL_E
zzli-ux`7xQl*A{e#rPn16ZnzJ2SpYA@!Y8qrH;>X8$D0j&cD>=IpiZ<58%1nx@&V#
zvoP|Z5NupYeFmWu_b$E4CzP}MY9(Qvu$*Avs3&xm#;K7C@hwP!=gGE*D0=BeZ^E~>
zu_d1i3q4?DFUKv9-<#}?aLWd0lpaUopw%qYJ3et1w4FGxHx?_?y){~xhyF(I)J3b$
zyacBN&a7-B83Ol(k*)K>75S@XQdjx&?IO|-)olCQN-bKG+!O9_>GjKn26;rt%s93O
z@(TbPivld_HI}b$X`xQ`7|Q5@sx-$6;f;5*64`Sc?1mVejnXvwy0sUrp@qwsQ8Pmc
zh$jPeyqXY+2m*m9svkNgr=uTf6R%S@-=2CxFs3-QPwV!Q*g>8ypvt#H&>?<#7aAGI
zP*fS9q9)LqRHsppDaUxuTP5B&v-k>$&f2E&#xCON=Hx)#M=HL_Q@8L}<aHONe87g?
zT2)cHzmK(3%L_tUqs7Uw>f?cfN^T(sWY#)Wkp!uH=xoF~_+ttpSzATjjG@8>+#$2;
zc2;(ndgZ{W2^~_z3yMsWs-H87W$4EVzY0-BM^<NP;(`J4j`IM|6kIE&-w7oU=~EI4
zfcyo`b%2~%kmr)ydyM5JL-!XL0s5!*rkCU-5oEF5gk`Zu)-e_L2gSo(qW{o*&gFmv
zB(w~*4if)3-f+n=4&W+~Gw$MCQi1as5N2P%#{Tjd2nxiKp|#Y1`NmM}Bi-113il=9
zC=D!8UnZI2(XvSc?)q=s{$^JKph&w*zLUcF6PC`jAnjZK&!zw@MOt#nk4m>#U>`#T
zg!4XJa&Fj(3I03re?|Qt6`^G_w*zST??J<a2K5*hlsrFi{;MnB^C3{z?TuR;nL2&9
zy56A5a6^|It|z#YqkP2a|J5-9P`bS-W&%N<8B6O6(2Ba29UI3y@wrC-SLqndV<6o}
zBa_cKxk+*kB5~&-naof(v@v-wf=T?}T<sQmP<pr5S=}>Eda_@yWh4@g&m`YdPokv4
z=CkhW|7$?`Bun*$%Ltw({U(vvXc;ny;c?4d=Dc^^9)b6GzW-lov!5T#BUMc>6Yude
z<EeFlEjSMk^WEMfGj5le+Z!YZ?|+B<f$z7FyvX5pwi-g(t^9<Gs|d*FaXol<EwS^~
zKY?s`AxeFVf9d`oc>Jf&a&<1y$)~3a&Pv2$z{Up~b@S(6c8U3Di#XLk-MFJhvPXM4
zdd4~qdI8B*!tKRHoVVJ4!Fvr~I`D4Y!|yRmV#7D%*-hcjF6eOcinUxAOT?c%E3AQU
zQ0_fT+CB=n2HTZM|ABaWA3l}IeVh-S&c~BZ`kO*6%@~0F_OU;#J96{#NZ!52(uwy!
z-u#^+^9$>~M_?-gCf_Ih1qsb-jKI_2s3(_D5Oj>KXjeDV)V^FZ<9#eOm;X?U=n|AD
zq<(RAHk`KaFR?^c4$I5kdU~1p2f|yB$5u!=$f^DnR6dqD%=)Hq_isMNDYjxUuQ;FM
z@36P9IUG87IPm_io&j5NT6s(0Zzla`CTt=IF>~<WME!kcY(;wY=iI-;>R@G{X^~Cw
hclH0fg>o90?Hy*n8bj!a{r?F-QBF;^<cay~{{vwg+7|!-

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/logits_processing_diagram.svg b/docs_new/assets/images/logits_processing_diagram.svg
new file mode 100644
index 000000000..92668e6c6
--- /dev/null
+++ b/docs_new/assets/images/logits_processing_diagram.svg
@@ -0,0 +1,157 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
+ "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<!-- Generated by graphviz version 2.43.0 (0)
+ -->
+<!-- Title: %3 Pages: 1 -->
+<svg width="214pt" height="784pt"
+ viewBox="0.00 0.00 214.00 784.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
+<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 780)">
+<title>%3</title>
+<polygon fill="white" stroke="transparent" points="-4,4 -4,-780 210,-780 210,4 -4,4"/>
+<!-- inputTokens -->
+<g id="node1" class="node">
+<title>inputTokens</title>
+<polygon fill="none" stroke="black" points="48,-751 48,-772 159,-772 159,-751 48,-751"/>
+<text text-anchor="start" x="51" y="-758.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Input Tokens</text>
+<polygon fill="none" stroke="black" points="48,-730 48,-751 95,-751 95,-730 48,-730"/>
+<text text-anchor="start" x="66.5" y="-736.8" font-family="Times,serif" font-size="14.00">7</text>
+<polygon fill="none" stroke="black" points="95,-730 95,-751 127,-751 127,-730 95,-730"/>
+<text text-anchor="start" x="106" y="-736.8" font-family="Times,serif" font-size="14.00">4</text>
+<polygon fill="none" stroke="black" points="127,-730 127,-751 159,-751 159,-730 127,-730"/>
+<text text-anchor="start" x="138" y="-736.8" font-family="Times,serif" font-size="14.00">8</text>
+</g>
+<!-- TransformerDecoder -->
+<g id="node2" class="node">
+<title>TransformerDecoder</title>
+<polygon fill="lightblue" stroke="lightblue" points="190,-690 16,-690 16,-654 190,-654 190,-690"/>
+<text text-anchor="middle" x="103" y="-668.9" font-family="Helvetica,sans-Serif" font-size="12.00">Transformer Decoder Pass</text>
+</g>
+<!-- inputTokens&#45;&gt;TransformerDecoder -->
+<g id="edge1" class="edge">
+<title>inputTokens&#45;&gt;TransformerDecoder</title>
+<path fill="none" stroke="black" d="M103,-725.64C103,-717.67 103,-708.77 103,-700.58"/>
+<polygon fill="black" stroke="black" points="106.5,-700.38 103,-690.38 99.5,-700.38 106.5,-700.38"/>
+</g>
+<!-- logitsTable -->
+<g id="node3" class="node">
+<title>logitsTable</title>
+<polygon fill="none" stroke="black" points="19,-593 19,-614 188,-614 188,-593 19,-593"/>
+<text text-anchor="start" x="22" y="-600.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Model Output Logits</text>
+<polygon fill="none" stroke="black" points="19,-572 19,-593 85,-593 85,-572 19,-572"/>
+<text text-anchor="start" x="30.5" y="-578.8" font-family="Times,serif" font-size="14.00">Token</text>
+<polygon fill="none" stroke="black" points="85,-572 85,-593 188,-593 188,-572 85,-572"/>
+<text text-anchor="start" x="97" y="-578.8" font-family="Times,serif" font-size="14.00">Probability</text>
+<polygon fill="none" stroke="black" points="19,-551 19,-572 85,-572 85,-551 19,-551"/>
+<text text-anchor="start" x="45.5" y="-557.8" font-family="Times,serif" font-size="14.00">+</text>
+<polygon fill="none" stroke="black" points="85,-551 85,-572 188,-572 188,-551 85,-551"/>
+<text text-anchor="start" x="125" y="-557.8" font-family="Times,serif" font-size="14.00">3%</text>
+<polygon fill="none" stroke="black" points="19,-530 19,-551 85,-551 85,-530 19,-530"/>
+<text text-anchor="start" x="41" y="-536.8" font-family="Times,serif" font-size="14.00">foo</text>
+<polygon fill="none" stroke="black" points="85,-530 85,-551 188,-551 188,-530 85,-530"/>
+<text text-anchor="start" x="125" y="-536.8" font-family="Times,serif" font-size="14.00">4%</text>
+<polygon fill="none" stroke="black" points="19,-509 19,-530 85,-530 85,-509 19,-509"/>
+<text text-anchor="start" x="49.5" y="-515.8" font-family="Times,serif" font-size="14.00">.</text>
+<polygon fill="none" stroke="black" points="85,-509 85,-530 188,-530 188,-509 85,-509"/>
+<text text-anchor="start" x="125" y="-515.8" font-family="Times,serif" font-size="14.00">7%</text>
+<polygon fill="none" stroke="black" points="19,-488 19,-509 85,-509 85,-488 19,-488"/>
+<text text-anchor="start" x="47" y="-494.8" font-family="Times,serif" font-size="14.00">1</text>
+<polygon fill="none" stroke="black" points="85,-488 85,-509 188,-509 188,-488 85,-488"/>
+<text text-anchor="start" x="120.5" y="-494.8" font-family="Times,serif" font-size="14.00">11%</text>
+<polygon fill="none" stroke="black" points="19,-467 19,-488 85,-488 85,-467 19,-467"/>
+<text text-anchor="start" x="47" y="-473.8" font-family="Times,serif" font-size="14.00">2</text>
+<polygon fill="none" stroke="black" points="85,-467 85,-488 188,-488 188,-467 85,-467"/>
+<text text-anchor="start" x="120.5" y="-473.8" font-family="Times,serif" font-size="14.00">13%</text>
+<polygon fill="none" stroke="black" points="19,-446 19,-467 85,-467 85,-446 19,-446"/>
+<text text-anchor="start" x="47" y="-452.8" font-family="Times,serif" font-size="14.00">3</text>
+<polygon fill="none" stroke="black" points="85,-446 85,-467 188,-467 188,-446 85,-446"/>
+<text text-anchor="start" x="120.5" y="-452.8" font-family="Times,serif" font-size="14.00">17%</text>
+</g>
+<!-- TransformerDecoder&#45;&gt;logitsTable -->
+<g id="edge2" class="edge">
+<title>TransformerDecoder&#45;&gt;logitsTable</title>
+<path fill="none" stroke="black" d="M103,-653.83C103,-646.79 103,-638.04 103,-628.47"/>
+<polygon fill="black" stroke="black" points="106.5,-628.25 103,-618.25 99.5,-628.25 106.5,-628.25"/>
+</g>
+<!-- OutlinesLogitsProcessor -->
+<g id="node4" class="node">
+<title>OutlinesLogitsProcessor</title>
+<polygon fill="lightblue" stroke="lightblue" points="206,-406 0,-406 0,-370 206,-370 206,-406"/>
+<text text-anchor="middle" x="103" y="-384.9" font-family="Helvetica,sans-Serif" font-size="12.00">Outlines Regex Logits Processor</text>
+</g>
+<!-- logitsTable&#45;&gt;OutlinesLogitsProcessor -->
+<g id="edge3" class="edge">
+<title>logitsTable&#45;&gt;OutlinesLogitsProcessor</title>
+<path fill="none" stroke="black" d="M103,-441.93C103,-432.71 103,-423.91 103,-416.22"/>
+<polygon fill="black" stroke="black" points="106.5,-416.17 103,-406.17 99.5,-416.17 106.5,-416.17"/>
+</g>
+<!-- logitsProcessorTable -->
+<g id="node5" class="node">
+<title>logitsProcessorTable</title>
+<polygon fill="none" stroke="black" points="30,-309 30,-330 177,-330 177,-309 30,-309"/>
+<text text-anchor="start" x="37" y="-316.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Processed Logits</text>
+<polygon fill="none" stroke="black" points="30,-288 30,-309 92,-309 92,-288 30,-288"/>
+<text text-anchor="start" x="39.5" y="-294.8" font-family="Times,serif" font-size="14.00">Token</text>
+<polygon fill="none" stroke="black" points="92,-288 92,-309 177,-309 177,-288 92,-288"/>
+<text text-anchor="start" x="95" y="-294.8" font-family="Times,serif" font-size="14.00">Probability</text>
+<polygon fill="none" stroke="black" points="30,-267 30,-288 92,-288 92,-267 30,-267"/>
+<text text-anchor="start" x="54.5" y="-273.8" font-family="Times,serif" font-size="14.00">+</text>
+<polygon fill="none" stroke="black" points="92,-267 92,-288 177,-288 177,-267 92,-267"/>
+<text text-anchor="start" x="123" y="-273.8" font-family="Times,serif" font-size="14.00">0%</text>
+<polygon fill="none" stroke="black" points="30,-246 30,-267 92,-267 92,-246 30,-246"/>
+<text text-anchor="start" x="41" y="-536.8" font-family="Times,serif" font-size="14.00">foo</text>
+<polygon fill="none" stroke="black" points="92,-246 92,-267 177,-267 177,-246 92,-246"/>
+<text text-anchor="start" x="123" y="-252.8" font-family="Times,serif" font-size="14.00">0%</text>
+<polygon fill="none" stroke="black" points="30,-225 30,-246 92,-246 92,-225 30,-225"/>
+<text text-anchor="start" x="58.5" y="-231.8" font-family="Times,serif" font-size="14.00">.</text>
+<polygon fill="none" stroke="black" points="92,-225 92,-246 177,-246 177,-225 92,-225"/>
+<text text-anchor="start" x="123" y="-231.8" font-family="Times,serif" font-size="14.00">7%</text>
+<polygon fill="none" stroke="black" points="30,-204 30,-225 92,-225 92,-204 30,-204"/>
+<text text-anchor="start" x="56" y="-210.8" font-family="Times,serif" font-size="14.00">1</text>
+<polygon fill="none" stroke="black" points="92,-204 92,-225 177,-225 177,-204 92,-204"/>
+<text text-anchor="start" x="118.5" y="-210.8" font-family="Times,serif" font-size="14.00">11%</text>
+<polygon fill="none" stroke="black" points="30,-183 30,-204 92,-204 92,-183 30,-183"/>
+<text text-anchor="start" x="56" y="-189.8" font-family="Times,serif" font-size="14.00">2</text>
+<polygon fill="none" stroke="black" points="92,-183 92,-204 177,-204 177,-183 92,-183"/>
+<text text-anchor="start" x="118.5" y="-189.8" font-family="Times,serif" font-size="14.00">13%</text>
+<polygon fill="none" stroke="black" points="30,-162 30,-183 92,-183 92,-162 30,-162"/>
+<text text-anchor="start" x="56" y="-168.8" font-family="Times,serif" font-size="14.00">3</text>
+<polygon fill="none" stroke="black" points="92,-162 92,-183 177,-183 177,-162 92,-162"/>
+<text text-anchor="start" x="118.5" y="-168.8" font-family="Times,serif" font-size="14.00">17%</text>
+</g>
+<!-- OutlinesLogitsProcessor&#45;&gt;logitsProcessorTable -->
+<g id="edge4" class="edge">
+<title>OutlinesLogitsProcessor&#45;&gt;logitsProcessorTable</title>
+<path fill="none" stroke="black" d="M103,-369.83C103,-362.79 103,-354.04 103,-344.47"/>
+<polygon fill="black" stroke="black" points="106.5,-344.25 103,-334.25 99.5,-344.25 106.5,-344.25"/>
+</g>
+<!-- sampler -->
+<g id="node6" class="node">
+<title>sampler</title>
+<polygon fill="lightblue" stroke="lightblue" points="136.5,-122 69.5,-122 69.5,-86 136.5,-86 136.5,-122"/>
+<text text-anchor="middle" x="103" y="-100.9" font-family="Helvetica,sans-Serif" font-size="12.00">Sampler</text>
+</g>
+<!-- logitsProcessorTable&#45;&gt;sampler -->
+<g id="edge5" class="edge">
+<title>logitsProcessorTable&#45;&gt;sampler</title>
+<path fill="none" stroke="black" d="M103,-157.93C103,-148.71 103,-139.91 103,-132.22"/>
+<polygon fill="black" stroke="black" points="106.5,-132.17 103,-122.17 99.5,-132.17 106.5,-132.17"/>
+</g>
+<!-- sampledTokenTable -->
+<g id="node7" class="node">
+<title>sampledTokenTable</title>
+<polygon fill="none" stroke="black" points="40,-25 40,-46 166,-46 166,-25 40,-25"/>
+<text text-anchor="start" x="43" y="-32.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Sampled Token</text>
+<polygon fill="none" stroke="black" points="40,-4 40,-25 119,-25 119,-4 40,-4"/>
+<text text-anchor="start" x="58" y="-10.8" font-family="Times,serif" font-size="14.00">Token</text>
+<polygon fill="none" stroke="black" points="119,-4 119,-25 166,-25 166,-4 119,-4"/>
+<text text-anchor="start" x="137.5" y="-10.8" font-family="Times,serif" font-size="14.00">3</text>
+</g>
+<!-- sampler&#45;&gt;sampledTokenTable -->
+<g id="edge6" class="edge">
+<title>sampler&#45;&gt;sampledTokenTable</title>
+<path fill="none" stroke="black" d="M103,-85.91C103,-78.33 103,-69.13 103,-60.23"/>
+<polygon fill="black" stroke="black" points="106.5,-60.05 103,-50.05 99.5,-60.05 106.5,-60.05"/>
+</g>
+</g>
+</svg>
diff --git a/docs_new/assets/images/logo-dark-mode.png b/docs_new/assets/images/logo-dark-mode.png
new file mode 100644
index 0000000000000000000000000000000000000000..24c5f3bc36cd271902376dcaebc7c667f9e909a4
GIT binary patch
literal 7425
zcmZvhcRX8v+sCP*cJ0ltTB}y6J^M3CQLFZficwNxC#0xq)d;G#7FD}NY&ApeQKQsM
z#HiX5lpw;B-oN|xyzb}q+<)XG=OpKxoa_6&KJU+UedEkb^ciS*XvxUP7#<ktT9A>E
z|0aEoy+%daGXKr}lJrL7XJ8jdM#irA=S7}pUXw~X$Q`6-8}tb55d?W0;7$gCK%~8V
zL4m->e(uuX0MB=O>O5p*d`S;<|F(p_-$ldT^I7HM_s7cfxLFLYM7j8Cpo~kXuNesF
zz7eZV$V{iyv#xVfH9t0gc@(2$4aG(2`#Et?vgKRZG7EW#$kJUI)=M#Bvz^s@Gwdp-
zu`HCG_tKB<&70wopj=|%r^z2bnzpj{HVY2Jk<LWJ8H7o0pohp^WYwcm6K;@H;SbCY
z|6<u+5^F<W`ZR)&?jReiy2#wVRCIF@N_Kkr%Q2ux;9C75q%UiGNPr??zV(qr?ZT$p
z%B`iO13;99=pq_o6#^ozj8axLFlA`HOjyw<Pk=PmxsCFnX6Id)`-3?~>^puH9QY)}
z#=`R^HyC4MV+S_G3rM^2<ibJB%qb!+x1DrMWi(IjR1hMEtzA*CC4Uv?E@Tla*IG}D
zcbr@n>FdwM-i+=l!YH(8t=eg!!jNmpz>&LiJ~Oucw$?9`PE9{(jkhvT9aOz`*cp31
z7&Xq?Il7Hu42cGvAkO#RNkl=e3PoNC8Kt??*XHdnIG`0%*fm055A%dUFc!MHy51G8
z+JJoHXPvKFhdFAG1y?k}F$<w1ADa)I`cHwc8ZT8Ab5&&^Bkx0I;wnG;jWq1<O=YV!
zapc&xGyuPRm4)=RK8ob0FG`KejPfdGyn~zF*_&$B=~t)Yvn<K=Tt%0wblr=rJ(P_m
z0-|yd@eq98qXMLmkdW2no7UJ^16DI_PG)AWE0j8W+nnEW^Zi{3l(cPh-A#u~brA01
zq+Ex-=9#Z_QAAdL5NEm%*LoMc0KCrZHt#wuDGh;x-s?w|)-Kf8_66S^eQQ<&e16<{
ze3Q}fjSUm(+79=%H76bFy2Y>=Y#=ib(I<EF`j*1<uQRMN!gd(B9+`L?u#_7$zRMBl
z^I^2>q;?9C5MD|8Z$F;UP&@XYe(L)5wn$w4$pB;IS!dq5qDK?)m3su2*g?-uUdQPg
zj|oJ0^sb*#6}M;iPTo{SZNH?+`?zoXs~+#Wpb+mSm+UAv*XCD~6a{$8_Y}|DJ668R
zEtgYetkuEJLZl_@sz7@MN&b1#7)DXvoxvzoXX+Ab<@>t2MM4bC|1n2Cyp$Fa`hL~^
z``~Dkc1`2>aeHMAjUXM>ic--*(wsW#_*=M0Pe(d>kpp)3QTcLWswhAJqgWq!BfDT1
z(VU(X$28Is0VIGl=6oP45|m04O+6ydLiGD0IxL+2j(*qZ!{9rX^DYd!TKC%YHpDp0
zb=yU~x*c=iX$^L?xX5}8A!O>%?d`B%T&O7qo>k;EH0kXUpcDB!=UH7<Pb~@40TD3M
zt@!&5GD6><NVz~{06lIc7@V8(=DLd8+GKR6ivojS=(j?hPBVr6YVRMoAhFgraLi0C
zv2sIb+XP|O|J$$A-eEKAqBpMVt4ho%CUVA~R^)aEKNpHC3-`LK2859X4`biO8dN?R
za3aNrM(jUx=)-oBMMG3Czo!F8GfPmxos;sD+Kkg%6Op9R1n#^f86&5u4+0A5LA9&$
zcLv4W9b!;*o)wOS0IB(1tl|53ojEJfI=RYdPkuk*A;RW%(~Z!9i*AuXyEIk)8OOV=
zrT7jpcRikmG)|v9DpgpND<_Bpz!7y9XXlZeaGuHRU%)%LiYsH!U=?fI_a3se0F`^}
zIqr*tVnib-%d*CuMC&sB3x9akKqw414jMn)fc(k(e~)R}xSp<V&hx4QBRh)09u_$I
zCRbfnbb|UzL>fqfvwR+NeN#Mmuj0?cy)FdPH#)7$#M~14E}u4DEJ>H?dgxhglQI$q
zDIKlw&oqDp^F_-tv$SM`q_k-CMH^I+MMLQfdb)jT|7|)2F*uRgw);6Rq@f$JoJqW{
zH7`7;INTdwd~-1FxmU0A*`13_<n_l#QR3!|;NV97EN3}U@I*E`#2D(2NjC+Tn=`rd
z@e0&N*JOp>FtppQZeAoZm$M58k-HcU`-%GQ4Em`D+;2Rtf|`08TPG(#@I))+3};j*
z_jDfb+6jDHog-=Y39s=L4TX(7cpw3>;XuJRz7Y{SW$(mtcBU6Nf43%sZp$3<L(;Sn
zg69|sV&vJ_a(t4gAq|No29neZAMO5qut%@6BWgUCBucpryA>pT9hlmrM;qUr&a=vI
z`3#h8`g^(3D3lMTo-yXB#PExGiBMpXe}z%VX2a?p4QVd*Q9g`k{i8pI8>GB{_(57{
zuWGV|@?Gc$fnb|4O{emgbYt8j-E%&`r+~=X)h}i>qV%JEjtv!`odhu<7x{_3zO+=X
zI?-VHPj|_5_LkP9q*#3i=opy<-Qmqu8M{TCg3q|7odk+2WRKhChR=Kv?u5|r2O_>#
zSkl$LKH7ILuL{sR;}pG^w(W2#G5D!J(Wt!FzUk&#W}%Yc^kWf?$D40grbkUO?N!U-
zZ6*vC=vL9;DMKZ#EO_8-YruoUZ72l1cy%<BbM#$(Vep1l;ZeE8(fOSY$)m}fDXZcj
zoAqU`)&R-*0Ql)hg?sy@EQiml;?!IW1<Vl5E2l09$T%~YS_yuzvA_2N6y>MQ*EFA5
zEKMyv=8FKJnmaqAyW|vCN8=D1;UObxN&?)|O_=JI73$)eAeXT~-l)Vb+pxi|Nk8<w
zV%p$N5|A;`Vyb1-xxuHc)A2(9KoV?qd&Q;H-gu3PdV<fg`5M>AJWqp7t~Er+_a&D1
zxdJmw{g~gKJCy-7UsK`Vo+dO*uvD6+Z#9LcLi#g;MP0p*FUu<63aNb`b6Gr?*Sdc{
zv2*w#))IsISLU#^{0BmUn`0I)%XNNnD+NL{wyH9Mn?JXU2H$U%qexIg0-ajx;2@}+
z2kN*@JLzyZ$j0?D_vkD8u12|pH_D0GD}dt{9un-x4d0!f{_X*)H>;%+J%R-Pbkc{w
z?UY*8g5Ar3_l0te-(US{IvX(Z+rH5}33xZ(RFqY8owAbX)K-cuVR;QVP|g0rHhYsT
z0wT2__HE~#iAV%nQlBIyZ!R@Vm=9s3oy@?{(2{hwkTk!(>1I#4mbzx;q61u9j%yn>
zUE+e|WGMpN;g837jxBUU`1$*?l3YGm5At<(UV+=avC&eeosusAD)S6;g6uunD`W+w
zuJQC!dgqKsjV)_Pd}Ka;@fEvYy?sF@X0X8%@Gr84DJR6@`dk)*S!b;eF3uVoGZ2rP
z?^w9^sXX)Zt6ouMU@i4UKoKpmx;FVY@YZT?{7>(b!k4p_ZKwxd^8j5*eiUQ}u#Gk9
z+TjcF|8RqjCw8j)hD5%=RVt%hK?9KxU4C|f(dyF=>0^pP1vgTh%t;CSn>Q~su2WlT
z6<x76rO~OeHYD%8Cd@o1Oz$v)_^dJDl1yD<DeZ;lvz^+XS>E8yZfqqo1-c)}JZRH`
zYEvxPs(y?ZsPjt}Ka8rmq60l}+6-ab0esJf_@D=?-9t>V;ps_End;9z`BPy4Mf=kS
zJ~;~?e^&|t*SUtf%;QyM*LyC6b3^HvcdN58UL<N1?Kl3B5tD;Yf8`phGxC`?7iEL^
zX)|sts@1LXcMh{Bhcxr;n0MJK9G=X(W*0?%v}yQ3x8Cs3i+`TExS}iXh0)^Mis{*g
zd-`!YA@SKhG~238)(gm#b5#5TQ8gsZ@pmzIDMujcJpZ2LDqq0^|37;2FOB(MUODM-
zGIKj{e`W2VchI*^pQ8$K5Sukzy=ZiB`SRBnG#tA5-sh3%thhr=JbGpJh0^0Haa+)9
z_>J>A_)~nN@qLnDxzTff@ehG){kETC$L?t;@@-=cpUrTMBJ_wqYP0$mT4Bi*?z#HT
zha1<&Ig(0K<J6v~de2$kJT;0-_Ez!#ZzT9PghYDdVhtX0?VzIER+@lQsB?SkMh0us
z`d>YMC>|Kyo#rY3O}CFLwY4c+8Ob4W`z^PAb7t_#)%6Bz0g_w#8HPWJZBrz~?r4iR
zoVXOK$+84yRED#PoxIC45ux?%raMi@3F>T!<nm!BX-#Ff&RvqhvewuVP~vlpKgW@>
zndk4Be{lFuQkM<CA-qvo0KWO8k6w+bJRkU*nK=|zmnD%{(`qB>`K^EoW>_2`%pEw=
zxuRRdei36&9U0JhG=>yNnh`F*B0qu|#_f_@NEhxigh8;v=0haad*JeuhCit;WA!5&
z+y`Vw!*(m^-yuQ2Cb_JJ14|Jg=wK<^URJiQa3Y*b;tv)qBo>H9Ro}@BH!KvQ=a7-+
z>#c_`6zj^vd*?lQr-Z3E#u8hzXW=d*vc6w)N<tlLW8t0|Hi^qM^On6gI(yzyy4Hje
z6I4s7bau0wE{a}CTh&cFyYNX5FsUk0lq!_yyTpahDIcTGv)(qx!V{sUSnzYq)W@nI
zxqUW2+j1fm%m_+_It;<zlovPo|Muq}dr10<EEb&Ad^SetP`}Fu_TmlW8-}~~|CD_%
z|HinDi{36eQO1aUO2vM&D#@xZD_Upyfq627Z&Kw;;aXzbk^q>!qD9`j&pF*h<U)n5
zon-sZTorzthc35qOI7OhdW>}df9A55C!8)LDadquqH}vQZYK)0SBHP}LK9C@LI!y9
zNH>rwv-EX&u;TOurSLuE4oV(ow?%hTvx6#aytj^I{m9L530Wp0s91#^kH=xoJ<}fm
zy$>axe5UR8T`;1qc;@{`O3m`C<z+gFZS#iqB8<IU9@dqr!WR0Ky%9<di}S#q!mgU~
zs==q+r&zV=4({SCwWs!pK>dlNIZ#jkv;j$awq701A7|ZBZxskm=2UD1P8w+W$H)HT
zE1YZiU$=0tCm)r;P<{*2RMOr-N~9v_Ug`>IoAZKX#3+xAY(;-dy5R3}1!|`iP7Wry
zD`4Ns?#BwlYJLfJcS&{cIo$MZkvfL+*Ba46W<N)1+_rX6DM3oq)PGWBV9)PW&#PD7
zOW$<U@5R<M{7U`<n7)sua)#56*No{XO}laX2K`NpPPrx_^hrft-9@&)zaL9iq~ewO
z`;3b~8zPejd>&tBvB$|#gY3!u0uHT6NJ%@&8Fbpx)y4+zb@kaM*}n=X1cUfm6C~%a
z?x$VPV_e&)@hN7G3%gmlNgY3@1vtEIX68vkQM?!pKcjpM1!%6+f5TEpvK2%RBY7P#
z#YJsN==TKHz#3*&)n{@;mvOG2-Z^X+-9J+k1f2c`v0b7&-d>|Ye<gQe<a)4R6^M3N
z)yvB@&wS}y^7LW$w^l)iKS{3>S=joQkkDJJ%$J&nIZ>6J-A82(q)^g&S-k&4=5d9f
z@|=d_8*neWK6nE`FIUWliAo@F*)&uZGy}II#H>|)15}Q75}rqUn-Q!N+H3xzExqpi
zrNkK?K4c%LQ$}o$O6@NS=T6S6zJ8%4%b@Hf$<yS=RINF6aWnXd_p4Xp1tA<{*a{to
z$6)&}!hBCbb?*|`t+g{hHQ;|cF*ypti&ObanL|H1?%DyZ8)dfJE6jH}9S$bm?R$H_
z8Q<OBMVFg$PUl18Weqp#GLO)VC;@mSKpxVut!%i_T1Kh~zjb~ic^vcp31tss7FA^o
z80&oZpxFLqJfyUPE$K*<RZ9u@c6_(*c(3CHhG-%G^IzrKzg6`gE&CG&24Lgkx$&}J
zm7pg1PrKhoM7W6f9#`7D7wa%_3t~*B{=#clxyni%kyP&e8BZyMwKPx(Xk4TylGSCL
z7QnsDsG^**RX@A&nbw{l)JLx;c4GfiRll?Io$h(0(ui+Lho`y9?BAJBv=9?=Ay9&C
z{iAxH9g)7JZTYN3t8^@OtE~EZ@~KZ-E55sMGYv}eoE^TB7#W8FzZ|VO(vaPf`)_dA
zk`tG1|I^XP0BxJ&p^yDDNfedx3DP2+@xg=!&8?W~)GzP{0Cd0+EH9dxRNYAap9`6l
zzH;A@u#K?KZ|SG*Z%I~cVIq;yjs$_a46R9~XH^2hq+-|L#=lRhT$BE`^~5FT@<PSs
zAq9b?z7WHWw+htITh+vfB~ykiBPmDP_e)45-^u;e79n!MD~r3xNRAVC(r>?8ri}|r
zP#H7$7`fB=p=AizrWNH#n}T!X8(x(YaX(XZSO%EJulDi+8HI(Lw>~Z<1N_EZQV9p8
z<d|!7awi&lM8fSh!0j}ah8u#ED;ESSxRZ81lN+I#DbZ*{$U>Ba1G$W8^2py9gpZrP
zZcZX;eBs5|QyJ!RS#+bwpE^A#MtZHDEaCaBJDgcppLVwH!5B_;N)Gou?syXEp~c2%
z!fCK4k#}*|-5kwW&k={AIj@cn3&T~3XCCk`MZGB7zL4nSdnr3fF`-ed2ziLg3dHTP
zofs`$o=k*Otc{UtE+ZqaeHo9u40MUPywHaNK*AXEqyr3w_M8h#I0R`aDw&07p3dxP
zU7^=|a&z1A!{?C;g-$);<8bnmFP9h00T_}E5f%;-cjv*umS)Y)`8!*m%aCJ)l%Ad4
z#wGvs=m(cRb0uPaUTh#K&u_UROUPLFHPzXSHe|>rjdtjAuI`Y(kHssbAAs)-A(S?_
zDl{b75+3iysWrjph9i|G!J-wDk>pPqdXlM+@fZyCc^4SNQ4YLPEsAq)V`3x=zd)MA
zh2R-`3zMc)A{-8*Of$ss*KrhW2rvC70M8G*-kQ3x29!JjnHE)|nIiy70g5vU!ou_6
z=ZD1I;7%|FA=9!<E9<)N*LODrdeb`-(oAL(h(x;EQ-KA5Dn~hL>~w9g-U`#>5j3>v
zxjO@J1g!?B9!v(4?HvQUt~QBWk5otOp?YeJh!G~YCy<lp#7G+Y)o<^|M1qveF6p9x
z&U8K4<6%A71+C++v};L6!{Mr~?Lrn5!78M7-TS5{FJ9GR5?**?L_IErj@$}c&GiZ7
z4$tdwd+ehp`>VLO#J?|&d!cw%tt2J-OJ=QzTYg-pG&0+Fmfmz8f4@fbqBg-Nf7_zi
z6SKfm4^7p{_)#@4dp(z5%!ZDu?fc47fTxxnmmbs#+VibHC39?|1GW$Y9Atm#u(gau
zM&S#@U<>nBoj>%3Aj?uafsucOFjn)NSp5|u?WgNPkP7hJT*@W)Bm`38tS*VGKLG-9
zD{Kdm9%M=@!h3$0Izi1v>?MqU@ihZ;d11?~Akju(<cz8jDZ!NkWETZbJf<m89P#Mn
z%scs4MOKM|XCltE?flrMAzH|%EXoLjb~*|ypp$nV;l{Wadymv{oYAfsGuJE_le8GC
z)7v`bYJR!6A{Z!jT6p~^?JHr)pK384lDQNP9TW-DxtQ6fIIubU+<-`N2$k7pSt>W%
z>Hf)b=6ZQVM-t5~6P-PaAtIZ|Va56&QZwTAVgrhPaSk@k5|2>E$8{2lbp;O?RB2CM
z3+=zdRqfvoZ`%cs=S#5w)@A}{uAN!a$t*uiqBBufX9V}GBE^l1@Z`hWv-x$njHe78
z@o4=B+Qsv0%-ledn;WWRKc>#yK_#NDA1nME?ej>{x=O!z-9JPqt5*%i8&^7jlzxA7
zdoi1k5qnfYl_$;;If+<5Vm_)0xex$%d{t~?P6>_Tgl5vt$11W$$e-bUnO_AEmYl+9
zIu@@e&NRyiS?D^&peliq-!yxE?X{JlDsF3Jxic(YMWD}jLoP84P@QhU!=}uohiyZ&
zCNTt!MZc!BjD3K_KFcox1;Lt(Y4Xe53WV&nQkf4kO}UYFJ{pa`nhxfDyI=Xe;t&oH
zfdPVXUD7jej5bP_Ney)k5S_^HC0tt?glZ8lpEaHe-wEh7*nO~9MTPry^JO?Ec7#|V
zyO|QZQ$!dqPD9^Jo4bkAR2~WATns*@Aj*|4J27d#MO_hf7|5@YT0?95);^xjg%<K%
z%HI#qdZ9#K`(AJreS_4$lItV2kPNdO9$%9G6eTi5;iVVhxQZQ(&91t`bY0|VyTm&<
z1abZvC$dQy&ZRt$!C0?H{3c_fY8>k_LM_kqsp55X4b8MOEi@e;UzSJVxoN@KXWY%8
z*2pMG2zMS@lLvcKsag67)#6i}#?9$Nfp`G*Vn4zYBSh%?z6N~|Ahf7~VY=$P#Lx{c
zIv4}hd%V@s#<46~v@g%V_V2suy1_|>pP>pj7LMB&A{<}sV4l6;$K5X*x!?A9#dIay
z@0@&*n-5wyB<L2j*Q+*sxnOo)2%W(dwlITd&t96HheL_rG&?AMdyK9}ppVe}93mKo
zI}1Hl**4D{ugqM7N({vfa6KYDf<2C9UVjj9&r`%%h~Z*F`bc>MG{8_7seb)ZrFTYH
zrj%2e6|6FWKdmVo5S*truq-RZAl6xbj(W4|ZPDZ<GOnUEryUCR;=cG+3G|<e4vnn<
z;1De(Xwv}kva<uSA{F@VNbdkkXoLd#Vds%r3HXBBhT2+L=Lr1MHbf2oOuO><CE{(X
z50<u#om}RqNh?9=fThtUB`X$AT5Aw@S97+2(SxCQ?zZI}rX?w%)P}V02?*3{$Yp^3
zU%##>!9Qs6;@*zK28Dl9M2%V6jfo)h7z1d+@4pI@V-q=dwbwavG(COU@f?+N*py<l
z0Wnj4y-Y=m#ddID$;CX+7E*EdEi)c{;c*_XFw;Ul&m6~hW!=59Pd6a^djY2z<i>gy
z#}<i8h?S2exoq)#aw?p4m>M}2zT<X~8mhGBdDnr~vle5t|2kLqNTm&iF+#(_`4>md
z2u3?&s-$4`(yq1OL`v8>6Dm9XL%OptmSv*be+|hzr1KMNetF@J%rE8u*)3y{bXd>e
ztIZ1JFGi1s(_uwE!bGSJaZMm}{(<kHBQv*-lr0SC+>=!}pwYryk*oP;X9Q-s()cht
zGCy#A9tUI=c5|}nrv~=XF#a7Q3Sa>GY^YXpmqCz*v&I(`h8GkF<oYrit3I5}wCohv
z5&zw&o9b1*e@BpI`-oJ1z|Gzn)f1Vvx@p3I#zFmT4q-0L_r|5bS@pUY(l3FI-gTGA
z#hTXukPKy9PE<s3#b4gQT_^^m?uSKn#qYunxoSNpli9^HiDBSkzppMKDMb%$-#L(v
zXZZ-vJL)WVKdn_<Nj#^-Q*%BJgQIRoo2HfN=-%(NeHSTyH`f?-@0C7)ajt-y`dGbm
z;n8gaQC3rn2;oYp{XNUgI^@2)++z3uZ@Djfr$iLu1dHixr2^bGC$&wlZSu6}He_$j
zY-qyIsq-cE1|4;aq)COwwB{=%LMYEE>qg@;!(>9N0l!#*Q6+y;(;js8l+D?clR1Tl
z+guy0P)Xc$D_#hH<#b5#2x=Rl;YZ^aOmQY8R6ES&^>re3swfS-DrKX}k99sivP8zh
zV`DA7Y}BLE&K6z>v<V5xmDE?0jv26uDeFKS*S%+EF&P>|vyok`qIN^Fj&RU)#=H-|
zgKrlL$zcxAa%9AjdT~#CU{rN!%U?!c*bUF9Pm#LEu?C0#cU|db&YG`erd@ZDJ)~sr
tDX<JV5q_OH#i^S#3rV(f;N(E7$@ulK__9wEY59`ufu4zOmA1=^{{gxFrcnR@

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/logo-dark-mode.svg b/docs_new/assets/images/logo-dark-mode.svg
new file mode 100644
index 000000000..004bab98d
--- /dev/null
+++ b/docs_new/assets/images/logo-dark-mode.svg
@@ -0,0 +1,207 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="348"
+   height="95"
+   viewBox="0 0 348 95"
+   fill="none"
+   version="1.1"
+   id="svg39"
+   sodipodi:docname="logo-dark-mode.svg"
+   inkscape:export-filename="../projects/outlines/docs/assets/images/logo-light.png"
+   inkscape:export-xdpi="96"
+   inkscape:export-ydpi="96"
+   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg">
+  <sodipodi:namedview
+     id="namedview39"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     inkscape:zoom="0.27368421"
+     inkscape:cx="283.17308"
+     inkscape:cy="45.673077"
+     inkscape:window-width="1416"
+     inkscape:window-height="818"
+     inkscape:window-x="30"
+     inkscape:window-y="38"
+     inkscape:window-maximized="0"
+     inkscape:current-layer="svg39" />
+  <g
+     clip-path="url(#clip0_56_2)"
+     id="g39">
+    <path
+       d="M101.92 79.3501C101.92 83.0601 99.78 85.3201 96.88 85.3201C95.13 85.3201 94.03 84.4601 93.43 83.6801H93.39V84.9801H91.07V69.1801H93.39V75.0601H93.43C94.03 74.2001 95.13 73.3801 96.88 73.3801C99.77 73.3801 101.92 75.6301 101.92 79.3501ZM99.6 79.3501C99.6 77.1401 98.52 75.2801 96.48 75.2801C94.58 75.2801 93.41 76.8301 93.41 79.3501C93.41 81.8701 94.58 83.4201 96.48 83.4201C98.51 83.4201 99.6 81.5601 99.6 79.3501Z"
+       fill="white"
+       id="path1" />
+    <path
+       d="M108.52 85.69C107.88 87.44 107.15 88.83 104.98 88.83C104.32 88.83 103.76 88.72 103.32 88.61V86.84H103.52C104.49 86.97 105.22 86.91 105.73 86.29C106.15 85.78 106.46 84.9 106.04 83.84L101.95 73.7H104.45L107.61 82.34H107.65L110.55 73.7H113.02L108.51 85.7L108.52 85.69Z"
+       fill="white"
+       id="path2" />
+    <path
+       d="M121.12 84.98V83.84H119.98V82.7H118.84V75.82H119.98V74.68H121.12V73.54H125.7V74.68H126.85V68.95H129.14V84.99H126.85V83.85H125.7V84.99H121.12V84.98ZM125.7 82.7V81.56H126.85V76.97H125.7V75.83H122.26V76.97H121.12V81.56H122.26V82.7H125.7Z"
+       fill="white"
+       id="path3" />
+    <path
+       d="M133.7 84.98V83.84H132.56V82.7H131.42V75.82H132.56V74.68H133.7V73.54H139.43V74.68H140.57V75.82H141.71V82.7H140.57V83.84H139.43V84.98H133.7ZM138.29 82.7V81.56H139.44V76.97H138.29V75.83H134.85V76.97H133.71V81.56H134.85V82.7H138.29Z"
+       fill="white"
+       id="path4" />
+    <path
+       d="M146.36 84.98V83.84H145.22V75.25H143.5V73.53H145.22V70.09H147.51V73.53H149.8V75.25H147.51V82.7H149.8V84.99H146.36V84.98Z"
+       fill="#FFFDFD"
+       id="path5" />
+    <path
+       d="M154.34 84.98V83.84H153.2V75.25H151.48V73.53H153.2V70.09H155.49V73.53H157.78V75.25H155.49V82.7H157.78V84.99H154.34V84.98Z"
+       fill="white"
+       id="path6" />
+    <path
+       d="M159.47 84.98V83.26H160.61V82.12H161.75V80.97H162.89V79.83H164.04V78.69H162.89V77.55H161.75V76.4H160.61V75.26H159.47V73.54H161.76V74.68H162.9V75.82H164.05V76.96H166.34V75.82H167.49V74.68H168.63V73.54H170.92V75.26H169.78V76.4H168.64V77.55H167.5V78.69H166.35V79.83H167.5V80.97H168.64V82.12H169.78V83.26H170.92V84.98H168.63V83.84H167.49V82.7H166.34V81.56H164.05V82.7H162.9V83.84H161.76V84.98H159.47Z"
+       fill="white"
+       id="path7" />
+    <path
+       d="M175.45 84.98V83.84H174.31V75.25H172.59V73.53H174.31V70.09H176.6V73.53H178.89V75.25H176.6V82.7H178.89V84.99H175.45V84.98Z"
+       fill="white"
+       id="path8" />
+    <path
+       d="M180.83 84.9799V82.6899H183.12V84.9799H180.83Z"
+       fill="#FFFEFE"
+       id="path9" />
+    <path
+       d="M187.59 84.9801V83.8401H186.45V82.7001H185.31V80.4101H186.45V79.2701H188.74V78.1301H193.33V76.9801H192.18V75.8401H188.74V76.9801H185.31V75.8401H186.45V74.7001H187.59V73.5601H193.32V74.7001H194.46V75.8401H195.6V85.0001H193.31V83.8601H192.16V85.0001H187.58L187.59 84.9801ZM192.17 82.7001V81.5601H193.32V79.2701H192.17V80.4101H188.16V82.7001H192.17Z"
+       fill="white"
+       id="path10" />
+    <path
+       d="M197.96 71.81V69.52H200.25V71.81H197.96ZM197.96 84.98V73.53H200.25V84.98H197.96Z"
+       fill="white"
+       id="path11" />
+    <path
+       d="M15.18 39.46V42.5H54.64V39.46H15.18Z"
+       fill="#E35A26"
+       id="path12" />
+    <path
+       d="M39.46 30.3501V27.3101H15.18V30.3501H39.46Z"
+       fill="#E35A26"
+       id="path13" />
+    <path
+       d="M48.57 27.3199H45.53V30.3599H48.57V27.3199Z"
+       fill="#E35A26"
+       id="path14" />
+    <path
+       d="M51.6 24.28H48.56V27.32H51.6V24.28Z"
+       fill="#E35A26"
+       id="path15" />
+    <path
+       d="M54.64 27.3199H51.6V30.3599H54.64V27.3199Z"
+       fill="#E35A26"
+       id="path16" />
+    <path
+       d="M48.57 21.25H45.53V24.29H48.57V21.25Z"
+       fill="#E35A26"
+       id="path17" />
+    <path
+       d="M54.64 21.25H51.6V24.29H54.64V21.25Z"
+       fill="#E35A26"
+       id="path18" />
+    <path
+       d="M15.18 51.6V54.64H54.64V51.6H15.18Z"
+       fill="#E35A26"
+       id="path19" />
+    <path
+       d="M39.46 24.28V21.24H15.18V24.28H39.46Z"
+       fill="#E35A26"
+       id="path20" />
+    <path
+       d="M15.18 63.74V66.78H54.64V63.74H15.18Z"
+       fill="#E35A26"
+       id="path21" />
+    <path
+       d="M15.18 75.88V78.92H54.64V75.88H15.18Z"
+       fill="#E35A26"
+       id="path22" />
+    <path
+       d="M15.18 6.07H21.25V3.03H18.21V0H12.14V3.04H15.18V6.08V6.07Z"
+       fill="#E35A26"
+       id="path23" />
+    <path
+       d="M24.28 6.07H30.35V3.03H27.31V0H21.24V3.04H24.28V6.08V6.07Z"
+       fill="#E35A26"
+       id="path24" />
+    <path
+       d="M33.39 6.07H36.43V9.11H45.54V6.07H48.58V3.03H45.54V0H39.47V3.04H42.51V6.08H39.47V3.04H36.43V0H30.36V3.04H33.4V6.08L33.39 6.07Z"
+       fill="#E35A26"
+       id="path25" />
+    <path
+       d="M51.6 6.07H48.56V9.11H54.63V6.07H57.67V3.03H54.63V0H48.56V3.04H51.6V6.08V6.07Z"
+       fill="#E35A26"
+       id="path26" />
+    <path
+       d="M60.71 0H57.67V3.04H60.71V0Z"
+       fill="#E35A26"
+       id="path27" />
+    <path
+       d="M15.18 6.06995H12.14V9.10995H15.18V6.06995Z"
+       fill="#E35A26"
+       id="path28" />
+    <path
+       d="M24.29 6.06995H21.25V9.10995H24.29V6.06995Z"
+       fill="#E35A26"
+       id="path29" />
+    <path
+       d="M33.39 6.06995H30.35V9.10995H33.39V6.06995Z"
+       fill="#E35A26"
+       id="path30" />
+    <path
+       d="M66.77 9.11003V6.07003H63.73V3.03003H60.69V6.07003H57.65V9.11003H63.72V84.99H60.68V88.03H9.11V84.99H6.07V9.11003H9.11V6.07003H12.15V3.03003H6.08V6.07003H3.04V9.11003H0V84.99H3.04V91.06H9.11V94.1H60.71V91.06H66.78V84.99H69.82V9.11003H66.78H66.77Z"
+       fill="#E35A26"
+       id="path31" />
+    <path
+       d="M102.09 60.7V57.02H98.41V53.34H94.73V49.66H91.05V20.16H94.73V16.48H98.41V12.8H102.09V9.09998H124.21V12.8H127.89V16.48H131.59V20.16H135.27V49.66H131.59V53.34H127.89V57.02H124.21V60.7H102.09ZM120.53 53.34V49.66H124.21V45.96H127.89V23.87H124.21V20.17H120.53V16.49H105.79V20.17H102.09V23.87H98.41V45.96H102.09V49.66H105.79V53.34H120.53Z"
+       fill="#E35A26"
+       id="path32" />
+    <path
+       d="M149.89 60.7V57.02H146.21V53.34H142.53V23.87H149.89V49.67H153.57V53.35H162.79V49.67H166.47V23.87H173.85V60.7H166.47V57.02H162.79V60.7H149.89Z"
+       fill="#E35A26"
+       id="path33" />
+    <path
+       d="M189.18 60.7001V57.0201H185.5V29.38H179.98V23.86H185.5V12.8H192.88V23.86H200.24V29.38H192.88V53.34H200.24V60.7001H189.18Z"
+       fill="#E35A26"
+       id="path34" />
+    <path
+       d="M206.38 60.7001V12.8101H213.74V60.7001H206.38Z"
+       fill="#E35A26"
+       id="path35" />
+    <path
+       d="M221.2 20.1701V12.8101H228.56V20.1701H221.2ZM221.2 60.7001V23.8701H228.56V60.7001H221.2Z"
+       fill="#E35A26"
+       id="path36" />
+    <path
+       d="M236 60.7V23.87H243.36V27.55H247.04V23.87H259.94V27.55H263.64V31.23H267.32V60.7H259.94V34.9H256.26V31.22H247.04V34.9H243.36V60.7H236Z"
+       fill="#E35A26"
+       id="path37" />
+    <path
+       d="M281.73 60.7V57.02H278.05V53.34H274.37V31.22H278.05V27.54H281.73V23.86H300.17V27.54H303.85V31.22H307.53V44.12H281.73V49.66H285.41V53.34H296.47V49.66H307.53V53.34H303.85V57.02H300.17V60.7H281.73ZM300.18 38.61V34.91H296.48V31.23H285.42V34.91H281.74V38.61H300.18Z"
+       fill="#E35A26"
+       id="path38" />
+    <path
+       d="M321.81 60.7V57.02H318.13V53.34H314.45V49.66H325.48V53.34H338.38V45.96H325.48V42.28H318.12V38.6H314.44V31.22H318.12V27.54H321.8V23.86H340.24V27.54H343.92V31.22H347.6V34.9H336.54V31.22H323.64V38.6H336.54V42.28H343.92V45.96H347.6V53.34H343.92V57.02H340.24V60.7H321.8H321.81Z"
+       fill="#E35A26"
+       id="path39" />
+  </g>
+  <defs
+     id="defs39">
+    <clipPath
+       id="clip0_56_2">
+      <rect
+         width="347.6"
+         height="94.09"
+         fill="white"
+         id="rect39" />
+    </clipPath>
+  </defs>
+</svg>
diff --git a/docs_new/assets/images/logo-light-mode.png b/docs_new/assets/images/logo-light-mode.png
new file mode 100644
index 0000000000000000000000000000000000000000..ac04c20f52f01fcbdca2913c22df1e0e9f63d13a
GIT binary patch
literal 9174
zcmb_?2UL?=({2bL2q+ec0zyEl^e!Di=|w^59YPU8?>%~yqcn*adQqAP1PQ%IJ(N%d
zga|>vgMf7DC6xO{Dc}FEyVkeXT^AO~o5}3iduGo(GqX3(?&_(Zqhg{0fk5XpHEtV#
zK;RnSxsmb=@M?lMUkSXC_!y|GfKWXwv%rVHpeCA*y1Jknz%wO?oa7RSl*j`7f=HM^
z<iDRmAT1K+lV<}G-Xk6`2o&W4B0J(S1Kx>0FMvN_^wB$MHp!{PZ16ERxF(zQ_!-nl
z95yV<pbxyA_0%x+0f88}h<_xS2AsbDX{IjsO?*vsb!6>5U_!PI9(GWnK$s_S6i7Z$
z7I=g~eQhCuFgJG}*+2!3-yE{QGm$LJ0r}11>#D$EqI(yj>fsHANC}AuiEt=VK_C!$
zZwE(NgWGCH!+}o<9L~PJp0dKi0RaI*0pda)-cG`zGBPs4B4WZ~VuApNpihvyuWg{9
zyU+DMLXPF!hWglhyLkG#c(_A|a&7HA{CpKSIEWLS{QOy`uZ!d9Ozu8MY5@ug6TcA_
z6%r9Xkqr!$CsJkgy<MQd%tZN$qVm5v{|D`89(iHn<o~glKbrog0;(!f$qS#DO_A!V
z)prUIi2bSN?OXQ)Nfy5;_>8KhV!UkR;o{;A7>igizN2eFMJ3CI-o4Bs4=W@uTOX6N
zAm;Nan!$VuDwC*0^7_}Cv~CA}5jU9U2Xf!(NV26M4Ggtpy=Y)W%1*!o*E9QLN-w_T
z^yv(21?w$02Qvg}dRw}tw*0%ZQtffQyq8ekYgk^qJ|fe@DCQq*HnR1p!??{@$z4%k
zRo*lnX_-8=<a*e>GDr(IwT@uveYvq6-^y3+Z<1{984ChaLLz~eE=wEHt#j|g$u1~^
zNdCM$2b0y@IhT2z`1Kzit4L75X7!76zd!wcp(CgCjAoKoKOS=-xe-LNwBW(cd?Mxo
zFO~c*XLwovW&@VzKw*chSq7Pv&*PFBZjVf|Bs3lR%@OeGu!oub5c|EPn2m2aUP}n=
z-mRw4`%5=88k<Ad1be#Y)`d(7O<Vl}0ak>uh>(MSmAZDh)uiSLUY2U<C>ylTu_EG(
zCg;n-=M${06<)rqo6FeE`%vOnT;A&nQw6sjDjeWx6NYE^hZ&{OyyaiR775aE8)Qp*
z$kZ=~i`yPZ`#z>C!$UdZ9ZEU^mks|qR(v4~<Z<t7b4W+rv31~&86X&ipZMPCi${vW
z_b)>*-~KC)XXYD`c9*lr0eY=x*LX;7>Kl8WLvkPTWpNRV%j?Pq=)2%+uDC*KmtCHn
z34HTZ_YDQJ1*VL;iPV?@dJD0{qus##LoTau=fmdNgDox|lCL94LA82d&kWQAleV7r
zcDKKoic<l94foRA7xJs)j>e9QW|yY^Y$7BQ({%P1!;`T3oCS)boxX6>lLS0UdI+{5
z{0Gy8i6#6ssL}}qPJLm5b(;}{KiTa~U7MZGVJFvd^I|p+0ZRp;df^?@1Db1o@y|Ux
z8Ut&8St-18HnO><KJ!-GVvg(e@=jllVsy)-n9C2hL+8j<@Kv6ws?R|$|1o<|@c#A2
z6Ti6nAFJPu86!;QL$Esq#>-V~NY<V~r5;5&&Bm6gbpOzwTW)$=&nr2IKIwaX0;hDo
zE5avqL-lj_dW=|A<mh3=5Yzd+Z4sG5ju#Oz7C15H`W5(idD%`18&OxAK$5XHA`LR)
zzXkmE;B{W*Nc?pn)l)a85mt_r`+wkhzu@KeRLuqnrd6NW;m7|wQ138Bs7jqhVpDP6
zW@<K;uUYN|>@DKk&D6galc6swK2nQj{WYJxKyby4c)Yx`y<T6vg7UB0UHAuL{=mg_
zE^mbmHTEPg_U&yD9DDY{gx(VJeRdf+YtIuseb{`mxo1(x3(4k+#i^!+?|W12rur?R
zWC+d|Emjg{zap(EpySr{LroUJ2^G%*Dt*34qZ2lo5>`~&lIgBTVs4WMMt{8RLDQ*n
z61vXZfF-(Ot&TSw1inlU!DRf5wKY8@b(2wgrq%XQAA63n%^LtFE;ot_oRYFrKqT(s
z(!uN}$f0vT(*7fNu)T><GuWHkaW3<9ZpXLTrq-?fQp54_CFdcR%1j^gU^<f>H%7bf
z@`|(Db8RUZyXH-%4}ZtkdDTeJmdKKLJ&T6&WBls~uSaW1H)$%I@^AfKz0E^#!yVg)
z0e@|rB2&dB2nN=o?sNiZ@Is;q7vtWuo|24!wU=!Fj5!W;h=?Q*2!-D+4n000eb$OJ
zq%&LT3wj4`P7kZvY-|~6Y?@YwW0OlQCdbBt%ZgCdt1TR>h+Mb88BrT(M#U0<F9}7l
z%Um;AR*lT-8WK1&UiIupk!ysiVvjj4FH;3SN{EOqFq6v|qtT!NW^3=&OkROtRGEKL
zxvlij^8~7Ldn^ZT9Xc~wx4Tz93QHDJl-+fb%VZaU!3)F2^~|FO*+nAYec;iogLSt`
z3(k3&%lTGJhu`NcP5fwyDWaVp1|J@i)JwUjRIGakb%Ih9#6eU2=0VSJ+s%3fo}^RF
znDKVmj~%W-l<sVc!;(Gcb9ASIT>d`RpXCLJO-wF3t=s5ChUx_X`~Ir$Vm+ad0>)PO
zUp!Y96>6d)I{~JNGR%Gy&xXSGIRxvcO~A{&9-}h+u6pnLO=~S9UM)-G9aDpg2(%xh
zeumfY!3YvD9!$R1Dzpdocd*}d2PS{bjVfKvS#WLau9=pH;-Id)^U_9^10ka`!R3bT
zyXkMr%L;ztre`{xNl*u$HP{q$2Zy__hG*vvUCj<QQ`v-z`4k@VsLQ~*K1%2w-ifR4
zr$r|`r4OvsFSy@sm2TVnNv3)!2eWI_tfY|lokAxmnjrqSbXalGT(T;vxyM5HvH7>^
z^)No3hKAyYcipNR_P1*-)LYP+iu#Co?)i>`OsdrN5}(7amCmLie2i<4p}%F-Q*?OF
zpkzavLMd)%t$HSCo>u8rvy_sx8ntJ6#6^mYWlU&@gZlF))929>n=7+at7t7#PrJBJ
zl1IrF5*azO!fMe(b)0fY$Tqcr{9EeIllI5!iLMX9Ou3!}u(C}l5S?3d{^+ZxF5ai;
z;2+V2<uL8tVfa}{IZV-Z12%a;85d=ZbFDhOuh_P?T|uU_|JA?lk^fkM{!ZEzixVh>
zkpZhs3O?`nDd*9MTTN%vaDuh%lqlOHym|qbvu6kZyf$$@Y!?c8a$r)9A%DDVQ*A=U
zYU2G~nCQ;~Eu6?YpAzbF)EW9hnUrD~!CA4N35mneZaA*_6suP(mfTNKey?^{7|0o!
zPx2E8Tuukz&J@hP?-X~0NWd0JH2;H#=wRhv?1)U5Nbqo!ru2h+x1EE3`nS*zQnI9o
z<V3PS@&K)X$Cdwq3k+8l=|?9$#?{cl%^T+(C6qIa?S0LooF#JbSc4;v20$%`yyo(T
z%@OkHSWXxRg*oH9IR*yuCmlSZ%lw;l8NLrJkF^eUAU<06D4G*8o6*uMq`mxdew}0d
z{QIU||MoM)-WYi08`A%_Xof=QQ{BwM%vw@<^>SVSwxe_d{bnlkAoNWSD`90|>f(gI
z|H~_szfl#=PPSLFxUkMGt{pje(=KsG$ik<?bSvRY3o6*SHeI+G3EBK&HSSnw#p}pS
zsOi?mH*W2RXALzDDLN^QKU!-DJG<;kUcF15$C}lfr9UueZyjEzIQAGnXlciC>KK}o
z037}+^`7#SgaX{B9^vr+h3O`DF~4T(A5YDegWn1dL+^n(X1?EB?rr*eRNddNy01sg
zOd!*#8B$oz9&qmdCUrVm+@&-i)~kI38V<7zB(E-k>>oo62^lvRkd_PaAM8#8Vub3#
zL=}Cm(n-i*asdr^{Nzk%PDsk;0M1rDT0bdTE+i)cC2FoaMf7PHIrC#gZDHiI@kgjP
zNj=Jy<Xth`y-kfZ@=Rh0>QVg~Y&K<gF-X(yn!Ik~tK|^kl2^!&QTG1|&}T4k;U|j%
zW-`vFoRcz(Bn$-mRR9o?8?SrqW3p=2MgV30!o}a!M=Q8#D=!@{mvN5f+Gs~l`Bw;<
z-C<_c=)7l?3~6n%^raYvi7|?5&5tDQRiz&vLa&F;-#5Qfa;z|1lzf*uqivMl=OjWY
zS&9K=K8yuR#T^6oLIFStA2arUa!mQZ@nD?U#PIrhr%XEf82u%81CN#b;g0@S>mmKc
z9;T(f8wU$(x~9}l87$9?{NH?WCRtistu9H={WV(adE<CpBxJG|x7X*s8J(d@r9qOv
zX4KtX`k8afr`(IJhY)Kq8P71Gc+OrSX1G=a_lWOwBb2jsBDW@xvv;G8R|*l4gZEi|
z$jG3hR{H~Ma4c1O>6`tcy&EHeXzJp!_g}O(?bsx%MuQB&qdl5eFB}yjPZ@qGnf(Q2
zlVu84=#oYO_4*ZD*6^wK$TCdOkH+1)>V5!T9k(djvR0tBkW0bylR-mPmJru_0~Qm#
z{{%ul-~0LHhs4}0Z6x_7vLR|dFNsrF8!1xyLPQgl5t4$_OUGw}iSuSF!6%E@q{Zr{
ztp*6?!W_NY`b9cuVD#9VyLS?Z2y~=a`9%QGx=ZX{pOPK`o?67)cjjpQKy4}o6eH!e
z4+Q8=NZHSULW-Vi<U|f-ml}`)q^h4kIW1uCUz|vOe?zMxB@1JB5qr-XI^V=wBv2G`
zv-O{`k6VZ}{n9bmV&#6~f%oR{$QFa^*3ZKFhXLLE#xL>sd*NKuW_zTBRK()SV7*U>
zWn(`&L_qy(@m#E$E7^rBw<m;Z6PLd+<1JQ;JU%H|G*ZL-U>*L4dobDH@+>DqbNzvD
z0e$hHYU(?!$?3|QomV(;AIGx7#%^K+<#L2ALw$dwmX~^uuNgyKABubUZ`!7@*DVol
zZ{8S2*6q<2KAaI5F}igXOc}#D5n|oZOELdr)kh3%R3WgiJmc|a2JJDUZ)s<%QV~6;
z#-m*40czDO!Rx1i5CahzF{eB5+8oxL6BJVax@M&l1mvgpjLf!|j|VEZ5n+96`GF2%
zAa3iE=P?}wrVzrI)1BX=j$AyENZ=M3;jbZ_!G=$BPc(a#-c(4a(;wYfN?jL9urF4m
z9TnVty8=}t@I>)<ZI;jMcs=pYkko9HK-zbA>@gqo>eXE$=sHm=yoK*&7k-1GDYL@2
zq!f9Dr0_Y)=nZ+CtH+l5x`K;T&|2xr>dx-&PSN=d!`WV(gu?2Qe(GB#)ZR7m&%HRa
zHWB2({+&qt55)neO)zxXHvA@O<*?|zrsYN!Rvwr*l$Rk*iuB83TE?BLb-S9`gTAV}
zjN?ySEjl{MmYT=(JI!k5klMnCx~~uPWx#%99X;rtgNsAILK=2(fjhect8=R(JrM`U
z{m`((!E)uEsAeH+MET~kFA^02al=(_zVvSN=2pGz|J^_x1?4?5wGeXS2<m?C?~P1Q
zoX+!;0DfOhH1Tjql~f3SXX`0Ks9l^8q{?|O&UXsT)_}i<Va0?_C>UOV^1pd<8Y!wv
z&xs0wVdSC0GR=p&AxQY1go5V4I<xkmLv-l)q3lN3e9H4(+g_|-Xj$Nr#ek*v@E(i$
z%-i@I9bN0ilU{L;Ln^vxsn+F37TQs?9UFFW*46P+w?2HJP7I43pOjDn7b)QpSx%~l
zdySR{RhDDn3kN&0x0=yeE%kepo&*Bg@ka_(Xz)f46V06e50+ap%XNdbQICs&5(wz$
zoPRxI)-$OxlF8Z5ZxOD1+wewATqault^DW6bLF&4tyO$8+kUQi3TTwz-Azf9*lMGU
zOg0*rPI%MhN@s5`!KbLiP`|Hq!Kuve@)H#RJ1`TcX2R?NM=UcDw8JDPc2d3-3963#
zV0fCicYxNs=a3amdP;f(KnKa=f>Re?W1OISvNNvgFOnU}s*Sb?EIpHs-nd^MXW*jn
zA(AxL>e6l|+lB(PaG03p`rXcs1pM`71_vvA1ucaw*?y~ynPm<e?+CWHPoGN?-!tQL
zcmF5kfrVPeOze_Z#@Wq=#UFbkp}hoefq&H8zuwtMn{JkA+GT$`mWecEctNmU?&6mj
z@X%_@)EP!r9<_#;o5aSACBX5{NfG-~6Ht!F<moBpiaLOK|CPt4!jLX}cc6UGJE6F3
zcyrH5p=sAk{Lu3KiA}OYj(Z=?i~;7l@x3nf(XP;affL96c_*@(cl~2W_NmeAfu1pr
ztGD5l(g>7IoBx$3W|Li9skR<CwMet_Z}|^_6#ODUi7-1@U-FiMg1@)*G(!awBc$=Y
z81KnGV@!Z}u*ZrEpHP&A!8wh0UMGDM0l4nd-%NO_t~S|TGkxiI%7B{!^-#}hRoPD}
zu&YZ>ZL&)3XPwms?OQh`A)%A0+JBtGkq>}6#~{3kFFE{GMIQ*2Tp0tLbaj(U^-D^6
zLvWk`n~ec5^UJzx0%w4OE)mj#xA|)L1IYr{(NT*6tCzo{S{A6$7cabgeH!Aej1bJV
zL)Q1;Qztx6zWu^9Dw&2@iY(MB3lPL3NT<3pms1wmBQZ39V><;7n1})Di6E4D*B){c
z-Qpz!V277Y3IgX(^B@XTIx$3_v)^<!x%n3adF~G<uqhZFa-hP`myUy0U-~>gk=OiF
zd0cK=2~kSD!xO}+J*F|_?`Wod%faDVk)E0MWP7_`+lwW(TxFD7n|=y(1f?t1#$P|S
zkgPEuz27}upw?J*jw)7awW0H1dizPCR@RwTS>N>7AM)g@xo#OFC=-uO+%7lY3%%(Y
zC@ziar8BcnOty+rDltJnrus#-ls&2=MFLkI`enAt!{r*p;m3&R<!XMxqS<38ps%+U
zy>PEo8sUMSx1^O_N>a<gwMr%(xz^{qk;b0PANycPFuZuz%-ht{d^EXgU~^u@mb3TJ
zs-`CkwZ<dIYHZ+frh0_yixN|A<@z&DYb_L_!jf|*S8KAk+l)0wpi_0K0*SnboU&!$
z+cGsaDXDtKX<V6<|2Vg@>yjp_9+kiEaN`ou;({bVp*6H`Be_qNT-q?_rlcLm#im(B
zo@ubKe`i&yGq?o)7)d{1CiISsPWX8SdS<+HGB&S4t51t*cynrQek7-;5n0NKOHm~_
z2P38k)SYZ)iS+U5Vl&$h=Bm}2-xwP2$2-K0az(_%1i_N{`6O_~hueLUt%$W;9F*|I
zza_XaZX^Dw3)g;^--D|Uar^JY&cFb7f@@L&b?`xj)ips6CB>A#KGHCH5n=XeXJ$V_
z@~wOl3nN1%VkQZ3Xu{i>^>mf<FgQ%GxAGegAQ+x7<go^Bn;bD<HS6%FFUk9$<qa9u
z33>rV&iiFxoCFt*5u{khT~jK%a*_I~?lSD0N>B`S?}n_t%K>M|{Z0%oztvOk*E0IL
z1PU>XpQ6U*TPQQO?8jYi(UK`UE}0i@`96Gq$Ef(QP0LmW*{r*G*J5SKkv*u=XV+cQ
zC^mwsU&#+UFz(p<V@ZitrX^wyVM0o?tNH%)YpK`=y8|Yz0>&%GrW+&ZRg(@YjUmVs
zGqxr77RJn*cSmj26)Vie2w1(Ml)ohRBMo=>E3s=u?q=FihUz?)sz2VAlN-n7ZxQ^u
zFKQLU=H$TQqKjdi@m`yu9kg{G-5#7#h-eo<Ml<QkL}vWZ)NU$e?n-6f_h%L}_OKXP
zApmaRrj!DXWm*i+lmo!KT0_BYo=FCWjd{9dSyiF~)a96(s=+8VW}zs?za6ypDRLIs
zV$}^;ub6grC2qxcQT^%hpE44g=<n4XF_fd!ZZ~Apwd<9bVER*LkJFg3=2#{HM(cz@
z<0dPOaKT~Z5c1<^gQ1FchU^skDHuoL=lw6by?Lq?Ifi%{e;Os{%CYhJTTX6@rb|{%
ze`zqy?5_vCEIm7`Mm(9!=fd#U4Aij|=Uc;NJt+Q$@-bmjXLIj*ng(Vn4T5k;BLfqf
zr<JjB&4WC-SW0J5>9GHD_nn7n4UML4zmHNde}fwy$#Eze(hA*`PL6SS;&_)yBL(Ri
zEVyH^$`hQB$G95Exx+2Yo5*}8ehlw2<fG7_oT%$Eri9n3&>b5yQp!W$n3qrKe`HDx
zZ0Kj80I!Z^4V&S9Us5{}s2+08(@Hzs)Vbip5;fNLwN4f)*0qsF5K5=3s=Tbr3HZxF
z^*%FSG-n%P2GOO^1^rfh8NfAXP9P1WUGtfeHj~V_;LM#(;*wEO@_E}}gpSl=(e6wQ
zKcBeu*#oPNIh1<ci!+xf%Y&p=ozFxiB)aU}H-u0RSV+2HP_wtaS0b)2QSXofB9n74
zKzqhgX0rC-EX#whD@G$fLqwxV7TBhOxnz*Vvn`)KQA9jtv+xp?gjWhU|8`|%Dk+3L
z_Vtg#JY@#od-nH6tPDAaX{IaFiB}(8dXcm#-Al-_2?p>k)q{-tmHT<n&vtj~rRE}D
z&EwRrCJkIN>GYP+s2~qD%lC7|8fw7>)Yi1b&H}5s=?t{pkDK->s(0RzxSqFnWt(~R
zME#tRq_beYpMM#m?cbt$GnP3vPBK}B8sv&ZvqbC0RGk2GckOLp+oy>~@P{Go=igDw
zC*&8+buU|=6HoK#2G=$v6yI_eb-5eD5o9UQl^KN;d!En!qC-p~%g7|I3X3;(*1Js1
zKk(N;e-(vr%q*lV<`=q+53{QnG<JcJprq<0M&_nuJ9p*vcf&1lSAx>=<AdJQC(^vA
z?wcsqtV8O_w&v&h;AOO)vNmWf<9x2W4mag0OQH3*T&<U=34mxMIJCW(3-Zm0$CI5u
zT(DN4T^_Nk-Pxm}ZEBS}=`vv$D)-S0)eVtJ6zi`b({qriB9s(0B(TiBROuHJzhnBL
zR^Gi^>nM_dBlc@UX~X&K7qDVrY?ZPD*HkH28oZEB<}svXz?W}Fc&nH7;C<{xhB>_a
zpce-g{=*P}i}@Np28mEst!51X-_q?fUUjPGt>18^H-5PP`R8p`j-35k3y)Ya9CYwN
zT`UqfTQFeIjN?k<OMC<jkXcSrnlsTgaPXy?GV{q4ntHZWko@T1C-S|qmGai{QqR_D
z)gRVbwRr+J1t9PMaLSm`8fHWPiOP9mM+tFwW2f>5Gj^A8pBlg_e1BG+NYw+n4wayA
zzF~&156hRi<}EC9MVK9~$G6z1MK^y@lGU)4YE@(HmdVsenSSSDSZVt@LpF77F6>>Z
z{|2oCq?;e>_&nFfp1~1kyIGT<?i1$9N6o2jij302vhaunXS7!iw>qy9f~D2)mit{^
zHN?XKN1~4q%-heHQ8pwtr3P??&41WIshNiH#vd*@7x`Hhn4I}y%q;n<=+f6YD|b#;
zOfHGdD?07#8LEaEHnUDHY_0{*`Lc3)MxLwU7TGjrqW=Z0%5Oc3^~UE_^~pqUwv-lV
z%(?Md&uHDq`pQ%x5S`ati$b%nZAkbDGpETC%OpB@3vam?A9#{-&osaLP^`2L3}d?w
z6IfJH$6M#^Rpy{{y?8T7u``GwP8g%cn2}3*<=H9GcC`F`z1I~qXaQO|Lx`XR&NNNp
zP=S0^$7kT6e`D-k8ztAW)@un$&LWo#It}~o*FO{ucpslZ6rr^758WrZziARnA73(t
z0aGgeTiF3{<`ni9ds^r-=}^u2N=t|-;ejp+!d@&3T0APxvNIwN2eL9tklG&5^DoD^
zT*ph-$E%wIfFil?aQ_j{%S(X~n?Ytyx0;qCGIkGgY!>}@@IZTlJG(ungNNc@(d5@A
z+tq$9p4C`2^M;sa&F~?45`N$iX$wAF`{4+;`~rguX-}?M&9my=h^c-hRVNG!aj1xo
zam7Pz_<a*w3{bW1=n_FTdwzZQRq8hT!2|x;(%R&AG1hLSz44s$=`D1fPo~io2(9ia
zI_=lf^23{SQ?KIG@pW#xYhLfV$`B+Hi~~kO?$618-!`{0{IaBe5$Sy0j}aAYE<=LS
zg)%|?NG&|d0>g>L0#i}e;Ef(=<@go;gL(_sU&=@KTyPRH2GXt%bd)U5BhQpEcRC6K
zSbfxuDC>jM-+aC-@QL*=jG6ycC%G{8cvtK+mMko+Mj#DIaa7le0ClVGMbyQ&Lhq6J
z1#+qLMb*-BBX^H)ps!H@x2xwxuTqhyDvyWMi+0oy5se5*ndGuz<kg!LDZyhbpRS9%
z!7!(ok5`?n3{Eb4&XdAQZh&2+N|~4Xb2S?aJ;SBWH4Sgxl;f-D>p=Gr7uk~-a{WSw
zp2VL@9PmRGa05kviuB{W)03hKpc;pHv{N3Hmp~j}1}<qbK6l&jost&G)0aOq0w>wt
zL5H>}TIJg=!HE3A9eyqTjl6q@;6Kh#HK$YEIew$MM3(#5oj`Elj2guq1RNM(v=Q^D
zR{Qiga)#)cw(+J|S=LwIkGzg}Z3!3y`|T3Lu>z+V^tAKg1`c!NYRx}BIvz$sHUj8b
z)t7VH`CxvKr5Esiu`CRq3lAc!tkvy1E5(2cIA|uj1dj9(X~G<m%1L`AZFc7VP5kE;
N%{zLxQ7Sf%{trs+hC2WN

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/logo-light-mode.svg b/docs_new/assets/images/logo-light-mode.svg
new file mode 100644
index 000000000..9dc4ae1de
--- /dev/null
+++ b/docs_new/assets/images/logo-light-mode.svg
@@ -0,0 +1 @@
+<?xml version="1.0" encoding="UTF-8"?><svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 347.6 94.09"><defs><style>.cls-1{fill:#e35a26;}</style></defs><g id="Layer_1-2"><path d="M101.92,79.35c0,3.71-2.14,5.97-5.04,5.97-1.75,0-2.85-.86-3.45-1.64h-.04v1.3h-2.32v-15.8h2.32v5.88h.04c.6-.86,1.7-1.68,3.45-1.68,2.89,0,5.04,2.25,5.04,5.97ZM99.6,79.35c0-2.21-1.08-4.07-3.12-4.07-1.9,0-3.07,1.55-3.07,4.07s1.17,4.07,3.07,4.07c2.03,0,3.12-1.86,3.12-4.07Z"/><path d="M108.52,85.69c-.64,1.75-1.37,3.14-3.54,3.14-.66,0-1.22-.11-1.66-.22v-1.77h.2c.97.13,1.7.07,2.21-.55.42-.51.73-1.39.31-2.45l-4.09-10.14h2.5l3.16,8.64h.04l2.9-8.64h2.47l-4.51,12Z"/><path d="M121.12,84.98v-1.14h-1.14v-1.14h-1.14v-6.88h1.14v-1.14h1.14v-1.14h4.58v1.14h1.15v-5.73h2.29v16.04h-2.29v-1.14h-1.15v1.14h-4.58ZM125.7,82.7v-1.14h1.15v-4.59h-1.15v-1.14h-3.44v1.14h-1.14v4.59h1.14v1.14h3.44Z"/><path d="M133.7,84.98v-1.14h-1.14v-1.14h-1.14v-6.88h1.14v-1.14h1.14v-1.14h5.73v1.14h1.14v1.14h1.14v6.88h-1.14v1.14h-1.14v1.14h-5.73ZM138.29,82.7v-1.14h1.15v-4.59h-1.15v-1.14h-3.44v1.14h-1.14v4.59h1.14v1.14h3.44Z"/><path d="M146.36,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M154.34,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M159.47,84.98v-1.72h1.14v-1.14h1.14v-1.15h1.14v-1.14h1.15v-1.14h-1.15v-1.14h-1.14v-1.15h-1.14v-1.14h-1.14v-1.72h2.29v1.14h1.14v1.14h1.15v1.14h2.29v-1.14h1.15v-1.14h1.14v-1.14h2.29v1.72h-1.14v1.14h-1.14v1.15h-1.14v1.14h-1.15v1.14h1.15v1.14h1.14v1.15h1.14v1.14h1.14v1.72h-2.29v-1.14h-1.14v-1.14h-1.15v-1.14h-2.29v1.14h-1.15v1.14h-1.14v1.14h-2.29Z"/><path d="M175.45,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M180.83,84.98v-2.29h2.29v2.29h-2.29Z"/><path d="M187.59,84.98v-1.14h-1.14v-1.14h-1.14v-2.29h1.14v-1.14h2.29v-1.14h4.59v-1.15h-1.15v-1.14h-3.44v1.14h-3.43v-1.14h1.14v-1.14h1.14v-1.14h5.73v1.14h1.14v1.14h1.14v9.16h-2.29v-1.14h-1.15v1.14h-4.58ZM192.17,82.7v-1.14h1.15v-2.29h-1.15v1.14h-4.01v2.29h4.01Z"/><path d="M197.96,71.81v-2.29h2.29v2.29h-2.29ZM197.96,84.98v-11.45h2.29v11.45h-2.29Z"/><path class="cls-1" d="M15.18,39.46v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M39.46,30.35v-3.04H15.18v3.04h24.28Z"/><rect class="cls-1" x="45.53" y="27.32" width="3.04" height="3.04"/><rect class="cls-1" x="48.56" y="24.28" width="3.04" height="3.04"/><rect class="cls-1" x="51.6" y="27.32" width="3.04" height="3.04"/><rect class="cls-1" x="45.53" y="21.25" width="3.04" height="3.04"/><rect class="cls-1" x="51.6" y="21.25" width="3.04" height="3.04"/><path class="cls-1" d="M15.18,51.6v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M39.46,24.28v-3.04H15.18v3.04h24.28Z"/><path class="cls-1" d="M15.18,63.74v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M15.18,75.88v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M15.18,6.07h6.07v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M24.28,6.07h6.07v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M33.39,6.07h3.04v3.04h9.11v-3.04h3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04h-3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M51.6,6.07h-3.04v3.04h6.07v-3.04h3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><rect class="cls-1" x="57.67" width="3.04" height="3.04"/><rect class="cls-1" x="12.14" y="6.07" width="3.04" height="3.04"/><rect class="cls-1" x="21.25" y="6.07" width="3.04" height="3.04"/><rect class="cls-1" x="30.35" y="6.07" width="3.04" height="3.04"/><path class="cls-1" d="M66.77,9.11v-3.04h-3.04v-3.04h-3.04v3.04h-3.04v3.04h6.07v75.88h-3.04v3.04H9.11v-3.04h-3.04V9.11h3.04v-3.04h3.04v-3.04h-6.07v3.04h-3.04v3.04H0v75.88h3.04v6.07h6.07v3.04h51.6v-3.04h6.07v-6.07h3.04V9.11h-3.04Z"/><path class="cls-1" d="M102.09,60.7v-3.68h-3.68v-3.68h-3.68v-3.68h-3.68v-29.5h3.68v-3.68h3.68v-3.68h3.68v-3.7h22.12v3.7h3.68v3.68h3.7v3.68h3.68v29.5h-3.68v3.68h-3.7v3.68h-3.68v3.68h-22.12ZM120.53,53.34v-3.68h3.68v-3.7h3.68v-22.09h-3.68v-3.7h-3.68v-3.68h-14.74v3.68h-3.7v3.7h-3.68v22.09h3.68v3.7h3.7v3.68h14.74Z"/><path class="cls-1" d="M149.89,60.7v-3.68h-3.68v-3.68h-3.68v-29.47h7.36v25.8h3.68v3.68h9.22v-3.68h3.68v-25.8h7.38v36.83h-7.38v-3.68h-3.68v3.68h-12.9Z"/><path class="cls-1" d="M189.18,60.7v-3.68h-3.68v-27.64h-5.52v-5.52h5.52v-11.06h7.38v11.06h7.36v5.52h-7.36v23.96h7.36v7.36h-11.06Z"/><path class="cls-1" d="M206.38,60.7V12.81h7.36v47.89h-7.36Z"/><path class="cls-1" d="M221.2,20.17v-7.36h7.36v7.36h-7.36ZM221.2,60.7V23.87h7.36v36.83h-7.36Z"/><path class="cls-1" d="M236,60.7V23.87h7.36v3.68h3.68v-3.68h12.9v3.68h3.7v3.68h3.68v29.47h-7.38v-25.8h-3.68v-3.68h-9.22v3.68h-3.68v25.8h-7.36Z"/><path class="cls-1" d="M281.73,60.7v-3.68h-3.68v-3.68h-3.68v-22.12h3.68v-3.68h3.68v-3.68h18.44v3.68h3.68v3.68h3.68v12.9h-25.8v5.54h3.68v3.68h11.06v-3.68h11.06v3.68h-3.68v3.68h-3.68v3.68h-18.44ZM300.18,38.61v-3.7h-3.7v-3.68h-11.06v3.68h-3.68v3.7h18.44Z"/><path class="cls-1" d="M321.81,60.7v-3.68h-3.68v-3.68h-3.68v-3.68h11.03v3.68h12.9v-7.38h-12.9v-3.68h-7.36v-3.68h-3.68v-7.38h3.68v-3.68h3.68v-3.68h18.44v3.68h3.68v3.68h3.68v3.68h-11.06v-3.68h-12.9v7.38h12.9v3.68h7.38v3.68h3.68v7.38h-3.68v3.68h-3.68v3.68h-18.44Z"/></g></svg>
diff --git a/docs_new/assets/images/logo-simple.png b/docs_new/assets/images/logo-simple.png
new file mode 100644
index 0000000000000000000000000000000000000000..2c14b4d9f9d8e38431c44c004594d83ad6a111e8
GIT binary patch
literal 2196
zcmb_edsNa_9Ea7cY-(l8=XAB1bt<(qEfs4EmmU{A2rTe<GGL<Ot28OJJhRlum7B#3
zoKij@P*d=Mg3G2k#fE|S#7f0PiC7UnLiclZoSkFY&gqYP?m74OyXStspU>y}xi|Fy
zY_Ea7xjqO4GVu0-9s+^D2Ef?8R1a9m7jLWvre%>{M`J*sjTtioEMB`E_yc4X>v<#=
zPQb**1xFtT#l^+h{SbkVITjpw+>Q{9&3@``4gxJ*>ka+paQt;;uP~j8HEUFlkTB=W
zS=G1G?0H=;ojO{)hnx<JsDIsDo@;m4<bZFgh0fTo6Op`nlm5EG-byy1@~5BWHyn*m
zxRO!+<=LnetJ6@9&wC=OH^P3~C2WdmBYBKeRTg`W$$QeX3pBD3;b;&)zKS99&{#R7
zmfpwu+No^5s%M-8%dSBeUAUmPVx_^C>&$mUosur#Kr2O{b$MXu>kpmR6V6%;D~2a`
z`B^BrxWbCsP+f?FI*U<l>q=J;5SpP#JtUh$*7`$YoF~-JQqmea6g){|hUXsXR8)6Z
z;`E?&a_D{pwU%C?HtNJEP0D(Ga#}G#8bf({t!yNATYthQZ_aurh@;&Vwv*`eB`LWu
z8%LI;m?0@^YDx7CsLBf0BShdW$hn&HjEb11?+MEOGOCXtKmtn|S+-);yNY^!&_I3-
zq_Yp@Ub0m@lu0Q~=`&y@+PLEA(_1Hqiia(avsns%raN8Rm7}z?_Qn{>yN$T*9vlWW
z0xYN@V#mkCK9)oI&|^3qTh@IwkD>^qDmPoBC9oQX6ibvSH+c-EGSc!6)^ZaNF9>q-
z&1oj1eVaMkc0Qix)_!?R*`!kS@}J}AItR+Igv1x6OkSY~r&L2_!m+@M;bQ|BN$M}O
zgVHH0BPy=X;!;WuJBW(D^Qd90P3-2k-;zv6+A>3FE#(1%D!jIc`eHLuC2oZas0ru$
zTe6x>n2JA&v=0SUw712c3*wCnoQ>q^R{K6v{uvh6i7YoNk-uu2<BoYY0Jgc?2Pr}l
zcnR`R%IBleK2O3=$03yK9a$4O$nw_d@adM$jGl`bHo}u9X+9XkHR)Qzz4ep6oCnoz
zJ*I)K96!E;#;UQ2E*fbqZ7A?9Y2K0tap|s&Z4{$>HtexZbVl~_yWq~WLdFvMWxOu^
zn1!J9*E8*XksbDjCF4cf9hU&@=SMc{^sf_%hcwc5g{{l!O!VE3$-sQsqm)ohGE&1v
zSc!<r#yzZ;D5mNbz$%otL!&2sK<n&k<(D8&9Ov?SP)*Kt`%X9njv^$(Aq|?v@9m%k
zkj&F_A8SV#thP=wYQM(1$K71!2(<o>^<?!;c=O<LxGL&8hujd^A@FFAk5E8@^y`BK
z2M|LAvi=<(;7`sOVjGHg1VNxC7+w4BUw+ECqDr7A;t>!aW9sl9;H(9LXO@I}U8K#!
zL3lMMR6{A!CV9V@5QP!OpNa97*l_wmR$?JTSP-yiy=bC~lqFhNi`!hnehlyGecu_Y
zfC8HB-yahHN1mhI88%A77Aw;f*X=WGN8QrFiOse%q|Rm(Tm^=kq5!V(W(HE@cZBYo
z+W8PO;-uO^JS8U9t!^k~DFr$}IHvup_P>8Otf72#p7}a2gqsV1o=ghn631VO(Uy%1
zQ*^J34=Iw-ijh2cF*x&ya=gg5k_O-ego;E!+&{7-yoeaSZZPCwA;X#Z)Cbd0>R*FZ
zwQ1?Xq<fBNZDNG!#NJFwJ)^+oPK-0wt-0~a+2jY^=rg!J7zU*gGFBNJ$wlgnYo#Vv
zDCE*zuDs||_LL|H_`geUmjAbI!?65CD)&>570tyLMadjGsd6R(ZeYR1fyHVOnLEx3
zk$+)(o;kcT%E{x|K<4x<#s<)m=Sm_F#I#RqNGy(;O)Wr9<&-TPPXoK`aGT5*Q{+2=
P7X`@M69#2}8+7g;FTJ7w

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/logo-square.png b/docs_new/assets/images/logo-square.png
new file mode 100644
index 0000000000000000000000000000000000000000..9bbb39243415b826a71ba65121b20ada5a7b9503
GIT binary patch
literal 3323
zcmc&%X;4#H8V#<vfo)fVB9L~likU${76E068zM47+wj;43J5eIgguf#5NSc1B~AAz
zvg<H(h8MyzAwh%$Xl1bt1SJ6_K%xyqLPFS*K*EEGT~^OHwk=a-)%>{is_w1(z5AW-
zob%m#;|n+E9b5Ib!eFo+E@w`C34>`IP>;WAZc@(>$KYu7XG_c(-*^}dfq6YNt}w}f
zdU8*KlTU&NCM*GaG42u!i^ZCL6^)7yxfpZF6cZO-JZ-NBgKfX=a_YF}<r2wA(DiX)
zVArfPej>`+eAUmyrKmLb{CB12HfHM?{V9U{^z0`Gzi9{!uN0hf+E4jogvp<;o;har
zjV9&Jt#3=6Dv(xJzd$v3?(?Mh&?eoi>9-4aoIvl|I|$jt8=8GLfDA3NUvJq8MIp^@
zH!sblC}Mi?a_lUtVH6+eVx3oOKZ+Zn4GI@T7{?a*Lu2PKYj1Bu3J8Bzagzypk6<sO
zA>!$<UVZ;1$*4%Eh$PiU%G%~`$P~n!k|TsM5*`B|BHvsIQc=fg9td-b`dD#yDuwnv
zp7z9h0c<_mLSGrkFU<Yfgg=6VAi;Te(^?Auh%FTp=3xe#IlDz?=gGP&yRUA1xbwj0
znHMz7H*CAU&*#|o_I1IUzb-|n5|@(W6UX^vk(}X3GMm6O)g{bY8d-WPo_CQ&W$hI@
z8T=s1!O_GTk(qN?jc-F14uv>+*)wf+>a${62jmYG-F0oDm@c0nE8F_Q03lg4EMPry
zGFZNupY9vDJPWR<GK4u(l3F&Ko)#54CUU}~XmV1sE&7$ipda+0$*%<+I-%@nZ#JX3
z!r}g8$7Xe6k_^jc3qJ8AvcXd|i@0A&na7scLgV@QJ=_XjM!6IWT=v&NPrfwJF4Y@G
zUqUUEfRSbdMlfSYjz+D{iMv^yq9^msmq%&{>Vr;k3p%m>N9-2XiM^&$7ont-@wBcW
zI;fYzmLXlS)wvS+!fb^usNLFFu&Fg|(0<pt-$d&zAaeo(38MiHN;m*@w%6~CoFVX@
zM$HE5PWYF@3k!o<L?Hmp!m2-9YMVB*+;?qq#5QIY)RuvbMAiX61~yjk()o4GZwK1X
zOi?4HhDC)vgb-A$Ksz&R=$Znk4Km=>Fav<q!?y8(z=)Ma9~nxnC3hk-2-8{Q9s)Lk
zsyjV{w6l7mGNx-G_3no<5@J^-yr`SAdXrxu1&JA!*&vP(dZ!VmoVg*h*<MEFJ!V6l
zpx%ywI!Q;m5VP7nDLfsfCTaj*`7AmcSrEP2t()2VJU5qHF-ULkGn0uLC0)^7M8j%F
zIxiS+-YOY;Ryo!aVj}D$L;>7w2yyi8GCdAxYG}v^O%-=1@v_(_Flf|*18#z8<ey4c
zG;_DPN1hZmAj7fvoyRu3MZLGce8b8RX4CySj_E|EG?6Can-^mcmKO6$QgkYAp~T5w
z*-pM?s=vE1Sw4JtHF@gQ@#aHPh$Lzou*qSul!-w@<W#1Y)SHp!h)@+9S%x=;KfvlY
zY9W_r#4iDDVy`4iLWo@?@RpTSIAz{DZ5}C}rwuhde--q&#??Crw;%|n3!7!GR2$PF
zaejM!F}A;qt`;40nW9{(>^RqGpQ6lG#p=|hsg*ok1-eVRd=0mxe?VftmFI_r`O$aX
zc4xGW3s|BjhyYJUtjOW=S5z-;gQKyc418*DrzjA?QNAp24#RQz298ZD>G_2{s=O3;
zn%v*Wk|Im9Ymh%==dc29Y#@U;?h`M_jx?HKu%LWM(6=Oa^f0)JymK^RrnX-s-zQaE
zg0w8>&YjWR@X7m~=S_mc?jTtCT+TMtakX({Y&JvJGawG3q+ckD;7ubetf*WqjG^6}
zO@=dsD(3bbLF45t!5;7Z?I1XaNKlslFq%Gj5OI(21K69xiqf}>3!=$y-OcbTMI;16
zD|h#RKN;=YdFq2c7*kLOS`%xR-DFC7l|oQS8(1CdyJ76uGAJzZbx=atVENv{NuZqU
zi|^`YMb*oS)FL6~uo{?)!*h>}39+FQXZQO60AnPOD=o$CNb2Jur2WUU{Bt3lYmwBW
z%Y9XMzl|~sO2$uxw0`!tA#rggb9L)@{?2zT`JKw&bM*ljI;|W({0Ij>^Hhgg*{_Q2
z9r#kktE4C&<rjMK_ErcNB=W+6&fZB%YV1OZ$$6e_(%+?B$LC$Mi=&el8*~k74i%G4
z6rNm?{O1G2um7tHyz|R{H;9kE^oEDC71AjN^ODHnf-!}wt5s?%Yp+t=F1i2!QduuZ
zkaF;CZa54P;F2AE)~2!MuK}*#D#J;dW-c?V``(wFTZO%+ujvIC9IamwaO_}aY&8}B
z56XK3;e>}-%#mld%()uhT7S~$`lNqG$QE9FI+bQGGxYeh8jNHx8ogVwzaY}s1&RRp
z+6wg`@JA3Fg;#kI-5Hz2|4<l{kT2ps_W@`raZq{{7KSWyKJniE|Iy3)PCyn3DTu49
z$Bn7{P_@N4?Po4h$Te%QY;lZ(|I4{L5af!}MbK!K;!0^lk%Je2DZ&w}lJ{@;k4wrs
zG2hU0dY2bKTOo3%hB>IVJik>j$0R^YZBVz|tN93-R@>wGGn&75`FUU5-@h`gbyv<3
zXIp@lwKqQ|oJ>KzF8EUvAt_lh+wUPYKdkmIup@&p#PGmq*!kQ4l6Ut1kT<}u2HEb6
zjpAyI8>4;vSRELdjD|`*q_7tIkFIa?S};7xTrmVj8m;G1mr(2$RZT}X>n_GY76tji
zi!_p?{n92?+d(0F^pJ%m`Tj*+_){dn&eNies%Bi&&cg<{xeqcP+{!yp>p;Ls*%}~d
zoyT9I4}`;cSrsdI_1O<soLrO8J5xj(nPn0QWr@U(SEQ?n<DFfMqFc$dg4mY~sw*qy
zi#@CJ%N!J}V9T122x-J_X(UG?R{lV}^WzGS$d=2aGm~Tm<%-7>qMs(g`ynC+0WkJx
ooF=m)A%l%qHm_eBt7g+qwHY~_RyAv@A1*K#C%02ICoWw57oI%f82|tP

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/logo-square.svg b/docs_new/assets/images/logo-square.svg
new file mode 100644
index 000000000..efc68a642
--- /dev/null
+++ b/docs_new/assets/images/logo-square.svg
@@ -0,0 +1,133 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="69.82"
+   height="94.089996"
+   viewBox="0 0 69.82 94.089996"
+   fill="none"
+   version="1.1"
+   id="svg39"
+   sodipodi:docname="Outlines_white dottxt.ai.svg"
+   inkscape:export-filename="../projects/outlines/docs/assets/images/logo-dark-mode.png"
+   inkscape:export-xdpi="137.93103"
+   inkscape:export-ydpi="137.93103"
+   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg">
+  <sodipodi:namedview
+     id="namedview39"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     inkscape:zoom="0.27368421"
+     inkscape:cx="171.73077"
+     inkscape:cy="45.673077"
+     inkscape:window-width="1416"
+     inkscape:window-height="820"
+     inkscape:window-x="30"
+     inkscape:window-y="38"
+     inkscape:window-maximized="0"
+     inkscape:current-layer="g39" />
+  <g
+     clip-path="url(#clip0_56_2)"
+     id="g39">
+    <path
+       d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
+       fill="#e35a26"
+       id="path12" />
+    <path
+       d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
+       fill="#e35a26"
+       id="path13" />
+    <path
+       d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path14" />
+    <path
+       d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path15" />
+    <path
+       d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path16" />
+    <path
+       d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path17" />
+    <path
+       d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path18" />
+    <path
+       d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
+       fill="#e35a26"
+       id="path19" />
+    <path
+       d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
+       fill="#e35a26"
+       id="path20" />
+    <path
+       d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
+       fill="#e35a26"
+       id="path21" />
+    <path
+       d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
+       fill="#e35a26"
+       id="path22" />
+    <path
+       d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path23" />
+    <path
+       d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path24" />
+    <path
+       d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path25" />
+    <path
+       d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+       fill="#e35a26"
+       id="path26" />
+    <path
+       d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path27" />
+    <path
+       d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path28" />
+    <path
+       d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path29" />
+    <path
+       d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
+       fill="#e35a26"
+       id="path30" />
+    <path
+       d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
+       fill="#e35a26"
+       id="path31" />
+  </g>
+  <defs
+     id="defs39">
+    <clipPath
+       id="clip0_56_2">
+      <rect
+         width="347.60001"
+         height="94.089996"
+         fill="#ffffff"
+         id="rect39"
+         x="0"
+         y="0" />
+    </clipPath>
+  </defs>
+</svg>
diff --git a/docs_new/assets/images/logo.png b/docs_new/assets/images/logo.png
new file mode 100644
index 0000000000000000000000000000000000000000..005b7c6d2195cf0af295c15fd2e167f3f30b1fad
GIT binary patch
literal 6386
zcmcgw2~^T~yQb+Iota#QoE9~8(oD-yGjheWnMlngQ!8_W%F+}G6&1~enx+YJD?@QZ
zO0(4^H`G+riL4MY_bnwcB{v8a6<PfL^v!qfnR`3k@1E)0!-4nkUw+H`d*1hbo;UI6
z5qrh&)V>3OK#GSQVD2E$Kj0wHx&z<H0avcxuB`;jx1kPR7!YVzn(Vu-XS=B)a8Whl
z(8&l7bU+05boe<C7K=3sx`4#^pAJ1|gbokP<yxzOKwEK#VLu;>&YK&Fzd;I&(RwS2
z>F^r-#o|D-N&1Oaxx)r+J0n*1MyNGscd6#xaCUfh?Zn>N@AbQ1XW#htp#0E*so%0Q
z<D4BFHI!S*tHRHuE0k<FokNK`yYh|yZN^qFgUc;L8K}z#^9BaeqdG*3X}|AUe^a74
zH!8}{D4pSn>R9B66Z4(uWlN*#J_`h?mg|+9X{V<(4V7Pf_j>V-I3xZVuoDZSX}mmO
zK^C67fW3o2>H2@q5WJBr)f4UIjnF82jSI6@P1}cB+zHCb3&|S=d~i9Y0TNb_5+0Wa
zxltwwNNFic!<^Yom*jGkMo1aofIx43A+ylt;NakeF?<xZUpS$!jLwBN1my+(#I0xR
zc2dY&InIU?lPGzmCKNEp+(5pDr&&>>e;x!ya{-6NAJGPZE@9Fh+sQU__y^-JVSbO7
zx489uKNRlk2!V_~4EO0-&E}2^da}x8qg&+;u#4<4@vw#{U=bRAky9;&2$$>s0B-Hs
z?3p5a<nKE5&ofczXKzmul|~;<9@;-n3T`@DAe=pEPj8I{KEgR^Kkw=LcRB*|3UW6r
zb~y8jJzAZ^VCR$2iTWK3`o1a0;Rf`K)=11<-Zo%O#f+REjH@Z0nH!rdfIUh2>V&!O
zUxC5k!lH>3H#vW*(KxcOPEWgaH@tvAJVa}A|Egqt<ZdOq-`HP+K&MEUUf{SuKPJll
z{cZCq^v(JgOWXm!=~HVKfjRnG8i!)I7J+ilRwhrjY#r9!FigdSw8i^38U+@bP2`s=
zn-=0mC?2j3_)4P?Ytt+qbUm;N1M?`pd%@tb0-IRIL2}b#^O$j}=%@5}wn`h(&iUQU
z-o`Z~V&>zc#e^@5kfcgTWZrsR_R>_HFnyEEGjDyz^!{Eb{WTI5I@pZy)OI*O@3?El
zk$tEbcNCr2YUvSNpOq1KsI)F18gNhtJ;518-|sq%dp{b+{E0hEZA9JAzmXBth4NKn
zAmU~;lm#TfRhKd>bJ^#xDsyv>^UpS(r{AZvYuM;rzppa9IiX5baURiTHmU*Usqq+Z
zu{q}vndyNV7Lh;us({I}P+7V#7=Hu?%z=&1H+%t;KIWAtjOtl=`AE-8FW7@4Vm|Sf
z^LS>TnK_d(*b!79VGmpjmMq{avc3=W({~DUWvK$Qo~*0f1?T45Sgf@kcl{$LmM*Vs
zyA`Y=^gg0xJujs~fJd4g>w;x*cmu}&N@m`Ll&Y-*9SB^eE~yezGw+Y)R5t~+mH1q!
zTd4FBtvjVF%NKXm43)`hZmjg`CZ%Dh&M*A0nqlFe1G{qn*Pu)G<xj92bcVP4s5udD
z974)vk;~ODy}BK}4!GwwG(V|tl2Klt8GQx@94qs*Y#kub4z~v2_rDwd*i#TFk((f3
z8pvIl=N3!mZV%$7+OeaV)mosFs({N66a;*~#h%s?IA>hgxkVdTB1naPTn8BH(vIs}
z{-4Y<DgdPB9rdC9bt{)W3e(GDEvn<J{Eb1k4$8cp#@?l6Eq?2<s<f{8>BN_W%jZCs
zewww8`EGP@#MS5eDuIeQpk-#HIB853q+VNP{=7yl=S&fu6?2kSndO`13R~9aDl=Iu
zY-%OH8j1A{9lJF{DVFFVg?&rw?bnQuI4v$tQkrxmA5PtYTst-x!P7OzP1GBVU_zoM
zr(7W3W9E<Kd`{$ycM{z58*0&_rD93<a8rV(*c(;3R3^WnY?Pn8_qbClx5!QOc0!(K
zb+a6*eP+zGo<bO#AKx3LJf!F(t9>JpQd(rkP}Q-fdP|+Q;2ax)08J~2?7^lUCI|yh
zcm1SnpG1s*udXLzcHWPLSH}B5lQ+orVDx5>HJ_lYnXm`BM%!%M7P-j1M?;F3>*D__
zUM0D~_eQs%W~ZPpyiZ=3=D}#j%cSuT-w(gFKnhE685CQ(@gbUg#nSTFyN4@-53blU
zhMIhjH15`xo-K;a!LO;zo$7Xsy1TZasCBqN34oz=#jidg1W7{#&LNZ?yW>oUvgiZ*
zdli)IXw>8pFu15k(y0Uo5MgPDwWoF0Gs&fT2=Ad5a==+H7gn-C4SaJLf@QhWO?4j@
zrmimAh}lnyh#$b@DX2X;SoO_jlyxO<8*=PsH~)A1%F(`rFH=VqL`tAg8?U^PMx8Ul
zY(I_@>K^i8_RXDk56_9%vrNwjl9qyf4O;rFOskFYkLJRQZ!B<VqJBq*LZ!EydS;nl
z-iYu3ad3zbH7pgToeiXivm!kkqGR@54DT1J746Etywf3Nwcz<6i5Nj5!>v{k<9$|i
zMwR=9vO&R{t}0>_sjNu+QaFE+=uNJ)s9!(>u7Rsb%I>pP1<Zw8J`IG`R}Us4SSGiX
z&+Xz)LDyRbLoUVywc)0%arp!z9B&98j_MzyXu!3==6aLs-ntR+NjmgvB4+uJV~M$4
zgqRcHV+StZZhEw!^8wYoz!yMPoX|cR1}oufwF)<}M0X9hpi-kPOuFcb)5eDkMNd=7
zu4UMsXbda{)6d@^-;U8?xA}4ODA9vv*mw8vwGTO4Wl1WqJ@8Y>Dtn*<NOA9dQfkY0
zg^hiO4M|nAt@qAOIGQ+Lr!O$@5?zaY_NsUUaQji<^}H5Joe<$Qm3noMWW3w0>pIbF
zC@lmU`Ys+wzn3z?{+Bd+@^XFIaHQ~dN!g4QA~@z}So?-!?B!rX7Blt$)}?2}l%3%v
zLaN)=)M6)9{W!#-qO~|-OZ}0v6YJiZjCd7_`QElwFj5L$<8XcM^F|1dYmFFR0hkY~
z<K;$hSR9K<epGGUv~XFi+bz7J2amQVu-XP~sQ3DCa3|l4YMk)Sizfdt9wWSF>f+ks
zWgQ-33X;Yf;s2`E-u9GfZkk`coA-;8_=R`dv&ZGTom9eHQdZf;5>&Ui4VZ2@lH}O{
zDsKL|ur|b!O&W3Ck!b|(q|7N2HLX{Bd#*)P0g%`P)-8!n#FG=%Y`Y_)cbkcWWEEr8
zo&A!s{u-;YFU>EH6mxpVCS$N$LF$6dj~CMLRqP_v_<cuOu9cR<Za-;)K+=stu3aR*
z`eQWPz^UB;?noQ1Z)!KN*FN+7xs&0Q$Au-NCh3V8zxN*^M7gewg7>hvLZ0*HWu`)*
zoi~3Y7`>Ha-t%SZiL6S52AHyI3U`sGt=4K4ulQ%jw3-C}PDU0;>Pi=iN{rL+xSLzQ
zy>$E-Qg}jv?E-|Q53h}ZY2{=tv1-nJKv;F8u@AaUt3V(n2*H-%?w_#PdBNi?v-|v2
zGnRW#Z=1D(53P>k2Qsqouq$R4yfg{MOQi8wpq8r?!-QRn(r0TWylsKwK_cZ5bug*Y
zm^&O9Q)Oxb20eb%V^5hQlULU4Q2Rp{7seUJOs-W%#<@nQ&|FYBY#SBrCzXcOPkfjg
zXK0a0!s9&0zOr_Cj<P*%jEaf-x6IhNqzB&|7&v;ou%Uss?e#Uai4-Dp#L~o84IoR+
zgV}&U=&LOztj@{`lB0*w^uOpT=l@z2Wz`J`#jH4Wa*ud|KbSp)O|5=yH@iFjTC{>w
zVK;7H$UU3kNNUdt792L`NQUO9#fevD8S{?IXM>&UP7hyxhI@@o<$a_bAF_gJm{sPJ
z6FVCt67jX-w|;usY*@eWY9j>acjAuOW4E<Vdd-f<e3mHngT&WV#-JPufM5sA-g3_r
zibfvJ57ton@ATmMsr2wkg7fD^)g=vJEYQp9xy76PTGtI&S>>XzT<WlbD|+QZE4OkH
zy=9rvpo5H2rS<wOci@U<8a|2*W@JrYCl~KV=Xz$guzQ$<-oQeb1}yRCA?ZX^@$kV<
zax1>7ObS|4yfxoL|ERkSlMGWKPbeStq4m96%E^?FwSi8Lc+Mq3>*s)P<13?A7O|do
zDLP@bysc>uGw<6yboHIeiF)lDwfq>HKWkO?(MQbDvfd~wHl6Bko%wt)TV0#2XaGMp
z?kabsE9TATrce=$Jxz*@YN#ay+>?Q6A|Q0Qdr9kQ&7ZOjDD$Q{Gdk>QV6ip|1)7Q9
z$Vhmz)#uQZo+t_R18E2MjOdjP3vP|L`hM^v0zClUr0z7A>dCFG^|go&;H<CaIwV(?
zX5h2Ks7jh4Kq<5hUF7YBNN%h2??-yj9zL#T_CKh13iuIpV&rRAHR?NBWPRpShEYx7
zwHi%}RpG3&pP_G!=#gTl^d!qt`vh-!GQU5BWtOFOgeJOgb$zv`?iu+#^u=BG29WcU
zjr^q|xA@q8A;m^K4(9jL97u+)8DzYhB$Q>zR1_FG6-GPtPo<bj55wY&OkOKnYB&*z
zZjz+Z1;;Iy1ovU0@*=nXsYxcjcDD&LdZ-gh3>N>wGLe!e9bmrYU50a^E7zlr;V_%N
zD)BW$!JGve{_cNO1b+&HznZgW*2?U3hm<qU+YDos-aiW`2tUe+C*dkS4bEEDs}J!j
z$5@s!@G>($sUvqTLExaiD-!`RcPLj_Tq{=3{H*tMQpNnfqQROyP>eQmEt;N2WC%>9
z_J&~(W}>0zsiaLss>xg=vRgQF8&<cwK-yhwU9AAG-UD6uwOF!t!=3KYKOYlE?h@1f
zm}TfACFh?_Df=$nLFX}sl#BB>@@<hc9e+lXm3F#RYGnHqc(2W}{zX4*(5-NC|7`-R
zq}m9pmNH+`8+{fvY-~>m8@(p2^q%JSMlUJvs&?+0iQnmB=-FHG9YS7rGB}0FGu&hh
zGT0*%%}IE~lvUFc^khj<N)=mIJHfPiHtWZ8-`6<v$W3*;ZD6pE@;1L@6ROhY<@Sc{
zLk<RT?P;Gr@yzX=)a(hj3~!v3c7hc^p<~C|!FQrBN1nRKeMvq<BS(^KZR%&Png+Y`
z-HmsFd2NIHsXbk~+H$#?V)I<Ig*0^fU_vvcPKTU|cg2zr15k{|!<!xrFv8)BjU8u?
zI}-8d?dH&^x{3EhXh#{{LzbRaw*O5LBr95ZuE8kV(S7X_l5*8Vis|$Q6&;sUZ)t(T
zbW7>@oVzHSVexRN<-Q~2DRm|qHU1P|yVInT-fIDu<1<UgQHE=^J*!Cow|pM%m(SXh
zH7&i^A05;)a4Y-UAGHfISM%N3?9v64x_RcpJJyXN0wT>lEsDevn=j|l8I@{sZWoGO
zBrHD;(q~k~?LGwJa>G{NMSQM*ENbf4`WIdaaPggWSw9dpEiUS7D@BqJZ=efZ55nL;
zODBkzH;Q{VNyQXjovJ%%uT!v)TQ?pXSXvrckwkQ-ulHHj;)+)dxg))Hv>3Mr--HjP
zD&S7ZazIqSiq0eU*AC|Kr=%S(vJrWty6pyScXxrcSQWSgNy06T){}|mu?`rFgXre(
zhoDPznMv#-vx&-X|AsVG;wK~Fa~-&keYLCmA?6ty%jST-WI<4A_9%{{CrSlQv6;S-
z_!lXw65IRDC49>%p~l?&TXR4&?39}rj<?V9or`rMsEdy9&pUc(ss>U^)NQQrw>n<Y
zG1H6l<`q$a>OB{q7z(=MIV<@kY1<M`6+z|FaPL!>?zjY%VfS6RS_LQ;w|qo_i}3ez
zkgf50{=DTLO1wWrf6KYq3r_n<<uz!OMq|WL`Z4-^=i4ztBeE#-vC}JI<(?q9_N(?H
z=Iffh`eoNCBLmn})3D&_las3zPXYrq78rBgn1v3f!wd0ots6MirYp%r|JW$sLBeEe
z7?Xg{r7+tO<cx1+y&?>}Dnph};$<xPe{=*sSw(^J`PSJn5#6VckTrGhQ23|*?29BC
z!tZe=S)0JTX-iovyV>iDHtW7j%kPt;ym|rydcEtbVSaBQ3nG6~;%XvFKJCq3_`Xxy
zfEF7sO>IjA3GKK*Dqj=#M<s~Q0=wgsjR6e4+>rnfmktBnR;zA8#Sh@?TV4uQ^~=MO
zg$I2&lV#3MvaZ*^5bE_gO#bKhe)rcJSbXhC)LgyENyP+wFNoPtSx>d#R$k(oRS9xD
z+E3LjT&@s1oY@8}QoEqXyIZT`33d{Zq1Kcu{m~*#QFHC}y>|0$xZ00}U5xuAZ~5fW
e{|6U^O+Qpov+E8&+y~4KI(+B|tn`;tzy1^S?}<qO

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/logo.svg b/docs_new/assets/images/logo.svg
new file mode 100644
index 000000000..440b3f0dc
--- /dev/null
+++ b/docs_new/assets/images/logo.svg
@@ -0,0 +1,134 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="397.00375"
+   height="113.55765"
+   viewBox="0 0 397.00375 113.55765"
+   fill="none"
+   version="1.1"
+   id="svg314"
+   sodipodi:docname="logo.svg"
+   xml:space="preserve"
+   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
+   inkscape:export-filename="../../../../../Downloads/logo-square-side.png"
+   inkscape:export-xdpi="270.25565"
+   inkscape:export-ydpi="270.25565"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg"><defs
+     id="defs314" /><sodipodi:namedview
+     id="namedview314"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     showgrid="false"
+     showguides="true"
+     inkscape:zoom="1.0083333"
+     inkscape:cx="280.66117"
+     inkscape:cy="25.785125"
+     inkscape:window-width="3440"
+     inkscape:window-height="1416"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1"
+     inkscape:current-layer="svg314"><sodipodi:guide
+       position="-146.75933,121.72017"
+       orientation="0,-1"
+       id="guide314"
+       inkscape:locked="false" /><sodipodi:guide
+       position="275.80377,60.668961"
+       orientation="1,0"
+       id="guide315"
+       inkscape:locked="false" /></sodipodi:namedview><text
+     xml:space="preserve"
+     style="font-size:117.641px;font-family:'PP Neue Montreal Mono';-inkscape-font-specification:'PP Neue Montreal Mono';opacity:0.99;fill:#e35a26;fill-opacity:1;stroke-width:0.612713"
+     x="115.73952"
+     y="91.977005"
+     id="text314"><tspan
+       sodipodi:role="line"
+       id="tspan314"
+       x="115.73952"
+       y="91.977005"
+       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:'PP NeueBit';-inkscape-font-specification:'PP NeueBit Bold';fill:#e35a26;fill-opacity:1;stroke-width:0.612713">outlines</tspan></text><g
+     id="g13"
+     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
+     style="fill:#e35a26;fill-opacity:1"><g
+       id="g314"
+       transform="translate(1367.0913,152.50405)"
+       style="fill:#e35a26;fill-opacity:1"><rect
+         x="28"
+         y="644"
+         width="28"
+         height="28"
+         id="rect1"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
+         id="path1-7"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="28"
+         y="112"
+         width="28"
+         height="28"
+         id="rect2"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,560 v 28 h 420 v -28 z"
+         id="path2-0"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,476 v 28 h 420 v -28 z"
+         id="path3-4"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,392 v 28 h 420 v -28 z"
+         id="path4-0"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,308 v 28 h 420 v -28 z"
+         id="path5-8"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 364,168 v 28 h 140 v -28 z"
+         id="path6-5"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 336,196 v -28 h -56 v 28 z"
+         id="path7-1"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="M 336,112 V 84 h -56 v 28 z"
+         id="path8-6"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 364,84 v 28 H 504 V 84 Z"
+         id="path9-6"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 56,672 v 28 h 476 v -28 z"
+         id="path10-2"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="56"
+         y="84"
+         width="28"
+         height="28"
+         id="rect10"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="112"
+         y="28"
+         width="28"
+         height="28"
+         id="rect11"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="84"
+         y="56"
+         width="28"
+         height="28"
+         id="rect12"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
+         id="path12-1"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 560,28 v 616 h 28 V 28 Z"
+         id="path13-9"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="532"
+         y="644"
+         width="28"
+         height="28"
+         id="rect13"
+         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
diff --git a/docs_new/assets/images/normal_computing.jpg b/docs_new/assets/images/normal_computing.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..a71483a68ebea1b02a8b3b167fa0efe57b498f54
GIT binary patch
literal 64918
zcmeFa1z1(v*EhOpknS!erMtUBx}_ULy1PSCKtj4xMM031?k)*wq&pN4;jRsyb3Etx
ze?PzbzR$hibFb6=?7j9_GsgVQF@9sr<y!3Psp~}$nyjRZBnScm1d;~+fv#skVj$?7
z0tXEZcLxp*?hf)Dpdh0mg9{oK+D*awUhr-T5x9{4S-_n@z`(%3!@;A#!J&}hVBnDb
zZ<p&<5IQ_u09*hR#9a_1Is_Ct#C1D}2m}Fwgn|I;`$vI+gMx*Jh6JA_1mxhef1Cn7
zp<yAQU?AbHXFy0$z)3VHG!O_PsAILG?}8|hzdw*)`d<6*xBmqO9nHrDn)my)9mL-R
zk^KPf|KuTr4T5@IC62DLjbZoyWWPW5^$=3sw!9-5>meTNkK+HyW1Qw^56f80I&eCX
z{QoEW{h=?I3QCPFU@;_K|6YtAVEM`6@5T5DC_feb_Kcsx@^9g9n)){;KbQQbsh{KW
zZ|QHE`Zp#&m;9!wpX2f`>2I3)7bO2d@SCRo12F#*|E8&bLGm92ziH|}0P|1rZ<_ii
zBLC_8Z<_i~$oy0Oo2LGW$bb6&o2LE~GXIeOf0w3E6`#+2!Fl5XDTSCkC(PqB4wZQL
zv1;6R(LD*k0E$YWk~e!!_?h0PFG?uN=V`%_rbu4E2||6Bt<=45qf$5_(+r*SW^Oz4
zB+HvAT!jugA|Lnp&b#!t&Ia^Qj32%&_Z>Gpqz9UyDtIXjB(wk>2}pu_GjIX@@S$^b
z>WkZ??rfX)(o8tZe{2Fg3_i<l>ICc-bm=}*oxkJPVmPKV&$^gHE@AYq!a$E8sM*7@
zueIL&Td?yM-)r1odc7S!cW_JwJ}>0+mE-h%bl+aKO%U)#^qzE*X#G{ri)dnd@M+J>
z<jiihm14rPhWD{09Q=>hU*oicPyAJW;voz(Pv&O(tGYy!K&1TpBRW?vW4W5Hc)lo^
z0vDF6t-UPX&B61kq4AVr*kY!hpdG&hfgad=5szVf@eX5aX<G^C9Em4HUF}r=Jz@R8
zrh9mjeOCnta;5SS9aoMNa>dQyr~{-P9&`QZa{@!dFXF@$)j)*iMb%!>!$AciAlRa6
zUytctgcMu5VJMPjpaEoO-CLi$97*H~VaTMH^qwmj?#NEy6C~?tW$s-GMZ%~$!Wu~i
zVeOfErd<O~H`fiBcv+7Zx);Q^yi{j7gfKjIob13&{whBSIgoS3FXkHh<eh0PdFzFv
z7q;4wPGayZI;7N3HMvCJw!wdE27zFwOLj9?oXIwogx88?-YHlNiw!P<bTa7*)txCP
z{qL4OaIz)lP3R(QH1D*0;Z)=Iz}K_d^`*C0<ivt*ht>Y_v|?=IrT@M2<NOupliCn!
zeAqR-6A&(OnznZ19qz{g=TE4gpPYJ@^uAENDCx~BO)upw+fNMB;t8_;LUL+duW`j~
zp>tn7R<~xQmbS@dm0$j^g(0XB&Kfq<<4v2;)jGa!v?rjg?0gMkX|2A5|CSBR9n(JP
zbGK{k=18uPkc*8?akWmXFU(LLYsMW323@VwF5Uxzb$_5<C7oc*T+dYSg|h{Vb1auw
z)#EvsK_C1#Xwu6{br1H%o%V9d@_~Ag$9YTwm(eE*MQ;TvRS&5%yErca@ek#{vaw|4
z2_KIB;m2RO#Ezl|wgeKXGzelo9#<wzVtWlX!DTFhplrrJ4>{DKITw>TsZo5SX4@cN
zw?~fE-<WHd{um~j2qoDi*|~SY-_9ouT=Rygn!kL{_5SJTE7ZEj!fes1nHNn-k%lj!
z<+wq*u8+;kPiU5^p-DMTia6S;f^81~{bx$_S6qTrXL+%dPG1f6&8ZHJ)L;0AS^fnn
zF>Bz~pl+CD+Qy*b#^u(Oc-?YHQ&ol@0hZHVGcZvf=TEf0edW-l7xA?`fn07wc4`82
zWqF*`YF5_ET`ze)_<!IBsv!sE)6sFdaC03w&%iDelYKPh4NAC({77n_tlQB~EF&+q
zJK2g2_79kKNqw(#!i@|?YcKPnWkrFQ$7g`}hf)n)I;S1UNK?%Wc;#(*_!hgqCe<n?
zgkeyRP+?<7LokaKxxUDB<I#Ga6mp!m?uT(+*q9C?vf1XJTQEIhoBot1yWunpl)w-{
z9l;uAmMg3<&^eU-`*YEwXL<|4b)&ozvAQKsT3>n8EFL!?#tf8~&xa6kHV3UWC9b`5
zjf<WK>O(d+@C@Zqjf+a*%jvPtXBL=~={7%po;kh8C5rM4M%2K?HnTv?Orz<Y>&pV8
zfzj+H!7gSg9TyAE3ld6TxS5WzM&rJ*^E1yeJ(V=U*MkkRCv$`zGE2}glt0?2;1!pj
zj_oUt(ety8bY)MeqJjPk$cr4C=(v{wf+>@(=A4e6V2%O-{PW+%RDVh*Il}vlr>0|H
zT*fA7Y>4_*jug_e|5%+xnxSq5vVaok&ly6aSO=#CulYdp@7WN1E7>D{$EzmH!RTE<
zcZSP6Hh!~Qm=*G8Ti2jD6N}BuOSV@k>w_;oy>s<rjyC*@Dap=X74W<y&cEm8544%C
zyPs5Np$6qgEUB+MKw5)A{_w9MVT$(9_1%ui)_foj7tZ%{s7Kq}SP7r=wq-6)xv*gL
zgFy6kV|g`>uH=I0ff+?_`^({Yzt~S-n6HoIhIXsHbujrrwB7tD6Bwt-N<Sl^xtmIT
zMq^vTdSfFf%T>CI#It@|W!BSA5LIWZF1GMI3Az5jatBcBze%zPXMTJh=}v;dFi)Kw
z!$kYaAB(E{XwHz5P&g(SfAIKF(b@c0V1Rm_L$$<{C4$PAd<r!d&#e;pSpEDfYOL<|
z5|5u4TsELQ2!yCJlg)HKzPQliv@fFS$$Od-Wd9!^B{k4geCLJJ>TJ2vdi+ue6YcWr
zyGNOS5e`#UPOp&QH3*Jw<7M(%q@o|y9A+splwaGr-~C2k1+z<h5OP@F_N`Al9Al4&
zi&a6}UV~&5-37i0Wx~C1{sC7A^N*gN+rVh+>l@T=(r9GI&Ovt6oj)q3nqZ9};u1W;
z_9`lo4##EwZ`&#+1LnE~YmCKkvw}$FO6k$~=Pt=`l9fm4n-k+Fw%-iNXZ|vrA8hqv
zYw+Qgl}q+BJ*%&wA<KGrJT=WXkb<Z@4U#V6^$aOBCtULMtg~cMd=~%m{g8Eq0;|oV
zu@aqi$FUMA2Hr%+4-3jD4zrBZ?}jTCIrlUR3M~oNW)jty+U01hH~&WI{QMuOnKoqb
zRu(kAZ4$nh^(Uzs<9H?I6onXWwGHZt^*Alu{6fr>72av}m}z@G%@38Yu8d{kycz2U
zY&nTNSEgYHhzXdZyGq5BF3#6{KrB@-$6|I~9yP@qDw%dNnVQE_*w-M|D-yt&5C@|g
zz8rCl@MeyG9#Cx0^dg!CBEb*kXZdiAHJzIs>lpDJ9V#8_!9g>kp&Wl${n)j}GB>82
ztgOn-SfT78*7k*QRNg&;_38mdn=gbpaI%4L9HI|eodg3(+&nLmpLv$*SyF1Me+$Ul
zP8ExB*|{AQ)j;JKPVpv$G*oQsIh8yeYzT0ee)}wKI4^VxT!$U`>dgn<il|Gcy`@-T
zAedCD&I)vHh>T_j3@9borw?U^Wz4?#(ksBl_?n7R^W&$Hx79H(n8`l+8zdGf$h`U%
z@hn8s>R2JBojJ%3svJ&p9i%eR5v%L;3gV?Z2qK}V<l(ZJF2w=1PtG7O@61M!0|?fJ
ziZ^S_KudrZ<T+C8^hMK&LXO=q(g*~?&A|8AU$x715h&Sxo)y9WnWttEtmK~satfo&
z#hm_K?o><McOiuBxW?YRSs^c0m3)<AMtUz#mKX@cm`0jZr}9`*TzZA3Z>Z&^9g{$&
zpv$MX)|FRP&f%NGpg$9=b=9LT$(}_UV>7)MJDGlbJbI_c{TFz*y%RD?jJlc5giEj7
ze`p;wHa;ATnDSRTb+~loq#skGSQ}85H@0U;EOgq}R(f>g{py7t*SXvbVXgjIM!+7&
zV_oO4!OUcu@{3mDeE#mA2FXt~Dso@L#+ZiYpP{dwzRpx{YqV^bJ@Z(woV~be0eiPj
zmcSJDC1Q%z7Q`e}*|9(EUnu}w<J$IIli-Cj=(G?qjaN=RdZFNhn4&_m9^}mnO8WhF
z8HOU<iS4ISLQh29K^vlNVDTF^`iA`O`CyjpD%y-+%B;U0c>gERHwQntAb{waTjP4n
zz%pErYnJQ1v!;9l)qIua3jxDXH`{0=0qa+^jIY+Gk{-)Ve3Gbe2_!KoIj$JW{4;Xt
zslEJCli?W?LV@Bs$#~x_!vopLA3PjHYGi5iktcW-IRgSP{SG!p7~L>k=93zBqA`Sq
zBF6@@PP%;3lKa3uU`pN~t%&}ld@DG7q5Cr1?wg&aSbCS>-cQZ9*%@a}Qs3u}@Z<Fl
zRk89;{B*`TRIL-E^<lDsN1ofU7_|M=oR>%e|AZCC#blK5oDh_uc!LHAD*d!R>G`_8
znC>D--`Q;a{nMt1@y|uo*1i<SQ)>%1len!3ydxUQzL-dPvm9x1lj?xwF!jn0>phZE
zHdIDfM+9A+Xx4m?fTTvTfXBG+X?v~s%yS$oa1u0I!t<d%W?=1QLbCP+;W^O+0vAI6
z1-!?{lk+GSsoZz3FVxV!koNRvJkXNb8EeIUcQdFq0j{jwigd&T`k99F>JBdN3BlT5
zu8!vYVKn^%$|!n$GoYJBY1S9id@0#*IQG`5)2^HJXqH2u@}q4q@qPHe287km{L~jC
zg)x9>?U(*~g6IQI3!KAP^5DyuJ?gRab=PC(hj`*|KyX?puLeN{WsxNB*7c!u4?t6O
zyj5@3RfGH}I#kb|*_Pr)a31C0{P;SSn&%Wj)oI{wn@D80y@~@u2ZYB+a^{uf1-0r-
z=5)zf!jMUt@6F8qd}9c>U!!5%_k<w|{|_0?u7vIGnvO}1oOqx3`E-9?>Io3%;Pw|c
zGP2U@IDMGN=K8w^1t`m{Hjm$aP}Woan%115^r!PLQr2lcV<;h_(|S<a<QoD}<RXA@
zXSO8#TQ)EN>H8X~cbRjm8!Awfg^{&RwlP)G0Db)xp#GMe`gq5b4!qpaQ_O^Q`Jh$=
zL>fYSAd~qStdPDCS;U5uuC`f2ACHH}QDUmohA$6CF6#IYo;+5`DQ|3Ie0;_)145}F
zHb{q%)5H;Q7;J#8U1T+x_pz#LERtJaO1C@_Jk34+zE^@Cl5-4`=zY19t7nzBkJqsG
zNvfu!{{qDN-`hkbMI=QWSEfp_Z5RS@Fw`Y5y=&A6dA>0MCFni>G|uGDw8`JJ(<Il`
z*I*-J{5;@&d{>;BdYFcU<et;r@zFfDMN$JnDv)P&mX32znu*s!$3*q0y1l?7Y`+LJ
zm+7GfU^fQwXH5E=p@Q`JzDd}g&_dK?#U}B-%DCrtnu9X@S0eSp&JCnhV?ghG*;|y`
zRPpef137Eo<dOLN1}fpn666*L<l?_|xp+tC-Rtcbmqm^t2%pgmKC1VhsWeCw_1u;v
zSi;v90>BXk=dsznJtoB_-#Rri+UFl~YbRma*-!m{ba%<~xqiBwJJWJ61Ag)<%yDrJ
zB54M?0?IFkD<JOHWY>!BJkx&-UDplzYeNd6$kRR^`HD)X%XYt7?D(@38i=_|?)4t#
z2eM8-Aw`{Li1Vd_(}Af{QRH^OVnB?1oNp>UvFv9dQJ-I^K3|#<<#GoUL7x1&=9`kH
zhCsx@;QWENj=gw&&i`lo3J}b+%+6ObXX%ZR=EyZU<H8_YWa2o`KLa~XNDGtzI4UW4
zH}J$T1RaQ1`hzD-3{^9#c|+GcWd+ua#GX+i@BWH2i{L6CkkdTCnclMp?(w(ZK|hZ(
zbDYBOvE$8j1|mlJwWig=afQLvz4*;}1I4nw@Qi=u&=#<)RF8YHBya_fE8N)(O!N74
zHlfFy!Pf!8kD18t#$5#(hO|zLp^oTC3ZCPnzVsP-I3Z~H?UhAT>@1j;0klGAsm6CE
zOuwKytko-&M;js&`-6ZBV4iu;UI<)1bkU|bu$gmG?}~l~3UCH??oNVL&UJ4TinL3+
zH4ZYbLC}cVCWau0QPpvh1>P+4D-DOHs*TrR=`aXl@=nii-gIfV$Ol5nS9fMeD61dX
zIea0w25HE=MdZ*%ed244LtMcDY@&c7peNRwq#MJ~TOQBNX`>(IS>#8NXQ&57lAM8W
zxVtvD)x%h6UPk_(UisEe%)_?i>)_P}m!Di5xjpg2JIR768}Hhoc*;O<XY-%?{U0j)
z>ful9{8yX)a=o8infT?lzgY1n@#Pm={&KaSqy@j+_7^MuUW{J^>SsCHuQvVFdcPOr
zSCRVd8NZ6uZ<_j5q<+VYUqtFRP5mNJzhlO)0tNa_Q@;w;Z<_i=q<+)XF9P+OrhXNu
z-!%2hNd2a%Uxey6P5mNJziH}Mf%<=rrow-?_2-wh`^Adjln{8@oEZ&xZXN~>8vI=O
z?L+6#kRae`b1YaiGG<nGdkjn*4s<qQC2SV*yA<~ffk)8cfk)FJU?8>4TqY7H%s8|k
zXsD|HKKKp7kCrg^r0>HZ;_ui0HUbw<Vd*vUX-n={i$0Gl1Rt1DO8Yvh6qMe6vldO=
z^EAjn!IFLzayO;J($!2pjgv5S(pPuw8U)2)F}>+E?P}^)CC4uJoW0Np;{Bv2S?81S
z5DxMO(yA(0KRaOSib+2ykW>FhjUlJeYf$UcaMd4l0Y<l97W4H=|I+k_I$$m74sxO0
zV{Wxh6^jeL22pxmgVa+-o(#)<jXTcw9fR)kOQA>~oq8mQg(J=`svvLQLHVT8$_oSk
zE*)$CH~ikvkScBYIyR|FrTGpyBveMJU6vX2dWkD|=4T}A+*To|*adSI-IDtmPi#17
zr$4(Hc^^;wIZVW!(;NrPVnW)z&nZ+7FSDB|J_S)>@t@|PtN3I(O<U@6m}zI_9TMNg
zOQO(PYVUsfRo*G5ROAxJ-gb$5mvLKAn9Gn}BhI#%=N}!%m|`g4%cx+d-gca-`_zn`
z+kP?cAcK+U1P{yJcy$`+H<2EboCI3w?h4Rvg#UQRdQWEo95J9TGQ?_JVl6_uJT3-r
z{0SZJ;0{x!=lj)H23wT`OwR36&jN@0u1LRk9>JehMr167Kk@=k+HA&h-(m(oXttY*
zK26%`IdP$ra7N!sW#%V7gnJ^4=!+7Nj1s9A%q}@~!*iClRa`oRFOVlgx5ap5d_@Aq
zIHBFqVWXc&?J@%QHV`b!4@BZ1sda_;<P(148C@#vZ<g;5OXtWBPZF&go=kShFz?)o
zc8EPg&R0B3T@WO!aCOO;GT=Y;>(PHdVa~3{m4;JXq7ql)P{RFanVdz*-5NT*mb9vc
zbzvo5wtl}=IWIJb0fHq}R%Ic|2?2d^22UA($T7qJwGlq_YhyT4;?xVKR2x>KBsD6M
z;WUK6v~XKGaa>(`6(K8^9Ktsy+O1we^ICf}W;~m|kJAX3a0g`yAqs-j(_8=yWkU;K
zH*@REc4;)_d(Y427Awp5TF=+Zu4GTI%2;x@rKww=t9i9h4mr(vFWwn>vwCA}PqZLo
z$!TRV1Fd+3ALFoVZTBS)nE2;J1nSabT6x6~?|EfC>Hmh?UyUE{(VJL~ZdAz5T`_To
z3ZJy9&Qr)6HZeig72#kmsFG;CWEo{ivS|eKBW?NcH3+_Lbg$CPL_N*>+y`S4-RI3K
z9o;H)n3b-9%6i20%va}hA1=x}BV$O4<;pMH>QSN3P}*%uZopb{$}H-Gw7<KUK469M
zrEz}%dLvtsC|p9M#S&syqQl%ba_u3~G|ww$dCYFT+898US{cVN0>Z@(D{T%V13|(j
z6w`3q^ADX*QkZkw)hweK1LU!R<x4Z~+zjz4jJyW*F6Dj+YZ^r!A?5s=sig@QrLg0b
z1z@qp8Gd;R0o_939@y0bSSu#O6`>%hdpahj3TOo9UD;4Vzdn{XOiX-Nc!-!jA=`-o
z>y-I)_Z;kERmnAIB2;RZDXV2yo?swhgyfW`*TJMg#c>tNRTO|{7C#_J$?S8oOG;v+
zH33v8$yX7=I-gPaxoMwchf<Nh!?VlI6#hYm3M<$!vv14LwEJ5jLeez|sa!|6;=#ZR
zrh{(eVnUc)+$xRlD)f?$v9?$>U9t*U`4om+g9frCl@4=WuqXp2y`(KBqKrPA*d}r8
zXdUy253zh3aA{UyF7NHhB69THCiX|=AAJLv<t6jkSpNn~RruDObNo}W#q<D4N%7bj
z!&JKi*{%{k=#5e(Kdh=!HfDvc{l0_gtlaaYHco}ik|ad&ro4mhZw1FwK0S{o-7`0m
zGdZ+UHEe8)KlDfBqinOGE@J}_q>>X>ARjODf#xaPqU5rUws&b3PW%66AnOrZN&4h;
zDE#o<cd^&ugQOyn?Tt-tmjsaw?rMqwow!YiDE)~)lZtMBXXqJAh4!r;1I=615F=Z1
zZTCJWr^hV1$;cD}8e5yZF%we>^2w(zr5wd_?z5|^FMplQACHkt&NTz+)wvk|R0(D+
zR*mW6?|Q7ooF^~F$@3{Q>wtL2Vx)GN<!jC;nCl&PpgZ=Tp6&G8owx9Z%%&Pv3P~I@
zgqozE3ThF<Cl3zT=Ndo#DmTBMEU<uDr~g1J8KCQ2Q`yc^-c#Cc8RdByT-mV^rl!%o
zH#<{v%68G%%;e%ax$P|$2)hY^t{gn)u7Brn@0JC!#cR+5TETaZ=GSS*nDr63Zv%%v
zg=<^GSw{AIico6tKA1wOc5fVujV7~-K3QyZl?V<-Ncdp-dr{#IctCv!g-xn$PQ}(&
z;lUkYBzMB1V+NwfV9(D5{9j?u-7~Oq({_Aaa$~Q!@P~=~rnG$$pTlhzQ~#lj$5$fO
z6*UAJ9b>BVY({R2+D%`;9|6XjU-xbtJaajoaPIS?sMUirtm?T2eF_9~iSC!6shGL&
z-M_!>EJdh_CH5Ove72^Wl5x-BJvP4wnNWy66mbJK@qtS#+M$eWdj5iC=SFM<t>-v9
zE;8i#roi#5u-vb5b-mA!)m#y#IpCJ#(C1;#D#i$~w5)?L@Bht`g-o6dk%kUK43`5Q
z5@`0m32rWML<qR4d6gKrEyA$s<u#}a(2+Ed!cyru|7Qc3Csi<=7A0H0w^qL&m#}c>
zJbS{(I1YK9^BCSfPhem5a7J%ZJ&mvs`BNA8yVs%i5g*lJm6&T}@82wWXlhw>0@?8x
z!~R1@X!!5|x)FNdjTwDu{(7Jz<sEQN=5xlt*aee%!B(`|MQ~@3<?+=q#SKk+2WPfS
zQRS&*go<++67JFpA#?+Fn^>@|G)%<&Wk%G~XeIQmXawwilh!6z4q<$azuOG3N{&Gr
zGA%ri72X)P6DPbdHmpK8Pxcia8`npazXn}S(GQJeST#jWe_y{wCZsSdj?YoH@9?1g
zk0PAXlMR@u${Ql(>U!dk7r3CupXa@y-xL^O@@mpG+{Y?U<Ev=$0Pf<CZ}cnew`}+p
zBOYip9}?&Z!|t{E>MV5;APo^sN91E?H_G?%3NfpFFh!b|i-5cAd&O_P)+yI0!!GTx
z1|K+x3B*@1L1wq8ia+aHuJYpKP3bW*GLyb#KbYs#S4^=8EQ{@;IpR$N(MR>`;$sjA
zfW@++*Mk;VsS@c^@ri>Dbd;O06D(YDhXf3cG+&)+N8@I_?he2==|D@wg*=iHQG)k>
zou@~qDc=SpU4g2Y^^8sTiE%JLp1ePtey1m~Sf$Jx({dYZ!lm$GLaR)3g85Zdy*_^8
zwWD;VHvO&+2znIv>OcqLE42X~NebaEg23p}W5#VVEmVr&g6;?F{t<!vmKujS4?9Eb
z2(TOecIc9h&(y+7P)z?X?BrvN<Vc|}3>Jc%VLs=9vD1FnD7u&~ydy6}oDndK3f;-%
zy>si{y<zC~$+XYkts15Wh)eR^#?W|7739w`s9J6ikokQp9rExjckk27r^(fISDLs;
zTqYQmD0KhBJU_`$VVf+bLx?|UmfR2V20$5-+9eJgMV>%B6cf#D4?tiho44^avcDyI
zMkLFi4y<4tfn#Os)~?&ckJCG|CpinT0ejLBS2=lx)f284u+s>`@f=?{X)ax)$bZ_1
zk{LZKS%<j0W+s6GzNH%F+q8S7$j1y<@GRJ{^fVFCC}#&M;&=xd{040nVtZ08RtTFx
zp|=T1l4aI;3rm2!E$!`$0p~L5ltE*967x}`lIASHdkQen7mV9P#nyT6X_<A!VJJ5=
z`2MMYURH|hvSAgOcWpt3a?*OOLY>+BHVe5u;TB8bZN<d^SaNg;d`RK<-`KlL+n8@>
z&Z<dR;fpATn4b=0W&n$HnD=Z2c@GG%!B%c=ChM_L#nIQx)peP7ZUlIauiopKoL8tm
za;Zsp(U9?Yb=!{mp$G5m--Pt(esPh3zFdYo|EAX(fj#fybs~-}S&h^<S?Hw7b+_y}
z)`cbL0$3HCl<-C%rRPfG$E#h~%^&UMcBd{JL;Rkx#dv)@d$cFkG%C7<iFFgT59JW@
zInQUCs2)?gwZGYxR#ejk6m1VKu_n1A72tiD+0FCjh*rKuh_C_YUr3o&O+d=3Ju5sf
zM&Y)+Cv}(pjl6#r!0u0m`h`}YEtyx^Qe|`f%e#2CymKEYhSH0RggQK8PuI8!K?8~H
zdvdRyx9;^Fe)k)q?C06GSzc7nJlAv*kfA%3S&o2o1k2y{@Erq$z=~4=v|51Y=(GHA
zCK5rqf3Di&%oOi>NOIB8uWJu4Wq_1hw}ra7IlsUvMf*!_VhFkRU64#^vMd?Y6A!jx
zeZD{Pvkc++{8}p=17vs88(NJS#GIwtjCv}Cxq@hvJr*<RpV{q^l<e|E6ku7f)6nVU
zGVu;P9V3LvC(Qk~8rC?)a-jBfP97jB!^`i$V3us<MZplot4c6W94WA9wy53nIautB
z8m3@_?`O!wK%}G<s~B9H!{@vOrIJNxKhO^GEenvNfH!3J#-&JF%9sQ=_)8pc<~n+l
z=!W3TgpzB9mCtxRf_Mc3FwiDj4kd0Nw?FEX2|62ou_9U%eP^DX&i#LNC=V-b?i;fG
zO~XWDea9((eFQVc?0q-ZyOs_}6VRFacrZgI=)T=yHmf)`KAY7-r~1uqgv}J4mx`N#
zUNFAmd-i#EFySL1L_W8&{+Gsc0Ibk~=m|{dohLy+1dV&{f0I8VXH5CZtXF+Qyh-wh
zrq;Wuh|jbNA6}+77>nrE`Jr$vk)=T9n(;m{*Sj6Y?2u1#u_~@E<U14*BJ4$mRr%`x
z2>6EM0s2NLSelLdev6gCuDvZ*lIIG9d=-}oLsl2K->H3wgts$_7M>9nA6N`}+7ct?
zgo=oPNCGVARLkL{7;$6G<XbZTAhdMw;bWNh46dslkf;NB+&Sy0bzA*6miCF{G%GhP
z%6sYn>m?kb9yC8GMvtp%mjLoXy19c(95<tL>ddhdCWEcS0dz*i1YtWW|M>?H;Z@8g
zsArU_ntii9*<61}8=ga9<F{izXK!vHfGJQF&6LZn;y$>$9Byx~!;u3^h;A7bK4Kyf
zRX(jYNz_F8N3$nYXwrn-4@FQwK+ixQ9_+B{`NUSO5R6FgZ$`V7-DeA@rc&p)29fOF
zNF%FgV>mVc=me<qKvBGv6G5#LsqwbKktOo#FoEI2LApZ-LFwFccGw#UVNW)TgHabX
z^cyFn@5B8e|8XXO7XOtP??~6kF9*1mamgz~j<gxEXLlr3Im9Dh{1msn?0@^(&hHCi
zU<62ki_ZM$z!RV_P>4`az@I#Y0so~_2uLUpGz=OhI*T#}HW@h^yNHUR!)s>N0Ia(?
z_k=}_aF$>x#FR{)j@(>;1};<!K|uGJ4amRaU|k&V3sPl7{(bm=9|zMQm^nvWmu|fe
z<Zikl|L-4`UL1;$t3gzr*Afb)Jq14S2R2ABiwPdlK42pS*<9I54y~=44kRjZ#+RWp
z_bS<#t&}9Ih&-YtaOapwF-C>D><IX_D3tR0?hxgbe3ZN2Uh*|4YVY#NSC%oaD12-Q
zGKPVyf$r(ckI{eoh|&U{q8Fl1d<^CFmIaHpjV8$UNw2DWSMzr@z#r+x3rYoSFqu)3
z_fD!oFu4qW^)OhbA&o^M7<8$*IbNvr_PhEyhe9RHaV-|vS0{{{;~=mYA8$HQDZi(|
z&j#1?wa@qhi8J}DEQP}Qum{J6zHQ5vHhr@aSBz@!A21mAmb*yNiqy!%b~;9-WEo$y
zgxhnQ;q2lS9{EXN4@SI6AS58gr0be`EII5wMOtw-ETJsFm~C%Ngrdab10R2y%u+?3
z_L{xlkVc38ptv-<nBRP(5nHw>hj8x7biIb!{`jk5{Y2zm;n^{@7Fl|#b*#BE3wdy8
zlaj?uNVZp-%%2*G(qZ;|gZ4GwMpa*#B1px<keER(*Lw8Y_8yYJM?Zho3QZ?37q)(U
zwWj4q2yM%-t{g$T#wken62njIs1pSVPbDzTv=&yDhR&`*kFNxYu0gS%|NIJURx1y$
zrnN9~b_liAVp3L%ail{rVwANzwzn`~sLm^dj1H;cEIfsC5WNJ*B~+<t?se*@b%mx~
zoG7!b#vA7}2Xtf-&29STxPWBw>oFBr8`)XKqOu__%t?3iS}vM}#&hH*goR${I5aM-
zDR?=$p$b(aHfl9(*~PL3HmFIFxwBjK(FkT|u}Y{?*ej*u&pDQ-(CFa~1bjt}$f;zq
zYF`k{5UM^*XT?=LO{(<=alj@@;;6SV2)QM(i02vh-uY_yjpvH}Zd%VLi0QPZgs~D5
zgg94ia_QyNce0Q#!dRVd)+Z6>k+f-8kW|LEGB_*IOj<^qkGxeYWiy82+_JFyiP@$6
z1f12P_z;~+15?i(o^3J36H&rY$u+*K^3H7Qg>90}jXNBR-UFU;b(&3h=kcvVZ>!^r
zs<%fXy6&?h2O~t$7R#Wbv9g4=;2hK{wZcbX5`Qe<j05iTZG?BS1nZ}FrnsO$+GWGh
zwObc~-ra>5F-mN27T=gDaSzfqWGWG-)^Bzp7%y@MXU}%unxv8oD+!3jqKb~fbV{^`
zhe;cwB-aRpD3OVSt;QPEsr12<2CG4bx1bDyIg!nkDxozy-Z=Zf1W%qvjM}t>O6|b1
zPTO|n+oO2dI~Gu3u=0zJ0c|mfc5dehnIALShIRw@HJIwsR{GPC;}1lxKA}0o1;24W
zFoJ-nFuUXPGE?RiGTfglA@o89io`ycHXEwl<>EdXiZc1THziLjFi;VMHL6185IVRL
zdZ8-0e7I*)<yIFl0;sK9y&>(=pgxmgxC)9EaDAbvY@}=#UHU+S*jag3;KYP(OpJoN
zmt<etI9?g^6F-p|(p*FVv3@Hgd8bm<{RX(xmy$GitP)Ams@#SAl+XqY1SV~=T=kbd
zL$(S_>z`ZP%5di%9Ii<tKf&^=?MyExEvD|z&_@dN&(Y?95Dn<FANX?(lvk#n%7u4_
z^fF^V!GQ_blvM3;ufZ2VjXyMJ1G@>Fx$T8wY_J=n@+&heY$R$f99r=<Lw26A&Cwph
zwHr5Jjm~)_nXz>*iHKsSA~#131ql)oCp6g1GjcTyK{#WFteoHU^$VU*umLH<Y%`k3
z!iLHk7ayiOAUzXNTA>j=`2Zr}t;+Ly-=F?@b&upv5;Bc#s8sTIj6il0s7I^|OaV+f
z-mXZ9u<&MBc22WhrgP9XnOlOt&Vj;Jy0_pFQcjRb4>i`jA~jxIU7{=unA?N^LhLyQ
zI9jpK;@FY(_1l5sx|B}V&?=Wmazs6Y6$@9#P^$ee`n;&%6~PoU*Y7tD2%^w-4xxl$
zS)$?V5NDM<NNa8WZa<unSh+@!L1b|hDO5xhtf98;BjK+kpa>*rQiQX+^AX}OQ7|LP
z)j#^ZaD=GTVtX?Ye?MiT^NU|9&u-60{;Mg*Q{%5CP5eEi9?o}TT*j)J7UPhgx8w|E
zDLu>c!xywco$`Kwoa!ZaCE_7yhDu0+urL-I`%y+r>?%J2Sdg*MM5F^@99mG*@aT>I
zJ||@Toai#&2HDc7<o4vzYIGa-)@Lkq6xP7{@rU&8mTg?dmR*`vi{bW(_;(^dp&~;d
zy0DG1tFyYxLN*q#-pl(p?^5`D8j0dj5m5QaYVU>X;`q+N5Dl#ge}h4M=^N7UbaGYR
z>Q{KHyWT;GXPypiGffH*G%O4r5UmamR*%BOL{@W+IHFS!2B-bgcT*4tyZliP`1so<
z#kAX0u}zX^==brog?uY4EaDmEt4>Hc;dTV6Li-!@FZZ=EF@!mhvj?{(<V#Ix0)bFb
z0PmLX4?oX)Vg!T?BtsfAM8=ztVTR}<GRM+KY&`vG7H5cYs@lV~qq47>0ZWfE<_;w_
z%*IRA|A4usco~{18)HN~ccCvvVABk#f?p_F6}?F^MXPc=7)tf~F>6<E2Z<Wh?%0V-
z4%nQ$RhKVh^{3p^Tn5k2e`jDpTy#!!!~!Hdfv}4&Mp8_VhZ3~b{4J~<49&9bi6$Ci
zm~OW9r@Is_S=faUvDD&p_tmB#2SoA!H_pM=e~wSyH5SIC4VfnT6QBLsSAgeOEZfn*
zFCB(~hos@;0T<jNY}wby@uqbNtYKV8T}WmemO56al0}`h3P;=eNR+^s!c#j^%&__q
z@qEH{!2W4a`sYrvMoI%BxhSpgs{ZBg4qe_=0ZL0W5KYI=@+4o+_nSeueW4eq$^ihr
zhd6FP!7Vd4puU@r{G6oLk8C<;JNfA6he&VW#Ha7T)(R(AEQcQs|0Dv$3m$ED5N*&}
z+2V^X(9jya#EGJbrVol~!T@}1ESl$2(rY_}w!6Z`Dw=ca4|We?R(shPYm>N(Jw$c4
z>5?7=iT=U;rSd@$%!$bI`K<Q%O@DaE15t=x3N9RV7Wri{drP#dcb~68hk0!ml&E_O
zmdes4n_^tX50(sN*l76iTJLg3<w&czq3-Ladmu|_@yW~?$%qyc;pIZ&1h4#zc<St8
znh#P6=^YV}DA|J_H%S92kSH}O7IKoJT|}K^jN;qJM(3sm!<sWz#Rt!QkF|!BTSN09
zC)G_#ZQjD^W@N<F9>EXpeG>Lh{WAooE2T0~CqgJTw5R=HFV(>g3zWiY+;30;G_7K)
z-F?*UUn^V6HLADH2I;4>xl*i@O(}_U_e>D<KH~&cJ~+J7oyyblp*U3fARQd7YMY-V
zt}T-N!w)j`UfUe)Mp`7>Jmjx8Zs#V-c04>4{T|Yuen}pV=FuTlQp|t`I5Zo~cKyKr
zB(nN>^0o<=;x&yJX{~rv5q-pRC#~Mmo*0k$)J+wo>cOAQ$L2N`6W`%K6`XV5lt1Yv
zgn<cj2A$fo8{Q!udx4xFs7nFA9>W35^SzKEg{*C}BM~_=PO!Y6U<}A@0#>-^_m4Mg
zcMzudJ#*a3Fr<64jnO!&A9<{r=ERcOqf=*-LE<3ti{ji5M8FxgnLO^EezE@aDI&ZQ
zF?F#q-8Nc<w)@@o&yUQ|l!ZKOQl(9-X;_(yLWH6-rf1+&IZ8PzKp-swWq8jmMw7U@
zG>(QNcH%j4X+tIYu_#_*<3v?N8*ryRVwvIgf)ZmMG^zIz-yxdUWJ(U$@kfPgiTZCe
ziYCE4kgiLqC<;J>phNQ)4G@z8`r!n0w7}|r4nw}5Ft2w{1^Mywc@IM11Y1xemaNEo
zfHLt7X%mnQ!3sy9c+v#;M=7I&&=b`$OJih>EWva9V?LD+hwgnJGx~ck;O@bv2RktG
zMbkjQYF(1$i!=}fdS^G|=#3ivAF&Vmb1x<YY?|rNf%~!`VW4k*IRw0C2LFl(Iusf-
z1{n)0CKkCev*=wVBN5?dm76^>BzTV;pI>rc_W!gu#jC*5$NfgGZ8N)JZ)Z`hc6v=7
zil^vRe6BazB^JhfV+(8zij}N4xkfi@M%i^@2uEGGxO{AKQe<!HKdL7AU?cEy$_w1W
zZj0LuTHLN6SDx`!V6>~2m#Ek#o8&T!dRw!o8p#`LV+{l79T>7qiY6;CjSUZIoDmkl
zcEL+@qHb>VB0XfT>A5p?k4X081GOAMo$%<BkwrmVcH#*gZSM^W@|;%jyZ5?UWGd#c
zJxBv`q4jqA?=7|+Ri0&y46j=(lQjzurD+zyF^><5?kz|V7?`+?BhT@n`E^XVy&ZR}
zqN`<n$2b3izI!9DS*JiZvi6E`{j5G~dCxE0815nf?h*R^Jl2$Y9z->#!aF9F2-V~j
z7J8g(&RQuL&%Y=}s;{~X+tf{Ew6-0-HE(;nbgq4=xUu2%*0rrbspORNt~l1(Czb+W
z4vm{^ER8xNOLdzXeT(yrLKP!Y{?*hz@*Y-k0Z%#GV=-MFaoV~1%lVqB%cpG{_<SeZ
z`P%9E%*L!6V(O4XXKF8#@b(zOZ6ZE|eX=+d%F7v?3OE}mdC&aT(~-i;g51r{^Fi;d
zy07xV(!wd1mt#sJ3FK79>C(`O^`-5g=tTQlmTCF<Jjayr5l9C0>GX<~!!$s7cA3a`
zJ<o*O%EiOQrJ}WbHT*CE;CEFc-g157dZ9?0fwaoFyDKl3dz!&`CK5}7k7*-)%$3*t
z5VNU)`|@$P;N6{7`X#cN$b#y2uHC{Qv&rI}7ilVJBY@iFeBNtsEyQfc?miB;;dL?k
zj!k8C<DD#3iyG!&ad$4T;x^pJoI=D<XR~NRgd>X#5euq#x=QHF69y@B>7h?h!X(nC
z(F=o&s@wSnA96hclvc6T=y;8CjN4Y<u2X4NS8-W>MSbGTGd-(|ca*Zz`dB|}N)HTq
z#lmi>gAfU!iFLz+jRTw#Io{p(Wz|<NXD<Nj#B&W<q}H%hzn^RgR>C|L(|R{dFIQf#
zQ+}hYdo%yzV0~0-Bd*xX&n?m1-*MU5ty1<?^(^nT0gP#W%I@<7cUCO&)yykQPkz|C
z2H6aGFj`@5lM}NHO~mgXX*zSVPpmpG&8~2i)e#oN01M5vTZGy>wCpD(mXT5Y9Q7&e
z1b&J{YPa6(ERy$A3k3ATrLGn0^+RCwCahQWdQR#SHr=R4@v@9K^Yq^MXNVM;uHmHc
z;sXt4(%#hBF2OQ9sF-?dAy);AQzaa!QMgWunEh$kaf8R1sJN<1<gMhIjn$v-@VMY*
zuPIfXc%(7@XB=3Y-CgmvSLzk}5zzrS?HM5eM(z8h(em@irIpOX;3=MDFjSNIs-#dh
zYwz_^tLsZi!1fibluo%u3DF5fd?QA2b)6|z3xaJ|4E}7!T+I6r$Mp-~ByU}%vqFO?
z2DbRv3Qyb-jol_53~x4ypa90gxVs<bj8{!<SnOC$zM!Y>L5xL+T$xE&xBc2>s%}Tm
zm9ZXIFXAQ25kt6L#K<4EJ~*p~yM|NV-=YL-sy#a&v<MidEPu3(k;B3BkGU{qLNw7>
z;Bh9AA^?3g!3%0{(hDnIE`=RBKjdn{TYAK0nSfiq>Ds{7RBiv1fca0Gd1P{wDm;{d
zvLJE4wt=w8pI^1)ot^!tt9;uv=)@nYx$+|cMG+rMLT)UCjl#u7s%15WJ1lq}Cq{G2
zn9qq93&)VBHJrKh^d-++&j&2(pf8bJRk>bHxg1=UFOh}ZE=3)R9vov(vPfU{s<`_#
zs5Fd^&n^g6b<3I02UO~^-c}{r%PyF0?bR2R!496thcH0~Ejhq(0nVZ}b2&swmaJ=t
z-mIZBo=W!dB~VD3sW+?i7GjFNjQyq-#VfnbHpB6d@RFf?>Zq>&Gw_PJNL3$r(9XUW
z6fkoXZ63TM!uP01Ij2HXJ9EMfj+9`7pUY*^qRP4=Jgw6Hk0#pMuRjqCh!155n0x7&
zd#NMUNu*+aa!K6<UTUruify9Vc7Z_UXbmsPab2kVmz4tl#j9w~&xF!k=z<p<BZa|t
zGXuY*g#-%)y1knj0uls8hR(uf?+}kcewSIqG4Lrg8mq9VvQa?J>o+Z!Sof4vMvR@7
z#8eLrbHCr+3@ZdFr0#SjQKISo_WnLjf!^|XS%c0qw@8(jht}%z46^kdzfb)}<F{?x
zx<fDD)T}<$%pvE8i0Xz1Z<imx!hGzW_Ne{;ze10NF%h*4QH@-^(Q&ytod`EMWyVcw
z$~gw{`hp4#3W3^7ECfZR?}FlF=eHPpNi7<8+0Gc_)w~D^L^&dTG-rd%7%ZIR^X5MW
zHPlGE)r@)O*o2)d$SkYdQuR7ZUvXcik@#v_-ml;Q8vksd@BRMmlueN}@$^py|ErQ=
z89GrP0^_(u$j{nakx&%Q8<G88^V`NAF_=}rZ2*61^AN#y98R0mn)A43GW^PAxL4|a
z|JN6<WJ0cL2bX00W8SjA@-5j-b26-#mfpgD^>T-9$jiu{qvulns|^WSU>5!MoKaSv
zG-tJHHiv@;djtLw@e?ckJnAF7&c~A<^4Fja*+!Omp6KSf;NvqsWfc$P^31$!=4a#W
z-5((srqWQ_Un4Y!s~61qoM*@9V;LjbaWAgi>YGKzV^4Xe(?L6!?v4UU!ztN|Sl*$E
zp+$Z?;SG9yd&02kCnQK1>_oG3M*Y1XzfpdEP}_ug((QfBSlvj?T*4#WOZ>4zm!f~i
zxN41f%j+05+`=oiUfgm?EZdZm#HiZPt8(jh(orogqj;lUS~*(bS-jhiMveSlI*D9^
z5PICQt>}sC@MsC(Sf8#@T52Sn*@c9O3<yjTjV%db#8NfuvuDcI9ispPvexiIbXYKp
zD$;68y=UH6T@VaI*u=*pXG2nc(OM@eNu#!c<7d<S#bR7={PDWo3pDRs&HGE|>!RLr
z-`4eRXKHHy5|71Mb<gD4MeAcLVi^7d<dY1Fhp{J96>v@p83T!lBr9L-%GVvNm=92o
zI<DADI`1+XAc)w&tjMgEy<+{MCB|*lra)jn7Mjk!RD>M9h8UL7Rs1GhAcLiD<|LDY
zcX979zvuonNItpD-BCW<@?(#8L?UGX5d$x)Vztj~21c(9TEh|%@se~hR)$E0yAE@`
zDj&>ggnP!8*ZPOnSTVG!!zQ5?eO-mG_6^x*ffluT7fxS)SiBVjXvI-k`LOZ|9S8I%
zX0Opyv1-U&W2JJ5#80ZuiduubGbbqq+v!~c2LW+B_g>O-Z6y?2F$A?V#d|+QFjw(B
z#(1Z#XdXRqaAw#n6qj0&5QQffR!IuWNV0K=A6C~2cM-;MGG8LFdur;*=RgYE3<u#C
z5pLhE5GDJD>!E%^*J1FGRQR$ctlD0GZX4uBk~rgM3T%B#)?@Jki`qm5t1(v;y!@#H
zLDBO?O+G07>oha&3#2PlSJc!sgQ(%u&SraKS+mRn)2FC!2|Z@Fr3ko<omb0dBN5cB
z+h5Jfl0Cx;@64KX01UhJ8l>A5Fs#6V11}thkaNc9n0W@vWI%@?#Y0UypLmrbXm5tL
zI(jeR!V1aGB^dP-#o1l~$1FMimC2n9Ml%F_g=f6-PXRzOIcOC2>AtjwWXir#^rl=~
zZS372%V@z&5(^k1VMalra^+XV9K>>7jr9dw=X{lq|7p||nbu`EkwT^mJt^KDY!(dG
zk94`FLyuEPb}qtZq8oNk^P*Q`UO$;5@}oNWbT7f;9Cd&XAC-DX1MHGu%{$1_E8Ir8
zCv1cm7r&_j_r@0R?1%5!#%&v(_ax%^KA!6F=$*U<nRu0x`4u}gze<XF=DTEY^i&zF
zFl&gs18c^8=Xmb8gQHMN9Wo|3+DT2<^bPIR`rdECMSZDvHB=`C^O!J;bWFRfR@_SV
zGY#$t@IN&S7|VR4ULQTX&_mXoi9w|mH^VmjgZ%_7i?=VKTv0T+MaXpa`0gtP1SpWm
z7pjIO6tUlWrtx`5RI=(LQT<G{9^!I+e%9iler8<p+OK>@4?|9Za35t*iDAweL46a;
zWUS&>3=f&98`;MM=*++Dsu_E?{pFpK<R~37TD6X(#S_I)r{&c7RIe)GOeAV|hbit1
zo^1_2GC<u7c=>fq@>b8g(s$r44N-lWqZxZh4jAq_e?r;$JRPt%RP0kcJGh`j=KS>t
z{AY&eg@O;hab?22Bu#Z?ojZ#fyCu*FlnW_ZjT9M0a=e7EYX%VBHTp_1XsMa8=-+j8
z^fGcO1FIJ>c9pC8J_2=wSH7I&t{H+j{F2pJyicX2edg1rftUbo!W28I#`C?49BL7C
z&jzb_7HXlbcR@y{)-2oST{14SVgM(=!sH&Z0{AgK&s-(cT$a*T<(iD9D&c2WbU}*^
zLdpx_VaQOM4V`>^YyeGC;a;ra+V+q3dgW-K(ymFbS^DIfX&D*lu!cD+-dL<h_inO%
z7~G*S$syQtas2R6Nt@Z~1YzM~*q=W6ICG}FVT_YPB8b^TR(06uH7PuI4XWT9|1x+f
z#`p2%E*ciXyy?}V>!q`tESXztXH>#&#t+UVj_O4r$V~aQk%T<{e3HEshTFhli(Zc(
zgO)qS=)0Y^ehDU4-iNmbI{~|E9<QUKGYpk+pR*@L^l;-On$S4Ur}u1RGV@06Uyr!U
zM{b}z;vIG$W1{q+Zo-!7(O23ZAfHbEL?{ox7_sq*(0$bJu3{C+slajgEws!XaIKiU
z7;dRc;+1(6UxW|4R#_hA<ns*t2;f%@(gaUO&7PWXVe1)4R-dLnq`17@WPbmFIseME
z6gT!gCFTIhflto<ooi6ylaRfNfUBIr#UgxqViM)fGD8}on$&yqV1j}UlauKTyO8d(
zfw-`6_C_&7>2IT%0v552kl~_ED!TiMY(D_tCI@et6Gw?c7(v}0j5nE#F8K2I7uPI;
z+C@vKJQDfV2Y+z&T)1n{<SQC8<ohM)5eDy=)uV72FW?AlqW^$!lUN!}&_D&1L-;ZP
zt<GaImCJ93>1KWH$stvox%v;DKY%KQJU@I&9{K4DN)92|c(**0yw5(gU>DJNH5WZf
z6&3AsEEzTJX9^+7BsUCNikM&C8-vO}CG!vlq8)dbxyL{uXi?$d3_k9hEgQH;E+xVc
zwS8|W;3t{}E%l^|y$&2UgTwRBD`Y901ahjo-{F)V6>ge_i{)^9wMOpvGS>*;A8)wD
zhjA|kx;55T>6{#~YisX({1{?uD_X-v{xFSlCf315L5-nfZS){H;9d#$>;^Tdb|z}6
zV-<zMl%fNLd}etYcs|#l&vZl4s9oX}D(1}8Ayj6|oZ+pQJnh}~nc?w}0X;4WUjsP7
zbKwj#H%hcYD@Mpzjdj1m&lqXh8-ctf>||c~)W#pp;u>@@@rks%QK0yn*zsT}rQ_~)
z#CO!2qmAG{5KT;9iDHqi?*Q}QiF&1hY4!At`XTC}4KAs)do@N>{@QKq?<Yk%$}s(6
z8DQH*V@5SX+QG)#wU7(*UI2Y1ry^<{oKSE!VR^cT1OhU&(UdFn)QyDq;^b$kKv?{@
zMtC|%!6rzg2y05a{&nt@vafMeLpD-Q!@my(KZ#cKdhv*Vf}j=4&+v%3aT5QFS?d!s
zUP6YWArBjmF2GI^zd3<KRvUZvNAAqU$n@gi1@`Yd)>jF~z-RDC&9VXtv7@C9Qi%nh
zh{xRvj7d`7>!)0%mPuqa0LaTw8MD5Z<ow@?rS#_MQ~EBoN!)8Nsd<BYI}W7!xDj=U
z*z-C?ow0SqoXf@>uX47%w=o{yk!%U3MhaWW=XY+-#7do8=EPH4bdEHP2u3B{pvF#3
z214SiaC4W~_u@>r*`|@OrtiQ6g@DM=wd|ms$MHj=7fyN&(rKs^qB^(@iTyxG<VQlO
zzl>r!?WZwVg$Ov<Np=;qUUluO!=;%NR00Ubhz<=|HnOEe*`4D)6KsLR;&^cb!86<-
zW6$WNH`2^z<kYoq(u?YG6XaSZmAo86pKX~bu2K!$<~b>vE!cTr|Dj72l~jE@m_Q!j
zO}_yIlKToOM^e64fS+5^4}PJm7e^(DSdG(hl7Y2I6)M&{cwI#C)x5eN)zwFAS90F<
zVg%|Mw5^$B#*}ds-{ZSyCR~I{rAph@GuF9!4PxRp*iO+cD`B<mfXv8@yXR$?+QqKA
z&?0Cb?2AD9PJbJobc<*R*q6+Y+9!qul^ZkAg_w-8w~%+@9lV5R^_1DFWW5cpMy}T&
zVeXyL(QrYU)Q#jOINOsR+Nx$EDhsjpH6RCE10xSzXauR(pezlm`#}0VY`#H|3R6o7
z?&5glfwjaL$K6YUr@qL*iB?oxnm4#HRXsFSKbgs+MSh_N>j7bFb5S);pNlS6#D+SA
z_kb-&h^-%tN9%zlcv}#1cnyj>gV%4k($@lP0-T5#tyBi)<a4!oMQnL{8MV4e9^>uo
z^nXV?fr)lr=CeQWD*-R`;iAVf-<nj)-|?=Er!*7L-C|82YgF{^1tQjJq#}n7bZ;;5
zdB(qdIqgglb=+;4cd0v1pZ)6HoWQnXc)(R{*ivdH0^gF<9YrzpievVA@hw(Z;Ax?!
z6sV+E_ps~k-YOU2z0sZj#vrc%a+pV^_dy=>_mt&~$|6_ckR4U;H2<(<e=&<2vn7GA
zjkdyk4h-qe5Kq;yntcv0QfOQ1-=ZUU9@<s!B_qCyQ64u~X>enGQXNbf*Q(cwD?(p3
zfMLsDLPSD6?Wgrj9W^s;6z|3OZfVI0`91xyh-!m}h(=N{n8izk(6{m2WgVKgZ65b!
z71Wcq7;rcO+#Jn4V$=*NY%^6<3v2(7!a=}mvIGbah+3;f6mB;vq6IPca-!JO+}T9H
z%TbNej{X;H>F3l%l-_AZ=j&px{hBBQM_9|P#Ji%eYQqoI754Ee+Kyj_vM;M=cy}=B
zIVSel@0qsHnl)|or~RSV=vUK)Cs+cPW1nt<&sA--H@0dXUPbenAFJCT)HzKiM<ZrG
z15X#=?I8zwW2}hKFxDvuI>GwZ=pMxqYdN^j|I?uvuL?6fMwI=f2-`=W3#T9VngQ3K
z)3QDu8pFIk1xjJ1a>K!n58^8_JIm)%=|xV~*PwK}(|&1Rc107IJUs4p44gJM@sfHz
zwj<cHp%TqS@-v!GjU97^TsQj?J{r>oVsd(TuQR&A@6ZyEq!y}&!Zpr|MGX065pdu4
z#2YB$-dCMx4jy!Xz2Q!87QQiR*Yx9>8Hg@ZAvA?;QrGm(9}`T!Z`MbHVU3s(K*Su2
z!q09Ys#;OpY(mVBFGS5ur*47*P>x;I&rlqv92)^{s1TzaNi}HLyW<ew7JN|p0AcGI
zWW=P8kCASc_)W*0SoBs|{!a#-52mucvsM^)9O`a2Wvy5qGCcwxA60lb@JuYzsr6ta
zJv?=ess%7Vyi{e&Ve$?4MN1t{c(34dhph<6l$1(h?)KTgjEEQbmqBEn;}*RoEmq5?
zfXt)a1w1H<0aldC%5PPXMpWrF4ZZ#Wk;%a~PRRCPOP&%UYMg|eigP<5rg4M+i@o;%
zi(*;VMtfjjhCJjU4;dxR5G2D8Bu4=OMKTBoN)`lh$T=edf<(zEU_wM4f=EsR5=04-
z1eBx#BHkWcd#&X@|2bzr_nv*8`#ir*(_LL%RsGdl@Ap-h?yBSOzf1LB>9;>>K@Bj3
z9!E3eBrtHsr3?OKYc|-m8z2?d;(AGg_SWG`(idxB*AoLW5@yK=Es}h&VT`oO|7Z~q
z%dlt4r~6`V$s}{>bxVr2J%z@xSsv#&s^b5Hv@{5(V;wBw8H(K;T4=C$v`D-ph}^ss
z|Ebv@{8}+lY+6(t8uaGwlyj5P_(85ZG&$AW&Z9KOaqk5E_NL*?xDXv=ly)Wrx!`ww
zbRy@cH5kbE$Jt+z?`Jyj{%N1`T5Wm!a;(HnBqHk>lJ%Mk-uk~SZ`SsbKN)U{4pU?Y
z8&A}BG4sV|Ov*?~eKIl?N~X}l5jCBRqfh@2_VqtCygzT;Xc=@SOAqE>68Nk7xF(hq
zc>P<i%4@mQJ@9kLGiDB!`ErSMRJBRnV$p?~i=BR94w_CI7oncEt5RPru|ELgrSNNL
zt!A;ehdLR4hDSYyC5b@!@_-OZ;cL5zr<dIf33<?))hbCBjXI_OiSM#6Lxe(e#yZ7I
zdHzgGhbhI;j{~K?8HACYpW2okQqq?7@#4r2Ano~Uxx&q%7USWT*OZ;M3xyz&+e2#*
zFJ9%v*a*A<2X+X(IG^W>-0JUM*!%;;Zib-aT}m4z;ewXm&VzMQ`})u=9fLAh%78}T
zvYQdc0w$@*$-^8X;@YJ=<ax{e>-k~1uA@%SQDd!_?)#e+r{|X?iosEy>|?Up2YPlD
z2h#q+U-R2S3ezUX+2QtFoqdm=@qT7xpZ_UBri2Z3Y8I^P1Kr1$gboH+P4#iF+;ViC
z&un-Hwn1_w`#&qqs!@Usd2THd4v~`HpKahd7vs!-bSKd3e~MBgN@D4wKcf5g=RUw5
z<BoJY_O93LIwha5)&&#c;J@#Rp&0ud-*>9B#8qCyUN?Ncl<&N0rgGWW)2)$s@tLy}
z9C2(7auGj~)=Ay5HGV16EZDb4ZO}Z@x&3cXfH~R6NiO?5a>^uQjXOJew2v1ynZ+<{
z@<s~j(u)aeBx%&H|6^a<+<g)4+$Tz*bILr0{--peJ*5oS$@u$*{w5y(9Gz)6-%Ik5
z`;5nI4E_-tOlGZon~FH+cH0Oe12~>~>CfHynIBp$u|a)<A-u-t0o|kj*L%aRnnu_u
zhdP~pgXagD>MtnB-2Fd^8a`c5+WN%QV{&5j*zY0kUx)Woz@ObjjwTTzB?W(SJ9;q%
zd^Lp@;4W&}K%p_JCSLa5VQKY42<)cu?4J_|sc<v!(yBf1K5<82w0*MirUkwGoTs%x
zb=+^S-$?wooOm8Gw7!{gDKR+2acZ9P9rDZW%SRnb&oKY@dVU#84zS|LZOpKsbDuM?
z)~Sw@_7+e<TjJD16@xaw1vU9b+mG%#ZZH#3#6F+#jm#hog~}BHw-W+mNanOR#y@YM
z6eW1JIcp4QN$Wn|Xmq7*rVQO0l51#T=o`pD=&<t~*i#^DLcoOoOzp2{{psa8h;)Dp
zSzn4tQ3jWv`uWhmf+<0~P>m_2)YkY2V$wh39NmvjrWn-T-T|UZ8-ZWrU1iTwOO5^G
zx!KE?;!xL)D}7mMVhchbG_ULjE_{a9tX|2bXgn9qpKh`Ig#I)G@;&%cMCQCxAZJT_
zjFcbE=SDb;?!Ee5*Ie5KRYl;T?Bow1?goXQ*=4eshI)--6019%hJkI9Xrq#7Ug{b?
z#6_%sQ@?m&wwZ3k)b0D((aZacT<V-LG)uIc-~6Vj8n`ROP<2)Vfmg0d2+L2;Mi2V)
zb+)RDwzVuC>{uH!NR+6@Cx7@lojlqZ27glQw$onnvH3f4NtHdqI5dv>+|>rIlZcs$
z<wuNpwyF%SZZ^}NMK^XO>$i6>%U*}xw4WXs48E~@lKGT^lFkb7;a#D%8e5t1_EO@G
zW^A{+%7@Rs95FsK?!c}{&1`0%T;%YIILDnreV6!c$ua`VO+m{=+aI_OFH(3$-6AF|
zdv~*#3)Q<rscejbK`v1fZ?UJSwH1+Tdo%Cu$X=6;{W?{)>_H4SZzEw5d*~yCqw@KH
z|IQ;U|0Yt}WV@*KowfCLr-PM7hOuHwgeFr6@(S^rg4kVpmT2|F27yi^{g|GK7)r7*
z<e6k0;)=P!L*Q%0<6<|vRH7R)pML$5jMPeNgf?zV!0mx(^-r}!WP<1hxVP^V69T6t
zq(Xd?rkBG*VCl=krg-q@)MM>2qZu=uCgTuI$~^ei6ExZ`9tuItG26!X3-7qhZ0E-m
zq_K1kg`jqzF0FWD?TD$`LnUgiejWlpxKh#D0O#!BhGRnw!bzR%Begc4Hxq4htSnOy
zlc*N)j&{we7xx{7H}jHs&0#CyxK2usgoL=yJ>>x^lI%J<jq`6E#GDq7GhFQ5?nSQ<
z<GE)hD5`P3BC+<%2QZOvF`-<`niW6uAd_+zKA5P+j$@eV=zFgdIBBFJjr!tpq-t@e
zrq-;vv6xn0V%H3H4DVpw6*X@5lUsg4J8g4@V~pX!Qp~d_gpjkJV*1WA?n|kj<c=AT
zd`6R;84Ry`DE&z=*zftvqPD$ut!LBV@~ODRdwMhaIpUN}NFbh~g^4jNq5XArkKOzZ
zr~n=s(}QPxMNU8pDRB*yB?U4mq>_}XGU4Gr0OtPi?y%s9BGw6cL+=ZBO%V{dH|aFJ
z9%H^ZI?#DEPgz)vO<Y&@F;lxU7y8uu$dPj5PI=gYu!f-4U{OP!H4n|jM<v{+Q0#in
zwh=EeA3VfEUI({^@Z2cSS2lbnooCVYScg5CNtjJoJZ0cGUZJvGmgyz_psV2Q+7)({
zJdNP>SAHto{R--H1k<3^p-LioE7!ie|7v_*Xc%L@Cnc3Oj&yDy@TvWIfwTP)1xm=+
z4|-iPZPBIy+QEYj>RiTtiYGI_w#Z7GviTW^oc4PrNgnEwe_A6D9~`#H%t!Nrj2r(@
zrEl4E)<TGzwxbnuiYJL{__K0_`M`IT+lePJe&-k1tp{2^AfR7i10d8}!9hWWOD`fH
zxD7@O46L9<Q2VN{UVQnJI&5St@{k{Z3dsvy3OfyPjuyL+idR`NB~+JVn-WvbZVAgz
z(q?I(*HbPWYuC%(VX4<(E%jAV)p%-n!sZPda{^LYRvy%`mLN(Pt$W<<`2HI&WMa4A
zqM41R^}$WtysfUA8<lx`{qrwy8-;4JO#RP^lTpDo3YqJI!Txnk-sfIt?phOt`ElAi
z+>75auEbLF*OhT$dOx3#Fd6j@+Otlii0F%KELr_V!aV+$3ZsJVo;s+D^^<qf_E?|@
zF?q#Ckn~h5QG~uoX4ti<51Tvjj#Yqn&lL5T@d+O(y3r#NBokYHgg2;)yk1<63h$L?
z2vk98h&lCCGb|wTDAokr>h|tzT5<`Lmq7gxtiY_jpZNI~Z|=6vhjlJ3pJmLms0j1t
zow^<VF@|=Ajg9oP@KzIhZa!H<I24&`Ge0<cFvt7u;?1JExH;-}uB{GHqC6k{q&y|j
zVyk^y(5`h`n=HRgX0i#_+*Ep&V}{uSzc3MJV!(xNQT&9igC*@UX=l4+xc+MmciNaH
zB-eGHSzfYnrV2SDN>vi7VL7j${<gj-<6C?j^?L^ediGW?u4FS0D>aRq8x8d!ImS9n
ztrYJ?07j6&x<VF-y}RCj8rku=6beVClE9DDH8!agb+h5NrFK>(Kt$|Qo_n>|RqbF_
zTUOjL`>&w;@%CC{ZP2<+H23?n?8^IkQb$BYjZqomCXC)v4Eix^_;_t=g*^4!>74Lc
zSx;-yj1?+H*F9lZI?WMNko#!>q=9g+XH2K05-~Xg9D=tDxv)q^a%1<|V`v?3;}@;=
zb~-V5^I!UIPty^s+#ev5e8w&57Axwt;K>!ne2))u)GITVe8BDG0AsypB`?}u@L#XM
ztBxq2a3z?dE6PM_0EVybGr#wh))tl*Gl|o8I%zk;kkN{&MmX)`NLH@}*>Npo+ayLW
zbGa;Di-I(YNPB}SZgYS;by+#^?xG+_xlQTDkaGLQ4ou=crqxs~>Wi5T%CvamRJb|&
zr@v~baX>S?fefX=9ZC0Ier42d2abry77=rQ#{Oa!lyrM-TWeg;TeUZm%R-ati!ILp
zW9Zz3iB!GS@Wx;=mWDqQh<J7F39b0BAG=lOJluJfodFYaXJunW1j;X&YCZnqnbSaR
zt7K46|I&*n)1oIvJFAZ2F?E(TbIn?v4sw(tq6e>oyU*Z7wgvRhjo2wtMT*~AN8Voi
zON|+n0&dA#w@5G|QN3)y$LiV6`eg=g7>`(I&8l5EyN)Ih0j!eM-$L|!*gE_1RQDwF
z8=BJ~j>I2yQ9hR@W!BlgBz@ETUJ^R=-1Xgk$GEPi`VJ-U9UOCsA79SG7F#|OdsZWS
z9aD9MT~DMYDyDw^ifXuge$Nt5U|L0QfP+(^aUdI6TAxeFi~7TIKsKwTFyy7=$hd74
z)b|HqxMGK}L<QDcy$DER7}B0~b`cSt@GLn@-C1$u@VI)Gp(dp5&gNfKEn0n)P{W<r
zRtu7W#p%O#|5|+iLx47yDl@V%9QGlbly~5KYWw8|6RCaFV2HSY8I)92A}l(MXp!sP
z4Q$7P16hzv`)U8WTjo3!E5ofG#>N+-9$i^&{&X=*k~4N5r%>Nz7H>rF4a$@F*2eeo
zrlaCoav;2(b|k%YDl`<e7jsrfP;MNp*ZikdKOuBwy%G~n8GAm|5BG99QUl7eY55?L
zY6vWTbLj>(X@qjE=~+HRgPDPNkCLpQWKIXMmW^e)?&dxpC`rwm<4Nbny8YCKUC+5G
zKQm`lsf{ua;E)C_^x)np9J`)9R@pJ<gNMl18z)=d>^9133!;_kDAmVWH7qS|zx{hT
zUxt#&Qs+~`xz$_hK6y8ad>4zYS#^wota)q}cVDvzyU=L0NUu~=k!w37weJKzTf&Sd
zqWwhq<}6NO+$cpzO4;N6Km^|93MmUY1ej6Q0ggGiMr6+VTD|a#mu3!nJ1tUp+W?v@
zb6ke|E$vxA5^rtU^CcyG^{+-@Jm-_VqQHBv^hK6#hM2r&qwuZx%9z{#@R~RFAw7FZ
zdx1qbx_!cD>lDN#?smi$EL;!g?%=6!P#O0CbD~UOq<uUc;oeu6Yw;eZ#^k$=6OFiN
zYf2+ecE~g2_k6x~55%<UiV>NJLru>e4|Hn3oqy#1?COn{#kN;Q9#Ny`8Hah*kHfjf
zT8EA3sK=+p`B?V~E(q~Y?}kj@ssB5T9*eu0b2dGMsdvPnJ?s}9i#m7g9Aea8s=r>k
zJ~ZocuQoUJ2XH-+HhG7aJqjLEK4MW9Z<Ez+mHy$hJt#U@n*ph27_X0A8UbPs#dRWu
zEc=l7*%iAPyJXaiRo8qQGmP>s2*1;<7Q?VGNGw$hYGv5_>sV_ZVy@HPKIptc2bsUs
z3*YO9a+CvDd>j$LPLde@(8jB;T8Z`E(0@1}{p1)p)Ng0<iRZvV#D2+2Q)`Q9>y;Ug
z1?{#i72ouUnpNizanr(Yq+PEN&)nj_>o1Ba5D7cW94CWxwUguuWC_ck%V0T<VHt}K
zL{i5kARFt3(hP-F^oGHT@?<XlsR(b&7CpnB7dgpb2vNuJ&L&K%>Eo!c=6M#!^4k^^
zXjIGvxSN!tPWJyozQ9x@DKzGLeL@05J9q9*;2yZfDUDsbB`D5{(hjGfQ;Cr{%lO{D
zJ(<QL^9i;~_Y{D>gqirC)h`pkz|@9hl#)x1ZgoFH9P1C7c~3D8#TV%M`~_z8CfD^<
zqOW`#4;e>^7DLXBd0Usi$FN6pnFISK<9m<)sxKU3?tGF|$j85UK~5E8;zvx>GpPaN
zQ6|+}hdD03IDb}FynfkgmKetCNEc!=4&zxGR@=<jQJk=RmU>xP^vEvn6y8j^qbDS!
zW@H!E4Ay`pvf=cx!S=;_(j3a#Q0`0k)bf|mJHzARF9fy5pMAU+KdW$TMOtisT#+k7
zeJG}zA$3mdZ}mgbYI85H%;(kX0R^FyF#M}yG%fz`Oj8~r<ik|Cr1L%Dhh@tl6GGE`
znnCi4KKtBmx)JX*TN_`R=B(PKA^{1scc9qrWU-sWz1u6-Q)!iL*PWSJIE;pvBTxH3
zthp7`|FSP$i0<H2`%1_i4GbkmeRIuRBUXI-yzgAgW%bw%N;7{$1O1(mf&cWNpW08R
z`}Ms@1r>?#RBgvEg^JTq<D}rodzAdtgA&04_>#mgZPV(998Y_Fj#K%9{A6G@ABP^I
zorO@hK-G<{Rh>k1-i}sT%+R&-=Wi!+`;PT&os^Tjeep<T3JqhABar|{iw9qJ$)HkL
z;K}o6FQien(s+1qca*=IQ}YsA3^jih(GkFTNm#Zv>a#42Zrp?Brs-H~n##l#|7WLl
zQA@I|-Pt44OTm7PzZCoD)fBm4QdtyQ+l49uz0`d5W~9QaTZyaOSuVMs_rSh^76Z9<
zhrJV5yBdXZWnMD7%_S|YHkOpHF8xOFyp*0$EV(4EEmxE1XQ!Z%#bCIt_7rV(6j+U4
zWIPpfcMWk6M{#}w7v&ei;D^oyw#G{fGE|4R?!`~lZiBs)hRu+!!+i14hJ?F^h;fP!
z64WI1yDBGQYJaIJK4o$drLN?)cGqO$wnoNlRy_l#w4Kz2!B@`MsnQBr;Nnn>r?z&p
zFBeHt{7VSD>FT0$TW=lO`czMbuv$k?O(jJ<L4$3N1ttyRu8y}-p|(VmgD6VwfYrHI
z1;*?i#g@KMs)?n)`}v2M^T#*Lc7vBMw!P=LbV0VZ=uvk#>~nqm6XnFq46WYc&#gah
z!@TFZX-YX1&;EkuUa&UUU6fBKL?*=dKZ38dvyjF_kmpmg>@%fLAnm9%lyK(ZkCZSD
zHSWi@oaj_p*hm;d9A{&RYsgjgPzd&vNc}qPWIPeXjcs=pJAdcK&!(bwUP^Jqd#mA{
zX-T#Quy1WA7g55+kCdN0SGCj|Y%}J4rU#s}5eIhDDO2Gvh`rh3lV_&OPPG~BE(++d
z5uyltn2`Ko%SjMY5l#td**uLTtL4#K|H2_Lu-`FmR%RmQYSYE&7c}SgG{{-XwW6Vj
z;_35eGjnKTJEcULhE>{qM!Xw2BLW$#<tEZX*wiw0Z~&?iE|f=dgC?{up|9_#vrx6f
znPmce1gG&B3IHj5{d;<~d4u`<2z{{w?=f+KFZbey&wKeBDaG6rYOwW-|IV0QcUPk!
z9qbESd61$WvmN_~6t;t%#(NLJ!4u>iD3VSQvfv&;1UKFY4TB$88*$%My5%qE1&*A6
zs{bWCN{%j?>6C}*GAI5CXqCEG3<s7$PE_mMt(L9jfo|KcM<vdrqvB$3(QWUT8|poo
z^&qtKYaRdWeAKJI-XSCB(BkAcODai4?RdMY|54o2v&p#<?$6U+!3bDvQrI{p3hf9E
z#eUZ0mtJ&|Xg+%SCL|VVX$Pw9g_8=mCqR9;;Zp_=p5M|Ynjw`bHZ6QsGuM#7u#=m~
zWcH+|n4Nci#jZ@N1Zhs6YHJ7cUQ3CK#d0Gj{vg=oJz5eVwsu$ZfH#yjIg(O`TJ0*`
z`FE*aBLeRtFSW|dvTY>oY9fS$BGTW<O2hEhZKx34lk4ei^QpA#JOeg*fmUG4KkG)c
zwoS`gBQy{GX&JBcO)hHiBCkl+o)qN<C2=V2bNIVRN8)(C)P?7qy*0EMHwkGT@bY`4
z`x3i;_9_z>`1MKClxsK$5Bs-A;L{WLslwT*(N75S5z?`%VSn3uQ*6^1jzR_h8BK#P
zmMoRaK&J_y4S#a)XUI#fn6amcaUCZND6;A_UH_Jyf28$i^j|#^uQa}5rn{Q`^*}@9
z6;yRpGIXI@Zn~iPz2Wfx?ZdkNVMzN20Dh!INm#7`AO=4HMyyhz+x|U!M<L-G<A2L`
z9QSWYmXK)Gzb$2=iho<u<c0sXq<=&GE406%{uPhkto{{`-v;EL@c3;&eg)*W0r?e=
z-v;DYJboLHf5PLp0r?e>-v;DYJboLHU-9^DK>i7j-v;DYKz<vLU-9^DKz_yJf5(96
zZ})<8{{aAn;xggH=U@nh!U-<)fa3OmECdBQMYad9;0XYrz60_&f`j^AKj8!b)*C<1
ze;)s0H3<%&qWI}w3cg7U$ANALdmMPAJn|*t#S_kuuA?h3r0~(@3;_O=4{jvDkr)K$
zAKPCBgb92-8F}POlnMZ;04d)fJ{`~pebC4O#d!pdP6Pl))^Jcl7(^IcwT9qeBFYAi
z91H-8aQ(OKui++3APXXoCEOy%0w^peir^HX!~+1n5Ok3p<%ldAFa|GxWgv#=2)#ri
z0pcScV*p3E@z?ejt08s_+Q5-Sz%ej!1~efMF%}F!OoSu2YoJ&FpfzJ~1>LNv#D(N|
z(4mHLgMu>zf;eG`z~kSyze9}xx`}28H%m$I1WX7U3h)9b$&Z`}0+v{E{0Nwi9rGVh
z+d(1ncp>n{e+{)4_@izQ7>X#B7~+Ko9pEMe#Q|tSR7jl`;RMVAU>=BY%S3nr!sNgj
zAOc`P*MAK)9zu>Mgd)c}5wdu+71DuFgF~TkxPdmogB)5&4nTO3soVspNw9Df00-B$
z1;7%(5$^xB{T*s>P!)WK9S49=<4FGqLWI{i8f*oisGmNi#5i^UMhJBTH3SVn&{(Pe
z4r)p;G8!S|1(ZT??1ev5A_AqMq~7Z0MX2+D$$?lg1439hK}?X<uviGS?0h7|>%V~-
zAW%XNh>RdgWdz6x>3{@*!obH7Fg7&C5Io4S0Q#6Qv;=g5P@56F!0q^d3pE%h5?Fzg
zW64n<0SFr*BR^RUas~&wj{#IbwjBa=flw2n;Ii|{vU7Ow&VL6r06B*1LV?kbN)rI!
zXiiY_qcn+rW~3CPqjMArOqv`I0r0V)1E2%2xc{2fM-nnp!huqNxCVMz5e4N$6M>$8
zu)2^NOa){+k(3uZAwdvo{22m<pqf#DtcHT}gF7KZN88^F$Pv__P7$I*D6rsAJPu&N
zf}xI}2G1JcKL|Ax2Pp(yAk;*J;sZAVix`eve?R<XK&&x1!VPE%<&mi*rNaaBCWQPF
zse>Q@05269#ODL(sxbx+#MFflN(sUZ5Ftt=p#8`10Ifg0z-Sf!aQ(-#o)G>&dj9J^
ze;JT}J<h)v@-L|W#R&g;u)m=G*W>(aA^$6=|FsnU(@?*m{uP?vQ2&a@|4w7?pPu>u
z`SLfbe}(WjtAEAgH>>{%kKfheuYmkEAiv`A+kpIv$8Q7jPk8(`Aio0g+kpIv$8Q7j
zD;~cM$Uoun+kpHE$ZrGkD;`8gLvz3S{SRGH{IU8C0D}`|82Mw45eR^w<;K{jshZe%
zgLC;0p;|W3v)~*fN0ayyR)|HMSHPz4kct03{3{snb30~RE;7kQCENe+IrgtNBVM{-
zb=+^IpD8cRs$7A+$)@$*MkB_e!zx8hhxN>Ml0pZmv@%H5Tt@^PX5W#~B}P0f4ZSC;
zP2zCl9$Nn5GTY68qIC|Mi(*nM&+Jc~EbICsUizpK-vc9^pNT}R93rOJ<f4;qwo1;3
zlROh3VjUcm=C8W)$CD*F2Su(Hcgt+vH+~5D%Uf0w88hQeuJ0DY=1lifK$a@3l(Nhr
zQw`wa2Byu)gM{DIQ|Y7lykEK@d$vJ2IZqk_l?u9gp=YXblgFj3Wny33$KGfuC^eTS
zI@-;(|Iim}dcROBdD%;aY5T?yQ(-oL7E5xo@`F$C<kVwuA_(gtrE#OIlY*%Hu7^^A
z1aYI1=xLdvLaQ3AZJPH4A@Pl;)vMM&04ed;AJPoT_++_|bXUm>F?LLh%6f4rOPsc6
zw!VB+q|`{cFOr(Sb9N<aZua{SAhn?Sc6o;M?m`7wbUMT<FU3<;OEXE{9r2P;lSShk
zW_vIsiDYK?Vhlu8Y^H(giy~YNQ4a*9*suDz&l;#M+S8CgT(OJF7pwPQXWE&j@Gshz
zUGzcrMNL>W<O&3>**^)Hm<>isicQKNmw!(LXfRbUK?1KZNq5zK37L!zZaO<w<tQ+}
zpLnUI%XG|UTiatmnfbxh3k8c&``=F;n^l^oR7#>9R#f1RF_bXD23<9a&KaLn41KV$
z{zi22w3<b}oSKg4aU@PxJm-d-hAT0!g`z*<lDEnOsf(@XTVh_ke}R2_W15=i*-8*Q
zF|Vt{-U=O^L;Pu|jL8omz;>*;dYn~4WvB^j>Z7z)p6@!wcCYUV`(|%z#B~Y!s0hiy
z1!NjkF04RzJXqONU2fWQ^`zt_vd=4A;f~=-l{M@NDf1z_Hlv^KT+3x%{j6Qdu05hM
zq?%1dLT;C|CpKi1GUg$VU%8wz*8KEIR5r%Djr)yR=hKv;yHr?W!~v0I0gd|da#4C+
z_+ZvanMeK~V<5JE&Fv>z3)9*dG%$@D&ufD$T%p15QOWHw=V_Y|*$%k+S#i6VMfMQg
zWISh9AKUE+Y*{R4Q11?sHW4eM(jjcWwr5#$N&QqLiJQ0(gg=ZAMddlow#g)v++Kxx
zPeliy;IDh4)@{!$?G=SSH;eB=RX?q$unRI=OBqxzLPseFOJeMIu07)-FovP4xP!9i
zX^r5?>ajP6+%5RFDoys-;#`ulB*%``pSZ?y+ert_L@{;0QR%q5YGYPTZPvvk9=ugX
zTD$}sRXp>vnBBsL<fvz>?mf=Dd>iDjz)tW>Zt`0Uy{SvN>H$I>WR&uy=e47xW4Fhf
zgE_6m-S=Pr)S%O*a%U^N11?b4N%o!g)UXkSUzsG0`cBX#Q^p?v?yzFcL7Nd9kGTAP
ziVfJ`sY&O3cl#%f@tspduMV1vs@()K0X7{?zq9?9T|q>#^|LHrTV>DR^uz4<nLjqD
zoghsd3}KD-ZJHX3ch+i@psk!Wy`RzakaHjdBSy_aJztemwAhk*_%Oj4MzPJ+aqJGq
z4?uMK?hT5`XAcIp)xL*TGB&Z-$li}2w=yLTto!bMOTB{a!IJcmm=>+pWm0d(A8k7Z
z!5qJv`b#!)2N&T@H^Rs74eK6s@rOF4KbKH<l0HPq$f+By&$6y%oQjS>+AI!D$fSF8
zS`EaEH8;5N%~Gljs#8~T_?*`=E;Fg|yI0F>xA~Mx7r{~bl(a`jgtZi-S}<*9AbPrO
zV$*?tp-pnc&&)x`Zj@Ud5N)CHE)8&xRYR$U)jxf*`*&+`wrC*Y*#r7!e`!P*CxR#{
z;KFB%Bf1jrT{%{gu}y!BOV@*JUwE1tmYv}0=G|E+05~Z=BdFl98+o(6Y9&z;NshJo
z?hg-s7U^uB=tyI*eBhc=0n?7}OG$ESGdz7%6aI2J4#-55dV0EayTcWo9G?i!w|==>
z>w7Tx+_viE#TaF=TZdZTSS!dUTEyCoUDrtjVjwott-0rP7`a~PW2Ho~XEygRt$6n{
zf!wEr412#k9h?y=$QFrA_ldZlSGX=%p@3~u^QD}*MOUUezgLZzGL+!s-k^||Uaw$3
ze@2vae)jWS@r$$i4@F#i6;HEP%t^LVdPc8a{9uQ^K&m?R(i{!4eBkv<`?X!f^_B(E
z5t%9y3d)I(Xv8P8yds@$hNFXusEEBUH%)2TjP)3M!o%cSihm>PvhO_Qz56@Zx*tH&
zs8QVPo+zt^&LdQ>u~1+qolxzQD2$pS?L;-b_Y8WGquKue3oXn*t<_8z8KB<p+bnxA
ze$@k4EP4-6CL~R;Nt<X3bneT0y`!lO3cgI?)I<7wtg_+E`17n1idd2y{KaC({pZ2~
zSRia3G^eabdiMOLmpHXt^8DSHU%0s^e{=5#KwFi<1{Wn`x_PqW-5s^e(@j%vtEL)_
z0sJwi`he5mYI=_(>Mcz-*?eES*VXpeCxEjONc|M5lTxO>1ikSvpN`m<TH&Yqodp!%
zr5)b$$28+yuT{KjCVqEYk=)Rz_5jqI2(SLjXRi#YcUALj@aA3(re7T|uxa1_BKW#q
z`Dv^}*VckeNmvPnHCa*iFY8%5xYXBwmdr+~K{q%{(}F`Z{lq0>+Do>u>Wi~3$WK0p
zUQ<~oG@&$|+gvAlu7(d^X)fXp+6&VJYPV0&oc}i4#7=T`c#FGIF4DNNFd$QW0j!m7
zJo0MvfWXY+F|-;^a(h~MOJgEEBBD2ty&L(UpxtY%GDVeYCHHSsZqfwRoH&0IiwI6z
zKOM(5twzF8_%!5~vRz`jN!2m1mg-#>-daqr17c?dL^HTNdUWpkh%p~iBr8l|{);4*
zddl0o2z#bzoyqY7I2*gT7(?6yB`cYlWA8Z00*QL`;`UEbw;jbrD@gRk?O2(D!K&!#
zlN8c3Ue7TH&a`9t62MkiU!gMKMO#d9I`!Rv{aWpF&dpA>@+fwy+k0ml0Hoq4T#O5+
zj2d#T%#HJ7md84C>z4{Ogb(y5P5&b7kt-4)k`VuHPcZn?)Yo7d8*5n{D5sw06CV|6
z@PiJB$Lu$?9798Ru$?8bF{U%~Ij%+DB{ZdbCw(`6TKRh_yY%CxEkr{mQ_KW4nr}&q
z;gsZvj(UjO9+wSP4yJMTz&@^FfnDM-!y|kpw(kk=HsF_7!MQ&fzOsvqC<e3Mjrz0=
znsrOt>MI&A-Ray*?(;+wi83$ka!?iC9)sRjd&oDc<tA-1$s`YWzt`yJosP*V(Qc6N
zF51yOS%a_XQI(+5%YNEd(!fMDRj8mMM;eV6+u~2xK;O3($S~(9IW_0(*F5D;+eC8X
zJ$PN1X+%tG@43omz(*`WZeSJw`V~1A!Z^iMJ&O_IP9EWs#bgQU<qW_@NQ4nfRq`+>
z5+V`UtcukbY$LsMF-_;EA9{2?=L2h~sDkX^P7vrf_%xS7{Y7qJtUV92j)^sKIv6jv
zVqKt0;g1Oad_npO4(WyLgNtFbhbAr~Aagh@F)>-|OLTp&h%W{E|E2@{Sg;)sB&-`s
z3<W<K{JA0;_?imXbHdo!qqxzkCLnp!>NkvAj(SamHA4}~&N(htS^s}KWM*E<2)X(G
zoxq+<*E55y|Cx)e)>}A8*#+E|7ZDk2GG15MH<IHrRw%^DR{Pv2cMU}HoE-q!)YV+R
z4Jjk;!4hA}nSDd&AO-Lp1j<y#5Wy)o$i3}|{6Zaj)K4mx*p74FeJb!FQh*qzL=<b%
z3)q<iRF<r<5;?3R;=E2$D?+p34%D~K#~#FNbDp7ugD-=;Mae(If4;h0F!woNxyhcs
za69$$w8u9JoRc|(l)6Y3e*N(FwMyF$7W<i35z$>}BJzyO<E$Z>#-bBfS7YsZ&mG@y
zR`627c<{cF7!G<DwMVn4c%RK9pjf57O@dPW)fZUx`sxnNO`h|<6;Xp>P;IxEZFdI$
zSY*V*Os+4av5KQUZmz3Wy@?EvG#3VC12%%)N@iQ3S7t-i_fkYjZxE5nP+VJY&XoEd
z_TgKg>eQvV+w_WDyS-Q*@-!u$<i|xb4=#Tr)-53CAC|}J;7Vf3@{UVv%?!0n5<y7|
zw5{u6&lS^(!QQK6Zw=a~**`i=m2C7ZQZHGfv6#}54-A~;2GoFn!RYLgJ7htk$UT*D
zR{rQ@SlJY$+)GJn9%aqyAeL*SfaYU?^H5LOUB(|lM`gaq)!xEglIBZDM?VmQ<);@g
zVi?*OHtI;zGIKS713Xx|(;tO?0Cqm8)^{BDCyG%mLyer1Uewtr7%9%FRY|eqs?^Zt
zAL*HNVi6rw=uH){1RruFykfu7*kd2<hvy?h7p7|S(bYErh!Yg|3iO-{hrYln!cd*+
zh=8fW_4%0lP&hmlnrX@$A`f6tWI|<PZ^acb^!2B%eCoW0StuNp@+gan&5xhhzf`AW
z>#0Y2)!NSoEkQ}kYg$bZ-jM9XntV{4z!b-(YwQ!x!pnt7W6xOJ*w1{L8f8V$^lKpO
zio>0hY_CAL7hXb}vtm0^Pk}F70lXffL(*!Syk#(g*opRZsLCGzX}r_MW*d$VuZr-)
z8;^@zVC|?c)^w3BNIi7yQLS=AF+eah6U3xufV%mY!Ml92*z2&G;^b6EuaW@LoGq!V
zgo2)*Euj&dJE@bu8?dpOtnL1j4VBd`CvkL|a-y5YqgU%NjmA7v0VOe)2Nh@B82nGV
zV+dsXYU{xk>*A>|`*0@Mb)}|Yvg`X_he8f2G|ibmB)Krrr+%@x$irp&f$gO{_VU-?
zFC?LWj>*aqq19L?x-GEmt~`6w4xxlh-d-o&j?S5Non40z5gW{V21h=<zZno40n^*!
z>J4(0HgG|rph$R$`PX7W$4@sI8Fsdgy-6MCfOJ(b1VL#);I9Zuu0VAv8I?FT-C~>E
zlqvNd(a4)ZH|lE$_z(`U8jp=cNh@0*hoX#Emr)XkrWjRXW1Nr_&D&-o5)NNvkN$Il
ze$z&M(Xu|noeHpHoFB$m4~z#H3B6Pedk+KXC?mb-!)MsjSRbl3+5lBAQO7O(m9s`4
z(CK7U*xjsy3)Y0)B&L3eq6}$-@oA=$1{Z3C$7(#l@n9a@>W9vFrSeaN5s^cYR6MeH
zaO4c7ueepPi!&4ExvBp@YxaJocI{~5UI_TQBPj`3JNz-M2BGB!P#6>Y4YaD4cNhYj
zR%Be?vN`m(iF*;mXh4!PB2&n?NG*z-;eYj%a(MgS_)N{U!zus1)#TNiuQ^`J+)=o5
z$@;SO+s2QNpL$RGUYT!N{QUCw?f-39zBR-b9*-c4puc)l^6i4lbOxGqc|ltq5vn54
z8t61g!<ejCcS;R+#cSKUBWz6embOAP{3Vj|wUA;*OyJG09<t+IHXF({GshP7Ln`eZ
z3o2FaQ6PGK4s~0NHHOMv6Zf_~3|cc;Mj{%mW`!cPwUQ&t5RU25vfY2NyZB<3o~~k+
zhN}syAZy4?)Sw<j7@|mW#DDc_s5B{$H2nnirL+hP>fMoIz-UyD+oQ;XVb~2a^Fm1c
zH-AFd;<{kw4!+2n@eD$7$MJ4<*MaLPFD^cQ9)5vQ0x+xlv~;Lcl2k3j&>LbwR?vZA
z9Y+-~b}Sd@_9C+-Hi#o@=}w}gaJwE8-kTf{!HlL?$eLqYr`9EWzvd3V(xBkf9FDK)
zkZYw$s}C%M_nQ>f&t<}-g=drs7*1vSaPRSr7C3HtHWZ(uJbMxwphi<N^`tkgmtU*7
z^VvpSegKs+@%BLAwC^?O$RG!ihHG5L`J}U!?s?M+>H`zt%Hy;1C6WEBIYVMuvXh<Z
zK~xm-jf?Ixk<x4&@gE)+&rT*W*}N9gEcxWG%3U#zF>o>;ZuO0JotP!@Zq_KMLAtF4
z4$XGgIp&`=^ay5sseL~^zxuLLTVt%UcNFyy&;B#hi2h)7XyDhR*D@QPR<tH^{3IJH
ziV4K`s~FS-9M>8r?|CIElo0&@?gTbJK0%Bj1AYK^Wf-g1z#h0=F7G!^?%D!}=tAqU
z^O(xbPY>7_qYO>cjbfm>R>wwASQYh_ij=xp1UxmF%<&4#-GLx{Tg5kWD%R{)?h*Bh
zZ<n3MjPqW8p_qJu-onF)>ujSPjJIh9l|`oVMKmHe3i5+HR&Rcq41ZC4&%`COhgF@D
zs*^QV04Wof_|Caw^k#(o0?AiMmW%BDk|>HHhEm*3q6nApY)ErQPcO5uoP2~Ri|~%<
zp~B%rl|aQF_h;u4ExC86I-04TuGW!g8Vd!*X>KTy(B)-paqV{Tr{}9diFXuPV>kUZ
zW{%eqy<51p{=EAKFm>(v58&YsK==pXd1!Gsb#V9lqdnVim$sgN=@;t8tK5NXMPRg_
z-H^-?RM0Qq%0Rfl4LF&p=wer-m>y)?QixyemX0Rc60u6i=Y+OZM$~ri@|88nBLclw
zN?By}-YDJ#5XMSoWUfppC~!f(?$j*J+CpGxZ((FD0pUyxkiz(KZ@ritH7G!hC49&?
zyMhUOdMt65=HT_Q#yWd)j`z$<Mj(Z*Zl7VL%<@IzD%3Qqow#)9;^<B8+nMz1R$QCG
z#J)iqCd1XvxQFHRhUDdIjf32pv8z-!v|N1i_&#lG)Ff9Kc}d#nr?5Uzp;j#ZLCcg$
zZucdA4VPLQXpkxD0QaPQ!OjA%|1diEfutUJDm^e0r*W}GUuQ<QW}qR7%`8c0PMLn^
z6@A<tuw0__n4)}+e-aAGg15vXTvZoN>s!K4UZnJU8mW?y9XZPTFqxUybP8#_XE^ul
zdT-+6YFJjVn8u)xo%ANLGwM}69qrm?rQ9*U)S-x`$7+tX{5Qg5WOvBf#YHf?LBWE-
zO!9LvCY+oPJ_>TMogXMIYp?M>yn8d#&mK9$QTz`)PL(39!^?e4i<Vo3((8|tUVK`*
zWT)><Mq3dV`{tOTpM&En!{7y{Q;1BhaqkB%&w105y8Wm<X~cy&+Zh2A^^1vzP)cV^
zvYcXe;bW@m8+CRtR8ddttVm^%Z22LLWOm3gQ|jP!u6-S<kwQq@qJ@yD0J}hE&N3oS
zLkl{)`oLw5JTkH(D^-H6I*r|eQl~^xMsJ*}gz~}}OVZ>oxA$i@!w~FV)Wab(MyxHf
zJ>~dw#r(99LSJ(m%SKc&r+|->{8jYOyV+;2`{c88Gq<x!E}PZyCL!-z^o-|sx&~p?
zi&KPCdEo`jJKy=rNS_dWw;pZLK;P0F70rKdwuZ?60mu?m`<--1_~+(o`f<Z74R>x<
z9-pgxi-zO{pSr3(oS>g!(0!Wda<9B@-n}nK909GWgXB)6QVI8>^P)J)^M|}7k`a~<
zwtS9n-N1e(Gq0+S-^{(IasMK<I)1t2oa5E-_nAfugDJ(N2qT<eeeY4W<0VW)&mpNX
z{7*0=7c<ayBQ)uXoYK{DP{FoKneUlyWbbT`Gs*eWF8F4#Ah$Tj^NjURo**myBpmL|
zQ6Ncrha_Wyt6Id>K#2&D8OZr!GC%%AwX)=mW&$&yeu^cu`x<AqSv~9g_dX-4ZWZ1B
z@u%Zlc}nyCeP7qL>JwGbgF5rW%JW@1U2<_6i&j#x^xX`h16?!<>4xUH?b6uXEFAF`
zYe`|`51={pM9u4pXCCn?ZTVDVLdkmzd|R$%$DCc|hp*e(d3uq!g)7zd<xs$(X%#oP
z;U;OX+o|qL`-lj3wZ3-m>%dk(7Mia<(iA^@(&|k<6Mc${ukX`J7b^K2tosd_W_~y#
z#pji%abx+UdV5@r)4a8c!(5&TvD8NHxdLn{m3wH$+BsX&tm9aYu_3PtL)CAvKD@cq
zN4m;)#8V^uP`R#M9@vFj=MYK8WAx9LI5(RXzkMlFXAFq^dgJ9CXxMAPGq?l9){{ii
z(&PRgPfKMm%!7K}e07Wnqt&1&W`BBgwBdck7o9a~n{Hr1&*J4qR_>ilRotw`TIPfE
zQJ|(KC9lpr-kJ8EWR!CI?$t<_Cf852`SptZ=BrcHs=OthdS(%@*~c+0nIncPG%)J;
z?Zku~{!M#IH9MptN%uMjiwWy-DI+yu^9NyHz6iP_Kg;UI1f_YGgvUCEkUXO_KEn|g
zF&3f0EJ@tN`5IH=-o|9eJYzp{?Zgo+A4DPVo)7(qUW>DiZVHX+Sd{YkB!rce_fC$M
zQ@jY4j8Q|B7G?-`eN>zenHCDJl4j(Rds0}QZU7h&LD6XzTi0f)NPgYOUAJm{RdhP>
zd6>U`N8)oVyBM!p2zw~>^W7b%x$b93SCTe6G5u#JPH`}GCecQEb27S>59D3YoXeuV
zi&Si;aI_e-KwC7Yl-CEOrI;oP1;jBkOez}apLf-rGe%h@$SR+*;w9w{m!?}3(Q!0s
zMsJ0Beqy_G>A5?r)28_Fn_xi>eC9b=o3)-QpKXq8vi6W1f2TVYw2^mb@tE_$xR3Uh
z)3$80r`4TXe5OzKLO%ARTHUC2cCQxqWADh4jplsv!la9|($9_O_?q)C%F2WJ8Q(ir
zsqL5crm;{DTPId^)7_AnghJ>W)i+p8aCQh8{f6$Az}mOe5Hifh86V!L@%q;c^06C_
z&EPo`Vhu`9&8irb9aqQC-O5pOG3r$oQKk|Ba0(=?@EA>oj5!0n`uG6_=%N348SYl&
zJn|TvK}r-nna4pT_h*&+IxuPe!9-)al~#E)4dqmQ1|aD8IBZys-3BZbmpR496z?zc
zV2RA}B84LQTWX4kU@f*{`g=0d<0RqKU0W0_pmw~FT$DG)5i@!st6FP$R6Ajj+m9jJ
zz^LIm1GQV*+?~#I>1>XynJ$=RvLP0F+J>AAQro9GQhL`H@{ohOPJ^5@@h`jy7A|6T
z7*A$VcP0eFJH82-AB)o(z8bl%S01OYc|rZb%dB-o4QnIQ4eqs3ThB>-^>a?tAI~K`
zXY0PM^}><w;MH3h>vy3eqN%pLT+Xj&!e}G)Y3r76_Y3WkDZOvhHI~4t31UQ!*L<oe
zr|)y!4ce5QymwB>)oA9qTjO#ebB1#Jmx><FQ|DKn=sYffNl)JM<`L#M-Xf}v7xJ@W
zh_Xh$;>~=}=n9!e!lXM7&otP`tKl#71yxVQbx8P%xH5h&$8!4tqx`dvIn^fOpWSo}
zc*s*qtZJ$irdQc;nKbikB6k6DUa6R|C4*MINY%*hlfuhRZ}o~4ip&$`+r`g9BH!YL
zTp0P!@N=-a&lIz8@%l86`m(A(bLscWo56-`)P-RkA!a380f918JEfRdIzgYAXY`L!
zF>+ENiZ!do5)5jBe1Rb#`|MdG7wbJxg8|39ZPmZNTAi%gA<B!eHF&icNbSo>VNqyD
zS5rX(Hg&e9uI+g~KX~}=@a!e<r?^KyfUKz>K#wQ5xNs?5m(gehF{cwONZ7%t^JymV
zMa08iz4>!kDU}#Tu^lF6G{~Z-r{%!1X0*20tZFEZOjf2#DIasmR%a~F;SYm!w2;w1
z_9;kL7fYC>Se&tPsDWL-fB7{G^*v&7Nx*Kyz&0p^S;&(V9m_EPBvrk^A?75fg`0kF
zW$rV_I@_#<DA}q<t|ngoh8J5-N`q5V@;<IA(KX!`Cp4oeUzhF|DIWd+4wxNhP$$s}
zNSpo;8Mf0GBJr^$;hG<rj=f5<+%Mo9m5GEIPY3ubn_N?Rc>nMf>S_X4-?!2OE$91l
z<5`Z)G!V&q+HPlJav#_bVrPq`&L4j^#4}HIh0+^4Vsw4CGGpR2&3I$-9^FiVJ;_Z|
z$`?8oPu8Yyk4z{;=!&+x7>iMqERf{ScUDy&|N2}Q;=ORL0CB-zvrFg|p-<5J6s)Zd
zByEy&Z}HO!-c*wv@)-&6a$S;vZM!b2lnZ9K$&#zG-*^Y(Gqe^1RZiyMk%l)JxTfQi
z$xnM1!%c)?vi#q_8CVIs!TFeACS0cy6rC}rEeOrJ$C}*<)aTz3tqKU)y@=3J6hD<=
zr5!PeXe?Blmxc^EvvT&0Wd}KwLZ&@QKdW}C*ggH|DAoCr%<tqUyk6=TU7Cgg=*qGt
zhd$f&FD`xz9A)*{q0R(rH#@mw_Woq2A*tuCO#pO;PdE0B^er^`+^W7|vE0Mf6j)NV
zboU*`ujEwkcARZ57_XEznw+Dfa#o1>y0O37F`7H?j>?YX3GjrEXyBD`de@CJa%RLM
z74GyqV|^~nF}>U^0Y7sc|Ez(7R^u!!t0TmdQg4`lxj0X~*9QM2A=a#}(8qdgYi5i|
zXvtzGMokQAH|V7qJKkp>IX9wMvQ0xE0OECNX5}jFs~%Xi!5d`^ME@!=MHkK*^Gbm%
zf4W+F{ABLk3;T^>w`tO^2A{%e_ZWS=VQ{7J+`GX*`Mmq4fI~I*d5LBctCU{ddr3Rf
zfn#=S`^6AC+7=NJX889_$8^2ClpE)HSySpeKZ%wJYqN2Y**O=!N&$2CQJh>x3Kng1
zHv7;sQ4LZtGB)^LDZP5_apWlr11fD!3n|xt`;u~l58l7Le>uMxxy4Z@CWMy3yi5@%
zj<R@vgVIPSChH@$T2Xd#C3V}Mxv^N-!^HP87&<yX&pwkct-hU1)w<=_MpyBn=>C^1
z5mN5L$f3{D<N1%Cg4r8N&*aRwb~dF}mNs-<WkxW)Ncn0p=MCbth!ub3ipr{#mr&3!
z<Dtumaj_FXGgr{{O?M3soZS7pkJ3A-$@b0*l;*x6pTnVsXp2wSS^Cv2l`eHw>i)F)
z4$!FcS-zD94lj*?*v@45o9rkxgx&XL;S|}Ipx-TK;hO-`WEVA5;IdZQb8LE(Vy+!I
zhKJLTyu2;61s7o_Nj?xf=l>eq4fYdzb?EgR)E+eYqVH`)w6%CfarGSOgq}?{IkeHv
zt)Q^+Tx{kpd6ccRQn`_pb4uB@1jkd0qq)mHw2P(`h0O0aaqG+}u4UYn{%N^1OXqfv
zs!iWp;n?8A?S`t<QcC0Z2gppj&sc6=4bMnjh|wzwdE~+iu$g<MD*B<2PLRFgP@|8@
z?Vc|-H3{l$?l;hU-^e0cs>5naOBJLorq-qx4PcFZc1qQ4t^VUuyEH!qG_}c|Bz@2_
z7TWy?4&`%AD!jmpl>62v&Av3UI^V?A{P=-O6B$xu_6LB31>b|G<)lvwef{FP&D5v*
zWeUbvPY*v`ZE<?8KI&T}r=^v7RyConY{wOC4f_~6n8;<7$&K8WSFsmkP^+ZF*99py
zg6)R8_U&D*#X*X3K0JIdqt7Xp_t43jTzc_-np(N&UB@Sr3XAmzULSbdCSU8S1r0dS
z^xOj5iigMU>Mqvl5jlQEcg0*UB$5}*NS87+5K})j!nH}Ib@E=vOO03~eFidbJZTug
z%vG`bjGeL{iuL%a{he0u%-HN#xjG%yijk4~Pg8?1;jY)Eu4c?}2oP5<?$HPrMaIb)
zQ^yICfz8MSY#Lh^syRK0t(;izy|E$Q8>~calAQ$kiMP%~5b5Ob8<-`RSD|RR+39*a
zRCup(#*kTE$mxR1U+<F2=$Uvj<IQgA7tu3ivhSUv+^#J0iVGY72o_v8%7*A&of<px
z6H3j`N5|OV0<wzNpicXG0%*w+BdL}O%S(vJ*mU0XE9+Kwb<GF8&scuES18swsAhI;
z<7wX0c$tQfWcN<GyQl{)g>_LMnkxNT=SKjrmlOX`V=;28NJ~V|eD(s^SNn42a^kJg
zz$V&_+Wz`oYeUfN?)l{Omkw8ZR|jj^8f_tr&N4>nd$%pju5&C!!TJ^I-nfDlvMt1o
z+^w`S;o9)o!}NDMSwDc%N8kCMuZaEtbS~}d{muLSht*az(E~EBygNhOy$n&*+Z0<b
z2~~K)g&|@QV)%91y693}yCMm26&hjO`tejmT>2SoRk^fk2zBYqttk0^J*z^+%8Ab)
zBMcejUR^FOD;zJGorA``X>_Fy<B`(ny7#`arqBHOMVFG7^5g)KxfjtZB<r24hM&2J
z(;uPBmL0A^dlf2UXwFs3FiuCSr%Rk@#ea_t{isStG}PA44ko;~QHpGv=42+dqSCDV
zc9qhqHlX`i^8*p7aEe-6mjI^I>|?~;lMUO6EnoPzeQvZ1%9^bMjqTQ4=j$>@>f-n{
z7nIn`WMQsRdya&bSBLa6nJaznC}np6QZ#MS9L1-_*GOD_>NKOFX8!Dg9XUB)yya;;
zj4}4U{SJlpV0gSHGamDK@}8|ftEPkFJK-Who{tWiC$#5yO^!>8e9F|VZ55HW@1mkc
zkpa5#0ORTIoAIeVZ}Re<)HSNy)IxoK%s!=CCE#iKxN4iw8J)MNOjx>T;*=n7Wkd@O
z8sU_dNzu%<sh{l-A|qknl;A{?7W}4z>ga4zBZxb%CMNtlTrpV2u~g1EGW5k5KP9#0
z(u((6i;Xts)8~WJ8|oIp4)!be{z+^ogzVkcT|%wH`np__GJ9_GwVIDuqcjtJyu%}q
zJ#Xzr)#B*y*(b}gTBKcz5Dbn3)MFiZH6Da^9cFiwzHhX;{~n!~7CEaYRR060$)#Iy
zW%EhSIPQ}>x5UuA==abxoipT_o{f4tp|5$J!Co<AvKIf2+bUA!^f#BB>d$SUEyI`a
z>3A`gcduRP4aNAicVgBr?PR5w)^k(p>(Q@J^gL&ETP(XZxK77%LNUcvZ60_bS8plp
zC0g5%W_Hkgxir_NQz64&V})hjW{sx%NZ+LIY9%Gd(Vd9CU3hJdRUi%zjsgn0%?cO_
zI6U8}upFg4mThj8BP*Etgl8jV(0Cn_<E+s^%0*0*_IAj5r{7n-*Xpw{D(=liw^z^A
zJT|xlFx&a91M0pwrG=1)>j}N;kW>W8Y;oyF7_;2tCz$LxQ2XW$^+_^^{9<E?&J%s3
z=3+a2k;tCooiBAV)opeab<_vD)hkIcOzFCxq}Dk5CUTgxF1L#(5D_Qg5HN=YB;T#!
zy=$;>{RKY9h((>nHZQzg9M9#))2=ZbRCWp*Z)dE#Bb~If-+vKnTD|j+z0p+Swi23D
zPN{$O=I-KC)CGtvXvh5(=kTc-X=uG~r!P{N_$e;IvJbLTCs9C-?afMjP;?%Zv~fZG
z{q6@<9S7-kGsb<_vQD2FKZ(NuI5fov=zR*Kpc^YW)InPxDymMMuG;hmI}GIlBq8cw
zFKcKBN)5cG1`DmF`u*Je@d8eWSo^_!dE#zCJz4&Ws#>tIGj_&rxv~a!p~{~}gkN$m
zxn=Ew{zQHJ>)y+3*+UP^BLvAL!?H^Dv<Oxs<-!oB@raoa$FhQ^!S`wzvb6Sx^gjSe
z))QEv2#xVWbe$lC$4qOSjViLr$@~mSGEpf#NkyA__trcCalJ(H`?mH^n93})j3~kJ
zubI4YKYoZG!}7WCM$6tf!L|GX*Y%SehR8jRsCF!!7<th`RI0SQY_<8-B&5I%ORdY(
zByY^pJ4ynOK`KIo4*ll(74;RLX!1er<q40Rn%l?4stN*aKmZuSbE^d9=I3?VJ*ar;
zBYcYQ=g3*o!P5oE+Anvx(5J~ZF*eZPqo~|GNZ+(W2PMd<0Hx?Br!Pu4!+F!~J>&|;
zV?MrT)#u7FiuBqA+2~by>Bh@>nwW?W2Y|U3G2=Z?v#<sIWHb7jZ!#gLWlqKzr9M$F
zrM)8-Z|hWp74Sv+d3o{{TE%ij&*F9I!z?!_1e^+8<20@b5X7D?tvm@~%v)S+`f#&X
z#PGBn{hB?0eYjtxZe|~;#RIUx^8d_1?^M7uk;e+mpRIR1DrUTP_}P-<xAfbyefU-H
zJ;-I?DEPN=S0@8!IK#$0$NYYtzjWcuhpb%({%HGkDv11+_q=IrZueNJz{KK2lOTga
zTVD3MRqUINBn3R)bL%NbqHqio(B`ubB^L5uuG;owg+@kH;MBXN*F%`zEagk|6Z{#-
zaU}Glyxgu=oA1SGWNPa)@A>%kdTbikB+e|&d#?@ePc-y;mVIOQx?QgxYJCbm;J0So
z`px|dmWfT|bF*dz8WgtZ)1xb4L7V)(he`LgFbaJ82Fn7_ys%}vs*AT*(3-u#oB&)y
v;}D!x&gXgVQljCinpL;6>T`{BJRJBN#Pa-&P2;sM#g0zY4YmnN`TsWo?Y*>R

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/readme-dark.png b/docs_new/assets/images/readme-dark.png
new file mode 100644
index 0000000000000000000000000000000000000000..d4526256d462dde768cbcbde886afe07bea2769a
GIT binary patch
literal 38912
zcmdpdV|Qg;&~4aJ$4)x7ZT9KdIk7vo?T&359d~Towr$&d&vVCp?~k}^jLosX>{?Z|
zs^*+4OhHZ@5grd73=9lWQbGg>1_luiy7!0u20B0VP|t%daCQ<Jj$mL&eg7TcL$)PG
zpqIE#qUugcwkA%l`VPimuCA_(<~Ej&hWd8KjJ6J@8JGNcU|_^xk|IBq-7?R!T-`8L
zUWbHC4;R<(rT&1y5fh7wKw*j~r{_|a8(dX-+Z$t68)zf^wmn;|uKFm>uMebj;RO~@
z-jf#`Q8rCwZV{yfl9Pj@1K{u6v-y!*91k<OQanvN@3Na<X(+z^zu}8nSc=H}zr!pI
z2Qb_He@CwF|0fT*ai}ohrw@+T+K`|Tfc~esxw*Qiai@y7AW_yj!0+e3Jnj1=-zuPz
z(!bE|D10u$2I}*bI@2Vkqk8kX{T>eP=HQzH4rXR<Mp%-54yG$YZOv-isinEe<(2tI
zc2|)q-Q?xW+0^y*EDdos!B(AHjJhz=oRsrayN3jbV!;+>%6-`u{eY?(YisJ@SaxPJ
zVn8VE)?bGkbw)f#M@PE;zjG)279_Y>SgI`M+S-|cWxKW7>JWn|%iZ0%GtI9-<KW5F
zsmgijQ)^#$zUUo%CT+>|8oTBeZRvC@EINHPuQ>~-H$iV~Og!n125S^u_c7A>REHQA
zCGhnzY;_L*_&oKpeMb&AI-S7>p5#V4z1^d%J-wL&jUdx98O0!@iyHUT^I)ZlkP{Hn
zSEG}KjlnfE_>&Qa9A=(J>8b*quQof1H*Bsdsinfp$}ax65d#%5_;Mk5w`ZI?poZt1
zy?u#MmQSAEnoLYhtB8bryecP55>PNwlY)#Xeukm+UMu6_;QT<R99Q4sBd0yFPvE#G
z4OG}@(W-Sgl+`8=R8Pvz{xncpTUv*eKF7m0PmynkQL4gaBOZVB@v!&Nwckb-g_TWt
zHK94+8sKTePSN}Ocg?9)3R{Yg<o*JPz0u}c&7Tx+rvJc+JYqfWB*Q#MH4rRB{9+_&
zKa<nu?TTe;>I3Z5HOi11fj0<O-rvkGU&aMi-g6CXS;LNeW2KNa0vnOF9;yLIaZdc8
z@#JhEnD_zS5HW<U>qb)R`}KLGNuPX&A&N^Lc#M>aSDg*z^7iq$<)ot*pOBCs==1(8
zU?-uR>`7ZJn?V^>VGDAzVCKs(J>HO^Z1xGGh-LtFc+2}kK1Ut($}@C&5u$dz+01QG
zeN78$dTji3al*k;S4_0bBrR;U#lH|6NvziThV5z^?hW2EHUY1<y1OTXl+1u>u2bNO
zr-kWR?`BrHzHS#eC!o;KKdnRolO%V#w9ce0Ba6p}k)JHGDmm%+pQ{@a)9C0kTCU;I
zq!7QI@p-I}tLF=I$8n{9dDfSCcvlAr7Z)fT!X1!}oWW0=cn1>X1#w*_d5*NLI5U{k
zRGg`Q3nWWJ5=@vlif`W{5%3<<sg|qnJ-u}e4-Y?|r(0I8q-12!?d|UF2g0HE8^#Dw
zBEw7Rhf0F}{_nx=e9oF!zpFQS*@rn7P521#qnXMQ5je{KZ=to4=G@YPbgf`Lzsw<U
z{=gGjgYQUHa8Xdqdp9~FKmKX|Tb4g$+uF$RZB6TVzyiJ9GG_3-%k&rdTes^WHllh{
zpQDn!>fr5ge`3Zt*3Am$FS5ky*x#RZMbd{=)D%@BeFrl%jC1a<3H+BITJ;7j5l#7P
zltY?MPa)B;Ar<|FdpD8=-7fE2am2ymwSG8+Ba8An5Vsk&ohR9;sctD{Wolrk9CF)q
z!T!z)^>CTE3XL%9Fx`6J!I5(VnEhxltQ3FSLgo>WypxBXBF1g_4_OXYnXk6b&d#dA
zAmZHzfZU&khKBdY>LX`k8{+rxrv_H1v&+NHmrohzajrCFErSXS8(QO=Rh42EeqYk^
z@?UE=jWwlVW(oXuZbC!0doUq)uddM9&Yn_hd`}m2kx~*TG|KK9Nk_q7XNedY%Qag=
z@0f-4Q5-BKrm)tAG9ui{jh7EMbHS|iCPBt$W4fbj881^H`@m(P%i2br6XAW1iE(wr
zSA4(3^e+R0RfaAI9JeB3@>LQJ;bu!<jM9JeAri86a`#y|eiRR))!#u%8vn@FebH#+
z<Q(ucbR$RnZjq-goE$Mc3gPhi{!%O=Dk@M=cB)>MG6SF3^Eps$3vTp<f)NX>>LNL0
z4#fwvo=U0j0?U_2g>hX1A1^D5jtULMX|vgAO=3(rD`lJFcbx2TlM>+INE#yaEPuaZ
zfii|fL8mOaIQNsNn<6NpvxuO{GP$J#cTA;Tf1hlcENUFcb!YL|9onhU36Y637y2D<
z(I~dSv6xUt=!B?aI&xE6Rf-Cr27Y7O4DTkiP^;F2;*6*+O+WGLF?GV<s!QN#kq?jJ
zRHXC*m-I>u>*@HloN;#aH)vp;l{X%uc(9XS=;+ZoSK18SxV6|bhOr}PyC_n|cgS7y
z#IPwkEWMd;iD@%SFK~er^hIw0d@TY*`O0OHv=#z(HmaCda?CDk)J;I2_0f~_vO_q8
zQ)E-@tNx`<ch*6pc*6xdVLy+k1|O9qO@o+sPUM?ZA2WY^MZ&4>aG`Vxtz5B^*Za-T
zMVr&fZ8OOK&}8D_NFl442pWCCJUd~0&G&GzH#OoN3UkLEP&N;$`+%c4ptYkol2E|I
z{QyU*kI(%cmP91v-yk_1Ff$<5l$Zxv0vrIwmXzp)APmz^T!&H)phc)Ty<^?VO4yhW
z#$`wRE8b-gh3kBCK`DN(ykKSP$bN)D@0&DR;aISAcJb1OEcCvz%*n~Y5=J1uSa{o&
zpDFiIsEvjK2p#U1({r&j#U;Wv-<@aVl(R4;DW?9jMz@l*XWA!+obmL0>>D4|nPp8K
z{@b)M#x7(Ox;TWC!U+4@@F;RM)7Df|7xz<%4-Ye!tLfrMhn$CzvhUCyq~|iKu%D8l
zZDr2WY&h<6-{f{S3B86|^4*@t62^9>LYu>$64$_9HFS>olBtKA>(3juHmk;vgJ8tY
zl72ZRd2Xs*BD$tlQ`;5n<_2~uvf`Rex?d8UD8D2crD)^LM>}GRvpvdQPKt9n*9>a-
zX#-&^a88)~CTx2Tzy7e@ach+V*7T<>;|n(I6lK@Ue7bxW!!hXyv$^dJ^j}<_j&gi;
zFY!1!pE;&aRu#j{ivD3V14!k;^#yM=7*PBWMt+DcWv|?vv}4LzNB7+1eOcS4w>@xw
zSxB3$8<q{s+?2>C#(!*mWC@RS3Y?sr5NN&inaRj3*>Q-xm3qb(J`wSK6k}g()v&h8
z9&N_$(0jkPKAR;-R=@qpc{hkU?*;qU)Ka5?;X-g-b#Li~Z-b3Y)l{SY%L>9*I35La
z{r$r_VNFBfv`b0cMwMfL6q9kLKwT#@>!lP0Akui^)X_(2BOl4~#QnET;9Xepk01>v
zTE2`|YyEQ*$Jz%v)a!>YAKy91f*#a&$9@T`8ZIjpq$iPtB*_YGy-!_yO<&LAD+s`y
zr8dS$T}%Td5&crt*A=v*ae*|xZ2G5aLh0`$E5e}T6lI|!cFXtcL(<;fzD`+%K!Jr;
z;TyhQFyfSd(>5#{1SeaddbbJ0Ke%z)5_oEuha4z&0tnt`_Umum7ceQv@8UIds9a#L
z0`>|DgaE0QsS~G7mqiWJ!l8kIlVI0cllZD1OK|TAsM?VAC6#E1KcyhTx9k2WQ(b-T
zuYJf`$NbzQ1}Y`d^;P1T!bM1PX!L#FJ_snJPkh?iik+^)fNa0S6&xii3k1e@$TVeA
zI_p5c^bddP2guTTeQBLQ0%9>y%lLg)R(596k@VgbcAY)$Z3AP+;==zb;m_@*<e>Mw
zCl!{D*~O$#f4n)>@^hcwE6AMYN#FW|UT+?H%Te!Uv|81q?u)zXU~DL==<DhJS?S_U
z;nC47%P;GM<*U2I!QXJ!%g#Zbjo3mDzm2XuxfrL^Z9ecX=^s`|!qi%)*Hn7EOru&G
za*?X;$Kd6#Qteq1;?q;Z_}II3R?ZpbUiCk5)`uPG>h^XN6<66|FP%k@R0Ed;-b!@K
z1W!aT5i6=l!kXlh*Dp}ez<Lm$P;FyCg&6m6aLFl?Xt>k0IneLd8jn>X!oj)2aUTaI
zR___AvtVk!S>#Mx_)GAKEECVeBf^#5ear2K9I&B&{G&VM+W0|@^Gqt@^v_LFiU!iJ
zTbilkezwo?aQp_(;1*@Wwv&7OXr@SNM4&$3qxykZTydX)PcZ$?TM6I5&Z@=WH5m2r
z*d^v)W3}Hy)G+l%=O6zqU6G{+OKe8%=NT5x8V!_9#zL*;$wS5lD$YjQ+8J)*hBky+
zm(IoM7P&~*^6d4us8IH`X0RYdh)ES$a_~^8CLNVp(?j=KO3|9KTc)RCxR5e7oFs$-
zqaV|GEY@7y9u{bEhJ%qcn~qZgQ|S|jYjodu*|v|g^uO}csen-Z+{BBh7%@Ze|KN53
z0Yd3WK{mq-LBOjW!>7rR@=m+aU(=@W)Fy!_y;huGcY0s1YF#^hA$rbt6I?{wze%GL
zZvz?76CEB>Nr2%^ewDJ;(rm;(Pj}Du{|(7(zI3%PWrya3EDi>yoqb#GTN64lTl?_x
zGgFYMqVi;b%8}G^&X>i{bNJ_Bz%M_!ATXYE8NP+@Km@AIIVdO57g2&DoO?KuQN-*b
zPr(J*0LkD8f1pcC4dUNqzMa|Cp=by9`_L~kKc6XM$Km4Qjz;3L^@bo3yq1=g_0fL2
ze1`SuK=7gnkv9MmlYHOp^Z&AuQKq0)HY89u3^+%lc2bFQo9C~7MO|it3HG6flSYO4
zS|ss)z1MhrSklwuw^?syXlZGQ|4ljwM?ooCrod~1>sTesKT=ciw2~k98xU@qQ1!za
zBHP@;^ZV%#rhLV@anxxlUb5XSH%a0*svt?7zKl9TnuDAOY$h9(9prpVHA^G&-${uM
z`TjU!5v2omRww|OaGN>zG>lhER5$`5u+7RPeyZV)z0h?oH78}+^`LDr@)~i^?EHlH
zd+L_D&ND%7`F?X+R|s^!<JU>Ya_+dsMoY9f+l~iS?X6C4Qv?z@Na4F)TXg-gj1c);
zSTl4a>t4J@=M4pI!DS_k_0P)SdJ&Sz7vXp({R4a{8#%d2QILOV5MW_Zf=pMIL+#x*
zCGey@H`y4y6#Jf)oaFA!KO?N0@zTYU^TWO5r{m7-d=Bu(%?2UI3RxvuxW{;W|9eoH
z(L@V6&A~rVZu#O?z2k%lM9G&0I_Q@Cs-YN2zo6iQFo?(V*hkj_cXUQxHK+7>{yUCM
zq+e@v_*_LlIr&~M=$Lg4a2N&5{5s{QhDsqTy&I#-gBr1bI`gvznpWl8q5Hc?$N+ay
z{k?uSC0YTY=*|bqGhNc|2}Ytz5kv#$pRxTV+4aj8!$`?N_PA=llk>_E^gcu&@c`eR
z2ngCEIU4Gt3a$P%RI=UZ=$O|H-G+cHKZy!|pF?uutruh$55ywmw)=gb%3`Rhf|XFg
zV^aHN-IR8h^(F{G$=<-g@CPl=0+LlQvk8P5u#{v(5NRf&%v3G~n0W4205}NlE-5F4
zx{B-;GLBmXZFYKtoGHWV*-3ZXoN$H6jn1Y8k#2IX;HMu?G9>`4e>)~)H<4Oht63i^
zGE?fFjMkNMTGD59uAgtaN#jU4fPQq)B@NH|<kgzJTmWWYQ&daAu5t_gDH`)#9wTPU
zHV2Gwl%9kQ)I=FcbaQGiG~n=4L)hp^pApOTabn@FtQUX8Z;09JRCcTdLsC7rouFYs
z@7M84l(@n+5^D?$6}prn*kcsTNHr@Pkr3f#L>EjbQDYup2<8)lwkHK-Ha09YiHNe7
z*ZWhQ=@g-$%lR@QK9>y!$SOcs5?)>R9p69_(wJns($X)?($b?b^SNmE7L<Ts#~nz=
z{QR}242(7RX`!Hzbu(;o`A0mkm=Pn*MRiMdOx1|bdr)0NN-sh(U#T!2_Mo`cI5ibT
z#t;}neScxl#zR@NhyP#Q{juOK$4$cHoEI7!FCM!tdvT)}Bw<qO-;{#~>fDS3;C8wT
zkF`&ui7hSCpJO(9|IG!D=BE!%;m|%osLALM3nu!1L0sV1cK(bC9VK-*p3ZBmo8&!H
zA`uY+8+GYxzw2)Kd#b_7<`f3)$;QZF2F!J&4@~;jk&yJTJ(vQOjy>rW6=1+h$-;88
zrcq~-<v1tHbx{a}kV{zEl5lD@7&nS-Vs9W|h|D8sq9=m!T9^;UQ8u10V9H1LjC@sU
z2=3=P<OL#ckF5Mjbhvnbxr<RaDmO-rOJ)5n!Sq(Bs874~aaz!P-;Wjt<_#=wzm)}l
z(!uH*{>+O`(%L`<SNj4PAaxo@5}2CKSQ*Eg5%(UEE>b{@oEjMt%_cVbhDS^KkBmgI
zw@XKmwWuKMO^GW^a|~RTh~|iz<R0~Upr*d(5A8=Ss7y|RA@aHwRZ*FO?9M{Gn-tO(
z30~XY4?JWbr9&rz-`L&P9}&v<LKyo}A;Abjk?in%`h34^yK*ryGNOhNNbpz15L!Mt
zMVP|$r}_^d2^DsiC@9KiehNzlhya>?Mlq0rd%0)Cfpc~D4(gt{5x2TD3^rcOlkv7X
z2KxWt02K!=2hS^wTSE_Iq<GF3htEzv4lr~n-{JS}h@_5Fe;0|1pH;hDtOkjYy9f%9
zc8={>&sppQUQ2A|P{tS2Z+8J%^OH$@pqSo9-FAbdEgmrl4fFq`8GE*v-xAuNjmOmn
z6Z5-yRMSII%E+R-O1OcO_0|7Wj*g(mjvGs1>YB0Q-1-i+hF28j7k{B5ABp6bh=Zs8
zbMG151pQtj)=H4LtY6l~OA%$Zk`(0MGheHgRo*u%ra2uPTwH2a>&>YCOC7hs6NcnV
zF1;U=<X5BU`<kxjam7Ve8qQePN=jM|^b#EvzQe{x$I|0vkqd>)l3`_MCF#-d)WG~>
zw4mQfT+m^$Fux)TxH*y4-fo{=&b|LtSq=Y4u5P}I31Sf6{k=l;N7Ev^;YT2fl-Jw7
z&|oH}4^U3AyU9I<ONeE)f6Wo0#lTD?E5((7I*#{kM!n3FN2z8O4pRk+%ex%#`)4kg
zAo&YS#!xD%TdY+wIMxXvqtb;)-n3ZBvgHf6xp4Bl9dT;0r^27j@H7dG`+z$wRC0Js
zM9IAQIE5M?miVMu@zO-?yj3``XkNiqZ@xA4*2?e2(o}^}6moODE>9O$5gHvE3}kDC
zk+iERaV`Z<jVgNDK_~{d5TBLZb+wH)nEo7qqrnrh{no!}j;8?BpMTg2r1VPXbo@M3
z(9=A?Qke5{Q8E9>3G0#Igirnn`0C=Imvx1bE-DH$g*SA&kD4p!DzcvwA6_ah3Ug*{
z9&l3jRzyT9cF-=EJHaqFHy8K?0AwQ*&g-u@$9A=zO0pnHM4PvK@?>aC3xIPTDa-b1
z@Q9+=#~1XegXfl=K6HJ)p4Ean?K?7mt@V5$rEA2%*$mL{uPwBx9S+)ulxA3L=HsB$
zQZVuxLwGm}GJ6u@8-k2?Fe3hAm1qQ(fL;zW8cAUDPE>DR)bCY&x7tPGLMJRNEJh*1
zGY(3ilt81-?EqlsDMMT|XzFDnjhZ5{*?Dy^RlUBbe9>`hX>=HNp$9___$F7N&6#a1
z^a-$+gbUA)U8eMVI9o&?UjHQ$Q&}&65=aV_UWAT{I+;omF#Pq*=0iu_y}LHjnZ-y*
z(LT5~ik*(|VT69h`cXp3r{{Ak*u|hHX>uq>uZn=Wv%{^!xaoG}&Fg$3^Ry{|3dl9b
z!-@!!BjB^E9O?HY^6oWUX$y-|;HDaq@jbXkoYv!N!JXca>#B5VJbyqtQRvSy{A+-O
zVKJ_!!FymPLSDaHe({zZuD&qECo#C_r0tL4fAlNd?OH-(A)yodVqCS7SY5)4I0@=G
zZuCn=lYTV~avf|PntZV}bernrs5b4du}g*ToWYFZ4VxOQT;<qi@N{@xqv*l+iwMTn
z;n(B8F`v|d#|Yg-bNa>&p3(&bt^Ez!s2aFULe!LyNxOgO<CtA|0@ddC7$=8^ahUac
z1IH5+F*@4c5E<s_R$jtcs7}Jwkte$TDzljmP1-oh#m_~1$7~jDClq^7WqQ}T|C^9;
zQz{qot%!O`;Z<+8%hSy195STzouSMd67|GfDCZ3JJ~-|KK6*QIcy#j#2?%`n4{D#e
z#V5w!4jL{EY_-}s;~IVoyj?-2xIg}VumK(9G%w3xV%??BxVO)CuKoI?=JwWs{IdVi
zgoLjI+}$sC6t8`JGES8Vc{{f0G$cj8YEF8}?n>VJw6k}$k%($aZmMTxY^Q}-`Q%N0
zzeacL{45{rGMrzQcE4-DK8ba5!yE9wAHYaStp0t8eIK?lSQI10BVO8?$D{Ui9k3v7
zP#*?DR#rN<LNtPGBvq3?cO1Oi3hBj6ObB(AD!*wOpO4MI7*wAc%3716g-OS0vluqF
zS23dU@aTL`&!qgEveKuDH}(7ZF3NOycQoS7CN@m<q3?us4W^(OAsY}Bwsj@Xm&&0H
zIh9T2%Cl_+*Y$;oFXt4<U2h%-!Z`T}W8Aj=U_MnyNRi*a{UfUIi0?y_(KJUSMf0!8
z(%CJunR*M4m9F~M(ThqrQ)4)k5)l!Rbn7Nr^7`38?i3WlEV5sd>v)@7V=rsnU$Z+3
zvvh=!ILrlFOkUA84-*~nv_v4>9Jxq1h$<>FqUQR_&i%nOpusy|O9RW_!LAB1?LSK!
zs5<|-{FF?SyV4b?PPI%1+u6}~I$JFiLxtOR0l9co&&2`@9Yn>T;%{__Vy$A&rUW}~
zfJH;8&RS#OvgM_5{uys;B|2_KU?ogbhoyn?mjV(7^s|8m3kWlLG5?!UjS|GrCXf-W
zDgvOxh*z5J7b&eyU9HW6c5mHoKOs9fYM5Z*(Yeoxr<@Jkrvl&O{zd{el|!=u%2+4i
zL7T!+EOEjx&_1cP3Ym&){BvTX5IeuQYn`n)2$0Kc&8O1Xit7dp(jbJp5yvX4=!?3`
zzlo~#WdPWjddsIt%EDb0_bdK9%+a-zg1|!UHkS*xm?1I;M@LZ$3yX;LhGVafWh7!e
zj1Ut$SegeyS?j(UBb}np@+GEBS!;3c>SfX?0m3s-jsnXyA62xrZNw-z{kk1`s1%U2
zrv#M4^N505ne_azQ4!V1o_7~bzRmB5<ji7thCpjt3z$+fgu(*@l<ELi$~oz=5Wr<;
zD=X4m$#mMObe?06yRlm{Vt#FuY$%UgTyOq0Zy1rCc0#OcIoy|5QMqyBS0AU`B~#%f
zw`jL?t^rh?a@|?R*LeiT>g2NIW)R3GSgjr`)WJXCHZW>KR)|Ry*`Ottn(q>9Q<M}|
zHxrAeWWJ5VYWx*vaxR6<N~hEnt3|`n9s6OkEl%SuFSWU-EZ1dQ!+nZVUK}!-MS?Y%
z5T9a!i^kxDHBUil`aNC_fXncGY9E#6sVK?R+>@Vhrqlbac>A~3sI$EA$YL24g?K>W
z4AFB&@Uy~CCkYkf=10jwhmP^s*w|t!xlA)x-mLsW6MIX+4=jr|B%F8S0zeyxA5(X@
zsCS0Bm43EWT2mwnv><fZS40%u9np<QWVJn+z5hE(WHu>6K_wpk`7r@D;Q5GAzN+9c
zSSzGWCyxBx_%;xSUvOk>YhR7WwBg=<vV^&46K|;SSMQ9|^@~>^H(+?ZB|^_v?`ImF
zdKH&Y7xA%uWEf?x`T6kjNEyeOdW(nbfS2k&Xcd*~SMroh%t4jCd^;*QB#A7mQWDS7
zbuY*7Tv5>gDTP4)3A}QqF1Hib&JQ2PEibN+_V?T@n<o5(k-)VJ{hzP9w3cb(lukP0
zu+VKPJ49;zc54+i?>1Gm{=O^w20f#|Z;#b%pc>7?xpvg`5HphgKiWhXHczbaB$YHi
z;f&sCi#Bi{(osZtg%C@kxinf|6kxi~Mz$ZEjwGi+{H?jYJn0%4JfswsWEp<lKO+#A
zY(F-Q>>!Fwe6$N;<U2?AB#pVjT4K;{9*s~$dV&Wni*?EO@gFo_G*C(?AwxNmu;QkZ
z^U(Q=>I1j|1xkvz0sY7PiZDra^(l-xv>>j}A-Jpz8&aVbk&|ETA@I5S_fq<z%#fL)
zD7B^T^GEh5g&(i3=S4$Pk|#w*Z+|dxAPT@vMQLrOHFl`oP)e!JE8>c$Drc3AKApty
z>qQi10c*9TW@$$I1xo4>jNW|2**FM=wA1^x+%M}K+PfkCLosaR2yEG^*3Rf_nmj{e
zCqplD-CL5~@SAFf^Annhj}+5Ei-vo9jw%!y003Z4lWD(s^o;B@g6Mk3yFZXgzeOSP
zzJ<<1f=Nqj<Mv!O+x$94&rbdPk<nxY*=@zDNJw$N(`z8;(e}rt4HGR$%3s0WYTEsy
zkU#s499cF+f~H0@AdV*@nSL*BzE~l!x1=WJ_UN#JIy`)&VAr9gdFRTt*61WJO0I51
zL_z`<C<QOY4&37DuKSueP<CCISh$o_C2VKKYEm-=5gTOk$e3{?7UF+Op7%Q?X=wPS
zPEVHOawBvUZ$`W=J#)0$V>9c8Mb1ooQ2g^6kkFo;M~vsh7c<iYC}h_of8+Z1b>*Vv
zB=Rfc$X0>omXB?4bhYtUYwjJCu~jn&^ZGEe$`RC_6p7ay+cBeg{_JtjUtBIIo=kc1
zX-2VI;5-y5s^ywJaSCVDX{DZ-nL$>%)Y0PMpw1tyNzL`%TST+ON=@u*Y+gxWT;}Vl
zsq&lpUoF5cN85(rNe>K>lk%ffW1{+qoLZD1w1l|@-}iK`CjJ&ds;!mU_IUlQ*DNZL
zy{$$8p0l{t;L3!};|LB?MQgtAcx=q)gIYr(VPPpELz*(XDl`9522f4B?Idzy19P~V
zA^tVB5xMXDdmqB8r1=MYs;E$CQF66QVGG=7^CCvF=8*R^gJ)rx5fOvB!XC-ubr@M@
z0du;D2P`qt$fUbb>3QD@>hMxQqD-?FBocR@%r+|`?(h$UOHDg+h}GDWk%*YUAXveY
zOyB>WpSivZnXUPSX2{h3XXdY1_5D!JGzX%nVt|W?80-G|&!lqflCp52+@0IgOjQGu
z{H>l`3plO}7xyfA@A$8ETi;gO=hLBhJyL*j@FRk3l!R2c7=^Q&kocPxW2ixI?6q4d
zYhmM4%K*X%tPmlbq;pgp?-V~9a8X^sc@*M?+rO!#Ma&S1_6QVuIb{{>@`z5%H1uA-
z^SB=*AeB<!@uwRW!O=3bO8ErmNSnb*1W2Et{+qbHn@|u%^Qff4(aez8h3Dd<^AK1d
zp`i7ywtE~3k1DeIFMLD=c}yzH4s9?B-&M)D%1GK?2DYDBb6M5PQGDjC(rN~%(c*X1
z(W!|>oIz4Xan5aJsa7wVn9C$cct&8EkCwnE+925}E?yD3r_tURnO7WmAI?af#dcd?
zT?%%m#-m5Q-gOt&TRuMjPU<@`(jpux`cnnUqz*Ts%%Ewlc|U(<Tpl@j>E6S;e_(L?
z0lJGa9OeD}?CSk!!j(4OR%&c?8X5kV6Z3*Y7!Wr22thQLFtTOdnIMELP?)&;6*1#@
z?jO~sFxbKGA6*kB0|q=8Bj~WtiWtycLVxrU8ccp1Vp}2~U5zzYG1J|BFnIDWeb)CJ
z5W1B8)hXX#_5JPob{xfr%%chAM!HI_hntk@<t@wEWLBg@tpne#wS+@rQAx(dqvIrc
zEg>sV5%fC;BU8XI_5Gr1m7GL->E(2@khHdu<*7K#EhR>DelvQQjy$D<8ECS=IkeAi
z|I!0E_$J<c_7ML3{M<%^jboXUE2Wa5U|D_bCJ>9J`fk1#G?njmo`tiSN$~UV(-1$q
z_~g1zlW*#sDSgOf;QBHXw)Pf36>?FqwrqV=udv!xcc6+h<6`n>z_&EbbTy#Vq=f~(
zCO!bQK{rH`m1{!1LuhIw1EdI;Ar>ji*R=YD=S@<&H}MHcv%+FD96K(iSA`l2yqc{`
zco;ZFIVf%E;uG?=w8`uA$QWAr8<2DsERM+6CLe0v6Am<W$@AhU(kV9ecsy;D6zA7o
z^9lzOMOCi|d6)HVGosJ&e&HxIHVzDdX&UI*%)Op<mP{K9TJk)8kzR6QmlX#CROxt0
z3eQnUQB2iM9U1QI5*=^&qNGIQkL9w#pus}XMz2{n{vySsW?=ZG2&J1tBn!jTz{*}|
znZj_9prnc*YW?+bH8~ZCvd&BSdXFH7%yBVsXUXK7#nh#)T02+(Yi*Vj`&A0~kt{Sz
z3m=b=l~UNdF;A#1B<SoHn#~DRJ;CI4Y+mPd_ghw0(h;N4U~>#yUj(a*XYk1`4sM=!
zn-gJzHh~=B$fEv9+TuJ&tzl{MI3HvX`3Jd#Bzumyx#DBlFh<45qU&kXlf|U&#a6i5
zapH1CMcz!Ui@`cYe5Z%Y;Kg2Np?S2)RNSHO*x(XJZGJnL`(=RksRL_>Xzd^G#8xIj
zj}R4x!^_Kqj9^0sv?2*FbmcF#@~A1vc(Ig}56LtZ!@=sF0GKJE<0LDZ9bhqg19%0Z
z&Qc?hyYiTk3gy&5pku~F<lj>3(ughryF<zF6<tS;c*D@GmK<RXB;pjA49*JCMI?sN
z&4%m9xweWdFZXftDWABm?WcxKvyi?ct$MKy*Q+gqd?z@$CPLo&7;iPJ=FyV5Uuhrz
z>V|^#HaI4asGi3eL=@)1zw>{vzxxP%HM_!df$&726a-9DC#2>dIwpL1b@jq#Y4Fxc
z=%#Y-20d{BU_Pv$0+f&M_O=&Z-=qLc!~hVLOXEe8KCpz89#iG#F+|Q(2y=>QGND|C
z*<ABbWlZ16touQRKT+_LW3_Qi3W#WZ?_7x*tI%t*>2kH{W<swLVsa2bgW%RH7nF~9
zl7RR@5Bu508@#M@{gjRb%iUgU7K{4Ch|++KA_QNs9}I&Po|mMT(JmYb^w%M^z=(sf
z_Aok>c=>5P1cD+E3MNx<IBqQp-=$SMD>kb1y2fab^_Fibhoy`$e<c5Ay70K3&IpN+
zfb~GxdJ**hD&#6ZtSRd2U>YK#1R-NSJ#1eQl-BN$m%i3yWuxomh5ow}m|QW_)6-||
zx>eLqQj5NMd%tD_;an`}K8zLgYF%A~3|*D*XVll(tsg$y4HVbBb60-L@}Fpv;B`@I
zxNC{f7)PuxGMdU~`>J^rgpqgyY#8assW<9OW7evmU7w*)h|JO6Zr@8_bXXchIV8KQ
zq{-2`gcNp+l`(>93>k^NZ_vIW9miCjVUo`6-NdI5-D`QjJe-GzfdWp^pKxTgTJ@IS
ztv$pmjD*A`+v6QI+xv!NRGV?!#Bsp8vbqFTmXEy@1;P)k^zYbe(+p&5!FpVp4$koV
zgZ$HE|8cJG%u$Mj=!#Zqn#MfOqQ}Ry(hmz8`#-|M8(QAkswCv#1%q9Ft9x!x=*WZQ
zs@@t0I=Mxkt5^}+mBGv8!pFQ-n~hSb;6?N(t_VG@!WEZHa8r{wBst|5&xuYfs=Z6X
z|Hkt97ZW|nI&`b5SFF&%roF2>(#7Pt4f4IC%2zCmS5{lu+PB_0AyS$?mp+5&epro{
zf4KjYayuB4IK7~*&w$=({ADUl&^S@Xyllj(!we$vLPwn`GVUH54O}P;+)iR>Ah1TT
zPSbcblPfXVIzIQZ>YtbO@T?mH2~u9;Hp?5p1ZC%v31c4cB9Ly9?fyK&Dgo5n5+@}k
zB_A7S4w1n0KGXqH`9W_?5ylIuF`rj4w=Ujyb$4GDSu#*r=*Gszm}^U5YI^RtTsppp
zoc}Oo=Rl3Y`P69dg$vtD-q7<}($d{WJ9H0Cps-Y!B&M^N9UlrTYGeexp5+L#(9KY=
zY+z83!<EqGc0h~}Gz}aSI~J2paAJIp4-ZCAu{j=W^;)@`7w@NLdcpywxt_c?pXnai
z6#@75QoW_`$jzhM=f8JbhX@FeUW}|@6M<MzjGzQoB9g^W<>C1Bt{#ts=MJI;QqlCA
zu2xEn*1s-p?^OD4NRA7H`<HK>*fogk6{bdn2}K1FYAEdhf*mLbT{jc*1WP_a3R&%_
zeQ681@2uQ`C7DKdATF&58}J>B*b=VU5h;qD21TW7d03C@M~jR?sn$-+#lzBjSGmwF
z-rJqb`KXPsjEQzbG!kwOG_-MS?A?@*Z}XC;1R~ebg>@cHTz&qT=ACd=i?|luZVprV
z_$UL`5Qpk*%}}(@c_3UVR-PuFfFSS`FB{UAm?8tzJOC}*HV@_g)4pva&<B9kB0y57
zw+~fhP-buNk>97tGOiA#h1pTJ-#ZJgJ-u#f*11R*3k#=d&Lzo$sjB3jpnz7G#wzgF
z$CHV@7$~hXi}c{7v%l5`q#29vgQi3iyL(TY`x{)k(82=8x1gX<=uwMkMmO=xhgzb=
zj-QhK;-d4_l@W2{PSN9y=KJ(PC#=HG*&1^H6>@#Fx|POnZZPn}$CsIF*`nb$3!IdN
zx-<kJw$pMbT30Z~rBXxXNW`P1gIG=bYw|sx;KE?yUZzntT;|-({Lts>M7OKJ))XJj
z{?|i`<^r^uirbIjD#PJlrFZ>1o>7F^Fpo>jf&h-Yt8e&}Zg?Xb&WLf>{gH9+{~C$w
zK^?fDKTDiA@cR`c(UNk`tw^C#u^buASNT=><1{X&!#)xxBNIl7UvXJqTl4vGr=U1r
zCCSku_v8`ss?K4aT#9i)$7!eBiOtW?^wGjUj9_MPYij^AlgP*mLrF0w_1X6#&Rw_b
zHO>o-n5wj7;2Gbm(Snt{*}3t15Z`o%T<#$)6@%aO^FAkpO*oPg!wUr-1!EQh4DaQS
zS*rcgW{ISQHU1L#f-1(nbbVA}|As<Ll_$$2n~Z_u_Og`Q;D*RlZgjst{RTDjcy~6O
z8UNhqFiEO4Dli+!7?QUq<12q{UAmZ1xVa5z59*fYLDt=~>PFa+c$W2ex$9JsYW~~f
zRgZ0_^xtrWvS|XvunCe=@Y3GKcyfq=Fr<VNZuy|XZza}2M0n&ZH_3Yh#hTl?3eOJt
zuj<N3xL>_|u5$H0_7l|-;;&`E-g9`$1Xj*gouQtjB})~+ZRwqHCx%Rgp+vj%&de#(
zBo18`U}qC=3T55HH=Vo6uV}&9@D8$R;s*&)sieq%$*#B{Vu^zP*T$C~FE27E*V(4B
z0Cay<vrQ2pvb7xmt+n2^zj6v3R9O{Viy9gAZ3)lE0w6`Jsjwa3>5!UY5axYsD571o
zqmBlNUbMmM|8<J_z@77=)O-5+OuJMhZ%B&5l3J?>d3qkHzRwx4X5nLF7nf@`NP)3G
zU04Sgh9-XLz6|1vYrU3>I=-P>J{=I4R_)IiAX+=>2Gl)9``<oG<PN|~){PD`hrHU!
z>Hi7(Cygt_9T_nNUp&u{q;VH9O!GZf3Fo(24$3=tn4}6CI0VpH<q^$h8q$gI_g?ne
zUbDRsIi*UENnL(f!x44cItSGm7$)u(xGZZ)aTcApSnT@Rs0nU2jk6a4#$^X%GTkJi
zSZA6tdS3ME-qr*qI;Z7CMnZZ~9yNptI0(sD-HBUg=Zv@B3AqFp6#m%^M%qEzV<G13
z?sVBkZ=$Rnc+!)@*_zswpQ!lp>m<vnZ!f`W<!qo}mSNl^UDv%sm4Y1nM;Eg+eDWf%
zF!RmlOTPR=N-=Um4kU;54mG?N`-$fa4nr>zHM7Matj6BJSr3t@wvix0?7B~&yOd(H
z)gbkY(+dJy@R;aF#0eF76GTO=lejD_$*9&VI10$TC1rT@FiN$EqmMuE%4vH(=Nc5Z
zp<Ek!;V2mGzA<6-$+_xtsx0Vzw7R>!KJr)oI3%Z)`RfP*q<I$<y>?)=?Zj})AU613
zo1bn0gq?yp*Rf^()xOBcHK<A1+8dp@7fCm-Bu&kaL1nIYE565qq=9B{pkBG){(L3j
zaPve=(cW(-N(f8h*}VAhxw`a%JG=ay?K-m<*QHI7{&7!P$bkWk-=^s;Wv@opkT|H;
z{ya_JYVGK6i40YL*)}p1U%R%W50n6}I!|#-0L9s;8NMMUt8x@XXoDrirPO35MqaH&
zONO`!!FIqsX?b#PirrP`$N6Sm%KJOS?d>fM1L9(cvPzO10WVljYkXm0VPbO9jzimv
z<Cv<?hc~pw3XcR_m5)~USBF-pXJO9I@F3VcyIY$eHN6SrLveQr|9q#d&JQmkSMZD`
z8^f{23#;{eDtyilNmUml&Wxx(MR6^T*lB#O_QYH%{=EQ~tCg`}S>IaEDP6!h7Hge1
z|GNiLRae~9%0LJw`v%ZIwkN}1%qUfdqBpNrzc_`DF<cH;fH*N!0W(JEIr%@RU#Jct
z%}w%qYhsAU>8#Y~m$Jz!d#QXjmCa9ML<J?RAPvP-=QUb;lLyj0zz!#QvDAp<!lWd-
zRw|phANSRZ1x4U#MBo4x`jUWs_PW~K4JgOb69{68Q#T?R!H+bb^dViXI<6Q4+rDo~
z>Cg(=yQf9|Mq|H1!yXF1eIAYxe5HniHQ73P*LGGYN#f19)yZDfL#|7}LEnCq4tGBo
z``t2K@C?SZnGn%e#AYCj`=qP*r!s=<GIL3E%7|&~vkpJwAeK`qbSCu~zz+m8|1!N@
zSP+bFSk(#1t;lZ3ZX!)hEr4n=L_<T9d-CX58*6IHup}Uc)}TgBw+dQn+>`0Vf?cN;
z`f7jdXiyzyU|YP<`PekKbq!5;Vt>>W$ERe~2b57b?RVjIaF$)V?v1TM06!n+<b@XM
ziqz-+BAnd6QlK-1nH(OyoKHO5&52>BAR|(-Di4^Tw&&>fKlki3^YcPa3TN~q(NuZ7
zv(EY3ouX8Ng#ixqQG*Wyn;u(21y%<6u6)O%mUIV{*m9_ty1-~C%F6Po;pK`GRQ@fo
zdX2m78W>)`e5hAoopa(#N?xRW_OlcET2Nk8Kxk05C{}pT5$3Dol2&;6bVrT-y?f25
zYd#n}qh_SpiKyz#D<4;>SxmNGdp>A7_};dSCU>z0BdTcEqAfua79LrkHbwYeqfgfL
z^~!dXS?*Yf)=QIbp{z}V7AA2V8(%8||8WFTBL0O5taY=-HBmvaP{R@8&Rptb$W^1l
z-}*S*kB-PUBeKk&S8C|f`*$bN+tZIPse|o5M|y{BEBjv!m5gT?x4&RFNn_G8`*s+P
z7m-24;ZM5FI8^n$N>aq<CP<;UC<2CpUH~4IM^A;Pqj53jKn9;y+B7mT=36dI{1lHL
zKWA82Sh4bX?_ou`2lGAbLPDZuCt;Fqm%bAVvQCbCjGSQ67Ai*EK?!__JN?-&V^lah
zRVFUiuZqym4S{wJij2g@1s6igI<1PmPGeQ*MybJX>;An_m|Y;Df2W^tu7)QWIOz{b
z9^kxUN6+>!tTsTca`o=zT~&*(mND6ODf{NR#5?Aeo6Ix%xGMD`YO`%2r3-rF%Hqb@
z7c9R|`CK&!#EP<7{zUcrQ{cF^{UAs-7K0LtQh56?B{zAB%X(QUoqxreCX!0jvZZqI
zSOQUoWRFT~JLQqXu+}(NsiOs|w+2Qw+Tzwr$u{Ip0ajMut^ZQM*~eurUSY@y6msdb
zIZ<*_2G7m{u?iUa&u9uA^P!8IqDf&PcFAO$Z@KQ)<nPp5?Ksoc7UMfy+0xd&i1{qJ
z;!sgDXEwt7c7L7XVR8JKuo$$c*>-4A?{vTnf`(S7&g(8k1^#l<Vc`XuoxM3X<hV~g
zr-pnEUsqEtO<nsS2}&^K>y*QZb`O{a-JD~)XU=Z|Vvtlpzk57gHW05_7z>rk0_>D#
zz9}#vS1MveF+}3CO$3=1T39xFXtOL#MQN@d^C1-V4%SUH#VLaY@M*fr?!HWAKI%1N
zZ45P5urGNkQ5Kgg*;5~j8tH!OrJ?4ah?$K~@-&V^!<52|HT)6r*n~qnB(7Ppc!>!E
z&!tRo%<Uf=S<>T92NkFm_3wwT91W2`y698%kbKuVb89LW<<WGZH{y<^{$u)p(KE;Z
zBnz7{y$_gb$_XR$9k+R-JCGA$Nw)1f;<Kb}_Us7cj3fh?xH*$@mKI9s;cGeSZCApm
zS0Js+!Lk8wrZ4~EjZp<S@wERQ8Mv~JZN~awWRc{mNJeOc0#Dag-Y5py8y~D${8AU)
ze&Qsj4G!u^`#~XNjNS)w{6<j7R{9OEGr3*giwOX}Yp7w*7sL7m$9gM8{myi6>w`xc
zo9Xv#Aq({Y-HZL@8HVjg>ia~__QoBe$D*s-qnCJ0Y6obEBHON9>$%?Zz=mvH3qxD0
z_FUS#wj~E_$@rV&BRyTw>)Z=~)Fo6rtSRFUj?`;)U<_1<AHsO7T!<7*QOr+Mnd>v7
zAUqZJ-%klFsST~n1^0`Xvkjubz|wU##`Jp?WBRS#!+RT-U9x}-UV_~Tk)YDZbE5FI
zVlt@vQ=<yB)1|T|-*^3WeV17T?u+UG2vK;!^j6{T*@8fuaxb$%yB3F2Z5HM97bJIe
z&jB2`92wniAwmi4DbfIEAxm;~P78!<7Vyn%7LX=NPJb4=*CzTmE4&hqgZHoiY-JXY
z_NQH-eM7#)qr^`d6tyj$bUL6ErA(^AO<P%67M%~xCeXS10~YRD-({LCAdD^@$ppP$
z^v}a`qY(dLs<Ih2x=j7eYd}6LE&H<)_QF<-)KyA8^$1v!hb@;@Rb!$-14^wah#!F%
z!3uBZ-`l%N;b<U{otVmKz`cn(o#2$Nko_5rAHL$EeJbuzyF^k*KRPz4>&(As$zFhF
zQb%1<_Ittw87Tt*;-%lwqJ|ZM8d*%jrJOH2xzk}o049ww<?nyK#xIjawVpiqe%M)(
zhm{qY4N3CR9kyN*-;Dj9xn_o<6c*hUL!*R7P))8iE{?H*!MNpeLlK7ZXC=q$=^sXc
zLKnSQ+C{*`PxJjoy|({Xdlp`pOr2E%l2<w@rE_#>2m3oUTVK)dLWwXDs?;8Q;$s?q
zd3A}B$EQRR#b1Fhk8}u+hgd<!cWTKFAJ-B#KQKqyx1{a00_aRJp(G+XXTHe!a5V(S
zDStkZ)0rdy2Qp^Mg7)uBrJnjC1_nOCx)ha~5F`sQ#<N-#G6lR-%V%5kU(H%Rdx!>3
z(leg?5G+zZ>2+>>AzmS@{jzz<sdG=`NiiBr;k$&qJI6~VI+X)~$&&|%u-OICa$HI0
zA0qa4OLGCnJE%j8p(sjNcG&)D^P3!{qDmV|MbuK)Ujo|@PkvG`A__Q!VX)%Y;wj>c
z=<}J?8o9r`rfn2m!Pnu-L2)~qmE>!w=WVG4J40_s5j5@l&E6U`mHx5B$;6Ocrth|!
zlSQ0GJl8R}XOERWU0Zwv24{?Gi_4Mimm1=0d9K+r%V)=%XrLAh%xAS)|5wC!MAZB)
z9)6Op99kowWSW>;Xz!`DO)tSu)CRgCarMwBCSUp;y_CUi1PQ58C!ekJ!_F}SQc%2;
z>SpK&KSuWn#9stT@hLtrIC85lGCr}Xs!J%IGF1Rj?lBI@bn7C*ok9Afsv&P0X$vU!
zdhm^NI8>yhV!;te@U_QR90BIUdX>WR4OEQY8dO@HH4V3qHV7L+OlV;_{cg0hR0kTM
zmQG>^-Cq?1$v;$+32&G3=-AZN7@?*55-_E8W~TsqA7_@A`4v_*;z1d?J>G5e(F9If
z67m6g7(<LB6(LpprVg5MMS>O?Y<}Y0CB@8IqGroMiI<6)O2sUiJ5eqSU2!|@Y?}=K
zfR6?`G;1&n(J+|qELlvdbd3+og>PE%q2=RxHEX~MlE^oO+Afq{LB~UxA<%9y4Tng2
zOLhow>$nl4S!Uk#ABZo&aPyH3=`D3uYsrr@<pzMynn^k?5o+^@Ns$i$LNs`s_h4pe
zdU@36^wo$)%siw7%^bdG#UL>a%}92l)#fx61^EDWtdYr^Ig#n00a||`3=<?z5S|8f
zVI)3Aa9Q`O4D!*veJ=dZ3^Vkxre;XXMo+?vsF%{$<=;P<g?W5*BFIEP$k3l8qgZ1;
z&xazyr`nxm=BIshtc;FeLhY!_*H2CQ&GWn(>r*#JI2$h>?17#L1ID~zER^fI1SI>_
zam75r7lPY!WT5S7n-+%Tba&&hEkEa4{-kS>Ksd$m3d!jZBdsANNbnd2dg+B!LRrc#
z@_~K$LFbPpvfrp1QLz?m@BKxI1K&0V)$)<359o19*x{ix2>nhNi$BjW<2dZ61F*aT
zZ&eZ8IkHfQKDvS>D^7LFU$es!6BA2rk+>#WnBAErMKY^7lw;yq1Kl*r{5JW7UM@L)
zi4aa^$<uE+=EiEsng=Wpu1w<%ii^5`NXX7<8s*D^nAmO;)#RKtuD~^>NYXCyXM4hB
zGN=&Qyb@(Y-%g@BNfd1T-MxW<d!do&xe9a(wdJ&nqk#H+jb2L+>y&xT4_Ez^c?FcN
z%9X>`x4A+J%_%w6MPoOKP}aD#3)3fXp}%s-#uF}FT2HUcrT!pCR+sTV&|25#;*YN}
z4brcp<-_Q=t=J^fn5_G|A5L>(z|DJyGC%)V9f0x9PobmDW!Rjl^RdlFW`ui98k%yM
zGld)8i?eOj5$zI`a1_o-egA963jA-ucH-Z4BdFAWeTk-G!-(R99^9qUK@$fDE>OtC
zp;c2qD!@rZrZcvgDm;Wa0{8Vx7SM*@RmM(&f`U{=7R`g(RzpzQWA^8ZIRCzotmpcc
z9m{OpoIGh|tfn7Mjxvn0-6(n_xKm&zJUo-H8dZTWLt6bG8cwxR`ve=HEnw+`{nn>8
zq?3tZY6430Gdr}cg!c-SW7o_!cHDQ#DQK8wyVW$~I=WIrZfv#_bmKQ|5T!3KxD<-k
z@m%}T9kw!%{5}&|-*KlusJr;;Sdm01irN!#tkdhAb7KBb@_Z@7O!=lw1*i=G599W+
zV8{q$Xq(%4J&~UO>occ(scAG*cJNhWI}D%T$jC`jMI80gxwSCIf5rp*Vqs}sOdQhP
zRC*y{1`Ih0yi72<Ai5xAB(eEG5?E$papAP@@B_@b70l?-{Nz7kN{FF>5?CkkL#EM*
zLW@VXGCH(?F|T)>Zg1GtMbqfTz{FYlK-fi*av8FhIZh+>mCmX0?M)j4vKVW)&Un@!
zqNSCvh<O}42Zjbxe82+$YGIQtVZ<o-_rHG7Hfob{Qj+Z_Uo+M;pr061#w@@NG6$A}
znQgDnT`QV<73_~zan*XqYlqU-1q#(A@1|u;*V|cY$($I`BLAK`I`tTZLM^h+lCOGg
z!GT9nVWNkK*U{Tim7Om8<G9^H!wU5Gv!m{l#Sxl$<_xHo0dSMA)BVznJsLX;`nSK|
z^Fjq+ZW|)t9bHqJBtP{tzTZco4%ueBx&f)3Z>PVUOiT&Es7EI=-@66~CiGn%E?dT=
z2bS6TKyCbPK)nIr7sliMtG&CkOxbo{(15m+7|z=rB)zWEqh?QM<i>RP*w_4E9ofDK
zM>j5Tas5LK%C`6mrTX*JiAC;5DU*r<^&of$`FE&qU;=+Y0_$5?H*zni@2P?Sq^g)u
zFK$_KSy^=Xl$T|SZFi9T>!-i}5x0QO=UWHP=ZDRAY*`!e({db_BX<5iz9xx-uNA(l
zXX>&JWFgbWVlIh1ioi|L9te15AG3=8)dIMn4(Kv2qYP|TlIkjwJb7F-bUx~$h|(`J
z5@7rc|Dzm7NDNo-)M;%p`8HHw!$~d-A>uBWBkXcGKEASXe*W}+I;I!scw$K4e%MD%
z==+94vG(?Uk|9&{3qy#nhWq&;A3*c6?KZUsmHT~!FPsska6FxBry!Ky%V&l5>+E(K
zp04Sa;n})lDsx&!*{=P(vIbckkkR_WvVoJL<t^TK6ZoQIUO4zb26(n`r&$T2J$cb3
z^_TUk&P}}}sDRyWxI!3Vm{~N&E?dGpIIk)lpr7_#rO*<Sf1{&tBCp?%a!pZ=t8MI0
zNCYa$oFWv;eQdN?j*cZ;mu`;>f?I7K?KbJiL0G}JA-SnHRK)9MmG_ILzid?%7S2|5
zsdz_sC5qOje?`W-OO6c$B@uqY@ki(6dV#|MZa4O%)!zb1v)A<c+pDS)7|ac3{Tk3Z
z=l=&nLA|~`eDTs{(q9quLvsCWVd@&nMG__gmjUJzHgN6uVYG8G`P`57)-kA68aY0A
zRKc88K`dAk$VEddxn)!--8&v$GGuzLnVCyMn6oknx3~74b-D$TRxqj>zUtc3`n+Zy
zkLSpHJst-vn*M@nR&EC(j`v)SJt9{iKkz8}yXi9*q#y6uv}GRozbzS?J~`#1mbOYZ
z)o#=h6nRYC+HNC0QUjJ9i|e&Et}^Yx4;B=kcHb2d@wZ<R#!Isc5CVqwbl^?*Q0~-u
zW=5_wn}oP=WXS31dEQ^cV!Ga4oW1`804wfgJEh5~f#XQB^>U?7E*hm$@-;4cOB1~@
z6CxC&^`Vqhprc}BX!jQu_DYpfBnn3TKZg+L5}Zgfq(*vFD3$pIoKj^a9Vjex7Dtv#
zxJ`PIQp~S@92S#!L6_)&xx@Pf*ni$!!OW#4T9;^xDI`F_Y8G^EZ(+ud#f%sbz=#2X
z%vn>WuHWOnvp6rcCL8lFDI@`=j+83!<P$%M1ybW9M1!`7)tn)9euxj{;N-<(7AAfO
zFtefR_+}$P3P_RgD}tq7oyD(QERU!M_M|1Ep9#h#gmRPgSJvnxbtwbdVF{_fRcjsN
zT11FOW`6C*O4S4l)&#O}Z6K%ibaC0R3Wg7G5+3Y-O;YYuShLN+{9glE{f7hH&osk>
z_15vst9t#@T0-kRPD@Mk2{b*iN&b34Ue~dgO_T8Y@t(^uLITMM`IZi!rF}l{2+M?>
z#Fq`bOfR$#)o(vmbtQ!m;<opTN3{Bdq>d|RD^kk6OP4N9YxV0L1M>1xvx3?uocVv>
z<p2BZegT(71e@B21}dD<+d)C8#+LoAzbisqdeb?<7mOGX2%x0p<W2^|gOQz`T@fCa
zuw<`PujnZbUn0G2s^FZ{16jGt%9vpnFm=k~{IYU2PNxfp!%j$WFoA&qXqraafpngI
z?j`<q{q>AGXE=&Up}4r1(z0@p5}VCNubw^V)2jy;j=q{*mE~M}hMVu^l<=_DL|tG%
zc#!Or-lCFrfwgt0h5pFc^CIlsJ6P!uW#*M{3$qU7dOriseCVo(kxq{!qrgj1nO^ka
z!lE^6cUAnSI4iAQzu~{EkgVpp3*@S#<zirko5vM}NmAnX-4W0U!DMtMS9&OUVu@HU
zVVe-l(Is&Lf(4i*s7NYNC=`-xBnNbbJ!XY%htq||mnWBsKAOrk2(DA0{|OzSb^@X6
zdV$Gg>WbaR9qer#A(;^CW!wIS`BzfHhFx|x?6PB0AR*36Vh0bAAsRwJQJIO9EDL+n
zEqI&gM3~^)60xQ5`iNV5R9y?D{{8pgw>e>wl2!3ed~D#gz&XczF0F+GRUlVOcthAK
z=ci0xcFf(de|C&m$x}b>+)f^oSjjQ>e-W!#?9C|+cT{+b)6gmlky5tlZZ^J@=cZ%>
z6NC_=N3U)rPyQ_;xGF|c8`{J1)R}!95y|O>hp5u6Q|{91c?5+nzuti7ik_y!mq@V@
zW^5M0dvCtVTkn2Ar;Z)C>2KFJI)34jpO`q|R=Os{H##OGGm8QJ`_ivZZ%TH)N2o<H
zY{-YnAG#Me{2=Sp3KfulFwaZ(4p#C@G|uef=r}0R{*U+P7nRtp3Vpj+$t_a(VO8n<
zh1sdI{GInNFC+n`j+8FopN}jSA6SgqjuYUJS$d%oE&wjjBwb&;NVa%Owg(pqn?o=h
z6Qh93k@RS-1d%^8QdU`g9qmD})+UgsTB{TpAJpB=+|`F(DQY_GO*cg2kzS)!@c7)B
z0y@^bI8w@jdGqFZ0|EjF4i1)m`}S=UYZ<w0nc8*S=v$-|>-qC_aw-*q**d(FHfzDA
z6R|k_pNS2-#0%}i<oIKe!5x@5bc9gMtzHM!mHj(aVdQ@bAsG?Z^=yZYDVm1M>(!=6
zDYr+&CFFZG8EiHSqQlLmtA+<he6g&w{Ez*vpZ8^Y&M7F<(u*s#d+MIYtXIz2B9zkW
zkRjH?_~O+8hgFbWs1hCB4v(kla3;6gjmdQAx7zFVVm6xz4Gp142P}dzmyQj=;mmKH
z^dhC4S$CUa@=lrcb7|LA+bRcU<a<{I*p$#N(U!CByfi$tuuP+S2RN9gqDv(&JT@U-
zQFz{NQTkQ7^u@(Fk&~s|;5{iM>rmxte(=eK;$NDHi5P;i>5elkK+<wyuJkfUN=6De
zH^9s}QWA9HHlk}AXquL0Hk(11AgBwJCL22>;8G!=W!oJNALv9;@avLpDMk#cY@HLk
zd@P+Ii8U5UUDwkTMUgg}t>!OM2ytSU0IFjD=kwFz$6a=VFiL+lTCS@Nk&NIYsi>c(
ze749~D1CmMXY?0f;&U4gZ2b7tPr5|Qdt2gsWo2cJDgcW;=oqy{+zg;=nzsB{T(32W
zjO-rLJ;5^f;cFs8JzmLkU*-)AkBk4i&_!}ik(v-6ZRMKbA-wi|;r{GG_57l&l+?z$
zbLHmdmWRj1f4xQb-qD;Q<MVVEJ3}=>gB6DM3m`eegV}7xY%#uHG|lKyttdv)r`zqu
zW;1R@UauEbJ#?g;*=)vDb&Qm7)`&oUOmmZ*;$}iw08!0ICM}XWJEeDjNlx1Gy4wrV
z)6;>og%FCAQWrvqnH63BdFSYer}}g;=9pdRl_**$D0A)~6VbnKyp6Z!6!qCkK`QY2
zNg?Tn|BRA(z`v7{#D6Q!V3-hGjbs#po-K0)Ato&$UeP%Ph0`RQE?@|TdmJ!ap(i;y
z%x1H6yWPpvA6UYA*9u7hrw8n2$SH%uN^q-&g)-0zk%15%aM&3wQK1^c`a4<qd+UT!
z_2}fGSC^ycU{>Oi<d;LaTcGM-HAA=qVnU$8TR+(*w0F(#nx-ZDo~kW~$ZAqQpSLW2
z?C5a<F_WhI*`J8G1^lE--m$dPe@cFRL7Pf^`m?Zlk9a;hoZIbI*{Xj-La|)Xn&$_Q
znj3fCPNbCUmn~b?ShJ3{gVocid&iJan?+!=2)aaD<If))^xe`;<yvvMj);NeOb=NF
z>XXIUDX9Q;MWnB&l*o5iNcSB>X<A6GlO7(rJd(&zGj@wGgrvztfW!D6c>Dbi`QN;S
zm=pz@&Bo4Md)Tpg4Hk=qtn3^%ZP`j<*IJv^^iMx$)8^k%HH~*(e;J3vj;5+uEfzE>
z>D1nW5JoTTf-;?`W)$FAB|T{HzLd48<U$BDpi3$BI@^`voV28+n{BtB(c2LR;Py&!
z)v(a77yec*x^=K(HVOK6v9Wc(vm2arZ6A`UQ>6~9tonIm@{b}8jWZAo5`tb563hx=
z4=xn8$lvf9<Aq3Cj>%;vSaTAELX4u)359rFr@Ju3XdeYatKa*R9_!XYRaH?|R<=DP
zq$<%Sau6J-Q)|B13E8DkS;rEmRc$YJ8iKNGB-nkdmR=1iJcinLSQ_lVWmGAvwmKSJ
z4A_8kB<vATH@f{4jADO>stA1RrHynbbgRbm5Eygv?VIh<Af*6GUG@LmTwciXLO~^H
zKFKzbgK7`=c(wyncea7HW09;Tb=LfE<HwFhNbxD+__e(|OfrznRHXcP-<R`G77#zq
zJowy(0~@Cgp5CQB_cT5F$Qb?DcU@xpk7;#Jn5SSbx~J?<QT7L*F#EvnwL20wca624
z;_^tgrnq(sL6=9bi>eX*nFT6RS2u>mcAey~iNV=Hj$7rnZJzMBt~ZCq#7_ZwRgW#s
zOxqY9({*)<q7F-FBI#zV;$Ys)67r=+!XP6KUQrZmHbaC41_tuOS2Gzmej?XheFb*A
z4U5G>MrIZnnOV$Tu!!rfz5;+1tA1t6)@@7}e>2mbeI6+#4!Z*om`o;g9p3n+fYXXC
zTp}zaHX|X(gJhNNuFw_FM|HP{#U%U#sb3ru6LWEPc6LR0Y{G3F61qN_zJFhjjtPm|
zsVXBl>G~+v?Q#lDCvC=(;x1?O31I2wa+fZ}=WtR;9v07EC^LcYs(t{NyhQAGn24`D
z6ZhmLLPMaSW5ZJweWw}e+Q^3Wo=?*HEL^y7`-BM-yg~>IQK#46S=fY&xWWTFbD&3i
zLsY6d94Laqih6Ck3A93&2;+ZXa1R&fpH|6F>l<p7izGZEp*=<l|5yb)KsVqm9jw5w
z0-hI;;dAW3Zf-6b;!`8vr9xUEl=|dXggPKT+^{&hRoImao<q1#jD59#p#gCa*s11W
zadGixT{nbgSy`FbzJ0r2m(Yn$>a6+SCXBr_ACy^~xV2V25>_b44=T;-{OnH``)Q1D
zlH<k3-O6*(VfuKxwRJ4pp&bUDcPzdvs@o#+Cy9@gQmPRViRZrgL%}#HaTPnnEcece
zu9*eu!pS#8bq26o1d7Rgw!<#e2^WXm_}2dl(;{QLd^Wgy!28i*W<R+O^2H^(d)Qq~
zB;E8D7J3T{I`D3ulMXQ!gb;YET-LYWG7bQz%Y_hvYp*gy<Dwsz(zj19R<2ruuIng@
zLRNMTJ9q6SF(H8;zBa-c%@#8jiy5H^bX^kQ7M!m%KT2nf?x7rgdTB;Ry`-YJD5dzy
zGsA{`vZUn2@b(Gs59#B0KQ7Xg8X4QAPk5*$zIxoUEfsRXZ)FopauPn-q^4ZIagVFL
zW>yxIW~c0dlR~nk5LCO%TBX3>Nt%O{lBA?0z_6UDsw%F!>Z*${w3>zJvp_L}ei)UT
zR2`D@pj$h`qURe&2!`O?m1B$vHU11Hk1OQ&WE*+KMi8~Y5Rl{QNUsE7yo65#Osmst
zbC`sg5*|^Y`c@z4(3yIB0ZLsDMx{rxJJ)bg`%>bB9UkZw1t|p*PZO!&zQok8Jkq{>
z`<uV|>Z`ps-+c1{pOIC|q_#uKtOdU$UO(nc-J*Sf6c-)u`P3m(B0fi#)6>41zw5As
zdJ@XU(3Q~=(NjT*M`-ON1c9g&$U>>hG)$&lWkuf1kyA&u+3qDL0u!IxaA4yngForg
zj=Njq{ANecF=~sr1u12j$z=NNSX}Q|P*}LH1bAC??`{jL^!6!?4BjsD%f`)LSGMAs
zvqOZUfWs<Y>fFx!(kb2Sbnj?=M3?fSyGDoG7yMRUQzNcAUa&YPb?zTA@mtfSHlSlO
z>NX0DE-$94yRGW2|50<h>PMvP*twfCh7MuIo-f$2X$wP78_ZQ>Ml)yLLWZ9|ikF}J
z7Z;p25~)ipRq>WOU^|Q?Q7@7orQ0JE(~AJ2Uq5Sg*0#$fBTnt%pu#05e)?nah^vPO
zTXv?o1EesG3$!WSRZWspq_TKp*@q=LX&;BRSNbaooAWbMcGkVdU#_#wN&d8mq@*OV
zefxG%R#qmaOqt?HO--E{85#Mh5W?sYZ4HB{)BGPDVuw>YLcdsu@UbL%HQ1HyUqn`;
zOW2=ZlldGGqVdXuxda6mBJXw|5n0Wh66WYeE2MhU0bbJ~9<Z7v&)t(xmzcUPcpdiU
z<5xu1o%hZhaQTSHP&@RAfx#UivJnq&hx+=;YPH@tY}l~%n>TO1)a7!Cyu3V-l$0dE
zh%IV-5+^46zMgX+GAQRVkaq()$8joNAy|qOH#*JU9;vhDKA!sJyq(Pvk+ZL#ZQAh4
z%Ap%yTlLb$uvJ?TV!t9-hR~;jctIeZ5%3&{_l2S?7E<mF4m0oC{QBy*H@?35+>P&V
zY}rh+zYyzp%jb`_+l|ZRYVf%cB1uuqwpOnnse7Gx3U&jmTfF$sXYmfL@bpV)Z{2iR
ztjw~R&6oYQ%el|vm1GsDj5#yJ92so744`*soAr{@f=@A<1i=oaLqu%+sDz%~ZtdNt
z+eR?jE=q-|bM@xG)rz3Tn51d-D`7|^PN$QqbMw>B%UQO31wny<%>M2N0IXIkGpE1D
zV-MfY%@gin#j0O1nG`H$v*Fo<`UZt2SgCs`lJbMXjC}_H!U`&$7&)xh{E!Y^dq#$u
zZw46NFTj-8!P=*Nn3=F(Q(Ro6Y1kEK1+#Xilizl`&^1vpa#+tj7Yzy6qU++}2G6<q
z(>cjWPE4wW%=YcuMP6Q>xbemttvBC%^Oq)*>1qI^1_%BJYaawxBc(he1Hr5q@r8*|
z5a-jP+4pPRVX3h`)}ZQ;UK_LCJKn>)kL7SyxW>drWPt#p8hKtpz%+->D-Y#zRzGL`
zomr(4w~s&!wn0!6?v8drwXS-T0#S{WDoXhpe!U0we+ingiWWkIgoTBDv1ZMh(Rq1!
z%J%KsMN(3dXoDVZ05-h0!J9H`{(DxGejuI$3XkFCvjx1Yqnwg5YyO2PU(Efgxi0K2
zuPzVU@Y?DL-G;@jQ53O8z@q~Cil&o@ECLdxl=leaQsH!N+4TCViECb2b4-eC|CzY^
zznfF@ikaTh=e_3Xb9$RB0UcWUyuP^ewj(t!z>?O!*0Dy2bIm1L2ev2*dvCiqEaItg
zQIpMP(QWa@a{6?!(LUT15*2Db3!tb>qkn>ptU{IEoo(2yrXL=-JmQnyU2J`XIL!i3
zoSFK+?b5quPcwRuNI(}=ZW<}keOe?cih{eUB+%~Uy<B?ng>>%RiN_wfpJl%oosrl6
zwt>{N1Kd3J2JRStGppCEtA1igB#I*NXwAFlRw7xfyUQ$^e71U<-9qd=V<IjcnrPe6
zy@NF|uUN&aNxH{bDXGv%h_Q<HVdfG2;%#h7c9NXoX5#4Zhab2yGBG09gcN)sgb*<?
zF@ZJDt68%-$w^LNs$AM4FE3Bz<>e_4KKS4>CX?x60Ll-*+%w>(D`CeQ^|Q9?pQQfd
z4xtbg=znam9s0%?f~<3R{e6X%^$L6Uj`#5Tv`oyyDh}_{CV)Xs`X^;^Mqi)Ru#}QK
zpC&W0A<!!tdPg_3s8$ndH)1(`PUd%P`a;<CE-e2W{4fxTcU14O1_uXEpEGAp{Qmv>
zYs6$*v`SX9d)A!%lv#8CVOI5C5*`Eopk>?JR^yWJlMwP2v!eIhKWpA&X|v{RZ>|e_
z-ODTcY<zvy%epY{R*)Y`<PeVZ9xf3;A0+Qt6>a^-S5{m^8-5Ru2kZ987u;T)O^<9o
z_NZ<ng=pd8wtjGB8U3|YRsB=6MXV;gL;NWbv8g+PJ9JG<&hS2+lI0;b!px)BM)K39
z3VL_8(W{dUt*S+?UX{DBFVjO-q003mLabe)Em%xg;^ML_U}Sa+T|WA)=4^K7Y$2+u
zTGaVumz9^3pI<;$b`DZXW_<Pqy?gZ_EHsocqeqdFdVqNg7W2op9o+NKV=P?!6I*`&
zgL6lmjizbnx{j0*i>YDnh{jl;yQmP_^SMTWb^xqalS2_O{^BsDuuLO3z{JKqPP%ol
zvM<X^QJKcTu6E2Ofs}A&?*LBkZLBrd?{O{;P&~gy#wL8KTLNCIKIcgxd6G7b5Tc$-
zd%=PQ-K|#ZBLJx7u;dcR-_Z1lO8mGkx<nY3M_=MXWKA%9ba4GKQW_c1b>|qY!=3r?
zY$jY(3hfWEvJKbZ+;Vs(3m(peC_iJ1ow8m#LyM`t1=ZJ@3Sc!s*Qkc^yT#3L*_7;t
zMHhmz#seJMrAwFRs$JaBwn_mtM)KDS@=|BbdpR|$VzAJ~U|=e+0w_7cQ|cfN0<)!L
zf~tycsk7!?xc~F{pC!*+&`A1yeU>LJw{LlU<rSM=U%l93RyGUC-!XEvS_}bwg`&*g
z^xDc-PFg3M&)xssoSI$CjFvp7*&1++>PK!tN|~FNm$&s;T<=(+>x$*?*G6_9*2DI@
z!=_C9eZO&kt90qaM_Y&pHX(!&8I+viA=sgCFi*whku3kcf+0Qa{IsbYi&^0D1lI&b
z7UZO^Nk^NpLO*oXErF4E;#H+_PposlEhs4D+N;J8-=#ApC8Y!f1=6=qFG|bGP!xr!
zPd-k1MkZH|8O>`i{+EobES{Y<g)@ephR5qg)3lmpzupXsWQNpOp?mfg*~H6rx7*CZ
zoRP1x<A9sQI4fKBxk%3R5Ep4CyHI7`+A{ja8=kGWNDE;>3MCa9xy33O`D%|Y9fDFP
z-4J=x(4Mspkdw(aM~T##A{2s;_dHq*g*Md&uR6gdFE3BHTrLqC8+#96D7&pMK=IB)
zpP-teN+Urfw9faaB8>7sJ96q9?gFg!<ygBQBb&RLoKk2X)Uaz#oEOFx89!wv$TdTS
zVys#NbZDnSSTmNIS*7(CfRO?9ujQ~9kM%lpj*m24_jRfC=;W3+Z!uY!2^*h;;YRfc
zlgTt@&YU@2uD|~J)GDha)22<MEoO52B9|>wf#0is{+TfTsxaN-O;jZPq$JU3mntJc
z1VeT2W$6M+B_bO^DpKw?>HMKPDh?b;%fiiXu8xzMep^VG`F35BcntEH%t#)xDcXR|
zZ>+rGln2hvYwZjE8CbX1{GaGBb-dkj_=WE4meK;a^wKwM(O2Kr>a9FAXA$1wG=N{W
z)oGRYXTanZyGzP7&z+Zs+cjODJ7-N907c~*B^5?d@XvlKX6$*P{NFbPoY6ag>_U|l
zTPrwsP#{Y;S1`1fgLONd=(<+#{ZXZ9@;u#j*+5~53U8zUxDM!wVKMZ*6UKMx%(SPT
zsChhP@}od?%%j4C_uW;quU%9mJ$rPkJ+G#tX<Ch#lqSg4bOMDS&`)-+4$RUkrB}%N
zWlBoD`;W~6o!eXZd2<EVjR@i8Zwj!Q1z~|E7OgMmj!|K(+E&Twy&MEN6gKQ}(Lce?
zuiGnm=B9QD(cxx1s^q=7p5Lp_c~VFoB`;hezc}7=_$yP_8tHbsh1qN#W$cl#=fnDg
z!yFJ9AgS~U3Mv}51l(T$y<?0b<<$gOAua?m>KLVb?+}rJ#(u5J@!`=RzZ}{hVto;;
z89#?hu~Smvsl9Uupq<ZkWX-?ExOPDe<2k*WWN*G`@LbG_@q7ZUMoFRkI$?YJzJN1c
z2Xi2RDK<9t5|_*MUX@j{%?rrJNSXe9F|e`fr`4D}d$uVtJMJtbcOm7KAR39SXjRUY
zy7+11t1B-#_2IL#j>Gla6!$*++riB<27lZs>gelA2Aw&ypULWI^){aFoUI@=bY1-1
z+SfYP6c+9)x#avlOU~#MaG5~h@k(?F{Sxdf*=XF$N-K2i7KOVm3+L<QrF3j(A-hQB
z)E*9A{wANx&Im?Pz+x$@cY`j=-kTc{7ys}qy>jNgrl5NAZKL$CNO2PqqR_XagMv~O
zkj=U)+HAHaJKLDdCR9yBRaJBeXPy<vf^-jQnI5jFv@=N5_niH#yGWJVjFPOB#lGiK
z6r)RbNT9;518(9X%_L`d_;FJ?H;xQpKvz54Qr)PU@!Fl$*H|m3=c}C3&G;VKu*c~w
zEq2$GK|JZ&K1rL!H!kn#r=JcILUaZUH*95{??JO7NsNMsKtaba=pSdqH=62v1XLZ8
za~lSr#D>&&Usw7p;?0U7*6Na3dgzSmBSA`8H<<@8y6@P1)`KpujxLSY#%h92VfFXx
zx+FQTezJ}tjJVJNRj*5Apkc+Vdv47MCAH<!6-DXecDqHLx{@vOH2-Ssczb!!=GRtF
z=$aYx8@gPMl;eQoAtI|8AcXjS^$V;2Jf4<TWK%Lnoxu%Y3yV4LSk#WIYKtgU)%~X^
zMDEMlU9LB_?su^-%fsfqE(%Ij1P~gipi8h?1iRDSv<o$JR^I>?tuLoroR#cC6}P9h
zv5mvw&;UAh>f{KEiN7N(rt2LJ;r)JxRKN3eaWB)|jK40N$8K%MrKbngL_X=wxF=d|
zR`e!%>IxySn9YXZlyJ^~0LGpd${GCv$d&cKuQH^@eBFIOTJ^t&$HiYA9+U7;`}PSj
zHBU^U0ywM+>-V@A*ws#9na069m9zQ=pzE-8zl)GS6D1WI9ixozzKnb?^VgN*@<^6!
zsQ4y7|6on2<C7N2leBrN-P)ZycXm+oH|_HQmOgiAMHumo5dn}|Xmqjxqrms>TqAL%
z+7&DW#77vVbO3V8>$|`KVuK;Cyw)n|)gYtTPjbsqkX<HmdTW)hs1kp^<d)Sw`A$A!
zS(mUUPp86DTNEwW1__Z3yWtqruZ&vfWmOdAc#562&+OTzO&J|-MwpX8b`~dA0ov0Y
zcHT)!O1fq0)Tyo5Ilk>P;n_`Tn`aE3e)v|&?)4rimSa#qQtG&}w<4uHxMIbMy{&$|
z<3Mr7{w3XebzA+!4N=1qVytZ4@8<kd%ye#NA*V>CeVCbVSC;aR8>1MMXeT1XM6XUZ
zb{{mvC4fgoOHGZ`0CcnEt9vd9yIcaF{aA8^=5>w#UKc~V33JB|RkbC@C=Ql<Q9`8&
z;!m|<w<(mCm9cI64yJ$lIo-P@q9_WA5Lj(Cq^{#~yNQg9prD`-o6W|mU)Q3lsxhwW
zM+iY-Q4w?JFJfOxDyI)Q6-9v$<`rSvs}m<ooLQ$Zxf{vHT7~98-lBNLJ=Y8m{dbpW
z3$uPMd89*!SX4(o001BWNkl<Z4t+8*GAf-O+yJU3sdVe~>|{gN;ljbj?fl10<y5*Q
zL3RbNDhaeJlvV0v6{_5JaTqgxELJn~J^!lvnx7PsCuy_PEkF*iZr!@F5hF%;gb-F7
zG2n=<O%Ty^NH3Oj4zK@hZ81Ujc7`}SSY!w-r?*N>wyzaoK{g|cJHNsZo%t2@v%1YD
z!v%h@j>Ryy9Kr)6A@-wJ2u`mgv!woc+6Ogk-&p3-$*YQc47JzK;MS#NZyuDnYpbNk
zg+gqoe?v%iiHb|Dy+@er5L(-AUf1<PRaHfGJ-no(B++K#B<~s%Kv>W)6wXB#13`8M
z90;K!$VJNCpsYn`Kl0x%Qd|1+>)u%2(W=LOEEIWx^j+kfFvzi^!_Hp<e0scRa-y>-
zndjSuaKzniE`9I~Tg;%_&ur;=`J$F(p;qJo{M05sr$<0!$L<|<O-p|D`+{Ku66`Eo
zUruC*iJqNoY)f&G5Mv=Z!_DH26<j(rh<@=l!h%hBygHrQn|W_R1?5_}sa>b64}xR5
zJwCL%?UErq4a322JDry<Tiv<X6q9<>$6Do2lgvS8JAxj<OpmJWHy@=s>6B>Y)i>T{
z?TQ~+y=EO-w{6Gea--`yLI{KqC_->BBa;=Yer41Z*D>zK>#$j^M#&%vs-|I56vD&9
zxckoA@p?VnanD2Cc$UDOuXCHJVReqLG$+~U|0>E#`K5c$#OFs1H9Tc!_6g|n!wT=9
zu$Y8S;b9>cy}QsufLUWef{h@%LXbltA=XM>iORNAH(g^blvV141S)J#b+dGH1)bVi
z_+fP!8}>L4YPuKz)FdFB6p|<TQ{Z;H#hY)w>6tWX(taUCFA&iERwLo2nw&DD3rv@Y
z+SUWrv=1`k0{7)t^{w{6?mXz(u2xKT41*%4AxcvVjabRLF74=GLtr}V6b?@*fIcx2
zi|+*w16`Nw%csU&4Tb>rwMG^@b@mr1h7D2FFHIz+WM7^}iOcxRn@x}yX%u4jMM}xR
zVx63_`uB(UYrs<bd|fVA&5fkZ#7R!Nr9CF?A;N>l12IUd41smt9!Yl)!$97SR9rkU
zZW&U%OhET#w19PP)9b5FwMg^#5;`^7ko`nJA`~fK-u%YOg(s754$XvTH>LePWAG;(
zqMG0BMpd0Pe{_NEP9Lh6?IDL6-x9cTeg~<e>*7z*7SSKN>U}Og);jjVD<cRAFmd7F
zAa<v_nY*TptRj^OmxM8}tDUqQ51I~9AtquY%<RqZ&~xZG?wW8LQBje~`i+}!9eve}
zv3oP6X-~SxxP@kUR!g}|1%3&O?fT@WTICCOn(C*!=vC)juyzHbMqffq`v^jV9GFZB
zss^eCQYe5REu)Z;0|Q9O$>XP=mJsAHVYQl3guw0bqH7wOt}*uqaM%=L0}JUG6+(GM
z!`sj=x|_Aqv#=yP^;v*|Qf<ZR?Ukp;M_bsi$BAy@_WLK^G2!VaCy}3Dz}j`|dFPFP
zai&LSP**#OfM2#&GHOT=sW~2IuP7lsUnM-q#7!535)o`dN{9$HCCpu0_CjRG?yC#Z
zlhfg3jO0n$IMs<Z0E-td?x`qBA^@rra(-=iKtTm;&uZA48Vn)XC)#lJ0#vwRPhM^I
zv!%-F2!@DEErguXS|w2J0<U{l@o0kNJZcJ;9WGMBz5;N1>kH8?5%pcx1r-{}c}k6I
zy{@%VO3A(gNpY2k46qo_)1Qc}^l0qJk>r#$><3n`8Qir6-2(#yFICmUY`u8ANJ>)p
z$4&R5s5^ltfmoUskY)!)A><OAd%okbab1pP8e3lbb%=l;flf^hJ`oTiO4r5Y<2{YG
zz=mX=bE~5Hv6At%zuF@^p4aj-YDiVtAL;eFmLH4j9S@4KQ*Sz$<LUeIHw8<w3sph_
z6^8V5aQ~GN+<r+IU;k3d_B1y`dOAqU@o>T5ApS^k(`(p7rc8c}>u<iDoZLJH^zVxh
zoPT<6hw}Va`O9ATx?pON+kI8_bxN{RUfQC2XU+5JdQ)Di5nxxi={y^k4D`?=B8zab
zk7&zYVyt_KRQ3=qcVTyJ=V01it{xsjzb*=qru`%Yr_#x>mkzdMx&)@sIp_eLgEEM<
zrxUImAU?*-!r#iV`sC*9k}BWo&OKQ5aZ(yVlSSE4Q>HH}y8oF^@_rrC(*aW4e8uIX
z&^3)m9)FUUm}p*o<5OlV6I^qS@x7)9*pu#I{caaMI@);Vrgog&KahxE6JAyF@uH#w
zbAK(lyeKO*zFI`WNg;WXKM8yH?iJm;cYlED(mz|0AZL~T!S1R!$drPHi8W>u^lWb=
zn*fx%jaG!!GO&Gc{aC_-C8A!LogiDoI*t_{L2{nftjO!SNML_~q}W+s@U#npaGw~>
zgT=LD%u4+^q!sBDRaUvX9Y*}8Z#<-?OAZui?8s55^!UFo6z_nwkJMhqY`*-*AAjst
zw~WaN=`EW)A(+ZN-y-Fk2yvuAs|s)hX0evZV=p|C6Kr^ORaYJTe^L_O^ym`@0e5YD
zd3j?Em`+l~gl9LU<rm9Onmaze&hLuZ(dzXhrLN-6`2!)u)|D$)W;cCA+a*6Ub?5eN
zdq=%BxA2|pLY4dyjZYRAV>1gLy*85eVP+~^I$l-s<?>R_x%}@u`p|t?EM@`&1KG0m
z58nUq6F?U2&+;OKTw0Ws^3T%r^qQ5llu`y;)H@dIuC;6R+62r|s-4aMD`)BKQZ{^F
z&b9>=9Q?^i%6umU3tg1`?51+L8}k~KA#MvF&o3e^*@Jhzm&`>@vKKfxu+T~RVi$Qo
zx+q@i!uhiY_fijHg_lm7Bopd{r)NrykF<(Hk0Qntr==Cu&ToUaFe~*fp~wX}MJgb-
zj-PnHzHa?SI>g43*foKEeS7ovyPq+0g_)u<jUFAWY)^A>&*kBq*3-evrNwMYcB1K$
z>5EHt{ql$N^up}4dA?y0Co?Ti(q>6YN>a9O-!As=->+D$R`crBs}BkxBGKINMGScA
z#4B1%#*fQezrLbbfevAYIP<kn9VmhvqYfEyp|!5$YH{h+4B^--y8f{e*q3Lx!Rrzp
zpwlg?`HQ!@*RLkXX2fp#HX+%UFDa}Pm=z;&Cag+m=9cTE6&lx!41|P;hQ}6EXdEmS
zc(wY8JPwQDhpAg8Nio9>=|=o1;BvX1>fgWrOQE5mde^R9^{>DFTIWP^aVL!rqtdeg
zK}*J;IT7+^{yXd2<{q$P_71Bnv+x(-j24{7F+~1-@Iz<6a=d48qVs;zX`M$8lDlnY
z*=VffqAhl2EDxljq9WkeUw?HDpYTF(k6{nwDrV~u#aA{LPvKs&7d(KJ@|7Qd{PFLt
zd94!#A%ywVQ+lla$KRrcR=RcGo?FDFLxSiVZ==$sGkawPH{AXVV{f>&=9uE*5@vq+
zH4)+A^zGA&J<0pn`o}iDnEl<)dAY@Rm*noN*B&i6rdxYE`l>t40p0tF!*5^jLSjZ7
zQJi6+)cHTH+*vH-#f6#sSNhH?G|2wi)mMzZZp7KcOzongC@U}LV0s2uT|S!V=-R#p
zsc8q9IPq4l8B#$+h?z>4&RcT|xq5gAJvv%hw7y)My}azyqMZG|H<$z`Es`f`qg0pv
zsm|D*K7D#;Aw(o#SkCH-NUI6@#~N)7Pw8kB84t7?`&1p$3Snn9RQQzj=^Spjs!fU^
z=&I#WkPQaL!N3kiOF~}(Dlodp1pCw|E~-?>DQjNw$fC-IBGO?o5`+A8vP+B-4C-La
zDXfmQGN(+S2qV6;Yh?XnU0$8txf=TmOf@3XZZX{Er*tr0uK{sJEzN3l74p~DJT#ll
zr=GCx?WF6iI6Ysru!sb3AjJ%x7}xCL(Ulnm57TBvBnTmHJl=C?3rw84;Xq!gJggqv
z9ea&Av$aJe)V4*`H2#!jfBYe(RQG4OlU8l3B+#yK@8#iqzq*vtN}XBDE1B@%^Yug|
zKv-BP_uqXdeR}oeU;llHO`Csb=k8?gzGM7(XAd9pTWD<Jkm|7!5s9H@wD(jYCw`<?
zWF$AEAGa^|BNUOP=)9>_c#F{QmAd?UNQZ=8b?4PBzQgior#^XOVPUap-h#z^_{nED
z>~`+HV|+ak2@v0<6Q6uCi|;pukz1_t?W$6)JS&(U9jz3WY5csU;+w`qB($Amo3X&?
zW{#DB3n&0mfDOPV{P)hi3b@?Q_H2wes;0OXIL*)Y_keXk0Pv=t{aL`2I*)GvE^gvl
zI^f1=*OrNqG4&fzcBq36aqbU*HBB4~5DvU}i2WMy5FC9SwZp+TV>>i7w5CoLUY|bH
zR&#AUp&|^+Tx6hOF-$KutdtcVLrg}87?#KCPBHC*jLc`RUSBW@VF=X5P=q1+cIVa?
zsOiND;SL?EZ=0DqXgZ|%G#Ij(4VU{N4#VPSK0VqAOiKOZB#=|;C8NZQE{z>llVO>S
z3~cy3?dJNmGreVXc8ln`mrF`Y5>uy6JuxhjF6%QOFKVT;2O^otq+3qo#ZPk&HE`pr
zD`QQHc%m&kf20(rZF_B9biapA&uMz}Uz;5Vc)ooww^=PTtKFm}AncgMM*@{ns;iI1
z^-dHDvr-qtc1f75>GHV|0|Lc;V<MQf+{NQhzs?2cj_`X-dPXMy`tOU(`0#Clf`TY0
zDCBRq-pNx>JQ|krz@u*gLjka;{RVdrxN>WXtEWrlns;Tz|L!pbMt2jfE3L|qysnmO
zIYLfQq_}lxAN%=hce=g@b^+)%{jNtJy07<)nP2ebrymg+8G+mF;fbfG;qiF5<f022
zoHH&qhR;5o#rO%gaZW#-ZgEypvORpVr1+PTLi3%6nonC1l7leB<dKpNyaBw1A;zjH
z2FCjtuUCyd)J;<Xo(5ukwrkXiP-`RS-$i{PIi!(sP2mRS1OLVl%l>HW+&|MiAvq41
z(8QSKz%lLcBZOeuv}rYubzQGbl@9jFhp6xv?&u&N3t_ke!t9V!4w+S*X|hX=M4awX
z606yeR3@c<VoP%|DbTGQ?8vT-#nh!Bvs4m)h)})U^7^rx)uPm&P;3g{SV&cq?9G>y
zxh*y4jtMb}uKTuh1f#B|Zy)lYgqmNaG?w7iC4f#C9v?Rdg~wa<8iG!k)c?}d=R=5l
z+ge;?L!{!7?qR@pO%DBwGj__Rv_EDJ{wOw_`<pY~WU)2;R&QaX=E0r29Vz9eWy_Wo
zH+{sPi>$Pi7s6r^(pLT9d^IE_DCR#eyvJE*4D~y<veL=Jlm5XgFaC$1pdfx)_6zTQ
zIGqW%-NflbPoqbV?gPSNx^ykcIglC>8z0=cy@i2^_TJN%6y4$0l?&g|Dt^4n6d2v3
zmBeHTI&Y}uS{Cwl(A$kZBjo((a1-lxIs=6eA~>pRcf~AZaBwiAM_tUl4?o86v(Dgw
zdnWSYfBwm>cic<&ZiyszO=xiLc2SXhHe)8^C*010b=zHkq<G#gw8<$lFZqy$qQ8o3
zyCd*C@GJF&-%7xrH9-NIvSge}LlFtE8pEP_EU*DrV^}7qG%@x_xuY3l4!K%sEw*pp
zUL#&MY}k-WwevZ!6GWa`J5Wj^`?_wsziNblm|(;8U7cW4>4BX&qCq0gk>IbhOhJV>
zWGjO_AB$>-(E6Qqjuc%=cI85u+xXmvIiO!`eG6<|ik*^OIkiRB!FX+>YF}Gj*Y}-}
zu5Cp;OLgK^t0Knn#7+Gh+q*nzxm|(VT7MqL0%gGQo<Un-%Rc2rk4ojC#&|p)G3K#(
zQ6@`3zt*0QJ8vgab0ekrv9+&tqEeil@>O<;^7AWiOm9>~>bm6NN&n!{hwi6yr;e;y
zx1L#F&Su63@9@Li`KA3%J!|>0l`9_#3o^ekXi)dfCMmn+mud{}7r^koffJ;nJ?KKN
zep{<Z-zvR+uW>ZwOPyD=a!(f8LV>)jR|k92n6rW@F4xhe>@a9Rx8GtSmCu6B-Z#hJ
zdEc8~eErS-|9k%5D2l?rpMM#%*~}|1{)fp=B~f1C=lc){4+~|+^y!pn(fbRtQYQKK
z>h{N9MMzdN7?_7)DFje~IgO0Hq=|D~*2wnbv0^s?&(b`1{NXSZ7}$()BY_?*eC@;H
zzWeSg(sjMcDl4G-ZS@Cex{;;5BO3}U8}`e#n2Z?0-qD8T(xX9zyJeh_k%9GF4eC-*
z=Jr3<={2&-t1&Bw?%91fIK2X=SKzQfk9LNY)7N4tffAP?IFs`lis{60pw#NKva+&e
zEi<_mWb%!@K{oaIZ7FQPeU0r54L1#wf)2-MF0D=1<2`}4!p&1R94IQ|<EF+1#U>6H
zCbO-naYsv)TN@$dpRT|5;V}4=0V7Ts*!X(rh5vhvbB3SAnWqn7@4o%K^7`An`O1r=
zrlqlH$<lAK^NMdMu`$DBB5U&XQ3Foz9q@X;cpIDdy10Hsh=>X?-*Lmp&?{6$yYWM<
zV)ts@)9AGhmjhDcWv#3tUzgX8IV-qV*Es8Se>*=UXj8J2zVSBtC)l6-$JlnKIcy^L
z;K75=qU@B1ckkW%_;)|d<-WV_z-qPf&ifzJzFibgJo*riJ@Is-RW?FHLKr>jqCO$*
zx+XSfZhti)35F%HZY!(Cgx^IN0<AIp2}%7B=Q|QAFx>COKqXCQ!~PThF`Gw<KWpTA
z8pcD4M#dgfq?ENTo78U_4iFe_?tS^NHB*vNQeU)zAtHN6L!TH3akQMJEzBoes?zKK
zz4Yzl9b)%yEj$tetj2Te8{4qOqgRt;l}ff{!k%0~nah8b#nf*>)><f4RXqXSWTwAy
z=+ABoBd=uAE&kg@N^wDh_MB}Mg?7h#3T=svY5H?s73ZPGSR4_p-xd+0zqXXJSXI>x
z$KrY?DyphAnho>A-1(H3S8&tct|z~sfJskI<;_=LARxd&?_NE5{@H1_-+t3|Ib%m&
zSW@Y+T$fp>5+7p`F%f2TT~bn^bMC-EE;=>PG-^o5Gg2{K_=Q&a>;H7;QJ1m!U3z&{
zE6*vG^8CR)>}QWUEx6mzo(>{IOca%Cv=1{A6K)o&2I+a;Q~IAeZ2gV5+);7s*c)e$
zz0pV}n)Jv6>`zVQhq((l?bK5^eaNZ2`R)geUbCd6RJWlwm4gAaofyf+_yQw7ueu+w
zz<5qwfop-zjU0C$@Oss8KVt|rAO9m!81aQ`{T$x`IK|KQ^}xZyc>+x^4)|Zy?*fPf
z&H)|<&h|6jgb}xSMGIUz81R>=`9{_GukqP_EASN1a!ar#m_2(oBS(&`d8{ak;jd6D
zAg4||xxk2347P*QW4O2rDg>D&kW~U9c1d(Fgw_#`fps#KTY`qMivGu`J|1uX{jZM<
zLBlxFa<@))nV{Gykh1EpEYN8G7-+3+7g=30sWchNQy|z_s0IfI4*^cBE*YXpZ$e8G
zjYS`X5!O-<DdkYVS@vxay?MzaCo#BX%9f0svrc`#U8sD>-+q_Nr6?A=|7?#IM0c0s
zE!>X~V&$@B%Z_PM<cY?{En61t+LL@&uO9XLoNw8>jVGV^7fa@ShtuWazK169+`p#M
zzMZjl5<)O>!mZWAt;u`$Ss#C5GJpKBg`OR4^o_Uk@uDKSb+D3Cq|&d8?V{v#_Xo)Z
z+UO-t&zo6NA9t%M0LNjjDJ8H-cQIdgZIy6EWQggG^G^wwv^Ud()gt(AWf{Z!2Oy>7
z_x&z3sqoPK_jGvTu}3;stqmi4p8MBS?zrb6Vq>BidCqVyzjiDG`t@bR*~99MOFM9o
zx$_q;%FRt_R+wH}w@9uA-p9y5ejg)S`Yqs5pcjo>2fG8^s(!CdSn<cGMz)W}zrCh4
z-P#h!ue!!aps|{6mtqLcqvb{nF<F<=s_TE|XMB5FM!+_kY15|FEcE{U?|%mfA&mZg
zo>Hh=hQVYqEP^Is2&wo8=pR?x-mt_a*pnyNl_MKiC|e56XH6;^Q5U|}NA0LP@-;Fa
zU6Pz9*_oqIRB4FFhydspXSn!7>_)OskgZ`qZ+C4*xyfXzm$iM|Gq(SJy#r|Vn6m~*
z={M*~isKvd(UJ&p0@Vj?BWz32Ui51C4MyvD%7hrjZ13CJb8zKuL+W0nlqX&o$}z_2
z)~8Oq_o0JX*|qv8CnuNj6YtxZlbL<i4L9AM_P`^L^T53m>Di-OgX6!SGmkav*3-RP
zBHw)VHItushItz#S%qGTD|Aw`y+ntbnX|T>znvR)aeJF-dT~y~_^nc%^rBj>X2|Bp
z-B%)cU#nog?)uuSRSdIQ#I?6v5ISZ4x^g<Tw@_53vGbst=rA+ub~%ZPO=R*vKjxVy
zAIG1F1TdLQy!pxty!guNjK2EDt(oZu```1xqrascsI`A(XXkM9gu61!$}5}Q?z626
z$wT1;ZU>4Q**~bx?~Mt-#w;8D+}aXoc7fufL<c4ToBixF11+)2_2;Udjp1%zh#_iy
zIW9>z02?-J@KWviw?~1cjs{jWBNMtVR+Awp`^MI~oy**UgRSJLE%%9uY|PT=yS==X
z47?d7)O5@Vu^TS$#HjkQoM4nls_SkSCJ3r$=y#jV<{E*Kl$3OQ&2MH)M`~lBj~$W*
zu|%n#EVqMb_5R7n3MnZ(-g9VcTsLJ)Mp>nN-`~FA-or);)1m5lwG7qqk*ccxWfN_x
zDbLxTvvu2!5ocd8dg=WSPtvD7`#htryk1K=aPZZ_+_ZJO_av{r=z{YYewP1|uZxx}
z<>{pVuxIZ+OeTdp$KS$Nvlo+RAC0P;sdVe8n#5)ntlQ<}fvX~~Oz2+xfk1wfjrPGy
zT3Oag-QDDhY@bwlPAl{LF1=p~!E~=F@S5=#g}tEZkecHq*r8Bdp%LHS!pBP`{my-a
zUw)j==|h^1Yzqtw;LTTFz-G1DHH*y5%*uWGyi3NE|KsU@GWB22alz;-H*Wo7``HB<
zdz<ZD{WA+mFlsK&Yh+*D0Qg1VH^c5!Xb>0V0Q}Yc`W@JJxDz}IGz=@OKL$0jy}1ma
zx=vPoKB@ZO1AOgg|M}4H3$-=y4HEwT`|rI{%2dEe2I=f$98D`UYDW6Dysj?tU7fvL
zZKZ5UyjsKF-XV>|*9Thb7Y(oU81ChkCL%x(7iz?U_H5tK<t;HvA?@>7FT{)hdu=gW
zuh+W=z;3sn7}hGY9>Fn<kE}w{poL&*euRnU_495MNMy$Go<mz?*8%a8s)+`%X10K+
z)~_EaC2pg?cJ}h+%eS}s_1XZ%*{Lb{1x4?D{q46I%N9=862kKt5O2FXET+qp$&Woe
z_WEnC^n2W|Yky<T{6)O`#{bcyoAEWbYj-kX;bA=Y>@?=iUBVe7Z{WklP72C2wxzg;
z3N<mjU%;(#k>;(pUmSLi8(#ZDcP@BOtDw|RX=@djtGjqjuShP?^c&6_6fkP!z`(hQ
zvDOcU_Hu}IyPP<zf}giku)iRLQP)1q4|9HE;_bJRo14evr>1enh>N)3(kuCF)>n-t
z)5Jta^XW(L^oR(I{3b5aH0G3euYJqvFSXA;`_P@9nX|6Mp=2J;Yx-vwlF)d4#k#@q
zl^9~u7ZVN30G;b|Y&FNq8d;ad>OgjBWP5Y*XN^pzehipD_qIaQH78q?cJ10p0JGUF
zJswZ(ihFQUeXAruRw-=Hg2GDKU`gn(7;O+kTcN>3R%v|!>#*q9EdD=xCZ*vVGD}*1
zPC+(9IQxqQEph7X$cBT(^|RX*3!L`fx+C>Hd-l}pl5yN!+s6yP?`HOi)GYF-Q84X1
z-g9V+jCp)hW`#?Cr@;=Bxw%a?ju^Ev4{vD(z*67-$bVHtLV{fpe0<aSp>cO#7Gdfa
zZ^tCW3zuAY!BdYt?7#G<S5=wx<TE^Q&z*EjtgXMeJ9#go#$3yeUAyVju>&vsCyB4W
zTh6@?JVCn-J(%@#32v`sSl@u)Gy6EOn#H)1?9|J);=SitwXAHd?y0#&q)L_N)lz+-
z?s`X2%LhwAuU?&OAs3z+XbyA;%Wu1!q~}Xc9eNHI-uNFjty#eX_fEv&aPa;|GnoGA
z=ZwAKI%a(E7MEUpA&*a)#y_6^XQQt}|Gs^gI%V>?xuxp%$PiQJ_)Ef+2d|8<#Dpt%
zhq=Xdhx3~L6@+9~Binr~1<iGDH}>H_9%X+TE0`X76z6mRH~HB<gZf{K>wrCe#@)s-
zsNYvN#IV{LmY$ye0fwa(g1f@VFTvzvX>+NtFJG`N6LQOBtHmHzQ|W2wR@ZdFzI;h)
zq2xf3<UkP|D3YWUNRsow>66i2SP8|4O#VE2bScR%mu%0L?8#FaWRD9I40|8i)f%c*
zRb6z&6<4I1&1M-K94!0x?R)ZzstyAlyQl(j=vsnp2~~Li(iy);LuQV6Nte{D2u+S~
z>>+Wj9`@kQ*$Pre*Yy)CA?`RLQ&L<~%Kp<A6-}J{e$J+>0<XR&-JL(;oMG0+Lf7B>
za5^O=rKF^$v3t*60K8r=&%N|2=L|oK9^Gr>DMLd;xciRrEL%LEm)@SkS(n^Tsbv5k
zEOMdivQKz)m%o9wp-jq)KGQ0Hd{rxFrdG+zTKS$Vsa}E9U-Fs)h6ur>E4Ehh{dz0i
z2VTs;i>9!0#jkw%*~eUQ*`-)47V`58c;cyPy!GCPO#1tS4D8>Ro;|wr_lNG|%TGV#
zqZu>#dHG7e&k5l2(HA47GBh&8^xLyD^9!ean)~0ZLTy-aR?6px^P0BxtnKFDzv^Z~
z=3^w@`1%|##t_^(hWPX+BzIK(9BUd|2po+&?HXv%g+l{A_1PwYnHVL9>Jp35cCuB4
zb@l4{`s=S3U|1wajvTpT@7}#1*=)9n0Cc((F8K+TkA(`Ka1WOXX+=g%WO#s)jVtMQ
z000+VNkl<ZsT^YS-^cn$@o2C&zhTWvk0vN=pf5xtzZ;|!K(FYQDNJ6C1j^l#!U`y;
z6pe~%dlU@+2s)3cyVaSTocy%eY?d~g?WBNg1Yv}`L^nC`ghv((i5t<FcjqKcaP|Y6
z^3uK={6?USrvU6R16%f9(+E}Zky6+7<*k0PHbhZYN-FTN=+|%N>=mW|K*(UH)Bj>U
zBQuNVUw(}b-+7%gPand|ufB=RX2WbY<954w?BV+{n`)zJo|^V=?!9XQ9pYl?(Jhf4
z-4c1|zPs?MDtq?qHSO57`}364G%~ZYDJm}3D=I3{b)5^Xmc*dIpgHZLB1OlJadhk2
zmA-v?(Xm5Zz4@f4XK)}rgHs0&A}T79e?2pW<ttY6#@p{x*P=S&>|xwF{uVy?cn0T<
z^jG8stQIQ*tnbV%`a?>6a&dOr{v&xk{|Z8~xhAJ#U@1n{b6whD)Moq*=#PK5abu#X
zwZC9HHL|_Dk?qaFU&74K_%CZT6G)<=@VS#?WTEZbx3A30%abmbORB0W-+c4UDGxvV
z@F0`PWSGvPhx7mKop*dw<^IRt&q>lGEv2l=6rqS9vMb_cPf-*F#j2oQC#WdH1u`r+
zKtz@ba(`UB0!0KBMN|-FiR_J_MW7TYvy|>Ow8?pXf1IW*Y0}b2OOfaOdgb+MPR@Dq
zO`7C<&UeyhFYJtj=vm;|AXb-<kZ#<}no)RDArOMCh0nDvilj`4J<nyCjsr$ppg+$*
z?TSWg9)EG^hImZMG;GXr`97No4eC7#t=<EB_56>>#Kgp*-MV#)4h#&GHk(blQw)R<
zKJ#))2a$~NoiE*Jixl3gv;@15w56SASQA0FfDu7JI!F@)LlvpgL5fBxqSB;;bm;-<
zy-Jl5gFpyXFo1vr5b2$OAXS>wP(m*u^j_|IzwgKWfA`<)&dff~o;lB%GkeZ?`SncH
z2|=lQ_u=d4Ku8kyy+#E|ni;{v$>K|l#L>$q8!{UBr|K+CfN`<rxsEX4aGz@yw8*5X
zF=h-pg!GAP%Cj@FWy1MH@|UV$uEV(yR)0@%9pkP9zYh#bWms5Oj-}gZ;e@kak($R>
z#=YNLnton}*}$R4qAhSfVgDETs7AuETPWe~iZtjNpFB9T{qw-H`FfX6|GQ}?aQnv|
zT+)zGF!1^i>?|elc`hI$nj<9Y9N!RKUhm}&Wj&Y?_iSl^4vx3%^p*|fn`ZlNsOf6s
z^(;-vb8*=Dmlw6u1S$CrQWi^2{^&X6Nt7>M^ArKSgj&QS4v>$vx-?mZSwKWE7)(q;
zLU+np%0Vh)`KQCkFq5F+`<9XQzFrQ>Zv-`((dWl)uQfmXR%J)Nk@kPZ{xKTkR(|~x
zbZfAHMQ=PZ796X6$Afn)E!#u1dHyOd&0IQ8bpaY6H;6BcZ|ptz9zHt|cMM0Gf9fb?
z$=z4dK%?k)3>xR1(|7neCB-8Yh>md)R2~Y-NO69ChqZp`xyJkV??0(AF0He~b$6@z
zb?%p&9?>()Zsy0cZ!U)EDOQ{7sdZeVeo`7detm{}J?aJFBz4w%89Zg<t<Hvu_aJb}
zPv*%)dw6jzA8p-?zS;=z0diEAl`dcDr!KrdS!uu<wOEQ<t5UbvUq{9@7Z?maUZ0i3
z-)Rg+^X6U4#k?3Hy-m@FajBX1Jv(jdJHH^`#!W*%L|aePLr4TdjL-!_)9xaT^aDAo
zYFpX^zq)Rg&&P>)i~~V0617E&CF55;U&1u^aoE!iw%B)r_;Z4_KPXBd23=6Jw7@oD
z!_)`UdDsqs8+f=y55FRwC$e4}VT-tOt}(}W;~h1|XuPNfI2+%a#hfke3f>bfZ-~I{
zsmRU~b3ji`vW_KY(O8dNA>O99$_V$rcNYuq!JjAE)5_NQ`M-)^Na2ul96jyNke*X`
z*OR;}l>s5xCNJ1K%*BK({M~(TLB*DNlV9;Rjz=Z<s~zeKZ@k+_-hSUg?x@yYcLw2&
zcJ@c@XcoTv3RG-%AORLrDr;IxLJt~4fv~2#$qhHQ#@qgmTh5U%l`VHmJ^a%~3N?Ba
zb*y_I#thnUhCllF2%5KO|FmlUSIDyub_uJ;PXJ5W1?AxQQ}4Cu#Tch7ozeeqv57pM
zRkOfm>Z%+lh0-cKP(W7qZ`x)ARdx3rNbSr!TD?4(uO?0}75E4(m$kDoDhp{dMNQ2K
z#R$^*Omo0X;wdQ}eC5uL!7$vND%ZZ|%MSYPzT86_gvVnR#&ShIQkK9!XJw7cS}%(@
z8CH4U%NSoC^Zo_M!etaZ%XTUUs3RTCCxedK<6|ZJu^+@}GykyBQ3+nn^E7HcO{7<}
z%f<PS>WRUoJ^>L4iN|;lt>CE@`{d+gJhOP>v`I~#8AdLx-_}^d32caw3v3DGp`|`3
zu!FkPn)(fVteItdu}OtF+1qrDU32mU9bYs<AG>#;l9L8kL~MS2pKR&sV8?KK1*^$<
z*i0_ho*01ltgw&dowKu>^_geoT!q|Hwa%&#Am%qgfsxfA`_}yg^KO9b_htR)o(00c
z`8-f<?2d^crDbEwWW}0QJ0q-rC-IcI%r^Qt;HIkh9Me-*3CrJ-^v{MCpcl4o$utrq
z29qP29pQVqz2Vjj<j*RfRBHbTj9-8WhJ1Bv&~*vm2;Xe29&cOEcX_7X?Owl1#-OzR
z@C)bYHu2%v(?e90^pE>SRMFN007w36pc}^ZS%@*trE>D%(<d4Q2?>dDR->qg)gc8E
zkx1@X!Cop?>PTu*%~{#Y8s*kjoyyf8TwJHBV>#xs#j_QBL4#L0Q8t+qv^>a4BL9SC
zn470UD{<j|INp)GQ@Nu{wUAIe3;`NLyO<m;?p^nty;t|}G5`JAr25g;A&i3Mz|>{i
z%3b~TAa!u0{({itaaT?<dHu;;z4f?DyUE=mitU-6Py+HM|09dVl~RhJp|$fS>n>sc
z!>y;K>5?)#<Deem6zkzpXxibz(o7Y*e>rtr^I233i9Z*O1SE5OuvvXsJ^r3#Cd_&+
z7m_h8>n)$i^Af(Rv1C7xnXp<V{YxhGa%Ut4DPF;0w%rY*OCmmv3I`7py=rSXF6Bl~
zjdgGK0`X_<_EJ}#1J<(>2_ct)_|7y9kGgNMFuL+BF_aNOje+|vw$uOl^N3q(fR*Cq
z<8vZ_Ojlo40k~S+>Oj_XA+Vy$07*(8KDXDZYI1A1<dSKiO$?$g)PIV_f4Qcl6ztPF
zXSY9P)V+9|uF~57;mhB*<xYZZMieM=W%lI&SxWF-aDwprO9d5kHS^ry+D|DdRrbf*
zm^a&XtJ!Ek8Ed9-^}rSi1OoXR8m4VC!rryBuB-eYZPQX0@=_JF+qe$&{vnM(VBPnK
zh{;b;w^Dzy4F@l$NMUn24lET?0kH6?a>>}9(F-Mb^42+k3Rrx3=64m#o%3<h|MGCW
z#;<&iUfeA3YLd^4T}(~Fbt)~G`mq0SPc}>AR*P78(dS?EI?eJNug=rlmxTu`=bAm3
zn($4mZ^w;lg1PADVu|Ka>==Z&&kd^Rn?NWk{AH>bx2o<HOHgRV*XaEzR@c4pTy$NK
z!1+-#+zsD6_$9Z@ZAXhjlgbwcEzm&fD1TL~zD84v*}2>~KE!niJTtdxH-P~;KMbxd
z(tdBi7YI8#O%5Sgo&^s#`5vWcLO37Db&(EgeP<#0PUVG+ZBowQa%nd&tkGyLQu{;r
zrPPgETJ2yl63s_svda1zg>SYdE4wC<FFuz&)csa&bGR|~wph1FtktSw1{BeKe&NLU
zOjoa$<#uuk>F)G>OTNrG^4~+B{E2=9t`pw~8_RVkM^_rGUJ))>@-Ih5L`3BB^YfE@
z2oFb*lai9sf{D^!@rutU>DbD#C5{3oARa|UMZ1)&Qh|fcQa;4--YWN2o|e3L?@5+<
z!ZLshjhER;xw0XPa`(EAyrJuOJH_NKF2Ffk;YeZj6~CsfjZNQlt=nSZ<@vEvzC-#?
z<I;c9fBN)keZTEzv*$J}+OXCRViT4jSMiwP2kvelQx2x;XO&NJS5vY<m(Db^8KWuM
zM44^eW?(-qk;m!32tgf=1kqoYewmvHriQ#aFA`|mUF;GMD||RgagLpSrxYx+S7iy*
zfxi-CK_J*3vv*C?^A0sR+1-_NCV0|y1evH17a0+fm5oeGGYmFuPWnBfXuGvWE8#Fh
z8anL)d13J-3($JEfK7d5exb)R2v_%s>cI{6&0`wqbNSYDS%yV|++4l2iDn^6m*T<E
z7mF;4Vo67D1H>T1zV$x1_yPxtkawhS<-9F`A2tUY=kMDKN~cx%az17Wy?eAG&PWf4
zoi2KaYu6lJj;9lDde|sonC2lM>=ou>Rvi+Cm+!mKZC@K-pTwMLP>48Ya!4`2FHV!(
zudgmSd3kR|qe1_QZOE`0J31CSPySdIsYjKiD}@ACNkOKb1KCK*OA$>XROBSO5$Qh6
zH5Zz&tXqk`-5SIOskFDNiCFy-Qoj)PE&+`BHgq&Kwf#szzxV4Acg;TD&-}-(aDlQD
z<#Df}YRGi6wzf8q_YfX6e}$SV0i#zFSeX~L`$j%x5u%($jap3yXyQ`$;4@aw)e)&#
zSxxdF&_W!OnXly`z{AZvB}Pt@|MP?QqbqweV?4Q-KP=g)&kEF01fctsamW|r%y7{;
zp?JJ8LRCFeX9F5dy{`40jOpV`>;7~}jYgk#t>mXk8Xe#7B(O>cInK|A<@8;7j`n^a
z(ZdZjk-y<fSUp&8O#aq^v?08St2_4Do<&nikJ|Q}<K0)Af?lW(unJp>A!@%m+3A;F
zrXZJW&Su~S;k1|IPZbfD{0{5u7v)wwA<sFw8A<><YP<In>bHJshmdoSHy1YQSy~sD
zx#JOWMXjKQVRhWUn1_o9URs`hcNT8Uau*ZMMb2^PRpGK;QW}^SIYKnKqmK97F80~w
z6OISl^YuNlx&pk4S*&OM02*qEPp?o-cX_!eV6$|o$;gHPD6wf+SeS5JXP${5<7H7{
zwwzH*@a6hgv7S>6Pz`A3=g*%N8igL@THW$s%a~vDqTfa-z$Ps0-Fwn2x`s@5AUZt}
z|N6K#9qMNQjsm38P_X{DO_IxFLPA1!SMLUeaYbNq*QuZU4Q5ULFo2+U&tqkYvC;k!
z$;!mYOnIS>2Rb|S6Q?o~wYdve*rCZvJB}=XoSRVPIObmrpp2Cun0m=x$Ex}*?}lF6
z%JqK=`#2fL2(#`=O!?I-@PU)gqJ6a_ED0r{2;KHi>F(wf;mx4NB_m&yD$S}^qzVxU
z-}ulT?WF~RL~@eUxti-^*VCX!E^~%>z(-;ETi4ap)ogU8AnCN^RYh8E*xfUrGM<nV
z0jS<k9s)Ks+AxmkOE|;kNd_FP4gBL<wph|*8a#V2D6#}#v-3ON;d6}d7i#+1ob>aF
zEKCCpeL%8oPCuWdPyz|Q1<EiNr0g5FLJ{oM5qX61znYVtyZj>x+TQt-T${uMBgp%n
zO(O5h;oIMP#1zJ&%r;H`b+VSvI@BkS?G3dxw}qu0c7%Dhx_zb-#C*xG5PQpIz~}nr
zd<yvQuJY}u&<v{Y`xVf#Dh~A8O@~`nb-qV3veO%G>+IuvVVX_OO6ipvMTcML!ArrY
zt^`&-AaG9vySNCDS&QGxuwmOH$=laGZx@d6d%k6|@V)8vmI>gkKUD+hmBt3&4J)d~
z94Vy{uU<#e=si#Y0r*hK$BzSo)YW@x#BKWw&wlgeATwU^7CEGQ1{~jZ!s4o_<Nq|G
zHrs)Odj<i0iDL7Xt7qN~O@##xln-VjEviS><+nWLd}P@YA$*$IGco{?Z4EePtd)5E
z*=6usVFIuh@b?E{L@JpFgWfS;d5Us+D8$&p0e|<oJyZ<jsO^?Gb-EBm3POm$%xvGJ
zlKr5iaFVTktKCv^yHhd&a8AL+S$KNm{Na7T;n<?teZqnJo7(@oyPI25l7XyI%t}JC
z*kJY2r;9E`I4rfs#|Ax>^X)TzyT)Mmd;xEbn9U!L#;ZLG0w77l*YCvMHb{avbmD7Q
zZE2A7${<3y29>2IpVJMjg4+E=7d8IVl?INJz+JdqHEw3#{AIO=ZQw<P4wTbW2>gTp
zPiybd+tysp#8RrC`#*v>v?jJD&L1!r_Yd-aL?u=e%<{Gty6-yUZChU}e<%L+`<(Do
z{E&cHY94WVSyhcyOyeGN5QhT4%axtz>^}DmeXGV5olc<#|HMepNTx2@kzV+Xh^tXB
z|Bkrbz;`-!b~6zfnOg5BPo8*HPMW?TrA0cVcTQkZK>6};1j0{VQBl!<^qV%xOs=IX
z;jqv^;AB>M`a^=oShJe>EJ-K*<-esS(p0x^7u!`B76a8KY2vM@`13v2>a&xwvhLyE
z%V=aO2A!$t>W;#Is$sING;yDp^hI9DhYuZG7Snw?Wuet(6)Xw0O739_>mbJJ@v6Mu
zU){;#aEbY(6AXA-$ak^auY341s3F(56shOAwdvoqs=O5}#gyEwG4-@SS$unU)$xIg
zl0(P@PHK2~_|Gd|LY%j4h#d57z|r3J#E|ZHdrzW(t)kVmouiC8y?LcF<7cDoB8W#2
zy5$~XP`=Fqwv3Fjv>mb_<YxMQr+33k(EK#cy1#x--^Qk3P_tNf`#f`H!yuwWS?%GS
z;KkQkgiA~*^BDRyGp>*;PG(N=>!86#)f$<b@0q^&eynVF7VeAN2o4_1R3DUVT(+vO
zse|;2u;17V;JAS@zJbVrb2-vk7=<4nz3t?=2{N++Crjhcc`B{XTuX@s@0SUn8PGuN
z#HJ)D$3d|c-6jBmg54zF<T~Q~(`}o3*=_LNPH`)c3~7i?zR}1p3CD?0IcZbqm?MvX
z0BgzE&YP^4YG2fsdqTB|gXiZ#?HhmFTq<Fy*#o&~)d^tZzco{(x;@i$zJ@=)GF=>A
zc7Kz)UKkPaf!|;!e<d#hiA`@4U!o*d=3&l$A{xQNoHyL0a@C^X7AJl@r(dGCJ=bFH
zy)pV5F7s?@>5)2res|p>ODaiBZ0yiX^)gfAtn<gXxUc%9#%2)0_t9hf0V?~`tJM{K
zjKe4B+ORO2f~Eu*)uOqf(GS)GI|NX-XIa$Y_-7S^XP@;Db4-UFbpf67`#a3=q_xaB
z{vRU@X{1Q43lCbbeVbmp%jheA@-S!T3S-w|y|2buj(R13zthm^$Ub=R$FXuU;>8HS
zwXl(66B81`z|t~3w?}`A^J!$s(l2_}=fT{|eq(2{X_$O|2ww3DfQo8)`PFsXJ3+R`
zZZ2-?33)`uZXJjkdCeL~Gm)#Ml;6~He|2E=rANpe75OQmCAIj`l$T^en@=Rb8l5Yd
zs!P~jcEu^hpl3H<_rPhemY-5Yj?c`x!7)OFvL-c?EmRydW=G_qcRq&PS0S&Lvvrgu
zcC#g`(BNOz^LbkM?>qCg*ng0>B=f`4?Gf2p`bMEJg#G6DZ0WX)ITUsxc@EWUIjz^f
zqw19FMy=z0Bc1abm3>r1sspJT0l!pkEJwvJ2&en&*z4M|th#DbKX{#Z{ThY8OicZg
zm=jivB=;I2?BN1)caYo}V<m;*Sl}A20zT*-33e0u8i52agxSTg2s64xF0fQLXA+i*
zwwbVHggR-;Fm|?vF*<S^=jLfFh4jQnyWnc8Y{8e*0VEk8L?Wn3MUeR^;9RW>h2&4V
z4}@>j)b>G>=|nPKzDD{@dFJDQN{_cYtD@<WZbt&WfJTefXULuZHiC13H~;V5$aax`
z<Nx#EB=Y}u2kZQ&2Y7)=9Z<Rd47~u8z4-5hk>(FTMgL|7+B@Vt+zkK!$+6cr^0}YK
UqCg!_2!QK_vc_{1*dqLY09x|HegFUf

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/readme-light.png b/docs_new/assets/images/readme-light.png
new file mode 100644
index 0000000000000000000000000000000000000000..f59fe8843787c25f5cd994e7170616561a481df2
GIT binary patch
literal 38943
zcmdpd1CwN3ux;Dsv~5jK+n%<&+L*R&+qP}nn)bA9+qPeQH{N|e;_Zm4I;Y}9RMgJg
znJaUx3|Ej7M})(L0|5a+l#~!v0s#R_0A2^cKmp$$`=}RyAFy^38jc_!NCW@9Ku2uL
zjDTO_IEkq{DchPjx#~L@gSfi7GMd|1IvVQR88g~Cm}Xw_;(~w>gGh=BskmibWV?C+
z4nIe>r*T}3uj2$)q){{q{J<3g&^gM~n-?uA7h7ul03<PRoNE@<l}oC=i>ggF!F~ax
zd0~FPjL=p8>2kEcXB|I}$O@y`{6K|)J^u7~y*tV}&N}u)e$IL4+J1zgp}_ioD>QO$
z2}=EcSCc$FiMjs&U9+_Rzuh#ou>bAf6BUIaDN1U%`i3C^Df8WQ)7o@dRc+DypFL;_
zs`JUEr;i~lkV9PZoen~oNMjbkzsZA<$>s0fja-v(($v$1hmSNIE9yMr-1c~*wbfdm
zos~FScqCs8TQzT0!(k)2tbJ+gEHLmi)>?O_Ru`V@--A)UHiq~&Icbd&eZEv}>87Kk
z<nYhn7X?M}t~B?bzkVUh(L)q}yyF>wKaHDsrFrD^c<P;L4l&a0;#ms6zLv9}9TY`q
zN;0WT_jDxSOuet%@!;kysyD)=<pM1g`edJ@eZ?N3YW>8A9(Sa@SZYf{x32BoWNpKc
z64zA62_}jb``BNv%i6JN1G;nnI#^lTld;n7cpOz$R)(RoydQ)jH_r;7DK4ju;P`={
z>dBikS5Qz8U_M`xjD?Mz3Bt_GjP?{h?xC;2M0?)9x0Jfp-@g}T*O#5Uru>H3F6lhH
zSzFd3<9k@si{bM!e(3Ml@Dj;%bZKGv9>bBq?M(0o?l5Kko!hsi_-}7v73~AD@Q)HB
z5o7(Yr(whWP8ZoSC5+mQp3T&i@^x#D?A)6TzZ#??`Wl0c9cr`!fJ<tKdj06RM#t8D
z!-EAf8+KBCw&zoPh|=y>XZyiQ&%F2l>~bkdUb&kUJ-+9|2}zFoQMzt@eSM*`jsRf3
zDKQwu+yCXw5IF#1LuqU>U%@QeJcDj*d^~m2_xtmFxk|&36)PpBtf{V!QwgE>k)r$i
z?|sdzb~ti*N2BTIQ-8ZF^asT`Uvo1Xj&v$xv+zdr$wKq{1MGkyA^yx{jm;MNpGPdB
z&3Wg7EI6Ck(IY<p#aFTX5$oTRR1}Je`EI>!RD8PL3UDs3;B&b@3bPjYHh&vLMW84E
zEYW_Ei+FG;K=6Jm4!zysa@CZWX`u=#m0qz-aSyR|=h3Tl+RK~C822q<THp3C01v2r
zTGN3yJq-=2O{<o+_v2!VmTk9l)%amX8TA+hiRe;%i;$EC23+vb`{x%j|L4Pemqw$B
zyMcj$k)eDs1vSvikg!P{k}Y%IR*hR$l5jF6+3IfjSri6xTuc6hd=O3N3Vau|G}Kp1
zNS<P&&wzbaCTdq6t~L7)zoW5q>N&29h1)(%%xH8C55(yr?G*wd=tKtFR1ypU2<TW<
zRr<5f%_fS)w*~<HFM@&4C!f=d`rF=<lp2oo%6{v*M$RxA{ZVdYMWkTV&#bN+MR(o0
zmZONhy}b;fKo8q2dvA+ef*~P7pX!&{i2e`g*i*>z0ki}xtk6(1?Vt1Nf<4b$o}csc
z^R`NWM+*{Bi%8Qr%9n_I7mgPxVg<+$<Tzs1nZmvtm?=4wMM&1;KC21@sKD#C&8yyA
z=Oqm~0;JH3*}_m*Sx5?1!s6gyF?bk~!Gl%HVV*s1fr`~ad9e$kpSM?EAqiqEKX-E=
z-Q7cz-j@}_CMRU3q@1epw?<ZtXV2DbU98sGHKoi!21{KpneWJ#v3t9S>+i1;h&!c4
zthV+`Ei2?wRS0|)8W^{ycXE&nwBeY1mFgWF6)zzc7v!%pO1^2=%5_({G}I_#qWbjF
zUjh>j@wx^(FgA1(Wjn#ERt}8&Z_GlTo$qcN&h{49y=)<bqb_GhU8FRe!^Sk-UBvoY
znAZ2!)GvKwe$F37jw67J1A~z!M&DaH`oXic^~mk-ZtPzi9BRNIRwyKxl9WU(okEX;
zj)&*jiekVU-@mx1m3Jde{e?TGCUgh|vxvVp2q$VR)IS=>i$C{U<V<2lJ6e3`yu(M?
zz+g|-a;Zu~Q%mc8p-3{3pOKM~U0O<t{0|r~GH%|<0mCP&5DW?Br~P8txgNE*sOxYv
zpTcl-tsUSc*nCRh_Yna{S%~$+i~g+66#hxUl4*|W?kO`XN%zR9dH9*wF;X<we!zTz
zj_0_D)g~Ts@6WLLnmV`Y)JQpHoP_~@CgN9z=ukn0J@~ZdjPhJ8Q9Z(+gAt%Qm-`(h
zr=v{XN<yg^(fZxlQM{w!6);j2!bz$)MHFUkpOop6xUJerX_`8gUss#Oi)oZuEaVZ`
z0*sb>UH_RBO3UFf=MpdGskxpvK|C#sR}rG=BAWZl-AlEltG7CubIR;rpw)$cmVocq
z>Fw=pEF8&{#V^eag*!H@TyP!%h-dbgG8P4A;o%>W%XE3?#$cX%J|Z&n+5q5I&HRfe
zyH`?Dns~#P8s}N%Zm4M|7m4pv4k~yMiwh=4C0U;aFT_w1hW+OiS4klExsF)db!}tW
zNWYdM>HFbQ$5QWEa~0JIlaFuYk2bm$>m!?IAbuMFc|Xtd23~%$xlX8u-qB)k{iFtJ
zFy~qPB%!RDy)>Q9{N0JmO>xSmmW7JG*`0{!_ebJxUrs=J*;6tGAStU5ZeQUfxj3q&
zDDqTimawk^QWVrqXW{lP{XCx>ynY=nb(-RGT6QAO#LZ`7XD=4^$n%Mzh6^E*aLw+i
z7_`&1tN8G*j@ef1a;%{u&Q*}({bbZ|4+aMffZFvTIkZ)r`Ds|;YP$&$H&eden2pw~
zhmi!yR8GH9zf-1LiS(txd897fus^c`uc7DPZN1KJJe9?TqoJX}0sv%-?l0zh71zBI
z)!ky2Z8=#*sSw0D!#?yP&P^A!a|lUswbs?$)I>7%xJi2>0fQNGOd6}?WyNAvHH<wM
zHF{CfCTMf@CAR?#uLKtCL9cdx+4EA*qw3BMw1yJ~qvy?i4l@pwY->AOtx#DgMmj5R
z@rC>I?;;!z%dd~u#MqW-Ygu8BR&lEihhCwdl}^hwe8MnnB;D)K9N9#^eSfcz46@*o
z*`VlJUtaVxV;^}wf@IiXB%{Vje(e{ic!7qG*gDL=-ja(}T5X}nbxH+pW2Qj0s=60d
zZ&>Hh>^_ZkuWv=jU$2p|>H4~g3LNKEsbs%UCs%ud1E1dsxxKZbf%#GEt9q+Yy#7MT
zr|8?WVf~cD`76Xn$7)c#a!`!MtD~#LVXO7(GziSMmhD2>!B)bjylCuaN>#P@Yv^UZ
zK+5H+gG$8(1BrW3eUQ;eUVB&-vZ3$urfYBA`eiJ7O$X9GjUQX1{{A+Jil-U68Qco%
z%;JqDNk6WF)KBst1ek?S*Rt)|BxQ=gD{Lbj;l>Zj*y&@>i?8?V@A(QqWk5=rRr-}c
zFFm2&jcU*bW>^irrm?rDX`^3gYpAyESnJlWwY49Sg(DPP_bdjTIUmrOEW4EX7Ec3Q
zEiUiEV<{8Etdp78^y)WbvkkXy#QDn4c=A^^zQz(vp?A}efa8&he5NiX-_wNn+2v?F
zw8;(IO{-1xl_SJ7I#xjaYvA1+&GF!E@6aOKy35P^Ace(rHV5xp!>)Ih+Xp2DfCK@Z
z|E;6MMBHDXvx?dmM^HPJF`S!I+Vye`oH$4U{g+1+#@vVJ)Z(ajAc>@E#Z>6{hNpnu
zWp|I$^b+Kz=d&jF`O!kW-uKC2t|Q%j(Sduu-J74y)Alwy^Q<bs@}~9`V0`ZSS}McO
z@?@RKSm5c>aQj-Z*1)vv!B|DR@iiIWqni@P+YfZ|&v|W8oAnl1_t(QzWhg}4hk}ZV
zcUsKA#hU>c^=E7tcTN6$QsQAGW~8tNe`)ATlx!`f049iL(6U>RX5^22aH2$q45Pus
zfF_oVY$?Xo9`E;QD%p%AB6M`JktiJHTpk*(z9FMj$;7ORqju@PI)A^~!Kk3%i%T|=
zzSp4Fk@*)2_?T~VViWc|kvpYbDK<z{_&V5*skNS1o3)OmX^d-M?~iSH=Ga@<!qfR0
zW#+x4i2*1gjY`f#`y4dYzt};i%u44d*mMgkmh!6BZos#t<G>!&Q-85i$Q3qmivptC
zaRp(ZOp8y=#(GR=X=8=Vi|fVVT!{W*>7a{rl$@~p?Ny*$X*xDyB!Qbn{Fq^ojeUZW
zfg2O%p}2rVnW5uTly+Ff6b>;mwXY*xDsD?K0C-9!3n^8N#T5K$5AxpqMUq=Z|Ihch
zRF_9uf>58HWc3OaB#}~93HMMq2aAa4I|QXbdZ6smaVK*-&{Y1Qks53-I}QV~KIUdy
z98M2qtXn}9&u4RP{p7K+F3+im)hbyjWqDy?<EP8TiaV<CCe^`Y-Z>j~E?FECj8JV$
zOK~--0^Hr2wXK&|Kc=#i=<pUlRfl8m{PjX%!j%%Mr&6Unxh&k)J7ix*|0FzC)~x*4
z=@BaHeH$@*qxklf(&`O7MMbqN7bSgyPtQ;9UYhO9?+tmrR6w(}em%0{Hg*GlQn$fS
zge5-G2)u$?_%%}?MQ^}ID_L$|jgSQbxp=V1k~nxM5)QY~V8A-Em`&LOYob3|OBdw|
z3-a$?^3e=xkRX(BASS<N)MzkTtJ5FI0iAF@Uy9Gn9FM|bkt>V1+HFNzmad%tE`zx$
zu&+6XjS4bCA#&+_>Zf_N_4-KU!^E;4D3WYvUq8{?*zi7i^;K{(9hT?Qo6!+0Yk~Qr
zFOC*3jIg{vH)b2A^l580ms3F|fl4)8O)Lc$XAp-XJXq?SStg;OP2?b;O;6$u;kM%#
zkHdOGMWKV~23rOx>_<tWDx4ALgbDXqMBW%Z*&G&SmV?~xa5&a!#lB8dKAlLX3k6Rl
zVjbjVoC$|xuMdio|GlP2Hca?6jUB`XP}rg$ykZD|Oofwsi#}MOf~-IkY+S#TuhML;
zkalr#k&~2^G~4WO{twxFl_4=vcFJ%wDbCMrV+GV(TaVj^=@5Ty?w^HUS%1G0*H%Z@
zb2<$lG0y$k<JLn@_FhI<JUX*4nmVV`M)cFaS&ipDzU^}jLD|fCknna8ZgM~iT}iGk
zoZ%Dc6^pRR_KPGpfcpbGYUnQ*)_MF(8^DD;st#Hz6^3*T4nCLjFJ*{<Ah%I;Py(4O
z_&^&L8~(&~BC6FaoN{vLT=W1TbCSJUe{i^WpkckwN_C?GMj|#?u3nto%e#%!=_DP&
z+-F4*G-udFO8vVk&nnqyg9pr%^ImxQk>~#IuE4haFlGJz{$AuUXY_QwbeUj*d9>pw
z0YO$-01~RXnPG(4r992nDcfN-jNtpLY~XOfnbpD3F*$m6S>_^xnOv<=oalDC+4tJQ
z&#&MgQ_MsAdO}nbg7|ce%Lol0!LiLMdY+Mpbh$7=;#Rr~n&CoZ=(GJ@e<&smv;gTZ
zsu1L7n7Bs4tQ;0SUf*r62te?m1R>P$Qsh4o<0Z4x96sHn$+48@$*Gjj;@?aVu?nd_
zE$t`F6VNrNNQtaa;+6Vcp9d1Op0ML*U4FzNrQ2K6k>a6ShbGSDJAQnzV|r7(xFr>N
zI1W(iEc@QhNi%(7&PYi~4R$!6i`v-KpzNu40AdNeYqQg;;0*a;q(up{$i?oS>)Q<(
zLI><84pz^Hk4!&Y>gn1?I>kjqrRM*teSj-9`1A!piiy8{Sci|y)DjBNukk-6QG?|p
zy<cpQjxTwpImq!2ymO99&D&xOgq}kSSoi|H7{n%*tl&D6$|p0CirugQVYD`moM@EG
zBsNjYG&&Wzz2Ou=8?uT&;keZ--J?D!GjqMId-{?+^AcH5U6ipCvBx)y5{X-A1$>o>
z?E;o}tE{!w7nd``-0!S9Kz=GgrOJAx==@Fg{%CTXfrdu$Y^W$HA%+s@y10iAhCT3E
z3rj6cWWz_cQ5kbfztu0Mjbj-v-3KBM9Si5*eM;60e-l(EW&9b!lIg{Z6K^hmtf#Dz
zP92=U_&iN)xWGsXUQu*1a3=#rKtJ+T3q-Ly8m%F*_muKo&$w5Y&p>Va)`>R~b+nbi
zZzcM-Pu+)72Q^9yi-&Tdu#;TwPCArCEm*CvLidM-k`BD?<rKDocdosJOt!%t&&|^!
zr%hcdM|9@cta}7Kdziroi$SxfZPVuVaam8#sC7LT1VZBFjiYWixCFWatb<CVUue14
z##6hc&ecIHy=boUT5VXFI?WR)>pPgdCl{n{)#{l+ugB{>43+E)G_Ae8U090dyhgdx
zR|Q)*tY%sJ1sl{9->XTCh4Ad_zW#gj>r7Zr3(ufF_9V!r^=zsQR@{?tdkbp&cPLr{
zX_>Ce?J2Xpf84=!u8Q4IdcD`vXn6C=JVl|wN(0qj4!o``=Z47F?L}Q^UiY?RASfbp
zzz#Q)_bV5*$3`<9)RUJi58|Tk!Y;l)d|+-Eon)hu=OrU)-rZGLdToF>A$B|p2mHmW
z7#8?;i=mHX9h$Iu-7VtUOSGx~_zgn{E*IqkF%mcIO}c5*>UQ1z{3b3g-YfNaMnOju
zK=i%zK&%tYPowX@q^p2=l1#s%K|)irN=4Z{^pi=;ZWb*J&XvufUE5cq$!s=+*>IR%
zr^yV}ZxvOPT(H%WDJu$t!+B?6x=<3?Yc>_++uBkj2IBsU)lE>T>xXsFOhCvCf^u1i
zc|}M$HCnj@4aPuH1gm0ETt6!&Nf8E!5`H?=<9W+Xt~pbc_P6p>{#gyPcz;UZE}eFC
z(o2`_^_(H|_N*`r$=-aY0Ln31>GF$F&p?S2&c=LS0H8UF0AEb?tgld7h<v?Bh$x?6
z3A!)RC>&?UYAD3#hxk<jlnX?8sD`92r7SJ;DGd_+Ee0ru(a8P6dU4j;$mn02r!UF$
zO+vZo`RSS`Tuj|sx4JI7^NherQ&>r24mIkW_zeVJIrzyi3a=V!-b>z8`BE~i^)Er<
zA8-h4GKABk=(g0f$O1kw8%*h+)s;n{9~7oNX@Lj1e_vFm%<BaT>>F>P3EhZk$;kE*
ziLv}-zx@_t+4%ZPgtzxule~Wh`PGDjk@QywoRx1v-c$7l>je;k{{hym9-oJ)di<X;
zJXh2;<Ed89Lmb^yI&lFdA2>zHWAmCpAMGf6)iF1JP=6Ltu}HeBs+D_^CM__~4fXV%
zqvA)$`ZTLmL25FPPk&B2O6r7eu^zx=V4$*`4C%F?#_9P!E~;H_=^QtpJ1!?nVvF)%
zZoL)Y>Y!JRKi_wbqtj4Zy61-ie*L3m5rm~Imc#2t+na%lP(+m=M}m+eIbmdN8AXAd
z69mBp>-Br_|NO^9YhZ3q!z`SSjEhYh6XsYQFVuquY6WGbAk#2CP3b#i@iA0P3{f%Q
z4-<TX%X&*flSGMD2U>wr25Udl%(7jpJ!zMZ4n|2G6K2rVF#J~^{b9PHvHGW=W`IfO
z@z?=$XYsn*;e3Q`@As^h2let}5O#vAFKvifp&O*6wBFcjgwWLf@i;)-$sinQ;}5QO
zY9f^`2Hf*BubL_w5xZp<JxIg#IJzs08=P6F)X_^me-YxYZajakytdnJFbk#_4$9Vx
za~pT(ZEWaPjF^#O16u#LEj{J9+JG8|<nWhTkSAK$6xr{n7cC2mvm)#D7SObdZ|U_K
zm~tt`xTPhRCda8Wuf-q;Vgm~~&%Hg*p6#!FJP|$l@cojOhTTfp`fvBgEJw!Ii^gfq
zWZ#TA#}5tR`PjE>*E}ciUngQxudlN?=Vr>!C{Eiq`rh)n++42{HRjxXapYqpH!tK&
zKH9SeC$_LY4Ix~d#(#@3uv2f==%yQ;*Ig<qDil?cADUFKm^6k7Tn}%2YkMTNx4IQD
zAP0ViMMsm6@aO1l+GCd@;f)>D<{Sl5*i6-iYZo;PBG|nqxuslQiKz<F)b@OE6eYRw
z>N@ERm^7loKyW@rIl9@c@kQY|VuAVV<E}X2F*DD4wcJSeeA-Y%9!={@q{KSszDZ!d
z*M_~`-r=Gd0^VF&8XKv@N`zBYvV*sm$80$uN>q{$h?G!oVBdpsc*Iv6?gp)7^~FR{
zX;w%urZSTm8j%5_M@MjwLa236>PTM57^Fxq;8kF)lA|eEy!PhOfhMNorkKI=<V|%+
z5kvViyi5+adqc-39beKQV3>6`7C&)~>8QaxN}GI9HRM&n$U8D@*7?P6_U{pM6#)RM
zp3kRsWiS3*Y8o1Iv$-w8Q?A3kEf}BKB^M`<tc;QvGfT1<j7-;vbRlqGtx2XJl!&=E
zRpbmlmyP(QQVOr&-CX_w6atgt!fMY?cb*9Fs_os<GE&hbk=<INB*S^luVcGm>J!|L
zuNXbDco>qC#YX6X5G0d-G*t66e9@9>+GBP3fj9y-2pT%v^IR-}H5xh+VX@G8YnLYd
zFZL>)GINCsSVL+_C~e)8iBvkWsf@cV^TIF%1==fC`cxg0CWbqHUs+rp7~!z6<m7{i
zSxKT+RtUl{0e!^X9#47SqhzN^Mf$cRSw0qeRJLBalQ|8<V#y>fSkS8ge4|a5?W>Bt
z>P#|141aVbi7g&AguCHeK07ATem0aYl&Yfd3CF`O@a$vcd(qPJ`#~mP9^nAm$YlBy
zLUy#hvEeXJWA&~{W7XlN%4(J7hb%^d{oh>wgfJ?2wZPs?7Wu9$9&u5be$Nl>_A!K?
zKSu8cyr;1Q3*%bWFR@xT93tuX`M)&IdZDFlLo*0uV^88Sbcs62X8B+XtYnF7db-6G
zn)CHDfqmh}kNat=_RTAoO|*Y;8_*Zu)HbuYP_50N;t}D>z3WOE<Gt%^F-rv&Esilm
zygmKbptKuT5!W$8UL#=h81O9^5Jtxs+1LU#JC)6Zcktla_9j+|_Gs#KfnMfsu<a0i
zDAW-F<}F#PCaujMTKSgT9X0xCXti<lzdjGfy>}l-&Fg+c(WuH|fVzYd4E)sO%^g1#
zT_)q<3?jXG|6I95vdZ|wLb5bUoMXaY&7W<JaHfCU8(FcciL%o<$v=nA)`A5gU~Rt#
zlVvCTi%Nc@oVqG59?#zsV>G#OcwA4wJNe-^_t%tsFrTUa%j*K~t+RxFEJjz*4sM>B
zs3IIMpOS~EY#h9)@%VP9l~GziF6b*NBF4r=xZ!EHJL#Usro1k>bT_3b$;(dMR^5=#
zLTHn1*X{5&Jc$Qb-97lqlgYgQX?@&Igk6YB3~8R4LmNtIP8@bm2dvRRG~MgNynL=M
z08pHiloSMyxTYbUr@JgKfhjC!W68sfpyXt0i5^3K?mC{{hvFX?%iGiTQp1&A!O(F%
zn=^6)K&@H{NpiT2z@+1rvCXRJztlVS)^dDK3GjO_xkqnv(Z-*Z3qHil_&V@4gi>E!
zC*K8wR?Ec}|4Cl{=r+@wQmD+#%+Q+LvTkjb5yC{=ZZFK6&E$BitLaB-t`I!RlMPme
zz4m(excduHjM~g%d=;LJt@(t(hy(ce_O&<wFo(uEm5m>1+couFF000`^${4VnLApR
zn6<&q9B*UJ&E%!cOr=0~W%-{0ehS}h@uIS^$$ucfz@y$FKwhxzHjPmI3p*b_VnmC{
zawi=rSMnPbAkPZn{ebH^H8eU2zrFjWO*{dgh(kGzH%tTDmd&K{5aS+UIhOu@l44A~
z8Qz`p1)Cqldwn~vb)9qmv;h%rzIn`+6&GL4bJWI5bC0Ss54Z}KC6bR!fn)zH)X%Ir
z=Qun(e53=y=>MSTLNf&SQAtaRf<!NxqzteLu81N`-rS5Ro<Mh{)7(D6u%4~<i{0_x
zkNh(6&(?LlE-3SmpR#Mt>O3{dv8b@8o4Rr*3;TT&6EfI;mnOBolQ1f4&QePDpVA(t
znv}0>G_4F#e3qWxD3Y6@lmcXz?Kv+#^xq1zJo}2hUIA@!-cR$>H+>_y2}yB(hbOh#
zo`&^)nU77UyB#*SS@xeZu35Ek$CenKZxKf>FYM89*unTG?L223U_-8G;EPK2i}A+T
z2~8pPc17zK4&U8t`3`vdrulCZfq-UI)O*T<l1=f3nj`aX{OPPQU$NR1l=&BTK?|#-
zVOTI$k}#x?ZbW{T)QPC=5^OJ0u+lX0?dJZvL`xEK4x8PvnH{(F=?lCNAQv5<9FvD3
zaktGmsgY2XYO4w+c@^Tk47X}$pPR~R-*b@nvxYI569#Z7Kw*T<<m=31hmP)4S*6pF
z7y9+Xh;V1t(VNzUy&SwQ0p`Lqho|GZaLjv$z80ALPk^N2Uw`mxEa4y`qGkR?)*=Q`
z&X0AUEZb_bJ_-pZHpP&;PG|`KiZ^S`R7M>_;HfX*AKZjeZNWvgxv|xsB^5Zhcl(!j
z=&-quGKuOK)xWm;6|aN~A3NkF9(ZuTm=4}xl`HUg{*Zay#k#u-@?;7Eq@r@{r*_I%
z;OT_Dnx~ZU_f=@$ooq5dSv2eIhkqkp-V=YP>p_y@uR--Ks_9HsTrbC^;OYWqE^A0w
z3ehV{W;;!h*=^SP!RN1b`^=vebag?4?(xsZvDB>Y9JW-m-$j@Q^HG>EuBsvo2H}Fm
zh&u69R~655uqkzXXv%45y59EmXz1w=e6M}OlJY8OvsqZm%Bl=LxD4^E)KoA#H$(K(
zfMuEbb9(56G39cG>^o8Eprr(jrC`;ookp{*0y4CS2PY9om^59s&r@C*)%SKr?$!Tl
z0i3knj72<o9QNnOuWV1bUEN6Bx4oD(|Bj_(cuKDnT4vp<v}`PTZ7)EQGMeVNK?l3~
z$)O_Y(eB{Z3T*(5#!IJV0vEl=<!&&);J!EgJc-?nY_eSi>(EY#6Ja#(!vi@=?Zxcu
z|1ZMn;^A~|Lw9l-#mk}2UCs^{?)QdG>&tTQS8~m{N1MUC8kl9qqn>Oms3|l}4c8A1
zSXtqXel-NkXB}kOc}A_eKYe~C!5kyqc^A%~hvgF7GWKCZv%721s@9CvDc|OnJV=-%
zK?+VIt0M-C+gpngG_F!K9E6dtF{}nVf|=!&f#MI2zlX%jgAl?8D1B0Z4x2FhS@^ZZ
zGC;|%ex}<k=IG?^y48{NoYL;Q_<^qzJf@JpFUb0W8HAF>dqY0;!<m~-HDsto9~Xhk
z7$rnMj~Wh$rUpgD=NrXcJ4}M#A3VBsCpxD5`|`+(HFyXyC9vRbof~9O;Y@TOsA<Q<
zWMtY(x}=Y%fCx=EUm?2=GIUeUYoiBW-aLakLii_!&+jt|);Cy0(?+dlHOsDV-3BQq
z%@(hLyx>Z{UK(on1U~w+yTPcd-iS!0TDtOIVa_yB{G*URWgf1qYoE{9OGMyDv)T;`
z;=CA%dxA;WD6`&AINS9o9*FJ+1?P<r1IEel)rfZjfKcE8SX}0sPNgnJKf&g;VtlBk
zS^XPDkBtayYaz@eey*T2l^BhI>70+7n0^%$MVRe&PG$|5)rJ*)DABwq+`{7m$_cdm
zZ&LLy=TWQ{ki^vn#^GDgX!7f{4bKU)HLqu1J<`#hl7fzJ-h2y6<^e*bD<!vQkJACY
z`Qwop@$vSip7&BA0shLUPJbbmTL#GQDzq=NiSC=l14-7zkamiZz2-#fxjG3h-G!Tn
zf;K!suo$%4>U7+#K=)WWbmEl&TK*V;SheL0lT2_87L~EkwbJ#*C1%&UytFOX&+A_F
z+bZvwD+ad9ZA~>CQZ%kDPW+j)3Yd_k^Zq5l%T;YwdtzPpSI`Tk??6nktd(j?vl~B>
zyVF1ZRBZO{<s&m73Q~daoE8<NyL==@T_8ja6(H1NwK<&o{RWoVVzs+As)hm=jGGiZ
ze_CY!)-iG}rik(3Puv$*RD4HF^TMCEC?`2_!d`6Vx@SqmP^qz1+L+xa=fdZ^AM{ca
zLbF;oKK+r*rqus7kypQExyEndlY_J=u2g}5kpP~&@pU$8)t>)4oaJsnAAl@&fN6{c
zWu^U9%G_|52Ne`)`(<ZKIBL|mJuhb{Ls0rNybB_oT&GO}gp@s~SuNg#7L~7ridPs0
zSm#HDZ2YUN=@`>*T_3po1Gf0i<e!GR-UM^_Q@v?y$nkoxv0G3atb>xvGc2`}QY|j!
zW&U~y?d7J@O!td8CSdMHZCO7t%hiLyRaZku+llV|pmYc(Vk03BwFe!t0s>Dj2_b_L
z#7QUjVK0*DYj=w0n%@6r?O?<+(IG!f*<AcJ>oQ0Jx#KdYlJLc4f6W*9Q{IDDz|v9+
zDTJkbd&MxG=yV;==6&AgUHt6sz7hJjr?$)ONvPiP#@D~^<RY^28Nltq07sDpNbOX$
zovVDSp6z^+<um~r(dX%1261rwbw1QCu9)m14Aj8$oit;QpE)w$M5-QxvEa9l1P$yW
zKm#wN??QlI#>t5cmZU#Nnnd(qER36X4npHNzKTch=MM-7rtUkYEuZ^b&g>~l%uRM3
z+Z%o`2r$qf6o?O@qC~9VA|0~lHT&Bsht@u90hKM?Bh{u!PcUr({v05KZncJ)c0V{Z
z7cG4A2y8_N#Ny#^?yo-6nXn4ThdxPUa5MFR=CD#^YpvXa#fuj07^X&Fh@i#GBJD3_
z@y;6DuX`Rx#49zRg(V@IXC#cWOh<n-Ioak&uwn8;bE}Yk#FuR0y>@PYUU`v-a)W5;
z1_~DXO6de2q*PuZ4e~@9v>n&dSXH9OgY?Dy>1$=QabxWkdiliAb>qm{@Fi#CN@XH;
zyB;^Q$w-$2Skv`RMZkpK<K6E&{$bxp%OVqx=~}yfj!Z~Mc-l=)x+`j7&o9iEK<ZG@
z^fcadxpnirky^uNHU#2z=ht|WUcILAkAGtXJhUmcH#J>GDpd|10>?=a-Rq_*7&hPI
zTMaudFyhXPIGE^(xTTxFZ0hJdB$%KLp*uAvOyQ(pC1mEwQ4p}-s0K%v9R}@c+K!;6
z_#Ii5#<=^S`xH>jGdM3?Hm)*4?tg@YqoBYweZ<j;a)!`A>=xkm$JgkKK;Pep%)5C#
z<$~aLC4=zy_6j)(x2^n!5u(NrgbB0gdrwa;;hE-L9U6>h=DajGcr6`9kPSof)P^4m
zkT|%<41PskBd|D&UWM&uzlQZoip<O{1qMy3k$*~{wXK|jc9unTjVN7<{ixeAL}Xg$
znjS~~(_$v~(ZKu`QE&tKs{tkHrgIhjU_;`)#G4*XrB66dr1_9DL4v0I`S;tW4=N@m
zpfw>VD4L0lAPWl}3lKV*gBDG-En_wds;X3e_j1m#Xhwi**}Cd}SeR1zuNC`*mxU_-
z5xwYlV9A9ibh9Uf=Mnpdb?bU9Vz6-kr3*ZaCxw>4Ds#T27T!!B47JD2c?<iV8S|_E
zwGX}H_`V0jZM~r*n~L{@Vy;BSOQesWR13W>Z|{)Z^=-^1r5A{Iqg9~+kE<M5n0>j6
z-$y^lw$+V;rT~@R$IgH9n0wWQ-%&79f8+F-)y{8s?k~>`eRp41AI{e^r_yos<4=(l
zbS_P7^-<k(84eo7^c+4X-vi=H%nNAq2zRbDnBHV)ZeIr+tv8y&5%JiWZ~O0Ia7g&G
zTS0V|nGpiBJ6#$}^Zcw|-@%|taBOG_&v#u!7`OymFA-Hft}89R*m@JX5A(~Xy%k(G
zyRI#0i-TUMF-elm{6KZAu*rwZR-FB@emz;HFXji6`_<jTFz_UF9#8BC9N`uBMT$z%
z#ch0`OS$Xxp&;6;kOZY=P!WQD#ipe4c^SC0c(KpLnCpc@%nxHL8My0e3c}190a8qJ
z+$mK7i#HjE#@ix88N-1j?ZuD{NE1}!1uziEAN4OD)rkdg&46QQH(&b)A+l!2-k5$O
zv%?W(#X%psEi3oxSsUYjp+r6_-90BzoLin!zb^(XW-U^zy!n{!4B!ZY*t8Z$0`d*j
z`WT9NtFGne(oG*1trXw6$YNOg#t43F!3nmkw+H|+$+}-nOdm)FDTe-8>9F&e`x%h_
zt!l6c9J1uoc9r857IdzzKT2)~s9n+`%7(cy3M%*aPhCHAxDkNbv|VX_eOj!Lpuz>;
zK!Z_7?3Aa+?mc>Fe176OTzwIsAOXNnGB_2C@JdM)lI`3!^r*<i5r~GbuU^<f^l$d`
zzC~44O}+Su`u>Co68~$IpDr+s_5LdQLk$6@EeAaZK_-lq03j2IL&R9i<$H@OUnA&(
zM3dj1TU*inkZ5Zm{Mm3ZI1)BWL-T3S<>LtR_pgmhK!u0Ny`!@m$?MCHbmWi1;!mX;
zg?sD9snpK!i{yz8TiwCEOp(Kaa`(Lt9s$g!_oLgCAZX1d_?+Bi1~PWYZVvgP48jy*
zu2-s+xn1*~lBbi|!V!3q{^<6*V^#wEW0=qut-hyj=-L~0iUXqlwv-iByBkAq(7vsP
zTohqj!lms`$+i)bAG3|7)9`rMoA}3k6k2-B#;D=Gqx&C3AEgeH514pTm9cuYMP%*A
zZHBW66KW59&ZpD41dpRJXt#%K*e1nh8YMxt9B&e_YTFk%C0~vYg4oyeHAex;h3?)$
z6arTdjg{mtwf`wBQhuAou1eTR?ug?UB_kB9LXePku@)q~d&y*!u&Wrui}hVUqZ1HJ
zCa20C9OddwZ}xU)&x<2Ko0$<Zg0M;5^qd}QvCI(DwvM5|%L;TBp3!!K3xPQoJWkA#
zweQVE+H+?Q4iEm!^A-5;2MIBssWf%1>AFCZ<8f9G)Oj>elS(|`OfLTQx3nad?;R`a
zFGNv5Z^fT??aeZ82T{Fa>r$t14XU^r<X-#l>L%_%NHT*OK(>Fea99rTU(B;jhSb<g
z-=A;p0*@9BmQ@axOKh8M4rlG_Z}S@vRNi#qo0lxEg2fsNSYO8qde9o)zj?h=Qz@-t
zYtH}Gc7HhpzrkhD<3{XB2J4IL)z?p*a@Ok%k3(gl*`J5mMlSschU%P|E@*=P_LM4x
zUd}K!-=mC(@h<t81+yY~HgbLR)pct<hg`afYuC}pB+34<B^M2B4hZn3QR=9PMGh7o
zX~dq}Gb3KTS`4slZN?c1Sma3>iCOMRsF-xNdK16WLHrXWoX-U8ABkrKMcJg149Uo1
z(raDsexS~7c#AoF<Xy!iVtw!QVJCInCud&l&PL&(MJohB<*gRg+9zD#;uz^4qx{l3
zj$pQzpt0k`H`6=_bp|NjdXO!TgCR~1i9l_534xXB5{P__1~AD8ET(m(UQpdDQE|v1
zl~S)-v8xmthhrf?*>hzxT!H&fA5-ut?muNFaJoCNUW;BgGKGms>_XnGU$*F`yV!4X
z)tjS>5BdB+NZ-r7rv=i2L&2qD$RD4Wl0X$+@C_|}DTUM7meFnsy08R)JVRtLdFv2p
zfPkz8C6Z|_Q!AVsW&g?6_OSi6{LfOnpBEtSrV|;&*I`X$ra3?I|F$|Gn}wr66bAQG
z|Iw+ewmN1-whR>}$v4+*`_3o@p<QMpM6i$VjT2P&!c+|!Dl`um0cte{Tp|9r^?uqC
zvg^$CYJp@DV!6!9CGu-j>qT3uZ+p9Qol25)6kUC5EpxGnY6=%#l8{-Iju5r*pak_h
zl$pPphzB&5qcNl{-?WyPlQFgJfK}VekM47HF~mqS&mhH@?Bw?^A1xAknY|yIB6~d>
zN&+Ie?yN&l+9L{9r@7beB?e-9jrf$tY_S=&-PFa3N8slaEQWLT&Jv$>5YO*Efvu5v
z{L%)KPGn`_;PpRuY3e<?*L}w5qN`$1eh-`*J-_9i^O7Kh6BMYsxTAEiSFBsGl%s+`
zIyVX3sT2S@(fU7tBH&R?<Xrk%{dpt-4dJ2S09EizSLl@{DsAHZ`}bL=?xE)H!omVg
zjOG$^(qi(l!@P>6ox)m6^|lVIW|!Ps9~z-F8xV#frw9jXgF`H$(r4ES>>Mn|_j2#-
zF8<zxt#tMvKV2Mllh4+u`M1u3UkS~WaxF-kK)|l(lJf-7M)+gKE7#9w)36=<#c}@<
z6O=G)zFj-4B>Y|Y=EnP8bn61v;SiNY0Zs5GQe!nf^<r!kT^T0y1bJb~Ob2Gl*TSo<
z-Gk8}jw&`Yo;QT}HEeY)rVQD#7uXmt7hb5w)NcBL8%!0}5S;i&v!KD`+d2g2Nr`gM
z?y8*d>()N*(s|c3lhS>6i&XCPu(aW~eqRK&(q5FvllJeuv3eUBSdsUa9~<kj=5P)<
z@+P}u-Z=|;y4s7ZcYsS+_<s`DRBNd21?p57>}S6KFUI@b%s^@EwI5T@ivi#>WS?_Q
z!#|nobw0kUbKWjnq0;oTE^{Qouw80db3#YpNc(FiO}}FD5xwFRDOzw9Yr+q2&0bv#
z5EXF18STGG;Puj+8AtMJUz+=+;W9;l8%#q1G&}jQ8RsQj2C3DpxdxwXXo))o=$VgU
zC@s5Ha6mz~PYU_sue{d4>h=54?fOJKUw1sYQusJdd>5d;_hhvjPd7J}>g$2NxY+mw
zTwkb+ns#W2!p(y3<FZ&y0i)0Ly!)e|Kb(LkrX%yo_CpcF+w>;p^ymWKV5P|&-ljD(
z_j6B;x<Z&3We_G9C4!idJ%)hCmHE89x6b6!gLtD!d!Ch7m+9~jQ&e1Bzc56n2c_i^
zsxT;8re6MJi4?U2P3mIdQdoqxSV?M%5MwBkDB1Cn%U;SyKW?3p@~*eG;&gshu6>=D
zPiViQ(ctxR`w5G!=ZGpQE{p$nNQpPCk4c4>bX&L=deBaNIsD)^E)Xx?lAeyHGu1$(
zkm7Bskav*2ocK=<mB(kKw+2{)G3KT-^xF22xyH}%g64o-@y>XVqHV>(*<C(F8i~O8
za?g=xrd~~=f2;n%wL_k{BU6SJmKp_0Y;kMUr*rs1m3m2~M>%AnNAM34elQgM4(%tM
zERoFVDGb{F;!4gI!)?{k9i{2+Cmeh+OQahSwfc5fA8eDc=_sq%wFM;eKHUFeziO&)
zmDxXvtZXxLV(D8?*krJV4yz9)sy>G_IP?(8!?oH?8aOyO$oJ{>b@qOIx{cI;iBH9D
z<}_Mb4|1sVI<E#_gmU66k<n`RnLHOcc9J?+1L~v*)ikH=M+|bj$4SRr=*UGvu<FhO
z`uGTdJuf%k^2~q+l-?fBssWBmI0a4Xn;B;stF_A!?1WSu)?ZgCK*6c*Gj0-uO1ilS
zv0q~F)F_zC4@#uWomN-*!|LyMXn#S!yNC$8MEiNsdDgA(8}y!UeTG8uoMGo>ng0F9
z0$R8PyGml6Wq0CdW+iR7Ib{1uuzuP&5|v-U5hZXY{I8}t@Av5BR1|0k!7vbD60qeo
zqZbVHdjqljMynj#4&UCc`P6WPvCuRWz&Qa<0wf>x#uRafYcg{;R`4LXWfL6tMTsKe
zkAFTj%N5-wLt~TtzjF=>eARe-y=$|83dF>eyjh<4sgdorSD(z>cS;HpP+34m(6ZbQ
z64;xUR|mip7&Aelc;<EDPXEE=hPM6D!|D4507zJ1a=GuN{uRiKzQ#ZR3b;%SHujJI
zSy$JW+UPv>1{<_LpT;}%9yFUfig76GBmMmFJN66BN&9QaKrtW?&tT~f?Qj&?OX7~3
zhHtJdE)3HLP}33f_9+_JT{#YC)Y^2idk_jL8bIoUp=p`&p4zQyBDS5N{^Q+~t?!MD
z!p$Er_&fb$|4VqR67cpLE|cd%Fet2OA4JYt43(2DytD>k6G7-uQ}T=&^iS(L^b&IQ
zIhA`>2l<fdP8vPMvd}6nTzRUM@A>Ck3wsfRdb;hGy>QkqN#)qsZ3!{QXug-S+?ZNB
zHYF)nW8*^(Xd-W~f|K7)$4PbD$6dR#6K>UMk#ey@mFMk=XtfFWrFWN<OqH!KZU-R?
zO+%KvLBdDpFvYgXN$E6`zg)U}5`nT#2$3P)hxbSr2#P)h&W-8{6okIHM}8k)chcO+
zZ;o;7Zb^@>2+@K{Fj-6pM9C@<eF<NxYo@BL=_`*9CA3Qwx04Qx7W$mg-y-ZAMu^IL
zx03vC!;|332^w7{Y_T%$INdDPK&3bd%LTZCsGCmo`Bb#&A{J2~>4Y+gEX~JWpsXjD
z0I!4LS>7RDch`eGq#?9;{Id?jCQoqfZkn>T%%>-FtE`M5pdS(a-L;vG#)Ox#hkPVK
z<M?hQi+2i?)SFp+G3J1|f2-4b#Aw|2@E}~}k4xfgKut(r#rsiYrfF>I<d`3F6bgO9
z&HL)AgLBcRGIp4rRA26{-*{ItY~O*#{-J3j^6s(MlA8vIwzqALn<VV>@&qS-<~q7F
z;b=%N0I=D*<Z&1)3r%FDb9tbC1KIBy>0ywCvdLOIXP!DDBf6A`W*djL9WX#W49g~0
zX8sUWU>WBW7hsZ#$QwO3Mdur|pa?=8F-M012b(*+JR|S$_Ha5<n08S&*%-?XwiF_B
zKc{2s-PBxdZBA$7X7|`}{mu4pIbl>|W2r^Cm2oabXg^(6$QO(AfI__vb2ABKv@~hu
z_`aA$RXqYi^rfU|5j~98eypv}DmZ(OmZ>{@>54Wf@20?`-$2O#=SEiPg61IWk`3bK
zCbU)tD(?<_rR+H|FevC3E|l`XGEg9b2KV9P<Es*L@*1ARiJFaR`Fkbl;n?QL|Af5r
zFn%_Ku*hgdV$yzMW<MXRpe+pakrc||lm1m9CZ_;^2C}Vxx_u;k5Pe>3>_96%tJDS&
z(2$VBM{8=6{8P<zIe-BLq<tNzJf=k8I_~eW<p=?15(?BpqOtjDj{{I=Pb&Um$8lN(
z#8YueMgUfI(2Arh<Seq5if>=#I7wiYpo-g+u&batShwl^oGrK;EKa~SNcKNgS-px*
z@$N$9RBFA|`FPPMv;rAm1O2-dp|Mm8)DL+@x2IdjgVuPc19}W^*t=hLN1!6B31S_?
zi?P{ck(z{yq}#;m!53}94ekS0Vv7ClHU6|WY!;!VFm(h)FnY=)bZlAZIKTP{yUVZ~
z5MnvFDTJ$;=!2XY=zWlvoHn;ET6VVPZj=<0s@|5JKgo+xgz;^w^<{=WRGE`I)@z4I
zNaMZo;17?0Ldlb$2T^ttYyQ}xVHU~dwHE5dS?j&>2%A{0mw&k-y>0|^FklA+m+di>
zXpQVB34_#dbnNbm$#vp(1B?Qn9V_Xb6(OUB1~~2sk#`n>04*<>AB3Cc5=<kN*9^|e
zO^Zdojh(}g5rMFT&?K?s5}e_4OTe7$PZTue_)SeVgmFO4YzAFX&HI<iqHYF<{^8Se
zSORy6yGlux|9eaxS)~SG@N^ma85?<XuS$!UfX+H%C?KD$v(NT=|2@WeZ-gzExwBQ7
z;BGBMLH5SIs)2t{RO15W)Og?bi-YVv^i#7huFArz2fe_elS^Jc?l7d!T`-6iE-qjh
zNX&oorZiIi!w&jGU&qm}Zz>hsZV|>pp-oT0C1FD1F4J%QBg$ztESF7Jmq=1{>pos0
zhg<$5nfToA6W8a8CB`qP6Bwb}mcssJ$q(8=`UpDg?h>x+merEiFg3vj3k$=b>PBvK
zV&gu}^_LK6#=46MYOAbdF6)XK@hQgCZPwR_XtkHN$5m}^-YCJu17~%dHnus071wC<
zx~ebD#FGL)oPG2!lkv7+c7m|uL6l2swe_@z^u;n_pq|p2>PPn6|Ec|`$!9?*+^nu;
z2tompn%6QX3)b~0V|s97{iw;-o`DNYuWH(;20&{eW%idPYJ0WLN_#a4>x+fm%SThL
zl{a+%yb<w_!rxT{<q0qoMMs6*_03ien(7vHEJSl&;B&I4)-K`X(&19Lv>wO!qo9JS
z`LL)d?;IsfLI)TjO1fY~)K5#G+23>bO~-@4PzyOpC}=dQP7V5jNwXX3wcQRBgWaQZ
zJ7=P&t~EZxp2Q>EAG|tzF7VptCryZ7sKHu`+}r~1wkE7ri>#bYT0VXqgU_mQy2<p{
znjbA9TBrbvF99%qZVVn^21Jm#Qoa1pltP*bgz_s^+8tiqKRjB6##zbgOo1)EqQp+_
z<N_rk4no;wgX^BeI-ZelleV{WY7VE8_mIYwzD%R5?V9@<qpQVVhWz2F`|)@Fb1jk|
zy2ip%E5_DHf%K~q=JjujuW#4>&b~bT(v<?CbR<Wu*|@f(SYx*5jiLa-|8QB6z_qTU
zDIqoHS9W0ZZTKuJ6@!G6Q0x0(=imxG@WzQEI1`5f_<^DNLx!6t1(-pABVIu$qC60k
zH)ExT=hBvkMLIw5*4{qxtzD$RX$CXDb<{t%liVI(l|8su+Qz5r_~o{=)rSrm4RKZ`
zr?whUD3VF0a52Ps>G34%X|<lA+P*ek27$9s;kiCqiH}=qY<f8}y#*|dhpEE9P{R&_
zT%<-nvUuzD_HQ)Mi#<(Rx|EksD0d~vLmwe>J0&<Bjv~e#?*>EDE`;r{q0!<4lxSXF
zFJQBnJ-ar%I2~bwfCQAX6Ni?+Ctm2GJO(4Ml#G7Up8o@w&pv~7<wA;CTd<}H07MTg
ztpPb;<hkz87eT38$`Na3g>_P5@*P95Ult>oM0&R2|NhCi+YM#xNo@f`O5wmux#OOS
z)<YaHXY4XisW*e~L^fM4xAg2<<0)!)wUo5+a_((XTqO*0*J=Q&g-jY^D&Sl&p*4Ev
zr}GG1o|2$mkHndPfoHXDULL4V`DUy#(ebg%==}431ZIu_$^BwYriTj?X!~>}<#u?T
ztNN9el2`B0h%ENkug2x8vn)f|gL9>8WI81P1neXU!zV?vr*C|86CWr;Ivy})>#e5V
zc)EZ+kFPR4a(B?3;<YCXh&9t)L3{eso6Mp&v7tK}XsffhaICZZcD%Kr9ly_&z57=k
zb1FN9ioHU?3irheiak@VA%8}M4JT>`frbmbcnaQQCoCA-t)nC%eOiaY5~(<^q|b>2
zm?9WEkiTFMcuznH(yFTHliwaEqD_W=hN7KWLq6gJidX8i(hLw~7drR=SSJ~*Tqe1l
z&kjJ$npF1e-Iv*-8rm~b>hK1A+GnVKhWNEp?$s06`aCozDLDu`hx2+ZT>!l|oTYZA
znd~Z=_oS7BqWoGy3@DKc%&a|uj-xE0S<_MP%$W=LAVO#9{Jd)FiSf;4o*7!#Q+3=b
zpM3#OxC|DOIumY+?bR4NV6K;d&8@DWU=T8%eP)wa#W?Z>0nU28CwQ4_Z;3m}xMPSm
zD2LF}M&2(=wNL&x96i?CqxRK4Ksncf4aHd;8px%ma)0@oNv^ZMW{)_81{J1f^Q2h0
zAcTbC`+Y21t;2M^(GDzxlZ@FiK%1JpP_$@6JY4WswI{yb_I)`g&D`yLskGw~M7!4#
z1eY}2mr;KN@dc|&gfm`e@t<bW*IW!<a((R}1rKxx73s9{r8#5O7O$Q5N~1yTDT#@6
zvW54f2+OO?5o9#}9SH3BM^^g3q{!CgeYs=TUkOrA*E}rukD@ME1noV;JaEp0zMtQ#
zJl8kW6`#?*Qxi`QThy?C2*kdK0K|(=1CiQGU0)A|W!?6OfJa&bv37P*1r<RBjQ9f9
z;%7OyxUI%fll6($PiFH@Nf?U!^k?@c|7<?Tsr1RI^k=W}_h^%Uy*xK3p7zyCwjf3`
z9mK|e!KbgEiQM$GG5=Q!fO$TAMI&7y^txj=y$|D$d?2GU3vK-HaT_rq$&yM~j^{^)
z$Puev-;<{;$A@p2DYA!btRcF2{fyuI77tR)RlhBxwKov~qj;KNwYv-eU;tiLbxz1g
zRgCvG9J+iYb5-yfQ@AuTuNc4Fi_fnTHYPeOoWZzkBY9om4UHubV-GlJ5C_!lZI((t
zfbn|5>l%6AG+ENLdFpQu66FL<Y2HHYGSevCZKB?p>1m0@6c<U*_EN_R;KH}CWx8<B
z-;VlbOrc%1-jOsxY?Ae~wxY^&5Bqkj*EB4MJw3>X&90RrYH&H^U#V_~Cf4<Nc`PGJ
zz5EAhjt1%!VgAo>7cdRz)^&#BD*ZSqlyb{L@zXPFRViAPo^+yN=V@{l126h?cW60k
zSCdTlr$Y5FcG7g)<50yj6^+L3=meht*boCaH$5HTv?#Sd;{hk}HE<h$7HIa12Jx~v
z-c`{xQch{o=kkblmK(yH|9bd805w6%z88J|M)|O93L%6LCi~zvA4uwW^49|F`t7&h
z_A7MiNQ0CseA}~qkIh|OCC@!E1e-;$d6$QbTrVABO>Et5luZ0`ZW&V-If)KeOB`k{
ziB9QSr$(V^+U!m8;FmZrlFzg%c1S<LH=G~vAnT4xjAUz_z(<k<iloj@x|_wi7Z(!O
zK|ZuLN*_{TcJ2G<=&lh0ePn#3>G=FoU)AVIg{TTx%&;TJM^dzjAO5JM#Hn$|MUfPg
z`T22K<@e<|8FOm;I5us!2Z)!>mW3lnib1Y3c?QXUX#D-Cz`a%*|9E1y_(}-Iy*h-Y
zvf1jJEkL46!Ez)s{4nLsp>jvdC)H!vH$v?a!4TjdDqIej+Qm)`9yD@feQEWTD_0u#
zEm$#y1U7D_j_JX|W|dUcc!V__W-Sfj=RZO?`y>~`&U12VFE>gH+c-^!Wt;3wnHR>Q
zwRZHNUzyF)MtmT6AX8|-uWI-1-A9eaT-KLAl&0Qv*^2_+<*y|LR0?u}s+!*2@x}MX
zy3%Cv)pqO)Zh3C)hP9uc{9Q_{7}j#@*1_ffoD`#&t!*A3>2v1bugXW)^`qJvS&OpV
z^l#!*x+M5D`T8{j!)=d`&&OgCbWgG3@k##JT)hzRoqfFHyuV)<*}cT6bsssgFa)>|
zpuAxBj_9PcO(jzA9WD0X);2+NQ5i1TQ|>42gub{2oWl6u7ooa6SZy}U;bE8@c6^#f
z-kxkicjr+S6-LLN-BA^Vs<Lu?&Pp^1(rTtl&mLTK!MQy2_%nR1Rdalr6<dbREy{sK
zJgn1wx~iJ$l7XdF8-3H@$lK04IfPluD*0|<Mb=lpRV=kz#NhKzawLX_sQk3dkW9+R
z^YYb#if^kcY_~V`XtWc_Ak<W~dGX2FVyYDU4>-MvW@ZGpNVw(kUnqPWF<a^^GApdX
zs5}%Kl+@xCP%L3Luw4k&3BfX(n`I448txl}c=8v{KyXu91UE>CInZl8Rw0Do`RAXn
zd%u4DdL5_$!m&mNHukp-8B2-vQ(PV>;H69WZMB2nRy&A~(inI`HT`<HNlW$-9c~yq
zN-9)#W}8{P-NvG|b_&ay*SJiMZ6o(f@ORSSJCjmYFJHd=(5kK<Bx1wz#+Oti-XsNQ
z9qzdt0_+sX4^_YXZ2Od1g)KL4$G&1+rs>7_NbOp)sVQZ?)<<a1{(}_N+~#9>^48V{
zb<b(3?a_jQ?DTiz;!?l+WkuDHxCqlnm!2MKIX>0Ks_oTUs9o81Tt~}o-BT<G0Wsk!
zabXP#Rsa(D-D>IUeZJWLioykvk{gCb;EK79o#`38GU~Zz+e#_9;F7DD_{OXB>3v+Y
z?Fx&Ex%|?LhzJXzn`s_j7nB(p6#I!iQX?Pz&$2x`>owh^=+QCW@Rs%HV4+jIxpP#g
zD%YgDGY9r{BwEZ0Zm-Vgzm&f(b?v?4+^kK({k9XyX7I{Txe+*{CTsg8pnEe-;{*&B
z3d3tm(+U5*h{F44VTrLLHWuOwPSV+5j9egWRc_7`3ReMF0ddE8QdwIdALxK1De$Pb
zZM;>zQoXF(+593@P^vQZw@{}37TRo^me4!3jgl+9K?K#F*zjHHci(;IcQ_n`g@wsJ
zefmftL?dtbk%Sc=uJEPZbj4kuF9D(s_k7v_I<QzE-}U)JX6~Lc<p3Tr?Pv+NJhyzq
z+6gCrmlAVmtdgoN?2y&huCXpHI27ZxC4>;Bi1@T~6%}`BQP!fuLiu@O=Ps`upK6O&
z6sU0N>O<GW+*wrNXWAks4y(fMLLZq0zE2zWqaeRpF1^p5k9EJj7>S`OE4Nqk?46f0
zV&v$S=8MDOAS7h()<u`gh0SWEZ=b$oFP}rGJTT|V3h6}{Y8Lf#Kz=rPUik$>`iF}7
zYpP^Kh$3zq8kzd#-15}Ia*dO^nyIRmXj1>RG&gHgL{irQ3VL5%^Ditf*u4YViDZj-
zb%^{H#CK2p!j(uKM(jl_yaqKlrRPA$963@b?z1>i)3^}9PyqvlviAg?cI+LxuJ2J*
z^%QLJU~+)J#t`h+3?FhU%K)9;r<$+lwrw$r)BCiMVbBvqB*W6`vbc3!->oRhk;@vJ
zLB^yRTe{qM`9F0bCyQfe-E#upXi|Q#`?ynA$>*MHk9}-oY{*haB}D4iw1%#vlzZ1)
zP!EBAq||+Q3N`?&nLBrGL0ga4mJmXSl+-RWE;}vs;%cvC*0R(Ogb@FVOiaxxEceI9
zN1C|e%rI7LuhxHFUiq5er+oU(^wM*rBKK6}WzKHcS8;CErs$-!WqYJ{S_jdxQ8-^}
ztd>5!NTglC>j$gVvQM%anznDo3|-f;+iivoDOur*oi4(q;4;O&PbP8=SSEcQO*ID)
z$&$RxNfC)@=^y=6dXg&jg;FQ-<;nh!@4GU(r?c8vQnHJEAjphLN_*~<Zno!pcd{~L
zsk1sFsq3Kf+?~shl?AnpA?2$>m<~*T>SvLv32qXCtAT#BLc-EBb47wC>5tCo63*fb
zPD97hR?;zlMfm;xbX8T2WSh`bD66r~OTg`i5DWM;D0dq~N>w1#0-^iw+5m&(MTY5=
zIol?mEjnCdK(B)!k`k!$NGd%7O&Zl~B5g)F$lzR>@a}cLL`~Du0jyT*QAZ}zC(ryg
z^`=WZfq0#xJJE29lzglR^^5dL-ycYc)OH*oZXU5>{kjSLrYFbHNME=$amfQyJ0$hH
zyjxq20pKgmM01xRg_wV|XYq$6r}iA!*KzTX{)U%lPm%AQyvS}tvWxw`!g7s-C=)T^
zDj)t>`gUnf##1$?ZELPSUCQx`bkAuws4d4Lexz4%)uj<6Mwv*8F)?|51*&SoW;322
zD^{&x?fMPG#l~PVnOMDMJ&!;90A{m^@`?%yi;76?T$k^ItGb%qyR)gNs9-?<epszm
z+-^5kiv?Xu?z}t-uU|50UOAgL`RFTJPUKuKeUwRkT1C#*M!MJKc^Q8If7HH@OzeEg
ztYwv3{&7i^8K9_KlhGmSxUx!p!0kgL39%`7{nGx;@8!1u?L=~a@$yA76T{X0(i1a8
z7&cCqk^u<%D9}}v#67%JI3v!$^E}As7ofT-uxp`23qOg9QQtca!&TlJBfC1{uwBq!
z1a7x`i`7~~3Zsv&Cz1eFUU0f0yS#CtiOmchA|Pu20-G&n$u;LV`Q*pF|7jIyl+h<>
zIhp+0!IhZQ16=n@T##KRar*<eDvE%V2yg`v$*6h(0KebA`M?G}GLSWS)|k{AFEarz
zgFJf3q{M^5&qz5g#2+$m!;~pUooMv2f^}Jpii_atwv3gG>T<#$p|q5ackrk!izqei
z*R~$3?O`d(FZFw*#Ol{2MdccQzc8|ES(R4ja9W??K??arbV}EO`2DgZKWi~SSYo$$
z6WUFsc^Plj_7<Z0CNI~;t5;x;akP-9myF6Eo2!WkQRv^pjv_!+6&!Y>&d68ad`I_g
zT^Tw0HBLOCFT3|-^UM<u<MDdQ$lS$`KmEdhQ~CjrmtVjgci%@!auUBSTEf<~%Wyc1
zeoQ74x(;72C}($u4_mn;RqWfdX}<13QS^5k9*am$8!DuTD9cTs0uY+eCBdX{Ygt~_
z>s@+uTLGK_F!)4=IQ2ODsUs&8vNOla30-XD6#GGnop9{9cn%9*y+S%M0+j25l%M!T
z<eQzuyMBSk0T+M`)MHnc>Zm|rVd1)nh#FEC(;w2muKzH*tofuHw-0s{K|&aGY7kym
z>4AbusPY)B&u)geP>8HAP;~Rq3cmX-jPlCH&w~(PxP+@Dd@JDNnsv$ne56BZ&9OC2
zOer!<=jI_5jXx-@hN4Qtq9}k6E5wJ^TU8~nyHt|z6f~8=tV`HkM!;vbSO`n2ds|*!
zz8=8i@f?-*aw(-0LcHGj#!D*&jDaIPWJG3ym;mNaGrpXbdAMiO4j4XS)!KCv`b}#~
ztE6fPKQv_#^VR^k6h&EbD8_3`RAg^0icU<s{rN8nNBRBIom=XC?zJh!=Uy`~?B4!8
zjCI-TlSrTHRIK`tQ;rK68J*Jkywa$!<w;igV_c*;9;hRd+1c5y$iyxmF4R3w4^{Rq
zwRWc(LV1=>q12gOUP)4{+0ZVoaVtLa_#;F`Me@U}*<5h$AZ{FH{B#yC`2(NN$Lu-t
zc>dWZ0hlv)KK)KQkx|b-!=KAnAcVkbv7#snilU(Fuw$!-huxw3T#zD5YV4H$KTC6W
z1^k`Et;$JBF{YOyl2V<~HZeab(lk3Y!4wI+?$cyltaaa18?Q_*T3+RmZ@&9|$>VX6
zrntNk-?XyW&<~)UNVek`L__%>kH;hYe!p0-U_nN=Zr!|qg@mBw-F?a2&UYF_taB9X
zDTCY~BIof#Sv5pEAT0(IL1d`T!`GMc+{b~5^?iU>b*N4I`G<u00@8u|B=iE1aE64h
z1dI^`2-rPxLkVGy26L8Dva?uF8YBSV_Cs+MBu5yQ#$6?ng3A3PfN`fogBT%?$5X27
zx_~3UvQ2ZSneY2uZn|{4M0^UQ9PT*<!7E^n6mmQk*X)c>=O0iE!^)3VIy@DgP?Nn{
zi`F|=cDujZ0mp5(BM_^ylo#TSSjmI28a8<^3Ncc-ZTv&$97MnS1gUxP6l_9DxpK~&
zIpr;%apa>UFJm$U)N1-RIw`HPL#+AnzG+rs!c`PiPL2sPeR=DU2+Ou?&&T$v>aQRE
zTXg?zd%TUjEq=+{^L5XIgB5FtwFRqW1Pb9oVUFfQRg6R$pVx34L`6jc;H<1fQIt9Y
z`1|6e48L(06TbY4g2E!=<6=3re?Ojiel*chk=%dpKLHSmz+y3CE_Eq!Aofek>}TDL
z)OjU%tq!r{er#B*^5GAq<5q2VeSP-@VJDVWYTKidQXlCMXC4Fw`5yN}$=1qp^ZDh4
zTb#gG!Tq-r$#xu55JCtkB^x(x6i%m80IXTFW=M2&^n27=pJGn|OE`F&u<Dr<hz~P}
zVTp?XiO>S4VeND7<+)zsGIpU|4Dxu{S(U8bZe{xKM%+$cV4O}sVRbX`vJMLb3~Ml)
z0B)2J35*a1kG<|(CzqU26*y*kF^x!ML*$ejb9M$<5fx#$?5(O{U2Td`CkGbR=-j#U
z-1PMH58is~t!KaZ;*09S!a^}}<Vf+{bI-w1R@XXx(vNex+;rtBQuxOr#V`)P&Sa*P
ze1Tc{H1ms@nfpV~SAMiILRPs3D++@^IR#`lBw;o)Cj3@eA@Y4I$L19y<z}IX<%%wU
z)f|pxr~LDjqefi*NWyJKtmM?Tuu5M2)PQcP$)4KU{RYERxEZN=kWwBqvOl#5Pf0t@
zuJE2;XQ5vapZvU{@~*yV*68$HF9UkoQdNbncB{h4-R(UcRxu(i(Za?|PnYlx9b(MY
zcAsX}t}n|=A77cDo*$LeW!!JN=aC^TSS4$r)Egr4`vVhcq%Nthu77@P+_V{o!$E9p
z3^RV1#a*}GO3xnM`Elx3{Oh5|m@(^TrhWAVAt4UTCKF(&R-5;uv~QNwSg-rmloe!t
z4-lEsWr#^=TG5`Yxn@;4OI6_d!Qq{@?(sa<ue+V)TdF&RJJb$6QmpEpLLZs=KJrWb
zuPY}1EnRway{fcA`?EAJb40*3@AGWh(T*b=wIp)n$dRJ3uu!<&Zn1jx>f2&sV!jhX
z7*;2#L)9LLI=<;nQX&oOU6&Z>lV~KNG{oLWWG#xT0*Kwyx0G=4=~aXQuLY6FT66>M
zl1($2izWO`8l?2`n@a*eEOALnTrICE&LCry)nr^>ml)%klbea<v?qcyy{@#R5JI(D
zt@l0p=%dr0e){Q<TB~GD>%&dN_8aMwzR%C>G3zF&D3=0j55p9Ez-&;i##+@YbMg=W
z$(%4FV;@PRSo!wifvetHI6+ibXDW)A4&ot@=K#q<5Qgz{b0B4efG$FaOOZUU3$akC
zc5hoXcJb&x-dd2_mKS)eVqF%ad|K;QyRj$s9?T($k2C^jbX`Ac>uWa$si`IXy4wf7
z_K$?;hMweDy*=AQW}c6jaFxEP)=0ZmxdXsv5ez;tM0kCY&Ix9#s+qpN@C@6kGkV+J
z6v9v=_oKc#SNBxxYekv5YP?dS2Nk@KNIV{6EiSL9ATPgwRjb$1AvuZffA|T2>gsBG
z_v*<nKTO4Lv+?=F$ta47qN+#<iUQhxY}se%)ga`vwQGu5$cz7db=;h^?%k#*MVqd3
zdnGa9DkpWd0g{wBlRY-VbkD`7hO%I-i*0+n^z3MTb;R)a_wE=Pbs>r<2_B=JNVemc
zK`n`_wMH&nxbVV|kdXHP6(nr^2)^lRyz&<}-HEk9{Jh-^J>m^3Wm7P#&@<j3WcHw}
zx$JsRk*Z2w_(u`rt}P?MbO1&>v*e*+r96K}abuODoFI>s!(z;*=RVBa)?m(-X7UCW
z+-Q7X-*$&>pHORbR8@8GuwlbKj*pL5YORuBkoo<;z{3mqoYXTX&iE-NEM<TY3=@*Y
zw7qpT8_BDR5PdQy|9El6q#4uq)rGzC)n6l4ja~SURbv;g6jJ_%5Vv3$uJ;E*B2z&=
zZ4su8E8kl9_PYNriQn(5_zPqBh*fKg%0;W!;5J)A&u#5Kn?g;ZO{tV}<)IkwP#`R^
zTYPj<>c5<MI|^3s^n7Yk1vj1>VZVNGMD(~HN;sj5jjl-+H7-K61C&>4M2D$t$aIq&
zW9Fo;_VfQXD7<T!L$#BdY5{QOZZE8o@|9V7%eSYFQT|X7#`7q6Kd7oIo*H-eZwnXm
zkK1qMv;n8^{@6EIvu*>GRaJaG@hd+3_%m!)E2j=Ph0xGYq|}YRq{O5OyunGK&0vG{
zu|;ZgOY^d3)gH5%mGEAjtm4j#Bd?5)GKng;PI|7Fv_vx*HDtSI2a7d2OeLq-N0nRW
z;p<{zIwzP}w6Ur{@t8i2Nk~02I=S=k+CJNfWGh2L2C)68eM$DYzW0%YS^~3i<3>?f
zSSUu19vv1H74@MIH35EqJP3>LhRV#ncjOMrin5uYOKj7&wv-46w;ANETWp|aShdf{
z2Cso4;J8c6VN@=hRR!jKSP0A#23EqT94r@CG!Cq;^5|6i>W?bIn1{n+%qyh{ZXqzN
zkinK#kALqIa;C$gJ7M{gb%#tQ)79U6^UV#lR>_egM~b5?TrvPFKK#&^{^bwfWE?l(
z9HEPoLA(NNq4h{*Ht-3^i?LR9%b5JblRLkhxqV+<*el*%(0A3?h3^QPeJ6;IK%PwN
ztl<tJ#QlD=zIN5yi-+xJ_|;z;Ycu6&zs3QkQZ`nV?tk}3W>m|dgPm7P&^;CSO45;1
zE}TFAz+!NZM3_}y_p|zjyqnmm%hw*S{^#l)#-*KiQV5ZuDxKoZB*mD>F7o5`Nvsxy
z(n<|ywa)Y<72JG&1UquPm{pXxs*t)xGQ!T7g}P^FP7Ad~6pT<MUDAj|ZrHe)&6~Hf
z^p9nfl$J4a^sDH)q{>x|#bU;0vvSux574t`ceZTZ&b2q*N^x-sH{Wm_KA#U=(~wf4
zDgu8C%Qb)x$#<IDFLWMmxW-s4SoRe8=n!kB!lmQaB<s>Wq$XI%DfTmWRTXFSHZsBE
zqfA^gFbq`)JU+?V9qvWC#s5gZ<Ltlo>|h^S+h==>WGmyfA+o8GitT7))LJ9KNCFu&
zXwW~YO`^#71~$F3Zyn`3jqK}&F(?r>NC=0#pqR=A8QR#)5M?)FBx_ONG!kzbBGCu`
zD1w_6*l2@vD-@cbs)i4Q=n&_Jt{&*)2K!zTf%1bAXQCW|1ToSMSwZI{g&QQZAvE2X
z>%L(9>yR)II$aKc!{HdwrAwD7g@uK_jT<)t$eIv?qn1(D2z##u)}DLrxg&NZho)(|
zJV(H(2#yC*fq2-f{7V^#T_86i#Gg8Hk)>4Muzl97{YrS*G;)*09Z`B2$a@rF8Vs1)
z&{&#?l#^GDU36ytJI*IAzW?I(<eU~Te8j4?>nHY`o)pzOv67vV`}Q<hg3E8UKHLQx
zkh%{Y=C}1&hZ34k&p)G=oy$)T8@6Vrdsu#{!SNTacOe8=OoAnwT%6g*&L^|W=on{W
zcY%+LTpt#bAiva)%cEmb1)Aor+oWY=WVoY~QXlwAum0g7HMCi`i^SL-xIH!Nc7uj}
z`}86^eG@_m6h&dh;`s<6m^tfbA|k@szJ3K#N`w%cd(Oase#<K=&~zPL>gc+}ybpW5
z*}8`U^tVg%)7LdTZZRoJQnZP0f3M(zej&_Q>ZIBuxu9PN-z};jK2l{*p^xZrm6&jq
z`D?1U@~m)n75KRDl(1_~=wiL5OOl0GzbeYE?X#Uow&R#Vtu?Z|yj-zbtv47)C9EA4
z*iaQ<GfUh)(J0u&<A=<Wx@4I;q(ne@b-h@(5RF~Ou_1v(ve*Sl5r$PW7!Dl<xQy3<
z!0m^Us=(DGHYBi(NipVG%jq1Jreh&9OLRP1W6PGR7;~@oHPF|Tu;#@&A}NIE`rdo*
z4Y>H?i<id5#mO2nf7Fu5V9cLC-%q{8{$nA8=<~O~nf+lNtJ^7K6_YPaQ$!emq6xn=
ziwdh>DfC*Mm6@pvyz)7DK-Ff}j#-+btNI->Qr-o0YVm~DgM#G21R0*T?!Uj>bi$)U
zs#-qfuZ^`CjEW888nfx(t;6=H<n#Ag)8!!Gsfmx2ns!Vj+Ef(o*||MCYUgpCtX(hY
z7s9u{8=L=r-EGWWZTys%Z>{FRt7FK_^HEu?Q&g_8WRr`hZj5E-(n{{RB8uGNN)#nf
zsJA3Hb5>+h+SkA9o*M=!frTonkgU+X_>h>)0--crHda;Z_nx&;Kqiw3m#doSs3@9?
z3=%@%_iOn58d8EU62=u&5rN=R)kb*GPcK_){HnV%&00vKLf>W;-J}ZqI;e`EuuQ}4
z(<!gg&@{=gK@t44tdgV{KW0_XG2Tp=T_H2i%aHydb!8>DWqazX0ksp!b{zE(LWt*|
zf4<Hd`N9h?gb5)G6>V28l&o*q7<wgw-7L_hkr<Ox5m;ZkxXQ3FCWqG_Q-on%+?-dl
z?Udlt17iy#?BFoh3wldwXjFEMZT*YRE|bUx+tONhc0>3z9a$r~Ed-=RH!j(gU9M5$
z3bd?QO@`$%Cd9aImk&1O)f3X<mEbM}TO0tPp`oX{-R?gM3kwBx>N>|JYA()O^V&x<
z?mk1T9{c;*pxh%Bu0%GOmvx9Bukf23lYe^Wr@ysZCijjSUCI5=;BJC5m@J`fuOBHT
z?!2`~Df8#gpTDW)6OJB8DP@<Q-9H(8Qpo61r$%A9pK6azh)p3e+Mqj{24$5R7Yzty
z(!6s1+*(a!s7hprN_MdivnmYhrNvq2$F2>L3Qg?x!1ucQ++M<*5HIS<+ZURPu4mfl
zoM7h6STi%1I_n-2Ex~TL1rW@pU{X!!x`yBHN0;#UO|g_$>C9VG#cwOTT;3p~+z(99
zT)J24ca;}ztP-NXS;`f4>I{k^`~Xp*3V&>NadI~svzApdX<j+^UlonjEXd6FGO(`$
zmq({xcf-wHP^Pgy!^3IE*_pS-wY4lS>$kc-TfWS8wBzVNZKn2#C!SdEc61jKU?8)b
zWL;v6A`Cb7@ks`8YPhgxml={c4H057q=9PDl;lpZ#jnH#m7dmRR!(>0Ozo5=ERChr
z_2VXl8Dz5|@|_wv<qfyU#{Byx1-kbg=0KOZ)WKa}WlK?%B#*}<f-<<<v9B13vU2RA
zzpWg*XcegQCEP&M$^C~Gg3A-Mus1jgv+)l#HtrEG`Zcj%8Qj5N>abzMRMliXyA`L^
zB>b*Id=+~D<{pc?z7n_hgDDFt%3>o_wr6|E%r`hxRH#CgTPG?^W!@UYI(fwzVeBsO
z(WkSObs28dnl(?;(Q+IP4InZx?fU+w^eE_ma*qN<`p@@Dao1;hRY0_4ID)KvA3L(W
zI9)n6i{ZX5Dk{R~+q=S+K(yFq^VV%hDe?RLC<2ymbu(>I1)1AD^c8{jz?phAd!+WZ
zP(?}aKHavTeMV(@baL8DwV!+a(g%i>b5oXwKB-n*9-RR_?bxjf*+o9K?Di5Fs!-w5
ziH$HJC2Zd1!KA?I9c~t_t9n(+dT&d+MY0`7JsJv-19-jO`n%Ge2tuiouc6v6aCs#T
zOJI^phe#vFFukZg;U=pDR3V88sc)MSVUR<gW{8r<gao>{qZ~#hnaT!6oQx7dpTxkp
z%L9cjC7C6XhR>leYhb)4fTru@Rv2w8YGa~VSP8pI>q(@|3|(UylT84ZSI6zEzeu5i
zJ)vO>pWp94ddbGfyB)Zz`XKc*z$jlNlrnUgL8WgK?@g)Pdj|)ORqrqDMx=k7K%BDz
z0&D;PAOJ~3K~#s4N!?aM9T0>N_pE+v(c+W;J#cc%t^c~X@#&T8)=%s=Z9iir1MV69
z_y_tc7Dt<x%k|{11F56y90O$$8>1{Mt28<}^}q3v>ZsutL^5x66}>uH>62<>d$yO9
zSQFD1J2|<VjkH7yeN(M;iZ@eOu5oHl8=H4|NYC{VS?=#8gjnDIlpgOrd3|g&z{``1
z-e0+TTdYj%{O<R9_5If<#=3ix>Yyw`r#wS4BfEl*9V|?r@go-vxq!dlbuU5)ygn~T
zu-dF3!Rz&6ci367^iRfrI)P1_w<3g*=~n9+H>oD1l!S+e5gs1S(m$7zXc8>_qmtW;
z?AWjd{E9Y6A9Hoj>M}b|C8JD#>c-fv4y(eKbIYHKO-}uy*eWs-brHRGryG+BX^9q+
zV@#M-7|_#>rb|AbQ%-!OiXsFSlRy!$Y)dt1iDn!&h4*KcZZEL~h`zTI$#xtYsFnQj
zcs$~hPd>>XJ9ex~2;sn<05K;+QH{cdl(4-}&^@j(X$rAG??fX;vNqN+qu97jV`}0L
zRRuam8Y1964R#bl_c#J{c{)YayTHLHv$(rdr*rhddjNKog4^GK)Dc*^5kS)=+Y1Fg
z&6s<~s6Y~1ROu(Pr2gKI34zYhM(k)XTwa}Rg@MQ7P6KOn-CMWYozd{)8iDd7ixCB5
z<lPS3mBWzS2n+=50e*ieK_Ce;yLs~FB}gWtt6%Z@gjS0gTs?NlwV?kG#QsY695P6x
zf?({*w-zoq<=<!L9`5-Z$=I+<ycicQuQI6z_FT0`rk&q%tAmF-cQt_2b^kGyXj6-*
zq_mNp6D&_y%z{#<Mtqcs@v};?m;^pea$&zvhW3~Iw5*bQuZX5&yg`r>qD(B<62hsy
zhx6#rj;5JE{=D*=snhS)b?V8zFf#4SLd}oOt@NChDom&L6NZN?QYb_M!UGnw;G3z_
zxoOySTz1J&>~<SMC@8{+pHzf`Pz34enar8<Gc$jf#j{U6j;bn{Os1NuI6nNE;m!u|
z`~5uf#IuaJISQ>pVhgfq7fGEjbXS?Csl#N(4tK}2E&(n4Qua9|xyy{Z@4Bs<NmW@r
z?OPH;(=n-nq-ZlCHiab{U7X&_PJXGM^_gz6i~PJdqlCB!6Ma%`lvQe+(aYX_?&_*d
zVTs*NcjoOVfOaC;j$;r51`MzPZj92^Mw!2&22yEW4L0W+1S>u?@MKjnVj4p#j8cda
z$S5&5Mzql#F(HQAyV7F_eQ(Y;^d@WLC2L*a`3*wo3Mv&G7M-}z11FP(m6F1sOznj5
zKtk;CYiui2aM#EJg<0$84N&6J$t*D@Duh9b6T)Z$l6<Gmo-zeJNTwvdX<-7l+bwG2
zB=`5y4jH}~rn1s>7l{7=$t|=N7%t}X5uA^zy~vZpM^hT|5g&edP!(ZTja_^n$T1up
zYjB)U#Ph)2hkGJNGHR@n2b{sJm~CyZA1QU*`RkBUZe6fo!LBwR@9<!LcA-Dn+fDv`
z$SI-b+U;R>v7Z@BoXlBSiNmJQG0seBrN&8JZTS3>UpGhb`RAYU`NXg2)2k<g&puQ6
zcG~nYHj6Ngo>ZtSiYQOd_Wi5IrR#}F>2}>j&9yvU4Nd6Kpp2}~CpqmnJJDf<WUu;w
zLbJeJW7+fhB_5wd*TLhHOkGsL^S8t^dqo*DCOk`rI8)7@zjhfl;x$bw*enX8Zi~ij
z5>$Eg086AB$%k5{UM00#$_vtW091PQr(T*|)IB;(O_Lx5%17hgej}=Pub#a4%76Lj
zgK>;~Zydj_n#OHIjlFDoo{wIgEbPjc{P2g9E6xa`Z>p_si@jx+hpF40^QGj&%2-=j
z?f6H-BG!&}>=&-R_S*jwLKr5qtZA_2{l<-SX}GRSU6M{wqLH9@Vz}Y%O)oOa2&5Mq
zBr~EWsJUaLK_J1%YECbb9G6gkYj=o*a<^fPtVLF-f=Tfc<=FrGuEZtDDh(ubc5`Ew
z_0A$uM<N9VS*}G{wMIs<f<)92=7=WBepUN4vP!_|X*vRY<6Dq)E_A#Mfbj6}$Im<O
zyorDQ`DdL!|A>?TdSZA7gn0sRpbzXMkna-hyq>3Syp@+HXC6o=E5<ImRe__0L>iFX
zy7HYRubp!LnL7{nRE}h9%4SqtI9Hq0y-Gs?AZ7gqZRoB44M`bvZ<|ZU`O0^rxrzYh
zH;i#RmQb3Tu>e>QncV5a%4&Vu4TB>(6jx~Em-zU<%c5AD?&gbM%eeJ|$hvAb3w9*&
z#e`3Y2oGn<chh;~!GDppYc~?%ijP!77p(FAyDUHBeJSfhBa5?l?2buDz2-wvHTO|9
z)D}^bauQMwW4<lMViEZKl4_5{?~{aC6|6cqJvzDsMFYQ->@M`-^5`tzSdGIXuqpzt
zl=yTII-pBT2uy%qgHS0rw}DGr2R_wYWTQV-mY4NI-S?24xmpO(Eg~`X@te<!cx7Xz
zC*zx`Ga_Gl;aQ<53Omv>c<F`zF#6RHHm>}F)C4n0(I$RgUdeGCtvq>QEPD$5tlVBr
z{~mVM>~w1%&nkT=XII9%!E5``vye4|{i&pJ#68&B(JH9bj|7N|i&N}&`>lWx7rFfL
zy{FDqxCI;YATi8HG->Dpj<myZ39z-$xPx~VL9YaZz=c{NHWZ3#!slE*LBq(!T5<5r
zdG+<Iq=cQtCVDD5VYYp@Ojfudqd4$7RK@7O;e4f(xEsVO#)TTzQGjZnPI{4HT{NqP
z1=SHmo;6*PQ=wDfY;1*Wh}!S9<E0ooW|8{ROD_!?IB;N{l*kcxt37`6NhtCMs410s
zFp%PG5KDOC#w&Pr@{0XAb=}xSrwJIx(YN;6kovvABZqq;M>4K|V%6FMh?OL=@5QGo
zW@~86Ee{U0Wf3))+ty<pZj|NhS`nL^I>c%g%dQz19=bhS^3ji_+&(mtC$5jBs9ZzS
zCBLP2X5xhZ5gB1@%AbAW5&m}b?I?=E4cA>AUsY8_?4A|33!(1>W;Gn6I6reqbYj{a
zAIPeS|2BmxHUzh-4%|*jwUpR}KnJV{gexe!Sv6#X7S43h8-<&daFm*@c<t)hXb~2T
z4^GisY}CEu%k#5d4{jcwm^Q#_F<rT<sxw0_jP0{-!v=1>^Byi5ay|pkH1;jeKKUqb
zzU^Sz{Bc~^FQl$q;=DCgWajy}^}+~t=6d<x&t)TWveKIrWIoy!vqnfsO#DA7L=z^b
zCfX(^>p%sNju3z7x;`a0H}|)OZBpanl6+?K?j}y3mz|R{q=o*c1D6FI-kzP4)2rt5
zpMd~*2MM!ta&j(h(1(Y;3mw<WkCXrz5>`klzt79d`@K1$oRXZp4d@bZP>6Ba+1dB)
z?Ro%GQc~6-W$$J_-jJP>b5cuvv?Xe!f9f)}Cr+H$Q3zpNYv~54*iqkD5k@SaMK!9<
z6jvG6L@ABLnPL}os!6)3g~I~9;$eG{VO`8BF~}uAY7~T8j6R|rjqiH9*{F)Mqo|$)
zN(noP1bq^9EX~z0YygknAg~RIoe)N1j4g-=3L)qbXT*2f%?8n}MMg1b(y+XCk851n
zro^SOt5o3A>VJE7v*E@Mu^98nDy_ecvdvJw6(WuUAUr&L5HQE-bRKz0H&5Kq3q_p`
zqBTnZCLxrccyjo8yfSfp%gtfI`wK$DF?}MAgpBM1fRNm<<DH+L?$#~`v{!7&P+o`+
z*RN64eOE=Xghn6AvWOnP$MgH481HbSC?|7sQl~Cs&N)70#2LLEm`#FrzAxdytD`9@
z*VtU%hl!t$BP67L&A;xNE4lXSE73GRpH7&>AIp|=)AiS#?DPA7OiJrHA<v}$Q*Phx
z0ps}-mE39fZYdx8P^)_5UNr<W0viI`KG?WI5L~1*_6qImB05udvru|xAefw-oNNc`
zq(q{VyFPX2tv8K2@%TPUS$R3#yLGh=IsaTtrpC1I;Ro*J!w?5QPJ4r4gTk4&ri%1j
zFLzuNg%B|7&&rJ4%ubDnYAyb1@;?ejP1=4)d9NbGf)2^a%Q___bggZWSy-40tivF4
z4PS%O($bpn0~+YxzrPt65`3P7Z-d4U_*^vp8w8IV{4!z0{&fJ(7VwZF#De7H<h3a&
zDJM5MIivV|a6c{nR+E#H`y=J?dzs6L9XfP4tVA-HJIBPtbfxy%g3BR>+YHg}aGN3b
z(>Dov#MN8vynci9Z7($9C~ILe8}+L@L>iXJT8J9?piWVN#~UKjZit%)L_O<*^hR~>
zn?_~{_#1ejgfRMxXoib~T8;U{g$9lpW>x4AX(SUhB#~af#<oI@okc3X8d364t6{P1
zn^a#3CoUAigRaq;S@&Oy#Zq?<9Z?cMPY#Pfs8ca2(X=*VK~ClAzsI!P6vF)Ghk)ib
zWk&;&OZ}nkmAhKN-$txhU+fg$?aw)iYHf4<NU8hr6l_3Bx&HUxe=ltLjAH{X>3i#w
z*`*t++&cYx*g30@gB9D{Y_IOe=O2#`Ad&TuJoxBSL_~!1{@DL=;rZvXWy?0+cxCjh
z(bf7ZwXLEPJM}Ut>Oyp)wn^WM<F%>-t(Upq_(gZ~v+kZJkiS^X%3POL{~L-(P8)jH
zZNo?B7ZfO7pN~OjpG9F|F~e`ao6@qfCi}bl&fB>B<`GPs=On_RaQo0m1TbT9g=bZ|
z`yMIhH+yM+H6qyv5(a3hx-2m<vEE`Q#FQqQTHJ1Tv%ALJ+}y!5VQElxy{W!hA5tEV
zlz(*S(4o~b|AdtHv^cD;-`iT_9SDLW8x%#U{~MTtl1SYC#%UlCc7t4XiK$PPDXWGJ
zxrW8Eu47@4styN`U>og#4ndjKPPal?_1<HYN?nFqyCFJ81}4w!D>_CQ^Ka-1FRs$q
zkS8dwHZI&^GU6tCB{Ytml(KOha&u^dkEQ*{lmN243LMvl19U_8H@j1>#p-t!#tA7O
zYQuRP5||X_yu&?*qXC<98KvnhatZ-(^+VGmRg0s|H{{;pZRp-Aq!7ncHJg@DzI%7c
z#$DQZFHI_1vUHORO_x+S&f$OK-p6Wf+5qs@yW=?d#1j~H?N#WyWX|0AAOxokIEAy%
zIO9$s4Dm)yQ9ImtZaBR<Tc1HHD>h00YvZ*ld~KX>_>1o5d)>WI(dDx#vF88&^U^3G
z6<L=&s7lVe?2@6PxVVJe+<cM}6Z!k_8~D%DkMq#u&!Fi|KYE7WIE>+Uk78Y>pKW`*
zyfwAB==-G=mzC#a%x?088vNynWDTNKRh?A(M$@z=$)BPq!EWn@xT=X`S-bP{_Q~zl
zO0Y;>pOzXIchI$36B83d5aPz5!^JeFVmGIxq#OXVX&p!@>nva%k4Ly%uKJyJ%*miO
zAVhAT;o{YsxXEKe;P~VkVRB*ApH44^?DE!G5fZ~1Ydz*x?7i>2pu#Cpj^;9~_aD2<
zbkd7eG+jZd5h6c6scDj`OCxjL)8M&iazJ$bvrpIcLciZH+-~=gLnfcOr61^b9)gi>
z;ql>T?X7{-mD{M*s62`xkpm9*9F7KD`{>Gb#T9buzO=O^b~{T^&21j;P#Yhq`~~78
zgW=5EQLtrm#NdyADx0FmUBdfg-^AR6ix*(l&vVGl&Eu}yZ=u@l=8n7XWAHg=bKn2n
z#k3hevNJt>dsJfT=%}Q$zoTh-W}c6GE{_)RQOY=}F?oaZO&zCI;b|i>`Gf9ehVI^^
z3tqB2lri^S5pByU_Jc0`5s7IxMkS_>*M&aiwKw0cyymLQx#Xe?xaa<d*|K#zy?gcK
zic2rzmDk5K+yAv!UBTTCz0R9cOV_Q*)OwfYWwod`dMp#kFA?H%g!o*-1OXEzEWn5_
zYz!$+OG!x?2#}YTmkw-hqRpj^j@5s5MH9zb#TqGLlu(qDYhGsxDgO>kqiLCyFrV4{
z!oiNID9TL)Sy%;p1S|<^>Huy!*zpb)wXu0KX3W?G7^Xx^By=4f*s!<?Hs(QTb>nZZ
zR+zkRV!i7)r(96wZJiaNV^pBKxzeNJ^*6mW^!Nis&1<E4S_73HNnV9<??>99Pm+=S
z9W0nGG2$(Qt+N2#Zws(Uy4~)TCX?yN%COP(CqW#FIB*lHe6rb3N1`D0TMofI4h6a`
z7`fo6kF7rx$jCLlsA;q)6J-gHKO|P(8vRIDNlD4lLowcw1}Rtgx_9e(+?%hCLQ$Hg
zJ{vb};mfbT;iVTwpzAshKJq00`oFvA+9j16ZoO;TsMlT_u%RUPtS-ss_y?|v{wzAo
z{ANtJiq9{(^}@&`O#G3WU_N6jzGug3m9nxG$)pZ^rMdV?ch5{sH0Oj`lo?l@6@Hr8
zB#4hx3AKyI9=Ixc(#hRy$z?fND}P_MDE0AYo_}&@dM5w-;BB6M?gg@T?dF;*FURBc
z^21Nf8yPRVcqmt1dPy&#w%|3Z!7)oDHC?|uJ3ITf?Ck8@a&mJ1z9%Q=mYkfNb2VM>
zh!B4^b&ino+WI%*n<kocPE1Tp3qF{foZJf|W-$O#h_4Ug5*lLno;`ck)V!AN$<3XR
zos)BoghAAjra;`>y?gg&g7)?e0vfrs{kr}?g!rUUV<``tdpp>*{mLt^xHL_x&!lZi
z;lk}Rgv2-GLV;5f%;jumBX+V!ykP}2wMI!Az-EH5AeV7f)BBnL5(!3}VsL`Yfx@f`
zq8-pP-blQ$1@-IKB?V5&#$3tHqCiVzEo5Ldxa{ogg{?EWR>BiECL<dqzO@}B7x2`L
zsZD*fENoFXgy?n1=F)}`5SQBd=ugr-6u9)!<r_*W*-NaXKnNielXY-g&qeoE;Vsz?
zuz1Ca6;0H3{c9j4IVq_*p_iiKVn&P{&AV^DhRI~&xlylh!Fhujc;@MZhK7>Xxl=zO
zl?NiLc_}@|*RykinXAqUJN~#%R+ewACN0sz;1e9y2d{}q>Yi-P&yw1mZ}=;H`EtK1
zXEq5BlJQy<3#IpyD7zSS%LS2x&N;zhKIeD`OE$Xb(ZS03Cx`UtlWJvsrZ*xwG3`c0
zGe2gx+0KoPi6%Naig(An#=jqXn&OfY{`1shd^Ke%+jcas=GCcV2Q$(fw=eTMW{G5T
z$ji$s@cDdKAw;8{qJ;jn|JR#|ZB$jY(Zfjyab=^1Qs(65=B_%(>FyOdIXMdv;@?et
zWUsEQJo7;NO-xMe3j`+a2=QlLUS7J-=bHiqXKkL`F(KiE10An5(YtqVMNyPK0BF^a
z5ya&q?S}O&(q5a1ELfXuxUqw4Vn^5viJjIWa8RGvQwxv3=`E`_Xz>YcqIBVb!depx
z86HGxoo-2bkz{SQAgfg1_KUjjD>%7l=OobUNpO68eBaiaU@JgTFXGT<N86Cf#Z6Cz
z(>VqdG5{n*X1b1zYdI9i%;iPDMsrL;TzvnOSjB8Rv90IfDcX$md(d_Ln5<^Auh_D6
z`xdXa@#fv@^>W|8ALr#4pCc+NlK*`=fr#)hhFxn!tVKsfvSh*R&bzj4xMjtXc@N!n
z`>kjE@Mjh4Gu+rMf)l&iShlsA4sm7_0au<G_P)&`oOwdK{7tR0e2YBr6r2*NGsa)3
zufsb+NNvnz1H)&AI}~$bl!+ZVUOL8`v6=;2cYDY!_7fi$f9Bg`UY)e;_g|k_zhd#w
z^UoawKyp$NFO3|*y$?Q$-|uJ4>o4)xQ~$wP*|-9%uInsVxH#9D6SGgcv41%t2~bc_
zkcX7>n)s+=?SFD}b2kE+jao?Al$E0~krd+FL!n!llauo~4O0Fapx=S^sj6yINjCoj
zC@3hXM2H!|A8JjXb4X#eTIC~;JQ6L0hy|cB3;YcXwic6NflH1s{yibNKIX8v3O45p
zHswhw+->)FYs!7x6oP*``!1Kd|EQ?e*_fxZr9e>ZG6qTrgA<bjzb|tGRVtqos`k{q
z4G9V9)^-y-Ae?i^M-<}hrVcBTK8JcL?Km=U@gu7?Rk&ocb;_f=^dBS?Q;Xu|2a6iF
zHoC6=($-@g$;izs`0uN4yxr*dV^2S){W^ca)M+z*kU#%2mkpb?@W_LK>UgTEl981~
zT54xrdub#Oj~Gw6mdv(2Ug9E5%v|bZPoWPXVAT${`MR^izf^U_TrKo7-q$L({;KbP
z)Z|v_=XI^pxl8zmNll+T@c6LVx=wtFQ={6e^TEtgIwzQ^a_eN|`Pkqbz;APY6vJ;E
zMvv}YiH(U4=&#ptJ-GMoyLj-?r+DSRWBiMjEdBg}ho5i<Q<c}pyu;dc8=jPM#XeQ=
z`b!eY8mw;OSYW0>)9&p-4c*3RX=#y2d0L}Zt>@O>E2NY@q+HwRICMUc{<?SXZbi!A
zcrLfk=liO`n2p@q0*3ER9;vM$rKC@vK6Q6NSXfxSOWzcfIO5lh`q;IwsD@Nd-y|d1
zMgUb_!PWvnVP)Hp$>7ZA=G^Aypl3@{?&a1%fm3H|p@Pe+AgD<?N;1BehI8<1MgmeD
z1eod#enN<Y3Z-auNO|Zjq`H25lgEWTpnLI%guoqU3Ogn%qO=t<i`3E0WpJz3kV6n3
zDe>g3MM_yDn|N0Il~9_OK4Z+g?>+U%lOy!aTeneET*Aw*k72^3uO2Jfv-8^jzV+^e
zFD8G@>o2{~C~OK~>7UEF;<}sIu``{4XP(Z&c|Xzh^t<`+XAh}~W~77=n?iPxpXe~v
za@iRnQ#2_ObdPe{G_Csk@mdvC`#>jk;3wV9yISQorxa)U{bF&y9*)_UoEDlqV~LZr
zBnzcZjZ=@avnk8NH_Jl#{NwR_JnlVWWB0B`HfZ3P3_ANPKAkZ6hvMw?+osQ)b<e|(
zKcnx-&Y`fdh^PNEO1|>O*b$|<=@a(#d;QB2NeOvP91i&9$flEJtS*;pQ0*J9*Lz9u
zc|cK4PR^kVXjYWsCXR(4NLN)=Rl|Vb)ZnQF1qF5X@7&znIT#t-4H4>dyKgv<F<S#+
zVPO*Bk3aq>p*Eo^q$7ki2x--ILnG1`<Y}^+4Edkli3X8U6#^>6v=*dnEawyK>aIm_
z$;R^Py;tZtP`LdHs$yJspQIWYq#)0wSA*?^G9cSMyc^i-#R0rtZ+`1f?x4_89qrbK
z5OLgnx4H3grOhjr95P5s?78WON-FP|z>xb^Y;ji0uY!+Di4Z1BtI4%B2d%mkZ+Rxb
zZ}aCjuLby5K}mMTE1!Hm@w_t!58c(h?<u_Y<~x3*yeTrd^C^A%95>{Hv2S2DH;&$@
zsBkjooe#L?%FF4JmI{E`Wa8iV-NU4>XLIHSx3FQiU~Re^yH)W0QYTlP6=v_(-8Nlu
z=wo_zw4St9df$CTb9%Pz#noLRb>7x0nJL|qQsT_H3DM&7xCr&fyDpCII&V!CiUNMW
zWb-Z$`BhOIclN`~{Qf5{8gf2ayLR*CSKlyw<}6&U=0jm`y8b#Ie&C)fqY_i^5Rx-K
z|8jEm2?GYw^Mq4<<HrB*{?eTEQTy|q{$+_I;BKiwZf@?%CJCpmH*{||lKqkJ-61W<
zsOybm!F2tAB7B?DjucY<ui^WedZS-9K3N=AO-X>a-+tSrY1%sDoq*mC1~#bj8c8-q
zm9o*s&|xu%RKE_bmm%E{MOA?$)~rhGmZtB)5KEv%(5o34)UAz9QO5OliD^t;rG&yt
z$;Mp6Qri%H9}ak1T3Y(2$z+l?o9#%++&&x@QiYJxV?BOwSMTSC?UTE!9rf5<tdG+4
zK*bQn7T(2VY5R~6PyTw)HFRCHDC6>1fDrx7LNNBy)520l-yW|$d~LM2eQ5N|PRSiV
znfS>^N#Ws5ORT;0`kQp>kiu=Z+*p_OJ!kF$rhfkeF)`7+_@Ad)wrC*{J@25z6NAgG
zlbPq^#4fgobB+(W{a;r`bygG+Uo5mU-_|NOexp|hNCZh>vF_z%-C4Q=-*AEa_Pmor
z2HkQ&M89sy=6+VQAg9=ml!C8UCh_!$SD80w9(Ub-3(m?)o_T&0FTXy9rGGBx>E~W#
zz`*lay=Gmr-^0_7KO(NU?BWkj>SmpL=S5N0*Z&c(-E~Q%5*==OJvzBdlVLi|;jc_2
zTYTV2DW#C|+eRM<ahdUtP?7SY;Exq44~bj50UBq7D#|{VPN_%7jvc!r#2G<{v$ON^
z76i8yit=gjF$n{bl9CQv<&Ihs<8ry`zyIF<f|Lsb8~Sy~C=mvstZG_UvbCY<kW(Ib
z%wdtZyb_OJ;?*QRP2%-Ss(q45FWAk26Wdc>U$?t0p-Y2kZ^)HomWW1K?Hw<Jp8u$C
zuIrmGzx?twweK5-8ngq1l*de#tPo=Lkyl9l$j7<&tle7eVPeB$_Jre)M>L~D(~77m
ziztQAo5~CL3u2`%rTleMmV3+MjV>iFQq^bv>HPhMVb`9}A*E?Y#<~p~`S{Z>xct(K
z1EhoI%v-=)?|i_9O@>-vbaWICKlm?JF8!VH<3DE?=doz3mGVlB%sd|^RSYf9$y!}%
zqyK#Ad1bWU>02v(<VcONS|yWo*G#wKJ)tBg^Ghkkg_<tOFY~iA$H!+2Be~@2e=_^0
zpP2UTciix|YcQKld^7cXZn*VM9(mwjjC<!z-W~HAD;Lk_^i%tD{VjJ;RNUBM37{x~
zM<2T1vSz1yK(Ef0*&DMw%CarhWmRr{mV#ciKi}xFzI#4sNFk;O2(C$;mXwr~nv|5(
zk;ZP|vh2LPLsrK$IXSt%5TenqUr4!6*;WmqY1;oaE)p;HBqt_5nv&Fn#<>vH1ZBo5
zhHTLN`#bhQKmsBzE>1d~PGK^c<Y%9K_W6DH-S?mnq65MN=S_tLH^A;61JA4#DBn|N
zxOQVh8e8RB3t6R&GqKC770Rpkc2?k@!|R7#C6KmH75Vlb)jl({abW-e7ko)XK~zam
zrIC2#)0=kfcfJ}1O$1SIEiElAeYvEhM30DwkhLKO?Wl)D?%DV1xkom1Sai7OaI~Vx
z=^yQ|nQl@9HULL*(-m!78}7U{NGUU-qoeoxnQ$zk#-sMA5aO|&`Kdz|ZE$^7RV_Ql
z$Hg|;Ue|RVeeyrt{`cX8hK6w8Lyz;sBM(wiQp&{1Q<(GPRF3acZ$DYJW*up%ofvfX
zSqwV+EOcFG<K`_a{c}0jDC;A3q-X5T$jsbP;wjxzjK1h2t?FC>=&%LIR)y{DlTSOR
z)8Mnu>>ZlclT%JUi6>q?mE@$vMsq7EEhR2ChV>gZlAe)CbhJ_Pz;3tk!8>nq!6jGm
z?g!&}{@JEwZhLg=ilUlRU!GhPT2$sA?NEI$=VoW`_r|xKNVb67+}y><Nl7I@K>CxS
zD3?ho8)-@+#PpgUYMUb^CB-3Nd{duEA(rh=UqT2|a#GT5jXRR3Aw=V3qUJlgwR`vO
zk8ZCiYJVv3+;h*#kt0V^SXe0CZnu2xwbv?#4I6fELPEkcK*byi7km#J--1>D-`<^v
zMOE%`0RO&c=FG5)h@)&O;SvgoVrni4u9=!k?t2KiW~sShg=l74wtKH-R`;sc)Vdmq
zWs>Fv4b8o9R}>evVP9vMbIyDJI1CKKfG~ohzMtpeQ4i;wc^P5m%e(yE2tY6^cVG~K
zlwA44P9^W|;%ry3zM?fG3f#)7SkT*A=>_r%?&i)FSAbhXXQ0R`l84O;iz~k&?+66E
z3$&XHkh0&%i;IhY>eHvsF5|GaMx!wX$%h;$Y4Jk{ZxWm;mU<x^iMLeF^pQiL<BQqw
zsI{IA{wpCVveUOJrCfK{#<k9?{y@l2fsl=TbyOAM)Bj!?=@5`k=@jA8p-4B<DJmk3
z)TQK-B7M0CNav+X!Hbl1H%NE4pme?a{r-OEy#Ky?_K!Vtc4nV_&a*qu%zS1h>j4nq
zeK>z9znIMUkYju}(CK$ss`y4xe-@`Ueokv5Z{xPh-LTjE&j30(=WA+OQP+-mYO7yg
zJ+oFvR%o!aQ*{KMFV?4p;_F)}X?PgyR0jtuO0Ufg{gWn9D^?CvfYj8~fY@g|T}1XO
zj0qPPH|xdFkWMM@PzE&?q~%_a>iSHX?d}G9051SL&x$cN+fqjOi71VWKmZ-?$H>r{
z*WHW{huNeZg%@LuU)skv`k&IW>!LFgO-R$<)DY(uX0GQPyw5?Z0X)v0N;{<Lr))cz
zG{2n}e%8*mx7$LwWQcx^F_!PBs;b)Y(^n!RO84+m!rt&c1+5es3xe}vv9(>vm>18S
z936}Fit`~52=2F<noSH2m;L9EizfW&M!fI%5VV>vBSF*#V_!?+dOb4I#J#6vr^G(8
zFXzZ#B)j%%Yh1*sxlu|x{9*fvFPt?GP#MU-kV;)WdL1k^lRrk<TO5Q!VX_HAnq80t
z3AtR1x7&Gfg?4Y>s-2ykA*#jQFHbg3vyBk!GNHJd-6t6jKTo(}NNDNW^S3Iie!tIU
zob}7<GI-1xSC&f|rY1#nfy3x|=hIaqmwU196lHAj!(~aJ5B*_lsG_HdDe&^tJ6_`~
zXYu2*cxX!l&PqR3JJefmd>Z@U)<0)_hpg^h;7&buSeTE)CC|I@ahoG+D_`lu3w-G%
zZ_+OyLj&u1u-*ejJbDKDk;YU`|4e7hr8?Pn3nd)$=4oHTdyvSQ)2JB<3FB{vPxuH4
zy1N~WY<AjUJ^`(I`Go}#fJ)E@Q=%Sr@qWCeFTkk@(@zogV;+^KsofpS;ew(Fa}Sse
z2vrF>RWT%Cky5ZPni*}*p18T}*VfXiIAt=%D0BDrwrYc9BWHu-3yXG>0mVBL@}SZf
z_jx{o(qR3_jf@X{6_{T!s>EeptY<janKtJ-r<|&(0n1xk=SScyys1Z{{gv<%jH>7S
zFl59b4DE1&Jvi?YhHw7|6`j}da${;|jfSo#BR^vFEN2FI0qg`nJnp0|OR6Y$n!zbE
zVpv^;e8W`GO`3eGg?us&<rYRRF_7f_Z|wk58|7aSLcKhK0c{}X5;37Ef6}8gQLph!
zfoQsv%RmhS8v<2i(ZxvUYbx)=2-5;S6JovEiT9jb@D!t!aQ)#&R8@KrjvC+bs<3GA
z8tRInBRJUKXSa2EeVfXpDhPD!nwI=#+2Usy?0ahN<)LSZtB{5}mpRL!x*CrA^V1*U
z39qr<^@*aO;R($wd|p;hOW84~+?bBJxd(cPX`d|FO?;)i;HXSx`s8+~AO!c!MZ#_V
z-M`%#hDsdm7eBUsA~TaS6W~duQ>!79yraK#iIWlRHoC^px6QQCR>U_rW@$BQKEcdE
zu2|r@vDweJS84Rb&5-i!sQKe9?7t0ShouNYN%ITstKCm=Q^Y*UB|ZOjBS&UF&DJvT
zJd6w1(((a|`(JEYrICHSW4$`XZ~H6rbER)`@7-)Z{{Ft>@yc!zUgj}vI5CWDP<ms^
zF~gE%$xF7tsduQiHI~VHZQknm-;X5{Ce_}aluW<#7bXhUyw%JmMBoS9($G-9k#pm(
zqCUJ}sb+6c_HKeyp7BYex6N)8eRp|r@8H71f*C1u&%V%0t|SPUJXum>*S)YH6AhI1
zGqC5Up-^^dQr?H=#i~+eO56pZq~v5|a6~=ApmIuIfwZWl<vz2gr{{v0LDIj%fX6S{
z^7Rt^_6>xz3Pa@vA6@rIMG{Zt<;{5{f}V8nhhUii3bn!Fg+e(5mYb21%24(^N!w&k
zLQ?ikrUu%kd`vWOiB8ss90tID>%%wO0j18GY^0u(q2X^4CZ?<d>0x!ABsIO#$%)a?
zgsZEoF~lNR1z(IS61z3uFUvRzuNJNOdgO$8<=>Wa9h`L)UWYOqrMQ$3y|VeS>XuEl
zjr(eiu@a~fI&M7EQ4N)S%7`3OSB|BJ(a8LZLn5pYVq}WVF)cm?<J5Fw%@TW)@xnGy
zn`EQ;;m*bv4ezFY;|h}S@-Ch%&c==AnHh<C*<(pb&AAh}Yh_48*<nh`=jdb`4Ie_?
zb_q9!Gr~3pr4x|pdXl*>@Zys)Vzd};b2Nv4^^V(BIdzlxyg9)FP8z#yp5fSZ$2E-J
zJ+k_K3ZpGItK|zwzY7Q%sq|*13U3NwFl)vxpxl!+iQ^zx@8eKji?uxr=13-|px9DZ
zf__QzWNWC)l%irUh3oP7pc{^`g>w2R@T<|~t09U(6v%i5@K0CajvXOEoT9)R9wT@7
zkQE}2$Uk@xYv!~<@eLTL^if(6IzyHjS7QMAp`lnn5HsHE=_(uip`oE?B#hD*h!=UV
zQV#+pKxZ0bacv_m*MQmW$4vsc{+PNk0hIAiD!zv)_8-Hfb1d3A%<v2>Edk65e<GDy
z17a;30yK7C)H_o`T*t4*(lL6bMdLeS&P&a=YgJWM8nveU+(~NhU0ek1MQORYxKv7*
zp3QRj>`>#Ghjm(es*vd36x3+*;NQo<X{ktm^|=H^e3B1+o{Ym!;=AfV`a3+lefYIi
z#_0yY;`!HwK1XwJA6|r|n~V)OS~?IRqc<}Ep^5~D{&Lh%`pi|F_LKz77rh3mV0sDK
zh`}X)6I#Ig22JPN#E36We(dm%1uBUYr1RVTXA=GBDMpYq=kaBxyA4tS^V)U+wWmIO
zeRSBAA|1#FEUy;urica-%3kd=ff#p#9oKX8sG#j`>9^660&%cl=kpd>xT}-t+1DoD
z4XN!De>sh;CnTWaN|CnVYs@-n8}l|)No$_FDbV9GPESDSsZ*TwS23Gdoc!^6si0?P
zXJ@oSr^-OC(1Q=89X{p=FdAoZQvB4mFv#C=1H~?RIVsD*u@N2-(Y=)YeZGQ`(x-EH
z*hLle$#%wQ?%}auA@$zVKK7KXM+tCpMqFKAA03~Z%u+vkB+=8^`6gDcc;Dm@r%rR@
z*VYsLMGT<>5@rsn5k^drA>fME(7^o<MkX!W((rd7NkEq}E;6_1B^M>FqRKO%E|io5
zfa{sEtLjy-tL^|IuuLWpH=6Sv&*L5?J@xm*=vPd}iCFy894`nhF7S0Ls89y}cmdLU
z;yZ&Jn`moma{)12Ta{Y_sd*bP%T-wB?DVvXmX;PGFOThMxXRjJt8X}y2``tHx)6~q
z!3Jf2O_hn)LJpEDK0C?%*vlBt>T<?nAz;$+dr#8283b#}+?M(3U^Wq+dK_@IpI&aj
zg9$;VB^%8>*PS=C6aP5yrqDd@{q`)_$FQZ9ZnH(!?^kqu@u%x02s=lCJZgO9B2xe2
zV8q7prW?f^y1ZqT=XPp3(;vk38T#SS#bak4U+!QAx7IW!zW7AGHz=^_xT{UbTpJDl
zvZ%fCZr)9UB=rG+_Sq3p4B=dB>r|h<$vW`yoTUWhrv!%EhK7i7T7yW8D&>31ri49L
zp?sEqaC|KBy~T|*GN8l^Qs28j<%Cq9Z{6ahN}RWW&N{m{92^rSPg9j;RotbS?1g9q
zQkb+gK|i*X>Kmc`N#3O7{)MPD9)>9oFRwceHa51RF)vs1#U?|Z4-YyO6-d)H<G8u`
zleM@;BJ@l*bl6#*2RF8O`MSHGSg4gQl4U05;|H4zN9*ONU;x_M+DTksTAOrNamk^{
z;T^rW9%bi)T6_?HVE`~1$BPoXrG=}O8>$2Hl~WV|I^bID^!$8@S5?oz!^5LN%xPAm
zuDbe><~j`Z#z*<r8usP0<_2k$@D_P|GV{GfB{tz@kTKu6WC?9gLld+XhgOaNS{I6K
zfEo{AASbU8m6V+Q<z`;cs*YH{jJ6xjT-e#%JY!&aUU5C1T_DZ31mlnQBT|vWUbn>l
zjDW97!+b1r5G^=qT|pU~%iBbv{~-JGx-kjPA+6TX^rWQPH#0SFuFv*o^atGNpPng#
zdSZ8dogL=qr!GIU`qh_$RXw2=U<m&CF@r)r-%pEkW%9MBoA3+r>{qoBzMp|Dd{5(`
zD<O*&;SV74z@qAVd}6UtjXpUA(73#RczW{t$h*P>dQ|&ckc2LHV1Tc~3jK+jqC4qK
zYjb_QZL!(WVUgFU621m)6!l?#p^ev7>Gv^Ik!3i>X@50%C_{2}d4*>Cv~zp=USnrV
zz@q+h=0Zd6+y{$4z0~&|`Ny`_4i1=#cwN!-v9{gX)>q$)KgG_5*+1EQbq^0okzvCM
z!6lZTUcE%S9*ew)&PZqQ`xp0eMu6CABtKL#fa&q+VQrSpp}uNL2Uv=@=ko7Xq+uxs
z3<i6S8YgSA)b@~M;8Z0iQ%V5jmzo1!?)@$&0GAMBv0<qdACY5iwBFbFuU~)W2V9?B
zOT^|q2n)stDiE?KOz0AGo~h}S5fXBZQ7fR0{aIYcj2G<6CntB0N=ZrS85tR&w50<%
z3L#^oqtzZZF}<!{xBe_9lE+d&<-~EGWBaUAVu^my^#Drw3eNmlHa)}py&A7tzW3Ip
z44(k&s}xS6<G=b6Z$QV8%beHFz|MH%i&&ZmwIyW+an*byaIVYqkdH36j1A}e?b=d5
zRtF1?eibA<`NA(S_t5b+yprs?TlnI#PHCGrab=U|+%7tc{4;0{jg};dw;46Z;oB;|
z8X|->9f7+yGN`rC&@Bv0Q#6L<GpkmXM8xlM)78G(y^Ghk;cb|sxZ*UwDtnv!>R*I)
zmIqoI$G>K@#gZ?8>ycuy;$aArB!T_TEvaK(jGL7kp6l==G1%1cG=&$jtD-=tTl0T<
zn#UFuaW&aPF@d$LLM(;Fy}4;wdEz8tVPOaup!){P{1{2<E_Z-nSN&;eZcYzI=BYni
z{8@SSVRK-V_E|YB$e?u6Aa7KqkoVi(f@>W|f8kGg@>obwY5BHuo8jogesR>x^WZ00
zOeXT&m!#RJ=o{KzW@O`po8>t=Vu3uo*aZ*v#oHMn+trlXy=SW&yDegW^G<qW5@TjY
zV!Da*0p`^yJW(GrbjTNzzp3euWyjpb+H@9P>Z#=2)e|ad!Q`xlOjk@eX1MiyQF68O
zU<0sjPt5YovfJCL_b;dS(M&{V|ChCOrikT<Y{&zGC?SstER9Sk6<Y)mWVn>{v_H_l
z0cQ4gzIx5-VbuD!n~Bf~H99NmVda2866=&}CABVkR#sL#z~CVlq@HtxhKECyRY5_a
z<j)^hCK8f3I|H7VW|>=N+(~_JLrIN*q0soTU?hh=p~!RHMh?}nPEda&f*m?80>D`l
z=RZ9xp4{a*FBBU(`cCN!(%;G1kL6o}+GGKo>hLx6m(cjaP*SNxGaWU1#sX0bM7~ro
z^Zdx(sSy~MTeG#b9Ue>*7sTY+BP%Zp=(>uJ(-6$BFG#$F>@Ww-9@-CKv0xtG{ib-@
z7@K~zzb%C&N-w`K6;xbr_MDYXYUs%A#6~>SX15{ykYAEJ(%tk&g+&+PqZ`GwR?O@6
z@peoY>=JrCwWOjs{>u8kx$4J{XZNjaLZE2@IslA)Z$H2>t9VxV|Lrm00q9g8g1wgc
zP|u@EEau^F`%heHKr^5t{ZUR;weB;p;O}rEs+!kbaN+zf+?KnzIBjC&K&3b&Omwm}
znEmf_A!2XLH!?`Tn%e;5p&v%fxiBLppvDcR&Bpi5&82~uON!jw+<&R5sReR6g)3x$
zqoX6c`pUrfjOOO%iRrq&S-Zuk_p^m}x7U9`3QtI@JXo@@R9iy>iN~@ri`A1e`fsUl
zr8>WS!UK<vD~-V9$d85JL^Owva=oRf3BhZ>jMg=7>U`Ck^`6;IR!%ZR&mT*t!7u9$
zAk|Uqb$OsAQGPpLac*H2Ym>>|F|0kw{4JY=#M#*IV+(wt!RvT;w467SXdBS%J&Tg4
zuZMa-sTO{E30OGYE@horcZ{$sLy3c#m%A7Ivd4FX-*@Ohk(6R6<m%g1{tgB?_;jAz
ztNjGDK4|YG<mO+na2{MuMw}@nTPq-7xhc46*=5P(Ky(ff*B-v8r30ZWeYfs-e6bE!
zk?=$+j&ioYd?`=5$*bZ3)ABD*1n5PGlDZr;=4>44m~mg&&cz&v+}U$WW4ISRby&)~
zeO2vat`0i?Nj(}WuSwnM5v0+zbApR5iCTf=b78LQysuypCAi@j#n~R%`}<;^Ul)r7
z-mi!ZP=t^w(!P^6b5GBuUO`t_4yNILw_oVn;;!5z9!&FV8=QO?V?o<GItr;169Zp2
zmn%ob2n29R=cDYFo!uWeKVDR9i%FZI7PN>2w**oO621oc{u-7ilfJd>lmv(Yihd8>
z*y9vf5mq0GFx~JIqLj^OmO+p0S(=f+ixpg2GBVxr%dp7<rE;w_=fy@<vswfS|0}L?
zcA4FT{~enSlacLG#KTsxk)5p|L?^&UJy@1b#k{9ueuFK$Xp(1I{xb5KeaKKLR0rea
zF`ntf?|@S7RGk|92Rl)*AF?WqT(k)PLRmt&%%$hhdL=FbkCwa0DiS(WiHfl3)S%co
z2o?Fy&=K=U1R=J92q%kfX;Z4<cugP)Edhcsq|c8hCQCeAJ)6(ku+!f%=NSEzz@X%P
z8f+UJZS(;~AFgtuKGQvQSXJauHVtmPDsTW~*Q{(P0w3Yaa?E{4)mKYN`zYAGmQ!>M
zTi6h$f>3xc4w-z2sb!8fH&)PN3S!~-Z2`I@kL-p0Y|}-;Oq&!=eDKM(U7~0R*OV<j
zk|ZTbm?VzlygX*HX{@`B(x-GhQ}X>(p2H~evCf!Rd<W{6TZ~%lTZLS4X6~W2_T9>!
z3Vye&j+qjl@60&WPghhZ@W1EOp05I!cs0x{sq*N@vl5GM+NTKHC#a~{A8j&5^R_??
zPzKpMo?Q;R>k$Lf6}yXz>-|18A#CkJg!&Gk#-Rzf0nakMID%CXLXg-lbovM!<$tfD
zU;gh6ZNcpSr4iab1V#UsiKeuQx%WRDgzd&**sA}J*2Tl8^8W=H+9)9YBb+9@`oaHM
a<PDSNm$7e|PI{66c&I9AJ+Dx-4*nnRE!XY<

literal 0
HcmV?d00001

diff --git a/docs_new/assets/images/use_philosophy.png b/docs_new/assets/images/use_philosophy.png
new file mode 100644
index 0000000000000000000000000000000000000000..23922d4e9476dcd0d3f457828807271ff58b1b01
GIT binary patch
literal 16463
zcmeHvXH=8V_U=m&P(VbQf{02JsY>q%0s;ckOXwX!@4bkEN>M;Sq<1MvAdt`rih_VN
zsR<<@ozQ#d#-B&e`G33j%UyS^d-yP{@aCP_v-h6;?3vjQubyfsQ(j`c1ONb~%Hv1c
z06-WH00dXbNbrAgiLu$k|04F1SJ5TI{{)lSyu^Pd@X}V62cSJSmH^-upz`RUuD{vp
zgrB#`Zsz_btd!jd&QrJlnyHo{mZ^v>x8|ujXGhd?8yYgSbb8W+>|RGl)&+Ta&s*vC
zCAq25R)HSl4Ib-F21anba%ey^mb)!hi8W{~)-G5%5bNcc@}kZ!bo;2g>F7I*<02^`
z*{yKkS0h(;^|6y)b_!L@KOcWJw}|DMiT^(5rnv<m;g=R>?)-iMAjCf=`tRfaQ?=Z`
z7~UF1sfMlfu%>JLqdtHzf&9xo9bq-#hWIr6p|(NLE6qfXn1i-gn)RxQvx;XP;xq)r
zSLA?Sjhvuff?a9r9To1xFEvjSb?jh9B4BX!1FZ`!O9RElH~EuU93pP^JyU;)6v%k<
z*PQMg1XwtmN}vosC@WpiUPHY=pPIC#Kfl<wR9*ehK)nd{dZ$~Q>hCbO2qV}I6;#z%
z!(hh3=Z3`g;Cw6MXv7I!GKcg$4Yrj!qlfF?@@rjkP@7`ehKHPARZ+8eMBx7AKrQs1
z40b>(C4GSUx8Qi`^Qa%I20E;6p?hnap5p*9=aU_1i42xB_RveM8~5jX5%csLyk#n?
z^`SFc;u2bHC#^ns#kd~53AuAla|YT_{i6yX!tkQQ`ylu3*WzJbULhm+VQZJ%7MaH9
z=hgjr|Lxr0i6>1Ota?^mp=<YmcxFivhy<5?T=S==qHdqTe?}!Bw&JKzA(|2pHSnVY
zT!fbg%X^YtC&NendOX<|eL13>vL_46U-i9L35Rco-}tu%%&2bxfrT7*|E{SkoCJgx
z894z=f23{U#;+OJAo1K^b^K(L17c}}+#>&2$zL%XSWNzq^S58Q3}Uiyk7x5P|1v&6
z03*8sG-=bkBl=^hmq`GDfJR><<)8UZ(Gd{l7&txi{MXtB2tBUe0@^NqsCoF01L2#-
zJ95B|s`}H*e`LF*K}g&p_HbL|&xFDa@B^EbT<nbaBik31aNruJ!z9U{fj3?Q1by7X
z55@n;79n==3NS<5J@$9t|1G6JM1UgpR}?DLOKCro_KtdeGd5nWz|<4F5qE};dKb1H
z4|q~8>9b33l@^BRn}!@}uy;k|3891UnwBCm7Iq%*s`$KdD(QwL*CTi5)cDj=-hHN3
zk$lUc?G@kUr{Ur6^?Q<yynV}rQyaPo*NyJ9CgKY2u<K!pidB?KIUY*ib8;#yFExJP
z1h%m@g*W=D+L)>bbG!TZs@BbXzIq)4k-rDlVdG|1c`$%$)eT72)X+O9P*;UGOT4=c
z<L^%JIUf6h2p<Y5EeQ6k({&Wr0qd}f3o0i5z?}x`F_LtF3pomvejrTJ$OYM5Gy_e#
z+S70Uj-xH;3&!fecg?*HSXGjZg;S#U_l_U>2p$!U%M+@41`7kjNFD8`o~p=^gV{~(
zY}(<kNJVax3q&f=Hi2T%ggw(!hn-(kF)``G-RxdYMNaVCXa^(OL&!AiKuGga*Rq0P
zd3=3_F;^2**QtK^v9VcH!JOS=ZBNf{HQPf&D!Uz*O883O4z`vgN(+J#)vQ3xh9R!d
z-zXrbM(aRF&}V!an!cJuK^YRV?P)SJvzB+Yq)*_W;YY!H1c`AO)(y(__Is?*(ug`a
z?&qk_N2N~g(z$&WpI9Ss4&sOo>#hAmP>{Vz8N^UXQzG${LvL>G$J&I6F_CG<rqyrL
zSaJ6KB+7Y#!rG3gES&5mCZ|y+%$96j5;yjj4+VV~U%B1)85=g~{-oa9nbid$DV&)9
zq`~wJt!`&nBen#sfPXcqCiCrgja4>bZur{#2c2Z?&~ightg|LrY{4^e(tlhw!j9gd
zpP;Wk<Q$_>01Ac9DqTA}ndB5EzIjiIOO`9e4tE*piY;E3;hB&uR6Abz_Bk6B%&7{V
zo2tAQus8?p9%go)XjYWvdH0=9JTZ~R=wLe)hJ2(jm2kBqNF;l0%C7lU-HqR?(Dj0`
zH*msF8OAzhtNO;oZ=zZK_DwOPJK)z{b66o_k@n5wbaERTti%$!!?eatXT9m%Fs>Y(
z$%6_gw)Udce7!Ekr9!05))dp=iY=OKTz;H(;5(?=VBnQr7;HlPuO)bg#-S2Fq0sYu
z(?p8pVXg;h96gx8#rG%nmHf+43*6D^s!zM^<Lqzsh3Q96)Ax11t=sv&tmhuB4^?yJ
zOF*7(rw_rL?Q>@vlN1cxt4clZ|F*~Y#p+SuK+<I=Jp-%1Rh3@6+u3VuXv{%Q-Zfjp
zzLpbdej4U_?`=mUu0Y{<-tlzX{QTe*qfsrp(b?(5qA`Kmni!X7XlcR1(g*Qwe_gsI
zj0BSt+k%fVL3%>6+HcRFQq55$kps;hW{yiNO+eI=)S7CwFZoyUQdjNwQP&;Xcz%gd
z9J|x&b~+orvzmsU)#j&8)l7YfZbF&3o`T<eV~816%p$)j&Vg={5ULwgE10Wdw^H3_
zPL0=e>f(t}RPOv1@H?74V(+^{lrwNTCSye6^9o8=&V`03Uz(NG98?^qbe=$+hlb=*
ztT9h~d0o_5W9wFUl1;V6x<-Or!@@zT>{aGwy5`U#4L+<Mn+vw5h)Im2-h}qIM^%wm
ze*^q|n6gjUWEBFJR<MRQj=e=T242u@IojI9D3}Z9>E3KC*FEx%>*l*2blIRo(A>7m
z9tUSVI-s-mE`C#1TA(hPsYnx<zxA}TAS{FLdf5YOcOSl2|BAZyG!6*_8}Ss`QloW{
z>fTp8KqlqtBb5><uj(h0<u7@F$xzEmDgS`FS;T;}m$f@Bok5qNxm_oy>khelanU=K
zZoT?B6W8td?47W|&LrE!BacKI-q`M4Yq?JL-^k~ng%AB?rAfdw0^=4g;+Dk|F{a<M
zbGTq^0~C<I=bLhHqv%oo2Rw4ve@*!t+ZB$8>e=*^-l%^Bm8&M?eE`5ID7so61zEex
z{9XjSM!!s4CP4oz-u2=Q`e&{kHPYFXzxvKWkmEh<o4$KL<UYQIoc6aY{L{Zf2O+Nr
zU|e{b3VrqfZd$bSZ75{=!?c(?{LwINY8Q%5wDs%GRTQdDPH7L4l$H6xUb6PI-N~kV
z&E~LMvLA7FC~LZz$;s|0_qxE(F>B7aH2&QyH)Tk5kgy0&P(%~rPWRizjhp3SE{7j>
z$Q+hMwYb*b@^-aT(nSQo8|5!A5%$UUZx|RdppU{nUHIk(;?~557}oo#rtYD+w362!
zIYMqo>9bQcjq6?N%4<kL-SQDfXz#SjcdR6`a*6jN?hdEusviCm70eUWJ`@LQMR+{;
ztrCb~jv}dCnE~$7nW=B4caIyt5p`6H4&zxoeb!rS3WsvzDM$x<e~f-j)=iAVR05fR
zOiQ?s58pBt*M7=%z8PBlvL*~%8azxFyoIDmM3-b1*eTKzf;AiuB=vdf+>rLiHTowN
z)A_8w;Kg4#=PRjP7}<#nQu6^EUf`D1ur@WJ$gb_N{O$xQRDy2JJ_lRe-Vhz3eQc-n
zVFx!4mGmFUW#oc=Ou>c?R^p$dg6Rok+$6Y`($W$qTTT7khaw)7+~ZI6U7x>t*64-R
zx%x|=5&C?6_Xu-_Dt$=LB_vfyia1o-Gk6?fmIC79-UhCwnVK37i`;j`*~h@aTmB_5
zbbUzrQL~ZXM!o@kM|h9k4=z!?El(Xb@sd|thr!z3aQ*nq@&sn(fw@<j+aDlfH6>d{
ze(p;?h1vrMr5{@l*my-15)+dYn#T<cIEOg73+g}0Yf}}g?tOQz4*?xUM!oqb@txw0
z(5C@3YmkmaW3fC5mK{FY35oYy1_szz5zRP6UXgxhX(@X?Qb}Z(y$f8-W(NsX51}j<
z-iRkxvvH`oR8p><P#(+H@dQToY)(%`TrW3hr)aRVNa6a{(T|Mxh$CS=889VJwh5n}
zanB$|S7q20SFjT(Ypw?wboBF9SB)!JcVoqazT@up@TrF5??lO+*w_@<8EN39mwx1>
z)=QgReO#0a8RdkdL{vmZUhCpyv%6T;yMzls$vfZ5Qhd{VG_+oKCC$lceQ=g}$2Vl$
z;}uwNR8X&s7@P+$c9rs*Z<W}na6L(=CUU$ZyigmwKpQRkd$qQV5#qC}zy{;H%zv`O
zE0XxMbj7*uUk*ujYf1&5E^?N=kpEXk0uWZx;M2vP6z+d~=vOY6MT^gE;l_7q|J%ub
zb@E?7`M<PwMDn;fiwnNJ$GN_Hm{j~YU{fXS_~cx5`s|e2w5%jaVsD9CXFWjff!COD
zV8`$aa+|sD`vp+s@NUrF3R>Z2z%2C=22s|y)yciJWS<=agZAwDh<ySPs;Jy?Tq@Bv
z)nk>(uIU{9pq7LwiW?jneB=!a?C8!4=qM@C)-ejuh)GUNnQshp@!1vUZLbU64m~~Y
zXhj7ZZf$MNq8GA~1H2ZNLVDxEpg4B=bxk1lg%j<_lR@D9hYxpZCOykc{7!w~UJHBy
zTOG2GAAZ`bU^}|Je0QYh8h6PO*Vf?}*k-1b=tr|c%D6w)e`%CmIi+{(_O{GKUf3d8
zl4XR*Kn%~p?h5uHhX%`C+rs3np<`TODV$@aOb2GEQx|C2z-MYRy3<u)TC4#-h#M&=
zn0jBf-Lz#zZW5vtpYl+rsACdQW;>7(mXX8~dtDs;z~P7PQotv7qtgy$Pxk>O*p<vA
z&(W%ry)m~`FtkqJ*BQ$fb~3=y8k&-wM7Ui<O@0MnEsE)WwMiEKiM@9nt-}{OrJRW7
zGq~eIAm!E<&;x7RTlJcoGcZf(l{vn%5-tAlwg|S+0fvz|ProuLtc)y|zMI|DlmcAi
zb$Yh|D#LyOt-qaQRaV(dV12Ov!k$J1@MV(H<-i8i6xiBz*7wRLQ=T8(Td`Mq7h9~U
zsj88ZJHVfuU|_~OL`Mfd*fp4KX`}4+Iegv}kD{y4DZaclpYGQ2J&i@hH1pUj_xS|}
z_6wXqV0j+ReW0jDCG~i%&VcW{XAU0Jdoy6`mFzdm(;3>ehX94ay=E9I&^98O;l~O@
z<<p*l$-;9XKM+29@z(`sOUt0|P9=z!Hrsd8LJyuER5U)gjXw;ySyuFjBAkeX;;s48
zDC*=gv&}py%Qui4id|TE`rMce2wYJ+Q1>7UpDbe_wAdcLNKoO}JI!En3w1EZ&!zx!
zFc02TOxqpeX(JR_A6d^Zj|2RB%5p3TuJQsHq8$tvnsD83oX@tnbFQQ~D-i)6sn=!)
z>VLi-i@6x?k$?F(jx5|#=f$o(p~b+HfWv^TZ>>~aQ=cMjzrKdeEfr^{z~S%;lK?F%
zvg!li+O?0)Ih&J1W!_P%x*fQFwM$fMsgUjY$Df*|du~ll<}0ZMrc<D6EHj-mc9Y-H
zM`eBK6M;2l`1R8+J=l<{BtVxexfE3A$l2r8l9{_wIulmMdJfdMRFp^jt_wZxNf$Gc
z*jeLE6JIUZIq(pNTQ@j^s^v&b{Iv|L!d>Xs!!I(Z81O2)ra}`Zkr1!UJEtbcLeE!6
zdZ~M2kq&JDvb!fG`+Vj0bwRsWXKb^=S<VEiJ>4C=obpl#_%QQDjkI%i{c<8ryJp96
znECXRo7V#@S(b(c4KOxE9h>>Ka4=5GsLmqwT}HXX7IY^%^Kzz_O@}P9GtX#@U65P7
zO=F2uLD?rYW1Oc13Qfo-)#!{$Cbbz@CX(PrdT`roxJJd+e}wJa?0YD&JX$wcO<U|7
zy`TNor)#gHIrBoC-4EZTeht_9uxAB}U{s)6ruW*gMpWF=QgAd9D$T|{zy9t-1O4Fa
z=h=@Y&1PC=?&P_nbtF`B+`{UKSv%5mrfG%^<%Q%qb4o7<RhlblF0-DJ`uU!jel2kl
zZ(M&~Zd!yL=vHP@FI~!^amA0*_+C*=8J{Q4hr(5whvCmZuw8R(irSvs3{@}KZ&dew
zzi|pG#MW+Vdp63=OP`l#NO{&i+~j?3;(q8i_iQ~_v$U(rcDjteE0;GtW2C^|w&R`T
zdNTA#S!~?8w-jc(B#F7iB2b20IAr*y=r_)rIGe;S;W0bdLG!?n)7*P`#2k)UP%+ZV
zrCph12ho2|ojKl9{dj3r$|C1U)3FS%4hSuF%x$#))jrPdRt;{@sA%Gx$Z@1DX2rD1
zIHhs-XvuT5_Uo&f?!E=vKtq<otwAoMolZ4R2V6>p8B0&h%KE35>!u|$XM9NoHkos@
zZ5}`S9;WU8^y;Y^im#5Pu)kpkqh`%aW}HXuJfmGfGjFNx#WG6mT3z?qML6AqFVCdG
zeIA~c8>95%Ke)W#EnG2a)+JfOCPNjX9z&1xmKOL&Yczd%&GUM<cI`7)Ge?@z&sO_T
zTp>q~p=N?y>_)HOFeiO(3bNTB;htUOYtU@X?gme^_)P{Bn|RF8dFCxX*H{VCMmkS0
zMqT4qp^ea|16HHpb0mZ;a&+v4p4#mq-Z#abuGsCekczk@vb#0kOD-(L4i}#7-&Bb1
zsAY0XK}smf3~H5iJ#g-ST4(0A__T|0l*pm4fJ<WWfKf#SX6}1y<F1UaE~yX8$o_TA
z&I1ACcqma%(#4u_4m4&JGiaG@dD!6AQIeM@7|N1=F<iv(U<4JYE8)9**K7W}PYUJ7
z(sIw;(>FqQiAh>?3iaT&-M6oMSn_jvUtqZqtu~uKqTf;3`t^1f{U`?E+3!ke5Irj?
zdO7tnV+P}pRzFw3{tG%vsI7TsXrx4!pKHAwX70u1rwCymxbxa&BZ;7($c;s+a2GGk
zV<VM9o0=|%jg%liS_j~Je;9Ut2UUZv72&eR)khYe>mL*4To6yMPRLB>l0FhT%fdJ>
zD^?*p;Np^BI%j3c!f!V-YE5pNk%xCkv<`$8?A;Lgelhfu@?6OSBj~ipx`fMDs!Qa=
zEpCj&JCgS;#)+@{4RrTllMvY<<q={#6!RjCQ~V!1QoUIod_N?pnEJX<_?W1WIl?jk
zixKXDn=l{mfOR>MY$HmUx2$jXpEVP+?1|b=eTms|>qIbWphA?0n|q*@K4(%!$8+|U
z{596SBaQ<_oo_;SSXKH34wt!M9qrknjQg)7pl9zltPBVt_ka(vh#uxXa~;zjrrGg#
zd)$4l@h)ogDhX#>tAchJ-b_M<_YfsPpSbxu6I{kLBIoy)A-IK!;%R=YcstZf7qseB
zTFO=f5?zwCE4Q&(Sj3nh;70>gwlS0LpKw#??=>n%at`?rpS_z62GF-Oq*SV<XQeAL
zHXpr*_K~fw%|*imD}4=bp41(i72ohZjZ7DH#7&Eo{J`1zt?)?iOlOTWMSEp3AC+z4
z_Oa44>MZI&#sgdU$w51n>W4GS00sM)5CN|u9O<KxtX13P<Vbd2UOoB{grct-ulwho
zuLe{DQv>?LV71}puR)Bbfs_>Cd%b%&WT`ew3im!(ZA8VifzRlOa*5}3H*62<=TtVr
z=>|#gH2uMLKt>VXfV`26Wc*yt-eX)B&wMXsMp>1OmD{wuTo`lp+6{pB%Y=Kc80U#m
z3kEq|u`;2x*od@daOr0nQ^45OBw&nPgpd96%S~#JcNR8D<;X*aRcq{)Zo0UXC--9#
z9di5PXOz`!(JRb9_+)k3o-A{xxePF;@?$||>sl;735jx&C5=#Hr2DiT^psWfZ+X9G
z$s&u9c6MS_0kQ+^emsjJLwB#ZFt``&E-zoCpJ7lA>Q!a}B4y1AOk2+A<;cl2xZj?w
zx3ukg8iGbvj-7hn$>g_-$;R`3<Ox3fs)T!^^y}MdcX|nX`q}-9CiaPn7rvg6@c19H
zm)zxb+;LeM3=4nKgfr`K7}4;B7(!*gVSUfci1Ei6G(k#6YCmNtvB`nFU`0fVH7N5F
zBTc!UU+;1<Wn^8}K-h{uXRPrA`pJ9~Zur|~6XtbU7!wAEEi`LY#@~1$i;VTdP7<Iy
z0<tUi)sVG~MOM^R+~xQk($H8pEk2ePa%Gpo=Qv|*&#>8kuJ9If^N234QKB$R2-Rby
zu^UE*(?M{be))x{D$>LEU|8yQ^=^;raQC)d7e-UoaU<qC;hbVk?^_WJZ=gJgZUME^
z_|Ac*@w{T~29B^Y+QUb&kSAzC9bIKNw8VrtJAR(-mORO-BwqadQuy8na<^tk^YKI3
zUikTb3KWFMeJBQZcbw4JV+^{>8$7qJvEj0e2W<|{;&1Ar6SqXM!Bt2Hy)MD9(=r~j
zMzPnh{05y!B8#}FBG@E|ABURslNc<hQAzHCqU89ljp#&bLxMM<zUyE)ip$awmhHs>
zr5Y?*YKRxkFQ5+%jW$<dt~mX*;fAIglR@(LI&loK>gwt#OCuBmYs1YN(mdG>I`qMn
zg%5GY*iV_W%VSDv`o8ss-~t~52D-sF#TO~U%c^%hdX$+ma~9Vpk#N3);Ftlc)Qm(n
z26N%Tr4KR%w*KeqU4_-PGJPZ@a>wV8xsCi012v<~8dGc&sqb5`U989FkK0&t)4B3Z
zhUJ$cKz*6|dQ-jlS^$is1tKwMv#TtqNkv`hy7)<_SX`QBIImi-;h90$mSyB^Hhhdj
z<#S=SL{-mk8oFsD54fBs-kAJRQ6^CV#zy_*3BkpY^d7gM9{SpnCk~&w^@f~V#!(MH
zbD$*e2^+Bw2<WJ3x>hGa?AR&2FZpm#9-Am?4i@^76H>GGKqYm|5sOHtbz+S^rQ{jl
zn{bcs$&~EMM}_JhIn``&tZD_~Hy@oHN!S$r4Dbfy`laepz_5VVW!t=zrOPrVOr!2N
zbg3NGx6<|jhVq9rh)SX&pA6mS@vSO9-|{j#*E<dL_03mrSdX?soj6mxiMx1_!cYHe
zM3}(J$nXeDBj;GE!>uC@mr<fY<Fyq#Bu5w%i#2ML@C}rRv}3km%|qViBVc&0cE(qk
z0x(5dn-b;vt}c$&WIq`oRY=dsAjxJOG`Rl*cn?L1tv8&@)SbY)3Tn6S?9&4P%j&OM
zfLg<>lNlD<nWzKaJjNS<pR2?@7F6Z3x4?FoK8~bmZOucYED!NL<Y<1GB)7+G<QCu*
z0x_*OINO|#8i0W9>tFEtqp!(;Av}F4pZCp=M8*$$$thk5V^$^981&3)f??Z38>B5f
zl(t^7!}op2eI<%1nJ09b{kF*#j<*jb!I>V>zOk9(nta18j@ATNa(`G0;J_Q4y|^$A
z%@!3{G2TC54_x~gb0h`{ALebg)>&i^G7bWq<AWw+2r3@6bMqe@dr9cmPQz~+)vXFY
z884oA8{)Gvs+9)otxY265~l{3WG>X>+;dcpcr?1cz3bk#@A{xA<GR6vtei;N4R8yy
zzO=ud6o9#PMeGKr0ANg>eJVX#7REs6W_WxS+jPcJ95cd5nbNrVdIf~);d*&tcborM
z<8)N(R1!WLXOLO>Ebg)kyVQDKM}6xcrN;W;&k^&>r}bB;>TaAKTjCnb1Wtm*-KWC&
zNVlKg7)0gY;8n=cyC8S`!sb)=Grf8Zm+XSbyaupe3Gay`c{qk&9PC<Cbb4ael%X$k
zu0Ek=@G9HCJu|T==}|ez=WE0I^9&1Wu|711cX`HPXb5*MCCRXPOg^mX;kgkh5f0{d
zZZXvYOOzHWnhZi2hLc|o_icUCxa@TO!Rpl#9xjj24(72Y$yw3+YVCK;g3qZkp%&=}
z@Yz+-4=W_`LV%qkcvds)h$c<I@)<2Xedo*@>D!XNYcIz#sjmSzP=^SZ+h``DM<i_q
z)6k}x9l@Ll9o#oF39aAlHXm)Nu%0LsG&j3_>`NE5YeYh^ng?+biMTtIwzxS&ipv~s
zo3ifPaP^#nux5@JsbQVLO9KM<lqk(*>>E!eF?=uGLTuxN&vSP*KeNlQVxoEaWk{UO
zmxD>qw0KP(Za%j@I_7Ht?@P==^~Jjp(V4O~HdJ*#Q>YX3cvHzy;~Z+<QWhUSO+%ys
zrI23kMh2oB^(Ze6dxsn{sg=VA%cWr3jp~?PBdwvkcdr0Ull>6{u*H_pw6ZV|AyJm#
zxH4$+hwv>7A17G$yx^f61x2_<{%3N`$19CxF_RS`%5QuELlS*nKC#GuE_~I{IzpJy
z5SDT(U165{G<u*S2aGT_8eV+=Rb9mHe(c5D5xxHD<MzD=nUc7$98b39Um$ZK$ae{b
zVaVX{PVcbgbsWw^hMgMQtFa%`%@J1|3xzLPm@u{0Eay{YthGeYaCV5DvfBxk^;A1G
zod#z4!hCPz-cubH+MzmE>U&!C8VpFxrA*tyb(dRrg%?61sle!lr<{H}*M*C;fa*Rc
z+qL^E!3J2VbS(!{a*>N1gVc_Iw;gbPV*kKy$HBfVr~PKo6JJs4$EI{*-x)|r<jA$N
z8|&KAPWry-f|{DW@_mT*J=;Dv!KT;bp4O66NQzjG7;p9nfk)48Mnrr+Lu2)rmZzOe
z+eMef9x|ENx|~`TzvwZZ)*zK*wmv8%Y1!poZm1xo7_bOvcUZ`l4Rk3KjOw3km5Ek=
zDz2x`zAl4$62E*+Q4*hh#O0rCAYhsIUwl%Ag&u0eeDqPy=Bhw010VbPC%5O<5WxG=
z7$oUgh0DRrtqydG%Z~8KP_N<XMhH4j)`}=dATRC0jT7l7v-iXIjLKvDM%hz4Uvfiz
z8Hvh|hp^8Wt`%{sb!G7VoU#;)AlTSILV9Y|(hqM0k2=xlc;Cpq(A?5S3X1p~I=kS#
zLOV~RQ>NWc;`TvMt#z^3K2y(`p}P;2L?v@zohIa%m?7rz*=Q~zt;NzS=<EeV5`Rlp
zTy>uB)6N%)6JB8OumE=4V(d&?=*N|CFzbd0eW$zQc0`T}C(qy|2Rewd(fUM}Cod-s
ze09d&Qx|ic=c)9ZyNNbT?0izTND6Z&&+!3iHZ2g?FOi70r`_c9&3s9e6EA#{#z!FN
zyJBd+l!~{TiSLq3%ccZgbLwXs(FICH60WXCLknH+xk57KcBXOMw)-hj$HgR?gojTX
zl(<pXXZ(tKts}l;*}>+~Xu*-LWp)wE!c8M2&v*5Dln0cr%8Eu^CeXB@3?Zd{q2)|L
z5_hx1I%7S<XJ~EEi;&3Tkc%>$pC^fxl#x^}eSEV-!~Y_}&uw9?ZtIk^O|mLZ6ehmt
z%^w|T3ceiPv&H;tcS0F1;{C<$9f{`CD*jaUyVt_Dd5<E6^@bO(*^`r>jr0?a9%LbH
z8;h2oGNo%Z>DU|ahrycdzXe(pO_WKi)OXM0H`})TV1CJ@Gr~75z_n2_>JvWZz`Y-Z
zKHg;-nct>)AltqnI7plU$}pPS2&PvX9d;7v?&qVkVDFUQsu{YK^CUw|>-D9W*;les
zx5tn!RQs#`ne7+I!ly~(R_QC(sE=(|H~_vrJmk1h66KuT88ou)CvKTSl;j;|DKY}9
zBh#{{1G2BWiDej2mnVqj(zTpvI6vdJjv5<DqRXI#9$&*-O!?MZMds4FC`J`1VUDa#
zRt(^PJ4HKN5LzrU`Vv~a+Yh!PY4KnL+}MGv)wc6_RPl28@~fJ`^=@bD&n_<$d=`$P
z3DJ-0S`g=VE{9W{^j!?Zo8T!}HtLYN=Kzj8Vm)7w9WcwPw4Qw*DubItS1H?H^QtDp
z-?UsK=vyLecF&7#nm*Pra2+*mA>fVa9=fo?QsWS#+dFcf^-Q=h?@&aE5zz5jO1&<)
z_O1-ptg7Mo?X=7M><+iVa{0U%GqoQ9?mL>Ofe9LSZ4@&n0+<PyHQ2aaTjSxPb!DjV
z(HoNpwO=NTFS_uSZ|W}U`S67sbJ8rXK(&4A#=7b4)*^Fgt0E&GG4lymo+(M&5@A-E
zb-JvJnB}L*_O;@QhVX?&`MzvXGcpMS&R(>9pE$FGh%uSOK^0pdL&bK$38V;%`g(oL
zMa^~9>=3cHH@?WT$kMoHu0tS@$Y+s$DA~P4xVFfX;u2#Qy(}L;uTe2)?tJ|k=&~Iw
zJ~!-SvwiMcLVmwVqL(zj;)#M2SpE<uvewgvMT<4Mb3_;#tkIipPNW#S+bxL<uw6O7
zwQ6jt*4f6pHXxjxC+(c~Qu?v82X0l3L6WM{-S<kR>^1E1n%`Vj(7H#xl)3+OK$<Xj
z<EO9vDd=<3vT1ED%O`}D9n`n&mhf7I+&~UQsS{SRTYXL=PE5(lYp`EQ49Ir%ObL3W
zNNzEjB-cDzDw*v}C7$uAz7*1hsfX4q#<L+^EyN&s-C8xdP|FD|Q*Wc!))FzoH$mEp
zA-%Y02Y~_w(#77OCG?RnhVU{7<<dy<G8A*_)N~<2G9p75Xg+YxA0m@SUw{%_TVi{w
z5ppb{@YFfDiuVn|X+9{9o@323E6$*y*Jwe=pY!cgg1$7i+}*$;mZ8ccnZxYc3jB3X
zl*~Mmb#YHj(kUF02q$jPw-haZc?c5!T<<Bm*h#^#v~dU2&hNj9Dk8KfNRo4Q&WWS0
z?qJ>aPH-F>%n^T+g?Vv`3p+%3<1l)|P$TZJz-3MCQfo$W<0yQ+*muAI%%wiGOR$;>
zscOrfbp9lWWaJ~#+3tQI5Tikrx7Yi{DIdfojG&0!28JNnA52K*pHcU;ex>)GX!0zq
z8c^QQe&CAw&|^ruYE^O0B<8`Mfs4H%tR1@9i+YQ6shXX6Ba<@X?Wq>WCw%1%E8Ny1
ze(`~TL^HpIi!3)-h@cPsP$4qWMZq3?(zLHVgRY6_Y8ib_2O?>c9TTM|5E!?$kQw%P
zIVKuqM$%UQ(yi97kuNlpyP<|TiOSpv?eHqof3V6Y-!!npyUx0eqNCrVVwi4sXQ206
zTU@27=ST6i!y`yLG%;7g=+@wO<LxT@*I`Q~Rd@)E5p^^iyzs2cJ32d{_uS2}35XyN
z$gs62m-K>0sexFB!ieMDCf#Vpy?;bPGNxo#LU*FM(r<YnfyF1T?#PrbCr>NYa+G-l
zBumEh%9y@d#hR{YH065bE5U^&`h2>t>~OzR#^T-4wl%(hffQdOW>2v6NLP!lzR-Iq
z2Q2TH{6_F;cHBTUvx-P}Ce7ZR-s#{rt$d}W?bC(C!U&nk-hKDZc5oj45;p2OkO@r%
zx%%yq`;z={679tFelR1$3^P{0$cec;44bfvhMe2u0F1@5Q&U=M)3$G9@`TG+&c&mf
zO;I?ETcA+gmFiW6k~!p@?J#awCVl?<W%2wjc^lYvv|7a41xIU=bcq$8;xHAAuVTHv
z)dr)bdL|gua(dLSTxgC8g}a^x1}tl8ih~K;X2=j%s=M<hHg;>srAPRQLz)G>hJB^p
zLmJ%6J$<Jg3_m*%<{%c)fguXYyY9_QNZRht{OM+7M~`yXK;^eLM{|!diL9uB)%u_s
zmXTK++?lI$C(2XueahbE)?i9r@y%71O}S(Wkc@qsVI}0qx!(z--<Hz<YJWOb0gj%p
z$5x-emkvTo?VoHzYnZ)t_U3CBAah|Y&tKJt%&JRV7K3Q5e_E3LS)Y~uqNwVEoITe^
zZXd3XG5j9-G=^JkILlo^q0x(vB44eNP}lZ(;YX%-kVlxHiN!&jeH^)H!foX73rOy=
zEYB^C`fVBeZrz1rhH3`)ijb_+kHc$L=>ql;SEQA^m!D;n4i@?I^NX6vwd!Ha>@|=R
z`yi&&zCHydIrh0c%;3#Y6d?#xcx$%g6Zyk2o@vE;CG#n5z6kP%(Z)_~c7&j^^0YBz
zQP*UE!XAj+?xm3fGRlR)^s*glP3WrANaB<{#_#754w^LP)-Z%7m7XL&jn?sN<dDGQ
z$$Nsn_YKQ?9>|SipNLT+E7;o4%CN|_&sreXRRq(J-2q17M`)Zl)m76+^`9aionJF_
zuX<~5*}FF}US!IpSpV2THIQlIwPU$Ai`486$DF#$g#{^EZ%T!9I%lf%$aE12QG{u4
z{1ekYgxC35J!1T@{<%7Y=hCQ&%;3mZS8vc$?*l)W`&pGKc)v9HNZcuN;c;@j8k}xc
z!H0Y-YT9qYr#N{1HFv>d5ZhS}Q<U2nW(UgJA7aXs#e4%;MPa_1roRANpFHp}q62M=
zrkGc@ruZFB%_VD>2OW~I^b^Z9=&utn`*fse^<3qr7ZQcYEb>%1k5&7uP4=MKzaV|8
zcro<2`XIIQ%I>8z1gkzlAJheW9STB@q@7qzyM1{+*(32h7g<0qqSB8Z(1G2!GgGwZ
z*A|UlsJtJK6gIo2_EH$6H)d=jI}our(-tQk(!Dz7C#I%7y0UF}QwzIoEM{Mx;!BNT
zu>igFTQ?4J9poPc_ePr4Z)@ruaxIYlL=)y*P(81AN2b=kh($<b^&%ssbR6&pEql#-
zu^0$0^5*DM)W?b!w}RrN$$1|01g@^fag<y@ilO&E1<E0L&F&U5y2E#?(9mncyXo)V
ztvIXsA_0e9SchOWxlWMuO8bg$QwqW&Q^I4usVlvqdD|ytRoAq>E2?GG4#3%m+?Gt`
z_Zq6YdN4D&ImGlu|8>Q#nZha-47+$ad~ZE|S)x=VeYEYFvBJaiiwqLrUJ*poN7>pp
zKSIJ?vhBz0Tx*aP82Tc_ke)_}F@ul6X~T~?Z1@0quqd4>k&8<jEL(a<$38iFB4}^%
zbdPxE{KIjVUp!11@WwCS+!4POvFpW251pAmJ&m3?Xh50WHpd;je~Mc}D|o-;<F~Zz
zGA4dTl|Jqv526gBV1lB4ytak<Zqlw{u~W9bltBWq%}4bdUDL9kLe=@;!XJC!a#gE@
z0-N#Id-C!=DR1_(_S`8%O3Hn-u7(mFY-fW-AKFDVp0zPO*lvCE0l}1462v9|=o<`m
zX02Z6@s|Qki5*nzo;y>F^>i!CN2NLt<#dzWl&Ft3fhIk~*c6ByPE-tcgay}De!41H
zNfu7(O=ZEYth}0a*=H1b2kDX(CfQIF%iCxM0aHiS$G<?=WiR%yZ@X-+j!JiwXF+z?
zm1U1|XBQSE-Kx#kU$uK{VOqPT=D7}Zghr%gSAKNa)~=0OguRWV0u>L?d)Kg7iXiu&
zSOcpj#|e&tKD-q#B;>ovcJWy`v6bvK$>mAZ@D@w3VV~jo&;s0Jul#MqkI&_+u9h<s
z&!~bX(FdYqoj+2P#X&%|z?l84gFOrA_~DIULvW6|bH&BWl~Uz%z4hAB8mPgvS%%=}
zRX?8%;rf;&QZ_ovGFa5?Di;w3LhnVTpWGGzFpLOf{>#HC3FrG^AS(`kuIoXNQ(pM}
z*x?H<>mfz0&Gs>7A@|E6QzTT3Y;UBtUd|%j_*rl_Kf_SauHG+>{+c|vbZ{8A+%L;B
zi!^;I$639dbQT%HvJmqgFpnX08uaPK(#x|QufCAxSxH5?Ok2;yu2h)M?p2Hm;J6p2
z#gCS82t}BX4X|U7jlbtL_nTgXP-!tK&&rnbwENf|FduObbG#Bl0NxvM7GQ(8!A;e@
zx|Q5T*&cep%O|AN{A8OybM@J9xfyHNFuH}kUPq-Iztu1_GMiaRNr2|se=i9Ak?Vyt
zWGGtwW(odjRI^gqs_v(bKl(-;I*HM6DB<7&3QFFx%)}6xJ+NBF>RF{i`xsi^Jwlm>
zaLUdl!lZyV4!NU_4>}nL(j=dhEssoMYnrDb<rtP>5kGwmCUXC2b|TyNBqr&LKV^T?
z0YngvC)fC?q<7xBs`&`;@rHN%^p1JnRwy#ziS)^}B{xIr*(A6^5MEzBT`G-odCwIx
zLq6lG*;kwi`Z{e!B;9`6EQs6i8bq|?0|{i^pdzT4#(yhl=f};sBbJNpJIX(mCMrJ<
zL%`Ka7A;JpR{O)0F1)(FRK}WLbGy?sRosRHMVESl3eu5cnZiJ**MN7QX)0gsNzf0y
zAd&K8bl4Ye0CQ}#QT5jOp)@S0cc4jN4~yRt?MZl5t{H-0z0Zc9p>Rb+1jk1tvQ$Jw
zIYvfr`lq;b3ajGds$D&*5ccbK6K^f{N4heHn~E+6xsZmO6G-g!$wbEq?z8OGd<(T%
z=}Y<A%2<@{lt9sljB99`j>63jgWQs>LQ>9`G{auwM6%UGpPngHUBchRkCq#o@Dzk!
zz;HtkpyMGU6Fr^+VQ1UsHmjJnh72K5b9bORli&F;X*iI0jo6A{qsnjiB_bC2uJ!D$
zsl9cgD@t#^I~`^XWDj&mQPv28xv~*;vdR|ZIlKhcprlN$S>gTTngg68P?ggAs%B#P
z#Q*X*98%JI^9GQ)dg#poXNZqymiX3+Il|v|64`G4{JK?xIyGNF_w>W@g$=xAGQH^H
zr5}Y=Qd=9WKVV_9X@d5yCTC(a0F%C-8dbP-(B9i<mQX|IpZ(_9hb%_*-j0UHX<1e2
zpB3Dm0I@E%VjzE~n#Q;pq{nk4L3muN<iPY0KAsPC@ipkKHJpb|xWb==^_?3H!~AN5
zG@KE2fnIo~M7kP0*}UDDfWKkqYTf?iae)b`!1-ZGvDH~0o@<o_jH)$UbfK7fdt1Tg
z)3X;_<6gaAZFyT#JV7<j0f&i6kw;(avofWrTcw->7{>Dbw#UP2vnH74L1MOWx|_}q
zUQ!akg^hbfcG}7aDw^UCkLUP9Z!1JoCikKa<jn%+lHd0rsYc$92%`XkibHkKDxC|s
z7Jo;mVqhxd9Vg>H0ySkAaOEJh=%Nbmp^8!COK4cH13BMbwCzgi+U}L+f1&sSvL_QW
z8~IBf8buCVeT0bM*s4d-*W(GZE;Bc45|=sB^tVY_$y|{>JVmSB;I1Iv5F90^dzOpc
zI|CvM15<4~%vT0Up6&Ub<W>UkLnW)P;*)gvTEiOGyJ(|drjZ3OoHy<V*P4XQP{RHD
z@1&tD2xiXi_&=zA4qW@x9*^cij>qv@xx^Te-jqz?^Ux0`v**di_x-)EwqgO_Lsr?<
zOSDk@41I`O+)-4GP)B?cwH+$?jK~6^p@eDH9l1Yvp|jP<2Cb?9I8RDUJFVeQAJ)3X
z{_<-cJK*1tH(+`3yQC6d1wD?h44O#yAo)`b`L+*VLH!&mbLlVJs+QpvP%k<|@w?Xe
zirlSSe5V&pLuhaQQ(28~-1+fA)vfLNAHI#8Jib$!`lGv2e|nuV!*hWqV+DT~jpA!;
zpVQw0IJ|eY{QvaY(u61AG+np<OSiemD+fGTT+ZM4(`$GTzP2++u;(x1=&Erz@V;+I
zdHOHwX`hf-Zr?8BFU^Oo8oxcvEqn9%Z>8gZ74lz&{MSSNKVC8s?@3xLTcZk#?R=o#
z<~;w7CRbi2%RM2Q8m+V*C{6Mlc9@QCTPAWh{S%)60t3QyA>gu?mr`On!x&gOHvKH>
zEPHG`tNkz66VJN5vr$CSGB#m?{x{2W=6~`xor#6MOV`?c%|O-F)&B==^o|P6cq27T
z@ei%>ii0bj<ocnY04K_q?KRq+Y)UHEaPV?7^+Y$gxp?A#e$_6bpdfmfFZ(ohtW1d(
zDAvr*GS>aD^IkKUsv$(d>~Ac;b)`uzk?r(8|JXi1z6z<tBM9-5FsG?wv0q1WsHo_q
z>|oZw75i@FZ@eUXbo(t^*yECt({;VFd6t;jd(@<IK~+a7a)%K}VwslrEsqcWMM?Vw
z1guk_`J{pcNt-UOgzA$+ELSP8i2C;JVk%`~Jw~lsE=e(XrdqF}j7=pg{Y~qC&{dxi
z2WuCksdD<WK^|Yl%ByIkHQAxdA^j3~9&s%4%71WwugKyd?4&$^^N(d(Xz)>AQ<vz)
zUtsvjiKn7w3$H)_L-+KT$HUJnr<^N)ute?z52u?|^O=7Di9i!iMU~c~q$T_VYA>_#
z*bzG1Nb?uNF8<;clt#$^g{V#Zu~u8}2S52gHj}!Ir=mU*eoFR7wp*ciG(9ld*7!4U
zIOi`M|44G_FGhXQ#8Xja!IKyN47~9ikAyTsa<~7;7IE;4ifZ32`Ipp?bp@YS<T2hQ
i{v+G}Q%X5Fzj6GX@NNZs9seHz02M`zM^Jf-Xa5V;XD-12

literal 0
HcmV?d00001

diff --git a/docs_new/community/contribute.md b/docs_new/community/contribute.md
new file mode 100644
index 000000000..78c713b18
--- /dev/null
+++ b/docs_new/community/contribute.md
@@ -0,0 +1,181 @@
+---
+title: Contribute
+---
+
+## What contributions?
+
+- **Documentation** contributions are very valuable to us!
+- **Examples.** Show us what you did with Outlines :)
+- **Bug reports** with a minimum working examples in the [issue tracker][issues]
+- **Bug fixes** are always a pleasure to review.
+- **New features**. Please start a new [discussion][discussions], or [come chat with us][discord] beforehand!
+
+Note that the [issue tracker][issues] is only intended for actionable items. In doubt, open a [discussion][discussions] or [come talk to us][discord].
+
+## How to contribute?
+
+### Setup
+
+First, [fork the repository on GitHub](https://github.com/dottxt-ai/outlines/fork) and clone the fork locally:
+
+```shell
+git clone git@github.com/YourUserName/outlines.git
+cd outlines
+```
+
+Create a new virtual environment:
+
+*If you are using `uv`*:
+
+```shell
+uv venv
+source .venv/bin/activate
+alias pip="uv pip" # ... or just remember to prepend any pip command with uv in the rest of this guide
+```
+
+*If you are using `venv`*:
+
+```shell
+python -m venv .venv
+source .venv/bin/activate
+```
+
+*If you are using `conda`*:
+
+```shell
+conda env create -f environment.yml
+```
+
+Then install the dependencies in editable mode, and install the `pre-commit` hooks:
+
+```shell
+pip install -e ".[test]"
+pre-commit install
+```
+If you own a GPU and want to run the vLLM tests you will have to run:
+
+```shell
+pip install -e ".[test-gpu]"
+```
+
+instead.
+
+Outlines provides optional dependencies for different supported backends, which you can install with
+
+```shell
+pip install ".[vllm]"
+```
+
+A list of supported optional dependencies can be found in the [installation guide](/installation).
+
+### Using VSCode DevContainer / GitHub Codespaces
+
+If you want a fully pre-configured development environment, you can use VSCode DevContainers or GitHub Codespaces.
+
+#### VSCode DevContainer
+
+1. Ensure that the [Docker](https://www.docker.com/get-started/) daemon is running on your machine.
+2. Install the [Dev Containers](https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.remote-containers) extension in VSCode.
+3. Open the Outlines repository in VSCode. When prompted, **Reopen in Container** (or press `F1` and select "Remote-Containers: Reopen in Container").
+4. Run the normal setup steps. Your environment will not complain about missing system dependencies!
+
+#### GitHub Codespaces
+
+1. Navigate to the Outlines repository on GitHub.
+2. Click on the **Code** button and select the **Codespaces** tab.
+3. Click **Create codespace on main** (or another branch you are working on).
+4. GitHub will launch a pre-configured cloud development environment.
+
+You will not have access to a GPU, but you'll be able to make basic contributions to the project on the go while using a fully featured web-based IDE.
+
+### Before pushing your code
+
+Run the tests:
+
+```shell
+pytest
+```
+
+And run the code style checks:
+
+```shell
+pre-commit run --all-files
+```
+
+### Benchmarking
+
+Outlines uses [asv](https://asv.readthedocs.io) for automated benchmark testing. Benchmarks are run automatically before pull requests are merged to prevent performance degradation.
+
+You can run the benchmark test suite locally with the following command:
+
+```shell
+asv run --config benchmarks/asv.conf.json
+```
+
+Caveats:
+
+- If you're on a device with CUDA, you must add the argument `--launch-method spawn`
+- Uncommitted code will not be benchmarked, you must first commit your changes.
+
+#### Run a specific test:
+
+```shell
+asv run --config benchmarks/asv.conf.json -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
+```
+
+#### Profile a specific test:
+
+```shell
+asv run --config benchmarks/asv.conf.json --profile -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
+```
+
+#### Compare to `origin/main`
+
+```shell
+get fetch origin
+asv continuous origin/main HEAD --config benchmarks/asv.conf.json
+```
+
+#### ASV PR Behavior
+
+- **View ASV Benchmark Results:** Open the workflow, view `BENCHMARK RESULTS` section.
+- Merging is blocked unless benchmarks are run for the latest commit.
+- Benchmarks fail if performance degrades by more than 10% for any individual benchmark.
+- The "Benchmark PR" workflow runs when it is manually dispatched, or if the `run_benchmarks` label is added to the PR they run for every commit.
+
+### Contribute to the documentation
+
+To work on the *documentation* you will need to install the related dependencies:
+
+```shell
+pip install -r requirements-doc.txt
+```
+
+To build the documentation and serve it locally, run the following command in the repository's root folder:
+
+```shell
+mkdocs serve
+```
+
+By following the instruction you will be able to view the documentation locally.
+It will be updated every time you make a change.
+
+## Open a Pull Request
+
+Create a new branch on your fork, commit and push the changes:
+
+```shell
+git checkout -b new-branch
+git add .
+git commit -m "Changes I made"
+git push origin new-branch
+```
+
+Then you can [open a pull request][pull-requests] on GitHub. It should prompt you to do so. Every subsequent change that you make on your branch will update the pull request.
+
+Do not hesitate to open a draft PR before your contribution is ready, especially if you have questions and/or need feedback. If you need help, come tell us on [Discord][discord].
+
+[discord]: https://discord.gg/R9DSu34mGd
+[discussions]: https://github.com/dottxt-ai/outlines/discussions
+[issues]: https://github.com/dottxt-ai/outlines/issues
+[pull-requests]: https://github.com/dottxt-ai/outlines/pulls
diff --git a/docs_new/community/examples.md b/docs_new/community/examples.md
new file mode 100644
index 000000000..813bd4834
--- /dev/null
+++ b/docs_new/community/examples.md
@@ -0,0 +1,19 @@
+# Community projects and articles
+
+Publishing examples and articles about Outlines are a meaningful way to contribute to the community. Here is a list of projects we are aware of. Drop us a line if we forgot yours!
+
+[MMSG](https://github.com/leloykun/mmsg) is a Python library for generating interleaved text and image content in a structured format you can directly pass to downstream APIs.
+
+[Multimodal Structured Generation: CVPR's 2nd MMFM Challenge Technical Report](https://arxiv.org/abs/2406.11403) shows that Structured Generation can outperform finetuning, and maybe even multimodality, in document-image understanding tasks as part of CVPR's 2nd MMFM Challenge.
+
+[Chess LLM Arena](https://huggingface.co/spaces/mlabonne/chessllm) is a HuggingFace Space where you can make LLMs compete in a chess match.
+
+[LLM Data Gen](https://huggingface.co/spaces/lhoestq/LLM_DataGen) is a HuggingFace Space that generates synthetic dataset files in JSONLines format.
+
+[Fast, High-Fidelity LLM Decoding with Regex Constraints ](https://vivien000.github.io/blog/journal/llm-decoding-with-regex-constraints.html) presents an efficient alternative to Outlines's structured generation.
+
+[gigax](https://github.com/GigaxGames/gigax) is an Open-Source library that allows to create real-time LLM-powered NPCs for video games.
+
+[Improving Prompt Consistency with Structured Generations](https://huggingface.co/blog/evaluation-structured-outputs) shows how structured generation can improve consistency of evaluation runs by reducing sensitivity to changes in prompt format.
+
+[AskNews](https://asknews.app) is a news curation service processing 300k news articles per day in a structured way, with Outlines.
diff --git a/docs_new/community/feedback.md b/docs_new/community/feedback.md
new file mode 100644
index 000000000..033e91870
--- /dev/null
+++ b/docs_new/community/feedback.md
@@ -0,0 +1,81 @@
+---
+title: Feedback
+---
+
+# Feedback
+
+If Outlines has been helpful to you, let us know on [Discord][discord] or give us a shoutout on [Twitter][twitter]! It's always heartwarming ❤️
+
+
+<head>
+  <!-- From Marvin AI's documentation -->
+  <!-- Their library is also awesome -->
+  <!-- https://www.askmarvin.ai/ -->
+  <style>
+    .tweet-masonry {
+      column-count: 2;
+      column-gap: 20px;
+      padding: 20px;
+    }
+
+    .twitter-tweet {
+      display: inline-block;
+      width: 100%;
+      margin-bottom: 20px;
+      margin-top: 0px !important;
+      break-inside: avoid;
+    }
+
+    @media (max-width: 600px) {
+      .tweet-masonry {
+        column-count: 1;
+      }
+    }
+  </style>
+</head>
+<body>
+
+<div class="tweet-masonry">
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">I am once again reminding you that structured extraction using LLMs is going to transform every single industry in the next 10 years <a href="https://t.co/xQ3tcWnrZ8">https://t.co/xQ3tcWnrZ8</a></p>&mdash; Sam Hogan (@0xSamHogan) <a href="https://twitter.com/0xSamHogan/status/1780637917737816323?ref_src=twsrc%5Etfw">April 17, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">outline&#39;s growth is insane, using is an understatement! <a href="https://t.co/rHCNWhZdCs">https://t.co/rHCNWhZdCs</a></p>&mdash; jason liu (@jxnlco) <a href="https://twitter.com/jxnlco/status/1780618454040797554?ref_src=twsrc%5Etfw">April 17, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Outlines is an amazing lib and more popular than <a href="https://twitter.com/remilouf?ref_src=twsrc%5Etfw">@remilouf</a>’s modesty will admit. <a href="https://t.co/DfHbMPIlX1">https://t.co/DfHbMPIlX1</a> <a href="https://t.co/mDHIWJrD0C">https://t.co/mDHIWJrD0C</a></p>&mdash; Delip Rao e/σ (@deliprao) <a href="https://twitter.com/deliprao/status/1780780217180598377?ref_src=twsrc%5Etfw">April 18, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Impressive implementation of a true regex / json / grammar guided text generation <a href="https://t.co/RX5RVYaVIx">pic.twitter.com/RX5RVYaVIx</a></p>&mdash; Rohan Paul (@rohanpaul_ai) <a href="https://twitter.com/rohanpaul_ai/status/1741099984299135403?ref_src=twsrc%5Etfw">December 30, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Most underrated Github Repo in AI + LLM JSON guided Generation: <a href="https://t.co/lSB8KIet1H">https://t.co/lSB8KIet1H</a></p>&mdash; 🎙Jean-Louis Queguiner (@JiliJeanlouis) <a href="https://twitter.com/JiliJeanlouis/status/1736857292581093706?ref_src=twsrc%5Etfw">December 18, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Nice and useful. <a href="https://t.co/LX72AE0lgt">https://t.co/LX72AE0lgt</a></p>&mdash; Dan Roy (@roydanroy) <a href="https://twitter.com/roydanroy/status/1691556956941525458?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">HUGE dub for open source AI <a href="https://t.co/bYKuiEUZ1j">https://t.co/bYKuiEUZ1j</a></p>&mdash; kenneth 🖇 (@k3nnethfrancis) <a href="https://twitter.com/k3nnethfrancis/status/1691304781732843521?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">This is amazing - glad to see more outp guidance modules! <br><br>Will try this out soon I&#39;m wondering how they translate from regex automatons to token boundaries<br><br>Also why Open Source will succeed. Even today I don&#39;t see any guided output functionality from the big providers. <a href="https://t.co/Ity2H25Klf">https://t.co/Ity2H25Klf</a></p>&mdash; Hrishi (@hrishioa) <a href="https://twitter.com/hrishioa/status/1691181499671080960?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Outlines - a library to help LLM developers guide text generation in a fast and reliable way.<br><br>&quot;Provides generation methods that guarantee that the output will match a regular expressions, or follow a JSON schema.&quot;<br><br>Need to check this out. Reliable JSON output is a common use… <a href="https://t.co/Bkbh8vKogN">pic.twitter.com/Bkbh8vKogN</a></p>&mdash; elvis (@omarsar0) <a href="https://twitter.com/omarsar0/status/1691179888214966273?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Woah this is cool! Makes open source models more usable.<br><br>Give any LLM Function Call capability (and more) with Outlines: <a href="https://t.co/PtPykR5ZGR">https://t.co/PtPykR5ZGR</a> <a href="https://t.co/RRQjWHnIxv">https://t.co/RRQjWHnIxv</a> <a href="https://t.co/BwNnH8SMwv">pic.twitter.com/BwNnH8SMwv</a></p>&mdash; Yohei (@yoheinakajima) <a href="https://twitter.com/yoheinakajima/status/1691231912466223104?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">This is awesome! Being able to guarantee the output&#39;s structure unblocks so many applications. This is a great milestone and a fundamental building block for more advanced AI apps. <a href="https://t.co/WdwMOc7hE8">https://t.co/WdwMOc7hE8</a></p>&mdash; Guilherme Castro (@skastr052) <a href="https://twitter.com/skastr052/status/1691239359494619136?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Juggling with the unpredictable outputs of ChatGPT API lately while building my product. 😓 <br><br>Tried prompt engineering to channel its wisdom into a neat JSON, but it&#39;s like asking a cat to fetch. 🐱<br><br>Luckily, stumbled upon &quot;Outlines&quot; – looks like a promising way to tame the LLM… <a href="https://t.co/oYQ6q8exAS">pic.twitter.com/oYQ6q8exAS</a></p>&mdash; Charlie (@14435635Sun) <a href="https://twitter.com/14435635Sun/status/1691439342689095680?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+
+<blockquote class="twitter-tweet"><p lang="en" dir="ltr">A complex system of LLM input-outputs interacting with non-LLM agents and models benefits immeasurably from structured outputs. The outlines package saves so much time, <a href="https://t.co/NhVQ6NpKDR">https://t.co/NhVQ6NpKDR</a></p>&mdash; Amir Sani (@amirsani) <a href="https://twitter.com/amirsani/status/1728734266568376433?ref_src=twsrc%5Etfw">November 26, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
+</div>
+</body>
+</html>
+
+# Let us know!
+
+We highly value the insights of our users, and we would love to hear from you. If you are using Outlines for your projects and would like to share your experience with us, let's connect:
+
+- What are you building with it?
+- What do you like about it?
+- What challenges are you facing?
+- What do you think could be improved?
+
+To schedule an appointment follow [this link](https://cal.com/dottxt/outlines). This is exclusively intended to share your experience, please go on [Discord][discord] or [GitHub](https://github.com/dottxt-ai/outlines/discussions) for support.
+
+[discord]: https://discord.gg/UppQmhEpe8
+[twitter]: https://twitter.com/dottxtai
diff --git a/docs_new/community/index.md b/docs_new/community/index.md
new file mode 100644
index 000000000..517bf59c5
--- /dev/null
+++ b/docs_new/community/index.md
@@ -0,0 +1,9 @@
+# Community
+
+Outlines exists for a community of users who believe software doesn't need to be complicated. Who share the same passion for Large Language Models but don't want to compromise on robustness. Together, we are bringing these powerful models back to the world of software.
+
+## Connect on Discord
+
+The Outlines community lives on our Discord server. There you can ask questions, share ideas or just chat with people like you. Don't be a stranger and [join us][discord].
+
+[discord]: https://discord.gg/UppQmhEpe8
diff --git a/docs_new/community/versioning.md b/docs_new/community/versioning.md
new file mode 100644
index 000000000..a6e12f218
--- /dev/null
+++ b/docs_new/community/versioning.md
@@ -0,0 +1,30 @@
+---
+title: Versioning Guide
+---
+
+# Versioning Guide
+
+
+The Outlines project follows a structured versioning scheme designed to provide clarity and minimize risk for downstream dependents.
+
+Each part of the version number (`major.minor.patch`) conveys information about the nature and impact of the changes included in the release.
+
+- **Major Releases** includes compatibility-breaking changes to core interfaces, such as `LogitsProcessor`s and `Guides`.
+- **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatability with existing use of core interfaces.
+- **Patch Releases** address bug fixes and incorporate low-risk changes to improve stability and performance.
+
+!!! note "Breaking Changes"
+
+    Outlines v1.0 introduced several breaking changes to the core interface. See [the migration guide](/user_guide/migration) for more details.
+
+## Releases
+
+Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/dottxt-ai/outlines/releases).
+
+## Version Pinning Recommendations
+
+Here are our recommendations for managing dependencies on the Outlines package:
+
+**Small, Risk-Tolerant Projects:** Pin to a specific major version.
+
+**Large, Conservative Projects:** Pin to a specific minor version.
diff --git a/docs_new/core_concepts.md b/docs_new/core_concepts.md
new file mode 100644
index 000000000..7f53983ec
--- /dev/null
+++ b/docs_new/core_concepts.md
@@ -0,0 +1,13 @@
+---
+title: Core concepts
+---
+
+# Core concepts
+
+Coming soon. This will document various concepts at a high level, so users can understand Outlines before diving into specific implementations.
+
+1. Constrained decoding, tokens, and the basics of logit biasing
+2. Different ways to define output structure (regex, JSON schema, Pydantic models, context-free grammars)
+3. How finite state machines are used to guarantee output structure
+4. `Generator`, `Application`, `Template`,
+5. Prompt engineering vs. structured generation
diff --git a/docs_new/examples/atomic_caption.md b/docs_new/examples/atomic_caption.md
new file mode 100644
index 000000000..a2b43e5e2
--- /dev/null
+++ b/docs_new/examples/atomic_caption.md
@@ -0,0 +1,189 @@
+# Vision-Language Models with Outlines
+This guide demonstrates how to use Outlines with vision-language models, leveraging the new transformers_vision module. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
+
+We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
+
+## Setup
+First, we need to install the necessary dependencies. In addition to Outlines, we'll need to install the transformers library and any specific requirements for the vision-language model we'll be using.
+
+```bash
+pip install outlines transformers torch
+```
+
+### Initializing the Model
+We'll use the transformers_vision function to initialize our vision-language model. This function is specifically designed to handle models that can process both text and image inputs. Today we'll be using the Pixtral model with the llama tokenizer. (Currently the mistral tokenizer is pending support).
+
+```python
+import torch
+from transformers import (
+    LlavaForConditionalGeneration,
+)
+model_name="mistral-community/pixtral-12b" # original magnet model is able to be loaded without issue
+model_class=LlavaForConditionalGeneration
+
+def get_vision_model(model_name: str, model_class: VisionModel):
+    model_kwargs = {
+        "torch_dtype": torch.bfloat16,
+        "attn_implementation": "flash_attention_2",
+        "device_map": "auto",
+    }
+    processor_kwargs = {
+        "device": "cuda",
+    }
+
+    model = outlines.models.transformers_vision(
+        model.model_name,
+        model_class=model.model_class,
+        model_kwargs=model_kwargs,
+        processor_kwargs=processor_kwargs,
+    )
+    return model
+model = get_vision_model(model_name, model_class)
+```
+
+### Defining the Schema
+Next, we'll define a schema for the output we expect from our vision-language model. This schema will help structure the model's responses.
+
+
+```python
+from pydantic import BaseModel, Field, confloat, constr
+from pydantic.types import StringConstraints, PositiveFloat
+from typing import List
+from typing_extensions import Annotated
+
+from enum import StrEnum
+class TagType(StrEnum):
+    ENTITY = "Entity"
+    RELATIONSHIP = "Relationship"
+    STYLE = "Style"
+    ATTRIBUTE = "Attribute"
+    COMPOSITION = "Composition"
+    CONTEXTUAL = "Contextual"
+    TECHNICAL = "Technical"
+    SEMANTIC = "Semantic"
+
+class ImageTag(BaseModel):
+    tag: Annotated[
+        constr(min_length=1, max_length=30),
+        Field(
+            description=(
+                "Descriptive keyword or phrase representing the tag."
+            )
+        )
+    ]
+    category: TagType
+    confidence: Annotated[
+        confloat(le=1.0),
+        Field(
+            description=(
+                "Confidence score for the tag, between 0 (exclusive) and 1 (inclusive)."
+            )
+        )
+    ]
+
+class ImageData(BaseModel):
+    tags_list: List[ImageTag] = Field(..., min_items=8, max_items=20)
+    short_caption: Annotated[str, StringConstraints(min_length=10, max_length=150)]
+    dense_caption: Annotated[str, StringConstraints(min_length=100, max_length=2048)]
+
+image_data_generator = outlines.generate.json(model, ImageData)
+```
+
+This schema defines the structure for image tags, including categories like Entity, Relationship, Style, etc., as well as short and dense captions.
+
+### Preparing the Prompt
+
+We'll create a prompt that instructs the model on how to analyze the image and generate the structured output:
+
+
+```python
+pixtral_instruction = """
+<s>[INST]
+<Task>You are a structured image analysis agent. Generate comprehensive tag list, caption, and dense caption for an image classification system.</Task>
+<TagCategories requirement="You should generate a minimum of 1 tag for each category." confidence="Confidence score for the tag, between 0 (exclusive) and 1 (inclusive).">
+- Entity : The content of the image, including the objects, people, and other elements.
+- Relationship : The relationships between the entities in the image.
+- Style : The style of the image, including the color, lighting, and other stylistic elements.
+- Attribute : The most important attributes of the entities and relationships in the image.
+- Composition : The composition of the image, including the arrangement of elements.
+- Contextual : The contextual elements of the image, including the background, foreground, and other elements.
+- Technical : The technical elements of the image, including the camera angle, lighting, and other technical details.
+- Semantic : The semantic elements of the image, including the meaning of the image, the symbols, and other semantic details.
+<Examples note="These show the expected format as an abstraction.">
+{
+  "tags_list": [
+    {
+      "tag": "subject 1",
+      "category": "Entity",
+      "confidence": 0.98
+    },
+    {
+      "tag": "subject 2",
+      "category": "Entity",
+      "confidence": 0.95
+    },
+    {
+      "tag": "subject 1 runs from subject 2",
+      "category": "Relationship",
+      "confidence": 0.90
+    },
+   }
+</Examples>
+</TagCategories>
+<ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
+<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG][/INST]
+""".strip()
+```
+
+This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required.
+
+### Generating Structured Output
+Now we can use our model to generate structured output based on an input image:
+
+```python
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+image_url="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
+image= img_from_url(image_url)
+result = image_data_generator(
+    pixtral_instruction,
+    [image]
+)
+print(result)
+```
+
+This code loads an image from a URL, passes it to our vision-language model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
+
+```json
+{'tags_list': [{'tag': 'astronaut',
+   'category': <TagType.ENTITY: 'Entity'>,
+   'confidence': 0.99},
+  {'tag': 'moon', 'category': <TagType.ENTITY: 'Entity'>, 'confidence': 0.98},
+  {'tag': 'space suit',
+   'category': <TagType.ATTRIBUTE: 'Attribute'>,
+   'confidence': 0.97},
+  {'tag': 'lunar module',
+   'category': <TagType.ENTITY: 'Entity'>,
+   'confidence': 0.95},
+  {'tag': 'shadow of astronaut',
+   'category': <TagType.COMPOSITION: 'Composition'>,
+   'confidence': 0.95},
+  {'tag': 'footprints in moon dust',
+   'category': <TagType.CONTEXTUAL: 'Contextual'>,
+   'confidence': 0.93},
+  {'tag': 'low angle shot',
+   'category': <TagType.TECHNICAL: 'Technical'>,
+   'confidence': 0.92},
+  {'tag': 'human first steps on the moon',
+   'category': <TagType.SEMANTIC: 'Semantic'>,
+   'confidence': 0.95}],
+ 'short_caption': 'First man on the Moon',
+ 'dense_caption': "The figure clad in a pristine white space suit, emblazoned with the American flag, stands powerfully on the moon's desolate and rocky surface. The lunar module, a workhorse of space engineering, looms in the background, its metallic legs sinking slightly into the dust where footprints and tracks from the mission's journey are clearly visible. The photograph captures the astronaut from a low angle, emphasizing his imposing presence against the desolate lunar backdrop. The stark contrast between the blacks and whiteslicks of lost light and shadow adds dramatic depth to this seminal moment in human achievement."}
+```
+
+## Conclusion
+The transformers_vision module in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
+
+By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
diff --git a/docs_new/examples/chain_of_density.md b/docs_new/examples/chain_of_density.md
new file mode 100644
index 000000000..10bda9771
--- /dev/null
+++ b/docs_new/examples/chain_of_density.md
@@ -0,0 +1,126 @@
+# Summarize documents using Chain of Density prompting
+
+A good summary should be informative, concise and clear. While large language models are generally good at summarizing documents, their summaries tend to be long and contain redundant information; their information density tends to be on the lower end. This is where [chain of Density](https://arxiv.org/abs/2309.04269), a new prompting technique, comes in. In this example we will show how one can implement chain of density with a few lines of code using Outlines, leveraging both Outline's prompt templating and its structured generation capabilities.
+
+The article we will try to summarize is the first three paragraphs of the [Alan Turing page on Wikipedia](https://en.wikipedia.org/wiki/Alan_Turing):
+
+```python
+article = """
+Alan Mathison Turing OBE FRS (/ˈtjʊərɪŋ/; 23 June 1912 – 7 June 1954) was an English mathematician, computer scientist, logician, cryptanalyst, philosopher and theoretical biologist.[5] Turing was highly influential in the development of theoretical computer science, providing a formalisation of the concepts of algorithm and computation with the Turing machine, which can be considered a model of a general-purpose computer.[6][7][8] He is widely considered to be the father of theoretical computer science and artificial intelligence.[9]
+
+Born in Maida Vale, London, Turing was raised in southern England. He graduated at King's College, Cambridge, with a degree in mathematics. Whilst he was a fellow at Cambridge, he published a proof demonstrating that some purely mathematical yes–no questions can never be answered by computation. He defined a Turing machine and proved that the halting problem for Turing machines is undecidable. In 1938, he obtained his PhD from the Department of Mathematics at Princeton University. During the Second World War, Turing worked for the Government Code and Cypher School at Bletchley Park, Britain's codebreaking centre that produced Ultra intelligence. For a time he led Hut 8, the section that was responsible for German naval cryptanalysis. Here, he devised a number of techniques for speeding the breaking of German ciphers, including improvements to the pre-war Polish bomba method, an electromechanical machine that could find settings for the Enigma machine. Turing played a crucial role in cracking intercepted coded messages that enabled the Allies to defeat the Axis powers in many crucial engagements, including the Battle of the Atlantic.[10][11]
+
+After the war, Turing worked at the National Physical Laboratory, where he designed the Automatic Computing Engine, one of the first designs for a stored-program computer. In 1948, Turing joined Max Newman's Computing Machine Laboratory at the Victoria University of Manchester, where he helped develop the Manchester computers[12] and became interested in mathematical biology. He wrote a paper on the chemical basis of morphogenesis[1] and predicted oscillating chemical reactions such as the Belousov–Zhabotinsky reaction, first observed in the 1960s. Despite these accomplishments, Turing was never fully recognised in Britain during his lifetime because much of his work was covered by the Official Secrets Act.[13]
+"""
+```
+
+## How Chain Of Density works
+
+Chain Of Density starts with asking the model to generate a first long and non-specific summary. Then it asks the model to generate 4 extra summaries by proceeding in the following way:
+
+1. Identify 1-3 entities missing in the previous summary;
+2. Add all entities marked as missing in the previous step, while not dropping entities;
+3. Make the summary more concise;
+
+The prompt also asks the model to return a list of JSON objects that contain the missing entities and the new summary. This is where structured generation will come in handy :) The paper provides the prompt and an example:
+
+![Figure 2 in the paper](./images/chain_of_density.png)
+
+We can now implement the prompt provided in the paper:
+
+```python
+from outlines import Template
+
+
+chain_of_density = Template.from_string(
+    """Article: {{ article }}
+
+    You will generate increasingly concise, entity-dense summaries of the above Article.
+
+    Repeat the following 2 steps 5 times.
+
+    Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
+    Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
+
+    A Missing Entity is:
+    - Relevant: to the main story.
+    - Specific: descriptive yet concise (5 words or fewer).
+    - Novel: not in the previous summary.
+    - Faithful: present in the Article.
+    - Anywhere: located anywhere in the Article.
+
+    Guidelines:
+    - The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
+    - Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
+    - Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
+    - The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
+    - Missing entities can appear anywhere in the new summary.
+    - Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
+
+    Remember, use the exact same number of words for each summary.
+
+    Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
+    """
+)
+```
+
+??? Note
+
+    Note that we modified the prompt slightly so it returns a JSON object that contains the summaries, instead of a list of summaries.
+
+
+## Outlines implementation
+
+We will use Outline's JSON-structured generation to ensure that the model's output is consistent with the format specified in the prompt. We start with defining the JSON objects that the model is asked to return using Pydantic. One JSON object that contains a list of `Summary` objects that contain the missing entities and new summary:
+
+```python
+from pydantic import BaseModel, conlist
+
+class Summary(BaseModel):
+    missing_entities: str
+    denser_summary: str
+
+class Summaries(BaseModel):
+    summaries: conlist(Summary, max_length=5, min_length=5)
+```
+
+We now generate the prompt by passing the article we want to summarize to the template. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use JSON-structured generation to generate the summaries:
+
+```python
+model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ")
+
+prompt = chain_of_density(article)
+result = outlines.generate.json(model, Summaries)(prompt)
+```
+
+We can now check the results:
+
+```python
+print(result.model_dump())
+# {'summaries': [
+#     {
+#       'missing_entities': 'English mathematician, cryptanalyst, philosopher',
+#       'denser_summary': 'Alan Mathison Turing was an English mathematician, cryptanalyst, philosopher.'
+#     },
+#     {
+#       'missing_entities': '',
+#       'denser_summary': "Alan Mathison Turing was an English mathematician who was a crucial figure in WW2's Bletchley Park codebreaking centre and designed one of the first computers."
+#     },
+#     {
+#       'missing_entities': 'cryptanalyst, studied, biology, father',
+#       'denser_summary': 'Alan Mathison Turing was an English cryptanalyst, studied theoretical computer science, and contributed to mathematical biology.'
+#     },
+#     {
+#       'missing_entities': 'biology, morphogenesis, chemical',
+#       'denser_summary': 'Alan Mathison Turing was an English cryptanalyst, studied theoretical computer science, and predicted chemical reactions in morphogenesis.
+#     '},
+#     {
+#       'missing_entities': '',
+#       'denser_summary': 'Alan Mathison Turing was an English cryptanalyst, developed computer science, and made strides in mathematical biology research.'
+#       }
+# ]}
+```
+
+Not bad, considering we used a smallish model to generate the summary! Chain of Density seems to be a very effective prompting technique to generate dense summaries, even with small quantized models. Its implementation in Outlines is also very short.
+
+Note that this is the first article I tried and it worked out of the box. Try it out on other articles, and please share the results on Twitter, or by opening [a new discussion](https://github.com/dottxt-ai/outlines/discussions/categories/show-and-tell) on the Outlines repository!
diff --git a/docs_new/examples/chain_of_thought.md b/docs_new/examples/chain_of_thought.md
new file mode 100644
index 000000000..5ac18af25
--- /dev/null
+++ b/docs_new/examples/chain_of_thought.md
@@ -0,0 +1,138 @@
+# Chain of thought
+
+
+Chain of thought is a prompting technique introduced in the paper ["Chain-of-Thought Prompting Elicits Reasoning in Large Language Models"](https://arxiv.org/abs/2201.11903) where throught prompting the authors generate a series of intermediate reasoning steps which improves the ability of LLMs to perform complex reasoning.
+
+In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to apply chain of thought through structured output.
+
+We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```bash
+pip install llama-cpp-python
+```
+
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+```python
+import llama_cpp
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
+```
+
+??? note "(Optional) Store the model weights in a custom folder"
+
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
+
+## Chain of thought
+
+We first define our Pydantic class for a reasoning step:
+
+```python
+from pydantic import BaseModel, Field
+
+class Reasoning_Step(BaseModel):
+    reasoning_step: str = Field(..., description="Reasoning step")
+```
+
+We then define the Pydantic class for reasoning which will consist on a list of reasoning steps and a conclusion, and we get its JSON schema:
+
+```python
+from typing import List
+
+class Reasoning(BaseModel):
+    reasoning: List[Reasoning_Step] = Field(..., description="List of reasoning steps")
+    conclusion: str = Field(..., description="Conclusion")
+
+json_schema = Reasoning.model_json_schema()
+```
+
+We could generate a response using the json schema but for a change we will use the regex:
+
+```python
+from outlines.fsm.json_schema import convert_json_schema_to_str
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+schema_str = convert_json_schema_to_str(json_schema=json_schema)
+regex_str = build_regex_from_schema(schema_str)
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
+
+```python
+def generate_hermes_prompt(user_prompt):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON "
+        f"Here's the json schema you must adhere to:\n<schema>\n{json_schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + user_prompt
+        + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
+```
+
+For a given user prompt:
+
+```python
+user_prompt = "9.11 and 9.9 -- which is bigger?"
+```
+
+we can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+
+```python
+generator = generate.regex(model, regex_str)
+prompt = generate_hermes_prompt(user_prompt)
+response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+```
+
+We obtain a series of intermediate reasoning steps as well as the conclusion:
+
+```python
+import json
+
+json_response = json.loads(response)
+
+print(json_response["reasoning"])
+print(json_response["conclusion"])
+# [{'reasoning_step': 'Both 9.11 and 9.9 are decimal numbers.'},
+#  {'reasoning_step': 'When comparing decimal numbers, we look at the numbers after the decimal point.'},
+#  {'reasoning_step': 'In this case, 9.11 has the number 1 after the decimal point, while 9.9 has the number 9.'},
+#  {'reasoning_step': 'Since 1 is greater than 9, 9.11 is greater than 9.9.'}]
+# '9.11 is bigger.'
+```
+
+We notice that the 4th reasoning step is wrong ``Since 1 is greater than 9, 9.11 is greater than 9.9.'', so we should probably give the model some examples for this particular task.
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_new/examples/classification.md b/docs_new/examples/classification.md
new file mode 100644
index 000000000..023e4543a
--- /dev/null
+++ b/docs_new/examples/classification.md
@@ -0,0 +1,106 @@
+# Classification
+
+Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
+
+
+This tutorial shows how one can implement multi-label classification using Outlines. We will use two functionalities of the library: `generate.choice` and `generate.json`.
+
+As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
+
+```python
+import outlines
+
+model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+```
+
+We will use the following prompt template:
+
+```python
+from outlines import Template
+
+
+customer_support = Template.from_string(
+    """You are an experienced customer success manager.
+
+    Given a request from a client, you need to determine when the
+    request is urgent using the label "URGENT" or when it can wait
+    a little with the label "STANDARD".
+
+    # Examples
+
+    Request: "How are you?"
+    Label: STANDARD
+
+    Request: "I need this fixed immediately!"
+    Label: URGENT
+
+    # TASK
+
+    Request: {{ request }}
+    Label: """
+)
+```
+
+## Choosing between multiple choices
+
+Outlines provides a shortcut to do multi-label classification, using the `outlines.generate.choice` function to initialize a generator. Outlines uses multinomial sampling by default, here we will use the greedy sampler to get the label with the highest probability:
+
+```python
+from outlines.samplers import greedy
+
+generator = outlines.generate.choice(model, ["URGENT", "STANDARD"], sampler=greedy())
+```
+Outlines supports batched requests, so we will pass two requests to the model:
+
+```python
+requests = [
+    "My hair is one fire! Please help me!!!",
+    "Just wanted to say hi"
+]
+
+prompts = [customer_support(request) for request in requests]
+```
+
+We can now asks the model to classify the requests:
+
+```python
+labels = generator(prompts)
+print(labels)
+# ['URGENT', 'STANDARD']
+```
+
+Now, you might be in a hurry and don't want to wait until the model finishes completion. After all, you only need to see the first letter of the response to know whether the request is urgent or standard. You can instead stream the response:
+
+```python
+tokens = generator.stream(prompts)
+labels = ["URGENT" if "U" in token else "STANDARD" for token in next(tokens)]
+print(labels)
+# ['URGENT', 'STANDARD']
+```
+
+## Using JSON-structured generation
+
+Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
+
+```python
+from enum import Enum
+from pydantic import BaseModel
+
+
+class Label(str, Enum):
+    urgent = "URGENT"
+    standard = "STANDARD"
+
+
+class Classification(BaseModel):
+    label: Label
+```
+
+and we can use `generate.json` by passing this Pydantic model we just defined, and call the generator:
+
+```python
+generator = outlines.generate.json(model, Classification, sampler=greedy())
+labels = generator(prompts)
+print(labels)
+# [Classification(label=<Label.urgent: 'URGENT'>), Classification(label=<Label.standard: 'STANDARD'>)]
+```
diff --git a/docs_new/examples/dating_profiles.md b/docs_new/examples/dating_profiles.md
new file mode 100644
index 000000000..1cbea556c
--- /dev/null
+++ b/docs_new/examples/dating_profiles.md
@@ -0,0 +1,230 @@
+# Generate a synthetic dating profile from a description
+
+In this example we will see how we can use Outlines to generate synthetic data for a dating application. This example was originally contributed by [Vibhor Kumar](https://github.com/veezbo).
+
+```python
+from dataclasses import dataclass
+from enum import Enum
+
+import torch
+import transformers
+from pydantic import BaseModel, conlist, constr
+
+import outlines
+```
+
+## Defining the profile with Pydantic
+
+Here a dating profile will consist in a biography, a job, a list of interests and two question-answer pairs. The questions are written in advance by the team, and the users are asked to provide an answer:
+
+```python
+class QuestionChoice(str, Enum):
+    A = "The key to my heart is"
+    B = "The first item on my bucket list is"
+    C = "Perks of dating me"
+    D = "Message me if you also love"
+    E = "People would describe me as"
+    F = "I can beat you in a game of"
+
+@dataclass
+class QuestionAnswer:
+    question: QuestionChoice
+    answer: str
+```
+
+Users need to provide a short biography, with a minimum of 10 and a maximum of 300 characters. The application also limits job descriptions to 50 characters. In addition to the question-answer pairs, the user is required to provide a list of between 1 and 5 interests:
+
+```python
+class DatingProfile(BaseModel):
+    bio: constr(str, min_length=10, max_length=300)
+    job: constr(str, max_lengt=50)
+    interests: conlist(str, min_length=1, max_length=5)  # type: ignore
+    qna1: QuestionAnswer
+    qna2: QuestionAnswer
+```
+
+## Prompt template and examples
+
+We will ask the model to generate profiles from a high-level description:
+
+```python
+@dataclass
+class Example:
+    description: str
+    profile: DatingProfile
+```
+
+We will use Outlines' prompt templating abilities to generate the prompt for us. This help clearly separate the general prompting logic from what is specific to an example.
+
+```python
+from outlines import Template
+
+dating_profile_prompt = Template.from_string(
+    """
+    You are a world-renowned matchmaker who understands the modern dating
+    market. Your job is to generate dating app profiles for male clients
+    interested in women based on a provided description. The profiles should be
+    authentic, show off their strengths, and maximize their likelihood of
+    getting matches on dating apps.  Here are some examples of past clients that
+    you have successfully created profiles for:
+
+    {% for example in examples %}
+    Description:
+    {{ example.description }}
+    Profile:
+    {{ example.profile }}
+    {% endfor %}
+
+    Here is the new client who you need to create a profile for:
+    Description: {{ description }}
+    Profile:
+    """
+)
+```
+
+We will provide the model with several few-shot examples:
+
+```python
+samples: list[Example] = [
+    Example(
+        description="I'm an author and former professional soccer player living in Seattle who publishes popular fiction books. A typical day for me starts by hanging out with my cat, drinking a coffee, and reading as much as I can in a few hours. Then, I'll prepare a quick smoothie before starting to write for a few hours, take a break with soccer or running a few miles, and finally meet friends for dinner at a new, hip restaurant in the evening. Sometimes we go axe-throwing afterwards, or play poker, or watch a comedy show, or visit a dive bar. On my vacations, I travel extensively to countries South America, Europe, and Asia, with the goal of visiting them all!",
+        profile=DatingProfile(
+            bio="Adventurer, dreamer, author, and soccer enthusiast. Life’s too short to waste time so I make the most of each day by exploring new places and playing with my friends on the pitch. What’s your favorite way to get out and have fun?",
+            job="Famous Soccer Player -> Famous Author",
+            interests=["Soccer", "Travel", "Friends", "Books", "Fluffy Animals"],
+            qna1=QuestionAnswer(
+                question=QuestionChoice.B, answer="swim in all seven oceans!"
+            ),
+            qna2=QuestionAnswer(
+                question=QuestionChoice.E,
+                answer="fun-loving, adventurous, and a little bit crazy",
+            ),
+        ),
+    ),
+    Example(
+        description="I run my company and build houses for a living. I'm a big fan of the outdoors and love to go hiking, camping, and fishing. I don't like video games, but do like to watch movies. My love language is home-cooked food, and I'm looking for someone who isn't afraid to get their hands dirty.",
+        profile=DatingProfile(
+            bio="If you're looking for a Montana man who loves to get outdoors and hunt, and who's in-tune with his masculinity then I'm your guy!",
+            job="House Construction Manager / Entrepreneur",
+            interests=["Hunting", "Hiking", "The outdoors", "Home-cooked food"],
+            qna1=QuestionAnswer(question=QuestionChoice.A, answer="food made at home"),
+            qna2=QuestionAnswer(
+                question=QuestionChoice.C,
+                answer="having a man in your life who can fix anything",
+            ),
+        ),
+    ),
+    Example(
+        description="I run my own Youtube channel with 10M subscribers. I love working with kids, and my audience skews pretty young too. In my free time, I play Fortnite and Roblox. I'm looking for someone who is also a gamer and likes to have fun. I'm learning Japanese in my free time as well as how to cook.",
+        profile=DatingProfile(
+            bio="Easy on the eyes (find me on Youtube!) and great with kids. What more do you need?",
+            job="Youtuber 10M+ subscribers",
+            interests=["Kids", "Gaming", "Japanese"],
+            qna1=QuestionAnswer(question=QuestionChoice.D, answer="anime and gaming!"),
+            qna2=QuestionAnswer(question=QuestionChoice.F, answer="Fortnite, gg ez"),
+        ),
+    ),
+]
+```
+
+## Load the model
+
+We will use Mosaic's MPT-7B model (requires 13GB of GPU memory) which can fit on a single GPU with a reasonable context window. We initialize it with Outlines:
+
+```python
+config = transformers.AutoConfig.from_pretrained(
+    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
+)
+config.init_device = "meta"
+model = outlines.models.transformers(
+    model_name="mosaicml/mpt-7b-8k-instruct",
+    device="cuda",
+    model_kwargs={
+        "config": config,
+        "trust_remote_code": True,
+        "torch_dtype": torch.bfloat16,
+        "device_map": {"": 0},
+    },
+)
+```
+
+## JSON-structured generation of profiles
+
+We will now generate a dating profile from a textual description of oneself:
+
+``` python
+new_description = """I'm a laid-back lawyer who spends a lot of his free-time
+gaming. I work in a corporate office, but ended up here after the start-up  I
+cofounded got acquired, so still play ping pong with my cool coworkers every
+day.  I have a bar at home where I make cocktails, which is great for
+entertaining  friends. I secretly like to wear suits and get a new one tailored
+every few  months. I also like weddings because I get to wear those suits, and
+it's  a good excuse for a date. I watch the latest series because I'm paying,
+with my hard-earned money, for every streaming service."""
+
+prompt = dating_profile_prompt(new_description, samples)
+profile = outlines.generate.json(model, DatingProfile)(prompt)
+parsed_profile = DatingProfile.model_validate_json(profile)
+```
+
+## Results
+
+Here are a couple of results:
+
+```json
+{
+    "bio": """I'm an ambitious lawyer with a casual and fashionable style. I love
+    games and sports, but my true passion is preparing refreshing cocktails at
+    home and dressing to the nines at weddings. I'm currently looking for a woman
+    to show a good time to and get a kiss on the opulent suit I just had made.
+    Send resume to this inbox.""",
+    "job": "Lawyer",
+    "interests":
+    [
+        "Stylish guys",
+        "Gaming",
+        "Ping pong",
+        "Cocktails",
+        "Weddings"
+    ],
+    "qna1":
+    {
+        "question": "The first item on my bucket list is",
+        "answer": "be married and have a family."
+    },
+    "qna2":
+    {
+        "question": "People would describe me as",
+        "answer": "charming, stylish, and funny."
+    }
+}
+```
+
+```json
+{
+    "bio": """I’m a sexy lawyer with time on my hands. I love to game and
+    play ping pong, but the real reason you should swipe to the right
+    is because I look great in a suit. Who doesn’t love a man in a
+    suit? Just saying. Send me a message if you think it’s time to take
+    your dating life to the next level.""",
+    "job": "Lawyer",
+    "interests":
+    [
+        "Gaming",
+        "Ping Pong",
+        "Tailored Suits",
+        "Weddings",
+        "Streaming Services"
+    ],
+    "qna1":
+    {
+        "question": "The first item on my bucket list is",
+        "answer": "simulate space but stay alive for as long as possible"
+    },
+    "qna2":
+    {
+        "question": "People would describe me as",
+        "answer": "easy-going, a little nerdy but with a mature essence"
+    }
+}
+```
diff --git a/docs_new/examples/deploy-using-bentoml.md b/docs_new/examples/deploy-using-bentoml.md
new file mode 100644
index 000000000..2926df0f6
--- /dev/null
+++ b/docs_new/examples/deploy-using-bentoml.md
@@ -0,0 +1,224 @@
+# Run Outlines using BentoML
+
+[BentoML](https://github.com/bentoml/BentoML) is an open-source model serving library for building performant and scalable AI applications with Python. It comes with tools that you need for serving optimization, model packaging, and production deployment.
+
+In this guide, we will show you how to use BentoML to run programs written with Outlines on GPU locally and in [BentoCloud](https://www.bentoml.com/), an AI Inference Platform for enterprise AI teams. The example source code in this guide is also available in the [examples/bentoml/](https://github.com/dottxt-ai/outlines/blob/main/examples/bentoml/) directory.
+
+## Import a model
+
+First we need to download an LLM (Mistral-7B-v0.1 in this example and you can use any other LLM) and import the model into BentoML's [Model Store](https://docs.bentoml.com/en/latest/guides/model-store.html). Let's install BentoML and other dependencies from PyPi (preferably in a virtual environment):
+
+```bash
+pip install -r requirements.txt
+```
+
+Then save the code snippet below as `import_model.py` and run `python import_model.py`.
+
+**Note**: You need to accept related conditions on [Hugging Face](https://huggingface.co/mistralai/Mistral-7B-v0.1) first to gain access to Mistral-7B-v0.1.
+
+```python
+import bentoml
+
+MODEL_ID = "mistralai/Mistral-7B-v0.1"
+BENTO_MODEL_TAG = MODEL_ID.lower().replace("/", "--")
+
+def import_model(model_id, bento_model_tag):
+
+    import torch
+    from transformers import AutoModelForCausalLM, AutoTokenizer
+
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+    model = AutoModelForCausalLM.from_pretrained(
+        MODEL_ID,
+        torch_dtype=torch.float16,
+        low_cpu_mem_usage=True,
+    )
+
+    with bentoml.models.create(bento_model_tag) as bento_model_ref:
+        tokenizer.save_pretrained(bento_model_ref.path)
+        model.save_pretrained(bento_model_ref.path)
+
+
+if __name__ == "__main__":
+    import_model(MODEL_ID, BENTO_MODEL_TAG)
+```
+
+You can verify the download is successful by running:
+
+```bash
+$ bentoml models list
+
+Tag                                          Module  Size        Creation Time
+mistralai--mistral-7b-v0.1:m7lmf5ac2cmubnnz          13.49 GiB   2024-04-25 06:52:39
+```
+
+## Define a BentoML Service
+
+As the model is ready, we can define a [BentoML Service](https://docs.bentoml.com/en/latest/guides/services.html) to wrap the capabilities of the model.
+
+We will run the JSON-structured generation example [in the README](https://github.com/dottxt-ai/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
+
+```python
+DEFAULT_SCHEMA = """{
+    "title": "Character",
+    "type": "object",
+    "properties": {
+        "name": {
+            "title": "Name",
+            "maxLength": 10,
+            "type": "string"
+        },
+        "age": {
+            "title": "Age",
+            "type": "integer"
+        },
+        "armor": {"$ref": "#/definitions/Armor"},
+        "weapon": {"$ref": "#/definitions/Weapon"},
+        "strength": {
+            "title": "Strength",
+            "type": "integer"
+        }
+    },
+    "required": ["name", "age", "armor", "weapon", "strength"],
+    "definitions": {
+        "Armor": {
+            "title": "Armor",
+            "description": "An enumeration.",
+            "enum": ["leather", "chainmail", "plate"],
+            "type": "string"
+        },
+        "Weapon": {
+            "title": "Weapon",
+            "description": "An enumeration.",
+            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
+            "type": "string"
+        }
+    }
+}"""
+```
+
+First, we need to define a BentoML service by decorating an ordinary class (`Outlines` here) with `@bentoml.service` decorator. We pass to this decorator some configuration and GPU on which we want this service to run in BentoCloud (here an L4 with 24GB memory):
+
+```python
+import typing as t
+import bentoml
+
+from import_model import BENTO_MODEL_TAG
+
+@bentoml.service(
+    traffic={
+        "timeout": 300,
+    },
+    resources={
+        "gpu": 1,
+        "gpu_type": "nvidia-l4",
+    },
+)
+class Outlines:
+
+    bento_model_ref = bentoml.models.get(BENTO_MODEL_TAG)
+
+    def __init__(self) -> None:
+
+        import outlines
+        import torch
+        self.model = outlines.models.transformers(
+            self.bento_model_ref.path,
+            device="cuda",
+            model_kwargs={"torch_dtype": torch.float16},
+        )
+
+    ...
+```
+
+We then need to define an HTTP endpoint using `@bentoml.api` to decorate the method `generate` of `Outlines` class:
+
+```python
+    ...
+
+    @bentoml.api
+    async def generate(
+        self,
+        prompt: str = "Give me a character description.",
+        json_schema: t.Optional[str] = DEFAULT_SCHEMA,
+    ) -> t.Dict[str, t.Any]:
+
+        import outlines
+
+        generator = outlines.generate.json(self.model, json_schema)
+        character = generator(prompt)
+
+        return character
+```
+
+Here `@bentoml.api` decorator defines `generate` as an HTTP endpoint that accepts a JSON request body with two fields: `prompt` and `json_schema` (optional, which allows HTTP clients to provide their own JSON schema). The type hints in the function signature will be used to validate incoming JSON requests. You can define as many HTTP endpoints as you want by using `@bentoml.api` to decorate other methods of `Outlines` class.
+
+Now you can save the above code to `service.py` (or use [this implementation](https://github.com/dottxt-ai/outlines/blob/main/examples/bentoml/)), and run the code using the BentoML CLI.
+
+## Run locally for testing and debugging
+
+Then you can run a server locally by:
+
+```bash
+bentoml serve .
+```
+
+The server is now active at <http://localhost:3000>. You can interact with it using the Swagger UI or in other different ways:
+
+<details>
+
+<summary>CURL</summary>
+
+```bash
+curl -X 'POST' \
+  'http://localhost:3000/generate' \
+  -H 'accept: application/json' \
+  -H 'Content-Type: application/json' \
+  -d '{
+  "prompt": "Give me a character description."
+}'
+```
+
+</details>
+
+<details>
+
+<summary>Python client</summary>
+
+```python
+import bentoml
+
+with bentoml.SyncHTTPClient("http://localhost:3000") as client:
+    response = client.generate(
+        prompt="Give me a character description"
+    )
+    print(response)
+```
+
+</details>
+
+Expected output:
+
+```bash
+{
+  "name": "Aura",
+  "age": 15,
+  "armor": "plate",
+  "weapon": "sword",
+  "strength": 20
+}
+```
+
+## Deploy to BentoCloud
+
+After the Service is ready, you can deploy it to [BentoCloud](https://docs.bentoml.com/en/latest/bentocloud/get-started.html) for better management and scalability. [Sign up](https://cloud.bentoml.com/signup) if you haven't got a BentoCloud account.
+
+Make sure you have [logged in to BentoCloud](https://docs.bentoml.com/en/latest/bentocloud/how-tos/manage-access-token.html), then run the following command to deploy it.
+
+```bash
+bentoml deploy .
+```
+
+Once the application is up and running on BentoCloud, you can access it via the exposed URL.
+
+**Note**: For custom deployment in your own infrastructure, use [BentoML to generate an OCI-compliant image](https://docs.bentoml.com/en/latest/guides/containerization.html).
diff --git a/docs_new/examples/deploy-using-cerebrium.md b/docs_new/examples/deploy-using-cerebrium.md
new file mode 100644
index 000000000..6fbaab390
--- /dev/null
+++ b/docs_new/examples/deploy-using-cerebrium.md
@@ -0,0 +1,121 @@
+# Run Outlines using Cerebrium
+
+[Cerebrium](https://www.cerebrium.ai/) is a serverless AI infrastructure platform that makes it easier for companies to build and deploy AI based applications. They offer Serverless GPU's with low cold start times with over 12 varieties of GPU chips that auto scale and you only pay for the compute you use.
+
+In this guide we will show you how you can use Cerebrium to run programs written with Outlines on GPUs in the cloud.
+
+# Setup Cerebrium
+
+First, we install Cerebrium and login to get authenticated.
+
+```bash
+pip install cerebrium
+cerebrium login
+```
+
+Then let us create our first project
+
+```bash
+cerebrium init outlines-project
+```
+
+## Setup Environment and Hardware
+
+You set up your environment and hardware in the cerebrium.toml file that was created using the init function above.
+
+```toml
+[cerebrium.deployment]
+docker_base_image_url = "nvidia/cuda:12.1.1-runtime-ubuntu22.04"
+
+[cerebrium.hardware]
+cpu = 2
+memory = 14.0
+gpu = "AMPERE A10"
+gpu_count = 1
+provider = "aws"
+region = "us-east-1"
+
+[cerebrium.dependencies.pip]
+outline = "==0.0.37"
+transformers = "==4.38.2"
+datasets = "==2.18.0"
+accelerate = "==0.27.2"
+```
+
+## Setup inference
+
+Running code in Cerebrium is like writing normal python with no special syntax. In a `main.py` file specify the following:
+
+```python
+import outlines
+
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+
+schema = """{
+    "title": "Character",
+    "type": "object",
+    "properties": {
+        "name": {
+            "title": "Name",
+            "maxLength": 10,
+            "type": "string"
+        },
+        "age": {
+            "title": "Age",
+            "type": "integer"
+        },
+        "armor": {"$ref": "#/definitions/Armor"},
+        "weapon": {"$ref": "#/definitions/Weapon"},
+        "strength": {
+            "title": "Strength",
+            "type": "integer"
+        }
+    },
+    "required": ["name", "age", "armor", "weapon", "strength"],
+    "definitions": {
+        "Armor": {
+            "title": "Armor",
+            "description": "An enumeration.",
+            "enum": ["leather", "chainmail", "plate"],
+            "type": "string"
+        },
+        "Weapon": {
+            "title": "Weapon",
+            "description": "An enumeration.",
+            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
+            "type": "string"
+        }
+    }
+}"""
+
+generator = outlines.generate.json(model, schema)
+```
+
+On first deploy, it will download the model and store it on disk therefore for subsequent calls it will load the model from disk.
+
+Every function in Cerebrium is callable through an API endpoint. Code at the top most layer (ie: not in a function) is instantiated only when the container is spun up the first time so for subsequent calls, it will simply run the code defined in the function you call.
+
+To deploy an API that creates a new character when called with a prompt you can add the following code to `main.py`:
+
+```python
+def generate(
+    prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
+):
+
+    character = generator(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
+    )
+
+    return character
+```
+
+
+## Run on the cloud
+
+```bash
+cerebrium deploy
+```
+
+You will see your application deploy, install pip packages and download the model. Once completed it will output a CURL request you can use to call your endpoint. Just remember to end
+the url with the function you would like to call - in this case /generate. You should see your response returned!
diff --git a/docs_new/examples/deploy-using-modal.md b/docs_new/examples/deploy-using-modal.md
new file mode 100644
index 000000000..15e200cb5
--- /dev/null
+++ b/docs_new/examples/deploy-using-modal.md
@@ -0,0 +1,198 @@
+# Run Outlines using Modal
+
+[Modal](https://modal.com/) is a serverless platform that allows you to easily run code on the cloud, including GPUs. It can come very handy for those of us who don't have a monster GPU at home and want to be able to quickly and easily provision, configure and orchestrate cloud infrastructure.
+
+In this guide we will show you how you can use Modal to run programs written with Outlines on GPU in the cloud.
+
+## Requirements
+
+We recommend installing `modal` and `outlines` in a virtual environment. You can create one with:
+
+```bash
+python -m venv venv
+source venv/bin/activate
+```
+
+Then install the required packages:
+
+```bash
+pip install modal outlines
+```
+
+## Build the image
+
+First we need to define our container image. If you need to access a gated model, you will need to provide an [access token](https://huggingface.co/settings/tokens). See the `.env` call below for how to provide a HuggingFace token.
+
+Setting a token is best done by setting an environment variable `HF_TOKEN` with your token. If you do not wish to do this, we provide a commented-out line in the code to set the token directly in the code.
+
+```python
+from modal import Image, App, gpu
+import os
+
+# This creates a modal App object. Here we set the name to "outlines-app".
+# There are other optional parameters like modal secrets, schedules, etc.
+# See the documentation here: https://modal.com/docs/reference/modal.App
+app = App(name="outlines-app")
+
+# Specify a language model to use.
+# Another good model to use is "NousResearch/Hermes-2-Pro-Mistral-7B"
+language_model = "mistral-community/Mistral-7B-v0.2"
+
+# Please set an environment variable HF_TOKEN with your Hugging Face API token.
+# The code below (the .env({...}) part) will copy the token from your local
+# environment to the container.
+# More info on Image here: https://modal.com/docs/reference/modal.Image
+outlines_image = Image.debian_slim(python_version="3.11").pip_install(
+    "outlines",
+    "transformers",
+    "datasets",
+    "accelerate",
+    "sentencepiece",
+).env({
+    # This will pull in your HF_TOKEN environment variable if you have one.
+    'HF_TOKEN':os.environ['HF_TOKEN']
+
+    # To set the token directly in the code, uncomment the line below and replace
+    # 'YOUR_TOKEN' with the HuggingFace access token.
+    # 'HF_TOKEN':'YOUR_TOKEN'
+})
+```
+
+## Setting the container up
+
+When running longer Modal apps, it's recommended to download your language model when the container starts, rather than when the function is called. This will cache the model for future runs.
+
+```python
+# This function imports the model from Hugging Face. The modal container
+# will call this function when it starts up. This is useful for
+# downloading models, setting up environment variables, etc.
+def import_model():
+    import outlines
+    outlines.models.transformers(language_model)
+
+# This line tells the container to run the import_model function when it starts.
+outlines_image = outlines_image.run_function(import_model)
+```
+
+## Define a schema
+
+We will run the JSON-structured generation example [in the README](https://github.com/dottxt-ai/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
+
+```python
+# Specify a schema for the character description. In this case,
+# we want to generate a character with a name, age, armor, weapon, and strength.
+schema = """{
+    "title": "Character",
+    "type": "object",
+    "properties": {
+        "name": {
+            "title": "Name",
+            "maxLength": 10,
+            "type": "string"
+        },
+        "age": {
+            "title": "Age",
+            "type": "integer"
+        },
+        "armor": {"$ref": "#/definitions/Armor"},
+        "weapon": {"$ref": "#/definitions/Weapon"},
+        "strength": {
+            "title": "Strength",
+            "type": "integer"
+        }
+    },
+    "required": ["name", "age", "armor", "weapon", "strength"],
+    "definitions": {
+        "Armor": {
+            "title": "Armor",
+            "description": "An enumeration.",
+            "enum": ["leather", "chainmail", "plate"],
+            "type": "string"
+        },
+        "Weapon": {
+            "title": "Weapon",
+            "description": "An enumeration.",
+            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
+            "type": "string"
+        }
+    }
+}"""
+```
+
+To make the inference work on Modal we need to wrap the corresponding function in a `@app.function` decorator. We pass to this decorator the image and GPU on which we want this function to run.
+
+Let's choose an A100 with 80GB memory. Valid GPUs can be found [here](https://modal.com/docs/reference/modal.gpu).
+
+```python
+# Define a function that uses the image we chose, and specify the GPU
+# and memory we want to use.
+@app.function(image=outlines_image, gpu=gpu.A100(size='80GB'))
+def generate(
+    prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
+):
+    # Remember, this function is being executed in the container,
+    # so we need to import the necessary libraries here. You should
+    # do this with any other libraries you might need.
+    import outlines
+
+    # Load the model into memory. The import_model function above
+    # should have already downloaded the model, so this call
+    # only loads the model into GPU memory.
+    model = outlines.models.transformers(
+        language_model, device="cuda"
+    )
+
+    # Generate a character description based on the prompt.
+    # We use the .json generation method -- we provide the
+    # - model: the model we loaded above
+    # - schema: the JSON schema we defined above
+    generator = outlines.generate.json(model, schema)
+
+    # Make sure you wrap your prompt in instruction tags ([INST] and [/INST])
+    # to indicate that the prompt is an instruction. Instruction tags can vary
+    # by models, so make sure to check the model's documentation.
+    character = generator(
+        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
+    )
+
+    # Print out the generated character.
+    print(character)
+```
+
+We then need to define a `local_entrypoint` to call our function `generate` remotely.
+
+```python
+@app.local_entrypoint()
+def main(
+    prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
+):
+    # We use the "generate" function defined above -- note too that we are calling
+    # .remote() on the function. This tells modal to run the function in our cloud
+    # machine. If you want to run the function locally, you can call .local() instead,
+    # though this will require additional setup.
+    generate.remote(prompt)
+```
+
+Here `@app.local_entrypoint()` decorator defines `main` as the function to start from locally when using the Modal CLI. You can save above code to `example.py` (or use [this implementation](https://github.com/dottxt-ai/outlines/blob/main/examples/modal_example.py)). Let's now see how to run the code on the cloud using the Modal CLI.
+
+## Run on the cloud
+
+First install the Modal client from PyPi, if you have not already:
+
+```bash
+pip install modal
+```
+
+You then need to obtain a token from Modal. Run the following command:
+
+```bash
+modal setup
+```
+
+Once that is set you can run inference on the cloud using:
+
+```bash
+modal run example.py
+```
+
+You should see the Modal app initialize, and soon after see the result of the `print` function in your terminal. That's it!
diff --git a/docs_new/examples/earnings-reports.md b/docs_new/examples/earnings-reports.md
new file mode 100644
index 000000000..26c2a9ccf
--- /dev/null
+++ b/docs_new/examples/earnings-reports.md
@@ -0,0 +1,278 @@
+# Extracting financial data from earnings reports
+
+A common task in finance is to extract financial data from earnings reports. Earnings reports are infamously poorly formatted, as the SEC does not have requirements for producing machine-readable documents.
+
+Earnings reports are often provided as HTML documents, which can be difficult to parse. Investors often use complicated parsing systems or manual review to extract data. Entire companies are built around automating this task.
+
+This cookbook is a proof of concept about how we can use LLMs to extract financial data directly into CSV. Comma-separated values are well-structured and can be defined by a regular expression, which Outlines can use to guide the LLM's output.
+
+The example is a smaller subset of a full demo found [here](https://github.com/dottxt-ai/demos/tree/main/earnings-reports). The demo contains the full set of pre-processing steps needed to convert raw HTML into a structured CSV file, and tests the results across three company's 10k reports.
+
+## Setup
+
+Install outlines and required dependencies:
+
+```bash
+# Later versions of torch can have difficulty with certain CUDA drivers.
+# We recommend using 2.4.0 for now, but you may wish to experiment with
+# other versions.
+pip install outlines pandas transformers torch==2.4.0 accelerate
+```
+
+## Load the model
+
+Choose your language model. We'll use Phi-3 mini, which is small enough to run on reasonably small machines.
+
+```python
+import outlines
+import torch
+
+model_name = 'microsoft/Phi-3-mini-4k-instruct'
+model = outlines.models.transformers(
+    model_name,
+    device='auto',
+    model_kwargs={
+        # To reduce memory usage, we'll use bfloat16
+        "torch_dtype": torch.bfloat16,
+    },
+)
+```
+
+## Set up the data
+
+For brevity, we've attached the markdown version of Nvidia's 10k report. The [full demonstration](https://github.com/dottxt-ai/demos/tree/main/earnings-reports) processes the raw HTML version of the report to these markdown tables. Pages are filtered by whether they seem to contain income statements, and then compacted into the string you see below.
+
+```python
+income_statement = """
+Table of ContentsNVIDIA Corporation and SubsidiariesConsolidated Statements of Income(In millions, except per share data)
+
+|  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |
+| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
+|  | | | Year Ended | | | | | | | | | | | | | | |
+|  | | | Jan 28, 2024 | | |  | | | Jan 29, 2023 | | |  | | | Jan 30, 2022 | | |
+| Revenue | | | $ | 60,922 |  |  | | | $ | 26,974 |  |  | | | $ | 26,914 |  |
+| Cost of revenue | | | 16,621 | |  |  | | | 11,618 | |  |  | | | 9,439 | |  |
+| Gross profit | | | 44,301 | |  |  | | | 15,356 | |  |  | | | 17,475 | |  |
+| Operating expenses | | |  | | |  | | |  | | |  | | |  | | |
+| Research and development | | | 8,675 | |  |  | | | 7,339 | |  |  | | | 5,268 | |  |
+| Sales, general and administrative | | | 2,654 | |  |  | | | 2,440 | |  |  | | | 2,166 | |  |
+| Acquisition termination cost | | | — | |  |  | | | 1,353 | |  |  | | | — | |  |
+| Total operating expenses | | | 11,329 | |  |  | | | 11,132 | |  |  | | | 7,434 | |  |
+| Operating income | | | 32,972 | |  |  | | | 4,224 | |  |  | | | 10,041 | |  |
+| Interest income | | | 866 | |  |  | | | 267 | |  |  | | | 29 | |  |
+| Interest expense | | | (257) | |  |  | | | (262) | |  |  | | | (236) | |  |
+| Other, net | | | 237 | |  |  | | | (48) | |  |  | | | 107 | |  |
+| Other income (expense), net | | | 846 | |  |  | | | (43) | |  |  | | | (100) | |  |
+| Income before income tax | | | 33,818 | |  |  | | | 4,181 | |  |  | | | 9,941 | |  |
+| Income tax expense (benefit) | | | 4,058 | |  |  | | | (187) | |  |  | | | 189 | |  |
+| Net income | | | $ | 29,760 |  |  | | | $ | 4,368 |  |  | | | $ | 9,752 |  |
+|  | | |  | | |  | | |  | | |  | | |  | | |
+| Net income per share: | | |  | | |  | | |  | | |  | | |  | | |
+| Basic | | | $ | 12\.05 |  |  | | | $ | 1\.76 |  |  | | | $ | 3\.91 |  |
+| Diluted | | | $ | 11\.93 |  |  | | | $ | 1\.74 |  |  | | | $ | 3\.85 |  |
+|  | | |  | | |  | | |  | | |  | | |  | | |
+| Weighted average shares used in per share computation: | | |  | | |  | | |  | | |  | | |  | | |
+| Basic | | | 2,469 | |  |  | | | 2,487 | |  |  | | | 2,496 | |  |
+| Diluted | | | 2,494 | |  |  | | | 2,507 | |  |  | | | 2,535 | |  |
+"""
+```
+
+The markdown tables extracted from the earnings reports can vary widely in row names, column counts, data types, etc. The advantage of LLMs here is that we can define the data we want in terms of the data types, and the LLM will output the data in the desired format.
+
+For comparison, here is how the income statement looks in the original HTML:
+
+![Nvidia income statement](./images/nvidia-income.png)
+
+## Define the data we want
+
+Outlines is often used for JSON output, but it can also be used for CSV. We know the columns we want to extract, and we know the data types of the columns. Year for example is always a four-digit number, revenue is a number with commas, and so on.
+
+We can define a regex pattern for each column type:
+
+```python
+# Define the column type regex patterns
+column_types = {
+    # Year is always a four-digit number
+    "year": r"\d{4}",
+
+    # Revenue, operating income, and net income are always numbers with commas.
+    # This regex permits integers that may begin with a minus sign, and may have
+    # commas separating the thousands, millions, etc.
+    "integer_comma": r"((-?\d+),?\d+|(-?\d+))",
+    # Number is currently not used, but it represents a number with up to two decimal places.
+    "number": r"(-?\d+(?:\.\d{1,2})?)",
+}
+```
+
+Next, let's choose the columns we want to extract. We want
+
+- Year, always a four-digit number
+- Revenue, a number with commas
+- Operating income, a number with commas
+- Net income, a number with commas
+
+```python
+# Define the columns to extract, and their data types.
+columns_to_extract = {
+    "year": "year",
+    "revenue": "integer_comma",
+    "operating_income": "integer_comma",
+    "net_income": "integer_comma",
+}
+```
+
+You can modify `column_type_regex` to match the data types of the columns you want to extract.  Adding a new financial metric to extract is as simple as adding a new key/value pair to `columns_to_extract`:
+
+```python
+columns_to_extract["diluted_earnings_per_share"] = "number"
+```
+
+Additional columns are not well tested for accuracy, so use with caution.
+
+## Create the regex describing the data we want
+
+
+```python
+# Create the header line. This is the requested column names
+# separated by commas, i.e. "year,revenue,..."
+header = ",".join(columns_to_extract.keys())
+
+# Create the data capture patterns. These are the regex patterns
+# that will be used to capture the data in each column
+data_patterns = [column_types[dtype] for dtype in columns_to_extract.values()]
+data_line = ",".join(data_patterns)
+
+# Our final regex pattern.
+max_rows = 3 # We expect 3 rows of data, firms usually report 3 years of income statements
+csv_regex = f"{header}(\n{data_line}){{,{max_rows}}}\n\n"
+
+print(csv_regex)
+```
+
+which gives us
+
+```
+year,revenue,operating_income,net_income,basic_earnings_per_share(
+\d{4},((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),(-?\d+(?:\.\d{1,2})?)){,3}
+```
+
+Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.generate.regex` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
+
+## Prompting the model
+
+Outlines does not add system or instruction tokens by default, so we need to use `transformers.AutoTokenizer` to add them for whatever model we're using.
+
+```python
+from transformers import AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+def add_instruction(prompt):
+    return tokenizer.apply_chat_template([{"role": "user", "content": prompt}], tokenize=False, add_generation_prompt=True)
+
+print(add_instruction("Howdy"))
+```
+```
+<|user|>
+Howdy<|end|>
+<|assistant|>
+```
+
+Our prompt roughly describes the task we want the model to perform, and a few pieces of information it may need to know about income statements.
+
+```python
+def extract_financial_data_prompt(columns_to_extract, income_statement):
+    user_prompt = f"""
+    Extract annual financial data from this set of pages. Pages
+    are from a 10k filing and were chosen because they may contain
+    a comprehensive income statement. Note that selected pages may
+    be incorrectly extracted, so you should verify that you are extracting
+    from the comprehensive income statement and not some other financial
+    statement.
+
+    Create a row for each year available in the income statement with the
+    following columns: {', '.join(columns_to_extract.keys())}. Firms typically report the
+    most recent 3 years of data, but this can vary.
+
+    Each column has types: {', '.join(columns_to_extract.values())}.
+
+    # Relevant pages:
+
+    {income_statement}
+
+    # Key instructions:
+
+    1. Look ONLY at the "Consolidated Statements of Income" table
+    2. For operating income, look for "Income from operations" or "Operating income"
+    3. For net income, use the TOTAL net income figure, not amounts allocated to specific share classes
+    4. Use NULL for missing values
+    5. Operating income must be less than revenue
+    6. Net income must be less than operating income
+    7. Ignore segment breakdowns, quarterly data, or per-share amounts
+
+    # Output format:
+
+    - CSV format with headers: {','.join(columns_to_extract.keys())}
+    - Use NULL for missing values
+    - If no data are found, do not create a row.
+    - Enter two newline characters to terminate the CSV when no more data are found.
+
+    # Definitions:
+    - Revenue: Total sales of goods and services. Usually this is at the top of the
+    income statement.
+    - Operating income: Revenue minus operating expenses for the entire company. This is revenue
+    minus costs. Operating income is also called operating profit, EBIT, or income from
+    operations.
+    - Net income: Operating income minus taxes. This is the bottom line of the
+    income statement.
+    """
+
+    return add_instruction(user_prompt)
+```
+
+## Running the model
+
+Now that we have our prompt and regular expression, we can run the model.
+
+Construct our regex extractor function. We'll use a greedy sampler, which samples the most likely next token at each step. It's a simple sampler that is more reproducible than multinomial sampling.
+
+```python
+csv_extractor = outlines.generate.regex(
+    model, csv_regex, sampler=outlines.samplers.greedy()
+)
+```
+
+Provide the prompt to the model and run it:
+
+```python
+csv_data = csv_extractor(
+    extract_financial_data_prompt(columns_to_extract, income_statement),
+    max_tokens=1024,
+)
+
+print(csv_data)
+```
+```
+year,revenue,operating_income,net_income
+2024,60922,32972,29760
+2023,26974,4224,4368
+2022,26914,10041,9752
+```
+
+Voila! We've extracted the financial data from the income statement, and it's correct upon inspection.
+
+You can even load this into a `pandas` DataFrame for further analysis:
+
+```python
+import pandas as pd
+from io import StringIO
+
+df = pd.read_csv(StringIO(csv_data))
+print(df)
+```
+```
+   year  revenue  operating_income  net_income
+0  2024    60922             32972       29760
+1  2023    26974              4224        4368
+2  2022    26914             10041        9752
+```
diff --git a/docs_new/examples/extract_event_details.md b/docs_new/examples/extract_event_details.md
new file mode 100644
index 000000000..0f87d9586
--- /dev/null
+++ b/docs_new/examples/extract_event_details.md
@@ -0,0 +1,34 @@
+This recipe demonstrates how to use the `outlines` library to extract structured event details from a text message.
+We will extract the title, location, and start date and time from messages like the following:
+
+```plaintext
+Hello Kitty, my grandmother will be here, I think it's better to postpone
+our appointment to review math lessons to next Monday at 2pm at the same
+place, 3 avenue des tanneurs, one hour will be enough see you 😘
+```
+
+Let see how to extract the event details from the message with the MLX
+library dedicated to  Apple Silicon processor  (M series).
+
+```python
+--8<-- "docs/cookbook/extract_event_details.py"
+```
+
+The output will be:
+
+```plaintext
+Today: Saturday 16 November 2024 and it's 10:55
+```
+
+and the extracted event information will be:
+
+```json
+{
+  "title":"Math Review",
+  "location":"3 avenue des tanneurs",
+  "start":"2024-11-22T14:00:00Z"
+}
+```
+
+
+To find out more about this use case, we recommend the project developped by [Joseph Rudoler](https://x.com/JRudoler) the [ICS Generator](https://github.com/jrudoler/ics-generator)
diff --git a/docs_new/examples/extraction.md b/docs_new/examples/extraction.md
new file mode 100644
index 000000000..dfcca0cfd
--- /dev/null
+++ b/docs_new/examples/extraction.md
@@ -0,0 +1,90 @@
+# Named entity extraction
+
+Named Entity Extraction is a fundamental problem in NLP. It involves identifying and categorizing named entities within a document: people, organization, dates, places, etc. It is usually the first step in a more complex NLP worklow. Here we will use the example of a pizza restaurant that receives orders via their website and need to identify the number and types of pizzas that are being ordered.
+
+Getting LLMs to output the extracted entities in a structured format can be challenging. In this tutorial we will see how we can use Outlines' JSON-structured generation to extract entities from a document and return them in a valid JSON data structure 100% of the time.
+
+As always, we start with initializing the model. We will be using a quantized version of Mistal-7B-v0.1 (we're GPU poor):
+
+```python
+import outlines
+
+model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+```
+
+And we will be using the following prompt template:
+
+```python
+from outlines import Template
+
+
+take_order = Template.from_string(
+    """You are the owner of a pizza parlor. Customers \
+    send you orders from which you need to extract:
+
+    1. The pizza that is ordered
+    2. The number of pizzas
+
+    # EXAMPLE
+
+    ORDER: I would like one Margherita pizza
+    RESULT: {"pizza": "Margherita", "number": 1}
+
+    # OUTPUT INSTRUCTIONS
+
+    Answer in valid JSON. Here are the different objects relevant for the output:
+
+    Order:
+        pizza (str): name of the pizza
+        number (int): number of pizzas
+
+    Return a valid JSON of type "Order"
+
+    # OUTPUT
+
+    ORDER: {{ order }}
+    RESULT: """
+)
+```
+
+We now define our data model using Pydantic:
+
+```python
+from enum import Enum
+from pydantic import BaseModel
+
+class Pizza(str, Enum):
+    margherita = "Margherita"
+    pepperonni = "Pepperoni"
+    calzone = "Calzone"
+
+class Order(BaseModel):
+    pizza: Pizza
+    number: int
+```
+
+We can now define our generator and call it on several incoming orders:
+
+```python
+orders = [
+    "Hi! I would like to order two pepperonni pizzas and would like them in 30mins.",
+    "Is it possible to get 12 margheritas?"
+]
+prompts = [take_order(order) for order in orders]
+
+generator = outlines.generate.json(model, Order)
+
+results = generator(prompts)
+print(results)
+# [Order(pizza=<Pizza.pepperonni: 'Pepperoni'>, number=2),
+#  Order(pizza=<Pizza.margherita: 'Margherita'>, number=12)]
+```
+
+There are several ways you could improve this example:
+
+- Clients may order several types of pizzas.
+- Clients may order drinks as well.
+- If the pizza place has a delivery service we need to extract the client's address and phone number
+- Clients may specify the time for which they want the pizza. We could then check against a queuing system and reply to them with the estimated delivery time.
+
+How would you change the Pydantic model to account for these use cases?
diff --git a/docs_new/examples/images/chain_of_density.png b/docs_new/examples/images/chain_of_density.png
new file mode 100644
index 0000000000000000000000000000000000000000..61e01f40459e26f082fc90ea2964dcc4bd2d70c6
GIT binary patch
literal 515603
zcmb@uV{~L)7d2X8$LVwj9ox2T+qTuQZFQ3F*k;GJla6iMw(fbpJH8+H{e9OsBQ?gU
zQ>UtS?Y-7qbIzRz1vzm<I4n322!tppA)*8VL2ZLT5Iiv8z#09CHwfSX!bwO{1qKFY
zbz6Q51R?-QiU_K>ubgGN=%B7{4o%Y0tmppD`RyMZOdAg7YPj3b;TmNW<$2!h>ZNKU
zIv^tWRV4mvL0en8saH0?St>6#Hcj*J`?~o?)w{8o>yAv86W<|ruwuR#7|#DhhMj2I
zfKf<ru(KPZ!~evGzVO)v_}2dh$Pzq2fZ%Iu+YgTaiO;H@bBq7Ei9;*CPJYk-#IdhK
z4D^@(Zxp}wuwwo1r@Q#F`2X__#ar9p{by~bEbjK^>hpG#po0i$X=$mc^Ybd!Mg`>6
zi;L@7J`Dc9-}z1LLs!=^&O`d-6#Lh4VHD423is_}a*y@lNAnYjiRWDSwwJ5T@kKe2
zM_RhJTy`4{7M@`K^wVc1ukp!X19fX#^k;d+W-T)*Ekr8bA)D1}lFPbo-d`fJhvLU9
z;o*9O2<P`1%s#!BbzKTr1y$7!lX)?w7PQp$Hm6gGWJBQWe*=@@nlocTy=X^ELnml<
zIGI{wbN#mI>O9Fls%P3ITr~8frSry}<{*?jfAZOzJ5t7a#2URq>i8sYp?(2NuR+Kh
zU%r!f+4;@BP2rQ#I}yJUX(f7s&6bMvTks2oj29<8#b;f2SLYbK!(7efMV)2#%galf
z+ohQJe{pztI65*yNkvswR(5iHjLZ4cWcU~1x~@RRW<Xrr<VYep7$^#ZzSH?w$qhps
zT!G@ZX*v6!Kk7tv(&%HlTbylPQFq)dSZwLKnyTwFLS*p;taI~dXlTKel7$K-DpXA4
zIkRp1kqVY9^$7KLRKm$hs=9TqiEfgXnr;|-U9gCrzDEd-+1a6?V(96zBwtljRa;*X
z`U8b>ad<tDcxRz*%_1nL6iVlABB{s{O9|+UD=SG!N$1y#14Xs9t*`N&K06)IQh7ZN
z^71<>BWr)UC7*tB)1a5<*1&=U5|Y%El+G@8-X1M23Dp<ps+Vm<p`d7KWx)fc)=3k1
z5XkD<+XKb&M@B~<&vIo+Tq_?RKNpzSR-|$jN<kn}QK)YbbL#^rQEiWS=KFg@voD^l
ze(n=5nc~Gvr6yw5t1Pl_h{rTst%Rw9ABdu+b<0O6U?3T`+eh6Zx}X8G`T7NLkV%0=
zS#4R@c{2hVL}yA~E|X=inpurzSLIkH84*$M>}*|54oG)q?Z41bRE$qeEiEXRTZig@
zmK6~fPaH7;`TzMdPk(u)ACQr;7RpN9=zb|`X1~^X$>bd#Z&+nVql96_LW53C#B?xF
z?lw|tNlTldB_1h(!_nPiTYq{KPd7u<W^Uuku9x)5G?~>E2!(~{;_}5y8_I(dD^t$0
zP-SsJQiRO?o^5M-7USw0k)o8IoSLNMNN3mh6=9G}wAUv&_IqYdZm#1@^fNHF=E&z5
zahJ|+j{Uut`{|MYs7$n!g?%>8j`J3)%I)3n58jI>_2J;6G->37MTB*I#wd~{ch_a3
zy`;+wMoN$UaJdob3BQK_{8@W*bEAfS&=-bK-`JR7C;uaj&Esvhaj_t`=DKJ1u>4bJ
z4K4ZlY`{8xn*S+TS~?LG4XuJ^?EKv3WmT)ZWD5=@JW77;r~w<ab;5IZ6|lapoAF6Q
zT4*|{*Z%Rq+ge|Nqic~<b;!5kvg6L4nZ--ulH(uO<A#6r1B<{VI4{r0kNb367&<i-
z%S{ot$AC$=%iHVvL|4m`%u1##hrr(2meu^=4lMS)?Wp-WV&t^x3=+hFjO~3kwegoD
z7?RWT#>T;d1GF`E9TWu3;p*nNQe0MsW#>|%)2N^(R<vH}-HGe-`vD=76x933&sREH
z>XrwxnHn*per;-U<JMI{nNvK0-oyvS0~E+3sTW@`B|rPK_V{T!lz30_M%PU#H)K0M
zTMi85o+9hZcka80d-O@Wy}eB=9&5ss3i6-M7e~fslhN1DmMfmMs2?00yt(-a4<B_w
zM?7J_b2ynj9Nh8#>3^M^f`GNfFg3#WG9(WIC62Vajy_5>HL|ry<y%@<thpUM#WlN+
z{UnRJ=RFyEEWytf>P5O(C#0t*w?Wob3<>W`<4<N`V<nu$I>{_8{v4k=zEij7&#7}|
zd!NAJrywVPgl|KYF*a6KbS&-e^6ekkd^*eu=-dcp^+UpDs&1o?2wtv!m^?T@H__6T
zR?d(mA~IH3-th}8IyiWZdz)G&h36IJ7ea|={DPEiV`tezQ&m~+_3}0!rGEL~ZtseH
zP8DCPNU36otS4l2lV41)qVhYL?(QyyOEDNWuD<>*(iHv?pDT8Z&*#G`X$X~87BND9
zByEMuun-DFQe436^C1-~vQqCm|2x9?Gut{%zti{fg6M6?ux^=Nzl+;z8!PeBNkvEL
zCGJv~|C62kW5wK<HdV0hAj!-S+3##JSlbQn==h{Ux5YX5R$+wi`5qCm?GZmhUS^Pn
zJW-oC-xxU7#oI2H8Wl4xe8iveJwq*7UVx$O>df98IroZFTs3!MPK+lGZEAy!8gKn`
zq`fwFi_<xQq+xQtyl#Hm^(ArOU&ScO(_&LX$!Ub!Ug7=%cf}fezqVqLboB7sD$cks
z^F@caPs%e6FrO4R%bW1zr;mfnNW?c$)|C$Ay3Oj`&R0TUSdj7gGT80)+>&96R}Jcx
zff%lD_r`#$>dO8zib05l2qK{2A)@7-o#}0FgPSEy>QJF_30kD{4S*&iVlta;Tt_HU
zfRJV@_hP}4hQAerz=_XJEp=@AoqJkRW$)kiua2k4IYfXMX@InF)eWD+?|JYNU+_ff
zMN=zhGR`r3W5?!ZfAv$Q4|jWdHd%Ije7OFv+~mmjHPH$jwC_!}$;e#-LA?eH-bmTr
zkAz2$sGT)6$G|GrZhy>;)dI(4HEJ%E#5YM3%U|!Hw=)5Y5YC^K&uF<?yk=fk_4k9Y
zVdH+&HgQYd;h&g2tO5#iFX~?mb~d)k%~oc@($OGOg5ODhE+065o}7`wE}$TgIh8;1
z{rBY<IPR{lyGVVxY}j5$KPQ-RFtP>g&=!4mHw7dumscP0r<ebXqDStwpzESXp;&-G
zlg#=qkIQfDmZl1FBH2z?E_Aj@?Nz+#;}4UbtF(uvITieklRxgrwUm>Hq#XT%{W7bX
zU;=}Jx_sZ^Kn_k$eNbtihhJeJ2iRIy#GlxW4ebu#26Eb6Pwm*zi=_Xa&8NH7Io)iL
z$|1df>a-esWMGz2>F}rR%>k<mJosEh>c#JP_r44K54%Kl=%76lYHi)|p@DsrtEXbt
z)AK8z_e!bmhj@-ssUoA}x^5&=w^7Xo6l7jq;$U0EF!;jUn)lVPq<cg~#o5@gvKpN>
zORon-`&;jRIwE4sH~LCPd}8vjk9V3^sp|zI{+%6Q1rg?5ij#aRmca>|@O`^W6+2xy
zQh@~TnckQHj@&NW><zIL*Nw35Uj=(}bhO=Fsbdux|H@7-=mz{!*V^Bt7e4}$OUlb>
zuQD2pmW8;}eUF5=G96#mOi_ZUGY4C<3-dd0Kbylp)BUkO*AnLx@pvAjC&$FhVIg}9
zt}m#QXr1@paiIJ@zP;?Ho5(Y8lUL>CDeWv)Mm=RGia_u|=ieAj;;ler;`A^7qwmAd
zbY<M)fxy2F#4gP&jTXUH2w%BsT9t5ZE!mkBpC2B#*mHRAks`q>PaSK`V3aSapCPRP
z4XTr&v;J|umwfjlqX=jI)o3+ec0o7Vh0B)yj*sp8;WscRxdu96thwG_S9g@&SS<b9
z+4Fht<(wozg11!XvlqTCLAuB3Hkh}6T4rq(?`Qw!WPg?t4<2+#m^A3(a*v$F^9hun
ziuMeN-zUa%)(>Nr`MQ}QF7UHVwL(IDyF1@+;o***LlGiOOy;cnuX2sWT7cvNvqr*A
ze%W{_9N@RCp*@~S$}uG-Vsh68n)otWo!s+pKMDi_GUSK-#%Hv=RiOeZ8mf+N=y)b;
zhrGRTg#O-6!i*c3AfciD>B%W2YFS1`MQH0vHDQog8<)1`!$5_OuJY^SGxab_%OCJh
zm-nP9Ed9Gw*;6$uYir5I?zTpIwy8xZ(B3`<-?i8C0dG^H-|A!TY6%MXrJSJQH1L~G
z>US>8Ehv_<cyMxlPvnEZeU;B<is!dLYH7H?V*Fd#bAVXhz`{yJRer(ijgEq*tf?tV
z6hAgDb1;#1!DLe{z>mR}j*EhM%SyyV%UhmP9WGWZjS``?qP3ug4uN|<6oX645G$$W
zdGIbi=}&-jc5Wt<PI6U+&h$Kd-311U86PhZ6xa3XyLb=Vo(D=mN0Xg^IX<sf>e_;0
za(w)9opDt6cUU;2eOCsUVBJ-AN-7w{xpl>FiAqzC7$63--xSj8K;4=X{udR}ay%{U
z<yFtCTea!K<NLzHv_7}?7I&FHeyVQt&~3n!Z@jbN10y9N1z<-AHf=q<jpn#%ip6bW
z-O%##@(8-~K##ihp=NLpFa`9xUL77kzC#BEyrJQtVB(=1Gs##W1u7DUe)?=vDNQ%l
z|H?Nq$_ztHnls%Kms5LqJ0?D|adH|RC}-bh-`}3a5GQ0$EJ;~fwDpbR^?F#UK9C0q
z2n+A_71MY6#L>}F8rB6=R75u>9UxW3u)WhTh8fo_Q(9tPf`fO7zgyVZ<ud;OD!*qd
z*WZagu{h_2#n}*bI|~aQ#{+Pt4#V@vqM`%l!>WP;QwuAr3XPi~j>l;r2}Wdp&SF$q
zTU&c!9@(!q7j;##wm$ElHc3ly1kLA*w?_YT`CD$^?D486Vj^Rz+7Q^@?!rGBa%;Z2
zx@-WKnwshKxM2(ufJMb7<z&UZP1Ddy`DDCog8uS;>G*l=L{`w=%!Rz(V*6c*mHsZn
z=VIpGoL-OEZ-4Kg3mzto%M!;7S%r#~=IwRWzxynM=O=Zl#P5XU<hsV%ADl+bx79u8
zzJZOlT2bLoP$2^VsPlbJJ8g9MoMwJx{M!-Z+dyV(W7X<%2*y=TK=1c)hcXxaJi2(v
z{rDd8;dL?NvNV1&w=4}T$l*JfL&ITpsC``l1rZY#ruF<Vwu~Prwg$hG)N|mKXPFo#
z9rT^*LT){jB>D$Q+ua9h;1$%=R<;}B$IZGt@So@i7-)&3#?7!XAF!BbmYpVKFfp_I
zJCC&cTB@}dN=tA0<=5SgIO@)sIV^jZ2!k@Fy72fsvJ%peKu;*i{&el<uScwVM2H{_
z2EWFT7R*mHv%2NYPI13iyn#yL!h-sGmdZ*RXim;(7yNlUn(B!jv$`PRuADk!vDR?r
z`cbpG_Z1nlml7=6+J`qK957H$&5^wmpSu?9_))xavKPyVjDfENM1!KRoKO99xNcAr
zR{4EAQKOS<LuEs;F&*EyyXLwUF*z-_j^8lZ5NjU-bd<5Yp!p2{!I#$}9ysck6%lK*
z+LTr)u2|mJPeWGovSks&h~OogM*X7C=cV#gmt_^`aiLioSN%A|*z$r_-y^TskRe43
ze2@Tf!?o{8A_U41)d)@|ZzbF9T^zhusLkPKS;WE}&u)YLzx@8PJu^>uu&>ecp2%}9
zI;KrFzp}o6zFeb4_=#pgg7y{;^Wll5jm<i>Kvir4Lp@$oW(47B3oG0s$U^Xa#G^)7
zrLrR6HN{48u!9%OW5v--ID|5M+E~>^@e4qLC|16R96r6Hu3kQ4=LEB_g)sWJ%VbqY
zAwv~GQCU^ME@GeKb*TvdGH&|_dq#oHZbw78O`)%-s5sm|U#^bN*YBEUV`KAhxf!R~
zY|s<9u%HI434DBf5GY2h&FL^UJ6qqxgd{{LK0bbWZtm{kL9fm%raLhvBBI6Za^rNd
z%G%262SR9M<iDjFV+>j?mY;hPA4Qih_ot)({vC{^GP|BFWpcam`M#ydkWHpCA3mI~
zLV^>LknC;s1OXo-GBUDUy^7EMDn}%Ex}u^2xE35lCZE-AF`c)wvjd=gNNDJTy}hpy
zBF4tX^z>Dv(n-TG*BP1Y_IR0cmz!OmS-f5#kP6j3aARLz-|6Y;;rr>CnUx0X#Yi-8
zrgNs$G5KshUe7xU8k&l{yucBYveMERU?3(kICqEQhB9<rT_0j`xj`T}IJnNvPNIn;
z;KK(51RP=Jj`I6`&@gz+-t3PK#bAF~$!}?S-XBc{t^%I0v9d~tk7s0Ixx2g5sWu46
z&Gk>q95<uast=bT0|rsQ<!6W=Ktu%#harK1$mBAbO-8;2h<m@kxM*l-xVyW%xUgo*
zsVgehG&SL~+pPg_B8or@0m3=6ou5;gW|<|--Oh3{b3S-}G9{^+4h|0OJN|VrmcjOK
z9y3!_S#`a^XBPd^XV^yB$?2(S_}W<f_EKC-jEtGvy!`QFrM0!uH2QArEJ$2#;O%K7
z`&)XC!CJf1=La#rM!StE(=2oC40EkFJ`d-^q8@;S-oBKTS0N+DeRZc(>b*3w)P!nP
zrHAMMV6$$U2L>CgrJbb*_SdJ6-D2uPc`a>8nrQS8p=GD7rN@#V?uOFvErlU(Ps^i2
zlX|f~T^=@S@sc_yP$R3qqgvc_7%dY`ncf}SHuqT*2lXJpAc!SR9Y-I2*e_P;AytXl
z*{*kdGMs!5*P99M-vf>>aoLxv`ys~+b(%I$EQjN%ifVc_w#o9!--oEf#PUgt3(h4a
z(>Z-^4(^v6{nJ)fjEg74MQ(I7$f8sg6*UzVG2fm?>yoP;&+m5P?6h^2sn2exVB`%4
zbC{o+@}@0PLbTdZZ0}rL==545CSslNNg00c29ZzJ)pEAb0^?6gF$Dtg`5=iG(~ij^
z-rb8-ynXbMJEuvIyqqm>KqP#U$vb>vp*){+7u(-nD-XZSk8a7EdmG(+2oD)b!hi(w
zXZ^@X-en|Of4)x?xlohGx_vKR(n`aU#_%jFEz95=dSB!Byb%f(%m4fLV9x;K`{?yD
z0@C|TAr>MuDd~W1`r^QcjEHcP?<?N8S>1G*x~3$ok5;(IPVK;Ld082|{U-csI>V3X
zBy^^(03}P!=ZEz#H#hLxA+oaLn6k^5smttko7!8PUe7zv7W`pi#rZ>N92@6z%cBzg
zN5EPoCZ9}A`!Uo%kS}j$r`Hoo%l(eGE_UgnC1$oZDI!<%=Z~}J!&1wqgNqAdZSdL(
z);lrXi})J&q-J>%eQ1=5I{DV_*ykW)>ZkN}#zzRiQppn<9#yoqu3X@1qpoC(0iZSw
zO*-Fb2QDfyo)oCnj9BAg-CN8UfAdDb0DhREMCD{Lyrc?6QA;b$%gqKw<MYPY#A2(#
z5)pKkY#YQYgMPXkwYbVw^*S&sBOw#uQbG!U*yDRZ*B2&(b@|aaE>T@u*4kW9pA1H>
zyACv)M+Rwz^bliNbw7Vjo7c+Q6F<=hAVvdy@R-mCNdjU_%oW$E8ZE!?fsh2n<86Af
zzQeAopn}9w$gH?wqQ;nZogr<b&*#21Wah-Q1%SV{x|Eb=pQY1(CxmJeU6pGqYkS@8
zdC2SrkC^;Cox-SkyjY{^fw`tw=JW^=)xn0FCjLTEgT@b!74^$<QLj(e%W-${$+YuZ
z(X{v2E6aHydB_N+rey)^x>V~UP2Z+lFKeE}=-pW>&Y}43Kkb!JlhSVb+xgK7_OQ&y
zhSfr3>KCrpdm@!x0SXHmDhnFV_<OUGjqXF`%TtQ`ewzWKI`mHMXM&hjx)`b`;JBak
z^s;`t9Q{*HR9-vyAx|wEH7!q8w*oHv@xFUAA*jum)?3(oy4w5D$mAaSP1%}B$(rec
zgkQ&q_cL;Jvv#KTstaxJ-(hMMqd1K%<q(ZY^$+mT2gb|U1b<djTC}U)z`;D~>!!c`
zn=5<GFL%Sl;FlAt8OW-YtX=0=;2^2=(T}(44<EcZb&kw|VSAb}dm8WG7jAqPCmj)s
zua=sw4=@@QEGvJg)sMWh$>wsP0UEZO<8-yj{`B+|9uCgi+uO>@%4{qJ1%0s6;>g)k
zS~{vWry#YowDk61Tv|pZ7(-1}b-l~)^W**1>2Sj5>57m>7?^oL`C$4H9ZxJSm%)K@
zJpv9|>+o`$DUc-jX1P$IIUE5nw_P7EqvPY_0|Wbu^ok-P{kMmcU?2!+1ox}09-vTQ
zV{zbmp-276%<KUAA@wS~4zEYK9Fn_(@${smq?wr+0GFz$s{>^Om$v2TR9Q#oY<nB!
zd*5wMWkto^{eQ9E>GN7+JhVmpaJGCq5QQOe`P3ir9o!xrL_kk(zdIO%3MZ<m`P>@<
zf29uw1~!~(q|xSDc_iI$RyUc!SqFS6bc!9Y7%^#C+4VZ}xVSiZ1%+Sb<<w^7wY9Zg
zj~74zkx3zNXv)pamGBsh#U1^$S5cXJd%ku0>E3HRzqp8jj^62cf3i@i4O|i#DU@#{
zDJ>a;#rQTrJt3fYbx6NnI4z-fDl4_kIQ*Exg;LimK>kJBn)-42?`l(ppZnno$D<<n
zW1j)T!xK&jBo-Z5Ku`Ab*)Mz^58umeW|f?H!uTzfv`W-}K+Q>A4%|(30k<4RD{lz_
zQZ0J`AHY0Yf9CF@!WBZap(6#20K^(c5n?bQ35)k;?(P^}wP-SdHv3;i<=xmbL#0)K
z7Qf{EVi+JMt6B)GRTU2MrpAjUuOoAzZ|#3rs+@E@JO(x(#1#Xe2QT-rKTNm|={JL+
zuqY`hNkv6HC^Y0<<u)dK!de$a{f>SG@OoL~*Og2_DJy6M!?A$;OTVwGH*-UB=^%g{
z@c2af(_&&|p+ZED&IHK4>idEKu?Og`OiZU2ohG6a#UJ>+9N^I{e%{`6xZ(H1urTZ`
zEC2}-+V2Cr-|Xdg1kyv}k<CvlE-Vdk0=g+!->C49@!ZV9f&!cuC+SfWCd*${d9Zt&
z!$vhgK}EvjrHvI<06MjUb1z14M1-h;R<F<N$~K)S2=}|!jXrbM1{dDdn^%U^u`qdk
zoi-0~Syj=}$_e9I4f_p8W8vX{4%+kctS)Unu{q#~V;y0WN;5NlhiHR>5|d-0<<Zv{
z`DpI>pnt=^c#7m4s+)9Mhi`$Ns=5aQL_oy!aure64vvTn++o%ArNG=QLcieN`Xl&$
z?Q*)N%Z>|gQrk4z1v^3$%ovxL%O-{)Tx7&Oe)!+booLlYJ3KrK#WR$sy4}X{og&GR
zsnf*_VzXnfewf+*o~t763u4(H&xa{==7b!5X!eU}Yx+aJvJNjwC#w?wwH&iF|1!dS
zHtUH~(h`B7842sOPdg7zd-ZH{D68E)45wyrS!@|L>miKn>bxoej%ibVGTv-NcowxZ
zpXoes5Rp{=dbVsnQ+46X=_{;8jvy@kb0?`u@g7~5Ml%*ymD|?J$b!;I-l)+w+J|Io
z6P-IY)zq)>8P}BsEQP?z)e&ShNV~6>7Dn-k>8#HKI2t*}g%!wNW|=V^wU?*c6KAdP
zOQ^LF=oS(pmaLkpqTk#gFU(DM?vk@;>yswVbiX1ZB4V@IEVa6v0*gpd8Oq7&wlXk<
z*&OK;?-x35h!8OMFnQb^G&Fw1OW-gX1&93?L-m%k|0!7+)dz}(hK51Vh*B~#Bwt}h
zlW5@)5NIeUhR4TweczfxL!m)i*SkaFu~&0~$H&Lv5qEXwlUUf;{N7Ivw`As1M?liZ
z%F3cv{~H}0E%JKH?<+w@;3)(?w86m6{usyqF*Q32XjdC?_yAkAvrxH<Km`d2+3x+U
zv~_TFPf0_A!(uVDSfw8!k_V)XpFTx$nO64plYTx#L_`@G8Ae7%1%-v?lUcYwexxM-
z7jAb)ima@xpzet8<R8yrLr$8Sn#(f@#2W)c@G)p@rlaTuZJt+Kg1WkR##YqS)K9LD
zM?IEck#Z;fMl~OAxAFiMPf9{3CeBYJmnS15L&v}Xve<fG9|Wk|fC0$k@#*PN%M5g}
z-@$u(1gjURS_AWmU30jYn0_u*8**}T0%9DwT*lMevy+sRl#oy_0LH%{;e@Zf0|OJ;
zS?93J&&#YWB3~dK-;OIQWd+SuSyGZy9`xq{z!raQ$ye9oKGPoo#iMaS4IuqS|BZCQ
z!vu4m3-;s^=jQPJQJr@q*wy>uwQ3ypKt)AiVs>bAx3a#p1PL-=O0C-Ty;=#Lf0L1o
zmd82JI?`=(VJBzrdWR<;=x_$=`?K`o{+`L<SjxiUT77NVzsDYC6<w;T5*Zme>?4wZ
zJrni6n_@9YAkglbIBgogy0UUdYwN+kd+u!%oG&*3o8g6=O{do#aa>rM8<1t-B0C<z
zj4{nyp;uFYoDJ5}*r>4_?hEDN!Rf`FK53ZYD(NezsObUqQj+oV`i<-4F9Cg*`4D>Y
zwlEANcu-HX`#rkVPrs+9&CVveTp!rz=~@2HxUw?qP(;ilJ~I-~*8D2SIUD33Cz$;O
zdC&K$*6C)-(1eq7`Ms?X5pz$3C>{g?<fq~ZJt392gqa)e&bWmN9XR~qe=|6~zI1vW
zu*gTS!h(W$%Ty#ZZ^8yJwF;`DrcLZ|oOFZXx>{0JD|@gOFy+-(m$bAdqI%YuIR+OD
zA+@y$7MO+k_4U7|f3Jc-%Da3-YI3rZ{^sMdvQLYGfz<FkQ)ZVw76pc?Xn**V>Bim9
zi~g?{;0LP_hhpRp>|}?BorbbTxexuJ&&GNNjDwMIK#NDfWdZqo)1Xlsi_5Og#fRoN
zVQy(?s4de9YzZ3(nApO64?6cUH8lZ5v8mXa%b;RR+W@wD7FM7ZYsipA4U|=uhmC(M
z+dNEkghj6M#F<DGJUf}or85cY$<a}hKVHJCPb(O?UGO&6)rj%wTseMERCe>U&0w$_
zd!glympdEE`ASPW-o{Q>->76Zt{-|=2*Ao*BPV$OyB&1*iL`yjkO3op-<O>le0Gxl
z)LV6Xy$|ej_O$dgf~ttiAo7t?oM(XY-4~M2BD9Nn(AjVhNlf@h9Fq!vvNLg$!Zm74
z^?4K6o43v<uK28^FYh)UNg9jfN<(%!dq}Ttzd@L&4F+Q2TKAgFEq=6*Or@7p6B``K
z{d|qX5pD<YJMdC{XSWwvOS*qjQg0sEUfB0Ln*4+wH^DVKXPD6Wo;iroz!c1brc^`s
z!JtopuPXXm6uFE%sC~G9>gF>p+hnV?x;o52MfPylLT!P)`yM8br%dKb_Xt^359qvE
z3==UjWUn{}Y%@u;S(}SvsCqG*)0E8IU;d2l@9#aFKVUI^EffCi4EFviB{h1v=GAUC
z@Qpz6=Plfn<t|^x07{mDI60_0T(1uXvR6XbeDn7Z&|oKG{lZ`=r(c+VH|W2H##*Ft
z3_0Hc7@OLq3V`lF_T|S@qhSf19v9iX)A}pMx#G~@GCjn)&!TtOzWL(u*tRBMh^-c1
zKE#_oEpw<6GVvmeC)2ttJpcmzQcmW69>}O)kpT%Pxts)q8;k2S%f?&%9N)eyp3-F1
zoX0kvjAL?^ERovyydPd&i`!~q?pb=*SyOd>@Uh$a>LX`1NiVtyU|78OQ~6cVVx3Q=
zcvV0n%ZBe7b8pDWS^<HsWIcHLyiZV-FQ!_3klWbt06|flSYG`1&$J=oJLog4GbsGl
z;y^Y(54MoAoTlYPig?t%gATW&<*xXZqi0u4Pw~Hln=9;d+G238^R4{X&@lsy46y%@
zIq}^prRjBbMBnNfmn8p@6uu_fMV~T5Q_epo9L)=+e=kb_18Ls3gb~a4#Pw9=R95<&
zpdO-lK3=S!uQq{!n9V16yft*z;o?73RaKMe^|LZF&ySBi-kuyGARtiERa8^}zJZ0E
zeROn`-~Gzq#rcA}+2_?2kV{LIi@`y3dhN~Di*$Fg;<R?YA8)ByS%1sR0p_ivz5V3$
z6rb0#9&kbQQ<>bK%;WR>LMDv5##yldl*QHc^>T+7JJ*!^b}tkqWhns5!9Z_sZ(k+B
zL&wdC7sW+IO_ys;TUuJ+(P=nb&r-6oD0c7i@`%{k*_l#l=;;A0@3jPOIuwgrR8&-F
zHva#Dci<&gp>lF^0Q3j>KRvkzY)Insd0#y~egQ2lE#-iJ|F<&`HEoF=<+3}72olKV
z^ZBYO@dJ>2yIfAuK?X-txgQ@ND=Qg`3i|q+larI{>+iG0@(m3QK=IsYu*PU7Oj}CV
z4-^z&!2(<kRr>s6`h0wRGGxl?>g=SXqG&g_z_pn?9<`~dnwFNwKs|E3ME|3*S|D%l
z&PLyrL#@=-7!}Nlg%zU*2ZhAIdNF&fnAV$@kUC3luhsNzxvvDJ?~qn2!j<;hw?xCb
zPoT@ag1H@JrT+Sf<fNQ9{g<x9+{?h>=-|Gb!8iKEs>}iUXvmj7;77L_whC3l-z<;U
z6)2WF8GGVB+0};}Y7!FC&sH%2noq%SfIx@2L5~+pkuua+R5-l#=JTHP_1L%9v*s%K
z@yW@;@?SgrVStj@_)V%{Me)q$CHFtklf4QwZJ{;1m-9=(`8it}Bd5o-u&|K5Qd?71
zHRgwL@*j0bsIa`6Oooqn04D=11)VkzeyWQLCY#g7w|e6^N!bK2(Rdt2Zwi_mEp5%S
z%?8_bGkaUWXhe$$1!a@KDT|2s^nC4eC6X~q9xpMhR!smX<EwT|HzVJNHd<dnn2b5o
z7KDb|4)lTJW?uyP;CI)>rP+dLPR{%ZT1S7YD?m=BrG$TTX7G-S43MRLI9p<Jz5<!7
z%$sV^L%fHKavD*`h{40hrLeCGkMXXqT=86SdOQ=cU_}RNDrmfMvtn^&;z41015?Jx
z+k%yKPB8|0B)=G_qqeuLio}^A13ts_6oI@LKst*P`+00{?Q?v&J9|hc#5E7{@0QPw
zN5)vw{r%?;?kQV>1PPKutvDj$nplj2vMNBeo!idA*G!cL(s(j7bpMvzRmd)KDNn{g
zMa$Rp1$Z<d11fys0_eQG7<c%P(uElX`$yKE<F6i2JYuS3gShUh{JwCa)cTY}*JIbh
z5=V@n?r6(I71iI5pDy63m4E+6$V+@H8G-9x7=)V#v{%lS=dQeIfYG%X?w%Sh!SLG!
zyd<~_ymefr?okngk!%cGPSbNXpO}M7yCo$Q*?gX$i0=IpCK)ZQ6|YIbyG{A*E@!p;
z?bL=dYiLlnyZA%h+3JPvZmF<tXfIy@pg6s{+Uj<h7B>RzZ4#Lu?#n4E`cG;yMq&_|
z8tLJj)7??$Ns(AtC<yepk^eOYfUe2caQV|JiPl>``iK!ibsayk9fLU;yfI=#xzR~E
zaDSwacP4e?E-Wa|U8kRvlNCdQGz^<W07`zrIv44Ho=JaEku%^IAnd;tZJg!~A0MWa
zVEbjgPF$6t5us<qwp5!BvxTtrMFM3&0WcCdET7+vb<To7#E81kW6ss>tp5DDs_M)l
zegr6A72Jvotn8T2TyK1wTVj&tRUhkyLrA~E<VTC~q&Zu^-KGwA`T6j}TC?z?Hjt-A
zLx9%gTI1;ERObZIBXeSv9tq=KrT*6FUrg2Ug9Jp?OcxF4Rq{)7=X{1=A@NaqI*{2#
zgh^ZAoEdBObymYLeTuvHG4uGR+<F@r<%^G~PmigOsa$0BZe7o@o-A!_D$2_tK>dS*
zN)*5I#p67^yqZ=?b=q8QZEa~mTR<;SEfB=uabp4`TYxDBd_M^Z-&8IH8@|!g)Axgd
z1qcYvNrB!%MMX%(7z8qkyNv?OT6fKMc6P51=Y4E0MKXdcY3=PiAL9V2oSvQz1_Eev
zRJi@Ukwon#JKM(>cMp$vz$?`auoh3c!2t3$w6aRe%#49FoSxZP$wT_V0E!R^O2Q14
zDzsbbW>jkPey%-op!M+Zh>3~G=?EhvC;y_Thk$?(7#N7^o=*p;_V)Id0NViuVvy6$
z={T%xXkY~vI>^A*HinU7d}2b(q!|hxy`;=YoDUTh^{;-H9Klx^8OUhNK0pAcq^8dO
z`vXv|AaF<xET<Jo&Q3)NuYYJaVoxhWmH`TB61`ZqI#E1_DNz%>*5(upH>FJ0*2l%Z
za97yp(?+}VeJ`aui_I43!qnE1Q2O5l`Yq(-O$uBS0?zsQR)CKq0`&?hAte;w{<D)u
z`!0YlDlpdsCxHB$x8vs$?iIZafyMslHHSQ>swXBE%Fnb<6GwUEOk;v2Oz_p<URGQB
zJ5mn?Eh&^WCfXlWhe&D;bO9wPtfG*W;->Ak<hYBqW`~m_FqP*7RGTSAPJLZj7nTA_
z$ol0i0uvL{J|-P<=m60~<Pkb1l7zpp6g}XHYpueANDc`rd>x3}B}pLA_uY-^(Qh_7
z3v`}bDPQ8qj7(Kk(UvHYUDr>nPEN2N@OYrbb+TCS(syx;VF+kxecrD409lU++^IKW
z!&IUGs3U!!RFsr*I72ZkC@7dg(YM|9YW$;}62f0x!vvK-hesXJ8<m>f@Mq6bxSg=x
zl802Xnpci7!Fi0XP+@i94rr&Qj)u!${=AB`v?2&YVNE2U^leWU>wBtNvRpUqhmDx@
zS`N%VYuST9-~pg7(-`lLUn8s8T*I;rvv6DD@L<_XfZhV;H4xU@x}PgFhfYx3niK@W
zsVAc3L}SzwW4xGct3<ljd>dMez0d(Hc;z50p6(yQun?}!X%@Yu6hjYqL3*)nE%+nO
zc#j=bq??BG#;NB*97-c;AEP0-<t3#Yb+P4-<0ssRC>CGMQsn28Fs#}UuwTcTGGgFZ
zV0h8I#%i{r<)<|6E(_NlPn88JxGH6o0`JL4F$+W9TM6B2Yolu$!5?!)*V8Oy()`fb
z#h%25=2N~hT=<Bkmpi?v)a>(Pmmt1zUmgjUQ<ZBqb>>*8TCL2a9PtaWaX1*GA}wHp
zUcaKTqPtm4>Vl#Dta??$(Omtv>tVRs*ZNcUVp3FpbHY88kstvzbIM}Mf&dlHbRZIq
z+x5&s-CsgdQZO)jX=Me@Ig98k46yXXiV_o=JIhOO5v?v1m6ersbrJvb!3g(N5#W0L
zXJmwgE|!*-wzlFBjsOta0YCR01sWQ7JHhs8o$I+4&`rBm$!lmV0Z0W56dN1+7Rot%
zUsJu+&o3&6Z`uu-U?>*QF9xuPnW-rl2pv$`0Oo+;+uX3|U7y3rTq&U3fx0(3ywD&+
z2F_-~n6V%?r6Bh_W@lzj=B2wIc_=AiEz@o=a_WE2LI|ml_yU3yhhc9pz*~aH11sJ=
znyqW=k@drn#70~=S;te$i`yFLPu*e36!6!`2!NVSL1#v8!-joZS^TcWjSRJ`$!r0d
z3=$&whIr?qm`D_Xb7w7r%40o-;XWC?D7OtPK$)jFKeUu9#v{@fa3~rTb>&K1>B1ge
zm<DCQ2V&?%Woc>2QWhB0flHN2{Nu;fkFPV9&U1s1a&UTDYJ5HyM(W9iu54;R2lINz
zc#@}r3X^%{?J1Lf2Y?+DE<(nZf_jAUUI=b3O>NB|kBlcr6QatHumUeHo=QYj2!17;
zjD`%U4muFu1rOZK$OcPBza`vADk@e?4J!%Z@fiNVI?A73Z&P_JP$;!#_kaK`*|-`j
zJL&m?z0n5<r0_v;{|*}oXScn9<nWCV=oHLG@$umWc#APOGcq0i<j}#vyRi!2By^Es
zO-c3S-9(tvxqttpY~fcuGyxpJpnubOzypDGoVpAy<h%1tR(<AJy`iEV+*je*cE6yo
z;N^}@#HU3uh|=-ssGJH*YiP2@NgQ=y;$uL3X(5dql_Egk$l5M)F#F%wKf@Z09TMgp
z`YQiC*Y=7lxBP;7@0AQuVsV7Jh7Vj93so0H?_z`jiw6KRm}BZh9m5XdJZnn}I{o^q
zY_g0D5`fQqW2AR)<y0S#tK4>k#jzYV_P5Nbs46Z|UQwFxYpzd2*g*6M{PPVaSg=(N
zJTvZu@;)2B1#EF{f9?hwj^Q+=#8JO>5^!Y3=sY8z%vhJO#21fnY1sQ0Tn1_U3SU#h
zER@ZER=TUiOVCA?xOpqKEM3N!Zqe|^)qBg_lqt%Wr6dX{0dPT-DT6~1?Z=gH7k(Sw
zT0t1l)r2m-j5B020rEV7E;%(V{Mh(8+%LP9v@|TL_<CtE=tJ1K`FUQagJ_O)0$)au
zKcMWwy{3&cAtLpx`-<daaySsI6pB}h7RjUrYQ{*AtSl}0>qGPQ=LiNtu9)Ur2N`;y
zf1DZ{7&zRRyn+dUKo(L(tkw%K<rXR`9DN8W@D}Fg<UqS`z`(@JJYvG+eIFSW6^|nc
z7{iC+@a>y^lo*f`{g9QGhEI5tR#m+R$hfb~vEACRppw#3fO!*npub`fh0GUpb2(pO
z#l_8hzCBD&OAC~nE>HF~(k5EpNw{G2{B;#YE4#D4O^@Jflt2suNwjkxZn}93%`9F{
zJ6=cZ@BgVpg-h%)aMg^Kmq`!WgvLfnq>L9d;1F&ri-hy5#p6h$0SROr<*vh6(FFOY
zmG=z%EvZO5Zc*u%J2^R#mKvRo*S55zoO8}c-Za1(JpF>_Qd7&&)X*rjDMt0@g^^;<
zSOG=}FhynjM}FIz6Zc$2DsXbW+1%BgS=*T<qz+7P0WRQl{$yil1b7c+5?tCJK!4k%
z_AB{&2a_b$ZEvJk5ul=!v;jR=9DNbmmd}bNjhd})W(pQ!V$XLuY-93~I35#Pp<HJ}
z?U->h_$WmxQql(7+ZEUK3P8AqTY>n9K01bLt>$ISj!*MVTH@dshOC$#u0PJrlQjt-
z_@V){5dgo>@}OW^2x_~QLpa+FN8T)_x(GGh=jjU)(o?-~TFupheJ%Sp9N9(7?*PN;
zNDp82kDg43HBH|dIwj2@2=boo1Z~62Uc%1SmN4<ppWh_PE3p{!|7xpb`K9OL5~K=!
zVGsm`DhXYV#*PzTSGC+kaU(}Y2<SIE=~b;{J;iW`IHe;Bj%*lE->m~{xbvn#pe*yw
zP}Z~zuD^YO8NVw6nAjjYb&*eJc)+j_PgFt-iek?vmhY=kN~V7|sW2_CiJgt4+>#ql
zL~&HvlV%(S3>KiVIacyR;!bOUCwJ~nY3Q)Z>ZLgqN2a^C|K-`F!{|t~CyfVZNt2sL
z5sev;pV1FS4INM?W1UX3l)=gr{lrjX1Na4`Y3!?x-^{M_LgavPMeOYcRTA!d%+TM4
z&)H=Ii{yGmu>a;G1qG=0bqrjvl2+}n4kb|-(pVySh6WhX3@0%um;za8WF%a_g6Z(r
zzfq)CnnA%4q5>*t@bK_Zk|x}utZAvt=5RV-4JM3i=wEjbJNN)`4gIhj3Iv0E7#zdz
zbTD>*x(M|*NQ%EiBZz>U+}zms_V!R7(9SY45cHD(GJ;q#83~_nt<GG23UB9!s?XaK
zN{tC7K$lmSkNZ(Gsu%hB^|TO(qIOSDJNAd80GzoE+v?E&iLo&o;0&OZIRW-jo$u2*
zYTN#c3~z2khB45yh<4~%N4f6_@ztipF!J&e`P1}z9nWi@m3vjcs&}s0tR%g8zXt{<
zaOZTt3(HU}H-<3vE_`ArDUTYLN5mew)i24)%7W@GSrn*s(yBaYjiME$q)xW<E`7pc
z^O+Fg{zxId-_YKTHUS{+aQx;g^K9LdbKW`tY|Ugco$uy2P1xSFN%E#qlYi&N%E8^-
z5cO6#V?|%59sOZMHi=r&hZtifAf_mgE6#o>!^(;l`gv~2!Zc;+YDDp|IK41)w_MlJ
zWy>vV1^BZa!O4|NuFOTcLk7J2P{LR_mcExQcH{HTO%>_Jh~?zQR8%08O~bSgH&-bq
z_db?2UWHYog>zJuRnWK*1_5UMRHYqFRRew17X29u24kK|p#<8-#-EcgbS+OXZZ0o{
z<|&^8Jv!$TnvC%A7F;?i05v20>RPWKHu2!!K}ia3)-H}?iikvQoKxXOACm5v7e9DE
z<C*!M^)DxHBx^fks#Xa+OU&>9A;n~-)*H+s#8$%0xdK*KdOJA2=qAngD*1t3L+?u1
zP{B&OrBqANFApi(tlZzbD%0KKy>5dJRzsP7*y&_5ltuAnWRLrHgYI4JTgY>Hh@5ZQ
z^^N7AG&U`D6&N<Umh?uHtjs1lf`z!}c5lc|g}$C%EJxYV$qC$@0~n}Uqhf4Kru+0R
z_#b1nDil%9F9{NBYq`}3xng2LCvInSB7wM3d<<wh2oUI=|7GAbsF@fHq@<)IQdLn_
z_OGfD0cd?%{#;*Qmx$d0@`89A9vJBC{9L_K3u+|@AgD+e=jYJ~k8;e}0aDEHw4m6D
z<~yUB#8IQGt1CFW$b?W1FKU2*`|Cp`up_8YI{E7}AE1O#Q=bCLu#}V(pv$f|I~unM
z*TFc5x6L*Vc7+G?&oVJFot#*?*nUthbGB4dONAXxsdXY}b%V!b{fz~35Hs_4D36wk
zGaeQ*hs9~p6IO(^zi1lRhUkr_&Ql!)!i3OZCJ*4NO~Cs!U~4F*2&_)O=2lj^?7xkS
zon-(zD75d(*rwvxC_T^58=44_-aYeoQ7-l;X0zXbsj}(UlZYgxrOQ1cz$;DjH>c99
zi?#i0GC!Pw<+{r_y^(@mZt<YoAx-Ud-qx@ITX67}Z)?u3MHmYr*YlTzG{WdQI55kI
zj3lA01oq-gmR(j8LI8Yz@JQ5^=vU?ps!4N^U#W!BQz0Rj++4Cmu+0bOzOImbLe)L_
zzjG1?Skvf4lNa--xXF<bS81?1)m6J~jcVM{BQN7_sjH~?A;AUp<%<h`8z1<?`o&JI
ztp$=vgM%#KBB)0`r$e*lld8>CXW|&5Ms7cPks^Bk4K9CeB~EE0K7$+xPzGekr{^x%
z7gVZdBwk0gP?4X&=7^rtsfYy(oITPGHtXl1dvJB@JoYbW-eR(BTk?tB(j;fhP}~##
z4_Wv>mY88c3=)eqRiT()V`GU9Magak(@;Odb)}S4;ey6lV^51sDX&<_G2`ZcsbkdO
zi{pajpW^yh%vv)QzfOvqtD5`{J4XIF<H(?}0C-T-g<RjTbutV*OMiIZBSe>ncl~gC
zH(X(+`hFUbs*>4CVeqG-eheoHGm|S&feR)2KSmVS5*HVjaXo*EgA$Mmzdb0CladlT
zx@Kl%Ol7gP66)}X;0GB0D=ID~=CzfyWEmy|L_1qssMi#6v7cuv8JRW2MKtK}EqooG
z_un#);Go7aE{=|xf%ajV8gS;`>i2;`Kocymm$(DIf{u=k_~X%J`Y2{^B`qy2Rn_?m
z)8NaZY8N6Rj$GPMBwRRp3|n#G;#m}z3j57YaBhEKN*Bvzaye{u1Gwk?{T<LI{ZX4+
zT9#K=Uk78b^*g<LOwjF<H>ZX$!U6gdJiuUf@*5(N#^#iUvi4AR^~wI}Bf#+;6p<a6
zkpX|ZmLBpHC?@qCq5j)UF={TK1FmtDY_FR;{q}){G!Z(rwA9a!rKvv2jFm%>b*|F{
z?Bhtx-D7Iy;&oYx-rJs4QB@RlQjcg@)whF6Pft5rYjoUgMFZSTrIwh$0qje1SX%(h
zz2OpRvn|ro5tVy;dt|Y0UT@L6iul*Oo;&VMB`F-vA(V+EIJ7^&*fK5D)f*dkFu8^U
z6Uw@ayQZz1(#j%JOKVzEBTV(3(9ld}cH&+<!TNMu?s)<t`!Na<l--TsmR1{flMo(R
z3M6$o9w_L^-CkFo<?QJWQn~<I25X}3ub_ZT1J?$OLb!N(cd?3;R*u;01ZmlBTRPA$
zJ*Wu?uU`F~J%$gO<zsty2d4nU^ww5{zKYqUxq8c0yxk2kV7SbR2GI}903$Q5k1E>U
z{(${ycU4(hT3LLqMvZ~!)aC^AyaiQx*+^w&idJn-I5`~w*%utRm`Nk3@edHN-|yo3
zHfmnkM;HRj4Ivjz${+AJpzc8-s~bp3+u4fdH-dpuWWWmF|16HLjhQSq%3?469UdM7
z<T{z%a$$fkWSUV;aG^LzpS2tKt(TaplL_752kbJKw?LKD@ryCja-!}~&Fa5>Q)z&6
z-4BA80?gmq*0LU%aly7FB0JY#uNl&7eWu79BWcNLznCkOku4gil@PfF(es>?1#PWF
zeNRJ8jI>zoj-l%ajp))2j%6#Hcxw)q8<iUI&U!^3ISoFux;@P7`g{ow5}y3{jPCNb
zySrYxc{-|HDyhZSkL*$vCJbJLw#~HUlZ~VpVO)`0ng;Yg!fR(>Ne#RmU^Mvnidvvc
zU=e`2_eONc`<Iq9K>q*zKrl%?1_ViD4grdF7$}M={lyIpXxCH#)=xx9;6t0XkRbVX
z798SVB=#*yI}M)&=AzoLFNsETCW7fU3=s)%g~GyufoSwQb%J}~qd9s2f>uHRli<y+
zcKIwPhX4eMm&jl+1pg#7F*St}dyoahZ17*Ob+v*DDk^tZ|Aj$~hNGh+P%PR2#_*36
zK%o6+)d_(3vcjv}d_oezi$J4ElCcPT<uWGJ<=s~S=;v7-?zoT4*VlE)U*a;j91nWw
z@fdlTvE_&u+!madgeb}B^>+?oKu_<2G}|Q0EL@rV>ZoKnrLyd#^XOqYTh(hkb}}5z
z*$A)B0AT4YtymFN>P-ba>>ZioZ!OHo_)F)Xb8e@D^GEu!<KsY`>Z-PncNQtl<TEAW
z!o@?8ES`-&8wG(1zRaURsrVap)}M})R2D9Ohs(Vort7RJg{_m5vGE2rhiy$=m9t;k
zU2kCEzJ3Sgl;nWKxzQ;uy5+r)ZE0l~$P9r^70{5-`E?yJo(<_<e?!K2Weq!W{2+|j
zg0_b9HJ3SQ_>{4pdcVOyQjc~D45yP$m6hl!=Vz>W%e;lRzX>gw4wmQ!<=A0OJt5Cd
zCKQvRqT}^^8nl02Q4%wA6EHlj6`r25T~bife*vYWWA%;Cy6UT5TNeU*+lq^vg0i^o
zn%?gNXuP!X1Rum{*g6K{II;=~985`|G)uc1*<ex?lI)=Ef8Yi<@8~GaGF3~@Fi%_h
z@aont2i2(b{zi9->!-9G8dnO)3k($YTjgIlnWDwc;i@m?Bqy)R=SFp**^n@~!q6s|
zBqUCA_4xCiwP03lP&x4KUJK$b3>81JENmK7C4pIsXh5Gd&Q^$!H3C004m_^3GRf(0
zu!yP*|FT0U`nz^+uX7GobxFb;rBKo6LP}{)`6Ny(Q_Nga#FEAC+QBCaui~kg62WnU
zp5yvqX(J!7Mggj4c9f(1Yf~;w{+ZQLyU$e7;p_!Ql@vdJB=%uLs8X^{<_r8QI%vjR
zO+kTy`FoS$GVCli%S{tZDk}j2DAk`8YfnQ$M#n@_OAD)({psjjjPv*_Kt@Z@QquHG
z{`|?blg47R#Ng``CpWBjbdNHt8io(x5|RoL;5m^0fDxe>3l<g@0IblQ$c;9ZT`$c%
zJ-NQcnE@JHTH0E_E6^{7bEgAxS~p{t+a-q@M(<cM1K{PZe(4YTzg_@x@?t<Z?C`jS
z*8GN!PBsy0(tJ%H0TRm;xWO8SF~>_UOn19Tz70id_xgm*=yk=?uT6*zw7l*EM~oVX
z+>(8uoT_d5{`TRxzTK|J&;R`!BX#%Bl=h5_{EkeHw)gjfBGZhnpAdRXmf-BA)uRkq
zD}ZCeTGvN-t9T0E@fuKD8X8=9I))}Teh&^775uJ$#OLSr3b5^nmmrBNO&pVHLFw9Z
zXzW6IL(Eg6*cp|9erXN*@)Ppu2%q2BXm>i6E~8MdwwsbUQd$ag%ys3FIJXKcb<Oi*
z2Js0}d)P!HZ+=ce9RKHGu8t@_GdBkDe8aDiX-58O%|pYW0ZX7qRmb7GX6FgzDeYL>
z14_n+1}5f**Y`r8ydCu>j8vu|VW*EHg?uYtUcM7f*oXC|fqHSS$1choFd`+nZdMxl
z*t3ksz7>u7(oUz6l5xS&6`Pj_`1>wWH1vE4t@6VZbd+3oA9$S?in0rVO^P58VD%A{
zQy}IbgeAT}MMdxX**@w}BX+{X1h^5q!0ET;c{X9s&fNpKbUzRAcofvs)seFsY7@TE
z==LzuclkyU3A520r)qjUha6Q1jG}g@o9BF=-ZApKnR}u&`3YCB(w*1%&E9=No{;B!
z8d;F=szRU~KYA#S!(K;N*jO9)1?v_gLL`mR_FD!IN+amSg@tv_(2F8K&xWBof0nAP
z@Q<|<L)V?@+L`hF95hQ^LL!oU_MVd=2hVHa6Lu!))57X~by8q2r4bvA{a=X+R2<qY
z0RkE3^x%n5)KJ#x>Rp`8NMe1n&Sy28RNdrRA#1^8?qi3|qk=Nj=xrH=4|CX!K-bdz
zxk`jJm5tsQjY~^T>D=-5lH|w$WqCVGee<rvDn~u@F2D3!!qzi!UtzEqN`h@H$;j24
zAre^(3HW*>te-I3vJ+q2)pDBd?Td;U9=s&)G7G+7?EW?UNcfYad)fRCdIl`r0{6#?
zO>$8YS;6F5m!BV_a72Kd<yVWfHGroKpoaWE|1$ytq@$ywfE_tFC?*0IFwXUL6e|E;
zRz_xSZUr;-`<FRX(WOJj%Z-lV{{M%)w~VSP>e@iJqJ*TRG^lh5(y63$cPZUScS%W?
zba#WabVzr1cX#JqzVG*aKktwG`#R2Wz!;o!_Fj9%ob!o!d+}U@iYxtFI@4!&_om-=
z2|}SJQD5G#|6peQ_YcS{-f#o~6zWw!HyPUbU|@Bo@s-8lY!zZWAYH*U5Cq`9fY6Kt
z1+wKBH0yjM5XX%uF<;4l>3UIRfc~Z}p7;+n<NgFsh)0hpGwiMNZ;7$-@rLc83y^CU
z<`q%X(KNgFT@AWfII`M&#9(`5E+l*rwE|B_pj)PbN_6*aHAO*`a_)6uACoYGfJiPy
z(&B0Lk?!@sef56nMOuXbEGpU?<v;a50|OoiE-xnLJT>o7pK9$9_oW$!ds+u`h&fG7
zM&SmvZB}{*3QS7k>AiazzTb3uy`x}rKG6E(da&r-wLIvwRUii&#bNqbuDI8TLi}{M
z-?0}lGQIyBD3h^dN^pZ)Zx}{lZkZ<8m^<6;zrHZ;WR}_s5zE%9wZe~&iHNWw=8^kB
z=EL}-T%R$%KQfYOWN#>e5vybz=DGKj#i^8-Xkkfyi1e4)-A2?w#Ivo@mqy!0Ti@oJ
z2%^OrT_0F)I~n8p_i7GVn5{#$gnVj6C5NuT?Pz`)&aa5wqSuIk4dFI5HkL}|*0R0X
z+r$6L;HQ06y=@PR`rYTK8_;F*%>NPs4H~%r^I+O@?U57j7k(A5Mf4f~(F^ZWT~m51
zAe@~u0IvcOmuD9isOE>Jk>S67j67Jm6Ap23lOoY;-We{n>ef2lOt4gGH@RKS)}qE>
zO%`vGYk6r;v}Nct$65}nTAS|q(r}m4(!SRD8Srq!vA<VXLwrfdU2Qudd|hE_IEES%
z&KVLaB|Ys=!_fToCP5NJ)c-#4T}DjFEIj@?pej8#%sDdbVL`LVmH}Fk&rM$d<E~38
z*}stVTMX6g@W-TINos0JD4hJ)HDCT%)Y+fW&!gM<zio(nbLI6e+2j741Ehiw2<&0_
ze4CuE$`(ecsi@mtC%||M2sKY3!Rk&D8xcAqU)Bk=jHf%J7YiV~_>oRJmBdRt+IQVk
z(ih<0uaXldHLNt;c7JOUH*>Y@?t9O~o)DT`T}6QDdgVC2q(6|9s3M)e^kyHK7~bAo
zo`h=J*v9-%ihe_|QP6#_9;5OLi?5$)bm=dcKbEwb*_j?}h9^s4U&U-pY)M~TojYE1
z9gFA=Sa3G<uR1^covAXWy&YuW-QF)Oa(JBKeht;NIq@};pE-3Fx=@8(`4BTTSzKX?
z+0)aIRk(Dse#^yGJHHTKCA?m0X;|K2;Gt-a`?&pRw6<4$gZ$irG09-h)JpHOcOGCs
zrM=AzmhxZ-5Z)+WFBpH&4r?)uInGJGeHf0b<ef6}<F!>k{WjpYe2{v5Uey~z&CV86
zre=viL>QtuTI<=mYUwLnhKzu)a%3Q%ukao|=-01tP?i`Di~Ga~AEEnG@a=4f2!WQT
z4?!~Y9_F80$cd^_XM6kg-5m@hAt{+Fg%$q20s{|wHwfUi?`1Yo;NfMIl^yKu0pNsP
zjWNDJ#mDCX0@ZGP?WTrg36l@q`hfV(#>9kPhKJ_|Yh~1Se0iY0ehZSqZ{<1@)f-X9
zs}=Qz6*E8xCSvXQj5nlnX=w>;W1^x|bdtYmXiRw54i_K8Q4<FSX_nmuobm0X*P6w-
zi&#t5aZ|&C?K2Y8)qsrM709IfO$6UsttIevgpp7pi6h!5Vqu+s2OE#6naSwnXy8M9
zctm`0K}JStZjhSf(jPaE!y;p#`fNlxe+Idr5cuAGg*=OB4_mB3R2rgHzvAKGXg)l#
zgHT6>vkx>ltQK?LNUjr;69bx+I)En!YBV=DzFw{A+a>^%_Mlfiz6+vw-k(Mp2&ATq
zgzC#=FP?v!+pL;%Ap8Wo4p#9d{k_d@)(QANbv3m`w`(7<#8<CQ0eRlL=0Jl>0I(zv
zi+0$K=0yw|?Qj1e5i{w9V~DCH41NU>u%C}#rnv*4+fNsF0)|HaPa_LnhhcVVPVBS$
z{e@aAjFcXy>94rxz!he-@vbGj6$a>UfVdqTrdFcyAnatJc~PH&g14MkIEE3yG2FU4
zQZgterjyOF|Kc|xAow(=4R^~ZRDJ|bACpBY)+@b|wxaxUy#*Iq_oixbiBfO4{J7?(
z%tjrrYMyy@@9)(R19gi?E$2tv{^4WG<L-M(Wkdf*embJmM#llZ5=1CGCWhDgE6zL|
z2K;>p>=SQD4h9j1m{=gt&~P~fL{%~JGCB<s_W@;7R7_;3_?KYs&Rrmul7(UOI*R_v
zg=fIUnctA40mMbb;>uw}J8uQvHa=Y=Tz14U85e!pn`F!sPLo~fS)U{;WJMw5{v@Ot
z&!5iBQUjt9x;QrV^45l@X2zBgcaXw6Je%KC_-^C~{6?M<bQgqMi3W6q!yK<p<oVTS
zuWe#WJVJn@%C@ux8Qcp$1p&FfQK6Lk)*TtZYbE<$TUVDblyVRmIK5S6C8^>1_LgU;
zXF$8YC((n&jVHObBImF8{z1`7i~SW5FCZ8J!jI%of6j<e6vdxrr=c4agdmZR2IKJO
z>f>VRw3<~`)8SrVz7%7M$;;aSZTLOM$IjX?Y$a<aj@pBIR1{*)?zVYaMve;?Otwn@
zm>k%mUp?L7MR%&59X5;lOtp491tGw8fR%-9YXe73^*nnt4+Nk|qf9SnN4|vF;)d4w
zx})6H>21Jr;hRF!K#H)tJDahJ4!995dqc@A50lAo-TRY-cfw$X{V7Hn7@z@X50<&n
z%#7OK$U?ynO#8lpeNHch-cGuv^c17$W$OvU2e>em48^5JCm~)l?G0dEHD?Emp1yZF
z`kaU=NFWe1C9_*(iK;biSrYuKKa;sRBL|&FD0vh+)WjQ?@f5e(rnj#)sTpz^j}kqj
zHN2E#bxb^4G1Py1<j1`<5#dLl5>r8xW9752E_(G=To+q?NSUN3cgkQvq9Mq6LxPED
z&zJ>q;kzu~kglu;4l{N-KDlDH)Cc>%k|DQWn{dRhZA0>Zdax=X@Ygb4J|`!<=JdxJ
zV!D@2EsFL^yl2vqWW|Mpu9eU)-mkr^+w}TY-_Ig(bHy>G)9K5KA4Tq5;a%GIKB^4Q
zdY8`Dnx~4ik=&*$Z6HOdV&sA%A-M58N_SXuUW0S>#D`J1<ObPyh)Js)a)~@qVaC7x
zTWKh80k=bWDVVXY5%km*Hl>O3ZuJjeGOS1Grs|-ENd*L<X`ngN?piuZ1}l!wN^2FA
zYjojK5Z-0d^Fk0VRs4XIDPXRmKwn`#3=Zkm&8%Cwb#{O00XzEo?zkcDqn*NFORj0@
zinCbc_$RLP>NIs@6O&hc!|h=p?ErQepk57-MxYzAS?_**&gAFkSE62n67W>+aHPeo
z81;LV^m(AK?@N=XI6$axw~0__`O0kY%DeUW^t8_Dj9hK@`1rfvMR;Un-S?H}%_pxD
z@I-;V)%|i6_@u$XL138!^hx3|w6!){D6>DZM~r|<;Og{L=ffL7|9QB*z^KMh$mFoy
zmT`CI_4@S3b|><B;kclH`n_4Wx3~AUnA72W?No^-43xxSV*o_6F);)pBa@Spx`4IG
z=S~3qb4NN^73z1Fu}L2u9v&Z8prHx%QgL^G1QiagX7}RaVogQG?2?i};vLwNmDzfS
zQt+?5oatWK(eXNg#T0@{N=g8+0ro&vQyy_E<DjI|BNj--X4NP{_m|49(8%xKBL%#7
zGg=yaJb1Z}xkr$dohzOEeT%4z+aItAU)qc_y^YweGL{55c)^I^)6byC*ERMVa(dGr
zco|g%Nx|kl{Dv65xY*oXapCpf^5*1$reyUi!qq=1a+C^>kGV-n`#ZXDb$#@O0zk)4
zJX<3m0E6G+ar`M4I4KZv+Qae;P`E$F*?|fV>i$fqn>nY8JolC(_$Gl!KTrnxef;x7
z_Zo~4kS5h!890h+xMsjl<tEJ5C8!a(5!#Ii%F4+lZ7^=MMr|cwZQJ&z)7`zYgJL2h
z6L|I8fa$=BZ&ZE%2m%BUwcOHNb-7-~)Skhfzt^m%9}PDvcN1nsq_uylbd9#%_m_Ep
zm-PE$H^IUZF_P382+Aat+Fq+hmNgDtlcBo}KurmE<@^_Bt*zT{l^l=*O-=G@>U)|~
zXFldADJZ2|5GqcvY<FP@X-i*?kHBbP2zfP0NqE^aA5>lt67KKf<3-7jUK1mj(gX>e
zZM~ZB1sS7&hkE7Q%=Gj*X?T~u_uzI)|Ia6~>LTrl@$q24dz|j>*MA+HPgKbJdb<g@
zDh)6E%x}!y)}~Vt%J|B>IC7-pf7_-TiUjEy0A}VB3WEL4#KgqoY_m<*qQKjxIiTH2
zWp5zH*wFLo`v6Wz_A=XVvIyu{7LW7s!hW#f!tbU-b5ixSfkfFTC?h{V{s<SF_K<{D
zi~XqgV(#893dBIIZd@OFH-X?(sbnsYlZYu^B5R%L{b<QSVrUY3PzhC0CGhR>@fAs<
z*HoRU{jL4p<Fx9=#z3&Eoh&=ci#4`K2B5(u<~l-+2WaHsg|XYV8F{iuP)o<xxaXqZ
ze|eR;2@{8`R(>*}wEHHcErMVB9I8g0cjVe}|At|hmKP)Ez#8-Mk$J%^GAF5X;Jvn9
z%;5UXTS+EdwvwSY%&@PbMeC#P(c-JeKWH1&57XG98VX<!`3XKhxxDUET;EN59(o}t
zCV5-^yD{!L*ZJ>yz8%=tr+JYSJ*l`(dYv?f-!*NW%xs-#Ay25~qHK$zaoJws1Ka`U
zuP=d9^FwWF2tNvnOj`Q41c?7^uvvEs{dmQjnHyZ*VuI2M!Rh~8Y?1Y&Cp8Z>uKLBf
zgqh+T(&-`+Ge_GH?}#%HimCoOtg|)VnD=6Sd(BU|IL-_kN-L-`7)&Gs-0XXeb1W!a
zHTEH&M4_Dn7U$Fv2W5QaIxDtD;pgIqoTM%YSqg5G$4Fl0Gd#L&e2>A6e$y!irjn?-
z)>*1m?0d)NPr1wq0CRpHWM9wYgar1Cv*Y8mu^sROB0M}GNSRt#40d+|8PG?_^UHf6
z=Eq*+^uY`he4F0(!n^g8hU2wjB1*C^{I)Hpj+P$q05CJVAL4rg@~bd>w$Lx{0UfH*
z<B3T10O;odcD%OK09*usXcEJh!+O2@kJTE$K2o@y&<O~*jYr>;k!1t<RH=3|C~3ez
z`}_O6Zns%^l(un`Jv|8V$7g3SP?1{IE6q0Pr6xBUo3g{>Zo%*0KLoU|6<UFs79p2?
zS_Rlt$8sdkuC6RiOm>%=n}H0>%j+|cPdPfWqf1TC&)+Z9JLKf3G};ZPeNIS704ghh
zJ%b1v<mKRN8R+RBPd9>o{AdF*)Cd{D>gs7UJNX<5D_h&(kdTY3tEsWEL{{?zAg=QI
zln1O&!ooy!zJQ<{&_U-`ZE?$;ymKCOOfse3IcZb801I_ac}Oy4nf?P`A*k*yQ9S+-
zad(f<584Ez$sekD85ypZ?b?|d$rK?DHQRR{`wNB|7^+jyag;E+^AX_=B|K$Xn$A&U
z?1ZB7)M9;!=~R^i4oy(w75&>U7+F_~F33EjJ$EZ6!Jf}FHSL$S?mb-L$E*!E2jAOR
zAJr~#_?ozcyqxQ5>}t|f7Yz8)F3R`@=C+QQuM%umMVh^n!nT|X)tMK7>n=s~C?GDM
z*`W`6d34qlIxwiIE(L$Iy6(7h-(xW->A&%~8P%r8gnBT?=+qWRzSP%&^JaKzVQ)Wx
z6-D$;adY6tX)!6a`Ft|+h+F{f0>^T)q%?PK2W~46`{-$PZEI=w6j1b5N-OLw=k&f!
zwpnVIGcjfQp>i}>`ZqLaM5)?xy3K9c*%&-zJ>0xm^Z{}HKCZlj`P}p+uN&s8j_K~N
zBMmb+L&<1IS`=Q_AjUkTKVZ3E#JSquvvER4+#J-XGV^j0Hx!ip?A|sVA%ePZ!jfnm
z#Mf9oi=aS>2@H~|JStip2oDbI%^y~0vScXl94rYr^PXZmL(o0gKQH-p=p0<TJ@jTH
zsx_<v!$*UGxN`lqUGk-V=#!Sz;c-)P6h*MCf)S}pMtpwHX~$U%-EGxDENZe;O<DE)
zAA{Pi6By^ZrJE~Q|G{cQDFU;TJ-B2kV(zq~T0T@Z;+_<#O!LLxy}cR6s5d<ro&E6_
zQ9o2>yE_@yGt<Gv*gx>Pc0;gh>bu9DO^52N#l}o%(HTGPF9-G#mIi3)`a$Y2ulJJ~
zI1qSyn0;g#SC+Z9t$2_tB(|wQ$qhZkkhVFSX=1)p&|!-AzdA}{cDId6Z)(T6zQ(=f
z@mQyrZy%kGj*0GK?JJegQ!wzub{ErQZP^Ygj?t*B4-CtHZ{qfDAl#kzv7}IHtoLL2
z-*7~4sX)IbKLd`>l$gC8q`A4QL6XyjbG&cd1_V#!%Tw^QB%{;|F(r2=zmy)7;D=x#
zCE_TClGgW`+YFB@zmo7JuMmz$;rh6o_hZ~pIU7mfcE4p!Kms<ejBjJ-z6n=CO)u-~
z?^Y-M`=ap8XL$VLN@IqrnPivE1;5p<=>(;vwmQE{54C!@?=XP@3hSCrS~#JgGU+3J
zE2&gDwtfEfyowwS4w~ekYv$Z<s4kk^v@l)K7Dym{FuC`}3PtRwH*%|nl6eM9_{W1k
zWgg9rAn?L|&dSOXq{CC2==?=CfK14F4%Y4L>@0+ai(3nTR{3112q+{fY78(v6SwS1
zJ1VNHvGqqo!onytz0?L|&IVT3)<p5})Ya!OD5U<>)sa(B013|d@yg%#pEA?aS2i|S
z3<n7i{Lk@Ut?ljtM~aW{-o1k;si<ndue1RX`r%@ux}01_R#qhtWG*j%?TOd#@9!5B
z4F%eu8q1}}Y%(ANii(WH<D*!FEAA;GkB>8Eo(04nkH;HBSnt*iYoLIy16L{@ORul5
zUpBK3LPuCwSXNZviV@YrZMTalYACO*{j@!j2^>;yudmM!7r4mDJIBX=-H$JD<06AH
z89=%a0u_~&wHq8yhSLS$P>C9VhRDo}HgQ0!(b-NyLIPjphsV21+jqVF*F%|q>g%H-
zYTVIKa;fP_8L2oz<P|&8#;`EgUH;TnUSpv_8M<pX#LbMr!|CT1*)vE=Q88f}bw?=%
zTe4|5_E*eJ6EY<wWrj>XfSsi@2?d*B*iLVb{=T)bKFi8_1DrPqi|KkUXdgq(zgZZF
z3Evdw6=5D-n(3O)tgn#L@yIjF0g63A&+ltOENra1Cgi-lb!O)8GiEV4b<WZ?7RC0~
zz$_9?x@V-d{Tl8fh>DWdB<14|vCN*H-rXIO8qQCpUu3O%@S8n*#iRfT8$~tI+CIHl
zo%w^m&uJXkV7wqcA^`!dq`RBBbFW!Z0_u~hx-t$<YFu)ZvWkLWwu~w*cH#h_$nNbL
zEv;@3t{zQe$HZ**GX#%Lb#HZbQIO~6oH=rn%LjJoQBzXB!QIX;EEI(wWBy+Iw@ae9
z<YqY98SH%IA;b^nxkVwX7?B1;PvddjVDn^f3$x^>#5B+|pdcqRG}t)C!w2`})s&*{
zu*E=#_S)+c8qsS)j8qwUdGU0atn8U&vm!GtwHim2f&%q7%s%|t<ukNyn#oFBYymXX
z^tdeF>b94_{_sRE%Gj>2XAs~=lQU|hjjw%oPyTMH6N8kVitIwu$WhV`Idv#D#sD85
zw~n2I)dL;}{)H?t#sf%pB6GnBZ7tg+ATN($T_GtdNujzIM;nueFkQD)9i(RXhuv<b
zX3rsACN3%R^3csup9~n=%r2J5i<tIRbN`*1n(`OvT*tv=W<8Dnr}qRr0ajLKBbE%G
z+0daa3bI`C?5xdAEFhl6H8!RZ|0r@di6MOI*x;EKejeNpJV7`^$luYdts60<eAm>l
zvI@u(N`dzki;jw(m>M&_Gi+z0p`;8B5dR`RWVURbojpS?>{m@#Os4!kUFHog;T<n+
zc}bI&qLAvjzVLnZ*siaJwl?u=tf<(fN)#Gu>Yv)rTS>q5O}U!TTWuxe)+LxY3F#l#
z4_5}$aweB#X3x~ess|BE2x@9RARe)SA})XW@Mv3qb|Fmdv%n>LkU^0ds{FfxpMm)W
zB{?#AX4t@hYiHL4e|yK>*zEO)q0#4WuS0HkK91)HJt!}@H%|s@W0QFhSX=Gqt@L3J
zm^VpSnTxva<Q|8F6l62q1tlM~y?R%`w@Dch9r>F0N`wUSTJPp&3s#Z!`*`-m{KWzR
z%K;Upn1!O2Nr%4)$-q~(9!A>P{lb1Wp{fpRx$*_OB$sV_!{Ikcu%V7INr#++bfL+#
zXs_<ApWoYf7yOxcp8hF5dt9K#1&7XsQ*pFIPDA~a_Ce8Blop!KQ@dc>FDA`Ifzn@o
zt7<?!$dbSev0Z;WJd)<-ExmyhF<<!#=_XLeQ<l#<2U!@IImX>GH_=9n$q40+R_QU{
zzBw;fV6@Z?uZdCKOFT$bif_eW78{!xMXIhYim3|Bu9HCY=9^@j<QqD6k9}(LT1unI
z{n)RUR;JQGo6fu=d8rP`e4}v~vwG|HQ?75!y6|jH4xL|G%D-Lmm1!V)U*ud<2D*4z
zrrOxCDN79}CByq#Zk9r7f8q3=4_W$em~Kz&9*|u;@@1!&){l;cw$9=uawz|PDLg$p
z6D3+gA><7A>{EUX{{pBEfsT*>1;{H*jEv5KIaLQzIrOh$>h&oNs2)v3<ki%~@FZ2#
z)C%(5*mnRit)rtufg0pP7e~ZVARGi<AQ(_60Y|yf$lG+8?X4{wG&C7CHEJ3ffG2S1
zw8JB0GHh&Y0PhK4a*=O5cgGnJULrS?!cz#e@pAbf*aWFGAg`I<54%1+x=BhRLtw{D
zh<7`|c>o1ZS=xL$0zyJ{f_o5G?gOy^kjrV%D+>$ffJ4dc?$Uhpb$%@{V+Q(!&W;WQ
zUw(7*0}vns(IqfZ0VM(SXUt4Yw{mQCd3kx=-QDf&?R9mWz??cK2gsb*TwLxpm&}@d
zy4H~KKPoaS{R?0OXZW7J@o@`#qupASEY(r;(zSgZKGJ&H6Y+QP<zM+jRUugG4w89x
z!_8)zn~^K)jm~6U;*UF&-F-{Vb`gu4pOM4aPu_&sDEmLu*Hk+MpxxG(qG#6HZMa5h
z7l*T#_DxAApS&*xJl{jDE3>;d%<!b+_J}z2?=ukrqMQSe=>tuI^W+=2CiTW|h)$B}
zGKUM-;zFu7N9L#a-|oKqJQfB9iVwPIzJB2Pt$pN?#zW-GkIQ&2k2)tJo$}O7fLI$b
z)ckZj_Q_ZP?y3tKa3DpilxC3ic^sQ9bh+^8XnAxKTyXE}?Q?tZo~;*zPk%}kM3zt^
zc*K2}3J618o?NJHe9r92i0AXD&Yn~NGWmIr^ZlZ&Z0$`33}hMp+%ZESX&CzT4J_RZ
zqGteJsX5}gb(y<kbH3fj5`}*SF)@wGQ-nUa%n1<R&spD+oTqxmhUkH+wXQHS#)Hdk
zlJeS=t$;9mdNta{dsaq9Kyi_pd|DC|?05m|ANXj#Bpb;n>^<xu>Lk=L$cJ(;<FTH!
zwGlbWG;$7<Mo+qQD>G)3LpmoUL*i2fg{HH2Ew0kY8YM=w_D~%6H;J=fo$#bSk32b2
zcr?BFN%7GS`y7|6BJgg8??K6eot^#uASB%bCtb#I_x$BZGtAx9qaSzJ((6tmwA&&4
zkDxFc)%EmnP`PkS3M05yudJx}WVfT{LhOCCb30HEjC997TLVj||Mm!2uSzKBy=Hs|
zM)Jwy>qk7Frk7eA#j*-AalCdvwwK1X8cj@0=jvs^Fe^75eOTPq-U!kDWNXIlG@!?1
zI21SAv~jZ{KtbZu^o})gp{KK;vZx_q$gm=A(;#O}7`5hpQJ}9Tu=T5%N`>oKjOO~1
zGOxP;oJZj7eJ&POEVt_^KQra@Y;;g!utQK&OvaCfZAWVA>LD~g_=lPg*>6?_Fh=vP
zgVgrOLx~AbdHI<a=R?%59?R%kzy-aGkB`?ABsDbLx}-<EO##a3#*%q7xBW+E{U0$o
z!gtsRU3UCnDAE4&pNYqHAJYH*Ds%Vu4f6m08yKDc?<0}vF){kwi)7uc{lEnk<_ACE
zn!4Z+f$=tZ>HS$*Q8u2%t+(Z|5u9@_I9sVo3L4f|tx<PjDKW9il@6-^c`A=JVB@K`
zcTi)ttWGL2!&goRfvF2nbXx6ri%2gtoTW87V1Q4&HQ~|Fw6~Z!=EKSn6(72GI*bGc
z<d^jD{~qD+6TPBhG)#R>!W;Z*kdJQfVG`knJi`MI{z<mU5@yL?q}qY%kt4+QF_Z_6
z@sszzuO4p#?90Xy4a>DQ?UAXS(<=3YM$>;ER#WeG6p`7e=@Pq2jE+qd9scX>t+MQA
zEiJ8*MnvC#Z#{3$C@(LF{7YKe=(b27&nF@(S~gMD^j272NojHM?kD&gxBBkw=}Y=H
zzTBcBiSya2#IZy3#K+uQr~k~xs(KE@n66HW=TPcUn6$+1=6uzJVpB$`{JYepMXrDE
zPiI{TxUeoiF#7Lpz^8ktfKaSd^JUkf@c+%|+r_3w5vJe~lK+|E{P0zL|1ZBam%f?W
z5&6i({CiK25F<7`DROeRO#t>HO6LQVR?_siV{=5TOfBNV#iMBs{R|C}7L~-2Jo>7m
zr*ABkDQ9zby5#FubJcpsdF#xHl5GAgE$uh^2X;YbO<+H-otgs4sH<;1bcS`A-fz}?
z`rnHjR=%lcOdu+uHz1Ut=19V?IjYfjdMDs*mFfRvwQO{zYM)#bOP}Pj8*Heq!>lWe
z6dfHsGBQa`PxWBMnDEW}u2i!&%c~FQ-p?8wB{4IB!iLZOJ5I6`sOp>`!ee(BF*A?*
zBk!U@zzLawmi8@%;UjarSoU*IG#{|lwH__vU}NnWnIi8l<=${<)jKIGE7#v`y|dQe
z+T?LrL57&ACk7@uvJ2n&iXpfWpS0bD$WUS?`%X1k0<YF~<s|NJ!OV$tykxArj33zB
zyNAD?`Y?(G2AZ+%U8io*Yu5{H_V)Ej;_<0TDJf}b4Gv7`uN`rD40{Rv?+D0Xc(4gq
z7%iFsy*g*e<?K{5NE9bF;4Y7N%)UQi>oPkW@-^8!S>Z1^Y?g{cEmuN`&jptHi|Y{~
zTGcXMPF5{fXB;>#@+hbC_8ma|bFX%ATL4EGJj6{jbT-PuWOX}^V0)XpHa|T4+kT8S
zUFP~sIx3-g4OrLH_Ddx8zd}MGV@x1aXZ>U6U}0vqes&`Ods$IYb+gvO2z@`;WY+uI
z+`qq1&X|CpD^gHE86;inwDtMx_im6~vAI73jVP44pY~UF-v*8VWHmUIGjTvi^W?rF
zx3CYDr}}?y6Ak^UMu*uThkRBo;fk0CpH~m`LWxy(AdcK=F4peRVRCr1GMZV{rW+Ul
z^S~)c`B@7nU=`5~<SXoY@u)v*xiVi~fRlz=nrw3OF}fn{h`HS#ZwJ&z)NSo5iTMmb
z^~k_Vw5KZ30qivIaYPk`oOZLyP?8TIAItma4BGXI>dF)e_1^}VknO6Fn~#7X!J-dk
z)JOsPA|+|cnhXnx4&+?z&hSuiMDwSg@6KpqW|BIe6Oxyb@R>|6tA5P6XVz~6HrvkU
z_$W}&<^Ud>b9-y1$71ahpnbYNHu_5+@;T2gWbc&}<AZCfzFM|Id-bfPDlrP=bxHI0
zH{wblu1w{%`r1DHIV`m?@P^xceEv71JPilIX+mW^X?eK~-X9IlkV(Cpy4%~N23Hmq
z#aF&>o4Eh)7;;Z)j{Z2^=uZR9|8WbGSKL?}Q)JeZ{oItRUL)XAbM@Vhikg1Y%1O^z
z*Z@033dXZCxCp#P%dMFjb8=XLZLm}>mz~w{<T#pnS9-<4<AQo+6>}=LM}9#@gW;W}
z#X=Gp&{*mb%`}@{hkKzC@wwd0T7v0gszi^i-7-5mpI%YyScu&NI7(0@S`LYBk_WG*
zh<aa<x2U7u`Q1;-3KVkA*@RV_o9CZ_VccBHS13Fp2K$xoJ?aaH1kLpt4thcSbaSo-
zSh`>%f&SbrTN$f5nQJa|($(h<BoO1r4)1C)pW@$SuwlUSCQ@Klvw!767?5dlr*-(h
zUCb@9c_K8&fnjQOHL~)&KYaLjv2p&_FU-C^8YrzH104c#&rYcXDwEb{5TM4`9qsfQ
zl!Yd9ziUcv&-w*kBk%me1*_lciXdmo<=>$Yqauc#o}ngyr3{S9eNO+dbKq_7E&g#k
zx-E7Awf78W=l0=uT|9!VX>R^fzh{becUf67zLmvs&dxsfN-HZf-knG36ovrUM#5XB
z9Ciqy5un48Br#hkFw`8z@?MfY_oTw;9L5a>`xQ(JSa;1UMy965&!OCc!286Mh?!2T
zFyeWEx4h8rl`Fzpr<WP38bs0pl8RDF)w^f9ZEGh`OZ)V1x>~P)$IE}l;rz<Yg$oZu
z7x+*dw681ZzVq(iCwS$1O646O0}lZ=ENs+aSD26T7*KbAKdfhP|9-Luf%CnDobBIA
zZ$=0SpL=;Exi9QcgNW%jF$6%i0Brc6pQos<Q!y)n!AQu1$($J_InL+nxU!xw5CH#D
zyDq`}ayma3(%_H0rat7+#J_QsQpCzFsEZiFu~c~hoox*!eY8o4$?5#JEE>+}jBR-<
z&Am!3=H$T6C#NR-h=+f(S`nJU{TG-lJ~&Nx_VT;l#-GLZ_V$MDFL5}2TC$|1{@)w)
zX_~C5$s{FA)~M&GwYPaI?|iW>b=n8-!w>GL*SKj?*)Elo(55@>i}~jtR1A%7sm-Ua
zB_&L4T%yP=7h25I#*9I=S<(PlG0Ib7s^a2OM^nFAGBU|iPe^>;G85(BjVX@!5gFNT
zWLwHxSW4r1uZR@pu1DPL)%(x(7DWHbci|o(NXVRO>J6x2n3gzW8ItM_HtdRYgF1}y
zNU+I$8>f6nFdgAKKzeRfJHKiE?_B<$hx5)AJe;ePg&WsLzX7gEi}`B%wd(#NGne{*
zN;QZC6MjY4CmX@*)2D2?8CoUv#UH^r<735K(6qzh`RPb83k|lo3h!HLSm^x1?(f@-
zMSl@7QATE#i4K?dl5e=I0kK8WFVjNSK<RS+B@J*F1BL6iDb0RETMVkTE7@0xDHaPw
z?P}ZMp;Bx|+$192!Gec}laiH?zx>4(?oeIbt#MRWl$JO;ILoMgZ%@d4UsQu`0MEXz
z|8B+GH34<`fj}WsC0f>luctd2a(S4&xf>lh9>T7Z3?Fk96|CLko4Fs)SU>Yg5>e35
zR2X$|o*3c$@7?>jCs~?_=o+ON8l@5A67H?xe9D<tQI|+&9xSdu`%5uIh6IaAaOw`e
zD_5GJzh}U7{fzCgDQZCHVD5qo$}1>%-jerPppL_#XMkb)rYl`s*|N^wCV}gYf=eAA
zu3ifB)x6vB8z@JTXluUX8JL_2kr{0#-u2Ixj|0Yl12kwnu#i+wuvn9sQ$1OnSmCi;
z-=))|EsR76I`B9=T=JZ>rZoBVtzltegLOV?R7pM{%2Z(kdA5H2nG>EO39NbnB?ZP4
z$jLz%a<x06>H7T;Y@($!Y(O<zt=F2wf?wml<0gVkTyDMND(fYJ1YG<G>Z}(wm&30Z
zv~JG##`bPIK5}clOB`^%ZMz%jT0@1vDR(n>8@+y!{}HEweu<=%c%k;K8aWM@!{|9j
zhDY|BU~oumiU<Hs5AD7A1k8V@5?Jiod-%G{qh`h+8_dYi6_QVBI^INl!BSNPp`j7T
zscEHX%S^j=!1zn&80z6~@15S<SDqt5%V#$cj8>A^G&F^f=Xhms>J}{?_wQ7&I>AW=
zHumV~WEco|<pO#$)UI4qQnG^h0-yEh`_aaxk!9c3nFj#+v#;<^1{Q&dkgRNaQ%O@u
z{f(7l=H^ET5{&<5#H?ndBKY+?Ji;*aNy;Lni-UoR3N}Retj^V2>~2_VXE%gPP7djn
zZ$=)8wrtNc&*9%cVd>czse~#Iveti*uh|=V1V@|A)E@a>;ny4v+T!uhzTKVao67Qe
z=5+)deA<-*LIdDDK}0An?lEbUSAKei^cho3{I5~~Y-O8Q&FlY}Z|NT0Y@J4{=;&+c
z=pDIPhI&p(4?zv?Yd9o}O#%ekBu0^pv3+X+N@nzIrOW3WrQ{fHt`Fd)u94EW212}t
zeP4$ln;u7La}lO}HPr-vUlF%z99LdQ#Up8O>nssx^lVk}bhY0%C?fz^QcA`kA)z26
zp<tz=<EOsCPlcy@{mS>z>*yY4iG_#6{TVME;c<|%#h2-N6mnq+Dk`mt;E&ej<sc?@
z<t5ey)sX)lW5ax#b7v#>42~DG$^fZatY14xrUBR(wj?S%3aA1-*j)xwT$Pk))?QmS
z2;6Ot^&Nme*pecI^~BM^HMFg)B3h_*J<5p<EI$nzN+kKe*|P0_duf&Yh{Nr%5fznK
zT9_%6E`#1)_l_(#YvA54z_*qvM2zm5-FnR93-HHYStEjE6g1F@aOo1XfaBr%wK0@e
zT#N{hVtl0Y_J2>@CxH*bV}YBEbw88lxn(copOPBxe>B`<3-6cY=eu2=+PTD{{O<+a
zYWuh)|KCpNGfNE04*$0^%o1PzC*zi(0p_mwcz+TSc1O+ZgAk7$8$izm>S!IKw53H~
zb6n86Dl1DuSXfO!vzV?AC-xU`8jhL*Vypt{-leh+8(7gB8(5O8>KYoLi5qCFmz74c
zv?#L2mtR)KNW*o-rPKz@EP+;qipsAf6ns9Y$^)l(TTV@&b!@z9D%KaK>mBUbIKi{4
zWQmtBn49YD?gv&2;33)GE;(nd{v(e$G2;z<(!NSva^Ou(jgN+bI)!2RZX3n=dZ}p&
zF|pP`BGObTp=58NtD6bPn(|`1seE`(Qt_%BuDnnFL{Udt2sAVt+4>vS<xD0f1$>9y
zfZ{nhIT1D0+Y9jbSK4J3-(=0rlM4#U8<OV3XjW=k?f)}9eiZEPuK%js^Z-0G=(pWg
zlX1nSh)|ci@x^HQDfoFv^G`XU!)Fq-7cUxVsOU_s4)@p|#|#Ge9=B>)Ee`Irm2(CX
zq#qmTwV#@V8iChsrRrHD{SsR0w$-BPY|%~Jo5RFtWd*53mZ#X8zUC=dz%T*tXtp1_
zWS|RAPUbS7yw5IVdmLahSwDmMu^1hz>^h+aI7`nyAAxdIJ4H^GzC<F6$Aw?<l1y-5
zZ?9*X12@r;3t`JsIF>1!bRr8d3f(K5yw!jFce$}{8v&;D&1ql3R|hE<-Hp^!VZH)v
z<DfwCb_&+-wd=wZn6H}GTmagA4cByba)Cpvy)he75ALYh>aNKLE4F0z1GuDl57Wh=
zz+N%FCz7QaFE9n3rW5H^Yu&-ef@d3v)7Rho+Rc7zbHif-fsUF!h1K-&hrVf(a(rsD
zGmrVCYuK3<-+A{wmn;9d=)-nTP^sEgcfc6<ma8+=+jO#f*6C2#f0G#PmRDyjMspx{
zZqAO=dHJs`!qOGy`bLeJnn3QztQ8KhkW(&RDu1}TIOM5gOWhluvG2=aF*g&0f#fMd
z2$}MZ)^Zt+cpXoNetC@lZk5s`=;w^P*A>}+!gb=9p-Y?`Zqg<j9~+w~Hx4NZJ`baJ
zvXz)GxKO`5yl+gjiLEZhLc=UK-aD+wMNy55XN|uwN1Y2xX<%QmdIS-h_Jun0?VW<6
zLbaLN87eb=Pj|mot;LxiOCK!i_MB^p3?)Fhmo`E`=4m&0X07?)4~sIED{-yA$vo2O
zOOo$SugzWBl>S+>xcF+<W3;xJ-bJ%+aN`8>6wZFE{-&sX1?sbT(yW$qQAr8px5}cT
zCsV3p{wRk_F?P|vFu^FiBkQ=<^0nGs{i7chS)J2~y09U1=6-XNE2<T1=_-(?zxqeE
z(WO>ZHSS>;-5vjb3J$?Y7as8($WUaY{L=Nyy!?V-0^T<eP?-qz$>je6f&A}oBe3X7
zLX&&`0+V}5F*QyQ2~e_maBo^WeLms`Ci|+i1~^@V=4P@Q0I5UA(~?&&%q-T|)hC@b
zld{kfHI7dysV<L{`}*C;Nyyi)E&;T6f&{xd`CRV_K*_fJEhZM!GTIi*eFlIkY;3G3
z^xnQ6R)%lgcgJ8>X%d>BTmf41_~a-Oa?rgTg20l9!qoU1Ts)P8M1FjD-?roK?eg-|
zL3_ePZn*3meFa73DD-kw$$GV{Y(oHVy?V*7sd*4O8M9hJO-&>4_pOsgN_2L-R})dZ
zij0hkj6zXPK}KfrOK1mIHJTu_sU$Sd=CdomoPiQ66!!){-u?sj(pOtES1BnXUopnL
zuV!X?l)OL%KtfF(DwfVUI~RB*PjaUyqav6&0vZpV+s=*eiBMwt`HEo0{sOgs_UXC6
z+hIVT0!+a8_=NGvN#7>a*yw0Gduub6OJImAo+1Gp?u^HW@B%<->DV!Zbu#AVOJC5g
zYuBUz)NA;7XkPh_lmYpb6NphS?USd?u;aC~WJOUT3Idr5XK;IZ2PGxMKw~CTtrh&6
z`{SQ^tC4bm6fWYTP8}__tD?dGR$i=kw2hXRl#1@%*m(PN|GB^idTR2+)=WddUyh)n
z8-RsO#cf~(Mx)R{Wh*qq<9;6tdQX-+F~tUXnv0l@xXb2aLXN~lstO9D`I%Vx2pMN@
zV#CL~*Hl&2e%2X*1R4OZ4W}nCo^jJJR8?`QVE^`xHqhmkXI0YDa*L5HCnm<@;kr-u
zNi?|c?(GEfJ%J%;Mhyc+jY%c!49=)nZE)OWq?hDm=RYGZMHGPH0c|*m!f^0YyT1+>
z)6tzGz5?Czs4xTn+rv{pK)_XpDXOQEnQfqFAS@sxQ_fhk-WFRZ!P@#vXBFK@p&(Ph
zkgP*`C3kl>!6E${E%sP`kieWS1~AG7vSRjWuW2y^(VmM7<Q(!&037hAe_-OohB#+b
zWb?;I%ge#Bhow)(qnW%teZ8Ng!5TC%R*WAlwxOU1+@T{AI={MJYG4y$bm71Q2-r}6
ztG$!L#MorouKaa$MMDM?Ew>yfDvd1CQ*l(4G%b$KD_DyZ|G93@WoURtLc>MzFNNrr
z<odbG4FSXmq`sBs61wkT0DUB5M}wY$nKlEIDn)Ud-m|as7?-hI?SCO8ES5Lq=3&qT
zMpbWE9b5cwMbjszrr=)q!uzyY3~CC-@=-+t5)^n{cSJ|TE6VmTvNA)j0JI?{Dr#qQ
z_?BoCoTh=&Ukp9~8|UdDk6!o}Kf<q80FWYwB%B*MxytDLA0Rm<A*NM_F}aS@H6Oo}
zo$o2F0tX=2Zz>EV(npMl44-D}bW3aV15!$)U}Eu8!v#xD<{R#}ryZ$%XFE;=+}>Tq
zH#E4C2W=Qx(oo#3=>D8f3JF2_1n+gkA1@~d5`e&r_lgq}NBnI-v!7fy?V-32%r)2q
zU>dJ|vRamhHSbUo?LV-x?Sbw}DbZ1dMOj&;xfr1C8J+yspv?JC-LO$5IElwAk>n6M
zTn-}UNM@**$iO&Pmq>}IivyE*{%0VH2@)6Ufz{XNbK8u>uSzN^3Ml)gBB<d0`J0F-
zjT;Xkz^Ab=zya4mqbFwKY$q8~63V*sT0^x50U>DF%E`~~7TbVO5YFekUry<^Xpf+F
z<GE)@RrdDUJ6_U!uRa!5rtsLPcU=5p-RSiupmzu8)&|Tp6%~IyJpX}@ho@1;8R$%m
z=-bwv7Z+J>Jr_D@%gOVRMqO0)1nxq*{l5@K%kU=;KQCU0+x_x1MB6i=U^h4)Bas#(
zHN4Vkj`)4Thu<DYlBUYgCpF>CAKAoeB7iu`?Q)#2E@Le=yNj$2@VGpMf=?vj5fP8(
z3u03iQjy~5;LVpc2Y|lEOdZXDW?OyW_;rZ&v}&Sj>S=uk5u-D-zmMh^5wxIwC-4xo
zCH1@od=PsQHN<YS@^)r0MBK2diuu_xXsw!pQyLlIHt%A#aGp=+AfeH=d!bGZN@AtH
zs{U^EHGzB`RohjiPt}@t;C|yI3<LzcgXCiu=2)u}!gclr-G;PpAWtyxniTjas-Wit
z3UUxhX)JBrNB4gCk)r>i1P}yT-^ftf9`62jnIk`iJk~BBK#Lk6(2FGVM?P#e=+FVR
zlkJ&4o0M3JV8U9*M+WG4OH~?7O3;)JX@5$c1kLQ)E%#{d)5xVi(r%5#2Kp=C(eAC%
zncrmF3oau?prk~r%^k*{etdJka2E!;xJ9<6o!vag3#rp<otOO!2lENA3r2zKvCf+4
z323%*PK9Cc#I1Omo>yF`UVC`eo%hB%5VR`}C7hNdIzlDlzykd#kEc|@p&EWR`gcNq
zYEp?4BO5djpiXB_d{`Zk94SIhbij`NRieW5G@V}oaubMzlL;$2T3p!z^I^XzkEGdQ
z>*)Q56JBoXvuYl_x`_1F4jT$VeAmZO!@0u;U^EE9UL3vC3a&g4{?|+SS4^2|QQfUG
zj9X(SoP7Z$jk2OLol2dstn9bZ>k$f`!6;w0QtgRmO9n`#&g30coYCY%q84>;<SDUL
zE%(Ud;f#k*gX7(9y3sRWaFgtU8WHmZd@=ibyt9f2K-<yKq+=tqbr??>$|TUXdWcW*
zx##^4{t?(Iv-wvLID{NZocK+pQm=0JM(wPbAy7IU157xzqEtXG)ULN?bSkg2z1U1O
z<h~)E!UH&Fxll8(_yny_wd?FnZ+{^GBH@dF1wSVoyr#9Qi$-VLT$%hnCCk+lN3I&P
z>1r1bb8|vKp8b?gtJM_BpKY;FX|Pw~_FkvT;|hDLBoyF3V3uh+H%I7va0Gh^tnS9>
zzgjqqC*X$G1}R7X`t{t?oC1G@{tMW<%D&&ISr_hry7%LHgWHwYd2kV5*g;KM`7+{*
zb;l-mXJxU#gqW0CpzQ-?ANypp!3Vo|T4@8a_c^q8)5=ncLN|*c(u4v^MROYR^7YO)
zZD6jRIEFySXO*JstDxJTR=JLV9bgQc+wR4L|B_Qtv3*#@J^89(1M+2hTH5DOX&>7c
zk-p|Z0yak0_cpjlx6ykwLXDq2)SIn8X*&RnCP2N>ipF2QLx&Mi?rHlu31*tbg>}}-
z0qFg_1z30D3_QJrrfDxH5*>DV$&vrjxGwf@`as(Z=##c|*V(z1wLgv`KXlCLqJc@y
zyXQ1B*JXJboDZ-Riu$j*`c|jiTGY|K8>g)13m2>;Dxh}Jw^6RrA7UC)e`Wj9^Yk>Z
zgJvohoMiPF$x_<PmY37Af1$LTYWs^@y4SLd>qPVvH1)ucj2<PKEcJn6Nc&k!_lh@w
zXmBA%4J{`jApsa1i1L$agANa5w$xw&)_c~o*T%S0)6;)eu97Wbe)JqjsU^4>9od_T
zBU;}yoFIZwx6%bQB^<>F@TNkEXMwa9kj^&gv^15pP@vFD`yZv?{bR-IAk^hg|30%D
zLVEOW5f(Nv(GE29P;1a>1dmw6We%K15oFptO+uL?k&$vZr5e&XpeD%MaPf^ksAs<7
z$w<k55Fiv9Nd%xhqq}B<n51II**Rm+%Re?xC*VQ!?6tz&S*SDyHw<?7s{OVN@vY6T
z%DySCCq7iBKzaH$H!{1nuMGwP)Y*u5^&ki&sb{ezAfs-yeatEdJT0zA4NGFDtBbd-
zte$T+EDS!!p&?C{nyWVT06M4i#<?D2Ix4z5{Z?zvWc#O1A!yoW*@$0<N!9UTH+ww=
zf*2Z>lFr%rtf$j{ji1N;c>yy0@aEMW&`KE&YCf928@)q@x_i6NzuP$7pKpYJewZv(
zo4u4DvyTh=PbOpz@L1YYLeBr?0z?t2Pe#4z8rxcQx_JqGwJ@>n7!w#L7Mllm3$i)Z
zV)d1~)uySLHIUtgrBp9!-h-0>oDgT5_xHCmn&2Qf?lyV?{VaokUE=k+laWe01E2H)
zpwUu>+umM3L%&w3`+kTe9Y&Kvv`c)38A|-|J}|dyA6F6v4y?`FrD|nsb}(A*kITZ%
zygV3@a)a(jC$6A@`^=`fix2-w_8A3uU@U;F+_D@FpfE8l8ake{8yaF`*1I2FBQP%2
zO<~tXPJ<v@DsglM%W6`uIBC8i(Sax~(azENDPgFh9)JQveGEuJKVdyV5-!m7_)F&l
zD2BTN-WX^i?cFs5N=MBK(CG)zTJzwJ^*2b_l*23=U2O}c8EXDq3!WmH$b2^%uD=2m
z^2XFfCvdH(%CX1YaUjxz{#+gJfb8&q%fZ?CX!fsq?D>T)DiKd~R3aQys&&`d-Wm3W
zvwLf1LBxR`q@SmV@Fpy!y`UGsyG$L<IeXWK{ll!2Ou%Sk(0o8sBuy^?mu@ywnc!pK
zx&%v}n<cuKmLD(zMsUslk~vj~^?LcF;;&0hsR|P+1lT^xs-O-BAdL69Imo!<M>7w5
zG%@{!dFy6NHsR;!)>hUHt2<@?_|NJX@GlJMH89~{h~o^_x(vv3kNC+@YSbIMfab(t
zPuAjXs?r}z=r7s_Wyj}GvM?M3h@#Y-mknJJvVZ+ORX&WIoy*c7+CVCeuB7>OuZrzh
zbeOWipohbq5;m7om<BL_Xo3tAe(1@my(2$JhVs0@5gB4vR0JfSl@mM$M}o%NSU5=I
zk7KmF;!CKGb5@y~2k>~I<ssd>Qb%V3u0Qx{7zF6{t$PI-Kia^#cC>U}Ud`-*kLG%?
zsd;kibU98c$4AV2Bg3b=iZ11t6dMz9$?dkkxnTnks+vPyOjML=6}uhggoyMe8Or&l
zEzj)QnOjBisbUTORYXv^BP2Wk90EcQg33L;3ipdqJDS+-y`ASRfMBBK+-zlK?RvX>
z`+%>NfVX>xLPSx#Yq8{_ASbswKe4h>EUrno-@Q<z2#}b~?**{ub{>v*<vD)Nl9I*i
z8=^@5^M63a5=7bfTqe^wRej2I>YuV@pQUA1ZGfleR$pByV`pp&If8Y0BH)7EpNs%$
zDO-epU)(UQDKs_t_j$Hk;PW^U)GKHEj<z3@2Q>+yuZs*Hza+o5yy)CQT=exz=5R6s
zBqr$l;29Hi?8uFt#!>FE)hAM;L>nI){>!>8GrCA?gwYCG9`EqF=Q8?4JT%tTE-(bZ
zNq}9Q=f86S@d!XA8{J&_2(<pQEU5=Z5u$R?K!sep`upJsOg-oeCZWOq{mUE3QLFBj
zmE?~DL(LW{m}}NrvY)?g5&MH=_}TJeX0Gv0{Kp(g*R%V@WYZJ?_Jb<6#j^FY<@VIA
zw;4LZ$i(-nD<OS2QX|`ybX->vs_5<QoiLtx0;~I}xg~1Vzr6!qLtjK0!GZx`>m?5y
z=d4PE*cUYwkPN!uqnYWLU7wyDIN7+D6c@kiQgUrpqSyScw<LuZjfaIHf@it-=OL}A
zVZJ!=46M4MV%jesPD~?=XlvmjJ(YFj%C|a9GewX{qr>~ltC_$aX>}Zy6Hqg4Hd(zi
z$@J!2(*+f<k<8}YJb`8ROl8hQ(WN|jXyfV*qg$^JcA>4Bny{pK@z##+`n#|f!G1`)
zaM1hb&mqwIhJ~eyRSDY<_7>P<!G$JejTix$fUEUY0f=EiE)!PaAxa5FixG1FW`G*b
zA7!ev8ZWV4`SNJgM;Ia2n%i6bHZ9BHalb6EJR;iS2{ke`R#LG3!cN6Pw+5>70QBDk
zAT*pqBQm@ntA_0o(%gU89(j_09JVz)gAIXj3RF62=xC-UNA;NEL4}jfK_^!RS_bcC
z4bnH)w|ZEOuWsmzc;EuzHRvA<JW~cFxIh_w!&QPvG=<dE#6-v(O%Q<WqN$jGMnmqs
zPYDRm)RfG`!tcphn<|TdgNWPo@*|LXoHS_9yZjM*WA1Z&hPrxoQ=MFdkADO1d?C4B
zqshMS$b5M`K$omYsL9#>Ku7u(`@#-1=N<VtqJb`DvXg843R>UXT=KXcUo07G(>7#!
zW7C5)ik$-x)hzM0Nl8<XoC_@^tFdG0G%`iXjS}&=&GdAdocSqcl*X-#veT39^i;%x
zQ{Vo+Ddo0oEW9R+7-Dy@B0<Qi=Wq~}?1M)_!3C(j|FE>)c?C|Xc@Z7T!NOt;>kUR%
z2p*TSqoSgnv1)BV1S#}WiZMj^r6=f`98SzPAR!6LRy|=o=;30Ni>3lT`~Pl(YgPrc
z#!lh&i1?ddkhlJOn)a#10s7$93+fW_5r1K6sHj9MR9$&frE>eeh=Z5166*j$Lfo&&
z#MmJnMt~X7l63$+pRJ=FbG%TbEj4xb-p+1TRu2B|UVcHrix>XZwcMv9pm$XaZCqw%
zF*w$?xvq$#qur*dcr*#g<zGSL3W{^=yqM8qSRrIt`T2{kbRh2D2-+%}0r?0l9IXGZ
zn3!jtz<B2>Y7Ph#|20q`lGc>0%KMY``?rdgR4DhiNSoZk%!y7{W?uJ4=!4@|?NL}0
zuxFH=xIEl@gGoeDkec-{2v))722S|xO(z2c-PT>bN3EG-D`lXjj~Et`Qpn7Dg9OPa
z%5q$HRg;GaXUzoxB@YOV+D%Dz8aKu}NV&LhAP^^&E;Qg1ptHJmd*UowLSM&^NkO~_
zYNqv$8#$Cty|Ey5QZ99S_H{q+n^@qhiX7=$_q~*LMM?!~*9I$4B9b8Bie#pH{jEY3
zOk@6Uh{cZGyn}J9J+q^2^Zn=N91QefUHJt{86#gRO#$}<L?PfaNUh4q8}AdwMkaMC
zW}vj9qVa}7P>{e9-vpygN!7j7V3RsN^!q0zO0Jl+vifH&UKYB{%qpy?o^v<+O6#5H
z2+Ev2x{?p4&lfc7A`PvTadQ2oQ(|2ku)pLd)QG<K&+u5%vRFudmR6{{<(Y;i$3L<V
zfFwE5(Rv7y+ixrXYM~M6urK}bws%z3<6Q`rq;mid{)o*Iyd#sxxKDSgr?YE{kCnc4
zu_Vd|Mnw@Qdug$0ILW2Pm%A5BFpMvS2Fn3#2S_j-sDnT!<hJ1Qy`X@GnkED|Es$*K
z?rN87!`xjQ9*$x`1caP${tXN?G%%jAi@eYW=XDt%9>SHU(5iEm9}mn@1avnL4eP9-
zcQ5q(Ba(y<ZQtJdm1X)ds52|OI$;N8m*xs^eqjZ=3ZtGMwRb3lsgHg9D7pW<6aeF1
zoz62d<Hx_3+}PaD?S7DZY9|a3fkO<x8M1acn#jBZjal2?W&lhd03<2RuE@~)sKG?g
z6K8p`f@{b6-z;ufMgxhpevo^#jNj*nyQH&*{k6SjfTY5De`aOh*EJy;4N$LB6(^A6
z2i;k$GDV;kKocBt(xRuMfrG3M>H!^~V|%K83q;83joVGQ|Am;w$us=U-|u}cY=kj*
zWTv9xAp5}1{DB=40&NNmhT=v?I>k=a7_{$*xUI`GBHtyvHQA7Cx(2rXmsSp@O{-Hm
zqI8s$FCem)PUWil`q`Uws%k2Qzu}>_w#=NTavA0by<g#J4dv+2mn9kqSjY{-)^VzR
z^hj_$=?%dk+Isd0sL%jw=3o74Z09EHSS(Pl2Vj8f0W~jm{5t?`X+;1Q5L#DTBBK6j
zw0SMYe+R4y%ToS7M15scR$H|8Rs<wPMG=rtLg@x+kZ$P~Y3c4XXb=Hu0VxRyY3Y`3
zq(izJY2=%pd++$hI6pWJcw_Iq)|&IFiBIsP?b@ZgpwFrMPUAvGHee@|m)9k$-@%J2
zcy5MLOe}eOxu2e)*WuipsXWF*AShif{#kjeW5)d8Ae+mD-HYL2otjXxd0es*@(2lO
zfqJ>i7h0lOLx_;FG6_H*n2D|Pwz(j`6ceM0?eYr`!p>GP56StOnA{dDjX=V}qg#CQ
zD@9+jZz8m`-1GNP5ZGx&t3_UZ*Q7^&hv*8-#>}fXX?1o=QyR2>G6<geR#Y?q&5OGA
z)a`ov){+WuD;7-5U_|KDkczU0Ofoi%`p;0y*o)yTe}J|AGIw<ki+)Ow*vF0hGB>6u
zl6d7Tm#T!knp&NcyN>SX*NWJ#j-|}s9U5e7kH3ysG2kFzNOeRb(ZU!I0`rd4{co9>
z7zq7QY~egLRhFmgM`A8_*U4Ku0lD8&aS=x%i1-WcT7aM$051emZXp>WpF=}S-`03i
z1&9zWOH!ZWNAdDi)<*HbJWA|;r1yMTRch*+=A6j>xVPe2Qr=7?Ten=E-$}wRb$$~<
zhIw*<`Iw6Gm@ER+MIbHu8Tx?e;hp=0_ezXcULwp~8_b65Eg)IM!O<Diwwkt?Ow6O=
z7H@vlU#Y`b@kS<75h^ruiw!{Q#PYo*r-`|8&8Q$9cC75!7NEg{lKxHSWBvN~&r@A5
zKSde<%vj#`8jgLhKSYQSoTY1TxheOla^Eksd5Tfp@P(s<-}QnXh;wBTFWA=f%B(|H
z6kiG_8(p2+lAc~iKAMzx96NekY%WbUxLlqeV&B9yE{cO~r?IV1NAV}2eM5-0g+>OX
z1&)^+x*EZG6PjG${9Gv*%E!})oL_e$`>+%-I&h+t|2#5%7%#?npZ0$jRGnyjNf1}(
zM<I2I$&Jk*kgNu_ue$6|Ja2v+sXS=?lKaUcUPI^zMo2vT5cI3Dnemb3{vyHF0M#=*
zSw$twqBnb^0{W3D{KP-V#EF&<PWec~JB*~QF`lQP!ERchRrQ;0CZrKTw%d-)JrqrQ
z;#i%BzgJvqEZPg25%BiWtz)F1V{@9<P;%XF`+2Iv82<d)o@hxL(_zN7dZC6V=(at<
ztx#2B-oy+FlW|>IaM>Kz<ly*Kx6Ka-o6UyW#s0xCw1U7D>~MLbhbIpX4zF$xMzg=r
zDlsUMQAZy*il?XkS5_WbUNO+|ay_GDXXWV~WqQ??{0^dQ<m`#LiaciLU;;Mb5VbrJ
z7Y{3s^B@1OB>bZF4L&I>CN0SdLc>oN44V`gV`WIgAoe|8mZoF2QtGvTmxZM_X-9<{
zhiddt+pWMtu>VOWXA2+t)oG%hkK7Kg|1p!5($YG!zGN{gGitrIv9>*3sVSjt(VdYZ
zbtq@)p{I^#KP900ob(o=K}IV+bVkGDZdF)il&{c8kBN$?sA@^7Nf&5xsQxmcGJ4K^
zQSO~tYb7xtdPn<ow}ye}bq!86qqM-8zUzJrE&)wV_GqXt6G9~CcwdFeM9XR*tJ=CG
zS6f{@>kE0`(!OLs8E1rRaN?+f87CGb6NSPwW%{}d0Qui<#U(t53jMyQhnv;8@HsZK
zl4i~Oy20pzdS=t5qg8Nzt4Yxho$chU`uGsBvvyrAHMQlIRQ?VcvsC@ta}fg|jo|!S
zcQELu45mLyu?grZTBGJ=DmdS2qu*Ts*t#^d8&o!Wd@iiytJArEQnq!C^i)(6O(J<b
zrzq$jn?spUf5FGW#dH;m+ErhwmzFLJpgJ~B_~;1^86m>F6H5pjtC3Mg89PTy`KkW)
zs9Mt(mz}#&O3<4CP!hFLD4L^PH=c#2mXdM_n2dn;(V!r+hN5tC*Agz_aJbRuN8DbC
zQ?eR&h79il+(+S+DaB{p{>9r0JEv4bFkT58fhjr^SCK!11aX(sg+ap6e+-a58Nb_{
z)2Y9e#0ESZNzY2#nx;x4n{)6j$cJ=bXoVi&<uT5yQy2)eVkiw+5;~ZuEh!a(64$vn
zesNgQKkCqniiigp4i-}Ab$O0WdayCC8aZlTT7l~O<2@zJ;sE{Umlw&ko=L-6P56;m
z<b15s<aoQMo_n(@RN#a=tn%ci<28j3`Yo2}!s=2Q0UGc+;!*tpgh7V|5q;>sfG8dj
zF%K?$Ch)|*4_qh^1%sf9Q^7pVXXO-_dpjBpItc)$ZFH(V!D^j`BQD;Se7;dEME-!^
zVAW<ry*K{!E&@|)uQjwql+1$>?$@x&5L=1f`B@aO<vY!gPWbc*9h>!G;8x<O+(Rx*
zq`26ga))3oy8qX}OEooW^3|jhn!n}usqT21c|cX?kLHdDiCZ!-G7>Ud>g))o=TeQ3
z4YGav{yosK7t@REQ35c!G_9{7B^4SSc*E#V(aE*@7`PZ9)y&_<Uz<}^9vaC#w_{M6
zvzYwB$IbMWB-F{N6vWt)zT!1ugKZHL6IH^4tG~TmDaRWtt_dST-Z|D<g5gQ_WX~cq
zK9bKx!UnTAx&}SnA0p=cLYdJy!^8I~(=ftu>_qgn^J&G2iUKv~PTt%r8+3%FXfI<O
zyq@PLr|K#y!SYMz-z5=AS?NbP{Zxjm@qkODm5I95VGL#mXBfC`9mL35E^;+h_*2%Y
zSV37FX|=L<pbo4+D_u&t9tYXFY@$Vi*NQyIVxwDE4&Z9SSkn}~W8Co2h|m7bIU)QY
z?L8oW!lehIAa1*}-b;x>uUYSdhM7A@;Zq+Cp>CsraHst}tn7jUG(_)8DBR&CRKrm}
z!~Yt#EkhY`fckc-#*-(>+SCFCvBQuu%HM5&pm+7f{|v38mHgakYk3+#cTgUlNd<Z4
zrN{{;x8Jn;a8wS_S)iNiFQ1$aI3py*KmZq*8nIsr92y>G-g=dnpGo80KttAices!n
zE7RC`(q$*sq~waPTtQLzFH|(~`O`C0A9WP1c(p9d%xkc4o?PMu;GFd{1x=YkIoNJT
zak{2ubsGf%RNk~i1Atqpt4=jy*S9SRCg$eX5gU{lRW*B%m$YZ23gVXk&nm0XF1ANO
zKFY?FIXNxPUb~z^zqLwypYYFNWkrh5$FgWQpxF$Rf`KoCVCMJvgoNudEDwP51I(Y6
zuCRY_xV?+X^=PpAT~LI!i|NOcrNMz2PUdHe?HxmnsrpV%f4rurA^YHNN}MNHrB23B
z7vQ0xnTteq2o1yj`GLs!p!Z$oH%`a$jn#z>g-j)%WX{_N9DeOTo}8khBD+7C$SK%}
z5zahK6Eo}Ok?HY1!Pnbq_n-s4M0rO?%bhsFh#ur#>BPHc2VI5X%=|xRCAQu3$QOKG
zf7d%4Y!J!5_8Y#F$E1;bGyzA}D+i~*hUxl#gAyM`sQdoLLsqA|?jaz7jR$mERyw}o
zbu{ROfQCYKJ1wFJxb6!uf^rrUQ0RZzJ%pW{!ZIt(j2!9R)IiY3z2bH{k?-N<Ub%Pb
zsdq(=_D)`jPsRS(3Ln#K)%kZ91$D;Ql?}3XBi3qer)$1@hT=85-UlaZU7gY&O4%pR
z%Tf{&Xx9i~R;B&SMklD2ffF4b9<FmEhFF}9SF;deFM~Do@%VQ2G0DhqKkSj<6alCl
zghwfyZWk4PuVeCOpnYvOl-2d(=We}pKYqybqyBx?Tgbk#o+{pG7*O762DW|S^m^l`
zAVaXW16agm1JQWxQ7^rPuZ{nHUqk_&yWQ6EAt$-(TyYt2_&H{4|MatnZ;wgOiCJXZ
zczsO%_~OVkH=_Ux6&Q+IieO^nuO!$;Av|otXECGD=Y_8BG`IqxZ+-1lNWwFjX=-FO
zUO8GG1t}m~Eq3WuJkmsQKbxBH5zm>LOSz}m;h{wvMt5)Db6Bn!oj-Muv{wePdcE(L
zB@0fixOjL>+c_TVze72zgzLx*_NK^aVh#GQKEHY4u7F1(^2|(s<k8&08Mn#s3j`FE
z?pq_%P)dRW0F3_Nd{kCcBjUEiLw-~}OFF1|c<=7<!ROY`xwJA4Lx2e>mJnemoaiNR
zm`%MV)4PJF1DH9MRYzzUfipcn+D1j11F%DbgLu7W=-^v(bI5_gsHW!m)7MKXIApB9
z-?yCtWPE~ND<dcK1jRMtl`;fp>|sceuGCzA`yA>{_C#H6bqUO?%(o7eHL4pv=)RDB
zih#xbV*)35PU>nsymO#yFdhk|Nvzo7_Lv(4_D&@_8jRNvk!4ls{BU#Y<>^Th&jR7S
zyR#sB2iHSj>N`sM!n8a-@9sc|;_T6@`q>Je<;S#iys?zbF|VfLZWnr90FLrRA*I=%
z#N6B*NbFdq<R@<V6i?)%E|w^TJGQ+fVCED0-E^@76Ri5FZnKHSJqQFOD4gl;07>d8
z7&$;k)G_2LE?&OU`fWy_4p7f7I$Ez;H1pTlQEw(qF4?(hdA8vKK*<K=eZ%h;#B^A!
zX&-L@!sEWT4%1|oPjX7u*ok7>(iL-h<v*ky49;uCESo@60Lb${dAqQFL~*1N6b?ax
zeLX3>%1z!lWIW7s4Mot7QKT6_;k&rB1jf<&yG?(UHbwy=2GJYJzdi-AeWsdm2bXZH
z5Wu=g=dHr*#XQZTv8Havr9o8Uip_}{ukf}a6b92TD2VTB+z&Pa#%n_Qinoy)^&SeG
zKg`FkceG|H<-9{Xgry5in}VLg=(3@qEWrvRV2CVs7zg(YmDJTe76=-QSiz$U4A09`
z=qRrvti{CuvNmTex+Hm3dz&&pu;lEjjokpM0Shw(&Zv+H+3CuwI3q`CDLEA1u2E}#
zpS5NKmSf8+xu)COlhrGzg>-NQDB7-Cj977S4iWqM^Ki_^>=YE~YeQKT#sk&2NMp-l
zvdKUocQ$HJq@he5YPO?vA@UCw`+<V=|0;6$ruj+~@Iy$Srtgt!O(}(UR{0eDxtb@k
z;I|v9-+C@BXfgmdyn*KTJx-R8(9BQ-g#yLveiV3rc;g-r5)#33C>qNf9Aq6_eYN`K
zFL-Ps`@zu~r=*0N#Jfa*&tY2$Ot5y9EpBX=)o**8Lr3^KE_Qa<p%M}k%b!W+MrT&O
zt<ix@BG6TqvR8+Qm@>$P!FcAD4E+01Ica%Dmao6$=Lk!$zOu>6CZGLPR02LwO*b)p
z#iFpXC9p?JCPvIh4uXCfemytM0De8SKzzjB+5S^RgtLQ_qN*a}Q(8t^S{6>usiR4v
z^AYbM!=gCdhddB@0ZxPC?IljvSO4-hE$FbY*oHR$31y$dYvrU`!)vAG$<B1uW`V)E
z820V;goM0JnCu}9<_*94US~V|8wgL*Ojmw&X{e*OH|*wC3KiSLqx4jJOjKg1z{(fz
zW15*AtuJA54F7`S!h=^^mxKY7Ym@Svs;a+W3<I2foFn(U2>K`p@o*0R0M4D8`;CpY
z<eP=X7LOLtK;cHuVxpqA)<(j_(~XUnV2;<Z^G?xLMHwZrUwlpwK%ZH;v=nUB<pnbx
z1K#l1x>rp^hWjh4w?T$2X{@PHcD??VJ$&Hx1Q_tTJwKfJJsT?NItv2yNQp}LSyOSK
z@>_ava7Iz~yBR;$%nYN9%rG3BP;5b!=ipW`q-d_WxGv{Z21=q1TT+Nxm`EThnEnjq
zdl>egEuIPZL`#cH5RxXB!V{6PJZJBwKNmqyt^88HcdTDXPd&i0zjIJWTc+fj$V%1O
zSyh$098OuZ3_(n6EaX~zyuAbk2>bvm5!AEb!=sC3Bu_<u^WtMvl&R(G-!nsC?z(vb
z;^pg6SB5*)%?Z+V^$M7eD2MySAbbfwC++FcjIQp6milI`%VwC<R@&p>w_k4F3?v`q
zm6UJ~crSE)ZHEkn<y&Au^|tr%<(`|TXVa8NRx&r6!!kaJRq7RmgEG{G?<Hepsd>l7
zpVHBIJLOJjXa~-<K6{88(i#~PLuz6pliFUSK~EcP^7>1__JxPL24pggy_oI&DJ|la
z9UAIL@UC&rw&E-=p4Cpr%kXY^TS-Zg=m*1xB$lY2!nZ<t=IlYwGQ@oTMua}(?oCo_
ziop4n^@%ELlkkAVgOM<ch+{oa<j$?jK9&j`%!*8VjNrAj@J)RV65Xph`<c9*9I?$<
zU4BNcH@&})4{Zi*G&POXM3M2kygYhEovGdtX;m>oBq2d-(a0Fy^qwHw$ZJb;#V_qR
zE~R(f8-#(ob9Ny8(D*7h*Id)`C~kn_*5*v`rX#Qf_=_|u=IF0p4W0d>jC{{0*4x;t
zrRGfT(?KncMu~yJQ>!d1YdQJzkLMd9RWY*@W{*2kO)X=-ttJOQ`2M7tdpKLCIXQwp
z^hf9edGbKObRZq%(Y2cn3kRn(W`v@4c2caG5;Dp%A|#3lb#(<msli&Gz*AkB`|uh+
z*J6X+KesaUfjH$IkaokIoJFII0Nx3!J<Yqj&*GZrx)b@WUTtF{{%2MA!4G4*jG5!^
zh^FO-Exjmv1fNJU1?iVbsnW9Ul}CQRGUay+H0O<08|}}|%lKh2`LhrcZ|~XEYXvTD
z&GBv0m<-DC)(*SLuYcNS-b-&ZEZ9QpD|s<0p9$QRWgY{FTMw{;-QC>-=jVYV<k3oq
zj5JF7@gp+wZCNIFu&IrOp`WXag*I`#XmLduX<F`Mg$M^{d;FUxs;Xr=j4tyhEFR@H
z&w1ox?ti82lqMlL%w5;w4B1+qFEzWQHKNN(OXKTq{Ll<q!BGR|xcDJz_?qXHh4X!r
zOG|6=)Dar=F>m==e~qvGp0RrFhg&L*>r#5&vR6}^@5CGS)1)Xm8vl4_#=zWCG}8$n
zX)c1S@^a_M_wSR!s0RDY&8(s*qewO_K{Wcu>0i>~>zcv^`Tx@buoK-6nVv847b1U3
z+s->N`B>nH>PzN$WAB**+T(w7_<juxI=VRA=Qssl2~E!0e>YC*-;FB?zxl`Ufvkz<
zDLSYu{2WXn_!1yLft2uT<WU_c%%{$@`K5WEpE5Ekq^`a;0~t~(_SRM&#9u&OB$!Q~
za^4SV{i>7$on#Z<D-o4Xkmcl5%AO@p8qOZ{qN2L`!^g-^dEaa{zqr8nom?-u&)rq{
z%_h6hhCDdr+f$nL*JiZs#|6pT1r#rc@7=?HfbIVQf_==Q-v9kgOZQD%OFL+OUPMLJ
zv6P+0z8F3?YtZ64-{+a})+PI%9$DE?Y#0d5RKm=~!QS6&Q1mjH)QD9|UBcs6a0-Tj
zff?aA^-uq1lC3!lNx44Jh?;8+T|-_7!P@gze0+QchKBWLqzJuMU3uO`b43kzpfI)p
z-n$1aSLASkU-EhF+KkSVJ=btu+*2m#-{GK(?LHTz_)OG7j>ZyrRUZ)@rLLl5u5S(Z
zOhs3j%VhwQMyg{)VaG?clNnxW{=9!cI9$fHZF4)B%@7mq-MH89u?;R8t~}j0kA-kT
zL%c`H1aQUcy~4L0h?aMMDWI9||9j>C{Sg2Cc)|5TTis`^f><ds^6pPJ5ENf)0J$JK
zDm#44aFjPfyFR<wsYNtwbP~|sGYUjJeiDHF#OKD}$tbt()d)yY|C3nJgs*KOBi8y7
zwH>sm$pR}1fRsz>5g^6(`-uq&S6fwA=N~Dfk|ds$mzVp@A8h*EjamK*6T&)j2-}Ay
zkM)Nox$u?qv#Osk&P~@(e=QV|&>jLWhL=9ezUe8MUu}ML4IkBCt2<HSmm*k1TWi2~
z8@z10nq>G(<_BBOOm9^|!K1=2yJU=70LsHciZrsn6%rkp5D}%mYwFMdw=I2&5Sw~`
z`~TkoP1)U}$cPABccPf&`237)NI)6J@WNF?hwF23Q&JMI@FkDa#&&bleB_byfAYP)
zxW`;y%og<;T{>UGVoyqXxIs+hw|^Up?XCk`RNB1KqVHKsV{bn3>rz}r!k!;R>6=#h
zMn*=v&vZ=G3SNjCni_KW^g6NziBjHi*%CaPjUE`&(k3EW<-h5<KUxwQo3%M!M?=N1
zAowUfo(H-Cg=4)DtG{%t37H|wQ5|>o@udxw4}6}BFR*v(|JrZO>ec-_8#@r{=o0af
zyw3G>{XpVirTcqDS@OY<y&@rwnEZCuDO=%e)$E@q%^KLgR>Y$YKRyc+_QZb4$*D6v
z|4MMX@UK2$SaEUW)t@fPNKrL;();i99a-vV(~H=-%8Hq*9Ry#*y=|L2vYb^>*5-Pe
zPWa!GDSSbOj`q%ugk{R9W!I<r@xKT9QM>ZfM6SHQrk!W<gmUmXOFI_sEz*d!X5xg&
z85o?@@^XQWg<hw4{hakeAoL;IQq(7rHt;OZqUwbR@<?D~V`5-qYMyruLL~HM6vxFR
zi|Mios$1vm<8%mNvLoTxlW<G_TWz;Hwg!K9aB@)}*)b(Xhmf-zwboB3*tA{-2w!`n
za3a`kdNLabS4>sy*CAc{_zx>4wBJ!h4o;+;4ik$Ww^A;`SQCV?L-RIke2hDJZ@G|1
zNkQUEjq}hIv0yxlQyNT(%ML}Gt8RA2{co-Q--@}y@XQK@1VP=yYHy>fUU&yG0XviZ
z_jX`Xx)z?CHu*QzTm*<G0|UzjeTO?y%4C8#TWk6>wC(44=b5E>^D#aMLVxA<rab2B
zPG#3HQ%f+>+BR!vvr}K=LeqnNh&#V3U|>o&0LO`o)xo1ENu+;#+-Y<8lx%>5)VddJ
z9j{8p>t}!Pdaa8Jgok4)sVXM(8y3gsU?lc<-@8k2#`m^u<r|KoZi&5=xSG7G)7D_?
z!L{GLwE?is{<$3?os4&IT2m<!d{o{{zwNw4DE^U43<-j~k62DcRR65s6O@VUfC{WJ
zbq;xDlzny3REdfCshrdN6^t&GQ&Lm9aO%PI(5AZ|;&s@-h}^`)02>Y}@-}WP)GX?A
z(nE|J&z<FGY6FB(=n^HRq_@ZR>CPA)-bQSlNClh^+o9zmBwP(6U7DGg@Sf`|t2#qN
z;PRJ%wR8D(K%<bj$u`3kf<I?J6{iTzTEztl_My`gg$E(6uN0E%l~i;v5k?Asx!ofr
zm5;E;|F~k#&d$J&Hn29o7`hnW$FKYv6XV;ERe8FjSOuK)yr0AHFMnINsZXWd@Xy`b
z)<_(?{^oZ9msxe2sr=jm;rip9iNV%a0yjO2^|z!^-K14yKlc+4W$YRLba!*^UxTY%
zeX-&P)BIu!XD*a%^7LwT#5C8sTW7VjCy|DxZlm6A&^wGbrlJJd>wNZ?OJEjHex8&6
z?f90fj+WY+L92X*EIHD`+%(yZRae4$2eivL2)0XU;xlVmNsSYzLKc30_^M<EuP&so
zLlxKpr#d%|3)B{x+-_jyUEN;0hM@Bj<6~b(lKI0WrMiUbBU;-;ZJ6%FgE0gkY5&cS
z*Pu=|T|Sg@=b1>-)$L1`X!%Ly$+Z_wxWr;o8ju&-9nhKUaqFFciW*A8+e$XMP;ACC
z-4tGngvuN{DYMf`8TD!pcf3mQ8AwV?Kklq?-XIMo=e0R<WV=i8X(b-ZJKt(+^>`V_
z!k>5Sdc%jF4NZDyVf{T=Z!k3Nz(lUAv%}+p9ly4GZS)u11H-x%-s-A46;(LyPfN0}
zglrFbQR;rDKgg~y3>aV7J1Gs_1a8LB+4)+-n`!N}B;8wNf|pJ`WbKfB=<5DMK7Lh>
zI%@T>_e+*O#*X_R4#!dsQ@IovmfZY`IVC_sthEJa$|KTfAEMtkE$Qu^O$Fqd<7eM0
zOL+XG`n-WfAZ2Hsz1t(yRKwNC`tP}}wilcg3Q2r*kWaFw+c}{s6Q#^*J>&J=tvO+^
z7?aoni-qs+{;x!B&k;w*{#&HHJqvpo8I-Y%2{y2jSN87Si6Y~34}N<UcI0?q>r{8e
zENK<EewcF?-r29Ker36yEG@VAC^j?OJ$}4WxKY1pA~;e`a!e*wE6u-Z0v^IL_F*@^
zms%!AaLKUI9DLQflo$VP6yRmTyE({Of<hMH*uNIIMX_?#+e}i%!GZi<D<MEx-R|h@
zwEUyV+uaz`M?h|mi4`s@EhVAA{ji{}pm#l5hO#3A#(#mx@{_>ebn9f7a`Iz!U1g6Q
z{r)c==fZ~}Bjra>O@kNSiXqJBYs=T=m@|KIH%u{+%MD@34bQFmnj}rkBkJ^fcM-Pm
z*|$HxF8lwahCR392H~k7DrX|_r<4~%Q{6YhoBy1f$iKXn(g)IG{3m#S-_XKLY{0L4
zis`hpCe1h2myC={*{>dza|ilhLFpRYNy{2<^d;*&z0074+l){XVy`vm)#T+zrdy?S
zbS}s&%yhj(9Av%~N2W0k4qcIdpS?8J@iY)u>(wZXFrgL?A?P);$(pG-N03Dtd-dLw
z-J|{P&Rz>^vrFUgc`x1y1PKzgcd%pS+Q4zLHq&2To@YlSnip)X8Rcm^$tVJFRF$0^
z?4G|_<mFjaA?o?%_JjH9<t&yQ&Y$LU=hZdySalWaBh3wt&gRA$G-Pqn=(0EEaQ`sy
zgV#Z7F^G&fLhkooL!dlP3>g}hS{TjK8JwCj&^NJvb<7E=hKNRqeFwB&kP`Y*{O2=X
zZXG@^l}K4}Ia>!in5&dF>!!yb2O<q?)_810<d?rrt!rmX`)^gp&u@}@<D&vzv<Q=i
zkN#@KyL)$G&sBveOgg=2{GrMuh*tA8D;K>Kr7U%tJ_Rkvla7a+H)*|XAlS_T7^~|Z
z$!+98TJAh_1$LLT9=sdd+oYyX+JKHpmuSP?e0g;6_Et2Y17(6eDNgHSe_vj2G8lLQ
zdU*0ENj?F+w~{w}?F=PI1WN-X@gWCe@CBLs+9rmh{Lrs~D*L|~1s~Lm4lO{2agg;v
zX>k$pm8Cw_eP(80gm=DdW$ituaA0b)L045>ohY52iWfbo6xlQZWj7iB;8w$Seo|7B
zh<|pTb{V!<eMwpQRJj?*R-<8n0)#PDRz3rdAWlE=7N9pFLk|>O4p$a~X5e;7vi_k8
zJ#ysrb`1v(U&*`*`m^YRC(wl;4UN6apLg_OPy~ctn3=6zR+q~s@{vN~2mZ9q=aIj~
z-M`$Py=$~3GO72N6rUT**C_|0q^|U$xn_K*FGb)jJF12yXSBGo%10LmJEF8)dV2S~
z(RThxt^uCa+loc5eA~lCmq+Dl$d~*~x1}a2I;;{f{uMC(1if`z=>Cboe2SpQ*-TQR
zCjHSttRf7GjO`^=|CojNf{YZt2I%NAGbu4{6k#mXz~BrV1kwkam(s`rSu3Yrc5WIl
zXR{{+MZ8H2iF$uSQ9~3xiBTL61<^uJl1^s<71dLwt5>FmWs6yE?Q2P}SwPVQus9lC
zf4lOcJn6Klfu)9)T9Nf1dqr>i&HnA^uKUJIi*5CC32YAQ5njBCZU3(Q)|xXslaIzB
zbOG_r3fO>6wYus7d^^v&-zRhY6#l$l%ukYz%PPve2?=E~oi-=Sj0aCxP+WdPt-9Qs
zkPsh5aK4-$7VQTQEk4IkWA8*yE5_Oz=<2E}pW!H~0yqJ;Nrs@$#@t$y+4N_V8(Vc%
z9IViYm>6hfwQA>AzJ4`<esUqed6bL&1JRm^p)oE}Cvn{OnUjY1NqtF3k^55QIGZ~M
zChB!qAX?St;~*Hz$jtbmQ}F&hWW{`a)3yM26#Gz+v6g3oti!+Y73oYxbJq|F5-7Sy
zx8jvg-~F<!zNfPzk=J2jh9nXFHEYCE@43T)aki}c;EGXI`O9uIvUV5UPJ=$BFKh61
z_ckf=%GrftAEY}dMnJxqkvPH>$U}x`aYF!UvnV55l1UOLa#Y~qvPvAa$i9W@3&JVV
ziE0ugxjvM&f3({3=zVeV;PLG%^19$prJ*$fjb}c%K~gyuUgcPYRDZA41bdZ?n0mZ=
zvCCEGv+_5+Oki^1>K&RFtiHAQ@l`#2yR~}VBpWdk=YsDOfQdS~uW)%k#k=`9UGLb!
zpB{clGeDaWy9g(z<+0B1JA(2E=R!-yBJq1HbWZO_8GLTTwmDE_<LyXXV!uDgRzfGB
zk}Bx1j|o5eRe0$=M^CC7F|FR!Q*h~votdgJoYhL%@@d7sA|4nRC>nth-^#;BbeT|-
zyXn#W%Qk}&$Q0JxD7ieU4e+rE`hyeJq#YxC5e0jmvZm_vPWr{M7<@N~_&yIWKnZWt
zt}tsG(M7)~nEreD^{O#FYhoftFgq{!R&<)YB3`mztq#1Fo9l46^_pIz-aet{_TaC{
zdObdeeY|*49{YKEdrp>#A>HKXm0M)CTO<5P|IjZxC(^%}BnODr@#^|ZtH9i|7iDll
zA3fw_q@%z=mdI$Y{T{U8LS8lfD0~=GN4)yLdGczmxvf#$|BUOR_mLARlY)zwn4e7j
zw2;dV41#emk*!_0`q;H_9cGPAK|^3xip#C3sZA`yzz`N0S!||E4WrpexZNO_cJvol
zGEe#Jv{MT#<t-H=1Y&&A1N>72e8l=q`yS)%DtxZgP#^smrA|&kaWHhDSw0ferzc)k
zUZ(fM-m2z!_3QHe*^*d4RaG*cmS1kJCeGVb^&9_&Yk7@a!qE^J9d(>J!ty{CehFNA
z{QNp%LzOzD2psN-<vSEol6+YE70g_y2-r}yiMzTtb?|>iOJ{a;DMP)%%UhkzPK0R0
z#cy)E>ZsTmHhzxhBegkxW0;hryQ4Iij5d|PYBwcTzwr=v<zx*jI{Fu?Un}$;Y*}(3
zgbxXld7-BCDe@D&tKK1adU|^c#zrx^w`XL|)mdy_w$?z#Q`ZZQUxLFiZPC#Yr(0)9
zp?RNMu~wiXN%f0|hM{*6!|>YgwAnv#qkE@2Az-DUP-$H!=xe5kWk}xPHyOTccmmn|
z1-;G0#EFBf#b_r+GkI&s_w$CLsx$x3Qx|(PFKKC|Gnej_n9BY%D1xWCtGCAoN~<Vz
zE-2tG4#%;Ei|FrLER@%}KHMHR^Z{ooAE}_Is&R{LKvP<zNLwEwq^(^Ij3;FO@kS>^
zC8k6r*1y`81tfVO4e)>Dx3097sFAFOYero%<Ak|vBHstfM|QibIl%v9^}W0?RmtNr
zQrf;c>JzhXr?ELTJ_I}AIc|zk$@or)wDn2S0qMfp)o+&YE>sB1qOZ58kPE}{srN*S
zkZ==Q0t*|zjs!>%f|RI{_X6Pv!ZE9`^S*HUu%ehbD(eL-1o3qlq<?zoF6pWHa%qQr
z^X9@Vbr{ha8j{NN7%hA2q$nK_!)DV)@A_FdT8V5|Xr;N>`%7T!J!f2kh?J+sOT}U^
zzp|Ce_`l%Sv(7T@c+E;drq9)zE2#HPsNsLkswR++kukT=_#Pj+y-$7fQdN!tkxlr`
zKqDpKcBIt3aq{GmKER6(rgFEuZgUIegGj!r+5<KdDDmsP57e}@f`^{SU*5R>#<y<)
zMabPm-f=UrC|!b@k{T?E5+1~`KuE*&l+}Iy{G5h495C}q(3^<U?QKO`+I8Fz99*1R
zUZl543G=_DFR^!f$Gu~imy~1}XzE_v-D86W^*!pNulj~oZ*C|&hm@b>e|Mlg=e<jy
zrFROihZeL@d}8wQYLM{!WK)bX{~hW1GQ(iv`5k}mF*7rBh(-M@u!D|tN(X;9;)D)@
z+YPnOl&LSC`&W0%Q=X%qsE;dpnL<3Z2ZyJ05XWz2k>&kd9wBE^fL30?SQRIrIsTb-
zXi$-`oYge6006iik?=HbZufR}RJ#n>q*aXmnLPeWzjtf?+i&|y4_s2RwPnk{5*G7U
zn4ceR^`4T76QEorYA~BtRg6kfdS2MfNlC`pzSWM4KhfR2tgtwg8!O)+9~={d&a}Qj
zlf`>;wa{BbQ`G?;vwY%FuYyOaDoW0)C-G2(X5%PcnhRWEWPCLy{lotx!NAI_&IUSy
z>@6oa4_rEOUguxgjTIf+nwrKId>EPDh|MYe@T=Y$6#~L)D^BuAe$#Ptj|*K{Aj}?f
z`(#?RSGO%^>cV^0Ow;e&8LJLDo3z?~_qO@^yXQZ{2>;k{U0&UXTKkwpV69|3*1IfC
zHWa7tlr-i}CGM4Qdv8Y{(^C~Sr5f)gdcVhQbGF~dc1DQ^AB@_}f^_?jt;64oh~GYw
z(JgwB11g@`m>3@1U}tEWSDJHcXcxMLy#GoDi#LHt<d0KXMETXq^g(9#w_J-8&$`^;
z;GCe;($LcCjllAwqwVcv|DRudzUcYq;#~6uAM)^h((ty0y=&k+zJ-8$g~hP*21s5a
z0G9%ymZJ3y^jVbBg*KL@s4{OKyp)j<`7HD?HjA&$g#_6(FhE{`o#Vn7-<&s~xzf8!
z?~`;wg7&?qpD<=#0)>!K&-`5}6oM5XH-V7P&Gs(R5+sT+bFdHzi62nyy6PYxj=GMR
zxVT85>jIDQt`WBN^f~cD-P6mM-Ma#M2;wJ3m3Ey7PJg+u)a!44AN|5YkO$bgcMv!l
z!jBwNes56;Q>NY2$##OH8xHE`hnp|``SNlLg4~ku9^MzQwSsLtrBA~D6$AnyzA9|}
zV~;RJ)zncuic+2W4PUl}|Jfh@H2{DNHEi;(Cq}+*A*}{Dhfd>`xK&g#a(&WrJNH%r
zYC#b)b3=s~4343-q;-*BEv>B)H<=jN68lz4PA^8voz2W>-p0ih&5`jOY0F0|4%kpD
zJg9K!LQ5&DU%ZF!Ul-WkLx-odd_RSQcq9)&F#VpB2L^?N+{!E2<%VJ0K;O9Xy=>T;
zMUaqYXkPBxRl?Pg>bv24_LCZcU%k{Mx|^og_E;gMHjdkBe>K8-&n9Y0SKQHw<z>B;
zX!ye|D<93eo0r~utUOOaO5HDo7FN8PnVDM0h(LMyMe2>Clx!GIh)m}XK2YC_h`-Ig
zX8<=4kAzszVIrdkK9zPEef^i!@sOJt^u7-!{5j~<nDnp_dwg!Y>%!7X_Q+Q9%A6uy
z%wNZyuX@@Z9`@$qFKjjHwaE~?n+OJ8jtbGB*!rdc-pw0G!z<sm1;rfp!{_P*C@h57
zXw3@;larGxy$J$_>!Ob%Za${JK48@vAgy?<2NBAj+$=F^p1AjKr)XT*yKy}=wTPRI
zcXVW8U(8m>Nlw>NhmW=S;WIMD@}1W<>4jMbBzuWy;>yZwtan6}lgX*6DJTF2Zdf_E
zw9!FBu<v7eO_|{NFkyzsuoc@sz%~&vbhwzL{8z&OcM}<<-eBZ>>ndH|Lf_6zM&?^r
zpgu4QFc6qJlar$YurHo)+fUJ_?s)H^Tq<+I0TUXW)dPhKKwm`T-+OOb2@9!WM=7q`
z8fY+FXU-c<65=hJ+wzjK))p4(YDy0Q5wlFY_N6#iD|^3z+6Vucv$K=2@rsm`l$K^3
z6(#lP$S(-LNK7QcB2@W%o(A}{l^hNzbX+>jZoD014k&!R1Cjf1Fv4?%&YP$(rep4|
zrl1fGbdZ9Af~?N<fUYuDE=U{zJAKXD54b$;I4s)8>h`uU0UGoSH?vNw7ugwOEsXW4
z85mFzSQ*g9>oATEHeu(}AaI?am~4E-3^*M}*&iCSHP5<NEw<>T0+=`6Lhi(R0uMGC
zTxCN;<Gs<2=H?zv5wF*O&z~F3(cMVt>+7L=N>4)(7{txXO-uKOR>E{;8J2`vjESQ=
z__Km3<2UEFWg|ZdDQMEGSx#xwhKZBoJ>0YuT_S&o`-Fz7wxI5f=wtX}K%JlRQ!%`2
zHMr&Q{abE<R9`31mX*~X@XpJ{%v%RjhW)d@W|+K}5KrN8G0sQ~d4J>fZEn`Mt#tz<
zla)I`qUgL9>>rBk_K+7^Y9C&fl(WW0e!AhiIDD8L#{~vjC27iF6yN-?%#XN$BVYt|
z0Kx{cMz}uYXuYpk*;pOgjUvOM#SE0po#gV1BHl#%jaU4b&CM<7Xzzs49Y8#)-8WhN
zERs36kBLcx&wZET-Pn)!%Q0&B*3TtVVBRZ4yu&%@)7cmc9$@HNtE%GO2sZ-!M{rp{
zHdRf+-`VA(r=moScN4-<{e)F+XBrzVeNtE1N4;((2lHvQhqGfw%AtI6k%BCF0w=J%
zg!YGL08vLz+d=oV<rOo0falW+_Y;#M6CN?j+m==^Fnnw8?rk+-v1X+BR7Xu486Ck`
zm~Clh)z#j4`)+fHbuPeF7n2!K?%oMvpBj)?Oyy#wCk+sZLVefhS@+1&(o$MhJ}M^i
z(AMSDDDK%qd053(y8J<ubL`#1ar)9)FcZSw@jFbOQ|M`rjE!<|A$hc@i%re8E%&&#
zL>V}mAL;yqwPH`<gY@>O6Q>1RM~cD`?wuGnI5GpBzye>>ZN=Zmji#5h0x3`M8gd+B
zuf1BEKgH+PL_ThI-3@8{u924=_l&w^(zu0L5i8<W^e<?+JOySaIxn;>jb4+GRJ!df
z%-_9+gbBpU7MGP<np@K_(y*~`ikgz$pv|XXx~i>9{*W6C+SjLy3|&1QOn5ixyXC-R
zgO*GY*bTZPF`D=CJKeoKqrV28KOW9qPZ&O=7f7)&(+|ew4WzvjrBtMmR}y8Y`Y*ib
zD=27LrW9RO|AD@^nn&d$v#+^BaOsWjMN?v=r6+|2cS@kTaVAE6KBG~9fvSu&3>EBL
zVL>we10(EQ93Tgc`Si)r#jfcUbKo@h#3cUx*!%aR!G1b6u7Sa3R5UKtVFo-31j+!7
z`QDR*g6+wX+;xCK)W0Od3q?mnp!q@(z8@@oNF*%!q}FCI2uRwDucpz_>Z|LodB-%d
zJnHf;M4YYREH4HnK}&`T;H2w(AIU6AmmhNx3=S$m&5t1AZQM8;wZfaLU+TSLpYLpf
z*2Z?*FcNv8p)xo)VPa`S&cJZ*-epxuiHwYLvsv6FV$8wG0f}u?StfX`t%@En9YHos
zaQhaE27S^D;pJJF5JFl&@$}QoIq@T`CIc3rN^jGhtn8^9+7R52MuhC7az6T{9x#`O
zqZqI}K!mR2B6yu0AN5y|63{&ik8nVVWk<Swa$=Uf5vUKfA_QJsh4zQ^t)Tcmc!`a)
zVBB~lA`v??cY9#G^0X-y@Ho66ti&Kn0;CrG`<egOE_Iumvo>Z{D9GZ?^86&}jmelt
z^xAwrb2XX~1BD_f5&%6Vvf6##Rji-(%hl4*A%Z%OxyUDIY}EHmE!(MS=dRUFAq3H|
ztbga%`(D*Ya{l0&Z_nK4%9BLq+uS7@2VH}J8$d8qU1)I9>edqfPvtSiZ$8uO?A<eF
zaqnP502*dmSAza!(T6MJy<j2o6Ajhd5|p)uPyEq+U!Erv4s=I-BrvSyOG@&mX;%in
zYe=B_F2~^{v8TW-(ZGix&1o}2OIQ|kX>3#of_<d089766PXUHYK>-V)rog?T^KKl(
z`uIS6Yg#w@t(4(muZL%E?ggHqo@J!JW8`sOUF;vltg{XDJHb^bxb#-iRR(+X((0>k
zxoLUd(*GQeJEVqUgtXqwZ?yX{Ln^-T*G(-DYiao&pV!eq4j#`W){8^8hySMqxIwm(
z^O{b&i;lYQ%xrTt0``}3&KPv7^3$BBvskX7Je0c~Y>WEnZ#cdvbtIj?!PtoeDD7Xt
z#$_@br$`zSQz({+g4_>S+B<De#tLcu($PwA^T$KR+{j&5G;+K{Df?4Tfx(~G5tw@8
zP$yx&K?H<F^PFwFS;#I5=oa(*ZBW*;eH#Qn!h{>p32tz0JzRdZvAOu#%(~olPUjX^
zY)VZ2?Hp~vQ_rBSD~8em?T@ybXVkY7c671*pSC`p6?u2!u?OxiZo}t1yU2GSE{3SX
zZ(-nsdy`!8Jbs|a(FBb>EI4R(0dgt$Gd>`k6%sVZ{g81RCLAYQ!6S{vNXJ|88+w$x
zZEhSK737b8Kx|gzKt_h~=8R8htAUiXG!RH%fBMJvCJRX<!3O;&(S<dB`q{$|Z3{Fs
zw4V?7({fnwku-yuskNbB$A`B3KTL_wriX`%4R`cGApxqZ-N2Wv=Y|2!>91+t2tOhy
z&=CnTtQ8j*ZKZwMF6iOt^1E|q!iW743)H-Ub9-trKHp2d@E&}MiMZTIxOYs3_UK&%
zH``Zo!k)6kzfFnJVH@glDXiC!mR5`vMqMj&>q^`8?EEk)1QaFK<CQ70)PAVb5CyQp
zcu9)?P?|al28{fpihggEc^$Z8qroYwptoybWi?r8+#Da`{q!eQ5G9L>)y>#0-3mJi
z>%SoGxsE_Z<3f-na_)T5Y9R~pU+^EsV?`A(OtErVPbTWUc&ZR#$Q-}Iq&r>i#y_Z3
z={zLJ%jMIN=5~t=l|n8iaCHnT>jNlU5+<quQPZBBFgDI0l%)=D)2p@bn2J}v_NAZ>
zhICso<XgG&fvF>zCt$kH?9z2-0I{9#&iwjyS+$Gc%6o`xZ_AEQ6Le3WGD{?u`H=7e
zbP))U$1OQSdLloPZ;yFAK+=*0e=h9(2@B&rB2^<!J6oM`LY{g3m4@RCxe?$`ar!50
zwl@`JrbwCKqy1A%)oaMH&1u9xipI%v9=wm>0ZYJw-7~f?8TN8ZD=Ufkf4gnw9Qu(Z
z@gE7fmA=~#iM)^AlHmqW;N8^4{s{}f!^IlScmGy2yk&Vg`;!PtI7|k*fa4JLmUZ6d
z-9OUB1kKCBp4xq%rjj5&G3$v%4CK}8FBNu^$)8GPKY5>BL<8FKQSuHSX-ca>zIK@<
z(2x+geY1Q^ATI6`tuW=h?GHw@=AWJO-9hi5Ku6sPao!RP7Ld5KG&OYkJA51YZTk$~
zUP@Z3*`Vn98zFK$VBWT8%TPmF=k0!6N_cA0j}%x5JdxKE-H3q9QFUW>KrM3CG=bt;
zBOC{VJC)b=Z9?qjiLgl-=`H}Rd3aVu?e-k9KE!#+wOrMGcbx{`6hA%ved0SuN{5Pu
zhC{h}6`U2NkI$ea0j^Cp`FEImPC?yVtZ_)eY+bl{^5=vPKfle3ME8wM1LJ6KU#vn4
zsavP(naZnQNH2N3ujZnzzd=G;z0=SVuj<-OyR1!IxE?=&hR(w=WmWi6;hN?x&^1w}
z69OA?mq5%C*L2*RtaP&8dt>GuuR63xr=E2?<1zr%WKvQUX>9t@jhe_NtwJmPrYkou
zPy8L@Loe8JCPx`ZxY*(8aNYP>mfeOoI9P)F@DIPkJj@Vkf=G~A${LZ+m>t(~w<h0Q
z?Q=NG$ZUM54FJ&%i^GJGzJZ$jOs|*32a*YXQPri**4gCud{3gJ3|#rPFuexf7fo^&
zpa4s-;&Eip8(rnF;Xkx;*)Oi56v?L|7u~i%!spDxa}-uu`bmiVna}BQ%5slZrR6|_
zoC$OM2G{meS{f0HySM?9l}^oL0~G%)1sM36pT=i(aJFX%b@U<$kN3}s!I~eXp<Qzy
z_8nzh6Q{<1IrScYug(9ncHj+M_QKE5;NT#UMX)=kr7ZVd_l4?dgLCVPk{RgE7#Q+l
zqp5aW{*he3iF~egyk}+H+>{QkSmbvFR^2iy^*?ax&Ec184V_J!`;~WBxuV8vVauYS
z>42dAX|LUzRMs=!P1ZB)5WIUNF1uqirDP&9lxXkVh}k~=bky~FXL~5qMcZ$~8jpYP
z=m9c6OT7h%XFFwP1|?Gpf9kh%R4}saY@_le3z~|XyHi))E-hy#(-41mhKFd--=}SF
zH5=P&_IS@!T2{vNvZx@Z5*67N#CEqm=XxB`ZsGt-=<}yYMmjn;GJ5su%1xFceW>R9
zXA3ny=|NaJ2on+C<~b<#Ag=qo$v|`-h;uN@YDx_>0DK`@;h3@@VjLZi$c7tG?ZqV|
zrPv6UvnFjlk#;Zk&87s%9IiD$)v51p?R^btmy~%zok>ld39>d878dHL(@Zw8)&N?x
zcR62#Xc!q3WTo@-a$(f4sJQ;B=Sd}YQ%gG@0@!(X2VMW*3-ky!!SFs5Uxm-KsmxVy
z`2nj3hM|Wc-{?+I$HVsOmDt_Pbj8N=$B6Cs5-83Mnd87;cmsV#_@(_v_TX~~O(e0I
zhMw`>6FgXf@TXAs8XILq(CLUiP<WoavQ$4hIOr=N><`_=jhzAGB{KQs%4arvov@LB
znmBd!Sm9T>&8Gh~U#wv16Ap}~^R)T2NTt}@&HlU-P}H$kl;ZBPrkR43b9<<UZ(8Kt
z1;M{~l2cT%rrX1#zaju}rlm1<c&0~7=9U4ZpQxexVMD?1241a%b(BxmriM#=O?&z;
z+F^+h&}*V1uScV!u1E*a(q!*Qy=AhW0C+T!-w<<Ohm9GSWFq23i0^|FAGSou7|P7d
zc+kE5#zV1z070tQF2r};_z-gM<wu)R?GCI#rRJxAJ#f%SOH1kfaKt$a;H*RX$HuU|
zM2&}Cq;4XT6_R3_G#vbFEHq{LNwkk1HNIj+f9(fM7wBdevQn#$$dYPbox&_sIt0Nm
zuxW#a5F3rnckb33p$8m;VUM4zxNIE(r;^pD2be-2<ww1Hhs1XG;9|U)!Nhp~HHAzp
z_`$%9RF$SNNisxfh;61$Mp%vb`TfSL^teCYgh<55@m_4SQ{H(@Y)sXPF_Ug#zTwsP
z^5u6URlpJejcT7L)e9{4LWCH-_c7~+EHLK7pn0!o{|owu+;t);RtPfYfKUJByN0-y
z*08HSo(}=%dETk<_+#V&TPFqIGjhIXNKzO$MNU1v4VBrTLEGEEr=*^flMfjPH~v`y
zUHaER7VS=Q+a5xfNW)#AzTIZxbrI?x7({%b&Gi$@BpxX^<gfBsIGf3iI3^Q-S!$D<
zg+8}0grI2`*0-~VbXEvas11;3j-S@%D)6}b$8xWU8=g{vOLr=I!Jb4y(ud!ze-$fW
zHZ9m8FLa?ECcwg|VKZyeKz8e|9w1Ql_%s+IWAzp?Iz~66C$8x!TR>h|aOe9LqsaH7
zU#`l|D}3lK)6dPV803(XtuqpkAvE$&0{y-6YvtMINh<$b<28zDUD4S#gQ8Vx3LZw%
z%$ndJ6og6PmYv-mfF6*(0<%W+g^!<;orgQRxcz07cBPi^ay@&SQkL0TBvP&La3(~j
zB=f83@^byLqf3zM{RgBEnT^x9^54Bw)jg+py}Z12ev0ZBMIGvOG$y#($N=7A<&^^R
zDB04>kig(t$D<P)0#D#~Ow<^wj<7I1#wSMVpmt<>i}Ds_XF@~M$YhDU-1%bI4?1$d
z)oVW7{JS+zBaq^AafnKrk8)=H_hs9Yv)Lb>u2=HN&G0~Li!16rnjAt+z=fDjrxC=4
zNnufT*Wo7jU>Tk1$K{JE)vkV>{-G^Z(!F{+LK5QglRV*Ye_T}*1h&)NK&IF({uPfC
zC(-ueRxEg8D4z6T27NCo3Tib$haF;``>JlCKel1Xen$|yDI`dj$K^t*f#I<tAM2Tu
zoK@0yh~=JzVL1LbX6#%@pt8A6iXtSXc{=Gr18P9HCmvjdpFYL-w0$4s;rU|IT3C=1
z;+8~CM{&726GpcJ!&&mHuf5(;)KH0x`($ovc<(mR1JXOpKE2qa;p#eCxQKl7!;j19
zl<}eNQm*58`1^b`R6Q;2yT<L!AY!%1{?=OaL1IuJD09&Q5Hs32*zwr8?mz&ND1V+8
zhLVyJBoF`}B1V;I^3(r$&&cYi4G~cSh$@@jZZqf=f3MD3(N6m+fQ`AfrY{%2N^<w_
z@Xv|o!-;=`*a#A}wN|?0?ws$h{h>33X)>F}lwo2w$nM74u1vH-NsS@|ItRyRo@bk=
zCDk5(H`n@7rd*szXZptu=Z)F_J?+)UD5%#P3oX=Y-2ZC4JsHR$OX-_+`27HWedDK7
zA;t5f&9N4f5&+{*Nw^$l1OK_km7k#U|1kSf*tNNhi-CpiJ8Clnpe;LVKNe0E0ny^l
z=HdhVhn(LF5rmnI19vUTF@2!k70y6%zfV<FXo#e=l+QUMDuO2kcg0U2+%;-fUQb~~
zZ1)|30@><oi=Xv!>s&F}c&ncjwMV}ld|;<>XB!@rc8$HDT&x6>|5(&{-`j0KzixVw
z?lD$!VWQ`D!;$aI858J}#y=MlxVk#VH4?K7izJ!<sG_XA=KOa=w#S;c-PTthMXC7e
z5aq*=C*A~}!sO4qwlS631eQJORNIFbPhiTxLKfo-Gm_W&1B1h3mBuKE&2ZIENhfb1
zn&WQ$6?XszknrQ@sR0x1l*h>fV^=u%X`{<+R$`u9tGHl1E7q_~O--Bii)vS0%3TUk
z>?1z5n*8bE;fIb+j$WK5G?nu$L{WWq_i%1Y3~O14%;nC_8!luw08K3W7Sme49`<MN
z7nPEN`V+oquiBO&IB;a*jw;3KU%7uQ|Bm2%*OGnT=VvE8kBtEJ<KeBwO?gJa^BCk<
z>{_tDgA9B6>OjE(ozWtrdSfS6;a|ujAMNYD4{~WtqGI(kKfHgvn^cE^w}}bb1}~o=
z4e3uFNztr911Lgv4vUF}b+~^2JfEyo6gRX!SK@#hDAjm?(DT-!mT2%{mJ{q(KvR<d
zeyOhRCYA|oPg$4gcMVH1vCN`~YeeTvb5K+yhA6ju8)HPHprA92dq<+wlE7vdoRkmU
zgxxbjvmE6%T%^Hf)P|0>T_J___8FQzd_Ma7=f*~kic&d!RjP=ue0|9M?&RNFs=IZs
zk+wH)6q3O?V+Y1Lz+oYExZkA|m|(#62caFh@iy+6;y;b<_uYTOY^EeCaPIe1-x|r+
zaz1KloVqv$0MM0hMk~c?*I19wk17-Tco6YgTV$o(9c{6D2^qe?G0U2`NSRf>soOqP
zmVixWJz>J4V!N!ooZsun0N!}TGh9VYHGJO|v{_Qpa%5i&p%mtZu5gfrg|%OT<j2k~
z#$5dFUBE2sq+C~)=N*?@9)T70l_?2vtZ!&<UB-wb<g-Re9Fa~`Ea=+jhDB$-y>8r=
z-24xJH8-&Q;jn=Q+V<&boPUYyg(okdH5w_<`f0?xaA)R!;BXU@71lUeNE|c{fkI(0
zS+~NSFli?GQ{)ll;%V#$8y7M@(xtUJgN`Ptg*J9h4y>V-o@B=1)vxUxscL$4?sqYw
zQ{MAQ5LmA;`An;iFkJcG4S-05;z#^!ZPQLSV%x;LVd4YnEr-&{+MXvg9Mg@xFnndU
ziWouSm>ey2y$s>D0>hfj?{<?At42m1YQBhw7;(=VIR^46tadH7ZJp~0GTx*4`P~Fe
zacu4xlL4LP6e3=$L-<~2n|>9e?I}J9yrQaNs;c9_SgilyZF>C-;*3an-4_cJUA)T7
zU^LfJEeMyA_=>ss**t!v*WQR&EwQ@QAoM(;bKRUanzOrcdn|Pt7<D$f70?elbE>Vf
z_Po%$g_PT9%}zJF-#e$?SEWacj8<UmMG9YWVMeK}w6y)uk=N6203OTrFeH-}z-Ie=
zGM;J9Y&GQ6;VbBOq{KAMvL`-*%#0ol3jwI#i^StcU7g*hENyVC%~yZA_7mcmj9Mf8
z(j}^*qkV*0?q3~|{G-p~o>_Olbt?g0>J$Dw*tKn@afGx3yc?J7SNf6*@-uIf5ch3v
zld`~+sc&>_kQp_TnX&fTf^@37HnR*zN>Oab=RMzz$t<cv7Sxbyx3^M@_oW%bxkkE_
zp%hXUsVyq%7<=K<DZ2CVd#s)k;bgT#N$Ds2JH^k&v{HDSefw5yrk#Fwln~6kOqRLd
zqfDH}V-y%17TpR9>`7)uNXXrhi{ei`CiUJ?l8>dvCbd@X;d|4dks%qzMk|Z=Tz_zr
zjn~7#*v@t#K}LbS>UVT|Cl{Y<u7m0^#Pma1g)(tsg4h<_dFq(<G49T!9q6+@m12Wf
z*Ni5ohxf=!Z{Xbn#OMx25(Hwi5Z-=664^TZiwj1k0aMInb`b~LT3MO0Q)wAdVuDN;
zo$a%!O~NRa2i?r*ydY8pygBc4&6~n{TUR~=!0xTY5sxr?0x!q8Lq%8&6C-L$S_J}*
zl5ma#g@pj0n``r$GD_-hjRdp^P<;GDj-#)FfG3mubRxUWRP$7;!u<cy^p;UoZr|JR
zjVK%t5Jf;h8YHEpTT+lt>6Q)!B%~CP5|9?@E&-8{4w3GbmhSF;r@#OEoZ%S959jQ?
z?^tWidCe>0KHPjYa8jJwOWk+(em^*cearR-GIyeo$!l3{G_vF9_{SMCTZZN#702{<
zG>uaD`{xu$Md<0y(c@WVWH=7u^n-p)SP$7=`xnjBOqF`!<M@mF4+qZIAQUW=cacva
z2mOaW;qUZqbj6HcGgj&Un%zNCC|F%CcF2@9eAQZElD0MNZEE!Lhu7gAIiWl?%Pc7Z
zsvoL-6)@^v`&tE@6lLnEz9W0sSlZB9f44!Z;DbX!{$Mo!I+zJ&J^E;_NG4`%;ciZ+
z!u3_;Ek8jp7tFApaP_d#%gOgXeA0n>xcQ#%3FtGphWtBJ7l5)vO6ru#K7^cHN^J)!
zi|g6$No=!ju6$@8P*E{<&B$k>+?Sp$gEuD+L*_lr89f&@0M8T#mI#~m*~0qIZEylq
zfgz-DSnsv-A*PJX-U7xnbXMNO=#R<5?zlUYr29@QWDXDWKz@OD&Nn+7D-H6zk6B?f
zaWyu(H()sefdGs`i9>LZSN8W%8v3UR#_~Q9uRbDPvNz=9<kqWr9nxq>r8CZ+E}cep
z`woVMrLDDn_6-ClcoUjRtgKI;IMYgxC7Xc{{=Jk<r^X?4XD&n(S6qto0rOHMj@&aj
zG@sna)>I<)k<R^N5|<g;Y|2bjL7Y3t_}4Y`H};orZs~K!+uB;jXezzXQ>}KLZFQW2
zi<oZ>wX?H)oPR<=4fDOU_!ckdaXyFl0Q(&XjX>R=so6(_#JGW+VBh2P5Xaoc)X>uI
zYhrwIL=3PZEF^aJ%@s!nhQN3Ms7EkhSxp%PjTw*qi(0$kfu)#{h=JF%l9EPTR#xkv
z+6F!joJkT?uRv!0>w83|89ww6fSd(Pxw%ytoNb~-%Zita&HQ=njS2kUwuyBAd`B5e
zzO-H*jBTXhPa87!k%VLo#u9rheAstSV_`V#haxH=ZJ;3~rJzs_6T$AK4I$Tr&iPs_
zjGVmO=!gil&)F2ez6dtzXv<<EEn%dd{oC#;uZi>2>nUWO<<K+1LvwJe@$^?@d(gG>
z*C+3*HkGnRK;GWtYxJC+(^J0{X<d)G>IJn@?o_SZ918D^zjn%%P4ORqe!$t++5Q+o
z@JxK+mCaaiQojAQ_uXWJBFI=paZA{qRC-*|-~{70&95dVwlM>li|7UO$Yj^=p+Db&
zhyGaTw(aRBtOHPiAq`3fgpY~ZSjg^PgYXIhsbN+CBD<fZ<+ZHLBWmQ@D%zn<#tiqh
zBtopj)q<XPXitoTvIg{8uIz#cZs2dlr9|F$6ri3Y9y`P#fX9C0o#9~zEf8!}AGyzf
zgIHj5zV#Y#gMfQ)8QC)HP*M3LTT-IgOI^yhbrz(!;RM-V9_W`61F*+T@LW-I0$$Ow
zt#SElcz|kd9y8F*w~Nm_82<z>owB>bi*It?KR;0m>Gr)|U9ccLqly^sNAc{8sY_CL
zqaaN6383zw11uVfE4p2P4Woy-0K0w`jhY(V_~LRsCUCwL_!*j`L(?!SSYiw7ohiF+
z10D~kCgGGG80_cf7}{Pl#`=6Zk)eix5eRHF1b2tNBCm>@nu!F_p2!{MDnop(q*qpr
zoSoT)6}MAde*8)39n?7!j6S*z=b-m4D;#8`PG|XFDk|E{pFMyn_2`KWInT=|P&_E|
zSTEO6=Qtnq6gH&b8eM^mT1ZIfndsHds~tF#74Hxd5ydUfhF_tCh4<+}>JFM;pMVH9
zmBZl~DcI$u7evQiU-d_tKgm)U8XdfwR5M~fT?rhuTR{(lyYK4vCDxwkyirIh8Dx{;
zIPJOge-|cRw{U$UwK$h(b^HD?g9oI%wz3wT;6Avo@?wA_d{$C|Q@a!~1avc9x>8_#
z3~6Y85}V{NePlL#|H1wD_iZJ@qaw;Zx?2F=YgeUTjng{5@m|S3q1v@9KRCCrFeWY<
z?;)|U%LwOVZ0C2PO8UyDomJg<HMJne2YT)HSm$n_4CEGOXD{y;i*Loi$O{vnC(n8f
zSVen#w^2X`@Nza{m_;Epr28?nzl!A3QYND#BQOK9c~iuIJ$GfhxhY0@hPcKil)v6w
z0{VEn>Ozv>+biY`T5YA<!Ll(|KaA#hS=I3uKm5)B`xz&ufULmpu~Y`l9i6Iul7s9s
z`#>Fl$scS1tNWUtm;MWD#nVuKpiPn}UO077lv_|zT-v9k37mkd%ND&-b6YDb{qhcT
z5b{-xYXcyeKVsjGA6RQ_ZL6iG!Y|rjq6{oDaPxl~viCVUJTO=h`k1@gpHv4qU`P*B
zX9N^~sJ>!-Uv41&O?p*UO_N%xkl79-xz9CmjiyIW$OYVw6Mt@@DdmzrYC!`8*;6@i
z=N4IV@|P$M+r1CpQ7-qW`YcG-xwjIyJn#8;PJys8C%jOs<T{<M7t&z(u}fk<Q(Sq?
zZvl$5;>B<v`Jr#9JTHrzjsE7U&t&EC_qtS|QOkPfXj^JELzmn(tZc1wzQ;vE=l&MF
zzW})`ny5g<(5<^EoLv91sT?d)EUd89)s@|e_XOckqzZ$mREJ7ks04U00;8k3rcDHc
zpLbRY+8HW6Gg5vfjV{DSpHW_i64rDJUeHxNpyjPN@gssbIGx?zy@qUX`NAGKUIuF5
zRsR%V1$Xn&j|J7kA9EClMbHX@VRizmGHstY74}W3wVXl}N_C1;dsWr_0d*4GF;1SJ
z|FvgTW!W`<l?OOPU0nmZF>}^Sg5=MfJ_d9wvDI$p!rSagV9>IuHXe>=ZqVF<ksK2h
zgO28R!4TdZlXy<p6_=Yq-@D|{?=L7U!bUIQY6mbSX#AeKk3#XFS7}!Q<4Y}~I52Ct
zMi2AW_3^o$U(rMhV38yyj*n2~0A)GQBO)KUqWTnu5>$!CP~7uVL9~cg8GVk5uyG@v
z%gV@x9ypvXC>tuZ*^h32AG3k+(=8r}GO@F;h=>Ramd`Xaz?rBYr#kon7zE4X-x$Vy
z*)u^JRz#KzG!#n88aEKEB_<AHQ**k5%(K8jmeOLf)uoF-pO*-hx@uV|TW5QBZ#&cH
zn25Z5`@YfPr>{fGxfziU(h|1|N|6*<3yr{Lqydggk>Xq5%Q=&I-nuI}JlOO381*NX
zRqKRWmz#<!#xb@q=n++u;?TZWD_+23>h4JpIO0b~fzmdy@-I5FChpIy$LU>~Ef78a
zPQ!7vE)gR_fcmwVxnUF~n(y82o%}xs>sA02f~sa;v#{XGdA-=Ty1(U;TDVhcXEBBF
za$1iPm3}rUNbV7(nHOWoN1QwNB&Z@}aoq!Vn7EdAU4ck~e#;-@<_!iJnGE_MSt)A3
z#(fpI=;tUou%+HAEDE1pNO=o+rprjme&ZMzn6H&Cdhr9ihz@$v>z4a)WVN+ro#J4+
zrCaF~?%j|saEv<ZMIj?*Jo!g?UE{HM+1UWLfq}G)xA5;i>0xNmu2l&}bXj`)d$_3>
zVUVRAbRxjY!XV&qpr-DOO%^7tdYHd;-`)L%1?xMkZxRBZrPf8ew}0;C3}bq3S@zLW
zlP*BL1RQD$QYR{9U8l)@!*JSm3lr)NxxTK5H4_&Q5^2i@$!1<wlqt%{JbU80G9gx4
zuKM)%+?}TuV?s#XQ17@lF}8d+P(ncSB8LS9i<YWRc7B0hqoMZ16`QZpj8|DLpoQ2W
zA$)wNaxZxTX<(DpP*u$?$P8>Y29DO#-)qZ}9n`Y%f<_y3V(7n}|0${85xdu)X@B`1
z6QhAkiRCyr9Z&^bQw~$c6@95yx=MeLI86|BN8{{nvW|0LF!izD153-?m=W%mE_aCU
zk|kfaXWGCWm<#b9F$(y88sPMHRignNaT=x9ol*8H_MSd`hiehww~;7E1qdub@UTKA
zW1Ff{5ByMX#-oddO?8cp!?2H!I8^4OKD?Sg^6clS_shF}bAf$jfhpOe`Clnc)wOSH
za=osqPS_1p6j7K%qzazsa1atVpo#(q*sJMqcw{(OHdum^li^wl+@-s|@E2!gu?SKT
zYX$swiwYD9bdh)n2$=h+AnwRm!ea%}-(aTgq4GhLV&$iXhp+*C`DOU8|HlQ0Mo0RY
zc#)NjjlJ^?*|^?iZ<V=*lZ-vFg{68zp+7Ts14~<%9Q12E4l6Z`oaLut&&qQuN>@sd
z1b8qHtk4mOVOf~4dnT!ysdvZj5gvZm(N%dQ82dGjIE-9?pY__aYu$B$z3yAh4FqHp
z-V8l=34$fFJ{qx7)hnwWcq;G0yrOqdMD1uWNHMG3GhD;5axx(2Pe9I#{^RNgpHS$J
z_K)v=)zKE2W%v)48WdLEKc1640Io<HSuC!Ez&{W%k{Bp^3FjsibzX74xDa8_dm!ns
z60A4?+htYtCquS4<^ESQytwCK2WYsJo7dsSz>lh@4xQ_!42Zx!ki<Q&2^bkE7s3A)
z3n<0*xscE|Yo1F#?->2n;d}RvGIpeuWr&bT?*|{QBq%2&3;^aMWRy|PExiz~)12x1
z!@79z{K0<)RPk$HwT&ug{;bha_MV4KM#Uuxd^cQoWP2c|EQJE`-ioD`pf$B5A~y?#
zbNbZ4@XGguq<{O3w7R60u5QhbyhpS`j={ZOsp)#sZfR@k+_v3zf5H4t)Tc*H=$n2k
zJu2ac`*RtYv?62CBSRz5cgXANPF}a*gFFz%1Q<N`p0)j2TYYM_>skXZElA%Sns^Y2
zH__zYsh;M9zJVN2;{)CKSJ8*iQsGtB((TyD$V=5JVXv+np#KA<$f=#rD1lH`+MDIt
zt+8h8iLm|cHF3T%#ZKTgL8i18ul-9YkP$vLg7oMq^e_mx4Nv;z^0&EDlhzC$I@k5}
zRf;C_+43YgD$r-e_9lvQC)MKP^-JDMDwPbIH04+YwHC-$U`Pg_BM4(~NCj-IjA?4-
z&5B<<llDF2<2#I<)ikon33Iw)%D1+gtRsi0%cy<|MWePCIzzmkDT&_RVM?^&CKL#e
z!`qs_D3#sZr?XCqins!XQJ+iu9R9j+zu@JUkY`>2!f!tuv|r=tH-QRSk!(7+>0hzz
zCw{@p3!IC8-WNMpi9k>`@5M+<zTos8`D6*6wk4RD!3I(-H*fjp=F#7-9X`;(BwkEP
z(sz7jO3nPxeTdKd4Vn>~p}jRwj=w%2iICvD)uVe+07V$!mP@@!N#W7eFy}9NNqfJ8
zVjR!*R_=CCPPRWSL!f`yMktL+q@r{slyYtDC2}v1SLe}z6$!}H?Xs$X+O7{FFFw<s
zQvSG$joEm{J#sd6lH<$FuP$>B`Tl13$7-ZV=(z!JhHq$$hN+XyuWD0Ja>FP~f)PZ>
zHxQZ}u>-`@4yD~Ru*fVYJUkkF@@#A^rNSe@B0eX#K)eT6=_3r1#BQF-EWTyfd~y}I
zVh%_+z$?EbHby69WuM^^VZFSQADgG6J+Z@^5EYg2nHvngX2u%{MzDub+b!5U*D^0}
z#N7$T``zacrT}f<ThpxO$-6b`Mzfi+j=lS7BM<~++ALxIt+TLb>z-H}&7@HhaLAB_
zb<Y0x7FF`Y=g9Ms^$>e&_3PWXr41mz^RxRjOY-qV`meNP)mc$`S;3U3qRZ6U;?F{T
z-dC=yiPLV<3nx4Q*Jh>V<#F-E+v7`)Q`?i}-Rl6~z@AR@*OSCLcr=_cc&QgikP_4A
zS0cV;nQrRNk2wO*$Xh{(h<aJ1O#hm$SC`m{fdkATicqFtR8E>khsO*~#krwamC{vw
z@x&xjH9^fr#lWBE;`#{tc7-lWl`%-c;phY5a+^uNXoF#jNM`iMmX{!Ue+K{EKSv7f
zj#&q!;dh(T`)3)>ec@X2Dzaw>l_J+aygC~eu^ig7veRLP2x{}Wd(*FY6Z#16&J0Ml
zahYgKar-yTzOEG0va&Bh%GP@~k0z(20)m5~LB4m*EhC4;%6TOw+&^6-4<oso?_Yq!
z8Te-lPpcUrXZteANpYS>bHgoJFYoXyAA+IiwHqbE86?r=ak;b^I7oOF{qrce;=}Kh
zzkNQfje=yXl<XL1YWY{>Z_SDNq(%;R@#rwN;2#{nm@2!esoJTt@(K9&)Gcl86s1z$
zC<6Metq9S3_abV@Sr&{eZ`)M&cI+=5GqA<Q3Apr3b_pFPT%#awb*zy9&*Wz`_W~ru
zUPvq?o`)&Rtsp|zd(MBr$TcIAMjHKOd%V2AyE(jbKGne`syOtQ9*lN4gw0S9*>5M|
z@9$0N+PS#6SX=8Ds}Lh85?J?h;?HKOlY8%A0G;ujg8-7En>|uqWa^V<^FC%|reKGf
zeH5Z%@1xu$jRg`jxY7aWUtcuKRp$H;s-mNQtf#N+mw_mq)}s^<kC{t_d1htY*>yrh
zX>dU$oSB(<CPk2P?=4vTsA1lsML|I@7kkbU-U27*{d+37n_$SFy7q46?#JoI58X=#
zw{w1$l;1=S{zB+&&X?o&I#{*6J>;J60n>|$GW{KIo*hQoa73hTWwnMhh$^CowMsYt
zMnU`sg?3T{B;J3lB1ULvvQb25ew#DMU>uxKpWE14OPlmJ9LZ$6P=w2ZmJGaf&OQ%z
zqafJM>Vjl5!^3_fU2}NXjw~kB;y(hOsXE?Px!=*vl#_3ddzUiq2+aZ!-HHGC_J~}n
z^1o}^K2$P15=Dx?ffB!3yEY(ydD*L&|D5&)-Ws85Zb7C(w#)BkIzEAM0f9fwv@}n&
zm}!WDtJo>|@SR_p9&BUH>sVM?eIMIsTj3?7drOBt0Wmt|Z;1I*Y7@DT-!+P&^X4PV
z3GlJz$7nUjFUASy&3khI_-Emx-E4wk1iE~ebeOk<DKk=9Ue>Es@p|YtQJM20r09Bf
z=K{6ElHy81IO00<h0&cA_S~Frg}DXPxI0-bQeTO!e*EBgsNf>IhE5gk(`*cREC)O7
zw;TWX(Rn!jJ!3?@iN(kS+<*Wm2~spE!D8v}TC#X!OUDfaGbl*;rrqSZ1WQ<UMvl+D
zC;f>HI0PNzcLX6v3G0!g622fxTVDG7Tz{{IjmKQU+dJovZeEyOoVs&flk}~UKG(24
z0Na(=NLyQLMLC7c{uL3g7nI&8K(5)3`L?(F8d&5t*DG2ovczYX=LEEwi_(e-U_P}g
z!|<M~@S0Y?4(Bc3hniIH8<0!`1nKs6Cfq3GK9xk92mr`T4{?r#Dcx4{LU;xBwUmuo
zM-D-soEvVSv@pBGOhq*}x8ZO)4uf*%(!VvYSKkYrhuMr3_l^$;3eXWEOWj|&ElxM>
z<r}dMon>fLo?EZaMU#M{<{NW&n#nw>fge2+lUrpqvlj32!q{^imn|?i>OP5rx;OmH
zRpSj#-ivn&3gCOV%%Xh@nfUpQ#BfYYTia^y1N$+V@Ko_K^3F$^XwR4b#-v^X09YVM
zGJpI+fdaQLFVhg^LdUUl`Dr7oZxb*oYNzV1Plngoum~LI2P}~k75Zti!e=*sk4~Vr
zvm*^`_F<YuO~T7dy;`Lzk}?dG43dxmO(Aws;Am9*C$^EYX!=7X?-(p^x@@nM&+YnT
z*X;~%&*p~VJ)Jq)dXT2x701WLB}C0Ec+WCKxorVFP1EG}Un(lf%1h^lPU&uCciA|q
z&V#8&qt7dPPU6kxPM&SN2+$IL)z%g^HC<b^AX!;O8iHtVmv8(z?|heT|BGr5GYItU
zbRp!v(|^5xOP4^PTcf!$qcd2IB7c|H4>@?9X6#!ba5pnM9oy3?4b>+&(z9*C1!ld^
zdryA9bez<CXfC%I`119u(y$;m@7a#-lZ^Xs2;Y7QTZS|z{Dnox**M;@{9DtUuN+6o
zOaNkNF)Xxw|Myxqzr+)vHo(<Iua`qJUCle*N=u_x>!t`EY6#&+|LXnbd>+8^Ub5=G
zYntt!W81DK_W91qgY9;{)!aD>vl}zBXqfMk3%?}e_1IgMw|+)r?9eu0{;rWl;phR@
zBQ7D3_1+;LF-ZpS8wh>C;(_;7LoLeQ*0$PXi`P#sN${18gqjUp4@i%(9+gyN$U#g7
zZpjX*B+dEeu#3=S$FBoDr%8Kh1#qWOYBObL?}waf)n%3<tzy^t-nU94_&@T^thr~u
z^P1D0H}9Lv6N{Yv$=!zme`t{B%%49Jsh{ECs|p0$!kL5pEwC>AASnTC_wRfOyCyf3
z7G1A+B1WOmf{ht2RpZ;#uf^A3_yzuw^K+3&b*@~N7b_58A#xjQx@MD5TwF-a2#=QN
zZr}&A;AK(~Pou@N0?<)D<24yebITA`k96F>{cEO~o`L=ZOM_yg8sov%IX|c1y5m+2
zZT{k-DPH6t#65qGxoCj-!dsnu5PIM09LdXL{P}YbXM3g~hayaE7#hMUy@$X`zE?dw
zGKBbG%uD%6N`QMd#FW7A{*`tQ%pUrwFVThXMm)E+v~_lI&QfYSm<dJk35|J#oSeGW
zy6<!I6d7*Kse^n3Xj-s!HC@)AmzVCQqXM@sLUoXk#ej~MA^b~1R$&$gIBdU*XZJoj
z7%KhL+WS68R@T~TZQQc-e|P9x^z&u=>AK_COMe6Z7isK(Sh2&WzL&W9batF|zZ;U$
z;BJM}{|Ji}z$DhIHr@?2@R}$)PVudkcav0?G#egjpTDI(ok+T&kLC{_#QTH^iV0r6
z@$R@?RY|kM!di65rYgZhK(o@XxDb@N-+jJ-b|fw#@z3uDo}{qX*{`_Un(kgy^#V%h
zq6!)>TNLrOh1|LAcdbx86Ne7M9$=wiN9@%0)&>1YtNLmM-#OWwzp!W)AFs2mFCyY#
zybWFl;04??(>yfF6Y=af{6E0M6On&-JT-ba^|!RNbg3)uNmLP9o%`Sbzgp|WHR2&$
zU;2$ofUc+_Lk(A#ZovR0td)t8rTZ#DM_Yq^CGWrBVF*kH2}Zw5pj_LG7|h0iytICu
z{Yt9M;8Q?A+o$A5S5pNi!SwXZ1kC&66WVc}=VsOc&jNxcCbVXET|ry)__3vl<-42Z
z;6#{1?>zLM2v~7~g9=&L`r3w%L^rq$ZrT={3hGNsXD1c*4EQ;=d^#XMFbnQne8u|^
zbjt7g1m1%(X?&t@g_MeBU}IHvsXOUolM#I0yQI*sc|qU?0VuQ!Pi&*9_Sg7AgWCHA
z`Wz3sVn#G>x$(gP4ntv5TY309{9$_jgfaS;aWU9a_q#L4wwjcT!NEsA&^j`%qNu2&
zsbf7J6<DZ%^R8(6nQVX}vB*oG7`*>nT?nPoB4sFWrNeee@c&BF;L3%BC`lJrRsH@g
zQNY~YH=gSQ{T+^fIt6QXdV!^BPEl%MkTz%p(!W89XjqdGcyrQyB4J<}q_zZKhT7Um
z$TnUn#UV>x-dZ~7?0Nj2|2Gzz>QnuZk?}MGGh?_#=x<4|M8VsTAlUiG&#8(xl_^4c
zdU~x229Oh^|I?}+cP0?l8Y!vGyWUt(x_mF0F17P3k9ir^_$=Vuy(+wYq5S9#V>E}^
z9MI-sDVO!6&B4dA2FU4E2|YauX0(!UELt+iI6lDxl^aX!ghIM9LzFU#SAGnu#6Xx;
z9%E1Fn+FJSt@v|Q&siXeKMIxH%;*>393F<rbNCO`#Oc(`+$}@v%S_Cd2IH2Tkkt!!
z8vejD6EWz8FNJmu)h9EKA*?lBsf}!l1)}gaS2j4gIdH$(TU+L(i)Dhg^6Ar_=@H^e
z$Ue~ob3^eEC<_0(36aup{qVey?<K`B!FUN=7w(Ckme)<=i)}S-9%R0a@G-#Ol#*&Q
zE(XmmT46df6SYZB(P;C_NXcg;BP{W}UYx#-_bJ1fyX5-j%5#fTGyRAVa8x|?NQw&#
zugqnCCkn<H#%x44Z@lWFzec<bsKPC*j6d8Et&=2-m#4YcB^T0t_Q8_kz^N=eE4MUi
zA!#$dMo@QR!cvW~xp72WU2<%A?5=MkR9Pknlmg&l{*f{2ACc^p%PSZXgNjQ|ZKzMQ
zx2NTvf0Hu9eFJ+JWk#S)vaqml8*StacM?)&l9_S?M-`X_;)*U-+5KQ8b*|x+(Z2$A
zegIN|t-KsTuqefsxCeGx^&MS*0tTlbqRI|ndu1(JO3FeCR(~6D<)`{j^%XRA{tcSI
zxk~2gyYAVx_BNY|nivTxE6cnIZJmgAQ+0e=LOOITv=vfP;x#Mqwfn+v`S2_5fB^r{
zuo^7EK8ddi+kHbyB|l3D@bF>3wae}gkVva9Iu6Rb4I4V*3+BDtL!)KoKPZ9&5HxrZ
z%5AkfO^dM(oIvHu{@_z_YF%cW<-g*>*V5SLyVck)ujcVMxvbKt4Dc1M8Yl1LhdC?{
z$B@hSlBZ+o7IL`%o{oG!Ma!0PWIGjHGH5y_ZZx)b>S;)Qlu9>Yd%${U6c`$i{e`_K
z0n5_#?mdVa&r~^Za)y^MP$G>!dc57RqLLTl-Qf}+66EKma(9hJKBM?7DdABo2y)=@
zam%!{A;>`brdMg&r>4QK%vhV=)ZX6%9T3<C21q!oM;i$z(cB9ZZ8&$W0)O!V3sR81
zHFclHCh9D>-|__&>ie36f@r_V?N-IRcL&()?=V>A9=##P3En4PD%)SizURv&4zz}e
z!RTi+0~SvBW|>eR^?nt;g{}K<9Vxu&6Uch}mB$)R75K2G*}o1ibgXa7U|@h4)WkBC
zg0<DO7$WI8hlZ1Gq5V50w<h|#H(QL^>!!ClW4okt8-e<(a8%yb19^hLHR7;y@=2j3
zUdubH;=f7tZ_jS)guOSPo~d|j11#WR;c%&&!BP~`vpqmHwR=hRm{ulz5H_L2hZ`^e
zPww6Rw+=Z$xp)f(uHxXhi|t$nwVuL5p0lfTbBP;wqY?54-0{9(_}cyHmR}|=xFW-s
zeJf;vxwnZ~C3qM&vi+Hw3TZa}{B5sq6ZA4b@4&;j`H+n65o1t`I%tbci)ns0!?+;)
zb~I$x$jH9X`kJ3d5Bc=44qmbEf`}HhknRgR$vlORBE30-*=RR?DP%i)TT2XV&x?4b
zJ3E;;I9~T{qckK)GSycr_~TNFefaP0-cM=<hTXH%OLCE(`uCdlzitvcKniE_)pZNs
zq4}=hinF?NcNB&Ruw;_ZZ;=Z)H+DDg{;g}2WV))&@IEPZpDzb@n$U;BC&m^3`^3}r
zIAIq{aI08!Y(*=iJDhoRuvwiJ@ng2B268a~v@PG9wzsixSiIP+2V!GjLZyRx3fkL+
zhOIhdo|(DcUT26roT+pcUzuq03JSY6h07Ko`c$X<?Y~>MPJ^?<Ouc)iDv$rBplemX
zF7}Uq-n%R>(1EUcN<bwM@b4E~m@qtM@+Ldr=Vxi`_{GWiC`q*8SUvV4ndZ_YE8*Lx
z<gn$LssFop4qQ)sk86T8Vq$VZFS5O(#1daUU5|T^`<44m?CSoN2RU}&v#8iee6_ph
z>qG)pzt6Zj5ySQlY&UVG!iew~O!wF3mobT81boTO&QApG!0;hs0=Ar-uCel~G4Lw^
z1L7;{l*j+~!~Q!Z{PAY-9<abx{+zV5EFs=A$|y}Y<A|^GPtM(Mot$5!Ms)pqCR7VW
zzk6C##PtsC6RfG1x=v)CiwiAmR)ael^zV1Qrs?Zd+LC{Tomqg^9U6Wh=<M&^KF25r
zslC?M_>o~Ais4r`qh*WPo-`W0N_N~2U#WfnzUD*43r|lIfsl`N>6y7tDgJ^=UJ2h6
zUTRuNKd@FqSvxsuo)|1EE|$?(80l<M|C_F?q6Yq7kC}>S?gl=ycs#F4;M=ICtZ}fv
z3xFxe#cek^ar{-aHBb;<W({fS+sHH>&6D~ZR%#|{Thkn9^bZ-Bq?OKC=2bOXvYA~p
zvsnk-0VMLP45W|Nexar^k?sEe|DFWZLNW%9|LvV)?49GwYLdOt)YL2V45oiKL!!+9
zn1#=;w;mw}d9|=EuBWh1XBN&qId9&$>MN=y-S}q7SyfT-!pgnn>~ZzYxgA*EE=9?)
z;W21Fw>7~@Z|5c=b#iK>{qy}Znms6zlf5@bE@6H-uB~G;QF`Row+`X!s6L-chc@%g
zfdTLJwd`4xtkBs6*%Q@yfS_Qe_eoKUq<lk+_Oi||<3mNXpO%tC&bzwM*y>KdPM*H&
z)(|of*DO3)E|LR>MHyuj1hOUoA1`^?d=F3u<j?n)e2AKwhGKll1B6};GcOxKpb09}
zzApKD|MpVP7Xe35WB?BUJcZZGHy&23d`+#}>Ywqti-(svC_qE=+kVOu!p5iSPs?w^
zE&1Pz{_o!W`^%M)M}1felw6=C*cpxobe(j%exkX3y5kqK&^+FFUUs^W^(-c)*22=_
z6Y##laK{&;!1bCgg+oCxt}G{b*m?FK96BM00tc$8LA_OOtp9rZ6T(Wtq@thb1Yo%0
zb{lz4&;v9kVTXl<_0JAG=B>W<TM6vW10e!;&@6yAN{!=xzQbOUp6+<`i2+>N+uo^Y
z>!TpSgk4u(;ym{@XbJN3)AERsqYaUoH_Z{@)E{kutlZQp_tdr~G_>dLd=?`Eqi(%x
z-uAA%{HK37bG$)?NB2ulz0KR%VH?xsmP;3AsGOGDg$4H+&CVuihq;_dMhn02yww&{
zu~N{#n(<us``;b;Tpt4Ki-milxnEzYbP=85aXvl$THr)<ba;93kCSTFT^F#`@H~m4
zWuSc5as1z@BN+M7W@e@L@-mN%PKbZcEh~FWTT2(0_>98Ut>S1RcDg_Oa}=W(m8bR7
zFYTtyEp+#GT<sNk9&qHy+~*jB{SdvVuqKVMC`Adt5c=V<tKa)Le!mwui4#MG0e1z<
zXwO4ytua0>=R+y9vz*Tz27NRF*3VZdhaOk~KBFKL84+<-P1bdc(VmF)(KoRy<HK*w
zdeQVj=>6JE?}Az}_2Kutsv}1`#al!euh`YzFMEINt#kv+12lJj0Q$VUiO_4GUQYlJ
z#{NZ4T7U!tW0zcnIJp<t+Qf)Q&7?9VBjwB-ESbP|%SyL(@ZyKkD}3B;<}6f(!4i16
z2$)Jig7&w9g4}O8pgu}7p5fu*LKSs2?V39++1UKHE#v{owIaV}S~~^=H=A1FkO$P2
zW`h)|;1{@oXi1oMJ@ZBkOZXH*%TdoQ;NvTG?PSOOV`4^tJK$PXO+k*$nBQeu4BLON
zu`@;;Z~&AIaeNz7`d8jx@1J@QhlE-Q9W^)ilObX=b<jzS@wus*v%V}rXY4G<%q>j&
z1{TxSCT2z^X8rAkh)=q2NS`$I|1AJ#i0W)xseorb)tp;opIKbYAk2`Nni?pUdE#CP
zY~|yrb%0iYg?gFrh^|K$=0ue)^S_c8(`Doogj`nn&o&ke&lUIkLD#+|<S^9rdxiB2
zCb+gO5)!2L!Z=q}<rmPoc}fZA!s5ojGx%S=(QDN+e?@gJE)(gH6YIBj&xS@uw}l+;
zdp^4NpQk71Vzl~u`MEh`gEI@z3>}m4GP2U0KPH!d`}TYqRa!E4NMs@XsXREcT)i@o
zft?E^0siN1$NT+#76;3$24M%sn>v&y=RM~P>yFOyk8%9>_jf^<2yYP(mY{J<iS;-=
zvwfEZe$MOli|KB=d-5+Nh9F7&>IONs)j-^QJ$AG7;N;|w?Z2<2-^SN>N;_WdVXLJt
z>#+EI{FqgX9t!IXt6zGbf$6+qHgcsTJoW1OOV}e`nhP_Bm5$8YR~+QsmCr|+81oAg
z1g=>3`agV>4hjrPlM5%WyReQ{=h9?NxJVZ!hTRk*b74l|#cyZ9`Nr|p?`wyr>({nk
z^r1_mLLw$eWTa+F2K|gZQ|v_K!G}p68mhba#pY~UE7#8!B;}$o5x}&hPR^n1)X-2T
zAzEps%i`qviTHfUp+~UDmqMcKf`XopPpqPH%E-xqfFfcToIW-&4_M|lXWuNIks^Qh
zT@hTo*gvCBeH)CgeVDm9l%$Hp<elX=8)gUz@lX*NnZSOOs}&K5V0izD?SEwt)xwMz
zo$o8~mjd(xX;{d5*YGnUY1gvzs8&v+cl#W2tsw)O&F$BV19NNZiDK1dbA|W=_3ZX-
zzYVVdhR7EAha&voiZ|2fZMWGgvi%yJijrY*Zg>XZmyrQpF9-XXxQm|^&c{XoS7>^2
zYT8d{x;R}t@#KWMFFchM1}2Xb;z#Wxq1QHD39hR8xat_|;cNgLrE?xu?0)~lpMd3A
znwv;ws+?VlK$LLDp#bP445qb&ufuAs(CXdJA2HuRlyW+EpAZ6?0vcrW!+u?U$4_@I
zNZ;bn&}3dsT^<Mps4yU+nOA&CbK4(V&zJtBWM#d+Udl|>q$U#Xni!p!7nfRUNp`ue
zB|MxloSg)LNJwnN#c0LzEB8M(J^gsgHG+EA_sP3mRCQI}T<wKmTgP~-TPOFPTYF($
zH^aH}vSmodI4v!o#|FT~QlswIT*EV`%?wdwGhhA-yoqoU7XYygp=o}b;J`7Ug7|o_
z6D9uuNLDX#gDKw%`8yoXc900JuGr4Q1=U`XsXV@OOB(^Lv)V**P?;bp{j0EQwZ9T)
znk}~B)~Q`zri0=S9ARvF?Ts273<PlqAKtrq{PX!guV?^jn6-<q&=vpxg6l%>;i?az
z(*NTE@W0_d@}!Dyq6`lBZ<0MA(wvoD>l?HdVMMFsq$FSj7qTB~A4<gaUV^5jg|3Az
z19p{~i(hoT1~+^neG9;t!VqFC2$R3;QO-w#nO_bf!sy2#N&<!mgM>vu(L7;Cff6uU
zQw@r+s%m~Cwg(6}B?D@69r9HIpiGF&kt6u$OVQd{upO9}XASv1<KyG%QPCcPQ)h*>
z5ZcoEo3?Mo2|^n0dI_8UXhXWKt?lo46Pj4rSeWLpp3p_MW8%04>pxebj*}ZHY0w0o
z6m+#<5c>Xk=T$=d=cw`Km#~K1+$!dFT@kQYIsFn>P?VjXnR>Nw`W(%7c6Ey*>H{J2
zo$ZNMxrH4pfs>Q5h=DywrG$_pu<rpIC!)3WG+#JMfPKn4K9Q4~2R0r+v49M2l{QnS
zl7A^Wq?<@eP6a2N16>;rEg2X}W%T6Lab1AcmNtJ62F$y>35f;K5G%3g0&k+)=ujoU
z0NkL@KewOJ%&pw@3JSvXEw9PXXX0m}tWSCNK?2-ZxV9Z@If;d+7|Va;ePyNs2U&}M
zaU{Y7!L)Y+X{dHP(^KUo<}|U=3k!u=D2Y6?KDXyF)(4}=Jd=g#KxSSVNXN(QTcMJL
zqX61Ah$8Trt^2_-@c#S(DM@m?W(pi`(9o_#gY6I?&Q^IHm+wj6zZVBxFH8zJN*woY
zyNjhk+jx_1?sa9aufD&e1d6Dsny!K!3L^JFK;^F`@WnwI4Fw~Znbd`~BQVbdM*PAM
zCu6j((hKIn<)?r5Py1hCRJz>~B<uav^?P{isRQRwG4F<^;D1otvCe!b{_^eaRo%{c
z$-mB?Cey3q>=V{-dn185-uag}y%%a=(Wa33*`Nk<cHgz>(xbGDvrivzKVF)eysD29
zH2eZ%P*PyxfZ-eI9{0Vl)(~>izHM0|xtD5P=W`Jpkh5Z8Z(?Z|)NFi9he1gA>lL-4
z&fD|+Q6pe80~#>hGd@yPk9|N*$8+uYKev@g@m^4a6w1pfL`UJ0FU?~hocuzJkATwh
zUX1em-2FJ1?$`V-^f>LgtKg?=B84rTKLiu7?YEnevY7|L%2xk-MfFs5SQ!{*SC^=8
zFMz`ff=bbov-?Gbb(9?ZI2fL2Sma0ZJ)wDP<B-`}{sw@?A7R<?FoHK6gm-8m>VSn?
z$3SzUa}Zl738qx9c4c6@XuFKnRQeK~m|u|jF5$#O_}WwK^~*QlV5_Y3nH=vgYXz6Z
zcd_Kh07;FDAfN=yHuo6l9|6^`+l63Gdivm0v$B$5K=)D&4mQ}ae=jSCY++zys;H<r
zIKO#t0;$kU(^3+6fghk*KYEg_fHpt8{=KE;5oYw<;#OJ#g#3nt$iw4Qz$w2Q&aR>o
zkuG&2<V7&l{8BNIQ$a%FCSt*P);XFAs}PQQ;Uqgx^KSg;VSDPd?;r{7N2xsZ12S-7
z56&}zXsPu2;L!Y>Z}&1)F8$q(NmA-$WwoVX02kQ|F&`2pWto+^f6;Tw3WTUQs<O(H
zWK%1u`BvB0uy$GY-D*C6cey3m3~W;X<!{YHfSouDK^z@}GcrEFZT#CR1j%OHEGUSu
zN0$VxPmPdXxSVZKv~d$v+TQIT&LMT_lT$Bek1c5<Wf>=3LElC=q;`%R@E)Kg_bS5r
z>FQZsasY%bd;??r{>9Sc$Yx@$Uq$7G!kmIXs;hMjNpx9wU}2_w|I)jhpPP3R>-ci5
zw{@oN_}s$b1v4lltc{Dm7zvZXt^oM#EW(CBT+OD1Js=+7R0QPueas-(?*xdj0D>A#
zo%IkFX_$#^noUWT5>CE;#B2=w^TDi7_Li3Tu5&Yztv$&?o|{3<<}CUpY+U+fHb#aH
zZDsMtb;C5QcL<R`rM7oE?1@SL{&b!vX1x<T@Z<IUz<u>tz30zK5U3-#IJpA@g9z|r
zIkcDotHzA`G(szps|I%QNE;LVr;z6fJl5mjnSJk`Dl9Lr>53&syQ|JoqlkNRD{gp%
zc(=gVg%mSx^3Ra2=FI2FczYXDHW0-8ImQ4l(>Km<cpXkm$Nj@rHu7U0C-A#nfX&AA
z2dRsx7K;CC%sfclrL*IM=OM&J^V`IDuXKs!RL9YqufB=c;m<&IG<4d@^T&o8(*WCR
zpPQWz8yR@QA!4?O_|zoW7?5|TXFp}tuaim!efdv&Egn%4*g0XC0rfO=I+j{);wFRh
z=1;9PSqOe_cmxx&&O8)yQ;3p{j*RrYU|w-k+8Mr0r_6Cto%2WU90zRBZ@E>DSDnSX
zkR=GX{U_Nk;zi)JHfi?i6+1T9usN&#RNe7P%i<GMIZKfEw3o!Tgb|I6N~`K?x$SLa
zmNCVqoZ=_<8k8D)T59njg6_nwS3^b+DFSuh!tctIzqkO8fwh9PA}YDaTeqtr${X)K
zAM*X1+$P!ezybi58d$jP`>{Ml3HV*w=%Tv1SbWACHs9WhPp9$Don%A_94B{u6(+L9
z^f-LPK<hbESO74e+o>Ov!OO2J4?;uXSVnxH#oX)n=;^L@SbT((mX@c<H+cwt4$CSl
z@Pt7$3<F`m(9_#^8<yl&RlPA)dCR}476&6N^f`S|jyoQ~yCB1l^}&^nmlq91zc`6K
z+w(&3qt)U6fyw4bt@OVh9Mn2URE#>aPHo_He`*TF%*Y5hbEInx(y{9C9#<Ds)%-ZS
zSOn9fu%^|R*ho@7<1dDTaPli8R0cE}($LaQfA|QE2Aql5I&V;Xz!+6oW)-|XdlCgr
zjZObUM(w7nUoE46>H;5MT~(c|6y*i;5v1gSiWnLO7)SVrB@;cdssNR%%Ik4oI@XG*
zDP1!AisYXSxvh(PeZFXjBC2orpwZ;Y-S_PnWMMkBMmn$^{#Zb*w;?aRzy$aEp;7tk
z&w;P`A(H@XPx`N2HU;hzkwUm!@8WB~xUL2>UICeg_{0K81U+fda3;V@uV_%CqM>o$
z27>ggx$Sy;+9PoZ(S7_C^#LeP<gZGWplWWW6Y|)jGOVzlJkJR2QGi7Y6d2!$STHI9
z-N>)S_2-=(Wt_wDU$yO~-j)K0+vM+tD(WUV3`5N7sw!~O*1uVM_bw$y&+wn28q?vH
zvoka;Eg}H-w`Wzajiv>D781(Ugjbg*S3CtqI<Va0m}3gdHLP_eYNdN0g_4wc9g40+
z>arZeiv_dccVYIa)jKCze@hzvuEJ1mX<kcTxzeM1=?(D|h_dpsj+PSr>o+Fn))!g+
zTspT+G-y&<lt&~+9#2e@B1#(i4Br(2_NwDr!UTOl(O8!u2l%f5<qG0x>b-l*pX%Mg
zkYUhq*=WN14|;6adq?f2>R)eFgqANscI{&-Dw8+Y?hpJyLC^`5)tQR@zZdOIZ8}bu
zN0-(-+cLoF2o4l>Cw@Ro<~Fxe1S+YtA7}8>!Q7)wI`U|C5MZO0Fbq-QuRxBFkZ5UZ
z?|^WN`vlRGwbv)(AVMzwHw5N86nAmq0i)|mr+0PwVcwKHMStN#?0Q0J&CEdR2n%bs
zBxRjYF_qP3s*;tDD^Y@~)`K*!C?CEeJ-%eVfmB^hx60y_pWC0B`Z4|qV@6R}R%NEy
z)j*lNd^f-}(dxs%@Zf)CksMMcDX(`4>tEmhqsto(&_n^4r>Q=wp{4<otaiK>H4SxE
z{mOSz=_D)bF#LxA#}l|@7}6t3D%y31tHkgfw6v!jylQfCr5UBVP!Y-~r&eUxG*>-N
zOvD1asE&>`5*M<ECuY^v>lCMU%IvR-#DAes@YJW|WZXwI9_s{naDuxj)2N-<j~6g7
z?(TlfvnBYV++quMQvcWKVTpBplaue~9)suQNTKfgF3ZkPPoSMjY896k&hC{>{L3dJ
zYljy%rQH=Iejdh{$QNRgqr?7$TGxNviWNss@j<{d(Civ&{S|u+>Ymcz!n!3Bxn1Q5
z<mZyszTRE15%(3J?&xec{{F{b5w-t-eDiQ=vsy8340a;es#5wR#2)Ere*qFIzV%O2
z&mMIR6`Js8VM%Dh9(!}}=)5(S4wG*jRUg+lZ2~B|{1rd=r`?NMlV1A}`R?1x-3P~q
zVxbcOZGUKQFF%~Fk8d|ltaQLf`1u8XZcUCVr0el|F3iQHCv0tj-}A^Zzl8XV)D54s
z_a4A`Xq!PpKxXGbdT(oK?&P8|WNLz^R#DFFfM!Q$GgXBFjQya>1b|{iq&=4$Y3Q87
zM^0vNvT?9*z-uvq?ae|h31M$C|EGry@goH_u1hOma}W6}ip%hgguM@=KYylYW;xwl
zVIWI(+3$$co8SiK!^ZjM1F%=sO4&|{O(NA91qXGfUw^`p?<U)g5Gm=LA5-J%Wc{JO
z_fop|W=F^WAZ&@Y`u7Gd*EP=bU+cE;QI@h=>{vn<@uQzMQpT5G-$3@4<)nlC{g1bk
zxI8ftn4YmS9!RRK1bg8%16?Cvv{!+2>drrC;DN7Sb=o1NGgtKT9s8~6ats7`R0CWo
zY!8v&b&6`5s(c6QJrkO}fU@BeM0>Qg-9)(K_b0+t5r?G{d&O;oWD#?B9RX}J3~)h9
z;CG1=qnxRKwF}q#_KgjQA?rLYj}Y^_c=*wg(NAUiVAcQJ5}MmT#m65;D8AhC>^Kv2
z`?d0AC-xcJqFbLea`HKx=>EgB!mP8Ctzm0I1t?X7=vYN|hDneHP?O^B3%}Ywi)Od?
zg%MPRpuUnro06Rt%urxIFPy45KGG|#N!VB@%4DTvgUSi(c)#-1Dg-Sswx!3HxhxYL
z_!Xy1=?XgDM@l(|`?GB0N7Cfs;D9io-ThrG50CTHruZPtz5e1|i(+j5R#Z`4v4Kkx
zLjC#*#njLF)>e^kqdn`J0hMKB%=f4jx<Niu=X}WQCs+6C{HY#zhCLK<d^g=+Y`FDl
z`HtOWy<50w!%XFMM%oW|`r2Lfha2D0#S8$3t(v)Z69g}*UX;JFr11tqwM<{IK^WFM
z<Gpe-zKQRZ>x|!^G)p2MjNfp`MQ$T-dMToU=|LcMJ(@^;HU%qB6gSZKsp8w)yX(Eq
zg+rvIpDyxie?`8#Hg4qf5$0<$Z~Y_;e-?o0`EFzFB3FDR3-|11I3x>$J6Y?8`UF8E
zEF|S4HdcH$I~RM$NmO*KK@M<PZSEjzYa~!G1JS|WK)_{dff8{gyWe$In||~hXfyy0
z5U;JH3D<eT+Y5G+jak|68UaZJ55T92cId)&w#KGVo*XZ(vmv6>Avr6@=86wJ2={cg
zWWuNW?_L313MDl7Ktj>}vnXkz#)}qep7AFP|9Ko{<>z;{c3v(H3RIWd0^t}hVSQj&
z2f$MksWWups>6*&Y>-ximqFe3EbqguzjcY}XBEo`1*c#{qp6+t&%LGSCyHBo<WB~Y
zhliVA-cfroGW7eIyn1D38U8JFXBU_H;GkQG(*mA3OL^HKcaeef>P9iV*^F(Sq{3cV
z+b9sM@JnB~&i(4<zgJ%((g@g$a4$7`lP~t1q!oa`7YB#C`_bQ)Mo$6Yh=nEci+qm}
z<B2_AZ^Y9j+y0f?M5sW{=dy<z+bJpOTpw0gSU3}!bi>Dy?v7Bcx98TkTyT4hEt(RG
zO^!;)%F6TD`rQ~G+Y4Q8W|p@EOMTJ%3yOa&I){{={Nk7|2@h-XsqHon)ARj^A4#<^
zXC=ZvGgk94sP}y33d@t0ra-rfZOQgC6uy8(likNcz*ipHNC~Rl{>{!_@<|(Odvt`^
z?sNNgNs1x2KOWm8&q{k@h_xo@jxo5ws~GBdKx8gpaB+LNXW@0FqJGUyWPt|uT)_%^
zI)p9m8%zWd(UE0c#5MUdGxcLG7{kDUC?Cfg+Po&#B!CFs!F=h+3$>!YaAk?fy8DnM
z??VD4G)Yi8+r?()$JRhx`p?3I@R$feSD;L8KRng38T;<OnMaGT{hC3?zHPyci9lfn
zC(^%r>wbOK&14QezaJBivQ%dZXsYX7oz^D<eG%LP#*E5Lzs|)i5s!%0`O9*b*7@pD
zh$#qGfPfO@Z%gK2n|IZlVU?jzb<YV`^rb)6{V8{Q3_lQy>|_H}1Iw&46u3Jsg$Axy
zJ4^qGB!iJf^)WfB==0~+UPj?nU)Ktl%)x-{W0SG3_^(@-CB#;Yh{OI7D!{v)%b`;h
z9>jce&@+?>FT-F<xbE%Y*5Q8qXc1@M=;QS#QkQrqSG-?)D^DY{dSfFZa_lYE*1!b$
zZ8!}5hbG@Q8GGJ%L=mEIM(YQ*x@hUk(dz&WM#bF*bc*dkPGRo4K-AFC$mvZv1W`yv
z0X`8K@9VuTzx;s#6;A%zvNBtrBepo^Yj#yv!MX<0!M;BqLNSn6+j#J?4)rg`pxgjo
zO`x4N7#5elA&$V+uQi&8<>%7Vvq4-*?~xIZWu|?DwC7vBwoW!Re|2vt(fxz-v?`w}
zXnd3-u(YyTJ-z{v+1*S$TMoQa*Uj<F|JQ6?#0G{2*SjslK2XJ{825+nyP0TvoaDZF
zcsmuh2Ely}5LCW-<D`N%a`(3(V(8=}g66w>^!ohpZ*dIpgT|}g9a4kylBT}0y5<aS
z@)s%=77absiOEUW?tc7$DijD}nUxGkR|3D~hvYlj%F_D42sq7%0>-sHv&E)4Y}8zR
z9_Jg)>u|im0SqM-l9HZz%Ev>F02id?(?bZRoz#6uO=YrjfCI1T7c2yJ@Y!u+1W{~A
zca%?JBM4Ob5+9esdmF~PFljsp?Ma;sANUbeLRfeM0Y|LY-|@iI_GR|fvcBi}OHUJ3
zvDd`NC(Ir2Mg;QA;MC%5Bq4((Iedw7)yS9dD+{8@mJTWw;arpNEp(5vZ;1NT-ysNA
zSJtR<ycERokBo={%@DbW8EV6q@Fr!KF>M{%N7;|Ev#wV;N@0Zp{sKr5WbF7h-ozsR
zSY!FG!qeBB5v93jy-o~;A&O5_hN<!^OULKWEwyeXPE*p-TH0E|Wde?07aU4*pbH~d
zSm$sm9FqwJhlXf=juj7UP)Z{~t1tkh<ViMIpgU|8Dkvt^KP;U<Nr9bpeBc7E;uVGr
zc#SXOtFK%`0XWK<*uhHQ(bX~Q<QQ=CMtoDMixZW|#!p!4HxV*B7I4Lr)M|$@Cv8D?
zm4>kjBl2nI_^CoB45wh~9jPiJm&ArP_9-?y9~7EM1@D-M+f~GLKMDe5tdW?Pr+&>a
z4+Aepq;FE7Rs!QG5JVnewLt8#f<{)cCJjm)9x?mx=HHWgq`*q1$@W@_uRh4E`0?%2
zeMGnR)z7M`mOpKVhkMPa%X~M0sD!aY{$_Q&AfWAq_4EDGJLD7Nmf#cs^dPxMOtq~H
zT~d<D;DEnILhQJ=*WZSU++ONRQDK1O7$951`XD7u1XlSBzvf#3><~I?|2wD%9eTC2
zOQBNR!n?t-*RS^{EIIGzzVkqtGd9*rb`%gGGRShr_JffWvqo&yS4{2XBre;%w%Mk>
zlfc$_)?@*FT@{U)z8A{sk69i`rT2q%P@UOu&LYjdaM?_lM%&LgHuSV~v~*mIPKn$Z
z$1;5aA-41_zXzvi`M5~2BcK)n9LGB39`%KXAnYvY>Phb8$yi8JjS%pYf+K4yD!v%@
zh%h$vHceGlRB#Hj-;%f2(bh&mB+b?=tkaT)hLqBSR8o)qO~#`zs4qNaWjkl+Xs8oM
zz+c=+^g#T{uU^97X4tJ@7Bw_7Jnz5@Q^5CsAt<=vCD5EPZwH_A3Dhm*r;9y*&eZW$
zVm|w0RbFN8ZuYm{TuUPOK(x&TV47i}nHbqIGBEUyjOeTBGCpD)9~og`pbRA=2kT)s
zzA*@c=;d>xj|&h<xfmaibd-@u^UvRZ*F@KkDvZGV6EQS1`uL+3s6bT?TUZCd;3w}q
zKd^f7D-E{G1UsBKxviCrf2xLDhZX(c=K<Seb%7FRimr(tKJelMcb|Nz-2Q12L+}z7
z|Fvjh+4gZig4Cvq_Z2W}5Mrh;?zv9t>rJx~@|<{hK_RFac9i*bJJy^CR`&7Y-J92|
z6M02jOs?9vsG8PmTU#vGwI#NHAs1zyhsK#V!7kNT>i>8x6HoyHd+%z8eP7Ga!R~4Q
zCX>1UJ#J<ON*qXwK&yR@tt^S_#sbQw|FkknA{!s>9LM%9LkQ@RnFIYhA5C@TTzd;#
z8!KB|bVRm@8#_=!MPseKEv4TAQBv0OGk%96DyEyzZ2GEl5=DQvjB9xCiEwm`I`|?%
zZQWd}_^q%YGCby<+V9V}gwiG9N$ZRIC@G0j=sxfQ-dMbg@#@t=-P6H4QBsdj90O!K
zA)JV0lm+y?ca@7r?WcOXt~}N|##s5ybB>R7wtS9`W)3O8ruDfKhl!64{M@~0R2Xwq
zb=Px#PizD|jKweNFOQFtf^odLE%-thWjFD4PJ4g76$n8?Y`Ap{AFL3Q5P$7UkXI2C
zqX>Tz<!EVbg^0SDsT;rk8uUYD@|=d6+H)l&{nQTDZ(mJuQ6_c}?BY27`GvV9Kg(FQ
zLfkFvEXT^vMUR$s*_D+gZ@HDm@VSz6d_gE^ha*V>F4OLuk6|S$`0K|%y~Xoa1nmy;
zSyQP3x;TKoVZB}oR$`%{A{jJqqm;gTCiRuSaVkvOPVdU9^V6;GsOU$538}pp^_GzP
zM{Qo~ant+vcoD#ES@o(9g()rP#}}c3M|?0pIyt?e8mxV@PL?%F%}B{It!b;O^5dAD
z=+~tfA#DW<pi{&DwR)BQ9E39^y`*RrMzGvmJl@G+Xa$M4xBLPE0t_z}9C+V)2|Xl@
z6{QM?c<i%}X$RT8(GkxwJgdu->OH7ggc(?bF%af%c^BIWUWaGIYsJFczIYD_9v7;*
z7b!?d5xZYLMQt$Txao4!xAA~$YGLTk+dB^}t+OE<_V`ldOpXTkbMu_TNu95hsncYm
zNj7Xfj2%H1m-@)p9LzHVDeb)SLST`mP~Jya^s1*uUjij51g)CYfGg=pKc3L^MjE<9
zz&su1CQw{fZfJdDQ6mkQTqlBowi@Ft;-`Ik@4};^e5U&?mCm$?qh;?NJP47RDSzt?
ztRcqew2YiS)_XU~x`E491Rxl&C*RmiIUFt}eh;9}SX-Pkg>b#B?YFpB&o4fQ{oi5%
zV%v2bl(+C|0@(K@?BB+bUER5yP*d|h4Yp>XDLupQ?JeMhXK<^Q^bNE~IN_4dP(4mn
zN0wt7QvvD^vonB(dYyY)PG%j$|EYdPj{@3RWV<QY?F0k`0S`1dI2hF@+K;HCiI@t=
zrGSv{s0YMkRn-@1ns&X#LK}2hJ;@(sqGUlsOr)`;Wv)D*)*gG-N<0dt$gJUelC59w
z40Q3dA9=*JTj{yC5eOW;A(~7-FQRKLt}FMcdF|eisc2$gg6zzwtFp3{w7EH5`f}EN
z{Ma{+)x&CPIcaLZ3S8cglOy<QneK?;A?i)Faw7LAvUeE_mg<!MtT^50I6VwLzigHe
z@%SLT$WY;HMoaOzv=8_@!eKv68#hp;6(S^_!G!FF%n-_hzhh@Cktmpc8;i5rN-scJ
z$<jvSe)r(fxnp5|NfUGw;7&ZcvrA84*e8yHl>AJvo|zUMT)pC^W<f(8?;0j?v0h&b
z=T&~HybH>a^Mv4m5ry-!%HPcvoK=Q+KbQg=&tgO=AfeW#{TJW=$JAQ@Wwl5B!W%&;
zNeKl6M5Lv=OF9)rKuTItLOPTV3F$_VlI{ip6={%?mhKRcj=MbX_ul(G<D5BX#$)Vf
z@Bd!wmq5kep{hu`(OmEHH_o@I_X}CUo_#s+n7}gu^!pmK)0lvOO;r=Y&xRfYmWYl9
zhpk5Ozjap^@5Jodzq(D&rrx~qn6$5}?_O|ZB@d5VqqaV(YX~IE`clqPPPpDcxz6e6
zVSOGJsu~imD1L`Qn#K_D&T9{_9Br>-e!km6^$v`wpn8R*UwE<3`aSaK>1lRZTHfAD
zzF~f&Wu;VqO?)vz+{^g3B#$0itfXOCePk~X!p+N$%WbjMZzc{OZd^O}1j|P?pf-~M
zlnul&v#0%SsOjFD2mXC3-=k37fhN^!!W4qNVwR+O$Zg6D3<C9h%Bn9KJJ_y{x1W~T
z28!d}ZAAfqR1J-S?EHFj(OUw?Yb9emLuWomH$QlYQeEG`yixpiy3PB??=vlJccG&4
zdKozKo<3saU$lm&!;*RbiR>YK;RHsxhd7(!%@rlAq$ekCr)~|TM8qqL%jnHm!KzAi
zXTLi?_&azt7F@A3>;M0r5j48!k@3#eO=D^fgB$<->&psP;mdH5maTE`MR?87`$nal
zqGZd-9lqN5ASaFd+XM)9uDCv1%SwMo#9<JP?}lv*aG1NlyUj6G{G@4uBbf?5i+S{*
zA<r~Pk>^?o1eku1m41Mi{*Uf2%`V_e6?Dc#K;?{3gECMH8ixRO-<)vAw}`pv6PF{1
z`1#A$^9%3`{30eI(R}taOFS6G&b}1Y$65WgxjYx+gRS@r*~l9m5k&N6WzWGF3FyBw
zl`sm#R?{;wz8%H975yYp=@~4AF&y^>_^Aoyk%7U%VCH&-u>aKp_>mvC=+EEq0Vkjv
z0jU+x&)0o+od@9yif$DrqY%lqZZ6n@Nfg=LbGn`0Z);_xUGaMIUA866Pz=W9b%~e|
zvjKm9AmU!R1^iaIs58$ifOKAx0&J5mP|9HOT3A~A=<bBnKVv@t4n5H$Qu!V7le<TT
z24!#{w*zVc%w6B?I#}ePh_6n{K_8%*tUGIba!d*@k+S)!XYJ?D?xtyCmv)Bi3HooW
zyzDjz;apU51M4#Kk3#WnZNz*pj!Q;+O$52&R*&G{x!bX14^N<&i5H2LeQ6Df56q#g
z@!h5lYDc)Qxbp4*&<}XAiQR3)(Wy+`1B>vc8!2)3%09g1|Grm!U0H$RQ?ru^^2nwv
zRR0=&!C!4YQ{D$wL)9W-DgmkQM>0<;B9(E;JW&z3(WChh9e@g8UN&D!S0X8^WWVkM
z7DlKVbO1KQarM_)wcczi)gx_ftfhBNH*v4~sA{Fh7lWx{zwhl4CwLZa-GUGunLFZ;
z-2+BXQTIPd;%=GIt-ng0{^h`^sZU6hm2<Z*{DH65Tn*<Yu2#L<Ev%mse4z^2_!!5V
zDZ+~x61=_?8L+Ovwg5A{;hg$F0j|2ni_`AQ`w{8aEi67qVD0Kuc@P{x8Z_N3uRC00
zL{7OIN~(bp!ji(<Gl&R=nm#1sBMXN>><m3S+DKwl4F1kCl+(W)-J$ZEz1wg^Us6<b
zckA1nU1UZHbw6s=E-WdqtG_LgbTbw?_IS;~@#NP1x4n)|PIpCIF9-f`Ny%$Vw@X@E
zi`W-t7)r=!la2RGI&=J^9NyaJgOD!ZXG4k9eHaBtSEj&J1G9(2llx$j1)mz+jE|ia
z<h#wSUgbPyeD}&Sue{<5+u{AJTR(IAj}wCJA=S>S)Oac2EF$f#fg|B9*#FcycA|W9
z`d9tlmYnzkbTw6VWFT!^__2A$uPh6!J-973+`P#Zz>F+e^5SK;UnycNnExlMg`v`S
zcfZ&(e~s$8H|vXP5@fk2rQYSTa7l3)io3U=!?8nEUtH(zrqw^whnchH3TS2@$#g6&
zk?y$&V>=A#irF5weg}aE?0f}91?WEV3$oB%I(06o@d`bgDfg2JT3GT8#NFAtjGHf8
z?FURwV!bbZ4Gs#t8Quj22H%co3_j`pp`%X2tIoLvGnf*yKNf34z=lFLs}46Fvbba%
z&zj=!XT8W(addUfoS2CCN&fIZADc}=4xAGMLW086DF6k6MS5d{yb}EuNm*6(F4gi~
z63UaI!bcD>1}_s35LK9+qalq{E+xv!SNyKuueGK_FL-oJ<nGbmde;?0lc&mGjxFG&
z4@5tveEz#q3}TZ<pwpged}rYh2TV>?Wpx0$*rQ~?WFg5K))?=kDm4v_%IH=MuQzze
zKe=kkhfAQu0wc7z1yg0R!259TF9l`#Q6M=EyX|2mG)^Vd`lXvb%yf1Hg&(wjW+-p~
zg{xO)+C7e~sUPS!HZlT=kp5I2ViVV}WX|N?<qWQJ0Sd<>dga?9wxg9%3%ODy>}UX_
zCPOIzzt`pu`JrX09m3aAMz<y^0nZIykWo(M-afqONUqs|UH?Pe2O<g--&U*XTDPHo
zcM#t0?OGTpTi=cDgrJ;zl!4Ka_<%)TEp5P&%*({ol|@5`+Fg4M@mG5GUw!?qqr<)E
zM^;4-L58X(F$<kj`iE54eOTX+BEVQdq&&)(&a9S9A6zl2e$Yvw2CB|9?q_8AmYIH|
z`Sj?*^FXnW@YwuC%BC++y?!#rCM%`m6t=20AkW!y$K9L+bgI;DkyT<4AF4RTJTIAI
z6<qCiCTclQfJ`y<>`P^LVhtRExU9r@{UDm(bf9!0-WL_%@dVguYI_@CD!2!u?w{YN
zzC%iXcK$Ati?n#!^H@nsT|!?qH>1>L=`_7RU6^|zHyDQP)_b;a(!H*47#Yz3Y+Svc
zO^aPfWZ@r`4mrB8alk1JdZA}c{)sLAg<(pNbD&y53R)`o?w@KrOH8_r`Mnaw&e<tb
z?T7iOAk%NL_HXsEilFlb0^uUz2$P$vsVv`heeqA8*JusAQ8QLZ$IYOoqr4x$sP|O2
zN)3A<U#mJ5269kUe6EM&p<<mnw1^JKnq$?5`au3^3gibHgQUCli-xJmA?uBRk;x;9
zUyYt6KqUj;m)qdpzHajFEG;oMI_86op^cz)oh6e$cnUs1>Q~WZ<E{7UAjMN81*g3b
zx#Bex&^+zJQ@}0g*^@WN<}20zqLzt<=B1H|O<qZ|rW&{9CFci>PJoY;&87Kj$l-A_
zfpKW2ITRPym02Nsd5IKSPJmrSe1P2t0XkY!u$-pNRFlIMCW0nXC-UXKX<o_7#ufp=
zoeG<MT&%>+^-aiD=Ua8^1wMyE;4jP%j}^3?`82iU%W|s2yBAP=LMtwk2ZE+sC_%mj
z@EkmIaDumNigB9!<ga9FQ8gN11wwgzko6v;)5WMRKSkeAZ(T;-;zDjAi1clrS9omq
z<a3l^f_h#hL2&HrkAFrte7j1+zvV#3y{NV4qT#HrDk`eWb~-#@l<K5@Q(2S!m=Q#W
zbp6jk*bMuT3dJQc!dtA>?ABf_)uAHh`G*iQy5vQDhHX)H$-DZRvNDVH&2MrjujY@{
z83@g~14e`nDW12u=7JyawZ$F8N~=WyzjJ9}?QqX&wVj@A{#Nw3RUbQ5ze(sa?ziG*
zY2~LTP`(NiNn(vlRc5A#g_b{XJ}HS}D4-+U0#XHhrMbOYQ-iH={$Pa$O`J^*PEYc#
zt-bPQPEnIrzbC|Z1s3M4T((^o4`{UymQLdKiusN0@%#RtoqGXNmwfoA|L+WEf8|H@
z`TYdxKY7b-F*@kDWZU+9>1mY35c|sJ)#lnZO~P*&MqV_yFbT3YS7b#?etfQV*b{E5
zoZbd_YekG}MQ+)+6_bhF_-}OY={p2>o;=a9wX~jFn!`fMe?A6X&edKrpmFeBY4+X4
z!@^y%uggL`2R9EOpRX)8)Hr?vXLb9|I!uH=*Bs$A0Q~LUNhgrD)<qBR8_fWnOmWwW
z{yn*f`6;Hql3s&7N8pe(i2GU%MvDNV(tCTVtgGy_)+QiVPl&)YAIF-K_30A^v}wFK
zTdo2rK?i+~UEMU`P7snAjHLmlcvfA0^lTOspWvugFu7LV!0v$|L}!BDG+810c~Www
zMN3M`X5uY$sWzG*>3J~Lk`TovB$Hut@`IYDv%RI+lT`Pqt`s#s(W1$9^wyVUD=HLj
zXRM&S3YVj0U_8W71xkLZ-i`NMSIO`-W|TM#pj8!Rk`<!1cznuEJhzNc50u5IR8d9l
zN_jTDC7(U=fsF;~?j%|*prZx7hAl8a_L=4wC@8_Qf<0=Lt3+|O)LjW#+la>qyOMGk
zhReAo-yO2mYhPfp0*Mw^g~Q0O23;7Y%Q*|ikWj2sX@#rZ+1_N(YCsMh0E+7O&m*xh
z5#B;w!;BIgwda|+1zQ_il*fBGt_C<(PykkLpyDazn_O@o=}G((?VlLj1i{@5%#2=D
z{-Bt6SMcn~)5~MVWIXD7tfnumf&}g0f?v~lsu8JQZx?P5Zwp`eI~q5%usdrJj;e5M
z<6YD<Gd|GO_!l{db}cI7V@?-(b3#HML=?yOvpl{Q7aIlK;${VD$uEjNKs;a|6mUI}
z+m2EA2s&hLCSFl4W^{n}o+mc&3sA37fkT#nu=7`GhiS^D^@Fe9)Xs;`$A_Ro1N7Z7
zu|#lzCI@~I5mAK9w}GL!SXcN&kTyk*Rwb32l}ji$ry7^>4kHuORFl^|7MjnW3R9$M
z>-ongpKIMr@?Kfl$j{770Qu|BiJaU55${88KGKe6lZxi+!O|1e7S<4J7A42c!HIva
zucNr%TBK;WlL^vx1{#_)%3||fG@o?3_gRDH@1hJr&zzc?3irB88{g&=eNqHU_XaD}
z^k;uBZ>}!Qg{(7&1_ygNu_@#5AiGfGV7CEZQQq?CGjirou>E5b8AG3Idk>j5hv`=k
z1cTrM@GnQf>HabB;yuw*AwfW4+Lta&K|%m&Z}3usQqPRy&#){%iHv{bUzvv9l;i9e
zPQei-NusX#R3dW+(B2i5yI7v&!I`AsENv2*9%whQ%DWob0P8h)Ji!P|@B)L3>w0S}
z2q4G=Jf16B7QD}Pef;Wl(s|JLRR*kJx}wf3VAe=fN)vQ_aT?fj8gN74js^0o+$%Pq
zUfbF_HLds%xBviofu<lLUK*4#voNHSB+-If<KtUOghiKMR!SEK@tVHa2bMoMDY;iE
z>^#u*?8L1}RZd1C9!4$THWiiRvmP_R6onmM;;Sd|0J`h{JqiA@&qjtpJgm=lBr)=i
z08nl|)>JQ2;xQ;wgcs6;&^2fZ6|40WfG@VVcxG@0>}MlC#BB$%F1RfyqB|-0=&0dX
zFU|Wp{MF7+KjM2S7_TmpNkQJIqobpndN<(2C1c~!uje{1MBO(%Un-D>J|fP}v{N5`
zNR1!cz7U7|u*n?mcKCyp?c7RKTK)wOlZUYSb{gEvf27Vp=@J|i1Z~}E3fXy|nVcQ!
z)oDu#DtyB#V;+@o)*zr1Dp&~*YzhTL362UNE}Ca3J$0k`U#PZA-{|>tS@Q1B-d;#p
z7)6)n8Qm?c|1oA=34*pnPpsw2do$JPIq~Gt?ieHPbda2Lkn(*Jl8}&)Y2C3=OK0~j
z6%BEGi%(L5XcqWu+W`)^DPC8<>{<1PQAhvI%nGHih#s|aak-a_SD5y1@-TbV5MALn
zEzHzdYd1I!p3goUR5h4C05r0_zde7p#QXNY531Spa;g%^#<vBb*?43_AH!eL^SmcN
zyK3u8+7=B>o2;sCP)P8VQTOQs?Bz?B1u_3mi$?@eW#EDZqZc$lrnR{6r1>1}+ff=M
z-446Ls@V~9UD|$sdAV_piOm)5h1Ys+&i+52OGS?-^%6J#eJ&jzGXSeXkGpdwW|V<~
z5D?6PQo7d9xz6>3^YLRU(PePe_9aYs=dXF+^l#mZ8hris-dx!qc|c2{4hy=+6W-oH
zWhckE@!#jWx4&zJS?RA$Q73MWf4`X*ngr)~`#0bOmz}ZqxPVt(QN0VbcwK$HM&${-
zLxuFJWJe|CKL2Q0d$-M{h1p+Ig+<vG7B+3zTVIc_D0*ak-~RXcGmAVu{qpZ~92_Y}
z8`HXSbsOzkt7%eIZ>4wCjuMrz$tS*Ios)*_FUMnVM`phN_jCW>KaMX9ytx0n)VFV6
z;NMLifc8m4Q{gY=g{YrE!^=dcry=bU^4Bl@Cyd<F4KbAx_T_QQ`hA^aaK@!4usmSo
zFicr0p{kyhc`i-luOQ`}Oct@N+qBei^2}rx8^f?-d?;6mkdW|v8g<{2k@2A{F$i^J
zWL~9YpOkEueXggY6)ykRGE0jSM^$M*-Zm?8<BHTJ9zr5@_}mstXPjPLhC*-Nr%$Fo
z@IKs?3;#JQDf3JPv~_AAZnf^6eb|i$_HX$tCuIuT&X!N7U)CTk3+tlcbAQ-cw0c>~
zl^}dAkRG|ZL(qOxt+loPgq|2R_+W@CMhwF=tMz}bgb&2PJ2P6OU$N3NbJ-Nug(jQz
z1x9GHrJdICCItk12eOt4SN&LvHwi+Q;XlaUQ=agUD{Hwf#gaSY95{Lp$PE4bxxTz+
zX=~QaoOD()pEy+T`SUB@KvaaZk78?oJuxmSHZ{5C)p+KPB?V)9m0~V>-c48>H@9<i
zw305;YX*&1ylTBRL<6B<qwanH&X!RYXkJy*gmhGI2X@h3@x#_>$-AB5<iyB~LqJ5-
z^P}s8m^TTkeU2;=S8V|@Vr4b8Fy;KBn3#OTA<AL0FhQJd+tt&jtk|zzVc)4rdff46
zA1@?CnZMF@#d}(2zbs?5>wQDccO$RV_aN&Nq~^C8*?ZK6%D{CcFlTGjcySH18$6NO
zF2+h8?(PlKg?GdhU0CDEZxhZ;E7zo%|E;t3SIE8|*mND?islHBB64YIIXrm!SYwsz
zQ^egAFvkRvtzNQ_<Gde*5g59chK?b__39~a5=41}#2+b5Hi5wf3=42vLn;=94*4az
zn@fv){|`y}DzgrdGnzNcb9mwN;iVt+Z1z*dw2=Pa`yK_l#zD$>b$&E>KQMy)%p$*Y
z&d%E=7an|38@64)bP`u<Q_k-lTi4}%4wwV65Namg*6|TfNaep$A%EY!kB*dj?tH8~
zDD1Ck_E>^$oxN$^B-^Ff1*6~g*|gq$KbK~ED0Gqqs(CTxJJ0aaPdzMUo`pw;HpB&-
z_LmPfz7NTEa(Z9I-eh;mKzUX0lN4O3M~QbJs0jk`va|DNcFHLAY%^cZx0q+W_S}k%
zirwF-9Hi!-N*A8KNT+3>g|;})vZm>yaq}Cv@!g)g$8})bg<lzt9=vpZx3lZW7dt86
zervehL~q{2Mfcgy4d1om-rBkg!Y(~-?>^FAy{Yu0pqRM05KM1WWFe==v|_x)SjpSx
z$EpjStr#*eGRVn2;C{UL$n;HdWQIWtZ&I>mAIg`+P)%QVS4=|>-2e9gfRS~&$%_E_
z@c41+iz!HJyP9gkf&Ez8<qV?F;n@Jw54^zlH?3SBi9^U$8%<^xGn2o+zjcl_o-wAj
zy4GFdTM?Hew{E2ho$ZX-(8rJQQ`n5ZJp1x{cXJ94?vCa0@emB`tRlcUTMcai<^Yg8
zSBb2oIOm;Le>a<wYkK(Qv)jUXHAcWidfmc*{Niy=N$x}X$+1t5)t*xj5Kqp`h`N-X
zrz8%6(*-4?F8cvB!;5Fn-1ffyDVb^l%@|JLdT#=6qBu2sqSDyd%v{tU@|%h%Bqib*
zi|&AJuXc-1WOeo7c=eFBwzlUC8l)~LIy`~w7qEMdL**i*Am6;9t0~)U`kANX4RPo<
zEDYR!Himh70fi8$P19a{0)iVpQ;iY%7FvQ6WWb0c_~hvt+RqG4IRmaVJHUPP%kL9B
zjI=a^f0ir}H44sOZu9(=u`0RXhJrS%-*yHsxhJ}QhSlt`JuKM8yNN<!(>kTaUb@~0
zb5(D9{E_OJYlTZeZRGBc)6?&3Pv}R6hXZ7!!RFNu%@NeI?BwLh@Ki!mCm5-7wN;1J
zG`EiM^{;*(9)@di<vwU~PftwE%K6M`8(t(HL0r=Okpl*i&-F4I8pETeAxZ1>PHrb%
zzP-yV`Y)*nNfO~FXRy0mQs}6wN2M9EU}NCsKF;?=@9t0Fbsa!99k=zo;OzY8Nqqi9
z2M=+u%2kvLf^`54RSCBvfWnn=m+7DbUiCehfYmRxi;}8mFCLSmC!~Cc!$m*^&{s@}
zP}4K}+GpAGrwkwT!28otQ1|Imlh1!$?pRnjW_TNzD_Tw#-gz=PAq0OhNWt(=#O%lD
zMt9;Ffu0(!+hHv>Qx6$f-s~R;?5)^14q3d)yH``?_h~V6o^yfY{@=fjpDka(q_+_L
zLCcEwCwH0)fT)9In4LRAP#8Kfd={{^z8)SH+}+pPzF>Dh;7)iAr|eN%L(K5`yO)4h
z5(cl%ZAG-V7Ty_ZJ|jOns|>?)y}rM4DN=@rzv&A+=t+UQWk4)gS`uCdtsiYWjo;Q)
zm@zSWX=m3|RtEUV#%XGN=zJHq46+2-UA{<vKiM~qjw}g$xknLE9DjItssP34n6m@)
z*Xy$NZYl^P%7=24=AF<G5Vyj}K*aR)xt#gLV(_V(;6)l)bmzeyPF~^XI|6N`{qMV$
z?16rW2n)|uV}QA|!c;}-QMhE=eJz%_bLu*PpZxb%93SVsZJk>vg6X~kL6UpD(&Gry
z6MBp}M>L-1Xse1x8@sy)mU!ElOA{fDhX=&)CMG0&!EVdDj(s;aG6wPw4jij;RjZck
z-a-_V+wm~~lWR}ta7uqVo4&F6*>D7ZbR`n(wJ>O@EJ1s$<L++mq5%$yU};V+4o(hA
zWm^1WK9}vizX`vjS?wM@l4j6YUlEf2_7(j`Al`>)Yjdmli(_8cp%@dpmzI~8?sI(C
zPB=J3yGIxd7YM#5P?9Y7m&Hcnudv-ns_D>908a#zCO|R>Ym!**Y!;FMQKLg=$_|R5
z33+*czrL;))NWLUOL@;Q3Ub&aB_-kN8_B%?4L!RSnD2vF*-G$9NYX@|K7^0r)#XKY
z8l#~6gl>ta#Mzru!l2b>^6wg|4Em8!n1ZK7(mfy!6)3^dFRysfMMEUa*k3M0|Kuv?
ze^LAodHxf(+SmJ|tF!Bp{na8nEE<~5R3xDSo?V{|18%G`-0)r7rz>sD1ov8UGMP@o
zoqaNvlDs`^j2o=ZX{pxfW_<pVI1yn`nL>jf1UCrm^O15b=LP*M2E1hd&6ziEMt+Ei
z{6KL7;c@zlP}8so+kY(;Jp<w;lh2+vIW;cP^J>{dm{MQu`FTg0F9<}2gXX|OYC|d7
zfxDF|YC?Fm4Nt(daHAFrufM`mip+sYnf>ELuH@e0q5?cz_sMBxbQo?Xq|ThLiEGP@
z7GKB;P=p86zXD4`w<aY>rMdWcuKB!cd&!4q3t?7SPIul@j!()sES;wEz5OWp?p+zC
zyl108qsK+bOe5Qa2l-jqcMf?^mzX2W@NAhGnIP>|%;OY@JOOHkLhx0Mjx13TbpZ^~
zmj)XK{J|!L1W`ab8X`vv`G48ir+4SzZNl+7m<a1zdq=uTE-I49Vs2sqRy)>nCoTqV
z$wyl1&z^l}qyE}HdDYou`R?62s6^Qhzi_3(26VyS-|zgIq@Me!sJEkzf&wQO^U>T8
zse_eu<a;_ACOQ@x?H4a%u0B^~mUsQF0#OclL2M{z5Cbc2ta0-^?g^-j-4i9bIWh)9
zNKxB&h4kS<P6w(KxSHx(O>c&9Z$#(w8`~Mk)ta+a+@-p$M7N$<)P#zdDy8>S45Fd=
zL#yHF=m<RKfHaYW5~nAe&;PN$ENKH{g$|VdnX${<BqyUFSzDfSzZx}3vskcnIlB&a
z2q<0U<>BGs<>gA~X`BCGgno4G-vI+btoC(N>h}X4&}P>*AJ*F~GhnH?Ew4_DOzX}L
z;8PsNshHW-P3`S<Ed(Bchg0eN&k*VL>z%;U+g_6rX}Y+ZJ|$7`O!W8^T~p&3iuioP
zQT|#Y!VZ0q5S>tcco?Xv@A0GizwSF&Kf1?upb??P#UoJF66hwlFuK!@$nkSEKkd=6
zp|V8N-X(iB{e}xVb`2#B4H=nrWrJ1hT+Jif3;)7IdB%KLbKt?#Yetd7l&AeEZp97v
z$B)<Fr@rRFH{bgbRb0ged_ZibO37hX-0EeRC=!N}lE$z+gN;?wnT;tHs`qdrUH%nU
zF6>R*`Ww#so`U1%bZIJ4*J!FYm@&G$@#ag%;d=>K-<YX)(^FGp7v&fq>S!s<<Yc`U
z+-VGYt<oU#F;cv<yC-aqwVz$?u3;Nxsh`$+Z>ZY+(n;r&5N4L@Iy>kdvq43oJ3|y*
zRt^r*e31N4l~&#1VvHJ}lRn!!I1~id%vw#Zd2e6(Mg;SocxuZ%{{-`G)5zrbNK-+9
zJIpP^l|3Tp5eu*RQ9Y+&h8at1#>KOxDR~bs=b@F7px}SfiTF6pTKAK!ndWh?ZI;X2
zH+!2)dB4!q38M0Txj@*;OfrS8o|xM@13hzAaSj>9;pkG;SZIectYoO~+F*Qw2}n_~
zTg-z6@d1%+4Fqiqr(mMB)phepw#|eD2P-JbpIrBAw0q<3vJz7M;U$E%B?t1=H+Y`=
zf;y_Ql88(MH_&1G=ikU(xu?V?P`A+HgU>eWBkNoebscT>=J%vtq2#0_^v=#3{xUb&
z_4cInnsZCU5N;mcd#Q;=mcCZpaJ(m9{+b+}`IuQaJUX+zwjeg*)>bCvenQ>{$lgKz
zDD2VGP*1SJo0PL0r#NN<`yEuSVco=BxuA^SWo6<;LBN{{R5^;cmW7-xc`vi`&Oi_Y
zxLfD!z?8(ABHDG;2&u{^=>BvMT1twGvWB3mC#Kf#(@KQRK529LXSa*EX(A@4MZ0z&
zKQt*t!C^<v^YS(_%*SmrRh^q(_XwOZb=f1m?Pd%0u|G`8^J*O@i&3W?PgkSb!TUNb
z?16h#XSsa9h<fLW4ZvX82#UC*wETiB;5#dMMz}Z~26CRPTmeM=>@meV&@`X>M$hJ^
z3!c(njk<)~qh}tH<ydt*g|i{@PLMS1=?Qv6<6)!sKHKZpWtCxt&`PVVS_*__#?T!{
zK@oQTsgrTfZ}+eqZx~&~=|bo%ypq7<1J4Ddxav0=up*!^E*h`&RmsIuzfFNp$a?Rx
zM!gXHN(<FZAW999y5a0J*mQACDNN?d>GG?%m68}~Vezg14=nJ32a@P?`7d*aESj(=
zJzvScP=1>!SGnw;;gr7|F7^`5B5dGEsdrnTR8#!J|G!#*=+4DiXx>rxso5whSP`4`
z(+kMs)()E8U?x>T@$pt;2>o`=2Of-y@l9ym6A-O)KmCOQL4y0+QjWtElWQn5#j<#t
zYeU*P?m}IS#|N!AR5hecv*+OGv7K_2ri)G&sde4@xmG}`I%KZvOeK09F?MAv$*W}N
z!u{*4-{K~K95-WLL!kZ_vDG9td4SY9{q^vvv4kHjD%?m$AwL@OV!BB=Os2`jqoa4N
zv;5mW3iZSL%y8=KDu3&oyWA=uh4eT`ncSYcxPd5iGr?IMM#(d^w$>ErT%0EO=F8M@
zUUl)MrS5OLi^ehK+h!+=GZ6X`6({*2U3hP8WPFm3ddH|$V*@i{ThHC?DgXususkzp
zbR8*E`*x-XTQnvQMO()A!8e2>t@Bm-kvi$ScZ+O#^F5n*usC{5=|y%f!ctwhiFZy8
zO=myVG^Z=QMp8FOS0=|fZ{NK7GT`-%`u88$Cuug94;D`rCJ}rx%BN3tu@T5<f_r;K
zg|hz5+QASyAFL^Gn!;n?LRjtWc;l>o`h0@|@q-!B*%?(_+bsvmE2G4mXB%F34!bBR
zT3Q-M$lj((xx1~EK!*C|ZPVmnDFY)e3X*IwYRoBckFl<(=`I2->5#Pmlkc%Cp_VEI
z>F=$2t3eMW!PYF4Oqx&ejm=E*7d0@&I>+aN$o_emqXfPG-tW%+8HLs5IU{a2w8w2r
zil=;R*Tkisfa{ZON6hPTVn`DT3jc<bd}p}#fP(9lU-FO<f=?ag7K)BQ#%s9h_BEe`
zU-UR&xnQ4IT;6{2Slb3DneFqh?cZ9(Rr&bXs2+Y8B$k&`@;cmomeA7BW^kf{^?;Mk
zhPur=ygrEu(a;>T<4-|XP0L~i8E+B4F+1CcineL>jwQXAvl*P20KgpdSiv2$!{M)>
z+v0Z8ZH70MLWzxyd2)hIN<^~115$lR1cOq^S<ltT<fEwWCCFK^d<D9bS8h4sam#6`
zI@>d~+^#=euwuywDT8sP-}2U!=+-)1%v<Kt57DV^uCLE~>j8`cZzeS1na;Vi9ZT6c
zJv~X#K}4hpf+A56o(XuoxH|Bvs;b27?a+{GD6_*;_6J`mu^pn{watS~4VI(2is;8r
zbkgIZhzSp&A*E~RZzA9cp<7!MLV}b6{3t$fRQuF0b2iosyW{`fT0c0TTxWj3bDvfi
zvd~7>)}|-NSgrJ0KIhdAcXv=h-^M~xQWDSbQ+**Z*RAtUA3tK=`0~MjzmGx;0|}7&
z8`ZbF{&d#+ps=XPsI$$rn2pk{(J?Y62IGCueD?Mw8T!%IRFmfrY4>{w;KufzHekPf
z_QpD!>vj|0U9m>nrJvXT70W5ne*gN%$bCoci4G~j>dEm84-yotD9y)1wF@2>NUEt7
z5Z!b1q<io{>&4>$itXb_I6-qhH@-TVZ{TBns9*2XYkGEy{!uxCf&#=tn=5^9KqoTS
z`Va->EgX>5m6g}N^fpAm`2W#Us90b0?0j(R9+q#(bp7lKS__1tWtHSc@6Q=M>qB=v
zkr0lb9IZZth<=f~MY%_|Qw^?(a<l<0?`^ED7iP&+Ii~bQJ>gQ6l_l1!DI1u-u}S;*
zHZ3>Q7j8_4Cs|NX>C7gEP!k6ITfrUA*8PLTRly;DBRaSo!?TBth=7EWZ}hygX`=&g
z3l(w1K#s4Y4Xqu#HtL0IT<5U$5_pu760`7PIE{Iq6iuh3xzq}S)<DTCo#bxv?sV3{
z^#2{v!%)iX*CZvt@Q3C5^l2=Lk7TyfV}@1h+K*+O-D6^$g3lyhguzTG`rV7_hiJJ~
zln5|1Tz7n%8Gf4U0@wkNu#2$QPc&xawx!|e_x6*MtjQP67WZdzVFYDy;vt4NHw7+V
z9epu6@^~h<V%!RHet!AE(E+`1oG66Qs~+8Jnsxo7kI$FP>kbfqYkw=_r+K4M-E3OH
zwd`pE0s_%D<|Uxm{E&GqBno0gP!I@_6uY8k)At@$+&gnbV99`|8BUH%Vf(I;X&ou$
zzu@Wu%d7szOX?gNA2!ACuhjS$dd-)jw%4x1q%NldYlELNBf@T&@&Gq*Xlz90yFV3*
z?mTS2;G&Y3mnU0YJGe8$R&Lxn#YK;spU(gxBkBx9h`F`pP74<WE|Q~?7W5xY4R&e)
z3Ka=Mbt9uE>aQ6QNKtV;IqmpP58mRM>T0mcp(1n9gnjf;a#^4+&3a2a)!?u(h}!VS
z#lzf%=&|o;b@k;p<hPf_;y39>xyp$@mhIYRL9-@pd=u$O;@E2F;AIpb{+G21$2p3@
z=f7x(oP4i%Cd~sP5<)>QbE?N}D1w~aJUl{;cCX(!<#OH5lDT>7*sUwKC3bhC=Es;#
z`llETBtg`P@U1vCVQ@5^wnTrEPo^fFDp?7?G)nJHh#or3;sRkN@zarE)0Jg`x>~U}
zcGrADy4Ghz2^iL;LC7}!T7xOJ&j3nIQ4xEL50UtMJa<qL#{4yp)1PWjQ2q^g|DyZ(
zR8?)SMn`#>Vsr9~u@LgHBH`D@EuZ|P0X_HaMDwYpERrGdYp3s<hEpyn&DCzUk67`n
z6Z3uXr5b7zfbVG64ELm&XXqS#6%`fzWo6t?PRB<05F`Ze0X~MglApgnBCaJ*pryx8
zfg7?!VQcbMQZYv<?Tb)IfA%?duDk;53UK|Mot*ZI?@t(;7|PH@(LSJ63X{d#{oq-4
zpEW+z(eyjFB?54ljDwRcW(%=x%a>1!!=6GlIO@00Pr|yq=k*)x8KdNehX&(f@X-;h
zY_W^4bBIA3zoLYM1c9NU15o&@M{5>*mf!21$lk<p1cD2aqEL_vDmNknkd{CQuDp8d
z#>hVPtmcOggq6jt#>Nn4_Gj;JL{JC;De1;`&WhLZ@igjnB;dxW6|WEtr)kfRxgT7(
zpj3Zgqc3vR5%(9sgAcR0m>)WkftkV9mi*?B!CODR3*|G8-U^bYc`GiC_u6sjqiit5
zG=dltH?V@63k4yA9R$$zCa;r182D|o7P^?2P<+fd6+Mn9-O|j#@6%*ts;&=*0z7Dm
zN|d%HD9Fwbr<ZX|Gakz*ChC75BXM{LGQ}`lt|8DJH$&Dr)RBH9Sit_gxv6QP!3|kG
z;iRKux`zA<kNM)jPXX6-f(W1bBk}dO;*oK2>oq(FOLww$eXgor0@{lV{QNmTzwGZ7
z6%};@fnu?Gd=f%wW~l?3<CaiA+nT}Oy9oP7v#R}?+o?nd0|S#x0Ufu`k7Nj!UEl0%
zk4P8zMo-s4PdAHln`3JCJik3Xw}zs&bJenYAzTFup>0;|{LwWR{Ptz9x$ZO+$kI)4
z1<ZOVpZO>mycAi}ZRT#&|H4MScv^I3g)1wi96OpWB|(KDQiclK9WtPtn^k~^0N~I#
zO?0FCzAtwDcBY5|REgYCq9fd$kqV5rNi*iu(pAnbdo8YqoXdIvpeT0rd$rc@<PE6M
z7-78XjQ)<(I4<hw=y=V?Kvneqy?ghNjG?t3dqFXFQ_mgrY*JAmrRMeDc`!C4pNnom
zIz_%k;x?=5Njh~p_UY3H=lXO0{%COYerz)n1kPYiXSf*;!~o#Ia&r1RcsCV7Ji)aF
zCEDOWO5*hn^xCDvxrP#9yuH0HGp^%WZ1`g>Ct7l8C_Vs&k>ODu>2Gwvt-(d>uS~p#
zkdZ-x*XH`=YW{sFcGT3gKGRQLh<jfk@IvLebWCV6Jv7D#+0K@GooZN$dVX9h*4+F)
zd&I<P2?NP)jt~%l;F-4DbrFt|`#NVwOM{up7A1Y2&fF%XFL7@*Q9w+$i>=qsJP_8K
zi3vA7eRGvF?t7u3e`8C;nq7*tSV)qxTre9&<9CzuJ{LE6UUim!JY8RBqQb@+Tzmn6
zL{D{nt%hDSbG1D0zfTxJ??IM^@+CPn^8L+J0;;zAMy!`k(R~Ja1^Hkjok<>eLfC7M
zd<osker&Ot1A(evkGvi`J9^EE-*a^0ekAR;u}%()$47p1#N5&rjL^e@5hoBJB1HsS
zi8J0O#!t6jii8ZFaSV;0_zU1X%KK>X_Q>k=fB^*ooM_&+qYw5nV8ec|>FhMqer?%(
zHZVAji7Zn3lM)g6r3EgC-Tc3_!fpI?Zw;bbx1dQJ@}}T@JR5&|-a)hp&3ly8t#^Gp
zhwHWy3JT<4%X_8M(b5{5XUHDA1<1t+`>$lRu0N%gQg}>=3~M}16RV$IVW?y|#~cls
z$qMX15vM}MK{8$^zY;N=Ew)alGO@6EX=wW5scv*~{LU76WMb^X@!sF?BJzQ%)$J3<
z;88-c$CpwH3YnERtF4|J2^p=M88J)^5BHEufy>@k>W@Y6<-_)F(+apCkfE^ygQR`!
z$J~f3((cGV0u&c-MYFL0QQ3UKmOj1maQ9U1xUOw%c(loJA0nc++0uW=De5=B2_!Yy
zoi?}ljK|0Q<42Ete~*oYbt;e1`icUl-N5HC=dyv9Po<@APv4ir^lxeMB#iPS#OqHH
zUD`$sJPZ$SjJWO@Te=V!tM)h(^gHpp8VF2zza1Vm<5A$|7UyW|jH&5p))>rKvCWF(
zrSl;f`NgB|>7ufjn3x10*$#b_{WJ1X%}xH8XK@J~6BBEhl*rU%MNiau{`cSzLz4Ux
z-J-0D{lO=)yn*|BC#WV5zR9ZI)o*lMxXZwJ@AYE;H2A^0y4mCuR6MuV+}GBv7ajgo
zHA`PQ*Bez_>3zvs>l+Z?*~fSS$5~(Oz>}8|d|8?2RgE_oH56J_utgNixklWM4tqU$
z|0G7XpwdJ!u&_vHISrcY&eG0}F)%Via`!l-cJl7bU$`87lzMiE9>WnLR%^>Hc!9GM
zClUU(?Si<v#q&yD)A?x0fgXH5@7ib5>g-}`<RxFg^y_QJ-P3jQ#k8V|hX-Eu7~rJr
z^>R7GVjI7TkfQlus*?U8Hr0~bRKvyi=INjq!)AMj2e1Ss`l8qf@0(B|bb|$PFYn5<
z5>czs+0V=;Ee;d8$tv?Z7IUEyA4u7z!O^J6qFd^Eu?{b5|L(DUUe9`Ex6_wpm(L_&
zqGF2uKIwOQ`tT8E+4EfewthE1ga1C~$pkQO3bf%P0884}z0lSigTP7Ew1zNI8BZ$x
zv+ttPhNa~{@e%Qn$5{)6N0h>^A;R)<jrgx)6$-@9RhH)YeKSQz^Z&yO;Hx|>AymC`
za#o;ZG-ke<{dfy6z0n8d*fGQ8v4!jPiNtR&CA>Y{Jg+u!P>7Q;BEl?eEKqzBYQLh&
zs-mL>{#%BYo2BRhT!<kGR7&Tx_|xV(ux}cCP7ST^)zwuio-E=P=IAsxKBOpQeRO-k
z!b_gY!t`KcflAtpz4pa6fWqLwm)iaV`Hl!aAqlXmkB2zX6C##{SUISES>b92M0Fa&
zUoqJSANl>T=kx&s_q$$GFllH$8^OT9xc#2;-ouAb*>YV(9!rohCVGW6z1fcVn~8_A
zc;zpM5u_!|FUvo%)?h(0q;Bx<nJmSwrZ=&TdI;$Nl(n&Kge36npP{BAUwRyGLL*Z)
z0L<_vwrvqd*!g&5N^;~dTNWNxCC})B;0{RJy;yWCo!cyI5i#sbo^D_<2ip$NJ2NW}
zsHs6FnZQEQRorD-5k-G}w=uE4VPk#$V?1Fd+Am-aSV%%gM~ErNP(=7JYQ*BvRx~Mf
z-fTMCyNZ0_f|2cHm!qXVrw_vK+IsjQ`wdYGPbj%ZEzU{FMTlH`*A6*W%GSE`4MY$^
z$TROi1PRhlo&*Pl5R=~G;-M7r_%#6Yo`ghsn}I7?5mcQc$Upz(>P@WoK9o-LcKj=>
z`eH5zH{-i`$VBXi>SKMEmH=<|#?Eg}<3)^3SDrl`gT9I{kwJPPhG+;>Yy!rNM8@ma
z?hJI!(A`_jK5P^sO4WXG-f(z`iyOGGK>D?TCTNE1USPWfFKMOJ^RKxqw7{{TN2Fn5
z1tW+?(mGuAR24{&6Bp#dkL-R7S(DlY_jv%IU=U@X<`3&$;^pP(_(~_)Z}mM8IK}VZ
zb09>{n92f*(ygpA)$-t;fS8H-3%-}X)Ig_aN7;4vhKLFUWW>rPqH``Cc2BPIGrb=L
zkIKCBckQITSt<cOOpHh<%<cN{T~WiYxI8yPEU+Cgqp6FPLEW>6H%`vjSXhvZK!Hq&
zc)}Ic)I2(Au}R@~uXFyMk`Wm+LS8fLeSS_Z-O<`-O`a}V%lLq?Rp*@M&ktix$l1pA
z3|;Q=qL)AT6uzn+_DL_~ZH|9w&7k8A-#&vz%$q2P2D49p)?&157y|TD^Wgz1LKpqb
zfZf&8Uq%{I@q0F>AY5ZcJ#Rc?zz+7D!9gH5$SyVK(go^m#>-l2j+WTw#WrPqHRh5j
z_g8iFcl7(x1v#Y+cbD&;o&IGe$7qw3Y^b(P;<R`NAag;%S<kluj}JHV3-Z(W8}Tmg
z-ilxSz1lW;usQe8@eO3?$*QF7luUXhu=m-6-#SgycIGB^)y$CuY+o+ISuHoCF5bb{
z!T%!N@}OkN!ATO{E4ocC^%;4`FcY=t$N*~;rr+0`s>8SQYiD&&kcJ7yn(W7U(E7OK
zkLeKww?czIGm209(lv1ij{R!!24oH!r>+iGbt!CO<FEOrzyRmGC1$@dW8}=%0!i$g
ziH*)H&wv0?d)~Mf(S-Pr2*bRq=4+-(S<TIt$m{x2&Sf!4`CoQ7SZPswdJR#1f`WEi
zn}quG>Z?6>3?b^Z>+>rP4t0(APqXo%i9@+WL?r1Aju%iFpux2DqgCWqht{6hqUEeD
zkRyVCL;ynlsGILccXw>8=BrnEIOMMo1O_<Rl_6f%`DpKclpKhuVfqB<`MOW=!go=(
z_2HiruY_##3k!Q(m_vhp+^!U_ciehoCx?b)tn3X&lVZ3uf>73Sh5jGoZ1j?w>BRH&
z!&CA{e*VBM(_9k)z5>jOKooOU*e^fSen3mBp`&SQ^XgMkJ_-UPE#7-@Lk6dtw4}&U
zo4?8<j*t1f!CL0FaJaXpkaqh}aqKt3*Vi>aJSf)EE_0t+yqud4{Ae>}EBL<+wl@=F
zM@GQO4Ur=!4*gUYlrw`8i9Fr{+MT~9lq6=Y@HVH*tjFG-MWmmh`&U<258XNF-O$X+
z{yoNWEh%Xo<OL8aNW!Y$Z(2q=J}Kg0|JU;t?E^+bn^%oC`zmRKrlu=24`5Hc@_QYJ
zumk;Aqh8jZ^5Wv!yEVt}<=+mO<2@DFIA^*f6%@)#Yo!5X5?8I69h&ofn^t@=HBLF>
zg&kJ}(F+z<uPABh8?1*WC?A~nnL@IfX7SZzzvcK#cj%U}u#neCaf!RIzhjY)xxgZI
zKhuRO+usk3Ub!?RC`_yF5|2YY3PRIV$IO1*k(HFB(Cy8(tPTpnr~z>Ev=+o2Y}y8-
z@sS6t4@PVTndp4wKR$lg6PBxzhJjG1hzUY_y1MqWGNZ}wC}|J_0tZEQQT|7MX=pb%
zh;85tme$y08~a%-#_Rcm*=6pWiGS1DQjs~jQ@g=F|8t1Fw`cyMCrfEn84W&$(L<#n
zt_l!dL3-6WwSb83R>X-v^QetU#ds*mFIdk5W&cHyS-q&S|M9z3^<C=9L#-lyX(-|T
z2+8hIO{qZ`VpI&!a09JnKu0QSTEmb!GWyepk`D)n8(nX!#@8!@kirkCAxupFcp7zy
zdcB-P)sm`4d6%zt?H~(kzLbF50xJ{)0>+td#?4mR5SMho-QVc3RCd;H1U9u)qV+Ay
zmB96l_mlz@cdHD>>m_^_f%bv%CE(8b{vjK4Ns37eMMV3?y1AL@(#@|&T?68(K%ziL
z@6@A$AJ<^7`qKUl1*x{yi%;Bdo$h86<o(!`6|IuXR$a59Tf7mP(&UKqKwM%|!0sjz
z$C}gvU+s*A&wsbMLm<w`N~{B=nZ@$%ya_Oa70YZ3_J+e4?GCx&H`SJ!13rEFN^wRk
zO;Z*v$wBcs!qf-Sq5-JgW1UU$XSVaHNaNpL{rB`n;9$DD|6aXA7{xzis-X4A#76Hj
z-@!YTzubGj{;9>979i}2<8`U2M1rZ89m`HXUYi~rJ_PrU``YpyVsh1w1270cugwu|
zGW7T{voJ}l?V@8eYsR_FbPzIQ4{4S(0YB8MRZAGr-dmlzy+Y-2-jfZm5*=+ef0G9e
z0s;iYX2#WJ<?_~-IU3)lsFyLC9ugc@);ZEL9YD$U>DFQ)0KNv`jL2T^I~G=epuOa{
zw58vx^K<($(8%M`^CM<B3}^C<Sgn&u=Q~%4dvmWbcT6YQIu<aY!Wy(>6ujsNWU(ck
zi#}wb5fr(fEVGg6<fNP~h<PU%AEnh0VB<6AqpN`viRM|E3fBuJon+OSQr+oks=Ylb
z0sg0B!<VmuU0(>gH%3dYZOwk80+<qb7v)5YS)l-}M_Ldp?0-dH`_@Rzd9-Kk33x4F
zM3bkbQ4l)f5fl`Zyq^j4T2EQ(Wm!h)Q4a8DoLUC$NyHv0HRvi14vHXo^psjq73dI^
zQ<T3u(bE3a<z|#P1LH&Y-Ni>9FZY~ls@=~_m2s-98*tCD)K7lT%~SvC?ib(^s?i?;
zt-HBT4Ycm%s?X%luqX6@{*LOZCkO<je{|RfhQ^ybD-N!1us+mduc^z9ts&BC{@M&+
zb5NTB0=-gi9Nf-%A<ej4`}uR1#@OpuAGi@}2DzDP*XflOJU$96M2D})gfsD_2*I*Y
zwGFuV8Sb_Hm<!luyW2J^!^g|h-6f}b+bq~PGCKC;r7r{5>Ktl=?0br5&-5(JGeG3I
z3$JY9TOWG4Hw*(#L1lh->+VplN|J~XSx6orm$OZtT!}0`Zy>Z`rpeP`2ZQLWt^yKi
z%FD&<-ZC@ZvoX&Y+xe1|OeQZEl=bw9EyRC9x?7HNB2oHo_%~Cw6frZP1{r$vp2?if
zr7}ki{()#47vUM3xZbaSp8eR-*3ekIEb3krs%uKMw)VUL)t#Z#Uz9Iy@1E{N;`pht
zU7duQg+J<`i_?7eL5|i{K;WQUf&?GPt@DP(;jlsqVM^PKT#$_llrvl86$lDV;bGw3
zC&b)e`L#?+0<Ngf3`Z5Ly7hLKE87}*dAmTheK(T1cSM0Cn3)NUwkbWyqN);bn2(H)
z@ODt#2r`s{^04nZi7krqQk|lwCi|yPi?+7@$sc%zUHs?3COEB|CL{fQba)sG8(TGv
zTkq-8(5W#ziqnpQSO<uYb@CWAV*w6F^LBT)j~HIJ+r`^|7Fd8(U)xw)BqhP&8Gws<
znn}24gO1I$002BKIF{HGUcSr(IjGAS2#KXon|>XC8t2NznIa{(&`5u^0dl@ED)QXt
z5fQ2`9R}>&+{1L?RNVz`Hc$-B#5A#UcY>Vz-tHx%9339lu<Vn!rI7>E<nb^l*HUax
z32*^t5J0wyvlTmZdi=5JRv`<x4}+!Or4$w>tj^|4!qfik`-8YR&B&mya7wDxCc^*k
zJE;i=!;;6b95sHJOaw4f{6AIUzyKK3O~i61c!4D*HsIz~-3|!)!0`n?Wz+d}Uwd0m
zLiZ9)#~mptUm$EkLol+0rK`DeK&9r9*P45)ESaah=PSnhiVySnd@CJrc*Vw!fR1RP
zum}1N&$cYL3-3TeF=&bqA3&~d1xqgy0+el|E3h-SA0YWLDKQM<O~0-Zv~xMw58f`;
zsxv-j_wCty!4yl!!eY#68Z)Qnd3krvt`-_UU<%DI&L@XBRB&B6TD?<Jy}o^Ye;L*!
z;ybrwvwxaSC?$$GkR=T*_oclRzD}oH`dIVHgXqqj?^)lzN}GJjhP(N4deSOKB}7%h
zlrsTNtA!#KcU3wa-a<+atJRKkN$9V5@Ze+Nr=YLDTHKT2Df0VjOp@I6W5#CrpwnFC
z76F0R<#Bj~K6O#SKWSsCN`G(n82Oi{pne0AG&DpWGPJ0%BS$pEdd2p97hj}_xsVg<
zJbNZVLR4AKepk-&O@-+}GYWyH%>w|*M<&GeYmSCL!F&5p=QJ#yeJUZ5k<|TfDr`nL
zncmaUA2EiR!#dMMkAL}Gwe;uENVuvJSZu13O?=@E6S~Eip8Zv(-QDF%#Z$NFG=L6O
zBsF-UV?wDaT+t;k=b3<j){5ED{j7O~?k<p*KI~C;arT4ks^t9zn0g>_R$oImYw*6a
z?Q9r#h5i7J9L<Hbd;*UVnZkik&%p1#_lj_`qh0Sjd{g1)hxRl!HbLoMTT=d?)e_H^
zxvbl?gliK^T%C+Cu8qSj{k8O?DV<TI;QKo~F9Ew)mB2&b5s-c)mPohKo5;v`FD+i5
zUrxmoW;}Sm&H5Nfa3)IAK#+`}knk=ZR_g-h{&EisD<edYOPsVlp#7RIR)3c`Sxrq%
znvgm%@%H=MyfJ+AW1}PRBch6ZE7kOv^ldW#>-otVTlDgc&p6&0XvaY!Le)Jr6}^2Y
zYEN|`BpfRAW*}bHU|c(<K~7FhO;#uR+e_fJq-Spp@KFf*LO_sBMlQ5nV}~lb)j+Et
z6D!>__2;rgo&T!^(3<`^F>&S$<{l39!Ouw(IXS9NYvHQ+E*k3#Hz&uOUoKC}LR{3(
zpY&#rsTR>Y!K?Z5RAU1sfHzM6KKEsn%>vKBa{R<}0un+xyYZNyeI|8WPUV@*>mW%*
zOPxn1#8<5ZghVls__JkB?J$-vn6Z9j85hon2Ila%8(M0mg@u`_dY15Q+3E=#(XlVi
z$LchBladf(Vei7LOTWy<a%UDKgWNnLi8oEX3<)?{ySni*2hE%AUjUB*i8o|q$!KVD
z!e<Qrl8?q?5pe~D4=mIJ1HWK62EB@gw%(wSVj?6)1xqtAu*AiEAjGEzig=IXv0j~B
z*VHfrr(khmqkU&voixqsxe%PcB3gJY0rg2CF`sCQ3JNkxZ`Wl;%1LY8NrbWrKAsv1
z5hoO6Vq(&@PZ=V)G<8CyX-I&2W{rdy6S@!wetjy=VPc|5NsgbLEO;cV^F%}XQ&vSq
z%}+0aYj5#yEo*Ad`gp7$8P&!2Xv1ce?=doRxX#zY8I<1OA@WcpwAU2ghNuWE>m#xT
zoQaTHEc^&SP`JJX7=9sLOwcq8AQN+%0xjWR$4W3wdKq%CF$yApt3WjQhoG;Je_ekW
zK!l2FL=7>!XfAc<E<#tBZem)Xn)n|R4Z)4?wg%I2x>WnvaM8`6No^-?2|Gix5cIcA
z=&I17sP8PhxZwLo8FC*|_#M{t-D5e}n}1|m)_^Le#^yzZRoTT(-1RK)Ywz0nuACNH
zx0u^cE8-644ln3IW<5CTohmkH%R+hXQuqXPa{A&JfZKT3gR4qeR*{R3ZPshM@FDrL
zta4Wvnb5=2SQ#He6XmX>=ixoJbjj>{T*)Qg151|9h(YOjMoV8hh?t?U*+qE!n<8Hz
zDJzH|X*}i*205~1GGZ3`N`$w?+-@LorgXUxRA@dx;C8&{LHGHq<Gk|cp~m6)p!I=C
z&-ffAf`%6OH(UxNY)wX!GBj@oCPtefE1>!imnO>Pd|*cYRv~~ZTH?QT71mpg%<kxy
z+_SYXbX#9dU)>q%5UqP#INitv4GyZBXXp9%&DZ)~mI0VI<GH%)nbl;!);7W%A9)>i
z&8n)cujAdIu{$c(2M30L!yhu+8k5j1u=k6(eisHkc^W$ac6OV=gCdRI@&1W_z#vj%
z_2NDaOOF9N-(^{DX&xG)LOCw>&FvacGwy86_lt_MIbNSmm89~cMC1;d!~2N7PjA`2
z4y>wA3h5be@AP!Q#z}4Wr$;?)t^QNx=9fV(*p(-ek|K6=Q4u&WVIxEA4u-E2ky{Cy
zE(W)s!8s)Gg7v*jgUb?02sNn3K10M=e8d%~3XtDYdY6D+j9XQM-yqw8-o<nQ=0V8N
z+24GU?)wT2`Uk|E-FUuWzc%Q9^T|c_ata4&RML=?d?)9E{OIYMuHTPVn5y&2uP(O9
zdrKKC=jiC#^y=g#YoJk!qxBPqQ20v^qPnX|JJb8yAI|eECUDu;&UI#a)9gsI#&0Z<
z?<#i^{O5vVXLs??XYKd)fmg^@WA>`U^2ZAy4%FVGsJOK3-2G1V>!=GY*W<n)v`D^^
z84a2Xn0=LTBI>0Q76Au(Z7n}iS>+6A3ZTCeXeo7>*n9+A?tOw_P{#VFrE~(lZD^ad
z84p5O+XT?M3Exzu7zejOxzkE|wAJ$IDSC-cmDQr16jGY_s;Ho7>AcIRjk^KFvhb1u
z$=tPnYD!3sK=Xk?=n*q7RI+@m71Ixfd+(#4cmbacWXVgVd&q&X*XCEf5U;4b!sC}|
zsSgoQ*V~G`23mYferSbK&};Oo`N2AI=t*UIWJ06)w}4;Av<ScJHs^=Wm6a5M?U0kK
z0Y0#ZX>_wbouugG8plBoX6%Zc+lx`{IDAF=Pqm+knze>F>AVRJ2}u!g4U(gci+P%{
zIu1jsIvp4BgKy;_SiFJJ2{GmHi_`a~oN<49S%G|c^L<$C4VA~z_CsoM?J6q*WOeP8
zYFU~ZRQTF<Cc)PG<7t@co7=pB4zO&&(fzva1)M+3iWTAED#jZpWH6#Bz3-UvKA*Dx
z*#p0O*?3ABNe-f>@uJ+M!wdP*%_g07#0tc_AKLZ)bWsqiudqB9@EmO`-|M=M{^mZq
z3vcKn5_~|ms`ulwKjlE&wn{Q50b<&pQe0Btxq0xYTb>J?F7&hv<}ZblBINR_V6N+@
zoZdUKKLK}M+STG%t=qmH$U2c)$G@YQ#N=-_MZ&yxwsY9D%jWi<Lzz@oR(74$UsJ+h
zV>sygngRjkNl?EJ^}GFP!b)cBu;A5{>AcVxvpe4vjghn6`kYNTKkq<VM8@YDm(^_v
z{xK9ms+a!*<L@Sbup0`amX^>jY$Td*YDG4hF8$<Tj+<)vOUj0Uy`VE*ZLtL-ZD|$j
zcW-QmTUoJ+q-3jCYFJER-eFBb)!<BDPhZ=Ry)fr3(yy`~r?IeLEcf}cV5S1V9}xX$
zKIMB@;FN+46?F??(-?TE5M}=SHB50E<PvkXS@CRIU=XSQKUBR3IF|9-K7JDkMVTRc
zW)>oQkL(!|%HGP}BS{jnH`yy>uTaU}J3AwL@9lT>{l4$}KmHw#qa!`f<93hF=eo}G
zyw0<t!mBz>%t*Zq+Vkp?MG$Br4@Q1+)3UTSohiz?X6Ba;Ky!6)>lUmJ{(mtZar5+p
z@6tn5AdU~WV!;ZPz*$32Z$g0b*4#90U`Iab)n^qLa?wbzeG^R&LZ6C{Ri%ip5>_+7
zF8%7Q#TCkHtLd69X$teVZYc`p?K_rVZ*R{wF;45aA$%!DD|3^Jmn&3J_H-nzvXuK8
zGQ%qd({%6!(gx6mfzu*5*Fc`cbzh^vLpeFQuB19iDV@CB>bH!VP$i^h|7PVu1YO)A
zF<K!@jhG`Y_{~JPovezitCyF5tljr)rvgdIw-^zvodc5#zrrsb3tsLA6r8UWq?J$e
zs51gcyFU31VBm}yAe=v3qz$gr7v}}-n7^b>`}Per%Hj(sd#lra<lbq5OitDF=&}M2
zE33rbM5_6R8F2riT@TFo=!?D78y*n_i<e%V8>7#xl#~>TD0ojnrlxe3PF%dXm1UPf
zZA@<v)BvALNkbI7>$Cp8dJLz&tJj_p^pfdwY6Q5^1}blhZouFJ??W2w8w9aYA$M(W
zAoTPmr?ulvfw#P3BgUr`iEmTcOjSev!|(*JZ`kI`|K3}kA#M()m$r9~p>NcNVZ5!F
zIT73|h1$D!uasQxP~W}6yli({l(FGHI0(6iBlGgBr>M$ZEJ86%l@^_Zy?ZzX`Boqr
zW+|{je5*V+mNOAws`!rs_n!bOJyrd#N~Gf+kL{Ne&=TU`cpFj58HE|t>Z09oa&KUA
z5(~+_M@$jspCY&38T*@-15ZP0UEoth1RA3MSy5T)2*u>nsiFF}4@b#DiqSFA@*wRY
zU2d0^cT&rRm5rI@7A9uc!XN=NLE!3aT3O5NPM<nfNu(S)0tM0J0QhiB+U*3~ADIuM
z93TBX<hlJj9)<}n1zH*CguHg_8<ICD6Gu@*Aw4Xh(bgDRxp8#y)wG;~DP1ADUsp=J
z0r^snhq?HO>&f+-B<U}>y4`>GvHhbtTvsOnS1umk?dW9DvYS*&Qvp+`9oO@w39Xy^
zP5`RW<TbSL-lqb#BTE5>e?aPACh}KF+|u;1M%WUqI&kJg^DW7N);JUHbJd^;0nLfy
z^SKoVp9|N&tp788@t?mw`ml*uX6(4(!<@=y#a^U^IwSD9Fv($UVS$b?Y8?qjH>vX9
z#S40Sd7qXI9nl~ytf(MFU<i(=@ONL%4jb7BV8^WbHEE^!roJam$KTCVFKS5Av$8ZH
zu)WvM$aCcI8j733c6xAl^Q=Q_jGcuX+68{$0t|B1t%2P-R7@~72Gt;!O?E{40*ppB
zNbLMA{oB?g`}7$pf+m~!<Jzm6DcnC+kJG$Bw_kTi`8TWWZ&(vHacgXG1@{o#=U?<H
z9G57`Ms21KhRbX%(0!|axqvM1WDW<F4sTzL5do*%f6)(nfAWed7mr`Z$P;x<eKl>~
z;fie-34r*Wx!VB)rC7+HSC7q;f-+NeVkkm*_lJ6oN<;#}((VL*3~h8gc$z8!c?vk>
z0xU3V`Z))%tixZgW1y3o_VyD|W&TG_Tya~F=&X>8fo@Up<a+s~C91?vE2}k_XCAe?
zn4(;&$gwkHY__eu#r)#j0r2R<JzOGUVm%Nj9#x2@seqIl3VxOt_H)~ZySv!<c#)5|
zhfWJkE)9N6aD2_ma?0+G^85Dm;)Vjl;9#o=ElXgtAtp9<ZwlYWOr+@O*l@Cv8Ea`3
z{)SuFMWc+gsk5bfj9)`hO<sRk)gc)sP6+tEm8F<xW#{j*jJv$v15IdUs(_WRNtib+
zjE=!73>Hjg3W?_xUNlB?@^T%lYzq61*Pp51%)kYCTw&n`013Nt#peWl_zpd+WIFMw
zSHI4O&)<4-6EGjF1jkCEzT66BgPFP18~>t4BPr<^MPxTm|CWKMeLi|IA~L$9B0m3t
zw)eTV1I*Hr$NnCf4iAqngzq#Yv&Adu*P36#h=5895$GZiz_0;rcu2v8<$jKXAes(C
z!y}}uBC`74x8({uQV&CwkOy3kAi;NTc=?o+iDe`Epa7a?>6^Tc&ee@>7x`OvlR&d^
z`mSa6-F4bPGMtFWimF6(1co`0&(OjHE~&4NgGfqA0nn`xMc98k4~#ZgSkmvPs6JIw
zeTqjFu2M=?r^5xoC(<EzO}X8Tfnrs%g-{*(*iLdg6Oz<p_J{YI2f!rngM25XW$_b@
zae!R_gsZ893>1$Qq^HlLyhusjAXwX}bFmg<m+}O0(|DBw3q_buY7C#no8dQZWyD>H
z7jCn)({SbX9ADoRvOjKplm0&U_w;Ks4bqTq(*maZ`FYHp?HyFAGdU(I7Tjx?Pn+JY
z27h=D!4qpM{$!~=s0b3np+m(+|7g$@3;BGn$;Vn}>vS$0#ywzk53cK3y)<*j2M}jB
zI3C79iTd-z0R#pBOi$vTT;!^IFd`pGgw&qN$w}#8x_Lvt3%|aah6aL&agv0H_=)<a
zXaq2sQ8$>KN2(9Nu)zNPNSK@hj3$hH{A}U!7M~6YDuvI|X)~EmG9gtsC^;LIJC_H~
z#x>>rJ@plvP)DKWy}Y8LzJId!Z3UhXoKSW^A+F-?LRU3ae|~y?mjG|}&$61n>Vcl3
zHyI3H!YkTNj&cSk2j&3D-aJ^W(*@xteD@nF`b;YzKNvEx5E1LQ{#@;u^Wj_tP_};S
z=WFlfU0p5?n~H4?-0*>YRGaBLE@*}d>Tu&s*<RQa6iP5dd~nbR?v3{IJresJ<78%9
zxEM0TSOL3wikG}?Z9Pt}9G}Q}U#->NzzhQ{eua_eczkels<Wy$3}$!9@9D{PviS7t
zRDV0Pf8{79U*p3EAw+OsOp2|e;K$(d%CVh;>2&JnrFTTcj6u!oBLN&3Y<vZ;jaTrd
zSq>WGd_eiehfJhZ=lhakpJ$zpk#yMCloy0|2t5uK>)FUgUF+R(5Gb37i>D1ab;QLH
zB+G;ld=YRrW-yUrAum;w&x6E8AZ;=14P}D>46qp}+MM1}Q$zXF&b0*P=S+ihT7K;!
zloG8tn?f$T+mv;<K%>P(Yn@`e8}Etx(6RjZajl@&Wklnr-ei4n*-Pc~zWuNH2R5O@
z%nZZ;0DeR=hRCG8hTp*_75iFjr;z9QdQD^?sDscD$LsUq{_6kKwbed*r7(*#ZZr3|
z6Cg>j<#=o1?BQZ7m1l;CK6(9qEjSSQ>fUtN^(6@yz+%0cg^(OBuvF*0C*Ty`J^OWK
ze>I_&8fzDZ;$R$$J$A5vY;0(Vl}T2*m4;|+f~~ZQg{4@#(zJ6_fg_-qUOBb#d5!gK
z+C|+x{c89W@SOhhs=7EppuJ6&8vpp`iI0y%o!8dG)b}|6n1PTH^e_IGI(q6J8;d!k
zw#?ColPklu!=3uV9tZ*>s7vg6P*PJ0dvE?utH-r9KtZqq+|Mo}6UO=V{ps*9Z^^ow
zI*S`}LW>=sNGftT)})+rAEsBm)Fh9n@ADnNC?F)jTsU3QL7Y{23cpX5%dK^eIZ6V4
zi=U7Uv&gRVfEbDSGOnSj<8<DqAGJve#*O8D$Gy6x?_3BtJ$61C!2DNu<Wksq&DqKY
z**}j<xA{%lsLGZJ-apGf>Kitdz>u-bXb$hfU)*iuMnpTv{!m4R$1p&QId2eLH=6k7
zXF5P=o((}~8--q!3JiO&5(CE07RP&rZMaWNe2#y@cy`Kua#>RZOv@W_T5unt-qG#-
zJ=IxUU2c7JsAw?xL0|UL&#@Bt2jtb#c!0se<%8QDEcui5=XCVH+=@%<Ph40weGoG(
z7eR`~m)##y2S|gP?OkU%p}URMTe|$a*Cle}>i*HMSugMnY1-Ai<{Mb$14Dk`UhK8z
z_iu>e>CCFn$>pPIBMe6Pm?(_5-KKVvl``7e@0H9D-&i|Q090bKuSd4CvK1`r)x@_O
z+Rix0YRQ0Eq^5@V>9WR5U$tcuB_uwkr>7qr+%gZrLd4SU=<rd(Ap)CQS6f%+AbbNG
zTT4?jqma00(sT1%PDAqTh*Uv=)w0%~BMTDjm?AQ6*_fu;?FQ5JNtZZLD#yVU7@&#f
za_MJ|nuzN*O%*&EO%BAt+VzLUvD9xY!fx7Yf4e<+r1A>W<WyV5xY0Vf65u`2&zWED
z{Y^mkeU#`j*O(bhTs(PxZbgmt9W>TQJJ{f)n#Oe(n6+>BjrKyl&;)@|x`seIE>SqJ
z^06lP^z`zxc47hV){NUWLHpRi06!c6!<za#0ys4sjo_<|BZyPwIQy-r$W|C%UD4)Z
zNl%5zpwRvn0`<8S&xXpq^arv^l{Hy}gcEbHaPBfZkW>>-2W8NNgCcAbUApCRy7Je;
z7&V@19koC4JXjP`PUS^KMFo|p>=1<Wpv#Xkv#@k@v}rsY-M)|3Ge4bSro_Gpk3G<O
zAVBbx)7i-_u+y%8S(l!UiW3NBjJbskx-TtbFZRCV-j2Dg?O4Ut4bl_z&Jhr~!JDA7
z6laV$-n;<*G<;7LyzV%hR3YF{T<t??X?k_s@7D95!WB*A1`UVX-c2GxTrly~V&^B9
z5%5li9vGMsZQva$dDO4t#$fn@<wb7X+lLwN3JUTBy>=p`*Fa@*POye~uFJy2jB^7^
zUa^<%&rX}Bn5I;o3z0UBA5_iDbBpb!WoF%REy0{1#mB`7Y3u4hzXy(Cim8^0M7Ni<
zj$zUUCVOFY8kH_0KCeIO$viqbItmphgYdt0XpI@@SwrnOlB3WP8oZ#FcVITBgeLuG
zCo#vLnl%`Azm66V9iw#v6H_sjvwd>tM*X`*FV-74!4S^@PNyEbT{Y>Y&tzq{7s4;l
zP}2;kG11-x9wkN%Ig(hcM0XPH(wLXp3%j0OhrxAFJT3kn|2+2*d3A9pfr`@TG&mkx
zELiHW8%03)v{?INDF)|%oE+Rv(81T^pHs;P)Z6dTKz%m2#L>x;zwcBL%c18hozgUm
z79b61+RT*VX#dFb2QEZ%<roNfvT`I%LC%gshzxEY>hQ?)W!H|ThP5Cf=(fqqO2<kk
zIemWhTVQRQDPn1M^jRs{ShWY$rt}RQjC-5#C$NtERrPgDcE3-5C~mxVaXB(BE83z|
zjH#!yMrE8lx;R*xosHk$cRbqHI8qwb(CkeN=&^r1IlA=*0&(Apr)e^FzLdD$@$8Rx
zFqn(&x}MWxn}Y9Xz3YV%9bO0tU%3^rGV-G2ck#6M`_K>=RxauZrG1AZ{)3jQ%s!>j
zoh<c}Bu!-a3&_;?NbwMj&~<a*5&KamV=AVRjI0DpsrGuXmThh>Q+MK{4b?woxp9wC
z{T}!nI+jmZ*yv3`+MxOX=X*)RFx%;CJ}VLy2bTSB3P!iMxjFmJ4le@5n68%Z_;+_J
zz`{Wx=x%6e{n+yOj}`AGY?DkXOlwa$Bod_-{t^qiUDj4-Dg;Sy^cR9v_SAg>xu;5-
zWyE$Ex0XGY#C!wdSE?U2DJM0^s(xnG<pz&L{kATQFjSF1XtK2>9lfqyL!@!YjKPPi
zTa@ay?eu}NoYJGd(L_F{^nG;`HMwzE1Rz5hSb)}#_r~v}h9CWaQD2M3{f)eWLZ_AS
zZV;iZ^~9LZfG_dx?EDgLsyN@hqO6*g-~CNlI0Se>bj%Ew-F1K7wMf%S_}`JhNE4RX
zJ>g^JJXUAQ2(qv=<o52mam}mVIZ;(lZ{B6iThGg@m_oReMS=UffKEc#-BT9C^Co7n
zyzJr5q3GPN`JL&8D*_@C@HPg|Sjat*jtwgR%>ecw|HQ$a9GQcf;hAF-I#=V39Q@#&
zak9|7_GSG!XZ&RIpx28l;oVdMOSSEsJ_ZIZMh;3sCL4-3k_`11+bbJqVUUri&KO!z
zemtNGOJlEo1Jp^Gs(Jp`5BGKsSK9_o2bZ?B>_nJYSh`b$;}M?@18oiarOT%ATK7l%
zUVY!a+r&5^@wR<$RQ5)<LI`6hxzk*i6lZ+7b-y6@eyP3v))@v5HzybxzPm-jR7Y-a
z?mE=TyOpQW9Iwv!PDVO5>Zt;$2$oRy$1m$pTR|*BZeCvKUH-P__OvM&#EDlZf`qPm
z=dy-|v^wQj)uzUq+a8x;ogFOwudf<}lJ9L~#Xd+#*gV64F_@rcLv{T5`}nPnzr-}D
zZln2NT@a9hmsMdQe+b1x$<_Oix4U_Qk!@!B<QJY6aqC0cKx!BKPfE|R&A?*u8$1TA
zL}BEQzD%7YHJhGc2tHBn#rq&B{qA}s<40)`dYB!c6n-~8_AJe2arzs!(v!~hslnE%
zVG`u-$LzYxFo(Te81Ju%Us>46=x37dWno~~r$M1pVrY3cFO4hX^ypzc@76If5mBV<
zucGM`35;7*c#YM&N%gM+TCebI9OkOS2ZBzACvZW*0vBN3zUGMXI$qGPos7V`+Q(^S
zk1EU~6ksZ;zV>W?YgC^9$vc@AL64Jn`MHlB#(!N|B=?4`nJ2_`F_1(`F$B}seO$12
z15nK7IRi?Gc2Y|C@To$g`zP3FAlKCqWc=7FEFMG~X!}hz?-qTjZV0E&{A|n5&aQ+n
z|A1?VfB!FE9@Zl|yeC$J5q8m&RHMeR?S}MKtH0gj5nfpxU*?OLZf)9mD?*wQ?F2;`
z-L?KHD}VtpyhUExnPp^4%Lpj=V28`@Uo*A93(5qZK$pxObSY1(R!PFOlXh&##0liG
zr7oc##IYQ?t*DTFPW7X378{K%-`3pRnkz@6D|3o|!j2|9xBJeIS^Gwl@FBzMmtZsL
zgT@|PXWCR5ResZf3%@phbr|9g(S_??oxTq$QS}x}^AS~1ef{LAb7*O)Jp{X&yGEs?
z6qR*(*jSO0-c^0LLlZ7b6aI^l!LMua+WYm%^Ur3*=G>N=PKw%+L)|P(At5)xEQ+Z3
zH<(NJ7X;$Qb_#f|?=O9AYdUjoGmr<vU3R?+sFWcRmyjqLE%0qU?&bNJU$QY?UU@m|
zzu7NRQZiL8W33eA_VK=yoaH+-x|`DK&)gTCX1EBj==tbvAiL3+GfxXQHEHtvAkl?g
z|AEJN=4*s1T=x9h8*4Rz8K1jx;qf6@V<jeDGW;iV(QENKeAYOD-HU+pr3luhic)=Z
zBmU+vv*lQu9~(Fkxm29m=)UB#5*1W(2NA^bWF`s*kPfUv;NIMBTdu7=1=<hgdx`pz
ze@X+a7*>|6)8DLS1gdFO+M=SZs<^p%AdwR0xA5P;<>SgmzK747T}y3vz{}M?IDAm&
z#(S*39MWkfOBW$Yqf_B+<*bwvo+c*tvV{9fgXgiQn5UtdjSV>hwrDC$#un=$6se7y
zRpju<WHYpolCFA#M--N)^X1Ui#<N&DI><ya`yeTvw^E)MtUk(~0v>hEwcRA~V|sP2
zDujg4Qk;X|G8jy$FhT6byWBX}aBbE*Sm7_)Y_Z~_k~$2_z@iH#&l-h=GWkjA>Boad
z?Q~e3jc6sFjkdk?hgB5pC)o5$8Y(M`w0dnrZnj)DwPqO3x!CG(7&s1ca%LG&Myjdx
zpDx8V2%-2&XPjc7<rVbL<43%a_^FZt{R7y}H@7erjaBQh64ll5=PVF^!^Xi3{y1LZ
z=H=#=1$WqUaVUD{>TpwO(hZpi<U9>a3!Rq2o33|SNI!MfcUGc6qBJ$qdsDqUoFcoo
zTl4mG#%exhGqA^MsjJu7jS?ZQr{@QLw2aVv2njtmGQ>}*cXe3r+jQ27t8j9<L=kg3
zH~@d;<w#9EFQI|{=hf>j^M}`HCA{Pv_H>UZ8oY@;w_4~IBE8S1n_xT7&(Hrorj3p!
zjO}>}ve0Yh7G85*^jE&0m0}ioo+hUX`W!5b6~L@K2^mS7g}*G_vv2kiZc<tCc~9V4
zuG2!Db%=Xob*!FcpdYT!zY~&|=e#j@|Mm?Y+bMGL#6O#>U-I&1JvK&Q0WW_3IPa|>
zPJ@_}6Xc2sB7n5eSuiup4b*~cbad#K<7<0Rah&h-ogYl#wuY@out;P5uNEM(IGUY}
zMOi@!-M2;hI&<ITy7HY0_~j#$B5;t9o6>TYEpVq15->q#1!yQ4!r#}sA1^Etewq1Z
zH9Hy4m0gMN!%W9LSGD00B$*zWaS;(s=6YUSxeA(e?AStdU*Y=NKxs)RYjA`vKu!(k
z;P<g4Y4qu4Q%>&kW4gCblLesK;bvy=?ryR9@BvGx*2#<V-tOE%9d=IW@3CjspiJ0*
z$5w}YZytNyJl<_sQeB(v+~?bH{nVXO;%DXhWN}f^;$ZW?#Du?VM|G99%QMBi<rslY
z@r}k{Oe}7pBc&}_UK#TgN*&}=bYx<3B>1_&=D^0Lki}0*0nwSW9UxLxcF~fO3c$d<
z2G6d?{I1T4swy2`dqb1YH@BlZ%{$xh*e#EBLRAJn!yhEmA>kp*%(&(;-}dpj`jRJC
zX{n3tm@oTBWm%=p<H28lH&<DG(BPCvb&i6UynA&GjFe#rdtE|=npLb485s$j#A6J!
zTcp2nSImK(NNsQhx8SAC)voy+Q#NgVJ@vJ}v)Q6)lODIX&;KTp#r9fST5|JnrW?(p
zqf)~FBh(_VRrOeTJQly1fM!j2Iz@^;`UmyPzkj36b5TVN<wybjjx@gVuFvF8IDVVS
z*lVDrsRdJQgSkUc=I6#$3wRTQ(K`z>GjWl&fx%2)ommyw8P*4aoUO&~%!bWR6o$aH
zk0x;T_hSpE-yS+%QvNkKi2h?=BS?}++T{}hlQ5cVRYiI=$43_eWA}Q3wYEZ&`Hx0}
zF9QGUxI&JxSrw{3q~RF29Z}QKvo*EUmTirRfK~Co_3TZ`RKu0|d;%nOksyZi_!sQS
zl@E9?Y|f^YQ=1GLZ^aa6tjw((w$2I}G4^H`<Wo5;6J=&*&s5t#)spdO@3_?trv4J*
zIZC8USIbW<bWW=KX`gZ!Fonw=UU{FNbb@0Frk)_g{5b7&+~Nu3^6A;v8DxNtV@G{S
zy7?Cf%G^C_loJK@b#<+J{WA0^Y|mV5EWE%)SMk%;Eu2^Z_n%$qa|2(Jo?{41IK&q1
zoSbhoUW`<^;IhXG_^c!ZH+W|R;({F!=Yu{B->xtBY-|d_9AIor`*n>VUWtS%Dm`BN
z{MjgkMl&j<QpSGL&@w2_%F}uFbUGvP=zW@h?V00P{Z(5*73DN~U{?YUCelWSXTUl#
zFtpw1MNfU-WNcM&>y1REolfuHR|fSb2$B;<zHV%B*_ww9D-yYdujlIO8aN7bx&6Fb
ze(1f&?%MP5K212t(T-g`3Uw<aWpu61%iRK`nn9D$6M6`bFr<X$Gc%8RDYGy!{UaqH
zn`i17EPBOfDI|iHc>ev4=STIW;S3@HZ^3m)J1+4=k6(%=Tq(KUar9~j5?4V>!qTX4
z?b-!MMtnanapfo_3BJA=FG(c-3)x=n4*b})t}o<?vp$W5G-%oB*S}sGPQ-0`cXNHb
zd2r~L>-kWPrRDy}uG{9q7V^O1vNUPZbIcLd@5Sfh*vLf6<Nd$cPAsQK8*DM}+v)I-
z$(pO&{94n+p&-aE2jQWv3%Tpr<^1tOS_oG8TYitv>gd=cPLihP)oSkg^^#`^30MSj
z6121o>^s{H<SCo%o_@&2yxZYV!ZelY*Bc-1NcfnPjAbx%x$bSw4+YJCdKjBiXO*fy
z75ioKg{7gO)oCW?;37P)9b4bo$h1=VmpqQdzE`U@nJU^DNxU{O`RnW94J+FoQ85+{
zro`z8Jo(_svomuPA5RK7(hAA-%f~ChrPHT2BCK9sh%4p2x#bsCPV<&vN4kQSk(O4X
zt5w{*XQmsM*pt-A%3DkYqHP&oJECB(*<B}~vKQl^Y7*_+%V6x8u07iRoZ<EQNuH1B
zr|fqv0qQaq+^c;*uEi9MyRNK;=;`vdSx5ka02@a`O3_S~I&8v==O()oAG?;I!2<ie
z#{%QD8~GXa8{iB@A>_?(v+K$?inBh`+Na_(TUue7&EOrY)R|Iif1#v~@38u-1XzhX
zA0)XJ*%jrUVBZQ}AI-^0LWNtbpdh2(XzL{|Z>+yZRYi}AzADLrkALfYSi<9s%yL#P
zmWlTM%kwNNP`i#+hFqK-&h5B@;G*8~pk+B+B9nB0$Bp2a03qs0erU6M+_wDt5m}YI
zGMs3~wbj3WXL+25zLHgKr5+u4T&^Zp#bV%zLrUc3g*(TyhMu;zS3lM*jB>if#I?0G
zhMHRIl(#nhukTA-gLM&CR%V`Z8x2{6$zG3^g}0eekn@0w44|hz_IN@Qes8*NYPu&o
ztptpyvV4*n`ihm64L$JfU=S{d_Dpa}FD&Kp8PoPy;t^$d=yk>W0*4f~cP=@>SLa*$
zK<mU_o1Y$Xn}Q@VA*O%mS5$Gv6UAQ7z0Hhw?;s&{$((yN%_s*pOZ?-7=1CoOsPXm>
z91k3woNXq4_w_xt{Q1#^Qk>_DPUyb-_3Q2yB=IO>y=IE3+!p8NoI$J>y@g-PztC_r
z_WkHt{1xtnhDt5`kdL3oul+RJP58sWr}$)D9nGn!oX%M<0fdT`h8a|1{X?0s@6T@9
z&$TtE;+p~?QxOLoSMY}Q7te1(qlNGYz)<Aacn*iYyc0LE5;=;mg#%I5@wy^IBnp)f
z?b?sRPHs}0j;;W3ihFsdpZ49Q`(BZcqT<H-O5{>^3IkDa)h7S;>PFy>ku+?-B3Pc+
z5n%A2Juq{JZhVfA{B(V1W?^Qq|G{0w#+xK(Wn099I3+tl>ZWWF`2LHHO+#bBSMJo1
z)#rag7W5$m!&iT*F1>0xWvcEvnow|vJQ~7z(kZ~Zd(Ga(>2jYhsp)UILGI*C#%1Ha
zG|@QDVI{2V_i%_2pxBVlwb_ev(PW6PVMj{~V@H4g`{ud0YFA`=u#5~*VIoK^(ic-U
zfdBH|v51)X2+=GMUZzP%=-@E8|B;m^5yqwZ%~m0oWPQ>T96^F;N*+JttI*Zr$m2Kt
zvA)9eiR83r=1#T}sSId{Lv8|{2-c&&twUoNx)Ox>?$fhzH2&%F_XV{3(`zAy$aYV^
z7hP>HLH<X|9nv(LnPmnUG2}LE#`r^Um1fd2sKaIFTSL(hMi~L{4dr8cy_WqbZB`Y&
zAm)nX-sc;o4@)qSKbwmmCp+MBP4N7()#V;{WiojCH4yjN;=%g#LD7{F-P?`wBp6;Q
zP>O*B@z@GoF)_R`YaG0s-CZo?eTecs9%HI=b5|;E(bIz!KF_w|(v{qzfn6*O;#JUP
zOP=s2IIq&&5kv4>Le@72(G%zkgW4sDI%NMuM@I=xJ{K3Kh3!^R@gvIjp<mjnswfED
zwRsv+$nBu<BeYl%g->_llS2rDBmq*f=E(uEM!TVE-Uk4pMBNAo38B~ro_B%6JZo(=
z#43i?<KfDU@vo+Ck{Sx)YWBvCtejTj5iy^l03hsAP*k~8ZNdm66}~HEPH;a$S{+7b
z{Px)m=O@BRjDHU_58gV_aXi>HZ+(*}`bpTZMACj`RS^+AoA`IRNBh55(X%n|P$2F{
ztJe|e=m2&2nLhWQ-;vZ@*|9%Knjw;{yIXC5^Y=ZRuvU4nH}M5DF_VS*_6<Y0*7pfl
zSQ|9u6`s=|uqC315-Yplg)>=(GCy$agZPkwsPe8lTvV0#<(J{IlE-DYnbq<T-=B9r
zLS#&PKREjz_4p&11JSF*Fp`0SRJ0C{sp^GH_q9Q52qZ^zv?dzrtb^;&(l!(FS%s-0
zS`$4$N{zS3?xocfiQ5Z4zTfcpw3wFtU6PSt5acxdG4QB}U1?f>)4@#9W|WqdpMe5_
z3u4aAjC?VuaD-ZDIm)kL+HOtD8z=bY4NvCVH-gaK$g(clSIf%yGq76~C_$vp+6!4?
zHpN?DHv@YxU+~k~I)0ZO$3$MM{O5XvD32t>#HhIiEn#_Q?+Jyf(5JT8?tNr^a(At!
zd{c%Yg&>yUg9PMr&{zWta+=Ha)yE+7(&4o;6B~<+5G4sY210LQ(j7@h_OB+}urBQX
zB+1?j#RGFk$5ovz#^Xh%>|2*-SfucL|2W4Gvy)freI<ns7RvRQ&mIgWY@7(-xuR2F
zVLS<vyHNYwbw+O~?JY*V8-Rj9ZbbfbdQhl|JNFN?u&59qzJuScKA%%c=cT^k9!MG*
zfExZ*ye#yb<Qq1P$3M9nM-Ddz=RE{!ob(W1;cZJxb6AloUXj>tx-lc5R$W2GY`(3b
zsw&Mf<}<3?2ta(T0qX1;{;YkCyp~-@TNlrxhfq_Rw6ofOns3q?4BN7l)^!Wl_ZPq1
z{U9HH=chB(ZR9iTCTt^ox!=K1XNAkDyh>A3^ThA410D!p?|NNfFAXi=w57@f8Z+t(
z#XWY+$MdjnAlsd(TI&Ni;dflIkgt6A$^{uBpAc}OBeHZf@k9Lw;sP1bhfd|`e!hoC
z1Xxlf-MZSOB+C%AQ>M*&!cRKnN3@vD_6;Y1rhoN7cOEa|<IMvBoo8**n-T?&$`=Dg
zFqovc-~>nUT)W@=%di9}gH#TnITzbFY$iP%BG&&HI#eSnIy&^a#2;{n^_MEA>I~2t
zPWaOOLvMWZ?2qml9-#PR@3t%*p$_|qfBKs@aPa33LdDZ=n}`}EiBh#aJ^Q2D*IpKD
z&Yka{TxczbrLmIw_S5+R<NPixkVP}ugPH(vK2vakw<J_uzmD+hRiuT(ENZ4eyIh@~
zuDrUgeB>ZX<ax24*y^XejF{}j@N0ENGquEo_^IGcvsTm?P_qJ?>_T9s1$Aoa8!e6J
zwf0ks`KzP?x>iDZ<U6UU_aYV=>fId$7mG?u2m1Z%&TXf$gWp5mVf`7^_coO);2hPK
zge`8BGVt+_Ka;^(|K`2st?*o2?paG=-6heYDP!2W1hqv2y<$*%9$D2oSX{&Y!OYx2
zZcQC8iH1De+Y-=O3_uAP0c2-)w>XjxuZ=^*vZJGr>`DY<q#w+%{%unpeIp~Ni1c1<
zZbRA~(2YJqY52B(aY21P?#?pYKf=y_Kat1c911Uf?VIrmaAHV4GDzbOF3Zo8si=}$
zNq3?VLX?)26h&B-)mqDlD%<<+Q|uLi<)caa&bO<pq{j1J^Bp8kk19lz^gh{kBHe3E
zBU5jGMseO%1l^D*)%F#G-9FDKCb%^i8=Jap8oxgH3AM{(qiMC3Ih_zC3}1$pSdkM?
zvA_aLG8;BfKouPUcJ?nUb$ByCc4_!D=(APeEsDob06Aly6zQi|7Rl=tR2F$pf3fiL
z@TuC}Omw<4ksDl;FU}AdAWof;hq<z~N=Bx7EfWh}ZSO<*Jn68rUERS2s&9Jw&oTVK
z#sVUmk`0`XK_OxCvIC=o^<EpOAU~;ekK|_f3?w81vE@{LW)Bj?n=*!mmZxgn(@eFW
zC8P>@=g!-xNYLiG;aZH{k)*;y@ohdy50=#Rp{gLrFL3F#g?#x_U4Bh@`Lb{6YW#0V
z8@g&Yu)lde!!<H7p(l;*n6<C6)3-CVocv+mW?YIF(z5vKuA<_pd#z`K=+m5ekArM&
zR^e}Ufdyqoe+G#Q=egvv#NEIDwZE{Py5_sb=)=qPsK$A(=nF?0yw|t{MrUCYs(xNH
z^cpYk9QyfoBMg2#kxOO0or@B)spm<P&yxV91*i8qTnee?Uo#D7uTCOP?AgOR7*ETp
zGr1oL4Yu`tZ~w8m6S)MEKZ)xHGo>#dh4o&UY{!W_3X??;(!z|GxEs(T0w4)Cs?Y#o
zguQ`1_S|-A3SL^PlUJysH>K5&j^OeVLjeuQKbz3r$W$N{O)ivZtU8Znl%!lXMJx?D
z!;f?za2#k6nFPj@fjyg~f&PA5b7R&xt_(mYiE@UP_iInxx#<FE!c~|oi&L)H$<;6n
z|8o=nD3|XAd}Di2@0E=nIe4Vs-bPF4?r9dW%=Y79dRh7#GU5REZAdGbnmTenSjIw*
zD1>bdIZr%)Q0B0+K5I~FSV9F<VPhZn1|C+x3Az+@Y7@$5nJRXsYo|z@feKs=Sct64
zFL}~bnA+yXw?_2RRJ4I}-mFUQ96f)>9@Vga#a0(0chn4u17Hw~ryDSkJiG<n8N5P%
zT{Wf!m`}i^6xlmI#|V4I&;Ky~OVHqwa+2_sN0qsGL4Y9gXYYh2e7Py7Z}J<e`N&_s
z-xH`{YWv(<Ox4S)X-~^8t*PbyavDGY@^pAa!6EC<-JOrFL!6@sBoM(M)EVxzG@ToN
z{ZpjY7zYaKa1lZ8)Aa00VAJ$!yggt$-13*`(=jIhZp^ud{Q;14FRyD6$~=MomHDFV
z;!x-WwMUWO_9`e!5%NU-^jye5{n$EB&hKow^Wh+Gg~TUoZ{iD(?e1>XDBr!VTEW7E
z?~xF@aF8tBsjG+A#oOtTs{K&G@S%N$a9KJ6EHYBEze`ns*>Vn<mHa-UM1MjT0Ra&}
zEL)G&81U@xR@q`FD84loqZ|kjPZOp37$Ps(z$g21X?^+gmoFN6>SwzLfT`I-@ESle
z)Na550aPCJ<bE3?I}Nib1ihMZ=ts|>($!L@(!$^{;5akh$7tQms+kQ>6yy|p+vH^B
zFtNi>ku=d8i22JCfYg9&fKR-7`RAP83N^K1U34ZN5jD|tGSPe#!5tM^$#>*){<fDm
zDki$|Y#TEnXD(Fm3nYXyFqF|ckb*<S$S^hn_q7DBOH+(q=Tp1<!h*=7q#I}+0%VBm
z{&AitRrM03?*Kg>^y;!D&42CJ$Vfpvg4K;5n;DH6u`f!8N4xXqRSi5z$|?1z@he4k
zTbszMm5z|(UklJnC;G6`!Aj5cB4j3{4v@a|bo-V3kTvg#lcsPrH$e*0KjMA}QvE!|
z(JJ9n-+LYG7=yakGC|U7b1Nemx?0O%Rb?;r@%@MHu8uoc0X|8!Uzfi3AHOx$AO=wM
zd5Jclw(P4|d8D;&1fH$RPpqHMO-ix!>Q|@d<}S}J4ivj6BcI+kez0}4_EfU<Hg>$n
zDIh-b1Ykx$<h&uK!h}AHJ!O;4do0?V8`TeNogh}XpW~6cg1M~y<9osee_YnsnQz5E
zYLBObiGhqej%U*nJb_Jr?Qze&4dFHd=m7ci!tNRZjUeFkqJOxC>`I?nhxeDzwkYb>
zzg!Cw+V{b?miY&o-MzWE^`mvdYX5y2yjog0?g<qh+$ePgWsZy)lq``-S{D^QGr$Kb
zq}HKjee*i}Bpqqy&FTKD59aQAUxqieD$z%_-D7td+Ww)0bGXa$tn!sD&>rCPJcHpQ
z%@PQNv%CbM<BlWOqle7Q>=#bI(2%dI)@}eZpP(1%lrRC82r$|YN{JIxh-Z6N`r<J5
zR?gK^e;UBOwai+f*9Is}jVUS%dnt*}=cKuy3K&e5n69g*y(}#EyOa3akH))g>To04
zjms;;^~f0MvS5DX#MQ&e7~k*&3UUkcTw=o5Loz{Wd3B?mH)Bie`a<q=tN{v}qd8_(
zMf%X64lc4j)zH}AIlOsEFz!dsk9@E0Z&FU)`{z0wS|11yTl{?B^JH?N$>$6qWKIB5
z#kH3K?5e`#X#W*hWVN$qulcn7)mbj!uUnP%KgM$h%V`>y{PyxV+O`a}$xJI6X*Ae^
zRYWOqk6vN=EEeSh7M8#f1<tLo2%<O6WF@P{;A?x|T)xIPj`O9pr?|bs^PgY))!)<6
zgtbzNn7&oSI82Z6(6t$@upN$|XW^YZ@VU3WfF9NgBb13FMSUX!k{(=sU7SxQ!TJ$F
z9(*>~f~5*f?>uhZML{GmaDRXEj{a78=LQ%oJ*%<ZTXAUc=H2AfVO6|!FvgW#Zl%7*
z_BT5)F*qf~*2HAk_vQT1a(*XquYlX(fSZ8`I(h)`99!eHuOY<N1D9OjywyTD)y_D#
z(fEfoqN1K*H&G5c;|!J&8|g#GWY+F10bQS9TfO?Ed+QPEgRP435*cdsf>Xj?I5Oz~
zIhz_y@o}-3O-Bs7*YX4N=|OO;zy$kvaAZ?g9zJZdiQGmgoTuA|yPsCy{C#0-I=8ua
zk2savbbu8BZ>oB?Ra^v`6nO=u3m_auXXnwy(<Eu)`D`L0Gra21-K*<Bu>d9@xTFv7
zApANBRV1-`F`Au3KpEG&?+GtBREsiE-F^=WWJ?#Vm720_nDhU9u?Q*F0IYzKv=q%9
zKheAVt}jIU$|@3_A%8?_&vbgvCkb#5@4X&CEz+&%BTT8SW28+?vKs0=JdcmvHq$lf
z@GQpCYZ^vv-$_Ys@wqxqWLE&J@d1Vx6nmdnx(_bjk+^khKZ^|Q-}!3K4*`<wM8x~&
z-&fI^QKN;=ck6i`sTDnMbikM5Kt+5b%{Y<;J%YfJ$CG?{`GpJ(9*17N+vwH9a8YaH
z2jvwN*$UQ-%mwjz?Gfl@z53ty&HClA0$6X~&`Tn34I@E8uALou9qx7qhNXH0z8!m}
z3%P+mMX1c-2j_%Rd@X-)h9cD%;$+4g?Y=hM^gV~8oxol~f;PG6!<^^Q>a3BerJA;e
zpu;7n5p9#$wKH@fj{rw*=+dAc{b|xTaYUJZHZa_uBKVS<SzSuXZTq_cNJdz0rCcyC
zH+=Z8+b4Db1f1;@41bA<x}>r%k13{<{23YE6F4Hr3HumcX$fQu^1^EBWcLK$ewr8%
z65`ajf0x;DpTHP}|Mj{oh%S%^W-U=|xxXQ)r~&tKvVGv*g>mO)!uH7<P$aMw9v+EZ
zt=v8ZW$ER`S*WoMc`U`hbuI3SdT4k!MX2_oOm~N4XW^v4Uq&f?9u|@Evp2`Q5wQk$
z$;mi~g2~rylJBkekr}&5nLxNsXeANWzKYtjl;p&m;sXB$n+_Q|Jiw^IKOaa=J#uno
z7h=4iC<7G@6%A`c<KH!r=m=0O1kq6mi3y+Met{B%mP>&8et-hQ*u>c5-%d;m`6_fD
zK7@9H$f7@Ys@7So|9#e1zNF>2V>o<f&HkU0T|F&W2p(SK8W~CGVP#tYHIHz3KhSCA
zGVG_~;Y^&&m7l*tb_1+va?#}Q4LWXA<n(ZLdM6YNV`g8u0RYYP(Gbi#!#K7N?X}YT
zT|F<?swm@FE30zqvI2BKI|;lvTxMOX0_C(>n(*qxDn_h)()0TosC`X2*dKK7kj2Ku
zCmRiq4UDxzBH%Mszk$NF*uQf@T*O-<Hw8WBz6M81$Rt22HnjGDn4f;_H7tWmn}rSz
zQJ$wTJM>T8lT!eA!WTPpcxWN^N4n+F@bDwxI{=9X*4{=jB$SmCg^G)XQHU#8mM*Vg
z5Ej#@=muNadnDwH1Xut9=q247pQ=SiI7>@GS@nfq=PntEw|qXz=d!GTz&d&P69t9)
ztd}&Y`V>WyG!Oac3$w>Mr9XTq#Vggfpk^!uqB?5m-_PQVUtc%3Gn%ULpWB&LXF~SZ
z7Zn+<R`9K5&EKlf?XJxAk8{B|KSW%>v<V9f3jkt-Y?UzE5ur9<^M$`3l#BCt5Y&A0
z)=iLVe!RI4YLbeHkEf^az*L_nF(Xs5tgHl@e#@zMVbm`TgEBov*|EC&|CQsAgrT9S
zd3LzT?h{F6!G_tM=cq{t<@iX>w9h#!UJyWOd4*XdB;-I2{$1I8b0}<QZS0A4%g&ga
zgpAbH!>!HI1HeJy0O_yVtmHO@u(Fgdx9~uj0m2_Rx9gkpifNy?O6>tm>+kw*Tsn>J
zGq-ng&3mQExU|e+FEf%5f!!}arsakd4Hv!<rCl!$dl@>an6;;&j$0OV=ze}Fbp>1y
z>(>loU%nbQ2*GHbkiU$XErse9q}+f){<|x?2)bK<$_HQSP-h-~{&D1X?5CkNb9?7<
zi8*S5p!9SjC~wWVpN+Cp40K2Q1v_pEJR&hI%@e2=V9RD;=xpkaFUki^0+?q#*V8ky
zGn=1VfG3NBEY;jN*UrK=CN9EsYD`T}k&T&WdUEL3@3ub<u2E?XDR8Hz$?h-Oyt?K;
zyEtcVWfL815SNr-`qK8DGv>n139w76mX5F__2Zkt{Wsa*jKXKOb9~<SwGF6WLc+iL
zLbVpTRruF)YqNRi*(4-FHdp7+5MVE1F63DPDJ)>1<_l=fV%x>Xd^;^vhl0d*h6Ks8
zva``fzIS)8)vFiGqPAa92l|1O0(>K_zX*UD=gB<eg)g);Gv?*z-dvvv6s2o$BU7Lf
z?WGXzkduR&i!gd7sk#9@l>_)pCT4bsmx1$Ex_8aLfB&$Fu|)fP&9}kH>A($LYJl?T
zzrl+{I2`~j?jP)bs~0rhf`)^hAqS?t)&H$DHoz2TS}V#bDD`&q509)))L{g60%t9r
zp?0>v3FE0c6P6K~1-Y)(b#PcQF`C+O@v!Hiikd!E0)HGcc?W`RUGUY-%FN0#Q)419
zlkpgH%S$g^ZOT!B0#oeTDY({v5O4DNy{w&-zH6GAs?uVguO9Wib6NgJv}THw+JEQ&
zF92tNvS$7uE|d#)0(#P0HdG<7z+QR1NvU@~+zkoNxR^sP!;e&tEgH|w%DeNbcgrp~
z@9hsy5s}XsKN^(&fgjlcI*{c_g*=fI)A+YIi~M-OO-MwzHB<4|j28{oYyJ;)1(93B
zWFTHDo<RpUVesGpyPlWmw-$YHzGVETQ2hr5V)E-fI@bKtYl{y5^2BJ7es^Sh+~9Uq
z(vap?l)pN^G4|7h(ElVAfdl^cga7-+JrUdxY4V5rqp4F^Hy1hlj;34w3(EY@=boIr
zAiwp$FaFPax_UAGf1d!~^cL3spNjc^U%9vU<G&~9|2`}(759Jfp8q>!&#$5X`-cA>
z&f=246ZijjaR0t3lG-RHBSCX{RlGa~Y7>wKUP-nC7TT~BIx+UFN>)oPy-=I;>gP%Z
zu@WWBOr11(HqM+WOid+u^^!9>A2VwbM#JgS6Li>z*A&FCoyTr%>UKdXxzRx{Qb$Ps
zMmp;>MPPIJ#psUwVpm+L<=7jaa?SjFX27<pDoP@<s~-!%3yKOdLU8sSoyOS~{-oHb
zGlWJ9EkDL@34Q+9CHq{^(y?}{W_~#pHDF+l6@-vNE|cBOdGpaqshr*yt64rqyqn{-
z|9D161v)$cSa{7}G;T5f80+`GDq?FeyL#+!{;51SwHYQ{L5+FtFyZ@uC#hKUa+bwH
zW3|fyzx5+&_Loe$+c$u(PU2sh=<9pm!)_qhJKbXnz0KA0K`_R@d|WOA1;=R;Jp^;M
zVv#(w?gL`UL_NFb*{{V{&y33+@P;U7UcFL>Eu|%tw9>0LZ6pUx{H3|Ju@P;C)ah>X
zp3syY4y6F3bHKo^!3q<o2%F9-Q)^z)({ZOybqUqE=n48%mW2ChvwH8GUaagjfGB9H
zz|v|>qKSqc?+Hd?xoNYZE?L1$SnA^5u_)DfblsUg_~n;hZSR2KZY0NkcelXO@-j#c
zj=|Y>iYie}PrdBL%Ci$KzVQjAga@W;Jr{+_&_uwx1zQWzN^H-)iBbn2O(!y!@Z?CD
ztoxVvE{rCSQ*J^?2s)_&0vf*;W=73Rd;KV3y>?Z%K~Fk)jymQangP6NMj^cjZe;Vg
z!=K!slG(KS3Ej>VNgB*s!Cwjswp_FLYS14&a$>_f+h5&WoZAO!l1N_&K%1792Bu^s
z+MXaN0wGLhwh3s>oEGfqNjHwSJPaC6Uab|0fFQFd@4}Dg{4VH5|6-%zQlPgpkdl#<
zj&1b*hk^U=RlNKVYI>Qpg5MefAt$H*5*l+lKcid>4KG{qZcs(GuUhWITKkCeE+hvb
zeZP!%Tm>Bo!(~BgphT^1zlWU}lt4;~sAw738CvIl`H~bLP3u!(H{4zd$T{4C8X79V
zhrDh4+6E`sLEKOdb{_1dp1Ar~^vJ73U33I!W;9eN`Kzkso&=e`GzG<&iL!--C9H8#
zQRdk#(k6t_0|#aK`LNV*MaxOmgTOF+#9x%k+`<z14r)455(6i0c<40p(}N?ufE_=7
zLxYI2=sGz)qKIvswa?<y%_;vAD=buBFE}}s1kE{f#l=BQL(1l+O=4)W;Mu~76~0ju
z%pfboKUf?XPPAuNRZ;m|kpETt3&>*Hg&zU<h<aPwx>D0jK20Sq)_}>wr6<(cg#wL$
zFd9L^j+~~HGSt}Ex*+?DTLfY1H>cyw%L~fOOM{r|PU8~n;1BJ_rI5lS%L-XZ-97Dt
zBO_|CY)=J1DX*U14^cM>aSBo5G~r;=WK$>r$rQ9>1$l*$HkwbJi$tlS>d~>{C#f=o
z)iqVAZ)4t+#y=hW-KL@!qObjY0Ep+A5+D&k^)jB+OcQ=L3mnYzesgms0Ct~OqUt6E
zqO|&Qc7PX2HC01@%j%*cU=Am*-ZY*cnVtPkn>D6h07y|#H|=+^Hi-76zXV%12S#it
zmm?DyJ+=^#J}6|UvM}cx2dLw~y$2F6g(P7Eo;yz7_>s-VrO0O~8Q5Eaa9v@O=MWo<
zQ(&mg*MlwxQl7DNL)7uYf7g`J`-!&ov`0Eu%hG{-jR9{Wj$1`b#O?9}^`Mf|BmT(d
z%JK&+#kmWws-Gsa?O|ipJWtWEE_JZibxbBr-dtUm&F~`GwdRp|x^c!<@{pGc#oVZ}
zy}cVZ>|;FV>&dhX#)&N~Hbq@B7AdJ_mQg!@NmOKchq@%Mzz&Gr7$Q?Z^xTON(VT(=
zQnE+foN{uFxqdKd7(kO&6v`9ZS&&}@t-Ow&E{o=ms-Wz5@5&-2OrPS^T1n$a&U}f8
zloeG~Er|?p1O2y@N%XoV$jhK2p*Q)viw%ZW|J@WW!?x%7Nf|K&kjEZ7&%DKE%VO<7
z($Eit;O!g9Y~38GA}9#F(GS-a`3~sBOvBX;1a$iXRbNK?`-y^^KW1k`vqRH~GvnX`
zH}~YXy%!_<?Z~I3ga*f9>6L{1M^kl9Bz%tGfl_rH_Z$p3f*b9x6cxL&8B7`F9m2Jr
zYdoh>WV{oEab#P?&D+hwgcsaQ&@mcH3R@^d+(SXJva+&W{jH*LLo|2^vN@)_ZvR=?
z>{!}^0BN6z23}SgQUtac&f`%7S-OEdBMg*3W1YObTY?B;LQfwBLjP*Zi;=i$G>uD8
z*<3%|vH}-2xAWy$32$`Y_qlQSX6!x5d|+?m0dNCTV*UrY?|O}Yt$4CBO+42&S-ZON
zBGf}u1YM?%6lB~bR8@Dn0y>pb8n64#MU;Z&Ev>}0lh|s0fg=3|Suwx#;+%=Ht5=!i
zh%X==%WEqUPw1e;9r($5pPqcoUm=;hNV#}(oPV4vTj{QV<>P-!e4(v13<Piq-AZ$P
zIm_D6&=}N!Qn2DoNO*=UPnAu)eND@u-_YQ32J6c8#P0Zqc6lHL4YAUKP@gevr(GY&
zrRi?&X|bHx^aoyEPOcmV0>POrHYQp?U_4dOV`U|H6e8h_U0r8zn4@crI^D~2Q-ocD
zRG83wL11t-P!&t6_lS|#2Sl>VMBgU${_UrpliYds3%e)tpq6d$K5M`WaelSMbIYkB
zV!$55gRC(T(01=}+an^PXxTHFjzX)Y3reBaE)I)(C${WLx-JwqFoO}__(8^7<U^_A
z*;LeSC#T6Xg&v5griSEw_)P!<h2A8%10IaAMl0-ih3PV{NEIej-*&D>S02Sg!6L0N
zTUf|}5_W3b_;d$%5a<XEji;Dss196%;HFE*kcIgEJFH&FnsbL+yh1a_0~dAkRK)gZ
z6QHvL7@i<v*=Aq}ZkqidNdsvqIL5so?W+#XIs4QfdF6?TZ{d8myMy!ujHA8BGhyN(
z52_SJMTNYMftK1mzbW1R&pZz~Cu`4MxJ$<*0e!={s!c0#!6tAdAJXGJ(^v+vd%u1|
zduP0~Soe4g9MTJ)SD_;eAIbPW8&LA=(Q`DuI!Tzxlq%HbZ!|k9!YldJ{(}P9ob<W7
zzfOqs<qk=lRmwLno)%(&{0$MM!Nc8}0UOi^Efw%tZ(E3Qb^u|spv&BLK&STxGiT^k
zgYuK0>j(fUqW3*;DHIiZ(R?SoiJjN^xNQcim6BPtSr0*p*hHIw0*~R7dFflAgRLlk
zP+XKZ_SJ+qwo{keC#Etc8@|%~E&+v^s&noi?xzs4FvYfsVuc*~^0Knmj$g^Q<tdTJ
z-luOg9v|#GhEhyI*akVd^@Z)U1U^-OIQSAq*5p}$6uv%L@lPZ%7#3I@y#~MY|5m|Q
zK6~^02}AogJp>Z{Jt{#Y6ZRoUqSG2_V_|w$YB@hP9|%KF&Kp<c$fu=%H*bDI3DIo0
z6ZfPHh$|xu7f@<$RNr^F`;j3{MV29Q#ql&3ldIDD`NrSbL+ERI4H_<<4p(iSo=ts&
zd4=n@jvF4_BG0i)O$o)RG-BwPdsPEE0kbfl*Xjd<gL-9#J0P|N1OfT;J8f#uOj*GD
zlm>wf2LNA_0`Or?2du6q`4QoYxl&RM)9o8K_YW#Jgqd{FBb&twmbJu=Oa_ams~rwn
zl0mXM>2{lH0u#!Tsc-g5%F0!xU?Sj6-pAYcStaz%b;Pr{oy*Tc@ulkPm#Po{5&AA#
zOPj)a;}?#Ug!sH}i>L)XF(62fP$UkLocbmNB0ZQz9BdyruiYYn|AKatnEgx-wAfeK
zTj&OjE<dOT&Y^~BEYz)ey|(#2R~va;?sAr&g<ZMHk03QC=!@Ceu#_Y{bg^)gqVM1!
z2x4XHs@&pRSY7{WKK*6ZD2H9=aUCf1`nH)|w9FVHD-1VG(^MYY?r%%&nNgF&k?sIt
zU?5b}Xg?K01f1EZ?WukppVPo&J7pIOLs@OQ;O5PlDi@$85m<K9a|-qrnyrvHE-U5`
zx#T+6N2$G4fG5v(^}_%=&qa<gb!595d5AVEO!J1fbq`}!IG@b9fjUHrKVfjm6W!A<
zK5qy7n2fP!U)9K3r(m30h$TiQ!gEragKNj8S5XpBJP3aBJ$kh~9;e0_9=c=uFd+R_
zN|M$m3AA0-Gv%i1QG7^Wl``%JRFjc6e?ikDA>q~Yu^mc)3g<lZ6pEtQ3)>3WFU58+
z1%4gj=H^bnrjro&rS|FOvtEO5Z0v04zDpc4a^3t50WsMvZ}_-k3QyP`-j{k3bhv#u
z47iJ<CD^(a&v=XO4J=?9L4tyiO^+d;6lQ?<P)XR<NyR5H92D|dSv5m(NvZdAm_PKo
zn$Q9vk(=N&VdI3>xvg^i<B`@0^}`4uj8NOL>V&3Q3IR7I9<R+Wd2-yOc!@p9tn4X#
zM0W@Q%Q%#u(Lb!@$zbe;ftt;to}5r}^ZrT@dr^dKExZ^ls+%k43nnLD?>T?1lAk64
z$P?sh92CUr)q9_fjaLoitSr3X1=tcqaX>vpyq6T2;euP@o9js}?1zJc@%1m70|sR(
zD#{l*bjX)|21mx3`5s;9Y68ByGTA;O;(6|Q_3}lqI)yE}0|@<#?T(#98^P0Ts^&<A
zj!5|)?|9cQ*UN}uh>sOD*gE~ob*H~%_rUyryUMPsqjT5lnY{*YpnQQ@*K2T{27!IC
zzTnEbIWFQU+*n;rfK&_W1MTT8c8619+tEc;Rkf$5du_7)XLs{`E&&*#YyUUs6u;2a
zEr1KS7A3myHDB9lefxZ(8*j;Ry>^YNACCpfw0ySP1-ejeSd<udbbP$_1B{p9)2icV
z06)Z@La`hrOXp5-d{YN2GutdTKli*|5m=y*kD=tl_q-f_^@&}ErJldCprcVOi6^@A
z#_L<Xa6Kl!`>RoZ88KO--D~fq=6)@dgTpD(*3AN%@cjH;{$HI<g_TA6&!(U-`X)Hp
zS0YLUb@pV#k;*B7iXArEoAWo~()eXxzP;hN&HYdZ3kwU`=3}*oehaE6KphcL3BAJk
zA0JI1vO?M58wrmqa_X?KO%1a4wt&LKDl=D&j}0lIc|mu=dRaJ%w_mwfRDH+gxF+0y
zBYBmrNXl49Pqoa)x%5Q7H)=-Shjw+<(wXNeOl`{o)xmXeaG;?05&2wH)DW3<vQ2#r
z`qkAnve1;rqeX9|b#$H>Kw4e#XDR;F-o~Y?y9n68)<11|ETf!cQ^<>n87z7!Or3V8
zw8zQ@M7^WX{dW2MwpMy2dsmLPZ<LGcQP78=D|;6U643UHZv^C9;D^gb{Xe3<GAgTe
zYkLtQN-7~C4blQiNOuSdQi3!R(jC$WiiC7Ys-(2Ev`BXeN_QjO`OSU4v(NkdVmOAx
zT6fHO%`2c+!pC&e(F3zShlaEWAE}zj@-7W<s5*h>QBEo)<?c_~w1{P89SIdIZF(tS
z8olajFn$OJYp{Ym=tS7hxV@^~TNv-bYDM*x`={mA1s1_7c)R%z0NM=&Apc!lGJVk>
zKw%yD0e?93V&Jrd%nY^AvE0*BRh|6R94JAvys+eIPPzMk%kYKwL|8!AVv2~*N$|8a
zpGNkSa5WC;udDDTk5;7=4GsNiQ%@2?r`D+v)1F|Cyb^FA#Bp{u#~^gR=Eldwa#`B=
zbo?E_h~)1-4&7Chc>46IuC6t>(GIEtoVOyzc3WXyFtFt+@aAf&&rDd)x@Xa3S;bmL
zdcncrMK;xai<KD^^52DGZ7^~021f0^n%a{Is5AarTgP#v%4V|Y?bKYLErztv-rutP
zeAnJXfQN@iKq0H?1zSaM@B@fSlB56n5ZpE{b?QI%-9dnOCD%x!KF<GVbYr6WOZ^ub
z#!9glL5CUw=N)?*L)jnrc(@9SY*$8m$lbmpNQxrWu>#d<R7p+{a~qG6d=Uy=MwgMf
zWeY8r5)1Fuw<Trb>RcA2J4smm$JOnOd;AQc@|mb&uZ_DwyTtf`qRM4>(9e%{I*%{X
zJp8U1HyLt%_Qh~Dm&K}AgKOUIQTvc=C?t4=EcZo;#4p5f1FlDQyoprP76Y5*c0}M?
zpT~mvG8z>9MV0Y2tf$c(=d`>|>aj8j{StQ-$!}!=oiSGCg4K(GI3)T~C=ihzkGyf(
z1W^@n+v~@Ear5ifBLXPT9&p`8kQFC)P_|Tzq-eha)6f4lDt#c(Cj*Ayn}+>cAuTXJ
zmTXD~t_~pJ?g!g&jzKiQUv}kZ#{K=<?Gzq(l?f<bi6i~bu9K#;_xbhAvaom#6MVsI
zK>GXpA>O@m<<mrwJT4f+;Dm+(*#`z5!qC`FztMH5ZvnBdUcGmC*t?!}O)N)MfX2Um
zJcfP+n4l?zB@P{aR6_%v9AHQ=sEC0;pN)}zgB@$&im5f2kpE5Ea$f@Fr-<6CqxpYm
zu@f+grLs|f;U?{5_4D(a3$S?&QRuuKrSi;OQgXhjO$pU6)kgyqj0>MCuGSuzu@=`|
z4=*>*yM4~g4T9cK`M={eZK)JVS-pB})6rN3g+70J=iVDSeh&`TXH~Z${&FgVfmls?
zeJf#4irJbgdP{FBE@?EM{$m*)kyeIdRjugcyAqXkOCmzTjIlB*N!rLV5-ET_ejAv5
z=D1LO{5a=Ju_R5}RooO%`~8gi|9#+`CFh@u2tILq6fmC{i&cnY`}lXk;K=f-9_B_s
zpp$+m-Y6c5z5nzB&hYRsz<3BXuSigf>Cx)&<kX0ava<c(&K3X=d3YbhfbY$@8N520
zhQsT=;XNW))@kVli-5P!F8S3b0Nf2)BDp)f+$;w{S3O<J;a9QV5GJ%X)`Ir#ri-)7
zuirz`s~}Z7J#|m*(f|1#kMxn*O#H`34kHc{k#f}uoB9i|A>S6$`K%7@c@Uwq;|H~X
zO5!4#%=KeJ;>J{cPIM+eP=TeC_<jeQvVPfG64a;ws?yk))}~GGO5dy`Mu2P!_a`gK
zB5_~axS|ZHJX<St<l0RWU`8WiHiu7c1;0f;UkdlpZg=bfy>EYO{A?$>h=P)mRQTl|
z>IJ&245{&Nru<`FOIv@LzQ`TxzVzRq5P1EV&L6OIe)r~Q``PcRMiYrbBC+wP9^f-q
zJkr@*cWu<6RLT){%z%{z-f$r7b)7stA067%YY6;RA}RwdM_5H%xjG~AEN#I3U|)O{
z*tx)BGLpd-_Zic2)vI_t6|TNxBtZ7%TPO_`CFw&I7GCbL0G_VMGj%PQETub2^qEjk
zR11t1V3l$5^>DCRG|br!WqEVb3XJp&kBaHzA#zf4Fgc~8HR`v%>et1BS1It3AJXUG
zER!qB{`~oy`w1=JOt^X>|ASAjI`PH^!1MO$_~FJxiT7HDy((iU8bX}pwep8~;kaqZ
z4)`=l4~;q}*W-~H&>PnrQO)kWhOC+KVz-OT(za)*d+~)Gt?25{pSzR3#zRzw9*TJ5
z#;baS+voz_pY~!8@L=Sd-KZ9!SUf;x78Yewrr)2+;7|#<E~$Zg0~|e!mwYG&bNeS&
zxXXWQ!U@{KY8@_(^}JEQ?YX^EhU+Pe)x#-8OFhP0kDHC1d8%`B*EZ2)#|jpW<o<wy
zv+yhx?kkB%vjLE(d~<7JS?^7m>7w-^P3mE0yFXgI9C4ne*MOy0Yw+CfN#I@njnVp!
z!JpVbD|ZI#Q$fqIjOGQ9Z76n`XDi+510gAhb}hc<4Xx%v3_z@^NTeM6KH#4kT34j^
z2%-b^>fL;f#KmL<^y3o8s?*#0;2SdEyFc(uUkOC6S6htzIa}qzRc>uQ%Og0@=EE&t
zxh3r6QYQHQ)%u9;bqFGg@AA0dJ#KH*efO@#sd~*#zX3#LDHMMruTrwIs&7BNaZOaj
zVa`h4LI5InrJ>DjZk#e*FT5CJDLkI6C@rt|K6#?3_og)cqAj8yh-)t7#QQS=8+G9G
zsaJ2`ncX*=E{k1fe}X-h>IqZ{<eT~AO>Ik7aKu-CQE71_qy?O9B^zUt5<gOjyh{}-
zY54a2ZDFqa+C=7dI|+0-L<&o2zCfgmm63LJ-EI#8lU1+xSC8BcG#If;17rfC(<g!3
z@v9c}(Ec@*xF-kk@Q*OhIh%>tNUn9+y3fTwJUlu(`F?A=VSTdH1I*04^1d}vWarn;
z0iM%_w-p)~#`EiUNYK*=42?0bnfBSbanOjou&Mwic#dN|)5Y4mXmnH*H`UYB3rRNO
z2CPe``9psmR^s$PEwV&1*7*A$1syF+)DM7a+n`<^?LP_U8q{Z!x-V8k{@fLG!#yFW
zbEEP!+d70pNP!-b@fr-f?jzsbFTzp^4nkR9!O#<)xM;vz*e7T|U<U(MT<bn7rRV)6
zH`fBNx1jy(f|P$`7d+f5ix`-;@x1Sq<t<Ku+Pkbo5frLaR=lS6LfsJv5LAV8t08>Q
z*Uio@rHKItFcZ)dMa3l#JT$1FZ1nh(QX>})Y{~C@QhRZdPG<Pz`bwbaz5fn(mTZ^K
zD3~$5&Y<zQ^pY+4M|j1^*C3sikP`&H>jcpWNgDlgMEzp=O1iqCU5gMrBZr0q^#$;h
zZ%1ls>d;Qg9!}xt8`1esp~-4}OkhzV8XA1K*v15V(!%;C7)ONcKR`!qgnSOB+pf($
z;?gf(JbkLpo0(sGyFd+Se|e?p==3*&m=np?bnYvWeSy>ofcCINK7EQzV7IWztxvU)
zG@`rtI9)0|XHeaQ{Bt2BBk+QRs)ZqwDZHX0E)7+>!s3k>(`2NqhHO&D%i>XXuX;@M
zpKxegK5{Ok8}M{4D<{0vxfb7Lsgd+KGsn=>^rxvK;09w}1$+;F*+%{}E+Hqy7SN5A
zxTmoGfV;n_ATK`nO={}>^hlb?doO}>bLC$tV1ZLX9w8Kvr|tq^cvqq#{v`rBp-xs7
zbG6*yU<c=?1-ZH4r!662VrBfbATvSZS-Sx@u($xvUcN*4n4=OZJh3l;w0)V$4-O6n
zKYDg871guya)45y;jGRwVIqJU0mu2EWH&hYkiMXYg(c4d+)(oW^nX>WCaWv|vwd=&
zHWw8Zi1SA5Tg95xy+{z7MEs1ASN}C=^#VdSC>&ydSOrzR@lA$?&k-~PK#HRoAJv3_
z^UmB{a4B5*2XSqn`onKJm>*{K5eM}DYf4m87LLzP^k80W`86*y!xXTFYuAF+)-p<;
z+UUJ;+OCMxBo3MEZ|@)cC6+70Eb--QNv@W*Hkfjkj^6bWo~#Cjvz+ngB5gfo=`YU~
zLP7x(s;{Sv?_y_RiBu=8WaeczCC7~qktm>pppg86f`f5hx0|ZWHwlBo!k({ZXBA|2
z%<l};FR_DsIDw3Ocx0GPQVw~RK6C?JZ1U&gjwV|L@Vg`>jcRC&08%L{<9@;RTv_=>
zB6&GC&oL@WRPcz=X!9R%d#}l@&YBdiy}bV)7R@Q&Io}nQkjD;&n-{7tAe|5^5aq)(
z!SnCpzi#&O!K+^V`?}L9a17CD_nfO!^h~1;4(yY0qWY{v<Gqv6gBc=h*^=R;OToye
zoFvUB(G9XKt*)HD0~}uNxMas&r#*r`E_qk$W|rnt4XZUT|1QNen9BN#pMM%0P*eZ?
zAw{TW(rfg8h!*ORjZPHK;V!$k=hokbTR?|x9!ogn<l^-tecjgqL&w>9O>zGdyprXa
zVvq-ok7+v7KMd~!6Uo*-n_V9YAi|i;Cwg4I(G=e+qKTBtwOI3ZVZEdTp{j?=)@iBs
zoB!UgZw&pfyGV(nUELD=&ph1V^4eH=Ixug%T`BuMa;vNsS-V_Pefj*&tgH6|hy|iG
z|Bl73T)B83)Ov10W#}OL!05Nrf55nU--@!_ulbN|buwywLJs8JI``R_>4KVK`=HPd
zS-h1U|CZS2k^onSk%?q@@Y(Kua=jMW<Rd2wOgcf=5-?R-ZI62bP9tj;7H}izs-85R
zN&1k(#R?l;=6P@B0}$5UME@DqT*b}z5Pn;V8wTgojXx(o^euNyJAb8vv&Tbw9fn-y
z%_Bajpp^j9LA@Ao67G?kA56bj)Zlr*E!Wq68f1N}S60sL>{NF3h1b&Df`n`ZO^hHq
zs9~W?f*KzrvTy`HV=@D!_S|6bm8cUp^7*q^f~dzog_Dnq<iGFh@1Dl|0L-+w_YHph
z%LVXqcjkM*y<*@ETu;X9HM~k7aQ!6eETyfTN4L1YGU9yrp#z2x<h;f+q5~Y1n}y(M
z7Z(jAW4DJ{PgdlVq4S~Sx9S%iX!i+=y$Y`@#SVOFiMBoZ@{Xp<3yalF;!JY!XZTXj
zo^5OvTI$x+)CjKLCX`lQ#a-Rh(bP0}RZoS+O7%1Nb=|~z%JOfArEdPMAX4G2+G5XH
zNXYT7r$Bjd1=W$7daQTKSBRK#vs)hw&cxy^(B>Lx-L*}kWg}iUvmGybrlh^zUZ3KV
zgza;K_c?KcqA8B^Ft7@(A+WV2>KdJJzrG)M{wHtPDT0=7b8ys1(#J9FyeCHEd3YHn
z)4!i9Y44G+R>S<BToWvjmU~v?&IU{e*7<Bqgl_->vE}gUV70S5Jm;m_^A+;koturm
zBYSvtvYv&}bg)Jxc!Vcx$FU?BdNKc}){Em@BLmgk#_-W24{tvGt#6Sr_^vJvnq{84
zur#Ix#F6$+xt`qwGUA;qeSVBR4sFY6-<i)5+xowb&p!5T8Hx0z@IO^h^~$(@s<55R
z%>|UU=n}HUBgDV4tm46oloTn-)W|p}{Qm!Ev7x1J%`CZXT)*D?RmV?$!eB84EHvh_
z@`{y>jVJPtlSIeg-@lg_6}PkhP82cdG}d6eu1%ch2yYVcb=?xB?IeR_FXm|7NRKTM
z&rb8}Fk%a-s(SQTL7uyk6To>LZH=pxotd5a4eJkWBu=HHMd@jxo*x!fMn}dV{IS8~
zs@Qh+r`tm`|0Wa!e9#6}q8?_s?QU-a%L`O~G(`6lNf=q_aNzIm@S-5<se*=eOfw-V
ziOi7e&!6Mwh;6yv%)-t4g5(O83Z<>Bt=Gf#;fArNx|VXE*SeD)yq>{_V7dNHth%ti
zQvur(E|TtL$QLg4YF>SrFse@};`T<Tg!^G}1vef3_NV#VfcIC|)Kry;`h@i@f0yce
zeE(A1kpA|-z|`StmoGqxh%a2&AIb<3>W61tP@?izLYpUheVi+O-w^Q$Jh@9m(NvXD
zlqc*xp(%v&DJA-BP?TgfQHr+r(_eqJZ5+(m?@yWAyAnvl!J{OR$BQMF79al0>N^>N
z{arhVS_`_|&Z(~#f2=&kNBy>o3A)9ncL*d$23kAsA~RHkT3XNHfQ!4Iyf@Q<^}?LF
zGH4#WjVG@wMcDoJQwvV&dTqk+E^8_7bHBzYe^q;aML8Gz*+5S7lh9|^eFPBJ_P|iQ
zPBrh{0#Uc95d6@9jRj!-Uw!6WP<f>h|MrpB(tb`Yy~=CjUu}K%a{N>rTj-3}b*)3e
zBNV<Tm<3E6!6Ps-e*ur!!q8MjL&J4zU=Qv<@Pzc$=e8aB>k&Sz7Ms#0IP>if5E_Bk
zhJw5`D|<}P#@gOB!pB{|k0(QT9qwW4?!$Jztq4<l%?(%rbm?i1dpDiRc_NcJo|vBX
z)Zm>R78c})G<KwR_kD^^+M*)tG$=zw%D8`(aSwynpyb+<qg}Zt^5W+xZjb4Un4`hy
zT2MeB^$Y{zYwbxBQ^VpavQ97B`uG_tMu^dcM1&p$cj?~Tb)byad5!sY+53&dgTWs5
z4}JDG+7%sL9Qua_)>aIzy_crX&Z0Y5?%ogc+_Twds#89#R#zW(r4m|vG_h)7DG*Ao
zu(aIO-MNEz)2O17ZV@s7z0QAWsQ7JybhmAOOVs_4fsYEIzQ^*wdd>Kj;TPyr?s&g}
znn~^g)cxdPq|)iiL_`axf58(;QqWQX6%V+hTpPpWy?bVcrdCGgQxk)`8zu+|s*sS7
z>b)8kmUd?smy_%79CQXj%1%j26v|c{9F-Qdvr>bX+9{_%wzA%cjEty>dh8^|x6zY@
zy05PN#eLpZ`$EJ%=y4&z-7TTlX@yx2TE(tLCS}_zBBKb8cAc_ssIcbcemcW&(2+f>
zYmHWYm|C~bM`i?8xH~^N0_kRt?TyAic6-|n_y_`W^uFo&?_XW==>dxWo_kC_tfGtS
z^o*?O`57c6cT*E$jchGjnD1nVJWbeugbWcB6bBV~GTCJnZSCKaJ*H(9u^MV%1Xytx
zfV;Y}#!~bZsFBH=i{eXHWa}Rl@6>qD(g(?5<A$uPtUyLqen}LnYb0-hE?3j|Lr(g0
zYp3qh;jRhYlqxky1xh`>Dpc|F`^EpTR!~q71@ZK(U4PCV3Y({_?0n^ou6*#WTuSw6
zD{u=T%Z8w7z-vpl!fX>p6fZ?u>dv<9d%MlaN7$BiN-E1oo~Z(Y`%EoD^xG@JBN|N1
z2wPo|CR53J6dG>vwU!~-?PE&y#IJ$dYIv3N?+-Bt(4$p{k`fam9k!S^|G8hzMD8)|
z$G)%WT3f`ba-H2=6KWbS{rv598`U~{_Y*CW53ve&_QMLlfUHeHb@EUD=Ph=a<0Ne?
z-6cKPbpGYBl@ooZcm0O7frZdgpEbD=xEo36!`zutw2YkjJvGUTML1(uos<;?qi#C^
zIrMMRZ(uZ<nEH&?IEJK#5j&Vq_p*1&2SYZk1xu%om5#Ngb)y&1Q`C8!osoai12Q;>
z^9T^6=^MD;b(^cgAVydDMQlPWz=rVbYHPvXI4(BU`}AkS=Woa79g=dtF-`@v`3ifx
zFCX++?pE|z&PPYZ-%n&!yfZ(@<3BJqHa0Rj8M+VWJC7;{9$p_<a~O%VKx`SsSf!+{
zIw|7Y@)Gawq@j_?5;=a<$7J`785`Akt_j&!<~AiJO%~>lJL0{?=bTR8ObVyEf(It9
zk@Zq{Vg=|VN=kzBH#qRoQ4sX@1qe`+mk$)(!^zLjt#{eU$p1iyAn%)sbc;!TlYi;_
z)oho$oB_H~M9flK8wPhlQi&UfKH)(*#j`<89uG_Fg^sWWVQV-$QN8iKj({hE0RJER
zTz<jPuC6ZH1|`Z-_>-`(ag`o-3AxxEtth+e1U!GCUT4;5^=`@ojwZmGs${>+EiZ#>
zncejeB1%%8rg6v1WYRIMH3Y%Na5OmLy)Fp%aKxXt8Ior%nu!O-WPr7eHCU2I%7HuJ
zw-Qc(v`AHKY%H8k|Ge4n&X2!r#7N)A-571<@`6l&M6Mpk^{jHp5(De7z?bj3`+W8~
zNwG2J$?<8a#p<1wNbuhD*PyX-k@WxhGf{dWAQ<VO<oX+;T&3DPZ(`#nEj;j^%*~Xm
zN(h?mEg-)gts1Q|_-NPe)m33ZZf9`gms>E82dm?X3U0|x#u|RQH{J*WqFMsRN2vce
zcoX-0VphCEKy<6h{ZdNZIkdj8V(5_|8!?~bDPdc2#nYai>qi?A(UJFV1%Y@24e2-Y
z15a|u?SuU>=TuzEi2Jfw_=-s`a7cLatiDwnLBJ3zPlJv4?}6MRL*z)I*5KGFR{GOW
zLU_$4L|<zNV7Bw0Pv>-x(&)ewyOQt|k}fOT@Yfe3Cg+cO+lJxD<ia?8bzf%@0)hgA
z!-^|Q?-%8-y`mcU_3J~j&!a?CbiO;;bAu%h?-ZgUP;`NvF=j-E&cdy<f#f#9CrNcA
zCn6tOUTWOe-5wM*`*sRGH@%972;!w{P0qT96>rPL($2$I34!#eNW1Ya1Yd&_98@M0
zmX?|6nQt_&Q^(zcm<+F;nUj-~tk>TQw2oxuwJ^Z@ysUeq6O4<fFP|+E5y9{i`NxoX
zqTW3$JUqU`is}gq8-~YucMIHX1|+@Rok5u1h51b3nygTktgf2rR=WUcFFW~^1lzQ=
zj2R>zaNplRp6RN(UM;mK{I+onN`)|ay7Vcpz#!dPgTwh(^|iGb#k|5S6(5!C5d_4<
z6IX31BSQo3<QF^h%dZhJcT?i!kGbgu39%Q~xQ67w3z6tAe=S@Yw0X6j9p~P<syQ)q
zjJ@h=&ry(k_8!RT08fNp!)o9{&gjvXlT#3)^n}vr^LwV}-=V<B$|>aaT3fK~u?E)U
zJpn=YKAm6vf7+SLxcO+JPw=P&hen5E6H|fSwzew5#lY_06R?XF(xg{cQCZ%75=BLJ
z>Au;$3Wc1y|9iLI%c%H>n6#Y2Q|sTTs0;qGO2fZK>cUT~URgtwL&^n%lH^lq1y$>)
z-&W1k(wdrP)$!om>W|X&`AhN67*Je%;T^|^+n8JCX1_9}OHbmo$^YH`?!ySAT9UJt
z78Vxzw<$u*2KYLtn@ER0CIICiA?1l_OOzWVjI<xxIg(QCoU+r(h#T621ztkXm#14D
zREA6Y%pB*aXp%os5HMxnpqhlo31M@TPt>q6fRpJqx53c7M$gP2>k+sDOYy5K6<j+?
z$yJzOn$P7&gL%uAGdblh$%heuUz^~|==HM&+!A5%!IxcKt@h}#u&_Zv`q6VlrYHxv
zK1HMulYcrsPT)X%KV4BSPtMK%bz9$Tn(4<x#1w5(r-y1RSmcwJEz?Tp%U*<>vP^Pn
zc7ynST1vW<hDrG0nfi>XIuV004J0Ch7K<QC<&>-k1u=cHC*;0+KE6VN{1cK94Qd9x
zx(<V~>JL<R$SwsQv7sXUV`HW-_bG%a5HTfL%dXzKpvPcg0aA+*YLtLml?uG%w9y@v
zn<5Tnl{-$ZIM_Gh<Ktr^<NT<3G5l{}-gxyO0jiJ5?|c2-&HZiT|6Cv2vqrCI)6-B8
zPz*QL<1HFzLgakJDCgY*pVM=q=9{;noqog&@MFvBX~7wS2x@mFCRPK&yM?<jEobh=
zfSf`Q#{agzPsr@LnVsgwvAqjq$D<*&ZfAxNwc}9oAn>m+c2i19icSojq?CdrJI~)d
z1x$yDIltUsTI7dEC%x1|TCsZ}LoHtk<sB8ed=Id0hA@EJHYPep-{aeJ`vVxKDp7MM
z@1)y@0|I1vZ+sf<rQf$;z>s<>O&Z>0Kew@Phw#ApsJ(|{!~FdyE;2)d|Iy$X#ZyNG
z1<=t^k!z@;qvK<dg;C&OH(6n>WuM$8Ao_Ac<*yYRPTL1m$i#q(C_OEs-Tw5L7}F01
z#WW<S2q#Ga{q3C^Wwmou$x16tlSf*jB3z3-<41Qtg~53M;awi{gR6b^LC~mq@!iKp
z7q_;4A|>ws1oO>%1i>e|Wn*q=SNj9JHPqDusLV<HzHA46t-ncz_y$yW#!q;L)ybGX
z4?7}sm7-sGf-7!^g;spxSNz8?AlBU6X7)o`=xC|+FSXcF{4$XDW#}Z|&&tX@pUhGj
zdJsVc5{8~0_GG&;eC+G*>euo<qaa^%a<`)bp=uJpjERjjl?1zsf>6?((@Av3vlYBg
z%~H}355lHXALh|OW|4bDB`oRvkfb1UaWR)g9&ODGe;C1TQqn(85wu^Kho&kc=C|@z
z)mQ3UsX}|gZcfuJtbF5w0y?^P^)cSsX><eyW8NVort5Qtd)m|mw>bbo)^a<oc$18V
z8Sb$_c{=z|0sq%g)w~(euzIYZfF^m1RD>vmjV0#ZOiO=b$OIB)coxPOfH?9M=}UjX
z1RJ@#xbK@BJ*;nH4M<79wJzw!*>^^ez_jO2rTIm;UWn^~ks#$(S8}6c>&3K`@R4Ax
z?eg;X=C{oY@T;D&yH95#h?o^Glry8FHySxr2m+|FF{mBY^{2&XZ$Y=rMvN6;$cBN2
zz~rPb&-PD+TLF^L-PDAJ{Fry2=EVomyC<68*}?vC!ZfUz$?CJf9@82DLE$5csG*UD
z$R$%5S64pDE4)tcK)?TfcE=x{UwM&{h7l9r)+&9;wTyOv-c}5jx~Rwx&WWF9f7jM1
zUFI?G=AF^_H5$-h1}i<8mD=;#+mvjZL9t|I$rGrl&d6kd7OPL^yM>A2SgWyv6RXwX
z@6T1!p}5_;Z72u4JTaV2S*-!rkE{fNi)x{DFTMO=)Sp~`{0+4!d1>kI%rx8IpsGnO
zHPMl9cx=wk=SCffBOb_xkBuXp`&)USgL!o%E-M{p9}KD7;V07*Dte0#{36BrkGyI>
z{O<8X5t@@eFWK)dIHg~Tvh5XrYqAkzcs2hp2)G;RhTzl1*lMgMMZSbpjZL>H!HF%2
zFVt*8hRyrSb=;FeS5{Wm^X1KD6|wE3rhVF<c8PNvTqFVHWJ#y$tuMK7LZ2qzpCeG!
z^61<1e)Hu{-`y`NOh%xpT3Etk=y7QvC9!if;?%S%lFRPZy@83#$H(R1^i8|Opcl}T
z%zc)R81!_kU`=G4Uof2g{p|phT6cwebSa-S{6#@_88Q4#oJ&p>t?V@EUmbof68yVf
zM}p1E(Ll}082?bY%I#WRI;FulNhqoF{*OZN>&(u`tmM@CG*Ne|3wB7Wq%Vf+a;v|;
zT}YdQ0XyzdQ++1H<6wn<02ra7_<=s$*@a%RP?GHD`M11`s6%NImJB$vGDnczuafxa
zO?FCb+fxVS#Sf;doSRl1H3lD+YUcO0dC6Qy`$4k8bIHt`+F`}_kc*AA=vhQG4vx=7
zy8mgVZ_~WR8X0{{i+4RU;lH04ng^Z~p{3DYD|x^9Y(&K_%QzpV^0r*;9th&2H{v%-
z=W#imy*>B)S&w1{K#GQ*a!X602uA}F(z4{)A(=1q`t@r#$l^03@X?9lV*~{LFx<-o
z*ad<Y1dBd1@!m%{e{@Z{1wk03-4<#Y*%?+1UU{8e8m+XI(m_J0ckt8Ko8*|tvxs;=
zK0o)(o@*b7Keg7y@>f>ZZQ3EeMIC+pY6|zqHwQdrCR8XRIWGM5Pp@l#B(>G61V{w}
zG!e1p^~zgC(}RUr7$OZmvug!wjdgD^)~o94-_TN1)3UuaOO#n%Ei1<nZNG*sIg^9~
zZWCq}LKwjP2%+U*Owxab>1`~D>p|wYhjsG$WNt1RpOIQbPj7Qp13x3veJ=JUW(2dE
zyit*!=?`nLx&NA;2`v|zLr{0$bnAw_OILR<GaE~BX>nO)>FEgDoc1o(O`?TG{35>F
zK}?Ewx>%V|k#D0`VNX%;@IZ~!a>zn~o<rE|veE0$XsL1E-r0VxFdYT#B$oJwo1SV0
z5))Aj9fx-2PywfbI7_D8_JL?aaLG$gi3bQZWD~^4^NDbg?o7=wL{__}@8MQF0!>>F
zJGNqm1q#Xo2ac6d-V4T&?q<&WG$*49K8SDr2W}v*{kChceg-C-e}3ucY9Ez{Of47Z
zRtFGB$-oI$gZ1W#)noI~i#C%ZDpCDMuC2y=ccf&8dQ9hcgr@?XxVY}7^->yl>+t4Q
z<tXo;g?qoDK@c2gb);Sd@0mV@A(a7%_wGdB@Mv*Hnb4yR^@b;&yMT<SL~(tZ3)opZ
z%O05Qf#O@-nfw15f5xT{q))zJg4_Su^>;Ll54u=c(EOU1)H+v`s}bbB3?{_9$rxcf
zm2pG0`RI)_r~x2*>%h4TQc@GwAYic_<sS7T&wZvzfQ8@|2a!IM>VZ-_F){eSm*VOF
zovFecHRrUzhZ|B*kQaWrZN7IJTk=5V!C;h+&Ez+avwNFp@Zu6UZ8o3+x>s5U*#Y{p
zAs+H%w{96`2r5H{wISJVk`yg3FK=B~uwJD_PDzxXU+u(UkEObLOn|F=PHv&HhN^Nz
z{LqM+ezl#VHo2<eb25H%JhN#O@qUG?nZ=OhoIIA0px}I+>Y2571QReV6)X+%rbr14
zwCe-6YfR+3!{%{v!juGe;>WLnK?5Iar(TSGcWD_3walPzxppn5;`tH|PH4NE_EYt$
zSGtf3XTpHyrAzO=X`l<`?5pJk8=ab;zy8<_m%O8=-C>MjL{n1Rf3E}u|EP$bpPw2*
z`arKDE}Vf)n`u>5(#iKtjm)zOa%_xDrZ_y6t_DnuK?-ZDQ}fSNpPym3KT~KDvgBPE
z0=*Lip8`4pKqv6M__e}OWRI36uP0JF&(<c;{*-{=IpEGZxj4+|9AIG|d|aTVV`%9}
zX_lm>#>&Y5DmJrv0;uCPq!t_V@Gv=}Ad`koO1sKQ+{j235FO<6*Y8rw;=R4CDl%FQ
z^J9w}HoR$M5phwTug`+ZGf875lvQFhRMllDX~kKtnLk%iNtVAQ=Itl}@|`ol#pZ-t
zF5cPBxMtIG((A|tRyYmlK<GVqYQkWXhMbv|vGVh0GoXvzo^xW5KB2pQd2aJ#1Ms?s
z*-E6s@=BV4yu(d@PyCahqnHY!`0TB&9-14R7L`_;RXrBWu%}Ie4Vg+|g}GH#05j{T
z>h_4~YpCktyhs5UPT@}Pz%vaZqFZ`^28DKIM+Th|Ce3Ute9ldyXQmOEwA^)-WoMJ}
za5~DZOW}!?h>VXO7#VD5UD=s}q+oa$^fZb9lFdArLJ4VlV<^lCb|qSgYf!Sf3uu*k
z)d~pKEG!zk^X|>KA4YtWW5+qTp?K&;AVl-QwM-PskF8lMxC0))Sd>$E^S_c{8Wjj4
zoF0c8U(^n8Fqd9ynwjSkxV=8bkXc<@Yc;e-M-@3!piKCrB<3q*fZxCH%ZZ1`1T51i
z*n5a&ETNi03>qg$(`8q4a+zXd;#gVQp<x2^`d?+;t*u#a7t~}u#yf0!Z{OYrvOld2
znQ4{`0pkZ)_**LmJKNi4mbsRexw6u2-$pk+_Srje4I~U@X=#6C$j$<3U8qMgmJp>9
z@l3<f3a~m%i#63)hWV&7@@kSKZq3Zo*{isAb;vCf5yr*Dty@^SSJzlr<^qEq(=W1+
z;I`_=?!)j;^XNHhR2+D$j?qvj&S!H_zcOcj90Zny^xZJajfjl|nxEJ28B;<~%mJj3
zLuoTO7{%D9lm9gzuJ*1zj{H2^ndRl|8#n9!Mjw&UfES#9<C7yutf+*L)<2HhnzC)5
zM<n5<x^@DE=u(`i2}8@*N~&*~A3lCvffGU)l(|PiaP}Yw$tJ0;r%K-ngm`5srC3>7
zE5A_pTAMIw7GvgOR@YN>i3^X9_NU+oD%;w=VI2MBz^;^&2hi1`uV16^v*AI71Y@?f
zGVbo|+!#RT&#B2|V36<Xg8l;qp%Q5g9+HlYC7hONb8>NkDR=E$;s;XXN0&m{kTic#
zCcm75d|b={ch?vH7Fc%%=7s{+oyOAFPOvm=sFgh+(i*_Rjbt5W=8mYrLBQp&xs5LT
z+J_eAzHWT1#y0^$UEO3E@)hXl@u-XpN$&0KsWPD^r*P2GscWjKsOj8n+CKZJ#L_7b
zW0XXa&>P{e9giASH!0NtUW2|uTKesB-%jKFcAKL9jF1<crqCkE8Eh1#udf`FklUIH
zwR!~CgvGT2d6;ghc2S3&os}vxo8T^SWYk#t$}N0q4zQ%sR@Hr?mQh_z{9+vf;-Usi
zCTjhbmf}`b8Ec-{<>wR#I?UaTN-W5#D9ouEZJKQP1%vx8Us$Qog}-n!!Sauc-7^V3
zlY)Ld6dUI!-IUM~&RauG7<mBGTU#|nj=sXxiCWj#r$f5zH77QfN)8-lkoDoW1p!y2
zd(4Wey6O_i-piLpy4F2wYh%#Sk_4n1iLf%a_k0v-`SrSc{p$+oW@riWs5z42Qs1F_
z!-KkUC<3?bI}lW=3QqvAF}HL4w+7FFSW87Etv0RXQyecHtLzgs<J|X2$@%-xfOmGX
zf>HwuO93(lP!Pb-prm#gAzriSgeXO629V6a561M%R&u5iwXrf%9fG>=mQggALenJ9
zuu*ZObEVr%c#9EJ`<&kGE1o-vgAvNYFKpM}H=wNsAzFR-$@$FAL#e%_s5mZprB{Q^
zAjp{=scTb1dE-hY0t@UW)(wEFBjvx$Y{R}GghZV<Vi`r7`rHzgUp07~*bv^N-{7T0
zOue@oCDO)_m1WY|A+KcbN<=^)Y`^r<+BzlLU-x-$d~25Jg9LrwCo-s<N9i>++Vb*g
zrKNj24;#GWo28__X>P24b-mIQBm?f`{9<FK1Yc50RI2|*S{UvKlWwRX31!;ne?QoG
zA0W$c{kP;zCQgsp;MysYWF{sbevEXdq$KSKg56DuByBAcNIxbq|045&?(C|&r>ntz
zZ%PhpH}A*%&K)Hxw^gbS3;c7q#34`wE6G2$w6dbb-#)U{s|Eu<&E}*?%4Q!z2;bh@
zUEPbfwJZB`MT>qdN1TJwE~mfo<)qs}_{v92^m8d$^5Q8c=l|6tObv<CjkFYtM`E%H
zLq<`1b?^J$zB7XwOyRlpUGt!4-b9M?8?B+`CX&w@@`wFV#3M?hFR*QuAmp}Lb{ED(
zbi#pGOg8uD;hlqz#bxNtM5=c^0W7(e$`m;NLoB8-a?{06vhB{M()T~)E4vGP7^xV=
z(tvch4`otxkKy*Nt`7KQyZq_K(`Qh`b#PD{uP~SKo)=pt*A?V{Py?IXV%0liV;}$X
z(|$^nK_hzMA3)W$s*jE?&X7rTI<`i~j`_cz=W|U>)_)g|7EU}G=Eddl6C6z3i~8&R
z4R`37|KOoC+5O3z=Iz;Q2!vpbb5FV}TIdpHL=v_@!usEjH}lspjQ@;#O`ee@X}vZ3
z1)-0xnAurvkM*KF{?|%0;A3M*W<H~5SOKHx71u8UiH0R+m-%tTt?AhuvRV;7_;6T2
z#j9KC)<%q0pY<;n0N(2~_!%iBtrg}K+(5E`W43?7I$hWDv-{7$luKKxB%0ph^744i
zUbxqLMF_9(uFRp5*4OZe3;(TTL!xnqkzBzM5$n$W^NItlUi(vcxa;&M?a1kIR|<oQ
z%B#hBvJXK7t>9t|&GpK1(pag{Ui9)mHrr@=!_PEyAGt0;(?`?s`}^m#{71Hj`=Txz
zXXE!~^{-<$5vlWBSx`KModPHYV|G?}8TG%b<w}}xVX?wmFYim%gXYsehDTD=Gua+)
zLOynMLfbi4r8Sn%+2-cw7j;rZ+m`<{PBz#i2c#cxqzDsJ9{<`}odm_t&fXqQdgRmf
z=WyX2vI;qF_~z_<+?vu>Rn^v*7S(At=6eNPK7GD5pGW7Oo~;2GbnGoYiQq@%u4IoF
z66ijAC+1^}K#Vq^h=YxiSNPknjU>@1Q>h?gsj022$vk<|H-7A6_gvKAr@Ib1o`Y~A
z|FNn_&hLJ<R;rl-qD)Ne8+^Rn;4>yVdS*ZpHlkl{W!NA=YR;@~lG%{q1%+nf=NSWy
z@$vBx1^HPT<Ay9uOp;K_gOGc(N4`A)@?E(yTEVk>B1?M}y4Rmw`*)3Z$|-BA)ta4D
z1Kcc`FX~Cr17k$6l*%hCsBu05Hu7prhs`)|!L@%tFiO<MsgXnpnmv3>A#_yYGoQNt
z)tAsZd>45q$AciD_m~)%FDDBbnpYuPU0FjN*W;i%E6b%K{5QA>1aP5~1i(1=G!7s0
zcmTs4(%Ayl<gih>ly8xmKc=Ll%)q4#U#7wLlZ30tYx+o!5>;GDBk9v8+aJmyftX!#
z3b+?f<T9VZ+&AE>2b>`wW)}OQpzt{A<m613h%f_2qr3Zw8nfRth3XqF2r0wc4BuCg
z+G*xL()|!GMP@}ve>+muul<Jmh*p}u7ku~R`}b(s_N`R`K=k(TA0m9?8y^*epB;k3
zU<z>!@D<k0%V`b!UM(<IN`#b%*Fve`;TmdEBnY)!CX7XrPtj5JR3z>qKiaw!vO^B6
zLuCj6TH6#{tAEHQmkHzzny$BP^JmsUwh9<hzE5C|IX8ciO9XxfLg;>tOm%1BsSP1v
zkL8|Rl8|3bzMH4PFDYr5sNVZF%>dgoJH)Fd9zo8}Juh7D?Y5&quwpyx%0E{=Wd{dN
z$FJSagbCppycc7-f4}ba;*xKAbR?w#Bi^H8kX-=Y6DsL4XHYr{U-9cZT(6a1TBfpc
z0?tRn01EQ@F}-~?2|Vu&8RJ{d&SmY!-p0SYKC#iul)QbUs~Om;tEkv;5%*zl`zHcH
zAMj%@QaLD5zoezD&BZ3A=<1>sruZUNwlufP-F__FbAo!lEQVh93E^3fLi)60@J6w>
z<jfPoS3l3zj>NaHaKN)%YQ3@#>;zKo2M9&^kjL_rOXAsFgr~i0d<1Ek|5qOG>R<Q!
z{6%)*Zhg@Z37RnSV<37NH1m|Ad%6wo=0>XEg{!QpT<yKet+M>|BL^}AxD>5k!!^J+
zS!}0k6_v$t!3|g4YkS3<UTh=1M&q5CUwq*t)c^N!-=7`xdk<Ai_Rmng86H-1aNLvp
zNe3hGK{LNMeOs2fc{t(=mtd{{p<O5d?_%S=U*Fs$&W2J?UcL)KTy1q~JufPPW><E?
zi~#JsuTwuuj*p7`IcYnE1?^euSwx!-;8TnFUy%O>-`s`kXeg^14Z)=lCZO26frx#<
z+?Xy`*OdEOl_~v1ZJ|BL5lATCu_~+vCV~j=hA4DpWYS-tN)}Fi@2a~@vAhGt5(y9p
z=y#dgJ2nM9cb(O^5nnm_&9Qpr$Knv8qbM)GJCV5#kcy&RiL0w~YNfIBG;yK6(7{Yc
zh+(sFY;-KYx0$PlhsP_B@igJw#jSK6BJ-js=v*UvK?**(RMe%fj7&savs|6JU#X^Z
zaf@46>t2BEufRW3_IonvfV~21P2m=`#&dPBZ>-3wzy)Z7vD9^$N>^7GK_HFKaclNx
z+U<<>q<E!1t#6(%hK#yB_vmh~(b&W(hRuC`&^>g;cZ1&l_ATPh?oNJ}Q&j(fo5*XT
zeXEl(c0kw=1n!0S8}lHuGqN^c;_tZjzKNNYIkmxaa=7KO-gB465*}=s^@EUXkZ#F_
z4|<$W^qF-t@enN<W%s=gXlQ*|tj(MM;2{@q3}Paqz*iil?aZ|6RQ)MCI}uSVTw~@&
z4`A9i>9PGCGMOHEjKdvnM_=nY|Fema1D-bqIVy%fkhODS-<d<kLRxLIzCrtN(h-7`
zzIyHU!~vkKMH3UxT44l%_jPplQL71WtKb-iF86z1Mki-i{u-s_m?1O2?C?#-+iqIp
zV<Ih=(|Wqa#`P(oCJ@PxoL{DIuyUY%gn|IHo*ll`;Q6-~U|{11K(m@Haqa;lUtdt@
z!%P(>O5LC4(U4=utw><tBfbzx9fy;*{YWUG+t%i2cpv+HleJ#MV0ZJ^tgPqqDs#IB
z;~eLH5z4mkk3YN76aGL3k+9q5QM$)L4L}(`M#fM{Q*6cBL&xzDM2pM<GDpE&Zxh%a
z*U;PX(K)oH9tw~vl#22eGBdNUuWc$SKCR?_n85Et3R|my93bEbLsVo_*t6A*&dC1H
zMNNf#;1B~ti{ww*sz>RM<@-NP)QH&3#nhALKWEeZ(bwlX)6R_eCg$!YlTJaFx|Wth
zQ=BLy3?g8MPz~A6KWFVwjZ6GX8QY?<MtNO$iJ_$)jEVvcq&Ijarn#;d6a(!AW+M`r
z&&bG<p65?a3^evgb~c;sx0bU)Ed>YSB2lad?p;Z?n~RIahno{WVVZ%yv$L}~=1PC2
zkP7m{axBK|-jDiv14h9P27NB~O1+NWq{9%lC%7eU>0bL=QIA`8%Z&sXy~X}DqTa!D
z(H)Ma?v)kgC)WaKtA!}oLt)&`&BH?w)I3sA0dvgQxC@HW=|5J117k8LK9eI2e&7>a
zuNVBBIExVFO(gNn|Ie_yw%8y6GbI3Z+P7c~2DkZ_KbLs=ve8NUpL4!c?1kef>7PP3
zNx^TA%gn*dK3(Fq^~Fo@=FO1s6ahSh={`FeXrGVgue<KyyMfW@)YzEI>*?M)!OABo
z#h=Ry3TkSS#HhbAF?Z|MT7XA9L|s}Hu+vW0lIO4U*-krJX-?=eBH+#D5Ww{o^5GYT
zb~1tpfQBq&ILLj&-rZVPI2y)Z?Ohz2LO#!7%gg@Gd15NKY;@dP_Z|gV*WjQ#)Ec4g
z)-78DRtO9;&#vXXsCjuKsV5Z{L{d*hY4Jq4b$ZQ_)%|R0=;<vL_D&RjgSP<b%6h7>
zeLn{^`oME+@LF&ou`GSP=#G)@+?Y%XKA^Hr+*nJ;fkQ2(9GU}D17C{ryw3YKq1T0z
zga3v3e;Bu~-5x{WGzqqho?;dtLL`pg9ajAyu;Jpp(o4>d@C@ua>$1N?N8YDfQl0tk
z3qW9t6Fcxd(&ki}85uYih9_F`pmzYqF&KUx;N3(+@?R7mFN`W|P5GdM>%ZIAyd3Iu
z(BL~lT@JQ+C<u7d{HZsY1Pffi5*RXJfm{_@#}`ds9y8-^@~DttV`G#W{@sOSBU2Nf
z#+St*OFe2T5zBL^kTVRo=0u&Thnr&J>|cK8RzEb4ilihGAe3R=82Cw-4s+9^&51dF
zuheKTRgaZ6JGNxU;lbfi|HQA5sI3L9$!Lk8x`EuIk`VxfqZzAWVl}E^KA)E-`@!Vg
zziFu_wXh)b&DZz)&gJjS%C-afbYwvJWo&6ZE9_`!shyPZ_PA+09yXTIJb~?mw;f0}
zAd%|YySEMdc$A%E*8<*N4|tnI5<0chc3Qu<_v(zBhp%Y#JB8opdSVMBYqmDQFx~G~
zx?ukh7`R<Weh21;bGM^L@sq;iH6(PbK*|Dqw93NdgtgTGCgM8N`?YQ--Ce#0S3j;4
zhVNd?s=_K?vtaqnk-DZ0Kw3$C;o&dRy@YNT0P)uED1mgMxs}m8H*d|%y!&HKMoyJU
ze(2&M`#O+6ug0D|ICL2a!*zjNu!8(tVxn919`h3N7BUhl@_b1>ZEXs72vRDIc|Z>2
zyton5Q3x{^sE$VRH9Z$wQ-I)U?0qTtyRWj0Ga?H-9Lhj)gN@+<LwD~wl5j4~azkwp
zffZIPwLFxC5d>glcuY#tCOjc6QPGhd-MzE(^M5xs);2evt3-sK5RMPUe1R(_37BPY
z-IaTG)Iw4h8UjHNDPWJ(DJNWe;WPFL5H;BCQ>DaQBn1FX|K>;mm^|1W&UCR%On}m=
zLZ07<=zto*j#M;EL6ZsvW%p87J3LhYI1-S*^>?z$WPHillF#}&vN2i3z$>!40e%3!
zV7<FFRXl3B^b=T6FpeTfz8i4?sOd}7m)|{mOpzb2fs`|{1bNH_IRymSA(K_!A`I+k
z1Kf*CcdBdnUzzyWyL<u);m=|=Z4kCWAB>N)lo2%VT*iGx`Hx{*PJZD9-<nxsVj_P-
zoiJm;dFvJKe^0Axi}591N8(|dfQUFS2_(**3Fy{3ure*g<^~T|Jhi+?o#1yo_XTDi
zxUx)znLQc%_+;T~qTPo8XaB&AyK(=66%1d16AYwEH5QYYRNNJwiU#j9aTp)YF9Yq!
z5^HGdrf<_r4R-BEJ{3y#k5}KZ^8M#%QW0l~m7*mhCxdb;jH@61u-KRw8V#7jBEFDT
zOUpuji0>!piG|_M#m93K+l@hN=}v~q?PwLKG_x@H@;(8}ttbJjy1F_hmJm#c5V0p}
zZ#{rUc7lOUcaJ8BImb`!rizWB>FwLZkLh<YRUnF-fv5Z+{lFP3IQTI*fj{}x+9`Cr
zDCtN*_kxre8$(r1<-v!6)pd*ab?#`0uOUfs%<A8<M{@_-Y#4NMnZJgIeL7e<$&3)h
z_YZ#FKX{5hBBrY92!0ji8d~>CCr3IZ#Px6yyQ>{XsTWYaLkWY9{{Dsvv!7b$dga!z
zx;p+RI$m^VXJ<%WYY0E-4xq;L`@U{wKw@KS>snSB85QG8y)sxn?$2x%VM>iDk%l|W
z#|^%v#WY`wbAse(T}nrhf8w4YnQe(!#`&qWkt)HQZB$$&4|x9F=6EdqxVY+_IQ{J$
zHI`R`11(bFsX`hW8h8*tGYXe)I?p4v!KR*TIrD3CgoLNLb?N?HOn{04_prR890dXS
zjfi&`mHcrA)lV_~Wp!i1m+d1OlSc3e_n5;vjBlziPwMJ|`T=mU_C!<!EMw6TAdYvp
zcg}A7plTWWxsmC^x!b&Whw$k~Zua{$r2oW)&p&i!j+9b)u3~0mxV?>k>XvsQ6ldDX
zMP&q?>b0C0;Pikgw<>|8F*!7t^!*$mEl?xMxa6n#C4c=YC-<5jfqRdAf7p_@`O=01
z@_qwF?(+yo_n2a-T&K8(-Sn=7VR!d|1h>{q$aT1A&=4k;&gDDAG(?o52h!O6QBkT@
zWlag1fcq>EDXVC0F7fYKJ7GKA@H1@oeK#<cn4}>#FHkI#F3Y7g9wD3BW4blwYCKc{
zrS0Pz_R`nf!t93}r^Vo|lS13PL$bDE7Jhuc>H`%-SVSeJ-ax)_@^~Lzs=j)aO&tps
z{gE-rrZ0b$M!#Sqj+N&Jzm%6nZ-v+p<X*J7jYURo+S`_wUT6}2O_y2c`;(KCSA!AU
z_~!?o*k|8(%KZ<-@n!~0o~q8|*oFJU$OxLOl5UlKo83MWtAZ-4<Io->o)Q!TRhcF~
zzEaDfzk$v}V2n)eMrmE0?E>mT-fKrkEC`8@kN4Pmbk_zV>wyjp3=}Xr&My-D(VtV8
zW8&;G*wS3aGt8k^V`6OfOk?_)&KN;6VZdz6FYMkOgV!L5^7gI=8{TiGPbKBo$jLCU
z_Zr{Gf3cb_QB;>^URD72UJ`i@tH8-AOI4(proRS};BH*iql~^6wncGEivCtY_u*Qi
z7W3m9+N^tReV&l72toiBcGj)6wabBlTR}oF^cop?zR<_J^jjT2?MAhLAUGfIrv!Va
zp)MfU?irI6>py@dUX(04H86GSF0uV`V{`@YJ%^VQHI7(qM+M(A)FR}?=fcG}-+u4H
ziG&O>j%2>Q%}}h+(5K!gG~uZS$qv_3)p_5ptRmR1uCD$f)#M0F!u<@t*W_XC7f;{H
z4*5G#{pS9?QwF|<>E1Vn-_uH}OLD9AIRHKaVSAW^A`)N#m2thd5*n>kdO5ex0kctA
zh4+!?Wjdxx4j_RjBWiNb)L-P-Vlp;|q7FPQt33J|N6pI}^Fs`c7KYh(yKX&-djRh;
z#V0>L0>7`LXZ>;~RDjToTbzhN+M3;2@i7!+E4ntlStgRg>2y&LAt9B5j(jaBkabd@
zSZz3|QdBcE42sEkTb`MtFBS-l&+?=rbH88besfbDt1FDR@4S6y-qWM6f6wb_geC=B
zFGj%UBHM8Bv0E=WwfGPCX>8w``5{Q}7?+lIBr|)_SL|@l(w7}VXkL@xvJaywp!qCM
zq!r~^e=ce=ul>sB%Z=Tr@32Q&$3cvHJuEp<^i^xZ{B;9NaDuLOJ2KQ4jMSHxR_~b}
zy7hmUWwPO#+40E<Sn(Q?NY#3i-=-wJP3dNQ^=T<f#v)g~c;_+Nl;8Q0@Z7;TNc8bE
zQS9$YY;;s*e`qn_{D-Gn{J`h*=M}(+O<_xzFf4nVkPu{OX#ZHpGb%h@*23LX&s(`q
z9rKcRWhs0R1f&B4_Q%JgAP1qDnw#CEWfuSRz)SCy#+^txCo(>lKXsLyB!HKlA6mlE
zT%3WfE2}U+Q#|Bal+xG5(M=VXuBxHIhc2FMyj5GP--H1*1n{@0guF%eM?m}x&0Rp#
z%X04Fc=p5^_t7-ieKcCRM3$AFNzzydBDsy169?t`)t};Uu^;Q}7DamXAK$-U2J&J%
zLvLe@7`1^%SPo}p0Z)JI3lTkh^rt%sc;XQ;8L#z;h|rsK(X6wh26x{!6M}ObZ&FHf
z;-&6%C^3fruUDy{YS^TC9Uy)^q$Ox=jE~Ss0c(DEy)tlUl*4aeth4RRXU9E~;Um8n
z3-SBCWDz_+Y5`qvrQYS(s(v>B5}~DERb7(#;bupWJ2cnDYc)i3pOBcStE1y<9*cg%
zff&qNei=Veb==#I`utF9{P?VSaq-DU_~6)BE8jCgc5Dp9SBjRVRGc;vI8}QZ+T;&`
zf(25ZkKHcfEIe#;&P8$sZ)j%kVB93bY3%pKQ}B|Uk0f{njZ0BhZmmIB9~2C{HW%+k
z(ar`N*Nh*GpSAGHi2wUJ=_1W4BEW9Ui)SALxP){ltG%Q1BU!lea}{6eub_@a0PSz&
zWV?S)KJ%HKCfhq&@bQq7Hw0$2mNp=1v)}%_b9{AI_*<Usy<7iL!PH)8k|RIP!ar5*
zi}X&7w073!R^MZ%>*^hQ35Dh#kypI8gI;g7whXl$;+aslWSShhJh$I@%v0`6g8dwJ
z?OP*RH1A_<ApT!G;A-T&8y~T;VH|$`cxhR`*h##hnfv0i3ezx;bD(?sq%$gZ)AZAQ
ze-t#I)38vb<o<CUyf@!qki9_vLR0PKE2G$`xJKg&xObAs$cYJw95=OqnE|4-fGCyV
z9@FK;J20J-xh$(JKOJ+Hh=O@znkxj-VJ-JaMN#1#HtLJ{)`DTcUE;2Q&tTi8Q8XPK
zh>3=DvgX3AeDBfW2MIkz)6H&k{p=&z5xM016=s)gh!|*bbR>3R0GPbN*?s8p;svmd
zAb6;>c%$d}!*5^xdw3DC2>G0^`4Vg#_nsyr|IpZ|iW|e5l~T{(z<ruLb3DPfU6UTD
zAO~z-8Tt$rrh76*5lpo{H2nIzPq43R4IN97vp!t-7~E(8=*O)inN69yz~TF3E4SM5
z4;B)goU#!K<A$6s-r?-sA5ajrU@>WFh6lyb|FofY-)GOvapkN$(+bM3MM+@#H+CnB
zNIzDl#IBzJ1ty%-hZQ3Tf`U@A_xk#iPxP|zt*^5C_O8Kq7zFNz3ILZcMOp;}`LUJa
zgFp0|PiQW<5#m#yiPo)ij<}Qjg{bhM7AYK=RazeisW{#U@<)7OdjW^n-pD$_fze}p
zY)pXu7?L9RIGZ~Mt!mSCHJdHN76+EUXB1aFm!iFuunp9UU5C1<Dl}nMbX0G^2WshW
zql>>7)%TO2lX`nqzoGc(ZbeDf=JNQzOIciewwLp-*~h<s$6YA?)OigJzH<bH78<g$
zwk7}sJZTf6f2frw8zw)Xa{=Bg3SxN6_7G4b;|l0Ro!a6K@N9I!=681I`}h%MyoQ>?
z7<854*lusyW&!HdldW4Gtf|NnK*ZopziKm8G&EM@-bD*(WNGg@EMJ|G0F0`xfav!2
zBN#auMJd?Z3MHa6JOrkd(t#b?KcYdS;*XF&iy((oA}=6N9(jGi>wd7?vX;vtBm@kx
zwEytv6;o<FFMrKO%N~MICozevF8IS-DlY@Q&~En*8fZ9aA_G@RM*%&CWh&&d<_WWa
z@cTot>#lt%LQXr~1GVQ8Mw0%}e-q2Ff|~T(ACKU(Fkzs!LbYM7Rk{0aTH~*8y1(vA
zhc*1kHh_2Pt%tOfm2Y{2MH4ZUC@U{Eg<o4d(z@z=N09JB9~eV7LlnrX5<SW)DsEs6
zY);nh^WRUE_(HO>?yymCNFm~8cf_kslv%->_0L79eZ=p25y39zCFu_n>`;YluyvA>
zlH&fQR$(<AqT_!0(`Rw;n){emS{f~^HUTX%QGsDhVAp_Fwe97%pFbsRr|OFGbM?UI
zE|OpwroL5yWbqN$O`HkcP{N<P*1QIu)Wolq&Z>DBKmoyAc}v7j;@uTK)bpt+pnFie
z2?+vpf}<SN^bc502OdtV${yard|UYcxOxkys=BUicmoO|DUwoxf&vnPfPjQF5(3gG
zB_Q1;p>(Gph;)M>f^<kpw{&+(O4mPq-|u<8Z~SMRaSw;yoU=J+?X~8-<`r9(j@;yb
z8AV}|(Ip#Nr2C9A`BO$}R)@J2-c8IaONm=n)65Bi;0fm6CL+_>Rpq+CLKQA(1CAa@
zKN_vJL|xz^7Ivk}O2b3`Kjd+8p6%VBzrBGQEVSG#`0($}B{+*dxx;+g)Vdec0fy4l
z{4R1L+7918b09)Ga_Apk1E7NZgz0o8>1VJy2Egd9>p;_)lcST9=Q+jIPJ3|5thL?k
z^o=+<XhcAF^myXa<$FD?Uohh6(Wxo(IA900#W2@$ep@*VvJ&(AsRgLVt1tc*Uzl_K
z?(Z)?=%vXGIa82VDoD@V1-+x++!3{b>Lcx)%}l`~O*x9`;MN8FjqG#`r{b|qATB`C
zll#u`2`SozrfTn-H|}H>DI}EpCx5Cx=>;gg(sWv%Tmo3}*Snx}OvTLmbeE%*c$N0E
zZSDMgWjTyYdNyhROhXLI{N%@v7nwCR14BmV=K^bsz{TyB`K49w%*a5eEG<%-TY0^G
z?mD_lVRcoN&C^YQG^Ra6mX_$^Nkau}PDv8%-6!1CT|P#g9~so7{%63Qa*j=^sW~xH
z<R4@6FB}*s1Mbpb(&Cp81iUXZGcy%O7J}9o@1XB#_@ToM`aqntO+(+=-CN%<UHDdU
z-bgE&iiSA9J4z^x@%iYTpHxU3C<)2m%Ug!`<w(xB^i3S)Sm54m$;*4_;N)}^?I;8g
z6_8#MSzXw$k1lYK2>k3OHZ5iCeTxf^P_Ra}YfJ!u58}_uA5Ylr@A2q{*gQ>qUx&-e
zY6YHy@&u8R(o&!i1bhYl=T{7bbaW^1T3Rvmpmhv|J++$|K?DSJ)J=0YSzY+CLK;@r
zOcZis!0!4(Bb5P}hlc=}DKIS3dJ0)|2&ClXvT`b}^;1X%3?}neJa+GinvY7&^MN57
z82Y}D?zG$A|42ym^!5#;Z&-=s7I9udVf)%g|E^wE8oAzX*4fEU!!o<`PfE%$z)k?4
z0#|Xz855Ya^$Y2e__r9nS^4YmM?qNukooge6H#!106H3=d<>>*?jMa3*55+N9PBU!
zxUo4o*3b1G$0zH`$Yi8^QwtDZSfYb{3BYDhgacO?)E4!N%h3U%z$5}k?dOcX!ncG7
z{yuC2Hj*^>@dIBb0H%Z;l17Hh5X%7nE}l+wSggAKFESUlhJQj$k@SZw3}OQjM`M`Z
z;e!&v9C+WlaEs|v-S6RHnn5N;T64A+MP-F7Kn&8B&78rFiz2i#F&Cz-M+Mu<v=23p
z;xHs+mDSY(hR4Ta!6{_AAH=yp!9R1D>g#Kh(gAA+8d_1X7oIaW=xHW~wGtB>d;_b!
z<At%|^3MKlVBkS`&KC$!YeAC(j$rcKOX$@e$VI^7it%xF7ctNfSDERz(P6z1Z<VHn
zku&ELE?GxAGBdME@C24CQH&z6{XRhXNagwSTYKg~fkDuyh3RW(D_3TeCb~H?I&gjj
zJC#L8PUR&~=o*6OF%c6}exxdkFYR4M+K^A9PzoMlJskY~yDL&fMEfHPqaxgqQa-J%
ztxU{xC!k|Jd8P3pszsmXy>rD~(eyPFvwwrVkb=6G)PjP^?~_=9pW7<_hYNr`+q|gM
zGlBPBjwD_2+EdB{ZCuPMHEnkh*Ed|*@}g=1adA&(^A(@#sK|*|mE<!qJs`n+agcuF
zrlPmbvsp#pL{;s#QJ>OF)*$gS#zCEIt`yHf^ny#dA;nVmZnp;d$$QNI=`%V!gMplF
zvnN2!zmpvw*zaU@!a&)rdp#|^m}<pO^o<zVb5Me*eY#^;7n**K69PUXmG<v@C+V87
z+(X$FA0-F|voPB3jFLSfb_1#jggog~dKhQ;3ZLLMK^j2ur?F{RTS<{r4R~JbCjVhm
zh{x)!n)oG;tg#x8rg|z|3ctQl2c2kdxi>rEH#PI0`)>H;cN{I*V{SDA^#eGJmoK+b
z5iltOG5oUug4A<i$Qo0Y{gN26D2eqS{@;f^zWwmN{x0<j5SE=A#+t>{`&dN_rJnQN
zy|MR+tGuLSV<P<U;nh3va~!dle%~~QV3+h0olgNf5FkhK|A(1dALHp0mQ|Z>kRc}Y
zoQ$jjBSQRS#hWqiiZ3QsHT3^6EvV=5w7DFg!2@N{9a-y)Oi~0mvt(=3$7$g4`{cJW
zrjKsg$8`B?@W(Tvy!>2~oP4ll+S@r7x`y$H7<g6Lg?W?QtY?xmtPk#qriYdi3i<&9
zGoWoIF8a9MD<<RG_vowTKsx-LJL`GQ=QF8NKnT6$!arlXyT1=UEBeLtEA@}MyRn~`
z-1%?k$hvfNp8Pjl7KPZ%*nnqr2r7|rjrb#`--AQMw}|hk7ejPH9CH%)-&25l%}C^e
zZg^frm_F4=evx_}!3N$aMcPltimZ3f#{l@Lcbr=+kN<`)@<Q=^t=0_2Tw|q{OS>Y~
zfui$8Uz!DS@*cK#NFFeO^VU#B@$mn+KtPdW^JVtwfa1Bz=!80}_S2Q-Iewr7d1$i&
zXmX)MF8pL1CP}%3?~|=2n@vh%{qF3Mdp7_N8no>PU#(9^Jpdp`-j>vqz8Lq7mVCQE
zgGWK7sy)Frp!k0en4T2gJ9i&2aTxm@y;iL?W(E!Cw(3*QVS6BN?0o)i-v0~`RO%`^
zIM8>aPkZOaWhc1JPxk%CZS2T^1ir{`)?Y)H4?n11d?_G!w^!=+4^B?>ik++YkgF4j
zYrMQ%SOK<)RSQV*Rld$7jD0RTRrPAoFSpPLZo=6S{&h~ow*xxkaO!DqwbD?kOeYvA
zd7Nf32<&@{Tzybe+6`n_c;HfAewDP9KQ*-()Uh5Ji~3)P*l9VtU=@(Mz@9KTVQ~Jz
zD-D@Y;(F~a`~u>hXziKTZy1e$!rATU(!f#&1^GGoyRBoA(7&nWU05&x<d`2BMblv)
z`{#z8Ss9pFiHY3+Qp(p3-`|`#*Y05D!OSB8$d=V1P8BXkw@A$FC~%Ff%#5D8Dsfly
z?R8gI_!wqPP7U$!-;)*>@0e`<R+qXvHy8(p9pVM#=dHrqK0YSF%e|;bOl+*bj|6Rb
z*WoU0Q~=_I_c)ouHQZQ~*5(kK6$fVsehEJVaTJ7#-+gSLk&T;^0ROJPpJ%uJ7Q2U*
ztc8X7JAydO-%6@>@oZtDv>xr<o^vF>K`n$tM1Oqz_|666-q-MQ|GOXe4li2&Ibs~-
z5(fEKV`(e>eWlfiKloZb;WO+(BivkmhhbQ-G1@se0*^#hMWq(|#&i03;ai`$jJ~I(
z)nuPFpZ*d@e!^FFOAHeaFFgB96l53lCH&9kWi`Z!2f_gVtkn+BZ125KrSeJ{#Jn7w
zRD(l9{BC|ENJ^&fdqal70r^)EaacFqmzH>>#MF!H>gF0heon-+zhf*xzYlbCbl*B*
z#e2+vi;2h-R8RN-<!V(w4d$PC$u$ojK3X0)t5F{e{S}hdpEEEJg?NFp$X-y=&)cvD
zpnnR%!2!P|YWScg2R_6wF?8_p@xb1_zlQ}A^9p05H2G^a*%g8=TPq7oG>o+83@kKY
zYG8Q%!2U<en*Oawg~3{Wpj|OC<`qs=`A>q*c5d}G5M^G-c4A|OZ*FcvC-?F*?@6Yv
zyv4RiYHHzIa}(>0t$WB%B|PqCZ90L+JdYVZ4)Wo>XX}}Ej7f4dRz3P$-k-?720~Q+
zdrUx)MuZfFCxJQzv+bvy>&xs(aiWZmSX{Qq12$z?cJOadZQUkXG~`ozG~@NU#~jR3
zn4j(8u*ZX-4cOZN^Fg&>W}0*BhO;u!5>RqwKJ&&Whi|WkR=44UPUp+7v8lQvHC4@5
zFC{kM`v!e|fX^#_tz5#lE@Deq^(lV8#5016xnh~I#orvpfi;1ck8iZ1;)QpkOo%uZ
z7S?s&nt92m9+RV4%gaED#vuO?7VDQ-AfwVl0!{*K*XnohI(pid1l9rBIL7c*9{EjM
zt)wh=lIe+W+x|LcOO0Hb0yi!;l8At~ou9t6AfyyliH^>|hf0nyb4Llc)-M<@>w1{+
zdecry;ETnSVxZu>2|Ua1Y@8gAuOT&VSG!HEquX`(j7<5UmTRJ&mPus8bs@jkl{6Q+
z7!BS_c5e<<`GNP8yt$nAkac8iXpNe1_=uQ~yc2FKQf&HV19SVX2O09f`>#po-#Z)r
z;t$-PHmj!oQ$uzrI=Tj@i2U>bTkmOpAuT26Xh$C@S%=a9{0~Y>%4Gf9KLtri{iHjX
zaB%F!HSqt-KAef2o9N%!-PL*OuB5I&^YlqtYU=K3=AfiAAwgh;-T^xZ{ggQFaJ*eE
zW%%&}G=9{q;TE5VJ<cgeh%WTDWVXRrgT~?0pRO^E8se_qcy2RSNQQGUB0@*w<yC`D
z4|as7{H`*#3JNOoRJgA}vj9X63Gv$C;FsfNE??}n%jYq}49cz4v-1i_4_MxmzMn3%
zAdlQTJUlEcWa0V@<=pSSzGf4Pr6)t9zu}LICRX`bd0h-XdKIauf&N0T*)$?G?{Zy3
zOe_p6HL5>!^@c0K@yS)7K0MGkOPiNecvyK=>@=*WV6=_qnf;m+Y-Xfyi4i`3E<^~f
zFVzl1+7myjHAj9~7JrP2h(TZ5vh>qizr1|P#uv<+Qn=yd=_VK+-n%sE*6Z@JhiBWp
z*DdT16C(>IF3u4JzwLrmeTp@a5uKoJvc+f5&bBU=hm2#80^l2u3H$*cv}#Z8$L#M&
zc6DA#be0(N-1oZG<Zr1{&&#=B0HZmp{Mg+_uq0tk=99a9%jJYuQcdj`?2ZTs=8>OE
zkkg{kc3)B2jH!dtYOdUB<&A%XV?*lP5a$o$FzUXMq33fl^jJgz$wj@~<#hi*kGJOL
zTkHCc?4%FZVq;Zwq&2}7W_+mg54svS!R<C;;N5YCVSi)VysFiQKj@<rW94W+s>lZE
z&t&e5{Qh0De}sc(@zwWX42(jt<e-g<m!vH#A5<84G@NYFHPNa4>J|SZ+VSZg$4)=a
zhkZaVpC2>$;c&pJi$O-VN-0?Jd1O{F_{PWem5maM!STuS52JX2OmUA$*PdZd0iiFM
z&+If7r#>N`bY+=nM<XSAA>y{g;hUw!Ivm#(_C}kDtQo*-zPf*Enk_aE-`7uCS;@~D
z_!FCx`QvMGvCDB^gUa%o^IWT$@0KBFIVpWRF{L42KM%XQRHRw6;&Im|xhg3|938|5
z6vEc`l5$E_O^M)~MHp`6`S^5qm6`Ya6W9k1z?T6FF5Ev;mfSzleP4cN2hX2kJ`Ku0
z#YyQXLds(QA}L?I2;v=20Q1+c`T0)v->BD~h0uz|bC@OM2jBkO%~R@Bo#mHMQ?vPK
zh5@<z;K7FH+7s{o1FomPg>;Xga;eGS<Q#esO6@kkqXUwH&B>ZV7J3BGlp^gPA>~Fd
zm14>q556BB;-aYc4GzU`aJx4o#;58<!e?pxl45g^s6S-w)uP+8-+;L6ihusMW;Ck7
z^8DU2c31dtEcnyA<6al5fleENZ{at%Q^45S(`m3unU2l~G9Wfa3arVOLQqM%kIsz7
zgfqUVSK2p}GTc5bNOzjso)QnmPgS(AD1h?Vd9J&gIW)9;tB?hZN6ve<C#xDJyB4*Z
z&%)Jt&KxamDD=Y{jpdS`8LVe~s=xyeTaK)kI&8-1H;uoK!p|YID;pxq`))>hmQ}BN
zLQd@7h_?Q^eZ$89!;uD0x3RHtMUH}yfFLL&2+~IP(p)@-W3N!BBt#6bJp4#-|Iec*
zd|~2`Ng@R$@_OVl-JA;&A;nZR?e=_ZBwi+AVPk57A3Ofp`$&aigsrEO|H-+#oitOR
zWa4A@y{!ug$VEU&fg?AZLxS~e1_<}q;;PbS69IE!CCjV}stuE#$MzRROocl2V7dZ{
zK-Hf+0UWy?aQ_}N#4&JnTPc{TgSl=%Gi?58R%;Zv*%hUm=X9Qk$@c0~7em(fLO<Cy
z4OGHZ6obObhMOMry0#o{E@>jx<ZI2;5U4>d;8NyXr>iEaWc!wv^%*B~dPeEwRQFQD
zf-y47Wh-jwf%0AcnBePI@Jinf4g_^RFul)W9Udk&&4DX81&>WxX#pxifj*(Gsh(kE
z`1WlM__~E3?dIoO1%$o?H@dZwFfeXUs&)3Rt95d4fce~BBNkRL7$YDEeB1QD{HO|9
zoT$-Do$Q~zJ2wZg?K#by%XQy#g<4ngsNKi;S!FErEPA7T4Ts+5{b<11!b^U}!QQvg
z%cy{~&>6u~9Nm@D-nFPI2l9_1BZ-OA3e@XA3QqPqb)<Wdp!DjZ2|=>Xg4^YrtY5jN
zWE8k7vhpT+3W3|drKCM8)AQ30Hdv*vXJ0Ok<zp`VdIvn(zpJ|dOB&8U-`;%Y`uf4m
z#AS|M2rh2xdcc2gzu^xPnt8#zz`OoyNN&~iHD7dJA2hG3ypGcL+kG%j|B@G^D6F^#
zX+6if@oRibjdY&BUF~f5llg0)JW<h5x%}I*Vfu<*tJcH6bBU3GadcvgMdOmmb;&~h
zA%cC!q=`DbEY>gG`{K7c7D%@dAq_?S@tD9~!7<1`f5wN6#Sy(1l$Xt_gC-=b7!;eu
zyZ#9EX5X)lcWbO05tZtSwi+1P5z4pxn^OBl3y)_G4{fRWJ5I(r%fW>6&iY$86@^ou
zxt@LyrF99Ra#<Ufjl0zpp@H#_UrI?zfR6Zz(Y@R{$$!X1bCv)gI1fiQCBI9OaT6*M
zKgPr3a@}B3BGp1SJ<&fd=Hq@yeVvjtfb1CINtWhsg94>Vc?~0u)SQy^6zW$QXu&t7
zIpg<ipEV*W)g2MN(d4A8?9e?i>%@^I>+$#R^?DC{8<0h;j^gF$ppUE7;@l)kRdlBh
z)2w@A@T>I^Far$1ef_fMqV>bZLs$l<zXN;jQQRho9y90i6@EUKlLp<L8$Xo!m2*eh
zQVgaLeH7Kh=hZbnK>OsJ|IRe{s=5>!zkT}_R(e)y;kdMR=bW_(EiK;>#g-vknt@=5
zEmdJrK4o&m=GJ=GoHc}dLH^gPK|%D%`kqVf#0W83f(`Nn!X=yX$*2E$2>Mu!^pldW
zVPPtOx4i$QlClCl^a*|a&+%|WcS!l!*%?(-#K#+^EywZ>kd(Bv`!p<DPrCF{zH%ZU
zU}g`@$WQ<H@e3A08h5tSUZTUru28&7`L|~-OhkBE>?!YP`Hvr-l7yuWZEf;+P2U2i
z4f-_16ouh1P}h<v_*(m=X=(Y;M22_Oo))Yy39hC^?6&!De&U?MSTp(M8dm4B@_u$D
zI7k}UJ3vKZevuB~R;1LFL&K6SA0L_Ib2P-BS35u$9nrS<dg>=WO&_<J_{`xz;NAS{
zx=F1wWt&@4{nmXtIu&MJpG!iL+nnsbNhp+UUsUxPTG5GJr=bZ-$YXyw?uoFFpm{+y
z@X<{WVg9Xmz|-a~FNJh%I8ZaQjI4PkPfx{5t22Il4ee|}>ZZ~q<Sd;k9s1S`*~v_q
zLU0biO3dYYfZnIb)H;L`hnABWO?z>fBy4|gwVe{Iry===1dm;mw!clNzOeAQRPMwL
z-*$a6^7S}~lgGTNU1ePu`cX_+v4hlUq~=3oIO;_P#K=P62<>$o#L#f5K&{H_1<~c-
zE-UOR)nkQX>Nhf<-~DOK{Z!!t$AKwXynE~b=nBKBMu~{XlOtka<b1L8MVAz!-6yk_
ziMc*^FWG8NjDv-Vm(YFM=u%MBx$unj^vDqhhuS<|49%>oOU){?Y_ei%X>QJGY(#|U
zoyb}Av<kfCJP-F|i(<C;vDK=qd;Xk-KgK5C;5qG&CUIX*gRqKnUq<eFAqAbw{Dhk?
z#Ry0#{)Ibna(+DaS!~N8Z8+OJz3<O$?a&uF7FHNW>?>^J<UqxvPl^P{=g#ZvGj;^k
z;J7CH_P01w6P4S~C0bPDj^ljWlK4_LTEB4w$*#Z}{T5)7<U7(i@_xfp)6_^WHsTY-
zr9f6*72ee&E!^_bnzZN!a3b(;T3r`jr*UzqY@8=Bj26yzdh@R9R5ek*XU;Z1Z^N1r
z=fCG1-u#5>_%+nx%pV5^DqN@g@E<Ft=Rbo-R$fv?#Y9*;+CE}o5J@h=yhe*GE)tiP
zI%@7^8ySKDT$&k+A14>r!HW-_O9Ia4=7}q-5l*)t+BY0k=f<TxBuG|jS&A<w>vpCu
z%tvZo!)|g$U%&Xfxc9@SCRnY^x-hen;d6IvZ0t>(aKiKR`4f+H2_6ucW&*<dQo?Ov
zT$x|2Vd+Sx;dZPjElTp@MLQb6HAb<5gK)Wpb?@u9-frBv)_jkK#$)2@oyq7~cCYTp
z(1?{r`KbiqY@Cp~$#I5@Q243e5vfM6-D4L2&IVVi<DT_PJ6$y0XJ9j2lNJR22SL(j
zVR^=wL3*Rf9<LuCD{ch^Nn-^EyGR+nSwa=k)O0gZYhG38fd(pFD-n2251%TazW=#o
z#ihl9;A7Tu^AH4F(+`*7<oxl<n8lic*V!v{DCy5<$U_{q7^iXeYy~o@PA#S*lx!Pw
z|H%Bz6Gyw6@w-t%-iV)Gj<1256HXehu&r%{7B;egg@M4n%~ho70E3UNO$zKLbi6x1
zgJXkjG&M8?wWs|hM45G;S$TyT*-7Vr|E@1@;w;D*z#2X}!m~j~fa#O;Dt<uljcokm
znRc5tggk%k6@3jZHjs_NFC~UF8K-}?sAeNOkD2_37WV0Acrr6HWv!o)qQh}2B_%~q
zpM5AHVRsqj`C9T*dedxwAq!}vr^?QdJwrkQ;`oHfwZ&DozTvl9<}*ydx^ySKdzZ@c
zrNb2i2IFJYVyBsN8npcxY;ZefE-Gc`J{|io`=7`8)#8N16L45kA$|icWF|?nw{N97
zIi_i<k5!7bWcJ8~-TUId&vjQxS(Sq{Bx*o$u@$xod0*fDR%PVpvFgOT8rR#UuY0h`
z&9FFLEqvv}yFtuQp8|nXnZ|}RL*Pz~j+8nH?dWfi<I0`bUjXeN(22*xgQ4yEf^H|v
zgZx;vq+IqpaMSktk(f!_>Adq{y`lL^^8pP*1Cuc`jswfEzot*ZxW*Ed2sX2<!)4VV
zcsxs?fp7_KpG!_R&y&vafka_?YF@`XO>6{_IT<-S<%yhwFRMswlXEm~1_uE#B=R8W
z#>c#zySi#h*fVyxa`iJ18~)bHibTmPSwu{XEdDvjviED&Q(kQTP+PM7_p3vJoh<v$
zwp;>l)Af>9UPFWZ;ZqswG#E<`C@MnR)y5Z<mF4Q-fb&)u4te9yw7aYl_Z@Yk;fM$%
z++(4I%9tTR@@MVFuU~#-sN6RL1A~AU2u}d9-A<ywe!cZBC60d+O>{K|nV8GTR!L^P
zd=Kin1r%Qm+q8YJ_4{i(i_T0pvEDw#jXPUyw`0XCUydD)EAl?I6aQ3G^T=0NoL^N;
zEVtJ?@hJsv?gJ;l`ZA~}Ln9*ay&Ie_8Fx<dg&@O@@6uDN%C)369qsf(Up>8=mg27^
zxDA4A_is8M3}x#YBW-P6mlg(|;TpGZ6EQzx1lNY$dKagL$@u68A!59D?#N~o7q87l
z8$&H<M`~VfI>p$)o4~4UCSnksC4R&Oxz?jXFS+k=R40GG(1N?)<&GZ3e;$eTaaKoU
z6vSlAbPSJsdsP*M8@rlU=f~2|3uTh)ocZ+<{mWv1O{6$I1fHFsCsD<_-FFtuIAo@4
zV6v|}yDxT{iI?-Yr)Q-I&FQj^=HdN<UNJc+{LJ4qR&g@L5E8+&%<|k~Wt;r8OVUjG
z^Y^HM+2?3DIDh?c$$VlX^B6mpvwp$zNy3dPE{>D^CR1s%4apU~(<)wwojk7!_gX2U
z^uxBL5_mE@mM<^}l-+Xj*Y}w2@lV<W46<R|^i4}ExIcSRkXXYaJ_2o=ej2E54?&>s
zUbqT^?bcATuc?K3898i5^G{d%b!%M7_023`iOXtfnfD1z@LiIikE*4rKGm|6!1yPQ
z(N%37AJ9@3BST<*?d#*S`3L<WBcr)SeO0=5VL?Fv*JX+#L`$I{_MxT1P;JAt!<!)x
zVPa<awo6xDw<Qi+4@!Ivi#=zM5K3VNW(F^FJ%J3zH$wF!Rkfd=n{BZ9H$eRfKPmoB
zCrwcL=f3LgUoLQJMMmH|kuOnLUsXk=2`5W+ed2|5s^=owLld1ByMwA9>_~RbhZ5Ca
z$%f+4(cOPX?;=>`uI%$dtULl`yof2T1`~V9lm~{3@+Z=g0bpW~ZblpV>9eZ5lGb<q
z_IJP?<kz+koXJID)k#+kdhr4RT?vU6LqoB%<nFxO<T-ZfuTifqw}J|efG88E^anbs
z(#=WHbhu@AdWc$S4xc0hfSdFCBwQ<}mo}zjkVm$(XoxYcQCVFHS@rcw&rul*%gTDX
zn$^^$`X_%oC-lPc1wMIT7xDVR!RFG$d$a;=3Eti6yjxxo5q-}Db!{FOH+Z8be2%{x
zu2l1Zh=gQ8JKaM`OKY9OCNVK_6IO~u7^(PXe?Np&*c=44d}kkkGhYN9C#NtM_~s(M
zBjdxkh`}>0kne4xtKBQA(oOqX%m4I=KG06y<}vQse&0}G6?C*#O|EPbF!s&P%KKn@
zUzq6m2vBa3yG-{kR(q7Q^>E=g%L|GsP8BGU1C6Rcr&ExRPep0zZ-kslry@SLxH9+g
z-Z#rP`i7QoL&U>bl{Y^B#D{Bd*}vFCA2mOAFf!Aes2AUO?ghrd+u`UK|9Pyuk_4kq
z4-bzEmnX9)`5+^oPR2&RcPKLdE!7*kadQhJn}r^<%V;3QGr>#2Hz>yP;1(t*rqXCj
zjAiyI_`es{c|bG0icgHN^^M5NNkO>4)sw05D&Cpg=tGKU!^x&LGUgvZ9rUbzjYY0$
zfSnyY{Cv2co0yq`fhl2JEQ%NGuBnB02r2xCnKO-+uF=}09g({6X_Aqx9Vrx(`c+m}
zIlEM7XeKz>_hn|@|H|AgD;x4*Ml3}sN^w9u+}4q5&crY!8S)7B+kU>bCtKeDU^Q*%
z%j@~uQR3L-yn2Zbf?EWv7w_}7J$rax1}Gy6&js!mxaG(A`2lyw0ixFYn@&E(^Z%N1
zfTqQqYh!5QHI!SrHOW!&AJf5Dccyp5((548j2o<jXs284`_=%>U@#>!nAFaG8J|*=
zZ8zWL3$xK_&GuXvViw->!y6hIZ13m+^NO4zD{vP};<5PN^tu|PTD}sZ;kIOQ^pUWO
zj(k<Ov@iwLhk}B-k%cMzxyO(lT#|{qkI3}}`zC(_MfGN`TXS=B)T@E(ae;YxwlFg;
z;s|FT3@Qv$X{(l;HmwVQhCp03ZU&p$iWLf@U;hY<uWlI_c;~Dhvxvtf6*|gWlgh@>
zGcko~CEmQtK-b;hFzv8-15VtIytk!JQkv2n_1|t(rV!ssg+U-Xx_|1jwuXj)L&tIZ
zLhYiVk<%u>l8O$^T>w|df446H9PrJ#E;+7s6{=UDGo|uB0{jI$Ql6*k)rJ}OboKm6
zaq3AFXqo8r7QKI2XD4`NBWUsG4}kC^meUqn#jRul=50PUNyD+rkBFs}V*J)=??3v`
zG$-2P0rQ!pjTJtVlx_%-+dfd^&&c$IK%Rs6Cvpl3bK%mL9WQP^gp@V)7cXw^pTGOJ
zdqRzVwS%dIs+(4uevgUIYCUFmFiQqn+rG1$Q9eAX^i6mT3VlKAwDLH(tHF2resjyc
zbp4iQQ6nRxh<Zla_YpBsV6QVeIS>`6Yhs>5y~h#-A<KCWef{d=Vxm$Mb9Z6Io05r(
zpF)0*s*?Ou*Y$VO()n@zX24aIHp|b5@c|c-v(<{1*A3X?0gYy6wg^W(brd%+&tYYK
zKpq$IcY;;g;wy8nc14iP`$RvUln+MiCpQnZ@5gn@cD7OQHhauhJVr)Al9iVBWD%dA
zhGA%6(!Qt5G-)fn&2S*eyVXUa%oCHhw&Yum;k<KEst7vJc<AnD0u%$u%=~qb#grkG
z?Ua?}dCjW=FYx<!cDpxn<oq3#-*S$D3;O}*2(nhv(iWVv!a$~KVL{1Lc|w&g5oSXT
z>u!3wpN5(F-vZP(Bqbmo8*7)m`n1GhGije3*hGQe>MmL3afWh78vCu_!~bvrK!YTK
zP0edo6+Y2+3zESk<dVP>2HBa&GlEGf9FH(Od6V@EzGK7(N?slw^^i_k$#p6{Y@CDJ
z(TW@E<ewg?$V_5@N46}^4_HRz>l@H`U{Up566Yu}Zr>>BZCtii<gmTAdU@koT1-Yo
zV|@c11a_blnT<jN6`t+#e~X03W1ss;KPvKrqpK3$CKut0hmhX7b7y?T4w)@~?E(O}
zU~<0HLM?pok94QX#kyYli{GxaS?k(DYZY+#^?qOXpV3x3|CLR&F&LO%nyX)m$mcd*
zf|&)sOWjMS9)!)Kz#oL9lP&e<=DAl)fVii6W4L#9n3}((y%jjZnp|mjh1uJB5-tMz
z+$DqOtqzZ?stRY^e-W*ENr}jg8T|g{S=Uz=yHYg0CHRg&n^p6PQ<cyX;eR`}nVwv`
zxIUYoh_%Uf8O@=id);2|m!5qbNOi!m`<BdQ)cC>3%%kj)JBLR>tBGnHpr#cc&@ooB
zIfe%<O7o=^Y)m}~pFF*KM(P(z$>G{lU>crO_pSIzcE5@Ho$8Bic@QR5XEglhadPYY
z<%UUfOS6x!Z`}F!LNJ9k?i}FG%VT^_qxaG+(S3Gy2k@Lp&%ZG%eCR9GLbA$qk7yZZ
zT|CbWz~tq%<Y;f~X+4A>^G#}NgIzn!2zzxZ##h3=T~UGtLDS9Or((31_}YIDnk(|F
zLEZ$g?ty1AvzRB3`_@1&G0KW(OxfY#ZDW>mbB0UJ#9Hfv{G^wiZ8Dv0=VOXj<p78d
zP0S`Ux0DOr6;gJfEKL8;FVJu@adK$)#rl69m)_sT5(k3vl9{jn_XE8Wx%9Vx0%t6k
zBt}QZZe1~T5_c~N?vVe_C9P1We6QLv4)*f?Kac&Z2&<WmMF4)5|F)mg#H&|=Vd<X*
z=2jZW?YS)9a(4}Gc>}4%sSDaCmA047ak!eMZ^5?K?26&K|G(GRPtME_mdbV-FrgLH
z#Q*Qoyi$Ev9VV{vLQzh|@odYxCZg=n(gE`y)En}GC^j;7S+yD{JrIk-^*J=G3I6to
z<-HKd0r*_^C_L}}v$K9{X1W*FJ3A#t1A^Jh0&}>10Tvy!3j^X@o1yKiR~!<vm3jXQ
zccI>m3{d7gs2X4Y|6M~|?=0wUUiQYv#zZ(1fphfS?c2XXZ{11_8ZeRePzUEgOo&R}
zDq-VjV1oe+8X|XPb21jpjr=>?mPb)gBZY3?ir?fb`kr@3Y-xV60;+f#nvMzMBb1c1
zyth4#A6JL_(W+%aaeQ+dUnbQ-Ao5GA0*J2G94>mpuCB!RBe&9JqYw>T^N_oAbd?63
zymVqMhhdHicmA{mmQWL1aZU%={QEL5vtJjF8huo*2>YKhWyKR}G@wL}kMoJ<G=gKZ
z661lfv=4mVq30KuCJ{qj5e<&}u>}bNZZ=T!LCt1wcCwChYujK;0NeT?VRN06jKSoe
zH(%e7*`7~siJKS<S1aFK471SR=I?#ghx}Z3(A27>$ME;NqT>*qucQ1d(lpbnUTT{#
z(+G6S84vX!OWD6vgX7~|4#OW@f>kM>+IHxDo9boxpA9I{^D)uKAA*`b)L<p5#S+)K
zbokxs!h_3VVqtdk1TS%YXnJ}#SM1OG^lXYm{(3|Ivu<Em0VoWf=9lvtb*|ptCB{lJ
z=W)|-P0a1B4=L6j?@ExgaKrln{@f2g9G>ZX)&B~^xEK6YZg_-@2-JTKw4AZ)YZ&8z
znvMDkr3X*t71cdUkNEmB=MC~ZF0lu-Ir&zrt0`BTT@JTNDVAq9D?K`UJOr*<={(_e
z1X)fzl^H?o0dH4xHrpQFBPV7w)a4@W{)XjKz2V30L^@3;%HxS>SdPIqm1<7Ws|VRg
zpMln<?%-v8GEcAy?^CxlPpc2YCswlin$6YVnSQZW9jtN=??iS)vnw50q*LC~Ft)FK
zd}f8e_G%-j$hX~)W)9ebMVhjpeP{maDP1+gaSbyiGZR+E(ss}v|2=KTCyX60ijt?7
zE-H^^ciVE*{$ee9?o(AsLNNDtwliaPgOnmBR@S43v%WbGaS*^?1YE62;$<MFNn2Y+
zo06FctK@IEI+S2DGeZBnm^z_UgMZW0OHF2O3#VmgbX*voTGfCyvcJJ&eQ_7)i8Ivu
z)^L}vzpWbA5ah)=3r~E+%vA5O)}<@_7#q2q9@>NO4S+>p{&fuj`72*p0r!(W?<AaY
z)PXd`+>$J)w(88-54}xH%%6e5e)Y}~sb8dI1|Y#IEZRRO+%6YZbuOP%N84T=Vj|E`
zLUH(IgIilIM_1c+72nNBObm^Giu%hJO(*K4YG)j6CI-f5Rdxpw*~jH<23rGB1MK#)
z^wA(bR(TX}1&8%Yu(r(d?|7^%IVwTVo)DdK-qpYDb#@S@$pgc)<S&9=NnFEMOUWNI
zdA6@E4Xn6@UMLPUkM2kquZ(wB12}Q$SZ4P|wzWcw{bZY3xv_IPyn-7Yk*lWAcvG@t
zvsEJ1`vyg*Z_z7K<KDHJii3+J{$hXQgk!Cvn!euB%Rk2_r=_K3{5PF-<kcoOhR?_9
z{!spfpo*B-kDiCx5S<N%KSV~EtYViIyjUS;M<%=h_vZ`NULjsW=fSxcetMv6E&EY6
zS#cPrGdFX0AH3O}scj*wIr9B<jMVf!3G6euIn|;oZH4C^5PRS$t#B@1myZUUFEfl@
zwtI)m91)>H9bP`NhY_7~muU5?H`K{YkqYb68>|_StRk#cQ=GeYnVu~y9Aoi$Bq%uW
zS2LZ~VZE}F)@iS#@BeJpS9k=0O4^!5xrKWr*H4Q3L`65z#|;hN)H_p>-Xek>&x6u}
znK&4HuJ1=m($XsZ8+z*(j}p?{)S!Yc<T3VbqD-5L$$<e|F8{9u84p?L(UHGBaxDHE
zudP;ITeTKVuj{j`%Mb^p7SyEW2j`}Lm0J9K#mQM-T3QOcwA(kXp{EDS&|`8f7_+Ox
zK+Ih0LsS&%ca*h*o9~4nPZ`=EQCcY=Y+g;lz74@<?C}M;sp9t!EiBJZE!v<~fB?9M
zadVI8nI5s9#?5f))><p7OGU;JN4GOF72y+QGB7Zv2yp>s42}}p4>pu_P}oVejfjW<
zeuA#^91kvJ`1f#NVBldwvjM4!j>6dU>|l8@paZ&f*WDri3>rW>Iy&gT&$<ZXI;V<P
zto3d4zDK2{d%L$)2Y`<a&_BBOzMvvszJP0rNajt*V1r#PEDV>7%rq&&oU0ntb@}<y
z+OkAH@IlY@IX54ewdMl@N+iU@%b8GY4~&i$WSu%t2GP^9xAl~jlyGc`{BuEq=(Q0*
z7*zcLbMeyy;u$mvQd01>CXU?rgE2miS4uShYBfbwKI@oCQYq$;e_zdkb4X7FgPV)f
z_+8h!LvV{e2*BYpmX&p25A;EWsD;nnsfxas*My~Dk|47-Bmmna`JFoqgx~Xm*ga8D
zh1}dEjFf*w;ETNy`SO6HqbT6U$EG>(JA}_?PvcE86r3t5iXEUP>*`UjwmDm<9=U(N
zS>btLfIJ9`<2o}3`A~#pWC$~fu4r<j8(y===gjPOL{E>Akr8$37Xv-rbxIiY#Bhbu
z($WwQl!}kt#LK&S4fVNF3{Z<zl(iva<sl8tWw*n76AQEaA}b|@E$tgQDLJ%C5JRY<
zGd4Ds3ob0o%q{tZuM9-s8H9hGzqgYwl)jPtq5ORiHxS;u?r>YvOd6h-vA+zzgGSm{
zn3$NjFsEQWWe_p~opHbqL_>r8bAoaVk(42O(a4RqTLjQJU?aaOs^6OCpdl~ikX3#5
z&JIQpI*Wt~p9G&@qkSGD4wXH`+rZVz{%}e^CA_1)D66o;{@J(aCYq_?E#B(rD@$1k
z3o*CMsGPK*Pqu<)K21x6WrXD%6Al@r_kHPH<|w;xxqf9uni3LZeyhx>%+FLUrDM(v
z4WZ|>QFhUA`mAR+Mp=N%{{9i;eT^X<?I8t=`vyYkeRU<>DS8<rpN5&A`9lEwuVw4s
z%2&3Xrk~IdnE!j`rQ49fCN11VD>&4j(!*)nY>lKSx~YnSBF9fJk1?5Hv=^7YZ$Dj#
zKKg8ILqT($b2XLPB&J%fE}5BNX8WFZLmTU(!n{Jg;;|fWEKE$yT~1-t_kBY{$~p>k
zG%WNpgRN(x5O*S>ng2cS)Ig;Nth7vMf72XJY3d~;Xo4@isA!If{?DsnV~lpVQ`Zy5
z6DQxnc66V{YT$#3g++d0Zl+=`92+$^`pq%Aslf;WVi$aUeF1km<KRA)tCN=4SFf)T
zj4Ug2#BR#!);P|UeoY_k(sFf3bTt9R=d{iiRA$Mu3$5qP;#mJvy=Du7rQcph!qpM)
z9Uh)B{t514o<tl?$A9}=wT%-5qi#2*J8tqXvW;&23y8JeUGL^QqJ{$k$DK#FbiN!y
z7Nh--07~WqV*$juT%}+-;>+i7{plKyAN~`P_U>1DKTo7Dp^t{e=yt+;0|~+Y+w<jF
z$Qx%+s#R`!Vl}Y=%nF#>h}F!IJ@pUMKa42)g6_LFPx64X<Ur1M9eFk>4g{oxBcTW0
z#jaD^Fls^A23(HAm^c2#)G%F^L!#J+SRI&he;)qI&do`zC0f44kLu;(Q3t0}6Px+M
zc6GEO%!I!?+s#*<`khFtrpEIYJp@zHW!68j{x67dT@F^J^U3Vueh&|UT;<Pd`#w0!
z0O?_9Pb~VE8w!M?fpe&{vnXN0%BY&==4g$axcHra8at3?CB#Ms(mrPYJyW%-DvRR)
zv@WxW$|GwTE!3(dwo%K8tRHRj$I^==#B}u81cA+ha@1hvMW!#fv$I>m%sfRu4}+<j
z&X=z6UUZp80BH$FCGvj9gWMH6r!Bq@!&JhA9rM>q-05A=eztaPX4V%#aSf|=b8#$~
z0#`y&c)l_Io+N|aw%*V9{Wu=2!m5N080}n^zk~tb)C3pm&+332m@3s_Vr!Sbx&B^A
z?PTK@B*NZ1bDFHYx;FH^sJ9uAThyOL8|8a1O#iGKwTaLolNEMXzy4sb$c6EhTOwZf
zT6>?loeWQOy8l^Uni=%=<HIG}dU{`g#^s=@{#|A5)#}>Ejq?7v3@EHOh6_la^r3@-
zjE9Vnpw6{re&?uPs@)V}>o7V#?A}1Uy5CSgb37gT{Q9q$hKjgrcL$vbL)-GdG5pb<
zEPikcI)s;G{>YD;!1;#mHMHT;d;IB#`~Eo8HA_T<s6V^bqDgv_r>N%u*M$qwD^_En
zy3I!x5|sBA?z=TR;*k;V$tqVl=~wALzINN4+tb`!nwmfmwR#t0FeP3m%+cMP_+j*^
z={0h^G|~+O2gK{m0%8rpL-+Nx-J1@lF#6R{kl5ucDAA(h;x-J~CV$D@N>y~SnyGL<
zN-aBdq@e-t9qx$<oa5zn=L4?ezWzZOV(I5PJ!9>WSHnpc5>2$}o-^O*=>=|`D9&>F
zp-ISrt4eC?v9uK^g<x^PIEUg%A2IStX82n27=T(}3erE+KVE9#1Qy_~wk=U7uO5U;
zj1-}X;8O7~EzL2~(%M4xj`Yc^*4iIFU5~<}66AB(BJghlrsAG8`9DsF1jgA#wFgN=
zu6Jus&7U#Z5~EfD`v&s2ThE;5Jr`KU-RT)<9z&&vx$q1HSl*RxFX^HsJrf1K8@1L2
z!B#z5Zr2HT?e)0Sj~`B|ze^JPACx_9)|$RwnCB)fM_yvF-09o9uWN^b=B1Eu&(c*#
z3IC7Cy~Lpij{T*^<^)0{1;2gckIJ^~<`%<7V|kKnrTwIfr=`K=Sd26>n<4Ob>USIx
zG4V-oVd(?jkksne6Jn2S2`YMmfL-N4+;VOIoY9j68Fiwn_bo=9fo-B^=o`|W)h$}C
z;>&eS4<gm$@qOp{90qE}>P&E`DeiczpBpWnDEYkj!frDO%+`8|%nd$c8*O*As^uR{
zO6t&i1UOTpAz>n$Jc;!McA(UDT{`_TJj}I9B@>%Y!ngW2w0>JoM(#S|q)M{tuSHTg
zS#iWRyXH*}F3mMR2Ul45Jx=$!OWdH(5a1x$X>6KC-%;|Y5kKa7NT^+)UTFyU@^lmA
zF#!{XkvSjn0#|LhqqPy-89fpIN9t_kP2c!2!fJ7`U2c=|Hc#U=eX}DenyS4by}(^x
zdrLi9OXPm_nqA#L?288_=?8Qv9`$%#PxYXLzm>rwRL?xnk^W-pzJk}6n`E$)iL?Zt
zKD8Z9OD(LLE7t}44V1fd%mzc_%v{67(1VYaT$w!ZMnSBSrbLpzcsvx;!u*W|1=a17
zn<OeQ8kC{MRCsvd@)W`grpMNJzfVl|jJCZ#S9@vC-|rSumxPOe<Vd^PMipc{x-Xx>
zy?PSuZ<vvJH8=3{A6Rm7q^P2BQ6hHie>YkZHgr&7j^defot5<N&}V^>1o91XSU)t*
z?;7yKUPVr7-y0tHC=ueo#s5Vf??$&HJEs<Gfs&x22(744bF{o1Bq>5ksmeP1e#MOX
z{@ZT3Zq}Jh&DAjj5^-F2+k@Dx8YOu71jQ@u!}*p4xs%mIKrG!e$YHd6t6%Ecq7UOu
ziaTMBLjUHs9S0NR33tj1llUO}TEt)>mhy~u^~0nw5EzMw;*g(E;O5|a%>3Qh@TxAn
z_r<vvPPf)3i^WOxK@>Q0YZJzKA!1cwy4GR=rg;Ld#tqpKz>plOdYh#dTJ=dC4;z*L
zPHoQ?a{%WN&`yC02N#&qz^3Nrc3!4c<#><z-oLC%xPpmNM`cI3X|Nvb{|c_bEzNW?
zNAbryhaa!o8H=|xQwv`B&!Wmm_g+J!?Ov&qks+ieX6~RhCz7r<UNI4NSh)mt%5+7j
zZ4$3bjdv|D3`xr@jv7FNw<NusEa(!@G?&Qhlv`Fv1#wga2;s*Vzxwd$Qp=B#sU9#<
zq>d{>6M~!@u&JvgEt@O<=u!7-=9*vQ2xQRQKpt_C%XUZlz-sO3Xr~k$Hdw!JrIWA8
z=Cll-u%L2ncQVId*YBtIAF(fLi^6CVu=9WIB4lGZ*!~UQX<G$=4K%`My3gDXo1fE1
z|A5N{Dm!vFx(%`Zv#brKUYtyEh>*PGQ^j}Jz)$KOB_Lr9SMzJP{f{CfGvH?UXBhMI
z$A%MXQvLAY5RZfXOciG*aDuokr+kUsR?|<RT(B(c*?)6}`V(sSeg!J`f$4ro^P)JS
z32d5!5oJ!UM+19W_IB^VXHY`pA#zO=^7R~Cn=FU_##DdIQGO1!U``DT_6wJDSxiMo
z#XffITHVja4L`7#mc0rf#L0QcOBRyPy;Z2j3p|J$(e0ZhYd2H!^EaB;uw;nMVCw}J
zQ%HXIxng+>a_7k#;URnW)SM1`&J$D)b1)73w-&$u@|L_`Rb7b)?Nr|&Bsil<{aQQK
zCa<e1frXsEk(sKsNDuuX#`;)ZHy|X)!u;)}d()lkE|t)MEq=@T@#D`XwJI1Lt`h~5
z^d40ou_33ACuR4j{@pHyR!b`z-a#a$DKH^Q6mXLI#Q5&@(Qqvvmw^|!zY%jjfXD)f
zDx?qo0rWR5mJ#EqBe<wD14H9<lc9z*K~`t&ejTIC9pf$H8ugPWS0BqSKOGHD%W)@B
zDD7Fgv{||TDx3Wi)SM6#>OFkFH;9lXcr%PYMq}bnc;kx(3M9<q>4n7lM>6_X9+clE
zrAH_cHHzV!vjqXA2io?0>J@h#5_U7}7~;YoBwj~uMq(plp)PoKNjhZijuGE2+>IBq
z0w`xD-)Lw~3|mfvaZcIWlQj7L+u9^`Yi%ApqP>a0%pa=C$?54jgU+gS49E61$+MX%
z|GLNTI5vhK&A@G#z3c<*SwP>QaSY#_NzKao@D~5puE6r~8Wwb(fT$<)xdgNuT1>eh
z@$xk|H)Unyh6a>i0?O?x+|$Xr`3j+AmPLeaRs%eiG0w!CFCP(io8&fdMgQI95bhx8
zZWDF%>G4(}^4h@Sy~bTd-u;SAPcs<q6!nTC70ye)vMtT73EjnaW+X*!1c%f)9D#@F
zhpU!B7}f!AfknGEFbNym<7nODTS;^}{Cz9lA9Peba0_h1h;3bs1lGcJ#NZy<eHuk;
zw4d>Ug+R3dM=Ak3^DEX);@DW_f{9WzoUbNoZSCDNZnK|&_z#R{`jCB5k-_lg>H&zE
z1Z5hbXIr8e({(evd+7^h1-f7Fvc9}euuh(UOTmMQfV(buUkMou$Ksfgh7|q{>4IgA
z2uYc`n5wrd?YF|P)hHtqhNmhjI%}JY0pssMWc;}sat$!NDBW;>l$LGr+3QawEWvbz
z)DNeO+IoGGF2>UM!2(zr>)nD(dTrnsK%(i~kum)PCNs-7jaI|~|9}tgd!OJOCH^dX
zf--S=g#oXK<|T*jJw0}E^1#*O!t}{41fatTm!-rkzdKj=+C(A2p<tvC?+HpoWTZ~D
zD^H~4qem>2?%TIC#Gx{6U^CxIL4o-;%3U7L3#beUio?Z%s}dMB?m*`pMk?@k5X1aY
zH%ZSIOr&@h^RW#kaDw5YS#_?X{M?KmEPcWWPkI-(lbNcx#3%v#wiH&n-@hkASe%n@
zi<#E`c?DgB+?QumP2b8JQi(nefg}O&Z{W`Wh-LwJn{%mCV$KtKSGQ2=jfOWRx)|ty
z&VY6p5U1*Il89Qd;rpD276Z;jKu0&27XZ)$;xe9a{(v+|v(FI`5%7ICQTnI!?l~Ob
z*ieIy)*Ki(N477Haby8i_<DV?TD5CsUtBgnh=-B83=AkpT+!6LksZ+JqoN3&+`5B6
z?4N~tJv#HNTcm!JlO-SJwn5Bziy-Y&0%;})lSv4{!oi`aoR6$+M<PaiV&|+7e9Vv=
zH!%%L#(-g=LuH9X96pYA$`VcPTl>2@>g@UeE&dHqi@jy~OiL{>BgL9@^QJpckw)6|
z!K<p102$gpqu(tLKM?u9Oe9K^&rzDw*5p2{++LUv@*swz7~US;OT8{SJh={^@~ok=
z31{dMA(zhh?M6*Wk{qF=qhMrU)IT|ZicnJ8v&5MhnR@K6^W=SF3lwc++f;rH$BjDM
z#IKIVUmdhhHr>IWPe~@RJPt$gB-EBn09Yf~O&^3Cr8_(F^UO|5zM5}0EUXFxRChFt
zqkC%KHP{P-`g?ZF7P6r_7d^wj*w5Xc0D1;ONk#_d-mV@XI(w)C*E4h7#H<bn7nCf}
ze$TC85%(UIlszf`QHC8H!cvq2mogzkM4PS;uDj4GM@UIOR|vf77OZ`{H{J#MZgjlU
z;M}ml;?%%mJvki@L#0QaVDr=oE%RJtznLmC2|?O5bbl1DIDe+9;7OZb?vkR1^3UW$
zZ@kKdkwljKL-@j1SWnXDSdO)?MYphXaEd$Hsq?tNjAiSUgm?45y*;IIpfSYL;+&W5
z8j>ZqC+@8VsxQ#*eGqO^U`+`OK#<Sfb#@nx*=Qqi-y*m(@!O4~SP9<0rm+e_o%m$&
zaxV$(=a_;7kr<K8Qpn#7`;1&gca`C#e&g{^7UJya`oj9@rQ**z++)?E^>bh2BJ;xf
zi2S&Uhr{2ovE{lN)#;DoF=Zd|<V%yR<}eVFDY>bR8qa(CTM6$w8ANXqEFXUAe4!tt
zs3cX?D&Qh}eWp(Xzu$%$2p!DKU445V8(A|^QG<O}TCXKZ<2oIj9LUz5Q6R8Fp4xB9
zzy$A868a8KO6W7#C)P5>+;^eN{}o1NVa5H^(o5Fl7zpKCudsCLFj0}2+?oc%j@>~|
z(N+)AjWxgN1a_OmDa1?~YcS$ln*hHfxF#S%V0#;r(foHTSX&%>;^gcU*7=j50Z~zL
zZ0Zif?+EUJYk%`78wVc=q7)M&aQ2x`FL5PfS4WHG2uv}-Ru3A0Q}<gkpzBhSc?Sat
zVD+K=^oU?a)HD;ctF41x5UHrXI<Ticn0L6_D?ox1WzYL1*yiikN0z#LtiqXT)Ykz=
z*H33hBI2T#;v$;nQj)4|*HA^k=S6zIykYyh-QN8CZHdaKsnpeBxK#`IXYSQy=LVOW
zvu^fJ;K<s;3Ua-!PYTExanbSDs_eBI%$uPhVGI2D6nAu#w9>Nj2FAwXC(?VXj~*i3
zHmr>}gaZTPek|kT6C^CU24+TqC@R{oG!@iwAVe%4^7v|_vYZA`W?=}DvG_!jT$;t-
zZfoGjGSx_T(KNcJq)Ztb2P{yl58wcEJcDx+2m0lq-=lxSKl1&XV)oQi62vQMj-=#G
z>aO5ky?x5R%%Q8nw$rVB1A#o7QR(OR5Q0E#gm+xSgoTa{wx+h%818t#<u@GL-`DL$
zO1f*JU>(Z0mE;L^%5i;0Ls~L3k#CNJj+YNgY;CWvn%uaipWL2WdgI!Mii)Q~2XWYu
z$1l7=-zFzRAKlW@4DcicZgpNB`k#KsgM_ZDcToDLyyqHprw2YojfZ#Es?vYB0B4s6
z6Guiy8B<f@a!G6|aw=J+r4$I9Y;%PR$1{v944i#eX%~XFPrO9o*JNVa)?owgn3SR-
zs`8JtG}QV^d;~wRxg#$jSk(%!oD#K1rA8s0Eh)y@jQ5y*9p3<pbuXwktBh($AV^f>
zMP6^ml7i~YH^)~;61Eo1p+iiu_9I}`1yrAkxDVoN0A`q}u#6QS;Unj#odcKLvqje{
zMJL^Mp1y>|tUl%fo9@()D`Qj1Jx2Y%4_7e4lUp9#hbJZ$79RQAZ4T2Kpjt(@i(+vl
zjf-i^LNdd33nrl#)DSX*+{M2E{`Y{-=PaHXHt(b++<E)uOUQ#>or@})iMlVE0&duC
zy0Of<Yh;8LG!QN%vXv%k`b3><ts`$8@?MPS#8BUZ)g&X165(oMfxr-Pa%NQ-?&ZM1
z{gdr9t<{aSyKc#skab{aU=$e>75gdit%WHu(QVd8kDhgmN$qG#I?Endj<Dr4l<-ij
zl5PEVTUXG0g?71quwcf(!hnVhQK3EjEgWHWXO*lpTV`*o{sTJFCc20J@%fz5_MaGw
z+{LZhTd-Tmump<}D-bFp2t(fQu+-0|w>6n-E!~fY(_*9Kh||9e;UYB)$q~+X=HGJ2
zxF74bo428O`6L)NrV5&|{owl2uBRwTh{&h3GX*lBdesC_Am1uxw#HGJNtIo>3N;cw
zr!lj=GwR&1&jSa`9#1~bA3ypy9^nUqn+Il67<yp88(Vuj<;I#y)5xr^uTQh6*xB!X
z6Vtbdkb{GTyU0CX;jqAqzTs(|SDOBR<v*+~aF6sMn5-=H<KShn{Ov3RA5O%yh#>+5
znyy1@^6Q8p0SITfq4kE9yUKlQ!<1vok_AYo;epZtfzm08zWzMehw+;hhAA`5b1Pc?
zFpDWGEo;<m?cGeQ8Q8lyOFsGLMDFAP(UTp1-UPp2S$@3?oE2IuSJ{E%Rj$YR0>ELf
zh`Y)Fx)lm$|Js+xe$Rf*ks<CK#|`%v(q?8RYKz%(aIDZIV_|rmp8B1#O8jrlwa;MA
zm7SJ%D43T1JXskk<Nj>bS-!#>l|N^{kb!%r*US6?>b~Fp<y-T~m^24Cy`EZ{M>KsG
zpBOO{Sv|OzGbGgN4C}xiid9}+N=oQi&`sOb<2Ddp8ir`=D8%nR?ra_nSXn<PF~wxT
zn;jI{ofZzZK%UlEVR*_w5E#N{lJaK|gIfo~FC)VUo|85Eceh05Kkh$}gjO^lXu$EZ
zE=dq5o7YhYQVztG9dMAz-FgIhWz4+jJ}%bkl9T)JNP`j5J&UGFzq}yQseT(D7B)N$
zq@>{k$=P1w>s8CPZl&nDC2bHdt5ao-8!0)^C(qe8xO8G8N|5I%d+!=ViB2EMJZ+*)
ztEru=Z+&cNr0f8uR(@3p5dWzmyrWD<hP>%b?5JOQA*Vttu!ePhif3w?d&bS`D<R(M
zAqlNF_DxK@@Q_NErGgr*@TT8X2h_HMUbMIH^^!psBrs`n1Ic;FM9JrH3Bbgn!wuIe
z!SypOfo)P$#_ba6r}s(g=cCS4|39kUI;^Vg3maX4fFhxQAgzL+bayF;prCYz64G7L
zU=Y$EAl=>Fq0-$*gLF$v-{JXv-+k`g56}6-qkC`GnrqH6-tor#K(Ve4!-?UO9qK58
zL@fV<eDG)2DaYTpB8&bS1`k@BZt*wijM`lPH?Lm{vNf<?3-z?DGTmAAEWFDjj7BQ)
ze*2}Yz-GjOO5g}0ihPKzc=_JxYi{JtWWz_asJD?{Q|?T4PKsqp{Ie8j2%uyY3Kd$!
z4O~?;0gh3U_fS0xzH&$Zjas6g=+{qw)6i#4=D`~(D!%t8;p&o$>y&T7lp`_H{Fv#|
z;jHqb$(JZ150$6p5>0RR#w!~>Jqpao%92&n(vnvfHa`Ca`&RO_#)X2z&C8ypz;qos
z2C+@d8eTrEn>S<clI>w{1jX@^-NomIsyaKoO+{sFTlh~VTTWg;_eYJWpU{{H)1+s3
zU}N9#_x%EwmD~pJcpU%U(e+lc&;5>{x9?}I(1~hYY}9ld@4r6#Kg^qxD-`dy21`mJ
z8ecp<w6<n|h#qMI9_;mNs1wu*ikh8EHf-20rCwq-oZzUhAG8gB_VKJvjmg1BXlT+)
z#?5ej!b5yrt&W7o#1(S;dh;8vSq7Ox$uVQ$8JCw=R8;g;)Wr-MWex&rr0Q($2Jk6t
zsv&LYSNW9@@Q`QaW)l12b8^0<eY%E7zVM~K0i%>>#++<yY|K%y38B%U(OVoHE-*HJ
zy{)YvuL!>9W;>v67)zIJw{y;fX`i#dBn|W?1J~U7DoZ%gJOTQu!jrm&>eZsl5iLEV
zrCVs=fkzUphP*K|Vx1Uyru3}3qJr{L@cp5`F?d^}h-$mvcP@!aFMR{7Oo`f>svtXd
z_WqTz%zHOFbgH$T4`~61v8h_Pg;yC0nn_4sUToCRmIFxrD4~yAlNH7W+O@`Q@n^wu
z+lZ&5?rjreMk2Bi-;%HgpPz<;4TqOoa#|VVZ?|q)%(I>i<=J}o@6#hdpLl1mf^|bK
zWa!#gEbq73##AUsdAFrNMkR4OKPBJCTSO^{=v(W@2u)_1STN4@l(CYK#J(AJ?qN(%
z;Nq4^CFejeOGS`4&0eDR)x#=H3t353_|dz<o)5jbb^hT2jaVUGs}a{_xv3d2jvpwR
zC`t=Xbc`grF{74z3fh#CxoUsI%8Kp(T5`Sg)-^p7vui>)Xh;p2cedQ?NWa+-g9lz!
zzf3I71NtK>{QZQyKed*4|0ZOkjk==~fgmGWT>Qk0n-&3cR_K9ER5{_l?mpL1cMKiT
z<lqZWiNwD=e^CFL1+-wiSrcfYN-l@8<=<27CHPEnSKeqolzYY=8gO_x)U<3R8*9A1
z>&blA&gir)&uq9ovEFrq!}dE>)cG0XWFzsgU1;A9E`NPaj<;j33+vkwXAFXeSf3v-
zG2L(#V?a2$xpyn_uq2+pq^9E$VGtN==@@A3JU_{;Y%(0DgJh3`A<>M)RU^|t2YV7W
z*GG++G>ZprKX_k*xMY#bd+gblmoKPi+Mv<^@1K!>zVr6hx~w(tEAAOxUpMHae|$eV
zC#9J|o7?AO?;HmYn|DGY{@=F~JSgXM4g(AS>1NKalg6m?YHlC@O1TPFP*V~DVKC@G
z!dt*HS(De12uU*Vu$S37WfFj=C1&@}s4=uvB)9K`$He@XW{%t<#MhRUrJ<oIGipc#
z=sL4@V`Re-8ZyIP4O$xUwav{>mrUxAyYraP##(Cp@+TeP9m0vpNtJ`8EBn1q#n6ST
z6in;dv9mUTX_553TNav-toA1#@-UgycQxKfs=m_py7f6sMHbXBo|=p|=ens>s>ol_
zrQSqTrAM<%iBXzHTk$E>cgI!QX$HUg>9J_nSG)Cn2h0`Vk!yk;>6y81CsWp_URHmb
z$HDL4lEo4HXG``v0d$11p&V(T_=7;Qg<a5725KC?cd1@kTyAtGgQz~I-+w4ke#3xV
z<~aI2h)s2P%(%F?^KuG|jm*Nr!_)l+=enql2Cz7osP`eW1Vp6x__|E_`A?GPfy#rL
z)9v1m&@1r@@OS_x{kLw)<290tJl!X*Pk#JU+NCKX?LA$Xp8Z8ot=<vG|EIuzn7#zm
z)$p$&s<kF0(N7^IzQJnAyF3DY&<<(3c>bCSy$&2)K6ut!o=YE45>S#*{5iLmlTuSw
z#zg`gwvMJb+~c0V`BQrqjM`2U=6n5c`y<`KtA9?Q`ub_;i)Tw4NkIArwStWU>FCT%
z!q*sP@nk6TP(Ay^ccw9eC<W{r^6DNGCS|3SNh^p^oIE^9_#o)faRu)akBy<eTwTkL
z=2;yr1sU)i9K`bSL!mKfYbJ8JtixmIo<1mj9fa&rT}sM|i`h)rhlj`<=x<?c(6elm
zl^x(<9~v6!U~i2D+V)0+s3BIL$|uQpy`7wEl4X29#6EfQcK1l=-?V^grg|`O6B@0T
zQZE$8hJm!*2<Ng5^UtW}l}Tx53EBzs(6Gzfh%ik?(Klk0HwB4>S&wAIKk@%CUd`qC
z4&twWBkn-f-5VX&yWpJ@8jXek5ZM0m=ioD2bVWb^@P*G^W4Wd#|JLaJ`!n97i5Id`
z9yfN1DJ?&Wg{FkO%EWB+zeRAz-jQNFe8IAq<37IcpyHZWXGOo>I)ESg^DWA@cy7FW
z8%rrLesV0KBr5;<e#2?+4A&*{YGSi9(QRdGk*-F50tp%UR?DXA;ynEv-|IR8T{xa{
zF-JJh$sn&@_M7cIHpVheU02VYYi@;^CxtUaK6w>dCh11t5E>dA+vwJJ--&M=^C$B$
zUFk=RUq&WbQC}m?C+(ZSaN#uH1zTba3<5xIZ_*5`un%HC0jCqs8g>arm82Q`WUYtX
z6J+OSx}ZKVii6Yx7*zwmmbf?e4k)@f7%-9kc}JGnYw778gm(zy!K5NKslTocb(QUn
z>*;Tqa~~S;aEKnt<i(>RCAw{Y{GA}cA2`NON|L*Ss5VZxSEuWrzwXdy52Uq$Bok-M
z3vk6KVP^-K9Y_(PE-cF{Dq84hZ{EC7s=6uc4;ED%w9HSmUyb&Q_Pza+q?g$XerIL7
z|DGG@oTv37?U#cYXdVm<H(%$=l5ieL3bAtHQ^SgWN{}wI!Pm99^Y84d$r!y_t_RD{
zKkavK)8BdA5B;-{$H~$5Pox%wyZ4nCSSiX6vVZ#YDI}9}l_XyGH$J$Gil_R)*Z@E@
z{Oi{er53=JzVf)=&u>9+NaWqf@F--5sH=bbajaJCMn_|AZZ0CycqvHw=lssz{r|YM
zANMUky+}XB65I{#nNNGSc;MkCNgF<___g_z-Qw!;BA9wKoWPWsn}_2hXAG$1B_;bn
z>c4Rf4A-W3S&qwdBc)!xMDeVu`WBaG_D<Qt$sm1Fa^?OJ+RGPRJ<`|%r!*{$MFkB5
ztrPfe_gDVm(^}$oPS5UyN`{4f-CRuqht`qNkzg{8n!LJ_s<iHQS&x5u@UM|d;IanD
z-h`@}?4z$B7XrIk&|pDH3oh08_zKbi*r4our0RbE{(T{>c%Y94bZ^#uc)(0->u8VS
z34I0#aJ;yc+sV3s697auP`OZ;%P8G35Ob@CzYS<gT--VQ1)n2}m2qL$8*oERUzS2I
z0$wVS6{8?VvY0GWR+B<OK>DJ}cx2V~y}kcbOcyK4pHBcsDI`k3?NZ)Y?U}ty{@1j$
zPoKn{o+&v!iwH~k5S?vp=876BNxCFRUTxtQ6qqUd$Rbrj@`a?T`Zp+umY0a@Z+hs+
zDRCLsjg1C2%0D7R^<ven5_@K#wCU6Ad(77IFkK82PaVhz?5HJVZ)7q;N^NbRpa;p`
zyVF-{wI*tbI_MzI0cI2&MqOPD#+;#R!UPp&s?_(HgC-;#UL_<1DYKY=4ogAx>3RH!
z@e(9A6FmknxB<%Mvu6(_yiZ{{gVX{<9+IyI1_xu}Vv;j*1UsF018Gw-Q{$uK!=vMK
z(;kUf$EirERW}p_O(SF|zmt!$b5a)9e;=$(Lp29OL57zc3gL~eqd{N%@NW1s7G^!T
zbzMFz40M41@or5(Cj%2PJ9{=2U{spv-(#dnOHWRShp=(g%}q|iXtFqAkLvo54g1H3
z7j`qrDS2tx1<La3z;~o!p<ChkX$bRU&V{J(h;mk*?EGB#jPCYsb~biF@M{5&2a`lV
zi^1TgcG6J!72wSE@%cd&k%8stAoh^Cy1U2xTkl;^WX5TXPGn5XT~yD8grJfz-M4S0
zz}f?{xFlERff4hRnNMMGkVH%RxzCgK$)~zDq>!;=L!OsLV-!wn^646Ca;|^Jk_{*o
zS9*>P$%Vd%-uo7P*T7&!Nl_Wi+kCQ)`yrE<rgXPU9wz-m7&~DX0RS;v77WM*P;QK}
z(EvD#fy#3lj5_;A2f=I%!}I$Q8XviZiMoWMG<mdi0{h#kF!WM6GZ>zRg@rBDNhmxb
zWcDE=x(#R(ABd`I$Ur}w7{ci?%1QQe|8m$j<W|dn2p=7T!0WHHzj^dbp1dty23MrW
zT^Sm4#N&a?5G`1uIwn=s)=tioZ@csQkXu!4%v&&}mx+eTJ05dNJshQxN!{^zeBURn
zo+zl+<@jpD$If6G=x^ikhLpwS5;tOv>?EwEqHAMsc3v-_5#l(zOl`*^)uNAB-&m7M
zmp>>w+8$T(m&QQcgz5vj$+alr=fyA{l!7JlvGC(PHS8}|d4-vD_*_Y)-q`CZN?PGj
zI-qo*n&Boy=+^?ES^+EEqL6!nl(V#)zJ0_YlBCnM4tx8DUt9=w7L5v}MN_>&>6I-#
znyyfl)8ge<92y)&7ysRrXul*WDN9DOKWFJCImKrG%xjD++H8E~Hl;d>xv931eiHM?
z3lW)$={SyLW~rtlBNYUGNui1n&Y5#w7*F(`^l$c1j6Qhy%2`JFaiz46xRU476P^Ln
zfKsn;W+^(a(DRBtH_NtdI%f)3mFN()*aeeSa--$fxw+DBXq;l-<JxWK(qrKIkZlgW
zessI_Ya`ewWIdr3?l&JR=||@!4;pH6Qzq5RG!+<d!A6EhOLe<+_|^mh4$T7Q@7z8V
zg!IwO#{{mi$4^0lJZ~<xvHo}GHvvIdm{xpzBvXvfhgpTkUqj{)B7X`3vZ%=Dt);aF
zyYEv;2w|8Y^U+c)7HPcbJ*aDT(LZ330#@fB+Ndqp2YS|vf)5g7{r&x8LklCj3`Tai
zK9iM>$8Di~GtG3${i`(iQaZ1Le=aHYucCS-sT?!o9#tSA@!^T#v9pDaj;5wv{0;Dc
zw)c)s{A%y-8_Uh9IaTZKI_MU`p)Fv3b$ECPN+$52(UZU(8Xd*=f4DXI`F`b8;eA4^
zoBl7Hk5@?v=M0UE{%pI2!IE)yNf5LxhFUAE7rxNY*ifnxJH3$|nwuNphl%0lZ!gmO
zNxH%9>Gz-w0ks(6#0`U_or&~RzvA0SUEph)Ti1nZ93&D6bf^XB;5(n3IXihoe(dWS
z5@Mn*EvizyM+p8&Q&Ur%aJ;1DxxQ(f`R;WoAI-_dv_HO296eqQ0Z1Uem-9{>8XMza
z<IYfG2R-CJ_Yb=)il39Ho<6bu$;nWbS$F%Z0BEEVTmbg@Nv^6P?~l&Y8xm%cF5Nge
z)h#CQ!2%;R;-mZV=4dsadsP*)R*m>`9tJoq6PMj6Y3`!^do9d<XlR)EsS!0oe$T{+
z*JCMxeegy2-0t5HWCzg+en)xbXmcgk95jvDzf;YT8KjgOciQ#<-w#Z=|GM61$NE*h
z(p2wr8yM~}tPYF8L^umq*NOav-u(URb>W<^y0;%3iJls|WD440gQ*6af{t6PS47Ra
zQjm`m`}0{OsSh~y+by+z3l&@Gi8)Grb{1S@bTR=Pwn>#<W8+YdGRC`C&&dUC_J9Yn
zdR#ckopeom{kv;C>whnGUrlC~+@vcGd(JByaU2}~$U^XYB4`cj9~yGmJK&QNOVx-9
z5>aOkxMG3B5PU!y6$efd#p%NzW@g}#O<jqkCZQnF(iOT}Eai+xzR76jdHLI55Z$&Q
z(vjdjpbIO{#Cf!8;3q3OPOAu&^z)Bi|6T(&IVP%|s@QjD$zcbXbuj0HE(Ta~cd)<%
z`kSypUHWIIjhC%WM#31Ur@4{Le{=m(SR1IGdRa}a&VM{Tyd3!8`m(I7{mj99+++1G
zl!^NC**-p#i@T}tD7lwuUfsU-biUb|+6Q}e@-&5&wSnhb5XOHuEOlwlikB$XR6lDL
z{zrpvrqATS$6Ajv!n;>v*U(QdjBc}Dekv&Fg|~t~oTa3Deg+w27e<dSW8ulWpgY5E
z^$5E%qsPMj_X=o40oxrz<twAd|GT2_-^-hC960a(_hNcZ8y=z5UoWJot~2Zrj8Xip
zEmy2@_kCPw@;otKc!xeCH>S@|5;Bs@&x0EG!GxN#f6TOZmz-IJF>-Zn9W<?%flSAR
z$w7U@ex#2G&CQwe3n>&7x}b3I#yIWNmtPHHz++|SZ1SkQc4Wr=oIZSQW6i|K=zMkb
zl(;{R>@?rV$f&)$cWECTlBygkM%RaW&c)_zoOJ!fReVnKF|Lg~3G-}AHM6+F3Rxs2
z6?dzD7npwaXno_EtCy%k_$Ql@i`*w@d0t+^#o1<VOr8L6_1}|dEN|DRkJxMZ`RVWw
z)hk4LbU0bQ@%ZpT=w$-iI_8HpqyJqCb@bGn<wlAy3EFTF3!{4CM!ZNi>RQ?Z+5vb2
zCFk^s#l=Kc8=a^hO#b?{x?R4nBbXQ2Ldk9tRuMTm<ipc&T%MNpj#0a~cW8R6dA`1f
z?{dt|M_0QU@&WP>XPB{t6&Q)h!f>+E8jtRD<rpp%?M<qBCy7^FJqk+Tbpg5%RL22S
zLg%J!0kMMqEVrg=Y?Pz#eCt}mq2!ZuYD%5-ND?2&Ps^$jS5O%09DHiO5cin@63bta
zPvmy{2PAvz(e;5@rQ`+91?}P;LT2@bpr7+89s-{#$BUh&k8^{5E`1TlId!hl-%I@6
zqO3IbHEM`c5_}SwX$BL^-Y?zZ1mTROVQ!7}&&c>40)+}Q&FOTKlcHELd7FtrC9-}?
z4vrxWkGK=}z<yNlB{pCmUzx<g!G=JtsQj7LjV-6era}?*?#Av^->iDfpo8@R`-_~1
zBI=Q&?C$HEHT;)>U6~_`vc%Uv*}DoOcge_Z-?;YHO=u81<}TUoVG{>jO1_g-DZ<ly
z+Lm9yxWc0n;VAOhx#Pd$v9lrc{_i5ZEpV>_$hot)maMAlV49e?Z1DE#?7Q9nS-|;E
zX~Xx|uQ}Jux}B^(|2}Wy6Bvz(T$~+fmYY4%ellOCb`#~zz`5E_A8c7$5MeBAZ^h>1
z<T`F$Ts+qiznLum2;y1YKTffUZ3no$e9b#__D4U_+GDMGHaytUcam9K^CR8)f*ohz
zBSI60-dG`-2che$uCAh4itF%B$eWWGnV#47g_POS+=laC5~lfCE}f2uuU%`=VcEM1
z-PYpg=l|v6Qsc4n1!7}+=dI2THo0vl#G^GYpFL};gpU8uN=xp>Ust7RsNJT@w#KS>
zc2{-u^ir!iGID6;)XUv-BhpWLnKl|_Q+~(V(>d})O;7R+kaP>3;dwL?8L{mX7eZ}&
z?pn((@$L5{R}0>p4%4~L{AH@|$LFt5n=Yp{KB8T`j<5eD|AZ%-ii*q3YU~w7f?#0K
zm!OLQ;@!Y%=Y`(tmnsV>RF_6i8n81`zc>AR8QtL}?~fk2dP{-x;pu?;v9MgSY@J&a
z7$!a>9A$9qSla8(1D8@meWSVU<&*=Bfw3V|G9pbdOKU+LC%3TBZundsJ!(kQ1+W_(
zPh&=8dST(p(!4j;*RCZ%Tdtwb^p1KfOR>;1mzNipl$OgSjNBtK)Yrecu>oFzTW0x6
zH`6%2L7j?w<dEopKca)3y=46_>P=e?*zaC%CT9_v|4qu4``oo;u(B*gBTp@chUYjo
zsv)=Aa?U%UQ7rx2aNRt2^1a+szoKs%&I~)ki2kU)=`Bvq8#IP!bV%0iQTESD82WB6
zbI=jR;ejd~!JA*Ala*5&Rqxk6@`)%!KG!DS<WDy!8Yk!WF*P^HHZf?*&CShz8y)Lm
z_3HaPY4dN0%|b=l*;)B0y=}m*W~_#efI|TqBGoeDD?VsMMud&%nDRivtgK>6w$$GK
z?qze$ygzGhj=~T-6jqwrT5<r@!ZVTM%r=L>$ia4R4+jTlO6zJdT2)c=MQ+;0`Z~l{
zCMC2&R^s(f7^lt{7=gu2e45PuEd!&Y@(aNVn(`Ux&n_1|_pnHDeJU$T5TsfYdYCp{
z>+Wu8#qdMcZ%R+!)<d!qQFL5Pf?C^(MhxT~@+H-?M>A40__XDgR+QKkRic+aBK!iu
zp-53lNk2_!zXbnBH60dq&VEDNDy+{>?TPclX8P}Ov#|{(%adSd6ltdc<8EieJPZFC
z#k8rG76rA~$AfM&%EI2+F^!R><g&aE-}7wj9caQk>@XBG{LL;84_%Qg;CSq#8>#-;
zJ<40SiHMFI-<p1m#6|hU+MqrTmF#4o_Y+Ua4^(3EkBC+A{5fAQI6ZupQ|IeKKtM)(
z!ssn|KCC*o*}`CL`5V^A^WeK-u_{)_-xu<0kI*sxl9G{(mZ&~_xCAXK?08uM*$x;8
zLS7Km*Vo>iEL-MQCm7>;-o;>q?>9Wr8P+quNK80Bp*lIixqO`iPBmD)ikf>Z1QDs>
z)zR6d;U7(&fp1Gv(YNn;?3}~LR<=rqM2e3Ht548JM#@_yykV}GPZEE6yw<U7g)+B&
z(RTPiZY5|w%ZCiU$jI*ePwKdc_%2;r*Lop)dy$zSZ*Om~)c3|H_AgBx^ezC&kd(Sa
zq^#=B>YGdx0u|*~;V&TLL>X=j^lT48`T2nO_B<!o6g*H2*rO{meR8l7QA1XUd0(>m
z99W*deg@W-aJM599_JCG|2^U4pyMelV5a84MCyd5MY40t^5T@G=t3ohwJBd%@(#~~
zhswbBJ7q75Rdo%F3j{$88=|HpLxVr987FM*0PzJKT`b^v`{#IHS-1j`)nR$j)jl|q
z>|9mLePtp4Xu@)V0Jf6ZR*zHs!i9Y>NPL!p0SV3{u}hnaz(sJW{_^erasfi3BlcTR
zq4YK}F%_C>xQ7-d`ydo-@*XR&ar1qHM_{05@5n~Qt?Ulgvv&>_Zf^C3!DB1-Ag0UP
z6huDz#BL-e3>CsxNa6TaT&RtPI5)xdB26b-S{yOZUqT*DX~s7$r(V?mf6f8=2q<Kx
zwaDGe?#&Vt)3reXUt*)~`n1G9T+hUrcA{6eSjGHTOlWlArM<2FrTJLo>jq5`I=095
zOEo<|PMyhdd30ArN3@piltDEc5i!)?NVNQ}w!*^7t|<J40YOI3{ae>Jl$mELHh7A~
zQ4v!W#_ge8GO$itDx54!4eJ#(wBH!CUWhR;n%SF^TkNgJT(aTw6JMXqjWy8%aAU(`
zyhFEipmDpP7YuCVFWQGPjL@3sn6-kb6Sjoji=TLHtSpoRsdBWLi#N_y*J;z)OWgO8
zVnO&^b2gRqTHtU_fJ2l^QXtBT03`i|y0i{EB39fBgC)Q0IW98Dw@%2AJl|<6zO&N-
z%CmQ;uazDViil{q^PYSnCqz6kr#3y;Fl#R+O0v~AbH^HiJ!mwG>mfq5)arsq9Yii-
zJ3oB?fF)RkM{oQ0habAgpZ&v2w;c!Qf$tH${>J^>=}S&wIBBzP2g{_RNU+=IOD=Bq
zsWOLZlM$W8rg;K3mfy%ddVZq(;{x@XB#+sIPP7u6y~w-?&+apDrxns$zlO|NRWpp9
zhwZ*SeQRi1>T)LW^5%XO&(6`N2tFx!k>was+0E*@o>(sP?)GGz@V6bEhZ(f3IehJe
zSS@AaH!s}k9nGiz83noau-IV`Y++&lnJza#iy_diu>nc1517$Nz3dZzM*$7jnWH7T
z=RrVTd2Gela7n)!#g9p|@uDx1=j&qv(0x`PzX2ZC`qIs_!)+8qKHKhe^D{Z~&cVP1
z5c)w@pjBbCJ%{3kB1(fV6DO=fV^ZZjpN1Pzv(gj_YNu2iMYnkE)_RC5tkj;aZC|t*
zuBfTu2dfAPg(w9>mUv;v(YDJPc6W2utah)}iVe}JG*`4Og{w?(!iaG(Q+|B!l+Awd
zmvMM2xVXS)&erUU(nbMZE*Z!AuKD2c0*Yv|X9veJT2lf=EVtQA%hE3Lrfu0CevPc`
z4X5s0ungqm$Y`dcsedgkH*H|5$K~7{a_Q#$phxAYJbQlF6GK*b;90b)oeo<m5Xo0E
z&)x3&WeP?~<O;rDqzQ+^%FWq9qt>SQ^s?Id=V;3BL8*o3!12|fF=?8&^7lv(vXd1#
zG=z>km&Xk3yq;WPp#Qz7F-ME}RmwMIkjwLKWznsQXajz+qKcAho-g?I#@7QMrU%AY
zzjmyI8l~b_l-2pS+pBG*D=h2JQtqRjVPKx7GD(t2J_xP2s=4~UAPM0zgEdyCtn$AF
z{ah<nL>KN!H7&rt&ThCoOq=^s^2K<OLDV2x0knAV;7Nq0RcuW}B}CU-_mn_bY*O;O
z&8)Y!hOf{)$triK_?@xp!PEU5XxY&4e|f>{ytlc;z+*raaCmyO-e5A0s^m6WGD0@n
zN{8y1?Y9dd!NV>)s7);lJg?%beb;|1?C!slA>(poMc^>hI&*J4p1Z*h$Xg9~VW)}m
z*=^b>+U4z4YcwoJ0iz$Jb52#YU&K?CHY!``Xp&1gxwuw(<M_?2JdltV>5~o@M7Iyg
zw+6??CX1ZajI1Cxs-J#^Mf<a0kjvrfge#RQ>`~x@vA!M?HfSS4#j^DB90*hz4x7*V
z2odSK0=lCI9Ws#)fByI}0590n%gg;wqMO%_Pynyh>ic@hPnTO$$<wXnU0j@;!+5kR
z1+&7=dwJpy&=JsMxE~He>51y8$*j}xZYb9A2J*(xRPXJj9A1I!qo8<x$CGs?ec$JB
zv9Za(wVU&hr8cha+{s^=TYIyN51ISD`Tpqrf~wv6X8L|qApmN2{yw<tl~yy|jkD1g
zi6)%>|8>c`J6^q6rvndM9W~=9Q10D-{E)5?xsZsffBz7i+)P(+UsKIozj>pxhoS2D
zk5qViyr7Gms*VUO)SH>Osh=4?Jxx|EcQ;kzGynZ76f&pdxt+-4ziXDe`#((<WteA=
z-h}xVXxKzK;>6lsrxF}*i^xkEcE<>>?M^eawo3FRG)lgb!bGwaoezfjsx=$HgE1A;
zUFL<9!sx?WSSooJpIhERRO6rJrJ;Hkhn^0GU&Sk)IdPW&i1=VDYOMm6>4Hc3ec=!B
zD~c~wj^^vl#USK^{kXj9?ZVm(+tf&Wt54qs@Jcfq@VM|!saIMHcaG6KEB&5N{gSl%
z)y4LDCcwXD_$n<XrQcEi0msSl;=;VLIsH3>A;XnJ+jV<jxd{pt8~nb%kcvf0ft1+p
zhi>PApQz})UlW|94A8q-$zBOh@)V8?4@=8*iP}=i;IIAKaS6I;c<w)yH?r0mWWD@j
z|Ch=%zoW17`~Ahx*=GEX(;fj3E)OSi*-p?5Q}H`uKacRP#8DEOo9GcRjbw=t(!)YD
z&lmWydWt|D7LXU$K_DUGVPkm4V#;6?0ZR<sQ(LG8JVjvd$Sx>ALA*r&p2`S%iLOqU
z#*&hdZ<;gEuOUtUY;`%FJ-8j)1Mz0BaF5ud&DL@P?J#VIYVNSzc6@im$@iEKoI6_E
zsbTztSe<nCq%D@+(0k7KO(4~BXB)YP?-!Xrqo~j|QEA$>Pv_VgM0&)&y>-$xHq7&E
zR(l%d9{nl$>9LzGmQR<hwT+#vy@aiZfli<c0Si9YHT1v9=!;IJ85C!3uiK8yMIOt3
z%tAp}7G?nBrTA6a^L-|=Ew|Qgi57txou+RzF!y+Kijd?E8gk8*!0TnE45j1dv5+K3
ziPa3YT_eW97CHj`Wxe}GLPW&U_yXo@%9+f<LdWGFRFam@9vJ@`SMSm<3Dj^<Q;+G3
z6>^)&Gv*$RT;EWyvD%*AApZB=PY6v#XSMeYUstm77_fc=&TIxp4yabhH_cGqH1~Un
z$dKS64KoQjzaO70oNmGX3`z9Y5z6PnXEE7xZQR5Ci04Xi;Va5Z^ZB}Htk1vyuI-~H
z%U-BG4@zWaq$d&9f3tY8!Q2}JoRI;xz?v@e+9QLN#lxMUeOSSI&w({r_^bN(9%oRZ
z#)}tSQH<XR2&@fZeyzq=4eR*xj>J3VAWdeGBA3|OSpTAi%O6Xo2%MKS1}nm-305bY
z#qF45lb_IaYMl3<O_3cw8`(X{9kradsv=Wnsdb+Jlvz6oyS2rv%crB1HvQS}dYa6Z
ztz~}j8{^%;LXdY#hnHe)1~w<v1dThB<<DrNc0}yRWs0Dv2J)*T7@eN9MPpVnJ8n$P
z%ecb0BB!L4m_-K~j_v8{@gGxpIW-b$V`GxiFNAE)u|GpnvQggvTBCgSS2Um3fM1|C
zb&Xl>2));RYjHHUoE|+yJ2J8hNu!wsjO4h;gA`rxa)bML^0zlaq?ZEPoxckcOldEK
zJx<ichrwijLR`4YeY^Qz0P;el7OK}Rq(HM#K}m~XhWlV|TrVbuCfy=kN>eKiSibO9
zC`}y$l;&SM^fer}7WUr9juqBfpKwC>ywypKeC#>bUF_uuGDlzW-<FLQ<3-NdgNC;d
z_?He_LXT7);cm_xwN)>`-DYFaNvwa5K8#Cyhx7D1HJ!C*?fcXAf_yT4_VLTvVCtT(
zv*xcnJNV$e^_t?seDJT_>ct1&ASb8!J$nU$1%>Ax0=Ao;+1VX{FD43k42TQx3Q$Z`
zqFqSf{hhcIXfjG{ZTTv((JZOZaqmw<X&WAlO+m|gWv$;eyTgPrH1OuXS6L)Rv3Qa$
z_j#f?jAMI8Y1!AE#Kb<v=9b!=fmV)ks_xSRbU4FNMXimaPg7R0-$CHazk}w-_Zf%v
zwBk-J=*yvZBK&d$Y>pmj8QK7nC+V{EMiKCb?*%FIJ$mD~**JOUU_yWxY=yo+nX8=E
z*qN?kb!`_#1^?=WbV6$3`_VnicS%UG*Js;$Mw||O?|0Vxo}~lR1r_2<_j;Fav9Zah
zZ~~8x`G0MLCmaG16}TRTJWURfq&a*r04EjpyOG5ksfkndUn1o-S`ViLz=hvdpDy-j
zTcY5l*a}+HRvjjiBsO@X{Q6U$&uZ5yi?Me(lnmuQpphe-k}Zhr4+ei`m-}IU+ql*8
zd~__`>{SN8ldC!N!+MwY-gRr$y<1R59<gMOkE=7_@bgz#3fRdu2F;sKe!@TzwU@zx
zAdPF5uN~p%hfW9g9w$E$;(WZ>nw!hP*^g&Qpx&iR-G|EcYeF?TIvUk8G#Mr5fWZG)
zYC)pXl9=0JQ5KrvXzA8>woY|5jynV>Nw51X%y8jajQ{L;2m$6Tx(a3GF2|cWvXA;7
z9CcOk==!B8_&BWFdhzl9Djg?(>vpk}Te-}rdda?Rt_D=7IPxuy5}R*IBg&D>(8)1!
zXFjj&g?qIBi<huVOd<RlqDg598!`a&lze8gh1*)l-1J!*zwqpIPfPD;OwT;r3ryrI
zk%UgF#tJSC{$n6AS7(V!2w@`n@`N+Y-NVhi)C`$*A2Q6=y)Il*3AW74p@9q(?qc|4
zmwI~YEYcTENjT?YhSE=Z320NAe0E}?BXqjjl-nOM4^TX5Q)wYfz}clHHI5NnWmesz
zWA36gm3|{eSZ#jYBQ|`F=cuBO^+^*SL83!svxgxdGMI!-A4>ehYX=wPSkE-vN@7(k
zfBOW+*O>0?Bv4vl9+`<vr~=lbAD)t%n@XP$_V(kM<IZ5QW#M-FL!8fwDrkdvffQWB
zuyX%~pe(=3u(v-4V$Y#rzDD`tWuQNp)i=#sKe?pxQ5M$u!z_FoXr<rGEG)R)_Nsv4
zgttDXkt_?6SzxnntjbQ7-*{WhXa4gkU?w`P^%0UZf86><M`7cCC7VPgAI9wv@Jj`-
z0JmCJcD3+6Vr-IMc!;j<!Z7h643waQ-&&g~bulz$=;C4^D$uTwJt=q+d?T?>#Aa>`
z7DgZ?+clDiyJRH7HT!M(42FPR)C*3v>&s9U;@Z30{gX4@0Q$6-72HKU3+<;d6rb|*
z54(*Cem+HoQV|uAq|5a-WMe@|@<^nFi!MBq09%!2tuEW(lggVnom^KsJE+fK*l_>J
zB_gYt^UBu8&Re(mE`s&th5Qf)OqIa!y9Rc)lBaH`jIf*!H#=8u+Sz5rPgE}Ly$9L|
zU~~`YNs-n7auY*SAxDX!QK=V#(^@EVGe7B$9JpV+W)P(G*!s~SFxKl+c@yHOp4?=W
z$)TW5Y(W)GYP6J;6@?#Ic@eb<jlNJ*P1o36$2ur2FDB)2=0%ztT+tG7XtVT^g!SH9
zoG#9cka@&alyM&K@4`r)kvq1J7!hqAGD22I(^G46U&0%qDt6NkP>}FvB?1Q*Qcy@g
z=<0AK<D|)!MF{JzqNtokO0?Bo^Zbr>2Ixc$uRw}6#o3<yKG}(e*9rgI|80jvsFbe_
zl?4QTRWxLIFv|PUz}5xi9Sn?9?os8=t%nZ{YL6}wVpWo`DaZB6Th)okkOMn=xj|*6
zeVfI+w8BiEln@wB+$CM#NB{7Ge`w%)g5*8lB^y&yhQD*4uOaN!ao0mazLi|{TVw2}
zJ|PJ6Y0*XVEHA&KE~%35FJXvy+8lj$mrNg2C(;<<x*+)T{``Dg<idS%ael!Nmr5ix
zGglx<r|^cJge$m+h7(%9K$_Gu2Oc!Vn4lvqQ}`pfMzpJ>lWLy>kI9|bHYS>%bD%W+
z?6-HH>sGrQQh%49pPQ;@rmx!ZrXj%PC9NssPQoPBVCl~N?^K!mxKtUP?M|63yE^IY
zB&XUs70#*aM|D3^GWJcf%^$hAK!*0huDU*ZunJvx2kb77G<UOZh-6bn>!^~yGh#Kf
zx&p@r;b1o|_io>)!da9ix=6yrk6+cWF*uHvOwG;n;ZPbd>mT?o%u|RUMiBwYsdNm%
zfb*MixApfU)%37KmYWi!sNZuYva#7$>Z7M;j_A}kIlxt<Pq+1P2<_9=ZuHw3$luQ$
z+u`K&1yVt%=v~30on3{M#J1KzQntU-XEa(`@vt>NVUi>;G=3XoXa49ZK}`!eGak|{
zEcoP6>kHGttRZ?4Cz?RS^OdFMVB3Nf!fNt2$?}`tuhCYM#iA!L76zZtuCenH!!}yZ
zN7)|R!Vd5EbXNZ;(%*s9fVkq&`ybUFnso-z0$cFd(&vR@<|!6B0<bCEUP&`)=ltw!
zzSWN@IET?S{K8-M3erAESh9bU>nbWU3dsHN)q*{mTo^k)m^v|=^d3|&*(^CJptXSB
zWL3vSKtqXwd|f48p!|x)B~+TW(RFG0P~-S(0#~QEO;IJ8U{W72fGc-q@DSxlA~DQE
zGCs4A690+GiKJQez#lx;R9(iL2rHY#X5x!V79?xoru*52YPUac*}J5#^QQOQoM&gI
z!Nj!0#zv)J5*u+B)?R4Cwj85XbOI0esSIaj_a|?ylBo9XiJ3Gv9?TS%mb0<;_qBJD
zz~|&lFz=~#1DUXYVG4U3MYygx;%&qu!jULm3SpPxK@#l0W~|9Z9%c^N(hl>3<z$YO
zS?V$=sykWCODb)N>!<s^Z*_I^%=iyow@?c6&VQ&p30HnLi}wExP}OPVjnM;lXGaGR
zTw)@8yNl8K{yEVlGacIS{E4E7j=UHFTa(~|c*^GzJa6Z45GtY}n!#2}@sYxZ`o$@y
zC%(dSlJxWV1Bx<Av$9HBrSBHF$#E;)_SLCg2iIc;r;M-~XvudFuMg4~Bc7^^a`}?g
zS~x~`E>8SaNTT_yEf-iWt>e3v3N&ky47PHl`aj{dvLN@ABvCx2%_2QQ6lZv;j6%2&
zPtE(72sy4}@~oe|xyn>;<5k7p#m^L9KDn0%oc3+bL1xjPfQ*2Ru@Cgx=~_GKS27-_
zjNU(&`1pP?;K?3gl9zQFQw9IL{pWqOW`o{7(2i`E9MI4{Du4a*;xdW%?>+i;IbQPZ
zDZg2z2$gpk8OR$;L$bSbv0vj_->}Zo(Z<F`WALQYJbm@HaQ5Lph@dCIefiwr>)7aM
zYfe=fA!25pS81V{?u^xG|Jj6xgX=`x&zt7p$WEtt@%5V7uFTrR>Craj3FGoetfRL#
z#*foe%4z<`A0OguPHr4jcR{}dN_XFu-vGdH)6sAsbGuaZO4$@GUdp|Jp}S8_QF-pY
zZN^`seyn1l7#Uz8yAj%{PmVhV)_C8iNRT3%tz~%rg@XbpIdrdcay`HJHLqb;yox%~
zd<PYQYUN9LBe;;k!T0&|HSH?b3)o_fqP#I~BTy%RY(|{gJ3s~-2m5GeCZ%S2@TnHz
z%IF_CBqBpp%m_Nbwt#b^N=X{-fg#^SoXeb~!5$HR<HI@YX6O;);>kzzRP`<1@f$UZ
z>&KxYP}Yy9I4wxv1v}l2ijE*S=DwT@1x9p?7{-ygAQG&=R3{aZ8}#=jKBce&KRtg1
zAP^!l(M?Oc?|O7jJ5aJX7+7vS_I*4t0c!7#_D(0_#BVVPp!kSPh;}>r`-JYf3%-;X
zeT0R2<qx@M)ft7EnPz{2JCojh{5k*hBN}yrgbxNLy^!}Hdx^u9gXC+K<NgK;m{0jD
z=ZWzz(OM8W`@JMJ`1|AK`A`2l9VeMQf1@zMCmpG@Yn+FsN1Pxj>D@cp3-17>jB`_0
zKaY!#PYjwvP&`enmI`;vG0($ULi$8RF1cBC=>9jlw-iOcz|1Q<kDuGE3Yqh)9b*Ae
zt_I&r!t0+po1$rHTmQM=BpK1imc$3*;QA0?ZOHL<AK>kJqcnB>3>3Ww9Nv_O+!7m7
zKm%#;q2`#|Ao=S5vUAa<x;XT$kAwLm!WWQp?fN%NjAP|roCsIGdQ@UC53;yKob*}X
zpvMH}1e)y^56rk%=EF0qPd-Dvr&VjSyOw%}a*wC6s0C+oAcF#_D{%F>bdx7UN8-&U
zc(~6nmp;gzW-2+F<1itTZ)g|@u^PTnDIDP=&FpN5{>UQvD`6Cb;Rn$)r40%ComaZ<
zc6pM0-3`DJ8N##Hh;4?GR>H8Ej<(X2V1k=*5k5Kzzju!8L)(Jf`!^V+o#0$O@Vx<~
z3`s?4e2;AghMx^9E|NZkZe79~1M4%Y%rFJTCJZ0GyzKmX!_ymQ9S>=H6%S_<+Ub$m
zpkVXNC<X!?kj*!i0!5ZAa`me%xjazjKyJ-!9~lDqhmhX_L$UI5sjl{?6l93idqHGQ
z<mSQ$>C=~8^Z|i^!0hVj=^5lG%{5h(d4a#WV``#Ho)|)SC`g209Gn=T=)XytEmt_p
zi_^3th~>mZEhl=oI{)H@)PhyDeD(o+Tz`#zH=t0!kq-TQ=|+~AWQO@S>6UR|e9-zc
z911Sue%~99tg<C(o_hT>pjxhP*Lz80V&Q0lYowX8urJrHp%gsO_m(y18tQ!jZAR+-
zlf6(#s}|wpQq@occmx&s!qxSLRp5}E(`ZpVTXB7DJ|p>wzAY~=Ib6D#?=#FC0+nTP
z8SD8c=g+3HOQQybIron5H9;k|chesb4Y+{iWo5vq1x?Ip>3N+H$~`9Po6oeWt<T(}
zv_DpEPx%~i3)3b#4K5KAp+ON?m|Lo##)tFRNLCC+Cm<=HAsh{+;4<6D;36Ak#PqB?
z&$x3-=jT`(0RsIXswW_SrPdmi4PrYYUqp#-B0zkCieYJS8O@VjmQ7880O@Z{fhD=*
z#YvpK7Oi4AWnESoD{OH7y?)XMXPYIeYrvGPf0h_owZkBdH&NR>|Djv`G&woh>j!JH
zz_&}4ZEDJ39mw(l=;rAMw69<AhsomOHZLw*A<JmbM*dOEdq07RNfc4%cafURDvt<B
z==~D=72*=21HC^u?u5_%RLFa0C8|O}xVrOuQ&my%10g5YNn6x=bqQ{S)3+tZQiHRM
zUT0(V;f#KO(?}4FJXfPU(YFEBq?|t<9N#E#F#f$-c|;%OSsL9+HR!&g;cWu)gC4b*
zteQ}U{A1cy2`KzGDmRB8KsFZhKV_5zH2>LdlR6bUD9FM0o`t>9ixL=p*=-E7)tN`V
z+Dg^aE>JytX=L>Vvfo^bP++L}9FNL%ZM_Hm<ye^|vG+T-RBxQ8%RM@0O=9oPyK~mT
z4sKvg8V$`2qm=%)aanF@YH(#{X}B0r{_9#8wBjh_u&*md`IqX|a^EkUZAh#&N{{j(
z!xCmUJ~uivU9kCl4(3=e*MyxK_HcZKL1d1A<qlW3pL!i@4O{A``ULaN#VWaNvEzm-
zp&ZV4f411X^*{Qnsg(Pv`+r<X(YyZsDVhIV<-w<I6-@M4%;Dj&?X8j?iLn}g^XCtO
zLe%32=7Y7leyAIwwW&L&h%=*m?$vxrOl0C2;*MVx!oaNBsCKkM+{j%X8C2P9DR*~`
z!DM1uI+|kB9E35^c_gIdN%`saU|)m4m`=rGG+8e*bVJR}O$gb%+`1Jl?EpbcJkEa`
zKDLo&HnWo;W>@x&7*apSCiR;8pFr@2w3Lj9&2KX`r)O^Us~tl|WTst5^KWIa))uyI
zAi)XZj~B57DJ@NO1<m((x!M+X|8@|xZjO7rzle;7`3c~0FrB6%!XGZT#k3PW`@;>l
z6-rwxUUIB}6iOzh65xqFjAw((iS<mNX_n{y<8;HyUgV9wk*_N=5wpfnwpeX*8cb@Z
ztW&*{<HU6~%%R$d7bx?0IynVq$fP%Vqw{`2R+U|Ez~i^~z1APX5Yo-rUc?ccCEJZ|
z+Lj_$A~LMrxZ6}GjI{Lp;PNK6cD46l8|RBhXlxW7a{G1afcojF(#v7#+==oV*48%N
zER3XV4l$5Z<^D1p^D%&qqKl>9Oe?o{*Z;bD&A|);W^Qi6i@K&>1D#Z37RluDckV<W
zR;#Po`%wiLr)Ry1LT<ov!^BPeRp-`M9t-qLHLh~TFmH)qi}(S*nS^a#a_0Mw4K7#&
z%F4?nY2^14qY8I<_ef%U9y2qAcP)9LG(ZamR0j=*Z8Rk9$-7cMqma$%RMVtJmgv04
z_*{AFjp^DnPsjdYAW4m{Yq^u*PVoZ+Z<)fS)^Y7%dD<mgkstQ%|I*ajg2)}V2S?I#
zx$Sl)9#eCGonfj+VuBXGH$#iB({18A6wn==I=fWteQA(5`@?ET=X3rc<srjZEw8{c
z!N6rJC$M6X^N5VfbdmMc(Y09zlXbx<&25n{Azh0fF^^8HjvHMgOH<JhLDxM+MxwTQ
zKSyhaZk9*<_S=I2*#vx=+cRj-?X$i%)Vtieag8t|&0N*CZM|=8thNm}O$!8JV`fv{
zZ3}wZ%|N7c*cwLp2kL+e&8viZ!o(2ZWf(&nr_jakwEP<ef6(<)2?n_m-fqz?1*WWk
zMl2}Cvw_M?5<L!NN~gbz#w)n<_O9-V2RE^vO}0&<A+3Rwm)mZ?!4URtH@%<YDb4Tb
z>};t?z3)hWtzjiWCd;P4d!V0ka0!X|OBWFPEy^$jCUFms4nEa3?##SlJQ=Pq=(x=n
z5*E4k#xg2C+;QOuO@V-bIoNUKQV$WWp4X{&@5;-;@Pw#V007eEAAy0eze*mpX_#?@
z-QOa?p=z<e=?pwW>?M(f<NEqWikg1+{x28c4=;O6TCx}W?`#rja*D10Kn2%^d)wxM
z&Zbl%Ztktae)Tnu)ixIncHJ)pDf`andIJg^+nXBxf%R9cDNly)g}#8TkM!;;o?^%9
zF&C#44UfvoRNX%t?rMGk>*}dzB7NN4?BxBGV?eY7v^~5t4GUfTS=;*BdWz~B4-Rmk
z_mK*x<yz?n1I*%IaXWM`tdg#V{q!C}kCI;!@l1wP#91^kp~HCXg=wV+pjg~63Iu9!
zhiMCbobw0N^9&i%V;NGSe|%aUoM>rE9+Ph|Xld+fCK481qh3H;$@U9v(C#8&_0ZVc
zyQx!e!P#>i2ltP0%Sx5vgC3Sg2gm!?_6|=q#2?7PR$~>?SRZe0mE-{=S^pLXYlo>K
zXOd9KJcH8mvUU{r29V0!@ozz}7+&7AwN)rqWhC;Cr=WnU1NFkKN>CfL*`boErLJ$C
z(0c$>hYbZDxJS^~^%s?dCB*8yG1KQXKF_jF^-0C+1e(^%o2A=Y0-*W;oQFf8t#dF1
z{z(l#KMWc`0SBj$P_4asg%xvYP0Y{viE(v6R6buj;rOP+#l`rKiB_7yo}06^W%hH@
z`?pmU!0blG#&;sV0^q>Q$KM^xmH&L~)w+h5#>-p<pFdEJ!zd9L^h_noK>CHT$>`#q
zyM%pYLPkM+L11~%!&6%}EAq`j9sew@W_y^m?yTN;I8{P<q!oO4%Jl1eL9i~$jmH5k
zQ@-g9R&KfP;(cbcESfKV6~!MDwC-oBJ_z4juN$B4FW07DH=CU4`t@_XiEUUg>yye+
z%u#`-)}qO3S!P#-Htw{Xu}!t>@#W76^+o&K$gJZE*{3wOj3W+uDSrHX`axPI00IjD
zjKjoR_5V3fzh+mh2{`BR{R@SMxS02!pj2>=7PBmvs4*6+ni;>fFfhI!)zzX~ob2@k
zFfUlExKRv5{`rhdd5lc@?9r->+-%%O`@1D&tayL+=l9mZv$R@JyHQw(J6~%iu3Pq9
z2JNVFq-#S`ST7_;^uEH)>Vur*wT%sj>z4kU*P*%w=?m=qsq)#B+Xc7|6<l29Bx-#4
z*>-!M21NQ!q@1XB-wXP;yz1-5wIut3y2O10+@OYO0!i40n?w`SlJDQb5^yvLUJiMi
z`~(P4cgi|9ryvJwFAUw*8;8WR_xJ95@LvG-NzObIg82L_$Fme07-eQ+=<8t_9vjoC
z)xC{HHx)3-%EZ7@SyFaaFsTrxNbAfnqwvjI0fco;p_L(V71T8Wd1l3`c>q5d7$@iF
ze@+;M<pm(wI@tK({kE>sB=X;y-zJ?~fxd~~9-HpgpXKEYY9Cx1Ya7?w$E^e33rsVQ
zWL~JMYCcUj@~OlTGtv3f(4ZaLKj$3!h?*g`Z2_=eYfuDre%*q!*og^skRr$otJh@9
z%c^$g<va-Il9367D-U=bz^I@OKwy53$HTyI(@V5HFK2Xk*o=U*%wh!Jp!+6$#P9j-
zPC$`4zPUCuO$?%<zoBPlq6_Dm^!}WLJ=D^Y4)4D;;hjyqHuI5=t2|GsiwN?Flm<z>
zE`=G0`Z8t9(&g6{d>?UY6vMscIb)UN=d0osI6ZlP=T6k<XbEtl#tJkUZw-SN2Q@V{
zd=dC6@SSpUVy5a`sl|T!El32B@DbvK!MO)b8z5*50+4{$4aey3s?ePE{4ZFfl));(
zr9CusEVIAbi`*tzQ}K>-fw`8K10*`^n_yT%Sy)h$fnl$QgA>?GF)@GRQ(WB#Fonl-
zsfC1e&TRZ2^Hvm-l@E=>Amb49ZRf7_Q7q9r3rk){@a)2{7}eDc=3a_@Huly)zs@o!
zcCxo;h&-RlG#mrjH*?fjflL+<(;5?bKB`58MrbKNDP`q}KOFiBpN1l8&g8G60`$$G
zw4ChhJJ+uXYbUh7**BFfD+j7^=!<liKf+)_!KY$Fg+(m68SUJ46`R(-hZf(jF;8cH
zX<qwP?xIx{6ii}r_V7RnQZ~Ae$X82eWJ{Th(GIt__qTQZY;X6JNQL}KC8e>yC-0fG
zIEkMLfFsiltb>((^tI@5K!0HJp~0P(0-W8swQUS6^!V?Dqy9u!f*KDZAiLWBx@9gc
zw~P#vf8UDZ+TEZJ55W9a?O~{7jpLTSo_Tg*VcQ#^Ko3g;D*fM}-$0YHudPdrKAbi^
z&96ll4pX35`@DE|QL1p}0KERo25oREgZ=T+eMSu9U}^U~r9}Y@{Aj`?U><GFtLgAc
z`(Zo@;cBJXfS?^!bzMy%4lShxGY~K$TZpb-%Z+4O9Vl0L@|G>C%am%S71S%=zZZN^
z?#Rob{8gwT>zOG2U>6PsyP~0i<O)*)dLn<060k>j1+~5k<K_7J#{8n<4IJD>&=rmr
zsgmPHOK%S+Get{7`!H|*69Be%%CDIGN1%p;pgRVRyA|Ic0w+CNuFoP~iTc#-(0X++
z{iU6qo$ZUngCRkEob=A-VrpoCKzPnF>5wZ@Z?4)7l87anlbplYhxlvj1~&r(10N(v
zFc364uJ>C2fC^SvPrsax7U_Q{e~wwpMOJCeJEcr?cmo7BI26NI;ejwemP@=d_+!d!
zZ1k{8*-}2xz}E`+bxysnfiMW9=Ykryliw%U7^NN9{yoWhXBDa5cnvNvy)4a-Vf&{>
z2w&aU5UR6R>p@BJ_TE?7<}wso_y7C8=93S4#paY{@5S@m0uDSGp$P)E>#F6Qj{~l#
zU<ymy_f14^0xUiyE{xPJ2Ov3sWIqysi|vV;`Eco(D-#%y(@MwYLU*t-;Y7Gl-Qp{r
z0*#%jSugOqKq&J4CmmSEa##!(@o`)|Er`Ct1mEqDaFrsD<I&f{Rv?eK!u$W}Qi}T5
zjUx&PyhQJE*gT>P&Xj+=!lK0_7pwPgg2qTAOB;UT>=F^Fa<DN3AW{zhKud$`Ma!@<
zocaFJM5ALK_~x;mZKlS?el5TI#CW#GJ^q^N8X4W<@+JLoDq<k{(AY>aZE%?}E+zMk
zZIzPJNM~T8>MZ01Lr8c|`P#3B{M3&^Si6Tes-4hni3|SfrxHcO^UJe1`^#Co&Uyd$
zOR4sT(W3w9I#T0C?YL2}1E^lY6uu&o=*8nNyIk($6#{gHNo9G0tWCrZ39s;ERTXU@
zv@Go_XSJrVTjyV8dLzVBl7>>qsQ~2Y!uLCRM^BF~+Gy~zfzBav`t&Nfj6H=L`r0eJ
z{K0{1e@`G0A&%e0cw*67i@yRGgu%40A2TxFz`4Q6!;_y=aNk3+d#2ph=Z-&_*k~fK
ze@H6hq({%4dINx|27^{0?(EEmU--OAy8CjWE#Mok1I^wmkHgdH!ZY{6Yy*_0<wFLs
zk7Q%4Z8Gs%3JNz7e*R&ys8_H{)Vq^4zS+ix?GVU|4%>VP0{pwiJdNA<gtA?_@q)J2
z+ug>6+Ktn7dWEOCzVAuh{<KStFo9SMqJV;ZotbZ4rwbx`9MSPAvDS6__Z@`y=!_C#
zD-d3{O;1hQ*T4X*f6IXI3h<FnR1(AkzABNevOZKg33lF6G=WbpT&Vp=jZ2A@`|0s6
zDu%`f-k!#NhTQ^TeJ`3gE+SLyx<!<;<Zrh&?7kG1IIyFz!h9do2I(Ij)?__?jpH)+
z=Z1mNb`sL4r7_l4=UVS{mZ0pw@=vgH#qeIRLHxl*iUcLv0@VkQ{{;1H*a@rF>ABwK
z_DM~CLjJ4qIL|oz`YiL%e(*EqC@9pRoq`&g<nsGa4i}hcq19GWirC&4qGkY)&Rq4+
zFF7%#{fBz@pi&`i#xIFn4ko(xPZ{<kw}HCW4hz#0Gno49aBU!6q128ORu_u<48P(F
z;F7R6X}@#W4idtg_;l@XXEHnu|9u^_Rv%A~h3${zfkg_i&8q}nT9(E?m#LCKz=z*F
z2$&6*dt4f(x#bK|($T!jAc;ZdTDQ&yiVNRK#~YdIr7J3fp$i<$TWieW9S&O;ZqVGK
zogIF8PZ|%7uP~1fN%H&^Rk$<dA?R_$Ngwfa*oYq!kS^KI$10BiSALqDB>xzOjJ_@u
zjkbMW4{t6@|35T+cR1Jm_y0=@x6qK08M61@lugK98CfBFXDcf^AuC(<$lh<2WN#uF
zA$#x5@ASET-|OlR*L@dnuW_E|oX1KRVVkq!3D9)~%Gxh(x<(;ka&cD_$POY_TFtb!
z#+|=(e)aA8&RNJw2c)L5nFfBV$$Us%K8vRQXQz$+$V?;R%EbO29X7!I%L<>=va(D^
zI+Eg38%_IPe}4A`&Bx2?yujEC!1S|``X}Y=M($Yl+~#xRFiLveBJFkCeg$e-I^JyZ
z|4EB9sBZtKIa;mtVTQSz5kqF}#P_wt)e~?o|8x*^KJ4J=CgOGer<Qg2`}bG9yVByP
zU6(886}}76glPZni&XdU*zC@+GWj=91%kukbn{aer>fd&E1^pzH)T{x5!ih^1M>JH
zrZ5Rz<*LVe(PY0<rYgfd;P3O~sWJHJw~W>=J;<rdP^&`w@_oExykpx(O#!xTF#t{h
zCY;`mnsd6k7c9dRkA|oQ)VVI9WbdfnqRB##HvHA8Y-WT^$W2I$>ObkgnC@geNe+|c
zFH3ja2Pi$cGntOPXMx)cM8sDor{omTva{cTOEYc*R$-A(Us0SfK`J<bTKf`EgyZ9q
z59Md$U1an;>45&Y+;3_m3dCWNO6u&io7cZhi#E(T2yk<&Q`+|fdaw6MsL>NBM!<Hf
z_ehta$>rlKz7em`?0KQQlok+&O#&>|-erEXp2MlLXOsN@uJffhG#^!0o)~uj+nE1*
zeYRw~H!RP+n{f6I6uP@G57(q0-bTxC4%wzNoy^U(_~R%*WYM0JYNezFIrgmsfBdIK
z8xJ;rm1Jhlu2hcG(DymiH3jC3ggi|<+#Ga0I#dFhR&V2x?9((!tI-xLr)ZOMfqz;i
ztmA9QqYp)%4GjozKF|>Y9wDS-K?&31AFy!7hIYZm#CQXghuBD!f>78-mfj8Ii;>Ht
z3lpcWo-xa+--qu=VtUlo1AdKsY5sswTpvVIBRd9{q>!(;qCT$W-_gjC94&!n&Ld;_
zZY!GbuaRVcPLG)ue&}^zO;fUl3mto-h4&5)QS*yu;{2G2)?g~h;Rfkmb^Jo~hXq+R
z_3pdar0DFc%POkVNq%qSFfcsDyE>u&U`$|UVeLsy>Mp(kRD4pB?>~M7Gt}-Z$(dJl
zDPZ7X_I9rBmIZe&IC{7WX@tFgeR)(`)R0?)GBmTyD5EP<?LMyYkzq^7PFftZA%VaV
zYzt#s^Fgx@U;OYuy&%lr@<vPk;cckRa2rO1Ln?G^P><gFsxwXiJOv86sD6vedT6=}
zJ!X<~%IA5#1|(D4I|h)Trzf(5yJBOX)V+}Hql3ev11&o%HQCmM(Ppo7Gcz+4ef<eD
z0ujTh$slH+C|>yZ2nE)z%+%5&HZ#penJPBF4}OOZ6TLyd2)K6c>9%EK_kz!Og6!ZE
z<Gn}>F6?7s(z?ei7zlPe^$TZ3#pJWu_?V;%>Go1~twCy3@o>HC(d^sI0QjDcZWT*A
ztA^f;-<_QEi{|?cs&*w?gQ5Q6hqT(B9n}8pjQB@?MPuR9ri;C>xTfQc4_E-CyS50h
zGBPrz2wA^|doX&F-nS}o^=*_-HPoomRAC#fwaN##v5N{{4%n070mryJW|nBoKrEzn
zVR4ZF<HqS7@ih?^LgALr0z3j4Nz&CAdbG{C9~PG851N@zDMm&rekdjbBa3>|FQ6*%
z$`CjN_&Zgc7HgnhUFu29YDnlOGJCOfdgjTv@#_i$k6e`Cgylu&8$od<odeLpom~40
z?34JUrWV1Jnhj;8Zk5;@5v{XPNj$I>it==+JQ)4&`nQN;GI7eMPTX9>{P=7M?IzAH
zt_3o|Mv27QnqyekC0JRxobIXEv6N-pxN%%S;jy;a8f(EGMJ(h1jK-1#g@<<s?`x*p
zg9uaanJ$HZH?`2>!fx9h=E2rQ$pjjw&taC3*cFf&7Z+!~_|+ReB;mHgK)y?hpHFZu
z`vW_XaiT=+KG@TNzs40VGbNxUOP{Rc#Hf+m9#;*XKJ6=eev_!N@oND2J!{zR-aC4m
z*MCGVGEpQuc2M+z$0!Y`;OSz19{by0_m<8_KF%Le2<*;{rc^T%9<7XSc-4*?eF_pH
z0f4jK6jmZUo^v=!ig?H~zETiq!F*SiQ?!Rt?O1rCZR=l_80P(Ln+m3xCbx~Vi^Jv+
z)xq+on+A|jdELLpm<m*e0$w{^^Bop3s{K<_`U7sfL%z1!P70p=o2m5Cls+4aQ^pJh
zZrs<o`S^AZcU3N*X&GittK~kPL=e~oK)Y~V#PFwvmUbL|SJx0(NGsz#>;*>ig(~2M
z_mQ_k!0o2kt6dL?q8D7=9f+UNR;t&nY1Gm7N@e2PvoQ_Q0YDY|dI}f|?t`J2o$krM
z*6*tJHr8>nUQJTL4%Ccv30P~x9$6-mSj77DZ~u{AVhV7ifFZcXo^TfZkeAgNSo?*~
zB;K5Bvl(OveL5k5j$h32^c%=q-=?#EGiq_&I$y9ZOA|dNWt$x+9}beHy>f{EH8wCj
zYEb(!^oKil97L4CVG45aDKFy8k&%2L&A9=W0yWyR2kCijdk@tLN~8LYO68P3=#Vp+
zPZiRBx6k<+?RjZ3HXSg=mR40!bxe1Ii@0aO+7FZlz)l!|L<q4NVdGQ%%+DyF#?PtG
zelpjl3;La$gxAr#m-!YjqC)jmbo9AQH;L$0fjlwB#-v{7o!zCOQ)5{w#WXrPd_nig
z>b>0BWmrRIn?v0saNGNm!7nfI&P-$J*w%Djt~pqq0EzY2Bcqnm(kMkci9KP@?|)#E
zAtNOU)c?+z0g_etC@Oh)8{Qgp2Ulcz$0U0@yC7zc@WMATP!F$Z@;I9AZ!dCg-|p5P
zT?t;GOw(StEt6>}-sE~8d8a#8oKAPe<0_%*xH6)96Vz=MS3D<+lorGhOt9B0*KxWW
z;zbzBGU^-aKML(Wleek|^a_=>Y2QML?*I7Wj3rqdA5rn1Bu#kAd@l@=1@v27kF0mW
zmV}XEz#VvR@ccpK1vSIKt>ni?Y~-*R&p&tX4KATIU)_xo_dc&O(CL8Q5k!Q*xp|hJ
z$yI+2Z=Q@yuz&~!0-A{F%4Zfyia$%XI1s>JnOK;t@JY_Mf0_do4M@>lA6NOeM8+?>
z^$<%g?%JA~pCCu}wi#cE^-bnN)#cc9^`xt|PB%DxP`1$1Nepd!?@S~fRRSG)1|rV3
zxETl~a%T)|w0UGk{hrFU&N{)aSkyq&nmHWygP?if(ycT|xovhGBnhxH1!R@s-p0EO
z<Zc1ys<)Ny$?s#Saw#<f^^~MkbmZrU51yV4z*!oN2cj65zRM3iDEQe1o?rzfyu4ui
z$ETpIR90F#IyQC_7w?hRf|G|H^b)QmF`1A3Tb^>;egi&c+&+$~Pe67+YHYTsUQ<CC
z5iBt+W2#}V%2nZH8Sesn8%_QsSeju3tSqctCQK-3Yg=>$azYVW?QH~kv~p^5)r&Uj
z5q}Nr(Y~J~2ysuL9%1)*M@P@d_0S!WLjuHX>2FMD@<ZJB;gvK9@FLz<R3s)iBJYw?
z!&}jGW7okuq!`XpK2}lnOqExjb&T3ds_4l{$$;6Se2)yNKUDp9)6U8GL+>I`L4YT_
zx~ekRJ2X0|TJ6ikoVT#J0PY8fc=}=;2UZ|lLoQoZTTHLtLlt4?ZzQh85Ca+(2WS#C
zdy79PK`>AJs14<<bRpUrYTWkT$@id4a4)UPiBqHlCIbWOV-<sZ;n^L^r7zIEB((E#
z@n{rwj@u~H2E5LZFyIG_xZ-_gULO9v<(ZxnXGuv(6DOv0_Nvbud%F(S75Jb_9B~0H
zIczE8<8h2ApKA7|YA}KNV#%S7a4~n<kirObe4C|(?tyJ47G?`PJUn#`0@=PnpDU|t
zDyyUX()UA&P0Am(eBQQlVL^aCHCpjY?qU5u1uG^3l7i%%A0eDA8FDOle(OA|)+qxS
zFeuMqn<eq)((D>tY^T%i?#`<>At_IkSebcQCDHzokX;@Y=fE1p*-_wQSr-UdL37vJ
ztH>*#Y(GP<R#gq~ifZ>T(+U~e*wEz1M-PB!u%=R=kj<?@gaAAFTYA=Q@Ze8+#86dR
z8LyDD6$}5SBa`ZJ7v2NHCwF7j;mHEBqmnIaN-OLGK{q<_`dW1RXoU_`VGw&iMnmPx
z82k7*%3GOUefL>swZW@>uk2n`>zAGgLF$07oGlt$i&s#k9B<qa6AOP_8``>%^+SUK
z9&Os(9C_Ni^s<<+X`wUk>+5WEhBe8E^Q5+Cg(BUVdAs4SbV;`?&hv#ecdi1QNMi6?
z7%et@C8A6n(-ivjoxC}0ej(w<`Yyidd;xy@wMVo8I}f?^!WAiCV*&o_zWXVyv)fG#
z4bW{v1PkHa{MA$qJ*~6^BhX|4N}f#9O7DYlV_$`^jR<Sq))rLF-X^>7pTr`plFfQy
z%L-*g?jaKk+hg~uz149tI|aSqn9Q$B$=ATP5}cE?WU|$&CEr|&TQLHu%Ysu<ND6c6
z*PgCEUS8gvgF`8DRDK=MAI7kQKp!l_-`4j-rkRMCBozU|3b|O5OI3%NET&he)IZxn
zXvkOB<Fuj5t1YfC`x`z2$P2I&8NlCx<ly8X!WNwec>l;bXt2Kj<$3)GP>Icj*<3QT
z;u01x1Rv(EKrtu+zooKJPY|O}4fL#v$|WH90G!z8Kc;(OBm0VcA*6qnrp}ci6;h2I
z+$~6eLo&f+pAMp|!ekP4Q&V)w0Dk)!{<m$m5^qrTqz2BGnNhhP+VoH_@OP6XMP8j(
z<sbX4r@HDB{5oq-|Lwv*luAu*l3`-JL}Fc<Y7H(&`QwqVvmP+fF;Gw84o&rLB+gJ`
zmTJe{EZS_CnVCKc7Y`2Qw&)Y&c3%|i-}6u)At7lLen=<gIA0@reyoFx*V#U`(iIH&
zI$3k}JRx<^u6!qydy%!ehDTjPB|kr3K~5o)n;20sR26g|?DF*~(TIyL(0J*hFX8Nb
zbK+_(3)W!u`qMvby#~sWZ8njL&-w7a5hGwCk=Lh}pIxG<{>&mH50aS#lr6AV?eCJO
ze0Kbu#>Nc#-es2`@b6T3kp~a9t)$Ep9~<8}Ju=l;|0u4NNKA?rJj%=SATN=IzW*Jn
zC-r;nv+`dbgSvk<2nox*4X;%%02<!8k4h4(ZUHU)b_RhT|Jmy>-{hukV<#buNieE&
z{7pe+z$g)jb%(kkpnX0*Ui+8cmEV(8bv><emr;zH+_6CLO>cG}DDmF;i;Adnv9|tw
z%G$_P=`8<{I;B@t<DIytLqY-A^1u#?la1;=Vu-@?IuXJ_4!0)Wn0!(n;;LC+mrEU{
z{mDqjwzJ&Hoh^$;^JhnH$nde@<qC~BBrBKcG&;%xpeB?6?p*z$0ir*O+JDe*ia8Wl
zSL0p?e;YPE%N2)JuZQ$P*23EUfy<;mnd&rxYzsQxKG`@kZ17$3MUc1Qq4jR~P1{O2
zpC0BZlVBnb@2-se3LmYmMsBCSYdZVFNkktqAL54oWj@2g{hYElRrDg-qw+~t*Ky_d
zqMx<|Ht(9=fLMnwQwd;IbVL%pIVGY08WL=bN+_N2+D^RM<rs4A_!(&pEaJ*{V`Fwx
zsX|uzu>O7F#`e$gc&gs<@$r$ol*W1tz8Ip?ZkBb5bN_GD?jUen!C#V!`7UyM`?$X4
zDj1a<_$Ygbkku#;1mG`!eM6UR1||if-}2Yg#`23<dCis>F#$g0ln9bV#pV3n+#ogh
zxQ+&~Z?HoVpk~4#pr(`gRq3Pdr%ewji&msH8}guAYi6gX_j<ME2JA<Kfq})a;U&6|
z6vt1{8k{5^3I6u|X`xIeVa;aHNN|2Wx_)2%1Ix!}+S)B{!%<mtY8n~Ig<S=vXI>-N
zAVHsOGBt8llHHbWL21!PtuLhTq5ex~)vfX-#|udj^r@I>4nZE##Y9m~2gvBjX!Yf8
zc%`H~Q{jb?R+-`#Gb5F%s&2LAiIAuPxw|$!DiQJX4sMQyxC0&qW;CSPaSIYi|3>q;
z9DzGvv-ezd`@D~L8Kl1M`CO(XKU-wijclDw6KR?{i($tKY1J)vbM}0}hkyrU0XA_C
zT>ggMMo3qHTpN$=<$@1CwUF@|tBw3w$$I_G*?BRlvgs^Q*pcru;(uUIi*B90mM+pV
ze`R^h5&bu=3M|AC2dB@%j<0Uvg*0AqZ12P_Wy|11Mn18zw7KU0db1R^ljsOtIPLNq
zD&jSl)5(Rpz#LF4h`-vut|Ztf@QYHojufD~T6Ll_eQkX;4CK^P2m+Hi9YaB>zD*{S
z-+r^7nL)_mJL!MIyEMFE;ZIIRZ%4R>g({8v{mn@UH`P#V0!zR#hNoBl4{%!9J_@p#
z*|pSv3Nj+`98Gw4+i#BrDe*Wy?t~Px@ziwgsKX1qt>084rL6qy;_E)&<K*WNM0CHw
zVaxdWOT8bIdoK)7o9#wPACpn8V`x<otEOXRW&NT3=@d8^rl8*zOuzQVpXC3z0J~_9
zBn^7ElDCe6#tImiSrdiv5jfF}P<XKrg>!|iW?;bWK3JHXtcN&!uG|r8pbTl{OzN?1
zyd~Z4CSsBEDaXxz=Fhq%`?(7X;WLLGEYFW8$A(d%PwtU%_eHq@t@G=w0~wyh7>Z^~
z*qoumY`3&z4HMS^-M?^E@9VW2rm4^v05y70`O~#8iK=M^o*q++27>h-%jFq4`3~y?
zcBYc(h)%7Q#S`nb4+llf{f5su5)pr21>_IRJ^*tew~5cmg6?@xPW7kH9@F6np^K)z
zD|%sPc-bXh3supu(V?NF#Ak5V6=LXG`ailYo)hdptF&m(%?!XaGxs}n9>&v<(3eot
zm$<vS5aM2X?Mr+9xF}}yPRr?mi>r4O<+8G8aQ}*P0)>d@H3UL#3>p~Wx_`~iTgJgy
z?V<{pge1eaidCTZ^CLf{K>nJU7kj2wZ-2RC&%=fGh~wwV=x_HW_L)QuHy2}7M{qts
z7??U{>K$j(iH1&<rL)4iRy&sKfKzgroSVDYu`F;o>Z_!rWNYR^_I^uRN~Rs^uv7gc
z{2ewwa4k5`o(>G2@gs&rWEx9ZOUSz4<`jCsaoo11f-=v61h};<7;QX649#5h)to{U
z93i>`zEwkHyJut_<f8mRvtuE&K2MB+N*G7|_ZzG;QDV}PrudP{(GgLS_7ccDBb9si
z$?5Z(p6RO{t}2Ze)ovY+zI`IMsiq9>(|&%0;h;J+>nN(F6qAAMTr}_qkFXVAV<r8G
zd})<^nnCYrGhS;vW2u8!+F21y$5Ttay-g*7Rehi|-tV+~Ircmw?<V4LLz<{;noQJb
z9B%8Ce)T<{1PXCiph3DXf-N>xG<%xp<2-J+4dF1o=kw{zgZ69ic+{4iPNP=Wo{#NR
zSdNu#Z4{wq;@lsro6S;YNf-w8eH8Q9JJk#)?2$>S*%bOmQg>}lc4rSb+#H9l%D0KT
z+|>O!v0KlXJ{(-22Q(D-VPK^_*mWJxYl7vWa|y)h0=}r-moBk23=Bs!{~tec&u&r7
zI|!uXWBT|+<+@yY9%6t)*ZU7KiE*d9O2kOSF4@FXn!pz{|Foz>_&}Bu+hJu#@I^-y
z)D4;7G4_*;h2PrX3bpF>{D^OV>+EfQ<7KplM%+bds^1@X>8ILDaO?h>K6!8#`O{gM
zkS2*~B;*d+J*b*elx+L%V{{=gYfa5KRSQ<53dFWy{%YCaSoVqYu<TK(QOjQ%T3XkG
zAE`NU6z|0u7#Sh|nU{~3Tzp||?(b)hi$d3;1V%;%<O>9Us&ek6Nn$knP&i!u1oQ2t
zqX9>zcnqX=Q{?U2u$8efrjDz{OrPJnEQerDJ(vpsNp7yWVj9PyZO8l?kxw1jX6c12
zc-??Q&gu;OHmf>gfxjd%Zb5=!C*ulkWF+#%q<khw_L~2g+*4%S>sSf~?F#9=>1G@u
z4>Kbl$@xo~Y0<;7txwhf;qEQ<VEJ8089k*J<Q1jnk~ya8NV?Ygxo|Z(56u_Sz99QI
zp>+BiM?~)n`-z&o`ueuB<x9AiTC8U@G~8!%@hJ#cK_D4f6@|rFyIm*h9gvUJK>1HY
zl1<+~w7efHdw2tLk}l@$5&E0LLN7U{JSZq<g=JM$_mEeVQ+{WKTW#F!gSnma&B%kl
zt|5+pFGA!fqWK4=4jK=F!G|AKh@+z#TQdz+HiO=mzuCboLttOzm;165rP2Yeb%Oo5
zb5KD!6#|JyflZ#)mKjH1w!&tr_aNxxh(aVPA}XM-k$3$f1gsaP>tN-sz{DqZC&^!!
zUz%*Xc0R1Ewk}ZzQgBaES?@Y6Is9xhq<Yt`mOC3pyH&Q8U2G&|z{|Sn6}g_iT5ut!
z&u(C1ofPr~-}7K_Utswc1oj@<%wQp9(}D5Y8voo%`{0?gP^MU|;AX;aTdr0wfNfS@
zG$rC+=96(<^<6@v&#1Pt79`4(O*_NyzTgb^_~TQLygozfJMszrsouNWHSiYK1^xx<
z;cJy98Bdhw>2u)s&&|BAIf*lJSXnKJNt5j9MQy_<R=p!QZA>K**<{}O@+-L^%7g#g
z@cd)}je<g0yyhsoVTkXH&Z=gmxChE<<(ny{JB-Q4-@yTf1*!9%Vi)4`h4Z(zAANJW
z4m^Y*Q>~BR+6^;kXe044e}&(p0k{ZJ6yrA#I7WeGhd8uf4W7MafJDScS1Ijyn{!SA
zn9->(ZLQG}TwLC-Vt2o9`92bk7joO*esGmyFc}w>CXy1Ha&#6^V9b84|BTE!PtAQx
zJlI!fZ~zy_*LQ9p4bE_ot~TB(;2}ZQr%_*>J02j3LRMrA*iXV+7BRyfDYGp1y|lKD
z2J3N`!|wuRF`%cPu>A+<z?F30Twg){EY7{{bDi5ZL}1dHDnH_glo7A8_mcb5tV$CM
zWGu1+G3N7#YS)8Xl$iM2$If|C9|b)a7zN{73UX`gjxMu_pS#=ZV|neo`ZlhbxkYm=
zKoZHw_<ZcT>;=)g)<nvb<CUypob?Grbr$W5zqNfDml?nsK<>K#J8k=^sgLaG!eFsG
zIWWF|+_`5+c!qzHt{Qonat!AIJ-_{7M1g(U4IBq|TpT=)Be6Mc?bY8pi79D<jvYTH
zhQuE*?9b@Q-}zUYU$@SZr0Ds&vifo{cmwYua-XciwUwkyH4XBb0mPz<y?=+`CSq5r
z3}QC5_jd^w>EPMwC^s@t($vx-1j8LO*tJ6|ek|4s`D@Ss&1k4ST6&qMmT<}G4o{G&
z3@#=2OM!oxs04)ZataEGDM|1(7)}=!709&HKDfhAH8tyZwp;WEQyH(u1*3Jm1Nk=*
zW$Y;>bkEx8G*0$>w>*ve4^f#Jz(Tgfq+r^->8#%9>l2BK+a!a#<@0QmpNGw;(rWay
z1iFXTV4Qv85hJ-N1bFS+^LJcas{VbvQnI-@lZFZMa&*rIA^4a+Ji>bLg32MpTfAyd
zwV$a=`oNt)wr5RwTu?C6CMu<I{P@=6a?Ek-()GW`GI}J^PXPWO(e;i-|ND6F_eki=
zLZ%<V81Un+`%~cJ3y|Tteo87B%xwPM9zi1JIBCffA7*F@-bgAEiu9O`KY!MY?Bjst
z?JNz_lgKBpt83+-(hYyR)>f~(^%WT#zX;?xPU?T`>SpdRQ!}y-Po8Lanhs5xqN3vG
zsbje(^0*G}Z5%A?7qZCP(5IHx_O<0T>)8d6=ox{7*%x(4`^mhJNg?>O0p6O9WST%o
zibQ!(aCn?2LSuZB8y^O&qGe-6m(vaQL~hB)k9{|bS3vU+KRiAX7x#X8a;UImR90HI
zJg2H>aQcHpW}vK;F>&x|OzW(Te&ZW)G1UW5-|u~CWDmILRgB=kQStwe3lj_NfCViY
z6WI_Wfr}~x$ydV}Xm-21P7vwZA&%818F|9oin?j}!Yi@;QHfY`07jkbf74VDV*7xK
z%6D@lX2(<}a*w2-EbnG&w&!Y&Bjl@lHcS95<%a*Qn89;CvOkMMV_o0C?>oO8K^{qh
z&lW|%BrK776eDBG32Z^N3I&urIyiaE`Y(q&|I%ftvK&3BM@$RuO-%pL`6eFT5!q<6
zV7-C_h3(-yWf!+9%2iGkNEa2_66O)Bu0D17bwA$9);#JXS_w-}Tzp&x_a}LGh(+=d
zCBHYEH!7=c^fZLhIFQ5vK7cbfKCN`nOs-;JU;CTwYmMP!rWHES{b{SI={K9GCe&lf
z%m2s9s`OPx>q>3519XpF-yLC(5in_hmVvso-yb+4SrB~X4@LlzM%Fix{4#sQf54p*
z`7$+HkX3=*`s&c_0`fl9hPi@>g4*W?7rdwDEqfv;ID4jgfAzMNmsQiYgrD4!Coreg
zwrFoXbGMl}rxyy7pj%s8SzfkuWsH=`gYLAd6DIh0!k&}$B)L3p@M?fApsl>BSYJjD
z`(l;q2umYUurl}eeswaRM^j*(Kti{4R2E!%F0LUYBmY5k>D-w_kID!s*AZyvK~Dz4
zxSms6E@XtC2k{TAs{@1V_gU14oV<Ldb7q|8VEGOUVNqFp8^TYdt9yUHE<Ezdy~`)$
zr($rdqkevXE0Fs`28OG7`_paCu<rR^I?qgqAxm?Xvx6K_!Q$}j6A+$&h@1(XXP!7N
zTMIW12Q|jVhEMdSiKR>LldfLzxn|WT{pWf8;{BiZR}IB_&d%L6Q<%IXqYr6m!Iw?d
zY15Csd`DjW1slx49gG_P_83y<!e2QI&0R~fzRjdL@23ZrsdKmvz5064%UZ%hcoP9T
z6oxuf#%E$57oQNjheOuT&2(Uf+&cBlQM0OCWGB!!($ji9@~mMVTC*i~Bch0!&5w}}
zuldKt-BX)<v{fNOaSM4D&NdVHkm{0%E3c!o69ng2=q$kX;1J09od0*|7rrRPXzG@x
zdlGd~5hU`J4||cZLO|u?rDwSfda^E|b#=r5rR7W3E)`zv>SH4+&z_B$zP229xz9};
zoLM?qz4i9bDHwpk%?|5&`PDg-f&Sl#BR$Py&2L8}!qoD4e6mMMjvT7&DId#uYx^=I
zemm+`bYNfNwR`O7NDI3@TOClFJNaWLwoi!3z2@S|GJ1=Seu>0vzt+!l#*UPcRZ`W{
z5_6tjQtdV40{eVbaRqt#tY|hKtU)v3U}NZKP@a2iPyU#q5^5*--DzUvOl70%R@}u!
zs9)wRy{p<+bmi4Nn=at-#dpE9vUIV080vSD`L=x?c|KP+XpnVT)DZEqt&}FRj6LdJ
zjHL^SAHLC$++#yLWLdQSNycv@1I_Bn;B<TZzn2-4Nl&w*b3Ns)?xWmwN2`QxlzeBn
zx4eRqJmhk8bkw+iH#88Zy?Hhz8*f2;4m1A2@KalZ{@R!B`hYj(RGl~Vb6PPLl+j8A
zdPN)#u76Ha^BlY<j%BnOG_wwu)64rTjeRl!J1k69XTjayPTqG}y`8ml9u4VAYPI*!
zUP@~<{dWOUGAK#DdgYWC{3l(+s#|{<-bF;XW@(-I_*;SwwR#7SQ0i9j9{$v;lM1or
z_o}!alN6_Sa3xSLAc2$``8>Y9_(6+GcNECiXpi>x@aOmJ*-p{ILZ3)c*uMSt&E)v}
z$wg$g@89<)b9T*F2j?s2W41h4v-SCxw9vp_g?fg;#Pr>dj4eT1&(Pf5?!1V3mwUD)
zWsg?m&OL<?oc7*siyL`4X2Q<K-tA}9r*}cOw)tM2uz9(=v(s?rg8G5<G0_2dNZ{xB
z(9+VvR>981tj_Ap@Sfn+iOVbfB@gj3slAZ(zVlGO6{d%@-(&3Bwlf3p->YF`hpS3q
zAP)8VRj_<Q^A9<g>n%%Bk7YDN{#d+{D*yY&>6UjT37%(ad1FdZ&RbOoT0B`YaUMq=
ztw{ecaxo!TV1HNzyZ%Jtps&v<U_0{P5%s%hf*{2g<vQNO=qdk5s9=0P$$0FvGWp+)
zll}$_Q6~5%hni|?;J9S@(k?V2D)!xNSzkhiD=f6=7QeHKKka$j%V~=rJ^<e8EHFuP
z=00(-w&t}m;a>N3@7a5)hgmO)?=MqM90hGae&HLLY&3s=t3S@P^Z4q0MySmE2Y%<<
zj-uonZOmg!heA@#!r?>5oHX_`KDY1O&xSZ1<HRwBenV_0Km7fI)=fHe#+MG}3#AjQ
zmWUsPl?|WE{n6BtuO{WQPeirFmKUjSdmcp6IGp*}8S;!<?&C2cIgZ0A5mf_Lz7n*D
zbNd3w7uVDMSqFi+XmeUgLTtXZ?q~7m7xF(P4N<^Z?cfM2Mqn6gZK4xL`DM!n;hon`
z9NJ$SEBM|cPULC%S*I9g7i>?Ci@M7-etUVYB*jldC|`&F7(>!uN433(m3#P=G&Y^Q
zb8EumtD=0xHTm3aBLl9vGaDY_U|^(+P(-t3hHG9kaAI$~qnXHV|4rMI0KZU1iuw+k
z|7d;XAI-O+;b>nTxsSy^A==f>Dw5&H^ZRnea&~W%lB{8Br!2}94%9-XuH~nj^yXq`
zexVQ=?;if?>K;%eZg=PZud0sa7uX*A(aE=7>CLw^K7gf%>PKZ`)zF6I!iMCR3NK~!
zCAemYUtcGBhbhJW)wK!F&4(Yz!-V+RJpBJ6{-E(IEM$<C`tk5n&Yi$jDTrRb<{zA#
z93ON)00=vML48S2seA(Wxq!7lZo?OBEYIloIC^ofAp8j_NiyQENJr(RzM1Du7~cD$
zk|x^}4eUvG-DS8}jri}iEHv6wY*^zWovB$Is^$M7{#k1+UVo`(d@*7>XPs`0{XVRQ
z0zQH>Vksxq-}j@T{hI2+&T*-Mm9;4@EGB|1dw{q4;|D$0G7j>&yym)pN1_0dF?=6R
z76(cpIsm}*^>J<xE)d<z&gSB(oWJ}2Mi5?OV}48;g+9gBTs-N9M&Z)C$NB74r<P1i
z9dLZ^hA}~b<kgIayraOJdmmL=rOeHF!N-dT-`TN(K1TjOdB%b8P;)W4hc^zAE8b^S
zRNQ^}`+bFz2=L5v3m>o_v7MU&bn^Pq?V)H|hz!lhD}wd#!1##1js^}QLQ%`_`Y2W6
zldCJYs;ZWgyR))Vzc)S!m`}^xx{{DBfODwJRlMa4t=YSRO0(ea2o91lWC@+}n`eVl
ztckPv&d+tw9yu*-{Nx<E&O-x&YKT}2HfB)n?HW>az~4=iQ{HG;Nx`bWO@r=(yuv?C
zYo8}bh@(Cmo7t7yV%cLSRuf)%Wio(z#j-Dno0$Ci<?lg%@jtf~c)OgxjSselc;Zm5
zzCJGmFeLca!K1?)JRbyXqPG7Oms!2E6P)+@bu@I4K9h58{Gbv8q^VI$Jy5&GkEI?~
zocUngM<c&gx-5V^=Sphn)m;~JQ9Mk5ei9nY<vl`Cz1SPR=8&bO5@=wo+CI<5NDDL5
zjd|mN*mg4PH)|L6Y%>^v-xF2dJ#TkDX#62U63DPX26GlGx{rZw=FILgKR>HzZ0qcE
zV!CmX<s)k0rPw!3A0aI}<`Gu(TZl)-ymY2xMQ`O`O9-d<EdH%!sdpoNFB{s|;8Rb_
zq>vpZrlN@nPb?NSb<NqOX1^<{-8B|zKYlI37px_=Z-zQ%LZDeHjp7^nb@{le`gCB$
z#kb4(dwD=(dT^7(oP5i5OyLx$Xm}K2X`{V?owu@bau{G~PR;YR{(rBU*PoqNCAJ$C
zTr)|PhemJT{mKzv?452{`!X17@m@{MdX2<n9)IKXU?DI6b9Ffi1i{Z8<eofX#>glt
zseZoMbLh4TiyI^(RRJAB=T2Qy+^>B(=FFa)`q(H;6r`BOe^$GfU|&2_;kAv_ew;i9
z?IS28q6x9|uSnvsXv#7iewU)31hoZYc!D)ebU>O_ei5Tv`ipO0`(z-C2u9hRt^@2{
zCtCA`0qc+@SZBt}=a~}!oK|k*P&9hlK3!_$<<(eKr2#A4kK!NpUL__MDrmY!X83%q
zNq6-&=~_Mb4G}$hzbAf{>c6S*hx!HrDTwI#RKyQeou9WBd_wi4b*+lWw*M@#OU}r0
zBq}0rBcjybC8P_u-TUH>(n0r+a-;O{sE2@00&Bj4f&$1C_oSS<g@Ex;*xQKu75A2^
zrmRvF6+M6dBO~{LN43)tDfXpl+}55FYOAB9?tscADK4Jfih&6Ro>t^b@qKl;tXtE5
z8kz>c5`tt~@YS}l%)mr$F81lSm>)sn8h530a>_I6?D=Ar9+_w|48-!BBZ&pc7+hJN
zF{VJ5eT9eS4=WQmXO4OP@*)xUr!vbyG*8rfbuq9rEPO;k!8P7omjnL7t)}mU-{Wth
zLIH5)ic~ggG1q=#o${s$co%IcpB>ckX?%4)+y?VBasqOqly`wIB_7<4i;t&^BE5jw
zA8@MrdI7qJpYA2ZvFZw72KussyV12r5^7y==ZCo~dzF-cWV&@MuJ$eg&=zbg91Q=V
z3J5&+qI|9PW^=p(?^9GwWjUS?Ww}SY81v1_pEZX;doQwBY!D8skJ_)KG^8{l0IMLP
zPZ-PGvYYV{a2={MRAz}L5!?ToO>ewKlc@MY$u_mFPUnd=^Ho>_v)O9Sj(w|`h&P$O
z_6&G8r6qoxkX1K59{74rp8QNNp?LBEGjnWwEHfJ&--zi0KXoPBsa@Wb@82121howf
zsdYSC{cdISH!v#B<cM78appVL9;Kx#F-f_-OEKpO*hq`Xv_&O}+~iQymL>hq#)qG@
zc4OA5zT+t5t6A303TH=Z*<kltF;4VvW=T>6l<X~dJ}s_GXs|-6IXv{J^yB8|zCS*5
zfHHorrMwAH0Ruxv%grkWI+-M1pNj1)xcANd`j2{VB2?@%GVMRQ<(|M+R&BtU3QRS*
zg!%>^*<XOMa3o|nX;DT(h@N>7pQZJImCZ{rSOMARMnVs{g?Sy=Hs;}r7jmuG7JbDl
zq4EU@WoHD030(Ihj_#;poHoz|dtiyL?j>dpBG0ex*kcC)N3!C1pCXm%i^cs;Xc8D^
z)FJ8#-j?uKQMGT=sHNMf%1iyvGN!k-Rz-~08Ol^!FE&Ld-H!vqvwaL7-@yxM<-oab
za+xUZb+#Gs&pKC5p`a8_O*!BzACL&*cvjFrk;TKpLCs&~S2P5-&L*ai^xqRgK#dU_
zU|{1t2l!p}nI$zn7mwez55`SMTyfW4dYhlv#{84?{%S~&Eb`@zAUE0rMp|52yrTT#
zYyKU}<Atj)k?|_K_{!MktT*18uD$OyNdX;Vf`JE_-$63Usy}7>w7_zawunzhH!(<7
zinq2ZE5Zs4EX4gzWH$5839R`|Q|<3Hr4?TM{dE&^+qnKt`fM+6G&*o%FMLU}73XN*
ziKSXzS?NvU&f{;y5XFUyeYwdE)d)CGcI(TtrXapw;x+kcuO5Mmn>BxTe9CWhcqIbt
zf$1T8<o;hneKo?ve(Z(2+sD>KKE#SIc8xk0x0T4&&y+~E4%Am$oK__GQl@Bk4q@Bp
zaT4?`vf7XN05=kZ_-8S*R3E_^3Y^pq6D9tKkO$Jx(Lr*+Ln>ilvyDVdQl`s&#oxx>
ztxyQt71!yD-exTOW2(WT=T@~fcJ?uDN>xXvI6HfR^Rpk3y*D|~RZt&*Ae@nZJ*2%q
zS<It1tj&At7OpEV{4cuLUqeG-t*<}T7nH}KrSyh|Mbh65y*oG8Af%_~!}C@OWY#>q
zd_$|*oiD)h;w8jQp)QxJD|p6+N9|tPLE?0LT%1m`_no_H1aOuWqe_(xd?dYuZ4V<D
z{d&?3jAkG-R4Fo2Lo=Sa=q1j+D`s6ig2c8J7L<Ugp0l$fgbhU<)<FIxG%}I#3lUba
z`5MW9>)*{;$7728IW$yQlpXX!41R92jA!Yy$_NR%@X%1WIN+LkXFJft`2{><FSvAn
z5-!TM68}?I>$31g%9wH_R#a5ju|;{W;1bb$5r#{D8d_Eie$%AyM?}xb85US~{J>pX
zgu+NiLn3cD1c&1DPCo%d7^<K?sdW}&5L1$f(H^O*b4RzYS=+wu(UbtoELeKsU}Tpx
zr?#0u7{I=}D+VHc!$aFbmN3OYMI!>8nA`I>z39UsvDtM5(hCX;nJx@c8^03)Mqe*5
zQ8`}x<{LF%xM-jCcVOnJ$c2YhypPQ$k8NXNVk02JPZ6-&@A`BzCg3Th60X-XQlV6c
z8;dX=husQ)u|d>5>Tfhza65t|21f^n1~&tOV_&1#?n=LpjkmJ4*49$1Dz70E_Pjpx
z2qeJIUzm&xO=v-o09_0r>#@)5r0zG{+h$yyKYui2WrZnZ-`XRtv+H%Di<QqC=H(q#
zioj_U4hu2nx0^xtcb*Q(p1+HHu^!n5`ETZ$lhbESP0i9Wdlp7!Z8eR_21ncqh0nj%
zy9!^Ti%y|kh(|xAK^NsGlI7JT&@^Aj&3(ZXS>pmfvPG%1u+W%G7u*X)C_WgeNIkqu
zK=9Yqwdx)Lfr@p8F**WwUY{abA$xcQ$I6OPn;tVs*}4Ln`hj607YtL7jtNZ|VNDt~
zITe)`cx*)}sl9K;CQwy*$Ir`ITvB46H<bG{X<)4D+S4yU63a4y=MyGSa&nu|>rPJS
z{>8<Lz*JyMc&4NAZPrQF15da1#r>gUIYs3sO5KS9ecj(CA96Fx8a;JyUlbqokB)rW
z)zfui+a)Dxurc}c1SGk?`L7BJi*b-pc}5ouXcks2kKpQOsW5=9OG&eBqARPRK}JI2
z+aQXPDW{?VfS0b$uUeYY5DxmnKjmhS1Rp=|bc3;ynu4CR`KU_tYwyqndo@*ER(j_7
zb+hO|4Q3`Lf_r}rQpx`=EQm?{fc6VJ2*kCYvP*(I0kD!)lCCdFUH$u<o`%(6W&nW5
zGPpX@n$=21Ps|eTy$SFUr$~^6%jXgt7Z(>E8&F(~6Dmcvyh%qV18j7AJ1a-WikUh$
z+AFGIa;08Yf_P16WmjJo!rKiYaPW1Pf`a9roy@^yrQTk`7j$w)_y56K;j!qkoi^5a
zTsn&8=F;x`7Syl((**gnJ}i#__$ENs_hSOvLli;6drJ#T2Zu*xesy{Nm)d7(^OA35
zOuC;|N=d1tuC6X16<5u*ps+Ol{d?GoQ=R7g9~YojQ(aS4Q`OC_!jx0@Mv$}(pV#@g
z>Tmv%4bCffAR2b<6ti-?(1?$Zk*kOABe|4VP*?yNbS_Sx;U8*gsli&+w0vB;jn?E|
zR~;wK7>iK^wIj9Z90v}F0ZXGklq$<^tyvy%a&lGG@Su~G0?9#&+SafoSti@+WqxC$
zAjBJqlG~Cpce^oFf6l)i-QJ%h=(;czv>Z|mF|3-JpIVCWvH6+VScg@#Ne5=`>$XEb
zonKG@&pqU>gD?JA*T<Rlsll;{tjx?itsGvzQBSTCwKWY)Gs@Oh*THuR9Ee<8t3)vB
z0n+{7ZY49XsCBFQ=T9w+SSxemmsIceZickhHwwZUMN2~qQZaRtQc4W!zqfUwAunI9
zIf*aehe~;P2pW*Hn$t>;kg!D^I#pD_D+4eO6wBt4-Vu3IE0>0((3tp7f)vq|12fWE
zTDe(Sg%yQ*@K9fs_IA|E=<D~uOJ7q{nyNr5^2Ii1={8}Ct(9F@t*~G0n>TY^t0emj
z1EYiQ<724qhQif^SsnICbm8Xa#=(<(DSCRaK&g(e-@fMckd5lUdPX#V7&nc&fBWIu
zZU6ZSLEX)aK7EwTpvpuwI>CvtIn$u8{%mQ(`LVpbG0bDod4Pu)j0$4N{@}<q@wkzU
zR&GYdzr&XAY_I2iLHX?H&Ai6rLCDiYyGVsfWW{sZ&2QW>E%+jiK1kv}-^05e&QL;e
zm$%e(H6ukXz!Z&q{qzpjabS*yFUu*`(_5U_w8tk!j)pYg!#*c8e4Io=ibpZ-2dA^i
z+}$iweB{CL3A67zDvIcI0Xr8%Ut@dy!jEL@bP?pAN^`@wOcn|5eabb@P;zprpb5s)
zQqy8$E_#K3m?pFKc-m&@RoVSvI$FBCteng(aVhG~27t>jMZHwj)h+CQ*8&UTS2(}-
zijAfwK|d9`xw)B>*aEb79Rx4a+}&N-Sj&i4IUg{CGA6@vH;{8Dl(8e4D5MqlfEe<{
zA=6_zNOpG#pJ!xLKb3sI=}+v^WS^i6Fy*+C^i_3vU+g<Xgfzb*4`a#wc7J*ErHM_4
zvGa6bGduH^+*6Jb(?3=6^Ai&V9lC6sf{}i1O1RaT6ahd)t90HI^NyIil_EdiY#4${
zh>y9?Yr+sJ=Y2S*1Ps?PxqgUjL2=`~egs^Rdo8%qZ*T7yJyKRaJehBlaC^>?RJ)0J
zb<an@XG$zBz-!iWyLN}?@96A2xm8L6dz!dpzM}%Hy!|N2*xdhoE=O*Y?8+4r35m6Y
zLCgu5u`n<*VoWwF-;;YO`DsWdB83mQbiaR8w-nkXlun^m-2Hx3$us=={6GwqaDo12
z@k-3}>tvGnf<CwqL!xt7*e?2{J_wZq3!F((lI@H{$1(%h{&hlrLEKAalYYc+KSYyH
zj*`l+_~}IlD}0iXY6E0^vT6Pn;wdzN!1UyO^do4HCZwRC&_=IzxA^GJ?8yeK?%9sX
ze%Cmb<OZKYEk3e0km&5Uu{|w?m#*_G-UgrVdOqkDAKbj?S&8lcDDro4ZrRd)&BtX1
zOpyo0oL?x<e0FPtvl|R$h0N!4wzAJ+e)W7GxBR<Jg}{+3;`njwyQ5*8a=8!~s5@9|
zvpHT3HB6GLslA6vYsRIk`uH*Ysc_a)G*CL|JJ?%YeF=LJfFTq5($)tBRUSV!R8UjX
z)59i^zMa^)Pa;^{Qlzdcy<o3SU;1n+hUCk5$DhrEy}NUFp=AWk?7z6t?Oh<-jMux8
zM|TT<_F4nU#eY(BsMM4}@0Aa@MN56jhH83jGMW0#4hm0eX`V2IN~vk6a14w={6pyT
zGMMGqj70oeCl;!!58%KPqsQ}2oi4s<t;LeeX{#n}P7p7%C)P02JS@XT&&*6i!%D>#
zT$)VDXJu>UdogtX8d`dS(Z2~Tz^i62&o)nyOxW>b!qK2lmqHl$M$2EMB#~Tj+JAVC
z3!j8d;XSO%-?oGnu5La3L<L`7C7es#P?*gzH!IPvH6LwY1<%11iMfQW*@Bh(;Jvii
z@})2Qtk2dD7bC!NXzBdW024-~@haP=rs}b_;?^;*W#Fl}BEE7lq~}V4gOiq#iqrfX
zG4g0u5PpRh<>Nr9w6?P%pIs?8-D@zG#6WpZmALmOs<uKvhBv!Gi?jG635oMV%^9$M
zVAH9f`#Vb7pBWJBi33CxG^C#@05Wz{1|0sGwfT|~`ku#?B@)vve=RA`mW6>6C-uD?
zoT*-T6e+OUfH4-Al#78|_^7jz2#3yj>4PiU<ELd)>q0AG#yXGw3|CFB+zmpVJvYk$
zoNo`5NpIHtQiM3!TW+`c1<OidOg4+XQn^}UpT=Tb-L3xJDuRK+#l3aGWK&q!a0oLh
zVrgqNRbku<gfQfJUyl`4nZ%$+w#gj&rlBO~##DU&{7yxMxIv3gzw`1R5V%^fPoG0{
z`THyhS0!6zyJ>J7eTT1LWZnIGldU7^Q%==7d<MX$O@+~$?&woIRaG@U2!=T3jk4+B
zW^CW*mKpu-4^HXOPa7%r#KN0CKs7y4Jw<!{^{~e0CU!WR(GmWn-)xgE%&i&I^5Bm7
zYN;?ORP`S(6%MqXOWUq<+M*lPffORxDA$Ezi2Zf?cpsEDpc=a_TJw*pHAnx=th|6n
zXoijb-{Zh#iG4fdEP-zA!JSZjys?QPll>03X}6XuG}R^N|Hk+<2u;jr9`--hl<xf2
zBX!U{&JdVf7#5Ti+NG_Vg(Af_ZQT=Vx0^(zCw_z$C`s7sw0K3`ceXklepJhkTfZ#t
z5&YfQIL%yWI)C#mEA0TbrY|S!ebEr8HrW{#muQaxl?0@HV`F6D2_b@DJ8xLn(tei-
zxIPVQuhG%LYY?&V+{BE5iVD^YTF<m+W=2C5=}1XeVIn~Jo}o&b$q~EB2L2cO+kkjd
zmEA@?#SH02^Uuhm`UigFBHkx#RO;ja<^9s~y$<V-TH{$u@J-LLFw;#;!m1s_dToDc
z`O<2^LIQ@YC*2*FK0h?}*;BuA@beOyWW(M82u+}caU@Mij-f_(QvT25z^x~L7OKI+
zSb2Es%+SiRHp4*B^SELF2P3xcxr(RILzeUT8YRDA)Y-dtad#8m_1qd;MWOYc*qK5r
zVZIQ*_=2uOCcwA}5O9O<X<#98_r2N=pzyivmBNhDY&%0k@9p~;9}WqsEMcryFmtD7
zUY4B~-T~mK1&#W(!^&i*k-<&l#3YilG_i>(E%FU9gTGdn9zQ%H$(DF`Q&Mu3N%)~k
zq<aYgn+(X%R})VN|DjGFD7luHJGyU?iy#*(2Yq#hKn%Xx?hO7RL-y#o16HuH8P|Vs
zZy-eZ8T>KVJh%vXZIz@ef&Dl{r9i_)Zn^Om+%qsLY@dfyl8KJ)#aJ2p`+Cn)Q1!o)
z&q>kl0~1&fj~17f8h&PYDCB<^&7tMqgX1WkSAPFd$2>^<{l0tQ?te4(T=TnoV%Ty9
zcmI?6fE-bZ^mX*yQNy~``tCVyFj^T+BPj*~K|}z9$e#KB5==J?2_73cT>lci!|~OH
zp!v=p5wFXxpQ&daBGd1KoGGU3O@5b!<fKVaKjyGzOz2+d*4cffqtwgKv_iT4-8zRJ
zx;0UgabSP4t(5=&_h8YXi>c7hmGyrNMyP6lC$sNJ`tkrsKw7hT`6Q&Oy0!&60s23+
zb~=@i^0HvfudOx8k)$Ys=Lb1IB&o`oD!`9PAxR0mr>3EAVs9Or_>p9P{|1DwF$^!e
z{H?gNySwo*Y525J2z-V$v^2B-u7Xwvo_J2j|Js}St*j4pYhH<oQxHSFXT%_Op%k<+
z%a|D%Qkz+eJ}>KLWGs4A_y^0?b#(jfvf+o(?${4;@kB%k)3>lz>vTj6>u$erUpQQ|
zF32k8xj2&USf{38Js=<ZP4h}bjFpBNOe7J)o>alBqx-<tcth*f+%V*%6d8r_Fy<{h
zxZdseaav!E^Oy3lQH3h!2@8(H!&9^6Fm|66=-fi7Q6i!LyHf21`GqH=n#W6DI9PaD
zX_nN*f{nR`hWeD<!mzPh8q}JD6tdge=ry%8p?wDSM474RWnO1-e`y|nVh3q~e8J`Z
zH%^>06^hQQvB$kyZyxTTFI+s6Q|kY_^NR3AuBG92i;TG@6wK6$_i8}DGcrg>mjoI(
zu#verSCuOT!NI4~1l-&n+-cqVqi%>6FT*`Bys*@HO3TB;msPa}R1-EklXJl-9RmX&
z4-1T2jPmj{k{4|4e7w9w2_HJ;o*X6qC-)Ba<6&J)H#`JqJ<5CqSMKYADrs`0u6;zA
z`ce!CNyoqR;_eZ6#?CIcbc&bA2E5u~a_I^z=kqB6N2g;tp7WFLTCm2OuCcP!Ic^`M
z2?keUyJ-|VHPee8hNt(zri;KFLY7tos=78YuZ#NaxP2VUZ@anlS;ywajguSBFLKSz
zN)20Z5uosbd7DTG-iO;nEiM~;PBWJO*ly0;u@L7HNhX$q%2a)9*f%)M$3`XpwoK8=
z>IMP~kzy1Ie62s_T4dQMAdep0w$m4L-ZK94YfszLxtWJ#>XFgYqG&+D-X|n<U&P%)
z*o{IqCu*pUPrAWHYE}}Hba`b{(4n|JE1xI1xkSlHp#jJVF1ZO?QLs3Mz-h`m1nL_4
ztLv6v{M=CQSA6^Zk9^@)g(udeuOD%6a&>gG>kHTuLOT<;CC3nccdDR(3Ee+8S17zK
zSFNCz)lc`Wwu0Kw%+_wHk1T-A-`l=xXvP&4iT_rxbZMymRA2F)8XztrB3U&7;1DXO
zb5U^#m1br-_VuhN5{BCk-<hk~`=h#jI{L)Q#j{I$PC%^v7KoGAfo)o*W%%y}Mg144
zY{8LaauqrD6+f>Z6GI#m!4CRnQhvBG!S(9PO#DxP8^BRfW<pPVX(~$LQ?F4)LsJ%H
z+Ty>JA`GXUlxdxH{MIezrpu`G!AYa`Dy8%Y^Vg1}-}C88Qkwq$@{c<-3vsoYfB6>C
z{IeTUdFtnx@1ffaTnwK3R4lcpFp*rXGGG1&a7wsLHsHIGn$joQV+%yv>uo71zDLkn
zfRWY>(M^2=gV?xaG4JK)v-ib)S)aL6z@XmXx2EPk6WDl*4ECxGZnISxs@cL&_upQB
zqh!SVHJ;wP&%*3Qx<FI=uN`eMY!egwTNJ`UA9@ud%Z@$w_<i0+2(CC&u>PLVNf-10
z7oQ2AKZ)V7twv#c*wJrATn~pp7Ex<^YHroy*}6!628;L$d!rLSQ;+i^j-;M}qdPFj
zDulJzV6*{0kVD`)>r*m*&#6`OmIlqsPv_2jB!npt%>K5g7Hk3Ak9G^q{UV4L?;Kvv
z0~qq`*|QMb`)1ar2CACPHobH?rE{xu`K5)}2=v7WhYj#6UmXhw0Gu@SAr(TxjOuNu
z0=_n_Q92Cwr~75eKJI^^l#T=IcHt@mAR$Dj3-NCq&AhnaYdacprHifiJjsHQ?+5L}
z_;^i7=Ui|^^H;V2oc8C4>-P?`g}-V0g5n<E#_KEdz$loAAOYJ@8HUQORLg5`(nv^2
z(}Z0?^azV!N`W~rH_tZdKQ^4M^Z}WJ8Ju9jwq8t3UWZ}g-jfzE6Y;a1Rn^p-P5HgG
z;Q`L$`MHz}1`VjRSjBNYSlQTo9s1qm<#!5KCqM{+-Op{HX)ruFbnV}Mk%t5q*~P~G
zW#i`7mj|#KP<;brEc6@Pn1%R^ha2Jnk7`)Evb@@41yv4vi^JGYTx6S$<-G60|3_|a
z;c7pA2MD@(8b0^TZD_2bB<Vk#b85@0!cp4xgZ{*@{>O~x>~17@)EG58gehBq%Mji#
z%gs~qER~KWWpV%|>(h6yCaa*u$hcqzFNsFSHqR3}8(5;$Zi$|Is;fE{bcFuv&&BUt
zd#z%iURHE*<(HgaUCp{|h4uPTozwL1@d5wth0D0}g_)e3Dx3bVrleQ;JC!y(Ti?V#
zXlo>skf!a3u~4z~nUx1__5a}r7F%ZWHO_k(yX|#6<I;I`*UeYNL3ADzOca}y1D<$#
zq3HV4-yR|kjcdA3imqaIrO%^x0kjgnj1Km|Ip_avdAKq8ebFH<{+?eG*f`aKI13gx
z1~xoz4X+<vHRF){`8hU04F{Ev9oC$n%e&YBTL29qmH_2Z41{I%_QUS3uJcda;Htj7
zYI$`_Q2Ej#^r@nt!}s4y-@h7JX6tfTEQ0(8=(WdLS=6Xbaf@s~2vZv^X|HC$Y3ZvU
zyV23w`)W+2W_^6dzzWVmqwC+Ww)t2-mbG=$a|i_5SE2hytCm|DFQI78<Dt{IUUVIa
zev>AO?Rj{>Sf_87AQ~i@JqV!G@xkezrLs5aEvHX<6VDj0j)J-yizlnEDtmqdpdIvN
zZ2Jl@jm`ksr=GgUNy@buC5?C)L7;`tywT!y>*D%4$*`Z%h<ONG1_zH=_n6J<&N@(_
z;m^#Z#xv~}?~O$^W=@<&x4q3V-CE?&e*KfFRB5Q_A>AGqhZkchS|u|R-NASNjd9ao
zUN|!`SzpflwuG%islDZQjtB`VZS65?b~(&OU$Bq@>U_p$9d$_qAmLMU<z`hVNJyA`
z9K&*XklF|EYKgUFT+9b2ch^NsEbn7tP+<XN3?fd@X7lilPJJXFTr%=DaT-6vLXP9K
z|B+^q@%`$!m`%ZojDq?F5H@)UYbq&?j{ZXH93LAUwOCrD{gnG}18gH<f{g$TIAi3x
zs8feeLQ^7{p4Q!~^9}7mt&^s0P2unf?Y1FFuLm=-2d=6Nk&;Iv2bShFqNn?To#wa$
zPSx6#tS{M={~t}?0nX(cw*4Ua+mgr#8D+0RwnCIWGO|i!kL*1XvdJcU6SDUxo9vaH
z?7jE;uD<tue}|)^;&}YV{oMC;o!2=Ey8B~BxL-X!>fW<_UlznkVN7S)#J=;+$<|ty
z=_ZF%eq(t6{p+~N*qG@s(=4Mw$(POS-N`=;<T&;QqgOw38@($>Ul4w49j7e&z;<A)
znS-T*>3$X$B>+K#YqM&tSk5R3-FRItGiu{p*}KZ3TM>OcWE8y$pf>zmWm%`R$1zho
zh;M8cr!Qz+HAB$c8yLV6nwvgWd;G!tUAa&vemKnwIqt-!<DR!xcSD7Ra<fL2{?o_b
z^jdnE^>M^e@!&yYuEH0(IRlPo%CF~Z+hJM5FRv`UOe=!3A0{q=%6?QgDE+!Sx;lS#
zDXOV0;y*)-EVJE%0*p<JUxmf2Z?6r%v;CEQ@v)}nCbU@&bGINB&f1!*Xr}HFH&0o~
zV@UxOwK4H_dak=TKxGxsB#%6+DX16Xqt|}@x~px1a9S8@O<*cq+<K+-8k(mr{bCr!
zn|)(d{DL8F;{0C6JvFtVFfQ(enO0hCY;3xlWOQiFnQu<)6SBm&IYp+Hrp@fZ&5ORU
zQCb+=*_W*wG&w_0@bYzAEj|d0Zjp-75z^DOY?_;&2oMnwKg`V7bBu+lR;Fl=kLO!;
zSG|`XUNCxzT76&llPL}kmIdWH5!QI;kHU9!-^Uql=ai-k#mW4Hno<bDbVDJeZ*pQ~
z)jmgm&r$G7Km^%oC`*C*4%+lUe??z_7rUu1<xkN;`7JBk1K`QZt&Hq^)|CVhv%-LX
zo{%szGvg-yMjw7Yu(Lj3zMwa_15-6`jH*K}%7ErM@zypqu)^w+s(mui-QB&*JFs%l
zk@Ne^_c@=&_?3m+pwhi(H-O;|UmtI0y25-DzG{B))BdXr*BkLL^1tPUgvbI<rKE%l
zjEZ9jwma0+Av#(|v2{C~O>$TAH5IO%p`-wz{|yA5At-SHg93#A?Htwm+S>WJI)J2y
zzlpTic;SA`s*sbB`PJH)l-$NFM4O*m4Wo^(Ki)hSf3D1+Cw8-XKcX2}q!Q2DU%q^a
zN4hXWUi=?ZaCB^>iJ3mc{esmez}y}|0p?~Vz#K;N5dsYv_$a0QCFF`**f0v!XMWku
z6d?T)Mjfm@0+gySiGX_d;ubi8-HsRneq?Ja1KN2l*3rAlOUm2$h?H)5btXNl5cVYx
z!~*B6Jb{R@wT&H4ri!|fE|@(4W#9QPa8I-E7a`QFLV+(cQ9ScY(PG+x2RuBiKoVpD
zHy8+gG&DMYwftB97Nlg}ox=lzwq=!z`5yKd!E*9aGPiEuF0I@oC%!5!Z1^$KnM8&0
zke&VC*I4kq@1T^XXHD$XOCTi&FPr~_hsekiAcGFj_J+a`Z4uKSCjZ`#_O?fNWoX6R
zG|5eWxbO)*5_oug+pl|QSYdtpFP-|I#;>}!*Xf?Mh=itntOK|{?X;9#r<!ppSMp!=
z=uT`<Zxwc-YuMczf_aB5xL5c2)dG;}!5G^C#vDpYqW;)Gu`3x^aPQJBw=flmdP#E&
z<huUi0Q+H_@ys-wQ^&Mx7VG~Q$5vA`wlS2yem7%=Jb|k01?vqVH?Ie1g(uG4u}3@Q
z`)6Qb<0o(BBuEq9F=qv{CRh1;Bcp@AI=aRivYRJGc!W7YFoXo*99U`z4<%qHn{!^8
zlK+Tn!Om~y@xRqd)Uv=lkuO+=QTO#CR-788X-<9H!QfW5PzXhUkfLk>x@T}m{^p38
zm~`{CW4TGqYR1Q<w39VD<TqLILmQoMUUk9<i&OoQ-kP082~J>pn$2BdN=elTRus?n
z<d2olH?)loT~rLrRK?hkjS#`eQtGF#lrj@XEL?0v-EHop_|4w7Q`bBzmAVnwO-6Du
z*_~E-gMzY|gYr%3uk0BLyIZr$aw`s*RPnJr9xFZJ=^eh&b(R^tOrMWAZcqL)I@+O%
z&`Nmz^ERI#-p%i=t9{;R_9#saE9pf6Vh&Yif)TdGapS1S*r284FGrvDIqBCgr?-O{
zO!E~8)r<@kHMB<EkMC)I9YYb-Fg^7I_diYPPDjGP_O`~;{i*M_rOll2+D;$1AEWMJ
zV<ltE+}3hBsS2&S>tmUYiRaXsjP)1)XG_K~?&J|P>yK?)TU+WvH!DjxZd(`rF)T{*
ztt%<vkOpU8=d))L8jRdR;7Ba!Vk`Ol`5B0Yz-oQrZ7i@|G&O4@yr=z}U96sH$;v`u
zdwytJ-h;m9PB~F~&#eOerk~^a*M|iE(Q)~|%*3VQcU&GlG)hizxM0_=_lcHtbgpx~
zH!7*ze0^kqI1(NFI{$@E=ugK?tF1Qs#`N8$n=MtiR11gMbXFn|1RVvFWF7z@^^Bc?
zk$HJ`k3mICUTqTK4~UR)1gZy~(;=}RbT3cXLJciVZ8rW`%$*4XU^}_x4N6UEY3YGg
z9JZKRmr37LN;H4EzrAjLuyjSvSa^huyL&Hz`Lt!xi*s|nucsEY2md(=;sck2T;%=2
z8F%-GClnye9wR_tltedREYHZus#~Y0t?42_8G4NWpB}b;4Z$<uo$ZvyAYqW<5}$;L
zbd&TYbD9%+_+>;M_=V!9EM&}4rQ>f-;~{`>F^^j0<$P0prD$Pfwzpevd(c!!Y5v2`
z?oe9I5?o}!RUc9yo@rg|G~DaRi;YVxecx<P2DrriIB|Y384xopnck+pcz19rJvLw+
zqn`b+Pv`3sH%cTUGZQ#jfFg^4Z)B<$CGx2btN!>8=R|?@x1srmeR5}7O&mBvb}&7M
z!Gp4*Dwt=Cm)LmdGGmjHu72*gTei=)-8^gi@~>F-Bh|vmFNymv<W#fD3lha?5(P`f
z;Lt=YW_CkEWMcuyOm8w|h}$j=QzPl6fG~pOJuL8B7f&DBb-aMz#i`RLEO&T_o8t_m
zrcD(SJ!=y{1waL~v=lvSC}(bLtleO5qxb=H?G~c`#9_2<*2C1?bh5;z@$1)(2*1j3
zFK&h>3thO4bIpP<8jFwDRK2Rs%&8LBcu}00d3Z1`?P701_GgJEgG`1PJ9x6=`!eLs
z{Yw{h2W*2prdM>?ax3SSm~Oq3U*3l|$D>b$&`B((?GWRqTGaz{cXqjc>uPDaw$C)9
zO=}--od_n(^#3k1SiXuIdI(&L-6$ai5fhV~<YPXjDZ{+_#D5nyW@Lm&hMeDyY%wlZ
zf4P$*{1iMe=YRxe*sBY}N^u&m=joyMgj|Pe@YFnGip20Kg&-;$7E?@nDaWIK3UZCV
z(68`2?2>^+N%R*;<A4FOfU_0Jnh%;}otRF2rUHXLS8hSUbj$CZZ^v&Ne0FtvuY)*W
zeELi!{`Be0NbByzri-?#>;DIyw$n4q6MMEs@~^Mi?t2{<eicDSK<EvB?N6?W$3j+C
z*AP)CM-uvXB(U8@?5&Px5))96=I`3)p=HXs`O%lVTL%4C)0*REy<3-(12QcC^mV{f
z!f;muIwC3p8%Te`N^9vuUmR-;&pXm2X}3uaAA9^B9IUzScma<5z2`gUYEL2Q+}OlK
z*!3sIKmFD-)a{;xOL9PD_;Btxww)8Wvh(34tNNHL5k#=-L@G8sHVhm}ew#_<`h_i}
zfcWKm<ojZ?g}HG7#yd;F#1)ZgTS6pkhYoQ(Snbs6CN+>Q`8u*0r!|_}vX6vKhn=qs
z-F}6+_)#KvJ4*<|b6U|)(UXUNO?9b$la2f(i6L1x`Qt~LY*{*Jo-X4kr#(KGD&NS}
zqNhmR=nKzEL)UU|%*IGqPCrg0Q;Sw%?S2$FblmN{Q;8bL@qb(Zj!MJieNTCLq)E)h
z)2|o7Ys2l`K8>+8ZJAk^D0|WN4$LBmF2b24!-a@kxN2zd5Gh9*L`(kIMTnlmLiO$~
z$<KK~fefLjp3KITT@y@lIQ$BgX*qhvA=xuwpZ6GuW6%+4ogzjHR&DdzPZbppc8{$P
z^{e;VPMWR9cF_$<ayHV;Q7=tROx&;j%G$}Y#&ok5<qVU=%FD^TJZy9$WqdGIO$GEH
zpgn}t`vG-x%ba|s(YeIpkc;2Dc)5vl?vxqx2C82(DL}dlnD1cl4|fo8Pjb(qH4q{>
zarSGE3FMoorb;)BJ(GcGY38c;&sKr}Dl-Pwd}@OKo?h|75D)Q$*u8(|?A10Ro>DZ4
zyTTNX*I8Au|3vX&r(b@&ncDZjLHv#^@Lfnh4{Fiir-1iUgz$1^1|8LOJ@LVxQNVV;
zC=FhqeMG2fnIGO~lu&*C;Lle1I!Pb(Vhbg}m(E93U}m^5w;9&tvh2p6aYt;{?LVi(
ziqNdKA*7My$5r%F7VQ2o;rWRzBi$$SqPu@B{g&hPQiFB^^n!o@ARSWD>6Wh-o?{>|
z)kpK>O!u9)ssczUBg^2)@9{e4Gmif2tdcK!K;h}=_7$a0Ie#rDSJ7QfDdIFbIwS1(
z=DI(u9TYt|tJ-Dfz;_z^rhWqfy_#YtC_pi{{s9yqV*;!VwaG7ZAV)dGDY+o$n@)q{
z)%-cHg*Kn$edQxc{vUTbw$5Zjl8?Y>IF8?Vsy4>U#ojD!U=_5w;B|l3cXP_+)ARHq
z&H9b$?i6iJUXaOXpNAZ&rx0Yx+`UccXbD39h8f%WBR4oy7Jd{ghi)ecmZ-@}@Bz!g
z{56N+d(hUlVlUb9Wt8NdwXE(#7jQ<O{GBB9VbHx-%F63*OQ;C-u^(6_ZVJc42Z({c
zfsf2aQWGpp`-LtpMqQr0M#nf|)qEV<sed+&Et;U9I%RmjNS<=s<?tsYe*Uus`_gtI
zf(1_+tZA@uoi>&>I>vkmP-~Q9WtKp6gHQ`|o(|NnL(mf+H!jkvQ}tR!>o1?af>B5M
zc_o{_zhdFk&UvKno44=N3Npv1roMcM$3yP=*5gUc58Xz3PjuLr*iPNrhC#@DY@q`j
zUOvYKqIBit)MM&nf@j$eLGgm`vU>t%G{_&=`6=fElSLXJ&g#a$MJ6exru6Ki6t$75
zv8sv+Y7=yRFqj1XVaXmA(myr?%s(NQE37#2y;^(WgVuS_Ap?u$^yG-T?Irael$3&c
zZFx0;jn?a8htao+i?dWBL(RVk9<}sOQ%|!p{Kpia_m;$dInT7cX05Q;-oc?VN6ODr
zZ(v&JUZaRGM@gG!e*V_6gKlWwyw&{BaqpB(pHnh+_^mlTfBoy!`Mtis`=_8Mrwf;I
zxtg%uIzi#`w_Q9l&v>e+saa{?Ow8Dnw3$FT)vz$Xctvx5=>KRL<&U=9%+WMPtW1=2
zHoh*ujlPxsESN1yt63o6`ZXRrdCc-l)>22<ZhtU$rI;4NYu=AapM-jj8*yAU?i^`v
zmAqOnnys6R6aVwiY;pDmO}K*Mh{kToHA7d|%Z?sOzjkJ3=2VZB7`8LM-sOFuh;mu4
zP_yH!(b0i)LQ6#}Pc-vrdlFbb=qJv_<77RoteaG_?IC1&=>;Raf;b4sX7)T(lvL=x
zNFk!iDQN+p@4a}_LjNGJ^)DNbD&CPcSeK`C{D*XOxDE_Y$vlXF&ol>11)+}U$Pv}(
z5cT_%w7zZKLUpNmW{ZnxIlDlF@NYbg5q2t~mfyLPNK^S&%@s%e*rxd%UYb7K@1Iu;
z2Ajjp*(WX1-*b1q{aTpTYnAW=QRpdvjJ8*SU5wSaw8@b?F}HSaP>8Y?6>9v(K{|Nb
z<@;Pdk03(y(+v42|9%+Vv$TjEMKp2*p$||>@oKD%F8$UMGpmi^*@-tF$?%fQ_Hd5J
zd!lm3<}+Pdpk>kEXRIq{;kRu+5UZ`8OuY8xy~%ZiquU89N{U;FGl*jy9pSa!?jRHS
z@wO~Kx)i>O3P}(Z8xu(SOhak9`b9gbWWPAPtiA1V^nAgI%WI0ps9JM}09mT;q5gv@
zXXb~v>0(M*E>52tt`n0IfYp-4Om(Au^Abd3g2PEbs~$vgjdMn{g@V1_^R`>53N3S1
zk975UAM=!O57(lfp@};75yXFd2A*S2--ds!=@$#k^dVK2AVpw_gm;QE5J4D<m`jB=
zzm2%0<|E!BQF%3TM4>6=SDeR0#s}ba%n7-GdRUBh!fva@D=-^xz_{%17bRI)9wcQs
zZusC=*3x^~3g4ni3?CWFes--<&-UUE2;9l>F}kr|&L}j}_2wgxKW~32BRhLm2BtZn
zx3xWyql7F!gKFWZD56PS;SPy3H6w=y%06=Gscr3T^j+-zKN`A&10PnJX=y6G&JxY+
z@<l~{I@?0d()<3ObAnFMU}GUxV}}mEY!@LQy;YbZ`bte%LfQM71f*8g6p9=Xx$LYk
zA_sWbzzlVB@w4CC<d=Pl;)5FAL8r2Z@#5oiynFg4dJDi^{3rl+WmQ@_&y<*u6(iHk
z3}LWf<m4KLKb>MC9%j=BCx;52-;_9_hw}}kVe(47()pAI0*crk7S*LGDTWVZtKowc
zD{Fsft3J<To(9*$tVh}>bVOJZ8XBt+-lCaGki8Lo&t*&KGw0*}*+$dAUzSo&z4E?&
z%ShQtWpg)!ipjBxFCifwB2`c>&k)cc<Uwj)r30Dledi4rQ*!IKu5gS?sY_LwlE|~@
za4<t$EqY}a5gn#DD)Q-g4V;VL8e1$G&7vb~WMmmijPZYWl7u?+2mhelN+iV~!hUD`
z*0)gM(d~EfA3vf-fo)ud(rSd^?mwUR8~);hLnFJpdk9o;keNwByyx!L;FO@Ou9B8s
zu=vI>@L3tsB&y5H%?e7<so|EY+-kn{#eDWeT0uOFO~N_|k_J0Hyp>rD1ROA#^ZCQj
zlvP!q9l@k?)LHS*E&?>(&qRt2n?iN4aF<CnI=&m}>3{xv!>vvz2<OoM#?K~qH-wt)
zes22X9|f#IWK;kDa%1HGJidiPe9zyY>pZbr`I+l|12y$q8fqpE^gtgg*gptpnvye1
z(|B=Mjt<|`bjQcVL-mZ!?n9aer#3cgr-sXLlyAQ?s_hHwgqIZqR&B>k15l*8yE$*o
zR45X&obG$K{%_9mg(fihzm#BUZ(1hrpNODP&Yo$qrpc?HZ;p(VBcj<u{mKk@e#)v!
z-*R%q;f`+b7RU9!fofgszMXaDV{cCv&93}I{OyiXZdqz-@%~b*yi%Vf^<Pa7$@2d+
zk(6(l6v!5_p!nzwar>V2(SfEpls737pVns6KZX5x7TEh-q#;+F36aVmL~@pfgQ~et
zO?p`=kiagpSsVTFf%isfB9qCl+Ve4x>53;TSaX-RRg7);kPb3WeOrY;Q52LUt*SPr
z!I8N38JiJO^fxH3EUubwx_fg;oZ4hVc=KSi{OBtbb2-rooPvrrWkeK}5Q$xSd_R@t
zbgy=Oh>8V4Zr@4IOu~Nt8!vZm1dBec@6HLn=ofl8{yF87a&t>4+uQEj-33|0IOhtV
zmlQQM7#yl7{R=Yy^Ae*HpmQh6K`taWx1I0Jv*dhiA#BB5qOv~&dC|DTUd<1|<(ZW@
zkkppHQk#XRL}YOF@xg-3{RlAOx(7KD<^sowwT`ctqL{euvQmU>ute@ZmC(Q}SBNOM
zZ;HIM4~KNIukmaG4Jj!E2;f(UXx6S*h8TbIa$(!rsFz9oRzJGBrrN{CvVg9<C{x}n
zLwzuw$5(L&CgO;S1^~N&G4}p)W+eCz70p$#c!Zl9Ntw#p`B=CcKk@q1j+>ho#j_@!
z4EYyngiYDqh*r$~x;;z=O!dW?%#F;%DcQKnCB@xg1Eo*`Z9@mfZN@P$wgGpFqj%jY
zLy8HzjsSh=nQKok4;r}A?;FZ%HyW=`pL05Fw<Df@3Ht|~|LCLEazG(QM+`C>r|nK%
z!7^LIf?ej0fh{Ji(6EHiJG*c+2b7aTpc5EIV9Z(PP;if^N^q2^F@^u<12bO4JU?FM
z!5pMlF`Lw@FcKYbA$)c;wO>RT%uGytLsiMhjwq+mz_VqKj>r9K!x*y1&^>ihe*p2d
zeLkH8C*S>i-m+9zVOXd7g>bUjW1nX?5&mFb)S{}Xkb5@5<1WwuyWs1=7(GIH2;R8R
z<hBSYaR5V7iX6ZNI`n*U@aHWopTtgGx%;q)9^7qT+*_DZhse>T#-o&tumGIrHqwD+
zYT!gzc220KY|!~QbfBCdCX6U1j81@3;}=*18OkSDE&o&Dv3(V^@d681TkKd)1Jtgd
zi~apZ<8)FUeb11J(U55j1+6q35vo3v%b&)N(J5mb=DLE<@u|``qlw?UIs1TaOVF9(
zyoe{LRX>F4(M4Z@0J^oT|12?&=DXqVKWdMufaZVFl?I=;U3qy}GOI5|mOSP4fyOfv
z=e9ADLq6NGa$sZC+VrI+%W6KXGnA6XUUpC5-3x8~IrW&MNiF=xAFFD^KHcX>C)=WL
zJd1qe*K%#{H5#^^Bug^+r2XY`Fs!AT4x>UVc?qiq|J(h?{J6-;#W87QzaiP?rKZ;S
z<WGEC%En<5LZMYaHYx&Mi@3TeI8XJvP^}CwghwH7KK1+&`k^VpOGb7ezib^{zii3A
zTT5@SJn~iZ?el5y&QPLeBH%6Nw~0m#$3nCe*X8X&j&6Qm;uiAc=O{a#Z4ga0n&urW
z+b&av*nK0;37Y7ag^N>UCPFJV*|0q6OZ3uH{AvHX7P1MtrtR8tUFx;?U-*9X^O4sJ
zNJ{!8%zL&($IEJHKUwwR6NQN>Mm0jMOm?LNLLh`1cKB~GMedgdy?J&CUm!CR5*Sk4
z*!7e9uZ{e>RKr!6B>bBu*r)qK{C{!Z^ouz4sPAxh`_q9*Pv3ee0MSv#z46!iEMXbu
z0UtK8jOf|!pKkat$D0KUP4ecJsHZ=a@!03V*aTMqq<?CVf1|`U6KP5rLZQxZ`SX*H
z>VYeSV|Z+<W$%}`Ejecl7$MXI7n?QfJ81%r3*yetRgJ9r>ok;MGdRO@rV<yUCUCxi
zaZkXBhlh<*2<SP^J~z{~ZOI_`Jgz-vWF(g~HH`6Ds#J8l?BFU}cO!ba=A)8SBlFHE
z8_7GQ+cC`67H1S#=0EFQDXD@rfh_9tOjVy_C9%rGU5sM0_Un@}fqe=4rNE7L`J~o|
zS!HErGv(&#UzHWGXZt>jdOr7k7A5q1{_03q(rmhzm%Z`Lug^K*%{L7iI+D&5X~OMi
zNuG@F&<ruvoJwa7vY%`<&4qRQ|02Rh=xA=$bA%mYM+Lo?)Yk6lX$7;%#l?JHxb>hL
z?GK}dlK4dq=|jBD!NK9k5zmLVOwkf_gtz>~^*wHR{HdpjLPe7MB#A{&U&<F4x^_)?
zZ+!v_FohH)KU{x7hu!#8ucU?~$Xqf>Y_G7CAO1iS{yWvo>U>SVk8zK+5q^z!oi+3~
zF3Y3AGIthp6mJ{!`melpoW-FmD$b3QOCMN$Y%7nwP0D3DkrCRxF_x{?M>!p#G^pF)
zY_q=QozJ_)@k-@E5IN>GdV}e=h>GZ?B5XsdiPFvfBy(nWZO@GF*O>t_LAYW22{^D{
z+A+s~Jhi>`VYXpIm>WK1TgmZq@+cT^UT@xd!?7|aB*tG|(s)N4RR`cYF+Ne!M_J+?
zY^jV<&r%<)%$Dx%4Gik{#c{5#cb_gXFf!RM>70fF>}jgT!p6uLF?|ANiN59C6W67t
z-Nd_zAFRKZ7H>t4{%H&buP;D5{+$N1b@_RDTR}<Jl{Ace=KY?Zn~+&5p__L_(8~{T
zb!l(!%-;0#wx4;-1x7h{?vR3<grSNNKGT%%pQY#~Xl0*Pi)8rI(ICU4qYfpQY*Kpr
zi_AoHA6Yuup5OfDF()nZn}GnaSe{-EJSC!`&la{c-XPPUyLDC_dcBez^Yx+Ml&>rD
zUA0s{jFfBSu9rH)HA`e3ro^2no|sIPI*0g`_XG$sW@Cv4hY#QJ86k3m{ASl`q&iK*
z&Jq|M-Lqn}QPEn;sR~BPNBJ|gj#u(Ltw!O|35_t;jdhE=|0O%Ah*ES33a|FEyh0yc
zVQa?o*(7^&>pqI{3me9WF5CA4NYmkz0HV-bP_Ij{|8yRolCRo*Xf$$oIWdlfKQSw{
z(0>2O?SV4=69!RD>HVWiNs!jCWYzh+oo$U4r6wYdO0n`BE(>aFdjR)3o?W<btf|Je
zv@~!-TUFrWQ5zekA$^@H^`CW-cgdbL@}`tqMFSLLdPLl`gqSEzKYvBSRbnE6l>+f8
zro3Anz1B8%sGcD4PgMY8vzOgnNB0B+ekz6CFnea5r_Qp3eZ#-gIB^xP?DQTYIz6j<
z1KfF$rqUC-1G6AoeSi)ZVnFJEhdgNWh3J8&&}tA!g9zSGU5WttcS^7bhZm%V<QeIb
zAfH?HA`SDJYZPJz){hT@zC{z`+2ScpvgSneUAznkkiuBl9dhJ~1A$jBSfgAFS0ByD
zCqI#XG4N~nF0v)$ii5P$XI}aRh1`7{(7p_AR|7;uChKcqX8rlRmeSo_j6_|HyUj%n
zsc&b0yjy0|Dc{z6WM;Jx!rF)?sv$WcR49k{59;>{!d@vfoo)4f_CR_r`9GjDynX*(
z24W{(e?jkcwSd?))BsI!<*AvzztT@%!KC8A{9{JQiR;x+)eXeTygSzS=dN5*wSt2B
z!s^yX!g^42(d}QH7(;pwP59@SqUk60Yu_$NmWO%RqcS3GA)#u$*9~yGt0S?$8qc_p
z6zNye5dgIm*RuHobX!x?=;lHb3LnNcM7cP+)R4qPHgvcLPhk#3`Fv^bTpAc3Xo!$A
z+{Nq!bFiattgH(H-y@@=&^)8F9w<wkYRge_^?&|EgNWiu8uvX4x0%`<TBf}jEPgt#
zSW{z&*{N3?EA!wE@tth}TNYtPQe>vC`_y#oDzXfRiv4(nvzeqX0TVH}<FQ|>Rs6nr
zkGpY<*R|kwbK{TWN5nZf3{*46BylngF59s0Z4j?)uP$-v>)Xl-w1aPlJoO)uFg0j<
zL6$&G;OctoOQ(LZ6x|gA!`88gaCkV%i>+-D*E%;sixv!v<jqtChG(hY#qMR{{eaN}
z%Z<ar`X5p^xoX|(U{>+6g>cs}{$Mv2jMhB<Xpc8AASZ8{&@SpMr_O9`kAo+5i4TL@
z=U>nbK6E+Gq;8I0+0Hu$kER-a-LJ#ZG7qqQeul;XSu^sb$hT(|(k#)g9NtiTiPK=-
z_!t_hTYrUrLS(&kVDrl3i-p$}RL;v2^uB6J`i@RjtgkD4hmN9jRg})S(bzn;0Q~n7
zJ!!hPVI!9Th+UwLZbvl(znQv=)>jIwz?Z58lLxCuxBZj9*7n`MRuxTEn3zqEk533q
zdU9Ln{YiRzv98B)(@7B=jCm$&k0&YqT<h-g4Kaj^#`3@2s0r45-*-LoV}^%zpB@np
zB@J=aS(s(36Pg0Xx5G|{|L`&?s#ibcOaJ~O#n-;Rx+ea~4-SfV3n#g7lL6W1hM@ae
zi$^Sul<*8cB^oPDZ(YNR)$ewHa}R-S;al<>7X9URkA?Sp)qAmBMYM}<j~iBIZ^{-p
zCTZ_a=#B|wJ+WU1#yoevpkMyxkN3_#Yg^!Wsbsl0<;xNm8hyQ5xn-*oC-TcW&iZV`
z=Q2)mcDNybnO9%6(e|wX-g%t>4$>0cqGpvRqY_j@T468loh5uc@Fl^?A>oGX&{j`(
zjx!1p^%eSg2V;gPDZ4T2ang%rt|FgfEJ{|lng>}7JdLT}tE$VgKUg_57w*+gQ5#Ji
zUNl8H@YbxIzp~ETA`vW6EVmuAM)5qnVs`$z$$K;!88L2{toHtc_>@ttVl<15Elp^<
zfa3ujN@ytMqqK;91AQYPr4Lp*03Yl4$X38*t_#CSQ^Z#;JD5tyY^<UAc2-Yuc?Cex
zh^KKxTj!XBvt_*TIkj=A07&gYUkxE#FNT)y#6&w)@wE#&+5bIo;(ql?`(8oc9V0{F
zvGzSG$?ApZ6SY5^b4Lpr7Y$<VcSwWSDn@V3y@jnN@w}Uh%m32`SB{RRX?F@Tdmfq;
zbeFev@Fxo=;~l^D>L&uIq6%t_Lg%4NizFd;c_o<4anr->|DI!qw!1i02?xptBynl#
z%2%cqf9Tn31!bj>i*2Lxyquo(MAeH&9~@ne@_8p-$A@1!ceD5W=;h&No&PfsA)idI
z@=6Vi&gz^J_Sd^+-L;=cKJ_w5DKdO#;N@P79}*N6=jSVl>q<!Ov2?LmxovH_z6PR;
zjoRNgPMlGiy6h6H_xDf`AzxPYk^q?p<F7S)68ZQ_bRC_akN%*2s652=vYUBWMFaE#
ziKkumAFRP{jM2EST|Q3seuM_gQeGzZVltP(&sdFpkIQ%Ky@tE~uSU2nXBX|U54Jf*
ztp`sA_T4oq%-d(6`{&)Z2JE9sQEWigTIpa#R?}W4VL#A2(R^ZG28n&_PUrKCu_#CD
zWSz+f+o`*=hw6?HE|S(B1j<64nl)Z<=e0nA5uHxD_Rxb6;5KOOpC=<=DHAk$XM)?X
zO&^31CmE3&YL?)9oR-<?ez?R6C=~8Z&*6ih0)eAv2~Jxt%mn#OAs80a#80&^w?te8
zT{rYWsSE!9TycUQrlgP+8S6}WwxCG|DzlS!=L7SGm^TCiJE6pxu5xOhii=ou)fIn!
zpi}vw8(`q1HhL8m6-Fj5@XZp4{Ts^m%4oTX%l+3**|G*)0Yt*AOl$>3mXJnOvaRQR
z1R>)6tB;8y1_W5d?VS&Phm&fsL(ljnDSV{=cjV<-aCYr_N9x{I%?(r(+&^iV`LAAS
zFqBuQUsSe#DkWhPai&PPT*F0D64d>gS2kCp!w>f=gEiT88(SCmZsLDl86A<=l#Y<0
z85tSDIB_;MH6|k^1Ij6NF%tkb(9yNa&-;ePeaq~Wlap0TQECq*orz<l4gc)1IpML}
zWw&rR$}32O{3ZMXPYTBzMemqO`02LsXn~gASnEwN+E&i)Ii-L2aCUNtm51%aa)X#n
zmONx0L1Idc)ACiU|0A%sgr6d{J<FN%MuEh<-m%H~7iJe7yUM?0g60X8JM_QA)M6w3
zL_`3ldYnzs1F(%1_x0-;WEc7P*jmnZ)Z**GB5C=ME?WJ_?W~X>C@AQ8$VOyeAGoWS
ztn;5wti5xs6a46nS~ycxYB3)z8T!siutA-Ljg67HC`Yi$5Z*v<C=2Y(C!e?<SaH9K
zmQgVq-uaOa_r`~doQ}^kkae9d8$1p>>}Ejgqing)WE8pi>4{;H0pB)&0+1kJYBHR1
zrRaS&TYXI6L>e1^_uhD8Gw_<E==gs-kP?mankZcV{W1t;6;##r0$1Yg%8W#|31SDU
zcWUfg|1#cggOI<N)D9`x?lA4b8>n+@r}J1?Ds<#*RCn$yf)lTN!eh1!O*P8U<*hZ>
zgBH@rIZJqk-6TEhL;39~9Sz*&c21}01@-%TC+{x1o`$>m;gbTCS*r6sojf4de_(Af
z5C%|>^j4^Cl;<-R9v_5J{8{I=9F1I#(-5dMi{-O>HoKGbP%b&fj^c0&X5A(I4d25o
zZnp(L=6Vc+`!=`-M@ELj>@=h!+|E{XUJ)}h^jq$Dy2Fdst^YtS)E;t=+jeA7H|^%3
z!-tYax9!w#i5^!CYd}Tkn*RxGc3^Nr<U82Zw|t<DVfi|gtriOJ?43KWd?Z?}pBw-9
zy2>VZ)G>^Os4y)Khu2%qEJdoR<mHyAUevtl`ZVB8a}IT+aG$FeKNl!^cGLY<jZ7t;
zf0D#q1N$f@flHgJ>L-Por|Yq|6x03iL9G2!woeqzR6nQj{b^cpx~~N8hM}cl!q+dp
zQ>Q$Z>IFF+cLlgiK-r+^eGZ}6#CIymgxf+XEA%=)K1HMP#re#@!t|VQ1+n_uzcoOc
z2Cu8g(#Y|gO-;4DAb2EeRuwOhj6={QB|7W*dsUuz0ojVSJcC9V3K457{ZHv`F=fkf
ziT>KJ8yvTMy2mC1BTC2DF3%1i#X}6@af~&bj5VrVTTiqnc>sj2NG8%G&o4+AO{#T_
zf|%wc;SU3KanN4T{A`8-J8<%s5cHJp=WM)OX~$WI%#yfN5KA3IY`*>dkBEov2zy<2
zln>u4ufeIQj|;m8w&i05T7*LH#fO{Rq27tpVCVaa=gezA+;0%Le29llDwtvkLPi-y
z%k{Q$RIjF<R^#)fLwMKb_D@zSu3iM;;D+4$y}G*9rDX8@`N8v;rq=E;37T7R;_<vL
zYZn^Iw9bCMH!&6$7S&W=9dA#XuB2S8L?s5f30wiMg_Vu*aDP{;(z>(h%FP!yL`6kE
zBcr?__1+^ICXOaZp4*(Qc7haYZL<Q4X*X&Cp@&6pAX%z4(_qhsj<eyB3X`2tEnO}Z
zEn#4?Vg1!;*Iy-xo7&RKm+RPPG&6NQPd|G6^CkYhID1fVTbfRbt#6|1ulGU`VQv{%
zXhWQG^e=Va>Nx)h??fpU3*$NU_yoBSN^iktuqI9Y0l~S=J){zKV|W*yHvh*37~zG3
z+PSJa<roV9(f1G9z#T-C8uAxvYXQrSUf)_XGt+Nc+yh9vHer8CLs`Eu4=7~wLqtT7
z3J8+Ul-q#YzSc?1&r-%uw_sR#eiP4c{T<o+ACFB59`A|QA#Srn_4qVJ#E8)?Yju^Z
z#)eSz?r)bGf}EkB9-gi^R+;djAo4Bg1PsD8NpfZ`h81XmT|l6$Y4g`Pr)51;u`!M8
z2)O@z8s;U5;`<|qC1c;wk&w^nQmO|-`z5bRM^kb!*3Nmc$X4kld_7;mC3s|;<;S{3
zHFW*c9{<j3#FEGjP%g-8ieDqD&s`^@5BT!+>isLje1%9q_2>NGHN}or^lMa@%Jt(k
z^Wu4T$Vv5mlhQ&;v(odkBEn0&;O|j+)wLDa35A?ZY4}l61xuqVOeTE}3Dtv!oN=Nz
zJ#V_FC#6}0c{LWM;USNA$Z?_$ym2nd@5WP36o!!5Kg<0duh~QW==p^BCFu#dvKJ=O
z#^gKqAk~o~HAKy@_p$a?y;b45?Rdls6X&HKf#dT7yv;g%S}&<Lc{!hsaTJveiq%1K
zw!OS*tY`Kgf_sPy>GoR|az?jCc^|9CCOFm9f}1TITBiRZ+2O<5Fe`X3m-uIW@qQzO
zjpu)ZV39Q(r4*%HaAzT6^6`8KiRyRBRs?=K08^HT%RuQMC+F#9s~!YbRTNkB_bbd>
zJMpLzuSnEOu`KPQSMWrG5sOJ4i?q$722t2sL5)2ulE&-ToT<ccWM2W_f$Wdl?`R+0
zo}W`h11d)0bX9tr-XSi2^PH8i&eqvk4J!p+GP_k-rFQuN$-=j%_aofu1i|Ql(4T^I
zfH^0hD(cy$p6PU4;Lph+*{4rYUhKz`k=E4mgF!V{+*Cn6RYdM!|8J>O1!q6>zBu7r
zv1RUaHXd=pTcQ6Vo<UGOJ2agM7~0e9)Nl;M&h8a|u!X@};S}8`Sg_RTC0bJo-LbH+
zz)C|Vt1G2LI3i6l{p2&u2jfv?#k8jfr+#())Phbkwa~&S#So@cIYoy0IiT(ZD}UnN
zxKtQ{r59{&7-eS+KRuwRFAxH{UT)gog*Sfs2Js0Ac4g(@C=$+PtE^m_R9Ki081dyx
zTJ(d-zR3ttY6#-WD^Z`9#3vVW{jZSrSO@GSApDF1Z+3XfOi%yG+l`K-)B^cfD6{ai
z%_L|PrLo&523=vQDaEM4KolJrZJLu0?xfJf!|2T@Cm;SfIsTfNkuf?-%>=%+X7uk{
zad#`X*7Rhv2bh@i!8RA-Ex@=8LJj-bBwaLNra?gj&DRqxw$S#WfdR@#djGwaW@Y^f
z87a7Sc4$J{%-A+CqP2|%QnvTg<6Y~qp&{crs}{ZD(CqVzapeI9e+imY1yBI_x48Mx
zgsZcx>b)}<+e*6N+$xkqZ2<a=hDJPVJa2PhV7Vx@3K)jid@wOKR#a2Au`iK)F`l6L
z#0(dg#1O-d`xQ8zX`8b)h~{+Zu_cK}d%fiB9|NaE$@_xI;w;ftRf6!0-Grurn(fEN
z15#O7@qngRj7FQ}<w0699bs{5u_E|+I}Uro`P5Hx8zI2FVPIeY220RUz}Z@&83C>>
zo<+QIGI>Ro3W}<foVNWLyXpLobD!|rLUV4*FO=cZl=Ie`kD?HiX<!P3P&{!(ML?|r
zNrx?O`Y{f$seU)g`*mX?;C}I&Y)zlgA3T_iOsi}8^GjkK1$Ahm?yjv%9$<o9E;Pt8
zccKcfgh9ycC^$1VDW)b_Tb(Jv6m@dy*pn>OQ>MLCEB|wG51<fHaKzs{**~15AdcWO
z`TYl=Q<%w4O-=RpXTa)%Zp7fNg}#yJXi$EHHzX8-Nt7sqN{Z5QU(!USgF3dKSf&DG
z=&7@0#rWl?Hh_TOj(-bEgASxJ4-Ji=fLrAD@_c4?`t;Nw7=1#TkQ)0X`sxo@E%t}|
zZS%W_P;~mYj82#=m0^Z>$nWZ?=~^1UWjin4-42ia%B1X%hleE^9D`qII#!5JDY&xp
z1i8!Sp67~geX6bzzdgBYsI7eoRnohpw)@juEphM8VM3AA(x%&BZ)$yD^p4h>4?Cvq
zVHQrnhJf4N$vW?L62C*Sd{IAqEZwTt?%Pj;{EO$jv^AOX^R9o_42^fb?eP2zJgk(;
zML9fsle}-7wCM-7c4v4LUgF@fLnNe?c<EYSbLOO5?JmG7gobdH=%g)s5XIjj9XLP9
zwGlX?5SQKCvzq>He;T-<c8*7hkBzxTN%Z}C=Y#DF1o0AebUhAWc~|F3+>_AxKELnM
zfE3S)Iht)}|7c_8(H9Un*sk;2tbN*5eh+Zc#Du!}bk+M`Eu^RD68c=HZLJZHmRWn9
zPXnH2UvpWufzd#zj(&txOKCg4R-=>Q_1tZJF5|-zi6<1lG^%ib(T$6@IURjSbAFL&
z3Z8H(=uBif3oEG^9aXMJD@||Tg`4ogC>$qF$Zb%!_9z+ViBMNw^cSKBs)1c9Sl6I4
z9c~@k3@E>E?CC^9;?hx+lmq|6UmqQ@yoek(qKy-;feTe=pNGx<1B*MGU7kzGyFh_N
zQ@6Zw%Nu}yZAFd6rN347tzS~#J?QK-HL^6UHJ%O-M)RZ99qRv+qSy@XTbw#fQ5Mrx
z6+X=xb*=>}Oo>8nX%LYc8HfEZlRo4#*~%M30fzx=M3j_sQRarG*kwzqcH?#RC3JEu
z5BB?J)j)`9Hr{{{+MX<2bNuwzkhTl{E&opt{TkRxctEVTeIznGG6FyVM9T3SU*z<M
zrR;Rf0SRKj<#!m$?X2FX^^*saU6>&Cay=xQXz8=JxV)SQu9Z5$`T&JzbGtk^vw6zU
z^u+$lDfe>aKDM|{hak+>30%OLVw@huGBMeBXgu(cFg7mfOMF5=%ANJ?qP|&&gghU2
zJ-Sb~?B^&DqX=ePcBC4lu4-OM&ouN1SuK1rlWr6i0?LVe5~s(-%3JMAw+AwblfzR}
zv`?&Neleo?1&du-Xz~^q^sidZEK&%4s7Os!#2Sny4vBtnK0HuwL&65Bi<RZY6Aez6
zr-5n%`B8`YF%8|F(=I!AL1xbF%28w5ZTt!<RR4L9^NJMI)U*vWbVmo9w#j`x<#|Yg
z5VOY`n3$<Rbe79?ctF5Q)Ns$pNny;_r{TBi0maMeJwrDa4|G~<Dc&mV1);0C@f@%x
zINgW8do{Q5@lEu^aaPa%Q7j?MQR%8*U-k<MiheoQaka@jX6AUX5EoD~_(NzChz<6a
zOWMePtD|<mI#PRBd!Wni=yexMXB@uJF5DjTDQ(s3oWM;GsFQx<$_po5k?xcG%Lrw%
z$HH)w<<@0JZt7^IVL!xSjF4A)O#aRb_y=AvRJSqW1YPsj!u8$PE>_cBuIm-fu5UnY
z=TvJxs3e<x#o@ezh6+AE(PQds2;;b3$$I<djVoFmnhR=EV!q+x-2`{)@^ZU+T0byR
z^s8jMTB*2TF`KO39v&YZ`2F2JF2+yOFWXf_C>{`sip?p>bVg_mR`=p){f@5|=NCil
zQ%R!1V>3Wnir;>Z1^Hx)eZw<Js0OWRJWvfN;&NZBf#s<fQorSOB~G*3vWooyvvbm6
zbM|6v>!d|rMFpm-;LRcYdDqm;tj6Ygp!Ta4(%~HEwtST?a7>l@)f_+^$mV@E(75>&
zvmS7M&`;F>kzS;7v)BDlwL*=B_;`cFE+t)N`#)Zi_U<tYOqhHtIG$~|;10l`2L6Kv
z-{%|2_$n!T|EIa#CxR%J?@gYiDu67Pm|ZvFkcJ$g*f2AG?9<#XCrw948QN$)-3BO6
zS{j_u;i8UbEGT^49UHnzxhm8IWh8eZ%pvEKVJZJx-Cqq?r8k~b6Cux}pel!AoeGnN
zI1nB62ZqMbjc0Sfs?*<covVMQtKaynmv*}5EUDfK{YBY>2iefJxODW`9|_Xt=Tv1#
z9ye8_Im7fomMJQaftgQW975+r!F@1u5T293?{OOK4-kj}rv)LViHV=rvxXT;unPhq
zBm!2mHL%UoSki)T5-#o`{QQedrBnsmzX#~}yoUNFw*p%s`0J_70mJ~lBF2uZuf;(B
z3R~T1ZEE_%MH1qQ5a==p{u4f0pd-X=^F71i*QO@2II$7n>i<4YHmXWI>9CD~_={UF
zoEY<xNXEUKTIS$=1a`s}mMJKRgv3p6IjTooFJFX1Y9+L(7B1D6bk_S=IJgJMn;}bP
zNYA<PZ^}~bgpTyH_Fo@VcS~8GA5Ht=2FsSDSm60dXAkjO%qS|o9-SKeh+`rbVoy(R
z@PcvGh+f1aX{2EcaxU^%vczfltWG|;qT%M9?Uid9p=1m7xrSf9wvYK^RrUJy26y9i
zT#Rnhob}>eqZIQ+FrV|UvUBPnwt3^Z4KGu%yGW}R9f9f@o`xB&9kAOfMXxp0<TN$n
z$;no*|3%7L!&rgW(uPvV=_Ue%4H9-;$*rNB7XWDr98<iM?f$dp$W67V?0Cu3{_EFW
z4h{rB^6xpkI6)p!KpR@9yiO<I^M({Q*r>kpc{mLoI04~q%H<D(&$Fj5Y#`PC%HhPF
zDeyYSZdMU}&*MwcV0v+R@#~T|#AmK&Q<<T5I>LV-I<ay~cn7qg4t9HoN9v8vtMP{(
zU379}WaO!=^~By!(U1cP7o$Fx52TN7(<1#NBVH9aAK`w7LHVq!maSd6K`?wY!VbI5
zO#I2TFJxt3J?I_QiIfr(>lqopUuPIP8b^Igu?{D+UANAxXJKXFQC3i&^XVGw7Yrqe
z($b~H#V@ju!Qa0>5z?cUy$0rT@VYAopsxmGP&49E(LLD^gjo@pAoVjz&C<HM5?H#M
zwMUhQa&8O#Kg4bxh<SYJpyR1U6S;<9D_n94CkT5+a}xr<=QL1NR8y`+X!2TA707QO
z<0vb8yeQ!BN(juruHMh0;}eKzM8S_mYXi<F04328zL3z!1v@vE@D9|Leyh~;V1PN!
z(-!<b+>b7}aywht<z+i`+2>ha3l<o&n1O|EX2*>@sqw(T`UVzM&r-vBRQuS&NbaR~
zym2*4bxb<dTUmKg@;zR6srRnP`;+#XV=kbd)lX@^Qi&WVBZILmSnQlR*MOD+7K?Aq
zHwPeRRTSz(2PtC>R(;CA|MeEiUc1BXbn5j9);*Dmt$0rQv%jm&AUgr%k>6zwOU>d&
z^aD-WrOrp=$*`-gt>u+OX9)F0TE?&9nBg9x+hoHin)LF;gNX9S<DI#EPJ&4xA*u>$
zvwob#SHejk;Xgg~0eMG60W?ka^M7C1!c2#O)$&jhNLpHK+crw7NPquty6z{Rcdsb~
z@zgYDQ;xQohu*!pXf$XWHZo@|C}Lms@I<%CntZL<+C~pjJ)h&X0p`bd(I3Oce}NBC
z!I(&koMGT~Z&(K%MBbljeC6O@$Txco*d3QpZ$O^ODaQJ$3Hi!;rbu7Shw~ZO#$4;B
z#)@<%MrO~^NMC212C10wX@z`5<<{$$(&%zHVCbITLqww+WhFJGsR>ldv6If672Me)
z{R#Y<S65AVd2ImMyn3n#bczQa!F7)aDDaLu{2%#+B%Pf^0>Ta~LLh|J=I>9%DMu2K
zsuj*IeQ-*azVk_^&pjM5agnf~@j3~$@ggC<lP%JRvq~I$$dw@4)Xl{PDFu&~**(Hf
zH-4nID?bRTD6hcT*>7;$zP~pw`j>^aP44vUbP+CAr(0(AWWj!NfVz>F-t}el$L__5
z#+}iK%e9`2)qQoDto+jEm0M!AW!w%Me4m|p->j&|(dRq@N5#-|b<NjLHkAqM91qD_
zzqyChU7$ke#Ew4(sC}F7b4z*N3Sa)g5z{#<sZFKO@XsyLJZRA>G?$KX{-1sxb?06(
zu`;qIG&-Vt%Q@JtaM@aAL`9vlHx@<@<wZv~_MrQV|1GOMLjPzPR@?wJVDWoznccNZ
zPa>jKXeVRi<YM^^VI$?Eu#j6fvI1fF`(N4@x&C2YjVu4%xVWA*6Ta&qNfQ+rrS_JG
z47MYD^893)%s+idxai7IzGSkFkh`rI2M0%k(7Gzy>40pQAC32JMCFdi54BeQ@-izO
z{l@3lc>>O#D1SHh0Y>T($14`qnlWU|Afja;(2W;Bz4MRh2?~Bhv?P5R?J^-g!Ps`<
zfN?&2j)Rf=9s>9v^3R=ZVQ$*vo9?N3J32S=YHoGQdT%GmsWiLe2nJQ$-6VjMugfTI
zTU+1TG#@b3_;VA<5a~PEKZ+DGJI2`zX(@a>PR8J*34ya<?n5Q${86brQ@BsXpjaP;
z1?|koQmYcx?F!>BOio|(PZSg-fZMJ9T18F{dx@J)jeV(RibrMP)tKdUgiuMAP+yqO
zCqy(^Cv`7E1JzT3p~mXeSv)yO=%aUUr2z}L%>$w1==kutM@j6jhI%oBY^e+-uG~`l
zHki2p-JB~XOk?OlZ;~)3!lkPZjwKp5C1DQnb;|YhiVVip`d>a_sjTgK3m9a35K2h2
zdAiR*oQ#^o`AqYEOkpZtPY<(^nc3d4={*Mf$>Hwostgrv+4A~&e2AaNyL@-al#wIZ
zG%qx*J;6-uA@f$TIPFVXQDytk?-N7J7lpkc;KRi*%6RD%T@?C|Uk22N@b)@AH$54A
z=%X^bv=EmNmlCTE#N1MEp$WOBc|V*Dzh_6#`cd&a5B}~~a332?GK8T`$IYIBRB7jg
zpySFxD_QA2xHY27twbBG;y<DyFkrF8KaiL0&dDph74S(_b&0PUP(a|va+j+*P&%j?
zno0*&{Ny(RPf`01FUB#*(a>DMZ1fXT>&Z1(!!fe3ou665!up@bRpqIBb>;Dfh{<~s
z!avKbE~4PW#pnF@+lD7F3?*^ywTvoAl5ry6U1r}(p0fd{FMbr5w??)Q^I`j`%(1A4
zyh99jHXb80;~x4)*jiCNXWsr8$`(gMln$QQ>4%ClO?&M8&b5gV9?MIJkq6{}#_K=C
z`=+?~!K#L)rijr=)zC8Ask6($#=vyv4>@?hvnuSzW}3I<o99q3>bixCJSUT+$1Hl2
zcy9T%M|>8~Kt49lBXH<SL*oH9!_o0cs|)=;t$J<3(g=bVBj9{OfUmY3Q#4~_Typ6u
zNW!9liHwYJLthKhnwa&<Ev~ykq8`a$3tu+68`nOz8qv<o$OPvf63g|L#jn}zaOqya
zBv6k*Qu7)J+Tc_QlmzE_9)1a$a5xpTO04W(zUKGttvdfR0NfNep~PE?-v5}>YLc3!
zogX2PZ)<Qwt!Nfi^w+#$X|^mD&W;4*F@axb@_6NCs9k<^o#W|5Piy$g%&Ge8ocZD!
z=l#FA&FrPRsqODte0wHiSmU!Za|foT2ANsZqhDeDTg;)Pq~&^@7sPmePqW#z%XQ@b
z*$v|BC0R_D9c_8vbraGiXU3+(nHu~#UcbHb7rhiDw5i`RhD#+K9{<@nRqwj=3)!B&
z!uT@1xwf4cqkAF<e1HQiT>I~(=!iLM&-XBDxIn5@nfs6`TMWryK|VAPde(UXB}!7q
zO|l6NDFZWA2JOJa!#O;3bagN4{}TxMMaeq17*L`?EZD?4gHW(E4(@8jc~S6cU+FJM
zK*Kr6?$f5rCP-CKQBrzuU;z_EC7SS+Rg<w1Uh3J*>2v(;iJA|<UxlWp{#h*`AlSJ_
zKMl<YSNk4ycQeD;MMfU8vb7eetNt%6Nt7iV%+WWe_COgP%1HFmNJ_!a{Z+eM;$)-E
zegOTxZe!2t=m<Nj&^{FSlVsOCJn{Ej@W8?8?e0gg>}`Fj?Z(WM7xt6`oMnDig$KqI
zpm=5%V?yZGQyX9&Q<5iF88oj!k7r_10_8>YH}$bj7*cDzZ@Az%jZAIO@k~BG6RCLG
zep<_MmCaD}E&64q`X`>7|NHQzXwcPp2u63i0JMQ|1Wf*Ti>PV$(Xop8+-#EzLsZ`w
zHHvyrF-4tmEcwfre-fv0cB(Qi;a(#rqiPd0r_%WI#?bVw#l+MgVRBf|Fuz?HcaMxP
zPE}72I}DUi5GcrK(dyMKXQC=g3r^0R-F95!*a_0ahi?eZO!sK$s%2*tvI*Zu7iy>o
zA;VkVp4U)OnwW{C#sH)j&%T1*U@<gVaDUG#I9Oi%hZb;<CWjIWQgt+5-P>P?lTrL4
z^T6S7|9u`qrPULl3({s%ehN=*C{6d4q6?Cj3Ag(sL*wLB)kdF1-KtBPw8a+L=%T4I
zHvGQcbwq&W2^WoP*Dv&G)JqWj)cb`_)&_@3wQPBz5MfnU6bgf+Ah%jW<Hd_5EIQi!
z#Qn&E;5&)W&#4c&<7AEwj}(**qN1Zj+@kwC4L5H4*>8&2Y7sy3QBYGp+_ip4@EHgZ
z`bVgr;{1yXf<XY=)x#_r7q#%t^qqmZaqs*trErM(tm!*iG?IrH!n!+jJUYa~ppnST
ztGQ`>lbP#oZ-_mo{Jess!nTUWuadn0$U$3igoH%ais$ER(40+3H_g}APeE<xD%^NE
zg@Pm$%x%sEw9hLKCM_?L*uRei<m&}Zxazm*es@|g(f@FE-g3r4gS+eBxha+`?C(ZG
zo~shDH$JJZ0Ss@AEvK7_TieSL*#kOtt^j0R=1iz-fR(|5#SkTbA~D5qlw9_-$4Ttd
zY&1N59YHoZJZQC0y*OysB@7Emuzdb*&v)wQEi5c$X9&fxf<>elR@`75V{~eIL1t><
zgLRej(X(@B&Xt3yI@GxMAm5t!^2}4%iM$j#^td+73J4>H^qu+#rXjvS@Us&WxjHe}
zHya;)_)YXvRCgO-4sXBEYtiZPe$hS!U3vT4O((q<IdyUn8B0KO{wasSL&TqkrRetm
zuIskYMXCk6vqGNuJ}<M!S3lhNWqQ1KWmoPI3x4;QoqPB2dd1gO3N{27fq-#tK*5bu
zn2p?<sdVf3yfRp_M>S0xBwy<IgM}-uPNvY3r0sZFa>D5I4JwC+()-)lQ~sb<``?E<
zi-_IUCf=!PG0xq@`DHS5?R%Rb9Fz~CtOtSIFVUg<fGM`iebspE@c`xew)W4vT!K?y
zr?nF+Hh-U|lYiX34I*pK7i2j9_c#Rvgp3bx#Yic-H<$_y<e`)z(z1Og0|h7KU|l7x
zR>024dY!hRa-BTxaq<ZWv|ds~?`V(q!_^cXn_L`tL}TL3d&CX~<WWuE_`iz=CC4ZK
zeC~Z9%+Wv&-N)vlo%_uey$=sF)j0^jpP(^iMY0lgen`buPJv?nn*5p9_iCT!l%fBr
z=l<uFx+`6cq0$qP;t)spU(H=(zoiEDSgZaQ{zTEUz?325gX@WAPa5ykjyD%8s&C-o
z>ayeL%D42=+v|THyG{-NkE*|bs%rb*$Kj1ADvE)Gf`D{`w6r21EseC&-JJ#@B2v=b
z-5}D6bc1wvcQ^lu_xt;d_hmTlP%n7)*=Oyw=6vQ8aoC?ENFpNez_6Co`KoU-D+^mH
zHI>VVN+?65^WG`b?(2~HQ?sg<=^2G!>s_;m1tm}4P9xB9+`5&Ck`fb5;`0uEHvog6
zE&TCW%=IZ+@PZot$oaRr`Z2uMuW=X&BJ%Pdmt3?jui)~_C3HhSsk^6#r&xa^fp?_*
z_y~EP*R}4wZ&LJ><LpO?$#9|e;b3!NWz9qcfJyeZc25%JR3cRYU_D9vS{p~!HIZij
zX0vhR;z5Vt+EDe$F91W`Y-lfU$~m@JOa8NG|GUYZ+o-z+PF{93)))6<H#tT=#OB()
zkK>3DZf#}!>UjLMaprQE@Qf1+JFKbK6xIOS5MJl>0|@MHA`1#ji}H-1_HHe6J$R?9
zz{ZtYnTUzNWPviv&e~E|MiCGz8E?Pd?tW$#CNT-AmA~BIU!>LZ_(5NKeXVw83s6J5
zAN^_!0L#ingiA@k&&w-)C26V-r704y&y$p&E+~bqgGtSrjDM`AIRSlZw?6bjej!4j
zl!kADfkr_mCU@~9%rDT;4!;PJ$lLa}u0kbREKY^_2yfB$?yfnY4x`mdniKDT{53#_
zowI~(C}j@Yew}xA>u@^@p086@hVH)R-*t71dr|gxwPO=`)J%SO<&@09`8TEpLyU?W
z*J=UhO)4&-X%QRybOpAz(aMk4nO?)h9j3?xIMMq&Md?3ny8xR!bFEzGa{&4z4BCoI
zijnx@?Ck8yMAtPqs$j$j`Ko+O<U2y_!;7m1owzp&HFN+wm$-v*M_>+le<cL3KPl?V
z7c0mvX&*W8sQvaph?9F5rfA}1K_JD_;bQy;Zs`J2lg@o!Jq~C?ayOO>&)6+rarkei
z+D{W>Q^R+9PEZ=9rp^=o{t%%&l<@lnlrZ-;JrMT|ICkT8pOf7GBT#N^cxhDw8>C<=
zg|V4lVq*69@Ms(;NdgnxAd{e;mgDyB9$lNnFr&QqEfzL~{=fP+3GA+-qR!}h{h8Z#
zaG?p=hRxuAPYTO+RSX@N5JeNL4Vhr@th@G&eL8(jFoe&RU*lUe(<5YkurZpQ-ZZ&$
z1kQUJu(xn|iIBle)O<O)puL@*^_nNB&NK^jsxy3K^11?w8V|rj&^C4dGvB)hBVU>x
z|35B3qyKVSAdK7oQuu3DSQQA8ykl0dGaZZ1OY^m(*LZ4NJ9s}uS62{oYce(m2Eqmv
z6sj+UKszWZ1M#}Am7;`Kp5H+8ZcHN8myGJ-lW1=0dOP2H6DaS$QMwF^2~_)|A|Ez0
zioH~q45^ou*NMOKXl+*w$m^7TMCz(a8!KxCmSgBwJ|M`KFJC`Zl$LXH<KTom8?xv8
z=^O=Oz?v_H!Hsz;<z0hK@#W=u4c`}G$L_YfMx#5FSayGkccCpjvFR;T&0uGOfQUNw
zS@B5M^Su1@<X0A5;LHu$CwZE^z5T~*Ci&Te(0^SQ+86S|R!|6qM{^^VxT$B?t-tmU
zdyZ;%&#0`@M845d?R^>4?tM@ueK_YcZIADW<5f@=y@^m^wM@CXgCVKBqM~wS6yZ>5
z(w}iKqJG<7>;ofa7e4p1WnJR2^2D{5*Z0i%Yd~jp4G}`q?)cjLiS$H8HR1xIBumqx
zYVroUUYw!4wc1^T#*aCoB`cKSF(KQprswYKZPW<REbPsO0Kd#DuB;vJu!DGkLP}O%
z^_ZpLBz*}VLY}7|#ZnB63NPdF=ro=|yY{@lwu~hWY%y%~mb3tOX#>g6cFQu+82bPA
zrMT}KYp)*QhYjYv?#!QjGV6(g07tgJ*GC}10`@+r3Sc&V7yt24n$luHus-;BLc^F%
z*Ioo%Cwp62S=mafGxo4Xe5|2?mL8Xb>OZffqTh8by}G*^&PnbxI;%_oaZukC-@F-(
zuoEN0{NZ){k(_p7ZWh#F4D9_dD+bb&UTOn$^f0wxfB-fG&yQVCM_z2$8hPe+Tp66C
z4)T?kuU^}xqVvUF4CJ06UEFlDpE7|!Sxuf5aPl?BFC?v^wIZ5_QwY>}CB8Yf3JPpA
z`Mj*xmcJ{k9Jzo_b^0jyI_#(5?F6b|h$vuGJ*#*_S8aRnyPCj#|276dm0v37#fUFG
z8mj?En6jdL<<Dokp9s`AdU~+*0YWK0pB@O-z7r6#rMm;c(~G|p_1C9x%Cv8*dp7ny
ze&Ur@YyDp|7cL4+mBeqs6wLcz)pB`N$fhNS`!4cUcGrM_UQ(I!>Xmmu>%?cjusv|Q
zjeuVpojU6QAg*gGP7}fzSO^_`n*r-^Dm^zjW-}y}jjBDibcjAJyRmz~>9h^%(7VVJ
zcD_K!K;V2s7x?tyZW3iD4K1yyT7}E-s%MqJOTfsG=Om##-~1}9H>S?;alzH*DiXz!
zjfO;iter`-#DzxE$f(-&U@D~O6~WpX@D(rjPPNysDIrJPX4gg;8%U`BTCu_&2cyNh
z*OBmuhxqu9+9KW4SXR(4Hzi>zIHN-Gm$fOB{<n%=Zz{0kQOMOdSKm604i1}$koe`o
z$4^Ynmi?-<2v>#vN`tT$Nc4e!1qm@IlgchV`M(39h;NjNt#Vt*O{)+Gf-j7A%C8@z
z{wl#QMfp|VOo)^Zs&vKyjnx8yO`SFJBc0l!wOXZBRj~R4-+FkknwUvK#8#G90%tzO
z<k*6_9CH`)+)8Zt{paxsJFz+~rnt_qdJf=o%1~d7(w+<L8}r58z4RkO-aB>IH;9qp
z++CfxPC$y~lbzExl~@T3TSEa^^yBD=ketNqw)nsJc=+|0HxW?$O`#_Fz-tKU&INop
z>xzRB{csBjJHwW;i<Y*WHC#O4>_7#`NT-^pbk&rq0!@t#7d>A=$#UIyR&-A-6O%*W
z|8>S}Gf;+g&a4e?JGTYvL(&2WL<qk11P>lQ3T{+YjReb5u+3}sKgIN|N2E1mGMe79
zva`b|$w%<xpQ-NL?QL|M=aQxba>^0!DUs*a;;kVorL}(q!s#e{U^H_R+7sKi<d|Rx
zKH1cK*ND?~eLGar*lwM(Y0B!4v*&2pL4L;4f>_wM4<EKVsc13|j|Kj(R)va%@+k8S
zC7XgG?0t)I5gC^HQVD+z0m}lPvhVel0nKDcKyg{$XCacIhC%(Cev*>a8tc$@wf?1N
ztl5N{3zDy9<cSICyx}Mi{baB@O9epGAvk^DUi#yX2(*V+%uRFNNRYW{Ym)iwnk$rI
zwDHfn%B-Q21dGaMeD7+2+W_Ng+c)1O`mQy%$D_Q;r4>O-=LHO&=9)4#L?eldcV0yb
z1g+K@VgSxTdGaiMrRqA9Wc#+nT{02?lufIiE4cH_|FrwY3|yA<?qINEo~UmyY7CvY
zcvF-4=S>=zadv1{!gWbJ<lY|LLe(J5jeC}!b0-4?=~@MEhJTo)n6A7e3Hef|#o8n5
zBltu8mFmiFw<`ehLwVyoZg`ER`ltfrF7Y*};HO;G(N4RFv-pZ4_|vEISg>xURYXBU
zYj`cFEmyQD7yJIwrQ*LYO<hf3Oxn*^l55&P`Q&;`S6xuh+}Mh03N{{~f`OsI<2#fg
zVtaeGDDm<6HxQWI!CM2K<nmwIuTn|Glz0djF56gIeSR%gX0?Wcq!)&RI{f5O%;bJP
zTW_;YkVL-XooAzIN!sXN#N?IfFnvDX#!-n}O`NW|i8_M}yqmhf`PreU{zg7KHPvo)
zR#{m;CFbe1jEn*m{`5l7-g~!xU{CUyP5OJw_lxfraX#hxz9@U)JNP;}>2MY^nRYO4
z)6z50hlr&_M&gP5R1y!yKt$4%NkW<m6{%@yQIQ16&yk%A0DJ6gmqdY}X>vX1gSso|
z)@W-IL6cdiQGMn8hqDt2JpN^67tCecgdB88HGl3{$i>^i{PBuQD=p1{>F5=N_1ZPG
zvnr5kl3YVF<hQq!0e1O<H|XDqJ3Ei-5|Ako_*+^|8&c11=2P;*=owFtBq1?VSV?tf
z#kcm?|9u@2%{UvQ!6X%?kSF6WA!KG47-J82Sf%8Gd9}rCz>}bWG!Z1gd#VRr8oS_E
zU=b){v$Bp_yw$Mg42R$JU?9|0*YAFQ<~;q&%eEB-grpX4FZfVT4j_2u+WT?dmuUMN
zU{FAj<qF&xRrcI932tzk6?KjUVP7KJe&dvdkx?dmQ&<T)@>4AqdVTQnkE9XT%tI|h
z0oRC}gpPjc#wkm>RCLvs>LK+W?nZNPEcS=a(^1HU)U&d71-oy?I5WXx1_e9<VG#B<
z7k6N~d?)MkjScX42elXJiRXm-uAtlVE~sMx)`g@)6a*NwJCt`@Yyu(N3R02h+ap!p
z$VUVDeOM$#w#%T)kQF^;kHtsO#KZ&!=6D=N4?P7EWU0Hgg~`ZL;#cM8=CI*4Z*#?n
zrNCoyWmwHwpxkz>+L0Gef=z+KM=(1xD6LDB43jaM3*244@j>!JOOFTmdR(>LBgOOq
z*<F$sNF=YI_@7kz&0AN1mzlh>1WvFhOzk6>GN65wmw8d4Q_W8FVCPwAz1BB;-c^|W
z5v?vSZ@8}hl2TV+W7MF62ZtXoLE1xivf+OpD%r$qmN#UWpaW4<7yvzBl)ZZNV<?Ok
zP=7^EoI^Sk#_i1|Y`jx;O8Y<aYs@8qfLns(8c1?5s|EKlLHvoz_J|sXdho`^X^RE0
z1x$UcCO2;)Pg%R}ZVaJ9SLFsy((yHnm#Nswh9Mz+HrG`2Ra&PkM!1`Rht1B)ni#Y|
zufn7|9q#7p48s*wi1y@k6Sf=dsJgC1r);skNqEQG8g%Ed*R*740citF?b8(?*XYXo
z)7nJx)&VdyofY;-PtVRa77wTQhW01#)b=+3`DHxg@Q2HZIs%0kbJho}WEjRZ{XU<|
z8faC_iDPx^4Z2qJlV^NtCJnT#+}YFPpTgN_alpxa9v3I)%(ckG*-Qf^sM4?#3Iphp
zASx)YDSm*1^}Nhh=|n1~#>RHJ&hpi_xKLBjm_|M372aO}0{4|)Jqk5*^Vr$S7Yn^U
zGaZBzlpz9IR?Ekds@+U*Mi&)7t#FvPaCEG1pyvkDLz#!e0Ez+I7x-~s|4d!I@_WfT
zY*xGn%qaS5&X|DR<5-j`Mjmvj#J;e09`PVlDRSusW0W0q^aH|Yz^=Ob_zvTJaTt82
z$WHD!)YqpVUA?_u6RW;_5qDIpmZxcNWyJEWzw+VU-f_#G>b`omKhdzVx_S{%E-th;
za{nZ%3?uJrMsCrK-T3^3{07x$n<0Y1F;9B%F|wznNlgDt^6}wkA#FsQajdMh|DdtE
z$^|Wv${}xKqpZ{Qj`e(7M1=LdeckD`le(#wXtTIj5FZiLICDS6Rg?<zXrhcG-F+jU
zQ7=(JUQslbLSRvol9V<rsYJ+n;0m$FJ%Jl%mLlux>%_#w7|lR9X*;&Z6}Wf@c}D7G
zX4WtJCxmLchOvE_iwa*ymvxO|zgXlKc=3Yo*M=<bBlz?<<5cVDIwrg&C?p@x9T|to
zob;?!9F(peKJ49(<a$>p-8tdJefcWHyAj|TD#Fe@UTs+NDQ3-{(Qvs;>PCXsQ*&)*
z=LVhDA9`WAYUb<ZPdwb1&^>_(-!oeB{Kn-Kx<CHf9gpUCe0r_GnZpS0+de@`N{Sx+
z8x?iAzlS~W|3S1G9^)(C^W^e{5<1z&oLCn%P0!|I+SNU4TEXZbssjB+!UTx6^>yZo
zOHKa5+xYKa#(Dhe6Z{NY$;ZU$iK=A`xK59I#{);WUy~qGj*vCqT+HbB?D~ZA$QyCF
zB+WQNAJJ9s1pk`Igcx1bP83(2<diZsvtL!$U;oi`$)e(l;REf^^fOAzgZZ^z^Xp#x
zA7gS#zS;R2yOp{*HrwEtP9MsI<sFBxzkQHJp=S_3G4y9{bBhI|nJ3<y5d(t_ad>$A
zjLBzg)b`blvxFuxD@#$JJZV`gUH$K)dJ)MgTx4)ghZdvAQGYNWW54lWrAACKbZ%LP
zv0ldwC+QKlRk}18Oy?rWgcKBXd-Go$e0o`fm&C-v#+UCbuB+R3HYp9fm$y%s%cwmi
z*x8tqSr$#(C6csfE`pgN6<R9?Ooz8#hQa=`T-sxNngirax6|IEFD+uC=pn)cY$jb}
z&cADDgtf;Tbe>Pdx}Brk+~8v+b03g0!ww1Q)T}(;{+|6f%P+$cbwgE6_6h<jnAf6`
zjSn#d7F8kX(9Vpn!QgT|cu1>D;w3GuPC>8-oaA&kzCPar^3x`&Xvo=I2g9l3$b*_^
zi9a0+Rl5A$tWYg2y82!_twX)WaiJo@_~+=*cq?%5P8s6NS{RI&zkESoXOQB@ODeNn
z7|-E8z;iT4j~LBzUib^cY{xHx$GRAgSBE<5(%C!(L<fIX5g)}6tZ7so9d}Fz$`rsa
zj7sytU<>3SPktw(@lAon8WAc?T?;GvHaxrwq5sy(C%y0$R)abRK22#3Az@*shVNU1
z|9nTsD(;Th5lM75T?Kk}$(ZWE14oVOZ==?eRF)<a-FoBMhUnYdfOLm-JihimWQDy*
zuIF>nl@DN0Kuqe50p4RLcL?q}9j1q9wE7u`Qa8c*{8Q!+&&dJGdw67c+_#K6_c4Gd
z0fr5@Uj(b1Y#wpibPe6&gHLwF{EDXlr~9@vEJ0jXmzt4rp&fmzXPm;qIi{w=3JMAv
z!}ia3AiqXBqnD*|ALdil_MK*^$GgeWP2=JNg7tMM&l~Ub2UC7kR8pAQc2IITb2WE~
zRs~8p8+4OriuCnbTYtJ5!KD;ZZ%~lXuWcrI&36d6k>~Ofqo95S5g~+oE_?J8S}9<s
zn5pxCK>9an+UHcx;Ss<TVx-dUSAp-(Q}gaG$Er*woA0Zu%2A)II5iaZ;gc_Ag*rCY
zQM^$J=Nuja!#A@L+sV0=CZ`rACZj65IZTRRN1cf~AQ#3$7k<4P6y0Z@QmS`EvWBB#
zKH>hR+UWsPm#|_eNdB-uApq&_1Qi!uTeXSP$Hug+S^$EztNzqF_^}7E%z}(4zhLzC
z03uE3X;q(@r$%Q(O!okUHQYr&gu~ut4iSZXCY#TZ!E$R&3AHQ8p#ADo<6|-ex#$0o
zBWIo7%V8g%HQqll5q7`se)SV=t(ld)pg_=`WIWj0+q>#=UUi)qCSHvxRQHZs=jL$1
z7_5i!wiM~VJ;}X{U#oo*B&T~De|I#bn>wnzv*|{1Rern|$;-^%-l^GHJ@O4LAI_v{
z2|87dHqoMv_$u)&?t*~-g4?##^GAu{#Bh9{?FBb1^l|GdFpCuv6F)oMxM!^~E?9pr
zOvmyW5)YF}S*56-(o%{)QgVNM{c7`a8NeP@8JPrPPHb%KhOdvE`VRE`-Tverjj1g2
zdP}9Dz6YwBZ@oBfheAN`PGqS=TXID6syt$lkW%J2woO~jaeNz(&AhwCNBh+??&14F
zfR%AmFjP-WcK>OR2o2HQTkVe!(JZu(x<%iP@~2^+i0cR7SAEPI{-ui2piFCuN*^RV
z^NOi(F!ecjtgMWKRDSz91Z2{<&bYNIlOG3#UaLbkC+lbMIA2-k?dp}zi^|PR<Z(Qu
zs)@2M-y^`p+qc=X+Bj}~mzJ6{o{x<vW)HEBI1g3UXD!DP4CaLmRMuDVzJiQEso1BH
zB5KtJv5}Et7YW+eZ{xS-%`s~7TvU{-txiRT*M0L57ZH^_u&@nHR+5+BUFtJkZA!rZ
zvHF-@Z#(Dx2@$u&oRgH@x6TrLL2r@y2{0({*w?TFI~5ppl+PWkjys+%PKhh#ZuA&;
z704|>y`ZCGApP1EywvT?H1G*<C!8lgTgvId!owh9&L?KPE3n8fL5CR@qQqVu?FD?d
z_A#}8tD)wbG}URhc;vgskh(cThF!Jn3|R2&@?O-7-wDmDkj`me8-AZc*bDUZWS9^>
zbTt0cRIAqsPSbL79X(xdFJ1z!OXwzb!*3)mL{mnl%x=amL8N!QB{3%=Bl8P-pkIBE
zW&X-04Kvr1%@@3VC#)DN+~97z+FRb5(Yw(XMMWZD^N0WK_F*^s8of^fD{K{obp5cd
zSMFq3Slc-q5HJ1y^Au^C)|vR}T3Y1260Cn8j6g~0!EfU2)1PYNsiw}KPkUr#VP6YQ
zZ6_C|B;a)1@7VK&1xikyLJ~;D%oX~R{VA2hcy<Q3I$&dm@DSz|7faB2)z&7Wu_r(Z
zU1sKS>!n_QvMc|vbOYrHc#nKe0ITRvt!MMzD6q|{!E@+YUr%mJV(g(!*$LN(`rKfU
zMLVe<`4tpsSzJoMVwn1Tv%vOTsk6DcIW$bES8hauw|unf06_>nJoXm*m!#eXBx9#G
zc#$@~Zf$GHjjG8{4`-YNyL`8PVje>9hVJvbi|J&y;<DC|^S##wtZNS=2BL+Ij;`(X
zdWY-VFRzk=u_|qi#gKhdyc@<N56^qY=B>@N(g9&-lAbIpbQN~(8=)X&C@z+_E`4};
zOfQ<t$jiIT^s6_AF*7q`lx4~=;X%s5Q4zlsQs&6Lxn4EtrN(9=?cVIfdO+j>5L9|*
zdfQfH;+R0NR2qcIB8TgP8&1sN+wMtn9|2!RR3t>~=(ir*6A{#az=*z0%pz+mpHC`$
zHHC$;%<`|uaJ>24j)$Ug>FNF0DH=HPVo9sNhXw_yRcjx-*l^SUOMgFqcie}&n{F2A
zU`45*kRVG{QJRB$!!sxrGKj&q1^bI2d0glu_6=b9*laKYH5P&g-23RoTKW1Esp6%Z
zkoeqOF^b@WwLuXcInkBr2ji|;0g6ual71KmL|o&tN|X^XTu{f+OJXG6t=F>D!aONL
z+Fy6|_f$Gx#FfU}km-9VF8)$OLuy+BOvpaHxWFFszjuuL;D#rJC!`7=*JQX{cUCo|
z&R7N)Fe=8pO^x#MGS#n_rNVAl;$MSddRF|PA{nO0=;+G73w*h!k%7&IlM3r&Cj;vK
zUccX(DAxoDQnai8B#DS|+G@LR%huP>D|SddOKjE@@DM@br88=WRi?Ff82z<e6e%JE
zD6}3SPjGx3G;ecGSvdxFGr_YRmMk(HxVir4kh)0Kz;JlD7@xx_9F(@XuHHWPS2cD*
zbrxnt7>Lb?aeVO~S{Z#uJFrF-JFcYC66q)&y0vrb{82i|(yzSbzA9@+q58xl79U1h
z+N^T>y(bO%pX;VInJ43vTGM3RC!M|FjF_GGk9?Xl752pA9F5cQnjhn=DioB?HY>dW
zy4&|JYoso7Hr?rmy5uQdvnKL9iP@E>N4^7f?$N8!akPROQi{UD<&~8rQ)Aa*5>4{_
zWm@-g%@@Xd$ooQBZF~BhjHxSdy02x(8~NwPAo0X;Rn$7$a18Q}M%P3IcXedaTXqSE
z%PEUe)AI(yV5`6{{HF&jATR^%#N9l)!XozoQr(gGze`Yb5C@9D%i@==$8tIQ4Bq2n
zWaO`j?@FyA4~%=E|C5UwQ>c$+ScD8mj%rp<IZKud2Nm(`iQkfhAA%$KPQRkyO1MO-
zZ}3arZVq(Mo+{r)xT(;-NuAms9xf58bKj@WSr8}?8U00r@?1i#vgqZj>2R?W!qK9T
z@b+E$f~}j8ct$Fbk%F|lHnz~Ot>``EcRFl(_bD><;=X|U9lkhCWreeomN=ew#X57$
zmm4i{?=F>&)j=-dOP&+9r|vA-uXZ95r-KfBL}72meNB|mo&KBnT@0_9y_q_P^Y224
zCo1XcA)98~Cos*`<ikMzG)iR5?K}t`goFu#>=AJ2#CrZ19wX4d@LL|dcfLFCAUTw;
zr>v_x=c^PrOkbRig^rlKDQfE-3^?pE(=(D7by}v|@LvfD5OH<3&C1NZ<G7nlY@hIV
zQM_uQj=XYZq+$ei)(pcX!DFSDXq~Y&5@J$A)rOLjXWp>jD{L(_%gktz8&^M`S2_RW
z;`HFf&~I+Ky*jY6Z0kN9<m&y6aW9sXl2%nwgD>7Z&%=J>)63fBJXt$CGbr`vZD|KH
z3}N@xXFQyppL3@z@+!yM5-aI9Psz-@Y`T!22T#bDd@wfu$i(*a6h@C$_O(`O&rGy{
z;-k#@!d1(P_XxEA1s1tsA3wSX3C;a-t4h?GCi3DtSJZ==3yeMKe!skN)0xY>pVjw=
z9#g^A-t8mD6Fy5gT4`unB+y7ngJ;|+0|XKrBx4YyMeGagECzc~>EYn(7F#t5X8*am
zzVj3m1>6-Cz5Ux0u3&`&YvIw4{_DVn_cV(c11o{dEMrhT1Vo13w_tSGDaRf#ABWll
z1Mb3Ppqy<Q%$i3AdMzwH@jTAibuc+8^L|O!%$d_@hSq~pTSNbD7>7&&6hIh68ejf(
z9(b=`szY$A8BGfY-iwz{$|@?nJv>;+RtB4iI0%moc`IyFWvT725V`!ga!Xt|)qNti
zE@QiJySmExhNr*<_Xp6LW;5C;qu+3O-li868noPBuP!W{T3&u^t0N(-7g`vUm{eto
zLrKZZg##M9VZ-z+UMEbcY(`@pgST3}TwLtc>%~C27;PR?R(i$FGp(SkTqfz<YEKjt
z6eRa}&9vrm^3LSDHyB~yPKKHBx-)n6JF6D?<S~SflaRBWIo8o~t47r(CRQALG(K{f
z{{Bhb8zM*<ox!3D%Ctv##hID8WzHKf42K%8X&SH^QQG6i;HZ?8oZ{w@PPmH%sU>Fp
zt!*C)6iaHTqp0-xLX1gI-}Zuz;zwX=Kr=k+H(gJ&6Q^g&Yaq@M9*;V$ruJ6%ZMu&K
zYogELfDd{{ifNvx^=zmNM)5-R<f6^gb=JEU*5%F@#0AynyjZhyPPQAM2!TAbSg#sD
zhmvJ8;I29yF5WEA%MVQW04cI{tsH{qKt46=xPwJQ&P5UH@N+_Iyg^V(D)dc|a&3Jx
zA7jnrORd5W)v@J~%li5^;Z7(WFu>>~BqSt{RJ(_T*@y6iPL-K^ZS4eRNwlrYtLur`
z$={>G+z;HIZizw5Xp)d3!@=*bL_5?nmRc1O;^_!^w0~ata}Jf+7L{>v!*d1e(Cxq$
zAOJ5|;qNC>oR}>fE3NrT`A&%UR3DY}?^vu49$NidHeK$|gBuLarJ@G%nwpxpTuvqa
z_2uVw?r*;e3kw_V22{6?G5u>67yUO^1{#GlYp3Sm&()c!UX63-A4}T)+tl5k@8rGO
ziL>9WQ$9Vzhh1JthH5stvQ`lszU_tny_WROA4}F6brmN)-qtM=z$Bu!jy`5bk2Zu|
z;uXXm=cwm$On5+SPm!gb<<FJreH~drGLcPqba$~>Z-WS$-_5ns3)`F%99U+X<sdem
zmNW}hAGv>=*E0R~E80Ie0fp7|E7)IU0qB{P)ooC7PMRg8T9p6f0vrX{ER&YgG7{WI
z4r<5n!~T3cb)QhzQKo|iH?CrTdWjTT-}b-UaZSz5Ue}IFyLwC-xY{B(Z6`V;{TS1r
zlkL>f0OrHH7v~UwTa%dy@f_aOJ^<svsaNfksdUg`Z?0kXmh;5UtbEORs{1So3dx#G
z`Wp#fzR;Yfot^)8$0xEgv{_-e<`(w)M4fM}NHwe>Y+u2BSo6@@YR74ph#gu1^D1%u
zyGosIRgo1renJ!+3q?Csmc|naGk%Qv7afu|Tx+d134Pwo|BnmMlZT(ps#iC;5+=2`
z-d*zHi1Hi3(C^eSrPsXfM|xY^witAc#_Q9BXWrOYIA@t&PdvbID`j?mrWE_~&fhcI
zX8SazMgZq}OG_(|J<c|+V)}s11xyL2=c1(68K8-qx}OK@6J!u0kr6mpSPeC<uXT!^
znkQvve-DgE$x1roVlGcil$ABPZcIy<7y1mxS8&_PN}a2$fBWd}OClu&g$GOU&q&=N
zh>*}zJ5?2ppf6_iyA7r)L6E3hZodUtO^g~<@6wXNWJ#=~sHFd26^C)txn<_8Xq^${
zMb?&WZPTn)ELLcrrlxh$G82O$ns2&qiBVyRwV0dhzBB2C0-Wm^t3(gjnH85p95F*)
zx&zL6VsUn~NU@>$+^Rw2!Va)^kSElf<i5d@(LuTe;O45MtGqEzDcsl~-VzubY-??a
zbyJAgKNev%7@=OV3-!v`GdB9!Nuq({cKm&OyoChvuY{Ap>l2b!Zo#(l<A<V#RIV$x
zZ{zDYah2rEL=z*eU1Fax*;gmf`j^*7M&Kz8Sisl@1zOYcXG5^GvQNGu@t&bxpJM2J
z1kwE+$PFVaqoX|?qbR5D=U}KPmllD~b17UD;e!}qYrA4OC8gH3X+S}@D@GRa9^u`4
z>CnPwn=xA@*4d<0pktCVf{hdA!Ry$taBG?Rm9l7ewf1!(p=(O|jl;t#AD(Z{DtpXk
zOFw}&2!Pzr-~8MU_+)}|#{^n`=b|D!818{%?{%r6jP^H+aWx-Z?#}t^r)p2W^n)D?
zB3ZzIuEKurZF7^LaAWmFz*Phei#57-M;?Z)XCz5ltu3u6&&B2b45qUoB80&nSiwuv
ztMSR{znVK7Pj?+(YTO02y4-`LmX;O(!)zzscy!srlfN3x6N7`-u73Z&vZgK4<}rq!
zu&rIbFI*nMBki#c0oXT1G=CVZZwu`xGVW-U>72J2J^(IFQf@A2qZJv{^9{$Wv|Jyd
zdlovLeSt%_P)CP7TVuGtt=OzND~x`%r#a+VXmBvEv&+TmqhQFuDRJwQB`Gkb<lv8E
zNqTjAX?EUa_g4qN;rExI)PbMnvYXA9d$rFH&hp22`D(?TekA@0qy7BgIlx$)oHo%c
z)-8RvDB+>MSYl&#{Ii@gR$*r(VNOLuqsHSp*w_4m3hNF&^CWO?Y#@G;N@u$tBMU03
zcg&9nOon4Pe`wYyh))NB?AB2-2F5+ntK6}iZqF4%g_6PQs<_uMAtl4i+*sS(IBCnU
zM^0Ql#yCtWvV;lcQmw-n$?q$o6cAS7u$(W+Ev`#mPrsrhCdyQ!I->lSGUp9*+>12T
z@F=jg77^c>(f0v_qb-it^!U3a5J3PB)A(U=2kGtYr3>dB8%ju;-DV;M_UmkFN<y_+
z-*l%~>>68rZDo7fOmnDdJ7hrK-54>O==KtU+qgIC^Qo|q4zRL7*O3ZaJIbC6`Ib-&
zTKMfmfupJF4m&J}l<?a~fNnuY?6bkDEN>C2$2{>Hzj;+_Dq1A6Ld*Amu#<W_Nj>H$
zG&|mHp23A#)4Np>YU+##P*Fs&Rh?U5Va@s&_~uA*JFDVH?JL@L0@^k<+9nUNt%Dz8
zg5==1<+C>77Z(PE%oa4*UJ_hIX8NKOf!|`dCRA!day$5)f8SmlX|;YHhg!jLusAHL
znU^wKvNZTxMO3uA>v^05@9ZpJHG5_z4P1?@lbw2jj@#IF2`gK!#<M9Mb6WEHXy$iq
zP-A}PC?cc{Ytv+P$B7VRW@p~5auC~gdg6gcg&LOpAJ8tz^Q|G(l^c?ZIAoZBCUfRw
zZ*Nz8>xK&g5$Mt~GQ_05Gs7hlg`U>Vdy73)MJJ$jCA`YNJXU0UJ<9Gr<HTTbscb;l
zL5=msOIc>bptK7*!oO)>-FL~N+ocrb<)5I{+_R5hw&wE=o^f81hpq>VA|NFWw>?e)
zMA}-T@(f`djm9naI^9qX2+1f4ihxPAINnFfaK7f1Q(<krBS^w@nM9l%61(1zQ?W6A
z*DNE;djIg;@Ij;5VEKx+c-?eYe%fjSM&ct*6KNUC<}W2tD3kp$&Tf4Aqv|p1*S)#8
zY>G{c*$wJE04+ztIZYF7xmy8Kpa}sZev8#!PANMEKjA97sbrXcoLwhIEaofyU(i-T
z+5NInYM`oq_GS_hziYMa0w3aV57Tb%bZdpf9MEDYreswVe;A!9A`Zu=^<@vP!zx~m
zj1+1epde2R5J~>~>HruU>bqOlkOXNmV83X0fTKk?G_vcaD+A_33atw#W9SbD^A3_I
zr;uN7pO)L&X%@SfsyKPk`KDwiL%8(Jw9W+uO=|kjy7&-6?!B2JUmu@?t*V;{VAt^2
zIzeWVk6`2so9#*Lr-P3I$orVc9}ef)@!tMGu$8eCmBg^G9Sqc_Y$rqWx`&3bC?}&_
zVACi3;rZkOk;QrMRL45QC0=bfb~f2{@;l)VI7-<S=0!1Jp{InmrT8ksysfu*SD_Wx
zY^0<n)@qR2@YU;BYg@ZPprVH1=hl4ty1#Q5L3DLf*4245UJ1)J%?@b<bC^%;Y<53D
zL^_+`+Q?_SgOB5c^48S_YGW0Z20UI>E_avgzyQc+w~ekN{=DjdkjIX8(rxTi)qdUB
zyl=l@?bl>Le)$qKF7Rt@L>W+o?F)^lEtrQ)<f@fkLtJvl2MgU&Vq?t!(NNtp)Ye*E
zUVY{?{WW@&2Z@)-fYj_uuLRWUoA(5WLcWLI^sJpIHR1eMDI6;q$bS`<lIn+O+Gq1R
z3vTw2Aj%^};JSe61axwkS;T3vUO|3Z4B;XgLc)8?-9jqDcu4Na#1bt10|r7^`ZiYf
zXzxKZY+c49Wd}g3B))MLaqJlcP(uLV9RykfX3s0geMU`6@-*c6>Bs?uAu}`gz!sgH
z94=H}5B4nA5e3Na&Nnm<39rMs!Qb(5=%?xHv19wyw-|`2grDlN5h?;XSBB=)oU~Xz
zp9jUXLCygvBl$O&v~YA6%<HAXhc}|CeAz2pVMTzFBRUP|GbEQK1XuWaN3j6VFHO}m
zfq$XZ*S{)y=&QlY$8aX6q0!yn_B|#@&!qQQQvP;elay#HBh{&yX=ilB>ilwV!6&vD
zso)?jF(&?Tl_yvYmEXuHT0uvJAfH}9_CuK?t;OypF}<xV6jNZ9vAw<H6QKz3q7%I>
z^mBR=F%~=hW6pO4;Ehse7}oU*OfhQQub-b=!2iLHyazn}3;H&!UqRaBa9K#K1nKjm
zh{xc;D#&rV(TmZAyif6j-d&-TfyjI0dq~)Qf|CcGB2-wyl};}uLR#7x-*znlV~5VH
zt2dv~^OcK{x<++qC<Ouu)y;)(fJrR4qNA%L|2>4Yw{LkU8+yRuzWR5=bmo_jFT4GG
z?@U_f12!xl!E2sSpP-$o?U4<a7Cp0=y=tkZF1It5@*OyGG&Hm$rA1Gwu|6FZy)(O!
zu#qTDb>vwext#gG<31V>I|&XuPS<UR>A_C^`16Mto&v=CAC-$-h`~<x5?rBs6w#7N
z1EUg5A0uO9l;>op`bz<IWOlF|pnPV<>l^Q<h=>u(o(_x*O-RU{U!J{({Xk#e_`%jT
z1_E)QrR-hT-vjbCZPiReCP9IUEZUTpN77s(57+O4T^%;+`!{4)5Idt`{SZpMb}Cxh
zLY-ukC)4%D-JNdC7dU7!QdiS%2z+=C%8B<_!OgfXy4??ymHRVDomq+s|B!0vT3W%l
z*xvkweit_Otm?AC15Gqu7eNnp0EM1!c4CqxGKP{K-S!jawX?wWZ*|^1RJM~cv$ge`
z{r-jy(?umQ#cVV2XENC{*6i1LW)NfuTFj^oKW;%MF_rT2@=VxLfl1))wD0fpx~aWA
zD>s*zqI&CsdRdN<=^+xIk&s&__e7TA_)x5$5YtB}A}%!TC{6p(ZIA^R+S*7!AaF}3
zFKOVL(@s5FKQ|OIe^uFg-CTfXHmk!Ap<msjcl_Ht<|&r&_3<(_j_;C9d1R)TGp+AH
zV7H(t=MX19fk05e@x{PQ#lzcNo&upZ4vvo3m=<pm_gV;ic*;#lq|JKSJP5d#3D+Bp
zSG|sYYAS`w0zGO#aEN4Q6AcXm3Ig4p@h8P^cL+B4Z)E6GQosmAk$#7eCqFJIXgcB+
zpP-hJN;zcNvdGPpME%HnQyA~8<Ud+@>J&=d-qwQlemW?K4ZrjfEXMY&uj{%yjSB7V
zLR74tRUsQ|FQev`ub=xZM1o?bD1#PIAvMBMv*PJjx(l<J{mL`KLdr`DI;??a<^wN+
z#)spveFVkCq%189^_d^u3d@1e0Uc(?^E7hX<=;@N5<Hr5p6S<Lf`<_vusWs~;o#t~
zteWV3#n9uN-SWLh5;kxgyd+OscC=G+Jv&$(yyw6_7$2h1mG+J<!|;&`9v)C-vj)rw
z_#EOh3hr~WRmG|<%+CANYXQJQt=UR}<6EP4H~dfY)UB&=k|)QUHO~I|tE+0r=1$Ga
z%0!IrBGrZX$yfLD#_wq|N3;=+aouN%%-e5xJvlp@J}^Wv+MmkHO&)Ui<IL%N^)a8O
z#o=Luxu-Wk6CNZoV?x#U1>)}e)IW-TY)o4;pL^rK&Q40QR4+fu_W|;l$H{8zi><Fx
z_DH2RBl5H2nLt&X$IIGSa3%I5Vu@B##m74REFvM<YLJp9p41JX7_4z+<?u8Ym9E(=
z*sf+KsyUdLqgl=5Pwa3ie&pV|wavlC^wAt*11UFAkhUn^_VSoH;7ET$&CSa$%BArX
zD04ll^(Y`sKuHdlpr#0Jek=0Lkd+4Ex$Qcd^$M3w0UKF-YpQFehqq7m7$M}n$-!i_
zHm!3P0gxuDpP{t01O<U!^0y8ZO%t4sFl-DGAydW?x58!c9~gcfCzjG<ATPQ|`D}LU
za{_7z7B<SS-nzw}2vR_Fkb4ipCcy(NJNZ%qx_`F*CBLMk1dCZ&5?)pK_TDA`Ci~#_
zX1jh^os_7kq-KSdC5)We)YjAV!`7$fu?QY8Vxh+gRj&3``RWiS<k2wZIczPrf4`}{
zXv2l<Zv53g_<<2rq}8oN_}0SUO)0}Vc=BD2EG<tb8Frk~#-*EP{X0FSf*(JiRk#rJ
z@On6*V`L&o5(pm{S#WK2=Fj9=N%^*3A8oiLK7x2W`5Qm{vR0b&vXaL7+je`)!~w~z
zbLSf{`^m{IA};Eh#lhGArIm$={VD?Y(&75#-;t(xke_2(O0+$@<@S($u!v|22QVO$
ziQ-6yQ5&w9CSj7mhqueIf<?bdJ0H=A!ERpUPL$XdB905syz;?15W&?dw7UHd$OYWP
zY7>JrG4=kV23%-J7VACr5i~N^1KtYOBW&?x7w~-TY+`0LfN2?+J;IyoP8?Qov$Pa#
zPdgHGseUp!{z>K!+L-e_66Hj4PV$oE<lU{Jg)1MPtTwYy0?i20`<_$X!@>;{LF)xB
zJ~nRHy&2um+|j4*`^DX)X#PYjJiMgV{ii8)#nge)@@i_b8X8;5o&f*O&gs0my*FBF
z{Q%67?U$A#tix9Hs2CVBa?j7v{;=ZhEqgu)6925tfAJpJ5!ZQLk~82_=rTiz196F>
z`Cw>dC<?OlC<17}5$uku5H!4KGkhej<P+i~K*Vclx9G66+Fe+<J(Jb})#|%#k8U{L
zKD@Y#GI^Sk6*Z9KU~{2|5$4QDrL!>p60tbSgqXt7vf%4LUpJ+{0Jmb;Gmv|`n=(+O
zyLJ%4fTV`#F&_2<%53{`IeCd_m}vqN9byxI+SHq$tbc07>1gk0-Yc_9N-UJcbO)jG
z<w$Twqng9aT&2RLxn(ctyN+EZo$uaew>IJcU346wIzJf&%KJ+<*z<)QaP9D$wal9+
zG@eppyvSNOyyB)up+9g=dcorv)WgW}=bOXP{I978BkVa~@B~h!K;Sc<a7|CMou4;l
z&sHHh3UUx$Vc&4rqbP^8sltVFX)2oY{Wq9Cm6fNEqOv|zjUtj<`>|Gtw!c8GN021Z
zR!ek+hQGvYES2Eae8*;ulLi%5YO3xkpG9(P?IrW1(n-!<V?3&9akSD!HOkIE)&ng2
z<>eJ!_i6#t&r>gR7>qxMnJMrtcB?E^rKI*jhg_#@h7o44Cq$%4G!Ye+dkiDbSiW+9
zK?=*V%As4K_s!>JH#mb7BJ^Iz@i-p?pjb(*aSJdiuwJ=reZm7j8Op0zEY7IfUzm4K
z17qW{;HNJ8K+)OhAZBdc`v2{M>}2-#2LS<S^n^^n2b*nJ)7iCV&r<~yiC?=T5k^D}
zKSBjbf}3kM=xbzgL1XOh8qjdkPhTA=MI7wOf9~C%heKYkgU(Nn+Ix@Ra^*Mb$uO`f
zz*(56d^xL0hxz8DH%-^o&UK|DpMEwJbkFTH*P<uABIxMbC~jfP&A;ZhI*vw-LbvF!
zn9Gv<JUt^Ui}vpPw?msrmLKf<<4q5BHdiOvI=3z`YH;v9M@52k4e)Yr@>#pB;Al!-
zl=*LI(u*;lr+jc*><8^X6!sDk&%6ddq;I)Rlm4xt(%96OvN%MoQNDM4_rVuIm_R;E
z`q!x*N1b&u;b~!n0r%l`OjYk<t#q)JwO#&Nms`7+=^i5oz^@`CdGR82I2Y@EE&t)U
zL@WN^x}z&>_C`~(+SPCEvbeaIA0LjJtgAfn=&JF7Wk{PReK@Y+%%7Q`ehA~=>7b3g
zbkBbfryDj@AdHL)T?L;46JGquJ5KsnD0FEG-sUX3R`XO?PPKKS*wtQSF__=9jP?@$
zNvNr7X2FF-KpwMQ3rP41IF&3tV86US%f$E4S)Os$<o^QcJ})F!qTv1efq4>4pAL;m
ztG>>f4@XSGmk{z9wUyN^Hw77(Y~i}NL{<FPr{6F;yv}MNjuX9d`EzH~hgV+c>azKg
zj~xP7`IpSOK~!9vKAKxj>RaFF+RMv==2#VsGE4>weGgsfI2idE&T~R^M>aG(nJ!;>
z@%0OeK<jPv+sgF#n)12|A+ru(aIDT<!Dc`6@7<p>6j<KmFu#u32!T)-Wmgxt{?$n6
z)z1Tt#KV$_yV4ISAN_j^sSP^J?z@;@yDv$8?)Rjm+^m&QF2DH(+ui^B%K$uvJg-*c
z%exMLouK#rziYS)#x>SEZbR!*{|c~0&8z?2VIq%CTpis7HhcPimm<yb3pi>pY}RjX
z=`dZV`Tze%I%Q>Kpo`}Cf8Q+k=l{MrL7Lj}NC-al%hx3C?S1;+-BmtxLAm~a=*RKN
ze}Mb(jE<Fs=zpI`&7FTNxCDMFQRN6}s((x6@}epJ@5cGROpxBb<r`Ij^Y5b5McLSe
zNf{~iVL+mkCiI6<8SuTL0EG6|PF`%xGqc&EfcLRu+yj~-lag`z#3-EtSIjhc59WqQ
zPk!*<x+$$7b6yqW4KHv;2x?|-%9=9c5C7hEBvuI%Zf7)mHym|bF=z;*h7f~#$%n=n
z*MrOv`%fg=)67NB{L1!#t_2_##J5yC?ZYaEk^$I$H3bWo*E}IihXcgQH>weA|8IKn
z_vq-MWaoMe7*>bFB**o<Fxd5#y;2%DI;?S8fOkh<Q`ZyH;NaN6n+Sm1*RH&PaS-wr
z^mr*5C<u_Yul&A{EGh+VXnY<nLDUl*--|pq`Ai^Frg}NOcrg{ci5t{>B*SEhJi+wg
z%L;MCe*g&XG!993W1;~oe4iwQa8?}nsB@QziZcC)^}xjryQiNLE^)o$4)kr0BBI0c
zNXfC^K79!AEc&*Uf<oWB*R?CcE->PN_Btr2XL0aw-t8mY{PQTtdJqJk<gZ0R{~q^q
z9DbstrQv1_<j!NhBQl=pt>A?r&N3M$h}j$KusWJ{1ezk3<2(tHty+>?o>^YH0Cw`-
zIXl0qc@0_dKvo}F@dk79VOMF`!RRA3hSJc^&Zbj+7r!rGwttVQLO>U0@KDa4Dv+2Z
zwjyDut?)VrO0eRhlGMM@j%Hj0Fu{R~{L*1A1$P*5{sPom{FXJQ53AAPWZ+C{pG6tk
zvD0-s(}bRM7%Vpkd%-hijhlN9-4iA<ly^2{0>VIj(%aJ$&gnXxeIy<nB=@<(z#!f9
zP1UZXPMgVwZu{9)c&ld|;9!?HEepiH|D6pa{1*e9+JE|ydj$BelVpUny1_vtQA?{P
zQjVE<4h~U5UNp38-shSaO|!sZM0q><T^ZHun?gWXSac+p<yX5~?byZ~ueVg5jyOAt
ztFHm7+N+tRo5kU@`|x4hX0BPH^pmRy0N}sk?}ahAYENVK_Q?~hav{Gya<aha;jf+}
z{BXYE%KE+{PL;C_q<i90Dqzn6eg0b%vvKd?l7Ac@r7&_oGb<~u#^G%L%$vep<MLy~
z0ZtNkY5e~den+Q&5^-n(^flI%sdpAff<;zG|L(L#Yuy`=`}WQ57I?pX?=5c2HbETD
zr~4*9xcbQ15oALaZA6AoAThFebb6<M4UxJqP`Gvl1pyMz=*`<#AeMlXek8IvA9x(e
zbZsM>4m|d_Fny(@2u48&vU5&1uIlONK>*JY%Nn=cXGw>OjmfR@8G9I|j*BZ@^k0Xq
zd6AJ4%ZK?HXXt#x{}2E{H@C-<%3%4H|ArX~T`&f1nV#+C6~EHch#w%b7Y57^B);71
zWR4oXkS?=?lmXdSc_|%#H*q%}1iL3@*3q!DK{8;?RRk^g1sLJ!Fw0bXm8%#G{sm+V
zaX8|#>O0OcG<*bmGkq+7>%}l9F6GHY*9F|UE{?ecT-P5G_?IUPoR=p@maLXr=6$?s
zpjQSe;g@fI8$Hbm6`mBx4IvUlT2}Td0(P}{e!?le_PAOuU%mv9Y^6j;XV1#xP>$r6
zoF{0fLBouI+p^DnCjM1?%S0Hi17wO9Qf@dB3m^oHH7No;;`)i9Fgt#?sEzjNo$;92
z*)qAd@EzCF-J?d0Bw6TRjE+u$v@o&#=5i-Pm_CcQ(qLiF!W{JSIt9|HPhGyw15E<9
zokwyMizU5*K|##s+Ol$gfb68K4Dj!8^O~zxP*RMkBu1E6%9Zypw-3=(y`=9W2w>_h
zEn@Io!(6F>ywccMkIQnjDLdOl{}ng#Bv1so_i(oh-VD%bq@t_ecK4ec&sqYto8XyM
zp`TAdYAP23h7%OQ-OXdy5RxQq-Q*NMKNZyXA3yRSFrx>B3DCRv@n=Yhv3x#$By!x}
zr~-@5m2>-Yho5)wy7U`kUvijTYm4MF9GpafEImIT-!zs*c>Lq@rLEr4@{R$WcP8>T
zyj8c-=*lbq9f3E_$ldsag8}!=+*3`HdOT9|`gMZz;S;jewxd>P=zyxNwzG4vI-%dc
zD~eybDytx)C@X)7lyQw)>)jgw2A#wH1_p9K=uzqXFC-8nD@^Jk3kmyq4)7vB9Ul;T
z*=cT7QInFc4b>b>xsDhcAHg6G*m1aslg-ZJ>N+|CWh?R=t3qea$<#C_LXJ|*jLGrT
z0D0zm&P>T2Us;(5t;0uVAR~YR(2br5*(c(YA_)YJ)Z*|^PGO<ZNG@vWm35R~K7Vn0
z<rVZK2s-2cg1giuGJY`0Bsu~%(vFTo3bwY^{Cw~^U+7-?bI_*w6}<R=G<mjbK@}Cw
zA|fI$3%|wiutdZh>v9D*r{~Q1@=4w&=~}XZA)m6A>)sRS_qf7ZPnM}Epc$KE&rgqg
zP8<wgTuK`Zl$5Fp3iQggFe&e<xn#fm|8N8F&i`cHeEMkDk`(Q<&J!k&6NTDDmaA;G
zwZLEGmucE4vKkA84he>AU}AmNkFPM$OQ&1_)JAYt*iv-_PQ*{Av5C~GKKoMtQ|DEF
z8y=q+F2~x%A9|@ZOR!NHtk+^A`K~*7XfflX^uqk+edYAzy1J_UdEWy=&DP6Leb~(>
zqJpD1%=?QKoys-a17d<JbQ}4|$vwdsjLly6!WAq`8IC@4)fL4BziiCYtC)7)8QKYh
zxeKM<+fzm7yM)?l$&wMz8Y=yo?O6bLF&(Z>OZoN<P9(_@YxfGj5xa#sY`Soks+8Og
z^Q!LiKa?)HV^6-7nyak6)jt^%hYiWEPU$5+tSWx_&u{3?jX3KvCh+1>kOUF3(}po{
z336M>fm0;GY4h60FD*36p%M>9SD>{5BS^Q<pVqA^XrH$J>VB!8UMj+c6o}ivnKg4V
zQ4@G4!*nok<^u5ei{JjBdW?$Fgt6j4mU9?15&&tyf~KJHlHa(428CVg-1gDwL{E+%
zWaX|}uQdSTYHVyg5{?HwkkqRRp$DS}z+$p7ISnXu+tcTEJ9d${r{x{>|F{4qS1jA!
zc1eE4IN4q8-~2hk=z_9bs@<7TynB0kaIDYgVf&dkAl7Q7hTgN{hKAG-^r11vEl2fz
zL*s;x$*`|HN0lS{n3?Jx{)OG2P<7`Y=3PC#hYAXkT4}J)v+;HPupKyWllbNZP&D}t
z%3JuJGRGrRiLZcNO9P>`uMfPMP~S_ggtj)fuy1NeQcsPYK(Jf6YMq-f8K!mFC?h}m
zO^;d#C4wWWNOOY_qwGE<xBkw4RUMO$1U$c%#VeCj^AgA7I1E6(^%<#j-8ff|f+E6Q
z6O5+TFAkTWGAOp4p#f&4FJCnXH1wDwxY_eEE(CDIg+a-~H>IxO_Msx{Z)stQzD||T
zm!-jkvkl{qb>4Ne(Cfi6_R^sU6akGy&2sh0-8gCKfQfvkrMv1>IKEZ$^uRKHYBU&6
zu^7jH)nBP~&Pphe-%{U8^WMLHs>>Yosn7sKQV7dQzX^{F)lhBC%gsm#XodkPVlbP>
zD7p%Uh4x?&ugQCIizA22eQ9~?FD{suq}=fiE;^Vj_*hGl_nK&Jd4jODZ|Ha5&=l^$
zeBQ4)s|5zN*vL>~+SZ`3;9Jg^=j*Q1TQ4gVzK29s7!jqU>VmeRd&veZMg!uTwzk{%
z@a?4@kH1_cmEmWOZ~@v`YKFcZ4kc4W``osNK*BrYGrjc^uc<AE%c3%VOhB9t2Y^vt
z=W}N2h^n_82N#2c`_H0-V|M!TwG9U4fGGZallIl;A|0KFx6uj^UG2(?fbVTmeqVfc
zBDr3W;ao%GB@1re-SBkt5FqZBsnr}#*W4JnKsPMyI$b^xP&*w-p2gBK_4b;y;;&}0
zvMM}PaU<i{KmQakcxb<)Cuu5C4>Y?eorTkN<!cej(cwR@28sZ*v|?cHykqy)Or!cc
z)=mX3Y|dAZezQ@+-OX<EoGJ6#puoV>nm`yTBxy~<h{%1n_IvMc_@DzAhR&~lr6@6*
zo1V|D<#_jx0(5K*sc3Xkxq%-v9*i3Qj(&7&zU_wVsY!jLB#vqOoz-y?&Jx(l=%n^r
z$anG%kG*+sDoF=Kaa&(j*k{+)N1Yd8Q$Z{!H-7{F5qYZJatK8c*&`z(qtLwZLkccM
zbxGUq(+CK40IIET-q9F3Fb+95G&MBLKs*Gr%szfQknFG4Mff85R!bjYw#P?C@j9Yx
zvL|E(Xv^eo^=W?%l%rQANA{DGQ&(12a#YISKU}c{aToW87lOnvnwF#ZM*Qt>4@<w<
zaJN|AT>kGf4%3m~nBbBAHk-Xd&${(sXjSw()bF@rYA7fJ`G(i7SGsxe7$$<<!3&9r
z+DBJDL}s>tV}7~)nqV{_@qDfw3x*C?!&+Io2BMzoc>vCE%)Plkc5(?^N8(ZXufA(;
zqI*-gJhzgbPdOEP=ImQP9kCM%o?ENqKhD8B7I^#ufd;t)SkrcOnsRb~dp-y*8~k-!
zZ_5yDcg-F9N%WY@W%E}{X3O6N7eTFVs4g4*@PutnuqUFF&dff%;c=?FcS+ph&>vlV
zr~T><Jr$iwq1M<*osQOq=U0i2Weio@VpZUbS3<?)QA(7~x;#}iVpFUNkvEKLKZLE0
zfO_Clpb`?IAg*KzhJ+d=HXk&^)f1$D8jXdnj#dIXY`6M$EZp9C<C`DyeDxNwD2No{
z+?^<Y03F-A0-fXCRW7T+CwP8nh@8ITK!3m4NWM|y_{(mlt;on_@k;%uhmWp2S5VmV
zY#mrf;{|z8Xkh62P|i;J{ZA<=$;f*pRp<5Tkn49Z!*0HZd@r}Pyz~zbxy<$opQ%!N
zq0!--JJ4Y295h6Yma_{4m9AfbbbJfnz?d0SFW`Xfc1ouOYZ7icfNx#WN+XqK*ANfh
zoBY<l8~anwfESTzY*abV&@rup`fC|uBwPXwi3ULIkUgT*Kv$3AIue?z@;dGBlFf1V
z8IkugaJOU2CU<zI+2T3}%=@hJ_4(~_Um-XA9c%$Ke8sJ5%@VABawjZA0SYvF<!9b1
z3#vi&L3+(De67^yYbPK<ATN-sBPqz0@)ezVH8Na$Vy3OSyquhxQseZ1{vJ9N)|XVT
z@>zLi?!C5aL|SzpQap#gQqSNsr>@p#HGzy(Z()}C(QS}DLqQ4wV92D-LiU>qW$~9|
za4S;RoTo9zWvITRR{Xd}>R(H$aimre7+#m!Z2)Auna1040b$qQ1J#H!Etx^%=}O1g
z_1_8atP1J!G9KVRoXVNo8896`I<2QA7xFwRO8cjn8$2w9+L4$U^f7$|EQRk{bCm+S
z5sUVX{y(o?BX8mEVnlEQ)NsZ8_+H)tIPaSuPDDJj^N5Z%fd)EoMxR8Zdfv)W`2VBn
zDuAkNgZ06}KmiG9q+0}}OG1$D?(XhJMN+z3I#jw-=?+0U1f;tg?*8}AonhvW8S$Ji
z_T7E<3I7KpRN)qf>nv{ARI6Yb3GB}^&!*tAgRYD;-g*}}A7If8g0gHAL!_UX%NAyD
zk}6_GfzQIP73;OqAk+4Z{Z143Fxo(<5K!gX3Ua=&0VCN)vz#7OAMl;-d;U6OY7Ka{
zq2trP?wFMMv88uB65NBK0snF2c=+To#qR0{GAh{0Kfd=7+9a6S*vQcc@J^24)4(|j
zqaaq$4P(a<AJaPY81Rtlqa}j_8y;U*PYVI!>gulbd?SCp4{^qO$AiB6d^)ec>d5Fv
zDj)(STp?|-`-|kW&w{1Z<%ui?!C$_-d4F@=u!BneI|2O@KYk77{=sn$op+F!<>T1j
zzoyJ*^7Y&qZgKxoGBDiNX)^@^gziu)9Ih~yzoD&6jcu<GTf+PO&juWn0o%Hmf7I9A
z9UU3zY*&{4+W6kSgWl;)mP`v7;^kl9srHE7bf+@8*hxwC5A<8a#$Y~q{P@KW)lU8H
zt@&D91U%xfe|&I{4_@N9|D_rmY~^Q!CT=GH&lR%#fWNwb?uoH921JCYz6ztf>}_sG
zp2So*6}Eq@^rsZ&=f8AQ<*joD%oG)ILcoEC5&yE_3QR4)#(uDC5#b9*A5??q%?5~{
zN0U!#8^T-1lwTzl&x?n1@Y!9oln2H~cZQ1_=V|RX3gpO(hT|}sz9n9?g!RuVqZVik
zPBuqAQIKKa9NBjL^yaN|x>%<iUqbdq{HJ`&U$?VA0QGG|Vz-#2e(`ig`S*r~N`Y^E
z-?xr6oRyTkOt!CJ_-K<30cZjig7A~xFFa9a!JjBN{Co7haUP^$2Apwk1BU~+8<b>Z
zEFj9$O2tMFOX2Z3SHETt4hwn_(hlh)vt0FeeM{vw36saqw7GI}3k3~dMt1j(TmF_+
zF(@_E*Ixpiigfn3g2IGEJPD>mQ1SwWL+k@g4@^OXxLD=x*UGJ97r_1k&IJLl8#cmE
zaG@o~0<hdEbqzPEWK}h#TEpdwwFHj;7zFUJ^E?e-n=POIx1W`dxpBczFG$RP7eN%x
z-s#aEs(}RWEVxZA8?Mr;0_xp+^#=<J0~Ik_@7mSf1Ikt<iN@XeI`O^{&{gP{g66l;
z@(=7ouVK{=&D(#vA55$l9{pMbg!ufXZ*@Q(4GfIxt*aVlAlsM$7v<w%rLV^?P;Wu(
zZ&*5)j3OU1mor3m<m4Eo&T8%4=k-iaILcxSf@p*NQjnM7Ab_PXu{A0hp3b0Fx*(<a
z#os^aV*^3y1V*ShD+{Z&fqqOvo3y09ga~J;D)aT_(OE=7gLm&Mu>L=v3Kf-UVPze)
zvhwad@{>?0vYD<VexIE<LX`xuc>G=qT|wn-1u-!dZR96WO8BtTkBbwLx9<TaRZtHx
zpZ5x2hQO1{mBKU8n_Jmxr;8dl8>!+o8MvhG>J7!&5X&2Z@6deuQ*KdF|2a1oo8Ebn
z6zDy~gCCdf+qi6V9|%=aWE!jEZ6IJf@H`r14eM<MWe1PDpGcBKLke9tIWuCd+FZ<Y
z9(^(&v+!R{<%g!kg3^Lgl6?yFcfiyig7;SUc8<nuoo>w1{c4L79E((_C`*ZX-Ksop
zA(+Sk+AEND`1`z_tpn3suamq>=zVMFdZQ-`gm^DQaXGU~@?C#L?KbZrUiL-}kIIIB
z0ZE~&$%MMjf%U(Ym45A#t>Gn;)R|PR)NPnnp^G5x_I^2k0E5EKVk~UbmPDS8{2r}Y
zaoKE3e9}e(3MHUCmE;Y^f7iS{aEbx9$jpyC`+=}YW>9Zp3yJ0hV}@6$nm{sDFW1Aj
zqYa5pO=z2I@UqeNZVHmKkapey6&{?fcn9~uF1=^T-pQ!~8X<P;+5T@o?t@tO3tR%{
zCqIA((vycQ4y$T9s@^*YRwboHPH)_Y-;jfZH^wHPB48t~=EjMj*D3!`E|>H999VLZ
z!=;DVKmP5z{@1v@V9>!oA#VWsJ(L1tX58eeT{oNy9E9ncq{ITIeM|@^fO{bIMO}hu
z-*R^umVw;Pru-#3#fHuszm@ReDD(Ez63d2%!=IsEI<=9!S06MHLHMsN%c|mRtc{=J
z65_}2#X%DD9b5ngIOlo+vURa{*jZc<2VNw0LcC51^q~O8shGf6f=N2?uB&gc4=(t6
zd7$*p051^K2yy=ts<x4{Az()jf|6p*=Ji#d&bme=HMjG`L%z0o9DDmi<9?k3fK0(C
z0SCYG*+b6~i(9?SpI!MAptOcGWfy0MCJ#mqrbLF?@y)Ka&%I+!ml5D6hDvoAhDziQ
zxA=Qajd0WlNlK_AXadMVdx0t-i2`6Yx+4_WiqqwLFW|D7_W8}gO9G)2B>-rw{k;JS
z(|@Ad3QvZ8<;{%>*_imd2qmNCaQ=d2z8%WMK53|qPy7TCe<5cZCZd=(i=zO445XYg
zSr{-|{jLG96Xf0S9$bz50ToeNwdwM9y};Nu1DNWZP(lWK14W`5!J^b*QvI2eq2p~6
zuhsWyUKFne>%-12dp$Xj<JCiW`dJk8#R+Yc9jQikWnjOoq^igRpfOyHvPm-sZ2EY=
zE69n5TUlQ20IDaaRUB+gYzoTK@HCzm`v;Y<*=Yf+*Ng=5USe>f!b-;3xhN#st7ZBW
zUS!}@2JL>IUkJw$9r-mSPVW!QnC8N_;Oq_yPD#r0!aA;h!x&<qTrvf7LY>B)`5A_r
zU0Ww-YI=Hc*VByscWMdMOsq_Id;xY<v1z(tUNf~&t7476>OcGjwt^_?>(@~WVW28X
zok>CR?Oi^fe{}Wq`qT@E1VB3fkFXh6pXI}-rlve-oeEYxWM~gJx;u1VUmWic5ggRi
zWM}^}%F0!4T?T+$zHb-x{Kkt~@i8Kyl1ZDx#@I=!-<QMyYH1nAWJkKU&TXN<-3y3F
z_kJ%~RbH9KW;PR#OTguvz-4ABCT5l{kOdqMkN_k|cCa>U05hhwkOCG9v-R#MguLf4
zyg=|`kqNbwvhy=AJL(m?aaVJxc!t<>ZHN>?cG}*;Cd61}wSh=fZqnJw;epBF|464y
zc8{D66~H=;nwp-=T{q7SKN=`57VK8xK^{*Kc8_+UKg3<z1QB+!FllA^8)7!&9`JSt
zCHdH>Z8^tek9aVQRJ3$7wY3}+6y?Xq>UVw~z9@$A0~lrryavn&oQmJcY#s~nkwNoa
zkazb#9Cj%4v$2v=rD4HKYzco;!hLQpNzInDZTyX@_}*oJh(yGLC-Y3IXJ#5;))^$z
z_A-Bgg#eNnaLXlf<mtW@!t34M(Oh0;34`fN1PalF3qYaJlYz0d+1VyI2$`GLtS2Ut
z=*b}c9C}L()H1w~c3L{dK}Px`P7lef0pEviZk)Td+F-F&W2Tilqopm{K2zsl1A*xP
z0l~2q_bO$i>(84e5HCOiFfsy+7k^u4`~gVFE>Ug#%36wuKe*fQIin{P`zpE+FR*!e
zL?uD&4IDz4WI!~5^c6%@px}E7AWC%o6Q(gT#^#qC<lshSHC>@Dpxsnn-#auMZF7Gn
zozJKtd%E)r&f>xf0Y?05*mzHz`8`l}f_zreNDjIX18ZaT!M~oF@i-{{r<%pUXO2};
zf)S0C`N~9v16!g4TKV!-Mj%cCrF>)Kisj+*|1F5HSFC`|^#Flol#}X#s&!buyl5(u
zA&>Nun`##*8PEU~($a!budSn*kf)^hpS}KhLLD>bB&+5tCL&Q0^(oLPz}RnpcV+F*
z*qC}kbx@E*L{wDSYa1ay4y3=<L#<^82RmB4fd$$i;7pA=;bZE~ZB*jTfek-U-l5>a
zV*#cf$<?Os+Iq(!s00~78%qn~K}F4$?(Rm<6Vq8HDPYh5qfm{75`zpRz~Y~X_<#k*
z#AYtVBmPO*9Sj@B#}0S{!Wm!Dh)ML8lv!TR<^t`>ZSWW*XFI&MhT7f&0s<M%PZ+>E
zqH)zdCp-7^XwXYEFyD&Nkxht7g49iZC^`Pk(=V|PT^rQ-(kD<=l;faZBSm`*9M3JQ
zAIK-W3}g(1&rir1cv)_?JwR0gC?lMoKoEfEyMUSw(z(6fOIqiQKS%sCG=jh8I=rzs
z?3?-doOWKI!`hBD68aW4#(z7(ckdMGh1-QonPeq}-^sxbCOVp|FiqsE^{5@5>l>z$
z-AIm?r|ylryS;}Cr~_e>7K7gtdCQG|n0TEXeHO{G(rivDk82?#1^+1!ouckmBO}n!
z7kbBMysqb7@+r#6$syj0i@|ZX`({M!8Z)I`|LO3bXOFJWSot9S<!*UPMRhf!MuV?N
z)(_35gkQQ1PoB}y(-!6D|07VUTk4<IWlQGO+cLc*)KH}#1JwR;NbqpMIFdkD?Slf0
zPu&N{ho`6yr^+3km@lrxjdh;d>>uobnFlI*dbGlbNa*X_;k7FK@kfsyC-tXry=s0<
zOZkfSf{u1*_gDj=3JmR)rxwY{8h!B^dip&3UfueoV;X8js^-rS0E*b{2Q8qdbE=hI
zrOoV;dj40dP}tej_4!c4ArjiG<pzKF*MzOhzn?@xUc6kXO}gtzF%@6FZ)s{-^|wD+
z9UEsi^d7|J^u%%5p+fi$40CLHJVW#)vRZAeMTTJaB{G{e#Kg2AmhT1zhtbi#!p;2V
zb>w^ZSpdchjB5g{<fOQQ{Fn36IYrg#VBy`jI#>i*yu>F=Cl+@;b(5d4QT$v$yrg1J
zZ`U_7&u7r6@C^ufV?4Uo(LqV?Xx%v`h1g?Y%6TGzd2w_Y;Tln%(9d=WpKkO>(uo!8
zs4|JkD)>5#0flnJetX5gL*&ka%X$5k2V0UIY>`-TelP1PJ6oH&wvEzY+zQ}_76wOM
zR)zo@;{5*SE6*z!AQM7G1hk!h&8Ydzi?H|%^kv65eowz;H29@i4v?Bpn@DSKFB$dA
z*Um?QyPUO+lcun#9RE@E*E$O~GtYp}=Y(HBi{$VA$2)gmvD+c^^d>M=L|C<Jxnjg^
z++r327uqVn(f?*H1MXnW3w^UvVb!(b(9Pocs5*y_H+)|Rtig^-e{u1)lIq6d@4|vw
zmFtIGf4on>9zME8BfK2v4#Q>j-1rwn{+x-6jqf$bQFtH~EoqroZwEX*GgfaDX_X`u
z+s?cAxG(QE(<Dav)l3`}L_=S?FS@aC`{3a{PqlZINo6g>F-TGxa<&+W={oA_Wt_$L
z9k5m&6h(`VbYp<Yi=v<8ijA3>Ri3<r_m!zl-s1jQ7S%w8MG_>DsjDA834SxG<l*%&
z<i1zrK!S840&!qph8U9G{#1zFSp3x&5ljyemG{61^)Z5`U>zZ=rBy$BgE)~jTqKD;
z55Pka6BpB{=>y~3KNDTH3p^|p!Ear#u(7ANo$nzk2$N!>dbChR>D1_z;e3X@;jn;|
z6YJo#_0YO1lUd8es|Pf0oN87Kp)S8NhD-MqX`dq2QkM`(zR%c+{Iz9)(JJ9cU@rPk
z#5%ti8G-p&O+=lRu^H_mp#HRRcW$>UAvdFl%Vt3Gi2qsmU0)?^i5kY7I^ESVDVhRY
zoB2knzxxBCCr=(fJi48`$vP5s^YXfjfH122YV06M24$Xz@<~a?gOtaNWMuuFozrvD
zaGmboqc-ULBH7zDwQl=fTDb#2kAXh3KmZGPF(P&oN?s#i<OvE62KS=Xl@$vMr<!kH
zh*${_iZPfO@hSMc9|bcIz6lAY#>ywrtHo?Kn$t^0DgN}YyfQi!AX(x+l0<-aCfcW?
zFZ77FPe`PmvbKF3rST-@eJ`b?rmjxQKxR>NE)Lg3j!bf8FqMXubY`_z-t1_BQe?N4
z%$6es+PI-hTSp#Fil+$2%E}>`ina|5^`wnutk_67CXx)|4LrPz@=d2v93oUiT<;5)
zM_fXU#|uj)jM_~|zQjh$h=`P=+#fj~+|lGwh=`(6QY5Xkw3<Z(5i!z%f|Maa!Qd^I
z+0;#Vz0bS;j55T?eC*e7SnEQSrFyvjBPuDJ&BMOkt^4eFp~?yh0e>E#qK5CkZnb&T
zz0&(*R7pwFA9o<#q661xEDMLmjDPb>6OrQv3gYiR2jcbU_-6Bm2ft)}9wIz12pS!S
zuTmPGb<&ZL{EBQ1IKOD*Tl9>R&Kp^CM0_2_oeAj?F-ZDA1b@_(RmsS5+ux6U3v*6Z
zXMB8+pZAdQGqE)frk!rqFJe|uYzsO$Nxh#4=*mwR8)HL!W#JkPa=5}o`}$r&V&w(q
z_0{*V62ea;?ocw#aMflKdqqQEH+jrYUv?j|fA}TjSeXZ2kXAjyn7Qm=*rD=u;hB@w
z^B}NM!o_%zo)-C;bL1FsRnM7($e`|mXUJjQJ@4^RhFOz2r8E?l;)56OUHETom}VR>
z#3gCaZ+Dr}Fi;c}=N(eM{x?)QqIzp|Vx1*$u#P&6g^3n;^_;~c<8xTmN=<3jN?+30
z>Ug%pKXw6q0Qqu_3^Qc-&1J7ixc<uw@Lw~JkGc7|bZKl$jB~?_|1`Qgdw2~2CF>y{
zAD<zEv9mPZ&9A3XG#H($D_C1o9yOQlOA?MV@oP0mu@@LC{b_2N2L?kTpY~+7t=rqF
zI@+m#`EyF8CqPX4i=RyXol$JSYAX~}yi)^@e=fSq2kn4?Me69+rB9WN*|#6G%XsOW
zm}RMzTFRNa-g_*C>^nBbr;l-?CYi=j^tEK|Ham2s=d*$%*Y%E<4cx|S59bDVLf64f
zCPZHvr`5Fa=JY8M8!P*?YO$QW;@f@nTeX?P1Gk&Axi-5je$R^o?w;}S5hh$V$03nJ
zm$!K=;u3T>-$VU<JMSXc1cW0ouaQn!MNaP|N@D*Pq?<YrM0OFt?U0?2K#lln_COFE
z=syc~=r8TN@+1u|T!@x`>?00Lx_3EALxO?0cmMwDM=(+ksUQcm)}}Z=MtSV@;<<mO
zt7qCdA%BC-MobV2<r9SPyLTn^FrZvyx5l8o01tdiQ`4BJm<G=qLpQhO^~3nm^0mtl
z4or@HKAy$=L$#P0@wab%4_Q9k))I0y{|ywz_88Go$2T|6H!J+Me{k@Obtw6@^&6eM
z!58LZJJ0Qrp9P50=^=Dm=W}9>%O@dAkCc#4*(cT?{p*IQMW*{(DKdKUdzx6@$H6%p
zg|x+U_bBsnkw5$0|LoVF%xiF)MmUSBG;HbDZMevNs_yoleC>_aF)@2_4MhE&oSXpo
z%I|$4sG*=GAdprbiZ=YfB-+lbiD=L<15E_0MZXMzu);jN`t#y~f+sP3erc*b-sVm#
zpKxxjNp7~0E8me_v-^koRl=68h5w90d^_=<{^@*x#+5pJn`cmSyB2hrihN+tJd!6J
zXKp1Se$!1GyqklZk*la6Cz6r$o=J8_s<UgT0LM=!!)n8hXCdsvq$xu8q|c0lM_NSP
ze6fb9t4>QvtK9yF6fAZL%Gas;N(yQ}1>*#?rt;eT)3P<ym7O?Enc14*A)b#HeYPc#
zk@Y@oCwwiJ1%c?39(Xb1d%*`2Raf_WiC7iIFao`lSzLnn$k@h7Yt(1RPgat;yL;N&
zsch$cHew=>_(RPs!xS26dD47&XxvS|v)g<Eegpk0N^>>`Jk+4J+U1}l42>3BzHfA~
ztcCea4-*2`HfHLbmd|^~ne65>^_xW6J&zB(-FM?OeBQ@+e0xle<E}O{sBb&|#xd^r
zP&=?)-T~OQu=b6P8u=*1e=op0fphjKFzXu8)=Iu*Cx?4ZVGQbZ=o=hN2GL$%Y&X%j
z2yaB7ArRzk?L4!^4X#*waIOi;iwny4G$PNP+UUeC*XjLS9tltT_y+^8!zUn|s-@W^
zD<~{f^s+BAvxT&!MU{j4&Zl3R^;0dD9Th8aD;h+ouXBi5%xiMmpR1V_{4P6y)ovRV
zxt*I|TG$D8vJ;%+JUrQSFQorhU_S4IiIDVvs#R_L9|eiNV*m3?>X)`t=MhT_7!rM}
z{YiR{eW!iSTmw=iWlW-v@o2<?F8n*-SAL}dT%FHjF)%K!BkV_VghR}TSp3ESbqx`4
z4C7xV@tce}l185Ka<6t+4FXF5k-+B^RTa1L2cKxIOri4`_!^N?Ff9F?O9hANZz}Zg
z0xD~$H23HS70++x4eCHsUGw9I)y@{KkrCFQ_StUvd|F(j+X1*Xp<!UnmYJn~Xd;U4
zK0Bl7X+0#_u>UO|g?zd*<?z$2RlQdX<KbS~{{M3UI!Y(MLn;vj&!^e(dCp#2J4+KO
z+id<&-r?WsGit!VM4K>d?3n2C?e`BHl!HvebBkMK<mqZl0)A&jE34E^*3-W4VGmDm
z6{r~1-pFVwR(#E$80ke;iwsyZFIKM|-QRoU*k9E@Hd%WvK%%EfD=R4IS-<_@sZ%8&
z(J>W$H`Mr_x>F3;ydXmWv3Gpvb93;ND~torsp1VDUUC}jn9y*Tb-mKqJvu(RmB0J^
z+5X`14)5N;RobUfr>!+~=z#FYevlp>Ha|i+IoSCti({VOY<6ufhrSRo6wd186!m=3
zg@@SA#}Dzj%=X6edl848rJ*TL-=2d^RWi|(wCrCg-PdC?*B>Uo6iB@2UtL{gq<vLu
zcd0F)7r<%n7c{(q1?oJ*9y(fLSAL<m7o2K|UMJ4h>^#nvv*dJS*9;`*fb+hOmnLu`
zu+|?P8uH`#$FsvLDJgv)!n@(7bwdYwe&!0!`|!Vokuncg2cBa;w=}gZQ!R^V*XKXj
zZ%>hh$iN@U4C9I_7G`G0o@*F4OwS49Bjct@oMVS&udAv~e*PWml+pRh6cZlqGP{9g
zAR11NwpXfyvE$0uvO;?Mu&rJAssEb#_eoXFgS&`qa*CC~#Lt#*=;9b4Tm(Hcho1Ho
z)U1V!!SXp%Zjoo~X>3M|G%r*JK22M)XXlvs`+QMUjF>KQ{)5JwIyP4IHRAQ|Ks8G$
zkARBC$mM#--Dli?6k`q^2cun5GBWN~^=}5?qtek8Ly~D=dEdT)=#NrOO6r%YfT?Z{
zeEQP2J3#}%Qd{TWl<4B%EG(U^C@L(R8sBBT5POL>kji6!+T@}xf}NP|bFJRHZy>U0
zTO*^R4>wk%*s|7h|D&7n$B%acy-pkyx4xDd>r&sy-I&}zld@Kx&B%Wl-nxJzxv~)}
zf$sJ+R(yG3VSx`rknRIGE<pkj%t3ge(4|CzIgA5Uc&5~-j?cU{)psDaA>L+0R9mVC
zv@WR9W2gT5S@tYCkh~mMAwFR+Yj~}A{()A5{Zjik-vR~7!b0jhzBLIr;;ik4%!!i9
z>B*TwPyO4xr5--Itnut_S?Ca%=%Bw-h+)YVL~`cEflyVaSF823leWGAC(cqD`ufw;
zsHm7TIc04*2z*1KEH5Z1xc$Rw^xKuLcX>8EOXTuT<Gcn@+9anNOI~sl>ZR-Gc}iY=
zGD1|lyd#(LLF%vG&S_%%o~kNmhbP~1bB&9+$NwnCLWpWcbh3hodT@Yn=I}a{A(A_0
z(o#!h`j8syf8h>^c*dowX6uO_1gYB1Pv<IUmDR-1A!gP*0>fKKh%A#<Rx!rp7GpLb
zJX}Um_}<r&BhkdNGE&6oIQwfH6ab^BwR~u{rQ=^HD$64fZv^i9F8(G)_|BRBOq@MC
zKUaCBQQ_U+6E~(D?tz-l=d(<_Hy~hve_C(L5rOuid|poi7^jBJWBaj~%NWrL!|X2*
zcRmS@Ff&7|7)~AG-dg`x7Ew|@%b_>3wig}=r_7?~{+xS6S((1J_d@m~1N0Xhu_9#X
zL8g@_sJWS!UXUt`e7e11|N1~e(!wGcBF&KkQc)-|-eKt$3z4IlNV8~F=z_xB2Oo3M
zetjxL1<7a9)cwxyL~%)zV2KO`ldt8w(cZJWcf(tEC+Z6FYlL-VknxXO`8@Ss<2dvi
zZ<S5Y%*@yyR#6?b`hI5qY@n{Ca{Iz60-9rYW94^BoixI9A(^)|N6P>HR9g3oAD)*h
z$dByu?R7|kdz3uP56f6epe|>|$G=T-0EM%$p?7h9zMx+v1qFY})Y2WcWCj|VFj8~i
zhJgz~It07DNcSyezI|S@Pz1TOa3|aKu})=B44n##3L@j8;6?qAE+sAG=+qHL)NNco
z!dyo1j05_bNt@~%$1=)4f7naZ=pTRnGE-A?ci+c=yNAhcA3aj8^$SX#RAQh$S=_5M
z8!IC<d39$<4pbBvG7~5&0)9Dm$mcVQl1N!uS=-jSKN^-$^54Loo9Kv-hnn1Sdjm_e
z?Jx68(8DK4h{asJ@=4r8(_feMB7V1zP4-5+F*Y{v*sN!<2=RA44f!)W$Nvm-Ic-H0
z&s#kCCN-$j#0jTe6e+tDE&VI!`UUh`1{x~%`ECDz>8Ge)F1J0Ti>9A~FZLYj<s?Za
zM6bb+T9VZmT($^23Ypp(<?+!wh{Qgh{glIz)#5R$sU7#@u{6Re>YBR7AG6#3$UrJO
z^a&vvo)|(_8G?FhGMKXB`mgQ5bSmWb<b@FncLMw34d+bNalqP3Yb%DQ$K4nmT&a#~
zjrs*2Q?{H=J-X_$Uou*hIu4gczq)Ye!tU)uCFEwEb&SgVNPyTXug1W}u9$?Sn4Zbu
z{qh$BEJ8-0r3cM(8pj4TJ-Oq_Vr=<~nWUs7@RhUDnw39n&J29Pn4+Mh)jIpq%B22-
zAc@Un>EOVeoV19$5LF{AmXi3Xm^ht){PwOrhboLK^AkpAxZOl!|8`yH0c$d!XuCGg
z>KDqlf`T6(2;C&T`p<;@y_kLR&+C(EF&^^i?PsXOm#?(m6!ULnXTK~i%xuu>9gwhQ
zcb>E&GS3eu+IHU&7LeQTajMGG)U38v=QsL&_MzY#oMIYmKlpvt3qSwK1>h>gBo&>;
zH?U-{f+m_Wl(-%WBm5HfA36p+eP{^K>!yZpg@njVlo+YNV6*LCIm3#pca*>IhW1Cj
zmb|R&Pw#;3(&*L8@_?V6of70|zE8;lQ1MyJIWogm!}NrnvZ)nwE-%*28GG2SA*;kk
zoBEE)s_=RonEtGfz*1&I^-4?U^G4h5+uMuq$H%vtD>-SKlsLHz4<qJ&ytLiCdR)`c
za4-^YI;v*RZPqdNL#@>IX6rg2bzIdZXLv0y6*Z$Eydvh`0mTX&<QmRf$H#~ByCrvP
ziF@BmXgPBx#>CY9a?XRtEFe{5HL3qUYYp4|EjzXn_yv&Jbautb!qnNt8rKU8W*07k
ziqh#%tqaap^E@S0JCGhO1(5ywL_3a@uqtD1ZzM$E!zTp=v<7T9yOXt&lnSEXRQP|7
zjg6V*V${>~<w>Ot&CLxAcCNy;H&Uux=9(rOFu`U>skc`P*t7=({)?G{G#4SrAbA%(
zPcp(8xvZ@!f9nR$f%UECD=ny0&y>GLWnTUAsxQ(!RW&g<?dm*jSvR20WA^_$gj&a?
zUgmY5%+S!gx=b_7NQIE!8TO;<%FeG4Tl3{!`Au`i^}NYdQ{Sj*xVC13iymF?(Y)fz
zC@H~Nf2*2jXNA@ryt0BVO%*xs=R%?vKV#kGhJO#yPLTuus%>8%l*awnEbz-0OoZnZ
zW^6)CR76ZvO47i!JM>D!WKsad)`qAY9z6Lp?~}mlbAIp#zc$>}zKj{pln1`4&jO&;
znDUw6Z2k%ulU>at_M%`ge`6dj)yXs%UQ<-k`sB^V#%kxbBuDn;o!{rOrd1x#tM8!w
z8cRjaE$<tS6Ek<uEl50<4iLgkoc~t%O^zqf&yNEkiNl`MH%N@};`ZQp`eRFGh>Epj
zVr7Xz(Ihtv`md;8MjZ2+<h5ob6QvE9yh0!tbviw(*~_u;HlU@7lK=fmyY@rEmm%A>
zb_OHAAan$e_r@s$)8zHQUBdyKJvQ$=Ki-hDV~xf5HPyaKPFl7b2}gR;n9#86jipO@
z1)*)<)-b1UDu7?q)MzhZ8Z`EYxk)HfkmnwWBD}r%{!Q<CuCz#@4u+>UK|t{^*-$|@
zMkmrwxKs{@kNJc>)9NSefBpUgqVu6k`IKL}eHvxv=SiXIe4^mD%357fHbGB9BJeyi
zzaS$w^Si3f{NEkBTPTQ|e>5a19sGhKYZ4~rH77h=`5L?iH@o&(OvQEeEy-y}KCJm!
z*UWpIUu<`FQU9HwMMk_?KR85vFV-6v6vDXrw^ih@2-1MEvw`*L`ObqpnktMeGLV>*
zuNgHQ(D{0gTXaNU3ZKzWu06K3>k7^g_y!U2I;wy_kz!s%Kga1u#%4(1wmFepb94eR
z2lEo~-VXFztN9Zl*6i9%&oHshgN%=JJH6hRYr)e)aMEqajA(0X?`)@n2^dyfkgq<G
z*;xJRjYS;|Vy4E6k(KvQL;qmoSR=QOS=7Izd6KNo_m^QN6LRS%h3V;G8DZhil}nd<
zf&WI2&it~Entw6)`hWh|vX~uH0YSKIY-~tx;U$a}@4rg3GB<9Tonzu9`&L+ox7{Lc
z8-Yy1<aw=4mMVbxBR9GO<DuWsULdp(04=4VdLa-jn<x{_i11D+0)FE^!8s77bEy9{
z;vDHE*`|lMx_`(f>l?=Y{@%Rdwa@P;@_$Dt*1FxFnD|%|JlEuLOTjfZc+2ouPiULn
zYr_!Xx=;I(a!12)Ykbw)2BEv+sIVjTW#cs;2@Mm?a*q?Lw)R}r$qA5ju|rr}CIFX9
zYdnif@a$dU%ZwCxN3ycwp&brg!hC=i$|Vjoyposdp9uJ!yu;4L8Fx+idwMqU@FHgg
zGKs%PgYaVJi<2MJWQ7ry^RqKbX`CmB_~Kd<uff{eJ+bw0=0|o+24q5xAr%$m2w~4K
z!B<o`0{BS*^9QbIctjuA(uO%&4Vn8H#$IU|UGI;?gQ_Gh4ja0Tw;6i}&Xy`~*sJ?A
zEIEN<Oj}fhf<q~VBSDIurQh_R;fWa3DUV;<%bRa2&OfO0b5ihfvd%95Ryumu+cRMF
zI+5CylN7v^U;jH`J?+RfmTW-|Pa8<6K3a_3gfk89KJ{M}K}`b0Z<ex%fv*!Evqz7h
z9Hd4^Yisx{TumbW!UF{pE@-AzoPML;QeIJUd_sE3%8Rk@b;fh7QQsq|saa8(-EY}3
zx)<05+a098OL>gt#tG7?JV@J$r!YT{ec$DTA<omYA+YnIkb;7Pvcj{ByoAz@4iOl)
zo!F*1FS7x0PfTylEGzuc$jd_KvETFThMY1a_Lw)I-4F{a-qB6iCL%7zG*5nWV?$$l
z%F|Ql;;iYPH*ZM$#q(=c7P{niV^PmAFoMH;jtNekl9Y55td03K6(u#~Of;;F%_U7u
z0+NaqHUBzmL#*H6H5Sy?Hf1nvwNmwk$8_sgzNwa_{?~{cZpXK{eyGA;R8up11QA`>
zUuAP8({#%#Q3v19&;&JHwX8rEU&ylK@6U&$on4I3vLW{Vo%eDYkGQ$H6)0{rOtM_h
zHwZPcFlygV7))oeyGlTMjkJHbA4<SyIl^}7-_Zd-<2IlAxfx_<7VHMUHKyWSvMy{^
zw+)j?f`?{j7}4<WDfi2w7YGPzz2rGe=5<g%^Hu(GxYFmjvKsDb(vZ%>p3d*lm%ji&
zze%a8E6+{CCM#Yl5g~qhbaL_c{4)j;HyEmUczD7+J=YyOroW!TxVkZD93K`ZVl2Z|
zWmBCxzGO0Sn#?Wxcli-jXXfH_2+=U3=>`o#RzqxKbtQSKba7#!nS~AN)9|-#H^7hH
z__vl&Y>G+(TRKu^^SX8Nwyyh#@!4HvMa4Tm17y;fzIj?}3ut(FfGbV>0P~=w|0b);
z(PNdOTI6Q8+``P5WR;)#rREl!5dp)!;~zBGHF|ryuc+zLxs3E7G;D)gr<r1h*O8$A
z<S1ZGTui)ZvLy7EZben)Y~yaD`!e&tToMMF;fc}FiBaJ^mXhyrcYV*VID6yiAshKo
z5DkeQv4y$C)8l}94+4Ihz6to$3I1fvE+;5Rk8>)2SV1J-kXTeeF5g4QRRirh*}>z-
zaKmzP$YJU6tiDRC*@?aQD+s_y*^71C_VhfjJgaxzhu4mvu2D5HuJ)&$uXQ^pHo+=N
zVLpD1%3lnlgKy6}7_a5B`^nY_;P}rh^9u{DZSnwU9$?$83y@XT9v&LX|4}fIIG)SR
z_fBMT(mLUdH;Q=JYnzikmEya&DE^4g%x;Su-$I=M8AaGGUu*}s6sOnP<}{|N!@#Mf
zMaMw#XAsE&t?|QKa*){5(W3C|8vCHVNZ)%^(^^_umy3NaIcZt4i1gdHIvnN(%%dA2
zdHW6ULPyd~S||T<=PNsb-Hb!qx#U^r`!~X<_~s=p)r7_bTj(4FI{-|XFxG%5ZF%Re
z4HPh02hQcT-d3TkWATfKbKcH{pE~oRZ`ki6PAy*4{YziGmzaKgmAKR?Q#*516JGTZ
zww@wix?q1`#a`5CH|@|G7bay1x#ff!Z2t}26S%9b(9>$Y(gNo}A{&4M+`q-nd=r$H
zU(6o)a2Q|QNk!gHMGm{f&xd^__3lr$v`QW2X2$&(hwbE;blAUF0ubZac{t3Kyj|=?
zY_H}`ucJ?db;!!fJ7QhEbXJ-r@{Aap5Pv?sA7Q;_l<K_L_iiBQk|a+7@w(HpF7z00
zmF2~wdo2ssC&Hw5$2~dd=~;6DUIeTyu7J=Y;P${&pHZ0Uw?%*bYi`yVbVWiLCSbDB
zI;BJ{fP~=Knj4!M!#<~YQ2QD78=O0n9|Rw{vlzn%jus^(3{RW!nZPB4QIDs@g-TsA
zjo0pIZy#_>9??(brI8sU8I60@WK}$ko<sc4pD!xM^+x<LTdGrHxAuU#rKqA%TBhl|
zv#2z%7y8tK0W3kU?tX+H#NVTRAsmV!%1u@%(+9^Mp<HL}<pc6j&%qB0Tqo*XmpkNE
z%OPD59q<lO?pJp)ygMquK`IfzbJBX8X939^-hDmG5*NyeuheG?1Swnncgm(cp`0X*
zM0H;Qf+(+`w9*h4T@0vIqI8<BRpkL>1M}ZjBf(m6x{=|;LU8DEPifOq=#D3n`_nHv
zumCYq>i@=WV^uYI*Y``sXl=A8_bMCgKm5gW?4Jq1v%9tm*dTyZcYL7^w(^M>8yihq
zyC8i3yR(*)&B1^yt#7Vp=bBhgmbx6hkB;tqw#DmuRd>mYuD1YxWZRNmv(e3DY3$jK
zi5B4xEoUDq>q$#hkF$w&c=^)Ln(p}moGT_C_>G#5lhCo|pYszrqy32Y*b><KYQe}v
zcU$bpa#0+NrBL08(~?z6Canjz)J9Hrs<@qd{tKy5VN4AQy*&AON#h)n?&F*8gFp%o
zcfEE0?N_QmCG+rluY2Jq6QNcABuif7BNXGIb;N6PzkL9Gd3cC3yYW8@Ld&iP3$Ph{
z*Vy)`w(L>pSYRDJX$~-4e$#36YTTyfJTi|?lcS!Zq^7z^+^luyqa~5>*SOdi<4oSK
zn9^^vzPAkW>mE8IBLEklt(=WxIdu-7Ft*TAly?WR5GLc`V5f<J)t|vJ8=E}+SA$?M
zlTlQhSMG9J1?OjA7j=cXf|4AUvnlb4urZgLpzM2Bz|evA4*eLJMfIdR5VQ9miOrWg
zgcCVVEV;(~+6vs@vNOEBX&pcgOB1GeFCMr%Uru-_PeqwmHn~+0O$E!`V^L!_&dDXw
z6CK+|?~N2XM8tdH&C0m0NUQbC`{`Nw&E67M7Yug&Jw{ogNpOogo$Fgi0*FSWsLq?+
z;$xzmd)l3fCcnU#-sFw{MWhAXcTKzpy|3BMj;D3W_h98wZX$ITfjTIfLXJY6|4v1;
zcj<!SRr-4avQ}u<dAv^+C!hT1*(-y?Pf1O_b;>z*&43^3UsrgnE^*|ucZqseV~12x
z$z>uF?euwBI1t40zZK3jPNKUrFfeqlg%Fw}bW0~b{L0Wu%FZm?IMFKBuwIq$iSfCP
zJG)xO?>pa6S-55EGR^;m&%X=VRx&!7*LOeWH_C>P;3h~?KJL*ipBQzfU9sylWKPyd
ziu~Et(!UX-CMY=5F(s}h*OV$Ai}pga_Br=lSpN>uaP%=R6*MPMKx9011`^mfS?KUf
z&Fwy15*7<xx(p^+Uog;6kRP>*h9v~{6o#28n^02{!G0WzU-TL)NZ}$2rS|IRWUNs@
zTs-RSOWjNUnHLcR{VGA9^+8ojI$cthndw)<kt^LI`0;M8fV`Z%a3nnfj!g&}##~-T
zf;JogpWj^T-`7yMej5r1e3+VYch5WV9voOMGwpmyN!gQ_h^lw;-OQ2eu7<Fkpn{gw
z^;B4phwOU}X6veo3L-9NE)phOhBSENICdsWM}H^vI`9j760y7cH5aS6ubk$u)x0Ra
zBdeMYisBnI!tZx-$SIi1nuPKs^soZ(Y){Sptu95!#FUg)sh8X1a(!%F0z4CDD7ctx
zm|%W4@D?jSH&ann^CO{*)aLsxu(gI-HCwFxmcQH38<XOfad{&VKezi&lE3Xse%Ue5
zGDe&dC)~AX1zy%{!>oX>Ak*mhe~`7B+>M7cQS`5ufA`0f<JJ@BiO<gt4iC4ds@%F;
z+)x|RGsC%8zr=-Gyh;P88x}Ai9swu=4-YRdQTnkr@!ZUeyqXfPMV*=3JBnQCl+=gE
z*S;=V^;Id~QN~1j=x#RXgTxB-i1~TF)*@4r{~KLrcBM|~rCBR0yMDIam&EQJ->p3V
z$eBURCnh15go5ur)sBV0{AUEpvk?L0Y!p|Xw6`y*X=uN`iTiXW`rit-ht{V$)OeK*
zKW5(#!?U3QVY~BB?m;vO^^jhzEw9hVZbN$}LtV(E7}pwv)F4*&>WJ<uRc1>W7k2Dq
z3i_E}Bpcgq(N8Uft7yr|*Ea0SKyqAIC*mt8C#SqM=&18vb`K8E!orP%9$EpU;4=A6
zEllRg`SB<9l8V9;@K330DgUR-Jnd?j&%#enk6TpV!^3y&)%e=^t(KLP&7cNk-9x_8
z^F@8rXj}wPDstMRJ|CKZ$Hbjv23pd-?pO<bo}<YuZkVgya7l>^3d)!1(7cxX62RBs
z!OPz4D`3*q(_>p!v@@qy-#=F8(G^}5*RgcDLy^NTEbO;zXf$DY0w9&E>-)@1QaBRS
zIR?LkR~u-SsaJZQN8L9~SX^DLw`-^1rUIsxMgBLo>WU%LH<;LM_b*Q^tLOE4=X*u_
z567?ulsjy9$FQ*LHxv4LTOmP`$W@c#d0<P6Ag}|T7uTm*SXod!=UVw@8m<R5`G&<E
z3fz37TWv{xO+oEDCAA<H%iz92pOd!8!i}VZVDZO^I)I^PMeiNdbkLk*6i+93<2_|{
z-M)lBP*M3KRFIO6jz+fyisD?oQx4=)mzK)O3ldm(5j}rohDWR8E4VJVU-wp7+5eCk
zysqevR94DhEs`XiCaW|5wJ7&v=Tc`l;XBeBEEWB)#jCrQB{kVG{SVOfNYI~E!kYgm
zG@U(I5*;1A<Bd6-J4+p5J>C5VuM>L2ALG=n<fBvGG-<0Y07`Rmdb~kIJbGXFj^6H$
z72BAFxKGT&BUhS7kA(5s3F8MiT{sW+2Asaj#U%_TvwMh0Yx5BWjLWLD8?Ke|MMi}a
z@*cqb1CAU-+`xSvk)(pHj=x)!`;4toX+le?nLpFt{WC3R_{}O1<-NCmKugnOfRStv
za59YR_+g-j)9VUt_2&eBkv3N=r*552ePCKh=aR?hJ!T)#JZyCAlN#@W%*?W!%22_T
zU56oAW%)I!N@~Nh>2?-ipuz52ND3>w^U{6Z%E~G>KK`!nvIO=(G7AVf{Q5VzY>|F;
zEwtXbY6uCI*mq3{3@mwX()>D=M{n_r3D%1M*S!;qEO$`T07ek~%5JCQi&h{4ju%z^
zU6B28_tH5tCp|m6Ej2m*Nr?E_xof0T>sy`pZo{7L9#gYI#`X4INX8-NFY}pAirwDc
z2EE@KJgewqb4%qa%l(h{9ZZ)0EiWPR;C`|0XWn<{Gfp@-aJ>3Q!1DCyIE|-4>mo~1
zQxfGJN<VAzz~EpwN<I7RfX`qzJbD}Ev?)te&jO}Pokr?d>A+gW^*y<hlafAv0(*OV
z8&8oAZE^nhqJrh+Wx1I6T~v(FjGKs^$G%PIn9S!_W_dk@8^OWifFe=6yFY+R_gS*Z
zeI4aO{_yZn%t?e32cE_Af8KIp1q~_m7w-DiIBT1Y2VSb0np3~0#8*~Tl;!^#VqV9A
z_i}SgxFv84V0qZr-Jdmr3m<GkT}(>pJaV;vaZw&7x%+Nvz|HjDqJ}jN8N=t+1-O+!
zpBE)X|D&j+uCC5rq!T(9A0EE@pN(l{rO|RMIjNk8`eQF;#LxAwcSTOVAbA;Ce*f`Z
znW6J&Y{0pda`TU>kETtfeNMYR`NWOT0AeWp{_M}aYRfmr$4?U^e8jpDQaBHJUyYi*
zxd7;?o*;~6^aefeb|Y(gYO0R7`F5?Fh2IBGeWomb{u&3X)45iFe44d+U}Lp4=c1}D
z-O)B-I_9VVo0%VM)<V5b1C4j?d>BGse{9+J>hN&xK7L5#5|4C7L~#oLGO)0q*Jfb&
z;o)MlbY%<e%vjpoPQh;P(@o{9R5yCrk57KHl+%w}jg9@r(bKQ6DnCw;YaL8!{>Ynk
zdHe=1)Zs?><zFU7uYp9i55B9?^Aqowyn~XTdJw3y|6V_Wakx|3fe}&SO<`ZZIc+zD
z4BWx6q5<PmQy&xN#1wiLVbW6mdimMA`Y>Qvb59Qrm%=Bxiy*X;0=Mp!mAFfAm4B;s
zSJbBs=i~GgM`+#xF&Xfq4iEP$kJBPQKsJmW9~}*ewLXs{@E~DfU9D*|JuxAiH)62#
z+}eI<K6BP=RQUSz#^^P!CEMi*FXMotF12w)|Bl<Yv|qaV1JNm8N7h0^rBbYd#pGop
zmLJRfKNmo?VZgY)E~)3C0a@>|M6PsNAdZGN7WF(q4^v+vk{5K}u=NCwErD?E<;4xV
z>)$}ti$Jc7F%%trMHK!51Iz56_$DBX)D$#Kc2{cv;xnc>NlQ!h_A~lgr#=ktG&sR)
zqaL0bt3>D2_;9F=9kL7?yX|^y$_^0_@_8TZRsHd|qL#alY<oRP<VlEDF=w8W`EkZ^
zq#GtMbMs9E;t{lsm1$%r<**h-VDUYCMg58apGALdo-yD_O*YR@O7@4OEncf83w{+b
z*++wsKd6N<g-(x?zf@ovU#{_!Q`ArKfgGV`LgxljPmwOWTZI;R{@hQ=3l!Fmr?cbG
zjOX=40Ry&bNnBSRqGiMd1=_=Vuua<OtgVQTi1?Zo!A40NXO!4CILMio9nqaaNd$RX
zwTmY#$<JN5wc?d`4mljlE#S?xE0SEqh|10(_hD5)c^0r#X2C#1GRONMpxzsjDCg9P
z-7{T(F4(4AjJ%`rxEm>?;^7C~MO;MwW$e8>8P(O4&mR`jit`ff5~cL^wzl@laWFF4
z_)Jis`F;Lehe9|r?4`T@8?B+CA#n5)SWGAgYNCXokvtDdacbcmjEVUA(q8Ri{p*m-
z<N3kr%>(U}!EFT$TMo_JkxVDz0*+tsz5iLp#h@XdNgQ@xU1<uIdyKe|Moui*ie^MG
zX-9cyUNWS#K1obZKNL`PbdDGJ&iz`6>W1e+fN^8gC}QhCLHLXj0}CqVnahyO_dE6B
z>AZy)E`Mz~)Nt4NmWDWgU1vQ&!0;*GsBux}Q-JjL{d=@!c0cQlsp<E35HN3$1>jql
z6}A?C?Id~lhiXO0uXO><?dhFWlW+BpgF=FwiX&b%CSYiZqoI9$_%>jfiG;=|uayK=
z6%Kh}Z0yy-?cNlB_hGiU`u*gD*e4`Bw=S3M<?DOb$LoPi2pAa6P4ktNm6ZVeP*fBT
z3@StJ(df8u$yR@PATOz?s3VhCS;=r9@ufG%0-*Y=gpM!`-`U%z1h6{<H0w9<@{P^h
z_JqS*F*zhDu60MQT7^fz0`a?dUoNq6m3-s8=KGflUqt)v@pN3AGUQqsn*LO~Qy;Hn
z)3$2L$2a%x$x2{Ii1$_qHD*9-D4w1m`Ey8<ZEQ$StzYwxdS)-6?v5>T4P@$X-+p|?
z{`UqiH6Q?)vnBHUJQnj`ZL$_P7{Xr~cfU2C_&nCuH#$DL-T565S7ie|%-#JPQY<wz
z@a3tz`R0&JN+f!_TFB8c(S~Fdq}~~JG;SSXW5t5t@}>A4_#}@EM31Qs72heb9w1tN
z8kY}hH$0#0meHOcl2ubmjh#gol-JVwbljODO?gF9HwD~ZgV^sm1=<?ZqEYX$jNVba
zJBK#N<(J+w<P=p)Hc|s|sb0{f|0pUH!J;1Hv|4Ox>^FAVrg!8bqvIA8lOu{%P*cnK
zP92VY-iQ^Ol%{8xA8z&IyO!05=h#4>3k-aC7!jCHO%T94?V6B*_#Bn2@^|U-U@fRT
z#ef>9B^^%R9`uQfo@+z1_PQ7Svm{t|a1c~fRYANY3j2=gdsKEd^E`PLuAmAJVQ+zq
zj1Srx8Xk7t;bmpP=#;H!tRm(jKy4*u<7H(Ym66yjF?X6ct8?1^*rLBR-O=4hK}DC9
zQKUXUgo8(9chpu3=iaCEPg2m(!kSTs<f1XFHV1TOfP3~O8L(Zu<xyGb>Df6rguv@*
z9+;a;sb)sgXSWsqpijV7z)Q#=UJ>&XPTTEEwjVnEyn9w?@E<Q9>>~^04Ey=rgXZIG
z<l|NJYvxg%Dpvt3tJyj!pm}p292{^EPW2i!FW4;dUq>FIx>nn!O?qE$L!moztibML
zovu86P1=3?U;Sw#rzK+u>Re)Ah(kb%DCI%Oy4|ggs_NQSH`$zwHdu#4>-iMF@b=3m
ziJY%ryWD@|ltn>A>OfylScuHOgClqZ!tuEeA_orv5C~f8gt+AEGr4DoM`AR<11PRm
zhlYvGq0UUoitqq!vp=i8dTDtX1U`Qt6g?y?B%Q@VU0Wa{4W!t)T*1t1{q-N(_0OC3
z=1comZiLeLvic`2Pw;w;0A@x(ahD;Dj7u5}*`muD0`|1ER->*_Ee|)l4GjA|h31YD
z*4libBGdI28kU|dz$*Ywy_{uCT<*ogXubcCi?B^gvRhvl5QY<zWdVB{e%(JKBe*6P
zD6%aJUu$#PhxGz{esW)v)RmR}rrw~s(U4bm9uUjB3q&{x@xWs3+O4S{+&PVT;u*4C
zPgeUMfaZ|G+uNI5-cwK>_~k;x^S!Gw*zqyg;2NR+nu?npb@=RfOoYY^7p}*uG{HFm
z;kT@ueO{_+emP~ec?VdMeIj)!QGW91x~o7PD^`hGOC91oXu#gB-hTfG2rV=;-LPQR
znzgn+b%PPKyh7FeHd01Q>l%3SmTdU#Z|>h~)TyxD`E^|{)Jj&t$_l>{TweC;>fHaD
z+t5<eGpF+h{L+O?gQ+_AnAXj()hvU6Q;`4Mp2_h|TPDS1x8*X_XSm!oGCj*P3VA`m
z-N49k^a;q&DQY99ggg&m$ICC+!bikkHs5=R%g>)4yL_NxGq?N#^N#O%!{nVicObIZ
z65HcuCNTxi;<c~~&^>@4bYcBzZvR+AA6yTX5A$F-u)4f3X>4&#`5Ncl#OzUxPn%tM
zM1-z`!)CD23cHPmeMT$7_v$0*C1KTRgD2weNc>tq{&%!cQ+E+%efh!>fdC*U($IDd
zCbKx$n2fsHFu*&eSL&$e<n%OBIEC4c;H~Hq5Vm*b%V$S6?cCnwgvI5?#QdvHugwR4
zbZSg{exc~J#yua`Q+EM9CXt&S!_pnTy*VQjA(Chfbq(l);#Cy&!vgN#e`lfms$AV@
zcxsB4n%efX>2$u}{MHEVp<m&Kx*BeGv(6~Igts(d1QUmO@BVuunF0lD?=qr>Pk$cq
zQ4jXP;UKX7)_(zJRkzf|FP}dGk+qk<=M5p6qLR{&@82&R)_cco(7SKv?<GHdN09>r
zZvxSY&+l?lhjBR`e2PqnnUc<_s;)+EL<N@K@9_~U3mbmB6Mgp8tF#`IQh<^4CO;y<
zHWHPE`6b#OEQjZxM?bp9^TD_Hdyz=nQ+hT!7P|T;`=VZ6t$BlTK-`*A$0H{lJ}%_W
z%aR4iokmQIM(hz+Xj)X<uUX5Oh!}YV1@@VV_wUp)Aguy+bQO)73L5cgkp+3*q%vfS
z>gRv__;H<<s-ZIvpilGaS0B~xg7xifMcUA$gp?lrQi#rlPgSn=J*D9)X^IOwBsMNm
zNk!+Df>K&RNK0LOF)YtT_I*M`1llviXNj*Cc<$nSO7lZCh6^*WzdOZ2`Ey^7nD5c!
zhbXM0@Rd=*&lYF96?=@!0l_KRJ{_xKBRBW*5}F8j&#PqxLX4v?-|5(`vyQO070qYH
z?<rwX;F{6E@{_X`eEunnma%}&0R!y4J54p&75O#yD&>ro*<!^*LN7adI}I$~R20^&
z++<PpOZ6A5&oJ9A?VW9<G<-cJhPD(A2G|V<%~V*jac>!rKSKuUQ@7y_-T3TMt^Df2
z!5$1i&mW$3<mUbxbtIQpQv=^LDNs5H8&-U1DgrM}Pr(DH)B3=cxPJenDfK?HkL|`j
zg9GmhCWQp~$HvF2s}+bZ&dl=*o{P1~q-WKRyQ#Ug|9<yG_Vdy2*a?6^X1MQ%ig@5-
zl2TLW7v(;-Jn5fK^Qx0FGFn|(e(~is6F5{A*qCy-?K~IHl*}6bIPU}8I2N4DPTe6O
zJLwZJS#56vfE2j*x!>5K2h0xsD{{*aXRIkAg55(br+Gleb@DCa7aFTl&cB$7Cq?7j
zJNK#Sio1FlxANa&!+J8SpdfKT*U!(JJo+5QPaKpESUr~9s{j=O9)2<qZGwUydbsm?
z&v)BitvkGVh75+7KyKcjkDB3>9~?;K6_S1XbGu`k?@!tjfL-b%UhB5X3VA)UD+P)+
z@wua8@M6^OiQ;*b<AA`R;4>Z%^yep$9&(}m{Qv>&er#1&*8m;xRKX}*A0S0R{ST1Q
z*>g?leoc0V$ftW7J8mMa-?+M}ijp$NnQ1IL*XT`kMc8Xx|GPEro7NH*7WU>3qPyv7
zt|JptxMp1QA$1{EoY&!oSzI?j57^nj9bHs<>oeC_T3~iDD%QlP<M11imjej#01Gag
zqP8>}K&6eyKJEF~;24381>~u{!+nS@GgxoDd3bhqMjy|(VvqExTT;CL!i2@;N<p_A
z${;3Z;*I~U=ios5_h_Iov8;s@xcuLN4dnqvAGT$JB)}^Y6B9?f5TNxVIoV{lJmxak
zFdR_*G3T4iY$UF#s+vBI<ol<?N)^%m@&*;?Q4IXVw?XAEZYy5iVyME(!&!Q6+y)eC
z!%KA*y0-a)EX{no>h)9lGi#AJ{WmXf&}Zu0e~k~OkC3hV5ZuA+386|0NRe*$mRcJ;
zh7ZME;xZo0np{(YcWpvKN=mTzF_}z;JNDvdXFBUk|8*ZE`KZN^73s~@b|9?kwdrYC
zjTQehPr>`&8h3kFw^WUuEhO+(2Kh03p-*S&AAuWXIlF4Vu?FOo%LXIchnG5SZTkjs
z^eK7S*(C$Um+5iKKKJ|k`>9e0$oM-U+zxsU=@YeAt{a!;9iucrlC;f@CTN>SLPbUW
z@b=-8+oQbu=cszgKb&Cac7tOIwx4iNvzz;M|92UD-V9w4zm<|Ag!TGu;5-#@LLfPG
z%|}jdL1I<<$W;xRpdByKvolwly42fQ(wuA~P20wz++6#%;Tsv1L7Prc3c^-OgS}XE
zxb0SLV%yo$+?>y4<9}$n%BZT^Zo5&i2$hypl#ng~Nok~!Zl$D=lm<mWKuV=c8l<~J
zLb|)Vk?y?f{l>U|3>@Ql&OT49HP?&`rE42)%=zHi@RrMr$UHOp2pB@`q#LfGGzU_0
z{NWKo7wA{B)49=;$b#kjceA-$b~j2<&f<>Rg|zJKu~Eyp+>;va0oJ!(-+wh^iSm1B
zgx9^*FY_IbaytYrBJ{lp;djhS#}qT!NA{`R!?3Q&1wraIwhy`4xN%rpm*=U6Uw^L?
z-4>^2^>&>)G`AW-MfKrznG-PF`;XIm?M`_Sb8_(VENJB^&O)L?De+Y?eT^PSbn3od
zd={FC$XmeqE|1ai_zJDsotT6m0gjIT$$r>9(>=Vsu&t_&NmY~;wtAYqR_`0Lj{~<6
zE~@JXK+%I-9{oD^jh<%UtATdaYO12gyVc}D9iAg^ZCa_$o22b;%-8hm#6+e4Y*3M+
zM`_{sQovV#m^=_2)?p+a)43!Dkng8jOGR5-iB)kw%S>Bl=EzQP#F~j{{Eh=sg0TuG
z4}ovISDO8wlBxMDwbp|N_O=^=Ok2jgb-TuWd_n!)yC<(-?_Ry1J5_A$qBAx&PZ?O7
zWa9a7V4z_A{4BxOHzoJCo<&Kaz;i1glC_s`u%$=kY8EARJiqm$!hFalHDEbU&Gf2D
z7d#P+hPWqS=pjeQdG=9Y<0G_&Fh0MCD13e$laFqHr_)0~;oT-}%@RA6jN(E17uG`D
zM=03dp6A{>U#@Q&8x~Wkq!ssSTopeS5Y$zT5li>|;%;tgX_@XYwf%D2nIZ4j*MR49
zA0P8@Li_{k?da_iIcCCdso}#g@-JRBiJf~A`i!z!RMef|oE~9*^bJHbTT!L02JY|y
z-Hi@5aGXxjN+*uvdF<^n{)Bj}GeBho^kVLYCSwrbYReXknZtm%?)?hPlw$3;JxM(X
zNSxdL4R@s9=i0vs95}cK+S;##8nQjey*Q$;MR6l8Xs%o;`fsx!pb`vMRl4H&h^hDB
z5;aUxG-do!&}>64bMp4?wpW|pB~9MNu?4#{6<gh}kXMhtd~qu-R>LMeF98U&&A_3J
zuE&5CbX(Lel`v%ir7y&a)rb-lj+tw1KU)VSl2j`v$cQg4xVAo8JK@K40U5jwSLbSO
zU}ZTRUxjIn>bCsjV;2`XW?qWIiqsLbQE0a+1kk+??RPuZci$$gPM>La0~w#u(8lU=
ziv|v8UHOd1Ow{zx)Kqk$4*6E3?6UKY`MTRa2Di4J>1gdGi=Q2(EDi!b$=JDdQ?Y?;
zcg@z$Rth~i!hW_cjqN6XiQZTeUI#wrtJ1AUWamxDPv+cBVq`LQ?l)cT)MfT@HP6iq
zH4sp?9G1DU^hViH#E&j$YuzImeU7}%CqyMl>=@Jir$y+T<+Sy0G-mujEqW?E@IMuF
z1@Le-s+gl-$YL*deG${Uayz4i1#{m>bkJU#adkHrW<-FTPIvYK%*mJ@Ti$vqwIJ=$
zyO7TTRM}~pc)h$y54q{Dtr_rWmIYWQONvX4-mmaGUC~mo-a~%kx{MO+0_(W{`vVU~
zHr{!*2)V^P$vrFB(ro{`E&gl!$ln&IiBcL0AKG8F4!}IC$aZgieH}Kevhq)6e{OL5
z>}@&T{Xx*hz|bHhguc=<oj+XIW1?lKeCP*jh>{AZES4JAKb`}!H(j|utNlSuSU>rf
zXFdCXAM|EFPESqPAs;fZx|;WWY%o_G?=c^bs*ddNX)EMPU?Rrm=9Sjx(^a#~L~)GN
zbov%0HpAaCGwOl8O5Lr~*}*h8G|U=f(ter^Sn9^sF=+ZHCNiP)f<rx9v*`1=T~2Q8
zALsobC<&4LiJAn7XL~y#u4Hz-AuCG+l`~E|lvF`&2C&c>bqn8u2%LYz;kfoGl*8|3
zN!$rYmd_g!+Vpauxu>Ux!l-O(GT8sL{Vj8G3A=gr$Ty;QHxTHn#N&D>y01j>%@l6N
z2~34vc6Cn|6c=M$Zs&t(sX}AlH4q_}#=E4c_J@b|37BS12p=^Y6~$q3ODAwYLTV|g
zW@@(Mpo%pr5|1RbC{WK!O`J~O*%oTBbehb|{=K}g5XWj=4ZES$c;0ncNJy`p-=*q!
z?%^lFM3|(N#tbv)a1kIH6XT3JwGT<oU{E+#YyTAM8r|$Z@y3g^AJpvZ?A^MRmB8xa
ztb3&^*Yej=4m|!+ZM(r-q^pa;jKjQnVYhbrBg*XP>j=1eklBr6?o~Ig_ueF2&a79O
z*a+|5uG=B^0|aYp!jRRR=1oHk*|C>UgNuUDG+VcCa6DH`+>e&bFIxEy$=cQ?&ym=w
z)n$iE4n(WY6$F331F$6}zV>YRV&ABb+K<q$qXI^HM|T$_|2+wm<&0rD?7f;o<iR>@
z>Qbj^CF989H?W8+hv-;5$^e`p{_l(BA24Qz`9PL($9tLW)rT-uvz$D-Ai)z(X9Kx9
zXRhXdsJ`9$uYpWZ=d51K44d@=44$*WVey`suxzqhN>b8}tU#YxeM^-|;%k7Rf#FQ;
z1|C5Iv=bOUtCj`oR-^Zq_eS&XM&wOF%S!d&PO=yw@B*~8vhQad%tvI^hExA6Q$p$o
zsrMxuu@S9E8yml)1_!aKQs27{hKucwMW3J{ciKJbo*Vo6WN57JjjSq32#)%EbQsr*
z{?nmzquy(<g`l>QQZ;pV_H~C{XOArR^!pGcp5#k|L8mgSG9u$Z5t4Vce>eAk$YuCv
zfL(Q0?ZM&rp-AxLnd?2jYL%J2!7opRs)bg6nq3@b%}nkS-TUJ7CES#$+Spk#MEn7g
zfw7(_v2xoam`*cG)X_V_BqkxoRVz|8vdbuIvY&ePuAtN<{fzlny6ZIB><h=ODFvHG
zl*d^d6YDuLT4S7UuWRC{G|vXo;b~WC*ZucE?^^Fx-6tzh+l2egYQjA5+l{E1BCDB0
z|MTd2g96~AjazD&8D!i?_<Jtm`8^B^24iK@T_#;x-?9QTY%Sl6u9A+5ar*GLi-Pst
zPze&geuEiH9u^jbExnW5s8hSHqobo>TJQh27(_{yCF%)a3f0p0`jw~6@v<RB=$-#C
zobGfcK>NzIdHI+-L;<J@*eKb#pYY|2x5I?RUt8P;qoRa5*Zqs74Y0K*Ba2sOlLp=u
z9yad9$y&#f*%n=Rlx~&1s-~3fK;_W+eV!lB4UV%20J#R583>&{>yJz9?G0G3f4*KQ
z8&gBSd-p~?FhhVihs++H0JBRI;hHzLVh8(hU_;m1)GSOSrv0|LD#kg3>o+&QaBgLV
zXU(pD?TeIyW&G#r*!{js<5wU_fpXu*e3J9--CfYrspigDnwfdHUaKT9{s8)6=2{fc
z#eQDVfFzE`W$zBFZ^g`^GXaTzZWM?%Kx}z%-G{Rh9TU^sBy!VB=;N3LBO}<CHINg(
zFMk`_JYIQr3*mCTz+k$6A90+jfs7UuaN(Xmv*?ivRbc1VRuOXshAIgCZLOM87TJAo
z%FgR06}^|;2&GTT&uIGR)AGeS`Q?5A5Gl9$9J7B{-a=H=EE*Pefe#Oypg)Y@hJJ-2
zXvt{_O%Bwdvv$2u2xdNVls=aizKoFDvaP6pH!=(`&S82++=O}MIsFqytvcM3cIut5
zo;x_n0vZa}Pd3LY4jtw2%qm<9dTt)qNANSb^1iJ59KCM(DbLm5d=BHH<p~W<ePiRk
zR2d;gAExWlQ;I$Gx#SV8o?LB<-58y@`LP8F{q!T;oPTz`{|$Y$vG#U%@5$3lF~J7q
ztBojIVK}(M7f0sc?~b=`)Z;jRi2f*OP$_JRG2VUS_@rxJd}?N?Oh`+*xJ>Y36~(uv
z>Ysaya0&a+P2EfCq;ep8Y)%>ttQey5>L?XwRc*rHGe2zN@s)?TsPexANuJakkQTuj
zgj1eq{r9M~Zrup5A78(I4H%W2n%d5P_~1YkN<HSW3t;{ryn$$-y&cykr}dv5OTf^A
zYw~Dw@^?<^+{*RQN!vd1>hUZcEktcd#r0?bYC1B|$9SpWu$4qyxsf!1dF}ZN93HTE
z##L2ExDqPV)RZs~m~#CLFtgP81L`rLz*<K<T_gO_R{W8dh>(Qa@<0MyW<b=w<-}L(
z))P>*2sfz_TRPlw$Xl2j!ymh2hYQ+svaCkeKI{OZ1<3s-2*pxM4#84$Tj;F2^D#vC
zDyZ9!FYV#XL^xNDHwFeiU=RX$&*A(gG~DK#xqJ?ey6|_HZql5_l}FOBW+R?sotg0B
zHvJDE(01H>S&5fSS3!M9HdU_$NFa!}mWob6OW$)=T8B~jn+xxJXq=Zo6mVJl6abPS
z7l%!d^nMSYn_V5$*O8C~jq=Nc+FsdIGKXSL!X#os66Y0naWf#yD=Vo~=+ua&54@L-
z=^vhUKaJ*TFFxtNwt&v_Tk#^GvbFAcF)<!5@sl$vqPp~UcXyLshyflVYKZ6c2NmJS
zho?=;!zLR<$S<uddKUpd7kSZlRR_u9SgPQXk1$}`+eyAj@_Mkc_;Li`y~~s@>0Dv?
zO{bmzQD(S*?(`4wqs4O@-Y9%QNc<@2ec~)X)sKo5PsO(%{%bh@eKddg5WjNe@IMuZ
z8$QhwMxdwK4=h~4f-wEv{#5$ggP8k)Vl#6eXt)j<LFGdohXpvOe!5o3bN@eYH?4Vz
z4)|T$o2+?CEg{|FBfr}dsx4xKo#VyWKZlD4yH22GgOLSnLQ*bgY^iBZ=M+D9@&>FG
z|M9_do^pBr7xkVAQy_2qGgWv0Z?u;Rj7-!|pD?PNC7Z9rd%@a0;xWI^GfsdX!NtY>
z&g7Mg@LF&oUywLtj*~c8#2Ime?mV`m+00GFJ8ImH?lSD+nLDDE@?u8PBQ@+w(*B_`
zc$(f>eQ#){te|-J$5cU<z-w#Qr#a8FX(*DZ&`B#2&dM_~4Wjo9kR4eMQ@rx1KYF}z
z+fK&c8XFJNJ1$39)!k4Vw4}=2D8>!H&vR{8koo@QI$5dKzA$}3l-u)HtMK}Rybr~p
z-XCEzub~la%LyO5Y#XMQnNd=APtV3G89q))%X0B{uftE8ea|OCv{x1;*+qw}r0clE
zEWfxErZ;aKBA$(N?(7juoVXgbW6WZfO<bo6r<mTvUff24n!=m<?ep({%W5YUHLcm!
z?UA!Zb^5pe1`QEd1A)4Fd1-0aY1x-P24Fj$V-kNkiE)#%)BI$&*RL}hQK)|smethM
z0EezQ3yi^Zy9he}pj12KeEsel{`2Ys*N5!vYF9X#K(rz5?)TE;)Hy*fgO)Cc#1aDm
z@moePoNQ?SWwnA82IzNFvXBS3*eVL@FZ53GQL4gu-FDaAM#Y_+)TII_ftCJ}Z_V}c
zf~jH(oW^0X(6c10TrW>$+U6p!lKW_Bvl{qCx?jXZMTc8o9@V4<5GX+;uG<Off0_yx
zCxIW|zvs^sG>^72KV~+#4!v%fRu&hnaQnwMCEYuPN%6~*l;*46$BzUwk)zt9dS!Zv
zNg1A>AKyy)s`j>50*j9sLve;&yd{+IZ@KgK@q#)d!T#Siw{@qF+2S=b-6eswwTy^b
zbWEai8@-3UU2!$<lqSz8;8PLdZs;uE(zZIYW5EKJyR@{dkiF+JmYNv7MK;c5@1VlZ
zuW#qc%_DN|_^2>&_5b#V-6MnGu4qiz_S)#QjIP{@zQ);=a7&-PrkNCz+(&Kge_%@J
zbU5no@86VS!0`ioQ*3uyUqi;pdwqLq)_c25_hy&YavOf#^fc3auVg5adl&Uaf+&i@
zG5Nt}pZ%X|n5i+Y?)c40sp6#@>6w>XckP9hmOuP|E<ndhFi&ZzRmvNRzHym}M|K)D
zjiid+b4zPM>=nEada-A>Jy{TUar1d5xlrSlql=S#=Fip0a*vpeo`jL)KXdtCNzLzl
zq)5K&nA%VMBPQiCt>^HjC$VbxY?#!#2g=;XWIs%dHH#dEbeTtW4hu&seoU{QC0#hA
z=a)OeuI#cby29K!jv{rSe$zI~dI$0YVBuwFn((`pW4!lbvaLvM7r53}R;wgpj$~`W
znkBY%ctC{Di0!woO)@z{9MBm5(xYP9-bHg`PWU#xw!%#F@@QKT^7yP<nPiLe1XITS
zvUFXXqow1*4ZK8KWaBAAC9j!hT<MKRT6YJC?Ujj%99GB+DgNpe&|>@rQ<@E+-nFkr
zjE`vq6Y)D**K3>xKu}|D@0|*?>*o?3biYUG!vqov0LTnz{^svT$CQ`P%!V0!A;hPq
zsZPItT)@ovf|Knz4Gjjj4J*s@u_AF!wAaciDv7pJg`%~$&ML<>qEXChV=3F)+qXqq
zpar+5i}=8`xuwm0M9j^xWzAcbq9k+Jv2rlyb$)EO9UT_OW7=>qct0WJ$o@DtE|!#Z
zQfoXTEAua(f3VXI4a-_s;wSfYUpaDt)E^QlVI3C|B1yO@Ewyii@@CRpPa@ND(<*H$
z1ju}g3YlZLUH$~-e}!CU+Q^v<+;&Ku+!hfT8r%VIYAfSk)Kz%I#~CTbonw6{EWg=_
z>=%Z$N315pGilk;(jZ3@Uaiz=>Ymek%D#g8QXTEQlT{Q<x1{lWf<5ZHCLZp;mmBCC
zj^=V`8k_Ctjg%RH08o@!?RFQt=<mK@t%HDpb{tpnMAN8*!=;#-NSr*V!&~`v_)uOw
z&X_9M{vbWMa5Y_n`!lvn4~<^u7bN*^F4h|r#8cL~Ud>ZhdOKKx*M3$6(|5_#%xwW9
z!<7_610de}`gq^02gCHKD*LU`Y5SAW%=V7Ca8j3#TYO6D%FT26s`T0!(?1iE^L)#u
z<N0iC_jPKVR#xBY<|UpT?sj&zyGNYWS`dpnaYli^Fzg)gRC@Mhs?u(i)4+T(dHc(U
zrVtdq%8!CVCs$byXlaf{6&iJFo%dfd)rlW7PH}$>(e{Yhy8@F{Q`6sB;ir(ZU`7}4
zoXyR0@?HYJQ|!EGc$>lFsNE<wOnc6TqNbGW%~%x&4hg=lLGafuH-Vbi&cQ~){qRUp
z!zVY$q5eCX8oi396x)R*yI$8k*<S)YTORulz<;@eWMpM^MQ(h&T>b|hbttc9rsI*2
z2-T3#C`w%)_-tgnt~N|(f8dBg9<$QoBWKP|H(og!u;*1ekIr7*JUhCDkl<H2H<s>H
z_4!YQ!s;^&aHTP@iIDxLJ<cUM85r*O<A$`!3o<0Anp>SAd)GSNglfgl$Mb3Szj6`;
z>ltY#p2t(VHpe}QjVFRk!P?R_m6eVC>y_&p2@M|zH@PN4k^GfrL1eVgUYO|zyg8M4
z>vfBrjwID?4S!dwr<LhBr33OEpB(9H%lvM?@eP%rr1_H^f1a(xa4XV<nxN)of{l~v
zahECXPz@6nO8t|bsVNL@xl(dYT65M%a=-D0_Czn(te9q*N73EnR$1)CHSYI~Tam?E
zGi6asgqXale<(t+n?0;ZH5%OgyOo36{5twv_A&ON(vp~Aiyn$?Y0pC}Z(-{GWw96A
zKKLe~FiLzuhYz;=f;qD%;AfOKZM;3q!OL3d-$Kk=?;?NA|DV^CO7@g1HLGcVxq`8$
z)U$r=iQ$FallJ!CJfRAi<opi*CQ*VL8TJ$8Qs3}=G+)_Veeeh`Sz2IxVoU<XQhZWT
zajEsbGR7@7N6+Kse=b`hV)n)>6u9^F$#~)1so@I?v{Ac{?YAhV#sv<8EUCgg1O)F}
ztQBi_^xkA1ylKTm)E0SK`Wv#%R<OV+(mHq^9gGwmH{oOk6aEt7GObwB8s5qTZkNR3
z?AV?>%qr5mh}jMNwYBwwBfD)LN9nyRPW7>u;B<dkQv)hK2xlrUwkF{*jdXoY-zm+|
zDXm-Winq%b#>8_JM8uKdG9vpZYD}N1OZsw0fi`T<%<Spb^li_j0N1WA>CP_cWk$7o
zNIS{B-;ULRW)VD)yD>iO^k*KA6}wcJCj#rAR`)Gk_s|2$D<={xYFvM@V8S<VRPf0j
z@#(GB=ktVGO{YQ_Q$C(MX#%(P<^X7nC{w3(_~JCmzH7@l#{c{g^WNMbf6$8mmRH|U
zpGLXC&|yrOI2Lh=rzKcSzoPXGoL`JWha<ppD&Jq^B1leNRW$})34$Y`*dGnWl+nw|
zG7BQPW55yxtHFF{369UR#=1LoBCcO4u`zBfj1T+$`6b{Hs->b5luNw2K~X2Zgm-Ad
z;l@Xb&IYMr*?{E3A7HkgxB-D(qTX};e={p9&(!Co#$MvbRjs?R78ez?cXcpDEbM5E
zjTsoND5<Cj;^Z@PGSJdeVf$g_m`h2vfB&xAQ7b&3ul7ED*qnook8fXJgS)cTs7UGM
zCRp5!ly*ZPtJ8dkfBW)FtQX2bae;WqvpOegGfYr|k_7`-AeKbax*~P&2QGeqy<NVl
zqGDHjyX+;mtbvt!WX!PA2(MeX+VPO9!f^iH1!bs2aVfji_Hejn=aP-oH;h9#eb-%(
zUvFqYk%-8vDNXiXjs0zZ1Ei?xB-Xjp0Nqns;7!BMe?zlad?0;bEvLfszC{JTG&VX-
z4)pv;+VyqQgnm}73GTkN1BfF{a&LN9;7|%vV3qAR0(VgDk)5lB^zl5M8O#HF28Pb{
z^)F1{YaiwHo3Jn5tIf|)<Clj)EO(UjRcyq}bPfZ4fQ;jn(BE0L$e1^+i_t{2)oAqi
z0VQQ6?lF7QqrLS_qdWQ<m97kS_Ql!sooq$6;A)-C(4|}BlwO`w%Y8HV`ddNhr+Z%D
zD^FV+@rDk)h@(3QD8;R*Dk`#!rVnnqO)<ue7ez|j#c96rcTm`|kY|5`VXKIPgOglX
zSh&LUJqD93JFOj-6(ox5<zCp|bmJ4$h9H+fDb4g~qN{y^*FuSR+EVU$S(@sRI4ypO
z&&5J;+iHvrGOSE2Rn(hZ!Xuv%*MWkc5QF2`HXcG7;lx%@uzs^=23KTEmk77-eR2W&
z{4&bWudI4OlBuxKC?5Jf-xg?f6=o&0;~_S)w}0WnL%cZnud7Pw)F(_b6%W&=&i`I=
zj-pxpTWsBG`s;QDBH~9XyN(oz`GH0H%6QwS0TJyXJTU(4YE~B&U0N#=IgJvRks0dy
zBTIuz$?KxfxN_4oJPm^Ce~fwl5`)}G*uN6p6I?unYMqexG{#245!3h%GLXO~7ZRS{
z`@P;%j1s2f;RuL2*ehZ6!u7*c^zQpQD)CI@Mxq-JL5O4mzsnOj`=}_Qm}tU?urL9S
z-924m+e);V$KKVSlj<7!2Jd<^b_cOo2m6U6d7qXkhJ<;K?Kednlk;ZJjZiG1rF8#7
z686zWC?3X%3HvPVS$^y@%}C7*!d-}x#slU+>}2(Q@_|96myWzT&#+&<EWLqflQJCm
z%RhtXRXQ9_Evx%A$PeevddMwIqN%8;G32eK(P#Et%*8#SWn%p0wARn0QCZOO;p5rH
zDDzv`Nu;H?qat)t`*{rdSjcvr?hG;7cs%#$#PB{<fXe=t6{ZI~!IO`wVXc*52n+<P
z(cePAV*I&WFY-zf_DMOu^M;NcJ@PB#hShW$&9X7<AXs@WYo@Ea11o*E)y7i%gB}@s
z1^syoHcmx_wy@fPK7DLUhKy=x-as_ufBNR;mO^Qmpc@nUv)Xn$G0-fWb3Go}uh|t+
z%zWXzaoNHSI?5A@FZ5K3aKRt%9}A6ELi9OU-HC`$u!Nqlk;_h}R^LWo$qs6%;2QxL
z+)PJ9FJxw?_$r~xzPRMp(eV}IpTT1!Bs|p5;Fj`(w$6#4$H-e#y)UaNvZRv+dp^^W
z?w9yA-ah%#{n0`^BOLT;?Nx)vfaL?rTwM=4PHdw;Gmid`OW*z8T7rM8I&~&gf($a4
z_o7JLInTfiLgIQO!FGYW&(f~9LqFn8UhY;tJ3U{uhT6bPb{W35B5fGeBK3{>>I~6E
zHCE&?-S3u!1d5>Lao?D)wkU{g>Ds#<zSZq_;WvY8_>1P}d}MY%Srn6qPhb#R@L_oU
zSyXZF&SOONe3GbNKuC`ms4m}l<vLyDz|FXZ7lrgnPGKQ5sO=@6RuoL-_OM0X6fkOQ
z1}CB;(%1u^^^bm^Sm@1338$f;NX-gcaC*stXJ}F|+BejclKJt${$R*`VOfypg5Nk4
zOJ5=|pa?Da<_C5Wk4vxygOByeDJko2A)jV71?G1gkq`N0qDfiF$!LhZy%2megOI8L
z-3+7mm!wOd@0X*4^8dzaI<CB0$oYfv=Js{#oK%|^?9I1-PZyIrH_z^yh2H%ZSVHMq
zhd_xB#6_e&I&k9QhJu#X4Ft;XUxuH%jGo}3-tr2fWek%XU}M|(B6KS*$cT+K%G}&w
zhhxbNDtEY+GmHe3gcfVSROweUR!I)*8L4MlmYJ^}11i7zm0HvC=jk_bsvo^gshNpX
z8!~9{V>SBlHYGvjv;h-TUKHH!4F4+ZoC+2EWn=f!V&>bPJx*vtLoe*ObH><-=ew!~
z6;~Lsh$riyv?$)9yU(LZmy+kNA^b_Tu<o<&yfMtORBqG%UK#=H<*Yl3#eJRR*)Hig
zB*=5GZmdDJ&>7Qt+S~2vl7TlK2~~Xse3Mm5%6ra`DK%_{RHmk87Pv3<XUA%B60E?U
zhIlOSK2G#H8yddHVKEcv4jCg`<F)VV^R4gcs8@fmG&16d={k#5?M>EPi79&uQL`uj
z@6j0_4}*MO&IE#VgPjl0rp_2XA+zJdCI>rzpfLohBH)U)?`0ne7zXJ@!%^roMnP_H
zhx6m|&l*8-{duyR^$<u@poZZmA}Q4q%U>bvDl`v$0OBc`3UM>3X=%$A?dL2{eOx6w
z)QYT$B~3GDTq8<lgBNhV1!7PKl9!kd-8vx6u8EN%P9KK<YwTshWkik`@JEUZl6}CZ
zIg?r0U~>~1W$y-BEMJJ)V&)|q86LA9jchBFr$@cvzTq<j@aXF{t4;H4pZkyUof}oq
zFd}Y55)y-y?m6Y58%vqppUJ2dChx~?5Ag8Bdbe^YDGi@PNF)i*{AEk=&_<bL>YJ9~
z90(JM2oHtsWngG{FN)7%^5@G`Lqo?K?=~~uL6-sDh{?hr5mNeZQdLtE1G#yA7{y}f
zYM}Hjn)ZF)=a&O3Cy6z2K0f2nsz^LbQru9P_7d-qD2U(Eo~S6z%*lbwt-@mF1lD(*
zi&6Z>M0-bjy?w)lh0Ik>JbS$bOySO88Bo8KGh#Vu^qbk#^n@;)QHv!1sP6|VvN||z
zVejcNcv-%s&9mz`G+o2TUFmWyLWlCIteey#n;{VM@CO(iob}(iJU=QeWd2{x*8bmT
z#;OwL8G-UJ1NYwgfVdnsgoUMY1uO-){j<n{AQe-+NsFz{=L0Ke&sCV<?GcaW-H+)S
z>>pfZ`|xw~>`TE@up9%vYg?DCt+i}VtB~ZQsVax!(&G9BYY1u*Y|rRpv6f(UTj^dM
zoqoYh)!qBb<9`o>UgSW=*CCg5qKaQinux`2);D-9H8u5hTb}0Fgmer$N8PX1_Wo=|
z7tVX$lLa*!-|;qZ;_H}uN2cm*JK%V2*2+!T-gFae0SpB3%q<;`3>8M+F6ImB6-4%|
z)Ty60|0*_1$AYTpug@7+wX}*oK<v!7D_sk=FGt0Ze$iT64vPuzzYCq2I3P#x66|T~
z+8CWaZ0=Cq;P@v$nNLyRcxZD?<jxR~I={{4G9HWewRvNt#TJrG=U|RO)YbF;=yNk2
zzpM1XaQ|e1lT6(ceM81}j@-NDc<!7}*<u=vZvPt#KR>nAa9;j-H~GhR-KYQ5VK+V$
zzU|)K!$g+__QG*I`g`S>w!My8C7m<FfA{)+eED;;`Yx)FkxJ=&l7sQecw!t~pqK>>
z2MEjd&%#d8(1yeD7g~ZMp%kF-h+x>W&Myme!L`ye=Yl==X#YZ^q2{vV33h<34Z3-b
z``eXEVyKe=>;7AkQ{<}ktS&VR`iiCH`OUFpv%QUvn0TB-H~x(nrW>Dj#VXOFc>2^j
z_OTfg`?t)^oeOg%=<>L)PeLw@yBqI6^Bk*b{}7fU-Muq@($hNP<x~vyvDI5j6k+Vu
zl*)P^gVqf#3>5?j{J^{OTObsuZHyG6gp`VwckD{bOhO_=Xwf{O$=*ZNDpJTgBIR`e
z2Q7#|<F9cBUdG_L(oKXl%v9aOEx`Ol(czxH7=~a$f?WJvew+Et_?zMevLx~?FD||(
z{z@eJCvKvDc|w_*CwY6-(dYDGdTC(1nLHzn_^EUZAKH^#nZL~LQt}jNhy_De{8Uvd
zBLyDvdSIA4;~4Ys(170<jvos#6rA7|m!nPh-S)$l-Ma>?>J*`l2dn<wr^agP>d|fD
z_I>hJe+f2+dNp|*$b0(-33=-w2BR%@_5#N^dC4Hz^fY(TO+)?cFt`kPTX^|V+^^P1
zhe*l(bfu3x!O4l`rEX^xE$U3rEwaB`R@D4_k*>V7FmOEH^_pzC{XK~4TMg6cc*WD~
z+fQ|0-l3v@8ud?EpN}H+k|~&p?a3{41eFA(*NMz4OrjZi{QWC}?Ate-S3JLAjia|)
zrtsF}oxf}E<Hw4mB+-F)b#IkdOGSiiyn__G!Qp-q-g@HTmh{@|41-}AKfWmGt%0l-
zZ@LkvA9uka2w5FTkN9rFA;Q-nQ5*J57ri$}>h`_Um}T$8S~89C4{F)wn;kx6-;aFa
z&?N7#_`D*))Yx2{z`uMRDiN|3f3R_u`fZFrPwl5GVXX~2;@N_OAOU~ocO=bvIcr2D
zi6+lpf_I^tpg1M=YeVHBIPM*XULf-FO2vjP@qP<piC$q1!y$c->$)E8TQxCRv(r=U
zyMuslQlA9t<Jl^IPyAG9x5Z9s)}M%6St5b<Gdc)CP@{J^`?(8{t+2}bm|?Wm;0HKQ
z8}AORghi;TsYjCXlxqrW=d7Mvo$r-9QrgYrcAaC936DRxJUgg#VG?6$_S8D=irr?y
z_1CF$Dr@rKpEpjEB~euB)XsNAL$oR!_V-Q~%ARi3IG#$<;}a7<+M4{hD%wv6YhtZ4
z6dIVRa=gUH0kAnikmh>2j)_2D8K%x7m4YqB7VspQ3M%f2hsQzWlWrU*dn?thcH;-y
z#YwU>yg05ePX^z?zCZaQ0>8p>?-LZS^+v<z&s!CG?a?Bah=CO*lzMREW7MjJ*cX~p
z3$52CQ8WF@Zq?3M<h=D(YqX~nOYYtKnw9>7>Ao}NG#)j&2|8jvXN@h=nK<|EqF$5+
zX2Bg2DBtk5_Q{3YipZ2lzU{UkatE`j$TUiKG_K0faOub2z(wVe&V9{W2mzDs{k!~g
z<HOJSpD4hmUESE4_MqW^68fq>fyX)@pa#|4DPSVHOQ}xXt@!}xopY%yu%xqwt^DL9
zn%*+|9$6gk*$LQOVEhLlo-$~I?SaETmpVum8y#=gTc@9a))f!$@6hZI*&Izir!x%X
z#xW{N{ofZIO9|qd+<qd3djE`&2U1O5w+}=~y1r#D^q}gBnc8GUoIg!C66ljeK0rhw
z5-H$9^y74DWSfEK$k@!}85QM7k^Q||yD-`GC_-Y6VtZ3uMhg<2(vNY>7CP_VJw!G!
zC)pv&a!_7o4vtc|&L*{D0ct$a_WDXW9KO7_prELjoS%P3{kF&y)|kiYNr$SHK@a2u
z-QOH>0o-r3LwGj4M3FjhD&LSAF)&(^j^gs^$s^@5ua+j2hDSxXDwnXq>FD_}FMdi~
zF7{Er<0oOd3Fu>>T*>pD5_|*|XAlX;18a5wp)AG<I8O!>4e3VUZF5^w5hBZ7(X`}6
z4+5GQ_m5w?ZnE!ng~x|3t?=Bg_zQ8E&_Mlof3ovCYg@pv{9{X6oC_AM!)Z{<Tzs9Q
z)4s#A-%}x<+{$ikCBLCTU+&T>HEmotd5<tVH5kuSuo-pl%SpP&h5guoNkQ5TIw@N|
z?SB(<(1)z26rg!~qeo=l@C@nIjOZq-A$8wg%Ho7Ct7-JZ%Cr|;y04%0H?xUze>5g=
zpzd!P?M)URNlc~&A`(mlLuPThY}jEl_Gc);c@1mfzut!3Da}gE2tOdx=CGk{80+Y6
zk8ZKG=~RV&)m_Dvq|tXS4(23$dads$FCG`qZd25~Ou)K+z)UFZ>%bTQCXbvi!Y#zI
z9IIh1qP(jA?rn~bllkpM)@)*O>RqIIeSOdW9{vyY&dbvI|8hT(ezJ7=R%KqDcMl1W
zkogp)aW0p=Wo=1y^@=o~vw83Rls9W#OdBP*h`ejpMFRzHn~?j*cY5Y`5FixNRNkH^
zSY;m%7Qs8sO_N}w+|Ihva2BINBBP>d21;S>Jvb8vUA6dRRbIod;72cU5vU0ADx=#j
zPzcO1KXo6S{M|d-VUf|*C0J@+P!4EHG799Y>z~2Vp5T6B6;#+A!69fL`c(ugMq5KA
zJVN(ZYNo>Efx8^b1F!m}@cNM{hj$7E6;ZD>G^miT&3fv+_|mqTJI(<-*bU$NS)(3M
z6hID@&tELUy3b6k`!x^xjR%dqA6Q!iwnGkqb$Q=YpSR3}p%MUR1x-#Q%AG~2zg)G8
z_<3>GK12)!PO{H^4$c>1r|zM}m6g9~$pXfzYjv!8$O!VZJPw|rkY0x<<06qUMAW<<
ziHW&B*x7UYXeBqAWw8;&i48(Mc4kqg^Jc2Obh)ZupaUfO$MPU)H0%kv<)malZ<>j7
z)Pcx|X66JXLET#RNMfB6IWjx^bGCqsYOSL@rq7wRF!%GeXqJjnK+F7bxfPnHtIns9
zus?0hRyqWA_PzNlA!Df}hSO9v8fK+8JR8<2Xg>*2_fo#`;}$K6)ZGh@=~u7wQlcfr
z<9HPCXKSONOJ#@%kvvJo`+SeeuPtlA@aKD!S9F*>#%G;Yo$RQdJ=GT)3Jj0rx^vDx
z(d5E-!1%j}b1fWC60z~TZn?Q>{%4^*d7#M<z0C0J2y#0iL;&q%&DXUm*1tFuWXtzR
zKv&Ie;q#Y!efcY`-P+zdyiT)|VQ%+#rjkEv5vP}6dS3=KYEGx8=J<;4_w;@<10pi{
z{Y_bE_mKk!m=LMRKEuvbx;MV`Irgvrx1<;By^VSWFJ8D7C*56Nx8Yy;G`V<<<8B)r
zEda2J$c*d7PfL?`R0DL^;!oI88ffPZ#ZaUJ1AqO&&df}l&d`-L^lY9}QX({_V|*wD
z*J&1bfAHGLUmLw7LJN_2_EoKzne$R}OU3=UYNGttVc?Zsyv&0JipDWgEE$F^E!4c?
z0#7ND<46`a2e5yHprI66%Et}@z(CHQ_JkkMtn!Z=ydCL}^rigNu+LBKvW~TOw^2Uy
zlc)vXr-Z7VC37vMPPcQ$p+D0{Q2$~Vq@)yMm_QEJ5FRex(_m+KPYZ?`CCbXYDcq<C
zfRExBk39dpji;n?JBd7+(~qV<zI{L8@kVfPLO*aCI%NU{R+a#Q4m*F@GuDQ}_ah=<
zrZP<1RY*_sean}X*EDaxDA9(ERGerVZ)_5h;`nxV3MIC-2+XeZLsIuq=`gGB;<kV*
z(TjJt&L+nCp0IDCrz*qFx;P8;s)3iGpx{}+!B!XsEz;n!kP#V|YeV<&ct>NL2!$St
z8f4f@>q<R?`8jwoBl7C%$_fgJT<FDrMZ}taC5(+#7~4YBc_0Sne1d!EC`?_nmoQqd
zEk2k|{m#!HZ2}jOR*03^M_;OQym;^o`}O5+Mm|a$Dd}`tbp(Spz!mKW6(V{M110%B
zoQT9}l?(cbywU1}U{Z8cd#6)xM4mcZc!ol6feZ?;(PS3_+Ax4Ky56!6jL7^MZ2H=!
z6xw_i8056d^U>#1t&WOAkQ#*($u&kOYlZ_P#;}OrL==t}ddcPD`H%NL+(vjb%>euj
zyKxVzVL^F9P5r^$FWiYc{b9Y<$-Ym6(sapt;3Lpna>!z(ce}rnLPrmH^Jbc-xX|Tv
zJy4Be&PK{WqY1)tTaAbRwzJal+<uY$>apXQ{och(SIM4N?x?_4hwrSHWMgwxUQG=k
zkhd4g<|L$>9+VOI7mvP*B+=@;SV+h9@<3C=5EJ$8%L}G<9a6i6V28pG7WIbUYS)A`
z6^^$}PiuR32UBmad~o$gawG&jU&Eb&yHPV5_yB}aG10ElQZq#5JspjtPEn%iSPTp@
z@5OzA|HbghrMiY!sIdVY0%4FE^3qw6*iFR4wt^D6Ea>(Da7G?e9_pKzM3sc0c?quA
zNFCiG1u$1sABXfJR)Yup{tOIPAS$oD@6T1!+FN9GsZTqKFDgD)mZfZQa4cYJid%Kp
zF=x}<uDGPqe(S?#)7b6cdEs2&%&aU5-Bvf}LvxOjG>4t3Dm(Bmb%lv%hpKRwj7+!A
z9jCp6vEsIc5XJkfeqUOHem$jzt{9TJs_aZ5=NT$VU)ejx0eckBqk2|;d=)vijKf<T
zYAI@o`<f9Dkq32EL1d?Ji)_vHj>*8dr?Wf$#;0bb=k)ZI1f-hk>P}k?cqCY=YDz!D
z<4sKUQfB3sm)Bi+3@E!&|DOwRF%8TE>Ti!Af82rupDvt6%5Acmba#Hq?9(exd`tGb
z#6+u58uGYa*ujULR*e|a^Fw7|Y&E8#p#9t4uSFa7hQ<Jw7`;|QW6JL<D+>p103`-@
zPN1S0d;!!gDoo#XEHEc0TC%9I8#rWh`gXU|>xLh1R6|+xVCO{R(<@boo$jB^EU5`k
z6LaWZ@e-GNSz>9wC(VpF9<7QQcV<l;vksits}j6d#^O0_OmA$Ft*R)!TZQAqbD<-!
ztrI}dX>YO7`K}1&2KUKcu8jg`s75eh|K50+Es7`9)#TjH-26{=A;a8(LnT5;NPmGW
zn4Hgbe9;<C0LhM4RK$Syd|CURpGXt(w_|`5>68v+=#;_v@tuV*yCTX@`ff+Ecv--c
zL{2Ml4fW^^W3xoZcMt}qY|ia_Yp=0o*YtNR&il`6>_mB$+%Wu43?sct*P5NWuuCs4
z#Xv&XqUpAUhcZneNAY80F2elR=%oL47N@|JI^|-QL$rU3`=gT|n=ReJI-3MIo-Olc
zA}=6l&EM2tky?P$Fa|-$0wVaH-#-p;BVVC3|5sQVBbYw<yV^z&?U8tYqS->h$!_V>
zk5pul_%A&aSCq%XD){@PJ`if16&v73y%=^7nccsHpKv&6JUUxD@*WJqs?4RwMi>5j
zz_I1Tv#?eC^>l!QD%RCDSF&{@W5cMM!#gdLj#!zFuTqwg$?9mz+pVjbLtyHzWXu>Y
zp6xLS*Zk(Hr)vrDLJ&`1mojwV3|HEYwD$|ish&5j&894lxbqj?KHA86tZpELeM}SI
zXcWgq`t43A{0nRnooZL_Wy(eFed~)NNM6HNM62K(=^MV`d2rB~mH2Mm)D+aog{O*g
zsx2isww+$$Ai%$fe0aF{Zs;thEpM7$gIPR^%Ngb(IXSjU(v#4+L%1Qf%_S8#o@A<L
zv)=i^y3_h(^S`S5o$>q?#l?0bwT%>^tI&QGeqxIGZNTn-&#Mor$jIBl999!-j=zp;
zR&PVDJQT;P12<gbK3jUvykH+B2xy|k@G1HsY%haJLCIERgZ!#a!e$7Ck8ogTA8o3G
zcdM~TybPwnC}te?YlCE@neJ#%41N_C_W7$mR*^v%$6fB0wKozinbNM&1gB4m`-$Dv
zyfS}XX<FLf!KT)x(bldK%z$RsOOB<33Mq(+_6<(B<q2>F?Q*_<K*b0EPlz6<Op8Ly
zt|5DoY_vr5NKJNY<>5kFR1{$XZ{e{fi(N7Q3LXi+`}kO2Vp3sYO$@rG9~YZYsPXzH
z8uH;>gIV$J@3VyZyfJP!kF7Ts%r>?GYS>7|-2PIPIzEd|TWuXKHn(9zBvdTNtHY;?
z8kM|H`e~Z5{piu164Mo)x1Qhq{7T+eR(!ALrfV6UK~dx<&EdqlWc_&qjOMw8bg=*D
zbFqto6j4FJzN4wz-|N<iT0Kq)?s=n9pdcAa{=ifTwv6jLjE(Iv78gzLi!Ql;A=<lE
zJU$bO<}LK+EpuRaxMboT)EQHT2ZM=hiE0<xjkNnt2AAuG%V|{i$y-(kaGxg=Og#=T
z@)S{(y-`>F>6yxMOM05mwTnR9rH*{25|{%~$s#*vSw-pO-f9$~nZk5zWF|`f)fh0X
z9L)PdB_T+UKf?APOyZd_*((i($7M7BmipYTHceox4v)7$0NcOhof+aO-5TomoM;@o
z;AEmH)y}uwCtu<j#3mrGQr|p#V-jpRlk{;aAPDpV_Cj+t=LJP+GIn<P;;sYVD%^FY
zL>~^ey-?C7rKkg=*$XXNI@~ZxL5C?_(<iu><|j80<XxOA{u_0C;LcQ!KE|8Jzv%e%
z=j}dvz%{S<{uJZ-#*7HXJAgs~4C1nx&yLUS4nL&3d5l`)6j5ohQroqm@;HneRmc9t
z*-P7XR7IR*V6&oTkKqy@pTMZF*48L{2Ak$$z1{R*doywAW;Ma(tfdl6TK@rdn|bYO
zmzB9tP?7oC!z!;gpJ#*PHg}0-$@><uYeO@Nt*Mdz!AK6LNpSrF(9#WVcGw+%zmZfz
zmmezs<sNytQx~upx&fM4z5s6eXkPG2?4uuHCV^uE)H!NZFXyK(zq%X)S3cKb1146*
z>@P~Qbsv>;-pRCJB|K=u5xc*d)fttzyP=^Gt;vx+S#jK(>*kB%DVazgD<BmPk<5qt
zQ6+{gIw{T4k7p-Jty2a}${MyQh6KK3C6+xt2&eVE^iDPr%1kNoa;w*yX>95F6)X&>
zI6yUaTfe^kLh)1|pkmanxN>N!DR0@Oi#TiMbGBJfY2vLjEX$2otv3YBml5(e4=-B`
zvpA&VxHzc62cr7(_K~*C(>W*XSJ=N=Ri>)Iu^byQl1qp^yV3_-HFV+qaGuV3ncWK;
z9Um+kz~J3a7S|eE<RWs$5#;jBv7L*E>HhJ_mlY_d#5J;nZyyTZ8re``9LqOcnOYrf
z%^-aCr33%8UiG<BS}Z1ub5qF2O!Du#HG=1jX!}MQn$ci(UC&^~X*FXDxX#!tlMt7c
z%D(K+81A9ojKO+_4S@SUo?WB@hBLQ&@*L;yW0!aLysPy)qH{i-cWYz3{K)b}GEtg{
z^PZcd{L+U~D~%d`Ir-i>$U3A51vG%$;H@rFem$F5{oBgXzz%ppSOKCWn8?A=W`Wtv
zlMuxu6ogH2;V>(Fi^tJ?|A-IymDHeCV3(+;I*%~F>)cVzo1x;BlazG-$McaYxP|A#
zNmI3VM}w7?`4NtGvf8EbKtPhgnB4U!9x`h&@5e=CQK`bTKvvfO6y@#rOHS$%+qkrd
z9jW75u0=O{LWIk-xbo!*1%)y0tX|y@ue@Mw+$!u#b|a$rf3_kqxoC)HWp)l8x@^tM
znT0nV$f{-UX-EqreiRxK!;?2Rq@3VRyotw(I>kM7{&;<|Xj+_VDvEUP$wB|QnOgk;
z%P!aShLl}O+ARVUajvZr4=9^kW@0g=?(2@Qymd|qew95KKLbZ)aklGP`ynzrons4l
z0Rpb<<?jynV5B6E%6H3)EZ-4`PFoodWR4fU0wILYEj;pIZ|L(W8zg;Tk@3A~`S{vk
zbZJ&<B%c3VjP9|vJ)j{?ea0DCG6}p_gB9pJ4!2P}E2Dq&w(vQ6aypk;S=~R~goxpp
zYMHtGu^QgVFx+pjUX7L(Yn@^WT@U}lHZn9YvXzqT!)14BSS8{Y*mOwQPC^i4!}mo^
zR`C~qbj||vPlj*6T6)VIdD}X8fvn}GrxbAukhxfF;yV)frM}5Yi0pTMeR4B4j)az$
zb}6E@p@HJS_Eb-oPn#5Nf8|>y&TlhQ;K3cAS+vi4+xGD*rg5AkJ4Wp3)Dl7#qEKbQ
ziYzKzS)3Y@V`El(Cw)LXf@Lk7h5^JPz{6Np4@hH1%+V)(d1d8W7a?Vvo!C4ANEeH!
zPk2_?&R$v`>=^ldgLR!dQ@H;|Vl_2Ho|nwyY}MV6E864cRqPQK_hP=f^rAHC!&HI5
z(c8hxospYwttQIKR+>Y4jS{RRP0<4amKuqR)M!PUDgA%YE{C(PpNr9kO$P+k5Rghe
zo)yO=FVt)M41j2&W?>vG=2V5u5?8Re$Fqj%d`MduuYb9g6Xu=8Qu#;w?n??voytr3
zh5{g@H-*$@N&%Af41ZN{UJ0=^RN}F<9r)I1{^^x`6T@lkyQo%ygWzs5r~`oCN62eK
zQP>Q;u(UB%_azd#`*scQ-ISk0HVQC8&X_<*Ql6jTIJI0akYKOCLQs(|__iRDRq=+O
z$mh+dI`JP}T-KYd5SyG(MR)@Npw(e*@QOM>=|4>aK8{<KVA9uvIS)&e<N*`tdJk^z
zjcO}3HR~}$3Pz6Ehl~gJL{){Y4*xB{*zR;!Kf04gSm9v=<@>~XJ=OqR(-Tz{<u(_2
z{Q@O-7rWA=r1=KDE!~GP_!sv?C=_HZKY3^Nk#~eP8nsN<JP~jNR*l~F^eeGFm>4>2
zk7pM8rzhq@hkCR<t)Z;m{I(a&2{xyQF%^Za#?&-#CGcx$bsDln!5sO0aPE$MM4KLK
znT3|Nf`Yn&db3_;Ys`d*udY{~`vVA0lO*?rcTMOqGxp<eF!nv~XrgsB`lJ4_<QGFP
zR?*c)x=F`Nt?B9LR8<uzvHfnI9fngH8yJ81u<c;^+C&CZS0->jP|p%;SOq_#hm?+v
zuH`MW9k)!{;q2z(ha*S%cke<a=!fN1#h+BhGy`)rU7BQK^<Z&x^~3V$Rgzw=#IrX`
z`E85axWv#g&18CMzBYOE$@6{hTI;{Bb=xEYGcRS9O5Sy`t{r1x8LTeNZS^H<x(hzR
z3RwEU)s9;msJO8aAsZ2<{DF&})_Z<=p4Vc}QRG$LUA1EFx1sdOaObw}DXS{-Ilq0c
z<NhR2%&;gjSqvlGKI;aFK(Y7{=W<ps9;)!o8oCuZTztID-!>Vzk_MxOr>JcPd&~B8
zu-!ubVas%P-3t_(sx<+l3e39FaLLqG)K{So#IzzPgMQ#N7)?k@DK4n>=C-%fCQ3-8
z_l3;hBh2i*_s5HJQi_W!&u)Lox3cQt*C7jR{}z?G3sfE8&VhrBIfqnjExu8j`29_K
z#^pUTG>vGYJ#BNV_qKaG(@A=k+HnsKsValMC<?yz!Pw*hSjs1`7oJheQ7#Ze-ef9;
zHHHtVqQEGPG5%!(!*fIC)r_v18s0l1pII%wvpt3SF>-s5E<VVjDMZ(uhk$$M#+eRH
z9whd@9k$ad9kdXX9ePh2pCmfa^e$#(zE6LZpOjczMf2}c#G74Q(#lfv4#E0|SO*xB
z!-m>A_9$MjMp8^#XSf$DTwKZyPpgGEqEclqjeh;1D$2v;)oo{5+LIchAz8ddm}sXH
z?rM07?r#A2ynRpP5qEGzp6EczL*u>4A9`Tz3;byyO+E@}Zm(MS*V~<1qdDPo$>rMW
zndo)i5p6TMI;Ss&DJ&o(^Nlg?zmKM@@Q!)aE>medPmXCYc!DGMjcYqOr}Jp*yPF+!
zw68_<_}okvnR-$K_4yLEfW!fb;lM3_+_KaDN=j<~<j)icVC}fWnhmhXMhOhOh<x{>
z*xdg9+-;tOHQ(^AD8yQECFo!8Z(#2&UcXXKQ^?#<WyqP|g|AdeX*GMB;mw-ecu<#_
zQ}sIZyc|6+YoJx5(5DbK6b0Sc-adwJmj<&HRh(F|d@CB7H^?0n@V2d9ym+y=&)Pfk
z<8pJlL~(?Xl(_bDqaJ4t=Bv0R0C##OA92w;Ttv?4i{Do?s$^sD<#FN7s7?~12=85G
zS090~6**6NRe7W)-2)(x4{P=jd55JsC1Wb*5jt0LW0Kq2n%FMvMfDfjkYIs-$wRYk
z*db#@RoOMS4i48>rle|5zoLHL!#NlDA`oFMgL!T3+K~U}@-AvMU>{INCIp$>NA&!j
zEsldnHkFBq>EkZbW@FmC+xHjFU0f1pXDnhh09@yHibe$NLndlX0^@RCAa}fzCMSFL
zrX4En*S<EGI&98Tg-T-UQr&#D+1TRd;zE)8rP0W0ruddg6QAQr)1T?musE28Mbc_=
zW^Yck?2Dr(6SCYVb1MHR${v*)G<+2oGo@8)O&&9R>&r}e#^@x5APp`b02YHgPg+ST
zBht?kYbN@f$%&84>&ghCy@0x@rXDTeyjMKtlaqzhK8cg`0^tkh&d_&Jl|pBev@KPr
zqfp&@;3oo59M+@J2#IH=W+qshGv^-*1iFSb9JHkrZ<bgT@x(y8rlG0M#lwvtp|8}6
z-^$7zGwvv!^qKNIopp}3B2MwtKVHM3>jxT@DO0;gDx-pg*ionm9FlG|ynuX}+lW*$
zJ3DiOf=7>7W47aI-LD-*jlVmIH~V@N8gJLgycVTj?D1>)bv6B;c+SOP+n=AQ`g*4z
zQj^u;fFLXP$<6bwsF$Kg7bP)eJmnP_#DmxEhs<NmfR<vcZg^J~=GdA&6~3gFOAsFT
zIkds~d|V?M;7{aNRx-o+dKRI*y*<1c3G{5I0;lh94NZdhiE!?s!f8&nrR|aArm?fK
zTk&gE&Is1^4p;0&D4}*})KXRk^MPGS-$iiC+^9J|S+IBy`vtX2fXK7;we`o0*i+xh
z6Q$Gsv!vkjpxFy<>yNlpQe1)+pLFnAgyMW-ng}sg`VAaQKauY>lnL3<v@Nf6Rh>SI
z4c{Tb0#}0}HYw-|p-+G{Dv1r_-^RK1k`FbgFJ25nb}jK#chXI~fVQ7CL1x)wyE8Ku
zw0G{{2Mi7M4fppe)$Xv<D$A)7myqEHz*%j)ml>f@lK{+m_?cV3`RI9|P%R_ogg<1)
zZn&*AiQ%LuWSaga(a?AQV8Z@8+VtsSYw*29R(XYY88%#)EAN%NgYQ1U4^V#0h5(Y^
zH!x&UTnqGAkaPq(6!=s|XLBy`<UY54GkJ#)LrhajQCw5%(dS#AL@3tQjXTC;+msXv
zWkT-0sbT#UaCp^*!^lOS`+S0uHQv0NW5{roM!2`No~o!%6I0VmEG4Q)rerl5(~9LQ
z26|8wui(?0z73(xCZ@j&58dC~RbnfB9w!0Uqs&n5b7RvC&TkQe#sb4X>YW7?1StMd
zW88YW(PGqm7rBG_846GTd%`NofPR7o=;rY)l={_&!s~cUEYG9FlCd@?h}}`SNMzxe
zgg(ba1CQ>FjkVvIHjC->>|Sl*bv|@Yy<|CiM+{%gv3`?HZ|h<6kB~sy@BDqJc^iYc
z%w!^0&g9YU;J6!}Mgpe8nkS1MNzN}WV2OOtP59$h^O;P=y;7egiUh0b$;UJpm*k~q
zVF)JwN?gZF?rSDhmX`@@G-BD6X*F%EIH0CFnhBc86CcQ|nP^{fM_$zfnQL=OQx(X3
zNqBG)Z@&D{H1ob2W`{svkF6ps%vcNNy#q&qjk(2r2sUzRj9$_!;G&~@-%hV<cl<Z!
z>+1z^G~G%n^n-I8E@IcQS8^*2YfrT-{eUHP8U9LrCz-q%ge%aZJ%7&7IG3;Ve>8n%
zR8?Kq_67t*Km?>yKt!ZPx+El|L!_moL8Kc&rAtHvB%~XI25F?byQRDPoA>jL*KvM0
z&d{?@?6uaM*SzAEX-BwS>6=|8Lf(8WD2qaIaE>VX{8Be3Z`Ybxo4;9Gx$^T9XH^IT
z`U8{5N{iB}4%%^nPDycUD!3~?nYTUIRa-_$G4>eI=(t>eveD4|$l!@h|J$p@d*B0n
zP%(SZcS(?Q(6^A4$7}7ev(pBOgWKB^D%vj!bH00@O*a1!H29p#Nh)uJSp}B}mOyR$
zVKZ{2(QzT+r=<R3&zyTL|1>1BvVJL28P@WC)@>pZhoPaNQtKK2P*$Ram7vGqJrU3G
z`bLxt2$1H_x&#5)tn}V@?oG1MwUssM&x2O+1?xWaE`0WkC*#c!PmLSZrP)mtEtBDr
zW)#KzI*pXql2kw?I<N_tS|4%Nuo1L?FwNPJ#kx5{K=as>`1h_Fv-U~Ehx2t)%P&wu
zNE}xusSzfVJwOaDJYUDe#6WOdMy&KP*oLG0N^W?ioZVsZn}qI~cA4#?wLZn0^>H_$
z+dkXg8>NZoDwg;zX|S0Q5Wm$2cPx^-it(H@$kFz=U+5pO&fVIiXd-BvaLEz5LIDLs
zLg}V<MQ4PLYHDWg*@l~H!*2$3DGjxy8_6O#2p?s*Bm<qjz}t(?`}d3c*XA7b>;D#1
zy<pRN*|cP7Y0e<z^R9Rc#rx$w&ygB`ieZ-R{-0jJrY^h;;4Q-C^Q3*VheLxfG5^!A
zvOjOV5~D664`?|(!vw5u5l+J7p0badqxD3{>(0aH&wF%BygPE2uFN7Z=XT56b90|i
zlwK;!n$#`4@$_azi4Y|(4rzS7v`nH~{gw)W&+sc|J(m5*X?q;YzaEU;Z6_*}j$XLA
zNm)-C0|ybWre6bj!+7#5DXF`8AfV!YMsDu6k_6}0ZztS+KTbgkhiR)98x^Oa^C+<L
z8p(S9#9LBbUg@}ySE5^%;ger#YS6&Hhpv2&1R*0s59K-W=g(hqb8<zL<lWCUntPt8
zV_$JtTA|C(&)SisqqCYn#n$fi^C8AR-kPa&h?S)4O5{iPdie=6NG4v`iy-XTWlr(6
z$5&nlImqgPzzTJFN(_HcC^}u`TliEdvz~O-Ij#@03TWr}ihyn0(?WM|k;gb6#TiF-
zPLbXcweyPKnYXF=ceB)tIW@%nltM{)75<eRXTh}7oSV5Uy;z^4lm?wT!tD1~2$HX^
z7AMwzJ5cGHpZ^WOoa~&puGN<6US!}H4z&1RzjF19$tiDVWwz%O7FrA^tSb4-F#OW$
z%w<mUjOy%?zr8uZY2I{Wk?Vyl2dN`Lkh;FUN(6gaU$!GT6wUH()Zp9>W(U*}4;=Jv
zAe1~h48-4i^<db>a5p&SX6V^^-7Ai}x5)Z5n69EcL)T_*(!CQ)8#wXww%K&yX@t#D
z>Vt}=W|j$`i4b*ByWz;-j!NrqQLN8SL=jmGRU>z^)o;bfLD)pJ3<DqK?TD5R!78XQ
z`B+MGgR|0+?3^6ArceF9ei;<eCO)HAYn{aS9Tp#-`;vu?y(rs^zuLXR$GO|=Zcm=R
zfGGFGk1-a3urvN5vA9=tH$@-vc#6=yy!P>LHkz(<*RMw_L9JiK=KXB9S&NQ}GE<P2
z7ZjD`XJ$JcE|x?K!Quz9FrUAg$gut2?@N5KuB@XIVjVIH?w?slcribI1-GOO=+wy5
z%4GCISK#DS!W^()_HDdW%xwc#z<qA-$-y!3dm@kTH=m^Sr2~bP!%?{w=o>FIb~6>p
zDUO-0mis|`Aut{fTS3uO_5S(|Tm(oYKG;Nb1hB0}NYYsh9^a@fEu|Lp2P6;sdTu2b
zDhu4-Z)~TSH=Gyo@VWovv|AlGYU=5wEk0DS_is)9d2JEUzDL?yb_^WPFGb_Z+Al=B
z+D|a?d{m7Rq!N{1YQC7>sf2ILYQtR(im~hB(Tf2s_#sSLfW==ZZF0dBRCh_Wk0!46
zTQV&_EWl#~NFq8*+*a`lSr0$Q#9T+>)cMF+Ywrv|Hx0oB!Tj;W#VP+=2l41}^24g#
zYm4k(lx3^FUgOiEz4ZNu7FISZ8Z6r02g_HF>F(Ki_g|wQAnXu!-ka>nFgASsYtYBr
zNOiA|(Co_O)aQ0HcpZw$xLrtrByl<<=zJpMq-=IPOvfl_)NBg}YMB!{h==;^QzglI
zf=B!6Cp&e7LFL;^_T0NADd@pb3CDt7ZDs%e4Z*@BE}ot7g^R<l1gF5gw~E|8x8eK?
zjptZU`f{P^i~8nDbBASjX@;$#7Mr)B!S4Ns>rK_-+1eX4*S@uS?};yzV6tDBN65TA
zdMJtu>bASV<U?;sf?$fe!tQN!`7M@x|1&1STlB4Y;d<rY8jaLs+DX`iGryofzCp<W
zFC&H1bv+>Xlj8#P(jLM0qU27*MLvkt8(nD!hRxibLi_S7uvMkePQZM$W2S&`B<Q};
zt+|$OFRz_}&d-;)*2MT*R8c|oinS&(qH#Wu&y;@Sf-~iNyAhYa5|!mz9F!su3CvwL
zVFf%ppQ`FwO!^k3N7v+c-*%kEFfj%RWp?RD?(^Fnm*Tcg8Dd{C0P;yfb^fONpbhYo
zD|gF7T7nQD-Lam#nfRmtJU&ynYgY&9n{;VgN(tPyuWD)b2`*<Eu9Iyf|J&*Q`?0SD
zAqNayN*7IdvsUkpsXgGqK<w|GA!Yg4tWL)DW5(C}&#oLFBnEFd0*kGz2$Q$E#N&GI
zKc2Ea-)Y$zpd-?FG%N%m5tP?>D@RvMKw7ADSp^JZ$0M5P)(4my9SR*h`Oq6thVdc%
zl>2_$XPdpUiG{NIvM%eN>eWvST`X+kUTV(>k7fU8tI$YZXkl7n-%Cad6Qi)?k9>B5
zqM2Wks(pCL*fDkSrIPic>Vw_Q9Cd9itq&Tc+0DITwH~qI|6X7FFxV*j*u4{b1I$1`
zPSU3D>FIszdakf0RdE#|&Kz^J##=ot<>fV%hUt8@nffF=p4{8iG$y5Ri)4V``|P5*
zV+7^ZX8Zj@o5Wrz7L3R4C$A7bw|}U$r7!4CfAUdZYwuW{!iOpG={;^XsILu7GyM4#
z2G*hFE(UIRjCz74?2n3Myf2UJV)gK0&P};tvn~1Ta&moBcLGefYo2>um#w29ugr`y
za&qwDMk;zR)RE@l{%klP0!|ObYvF^fA%A%~6fx_`g-`u6v$)43_3A42+rz6+ulO`~
z<$@0GY%e-4#y{1l40%_da=bBppPNOOY`F<0Zt@DDO$4&fo(->;J^ZzfyZn;SxN*2(
z14)4hQsAeLIdQ*3^HpZ51}3}R$dwPOcvPT{_7Tmz`Frtuz_V&L5)c?286Jkd%oFHn
zu0pnA4k&6FsuuaWfyM4+)Iz~6Pw<~<vk(ISt1{X35uZCl^<i21-GJt|eAf5#>Xt{~
z^j6Ked-|#q6yrbyUtfE#V9zb2566Kt0|B44^B<$Z;b9Os!ZeDA2;+kG<YF8MX3QVx
zwmAsy%cKn`w@!TQfyRj(!?<v6|Is66GMs0PbxdD~UtbY-NzPdpw}aPwv(fN?9l_v`
zhQ{tomVbB|&X)r{D+bRq`u*0#ze{E-4A9Mb%TMM1sFY!jvFv+7!N~L_xWqy`48xc$
zpN-&ov%taj$8<4<*}dmGFRK|9B(5rxJA@Fvvo%Kxn@gu}?27?rlfTt|Uj=DD@>OQ(
zTv&2o(w$|_`#&y#HO8)>;R5!l)RSgMI=#d5m2M<OZtjK~w_7<Z_oe+?YxiB*A7yHf
zF7hPlsHtjbW{9g=e2qm_2SGr!WelU=8~BR9#Arh6)Xq#~#IZB<yfB_QD`{BAqpWmu
zYMcjzr^K;eVnAHKWDLKT3=7*MR>F{G+!pcQOmLL?smJW=NyujO1842$OHk|qLnsXn
zfd5WeD2*s3eekFT@sL!bdBDHIUUR!$tUN0ME#uh57zc^1>_h4ZE558Y(3N2wiYvoK
z;He-<Bi^ql@5K#eWgo_g+cI00?1`~mZ&(*RTx2(Fc*0%dFs`bi(zd)m7KZ-*87oLb
z^}&5+=N;5`@oVG?3b?1tAx-n3WB>UwwJ>ws{M$eHRo|s3O6VzEu|;UI-VGWJ>w2Gk
zh=_@7lo`R+-Rg0?d&ASst@(1a+>R2LSn&0tGE#n*U_5(Y?Yr_zRqbaCcgXc_sL09y
z=9Jt7EHIdM-#~rW_<HEkcZdyo^yr%Pa+HMwQOzkX&c_L99<q{cK0GU$RdV2d@#P=1
z?*K~hRA;msMo{vs00q^Zva8dA(EoO_Id=&(?F0Q>S4)&(022bO#5+EXtaoHxP5Z3&
zeVT=--TGkVklmd5+2JaY{qa@uqB*c|fTjeBj1S)96WTi3U-f{e96>be@>EYDOzs}I
zA)G8^9r&P0r)B=t;`14XzRBl9+V=qq01aR#QJeO$o-<$k-fqbLO>`XwZ?E~||I_i1
z3UJKp7faHa=5j3Z$HTi4&B*BJ`Kf+ON=;&Fx;wM9^hHwIz1dxO75G4GHi_DN@uFx#
z)LCbCVfEjU*NlH#Enjzp3-j7aRV%oOrr2#iUS_8faCYB8?0|@hylNirg0F$?Fkz-8
zZ}}240`-NR<hHZpF@|qxcer3#A)4<_YqnDY+L9f(1K2LaFRjS69{zxMi`1^dnxyW0
zi)aBCYfNOF7xUvx*wKF6;e`~}co-liOWPATxKQhaCkeR#dFJ3=Wunytv~}crw1>)i
zYQN!V0LH8GB<~~a$Jgi%fUKjW6e^ZG%+v=~9Y@i7N@JeGRF-^Yomr9N!uH^->m9Xl
zrIY1$rTuy#xd=@_qj0$g8gj$QN$~iF)U#%=NXEe!_+ypTjQx+U8-0n0noLhqGoaB3
zr2je_U|uc{H;9hQR!Eo!oaSI-9~*GqaQeGf>v%q9-NQdT)e{J!o1@jlmw)(PZKgfI
z`fVmp&F@HCIbX#z3%X^W8UD*Z9z+h$OE*RdlOvg8<l0R7XHmS&<jEqAb^%1Q881!H
z)Uz?kp2;xwFil@_&`7F(N`WPVQg@L}kHLXMm*(lMDA(puwNRKR$fDQh{c~f^Wm5f|
z(3V1=KzF`$*7o&5!+D2+!wiZqhK9}41pW{`FQR3q`NOe?a}Lk6@iGXQr1b$Q=o@ZS
zJm!ColfRWda~Of;s`Kd<lF71AXG?J&cZlKZTke!F?~=b-&Jv+{=U|T%Ie2_n$+IKm
zzCM~aaK-tk{l*e9n?H7@S{u#CV%*$3`sIg*{f6Abx`7MHNgiTE-cEQ`tfwx$VUZ`R
zDqEGW;;~v3SzierZ@~L-r$I+o5w;gk{itIxBu$$))@n{$^VEtewn=ZCJ&j0Av;K$z
zyWv5Q{#!`dYRYx&w{ht;jtz?^LYkIDw#Y9&2rVg(HapZbI}M4v(pT4`6x{1B4s?5%
zL6hy;-ly*YZUpGQip#<@-r<<OruJ57^TQ~01@*BWaR)bi^Mg0ETu3In%R>lOjpyy*
zg{jlX-YtLmY&%`^KxD2$of+5!dIF_7)o-Z7p0Tq1yG_aYR6@dQ*RKMc3G=e^O@4kA
z<SBaU|C|$c3^+S+bzC649^l+9+{yH!ENle8h^g|!=2EYf6z_dAAMDo`ESwMD^we*B
z(?!6Z-2cp@Ctf5+3%w4KBR9r;7;01QATkX7fP81klLC|veZ=T5V>_@J87_Q@gS_+b
zEBrVe&*{|g6<?x(K5f`;Z0wKK1B^uH%YRe|gm1YWzV+euOA>iLN#V#s{}>yAY5>M~
zlNo9rkQ2y|bomkMhLc(N=1j2-uXa=HS=aH_=J)dV!sj?qR8tWfc*Ui^BBz3@NgPg2
zEYw{=ybY!n-+LY-Z=P^(3|2g6dgT`ulabuR0#Gcnqf;p$GIyG5>Rm>kN|00wP20jV
z+yovsO80dbF9MLFEQSVxa8Q#egoYtv%rVr0Zfp$PIOujZ>yz$%eiZ`a+FeI!MmJcF
zq^6i7I`cOQ1dkuox}zaD7+x1EbMyhY;$^S93=Y=1s$kX*pBK0gn7*LGAn8mXb7OzK
znYo!HTop%7flkE_>Xc3e1(=%oecj6HApo8{sB}^PlK=r$v$1KdYzpKkF3&eUpy@cf
zxc{Hfbn1^6+B>mGxeTaVtL_poN6Az=Zy_@QH~tXfI~^=)eNOoTR3KSnkHd{=Yg3@2
zfcE|D=m-iewxc;h7+-bW%2H<*cKg=={TMUzBb2(yL6fU*7U8##cAP{E2!Akfvj}lM
zB=fJAr{pETcLT0e&pUcvJE@e}T;LNyzXp6l6fp+kZQ(bI!--WgsT(#0mtM<X<4G@d
zj{zpDyzbpKep(f2e(zf?>D$?zAT{Zl_Y2T&q2%%YZ4$=t1%&yuJ28CC>0<D-W5hOv
zlpcHqL>w|ZB#&!9h}w)zOcxZym@W{S16iWCFM9n03G~5%i0xf?fh0ac9M%&xJ|E5|
zyVcy8PlJGCftv23t)cAK@7=A-uJM84@n5wrMWR~1GauB<*l$qrpN81a`v`LQo*qhi
zqe6u5Q)$Jk=Q5qwOlL>ClLX~vz)*lY9vpzQu3prwg+1uejjRQtzGZJzv{Le)>{uDj
zGKQ7bHc*&eIon(XKs#E(J$U8((3=znUM$3#BZDBiKz#oc;36^^v@vp~C>u-Hr5ukz
zN-2aywlM(qeO2=tym3NoVHm2A#6gabIto+{%}iiVRbJ1k%qG#+y#g@pct1|>Ws}}d
zXHefUBCqUU77N3mIZKfo8yT^ib^`{@ypi<eA0MD$f$sgXs<ga<Pt@&fWAekJt`|Iu
z+M@-6F2Dmg-X8zeIrD)2k=a;oSYY>b$#!qpk2hA&_pgn&^BPmf$IFt1u_SI!O;H8E
zd*?@xI$DrDIt72QoYQD&nNodG<(|8^EFWB5z2w8h$2d5l?m6|EjEyDveLSAGudhTo
zR_oB-^wS>8V8mv6EQe9tV`bHRj*proxqNw+^NEHj$pcJ}s`{0o-2oEMOaBcFylVT?
zgAFn@KM8TAZxw<1iGN&&mWQke>HX=?U2app+85av*d4Gb0*=zz(NsVmTR1O!GMV7`
zsfJkZ(t6TLOi9M+$0r@nBHlLHSO1Bu9+2^%Y?-OCT0qauwW?`qif%K3>p}AG)W1Z0
zl|Be!DTfRCdV6gPdj@Zb!|b>(dO0CJVYxe=$A0qbGW(@_zYaqVLF}`nPbq<Db<<BL
z*usRIEKlwsy-|a71c6#k{74;G<Z6W>k^0eXXp@q|OlgcAT^`!n3V+Auq^ZMdr0Dy2
z0=8-r2fLi3<Z1U2pD(+g#M3A3E12n7oY)U{Lpn8)of0m${Nv(JnOwD4O4ACvOFMjO
zUdyisO;8Yg`aPtU(&u;3kYmG^@13Z%XsR8iO){kEb8{D-ZhfO$(E|}3+X{|MA@9Sg
zt|?--_h&zzczpq-WBGVSR2ID5CtYvokeBTV#~Q`C)&)l$wh6Ck(uJ2$M0y-EXK%{h
zpu^kwgD*XNsQ0t&wG(}nh1AY7b>sCJyDCzjDe>-urE*LPn-?F5r1^xYdS*G4aSh1b
zEN5mnh3ed9Xy~!TW>IPmy2iEss+o~0+RhcT`BZ(6nYx9npBENuQrtp}ot}N~r0|Kx
z7^ke95IyeFpD36cK*htIUWz3wX2_-(#%b9Nn>dn`^s={q%jf(iR)MBt_Sg`P3seNi
zZws#AemIm+Hf5&B$@>^6&7uxhmP!H|&!6Xw{E^;&t;E|qkdvQVbFxY9n(jXZKfX2Q
zXfcH8!rW$K+NZcO^L0<S0*NA3>89QO%4!IS$d#X(g{{Sy$1I<Jo6rxB^@`{KVDt-b
z3waV=pwuY1^o}`7(?z|Z#KVq2d=qs>z`gcHz<WAoehA^-Kk_1qiD|jLCE&7mCw!sm
zm<Ie<5CrvIN<mKj(pz5JiG)~#6+WZ#f!4Xby(y0itrF`wcux%X@9%6U@XcB6O-r;S
z#K)J}X#e(XyJ$%L1RtID`hng5?so=8=C`|@eO`jR%Zp1kV@?oBgN_7BqJ}TT7J?;&
zRd>7yR*QX+M+9usD6_yYHDlAwP-X%p(%WRt`>ga<3#ET#>X)z5%*wWedi#}Y-RGL%
z;zsiXE*ruZwdMO(7B+e%-a6t&{eMdy6&eU-q-D5W?S<hPCim)q5e8)oEIQH)5$n<3
zr)=T3;qbZ;mug{@^a|3eaIDD>b$spI{RdMCYR($(i9-J+q8KQ44QfohEwoOjB_1d%
zq+XpKdA+ptu5Zu4ZjzKDtYP#xKS2{yl^bt-Fbyf0I=qE1BBO1#7x)Btysx98Sqlo_
zsp(2XCpESGtHuXadI<>aVx#nb;=__sZYy(hd6|`~=&b3Q5#epHG2WahF+DYk$p1DO
zrgKb-*#t7Cmfa9dJ2LW>uCHA1cbzgC!1GajBMfG-XOODz{%qd_fUH*8|7)*65HH1a
zy?;nY>5mtfkm7Alh*+Ja7q#VB@ot+qGFA@|0)J|bPhVZvkNozO3A_wGN2pKq?#bBm
zhY!z$j#o)P3LhTH$|^|X*-h85OJ!}({Bc@sr~t7NMd{X-oZp5)c<p^@<CLrg5r^E&
zkJ?TOn`v3>;*Io(nB;rsSI5Kqv39QIJ&<YsT%kkAeUFlEXt+R|%W+4HC#0Y#4;SgF
zHu&Hbd;|OnGo13ir1Bk83DlF|t{9vCY&GWjATQ!&EsChW(4gV#XM9^>IVP@v`vf_;
zLtGyliXdPXT=6?={NN|}Yt?TzZi%|Lzji=;Wf;IemdEHsnJez#3IE|**n}sVcB0Yp
z@S%!<k(olmLBmLL%Yp(W@7Z*knKz0Z0)%+0T~C+$Qq7fTXAgS3roHz4xLsfQI6O9n
z`;MnTC#N7sMP2uo;riC?zP0t$&g91nCvnf0mWU9JA3ifV(SO_8+JTGWG41yC+qiRg
z@v8BXh{LTgM#?**^s5xISi2vnxBg{5(`94l7MdU4JfF1tov;u%N7zz%3kwUxg}OCc
zTDq!?v`JFbTR;<ueo)?g-}91yu-3Y-8qbgh$llmU+BfM?QjxoeoLonE(nUuaN0b<;
z=<twRI!J-#WV_1WF7V7Km2M&laRZD&!GWK@#28e}ec)-yw#``i1P0qAq!$H8P0h_m
zTceIg^CfU4uyKb>8)>B^X=7FG;2Os{U17aMuuGcpO(9cbw4N+o-aeS1xyQ&@`D%;B
zLCu)D+v8YdV~%jU)K-GF^Ba!^teCsEmY&N`!fjT0C}(SF&FAsgM1yDvJhWSS^i!Yy
zbex-vAFm8hs-E~nNHkUDu-}}=3$$+YhK8-*nd3)gD86`e$)Hp9!w9L2U;BjaP`b@u
z3?JIttDszW>r)H``ib>)h0|riEFQj!L9dd<!?SSK_3?<z+oZO<t}7`hRYx>SqFS}?
zS|I)V<V*Uv-M%;^Ej!!X()}-Pk4N+U45rTSUK2&S$NeKiahy(cL6QPqSMAFyrN1k`
zzng{&t~j>!?gVy>{JHa?i6%mqjrYJfePl|Sg&6xrxOBjr#Dh?$jOTJDON&znl>sk~
z4Sb95LZiSB{sZ#Fj~2a75;2u}8~A&?-0kflFEoE4&`_a^HMYp@?|o0H<WPzQuGgOl
zL%M3_++12y-hH2gY`m77yqrSq(l?u#R7{;4tI-q*@x9`CnYq;_^}4uDC7GFwG_@n1
zbNKf5KksAzizg*pTwD_MJ~OJX)&dg)Nr`^9^gk2kW^nU|=5HkEAMaz|LXU0e6|<YG
z9t*zpcIZh7Ub5@{$X?^Q^Q6a#V`yG<P?Vve%5I<9a$gv6jGW%Dx~cL0>h$7C(H)zz
zlHy+D(x`Vy>F-Y#cWqY+|7}+FYN9+vez>2Co4iYHq&&x7$yIQ%)?eRTV83@y#Ai$X
zUT*K9j!W-*+DG(%T<a-?saP39HrGi`vLuVIF7FT%r>b5%z2L8slGkmB#5acZNZa{U
zz`m2Ih0WgxLBIl@eq)o(;5bz^+w&I2Qw5E&Y1<w6vAE;k1ZWmrprOEcrH!6y0~A86
zQ`d(RW!oY^WQYu8Kzc_8o^lCd1<59-E$r=K-!LeNOCU=Y_KsAH9qd;xDzXc_pp5?f
zZP>;T2u6sQOki$is<q+k{+4lZdHK&`v>aObb!FSgtAWZG@XAFyKlY+I`<Y%0y<2m8
zgS^;#IyyQ;8BB$*oT-p_9qs8GwV1xJDFp}i;v6`piAdLa9UET8n=;U?Iw|XDGq(p5
zzh(rcDlrjJvwd}L?wQ&m2|oTf90RQ;zmf`_U7O5meR(^GyXXmmV2}fkd71`6((bZN
zJ9<XO_%BIX>bkk9`6qz?Q?_hwVuS<-dnzb)b+@T$syQX|6E*(O(-{f)D1CB<2^$*m
z)x%#k-=E5=)Ejr>;4}t5m!lE#p0+^`R+c1=@0q&XRTj=OjIWoh5hmw~%4g3=(2K=J
zjFXgbAKom`O^Q}YKu0pUp6Rz8>+pxhJ&V?S$&TSG@cJ?}WeA=#OOPjwB<4~}rKR(E
zeS%N-?(0ZoE&WB^(2bs5cVpWJqOW#GDOi@P%)@T@MmOG8bmd=vx)!Zhs?#b-aMs?j
zbnByGn???A^j4;N2AXg1*SD>oH5(Y_yyd)#8J_gC=v}H)4%tMpE_PYQIXy^YZw+Z>
z<4&Hq8sh5sppp>+uTPrATNyR9l0ZZ&PY`d*C;R5TOlnJ`V`){JBV~)%Z!}0lOEx=E
z<o&i7bATXhVSJ{{Rzlhnr*0qZUg&iPG`x{kz?4$?@Zq1s{@Q-oht2gn_f#o$jOKne
zz4d;acj}g@gqv2HnxU#E-lgCIHu>gn>M<##)^EqLTSGf(pIaW3|Ghx&`XR;wkyKSw
zOlxPiY)|W_&vy)JEy+HJ4-XBK^ExSxwzU4uWK2(KGT|C2)U~!Y2cxWwpRP`}-cQSh
zun|a;0hSwZV!+}Brq&d6At-eXolOm$>7}J?7XvO`O|QYqiUMkcRRV&)&oIv1E5T$l
z)g?=!Z*Xw3{Z(LK?~0?WF}?p~m7_Kpj{AtV_ewl=PNR&RT)tMtdr1bsy7epO4)=zv
z&J86{{rt47kJ7Ph(atB3x?EFi-tAG6yBCIY)mjKRY@;F?LqmAgh3#!epR1((c<_>|
z#pYQ(Y5Quytf7hz=u<&5OhwtbIXSSsrHs`zdVQ?O;Y3W9|G7=OkrKvoG&v)mrr;c|
zaqJVKY)PwWdTsKW<Swy=MQ$4Ir<RS?+l?RVe%ymWSG~#b&`?KPd$s2wIr4=)&Mv#i
z%Gx~g>u2(le|iw!G{cQ`(9mKqpHLYVpZja-&O~4<yT;HH9xs+Y^i9~!|0*Retkb+V
ziRvBHR@Z%dbn?{dk%PBTb{62#{vKsPz2=l`+ZP$@*;!F-iwgnH4&{>~FNk&7j@O)u
zVBR}EF7WUnm=D}QxPpr9mLc5|g4|q=>j@8tD%;5|M6Ev$+@x%&ZTh?4BfavK->rIr
zDGMKojE=qD2|7-gJ)R1F_x}AyG_hKd7j)e5Wf49y=C0>-j<<>jbcB3vGhfAiJ;`S0
zG^C-2lR!#pYh=W|&5xfiT-q}n{BhV>m|tkBJ@vaQE)FvYQ)~w{E9}SdH^U_9?8|kF
zmvzf5a{&9CeRl~tg|JT>&lA{Sj$VlGe}C+4FRSJ7yd&Fw093=>i9+fx)$=nm#Tm<T
zxB<#6)T_AB_~U)k;@9%`)qdji5?<<mo|AM_asQw}Jj267!o%>+Nly)m*)G;8!}3E9
zRu?e0CVhTgx5^4rZ5m^`*o{^Svduk0p+6`oo&53obZ)^+i)DC>5L5VHw~)`O08+h!
ztE{P3kXz_>HBW@lVhJGGxw#eI>uh1qAf|^VMsDdrR7~Om%?Yr$&d&;}klyLkPZ0(w
zHx*oy&K;dfwi(8bHpe^X-ozG;GqJE#`y3PF$8*rNwuHX@RYp<mFcbCnZ)O|e6d*U1
zo?$5yIE3_Qw?4exo6LZoR>W;<e|C2txC^J~)!$WfvLZHpj^>A^{U4l7yxg9s;D;=~
zAKAd!i%&?T+5`d`K@Fp*+m^&VmwX)}(+n85B??SD>-`!>N;UciWIBzUKe3!8Y?P>`
zL_3TvAO`VUekbq`6_acC?g=%RrvE9)a05y1Z0YEa*A+j0?5yEVNJ&YtB@64T>4ey3
zpS{##`NnJ2(S4a=x+H;#Tm5FO^~Ayf{hKq`&z&9p6mmT{X#Op2!97Q4#+D@FJqu+c
zM1Aye?w+20@W(i!=-n24qW3`BgXBSg-g)@o5O1gpg_Fs3IC=+2?7+(4NWqHNW+ov)
z^Od3P+qr|NUA2F->%MSGtF<ncJc?y%z0OEbU?5N_<9|EP65Ao{vbQs-rN(Y>jOuF|
zBAmYRX`N^Lb9BsDv8`#(NK%sVX7GJVHnXn3R3^9Fy7r_T%#>=~7kChs_-$L>CTDma
z!pC4|8L#phXg}0JZ%Tvwh_lfvwdF$S>EaW}Xr{}IH!~A|?gl~J5!_ZDM-3z@qT%!(
zt#b4JEiR(eeZkFJTr$-xn31Es<7}%}>wcOl(b9v6f$__$iK}yG+rol8L8mP)u*b$K
zW`bEHm&3TEi1fnzBB-{|&BgC$Wo}RXTn#F-Q+>+}Ub;EqPzZTkaUrCeTyIxX-GGz@
zD<=sUTS1K;lO*|Ko@9FBMo|G(<>jMY6UXc86ekj-2xAz{@AEpy?&vcPaHbbWfiRdc
z$eAto>>|E*Z*4j;VfE7|s9yaGTNdD50X<E-*3?D&*}w8Wwyf~2WPUrWPns86)-%=N
z^(Of*tE9BG*ZSkqh%_`-KBs2naM~7kJn5-=G-8wZ#H;^mO6RYFv)Ruv#=jr)zOd@T
zE7|H_L4Alhainw+kAu*0Ya3W$dlzfRuRSSoLVOKJfBJi*)--_(aJX0O_~g4V$i!vp
zMqwX?@YTOI9!*V>yahUjDOn+G_oPSh^r|e|#=``EIqYZ3lX~p6PHRy=M?YE4?0c_7
zGv%{fPY6XH+<ScGW)n@T`@N_ZhPJ@H^4uAXgV@Laez2+b`#R2Bn7(dX1FT(LJlms6
z3<S`cjM=cF*I=FA-5r}8Y#I?|di;n2VXmFxIXOY^fB#{*qG#pBsW`_+9P(a`1TZ)6
z{>CHwOnVKwWulm&Cr@l-|66Ws_f|JPSP>8yL0^jE>+|y8RxPS2dPOZwn<yT{PuG`T
z+EYdZJ+D6Ac^D_J&EfJQ>mFym$7=QMky=lXX`}jnjAuT}$@qR7;deVle4Tqf&PckO
znxvy33f-u*4<GJsH0OOCeg6f_TE4tMR<NA-+&`yXub1^0oC*7^W@Xcg8lCjo+onNV
z04{Af7=-7}_ho4BKy(7saSm_y(H*!i1a{vf_f6jsGgRSZZg_8DgSqhK3o%;+P_<a^
zu}!zXdPSO<pN=4@sdQGFY)=j0dqz?!DohXF^Js-;{(J4{k#kkB=6&OXIJJe{F}$DS
z?{kl{6mH$d^0x~RR^Nn$&~2-x@`^1sAv&Sw6$f-#h?tTzHqqQR`_kY3kA_momfBY0
zx5QEEL?1ByVO0A@m&sTduN-v)`PhGU5s5^q$Mi>=$28xKDj=hhWks?r3owwob8Rd8
zBnyRTNJ0ySZ+5>qLQB|+^h(zg#9))98ScJBXjxf^)TpmH4z;L$*Tf9mLUb?e`nsYI
zS_*N;=3@y*>#T3Kd$i1QnNcXk++B-mNQ*DOg}`T}&xuh@)F0B;>s5Z)V>zpN_oLyn
z{xbmPWHoZa<8Py;ic4+fnWH0gck!`uT^|M1&;LX>dlk3jr_`*HH)Er}Z~Hqaw2v~H
zjQaKO_cDI+#VK3q`RRe5w@pk<C8eYw^zzoNkdtTmQE#I|i@A|El@!f4=jV-N(9z;L
zS6;X(ynNZ(+*S|OdCYp*r#1!$wdkK|iBa?E2qfp=ly1@-alT#64O3?xYHx7dFegD-
z9y%IGgjbpuRM+^7AQoxhb%<IQy(|)V|0%EG!yL~{QZ!%Av~S)WjQsR)3#-=_Tt)w_
zax(vnCap-tZ-2Xns!LUMAYk-1?I#)Jl}SoKFg-nMOjKN0E{RljYX3j=n{A?~h=09U
znJ+Bqrwg#=!EIFgY-Z7%TEwgVdct-`tm<AV(=w-hP-`}awxDXwRCwB28OP(*I$PQX
zaR~*YhrDfa&+E$${TNwvHT3ouD|&vakpR;y1Tv2u?*-VRf?MvL@4h%^DOV8id>d(v
zr2A<2i>u!=oRN@KXKjXEXG%{45=$5V5dm44Gfv+0z7$U5bby!kZzIm1R!$EnTYlFl
zJy(-pU3-qxj<x2oIj++6ri0=CxBvnVc$2&4kUFSOrbWB~n7BZlZtZ9r$D`n?xA}ZC
z?1hF#nEDVm_h7GP{8}!!@$y%=)&A{+e*!BVY~txsQ{8m+J&jZ$H`d!Qs$|Ka00>kK
z%iJnyJFeA})qb~^g2ThW67O8lYgK7(gr6L~kPjc4Wrlg#=JFzDtl<1Ae_y0aO@i3E
zPf6d3_AZ+FN=lZZ^vhw7oSghoUT(B%O_i6WDaR+(e(z!?ea)FjSS(As0`D=AyKbAu
z;<(V<!6HxltB{vxmz`%<?z%J#`}7^@v}{^U2uO#qztL}xdAD~h3m<%fL&Kh@V_{<k
zy#F;uX;K0#PImpDHTl(kUH@*>;Kqr%WWK9m1ItVba)$d11ktMSo~}8?9|~MH&Hd3V
z^!Rhl_+oiYh>MnqR4?=#;vBh+%AvTFRx&tPhvX2gFs5b5Pz@!1({h3fkfAA;)*&u#
zjK+|iga|3w3Vrx9m8X!z2D#Ndyk*t5Zti(pX69|OpvHEDga>tWcIQB7F##1x>`sg%
zgIr)pS#dG=*};QO|F`hnr6qtfJK7&<v3yxh4Dc+uiSEXKJ3#HFZvW`u-6PKaF&k!4
z3lslLf0>&R9grANwv~b%=3DoQRly0lga19OxJ;+!wS$zY20UcUPnv$F`IPiHV@aV`
ztc<=NuN0tcDchKg_ywmgl<Hgjt~doW54v^TH}_?p>fg=@a=s;vSp{ZXl8Vya;iq?w
z+ANxOMo*f`w#C5~i(?cD_eyfX@v#>6$c}}ekPuu70rRtO&4Kd#9&_VE*bgF-E9zc5
zXlrYaS@Q%*TC$fgAk#HhFvcG4SBJZ!g`;z5Y<I0|0<|GFR#VbDoNQBwz@FRq8NauJ
z+|%FtUsV)&g|;&?%uwrw5?sLE1At^eHBiL7B+@>_wWX6$i^?5LdK^GZNosy|!}ejq
zyUQ~X&-sGPTq8qUf&VN|wVejmr>ChBl#>25&Aq84kAN2!cAXFFY={5XX_0=3B#1n%
z-}d>gfS1fS0jZv&gX(%}6y($vCMgko<kYcX`|DTA=RzZrz&)*?s+yYm%Q8RD0s6C*
zQCFngbIRepzn<o}$wgAN^Fa7oJUNCx*(pCG|C6a!<!h_A%pVPH3l#;AsM%N_v9mt%
zp*m1hlBV$D8gJh1{?iiJvRi%LJJwed5hG<z@<kcNx6Oq0Woc%4dT7Ve*%>D5H(nhB
zB}||qv2C~Q^LTIOl=XazvuU|kRbQ%qOr1)7zVp+i%!OShGN5JQ-@hf7udn+`0X?Ru
z|7pt!m#AM@zTk0kc3iPD4qN;KwGQP+yf`5TZ%j^gboA3tBX)4d!{bZ-B(3Z4*)_2V
zfvnW(w&lh(J`_@PZDarb5u24gOmd+Xh_DvCvqxmvTS~pWKH@lqS?TUyYU>%LIvCO`
zaUkk*Jx<7Y6FH(1z=zau!MfMQA;yaOBJ;adAKoPu9via*IMMeL>tW7n*|z9y9B-fF
ze^RI@5*72it~1%h#aoWiWZwS-BeP^)_(6){Daj2Y)BBP^bY8)?KQwcoB7q?x1zI)b
zmkyOCye;0ab(pd1dxf5C@EN(vX;<b36I7~8pK7YS_$Fv|<}yZqM}=@_eAVS39d!a#
z;=wxw6I_}M(c^B<`wtoB^7$LsA28K^uao?Hr;$UQ!d$TZ`oiqex$Ka*?qo)B(80k{
zTEDf=<P=5bhU$6Iog;67U+nYQCDqOE>IGf-?8YBOBz-9hX~(C->f&_}rF2?z+L}nI
zc0qpBS$8WDtrhj7{4Kh}77?;H(-<s%pTUs9TYc(32E*fTg9bmh!-`xQf<?U8!nBUP
zC-OE%>eW2aal{&q(@1N&ux^CbUEji8Ou{uXzs5Rn-9W1w#?8}9l&<mKXS@GISb#?a
zP<`kKTgk0rOWqp>9N9vpbn}*X@~3!ed?G4*qCfVVw`#r!`>?zzpF9$+eWX~W82D_x
zw;m1S=zP7mgSR&9V^V46UnO-(mpZ|;@0_R@Nc4m7A#>@4+?<CJc$vdA^i4JRU6Au-
zX=nPvV0+sgMYPzi%1?LV2?ha|IOK$_D7<l4!lb`5!R1?d8F8eheD7*raGZ$w#A~$h
zQrgrsoy4vD5mRfy^p8vSn(gzVaZRZS`PSsbaa@wYK&se+txk+uyvv&i@O23@dnA-B
zt=4N8Gr&6j*2%3#4$78_$?bki9p6e@X#n51rS(zX)f*Vi!x;d8JMqcCsVSY<6B3MS
zJN!NOFK#Gf1B1$sSRdixM*N`RS#*%s-8ugOj{Zlu3@KTc%7s4T#T|C_MGoBH-^<J_
zr`eq>CTg0Vu?8joq2Hnhif`k;J07zy_RGZd6wr}yO7q8G#WR%Q1V~>wh#wy}W4{Cd
z`=I^Y+f!WC9CV|~KVCUd>#~t>`tgN7CdGl_nE(>oGB7+iTH}_O{{BWl<Co~@CHB|Z
zG$&_HM&a+|zJKHUm6e;D`{(E#7^_w1<J6*Rd?xB>2f$EE7SO=YH2LYpl$HKpzphkj
z{C+a1!A}_;TI}d5neh2Fe3bf9G=1T!<gcG8@xduMZ=?YA30j4&t%NZlvg68k>vZ8u
zA)0>mtslR*_xEdPk|XaPJ>;U@VKs^sm2U9{53t?VNBmEO*BnC2ibQjjP-M}>+}skL
z(D4L>Rfc5;-u!6jsm;JDyzle<@Xg`BM4N=+^2*TAFeMyMLW0Bl7|04_jm{fE&z2xq
zWmD9)uv^SMg#HG;Ov}>`QWweSQq5ikEJ#XB{Z04tX}854m+eORP#g%T^8#R1OhJmQ
z@dr%z(&nROByWAlPt6ZylA#Vio~)F%vC&iBIkz{~Rs2?AVqq&sNId^&%=trX5vsGb
zmdoCxDO>FN{K}iN3!(#EJw2y}mzn^M_bI91$AcaIytymi_Ef+RMU6t`B$O0LwlN%4
z2AgkXD;v`tR(T>~B5Z8zHks!!F`AA~sNp*WUD?^olC}b<7HM<0NW^0A%=~bVgH0o#
zF)Aho%ahPA$)Q3>guc%3_uK>vI~&{I5ux1N!kpaPf6jdaz3#ya<-a~Pz)CbRG2x>G
zH8yHLDG4zFU36RG(Mwz>4LFY{-d*-8$;x8EnTtY+8`}YDwdtR3iRI;da7#qqS@ytL
zdRuY9|7FTHTtnpr%V}oM%Sl`EQ-5<SHLK?bYA>E?KbJs5Ru>nN>1r>sA9%vd1hC-F
zB+mN+JoW6fE7IXLU_;($kA8Zc#2J5NZeb%eQJ(pak?R8v0^I~s_xkt1<6kh5UD|fx
zL}urzyI*4Xc1vpG95_KGwZDgLefAO+QN90+nA0chFeC^};$WCXv9?a<_ARMNP5Xtu
zsw$81S`(St11{1~@C5V$Y`CjqWMl{*|N9f(GMLqYQ#J^<Lh~sRa2+P3yf-dE7}Rw8
zFr~1NnL*N&<11Keo3nq8ic<h9mHYRbnwy}(g6SiesC~q6hXmb6BeviB%pXK-42=x@
z<;c&-7;5jwnL5bKGz|<4OrZ}$R@T?dEpp+)haQ(0h;E{E>IKwJXkrq^1CdWJKfHer
zkpgcmEAQ%k62C`(o(N7n+4w3t$9Gmj+t06dBGQf7nrGe&Q*z2p*pDSRi8I{Ow0VH6
zq`f-jUNDt&n!x2Ja9z!NH56UEw}w5c%M&Mar3zdQcU1x8Rhl`~R}%TUo2`SkZVpoR
zYC-W|nH?2mk-92BJmkmZ+c^{eT@kG-;f=>To2~WhEIUT~ZqjLrwViv{%A~fm8CH3d
zwE<3FH4^-}b~0Fqk%J7!p(YK}_Zhj`RsbfmEOm|PI?HQ$DlyBwAsZ)O_XxU2-QAJh
zmTNq`O;e4A<Z9<C*dAg(#Iaf`?yq#-2;0zvT9Hc&H0!F$FDC~l<-aR)KN&2)PC;df
zZtLt~PkAn9t6hbP7#R`1D4aQaxEJy&JTs%@344*4tcoERmLqre?s=XkNhH&WR#gbx
zd+Nb`;|-;fqOAXRvs-)nqfu*~IC({d4qv-HmS-Q`Vcf-07kTv>)+Iy>&KTn-XI!#M
z4UY_Uw{|Kj^f;1(3>eB<1=)_Tf>xG32_nxh+~H1@kqb<-z8m|ww3O>E(b7{Gz>W-E
z5fCQFCjp=kyl_(JCC!|dJCd)PpOqHIMnbf>6lt!i?~M`>P)1VB{*bk!we!oDRC<SM
z;rE%9+L#ITG92j@70uqGujry$UvWTNZDKn9MJ05HinBhz_g)AwI&1>~(qV@LPiWuq
z>OIuVWtIc#i^Kjc6yJ`ey)Z1q7gVXviu(Hc3GMoCwFQa^Pd9Ge`%Z)M3XoXIw;5)y
zp=zKcI8#BS{SCjuzqQnUW~(nU{<@4c${b~_vJ*n`U_HT1R3VP<OS92`&?cO^^p7Xn
z8*r$~9QOCZa1JYU*|eJ`%tZK~6y)S+Y00WcY2`?>W`C!qNdEfi%TG9hL&e07s?mQ(
zg1McNrXc-eL0)#60{@3OWWqU(J`cPT)8^ik$HnnWD_uIZMJ%TD-npbFNv6CV+p@4J
z>Y3`bCL{Tyy4t93??3IXR{<+ys1ucNyM0?bE~?_W4aT3E%dUWoiq#~l1ejnR@C!g|
zAyVU!czsI`an7xvr4!$;x*YcUYeFK8DWcFg_ohKaO2qx_6Ag~n#P_wpCfim~!+YOB
zW(d>)8Ug3Z!Wu*HVgmBO#Z91Bq@`!~wzN+gj`p?@wsh~kWUedWpSI1KXe7?h&DF28
zO#4`AN%JTK+ULJv&T+LSPvBj@tvJ7(Snk;MyC|j^ij6EqYTgcGGOoBjss58P7EjD#
zoP9E76Z}8Y{OjooyGQcOQIhj5cPWxVpx|jNaarTVQkHLUx+QTk7BT|Sc8G43Th^bC
z4kcn$YLvGztz1x~Xzz^95KqAYVXG^kko8;uhlb#YlRyJKzBeLSU}msTg}Ek*oG;>m
z82&$LA@8HTh8+53_8H2}`IC7Kv9kFpxLb=$xiBV<>}+lS+JD=&3G0{>@mW27xQ8vu
zKrV^+?(f+CmaO9#1X-2yxu7BHyH@tMnhQ~lb2Yf9b<1i<O3te0-9#8^cP9(HJ)ce!
z#*BEv!?QVAc~xYKxMt}Eb}W=GpSPJ1_}~<aV&mVuA^92l@<hJs<GMbz?}qrW6@mfA
zHL+5x@R!CpsO;$mY;WR*`2}`d@eEJ!6pWRwi^=QKg*5NEoyFGSPdDplUk)hhHD2xC
z>CF_Pls=1~+S&4YdD0>SM1cg67cI9pYgV1E+Z>Wb^}>CqZr+g3ukt^rB?Phr+!af^
zt`@6WeBpUakAT%uQVNM40z6kuaUq~+GU38PdKqw#epN?K&OSiS0cf_g$Q>CR?26|k
zbCc4Dj3p!{$j#3c@wy6*&wcBpz`FO1%efdj9+@6NJWiavf;=d=?hxUVS=eFLw)erg
z`j2OM#^Xd?Nm_8l<|YD|MAP2~+FD<Q-Oo~E>SqluCTp(0iqQf!4R2%S7~Tdj!xpBV
z9hQ{l=jTH9FCqT-hwUAZbO(18wUBqs_D&F%Z6SVg8knl#;vS5j|5&-Jr)8F<%Jg6F
z=6&>bq=sJYK}1Ifd-?HBTxERla>@SPTVq3}iwkD#c+`TzrVXPhv&R->T8i`mz(|wD
z3c_y;z(B8^sX3p0|FaO72uexkU}$T==av2|mwcC9CcUsceQ<wE0{Kx!k)N9yYWOu|
zAaEa8K~r9mw0FKyZL079K;I=F3gCeArl>%}^p;xFa$VWI@Y{`mT0@QyGo3ilMZs%6
zKlm2`m&j0KDaRL178^DLvTFO#QU|Kv-qq8jUJvz5LV18mrgMGRIEXRwm?D>~1V*I@
z$Iz>WxoowFfu!~4tNkAs@T8LJ5MrGJnb3^f68F;U)OT_`BZPKnb*PVBkgujfhmHr$
z6K!}f@#JnAityr(R*yOB-0XJ!!-*Q4!kmeE$BDJ?chb-s&Mk5}N$O0I#B|Xu!|{RY
zGgL$@yv&Ya8taZ%8eQgwEf*DnlYVg1S1PvI7H=YpiEgIVa(vQ<HCPSCml_kRvX#6E
zE&R4+wacGqlVp!gEx4K$GBUcBm{aA#<Bud!?W0bs(V|+GF<A6W|18>Eoqzhr<)9Uf
zBFDoQP7^A08z~7DK+FxxC<FO05PuDf+bSi^=)btL^}1X1Vp4(~Ty=>!GlI^H?I~T{
z1#0puJfGK={Ck_3Gqmc|>F!k-&LJB~Xyx=o0}`$QF9JT;e{-o{H+b7E-bY48`W9>y
z-4lF(7uOmZc6|%Uu^`KCjBud<!`6~lt01x>J0??OXljm>kkCuCy&=rS^*gp;Iz2a)
z+iKi|T!GZdRhY`=stnxxK<;j9Yx&GsqvBhwZ8!y}A}NJz4PX12ct`cINe39h00k4h
z7FtZz7pmpgCvGrntVv7=pwFArb9^6(H@C8N$HoAnWMaR57LWb>#=*&(@S>)?7zgPk
zI?wY>$zFL0X0j^X^tyVKmj<<7r_5q6USwueEbab|H8M(l_VEninC?1MTa2YhI_AKk
zC&1#h6E(NCuJpRB4kTBoYct$1k7eiFg`Pi3CjG*%0SJ7|?Q$})X7lBJFqaT@+Ip+!
zC1Grw1{e{<qxB_Pc}|c(=LA9r(G-K^g@CIsob5Hk#BgFl><SoLY>tdD(=zexIdtZ)
zMiciWqv<IXph5~jJ`PficbT1@J-feF@-1H;P$T6X4_-|(6a(B+s@VmSm3?^c&zNgd
zvNOAwo+cC@uGh^15Ed91%t03FTUU1X7JA%`wOe5C1YfIBYHHpWc5wkY_&W`~2(*6X
zmSYXRb+hejmM6}22S>Znk#{s-sFNDn@!Ezyx77ao;Sn#8RQ9lgKYdaISRY0rnC}?C
z+>>|a4<5ymP(x|0f{mWuQ^TmpI~r12L1MI_R12O1>sR&0X5hz9L6Q7YUnM6a$M#Ag
zKA~lyV4^zj=g-x@O9R5z><YLgz&2qZu<%u&LH&qc<A#}tCE!%&TBZWuhp@-0yO)H7
zqU}>a6!Js4uiqZjelS=z|D@Lc@iw`#mPQzk2JGl*TF-X;-YxL$nlXfXGkaRPV|+B6
zH!=#m{rU6f$e1tBCK%C#6ZhJ4C#x^uky3Wgdm68)SrYv#?pE#{u$2Ix85!VJJk^&~
zc}G(m_H8ZSZe(y6cyDbT?0nUB((r<+U!E(<KJFcn;>km)>lG_1g|xCgY-}ieAO<#~
zB~6(L!@V+4diz#z@E%-RDfxVOh!@QE9n=;&?T)Ml)iwAi6(Im1!XU25Cvs;0d_0~!
zNLDY`<l4`eT92wAK7Je$d8#kOuSe`=PK0BkqS%VE)zu@jl$ikMMR{5e4;MRqt|vSj
zgUe@S>4Hkt-)p{q|JgU#bTj7{9);1wcmJ`2pWHm$JJshuCJOZJE#gxBgenzpAoOU$
zjzLcu?y#_S8Z8hKMytYZ`QMsjKjQP))rPNE#XjfTHyrKKpc|;76nggnB<mW&!n%X%
zTajbG$BXX^8M|QE6<{dwaRPVR<A;JPvmkFo{VN&Sz5%XZ|Iz$1GjkiZ3HYuY%!2Ya
zCJ#DUcr~-0mR^FibkxR`E%v-!^iBF7J!9jAJYU*$S!_i1p#|W?7uH`B;x42p=3F(k
zxOxYq{w^OnIz<jU%4uW7$U#H$)Vi`FM+ItMg7_6nA;zrqoE)b8XdX_U{M^Ei$pK)u
z@#9DD!ZJG<4#YHYJ?0F)p!C;w8XCBT9m3i<F--87kly0qL+;`+N}i8~D}M1Hm<F>P
z4{mTcadJA8@sZQjmDkiE#A4(8FPxcK&6r+&%~0Zc?PuFV3uZaFi+-^C)Odl#oCqc9
z3ui@1b#7*S(dVdgQ3lG{S$SdP4U?TGB)MB<^!)b~G){|D6H2YrD3?*D?cK9fbATQO
z(^XjOwM|W$>MzD_|5#lz_kLgfO};0>?3&=FN{(YCgJ;#RG%RfX;}U~K&IB0-?f3-6
z1feTng|mC1MV8pv8TgpM{@iM^*Uh@lu3Is;xuC=ty=Y=;rkF5FoE6!7)MZsPx#PY!
z86eKs(b3VJ{Fo3qn>ZNzW8)7jVHxq>_I_)aslt7L^qO-E=&8zNZ*obbZ$8j?^N+Vs
zuf}Gi9(GCCdOTdLk9M&==hl5zbC*j(9(Y~)N4}tCrs{v<Ol=_w3`gM!UMd0azoI*=
z?&g-(<*%k$fO3JbXmJF{)199D5pmt%{vdAeNLp&N>SrtU3ri9;kuB&X=l09eQGdA%
z;4f_oIs%ATSs}o##eumw9?N<}8!c0V@-!ek8M@gpiiKaE_7sLc-5Qzg?RV>Vw}I;Z
zj4}GPVgk3_5jRdFAX1T{jOvA7hF=^Ej4IPiDGH8H2)iDH1}u2xcSdnR3>m4j$#wT{
zlA=H7cc`x(1~al$d;V+zb__y$?_P~V#~J;DZm^hu-;~v0!+teB>rP36r8J+JkM+~p
z;4-&++k2v|{U#ej0jR#E>6y7-xiG$_WeSJMQxq&*<nXntPE#U-_XklKMJC8A&3dvL
zkyp&dj@@&lT*nivl!Kd#7QgbQZRiMF6V40Bas_T8j;rG+gvmB|GiB=w_YjWAe%!dd
zc10TwBrwuqk6VjVE}^uBNlP+`@sS)ciZ7bK3xh_6C-dnyz1a1oxBNvluZF4=I%W!K
z^48Dj2H5jg2$4uE2DkW)-=VSp^~Pa*<iK{y3pc})q!uN0(+y-Wxo_9KY>vG8`5{2b
z8Hp2%w-G9Ma9@1%)$cviF9zxsp~2bxtji%X6&BkmHj<&P_vijqk952b99;bv9X_%7
zfFO`BJUPUMuplO@4HY?)y{{UOJ>Guv`IDdVCu@4YxXGxy>AkB`;!TB~RI+T@S03li
zD$J|z0_h$VjF>z{hkB-bcI#USS*R1_Y56TjY1YGy8#p#*lWYJN#`dNmCa1o4j}6=g
z-b=N{KE{CC-{&F2s~d1vzLbe{q5#>{O?b{A>!S=q@9HN4f_Ws;KzX#LktFuZ_hcyO
zpwf$=9(DTHKxCq<B(Lr`NgA7hy?j%#qoLDeJ6s{Q$OE#6$PtG7O<9VR>mpXn2#^L@
z_TJRGohKR8Q5?sqfx&e3mhe9vadgD$xr3Wl+`w`Xq|&acG15{{!iMzV_Vyq5g(Kh}
zponQmt#<DIKbo!rsH(1OUr<3sP>@neq>+|x=|;Liy1N@ex=R`a1f;u5K|rKKI;2aa
zL-JqmH}gA=I#=M{6Z`D7o?3Uj8Vx~xz^#Mz?5}yWJm^mukQ32#+6#+|8XCESy~Bho
zeweBLdqRwafN|5H&jqrKaIX0}XtGsW%~l`Z#Kxk<Wev-onHhmUTs{Q7Vas2B5vAhB
zl-~T>zSw`Xm<pKOI&o6w1mih5)Y9093Ic;>=gw`O$HZCuzJU_t7@L!(*F||h#>3N!
zU#AVYgA671sVVa#0@ezM-j{xx`{FGPHn&cd))GJeFz?lhHq0~w5sWXW+o1}+d2b7#
zPrZ%$?(%rI-+eT+k*zpZP4Uqf8A|w<mm?we!xXtB5j^sho~CXJa?zgdtadNMUsZup
zVTW3-MEfmAc7p9~*a@bw?9Z5MP9ac4`MslwZ0qIktvt)jmF@;mw=R=lM>tqoSh>9$
zs)O0;WUtChD8;~Fg556#GNs%x3e?lRKG(X^(!A`#Y+W_Iti+7VzryMsChUm&56~Vw
zK!adRBt%KStoYPyq{H)XMl2+}r~EG4$nOR5r`q>M1COA+Lm-|tt)J!9m5EaR4}FK|
zxF45xMhjCv1btzAmc*v{-x72DAouoT0jH}+DKb?g>_%M%kA2Fj92^J;(k?W$UffaS
zdtVlXo55giZO-sgXlHU#Q%o#X=4Eybcgal7OL#b9HNM5)T?XMMG5-O-$06@~!3v^x
z)m9_CAU`3dyeLOzx0<SO-CH;z|8nWS_M_toa-vkL$Bl@j`E;Ke67A2Qe_vCQaJe&F
zDJ8@_VVk=Bcgz(YyZjaOy``o7pri?EgRr>?pmud#Nigr@r*$s9{@wu%zK{^w#=s$L
z+OV5{X}F8|>(HU<v5nrz30nW)KzxpL+u>sXKAEBPNF9P@0J^}>bWjF5zY3^y*z&(&
z_b&5X=(3-KV7%|26XH!_=unf2G)!Y8;q@3aVusTSAR(iAlh{|J^R|sO$D}fG5z%1O
z_EcU+{!oz??`!{z6?LfN<Hb)i+Pzu}8;uik145|j8OwhkI+y<}vS3qLL)BFikIEc>
zKTpD9_B9nJyhqj^)vxPl9u@tMfbW(CN&pQbqwCS()>jTv5ZDjMJet?eNGhF9*Ly2b
zAkE+8IB<KC7LA#uLnw~_Y`Y!(-;1w7Q6cwt+cpy6RkED62p9WR(DoPo1Tpu(&%q2%
zoBjsoL2z@Fl$Y1nvATkYZ{Ay>0B`a>>Y0_ajFkKN_shQn(s*Ta5g`};VW8Fkea$_D
zp|H2(>-fv)OeDe8jby0D*TVVwWg`Fbc}>jBSa2A{6jr{!^v3}o6zJCb@I2@Co75B(
zjIF*1xaZ7TMH_=~`XdZGeLLaCF$rUd$*Ipbx6t6~Px41o4jl_-7OXeQbFd7H`?SFZ
z%xdhPR$ty8=Z_5cTV0hM(QQ)BWf;W~n9(@C>pRu^m%n^jGdnqmmzdB7rF&i>fA=r<
z`TUHL$^-nM`4%rq9x6s^&JR*~u;l}mMNrBYh4E?R%ijNM0XkL7)t$CdO^uCD+&J*y
zQ&1GMKip&o?7zM&{Tln<KVfwd&?BiO#S(|B9sSZ3YmU(7v~>YRA?umwmBgH>oU@~I
zlf!&*;<sdeV<C6<dyA8!1nmW|NQG6t_egm1)LhZ&{rOLs12&POB8xW9k~eoQ!D~Xn
z%4@LEM#lGS{P55LYR*-A_aof$<vy*{=LDJjb+<v2*MHJ^K{)^@n8}b{5a;=IbN$DJ
zghL{gmz!pm9y&ZE|F|v!<^4W<I4@d5bKMgN>)z|G?Smn0aI50rJ=LnR8+0D>H2dMo
zMhu>m@^pa%5Cg6z9E0fpPcV8)y;&Zjucv7SW`md@nC5Jlf{<8rwr(%zLuHy=j}+h9
zNmyM?g4$%-Dhia3ZJksOM-N?WXTd7&`FGSA5tH!b?8t}7{-Omk7Qru&%QLwc>_b=n
z<*l09>+wm^$$PB&OP-lYPqE^Kgaj~%c;AXkzXp#|=xe;%TGlchUgw?D>$te8N<(eL
zr`kmOi>pM)M&MA}=t7&zYCn$um;>kqfS$oo_;+CLJ2##@$wAX4HQ(!57r%igvp0Xy
zRQHV&d_>yYCW6m)IlXUSSTV7_Efu|}yqs+Nl6!qo)B3M!Pt>SPG0L3XU~~=MMeor^
zj}iOHuMKOCe!RvjSD;Q{&LxCl;e^_hH7*;MOjEbhO#{s{m#%*-q7vic2w5ux5L3sM
z%F1sA^S$8UN^H#pb)A9afq^Dv+MpVQN5ObaL>;^LF)vrGVVaG4b%e~z-0N2)aOlnT
z46U9dxc|7sgtV;&OA0|}GDC2CnHH0fu<z-kCx}3acfZUYYM+b8juot^WPi;uf=Adt
zDX_nPP`$Zl9KJ4_#j+%C+Coq<ZH4@PpWr0do)C0%BFIQYC^tS<R_q%+N^hg9z8Sz_
z$US2H%^BZI9^czuJdQjH%rYHr@mP%;FLBzy-{WGvJ4t<Oc8vlKl3zr**OG7W-Z@Td
zr%O{PdMcJ((nOn{)7$-WIL_El9xyh<?=B#n5f8Z|M|vZ2lR`)tcs@&uix}+E*8OKQ
zW^`>I-R*Wt-m+6I>BIHrOiU-uhBf7z{*^ibXH>m|W}(Qozzgr*PIo$n;$Zp?)TBP+
z>S^WeA@7sfZzk2uk1nTsKMa@{-bLODeiQsXjORNpAyyM;SmBP8C!X$GvO{!oUq9(N
zI87hC#gM@b_tO{14@6-JmAEPlMH`TV=Dvqpq#P=Bfp^>=I@cU{(lUI|t=OTczp<+0
z(q<T3asCyeqjG60t`!cmx3X+|vNT8j+h|<3JW#VVZtEPb<?Z3-ZS!0N@h<6}mH(UZ
z@o^_t*FDKUj9%B4GyhBg(EUc5y=;)<T|yW5;S0z?MCzE3n7%?y1qy6r#5;9SQ5?N0
zGKQO1zZ*X02UWzpRdr!#(ypyy#lyiQ#kzm+K+^9{qL8_Tjo3PUYMK^k7oTr#NsVlw
zqen{86k4%sK?^J`oFSEsjhmQ|r~;)0ZD3qN99XKrL#n2t@_2JnO;TYgu>5w#)H+op
ztNnY`^dH!OYHC=)ng=F2a`mToJMR~o|MwH|k;0|Ve;QOxTe)a=OjaF+rO)v~e#65+
zukEX~sdEu$wqHMg7D3;Ii0Crt>7sIZfyVR*tZyO{d@ERSB4!;jl*Kt!MU9PIoaD;V
z!t-<E$=}}n;qyZN=Lylru?oy4`C`yyG3vF-$w<-AGGP9O0?7pSwlw^$SFdP17Q8oa
zFG>??utO#KLCY^G4LT2!;qvmbC{=Ia&iA<mcR6iQ5D|hu2Kvgsl)QX$-+b%Y$zqs1
zX*}SGh1gPn+lZO47wvcCEQ5gw#Ea`6Cy@Sktk7RZXYvl2R3p7$Y=6P`+S4zx_I)Np
z+5mJxkFcu8$KNU`sOBhAgE$Lz6lz+L)fHY|c79BE#tx>Ek{qm1LvqTz&!#2{Gzl@o
z`*EX(q;=PO^^Q}LFWym57z1_EjKig0Jxxl|(e+)#xt1+WOLIuO;U^xA2f^VRUAemG
z#K5>5Aqg|iMN%`MRTLd8^em-`tOuqVzQgR{qr3SXMpX-KXew$dl~qh|MtNN>gka_f
z8wkC?tQBZ0=(#)g;rnO*j%6SA3E<#u7rX^0$*e^?D<HZOLpM@J_7Ui6l^;@&GN*JI
zDA8cEr$w4MAgF2Rv-k`T3mw(&;t!ZdudlDenBU}vyKQLJh%aAC9^PUbga6Cy6HLCT
z+S`}xt{3J<iAmC%;=hFyRjjnMR9m<CY7%yJ9oe<~NxqvrS=giV*Vo@)TN{1Ovzd4I
zdzAJ7Kf2cW1P(m%(J_$L#g-Ol#N4`!!x6FmE$LAoDHDZgXlPzB&<M}dP|&xwwJkY1
zr5A?A=J?OmrnxTM^^Hl2Upv@OpYwvt%1BK|%OJY8WbbZ&pr=8nYWb>DV81lxNo`BZ
z=+S9_<Vp`1Ji*3l<j4tTqMBM5i>2JEcR}SP`=2I=ny#$0R9Z@|nw0|w50A%j$zps1
zAC%`o-QCM^RmTw}<&rc)%G(7Yj9V!hxr9V9^t;>JsgLzB^Sgt(gHVid5`lbq!b==3
zEd|6_DV+=?EN(d_fi)Bq)`<xpRn#Wm-4#-lqAenP6mW5YLSC78&K5YZ{^-wSun4(s
zgGczt&k%VTEGthm$QBg>N#A9^6I)MMSjh8dn{eYmQqEip@-Z(D6#y}(l~tb<sUM*n
zSt`<cKqQ}ruI|W0zSsq&7z)u0c4tM-a@>`F&S{FiBPVF8oNL(V2V~b)jWRth+iVJ`
zPR>Zl$ytS8sC-*;cp5i)yA~8CNuE>N7#nh5)~*78pSGtOY5X4r^7?PZ_tAKCzI|ni
zIKEu|&h%$88QZCB=h!V&_qtp0R#OwX?SkzIH%~%GJ?=l9LMODB?<HOy*J$g#iqA5=
z*p}Xsdz~&q{)C4uLz?`F@5djT?kG)N@lJb|4>E&u<uIB23;ObGyw%9&<h|oG()bh>
z2b3oo#`@pkgR#$dBLA=)iCIrTi}Wt?(%I>dUT3-NTODuUpnSYC(~?MlK2#Cy9nn`W
zKpGCdP0k)FQ%}}Q)4c7akw7%1!oyfV6Lm+Q_%7@6P@G<Sgr1h4kAC7o)0u>bh|!4G
znQwlm?`6lu*#Q^#M_V0xK&wy<B{Pzilh({UM#6njuB}UeDb+nkL3};zBKByS0g37L
zY~{`3_Xg%%F)prty{2_})75rV@a(~efqD1+&)L}u01FL9hC?3xh8E@x%?tfXfMgMV
zmk~7JRoGzovw-fxc=POPF(#Mv$$jE#N~hxcS6T77TcV6NFJcG>%&e)h4x(?Ibh&%{
zCtsEMwwBW7dl^i<m%Lc<ILC?0ZQS$lPF?y;>@ylJaF1MkY_a<Sgpt17jIEnpww>Uo
zn^0Y!=O(xoh|r74{f9o<4zJ&0JD;GZ!9*Za!}-xU127BBuj}}7i{FX?tuLR&vXCGB
zE6iC7-t;COMsR_!zrkfm*~v<q<W(QXHxi#n82JGf*$0L?mzr7=<KrMrZ86+i63}VT
zz0wOv`21a7!(~U5(PM-@|C!uHQZ@c$`k;<8?`1XKNZSg@?N-EDP5~G<8wOD97(EVr
z?-M(kTy4=^oelSm94{|M)g1Sy9DUSGP6IUOqT?4vlR&HBmoG&gB&S=)fyF8+?FL5m
zcu>Dz{tbYI#=<hn=!JZ~ozm_H+Sci#Bn_=p;H$zXZ6K)icB|uL`S!VT;@YhfE?mP~
z%kz~Y*4tc}8>*kb&^|15Z^OrY|G)PY@B@8#n@XUrQ&>^aHn<J1wzRJ{4hFtVEZz(F
zCZZbeITF5z!IJX%^NX_r$B}T|;~c!!&Hs)v{PuxIU>^MU|6V#D^6Ojr*(-UwrS0T@
z7$(Id&N2BQBZPHwl7qFr=E^JKmbHUlY=*0;bh;sz>OGgX4NN)O@JyRywtwIfWfQJe
za=Yt3+jv%ykvqHd>GvYeQqJ1WAHw<Nr)@4c?@ofO;fu0f-ZIS3&CQLz7{0i)MUC&O
za&)S8bP|?(bQ5@v_0$xMtxhNTnRmPGTblBNx-R04H5zSR6)D1JA!+|AkQ~1lavdxW
zpP|+3L{l(XUv357AT{--s!&f^yQBOeeBC!<iJQqUrNy?^yK~Nl*49M1Jmpchq)aLO
zr@wSHRc{I$kJUbM&bymTHCRKEkJgXieP23T1&R?ND&x|Y8)*;6p;5hu83{z!R)+!~
z4Ty519E3E_QrEH=JT}hJ2_y4^X6}?11K#4>MCbp4>eb5S@n6WIIQOmbeMLhvLVES-
z{k4rYKmCgi9k6uhw(Vyz-jAe|+3L_qFFyLw%)xlVl-aitb;;klb9r1=yJmajpPyU0
zb$M)aNyRd%*e1UR{;8WIoD;22K$u(}GaR0MxcS`<k?yT~?xL`PZd8IjI;Qs)skavV
zwIv#r=dN7<Hm<IL^qreaada#0@X%M0P*5}M?|<KIyb!K?r6%|k^QK1|?^e6ptmA1<
zq`jl__I`Tjl1wJo?cnvTpTeV6owUx~k?ft7f-lTZO>yv&Zty#A{htHyv-jQJ5UdQo
zY9CHJgCH_SOiV1l7-7X8Jg~m+<~!Ll9c1kZc<1TtD-ou*>$giHo*2ZBO{=-mGf(1=
z`%B=bbvX~@U*EysW4Gvv`)}X?d~eUn&Gr3T^xt+|R_x*xJDp(alPBcJzh0LW67qtt
zlE&@7*73IbvpO9>B$n%o!SM8B=|Vzg?63I8AYGoP95_7K`CwQC<GeDM$o6;lj>FDP
z-;l?u(rHu}^dbY*ZsK@D(T3+ag$m?<+Q7faw);kd<}p!Y|D1vX3+T5vTQmP9nY8+c
z%sng?9lK3JsY7;Hfd3v5IJBp!?GroFV=T8SyjZ*M{?*Vy!)if*!!}&u`xGxmB;8eX
z-nr6wGnLq*&i|x*qm%6cJPW=AphwR8yeUH&ev5jW*wQk$<UPy7sY@E)>m>PxisSYx
zhwrMK;YpRKe%8xQezBW#S{jO9M-0`$2Vna_@BIDm&d}nI=P@B7G&CNURI&pDz!RSi
zKR#}MLSOo&p6Q?Oz8*-HeIi0;>da+_GGwf0Yt~1`@D4&HeVf$h{luQ;kC^!i5dROU
z%ddS^WML5)_$5+y0ESZ!4fftaKnvc=!?CktQD0Z*(9j|~d$zg;4Ddqx9pv{fC`74_
zi+s<P1@0LK3cQO5>M~g2jF!r)^Zj8`?aoUaD719oMu-70J!3M@QO{$;m8W2B<0528
z6_O;!I;0KrQ7jyvP!%)RXsp535fW;-u7?sDJd)7v!!5m=wZaK2Z(-&qXvb;1hK`mw
zbrm(c-!XxS&1ZGk(!yeAo8a?jka)E<C!3pFmlhSpChf$PZXyVK!1WhgA&>ZKP<eju
z->bVpNP-|J((w)esU1kaZ>`NWxZ-;i|C1NsZc94I;G)@iHOzlnatR38%~`hN)MkJr
z{y2%xRaFis@r97aw~%=+y5|)wLzjJKRh6QrUNv=Rh^%E6#|L=e5%%D4lMQnb$|=}C
zcrU1|pt{=3ChC2sOG#N<Dmw<J-B|%N@>(DkWowL1<dVW(s|!Av!+FaiB^Ba2<zUx1
zw;advz@9nV1t;9Y{L%^frld}!M=OpQkdX;j0?q`S4<d_DsXX@jnuH{^9}?2_zG#S(
z6LfGgDJmXg%s)WDsj#%%0UPbbnVMf|X#^lxT+>-G0fsCod%^-a{-5Zw!?|(8p>X(N
z<PiK7U+mM{9|*q{Hw|}D%U8m>a;hSqFWEBs)*Ltqu}wAH@4?AYQdxJU0=xsU&&HLf
zbOAF`m6cUjILnH1#=kAn!`Wb<0VN?V?N1^?;@q3w6?>J0MDF7-D39UIk&)u#6L*)G
zwy>|bimn?3iTtv_G@{K=UsI9}ZDTXv^A!Y>@8Bb!46=~WFgg64p9Ymfx-AptIiOB4
zHpXGqCk@OT_ys|gJu4?W{|$JQm+gy?m(gr?-#9|8|DsIW-o^ZA<-jfEN>m9wzUyIh
z$g)GD6-L3z_)i1x`?q%yRxY!zSW@QNbRY}}TpXKD-0i+~qEY(F5A(7sD~pOW-@M7B
zq}%)xRwQ9ZwQ9!+0Vgt)(xMs<_{<ks^Neu_2O$=z`HscS&25(J;`YMnf^-;pm{Ed+
zB6;Zj&<JX+m!6cdDX}S4XAu%3pK7E%_xNmg1{X;R%3c*loXu`rU!R=sEF5ufIG?&_
z9B~eP`%N##2<hF`r3H7s=2^aPX|`toyAc?p@7_I!Texg>?pJ7+LDI)W(2Uw22}R$(
ziyY8PM^C?Z&^&A?z@kyYQZpDMxgT$4YYX|PtLuRib_?txV==DI+n971X<hp5{b&ri
ziG7Mjn{?daW0Hu<DfpYAm$A}z?v!|d;8)$R#4*~>cgZQvn|%*y%@WzHYvZo4;*h-;
z1s(l5mn;4qc#SC+jMq$bt5E97J&-6Lh^Tlh_RoPceJ-d416N1mt%4u**R;CG5!G?+
z4x`t`G@~z#V)yhiwX%9pqM9mbZE=9@l@lh8A5g|oq+F7m`|4uCgV@JB-=y>KF(ggv
zRK-<Z-7(O$nB9IpLy+(RvRl3x{I1}9SypXruTB{F;)MwIRU5+D+IfvW(Tz?wmUieY
zNg{pmT2T(4LrhtDb)@=}%@Qnz&L9hCC#Ub-6r-2+3Shl-3O1jd_*GzRWMOWBjg4J<
zt?-}~I5Ujku{oKQ8^I+I564r<{U(bKwm?VmGtrcWmFn?bznU4p-|RY;9E+KOv@kDE
zNlr`QO{Ns4nYFd0*&myS$|53eA^9jM0g{rkwpJF~)ygpO$amF_pM~ocRXx}u2>2y*
zYIAamYHOX3zr<d7d2OXkgK7`Kq-<Gn`nI%V8A}=HZJ}|goR=?IM&l4~6%}Xb(Etj9
zi5TSM>YA#Vi5Y$9;%J1Z9>243R*l)le?3o2U3nURDSQzY9tsEZ3CrQli{J1%q|a6)
z4|ax-KAL!&6qjITlTSfeaOkjPkn@teGYEvT-#;sBN~6`Y(H0lQer8P+0y*)z4Q9yQ
zj{7=9SfR;!5DM#uVDg%JOG^4}J9P%)P8RGaMDIDPtX*_>fB25QP?ASZ0fEEnR;mVy
z7M5BB4@E{x2Us<wr8gcgsLi)r?td#So)KnkB;NDA3t{Pn*=2c!**PU2_$#Nm+~?fi
z<h>P(8b<Ov|FCMK;q@2}lFjX{i3>xh2w`DyK<5&$@6Vs7wH4|J8w!nb)G8^ds46Od
z8Zir%ENnS~0O{>We7TocptE#0S7lEd7#iY`vK7(N($W@UDQ09IB>Cl$)ML`#H%Y?h
z9s2Fv9D74UsjI#24ho`JmD7Of-Rs55)>gwUgUj4Z&kZ`ei;>chikMFg4eD|>{I;Lc
zOCDXE-tqhVp#d&SUEOKJ2lARiq0YFhtoAPrBBC0Sq8gQrjsG0AUwV*JOAwXu)il+?
zuQDxN?it|^fq@nl+oLQjzS%$T;C~fBM<!-uyu{;74G@oNM?-t;eG+m;U3)e>@*|qr
zR&VY$F0;|1l;Gj#`}bRXPF0M$y?Yjtr|u#eoX?g@n;#?n{`%LNvU>&TUDJU7L|xq&
ze<mu%B_#c<9GvfJu_vd<-SzwUaY9%;Dp=gdpmG}K;>{%!{UxBmMnXVu&%f%;QrqX0
zun@9DtTmAvuVSP7jdbtnvy|!0*l*N)(eyi*?MgaI&5K9s+Co@}-xX8lO0x6p4WF7m
zc!%(AI=9r8$_i=s8x)!QYUZ}J=?7U;5v(;Bpd+4yejiqgZorffS&p1N8`3*8t)5m?
zj7#$Jev$^62Fu=Hl^%up6>y3ht4&W+=C_@-8WInusuo7Q9$t8Z4ABv<d?3qmY+>PM
zb4wb5U&J<b_5p9^FB}H<)=n@h^4)udJF1Kv9BC0)GyQrUwZ=X(J&vEtFfL>ZUY`>3
z`ZpBN#78!l_PUbr9z1S|<mZ4;7cKU(XUtysh(CO*M5m+<ePWX_!*=5xs$&E5|5<j!
z)$ZJ}oxkHZWFH!zKaKeM@{((?kI>hhu&%zI1dQ5i#r2zeoSh>On)QaN7pxm+M3(?e
zJh*$cjU(MCX*2Kbv{co8@SzgyE8?Vh4F48ej=p+yv(}@RouEesM`wzl>@nd-1U7Dc
zL@zUYCsUU^uw>^IO|nqCSX@_@De>aZbY+)YhdRQ{+LC9}`AYb^c#ZgT9+Pp5v&azr
zl5*QA-zlmKM?#>dr)c6M>TPs`<q8zC3{Hu8r7X=cpW;+jvF@2B4w&?d&anRF;m%-g
zcXNu5j<&7pvU;r)uQSVTZ&+7bo2EG^7MYd$(f<1eYZEr<ll^riN^9M{Q68RC8cLhK
z)7A?zXB?bWVt#idL=*Q6zwTudbM@lNZu1#`!C<azIk?N)GX8c3ql-rKi}er-c$d|b
zAS1qDhlL-FKEqWS$^BigzrmmjsR&322yJp^c=ZGw{Ap$9i!JsWNHh70g$XKsNnVT@
zlk#&7N57X7{rvjcifDh%3EqC4RlBlXc(7@tuBoLpX*-RX74ryDT|&oR?uz7>I>Wuw
zh%rZAp0yylBfs~qQoUeM*Xxu8A4gZ*^M}h><e|g$=xcQ*+d8<Np>!QH83f?2q^vML
zDZVn9t-ZB%W$XIC3bg9*VY6tTsHjLL-~GFOiDhZv%tZW7U;O3hqlWam|De-Mia~eX
zzwUdFd=^9o@glUjU4<<7UjRb)WzKbqiiQgw%J}`sBNxrQ><<(R!O33z2Ul7u#8!Q$
z1(u4XWttzNJ$lk-mo!y?mMP#HYi&)26^s7RDgE5nsHiB(F)O+P*#kSRtdE)QUw1ei
zVtd6Y%bhj!ji;wC;d(}8@CMu1L2d^~q4S-Ko9+3oQNz2qcU(^(+hNm_+3L&0e{-SY
znhp1|;VJPd1GnAhvg}g@esoW{PAj23Ut0Zcl(d9{M&ohsks3vwu?VvMSktdD`hJ98
zuR#W<_Y&ksnMreLi+cuNqG&ubhbbZq&Wn}hUm89Lel6E*TP^sR@-cSJb4kirg$?FI
zwRgL(I~`T*f0@5*MJX)SJ>r??;<TM>^x6$-oc^4l#c^`O&d=KzIZeF3R@eLug?(d3
ze^>u>tWln5WG~wwwNScU^T;Z@p0m}#JLcp=uRhXOR-CaNT%YAD1H{`o<>h?`V|Dwx
zh<Av76`4D9j9~|%LV7W{F--!m5r|O>mLOfW5ZZ7Wa|V?s36FnzR<j#PLtTJ)dmp*j
zDWfIgjI?9TW1GJIj0ao^2)_?xWs_RF<~6~6T}~C;R>X{M3D<;({sbz%>c#--#GdIP
zjv$X!4qRlv8o$+%P*mJUm3I&rc1D91tRrd5L>(ErjooS#NMG{`3o@KwvFVL3c`NAB
zVQ9%qP0F~(nbr(h%p12UMcMa(xb|UO8|mUS;pxYUjDi)$;b|C~y`03!-+NTvaCNDp
zL)AR}nuO1NI&vi4n(z9e>ou9PbAF0!o99ZEe~Z-~qk?iwv+v4JC$#Fu>e7kkvIeKq
zh8ArSNgrHjbS6eeWpeE~(?5Px{VS=iC+~VuLj#h^SLuYRoLn*)1|6~;MwO5M9wf8$
zPWhA#oP?Wx!G8EK%Y@xMFuwB}A8}Op`jH8<l>FrFviZt^VPy9+-9|fyjw$9=JV`~N
ztBKkpm%0fLGH1pNKIWe8uuHRnBIDPlGDoL#&F=Ht8n5Zm#Uk$_=y01mdf%vuSripM
zpe=21+gN?ZZM6T83Kbq};*a-M286U_j@N6CUVUU>AuRRncP#JwnK#uoSKCyk>C$oe
zX|N*u)P?=M+Mr7_+XXS1_~zV$1~UU*pU{BTPGlaN284hB?dPoW=%OMP+qlxS$-j1s
zOj7c$2SUx&RMB?T$vmDWzkfNd^gQpcs@iXNOD!gKHrsP4i;EqsAYBON_O_J{3BRY!
zWyxnaL(Mmzn|aybGjFH<r#BZ>jox*fq_Ek1mv-oyr$NpFf+W1h9fz29Nu$u#b&hxj
zS~XHxrl*D`d+Z`^jGpFN;>kPK<H=A~x3*@8S-x?_Y<bv=c&oR#e<5q;>oR4;VjdVD
z(2%_O2<`Pa-JE9hYeMg{?mYXT+@5H*2H&HfQ_Q&@zBp+Er~EahaR=u|1yN<{+wV**
z!LsX5+-PmLRWhIBuEv<Yys@|Ht;S8y?|wWip_4&@aXQK1Q(6lVCo4QR?62vmA~;&S
zZa&})5lhG@k2@}62;cFBlZ|?ItoHT`0c2^vNKE}0dk;Z~K}F9eU9!XCm7e`AKzwL+
zmIv|T_qxsl`^s+t<37j4Ko5=e@HXGGan$0_PbWq6r@h|!8K?vaB|{ZVo^F?^VO1}~
zm}ERdCg1BQ707VAZvU{_*+N^N&e?xsQFLajn_|Iko>_X5s?yiM_}!oqwgR2>jAQZb
zGQH~MAK&I1Y;CVlnEY<<Rd89E7v6+<UcLEJL|r-k2N)^#G#wqK@P0yUoTM+*ub>IO
z=iWF^Y_p@I=96W&)WRjUf{yg=N<9ExYv!ASsmsG;gv6+b0xw2}WK*62E0*|PUY~Vd
zgUT}!eq3B)V{2pkzD3)Vm5%)%^85uuxyq)4dUDrsZTgW%Z0%!Z#UHyU$TR*n4e&2{
zl9Ijl+*z+xyO!%d>R0`LEdae<6ZM7TTw}NzxFeSu-2a41OK2{w+8xc@E8bck3h*yS
znL|cU#&0=pYg=l~j&_~G59RZGHkUQ=<jL>F-6cwL>iY<WS4n?Q?PmYBz2^|pSFk_c
zI3L*2Nx;UTeN|j`Zvn**2#kT*Qy6DN!qvd@yQ(}c=SosfK;+#?)S5Idd?(mC_i&?(
zEd2e2XJ&ckI7;&rsr5!*$J$5wcjRPT@<Q8P*K9jgKbFz^j(DGY(70)SqWRCM-KDM~
z<saeI%YABdnqD$gSEakuy@lp;;@Nv-au*k&A@}w8eV0F+y$|ZnC-TI|PNQ`?Y+v?O
z688*&)ri<X{`bnsnR(jC|GzPYUkYh+7j9+SJHWB*^KW~%B3ttYZ)kJdOx!2J{do1B
zhZFt2i@ri-_O~y>o~~XY=Raf(Y}g+l_<b%SY-XmcPSawT5Xm;oIEX8TOCSs7S)aQO
z?A7?U{0aZWF$cUf1_DBZ{xq1uq9WrCu>Z5F@m-6E--8wo0FZupA|6)zj;Sy5S-^Js
zDO%4uzqoKrU&(Q=zq+<E9dNs5KkhL0P1vi9-!P-65C5}arM_~t{432z)qP_k63RG4
z#k)*;3t@hKeymt(MydcCRn6HM7+4CJiM=@u93C|XJG=El9S=wOSLa_<@#2p$h(F)_
zF|Jrl^jml>be$ZN&|$>V`gr$ae4?a`-0r+8+JE7gdq=!tmgWbrcr*IKQd2eM_nv-9
zgai{_7ZNaYOxL;3zvm++#Msy$QG(745fhhMH*`b|jHT<(%Z?O7bvl>5>|Dr7M8q6Y
zT>m^t18eSvJPDuURmF#f<g<g<o9VY`Z&ehK{o<+|5C9DeH#g)gD`JIyh?+y#9sTXp
z7Y;7bth|dLhMvR*sSq1!d`2K~e$A2t48r*wLR9-d8+ZQ_w0$=IGvnKkrO9P&y$c<c
zwigxxvMZ!Q&b)ZnrEp4|G$SH#WZ2l=h^i>Yh)08l2)rpsh(2?rVKNOkJMXHe)A&xw
zDGFu=Q|;H!mqNu>_{Dtr?WRx>F(FD|NrZUR$r4R|Df#-mypx1-J;r;KSPvc^I^^Zy
z3{$-5Ub2Hz@}CDbWx!-_&b0wX3B$HQ)$+Keww{E~C%-Kl6Ol|e!9Qfks@WGC&|I#|
z$!2``-IOr^v+v5%?J1Va<pLQOS4uISoNOv5>6&!YZ6wboCQ>x2Uz#g1oqdH~tHFc#
z-s4ugdpLIrG3Oe*7u_3I4k^&K9(4oxH+;X*jp8>zj=3k|a}6GYoz~t`7Bgknx?f1j
z@6FNDGCbqSzvLbxY4JF+S(#Chi$dFc)SYsalxohpau5XlEwq;*axg>JR%qdlpf9I3
zI;OK|(N*m@0c^Nkog?@^HWw>e8chF8?8TS(G#|*V3p?VBxw;JfL7Ba~)~<8yLiTfV
zd!(geIYgyNPX9=E%6t&xr*HFq*I`x>5vKU&#hd=Zp^R%IeSM{ouEGL;lSMV`gtHs3
zZKtbyh%Yv!8aE#3B$<1=b5zV|i?WKWhhba?)R)iB`1ZodSVh?~`<p1_W6>Gh@RD6n
z5pl(e8~i$RQzO_GXS^;7$xM=FlEd4di?9KTtZ0hs<U~{xGg9XzmX10*Jgd}}RlkRD
z*f}~C@mOYR2v?G)C#u^QnbSagHEf@r)hg7&e-*KT?w*3@MR(dS()C~jO?Y9@TZlrD
zNUx06njyfK2tsIuU;M5Ow$N$wiRXB$E@1~tA@XlCNZyg&weEz~7AG~k|JhyRlP3BD
z0s~kXcm2B1V3<wT%x{6fPH}nQjP870GRzP_{DyAHjnHO7rq4O8-Zh+&>XrAO&$cpF
zp!5Rk$FfLnrP)~!e?7ru**2vQm+NW2U)mXn%pM|A15m+?V+M0d*1W%qlhdYT#P2ki
z@k^oo=LbgnV}d%@R6lZEye7n+)@pNf+T(h~sEfXtzk^NukB0X!9v)gYv)_NcN&=s0
zPB(xXoVzRkF09Qs-0$j|nOpRyG7p&l@^I?g@HQKY6FrhYZ4xV2-?rOZ8aqn5QqXhk
zBzT_oHoon#PX~dXwu}=DTY7q=5t3e)Y9p6-pOuYGT2iV+2Y8<*dNZZX7OR7Fjn}8Z
ztWj8wNp?ea)hvWwB1qr7=`+tz3!f3LO-RHx9n|=6Gc!AY`{}>$2QRQZ8=FRz1O3cI
zI=742$Q9R@)l)~)EKWxxjicvWLct;zwEAxQKOQ()yAAJfs%Zy?pMNJ;+b|x+18^V0
z(1{NZzZF*FxjSb-nFFmMG@1fFxOWkajt^l!S^Ib;PEAd#yU}E@m~FV3kD>p$EUGZ{
zKyjlMU`YH&>ow6E{_b^n<#&HqsMFAd%4)e9o2>1}_SWBF09<TcWgjrMsj8)eqZfdW
zAib-(2{|bnv|R`y?WF@tcu%4wAmu>wUomFJCdJd!TrVAd7pIIKC4n@Wc%`wJ4cqTF
z!&}E%t|T^^^*_(9aIiva+_H(du|t6`3J*0ga@IwAjsKrl;NHE)2omad8+cH(Dh37y
z!qEkpnXj+mu#TBo(Efx@P)bs&ft$xZII`#CiKSZkRE04l<Y42%(@s8%z02iBu}lUR
zL1&+Q%g?+0Y0@5{ZX{q@6^i+gB&RStyNtWjh4kXlP?n`|-hHpd6#!cn?-2**9t&G9
zMc=8+g1!iJ62+6&mDRN_&MveJG_Tp1kr9~sq_4+9A>n2&o)*u8EVze)(l_M%OHN6M
zkx|>kJ$Ow0b75DKm4z+GU$`R`7>G1;d^+}*w5ft$n>l%6L%vgB9Cdb(2@44{*g8?f
zN^Bktykmm2%Jpf@H2D<raZz`eHFO7&F&Og%Pp{?cAOHj+kmIy&ys;`NEy+DK)b)-j
zZKBlY=^_87^ADDuz>%yK%bo46E|DbeDMO}Y(BsvRpJHHz8kuk(T;*9t$}$ZOgE_iQ
z>n=)sqiaEPNW~N4Xdb84yWCzuJ|R|?wg8f_4%)FItaSI58w>`DIrhYM&h#tn`pvu%
z_tB#N)JfzRCIY|(XIY)DuHp0XdIqWlE+;u_`nPwQa%ukL0TrCuHa0$<lg`hv)Em5S
z6*xe5^5z@j-Rm)j^d+-^5s$sj-0f{14(SjjxCUxUhEEW5>873&d)o!coJTbAPm;3B
z3!1HZg%BYZ$6TZ|c?}K9o+0@`O`7iRYZ;f(U0vUsPHxJ-!H9^2%(CQ=keZr=l&$aI
zSCz*pORcTJ=ZyI3K5gL4hQq87n>V=g77Kudkqa_#iFf4kr^u%yB&TcYC_(&h1Sg|s
z^EB?d(yGyCAa*F_@LjLh=a2iXM97!y-B1D=-Lf-AIE9s-=Ee*^hU;&=yGzK9N<(5j
zy-7<;%OWAGqN^(>5%Y7&E=>BTV%z8Nko%0z6+KIKmTHWTA6)O#a(@3cO+(6@ms1rc
zl}D#nbw9XMMO7t>-?dOcg+0UQ3PhN&tG8lyjzkjqgJ1GuKZAGWjF<X6b#ZHXSrPG8
zNhRtE8<1pzvRppjK{~`&fJO-FP}<RTYUe8PCI{8&P!Q+;{;`*6h>~M%Z$Iaa&h#S2
zp1TyqPLyig-03u_u4QFQmRXn7YX_!>Jn(>l&R?z|O*S*c=&2p2m8vSk%w_8%j3qnH
zx@wabwEDjsdgf81KD0$UsWa8%%GyyrjqU^wPdd+*><cWPY2>7$DlA08$B)3T52mj`
z>veV>dk4xG2<DyRk-l>s3ymGJ-<N1+NgBN{={^+8FQD9HH8^u8-lgj8FsUA4)o3*+
z46Dist1^(Wx=M>V1Fuz!BCi0lnw*rJ>Wt%J6iFzfz31?-paAcolD5$3JM<HqcuT1L
zUG<lsh!w6yIpY7>|Ff5o8VeIsx7y+vakQFp4BF+Bf6Ux<g@w;byPdeb$1Z+BJV$E4
zucQ)nt>X+U9}Z|Mx~V|R4pl)W@8c$i8zT)-ivNmRM@~X$_-1Bi0H5z(IS61Ihp=oA
z*NJUK0%$Za!0O|}j~h-wMYYmqKG4lJc&*TLiFte?rhR;p{gxUEOb{^db#ydh9pgjf
z0bRfBNCA|t%8-fJT(LhrF~*Efhv-k|ANX!Y82tLw+RV-8&C+EEHQeV7U@xRDFZq1F
z<v0glo3o4sQWgjgL3;PgL0kpL7!z~JxOn2MJ-u$?b$2uyz9}d7xwRQ&n7XY}g%{?(
zE|70_xL)s@w1o=nV@8^qvUbjw7t!mYduE$vPUE^2d%0>(?sOzKJMWbh-hT55=_sfg
z^31_F>bA|UpBr$yi`eLo>Qh=1lR{U+RK+5F;K-T&xhUa#5w<=5$O>g9Ij?z0^7?gu
zPx;ZJYOV5F;j!|VYBReLT85}^%*kEEM|8)S)ue_vtl@5}tO732ltBZ;mp!uz-!?j?
z-9b&hji4wo?zC?P)l-CZb>}gjs*_w_v!BrDp(e&0zJI_|iEjmyE9&-JClzhzi0Y`B
zb}<45w7I-`q1Hgy^{>XP1#s2BCVrGeh=Ndlc@F`^P1VOPX{EM@&o6x&yO5Hz`O$G!
zgzd*O`h(s6aE$kGRj<rjD32&UKse!Mgiy#SC-1u}P?Kav7i*r_F4uED|NJI3N$&^2
z7alKl)C>%S#A$X@(?kd4VFzp(M~;n--*0e#e2wuCLt|-m-?9m(^^Ug^(x<GT+MVs4
zk$Y;Mp5n$T&RMwJ;K^K{QfE#nnL7UfTCFmp!{{d`TaF$kbpYlfUtK{uo&R__)+<m`
z7B>DHY!%=4#4i;%oa(LKbTw8vnG5u;c;G=6q^9}{cDT^+P$WTfjg-H3ySmFO73!tm
z0tR&(MyD%Ca@MDgNtH*(Wx!`XTWhTZ8f*iH?Xj{~&%OEP0hIh(N|Yn#ao3d)2OF2(
zQM<xu<Vrqk1t#m*EV%o1Kv3Ej<)7;(4EiK5VWVwWg3T#OP5GyUG^xrzc;w74b#h;p
zRhaeW{JXmHsyvXCSV#<q6}BlZG5J_`bqx<VAVr!4e&+;_?=PHH-n_BcpJlSqwN$0w
z$#{YUjc-bNYPa<8{<xf`*mFt4k1t}?aMN1%h81*(tgCyU!Z6MZIdLp~D51e)FndNQ
z(rak&^bb+{!uuWya?-kcmCo9+>BLJdbFtCU{o<6(+&uqD^i+O9+*ZkKOKWA}mi1=v
zgOzSNy`~Tea#oHp;Bc~~%hM(w`AAE@Lh)&^F9Ep>$fXn%v7z6tbT<4t;90)d@)Y~?
z`Z`X;YeF3jl>s7lSeP(>;k_yWZg^<t60*(03I+JJHlN7;WoG#!Os=-w22bW_*wnsJ
z9D9p6Sa2);#Q%!8nS4P(6ZMY(H(^+zpe3gWJR;FO?T@2Vc|0jI<n|~jr(gT*pNM03
zzF@;a{N7kuA&WPsr<cJkZmD^sv5;ZFOwc8Q^#`JzglkZsqNv^*9)ZDbSuo27o^Tyr
zVUdXxaWXYomX7CBG+9a;UUC^qF-aLV%)dM?SRabMX#aE{K}%a4QJ`+GCt+!B4hmiJ
zz@jh~3zd<q$`=cLWnyxwC>nf%<e2fj$bMj(K)B&t{T-NP@{^NYv6*QQxaU-&sG=ZN
zAe%t;^A!~%3C^7^R_5oX2CQX6csUcWr--$Zk(2*eGdp}3hIrS$Q5g{IjqrmQL_rU+
zx_=3hZiK`LP|qmHj0er7)d^V6Je7Yj=X*}aMxR?-ivGZn)H)XgltztDaoW-{p2MRE
zvVh?}8z_(wT0DM+Ycl7K?Y5&U2Q`4Fr==yqS>W6{(dYb!2!DTN=$m9j;YwGi)A6%B
zTYE_jEan_0#Ty-My~TGdzn%wv%GOfQ=pr)T#1NKIzIzc#Jo{EuV@vpZ4-iuT=D6|R
z0MS8+frfyhPa%u{j-ODHAU9TLVOgQIrS0B)$9*t4J$yMpr=ekv*c>Cp3zd+O?Sn93
zV<p(d`#Ld4TYd~MA~N5Yu?`NKrITLhynTBYfybcN?nZ%V=%j%;UysCh3JYmXjZ7G|
zTHi~M1ARYKTe<vq!Uq@B5F*IgEjfq(`v`b=S>rOfpgKyGx$<!2uaV9`r6+6uQxG~a
zs;H!eiood89-b0v&B~0OCy6qt3XQEvAdj3lyjt6d{L2G)!n4n1Gs3*OM}+&Gy}wqH
zdtXuCML^HZTGu5aN`w7=-9!`2%W;YY%nr^fI0(p@KtkYFM^`5oRVhvdOFS=K6J8bP
zCBEffCQA`PMzG-+6Jn@oY57jyipk2-XABCa85#M3`g5?dvI#<Dn~Z*qjwiEq8f;_7
z>(wcXzQojP!}v#-Bxq^=W&&qwQ8mkQL`GGWo1QG*RhcaE?jsa3s#h6c2ZWe(ncQEK
z9Eh(SdBttJJ#@R-nr0aa3ab8D>0QIM4Ejv?3Yf7_Tqr*RWbEdB{eQJAHxK9_?juw`
zDqS8WmRlCPj@K=ojqF%ReY)Sc$EYSJtjkjz5JK@oX$>J5LcVHVgUWZIn1+Fn@(&9M
zjrom-xAL^qs&ueyXg5CBJcok}=^0|`%btJD$F^}J#Mc+Ga!CKa<pdyJ<g%k7#sgJM
zE!Fvccv!gD;t~fCFxobFd}q-@x3o#5_}5hl+ph$g1SBPj-$8g4N}~F8j_@xqQ)OEQ
zhSKcuqGs2;a!%UEg2BzB+7A%{qTIh&YA9;{8X;qh1xK`$cK3IV^mK-iiD_rO5d2-Q
zSgl@A|5H$v<i`05R~!U8HWOkXp|Izii(tqQMx5glg?AZvX6t}{V)6m9!_dD=pSwOJ
zEtK!Mu9^BAZsL4>-o&O9_NJ=H1$I&N$Yrk3_Me5oW6sS=gXC#iOp=-)r7u|=RBez|
zrO|GUNSDGC+$O*NfDmZ^^dhl-&!Zgqy#$Q<)kb>};e)S`Kei>UT9UzGI#IMHjlF6N
zmz2x<)BqTC6~Ry91n=hkzIo|10Pkiq*W`IB(>GmhQZZeEmQ>z;A4bE0R31Z);b^_z
zVX;O`S>zl7ieVuD_X?ZX%a=kewbIgiHCzsxKOVg5E5tR42W-(xa){SsnQq4X<iU8$
z{G;p>k}KpZ#MjnYk^GWN)@Z1F+wg+0es91`hV@dSnTw|nK0FnFtVn|lp@*PUm<uOE
zLPE0n*DfxumUU{nnss=x5~X5JBZ*_&RgCms-~m_<qda#Py?;x{Sn(pJ0CnoP{OV6J
zD4`sDrYWPW{9%5FpTGRmTQN<G&t-#o#T?)^Q)>SV3Jf-Owu7UiBaAedClo3nQjD%C
z+-6G+#oOPIwTu;D-C#Yg4^-OE{-$N1ynw>{2(SG5hL8{*_b==?OIk{f^?EhMG{YqR
zi~Ro}?vu)4s`|Usb19bOKm4k3s78Q0FZWk^vGNa{IfeTvGRtg^2%TUtXyiMcYxG(?
z;?v_k#-+L(NC^V@tr(ncqn}@1aCvw#3$~!u@fu(Y<KvWI*}vH?gXhSOcH2Ss4;?Gt
zfq5~;Kb>cZ$cUP&D5BS4YEiB7h9s?sR)R%CUHnIg3XOKfa$>DJwM)x^E)r2FNWEyT
zIJwgC7qeqsBy;`toP=k)p0BGMO6UqwQ&PR6WPe+*+CkBEz^TolIc(OF?T7G{u@Vv&
z7l*158A;%m6Bf{sk0#NzEG>3$87l}3Ki%?D<L|y~qD3B7Qp>5^wU-ZXQrDKjo@QP0
zFP`uJD|MpOuM%w^i#wF&sz1P*no;V!T7YjXde7(-|H%45Bael)-pt=cqo3<`mg8ud
zsP`_@xglS=-n}*a!w1fC^*34Q#(ajUpb)Tmaabuq84n;I&OaK;m$cjzY}ejy<Cpti
zjtZhN2#QXg%1+z^eMd=&x~i^*sxHK5J%5gSK+E~Lwlvapslyq4e(`g=y$(0F7DxG7
ztBcCc<5+Yi$0n{(9jvEM8<%~LG=E!{<#VO}2;sdjQ9E7HA1Ct;Ktb3>f9j3u(pQ*g
z#(%V(>Qi_3`3F*c5)fu>4Aq8S;OCa*zoerhfB7;!F{9ai)`p+OeWgi-Rw@j&#mh}c
zxWrY*+?;<6a#!_6S=v6E5+Cv_=qfb+;fX)G{!nn%d$y#|{j__gzlIKCZx-_KXN9XP
zrV)PM4pkqI{rX6_fQ#_k1-V2ii+@w<!=0V0wPOaaVN|y^Z>(hlv8UZZZP7mld&6%C
zqk(4#>husNsFoE~pBB;Y*o{CnG!j2~MuJH~hk=HZd4`!opUjJqs7{Ex|7c71lSJZe
z!~Sm5le0?=d0MTa*j;?qvqnadgohpk(7KvB^A)jPxO4|>9!>%NSh0M+>|e4i!GlhE
z0;wG)r?$v$waypXCJ$9)m6UU53GPEM-tcsW=S_Z;$z~5L9wk}<<MkEw<#c$<q(UBZ
zJ<H$4ArHd6CuQ^_^>$-fUAFeD*1JRq!I$w3?5aqgnwo+J4EA_7=WkbJwX_H()9%FC
zpEGi6f~)4uH&XZw(-~r~)8~oPKq0eDaRXo)i2%A#s?t-chz$OLh|7?pfbyB4ocw%M
z<>OcNMMXKv3L1enVICqX&ayf>*MF~due}5^20x8GyV5}po7Z-@x(^L`8(!JDZo^9o
z3is8o3<Nt{@9Y1nOiqByS-(>K_}dd)tj}+$CDK2*vVo>mJ(GfrhKmjO_!K@KuXm<a
zUWy?0Lje0i#J3b)<kfL0D5fg6c^Lu{rJp~yi7k7;8O`s0u-}X3!`}uNVUx9;Myrjq
zK+$Mpm+c;E^Eg>ivE|!l(m`E`kcT?U7p=e)c?s~si%`9_mCrX+qJY<D!jWj0yf#z+
zeFwzOPFdA``62;PQivz_-9+y}B_5||N8Zab)>t2i_>}YcE92Mb`_B>Q*iY;ABdbFy
zn4X(JybU!aNeUh9OYOJn<I=d0oT3<q1<a^h)_!e@IkSWgJxBQy1TY$R3zh#SDN4l5
z3{QnTM-B8^d<)$x(oLqgmAS^fI3BP_@%pR^nz;O(b=~tVHAQ9e^F>yFq`Gd4pNr%)
zuAmqa70Fosr6!Y^ep~4gi*j~;(YUE*|KoFi5^K9(Dv6;Y2A-b5aXH9XND?wqq(`wd
z*bn2oe*XMvzPD+o&G{ISYF^WbmefT>Uad~Yo|ci4(PzX0DAXN6C8eVh_i)1%tn8i4
zmdZ*=;F)4#UIcfBM#^JjVUYvsbqy$h3}6jCWT9z3$VV$=X3@jMIVa~DZ<StKU#Z`L
z>A~aU=^1I01)5Hbn`?fb@(M9=i4#Yi`y8O&nF{{F#s0d~&FyM0`^yVG(6(iy3Aprc
zLdjUPZIRhw3B$TU1q1smLQuipT-Y%8!w39hQ4S|vk=sv+ieoHr^45$#v-w!XMWAs9
zyk|`##2WxGcvCxzix8-rIC8A}PAFs?9KIDn9t-u?5-($AWFxY8A3tpIZqKi?0Tl72
zEW`<dC>WF;hk4=fXRY7tBJ0lM?YxrDeuRG^vESm6VsUY?>-291a_UXb4$F{_C<s7!
zACHVad5QxRgZ%?Ze`Q($5|ZcUP#ZXL2S|_~-}LST$#w6}%GMTDy+rVO9U3H`K7sbi
zcHVnxVd4E?|Bq4;h!>vW&-<JYl9wU}-}~QEXEdLe1`1)G$7MeFZ}3xVkN9$BJk|4O
zksPZ<Ou2=naOw!Lx`{qa+RmY4Jg81F*MIHU1llyf&rkwFgPzWCeQJEGq7xsVR-ViE
z9ctv@&PF%3VS#-YF9VjOTnOl)renat#QIZU7a3MHnp6`37LEW<K0sufaT(Tb>U{-B
zUwj2z<Lz5@I*Gg-m`Y$=tzNRw`rYMDaOayeL2}9<)eiq0g9i`nKhq4YCR%R=-V;VW
zv{joY)l3DUqUOTFf*fT$ltAz!dyV4YP0A%77doK}fkXztRA_L&zGHcEo0tH@l|g~1
zDZ`dGqS@q9M~E;pHy`-fi}P7AR`x5naY3!a-$Ql7A|8OUbj4L2Wd8WgO5JQv-ls44
z_i_qod2^qqa2r%lXGAC9+$&UuC~)YE-@MrdY90w(hH0H9XlM^G@4F!*f#2Qn@DQ|+
zJ{Gul1R=W-@{LDXHVqmO@9IvDhtDv95gub<k&Wm%>*<5pE*TjVowz}`!QzQxXV3J6
zMKYv5vA#CTdbKRpw>eoe5!G%^DTI5&bo<Nf86V<@CH1EfbL^KCx$MRHrBM-N=uh|3
z62BTARAE2c3%$%*Lz$OoZ}vJjFbq6EVYxsuca1CNvE~u^D0T<&G7fMNEZX>lKy3Lm
zk2_ym`^6ozxZ7y%uQ{-Cp}UtxsQMuLn&_}2%))`!6P_L!a?q<~nJpQoM|_G<&D6;%
zemOqvXIA5?WPpsoCjGq67e0XO=PF9_i6kfEUSuH6FLTi4_W!B@gwRrG&g&^dP?E@X
z4`(X)vrz)F^K%HbvVNe;$lP?iW3nnKAlsomvMT*|ejY>g@B>h(fYrNu>jZxY)Jm0q
z@DKxb17K)7>ggsw*c)Q}IhgugcZY`0!T#M@U2GF~9oolJA?{;TBBvmD@(GaWK^>dw
z*?vk~ey17ea=D=&Ib&NOl_VVFM?pdff3w;OX+$L{A2|t6E&mD-3lce@sM*yFjRpmZ
zPY<SIVqIi|Ec5%mzG)PRZiACEBi73-1B1y%Iu2I+6XuK87~#&kf7RbAUk%ib)&tQA
zZd8+as%|QB!%Fqy^MFX9|JMR+wOZdisR=gwj|2U2NuD8B4&7!^*_)EnjfT6xD?3(8
zf+v>Z$v#viod%U!b$j;^MpBgI$9G}{lX+&tJTYqkoaxTb50=*jJw|^C;cvUQ&9doC
zs;=i1W2TH%Q-4OQ!n(UDSxLzRx;YtL3Au!*_(`8d_cU`e%1RHxzzj01`#mkE?jBim
zugXF12LXG_Ga0*2EsAKtv7nu$*BT&a`lGk^Q#svsGq@$eLCeLi`U4pV{F9{PM`c$3
zqK8*DL$0u^4W^eN0+50{@ArY5nkfjbBkK1Sa?y~!PYhL6VPoPhlf@%_N)Zc}m#M2y
zYd!nKkv5RYHQxbndbq~X(oi;x2)e^G{r!H*1JDsQw6#GZ77{|Jqhr^~6yQ}?Cnvge
zTP<bnneY^H1Q<eVm^cRde{OAVhJ;1cCUbM~)7(QWKaG<#8U+Onyz`13rxro+V@z!&
zSw8uk1jeoW_=v?8&mjl~jle<3;B%OO+BDvj;gkN@@bC{$Eed)HcdGrky?=e6)O?Th
z=r`m0E|c)ko9`~SLqdQPoTRx1j4MO7bvMCt?RIxHiMOCF&1!Z-M+{3%z1b7=JrR2(
z{!umNJM=n?_o$HxF~o&r9;H0K9C3}3q0|~mb~*bNzw3!HSeKa@pLPpkQs{O!<lU>M
z{t#VaO#b>cJw2$Xp!nkzb<T(;E9&EpB|8~h%h|y@H6K6zbF{`@Gj=@uKdQbmD$2GC
zcMu6hLIgxWL_|<273prIyBh(K?rs%n5$TfdZbm|oZjf&2?yj?azjfAG$Hfm>z%w&X
z-1pwszT$SDA1(p}w$wqS_sCC8CJIoOjKX2_gR?{vN4z<)TEHzNp($Ag8wbQFK_KL>
z$9jm%h)d$M0(B*=dH0L1lVbEY>Vd0N^V7H<^Q{j+bCK@xme=jz4<0=PLF>Q-&P=^>
z+$h74F*}8)nJPVvf8>e>@GCj%=V5$SHky5%+O{1F{@#__qMjj~tLrokz1yDiIOKk$
z`4FUDIQ>&lk-#Q<f8jKDrFS=lvJr?~zzT-Jex3C>k6*X6q_pshtY`P*Vj>^w%|S^8
zxFo(mfz!E%sl@u?KFz03AK7Xv?hkke1j}%_&fSg>*DD_Tq9e$*vIw%1wfS<~Gn5<T
zp&XsZjfA&e^?yuLM2K~^1xSbukiK$jbZUZG$~o@2Bvaf$Mxgd`#aE!XhA;D@ux{UI
zxD(hz`wGl^4(NFt_Q(Re1x5sR!MbcU29UJtEa$gW-y#<OxZ);l?Q?P1)|}?DCJQ&G
z+nOD~V-=mX^1ZzU2KAN`&J8xPYVsXX5KtZb$aL0TcSJXse?a~lVT2Mukz_f{#4%T^
z%gh%P_CpJQ-ppXB*z9ACPmCXm;*31xk%YJhzh(dt{|#=x*5rJwQvJtn@#aAg2a#lF
zQA8z#JGPyF4EXjJVki;17CJdAlSC60=MB+vE|uHdJ`5O}FoXMhPXU2o$9H*jRs1vM
zwqJ9BmCe_%FcIH?E>VxW1awd6m9*tfPfWgcOLo4!M{_b$75??R4NzYI^EVx_L_uc~
zx4el^jg@geSvZB<Ss|g1U?mNlVO1q9+lw|2>sS5%F`Ig@he*r}O?q@+r}DHQ8t}Rc
zSr?#(5B(CLy<L-%L>8P=@^_te`=sq%ZCK44b=vW-2hS12PUk#(h``Uk46jK+RUoKk
zU*g|i-Y@|a5ly3o#>I6Yr@*)k&}@f_(w`v-94?KODJ3O!b-Tk;em(*|tc3}F9<v8y
z5{mrd0LH#n4-%^V{K3Q{=xW3LA_zw6We0R8!`d|RzmY)U1&e@}(J&^_yyxosM@gq1
z`>@_R6Z$#n%-q}@;$T*A+#}8BeND|WZE*Zp5cM*XQ5(hU@W6zMGVD~s_>nWa8aM)n
z$8E1|fJYeJ)2C{(s$A-xu)1`#A@~IJZ5q0fN&0_*;NAUZe4-CG;PxKOe`+Ui9`moc
zA;$!K{>bzfBYxyFR~>^Bm0N&fJ6vg$*_O?7g4@Dx<)-y>Aa%sfKw%I<b%0mKBP29e
zGz#phFIg+Q2H3LWu(vn*_<(UY+&mzqXrm-<TxU%Ny#=5rAQ*$Y>b+Y}g1P&_$L+gG
zOu&YNV7{4}V_-w9WL%?hk@M=s&%to#GO@KQ(@o3xEWVu#hZM3uTp6E<(K9^H&(2qn
zSDfYeGdR?*ET;_0*YoXNZ2!ov;UXL7((IUi6U*f6)OMzEAlw8`1Vnb2KRQBqm7m$6
zs$E9ZNQ8LsY@XOelokx;AGpRGvgg`I-L<-@q!+Q8qmZ;*TjBiE>T~SrBBcMQ@=O-T
zHEn4Bpi3*`pINjzu#RGl!zU8$dcd_lo0-bE=QHA0C2?Df?A#3<b(TXP4UsmS7Bf|<
zOsG8eqcS$B!INhGv0Xo_*yQJ5`s3X8WQ>+hueZn$RUw)v2&~UthyL8QjQhJ8vbgqo
zL*drTC2O|i>N9*U8oVWuMe>gR#$_@BN+lj%uZW85HxDnfC3MFV67FJ15m2+;NtkJT
zf9v_3CcUDwfhT9-N4@{1HAI0C@!v97_1RwI2JvMMayvYlT&n14ocu0C#lphu`jE3Z
z?sj1F-+KFKo8zl{gio7EReB>-6ccK#n!%vaDVEgt*Q|Aw#U`lr4@7W0FW#)M-XcMK
z3Gs)f>2xK;Q^S>$JhZYzYcz0l|4nJTsa7N^;!#Y{FWtTioz<G^tSYJ#^Od<v&dyF+
zSY5t+*}0oESi8LLdes3uC6N$f>CC}i?Wz*v)<~?(Q1-$xN#mC;V@{_tRken`xj>wY
z5xQMvzYW82&<vqOxubY_(d_Jo`6f=Hdr7FEKH~E5nA<zlB9Gff(=N!)5C8Ukw8oZ~
z@;%P>47;ZBGX$o5Him=i%jlrSM*Zb-3s8t*{0y<2`dI<$F-0LZTAFQX9hw1v@?5uO
zUa1N`WZU7P?0{l@V2sCp18+PEBpvgjM@%tYAn*p(My9b5s9TZVQQ{JvY3Wp@HhX;u
zBG8wG)cQ>hBfClygp)DB-F}PPj=<3R>f1GZa3J=)e$WMrwZwhLGfS68cgr1HiPj$w
z0R;5T=YqXHx9y&-ncKIa0~5y^j7rMN?VX+OL03VNboU9PoE~zQ8zVJ@4I`?eRo9{p
zdUyR1e8w-QKM9{Cuq&%-HukoPQsLt`J@mWcxBMs#whX*abgDgUJJD4#o3T#X7C_L!
z<GxFRaM(N^M;Zjqy?3V!>+9#Asou6q+FrGG0%8Z;0_B#OuFi((@ERRac5i5#p75%v
zN!NOeRTdXNr{fyn-8pJJwT0%+*GCYC>}aggbiI}|N<&@!^z>AdrOGzK>k1PTdLbwA
zV^zb>SDL`;qNSrM1D>8fA)^-dWU>AHJn{;bMfe9S`PEA$;OxYb_-^qy1$=np*^~F5
z@fA~Wg=_r?jM^%O2f^?k>_~g#>;DXibJwq_H{{Q2pQAD)r4O#Qwct?}{k<2SZ!C!<
zb;?dD5to)`S+aFgbt147;KK2{WW0TXJ$sG!Tj<ldBKOJiOS3FxICCZnG^k$1{i9!t
ztmvh%?UEbI_dXZk_36mn)%)EK%cosY(m4>80I9sVxL~UJC}NmK^;U=yXQUbzWLO#f
z>U~3>a!c<;h&z)XEUoyL5}<fDV=dEbzxQK7U6v-_<#0Hg3t!&cd<dX#2DJ!n6g@fu
zKg{W>ta7Ay@R*znYB#(en6O!IOd39W*XzJqPm2Nti>RA4NCoq4)x0MQ_jbdMx9pBN
z95{g2B4RrB)8bIAKKJtbt*NRO;QeKKv_mC`Pedfdx=ckWC#@_Zo}w`6FG<<n-XZoS
z5Cst-z&_{Rur1y&(RU0;kxj|#R-A3&BQIBe2M|#06A?zi7raOU*?mmR9Y{U>VmH>8
zEPE4dIYA!<iDUBrxUBykDjw%wdZqSd|Co_N>gu-<Kz|hS48cHn@bCbH10EtEJrhP_
z=hRp?F$&+u3Lxk5Aa`27!uLDfv$1(qy!-wWSqAs-^FE7fQb)$~87e$Xh^CJlK~mh0
zr)+G5H>Q~ml90RXY7YvkQhvq<ooAH8mhW<K@YG-OZ;jMVFMy6x{tf^gdPR1Grbu;z
zA+-CPJM>d+C~-lm`$gN8DN3clj3|#YQ5viM$@owwh|E%XTX~_ekDI6ouk&~aJxsXw
z_b}ihkZ0#As43A?QU<*dNAU4^dA1Hq@zC^gyYGHzta=4C>7!lt@aT9Fd?Fqp(<0cz
zU@KO+3!}l$?e9{sG)9y<VKAX{vzFbuMMhJz>z4<^Na+Gg<stsxzc44==kX8``&#4D
zjE`h5{TJW9F|oL{>b|BRNftc(x8m2~LA3-W<z>xFNA;gSl9OMt@T{H%a|-zQw5>n&
z=IbM6|5=vT2<oFd=mxKb4$N0D7M{mQQlC3*9wXhj-_TKJr3wnGS;x%#tP&Yrwo7O!
zjaijl4XszInjPIhj?K4_AfStWyiGv|&M)+|t@FE}Q&{Q+)p5c(JHs3#LIJ4>9|pUw
zD=WUOy%~}$kztV>)(Z_VCg|0!Gb$bh8D_qfjqKMbR0Q<x^P)JKwI2J!Q=-Io-e=`H
zZ%uf=%QiDoA6PvA0}!Xvcwlyj-`<;;V*L@$geX$Y`<O-mjsAg&yUuM`bIGY_XboO|
zpz;@cLzY@d!AMI%;bjz@)y8U|1TzPiwmz7Jgfc-5@v;nIk0AbP8J1T8K>_?BE&Wdx
za>@Og9>@@z8nQ<;%|ZD0LfW=44^PP|m4S*ejdxv_T1T5$c>4m#?S?B@?w~WNd#0sV
zXTy&Xs@pXs6$K15th`LL#=SU|*cuLd9OcnG-FZb>R2zaHf$9oRE3_3=q3}_Fp(rfm
zzK8o$DqWV0s7rVb!@&}1HFmv;#D#hj*7PWLh7GrmC~$L{Md9NS1%e2ARpAgs%<G|l
z=%3RI`dY<~rLjKV4x9WN-S1*+mX}!P<W9d+(R2gZelK{E>qtnXnK<+jr$^qex7(<h
zJz9Inh2AAAp^u-Ix?{=*Osuz^o11AKyK?+93f*C6xQjRtSy7c_wx6jo8o6q+-t~tt
zW_VLrx3KGvS4?0<(hM;I*yU}zj{ZD8JgmSsoMUz?bxNLUT^aqIMj(vlj9Y2Hg2ma{
zal_Xes`|u(Lztz<#l*t=@Ue#Ux~6hjMMcYRYA%;En~5d+*53Hc!3THHOC1e3z{C+A
z$J|Ix9uFddq-e0JFPm~lRUt$1{nZt3;qf;psqA~&RomR<gP|OO4l`X;1h4mun&b05
z#o%V$3F6ITCQe+7wZ<HT*VUTX9p<k?yk=!l0(e#N_(fL<$RMu+WnXO9R{OQVsaR`6
z*ONMp)#DkW&o5*bkAtRariySgjV_#c?Ddq;a>&k*-MaI+7nN^ZI(_qK_RGwCm${}1
zop%HVl+fdnX_)mpSL`D{D4;Z`>RnvMpdg>%n%_BW(0#e5SbaHPSvX{ij-ZUpV#wpr
zGS9hLe^U31@kZpVVc8I$$Rh*+8levy?cVBYoE3yH0EpvX=4tS3q`kad{v8ZHv$;O|
z-4$x1EL`T|Aln?2%%ms(kTj#(*BxRLx<gzV(`=+AQCq!HPJ2s~<VAtKX-kCMYZrIw
zi}{A==;)uW#|6A{_Zb~8uL>smbpJ4fO0hVEw6**mIo5v(Gb#XNwf^kf@N2$-J}AY4
zdn@FI7wB2ST~U-PuuPGfjE@)jKrQR~=3!dFnqy0=QQZ@H{8On}D>m4hr~`+fVb2X~
zqo=|NTtt9xowk%PLskEmn?6-l#DU%5^|T&3Nf99c>o{!Cuy%NR(ct^9kmJJuwNB~r
z9GWnnS5i`8bxe!-{&vDmU}-T7f<87~{wCNBRJVxA-oMBqf+!qu`4kY~OrX{J3Q@Lq
zwkwYOVPR%|;*+sV$3|um{1N&B#oNIm>r1R4iW8T3>tJUGS;0FFtTB@K_Xq*UNgZ6B
zq{D;9$C{1Kauar6;kI|LORkzpr732gKBWTUW=$~J>liBP7ah7U!4vr`JA;OS+U0sF
zBEpo#_^RQt94@zX`q#H<8Aj@?^PF(n=!n+O+XtRhrS5{<;RZigOK5CD1qpp5;8`Uy
zBI&Yl&?@t14AcRw|G1h}uhphL>r|p&s{!BafyrCyK&{h!)VtrAz4;(4d_yJQnr(}+
z;hgj`BJvwxKik8Zyf4^+^d|qk5CR_yEmw#{6@@;AhC)705JO*0ZLp&{_AKvhJ-y`n
zxRyuw$+fBTXPcSVz7K<cHGI)iPd&6dOIjG~9PtZvm=3GT)BIq`dTy!mH*>dk$t@|=
z<?UW)_Eopb<>60{d=yWVUzSZr<a^V|!Kt$Kz3XF%6VvANMe)ns-P6|o>kXdPr+JPO
z$?4hFs~fvX^u|eH*NTFG_j0lqI@)M@DUw_p?;5+FCJ{4T_pn`z5QRUgHE-4Kg{EpN
zZkX~ch_!La+!cU_>C!24KrDDUH9RvHxHGsH-2V#q#VtzQ-wk+hWqUF&vBJ1;*lCb@
z{nX5$EPcdqV1-`f_B=v&w=-g-ev*e}ceX{Ypngoe-fM7~VQOm+ZNdFLt9<JEKaM&o
z0$@Md7db|~v3G%%QW*e=SO?Jy=cqe7dV(@ek0vWjJ(V32b4wLOlvx-#`%(|FxwR62
zvFiEp2T(2oKjqn6JX5fUaizO+-wu6Vw0!Gs_ZyNZe%+VRlY%zet*D?=XH{XYeUuC?
zFG#he?F3)l*D-CJ_PqW5{b6w`efd<i#{HvzK3Y7k@$NP{9`%B13)t1%$`yZ+;k=EZ
znAGU0%j4yhju(pgD7vs>!YJU9<~?NT#vEPFJwlwdE7(;ugib{E=(_1(5d}6+h);>p
zQgL<!S~P95N<AqaGqaenZ#q^xRyLhJJuwecrBe=7yU1FQq~1Tj8z`%ygUhQ+eSOQ=
z7L}Oq8kpCHh8|N6?Vbd+8(%#RvW}&O3e(S6*AGYb&&ts+(0<cUx3`NQx6UbEckcYZ
ztGz3uxOt!QI{R~g49sxf0w?jxN$^tbVodvrrlqT%3Fo@J`Q{n10dG(PbL54>_C>-s
zM+w*Ut){Y|JRp{Vi{da-#^o>c)1iv9Ce-;}k8IDJkI)K=RA!VjO&o{p|G>|7WFil<
zPGk4Y_pFb&dU(Mg3aLGiGUBFNDGGDBB<{a9Vc+)*mw9>(L$IKRA7m;8QNbG1TYkA+
z<W+SaIn45J!EXLW(|BbmA9?M-4|+3{3r`fH@TbooA5}`cg703zFTCSuKXK)sMY$l<
zeFe~#(1LaK2`}aYlWcQ~^=gk2_^WAf?-!$z4-LE%CPon^<|*e+{Fs`B=P9$6go45T
z*-)dvrMw3jAMaz)2=`a#)1MtD9E8ah5XL{_=lD~#Zu{(cA~i3Yw{3t|qA6eD<XoTj
z*@a(7(N0KlMmf70=6lDX6`ct5K2C@QaNg9$)kXZ1_usz#DM^gtVpIN(F2eK9Pvr@W
zosnc#n-k;@KTjlqd&6(1iy4MS#FfP8VzKiGF8RZ8<fUa44q_HaxSmMN>nOdvmpmTU
zY$YN=dH$DvI~Z)SC~zn^Gy;7mR6tr}&GNq9|GtG2b7!{x<24!QpxQ>gme;fwq(snV
z-p1UyzRJD2_Wu5TK~W*RUMt{Q0g5c)MEz^2py5Z}Zmbu6FMF)YU|s}2L-DGriUvdu
zB_{A%f?{?Tm8=`_@01^4q6hZ(P4`J!UR|nS|Mv#2Z(MPQh=4`{7>AmbfdJD;5Os@&
ztb5#~44bVp$$uZE(~fe0v6j<O&?5~W7J(qXV*~;|cokj*<ep%-A}ze%y>9>IvGvHg
zz98fX<eL`WEKHfCp^wfhbFXZR-X!U>oiRFPp4WS)2X0<}*Ft~rygeP~i0bvK<%kKK
zPcs}1b@2<!DE)tD;CfrPB!)S^9KJTY96%(mwql7<I$t?+xsLh#&OP@0cOJb<$2!b#
znsgD@SSQD9BoGq>2L_#}kS7oa)M`ex`L&NME8(Vl{;aJtCQj;XBK-sv)BTZN<$JYC
zK5TntgH+z8<Y?kh6%~a`9<sp8PNHT{B61|I_D_<NN{QX1o8DeP>fe$%bb|iX6xqz}
zf=AlmNpO=Z@im1!$G<CNIWHkq^RX}ozLseoA)GXb1Fm?w-A^k6S^oFV9Cr%GQ=iio
zevSO5s4SnAmq*TNqzKeI<HtKEr?$p1#KBNp0~xzVrv?<l<MjDnh-)wufIh2Dy(q;e
z-nVuN7}nql3B!qxA3avMP05HjEqmy~{qJEqT`Zsc_9ArzUd}tiG}b(89zX&iCrG$7
z`a=%8)7Rc?Tf`4;A=_VE8qVxvIbbK_NdB9KDwPd}Jc`Q7#)}IUnn-aWRW_=PCYXmt
z4rV|`(ur1GPq+2v+9HJ106o){F0^`#>aF!Iz?6IF7LGGJJMDqcsqwGn1TOUg%`B(T
z(&^+t7$|Q9zZ{0|K&tFL((Q_0!tw%t6(l!6tcAiSCff(9uUY;N5QY9c+vIW;i(V=i
z&7m}cfDjZHmjz%ulbK=#sp_Bz0?|lWSs7#?x$bdce%hWYUIb9{ff_sFrM!wUY(d(s
z(kKn$nayA_2^lfeBJ<#8-gBnKZw=xAWS>VtR@Rr}WbS`P`A);zFIXMfVdzlM1Jj5r
zvZzsGIf!Yp_rpbc%=awWGpv2+SR_<}d600Zk!rMBv`jMQPG*LoHr7wAQ&iqb#E*=e
zpD-DE=*sPkx;(1dzZyi->zg6NbPsP9_ZF7p%*qS;rxD-x7FTZRAo?C<n;M&9BP6M9
zCkhIUYJq6JX{6C9CXqazo?&=`F)b4M8GEQCiB&^exuPWJ3;@nsZ=ctWfadK|wW(i9
zN(#i}h<~`Jzm_}jgJ3XEDnmTcH~Q~mjI>t*Yzy8R!1^U3B0L8;Iw5?d6LP!;uCyi!
z&?aqEb!ixBpxtzV-;UJ*`mzkMu@&G%E#<jo=(uo=G?8PzoIJ*=k}!N7U@E;twTy~X
zg5irh=al8N^Vdn{29dS3u8PDu)B5{?&FZ!K{vdZi2zdl~T+n@q<BSkcd(Ug!^^BV+
zkPsqVoz}?{e(^O?)6%dWxZgyGNxTJp(73@d%7W{3pYHeC1Z-ixA7v19yaz@}dGjeB
z>d<*6kVU@-rnGK9qoig*cW|U}(8s_GcsQzsNkn;@GT<RN`_dJewbpJwx__i5r$X*d
zet68Ye|gfnHR;c^+@lq5+Z%72UY4g>cjLjGAv9YfRkc|E=5L~*sZYpxn{E6o)sd91
z&mL41nVaWPhdiuqvaAnF{`G!4nTw}hL*@RpyI-aZd0}d5t*fQ1u;HP{^~3(&M41G(
zt&8;a$Vj7FS36$krN-71Dg+-5jo;WHO?Gbf(D0B}_1zcls-Io4<z;_xhq~5!9KL>N
zzPa${05aH*R<?#yT`S!&Lf<}(=P~4p>O|V|3iLEp+0Btrh7ng8#|imBW%<zKyjTK&
z3Z_gco1)SsD`z>Gq!}zvDjiayr3|d)p5Hv%I}?6W?{APlDl#^e+~{n-*Ama;+5Bs|
zrKxs(cXl6c6!^`-Oi9dI(mFVHP0HWW`;3e_Sp0GXh?h21_x*+34wo)#lsduW{@yov
z9H8!udlZ}>ZNV-HmO+J@^;2E?4<3-5&?PR{Ut1X%++zXG>0_{EZT0CzL2$5-%fzrL
z%cx}5WGs#6bIPL$aD-G4*`;{%QFz`Kr#hQY$>elB6Co0z4O^j4>D9hP@P>$pIVx-L
z4~HrH+x-z1u%q@dVtyeQkrOMIq|{uXoSDsT>iDgq!=jDeGP^H*eKdQb&8Svb?oH`q
zWPnzFVVZHS1wE(iQ{;>K;y#h?HHlkt1X9^jxy#JaS;T5Dy9^nNKW+SuC8&!oOwHO%
zTiij~gMS)@&YVhXs(-NS)J$ompX<oN_cz6=L4?v|tRJdIW^4IHhb^ba#@eTitzDDQ
zUi%Xixzffp{krqUPPn=0{8E*kp2w6Wfy32gDkr^`YjQCZsk;@_|HN-|+EuqZV-qgX
z1ZyGMKlSkX`YO1fBmS8a8A&C595cOxeSoul3}aXAiVyQmqH^b#6&lfT`a@z=16GCn
zt`+BXZkJbmy9&Fv+7XZrG?^XVV)DL>T4ZQ&=mD#Xkdbl^>tRTf>(M102X}<{F3F#y
zZX;8Z4`kf4NeOZF?-tKYD+Mae)dy=eqAJwc3V&UqI#^zQ|LV4*x|3TWscNcBH#|5r
z{j-Ue>u>wQtr^F|H?QBYxis*yGaKY@hn(X+IX<`Cp3bL5@Og`H;xDu=?z$;kauB&$
zOy^~%*81s5)l@ksE17rS%1ka;n#?{djLi>TTw4qTg*rI1D2?>Zju26EGjSeoB16qs
z&d2kMOe-4`c>==;`S|&NHPz;382LFo)MV&&U76L=VEr>Z^pM;6NZV;08QVWC7)2Lv
zyRpEY#9SwkY*jzwK)D;oB1)9oTq+|2I+-znULNjSH@Of9e!N2d%j3<l$k;^Zm6PY$
zhWSNV3-DUFN%%GH9v{JrWm7@l2DS8V+!V-W(gV1EdiJ);_)DmmI~^>T^i4mEUKGFq
z3xR?{OEEF==9(7_^x@ijh3_q-{x27RoTI9uz~eGf3kaoGoU8`dhS2#e)Z6aY?uf!+
z#NKwYmCVQl|N1(}>+^$rODl>Y798@+r6w*m%*`{TNJ6#rZP})qYAY<YTlLGyuy`EL
zw@~xk+J1QCMltf(UIp~7!BX9gal=9G89}n_7mBvg{wHbNGKnu^TjwRH@DUAz#?~Vi
zpYCFQbl9J&bx+MM?mz41bUfKypPQ=~Fn+9TVQEUwy)2g6&oznsYvqLS`sW}NoD2JZ
zoy&a1ea4x1ZtJfCr#;>YK&RSw`iULqh<5WKxOB`$hvm}_SC^)KR?jse&UvAw$$<(2
z%J<Vv=&WyEf+#AKe8&@S_haW`Ffj))qOVrtJ_hFm=kcSn0-bgv+iB0BBo=2b1fQVb
zWnbMCqjsHIxrc;{$dPv~Cn3T6{2JlA|8Bd9NNgjUj<ycCMRiCd#Q42gOvlr}Seab`
z9IXLVLqSYpR=aL0tB%(Csj^3iv*RxSF5otNLe6e&v{dFl-|EykB4Tg9S+1cclal0I
z<@gl;i@8k-`Ig17GSPA46S&Kp-m(0nag-`3%C1ZxTTl9*9Ni?t<Z?3mJ1Xby;?Sy|
zj5s8j<kjsRPWxUa*ecUp`$21=Gpg%&W6CyDr;lUiA^MpM_%_`@e2-9xjv>Z=FuKMn
z+AzP|6)mr%g@t&@{EDZ{@;J`%QDnh59^wAqhUo;6&(}=;xYv+B=9664vmFV3o_y-o
z!wvz3(q?~Ig$cQB2TGzMS$Cq{Epf`(mbqTqeeaBzwou`w9a9=jk5fxtGUhz58!um7
z#PJK<LKPGZu9|QO_!*)xGk?Z6fu)7<ImD)FWqMP+J|I$jLtajycz-cb)g6gt;cF54
z4Lzayo5$cnnP8HJAbl@8FN4bqT5WZ$&)v={%e1xVpF>KxmS1oEvP%~FqE!<|YT?k5
z-J756+tgxLGnwkyhDjZIiX8g40B)Y9vyhPC9KT^Ha17mYcRh9@A@!pv3%Q~?8w)6}
zgZT%{J(C@q9&YepaR#*Bt2iylFEl1+!w-t6a$Kh*3j8mJ^iO<bnEn0{=o-;G<EY5D
zxy{7-=I()p7JMVmamd$)hm<4~P!V|KyG~Ujd}hZJ&FSHh87--;{Rl69R4#=<rme})
z_eRCqKV3+V&l2p;`~UpeSQ}Vv^B>eZZoC&R51k(hLb&<q+xF~fbNxGyI94{#Wrk)P
z*1@KXh2{0{`CSVw&5)a?AYF_mCcX8kF`#43!`*f0&z~vG>cN4=E8?Vsg;m>Z!tCto
zvHt0!j>E5z(m_p~u3wn?mP)^ud@HJ88Zuzs_I8ChT@bk8!n(rry0mHQ6Xb*PxD!>H
ztMH+QgHCPzuViY!Pp_ug2P4CGi<5Qdv&dqbr}%H(1Z>_ZN=iv>4v|*cmz5V0_X1m0
z_-VR8@-s-Vhp)m}CY{iZoVD^ehu@A3{^*e0-6gNA+|kK+hsSkQL0}Jwn^d$H)>PXQ
zU;iy80qBozbSLo{88r#PLzNuaxt6-6>@6)KFFHCzeJ)<6CRS%d7kn#2RRXKjbuP4E
z8dFMATuRd7Fh-w+B_l1vb2STYT?K9JUyIhoV}yiJ`is68F4WYS(^#SJiq#WaYdtQv
z=D&K+_|L@0Bs48qftO!#0SK1Uqj;#Pgt<Zrva@R)m=qyQL0aV|B8ZsdMDN|7&XH}^
zyvRP}++svTt!oFfY|1kP#YKNO<my4f$=9OuR1v4QDz86D!e4WGtX;I8q>Y$U*UL54
zGH6%Fe2eV=)5T%G^*TblxlxS!i_<`+v7@2kQ-oJVWCl_r>Ks>N)6g01)AQ_ANu$(_
zwlkL+Ev*FHhnz|Z8Y=3_)?0tef1z?Mj`kN(3nXX#D5`>~A2o}`O4i#OzZ^rV*4Niz
zZuE!sWnh3rRDmiZvz1#iIsw9q0y9F~o%lgQeC+gOm$>*_iQX=G6%|y37#<bN)?wzy
zcZP=PUt0U$K3^qSTSFj-at(hC4E@^_EOoW($OV#tvmsT)6BnnZOtj?q$gRQ*&QhvD
zVyFp|1<uAuW!X<k`X>@S1aM*?vjwJ8>Mr6RIF7<^AXfekKuq%L?^P^V;UNg}vEH#C
z_G!<4+me+$+|fLhI0m+%?z=A@KNnV0d;WKKaN&;~wmrIozZivPjRQn8aBx4NeX(@f
zR$*4v-AzYI+X-iQ#PZWqxNXVx-r<{6jWj?UK~!V}$RWT-YB;?#ydZ=nAZ0EzMdZ%I
zkL8C}jtVKXJpLZ&h#itY+fA9jcSPmrlqDq1CjDe?`C=TPeXU0&_Zu7<zL=8dqrelR
z?NOJ0hPaC+#~yr|P#&MQ$V5*6h1PR^!dz-_Z=h2<6W{dPGLPD4t+lj8Ohi>2Df!7<
zADN%YRc!^mquQVG7dCSf9ig@lGmX1X%Nk6EGuM6_zA^<f!GMq|&R6LSm%qJ!EG-(M
z)H-s}JJ)SBE5js;a(u@o4)gA#)%z#!&E5;~e^*2g6Ivi-VW}*h9sYtP)hRJzujN6*
z>UKQgK05SNM#WZW(ow-6Cu+fby>3pb`F!yCea<(=pujo(bn!=x24r8W=6VWA^-h;-
zo35vax18J{q=P)Pw6u79uF`WWFI%<xCclvs$?}4(DCON%J7Jc%ClejUyFYuRah}jf
zwuZdaozKb1A>$l%u@DaaMG?Me_)^Xw>jf<>eLSba3;KYRB_nEhy2JW=iMo@9#@Zi0
ze`@}_?^o~q^c&_<1R|qFBy!1!1_wj`5YiUY&aDX0fV8JWK5*-nYmxQlUZ;vB3gR>N
zgDRR?FS4)ziOy&5N?<XxD)fF9aI`kpvj{f7;&*SpVq#_pv(9)1nHntbyc8kn<R0m<
zdIhZ%HumipP3``(9{w|a9HJ%afmH<J<u(<<E8E}NP>Ili1PvjpuB@m3?qrBuw(Jnq
zheBTAyBO6CgdNq^bHApDSoJ!r07`@x+zXo>Qc}UKAtbMU=sen>{Upi4%G50Vx60l+
z>~mxFLkE}*DJg2|n`K$P`{9?e^0LrdjlJ5gMqf|wRfcM@5dJAuhA*W%7SlZgQ(COm
zlKzn)xgY%=Rjk?ipBn24O!f7%Bs1gUV#;umGifjB{2Ej6Hv|op%1cV{ar^}32+Wyg
z%2e|Fi^2F<Ce&h7MOYXO%XTny*Z<M{JyMeB%k!hU`&aKrzj~v6YL+)`tSkc$e;D5v
z6{Z_5D!x$CkVs7_IioFyo@Hn2GL!%E{Ij2y#9N8$o#(s!zYQ3oXgojANb9Ip+N^%}
z2R{;da&G5Xm^Ti6c*ci<0B4=vWvovE1j5x^JP3raELnqIlxiN4Bn=n&BVoBNJ6e2y
zy6*1beDw^W%+zN-fU21DjAHr(JpI-xw;3hXmT(T^bf}06K6R77GTWK5OtSeFn(Z&A
zWFP!TTuEKoLf`DfK+FFrzOn?IZV-;&ym6BsrJ?*edBhb7v;>^&qvhQ=cQHTpzhZoM
z4$yIi?6c;l`G>SGa`A4yVtJjFQ{eAJ#_c$)2#iU{{Di%gH#pe9mTi<SkIs<%Hn&!`
zJ^27X9)Vugi)%MtC`9%`bu&J6{K!g_u0|lNOgIY_d(EBX2?QJFqN?feo_y0=^3}4B
zPYuu~ISsz^LNF{1fykO0K0O>?Tdz|dS3(VapmlHugHLvbJ4bQo(%o^wbQInN!t48Z
zS<q)QZsCMm!v;cMoDv7Snr_hu+HNJ0IqjIV?|+caFG#5f?Gzxxsm$rbPcmLF#+mQX
zRw>FQW8Od~b`7}O&QVDyiti&>vc0iYj)rI;vi^ua=a=Ke{KCLBG?>%H>>IZV+P~fq
zlh^RSjCi$t&Xk(ZpSOowj}F1fD}9i#_z6BS)Zdk5Medc`*?Boozc&2wH!#$ImfF~8
zF;BIm!Sh-#Plz%SfGC5zg6b0QT+Z<~WQP__E5+aGYv>xs!UDZ)ZTb1m2C&O!22&?G
z>^0J)0J`YaJfW?-6WVH-fEM8PfB^%sb7cUn_xjc}p)j~<L%KbsC>4(mMB(OVX3nqZ
zE+eNb=O!OOtUcV92t=`}K>EcY25M+)VDq<%WI2XVKXGg-^G_u@yAMwco`g~8T*gsF
zelfId1M+LJ(1js07YZY`ML-IIKWsbXgh}__h5Ryp@yS(F+sCx}gUP<7*8b5^Dy<m$
zjbVd<PtEDEu_P?Yk;{x)GsT9rVPX7QTHFrs&QNqH_&ppb^^4h<xz{h!>;Eo^M|Fov
zUYPEQ{r;w6Im7$qq_b!stgpy(%7P4s|EqE)zGSfuG5>G;U8uV(4TIjRe!bhJ(r9W`
z4+RAJ!TLk|qgl@JuV1WRciswZMMb0qs@-;&yyFM|{Ycrr^zaH&U9~hhU5~cd4I86Z
z_ai&bh}_?909$d|-EM1PcDQ*_ceWr<#bUZ{Q2XVLL@#<X0^z%>`+KLpsImm+8to;$
ze`dWCfpKy$wH1{ZLr-vfKE;qz^pxu=sxt#d;Pg*S`@HI&n>>KL;QpjX2kUJ5#W!j0
z2@{7ld1lv!_@rR^_%M1fjC(iDrDa%o0p--sjfc};h^|&GU%vd3kNG93PW|O9bC##G
zuwi8Z16E|dNx~<gZ}Hz8vgvj2K`h8U+-A>v_RQ$yxY!zam<uxeL0>SU53oC{w}0q$
z{wOojq5r@jg19ZK$aqqk^(x!`f%_fTzbNDgzNfNonnWGi-IjDhh#W~}`Scv<6(GA^
zOXJdXFYY#QTH-%+-%DCldS)1*Cz@sUC$H#wT9@Tv(9aeK9#?nIXt5rwza4jbOn@ap
z5_hH6pnJcY)_}F(Cg~x8{gPvBc}=Ih;q5m*f)rnQ1CvVnp&#~}!n#K0*V$DTSC&tg
znd2Tg)Ru<(9-uL{bWb>fEIPfkwHE!*lk5^qCBq_^^d_NRqOSNgJTiC}Q`klX3RG|&
zVIt09)UHKC;O#hRkkYO$o$~*lcfXoY7+?a)QrfO82nm1sVLyLP-&*l3gSv&xA~zUe
zV$1g)BOs-6n~V6B-Af-zz&gLx-v_k-VHD2JMnY_2Ohf{>N;jFAb#!-QAl?}*cE$4e
zd$k(4@`}1nbZrcWV+~B%Bax6cO4rp5ATa`gV_x`c@oP-P;7lp&k|YWun$z^dTN7tV
zGfffYo}lb|5i53*#UV!(17A5P!=i_m@b&d9sHa}GRbslE>ip8p!XbUVGvy}}ttQTs
z?<&#f@fQHxSdumQ`v?rqxPg+}8@Af~ddNW~gcq1<aFUub-AfJ1i%(l9LrHW8epb4m
z+p&kcp+nVCX_J1KSX6LK_O42|^Fx(CQSaU@uB@(!8=IROdOOp$wRSY#!2-KN3u^mb
zlX4XISG&O82SfN(yrYU?`Q<!_9Pf!?E|+1%#l}8diC*5rfH1|U;cuZ89qkJ_a=@0P
zHX<bq5PCyGnEP4Sg>rTmfiO3x#}9*a?g!+%R9t`Gwhh^aI8@k>9f(n-Bm~=wtGAqo
z-rDA+{;r~Uv_64V+W$p}GKSDJz(jR=y~I+pq_SLE{B4t@H=LfEH&AHiMfVp=eq-K3
z)>~~!J`wc&(zdcqKJv|{^&aA+^F?e-1e4{kBB9};5fK(*c2<{`=J$}}$<c2x3Qezm
zaP(7E?V_9u?}2UP_l}&{C8NmHT6&_SiC;bF{CFmB)-mqNRGb&kKcnmEWc*HKWVCp&
zIh9(KcE|O1SGOeF#&)6V7XmUI4Bvh}xKD4B;joUTdA{$t<fpz&^St-`=52&m5KV33
zUd()HL@e6-^YdJO1r<u#cI~HkzaS8I_ad%a3?7oul74p4iKHJ<?xUQ=1Usw&5<`)r
zLc&jCC<wa&J2uXfEHl~{VYd;gOC?r64_7qC9H&)0T;vH#_s#j9=Wh6nw9n+r{VVr^
zbYe%cxy^aAp|+kEah0GF3*EAKLm~x*L3paWxccWZeRot_nFep}33sq~36$u7A|N18
z!qH6aKE8vFpqBgehRJ<~nJ-Rc^u}4Rj-AoP(%@l&{YPsMSwV6^MV~Q!g|V2lRAHep
zLp0O;UOAr0<x2GZ@Rs25qH&tDF)UlA@COPs788VhBpC}6ai=u>y-U4`pJ`@~Z<1YX
zZGioAYY<7D<MmwxpPT+b!<Bon;Jsd*x<nqsi`WQ>RQ*IK0%~gP)Qjxw<67!omkCii
z1Mb>b%|wM?ToGxc64f(vI1%J@Vb|5Bx2{a@etMQ*OVg?+h+Z&WMEK8RP)=5M0jamR
zS%g8|^7+YU?CC0d)?uor%<cfwxz#m{%GL+kW`>LsKzBDZL~Ls;m0yrOB&&dQUq%U?
z`EY+wRz-u$lc=<~_z4TiPO$k-#QI|tWaaSi(EX_IhX9WZmG1{tOE}LQWxb$#5yNZj
zBDo=PlV7{ueH*Dabh#tOF8EbUh%PqIyN=wOufeC^XmLR=)1-qoM=~flLyCOn&!0Tt
z3e0t2o1sV&bJ|Sh7|IyvsKV$h?zP@N&h@rYF^savwwSZBt79Z7%Zu1w74J<lG;V!L
z!sU7Gc~tgsL2+3+1OP{R?YfG*3r>tlkvpVaOIFR4yWjl{3a0LuC+xEl?G2G8`yKx)
z3wjuztxc7>CM8Z~_G-PrFGfKeVt27x^5*TS5gfizc44Kb?;_hYM?tiRpbI49Df|gv
z8O<;n_WeS$uEN-)xlGg*<FV*sN?H27?7$?vWqFZM)<khzPmnJzqN6&zX=7EPNBgI`
z%FOJ0%UmIquCJyMu*w0*8weQk{#AGhzI)XOg~|Jg?`1)1gEX&5NdXIRIkak(4!HD^
zuvK=L)=lz&x;91jxx0%Cw5rJJY9gYjva+JZ(Zgx?FFM6`zrG~<X7p+egvo(k*_o@Z
z_pfuz7<<l(7XYxq2Lw@6M^F32Praf1KG)CaXQ`rhrb-5NJb0&@{P1qwBs4TcH0&h%
z0#OFmR{{BecoX1EquFNqO}l4hXK#&VW&+@7-Z{S>xsuB&c)vI9IgG<8`%|1tw?2uA
zYYp+V(Cj(=9Al#HbfMsJA!Bm*^M#o<@^?#ZNkv&Shsk6MJ#r!Gn{S%{75;&p?n@G=
zIx)Ok=Q2k_woOMB6!_<b)z(qmII&5YU--6Le*f_WsjY%&n3}YD?Y)t<iWt{6y}Pp+
z+BA_np&yex{k?u$Z&tfcb;Tt;ltGKrphpz<(!$9rS6{{S`FoOF;>XWyo-ir;W{5nM
z-~3~aK&;xUkv$d-%R9;D;2+mxM8!q$S=(mE<YaS5Nl_4ed^f`^JMiSMfAB?1^tbzK
zisgh*j#M!)SZ2TEGCWRxp{<+<Pv{zHbghdWD~r$E=3~4zOJCg&A2@86x<V9duU~!}
zi&QzpfaDCyFjW?&H~vB$J>8~8<v-o_rw1ClS~~tNt~;t&PNK8Lhb4RH+q}S;{URhF
zAb(qj0;oG5-kU)%A;={K{viwHDLy`at}4vT<)CN!>3*FsOATlm7)%06Y@SE2F!<UB
zC+Yj)Twz|$k`<dHxsdLdCko<*_bjAr0UT+e75|(-oUc#L#E6K?yKl1VYz@o2Sk=tX
zFf1XQX6jjochlmo^Tn68_r<E3Dl-%Wn2;>}1P@b~Qf?z~MC9|?)Rg8^{CC&M^H0<B
zMa!(5prn<86dA2^1p!c4+JtURow)rv?Hob7k$myJmy3sYshL*L-Cbo_8vj;_>o$s6
zcxcMXew3x+_Xt-9M+I3~HkZFaZ-z|-^aG$_qY*vd3&>foZ2WlRv60Y|yKg@UKLuMn
z!~`PsJ+Qmi$NbQZjFdydYZE@#xG)7qM20;d0uocymu7(Q4zV9%XEJUo9~HO<$=Iy_
z-Ox4Gkdv$CoJAxt_Ud_e&jG!vEsL1Cv>1%<1s=E7y7;j_nzM4M^O})10Y?@UT*sOe
zN*{{V2Woc0!Zj)S?t-W+V*sv!>%J(bhP}OrJL<g&+dVWN<6a%`bE&T82ya{P`tjrx
zZ4ckT!0qJwkF>P+CKee#_&Pfof1lBWi2+m>ZTrATO-xMkZeTNDBxaMQoEoN4J<>P4
zbu&F>&CP4$dmv#jZ#BGP8^_M3Y<=PV^}^~t`1Xrc&Ny-L@VpWx@y1NesMrEzvVb8W
zo@@iyV~%GSZxh-Oh{Xt{FG4oQ{8YK=%#}XL-gWuqPqpNVb80dw3dj`{)m<WGsm^wu
z%Qf6TSY1rcO?BT?j($`1Rz##ih|=1Aiz-qq1T6{mS!{It3nB?Hq!RD!a0PUys9Xf^
zx8=;^Q(7Uq$lt$P)HKxW*8iq}55<mk*#Uv7zI@J&;HAm-;(|oyxizU`)%fW6fTlSr
ze5zOMgrTjaWAS+#nm4><Rc2%%KZrU!Tz=rh4Fv}SXzLKB6xu5Ts}1e5cctuFnh8(!
z+rz?*OKtb^=(__0I%DJ1*vLY9@4Kz5ftA+mE}A0;QIH4)-p|XO_URrGQ4IwOf{YB~
zz0(^6!r+#XnkKY8HWe(b!bRl9@RV85gO{7TCi@xfUH{*I4kT}Xl2lQAS6&2LG(-l$
zyXJd#^BGA}G<4n0T-qW=6%~<7U%f9!PPfQ7-g)^K&y59+q@k2rvvGy?N?bKIMkj_T
zDJc!+W+JO&-oE=Ys8PAyp>7MCo`(A9zw$MacX9XhRj*Dg@dF3@O-U;J%j1U(bPTIs
zq_3W4DSklCmu=!nP(5E|=bcqOX7o;ZFXO1vh>Nz{-c9?p<%T2Np`$2_5>Ru531Vj+
zDRz8H&Omh~wo^%c^-M^df$7@FVbp0|E+qDd4WTctEkxM-21}JDyl2Y^S(!NTvQzH`
zj$iKlFa}MMN_tk_#@<*i7G+}FeXPG7Grzz8%$gsz^eE5$rm#6*QdB6%h=!jZsc~HQ
zf+uI?LHNwd^%bA&#VeAqHFI`F4SpuNWe;}b>xwrF#ng))+CqU}^JPX7mft8OWJHdA
zO;C-)SGY(CQK>vUOf-0#dAM408GK&yiL9`wSmm|F(Egf&>0l^LV!P>!_z$dlUse&>
zT;3QVydfS-lVk0FMpqnZNpE;x@Xi^yr5qeC{J3%lenb7l&E?hA2c)DEVbjozN9xEP
zAeT=>Q!{gImwx{;%w-{n`hMwViPzRENFA0aI^~DE%_yR#w%VBe#P#@5Sa8$<#@_{9
z#|v4$ZSQH`nW5|3IXN#vx2C@;cr)T@RuhI?*xBWH-X=6WTPSYw$lGjc%kui6Y<ZoI
zwJaHC8t((O=JvN4)tBeH-;}f>LyA9KHtkmYZTehgj&!j&db1XsvXXwVzAVoeRn+ml
zT^x&}`^Z}D%V`e*Ghy!+&8*a{ha7w2NFF%?Y-t|HSDP+Rk>a5RoP&CM^2=zW6VG2!
z^2A!LsT+yWS4t;jR7{&U6HMG1P47C}AS@;=A+_N?K+@YM)}IbB(0*=qF1&|8)J1#F
z4i<miKn7n>n+xaXPmP3@nOzcgo6@OX>zhKeCq_B&BRjPGNPEJbJ%%sx_C2Q^Rcq_j
z3-&=RFiO&S2w2Sk3OvwALKocgpEI6FKjegC+{JLI^mQnh{lIW`tG{*v%B%N3ye~{H
z*ridf(TKX^R>v}pZN$$WS^W6rMLUn%%6A_<^&e_-o9NEIN|ImPZr?UF{jWKz(ubMd
z4)I5Zf(lCE4%q3lR#i`X_M_N(ulMlp%4km6kF6DS*8O~!skd*}s3Uh_$eD?s9fvHO
zm~@|t%BOp<bS%=zG-#>1xqn`U1;@|lBkkrz4M@peRr?23HLdVb)ITg1`g`9gtD@j)
z8R=v9H81{W+V=xM?&>wZZYwCuRaesMkT2kV6rx;PHMz8S4n@*Un2%A^<GfEjHRh9h
zvu5HK62wnYkAIV&GF9KBlHi)213zR97t*p;-@dt+JF^WF!+S^{JVTH6-ycG=KC=5n
z4-*hT`2J$zYde)N+W7ybZvPIxw|=5V8jPP%=TY7pVM|jG&sp5<NUzK`?mBuH_-lj_
zol))OxueVuZFF3WW<Bvms3WJzWEP>s{fn*kOC$sat{v>6jxgLz;_=`UJ!f^Ab+q4f
zJ7`cDz=aMP{yaA)M@(YJ=>iT0gQoEzDt!ML@dhs2@wEK7;R}gxz407(U7uyhJh<Lo
z1&iJc<H>Jmr7Da%lLdI8tp;zi>u<VBzIpTKn%m?t-WNB!^5lEi|NC5Ro`Ff+>kFk$
ziK)H`I<x!?p`w&wMU5-Qy)!@0Upg-R;kES@r4E<6`V{1H=W|ZR()IGSjBAyn7MJJn
z^p8~VoO|=4ygKG^-B4Cm_T)p&YuK*z&4j+cV*T;#?B(m4oSGV~?kD^nFok1$ekrcG
z3{<PVfDX^MFmEE~4)LhyFb*T}-Q8YYhR2n5*g}tm#l!je^QQ@7W|koj0C-L;*jn9>
zs?UQ`{^I3p7)oEd4hT@-u|u(m%1!)|_}^uPt5e?`I`LjnN$LAaKnRY5+^gblFvs+2
z{_u;znoSeOuXAx%u`TUr{;ofyJ>0y(LrWXgx{Qi|Q}Rh5xs3j^cb&GmR;}!_?@zpn
z!;&L2Tp&KrjQpt}YHiM%x{?y%y(dirB{pNS*|6V3(*?4dkS=N9m(+QiSek-s8F-4A
znb9%5wuKhiK}knOMMV?|W{euALGABKDk8xlEkPC#aNw$wno0<R1L=i;q{@u2hK7WS
zMqtz8(&Bynr)*?Bd^$jP6HyV>kWry2{80qfB^jOv2mhA~psu5)*525qNErt99{RrG
zfY2)&yaYD;!~=?hL%Rbzd-#86N4-Jx57}qxv1?N`HAY6uMn;CH2=!6QjUyDqj?>SC
zV8TxVu&=z=D=ikd;r$~sS4&$?I9)a`pKj;hq@tj#y|Qm)uz#rE#KLItK2duHrT$Ar
zNNcYD9xk4)s-X^D16mRzp&%+Zq5&q-BqUcQ%Fu*?52L9Bmexc1n_&n9G+;$?Ma`NT
zDk{p_{T7%v$Sp061u4TyY;3AE6BWq`_|-?rFkCglWWPs*)6(XbM|>*5uPysoB4Sir
zUNlB;cI3!Rfdz8O@`^P1?2-T_Ib~`*Oi0baabS5CVDxTL_pJ|{Sns+5|KhN^#E=w9
zYy^c;R1JwBzqhis+^>18@%&d&Jqb>3<QWAmEk-PUsN>3v8GKjEQZJ|0-kT~Z4?(_P
zdn7Pc0Dajxxrm><2<g3h!`tm1nY4<=z9SaOf43Dh`RNx`!A7IAD5z(L{r$sx<Sc12
zGPsC_){kkkO!pVx6c?9Z9!w-g={&>3)U4p#yv230G}X$^ltF=sZ+G&sEEYe3+~#m2
z`fVwgwbTs_eP-_Do<ve-W=xE!E-n$8P4~N8!R2RCvHUnbp|<aIu{Bkb2Xd*))|&_~
z-`Y(eGr)8jM#O@s`^ZvGV90Q?>cCL6`NQ0v6Sc_a=5#|1hZ9u9zvUUG-fsFOpQ=qz
zORX1c=<)p*m=f^)p`Ql*>gFvrOhlio3UFyg{&YbmsH43#hwI#0Q``H~R9HQ%7cdLU
z4kRBwG*6bTGI!JkN|iTXmHndG$`pnXHMKxkR(SYjjTGlMg+Ua=6w-cd3CY4tA1(su
zFfqMWdGf!Ir5{;YR=V5LgoF`)2L_~Zj-aQ|{AqKK9}j?e!}$*Ym4O_NQzLF8{->6!
z??Rd}3-c=gGDNi!_|xSb&8ILCUn$ytHq)XY&RjVBj?a}w{=6oQo77B@iLb-kWg|fV
zWw<~KQ+pX{JC;%{E^bF2x()0v7m>(#wca>p^Np;zB>Zn(l2Sm7>Kr)<Xx2yY6_RgF
zxQzfigZ6npX|(5<CJ>)|LeU9{2(dm<j=4fH`totNWN_v7*zw;*(mYLITOU_Qy&<tb
z{dRi88xK+R{*<F@i)GSO_PVZ?_|8y3M@~8|s~Gb2D;6%|Dr8yS&-72V#;qBsqCN)t
z^9Y2kk)&Q$PTzY)Z@%}&1`%NqzWIDW+$wX-%*YV&;7_(S&cwid0_l%Oj)^(-eZ#{X
zBnvtU8WW?V3zK8Jzq7IAUIBsaUQ2LU$xA^w3VtdW<wZmc(T-U1<NY3O)1AmDcjEH6
zTr<=!7NQL6(kQvT{$XR~?$lHb$k2chU^ZE!!SOq#nhEZ*03k{ZQ(8pBP+W2`h};pd
zMgUnwkOB|Dv=-?^dS+Anos)zKme?3#W5Z>((TS(&Y3Mf}HZw?(uznag=?5e{ns%&W
z${Sb|ye!thH1SS(Ca84kDDnS29ZN243G0s?Vb8m3vwW4>D#JPOE^d1qpvpvcPc+0!
z*aLwN1dWQkqUJzm6b6D%P0c?!M_pds$jGSC$aaDFJ*>lr_Vz?{5Xz!q{4UFLQ=lmQ
z(}N{T$}lm>AZ(Sqsi2gV8YKy1$%HRs?}>Vo2zS}eRCm}B4UpnmN1G$$1K98h2D!`G
zGqa{htwU@{x;p_kf{n}#tKIS~!2<Df&3Hbou~tF?zzc6lA?{`7kQPnA+jI7D*V<g^
zI}a1*PuncK1i}wD%>R*&F1J$iup~L#s0Jq06lkQAbd<DE5yY(H8VZ87#l^R5qafT+
zy>9qEqR6fN0?pQ#POX?6F}pT!dQja7gL`yDJR8AFqlKNORvCCQyJEbX<{=F1BgL1*
zf}Ydarbj%~X~sslYcnH5)q|75v?S5Nx8IBpVbNe5t>3%cXs(2=zdf$j>NGw^!7QI1
zx{mL)Wgg`+^5U`AbxjB_ZSKkAjh$fE%Zl_*@RUn*zHl2UoOIk7oj6Xa%eA`wN#EGC
z+~zVee$owHka?j)t&*WfwP{ge31-rp%A@XxhF=X@1JwIQQWE0QQ#G_OzJ)2D341_F
z>3tYKO;(@AG#90mxP<kNN4IgtRh*V{uG?)It(3<l1N68yREjjK-ad-}Uk_fp4FZJj
z4nk(eo>sfonGp{Y@8;^-8kgs44xr*d4UsSoi6JHhDWRHLEY`<HNdCSpoI*Y|b>V1H
z=dq4+>=%NSuU5xK%W3R4<ALD2AJyaST2beG9M(NYf%$iBz<GDip1t}M)&o|?B<rHm
zw^F@bos62bdh3@HDs^1zc1sr#Ov{B`N!;C&U5XWMnb#iI32_qQof8T65VsLiYgG+c
z!@^};$zqcqPDOD^H1fS6Q($_UP`$^iDiz#g7#_`ARSEAKcdO6(I+q0>k>{i*qnN+J
zYm-UI?m1E3q5EIuh}U+l;-RHY17Nzrlcp~$_rGOj7I%n?M-mwU@}O!L-Mnv{gvFBv
zfvy%GG*vYzGjq?*J`=99<YvFBaa{kPZ$?EF_`v*^{!>hw>p6oiwr}Aod@71JN397p
z)}sn%^z0XipQJ$!?}ZPgnWL7vGK}HdfPUxIIiIgy4|%+f+r!O0!?B(7uN9RHNg1@*
zM74s8-Bo_v<S)X9>zi#718+q+atZh%O-pR-qwNpW>B3;IzTcBDH`yL0N%?6CtG64j
zwQjfz{^gzAqqS~V8Vt@2&z)A^-rDnM7@t)GuE7x6s#_ND2WXoZcSB#zOT)N)B<n3C
z%0!C+bhNhi0bT+f-9?Sv8R&Y~4czIn&k{J<#kW`2D<9o-;|Ew~*#Ap2tQ-$t^&t>9
zZ~R)rp1`{`0Ng!MR}Gvfa9Gg_PR`7|cU<netZPY2?>X-6dgdr6pE5BqE@piH9;qk=
z{^n+0@nRSt1?;qXE&OH>>1=dvF=JqAFqDdnBN2=cqXIMhi-tZYuDU>pOlaR}C2ZkK
zSt3x#XLPIr2UNYj2{iwQsjrTzYU|$KC>W@KN(mAQN`tg?w{*93gLH==f~2(4-61WF
zA}QUCh@^xdo!{i%`~Jo^27g?xaM<VUz1EuZnNQ^B7B0Q|9269EaX8%+q=~m2Quxry
zN*aDpn`i@;hUG*mrG}y=yUi#Io(TohM-K*WCT6bXu|U+<r`~9wv0ZC<$?{*{aKYil
zX8fuh{7+F3Eqc6@AhabpB_&#FdLu}bqU@j6;+kvxo9c||3@`2bRi8sUFF$|%U`i(I
zl4p~kPUmu}8~aq)S3ILF)k@_ay1(G#h|DzpOq^dp4#K^2T|~ZA?eU!|$qa?7XH`HU
zBhSfZCQQiGi)4huPaeCcqYL+4tGBe4gxSj`-a%BMBmY+Yj|I9!mn(EtvCU?tLR4Ag
zNY8K=D&?#Qx$#65v>miduc62H{2vdTP7H}+902jCAh%FIH{Z^+=(U}4bc}0#2`az5
zf<oCdc5Uv9l0r@0(6ke6Qzi5W{`bf4QYpLe{;ck*M4|XTR#zvQDNv-0_L<S#bEu3`
z93&(l?6isqftX=aO*#ab_(KVY@%dQS1=X|iTXAFguWtI@@E3f-NmM6RR&c{i(+mhJ
z=Umk+Y;4BMODKO97@8px+h6bjrwg3VMP58GO`{A#Ed9Tu&I!o<B1t){mQzUoy}H&x
zPiUOLH9i<a2v~l#>x|@Ef(2Wp?!5B2EEESwcgEx=zvtli4+?zQ2E4y`*wicPm(-Lx
zv?|?0+h2uZZD|g4v@$h1AQpBo(1yA1IStf*)FS44#3&-8N)~WOUP<W#qs>P(+NV+C
ziWFfnd=>nZ*iz&y)Ho;9tdBe9H8Ga1o*XbTvL!yVB2Hm|G=$lfw;eikwA4D)uFEx1
z0dQf?d!m#o<IRRT4G)gFHyzs@l{Hnpe~ABiP4)$bBAV<)AAzRo#D5!H+R#EpMdXzz
z@0pqA_QmtMb_VUiN<omhz1!&iz9VLq<_anbkxm6^X}#vgH$@&B8yM&-d{1un<8zgr
zcvvh)nO=Ql3!_wm&^E)W%_({8K)&!{vH&knHx2dp&=e-T^26J}@}^mSRWn|gz|RaP
z?K^s_uZyp;m^V6j&gxj$%dt?uga{+RJ`sNu^-+yK(WWpQ<_zGV@sd#&JO<msx6ez{
z`~|^-kH`DyxDIoBgZ`|o9P@>-frQ!7Ych(adZoE$ZjaOM`2~}Z+k2j4MTxAQ_neBc
zpex3|S1K$ewDJFy2P3Do<uZ4=+otP%BU&^=2{P&%phW?G90~%SgR?@{n>_x~vzPG8
zs<zANmVC-plTkH9S0}~R@4cl+5h^4f{o<)XTx0EPSn}Y*1~;*}@p2SfH|S+S@#3kX
z@MA_?au=9Qp|en{KXq2-DKCX;%u%(u&@3ub%xkvsg2j6GM)!o6kdWs=Rp?G%6F%XC
z<6=#C`cRP8Cq^}{{=;likn9L_=<(})vpl78T-Z!ED_rB%;uC7>Y9vM6fTT!i!%#@0
zcz=P5ondlP4F#F4KN+uf#+1I}h#>ETwQ4NWlqQ?})o9P7-?M;rVc)SIk;N8)s$GM@
zo9o7QoE)WLgm{Q6zAGk4dZ92y6l9UZ(NKGQZG9FZ<4f44pm!moX)oVM#yB7te)Rf@
zKH^jB7Eq>Z?^+LxywxBg7=tJDbawZm&F#NWlk524eJ4UZ@wqiN!e7gDflC?G=y_p#
zif>~{w6<otHCt6WLb_)~cxS0$<jaIaf~kWG$>IK4&g-+2-2D8Pe>Ljt@<J=0ApE+r
zCcoQ8#`T)P3~qhegF5E%)SsP+g~s2i=RU-+m}tAY9M*FHkGpKJQFl^)rd%+jnxrHL
zFZEo3*k@&E7h#r+_<#pmY;^1f#&awXh(LSkn-0XuDJQg-j4J5ClA;(SWqRMOF9(oa
zAYSno+4(u}MvVIbQ}f^M?W^}tU60ox?{h1T_QMdcBTt+rQ~$=5Y)x!h7@NO%6g5<C
zhak@%Vw`b_qK14J#{+(65|3?Q`?B|GH2(JCD%@xor<9#z01yEV0<$$nXWd7<J+!0m
z6!ET{RcSZ5<kzQvZR^@GUg43Ma?L7H(9sb{QxGgp?ChlJ9~=a!Pq_*T@=brc)X!ZW
zo0Xkez1)@yNiEA?(lb;7f<7GXTJ&2T@5&R{3*&h_ci>w4wCl~y*9q<REY)58E3;bD
zGQNB$xy-HE660=kIW{$Rg055#h7YGLMs_ziJ$~{x7nf6DRDI$AvK-g`dK59&vF4iR
z{?8H9Exl!5Oz_Fc{51^5&qFyOB?TqkXZtM>8ZuU-v$QrXA@L4!vkOYb2r}loLZ^Uw
z>l?lqCr+-`MtS8mm@G9-wvW`@NB9+mJbQLpgO~e?iwgd#H{3)>@4egN4IHmFaThVd
zNh)?Z9X?vNEg{;ls<za+ipZ&?E9NO#wq7IVj{gu%faSnqy49=L?tAFb_^xzdR2$`9
zO|rn4^;;s0N&8(NK8SX4;~DO4b6bwx({$Q&{JqBH`13hDr36<6urQs$sy%(tV)0EF
zugB>w=#W(VQVR2np)>yM!Sk!mWEY@%=xSsiqQn1{S5euSb6!ytv^%iYR-xBKMdak=
zAK!g9z7AuJj$Yt%2n&lIZA@EQSoyav+zOWZ)NPa#hQ_F*d?~p2@ok>JRBh+n;55x3
zCq1R%mBXhxyG<7-nZ}wtTYX6lGc%5=EArTHw7BBnQ_D+}A-`L_K2AgPAt;FHOckx?
zHUGKy%ha5+IuW0L9%|UMuDwc5PA)AiCCcNpKQBA=O?UF|>XP(m<fpt(>i6QQ_bSC#
z{mPdKn%u-T*5<w~3+U)UnevWiO4R;>ZN2;;nm|K~Pi&xhdoQWDBu8<sxzOfiq`0sZ
zUD4zuXPWr8m$_9HpBSBQGktuKr2zE4yxgVir!ga}j*ffO)LlCPa`GxmJD-7c@<q<*
zlDyevC5>&$0+NR8>?Md427o4`Xr$Q?#u#V>cVJ?|@v!jvH6d{RkYqJ5P0h*8RfD-h
zx4xtB6sUd7Z(oaR2kVXY5TfS~H}}c|Yl2n4eZbV*{AEN)X_pK4EI3sGx3DU!D2oaU
z9pvL#mzB!!=xDXnK?Zmr7i7##cjp)8Zw899P{XF{#o8PAF^}=1Ff}DL?z?mJN*JTy
zp^Ff`J15(i>*z?$`>3p_DAl~+=uq*oTeO=`Tu8g8pT8-yB5`C;Fo#AmC!0S5W<ta4
zEFf&qcb4=em2}{Y*!{SDH-faxoSmhgEUNQPw`d{By?2y6{p9{V#s>BPAVax;dL@nj
z>HGIrLf^J5WLB|FDCe}jG&RdtRZ<2Do|#g9JV6FnKnp(|Fs*PBdTEmP>-V>ge$WcF
zx}B_P%PA{^eJK0^HYRecKj4}+9^FYE`4jK@m(0N{h>~(NE6Xs<b&QnnR1Ly+wugP(
z>%xK&4tfQQ_>+_Ip`p@)iPQL#0HK+ibMuyDR@Beb!_T$qzpJ=@;}?5+e*VtN<{b{4
z!z*86H0S3$*)EiUOk+}FZQOQ^Z=&y?0Yb+3^5p>GXtw6%|2hc1@1;NNs|C08Qb63)
zK0Eq-+2qVL6BmmLOesbUhcHkDcF@tMaCQzoCVU!+CPu11O?)6j5GZmWNku=!h>l@{
zT(49kQem*Ma9=r7)6GnJ^*Vic+z=ZOH}Y}|BV)hs_ge>+?Q}I|jRQpK<o=Qh4&Q4_
zCZo8h0)|9-lUZ~ZDyl#RS1=G0dq11OaFjn6HC))geSV=p>no0P3uz9qRVd9`)FR&c
z?tVODTDh_55B~J!#B$hYgDh0bI?RNGSyj)7Q^I=$zv7*$DJ=l)o}rTbWbZrIIQ?n4
z9QnO2?G32Cy1A(}ALlf#qZ;Mq6<JjO;T(U2dc`b2*muR;?DYS`<6rCIg<sqGTk&b)
zUgwra-6h=r-kT7{aT)U7)!6@4-|y#;+Y*PDBLJa#it~SeduJ=X;7*e1f585E20tHA
z0pO@Q#}2#{#BBb5(7gL$h7}7uPc{sg{tUiF=VB{hnb+}2h>>ij_W5s9i?e1d2oi5_
zciLK<w>CCEWlqm8U_9R6A^mZ&vbalOV6Hz6G?cB_pIZY(1qHE77qe4d=>CJ%ao6`T
zFJueV&<J{pg6+4GXCQs{Jj<{Xh*E0te=%ToxM5j%vb(b~TNF&NVDn4=LUnbfylx-!
z?8Vt~GYS!Y$hH50*Igx3N&xYP5V1~Ac3nBiFUa6=8h&+jVU#1+;l;&XP;f!z9-`Y_
zCSng{4ja{#%_V3k$BSPbMWVr|_uTpd7%Fdd_(1p40VHWCQBOm1p-}_m>z+jBTv7Rj
zGWW@@+H>cn%Xd*~+^R2jy>5=Kb?)tpr+a@^*bL>oo01YrPk-|HH756NSS-iV#V@nw
zUW8EHhw1zIosSfuBs^~4`oqTfus&I&T<kx(dJ;;+nhua))1yIEAYDU)aXdQG^IMgJ
zWBi;+?>t9d`A8M_-v4GNeamC583Ut(S`8kyf(l(btc{TMeZLLfZ+$u(wS*JFq~N1*
zR`v=YJP;2+xY=aP0<yWa&u2TG`=dOrcn<x<ResJhaBG=$iG6kT;wsRlKEFC@>pQr5
zL{CR&+0-m#oZY5N!0+t?bq?@{7<Sdk63$OaOn~JRrpm_5>h$|3x9zcqSyi3x+sc^M
z_X20c!rbR;BVT6kdzEruWSE-e&cE_LIe7TLdgsDVmw@?meR|ioaXDSt>f**U=f2Is
z=r(}g4*Lbpj(1@O1J!$6iND9kh6KShT1}SakpxWOj-wzN@dQzA;Z;>-FDomi-Lc-`
zbfE;afU2vc$;9T$>-Vt{u4kh}v@Hj472?9_a^(BOoPpmLAl8zLhwPTcU}vY^;GsN7
zZNM)+t|S1++6IgH)LkG&bGEa?Lific%SdT&XzS`EAN(NS!c*%gf<kbxW_6fBS$dc-
zRRzMMJVvFKO`d2vtwzK1sGL%$j=tme7(G)$K^T6{*1MhP<Y)B|syN!&$)ye)?{A_Y
z&z-k5fH_`q1c;Y5Cx?oXatf#WQNct+RtO=_*U>gfG0pEYJvdy8yNtY4Yi6{5X|<LE
z(`Nx;2Jh0ZW3K1$XMnH=6#)VEO#?9}+f`J=WpAm)FR=ZFFM`n0bw3Ry33VjIau9aP
znS{5u-k}Fr0+^D3CgoXOWMzy9z^lY!d<k3@5c{kE43dMdBxiqu)>a3sfPgP4xf?lH
z_JcWY{2|7r;E>0<qB=HKRK!SmCFvFI%t^ia5L21m(=fsZ2W#3QMkBja3nXGZ!688~
zuY-Da_C^h}Ubi?InID)FN~zlDO;i%CTe>-HIe}CX@Vn>Z8r7dT2&*^S-oA3S-y>!d
z7Ez~SZ~7zJkpyLW;dF<b_fmxf1iZo48CUF&jZ<y>L{&uv>cO(ScM)=wwI`%M0kAKe
zE|={~CZ0i%%F4Z*KHb=)nI9<0&_8C$gQQ3u-s-2@qi4ta=|eamQ+*mj<ex4sKJK?g
ze{Lg?p^zg@yu9*~B7)SmWOmwnyta~fgQu{xG(s@#HX+md%cp}lGA1cld*{^_r};Oz
zn~!haYzR-EBT0vSjg?hYsr&C@@G>F(%Hks4Ltd|Ib-!Q%`*<|~iAPYs3rxG7KP*m9
z{|ZE_!9nH#Od^M$yJ+Fo+DQ*LKtUI<7z1ZzB8yiA(28Kvu<0k(nYW7g0Jy-;%E}iK
z(srDwK=oUNz5&OL57JQN<@MTz-#M|VJdU7TI{My0!N^zN?NCt(ciGBJuT8&_J0!&I
z<orI;+%@R&^Rs-_Xe>6m(ZVDSg1qMcwDrT5<81d}y+=+tf_D$@k51hK4GoofQ=6%x
z0*mS2)%M0R@=9_pJ1i>H8=1Blv~)cv2zc@i5`sl-o!u%(U(a~lo$Wr+l;m={F6lD|
zqaow9wv|9s2ME*TMVq-!jTfO*`=+T47xaI?l-=3sde3!mD_BEy3D6K;56NFUE+ABg
zk)_L@KeoH&crEqU?8Xye>rfdR<_&a#QQy}~h{$s=cQ`Mqje}!Sg1iI7XPJFla7PWq
zFvrG*W@^nY8#m??_^<{)-n)T{H2VdFL)%yS6&@*UY-}F)ZS)f;s@BU5&;7r%1K}AG
z(#GY<Y8@Q~{*iuVbeP`U(d{$|XkP#qhx5>65S=u++x;Kr1YQzZDYl<6T(>WN*2|bq
z7dT(#cE^i}Ha0Yt@>36az~RyLjnV6MfX4Xx%oH2-!0PqzYv0+(er(rOS^v0&y8W&b
zK?q{pc743fHH?3}Hh6pOQe-+=KeMGMSeE!&>^2dLb%iqsB`q;bzK%!0pr{z2M&zZL
zHCw7~3NY1t!xBMYHs;l|VL;KhwBrJPjbi&-9It*EUeczi?qgirjOVt06qGLfc>Cp#
zO&~zs&tWjl%`4KbR8kc?5$zyCnkSW|tcvKIk4Q+dZW5}Tte9FPaocZe?ojrd>iC^}
zpET_-+&N^N#eAHMAb`#*<PF2{3<4jetXL?6+8W$`4-{Fr89RxpU_fHnl6XR2BHQ+K
z)^7&gCe0@Ej{rkkuv^7tgaX}qvi4Yqcr$PHBLKz-(tRb1KLg{z_|~7&ArDdn8Q0LH
zrW#-*>`8ZJ20@)nQr(-DdBEmDCgZX)PK6**6Zi*fMX}jS)G7Wjt-hW*BDw&y5z1Gp
z{QU!f9fBfL_8~?hhmE@DRYj<C-4C&t6{Bv&*{rOtS{`U?sw<z(D}KeK%?6F2Dy=q_
z131tbjeH)Mjq4>&V05O6YcH=Vd+zjV>q|xslyGn$b9+&~9KDi~zZP$;t)0llHCyXx
zbF$<pYp{^a<Got$G7CHR4d2@j#{2sDx7Pg&wCHVM2|bQjS#Pxf76|_{qo4SILOjF(
z7r+lMta-W@=VwQ3+Lw~NPJndY9{#}9bS{}ASOu^+G_=W8gX_M1Ixi%?v#>LBxRSq`
zFJp!M7xD3QuQFO*BI2-p`$bN>PCqYlFf^y{Cn%CypWJ(w87#F7aJ``w%l)>nL~gzA
zWFNmfFP3~cnD<Y%)K<!I1+YLe1@D-zjUtKHy9S8qfWK4pXvxhPKRVVd)ZNI)$Z$M7
zG&6eQmOAG|tc$hBf{XlkX)`o2z`HgPY>?9%O33>7Rp02%p{4DkE1=_o-@jx>QBo(f
znbpU~uQ>;hMv6myvwyIj6qP{YwDoARYABh>b{z<~>|1|KO4Xq4oh@~p;hak%e{B1$
zUjQ~GGG>#MxqX^3AvN03=$i&k*B=fK4nKD#f3|CV2xANx45xtojcxS?Mo3-gS)xtk
zVrE+VDEt^^WWT4rx6IJ}X=X&4M?)=4S=I)kOjucm4>o7JjO}$Ay$-HwD5MXClc2kt
z?F;eu-vwYuTa(>-?*8T6_w6;~>4IwD2!N#X^QFwr&aw=DG<FH<%fTEk7$M$`7=RC*
zlT#z3K0rg`gLbXc_wM0w=b5gZ9m($4h~$sj>e|fuD=)>aa{-^kB!!qhiRf}z_;WF#
zh;Yn#qT+vCfU!dLQkSVseuRJew=*dI;O7`&@ZQtYgCJalzb_dVm`UQ&eGkW*`rfs?
zE?+ts9v%gPyQ5=U7_pw2_3_H~P*DC_ZLJ&P1LOC)1Pge|i8MzE2_MWh+Vj2Ys|tva
z2STNVDm~+~em(~lHF<g4rVFRn{8$fgu7cG*fZy~PG}QV+p#7A}^4vS-!3fvfC|@z~
zh#VFL9_VsUGJWNM%N^qWi2-6}PMe8O8e(GZTjwoG=&jCYl8)CFevTyYz1lj*dxZ6;
zb~64#$>#D?>Z|^`S^ety#l)fYfBRq)@zZko<2h}_CqG|OOyf*}@D3=}--Z5FXPO%y
ze+I{oSya2;;uh@c;*#{#^k~R*y{Gef?%fTIBF%=|0WBLlm|s2EYinvE#WS1z+R5)p
zn5iEDW8FX$BhDs+k)4^2nyR(xx<G*Jy?3aE577VtPUiJS6L=3&A~#Gzrc~W6kSe|~
z9)hw}?QCed<ju_I)Z$_l-MJrw6Y{#UV<tUMF#>KOQ0wL9F(QbhL+x1VIRr4a%*v2(
z6L`Fu+XR&?*__LnrkawzzF}a?eB#56_OORehldaD++i(M15gSDvfNTo5&S(@RCFLL
zHhNW0n#j8u?*ptf<#@P$BxUL6xjI<{6?SLSUC<k-BR5oo2bs%&v>m5f*`xk-x|@ZE
znTEXtL4MTD=+;hHzhnaokInB`90p?Hf$uyYoUNlFKOhVcj?RPCMcZXBU`M6VP#<q`
zbz53?clT}+vyHb+?zCI7L_G~@^rE5NZw2>_=GPeuqTzCKskW~5@?>uTX{7pT%Yyuv
z?CA&C8DnMaWGz(go$$`4&}l*S7MkFfzn@hnYK}lGkfA`OMURPK?d@_@y(tcOQ>0OK
z1LAm}5dz;EBYdjiz-PT5R1>;&ILALnG?C_rg`BPF`jxfy-A$qT*zC8yS2p&h!n@nr
zpz;e(xtCSUv$T|rS27D@JvwUoq-w(~ABN^GU~EFS5FNdel=RRr;KOsLU#&FkAhSS_
z>tTuYgmx2yUuh}WNg_x=2*2{@<~!lY1l4D@>*5ANekA1s)3+O@hKEBovS4<v#mS*B
zbwl8N*IGPG6cGeA%ZI!U@3Y^!sk<6AH8z05{wH@g*vh&;O7XrhCNrL_b0gceM+<xI
z!1DYFD-rUE;?IwTlWHh~o@;GhlseDpNt0V4&2viPl#j67wv;o2(QxFk1hCz=k{5@E
z%4a9~ik<g$N`vpXZ=q$1Z#5j~+<#d<N8-nU@gz0Z<)CNlD0yqblC7M%oUqrE`7egw
zm(QMCj}lr5c<gfLirgI?J;Gm~p)_0h*b4|m^i4=1@1qjN%Io|B>MR^APcw2DmU@zB
zTLy?d$=+f(HJ$vh1(q^$m{_NzrKO;y5Fed*d@St~svO4<Jn($a0eChi7(^5|9@B6g
z^qz?cU9W4pb3@R%(Fo68r13N^<~BkuaRm{$F=08{X2k`~Dnt%6G;oK6-WFygKt7eO
zJtq4bj`{>u01hMUJ3qOcx4S@C)qk#Pa|Nb#H`6|e#MB4ec01zyah{Qb{IK6wXyTRW
z^>ov&{TWa^@F8MW(S%N1QpQTh%Th~F+|)&1LDI!=w=VSn!k#6A>Zj{Zp>PLnW!hnH
z3k5@AQEe)|w4}M4oq2qA_Hlc|>nlCwCZ}WzgEQB|!57KIOz*7*=QsNwp<D+r&5#wC
zt&uIcyO5pc(vy6(ee|^P*9C>B5X(KVh0`ryA*7K4^R$7&>be6GXQz{Aw_D)rTow$I
zR=jqHPbG+g1^|1)B>efV>TC|Jnq%2lG)q*tRlIaUB^FoHX+4>xk%~t^-Sy~@kqmmB
zaZ6lPnXtZP_KV?Ns{3x7oVOpa-aS*oK;9nLTC!OzOwRR789fo1s8Vp8{p{&I)i$Eq
zn{va8WPeTC$29vUa_Jr>9wbNE-P^g?)rrDyq#RQlzcwtf03jdb<RG+`OhbGk(gb9z
ziP8HJeqp8Cll>zD=?ZyJ`wGMuGCi=wmC}}eLIcY5M}y2vPo6x9!FJX0)55`VmEj>e
z^6+vqUc$p%?d$FpAiN^Lle9e|Xp??Ki^rg=`S)b7QHDoQkSv+o{oS>lK)u`&iuhIj
zNJDdt+QwuI1PfJOQIRG-=n2+v@K2)@AkX1HSRr}taj(JB(oz^R5qW8vdQdw7rZ?vC
zjn+eX-@Cq}_+UqZ2w`PgE&AXQgsbOKFxzKlW#Qj;6A~gr7t#J(_gDV6Lyg_Tt32Af
z3W9=P<#ZWHpON#B{T>*?Q|%?eyc2%gi)7Zrt7ltrnXTl42juSLET>;GOlm5R^ff5x
zo@Aw`Cmzm!tjVk>EG*R0Ri${u_>nZQ{?x-EA%a|8ckWNC#Z6_fRZe1&(UcN`KcOt5
z&Z2({jjW$d7b-&sbSvr5-H`6=J{6VpUToRJJdEgwSQQn8UE-euO}Z)B-^H+?bfzQ)
z&v|+(Ds@#=3T$+VNK|yR8JQlk8_?Q?&8#D-+1dM@<*I(u$4y=*GOA)z)5E{Qd!z>{
zK+2zzEju_Y7YBK>!%D=wUgt&)hNh`;@v)DN7osBL_m%HWkp2q5C@IV>%+D9)Lc1c6
zFcwYvM%~+caf)|VP8SnxOqeLJF?@wUM7Wh@OULxMELHEL;ZO=!q_A#mP*4}u{dOR=
zNlJ=C=r|gxe={BxJ-zD*eaB3Zx1mE}Ox9(XZ+v-nPQg{=mRIsSw}95jUtP8?=sT?!
zB#MsDAAV#wi8_=}qIO)T+#;&OaBwRBZwbtuHCjewW*WMOt@pQh7?8=UEBjZWkr=%{
z!gnL(a9r{Y332%Q?McFDQRYDI_3-4bj*hjhL6h#lv>ra-6e<i2naCuOb{Re*@BI1m
z-T*TPQt|N<GaZk?bo2z9ingUyVJ!VA2P4N`H7j9)`CeZ=!O9|R1q))BbDynDJOuA0
z^fcO?f1|*}y&Fl6XzkTK!D88K0S~@l4U!o-P1QJEx!m-0OW^r<4TZg15ET(=&(6;F
z-mWC|-{`>D`IrkZ{Ur9+8|XI)IjrLwk}AQ>WWUE^Yuepm<wN*~zMy0?5s|xZun_Tk
zZ+?wQ&<#vhlHRxyZ#12$kW0BE;<fT5l3Ze^Wn;YXbGUBWMM-|Kxh?u5iN;%Q7n?0Z
zt64N&NB$zF?n71DCcP(4<<%SWrhJZr;~ckQ(}lSkRc*N4j=v7fRwXoq_LX|VMse4O
z=@=i~ilvWGGE0HC!WY{s)PnBOBiqv-#c?)QoYp5(5>!8#p3@Ep4wK*fl~XJq%$>a<
zVIg@13-xv6NyX<^;mM61hwqP%2XUH0x@jFm-jGi?yhb`O@ovs=d#A)p-VOa#YrA|t
z-@paU$F<M9H_>kj30=#~9H+q8s&T*A9_2d#u4$oG)t_LlAkx64J6UphUrVA1{ROkq
zI~XJ)zwR3z9GQ@Ksj$8g)cac?lBMuwPf%Vy{g2$=odF)w<NVZP#s?t=*PR>g+?Xu=
z&2O!9?pwn=d88XJTkCpqPEw${KI!b@Bb5RNb*y(FI*QL$3}HIMVIDIc4NXDgA?*b{
zfw?ZngP#5`e%P+{R^A&huK4X@hW}O%zQ&i!$|~@g2kTlZZZdLWXLR;_zwm7}jlW0?
zR&%aYtFdP8k}|i<#k|l;Us>s74h?<NH{hvvcYtz8z(a8G(<9<{5>$>JyvQHyU%$^e
zc9#qQG{cFB5qP8rUt>124PM*PAr%hFj;zSankmd~BFW8I;;XXTlH9|Q=30w56Nb6e
zdA3;X?)jGcX7G~3pzt>`>Sq=zA)bM-7WzXSDG3P+OH1q<YXcnw5)wDV3TZD|JYr=f
zrTQE;XPD`E8L6MbCX-Q7NV_P6E4wY+H>j&^Kn8XX_g?IFWM}1_&*$9)Q@e6|JKctx
zE_+XZgnIvop934C0dn%E2?Xp#<uVxxfCkK6Lq+j9Ty`T9^1N>#bm5cA=FLnY!Ssk~
z;_EbO79|9eHPL^tKOY=PoUVWK%`gg6s$P*2+knh-_oV(X(6=}YH(oV8BST$B{<U53
zjRayvP2z_s@4x9w1d7U=ndjo*;-Vl4t&vd+KqWRdTGrXCI~yC_gX3X{xopLuwl>~e
z)OhqvOm>UUjniuI`PjK<>o-$y!Y(Vbe-U!KKTc3xLO!fT;)UefD%?M7#x@0o-|E~i
zY!j}QRj`aKF4nZmCW+()hTgdYgi&6H$sE4Oq^N?T5f<+7c18a3)5*^-Qi3Re;Ay!0
zOWx=48b6)%9@vhs)s4J(a8si5>h<0G4J?dIA+j$-r@-hPz#jly*I?aV%xA0RzfLX=
zsn~_a4DAYvzaS$M3ml%K##Ve(yo@#8O}q}AL^JoEie9@(B6C(la1)rq=~k@DIGmwx
z8^7y*3hZ0<BIjkCIBbdYmCOQlmhN7izYe-T`2^F;3G*WY0&{M(qx-ze{a2exBVdfX
zkEcpYNnQv<cIT4K-UeXZIo+T9`2K@nq@lDG9l<1oCy!gQ#W(Gzw7=|!*vdw?Ql2_4
zEpB@jW}2L`vbBsKEdM;*`1^OLXsBSH_8b*!wD0}lef-C!wIZZj!1_!I)Kr|Dxp|<n
zq=pAG821~BPyMNHk@v=#=jzvQnbGjgkAry;kta=bXN$gDti-Z<=ZOzg&#WQCM5;T_
zxbH)E^h;n{)}*C<nHhP)N+%Re`KBcziP7;Z^ee{)zsbn`Df47C+(eriWq>CJ>!QKk
zkwRiH-y&|AK}SruH8VS>M5}6Lx|IfsEs<MGiGMh*BaR-8b47Jhcs)~<Tc7wzKXB{a
zhM>ok&+K$OE%xt!7dHg>3+k!4L@P411KfGlC+#({+r1^K4#Jh+=|`lpDNTY+)Ogr9
znA;D<u~37bsH-XYGP!xsWf>Vk{HDDZ^!9(PoyFXhac~AO*P&$y4SUz*xQr4i^Vfj{
zYfo>F25)&58yzjn+iCT;(p>-hq@?>PA3QdelFqoQ>oJ4j>Dbx97tfN@Y1|H|BAy4A
z%!$^o>*p;_bYJY}2P?ChdO=>--P`+s%lre7@Xdg0GYvKL%6XCy_e_Sq{>(8%0d(=Y
z?&tRmzICZx#>3t@^1W#ws@3RL-by#=yEQ`SZLJ;`9U1!}da%5%Wo2Fn*r1wP3g(uE
zpR7m9c&>+KXPKB<m=Ub}PO5XkK@1HuF*hc>*4A6!_rIc7Ef#YQfRFd8Phnw*U(y}q
zLjr-6lvJ7p5z&1(xmMiw+J0?k*#AYRrl#?ZXQLvopwf$S`Ht5oD=E|9c%YXchKqv(
z{$F8#_R(+LCg$ia$gWZWaezd`825<9Et{*pq8NYe99I)!BRD`!62BJDM$(~4_}k&j
z!DP4DeW<-|p$ETp!Kf%L1Nu_&2RmXuPe#4&K<Jl&+%q~_TM!zvm}cNj@vG5w`5bCj
ze1~Jr3;qfUh9;lB<?x3~IK+2ZorXURmN9v1TM^5qeg2H<SXLJK82SACz}k`=_JVJ3
zv`?@48p|a8`QFETW<A5q#`5a-*6HsZRx771iq_UAGWg72vq}`9<A&EaH@6rt7b|j1
zuZ<7Nd&darndYC=XqEmtzkW2r%9=tryj|yb@Mo6D?|Mz!w<os=p8K>$dh9=Pe{hsn
zS6@j3A3Oxs@l3n_(Y$D$pY=QPXe_RZh3k=&Ra8?mOwIYdx~ZYAoJGnI9_z8DZ=GLQ
zh>9pEDB!|4Szvf<EF+OsFzpZ6@0MrFQd800KcfOhTs@qMTT*zNJT4)mfe#d)!mtGh
zmy6$zPv`te-*idV+sKz;Cs>(Vj!$e{wgvDtDLOPVlEm$5=Z?`}d!l2eKJ|c7LnvqH
z$%l~l{Wk`PIXGl2we($GTj9fW=du9a`=>u0WbK}CBtv@m9GbrzrBZR?H1sweynH-l
zpnjKEm$;kt%Vm2~S6kXK)q)Rbgi$$td|vwsrxREGW@$R?uitt2(!@YnL8EG~q``E<
z>biSE&`wat%6KVRfP|#vdChM~nq@@Z_={0f@4!T7qQC#sa`ONfqggwRMhujWjIc83
zoWKi#EEi5P2w9_XJIC}Q(a~kVXL;qLJVyWeOB%qQWUalrrlzK;1Ez3Z7TK6m(lD4j
z!w*m9UH*#(8wWuCuy^>2v;rj@{sX?YuK3BDouhn#@mnnJ5<2=0=xfo?{E_CxuGTh+
z^c;(8egZiI@f6c$q4nJC+>h6UMsFfWnjAjvT_Yp2`Ijbtd1+5a5`5sbSYFOAT6o_l
zyIPIXt6it|?Fl*Yb361NV}kjOyzKnPG&E7~KNMzW=H{n^ums9yc`dDb1h^9=HYL0=
zJW`TU&g;B^S14aRjS|gx7l~?bXarv8{as?%=!;{zOFDlVa3A1IC38{J@h~$|jbz!B
zSYuR8)VTg;-{8L*2Wh%jgal8X@4ebwP|{XxFc|Y1X123S{Pt`3Yq~y~tF@Z2lC)Lm
z!+`r(_<~s9b~_iU&nl};4y-*BYnwdrJhp>K%!&L3Bcc}SZ+Xo<7E^%CRt=V}Jd=aB
zZ`)YCErnXpcW$9WBKwT?3CG@gady7lpALx%VV0CW?-Tl5I9ljE-3bZT<{B<;(EVtQ
zi6UHCOV3Zg&EMJEDKy5(MNtg_VZ4v>Qgg=*e)w4Lvpq_?dp$f6B;TkATY|E@Iu7An
zH0F&;v^&lUs)|XRw(Hp&b*<mjg%7WwnEU$I<iTDBC?zvfe_U*AP^i@V*jQ>xN<kVO
zs2b?w2=R9tAsL3DiIjr9Z)j$uE&qG#cQ5cyl#rnM+1L8AAM!FZ3TtcLeEYSYs8xO5
zJF{+8R&LL3c~sEYNbEUxbr$U^qUieo`yMsbBYE99ES&qQs_ou;`;gtPsv6h$;^{)4
z_H~89K@r&dmvam9X6jE!7h5bLJa}^Dg9~i)+1YI-3|fr`fAWBD67}VU_zU9sjja51
zS{ky;yAHP<?z*Dl6!+ul9rwxc!=1{eccp1lK3+Roy_;uZA~g3=e4?@IGAop#yT$9f
z3)kU!f|u1GSOXjgH8Lb5jIJctQ1Y2Xq%hf6gda*53F+^hJ!2!Ic54&b$|5zOgs!i@
zGDAF2YFk<9xU(}2Yb$#T`}i;*a}`X$I}Iy3&~!93aIx<#P3NfU<|Kp6h0AFJ?~bxT
zxjM{gw{J@HgvZ6|<Kf0+W|_P)Gy;J=-0{U4yNCfsSTJdN^1xrLsCZWAb~aj>;0!tx
z*bnse7cP^33!u6P$Y>}oW=rYA3=X`4v|BrAclQA2q`b^NS1CU~sEk_`BTMAEPhz5$
z(UUB~>a8@sE8;KzxJQy-?hIyn`)gLs^9wnheTlc8zRk@1bvV${`mkRd>+Zl!-R+Q#
zc72IS^^e{cfA#;ZwD~ZC%FF+joAgcEul)M>H&Stsv+k?zL_G(yQZ)93tXJmQm(^8s
zlJN$s!w-3MgK-|$ZBk~E7@hCNtL<(rF3vBF&K)g_&~Vm0#@}CnDssAk-a3e0m-M;)
z>gZS(=dSu~hsxuFjaf5ozstfN|LqylsB9eN9U*hnfPL~IRC&z8%em3p=h_kG^fhpo
zGGWkJb>YB&>*?*lDJ^tHLvg$9-@W$Sk&~6A2lEGb4(T(+=+8w#L6U?+IpO?AO;2y?
zub<begsLrb``663TfF4Rb3GCaS^n!KnmR7rL!O=%EKRp^rnlP7%nfA~uyfY!>7rot
zzZ`GE^B0`&kaTvU`}gmBl6kw#d3mX76p9Mb7rx5ljQ{NQpDsR>kYwmFedHB806fUa
zX=rlHi+W^vg?E=P+4r>l`orn+2Ai$qo`b?8&v#O~IVE=cbl!#WoVK)+AK1*af`clb
zs{yD1zu;IYBc;ZlCa2;423Roj_6{UbotCm3Qn=?Kob_|6Yof>xs*aO~GZbWh4@&xP
zUz>s70nf$WwPK9ONPYN+u>vvniuvJ|(sGM8dgIlCSqiKp7ep5W-~N3Ijd^gTX$n+;
zRouI`;ik3a0OrMuwbGgwlNP3_VmaRWOJfCN=NvVzJqM=w1(4dHySR>yu0=(~Hu+Mj
zG@;3HtwRTPy1zUC(f9L*cSf%iHpV}5`j$H{#CxP-Yuh=5S2D?=-nqQq8uOcNz*H%$
zuzM_@Fi8f8Tp6slEN|YNho9(|T)PlG<!ppvFh<g9GIjou4gP+LNQS&}j4czvKOgF|
z!mCH7fMkKmKX8FzC5&4-S!#Lwb-*W=pr)b8qo#p;aaCp2j3i4oDM7*X!otKH(h`8&
zH8m5W6N6^;pRI^qJ9gmYVWi3m>$CBC?8u5kAE&9J5)0019(-adP$hwVCG7MWM_z0R
zz;=^}G?X;gaA$J(^=rxA(1LFk;=;w#<Y2N=QdnrdQLdn1X=b+S>nkWv1MQ7OWQdfw
zxomTKZEbdbe(ljU<k^GUMn)o_ZhZ1Y<eDD}azxT~9f`#Dnl457GZdsl?&K+%LpG?V
zkdS+ED##L{%T#fav6aX(%>mzw<-ZWS<KS5FysXUi2o6UPBsH_6yyBVc3rWb}yyCMH
zgwCt$V|}&wVE}1hyR5EPy+%=deo*@3lFUpNE(+C(`$;l#7w;s?)M<gq%$t$|<!JKh
zmlscm_+_gz3kzGllse(pAa%aNCAeKtPF|r9SXZ`FaMcE6SJUJ^M8<~2Uqhg;nXIh^
zUZR6fJ$Lm{W&hBKak&~?(UYu;Tp!qDe7UO&>(fOXWJN_aK->yYN-M!WfQeF4R}o09
zU;F)gKt_<V5Li(8`M6dU%kld7`!>Q@sIsr8w(b5~N8WV|;R6mE`c+PggK;Hi5Q*T5
zz5}S|@w$I{?iYSlJbOihNb(-(P?H{AUwuZ5s!?bJj>nSy2*=KKRCH#?$DezxJw*Bk
zhY#()*8;sPKi|Bd7?(!H%7p~?b(eah%j?h>BL&I)#$p)<%V_{K(#P9B544Dd;}hTz
z6%=P{XsT8lwrXHcl$KU$mFwGrOG=WAqPz-tn3=rL`Wnaj>MY*~?)M$Xy_u8n>3aK%
zg{uICiS3`cZVs|GX*#0-X+ggw9tTo!95}T#6+K(mLM5p3e#*)#D{HGO-$B4lQ(0L$
zg4D{YXRjc3$h2aF2KT-&nZS*lQx@IbyUx|;-p=dwPgbyDC33M9NlW*cT9Sv{-W|1?
ze8h{#6!m4i(X}W)2+-?VWxQZ1)l-?20}-1uZVPjDR}lk)g_XOru17}zSpXc3;uG^7
z&ykUxm2Gfqrhp4s=W^htB)^Tk2)%=M!4%MGXr5cJXHW7Ji$#PT|MoEeFyKG@G$#{e
znS3t0yZwwL+qN`t0^eHd4G9dIw_5}Z8IMkLgxh785qj>cbyY@-zt~t<Si<K9R(FXK
z#;$-F?owA}g6>Jv{f5ixf1oAc+qWl_G&IVpswGnu8E{G>ZqFx6v>L%#SX)p~FpR9u
z{0!Hj-FoKJ@unZe$}z!%H4@kr4-f|AeRT9Vgss4GkOBLk()kT{OwV2&!5lRfZzoK#
z6G+9=w(0@)Gi>h}?<a0-eF553P)Hcuu6yY3bmNL3Fbf3%9ob$%97RqhAU1a#N9efr
zg@s!Ktuc4~&56IZJX>!H_&(onTG3mK_HUy-deq1*DugA-fdNRc(Mj(aRDi<~4H=@h
zIp*z&8@?~TTB6fHK{VM5lexWi^Y=6nQdS-zsJz!QZKF&sdqymr#g`^$Y^0`ruQQNO
zhjBHEPs0Ql2|Rg{VCdzFK{D3g56NIAJs-LkyKwO0pl837(4Lpjq#G<yVRxM;X+D+3
z_k>vy@&PKb&T>4P8CL`W*B?bi#iO72J?m&vC_mtmQan^OB+Zwr?s!#iWn6bb_b-F4
z5*~=&6CI%uH85k^ow(JS@^iK&2Py~zr~~&aP!K4vZ^fQ^{oVO^)KqNB$bS_9rh$CB
z_0(r$V2u@AjH4q3s<d9G>x-b*Kxv+`=+DP#3d3iY?skQZ%3sj#=Wfm3-X2sVS|=xi
zFjYzY1AAjz$Gp&2vh-^h59rW8Bh3s_D#X*}XHbvNdDFkfCVs$Ud)&u48NVz`WZmf9
zEp7vmRRi~pf!PaSn=H`;A@PH=Hoi}l)SVsT7mb$llc7_(jr-SqMw;l7`Apj2+tv#G
zrQvsJcjYbEEwB32@JZur*jQR7K0h--KHDC>_`C4n_VM~h#Wv|=|F_j44_1o`Gu3o-
z9zA+g>b19%<c=}py^D=NGGM98mxm5}nnO9Nn6dEWsNX{PMK#=xqqkWILP>8xat}lJ
zYipeh*sSuD)-Fd-{e+IJBd+B+3>sCOw2#pdC+ib~HfS{uNYH#F7Q}SxJ<fJ26OjI%
zSrutz5$p^O*JD91uSZBCw`b$gArVZ13|NL^xa!dVD=aYQzZ<ZSzbBVSZ`X(<*KN3C
znw+w?7CL20M*pa@E&tXexcP$sAJ7?g%zmU<xoL`tX%KoROSS^&^r8c@(#wK64PhFw
z2`(})JqQR(7Cfn&nEJJ}2xg*Cc3tInF|06P=fcjQqNd+=AOW>00q)%=PapRST&A(K
zjht`!wx(6F{ErLJt6lHAzp`phvxA+k|5Gjo+;c!FOG-vJ<H0%De#LLDdfZg8JWRTx
z4Cn*NzV*QIoL}Hw8J3_jg9dKRJfH6ABn#dHqB_e(H?pgqW@c1S5=3>rkJZlzeBmkN
z7g8|{C3gIE(?)(wu>h_ElJrO9FqQKYdx7O$UlfzYFP9RN<fn0$=suspn#(rHs=nUa
zpKi#ji-GcO1JOXfWYRZTSw=%oL-*=D`RIAK2XXi+=4uPj5N?m1hYi!-D=rpK0-x%F
zN%+d|)sTO{j;_ugF`>}Qp^7La`6axtvKss=x*~%j-d7mIX%x6Afa~Tu-i=_qs&yO`
zMDRRSSI0|0%|$7A>IY0OO&9hX#E5Eeq3-O!zYS=Eg2~vGf4ldSM;@*r*N`<r0^wC=
zR1cJ>POJu)Dr`_BEK&+Q5QTzVGkW3YwEIB4`n%4&2Orge1)hEeNJcpx?Z@ZV!>_fc
z;L&2x-mZIf#3>P3R>s7Q=C6y_eV6EC+!*%mM5+A+OfdNR;<Jm3moDf{FVgPo3*zWn
z0#%EJV8gK-|9xS|Z3G-EC3f*iaPPJnmw#)A54JqaiZdnz{^dpTv+Q{xg7{B?loArQ
zdH$fdV=vNFS5yS(>qXYNLEh6y@^br~E*fesHCYY}IHar2+5yFS_OW|ujD?N|Od7lr
z7nZidJH*FEMm$$ytH>mw4(#>Kf~|G7{@fWi*ztF-9{`+ClRSSY0}kgO?t9X*aN8}`
z_JEsP5?EvE=?&PX{FO~^K!wtzD`kl;nZ;|jF-h}=?`MyV(VAOvH`C-)L`7x7hUy(_
zT&1JRowoRxAhASt=&6~78S1sBIqyS5QDc?sTqG@@yzlywDn-Bg`tsk>?{=ymBf+&K
zK)G+GVIgX?v6LSD1lSp7r<7EkKu+}NJpwEm4Iv9~b#=dUuOa6BhkElFW9pd6%HSiC
zg1BWoTprK!{E7;0c0Lm6BeS%yK0e3&pLykYP-pN-r&JtQx0Ww69IZ7toou#H^yw0f
zSJF{);`U@w(=c53nH#mf7xnUxP#_1&0aaraCoVoF*VC2V?(R7i8&rfxd*e~^W_q=F
zq$T;JOvQ9IRA$)Rf>5_PIy&k{K8TA^{&RFHt1~n@Ji7gaJYG;<;X(_ifED1(Dd)|p
z#{0K-9aZTyzmd>9ciORZbwx{P6^)7kuCt+}q-0=mhH0g3Hfq$Q<?6Oi^Lw9WtfzkS
z!$Y<sL9LJ?1sM0Gqhg~|3mYi7^^AAg<8ee<TOkseijIbcp8m}6H^^xxZmIU_d`>6r
zO9BtaW99GhIV4}7t@DMqMxIs|z6t!d+48g?(&&$q%-Y`?yN*y|2Pw+7r@oilZm+IV
z2#0B&wDcY1RgK+M?US_D5Tv=c%^Y@5gcKXSg)9SnDko?qkQ?Z(B_#zzt<%^<Y{uW@
zpp(9=bH_?F+7_q~K$_lN)V-u)`>naF0n1KMe+RJmbG33oQBeW}NGUrzx87oI3tt~s
z)9f{ieY9xL%Epp9Yufi^aq+UMY6)OVngmX5=hu{0Z(&fNX4m2(rYr;+GMdB7`YMX%
zmZmVCoSZ<ta25b45Tr98SW?*KaV}Jq6=Q4<5EYgbe>SjzidbJg4*j!#&pT-R^P5?c
zu7|%yn|i20KTCV!NG#d?F)<SNew>eWcwnySBdUkLmV7Z1PKKgAy{!bMpoD|4AXE1b
zao6`+CTEE^uX4|5qog*b&5um^#7}C(lHWpVH8i@D86P+rD810qaC@52dd(;GVBXxE
zdcR6#WuYa&t+uWfLHq^P8-A+7cmbiLqH|v$>9C$7;_9zVN!c=kW5-Dc0~%AcHJ9}J
z{|x~I0@?<YRL7o=t8*Q!W))_3ynI$XS5I?(IGTJeo5<T_`o(gYw?U1z6ILjD@yXdi
z)-Cp}@uUxlOzun3Iy|*TMvG8xfITL_*60XRxgD&%GOLC7jukax+1I3jp_&(05JAcp
zFC@1)Y8kj6OA7j$nWi>vlz9FuaCkYU!cmir#=3u6Q&U>{=-G3JtuF?EhHlP$zl#hj
zDd)&OfaD=nZ6fcsX^|d2_oGQ^cg+Ij@7~&tYnZ^j)tkz8Y?@Nj9330`y=Hn*bJC3G
zYzcg87&7c{Oh-l~Fwl^lEpPh$T+k(~KCd+R%V|L7*VRQ$CUgezQh^K?zFU<|Rp(z8
z?J~#LAxmCU<3o{ljl;%UE2wYuaqAl$x8}Qyg4Z_+Ki)P5UW56MiTl@Kh>O0R$XOOg
z?jN4gu*>1T`qPL{(SU5Oyd7{05UtYd)*W3^h(2&w>FeeY=NjaIfyGz%XS$>xc-{3N
z6N4n-LaWSlwi)G2E%F|KdZANJ?`&E>-imX!n)?*jPWJkRq{LOkfhg{bv&jOF8F|0l
zb$0wcae|p*c@dBP$jVB_I{9IM{q}Ce=Qrd=BE-S1$5uViuKz`YT1L0Vavq6mjEuZz
ziF<NTNjN|S3$)&%T@+fyr+&l_2{xy5@Ccsc;-!3s+w{8lOj7E5w{$3s{{lm05uabt
zKzMzc!r?_qN2VwyRbH0A2GUyO(oJc(`0TeTU3k#D1v?HUYOA5DBm|fk9<bW3a=}u4
z;7|)rbiEA$llB3h$kio-xa+ZQzdf5`TyeO=8uNeE@P0<>b1p7E_x(bi0GVVC2977v
zyMt|$b|-C>XFPA=HcFMM!=g0J2b!q-FRJM1*t9rHwqO{V;e_QWQOgQ6W@bhoQXT}r
zOi?N_CNWZDuD^MK3VKWVuNYT7wb_CJeH?i#8xT@a5mxmyg#ip}EqIUpzQ+v7DO#BV
zwXD1N?0rclD|fQs-<z}uDb}#6l1f)sm$>99S=KFyNI1-MJy3LgB#oIq+C5a44>o0q
zk4WQ1qcmH6U`qEV4BUv)hE`$t1=!(FF5JB$Ko*ijAw|Pn_A&kwk>tq73^ZKT>17HU
zl9B=fyj)xXSm^4Jk+itClD4B#BIA}vhlio<&p@S@45AyN93I&gQFj4d3$`mdR59ta
z15lpPN_5IP(IVpF;`jtBuv?L5WxmtR)pWr85RZ?Gjca0POhvWR@Be<GBYm6eF{(hI
zO2?p_YSePq$jG+h;9$aXs@^}CL?~)Ra>NLCU1;J%F-2`U3ihV-bU^_Dor+W3P)Sgo
zne?yvw?09;&~lKFRfTlO{IYy%o&^%;!f0$wSussH-IVCW$<wfyg${G`Ty_>VFmD<g
zt+H)@3|+W}hNJ`YOZ-t8c!KN37y>zb7_bs+9DbMIQ*uNN6(!u@fkU~roa@^E6*~U-
z(kARE|BNga#exT(;qRI)1jwHsDWd$h8F?36wXnG*;nf~%a*4Nnx-jVqs<$CZ%_}N!
z-~Dw|REH@g^_^3P;J>~~1{>X<jAp7pth)2QXfRWA)hblNdUxfMTiP6*3N8V&b=`N8
zS?2lPQ-&-@M`()1bRaq2T(9~Ubv)w(6JIcs_B?9y{{41jLu)5X==VRM8~(nLDNBAn
zQ^~?&#ar^u9<Kc#$en-wsM@lm^!Mk^tL9vooBg#iy1}m-=@AxLr@wP#?GYf?y?GnM
z%F^=Gnu4`|EDL!;^`vq^go=jd!OAMaFCt<DL^z7@h}g(AbF*_cQgJQ54%?7Bcc3Kp
zu9jK30gWX%cFG^G;f~YBRw~DR3e07@e+EbOxQ6o`^qW?urmlPNm_QqkfTWP<0ntgN
zxwWyrmbPLNc!_uNj5ETd7gEfF6y%)gX{3V7c7kv}erGHsyCImH+I;4;7$7Fvgc)qw
zD5jwBQ|i{g_oamwGv&hB?n6L$O5d4Teu%>qjfh7ReyIP)cge{z!ATPnHEHt^WcAmh
z$SqGSwww#P4@^=^su#nRiDwrYm#;1Z&%uGf^KaM;5rd?Vv?G5LFA!%Jad4Vw)sI#U
zhLx*rG^WZ^0zTMXLJlQ5Z2cCy&cEM;9lGxCQuOJ7G~##E){YCh=D8ZH<EU@+u*3FZ
zhkeZyd(qw>Voe?SiJx@JIEyG}$1vAKNA;b!73)<LFPp$k&qhE0>*FmVhWz_hQ=gY|
zQt&byd$?v!YnZS2356#=ADX~1)3-EiHhcz62mr(5aZzy#{ke0mSS$qeB21^gPCH!m
z(ru*N2>rK6al{h9x}V)6z?XOn)i-zLGjZ6ymDk!ytLgBFQW0Gp;&V9&=)4ggcnw4h
zxCF#Ek&7b{Izhn}cg#P>=gV1gO)`Crn8APwG3u?0!!Z@6a`RjoFJbcdz7kCBO|%ue
za)N>agWd%OJ>cm64ns(jYMlEffI}r|`+MX67_2U=xZQ4gKEg5*(`0sZ=gyt<4c>;@
zh5sJB@dM$Efrmuv@zU0$Q-AZe@G@Kve;q+ZiSlwaJ!SEEBFNrZysOjFY<;9ioK@r2
z*5b?YiUYHk@mGF_{UL^c_{69L@1sdzma#o0fTbtj^46iUf>dH>;yXJg2|FDPuUB^<
zG_X*aL3yJR&t|(`m2?7BohXVB3Go+p&W;n_hr(MrGk}#H5OGF)jN!5A4N3jcKS1_&
zy7uI~XohQT9o$W3(r2g$Y-fNtWaOtS>FC(*wI!`!_KAa68Z_)M^hsg14@xd}a+28$
z(+2k?RWXb18#WHGD+eTmJ2SB|67YUgl$U=*2C2yTx-!;7i3Dh_o4;-vKtTW#TWEEH
zgJrSj%+dXSZmtd1sX7yc5)cqOE*{ejgyg?{D;n1*1ND=+xj7KkBR@p@?*Cpj4}Zue
zEUnDJ*Yszyhs%^tR=^8-fA38WpX^_^9X7<W<hFP`9&cVtA8HTT-JERYreHtNma8*<
zb{7>GhwP4nYHSI6f42XggeRw;F%1s3H1HJB(ybky0>`YuV`Hw~(KFEvKGVS5aeF-a
z`y{cg8?!yVEi=j*K}qjFcq|?l!MaA&_J5xO1PREX-T5G^BWp}!gZpt}!<+28+b-xC
zZ(Sb{K2TCpI-aS;!@WBaHi_OIN^EUop{=IoFuklLZ+o+rpHDjB4aRej4eH64uDAdg
zKQF`s_zv2f1xS>^{Wtxc_CxiLAXuTK+V-6bNWtSG8u&UroK-TEd{s=u{gQam9?`Jo
ztl6CdMA^2GKgOlxTf8nWBE3r<b{m(!#oVhXUDMFjy;$k|o5P@!aRwdCF0eoy$LQhi
ziHNj{$tKa!(+k+Bt83gQS?km<g}jj7#DgLLt!<}%E#e*OG(^NEQ#BsIEa_<|VOB{F
zSnfGr%nyy7Z}p!`03%MgeQnj0YtMm}hJnxZxPE1c<i}u6lYPose7rqgNrlOo0kr#D
zkNhkPYXJu2b~#<22p9!b^%p-aaQFq)*Ou?>+-J`wnDZfUz1=<E<6<Ig34ds5Yug{!
zmzd&5zi!<8^eB1FoH{FOvE8#bD4zugL%?rEZrr%-wLM>jiahmWG|sGsA*39o$Ij_U
zlnH}~^Ey;xi6^^2WD<Ck1`nyld6CU)(Tua&m;053&VV-e6XO_V0kd&^*85E#UWcur
z7{*^FcIG_D1+%}ndwd@i0pNJDM79Ks8R?*-`Ruy<1(Rv=h6Ud>zvArl;fc!Te<A6{
zM{CQN@#pe(*A8A)uru}Zd98VPdbR)$5T=^ILC6QNl81)N`QV(4gQF&3fBnZc_P4md
z-}}l7Yc&!%-M3G(3Ur9AnX_M*Qa2?eB^R?z#>PbKKl~+$^8O`Lz@6=9%#1QpebY62
z;jjs`iRw?U=pjD_$IQmD!hr@yug%ieHTd``z1ODf5VV3XA5PX~l$OpnM?^v9?}@Xu
z{Ck_Or~DZMZ6Uj}pZD3pUwPJ#B4R+gG$AM`zxjQCW;9EmK=JM4IX;6+WOclOj*V9j
zE*2bzuRGbq0l!Uk`$zHy03u>xDPBA4i?Mtv@Vxv#uHG^#%dTA)z7PpfKuKw&q>=7W
zLApewLqIyD8$?7(KtMpcQ$o5V6lsv|QV{9x-luE5Yp?x%V>tZ5^9VOr%z4fugX|6<
zh+WzZx6!S2>usOAnsJLW%T)^3Q15Kp@bK|*{+_e*`9|VshRJn+I0BC@<0IoM9k+g0
z?7#P~ug<(L3OEq_+-J<#-lLE`UKSdEtOoV6rXO^!_Tq{`0x9k|`QuR6iXmFiyXzRf
zG~0M&zEZO1aj=YzfZK2FojRzaF>q4%>)l;DIUF(S4j;Ee>Wu|h&G@&b#T_pl009Qe
zN%{r5ck(|;Hq5=AIOGdV9b2=dH)i<!r*(DpPOmrN8eB{n@v6r_2N9GVARt?tJQ^cn
z*RX}NOP@i*OsTtGBE?O_SQ>{A^Lo2d%1FNozOyw&D%D*)Qaqwf!4CB&f47(dLPP@K
z=y8XV#QX2pt&SJ^#?l^Z=2juIdS%;7fyyao`VE5Ky8;}2env_OkX+~XAraofW%?Mo
zWLQ2KRd!`{i>Eg!$hgbP;FxA-Wn*D$N#kKPR2Kt&Z&;STf-VW@Q;-p5UU+*r8#GyJ
z+1SjKMw?Yu@@lXdygf`zxND<(HU+wT9IB)IqI^+j6OPmFjN%**x2;qdmH>~GRqhk1
zaK#AG4JvhRv1C&W1JMX{9aO(;VZQ08R?$?SdC~RqCOzG!cBSN)B?tb(6YPzxiJ2EG
z8SY1vVj!Ne8{O};xs=L$nm@%?R9f-iNYHZ!;4lFHlSCRVNM}H+8HhvabUK|4r%Syv
zVGg~0?nyi2m$l}X)@*EHkWz=*Toq;o!A}1Kl2nuRwD<Y9I(E3VCImr1$~iptm?}6t
z?B=Z-Py@@y$!G0zNPp9pkCzu*nBQBw$KVacOZI@|KzK79K!h}ap9|6Lunc=~DpF&<
z=`{-dUW5Ev6jYH_jqD@tU)gV}x_~gS@~iD&NP{iWs~ye{q!nTip{c8?VrOBw|A3C?
z0bNB_MMbYL$=Sf|u(wfBkJ`Jx{r78qDen`M^zFZjv43^QA|1VbO`aETXq4}PgrQ}I
zmxmQEOh5)?nO~~y9alVOj}rb3L<ivP2}4g|y|AIUcj)T&i1DK#7s!@k7GA-PGpM)E
zZES2bFu0FE_+0B+bwp~TZt0&mVB1le0n|efg038!dKho6Uwh~vHwlB79%~PDM&3l|
zv$MP8LjEb<u<rm|<9Rq=S(nh2aZ;YJMJ43cU%pgQGWYc+0{6{v-E?d$za2(v(B<<C
zkEArXCC8@3e~co%LkR2uOXCrIN<eUFXfja@7pnAbwavv<e#vu#AbsRRO!Qv!jE1Z#
zCpXK~&<rp8y?<CL($Uf5usSj`GqMPMl~i#GRlKC#HWEz``~Oz@3F>Pa<rl3{29sY$
zS!3Z96N7Dw)Bc8%Xx+i)#6_5IT4Z`-e17Mz-wU6uERkR1`OwFT?Y<4s-{0;E_N*OM
zCJvZ?1Ody&)^y^xG=3hQaFP}8lhxiayE&9*15ezU8m~4`&Nf5k$U_mILoPw+{#ajW
z9J7UO$92mT0|9aTr(37M)8So@j7qRIGI`*YcfX+a?$(<*xZoIw(s9yU?FMc|c~#Xv
zK-@{ldq$jl&kc7+Mo#wOiIvNysMo?S$s20PElOc`J{o3pgi_4?#S2phKru&^wDr^+
zA$0P^W08ItDq%8~8w`6VpZSeeCmIs<oSnn^YUSx<H-()tTuUg;o5HJdk`*=jUM#RM
zv!MLi=1c|2_{>*FDg^V<T>`xKRmv%iUtVm4<V<x+H05Re5T_2hjRkRM#dCF`ZRNK}
z$HlpX4)@&+pdP%hqPm<?AdSjnA_1(iIcAX28Pk5B!LtVUOXX%rTk@UjP%0j}*_?U5
zk6gSa^hxHO>%T!Vu`@($jyY1QW|r(B6w)LFnyH+&cP(uS4}*QWx>@?)xL6ZeEgK#M
z$ZMqMOMn_{s@k!;*gFv1;+Stbes+qw?6M-{0v@ZUt!i)x1WCPJ3GAMV3yDSx_#yK0
z7W+AdFI8mjB3>e(uvXa`O%B|-AqyF%6U=XJUjwL$gm>DiJ6!ZPRb#{$N6KXR731oa
z@-7T0ueQRy10ysbOD6AGgdb|v)-(nJGH93pee`RR`iSXrvy#P@gdyIFG?d`>T@ja$
zt`2eq=F_(iOIl)$WFN#Q>)rfse0RNlF#T>A1a7L7JTyPkGnA1rKPD!QiHr~9I|gZX
z8l!)tUt>YY;5%N$h;X1EpebU$yY`poMKyiXzUw9h9_h~U#La<yzlca=`J}d@B8iqo
zM@K^*g!oa<um};%H;im-()U6V614de!^-kN*SB|auHH@D>9}}T0Ufyik2{eBQ3$mU
zpPVCnb3BEdJZD`&!N;iHNM2D<J#2<Q(r40ttKLih4GGjXmq}N${I)Z5=j~m4xj1M|
zxSp3?;u%`g_x2vQV;#R~5*}57v^Xs%g{KY$m(nXqiKO&=GA$;MfAvu+nKh?6@$=_}
z(7%NcQYhq7j*2|?7RBpW3{8AO4hSn890{<bbmY~ZKE)G7N0oqc5Tne$eACr+sR}(E
zow&rrl9F8fyXXHHl9!8FDwqiG2ru0SbAfm1jes)Yg$01!29B?m<y9F4piSi~7wm9c
z<l;^=C42ttc}`Al=F%aN)Q#(2;jGeLsv5;}LcXgXfW^<M=;?O+)hm1?u8&_JNWs>-
z|6akV0rSmm?{qj6g7PIQz;Qr6ys!{3o#=0QcIMQWk1FKu7*OBF-t{;+-RS?eIp(d5
zf!>6RdFzg%5U9q+hE<4XN+ANlxk&l<LpeG5rAr>L(FAvztXf(Hypzhf=(P&^i!!j{
z^oH80qEe*?bJA5f{w~QMVuBDZPU$D@8L~EzPz0Kdd-*F2AaKndFt3}`hL!00y`JMU
z`EM8qdNR(xS{xv^r07F~(}I(loNOja_R#ANZ{EQ6-Cd_oILYo%`v6Nd_Kwrq)_WXU
z*};AZ<(|~`;~#%O1CLvM2}8*8rukfd|LeazBqYl~&Bt^nwDB|T5;!?kOcoLz+NBro
z+C&c+{x_9(^#q(<AF^&|yS>KU2@fYY=uGa+Ux8tic!Q}V@3z6uQ0TD>d?MgdURhl!
zR$<l-kfFHJxnIgS69JY-8e!BUJPAg+^ocLvPLU;WvItS*X?S64>Gi7bQ12}Kdcao>
zmJ$Ki-Tt{sG(6<Tx%O<C^P?BKkLFw2`v$sUR|#ccT^Dzn-ug=ceqU*fseJr)cAZ+v
zuqJawYp?Ii!Q2$qMK(5dwwmSVdcQmew3WUqqiMbU9xE(P5}K}(XY<G5a%t-0;Y%Ne
zt9WKvC&E>jKMmISDo3%LR#f~uOwI74zRyPu{>?Q{Q7IpptP=c~=lqPh(p{hN-h_(_
zVI+75Tpw=D%$&~EUgb&i*NdJ-09!Y>s~w~yhqbzvSbYks-M)>~#lF4jB4Nys>D6Fm
zXl+B$$a|)R`{3KQx{NwK^@yTV8@k%@Ch)BWd8v2!Mw8pgDV&f1(M=q?5?#)#RO7dk
z6t>G%wwxIGm{SblgB_!T(_Wi5QbjyLF*rH%?dR<1>GrI+atf>i2k3P#QX1W2QVoJ|
z>g~^b9OM#?GrUTUd%^cVI6?+|fL_xC&#Pd%&VsIjzZA2P^>@zFAN+$n$us{m8ECJn
zyl1%2f0bA)nfT$8k*(#q80)o$XV0G!Eb?5?VFhI5r^iAM=kZE6h-Ur`vR0=iz(CPq
z4&*t24-iRH%{QXFrsLV&HpwMO-@7%chp2p5V%{Z?#C4&f{5&~Rmqj2b_;apE?Kv|u
z>_>3k=vY|%^B1<WrO7biZeot*xsr(EyJ2NzPvHz_b>&)eF*j3p_G>-Zf68SOAD<U;
zccEePb^{OBu(bFtKOfYrn{YvNJOUxk;LiyXwDb1%mbJ*Q%0VYx8vM(6|1CXJeqmM+
zQNYOPsAQ%*Ui8zf>@4v1LDDVI-%m(osAaimx#g*Ee~Kp!#^Z~CJ$idN|Mc|qZI%hF
zd_0E_0UT9SL<>ta-<X8{UN>xq(2(HAep2r2)m6n+qJ*|hZ{E-V@sPkC5|vS0MdW07
z=X3X%ExW;2h$y{k{y<aeDuw`x8f=OV?>O*e9eJ9FjDj2ph(JRNnx?!=#%EXwsN>qw
z_9icxo`HGO30Q2{f2)g=xOw?Nd?G6&2LZeUGo^r3JC$l1+nPSY3NU|uCs_&N=80z7
zijswd`IXfkQ`nG^ELToQ-3abvX7180*45GyM8>DI;e}2;YKTpV1Z}gSNm>L0Iedj5
z+NQJ3B-Xk^=o#n}<L`#*E0Oba@v*B8k4J!9L=pR-r_lN%_6*EJ!N$_3`VFceA_pJe
z?cvP6r;G$I>09|8SC`B+CW@OqnVS{0DP-ffnWMK%W|ijS6wEFtAk~X|*XAop$Ta6L
zVY;4(7xjqnrT+ht$-mG0uYNQ+UPS^D4_bcb3f>t#JtfKs<B3*XU1P?NAgH<Gv55jZ
zbGly&r~}7qaw?VZ@S+<6-84VNJC#>ca2qiqDi-romn#^x5RXn_)4MsdUR3X*RqLf{
ztKS$L>GzR<OgrKySajFdhBFfz*pU6zfua&MTmmYX_^a#vHv+AH;QWsk0Mm{GzD2tK
zjCc5^6n;1inBf%DD|~6G@9Gkg!@vrdpGR1vRkyvA9r|}=Z3W^uJlt#O=@>$(MxB-}
zX`$NyMkyZN;kF3M;@r;{zXrGwMle0Hu_cNcmjy?D1787@ELiZ6B7hlsCwuHWucnuL
zrtOP)JhiLKj(pw0B^5V)3!fSyCcAd0Mz=BKydSrm3v%x}RVI4~96GP=%#V;EGD@;H
zBi`#*8bQF#Lk;=RrmAA;V4pQI(&Ez@(2L~wCszj4uAGi^k33UX`LGCjq|89vmDye}
zR$6s_uZ8mnBlAE444pCRz)mISmi#M5^?cRG2jY)yDe`seQ=x<9g<(`=-KOYF=-^Am
z15~7G>Ik4kqMl?ybHl6FCO|Sc2<Au}9LR^`VWTEj8sbxaNZCL}O4+%zm#akTSww-q
z4A3SVv)kU^EdcA*UZDk|SpCZMxsrt8cEMsOM3dKlO0=E(O7yach+>-)P?RbgT_8=L
z$*au_%@8swmREqU0P2vq0TH3FQLIL}fM<iH%ko?nRG3caEg-cX{$V~;HwQ+ITl-Fe
zWo7*yX@81*KTIKJ7{X*wKLIIOJa})9ao&f?Lp9W;_X>w8LmQyZg}Rq4lTaXRJLf6j
zyLk&dVmG;?4SX(FMrSb0pQ>A-fwOq7t;+$FIzYYwjPjmB@K)I5(7>qVn3SsOL!HkS
ziAuS!LF!HMD<q02!6zQ?fYcZLWly;%@X-4>`X;Oznq`*cmV51-i&q!gai4JBBhcTe
zzkoiUK7+4!=mWz;g5~A=j=d)yK^cV3TCwNlH0T6g4M>N`%bgvpj5H)Bt0_yvzQ(t8
zdp+<`?m$-d`2Ws{fsg-xE(-T{<<7_$@B@b0RHg6)((0(Fyv;#XVI-iZZw0!1a<Z<V
zV6}Sd4LOB199{PS!M>U%V{sw01QAXX_LtJC{k!9HZ3Xhr6wV{@b5$#7XqaPjsl}VK
zBfu;+>)#IM9B|b|K8k8rb6cR2$s7QZ6`mkgKFHg>$1L<*R`nj2Q1REVkEEnJuk;=~
z#DowJ_y;e6f*BrlzqMXcoKt6Y6}dcKW9z~f(Ao;ZNhhIXT9&=GE4|Qo0^=thYs=;^
zYD7JKs;(YW$-@g8lc*>)&}R6*dF*+#I@@GG1uTD<j%$V&5isH4%GiWfiLJ5r4vDLI
z!X~g#1#b<4D!_(Kkpq_>uG{yFc^l!QbVitJtG?thG6_Ol%Ogn8yoGrSV)Xz7^dKgf
zpRSLCrzK4VW_FO_pdw*mk|xg-_&s$X<1OvOgDcl>@B~8^(CTWzrtJlMtbV6^fR_Qf
z0MsKe7TX=>fcL|S5ku%Mt(^2y?g9ML!SRvbOiM{g$-r{Q^KxDjLS6tdm^;SEs0Kjd
zeqRjP{K{<W^q-#>?Mv+X56|WTrJ!Fr!mqe$RL}qMcQ~}Y?NSDcuBU5f|JBCKXHN&q
zQ?QGH<m6xaKhv?dCPkCd!-tm#QxfR@_r95+2fXFw!z-D+4l*Xsx_WTsLHyVqHgpbj
z<%eQSg1r13RLCcVcOO3z$f=6BeE=a<`nx)9P=ShQz@zK`%}}!a$Ii){C5}0BD@!m!
zlZ(2y)w=dU1MBfyI@08}n1!zgxR0zdy1P0)#>TYXXZXy>;$Bm4X<-SSFJSN(E(Q5T
z#tswcjTO}we+@+U$HFXi{Pu@mF*|tv8a&8u<CCSv83Y-z*j@O;lN5;KvIU&dcBf-%
zd0f9V0U^OWeiZ2aQ6%1U8kTC>p<Y9wW@ZS7@U8>W;m@9c>Nx;^Yi_>--?kd%A+)*L
zq9(<-VS#KUB0LOVExW92cYpsDXh-2h2-OMRCzEq+ldwbOWaV^rJA#>@%+Uem^~2n5
zpW}I8C$R#uj&9@L<Q@Xd2IT{AaD+u1=2B6AK98YbVPWa}Pv63Jf}mQDwav%-&$rKB
z35D@l2L-B+v4g5d#PO2r)Tpb|-eC_*rHqU@tV`{~kuUKNVoYFVQQtiB5E7~p5QamY
z6`yr?PZ!E-b|O=7hQgx*;|{FR{k&LMI7KBTz6T7=-+p?0(D7X>CV~?cfWH}Pl2Yxl
z8qaVj|4g<Fh|$tofj4?1ULCj~aVTvDM<W{((**!kv-t&#p6)gA@pmEX1V8((ou#47
zr||IDO#h;sI&BwQeUB%miMh>)1js|&^VW>RuHU(nyj_$VlmD3!#sjalHU@_5pvz{V
zaV^)<2Wy({{}Jacf#W$WJWP73nB7|{8vV!r<-W6(Qv^$T3$gAVUO^|1M;)%);jW^)
zdqBm<k6X3IYy1~5ik`mA1`+H<ebubzwkCmV^JU8uH{8v!F^f%(ZeX;o?Ek;at32cR
z`*qb+INiIHZ&+H`?iZ1Hp3Sxt`EE*0fdGt8$@&%6f4JDyLkgkr=jCXxVIVks<z&7`
zE8aCZb@z};#lT>$W#C0n?ipw($k%Ng9QPi#&^8q^!U_OB8&IyoE0#RNx<19l&+9_g
z-)C%irL`C5<f8Xw%TF%7AfM6Y|5_NR?t~8*j*nIRsjdWiB#)Uz@1TC;iv6yl?di2Q
zr|$xl4MaS4&l4|GA}%mtZGz_{_ow!2Q;io_?vOy5(!H{-{hx!8I1LQbUrN>&t**j}
zcem~9O{tBpP(XO9D4Y4^3U_zc!6qm2e;gX(zkM105xnEBkJ4Y~)XDwtbo9`B>~2_*
z^B6|87uEExsJmNV`_|#+2pAh0;Y)S~qDvc#JEAN(9cstZ5rsOnPABtd2BkQSW)V|U
z+TMN7uj<|0Yfe%`N4+8{J`&U=Q%>*3M8<9whQM8dZ{X})(d5wyvd`|Av$~f`P+D?P
z5N?`%uqV0!0cN4eASPh&A23>4bTrs2e*R&}XLK1AB%|&8ZFcfL@w^RsEEARZ;7%J7
zH<q9&IcVY=<2Xq}Tb~++h1zN9s?zyF!}{hSMgNLZ()<6A>Q=q3VFPd9NqgL)sJGgB
z!NlYeT1`h!KU$=H*e3pJ|7|?zhkXy{4_rSJM&y2RoO;)HF&s8^v?Y3+(5d?}E8CoB
z($(sqNl8<3I6X5rVGH!lJ9mX#{!U8)%O$bWQ`adYKV)uj_1R>mvU}d(ug?q@z`apX
zQ538ubLk5MoTY3o#g{gv>3;g5$%9XyItgVqELJTxZ`yzmaYNjv!|>?$Yzmu;Tp4^S
z`o`xmAor%4X*}&r+q-X(_t|FVOI@WaRFnWZHJ0K4VCW*JQz2x>So773wi}D%WJIbw
zJzqCK`)`Q#%N0IFue7nLQDH#=n6L23@t*L=+ZJRL4#P1}f*KRk0~$a2m7?fwMA3<L
z`Ax#)qo$_!rt^JB@X*M>wbzig$fsPCpKfYuN%L#Z|Iv5LwN=S8@B@J3=la9z0dLj+
zc{~850SSl`X)nNtp!e_JmiZV0&_ZW+34SQO#Y+nm5ZsS|PIj4=hZ*8JpNG>Xf;|qO
zOgV~fvgC7SZ)6M#xCd8>iezNvmzX*LXJ7jBSE~K|qOjG3tA1kQOZNBxr#%c@85bbZ
z0p}V7Yye1h!$g|>Exi5y4cN&~JgJCOpK6c(-BCc#O!Q}yjY9FeS0|(mhu+p!dWx?k
z+G$Vo;X{k0czHZ|?)B*D-UY1Sy6oVTeZVYK<*uZwh|LS<tT)ua0*;LG>8VjFKYXl6
zh=hn+65k`h`yp<7pCQ`9)QEsARnx#A;_!<WD~x4%p9J1eYkw|?$(Ay5z!IX3$ffR!
zkynsIK|nV3<~z*|hH@+7J;*--F{rA|OOq@Jlm)LaZKn1U7S<j`-eL8dH$xLSR<j^^
zT1!hE>UDZ&W4jl#a$csXrEzP2iigcDsN>UMSq>lg&JXQZe=c@p2$$#8NIESA3Go5K
z&cL32k{@>K092-+^tfuD)d-ZDsIW7b`ZmZU;~F$G<7Z^(YUJvgX@r4p^g3_cPllF0
z2=iJ7%?#TYl<vc$_U4P?A8&S!#kP*7_O=B~wi0U1isL>1CRzPtxyLD6mp(C}X<=cJ
zljRdx@$x0~1wTqp%d2^IYIt8~4FVRGY>FC^u_+@;3JQu#mR^|2(q%vxS!H%WKrKnI
z<h9p*21D21pqH8iOIjOQo7<WuB*tB_iQ?DaB!Nuw7+Zq>LX_Bq@8inz^EW{}-~$2z
z>q52z;Z^6dO4D*dJWy9Xo8T~dX)z0$m9DO>JY?@?$d@w0;8v)B6s>oDQ8rG67yY26
z`S&Y5Vb{UIL38~tMK6AQd3b{=x`-9R)D#{llJKglJ3EiO@3A(feGD`VxFtz|ZmOej
zdV)h%D3P1V%BjU#m>JDHe##&VZDQjSntR9RxHn{zCMVmT#stwt5glZaVZ}a@mA!!4
zpwX1-4-0m007^d+h<=PC_~*la{<{9Av~=eU+Q#m=@8%iw<~=#>s8X~_Vh980t*EC{
zvgpu?jRN_+9H^lyT*v;BGIW3cZap+fXgAl&9nZSXj&Gc~skcETzSkOFsNYx<+K?O$
z=IAn``#M&h!ii9D2KfpgXD)YGfhHwLzwX<d`P-8O@$XO7TCTsgH-17cToKrt&&$(4
zS7os=R^h}WZ!cKU4rhoTU%o*h*3O#=fdA}wZXo^D$`PI6IFQW;#3A(Qh1`<jDam=t
zv}tR$E<6#`I)7503rBJ3)~}CcmBB!E-0z3q9x*a95cxc?!F}Tq2F1oh)KezLe*Qvz
zeI1=MdNvAVUwCzP=argDnp)mRD13$-S>=2Kknm`Ao{g&8uwfGtF2awToJ_c*3T{gJ
z?Y-s6+M%$|ORH=DN@)^DOTaf4r|1wVn*}uuiSY?4Oa!^I3(%w9wm2$3#Mp$3hE{i$
zJv929=ux`)&s+CBg~^^hhvMmt>EKz>x*Gu*@k0!*rPbJ>L2XzrOQYBa2fAv;nK3^@
z8-2PZ`1)ICZR>`9d_z4|bRc&HmkPrTCT6Z2Va5GJ_xCdJs5)E@4+yanXQJjPkczvG
z4x4cpba!k#wu`7P-C|*GLIwJ$1;QqOp9$SYlqhHHUf2p5jZ>3`)j9pvrI&gFNjJ#y
zipzGRbAVj->Wa%<LF+HUrS~i;u3!5-T#5?SMSn)^+rue;KQJ{>F@J2B^D^}B<kUp?
zU7=U21k&Jn5DE)tEdm>2ZukZ<=35G<3JqZnW{#4MUJe8aEX{m}OOcgcoHO-s3aW9b
zZ+mrje8(Dv@e-=Q5;ljcT_gIp8G|L?wEV_dpIzm#xE9(<DO7U>Y~h>7w6{g;d@A>#
zPI9eAS2~cB!rA%w+Cv#lNh>K%yU_|1?=u-PK3G7PH+BX&Uibbk)T^zqic{KK>S1JN
zSM0O?_37@d0Jx)2e96c~0O?rk8`ee(q@N}PxpN`xbGD(5iHVEPPE+ss^OX1)R0J|c
ztBduro_0JVbzVqVq~jv7XCKND5PDBSLO>dxvhIE{xKuJ15+c99?-a)oUo!Pj8d?!Y
z+d$IDwXc|EeN&gm_j!bEGN2UL*g|lTpdA+ar>-^ke?#RN8G+<maPF8YrW)CrzQH7b
zc+|!3o$SfNaClfz!uLE_ey5dl%UZvBkb}CWINZw8`fVKn@};UuV9@~<u0P}7X236z
zyl$<J@fK-rM)4-6XfwCWv$C)lpPJ~;tN5dwI>44^2;qnNPoIWFx*}hkeosSkBNQ*P
zA0`w1OpHHYDd$18vNC>YZSw0rVKSV$x5ER;B4x6MEY_Y2he~E}Y(Q}J)@!-!pYgxx
z7OthsKv0|FAVc!W=LejFAYTC1HZdiVkODvOa&sip>b`JdQ*<%8L&v?q|NaYeOPnsp
zNA|IA1gV&0lkwC-v%yj33r0o-a&edStaw%;vTW&0RTK$z^_K@B^j7>EAbv%A&C3#3
z3`*SYpBG$|l~+NZp#}4(w_Es#^7qZ_HVig4FFCzK77Ee22X@bLoLSn#NnIu`K+-r<
z_oDY(X7O)TPDDjttJeLZ3k2%OYZ*E&p5+>&4WdSg4<D22?2bnFY&VhH<fIR2qTt$>
zIgjC%;hcJSki<{4E9>13ZGG`uRf8kNq249B(+H(`Ls$%fd<`!37^;d#P;(bjwj16M
zNRKD$E)_78bt5QHyV_E#r`|0tdI_O^z}U0UHPQU*OV^gV8bjEXzIvs_nHaJR*QQBK
z)VG$s%0_!_<)ZoQlJ*~9IZ^sj&+U(n>KbPvMSBE&q|9f2YvRz>WZ?;Yu`Oo2P#>BG
z7sJDSCNC7<p~Hr2e%r-;3uh@~plUe0e4Js6u>pWOP<~Lwt#SDfWFfEx0d<2saGW$5
z<0dv<5JM!6ijzx|4LG;vjb~erHfD^E4i#;3ScpQY#N1bS?9aP3P?kI1+Gl!Gik?+g
z-wu{MrEu1;8)v_bsV1IlGxznd?2HZh#O-M5(6uCrp^F$!f4T!Jz@ss^p+7n#d<Q}f
zScwv4=w|C&e7@HVC>tE3Kn~Nld9-rWH{X*?*`iV|1r<rRoR0<;%s5b*$EKg@lt0Xy
zFYYr2MY8A&4kO_$sD49z%}eHrygs^USEib#<Vmx0`!j{&?TNcQ)^0IMW0K8Zl9DQH
zRWMbVXjxD&l+3`cReNtk2c0saY0EuKP3QhX47gD0TsLkZ38bqq!U53;ZAZGcqVqyW
z<Ra)u${2_2+;^kW>P?}$B<H*Hgb{x?@4H>|S(>FqR9Z|F6X%zy2(XaEu@VpbJI<%)
zgoqm%zU#`mz4%oc+{xZO|1M?_ALOmo<{sHm2zcO?mfFn!(c-7DDFT{31DypM^&(vU
zdEwIV=(_-E10m1z>1-=&KC?*z>1&Z+JQi>$BbHV+9JdZDQ7$~B8@%t+-heD6NbFs%
zwg<a@;L+Asrvr3{&dzQY@K+@X+U&kR=~g)k+7_aI`-nmKJ;g{?^*5};sxzq-|F$RZ
zZZ7o#AI0mys@>pK+VUxK@RpLJwQjkH0PB2D01I;alwq|CZLIkN|DgMag{GT$k$Gbe
z&zh2N{_p-Sc-e659R5znz~Gr~neSNa7|OjHvP|QHuEd`89+VMEx%USBdW~GGs#|}*
zYN}$*&&mqAaT*(=POWjJKT{8HbjzW1+Kn5CQ(1Z2g8iL=(&eSZpk=6?11MwUftFFJ
z0F4mQ9f=b(DamVo3N)R+{JF}?`s!<ODaVx~Vd~1mM6j@&c+PCF;eY+3NaxWDRdYWy
zQ6#F_55*!<UJxkQ*yzD(!hdG~Cf%)Be%7x|SKE=@MS%uy06hNJkZZ#xcMT&gE6?28
zI*6P|`)ziB#d%;vM-%PI&=RPyNX}mSV9(UKuro7nOf;~ZYCU~Anq59~h8qkY>VBW@
z>RfJIR1{6XBNRjm9o3#EUQ3WAj%{#w7?4f5<4@<^vQ0RcF8U%R(MA8_(>JV(Mn=^A
zbzI3T?u<8bM{1G-rCzo1hmZhZ5tB~7=R_R(Cz)2z-|T8PHGVvL2nun?Cs{UAwYMPl
z6#g*)f_4$gEuW#!vX47@x>9((xhNw-m1X4mI-*2TF<x)#2?xI8ulevrFbOwUR&U*=
zg?7$mYgABdht@O$QpNl+_R$b47DtBYfTXM+M)uDXk{Tg+YRPQrFA1sBkKeCC|Mrmz
znKy&N5qy=74tqWmMbvZ^>Dw@4zQZNKL`=9QezvV91l||A<UCQsj<%TVJS@pqY4E&4
zIiLf}5F(K~G-0h&U3@6jnmtoLnP~;UY5n+A=U;R3w)b9Vo1(c@d?!O@kutDF<Nhgs
ztvv+Mmp;+tD-#X!PfYo$_PRGF6(`2<G2i|?tb&J0)2@XW)4Hh0Xkq?J>08$T!w<aD
z@TxXnVdDSY9vsd+ch<<mWazZZ99k-v!pltsHl`YQMRqr!JWh+1FIfo!#9PpDe-ytD
zdjD-<r>X7Q{>p@R%pT4D(aN4i|2tYVG@%?m8f+*bTi7e@q>E_BK$v5&6k;{$h<LcT
zjDl-=or;QBmFmc4PWDgMs`p)#lv-)-#9oiZfJVCPr#3c)6_pkJk)K?5%xM&Jhps=R
zi<B6zLRAe-8?r$9Gj)jHFYe&pBE$;dvb`5AW3A72SY8~?a_8zryN^(qsbN}R{6ky)
zhc8}mB4Bk+5wsV8sK~uIp3R6sGK}z)qzAUcLHGRm3biiTM28M~S`Rz$c%>=g*3N`p
z%^#PhiupQ|^H27P-&K>e#C-*!_^NL-S&6O!JEF@_g_7S<#F?Hh^CZOPpdXHsD}nKY
z(!~|{Cx$Zr-;>xSkTT9m8wiTN>aSmS{|NF=$g%K{Dr(<+*S9c~$Mca#`BWfPIddSv
zgb=xY&ey}rQS$mxI>enEZVvRW&9=bQ`?TXvVbZigUQo{BkmhRwu6POR;k_(JmS%xo
zDwc(~=oqy8;$MM6P<Y(7T3l1ZQ<A3A_88Oc9Og%d-QBSwZFI!hc^W1<+HiDHzB7I-
zY`^rhW~=;6D{RK^9~+W+pR`Cxwn3v$)zuA$^C9Xx!MD(Ad&>Wx`JiaNn#X^_`ln)_
zC%@<x)c?TTZ`L*}3wb|G93{6y=|QA)MO#erF#Y$ZK%95GAzR*_S7v4(s$v=96_^=W
z*;z_}AcclJ#Y>p6AOH+I`o{K9bi;}o1NhE2#eQA;$Ku0_ZGA?md(#3cEknAt8cX~W
zy3M|5T0^YwX`)<qzeNQV)jL$CVrg{dCJjggF7d?8M$>vJUJ$NRQjqcfk|DYilGqfx
za(SC_56>RQXgq_YIK5h*`pz97GvI)4H|2dBfh42R5|=im8v);8;uiNjNqCZ7th4bI
z2G(~)o|Dnf9YuWpt~s~#UgV5@_;x`oSYos1aRVvl%;OGw(!!5!@%^`U6&X=t=D4Qo
z;O|n91<@nYN=X*?ZUk$NKMnmLLziwx^LS1YFZc(HtD7^W77*=KcJsNA6d~bDf*W~-
z8Q=Bgtbs5f$v^p`NiLa~(O_3l&{Y=t2A27!+I;2`bL-RR+a1*yN46}?%okCI_jwfo
z1hKSQdvFu_LvFo;e@l<ZNI+T4m+!ex^R6VD3uA9qeJ$Jd9I2Hbn{@XE0XT>a7!L1H
zp12&oXOfc(feEfMa&8OhW?>!~c~+WAO1g^W<N-cBtFb*dFNWxY^K5a3lo9f#^+CpV
zD!;1O>EI{-&pA`;_~0Z0w76>n7ahU=ttRRkV?l3e5Sh#U?3DIIPG>x5DX#&$G@G8l
zQJ<;M-x<$~@H*790~7=e?KIt6*i<R;76KwN!kD>SMf>ZB2&K5=>BpL5$v;m0lO|+{
zn^z;`)nw41BK@0W*9p?%o@D<_2IfLre<|*fm26rkS-b)|0)$ny<9(HMZ@CLDsGs@R
zm9=BMb4;~5f1aP&N9h(NV-sx-5Bl-jh&?mq`@eROeh5Dq%RI~GE_WK3UhdamRDJua
ze;e%hpyGzYt)*6gAHMGYE7clEF9DKo0(2SNNg0m~pns|&H~f315sd0-Bsp~nt1Al_
zuwbJgfv8vk@7otTO?JP#m($_hVAd20bJ-ZQps-;>evOXB#l+nVXcB7$M#_^a)$LF7
zPasSf?uBSykM5MDVTGR*j+0qhXWnc6RNR+@)A~WA%=Ony_!L>ILZ*gIvmoyJx4`3U
z&H(tp?^ICfz7vlhC^Vss`dWn%($l|vzlVaP_HCtQ23R9EiNReW?6WTOV^d4(lk>*(
z4aJU3<M}$5O%q$oYe-1Qt#<oN+Dv2Qm;59@a{yPGFBPJqrcS|?=bM-vnH<gk?S+$x
zlY*=TNq_<;H?L&`6?Q7egZ?c{D9=->qu>qHWDO4wOXO?4qqUxOV*#34pLUrcdHF9A
z5);ypDVPUa@V7PfsK*5T!#yDNdtnNnB(RUzFhkd-ZORA##06$h=)>x`r5mwXUGX5=
zz5!++U>$H%ze-#c&up`@l-%|E7_7Y{0k^*{f!XrCjp+*P{BT&J4aEKD;HZO<=fp9m
zyGEZTQc|-0XZ*@rtoFh0>w^;ugU*~2_PS2XEC7k|g(`*nUITrlf%4M7@##_zTz0H{
zXL0<(-2<enHM%&tL>CKJ%M#4qa{YrTuI$x~m1>v=2R6FYwDM&$NmDmIx|i12lx&M9
zZeGp(+Hw2zqGP-1?;aW~=6lrduXJN4w^X~0%C*+79I>1r@O@-*i{l}5ZQ|rsRTXdY
zr|wz(+Bqq++2Y8B8qfCM!&p}pq-|I1Z=n)XjfFz~l&ZDlY{10C6rK+8uM4Wro&qzx
zv$$5>-sg=1sd>Pcxh;(z;21yoAd@vXxEsC1jsm%ejqNez2vp9=NE&^aGe!hFcbG%m
zu5Ij*JOp-K^ws{Kca_~5tU7508F&<AuCp6L2*4;a7`lhna{^w=)M#iP_OdINw8d;<
zx~EHL$3H;1vxF<+F4Cw%=af(>Lw$R8`|Fy^=8p#N$ym$((E?13LF1b{zpJ8Rm~r5t
z^@Mx#WS3F*J9R{YArc8B>u=&mSPjYrg>&N#r5F03Vti|+bY-W3EG9<XvVb{m#x5P3
zqtD<3mvVUP8$|?=uP`Qo3Gb5}d%WbTiIwBVyWIWvdg!{keH7#68v_lA)_j2q46wtV
z1op_#DegjQqQlZ`i&{I*tJ;GxyP1{QO72Z*2xF>S9&Hu7$9ZC<?crJWXNk(%ByDTu
z_w(oa#YOoi%PZZ}5HYe{cecG-i+^NmP*k#{=hBp&Q}AzUPCkpsF*896__|Z$RaH~*
zF}*o&!2#SyHxXC{k3>uEmBY-?VLff<ChBpzI}%+_z>F5rB1;!#CYFNt;rjIe8WL^9
z?RbB2(eXat*7#V;bc$<zqxarqWD7cXyn^_Zp#o}+ER$iY!`M=1ZL~C{TWz^uCG3SF
zVpJn~KsdsI$V{1>y`*q>upbjULXz_<B0+PfTEN+Nhxu23BL!g9*G2r~vdP7rQG^$%
z?h1KzNHpahNg1j7JRI2NO&mC2XnW#nXTZi$0{xRB&X-)fsi(0i?obD*){Yw+69=O7
zfUm)BJIv5oV*OJcx6fOc->ff*nIQPh;}++NsSm8ZKRY*uK6^e>AV(zI+t@yF(+Ud{
zBUN_#{D3JTX&*Y_P^I0}7T^0<A-?wZjPpo~V|%PKWP;qg6AW|plTABCnl58gu+5M7
zdsc~WAA&BDHd03Th~gnZ&suYTX{PjagL^=$=@~`<1jW7xgB+y4mzie7eUxV9g4eJ4
zw5POas$`A60-wo3;8CNGOR5AlkRr7ywc}p}4SfR}gLydlzS<v%l#QD;mb_#Xf<TGG
zwIM4wP!D3DzG-QDPkOi?kg7>zTg2~4wl`}(y&M<u6~bI2`ciqk#SrVM3Zgrl1YGVz
zb?N|$I0{roo3ZAd>sKE2?_gWSawPCW3@_7f<K6snTcp3z@xDgf1=Qr-hKfi_f-{`t
zps>&;x5ZOPETEjJ*HTw6%F1$ESq)>LGOWe#^}XXpUhH4>7p7uy>~xuD-6wtFg6NM?
ztYWTAhwW>}N0kUx6xyCD;(ye529nEg4~m3TW;PbU2<j*<?~V~2@VU(W#t&a;x1JlW
z>dw7|8Ss5_QV4<J6}FJiyf7VWE13g+cpyTVD#^fW|M$Y6JzO-9^F=*;^vsd^cd0rq
z#=D2YRfUNUY@MP#k0^yG$%uWSapO)4diC)u;xR|yQbE6;Ea0PXR$k%Vm;NJysU2s7
zgePswt}%I`!d-;2@2Ipo;bRFs0&KQCT@;RxxYwepY7P0csGDKa;r(xaU;Th6mmp}A
z?)Wp;!~IC#$1f@)qe6I_3L~PgWNGxVv03ZV%^F?yip;ILbssXcR|rC)xa(}V-C1r+
z`wm;I#%xObh-gqtyUKhMcE?G_iy!%W$S=iTdvgjJtVHD#<cP{M6CSNcw)kA9eZH7p
zV!jm%PDl`@$G!S`EBw@_O@$LREz><&vbO@oafm|oR8MAxf5@Ev@|W%_nb}U_8hEjH
zs_y%MEmG!Va&$ZKcG#_31f;GykQ}jp+AFMfUwOR7Maqhw-JqNvV1U(t=&PS;A0TsY
zF(!t&tIMO#w{zTRwsmXTHA{#1EYvLgBH+)f$ALK5H*Z??W#5b~vJkUKAF{ZG7EmX|
zo*TdF_wHSt3Zu2z?N|eZGD1T`ZP8&(M?t%pxFh!XBAwus0KJq{dftm5e^?tqWnlG*
zHC+XYIv5GcDsVdb-lV5VWUIz2nBq@LDQiGq6MwMzP^Dw8ao~|ceCc*>jnGSadYVWT
zwHAU(V~YT#+>x=pewv<WPks*l9n3ha>S^-p*Y5pV^aojJ+OMIF>oIbwCJ!{N!}ld4
zC25#hn1DifM(~UojYX3y9>NZTMhSf#xh8<LC?eBPPOYy<?t@^rTj-g<3y5Z8z@}cD
zsgJ>zTU8qnElUSF70C7ayUNY>Mux5+k1<q|h9P>yMC<+Qb{1x)UH5Z_Sc%*LfME+(
z6Bt;GP4Hx;q+oI2890UdZ;(Z@vNN46%=Zt%SWkcz-<xk^#@r<*N5_>9Q)u)8fd12`
zr<1Ys%A5oQ+t<C=a;uU@Bx&vt5)iDe)|?-*JqDc}q@pFZoFuoglevlLM}9LOoa#Vu
z0@XA$7Do2Q$(4koRZJc*K9f_#R%?7UNe-ivkMdS$O<GF(-md$^Pqfw4Yksc{xW-n@
z+UuK8=f?c*Zq$36%ujInhgZKVu~?DUu=(L5wtskRwU9pdEqI7n{}jzge8p+<do8zF
z-z!U8wQ>L(p~L-mh5Vk{&Ezo<cGOVL%>8{l8C-MxR&na6H|2K)e8)L=B}l8N1AD3A
zh2;_;zCcRi)+Jqky_Xw)xI8M=7B(g#(y%E?&TTn9OObdrEPp}v;bWAxUW?am&yn$N
zE(5vZ_jVJa#D%?Wi`BRaBEf{9yiWiL!XQdYTCwh2v(>+2@XLJEcJ?G!AR$m$KS(k|
z7Qc<xekY~cL|s~R#yHSt{C=bDUh&(APHEhA%LrPwcUm`Nm-s`<ptDroaR@jElc}eP
zhK>R_(WZhNdauuiuI?FL%m&k*q3^GGPfabs`+;mj0qMlLrHAQrKs$z~T>ZYi;1Wn!
zYAm@Tm39R8N$R|Ln{Y46YYpYiOVc_92CL~mTF9@X^%THTm1pX`AXPQYWAo2|Z{4d0
z08YO!&TUTpWezE*-@j*qw7va3>1D*<lkbabjXr+nVc7m-Gl65aZb~jr60?Z!_1}?p
z9C{6YM&(hmSs*0*84vTlO8u(@9v|>S70sLpJ$Tz`5_VTeMs07dQZcdjH@UdG7{F7C
zJAkA9b$I2Pc&6t=1p=t1o&nvj^L|n!Vk>1(wV-b^yQ`CP>)#P~lp&)xSKpV&sI(>{
zhR%o2B3N=TsUOhjyUsL`GCqz!+wzMRE9hn?=9Q8F$q{$|dshDQ*~U`VGS6hU^z5uy
z#QC{D{2DoVS)aB!r4HGkh6xQ#N?ZyWGrUDh+-q-Ke`KKqlK$+ekU2gwpoj8H2a0XT
zAWUh}CnLADu<NxX4HF})?GAwf^>KbR)Giij70R&MMlKdEIOy=;HcGrUF;95iYIGHV
z^h}?6Z1%ZbRhgEuq5ppJ>pbgPvBb6`4`)RC=i|Tzg;pc`vqWYD&cdl1{qR1V-JlS>
zBn36%6~n{5+_b_%9-no_Qpxv%4Mw*^BrWE++zv!&X}z6!d7XfJ2=6pv>R*9Y;<-U?
z!6a=RGI{0$asS*cb#?cX0O(v-){tImN4tVG00gALl18SMg;{w)IAxmFz^_JS&?GCJ
zogXBsFs_?e&HVJMfwD#5N1A*~d?@ZZOkBjq=i5Vh>F3W@H3tXqL$JX71zR6hz$0vD
zXRi1kM&-Z^t12}AbJQ9#GN#SMDKLCa%M+mG)=`ZdcomZV`XwGB*WGn9G$j8SP}zc|
z_O+Cvy7Fq7jVVWvX(6Rm+xb2J&6{foBad2CtPI^87nc?*PMBP28PEMnW7qo#fC1w*
zxPUP?0x~xAx8*~efu_D{dU~;~N#2}N35}S+hsc=L#Zj_V*N>Et3#kt0To_OZu;?M~
zsO7i0l?BQ2+Kc$UjxN*Eao}aYXmLjLU0pVWJi7P!b3XJT!mp#9keFU(gAdT>hRzck
zz2wMr6rwf!?%!i}E06s8`&A4c>tPGew$9=pMn-eK3YWuwEX=JO9b+Vryu3UD%mDEF
z!XIX4aRtQzc_=Dlk>C5v?@)UHw}cKM`kT{H{3Xe^yW?80<l$efTX2P{ik0N$qZt=>
zzWp998;7((=W8ei&!OiDB!B1#S3I4!F@RVuN-3aY3q1^O`Mp{*y}z$ecWq{RI4c<)
zj=LH*Y`q*7D<DiGtKRTl-tbFug6=BQ=Qz==kUhc8bNnp{ddsW#75H*vpJZz5Bp54J
zX6$+nPiaPK_866_;`~DgW@_+)>F)EF0H3Tb)%s)nwx=O>w3bG@`tzqxRnp8}E<IP*
z_?%zBXCcY<Mky1Z$rln$v!(WmQ`aKeyF0q_OG=;^vRQ~7`SO6f@&Pxfx3H~O{QgC8
zEE2LYun}N2<z(+2djS@&9oGA{yQri%KYhTwfVm66U0#Kp%9wx(n+NfQ)92SBWlSty
zFnE(A4i0-0<VvA@EsLYM*=9klM%KnPJlaE99BeGo-(F#3`<OE{y@%5V=mThi&0kuC
zpJ<k9u7e6<n8PAD;UG0;?lbh1Clp|v5<yR3Fg>1yuyl+(O(ZBHrdEd97f0bf46A5Z
z4c_Mw(N*)8p~I5O*xkh<#|d1IZ8e2qGS4FIQbjfL#mg5sh_fy`DLW43cm&;n_npvI
zpmN;Em87NXzh-c8G)WjJYn*ngw=bHSF6QMII1LX;g~tNn!_^4m-2r3>!3t&+d^ZL!
zULm%afE)*I2Ru*L{Q7^k$(<T&v@g=kxwG8dJ5<4gtAP<-hjsnC-^J;tPj^MV&d-76
z0hy?MrZi7ZfE6d(vv+yK>QA44r_|JI`QuOvTcimd(}6rGN<?*KX_Su^UtlzNsM;57
z3R$y-84-yYv0xb<Uf$;>53-ci#~zfIJ9e5xRaVlMQiRtPo3RH-x0+j73BD(}b8z?%
zfrl?`&NUE{)~gH~1Cr)bQt(_|^&u(2zt|_Je1L~9+@nrL9~Jo!35*noL*aUZ{FH$z
zAta<LBWkdTgJ8R!JdFJ6x)>DHFRrCNv#@z1#O}Hia$xnGl|;x~!ro%#9h>K5(!j{5
zy>ilt8O7Bl0VpIyUe3FvO81#~qIdpdD9?ETUAj``B%Q$#60R;L4wh6DNLc<<vyjUH
zsYnPCDFajt7v318APf1dUL(EMh;s`C0apg-;a<L^M{2y6E^X|n{z$?b!=QM#-eN|M
z=X?Fy-f_;D=#Q9anIB*gF43$*0$W7gcFzwfQU$q_${ZZ#8k7+b)wCpP^zoBiRR;Y`
zwuX*IOaua8AlehKc&L9CcigembQ?bGT_dM>g4$fyD-5b^J6%z;moF}V?d(mtdLh&Y
zdt(-ku?X___q_9wU*EzK^5K{hEm>Q{_Yf>qNLpj#CHT5OJ+B57u(*iq^kC%%qT@;d
z^ut8Gar7D6R#}+?@`=~GD}hk{3weehA&~@<UX%4NhRGt#9s%dn;l|`(ZM03|mX)n(
z)W<01+nDoWbP4gKzCe*tl)T__-4G7hz^yEMAnf8+@*Jv0ASLGvx1NZIxA2L8B59NT
z1{RNd<Hp9`pOeoc-Nf*Q(UfKCXq<)SMxZoT@AotWhQC--h7o_Z-jO>};y5;?)?wzk
zu+AEUBTzS?L&$^u$@}2a$aQP4)ANJoJ7YlX>^nj)GaOhRI|c*h#!Xz3L5N9?!*6-~
z0o3DFW<hhLeE%gr4!w+>T?1ItWnGkfkDU@JEMm>7#YJkamM6k=W`oe6My|wQWBq|C
z9oKtB;loVLu$j8aY7&h*i2hfs8kd&6pV<5>MmvY@PD`_yxnk88&zQ?Zudg&HRc}uF
z?(fqhU$3_B=#>MdF12^H;021Us#g=*JgRnArXij96zB(AECg6_B@;=pDI@@wj_xy!
zjPRa?C&^grPOX7n1V*UaeVQto6>yFOJ&%)Iwg1n{868knA%<g(nT(gsda3QM-jt}#
zT4S`3)`|8yV?IjnEW)Vx6d>((+^xrnFZCO-zPhwH9^SeSk-dC0u=C6uIoWY9!R`Q3
zB7vmIK7&E%HF`=m^+;|t&R<&D=r51!XvLu<+D_vYCfYYZ?KbABbNBNhUmV|z-RSef
z$LI%hO;#-N3L&A&<KE3Lx0`(uNJ$x&0QBoyi~F#i>}$9jAF9z$!yha;UFp1#cjs|G
zet2DqRB{n`cLyR;ba5%gn|$_g%6MoZ1)O)^g4#j`-vmqikPe}al=-Pw2ZwDc2OOyL
zbTm`ct1{R3v}}?JlB5~qj?>3@mxjSbKrmXQ|L)f1WlOI95#_b!+_p={pFR1E0#Mg@
zALQ(jX81^tK~)YobWc!~AwG^Mv|;HNr?pu|NXRbaC}14ewT!Wfa$D^kUI6}qd@=la
zYeg+o!`iS+v%{vKuE1k5i5?4UpL$LjxlMTSbMa3tXXZPwpXfI&Vk6`KE=5bbl5v~=
zhBF>!z|db-z;Mo7>_i>ToxDYH16u%I!d}7kQIwx6TSE#iK>Xq{<N0G@=h1me2epbA
z9fRGRBj!bzTt0}NrfTuW(Tx6eWBQlcQz6a6KROn61@}wqA=;owSj^nu4GsK<nGi@(
z>0+mcqsBGAFJ?#V?KoTCe!uddaAMbOOi7Nuua^j~tpwF`ZZ+eGZnFUoq=4EfY+pYY
zP_mN~3|L{Pg&5djc}ASpEPXHksESZ>M9Q$&&w1?cqn$Rwj=DKrhvOa!qxD6gUa4}H
z(|snqU~qr$?cqL|-2o`|%+)UDqV#j|3t+ENnwjD*02Tt&aC6-8lTxlM9UeK(k<f!b
z;CSQCQTmROrd(x<ku2?Mhe$kk*=%fSkK9E^1m08BLd2Q~WyE_)D5Q1Y@fs_S1E3Nm
zLfn1b5N0~$*T@<*s1l@%I3GWJvERXmb>27cYp1w%bqy#&U(4(Z;tvKYH8w}z(7H$0
zr#-VoRXt&M)($0u8|r9qbt11}O!@M^xWg;w+riiMlCMNL6Gm0wy*qm?uSYR9vr#cu
z4(fQPgT||6MM)my_6<oSC@B11q|ALi$boSI(8l@b8$JSBpjw~qDT$_qhAEBI>3YF5
zyRFx0UvQzK_%_+mZ`cdbgf8aejX?wgv&~WC6(b$%HP2t68YrW&MDXINn>>IIrhkG_
zc|VnvPZ(sEoJ{n>{1Hl-_b@_;{3IvZ!YE6vcSMi}{M@y#x)#4P5nFS2)3w_JZ+{)p
zv66SPTwg{vI}X(l$uOGz=Y0Q(UF}_81$)lApj)9|1I=yZSI(F>&<MB37WS4&_{6Ux
zyy_!6#?o4!66MVPLd+TrVxQwUd`^yk0*-I>smF>kWy9UTky}kH^@?HMMiG0cXZX6X
z&;E}IPys=fjQ!DQOz=WG*wa8~TictPEZ|)VM*i%*jQdBn%v^VsQ<}b;)9!D3;-)^4
zYF7e5#8;Ecju(%S*X>JolcgnjHN%U)Hebi#<|w7y#v}4$_>;#77Xey5T?Q8=fZ+y(
zgU>?vx(J}FQ!Z4jY%FtSFDl~#6JrA6eN~pr<9z-#Li!$$uXxyA?6i2=)f#-Ms=&PE
z?zptsV!&<j7PRPk0*gk*FDL%E>3Tf^K@HgDScsLvNd#Hh#ZZI^;m!;a;yOE52F_CA
zHi!b-rj*q9DKQQg0fefPDIfjRV{GQ9)0%aJswtPDwF#sl0-N<AKE|Ki_$F2e0l~6%
z0;!V__Jsc4YdnZhP8IhQ0{ueelQEunCATvX{Rd{~gDGyO-*M>&lwj<ryGDWZi>!_&
zp9NE?y}Px#bv9mK-Z0btN7)DE+p?VU76u50$VLU4yu6l%OZ{VF=I&5oARzveO#BLa
zb|$&D)+ZD=woQ~CflKS>I*Ag%$BDe&a!R(Bo2b%cY`u4pbN@%Fjc$tqKJWLUPC%Au
zqnyIlqX;`ah;`)Y@b=o&oq{Fj=RK+_3cdDDV{kWhb5$5;kEl9>VuPAZ%0z5EQJd2n
zrd;jjs2eUdbf4C3ZQB!tdYfD+ao~cI!Uc}TqP8;+1JDuJ4#uqyW#dr)<}g_EEmc+H
zBG1&-Qw1FTx{I2;W)I*u0C`bQD~rs;>00%B(1V2BY70<f2Qntq>nxQq488%ZMoQ@)
zIFcOfU4@G>ix&5m;nwCkJrBes5ddI$g%;3u@ADp@(xxz6MZCUF1G`Qs>Kz<u-t@X&
zOc`DnczkESdR%68`!q_tTK{tb{T;}wyTX<e8MS%as)V$ZU)6uTLbmr?KLa4aovY|`
z*}n#8n;ZG;#%eF0v!!SQ)iZSskRJVM8*yB*X~4;;4ekMI=j~ecS4dV`g=cnsMXf(S
zWu{(NV6hBR-x-cF!b3T3L-(D-)n*Znv9ugrl)Q#W|3sqxk)`D)5F^$u`rL!HlSXBm
z3pHwTLYqSL%T1)igo`=b`E#tS!EFLkXytbL)65R4&%$-xXRR=2Ylw<CLy}aZ>ri{Q
zo|M$ec&FoCzgf~KH8IrY4-YTyMDm*sKe!?m8w|bfch9azK{1?sJUgTCV|)z$?P+SC
zc@L90xVE2igP&R4M)?If<&QCN1u9&H{Ul`0Kkf1ziXvb4Z`K_B3XTal*$dRKp62+q
zQv=&~`j(-@E6ry;^)RW6c>Z!<n?xgg`SK{yRUk<lsd2m-kZ)?%7p-sba5(A;hz8)o
zJ21SPXh7QQb4oD|O__5=rL93zjq?KOet2Gp|DzTv5qHiP(~L{>d!_yCxg5Hky)2qA
zJ?>$ofX>S2@*9k!{o~K#S)WV1`qcex20rEO;8x&bt|6%$uN<40@Iv`Z44l`uDOqFJ
z=KQ83{)gJV)WQs)DWbY4y~^ETpwVU|z#$nnFi^%tfD0wrV;B3Q_y-h_{p+f-s`f*t
zF_PErHKi@j6XO1mOY<X+?(FL5TA7^qx!7;pBP}GrdJSO^3||{<r9?o0+~jod#N2Hd
z0>XnNuOXijNICR)h>?FF{_={vMZ#lr5JRRBS_p+z*&c9_u#=g)<>pR(>-lkavB|TV
z3X|d@+yfL;NLoe)tG65@Ucfup@IqeRRCpf-e_-+hGiaDQ@zw(uI?{7R7jk~nKM$Z$
zct5}Zikg8Q=Q~t?`gFS1**m&t*nSO#TKIm+*RN$daZ2AxD61XglH&94M9O$=o+U_7
z8zypTvhry&=+!&qM@7xA5{81do#Yl;#VX^1MQ#K|;likuWkGb%k77L`sVD8GW=n8v
z2Fsir%4_gt(Vn^DP>3M)i;o{YRm8?`CI1*ifAen-oFmj_G=D(nxr!?0@f*=8exNhv
zU)5xN9Ohf&6#GpZ+FJF1KY3jM2XV2R`}AqQvp<a%c`GFu{9_)3d$-o-F#hG1S2=D)
zf{x!}DKa$x%&`cg{K+Bz-KU0($a|>vn*G@}`}41}A!d~&Apz{|AOMn_JsZkVa1wRo
zq!AkJ=%=P;yoCUVJ+7&vt(DNUi)*>(t$Mu&4ryTGgZ-QOxBvO~|HIQ;235U&f5RIT
zQ4~-@kQNa*q%@M!NOy;TD2;TNNJ~kJNK1D&(%nddba(f&{5|u(XMAynIl^Xp?d!YN
zC&?if6;w79SmE3A`*%G_m)ag39si4!2DxW?z&bgolm9?k9>Vcth>@cpDYv$kH#;9L
z+i^3Qt_;Qw(+_XvbP$DR`?Vm4X2C}bd?b3y#~Y!jz_j|J^dUD3?O5NRUt$$~$v3i0
z!)4{HVpFIf!%eq|jfjs)h{))8N2kdeDn&`J$$EO;JqAlJbX{B!ABXV^+S&E`vh~65
zW9FfpLYCI=rfZs+^OlxYmv2?0ec8m+80-FB7w*loFKH0-^yiD;W5MRsyU5AGQ7_S`
zncTNBFEJBS6hZWc?Dc8Dvpy4-?mNHfb@aa+kDO>mQDWF5C+iJ~eWdd9v#hLmRW&p)
z5E5Myp*yon`;N}l#KQVT<letO*wm{k+YJn!U_L_f^tyAAM2_Z!9dPv@f{c?DPJ9Sy
zkc@N!&pZeT_2%Wr>v0iA3|ziP-)I{=`ulpzDoUzsSH>+ZszjbVf7%=GQ?8HxY@Csr
zm4}C>tz|qD0C{X|&^D?lm<`^KM;!Ba#8ERg6jr8<3{4-kz9?X~>6jmZ2R(8l@F|bb
zx$>=b@CoteJ_sT_?KMV6UV&!VSBs7b)IwP~S!@}quM*aACGkL3Y+GD(a)_CqdBuPg
z0GG~^X2p<Gc=WF}n$_0K;`fq@p#Phint;{$jZsl#kJMKPpa$jwoTsGxE&)WZ?szvN
z8uhlQOX)2(-nW<F6^PckapZ&>m^>Mk`ep0>{bSGc$ldcCcPR-cIOuPUg-}P+ZKx1Q
zY*V(r^}8_J3GGKVlc#_<E;>3239+#QH$!SzCzpoCVta74TRYW`&;qEF2JrGSE7Ova
zI=Z^1+q;Q#^pA98A}FE5{l%SYAh!5j)?>b1fF*M$uMQ`o6B$AikPtc!q#r)8y|M)%
z9XmVo8)e9p8`kpPUGvh03MPV$?Dx!7Fq*@4UR-QbV^>7=_yMHH-mfrIGc-yn$jCXQ
zBqv}S9j|nJ8kCop$2MUF5D!Qk{>4P>RqQb?7(vnIN9jb@?(Wr7ZSi~PYdJZo!LC1U
zmdV2X|4nlHG*EZdkWldzpIt5d_nS>|>#Krhr_L+baZko>NX1rV%H5ogEA3OYgG%{>
z0G9_u`QTyTjix1)AJHuq^?x+1x)(QdA_Aa({_;hHOcn17l(NE03_v8!$>Y)U)6g&@
z5bOTnND3xqo93ssy`9Hi%NS1E55X|_#HppH_XP2-2j$|<_NbVROS5Pm_Bv{L0L&0~
zE>E8X2)iRa2XXW}W_8%wOwWfziv^<t)g?2=%`nhB$rH}sd!~F-$Yp(W#ElG8i@__L
z$%ykRmJD}ssRK;3V@$M{5ycAsY5t!UKpYgr-$Bm{wbAjrLG6p%+!GY;QtM`xe!u*I
z$y1<t8kD7khUF`lG@QX*W||MCjG67p;Dv9@Qf^vw#sfob^ZSUTjGSQ#B{>0j-OzCr
zlE8}19Sge>4w}3>Og3`ta=hhL8X@mZi_F#iQG`73o<0>~i8q#C+&%RY{{5KaE@DU@
zEt6iWX8ynHXVU#0$evxhYYc8iN<o$psq0Cp(#diX65Zsf2AN7|3VaN7bfQOrW*txe
zvX~#e6tLk5&(tnnGrGk=3}lWRoBjxY5*?!olQl>TA4z>B4Vve)JUN0qO^{_5TzOOD
zbouLfL+%+Rz0wHJ1Bz`3$`53>1vj~}GOop2V~8+!7vp8EQISjH=DPaA!-4?gu<gL0
z=#IfS153*zo#RSv4UG^H5+lXR$Hq%~CO93oPc*hF!>Au@R02#BruDG+S?)HfZ|%O*
zQSL9_fb@`ZjO5Z+eVJA1#<yv|9d`eu%NFeH&Qif%br)flN1wqhZASl2-9=7qL_s&$
zaPeb`8`;YH4V?CYz0C2-GdRY`B>aUQ09a*^mzT1Qydq3VB^x#&w-{28K)+qw=R|-@
z$zR`y3K%6hYGLDKJ|corlnK}t7UfZWCSE7ghrA^(iG@7uS0#RJGFYQG>9i!V4>4I!
zAin2=jMP`cVUx9|lWSpvA7|tmlBtwT*4kPJ)~@}w@q{`$ItW;-W1C~2zy#8n$X75`
z{Lrr%)+BSlIhOte?o1XRqCUF;df^!n`odt?oVtc{n_mkBLxm3AQ`w(uh+t)@R0XEf
zXY1`cm0LP$O60h*Ldd}ej}Qa9nW+3BOaznesY@zT(I;a2r@mhSqgP?j#3wm_n#VaY
zH6aD7kKOL`6)Q2y2sqCAseyA|*@1!fEU0<6<X{X6A?3ITyv$q4_~gkluuGkn;a><w
z1Hj|pHs@#%(Uv~`sQhR7Y^-TeuV<Cf_*Nvu_qW`@@bw+$XWcP$dzfggO>fAcSAhW>
zX?=rWl+Z!lp7ZF?6;*d?jVZO|3ZAU_=wL?f72VK5frWTP%=4~%gq%b^@Jo|QXBsct
zC(h}g#PQ&jf)b8nQ2Lx<l4|#BH7)LAw*^yv6GwQq@(M3n8YfRbB^{^CrRyD3tF_!E
zIgHLdsoV}m?=AXGEulni;VAGdb@l10b!>H7A=9CL(S1aRd6e%Xu*L8k?n6D@Td9l4
zJ*EkMYF5xSxb>U%`k_gCHqSc)Wp~Dut{R#j>%#Bl>1SV}fe$KShb>+y3M_Ezn3x;y
zIRBA$_QrIE^lkeHTuUWGmFSgtkFU6&>nfGxOKAy><||rzUbtYR-3@MK@D=U8T6{BM
zgcVaV9~B^sh=rBRdR5?!Y3-U}(Ya?D@u4aYiRJrDP%r+B*L`JmHQxC575^<+{hNVb
zCbClndy@1<zo+Bu*`pubYwl_yvZ59YH`>g<19U*2<O(JsPa{8d&+rs#&FtjR)r3$#
z*bhTH{{uVS^_4d^|F7v9H)&sI9P$F6rd$7GaSo42O(40Xv@7cx=l@wO7NR@~NVJQd
zLwVNDtWbNFQrDn&4cPE7s(v&Hf1jk|5-`9>zeFP6Zx4~AxQ77N827+~vQr_Q1m|ES
z@RA;+iee|bBil(V9|zIsE&?CL3q+rgHsL&T)hyN$MxH;~7t(Do6s~iNxi~j%YG1k*
z(^|LX!s@v|3G=7$dPDvAW?YrygAn`tXxf;_?-{VXlQ_Gy)CS###U6o%+fk|K@=GY6
z&d!i^AEdOtATNp)Cll9@eYlLhIm^VifcHzDT7G@S`=3POp^qSWxPQmR$s2asaeb<Q
z@6CV^Ux<g}4cY1iy<#1JY^bwYe1GqLJis<brbjAe3rN=Pd+pKAG<j5Ce@HLV>$TSW
z&s|zs83hrEgImYL;3h5=0{ZovqjOll06GEpiSIpP5YHy|&(+wJ2#E1+F-XVBw72>H
zm!AnLHU?COJB@ATF0+6EguQ4)^8Ef?BBI}L`7}GP2E$Xh_+$HujQt<H0!Lw1Y<lxd
z{jBmp%b^pGK<_K_D_&j#5s@}ecek)c5=2idmodalZz$aOq4=~y;ht%D)R}s%<FdzW
zp1DX-!#!?7(^M6r|IU$r|JuH+|CKID3i5bZ^vvl)n!VfmZ=V<Vc|6eA{Leib^_;ZH
z<FszJt$xnotYw$~BbWp8jMA=e)lIEC2HnHNg?VRgc|#}}_^kSoV!#u>cjlUgAx;?%
zq;xWqk&L@Qa=K4ka9MhE^apvXx6PcD$MYXyPk8eQ9|KkLY0qo7tH*&&d*}a<x$+Vu
zgcRMaP?H#9ZK(6<WKq`)$@}-q2v0RSBq9uY)70Z+;Fy*s$#B@DUhi?1ind$cJYM1%
zTEg*8_B?sxc5{J__#!ypo+gVJn3$EZH!i&wAdH2V`1pWQA3TvD(W%;X>TQ}XtxW-?
z#8m1F9*kA+_jq%U!Cl`nPk5nYAOxCn7=i|8FM&@P#QxQ<@s}>Xth1vBc+hHMF1Tyn
zC`;q6OZ&I-%h`STaETnI2YxkzHHXroa#l8`E=IbA!*ehJj%5$_{UJaoG^9aA5R#|v
zo4xV3ETPpuQ9S$X%hihRH2Qs^K&+X~l@}i{BF^Ppr7xlGjYL+%7Q4<`-3j)MbXYl}
zB<O)+Gp#cE>ql`LE2ScTN%0P1iLDP0A|ezoPq#;FR59nq1O_TmCD`A`?l3lY4dntS
z4-QX!yhi<EIkPNKx59l46`pSviPXr$TiV<lv7hjow1jQPj{ywM!NgAYAAOD<``Y?~
zrinoL9~DQQ`qErvdUdE)hJ?fjabX<(hoQ{^lF@(cjsNb`CgAZ^+PLjC-l%B;5rG22
zabK8y<dxo&9S^X(;457M75<q1+X?+0?$N3UbOiifz9a|Fi+<m?;{5ToJtf$hx7u!7
z-?Cl5=!t-LzSyDc2zH<_d8FRmOBr@D!#(N&lX!&%2x<ZR2sp{UpjB<o88WwnbJI!2
z_`6;Ax{`&-jUAWU%#u=8I3n6baR@i;BEIoICwxOf4vTnw0qCCBtAJ7gARfdQ+N$jk
z8Wb6h3JEPVQOF4026t117l=2>R~dd{Q#?gG#K9H&{KhhQ%IVpG%->u>oi8yEDU<Xn
zpK~>+vC7uP{gIAYmV>V!#YusR-pEE#4TyN8zf}ddr`a@!%FV;^Yu~YB&Dn<9IcZTv
zf!DeD`2w#aJ=1Dau+C>(r6oPq@!nZ}tg~NM08E<dE8$BVG?wDxq8iwFtAqMiRox4|
z5myXOPb-E021%dG(NQ|+bnf4k%4#fIW0|=^DJixQArJTUy#NY<m2HHj=($%9N#nR2
zDFSFICrD>>mG1Uz@Nd{~)>hGB2p0#jG9J#;2@Oqs%?_9=v_jb{Uqd-E`ti+u12J^n
z>Q5Nq!f-x;ooQc?fKA0rr$Itp{efRI1i%48?@w^m<irB66CD9b#~lTj^W*cpyNmoM
z8-rzF5#=G7Ydt2?ctZa8K5$EfBMp|ycEOj@1x_Wu5*%z|q;oVi)pK>ra#Mfi3zI)^
z!ggBUKxZ8Q>}Wo}KLRv0;BQwMdv!`=D?r>*t=!!fN-RS0ByqX{1%V`xEH3#&Fd3dU
zxV(p%o9a@}=UT@1;PQXu+cX-KF%Xf_tNxkqQO*y%<e5U4cSx;Y`xw3r|2<yoZWy|w
zjr7-avH~`x=Xh2;Bn^%?X6^f_t@;llI;)}ciSoxR0gQ8FxmqS4Oy~f4xJE9}Y6#|Z
z0ZD*4;TR-p&3~#4;|O1!nI7C1a;eBjbKjos{3UkE078S~o|jUUHEdD|JW$}iPC*`^
zsOd7=jP3IU^ihBlIi64UzI?a}z1Z%O{o@`jxNe0reF^s;1Q9Wrg~bI}*-7_?FMb>_
z^YUzm{yoQ$z2E<$<E>_UiHY0iKf&>qCE~v)+o-CdlEAg<d+!&b&~gkCFJxhz(&Wek
z$=MFbla+rxP6aXnKC&9I@e5XGXDw-qRQ;*r{j|{xk0mK=3Cu_Ko;xDnfXoOWYIuf`
z#Poyn;-MQBA_;It^Y}jinQujQd#OKE6%U0wHz6ynqtdYz8dOTlre|abpAkT;`p&e&
zU8jVww`u66F)s`+r>{QS#N;Pq_z08F*0`A;hPWwGadUBfZCl*mWZxBJ<Sg_c1@JvG
zDoU7qA6!m_*F$-Fz&eW0d9{9EwAS$>o-6#|gdUNVp7q*R&ZE(5YBauh`87PC^R+m$
zQuFM(mJzyS2+x<af5u)B+S6^UZ#Zhb{%AB%R9b4*KHJ>E7$cZ$rd8t#&oQ!aV89dV
zY?aHY<bfqw?fC>7lXkh|&%i+I_ewp<1{w0Hg|9fY8zv_P-jGs!EA(FrQq6yOw<!lt
zX7;GsZ*U`%0SL45_pSGyW&I;z9B1i}!#h?FEuCkCTug(5AITDPzp^{g9|?d=<H+v3
z)5`6-cRNxjT;^+kkw6%FN~lv)ZVr7xn&l=2VDJDm7e23sA53a*3pah)x}TvP8Rl?A
zMMo)SMMCo-Ch;^WYGv(bWRQl&)?O_i=2z72#KnCFum!X@iJWe~PwfUkt%epDC?EL@
zs$-JoYMtFdq<TxcF=PD>y!2ce1P!-88ot6J?lfJksHDa0^4pqXm&18Uu9STg%w;T!
zi#@qNwp}o$G&Jws^@A$5t*NTkp(d%ePZ_1TS8D+!CMMZncK~QSe$VKixt0(i^$Gv_
zDyl*E%`1}72yqHnS3zwVfZg_c*NPSH+ud1@$7mwdKYGFH>VH=dZn|r4K{s7g=C~g4
z#DLnsN2FCyuUT!PYltYc9q8@jfudmuQPwZQ@61K^430uUv`|qH+wtbS$-usUm*UlH
z@YMrNNXqX9;PGTksU-+Oz)UVj_Eooo2gA?hs)u7m+tpU;M?$c&)mwE73#g!VkJpO@
z2Ymam(FpuRveugj@ic(8=8kF|I+jP^8YEBA=PWA&gSmt$_LsVVz?h9!y!k(b4u@*5
zkbxhC&?V)_C~)wj`qUP^Vks&wHtIzOw5L!v@h2F3IV`588gvUPn06b|+%#+7*~j!H
zxaX0WLx08i)XVfVi;1F!h=ksVRS=BSf^!V1@vp!&aM_WO->Odste_w&#@!2YAJ5+c
z8kwT5c#xrl0D{Jw9c62_o@q(?{q*Hs#M!rK3<RvSihOR*zxhK?2l{<9>9bg5gwes?
z(efrZZ@3c5t+oT@U-<H!9(yd4er=vVKnol&$rdIT{dKvZ{N`N;Ayz<I{;!LkQ?|r@
zF?kKqP@<<KxQBlSZ7?!+HY;o)0xGWnFKBA~{d=?b_>UoUDs6mXHl*|Y^SXO2gIb4s
z!i(fyZxD!82JKWi>OU$gK<1z#2u&X%98dXvFE7333Kk^~50g4h^!Zl&EwEStJ;JVI
z8OKNPKiV9GClCt|zyi7__q{W92v323qTnbItGC9GhYKa%-?JqQ%r$u3S6up6AH?@Q
zLGytOfC$R3iYhABw)of6V;+m;yf7A^KMcK#$asrQ&?JAmLH-ovrSD%2%-(I$`R!3z
zel)zw=<!x^Y-j{HXE=)giyGNmf5?w`W4qLz7MV)bhum1)3hDG)zjD)*<kV^nSCJq%
zfpsL(zZDWK9ArTq0(mtFmmzC*15WFB0HJ-5bwG|gqMPi(!MX4id@d&33T;3z{LH%T
zz2Co@07T^o@*fB<Z{Xm<ypN)dD@lR)`|88eFrGW;F)7HO$TnWqVWbQjM-UUgp#%_1
zTdRM4-5%%{1qv*bdkw%i!~A0Wk*QC;6w3yMfZw9@Z~-2FvQD@2i=&$rK|zv%5)uvR
zA7LOQ8ZY=A&@<UY<k$St<H8lXMU~D7|Iq!jv2HIeVAr~lEaTy<EG>iL4a7a7hyG!O
z{@{u8^A$VZ9$j=%wuAiQ@7dXA3g6z4eZ}WGi}Dz_Nu~0`*gEA)PTmz%+yN0K=>324
zte#LFnE!L7y_<0{HBA8#7hJUnPa+iPt&-mcWAsGpY<^jFnnD)KFX?Kl%N@`N)b8>(
z-H%~UuUArv<X$t1iHtU%DC!>>&&;m=(Cs^lme(RJ7R#!ao9I!vyGcfbk&&BwwAE@_
zq~?0O@LyO!Rs_}EyMBhNOT9l<bMx||BI3o8<=C&+md1%d+91Y)q2&D2fiyEab8XFt
zUj3DrC<kA!@j`q@g##%MR22yo8{sf7uaq4k9%JKq3?KeX^ymOe2;`5pTQ1<~oHk$S
zv6T>)s<v+~$*mP6dtB>?3r?FYyJl(86aTwS;Jv{l<#ubm{?n~`L&xsdy8^96N&d!p
zxIB|t8Y(*T)(2WvR{sU=k~}3bG%z>TAIUD7P}5vr-@w6*v6!xIAMf~UZbrmfU3iW1
z_-k-59-Z32Pszl~`uyzt^W{j^kx^fC68r<oq3IrB5%sK)Ze>-)%h;=|!ptwN@6S<u
zzxasJf9Jp4XVIm@fF=G<rva);{+#o+uPO+8VMcDFqDX#QTr(9%yHgR8SpOHCte>0L
zCMPvYQ;z}s0u6(HQOjf(5j`InL^<jWc=ACDLWARkkdU_d{*}7XJ?`InrQp|sj)tn+
zz4ejKkUAGTOtc4#|IwRn)34cnZ(lA!640AEB{?~Ms8#(bA57ZcpEGK|YX1W@VLydH
z_-nUq#^Ku0sCol|`v>bmpOF(li8fxn3ZpBqvzaA8-rfi#_q$zxl!+(8CJfE8DN)<X
zjIA_QBvLJy+u1%mt193hdVuN)-u;Vu3^+XZh`H~v2U^}~tk>2Y8BxOHDyReUEjVh~
zHX{J$Zlrf_=4OWzh+=Zb*VeZq5+Z6$&QpLQWhRI%{`T#1nFW3pfFN6|q_v-Rq!h7V
zQ25(H1`7ab5Vv*D&)0IQ6fQA3S}HsQB{%F@%zsB-S3n5GBb18+!-sHQqxJ3qFAFPj
z!A$LDZ9I2q`&7AAZ|X-z6y)at!dg{kPkaO?O3e!@`WM5S{qUb{yx7kJcopC(vpJiN
z>tCv3br^6-fsqS=cnStb5JQP{kIfxahqf<Pm}?ylF@oLe5y61Ndedu=uh&{%mqW2~
ziW3W@KzJ$azBlDHdZSBA8)T5bny=N604vV=RGBa6sYXVvRC8x8Kyd~-mm#ClmNXm$
zK2PCo%f0~#kNXasMJU|+6P)9A?G0=I)bQT8xx5fyL#Wl;3h2pOz1fQgq#7P;Y4QD6
z6T`(ehy3a8alFl)9Rq)FiyTJv<jr+mq?>vMqz2GGd&W|ysEyCi+v^!PXloYC=vSnS
z&%}y5a6~+MG<<$tw9koWx1_UMr{h7oyST`*Wz*MJ2AvYyWwUG!gMq2-{iSGTv%l;l
zg>(39F4Ee9ubL=(Ih{_(XD0tuN{(4f51N!|<dy60?&N8f|9wcqZZ>0g(uB+Im!puS
z{e##6^2ppyy`-n?8yt^6v@ZC615f%Oj4=`xGj4JkZ9CINGgU#VMMr(aQ^y;#FAqkl
zp!``peF*MIo#1-IzeX~YF>`Kvjw6Kv&?Xqn<Fs?pTNC?xb?u1|(bo0~o!V2U11<6R
zq{bEL$LrcHTgJ0GKlLJ-;1HUbc7Nw_?>Q!{Ko|cNA9Oe%;70o;>hNiv<I&PY&w4~9
zkGqToAKJ{tRqN)tq>6Tf9f#S79M=pN17DWGWjUIU0>8g#O*zDKtCf+Htu@c~*)L8m
zD}JQs#B$<V>Al?hZ*i_RC`!Um>Yf|Cy6}v?<f&>~lhEeH+4@j)i_sLDj1vP&Ux&@1
z!OlUw$t@J@CQ<NDe;+WyZV%R|G^3Bw=W*@uEDT@)2Tox7qT&$qJ<PQ!_R8hT&XN#l
z+hQ(sx-9>#<~Dh^!fd!YJuYh<=Jk*SPE|D}I#p_0jhV;*0|qgX)Kh_iorF3b-VmZ&
z8T?3Rd|ybJ>R?pEJx=~(gr&%Q`*aHZk*MfW<FA`rcLP`ui(cLJ>WzOl&$4+RD*%UR
zq*##peaUM|N=hg-d6>wIWWEIo^BQ&mjXd_(Z1iBU_ifc`a5m0q{D6>?mseNNSXo{V
zSx`AZ!Cf{B39V2qp+qWYr_)Wm2hxxts|V^6RnQ6pRkS<PS-ij{f4_^u9wSac%10R)
zUF5C7ri;Ub?e?o=Vmc5lxuF0v+e7_X4Rqu+&W*j&6A+Lu63mCmoP#{#zqYoYBo;#V
zP6)Z3@kv4*9BiT_(#<eb^9ErmR?Sjc*gp&VOJ-_Hi%3%V51ohiE^ZS9sxgFGni~du
zZcUMZ?oK78`;gRdhKv*;fN27LCZ1{S8Lc1PT@$2>*rO(Puo<lM{uE_mzvl^BL9Fm`
z_DbBL!ZgG81cc8()NJ1Pl@u{q_TZbiA&63-K#_-!KTUcQEUubr_{@EH%BriYc;yzj
zMwY5VcWE;-sTo&G&9PgOG&vsZRg{*3?NeGxf{lZNkk@q}bKZN9A*D1|iz54%re@st
z@9%B}xqZhc$H!mAIRWDKGF@AUyC<Fv-OfKKD8es3M@~&$;j`#DW-byZ5d%dfZn$_)
zSEsau4BIb!#*o+m<b83&;+|Y0Dk{H_?I~oyyE;32ZLF-aTa6SS!rw@fevDvvLCeki
zqI;^_;O58P)86k0&nI!y<#v)$61a(BKUC$$JNl{gPi$LWC;a$ELKs7arL0Je^EAS%
zeNiOEBs{ZPHIxBDW#JFGJX^4YA=5KX#QTncI@1hHQtWREL&7O0ir54}oe-W#E?*{?
zyoEJ6>)ZOC3I80%{+EVz)ea|z$Kjptr3pE#?W~|e=7GtCAVW-h+smaDybHFY*<$yX
z3+7hVwn>AT@D3O*F9VZaSWz-)#Og)=UC-c1*{E!CVPObv1~y9qD4oNKP~@G~cCp(<
zMpW$b2`{$3A5fR_6{f)r$c7xZCner-aar*}C2!OMI;jqfI(qt^|K0yB;Kj}kt&ja3
zlo8_3%|Acl#z{*{AI=>4n-pE9U-NhWO!TJynfe5r<e-wpN!%c18r=Bi7ZIKC2`R;V
z@HFTp_~7>+bn@^uxQXw<ZQI*p!WpZr{o`x!SIFr*e2A7KJUXjG0=EugfVuEll>U!+
zT?j-nv9RECK1(W;Zw_;npcwiw&4ahP+Y!Kbc7?|7H}yO%c$sQO&gpdq9Q;xPZYnWU
z8yhGp9QtQ7s;lY$!Old*k=VWH<W#-*sS;|#0)$8JB73#I$0sCw5obJH{c>iqI$XE;
z>T}A%5c9{6?N5a60oy1mTd^Wd2ny-)-+3b@(hAll_7*=K?)-yWQY`QSvf;BZJ*_o)
zx9l!L&rsRQDl5ZJGpBI{K4tnvUwY@QIH#fK>}Y3@?Vsslxy%B0ZDV0+Vd00aoEiiV
z{<7;X|1Cfso*C{TWXULtjjgd6O`pcBZl)y&gMN48N)Y)6=M5jrOByZh9|;LWAh??K
z_}br1`>C)Z;k9`RQZeS;GL3}T$R>T}o?fXU!{0St8M;E2+?s?9zI;6Oe+^ty#Fb{s
zZ2UwHKj`OR2a6sb*op;Fg!WgI7o%aI-SzaH)p2!4(UtRRo>!sQZp6D!-PT4tRl&x_
zBoNiv*%=xvNlkCSVM$Q<&Z075+rf434#7DZEiEMIrzuRN_^ZE6U$!;PVMrzxr$}ii
zkd=R%Rao3L)Z}`6;FGl;_sEo%)f-2`&_u)6yUsj@Uo`KjWZIbJ^pk*WI21vhg@v!g
zQs)NRy9a;Y2x*Z=Sj^<vP%+1)jjr>xy)2CV`g4YBao^EHmsIlMLxaE~**gX2HqC<-
zM`uhg9&H-IKM&lKuI?gM#_F&AgvcIq**yXc<+541oZ9;C*y9Pz=;$I5a``te_(?4p
zZCFZFym}4)_vX3_z!w|n8}h7e3t6Tmg~^NDrv3<%d-cvsq$_wd_y(=^3x0)<&sJAg
zwx~P$HG%V;H*eMB;H|6-Pn5mWzsjiladin!Q1sIkI!d7|CHPPHCgz`VQWmhXlqO5x
zzF@?3Ku}ZDsdV7de`mM1vC|vKi)VX~i50--Dg_wELz4;pLrdDVKYtJd6pe`z>Cj{E
zaTY75r7$^=@%!ra{7yat;@wBn#oaMEvd3-iUq9MZCQMe$RN9|A80${YXw~w}IJtVk
zo82^R){*2+izk-EQ7KH`-PLDjZ9PI&f{MV#{vP4&X5lViW$Xk`bCD-3L6TjZTJ_#_
z34ZaWuKLC3K>=Tc(-o**W=u{_+U^boydh(KV59^GQ$fYT*qF)iD)!qDGaeolYt}8!
zY<t`DV+Mrn`pB$&f5m3J^-!{qAC>G+*Hb>k{dm?p>!UXW1ddJsJ-(k!!R2NzpaqM*
z)5*}*dIW~p&!(>!Qqs~pBDzF8*2u+3NC+x1zv1>1dbR3{=Jqcdx;>Z{Iw3qx>@*kU
z3h5Wcxn9@K5a*z0?RYo3c=uHLKjUYpAwLiZM%mLvKl_T!A=NyU-D@{RYY8rE;$l6f
zf}i|+ZYP3w>{;PHdfM|~m)8jAqQun55B%xDarhXBF~*-?dF*!Dwl5I_$=dPi!&Jwl
z8mU*`SuG;{YNk?_O&h+i=H-0!m?~CGjF_J8+3Gt!yyF-8lEnr2|Fi&?G<6W?AOwGf
z)nVTHD5I!o@R~5@SC9W20VZjq(n}6TbbR*c=4HE`5eC<TGp8Wcjj3gVH>V5<l#n9t
zxO8`9O6p#M6y=NMCm0~Nq`WS7HWpRE2+XK7IIM;+?Irb<g)PUEv)_Gn_w+)_bWyzC
z3>dM3CP_4PCWMQ_<m%|^vmi~>!QP)J|5n;?o-g)h93Fi~7H+=CjaAsFq`YoVoDPT(
z<CBH?KZ1xz*sC%tkER2>KQijnxm8du7>xK}YkJK2RX@QMpp4+1)(YL48%);pNYlsX
zaCRKeOm1%eg)i4Unomzsa0dkN!<4~g<;7Y>_8?+Ucnt}%?E3pOw#`9e98^*X3E@%A
z`pgUouetQMB8QmEvjW~8MGa$T^mcYL^0=AZc9!7H)!jN?jMtlT?EK(y*}rIwh<Dg>
zI%GIe8L8q$4IvNK9P1+nxi7m}&wE5AuAK!WUK@G%v@h<?)v|uPVSc~p>^NtS70&9u
zVRrrY3A>3}Nn7xTUfsZ(A6!g5e|*bKuA_I^3zgrM;7XD^pOAW`(`j@1NhCuw!xLyl
z-|&u)5&N2OG4nT)N!B7Xk*g*;pdG#jH5J!O6OxFvnbhtuFE=sS`|XSS4*7|?RrZIM
z<D4U2p}V5eXo(-m@ZZaIKP4Pv*c>q+E2f(&+pehmvIN~=FuIpJp9@le|BjEsvFl6W
zyk5jdsdz|Py_i~=tFTa$rodVS{ptG~vp6jfuW@LT&1faOz})%X@cy{{O!zuPq+29j
zrcQ3*K;*Cms}mH`zohQl*cNKme^|84iTSX)0*dE(sU@##zZVZ2n`$S9FEG&wnQY%h
zAR164$LO?Q`L(aU*nBO;EHrtyH<l}TE{rKN5;thI%xuuCBQxMaLWqaNV}4}bfU0lS
z;|K>qj=2MP!z>i9Zc<ktpV83K^;n7}uzd2?O6IhiZ<}+GS=ca}D>H~E?O*L*SV_2y
zFW(kzX!vR+7wSBcN^+bj#(TpIp=d#Y64%$RX69-jP=omNU&DE+j0qPL$Jy;37*If1
z_vxJAqRy^4`%fNn!(|8NITzcJMd7E}M@N?%$MF~Tw7MuHesCu&?xU2mvj@EWFV=gl
zLJmUROy}mKBv-Kgzl2%^tTKw|xyN>onf)N<Gd^f3T#78+<k;nNyXpF*LXf9&)Txbb
zei+6<7A5$b$Z&!vUm|@7|BAIjkBOeHXIH0SpO5VEgV-^xJn|d1Z%<5VY01MYO{eZ%
zOZ%jH^ZIqW9Bf6!_+8G|ac8$KaO~>$>r|S>BF*;o`L^S|BUEAt4b|S5tZH7`Hvq+E
z4ArO7itekHmry<4PH<qIra-Mo>3h8Pv};v0$=~zhx@G=PBsGqT!m(hzRnbwG#2F5H
zUz;!6$F|64ij+x`tOTQ-gEzf3n158F0h>a@FbViizp!Xv!}gi_Ydh6}QlmeP_GdVz
zwB5|?@ozH>y14UHXt73P1IfGuFv5r!c3X2eV0qyGbmkZ{aI(H0C(a$ME4OB3DP{&R
zX4puUP)B;s;G=uQLCWW1MZfO^n!eMkw_bPcrMi?dc=19^>~e$v^&P~u+tKHxqX#`3
z=)jARVvbYV<IU@;NE$T}8TB9<-j!CqGhmwVv@Aq4{rmS=xV-Ng2$6y10^{VKEvrVJ
z3Qd&WsVOS~s|Rdsl&s%xT!r-`bmhitT<}b3C7(tlE9BtYntr760~CZH^P`UYJTykA
zFzSjoE*dOi^=SuYw_+WOM6t$Vz7BlmwKdD85v)WbVkgQlp+<p2C3XqWz&X2_oJ!j-
zIqD0^FDe->c!a9m*^)V<I@Qjpx``yGq`b#RpKJ;%Y5;=KCGmH@qJzAOcCMDnJTcx4
z`Tw2JhQ}nlPF6PxkwGN}4YM<rb8dGL@c6i$h{|riz3(nArl_S+ecAZrs3sR<A7k*R
zHMD?6?zU1kKE`#sgw2;^C*S<D#?Dlg)5+&$IGBh%K6Z<xO5+pza*$*uuowiQ2c2u5
z1Q%%J=dhV9P{o*mmHI+Q7*VCAqtf?`%UORW=q%%TOgGXyg|lhK)KzF=ck?616#Hqi
zwm4H|P=hUYf<erdB}jc`uYxu-=KuW%5wQwgNNmZwXDQTEMdz6J1Uv#)oY#`v^N{1w
ziTGqFvPCKRFI1T{IFXSGhDySkJY=XEso0sTr0VTf>lsIF;BrnK2C9-2ibe??orFoI
zkE^mZ@)YD%xp--!iuJY+;NkTGZy9FKfjraPVLPsw3Jte?lcQh%1MNm@#KxdbHl+o!
z4DWgG%<niKU*zp&EfS)h1+-%E;;L=uj?hGDq^j5I6SpZG`C*KiFr9ML6-6|-p9&;;
zuQLi|T)K#J71d`)rO45gX_&VxL|xR2&-JgG5;H_}JGml#DD-{K-{Ai{3AMuFd}Ri;
zOcX*%$bwpXdM`uRCZbF^Qd2BloGVdKQAozi4BMt#W3|cKZb5<jiiSDY_=0VCozI0~
zEPH<l#YUogV&?MEs8Jb=KMA;e=ZA_K34+JCowZ<1g7P4kiZ740oG)jE;<RoYf+cxx
zc>n%YrAW>Cf9K=gE2k@b8<ydChaFOWJFm{9{#bUa@g#Tt8inIC(%+b^cG>w)+gquU
zJ#>$|&S{vLOBS&sHmx&qf^6U!02@0kHwR!Sm0u{pO^JY!b$5^r=bv{Qeh0C+>Z!)z
z>jJjtEVQUL-CLt|EUt|Ar3DoN6X+i<79X4(7-8+XMI+_OoMz|;rg}SIWyuh*3iuSJ
zmDcuB)iV$TPnNiRv=yGCXlzJ!xC6DRv602*t9(9MPz=EIoM~csnD`Kxn3P@r_w*-3
zVToZD8$nC@M{~n}^X;u_-WQ;9hT(*mi(H1iJ9}}?en5fh*@kp;aiiYLmkvvFjCeW`
z170wSC^iC7=VA+a7>IY4mV}9NLIj~iu|<<N|Ji*@_jw&9C@?+L)+P!EP|p|h-q$n*
zh2idr9}f(~G}K}NX1e!s1s7*>vJfUb3d)SOS|>0t;kB$B(i@ddMn<xJ^$vbRcC<bk
zEJf*I(yAI38#y#ShZ)q2H(nc16W^^%nA*Tz&Xo`%oEy<X6V-Lplgpmi6!}fO1**eG
z<tI&_KWFk0sId6E>~(Ww>mXzS)>08zE0Usnb?e+uFM2Ok=;I-gYy}U0Xt1)13$@HF
zf7FDf#F||8r=*k&+z_Zx)A#_Pq7sdGldj-O=uC+5N-yaY-W&sdO0wo0dvncA*%ZtF
zTGN;(px+Pmc$avxNL%SBPeV0B=FvmK*GG&;n!GpCYqsx<^-fAm|No2Se~&(yu+B7a
zy=igT{F@1>LbjgqRo>p3+@Ey5x;#@|<Qi?u(C!=w&;zW8x^OPdD)ps;vAvAWQC<zA
z&~<+rI&Og$tghm!sp5Pw=A5qZ`omIKv{x^*S*`z3;jM9)C^V!(aE?zsm+)}lz~k1O
zI=5*;BD?rAv-P9avNw)pwLq9;vdm@bUXieZk`{SPH+p|QY^a%QQ}KWDD%}tB10Li#
zoH9V>uwcw1p4jivOeW)hMV^#^8yYaHmF*g&cYB_E{*wt3sTf;hw4Bs(WSN}q7d#M^
z^{w`*Nm&w)f+9UO?)xmb$C&<Wo#{o{uW)g38pz9&+hh=%fy}kps?&k*xDNpj_|nwH
z$6EtEFwdi(!UxKKVi;gx$j~1xeT#q<@^z3~-Ou$nOj6#ddgJYcOH6aMp0K}3={fFq
zJae~d1JUL;3?BltiX6}3cCiH+MP8P(tpMNMOq8twZv)|n_^Le9O-))E!v~I6zppQe
zUb_=GIhCSY7z(Bu+f52wyKJK7TYoBy=e99<YdldXaGkGJ6W}U^LCkGuv|H|T_+bSP
zryTRapLJW{IFLM{NBT=J0Zl~bv{nL43xdh_3E7U8;)0l~u4L8yJ~>?hdkb&XY~AfR
zbD?@q{HZoq872nWSEs|-J#UbF7pT|ob?gtvT7rN|Z@BFSXq90j)&iY|e<!9U#LER!
zWM{S>DYQK%I?7>t+N&LVy4H<U`(nfW#;|7!tif0HPBWn7A|l?JEVn4L9ldwwJbu7P
zkt!Gk@iampT`run&hc1-|LIVopuY7+ihtZLUkkK|!|Pr<q;(76y$szMFAk9+k1Bc1
zTkRVi89h^J*WY+OhsDC~=W?Sk;yWstYB8Ox@YUsXD=_vvo%vZJ;~so~d|v^0$Eo#G
z?*$G>G4!ok>ilE_CA$EtgGZ0tmtrW40K<SCEna)A6Y1Z%u;%5_yF%xO1@m}JmTi8)
zaK~9LjlF`N2B8{uA=Kj<$7L@35fU93k{8xr03PZ8RGTzU6D2S^hsh?-`HJIrVR`PY
zQO^Ina^aIhyuaIOr;w!t0OrM$-6|d(867Vrn$$rfT->Gu>%PO70>BU|wRB*0k8g%j
zrfR4s_m<&cONQT7(_tqkHy0TZCE>Ps;`V|W$ZkQp@EvZPxHzb4x=ZdQOkrJB`?Y85
zlysp|izKF-mW~Fpz+LOfXu0ut^Hs7A4<6ba9N2Ls2$R>jFUlCdzJ&7?`m;XvmHo(e
z4d#)PT7ApZX%Rq>C*|x4S++0{#5O<{Pwzj8SsH>YhSu8VkV!D?+t5Miet_uvVmY<*
z=j<dXco_=Rqoeu^F?fd;+B)t?@RL_R1xlJ+MeXoABO}nN$Z0tkU$SXuDPcXhi%inL
zd@wY`0+W#3CK_7MduAs%v~V2MrS64`*FJS4JYPY)5s=^Sj;(-M1K{M+QeaXs3^cqg
z=S6zrr)ymaDI*v&$E$Wf{S%LOEFHu!SxQSuK|=0LbGYT4hZ2qQ=Rh=u(0)9n_aBz;
z%eQBbLRia0mD9a7_J@IiEX+1A>?<k^Gm6c);t|SHU>%^Lu@Usb7EAuL03<4`G{;Np
z4~wkW^qNplHtlW{lte$2C!lZMq-LmXW@ifzR_w&Z#|;^c#B%In62127l71`A_|ZmM
zT3WZQ7W(9^v;?sNguD8B$BbA<^{wn>NJJ&f@ip>J)`Re222GH0SZav_CQDgG<!$;9
zyqVeA+2Bl*k^xc_;uDx|Ujztq6?*KJHJCR?BI_5Wm43ZmTek_%)R21)#>`NsLz1@!
zuj3<r#s!zkC}h!bFg9dX0=4h@8;YQF>FB9(7X3mK!5fH_gCE-8-_!T|v*wF6+iLjY
zJ_;_~J@4ja^J_O<W2Y*wKl6KzG_A4S%Mff*?|h9Bpa=k)p`kMTQL!fpg~0y=74RiS
zhsPZ%TU)q6<&PlD<I35eKc{?I!l}|iLI4`{HP<Xo3Im}K$4bRP`j#`My}@QTxvZ=V
zpny8m(0}|0r%jb$T)e7h+7rat=fIvVqU1s>1fT0yS?TV=l9KRFTo+7|=7n6)hw-}X
z6Ifvadt2vp`g^ZZ^*-$`kK<RB_;1Ar&|apj91w=j<8cf=P;#$ND@{>f(jseXkr_gD
z>9JN<@IL*6ewg9yC@v0cUWRX?{125!3zC#*#d>Su{@m~1C#tKLHuEvsT>Y(oX(_a`
zx?CL@Lj19p_!~MNK|Moz8HSBR4G6A<42NKVMkGIqV#MIZgei$R<)bz=5RLs4q&Zei
zOuX{o$!YK;{knOi*tq!cOA`(G`elB5!{bs7bK_^4O7!e3EY}&`L+iH2pTAQ1^bg;C
zM)Y9Jw~hRXd)!Z)E+50uWsQVriIf{h3p*s)u8%<zoJ4Z|k)-{3B)5;hL@B%0BSPB=
z-I%FwJdpJaqu!9+CBlF9?0T#EO0pxTry%S>?^h}G63q>voZ%Jam50lMk_JH+*x_8r
zF0{!WN(9Ptc><)%W-`jkY(Dx@WLFPD!$`9fh;<_YV2^1>CB`SbMJ`$7CmvgGc;Lc4
z1u7Pc$@-e&VwyDRf$z&f&PRkQvKM>1cUtW(Os+tvh4HiM({f*(7=KNOU2EV=kX2Yi
z^)r7B`SF?V8*p`9oUXsNIFRa{BQRdOrN3FOl%c>Xaxt1pOiw-k&nf{H%fP_zi$us?
zjrv2kk>!RvO}M)Ti+>6<6lg!KFro)2O{R~%L-;#cUnVk`F4R41|0ZkT;f{Y@S|vT~
zkwR9ca$S3-Z1CZQY2aw+s_))Zd|tP1UgD^2SH8>~ro>srySCKx%Z1RZKlZIdDu=fT
zSRO<`4Omwp+Z<d!i7a{T{}&THq2HpS3}>Rqz<d4S!!L{z-Tfq48EJrguPn}%strIU
z@sUwWo!P@4RO2`KJRT3srmA+KgB8{aVm8;k{`~HSyCn%Hqcm3gBdn;yyO#P--qSp@
z%Rom^ukL%`MzMx60u$4UPvU+T@Lz&pUkRoUuhu|R*Qb&-!b*U<IB-Qnfn^*so9F|j
z+YxK~jDrq)M)gSxTl;qy`ebg65@Gnrck_Ldpc@G*|2D6q*+xqtOj%c{xR;uHmhE;I
zq=@+D<$aWoA=`}9H1<~3S8lt%ol$yaTR8gjJ<tdb6UB_$?ScI>Df^yP^7Lw}dSnqi
z+&PCsJ~rm&u@9qjvs4GJQU)^Fq%P*ly{}*HZHYYkg^vsoLetMF9CvcFeo<1|Ik}#N
zAte(F)e<`1Y}sEWydsZ?Vfb;#h!gp|gDcu~@~fG%$6nbz6_Qa2rUd<K%qU#N3-%+0
zIT!DIdpq>&nuD$7+1}ZfzV_<h4FjqT41dfg;h_$8#=S*#di5ec?o%6K5f+x=z0SqW
z6x8tU<#`)C%xB{clW9d|49Y>R^Cfs8ge>~gvGX3P5yq@ECmy^T`~04Y9~q%aZE1d+
zmjCByuK!CShW^%x{jGg(G#~w?@hpj`Y(O@nB+=Isg1-izY>nSX9LD6{Fk^HwBGDXs
zxT>Gs=EHNAb;@Cz-)-kxT%0kmfht>5&R3>k-MHAl`fg|rkBUbLX=%QA@O3A8o)%fb
z5pr@ufd(MwD^4&IRb1Cs6lSLBSY2&4_4IsE<6zjnWg38z%o<yqp!AAe-rC!>tnepJ
z@-1`-2x%GIutn#Yft$xwuvj$Jcx~Izpc&Z{q+mV@h9)NZCMJTZe_ZX>ZL7C_@C2)D
z8d-kS7rf({O(Xv-UWhXMhMArVIi?+M$i~Heq+nP2p59|dC<ZTnt%c?@MA6uPrikB_
zvY-&4WMcWz;DZSoLj83WLGYiMQ55-g_@+=}9f$_vgM-s!0lGy*k{mV2xDbuKW%lOs
zE`W2=<Oo9r8>|Mwgy(v`7tC=yd`tmSlpN;MnZO;=g>${0G?+PfiDM$d%zrH8M1<gs
zoqV$YWxC51@rl(A;i)hGK&;D-)$X0bGe!yhu2wTm?0EYSwrB~O;0=+tk8~*uS&Goh
zEq$~eMG_cdF}B9O+iEw8?U0tidw=iMHPd#dnEloM5(^?sIox~82pRw8o=l-KfigK`
z+*?D=xYrvQ9`D=Z3FDm9s_|5C_usoB_6*KmWj)&Rr4dmse>K+BG2J|Lt=t*^ZDWWX
zL*N2+D-L&J{j-#Zvyh8o4URj3kNAH2B|eFBp3G(Tl5~1XcAWv?G@ciEF^#pt`8wA?
znVnH@(jOwa_Toh~r4LbpO4Kwv^M77uFf+!*Og85QF%^mOC^#C!cNOL-N-!w&DME>c
zl2aG$3mLM4`&LGIL)i~SMCfRS2YZmT>GK#|6i-M(n=2g543trAC`dwe)wI6-3}2pU
zWZ`8YM8H#7t*3t`Dn=JFGQPC@+R+64Dlrf5G;@g!37>ntQ~S@AUV^E@*eWL!iKwp8
ziNW!4wba4)%S%X}vUg-z@qlOeFG|8I30b*$_Rqa4Dx9ad4bImxsiOyH$hMuoXP*xD
zo3MUgiez>785o-8A#23+Q64oZi<_vr%QM?OTvwRM$;~m>IXK(Bh#CGy%7_(_G-y<2
z!thjb0V@K>+9xhfz2p~COZB;a=I92Toot$LZ*T28vY2kW_{oLLY`luea;$p?0~}Af
z7^%O};^g;*IP%Q9nf=g8{$yj-w%~^kH&bs{r*h7Sdp%!|tUX#&7Y)0gsj*(4J~MDT
znfdt^v!z8<PUvN|Dml_pGPpr6rQ=6NI|Orv+pMtmI9(N04D!F6W*-HsWSE=J4fVA=
zo$<u7<I*m4qb0<Sz9|WqUNY0mfAX8UrKqe*zE^@;wOjGb*ysX|hugdMQEiDs+K2kD
z_bGU?B5b%7t%HXpK8V<t6;DoeNhZtl(L3u&us^WddDA%dRW-f#z{%<>ZfkL{o7_x7
zR3(;<)VijN;*{v4hDo7ZWqx{R`R90(@#82HC@G&gu`I0Y0t(9duRpiD+mdoT9IJ7{
zG?!MBRu3*x;iq+0PsgeH&dpI_oV{!IGEBLa#;G}lNpqr1d?lTR70s-5*ociO-Lm_o
znG1H!yGxb1JdZLpsExj1Uda>95wZCqk<7{c+SLLtH1T^h5(3`5N6rb6Ta>IC0(wmG
z&GgiC3fW53pV54edbSR%%ZrMvA`l*E=^r}DvF>*SuI5<1qEIl_a>WW3HOKmU`|?^J
zwrubt{`>#hZXhI69b)~^(uaCS)#;v6QKo;72fA&q{VD^h1_bp33$wc>yN2e*#BOUc
zQv_pweD8bD{G<NUEl6#ET+$+$>lrCEHn`Q$RitFgo|^3bti^DPpGXlUVa%!l=cOUv
zd+D~_vYqkw@;<uFz{p~vuS!mDN7wnr%jS8<8^y-~+43FP&LfNM$x1X0Z2xWS=QREf
zIbMvr3M@2a%8(&?h@k(@rc=F90cERX#J%6qr_h1|$MUplPvt25no&HBBie%r*zY4$
za%Vthxwsj4&IN*an7bjRd6>O&w5_bWOk;ZicN0_s<R4^jySWGwWGY;4!@K1<!j>sX
zLK8VmHDEsZhkwum2l8US=oQCQo!bi<qxPL^5Q^mL)|Cp0ze4nA7uFU_*q(1O0Z9y$
zH9&W9i63&p!w0A16rE4k&|uL05ircBIj|Q0iBY!%^mdNFr}rS!K}9v_pT6wplah=(
z@EJ(pb9tVUZwhB!m<DXAlNB~Li=ysxCQ4ObW)^HwkGQ470<KlWlpcTjlB3SC!l%ug
z+PP$6c-%PxiMStjzHz^Pf|!kJJ*naQL7(-S%!<XKil4Kqc_@Kuxq@@%cgP18{!f3d
zGP+BX?s#U_f4qQ>d(Xc%+BgI+5k1bu_jx<e`&(HFbdllROWYpnFBxm)qbh8HOY?Ae
z`y*>yEtTzNn*%q-OA@j5i+Y*eZhrf!>4peJZdu7w?|^W`o3;r3cAnFZXUL8Dp{Of-
z>bZBw7T>bO%S@2eR6Hn=*BX~>PMza08PBV|MErff@%L2Br>{%3!Q8P%I&~Ns@kESs
z>~j3?y=$bT<B7g)qN!T?OEQ(W(WO5eUOiXH%>1#&zan12%avfUvWNl`TM`kq@Z#Fd
zDtC>ou;Fn7<a(}RAjSvHxN2-U<DA_Eg48u93x+pJ=P_SkycAuqatxM{d?A18V}W@{
zsHbLIFjwue0L>~L9X+DfY@l(1#5RPd5~XIX>rDF@0XaW&91)W#WP*4@%hQv|q6C?Y
zAg5~`nuz@~WWuL5`+;__zhL{FIyzT&;-V=LVf{^T!AuT|0i2}%@-Gu2SGy=v(eE*0
zyw*)t-<k#?JXakg=<!bF8b$-qceR8pKr9(3^cI)b^GDYyW#YZt7h~vR>zpo0$~4ds
z4_ACoH)rnTJRjnSyOQ6;Ib6A}mYy1%MnXWJGVtv7)>k3kQqCVvQVdP+Cd%Agxw(~_
zsrld(K(X$0r8(XL>RnIZYNZo-Fs}J*x0YZfK3x4WS8GGsdqgrL7(@TVv@$nyP1EIs
zv~}%jr6-oh@r^FR@p|RjH;=o<*FH=F^0JYFMR{GW$h<A`hS#6cX<wgQiy__>#lj6P
zhx!*Wwz=^*B+?=E)~6HmL-M+7Wq1BtBm+Rki(^bHYUi?s*KLztV0Xj|{^>F1Bko*n
zabY1>>`Ri85-**)!|2C(?~W&2*iO|i;@;FxpN{y)V`nStBP#i+vx7{j?u^PAE2Ub>
z*2g{*X?Zfj75sW)>wT!V8?Os;<B{LtPn^25y5_cR(vwL4ZbVYU$AV1vk4R*}iG%z4
zj3JSzh~`?<g(MlS?;Pp7HM-B#(s7U)OWUxI7;%+wPXY*{dt`+?*gvv4aV}N+6t((%
z&9V7Ox3f9;f@?K|KA0^{;C99e6^YFJ3f0oI07ploVe>_p2_b@uv%HRF4Nak`o;GOP
z-tbL&3L4_q-LEy@oV3aCQg9ig5@#2?l2SIXuzqF^Wycp0@Z-Do7E;nO{bx6LP`1YL
zT?Iy_?Ww9~!5twpsQELM-m}u2L>2%i0-<%0KifXsOwY%3Zs4k~qvI8k>MgD$q?Yx^
zNF{IT3a0Ig;gON|V}IuAHz`5i#%HiY-V&Bc%8!R=(eM16o)e*9a`9(;+{mgWw=ZGx
zPU9r{6hMP;wej08Jksyk0)iGA5s^F@Gu2AX&`$(5Hx^1iUW!UeK17I%OQ;vVXDbeS
zW2yHDvM1qu(*2kBAVyEmqTKje!eg$VmYu7d={z-rlIgWw>-XUCyC0d?2>&pgp0m;0
zcu=3%{~vV(bSUUn61Hy@iP7&7k)R^ZWU1^(EFibjuw5sxXz<6gh}HfF?ME1(8PA(~
zTl{}o0B$SCP+g&24M>W7on{;%%4jePcF`q;k>Iq*QjG1%k&D<%MbJc^14V~%Qobrf
z#8c=3_1lBHHJC}5zqwCI>zuFqx`(ejXEK>y_$sLc$Q8C)mSgzc@m$k%0ARcAEB?XJ
z>@ma-f<a8H&i>(`>-faP56;@vzdL!ld=R;W_zQszkm|1Wih~G|l46;<n3B0!!KtMK
zk{H!~k$bEM|FQxQ&91@rl;scf3NRcdP!R66iC+p;*GRe0EL!$ZJWGsY7~X4X<8ZFZ
z#UylpUb*8bk-Fhxj<4A?gI67l_p^fflZ=k?Z+}f}B+oBYNC??B52d}MQGe9MYf+@e
z1JzAAt6ec(^R7n-&+FtVd4Xu>wimj_bU)Y3w|!3deUwEoAO6fN+sF+hLhOBHz3aIw
z@DQs>$A=?MX;VNcxMG<!l3;l%Je2!&nWBlZsjlTcM3zJ~<%bB2zrv??Kngn3>*0^a
z%yYpgsWaDA)1Nkk-K>A&FuR}hm*u$u7Kh<XJUuH`o{_{x@D1Jwmvqapp;3u&;}d;}
zxfK}+))(;>?x!w8m18?k^D|pAL>Nz6NsH2p9Os?dtGA=FmV*)tM4#0&xZx$(w0Dgk
zG_Pf3IFnTWa3ao`pKn=2#eN|~^x3B*u<Ms0R~(m=(<|<V;>J(Qs7Kqc9afYOd;2F`
zR*wAv%`Uze8-5Ji{5a=Kd`z6wd$91<R9J#^1026?<s829Fds*LvtOP>&yreM-#MY*
zrYe{J8q1q!S+Mi3xQRMj^j#1^UuycpRol&>CkQ|oyckf<E|ENCGBm3<IN1HT`OI3j
zO;>-(l2BznhDJG^jNvDHj)(!1vQrYCi8{CDz2|oje+P;bFGQdmJd}vN!}l0fzu=CT
zjFM^)NDrKE$efRhQLqBGAS29IGy9Lx(*ia4R-!#W?GL-+PRuo_GJcVATVWuSXhv7I
zT|n6t(ra3v%L9RWz>$AGN<;jeUqM6oh5W?1PXjNb;+=(KyS+Jj%u5^<@W!cSDI{=J
z@BV)@T?16!Z@hoQ(pqiHURuk`u4UV{?ONWJwQSqAwrrcrwtb)fz2~0uo-<ZIJorA}
zkM7o2Aa-Ya;^+c?-^d_;E>$bi(g;wGS59^gfZ%UTHw4<Z@mFrcDW8VAMmlf3?doum
zGjbhpB>hb*KA%?P!o1X$R)2!Jvf7zX3DFvAZ#rK=(QQVx6-n1K;fHkbZR}6N%6x<@
zkG2pcIrR1ni?i3*$rjKW#2GCN;jTc%30+q@#aN;n+Q^#gFI|1xC3qBQ0H-RP^4^`>
z4&C9fgTo0y<`!mGg&$2(>fi@LtH1F~?qA|NTEJ^QmdH?q*wH@Din?$^j)GQ(qAmiM
z_jK9GdnV_hi?rKA-`O@OQwmSv%6>^q)gpfUS@sNS0rJZ!u%0T8Zo1l8;SB}z#eIi@
zAh2^o*i$%8P8yrEI*We4t+?VTgK5+~!-imT>qcoNb+pbv_i*cw2Aoc9B+OOYW+St;
zFX<E&;_k9lyICY_(PH)ce_>0LnY&#)%RY#{Y>OD!J&$l&d5((t`WUnLplP<;kIF%X
zi-VKJQEl|s%4vV}SyX<afq>v>ER7vdZoKg@FDGaAKxBni16!d|$j>%yEdhn^CiQ?9
z=r0gFNdzp$j)cQXLN7PDR4Kv;)U|0NK>x3$NxRAAXCGFr&CJHZ{U*wdi)Wi1CV-VB
zB*ZO=KH_`bUx7h^fwbK$%{6ZMR{Xk$dM<%JLakX31+v_FnOR$k50Oc0ON@#`TdZ2S
z3}gd{M;tf;qBlb9E{5L}NwK<p%P83;v3&Dcqshh*bi(_yU~GlcLcvU^mZ&w_{$8iE
zhX-h3pbW@*KY$I`+N6c=Pv%>1?z`|hY}G52in482i$aB-;pJ^bFL8uO?1yZr8(!K`
zdw$$+kJmewfNi8qtD)&6(rT@CXZW5PuxwxD;`U}$NCQxuTpY&DY5{W7X8-sSxqF|W
zZ)(W|*i-&Pv)PKty3l|FFhvom+2m5AMsmO4eD4DV+BWiCa7`^uAbcEP0AIek@Vp=*
z<IJm;%zy{AX{x$K^h+eC&Un&?J45#apjtj{TPg%!b;emje63_5zEYoId__t)Dm|z<
zTY7!lYYTN7FW)2l0(jdmY2>x4T@a03Tc=MOk$6q|my3|k0})tU-w2S~sMSc(#Ak+F
z?=O^Dn!)EukS9>%J}^Td=w_#y-{QU@Pl4Ndx7vPdz2^nq^${FEN2a=@N#eZkX&9FX
zISXxvFh`s?SBh#90REFH&~G$O#O)0S0nebaO5Z;m|NDzUtkP%FfA0v2qY$?s`Fx8g
z6Jpbl{Eio8CtMmOCBmbp$JpM@sw>qzTXu!?lBP>ghyWUEb>=;n!L7GmUz|#Uyf`{U
z&V0s0u%St#alB|Yiqdqjn?-&Lm*Lrr3`nxo8>gubHP7m9nqqE&&VxtSnStrxzLHBg
z`y=G6;1SyQXHgsY(3aK9pjjV_xlOK)Tcdw8HkqP~3{6i69)kMk{bHDvJvlV#6?SR|
zJaa36x$dp$$9Al@J5r0n5bV8~h7VH};M1bHn*!iTa-V<-D>~8QXkajA!Ws^h+1RX9
z9<L~dPLjSSd22F>?G(cDQ^NIYT|>@7E)pXo^QU5v6N+X&IEOA`m?G842kcJL!~Bsu
zg$k-~iWH4}i=&Fj4h_*74ubZ5(P<m%!G~b~o7rt>SUTC^OMUT=mmr!|Q2`8nz?1)y
zPME<q%$arqKo1Du(y{;KH~qTDPn!bT7IZr!9#X-EkSK~V;dE|zh*R_RusIsTuD`|~
zpJXQh`T$lxM9<i~NcSBCP<jQ^e}ZrOkO06U(9AbyJ^)CK7}@0b-{&!{#?BqP8pq34
z?qx*SA0jzPjP_v_kqs>+L-&&MfGaMM*SXgVm}U$yn8;?UD?V4uWR1!IAs*Ln*U!z(
zzr2UWjG5?=l?4yebZ6Sd`X>&u<$pf64cViEMh^Id1u8|_X=$JWlaLtvO^u^L1UlKz
zEjM#LJ%|t*T3W)jCV<<Y!L~{iRabtkC@tMMe<KrdV|K!{H>>#hzr;f?59e2}G5%UK
z;Au(;aWca?+Bw+RIhd>#o&r>>Lxog^31Hp=J6ZDKmNIvYG>ELMLR7XLfP4Y>$ev>_
z*lCU}3?=fi4dr4^b)43j)=ZH8(*D_@acoHH2mujFa5;v7c#*scwdHVv9VBr9fy&a?
zG-KmH0kXHfYVFfGK_NT_&$cJqi3X)-ki*blnjdF<*|1$@Y~W&8w&VGL0ZClqULF{>
z7hH;i{}|!bP=!&ARGUa+#xS(7Y%ZOWhsBgrS<=@+G(YFwqES+=qy%~*ytLlIi~iF&
z#3+CFu0L3<4X$>hKtLxPvW#mdo*ixGluYISrJwF=a6+<r6byvKf6ru88>051cKq+O
zx7QLzKYCOU1!(#>P+wY`m*$uDa&)%757fiMW#W8{&?sTeq+w;qYU1RAf<Q%9uz3CE
zG&%EK-CLP1@z^Ep8*2@6pQ?IxMgJj@S8Y$)u{G^PmoS%O3A~3`ig!=6Ii0QMM2eIs
z(FXyC`>c%(w-bcKKmgd^&PuZbh%o^4`^N7#V{r{OVkkX<d9mg?jFrV=^%&pz#$SL`
zcXd2NAby?s()Lq>72X#JK`|0y5vhKGLFs5BmBeXYs!>*6j`?{1y~Fbv^9>aS5_J@&
zL-)8Jkmd-?_X7rW4rk>;DiC=v1~c0XBOvuO>Yfr7jOu`qQDlw-_0qmasEIe5Fa!qc
z8{fV$9bi=l^eh2}LQj3?6$fU3mI!v1m8RRT|LIwdKu92v^CT*-%fb@|S}tVc>=fUw
zbzcntk!DOsvr%aGMs)S!dOX3=ac4n<ZneR)s-qyv&8wC*TbU*lyVlk~4Lkmct_w7$
zm(|)H;`#%n5>=jYsEONCg%Q{11Zt3c-Hyl5M(8XGdL)n6C*>I*S<T^YJhr3n>sg0?
zipp5j;-d4m5<0`j@awYM?mUr~95~c~J#r0W?SbztDd~mJlmEpSAc&vv*r7uZ+&p(>
zj`O`9ehZke@HD%W=VdY=H`mR8>t(ZAL4?$FQINjf?LbB^C(H>mpD0CiDM}q#07q*p
zE$Da!EJ}j;2G!w1*MXva@PWMlcgh7=A%LC!j?cf5N$v;_Sh8Wm)`D|&P;n3BPv#Vq
z@3&8&{@>c#*+let-uACo+cB!C@+Q5Wx-3D0`V-JAl!B>PT|~K1D?mq6PS=y_t2LYk
z_2-um45aSn5;5>ZHv0aN!E@A9YhA3d;qpWiR~-~s!HrZ;*a3TnVz4BT`AIdiC36W@
zC`g?O3<Sf07&4zUpFNl(=pDAcq4`h39rDj+J>-R%*ekypv!wengN`rRNugwpd0GlO
z!uNy>O{|cK;ex=g)L?egAB(YEZ%mKzr-j*TFj{xlhU2@p#UT*A(U=}bJq8t(kM!3q
zD|gC%u|pUjG!10RP)5H++W(Q{hy9;#;uG%op{)>L@;$8G=(#<4pGC0bRe%d8@@mOK
zV_F0CXL%U8NIMAjN1ngGn9|7l8CVg6g|bC*s%omBAYl@lEq|{~qzF2^r%the!vtV1
zBncrMPuqYot5MHd=R4v>#)RDL3^<i)eLBs!UQg5RUeN5NNp9RI5I{^R%lQHKUl}}s
z1|YKbd>^`?0MrpLO+3s85@Bf(-S1}j$Hx{%0RlyqVs2#dig>c)s^fns#5HeChML(j
z4D*P^HznNHBT33lym%=mh*V<NkMgJ5!~C8b&)g$RevaERRh!QoeSEI&Jdv5{;0R9P
zY~=T0*3}x(mLo}iv?$Zajux9?ykjPp?sv|PulW78F0<>|-nsL=;)Bb@vwBkJSoy|0
z4M>7yl+IIz@%``@ftDWO_JT;-hrOckp#}-&UA&ljI3}!zA{V5Rt%~V0_5ceKH%jE9
zCfzzS`>a$4G&q)>z*$GM13Ude4v3z$|A>Jl2(N7C13DdNz01!HC%h0}(4#j@O~j|O
z9g1k{XOi!bhXa(+5+M=@5EDT`NJ&Yd2_t{iG;4(29I17G;aou~$$G(i#~V!d{VOmu
zpQ>L0&DgqP;{Qk-AA^Gmldn1w2064KnMvbvB!rOwiQGDWgO7rWOo$|vhFxzk(!1nJ
zwA%jo``ag=5K4c~s*rKAi`CgtoMZ3m>TGY{>_*<4Izs#&0n}*^Oly;qoxeo3?cDmE
z(E_`of`sBU!UxbK{^D~DkVS{TtI?qOaDXF9Y^WD5VknTKuC5*=G}BKx`c+LTK#8iP
z$5Q%pl-#})+)97Y@cKpEAnqmuNtBsoSFD6FG!LDo6Q94qu+caNz7ByoFQ7Wm&<?wE
zL%Hw~MpC>#ql#PXTpsT3&IRS}Arxcqbgg|(SSdbZFf2rE_x+h0wBPUa^rZCkq6`kd
zZ)dmHAXQTM*3y~tNZC&T1)sP4#o#rPQ$>kZqyp)A8I-1g6W533=^70@ehf(siEJwd
z)-S6aFaAvl3E!8n0LrsuLw&Xar4}L45;EGEo2Dk7zrf!|BX!X~n^1u{3SW#E#pQ@=
zYiq|XEaD?=0TYP}^b3}jvRDtk6(~Iawd-9T9-h8JB?!$O8l4p>o3Lx|)zo_XOIFA=
z0>y<fDYd{9J}CIzPaA71tL|;YuT8*J#-*MnNoo{JN$&?N<l)p7h?A!FnQ`BN+y985
zMQ8h(wV62VgNcWRCMoI2+W~FZRL;Ux=~>AK%p_}Z!@ZT?b#*R>0jd=>b=AaQGBi8`
z(EXpV0DfDz(p73CaRY#q=f>@)kS9>ko}d32BBxHCJyo{N$YJtc=+CA)&1V$&JmVvc
zjRK{jGdJ$g0RRdDfnv^tUfY8)-&_?3woZpvMq0{Tz4H)I2Z1bWbysjqJ!jVf{H!4r
z+dJ$81c)=S_veqnvnug`!Q5deP6k#7VB-&N%<r2TE#@I#rvbwNnZbXk=nLRHO2Iu3
zxd*D?1q-PVmr^^{`};Q`hJSvqG_B#k=VVFM)0jH)7w!Yz6g|ZoTm&kAfdb8{g=Gza
zeglSJSOIkoyp|4=ItHeDhdKse9(Af`ON}XM2?*67TSgs;>et5szCxjw0{x)}9}BP?
zpV-g=O&8b;;kmfH;*b_`t6p;5qZEpo@tt8J`z-4g9kD9l>MZ0%9mXr534%*_{qa>S
znPIsJm8~Ajp8@p>cq`%Z7>I#b#AW{>OP)|n<usW<t)yB`0!U&?3h;zxq9wf)wDQyv
zxahm<?xCJd<FYTLs|vhs|G}#X7bQ;SFW?vN3cjYD`9N&yOgYmuNvZk6gcx6%05crK
zzj<d#q+kjJ71I#RAYKnA<h`mBgjdv|LWDkNs94TWYtnNjJ5?hx{plG-G{h*Aq~BkX
z#rY5h`IcR*PQo62+R*~#EA<J+S0LxE8l$7{ge%o4yUj3#RNIxQ2>Eg)Lwe7i9iJ-s
z7lpaA#+}hW7>K2gEoX*(9tb;G<Fc7Ps67$IJ)HUd6o@OqH}^h?lHlmcqtMZVrixDc
ze5QdgN(|UJCaOXEYLr3waz+~iqIO^`Bf4RZbBar92fF{h88g^5Gp(-I)Ztd!blVJs
zSfL@$w?P_Z&=As`G3a3QMrXDt%N44##k?3X9^A{Fl{Ak?iJ|#NSKMmyihH*grFXe-
z0wK{rk9kAHYh!KP<huJAi=wZH32RTZ^#@;v9go}9(eV8c-Dt$2Ds5ejF;M-~&w9fK
zir1}f31s-I0s2(bPe%JE5*?Q0A-u(M{K-tdcrkK$D82nq7S$RX<V(WFrWM?%?8R51
zaaBg<N2CtapB5`qWue|GYU&1YTL!0_BWf9@e!TuS31~PeqGl4aRc(8e1il(2pFfB_
z)ToumqlOXoBq$5HEX|G8?4Crzt9GpB>V847>o0^4?9$Iu?WclU!zNW7s<`1<8E66>
z?!&JEl(nP4!;P{&TP--2q)Ei@g8$$v7E^(!VD@!MVAKzvw`&%8Q6JQOu4Zsq;_uQX
z`qszwe}b+qZ!mjy#KBa_(%nRt<j`Q(E?v2#3M{uy&Wtx^r|gR4OKrLy8crFLQ5&uS
zY3MU(d=07EZe84gD)MBqsA~m2@)n<6gDHZH?AW4A1(>K_i>~cOR+vM+EzO+Xh>&&!
zsE!f&T+*>Ja9kfh&iBdcf-)zA$NQhCm8}ovi?wb+!>IxA`l&R$g?bh;G|XWc!fImY
zI)4N0p_u}tM7-WN(L!V*x*Wg>p$cpx*-HT^M_f*Jul!wo=CzB2+JY1jcs)CI>d4;#
z{*e4mbE4c3Xv^E*wdL5ZXtx+dX_QgY(52%O>Syu-Qb>obwynO6iaPoC%aflkc~epF
zfL*#!EL;PeJ_fcfo&!lL)>_{}{v)Gvcqst(Vhle$wJE)39h5a!PDN___hS`1xlfS!
z%EX~rtDsO>Mmc7Hk+a*1gqlbXF5m;v1EBP3HRHa6fFt4<N`~j|OOA`RMoLNw&=5|U
z(Ec9ShUI<%qjSf^1jrB-M$+LRHM{gf10(Rh00}9O!IdP5ij`2VfXd+5{Va>kRH#c&
z66|z5DGqT1e3Tn)79Jw!b21co@bx9Lw(d_?Z*IgQZL-k%A`BcGX+D$E@;HtJSexZ+
zZywiuU-)5B^-Jv27v;fsnl1)g5xPCI?@B(<zJhp?3DM(;O;gbIhk<yw+BAFJM&<lm
zsr>s){&D|)NInRr0+iPvCj!WQcXoEN4-WFE=iXBNYWwbFJ;&NVJUzXzV88a)5G+^K
zHFY=~Hn@*V$+0)Ye?v)4aH0kZXN--H#Mdv!$w8;9LpIVzn{%>IsZN8**aX%f<?`}c
zy~<4|PE3d)zkiFdHW}M&1yHO}P+WwdC#I*nM|$p0e)e^oas<&4a5Y{I`<HmXYkR!i
zCypD8NaGHDGG2shNJisL8p2Xj*8u*@01%a@ihFl7X9@T|FahDhL`c9d3p8qh<423j
zCg_Zvh9AZQr74r*RQhwEVTB7gCfm-|{$}L|vg`6u89;<2<N#e1P@A4ER8@J+;g;+u
zOYYWi0N;8Ns84^^T<<sBK`je*6B7CBUhB1Ud5Wl@A*>N_C+in?^)x%@Q&_(@&CCLS
z+Z&1?qo#z}a8+7TJEX1D+6aR@s|;JgB&nN&>lW}FKzb@*Gy&f7&iV;pybKvlMhc}9
zPIpA+`QK+oZIXX1DhZu>P?mJ!9N(`ss-6H?%I_7@B%$wayB2px{%E}s79al=OqgNP
zXrf_dI6a-601cjjH$(UnE=X5tKmEs0H%;UW6*#nM=MJDTi8v~976M7#Dza?%DNxz{
zH@%9cK@-P`{9$V$6s{WZYSgR91jvd(xV%BI(dKL#P@+TZ%*Jkh<)Q;=F2NqqCO4eS
zhjL>9@&{X~x`nkhd2zW312))>zb%JA-m_ZF1FhXWetBBi$gxKx3L%Hzj&=bGG3dB$
zUfcnm|8naO<--PSlqj)0g;^Ti`9DKI%v+rp25c+1J??QKz#V3`OzUWWeRT8u_}HrI
z%0wC$NOMjdro+ckh)wc8Gd39v4kH8$0rf87YBLZ}yy)<7#)unnu2_IxXHeS`5Da=>
ze<ZDrlKNazTdUS!E^45F1PKE+0(ZY(a%D?KN)lZ`UK}GgCLqj#ge{7g0YaN{)$$c~
zw<|J<3|@CFKsPhqwi1PZOldeJ;<4Wui@5NC5`z9av3g@Vw^86Hk#+ar=93h$Om62h
znSTB!Y7_8q4MkDWkV>*`S9shXI)7DV9P<O-|K`USK#V8UK17G*{IkZ!LrT4K{0LST
z^SZ?(N34Qe<5JD0MlZ3ioO^Fkkc`I5K-UuSd*J<-KZ+I_B47l8Zlsvsk9IlTj_2QT
zX-o!tpLB^%t8xM>UaW_@`o@6B7igoAtwexG_+zvJAl>ZDjdRfvBN7NbtEcrRth{Wy
zvm+hHe(wJ#eveG`th}FM5tk@ab&>b@X_o-z(?$@)w~5;mtKG>K!k+4uouk+f8C!ju
zQ4w2$0VT!PC5y91E@Xt(C*J?w$X%mZcd%6@!QyVhi_TQXNB-%%IklWctnCo5&t2XH
zlU#<IS~KY>9AtdYqSj##be~i$@c}Cd^8R3XwP%@w@t*m+P{EWwI<dSun+hy#tmPIs
z%jnBG+aZag8GW{ki8@@U$xcx3Ohu6)mcoTWhJK?THE2|b5F3;o44lHq$v%UEx0nUj
z$wBA#&o>A-q->%o%Gp5ltgK)mMdnGsLtspI)-J+=8P`d~e^f75m#JHXZ5k;CsH~xa
zg|;r=0Go&ilpDZO&oc(PuDzv7s3#n0Sh!%Nhr4}AQc{PxaTK73f0x0TrFaVkNOA!g
z0y-b>@BIPp2wEj$y6rmcCZq;dz~K@C0Rj%7;zvnPl~aPvc>4U0o>uFN;99_HIh_m;
zP3}>nlu^KG9PVs9J#8NXLa%%mJat@gEnj*oE+lzHck~=Z#yTURySsWbp*95Kd%Im_
zpW4lO!(acIwdjWvWN!nwo($XDE)(D7;GSsWK!0;g$!~zud0}k(so0-4Y&p<cOS}q@
zAf45&kM`=S2{N<jCARMv{Dtfr>_*CledO_(ve$oOkux}(m*yn>q#VfAAorI4I5p7p
z9J(<kYoEuxH3Pt8AFAMo#xo921s3o0o{U(7`N*eYfg}wXQu)X4SFpaO!u0D?%MHx+
zc3-QgKP03{c5vtHqi-d>fpCt$S!y6#5AK;Ew@AzPiz^T*S-|r7#b4+0Xq10To4KE(
zQi36o*G^Jr?(3uCV&iGRs1Qw(5u6_m)4oK^&OJIjeDxA8=?1N%i{rOy2F!2tzV3le
zl`m&v0Z4geQ%yKpZUu`DmTAHC74nf%q)5(p_hyjB_?hf;po>+3Qn3MFFU&U}6#loQ
zKpQhK5nt>jhUe(W)NL6-HZnUKAoZC#ZV(`}z;Xx7jLHH)O9KFj8VT_ExV3XNc5DQs
zG+^|R8CBsoSfW%6v^zyC$?a{;b?<Cgye=BO1$zM-4agyLpQp{33r1gvJX8r(lyB<h
zj-lzanz@t|ev4GWX!f0)RB6h8z(Z@ngX!tc%;j%xX=&;vxAF9`VqfM*>{|<CT07(A
z<xN0)JmNL&@F;WT!SK~vE?obTxy6U$!zQ|UzVo~e(_*{g^~KY!C;E+oLMce9U}a+i
zmiTFQt3+rCa_Y**_Y)qS^YYlDOu%hg)t1NIvpxIAQ@u3Evv)*##06)fil*k_B#!43
z7&hS5F`Ue6{39`OFIpMT@$v$Z$GUDL<1)o}EQ7;`BeGAk-YosBC~UYiP7(`viiM}M
z=TudVU}&j`d|xZ4ZMoqJ{N+Xd9sxv|gOYk;fC&o<_@}2N;Z~?MG}XnmSS)m2m&DWR
zG-aiT%BF37?VmPUNR|dEa#Vv74HfC35#h39{Tt_Q4%Lj%?Rw{f@3ON=Zf$iOu(=LE
z-~~7Dn*JbuB!#Q?s4fZu5{Asd$fK<7s!w$_cCBe2k<bpkZi}|&P<{1R%?@|ff&x;i
z-GOJkPWZSu^m=Qp2E)^=v!E{|QEty0`#?{cWqdH0_cp!5^AcF4L9qE;VzazoPdJjp
zmu?ikpc9Q@fuiWx2{7SrQj-*cGy{|X%Brf7i$JSL#9Mg?aHCVc?|^y*BHC?NE5DEV
zk%HXt`F_35+9o^|`V4OU<)=K$&EfL(|K|c!r1F`Y+l1UzpLX`qXI2+K44(_4C~^k=
zcT$<0Q~-HyfRYNHLI#KRMK~8Kg4<Kn&z2S4c46|BCg-l%aLa~Mj(O>QH}3TCy!(*4
z6yRAD8;4G~()4iBwOe~7^&jO&y;{#@ey)%K&2*NHIn$;K9gVBe^|4BEm(TqXTq_er
zstS#|-RyPoF2yN`FVuhOt)InIqPorjp;O)DWE@gSq}0@x8|T*RqO!PrCVTIf)Grg}
z*qFcc5An2nu(0w{a+4INESIY{ZGnuRemI8%vOuWk>*igvBdb#2PG>TzJG$+Vmq#{N
zrgJlUsmS)C0(_FCq^w4><LEP%p?vd7Sle&iXvhb^TZV&qkWQ-G+skUJpN$8{S%((a
z5s)Xx_qAJpxW6!wy&F57$KCBE6=$?>4DwCV(o#p~1=yLTH+L*n-O2%iNAaJ6lL}Lr
zj4*p<G=g5wyQ4=6gk+n4drtTePu|w$`bUUJnp5NqH71{vqbTfbZ15&B+?6Q)JD?*l
zuw?OBSnL%U1-@UvZ1e!@{qp1_<M3*EbG^%Pei|@oU&^lo*(Vxp`gjmEb!3hI>-D}3
zl^|39IH#peMO{J&oHA+2n@s!ie~wagz`}yI%Un$#)>wxB2b{H@hnb;$r-#~wP)*II
z_?>)^fH#uZP+D0yKe-(CS8rIOvbCtl{ckD`-fa1||E?&(NE`$&^iZ>bVhvbCNE7ni
ztfRSJ>021<*Mm4|-i%X+qkq-Hj^K<04S9629vB|x*P#H}6@0BeyIVMK-8^5wps}!6
z0NIl|b>5crhs(W_o~zL_fR_axw04sn(!kKWng%>Rm+6bT6r@mzTr%}rTkR0Zy{y1j
zK|h$L*J5mGwml2_p_NT*GdEaBW+LS+2)hCiOiD^lujRCVvvf0>J0_>)G?{Hxz}wiH
z>tDJh0!s}pUY$C33(qYA0@6ymyEgshSLf)TyP~#h^nNV_X?S>en;)Y=7_<Bp5`pyy
zNYkk)p96mcv_>4{?>0Jrf#eKpI8EQsnD9ON!}PPVh>DA{a$Rbvl(;$0h5~RqI^ngu
zy#HusZlj{Di1r?xik7^nAm<I_F)hN89p5Ui!TJ^oQdz|arkoaA>d?lU8lSc1K5!G?
z<i4N;#>7RP<V_7ms<iwFdXmWLQ8lz(;=#49FxXAML4w?99)H>h=Alao{IcXtiJ);<
zw=JD-DrmXc`octE8F{+%-8mmOTx>*t_qc8`pE7GgSAngF(OKs;3Gy#x*XN;x8j`DO
z3jOC0d8{-UAJWMApv6{aL}utKP>a1<6`ePf`R<jlhNoc-dH1&>kz;FL%d5&jk-$Fb
z6gC=8S3wayy^!`u@xaaKvjJHdk!E`3gvTzT2L%nDSrz>8Asggt6wB2JD*MNSk6P%2
zO4!AuSxG8sO@BQe#XSn*L_#A2GsMeA<r+&RyV0hTEbg8CzXW!ql)FfJd$ZiAI4i#U
zDb$g%^;t3jr-Zk<FJH(=)Inl#Ry#t9fdMk#;Xw@nIX@nUbbU<as(|#&SmQ8AO(h)$
z@;<(kV>mrs`GddL)X?}%+GVRq@t<d`J4i?LADVlH5p2|=Th!RCcVzzD3ulL_)9xMl
zbfFZBkw9maBP|_v1qxrEi{+f`?C8E-OUvT145ptVIio4pOUI9q|Gk~TPMR-d_z(~v
zGK(qm(#L^|l=QT9KLq^ZU;n%pesf=Yg-n;gH9uPX*pi!<_n9-y`}t%u|JXws9@oR&
z2g<hzP;Gr9>OVt%i3$;M9(=@0$SlYq$Vhy;E8>zA(9)FLl(p)#w6ui&V@l7*8~Mt@
z`28E-O83?@o-1Lgc2jQ3MFb(o6-YrP*8c#){R)fBmr^$`J<2o69U7O-CPwZa=_l8u
zyg4)@h&}iw!ez3(lC}{cpBY8;d!pcfM0gLLBhp7Gp(zgN?^@q1I8VNOg_6BP3-I8g
zVQn9JaHBbLqT`ALA#F0=`NZYC(xxO%k__|=i47Eg%tv-<!9eH=CB-58E@(-L8?6l$
z*+tKq=)rq%klG;GR~#8eW&PH)7wc>r$AQtQsJb`goZjT@jpx;}t2$9eZli11#!YvX
zd}-HYHk$dnvhYpa;ld3i(l=KHl=L(haN%?KHA@adxD~XAZ4E7+&T;vYS%u1wKe=BU
zP!{c&kFNR)VnEge0u^%Tm+)0iv*#o44rg8YjSVSfjP2#tCzcLdRYm3U@^YS9+rS_M
z!-K8!^bAk#Rkzvd!{p>-fld-5V`Kc@o7P)j+(S)zJus(%Da{-;8=vP5NL$kj3C6&T
zTW)={+wW`w+!B$4l76+-{dsdF|5Copwaz9cg5C)3!|yZK7bCzaC3g#5{zFvr!}n=K
zz!I%-%$TxIc&kfls4&F@)-teginWxL=#1!%V{B~$Z0?4TTOv>Hx1@xS(ePuW&ZbBt
z0MZ-#@W#C0Bh`^{gWh<9jww(St+%QtRIIF4J9d=S37fg)e*2WjVEr+=<>0^b{Kz9X
z84dXT%{s6s=4f({Q=XZP|5=&20#u%CC)~f|;?%#I_C~s#b`XIYOHo0C5F4M>dJxt%
zvSFSdk2(|><xCW!<HkiN#c9i{TU#0ZJ3f3pI1WsjqA2-74P=AZcn8!}gB|X-AArgM
z4g(!CKR+*&3E*=}b3T=6EE{IYDVl0D!Ot>%pu*zs@5rY{W@ddlx_qO3b8+3$^VY<$
zAX!rb^6#Q<X^&>(i3aW+&^U@>vVo@Y^%5#Ui#zmRt})plo2R{PVC;h&|4WGN1QG1x
z?Yhu!Jlok7!$wR1uU#W4CC}98{N5AlPn)PLRAzZOFDVlq#-*)qUX!|xH^0Vr5zU^5
zH}e`vq&q*l%&V)vhRRCGNRA1WJoAO`>|k<+yX<sgYz6rh&rNp+Ee$@OEx<!yq5lre
zou=KK3dJ8cHQm2^7alh=(|=N1+#Dd0^|W8G))x67?g+ea8`;qyPR_OD>ZPtkZIvfZ
z%L`FrWG<UU#P7d3Sa2pwgP<qi96$@7L=utTl$Rt%rqk$2a1A?cJuDS|h=<!rnmta+
z`{)VY#E$x0iIq#*X+kDQ3nh*)nY^pYX!UAQHLI?v^1NMVVqBr3qC-Rstz>3<K0(5K
zRI;|(n%B_G%{MhNG69YRDBlG@FYrg|@_Wsa-)BJ=ArgJ%%7qhXcb?u77MziNhBb#%
zjEt5GR=gRqS-;&qy)C|fP8cdHE-o)B!$5$YVW}%T)~Z;}mHRR%fuuym(vc~r7T4C)
zl-HV#ltLu;#{2H52AJz9qUH4OseOPmOl&(2an}5&`m<|&UM&d_=8TO@Nb#!bEuj92
z+VV|DWVCv^TIp1j6{RPpJ^(IIB-`n>k@23BhPZiBT+b(`=l5LHL~D?(OJ#Ls9@_z2
z=6f|{#<Ie~uyDz5E4@RzL%HXMEuh{yE{#m*41CZrbK3&ab9q*fp7jJHXk?r#j2#v-
zNHvTe-?s}oRp?FmRZ%g?@qdj%_*=g80#)&Nyrhy;Yn?w1<ZDlN7&2UAs%4Cj*Zsjt
zGDc`!9rKG;*cqskbMJ{V619}H3<}peK-JPa$jyFbDJYl&hS$(`fT)nBX7V#N2QI{Z
z2UV>6sl%@{EG(@q&jO3s5PXm0Qrf7Tmale=tZ68M&oA@*|4E$Bp??S*mC`2ABntvx
z6tYQRsSfg`wlB%Y$Mql}Z1voYp$d@(F_<Bz`7IJPGS~%%c4c{i11v8?lGrEUEoJ?(
zjI)PlGgrT+J*quAME@y9d{>4;w6B}b<D8R%4fVfNKN>o^cgTnU@B#%FP|h}2R$?#U
z(&%!_N+@*leI+A1qV^ZS0VU8CB^p2&|DD5f`}tDLt7{@*V4^~L10ED;Fl(NlZ7d85
zb3f5jksv`#V8ZEBc@G}R<)oyXdZ)!LS#1JW?>WUM$SBOir=s@oHI#pl@QPdFf9yYk
zo%>J_W{kWXz48-DkaioMmNk)0*^~qVlnvq<o0WkIL0jP0!e1zDzC0?L;;`AO5xOSQ
z^<;76!ZAm@wy@m#N$Z{5F1(9Ht7gqR+OvJ@#e#AZa)*cF2YccNPD<$i$O<=NYI}=<
zDBohqX$qh{-<v&+3^B0PSU4OhTaka+AkC0Iv8OGg%sO{jR4n+1CVWj!9frw3X&|eh
zQ^=$Yr8ks$jgcym!TlN?P#xLPhVHAEGd9}!U_h9$o5_I7<5W1PcZERx#*rfcq#@R?
z%#AN%tKk2&F?k&JRGgHnUwYM3k*)_xkiT(0Ty|)MwXgPX)68Z(GGA}cMSpQ>ssCj%
zzC4DNT@Hu&>-SkMOYacM!pVo6tEVc8QKNJ&XV55|GdA9)Eh~F=S~M8cK*n|ZyB+jc
z;yOSHWR1?(7s864IrXSx%JmGa#)<#^fWa{@S{H;THH-nigjEaIh~fh@i09IHDrp=(
z{$1vs@%k{~?@*5xZYG8ra&u!rw6nO&Z~oc%9S;|)i;Hu5D!Qzx{Y0VBNnu9}j4iN3
z1&7Njs4A-0S*`CybuHRdG49&ClNqFJBI43C+H7Y41ODwx3ZS9Lh)$N5QzPrtE7feX
zvY_v5vtDQfaqwM3Y@c|E7lx<Z7Prp9(2x(Vx7K*vc`kT*>XbfI#W}d;J8@{wj}LEG
z0^zD)8q~4Oct{a-0mobbiGYwMu!p7MyAFs+rm-JusHy^lnuKzox~6`5!u+t8kx{{A
zHvaj_3Gz){{Cq14#`T1cuA(rNTc0LP<o5jS?}V8BbNd~Kd@WwpZ9ICsHy`eO7l`PD
zw66y)$4s$CkNGx((sIzXKSE>{x$4d|NF)uD8>QQ8R!wG&h!8kK+!P>{s#VJ9ai{iL
zZw)g`?f)G`f_EhnE24-voralmt6?pO!jE}j0?8y>WPPJfnK%`s<o<yJGS6uSlQ8BL
zb(7!NeszAQscK*w#~gcuojI~#!<Y2!C3aJBpvWQPNl*9B`ez+9**`h&DKoq;=eKG~
znwv~gAHImDq=^b@1D9q8%gb=1D=>9p@+Hq<TfuaQ6~+)AATi_pKpyJ-Fk)EHVQ4P6
zfV@Jf30JnZe~PVxAG|f7ibD^|nERbKg$&BGNu}?33LwX5)io>LJn^GHb18V;U!s|3
zmx^Qwm_={Z<N?8FdIAOrRt81~XelXUqT&VgUl!M-D}Y@ONE<uCQ7n*6@*S9RTDUU#
zEY<|qrKd^V&0(`v8}aQEfRsL*-CY)>qD3n%r@Lo{JL026nrP!mIv$A_(X$#y3pz>;
znc8+-T!cKg)O%mtP*YM0M#zA`F0d4JcwW;{lE=vuRA|;YUHq~SYG4^ZrT<JdXi(Qw
zpM-rdBqq+#BLp~-MLF<*sHdgaMK~_&x?Sx*rrU~wj6zLC1?9^^ognumHmET1;UU0K
z=8;afrDyZJNsBI6D4EMKIW_F61y!f&Ts#En0>nBxU{Aq>hIg%xrm}lpo^%0~;`kW?
zTw1%q{txpQPFB=xgAKr@2`l0m4S$7wQo@AC#cRg544rZT{~10q<9erDRCakRgY)7=
zFmQw-l$_W0##V8OI;wATqEHFaKQa^<EQ!rzF<+N*+Z&Ga;XU?`)@pScd$2T1?sEQ6
zS7}frhJJr`@2T%tbV5uw+U>G=`1-{Ygh(1}e&nj+riuSmiH+1!E+f$TbFz#5G#~(O
zZ@@BwoS4BRHlS{N{!{dCV$6=FA{<S<%loH}kwEZIS*5w3w5qWCoHoJ^fkt}_-?eky
zUWfMuuuTryI7KMV?w>$oPDTO;tOKBI9EPhXF9#TU^Z*%QMr0}%NPDQJdKFKTAAZ3s
zQDe?Z<#UE@0aoQwLcb=&F>}#=je*99f%VAl+YPj6PRf6kWZg*y2Avp7lao7D8|Sl^
z%L84Y>7;~Y{JZFU17l@DCyUi!&gE`=W@b(@DAd&_^fH(UFlU{4yu$i6y)JIryvE8r
z_%p`O`himkeKOyW{a8JC55VW_Ew!2VTX0cs_WUidh!&*0vo7Ow(vzz$AlYCQq^$@W
z8+&_a6&EvJunU49Pcsw@5nx0MYS{s_UJ;SL2z*{La#|So`)3V&4ttmVZ1-L52oR7f
zyXp<7@x9;Q+~gD$CG~9*Cp`M~25>dlnAzBjjt+G3YC?UJQ21y+h%i5{CmtjzDURrx
z(rbg!4#-c5iDqTL+1Yw;&4D&^smWh}*wkoyypRB$BBRgJ+`PwxgEu2`7S)rG2qG6N
z=Y;RRNveV7Aw(JoB=%%v-$8O{R&&0fL}`0FlsHxu8QGOn-D15l1oF?)sNzTD%hV^*
zo(HxEzE^JGF$+G;oLtV)UDM(yYh1kE2f3%3`bYB@f$H{|RqNk2QBWg*F-!Bq1L)7(
zDtw3GR~80KVBH!amMLS}jjP)cDuDr)QxNS#c;BWfjdrEMKJ?Wt=&Do>0XtXKd)MWf
zvJ&o6%{rw!S#(7lVp>urwhFz(wzS|VsFw2b!a~@{zM;XJe{WzQ0gMlXitT7LT8aXj
zEqzeFK@{bTI?aP_<^`L`jDmNVh{Zw$=>4!4Iwi**W%al1i({y#a<tkuwPsBdBv(s1
zJ)=6?kA9L{{4#e<aFZ~VeD0pBDN?e2-|Qw=gJrTakon*R4F0U><t|o6S#(wMRY}`^
z^mDQ(tT~HYWm@`i%9}inX}QxGRF8Q7`<U?Rr%1Lojx%P9u%On8tqY#9?xNgI%x=Zd
zq>j~BREWz1(fiRXI>{yH?xNK>LO9Z&P-;eK`q+3l*tes0a9yXCs!q-!5_w2V`Cdyg
zq9_fgRI!td6Qu3V+#nPV;7CMV4_sQsmwqdq2?SiN6>LdZXc#LOkI6)3JS2YOGiD$g
zX?Y8gfF|<tFVFnxUsjc$)}`NL+U@v(&3+Rw`#<X6-wdo=3kc-uHv`81iZLiYn4$bI
zw2OvGFpzwUWuAIXtD}QhD#bj-K9Tptq09IlA08EjrQAX3a5NTk6S&@|Ph<Hd^)uX-
zx&OMabf7cXAy#eM96Owsx~aL7oBa^fw{oBLk1qSI4FU(6qQd;fN>R8VqAA)D$B}qt
ztFyFK=CF_YIYu~N3rO!uhsGm>`M>eoi3{#qZ00>oK5q_wbc1fE(Z)qQ|7X|u=Ai~_
zZ!Cc5*N;|wOb1QuVk!wuwimUE)uuW}PEpi4%A(-lV8Z1k2I04^n_!D}iTf1wEs@Rg
zp;g_wdc1SZ5TeIAjy;i_R{8y&WSwuWVjn@ipg7y<#z+)3mCN#B`KBAk=>aZ_+8>?(
zB@mCnN-XkWJMCMJJXf10X!H^Ep`+pP)tiR)_&|Gy<OIt~Jy_dlzVBNNG#%_8mJydE
zV#_ytFsV5#Uull?M?&j8Y#;*RoNl}AOYb!gwIc$n4r?4P;>%t}J<0yP1vW>g3Fp$%
zBiN4gtxo*nUL`YwCUe9u2!4Rt>mQ#$=Afw%WYY~j?SpmH%nAByX7G=wW6y{gml?1v
z<#g0suArU1mR?+5&f$EMJ=7%0xtI(Rr0kb#u5Zgcyq^#uVsU49Xc#WqUIii6z`EDk
z{8?QhpDoX14v@ux5&x|SDJBf&$B)1o<m2aQz+>ysYz`SL<_K+PBJS-a2Z0>O4Go~W
z4jw4-883Xau{k^_6d2rPrm8E}X3})y7bA(PL{P!Ni0hL{<Fj8>mKF85kG|(>u->B@
z#e&NXq9Z2eUG_c$``-?&`p@N@UkRyX9}$H)ed(6v-e}hnBnd5oUN<=L^@rPoFNA~P
z-1vU>=EtmUF9D?0?RxfjmBl`P9z-4yHS?T$Nj^IlmKO8cKXew=U^zS2#?vOZm_GJe
zx1but{qzQFDzu+3dubWooZBOuR+~E(<aah^bAEff4*TlYXm41Ei?+7B{rPfFiANF<
zVdUKOG#NfChNQ(v2zk;1M_@!Mw6EXi03MOleT(s`!-H>T#bm(62%109;E@;sl?2H@
zOos%#2Ko&8M2Fc9bT@7p@ha1{dEO^(u<v3cKg1&PoMya+%4ukL9~=qe#&KE?{sIZy
z@w2lGIRVbLo6khBP_m7L9&=o$p`yK-c)~*^3=QIL8&o-Yd>I$7NN)Jvp{FN<`4$~d
z5R!;H%05wXA^qz5uCapUK;El$TR}DZ?i>0?3~UF<cUauA*AU-#`6VwY9$#NA?A%)c
zc^h=e<cnS-d{g)^0vqe%H09O(U!0@i4+&9MmMOvxzruwEh1r{#E)N)8fU{pz3@zl&
z8!>mH5Db~}(&An+w+<1JLo;);x+Q$D8+9*Mk68W_RlO=Q-;dw1CZ6K7cV63;09hC_
zGfFHh)8iRAqR8PvM_(DR4NSwm)zlL~>tklx#Wd)gh2;BmP*(q9h;JXgIy?NErNh;9
zuyrwDJMFQ!aT8+pSBiRuhs{pRqSFx+t3eZX*smacfCJcmZoYA=`>VXh0emSD*LsUJ
zItE&wQHXv^j-d!jXh_6n-RbJ<m*1S|O|DDD14ZWH;i(fji!L))@K-6k#sZP9ICv=E
zSl<{JWIsG#Z!FE2b}T`W<`^31=2cFD+&u)>dR1FKWIO{lgZbYH6&@}^AweNHHhkD?
zbdHARHUXu`?9$R8<Inwl|G*$N4$8N=-ve9WhbjOc;?2m2jedC9G({h)!$T%8urxJ9
zga9DxgwS^0*_j5>9x!E-$+~-lg@qvBWJM*Fg%NqOOE3=(-iVOI?ESBm0VIg2Cjh1X
z!)*A01ww+V5*)05xo(Y?Li9EH*W9TEODyf3k(0krcRp`F1~-~NQ{l!9?q0?yPJ(=-
z=#RMQHv)?MUS5JOHiVF`Nd|9(g+WlFTyGd|KpZDG$)a~93vPiC<NeLdrP(em36%;+
z#_7;mLJ4@*!=qU^1t~qhL`Ar2Pf-htsNmh&<W&`gufx<>K}NFuvYA;z>0xnrv&GcG
zK{Bx6vO0CT>d(5qeg`<{d$NR)eS6ndKfC<$nkXGG<duFaD$C1A$$)LO-^_yx&uSLr
zU^mz2i_lL!?E*G>|83_%7AhYq*vASY2%oa%Hu#7@wM-F-KRz`hqVNg8fy;H*w;8fG
zuSClfYis6JMOk&{&b^OO7J_Z2?NtiQulGi;s=xCk{YNj;%h-6T$v(G|IfS2AXEtSg
zM{6xY_1f#=g9oRD4yK#xki^NUSS$zRg~Ob&)9SnHlN+x)`P%wgKxol_hz;)GByH=d
zu<r{AivSIt-r4&E+vO&wo4cNxK9J2pJ0hKvy*X|}3#qNG|L%R2&f~ZU0GX;KptC*U
z^>n!D)_=`>Pbn^LOy2V>V;wc)J-xF54zwVZjNQQw_2b>Y%lm#F)Mhp6oo%e;=EuD&
z`vQ8Xzf^-E7Y@~yzYG%OC}VR1Z&c={!ED-U9P&SzsR1GK=DcSA!X-_CejyC`3OI8S
z@wrbJ2DEuTv)QWgSOpP;i@ZZZarF4Fz~-*Cg$KhX2~?<{7=-*OVM|~2hJw8RfKg+2
zGE9vnL#+BE?yw&6r&{<D`P~DIZ*A=s5Jyvl6Devd9_~4DlQKQ7b`|W(aDpr<F-dVM
z%6es*$g$n?yIUI_WCWx<x&HrJp?}PMpx4Uglam)grV-I%$JDK4W@%V$M7)TtK)17h
zYHDlh@)OtGC>AD$q?j1K6R%&lR!8Wp%1IQOsf*PL*ihB}?4~3or#iPSZjypvt~Asm
znX!IGJTrV8g3zX2&ju&3MeNe)@EtW$fBF(j+__hfv+Z=m>@s^9b9>Qq2|x!=PngP|
z6sZItHa$Bjg7etY+WOri?)CWbz2fa0Yy6HXU8<CXOiexh)-KU6*j(`Xatt<P2ssje
z$I_5f;$UU``r;cOkI9&Hb#}82LY6txOaS;wR^Gzz?%u;t;00i^qVjT9R%S5?89`(v
zHzOxw<(Z_IKu5<*At50Uci}O;^+Jb>iwl|t3&65MNGa?`YwA)>LJC`HI+NpK4#i0!
zTubwL@M8&H!YB|no@o|+(k3{ry&{zQ{XQSycK`${EDU0gxj(xxhU~Rvg?!b}Y;xm1
zc`1_w@nuf9Ba@w1wZHd8wY}Pgs(u1u327i|)9o1|(x>B>i{s`v$?L~kMTJ<O+x5#)
zT0GWrP50UBgTOIf6-<*`8?7*~QjN<t6gHv{oTH;-YB$;6Zpn+j{*klKPs|Tk<ll{1
z**z#dtP3@bXiDc$QBpE6wD#U#v_=mbc%E6J#Zws)lcek)YJ1%?HrKEBxt%541PbRq
z?a>T77bO)XtL^$lbxlGJcTC5)kce>2X1me4P&z9Y9!UYuf#MUDI5&J=3bt3!qX2{z
zR+XKM)+azmhx~hnrw1+al!P&1A>qRVv%1>a_Q$h(*{#dDAAOqxVpZO}X}niKGA8q{
zwtCha2uq#^f~tB7iK6gC?3{OhDn`W;#-qo(pK>3@19_kD{QB0h4#w0A`F3|>4!Y=J
z!$k;qozBJzE*qfOoc831h%#0|pw@kJ#<cJF)WqQ6fS}3iq(XFbbm`TV)RrFjk>SX2
zUwecuk&xsIzanRokc5O>fR?}5cif+;rD5q2m<2@~RG%?pUNfi6z&zz6(;JBXjcsJ2
zpskHY_#v+(CoVzN_v+Kuw$U5tzpjSd)cr&IX6BWLhlU1dUrBb#9tC&NOy7ovuuLlf
z-=b#XH@_xyWmQEqs!LlLDf4Lm5T?l}Dkyz}u7zZl-n~?X{XZ8VcYJzUOGiF<C7Umk
zNKvY`K0YlzYG!g$faC)U3n!OeLW1h2V?xO9=;-N*Ni%EfkrC=|=VP(x7@S2uj!jyn
z_S@S=&UGvtAh2|zsxwqy8N=9Fd0T~!?LUfME8f3zHvr3krF(h<+bB0rjO^ni(lM+#
z<p9*P6COuuVWQiWS$<+-BRdlty{?8tv=|v?LM#<kzazUePx1P=IkVu;7ig&1*m!_u
zfGP^f{|Ct@65Kq+$B>X9Y`lGNSikEws{ftc=UU~kFQAM;UXVivx?vpB(%^EiSpbJX
z@Xux_kW#ftN<`<a=~gu5v2e1ME1IgPDy};;b$W@2YNEaNeTl5Av&#J&@!%Uv2K+FI
zXe~bgmxbUc;v}gDTpZF-bQt{d+BPTL15nh5fS#c^rTNO?(I<eDgSw22q$;<RJ8wk-
znrn3Xf`lXoI5x85%NQ%yW&iM0R8$1NX->9qcNg{&{Fz?vqM{N}Scm{JnrW!w6JoX<
znzk|GKxXj_3yYNt0aPR#6AST2L~!+>`2NnR0n&zeNZ&fwG&et-)>kyLPi+pY>^sIQ
zDXBPf<`v{9D-NA}Ch-uZ7892cf6+gu(fY<iPzGFbS2f=hmlPFJQP<Vhe=f*ZZQx4N
zZrQsyud&s$O9WRTZ;wcY^&>dSF9?AYL|w&?fCutH|LfPcG|nu_%Emn!F4|IBKfX>;
zkkZcR$`&-`ku!3ZPsu&dA-J_hMYSXRj7re`gs)JRkseVSYb0Uu9i&fZ|40q1b<*-A
z@fY}z{uQ)2H8nLSXtD}Q($ZoA(ft8mfrNx4(G2AP&W)S*Gi#Qj6Rmjsukr}Ygds4j
z_IN*rgQHWz(6(JNTr#$bvU1<s9)5U;HSIfL_ojp7MV;RH1D@`lq*ER4gPkAwJ3AgW
z6&_)-3p}8@F_FP}>=?Pv_>*!=6q&{u=?wx|jYfyfJ2*wX_5_Lko5WXHsIRnIjh<b{
z3zAolTOUOAITp*$8XTBkJI-gH*BKHn!D0yZWv{!T|HhItOI+Gxs<*Sqi{PiIpun5c
z(!#>0UFtY3@Dvvw-y$ik1^VE=5JvQ7c0`2x4~*rW@%`ls%;DGgXLiajB{jV`sikJ~
z>Jys0xCl(-Kn<yRSaA_uQeg|2P1y=xzMv1eg1)YjjyxqjH7NzfBbM<dMO1YB=sGiK
ziIAes@?Jr$(;qYgRZ^U?GSY>En%>_I=midUc?NqR@d0}2{>+YmW?UC4I^tl!P(Lkr
z6nxltbzx>^&CfGc=-={*9niP(B)N&6nV5VE<SH>S-kq&p080x`fZ9VvR%0&u5<Z40
zBdC};wdHhl<mHrxqu*mitUauCXZ$mJd6QM}B7ApvZT)yA`<Fh}<ptGX>x`I-)fVhh
z&h7R)s=Ilrdnt%7L*U_rT<`=+Jj<^42iE{55)?4OHMg+fb-v1s^IwoHc<}1(B|mX-
zV#PD_Jc+#=E4a}%+TAHzs#TO$z6Xwh+S-8klmgV5^t#-(6Hi1a_vak~GB!FQCN8S>
z3^J)M($YhXb`OPisii(B?II!|g4R&*IYSoQI3kk(S<C>)l5HzM@aljq{z9$K>D765
zaXmLFr>LmBZ>;c&`CpXB{Sezw!muv0?d<E<g+AF8-Hyjbo0-W4ru-LSj@Pq=-LtBx
zew*Z_MQ|{JUH_xZbwg-SbC}N8O#UjHC;f`FwT)E{u2)=sec<`9gOi?MzS@uvv-4t6
zI05SU>etwR6HG+~)bURTR`u%-$B!?5#K2MGh~b6_at9v(YWB?Wx!2S4mz=q>7V8up
z!N7S!m*ts#=nmh1HRY;(5xDKHyF0pR3qC{-?8&b<-YG4$Z0QJIcRs$Fw}cyZe0dEk
zfkd6Ar}X@P*ZIh#Ufx#e580qow4=UCpSgK*;1p%Z?!LYQoQXA`E0UN2z}*73zkcE7
zLm6sbxc0hCY05Csp_NqkC;);ZS4VpV_+L~5Zkun<%*EV<RPje<*XEcV*wya*eZb*y
z)vt?!W_aX<0)GT--ks@{=t#yh)&3v_^x(V~oWX%{Y`I4LnPx;jfa50i?|Pl7UtjH+
zfk$*hP1RGK_U+jE-_2cUKfTADYY+hU0fDp6gy*{*YC@kN!~2IbOZ(ksYF|B&4DBzx
zCJ)zWcQ0aSxCy4tKF|D7GmuBnByw%?bxu678gBr!!s`iIzI)P^VSZLZTRLy-Wp6Ai
zGUq-6D(BbZ*(+h<i#*W_>yst|E@Y34bvp1Bw@?YGj)(i+2VaYj{tG)YZR9?)g^dos
zjSdt<O(QWY5*4L@@(WpMYqkQIq@#-pQl;T)fK53xLM;c*4=M_0=WoAM6vqC3PsCJB
z`=I0K=%{C>X0G0pOExsviJW5yl<t6@GH$!xe*)&AzHUyTQgKhFU)K5u5P7$#feUt@
zszwuYoFw~maWHK`jU?vpRavGSnb);x@5#G1##dO|+Sx6pD=UgeAS)(OWHmQz@w`d(
zAr0iZZJ)b=3LTtE+T?Xmkx>9@$~ub)DpU}34mf?w!?2DU{Ug+OxHu!*4&lhxkNfr8
zS*@)k3>++9$WlQU7!d&uf{17jkj_0TyFFDJLvEcwO<iygW^O=5RnN>OIo2N%5)7_H
z@W=}6$B=KI-jfyN0)O<)b9U-beu)qy$oTj?_6MAtlCq@4R3r#!5d%P=q_h~9$%6gj
z%lhR#h`<BKFQbnGvy0G|j={m<6~=Q(*m#u-*Zjo#$edpDa$KC@z!%3HN51NurS1r@
z*<#(L-?3aZW`LBa4s&yh?jDL}k5&-B(tDQW5>2CGXXa)}J-2N2DO^PO`Ii{kqFNHn
zHz%uBA3X$!k@0S}6*lr-pMf~!8z_A<eBMI<oSvfyAW|t4&2N(co$$>ZKS;(E7mqKi
zd-GEU-ghTHe!LPvsbT%Y1WuBAGOopr35-YBzm&vO1$p_Ij*Rle!hTWuj)#ZuV6PiS
zE~}3xz{c+_FHuyAZKz_#I+3ZXI|APmH1#oz$(a-l_2DxHi;MsKI}<?>6_gfcNA)>s
z)+$AX5A=TXKQw)1RF!Mj?t?*#h=59$Af3{sAfU9Av<OH^cOyzkgLH?4AYIaelypl=
zNOw19y5H~Y!H+$L3s&7R=T#Gt#^PiOFDg`owQzcUzi%}U(j3>#3~Z&>ygaHw?xu$U
zD%i7Aw8z}sI2Gla?y+*<vV6=zOUDx`N`cMsSW!`ReJ6VPw-a|w0mLOEAZIeg*EW~s
z<5LI!V*b;{rAq5OP|krQ1cWiryDp;xzQ#{Wwg$>V5q@rAX6nGFv}0qh4mgtO{RS8I
z9;RkxML$h@6`w#)|4To+O3MOdn2>CW`aw_Rw??ewqRn@Z8o{)JH|googjr2y8n7YQ
zc+1jDUi{k-t$0%)n+5)U+o?7w*ic66sm%+GY#iK_*u;qVg4{I7Hz)uBI^^_%H1@>{
zOmdO4-IH_0=8a;E2gvR&Rx@>al~zIc52UN@uIv!#S|TFw;4wDeCn+69e&2^8ZuX8M
zh?YUW0aJ3-b2CU4%h7uW(BTEq=dM`q1i+uoxdS=`&}_hY78#*r!WI8rJ>jl3Knw<U
zOkl+=uUNLmZFxxZCCd;&mkGjVSLNp^q@d%GmsL@h&^T>2N{Nn+rmRqn_0$#9&zOlx
z1)NJy?49d_w*uV4?66UKmzEX@QoD|}vKg5lJ|Mdz+c*z$E(~3gyD^QkTxsto`}*$V
zl4xjaQ&aYol~=6O+#e?ZETvx8dv$FMUQSZEPnq3BPnhRxe^H8}lKbMck<gw|r@!ed
zyN+Bwp}bb#I5JK}#Q<YCBpVu%pPWgfAVhqLf5Nsw^YNU}to^mwyCRTdKv}zDYWCv!
zbC3O9!_qZ;f`xuYIF*!TWcF)njiSTY)+!&uJp^inF?)O1nu~8~@n5vjDXFN~zm{?m
z6Z;VV^%ERDQt_qfB@FEJ&oo}VRFmxN9EFVSUtL45B=bI?GZyAxV`9u(#dAF#0?sWp
zwd2(c4o0A{g~>{GhM_^u6)R@^WhXBY{eKm?BHd*IWe+|l=i{B-liiJ67>N7PS8`-{
z_$U_rbdnM;z~24<zCKS+`pb=?(&Da8)}P8!t{=aCD*04Gb?UPDU!mp6guif_Zc&+5
z{>8phKcQJsmYMkM;`GEFGcYPX8ZNHQQ(siPfZyZ4z~L%+d*g}RIF!?%tPLWvm}#E{
zS%CD@Yt~%Q9GUTK<mcwXqqJDm66qIh&T;1?y~P@mYn*NgJ^ekJTTri8nZfzSnM}ky
zFr`Yr(>DMo1Ep~<<7jr?{QV&yP=~XTw*$oY&yK&g4M5ud)LGJX?}_Ob0F#IY*7;fV
z_K%Yx{3raRBJQ)GXTrw>8me^69uAC{p5*17z;n3CVc$(~f2*^(V0pQsQ|A`$TOBV-
z!Mo;i05t_RDx~#;<zK(4N>QNV9jK=<I&xialRT-+FeNuIbvrcB^x;k6{SeTm%j>S3
zJ~bL3N<(HPJT^XF?RwTa&vGx!rlB?wYG!3=={1YCX4sc=b6q1NBcMwSYf>~cJg|^7
zg0mc|KKp}%@P2EL%O~aU9PDK`(w$3*h=_T%ey4=RGIc~4zgtMOt*h4WMD;7CM)}j4
z<kw%k^AB!!J;YPoCKo=_;&sROYlg`Uo8Ok`5dQ#a_(8l5@?yGA&40-sO6ZHZOWT&F
zeK4OCnuhIsdMfdpaIH@<2D*-=P>X)!Aq|Hqcr751yIq)qH2yWKytA|4F|id{Nsc56
zZyNlFK)e8<q2}I(zyDb!{{Zk>%Ng68FJF9QsGtKG$QdflPCPhU11s(4W8mQ&$%UV&
zkMEwX=0>(UrkIOd{Q>@WsrQ7Li-|*s6AO`%4eA@{85L>oU*Sch;AoGIHp$7;;q%0N
z?s|;ovo7ES;vZ1hk%eZb6}U|QeYCsna&Xw(!C4?q_2j8mZGLUjYte>17Re3+JVemx
z9Q*87lUtI%_#8-+6_}*YesCiSI&coP3F8GEpPpPF64{z`%uvkbBKr2V8ZgV?!OJAj
zw|D+6(5ie4FDSH4x5imlOuL+?`{&XgNT27(?^7`_G|h7DIMrTvt*|%5@||4`u8G|~
zU)eKHw+Gb+5glFUUz_4MG(<u3x~P)U9(;iONr#&)`pV}=<h3wy4QiTSt-opu>OJO1
zgdo%PWki1UsV{u_hZ~cEHdXFNW9`3sri$zx?d^lBcuUjLX3ty9h(`;>WwkU^>mBAp
z<|1CIKUn$obrm%0$^0&TKiiSvrSaei5Qq0AN_=@4LNaN_ZcKsDWM^H~+dGPhQLjSy
z`%dhju(T~FBz9nyJN%6ZRI^OWW(9rQO8R$X8!QO;DfppnM;wA$ZclfAP}4!30eS$)
z*mf421}_}hOdTyPEdwrkk9O@LJo;U-^?d;wNKLfnO{~;y{8kf68%$12a5s=}0oY*=
zx1+DTksV8WknBxzpTu`OymyKGcj>pb9DeAw9$CDzeooKu=eQUQrDw32_{oH!h1L=N
z9@t(2^fJp5i@GKb$m`aHyCaRXgr=Q1?(dJNT1ZQGLec4wz>k5PU;j`zSta1Ho&!KY
z+5PX%M`;sNW=zpH{07UMT6{qe3bW<*HpTeHbof|js`e-t_;C=oh8w3I7&p8C1AQ@}
zG39yAMMrlPe0F^{FtjG|o}NE21}&|0RK2W2Kk+eJ{=U*|TShJ32>`2ZzNK+k7_L$2
z8n|bmW`C<(2jdZ-YTMEkrj2}V_?G~5=7xrux>Zvs2tX0R!6dLFMcj6u+~EuFe}KT?
zns?*%mBr!Ibc{+Gsy)?I9>c>o!J_36T<fqm9b^(25upS^b3{~4GOx`XyH{S=yk7$a
z&ZwJy9#b1m_v`k*YZb(Q-F1$k_KyM4;T||DFm4D+KZU8nwJFIxEu3wS@Ly=SELun|
zJ;`e2B_Y$a(`|4f%4Mo^Ih!3C3v?4s@oL*Uc7tOjJ5(S|&tvIw9+*IdTzu3d9d#9R
zBXj44zGOK{^~>U8!|#Ck{I0OxV0A8wh?uL_o(XsdQ_n%=G)*ir3<yq&NG+f3>B5f+
zRcH=QE4dD5um#CP97pyGdJ0}~c!uR#XMgx${;fJ9ipapU_D#pRu6EYHKHPm72e#?+
z*_kEk?R&r|Ld2^!zvDG1_A-TpCi_G%tgYXR{&6~Xau@qYGaq|wTL~8>bpECCMi3C}
zqx_M6Eu}0o3ICKRX!E))`rwglwtlvmsi_)^_VQ7`{(LpT<MMYgrRLHl&+oRnxKwRS
z^^a|ln&%FWlrS<d>caZck%7@THxZ!DkJ5;dcUqMLS9@j)Uw#u$5dI5@S5yddopms(
zsv|`N?Oa1bwx+&AXqoxkF?&)0D{G<C#o-6v>aW>t^GB@Y05w3a+5xT21H3nao-4l5
zS*PJ2T;+HxJ8q%+d=&9Kkr2~9=-xvsOqn8oYk9PIL1(c0+RV`7YVjF@OK|6j{ifUN
z+tGjKYt-P)*B{tGdNjaH*WUjb8k0W!U@pfiKxRAcml=~%?fUWKGwh0Tmy7kwPCCc4
zl@+ae4?3PgFVV*9qM!zvA}&C1$?p8UJmwCU3JM88^U`!A2g9Mf-TjNzv5TcpOL_)|
zM1I%euV0A}7#N5+dF(m}*{=WcQ34McKR1~*74hh|$)dM>2zF_uSX>H6$Lx};KHxwx
zg(4a?0K+n1KHxn^*+|GYk|Rmx@?iQW+s86&9?A^vs<u1)2}(BPHMEM$Ls0y}HDPl^
z%0DtNU}HRkGZRV%c=aOJV^TX|dR%n>8&vR3jxOCJth1^qHNIp*s-PKV)x63*Thrin
zJvs~qUPkn2WO$SuK>g;~&6(Q$xt$I3gb~0obZYF${}oi%{p6vpETy}B=DvSl$mSJz
z0^oo`QmZoef39G05gm**Yl7EDyWUC1?agimow>Fnc*9`Ov472a&TSHl$=Gn@&)>Ds
z5T`R>D1eV*SM2|N{8Qj)YQexzN2uZI^yC6#UgG2g&3$+G*#)G<ODjIv3_kl%nvIh&
ztR3nfEFTmgtpGivlg(&{b+#x)mGhBasN;7MfrXV>T9%$0Uc@oI=@Jjv?a-014ugFo
zI`enX32BplYn6~&9!0mi!ESxGHTaqr<LFjqrr|laeR|uVt33H+SX8+1rp<fpQL(;W
z7_FqDHo87tzcUpmxfyCnOoR;$;D<~G`NVn{zFX~aP;$@0aZ{&3lqwvDl+VY`7RI`0
zy;9~d-*p1>SPz;5&*R7B0`?(o^DyB@GFu<C(g`DlyC>TNCRaTpsw}4J#wjCI8Tu(g
zwE+eo^z(qjy-I8I0<_wH`szu^bwpOapkF6yhbRyDy+gOZS%9NvMPK}!x{g3O2Etek
z?QiDE4=gN?diFTedlbp0J<n?aLOW`!1hG6c_27uS1`VCtJGb;tB@u0rHE2f`OM54k
zn?kGX#7eY$RU`uF2!N|O!$2-zAagQ{{%K+$pjy>Hytv2d_=8r2f57nU8CEbO&-Ic`
z);cZT{}<ut-#M=r2l$OqhxZuk<YvwzMs6-Hm)e6HUcJJ)VCyM={8+fw#l~vu5XN#&
z56u>forR(O1=*eA;+>eUX<4c&+EeYb%Bm`L?pL{G`H7wL1cZcUra8M#wIJiSiI0cD
z7|Un<`bQPPFiQ*#CuN!4^he|p*}<(_pz5u+hXCD6ds*3oB8#q@<IIwMdEe6Nf1K5w
zfkh7G&q$plGxIk$Vb7T#89)bf(&5yI)Nk0VD?t4QDuI4&V@xPn#PqB?4!hUn$@q#)
z$I{8n^jDhrV1)u1xo%hc{rgz|3iQTDL;W8-$c0E+L%KS;8=RN!AfIE@NQ9h2p3;hZ
z_nh~cNCFGp!|co{*L{4md;rxiE?vmvw<tw;_j{P>nFX9%3H+L;>RqNM_I$pNHNsp#
z{Bmk`zQ3m%XdDAC3zMfk57sX~2jHOmNZS&2+mtoCyxX_5wr})@%|TgQdA-M7&YkMV
zlql$;pwj(qWJcGedrInSak_vJnq2KUTD1tpF!+j3h7DK$YXK}My@c6`umJ>zZsy|;
z1X5;L-|?j+#H8pdh-bXama~CcS!_Mu(9jU(DgoaJtqF{uLO?jyny-xueV@|HL6p~v
z`_Cb~8RUx^@kDk+vJ|*~tavfwWGEvW$~PqKkd$8C2A4RRz{k!_zoD4VGc!e@5TN5}
zNcESY1<zo#C`DH?drRk%G%Ftw$l-@eo*8hMtS{dIqB7PHbJv_QlD{p4GD7MGjCD`l
zUhQ_xqLr9$3tlzMWG<`CR-^}(7$Pr!TN?yS%FEMHvfo7Z?aJKFp&e^7&+h`e>IZmd
zBxH128|dxYb2TmEDNU=U3}}b%?p9zNVF40j&Kw;3`v$TM0C-{qA=w=hBXbPIg@4@G
zIX~mu^9)53E>p&sz`qJ&+63mTM}xzlB=-*rA?2GS6EX%>Z#BOL!j{^g1B2GgnTcUU
zFvBh(O}Ny|+e`WR%M7DszuLt0p^o_g?J^EkQ-j9{qKx@PiCbL!J`L3!1W>8%TsX|C
zt6lii<Ma+45Vn53eJI?441qS?a3b5MoRP5emy~=m+h&h@CBW1=6&}Djrd{;>gDu*`
zo>Bu#s1d&k@Du!C?(V`J!OP_~cp7y_ksY=Jv`HUiWFZS})$kw6XT*Vm#C0qin=Jl?
zO;13<!S!f?vE<Nnt$OC85RuU2<-oA!6Uy0O5}YInvw4ozkk<GrPm!t5pY4#o{*i0P
z`1q&xT{*MDy4O$|J&L~ZM$-2f#3z7+{%l;|8LVK_4Q^2NhAg5XkLc;wc32FAKDN!G
zh$?@ioP1<Vcnu+`FAZ%5`Ps`sj#{^Kw24=L7YRVLA@#hZ3oY)ctDl;h(i3epN$F-*
z3SL?ky`o?OWN2t4Iz}yB{1HCpVz5ueW=ezG;N)cbjPC6*9~4nB{rORUV?;qAEHLQP
zO~hU$3bKH^@=3s))VD2#fQ^Uay(kDwHe43(N4)@p2DE9wonoW9@BV}~$6^^*6L_xO
zPrtzb4gD2WINXOZ%*llmtURDa{2n+8A8#;;oQBI=Ac%mUh5P5)dp7gkkK6vsWQ@3f
zmBi}0gDsk>`Z1o<1I!rBn!I%n7v{w7mrX1_&^EUO?6}WLfLNN#zY8@UgdbS{5>xAs
zmQA4F{?|YK`@mFn!XeQ1>$6vPzcMPc8}KLLfN>^mFg1bOxps-pvR8f@e4DI{g>aL7
zHhBK(z~EtK*4N$5zZsuC{Q|5)5Da0ntE+2#Bq2He=q{|W_gAPHzlAq%SWY|pXIam%
zq9ANG$GE5;z^1^Jn!=JE>bm~XcwOO<!EBdVP97qvsQBF2>F3`yN5b9|A(sUOvXsKK
zg6N27s5F^DfaNe`g@CFzc5OI;AW8*V?J36b)O$bbBzu4TV!VaAU*oZ3$HOA{8^f-<
zYbGm9DiZ@I12XOu6&2$m6JQKBJSz6h8`!@b4AD{)MX&JgC>F!Y#K*ju;GKATeY5Vj
zNI7Xzh-Ax89vfcZ*dFDUGe+|UJ)j3d&&%r%+k4T}+*}<61!V<=kD2wbuVYw~Ly4pl
z=_{ZjfiJ^i-@oxs(W(YFa>LI9zTtm5`{0e~Lsx5)xX(0LNd0kT9hK4`2ZCth;^PVH
zO)l$1&yCLY+H$R_vIYiA#|r7Mf<zm*EA<9h=Y!V4y+|IUy|aTwlS@rqy||#@^6%PF
zEOie0hhl?S5-1Du*Ap`rM2aGSAdtJy^FsV%cw)rmpPf=zMItdE@a}-fZPT1QZf=`t
z7mr;Fql;YSUw{R`#OkGc_`bf+&}TRd)0^2`Cjb={xIO72Gc)<_4~W!2hGb@8A^cCY
z@i|(pN~azLF~Bov-S21*90r&3#oq+L$WClMIy+hMLE!t6-nGJ#-~R9=bAtK1oU4hU
z0JtWcPj23xoGi7q=?7&v3wEf#>4FsuPX3uXs0*e`syB8%AV>mz8<>Phfwxvw1(JUQ
zI#jz~(>*TU&Bls~;h%?sK<phAr3R0AJhbI_aY!Dtsmi2{OF(+>4#D6swXyPQ{!G>@
zY@+u0Js@;IdV&+z!K}^2BTT6-t@IB@mzCWZPB>DaN1f2<j$>DO@e+y{Iiwxr#`pI@
zU&qlRj`AlXhINFeon`<m+WTnB%&`J@F|uoPzN>4f;vnEo198|*FAWD$i8ZRA``iYT
z%Xg@#_J3v$!Anz7RmqrlsUF8#f?+Ey3=9mGQ?8#QCp1I9{@1n~KY$x3>y<9ed@`)T
zn+VLZ{x%GI9eDGyGVBNH6l=<uJ{%51MIzHG7zh^w!S}*QsZ28SrYHVEub)eap5ZAs
z-a|y84@de|xE9vE-i`0nMShG4d!1e*MMH&byN;KJ;~H5^m2IvD3&<zC3Oa8-p$(V7
zCN8m_`b9S1lV1L}ik(`(eM?nY6{S(2w|7fWfE(3zoQLmnk_?Z`?>Zu$h(RvN@sMZh
z{6{(^9gnAf(k%q`W0{88S%WqfGG#w<FiyVttJBY66=iTr^sy-JOZ@+@1qj$6yCNWd
zofpHpnw>q^_j}ppWD%^cM}S7W9kQCmsz2&uDcza+9TF8aj5Hd2N*kvRmjV!Ma7bB2
z``d2|9Xk4sYhDh;$;p90Ng|?}N@B68C|<w5GW<&fm@cyCaVW~E$}BG@=tq+&=5{B+
zu%;z!!#s$Kt-3K=W3#(k-FrRop`JP4BR0U$#mhn}-kNIo)+Xi$6r)6^hU$8S>WWHw
ze|JW_?5-P(8*(VzJ@OwY*%gzNYx(&QQk@$3Y4+On!~j{jgLeu;4hh)rfG?8*58X|e
z{M)d6q60y_Fb@y;yL_avv+b(Ory|w@2ph)&ZPA9M?)c>JI8tbyY&P)_kfH$@Ipa|z
zYtEd|wsgI*z}@dE?6pq)$B(=m_Ie!8P!I?$vnBefiJBDROMv62&(Q)!OFvCO)?@nU
z8whTIIM6jqdpF%s6133}(19#_TRzsUMxn)`YQ?{az|@3|y3-4^!Xt8iRjTD0qF_7v
z=K+Vbv|?wx4D|Fuc0&)jg~PqtsK>v=ESs($4G#7X4$D24?d@jt4+*}G__0$52<@zZ
zW&uPWaBQJ?(bLi<@_r@fA}j97pzOHRAKh>#_2Aj_$6#imKfpqdjxcb?z}}pZM?<)V
zhx=R=WR&sicF}Km!?B*A?Z<3X-Snvif|OrCyaYHV{=pAsV1e?F++Y_m-lfDtr^0pt
z7Kz&LgRd!b0<$19BC4kt;9JcQ?+vXIQ@5<^Oo(<`KJXW@8zh90^u8ER)A*Pv+0}W8
z!JPR@9~An1p@%WN<jAk9*wnHzeA;d$MU`8K?lo_pnAy5}W5<7w_ei=^-?zrEnnGhP
zP~>>#Coc@dLB39LJDi$ip(2jIPRtDU)2La1`Dsf$?A9zHy0r*8>R+9mNrJYD5f6)<
zn;Dqs@n1Ps&d-7Z0yL*JBm!l22V0(U(Fd8O`W8Buv-&JhOUj)&Mn^?a-yVkS5F+AL
zaz5+mo>WUfdj6>|38WXdgq+H<DjdKzER>a!pu4@<Y(yxXo|Q!yCvDAZ{_=+=Pr>lu
z@bs1)wqNtsWc3p{)x|C!#t5e92ybtzB?UF5@?s80LAUgV<juIde)7rI!BEu++ZC<F
zWx&|GZ(j>g5qkO&F#ZiGN9opGC@5>a>}Q!BG1QEWf2A3(TVp@ddDK}EY<9HyyUbxw
z4!Qr<vJ%w9Qz4Qzd%H%xu`Z`ogAq~0R=lx)*sAYJavC;O`M1YfkQD<U^}vKaTb~7=
z!*Hvon<@Gj{4z-6Z-p7iv2eYe0xg3DwCB(g_NtfRAsnS85Qwm_V4%ocJVuHqBJ6<z
z74pWfhfSw?VDX*u>m~dk(;y)Z*3l<t6Ql^}=VsiG-=)rYu9J&oLpO5nzVQh}5}yL4
z?{z2G&-u@PD%S+mlty~)JeL}BRitG*L0ehxvC@@)(&?JG^`8u5c8QX@`}7|Dz{S(P
z2X@C9jSjZ9SV(H2v$9`0RgfHco3Tj6_b)!Z|2(7bMM1MHk<tJ2r?{)q#xl_?R)2hE
zad`9m4mt!`sz2Kb^b)#*R%gBg$iHML3DE<BgwKhm{hdb_lC13P{Qz*=d)}K1`Z@8?
zggupDefQzsQrB`(+sdP<lyp8TCqcxGI6nSIsu&-MaM48TKrOYtla`A8t*@WBsj&S%
zIAozB%h)c`vKs)aonp0&wA|%?#Ku&aw7Okbq&PC5%=4K2n%C~`FNmUUaK6HrH||3-
zQc}8$Z2O3R_WaE7lw0dnAtCw=^SbyEy@}xwa~GlKscS7Q0+qr<^t2MF0Kj38)8Pfg
zPGj;9Ob)zMQtC_{d?tQBrx(h(8qd)Yat5ODDjaN+NQ7L(fks0m{f5y6x8dIboNQEO
zaZDd<q`TXEY`Ne4`s`tyB)c`tKEE|nVZxlC&Eve-G(CB+1cx-J5{^&M;DDsKMekRf
zfxzt2!5n2drNQ7%qsPO;MQg<cc?DVC=YJ4UiFT`q4txRIrrgw`M;!bt9Q-3k(tl8v
z?)IL}GzFba*E{{hrD8(AAvhuf;-q%sJZPC^XafU2vEAn~PPso3U-JDy-k!)%xq{zQ
zZewQVS9I$t<(0aVBeVSjsNX*J&x0(Uj!Bn7FmP`CCi%zP)3ryBxP|-m*4vpI3PA%I
z$L+ckx;U%(Por>h9M||qVS$IoV={+K$@O$jMM0|S%lF*Kc^1q#L8pbEa`#?yhHm?d
zZ{5sa^n6WrEG(IE4FNZO2fYb|l34SvdcVu%;u?ZAj*^<IZrp18Wa|gn=j6W&2CPW}
z;ZaG%ckhv3KIW(3wcC210nKt+F^nONC^nZno9`1l=Sb<5{05!o2{u0h@>F4WE2ZNX
z26f(Zzdr?XUeV2P+Yu3bY+8qFA4Ifpmn+`6!^H)j;@cZDCl?9DM?MM?#$54Mq}(>V
zAXkPv@=10e%o}BM|Mjz+G3MN@=gQ;UnUX*_h5SAV5y>e*XtV2y8wNH`q(Lz{<T^L*
zYzG9YS%7T4?%@1$WP+tJ8-k$UXzHV^0Q=dmo{nJuPtJ7W;BDZ#1e^hgZU_Sd87myL
zAAPc}LhDL#0y)fxkA*t$Ie+#EaSc%54fs>hE{+#W736JJ7pxJjjR#M?|CFBNpN=;$
z^-#*uQO47g<i5JH!Ib-``;4VWOw8!^Lckecuk)sDTgqHxc%-CU;<uc99M8K+aY<$)
zDdqjm!z+tPXaD1-Lz4yXl5Wke;*;$1KV;6xsj}l`DFn7~Cx9N73Ru_Sf8eJ*F7d^e
zpgp^Zi?i8*l1~u>CHi!|a{FYEu^A}c$@O>`<+XRrQUR&RYqtBO(QmCyzC&KZve2Q5
zvA9+HTJ<SvgDVO_)G&+O;WA4%^S7RmLm2n5f7^+KG{H-z#&0ulyFGSP&+<hlL&419
z=mt+%0{7V2YIhB^Q04XqbiG=cgC>h_TOkgxgo7i3E53l+>?)ZQt^}DZrX&{n7k)dG
zGU^ZR(fR2fl7I??fr&xbd2wziV7|W;K(Utt9M@8Wp<Sx7nUaU5J$OZBxE$VpU0q#H
zl+E6ZCPr44Q7c_mS^3qEZ!xU6Kk2fpoF{85Z!Y62l93M$0muE}#bfC3fGPv}cVuXU
zl+Rj1Mg4(Kyqcma9;Oe(zrjbPcSrGVb~gc|@lXFdd$+w>-V6&9EIjQZ?;qBC{$hZ?
zrW(#+m=^(o&b8M<+Y;hJypDc-{=o@~HI=OVAW;g@I}b{2_p-ZB95yTSe3Qj)7ptkL
zsK1P8owf2#Zdlkk2~9@5TKcM`^>Wqq^2M8%U+Xs7zXUF&RXLSkNEC{B#m_dkJ93Xi
zBipnLgnrF<0Usg6jFHD8kSco)KpH54+f|KYU!QIs=eoAw2Hc<u$9-_uabsi(x|C2J
zr>js!_+3zf2lik(lRC^F3N2JqeN3yycE$q{u3OnwgwmknrjXNG&xOeFrCYsM+kAPY
zqj<|t6vReCG(Kcuhzmi-!MKT#U0qz$(`DBXnF}A%Z5nw`+$|ey&8r1=#b<^D3`UYh
z+4vs{GQJ(OL_fLcZSoU(gM#pVKj~&>7N<(fD4mDII?BRkw)@yGpZ&duud1qYy>ULf
z*ZnQ)=|QDfx`&LM?AFG{Lp%%s^X?T)`UhRQo#MDogk$pyaSN}=D(63X8ocb?UyN}d
z8#mx%Q+m4esKvwVw+MpigdY>egpVn}FQ9k~ai=OuSjae&ECvR~0&~kf^sw1O&ud%O
zWC-pt4eynCnonH7qg#>PQqqWj&?k7h8`p%FPdXDkE`u+)FF_ro$@A^mm4bocPY6Q9
zAs3-$rQ>mKs+tn;C@QY_W8(}E;!AAQ-BJLC;Fm$PJ^ewla!Yu5b0zI1J?d$7ANgkX
z*Xp;eiI(nfnb%iVhDV3f6m!itHu#)R5QXD{s`Qe0DT-L4P%ru0W;tVdU(Len8XFT;
zQ~qizVsO|>Ny(rfX41CHo}XsKkAgYjw_*xsdb)aiDE&s_<N1-kk%7+?c>Px0OM8BU
zq3uk?85w)cnNFJpa(2}QP6jeIFRG(_;y&#z%~T8zmm{gUX>yd2itLD2+ut8ypI%Q1
z_sd@M@@>tk?REDF5Y4cV?j)DMeohoj&$uT@qpT`1J~;L4$&<^LH7gSndJ}rz;&7C3
zsr<Y=$&}ZCrvwMf;$x2Gh}e!u!dz=O`gq+pOMrOkUBEay+JxF-bA5Gnebw9JLeAzz
zyuzb(;z1`*<HM?B1851mx8Oq>e5lUUeMbV-053PKc*Xn)Ww;cs_x8YW!fi64Cv6#&
z#<7cEQd$O@@Da=z0&26G)FoC}BICttj?3gSFF84Cs%rwopXw_O`Y`e|QD^8$Juj``
zy6%;^z_q~Oy0HjTYY3D1Z58c)Ko3I+d7e+LO;)~`RrVQQ>HdhB=m$&kg6$(*%xYJc
zQ=;1+Kt2UgxuYqSzt|1I=l1mz9!FA?{;wBOQvTD+*L63>HR28Mp5q4vOVKQwmWE+1
zr#wS1GVesad6SRoeNKA5|783l{F0!cvgHs%0+sZ6P;btiYNrT4p16-gWNHei)6~~5
zk3Qc+kq@K58&-~ze({TI2*MUd_B%+Ew5hQV9&-%^21=udCVcsk%Bf(*gebgJ?(2Vh
zpe7a@mvmN;V-V}Ov~!{|{CJtiulY{ggVI;=TtwYX`SOa-A}CgjOYtHHtE;QwX|)nb
zXiStH@3q0~Q;_0T93IMRldpJHx>ceI(X{huhR|!y^tF>)nPpzI%v(1HDjcfyzcq0Q
zC8owy2T;kbyc|^eFuVnv@Z?i9(ecSqpCf#2Sp4r$OLTb>k>KQIQs78%GHvrtjuS|G
zh`7zwd=p^mlI=5X+qGO$I$Jju_8=q-)GmF~0lnGbrU((qE>Vl*t2#G!!a#t$XAUMP
z5z)9%8=Jj*%$jJ(+9=cD=<ta$`y+#0Z$3`<UR^$?{e2)$C4$Sm8y1cC<_wpo&ffP~
z_LJfsGf{C6B5v{|sbL+Ig-x5sB4xyekiZYef2!w{hp}+E3twyZ$x3P1<auRabfJ>!
z%b-Ac7-0Tjt$&G>mXZ2fcAmDvE1;kGZ{9qKjvfiDZ)-OgfcZv!3@oP~Cp?dD_{ac$
zaJn&?;#)P!!<i)^-dc<4uB55SN=X@>7+;Xr9WUhm$(YaXbw+@UB4PIxtuXuF%Vs)=
z{B*0Qpyr$y`fUtpblZG+m}b(P*HA#51<`CrH>lpzGk&P;yaV4~Ve_<|gG0~R+t%%x
zw`+&m=`%DDKNI7Gkv}V&Zi8Ew@YXM`lbMy4{Ry4nJgne$B)>83iu+38)zA4IrY;%P
zeZR)C%5n_1({tYxPgOZGzs~Z~Vhye<Vxr(`gZ@dZ7)vJIvkP>w>q9MTs?063R<*3d
z9ByAF!5Q#YKE>wf$sfpVgadn!v!uGlwZeQ19RW2YXSVF&ZKv8dQ*W;eCH-9B!h8N^
zW$rujIyL)4L=3Z5#Ynp-EB#~bN4OfB$0)-1Q<6{aSiV92bai~d=ZF)f0G@c#^B4ex
z?vd6P3#c69J+0#df_#tNe+)d1Ct3w%BE<ebql^3V*ZIssR6}9}rq2op)Bvq4z|Vh@
zQkqrF#`#>fv({x#PWc^#@lZ$1rZ`?)c2;y#5x*B{>8V2tuenjub9vd8B!2Bj*U)i|
zJtNZiHZv-yjO&vyXCI3`Xe8)!AMl)??@bf#@0XO??&dRf5ke%)ew~JfCb^I^VN95T
z={x0_zS!hAbA!~-tUfa}rm_8bR>_c*fDl<@V`C^P$OKw%wXABCzw6H;_u#YrcL}f7
zq~q0#7TN8%ecm)b_+$ykQ@QnCFBB_M9p!IUn(7^ne?EH1NIgHdfD89Ke1hi8vSk$f
zGSsw?O!APC2h{!)x}5kG&CT5_9i{j9oGtIMXlAD~5gBRewWkeC4)HQmuk9X8Sf;;8
zh}X0$SBJu-VSW0lQ$9gHr6w^d2@P4^(>toCh64zowxvU61d_~{8L723)f${yT<B3h
zbT4WN4#qgjdrKuQ!zZkRAa3_RU(C0#aXLRB+F@hMNiE6&;Z0K)4KuY_?`6gWcOiND
zwi0cK=kc*R@JJ=Y2}B9;utEU0^&Rwe;8q=l#88rbQ0sX2E`_oU=~lWtK#Iz^;FQUd
z+iiVQoU!rl+O-d1mx1C`-!v+gKW_O96IaB?-~W;YQFr4}(!PfUO|RJ=KQ4bSpVId8
zA<^=RvdSo!kj8>bqNBo55qZZLuf|k&KRK=M)@bwzsH;gRe@;64K9>bKwmd}s2w7m;
zxNDT!bj;4z&zMi?aSoKgTc{}TfCy7#y+2BY^NF?RVv_DCh`R_aZFaijW(6~Gb#;yW
zsm=|r=n8JcKSWlv#=&AH{Y$Q}<4D=W(KVRo@S|OUm7P9HiiSm7kQ^yfkH1?h_#wGL
ztIRE_IQ)st3x4P=Ub26b=<XBkof@X+613^xwFma|IX%pVUpVLEL}?sp{Wa1GaYKeb
zA|5j_2Oobl!1OsE3sSx!tJjjG#>afNE2Pj@t{z|Cb{irYX!*I*igVQ$zVIzS=Cpd|
z&-%x&AwW*Xxp3@jq2;kR6+J_{?zWK2=3F@xKs;RB0lOAB^#|maKe3Mw{z_<FLm@{j
zPN%k3JsMuCC~5k&%|0~Y(k4}vx7^wb8x6s!fQS#{F@?r^kYpw7eE8H(@FmJMFjT_T
zd&-&AzZ7-!B=^RhaQ2<&^y$p)=7pL0S~@yC{XZPq3NGNVE2t=_&#8Ln-<jdK8nxze
zUALfJT@5G~zgdNg6faiXT1;0?E5odU0+eBwrRfg<(}U(YW5fJjPg1$TaBI9?zRIV6
zw}S5b-lagny%zse=>a!d*T8gr|5&zZC6#CN$A-pAN~L?oXY89S?J#(5IU+(hvJUO9
z-V-=59x?nvYj7Iw`@Jx?z~&H?2WC;=;%V;cH-}DX<At`}!Y+CmL2)8;NP8&n-4M5u
z(LLazxh*HHG$pFb!_Ph2{5$;v(D8?H>&%!2jRwC?Wez2Z-z~&;g%Vy5Z3aPw*0vB4
z?tL(Fbj(_5CrO4LSdoSd7s1nzg-V`8p+2RJ3?ZK;-enMEr&mD-R4ThSfm3YZ`THNQ
zLpEnFD7H=1ogaejhoeu@cMbHik1kspZ8+jemi<aI>Ov}_p))_<v9aj_`<$@rB|oP?
zj1*12b~z+K1%P3h?GNV|@o<w-3C{~}VUI+=pfY?UCoflv&vVYis(r$n3WG|o)6tC9
zIaF*=#1)*~H=Tz)VvHowG;i6fD87+>_RI(L6aPM-fpH1iaB&nYT<C`2)9cUPvfAFY
zyZ7y>UhymVEh9-+!-vn1S6r6dWPZJ@PDVL}HCEG1pO4f&MtvCzzo==N`Du5wuRy=>
zo_X=Lk?rd(R?BpKSJyEvhL%22a&uiaqpakR!@NF!C>5-eEp(DMvsxt(ujV)Iw60=T
zL7$~4j%5y8--wVWX*^T!fH#Dh!sic6Lu?$8UtPfJ`Z|%hr2vG?fx)cI462;BwA4-0
z;vkLd-j3tyaAm3nh4Wa2^M|RNa5&;zU0j5wv>`~sr-_Fxu7sRGTkFcseKm8Z9p}w!
z+IQF9TfN>2QPZ=0Gaq^EcDbJ$31@hK_e*fqB?&k$k4&Mx<92O{@BxW6CB0$lTA!pk
zuOrkO-QBL8pQENK*Rm@rLe?rbUxd*VrMNPnG){D8<mG{Ho-i%%+q0O=ALKW6lrmU-
z9+CNJmhMg@PP$>E&aW;|@Oypv5+HsArMFV_>0os5e4De|$f$lt(|01%uhsz_=H?qO
zUj_x6#e|a1`_@-}W9lwQ6xiY7%`Ew<p)LD&;+d_&u^v)X!PVe)cs%$g?CWZLxay(>
zOUNVhTZl|Dn;bj{&bO?rxA!E`<Zs_o><R|~6fhJEV>b{bI#;Y4*9<=H4vqv9t(rnN
z>cXF&8(C*LSuc5}pWucW9o_2vy>wFu-Oip29LEOuSaIT1qyqap))H!de>uk{$~<v7
z`$h?ZQ3LO}!8C1~J*<V;l7_9z?_bJcgn%bT#0}IB1VZ1hyNZ9F0hw%~fc<fQ<e&I)
zeyZ?CpW1?~Y|0wkS-s?9@XM$BWOm6qXRV+YT#1MfxvIhl{GnB4bN55bAB-F3(OJC)
zmY>6K0pVT_%Y*s|i&^%7v+eezJ8t{Z-e~@-$=^eX;ZYeQS5~*)2Nd@QG&lJ8N;){$
zE43ShbNU@svBN1Mwe<K$r_v+*PpxHfgS4RPXmBH)9vEMpT@~yZH0OOX!fmOQk)0PA
z9X{4SmYYhkwek3EPz@H524uWQ(MAGKE*o?U&{hW9!-z*+xO0`do|rh!vQ*(+vXWiN
zD}!_*k*z>-);kf5OisQreOI;=Lw@TF<a2));CS+kJ&2X~;%@gd^(8&TKxNc~xmf=#
zdDcpeEk>Qr50iAVWMUVDLcV-KLGEL}?aV6AkkEiK5Vnv@6(3MmSam^t(yEz6s$nrl
zaqGd@xor2lLdbc>CPIQfg{&EtF7?K-8+ym2r(Z*W`EakPXut^!3Xoseae~bYDYh-g
zjb_KXSgnu*z{K>9*N;W3!XeeL?F(ny`8m1oGx0j?>ITz`=!(OTwo9JGR+)I-Y1_^9
z70yK#6og-UxA+PKmn?!Vc#timi#jz^p>O#V#t-CcRJ6NJ#$pUEUTf^X?g|-VZoRrs
z)ZUtye2!A_A41WeOoHID#xSRprtX#<KiRcbxp)i&4-fDC<>=O_NVUfq8FE8tUtret
zQe~MKi4`y#-~ED>)FyMy>$<CFZtk~dC>xNw*h?Z(<kWI{d&(5G*tPAp{lTsDYuAx*
zDYu=4>Dv>I5Tkz`380~D`wV~-s+Z^$F__|lgS&g5>TJ!FUjNu+8&d$7wdJc)t!L`b
z%Zf763RD4GowY%EJyyAAU#@<A)XjHs)hF`y^$*FY4L1;5rTGlBqpIKpm>BFiqoS%q
z8E@-PblBFt25L*di2j~3tFb0_b-PXsb_!p#<#veDn3Vl#6zClF2;J>f{7*Zn4KRS@
z?e(7at|!6y1z9djrzh-pvm70dQbRPWjW6?0$dAfI+~0Z1Xw8vc`CaD`c3r4+yebZr
zxO8+Vn<;wxy?6atIx}48nO}^wt4Hp=l1%slQ^;9x(eK;w2TKsBA0>*o(;T;~wLR(W
zPf=!Q=^PZI=FBe2af<C`ln9EwdXSlUzn2Ur3Z*QjyWaY!gH4ST4UqO&KEqdmu2<Od
zAH-lhP=H6F`pyPhDf)%y{8*qG0AbVftx$34_(^U~+BOdTUS97I)2-2Z^B?wfhwx6h
z{*ec<S)e}J(d9wjg&#zjoP{R39i9*KaJr8aXziB6z(K#!&tn@Kq(_#z5=5r#S2QqE
zvD)?O&Z)ccjNBPY_>^#qZhPGYhP};mlkX%-4(Gg~U{SGrD+hfVv7Rv?MUZ(#M4VFg
zZl`v0)56)AuSSvjnOfC6IULfg;Esh?2}s|3n@g8^E#5QQn6R+l^#?VQ^1gqmbl|(9
ze3B?^|HeeUB~g9w*DnTaY`eG;j^KZWWNYAYEsTxR2Oq-49QcXs!QD!*{gnl+A0CE)
z`497kH!D^EgcWU~2~O8qcVo<8O-%`5E%xA}e@v*{xZyPy!6ENB`q!+%^4&W;%o{{R
zrqF>Q+shrL(ng`Ii97=Ie2^1vJUgkrRzEY6ee*2RcO<0*+1<O%Te~${^66K4I-Vrj
z1cpvB2((SjELmCTHA}2EhZ7tT3R-S!-`>o-_4j<MJS2^b{tPaOn2%Ylt{V!cG&ud>
zk?m#{I3iiDIibt>_DxMW?Cfopyu17LNcywo9ZsAWGiWWMS;4_|>%hJ(^Z-E=N5007
zpFR-2`<w;?(??51B{DfqHZ4q>gcNBM)s2y%O`EsfosN!Do2);3eH22TXJ*#<i(yhb
zX;O=1iAke>YKpT-%5$&vadxkE!PkQR;kUWDkEk2eZ+wuuvY7e>uk$BA5&N5;=XE)`
z{!2nv?n5IZGvL|(yN0ewoX68&{EErpwXqC5!vDmxxq;`;#dqn^r#e0m*$#H@-YZVI
z5cuU7@4<lqjVh~Gii6ivP<d|qLtvVsqK;#ArojuBx0dp(aWbFeWpcme73SpISlcNm
zhrv-n$@f_QW4MQ=rlg9!?8_JN>FG`Pd5n-Mw~Gn|3Q<)$fJ}2&;$S))V3yxDHUJUE
z{kspH-h;K33ftcIvp4F@OiY-a)E2nL|J{!;dq=St!}qlM9d}JxEfup+Amy{8eJjdH
zsU_B##@B4XA^ZoN<>eVAW+!A-kHZ^r7SwTh>tM8GbRP6b`Gxbd;nd^D@SJd6lUejo
z2VsMR&AnG!g5|X!Sz$pvI~NrmGCn?`sX3vre+)+~DDtFjmCyu;1<4xDj;IMP-rLa7
z&_MXE0{_vhXh#1o)->h4H~2xc?0f`>sp%RZ4I~@p=K0KkF#=TT&!sC=SLY;n`MfW%
zU~{uIi<T<<KH(jD)4kA$*u?R{34zw`u|`o=L5|b8s03qys2B7s^kNc>uSlY$XlQA^
zP?`Nc^=$<mF0mQbW0fc!ZUG@qO$yu~HV!s2z4p&hxdpi!8-|>QYF&PO|6dDW2u<_+
z-Lu-_^f&Vd_V(EOb}*;R<B6WW{@tjjJ65(50h@3`!bX!(j~lN(g`ZEqR=d`<viluI
zR2OAsnHw%0aL2a5rylb;+PR8<?3E_Ra)h7!F=0&Y(L&RkVz?AcR@C2Y(Xqd|B!5a3
zzKCNY*EGM!Su*4CL>%uLisjpP6fI%r<0@|QPxQhcz0M3SZ@Gsye>KzE)b;Vpm;UMg
z{*e(2(|1+Y8=r1-#{SnLT(Ko3X>X7w`J%csTbG?Mz#1(bTMr*0)U4N}9^FcyUzAsT
zsgaeRxhyyyB13z1kQZ>GL<5bTo54R^goCYoC;Qk&H`~+quQXvhyP>Cm)jD~7;r;(s
z^~}7;(%Q$CH-tUO-UHp2(zs{*0nnEO{rgup?fS;im937<gOui`u7$%XG~F8fe`hH4
zEec^BnVwPJ|9jy7@GWXjO0sT|rmqK@^$vIbdsI+tZ)*CZtMb6@22nX34>Plx`ed9^
z5wn4YNPtOLV$Ad`o}Lg%%)|S8pw35wfx#$Udez75B&hTBB?H3y+ui1SVMRqjPNqx<
zeO`9nMK4d2$a6y#dv)h8>E5;6(tBnxe;FI+_wM2;A&t`x3o$p%N#cP|2~QG+D2van
zYYqNTOAeG^I`b|1ssFxw2$pBA%CgnfB5`_~%kIxn;BQvRm;rLcT_~wgJG}vgJ0U4C
z{(Z9^EK?DC8ABCW2mpXw66jcr^=E6|;169xpwm}ip<%yjKA*7X`oB-3f#e!l;a4V{
z`{z7XE2|q2gkC35`JF$MY?1!trVpI-kJ8UYTBk20<Ks2Y3z9|^)Gn4ZbE~r)0K<aW
z9G8_W7YBLnpQm-WU$8v5d$_LD_LJ@l3IC!%>(t5UhUV2Nsiww66IFU^72n{Br=~(S
zxMSu9I3C{g3H9JyTC6jFXA%9D4b|6_frY-as|SWspEpv2b+1>air?9LbVSzdD%x9R
zbf}(@j6}%mOK$Gh^nJ(S3(JqqvlM#%IJZs%V_x*On{KUVM$lg$4g@Fc84j6b!>xUL
z))b-3a<g+Eaf!AhVMpgVr(SX%%j0pw6W&paso)UV-^Cke*7}peu2ydyEZoI1{PG4z
z8O+<)wxN$mpMi6v#^aK>HLg%j)3t46qvjF*ZeFT<bpZUi)8qyY<3{K0Ws|<WnB}AV
z6{V3vokO)xifS|46-BJ;^n@S23_g^+zdCPKHJhlZp_&|@4CVS+v-Y!=F8~D5Ifq`f
zO&oaQ7Ki-bDm;-!|M>Ko&y}}fho=bfD9hIR>Pr<>vP*ikg0nhCh7&(~F5P=@RAALh
z0_!-+>>K$ZsQ@SNN%j@98kXl>_x%NRY1_dDY01c>ga$ZcVf)kHt*db3c~$umB<U{m
zsw^=)k7l-$&Tfk}XP~Ya<32qeO#3PF40QMq&4XqRCWDhMlgO(A&2y}km6fUec{%O`
z$Pe`<o6eFxv@i3^i%bf(mr(8Lt?{@TH^kpdh8cAGhb0Z^A#bO71|c?edrl>G>x5Ca
zhW*mQmqc1nNUK{$5~$i+lmD#!1k!}ZCnB};SKEa%A}~J=(5UN`#-&_G8>h~^rgN?+
zp&xW>FLOc`N0QDU#Pkhm_l)-|$^YBU=MSx3<<kV8mP_vxrdDThb23UQDn9Z1a<DLl
z#S*&ocZH=S?;fVl_Y=#=mjgc6($d~<N-<*fH)$)&xT&e>zI>)`VuG5n)!Fa|Qr)a`
zlM8T^R2NOG&|}*6tn6Q^D58QS3gWCLfkeQF_!$_JhETAw8bK(K9F*^AcN77>eTCgB
zs_Qr;mDjMGrKqk#dWS&zi$ZItJ-Jh^uW#$;GHgD+tSBFKctSlz_HfbdI1ZHWPY(9?
zC0c)ih7GzN$uq^A1q7Egp>J%=#L`IVxPTW9eKGCsZf7pSz?)t?ip4@gLLnI&0K)@Y
z_(~(5CFF291WIkN_AX-r#3a8|7ug@EriZXu1Zlpp(NL8u0P%iJV1`uaUO^6xFuVGv
zANQA52#HgMrlvOL7QlYsTfjjWNPy`Bl@}CBKxP1Vl=Bf9O>hjmZ{+{Hr^4`9sI;_H
zT3hxZ6@x=i0w-bMEoXwahI-s~a_roGkaTmf-ogm5S}<%~{~vGKT&h0dJYt~_4wj#@
z0x>cXP6fC4oY7|`xO`!$elaw7{+wOd1=Ltm<y&k%GA8pCw^!R@e~UwU=+EaG8m2>1
zjQuFQOciw*Q0|VB;Iwr+?_1&7Vh2Mq5nwUTJXhBa4ro()<>mf~t3Z`5mthVGa{Y$@
z+k<Y`#AHR0ni)6>+dV9Fr}G629bmoO-q@I#B>>JK+ud!MK4;9e`kTFg>Y<dpyk+}Z
z(PFeJYSQMyYsS$lK?qN^cXSjG;Qt&Q)%K7Uf{vWw4NV^HVqxBm99Z}p7jjtf;l*_S
z*n>L{lxF&*Vm{khm&Fh#&&#;I#7yYs=6nzy9(DI_!t&!0QgbMK;Xm9c2i)#%=Gk!e
zZ|3IzxN1o>E{>wH^=+Ph8oZ9N{!3ZqcZ$}z+|q;nKUBGr!D1}g)>=A7f+}W{?>8aZ
zqAP1Ym5}qhVAI0z*)uvx0R|tA#3m4r(q_Lc8hdwLXiAY=_4tw)T#dk^6A>jyjJ<kt
zRn)k&r|Sb~&HVDwsH9k;&-YwNYKa&CT!Sps()5i1tX;D0lHFgm=aoc6tDxbMPp(^D
z-E^=iB3|C!h1CKfC7tsR9xUG?5%GZx<cX%t&vwEomU?gBy)|Vb$`~IX*Y%YE%Xu+b
zLF}yIC<`-V@2`%|UmZ$eF=D;mu<ini6`}etG5KDaaXs$bWBCz1Gb|BfOEa--X)poB
zR{exip6dN;Gku|=1#+L}S?F<rt>#(k2MPy@UGx0uY>LdV4&V`9TsWb@<oMQyG)E<4
z)#FSxiX+$UN5B;WA1Wzn5@`5w4l__)GF4SZft7YV+Bg21$HzB5kM=4g)5yTkU}?8X
z@TT*3b}|;cpRCwyboiKJIz9jd#wHNmo+Kl@^DLtu+}5d{#<zm~85j%MhDOVYhnJqk
z#^@@j>hyb(DyX_xrbuTfqU2oH^q`1Rc?>xTQ&Uq+%(4K>oo_R}0a@5+p$?2Mf@o$9
zWvx|5*4JxU<x=QTQBg9%!3YH-UE$ZS<D$B1AQ4Y1rhoWYh=X5_0+-?ks3sgT@=;PN
z3WOmv=+=;NOi)lzu)S`C_<PZ?=;WxVNGxPAYwIs3&z8CKvoAa7VjOY!=W~ydAZuF)
z=DMHW1<)Drk{}3pPYkPm>|ID{h7dBu!v&wFy}dn5mB4XB(yks(PYl8v0EErJAhRJq
z@an<;ngghA7Fj7->CRHn2e)UWW^3kz&O(Wn*DA`(D=R7v_7A6PH$0bGZ}Mg3i&7i|
z4WBj}E`ANguNnS{h-f)Uu;jLEYOi*+s>TQp&s2?Vf^jtA0|Li`#UyB<p3|Ga>4aD<
zO*adS_I9>Q!5FpwM%(jhv1s7g*%dwGk55jXewfD%T@GM=P>KUl1TgeW!G=`J_RWQR
z_yrmbQO&bqQg|R&ncq1{*Pap*ay=x6*bf_fGWWF}_cCihfP7>Ma`H9G^~|KD^Js!c
zhNs+5*T2~3MfW7L6CefCwUJj9eJvf7?0n@lHE0M2M`^=M5fQ?o^<7az6~HBaO8T1{
z@tkJk@dPv1M4@&y-cbTX)1)jZ=RVX5;Cq6G1!jT}9<)@eNTX1&P&z|x0|BAp5l)vJ
zWjUE;85y>j<t;|`TzKMh{gFoG!ke~^<7FPhW%PFuigVfhTOwDN@%(|;`~{u%xd0HR
zr&fBYN_IEl5$?8hC!@s-7P`wa&rdXYN5{~dmrE?5K)r^-I?#Wd+%V_J3TI`w6rQ&B
z6vXB%$&-?~SWHb0jE{|#+YeTcR8#=5=jPN13^rgp@V-vBt!IwXHI=CtP*3$t-*G8O
z4o*)O8=stluLRkyA+|bdy$QtqF;;uKl&eb>_W1AN;X1v}LkOH8Fl;F(*n})<@JB8r
z0P_gt=h3<d!^4(XKJ$_s9UO;@WxEdI$_vgCn?-ld=Bp<Zew{bHiu@nV%%c^(dItv;
z6v%J{DSwWFo(NRE)`ts}_+4{gItEaHKxLT6@@((sHN;<DW~{*o4G|Y_J+EDJOjD3r
zRrC7P-tAg*d6weh;<{qj0sTI7upU$OEzR90r3be>@7_XYSQ1P1_71n(jl)CUqC-$>
z<<@yHD=ypeJ|NWd**)X}Rc(;C(h-zN!^}*5_9KIC`40#RyW@p#BAX&^ZcRrpJi_4D
zi_xJG-71^Cg;{rKJCN<|ol<JL9@vRQ!Q=h?^v=Exn+ufTgzHL>!}w?2Cr_EdhlbMn
z%UExBi*+xnNV>Mj=5OMdFmZ}Mg<40%dInzHb5>Q7@+n433m50jw94I!i_zu{>%g$e
zZ_a>anXVW01}giq9VI%ecFkiTorW3fZ+evlb&0jPO=*!pY-7KZ)5S;~B;NrMAnIv)
zALgDl&bzv}LO+RfQI8ec{BMV!h5FXw#6*R+-bIfom?9ErHE#<|V7o6PX|&{RXhuw&
zAVm{E>~Se2w66|>2j!E$!;JESTg&R(heY~s-hGZra)mLypg5uybMOBF*b<<|*dBJ-
zU%uD_SO$Y^@G;|>XOruk%t7Z01TpE{Us#>}2hB6$8R6T%>>EJfVs(|c(*vrPkR*P|
zh_F-zxaDOlxK#mx^Rd|HU+*q#AO%fS1eXRVeZGF$UC}>Ntlm7kO-?T2{{AUtXD7RK
zCxcyCc~6pHP#GCoNTS;=lX!$g#>;<FFExwl@5`>@s*BQ1>3VFjr7;D042&maA{688
zvsRvCwyyz%`VzQ^;`Q-{HR%6Xbq*h@5_WV1s?Ab4Cky`QtkO;QUF^A#%<k1mS7^79
zSp?<^d;v>xWD#Lu$Lj&WGt&!96GUl!pB)_=zKK8|ASs_ry0qeJ$Mxg~>p!{=ylGj!
zF3(L)QlAX8i;}Mxqt++%-Im5u%FNKS<xTLMjEheW_++=U$_#MV(bma>&^EtX9MA3L
zm7(YJ&D8oDniI$$ApIQ+bdNv3!9g}}3c@TTjQy+TMvv{-!U*r$(hT>fFMr7kc+T$}
z+sKRo&JHj+l<fT^|FAu<D+07-V^dOY|4-l~xIfWzr6p6@(9<AP82MjIduPuQHL&Gt
zuK$Pr{r4mGdozR<Cl(ufZH}-~V7(}Mv_hCXc%2{$t?jmjFjlR;p<Xn^zWtJLhRKEg
zVU55&d|>gX`-4KUVE_B}Ora;4lbWo`_tHG9F|B`GeeF6HvTM@?kAxrS^U_Mw_2)Z(
zF=ol^W7axe>|7)aX3If}gH+3&;W3&ZOw5d=bE?C1VKYp>7cWZ%wgOm@D)!wMe)7>w
z@oTQ{u8KdnKCpX^@H(yAL1qS2aLFR|t`~FkOoXnh>ksjd?tHW5=6;L@5VPkgIwFYU
zqNu6HYT67t`j4ce9N9D)0>rwxg#`+NCXfIxDzIKsgo*OK7n)b?xp%6chlu$UI&z5;
z8qU(Or2`!2D_+NuH6s+&=(kWAAYLtylV2XwZDF>B?qO#HU^0rRfqsL*PA&W?kfOW)
zJ_EhhzvVu^n3!<oFwMjH_E=_hbCaHhg>;VPCAN#WRBF1nAfi=g#X&`*fS*grC@>07
zL=#N-$MLW8pDAF*;C5JI(Iv0`O;#W`%90!pDe02;B0)!CG2_nGY=o!LN2q6}e{g&$
zi%{SNNF~aVUVK=Sg)vmE<((g{S(KJ>fkaTenDSDmV5T}DMwhL_bsbKFb4%Tk!C_(B
zSyNeaQUt&adVDBjCk`}^7JK7n;iKIVa}5tC0@Ctft@YT^Y0Ka%CT>mv9-amS(eVMp
zP)f%%bl<WG9BHL_(wTPukEg2)3u@c88xS!Nl$I1JQ9`;?I+aGc1O!BoZb3l0LApgs
zx=TbrK)R$Qq&uYHP40c~o$vhcoKJMe+H1`@=7?vkxFo3yo~KveCCY=xw2h^C*O50f
zckbTuwNPhrKHe9Ul~q>ZPk&y0de{Q+)r_^q0sr<UjB0NBeus(t{=a`%I?SAjWF&C_
zL6wv~1;;W4DVdGc)r(*Bq&=JN{>P}z>F!O}F{WXxhc?g#?kwyKc6PvOv2S#;wLXqS
zNB)d5r&d8$@3;9=t}|<$e*zv*6wy7Q!$!c5CoGBEr`511Md%p*+Kl5epnzag2(j6*
z0845+XcU4j;cKpaca9n_b}Q&(-j@PYukVhnz#+k79lIR)>woxFU3K(h&n`LFbwt;{
zNyO7nLhPom=u*D}=Jl(xi-}S*l-~7pbxhVY6q2pH+I6O;5aDmQ&xM$h`1lqAnmTSg
zBM&3IdxQr>`*&N@p42q0H>%%pEYDg|p`HI&C=ZV_K<*OU!{6m~`;dOo;yOw(*|@&B
zinsfifm84%qFaw?VQI<A$$A5;-qmcT$agEypXvjAF#7_P|DhEuYyPe^Cqs(aIZ)E_
zY9+u5+64zpnoy9x_^PF>1DMyR4$opb^IkJ69xjD#2~@ksOJoDb=Tp?7!$*>UbuYZX
ze{}^LnqD6p0+u=zCM10Gf)ZQ@fIOmv+1Bsl3QyFqZv|Mp(7+*m!&B!I?~O#$w_!lg
zOEHtoZlG<VdJse6fjrkyzB&w~iM!#kO{n)4YE|s(yy%b@@<hz6x13;C2BSaoZ-rib
z>Q$H9$+daAeKZurv7fKw_+`uuz=rN<=Kw6~8@{4IX|aQNO%g1BS8^vu_<E2(J?Db`
zI#n*_Vp4w~B-P)u2n^^3O_}>wBC+4-?Ev?pJJ9;}t)Mqk_tg{@plQo#rbKx!0(6il
z=`77bp%sgOeZ7&`S0dQw&)X@Rxm|Z3@amVJe}QX3@si0-(`m<A7pWuXcZ*C&F@J4|
zclL8@#YYP$7od^ON)>%(_cMR{yv`HrRy8nTriLcr-54%cTA5iImO;gIa7ZEewXo0#
zOoQwX9;KZ|nNmV_5Y+0}XJ3hkmt&y$l4wuI-8vGQyM4{kT6e$6{<yOS7qI6s=VQIe
z)=R{YEaZQ;;=h01YS6SAFe+SlBHatn0!W6GQ{TNYt!sol;1Iw?GUjnm^pkS<bd;2U
zf5SbEi_FbM^#cbW1wxRX`sC0o&kv5YxeQ6k^6Cr>o3kV3j&`V^G<v2Cvu7|c2lrH7
zA|%}6+3&<<D9Y%lKNo1!{`q6g7voo+MxOBTClRtuycXJONGHEdOr&)$sP*Aq(oJ2X
zMxCcn2K{Io_ZE8yEkKT`G1@oGRB0kizwB)DH?J@e;8uUhM)V^OHEura|NRrFg<$sr
z#3(T<Zu>bW)c5)(=ASbP;3(?DKqCku)94mr$HT%#f|fj~7+Zm2(60X!^?h1hU0qq(
z?TSk{`&fv6!C$cKM7p`h22SGDHr=+@eFS@2@R4yo-jgj~s=GHRbWyDOhfRPEja*IC
zNj1Q=u#c5$SNXx87mEMR1B}~-$3zrD16pQ3%WV@sV$%~gH5~;<1I?Gj6dvfnXyCt?
zpZic;Mz*q`Pshl}G5!W`|MsjNB_EyVKW~!^Pq6X>YYr0ZE6y5FfPt6&l6V#M{R4WM
zh9Yo#LEhTfYSh?8)WmlH@IxH(cwEqZ^%{pNKA9CG6`H`+vPZno-<h812nqGOTDrXA
zj2F9N!M43uPh=aEd?7OZ{g;@y=1qi`mlt-bcY8CaUZWb0q*z=}x5dOw?8QwuBBYWQ
z1h9|-#%`c-j#)amy!zJ{BQK?}v9eJ?ViMA<zb+91Xc3D|z_ZlIl!&RR!OV<;4+N|2
z&b3h9!_tHjS!x_b|L=f!Bs=jfN#E7mNMs~CJWkh{*>(|TBD<5Tf+{Wb_XlIxLb!&9
z4Gf+VsT2hlkY-UF<B2_N-h#nLqa@S;x2N$tIy*tOXoQQ@2Lz&f0ieEXD`!sMCT#Q&
z8f>wLEc85ug@t5@Zeu9ob#uNQ)r(Yl<Z>iiS<<tp;A^h6S=RHP{N&hVel7DbjZz#P
zZ1I)I%($K5BHm7mcm-t&D69F2Jv^^m%OQz#+t{!~9<b7qBcR@ZC2lx(MP$8dlkT5j
zFky4ivG+VM%IJMa{IZ0|a;D6y1;s1&GcTf|HVl~%<bs>7lZX5Wj2$S;3YM#{A)h$y
zACmh=MHxMO(scA8Jw+`dpxf?h`Mv^Q0pzM7XT93@G#LJk`zpuvlC>MMI6GTMB$R$l
zM?sX7*to9-z+FSXZ}>FB@Ov4C#xPSv_4;z51}co8zyL{!x4_##u6z`!sq^wIpu8Z1
z4LvLSc;{IY56|!BxbFihk}e;#95<nErKDvlfwVqEtmD%M(OJK%WL|0Gr;l)^F4FBa
zQe<vW6Tv!*AFFx6vcrnDdZ77wZ56f{=lzATkxEXGB~YGmy->WvB4t?eyK$5SbJ{m$
zUwV2TT$N*FM0WB(%!MHS?Py;4U1qpORM3|Jpa6@J;<cHHc`ll;y}Zuz?XLM8X8zmG
zAw!*CtUON-ax4fHp=5ya;u9+?ct56M-tNBjkBt9fdxs4;E3Rh<POvAVp&pEBt_Bs0
zU9)@WrL3MEe`h_h`?)x84Z=US5g^@XFd|-p^I8+*evNnu|2m`&k1Nv--n|$0y1=_7
zMDEhxHf{@)Rg^GqpqgPIw|x0c=8nL%!ZTi$IJMWbYz$xxsym1i>6<?LVhzkxY^Vh9
zO#U934(*g)D`HKSK=XRf$^GN-jyRgvX3AmPjjf62k+iQ{4A`Ub_ItBxv{RlxSpQps
zp(`0FK@~ym;yR}Bt5|KNFAg(B7(vqiu!+2junN5vA@yw@1td`JD#`mV1YOLj*}(Cu
z2xTSmDQ5hA?<;y7C;NrlQ5AHqp*;$qb*Kig4P?IlEjRrOQh)9kgx~dG(O~f7{)Ud}
z(+YX^@17vqjhl3lETmX;tbN}MpTbix!SF6O%6O3ZWvqYj9!cuaMDu)Mp%rlZ(%w>Y
zGhm(65uJ3uz9xz?p!ug*gC5)#Mr$$t-G~=+DZ-A4{vafCAQ|9<iv|z15V-gfpN(fD
z`YCE%?qp?q{t%Xb-#$-oEMv`Fmt7$*UD9iJ51%|n`_cMZgjFIJsh|xlln!H&pNL7^
zn+z&KD+y{J6wZX8#%Mpd3#Aj*y%@i)zvSGu5Muru^EwrCyQSr8f(_9-G!t9ST*c{T
z?0G?%=Usgte;-lH-W~*c^Q~LEaC(A6!&uE)RSOhH+@zr$(c%}uXhjbgyO+1_p&~Fz
zK+Imh@k8kaOi&qbDL*?KA4vh@!1`N+9dclH;)T=crtv-fcscU{BcphZ?8wk?LqMQv
zc-#{v`X^7WHV85+)0X+Yi1*W}WV09}(GnYcj%H;H;LQXJUsBQg1z&l2WuQa)w6oBH
zj7;DhX?hEw)|HpFibFCS9QKGXZIpxM>Ga{`-#_7_&6GuOfSA?20Aa=XoD&Uz#t4MZ
z<C4XtJzyrpCo?tsf{g$){j)Mcj+W9f#iRdx*J%NsH4+{X1LF*#G~J389)Q+1mMEGC
zZ(R$ZHzzEbl<WP~=x+a(VNaxF!2j_oox=h4{KMw(D;B83-J55p?{D8cd;aXn>xqQ%
z+K*o}03Q6Z>}0&`Zy9~Ii4MdjpRT_wCMMT4SFNUtSHJZSLca+A0`a-%bgDL{WuNx+
z@yGw4k;PF_iMP9(M@L<%iYZt0^e`de;SWSF&+4kG6ixaDyZli0D%0=fie0Cd6y(tP
zq5l;0_z6O$ufQ=`&0#vqlo|d`qSdw98g35}_?Q^pdo9^If7upIf7{28AAgA{BSn3(
z3LK(-!W((4F(-|k8C_{()UUr>=zPD>(eZJ(lano`H=ltE6|*$8l7$vYUhCT-RU*tq
z+47f=CgTMswpqWuhFbyismbRppF1nXe)Mb<G?er)%y$bW%H_*x1lU~X_MH{UC!=E`
z-cvS3U;GPKmhd~t3B~L&7K6m1y4p~RaSM~32;rMjVN|V0VoZ!Lwhk|$cYr?rwUyOG
z!#oBTVY_Ls-nqxok3UiR+nSG_-_v{W$F}Ags>stc%@dao-2}IMmya$K((k-b#_|tZ
z_koi_e35PEW(S~pX!A85eR%A4%=_TVv}|(t-HX}Tt1$z6K;ZuURXsu1Go2UGY6`l>
zc9u#qCtLoZdV-GL<9beZpQN^U#I|pPRwvZ;f>j0R0t4~$DvwR3tMX-Ky84)m<Mtar
zct)r0%=7C)KhRDCge}DDkQ>|h?IHKFxVC7dZ+r(Zqi$TQU;1!=*$ZWgyIL~MlYeDJ
zs5HBxu%Y2upor<y<>dekU>yVVAM5d18V;5o;dVb23sG?FX9;<E9T9{6tef-j1qr&H
z$vwDRtdYRYKtS&~6SgfNu-!iU=F0@AupTI>-ytjFV&OUHkJ5*s?Lklu;9U4?oK&~P
zUZ@AY%+1NEs{EQFk^QrSC4w6UtOPYbuzn18G<P&}@^J}qH9c4`(NMAEPo_Ku9%75-
zR}nKfBEId#rCV|?ogVCct0eRLKQ2I662I)js4OD8`}fVW(#}ZGey2}6I@*sn%=-(u
zGR4br4vnHCRL{;&QNk=<ua&+#!4uHNB0LvPW+jc;&oRE;fz;{U=ywu)+Wp<@@bu}p
znlc#z6bG;rbl5tYjT68aU@N_<<204EmRv7xz4vNzo%c@2>jdxv=vjwSCg$cTLiQkd
z<A;8fmI)3~o5RMIt5L8H*-jNzirUriALj6Sp$V%??^G{5GeH_xP>+Kc&P5^L+Qyak
zg=B70;@x^N9R<@Ip?8IpLXX3Nl2?a8?;{5D)r~1n$PI3r9(mO|;9K@t(f5Kz?FX<c
zK?LY@#BIhs48VHr_;}xzKMp>L&?9ZsgtyoU4s1kr?gC_2sxGuEffLN+UhwV5u+b_!
z!W$Xp!Xv;pD$#hM^jHNQ`2yKcUkYgNZu=;GaLY;-WO)W@NIzf9^j;cRTW>X-zCYhM
zsWOQq&nWZr#EmY@ZJc2udp{2JZ{O}&uqqgx^&#HwEw$U31o$!C6FR^TjlXESz|`Th
zJc5wN1!Y`frpAs`7EU;xAMK?yO@9Cs4cVSEAyPsYbR)If{na{9YFkc7r6hw{lnfrH
zIIbKV_+UfK%8rVFN7Kx-;!lVaA!YRzeuMXDgxNbgo05vL`&C)QsVYGDkpry!<S?mW
z70U*;CWtwLLoiX#zNUFXxG5=<qnRYpZ8ls|EcKUa@AD{d<9<&~Q9YsR9vC2ci0xvb
zHPOLFSTt3ipILY{k?w2MOTQeGjEsti_X*|`)%8dFcH^0Q2PMATK?uMZPpF<{D@_yB
zKf6Ud<X6kffJn(HtZi&SM#9fTi4UOuttVw;q32+Kc<Y2}2<$x}aezNU>SJJt<Sk;m
zX3CV#2sHxa?gj1VoE%`7Lovu_+8+qTt{Cgyu01z#T!s-YfQ|m%t!X$JS3~~A2F4Gw
z@8!;bt0SH|;*Oe@d-1h8Yg?zpr!Zvdkp+9OMHxE_5$5CB#AOWxBzGcIzn3JPkY&eF
z(Gxx}a-&st_(E=Oh6-|izbxuPh}vM@VniSn5D|`*IY!?Kl<h?R0D`~QsAGTh1EdXP
z=IB@b(((!uJ);oB%-ph4(^hggI*KPO`u<B!!i~*n6%+<YTFK7&=|kqGKk#i}kqZdO
zt?}Ub@*cH<P!G4g-vRmrCc*gD0n%6Vm-NpYfP56eqiQ3)0xE{yXQKY^m{NPKx0cN;
zEIi4`$3~`6kbwcrEKX2tuWqj7=35IZNKblrs^YN*^e$n$rT*J^co%rsvzV+Y#3o=+
zWt}nqL6kDA*OTnu^%L~@+q%z`F#V$O=AEU!ltyMG24&Wh?g4od9o_kH%)oE(w-S5!
zLPWQ*tnA5SP7!J%*nELPCg`}5;8(l4aC83O)|d!oR7Cq+6G6%UH@m)fZ#bEt+dNHZ
z{@K-4__9e{TAK3GiV`a*c|m{xspp^a)~oO7>`1J0qQUZKre_3ac8T$*+v;yJoJkVc
z$atKuoJR+gg7<R5Fa_w;UZ@TAkGQC-KwbL!{{0m|%>5mUpq+W_i$i=5pP1a-#Jp#*
z7;4<?ETXRpSQ)rzs6HU~JC!3A`bFHGz+$qLf=0+R1E2E7@_KY^>}zv#{Fn{pe9tv?
z#;M}JOEcK1PXFdqZOj3q1tMNpUZ-oe?hBiaGa8HVdyT-;pY&pmTA!@<?HYNQ?CYP;
z<|DGTyslkpOSSLWGdwcVom7Q~5Ok*scoekfkVXKNx3qPy>b!JR=ue5~HzVLo{iIzE
z;^P8^&fzn<w3%;oytF9zMq5Q&>N(<Vt1XT<Y>A;*-}FiKoXSa02Serxn~hG`%K)ea
zqFAQke3NFe_RGKsy={d+5xrWo&Azz65XH&eL8ksxz(Ozt-9eS9C*b<DkAjhn@31*D
z2kfLD-M@?UnW;dbH&p&hylb?hg9R9COvwUF=K<HI(G1GQpu6<!8bA8(TxK+f<MT{6
zbA1t3HQ0T;5H>j~&$=lzO9t}f7a^5r<v~&pAQ=+s<)Qropx+j1*Ge4)OvA=m_F2Bh
zZes+rRn{}N@48Fwpri6*Rhgf6g7T`-{<Nq~!q3_IHlhzWjz6dPO;0F%|2c4GxLJV&
z1oj&M!Xb=m#CjU*ylZ_cP$rzaP^adHdHK}p-yp^0hH}dTS^*s^yhF#q6IBfcwIlPI
z$gF;Nw>h6pCB#}EJ^tCm=!AhtQGiV_FYx~{!cu<elq>1TbNok0kuhodOyB&KceDNg
zYiVr3xQ43EJRoqy<STFv`HGfQm4QEdLEhJ74$t@xAFjQ==HWT^mOCjf=k#FW5GH@{
z{Kb;ZtjsU|^bf)(47w(Z)z$D1TYOG#r`R($1<yNPoAM3(SIX0qfe*kTY%U3Ttes_?
zZ-OWG!utx6>Uo0HKn*&`S_;ml5h{v_td!f*Z;kCH>#fzRDk}2xb7P}p?=0mG4G*iS
zgclX&r`9>)`!IDZzI~&^a~lDu3M%B77(Dy(3RntTLT+#~<?KT!`XFBbf*r6VXsACA
zjSN>+!nl&xU9Bb<&a~y$uHx?*U^o=dkK)suW%>%PnDnGQ{?qRZn^#MmY*F{PNx!BQ
zFHgjfz~~KFqEPH;H##JW#F3htl$c$tTX1cC)o#-C&yI<Rd6pjj;xUa??BbC0Gz?Y~
zn~W<`zA8wlVMRy*PJ$rj^b$%k5L4Cb?X5RpHuS4Nd&n+hPF$?j&rh-^mCfi=JW4KJ
zptrZ1j!pwi_hUp$*tz886!6h8FYs(`c7VwN8nWPb{FnjN`7h~M)0&RW%Gho7bRk=L
zphHP{x}5y>5YX8GorAzl7Z&S|*8*f#cn+PzK=kO;e6F#;;as__sOU%-8w=`qOdkbn
zbMu4viW~mzA?2kX0|N+K@u;GtL2?w(VICA$6c=Y&Sy`DZ<o*GqlnMPqzaO#jInEAL
z$bGhywU`JA$}wkO{k0MWBz_>7f#(khO+Zr(OSFMs;LqUexpd$tfbCo;#S#d7NY^C<
zTy`x#L<l0wtHsYtYhZ(@bhP_D5sK-tb7Ir?j)n#u(Amz;)i1w~el9z4sQv{+bTcId
zyldQ>r!HCEzI^%eHX+s%H@iQB`U`us|D|u8aRFb_&*RD7(OK|%mFFuf$z^>D9yhv6
z&q(j@581&W2#B7zMwy;}K7*}V(f8@)wbxrG&KCCeg<r8OKn?`n`d3Y`Yc)^qqbH?2
z+5qLiPTi<gh2v}mZK6`Hiu|2}dwwUN(1BA}{*HZYtY1woOKNXiY;5UMHW5vyBe%wa
z;Na|<n}KgWHtTCAlPyQGKMu)zz-($|4y6LBPxGTE^y3qu;8#e>nWzwF=V&uoZ&0D_
zdE^JPSdbPFliX*};kW)YGYl4kFj*5tz*jG|(lXi*bZsvQKDRCB2Y1beK4k^5i)7n_
zhj{q-lSyPiZYEDHD#~vFMI;c;`p96=2A0f|C01qL{83)*b#<^kR9w4>lkjF^swPAs
zC+edZFt}s72<F~$rwjKDjDOBe|Jpsk%E8G>Ztw+uotyrK|NC<h-O%n?qWh~_TJcm|
zDB$PB|GjT~AW8Te2t0WC#z8!yn9T06b7uW@x*FtngguEgx19GVMUgA|3oEM0I1DOL
z)BAoIf5`Twhm2y|8jz#HUBLr{q3{=?#m{Xx)Hm@>vLj*_QQr%Pc9d7LgMS?ytQEdv
zeT`2Mee*YYSy>R~3g*#w_U!=s&FVMPU>F(e*Oirp<`G^A+y~=K;ko6y>3d<|S^*Q}
z`o%L}3Ol0r!cVz4Mn;G4S=o_W^@baQm$Neu4-{okE%1|H-lL3mXa0aZxR(C>M#ka^
z1HWKub6R2|G0E}@=D2vaQd_7^R79*{8IPRinC4(kfnI~$kTNOLsIcBBJc6_K`W1dA
zMID{tfl0hpiP_z1xHC0^&q_*|bQpY0ak9W!dPm-1JNA(38N~Xgea<s9PHX?W3$jP}
zT|edjV<1=zB3Zx|a!VrW>q#So`-WegADz39VwqYRdtqvovk#MPSOMcm_`y>S&X-j#
z@uy4LH@coYJ~*(0Qb6p&3(Pv8?MvxkXW?95ULP2GKH5KW^$Iyzh3AY}zlYO5xq4xp
zlXyPO`o$U$`?kgP70B44>ET`T@x&w{<*d-M!hbQ}{S+TuT*}^J2t1>n0VTV1=&lMt
z;Bj1ayS}Y9bLkDh=TvC!@1F}(<0MN=To#)%ma^YRRl!8UV`49MPVI5q!@vg+ZvLIJ
z5wKYD%QA%@qz$eN0?bU#<tLGu5c+gzv+C3H7ixgr!`oM;bqd(kg;fNuefPGVKo73x
zF}^;#3UD|OtP2@`0d^N6!NJ8Lkv+F}Y*(A*n_I#uiyPEh%gtl<C4cAGO`6yWEKv?g
z)Po*Mzy_Y1_ZVW=06udFp6C^_jEpoyyzqPstSwq*-<lX_aTRLi=C>#*sXS#PqYO0J
zqlf!T45A8=$?pFGGES;OhUW`daI!s&fG4Z*k{EDYSi+`8A-v)*17}xeMYENh_#1)r
z4g<3*iILrr7QiqYi#kZP%~(}q`U?d>7}i(dF|_JGCB=%{+p&3dmH*&d6Y%^&bnGj)
zCWGN<>G={3*l4-f?l;3^kDZHy+~6b1x7i)AZ2=;7P{B(mfWAb;0w92zj`+8RmRJ6+
zaL?zIy$xZ%14NFdk7K?s?LbI)eD03pnmShb>w3rtf>vUaOej)lBZ2D%j<l%qGxv-*
zx_G(BsF;|D2nUC9bzqA9vjJ6L-$6ugTKvilctf0t-Gn!Qi*;ac51$L<<>XnyVc%-_
zL9`Fn(C8owtf9WqjR_M2naFSmwBTg#dv*Xt@bOw`A%lEKdQkh)ZLsk%b1E;XyqW}q
z<^}vj0;mm4%-fdsR8?a<f0w6Z=TcHpS-p83YzU3upcx<i^<;_6PPTZ0SG2K<pC9F6
zA>4f9Y|D1+9xoFTH9jSRe+o_j`$&swD1FPTcR&J!3Q4oUkfn5FcE`z9?lTyM<OQ-K
zkn4o*%p=RDY?7UBYZBREOe2mCwj!|fcs?`4$k>xx6@C32*gl^uIdSk03ftwN#qJb7
z*V^)EycijlzCl%GRT~>}#`b6wOaux7j#!2yK9h{mr!j9lL3yw8;@bldM^k=h3;I@F
zTkGfRduM6&c3h#my2@_vkk}Qe{UWTPC^-MM`Anl@-_(>+V6^uKyW6ONfyqQ4p~ssF
z@GX>bl)CwdQO416EA+7C+sOP2F-518%eo@$v)>l4x+mEzMhfCBHuZf%n#jI?7riGE
zCizqNVH5@b9kTOTAt9lyvv_4s$Fjzd;-8-Yl+L(3S7S;(Eh4?2ZCXE3yXKYhwP$#b
z&chMIjj$uv+HsT5_N<Q_PkgQlIeyOl?#wG-=SNXwpY2*_&HM?~cvvtwC^yh41Qj^h
z;*E_pwrA&8?X|MAsr3Z0j!4`e7B^ofu{>jd*BO(Efly1J+t}Egzp-fyX%_J=^RMzx
z+*>^GXwo9!yim6R<H{Un-ONixvMJM7hG#8Wzbh3i|MLeF?*xUP-kwdm#GV&wJm~Fa
zR#8)W%5}c;a3!@(ilJjhCheN9*aykBgm~gi?nzm~9ur)4_{la;^vl_mFMpohL(cg8
zN#XyXCT1G=+~NEUPe%_0gPfVUYtifivYS{;CAEO8KEwD^>eJu9ES^1P0=LheJ>^Jw
zz)VNQ^zY8-r^U}vZC`SWzeily|DJf+(J2MWzx6GO;<8Fu!R$wg>~`>n-r@8hBML6d
zWW>aXxwID5l4X1PXZ+^~mu2xzl}~s7fA6Kat;nm{SV}hF^1$|jS;Q!KJY^%%qY6(5
z>H2!XWw_(bGb8)5#IAWpcS^&myJS3}sjmM&k0BM}FfUGLZr?LV@8`xcg%?b?{(Iv8
zo)G8^-kwoOu>J4-{r4tQQMJPS{_{(?G;}4LIm;aKOHFuErKn>s<vH<jaL%T1?gT6>
z==+E=yJN!(cRfAnmJ5C^Jm`r6q>3%u-PTqyiTR(~Lf!z*mj^qdqMqsbsa#Ts%eY7=
zDyHV{TnnHOpExF$qE@mLWnIfD^Vv|RZ1doQ_V}h)WJPv-n%B3j$oEn-f{#5yTmHlb
zr(PDRCQOHGPO849w)#SM>^b=@T#icUY(M(o-PeC|x}l!)n3?G%JX}rBWt0pfBWCp&
z@~X?&8(5U(HCiS5ZTB7jRuC+dFRpzI<A~rm$sr+8As825bx^jCU8E>2`?{K|P#_zz
z6lbqyfAsjBrn<Pyv)a<MjE2%~Ax0t+PreH}9%4a3T?5MI>FJC2&CN56T}>Eo-Y~Je
zh-elORkWVjiuDbFY;CFVu&}DK{6w*bmrVEZ7EDy5V}KT;uI|{M6PbDwMnXtXYp0Z~
zQf@g&jhd7~do&nrY*=_hbZ35RC~I9>a~eedSy+g^OGtB98uV>fYup1H81D6?6T;_B
zK~YOggDZCXO|grgE>R_Y#7Yw^d08djF<%MO|7Mr9Ul4@)DdhM|{6M@u*AlZj1_lS@
zDCLsBVJhJN8#7GTqs=}vdcMS+Fg@J}Qk4-!xSt(fqe~eyl0+?qq#AIo+4uCy1oeWm
zjr3Sel)q$4zZ(M}|3#>Xq^pJCbh^Oyw+FL7;)4qKVbpNR5xNM5U4>Q~yharlOrd@q
z#<auPSsbtS!(B{{vZscUDq21CyGIr8Ymk@4V)ULL+Ok(P*jL@!vQAp2>)Z?3Aw60d
zJMi|xjm5a23|&-~l!nLtG_KEb<9<O<uZ*sJluH2~?Pamz+-sa5-asz7Gl?G=Kl6UC
zrVtCY;AcjEkfLG=)9gA~$y?puddr>~D65!MWA@6BomG-e`2LNveN4p7j&Qcb>3Y;o
z+Vaf+g@KqpmZ-!JyVRpO$8A<gIRa}r+zH#A%MB}%ql(*PcO;`G5-z{;V4TB|sp=ao
zZxV9MeJr6l`o60D_Jq*q?f?TD_Ay51E~63N{NI=Hna&fRD)?U&#`*CjW*+PPD0;_k
zTRLEv@hs|Igjn9!+{}Kn%g{ciPzRDTb)80!rH;Kc&zv?-3Zn}46-*a%P4u<E_tQs%
z?O50R#Z!14|2oQYy_DljR8iACCahY!e5O(9bYX`*+RgXR)VSf25!Fe-eyn(e(7#pb
ziwcv10?Cp&-mSkgcwS*)ht7YJcLVcUpPYHtgrC<{a{tzZg>v2Jg_2rHr{`YN*oPN}
zNtKI#-?e8N;bM^qoz|Hc%pYDYMj6?&H}t=cn3U#3!NtAa1jLL@TCp=XX``G^&hQbU
z{A}m7RZXPY_PhsPy=-WknfFCLog)v=MYE2(t0Wzc4>~EqTGP1cqFva`)U!*YYTNlY
zA_n)6?VVlTajndtIm3*VfP09(fly4d{@I)Rmo`u1f7Q%(rNn{-Dd}}tQ&xx1q&MEy
z)ap;Jd+L2^Z)FZOPSNG`Xy)QcpdxT}_+*K<c#=b!05r{+>JV<s5s@*qU`afrR(fjh
zyZf@7L`Ug5#k2?ayvm+%7yOL7xum(=_%aW%!5f3cSY6~UH#48`c`sRpW0Z^AJSXb6
z44Zd58fVAr`H9o(k+V^qnxDRRvRSVe(C3BLAEjo-jwonUNSS^`=q}AzCDq%vWCZsA
z;@=m{9Zd<N>rfya-eB|of|-X`w6L|Qz<FVF#w$55V3ZJ-L$XE}A;k+3B2FsnTW|U=
z@2yYO(DPOOkqx~*>nT;FE2A{`W4590AI15lf9COqMx`0KPmf0NP}SGQ2p=@X><iiQ
zm6;n5V~4@Fwj20(rb6o@^?*>V=@H3x)1F`{90tw$9|MceZ(JyP3sW~@1wb;`U$Aq6
zJwVr#p*H0oXMW|^>0dvp>vHIJ<num$+-6|LM)B}Kq!ipI`hFYpjFsosaL{ZNeyINa
zYlYmcW%}dE$qDXs78n%60n79IrMZQ9@oHx`vz$Wr-Fx>Ec&5&C8vn>$XM6Z~)5s>b
zG}3!%h(0NawC4Cq{PjAoleX<y{&83<%`v(-6s!b`v_cHq%F^cYR>`H256^=H++bHa
zP2%5gijK}|=%!EzBOy)%oS&DsE{mFm)Ur^cF*aWAm9(b#K;qDQ5%2Qotjy$?{bldI
z`QqPKa3Ln<tNJdiJ7JA>8=)aFzSW-|MRhP7ZL@GN`ba%|^5`dAa*2xNC%MJMp(+Sp
zRUwvF(w-c2?^M6w<KwMMX~3k#3FrxRfZeH2{JKws`u40<8y`-5LAqeuSbuDG9UdNz
z2sKWQYxVHRsAMa3pBcN>lc1Y^E(r8&y8hx7m6iA9pFUT8A>$VHn3>K3q%XD*A^F-&
zOb&Q?SpICu>^718*X+)99CV$<|59h(OiD^B@@_%-*4--N9{gCQFt_k_SRiyT5V_MD
z%=^;aR9ziN<~w3Oj$_W&N`soD-9MK%Hv<9`^YX11W*1r6I5_)A!n_m2#o9q2q>wzP
zqNElP`A<zs>TRw6leM#o+!`A8hqYBzVL?G%wxzl0jZFF9fJyb~16N_;#@U%0WSm2f
zo=G}k$9t~&*2zQ_#4`!`K~&`au8GHuAk9mv{Cx4Ypd_i@J95DGfaP@??wv+;g;#^8
zIW$FCPBco3V);Q|bUGVKm~1A-KNPZY2-NIMBF~2|R-9MXI7RCs)4tnIwa(Rh-bfpf
z-4sZ5uW&9NmhT=@9S)#z)0Q_>Q6+sZ!5%%3m?IFBJ@w6))u*c^YE`*|y)?<|*W(70
zNH-j$Pja*F#oI&Oek!C|(@8N}>%;B#3Wx8H?MCwl!t<ty_nTt8Yf64C$#eU3IK4b$
z&F4vMOVj$9*fW&m*D{AZsNxVx=Nyw7f16bOZ{?8ilpYN(Q*vabk6>^ps*m1`IWR9U
zQK*Lm$;h8yk@IU@aFTyvk31Osc~6x^+CEoKf1LiM^Orh977gT)+38(X{J4>D%I}PB
zt5fJmAI|V6-Zpf^_T?`8beYzk@ln?E$j~E=I0KFlkxu>R&KeSfeEbAkJZ-}g{kzPY
z@`40$)Q!?qIMSUmw|cNDy}eP9zR^(%g8Nhq3{WDSvOX%WY!uN_HVP~gxN`k|979Y>
zFxL0mGL7qk@Nuts$3#kL7+R3$%Fc5?Jz_+OWcn2+j0;2Y2W?f~iC$vF`8(d_v;Wb4
z-NU6))RD!dFSN6A)95oIENAa#a4H%7&O5rXFmm`2`X>fcTeIEGlZ!SIbd;44%{AOA
zlV6Hpf|zVuw|$Y*qr`vxFy{5I?4&j7lkmt3?TvU_)Y2ISo-F>1g1YZwBR@qi)&kZ{
z%-`BuZy56^EAtp3%e3}F^d86Q=eUHSgVL?jB4gLXT4C}SnKs^d!ZXgcdP<BYB!K&8
zK`*;WMN{e~{($_dhZ>GbpV|+frIbv87O7WewRMNHFZfD~R*uk}3(-aX8W>Jy<lk|7
zK2y45ag<-F!0aswqBf49VaOe5a6Vt%|K$jk50Jzrhcv9uxOtOP!lRXx)KCx!+<kFz
z@w&RY5L%+i+itH-3}`{Va<uZqxF{>@i@vE9<)cT>RdsO55~b3We6OMFB(0ZDFI97o
z|6RZbqOs!PTY1;Sm_)sgd2RYnM+jb>xXc3__%LIxF9FTJm6$~Sli8Q{Yf|!EkB*-1
zFA^!TtEj#R4DqC*p?M$spw~wf0C$hd25@5mPT(;*g-qw;VvThM=CJqgMSjQuPYpxs
z38TC{a-^b^5FXvh+X)%9l_g2PdV4|Tc|YzGf9dy#yg(>!YrN~Vt@SR~;|H&}Q;kJM
z6xjo2b~NI?xyqyg)B|hbKSA+jr(>E_={rWPpwM&nJ!5G97ez|OjoUZJp^ND1aIm-c
zmuyY91ru9p>K`C%e9<dsQ%Z=3_c1^b7|ej=V^yjp5F44<xx9WU(Vv5(_5yZmu+w*|
zuQ&M{V-sL~msy{;<7a;lF~abRR$5j%QY!0D@s|kkxw*Ln-d1T{1wV-#(^n>@ON(WZ
z1&qJ!?AloG2L~5evLn#X$gvN3#wD+<uHVHWr=g)YFy!A+eap?uj8BMLQB`s^ld;~q
z5fo^0FwF%5Bfc?9rT+H0#tRf|F4Kn*`@HNKP$0+ow@-D~WR^V<WVsL5AP?5$z?wuq
zI{qb3XVTDX^UekwQsx#G$}18`3GQp@C<op8F<ZR(rswtJP9A4;NZ?z3NJp0w<GaFb
zC65z8ihJo7GuGTABd4<eckOBB%@f$z%-P42Ps1cQV<KZB9EvM&?%jK1Wf3MTx4FE0
z4FSxJhv&Y50p-I0<5z~3uTTpjj^%%STGUllj!IPjS({OqoeqY^L21u<lfQob{MyQ-
zr@afVBQ8AJTUUvmUYbO94^(smySKLPr-*uckMSJmTC<@DOTRD(jGFrPN8j=+k=deo
zeupWv5*|M0b+)GBkHaK0aS*eb!H)no5thqz#dbLtq@xJYz~Hlh?3c#Ii=(V)(0bc-
z*<q{mCf~y$=V<NFeWp`bRkgL*LPs^4QIG|BWLGO4l%}|GV(Ci#_TWdKtDJOne7s6R
zTw+pJ`hQ%2%y}FPjGF}!=0+x8vsqrf4G;g6q^7Cb@;Fc-v#YBE%xjb%MUNW1Sk??i
z4p{PF&1h?DlO-op&i0!bC&Uwh7g;+UEiitiWj;>4)Pe{Wlg2I^$9F)Z1+6z6Yeml<
z67=kCu5M!83IJSlac(X{DHl##RC50KIb-IB5BHD8p0Hc)w}r8>aV#$UdBkuM&2Fg)
z<Q|JRyt&KD($AjF%Q0gM>Q3(m7tBT4R?&x^qoLh<Ve=?4{r%TAq6){c8nEsC))w~Q
z@#B40AFi?;i{-U7z54G>6@~5vv9H{X=A)KG{QteFs;ES~7KW-KSa$Afj)D4xn$;QZ
zHzCFOuuUBEI&Nxo<Pj|g)swo$l7OdD5AFtNZoMng<ql}p2bVkbqKWab;liT)!O4M>
zb(3x0Cj7E`e&hiSqfz^G%hjkmIXrq-xh)5#;}3BH7z;P-cpI_`47Wn%foKCA(121T
zGx?N=(bm7IZ*a2nRTW$sm_rN)FqQ80?|3|kW4)i+`iMT?>1Y8xM20F-3U&}ZTlSUv
zTa_DvfXPSW{cTwzCKvSZxL7dwVJhJAZ*{!84vGorCMaiLpkR|H3ps0e;_cLza5Idr
zYJE!lK}U;w<44=)SI~e$D5QWV{`amfDkfS29CG}(m;Jynxguiz`PQ!tQyI7tL&J>O
zh?20>rds!dq{zAFZx1itzc*=I<Kf{3@xHFV#B{B#LqR2NR0qHPrmdM+a8g0}hia#{
z8be*9=~Z8opw7JR1xwTWdxi+=PXAz>nE>tM$KKot(3>pbyZxD;mzR=JeIEYg5kqOx
zRZsdX{dwG}6|gDP)Ri|@H$Cp*)jAy5!l#0Sr$ij=+H0uFTfKIn^%YH5GH2NdfE+bE
z0vxD<v`Y;cePv=|BQe8*|B_U!?JNvoBE?CTBU6Kk5yvUQ)VsUe!EUmu_90v2g=juL
zfwydvb*2>cuDd$dT5aq3HtXwN^Nm$k;ixKysnx(T^gt0&T5k`&OEVB8MyVGYtdWyq
z<>Y=HnH>N3@R40V{}z510V?=7G`Cr~LSArSsiBc_S7r|LXu(TP(~<rDpPolQG+E=p
z#R><f=S#lDwq64msp~a3t1BDHeLJ}Qs22oyT~G29NT>NkdB}bM5%IeX{f(b6()MhL
znWt&%-0ez}U)=b93SDF1-p5thRl&2PeX)__B;StL51*|QC5v!*E=U>m$<#f0gE#3u
z=YXMkb3<Cdje9SeA+`Kk(NTzZ-Fu2un%)4>^_ICWs@Q^psst+0f8(%@wzr4sf;9JV
z+XF2Ql&^=HCa*oTNBdS$&&SQ65U%Xi+F#FbJD^aSG!64P(GYam8l-okI+E^}KK;Tk
zr=t;b^DW1vgObkbot)c9_;7RshjY;q(dZ##jZW5^haq-lf8K0<SJ9D9zUN|88LKYh
z`oheFdd}?o<BWcB<iYcsOw0{R#M8twt0>=M>a}!<H<>3X^q+q6uFNpfG`;_A{h>WY
zFE0i%I5c%vGHNQ)@!=XjQjS%bo>eo>c29CpWVzgBbvL(cNN%J$ap)cWVOv{(hpA;~
zs4a4NAL6G>nY3~7@WP{FZr%)YAtQF)8lhoQ=xMyT$miOEAZA*coWF~n528BKXMxsy
zuR+@tJ6TG)3zlMttW<351y6RQN9VWo;%t@c8wo9A%!NrkCVDjeQrC!WQvNXzNa7x?
zXxh3b<b`+U^h9#m17D?HRbbdr=vw88>h`-`==_9^j-rKW!j~5gaNQ((`=xeeG@d}p
zU0`@*EAEnFayBBIt}KVP@owVQKDFKWO)^5;qSR8}@-e5MZtHG6738+o*tz~Fa%^ML
zGYe+kT`f5ny@N4*nz>aA`B`+W%!_`fM6!w+g<EI{pdzorNE5DW4Si5`krIGEsCQ)7
zWx;_YsQJ3vjo@7AFYCD;+dDU76tY=X((xmXg9dqQl7<C1qA2((Fe_g^{xte;IJK+B
zrDyT?NkBfzAEnKZ$x82p=~^#r>|D&p$sJM!vI@y0J~u=-Jy1L==7Ie@F#QUB(o<Xe
zWlwW6<r7BV_IX9}F-NBvto!%r1*U93%Jyl_5*tw?SiH!O1wO>P*tqsqHpyf%OTzBG
zltqYG1HwBR+E*w^Zr#55^-#|qJ~9Z8gQ|CHt<16g)u1#k_QphGvuAdU@7Q#5qNar*
zHdoy7JJYcVjW-kJkX=-Rcs1+)+ZeZnu`u`+GKR2#byw2bhA%5j@s3YV81!oYoU|%W
zv+<icJ4?x)2c~?QKC-!J6g;Wr{le?1mkPZ7j4Y~C$Ia=+7vh>PG^K4WzZOdV+Y%ei
z;T{vJc5j6q>Ep+r{}|UmdC0z7sw<+~Gu%OrKzz4$@`g_G2KtduBFMpi>hQV_HTSqc
zj0j}Hi4(>#CArQIm<#BTnVWZHVmO??t@-?bnS|)A75ol3><0z3Qw6K(uUODvVUrK8
zv1RU5b4?Cb6KRFn?<cL+XBzAwz7lFhXvjvHlQ~()=j6%O3&<xM=jP0E;4mJoypt0~
zCLD-;;=<b|as7&2y0&U*ZMuinq5vtlhi$FH5cwu0yRs78@b=_}%V(t6d}@^n!(qkD
z0*+?5-e1i<w~)5i7AL1C1)3GVI;I!58<T?;ds7x?Yh7A#5Xd%hw>vzkA7i_Qj=uOS
z;o@l19zvn-<*7jk=HY>6y?#&YpyawGLmWfWJEuzp%<C;(?Q2_WY)0cM(cQyi%`>Bt
z+Hxu&hn^WVoqwrElHY-GXt{Twxv;ukY_suL?Y-1vLb^I53`Z(+=js~fRK7~r2?+oa
zaOVTwqo~s*-f_p53#L7S%iGiPIQfmjtXNeut!sAgSb%6b#M5o->e8ZP@E}v_=+Ph8
zcySQ9r$hGkmF<QlliJ#5zcs?lGPZbzO=|1b;?$GKESGvyL~;f}H@h`eyRbO-<#S%W
ziO)>(8sP_jY{qkUcjzD7*E+{cf)<@?6<IjqZ$Ow24k&qGL7BWUbld&e$lO)`_T(qb
z4t8yA;oY=jZB9c3NAL~6Hthr%=$`nzWbv5PKw!%O8JBx^p2_3RK(?GZjO*GQ$?V4a
zYjb8*_VWqyYie8Ll?Dm^YX$d(o+nn7B#;ul?VcV{64Nl7A0LkErD&ghi#(N(|GU$$
zrknBxu7YIrM3>wC+TW$c#ai3VBY^IKK$26-J5}Rw(X|H@rj92|ih$WM>poptexFQ5
zK{Ga{F<s}}@#}|%oSfaodc5M+3?3mN8bXJm<*7%(P6e}s)6<jR6B;PrKw0^3bJEhp
zpwi)B_gxMai9Gx7@;4QbexBLT4YdqgPEBtr+g|4}E~naJ7eV(sn{qU-74om}t+8{l
zjEoF$*IT!Dngl%`hVl<SE)Y%Js<yG&IU5_~*QO(0#m2;{<>gzSEY*GO#HzhjWeX1m
z4egQ#|L^64r$o=gPtSJ}$Bxh!76}6b1LbTgo1Fjt8XxXLfw<PE(H|OntTxDc(X259
zYUrpRY3RO)N!baF85*28<lp`m)b8ly?0hRokTT(PywDYV&pq;Efsl6AocsiGb%9d9
zvax~YWy<V>qJI!b14UN+qM+|z3RCv*xESYu+yACW2tjT#D{rJ+*7!9P4sTGT@p`VV
zu6m?K^>(%R-1r<p$w-s?^)oSag{PUAn_m_BDSKhd*;Vzp?qjpQj}AEJ-1hW0FN?&{
z(+H`BuF)+-*i~|=BGIu#W8G&W4t6-JscJu&?Bp6Jt9|**!4)viy@q-(tR22iuv+?0
z>D97#HoHN0Xw9I*9w+=jx6y+HGJBP@MN7-07xunQa4~_)&~CVIjrlVpw)`)lp*;Z6
z@?+PB$W?6$xm-s;w3mL51xlH)t*L6;iysN9U9dr?Hr*1hdeqs4`VEBFg^b->Gqq<&
z({~Q(6_DpI)d0<U@1J-&=A*?-F++(H09uB-RB6@a!BIYgzkV$D9PFJsyg~6=DiVE;
z!@I30+k30pgOceiyFA+1^HO*&Q3Zk<g0i0H{;O1Droc$zE_K{rbOPT7<Y}{TubDTB
zuT1x^lM<FL-zQbquVcW+<!O7lScXwmT~nir?ys(*h6aQR46n%2-D35yl*y*|t%gEw
z3=&IsU$EQ+@ag9T8SHx=7u&Hg(vBB3V2zif-TM(Q%T8?7`!D1@%IMt^+rJXRTg%!s
ztHHxACx1qK2PKg|_8KaR)<rO(uKr{O_&6`gmqH5&_rM{<<%nmGT=B1%GnUrfslB)!
zlQ@4q$JxMy%@&)I2FPvu-04qui#e`$^VPNlpMP}U3Wz_hjT@+9F{fM8z2#HeZzdn0
z*dv|q1C-Bsyr_;W7PFO*co!VrfU=euMybdJd_}8YKP{dh2y4jMpLu0xkFidC^p(B2
zxmUh&%8vQVxu^&a%9z3>vALi$@>45%d__;WVHYhwgH9QZ_uL{jRBt<G`_6etw-2Ml
zfL~3J5^h6FvDeL`?c%UW41|;$3%ODF5!0@Lqq?zGq{FYrc=B<*kZqBRg*=AWa-P$z
zWKpBOjJ$b7KX#EFsa8IEkUd)n#aNO!{n+g&!qCL?JO(@yjW29;CJJ~+@IAGS{i2oq
zqM8l2`}oyLp60)PjUs7Ja(yD1#!^M4+sk@9RvC1Jnz`I{wJ+xE_{8(e#?%AHg+i`s
zE1nwkRU?%!vfuV=oti*&sXg&?+<IqlbYN~}HTawYmtqH}a9ht`0ueS+ekb$tI!HY6
z-ZJj1KESwG!owP&w*0!Bn;_&cBGZsv5L72$<!~~ZRKl~X`ik`WsK`36sp~H-i<9=A
z$c9%FXP9bt&3=TD(I9Vo{yF7%b}K~bVNwW6-N!uIzKevnDJ#ed(uEoY20rWg9t}?t
z`6Jr1<XmAZE|TLJ68hu!R!r<vBp;Vsd}Ny8P_VkUw&!+J>_~VC_kj8{F~;v}=mS_2
z@h-*Y#jP`MO==W%hXX|csRt|*$07>F3oQu4E2}noJcfeAUDkPrzg5SC{$K?u7BKM~
z)bEmBdnR2*b&DWpO(u9Ep+3i_?<Tw_e#E$Hq2eeej;X8Jpx8Qb+MP&^)o`A0Tb+lP
zUi#)8dk#k~Q|Ol6-OulYbOTu^Dj4(G_e{rip7ph+zt8&VV0&gD=+b{GCiVfy_=hFc
zDlcVhrxtZUSDn|{31YV7nOivV^{+Q#&m|SegzQDMWLug?ZZVqJ?%smNd3OG9NQfsc
z@jWU^CcpAfl(dG-mUhQ;)|ms$>qc$tjtUA2I>{vtoUzg!=2{M$hcnglR*QAOn%aH)
z?kZWF1gWm$MHh6Qonx68F&WR9i)`xK?1i%Ys@7tlO9Y0Z^<<@E%gB@~HhFKylWztk
z+@w7-qo6@>{F@+uij~amxl)_n__bz5O7J*Rc_xM16>MQ6l?WO3w2FTfm8|TwYk}!}
zM_|SDZ8xy&09Q+Ku@An!zQ9^gR@JF<nHnK;#R3}Z%A9e|*U!A7_`r4`xOZ>bb7AOx
zW{%p4HQ)r=wQrdaJOcOHcUvGOa!RGytFqLF7{g%gJ{wzGEKJO6z8{LF%J=q<pTH9f
zj1qW#@N(ua-Yxs%)jRnkZ=fTHjEfRyx(RgkAgm>3-G&PN_|(fbGAhT+z{rfq3^&W~
ziB@&kOotOrn*#@aezYp+Q611y9%waf^p${Cd37B@FnQ5HEiTqZmkl9;6ub_!O=(i9
z;x`eu*k4OaOB0o+eL%iF=d>ieckk7X*~~RwiNWFfYx7k$BW|fZUzy-F;LI7${Q70g
z8WE}_r>$frR_<Lt*=kWW^Quu&mCAUD;HVJXczX?yPfG%52<duZySasc+nM3vMdp(c
zOBfLD9qNR#W=8OIPaQ1h?u9Cwj@QajXs)$8wi;q#h1zR56wZ{MZU>YBvJVdt)oU-#
znn`0(GFz$6)uq39tlntm#$T1uq0WvD<XeXG^m13umj`;9IzBD4XKIWz%8!EF&d#_t
zA+9~QJ_+;XDX=A3;(x(_!~GoX9YJBS!+y2h0Vg7tJolru!})MFO4Xu@-xdE?o~VY@
zb3OGX9f#>fveOaE-yRDKHIW0m5&q4$n&g-1b4-lW)vob7VIsGeT!L=gPUdzshZXKY
zQaSs!T2(N}kFOXXNg}o6-1=?3#5!g1FY;a|u(}{<mF{l9lXI>uKDXVY_@Mdnigl|(
z{pD!B0{|-c{*m`EGYpIj2=@;4_q^1SrFumD-MZypj0NZ6zk|$sAEC4^`EEU5;M@h;
zVtA4vzs`V-fbIqFd41Z`Zg`wAETwVb?=dGu`!!k!?ZkvMq!c49t@ila>L?J19Ux-*
zF(7b#(lINqR?Q3O*^4@04I=&ITQ3EERI$g}ru@VqaQD;+PnlgiCWorOHX6V@bR?8q
zYU0r&YE&d?!wL}wlVW*Qo;@9`32qzqZ}U5rrtFuQz<U&OolEm6JV=bTV;`?`_2c)X
ziXCoy&C12aB_u>KPUG>I$%JN1QwvD8&@A0o1nFT|Yc?E-R*S9(2)>&QB5ZD2fAIaV
zZ0Dle>@R{!*;IT3x$8e?cKA-01Soe;iGmaIN|OrMfvuVU1%ea?1tyUPk2z70>8#Ie
z@p6D4l9O*z#Ey*4Y^@s33{NE=FqyKyBr(B{0W=-;?Cf(CEvi?3nfvx0hN&VsDpKw<
zFnmA%0W$`o@Q)J#+<anq_`Jn1S0$mQFhASpG~LJ);O)?E3mTdqY-P5{x1k?WkXMaB
zA&&BZ>k=(S=I@_{y979v=2jEM_KJT7<J{cbAd2?P_8tur)2lQta-Zf0|LUu=Fp&Ya
z88ojiAElgBEZ=HS53m@=+rg%T+{MKQU49D@2R@Cs^Fl7yo;wyI$I6O#?;dOrqNcJ!
zLotISLYo6)na}_#53sEf6QiWn<q*<PdbC5et*a30<AZjv@C`vh?(UcxnTIy(w}lN#
zEEx-lIVxJ~USRNO-WNybu~UhSjuhNwG-l;tZUPp)4RnN0SzAT(I?YJ<-_`O738g}8
zX<+F?1QX{5fPOS62oNP|?GDsj+UgE0)S$nD-Lvg;ETC;5uCX?~gFssV==vu)TG<YF
zWA5<@#0&N?pog%=Ybz`Jx6en6TiRKoBGmQLkllNoLzD+K33cptWCQyo-^XP2J!@BA
zhUZOI+gV5G2=x$c)C!)oZorqkYLa6V3cf%%rvmKh+1pIX{yC4AA>Wjz?jV8=vEgRa
z`~<0g$Q-&>45dxS)^?v#Du^IvB$gNm7O6j8kywjxRLi?CT+O8;Zl^89n)h-#cRe_{
zeZ14$R+Byrj`$l{;0%(m!8DJZw!we$HsXXyjxa2A7vM9(+Bzj7M3!j|L$%Wai+8Ex
zZ=MgmuL{xBlphmVn3iV6Rrvs_?u^F-$PxF^n+1$XhU%`6`RiWQ2|p-gLPsb)C7evy
zk*A-6;SK;mx%KWgim=;8V(RL9W}Ag?Uj-((4zMw4>!roDIQOs5`o+hI?=?VmZOn#?
z5D`&Bn4uac5=@c9NJ$ymKI`Fxd4{hWeH~5M*^$QQ*r-=ticMBhTPs`{LD<EIBWCXN
z?$O;i6r%dVkNHmK<E8idRE8uDzhT(q8#J%nB%2J!BU-A<KfBwJUs(||s`9YXZ*3Q!
zb48}bi0O^Tqly@U^ITsG==G+velZwKEepQL$eTm-OM0|HWc!n5nWYUMnR^|jA+MX=
zB<HQJ9jS<#OMDP{`bqAUm?|c+uY4UI0RP3lXp&@bXBy3DPRZsjmp1(mnk?FkJSstg
zsRMPFmCJW$s#-a%w-*_7pJE_@uwtchk4}vGmuql9q^q-cadCK>msdg{YqCPl;0h3E
z@NQ+7l_ho<L#87G0Tta7DtbymZmHkJukewyY#&%%c5*|%hajKgsvb~QQMuN*eHpIz
z{2Bs33kU?Xm=C{;;$rS2!nz6ss*H>@5bE0=GThnIZOTf7JXh9oJe|&L;u^T~IHbdv
z!2NOm{rrQu{^al2sY-BIVTEL|5y6TEpa2f^syF)>2qh&70&aNn^1Y-bLt80rIawxl
z4)$oms`iC2bjIeYY9msGGN-Gh!Ss}t=%QeCZ6#V=&AsUEGc+A!8$5?ix9oVQgsr85
zc4wmq{jS#rE%km@$|b4Xq`dnPfV1bD7=rH-=c+xrWnSDfK>qywi;e&U1i{l_Y)d`R
zQ5JXE=a*V?*Xp(T%K9qIq|B2l)qq5G^kX5Z9(Ab1kwVbM(f*%a1L+Fx&hGAmc$Bn~
z8cqL<I|!KnK&bcu$?N1pOl+~#{uMi)d-%oX?7T>pY5g<5$M`WEh8Y=?latpT?j(kh
zs7?(Hcr(13nBeQc+l{MrGkbAS0$v2b-9kY^<j~G{%#3%p-^$8L2L=V%+u1#AqvH^W
zGO=bw<>ujmL+Ie(Fw^AGOPf_N{6Fz{m7Z5S;sjyN810bG0KWq@wNWKHPF}hfW{Eg>
zcVPhX;k$oBZwFY5hfW_+j;A#1gAS*RuxNOg3l#x_J+B?!aU?uB3cJOcM)dNVzLicx
zd_X^t9)<ASU7`{~9ax)?EYZ{hCJgE6v0)AgdEdA=f8hrigijxFlOCIskA(w4UQJZg
zO<(_QW)rhpij^H`ZIuUDx1G+ep$WVjl>V6h%x6+qF>ZJm#uMahmv0}$MjSU*fF(rd
zlRWAltr8>ePqs^b_q6R^6#frcZy8qQ8f^_P1p(<0ln|sFln@C)=|;Li8blhT!Jv^Y
zX^;l#Zc!R(r5ow)hHtvhIq#p3YhQb_7rJ=X6Zbvm7-NpX>?=m$u#X-={oZC-H00~>
z8mNQrq5o&@5q@EY`5JaB8~xeubC2SM*EV*{z@d<d5bHXJ{nPWDx?cy!3v+W|NR*PE
z{D8v><U2Y2`eLY6=EL{|ajW%)rBR=HuO#~47SbcVFc&>6ex7kN`yyu$wwr4=-nW+*
zon6dh?yI5-3dJ8ASjca%!qi#*+!fR#e~_};BrSY^1fpXl3-cI!e4;k#m}uOctqpn{
z?ZU6&N)I;sUFBXcNwh^u(I9N?OKRU8SZlcvc~U>@&E2+^&tNQU_Ws_|cRp0f59X&J
zO-vSAP!88Z#=+q^kA@)IkwGziZy{-6q1xKoK?~`d?~k-{LNJ*z%D;xS0oTCnNyN)P
zQHuU|d6{Hgo17ex25cFJZ))n!YRfc)B=z+5N>qIGtDE>RJqbADE7Q$gX{4}@TZlke
zl=s)yUI+%(g4b&~lF|Xn{=HhDzD9sF)RXWmKsFnJ=r36d{}O=4e1HaB??mZ|t%FQK
zMKp~TVafXn6=dX7|5R#nzv?DQDG8TnBV!DsGC+jN(h<CO-QP4}3PDHMaLz_ZX5tAx
z2ZW;b_8f8FCvqL{o9||K{rd$66<%}PdJR}oW1}ayssm?&gb<=g?T>Wj%yCa^TKYTp
z+KwUjZ`1XML9SbspwCZJ3`vQrFF0sKRfrts@K9%YnK9AZHt#RSh}PYF=rSg<E>)!9
zt#rAsX)$oQyR_KWLD`cL+cH8OpW{zX8^(YD9bhv)QW)`@+aJVR4NEwMiG0-C&>Zes
z1bnt2x%7Az)b{$50L7N%c&bKRq(<Cr1WIIG*z~q}AXaViIeGHlb1s~U8r0rC>)xp4
zC>0_VnnA(-$CAV^eY9Ox>0^zem-<H%BK{;1s6OoQUA&F3DWv4Q{hL(rj~yeHh-1pf
z_frUyUH?dLn_?ey52Z`|v#XbPjxPhmNYZ07a{9pT=fKv1nURs}si#_mNJd0(FotWS
zF#EH}AaGRKZ9o^=$s2y9RBVv;q=0wDv_)4RDK&y~@v#PCuGNl;kX`_#BXE&Sj+I)(
zPoZ0MgdtPl<E>xFJCB|4xDgC*ZvGMrRF~ywX}(hYys~0A<GFh3)9&!fbxb|huGZ>~
z3ufqDTpjWzry;lBA=7`!5wBe(mRJ5Fy|-4v!DMy_v@J!MA7~Xbd4-yX3?doZ`%gF!
zX^$R_O09pKFFnT)i3{18FSh3b)Cguc0r2GEu6`tRKt=`wuUi{7mNuw}!?+hT91k%0
zl5ZhYn6S2FA{OxvM-+2aq6sF_>$0ors<4-UqB0*AGke^$f*@>!E)cu_1M2$^Eicvn
zz-^Rq@B}65E~55~;+p&;%BNchG9nCpS*%3oO^4J4V@o3g%je^RGj(6f+@K|sycH&#
zUM9)}F|5qRmx_C{$<_mYSbjXZi1#b71l*2j&0i21%$cZ*_(4dW2j%-f&|k4v3<WmC
zm?%Swurx3bBK{g@0kA8ZF(~}wYeHwLZkDZ`-Pa_c$si#k+whVpo2WSdg80C_-;ALE
z;sx3$osbEzv9KJT&NyKUbZiU+upb2uAelhCF%uzT_2;6`pE~e+sOWRq7cl?(U`x_V
z?UH(#!oo?7*|sLV<GJkB<AIN1Eos!TE6iyh{F<JaGlDxxa0;Hi?I)EFxWQp)c@(!B
zS-uhD#O#(!r=Wq;+dp}gS`<i~;R~igGBatI3*Ln1k!*0kwB*0EC6uLec>Q{};MQj~
zEftcj0)N=rp?N%UbO6INhl=WVc4a;6Ec?)}-xrooi{U!h@N+=m4lX`ex=XG_rF`>j
zQsvL@TZny%JvROwPkHjMbT_ZK7bg$eS5EBi5%lzo8n6~xZ5Q?NEmqJG%V{(jH_e_~
zQ3T8QM?QIg_`V?6&m>ES#ES+KF3?sS$aEX0CF<@6|DP6sJWVk#m(xOp1WKJTF#gKR
zXYlor`15;&vkCVuR(Vy`G0erYi&5bQ4weQ2XboT_oj3vHM-HyBkqRNh1c3K+IydQT
zOD{i;o0}U?(b3VqV-Nf3T(!Z4I<W?k)N>JmHZyMuOcdt4`v@188s2BS0qlT!KHg9L
zaXaE!#SsY^&F=p0fQ69rc6WE!-<RKTDcFmHqoxbyc9R-@BsKiY#FaaGlEDZS3iL>t
zUlJk1jH+6n8tINCISuXuOcbKvtXH|PBe}Y|0Ppd&5YIA&j0SfA`IJy(PtMkME{uFE
zCNibesDJ)DJv(G&Ks)7gqwyLRaM%XysYb+YE^EL-2dH5$7B<Lv?{`aeI1>N9_tX8A
zDM?75F$b8Dykg(mMIg%hol2@_s`g0nPzI5EZs<*@Ktp{hoR@zcR=#wyJGXbR96RD#
zZU%|>nWcHIBbWW!g1=d34^e-0bb!a?^Ji1W6RV%+W->Ex0fr`HG#vg(2tP3CNYEKM
zjJLs~0i0Rna|qCo58sD#GrI+d;CoqCw*UOu=-N2D9wQSh(`|2$k<y6XWV;mWTgzA%
zAxbyA>R3_rQif%Iu&qi$hMO)bu*uL|B{(=xOFMzSL5h{-9FplxxnAD*2Z^m=#oyVY
z__ZW^NYmk-Tkn<&V6%mswPgQWFZM3xTL>*fk~_h(UQYLif^Q+1lDh*JGoP#>-rt&A
zoWsfsKr0R!B>KakaT$x$VPNZR?=@*wK+9Px)#dQ@jr&GfLix`9=b_(1CofIaYes_y
zxz74)6m5Cxx%-WNMT10^SQh2Rl{l7N<8b==&iw{Z=JC*N1$+xyYB_lAhn0+Y@lgU3
zfh{*vIq~|eSn7h1#+7y=FxgqRs8pcplJK4<OX%m&`$YI2*Q?kR>^rf^7eW-vSXg~F
zQ*9j`s0fyv4?-p)=m^T;)K(Sx)7H}+N#4hDo=nUEM?4=iP%vNPt|1UEEY@6=pD~p+
z7u8cfFdL4{Ff}x@y4v`R%3z{6uxz|FQ8>yDw3MGD%iAbWuS@4VzwOjo2^*1|$7?5+
zgI{D@#GNka24ACFL`D+{$ThaT-QUKJaG6ljLWj`KIwZBeor7tyO^u?QB!Sr^#A^GR
z7n%3YIm6W;=f~^KM;1O^J|X%)^(&nqwxE%4w<PnM$Jg|xkUz$V7k=NkE1e)>p#I&)
z4BYqkU#3Gr^{u0QB0as=R@s3$Q1r)q$qB{te?Oo0R}iK)gj=Y5pV}1ly24f5^iC2P
z8PqGr9JSzwp*lS@<d+Y^igQp?$jLoKJoeY1T<6hF&wC=SEn~!mVO>Egk&YGa{<^}c
z!=Uuz?c0NPfjw%4&i1J>;n5u8a$?#t8C69e<a3N3f8jr&(64vexy~@7oFBBvn7k8W
zpOm2%#TdO&z4Pw}gXQYQBGd=yf}b5_oUjlyv^g1}K>?vg@wqj%&nC3ALBKw=dWckC
z{Z;TuvT<wAPXZ!5hpS`HdvDOYM#g@_sty9LaO>dYkfrp-NzIfS?{%l8x$g)2wzC)>
zeBOD*iK7ir_-k8Y=;@I!-bI0<-IMB4w?+nxu*%saraMTXg+x8q2^Ob+r_bxcOVsfa
zlyZ3d=kEOpC<svCJY$LdgjAgW6%%|iLRF8{spIc(aljl3<Q&(=&??zk@UHKUrbqPa
z`8^WqH9rvQM#v)HF0Cv~4hy?qSP`yeui9H&<1E(dGs;z69lyglV*(?5aB8~OA$&ue
zFPV*+)%RY5sv_Y}a${mlx+8@kCUeqHYATWV3*R=tn&W*cI|Gg39=iAa{&7$j!5rJ)
zC1VNNS@j{cQN;Ui-ykU)<8Dv@8tAFgGt^rJOKKt`27gaMPggpYKE6EjGbbHy@SoAp
zdjXT}3Gcsa!J`bEZhhQNM~JkdB6I}Uo(;@wX+J#;$u1HS(a^{(YO?F;yJ90Rhg`|F
zmX`TVbBv$VR_1DhFUrxen_Qd+B#j1Nq^^jM%$acvxPXiqmI$}~%_SbjHE*&4SwxzL
z<(E6YK43u>oKY1kZY+BlXZcVb8yl%{PYZ0`Z(ZM@@Wv~Ol^QGx1<W*2V&Rb_hyJ<>
z5FLu+vQ)AsiI801*pWbYB>K4x8Yw}>7}<0%*rC!JoBI6#pq~>BP(lOg<175!`>g!~
zbq<V2ySmiC^gHA!fn}yR#VP(>fVqFZek|SHO`53X;_4Evtd01xJ&%r1w?TQuaL<`1
zN72i8FnW7^^F_|E#Ns_25nVYANn6`++{9fJ1mppZSk3q*^#N#v{2q#PFwaHcl9Aj6
z%P10kUu}|GFxH3Sws>)YYCYA2gXqK>uiu;s?PH^hBE~1Wi-@8lS=c+xU!Tyu2%D+B
zk(5^YG1ETed}jE7?#=*FQ9(Va)jHtbw-BX-rgZIy@IyMRc>Ts+k7|cmyx6k!S7BFq
zv*CCH^_vZcpPd0`gU4&PZT!2lzp@PlVt3ybP`P^zf#MnI3PWgSof8S`2smt*kuhCa
z4x{rCv$4?!f({i`knd1a$<bEHakss)39)M`>Oc=Nm%Au+2wdVs1OgTaxSJ3NtO-04
z=Oyv)eueIx#@*1$0cOtKp8F(TmMVe<a*T5y0m2HW1d#^1Ixr<94@))A6)L+lSMe8X
zg^3H$rGZ?^@GF&(KXsbii6hc`uzo;#59K|MJ9P!e5W82o^Xh_g?>JKQhrj0)W!vb8
zR|IOMsuh3R?Fg&9Maj9C>I(!un(j#Ro-Ps8Ud`V^2r#?eL9j7lPq}Q%8{S9QF4o#<
z=xB+cP?^6%tVQ;f#ZzLOWIa^HjZV$tzREnJJk7vEr9h=XL_5Z0m<|yOfhtMYgUsUr
zCf(dm5z&E1)3N#^{NV)QmC-!x>kMAAk^Gu{(Mzw0o-p!YY)6>W098wSk_|_Bnvls(
zyF=qtO~8kklj%Bp=5i~|q@*(O_mv4g)2`OF72hX$vw@QwibR4cu3{$@)DUlR8+{>2
z-*dqhkg(Fds^1iNFv3APzly;sgD!ZlvZ6|nLFF(%lpM6l!A{G>((6Y`=TB+29xUmy
zxQZK5a!ZqO&9?Qo+*f@_p@nz%C(@_XsMIcFy}rh~2riT@8SVnSN9>Ps*uGZHdZ7Bg
z)O}dJ*E6GxuQY%^JZ4cAC`nxyIQuy=9Ca0yB9$Zl*E?Z}SFR5bHTirkC#aMtpI#EP
ztxOw3_snM^vT%Gh-(%^%@%o+!59C?K@3{T_{W)ZCu6TVAfr#pAa5*}LNz0K!y_EQv
z<mjX(<Cdg&J#ygKfhPfR3dlDxJV#1m@mLo1Fi|iN_t0rsINmKgo;315rO(Q&=Nl&_
zS+$;af~*v56A>CTW+6;({odZs5JQ;vEwi(uhZLV&jwXVmY^J)RqVCx^;1Zaa*le1Q
zp`7j4tcaAP3VCeX+9rxrq@P=0lT>X?c!}2e-aIJk;fEIkBfry>@+8|{NK5YzPFo&f
zg@mOv-u?&cP(}RyJvAis7?<>>y{*d{#4p5}JI|+epo}Y;t{{m2>477O(cafyR#kKh
z%uI#e&GzW%DktjizSOt5Y_BW`lIdA-b?=gtVLj9w-o2tuOAANKxQjCRv=GZ<Us&1b
z*JlgmQ~{hikE_tHA1_gEVnF3=#;(WCS};~+u>~bnzkbt)DHMgLoEAH)j?PX{=0ZTx
zVg<JlQCvG+1Bg|d9X9+LE`I=zW>E6af;>ZOVEG-KT7cCqJv`virar9&eYp4e*avu$
zAbkS%;jdq*Ai9nQpE%i^Sb;0)Nw4Sny^fR_J(vZ4KHY!FYVGo@wEYk4^SatRo71Ih
zX9c*hwidBr^Xv$`@Ir15+^Hqo+1aqMu!@R}oNDwW;K8{+LBD8sx)WkN(wPKG2T>}@
z$<w(KNkl-POajDwM&2n$MMXV3J&k2cQdU!}?DpooY#?^9HUBG}DhA-Yv|Ru0eiAwY
zgt=b%$Ug1ubWAaP^4c56(=Tp+s4-9BiRuCiKiY^+$o~b{Uz?{0X8UxW^(R927F#u&
z^brz(-m>(nvuQI>vG{n@r+I;hy~*=P2NL@Z(s0@2*&x6=WoBmZ8Oy0!L5A#SdTXtX
z3qGzN6ZRbBVK~?!Wb1XQYcDwB$;b&Id#B2*Bb$fO26R?H5>{_crAEX|4^J{PF_pa?
z6d<|-Su$$cv$zCtT<(4=0yf1dDSDo|51|Gi3OpzWQHp<T`<M5q3ILympJ|JkU?HO8
z)cN?w*tEISRW%1eqkze*SGp>)NuheI0W_w$#@8oPqocref+KIEnTwYE)2+?T$jIoD
z>PkqeBTtV5e-uEo8X5&)CevK>SDX~pS1RxkWiQCpDO*78gQ^4xmAoH@dl$zlJlW6t
zsLGpSyb?>EZJydqK>h;saX23P3%|#EpfFhtqsaOCO+#0?+WMd<*vwBj3T^kGwm%6>
z6wWVD9osE34jv%0vGwKY9h5v^2^38+&kd__Z;p|{5c0qTL1Su^1YH!E>op2Yhcj>c
z9yEX(spr=6w0iN+*0xqqSxD0C<2c>b90?sZWfy<S2wI#o7ym;+P)uQAIXWDm^YZda
z>@1?lQG;&<EbA;H5TBJmX!-QV?n_6#7Swvi;ajgk4F^+M{8is(n^p#Z(b*Ps85;Zj
ztEZ=Y#^w&(R6#&)=+wP)AG8;ex*G@thaP!KA?e8H>$S9S3>o!8jF14k#D!<_+LlGt
zql;$64fju>h=JuYFaYhM`1lqAluLbmMNm;dVfOL%C7~*JuDZdO%-Fc7f4e@!_-JWO
z*3{I5H>d0^R=`IE>4-bK1{*7@rmAkix6BfcuG?x`v(eHJcGNb=#`M%+qX-GeDmZ9R
zatpl;uFTSu*L(T>s^cXk9tXz|Al$%`1BF}%Dj78W9NzM5P^iN{D@*qcA3p#s!+L^N
zdw}~zMYSf-<sUz=^>E$gt&wt16azhSk=E$c(Cwc|O)KBT9$=1*Yiv)Id0i#M03$hD
zd%^(8+>IH*(pjv!&+ZckKWP@aw`80}A>_{Ou&rcb(smr$7aA%LwqC8g*0T-P)62;c
z@=5NytD4G?Ci-@|WZ7RnH2n2LU0pqg*fh@kan7953df`NFz~|yTm&Ego@k1Mc)Q-P
zRlMAeGRcfQdWDg0lF+-@%8z$YKGm){+8O+no;0&wY_066V>;zMi``)LGoYZlO?Ige
zq@UE^q&2DTf9IEY1~+0Z_iax{OX8?nu8Sv2ptzP<&ARaRQHT`3E3Zo-Azd}zgnq0q
zvX>iOJGIuBp3ki@tMl>R<Bz4~IHpOv9?vBb6!QJ2E+!5lU~DM<B<?ti;trysV<!3C
z>g)Eqh*b^cttY14e;TyIg7q_>HnbVoVs!L^`ZQ0y^D>Sc=O(D^=B-unUq8n~oQ;LD
zJ&aG?Kk(Nh35&xB=ZV-hiYinqxWY-OJpPFNrt6aJeA7qlAfq3yoEA4VTHS<4yC%K9
zD*c)H?M<$|i2D1e)E{XT=c>2Ahf0xV<S$U=8^+kI&Xv?mG@WZE47Uea+<HHVw6Rd6
zgm^*p<W2Qq$%Ahz52{~OYtKe$I$vr~zHEVdu9#P*^IlnXTt|jkF}Z*n9mAv7hgY2&
zFbD@DL$fP?;g0L;@AtplyNV8diXB{U?rgtLkOLx-*wtU4grhtFd-0sx--jdPS_tP6
z;eQ`qYknft$T1X&BlmV|qI&lb@oM!Rq9)0mz}5M><QWn#(m7@3Sr$yDPQ|#H>n=P+
zHA!3}Q8s#mA8MRbr)sP4@>QSII9oPd%Vvbg(B|&a6HDp>!&~oP&?4SnIXIIM<<O@`
z3GU!{;g3A{R-@}8RLVxXJ6wKe(bPM>qN`JAPUN+y>gL9N$r{Oa2g^xwo=km`Gv_ys
z{HQVR7Y}s`cRuO#e}0Unpz%=B%Ei54*Gt%nBSQQw`P(B-jUt9zM8;SjRN^4?@a~_c
zC-rWJv1EXbU}UqiRQaWW;w#CTs%*c?ELWAcKPGzl<5<3MLRL4EF9|`a<JIwVzT}yS
z2{k1(Y($8hoaAs>es*?x9x;MVxAD*L<c7J~R<*s9vcm?`O|Mw0bZc8180P_LbQjss
zPKx-$i`BI)RyM|rVPcRNz;A@dHtI9|gVnXnOp_^07X$+NwjDT>fZ0O)_L_LkFmze^
zO<t&)ktat-_v4}|`|<Q|t$7N#o!fI2JHymaqBFUyL#m;`;BIG^1L$lYj-b5Jtacp@
z%EC}PV}X$pm((cd>-~`9P<g~NyJJI3_xj{Y-w!GBo}L-~>sUe+PH%vpcHcb(I>uBs
zkx%l`Bkxosq)%{YtUvu(E03@m8Uz^L?jAS#$FU~xjK7Tpksi!)j_9H`abcWug7xJw
zxq#qs&rj#G1sMFNrtcRM??C##8K3fp+etzp8R$CzVHcg+LB*y7sR%An`RPT=#x|U7
z<JV((!A6zbX0i?rCFSQDI&SNmOJER7%8YTfw}>w2(cpAKN#%%m#P?KDPi@4$Ebs)g
zKV_x~rY!J8yF0rDT@DX_t!=~TxS~Y`G^8m_Xr_-J>^STf&o=Q)|LlcyoT;o0XnZkN
zdhH$TBt4?QpHMvg6->T{JH)mdlso*6T_d-KleV_DHR-VX3_WA_IZ)j2(dAeRSIr>S
zFg)!M3}A3R@DEF!u6Y;Jl+t(gJx7c}21<ZD6(%=a<lE`>X67%dsw!htRc?bU(hO_?
zuD+XrCYG6*8G8P&0gF%pZSd^qam~6z^-05Vc2*Gn9#v#7J>pW*%8Sco_mtOH*v`pL
zqtwwrMEPJwgnlN|Yh;;~k(J_s+BF%mBMLPwtB9;D7kx3V?<7s6q&cU<<zX8HPFVJI
zEz`AwZNUcxy%G=7nCRFWUba){wg9WBt>=U0Jhj=w=+C~G6=Y<9Ie4wrNsHJ)0|<dj
z9=u(yPl{_Xpq>QuJ3<n0m<y&(vGvJcji_Pn{f9bmEb7kx<~9^7?jg+IzJUnVi|m6m
zn}r`SiJ+Fiey0XAJ6znA-*n3FYLaeEIsctY*#gOxjna_%naxbKL-|@}AWZfEC=7%q
zD5;Ml?>L~K+=W4%S$D{gKof<BK~q@<!*c9iT(lws^&Pp{ri%f;$^x)w1F=+h4>J^`
zFx+(q!T#wdWY~eq7693Y{GJAr;@*b|5N46)++6YPYhmLj@eIg6gUl^%p6ajkO4_6E
ze<uJ>2iq#3=qKcSrKaF|J3NiFEoA-n%}iC7cY8%1c&lL2mDgl`%i5akTlqn4G}yEt
zqil-+2>9P3fI&y9|MCC4M8-~6=AO3Q%@gyuYxhJEp;7xb{o7|1E8S0#UoU?%%0u`0
z#7^pm{_1;BCx<C2yC$CL>Bu@JfFxVn%~_ouM$e!e2KqT=TSQ(Yi-wkbx<2KUuCgrO
z&p&YzBqQo(o&IrAjUhF;=5EXUK<^T#SVK$&GF3#>0qQoUY5$iIp~rKZuM_v!>N6{J
zl#MX=dwPB#drnU^2y1V2UF!RIUXs5uMyc>^U$m{SvU<@W1hN;VA3@j0wBeD<#^q9=
zErz^4XQv|wlpNELGp!YaS%nEvk*mz;iVwFS=Ja!8;UAnB^VT-8(X+zW$yVlXKJ%8C
zX{)RAXyF#=m9|a{vCy(c$<je$+Jx4P(D61OOMVs>;`D7<&yOJ)XXl_tK_4G|9rK-b
z+S-OMY?XWUuSB0nMMQ5GP2YLiEF^wnPk8GV`^oUh$?nw3WA}Ha@|${n8^8FIZE1uP
z<Y&uaNNp~>;!qYO4})1z{`I8fPPy6L3EWorGGgOkI5;OB{=R=_Z4ygikqQ^h!Blo{
zdt%9COj|H@w%8d*NS{LLIG%0W`ad%-(^+Tx@vppI&+B4h)W<Znn_Q00XMWOWw%ene
z>@0Rf$Al<IDyb=wkBt`R2R(GTAP0k^@Gk43J}{A7);IFdo#HotldGwnK<xwGuzh4g
zwkIt{5G<IMs_!9eZ>`r9n{A4Om6yxVpYSx#y7iTau2oBtg#dYKuPM&Msj!3Tn=89V
z;_4C?>$QtvpF}vy?H^#S^*z*;`;4`R{ptMVcy$U-LJwD11oa5~!oTNK+Z2=gp`26v
z$N&SrtTirvZDt8n{!7i0$|}qn+x@-a6-bf=XRCg;<OB|P>&@JB$X#|;NR?0!85-hz
zTgbXmBq;R=mu&N`GEk89u3k~}xq6sPi}5$p{uT=Qrp-kQ<*He#e3s6exl2L%4=vWf
z0cTD6$5G_^5T)rA{rKm$JSF0LYtt!-@hamj*;mKnWNbxj#BV2>8~q4`J72XtdONm=
zv8Nu0yl`*QZbFuoKrDmG8r;u5aEYA{rX|J^6nQg+-VYQ*k~_zIyvC~8MM_yQVI{@y
z<X>G&At@S(uG^*6hEMso=#PO&j$14aJwcbA8+jPYY(wh#7e-qCVS}K1Qr$X(Dn6nt
z7BTPEvFo4nkalk@Ci9zD&2H6>f9q?ALBV{wrOx~}ov#bhSCx$OP(N7<$hwmx5ew7c
zK2+WE`+^njk%Nck)A7mX4$X`#qf8W)^3fM>cJG%oxFVN3|6VKZ%#6sQ(T&XMY9>9~
zG~G|GcBROaRN9`fevVl;A7{!d-<Bt}&Jk&2jzy6DZF5hg{^PbFF1pxs<8v&mL(D@b
zhyj6r6Aul6$JIGU2L&ZBox@0l)UysS!5b?xP#s9VsCB&~kNpn3z$nuVn3zH?6*N=m
zswW*d&l7Cg<a(Bb`;El@(CWObeArVr6P(3b!2QF}mfr~#KjQcBcZrbnd`b1y%%!?X
zhw_x9(A#oNGtP`dq|!Z0g&Qqtk7-sMLKnFN5*X{5oFlh1Tp4TO18^Dt&Y4um|Ev1&
zvjCl3M0SKZHH%-3g%@dkE5Z1t##1<z8yuB;>AI_~0yGzd@)?WYwrk*knvl6Tw(N_e
z3~G05GnHi24u)&0)d`S{f)`frxv`7rt7DmY7=5lJpD)@Pk?*}#%`yHiDXb}uwTFZ<
zv-fup=?&YXJyTf&mQO!c4F-DUHe`xTpUIQIp6Jlo>*=hi-O47k52p1G#4W3=N`lAu
z+U=Iy%iY~wG;HgkN2P-VIg6F&UYwwKakP6!!sVeuumV8d_Z{Q;DPqsdx?-K`C5ulV
ze$RetXqbWl2GX(g?5g9<r8Ga0V=l(^F9uf==dVaWu=BHb@?>GlBB}$6lq1dL=<52Z
z&8!3Hc-EBL;aR*%j8n52J3qDPgh7q-qwT?Ob=v+KEPY+QsZY%m?d-tXoS&t)S4)rn
zJ5xbEHaV}RnqKMKc`{AScrcJ%IQooov@<=Td(j)%w(z^+vqkxR)za~}e)6^r-&=_9
zHb%4Qi;)JWlTV9&5ZIHBpgSaA&+?9q>tLh#Ag~Z~3d7ZLHmQ0|bmD%BPfGhqwSN+V
z=kROXUYo5BV!a6giiPk9g5781FR$E1-e)^?b$6<$mg@U8KIFY-=jAn(ep8qJfxM3d
z#fvd@=F{Gzw7^Ai;rA#A1a2!l0hUkS{^U%$g_sM2gyMwANb>W2xfibo`inmAc2Gq~
zP9(fxEk?aWt#p28*xj8l8b?3b+ZLbo?X9u-?m=Z$_J=FCxY!sl6RdT1w3<x0^nzSu
zh;K~sZMZyT)O&<`j<Y^mtBfi|@@RJh^Mpb}SJ%qQVz4DkSNr4+0&z|7Nc_{a-UShS
zOT+$irU}!1a-{K8NQ4IWx2ozy-Tk?^U0dga+$ib{ww)r~hWd9F8`xi4kj^Wd?t7Vu
zQk53?yQ|}zRBd;rnjG#@T7jtq;?%}s4P=6F6QqjJ5HWUgPS)Lr|3e_=PLQ9Fz<<;7
zU<Noag!Sn5e=ol{_)d`k-@S-}EFdd64PN~JUcO|)&A0{MCHTnFS>(Rt`1dyvh~0u0
zpW)0Q%<(_G=u}dBv!VCj%kO82TFv3R_jCr`91Qm~aPa<n`TxC#sIqQlVf}xvA-v4~
z=Z#4u(!;sAABiLV_p%pd0AP*?L<j-8*c+Cms%`ir@O7H}@_&9GS&X!S<naIdD9$MV
z^HJ>9^B)KNJ0Jw&QtN-;_a$8f7~R6j`|ssB9@PK&qfd}Ci}Hs?lK*=Z(aih5zYjM!
z;-f%$q5Hqjia>Zd{O_Bjf6avKmjB<IP=ep2HbLZnhl+Twzug}+Xl`k7xV<(Em+fS1
zLSuYVtJ;a(i%`N{+fwF0Ew6jvMR8cF4TX@t8NENnRt99uuS^!syTMGBXM@1Rx=)48
zN<i3;ffqX$>RVQYx|7uxe>``$Co~n*wf75(LZ5C&CKentdpvCr;D<rQYH3wq<eU{K
zOlaHG@<K#u_{Js=k2hvA>f=5%Sr5;eTN;^|8L3vc{%Y+km&qC7)1_<oD^&W^Z@;tp
z=8rVRUwwaBy1at_cR70^v`=9i(zmJ6bJjZmZrP^7uq->2b2n-~Z1lQ|65pc%i;8rS
z^Tn+ToYdu|`)S2E_VA9Px1X#ztizIleEZf+$Jjucx&j4XG7&lP70u}#w%lr#lmF8K
z?2M&1rhm9T3s(1~!V}Kb9%bXLbxmC$@UG(ABHQ{_$kN0pE7#+4y$}<oexD<djQFaw
zQhqKV{Zp<Ay721JB+lMlshW}k|N4wZEMMWS-ri>WE4n!Cke)v(npYQxDk`re&y(y|
zW~i;5xP?2rfBdRV{(1DBpD-o{jna%B72Wr8{ixp5vAw?<P(71^DHrez1<D3|eQC2y
zb_7L#*N0WbE+-L3?M!+9e(@YWn=)Of>*-%j%jDhEubG)31fO4;_u_r@n3nr0QGHq`
znUIiBui0}pIv<LWrbaiN8jr0UzJcnvo$p?a^q1V)N>o3?VbJX{4<{bFZ*P6K-28Oq
z!AkUD#o^~V->CKwfp_aV^459;%m#T37e}WOnA=)&TixAEKH)DOy;4)tYIa+z7!URl
zHo$PN_t+Cqzn=p$2Y7*^9PvpBv4gEV=Pph#f&~j<BtAcB0XAD-OTS3s{=3^B9@|ZI
zX3+PIT)b(Xc@7JZthBtv{2=$IVujJa{duJ~TQ{H3?g+=)Zy>Kt1WB@3Z?OG$%fD~0
zbR?4t+t#~sJNokMov}8BWR#DO9~SwO^ra)*0A#)pI#$-_UWc|$PO3UJ&K>PD&+8@H
zTE*L1vDataJC?a?8jSpp>9Kd{aM2(mJgOsXro@er&Dp_9$YQFKEt!8ozuwhVL+cB7
z1}A9O<TC~>D&(!%1ocu}5~b56>n4Ke!|7<nat8e{S1rLVQ--KFk=JU!rZqF#nSx4Z
zf<KwBEKEE~7;SB1O^?q(MMRgCa;-eoruyeam2#enPdH5co(}#*J>K!1-Ja@8r=?~I
zXdNL&BO)ZIzu650Lm1ZLe5?c2DfeCDn|T6W*R!LhI`}~)X*EwN6DPr<{KJVGm5m|C
z^XNb$Y%*R&Wo2f@%X_Ic@MlW&iIQ8+PtVF!qMO<2FMs_R_%paqF&5UbDC5=Ylw#4Z
zSCikaw70)G^`l5yW*9)n`|SGFt`{>Q+Rmuct^<)1Z2I-Q4%b6;+usVa2B#+vB#Pw|
z8&~0q6Orp4uxOTE|BVl}8gd+d63=05FLNW6VI?Ns2CMsjm*$$~cKWVJ>ump*=wh5T
zEXy^D<mVm@?<k(ty&WV|jg<!LGvR=iaJ_oNLCD1D2qVuf&WQ^D(qMIR>vBvyJ=`CQ
zc&w|}SCqz2LOfN#J))<Od|RlbtZe_dvDu$KTr!h&L*07PGc7G`pk>B%Z0^?jn{QJM
zmk#%l3=J1|1*6)-q$OoR_>}P6v+UcqUt9G!?E3ZNFN7X$TKVA+zsJQ#7Xxq5h-jSE
z4SE=Wpi+bz580yyW+3$Z8m$g<Yp2$cB;+FlTGKzB&*O8e#uipC&EYLN*dO2t!wn1%
zLv&0`UcO=f7e30zt20HO{&>PRv;N?R|9xVJEbLlRTAp}bXm_prK5H#9KoK3EsF(Ya
zx|I|KZT}iitDkwagn#gKe|NHMTMKUB!2y^1W_68)M<tX<T)$NTw(iqM54b+fW3+lG
z<VjBXcV<vkX<>1z)Z?hVgl7j&n3DZREPF~pcDHnYX99<TqwHqRlI2rMA96u#><0Vx
zjslFPZmCpdl*7Gx+cT|c|1*?VdA}omqSRK|NYLQU_)DY+I@i`^qcqdpLzloN{at7o
z*WH~D1C)~Ozn}5*p5m<MbSpuP)a<gvNA8xEQOJ8xEOij~?Z*Ws1;_iJOEV3g%^$K&
z9(-mb=z!<L)YP=u>m(g?9O%A9TazIa&t;{9+*Ym^ne}aCxK&5KL|nsw{M4ImFq$Ft
ze3mn8ZX|wQZL<Pa!Lz;+<j?J5I~KWju!%WJip!&(3o|_7&Dws94wk@<ue?t}i)io}
zshF03kITHpMy34S(eM1aS&xO#gv{)7P~datCuxIfZWm7M3=D2pRZ?bwFCJkM4|MdD
zmq$mdh^qWrJBaJ)?#2^tbUz<3ugGUQp1b7<No8j}qbbcUzOzGlpLzx;uk!A7FIn<{
z?0g?17)~N+uoFkv%DH$FxxD_)-D<=W)^>i#z`zha#S3%4C8cGZ-Cg6gE3Xvoq>ywY
zV`5mD*aW@vuBpLEIqvptiBlYKq^}vwpvDgrK{*BqSdLOI+%;g;z{tk-5)B|UWo3~N
zNgDYF{9mi8h$!#bQm^*b@1o>$@NmXIHm&*C;2~+QF&&T@7AL_|%!ZZ@h9ZUhC!?M+
zvwBU?iV$Opzj14ao35va8CN)&-{}FjN6E(qusOCM*1W)Z*m1!gFC?R>iAU+UGe2B6
zD+bBf84{SK<)uYSdnwOx8bM=?ABZaq0HfFLv$1Jl6=uL3-+Z*`r%MUGuQ2&jmltt;
z*2eg}=ZiTa9rAY1baejs{gDALd@#6v?9qwfh;KNVQzXMle&TdHh4^G)DN)<``kIC#
zzAVEJ$708r9TyEATf3zRIf3JFQ5x*c*~!Ur&Oq6qt7zNH->w2=JNju^>E}CKzw6fJ
zL){P9rY3FDDk>bcozefTA9D(dii#~O7&OgH%#7Six~xHi8k%1i8JS>tcXfZay}jkS
z{DaU3H{u2K4v`TN<4qecSi`E#F8uBqiEj#d+@|E3@=p7d+?(%|Xz6<Sg!(d$HL3jV
zkvNH5_fUiG^n;Mls_gP8Pfd&Xz|#3+Im@cevHQp8x~OsQfChTVX@}xPo+c%Aw6)Zu
zq<FUmFFHP^cix_K`^_7z_V@#JOsqpwg;bqt9SQB8jB@R{LU-`l-x!~EKPvM@@(=r|
zz%7&&)LWF7w>H*73%G?OO?PjPexm`Yu%zP4momzWXTgmw)zRQr^^FVy?a1U@?vt<0
zZ;$j17O9Hc{w{E!A3cb(UuM_jPSY>ODu}~=z!%$7_>|-KY^llCu`a5xext?CW{uIo
zgq<=xdrMpy*I&OHjvmVcfZ*aG7}l>}YmLEHU0N0HgOoqEd4d-adR%qNw6bzPHahz3
z>|6%}^NlnCJcEhx4+=HP0XdK*<F|*t%kru9Pj*-5E&GK-U)NxDtP_eFi>>|ZYa<tr
zrpm0d*du23qDeA5!r``QtCbO_MY?|3{!XBj*!n^v--KXdbu*LH4sd464-5`!RlUjh
zjA^v?aXk0#7yDmDx_qpfwR-9j?vw3*ppb%=L5VRsS$m|-D=1aaEFvgqzbd3aOjz;f
za5h~isfCSYqy02d`YYJB#RbE>|F*_2Pa_rUH3ut5?jP-Oa6T#07Bqg{e7r@bRyeDr
zt<C@DG@MXonN_Qai1JWSKuKR=a&YGO*BLWf`hF1Ia}?)|i6zA-de@ML+!g9LskOms
z(dsJlsT^Kc`LsYC3Zq%Pd2{;ipuzp7Yk_d1L10RqRTuYus)mNjFC|62di%X9-HX{K
z911?;7tS{Q8y~9igl~VCgZCeaB~tjP?-3;P0!{UIZ8kh+<8=<R@dt70?D{n>*T3L2
zau3jI632HlRpm3u$cQwSl`-+s9wj~`7eG%r8x=+JoEA5umm}dRGeV6cwB(6VSLr*q
zT&Th-XqStpB?`8gX#xuMuepu;miQ|RO<rLJ{+#qYjiI4KD^j)T8YiCJ<LByTIe0%&
zqzoBgm2o}0!?u+?HLbJJWU@MUjdXFORc|5j^5uM|!KXg5$sHkkLnpc3oYU6(>BXv=
zEFN-uts*RsSJ`5uli_Fz`PSSeMHk?ah*VZG`=$N$5qq=676)H)s7OWCNG$fdr<2YU
zH7-l-`(h5@{}EE0e3J$+n>(v^V`)(Dc$LAQ&v^t-ZEp4(b3n{xF(oM`<_qDhswyhd
zbZOJxhbjs36#Q>Arpd(7(vO(q0$RV0W5f=x9Xb^d^*nxXy1cWsvM~+Zo*aLE{unHt
z<29yh6}-WTE<KgVzQQ9aH_e~_^(}n-<INYif!<5@1+)%||8+p$oAaU)dw9GLuO)-C
zgAP}>ACm1WJ;D`mJ^nG8!_0Os$OkQ5*^HeoC)3o_WEPZ6k@sO0Zp^Kr5o#`}sWI+I
z`P8FkBZ`)lWwy|Cyt-ZKKIz8fYnWCXmmJlzc;J5Oh~;^9KEF7}^z>pr$}B>Wg4O|x
ziHT{l+Cit;WChk|>I|W$3F7Gu4ODbb*?4KU673@-qvP~l)~B*y0q0y{A70~fs$9Ng
z?VGn%FP?qsXnXUW*$gbgcLr;z^|DmuVtjEIFs`vv?jn2W1mFrA?qwD5h+{td9a%6?
z0)=o~(F_9%lc4)Ny0RLP!rir*Ld&rY=M!?lJj3!-Wd8=7u(!uFq1>cZlj9nwx<bu%
zy=61cJsIwC2uo@u^hL)Ml5*VJy%U)M<?ekpeNw&}ia=3V2LLhk6>9@`1+g|-5U$4{
z1GYe6w0Qf|HW3kFYW=;mv`58hkHT=*|NNw<IaFOO;3304*}m!sluE64s~OcUuW3QX
z0urR$-QVGU@$&K#J#EeuTHR#yj8fk3PZA1*gmE4L0qCAb6P1mCCH9R-ztpqdD7soA
zjOj{MR!FV?T~Ipp=YH{wM#)<_Q?+-{Vhkjs#u8=s*1teY-%OhNGFEeW`Ef_#t(}FY
zm)A?kg@Ucm?mY@+bYfg;RAT?Ut&!m()O{S~Wu#UgW$8e>(?(CrQ!3+m>EXO16d9YM
zsixNCIeH6d>4#QwKH-uKOsuRzkYfVDbiT}^cJz3{a0_Rd8PB+%W@c7wuI_On-~Qoh
z069>yZ<Ag;0=L8OjfMQrHkP+8c7A4{qX!ouBa@J<@=VbZ6Z@q%%@na(`d{Eo)tYQ&
zSN0Hp_Wt|~JiUJPutWO8n;LSeh_R0Tlh$~Q(^a6!tc`PRjMKt!rDe}Q6UevG(#+4q
z#>Q~JiYct54V2|`-+QB}dDiH!fRBG8*a%nUNvWOYY?~_T;Ss;<v7w{vTd$Y%DOVCW
z5BbH?m2idI+Gy9}H6e?(`l7av$3S>!1x{Bt)4OT6Z!En|72GG}0&04C5wR(D+e$AK
zKqwUXilOdgt}D@b9<Dl6E42DM&f7vMyf$kq(G?#*{;>S;GIrU!f9hLqu475LW{L69
zuo=6csOXYsBcV`EW?O4}OiY}&gEk-ESdm_HWo88nKmEyEL{ubBaJHY8rluxbKMT$A
z`9-tqWCWI=F?Xze&xQp7L3`*cl~?Bf3%gtn{{8anjzji5?#qk>p-^#Y>F%Ck&+&Y$
zd+29d7rR*)rka{x+S+N+e5rhJQ%-zv7Y<G%CH8{Vt-tA(GOV7T1$2Ia=JJkz!@b}(
zi#1}c1QNiMyeXRVnW~~Pl2eMF_PONe6lCNSIGFFmnb||MnZQc~B@KRXq)^U?r?~}r
z`YVQbV#<4g(5@07q0VApxgS7`s|*GwJ?dn<w;Y*3TK_Ls8Qa!%QkQPTyOWlvJtOfi
z2@kGBWrQ-)=5{bX;Qo-vDK4-wHh;uh3N|@T^@qv}`1leEN!EtObFXdqjbG0ih@i-(
zme&OJ=B=oVT2OJE8)<6Rj^YT@&pz{JeBHY*_VU)sj|@Yo$go_`wD7eV8CbrptR6b*
zHJ<U`k~TH1DymXz<mMsEh%oBz?(UrHvP4TyN}@AKEzHdg7jRkmyPTNtfXe64(J9kh
z35-mBEXisTs}dE=yGhavWtLLQ%6$$L{g_0?$@;w9^?vTHIHcR7g*qM==e_~CUslo$
z{`KFUq@7Ppx{U4Z>@1Yu#Tq;Y<xEEi$?}a;e%b>{5u__9QPN+*Hpo)T=Ft*ogOmO3
z+uxRyj+LCuY>dPK)vjvXuznk^_|AG9JlNI&@-{5h0C&fmBGykAC%jJEO4FV}!7|Y#
z+S5{siXXb7<@!5+G{X&F%1id}Vc`zyGMBWnNJdGCMx<NA;AMv+i5x^(U2PQpSUjK*
zc;jKJ)_8d2DBaUt(1c>m!C9`!)XkNkA}3e)is4tB$M)%Eo&FanIjc&`9z{rUaD2|^
zdr1)?d5;V~$ZUuk-G39}Lq2$CNl7tw_AoMZ#>K?wH@lim=hHe{>jVURzUz6k3~6?9
zwOCs6(m~QW!`HrkUUy_1$}w`VxA;8-1URzGzAAlw<nZ`kRr<YILN4Y{wK~qMwXTWr
z133?w10$|v{yOR5HEo`6?ah&rqWY;trFM5u&!lPcS8!SYI8h|Xj*lm)b|^`F{LYw%
zJq$o)Q<Ie-Jg8u@l^A$zW;e!KAiyvuGmF{>`S!g3Zmxq9hp26Hp9OxfV#YyYf-nnF
zaAM;`o3p{=4=->Z5|BP5a6dXP+_<mcWak_ldgg=s!;;bfCSxlOcf5I)_gqGiY;Y(z
z($1#MUo~2~0;s#&X)dsFtDxvQ@7C_kGh5h`!Vh^V*A8MVdHTsFfjkQ{%IR|`C}`$=
zlVmFj?Cyq<1oqV+m9!JBxBSB=*7=a%k@0C7a?I=8YKk@L&n$<J$R6rHdW3&X{z#k#
z@6J(Sn+OdasXXsxZ#;EG0@vy{>l7`ji1#9~w?9BpFUxU+#BG4qzg@T?<Q0@yZ!Tt_
zxCh0^M74uxAuBbf<zeeuHk~+dCX3=&M_bF|yKmF7Qw#j&DIc$}-J$Wm=6|cGqx2CM
zpW3d`jw(V!hv$OZSJ*;2Y9ZD2y~1@8)0%>m#khY!r`yO^I$V;gqy<~bQh6IG(a4fM
z$+}d&Vkd7(`~WID5eXlm&B92@#Ez9wtOEimn!}?Xlqxqy*V)`X-AqhGY}$!&`g1Fj
zR;zPfPS_$WU2?+=A6_urL$BT((>{@1yi>Kmk3lD!Z1vW7?(gdrp<I2N=56m@t<&Hp
z-VG)7EANM*pPs~*^$_)0{2#bXXK}56@Z7VJU9Wa~u#x>*v&mFh9mko;cj!BUT*g7!
z!otl%cxXIWJ?@*%XsL+%?GNB`0C)M7n<HH0>5E-Seo|)DSafoU14g!&5)z!$oESH+
z@}k$**}Dwes0RC=-@krA;7=+gDKoq?+whs-$%3m$q)rZ4kkNX2@-5gGxOxcs$}+Mr
zg$64eZcg2e+t>hWGz-D26|GvB!nCv4@mX1u?0gcR;fHlCwrIJqK;Trt#aUj~oTSk`
zOMfiUM=k^iA-iyZbQbt$#Ky)_`jBl}6(#Yr$>l6UBTbq3lTw)hEgcqPR}bB?pqB7=
zjwPOuxY7wr0arqFNadY3pJTFk<)oROvd#ZG>tP<X+Wxyd)Hz5?BLy?u_<^y@rgnRQ
zXq1Af3XgCzGc$38Y4E#VDCErNj0eH0*Ue-^jdWQM5@Vv=SC~Ym9Fz628j^36hllp=
zsw=7dwXHq%8T~!7e|Cw&6k+h#v4Y!Xs!m%&<uh(sb>%f-k63(SeCOi_cLTdt*zd6D
zRE@Na=WV|xFdQ~iigajO+}qIAS#$rx&7Ehl23_*$TVqp`o+l=I4^J~<zLb?au>6wU
zTak&$GPHhkDd$tM^x^Ie05=y-alf;33ZibDe3E^F%9QLr1^Vyhb@ujDpShuutZr~T
z7;*_3?(gpW1<_6bcc4+im8~uC&aA3>O3OXnGDCF3#40yiV>&*|$?Uz%(ne1Mxn)u_
zmR1%i(MkJG$A(v<ol{d>t9>E@Q(ZG7`P=FRr)II9@{+P()((Cgv*ZBgf}f+mKk(Ck
zm%g}MFK_zxSbbypQdZXctW?NXmde61zq*0^-?{}<$k+Hqa}`xhXam8|CMvqY>q3u(
z$l1~1uKU4Hr+5C?#E-<XP^QrEgT)%%PN9%&y+KZhP5Foae&!7ZFzoW9uqCs@rLjBk
zZC;*kzmj>6+ImKpP1~VFPi%HE5gs!0@v^RY8ZX<vcL&^ms3M+~=(GF05;nAdt8@Fj
z>j74r_7x@7o34+L-#2}@znK&s{h2=92dFL{x6_u!N$csRFtptIIQp;-3|C@5Tt2V!
zA0t@IVpNKzYN0o2O#nS8$lOTu4FXQ(j+279K3i|yt;);0{P%7~!B|pjW`Ch=^3B?D
zqr}zuVG_T?8%gsod4YoiEm!y1bP3`H28@h5drI;MI>0-CpHzlC>_m_Xo$~O>)6k#u
zNqluXKVmEe$cr1JGCrhiw-?sq2}6a<U+29tK^RQ+)zIzeP)e@n)U<^qLeg<<J>;xB
z$Mc#^YJf)^hk^wc9m!@|4DD!R`e5rq+;0YXrp{quH{}WSRfdruG}E$KN+4Pd#qI9x
zWoKg2<FXOoJs4n1o*EdfRfM|bSzJkNULZJYG(87IN>xYce6y{`F2fH%7m!HD9m|J^
zAEKeDs*nCkJW~H2qU~14cU(2%ADSYJMoX6im9-XtsI|ejWHsaWjQmhhrtAOdDc21g
zp2chOQeUFZc<#3L3@2;r8W|cx>PPhMVq(k8$i%=49LmhE-(W_S78kwEaPKJ5nd|o-
zzM2XVG)>Kfyu7mA!A6&}x{};6@7MakhBA*@-}Mz1(%%Wleb7~ru$iPoE%Isp8mS;Z
zKl&bFbNv3n^YmAzytIFQ{jS-buz!>PE%f08wGZ$lDw<3+YXc?KckcM+^y%txIipn9
z@d-7jIrncTB&-u;{}NyIr|59UNev~QZ}#dPIz6SaWDbt5IIJ6`;^a50xj~4Dxmxf&
ze@<QQN?ZdFtU60Fz>{5_o&TsI<J@MeQK1o;$hbIVWfNLDo@-YDhwJXlJ?ernGHic-
zuV1_H*7vxH#`8uy6!rjwoYGX@5%QS$2QO1CJcw-|BcKm2d(0DK)V)A(At=B^&AIV6
z9z%xc*Rt)ivpc>&uTb|Pv~0N6^$dOAo}eS-S3mnJC8Y+pEu!kWiT3W>IQNy5P42Sx
z+N$zcPs_>ldmJrK@3%ShCpG^v%me1M#H6Fi&z+Pa^_j~H3ZK@8I!RfXnQ&4d(|GVn
z=_+bQJJf055{f5W$EWq@4|mO4{RstF$LT?((zXEja0)?S8*IrTAu?m7E02V8ZP@j7
zA8uj96(6`XSnJl)(9o)@sd6v2UX$HEzmfZ2TJm}){P5^VB8+@GS{EP$^>e4XpwKE)
zFRu^AnmW(|(Ln1;goeBpKTbo%D8NKQze9r`2pu56#<;?O&J0;pU~eS@(zS>mdHj&i
zZ-v{o<`49UAZNi97NezIS|WrM9*xr4=_qRMK3I)nA3v4+{PgWOqo#X+Bn?qqEcX+y
zmsEJuIukO(Yr^S|>7L3b$j4<5%Sg+>HI$+;wD$TyV=kN>4sEkN$*;uS^&@_c__x+?
zBEMX^g7?$uW6L*hKz$Kz)SW;=nwOvN5u~8g?21ZG5+0R!axIb>kuE4BcD7bNu$a@o
zNKeEOA03k%>>!(&Wi~RhPnBV~TXWY3_rrT1VMSv{+2Pb}h*6T)^OAluAtY^o08yv5
zAufSS>+1>Y2?_N5daG-1_$Mau@A?1%l4HgdxlO~dZe(JTRT(3uD7AJBt$}pUu2}QM
zT%aTk1i%drsq`~5=#<-$%HLMfQzYK!kd={928L(ZXlmQj-_li}%lBt85QMOy11Lir
zk7bhOxwjtz8Iv;Z^nm%ShNWUZ<PWhS<p_+9h)+(A3Rbv(3-1SYhO2;{D&d7@Co^=b
zsQih*Y5;q&aFlp@yX7K)0kZtyFKatfw0c}d>%aOE5nuTAtFegzY{zlY(SQZcO;WEp
zF!}$y8}>O{N~*S&55xDP5j_~?x<t11Y=~utA7G%v{V-ZyUmaAozl|h8!x*L4;1Pmr
z@8D#b$B-)EuqS1NQPk{0Y-(U_Xr6Z8apmRrkVh64b&+2UH%Hsrn$3)JiwiPX<#KYe
z%)mn$s$VWXRw@p0Ky2H)LCVF>%G}@i>xV>KWP+Y*8@BsZsBro_TRRSp0R>t4@$#Eo
zntBOn^2JT7Vrt_dB-KDsQAb!QqnaW{dJQ0k$EzusqKb+$?cGZ7UhKt^GX1S`IoVn2
zYH9$@czq7Bob7A*_4^r{pxs8_l}j3OpePFiOYGFodmOY(bXEC5;JvXU=mIwu>|2n_
z>9{>c$<MQ6^p<Rf^{>g7*lfShY`O^1^LwcQ&DovC`#Gng0>46DJhHEy^1J_~|5rD*
z=TUo0U;7Jvc4}tJj_psgf4zZF2<wV&D;VWwS#yvESw><(d1?Cf;#@c|PBa8H0m!$7
z8K^k?M(18ILBM^J+ttjEU%zKY_%?g%!{dsBgJeLa1uaA9@w*{MKL4L`=Se{dZ{Jy4
zOG#O{nX0KAN&pN@K>ew6bu8H(Te`@?!eYE*%!OHolUB?(Xuo?*VNW<hQeQzrUqL-V
z9?mWkD|3e`@NEHY3j^vlk~D@L*lb|eDiN9*(JJ0$tR$>t4?T}?z>7z}RYpa{y+zS{
z8$7H4c`$r?%JTOQuph04rLoz3UV)-i*=EL`4uX|j+_IlBv(^rirY?W{Fr0GymGewc
zfA?BMRMXWr(bC<DwKol{;!u$nsv9nF8Sw-8>i9@Wi8IZNt!5ow?H)vRnsZI;QX-M?
zZ3!DVzWqv5BuazN|7`MVGeG8lixB_0`FT9DsmzAR40?GR$4>Au>=(_DnyztlTdbO%
z#r*PykcP@`GuHqo71_*W1;~W^3KZiNRwo6cJ6d`U1d5E&mT~giytbPbMlIaC6!h%u
z&R_wI@-8>8@VPs$B);14E9JGVjq#Dj{}EPr2Q9B)!nc752@_Wm^0J8nf^d;892j#+
z>F)&s3l0XOz$j<}gJo>|ly=gQnFO#&Q?<T>@VJ|M_ElT|rv)(0NbAK5L}rz%{D1O}
zPygg31AN=hU4EE402v72leXR@&oSAv^If}=?Ckd8m2^PS5*lS)IK#XC0@Po?V|~TX
z{QT1mHh$oLf{w=jIWOM0{{$UcSn~A;!E>}TH&>tV(U5T{cr1h&PJ7XOg=5PMwNL&>
z@pAc6tM0W`Bc$tSn`qEPbvcdVT=#|th%&`!IcG4UAq!jT;^7C1X1$s?=i(xak-o~K
zjqd{LdT1!oU-(QZ&^V(vtkks{%lzqOAdpj$AywOW=5!FhsCR|+-@&~1me>#y7%*4r
zCr4giTd(NaV@ct0X#aiJYUr>TWPTmxs2?!>DvXgRMd_j-jb!&JbOz~=>p5ty6zNR7
zx*_O}n_HMQzK<<S!m|(drp*fjw$&Tgxp`$Ip7#O39sU2P`U|KkyR~~9zCl4i8fm0M
zT3T8G>24$yknV0JrAxX*y1To(yFsM88@}Z^?-~E!x5pkspZ&mg?>pAIt~uv5XTg&!
zMWi@*`#)7_d*f4Mj|$0OF3<ZKs*A1GzMHEznFD^nfy*<Gmx~EFzHl{$mxRxNf7G6!
zs6kGx@%zBklrs2t!S?wi#E_=~5^lk&{Ye8bj0*~Aj4~HSXs{_M`y3q2=9LcN@7YXD
z-v3YkQ*x^^@>PVqP<j4#);vs{5;R9tga~A>Er#r9Pz4ZCHL5)6g>t>pQrR98YeD5R
zTFP~;T~~KAvBl6W^`&OCh!-vOA+a(Kw5kDD7R)8N=)uhlE}A+D++P+a{Z|Y>f8{Z~
zUIlQ!$5WIzWukylnSqK!;M4$ap;iMoJ@3_BjaD6wc=Ye8<M;Wq=6sXl<zt((=GE`t
zfLXK872M9x;&k~ud`Grk`UznsoRgnl6h8!T6Ip6HReMTj@87CORaUVB{0+2kC(c#D
zu_XgT16F&UQ$WBP`ZVyhf~v*)iT4$fhEmgsILm)8tfJfm76O&npvd$`F0++SiAkq$
zu*poyr0Dfd_ETfix3A*R@gEe*!vSmumJdqIori$lzdw~F>Iw@(jJ9O$6O~PG1y#+{
z72l-r)aMuE0cbbC*B{ZZJ=;l9N~xoB_U+#;`sAJHtl#^~Wj?QBP6f#6Xu09YGIM_c
z01WG%PdUm<StiEN^vf$u0)KQ63ZcEYq~0$HBAn&YnpRJ4LaVp+jP$>n4z8c*G<n9w
z>3(~Ojs1#&QCe8HiG_{Bv9tmRQNGt4{+Tys&l^rzRQ?$Q_HFQ^GsXmyk{}Cp6Rdx9
z;Lo)mUApb~Sy_g|vZ%JwUq2dV>7tSJRTLomtc@=tT1AUL;kkmeD!vALWkvZn^d-)>
zeBdyM>)1Nlz8#&&%PaB#FPu7?m>d5pC<twKTeY_>9T*tNk$(fKH9)ea3ZNAg6?HMM
zVGVEClJI*hH>U`Hv6Zy3k?45w&U*Wu#Pbioqm!h}wY;<I6N{=0bv~s3_5Ku}aCv;9
zsH%8+x*vkx-!}xfD<K4+Pc0@sylw(uS(K{*UI;4F0rL69qE>RS)Bw8BNNT{J{6XnZ
zuo|Na85(Y^R2Jv=+dA0+v~p#!4B6j)hAAa;=X`KK8R!70HdIv1>0-RmLDaDfQQ9A4
zD>KtC%9@&YKsW>%w!S81Wi7B9@i<!N<rX4);?Qxk_4i9^^El)d=egXoKMy1<X^9$o
z#N@cHCo8j@2m@DqJ<t3AE}(I2E_5q!{@7}0;D-;tHZ(NA#EkBs$goMTj`a^Jw{;g%
zyt35J%KZsI`^nHKAOy?JDbQ9@BE09T40>>)2MN&!3*1Q0y}{o1vYGfb5O9yae*yvX
zoT@m}v73AXPoGvT`X26^nmwDruGJwbR}5mS74sIjfYnyJ7uXDxk-D0c0^&YfnVOnn
z2mviEFqPn;Ru$6>7L5kgHW1>7$)#QGjuizBx!{~C>V%n?8?<hyYJQ3bj985a?~B-=
zQy@0Dc83uX*9n`O>HrpU0I_Ux!DZSB53bPvB&zD-@)0eq7`!hzD>W0@Slh7w(AugM
z=3rHmlB}gC;32S+d`38vI=Tn}JlN_LD*r}NpB5<O{&R1&re!54!Gp$Urs&hDOsZdA
zbT{yK_;Ub6r*oe)Su?Y)4rgb(mQZkA9e3>i920Bc<Ff`<+6~OF3B#}20!hS8bO6v1
zV|;%L)(i)F+;v-c;oHBFO@Q1tGc)sOaNu{;Gn$|=1ncJI(LCrNkA<Ci#Tw|h!suz~
zNMS!u%x?gln-GFc?V?LXcKthE$z=}9^S6Aw{_jcm7s|$Jl4!n*S5}qrfmGSw-a;LW
zvc!U}Q{`|5m=d~jTZ@4{QQYqo6(@)GRPB#8>s$_OmGp%{y@>=aCeTRndY#JCaK+8f
z&jVrV(NbC27(TEmZ76RJumFuS6W@4+6pa#AM&kT@`|mGWz`+Ly&y!R3XBc#I3o|L_
z(nKhU=W3QZnkdN8=eg(81nT8fMd4f6y)G*y)<rQTC>a@tCd4j*LI0X93JM1(avwch
zWfR`kTgG*Vou{$s?`GO$^iqz`DhLDGo{{?r=xsSzCzTBI3U#Xt=GwVM;a!Is<~E0t
z>x~aXfRZhO_9x&HWbk+~<My_V3GjE=63PKn37(*%s!bLrU5%nTGXe>=QoHFBCI`v;
zeUwpIe)!AZk1u@pj7@lE=YV9y2&n2Ek{m6L>~97YTwJ)xBWoSj=>XbeQ}^4u9h@)e
zkWlpRZ@~_6kTgA6CradQGP?pNvdiHxK8qc^1;Af`mfO*w?HYS=2Vn@m7)oH-D3Mry
zV{nS_^r0Y@v2`Pnz!VfTAd*>4RdK%2@}rV%%H$4a&N7ps!PlZsEmPC;<FmaGS!O|o
z$3eRU8x{q~LD%2)&C)F~;T5}^o1n4!RHFqhbt5wm`zKU5IG6kJ_BuA|5PVtkNPv5R
zY~SF$4+uACyER@jU&$kTf5z~8*q8jRpzblS{M1k==2XH#kAsg-0A@Nll9>bBBEVhO
zra(s=UZb~@7`os{;fu!_!ZCo$YG`-@`pVjMv%mApCL2*V;AS37<TP1NGz54>K^!wX
z_zvTf>U>_)|5QxZ&K66h<vvmkV;uqK$0yx1m#205*Zv8DZu5mESX4z<yK8KvB`e>x
zlbZN=!QtNBF{SVYjio9>D$SgjhK9+&ZOru>z{5B=$QWX4lA=#ePAa#*OMH?8NdJ;V
z#(y`n`<IE_VbuH-N<oq;YWjdvbakl>5bJvN>J`E6UJ9QZ#u54h%i(<%<Qq{~C=EDz
zfJlP1mV1JRgxiz)?KZK{8wz^2p;NyZ%jLYHA_qGMS($;^s-w3;+W_x5e65Z6+*_UD
zaBpj1{Uo;d6WE6>7wg?NXFUPf?{g?%Ug>sfJE*NWlh$Ot(U!qnP&OFPb8@|N25=%_
z1Q4P_v|RJhUBs(pc~P`*b$4~ru}+GIY{HCo)$$*8L2kkPyt>25PVkF(dux*!;|z0D
zqmv*9=Qjc0n^WTg?#h(5FoXbgFO+Wmck}~A_e0mFHRatM+J2I*e68)lSHNl5U#)db
z1n@f8dp!P8c_w3ExK*rGgKt%+-k_l<OBtCe;5Js`Bo`ZGdweGE#D%M+qpVf#uqD6~
zI$hD=wB&kz;Q-?2sb$gv#18FuSIO}WCn_4QI!){Zax%$$g>;)YRn^u{BSuTXbJ=t0
z?qXM=hAv&6vLhn3+8xj>(&CcKRSpl}Lh%#&rzQCPMZjXY8TX$!tqSaYO5N|dXBPkv
zd+ey=2}2A=@d^p_@6%Ruf#~_Uxp>_6qmfuf=t5}3a0Pkg0PUzbT)YZr3KXgByx7Ad
z@0$as!WxaFB!gzelWk^aY+@@sRe*%2wk4^+Zs`<Xi3a38-S$@{A*9ELr(GV-;yK@W
zT38f;r5td$pgTtwf|;>$xUi;)?z>h+tiOW-JS~@8-Wv-+ejUJHv<#J5&A=cROxt^&
z4t{i>L_#`MG8*1Gg9E)e0L!z+GQmc3cX0x?==pMeX24Q+@3tN?hH^$mCN!Bo#MTq8
zfO>0WfEgrGOFky^*zEs^i2JPRbSQ?v?{WF3zfK=!#s?@S{DdNbC^x7oiH)@)_f`Le
zR1VRB1>ojMeO+%|!xP@|GxB4Ry}I3%k&!_W`XQSklknEHZF~xl%7DqGq5{$x=QAcE
zvY#>0AR6lS>V%NT6_Xftq7+Q?T^yhA)Ee!9At^xKfg~`JgTNOppgrJ8skWGbi4-r)
z%(Xt=re(v@w_{t+9LWO+T%gyGgQbz6VS7B#yQqd4B_-jn)Z6I-EeH^<05!pT1jDa}
zB|6o50B)%r-cMfqx7^P=T>}&~${$R@^VuwlYZ4Z0$E*9wj~p&5i~$@&mxt5rar~d^
zcXZH2m$UWhH42Bi1L1#e())!({ERhKwJqFNySv7O_yp1fg1bO5Q7~f#h!G%%at@fW
zY8iZdLq`X6#_th8k}CKs0D-$UK3gp-ym<v$;PA9RfF(-fbXg3dI$LhMI~L$W61*Vu
zHMc07B?>DE$9=2{8mca5w_+C+&Y7sT(0TxHDIus4G4bI$M&HZ%w{0KHye1QQ|J*ks
zA<ss{U=Sdi#oBy%dFipyNk<2B+*BO=WGwh(%F4=7F)0mJTiG;iWF+Lj_hi4`L`tby
zIoCE+*_B2|{qAQmG^uQ1OZp-5`y1g0hiB}%VPQj;dt)e1faNx{Fj=+^@$XruqV_0m
z_>K|S2@0~Up<5|13O2v*2L{O47pIbw8&8A!fwAv%K{O8%VyZqZA9SSf5S<G5wFj}b
zwMYQ}*`P=Zwmld6tnADX*7E8V4uvmBY1Fc;0#We`PHWC-pai;cZ<e3!m1goC&0Q0w
z#A*g%7(x2%u<6?uGx<otK_=(1*&!z6F5KD0VK_>v@GG1mu-~PrlL!dAJG(uC<dV6}
zb`!CK+jW3o2IFFT$F8q|(kGsh5apjVG;SDJp%8(8vYfR%%C{N7bbRDUBW-OeW5$(<
ziJ#c8GOkF-PXxeS!l$#YbZly7cJh;`PqZn&zzh&$LcP=Tr$Fg`or@s^N>O^LGtfSy
z%g4v1g?FIy3A8l+_I`Bu`N3J_ygo(>d~5<@{MQdBSj`rLLo#n(1pz3gzMcSR;uJo}
zsfnqIW(N)71a=~%55RnvK^!IZEibow5R(J4lb)d|m)oBwHi!}nj;|x0i!{@`LTXv$
zgZ-djgD50U`U<qXL`U?D^jHYKT~t}la(p<nmrCv`*b4x<_D2?u$s(ts@}uZNAd3{&
zQdf2d!qj>h>$)kYL;l^B93y*a1~8F5m9enw(?eO}86<cA_1V1_S+qpNTfF~X)^~|5
ze3GW0{1fR4q-bh7dgPq;M|HHdiz+J%8-HWA(4o8Ivy-{>_5|;UpQf(TIBLcJOlGwo
z{7(+tgWC%o0XBzm^~Ug@uVQ2~Omg%7q;1E~9{<=D<=tFAkf4sXC;~3@0oL>&9k`a*
zfnz!gNK!T<fB%UBTNqI5fwkk}9N4p>dcF)FAz?G=j$TE_HOh$&a5`En59|AB@k@cv
z&G@*r8N^E6*zIyX*)m}UgE_Tkm*7`1bTT<oVTV9c5ng69)PMuXc_4{$CrXY=1ar_q
zKJqA_7r^H}ECDDIPhDLyzl|Q)jmO9BL1|%VHk{n@#nimk?rc6<>~ee8m}|OH{CcTM
z6d|(DU~R@Y80dGA{6fDjd_WjVZ86^+guABM9)2c)6}6CE7NKAk6%{3fK>Oy+>dVaR
zejYDcXk`sQK|@$nrhl;iaHgRLtd2lNx!6bz)Cyxhye}?}w1cgZ@>{G3-DO$?%arLp
z${UG9iJxQPClY{>7InIQF!^xZYM8(p#e3GazIq-a4S;NhDEHzq!ICa-s)AwdMpH}E
zMzg(otN_P7V>5E!%v?qAl>ukC2fElTCY^^{9Sk6#qp8{9|7%?f^blK&I#WiYx!^Vh
zo%*EHtEdU`Zf)$~E?q0WlbH%ST3mA9PF=QJP$j$Uugb!G^eZe(4)F?`zzF;Z2!WLz
z`VzQhz-N*(_t;m~P+<m^G{gUDJ9-~NivM3g<@$=lOFO6X9QmJyJai-5qT$8)Nd?$M
zB&4kLp6>4ftk<ciz{Fhg9)XsgmfV+;TkItgQaFf{IDbAqKKah=Piu4zSs!iQ9CzKg
z{SJ_5-z)b3(5c*PWet4EP!irh+kq7b=|MqAr~!mTr>EdlA6>QB9!zgu3<wM9=9R01
zo>H<ObQL8ljdl;4uL)#KkHSJd9^6MfJ=*V6W3uY}>hk1D+MG*6o>m{9oHZD50$W1m
z4YvLQaQAV4jAM0mOi5-XK~viU=nlYAE-|qK0Fa}#bq~h32tZO3c-%>X8CYq*whwf+
zGv<ra@A(BKDcaYp!RUcOtAd(R)|fF04Uvf`mW)dkl@%=`eL-HH%P9jTIqT+~T5zzG
zY;v>ppK~}BmFc?m7v5<B-S_T4hPuH-Oy}%^6{*hempKx*^8>dlimIw)%srh8(%B5C
zlvT;mK1CtO>%z{I|4t|u>+bs0UQOboW4ieHabHD%Tq>~m1$B8C0jI%3YzlW>9gtU?
zRwZ3i=;ER}u~V4`ppdY>kJ@br8ftcXIswvFAh?SJ-$$aKf`Veg+B+4#J?`1<HJnz9
z-xiJI&iP2B@fVw>j8@yT)0O6zM~4R{mX-u~_G+`09g(rwY?)brAV~DQUeyH*t++us
zky3UM?|)Id$9F>{V$rdwbA3Y`Ok{yQTP^Mn!y_}i8+#%-u4UZGTu1F<C?vq*PJ(jW
zmxTgw@F#!7uzJGcdF|+iUr19&cakemC5589V&1n`NQnX6fNw;7;x3q61i=u52Y6tp
z<HVVM+LCy&CA=c~nUMG!Onwhy9?%cM3loq{b}b#8k9kmHA|8>%MZ_nmm44}K1W+zz
zFc<gy0`YtU$`)d_{F^T<z<<T~*I45GC*O!{5J+MSRQ9I%4BXU8r>0G8{=|YPJ9A3|
z?{#s>GBb)CcDHq~^gC4Ra--5=z;XdEoJiIYVZQz<9t=^t)mU}%XDZT?M<S4<<kRhq
zCuBHV5&+%=^9ObewsTTj#Mif(CW?xkB?|_usK--Jqi$P|`XtO~FWHnBWkPAOM9-Yv
zg|K4vXQrno^Uls34t7bJ%d+c5m4t<L!X&cf*(Y11rGJfOTB9L8hfNXHf13$#>yj&Y
z5;B9&mXhR{8ykQIXGH+RM!r{ZQNJatB`7h|n4NJ>P21x!oFgKgW`=3dj@o=GfVi^K
z;E{n(AXtbxxa&{Q*A1e!J}2vu0Ob9G!a_<3ue%ImA%k-<=8<7rY3lFq@zw-O18u&O
zD5$EA&rIy>>@|U73FNJ7Hthe(RnFR$@^Z|7!(98Pa~bcrj>Erb0%;g%`r2OVM~YLz
zMyl&*1fqlGPeS_kekQXFxS5GihLNzjPmMGC=ynRuS+FUsZv<}wpLlX#pu_@8q94T<
zbJUvbY<MFoE)EtjuY=)L?1krdcfzkeux?_SOX17TF?n~f6NnzA^>Qpcz)HK>(e8bV
zPVum8O;QBS%QKSMqs^%bAdQOEdUsq<ru-`Dwd#N2@n3U5dbepONsvN)aX$CbcND1D
z?=DV5<)lRs0=sf^bJe3rV>PuI6QC-ag~OYQ$9k6w2B9~Ac)3ZGsl5knlbw-llf#9U
zgG{J817N)cbuzeEZ`Z4}U#K$c{vh!m(MJApbDg)Q17>Emm~hE{rbbqoJ*@$-q?3jc
zyyk{Vl)$TR+IoWpoBpdb*I=Re!9i&j@Sbj<>8!C_iTf!c(k0T<naXeTt!Ev;|99c3
z%U<V|-yN}{1=0k(Zd!{O%{lM-FEoL<hE|jN8i0XwMzlo{uyJCJTYyO$!R02%u>r6W
z*!jsLw*jpS6sKb&U{aGRE6YI&RI#Qme7>W@jC+&nbO`CcfBzXWi=*&e789|UE#Dhj
zKm!m0?g$Z3@@FZ&`EW)22Sx<2*lNXahf8JPnvzh+vl0|peIz8<wbE`;RDAM_4nP;;
zy(B_HaZ^%I=v!0uItVNU-Or03V5)6$dJ^2xk|r8A>H0bIyu6%z&w`oJUtwd<Siq67
z8;u23o)3<fvo6nwv;!$ZW~Ov`2XHq{Uf^D?cT62w_X?7W=kUT8GH%;90Xhz_QsKy@
zy!*P_$>IX%r)0$R9szW)jJ%h_Nf$FwHx)rrl3eCX*ckZw`i5nuJNH&$mIs@!F(O%b
zuZR<Q?RG)<Etm+^`OpT2n{A{iw-0@Q!TEBdEqn=k<<s#Ayln<MDHGYrVpl=Hm1W77
z3g#Ej|24!L<2OJmBmh*ZhK4bJJA*`igUN$3@euTSm%G%I(5WgZ8hUy<X1GP^$K?6_
zHN1~qT^Rvr#N6icK=d_+wROnv?mbYV)j-SK!{^`}F3wO~T)gac{X7*<T}eq(_3>){
zui?hVJNAU6I1+0L<ELR6utQ2pf{9HuBWA6(I|j`ro5MlcnvxV?HURjQE28TtKeZ5f
z1qGWW>=nXlz_lalixCOdMoVS^P0;EVGa89tDD(E_Yqu@wG*;8u)#Ki6yExe6C`I5)
zf{hJ^Eja^x(kch&=?ctD;b1@8NFWORP#d6UE)hWh5{$;_DR4PET4{Q1b$5->=7E+^
zm))ux*t1g0CUvY?bDPT=8X7je4h{<hT6BE@&nWi+j-Y`J$ztm3k@0C|tN+a^3SzW%
zbWpr|S4-Ovmiq2BFv1w8MtpwmlTHPMtIIEjNn*H4E{>AGk&5t3*olhS)10+$is8z~
z4lk}sv~-Dk^Qzpxo+nI8d}0U9yEr6;FM0$Qpo7gTb-&BYEs_2{{UXtxQ>}9xs<kj?
zrxnm0{h3w%>5ozPq1YYQ2~@-|n3<K;!aM@zQ2U0gVyFo&DlP!cqc%dK+vx701^XBN
z@rm^R<<mS)z6uI5t6*nUADWwi@s$!n{r?U&w-CT!#7k!<5(M8nV9F`6dlAW0a2z9?
zXS`@vMgHtQQ4gy6is%Q>t8~kx&Yfl_UHzQ(H=*@hLR3?hf9GH@0tRTn1R8*;k^t`=
zm1z0ZG{k7geO;}G;M1X)TD3~CHk)M<HB-Q(kf58r;CxHqi#PQ29ffSd^1lz?mf>)Y
z<z~+aXUT&&bg*j@;a_0DvbSK}mmu^9oiY~hwEQugT>$+W=;-F_DgINxdy!KJKY+l?
zcsve{G6}+vj2k(8PJ?gYY6TU;>d}m9PF5a^V`fWZqt++a&5dvG8z)(ymKQGEo4Xq?
z1#_~~c-$g8)~X$x-YaqNgl%$RU2ZOUXPNzP#M+A;GqToVVi(M!B$c0~5&s)LKi;~|
z_`<GYe)*$JiA(_X-^Uo;OeWgDG+!D0$-kgijP>s&&#L%v8_-ySJu_;m|KBT6Yl*6Y
z2L9zH|MDX`ModCXMn=)xcEf*X&F}qzB@X3+&x`W}4hu!iD};Z)usTw1e$fp&g|&nI
zWa=*C9~MhS|6Z>Yh`c@`TA=H1nPOG>_X6~UyQ@7@`}x0P;{SOCGIsm_|1Uv|jIx?4
z%f03Q7B2sO=swkx+~~NnZyl5+f?YJ#4rwq@iy)F;VR3eTadvibH^7TC<lh)zg26lY
ziX=QTfpA?VO-&`G;3p+;NB?^abS!XGRcWEt1Pxj4J^R16{w8`a65=nEUZ~q-zyCME
zAub3E;j4bgk_WQ&1M@<23x*4O2apUA*r}Qn(zkvhHPA09H4r2P9s@uvE-4BAE?I7O
z3N(6Pd~*O;NBrmU4d=G5ss^Zp_f(luIlK}Oh)p1<sI7Ch6l8veIdXo@?tX=|fm<b%
zr8udqagnWpi}0Mg?gsQF;0c8EJNOz233H3PyVU~!$D(}`93JTcBN1RLP@)Or2J@Sm
zS47aUm$aDX1eh|66Eo6&T~vL5m>@kpH3kSxp<e7bKudx4?b<gJfK^~H0V_J{^RNH*
z1JJ7O@dOKj?<i2Ffx{`_b&?N=MF<!t<ctA`=xYBE$V-nsY!jXkV86F2Dga4_)8Qlx
znX-}+R$4p}7U2~O=b^qiUs4C~Q~19_=_xl9IEIekoAuHtCmUeNhBXdPFwheT_=5ku
zX3h?%DLRbzn(v7SLcdy>h2mCY1&Zw8Cq#~w0iBPpm`KO<e=QQA_w=T+A@F3K5wXja
zj&r_wcKK$e^S9{W(lG(d4UqaB{M{5HkDRY(X8cHriE@4cm@D}6i-Io|D~%86x4T1o
zH6)~7|Mx|IZ2&On1H#VEE`uh!5vUj}JP@F7PjBd7R!FRifzq?uLCg~o5F`J!hgdtg
z#!whz)@cTs?Ci?I!lJwXE;rD%*VH{8icoKcT~E5too>BrMo~B{bi4wzDO@i1+kiiN
ztdKDJgb@SiI7eF>$3(%ql3L$PfMN!O`#kk5YXpb^XRYgfOGmD98z(0ip+!V8V{4Sd
z!)YPSPrt-%5Wl07ivPfJa=P-ZZE!1Z=sWcElAg_$7ee}eTy-YE#}^&R+p02Y`#&wf
z)^Z&i?-kcp>cm)J5C<3v{oms$gn8T;(i<6O1pZNTbvH`d7_VOaJc32EDN9-41oSJ&
z6v(UI(Ddp0!ZE4LKB??zwes_4)FB~ya#Sq$lxRAbX8ym&{mO9`0m@5Mr^l@caJ5-l
z9mVg~ga0QDHVd1DuL~UXv8yN7dDVtNpFe+QVPj!uV==e1JV8T!iGXBww4H34A^-n9
znw_ZEgt<sajy6_?F(1z|SYQ2oV%^LZElDA^RByn<Htr@TCvUXd6J=J*8Nl^qot#N9
zwfp~mUB20@c`_XSgN(|}K3dPPb}i4;)Q1lrZl-}TCJ}EXE;ecRcensHHcoanPVgZR
z5m2wtfMJg5|MLz42IH`z>O-sx)qRfn1wP*1rw6U;8r2r%9v9~@bfePf7#M4>HMyMY
z_&{Ouqc(n9ISKsG{~o6vEZm;#2n(DG)05L5yF29^vcP*c^W4_6g#(Ep858G}|NT>2
zdwaR{a&ny=1I6zzX)U%zMP~7)!T-;PXnskpl+mX5)<}*7m|HP1(I)M>f<x}kC{9E~
z6MmS+Kd|2tp?0^|=G`%rq|e<m&d0;qwi9Ef;&7m>>HEJA<L%AbC2eb~MIPyV@rN1M
z+bU@@gR@-H&MtGY6?Rp`Fr@JPt77Y#wQB4pGtmLlqpPTn?AQDuaK!%iNRXt~0DA<C
zXR}#L8c<;<X>)*g=zAXAb+}l+wmdWd+;o*J23k-UaCip>XZzNc!7fBg+b>te{l7~M
z{HfGoFzoQ<cjg<f!@)qdEzh*wdGKwdQGb)Sw>;WQ3d<Pgqug+TYi%e5MJ5zEN{KD?
zT@xJ>W7@xiLXE#t-g=piYA`1~t{!={k|92BT$vkx-RNDzPLLw<r!>_fos{{`1&oAs
zf?#8yzXM@aa=-mhJVMD-FueC+gRya>CBIXp7$RQul`oFRj&FK32pc-7*jqF{&zaTI
zD3B=Zt;wltp(<tl86M-55H)0ki>Zu1J-K2|WkEK?c;P|?9KJa@Im}FQC8Q)g%}Jhq
z$wTp6J(%G=at;+9xRsUFS%wBDC#M92yB<vyuW+!{RHxq1_R`bSN;`xq2&6@gpZ*!Q
zRWbZ{ZZ@}~cwAkZUoWa5@j+rO;{dqo3#ThG49Rf6o6VVDezIx~0-{K8#W&Pgd7l5h
zQC35TgWHfw;O#lRPH)<txLg3CpCQQhV5%2B;G=-G^YqdZ+Z0aveGY9&ZcvyMUN$N{
zEv<Sye#NS_uHt;xUo(8hx!&>#_dBa!V{h@0w`vZbbv&E!Wk>0>kBrg&%d}Gdt38aI
z33ki1YRZ-`jW#=EW4`F_@x$kRbMr4>$PHLOep38GPB28c@*d;!2MV8-5MmO37loD4
zTvBGPIszF5vA(cTA6IfV8mSD4T)AZoSL>?N@^BOOogocgA++J>s#~GZ?ChLrmZL6~
zrH97C#^_m6tenOMTe0mAj2|wT$Bc`!mICvnj<q*WB|`n%^fC<%3^)F=;UORFB)am<
zt25w0?$3nCGk}Ez@7pj@yNz3IsI>C!11d-2_wQ|^KemC(EwN?sT*aWsN8gq<2~Zsu
zYqjvoD0#StW>-|uGqK=nbnBIZD>acIAABn;J}G7U#)6!Y>Y-{~%?w&_q0tgPcy@I4
znV5XOozk>s6&IJZIKoreI$>~@kg#DysX`>)BSdc>7_j(ie!Z5>)eJ-4-YyDXA8lx)
zD!cp;AcuOmbmB)K;~--ps{;eYClP=c1R(S`+S-I^O`gLc+Fx&y$}#%N#m)(UvwS*w
zS=B7+$RD}t;ar}))#i9&8Fgk)wFL!k-RwhGq9|i!h(5x?zyva3wXWRu)O2q*(hy9i
z6y}oBQ~2>Aw)mF*z=KMw%j3&Mli%t}Wx#w3tuDwn+Y{=@$gU3q_gfkwFX2L9#@*JZ
zp+r;RM?0{X`<-pZC*kqCoy@O?l7!0syR#mH-|>0e+{e(L$!Xx{iCtNRJOyzJ>?D1G
z^!!HrxeOmyvXNxrh8Kh`pS2<``TP)7n$osdt1qp#I^Q^&=^QArNG(**IH5Aa$HzxR
z4urn-pjq~d%S}rCE@^oJ_L~E-46bq+_<n8ii`j2+xN?_c<XSPP3tBno28uEnXHRss
zo4vIE+OTZQa}J>Z;LTkscXCa*$7_(8LFm!m@j{Dz<@Wr<XRr-1ML?(L*!AHRG^?(W
zQ%`)P^hw8NbsG)1e)gs@>olx1*xcNm@R*oe);SJX?i0w$j^CttjZP%6M0RjlsW+J3
zd_l$RnW<(2kp?%?u`!X4USr0Kjqc^;XWhjt9Gtc13v+U)b$tc^kPP1(O1MZDomiOp
zm#>FCU{0-&vGy<Y46`Rrr>D`IJd^ixS#1|cNp-zzIzf33c`PNYKYhD>;V;_VX|-T}
z=V8i@VBhU*l_|Gx#kPHZB{abb!Whxg#D4_KXk{Rxh9V<{;x&BT3^9rG<dwShZU_<0
zSaMkf4g;6EGsbW62GM%ZIM_L{8}G8fB77q#q^)wZI2Zlr%Wox%UN2nC!1dMU6wa4h
z%Vl=r1HGwW_j}ZQ)yU}i*pD61mCEfPBB}VNKePcmW(tjj_qAYSNqsgaEtR&a=qDxZ
zsjhVyC#P_r2E}cD6Fw@IirYABLEj1u##Wa`riDMJmNV=BQ6M#rJ%zJ|?iulb7v4oO
zxy;`-M;^1J5<o5&7%f70n-ou9lC@HFP{0-U*0IV3M**vX1-LOBuhajGE!7h|5uKcv
zAjH8}(pHk9#0*8YugWnadxp%M_O1ERlP3iglAzrIPw1^twcl#(4b~sb=;){kV|!OH
z7iGr9480vqqNSnlScK{EmJ*i)X@!5c*0q}rFmQ2c%0Z?bKNxc>%S%WKHfw9-#{84!
zDS}(sgpx<#F^9c|zWl2hUL!k1{?L&Bi~i>&F8v(+<+H(9A_vcyl!@xn@{q7&`g*#Q
z;6=i_q^w%X>HI=H9wI7^Olpoy@_~N$AWo#Sge`(>6v2LvPtz)&sXvE&zY-oW;xuxL
zxpjzah`D}MZDP4fZTsxKj9igeU^qj~IV|^H68_cREdfK;peijh$i8U$sS}`;vjtZI
z(@HiHm=+WMgMuiz&glLY$LS}$vhSM{+U*~MwK1Q?(iNOcEv8>7i&c-~pR4&<CNR#~
z!*>Xc8`O<|`HJAq<3p_Q`n57Xr!>hJR!W_yA$J9BH>L_!(o|Fd(=Q%fOVsg@?=k6P
zRVR=hmBr_bw<N*%uU%ptRQUSb+Mw>{9bN(R(3UPx6?lHe5F|MYrwO;VE`7Q_^ZT~?
z7OZK0XJS2^3=Fk@#>-bn-;l$!IFGq!K0`)EzJ-6z_EMwDoyG1PL<daG&2e}YtEf20
zC`-fQ(wNLe`3(-wElpryV5?~>0lN|&tFR6AtrGuk>Zf0$smlVY<YSAI#`UrAc1oIV
zVj}ih9v&ds@6$W0u%DUN*D=6>q2<BTLQijWyxo4UyXt$}T5+lX5NEmH{Dp!3ZsOuJ
zGfs~y$Hm3ioH$P`FVljL=<;9;gxV#@!6EoWL_Z_Y(~LNR<KkXG{{Fqop5KXx_Bmf}
z-d?|mQ&R?`lk*IvqKyHf#+BcGjppz2pl}$5d9f~kHKfuJcw&o@{K)QCz;#&PWNXm2
z28>ka=Xp2P6Vbv$oL%msX<xlQ*r^zl0pTYsS;fsT+*(T7DrYHf2<195D)=Tf*_*N3
zv%u)I*8Ieq`2;TB$|5f-JyA`bU6o&qfRd9e@7RI`Cm58Mqp9hB=ezN>o!!}+8k^Kh
z{7r4?7=<E#Fq3Wuuj-47Ev=q@U0%up=q!VOs8c-6h1uT0A>d>8<n&ZUd3hq<j@*Q@
z*dQu0;?d5IHv39_Z6zNQSwcXP*<2JH2rs~5@qCz%iN&GjsJdv;ZNg5Ak4n;gr=LhY
zF389z9jP9`M7)g5<I%-RNd7(IXB0@wmCaHdpv#X4$Q&Nr`SC(bOUU+%EkUR(Gt*}Q
zL5IHvS&>-4jn-GRyRy34#y09}Lc-gu_fE_!k&%&A1+gEyr~$(g7!<@%q49Ql$)~do
z6(TXU=*lu&!(j0-U<KKj_jRh}tXQ`&n6FmlWClt#^iqCCF_E^eEHAe$Dr+kd5@Rnc
zEm6?Xec@ja5|`H_hs(}11p7R2mNKM*+nd7~oGlS298O^~GyZYk%@rY(zP)3dltcgn
z`Ly*8F#|aVRdgQ+Fgx2+Cl^yuRRP(h{R4gZd31HPO~4D8gN<c!CWRY=O<i3bBm#hH
z0n!`0=k6WB2^F<FJO6xGVf>u~cWc{9)zb5`>8G!DQuL?+!c4LGLE|ANiCEk-<}WAw
z&WZ?=5<zK`nvk-$v)*FZ@d?5sA~?N7F*mgYZf_iH90Y_x=?ytOJ<PdPMbE;G5?Ev=
z7AFyQrV=#j`b_YbmiQemP<{aZ&ops2e-M>~07F}CrM17`)HG92M(P=Nem3aSxm7?Y
zjl6Ep0&{MDy?>1OVwFBy<XaP6UV-^x-70%n7_W_cxrM9O7A7VS^BiCYNMHba!X}DY
zwPpE3y9>&jXz!59$)N(=KheMml;6_S@?Qh=NKZ+b1%0baIN{rygG$}*HkFLLg5{*S
z9gVF%8dfc3<bXn|=8)gq><Hv}2t2kq))Y&p&1%jBjBZ!XS#sK3NSNODdrUqyXoP#R
z@I55;;X~q<(t`MR72k7Z4dpmcear}qT@s)2O`V8Y8%bIQWj-`;BW7ms+{ty`cWR|g
z|3rjn&QSv+e7SN<>S8{}O(iDHC1)^av&ou(5Fa<xRA3_>gO0m*cxZEWUZ$bvaeGaM
z48MN3F_0X6-9GnPBc4UZ&OMZxnu(T&nivX_?9Vsbh2G15^C}cQ>!u-nL+L#XQUyc)
zjGxludcuDL!hZ@HFIY&%VJg1A6Sri^O6%qznt8g;;wDbf5_uxvF5tzv$X8a5M2y?s
zzS%Slo6h9vdo`=2UUnpi9hF^|^9~z=r4x#6)OZQe-Ai`8&H}&ZTXf2TIrHM%^=)QB
zUJWgK|7!23>4u>`M<U91-X*EVI)vnsU_OJ|%$`IWj*DqDo0cG3go0Cg!6C03vH632
z8ANi*wft0qznHobO)9er?Ce6GKktwO+KS|L#-$D#e%fXPkE1|n<*#k<h`1v6q6uU`
z@-z5&9B`BMy-V*HinD8m1s2MjLEd-?kx!fYb`@PFolt_>ccF~s5yyAg6;IuTQqX&&
zQ7-sHSM9%TBmDzmMqUx?#%b$#4V>}4r*V{cffuwd8&!j>5on6gUVLj?YjYtUSSO4!
zLNI}8eJ~GD>6cnEPX>7zGI9qYThz}wd=!&0!|l>umdY3t5ynM4hd}oVbZSs*U&f1a
zyO)%do$<T|TR4lgXWu{uk8x%Eb*7r!0pj!f)m75*tJ~W<7`?zD_f4%FE&&+7k5s@u
z9IWXSpf{718|t;K^?DZ2p`tQ#C}>;X(&CjEE2{0`6MUCMB1fcC*-8MhmfMD<+~W7N
zgZ6M`a&rH0qrsx$;BsgW2CyvG$0rpOl^BdD7a+4sjfq%|>E7Jb{H<&j0YOrM*)E7P
zAA8ntbM1Tutl|2)@-j)~awf{u&Gq#wz$Fh{IY2fWD1;zV$~WJP*~!)va7z{qjEtfq
zqs~vzF5TC!+Q-6td{U&-XZc<GCCE4a;&gUWDFy`m3ne#Npw@(HUDqw;#h=sDvyk(s
ziilg24~4aQ>wh3JnpFO2u4-v)X>zc;iv!lYfk9NH!0qjQBm|^A5}?Fj_477nw?XU!
z#@Gf(F3!#d3tu&}MJbr5PX})3MO6P3<Xi&Zo-JnW5qf90+v{}Z<U%r9GA3FwS!$*3
z;{C%NZtL@a{M}C|95Ktyo^n<#9Q1L-^EFtIQ2TFi=kF2_fLs?b{{RzvS84Y@138x>
zu)uQGav`;&WAZa_tt>vY1i$1m9V_@`O#}hASYRwZCHR(<mTY=#Zew??$#(7!#S2G}
z7mCK*QXbTwy@>;M???;+u-*<1S2cJzkY8Y8;3z97Dyu3xU(e=&R5i)h@DLjF3g~3P
znk|x1=5_v4G-rV!S&9<<50;Wpdj`y}!p8FEW(=sdt`_~@t<1vnFU=k&PE}=w+yj2h
z_&?|R(N8p^Ph5(Vle5MBz{dHhc#1w2YF#&Ce(@q8QS;LStgwH881$<=e>09GoQ=_a
zOY96+4cDsW*Iv!RJ8lV9#i*Dmj*_JnFgW{V*gRNnT1`ShfS#)*uP&=DYf4@>AaBg~
zW9Pl&4gqRFG&=6t37Y4{{5nWJPD>^<yV~9XMJ5c4rz~dQ0`nVIp2KMpT#VObRE!jS
ztkn$-Jqv=y*47A+F?(WN9)?d_t;11zoZPn`r4is+qZ^TwQ<Cw(=*U?;ek$1@CWmt2
z0QMV@G@aYg0ajJCBW)v7oOO8xjfeI*3j<$HR#r>`;*Hx=7MehjoQWO*??Gwg1OgdH
z%jTpwG>9EUyxtXlYIgP}2+hs?(Ok?4**iE1+Sqz(D#Jqj{7ssUugxvY)RYv83#_ai
z&u$-P*l@mfL!+a!m4;g%Sp9w%W{xf_?Cl&_nS6Ey@Ege8;-@D(`I=u;BsMr}+<=q^
zR{slNCjJ5*K8vzzftN>{zcIlH-A}U2?Oa(_m4|_xhn1SR%yb~zVdZ^B!hNIcQQN?P
z-NBXJO8N{H5>o#PIprJcr|Fm!O|IXtLs!q>eOTBy?Hyff?G{AY&4~MJ(rYu3p?DHx
z!%QleUqwwB6$Qb5esIt@Yx(8DJOTFo*J?jn$ocMt*JT~jLhu{*AiqqB%gyI4hQ*LI
z%>)R`h9FcQxNnQq-jyh04sJo`AB`Lp>W_@L-OC3HOlmL*6^}HvY4n~j<9(iuph5uD
z`M3J@v9m&dtFS|K6PGe*_?wytr;BXeZe}}f5CeTW1m6^v=01lObp$RQNkW@(|7`i+
zgUA+wmR4O)zGjOGMu?$G;~UZGvXqyKjJ9(`95|!U4>{?b{e9SPf|kn}ig{q2vA^o3
z61!VSPR<17B8vbAy=l{Y;;=BsvQ)~7W_ZsU%s1%_5Npy*5z36i2m9^CT$uR4J_7gq
z7yeI3SM3%im(w-y{x9S@lRu*J6?s*j$qy~fc{Qr({ik!_LiHJ1NnETf3F$rE(lqzb
z?EqWJDW=P5O}0q!iBpx!G~ej0Qac>fs(%7SjQ=dtHd}9v-1M1mfLZ+zMgfUi2RRYv
zV|>MZ{w;Y3F(tq>-*L<9bbP(^9n@>(wO9AoaH{xzs((NfuW}%d3+7j2LwVAo1!Z<z
z6E~T86`G=MKLQ-oo=>ie>mqI^p?WY}GMivVW)e03^QLfhYBp-1fiBj$gldgjNX;cU
zQUqqzX8G*e|L;Z+r$*8`8)sLsn)ZRhAFNN<-^~f|h_&)dOI@8lbP}wTANI?AlBJH0
zgrQ+u^zlnL8S`n#G0A7Uk{Ic|v~}4$x(SLFF*P>t9~_2*K74@5#Pv|8wynWwWW2m^
z-+AoE2<QOW2r;oLP%FP12;N%85<^>TaJlizNU4qQ9dbhh17gr~KatVxb5I4i+8@>5
z(f7dF74;e*jZauEmXVN0k35}rED^?~@;7#XEBRuR|F#y(77#Kp?<EzJK`<l)V$)+$
z%s^ur`MoJK^Bn}1{7W=sord#*0lNS~HBZ7K^RLz3$@?Wnay+w3lk#d7mL{Qck}wbs
z4iR*Iab;&?^YOV$3ST`56ak}dWTa;7%ntFn*^jK8VNbvu0KJ6*;=jPkloS*wld==0
zb!k9NLgKM(X=p|^9GwLL8cr1;__J00rSJEZWJGifx2?6DuI|$mq0>^6`x(;mCym4f
zhSEo5eStoD+z}9F2olT$>%G?f)*lksvg=%KNlkIX;9fu`I~v5pVKB=j068QiNSa_P
zhau%-jg`(O;;y|r@kuIXCMnS=^NkS~DSmw2(DU%@8GxFsJx*p1y7Ym*VCR6Vy7hNd
zIaD&kG(z&ZY!=|D;yDkLv?d9%C0Zl8^~>|qmZk=8?fj>EE=BmpWJv@oowjfw^|wx(
z<4p<61}(|lWe(BcX3{N3pEEEtL4`m)w;SrlPH+&%Mw6Vz?|#3E11(lKiTFPIt7B|-
zgm7_HYxm#;A)tg45D|Py1(9DT-7zc6N#0wyc(9Nx>J{;F!{hu;q+f!Kn9+>LZb4x#
zSl)1n2m;Sv7*P31KCEx@L!}KK7gZA-O}sVWTnO6Vi`=WyIn*9olz~`?hZI2~lEuH2
zk>hN3=r$xE)D#^4t`1FMnXa+NOS2rYr0EivZ#7+@6@&rUGt<cHeM?<_TRkTki<IC6
z!mmn{IbVsETj=RJ32o|6PS6s?o1&D)Q-}`X?&^br5S*^gyQwIjLB!!=Owa%9aK<tc
zPCvR{g1`~E8KKL$X2S;&*-l^43}7G!195TTLIxXuPfkvl60HvOoYq@Y51`c@ejYa-
zRvu4lW8;-Wi!e0HIzFdT?F#SUXy0vBH%IfEwQDlyE$6t{pn8iVmF*+|q~PJ<oh(-V
zo-Vds_&xjn-umo3uQ30+)0L-+gM}@ycvKS8O67gWy{MmLote+Pws$*5+B?vf{Iqf2
z$JXNEWA~3YK%AebVoJXIZ2L^*w=|9j?B@ahs<R@a1%6io{MTGlo4@V{T#l3&v5@?N
zI@X}~s-grpwuWB}(DAVZ-@ZMb+Zx9l|DlN+H5ox5kHN~$`W)h))S(NNG$12q6B(qE
ziMWv|_!*bQX_(EP9j>eQ)$X#+SBjUNi&5>9JM-DvtW5CVt*$AWlra7ofyv0ZsI%&f
z=aA?&#rjW9p$sr64Fj>pr3MF2^uFg}V5O#DyOSzk3S}-bU#JWdC(V6BdwO!}=m;hP
z&?0DG{mp93tAk6JZSNS;)>PBgoFQeTI>Cp3o8y$i@^0AMP^0iI3FMmR0fwa-sA>#L
zL*shcbRAJ4^QqZVc}pgm@9du6+f>$&p=T`Gv`i9^dQLg($<oi<)*yKZog&aV<`xzT
zm3{W@c-I(ICF;=oAKH2|l|>+nF|gg+O(Wd(O_LYrEBDH*MyZE{HVkR15Z`ytVW1s0
zI`^O-R+9uV8;4ZO;%N$hW5U4^Ta3q1#Qu<_WP`_$Ai6_BOj?faQ_v2hm)9;N3rhnz
zGdpCD`_~0(4cdWtN*vdsT~a5wl4Ipf!cZIxc??*H$k{Pq(Fke{GXkF|aofe#?w;No
zMlb6W6B#~(kxx5CH>)(m`5s<`?`3z%Y5qoVWF8g4E?r+9{=>t=dR*XQ5q;dewT-Q<
zjV)dO05CeZLPM5+^D|`790IZO!u43pXmY~(cAlRj4+rtRy<s8ijoT%Fq=TtdS)#cB
zcVgqCZ3bG#7vC-~!<@uy0I#8}tV&PU>oEw=>epWRHz19PGAom8&+5_UgP0o+YA_tc
zlj3<lx_p%_{dnJTvYR;X-@Vzp{SKBG4booXhg!V}59&2Ji@VmWv%BGEEay<p(V+LA
zL_w#n^fONQs~#})kaN#I9vvOsNgGzm>-cfXvEs^!$^U#T3Uv3LylzO53DoU>{Os5&
zY{~UOEj$hqf#qKFw<b%l*HKc3SJT*j1Z5%=k($8BOyh&xh0B~6aGqe2xC>CnAM_Cs
z5u^~ny*0Fza*DIdK{Y@V_#84|<>sY9gU~<thKH-k8Hu`ok#toJ^KN8NT;1J}5~l=H
z(88jh%0Y6L*B7*>&cE!P%EZK$C@51WpVpURZJuwpG>Pr*5&{4L(fNK~<16%wl(~SE
z84TRUq(obN{qn;;>B3A67&d?vysR9v&!E4xm6YurJ1%a{_vJg(n09aPEJJJ>&-<yx
zNqH&ZgPAm0nPaT%{2+38A4H^3dBvYS>tb%^I-YCj2@KS9-p<w_r`-OKXh4p7gB1PP
zv8DtEJ1$mD6-6mPf-<c6@e$Pj8WFJ2UsQ5TgqBo(zg)Z{0>0PKbGU1t4)O^4myk)q
z+(J#ESpKhEQll@KXb`pNZP|&$vj-onyqWeJJ9HIsH-A+<YiF;goRY_PXq6{{VKTY-
z#U_>}MY)CDU&?}Uho5}|&=y-&4#;up>7*em%WbG{h<xz#eGawm_(=&1NBbw9b~_Yw
z=T~}}b&sOsBOZs`H=muI?CcBzf<<OzFg7}-`-=_Ks`qxfM4S6Wk0-jlW)$3Yz3-1&
zmtS8UylydE0;K3827Mf(K&=z+>9dg2C2|HzXHe=6;4;^`9Fd)Bzl7T`Ywf<gLM<&6
zAwR`P3bwQF47<OPunMzzbCj?Cr|Hl_EbzGmvW=WG8qVN0FAMH>EC}Ae$1NsvJeLsG
z>9p*`a%TSL%%@SlQ6#9y$HMR6<1TEz*w;h)-8U{8+n@~V;vTOkG-T+=&k$eFM^mQ5
zPLOWv=GpD?ch^4JyX!o5cJKDG*!V->G1a|>9%lr9W-rV9g#3~0P3y|s)P)NH3XV?~
ztUoHUjw-S9)`<7FOmP~FcM&$Phy!r9Ufmgdf0%@U(Dl}hVn-GyB@SV~1s&$Ul*m}-
z!lY#RPwZ{v*(rKm<U^!jk7JHa{dD~|i&q5L8ev)oMtvtn{<PV!e&-2okz`QrL6PSD
zR|`wc?|1<}DC6E)NNau05&xeS;CSFNT^7>2d`>)l)dH_}HBGLgt&`<>_E%JQx}kRo
zLYHSD2#3#8BdH=;z8G(@2o3+>KGW7+mrz?L7C`jl?@8YRY8^lKQ=_IvL<W37uQ<Ca
zFe=|P){y^3l2BXY(vi5_+&p@0jC;iewMvj5%s1QIM135T?M<G3am2Isj*~PkCBVNI
zdaI-voRmb*!Cz6-^sKHNj{$yJBJJq4bI)+w(m))r9~PGiLrMCk9HhZ)uphc8FKmFM
zINzC+f*2|*CpFTpSEd;Cr-E^ibH12>@i|y;ggsIMU$R4h7M(8q6f_tP3Y5zksj$Fz
zfk2S?mZlaF>6n6IYpg)-c3?`3XQ<sp6~x$j!jOR)06c+4VJL{I!O5JMh)9SBelIPY
z*U|Z^R=E3;#6Aa+|K7=R;T>3iVkDb@QGHWKMRxXPv=~}r&4?RZXwKOEQZ`73g@A90
zG(KEHAB*XI{4}_n8Q$SQ{WMnT6fc}ARy9C`bF$7f^ey67YaqoPO;mqMN~nyA#*T$e
zATS#X6(y`H&B5)yZ^T5nJ~8Sd2g@^w*9|`he{{4+R0R6o{w{F?Hn@PkMW!#Se}{)|
zZ*QSrm5tuqGCBYqb9Rn*S$k{@(?9$r*Z#0&HR0=ZD|mGP64JAZic;Q_;%yIUfl7kh
z-0<thqwR?)pno!bXWjI10O$&UIRKyc3c?b%xbOUa=2lG+UfaUqpl6M4^UOf@_5(X_
z=7?Td#z;4uUCSHZHKXGLczrW9rlP#O3pAv=ZVnN^Pf;>3Fyw?0>+wYXc7Kvi%f2}u
z(>78Yp^+UPT=9JMQ$j>VgO-tk&<5nNbS7!5tN)0LGdkxtI&DT=87y^MP$(C+n^<?b
za(%<k=d=>uE3O~nXe+|IipLcvkCs(aTY(HAr#!jmxhiZEXyGrr{=y9J;2mso)1N-V
zYdtPK-EK;TgT&uPhJOI(;5lrJq@ioq)dd;^&FJd>pov4%j0*@LgaB^1wU-l3l}*9C
zLE>Mrp}zt5O-GU><wMVx(w`WTUxblpzGu?&=<E}A?mbs{1L--3I7Oj1kV5g|c>X1q
zDZfy#*2?Efen_E0Z;l_k!hLoKr#L53k8TGa7y>}+T17{%WuA&U;g87mggh0fT$jX3
z@F}-GC%d{vCrC=j-;F&#3@<?3+afn%z`f@lSCf63Pzn$FMhe1Ov{>&9N6rt9=KPI?
z^*`xfM^BMqA_Se<8e+!)Y0i&eRR9(=b+Azg^HVBPL#=}5xeBlU`aE_H4|lh>v$C?D
zZZA7TqK)f;IgJzn$74`se(cJ^*Wf<X@dSV`D+lYU%gmOx?z>XicVXWXgE0T04nVL_
z^r`tdBzu2tw$N})^dqFbZTK-YGDd~tP^a0vuuz9-q_(ck^X}o3y1`(Bq(>zq`Uvi3
z5swhz$j3u|AVNuhJR~U=L%z!)f9WUL<ho9b6TATeJP!9QElv5_)GNFO=+#CR8@;im
zaPhZyLPA1X@zR3=k{SvM?gsNRNRVEL3qKU~fH9DT8neAi#yv2PP+@nr#OJQl=*R`e
zrxhcBUX`7|0$?$SwBFaIWO0vDsSjGol%L5yK<cIyq-3-lWV9E5$DWC6yh4KO4v7`K
zD}raS{UMz_X8gnZ+x0bR;!@9{-w6f=PQCMC&tYn6%}1)hJ`;e_0Imu7)bYlETb;;q
z*xYXS?bNJ1GZn`(jXv&qK!CwN^JzmyMFCU0wo&5iMt5OhGU75L$`NrMr^l!-QH>f5
z<zy_D(rDWCpl@eB@;$wB-3)BeX`b&cm$S#RuM2YyXHMYn+{|YR-)}aAZltfD9Ui8x
z2V61gmcIpJtkDxTFHSg|2D623&_2ZejM>`6u64bwb#*2r+%=JV+Lq&c!OM#poHe%N
z<KxrIA9mkN;f;Gn=fxj-q3?d-%q|W1fRm29Zht>ulzl_Iy0s4AI&5s?$0NPAHO>;<
zbz?6nRngl4Dd7(;Xp{)XTJDM>%H}36T>xT(X`OwK>(k~QI54;}&N7JzB|pdXY18XY
z+#z|K@Zrg3O0TZ5jOW}}!a;!aeI`8-Xukg~-$p!J;S5EAO67-dBQIV<wJ9gx4zfod
zkFoHOyUDIJT<v13N7+$6RM{W_PA#I;??ZYds7{ND_*uBIp@`h#G?de+;^OAd-W{0W
z^l@vL!27Hn-~#IW;Q=6^00l&4ZM@+Va3rhp%XwLTawYGk9mEH%Iz;xEUv=HJhllIV
zOGoZN5%OtTOG?673c5ItJ4bAQ0EH*b=4}3Tzz7-C{AD_;Zc220NqvkT`)mXT0&<_;
zVQB?;C~S9MFx?5^CP2FMi{he2?DEe(y|;~$4^XS9r$FPxb%cbbcrPKeodZ-3d>ZtS
zHRI_%(JoOo$`LZZ!C^g=(C<syI?OykVdpXZ5t_yFG>GR!hF)$I0#M}8t{=g#O>c3;
zwQ}0ZOj(8j8hn)+#{N3xgPf(fuziQzm(O3pT~?x{G|mbl9zjZplDhIKRp6|k<D1C$
zqbc04AwCaSJY;x3p*qin!}|?FZS5t<dSv;eGki<bU(CKT=V-BBK|v>0HU}+|76hBp
z^#}DTKyTl^yS|&!1k@h~XKqgb!b`;b{5LFkAp<j(aj`0ZS5s4+xv#^2cI-s4wG8$r
zAn^j8*L7JO0xWQVH1*e8M0*5_=AWqY#QAy*0>*a?yA?@?a?e=suDA5>UWBKrDyv$3
zYPkh{e6I2?Gj4H6lgb2J1sd_)$$i!ymU}WMR}Cop1>JT&Y0j(?g64fNF!sk&WQ{I*
zsA$H-j+}0zfYS*Y<=@eF{5^A1Xa7H*z5^P||Bd^Wj8bGvMkvYNWG7MfmJnrU?@dx>
zc6Rob2-$nf-g|G^d-Gnu|9jroaq86JdG6;P-|uyOuFrsb-pp+CE$PQNQ?>QAZse1v
zCkl0FmY~*ucs4akQbG1JE<vp5;7XQN|4ni&AiJk85yDD@e7bo{{5u!`^Ma?l-UJFV
z-nJ?jbh0-%#6un2T7dIfmAY_^rAH2eF0I&bNug&V$3R}?k_jZ}XB=^!FdAatj&Po9
z58RKkfE*0c`>W*BwQ)clNb=C3R=bP(+qpSxC3qWXUGJyMb)`oX!Qm5Pb)k7eX4ffQ
zS5NMrxAwG9jP{R5YVSmz;vm5JKt6#(!|%~=xnyEV&&dX67cQY|#X&ychk*kH%0nia
z^Um^_P*>?h!PC`!k%P&|VO71FA!2v+jkQe_#NNIHm-Ox_7a1uvjapOJz{-wa(cM59
zm=~^(<cXQzT{dM*>*3_)qA8NZ<rp1(6Kad`{C((ufbaK&n7fGgS%t2~A%l!EnU4iu
zHcD`-_Sxn}_oW2DKebkZ3Q#H&aS<*_t@Ow?4d5-Ei@kn+wbsZVB%A-P(n;p5?fz2j
zHyC+F2Aagbt4OHi<$Lz>t)%tJa3Hed6x1Qya_PUy@~SCj>b(z+*pJyqA*9C<%qp6z
z@8;c}3{0ms`S?V*K;N3UCn8n)vYtUgdyLR&i7SEB689OfLcWZljb!JMt%&6^=9;u_
z^uFA3f4d_Yq3skCb{An(D=d$+Z%rQ^NFT*Vcy~6AhIxdIFFXbL_pqjGtap9={kAv$
zQHMV*F;Qhjt#qe=-Y__^Mp~vvv%(o4$;zgu`5V_V=Y#KqO~G$-enm6$Y4+%-n~L3O
zChylmD=y1+PW0|1!SrcF8jqp%G!g2jR-5g-OGl3R-@L6($5Bob=65FTgw{SfsXcy@
zU^}sQ8xb36VRA1bLb5pN5or3U(Y#^ycW)q`frI}KpW>JmEek_%#V7hIEI}G^ylV5^
z^WEqoK9(+((+I4IfG?wH!T&Ze5I)^1%Od2rg4^cTMPWcp@gMuyhPzZnJ?jZ$KiAu*
z-)rmI*L>b!Gk7#)Gdl)!5A;vz{s9Suy!^|8INAO>&A1~uxt2^vZ&!=k9iTQSf7X_`
z7n~WsLw0x>h@Qg7|A)I)ZR6w85{^wm%i!b5_TT*J9(q2*2KkD$(G@q>B>qi`<@M+J
zk0~w>eC0%$E!(Om{>nV9k1~JM4r%EXAqKK0S8qJmEXIDDs~E_>Q?;MorR=6MfiVQn
zXdcBi-2{`%TaAfm59W8}D5xYRKh-mX$VpZl=60^Mpp0PNvcL;dJi|b*to+}8Ch86-
zSf#UL{+}sY8m6J~Y4~EBXi!!Zy@NnhLfLhjv5Su(P#xm6rhuFU;jVhu%J;ewJ$NCk
z-tVCE$FE8pDia{h4=I12TcF%mQTcxJ<P>6W)yzg^ljS2>v=%r1t-(yVv8!5VYa=-R
z)mzDihCw=7A>)OkOMywv{{ApTB-JA&IIySH+t9)ll~Ta@HUb?Pl2B{?;znVOm0df7
zBMd}XSe_KY1Rs(|2Mub1J$8Vj7eqPj^WS{)M?`q*kFUG0xpTGdCq^sEzxiXn&y372
z&%dqy@U+l4bu(vi@BqFE*GuLt%a$Wmu{d<o<`*V4!@}mV<+Sr=D%4&-ugNUUOfFCd
z<|+`m5X;57EkO)9A@?JCI?lph)w+6mdN4kGbo16F*4w>P1~r3rOKj{TLH<&t&Oe=d
zE#_G=ZXa%+70^5{T&u|AwR}_WAtVo0$831@SC(0{s_zFh1BN$U<xDZA_tA8x>F={=
zT&Pa&;A^ERbd8VhbvoBHmzwD=<gr#UAQk29O^filZQSb@zJERGNbOnI-E!2z!^$P7
zq%2lelU$IW?=e9;;kQ`O@NLZJ*$ViHCGb^TLo#xv2OkSFt)kjYTXS={#Yjm3%!zz3
z(0zg_e4AS&PS)2y;_sX^q=YS*l*-c$E=dv9?B1=i8c7hTTkrm{%hp3g`1mUfCM%rI
zawn=ZG&TQJrAWfPaLb`9Np5d_-ZNGBp%c1yn;tUyX8|Gd>Klr&I=zrXSY@Q36UfS|
zu^Jm^0t0~f_)9=^dxWB5{6b+e4D*($B8MMrGD?z)RV8_IqifpcmsZb#*EXk{*~O1T
z_p}|d#YcEJA#Z1OJ+Q*#U-yst)(!w$icPy{i3Hw#3O@eZ$d6ba%qGB9;T3SgB_Rv4
zJt5cN1`=L_@+%H$)*^#0fB)E<fll*-gTtGX9;P}vqE>$YQMD<smeu2V3$GtM3VCsg
zzDQgJ9xCWLpMr64a2#giMwHB&c^ZW24%!bZ+xD)eCZNna`6dbq51P~k7y{<dOGYl!
zzN+#9;?GGF!E+RMDJ&%<>zBzZSs7wNTy73FvdpZ}iZ2_HRmZA$mM3l;_BTFRsOy+&
zeSTP3N3|)~craI~j&tix1IngW57s+MUx9LUx2aoGFMx?j<`^V6Q0f;ZM*Hq-)$r%6
z@v<1|0h!0&Lb#Yve*rIn&!^&zy6=Obk(w2cT{h?QGjIbo`S@UT)EAc1kM!-GrKNLo
z?;mA$#zsh8E9`S~ayPd;IXpTnC@3&C&Z1g-h5Dl5{2T-LP&D-{7dKw`czZ8v^oUO$
z^7HXG(&)y;YQU%gIHhNLk7AjvNA_<-;f(Y|#Q$o~cjaF|K{nzFygt#~8SvgLq(rTo
zi*X6fQL$GmCvYmcQ}WtMGFMgIn^W1$Vk%DY8V5@=y-N9q(l-H7`!>6rjl9C|+#^H@
z%j)wC-A^kdo$l|yb&NbaQgm*9l~S2$@e(hK=eV?t$Yr;1&vf`V+NZM`^d3BIDti=-
zVukmg>ve7_e@WxkpJnBJZhyr^&k%B7oRa1@wQXPA&@x^N3z1M^k5hvVE0Fh;=O1(p
z;c4EdP{8l%Xt#Ogf`jf=9&Q}mtaB_o)7{Z7W!Ot&@<H?`wZtk7`iF`R0<(~&2^cz3
zJQliPdhu#}uQ1OakYam*3E5vvfC)!)lPMD7YCZ62*0B>a06J{_1__j&a(nM!Hh9Tl
zzg$chX&>1WUaUBGE!NF~?iCst8rW15pOOClV_LAZnUv(~{GS%ReF@AgPf!tH3jlip
z-I*gj{n4eDXz1i)-O$j;{*j@eASqKbGY;-S32!o3H^|B3mubD*E;0adX<I${7t3eQ
zt;&JlSa905cY8X+(BSy!2rg=ObVhcnAo($Ki-ek*gQK$P3tbK{umaE_Z2u%2dK@%#
zA9##HUu^)M5-nkOeDK9vBAl03=wQXn#KdFa!7Fl#l34wWoD_ol<Z22kx>@<DIk~G#
zpa=UI$pRoC!R_W&VUmF6ytg$Zh_{c=rB?%xm0{oq^fjDdf9MDcN(xC@OOArN0Ro1_
z*!6D%PHAMM8qmo)+F#MclwWIVtA@s(^Z2zyGQYb=9@jfC5THqpkdTnw!Mu(7=_OWf
z=-8VQjU!uI+Zrny(W@VrP$9$O2`T2HpKlDS7PYu*rOIbp?Y;))$xQ+GGvKv*-atS>
z00&~&9qpZFFAHRDBU3nuk)e?{KJEGEq<udK>INICtejjU(=54%-^Bl{t*zY+_?B&9
zw5Vgq!@&id2N7pGJzGr|iOd+k<SbuS1SkbCo7aeqd-yy<7`^{isQw-F#@5!Kse*wq
z3DnGN&o~(56h%$P=F#Bj+}th$mNhssBRP3W#af>n9vu8GD9HXvziuRFNE9F-e13K|
zHag15r5v;NLt|pRks4@-<K^Y$1{%?9lwma}RC?@ASCeCHWO+O_o{f!-K=%#GX9XTH
z96F@))KjthHJO94ZC;MRRd01+4xb>0;SnPTio1}WO#Q@sLuqrT^APAy#99xJj{N-m
zzfIjOZ*3)lRwblq4rqI!eY<ZM?N7tL-nejuO%zs&mAC4Wz5y^FSXw5eZ~IHVCAxiI
z`;9VQVl;@)n#H~Xh#p+E4>t=fZRV|8x4PLm#5Z2Q?jXP?e12glEN*Y^Zmj;*yTZRB
z=Qg6Jw**51xsiT(RYhN4|4Y`X2;blL!q{#j6%{TiALb{<C;V=v{K)0Bxom~bIwvn%
zMM0G;ZV1~`HuD_)3veJiUP+BJMr+LU5|H9H_q0~J{-s8;a&ulwzBV!0eu(EM>A#mD
zex@s|`^UjnytT!_*1iDG4;)E4*F1IktID%O!%_tH@hY9pkh;CcSTVZ^LS8OML3tF@
zZF|L1E71<CywaYavM|0s*6MMxB=zSFXj)iM0-7nNYfYN_Un*tYOO@t(@j?$69@UpO
zGRH%<iIr1NYa5w6=0q0%27WYo$9Q2|V@5#WnAsQY|Da{e+ttaKjxJnQ!f<6i;=zBM
zb7=3gOVB{N@pJz^X;*201P75|{Cx?Wt(Y*S)TAU!E34@md#fe(2g*sOQS~h?&hu?G
zacV4>N#P^%S6WFp3~)#ucgLQc-n@abd-k6k=QJd{4<uV~)nI&bxo7=9WPHz>U5;}4
zc?MbP1w-G+ko#$qdEQWNHup_5eY|KYfz7n7WHjqxB{R(S>$}zh>`&})0xADsqUqme
z;~qS=^QiA=6ddvfv6gG5>p*n<c7bk(_tq8#<@Ic~`Av{s-3BsjcLHB~3wNJu4Bd4m
zOp>mwm%qgL%~A4fA0Hga{wMQE3=S=;2`#%)*6`fB{z6~7M@dASF@BridvHTbi+_zD
z3?RQXem@1@SadHSB<2GxK$+*p?RJmPch^<puTMG8-tNrQH*}s14Pa!5qieq-xc&dX
z+Ie*7FCp{Tr!k)2rm)uIAq|6JPTUQ&C2g$#y%-MO4%w10+JAyEEMce9o0G*hCk>0K
zbaXRR<_SXZ=(H6U4>2YOFB09l75ovTY5L~W%(Q34GS<g?>5ADML@avCkMH}m=fr+V
zQ$}-&@e{iz79$3TElhLn9NLAFy`K5xyIQuqOr)gSD3>vR<{&~FE?9kO#XDIs)KM#b
z>V#&ozlRm7|N7Q$aJnD3a|X*v6XD{P#2&7>Ziasd^DmqqoqVw{yW8Sm<XdPk)W6ML
zol&aYTUS@t;A9i>ZG(;uD%w8H$r+>$UNNS8x`vG2Zgi>x1L4lq;2Ki7V2=I}h>l?1
zyhz_S#32{<N>c&cCq@&(QY?fM**Ul`_EO}pC_ya75H$v+(e=T4It~X-N|GzimoG25
zOG~V5te7;aKAmR%PMW<aFKQL;9hj)~&`HYAPpsE5GS2akPTf7iqNih242il#C+)(2
zvU25#0{%#TMx2L%fe>b(U0~Z6Wx@#&Oi;PveMkYh*!5~1aD#!Ty4+JN)}MVPfq5&~
zU+UHJ!osBisf>uck+VTiS=p$ESn7RJ>Iq?c>e<}|6HWk}6$*f^&eb6;s`>471<t_D
z*1PW7@^2P5mS3hT{?+p@&+cpGjBU^Pc-|^o!I9~K8c55*Fw(a+1hdU2S$$<Lr(E1z
z6M3ifq{BD;q+T&I{TwZ@Yn(B+vp)o$E?e)wsSp=wrdpmN-aPRCwBHTd81kTdc`_E^
z`3-OmwJs-9`pnJE@!MmAyyn#MI+<$k(kDpDjEIiDO~y`1hSrvE>Z$-A$yG16RpcV2
zS7glTdnjEEiv`)n#t6%G$@b3Q@d|Gimb>7saEpg{gfgnzIP219x=P`S2uSsY*bW|7
zo040VIKgla;<!J5U+#nkb;1#&61VE`1BVR^A8&}w7Rv5(<P`?p&cv~b>X5YEV<8MY
z42=D2mPfTotGfeHPy$?Mi-*U92N-cya^7BIzmdD3OPF{j)Z^LQi?y<LW}6F{EGVC3
zGc6|c9uSaT?G<AntLsL=`F*2#>hCKJUh2F$Nf$PkgO|p~#)jPDx(4<LT{eK`w-Tm7
zoyC$Upo1)&+wQ~GXXkqKafv<db)ON1u-o53qkG(AdDGRp`%iE>+y5RPVV!BnJYk~g
zXcc~=ZLwS{oeDb6TQ~<|L>c3e9vm}L3wwK(GiM9)TV7AbC-^hQiD2crHUEK{C+|np
z_j`#SU&z;Bot&O*O_v=&RT!7qqdWnsN}+}ZQ6{Eez@66A`~h;O2GP%*GVr%Kh0sv3
z9WSG*wnfIFX@FnaJTpD>;So*)Vz8N`zr8s%-FKe=4@fqDKR2%Z>-^2h^Il&c4Y>Zk
z|6(gZ<@Bex^c>?WuHP4x4k8Zi{E6{#ErF_uasF3i8>&T}h99qE$XT8~7Nn++3=g*%
zI8)gkQ`H!`M?m^+rnjjDxY{3dOp|zy8fj9xU01D?G~X_-Q%+7yNa*PobMnBVdET(G
z0MC3Q^@j=<=MzV#<<62`Wwn>XwU+jBlCM!e0ahip^0w0YDJ7+aw#RkH2!$B+`}5^P
zL|#Sf0U6O#X4(q9J^>_YU=7BtQ121k{u&qO0-KKZY9N%Tz?)0eT{y=;*H3#y%=DWG
zmq?Tw?9J*ze8K8`E)<tbWhp&inMPqqvV(y-WnM4sF>z<HDD!jKcb-8<wy?t>BFXa8
zCuqQ;>H`8QTOX)zcXoHF(Ez$CEh(ny0MP)wq<@>Ws_Z+-w;Av|YzlKL-R1=B`U%C4
z2Wqp6iaMGbD^pVk21i9Sr8%kjWL4CyBFq1t9MUu8J1*dhqz*t><lZ#l#^$<>mCHcy
zpmg1WOyJ#nJRu1Q?+kfZ@M$i~zAVDxa9LSeZE0^|ZZ=h*k7_MpYj0m|?=dhw-qO-y
zVEjHMpFdn($LZzJgPFRy&^eeB6qghe6A%IEpSnWBM8ieiNbIFPm*=NMUE{obejDnk
z?2<ef5SCVE`b7D-)N3*<-#hqP<Z@kC2Y6PhE}mjP$LXo#NWMB19o;9*)}#>MgIrTK
z+|ZX;SI?Qbao;Bo{^>M}GgVUL`1<vx#@mVLxPFczxQx)AKE1X3n0vof=)MqojI{ga
zRYpnPJE<4(Ah+0{B@K-H5sMoa!pSK{M2xS+>QM6M`y+&tllv6?%PUFWE_M#5vw1ry
zlL-(KbVM<NQ~fI+oZEhF?Ol`<6sk%JZ$EG(#s>zDvHh5;;^VJm?EE%kqI~-+9(i0y
zYyy{PDhO7Xe<v`Vs7VOpVYK?3ZPd*sMLCE#=@WX@heUpVQ~lu<G4aaDVfw%tfDcGr
z_pb-|VT!3A@JVdX+Lr=~IR`^-(4@?ROpxS`chJ#+lMUmon*vA`6TP3m44tEf^!%*j
z?9Ls8U{UYaFUAmdN<O~6wba<=e}qy1y@V{~ZBf9$7La8a>>VCqb^c|loBz#UYA&Wp
zz|UU^q+Vd+?=Ug~QUN{W8^)6I@})rSUPhbnx>-wRS-=w4DFmH_IK`%m-Dq%lcn}yZ
zq4|h^S5t)r*Tx_Y5}ar*2ZkCei+;LH{acC|D~H>DQd6_UBr?Fy9}#)=M$c7|9!SRw
zjsYaXNAIpo&9y7(o-z~R;#+L*bdC+#bhH$|VLaL`QbT?N{p=3<z~JQkCh?=Xo>J##
zJj8Kp1Jo_p$ecofKA{To+kA+<X=T+f+(z*uA9B-2oQQx3wN89O<;NpD>uSY;q7U<U
z@NhMWy<FDNX#d-eh?r<pW|r3a{_Yhpk;)`yzAMuOU@Suvi^2U7we(t#*=7?5xGwMt
zDE05x?vIf$+5>f}kX;e!1ql=pu9EzlFU{{|zTJTf8<P%&KDy6KBBF@dUB~K}fnUE2
zsH({kz_(sBP)oJb?>JVam-~f!G8f)*5@D08HaVwF`_q{_Irxq7nfe-Bxloz5eRYt>
zKv{@nSs2~M4#>*>WoT%qp(E%L8;Nf12Vu%aj>?kG8DM<spML%D;fwbQ1s*y^*L6wo
z5$3<{r#-s<N)x~R5;WUz%TIV{PP&}v0lN^ET&h*FJO7Uh@OjT87?9xNnI5;qt!z3a
zG%T!+WH)|*<(Rc9!Ob&coQ&YR@Z8+;4Rjg~{$QUtDUnZ!&lst~|3bEOj)f6`Gew}q
z<D{;rsj)hbJY0xURaYrGud^bJ&&y4-oLw~rOW(;s;3==y?Lt8bI8nlM&F{A8+23t(
zYTXOFCnrtcVVQ^=E?FG3Wh9yy8A*&yI2>Q2!r*t&a%&vc0;Yf-EG474nzePk!1d(j
z;zmBfZ%}%a`_}7#-4x`rv$3-`2&l=vs}Bt~ln^(93ukvX@Od(B0C5+;-2r0r<y#iV
zqkiocV-VB|U)sm`_y7|y{ld)3$npDTEaAC>#ra0b;lWX@-8K~igTl+#Y;4cEI-bn7
z;`@YB*;;>eR)6^zA&B@sJUoP22aKcBPYxhk8ygC5s`rH2=x$(lUcxxXpL>DJ!__{}
zM?u&A_;C|-IBuw)^R>=~+C+|vm|QP5hW6F+V3Sh}!-E4!5V{K9G^eH{>DIcS;p2WT
zs~Q{8MoNgE7@6R-9vpzkYiX@){4+Mq@-gBBV}|^xDon?!gzh_}fhFg4eue&$GIlDk
zp%e44P}U}aiXV6C2?u{^N?}EHYPe`}+bb#dRy%-h;m|<*06Wz!M^_gKLtD-HA_)h=
zMSWw=@q`yoKt2Jh;64_(A_nyr-i|{XhUyFk5&XyG5ANSrP!8dDooD3de_*3zVKL)v
z=EBccd0_hoOoxCy6ihK$WF6Br+Z$5!)6mxRa}weKsw&VJ^7Ct`+EYAWSch=9IE}YF
z+q}KPtU7v0aS?Hp_sEVW<TCG-W-mt$Z@L7-K)9{_RXpGR10EhVeU-qpFVPh_lKLuL
zH8h!Ydj6rF=ITybuigxn$dn{jc>$(D!Fz^E6aeKN2xt$m?iAPV`^4t3BPTa+{x{4n
zbx2FohX7Y9_VTKff8a_0gUPY3j_JAkD8Ue;2m{<$M~wfvQ8{U2GZU7dS)6SBh95aW
zB?lxUw3Ab?Zb{&;MDzYqYwvvUSN)4{Vr|&>ux`K<H5Iv413|JlvMsQKg$b2oWZ1pn
zAQ>QG*0^mRue~KT+a!lm4Gk&t)-=p(H_7AKtRSbP-k@F+Ioa$#1d+G;{ioRWN8nfh
z-Mb493n~&QK4G(}_d++kEc%tIxzKs^k2}i$cKkSMA#KsmAJfLhO1!zL(&dzIvX;%h
z;<2BXm)Mz<7Ko#NRbqoQSZ1UjIA1;QDo)V9R5^a&agQ`-Bzo{5QBy-rE*+br76NAy
z^3s!sXM`<xPJ)3~26Z$CiREF-R918u^q8OqZ&{86Kn5WpCH41{x{1(7v=16V(__r}
zR>=k{UbZyBtbWjlcXSNo0-p6o;MQ?)B%#WzUHD2<`cFZ&({^$=aQVc~Dgpy#47)FD
z;CATqa^b6}_@O1ZfUgAn0E+-A5PLw#2e21H9$v#XK~MklnDUuSvOJg<NVT`ajpSfD
z_S<fN55dF^V9d{g<_t(-nSDlHUN7LcDJG_x<}A@Hq}YpWP?7+|uJu3=P~FsMrsB@7
z5aZ%<^RWGKa1=LW1rbmiX4lvdK611*c>{SRtPm3Qk~4Pk`$6HJB;u%{`BqB_0|CC;
zi+5EE;fkQwKuJpaU6xWvim2wlX?T4D^l~RnLjyi>0aD^`*_o-vx*E^VBKDp6I4UY#
z;1m5{Fs`zq0-gfcKdg8+a+ai83LpomYl6qp(kKf`5~srz*ihPE6&Dw83%cf{7k9~P
zXJ%#|Pk#N0wfr_-(2d|f@B`MY;oi~Qhb*w(1!F!2Vjda-nnB3?`nSH$!`}Prrv=G5
zDyJB2Yt!Q!tniSc=&TIi{nUtpJbCgL)*Yb8evG_7uemU{L`*<h@Sl_cq~AiW^LNl$
zz0oU9(JWBC8o_5IfQ~(3+P!p}e|DE~uaMpEZ#I6FG~iM&ZC_b@Br~Awunz*LJ8rv8
z=G6Z-uZfjnM%G^F>8IssUzu%GLi^orsAHPtu-&4sob>!u0|qwal#^lih~uD^m%sWV
znf@^=dvS4zI4bhzPkJRfWj(k5J;NuH&mxb_-(4+r-a>ZnVqWPP*PzsM^10P({>;SL
z1|GlDhT&-@%R8{I0oBV1VD!K|fBo7Y@^c<P%B<(_*$nGvw-{NO&g&^JFN0o!y~r9R
zDXUCcO34k*gBr`#=Wmo3fQ#M;^=zMx*GL$hnXWZFzZcUfFj9>X?BgRLOEG2PaU8p^
z<N^$L;(Llp8m*Jh-@NMWyDIs*rj4g5F4labt{J@ls!sy;InH5rVSPPBJWI*ODOTcg
zaueCuepdwi^5LO<#PWJB9A>rZekUKuCMz2+BT_+XV3lx(goJ#{TcR08L&S0$dG9Wo
zcRe=Z?I5p^xbE1bVj}5zh<U+p`Hih%1qt#>_>z|w5mAsAdrrExwPj9G)x8~zYC)o`
zM`mbQ<5b)HcGAISM(H@g&0(^d+i+Q#HsEjy4}kgd3MZh0Wd@ti)H)eAhoO;^e>d#W
zMno{aHhd_6oX4$qwT_p2#tOKhU(yE0E6(yIeVY*xeHFX+R({#Ly|Bz^2|3I+#H1?4
z)BY(5?GrMTBiplg(K9jYTX>^<zU%kbAzypPWw*0`5~!tUXg(++pb55<i4RL@*w4h7
zq?m!27i6bP<c`HOnl<kawp0m$a*u`jO@_Nuq_EhBCgdG$+*itb$6F3&JgXmvR(}Bg
zM%+VziVy|ZE%e`KV^iS7>Fi?RB<0hC;|F8HiQfij$Vv72C)w5GBSD|W`|mBzJ5ReW
zHKO|N+>YO@U&w?H$rLvnFXX;67yt9cfnB12%k*S_f8`D&+eI#COmOE5JM8Vi+wzjh
zis-yDd_HFh?T+G1mFv+?w+aGSVs5_0taW+5>m|x;Q>a_;ww42#aK^Toh@3<~>q2rK
zn1MchqphH>um9fu->9dQVK5F?uf<1mzd*T-OJ$%Pz}n5rID9zQwUURUtfG7~?skCG
z^ohqke0;Yp=j*M&0REDTb_1I&L0jBG=<Rpg)3MyD6PdLID3NIlZv=?|e@_x316qeX
z?K1n_Khe=fl5fRT?Cv5>PcoE82I(0X$^~nV8@kkW9B&}tsr@t?<icKpiGsE@4sOE-
zPtD904JoFlUS66lj1RP8AJH8Ffg1Oy1S|w=o#zJEbTMdUNe0HxHbtyo9WEU9J1iA`
zwa~6Ot=%J8VZ}qH(QiV*%O7L|Z+=i5kZB;37#975h!9v23$5QOEYIkWtel*rnp(4c
z=lWZTmD9)eE(fPoK)R%3cxvq=FjKmHyfulpb9OorW$w)P7MY!29vhr2G*}7)N3xk(
zw`On)U0N@NH1sU;k0!=sf@N5U*vpqJiTv0|mdcES{brJU$z{nL*meYOX%3AHSlQcI
zn3>`K`2+JX9+t_3XEXJ4mZxYhVOK(a14jB7YZ<9q^jiqs5q>1|J8mVwq;!jt*O?SK
z+ME<U@n=aB*>%`neymqkYPZ0mwP2Wu-Mv&V0|6k&Xtgf>_?H7sl_9PK7B|0}pC=ah
z;k1vInqWn>R=UmH2cZuc@?A~2`v>>f$oU$8Ysl=B&+N253@&nV+L4|4p}5UjGsolc
z#+lZsCPz91z$E~mS>Ca(G_0jMSZI<NTM*wP`4Oj$iS$%!tBedj3$n3vvK%<<V|s~p
z8cc-?9|ZZ-?$OA?(38M>uyaJHw2(~kPRM8w0Bv~M7OxiH;oP^i`vYl6u$sx=en=bc
z_s3m{RtEVL=RnCu(&=?>|3jc~yD#3v3q5D5;v7m_XcD6G;r`vJiSItnle7Bwb4knU
zbxd*k*qj@GPa+?C!*rx*y7x=hv-RTZqFtM{{|M8TZDaTt9y*xRRH9cVov}ivoqjJG
z$R<RcpGzrCoo;cxe``)9jKfIO(cb_wT6&(^I(Y7fxK?;LdxfoaR(#)cTTHt_c-_rU
zM?par2~0w4_YrTLA`}EHZl9xijPO5tlyRf3j4c)UB=Ydv?^4uHi{Cj{WAUWztO<yT
z@d*jpxY-ee6mZI*1#++o;}nrOJd8uq^o%XGLEsk^AhZ$$p_-o*8a=j^HaE$g;KZ?Z
zzb8OmNlB?k#~>n#k`+3X0u&ZW`4cbDyne78Q9iRfYZ!8%{H(^TRc1Rkxy?L!(8)*?
zC+OIV2$Gv2tXynikzxOm<ZojI^kEC8upWHYF-Q?IdKlHARq?SAjNbkIrQFZ$Hlu4<
zpG!(h0#DeBiteiIUxXC(=L_tOX>=qE56>`&Qj^z0)u<`+*pmt(Y7ysx5^+3e4E_E5
zV2pNMp?}=?q(@9ljQhG`SmgS_%xE9uRQI^#3pZBejYcdAC`B+3<>p48F$gxp^W*HF
zCOS(>eI97}rJ4!d>priD%Fj6d=;HKK&#4Ir&UBqYUtA<aL|ib?kdPOi6=x3>aB&fN
zB^5w#sUP7|1Y;n&;&*3?7I<U=>+@%BLho@yM8d^T;o&WEaqLzd)*Slw7k3Xszqkqs
z3sK!ipr+ECsus|$YFXHY|Kbw-=`gp_l~7<ezydaD`4`heT%4k%?h+4YQFOXHjU5Hv
zDe1D_9+XKP0FR{g2c%ZKWc_D>6uVE@35~LFh=_@??qEMi%J_Vteb2?=qU7Ugihp~K
zn)?+U*Bno!#|LYA<PH}Z^^+$7^*Z8#C)H9Zkp{Z@RF9wDLa<XeP#!Wo36?dyx8sO}
zgoY-tnsdS_iuteF#Xzjs3I<wnm}7+*sOf1RvXXF#4qif`h|~>_e}47op5IOW2R|u3
z@P6B{WqMAY<}!ov3FTwX+t`0oBVebAd+UoF+ivaVBv6F%vgm4!FF!6~4)F7m5h32B
ztNB{x%CfSmAX!phfpInp5?^m=ckKLcP9|CL<}cWBdPizz$Ok2#$dYilwTl@WKYjL0
zeUX8bIPvCxwO8KOySAg7{|yZJG!ADyjd-}Y2$j7xDR7)F5fP`C=|3x@R;09VQ1q<q
z53m;^h}=V@E_W>)qU!w%@otv{Ry;xuoF>okT`_OF?f#Q_ZCvhs_VDCv^aMk#K0jc$
zITg?vq;9hFrjs`HkfMT>@q?Iet@wzDvY6hup2X_MFSdNBHE8tjeH0p3x?_P=!^FVg
z<+?kM7ltI2mp2bht&i;!lQCmde5CYa`<NyO6Xfq9JJv1)gcqvXau~?`d^k!{MaA&Y
zR7c}EA)k}vM{$-LV3eTO1LhUC5ZKT-`KDUhHD9Xxi6%pTaYN+&`+}d9BT_S^;D+98
zV7Rh4GCV}1^~Cz==GDPQeBiq3FYU_0>gt7v8GiZ*bghMz>-)l?`QeC$3A}?n8NXXW
zEgCNM;!LY|F+(JC>@hJ7G3F`DLn<ow<)-JQD=d@(?&HU=WA9OU;FDpGw6uS2T_`RN
zHOuWpqx=D3s>ft}eZ%6z14H877jB;QIydTwk2B@7k%zvY{X{vEjs-Z^iB8gz60>Y>
zqWi>GI+C+SQ#teHnsd|KNUW#0l0-z(`+9G9j-Ikbic#BZUvovW#PlrfLi)zN9V*`0
z7dypq#~=haoBBw4I|3bP$~$k?VT~`Hd@Rl~rOo<voh^#rS|_R~z{T=ST4``1eE-PZ
z2u0=p#ne?R*I<7e4H2s_&8o!EHlhhH-Qs(rtjTATG+u}DamwtwGv>@Cb6VOXmLEcA
zC)pBkuEkY@IymGY%MWUaWAi(~2IYxELX-$hSC84jc;yCiXZb<mGlhwh4>Zz}ulXGh
z?!?#k4h+%SW3~A=eo(ljQvmj60eT|ZBmapv&lbx&G?~@3<)V5P?1ZTT0v~Od8uu$I
zBA=Rdc*(HG2@<2JF+*~H{%STTa3Q`r4H$9Q5+<SwJvg)-Z`<@>%wloCrwzD;Ji=3W
zC_XcX`bIq((tbLBEH^D#quDv1G?oAMmD{r}9DhsdcIS=X+11J=ZbgMtX^oti+I!ZR
zRpzE<vA%Vs6&Vyt1U9Ua4L<od%Iub(#*;ihPOWrG)1o4eY6<oq1adS{EOeHh)j+j4
z7Ve{&;FOVOkd|g3V1G!Gl4)~#;kV`dljXDNB<;5cQwxVUS_RN><>jIL>Kht96C7N3
z5IpT8q;(~VBhzBwVtVoA5p@|+aBJgxEASATQ;=$TZv~v#zjX#hp*BXYAJ|c@UCzY<
zj~JLXJG}aF=60*sN&k7heLHbP!Dn^UZzr_V8oX8|Ba|3iuOmds<bt~p@Pgs%XLP;#
z7s)o0seP+Pn0SPQFF)cco4Yv)zKFPTT4tw@z(($?j{#8fx$tE$XWAOkw~DDo(Nt$?
zSs5DK*Pl~E+%t)3gD7nve<Qj~j5#Xcf-E#KNhFhq5Z`AD6O}~EjET?7Q~cEplxss_
zdhK$nk=7CIw~~@-5<fFaw&cjqS>_D~`;fXXR@jS2K|gu6FAtS&;qFnQ_%xDFR=Uk6
z7BCa-QDOu%Gi$SZ_||OoDXsLWJ*N^q!(t@*nUf;mzF~yz!(sU;_Mu~PWt5zW$%KD8
z*2u4Ss!AF>7Jp-Fe#l5l78Yced-b21-F=q%aB)e`Wc?pnT-zcxB9h2iIbBklwBSB_
z@x9p~QP9a~aOXYM`%nql1c`O<0{8r6SoDnGw#qx%JQr5dr%zv#5p^u>LP*MMS?r3M
z>W+@~Bjze!-?seP+zb7oey}fWT2ziQOmLtSu(Bp&V0m&(js|~XA>;(~XMg7`dl||!
z^FJi*8M3Q`b-GU@bit^fboK<G3TX4xsw}Z+XQ8_taI(0t?s0j9TH5MRoBvNC_pY@|
zCN&6E9&<2AynEN()VAX$Z=_(AonF2MX&Wdy<CEz((C!ljc$!njOpLcpJ?7-$s&P6s
zH8Ifu%&%j_0+v?Isj8!m(a8Xkc%@_(w>c3d-;8D1>m+|!i(-q;_>cncq;xxr2h2SJ
zPnG`-%d9W=#Ds>&U3X-VCsGE{n)Y_-tb_(U@TwPZ)+t?Z^ocI}GKCR@WiOuC6{o=$
z|3Tj94roAk!@}^OHPOXzSYP{iTp&JbdW5=?DL}D3JY;j`%tuEjUHW<c1&-3^?(>NM
zX54=bif-K$J(fL|v3#=u*Pr0FUFRz9XEHPOXHM^N6VuYusuBd@jvRAX)l-(9<dOR-
zs;*E$>gvE2g6bGtBlqZDQ*%=Vfrxh2W!>!Wh7>$Vw$V`QL8<FRW4IYR)=i&rn{Q-0
zmSu38?z2g94a;$04GsbXAovDa3>XAMJq&8_1eaRzdU<CQr&B+F-=jCd7bZkqiGmhK
z2Q_M~Jb7Q&lpi0w0h%P3cvaf0w^00hoA$6y=Ss)fJJL@q<x3gwS4ly8%O2t|O1Jd%
zOSgM|P~W8Gq?j6-scS02Ok|6<8^tVT;Qsvl2f>z@?WCXYZA)V8t@#mH(u|)zT?{@Y
zFAZqcnGN8gW|`<8uW)J{S4g~H{I|Uw*xMn0<0+Di*J3sa{5=!$Qc^xw{V>&4+6O+J
zeCL`FH3NgLg&{s6epEsXIerw~<Hvrc%7DV`ItutSmb-3!>8I2+;Nj*%MdG^Nv&THb
z%Ra{(!z#8r6so!6CF^cHO!)Tap!sL1>x}!v<>^eD+{*eRovmDB2)6rY@$?Jksl!8B
zTIa)y*U6v7+Z=Rs^{*DE4`GwJWbzHU<}_+|yXZ}T?hAI&dY5OhUu?=Xv7^?HE*>M6
zyqbT)8s~nGnR3Sk(gv<RaqfBWF$o8`+7>Px+khek3W{E?TN2pvH5M~Ayg$5KGD3D<
zAtKWZ)vmuc@Vq3_`{W1j=p}w`os&<Jgl@}8U^)Q8Y;b1xo12>;N)u%kc3q*w#5nEf
zut?Y(y~ot+N9OiVpY&+GD>as|k@^XJ_oVs&gwFM99jBAP980LV3G0fFj$k1f9SYjR
z!~fhPk)36&hk@d1N-EFQg_9V}+5E-omaO{><@CHk3LtdzE10iTyYFLH{J8^svk7}j
z0sFHX=&Z{bEgeK%Ud&SXV)`nFg<KEREJpo1iE8|Q6+b2l=;~k*%Ql2*kNxV&SY2>3
zB9b|fY$CpRErj_pTZHH~B<Mil7oh}qhqnBen#mMJN!i!eb{*zXhCS)&>CjHZ%yXh3
zqo&-5W!YpmcjH1TM`~j4`te7-FO;~1_(Z!;A={FXo%C$S?ZL}Yx8pgQ@Md@k2`f%{
zn)NCcis0rCE{?ULCcK*F$2PZ#&=`zu(V@xH1U1QFKMo<<IG=SBE3)If78e<M*xpn7
zCzmW9#jE~dqAE{M6d7}rWNJLGi&N_~O?A@&3W)f9Y~O1#mj+6^FI-a~Sz58a`BP=F
zIsV`N5CpfnJh;~cL9bQrsBYZ$py-3Lyw>XUTg2BGogjZca>^3{q5PyQUT&&e6`l7S
z1t@ZOnWIK7p9HsZkfmTCFy=yg*$kCkbD$`7>hECy;GkvJWFFaXHW|F|6%%DCe!b0T
zPk}(y_;Q?V@KS+LLcE!?^bb(~eZ*b=*)u)(4;3|~36Ein49*v}brL%gRkocxo3VGQ
zK8T14C+GSQy-?QoIpe(R#ut`sj$T6T(^CNAn4`{5vRfnLF+xxY5MMt(6qs*4@E?7=
z`diRPb!+>DNvX-<ca*@C1eSM2Z%k>!o#xtqZMtAbeUH3lfAV5~dAB#)dt&0oC!Z*s
zkBO~v+LPtmn4|?x9-Kt$o@LP_2my$qc*%qkXLSEKUDE1IhjH<23v5X3<A*@JJ3OqA
znRSog2I7Tr7mCX^l7C&5IKO_GO#`CS+DGCSdfd8;?<uZO-|J5r7XRxpER}7Y%#8GN
zyN*5DRp^L~lCG5%n^5BM6TM5_LyvUv4Ww>8ohAIQUeTDE#q><gio7}4tKKOsLirR`
za|IfslDw<g-6C+;6#XZarm?%T_r7QhTqru5P56%}p6>=L{M2zMzRB&pY;T-PD@Tfh
zlx{y$%I182#tLyOw)W;d+R$!;v>jDKZhPG0o3y!FNYsm~UxJAV)uP+<@0p`Cw3Gs>
ze%?$SxW*UG4?;(Vb2Zsb&c0toiu#5mlB1!Wy6`aD$=zfSYfBV;`SP?oFks+HI!BpG
z)SXdKK;ou7CtZ*4#6EG5(#p{AP%tUmH;8)FC@Sz09rIZH-q;j_jc=GTwr#<G#4r7$
z+<w6xc~fYa=ZpE2Tga;a8=9K=<V*9T)j*HUoOI}hlIIk{4Vhvq9qn9PS|LqSM&=vh
ztkf@UO@@U?$o`09HwpFIl_%6s?=RegtBO(wLtc<RLW#RD=%m25%g)}9eEMLxF;<|i
zs@k#?lR41&wbpbJ8v(E3ttm2e@EePc1qa4%?F##=el8_L7VyT6=Qnvv8LkR(Z`*>l
zqC09TS^%ZN3@EU1?ubl4$Z0vHV`SUcHzbH+k}0}d#?Bg^8@S-Q@dogb)-L;r5tz32
zVU5NFj4U&&b9A=cMtUpDV8KsGX=n}cy}uvTrCnjrouY#5tXHVqxQ`b;0Bq_*+dnli
z9vD;%G}R^(w8~zx97HvDF%>BVFdyH%FW8VEMxCy&Clk#+GSp@Ax1k&u_=~552hu>c
z)*@$Oj;eIs%+NXyPL?Ae<9-09N`FRSTX<s5vCNcMZ8X#uzVf)KIf^CbfA>Qww-+OO
zw^qd*tJU&i%xb`_n#s&)n`PL-8*CpNSn8m53;w5(7V&l8z%L*mJ<Wi(>y6_@=gTx6
zvAd{TnHmn1GCzbUeHvq8NEEXF=>PQCbT6xmd(Eo!(#cFNzV;udZnF28&fUZ3utC*=
z-vLsT@eG2ohlhAf$4hv48QWX_9s9OD7|Ev)FoJpb;-6<y2HhGIWO(MRVrn7d7Xcx!
z9ud&$HbzcytN?4rs;aE)jq8W@Y#X(tk%#0as?2Dg;0o}{#``d$IZg`6@S|OlSKLLT
z6-TFUmjtW@F5_608|Cffv%#tL=AS&-YcW|AJ=!%Wh(e#$s-d$%iTBy;uCB!X>4nOA
ziF91i&Ge7r`n)1CcKi7zW2W8xLl2(xXdaw(@jrg~byI^|B%ic2v~g4Hr5H?jXr*J$
zdz;PHQk)RWi_T^ky5Kf$9@rL7v~3nG=fA147R>oT5sZa2n*Pd=i&viM<+Zumd&)ul
z_)$8^oj#2kz31CYI;>*B;DKAVZ@GN2R78|p=6G0=o7>soR8*t>NxP^zTEOaJHS`FK
z+&Gq5;)EQsivlXs(!SIQeV9d5Or$vIBe)%Q|FMM|JnPc*9k!To!;E;+ACWuuyX*aY
zCsQT1y-IIXelf5G1hqC{f^2LPDIHfi*mGTdHtPwEgvYV$KO&`=iJ=@1fT(*#ijzVt
z2Zdl_f7fR<buQ20aC#NDQBN;TX@KL2bj&*3C4NOw6&C68bTDB3{A$i^aj~2MLLHwu
zZb+M$#LeBie8!RSfi=1r%K)|RmrM7+GBXzsm$xvfw2W2kr^Y#eYs41}9bPsnKapT0
zTHou$HxvCQ>h!E!Yl7R_sY+$NR63!&t+7KwR@J^IXl9m0Z1glXyD#jk%{L!X(Dup1
zRoagKNN3TmRZvh6Kavbs?qO#q3yV`@wL1#?+A#vnwi=Vh7i!918bvTU5v1ochaZQd
zxO;Lh_gA`Sm`2aL@JDy|{~s5itQ}6%EhRp33x&VhHSQRfe@cshDlPgiXKOngBs|K>
zA;-IiiQI;Nq5k}^fCr5-#It}1bv!}}Wws^W2Nt(mornvMQolUqH}NZo?}R1J7?5dR
z#-qS%9|^h1nxYeTRaV`WZ@2E=-Lzn<xgFIZXn*zrs0<!h)M9ON{u$mda1*?X`4#gN
z`^2?oC13>3=W*#n4;omrw%<7&loDLuvp~JkqgCM<$@5FQ+Oht}RpjbVfYpD$$qk(n
zMHd%u4f7QvaOGrs#>LO0t?<qr9$>!s-8tPXOh3-QR;JDd_6vqo^l6VC_Y!YFVFne9
zqk#KAz1I2p{t9n#j)i1;`gTxCTI(+01^7$d68l{JE8^iJLvdAEa0i;Z=XM|z8=Ig$
z(`zI!sPie7KCuVq_O1M1zXk?|N=l2ebMgWM<>VEW$prhsy;7<@B}EsCo=iR&m^mlQ
zX9Gf*B=QJ6Y37wo?PPwHMUTQAUELFiMiBKqJ;SiX!PeK`|7~mx1Q)su%+}^s-KZvW
zFpLf)y!~o&e4L2rG`lzuLQ`j1KRO0Kd>WxL1X5;3A}K}DhRIG6854Czdm~3Ztrw~B
zflCP25SY|Hef?U`uWn@@{84V-4j3Z?^W8;Emx-;Oq!+C{fF3(<ogsv}l-a+22K)La
z_XBytYjZPK!BGLBd&(3LBm;7KrFgRDCNW{xJNm37*fc&JfB%O4-F!~<9ymD0^N!!6
z)ZBI+5<G~m*J)~b5}~I0p6v}JSHjBByRrrDoLf%CP#C#iL@Us}D~`>%9a|q6sj^5+
zlzcE`RQfsP1AT-6uh=#~7?61sCv4aP-I;SO&;ud?i3P%BPPpz%7-yNqhbE)6U+g`l
zt;s6W`xN`{q##v8V*+?dkD8iW`MzNWe4`Ms{FLnJu45S;mh!ng+saYzk1yum?iT-^
z6(8^a%wkQ9KYk-KSJR%kM~MG~<q6O5`$vdn`DE@70Raj_UNMiJ`UQZ=o?vgkEGQT=
zwKsKDRCt}8<?9<5@zuU`wOQM%aSp&GxC}rJ)HDC6sjL|r7MYyQHagyap(HKvRkFCa
zL`?Q$XyONB+luOVF6)ceGCk5#J)}t)eC``r!O6I;LNiuBUN{g+<NSL-3hM|{mo!0T
z?7bXJ{s&U`jn{^`_k4nrO^IA`8+|=d2Lby7VE2D1NRRQu2ZV&<0@KTX{9uggxO+D^
zCnpb96Gb&jvczac8m7mXs~SKoBqWzIWF5q~OU5=CG)RcbODm)xEo}=;e^zbNlXOh<
z@cu&o)lo+>3gIE+&t=hXW?NbLyo8+7s;hpGh5x;Sc?Xz1sp6yfeqVBbQ~~tb^5Sl9
z@g9~x_&Szp<|TV3M0{1(W>1kf{xe$El=F|23_BAriiT;`USek|Cc&b-WQE>}ipG`w
zBI~^{rGWyH+?oX0;BN+y3c<{lT~QG!@-Qohj)tQVrj?OBVXyzTw<YN+po$O<7L8e!
z-|~A3lN2CLG)xlMwZH@d5Em`8U!JH#4b(!1M$=htElrO>n*e?6=vxg;^u?2=#bNHk
z^zx^LCBA-AM%ziFZ`CowB1=k2nwy>g8w4s}?7o4aZ3Apg#kZ>R%G9gfGmlt#rIVx@
zKK%B=!08<t7Cmg9v!<a=W%!xX)g|q{y7uLzc!0leQYfW@f_nV-{%7uQqvPP1vA825
z)W8-DdOpBOAkNS;hy1Ps47GcaVz}4k36-)TjDi=+Nowg-Yy3Zd-}y?t`bbl%p&orp
zF%1idU|>5m_RQ%fb^TylZ<&TM+<8rlgOgRFH$ruEji;xG!y1=bnx7C75W+^UN@Va^
z1~394xw#Ry5HTDXyii_)1K+K4R`uMh)iVEfdIg2P(4&&6A`An)rAL`*KXd4w1ySOY
zQoyHa4JHrVF3yCD^a+R3*tW%z{{gvK*Pm~>pBp|vcZmMt;7oDkvZ1RoIAo3j(G;*D
zxXW0*U$kxn@ZyJUA<H}Z`*_r1uN)w(s+&Uipn23$^ouJcs5zngTuas6sl`qX688<M
zC8P-Eu{*+p<Ys7+4@wz@zrk|KFtV+<_D4zLC?Fu<*Gv^he$wh-0jkI2xS`<3&$y<h
z24FignA!RTKuURl8jQZw5}4gMYvB!j0V$@ONlra|t&8s9^XkzLL8sMH@^fI-pKB07
zAx%xz`iyL}C{W~N;1<Cb%(bsjtH4^c#qU}eT=3`%CN8y@h_O0ad9~ZvywfF9Td<hI
za|)kUDAX7=3&1}N{4O9i;2X)&X4Hnq9XGA`iZgGxfq-{LAK0S98pV>pqTKu6tF+aA
z_wa-oAL>;YXx2EL*L3@uh<>?PQ7>_vKI3Be@~q_R1^Jukn8`{j<CPj3;7eUH1>A#?
zn+(wCCD3X&n@g8wFhQ4KXX|0=`oY|tYNy^P%QL<F@Aj>49yh>9SBrA`4(uA)zaUBl
zQi|HT{`>Us77tGSC|Ij$Xqepg+sm2wkFl)x04z+@Z;jMm>HIN1qlkIxAugqKS^uj8
z(zThGpRQY|!zbU%Pj->byM6*ies#Qz_MO3VtQ(%J`u$)2>rejo!pM~jB_A+5EZ#l)
z3F%xjpD}U!EkM7%;B$mY$SHb-0+D`Onw2`n%$>xv|My`2cPfONYFmQPso;PX`eQ(~
z3!Fx{(r}~$A+ORnT5&L~@!{<qXaMy;o!l)V&Ii@#A^d>*?eVt7%&lo)uDMdkiHd$0
zD*XCc0C$s|JUF2YLQZAY)s9P}ujyjcpHaS`Ik2<K;Jp-6y%U`1{Ksx7K3+(_U+&6n
zed`4=Uwn;o^L!6S^F^zlum5XV8HtV;cYTrw7N-2<r2T>dc{tca$KtQB%F3Kd{``T-
zQl-m{_(Tl$^2=-=2YXqH;8uKMVh}EP#`r0)oU19T*1Ar`{8oomv2i@&v&`dkeSm0E
z^^Koq0}14Sw<M#a8=7+`r*guX_J5Zi(y$Adx~8O;8y{;*U6W^Vj$caT@se%D886oa
z!}FW9%KIlbJ6Xu&Z1JjQfl@@z)>P3t_E+)JPU6$tH#k@Ne56Q7WMqZaX=lll&z^7x
z+f%W&Yc{<3`KWFRjx7y5fO_2tAmw%EDqd3Zc8|cj|4#2&Vl{>wm<AeW;js6Pq;inh
zqAUIOtJiTqA)%*YzS17(`~F?!<m4>ahU8Rf&1i2A3x;@*{6TUjBO+ZLhrFvbWT&iY
z5ymt}rP1+KZp5$5Ewou3476=IdA0<Ul~iJ&ne&jkl#^FdOH7EF8Wsot@MI2>#C0Rd
z_IB9mABVoWCMWO8j6Q+E-{0_-V=amOi{|y=;S!h&U82r!3R}|)a%jND-$><rhc>)k
z=b%P7;y*7hgMv}Q2o<Q-rtCO3$-f_T0=l5Cu9BLQa#0?X@hk8G3bB^i=N_d&8A(|u
zbKtAuU=|S`eyOl>2Rjg$8-EH}*F!V!BT)VQtW?w04G$_&bx0S15>d}~qrl6hV*5vU
z)b_$U<i~n^52A%CPk_CmRkC$gS31z=YE<!(>UT6FrM3s)@x*qVu0q${UQaklWmH}|
zt)<L()j;w}x}kxCtvqbNK4WjMJywkh3$5K;FQF0q$`?!d4ay?`ol7@`9(5i(S`>!<
zbYloB69)HZ8uzHKUqZPH<E{t-!u-jyj+Om$4Kk`r^q9$nh;X0AIf}^|5ETvU{tQbI
z?bS%8-Zq>A0s*o!(R~R7k3<W%rQ^JcjFMXY3yQWydke~1<8bnGjSvstm-_}xu{<3U
zt=ngt7}XoSGas%}1uAR~<u-d3b_J@n@5(D(V(6vHXRXDIk=$9H+hhQ=)$Li6j&H(x
zqXsY^up%$}b;Ch64gZ8$27~+9MYHbRy9d`X3Ou}=?6l0Ge?MG?6!opz!eDMnP#eG-
zJimkTpye0xs}RqO=%iI^Ii%OEm!dQ`>uU=Des!+EqoJ%x&#H}=l=*Yt=AV%}LAw4D
zkQ!i4h05yI{s?x#Kkn^mw}8cfi5<x6{7=-Q<HT!j*W*pxDgTKYxF(7yiq4-k4@?}B
zb^q<y{%@f=b-ACEVZAwT%%u`>MI_ed5j)0_K}Xj{F6iWW=r!H{<)V_;ZSJEXpS{p^
z|ImcdlDD=hh`YgsMA+${=P;lPK<)b+3FI~ycXoD3XUN6)4vV$Y0>K-)NT?ps5wIgU
zt;Na7&JGU?ixsk)f8|nX;T^YRS?>8fney|CHCyj5WuwYZS_`mg6o{yV68}RS3Gk)|
z2O(8Q*l?H=iv!=5Ah9R-g5X?;F0H&=<>gD`v+sVxk;_K-ab50|1ds<0<V9uW@0onh
z#^1!CikN9#!WaN`P}#62+IoL072LXYX}k_iBEw%jLX3<|E4DYl9R<*3W~RqOQ8yUr
zneC2Uv$K=Ub<M8oPVdLdCi0D0?KSPLMrH#$#KE)A>J}rdl(wA2Y-{NEu%j)o^<cn)
z4ZIRZ3rtv`?p^Lm@a%p~TPo~+R`t8==83D6%>al&sCS*II2c-6dxU8id|o{y6L!P>
z5Ah&ki3X~?^?3rv|4f#!+TRRL_FJ##aU^`2saz<Xwjqg7Qx<}k%By)Ut*qC%8ahDB
zpAu)i@2ak;Sz)?+0q#wuD4r0R$`IMf!@(uawa5LCj*eGaRMd6qpT65S_KWr&VMg}Y
zo|Pw=88ma5yfVKFS#4adQZh>1dxC^Jb*tSjfIbCO3&?bWns(X^GcVsF<okCLj*m@g
zehWLin%g7uyN>nAq1!^PY5hhYQc`4t7RXxpXDxZLqg>7=2@&2h3k(XMpE-F}_Eu`j
zAv0y}T*0mUPDkwMUSn17yW%nApi+i<MU_#Vj-LLr`7=&|f4W9uQc^#`tf6Xb08L3~
zrW6Hu`y3h1KnFYRvN+!NoQ;F)EJJX-@|Y5uqu#>eh*KaSd2G{?b^9dshV0PMN95!Q
z8BOP|IaiDu;fMb{@VI89=ZlmzG%_}3Ob~YItwTXVV&c!+M`oHFv%bds-8}CG;uh|a
zeFss`3zgTI9Kgc@1~1!vX+f=cfQ{~w^=9vZk*4FqH5J5obj^|6$5!Wc$Vf>M-3|k)
zr|TWR-^B$4#P}{(^IRyw+7(85Nm7s>SM81;5)%RbOr0`;hO&vdgLVxZ#Q6p4rhlpo
zmJ~hjWFc_t_4ki2FD>{qz9;UA`+7q_!rlKa<{enp=PY@%x}aPJbkGO8^KX~KvFljx
zsPx0xuC8zJqfAs()0rhFkGtPhf%8W1Ub;!V*J1@%hk~|jN88@cp0No{$KQ6iNB<8|
zUm2BE`n`P+K|w+#q+38zQt1xqP>@Etk?v4rR2n6uOHx9*OQfYiKuWs1q~FbNt^ayw
z`DvCj%JZCa-+N!V%ifmK+ts=6aY4w+%*1qC;a!#P7JyF>qg-f9x-I0`)7wePZP5hB
zBLTqzihs+ZXX8QWNBOAFpQpaN2EHF<W{R%yYB%Fh9VpVv2~}wDT^ADvC5dk1q3HAH
zU(cwnyf#b6|7gdTW+_7{<(l3eJax-#Q%y$2K(2v%E@fLqN7HYyO%GOrA9FMF+6xOy
zUX6S5kOgQ|oI_F0Jrc{B%s%!;mI~(K+8Qs1Dj-M_=|i0UF}L19cr})$uw5l3rTWFe
zp}aleUw1AJGt%!@Kk*r4?nmB60GJ1QaAf2b>ld#5-hg(rz*e}He?mgOV+n3z3TXd}
z3~FgWw!%I4IqK_Ah)(!##l=itzVDYbwuFYX*`KJ7H@dG0-pa~K`-!^C=$YE8cXTjF
zz)A28m*OTuhO->k9{e+k8=`pv{hkj%(=X_-amCUU6?+B;)#lqm9LbjZV<o*_-5Wq2
zTmBgHZ~6<rq2VT#%)EldPX^4u^z?l0R?D}*!XUt%P5Le5!SZF_8<OBgp`kH?#WOqu
zAM}S0w~_u;tZhwSaFy8*>gX%E&ep}Pq#+0hKt!lkU{eifR-qKMXyay}ZK0mg^jR6G
zHd?)bec1dq8pbhD3$f}yzI8!&_qGGZv#n%dZ(+aQd`locWY)X8nDtNUyV+UVW?%Eh
zm~mw(#ze2)Vwm*%2IDe}wMBY@<;L5yusdGHNWq41eQYr=D?^0IVI`ILyQZAk7CU{u
z+T|t(M=vkGZkmOZtm^7$IWVb&_ksLUjG2!N2gtPzXdxgBj*kA!K+VSH?xU&k{5B?r
za(74k8aGNkt*845pkL#EUfyUrYl=``E`seWoWe5YQ%v#IPVUsilb*yQY=O1+;VIQZ
zey^TN&WIX3p=DlP$-VP}r<^&K?Ixo6r0>1*WEuB}Z2>E-Nb6;V@l|GNJB$M;O-|dI
zmD?gQ(b2`+PvGH&h-!YH4gZ#$S#M#=Kpm2&qP8Cz>=bOo0iT_{5MWRwlFuA`EB<5T
z`3vzMv=1NB$w8RJ-ky|{l&@AfVEn-T<>c3?Y05CO3FlTcq_LxoiKX2Pe8<e#an^`C
z_<K-R?pb$77qlyUui!Q>9)q(89w-V+`|QlL-L@;Amy!oVGL<H#rci<ce+*1{%z1g>
z&T-nq1g95zPVS$Haaaircx{UcbDfARXXNbs&WAJ7^PoyXs6M#aq}a-84e{@uPxE}P
zp>Y%O;2Zx`vX?$LI6%88g9Gxt{+*oEh#gS%9u%hL=H%-)xN?QJC-8b0Ia-~bbU%U1
z-ceFczAq?L8AvGRZ&sk>H*>WUdo|!RdP`01lNj~vGW!hG{o5@Y`EHa?dks0$aLe3O
z*49Z;AcvD?b`_X8kW!wVG+u3vmZ$6OGa~Mk_&;V}!={pQb#_P7?wOqI*tj+bl9KA3
zg&2xb--|ChB@6lOHl5@q7gj{hMu6WDzT>DEO&i_HpuTyuz;~!kZ-ge0_q3Ijw*uc;
zTVYMZZ9o`|jS;-C=N{Pl5J;!|yuzN|p5@(t7MJHcyIh)X>>$%%mM?`G?NFU#-A3m>
zT|cl8#^>59vQGm#Lqe4oURT|`@kWxyx0IXlxs|$;yse&(tsca}OS+NwWLj6wpx?wM
zOU=suc0tW}U@ZIZA|dwh&>0TEHyPRU_dmgN1jE(^>f0YbM5+bra?|sN$HwsS@9XQS
zpDpCR&kJ(Y7gGfXL`liq&t2oE|NGBCwBTG`{yZvLztQ8EgYJBMUO+AglbfF$jS@7x
zYET~@7;Yf>Jfu4yq28Kb<y&uW55s+oAWTe*XO957XP-=pj>fmPTK{c3J^ts9nYnSN
znG=WrZv?bZOgFsMwTVyA0rioF5=C%p{1Y#J=T!^GJxq9pAZMbhi#hWvU9fytiPN4C
z9O8`3U@iSq;|Y}gcutQ>L?JPiz!UqatR#>K(Qa&NYwPGGdqCp3NR<omG3&+DDD<63
z@snJd2bd`6{@6lohv6Ayb$DbX!~}S5Ao`XM-p1|hGb)ac*MW~y>{W4An;~q}!L8{>
zAR|_=+XhqHuil<)HH&2>awro~PC4r>gliuTX>cAD<QB}oehS{hjYV-mwjxL{%?P-G
zuoe{1kv!|aw)B8%98dv(<(Z)=8vFM74#xrHY5}g{$(I%%)V0^TpRJsPf0qb0j{mwB
zlLfuvWN8X@D|Y@Cb4_hmR#YUn&%Z)c#=E+&W?P>!Ke=jYSJBdngSX;WKA9~mHfUY9
z*&r-$es^$C=x`e@f`p1qG_0q5Vc!bAHF{szX((MrUI;*ufE*JeovxZ1jSwv}^`mH6
zI!8V-?h>#tEG-d~MRo$uBS^u{EKpgN|Kp-l@BT-8yx6L<CJqOO;%9hm{{hg{`ctpg
z9dnf#tiB-Hndn$(%OV7$Y4m4Jb)9AJV<;|ha&_=qn`LEv`0yb$R2{Bv-S=#8_gPOa
z4JnHXjh3?VG*i$ua(Jd?Ko-$LBXaNdKNtkjKrkmIc`PLZ6iW|WB7o!lhwce-0P0!Q
zTgP@l9SJ7&!ilP>p@A756nXanGjmZ#XZM49)YKvk0qwm%8R&6K%hd){uJG|HEzcT)
z&BNfBdHpmI!m}P`eg)17hl2YtBc0%m@yg;p!_7YvB52|5fPz<+7jG11JX7k1|M`P|
zSfB-~+r6TC^hirv1Io0%?f>gdeS->^IIqpPR5q71R;lg7&y>OrcQC?h+~(CNd**zp
zu8hdBkq^PadeyHTm#1lgK>r`KB0_;j+j7S|OioVr;QA(UHZpm-a9JP^PfKoL?llN<
zo}tU<mtLJ|dOB&EPDAsgC~>plnY{Y^QgkU8^hzHTZZu2}|CCjHk(pJ&Ccw})*73a|
zjf0g|URT%VSI#V`c#=z*kwgKHmohq^<uX`x>UXzLV=X^390`jBDJ!7+LrV@0$x}ec
zyc*mK(l~A-InU#R6Br^zL1Q0PmpGYzQo6*d<+a>A|EfYmfp9%CSOA%<^VU!|ehn`P
z3`!R>e$GTZb!GW9fIuodKLGg}Y;jO%4YalA#+tChIgud6Ed<CU*@XR5#U?m*t}mw&
zD=klH!gx&KIv@Cyx5vDya&u1-&_1w8!5a?#i~$wO^PNsEP-Q7k)dMGEJdXzvjf$b~
zlZ8BP2Q(8AuTcv440N}<0p1}SvN7p>%y5j3qax&UX(ej-dLjFzmq0>W8@;vs{;P_A
zwh5ykeC^ZyPQxO&w$ngf|Lu7`%P%;j9<b;zL(*G41t4}{ykOU@KA~p3xoRQ&pCr<B
z@aVH=V-dh<op|k*mU(>%LbNb{?|gKkl=u-DhK`|})-RYMwBY#E=|PdDrEO1FXL)6*
z$L28(!fX4I%k6jB@WHR1iQ~h}dxg3+D)7<377Rqm?dNZD&~AvnaDh^sj&hx2c&R~H
zU071$`SZ07C>)wj9$kN79lwUF%FAur>P&qvw}J5*tF4ytHo`<UCaTeLc?Ijd;io;*
zI0P{)@)%$mSOKy!&F%g)^tB&v20gq3xHb^;@W_m^C4GvFNd^j{?xFRTu(Lyh7dG-d
z3Lk}PPkjT{NXSKg_UxHs)ueq)+uLfc2fva|hNsRNO0AC%VFR~EeG5z|#Z*1hBWjSL
z90U3!YWM-rZ*6G}UGb8ziX|8^fTIeHZY{i!nM!V)EVre#;^p%|E5x<8w6@f(b-4mk
z7N}Uf_V8!&eNc6$`a;#$`Iw0!ywPL%m5S_70CgVTS^h834^(s|m8HTCCX@&Oj=<Hm
zk6v9220D4I-yy<6AffHSRxdnSXE;eAK_9)gF03Y&4uOgX1qQ5kTNC93i0h^ZA!!QH
zUHrF7?kAHN2oSUq9#W>qc<ExGrD*Qnyhrg~r@~Af{+*iRpW)81&t&bO*xDNNda0@N
zt-UEa;tM9SzuYUyaCH~|hasmHOGVC#+nNPilF)BydgZwCFS1kzRT`w#K6#(#AL}rK
zsu?`7;NzhnBiz+=<_2pF{DQZe)OTVe!^YtyOKdwAC-=)z$7g05z5lV+W*E(ZnkJ?1
z*oPECUvM|bc)VZ2%oX{083hSm@97X%QaXZHTYZ_u($-eM*WLb@NOV?}0~%s#{%GUc
z833T$%eDn@B6v9x`D+?u@wBzS968Q|e*^HJl&5u-@1Lc{<?ccdw;8+cBWmh_44Zoo
zADV8lxh*a(RBZ8oiTVVe&;ejbSoC<7$2S*6Y3}Py)mu297TPA5DW?2d-iIo!d;d6`
zSAXzn$CH$B4ocD#X9RpQQ1J$%3gGNFj+uvr9iAq>@AknA<ToG41AEroCr$~}3tmo-
zyw4d)1tVYJW68{<sP%+|7Mg;a{BrP9b)9?$@z#E9Au9_L1W^I&>W6ptHo_3TQCjvG
z5TvcWLxC0X{~a)-h;F^RoU%-9MWOH3+z5zy*bO)vk8>rpJfvXB0g)0g+o-p49@1y4
zF%D1j6f6?FKzY+lLqoUhHA?r8E`o4yymhM6fsEv2eVtT^Y+AtXSI(2|8z2g}ck;*%
z4vcGm0f&~F+8h{p=x7`|fA9e*K}Sc2Q1S85K6<FA^=#l|+C>m&85EtcnZY46lbtaz
z`23cCULFZ#5nkryha9vWegk&p8QrHQ_O$%;!TC<1M~^zObzIF&m+9#;;ZXfM?qp9A
z+F-LQNVmzJ;JUCx@J>ehYJYd>ath$nyASYcy)SQ`A|o>;fHl(+n_2s;tq7YCGfH;!
zwZM(CHB(o&-o)`r@-cNiEXnlBWLF)nkY(F7*72B*<vs$u0f5yddzfEGMV-(QpknZu
z%B#v0lx0W}NXFX(;|RTvo~_tp*#K%!h!6Yr+qU@e6AlheXGgMNvp>XId^Wr?3JM3E
zL(3M*f)tu++Di15u~E@naSCW2eBjVP3_+#%ccNNZP8%KhLPVObY~uP`103Ayqr;PL
zv-06^Ss4VsMq0NOEh`r(0&6_R%^S$e_yqX*k20SkqGI|uKEF!=Ht|@}Qq@93fYn9%
z#nMtWZ84)Uk=#`+#Q*1G9*Uqz{{KkUZrwy5TERkyU>O<NYXvb7Jn1$#w~#79ir}6t
zJsDI4F$iN?eA+vE<g=_?j6sf%87<e{+oP$fQF8j5dk|az8q)k6e22~R<}NOW5IZzf
zn}^_XesQ4tKP>=)#}Lvy|Jt2QNm-K_pbBa89e2S4%bvEb8;ED0=aBjVd*DxT9vyWZ
zO$}9I(v+M29AxRr=8wyGcwo2$(FZ(Z>UgwBM0E7P&`@U&GgK6>O_Bb9RZNI>eYC)f
z<yT1Pe<TVdBGQZN_pCS?zgav+3@z=@kt}*?=uB!hgqn50r5rqfo4s23Nc!iqfBW(S
zd{s-xGU0H>rwE^^KGeiVpGzowS?|9mN{lriQOfzXODy5jr=e>aORw7D0rEwG5P?-u
zJXT!MK$BFefs4cn%wD9|<-?uOJZxxZ7Q%E;8QOSWQOxBiXC%i2sTE~3qshO2^MWC%
zZzDLYtzQ)PP(c395G>+BT~;e=pFqGeknGAEQ@s0eMv$7R(7{-my6AHn`YGTC&ggpm
zRt>tX63!@pAp{4`a%uEaV=`$Rp8l(>+QnlIk2!Z!sn*20^RsZ>eas*PIh<=mC8Ovw
zEl#8lL-kM7E2IH1T`ukS6K?TQQFHT&IJz=+R|$g+H@#@nK+K-sZF{@w;c#1S$JC{M
z{|DcnIshD`T$s7@jgJES;qOB8*-4MQ;A<3jGvmP&DDi*ko!6dp){{O64aHK$m6KBf
z&IrvPVyv!RlH#XS0Olg{guNUTRtRzfmv&f~fiJAZs2Ene0CemQc1CgK&v2<jf&J2g
zt2GgK^S_U%{+(g-kTD$;RPy$%w~qvEQZGcZ0ZcT%>#nX&i$D<1z1wlI1|5cv7Z*(i
zjGqJazLWX}NP#%5>Lb_R^^+_?ks(L$79WD~Zl&s7A8nMM{@FFt{M2}8aS^mZAJ8#r
zB0t=SBXtoLTD>lG_-|!>Y4Jg?R$14N;jzhICT-6dxN%bP0DjGwD=96$=?^i?Fc6Su
z*}}#G>Q|Vd;Ol^+4ul7-gMrb}BjBz?JLI%HCo*Bm{8kVhog|+zpi+;DI7<CAeC+x=
z2Jj_Vs>I2Aj;Yw_T~{Y}r_R7JZ*om#y*nqmHUJp*RS4Hv-MKf2+tJlES><gu+K>T9
z1B|B;DI|a=hwDy)1lM+=j*W&19%Cg}Xp=#K2f05H<r;>CU;pZQD|eKOcTp(c4(#D)
zW85I`fj-yg(a{&@GcxV;fMZX(`PNuotN~QW5vOo{P`$-VH>6Y`yi<=!>m!F7LG<=m
z<Uk1vQJ}w5K=(1C{{Z#<#RYj}s}ZWd?5O#)2!X85V^qn{&+r32cN&5GvKjxjy}fw>
z6+1(OqU6_o?J+weI9zR}d?K0+|GWgKwCh;!zak74CMHhP!J9xx{Cj7zFv@FNGXDgX
zxDNYv-Rs4es;V#1fy4<tTMfDD?w>^e^{QjIN*o_%o3$DY4CGwH-##YKLx5ca$UyAX
zJ;cJ=vY!}>?&Du3iyMI5c{TQf;}Tp6J~IvBp~?Uxcpm(;g{lomj$EV1K#=fbG>`uB
z$3$2kWxo!rLV$s#M3rD+Y7oGaDE^&>n9khb0B$7i%w+Fx82Rf}#?39ZX0?-)68L&*
zt$zRAKla4CQx8Iwwtau-V~1U;D6LSId=ZW8CDJ<_(b95H{L56oQy%mCg}Q;=4hZA3
z!?uo(b^&tf?COQ4Icr->__9AXrc<4ffYM0h*f;ZQj-j2^d5?m!-#tN`_BE!^Eh*~9
zz2g6I(cQ<;nlwBcL}`5fA6DCgHs#-yOy3?T8B_(e8i8)00T=7hBj#KFHUIqw%9Z){
zFq*r!``2a9%Ku#14L^TDR&yf?@qhbP(D^X+H!o*!4>|(=GcN(_H}x3kFqu03`%aO)
zmgzHHeRUoB#)Gv)sV9uqmr5a1JjQ|m!jhBbYCkPh6P~KpK)k^R@@VAzTm<HZii$`S
zDwa9*ZT@%)W##4%%_b!gPspJ2gjOe<gwJiRfB!+z3;-qXWCGvGOf@Wf@2y<IhLUpV
zBM72tLAKaqa()t!(B^VfRTK)-FlorE*B2M_33F7Zmv|0$%vdiU&lV;;BK9;nFEo%S
zs1ccae3|kKX$5jTNeJScciETN?}81i70}j7yVFG(Au}L)1n!PK#*Cf^<EfO?_>W1t
zhu7l*D>H+#rjERU<}D;D8a|D_WlDjv<$Zir<h%PNSvwooEwqyvg1d7@M%}T$3>r*)
zjg8fz)ij#)7{>89V0TKIAn<|DZSQsXarMy=4np;LLW_krQZ`j3$A0G4vuCmjpse#&
z`17a3LK%Mv32z5|z-<lRu*7^Uq|9|w^onbyB~IbmVq{bd5`Dyf4x4tiM<E1fYifGm
zzeri$d!`{<&GOE2@44+fyNTl&=@qVDeS=xwm$}XP2gKy<W~H$IEC|1&+u8AKo!<pE
ziqmX(2L%5qMsu~Dm!`8|nfUUBF;v>#nwI$y(|PWVa~hZoIS19i^WlLeb!@LUU9;Yl
z;PwVd&CGpchU3g1&}s5XmJX!UlY^Lgs_veELS5spQtDyKvJH;=g{9xlV71)QtHH$#
zytZ8aYJQq1Qbo`41q%UZ+CaNE5OFUE9$t@00)A^9%eZb>gm<R0Ui;lMl|n--tn8Nd
zPk<MOt(OThyrevaKfKDxTwPy<&8ezgwLXU=I+|bkgD(*KU$e}uEb~CHl7{z3TgMJC
z?7pr?Um{%((VihH8XAz#c~&paTaNgb3_f96uieKEBNj3ON6T{(R%U_Yr70PhT6Qry
zMFh#&Kez7ruHHdL*^)IV1^dCgXn#?+Tb_p>QU(JDOr+7Vnjmp(RP#)lXE+IuJ}gKc
zxdVqb%vpb{Q4yHIvU?4Pd5p#$xP<vJgEmtbcRoEY{W)X{IgZodYd742wje|pm%jBX
z=-~s$Q8a(+2g<p)6bHXrYHQUyM%Pr8GzgIuEKoAvy?1X_*f)=n8AVj)R$$BGoC9!5
z;6p*z<2mGkK3tIaeY|-irEfh|1$^R!!J+c9;u^jN0(tF2rJ1lPy?&YhwvLPw?SdpI
z2vdjma&Mx@8klx)`B&9=2{}5DnptN~jwkZBki*1+>K_uq{bnbr3sVvM5KKfXWB($L
zTwyx|oRl0ts3J>xTjmk-qi8^Qt{DF-DH%NO+|m4Z6?ZczC`eb(2Vcqn58kA15L8y$
z>N9{E<>W-H%Ry|7jQUC|FYo%#ja(NFZ<i2>ggS{V1u|TuqteJqvyD=V*b+qZNY!zF
zV2{~@*eF)L`rT31TC|iiZXR?Uo%jrqh0Ys@0-^QDDXYAuJ>a9N&!59hO+X!a+g%Vn
z07Oz6FI4TUX}#P}KnV|8ArR+m3pju3TegN;gxowR+BEO02NwcZMhj}gX@px5|AJ7E
z2>^^2{!#wx5HSp)h@rJDH9ga*)Y<KKAEAc|co?`tq4OCe+i#0J)V<NG>!;d}ZEpnv
zg?JaC4*{$vLuL%Msl9(EanhSDc71M%&u=w;nO(m{^B4DBMtchhbtllc1Had=o{^b}
zjX+4f3hQHH1q8xq3i1jI)`JSDu6@VgtqaLd1j!hT`C^mr#87XIy^L#bW1!FciVaxS
z+Gnj>l*rl|0MjHXLVhTSiK9S~)7G)+EKD-HDiX-f&JkGDU%R?&3#w9Ls!D5PA^@<H
zK-Bd_$BU72J#1;v&nk(D`R2T7D9`k;@&5GXOHRorU`5}8B$Oh2d2Or5q!{?Vk>TIZ
z<ewdG8MZDuj@nOGzdm6%T_nKVb0l1%hlB0f<=Hyx@N^+;Vd<nK^*fPbah0ypam$pT
zep#hk-JaJiL|OtH02eIm0|NX7(u8<mPr<M4D0iAD^$mwyBsgfr+6)UUB_p))mHj!~
z2cKJw)Gd>`8Q%(IS_4B}QCtGSxUCOODJ5uDoZCR;WjE>ZA+cK;rxFf5IHo?U5B|?>
z<hvh}FSa-%FHic){?@fW*b;15<m9o1Qw_dnVjfZi2$e*D+qMW{Y!oc9Kt2<)oV6C&
zul<NB3jx^_P&5l@e7=dO$kPxYZG?R<dstavzl3ubqi}F+IAklb<rS)^gt0n^HzDFE
z^`0-brKKgpSd`DiTx4OtE&2(y2}0aKQjSkS%u3H>Wn$rfpg84%Qj1HL!0q!UyvWbM
zF8fK1r?{=29_YH5C>I{IE$<H~QcDrUc5jl22=L$R#tsd=E;Z@z!@&$3D$gwj%hO91
z-qVk2+q(^ek>Jo({gVVFgOqgpbj@*&dW~?1ehwBOWCDy)@qhR+12K^d=|8BVuV0!x
zv-Geg9S60wN<e-8<RAnd!!I>Gy*miNxWJ@!^tGOBx!}fnt=L}`79fVN?B-@2afv6N
z$;yJ!4KwgleDe78fU3S~=cdFMJ^YwCB%4o-2M^Hv$|+FVLP$^$<z!Fzipwf0Hpje7
zR@V&VyP5ib7scP)Ypkeol5kPH{vqZ5vG!m3*$Tqeuc@K7uzv#16wn02QG?+Mwt&>r
zzhchlEkkDn!L7dFGS?y1(t4(9pqw3>AReZqgx4<O=Hq_}h~1ctl8yKM&#mH$cUC=#
zLVE9kcnHLAum;z?rHT3v1I?$jg7d+H2lIYHxSk$Ti|GXGt{Xz)cJ`D{MSBJ(!7~K=
zDTrm?Hyaum;gvY;^D{<eZ|NhC&4yo#<-%*F9N%ELx)$n#F^Y!3Y^|fM4T?{&42?}_
z!CQxd>@9Afy^lOX{+{gd#7*j}JU{2)zJc6~sk@@PVsPcYeeYi6o$T-p(C#TG&%%i9
zy)k*Z4#47>4kmbCg8=Yr<&pt_<&i83^MXjnTT^SaZs0~PbgOM<*}ui0G_6Fou<O;T
zt7<7y{UvV~xP2s8W`AvX0)A3LT?-rvJ)|16xT!^AH22MwoNC+;kKUR#-o5S4n;=2;
z7k6ViY<jjg@oAOucBEIhF0j{V2sksq$1a)x1s30Z`qJ79DZ=%H5W}5Rl$FxPukU{D
zws?i5G+YDt&KU4}z1%L*k@{b!U~)(zYp=(}951tF|0qc!;yMg0-q}+fB0-D4wYbfr
zc0vdOF(Iq!3@H8;F>SpCjrz5Re`6H|`t6L#rTO{70`;?%hv=2DFoH3$!{HanEBji*
ztzXO2MKMKF3#U6v>bUlAvWc8CQ8d(^yKgRHB0v`pWh;5kU(LcHKY%m)rr|9vYN@$`
z6uat72w}(<0qfIfQaAf4rv{fkkE!^@E8}ez)by1a#Jd90homJXRc=cC++y31RxRfE
z{=K#ql|bsI7{26Pv6_U;TI1ZF?e@TnBc9g=J_N?e2DMJ+TZz@jzmlrWdxdU=w>-=S
z2S@!%X8`F`l^KiPLR8|XBmSocF0XSGuY6(u0#W*T*=!Ta?=`DDz~k_;de9L0;Aw_%
zcf>y?%Trvf%Wwh(?3usYPN&Xr(NAk~j;AVZUu$i<t`{yB)F$T);ciS#U#J-YJ#YDg
zC|rWUjQ!;s?od&xe1p<Y3iK#Vb0<#lP3-iq#~nLez36@b!g=2X840Sp(n?=Y5NPt7
z-mtXjZWfvtwRaG4nq7;#qp6zRm&6lV%d}Q%^z-LKo2lyXv*XR$50%?A$ltcbAvCmB
z_|j$x(Hea5{+(hYf>kbBiZ48l!xU?e$Y-|<&8!67j<^w(=MiobRYyQ6vXM@zsQNK?
z7lvILndjy**}q)(p8F-WT-e|1ih34KDny2O#u2d^2yO#h=k!1zBR#T(j0}BZFN3e%
zD{e!?5tEMwhOjZ5?>GE0jAiJvvu@R6kUwd~*yOc7HCY&Q{M+$pn&=_pYIxe4>5+Yg
z8UuTa)a;>3<?XS5KgaKAApQWXmC9727!KcVPp#ap*!-gmZvUgqS&2<nR{q9A*Zx%w
z2aIb=ECJ=YXqrk}IPvVoq~8Uh+x?GKoi2y1%uj&0<h7Y{dA$u~Fs0RT|6Q_at#F(E
zcsZTE7O#Eab{N!(bL3hl1+uS|ZTis{kQ}s;#iPLzXQ1+&EH)x4CUUjJ@zK>#<~SGq
zwutxMSni#6B|XG{sQpOiOF|qGaq6_2_HZ0=USss)<hTJxAl<|4%`vBlvnMQqoGgNB
zYHB~WJB4^&m~Oe6_}~%Zi;riah)PPj&sC`^C!aV-gf#ljjwt?qP4QF#1coL%#mkys
zMBL0=*$yR~$!eM(;a?Md+`|G7fBG~}o@P>{MRdZ}<-N9Vlrh2fl_<7w8h9K1M$=_b
z4K_H-Lr@nyDw}`Y{<S6D0u`hv${$q)3o@L_KA-M*><M5rzyzs?MBIHK@ZH?&c*bWy
zoB{s|?bTEt4L`3i9k;H&Qp^)Z7jZSr!7@<IzSPp*n5u5|*-iuIB2|s&1p)qAZ9=W;
z3#n=YgFEco%f|FQ^iT&$uP*u0y(cd-$6~(b1vn{RuGLX&+F03Cr8B=GzV0079kSS8
zQdoB4Nsr0m5AKPHj^rI4RaaFLH^6~_=voo0t&LSkD4<NAmsU)~K%m>8br5YP@y5AV
z)A+<hqsRWza>~E<0kQ1b&B<eG{;jibKbrno-ra$klTkOyA9qpbN55ud7<Y&H3>5DQ
zv386`M%?w=J@8eQ-~GIEh%4xRz<s(|XEO_7CXJ=y5Gw+P<Mw%xJu%S~Q4w)5{jdZf
zYJvB1)|U6~-{<ZhTAX^{W5z(23APW+J9iFCH-<uWc&<krkd8pN7lb+c{PLnMx?A`T
zMhY1AjpouX&JGtZ=#;X#AT$pqFi;2eD1>?h1Ff84JO?+&izouGIFgLf@vyM4vJe)4
ze(KD$C;mX_TZaj!rzUX00Jgyp4KL1z+jrFW?|?AdrHrTco&X2mh{vWG=z*A*!{5tD
z4iJZ+{En%s<|QE^HHKtdg)GQs;eP~TnUMz|s}KH|R^$g;hE(fAz=%33bH3-$30)I_
zWsT7f6MDizf*#(zO}J#X2qk(_xa_d+OO;6OEvxOrjv#CDsV+Z1xsZ2M0W7ZE=ChcV
zk2<q)?RQU@5<kW3KUd4kEi^O9nHioWONn54^yrnDRbs5cSpQf~R^D5qb(MWgm?gPK
zMj#jzNv$Z{l5kFZSLN>atwD=#xT1m=)kn|QOqbl*)5=O(!Weo1QnUUearDc{xgM5O
zG}R+xK8Ytiq-G0K^Fol%twF)cFv0_KyKO$6?WIl?+m*L=@N46Dj0W{9Y9}r{42Rt-
z14nIXRfmrdGh6Zrpc(#3+gsSAeI}e2^^_|6ce<FFuD^tUQmTiLl@<|kw{pw8xf~Bt
zB>`v)3i<&4qg_n@3O}YVZyM~pbiA5HoJgrduntu6P`-|o#SkiDyU?NvO+VA-3WX$$
znSp_{94dP|v!I%W;^L3-22?aG`PtvF5ujcnv#m~hs(*}#{4*HEu@79z%emb*<;q=L
zAORTkNfLVHXLa4Tql!&`ey6t9Wc8vXW@V)6pXs_k-gp-+P|A{6>!x5scs2289CK_u
ze8A_bFD#xvVtZ5oeGzIAG7d?1g5|8%*I;I@YA<=251&I`o)w6t^1pVt^kfWZs||61
z2<GNZ#v1-VYgSB?@$-0}HEZpXxSZC49`2tE#3(5mbADDI+I^~P&!s-S2|6taXI&qO
zfa0__HSu!bK7w$Upik*=cbkviVTPdQjxIVsHu4}C|G&PIv^q7v{|9RKpvMHZjqiM9
z5K6|~m9{(nYxqk)!N=T^JUMakL`gdSka#TQRZ+@5-gZWMIwvjjt?+hv*$<>EtFQOj
zU~7QPqLMExK0--b?V^%CJ`%pK24qN|Nv2zlroM+QzBhA8{zLF?migT4r^9tn{$X_g
zBOL=nMs`MM`@HI@2cH(Pa}g|-ni?<mh5i3bOaKWi;^XM^5gv`f634BavqyZ@9^-tF
zIUQ2BP&EOP$CoTI#gWZ2v<j|l(bCqy+Q=Bonz<^0cQLFY>q`!!T1d^QCmt=T3(t!}
zBX}T8P0T9GE7#Z8Qa`9NE(*Fuu&^*aQ|tz4U}LJAj`~qx%k06H>03J|`p?AV2_V*E
zV0~z6IZ%6MYD9)}E6{_NcX;{ek(D=RTz*<se3Jad0V6r9;)2rHpWhHoH##_&m!EIZ
z=HofI3abdzT}TC5TU+;abS?QHi)#D1Cus1r=<5DT4=1*F;^F0$YHu4_#bQ--DNG_S
zHjI<^@L0s_BlT`zO7&gNQ?c@-4`>#}sp}95{YICB`XT*B<&#pUTs}Szy8E4j=}$Zj
zeHhHybfuJH5r4-}zrzTTehBw><~N>T3;%%2A;fwc9o-<Ja{*Xit>g8Hp|^&H-n8V-
zr@U^s3v#E-0pk{rJuyu(27V7#`5fvhckg+?GiYVKuAEr?B{9Ciq&e|KiL0@(w-MVU
zW7o^mKu!Go%=%1Sj5i)`Xtx^Mf0uuJ?}D$pdU^G5p&2`hb*1kwbShH6+Lpnltby_G
zy?&okO5<RA-DsTgt%gQ3tPn~$b*KG~U9PIj>PGSz!;BLyWBrKEukbevCtK?O-`|k@
z-mBRm8KK9h*lqUMfeO=&UT1z3R`*cnap!ND9$hH!c&^pLM+K3_Dbw{!;3yfIJ8wJx
zk#x?yYyCuBOl)9b*Z7hS6&t}SuC%q^uJYV3oL_mA-Dd!ohMQLL*X%&JT=)UV-`jIk
zrw?1V6!WF<doCF7r%M!BWxLN0kY?FA5nMTv+H!k8XMGH?Uj@F|U6-K$Ea+Mdm;Vk-
zd_t^u(A(JG*W?A&PDtCu{5FWV*FpvYd(aPp1FMmG^btC;F*XPBiZ@S!B6p9CEMIMH
zRe;B}AscNKCwywcQ~TW&1**{0d$%G<N$Dp;+Td$BJL14wkrN4dv&%eGRCF?Q>xrj!
zP)Zl2HE0Xrj(1M`oS1(Aes8{+L@gZD`S)p%PCd6HZ2GV@neZm+d?TQ4n@%%p4aave
zZ9cwv(=4T?_V$U%9{Y!ycuadioDh4%l?X8*)*tr%5I1b^{4ylvV?-)dRwfZqqp;tS
zHiJHZ88AuVaQf}eK7Q;=0?A4blqj3AxQYCTU?4pX>yh}^z5eF^|I^%^vLVxA`WN};
z#DHTrR`abmH#gg~xe0+pJ6=IY!w2@#xL6?~=#XF9U`tC8j_Zl{XQ;&RfJoRECD$^J
zmhAn@1CDh|t%gXzbKv;sN#K%@Iw1{Rg(N5@aO?AXBqoQsDjW1P26P&c^Fxv%ul?&;
zuG%3#DRH%+t8OX@)cR-hGRMwb-V~ILvALGSjlp){j!3R`PdamhS*4CR%2%DApFb>F
z5cla=h2>Q25jbveZe#Is4yQb|!*`os|64ivZU}DW5Q^2Yq1Zjm{^8*ewO(2%)az8F
z9xkO(Wn93vPO2SeWpCVR&dBbP!xZ8A@3f~jIWu#SD_e1)$MP^3FE-*}Z-f-jIAGUU
z`lyU$1}!CM7w;dl4L4$FlH+$W$uZg^AlJ?0sZAKtm~^S(b6joT`WrDmZ@!E1E=gw;
z@AkW<Q9tczQ7qxkZgxm9UKy$#m6?Fv20==%6d{N2*=XndX|s`mQmuGxp$qNOPD|$m
z)UVROa<OlBewRG<OCo=VYm>cs2E@5sz7a#dJ4`YCtLBYQpHqW0bwHF1eoKFjN+cn|
zH?m0W4i8<5D}n^PHBbR~%4rh@zfjW#F|67kCt`RJPk=PmhyK@TLVdB(CatcgPKp;x
zMYV&ga}9$$v($1?wDpjBND+ReM5<`!+Dprx(kt>bG3Uch&a4E;3IFl&GSP4KUbKS+
z^@5wn@umBM@%FNxTFt|+?~sVZcpSu9lYRZw9c)n@uUbFDh=>ToWA$CYF|L}Zj3q9v
zx}@<4DNcAdgpjJ5e84@G6y!WEY5MI9ubN$XCw(}%oFGG4I+W;io=xP@!=Cmhsr$G%
z3=GUdO*IBZepw<{nyC&6ohFxCyzxh2IKKGv3_F!$VB?U|S{;ltCY^MJ>hOay79v0P
z7BX~@IETre2ipmSG~(TO^WAnx`vyh$^jD5c-roiK6?TUu%22<74)*?oGUqFf0$W>U
z?{jpV@0T`@5?Zz_Q5wm6w^`1M{9%Cuw<ZO?sJ7hNk5yV5-U-*0v1GfDD+-+V`<c3y
zE${!}Qd(UNwB>$oQ%vIS-7kJpq|^RH{e>PL5d@5waK^P7L3A3s{R{|ycky;wMIJw;
zezaIRz<FT+fBL0rz}K%|e+CZ8AWdc+>{jj<mN1G%^nN+yq!2MPaj?`fRQXWyiKZxY
z>2CwMh<3={9wr@ntiUm4k#3FQQOF~qDRH8tb3Ht%XU{&Cw&QM1dOe<#P*CoEyZeOG
zbZ(5NSAO@OhzJfoIy!Rq_PYYoQGgf>t~B9{Yx9$SW;qV4C!FW-gduLmkfpOb^x!`}
zEZ6Uqwa$kc`eM|7gkNpsz(YN-U3dOge$-B;{U>VE(xw+8`b_mz=)M|q8w1<#<g_QE
zxq?$o4Y0Yj)S>JDpHJ6KyG8x!1{jQOcx#m*tvE~ZQ*^YL6x{Jowmk_6Vu3t_xft%C
zYGL-calR+p?lCdZK+(dviKfJbkI%o~0Z|?v9%or!P5-{GdUH{brAW+9&x8rbWjnjY
zCMV1PX#xI@I5moL#+6xI+6X2WO*Zv;Ul^d(n?C;1`N(pmM*1mEF(r3IEh-4%T()0z
zOuW8#Zx%#|j*TxsF)Fl=&B(}d|2{q$$1nIvU!8y12J|iEw2qO9jQ|hNg~&$0VUqDc
z0VDjw-XU(;s=eu}>Ghv+!I1F)-3ss?pd!2p7kh^Z%f9TEtOXFS&XuLKaQb9m6^l{?
zUe8KfTU#0$8e?hxFf!pNHwSecud`V^^HQgr_%$oGUU}26jY0$tw~~Bk_B+pT2OtFl
z(rovae*OKAk?j$yh~Ei-UzwSK)w-I?V2J{#A>Ojg6#nu)m(V(^qV+<^uNO-KRfwcu
zv2F7V(tEJzhsHV#2kORU>)w6E?`xlibpZ~*M1HsG;@t7xHqTCR2JSfftbt^A&`o|w
zA6h-F;de4q8c@-ch#nn%B6xHZ8Wd*wIF4TWKls<6`~J&&2DVD5LSqOcZ(TkFgR`LL
z+-LcoY!yaX4uw!`uzjBLZum5PGmMRvcSjwR^8RqN>s9TzD(=D)4YW6GnlRMSM2;2N
zr@%YnQ$D^e?4I~3hQoI7AC#9qN!vF0nyJYWSCoC`vRzR1@9V@d<;ch=EJLBht51!-
zpjI+I<hdxWwAc4d`~E#5DfI9l7q2?#9(lsw`E`Srx=b`I+*a@AbM7Va1AcrXK-~LW
zf6#&(ddO~h)xrg$-oQh%kbEISK(H9g@_KH{9e@iAjNR(eEEMF8sD+IczsJ#io+42l
zdE?N1&Sw}7i2KQ@qO8T!KaFnepW}w;cb)k8lV&f@TPMCB_mmO3yjaxYq9-LK*>OsZ
z&+7`%iG`T8Y1DRow8leREDC-T6<zVc3<lx0|InSxl~ECctEQa~jP29`|KSinfBD?;
zau0|XN#wQDF=XRAmT<0*=HK+!wYmg63<?<_JaLei;9=_sv%rvXY)rI>=bj)39~z<%
zf|aF2^8Wep;<vvM-;Xo?!+jX}VbpHQy!a(vTUcU$LAvmv2ROwlA%sS(`RdFW6z!f5
zc3+(lYW^>JJ8I&X0*pEzKL&-7a5x`)P<l^`^ZxoY{<pgSUz;>%oQ?66ecSLySdV9Z
zw|)Tp{Q5umJ#zguz<)5jqjVMN6kGU!AeQBe!2XN=i7!<*2}~a@o8Ea56^;D;87EH=
z9W77z^yD8J9UY{uf8%G@$r<DM0clafe!C@AUlpE9y8p^~1!F;~q|4^!re@*HM`bxx
zz>tLv{?W}z9P|hz!*IK?GW)qL+^l%qOEAp2e-+j(iDlE0mjH5^JtiXkoxdQQRx>Nr
zv<`Q1$@tadw_aqz18}MPvd2G3I3_I2&+}b4WMPBE*<YT-AoJ9dle<r_xUeL2<eMBB
z2U!s;zk0G1DIiU*yHAREWmT8ot*)?ex}E*+5_w;On_#wboG9IWps}D|iUKbdf*s-7
z%Tgk1kN#Tq^`MPh*-BqHq35@Xrp{mP05>thTTgv2i|Ue3?tlC_KE&XZ5&)_WedUz)
zhxDkoSWdt6#0z7XmfKrf&n!B;6csgWCmaR;jkI*@9I{+vhUO^;@~X+2^=H<W<b1!x
z)WiiJGyZ(_eM(beTgLlC!_|%@@DgQF3-25<Hflo*yl^RCTdc6tbvm2QfaDAhK4SR%
zBX?#ph(y8~?Ftzg8DAT2hk?lEpp1&hj~wz=^lPn%%~G5HeFLp{g0AM(Rznd{9jKQB
zWh7e}==be@b+coIe-L*v#CZ)-<zSp}mwXNrz&7+W1os*uNcNqw`usW?L7J|K<<{hY
zcE%e~@TjitYZ}etp#V%^w-~RO99|E#GGKPBuk#tTnVkOp8wUZ1RX%YP{)mKxxOfs$
z^hTdbn`moeV>~1(Duz{W6>=#CvMb{v(|is46HM7iEz1ZIIES?xJdNQXk-y^v2{@!~
z6jhrCz+E7Bi3yx`R#wTFUg+y@HASlP@VIzaK*M5o#!s2iz$%yHSMa(fHOtD%jaM&I
z-4|vNbnN-h7A8hOI`3d$fO&IkuzbQmNOkP2gNGEN7akIku8oe2kc;@tPG<J1(BfnA
z3DT@dHGP9Nk^zq0!^1<M!gE%Vvt!%S@PNWW%|C7XNuNLUz2JM!O@gFa=Zl5$Zt=Ho
zc`pk5Tlm}$4tgeVtFl|>cXcYZ3vzPu3JP$2&$bRum}fHhyl}~CJ@?t-6t+x_@QLu}
z>g=3uC54gLw){>?+otI-ab)AUYD-c=fiIvYC$@+`Rv{)TM#N(Y0L=8v++MTNQQJSW
ztA`5G(%KSwo7FSjhdQq5F+~O6aKk@fdS93UmU#c;OEqoJ_30LW&1M8L;CSsPoIvb&
zKQbpHlX!>8eK*L3fFK%VbmjSuQ{b%3E)0u_&J6VXYrR~aFyIXcfahl0CPeM;m2l#w
z2nGj*E-!oXk%{_TL~6j*3&)q0FUoeW14ifUJ=ObjMgaFQkJ?~*;$?j+c*sWYdE9Iq
zduS)?Qu`)e^Df1`MTZ3wPqUa0-(Nf1oX3!RK^M@mj8yB-YMa;Hzq*JGr;5t*y|<g^
zb%&(AjY{@wShuJB9uJSgO)2Def*z!N6A`l)ho3q*T7uk0z+aM{od3u#$z$d4*L*()
zVX*1YjNEg{e;a|{WR#Q7%mN=hV!eZdF+T;RWcqd9Q~muPF5z`N)S0xlTFbMxGPw0V
zx$)rYy8x{Cp-iCYJ$E5AU!D$toe=;-G8+LZ9wDd2_@43V=a-ovnXo8iW247JMn-tx
zhh2@+%HeKIWfY{ztNMy{VOly)PEM|8e%bEdY{E&5wRAO-W$guH6J~6v{&XJzhfoxr
za)O@!vNZQ=85zXuboZ0MWzcsA^EQ|P$@v`ESZPa1Ed}`bf6%6PbzR-Nq6Whr^&=_C
zmrvP*Ek|ltLF9E&z|zh9?TAwNg#RDwpJm`J5^7~6jOP(%c{Gpe^~!DgnF~GZ5_kmx
zB~t9}Ac0RbzwXC;|HsrqCFQ;j%<vC?8rDS>6<J{`yAwXS=x~$gnU$xsyX5fiq_cQS
zZbV7eq6(hul^if%VCzifdYl}&@`4l%nk3wW>Qx2;nV$&NCy$Iix%=^*&{_B%I}D=Z
z)$*fY=z%I8pf;cYSCsB9BfgxwBtFSYgo;g=gO0ndE?rGp`t%zC$;v9I8VT?Ti(g(j
zKVLxJw>(pj!w&zDw>{`|T^Hr7Upqdu9#>bF9}*H3Te<BRi!B@I=(w*@W#dgu1kq0r
z=dgG(=2gB<ctCc}^i(iema3z(i>me|gaigO+gRA>mfP)v*#jmAFv%G-c$PZ$Wa(N7
zBmEYv@`-va(^6lEh+lq5fQOD%f3@qI6h|?JurGTw0RbmxY2~)CKLsG2Jj}Yh!be9u
zTs*w%rsw_Z<28x`5j$_+-hgKW9zghQuOU+x&?yKraSnjl2!gs;>Dg4pbvIt9bimvU
zD;$Io)HJ52Mtl#otA$U-=JdEbbIEq5T!g-m>+-pcu93TBRKnRMfX4hum(=`reV4{7
z1AQ;B?nhQl%y%5G?E7`hEXoXd^xXWB<??O8)U<H2Y67<CbArFS|5(@czV|y5`7yvZ
zGtUowyo(R~J%-LPJ_LQ}=#b6$djlARu3x=;o`!C!4j9+EHfWJ;-A6I(yUd`hR)aJl
z_YFr<PDc<JAH7~?6~ClCL6gGW)(2R2rr}If*5cp$m8u3G&oi{T!km1PCuTQKdm*XE
zWwxtl)Ly(jlp;Af5+bQ@O&n@fdh+4CFRdhfsihYxEDye!?_)Lsr0eD1Vt}=1uHXe6
zeZ_WT7B9t&#CAO0W95wZqpbH5>D}0f$)cj_V<$i43)mLCv*TJHW^ygFo>^H#;BavU
zqe)kTL_1*X+fFm5t2j4q+<4GO1_g5SSx=$F02(5QNmle4T~b<RZ6@jp?Ng>fATpkJ
zcnqGLcpd_e1MiRPm0PGyF^6nG1OZ?xpETOu%HYz!iqA=0odR+pl9l!88c!u@?I3|h
zJbe5_PKS*TuW<f5tpSw%jTrSdjd0F4MHJ*faeBnG`o9{IfM)k4($uuVE+O)S9tsO%
zA@jB<NEC)+G9!JyLCbc`rUH6WzwTx{=moT!Iyvf-w#v)I#81!@NRbkutE)y&&E~qb
zKuz=JyT@5F9GuR>Irk*sM7UY+QwbhcGMxAHL9m>vGe((_^i2MeG*C8&$V4(627Y?s
z*mwgqZRoHL@#cq@VQZ=`CokO_zyReUbINoh*s4fKQ=+0iWo6}g!yj?uO=?cgU9$Ux
zN0%*So$~?W;&(8EZ%yEgmAt}+@(e~we%I5J;8+;q4}I=+%)Yj<%@=aDUtd}IB4|x5
z5_|o<gp_?9qA`coC;h<DVU=6{pFVN&Ce#nu##Ustb(!7?mY0Jp+t$`rP<2RT55=-L
z$6p5+u;^E_LDw3$;M?_%9j({yf=CV%Gx^bL^Wt%?Wj0pkqT6s99|oU9L0qNf=k|m+
z65NW?$f&zObXvhZeR^;mn?e{|?mAKe<N46_8#`)hV)}%a_v6Qp`t=?UkjW|DFY!bi
zalarWZnEL{HLSVtT3Qg#e?n8zi17hxmTLyFwCuG1O$;(7xbzeVndc<RmT9+LzQum-
zq{Dfik(nFjy0WrDG;1N^GWc(0?43(Oh2P3x-@PyiSO|A_?>zNBXGvM^>P>6#x)RfR
zwj1$YAX4;u>1Ly0Za3^ZLax<5hhxz03g8a0g0#31RDE5>zmRVnQ+$dbu-QnQxu$1A
zHi-Q)_H2GeLG`H5Hvl3bcqKz@uzlMUyqOs-RQMAe^E((<v0{LU^>!29$B;xA*;u{Y
zmxG`d500C3<9R5Atl32!99BW~pZlc7#~0T#afDS-8prG54dh77fBZP_t!=Hr@LHE!
zJPTVkYzz#&x>y)q+bfTWko%w15yPdjXR_3&Hv*o~(IoU-F5yP3=+-NQY}~?*ii{ET
zIC6v+UBycP5p{`Ahy&4FVP?6@+<^yi+&J9$upX{>`MZb&A6jeh^6-SG@IIji9VG2g
zhv-#TPwjP;p@xqTIm2n0hrGsA-Gw3S2#{1cu70#V4Rq4g@N#Rh`tk{WNqxb^!V6rG
zd3kt@13EF|_wQ#{N#DXp<8#yi{RT|^zub85Tk3Ie(EJ@TWV|hR$=NG-f5S<8$sQ1O
zMZ94%vE+h$w;4GdKJfc8cH>3>@ISV;`M6tCP{ZQ7B@`v!6D%(aH{Eu#@pi?LYqpm9
zz-PDe_200m{H=TiJHv_Rd}T%ESJVF0gIe4=q^Vz2Sw*F(ojn!Ija`FtD{rO{D%OW^
zt(@%+E$cif>||!NvCBH#9GVz0MSCOq+L<US#o*m?wLVRx`_WGn%)k=kzUC+PGeQTR
z6`d|4`w!kg?G8Z7dT<v0&j48K!A{3b7OLdP$;rc0elK!B26v-k%lu8tZ2#x((JN5E
ztc{MI!=DRl&GdcqLOEyDgw|w!%S0EEJL|qaA0>al%46R$t)u+~6?r3j6Cr2C){G^9
zNexUxy=hA$BO}2<LEfcC=PH-dq}T{dVt|&InwsK}3-uToTNiNnQoX%#gQ9jSpM}*^
z`z}@p1?u}2TW7ry;P?nB#;>oDTNai%5DPZAidAIZu=22|>T8rF4V3%1_)#sF4m_x_
z;r3zm1lk7G|4lLs1y6(K?xjRvApp{N;gF&Fuf>xfvI=s<PduxDS%k__a<cg3NK+#l
z?(cq$uW$lXUqG1`gb<trHv!t9ZcM8~YTcd|Ffc5{CX9g?&Ea7N1)(5XvG+f#ogVVY
z6t%+RhNR}Dvrtp(8_YB*2fX$lQe6GVK?}ogpt!1r#;-02sIayMHYsfh#Xo#T{;J9F
zT1kfvH7*BlQ1cs}<A(nJ^f}=cVv+f&ENb}kttVBPPj#2Ozjb##w6Mf-%X{oFnWHIY
zFCpGy(10nuMM}Q8LMr@)gcJxGC>67{vN5qT;<ozNAAW;QnTYE?0u(mNU$415xNkj5
z&Htg@oc8+nuGx^PJ>puwEBmYu@t>|&P`MRHLh5MC4R#5uJSO00VK9URzi8&&jVAC&
z{jVhhPN@czA~xNIW<$<M6r2|yoJ9?kHD5>(BUlG6D5>QI<jDQ|pCKH5n5e+#Wlg{3
z>W|(`ZrLH2AtJMYF1jHK%9F%6b%@=o3KC0G0U}al$Gq?~zX9A05wbr^_PW4Wfji*v
ztkW`>@D~t_QuIm-X}|k^kZszGXS+$8xq=8!(}w#x1E6L0!_Bw3A62zH&>Txv{u8S}
zWyE)Mlml^#)hpF|NLjfX|0R{Re9CkdLPbM!{`a@3v#g6Et0VfZ=hE^fSlY^BGzb%t
z;*(!lzOpm3BHN`lYte8smhizdZh<q@xs0b#%bc5M_$7;mqDxP(|LC(vmegspVB_sK
zy(1k$$%R?Nf-{qm0tpS*@!N1?cWzcoUecI@obl}s8r}p&n){ZlWeZWxwQnwJ*R$Ef
z-%a6DzhbGy&wy(l4uwyj)KUB)pQZ`1mVZS}jXdrjf-tS^2A+tt?{(>WK9bMX<g|Sx
zS{5*jA%!<09^w|@QvLZg>T8rUw>V8WDAu_+JhzS=VWw5~(T92PDL>XV!Pm~T`*6eN
zRJ{X&l!kvUl%>xZAr|)b+g5D<`=$}mm$I_TG6K|(GQXBh{+PtN_0%pBq?8pg((nO8
zYYMR9rqBuHE_@{YU-!|GD}O*_WE?A%C};->@b8MEzRfMBz7l6*VuIdCaLhDNAXI(#
zP=)kpQ7;aD<wNwZgwNFX<4BYHtKXO%?t<=;|4%$h(R99k@t2f~f$#p(zCZ^-fvnv<
zlA+VX&45eo1CC&21tkTg`F3gbm(T0G577@M##H>Sk~b0mEuZd&t5I$M`Z{whk<_d_
zuu9w~O?hhiTU;G4<MQH^ABX%`?H5&LO;Xbh&f!s?{Ubds4?gE9U64aqSQO;tq5DIj
zyh(<bb&w2}+h4X|>9wfnBq&Go@kB-96gatfpw;@q>XVqvd$Aig;4>@h$f-)y1t*dl
ze?AKTZ-)*lL`5Yc_U?nCZ$<Xj+c^bAS7xpM(#}V^*`&i?15p%96{CM^{yntscw7#@
z`Xw6hMvU&RtL@BzWejD=%sp1@!=Rw9v44{E0ps02SAc(md?*qtD+3?D)4`wbM4M2N
zC#1~Vm7LO}e;DZ@Vw)3EEXX!C`=b=#+CI;1k$8`l_s#k?titWd0YgbKR9CODUCNyp
zaIq_{(@p`D`%~jIvvJa$V99M4XhTT)A_wwG4ehgY7;zCSepIAG&pvVLSB7wP;nval
zK}}V#3W#9uw}=|_HBT$a%6?Dq<X=hJj=wo){)f1J74x<V<ll>Y>T%ic^?4sg8@nCe
zAW^8@@EYbgrY<?SQ!BSTmE5$wVITLyR7OY0{L(yX-bh==mA-E6ERBU`ncFvSY=6Zm
zN3V=_Cxu%vl!5~v`7t9yKuGX_kYL@ydTo6Tfuz8bUmo3XoTzGRh|xs)%ZQrziGA{O
z?p*P?D0T2#xf<q%dh=Ue0#OkF)m*BJ`xc#+28mJK9Ughk%)~4)BL$~k318xuv3G-m
zj`+lHl*|z%*?+k{rl?T+^fu?mj{0=Fr+z1Gr>t8=Z)Z4(ux|AA1pk=vtw^i-^nE<Z
zEdrDo0rg2p;0*+FR2(O&G?X<39<t5@Q3lFfe$Ty$Mp;?`aR`ohuAtl4u=HwM1Ttc8
z_`~pcRTQQ0mTmEhqRKXqEvVCoZ*1bQ{PD8+T@_Ug&oz(z@r()ZzZVo1PSrS3y0KD^
zx=#?he*IRl#dIM~tW6}_bS1$NF(%tmt!`Sl=Nyoj9|BYoct$PUzh+8pLk<EwF6aRt
zHw<Enrl4-%tUL8Ce}K~YvIW<qLq{@<?w)ZA3o3v|#TIC9FKsnnshm7|P&Dm4RY&`n
z@m+5!-tDPsoYcsZ(qnP+dr7;i{ScVCU}!HxA6eBRU{$z<y-s<I(I+mYPfV{&gpB|`
zA*<L8p-SbH+~2)qWZBAMesj>0udl6hm6V)}Q);DPq<H-pTy;kN@|WJd>iIy;z+7t8
zYE<^rdhi`bN0stbVwOukF+xSn<~uta$wB+RyaLLn-)#ywdk-FX)KwT3(m3<r<JFW!
z^=y%8L-Ymt#%5J>2ftpra2#XW3&Y9vbd{U_%PlRjFt+}JC0_i;4=t7@8QiYIW%1GZ
zs&(EIitBeS)0bNphJoHB!LnEDg*B(A$uVg2$`X2thyR+VCK;oNi2V5QIZyshlYi5#
zG+90mmbxTrq>3k51I_NJNPoKWc7xBa^Bel*<>d*7oF_||wH_^+YR@49rgU}2es)qT
zB?fl;jXxEUJ!W8bOltI|1ws&TGd`~D8~z_7n}v@Hpm8Ez^Yzl)$iH_BO{3H>F}+ZG
zo|9Kd%;h-NF-ysg8u{rH3v3Am-)=z#iP(1PRVq3pO5xqVZ)0n#Q(;k91Chlt090dQ
z-YzSzgbav`xTd7DaENh&K|xCL-GGca=E;0w1uV0`E}3wi%*AEkb9(f>F|K=(>hIqK
zH$t`&$+ggln853AN2m3Z?KJ=VE7sXaRq*(sd-8)*8P)&))%F%#RdsC}=mG?!6$zzF
zP$}sWkVfflP(m7{JCu?Z5h>}Gl5Vi*ZUiZ5B$RH>&GVk~{e#0`INXE1SIj-voOfIS
z%m)Awst3N7$$Ww7|6xpX1}#Jgzs`MM)7_LPOiu8{;={e93&A5>6SZN0gHkoraNNcf
zP5rVs-E*6GaS2HWTh?=h{*JIHxq;z9s%N$ZMfnH<%@&bf8AkfMScr>@OIqCDLhlDr
zxxu=Mc7_g5e*SRCl9ZJ6K10>E;Z34qzTX97I6CSg_!6{#(U3HDLjA8#yO}|jPBl}n
zfIm9z2)PoHadD}n@)TYW#Jm|Q379S*X?Jy55oPZGQ%l1aG#Pn_X?gz%XQF)Ga7sMe
zk@^F=R}UR)zputbK}*x&-xJKJk2}5>#$SKk$q&C|)g`k+=KfI2(%b+cvd;jtfL428
zU@geV$aS~;d0qP2)5y@!u0N`GRzEc9Q#lB@>v?(kiqTRjA^vT-eV=Jy1P6FF&v@Sx
z5d-|Kwg$^l`Y}gw%X7uh@AFVo0V_KQUm$<e)EEcFeN3c7`Diti0D(dT7i>N|J<)uF
z*icEJ<{tPuQ<9YhoU8MJvf=0RG0-Ct;fnIR7V_O=Z&3i*m~B!pQd)sfJw}C^$z|S(
z$Fb}2^NfbDN#(m{W=jZ&c7PK$AZiCVh3#agOTIHB{$oQ!gM$z_>0U-+S5;P~@9P&A
z7k_aw|4C2{ViN0spM5K3@@sY8Tk}o(LNG3s_wqL$k(BVkFF^pTai_kc(966!)!x&g
zY^0)+kx^DqazRZWOz1}+Tv<8NQ*i5UYwH2kBj1DHRnUu+lnf*$Ch|9b3@wM>#mZ_D
zw!0^zdc3vZJ7L)3Bf6@v8f)Hs??4nPQrvrJtrik^z}tHplPKI#E$Yebca5ySF83B2
z9o;uXS$E755l_vVF7%YM{sU<S@^a@*$m_;mqvUQeZl<JZqPn`e*v%RozW4m9sPV*M
zRMvxTb@&kiDCAqu$kFdL`h-^)XaS_ppJyy!9zy2t<EL?3_)Tsu8PXUXQC(G5zBbF=
zlAd2*FRiCWiGZyzR2SekDJl8NkPYKp3iDkj!{P(NJ;uuNPl6x^)8u5^N}9(efdK{^
z<7%@Xe>JwX<qBg)MW(F>zJ4mQ=gDicF#?SYn03G}_wdj`9`DaKA6iT#6o59Bak(}&
zw&OiboUUiTpV{MplZ7bDeb&N~5{KzVr*D;Y(4|BMrh2X+FqpcADjvHf=GNp%h<B41
zF#Zu5VJ;!|76|d)c{Le<<sPu_5d=&R3}*)t<a*O!4|UGF3nlq!3Msl6j7e>jFJSWt
zDnpHFr$&vQ-R;{s{xm0Fj*he5Mia(%W~=UsO11|#g(Ln+le+jJ!-InyPBTMmigJzx
z$56meBC}$kdm{T;$cTy!5-(R!^^;ckRItQc&Pf%5=?~8u&~>1L2_eZ9TfTe#rRO0h
zhCd{oZwLnKM!aW9Uwtt#kBN(^F5ifaH#+L!^OvM^thk>g@KWbW6}&g@kJ|uUO44M#
zk_qpwv9$?c;pnrB>9UN$*APFA-)re4WwKrn`AlttgZQGng6eNC_awl8+7(j+#EEEI
z-M20u063+0$_?5X-VcHU1No7WkrB{g)=<~XF?W`JJ32IQkByD>J}Wu@7f=5xGj;h#
z&UGOZQ+U_>8&VDb6JRg+YCnte<yixYk?0-&%kK1uo{jAHmD!8jfFyi4L4tsOHp~a*
z4ke_cm${>}SK^jH<!y$Lr1->?fZWx56coe^wb}L57w!ijHB9}6zDGxiZ{N}ciHmjr
z>(g%>o0xc?B9X%3)>-sllm!pC&3-WW9OPx^Lcj*_JTWm8R4OGpmo`1HL85O{sBxl*
zfA`^Q{}L23!>{ACz&tIl92gx`Reu{D7P{h+dCbDfT5Qo%Q1l7KfB6O3$<)}`jQpRo
z$c=zV1%*|68JUnO4I?p>S0j}X{>jY-p8p>f;LBWf^+BG9FFjMw92bieSLE`-;ZQsK
zGiW!!(gi#vkgb%K#s(Y3;XyKq%3nZk%#;A%`uzOs$hD9_ScuJ6l=T`uFsj@$6+=P6
z4+#h>FLzRL)_9+aVfH#}nfTGZ2a3V>g_ge?zrnIp7(j)+#)nm_!e1r{byk&&s2BO-
z<$AmOGCVy*ujY?pD>QmvQgAIvIJ3NxRaa9Xymbp5`C5ucCW}>vW@2b4rf5f1u1EVv
z8ny^m{QXWYBP74Lc%>;Q?a?mtJr-;qQWdBwGBEIesi&@49lVtW(T~<wdvLgKZT-Q*
z+7lEMh@q{6`ehAAD=8_7tK2cSR|n9)U1Qa}Qs^E~kGXO=?h62wTynQZD9v`Wzk_L^
zHj`Ihzg$+sj=bbjQt4PM%L;+cZln9Lu7*_TB>^#vleqpbSASxwYhapdmRC|;g9ZVq
z2VCQG6%Hi*o=X3@;fobjr-Eb5=+tKCO_JpA^AFkrWM-iMdEfHRJzNU1-XqfqWiNn>
zo}2`OS1yYGISo1D{%^+x4x-|)CQ)jxuYiM#eTT#M1S&{+Q<ICPW^c`I?Tc9D1Az;A
z+;?-9*bSSPd5Os0P<bAMi8nVlFTh(Y&eZb;+-|OJ7&%6YQL8}9=X^D^9~Mh4&I%Pj
zXD3={p&x%UT~URN$mbjlB|SG+&E@AtKlQ2Y6>e+l+~NjsKvtBfL_7}OMH@Z7g32&G
zH2u7Jz*;i<(H@pn0%_0_S)BLj*5@5__K|8o+I);<^wXv~MZxdM6D%T*YaS?%e3siA
zw3iRgWngvtw(xKJwKobF&i15nEu1ZpZ=K?2Q8sQFX{p~>U2d}WuUZ0&5W8s~e!n#<
zga&+oUdUHhtPvpJDK}>ueL^^cGgFFLSlgn=@PTcJkB5cR;=OsWRATxBN@U+DL;fq3
z!vmVya{c30<u&9UQP;a5#e2gm;(3W6;Q)+QwHK_Fm8->hg()c#ONxc{UxzYl3d){8
z)eRobj6<HK=Q(ZuF4!V~GQNk7)Y56kDm<Gg{t=ZU!Bu_tD3thWOAJ+Xb#FqFf2Pia
zfb!|f*=iJl8oxCKEd`C%T)&pXm_g{PfI^O&jls6V(GS?Bu9_0{8UR(9@H<*;yjwWX
z2kWU_^7zgz13ESiKs`m0ZGAE4-MSd_;Th3@+H_v84zza4d@BB=qRB+32!cQ|U=n#e
zz9lgApYhgjl0Y|3WMt;FPSn04noML5FwTH|3235Bx`pbW^*x<Q#P_GbNa*lNVsc`M
zLDj}Y!7h1qJ+)A6137u*r?8z5cx*uLnD|`9Mf0cQIru;jpd+{F0{{WVuHRf{GwP|%
z4S*;pULenUI_B98F#s*b&#DVrc-$SXk5?P6`w~e=>so!Ai0Su6)90<t5WyxmKqImY
z>Kz*#Q&-W<ep?&`SN(lnxpKL#B5!GVDI`cnF^Pj3=`h`CbLq3_9MxKb_`xiI{N%N;
z%Mib5l75Y)7f=7Q@DjT{S(%zhYjdSE@ACs6fbn!Nyqn6c0ssdr6>JW!O`aD&K<Wt$
zFo5&IsF;R{fF#S-6z}!x*O4!6Ref=B8c->0!Sf<OVDcvG$AiNF^5t7yMRlDQi`&mP
zeNv1u`r}1&B^Mn%eYUxSK<)W==O0)&T3Y2pO%%7f3d*QxnA*a=Cm;E4_nJR>B4^aH
zbUEk>GDd(pLFOQauBvjGa@|}ZY4W<9hb$a<`KL>V(pCZRtiKK^C*4+$pjCtN%)ct<
zmLI+uC;)+erC3W);5DsM?aX|dQX2MgK=?pdz1cvpx@M(B!FQ6HM;hDu+i)_|qe3yY
zDWVS@IkY#FWOD($Y4dOd&>rD&zD_K>-*}&$$y(H6uE#PKBHiSsXXqd;aRkttjGx;z
ztUOR02bc?55o^#jbf^haKk|IQOdly|4FxwqkwE*MJ5=&ly~F0F7eRpsI?jhWP&I3_
zTpr0@RJX&q_4lZCcO}2zftz2+zTPymN5oB832A(%iRM<9gNRN5I19hrBTAH<gabqh
z5jf_1gliTx&F};!-6robswXQ>y5jZQm%o%in-;*)N_X+VS?LwWnXR^QE|}5Z733#E
z|EHn;)@Y&eT4L&Wk#4p1?EF>zyYA8*t`8EFUsT70EF`zC*j7PJ;rw3n53V$^oNT#~
zhYlMZUa5vvF8jlI@^hkW1e_evR012e@dc<zpHcHGoAT?Xbm|%^JB>b9&mbU{ruX-B
zKaZftu^DiW|L8yv;JGD8^m<WJ2tvIFp^16J#vGqc*Btb<p-6k9(JmZ%s|hjRVV5B7
zPjg!6L)+`QD!+#)-4-H5pdVy}+2XTI{<nuOy%1*)Gz5BV@9B~M=^nSsx-AWr+OcyM
z<O_f^uTH$X=cv1gi(EtfYB@Fn9F~;4OYtn7gk*(8{;}?<!;!k~fK8iV{IXrYJ0ne>
zaImly$wVi+r`-NT`TTELND*-p;1wB!qRFk0L*F%RC~`pq$bjln9akjykQ%RVX6B+d
znNOItt@Im|6Ttvf3JnFLt%k#7T@p|(WfW7lZ%@CCH>!nB?=oA>L>h!bN8x#gUMK5*
zgv}SPFdK%J0CU6-0&16^cQz<Ml}rXjOs=~tImt3<{S&knq2tTrlBOeMlV93>ZlGZx
z(aTIjh&80D3Y?2;s2R=88JVAS_2r4`w!mi?IN?vj!W|550v{U)$patp&|gbEOzfes
z@|BVTRyj!2FB|ZIG|Sr-BvevWOKn~l8v(XH*jdUlyaRki<3|=ZFKazLd7Ib%{{B$4
z0B=&rGI@y2D!X#MCTv_`8-ym2KOzMcV)O(SmIs!$)@mx6l3WwOov1#*pTUOR6)MfI
zU%$?63+|{oEd+tXqL06zm;?HZhKy~$vLAvtLz2;BnmN&w{tE4)iKFa;g9F3dws(ZT
zf$szqYY1;G@Qt*z(&8j+?d?fPNC1QKW_jL*SBN<T8#$?;bbplEX$<(N`Ipbq$A3lx
z!pQ4*y%#WTErP-eLqkJJ2_L)WU$UGgsdX&sTloU#<W)19)yK++9+1xh=7pHp*WW{&
zH=onm#Y18DS0gH-{5E#UpheAS_U*e=F~%5}dtfmNDv%Jdm1T$obV4dXN=PJ0O90V|
zTsN~jvyG*tg6z`_03vz{c0Ij(6Z1<xi*^MSf<~f)H&el&3W<yHy{lZwM9m4tt0;n7
z^x!^z2$(l&s_IHJ#)5E53@-4x(4|`SgJc9!J%ma==LaUmD>23bSpQf>bc7NRcAu-(
z(vni|Gcot}uEoYCAFfHNi|gaw4O4v5onKHqJUXVTq6x5qp03=k?oJwd76d@J*x_M8
zK|x-JoXZ9gK;ecB7z)jSf!DyyPB~v!QccYXsD6>sr|O0FvtG_?lg0F>(;f!2^?BKq
zjfv93L&{mo_aIRL%DH1X@L^S1NSKBA2<kCNxbf#4_@SvrR-9A2$A#rNI;x%^|H4wI
zHBZW70~HZWiFx}TOneyW>1~Zo|Bl{wQ*+|74<h1xqM#5*{vo{6xb>>XEmU-J0Mc@W
z`JX<#dp}2M3pG$jAsO)}FmN*#^bq$0wN;W%Oj=i0K_gL5-9RllDK!)g+~^leXbQni
zspR<5KNRCR6c3yO{8Wx|mxp$+?^xbnlzf|#%yjSZysHm<Yv2GC^&uGKVL|9b6JG<o
z%cpO$78mRMti!{LL`5Z%K=YYtl`*1=PwGE19qJzSb7nKJRRCaMb|5q=Vfq!0+ea$N
ziqyJ=0JGK=UqeGGk^Hr1=hJbhg~!Sq09OebGdv@s?%SH&vLE6!u47>(B_{n%a)JII
zSB2yIIAWk<DI2=Zm=DL-YpY92iixoaGNyi?*Vawg&HZg<X-g^OfgjuHeeicorm#>%
z4j}B{f|dCi*lYY`S}4dq>d-;-0aL83euF9S^c}4*s6qhoaki=t@ufhWI+)u4D@=N1
z%hJZ?L1F&lCYX4>9K{fJf`Bh3L!@*F7+Z~J-`<rwqN0@EV=v7r4z;_rH}c`bJEB{}
zp-~Dqh=>5tPg?pE03smm4-S#63V1Ued(^^rMQ#0fLB5==`Ld&v6s2^WpFIV-x_slU
z^DhZw)yT<`ara%IlmrQRz6UyR51@KH(ugV;@U`jZ;64bQWD%@iYI0n#0t;y0dn3|!
zR{+D9Hz~Tar_aG@w%>ZI^~+pZDL(b>+qd69TLBPQlm@hie)tKI7-$QppsIc=ub+#T
zbB>9NqfH6<X#ucz5eR@%KP-g<lBKw$`2sX)XJaw0FKnayw7}o^d10B#Vwxr3Yf&0!
zmX=Cb3M*n!=I<@E($9Z8ZhjBmm?&=!nflXD$0cOGk{aOt=%S|#jfKSef-@d6wDIYU
zngwzj?uopMXpVv5CWn9OEhX<wZr}D6XN)=6bo>rXliS|L#-=Zy_49rk=R|q_?Ae*s
z$jF$=z4!eX{y$d5e9fl7W!ehBJ#*Z<S|o%ET8rH$LlyNz<Y~1d9zNRc*}1`!lN#O4
zE#kO_doC_Ly`IGJ?-de6?01TSdV4GH;-vVhq4n0hYRx*v{r%e*+Of^9yE8I(?r}V9
z4g|vfa{@P_K?DsgKnZH3u673fcnYsyz51s|!%Tnvw=XXB1M0%1d@{H1=i**1ld3%$
zXl#zfRS2`9w{8CzLRZ9#O%h2SY4te%w)`T=ZhR<@Dx);$IOco{_Le63$k2)L+)MDA
zs&QFJ0-Q535%Gcpe}!S&><6l?L-dM4Zz^CRI1lOrtp!RcnXA^fZyN!7b?Ydj$zoav
z1Ea0Q7#KVXz1$+h=FhHvyu+GP`)clrwtYD=<2c_5>7Vg7S3Yr&efnmXS_f-kYG80|
zWKBg*pWt<|%Q`-6+}5Ou|AI4s|8yO)uI`nn=NM$T5%w>Jjs_uQS+~6Z@c({1sd}Q9
znIR}L5^+15i&1M_8+iEe2JH6-_4ZQt&*6Q|i9t^0(3BMRSRTJ3Zk^G(Q;4XjsfGBm
zf8-1Aw`muRgz5s+7oomewRY*|^Tt<iZ|m6z7ISNJ1Ew#9-7&NYR5%<UKX}N;b$<LB
z9&_--E6FUX_3iCSa~%ROF%VZt*%Nwu#rN|-l+yXcdu8Xbm>9GJYtBFJe$Y~<67ki1
z`t<iH{iI>DFB3hN7qI4=E+c(Y`JKx0Yn_*d(CY<tMNf<e^Eq9Hv`f6?q|8P$Ow7S=
z?N^QMnZ*iZ%D2qfKuF7Gx^`IxC-~Jryl1>grSuH<YC+~7EHEm-qu}R8DhcL-))n@j
ze60A}Si12uZ&7{<yUkB#;5ZC8FfUya+=$uG!0h+rHPpzL_fJeVZ7$NzC6ymPmX&=-
z%aK!7U<pGk%(~pZxI`@3-ZnL<eW^uE6(xB5X>NffCyx@Jyz@*q_7Y!H##YaOtLb<+
zGpuVpq4vj`8l&%}4tV2S0pergrU%79QUg*T-#yv(Ewm!wT>~K+FLKg5Es)*tTI0dq
zPP5bIZlR7PE31Nn0xrg`(O4HdXvm?Iz6O-?$9PsrJ)l+AY)o{&Z~~NYKzX#=_X~Qa
zPKF^!^U_G_wuM;e@78%P5sKXXy6{h2{54=1V^qq{7}PFHEo^PWqax$u<Nx^9KzGtm
z$)^qvY^?zSXF?ZK>k<%dqlLQt;lo&NF5^kd2Fplm+6%u+7D)9W0vBrm{gukf_f<JT
zfO!!yK?4PA9*DID?)+i}q&M6|Ip<K?)~}&SGD`B&PnmlGx>RTjLSJpTy#Xu%7Afha
zg-t&FU!hO{-$mAn&Lh7*@4P$Oo~XDHm)MV}T6}f|$s!ME1xv2@XGgYw32=e$3@mIS
z?8I^Dh4pe&M6<yzQb~Tx@{S+IT2<m#sSF}ioi|69STAw1a`Ki=66TX6Lag!6ZDQWM
zxq-a#KD^cpnAgD_9_V`%#fJvgY6I@s+t`sSlGkZ6JI${Nuwj5yGkP$+QGXJWY9JxO
z%60{dG5L4)lL#9fkPU<;zX@UOCTf0BZDDKeMNDLDX8F%j#l&Pdxn?^Uw|dEl(1nLM
z+Q3d@ln2(~(~S5tTXn5*V4P&QO~Q>%pGJ=XqR^n~y>&W@ze{1uU6;X@omtp~;MPzP
z4vs+*U7YY<eD8B`5ORtA^S~G1=n66*o|kI-<Z4)Yq<7g@ZC`l)=}MGn?}I|`mZi=^
zPh3>07LS1L)tgaQ{Lr>Ej8MtRm91*mg=&BfKr;w!fzQ9wv>E3z9YB_}Y*duV14Bby
z{Y(J6>48+FvtSK8Z{tIUiwZD2>yekxJLk)ougFs>DYUQkxxD$3|BBtBxG^guZPen^
z(%49-^o9J+DyT>($*G|c`@D&QjVdK2R)&EPme}RKR6w)!M4xf%h6CDkn?B3TG+VP&
z+vW-^m*9P|g9j=o!a}@RK6D~K^jmz4wDg>IJ>yF?YfDOZQ5p==>{?Pz=8T7`HtO}U
zP(sleEBLhQmERB{fBO!EZ@U<0ZY9L%Mui)eJg{JSf9sus5fK`4I82=q*1+Fwv5w^U
z%Ap!OO_JmJF>10BjQSX}V!@(k&HBDul3-FL19y#fadI{ve#)x%w>Z550?j0L&0A^?
zzD>>LRrRH>J?7Nyk=od058eyTN!#?WeU8&}YOMQoEV#(GZ{Hfpq2dh9OVXQnrp5B)
zp1uL$XJ05BxO=+|G6YC`Tg45Y{&InNV1wxog)D$6gHPpS18_BQ|FiYoZCeyHc0h)j
z{biEnV0*;jUqInPH5~*ovSj$#3|1vUdJ0fWXB(~7fV7q;DXb~@w;m@~%Ntd+@e4Ru
z=1Yr?6BHU6ORc-+SKy(?y3)&`T=7t+{AqLWJMItiSWhjWp~Uo0pASHmRu9*S!+}a)
zhiYoN^s&LcI6Xh%G0on#c^MeZW{(~|Y7+l8)1SzZQ=fZKl1Ga4BnUbzk8J&l!rgOz
zA4detM&pIpflu{^F)9dj%|kYB!m3X|CEDB3!$=qkS>JMhBJKVAC<sB*l`X<M(6?pC
zp_vExD2LtNOQhE6GRE)OLr^aW3zd+9-GXwS|5^$M!$#Bi@Cvl>P+y}0C!{A)V9D#l
zcP<?k(0FWD@_|;pht5w+wYF%o8Ou4nW>2*6R*_o!)0kT;@6?8X%2QM1CiW*3LMJH6
zq<u&&$SC23*v8LA8UT!kbN}E80~T|h{KkmYp-%{8frHz?n##OD_ctpXBcON|_x8Q6
zvby)n{sX;3D#pC*QoSZzol<?iZ5yktJyN}v7daWtY-ybN_n3tnoq`n28$HOY59b_!
z%d;f6482KBb@f!RgSg9+(rFCHp!y>H%<O!Bd;#w>8&@{s^HeUAu4h8Y{bSyn%S|1t
z69ARUQ=sZ^NcldmU+YaTa^I{+wjESeP##OaIGDzazNI#V04~VAFXFy*@QEykVGC9Y
zhl{Zj*XEdk5^u7w?amQ&SY1Eay*MyXDh2UWin*d<7LwkSoOGIg;qBkYKpa1gPCWvK
zpNCU!)~)Byys=nb{B4l{(&9jKiXHj5F|F){;*TGaN)$68cPxMad;<`bG6&mGTiLc0
zgV$(ad<3KJRAK1jnGB5`UlAJEvd3*+hDU_mv9dR}H{WG@G-TF9zT=!<QUG6mv&&(|
zW%oGbrhuij(8Wg}DF7P+JG)$9y;z<l+{6mKUuaO1iKXpq^G&*D8l1thWCRbY9v%|l
zI_hkX4NP=IpScNB<6KwvOe9m;#r}0jWoK>8$IE$-xey?vQ2Z7@Jwk8n>gX98JwD*!
zH(y}y#nGN=vzaw-$ck?KdGOaKS==q1&f|IAYFI61qG144L|1XY8i@woZy`~`F04*z
ztsm1*b}2Rl>hnQzT0PYNkN#xi=5*@j{OMw#Y4!eWf@Um7Yy15wK}zB$*s_vZKMWM>
z=Z`GB74(Y3@(J-@8W49SbS7Sp4a{WzUH->WG)@h4@nPeaT%_hrnx>LTJ#DHapYVG%
zEDwSfCV73L`yPYm``B1msE;-Ad4wRx23$RR>Wrv^J39i7A}-0UK%O(1F5Ui(UIDYx
zpk_m|fi^5$t|_hMHs?fjR@TB+Ks(^5kb%L$GS4PGVA#Suf_~#nbNB>cB_>-ErJZ!P
zKcvR+IBewW3^QqHjn~?36)FGq+njcJOk_!p{58L&?HK=oP2i?pnIe%S$L9r`KJkr`
z{t98?sR|(#v9ixN!r_E*Z>3d<w6z~sPn=x1yDUADB63kqOS|0Y*=Ri7=n<md-5^iW
z<sJL*H0PakmVm$n{Oaf?6pD6?O?GvDWfN_A$Z<cX@yymK@9gzAEVY$Kp^@9;-R+%U
zFV1qT*L5x-dj`-SJ{1-iYgdY$dwTm0j!}BuidFr@PsB~u+npth{K5%GkQ`MHc2K}>
zeDn)tZZLLY&51NOJNFcyd~s<W`m<-$a$5I^u#tzHi~rp9VNk{xzwXvzWXqeF(7$UR
z937-huEjVp9UdjGxzH{z*QUA6<!DjShDU&Ytz=p`Ypo4U>EleL3tw_CnSl38+O*4+
z^~h+G-yOL?JPgcUtn*`m!lJZvBOt*<$8FRL)RE86ccX+3zePmgQc`a4+WkC!l6Kp9
zU0+hYGB)7+<DJ+Mh1TL?mX8!0M`xJ;HY+c$ncjnB4o*1g=Fr#Ao*=S9K+Y@|qwRb&
zfmJ@*H#P~Q#lCH3tQscEa5AzDd>-8z8?nA5bzN|>&FwjvE%w{0vYz^MUTV8FLy%VS
z?EY~x@=T?@r}Jw`Q4tHreJSOgi{yS77{Jiw7G_Q7tu%DTEHEij;|ZE|ESxy+V)&ci
z(NTMGv2CcGZGLJbJay^0?Rzw$g994^;%}t>G2uPo%MH_xR%Sc{gRn4kp#utr<E=gO
z#8~s5=rG?$?vK+zw$-qXMDUNFT(;_cq_Kw`K%YRR8koks$+rIy2m6@(Vg^ud2C?<8
zeBCFamSe$=Wy?T_$9m|cRF2o(lRNwqQx{d<8G^RaH%`d*_MG`jbhbo2#=6f?KbeT-
zYc}|-P$88zgAnYrtUUfwEOaR7<&y6i&qYgmmvGJg+)F9(AqgUB+&TIuBffNIk6ie5
zrzz^=AbD5!6<)>w4wtdlA<?NnelXRJI4LSBVESW}>|@|<belee!K5V&eS#DICD#1?
z)1`@+?-+ZV=m`5Rj<NB=WgYogcV4&5g8kB%_woqVgE*_trACT&GgZkoUdPUBgVRED
zN1ff{;_cDrO^y0TG>_i(zHr-`n9s_AmHElo{?n`>ckPB@%e~Q?$;g~+&2&i{n{TAF
z2y^GJ1NZ(gR~(n0^|=%#$QxGgb-SO--m^5qc*6V5xs!~@jg-smisF?uLzKyl8psD3
z@wBS-Mr!Po%u|OK&JyUc#I!i=YRMW9|NHIG3SX4MciwWUonjl(4WViJ7{!#@7`J;N
zaIfBNZEw~HkojtW_Zbj7@0C}mclqbJ$3HgK_Gvv}Zga_laS_O0Z5<A3J@N{_lK9q!
z3%v|vFp)Q5z99UwrQ0&oQ?Ovktt-4zZ^9@~fGg5qaq$QLK6Z}dep+}P#>vTvr;sc4
zbzG4@OFs=6KH#oSsbwhRl8+cYh%;<(V7|-KlwUjXw!ORCONWi<;`au(sj_I)2hE1+
z;<S~fhun*({!J;pBcZ-pB|)5t5tr4k$oSewk#eVH5}a#s<8ll`W%MCy>+5@~%PEO*
zlq5IzX&!~y7;^#|N~_~=^B0s*8T&t8t~%J*c#Ag~t!p}PH$Ivspv3dMK~kVvpgJ^!
z{+duOTrj)oiSl2aAn|$81Ggdm9Q}LsFOMgOe=G)<uu%&C(s(eM)n29;b+CPHHYYYu
zo)KuPZUrl6eYsh$z9J9J+ZGuk$?`ovVIQAa^Vvrql3S`4v=1<qea>GHZ5uE4P+O^w
zIzC2!_0|0Qx2%r0Lo6f}pqyzE;u?T?;FEf2JolNSz_R}bVW=!avzI6ZqRD#tedNcL
z()aM#^ypdb#z*}feJ-D{mfYN2&vT2}4CJBH44um`)9bI#Zj+No9Z-5g=hITEbwcHP
z7Iiyt!D5PI+oMtZ-;eAKm6t1>jM2p$9i*H()_s(xS8hq=ingX93OhO2Dae0|N6q<|
zE#9cbLkL(GqAmX1>URY5MG=;BNCq3@w^?_^+(&|Yg+BffEYaQ(1;Gc8VR*o5HYc~?
zyB`u=eoM>PpW4De7=C)V#^F;gipVlQV1tbM-_AbPf^-a@Z1y|<bD;$JTxACA=!mFD
zugQVYrM)^F<51II7Dj{-PjducE*_nKmhH#ilpQrI(-z_@Mfc}e$OEe#VFlT5p#dty
z4`p@}dwS&gN{#NosvR|r`M|V1iS&(?K=eTiTLPDXoLt4S6m8(A7dED~cB85-Pq-xM
zI7Z88)a5#f!~C{_%XErzBgw(K8qKHa@0$?1vf6dm=0&Di_;yXpRM6rJ$0eR@i|-r_
zb`{_8cy4WS{&(<Jfg~!TKvq}&bAx$Yd^<#{{_OY0MeW0OSp*?lcp@pLZe7@CNVOR7
z`}Zv*{Z;*At@wnAqrA{3qn(5N0uv}mq;y49Szl-G!frT{?iO8w(ynCK;xXkHu^sUD
zUZVU=PQ{8)v789Qy<=zP9m{=VFovRYaIo{BLgevM<njhlXjU{z#B*@_!`Ds!6xiMW
zoO{00g++!a*ZAp7`Fbo}!$bCpr%aso`ZVVMNI5#}bu%F}8nQ;)QXu&f-lsqH9NyDw
zZarG=vAWrC9unbLSz|piP!%6t6S*(r$gQiJbn0p&c*}8ZxmXA*w3C*!G+4iJUC8F`
zZ2mk*=OKT8w{}}rTp~!FK0Z?Bpfq|)M*dH>@P?P3HDHbx^})x3H91Yh&9XDi+LV}c
zXh>3%77ejxmUH47h8=IqCwRok)zNye(wxtJ2o)VbJe{-SZIk8Mm3t?>Q!0Z>9p^^p
ze7v2s)i#3>449iXMbyGkak4oYq!#w(*5=kJF^MT3QpGTYc6Yp|2k$d(f!mFs)5ZF}
z9$cE4hydB*x^m6nQR$)G%`nNee$Fdm8a)FJI8sVr@9o%JdYE61kJ;t*8{mpK8M4GY
z>h3rhAAI%@C0(JA<Z282*lVoOVX7Lsrz8DAwSb$4hl7U$0|8e5x}UxnZCWG0?{DX&
z8FT7yrVnAi?kpG_8p<BD071L`JLc!JVd$^X<qQw6*$fu_1AXe-xWTEZc2jV^K11z#
z$KoP!#@KEoJqhi0%L8_!$_@6or##+&5!=2N&j1o)oDaqNm`JmPpITnd?^MQbEnAlR
z07ihc71#bjhTd-OVo-6Ui@Ry1Bpo{9=wLw5q!2uE{{b5#v#?O5$-@L>pkBjC^`DJx
zG00en=(-Fl;X84sMJwhJY>XG0h39r_7Arkf%n_v6BR+;i$++V&1?}w&`?kI@@RA0f
zCibsC_ilKWGI2QMn3-{t6>q1<Aj*b@CCU$AeG>Gn)hdhEFY)IqxP~C<D2T^;F4roO
zei4)S)qRR`2G}lhTfJY+`(_EE!`_5Ny$PFf`#Ul3Gj$a!wUM|;2fQ!JpW$Y+iaX@w
z<FcKa-1OBpwVDL;xe#Qa;<SBOSdvtzv{}&}b8g_7MMqp|+HbPkb#EyY(CA5#hhv<m
zNV#-u#$dVPu3H#gvU}36V8)WTapw17nkbYDM_2|MjlWo@9{I%`7IE*UG;QqL7VnF^
zds2`qx~ynFSk&3dJRYkC=!`V<G^|A4&S-9(PpGk-6azIY{>{Qc;)6@pe0+94J?_zb
zPOc%XOIcrA6P%cO=O#GrGV-iq)=P23oI!-v@^@vOGg+~Wn>|eC)A^V%z8W)D)?Nd>
z@sx&_OYp#OI4GI?PNm}a_Fh@8%C?mwwu~aEa};(y{<N=OA{MWhN=CNv`OUq3Mhr09
z*x=ibl`h<RiDyW)<%NGOP(~!;O$3I26s4gN&Ek}fuI}xfTI6w5Xno#hURz1YE-$a9
zZc@@Zs)Brk=sry&)?*%@zHjXu4w2;b)fO|7_lVwY@%|9fV&3e%A;qxWX6oIiO@b@v
za}e<a0x5?>-=pc%dFm9`Na?&7Z?^DOhP(h%Hmf(?JPDk>8-wX{m3t9wkz{Y*8#z9m
ztnK8|c(6Ly+Ypsy>8|d)pNAGJo6Xk3&%|(d`<GX~f^|(7u^|IttiqHV^-JrSz#mFc
z^HVKHmdJWC!?{sN(@ux-`c!<56eaTVbB&Aj%zNB@sjZy#PjaQMqf(+=J%c0fBdbP1
zr#hNK*kSH32fB#d8L)AR@8T@K%~-x6KC9YI0+I=2HP3hPKcR``^U|3F@e53n-`Rm0
zMdgO2@eXV@g?)n{-uq%@5h$DQT37makuQ^zgdzvZR;H%>E<?46YD`2VJ0K2lLi1>S
zOw{f4V5h9CTwcNNgOT>C{`gJA@yRRA52MG}t@eyvEUaGW$odo0ll_A_Y7&nXLv4t@
ztKxF%NBd=Xcr71R4{pbeYp?5CRPc)@S~w0n;Ol!K`O4~_@03yxb$ow^u^^0RIAi0<
z=l8FchA}EFy??A>;P6^N-y8RTH;@rqOvCN{ySf86<QXeqgBNb_>NYKhke`Ir$Y;k{
zc2poKQqL7i0>=;&I!I{Q+A3a(2uERHkP_{LqoYD>kIXsLW#=WU7N~zxU*t%>=cU0G
z+X*}v9i2PKTSY-W@ynR*B_B`MWDYa$pM_cwT=_p$Fm-lz4!#((A9+K&wDb)vBpMwf
zP1xx|b@V*?)61nv1?0$QBK^Y0lYxPe`VnEi;-0s3Tgd9_d^3d0y?SoLT%4OWwi}y<
z=Fe@4iXpR_yqrvmaA1z_zJuFleo2;>RouJa7>@p$%0MXgpP3owo!HK^4b)v6>?i$M
zc-9}RKmUYIT<}xUf^{?{itOWv$I7}Yv?digckhmp-3hb!l`j@;H-nCTb{-OKJJFb>
zGc2va*QJ(r6C)r){nOm6KzM&{@4WND`ZF6Fe#?_a2^2Ad#z(N@+t1X~$96WjZE{;r
zUdvFH%v~*c5eEa*P*ZGFyHTol`3r~N)p2iIVl;U@A$fDA+V|k{L$H)2+&(2QL^0ZW
zIwIX2sx=QhFV4`fUOmoXK5wxpVy$p;Oy;rN)4QDcyS=_Q$e%+@j4Fl|pvZUe%g$l7
zJs`1>-M{l$1O<nC|3M`6CPPFsyXFU%1H0KbSY(0uL!~uJEi)qOGOv`2p$%&{>w|!c
zOpjrMV<%|WLk;p8Lhvn%f$+&mh5XQDP{g&EcS%$?i4{^0JN$WGG4+dthI-wgB6eJR
zfPfn6(su&FDYwn-4X$V7GZO5E%kely1ecu)j-%+Xz&xVmIBqKz0>1b&&P1xM7EMY~
z4~shIC29A?@78>F?LXSAEi9b(+=UYz!e1z`tS4Zm2>{V|tfIk9fvn7mlLwRk5W_!|
z@5SA}v258!9J7YwHpwQ5mADUhE{{tIeDd+-Fp(~^rFgEA?Qmg=Bo|CX6@Bd+_H$x%
zY=+l`clb(P{Cac$4W>Wp%`_HpcK!RD-DufuW6W*i#~lENNwuN|i5G}$Xbg;vMR_Ih
zThIKKkO_bQu-W1mEL)pjy=smP@R{Oe6`&|yZRao-XIm2KpDH)8KUEawvz^Wnp@#{Z
z;)T!1m`8;th>nOD2%#Y!%D_lckbkPv;6NVUi|QZI!@|k(#ynmmT&oC-OYXBW(-U{R
zY4^H?74m)XUFRZfy0WtJzBj1u$H*d&4MaOX4+CH6Cd3#3K6)%Chl8a4XanlD%}le5
zgapV>(lHBZ6{!2VGLbRM;XrU+u-5<nX-<6m6jnu3(<9_r|0{Y3i5!$}`5h+z`r6=T
zHOW;z`s2rS1XK~|GO1dH6}$m>?kRK2L;`W&<3E_sVuCk6nI#cSQ7j2+so4exJIIiJ
z`E}gePDOi?M!MreEgXstp_CX&VL}`QV7lg!mxqQEkpq4m`xlScmF^LK4Pn<LCugRL
zXC@4H+Y&zF=2#mTr14+p#in!dT}}6GlMY4?I`hFqMQG@`k|pJagyE{`GBZ21G}CUl
zIBCVL?}xvybS@{YGhkrMla`c<jruS;lC6SvXr!m5u71|!6J2i%mIQYhbBd-aBf06J
z^!2XORG_ZE)BIm}SZCLp{M9%4GV7VQaRno#lNUvkdS6hBxY7x&91A7qD(skWCZ6I`
z-94Y~IRE>jzzGuOU>T7|UKiz_X2d^9wiQ#68!sOnImqijwL1B+wma#z{^Z+{MHC|k
zN8RjP%tBsY!Nor@(ML8iyRH1=F27o=cR@BODvjS@cj(3vB-+jHUu_I1DPb2_&Eqx8
zghbkK@w*HUJ9Y-OG$e0No)i>>mesdn$6L?ok8=pzWqrhX;;X%wQ^+@BKjU|F8v#;x
z8vdizpBV<A;`!y@9ANoS{G+N509@a3FI(f0EH1U+Lyo)q+5-h9VtEL3Dh~c^4mTT+
zBe%KJm1e@@Aq0*};Rmq3{uU&)o%VeiZ#$Qe0m$(o=)}f8hRTv(L}_XX+s~(Y!>9WP
zKO21r0=}*e)tTT$kB&wCx%Tee+e$gm@w(;@3HgD$J!4kCdd&H;a&>Fs3U}s3`gbKX
zua6hSRHa)%zDyW_5&UP|${{%w#lt5An{nFjhwj#<d2;NSaqBcrUOJmM5crC5aSi_*
zW-Olh-0Ih5Q{L7#`Bd0vKO}-eBp)X6k$lINPe>=3HFNiIMV^(VCH$5cmaPI@74%J{
zdH1)6oi_64k;k4<+|Dld=SU;(h(<PU4gCbTnDZfYuU%Cg*AuXYed=9PMLPXP>s!n(
z?bl$l78b1ZqPXR^_05b^krk<QwB|KleY}&@X{z{_yvm_l(Z#95i7DE`%8E+JVZE!W
zMnY&?$ejp*3V|Y}=)=Vt-)O4#nUj^7O`*f{_9I?r7XdP7tB&o3lHiX=zg}=72n0f)
zo`xvb#p%HIXV}~$>avAeP%j<axD>t|GauMFzCDpK4X||c9)s%${y}xfVY7%uZ!)jl
zc<$&l=DeCYE8>5$|Iy8U3PCPRyu-uM`Ru0uQ3_!pP`UoV_R+dy0p&|XUHC{EbKKKT
zx;qT35*$u=dzg5ux7>GneiK{ED=28P8T^Hc-Se9STE*vH5ExT;X>;6(0Hq~1y*jpg
zM;nQ6<xHDb*4=O*6IAGZa*;(^s_pdK`HdS>zY{52ECag3;zTEXyCT``Rij5f@=dU3
z53Yeq`XLj)Gs9bkxa&g)U43JM&n}s`BU0x0-c{z|M0-y(S?4j2OAHTjwaF)JO?NsS
z>^<q$DmMCTx-=H)CnYVcs>Fznz>3E2a)273*y6O{y)*<P>m7ps%ZcXMW^QrvcyUo9
zXt?&xz4D2o5UPWkB99-N-E2wZ?CQ;sXbgo^0F72w2@EqvdI>Kag)nD?cuhHE`4CX!
zC2{*uA*<`FJt;h`#U8hg_I#bLqaeNCzteMbS1zN|Rxy?RW4@St>{zxc#ADp?1-N7!
z4|ovi>nb^o`uuh~Jv~JqlSudbj6R4~@5Yy0RM~NP2m>XWwq?Ii$0MKQW@zg-N80dS
z&Vg~QQWj>fyKzqiD5As*-aeh7eEY61oxIs{4k`~rT%1rcGGM~oU&UQo5--y(%E^(!
zSiKugfiI(%jD|djdm!3E)AF|eMHy^pf7aGC*&4GextWFHdyNGI1RfUtGcvcd)RB@3
zN^MTnX9Q6iZVq1Hk`|l=bJ50+*-kG1%$VMh+WqA8_!{9WS;n#F)_<&5TMqH+N>k0>
zzgR1GUMC+l;V>k?%}{<|h`kFJLrJ;oHp8yfwHt))Og62*GTlj`@j%1od7~_&o$2qc
z=MG1inT<Q72N>0YHk1YWU8JNOYx10|@5yv7XY+BZ+>?5ud!%Sc<TxwJmmKR-Q%kDi
z%y{&OpX3aU?BUkQzj*QDq_0flK^)Y17*q=i%5El-gfzAkZ4iUr*YL1{f?w`iW~i}z
zYACrLlo?TE6ZIjtDTNhTTolnN_%ihEW1|?&y-uH08S?>!h=_|R9SH`KmE~oQL|l3a
z302!J!=_J}zVC+IAC*W^I_0P)CdPqXjO7a}I-;-#<NaDoON3>CS-l?SL=%V9kzidn
z#F;)A5s5qdQ~h9FL;3~pm2pb-nf*AGuZn|;N?C4haAHAd{c{`@?GAR5_wb}AB_^s<
z{m#y><!9oH&dM<}G08DBI?7#;H(L)c#0Is8rKP0<*Xqg1k8rZ%<DeJ0jDmubXPL%v
zs+~&?olBY>?5J^J?F;Wj_0jL<{1m7IfDNw4uguKfhra*Z`c#%qL!ai*p}9NpQ{+6o
ze^YM8Lo`y6k8_cDCtRB)_f@m6Lm7g=?WJd{XDHvkv(tfo(PxN@v9RerM4ex0W>)&3
zMS<oXgvMk^WS;f);xYH=x%A}JVsZCkPtStHGd@>7Bn`i7{Ld?w^dT?ryY3}avCw*}
z7%9R}ss*4|uHpAVo{=6e2p7*ITQgQH{c3v3c<mREVgnCoRf6w{BLh6K3An*IW|S^N
z$0_ws*ITNis_UCI)H|Na$uZAn4idcRj%tG7?(Mj>7MtDh`ho#B)U7n${P-OZ5{-$G
zc0Tfa_r?-8H#bR7qNa<VZ{Psm!K3Zg2`nxV7wrDaZxndqRr>ng;+n2oplTIFH`vGe
zN8i9VLy#`;hWP5Pk8~79d)Dy-eqCJ<a>br#$%@nCg7SCTmyr?E@eeKu>0<fSW%(w}
zR>#yvrhV4~m2F+;WamD)_=VqncmGzK@oNm)zyYP}mow*KutttuR&9v6FXk3EK%F%m
zDg`$;Co7Xgy*@QJQ;oi&37pAN2{_q(p{P^liFWn6IkS$73Fqlkc?EgM^Lg0uXajGY
zY(0xwF0>x*D)tx#SwM4qjB8IXhA&A*F>H9f)_;^WdDoAb?pzNX0K{v9iKb#-j0sZ#
z{x#aYD%h%Hj{|OsU(ZOpb>MPBGGR0Jmlh3;VAEH)>Fy^53d)Ux?Xw2AXuGSsI@l%W
zdELVpa6|k$+(9&^6ZdRXS=k`;xN0t&q=n0<Gk<;YZF04rdS-5|;6igf!wTS@fJrL)
zv-lWZ!i9=o_wh)`IvykA$#1c5&lTWeY~r%Nd(wuByElGW<-9K0(#6F{7ic94IX(|F
z$(=jDiQ@kI+^S4~02wy4)(k}q0-`q7X7iX7)x<+J?Ruj&rpLO91+IohBm}M|Nt6`g
zaAnHtR~LCh+=y`5;$Bu>p^(o`?OfLB$>Cb(i?V$D6>|{w%2V9aYjDJhZqE@U2^?5)
zD*01uaUuUH>+%0iDF-v5H6k)HSBZ+$+xs#d#pf^Id?|QeT5TUhqMhz%4wo&{?@j-i
zn;G+qU>{YXxw=2vz1F!{tKw%J3kp=C%Ohpy5}5nrBR}MOwFVnx#(ji%=ly%HpUVd0
zQ4Lf%$^75By1!5;uzp)2lmAQZadIR!`#gZ3g7wq)V~2@JZQGxptzd5gAgy})>l-KU
zEVw73Lq*P$uk?R|ijt63;Ig;Be<qD5O6mR*@-Uamssj9#)7*><jv4#KuhV)-yrA61
zay_H;b?=K}{p$3nx;s~6c6G_>YVh$KXBTJC<aHGMn|^uQh)eAjoSn_ZVtn~1!{9r{
zzWMyyp6llWxO@BKCHh6hy%~Zx{`+j8)io#Xs-$XO^P{(~n4-6)K6~n%y4t?Xp0X;>
zH>tiT&GBP1cxGiRe|2vAuVZbVKfl#MzV_c7QjUCBfRFf4q$t||%+!Gb#F11hX{kOz
z-OD92UzAa?Q2^6A!EwK15DYx4D2PI*-BYM?v=(t){49)#nh!Jh>KZco0|k<{!RM>r
zZPoreCIQ6zj$7ZBybiw;KfMbFp`-jY6-6yH1dI3j;~U~5w>kD|0Ooc2JVqV!^?%nf
zu=0hX_b@~;uf}E&WLI;~GJnfy*#+u1?m&HWR7|DFrF+>oOM1k`G5zYc<5rzdz8!PP
zoGN*x9==}J{_hw!PE28DN8P!zBaKVRZ}ajJO0PnP-?x6@^n?14V7`5^iiFlBM>90%
zt}a}i8Fv4(s<UX0TQ6jDRAK%2-=rIcC%Hi;Z`fd6pkc)6fO^*pmLUkV+_r;SnHW?k
z$F;`uGYYo{UfJ~~gZ$~$=SJyZy}wN6vi-L+`hRzn_m!fTptWsssIC6;v)DNx%n!fQ
z!NqtT-IHdESEC${hmK6+qp8cW(s(BWng5QX3eg~BTd;s!Ee`(;t&+b$5_N10ZE3vH
zI6QhmbSfj?+spo8LD@}BZMfwRNAUjs-pStjSf05Xy#y7M`U9;RuzcwOOAncz8l`M8
zO`N7OCF<3~IG`kxq3pfR2E+8#(up6w71x)*8#OKi$$kMg5n5gmjx9@|g60a=TRksJ
zDAuc9UHsW>Yt4s;hxh5b*mWQ31o>qMCZtc4mUhEOn7g}!)Z6zrEk4iAJPha$9C14i
z`aYN`c<;XpPm^*eo7*^+TbBY5c}kN76%|!RMus|ObTdzoIP^Fv$<zLFbKK`*W)c$8
zVkj}V4{p3Ojg||Flv52gw#8Xl3NHNDGqzQ|#fJh5oE}`hb`ag88x`}R7thEa_n$Wo
z4ZNC5a96&r@{GG9|IkL)Zob~M@KpsA`{38U!LKFurCUh4Y5_Emdp14-q6Z?jqK%L6
zukow7Da_?(_EWVP)(gQU3S@kXWUFdEAE5}xrxkCMt8`#gV@a;VeU!NM0Jd?kGVc%6
z|Ky?a-vr5LY>T4M?4=cYC=-An7Tif?Wie%CDo}A4VFqa(+B73|mU|BfiHJ!_iE+Tq
zFFXto^MG+PQw>$8i~FShDapH}nl~>h5q?7Ba-<}*;-&G{A+l5Td3Zxy^gXnQ85!;x
z)jI~PKfX7|XB7yYFwX^&|97c4&xSL`=bfg%@+my^-S`tIk_^a!&PBre$4qgu*@Oj^
z;<~gkPg@B)y+UM+AKsS72xv8iN1~@!Vc(g9KH$d5dbVo8<@=xa3)V0&PDqugvjjid
zZ;H8_?k!A|muvC#e4_u~yfc1bHI&s*QxNK~@AegT<m(yKv*Gc_^z91IAKn&y9tC?S
z48ITiDp-79{l{hbMQ%4qpUJ>KKjEvt(5idQ8x>xCKiKf397U3exe)y~fY*I)EL*@E
z&RM+cU2+Al{rBcTk?T&}|NHaRaYw#io$<djz(4yi)K^FR@12suJ%j&SHLqT(5d8oA
cv8GEjp8M_3HQ9D=A@K7=N>Q>z{MqaO2M~=0ApigX

literal 0
HcmV?d00001

diff --git a/docs_new/examples/images/coding_structure_diagram.png b/docs_new/examples/images/coding_structure_diagram.png
new file mode 100644
index 0000000000000000000000000000000000000000..c30e04a5b07a9bd962f4e91d1cfccf22d22dc702
GIT binary patch
literal 73586
zcmZsD1za3W@-~*B!8J&57Pka~69}?+Ah-otB(S&!cbDL90TSF{afjf+Z3zw`I0UzE
zc<=sq@7;a*&BA17db+#1rmO0yrwCI~l6i*l90LIX;Tc%=tttWnaue{OM0*N+Q%8v*
zgn)pkWhp7C0+y7dR&lU3v$TdFAb>tQ$F|BNzrpQYPPOpY0a=*aE27ZI!So;`1bDr)
z8GRK8)mN`q9XOIPX+fP29fK2cd(g<tdVjuf4C68lK8fegcCD2Z^Imfe$Mi26og?ZS
z#XAb2<!rc>bMqFImzEdhlKlBVYNYUC$`q0%`*nN3|8QRWObF!gGjd2@=z0=yYVo(a
zDZlfvp!waI>w8-cL$jA+-%KK#sdC<iO{P;Nw<rp4OuXyyh>H+?Avb#Mx$uYWIe1Ne
z4bd!BU`=MtqF#}+-hQCGvVwf$?9QF3hWbJ!=EHO5o?JDi&whwR@(-71g<@gP?5sO`
z1;Z;>#v@ReDfmu$YV0~dBcQ5a(mD>u`%FXmlBd34#rwtrvJF8Fgn%)2b_Yc!wLWtu
zxYJ1X`@~O`dZAXL7t_86UNk?G{C<%BfRU`VGQM3$LlZ{WK2x@eFdmf&9`Xc4tpU*j
zn<*+HumI<12#6t;2q?fABJg<*e1H{+3r0W%e&Yk5w>e1vcNMuQ2l-#;l#e&QQIiCN
zf!}H-4iJcqqlK+gJUlN3c+`yLdo3p|MFl|<TPTN-sjV@D!wqWp*aSh?O%OPRLY$1K
z-JsSsj)HC?w13?p2%JBD%}GoB*DX#~BD7kHD%6s;4iIWS4lWKZT2TyYYHDEzQ!_!;
zx6=P?4*XAq*22ljPLPw+)zy{5m6yZT!JLy@KtO<#i-(hkhaI?s-O=5~$;ge}#*ywH
z5Bb06yoETLI9S>_S=!oAKR(yU*w)!egqHTPqksMT$2cKwmjCX_#_^xi0w&1$c!iUj
zgNyTD&jy+bKYlByV(A94)_!XV1;hjNA<8enFZ|d2|8wQvJ^rnw*1uZ{yn6ZfmVdkQ
z-!0W0Ar6wZP@qdE(SNVZKO6u3;y)V-b3V@eZ<6?jnE(0~kh3U;Fz3J4Ocdi;PuUd$
zf;a;B?VI;*h<ll+F!J8Bh7jXP^j<;}K|}(iLf1&krwSD`xsu<&)b!%dMG|lqvat~*
zl)o{~iU0cWssGP!U+&&Q@=D9j%k@l7x~*t4O;$^WvixQDERQ;fi_6y1^vAIb9yNt~
zrO3A^Z^i%n5De@{pq#*Qmi+Gvz^!bE%uP*r|DQJDzJyEBO&Px<(L#Oy>mbx$LZ+og
z`mcjhN<7>~MqkNU`hU9#FF!!U!~L(rFPfHGs^JUAS3$b}eikj4vj)n;^=$ia-K>F?
zfzO@u>3+kqnTWzBFj5TJ4^2CTysTxlaaa4cK2}?4R^>?bDvJ1jf4<)Wx&rr=P?6K+
zw5B1ne$l2`uD8|@<!#Og8e8oTiLHWU1D;lo3!9pxlE(j-NUQ>N{O*a^{c+63^I938
z0Tg{D_xOh-P95}S+U`Fl5sLRc|5r4n5T8qFeV8wr69Qs_e%OD=$_d|3A^ad894}Zj
z_B#T&B(t(x(SANt3;K^4?<-Ts=PlkZy~p|Ag=h*A-zpvArbb4^eksq{>SUe#*cV|W
z0(kSM$6>2~o!kiCe@y%qeX8IWHkI()Uh`qwRfF8Ca9^|yr4>K~@p<V&>gs>*TIDU0
zgFeTa`n&Z1?f4@jusUW=f-3*BoH?N=`M-#di@z!Uhvfd!h*t!$dAhuUlz;E#AF2ZM
zrRo&nrar~^A3}W83coaT+3_;6Ol-ox&*ZOj5G|mfQY|bW<Nrgjk2)zn0t{m&QVA89
z;D6~T;b%a5X$^?#!2dgVGhi52(qECK{^vajmjyH!uOfVb{y$ddagr*$z%X8Yne_YL
z8B5UvathUyg8qm8{WbV+slYJO#&^X3cgD2%DNUhKs{bkX937;1Mn#_P;Agzc-Z#*l
zAL^R=(|>knD{9N108%=j26_@IuldCM*Z6)&p-_43aXa4ZH>8soxOKMNo(yeey8h%A
zyX?dn)^i+zY^9iBPd3;Z@>=|$uIYSAo<e4pk{j{;TS=zU%H72>zpibobBs-v^NjWu
zN&BT#3^Yn-^_G9xvxbb@IM5w<?z7#)wwv-)mjz43hx@ycwFpidv1DBvfts~&R=#4g
z;s#{&NHX3&J^NlFvvB4@E}{L}c&}*J5B!e%afC~H_QgE^oVf+1ACBl{2Zhs*H(PP?
z9L!cIg9j0uZCVT7*E?TzH!mLU4<CkYn)fvp{+j7;bU2{fTd|VZ+!xC}K5Ly7t><a#
z{`!7V{6FV190r-qs(OhX)Bi9dy+cwKIZM!XnTN>Vct6}MrzabH8o*9V;QN>x>1*sX
z6l)RiPvxk6;fMOd^T+K`h*%^`Be?ZyBVKvfc~(EUw%Ph-Jx0uiQ7J>KUZ0$W0JnKC
za-pgx-L}n=Z{Cz(m*#WfM&Gr|q$0nz)RO|K2}+cYUf3CqqXH1$QPG>dYUsj`cwnKk
zglif9sW?!X72t_{i?;2pZ|uu7>gUbUrzGu~g;xU!hVkXujP%*7^uX14xXsoX3w2Yx
zW5OmH+7P0EE}a-X=V^7blZ}D1_qisR3e+*%2xFHj#*1A%OKlvpn>OvpQ$Bx*)UTeY
zay!c-jsK_Hm(mwKEB6~rVpU;o(lJ!fEU$Cz(_dd=o*TPmS~qSc*;MWEIwiBR<~$2S
z>EXarl-YB;Tn!lsCbvlugasB{$cSBkPW3vU9|6<>0%N+w4n|gOzdx&h!XYxg@0dQ}
ze%}@7;JpWS*`0S6fX0oR=(BDDuU-=GIPw22FF8`S_jM7Opp%8R$HRF;dur?XLW#8y
zxq<s8k>g%fe_BP`_297Z>F_v@^(~-A_!pKkf{0b}+moXnVz)<<=|?Q@wRR=}p;|_Z
zUR%_xM+>Fxe9dkvP50bZ654<12)FOo4oW_`IcT=k7J&Z3LUI>Dozb-w3E0fuzMfK3
zhMxcQu&n`1fy>kUr{@ykHpr{RAOCN3Fa1!d+;a8^O*0i$9henP(zUvZtHW#G0K0o+
zM|tzXN8Z#Bp&3ITuad}No*}l?@`>x0lL{u@7tPD5ORmdH6wXuc%z}tCD#=pipVU$#
z;USso^a+KCfn9*Tq^3p{{LhX+6o_#uL3zo!v7s|k)Gv@8?l0O$0F~9!ATvT5UXGwm
zm%9G5V>0@BJ(5jI9k7!*(vO=F%Gd6HZofHbD1buMRGZ~%Ck$f9@WpCh4RN-=gjoc4
z)z7);<#?R^CL-ov?~maJ_6Zi~SE!oet3-V!L>yvEy`oe{LIytp?=Y*Li6B`0T;kz9
z_aUcaoJL$l`~AYDwS{h4x{KDb&%=$OWbpgKKqHR#dOak1c5k|Gzt)bO(Fl*$@%Mdu
z?gE1;rEulG7x5b9+2&RdyZ-fM!k!J6?zx<n&*9J35e43n_bhxi@YmhcgTD@qn|CIy
zYevk7wN3GAfc<B&F-86f9j2b`maX?%7}MVI>6lpi2#111&ByXL`5)73!!ET7HH-$?
z%30_8j#WQoEl3X;0*3w<^HJ>g;tgrmZ$>6m8ASnype_+gQK8Vg5dP({F|h}@gTcq5
z0=Cn+3lS^+6APhjhh=`(pGQ|62_Kb3ukvYx&Ks7!tq%1M2)4MgY?^l9Mg_`kV(?S?
z=j&AK0$l5nyw9&&CgfPpE~68$!`Wai)29?@*+%!4c7|p)e@Z1hWa-gDE`EB9Q3-wM
zK&bgMA)l;&S}W*YIKdIP{bcMJQ{@_kFfX`6>gIJnuz!(xJ#Acne#I=+@YzYd;h!7K
z%ajg4!PQ)Xzh-+ctVnB@#DdQox3eOs+9s8Rq#9oM+#WM#Fb27MZKs$-Q*DlkT<DP-
zo{#z5Pg1yVeBJ{5l>Mq~B7<0!nZgVQIU(14L$mEn*U=3Z#U-DsA;V8onG7}ToV`8D
zKkfS|A>e@9K@DB#D@k3uH)n>Ir^bK)Zc-n`?KJp=SgWJWH<^!~?Z?5<LcHhjGYVLD
zitW`f@0e2wUwuNdTEaCl_UjHiFJlK_l@=P)Om0-HeQp+)7m}CeyzA}TyY>K&r%aA}
zgCR{IJ5`k-$t|kM=+BVheOzxa8uE*((siHP^%o<6k(rKl>|+^;KMf)=n4=r^%h$fU
zKC20<fuT+oDy)67t9Vc6UY~gyyxg!~zc{MrzM1F}%dl^-p~GXbz5%>?)sD|6m$mll
zSVJbHTNAgaa27YCw|RGY6K=xYli~v}m*oPyIa`kWGncQfcPo^O`@EUJrG00mFLCQG
z*CLYP=NI$lFS=`GJo#e;wn)2cC>aD9g*hLNV|?a1yXNktp@3rrsxH<<X$nN3Yt=G!
z6NuZ8@g?z;mb(-4{!vMq^ZEnLD#Ua~+kE6Ozqp7U{m9`31}EXoVY`odc%UKbOwz}o
z<V=;yoWPS2{$)~w6_v<%6sgcd1pNA-^%_|>y#kX;x7gl<<PA3ex_GfdgRZoOzUNkR
z-NhoDsJ(MRcae!<b-cnHMfX}>T9t0AimLQQ5tmac-y$~5dNMTdODkIl<F!JPHvmi2
zFhMgTQ9G@nOZL=mBG854VF5%Y>glFEjWrmEEDQ;9Sonc9x_b8s7;GveNxLtHOCxUi
zc8lLd{8u<@`8SIl>$i)RhI8HrK3h+jl;6a$HSbk3Fp${BQM$}^ezK)cylCJ>_<p)`
zGQ`Obn8~6$%JYln%}RY6j|KC~Osv^%6O%E`_TlV@TXV-AJekY%ydh5}vLo2(sL(zF
ziJWEJ<roQ}?$c7V#IT?rm(G>hH{qv8`Zh_j<4|JDOc@lxSOg9>OU@F?xy>$T^jTIm
zdYH$W#()JcU3uIA4X)1bUji72Go(CDaXtC6{;I!}Z3(RwE4<H2=C*>N<U%f`WIU#_
zQc^542=G5yKKEB~dE~5<*o$_MttT%P2YUS#t&2Fd?D&4Vtq+I$3lX!5q?!W8|9$r#
zL9j$)-Hc9&sgxYcY~=n0+);Vn@*Xhs?>8lMG~V<*i`v>t!~66Gwhzu8rtzErGZkPv
zye!$T#Y?_=$&kz=OrCkP)Ze6A_5ty}m($y5y}CrCo^10<DHCu{)HA~m?|+w6*swQi
zv!4O7u-$U}z-%7t+gbJzUe23Mt*R1kgA&dMc;%*az7g5^btrcF*T8FS@{kchrm-SL
zqCbU;6S5|c6`k*hj>{ib@*CQ)H1j1M)|y8jdlY<=^Stw82$D?cLRrJlh4mkRH`f2t
zP=wFBA<pc{(a^OV2d9u#MQeSljKDG^S>G(N@bOg>%U>NDen(bAJRxd)8al+_^4U#{
zP0&-&C>$r+bwbwVfFkO0gJH^bej{sj0XI8zk<P16zJr8KX}Ft0jWE=N5gQ+eFu?*A
zMatbJ2In}HXdRa(mpWwrEfdzaMu;){bch03WNPfxcH{eJqbm4Jn0#a-`<Fm|o#!J0
zWp{sUf8@NGaTpZAO&^ThiNh8?>iD*3RkKJHtoOqk9va5!^7A?1{CE`&{ds1Kb#b%f
z!nIEj(AQHK^ZXUd^FhC$Y!$?())E8j0v(61FJ5<@jz4$@Z=cRq4fiGJUB7fY`ao`_
zfc><s;npVb%(!S7Ffj^sV@AK1&C;R6o~I+`sKXzg8{muvMZC;q7{O(38p$bj3>6Q-
zU)hvU6-H`_H8NU@LEj2{K~Eg^!YR&@o?)EXKzS;!6y^TUSFuFqCW;7Szr6Fw?E9&v
zN=V(g(8X8?)u(R?;|%o#yXLK3@)R@FqM0ib3)Y<@pIrhai1pzFl9-H}cE`Pn7hCDJ
z^@EL$HD-pqbM{%%7zfe!5un*1kIop&jawLdFh>L^$L*SH%>0*PN5m@)BcB}KPj+P6
z2G53gW!L-7)rdQ%6Z|r`QgO$cr6Ckd0-6f4fzuY2FSoUBjtM0`V^&<LDT_eZ3qb}-
zDh75Y&KGTqqvos&qwbA3=&K)cdeEybU@RfvZrE&Qb$H#yQ8y0f%}>X%gX$p;$WM?k
zjuC#i$-6P$g!<RJyvxiIqAchK!`xPE0ZPK#T^Z<DW`KDtgkUgdVQ3n9ITuVOo`0cO
zf3{N7>y^ID*d^rI8L<xd-8-KbJC?Jzs|lBcEsHtue(@l!%A+rB_uB*h`XLYgm22_~
z)~EQFCT}S=>t@2X1+uh0-n((XorfeWMiFi=f0~}NM#3`2y$RISQm8aN3Xy-oNu>qS
zWfo?l0*qZXsiW~&I&;rv6AAdCr6rKk>(4kOJejrI7tuZ}(DpJleCgihp1!2g1MGSy
z{<*(koP@Cb=t<ID%bdwb6#9&&upjKf6G9@t!`=X&FhK}<Yt};=>|em*G6H%uG&&I`
z471afqinlXdyOEqD!*8^GYI>n#b}kw%9B9Vw^X4jp*0&+es!Ft%-97)P*>=RcURml
zCRwUUy9aFIB(vXHY-j$vNK`?B-!rjQ{4DBQzK{Tpd-wE4&(0?#3Cv}79n*p{(h^yE
zx20lkojG{OYE?f{D#gIy0n+!|21arm%hf|2X!F@6b)+HNBh?&I{Fn@hK%gFq9}3}u
zylmCsAF$Y5AY~AzMXNSG>372EAvu-_UYI*WV_ii}ea(As$Gc!0!M-QUcQ6`v&L~c-
zLWpl%Bcb_nmKTnFwFqnorPF2V1H+Z2wZGxfPqJ(2oFX<E#1se4Vh&$2kn9R3rLy^}
zGbXe@+#QC=m9xsLij?|bjb8dMRvj|?GH|uT94pAGI_T?b>;KZfD$VAWYDgn-&T52h
zeMJu^wt=_3OsnzXXoxKNq%h!o1H^=mH#e)So*<6g+PwW*KXapR8^KV_*&4U_Bd>mc
za%>bMBE{!$^s&ibXIc`kuZh(h^mTucxZ{0i+{eUV(>Q)f6W}!LzxSg1P?7W;q2^<o
zn77w7oBViLaMIvo{4O4W&gDu)S9yEoKb`cNR@yXdep4pUU2HNiIRV&n$IQf*Oyd<x
zBa%yxEqlV-ab^Ye&b}i>1o4Am2Q9+}p2xi+2@&}x?l#+-PcNovWl57|((Ujf-Af2_
zTcd1{2Mfg1RTr)*Q?C3IMF+lgaQ5@qkASAzba`@#mF(Q<KkI;|8RalyD$JL%<L9U2
z3$dDb4hZD=kr6GsJo|I=7G+1l(Bma#69`Nu<wfG<d!{OSvoZ^z{o|zr{hO2P$FOav
zc#XdSEu5o$XSIWpc#kc0#OMArd(jTd#}1wUP)^nI6J0dU((*hpIjNDyPg*H{+q`0B
zl%`PeJ;;w3^<7#*_Cpp<s@oseV)a_??(RM813tM}<U=^*U+T54Wj_Y0`}GLLozwfW
zGH_05QK>AXAki_l5FV>~Eqh(OdS$}JDN<!FQyLZZQ@XT_)9njat5O4oy{C4d2b03j
zUd+QvqwZ=7n<3oj)rg8;*BlAuSm^Cth@uS@WZ6|$QuY-#`7hHPMWFq3-;-6wq3Vm4
znyoaru)Vx|#}_Ay156n7KdFo&S+R^SO2!JchfFiDpKyk<b=^g&DwMXw*Iqo;WE?WL
zj<#%>NTX48N`Irhjq8Wafa{o8_Oe1ZBoAcT@SRw|ZbkeiglEXJzej5Ago2=&E6y|U
zg`3`wcW$rj@aU$g7D@knV0i(z0~?<J29$TfuJo$qM~}fu9_myhx0OVxpt9QWQH<C<
zOs13)m9PLxM82GZq)otT)QJ(4>?J|-kw}6oqvfaz)28YVSq(;_3!jnSwk0S^!wxW8
zd&;kN9mhmmKOSq;wPqc1+2@B)x}==BggJ9i!0jQ73MnhZy4Pei9erEc${$VL6y0q}
z#k@V2ke7{GGnTVs_=`l`t&{W8QKhD7ko4uK%@RKwzaLKb%V*X)yzGXAT35vYp=)iG
zmd<xRkBcSOMHuC>9EFHDIi@1Q?o(n9W6rQdjISS7Ax|;6wE5C$s>bikGWdmM6jDD*
z4{Oau=arhE5B;u>tmCjX;;>`#&SPm985`ZNe|5@R_^ZqB5NB*w&c6YRhEWNP6O753
zqxVIixqj+Id!V&Z9a;hDh(`Rh)f3e?G}~3E-rqyG^s#iYO28gHY%ngCrEIKh7!gpp
z{XvUt;ALY$)OJ!kCUjwPgy$Y{EiXOehkvNa?~~T7x{b?CZhV<~FL~S?)S93Ayl)7@
zW)RCI@YIh-UA~5JNg)a<L1VBYQM6USC0w6AgC6l+Hw9+_tT3)`ksAa*Nb<Z6hb<$t
z&{TtvOi#~vuXz|08KfG<Gw!@MTH#92W!GgdXj?cbau?Z=(~2v>E+cY~_tVum?y?2q
z^pmQz<$W!ORsi#9S6>>~Y*BS`?EEJ87#!!^p?%1ub{|jDwXtMx0g0{RFa(e($af@A
zo2Ka*hcPTveEDb}hA?Rr8^}_twCbN45}~v67Fvm$PRy93-SDhw;tyk=LDs;6&e(^r
zY!Yg|(80TTxlR@->2158A6iVRK}~hFt>YP1mTEI53kCa3nNF(j2Ov~X2MVstIL^S!
zd<fV15T8ftwg)KN=iVL0Y%gLKB-3eY{yA@PR8tvB@`z(npv49nH}BWiB#CZ4jcR;(
z8V8ov7{qHQchnurWiHM`M@g9pDeX6kV~Gp69F!(Vo0w4Rc?vbw_?g64ox>E<7_!!-
zt4O?7Ti&bp-O%Um>=(s1%d?Hvn8rX!sS5zG;?qJgL1sG3Y9G(og&Hm#@BeVA_TUXt
z0u3)Kl-e|d+4@A8nPf_Bs}#Oh9UAY=<=LcHe|x6Z{11E~O&+0vK`*Te&NCtf>fz2s
z`*PG*xPUff)e1xT?M{PHv&q+-oICrbY&(X*GK}FzQQiHV{aIy15J}JIcqCMTflw}`
zQxI-A@`;VAVSJWR{zeu~jDaB<{*EqxJcE4POw(Mb_O6fImMuLxq!TX9+IS|6O0hg=
zzSl^l#7`(R3dM{sv43Wg<gVF9DO&KWzFS?D{1J!E2}2>)wdVc{DH`Eh9OkmM1t}m`
zKSs1#JlU0{GZA2FK8F}*x`;?}eh&F6_+TKXj!9qSz<XDQ-q;>@1(G$)nI`GJW%X{3
zJ${=YL(k)&W%w_LnO#zm*Xw=vcQ@&FpHlm&Jcc(4_Edd9Sk=9HY26M{dgk}bM_H?$
z<{9?Vl7%UN4<=pxF{GX#jsOE(^qhTK7`50^-r6@*m_j=+<@jiR_FzWunq7*2Xj+r(
z5%U4uA`lGJ6lvYBcVY(--H4SO1_0Q3i0MLmbC30s%OVs)83vh+hG>iwTO?)X^gUaJ
z7nu^5<qs=nydF2hu5cPeM1Gn;bahtYqhDW^^z5uUqSe&w9X$XJz;)7er?~?;aJxwd
z0i>VOrr1*Cyh{l`C!e|cqaZj6b-(@Lew^;nG{-j}0lv+6^hRwJ<pR~ILh;Ke^}YFc
zJ_NNx$DFjG&X=OA2vd}<@XwYAzZ2Ayn1$3yv-LD`p78;ZTR=oU<6j221P#~Gb>p*)
zbQ$R_%w*sL3oY0RJ(etfZ*Er4QBkUd1RMZFUQtxf7;7xpN}$8lE}My(X=bCzl=X<3
z;{ccolexJg(n}9}6V0G;JX94)?hT?hjA;E6w3))PwO2F7SB2G?@N99U{^Q)LHXg=t
zj^!ixkd|)K#9St8b7PE*L|+R4$df#o>kC#DufPZ26=rL?v4s+*yH|83xXbMn-0g2p
zF6sT2SxQP(HHc=ojNJyT4ZyGvld9aR_2&ZuE^WdMM6LVYg|@SYNAthxPX}7-&SrFP
z()Sz|5M)}%{Ho*C=j;IRGdXdEUwA(?ifIUls5D&Qov1JL>dTfM<<Y=ec>b3>;?Wg=
z(afk{eHFVaO8r<|G&nkn$r}D<;hW3+n}t|bzX*<|?~y?G#E5YNzzx>hy;6vdGpGg*
zcI=d|?Q(LhK~>58gR#B5kr0Mx*7LQuJSDZsJx3YZvnHJ33LZqKjBdBb{Ud_Ar6vFl
z(xfo`MWN3C3rF}!oEM0H;!Rr_>v2ZgZ#Owd@MI|PL^NCXjEP;gB&9!NwOi3=%zIrd
zK|Q@{xu4C2ZnC;8Q`mJN^2OiGRhfJ--+Z2n3t;K>X4zinVM#{o09>S5A9EA&`fWnb
zT!hfYC4d{LyXa++yFCpKgX`US^4r;({jT@9+coSPa~pc~<<9c~b{O~D&be_++R<g(
zvd4UUk;+l>0m!!LezLh%aBDlhZ2<&(H{x=YYM#ZsxWzYhLt%GjdPhGd{p*>-w7TZ}
zA$=A{?h)Ji4Jk&WprBDx-N{Nx<yW1@ZSrxaJVYyP`1$Fa;S|86YSqRsRVp`LyKbeF
zBpGT9T@k=m{4wPc^Hz8cs|XA}b<xw=yvSCFI=NYX__;h~;vT_X52~4X3>wTXk5+?8
zw9AE#^)rxJ^?nlH18P3bGUKgER9Xj*0o5t^bc)CfUjo^V<uCi%8Y*dF52m8;4ehQZ
z2l^MHCIkRvP=}S9qt;uuGS37OsU7m$NX82U&7Fv*4XZ}+NMHgykkWbyyZB%kHy`dG
zA|KXrSl(uc>NTJ;TDuK~#C@PZ1$$x)&iTIpF~j?lWV_@yWJbS92&JD&h~8hWPl#w~
zz!1sp%C?&5nWrT~uwUKM(EsU_7!e6zXdt$v-_}`B8{w9+5juVD=yiL@8_u@uN+^Bs
zhNo&7@K(m{aQasPB%BY_1-QW#i&N`0S=-_<;XI{(eoqg1o68n&Vvbq0ah8-f71Njo
zZ1auDKMj%SydygdQ;JPuZi?jTmtL3nrdR2L2T#FW!#76PTPS#uIW=l#KtL5e-1%Le
z4k!LoASw>0e#@`YmXpD=_tQU28fPVGIwKz83eK=65sj$bmYAf|(nekQ1dJ0qdbDMS
zWSmC1aa#?8nryfGY^M?dd7t?>>E+l5Gah)pSGB;8x8>>Z6UOTalwNn{90mKQ+qe41
zMdl6FYIAGA@`J%TU5;x!6cJHn$&|}AD&w1Q?sKE`v?`@rQ8~W(P?Nn&_J~&58)(2g
zh4BKc@B?HV;SZ#;m_eP`_aDjNXuHLbOup0R7ZK{`yXzhV-mbTIHe`^Lqkt5^gWbz;
zhq5{>US$eC+@6jAwg~>ESy;Kmei4FT@;#R3+yK(lrt;ehJMc_hXInd1$bnS4!|NAL
zSI_TNQTmtkGcj6MdA3&b<jmX>k$La0O{h4wnz~D)E5tZBr4p3I?rNu!<#1YQ>)Qzy
zr-rRgfarTMnd{))vU-t5`7nvBzLpFr16y(6fnzQ25Q4_Ln=dX9x{jA`LAAN^tTHiF
zA`3dp&tG{-)DydC7U_Il?A%W<*Nh|Tdbjb(<juyDZ*d=%ta^BT&1p1RO1qpmr(3=m
zQ$li{vMSYeT~J$!R~xwWC=v)I2Z`)&r5SjgO;5YzmY-;~f9Utnuk)`HX4fF{#xlz}
zGD|Eit)5BNu(>;%ZOB6LoMEX=UiKihfvpCIO}~tDTmw5UyxXX^tsR$sS`l6906>cK
z%Z-)70Y<_P#;S*o)Eedm(63paiqDF_ZQWD?h-Qj;w$}{He&TQ2<U4jXr;;n09`+En
z2-8KAp1{vbd0`J-hKkGKGcigZm21iek`28r?HD9UrH1I_DXZkCx<RXNP${Rt$x>cB
zxJZ$K1l`@33LxPMvd;7N03))pZ5iKWYq&&=;O}qMk?B=6rS%IRXHLYeo!Jo8_ee)O
z<th3L(QPJPc>Q@-B+FKqo^q%Y_*OydP$fFf^lPR|HRnfDOq?;bouC-4ruMsi+pR~C
z#%A2;eBvm>s;n-luHJY64!~|*O9f{-^Nkplrl%)lI+<{9QTx@t>cLl0BYu?CD=Bum
zB3Txdblj9-DJyY(X<ncW7~f>Yn}Da|8E;pqy5u(i_Y+)}{>lBICoZ7CosB;0^*VNz
zpO(w2a{I*fL2*9TMep*%y{jY}uN*kZ?Yp#HbWob<aUs^>i`j<!G3nY&$FqZ_=6>c9
zC<!WWCKdNXRj2N`NkVsMFK_Hl2q@(1o4t#e6Yio50o-P#s_haV8y>$?%0=&+eMPqT
zM%D<XHST^2lT%?_j5x+sSLT`Wqsx@9hFZ#Q3I@Aso)@${jjE7Hr@pN;;~pm<EG%<!
za}s0zl_`WRdQ$a*gHRf-8tR2!v#0O28e|6KGoxmf!lOX+HhE^(4nV}xkVhu?RVGJ!
zHt-Np><I0^R2cUZmnyHvnL3g>@9iv|n^cZ(U{{1by9U<p#=G-H=;EzlRntlk%?=gP
zc7|iLJ7CWT1r=12K9<@+h%6+hQP5_o{qP4#vS{4uk8}qQ322A{C8uuuj}Mv;bq-nZ
zcXP1TMn{M6hN|p>i-kU~B2;JdWb?{ByOT89);CVlHXE>WI;REE^t|ZCCs9biyA}d`
zk?E<~PxGtHosJYF_D}gOFb^rT6Y}?3N59fv(1arb{=gdtpV;0^dV{l{Oz>qrl0d6B
z862};(YjuBsrfe+H~t*n9`=2`zaFz2<o})%d3ILX!DZWG${cTT)HQRNS7Myl(9_<d
z)i^*V8ePr%*vmB9%DBEp*l^bi8|J8a1Lk27My(X(J*wQR9$HuqlEZMU+h$h1k^ua`
zseK^fQ>167-^u=TmqYa>q*_}}orQjX?@bQbY+%Ss<Tq#+c5L#dG1BUd5?HvzSbCRa
z*2R$giOg!ea1KX%p4MPat%fUaxuy_=@IGMMr;0K$6+hV)dvL$#P*MIDWdY+&4y1f&
z`kbnPB2MJB(M$hv;xM$@ID{lNCd)~B7|*nKlOWj&jY!K|lyF&u2KuM8CN`D)aJvAS
zn&Q7Pgwnt{5bb2E9s0p?POZr}IJzuq^bl6LA`ES1rg}<Mem)RS0WX#j=Dmt9IrHk?
zuDKJHk|Z}#LuX+0bCQdv$c?P&+h7SocRiZW^$x*ck2h?@E1zYkst%Si2xfEgwMgI=
z8LwJwm)rPwE0>i@vtdhX_(AEfIo`ou*(q&3lXsAU7=J1i=z*+8_t?YW#zFKyLpofH
zo6ItuXC`*OPYvD!sO<No)rFP3h-n@|O|fI#d`X|H&Ezt9+!@8++bvomyY?i(LtkP<
zKOPcYM`My+U~pqvo^=f??z|C(#_f`hBlfU2?(lzqt&|;>{NZ;HkcemBbFs|y$nx-Y
z%zb~@ayr`BD3qv__XO{|AHuQ>JR!{aIlqlw%<}>1h(+rO*L)eHvVW6H#IaEp<i?+^
zJDFf0+2i&o#Bm5Sg4|&`dA_}Y^n?nTelgy0c5K>6?>tWl6WB{>!Q0XUG~@naP*m-2
zF~TR`Re3`_q`A+6w2;NJ$cA3aQAq5eai{%3g|L7D1_Ox)l}{`}5Oh}GwJzibALCEa
zqYh+`_p5|&3M+ic2NzaDUi|!m#ro){K})}8S1nXzqu_vohto+g?tr}R-AHZ-l};VL
zET23+jQ%Y1>gM(ebmwU3Fu+itj&yJ0uc~YfW#daa0ag9UyzdPR#0@~jsYjQFJ=hbS
zll4ZA^Wj+#AsWYjcN%gG7r4$i%Ags4axiiRTW-Dvm=^zHz9gd6M(y%wbgQ^OD?=t=
zRH)S6yLABJw+T^(F;pV($h-%EOhDFcc>`6e^ha8BlRRv}XJ*>%Ku#4Z-FpPDT}Fpg
z9*rwAnfjb2x)3V*d?H60y9+b|wI~S`U>BmjdFK>PwcUM!Z>BF<6v3QYPFc(T-eE~p
z6ztOT2Z5U*`7Yyuzx^tTrc+iS<zgr^o6i{ijJ=~pJRY(B`PC+;k0rZG`5^_lfz^V-
zT5ZY)s7p1l>W#e<OnW}(HQ8q>4n^6W#7r6F7p9jCVTKov2x-3Y0tZqMb;m{g>wT^O
z*03jJQ=k8~>bz)#2IEqly-_43q?4EeZ#pq~XSmON<ym9!19>7ad6QHLWeq#Kn^yGP
z^)o^8O0vCCMA0VpH@#kD@)ZONfe|)T5<-;{LNr$798DSIvpYuJ!HEfUrJw#xqGOE$
zNn=$(k8AL9#_cMO4~Y2yRB`X=3?sY2AP%RWU;jl?-U(BhBxceya4&e{E+7AxNG)Xg
zs!S&x31S{D724&P6y=&_B+$G;#!jHg;#9Q2*?V_sC;!z)>05+VqEsl^pY7*wb0N5a
z>sNosOc-!Xu*j_KQooa7p`+?sAm6!O&;_8vIMl?*4sRm|6VA%IOK-Z~KU1>vO9~Yq
zN9W=!d1!IoXUGWLQKtWw643szKnhZEO8u64Idl}WgDXqke!gIp?s3rgwZ|X!j_q?>
z&Pwi)KqEjD`ZO#MGQ`f&JSrH%Frta}!t<z8wNK@@49%Z)xv=rxvN=~oY35-`i;x-?
z9{X42J)`X8Qc0wmTQQC^`qyE3zXF}^yz@BN7GSJ6FGS%v8G-`9Yb)nmIiktq&dk7D
zY#U@$>FX$fNe_Lys*JB!s0{E$$V=qOj9r#)s}D#C)!>m~554Tk?ucHH=?05t6FAca
zqIJeDopmMcO?mGySQBjbrDtRwJvA<j-N|d4V5m(DP24@&aSg|zLyWOUS!lKGbSU@I
zc*$XwaFaib{UB%b(}>fPo<jMzIloXInLCn&hbp9%rnri<|9(xP=vC09lCh1YFds5b
zwvok+sFlWJ)FADL(&4X?+{RNb59)>ye+TAEYQ3~Sl?sbiv{BIZSs+~VJH>6^2TxA^
z5G^fGO0gLrfmbyE3OvDB*mS&IAq-#)a`uhoxQ=@G*-Zc)hDqr#kSAPqE6}|wE)dpY
z$HB4n>dzB-PY}I>LzhS_y_UYo$*$o;Q*COXP9jP~WyD3_^Ig`#`zpy`EM|S%L0;Qz
z0$1VPHl*G2)13shBr-cRBJHQVFjPxkn@p|FxE?W70}OmT3sfX5(jfGuqY0xA(fL03
z|75%Xv77~=UlF;w{j0M)+z*XXDHe}hDh@~1zU!d85C$&BTn5~Rar(~iVEm{OHtt;B
zk(cq>N6kH|RgF~k(^5s9L8O|-h5}nbWd<~KwG}^_6~?MQ-JM>^C9`^Tp)-+|hHBki
zI!Z8_8In0I4p5I6f)o}5d4hVM%QvJaWIE7?%Ef7ESS+(d?>br2LLbr?r^l*Hv!Vy!
zj$_ao$#uqIuZ~xzyI3*DiSG}dEbCIU$un9pQEVr8!`m`1<-RTUF%esA$0;#F7jSIX
zN`akMUx*ao*WC064}mZb1TVL-hb;9`!r%J(RR9jp81}a-<N)FI<EA%wNq^fQ=+Foy
z#u?D@h-D4GGe$aWRZko5yd5|q;Jn)O>cyFTEtWOQNy71^<&sz`K}k>Jowr?uB$-Wq
z;7lXnlk!dONc3g}sLa{FI_E|B^8NG9>qzr+Wq<NNK%&>;1GcT<@%c~zJFgaAjsMNI
z6ipO(4_2v{Ru^9l>f)dxj@Tb$t)S3&r*m;gl+a%~e7AlpkP%O)U;3bxI{@Dc?8Lu6
zqVgppORy(bJ`uO;zuG<;+dtn`LkXAjy~#KvB^&jksYR+lKbQYE2b9_tDac+QTd403
zKfRFy@yi5GLVtj1M#uJXc2dva3+k;c9mf}?6den|7B^_TCR|b(U@);sNt~tB-Tmrw
z|M9o1;Ar-J$n{F_89g}VG!Cl|)U8D+$v`q|LL!s)9(A$%v@o#@VpwqPrFkw#JZ2X?
zR=R(@d&aAKqp@4b0q+kB!q$f|_un}K9i#(D!x*kuCL@PgI?sLMWc0Ww9tZ=mMeZo=
zI4H%n@KgMMu0&T}`c##u*R__Eulw6~9hY?X=KxcTU9Hq^--)B>YlcHO4W1ma{jrJG
zxhKD1f{?LR)!^CT^0E%Rk|i#mg*lpU=&Fn7;MRIuFOiVS)pPhs%B8js3yo$Hl*@4-
z3GPw~dyI^9>onfs%dPsi06Y|3P2JbHH|bkHb&GTSFmsWpvcj=Om0-3LJq^(Zu=Qw0
zVtxvtpPGDkj`WIQ8^5VniG7BP<gL$IFMe7bwA~y~1z_~3bc<vl&rh#rcZ)SFRw(0v
z1Q8t&xe`-f_`e}GyN~O_5C6$jr~H3>)*KiLM%gd_C{pRdGyAYayD|ptDQ)x2=>V1_
z4gNEXst*7<Wf60CkCF6}&DWb^8(@;Ev-q>}9-D9nbv2x9eO3q>o=^%cQMBSG#${`S
zA@sw&z-*g(t*cK_*_KTIND1I<SN2Z)3PeOHoI+1DI=?Be26NYzXGwcqbN8<bFe+!k
z=@{oVO}nd0Z#1rs*ZaTnJp7qdhSoz22MVGEwlm^!Sb=cuvo70>>2ulyJ7EHgS%ArB
z`mhq(HKf;ekktg>83aW7%V~Sr&3l0h)JRVlI&!G0cD%EY(*|ozR!>`U^i^v9rIQPx
z;;TgZT6q*)?JOExfmh0@nk25TncDniJvn9qxRALZ_PY82M{LhXi?t~JQpx9Doz0V*
zfuwh4%w<oPQBSD&E(0JrmRVKSXiX~nV9>RT7Q*Z&w_};-V(#989d2Gf)if>j<+%S7
z$vaxfQLUgJXeUZCIVc!_4)`2J)}ijsYsyvY^{>_mf+)}F9Wa6foHKU?F(3tQ20nJH
z0MJl7A8XQ|BW_zgz{sGawK!lC1>3M>V`}`ouD;TO1sl7KT!N?5m(Yr4?$pKBGHYh=
zC^Y74nV-MXy(P7FwJ9_(rm_Hgp?tbHfXW6-?}=m|j44*A8}640q`6n3fBxV*fS!d7
z=dCIV)g$wIged=Uh}WSkNQ@V#X_4Q(?hn8bookvDBr$+n4G2aeg&4a4WS7K|x(K4E
z^^6?C+zr~cI2l#OuRmbbAU9C5lD$!Lw$xN&$!5Iu7)+NARyi^Dg9VI4XZTqk$Uew_
z?aGriQ)3#wkT2z~2B6rKN07@Ubst83y0&+boC&%fUfspS$Pk0z-Fy#UVzEE&0o*&U
z3;lua-L1i^?BT=a-^C?qk6ye@pYF;d^LGes<b$6W=?d%oeMk6E@LqNRgsuO;o5sp5
zB{GDkElwfb*uU!zNOdH+umhRAf=vz1AN49Kru0#C;wOyBokEcwP`LJKQR>d1Y*PQl
zNU~LVqjY#V&fB4X`6kW;AZwoDN^?-YO)>HLXJ^vh8T9gkd*KpL6Ud!DMZ%te{1{-r
zTG%vh61&RAFuP_Le^+92of^3<G?ik+<Rvy`3u((|a6cJR*AZRuMV7<hPXe<|19aQ+
zI!`;LYUOpEJ(Sts0OM$=Y~gNnEyN+ns)1muet)r{LcjNJmiZb6y_rPB6AHnk8E8|1
z+cCfa0t=hDCP{W1c};sO*%0glLB(B#&)rbp*{Y2aEz@@Sh)CS4pEC`?CeVT`Ah|yy
zeB498XTpDhtzpB2g?|Zdgc<79YY}dxkT;7Ra$iu|3~H6bTpO7h4ev;I>~cCY)0w~O
zMe+<*E@NvJs%LXY*pw_xuy1cRkLs=UQ#eKn{K^T~I~{oTtri)1?MF9?jj8vx5Y4|U
z2mD$9uO4t$4uOJ;t)|^_))A)glcru`-Go9j8dlzeY)~N@XTWW5VDj5S7kh^lKh)=i
z&2|{;VN6*TmzYmxg%*5sYBy-K3f@^iGV;i+!c`sTj6$SJq(8Tw46@2kJ~3kWljutL
zez^{BmuY4#`)<#?zAaZWWWn3yDzFpSprH+4JagTy0-o%Nr*?17G)D1OlneDy&29L!
z{BV2`C?Ukz^oy3<gYqDv%w!tEg(8LcMmw6}#Q4R*5G?9OBJyQ<KWv;13p=)Tf#fNM
ztlidu%wW@rw?NXnV(=$og5|QG4IBM~d>)Ipr(?@|YQH_H5cPb+&bDIwSS5%SR@1o3
zR9-)!(9o~m|GOc?L1M4@bzNRl)gxCmiZ?4eV`AtT6nh7I{Y3$08FpLE_9gO2q7Gn=
zUd#pMIRR*!?q)%M^g|h!4<~2*BcsyVpN;Q}lYmv8?fl6NaS3ZZv>;KV+NKvI%^g9x
zSh;7M%UDj*nWuon!qhfY4frE7a%>YymA{o96GVx9W%J6T8-sSDrQaWk@Hkr;%VEpg
z`BuMbj*|5$J+G^#z!2H;UIgUq+#EOF9K2Df-3CsLLyqP>WBvMg9!e9ME$>Gl)T4Gh
z^92qfQ*PZdv>!q?`_Ww8b#aMl#+_|SubKXM_k1k%eOv0{yV1LpZL=)h?cZPoevcM|
z(EbW%jIyj(d^S;B2x2IHym+&BfcW?i@{6_qWr<4A6pr;H_1$CjeG&K-#iW1iZxyf%
zU~CQ6p2S~Czk6#xFEvh`#hzr!Vohfs|H$Nbo|LMRu#9GwTeWvDZwJi#V>wQ>I!9#(
z(u^G7Ax>_%hpXz-Pp^4uIfn_EeWWRsXv@=Lf%iNyF4kXTC`Yy*L>sWOW9`?clwx%X
zC@l!<`QSKWPar+8Z9pS$9fvd~V5hm&?AI)S%mCB)y2Qk_1<88zbj^A-LnF)ahh~Bi
zN)X(gVD6dvJ>7f^dCD;k<cQIX;bQ2$jyC{<$)k6&Bu?L3qqJTBX1&3p#m*FaTK|3J
z*)H43BV!EP0m8f(&V!{CM1>98y2&Levq&CgMw-T7;ZN{$5PQRL;gDkWW#N;$0g%`{
zL;0>4(EpV)ko4r|tn6ybnCMOEGv<v<yb5+hoF9%%<ahNB`lB-ihw|J(xN`n#gl)_5
zZ9RY3%d)zH*gC2dI(FDApqzsg;ZPx%@PwoR)rq&C_^PQm-2ef#bW)EXJtCbI569uK
z`mkpsSas-Cui_H=!I>0MwaQKEw}^#NB@2>pj^k3;kO?A0oRbkQ<dzw`S3cn@w|6OR
z+g&U_<Jf*j&$Hg2895pE8;wgAoPr{(ZD!#7RDNaPMS%N;SqEJx$xDMY4Fh*bx1)mU
z3&cVnOvz7ZWwe|D0J!wlSLK^sFltw6HL0JT*bx;iKl)QdG4R*L>~nX(b-R%LIB1Pz
zG%;Dbu3Lpe7n<JLlbP)uJ5+$@Kg&Fz$^9!(g7$qE9(vKFBU3(s;h}{T|4#^%$nO@7
z>x}5&52%U$R1UzJMOcYVlDWxeT@>3@**izs#(a@3$b8&HR<4!yM8Xw8plCM=-1FFw
zZAmR3kzLSoVBn=Yp-LYb9t~VVl@f957F*JRW!m=wBMJQ==A)q&>vY*2U@dy-Iy}O~
z0#MoM8{}u+W+lj=uP>og@VAUsV<1<w@Lu`%8;=v|x3@;=nSF@U7Jsbu=$Tw`{$^M8
zR(PC}Jz_kn9^gy@6=kuPyEs0V0nk^+!yJ~HR7)$U=nWn{V+fftOT}M!(QubX<&jd?
zz~$Th&COJ6JU&2{Rj0wTp$wEg{EBiA$mEgOOneFe9)p|Q0`P!mx4s^2E7<;ERxbm%
z8c&e4aQh?4)N<cLo53s3!?RK8;Uwj}2r*395+KD_^Cz|ds9)Kn7ljN-Y?y!g6mcR!
zuc_OX(a1J#52(^GGcSmv#bppj;^yI^>nSrw^N5*Qx9t&>q-6{ItijfVk0(njZp?JT
zzy;M9c97S59X1}f<*8KMFusVMjMn|uZq{Ma^40Zi@q-5vfH+hVp9A29SNl78K!A!g
zzB1Uks4eob)P$^7bo5*7jsU1U&lSS9y)LjSmVFEwWXWJ6+)_^_70Lv04CSum4Ic{f
z=B@Oer7zzyvgMs&03*~u$X_a>6W`-PPqDMb`g&lgU$SQ2=S9XZ&PH4HZs_vefNrWy
z_zME>qb}y`xR{A78z%8c0Zv(kY$5{u7^7A8xSqDfN6zo(w7D55O(7<0$Hx+&j`<s|
zOQbnAake~1BR#yp8Av)l#i#z6m`bfePtNR?%dKbz(#~$A*FR-YnAkWxO(wH$1<Fot
z70)vV$k0FS{pup(pDtKF0}_5+ZOUt}zVA_LP27>azpd6$0!rifQ=`snQxLSX2Y7}<
zjlM|ppZ=yDsFIGRu;hi@obt}<h=CQpH;u$(iE%oVo$bI5XvFbFx4!ok2n=ULFOmiD
zv^Wt;sdPv@V&+|nIu(8h<#whzO!io(ZSmhD_m+l3ynh#cWn^g^2ERzR%nfGzg$v98
z3ngX$i?+l`k&5xmOeNOOkGXk6E232mi9KVzD}&|j_t$d47nE4=sR6~+q5GRd+Z2H}
zs$xLx+8OWrXS5&CS48Ec)x_>R$}IF;-ZMAW;qM)+tJt~1?Pb#Ogz5dJIWl5|*(Qgb
z<E(-vUvfQ5&gVukON9`3Pw~^|82=8}8m2O590Cozdjx_qr4U@_RBYO=w(QjA1&?7p
zo569tUIK-(L(r>-c>2!wpy4f=fe@fF%JX&cUzI{YeGJ@xV2yO}*Z7xb2gg#MxgzK%
z*G;SN2@CG_7qHj9xPu`q87i`&1ZL*XhAi~|qWy4Dx6kNUomspdBwoIHNz?43P~CIZ
zPq-s&UKCnaqF*1>?i|=@+0=Jio>fRP7*p`^u@29#3s1{&<kiTxcV7k>Tg;uPE~Yk<
zbjvrP2g~j4&J}0m{&k;8ox$tqxGZn<I40j+IRIco3P8Hxt?{6l)HjOiKg;MOjEk~|
zCTkg^cYCH<O6<Kpy%VGfc&!#=O0<<gAC^tvAdGvV1txt@114V%0wN%C<Fr(hlriBJ
zN0v-6cnm8Decq1OT*-9p02Pb$gV>H5ajvqr8&cLAH*^{xHLqRNuC>?5H#Jr_$f9Si
zMKkOk#UIZJ>tAQV(d_yJBtOW~`Kc|vttD8#*N-}Wgv0F34jQ-XH<R@oz~ucqv-<ly
zHo5SaVM_TtQJ$H&?wsdRNM;8=^DDf?EQD;ng2ME86A0ZVTnGnW`aJ}q6XZ$3l7U*Q
z?rN{VLIwUMC&*&SlJBd;A0Wd}#9jy=IBv~8)!zTma6+#o&^ajSLm_W>Grb}##As-b
z0Olf8|5$z6YjXFgDC97ZpbfLFf#MYD^ln=iVCj^dajiA>irbf!UL6Fh-Cu2=EfShF
zf7Nr_clYLV7(M-zC8S{V>msoTxkgm^()*-zTi~<QrGZ_Az<>h>b5nGZ=wB=#N%TD+
z7(SQDAi*{)Z#~z!!B0DTvLMDc1~I6YT<v{1P{zkra(0cL;kFhA(Z~W2iEeChF0D95
z5WX9BpuO2W_i*EYm)o!05NRUWIBW+|!%8PMAc(jx?18M^b={;Vg9!stxNv?Q_rI_u
zK<hiNc?L-?T5<s#)qL7UYZ>jX;!(pib?9TMms*f|b>Q9Ywiy2d@IxLdY0otO1F!%l
zrv?cwZkQlEpfIcDkv>p*vb)cs3-p3HV&Jh4X>g<ncn)I65xcDvK@I<;pmP7=U_rS(
z^*XZXPn_}lFg4S`L@mCnIe|H8ec0AAu-%89O&rij2B)Yg3H1KiU*s=hX{OwqWB$k`
z4CK_XpiCr4S=?YK(*weV_Jl7cdZ_VF!i<DZCI<W)2tak8uIgesWE;p!YA1CjTL#Aw
z>N^DI;4-4iaV;OF^uUvVg`U)U*BoX3KodftN{qAjqT|%G`x$?&(d)#g<Fb9*tnrJ`
zhRjC5N#941U2u%({Ln@)jS=vQV*7y1_4b_%B?MdS)~<Q*gY1NB=6cr~<TfUEW!}_T
z$B><!E)Qf#?SwcE22Vh$CkdG;8tO0w|HG*-GBy{9zSBEPf_|P#;w?I6qDvc^#iJab
zz@&equoj{_9w|5H6Jwaq!@_jfYex8V=VR7@>ZRr1X3o=BsL!68ZX06@MODYrE4_Pz
zD6i%qkd@3&7*LR;GYEIz&V;Dn@0p9SE^f8|NZ9|wm$MQq$AO`$FtGU~@-4Kd@`khB
zz_~yQF2Kz1cC^xyJfXV!oWyOuSE3iwbt0moB#p#&jr9VkP?);3ERW!93CuH%gL8vJ
zm$r3G_UVJgplQNvO-ip1^5d1NIgWa~nV1_BLiqQY@D_mbZ5v*{ctt+&yA4u8d_u*9
zUu<0~lozD^1=I4)%-z%o;V+X~=&ykMvs|-^=wlhL5U9t=-*S6pKKxbNZL!Y({Vf2Y
zux3Ns9TXyB&vT;i!UO5fmA2DOAP(9$y83g$aMqa!j+l&MZUO@eyGex=Z_s!OQ_f<-
zf&o=O*UvChJ2UU|MH-UyUFOpg^h*C=(FV#L=vb9an>>xGR4g|HXV<}5!Om48W7|q>
zTS5^+H&@%)w&~wpKNr=ppc@dadVNPQPm7b!RCQwgJnj7#nGFG{<<F@<Q2qpRe+UiU
z{$3q@HcV1#pL>(kZU~Ta<}AKM4YDYvxdf7UvT0~B7vm+!Y@G-LjIB=;eZoI-^?(#s
z5k@lWiwK6kJQ9q9+LpIWz0ZR;mjoyCR-|7vxisI^dFT1&Q3TJ#%AFZK*<hd?dZz(a
z``#s}(p2SA$n0~^4N&x~od%z`N45{^<!JqW2d_bc_Uoh+QvA>;+@=ZBdF8dqKtAgg
zJ;VZ_N48R%gVR!XwV7jiU^cQTpi+|(YR1q@&|nUCmNfdnOp*WL2ZPz>mXL1QG(Kl6
zP++%&ofWvPq{Gx=yj#|wv;b}3O?e-I`y1~(*&`D%MY@Sfb8@C7Tj;TLWqxPJHtom@
zlG6oFN>wN2LT{JRZh>wz?STO36hNf}GY~lWZnZ8k=*1|s=1x+GOQ?Xt6n?{bOr)fE
zxdY|147L7qdR#{W#$XmgwqPxG<8TaO9^=LD`BBcutOlf*vR<&9T&-qAY~!ZNlS|I2
zmL6kB&nph_>;BDT!=EPna?}UmZ7L+EA$<wd@WLuAZqgpt0Jqi@2)<M}OifIBQcR|9
z2=>so5K!_u01cD9Ha3dA*O-W9-j0E{<e@rc^L8r28QrGkXr1wS4ZX=yGk{YPEWMi?
z4cdJ<OH@Lfl=*`}L^vtm0v3<YFR!bTCh`Ax`^u;+w=G;iKtj3{Bt$^ELqJIZ>5wi7
zK^mnyL_k17y1To(QyPJfE~QhYL*TBLz4h#U&b@!{7!Jla@a20~&NbJZ&wL){#>4~m
z#1+kc?F_a8Y<sNxk4I7yZRR&v*s{KYBp2o0L&j4W$`Oax=Yt>*CC3fA*1FR0WsfcN
zWP~A`q5JF2Kb{8q-qAB1kUqsU6h1;aF+%2b`^tNN!4nj7*PzaRl(v5?Va<rDBGb0U
zvjKvIC-UYoU(s`*gtBIW^BSv)qSNKwC;Bs)Qyh$ioQ(o*<W9P&xrVC}27OxU^0jiO
zw|#7zmo(AGemLHlG=36?y?0hB_|<-Bl;yh7wBK3K@rUM;$I<Pb+P%&(n!2ctPEY1H
zjKeMlH4&XyJ*jsr41FrT{#3FPrM!Wm0eHI5=<P!EUlbSbW=#tS=GATP8aOEkQ4K!8
zt|&kCp)4looc7y$*Jy|B*^B;WL#n%fbBO=@+_Kn*`r|1>tOCS_{g!iBi3@Fgr0^%{
z7j|Vsek=iOMzqG8qW3!7F>7`eob<c$mnR4N?*_s%PaUBVVnwDa0IC4RJF6i0Qy?67
zfHZ0K?TI1V9tcyErKh3zv5f5au&HXdlvoJvly>Z(x1`ki?@G4A0fRlK7a>{4Udxrk
zW?1~%Kv$=nF=RQ}P~m}{b5>HBdB3#%sZ0i0@gjskBeSm{m3|z}gKze-_Sy;4Z!GD|
z=$BC>!`4Yew7$JD+`DI1%}d3yb9vS<Q98Ro{ajAmjy^8y%+%DS_8o@%v=p7D7r6Cl
z`#OT4NWMsEW0B*IQr%VUa)<fn;}Oey+hn(1rm^aE|NMGHL33Xqt-bGL85GE27D#I6
zkasFH#J)xQCN(^_e8nYO*g6Yevv|4ZxDb8a^7$a3Q#I{Wm--3EF#q>PC7eiC<XPee
z!7h8)^thVKb=IdIJye9Aawx*#=|TQ#Ce&BG@*bnh7x?^Gv)M8;X}f<#_>v^xcuL;z
zk(c{+<Hs8xC(9J1=}CrhhQbkcDmZ;Q(h*n0ZI5vZaRzRH1tn!p<@r%6+|O^_xh*Vr
zN>2mq6-+wVKYoii;+AZGtgDM#hI!tOu4eK{<YxyT2{>k=i7F+?oI6XpnU}K@CX&l<
zQIeEqUx7kymRTPaw(J$9u>G=5v9g%$QiIqqRz~oP=r)$Z2K{Jr#&TK{4a8*a{vWO%
z4IB@@NJ!isRR~mQcBhW9)xV@fe;oLMg^hv#LXkqWjX2nswJ^b^636$x(G1a^J^lgX
zJKyMMs%avfdX#ANPu$Y1aQE<KOW&Qpv&Rt|j$yqb3}V#oFqC}0D#)OI+_E?>iJQmT
zhUdedHX7&~RT`ksn}vzIa@PdCC=r6|j`zPeG!U6}1I3Mp44baF1wNK{yt)`$@91V?
zj8elUb9aN?_BypcbU0~Fg5%HN76t9~tHsa&LDr??4DHuAboCmvc{kKAcG%wP-O{?}
zT6B#g??F9NfUt$pEI&Z(WqCE9mBq7l`QM#Z5XOkpBlsBn&0jL}vl&SrftogsRr$7h
z`iGPw-(}HB1IEELs6yt;D-{o3Mz*rMD?hU41Zn!&rL;GgKVEPaXPUkvz|DdY>E7*d
zetdb!n8E5N)6VXstk;G??J(HRUT7$ZKF@J!!xoCem?9~n!)C7V_|!MZ-U;iic5Y|)
zkdF%cO57^%^Gi1LOi#?1sF_7S>T(lYO*PY$lJ~5^%BQ5Zs<W)#_xl>ZXL+&ralbs>
zMZMi&u*lX(Gn#$;K$L~>RN=1IXG#E?)zWYhjb3hNfAx#{-is;u!pkyOiYLu++hDxH
zKzW+%%hDf)0T|SWx<5`H?V`^|-(ziCeG>Hi%b<v)5tOV*^+9<r_}nho*<6cGU8J4v
zS>^`!Ng$O5y}M(7tS@qq9^r$vk*<~ceSL>r9mn~qf$$!5O&J5p-1bb>vR9AQ-70D?
zY~&*Q6}D9D46Op#(5;<xBVkkh94}f{9+YC0VELe0VNO@tD#znR5sfK~;@|6I_kTm%
zoWOM9mNgI+tl=~-aE)ig_mODO#<HT;#$Q!>Nb6Y;^UwaY)w$~s!?R=Ga+z(PFlFZZ
zv_FcR(##+Cyl)KX!sNg=R2(!w?Yeki;c61RmdJ*>oMmW4Z`olGd|~m}#5d}3!E2!`
z+m*VOFT+W(D*V4Z!WZGuJ;ImTNsWWo?+tlHv#8s4c)#^*X(_y`lj3*d>7}Bh=+$-X
z_FsE<+kYaz`!dgz8?CUu8QuOAVSMM<ZuU%s`rXw9OL=3Y-(E)Jp+B3O>6}OwowA*z
z?z^ndHB<)oz9CJhxs|f}_xd}Xq3XOg)Ob}lTwrQDmBQGk;PkOYZf#RRgsAULi<U$k
zis5^SVB%GW$)jF9ya3PkWBq``=}G%=dQ2K&4ccaN#tRYihqDU5D(J6Q;D7kO#+;~A
z@ZRr#r4h1v|G7>7fYQs1z8{S}NYnD^&&nU+a(qf;`jyO@6l6QOrq0;z;YPQrBX0jq
z)!o}Pw{^b(Pf9gLIhlaA3xZ*&;4Cy`kuD$%S57sp67*JnP!VjrzB|&UMOA<sz4?rn
zAh4|WpsvII;lv`*H?w<p-n3uh(8SwboW3Gw-e?}ZAJlMsa?#Thowd=OYmxwEcI+O*
zuPGc7=;&TKZ}u-ErK;NL`-x{iJqNYhvV`bTMeE~l7x6yO)i?8-4Twd%(ym>dx)*Iq
zr=<&<X1o3*(5c1=lKw{is8P>vH&~Vq-?gq|Yeg$rJ4r)uo74+6igm78)Nv@Yx|1=Z
zY&+b}Eo)alk$E-lpqe1iiaLnJ@_?hMJY?Kf(B{cxjjkBa{Woa~lrqcW{356XI>!<4
zF~U)--0X?zn7$oqc4u<zcl{jY9#cGXPeXSfKp<(j-ND^(6vp9$U#pele3zBQw&|V9
zs9LFxyOm~RCM&jDF+obO#~!%rOwvSrTX@@^8gNZQR!_q*_##EX-R0E}nuhw4&wgnP
zPD!0lXy$l3<m;IIpH(tL*a{sX3Yit*m;fHLtoJO!mA;6?;cM6^1QJ7i{Xr3xpPbaP
z!DY+Ckukun{gk1w#-s4f#+A6=#2cvxTjr_>I<pM!9zVN?jEduBWFu70kG9oR$H`0S
zXEj#+ri9dz1MQ*)C7Ff}se{9&o@(qm8w{75ZJVf`z2SQ>6x46J!Q6K|g79J6_VKC6
zetzt%O>AS}xj~@S4q2v!Oe{Bjq!bxu8KRm#Y1D+F6mw6i$@Mp)mRvihu@j!{x@Csk
z5dQ>&K;fPd1UbT^H(t0{Wb3@ozR?TrCGd4j{g7hK+EAs;gaVV4Bi@K^(n;I81q3?l
zxa%_=k22aoH{Y-jT}{WfjP^>9yRQ)1UQh){%9L)Mzl*}yhI{AAzh~LpnVqEX`EYNO
z)x2&oM5@2`y<a#izW;4)gkN|LJ9))>0xs2NoJm4ytB=B;8zUyl1pLq#`y@B2Kc*9F
z(!c$JQIQz&p4LvwPt6Ilx(k`z@Z_<Pp@>qx!U8~-y?OCs@*WjRioC3)(8{Ypmp#2S
zj4o%S(`e7WB^B`n<R_Y?#~1HZh~6F$UtAd6OTN7PTCN@Q+t*^ISfVL7X$b6BW1!C7
z=+=F&Io&Pylm}I?od^d++3U-Q6xdVz4jNQDXeIL~OfjM<7iD9el_lTXLSf{`NNkQ@
z`>PqSZ9P}1?Lc{fy;%O$;c|4r)5pgK0?zv@1F$zqB6;1E_sK|#FB<2YM-0}hYtA_w
zEF~-b(^U#O_kL<_o8>H@HlGFf{MbXu*|V9B?(oppKd2z(&?Cf*PaEk(Ruk=tt2dO=
zaFVMbBU$M^xL6vBIP9(Pa`x|7#~K#<7PoEhEzf6NHJflW_tkmG3+06@Z(F^5dIxLH
zJHf{RDw8$G>@B>uhW^vhXC7)K;t#|obx>~-wL(NVAL^{R+C`KwM%5XML1`Y!gl;si
zuwnU(SwR$c&)z-lBZIxlEFUv%iZ6qp*vB&D@wX337__-}y5F-bz0#DEN+&euU@i$Z
zvCT`PL$9sG_1S0$=!xvlq&Gkf-YR4DUw=Tx6UmBaTic?t&DMo#u3J!SS@3w}tuFq&
zy0~@A9;;#bti+U1?A*w%ZHi=1_ul*)G&(hj9ggT;KEh}`{^|uw+A=^P(r}Me&nxLs
z;#ovesuxF({oRPIW1fqOxM5Pp;^y>zYs9hUGauYL*r!%QB$kzm(@zmw-n3Ggr?aP&
zXveq{YZ5}iH|7UIey|ajHY_CuYGbgf4y()ElU%IuTPlVH8VX%&iO;7E4WpY_+axup
z?Jb19V63*@!eVqiL(Re84-KGuS|(F+#TKlKIbUVSvMp)(p=;Uih?+FP$d$3VEH{vO
zxOAj`2u_5Hu#HhhQ!#t`mfM@d+g0rXhDkE}&8Ze9U+#p;OJNg&#KFB&)E2BK8SRTz
zrA*F+87m7QbfDUsC4jslf<%H9$J)#pYCEYZahhl{ZtGtcE-LL{5@#gE=cz|Rboxej
zCD#8F{=8;Ma|qr|uc)x=Il5>Zt!biT(`R>xnvIWZXxI4;M?@`gqe`EC4UzyHS^fAy
z;10`ppv<g)88x9;-a$k%7vtXMvc#}_B@hkb8u(W1RQ;xh?ootVWX*FNt<Mmm8NbQj
zTJG~ds@sKm7`%NOVED^0VH0gDg{R+U=e+yddHJ|XCVSC`Ai47Tc4|V`&4c|S+bf4?
zNoFE??_wDXZ0f{@$tVoL6ql{M@_`}Ab=1%z$IJJ4avPt{h<MJY^d6}0ztNp?8BTI=
z!%5~ZodSi_qw)7-rw9^=rc2%<`j&Z1x5!v%5YFEvu>7m16xiL!=Ngt;UoK0ZRz9dM
zBCy;Zx4k+W_N-<z*1vv^Uk=E`4yR;BGvAOP8AMN#-JU4JrJvPT88C{Yq2fc$VR&%y
z`Z1BECTY}h7qW@L7;&A89!VdBXSSl(>Tw|m61)T@Q`)N4(HOv%*-d-hR&6#5H-gzL
zxP&|~=Eq;#|JcRuzZzIcrrDZbR=o-fr7B=JvIgO<yd2@URTdgkDXg(uzN*d!Toq1%
zo$X8@S{kR0XR}Bc#I!bj7juB^GD?t1GZ}+!-}aH^l8<o2)Y82%bO?$Q@F9P&dWSwd
zef#;w%|#wKZGam_3Qy`C`K@ior1$b{NRJZH-Wl#So#@)mnC6`MKVp9Ax?7bUziFll
z7!wkbI8+~up?*rt4Br78#z;)T4x;uc-~Ec5z1r%_Jc#fy9G9M6Jvc%H{>?1<d}~Sl
zRz7=-MGKR<@0(dn-#nR7#+k?zPE#c(^=ZBry@}3iaYnQEQ0U52)!zr=&b@s-Y#`|h
ziqv)!%Zm04O{P2B1zip|vblq(1n=DO6`{;iFLYJkdn9-!S=(EjqY+4VQ_)5W)$Sd~
z$}uv+rNlaRfbiOVLdz|_;kF+dtsLCk!AeaS4EwgPe=M~bElN)=PFa7LC3015a0kqY
zMOuYhF9NMpYL(UhPU_D1v_hn09BS#gUxxx_ze7<#1+sw2z`+Q~;n<l6y?hYalf?Yp
zvO<HSd;AU-_?S}E>#Ih-om8;rk6~f_+`CzyqrOHgi*|+YQhhlCY17EZz65=bSgzCy
zd*R|~^l234yqtfA)}`Vm5LVDG`;pj^(qzIFsGAI~iV^vyhO_3rh2R5*YIzA)U5pu>
zMo^7hRKE45$2YdOO<2D01jQ`HM{HjO=E0V>Jho&Mlk~wYGbAaq44{stEr=HW_@-EI
zA_d@4%gdcDKhR}K0)ubcUFdX21gdh<|3bE1d*a93<96bd+}UpW`CESl6@EC83de10
zCpWU~L!z2B1>)5srSWylUU<WkK|roy9_z|0i4q#%jPdhV0$^#0z?7aZF6_^9s{Fl<
zW+H?v;zOmL0`!wDJ$*U^7`ex5lZlGgmipnR|2yyWj|@+kGi$$}8NHS5{HD=MI$9O=
zSeucnnf=|jc>DL{UASG3qAhOMh!|`uECKE(T_1i5s~$jqBofM)MjHr5pnz$4X4R7&
z<3+AA+jju0cV`MMC~2Ny1O;<hq7a1KXFIE;azwB^baV*BhKH6WP}c!Sv_wKnoBB4X
zodS(C$*N*u9I`Kw>>P60zKu?<prirIj--)R9qGw2^a2YCuZxuJ_NLEXNW7@AYwtZS
z=I%wd=EuSq(s10x=umQA>W}z1a7=mG$5WYUv+P}(vE!+Y>CU_b#f&G9sj^ECX@9g3
zAP@g2{qdAqA<;1T!PQS%F*mG^_o}b<#6B6)v$B&!yiHn>`+D-RUV#;hcLlp;CYtFx
zD4dv>*?E*s45$Z5O*#`E%7eY@?~Ucb;FgCUD5AmG{FTBYUb^C?r-eqK)yF`H1rtI*
zfwOgK^$Z}3*g!~6w@deAbF4>CML%+0O>f@2@k6(+_`HWe!0o~O4TGwOiUH&i{=QVE
zQV*`5SjEK{Wj~rR41MZCJFp|s_DnIKYJk9HzxR<ukiJHihbJ0_8aAQK{3=8Jt9xQ)
z-rj;R?4G4)4LV+eK8b;-=_HP0NkQkYLzFU6xIxrIHWn5Asn5Tf4gX3i+y0=m%}`x-
zjPr1is}N>}Ur-TmN&Eg4@5y@ksuMf5KXbiBN4d0&;ljK0zKT{O0pk&RLisp5-AI*}
zUsmH^;WU!hu6MK{&#>8TC|txSV<lq{fszzaC#ob4Oa7CGx}0fI$y0CHabLf^C10?(
zsJ5~`L~LI8#gLv0`QhElLMo@2rM~a`_usfzEJ&2quh!qsNVohec0xIVg!p{CGnfA$
z$5_#Z6qix9R{Zt7G{>G=c}5G`U>J&3Vzcu0)j-*-MMZG1Pf>QtULx%S(=sF4>#x6>
z_og!C3k%dq8H_41LVLD``UR(D4+wSX<1X-cMjUX%Z+)eMBRN+H{(&&$n5rUGY|fC3
zq+eJ(OXMK{FeqQMwI>uu6}l6DB;|?!dL^UnZ`N!MAt|f+)NC&mvLW6=j#r<<%;Kki
zls_RCMd!FB8x_8;GFwV1=vW)$#_-f9sti$6KZSP)OZw}y)^LSfpw)Br?cY5!MvQ&I
zDeL4+g^3x25l9e?r;DxlrYz)mZ=Qem$A&;8`kemTBBL$;N+RXsx4PO1yw(<)u^W<C
zH8?fTi5jW^NgFi{pJaiLJ1s;e*HBfPYLboG_(rr-2Or%b>r`+qq)bR9<9!EGjaj6$
z?o7npQL94Y(t>P{bAMmVE&^X>$*CSa1goY_nQX}&qpEQG`4o;Z7->Pz?lOmi%nUjU
z%jZ&kI6oqb^CX(>!io^tR$sBFOn&T7US~Pj$}t|Ny+3+?QP)yC3ChN6-l&DzpZkuE
z)Bb$Qlu!D;?K0)!t_C=TGA3Dz=GCj~dfvLqzBqRd`5F@HFG3w;dukChuV!i)7*%4&
zlovYnll3_E?3MG{IqpPCmeuuXx`3leMhYG8{ldcwZ06-}yUD&9g{@zIUY+AF%FeS+
zFJf!aIIQ+QN@1K8<8H1DnCZtHK?a^&G!BKcDB}S+i+Yr<h314wQf(Sy?Ot;MHOp8h
zWwn$->>Ci_IsQRHWr~FW>ha51wym3vqV1`lc^mABX15CC<Vm~{YyuXjDCqSmf`)lh
zXa2tSvo;E3Y;;sl+6gyuGlFu?453fFF`W{fc+2mS#2)({H-hK6#c_NaiOnBcrPeC)
z3+ulb_Gte8rjrSKHFCRR$+4_MvbRT@bocxW)Wm2O@Lo!MO>23TN3+He)IO8%5%GQ#
z;L=LVzI762AleGz9_W}Vq%Cj*MB}8xlCO86rh!!NtKbHK@Ao_7Cm+7iTYBdjef&d*
z!m4JpRrl$-#k!%^*_`AzKN6ndl0ser?WYitqotn*a>-WRde#PC`Z>Ng<|9DH+QIS$
zV?H8X>lP;|B?8i;u<Q-O#!a(i$S|!)rOwq*PHw5h{RV3OWv_02a>jtH_}c!&^dETZ
zpAW<`fP~fuzQE-_2z#ht8gf9_aKqAPwHNqq0tA1fKmRs@K7c^il2uQI<v;ZYzqf@p
zpqNbP*Ln?kcK`lo!U~Z4+AaBi*JR)PrwK0n+TG*zdmvBl88K8ic4K18P@aEbnmAH0
z|I^B6sm2IB<erUR<!>q;=H-Nn`{zAw=U)H)*1sr!p_{BeKzKMzA*ne2=WqWa-c6lF
zePyGQY!FCSpA>P<f*|{l_3gK32VW$};{e0U)K%FP$b-zZur!~xyB)#QvX0)w+1KoX
z{2MpKeOcuS^5I!VO#dE+M|gmd*?}GINEVS;atRRjaay-rRy$u`o=697pZ;7Pe3as5
zFt-i}8qGi|mxDvU9q|B^B664W=4<`Nv)#0={+$qwm?Pj})qxjH!uKvjEG-WKM7Q*1
zwB?39AyuBV`QMvPlFycWunKK1z<XX5DM8fK|NLh|87BE9YTHJxQbkq^r*-QbP$0wu
zJN^aY=`|kDRYX2{8L5739|+N+CXEst{hOYf5imy=GuGE@@x{TFl{Sl9fCzj1mkbH5
ztm=r1-Sg@``hr@qf53pDqfi1H>i#o%7%I+bNGresn<P80-7`<?k_r-${$)|qmvVLZ
zr~=g1JD8uqtpBkbKq>^RW>FT48poi*E+}{H44?*uRmV|kx@-Wkr#WkI)`sPw=W?`V
zlpSYGJl5c7%>fdB(E`)m_192}`QBA_GT#*eV1eDdp)^3oARe!l8UHt7_H#;XPjpNv
z25}y^bXYr1Er3l9ief!|1t`IAe3Z00ce+3#IN<>qVOG>I1ZKY~OEMAwb;k~4Ir8KU
z1M2{K`10|DS=hgJ6n+wzDmAuxv6heXFaca)-mYVi-sfRyIt<tMm{#otaBdTb2$=)K
z#p8gl-E%bvk`I0@5J@I|0i!(cYGjdF`Vhmvc8-o1#K^7kY<#*M#|fB?K>*Tv4Y4iz
z7)T#48>11bGZvt;G?r@EeG!vw@wnXKn^kzFXhZ#9%fIUZ7}m#=B1~uG@M}3u+r<q7
zP%V3u)>@urCH6_+xB3wg^f+=rRR|Fpb4W9my}#@rX7~>*_pd!k4z}>@x~9E_bU+S-
zQ7SZ${>T()^$-ckp_e*_Q<|wNQUfqxNuyal9L}r%QziP>YK)fv2IW9~Tju)V56D^u
zG3|W@L`AvaHg`3K&0TxnFpd=5`5_ABK{SR!J;Abt8^)b~tY*F#Xoxjp_)m~cREHth
zG+-t;Q7_m9tRGWhV^<pzg6su_<$)Ry<VK32=oM2H2Ja_&f`6MET25$5mopQc6B8D{
zonv_@5yI)X;G82%c?5ES34rDG$)rR0=H1C<i1zvp$UYVV7GZ~(WNrEMzJKm4@OdU2
zw5W#3HCvn~sukibwrV<3HLs%P-v=DeVcTcG^|q>*v+0imY?_02h&eyg*8-|smEv^;
zqyL(sBw;Yq43n!vwNuW=@0`ucrWU|lPmuC{e>@8nk(?|;IB$#6He~K0?&0GC=6`G(
z&_C}X>c6)6`T`v6@D9-m$Q{pXT55<-4@!eFR@DT?KDugkYp&LCA|2SjdG6<L)|k%o
zaV-fwU=Ah>T!EcRc$Dna&WfR+%MO?^5^h*Nryy9@1c7rOUCuy(%YS_d<wzG~V1IUD
zHC%z@<kXpkqD4Rj^&YqU1BQKwTLjQ$8dfa0Y{%&~xvXk5ZT$~Rxisj1!itEL8%<f{
zfZ}-v`IW-+*)X`VtEG@%AS|;YkOCGARVXq0s}IuvS0@859kXF!@r?i4FO(xzi0lBY
zz*bLN|LZ};0DBO$GcZ&e0KJX+>Pe5_RhgpkoWVRUWHte*X>HXU#Er2QAurHd<Xua~
zv5GY@Bg;MA378HYSLS}?Q~uX5f4up6LY?dTpgOB&uei4H{9QSe($+%y5G!+K9IKDq
z@*$e!ncla~faCZlAaBXteGHNOLo7i@7PbRy{u+++_LjygvOM$S1Hk&%O_cois(J<p
z*yHf$A%0DIDVVApfJ}HUTw##%jgy7=S_50k1`sMO8i44ku)m@LaTsv#^i^u1uY3=A
zkQ^=ZdHBh}N62mtfGe39;L5>(L<=U=4S$t_$j8Fa>+Qm702*>@Osxjh8=R4FsNZy*
zz9L1eyp3300d=$+ij4iK#-X=tbOVyEA!^)~2UW>iw*^nVH2`fchV9$`;L_{;q2md_
z4S1krAG5MTqe9=Q>yNhEuHz6SDc<AqJPxvNjOCvxRZ$_#n+D*>8UWnQ;~5q;#jT++
zv!5PHX!;&^5xuOG+J$)0S*h+kD?HM1cx7M-3dKb^fKBJx(ar}EfmC*H0R0-*2E!MH
z?jVG^e|ww(pnE>mfzh}8AlqcW>GXTDypJisZkPvj1sGPPP-JH>&dFUyok-W?%;;j>
z`nB^A?>eWt8(^Jo3n#cs9ZRMDLOv;vL_f;+rT=ZK9(A*_(e|~Y$(KA3#g^#o9wP;9
z71bw$C1K!YDBMAK5@g-@v)hj|n{4|H8HU-)8EV13Cr)q329-(w>}pFPiU_|C{L>H?
z>sq1SN#6;;C<_t9=UzW73J7^18F|{F!-CH)n34Et)e#W0S9o@Y;Ii~R@}%tp)+=TZ
zmy~QV@Cfoi8UcvMVQC4>A@_23Fozqv8q+0!&tdJpteB|=)yxTi7h+m{Hkf-RS0e}J
z46IT<RqH`)TNB{huJ3!uK=i0EGm0J5s&elfrX@IZ$YlopG0kX@)!DAW_x&LFvXnjj
zS9li@(p|@y#&x};VNfV&qLyKiahQHLg3cx0G@I^H7BCSx#K{{d&I@SUPj{zi(M;U!
z-vZ8r;`s&02)Qy=q*%UmKZ=?kVn^En+{M<Km!n;g3g(4neo2#r4F0(9eI27F#8P7s
zowz~yu+>^&#ZX_B_ml5VG?hx9XNBZGOzOHaO&VjPK{5d=nmP1;yhU1WZ@Ok%f`qvb
zH0e&~eJsT_jg1h{CHQgDmJEg0sy~rLwj1T!alpxBfLB2~#0?+I9VC>1eV5xJ_Q5eH
z4HRFJJtdkYnw<bnLjuu*>Joz4(@Q-e)&iM$&R%Vgv(+!~^2wf1kqT+!fM!bH-xqr?
zluAKwKe~E0&f1wz&jVxzYM<cQ2JTBcESWKum(^Dd#scc0>25}jEWEMDMWyX%#$5zz
z%KNnQ_bErx5I?}0fS=Rn(jznhA<d%7wNxb&EQETWg6&~J=2O~BTh~MCn0VBUdIIV@
z|2|z(D^NLAzQ~G74$%K>p>3yaEos~jRCyO**!5I>Gy<;7j%vAGDTst!ifKB}aLo4b
ziIO>3cS+Au6QXqJ>XPgp6#y-s8@m|B;@wVLFFD^{5zniy=oN4&hDyd|7J2WfOC^X`
zIc~*^vTr7e%`6nCv2441jEbJxfTA^pAX3QvMfKB6UN!J*D?@~M=o^%5v>aPD2mb!v
z5llq1l8tO(s$RUfz%$WJ`{|&b5K?mvqKKxuF($Y|1TcCT+*}Y#nT4`zaw(n40>A1q
z8EY40xnRnscPnq|2X`Vx=f8C`QmgSa{`RDVs1YoSHdMC~TT(h{UcGZbW9$0?d}XD+
zL>ci{*P%`Zk%j`(LyrQFAo_UM`}TNv(zrx1NpbRm92HtLx<i+2t2nms{9@HFTRc2C
zs%{PkF1!uo{^H2Fcx$Xbav(Zlspk0efU9V$5csQ;xErsG)TFE?7bdo9|89!1tX^M!
z=m`8@9rmRz?WIYGeR@2VQRwcJ3jsfk2V&WbE@t$7BIg`xrOYz^{yTgKX#7nZz3PVF
z!sfw<X&-pEwiUX{@0x5KG2jTP=~w}2nP;h5D0YG64AeDRd+JBO!*~0@3Fdj3oG%Nl
z2Q<;td1zLB>!QX1^7=1Fj308u_QLbppw@*Y2{l0TQ}NO;>W^><Ep#nR41nej_~SSx
zJW&56<~{!C=v2N3?Tpr&l)OQ_)*;S_O-F^#&RvN7zMy%_LB9BllEQOpXeq&ar!#<;
z_@HO2{IRnCluPkGeyE@fz2stE+N|vNU_innlrKrEZn{D&lxhME2D0`mmESxLI&xqz
zViw`Hnn61qacOS`ga5nWk>=l<fmjwikbzoo0}Z65^;U&if$n0MtMPj&!b6TntC%to
zw0lxj5sH)?vQ{Qsy6h7FtSUk<UKI*Te{tpMAq=X5CmenZW)}Yr-U&lD%g??F95>$)
zg&3c;Bh_oUsxndcx2a5Z1yAfIn-?YDkh0NxOZDvm7c0rdEh|aMh5={vrhPZp<3{l0
zU{K~3<UM<1QiNsiB)@4#%~pj5=l0zFt#^Mns}cv`;VtdT#+9J~H9_{KGI9i7LpCAw
zAV+ch(1YKu%@Po^*&Z7y|LYH_!(zaSQR9jH(>(_8MoT37q?-ja<0aGtjWT!bXS4Lb
zAF-nauOfBPhcQ8~`s^cA1U>O&qX6^ITxx=wdr?&4-zlaC3PsSWl(xEK|9e%UK>-18
zmXP=xaS^cxVJcw3Xhkw)JN$k36>`LjHQ09?oktmWz4=;jUG>8vw_*mE<nnJE1FSw!
z@!!Bmn&=>K+tsQ)zxon!gCtd99AJX(J2w3-kOIDL1)B7QXDWar6oMdlgHn$;P)JN`
z%c5ul@#}M39*(^h5J&oRjy+2y0}4XL{>dSbv+vDJRSFKx1b}PddNG`}V%(l7^7m??
zK;Xp-6XSWEnlA?}uzKHCpo?L`4C2<Ed+94=paQ`d<XU7IGyZduQ;wh@ly7jq%in@_
zmp3zo(7Sv9ccq?r@wzuEoTMuw_|JD^(6HM3Y^xpp^$SuNy?&+|&K1r^a7s%zMN-$i
z7eW8?EE6Tg<SL=}r0LBDj6nH#pLcZ5ZcNtrix?3XZ3_^E+RNO_`)$Z65G=*Pm?Y@1
za;g13CjQ)vD3XB0%-ESp$?p~MS2PCN`UyB7R%1*gGx%SZAg~Oc0~5iM6neul_~#F9
zK2+X=R;EMSxc2YmabtFFR<|A4O8>7{jXe4LuQq59K$JhM?tH6Y5Xjd|%g(F56iYQi
zM8>>tA}4V63exSVF2vybZ6{WQg~rfVg%$0gxtF_fbs${r-ch~?7;!rU{+&{vIuNcH
zLpS~1NBl%D?HF_q@9`%%8>$GvCseC((77b^j>dHYkT)$Ck0kBC$7vtP74cvN=7YEf
zg(JeE+kxz!{mu9GAs)^b3_PcX_Ph{cpu7-UeOHUPSsIERwKrXMY<l4$;HF>;Vl00?
z?O+KypPSSQEndKFXkH>D@pt935Iop8OiN;l5#T#M#yxSix#wqDJ>6{qd^)4qDoESK
z0RaRUO7Q1rer`VhYh%2_4coR6X89O@V_N?#3Fw0Y2^jmTw_}R``t8kWjszLxR3h=g
z|G7Ki<OTynJ5#$!`@a(g-j%L3f}Uag$DjFozj~pJLrcnX&hC#T`Lhg!grQz-_$vOV
z@cpk*0zb_jf@W}A@k#y9I!sLkrkL=s{f}+_%L%#pSp^oN76}jO|E0$YU`e67=#Ml0
zZ{ufxWh)9!2>IV0(?TW<sfpDr`hOR_C1}P(x2(kf_831BjMF>ayIj)$+jwr!4D;pe
zj6WUy^8rl?jMM-3=@NRb7y9&=HAcrlqgLa0cH#Gk!h_z!V;2Pq;h=(u3(*<~vav#(
zjzEA*A15IfQi1q0%^!sH!7wrnWGRR~qagAA+TkxfJ?JNu^{0Q}Lw#iHxXz5Vy%yq3
zU@QI$8U0yEfCdUo`jiLVRKvWu{^#@Hb2G3XC$5B=xZdq}LSJ<ZgSjmDvvvrtZ1Vo4
zP6q6pSdG|{y(UhsU+YS5gyma-q;jC?wb)m#h4C!vpLS`37)(LKno>bV(la<CN1*Lc
zEnU9jFdC)u@}KYi&xY279vjV+oG70K9&80jomAyg^MPu`fAk(WWFXeFu9-_ssetDV
z%F1U4z<t?aD-EKm%;vNEcNEVK7YuBY9TzO7n97TUqt6Ni)dm2LR6ldaVbwBKF6bZM
z>;#sX3oFpj+m!;%09;|8%4ds-6m}u($?t!kqrif@5g;=m^`^^79HnmnhhGh2Ls7#W
zzKVY|R0tlLqgF8myk`_F2(%4=6N_7Jx(VEE5ArpS{@I}qLbKSwsB_0iVGwv*ASRYW
zWOQ*LAia<X#7d?ymFfQTdeDvyY%r+&D<b(M^TJ;0bBy02=bPOxAiCRLZ4evLz<hDz
z7X4}Q1`722H}-tIc@aGjd5&w~gDJm=gUC<n*|12sWuoT?R+Rs##+Nz|jD_V1L+){q
z5E}gJ>7>`m&V@#5Rs(dUmH*slsfM77yRp)FAs>Vi;rjMKdB1t)PWPEf!9T}g9VfQ_
zpgxn#p5GfVv-U$t7~}T$sK?g8H_DQaKD+qtc|EHd+kAPfHM>yAR<FOFSPs#gYxOf<
zfa=TP(%6arxCM|$_z-&QhyJOgBqwlR9E@c%93Zi7)K_1#2BCXUAJa^CYZP1wI~F1N
zE`{?`3!Vo-HelM#thg_AUnkiQ05^i#sKz+(UlN3=q$rh(vmX2kDkLN#OCSKr3N~wO
z9nh&87?T&YY`!|*9(B%6a?s=eIktFTLe0+uK+DUx``&9#0BNYW+=dmX<!RTj=;4S@
zH-Ru=VYzgBJ}4b&nk~D@Uhv}RQ9xJZ{yc7#Tky%VcgE#FMf!1sk;j303ebET1ef2}
zYbl-{7p{*C{yHlbftz^o6rG76tQER@lezs6J~&+tO@J6PtFe)yoyMkLr_2v<&st>5
zrwAPPw0I4&yozu#CFV2<Q*rPIuG%cEQB7EkR4}f0^~2VKb5N;LHN0MJ?j}qkQ|<mF
zibgj66_CK%XyWX+!=YY6?m-Uec5y&TRiNyQs8a1OW6YTh)_ga|#Jzl~AAmtQDsmRe
z7w{Oa`RM5T66}2D3M|ePKu~hnUe3t?Q}qo|p)WhZiVw=UI>B2GMzunDk?v4(Yk{*B
zybCC`M(iOf@$s(yu#e+dsK?>Mij{C_%`T#nM>i^$cIaTs&f8?orl!<VzXTG1#}e^t
z?DS1QgiSMsQQLz03^%Mifkhs6<G2peEQ#2)AEd)A<8+bV>IT8zN}vN)0(d@;YwQ7L
z0|jfW9$Joe_~2Vf?RjS_I3V38D9AL(P||2nR0AB*TvBhEHPumd&<T=$wW=VY8l0s#
zO4FzCIdCi>)RonpdKljL={`oyoK4W6c^PnnNUloHZ^GHBW6*M}WS9}Zd=Rn?;2dBz
z9DHmu=*QX}m1eQTv#3<$nQ6{ba+tls&A&yda2KJLSvpWsFNAlE^%oBEAqcL<LY083
zUz>h*lUS;cK$H~+JSL^-v6J#e;GE2bMt>`A^8+8M8^A2C0cydDvANVG=H}CH#(7!M
z>|lAiCvk2603QGJACLd|p^yo-=K$R`5XGBm0m3gZKP<~Cn5uEp(L}|qk=;p-svA(w
zZ=-L90RQbhq=<I_)>IrIHlJO&tMn!TPN0J>vAq9Kwy!%8J~j|CTS5tVdgo8eIU)+p
zKy=wpAssB<o*Q{Vup4V6yJCl)-`tFMA=OKM-70R2gU$fF-OzQwm29n(Q$1yX&Od2M
z;*gb*@Ufg9C^yVX&^3^{=fQX(Ge?UP3uG7vfkcV%=TB_L19#28Hf^K_4S?XC^)Vm-
zsQ~qaF@0vh4Fkb?UobKbbxM|VL;mC*QxM>34TMO`1*2=}r30mxuFD)B{ZNy+Chw6)
zzohrYHhPIGA{i|bo$31{V#l}qSNZn~Ig#>TKPrM~FNnn`%|8?ZGKv1^a^?5Rp5G5H
zlFm|GTWr6d>^ZJ2OgcYlp0BfgJ(TD?RP#M)I5Aj{r+(5e{DBu7LKptX;x*T#_Slul
z68)gQAYj?Pm@o2*5vB~vlV3ogP%wc<m=_5Oe@T(*d<mFxkvTv<vI6AUrf3v6Utl^>
zHVKuIZ1#(O#TJ6|PDiI*(g7t6gFs}}{MLf^t=sXCbx(fkO6)EX20f9Bj=dp%iPlx=
z7GLun*hen2@<tHkVYe5)Q(~Bf)DT3(!W~4&B;`!KHCaL)l)w`lnNK4Bl+)}>y6<HR
zELGSiay@GMwu=In^ZqM-zJhp;e0g^u4Qe|Wa)~WO;fHYEkgFI!2{aH7z-EsFNf4M!
zENfdL5bHFaHi`-UdQb5cB6cnf=#*Lg3^m^*ByXqfvYEwgva!(Aup6P3E@~T#fUc}m
z>I&R~go+_PhN3X5IeejFkny<D-rSNdZZ=nZo*@y+4bf(O(T?>{nuHsS%A?S|_4yD0
zZA`r4v_}Cki8a_APY!@hhjbqE#R>UBtXxzu?`26tTUZ0E=N$~2`rY(%jk$hu6}3JB
zd2cA(Ro!<E5~w>q)gYxcln641#u>rff$OM>W}w|@-X?QxDoFPBi7T~lmj@s!w0KTa
z1BXR-$9HpRwL8pHIsgu_Ycc>8uQLwQ>sS`^KY$s4&~r`o+PBXSLphW9?Af7I=0&YV
z2|(zTeF-+Q2`fOLTT&L}lf7*N@Jb1wJd-KMX6t${gn!Qj5+)j<P>ajw)@VBfas>Kh
zE*ohh%Wfe3m+g6d#l3U|L^VZf0`~hr9d%!IB-0f_JemSl>Cwp{#BLAtP4#}s2XcU^
z4<twxuQ6b;+?2QI{D5;TEzh%%#P1j)#xXL=&~@;Ec}iuIL7&qYHdk#rM)TExR%pCZ
z4Q)+(v>aLQ4CF2cf#uy(jrny1(gxgU?Vvo)l)3^mb$Xvgqr7v-9_*jjGsfL@gff>K
zVDPWs&vV6O$1@htNWfRevxSoKb5Dt19w{3(0N>&Ec4caIaE5rWlsZuk&s!GH#6mjr
zCrF1pxoRc4;xE2zy1uMx1o8)dmKb=$9O>g=8{{D%)<?M*NjvG*PXh*&f<X}de|3SL
zFVCE}kEjg*FO6@p92y*U&2k#h8l=#iz}UXE1-|fYu3WSI#;<-P@ervUP)aEovm<BD
z`|fI^uqDucbOZ!MQ5|53@zABrbfM9?bcjhNfqM{=DXyN-+8!5cyJW7lbP|K}?jstR
ztJkTwV|$S;!v<yWc>x>J?xGKYs(RYKd@{!~IPWQx5j$S)82WgmOJJ`Cq2g8_?W6gJ
zpM%%z>6P>Fcj%|a+=9^FfIx_DZsEg-(DM!YFgGx8xbX`NKbX{Ps_OWf&wB~^o&vp1
zVW>*_g-KWe2L8^XO+QmxHF2W&rHdpOj3flJdlmi64Y9noYq5}db^|UC47E+;p_dR9
zQtp@F6{7%&H2jtP(2#gX`ct}Y45uel`6ZF@=o+?+`GhSX9!h{(3Py4EEJej9FH;je
zm^<I2LCUN*6+RtkEQwR8L|g(8RG$g<L#~r~)10K8(D_r(Yg#>+p5>Np!8K3C>tXI-
z5}s=K6@Uv;Gt`TLl2Resdy^Z@z{U!{$un)RGt5hniBU#A`lOb!1xe%<$(`k|&3NSR
zUYEtw_jhBcW_i?tL$o}O*ESo9KMw#qeD~D}AT3i*$(Bus2e$$NLk`$|;GWE{jg=Wn
zK!E~e=W}$IyFbC}W@sYvb<m{i<;)E7%yZby@ilZ9jyR6ZH#q28&Z+b7)fGEe7!N`Y
z4xYpF3(sXRBQ+^i=95D%o3hL;95WIvu4ZKUYs&zD6JuFj+#K#wdN-<VROErW1_8qP
zb%{rVi(a<g2>|hx9sE%3>a`6=p};ov=JWInU=FZDrPF?e1QJL0U6->+WR74DRs35)
zve?4<7?OF6h4?v;dAlATfMC_MaPY52E}L>>-hS%5Y^6buC?E^_JyJF%;y_hz{d_my
zIhmwRLpG6XI5Yg+<vkALyGUss^}uDceDM<L5xAaISbqCHIdnd220=TxftTeCfW<~T
z62xCPE#T;DQp}Woc~n?6>B2%!fERtBmZg0>{8RAyER^&){c5+TXR1I`Jf8(P{%bCs
z2nx-^*n``Nqw()DsP~V{j6H$`LAV*JhyneDgnG4baKrf>xg@@-ZV1uBz9sz&9O<xA
z8HU4=Z-7L+rYR&Y^5vu)7FXmUw*IfvpANid<T{QReL0Xm9z?D+y(%-j_*5IJH6MUz
z!Ef0o2#*T7prqFIttQBiF1{g(Wgve__2s)Qvm_Jw;g1Z*?+bN1Ljr7sBiC3@-IPYZ
zw`gA7oK-exBX)I(=WI`_lY=(`ip+?VrA5O?#?a$L%cKcb))au|S9<DH77&bsG0`;A
z-<8I+sqz*VQB{J=g_}UCBfcZe<J}A22OM@IKt09Y5PJ2c<Gmf-Jg3N(XpHH;xhNoZ
zr-jsAX5!bChlN!la&h8SwDKs_=w{2ujiXJcUs&c)C|AjNgcW#ArChwIYDA;A*C}W6
z%wIOHZ>~O28*D*7dSO{9P2aL6=O8KDV^4ahM?o(+Px2f_mwK%Q0aq_uGADFBwnMh?
zzZT*MrOp8>w%Uxq_N3V?fHSEVNM0Bvz;*w?r?tUEfZbgQ?1e3vYP`V8l1AfG&Zq*k
zOtC(ge4kD92)X1{6_&#hH0I&xO;nk3%qY3q%cQ}kPbp%8xw;n<>kO`MhsWPGOY+!j
zJdnG{_Zp+-I)l9iK>+&#Gk2BX405}IJ7}Y@Yq>GTc>pCqg*R09bH6yQ>Uj#ZbbUf~
z&NwHn^YM*^XugpA<^qT|o0XX?sD>n|F}c-t3&@^**2LOSb}dL%l&m^Z)3J~0t@_m_
zY@?0B@A*{kF+kqDyZx}f8DOK*nOe+rk6hDcw5gFcUBbe*(w>EM8So(5;Ewo|0B<DT
z4})YzQ>SHCOzvJe9s&wQn7`2FlVhVvdi)WLl5>o$ZpBMZNf?Mb3l|p`qkpJ;;<%^X
z^q!{5$_a41nYm@_4QGjB@^;?2%i7vhRnqrTR!4<DM@Z8<=sc*}YMxKO!Jt|31&qzG
zI~FL)Qlz04AQ%r86p&LRb<Ee>SIAazv8puFMQJpjne7=)`m1xLZb=uBSX0Bo&CT$6
zoI)m`L2*!MQaW*zwZA%ls>E7AWg9ij_1baUPIf#~tEz5B^&oBwUb+qs-m$Jk=qz+G
zRrnUaP*gYiV~|?i5#<p#WWKgfLqt)|fa9u%Il*k7yq^ECjcfzr0&L)t=^Rp5sx^m*
z?_>&>Y;e*WoFBC4rcRirlz)|H=8=D8c6o6k&r4JiKX`sQO~|Mow#6!@RRV9sd0kXb
zU1_~2aQGOF4Cbi1p6SfNMWG`Ax+36g_Ht5Qp&#TAwXs*WnxymJs*$gQ6jet^xen%m
z@_NZ;k#=|m`AYo}v9h##w~iD{ZPpzCjI!_*28*ve^Iq1+vlT=kp~53#(^i2-XAy>O
z{^a!iu+|6L$RQ?;RwhM&k*#deH^d7wT@1P&Ey#P|z4mN6uke#?$RZZw;!x!<J|Z}}
z&@i)Gw4stM%FOONhx~?JJ!xVMt;=Wbx1(fl`Isi16EJM_QD?FDQ?q!I2P^D^Z$H{y
zXi$rC22RPIrH1Z{q!vN!ad3R=OYCDZq`pImne#H;ds>1LsU?&#ynUTQ)^0)}WIy}a
zc*n~oCS@5MdB2cs8u$aQoo$gdP&Tr)s-8&2v-^VfGJjU#s5A&EzHdJW;dp2b@`coV
zN=*>I*-4(|TIZhEn7gFeRdsCc)4^WH3rIX%um57pl^YS?f9_+niL6@mWRA9NI+6*m
z!ealR=IBD{!iVQ<Hqw<BcDDV!h`ch{ZdCFf0oR%EGjaYh90V~!g~>Zra9=9TiFuZQ
zfDGe4PA?f4%~MbKc;@RCBxyQCPIq7ANMIvLMH1k$r-ET@gcP4IuY=Za*T;KuG3HHM
z*K?YTQ1DxCLurTwaHl60E{;cI_(3NzIxZ0(G&%SsKL5^EPXT-~ey%kT%xp6zuBDGr
zPS9X>HcRnEC!*QOzs;p6lQ;B|YqO5(GUdqJ$+p+W9h=tAiwrxzAz}>%>0bxFMcID_
z2ahNMJOvCR{VY6%3NZ|UERxyWiI27A`<i#*-_2b+Dw8W;lE0B`)2)`Qs)Ya@r3qH&
z<G)_|@R|zMKyp~fb@AQLA8tqU+3(K4kn8wBY03hLOh9D*{doRU5NW_dhV7e^zu1i!
zKJN-5^507~YdYsIAe6jiN;jCmGt6l^O2rxKrkZW8E8uzcE3rb3tWu)IGmaa>@s`!9
z+Z%^xsm~GO@wBnV$6>-g8fJx9xWG^-Sb&>@0ea74PoyWr7^+-YdB`~dKQ|Q*u|cDs
zMM8IBRQDGbhno?#!pPr$DaLT+m&JegMPlBn*jB2i=(Bv0t2XFUG^3pb<(ejh)Mkr7
zdfSCgb6-MGj%x{4s8OLr_7aOqJGFqOslFFeI-Xt@YahU6Fd*^asm;jtYTNB!yaoU{
zE?-mMUC_<>nQL>7i#0C3reQo^g`p<)M?OvnOh(Jwz24ai&*2*ef)Vb>EdkhuF1X}~
zU!OR!zQ{&z)5poB1gk>rOV`k`GMGu43=*rOk_RfyaKcvKxURspdtj-}$0rKh8(vPp
zC(Go>CJ?T?1r$d9$aP!9OYdYY3)ed9vMo+vZd1&rc9vYS#Y|6up>CteLslL|o{sJ^
z+T!7ngPkbDo8_-eHiq8kWrrC%u+k5f6j*X%tJz2~w+M!XtVzgO(G*=)g|0mxdUTeZ
z4=){MM><n&X|7pqVaiF)g6d6+zM=oIUI=8^<c)z)jUqF>ktm!u8V<!sQ5d?$#@Gl-
zX}sKX+2{@{ICS?X`wS&;!h6>j+cZ0X{+W}M&+ha(%9DlEb*BvVJBM?Gw(SMb%buK)
z<%%$?-pR46^1X7fnl8Dgc+hI01cGeuuG-F@j?6(KOa-V3SSsF0YhB9HSTV}dyN?8)
z4^|n!@J^71!YHKhGfk7*P+Y73Yso<j*UK_#NOG*b1Gz?*Yqxm1D*jW$(qo-7E*Owv
zk9w8$Y2%Wr!Z3ul+hnQOZ4_)k!kywTMVD&_d$*8g0Ai)gwO`2F8xdWRP+`oi3VUpB
z7@WNpkwH?^*g~lYh$i*g@I+-JIjNf3TnuJ2m3h0;FE)PlvS>nw_7-x%>*3@!-;wW)
zYl@e^=eK+@vgC8k1a8+)Q4i*5Jd?MY*20ysSIz8MYL^f|oMZ9sQILvIQG5%yWHgBD
zt?(GiMb)${eD}$xsm<q76s!fV>f$1!oyPSrSNMLb_bs2{VZLofsOWT-rp0omlI>^<
zRv~Muvb__JXl^(d0ih;2rsUDDQZ<Q-wAwN!=!Rt?)~f9@Dz==6&or=ZjkjpLV0#LR
z0=bY6_YWpS0|ADyKLoev$SjKTf~k6{Jvk^|K3}6=`s#CcRn$wPG3|D#{3s>X7C<jP
zyPbdLsv??bt%NrW;Cj=}!BH_oFDE4y(6W0#SS}&1Va6)Txz!8)j4b*Rhc>{g9(Ut=
zswRWo)_Cv6cdDBM#ShNgicQ%tIR`lt@DrLj{A#vg6J!b`r&I_QF1~BGKzvzSojJpg
zL6rsfa56Km?iZa4erp}Y&}!Z?EvpJ#MyB4-?sB!{$qcv!@T+B2D=x}i=5mSCco!cl
zV&}m}$^pggrYcEsGgw+msomV$o;awevwIPxg|2dcL{D34MqsSv(Y``!G31g=d5l>V
zjh*EkbHVJDEpx;NR0ltWBg~iDfm~q%M7MTIt8*3usLr&C0i65^$|)2LJzUO#c#Ic1
zFSg5dR4bg3zTiEQt86gqw7Psx&DC^l-qumAPl}+3J2Lm}+%uC9{d`tP1X$U?2SQ`3
z2D#mj<uNSM;Y~Ry4Gz`@4c%z)o4DbF#R~-D&<OMOleJ+|6Vb|sxtfAV0{K;dndU?Z
zxz1xGn%Npg69VMi3V}1pg;^5+nCnu7DRuFV1HhZ0&Yt(|$IOdMklIHo;=K~g3w_Hi
zIp|J+pFeF;-?lzeW&S#W8!9$S^DfB7`QoMJF__!O=6-(oes?U+)K#`OlV+tjuXKWr
z?j<aL>$`G3ol9lXl}9j;gGzkceUGazeb~MT=ElnlC#UAIn?o}M0B03dwJzTu=@=p%
zd+g>fy3MyZR3)_xS%O$zwX;pRH-ENS(UHP4;ajNQk5O}}12ltjOPU0KXeJJMms5HP
z9U^W0n!jTvc^D*#Dovg0C2@(OlJIPCPAM<u>M@!PY(R<Ut0IPtKL&~M`N{VbuZ)~)
zuti+k>uVdo*gfWWJybau$rWCjO*m@Oa`n9VEV&<`+RP5rQgnhp77Aq5(QvI#U~E8e
zdWl0FfZ?(Xi(#%k49yab8@JF{FZz^pxdyv(h+`}WC^oaRmn){Tr@1(r&CSb<4`Y}b
zr|s@zNOGrHRkL-BlD1vkf|msM$6um^b{ahjXDZUgizI`ag~V$jnan!%Qy#<2EJz0&
z#shDxD_m4$&jGG5QYGkpA!chg!Nb%RwtN_@gTX#V>g2l2KrS9n7|uH!y3S<IeG!O_
z@6B@%s=|#TUZKZU2>Z0FC#YC!Q;(`{6s5AyEOr0gT%$buyL+^)N?z$H2GmFPFYq=*
zs9=xtJ<B|7Bwap?*4uAuI~0`x($J#_d8M%hi}({`C6A{H3SL@)nG`Mwmeq7!X^)W+
zsBuOX?^dM(StX(YNR8$K7PHcN9}ylgU))Qm#*Ep1_t80sNlgWFPrr(o;R!h<%)eCF
zjf*$9KAyEM!?22ovwsmLPazrx@WQT$Q25XWwXkk2{usU(ITn-y&3_d1Rjcdkd<$+i
zHi#&84C|U}_B3XgP~zpD`zF;%Q9r{n?p|Tt^P}Z=TBP-2L4@MAO3c83i_5a8Fr=^6
z2JfGO9X7?Q6S00Y;5Ug>w-r&~ZN+Z+E`E@0iV!qZONdGS&{4|MU(y#xVKC>)svs+B
zExo8Ra}TBddZck=%TW&6WN6evsamwXvFjP$I(D|B)t0kUNdapZu+OEo>qAxUZsXT-
zgQzw04q7A|-Fxpv&<dNqr0N{<PJFawNg~YNB!r1ppaECyv%1h_JM8JJovE^W-vP=5
zopvLtEMZK;eGqnAU`TQm$|vkDnIqP}Z4Qql_1-`Q!PM{Zky+1(#gt!i=O2BP9!<=F
zYK_joF}E9Y1=Ph}fys)VOyyA&JkdtpruZCuu9D@3q8!o+-HjbU>31J*8cJdSnMeVD
z$lAihNwV(IRR@+}r68IImxoVt2bZD9{e){XtN9pRmtK<)+d+=MI{W~17^N$AJxx!s
zO~D_N)WDVoic5hzdwB-3Iejc(mCrG_J+jXtRn1k8Sl3L}f8#vCIL*-Lm3|^|UcEP#
z9MSFvx0Mk0jRLxM&YnsH2~WO)Z!t724^8lYf)L)~696nltUpsM=RrVm0ykGO@hWDs
zPSK-lx;2wh-%8kvMQ}<E3|YH?OIdfd{99(4>G6#%n&&T2HfKh)d5J8Kg~8mUGo%VM
z0H|ohr6L`d`Mb592~j=Ru@yi>q#Za){QP)0pWR#!MQD>+o`lTjrf*2aVdpvkH;?`r
zV}z3*fNTA}jE(0$lEujEUFBjO1E>qb10>z&lQQOZ>%%G5-5>86CfMPv5bmRe5djww
zTh<%IFXBQ-`DbWaAviyTmvpC{tFv8K=X|YNVP3^b0u+=Pd(Z>ctdX)rQtOenR8KZ-
zk6jQN-+UFBEqT;OOx$(=)oMMI(*@CVm3u2&=dYA93WV#DUAE{*R?O^-kdp)Vt+|0O
z1lHc8)bAfatZHEn2$8OV{Zu(0&KX~Jek}epFR#kjV`M2DMAQRa-qMjHpmYj(3GpBk
z<;99AR~Q}T1@0?zK%q2Z#sZ`MXvo9*#B9!16187kB(BC)2PykCWrLYFRna22&<qoV
zx~Bq<6#UfXkQGD+BS6Hm&iyoR^CUePe<L3jeNGkuZBsWS6uSe0k3X+m49-EQY5Ef}
zM;;wxPhNgq`8#mZ=z^+>Y}2vqRa$YzkUR><!;LT1^2LRmXpf-7S}4^8X*^Vo$(1vu
zvHmLe+R743raupsE`~l<sxO-1|APuVL-7(!kQ<U=<26X)?-C^^AB(e@N+HQ|rRL!Q
zh?j_#^t~<{@iZA?9SwIIt{?cqR`|szMnceXc#yj32hvXl{KVDGCtOmt$`NGs=6W*0
zWV>sjp=KI%U=E+X1fz65nQKBlPLv!@lNhQwiud}dyDHNvC}I}feHD+MNdbnZKJLM1
z*;`!_C=^~oQD}@+<4VWgAPr`1A@K^p_(LBuq@8)ezRBEr@#$tzfQ35v5_Hx)mMzQC
z&sZTtwVrYbg6AKRQ-Q!=#*Uj8TMo14Po1y=j|s#&F+&<dXe=lZhk)zF@8Dysbqc6!
zA3DpKH5UqW@9Dcyp-ZNLfgl7MqpViOmnG9ohHF?C|Ke?a?>>}&nW?e1uD9D%I*lj^
zyN_zaup(kZ47D@>as0KKj`uji20+AC6lF`<ZG6Pa*pq@ZJmdK6&AKm2AN(Ko-a0DE
zHGCVDkwyd+3_ut~qy!Zh5J4D=6d1Zg1%#oLlny&kX=#+sp*u}#EEpO=rBg}ibH5;N
zxBHwwzO~L-XRYu1{j(Q)oO$2pectD}pF6JWx{>#M7#qi$q((8SaXHOL=>CcWb06NT
z2Z%}ev{~nDljWX7MzXpi6(o{i997qPvYZ`2QAcG1&u+!NixLm=Dh((<$wce>lbUhp
z&7dB0egY7diWY`bErSDPbhVVGw@ZwffOfzv4|8UQ*|qpfqngtPAfHYQX>gN2aRM2F
zzOVCmM>zBbjMU9D4~N9gL3=Pb1_JMAdLwUcK~eYdVdEmx4I1a6EI&E6??~11RKUqy
z^it1EvLMAAz^7e8_buthPoT{Cx!Mm0=})A;_^|`l=TBNT#}B8Iu6=vseDfOwhq5E*
z45Yhx)sxPp88A_jk$(kq&Po{7%>LvB?(!U6mdU=|Xig=r@8DfWLcWrCPGHN?>S5+X
zbPE=bPWyHFT@9px@!|?lBZM-@<VV6@?+DdU^}|VbBQ+gxmN*q&KR*JGBn<2JsbTgR
zW^GOuG0RU0xOV14v0#1b2b$8g34VRr^5Mlv4U1p7Ycmu08%oeH!Tr11z?C153%0d>
zI2DpG5P;!7ry+YgOINQ2UMCamwfO6r3b?}*rAm>a0yP0b(l;q%a8+GU7&tH;oo1>y
z`}P!AQX2d8J~xL5WC`$+8)!d+c%4PeS(f)vWZiM5p?BfJZ|6uvQ};r|W(LAVgU+Y*
zR=A&GNyKWe+-Qz+0}<;?KmI43ya-exAZj~bQQpO)>!&ggHR5ZF?rM2Cz{BP8cnaoQ
zCRHbxf{gsqA(+Vc2bw9*S0REE<)58>sIZ8fO&U~47E0s(IdRw;u<281$%y)dJG~JQ
zBi2!rx>2v6GWHPX5zh+@EG>gWFt-gU35E-YpK>o0XmA_N+!HRO%7_dA&gs)8(Ej!q
zUbp{E{oNPhN6W@?3-^+jsur^Zllfk!g#|cW+OLWCx6NpV5@PujZ{IsUbx6f=w%Qdx
z8I&`ENrB~bPSpgccEWBy_ro*CpW!djvQTx5zW{H|2Hu*5aqRQpcNC?nX&+aB@v`q`
z8jp751z1ine7U1>WDNR6wl-c+K$~F={i)uaT&!imB_(nP30D~*T=jYIbewk?wy<*;
z4@D_$Ww?Y`Cx6}C#+{jDOvVdZMJG;Qzi8hm9fk<H!i6{%uAJKr%PIu6xaRft?mgj^
z*yn*>E3?|D?i7P<^}aA>G|mY%vMaqs6RP?l2}?{a?{|g>sct7{y!#kp9*-=|8ikQc
zk>@*y1ffbMs&r|!%g^wV5aP3d?cgU9tIuJVl;r~W$9ovBD(|j-WPAMDu^9=ri(BYd
znQkQ{zom7CV_=wv>5kC)R3W#u4gSV6drxx@KdL)zNwe?f4xXo2k%Q5G(cAYSc46bm
zO!>2can^459L|USeZIDZ+JWOcu+`wXoTM?@m^`>6AM=OW28EGi){t}lwi;RF{aADi
zJoJxp+h7)Cr0c`o81pT7EZV(VoWJj|KAu<VV6kIb{nh^ZJOQR&iR>lRPXD;Zmdz!B
z)KvfVITPHJ*F?vEbMza}%BTXb|GJTw!e5_1IRiJP*p|Ne7{5mPl@}KM`%#(Hzdl!k
zX-)FEi1{@$8`t*1O2S_I=Gx8k_vf^5Q>J4XGXFd~^4HyDcz#I&YvNy@%Yj1{iMmbN
znDt-R-h%qUg#O?^=oAkn%zsT7FC0ed)!Bbf?*Chpo)UZM6f!NowKR|;N86h{$Nn>b
zkGtYz#N1kM?)+z!tPidac?2r<G6>XukV9Ept-P^r9~rqCt%!BovHmL_8xMdrfT4f0
zPvqf7X&(NNcSWTMb~K-amGU0nDLTi|mdxc_apW5}>*K|8bW4)8?}sRu^FxX%rvr3<
z4F`@_#L=qd+!Is>w8<H%$!APb!AT#0lX!be&gMOkcjDI1BZksMX9FIJWd*4SZLY*x
z8K0zvc8eT=zusgY{EE~29uNAj&qZNX{m<9`|CpKCy!NLy|7^qOT*u7nnewiFc~MS_
zv1&O|=G|TTF?FkMuIar6fGjqG+Z5SaVmW9|j0nfH=x|PY)WS>u$_e{9*6O_c=7oI8
zeW8_e`?mcjMJ^gj*^RV5ewX>IUI=PL4mA{e`WzH$Dgl0{lii3yQEOCU9akeXD=}B@
zZW<LP4HJ;m6sC3NR*td&6@P&|sI?A8c~`j2UVz+;!{@3G63%d-=Hp3xgVuWl^h^=j
zq_Fp@hXbe?4MZ_Wzz+G3)i9xWCpmpV7^<MGV|=*ot1s#K&&Xg(1@!W3hYfZgHL^x9
zY&W_POpa}LX@4+-!~q|y2gUQI9h{Z7gV{FMHcuoLAR-U_x&?~5K2er|ugb5l`~0fO
zo<8vDni%_JzTC%AlGnqO=(D>#y8ru`F8G+tR|S)oP%MQU<Rrgz6sBxRv6kky)p?`g
zo~`(#VGTvj_#L*1Cw=Je+1yK~VFmxj`psDH{v&JvZ0`r;iiMl@0xM3QCeX%JEAZBC
z-`jKOMc7Lpt7L*1HwRp+fEI6C0e3sELt2n^SKpg{1yyBo1sH(7o5Q1jCf%@_q6~Ub
zi6K11)O?dLbkR`N8MZCHI)F%evpA)Cee)<l7x?&fBd{<8G|$XDwzDs(a=^oH3u=uS
z{h?osZ1ct`CcY3`YiYpH^=qm!qF7KAU~C1@hjoA3HC3jRV7Qam{VKy?8ea?WLu95)
zaeSM1ii<-5XoSX|ZgCWAPOVW6mi@KyCFS3b_LG1$xMApL;B{jM7OfA*f%WS)q@uHK
ziSFiL;VAd(+Pjvk>PTMC1J5htvHkZ66?_&gSR;7aR@_VC#9>qj60YxS4aoa##Bz7u
zb%T*dlKM>{&XC&)$0X{@Z-a&tK&jHh$kO}8o=l&MLxS(ud0baE2acXXsnWqebFT59
zCEIZqIg*bE?z)8(b=C)WpPMlRnc?fLPos9Eh{DM6WnZI?Zkk2LgSZJ~TKb-6L)OE7
zH#m6^F8^K+^*edAkxA99l8vHAslo~F|3Wiv+wTQ&j$C0sycL<g7wX%{`7S^z>4hLq
z@{P@<*vF4cL$2P#|9TZFy9#hdhin`DeRaOwJkH3~UsoAKokVfL)dk3e?);vYuROR6
z<kyd?=!5t9UI#zr&8VBI-sY3XC7^C?hbLEbTR5>7rO6M|S`ZeqOLp_gg?8~E3nDUH
zOdqz_D$LsgYv9(+2gm(DEl`$`moTsKzo0qD$O{uAXSkjA_t4L2qg#=im$n869@vB&
ziLX1^$jru3quP<5pD(61y|)Eb0PCqm;T7fP#B4!TVw2#VXA#~Ap{Rq^U}Azh*yuJF
zObVlK2eM#jOFl$VqK?C2_8-0#-?sT{Yzc)L1zfzpQizR%jC|zmAN~S13q|V;>eQ0L
zP##<d)<G&Q9MHw@8o?;AU3;}737p>&bD|eq4f_tBTD<AC%iEJqBW*9FfwKQbAZ*$?
z^oXw7yO$2uzowGMry7tQHaEKO?c(m}V(Ob4YtP#v9z+3i^`rtDU3FvIb7=T-sp2gJ
z_{6(AI4=YXR>V+pKmi`GIifj*iQhLyu2EJEFnTLJaK1}jCXL3Kq998VtO^hR@F&|}
z{!QF>)HWH->+1HuSL()I;)#b9yzhs&$mWas<pz8KOGRg=@Mv?6el0OGWS?_N-TBJ}
z<3Tp?7p1x#n=`(#sNx>Ne3-o&{P5RRZta1ql=F%o{9~fw2LI1P2Y#3sm+a<B+*onG
zLa+!zcf_9k%kO0}f(iJZ)c4!|Z5U#l3@kaJuMQS}UF8KB!QG#O**15{uP6AARrz(5
z>u{A9k%63lP1in{u7<nspZ#?eTErIre>drnGQAS+{>ZL!At@Xh%S`Oc=0;JQ9?Ax=
zR>b!GCr$GOSP_G)y6ZaBFNwgR-&b%+m7SbV+2jVj=rOCz0<G84d$^%oA3*IWy)vWT
zl&a1xY}Jx6f*U-t<E~FNR1&Qbthp8-Td9L@A4KM$fDWmynfF%vny)S|6xcQHhCmEt
zFQO$Pk!l{=HGl?y4R_Bq_3z-6VeAW?N^>|uzPcO9;1FvAICGGdaf$OPfF&6F-VESX
zKPc?DKAafV*qo})-|u+hG2hO+=i-(C@u;;!Jo!G>ksLaUvPHkZ?L@4=ZePG?HRn=r
z0QBHEIjC4<UNTjTp=Q!V@AS1*egCBKulY|?K=1TKJmY<KNUvxp@yDeJQgT1Q9;X0o
zhsk#Fz9%=&!kxLloU()c05#V`YeYOLEu;jH?avKN>u+Pn$c?;LoE$)L=xtUP(_UXw
zUpewp!0IXshP6XA^8z4ntun5_-t9Yl%<gkJiv&cQ32#nnA#<h5EVmj-0`D7P#GEfE
z4yN<fp>2$o5DV>3`ape%t~o<N>9gPbU%P|)l^hsunLQ7l|Fu7w1z|On?a<#?iof^H
z|9qYLe|ctbRdY&hsNdEKy~`JJVUb|k+m%L}Q!kISzMs%)=&1sCF$oH<;cTLgks!L#
za0D_5UWLtSid@#<N}pMhYQ5Z7=6&Yk!I{I+x;O93t3qaBD;eHV7<*~grk#OnBN{OI
z*6p8G5!o0SUv<%8Xe2{BC*2>cp4(pVrqHm=nSLrl+S}{<0`z6_o8^wWdL~<{=@#VE
zj<&6(#6PNcEXpWcf;<utM62>Tk{MYibKbJol&paK#I4#uN!S_}5L4?nnEh)p;apL)
z!3K+vk7|T^?C9r~#Zr?R)HDFfC_86C0Z-?hSq<rkm9chSWO)WlpLI~Iw%}5-M9el2
zYBDW^os}{6sXv@}<+Wsyb$t!Eq))2A#IN4&`@QURd<Td??F9nvhwGqDdZWK%x6EU9
z8zwgXOek@iLn$oPChY#wG9tQyfLIVy1DVT2#P(N+%l<ltd}()|05-B;o;zfHh~=)+
ze)Ge-C#u#~d%ppGtUBef-ElR*7B>Ra*34TOjMODIKS|i-zPVtxLsxd<w)~psltW~*
z{)U!`A%5yC7`#Sy6#fguG>St(Y1ks2RLFOl0|z)?8YD1ea<-qXglWR+QslDIW@nu5
z@7>&iP$m`mGzFOPE1<V>QojvpRNK6}Sn6l*jhcI@!+8<dCCicK)Bz~z3B`I$-3vc{
z`F7^GAyf}9EX|E8g$Wv70BH3xu}Lu9H9vxyKlR=WB*ocES0Kc)HS+jzei})mf{0eT
zHE=p~G0E?e8b=}}Y`acOLz<7uEL_U7^cn*VoaVNHvX5F!RH4u<f)tln$`wv89geiV
zbkcFigesU;sQ&U7z+e<nB2}lY4^l-97-+x`h<=jJU!^s&I6w4}^NYipKP3;Vu=zv7
z0?1BBLea2g?Wq}H%$~rT^hMV}ws*q^xQe1;ziz@fR0&^`!4%S7P)~%c^3k6ukV&qY
z%C~H41o161BM_Z2F`Rk<x?{FW1<n&);tTN~=7YFZL;*5a6gv*fYyi6G%C#wz_4$x(
zZG^71q1i^A1r{G)M0rfr(ToQ5^r;O%zL$?Ds6i!Ouq+^~$8~Uj9!Or7JZ=QWl?_5r
zk5h5@UbD8Fl7ZbClH}QaB;5*xpWX-#Q}oq`K$+~6+z!C{_{+t;`ZejoBucj@*#I%W
z&29&k!<xcV?sJDe&}&}eLM_k%Al&<`{crbnCwU8WbB7}?*~GB)bSkJi8BRl@vD1cG
z4uG5QoEJcgM}*dB1+9I09P+ewNGm0@9mPGCXRhF&)s!0S0Y(CM9%yzQCNDV|YIT1F
zDo+uPP3--8=6jAM@f8g$ISoT9MKt1e9er2g?9^Nn2$EPnlAMzB)@*42J(CMpw;Ar~
zS%#%0gfx~o=BWX1Zwzyl>W9@(_sRK@`PZLadOW-FkQ|JwMy;)B$5-zQ+P~>(;D{(M
z2Q!SZatH$V>gB|^W@!^*JiTFAlaG?l=_=hL<<A3C41;fR2S_(1S&Ar|sI^YEXBHiV
z>?I4pNf?`;9Rocg_D%ym;uFnXsaOfa`hrIMnal{&{qGZ@_Su!y<VIA=NEje8NPhR}
ze8m%|`CWju0(=RPpJM^C>^YrO{PsE@&QiF-4qckNcvJdde%5O^)^8YSKA8Lj+Nq((
z5`**!R5UDx`tl46%R5TDMav;(zX+i%-8VVG-y6*jOE9YN)5Bu+XmND*<%2Q();d?Z
zBN8Ct+TX;kXwhG?&rHM{Y32<FMr<1l(g+0`daCx!+p(5R{bP&MU|x352+)g8?Y?hl
zj$x1jfn7gN1m`ukOybs(C*bGCBlcKD7q)tp9HMxR6!S7qx2~pXY8?ZTNGoMK?{}w#
zbV8;(W1QX`&91}boh7#bM$jw};k`N?Ti94byBka@z>i8P`VR?jTJqj16bl*;`3>q5
z%Mo%#YQDi%eq~S-X(tPrY5)5A0rC7ELsR7wU&tL`l9R@I%7&97OEVe)P`TsCU~Vtf
zSC*^1*6hag3W*Ud*&1U)qZfTCqt(>tuck>AvryO5XD5zhu4@YpLqsoz4~&XxCg{G^
zR>P*(WCx4v5AI<S&)D5#S6B`+-?{RW3d!t!?~yfmT^DJv>Gt@h?H=kZvctVh>5wGB
zG<{nvDC}tAqR{bbx>96RaEG>wsXBEygYtsC<;UmeZI35w@5#0t-0#wL+>AX<z7qH$
zRvt!x*G6koY31SlsiXNYHXR~A<(5)u><f8GwRLIYEoW(3U+FIt*>tv;6MARCMa*wt
zIlhPx=;&EQvM0MSO{^4O1@{(lOfjISnGC=g?@ciM?TjyLW1Bu&%ot+$wZ>y*gq&|^
z5RvkWJ1G=@IK9R{FbJhjQL7@gkU)G%p<UO>{I!B(s8lNF;zH>_$;DVp!<_oD)<*mm
z0!XnX@SS<#=3RA@0q>dzUB#2SUd;cPZe^0izsMkarJ+vnBH~|U^|cNxVybu<UQodi
z7PdtxZ2Ifve}ef6pHY|e%;nW$ucj6_wD_N<z3E3Zu-e4+5rn#gk18;Pdapx)<!a>M
z%1MWTMl^#q-M)<PyR~#KQcN0ty>z@GgEzunzT}b2I3)$0L{d+}0@UwS+RkmgU(%QL
zBaln+p?=GiaLQpp4%pUccW@5!#4a;zE-Cb7G73z@W6fc(L`6hhOtf;Ec|(USuwg{^
zstIH(;sw*mo@tyd|0u)S-ipw0Ed*Q6BtVjmqJ0^bNsaxKAO87EwVZ~r*6filzU*5}
z1DB`kP2G$97ul9TW_{L;MEzbc6ObdeA1l)+TZvWaJ3<HFMO%sB3(7)SEXBKD`rT~2
z4q>=BoTmd9-u@2wu;VDCqhUKZ`~FE$stzp(&1AcILQP1p27f1{e{Mc(o^K)brVTg)
zd|0$rscmRd3{X3?S|uO7Z!qf${v=-O&Uu;p!Jy<~YMa@dY&66mu}bM$nJ?~NY~?Ku
zmd!eyC8TN+!S<?j>mo6vRrkvNqK-;DqY&n7D@NT7NP|SsAu|b^Rnh4g?LCWZNE4KZ
zq2OMwe`+jWP+rq!q^VOt2dcTfx}0vCIB?5J{>H)7!(vA)o6W<f_CKuW@KNAA_dZs-
z$1M8$$or*HZ`_cORVk-*CQN6I{;cj?q#hN=az&i^3tExc?^p(!WnRzaalHN<A)qP~
zWg@mGtK|+wm)9`~5E?RZ50O<Dg_?HUPp}QI#RjN1-K~BU-g%7j8c6OZ`}HcuQ>7P*
zi3?QBe5&E|&teM&_TcTC1Me4<k0zLO_$y~b3_=sySCsy%Ifx9#E>WexLf&)Xnye#%
z#`>8~2+VN5gtH?@r?}&uBRY)bE_Be^GJbz?Lq!h|g1Rng5h9Li-!NK`FL<?TP#J+w
zah)DmJ?yO8BwWH`fzfus^hMa#6PNTab3+({Jxu{1N}RF~f|yu8@bR=Mgyf`{5(pEm
zr_ID%vQV*a{8?Muo&hVN<|O8t6vD2GDo;BVC1M$9vpN>>rj)*EHr=VSdCm<^78VjZ
z-2+iXwyEr_^7fX99RsJe5lRti=BjsvmYPthEi-Tq7Ov<b5;3f}?d+0m(7bGW;Wm)!
z69cWyyn>JAg%MOb3_C<V<tTQjFzm9I7it6Koy{uXR*#KDX566z$7frvgD2qz^JhQG
zM1{Fa69x=-&y(Lj0wIv8N$&5g)1h9Jq47(*Ph*5ug96(Kmdq-sS&Tid@((QoCo1`f
z8ZOCIuWDg{_1k=b(&x&;dD?`@;&0m#!P2r+KN;z(%?c!q5uLhJMzhZl6<eE(u#*>e
zTqgDLZPCum8c00_KKj@f>D8Z?g|GHj_dRc$$5)QER^GM^Hc!O&k}fRx(FvyHg!m1D
ziUD6{0p5KDfbMr~03nW<x&%=nlPnZWUSH#addr3<f<A2muAi`%bQ>VePnh-SleX~p
zYS`d))OLLX_a+jY)=r*k$GXY9^nhr?DOrZO$ajkkww(}Qca}3h<w^z!+*8$f8H)53
z4?0zIC94@=CQ%a0yM32yV%J@x)oT29_E4rcIJKL*AVfag&2;+WQcJ4(+bk_(JrYkk
zjlKP+-isqHW#Hv;dk23vVqjYqT5Q_RWnIe(?%{n%K}n1q<yCm6YnrxwsctK)#kne=
zjhF|}MZ!^jl1ypoiKLJzfQAu9eu%A@6tc1oa%Mw9e;=`FM6I*Xh6o(ZuFm_OB<dHm
zGr+;OLioR(mbYF+NbF(37Trbg5O0luO7$?}@D(>Zt-C^_dZT4lA(~pp0gQ2JXgKwK
zhBZPv*WZ=VvxDNP*B18D%e(nCDV4q@NY##a6wuk=WB4RawOx(3Qi0NcziUEA727E0
zBt#+{F;^|R`G;bBZ(^eWKg^b#eqQ(20U0+!-fX8y7V4!Cu2Ghn)OazTwpHuVnP72J
zv(##iCTC`cjo97LHQi^Sk+Ae}yJ9UX0S{`#*oV*hrcvXTw2%Ytm?<~FO2^9E{>ZTT
z=A+U+N+C2Ho9%%0H3_GHdGtr99M=YTHhrEgDCQ%_+~+_SUa+j~aC25kuCg}fhi$?T
zt(2!SJ=6;!M?MjVLvOB>ExyGxDfjYZEL$EcvNYq{vM){u5Fpud2H@*qjFD%5=h~??
zl6+(yIy=x<Egm3EJOce9i+me^><1h9uylWlbU63V6=|8S1Lv{PkAm)Dk~3{&X(~!3
zT>T6>7pz%8g$UEO&5!_#wnc?2*Dw`LKQE_#r37pETG^%GJ-`2X-ISm+A&}5xm-;KE
zcsCM0?#kNRe}2>8Y%F~_NEt_l-eK5uR(@UV>tS5Z4OIby;txg-AZ>bV#PF3jKY}hx
zx)6Rz3bImpFKGFnLQU!=0)R|{rN85-`0HQB%-4e)De&xuN>Sbip4X*z4Tt;7yhYz$
z(uaC+81{V2kTQr0Zp7Z;bN?tkxZ~(J2=$tP@<NOYZ`IO-#5A;CY_Nt_BIXDV0lpL|
z4iF|H0w6r;GGV>dup$=3gg(EmKXe1IEfEkZp@Wnfa-pT3_HL%_>t!IEz3-WBs$xHe
zeL{zduOPiTx<?)wBYi$h)_A-hU-39&4Gs{+u9RA=OI;ur*^`)=fT=w{A<9sK7BNcD
z<f;(?v1;FbEID=e<5p_VXMiER!aM!x(5S>jaeN67=`S27AT&xl2+ge=zWIwK9Z?jQ
zzU#J7Z}~E3eL@vDakWcZs$OM@YL70Pc}F@ntgf7fkiH@EJtQSQqnJA##XpB}xg}PC
zmxuJho)T<BfRIeZ3#p1>f=Ff=(HIbxUR%y{1$vvm-^DN_2o`o4(QttO@1$w#UOZC{
z^*Vx#nVKF1gLDCAr>9{Ev~Ibkw~_oXqb@QH=jO<H_d@kA3Rb~u!%ssoACRzsv9A#L
zpUX=<Yoe^S#TWh07=v8S04RN6fK6T)3(#+ef&2Vf#}IXAe;eI(d<Uc&NW51;#WZri
zb6ZV}2lx(YyL|Ixw)o<F?^~C$xg}*BN%z4YMw&;)Z3D+X3}Hf_xnW?_T<+>145jlX
zf^n5etW{RYo0d6Bew|;jAb80*t0VsN&tf4xQkf=j!WGIuOeys&-3ak^ZlM`12f`#d
zh~PydNFWdqJz;RU%vuq%o}`|v%qzR5F5Qb1LMe>^hn&5lfaKt3M6Mx79AX4UF?e;U
z8^J<LEq^~-<^{?@#}TTlD=e_FoT{H+>5!l|9PIN`0C*)!W;M*faVm_YarbA)6WtJ{
zA^iy;`W!L)J~T%fovd23?Tf34fzIlzOZ|wM4Ucy2sLOziR80~<vqb<(s*JPs{EqN<
zt2X(&ZKB;T5Sc8vw3V2uF0>jl9cfWoi#i``B_Uo`Sw14`4hS!*I}xHbddE&^fovoZ
zUcg?DfulSrtAO$umwSij%S_Pw3QTA6NXvH2dxRD>-n)CjDDP6(Q65RUb;Fjk`|A99
z`-1Uz!eHE{Aw3YGZRGm;9s9E!0Y~GnR1PCz>F=G6KHKrYi6a*d=Lq1h^CsRhPsHsx
z;8`^++cOPu!?98-Ty_fCftBC}PoFnadm(GF09SfAp)C^Wr)Ou`D<2|ks1~aq{kT5G
zovE93xDup?3v=gB)-B=RP<B)CmQ30f`3n;Vy;NLNCkaz(;ig&BA6Ghnf_!;A#%s~i
zUCyb7)+%wJ{v|&cMlJM;apb4a#6z-89O($1k$n{DQBCVcxoz*0tJY%pSig^uvXT0t
zc%F6SaTn6&8<8izxzfc-VUI-9!`)>?`ew!)Ank&rhCBO_Opz$ji7cw>8HxxRrvm*P
ziWL1L+AM{HDOG`#h}EQsm@D8Y4}fO57=ZO2c<YpTK$sW-G>`WdOs~9`zj2sD3S$70
zd@60qHA!AbZy_Wr#O)C}g2;{U@klsamx}si62B&pvJIqtlElB><s4(7NT%C;<QP4x
zX#ONZmdKouX#k9B_Gr-BM~im~i&HMfcM-ysXYH3lO>`DnQLtk-Y-i)3NesX9f<e3M
zx|P05@bR{LKKUKs+Yz9_EdJ%Dqj4PK#dVPFpFK5J6l~8rx(Z!+Y>xqS%pu)Rn<92`
zRmEv7j-DAeYINS0pe8y`mUd$AiK}t0K>Z)DtXt#h)mRYuV5+l-FcDWY(^i}&gO_~_
zx9X3?oF4{-3QTE{`jnDGMUT2v8u`alcm3x25~dOP-m8;w<1;=nkSqMiC}*6w+h1sR
zCKC%uOuTlMAxV0oE;gNJ0;&nIjD29AN|PXITa!hL#yCCli_)-!<9d6gMSc}N)KG0p
zH{c3yUM_wgJP6-p8G}~wXVZH-xNPDhbcrgFQO)mXp(~agwLNq8qWtMpKP*Y_C?Hu^
zWoa6a-u5q+H0&V78ODf9YquAbXJ9i3%*0=(3;1+ObNkYfR;y4e-Iw*fQwsCpowwKB
zJKvu)LlTlO%=c$%g4ws+p(yE4?g;_`@8=&~2(e6P{=0KG47#O<*k7rn9gqS_&<)M-
z&r}v2SWWhyr_&~smOIS$>oZfN?W0#b$?d(?XESpuL(8l^$zWpMaa%!D(^g06TGB$X
zyI}NVJ}Nn|w@SWg+a0^PK5`fmE7G|G+?-Ul&X)KACt7RzZK`-$ATWX7E+X>DU_mNv
zE+rg3E7%PU=53G9XnONtut9Si(q82@FUBgb&YX7wMWjPx$y1hT`@@)yrMW)AUneaO
z!MkHX*<$)t9WF|zKAcERZb{ELDMqU2<B+xsQS|i!@$bwAXwk6W)CnSk4ktdB!}4Z7
zg4PlT2`U93>vTEZyiPE6pXPx;Nu=cqn!oj-;k&wmAGYz<gz(OzVXX%BVGY;XAejM&
zlp6}Knc%+NlY5wSD!9d6yDVKxXOUff&mm8-vIH9gZMGui8`7l-OKhHNqpPQT`WKBf
zl?aM>p$zjT1(?W`*|5>)36riAW2Iu@LT2|Xq*bRchkc@$T8z;oX6P9YOEw?TAuCZ_
zbuD%ZuQYe4N}e$D5+@y6fDCnRe_(XEVRhv2kM`|k<O=+_S>G%?9qE<qia_v8THRYn
zs&E+4tXy=c^A5ZYP?QKnlHSvFt1+hTn)gqplmxnZq(4|^)l0KD=|0+Hh6>wrk3tbC
z?r<S6)Cvh{qjUzz%KJNAWR0mOe&W4Gsq9axgymNy(O&iR41OHJy5L(HLCt({_`y4_
zsx^XsSY0}0vVEOrK<Rv<2XQVUB-eVQoE;}b4q{XitYelVr?@K8Frl$C$CjaunXud5
z3_b0Fxhak(OPC^156@mPET}?>Vw9YF*7TC4N=J4kwwMZ*96#T+<VZLf8p(CKGnMWs
z{*05|*9n#7JCmi-Q~WzXEMfZ{)U|n>!HTV6(NUY^YiI4{+doeZe}1oNopNL(esr-1
zzio$Wy8AA`t3OAkX##S?n%_p60Y+&J{zJ_oI7yTZ6AT2clzVpe=yabSkh;adwv5@|
zrf`D?2wRgsVr@%pv~<lv;s|$Ih8wE6Rsx)cL3GLK0S!SwzcAlZvAZ70HVBT?B`g+|
z_Id9sb#fy$U8I+JmK;;*TcuRk;AIjqYEXbc$*=0ne&Lm^?~e$Hk7C=kzK%YlSgxF%
zKj<~ZQEjyXt;+hnt-~-jv5KoMgb8{?k6;O_k0GQ|-k2#<wfs<jA&8@oPiJoWnk$c-
zhDnx>aK>1NvZu?eVPupDb(msWKatDKy=qX5&~nxXs-ev%GAOsqa%^vI^I-;WT+26&
z^&y|GK4;%8E@>=@`j+7E<7}(1>6NUDoXOQV&r7p!b+^j>RI={!sv|&f=RO74Vj0PD
z(6||M$1JGE2E*o*N&%Wn8rQ6q!j0%PqYrm$j+>m{)B?CAXnD?pi#{%K;Df@5%2<p-
z2lHp?w2yOJ@ssgUdsdD%LvYHO?9qIrflK?r_~)<@OsADrn*B94#OwF1-$}jnXwJ%n
z^st@2ekisC?_&7L<bt53NowD5*_rS+`&H);cB`c9%VTdY>+4Z0c1krOHMb{fY*iX|
zICViKOpSOWUGHMIeu*BwTQOz#$;jaGmCuluVKki_7^c^{P6xi1gTMLL&u@=FY`Bji
z?c@i|aqBecU@!XT-kvkTYB#VJ!KcDz`r|lG324T!4C6Q@_=l824Ah6S->;zcF)tM$
z7F21=2dNR7N%e|r#<UTlbzIWkMOjY?Zh>Qg{EyvZhtqq=J2`&xH?7d+7#6l>Fupys
zo<Z{++4;fq=9*7ibcNrku;xYnBe%u}?UJcJk4U7G#TszvtW6t)t>|gBhMfp2?>fPC
zacOZdsYWPIzNBliWa;RB?(TW-()YQ%oSiWCFBEB|!}e(=qT%8T3?uT~(WX=f<ZNXU
zR%YI=#<))$cYEOisrExxn9EgMtk-xm2R^D#Z6%?Dw(aina>I`X4pa%4>49YYjhA(A
zNk^#ot#omDR?=8Or?`pxVdEHwr5_uRF>DfQxYA09$Z0L1B#^M0g%3%>wlXLeRQNSI
z`kL#+*hUZ3JMm`RBP?YUYFEjW7TV-3NUE!^dA;X=blnB2)l>SSW=MR)Q;F4V9U6F<
zewunF{&39*Rr+Z<5&K+@SJ4nX6yD3G43Se_b*_uGyz3R#<&o!=nSu{#qWEZQTu~}r
zy}F90G^|e;=uW%kv~p_(UCr^jq%e1AI8Twqg1=EcHW_aog0E;0ZfdnA=-OAN)(7A{
zr&h~1@-FB-WV2$V<}-rBWg6WHj?y1Xovb=hN(=#*A@#K5*sjWp1eT}j-6g?YTj#RJ
zzve67xKc`%=eQP_rXSH$z^6PYtGgqG5|9*ZS-AtrBo5J)9nBmr^c4h-**mJ$>F9T-
zcu8Au7k%3SQ4lay>ouOpN@}0ff1dm>buouWo#Zp}Kt?IjK%HZaA%Ct_U9oCku<=u*
zeDSU-IYNhM88*2~E9A8LffCi&+4E^$Z^O4;NbH3K1Yz6#Wb-eA*IN}_y!4bYS7n}6
zI_6cFCkn<slKq7X*X;$@f<eo&nvaO?WlumdgS``m`R_PlU-cZe*7fu;RA5hN%4|(H
z55kEZr63`NoCGTSPo80Pj-<xUs>f6B>--zX2$uoA`}_coq!i{*Le6qe$9oJQb&)11
zUSW@A<eJgjB2v1!R<3PRgl^H)`Zil}MB+dotLN2Q%v#h*^lIP3B&B0IjR;WIc;tC9
z9jmr-FGOV4XH+bH1w*f@qh^zvbVAK1%Us!?1z7gssA20Sz#@d=yfAA<*(RqutR30>
z1BbkS@KMhP^}Ow{Gw-weCKEH)dP`splCT%kZ#cCZ36Ja4Pf<{q{X+GrRL9O6ak7Kc
zLTkM;+2EMzA2_+S1&n*IMMS_s%}*=0-iUx8mTV&wQ-bk%av81Ic)=2?*@+5w&7xy#
zF~rP4X5^6}#_xCfzC^Hi*OlGZigq4rJ8r!T@nbt(2Lh#$iWDO@-7;Wtv9#~;(-iZM
zc+u#n847!#+f$`u9#$!`pc@h)1$6LPc!B<;%3eZ=vtcHIBfJ08jIN5^N;#r#|DMQ^
z`Z44*frUcN;k%`vR2afR{-zQgH|rH-jTuyqag%EjPSrZ8Ic0Un$;fLhs?XED#60nc
zvAwJ7<8>gD5BaiG60ile=$R<8+|ZemMqz>a@`+^(xbX~15W<<CfSQ(}Xid6#!1uC)
z!|a^(!ynEv(7%%>uAk~R1=VQnG2>6xj!D%Io^5ALk%5%V6;~4CSI483<O`nTlQQlR
znLji_Z3Ch54ZxXAXRkDW$6$%?gSvwOg^t@N*H%LpilsEsCJyZBpx0LR_9TO}s#=&R
zJ7w{HCW+#ks~@BQKt`j!m6vDc&MWDuTF!B-E}kXx8!&@N6NFW9K+Fn5V)*h^TiV=$
z3PrccKtEH@eR#vRa!1>;K5fjT7JbpQY7m;_W93zjG*cP6_Z@;4%R#hPe~9)@XKh_5
zwDjF{a?67NfxMtaXb7Sjo86fmqfioi3K0#DNSw<)D0{y<ThO)m(V~UbI@|Z7#+lV7
zgCb8b!_EB;{eAG#k@OCC-076A@@zMrjLB2o?kQG*mHbX?&&*$YDP&z!a^Z+0OL=$V
z+rwV>MW|X7kr7ZP-EVntGyGIDz2M`vOM8`sk)A8f?>~7o5$eV*h0f{}x0A#^6(32}
z3k)e6LtuHUNcyPX&JTsAAL@-!NPF{<;53t`PM_pg>nLB7=oJ|B=vneTFSYS+(9y+o
zb(Y}WE}cc&s791ej<|y1u*c-q%c@)k46NHyvzBiz%80`D;xO=e6o($?LRTP)$l+q!
z7-OZ?394^3GQ7J*1cEMTNQ?$nn$8CCS=zgMw*MUbT1eW}$NA{Q!M#z6WFGbcKR_-M
z|Gt|(_{*FVf!3nC%a`Sznl!D?OO)Yw5D~*KyU^wO%j)}(&+K1(wM&J(;4va*y(q8D
zqH>maKg7s$mUQ#6RT%?<f@n;T5Ay3UC=i_1e_GNAii;1TD9i1a!lpC@hBbD4DfPY8
zeK|#~=G3|GRuzi6%8zbsL483nw>#>3ni;C~Xh<*A1WG>%fZX?rxzuW|RFk^Azrs_z
zl=jL;ci`9C1nsur2o<Rg#NP=DT3j1R{Q~KMvs)!7jA>snQ7@@z8%pUVbEl3Nr+Zb@
z)r}L5=4B^zXr4;c$h#;#Y?)mj?9te_;JM<dCNL_WRmr5G+}h0Rs=r@$<(g(*x}Mcq
zDVLoDqM>4<X>U}27L(?t7H!+hPq9{aI<v>4VgJ!qa7(62?5Ea`LB^wGYZ8K_ihJHi
zzaCw>GSaj5vvX)kwdWT8JacFv(1j(Q4uD((n`B?hd>rdT`K4xOjyU_#)$^WOFDY$5
zx#Y6H1Hb2+a~Grb_Nu7W+@-FH#leB2OG`gTg^z1~UR_<ymB@;4v@&4ob{VcI34EtA
zuUguwW`zk;IOQ2HAqa_7MnT-*B)4Wev2=foijfs*Mw@fXqP&vM%4*Q+6;-{KAwLe!
znE+K|dQ638UTd~=FkhbEW&aZy3B3xwS(M(_A>gsjyE0Dr7_9DD*S<e!_2Um|B;TVI
zf5__V)P-0cd>oBufG0E%O#%*|Gm~XSlH$nQc+Y&oV)uZsam=H54-vnDta(JcY=-NZ
z`CAs7lOdN&r&9L?#1!M#L~T0ut|=bv_FBB46iZTZfed9H!|{B(X9V464ceXJ3$f|0
zrP3yMrJD8>^%#8ai2-N_7L7--bhmPlRoQeFnnV1h%5R>~5P$6))UGa0N-$+xQr@$!
zv=gpiz>OaD<ai$~(P}|{c|U*Tw_w(f^u?O`;hHpY0SO@*J3du;ui5W8HTiAyncP%w
zY^nB#Q|rnC0IGPOo@k;XU2>Rog1hywd62f$TWxG+^~r9H^L1t)IhH#jE^|)?O;prB
zc`trwFcaPGFBrY7Yco0fN}@#TgmtV(mM(Ao9xd~hjr=tlff04a8eoKAQ1n^mDR-?t
zU7*9OCS{q7Bqb_@_a(G*E~iF{c38ysDP0HiZN`*Ro77-7)At9e^!T2u0&&gvl{f{V
z&{y?lU#eB$8wpi4lb&a1@33@c1JJ9U8|!boFeVQ0;9u}LEIEb9=lgFiee8`M=}4Go
zd7>CEnBE!wMi@!c`2|{46FMTqsaW$M3&{m!JOLVEDuf@nk;}gOz~>EmAGEX?2I)+J
z+(kAkH+iww3Eu2mskY|}U-~(!16j%`1fukvB_;NbjHt#Z6ZwQ8x~s(J-u^rY$rLX8
zn(>2N6QmOhnrI(^^27Ew!;j*RpF)z(R2H8eG6UK7ZVV(*X$bdZbhk&UrQAH<i_@jO
zO3Wp{?65HoJ~=GeI0*P%@pjZpYq#ktK28kgA7XLf3qj!7C9R>tzl(m1JaImh(FXyV
zElIxDNu0g)p4wmVLQ??JM;>5`{HLz4K2xvG0vhpu>FYHqU#~^;aQ!pWlqe+qiEvH2
z?Lqmr-h@FER2dYQKoxT*fo~Auv##GtBNf}4WvJ8;Dlq<jd0{$T!?y{5#wbHCs*PDk
z^0$04CIDTiDJhQVHlctNBn|_$3r&Yqan)b1W+TZPcM=dM>EBK|gl@j?^_xI|#yF%!
zOdL4(uPTJ^2%|}J|48w>sLfHI)4;CCfySoIG_(tPS{KC4>d(@Zq{Nzsw4Xr;BF&kR
zI!8g$!%-N@wY}B|Wb1{@o>c(YA!3RaMd`*>mBXOxm;G4(=}D)w1<0j6EQ|)c7yytS
zLG1f4#8_8S%_yXtFCtmyP<>nRJV2F%8+)&ike*Wmkfjy{s(PVW%OXO}X}dDfRWg8R
zLQ5lwL|y2)Km@&2Yv}Cu&4jh6cgh&*YK@43Y316gJ0cyA2-gbReU{9F)N#9KXMxqA
zbKN3F-(hVFk~%~hSQY{}xWp3Vbu0n=+75-#3m|OJD82fF!P^1oDu1I3eX2q82Sk0K
zynE=jrQ=xJ$1Y&qm_sI%K}svG4Qxm63N(+kUO)3IcIOi-mnN)0JCwi73mOGppmt>D
ziG_~Y1Av5+Z+q_@f7u8ErzGH_t>PV^?M8c?dJUpmX|XsneEIR9C<`>rHq;2$5Tz)F
z(DUjG(GC;Pq#wv{QWi$~1$UAvOPBTYEiOJLOv@=F=~>XbHEcxO0N~OEE??`IFL_$F
z4Jr=jK_lCTW$>#M%<U`@us$43g*N&{_aS}96NLTB`9g@<@8$1pK|1^$q41;CFwX=t
z^FJxbki{rM^UDZOW&uyHn=cPu4mbH7MD8CjdNfk4FF;s+CwU2YFoh+6Nb!3Q??=(y
zK)O~FmyWi9LD3hgKk0(fv%Vs0#>{RgZZ(0-*nCdOSZ;$gE(5La>!MQ%RryA!I55uz
zEPyj;lRY7YbR+S9_nM28#y1J_uHBhwvlwj$vy3}i>rUI|wu6j^wN<31BJhTNrw~kh
zYi#^4OgxsEHf1;iFKoCa#k@F8Ktug2A}%3nre%{ASSSv+_m|dugjXPyw<RK)yTz7H
z0KN2@ZoB*#AbD{7M&U+X5qw!J@8ecu?{0P#3<1B}6TtvbbO}Prxly3yx@LRuTSVey
zt2)JX0BLz<BPqr!VSrt=>lD#fiUFmfSjOz624jhI*J;QWT!vs;Tt7(i(yO~@bXW);
zCoI1Z0qAVWVrbD;5T=@E7Csym3u+H}fa{}teYK!t1C7-O`<#n=(C62F#*L#8j2gXQ
ze_V$K#eG(reAVjbPDlam+6s<zXCAigHYoxi<mgQd=`&DL7)~vlaY#k&w{`)h+Eb*u
za-Fc6kGSmLS0O!|yJB{O+56}ic!UE8I#+99#rH<Ynm%rP8o?$aE=jysH57M*!GDu)
zyRhP<DLT8N$zei~g>(bHVZ(GRNrARF_WD;Eq`E^9nNl|edvt$!6{yGi^*honM5I5?
zRur=BD$emtw|Z9DjKJlW+8(ls3F??b5cW5*O!LTh@Rg~q5GpfRuJelXUX=t`#tjla
zZIqk7-Vrd7l<D~h7L2Ok5z7ToQFufP3lLCfHV$8RDW9l`0`(!$pBLJCz3rMWv?YSG
z^b&blYhU5G<<8KazO=_@bSp`en|n{SRYe3(vI3vC4o4T*9f6e7D@f%Fpon374@^8O
z2fPG|P~(U+a;-jHTAdf%+M1q#Uzfu*BBPAwBS_2s!ejV?iX`=0LEa}CK1`v@CJLUH
zf|`7(*eztYutTS_6*<pyAxy!RjQpfHNVtC^qhjkN<TkJj>zaK!uHilw_7-z5E2pW4
zf4Sq*ngeO9)V!S~pHF;&bJu8F>L<`SU|$S)X9d7w8ktxv(6V~i1i4WRDnCMnzZ+mm
zbFu{uh}g+^SQ$vpcYMrk3h{;7XPhG}oW#5VNSMQ!3;*#MR~)6v^b-E!C}~UH-)Eyy
zmNM+A0YJmW{w!!4H(}$iO%a9{Tme(wVe$iDt?4;RhxtWb<7`q3w}^DFLW=UjLQaHB
zS%PIalOw$JmTAmnXAxn4;vxt%Z{HrT=d9h7<x>fQ%Ffkk9`D3!qW}(2s|ZDO*B?2;
zo5$jl=xXRH__TZbTBjZ+!Hb^IU<y2l!j;S2?iC2(&^t1F8G)inCE|ByMD+HdwF_$W
zt#i)5kV^fG(1J;_HH&?E{k=D~?QZsvfLQm|3R>EF>C96r<%_eHT0wT+Ggn+BDLzF!
zlE;*b7&`NnAH+%dHhZ6{*FOB{;qN0i18c7|3gr$BPMng{)J5OK#h-`g9W;+6B~>6;
z=`&goW#*})77AXn98pkug287~zWS7&v^?CNcjZN8|8)chp9hHeTkIW^VWac0^qs}H
z0jJTH0?$XJNgef*mgJV0>6W|gg1kd5`Lb?talKFByzfUms>c$tzaRp@g2!biENB~g
z>m5h9pBiV?zw+GGMDN-C9WosC7AFsciPQ!SbZL%Hf+pMe-PbJ4ZCn&)S(MtY__Ua8
zufhn*+C)cBi*C`^ZX}*h7d<_kBq;&~5rIJ{b`;<S5TzG5>#{k`iY%Mn%+-;64ALxH
zQ{KVxfxqfOeB70(U(|aFaVPW;bVW;K{BJJ^sfyDvHM5Cos&RloUnk3~Cdow3OU*w?
znlCSD=cnaN+gF#B?l^$gxLDRE?$CW7*7{A#lEtpcI}x_W_`<YvjF|7<NrW`-XXUCs
zDaFETqbsc*jTr5$Ic=MwB;!S>>#t<k5XTSPVY?*4lCL<NWM<#_9m1s7yI_`L1jK(c
zoSvcX?0445UWJ1}3||=)rQtsGo|sk{E0~(3jk#xQdU-ZglRgV?NyWfECYslf_Qgql
z#T9Ds^%v#iF-s2PNs&f8)@x_7SfxC!pCJ}aY4v)74id$kk8G87>TDDoZgdMzmkR3B
zE;6{Ec$OMD&=|gpO;zbFB77}{%**>2&drl0gf_)ELuN>`>2l)d3inQ|CMOmDUCo~N
zkV<GQuFOwM+<(jj8|Q?iOlRmciJ1|De6yesKQSt^b+xr>)hq@&dg_n1fk-}@!CxbT
z<qIO1DjFRF2X?n$$!<)mZ<OEG{+u|hk>@Q)IXTAXrpp(+TLF)O#(uwu1ui?v_5g<i
z{ryBMN8l2*o!z!i=+!b5SDh0Ep#t793TLKRICQZjdSy{<M-(?ozkc5bcr0B2q}EoO
zZsdVr7u=*t8PHU^N_wR;4CV8D>5H9d^0|?CS*1(wLGi{U59#nK!?7MnkR17N%@)-x
zQ3|8YV_KtzaYF+)=!KSg&dg7A+k6<9TLa-`zi-@j`eXBaFClZr?DiCj$e*{zF)v=M
ztv4WsghXNF!Uo1P!?Vh#6qbTK_EMBU!{aO8dk-u)d<IIBB_emoa^(>)Za0k8A?l3s
zsro`0ZDJ5J+pR=IIuOzHSElNdk_OGm?&q2P%_67k<fn^`poL5g9z&*u&D1j{Da=jw
zNdIg!{(#o)m;=LEcsggPaWZdIWnq^nHgxV<L4)m@Bf-n7t~2$GJG`O`Mm7MwF&*53
z0MM;G2Euqt*h(Ts(JZGI;2h1zOk%b#&eM`gT_E!$2?FWQ%(i#lae`GL38LI$k9F0`
z$AlXekP<p_?7w)!Vs_M;dgZ1+0xaD}XQ6NpoIUP~OVEb))|{|H+g&kuJnT7ga7Qz5
z1~_BnngA|#P_1<N!L!3J(lRw|o+VLyS`dV`!3$TS4&FgvH^BRP3b~sOo-NH-MUX#N
z&T$&nN^fZ=*;he_GIIn{$Cr**$8Zy5Qd&ySBp47eI}64D<zxHcnPqqj@K5wB&jhDO
zxu(=DBM>nh`K)})?~qnm`vHjLt<O64HLZwMOXE9F{j^;gJ!(p_39%{MAta86>9~F(
z$)Mmv+DZ!pQvBit6qtq49tB9kY@RI0Ur!K1<)f=(>xh_*OCldLDnYvY|NiTqjt)03
zi~WvXd|%L6thPf<vcsJ1R!t7A_icpAn24}-ZwbBq1z~|Wv{tS<b~kVQ{x)j(AL<=(
z!}ip)>How@>wkTPFp#y!O1AD_(GKX(LU42zr}Nj)84+cvpGFc2kN*8gj0fQ=elLEL
zw>P3-kP-rh@O+1={qLc#N5{U5a1}--t=|T8V|HF?Ll{}v`Y7<XG28h3CIn=+;}ACV
zpJPW?!lC`6!ucO|{>Fo&&%ss5Z)yHH<1ooTm{T}zJ#Wp_7ygX6Cl_u|(m%&Z*85=3
zQy<aEj(g(@9KYm0NqMv2sfI(1Q7Y|qq5<_QH4|X<<Qx&ZC4<%lD&YjUQ9%u6i?gHJ
zkQ@OQA{(U`dW6^db|3OqnN-mWb}b0JXAV5WgX{qr>$Cfn50}S>$+~i;<gdwQ{Hc%7
z6G4frcrR=jO5acc{m?HdY&(&csAQ5DqE&ArI|t5tIXObKET6k_g}nX%<2`H<)WrHv
zxIDUtCDDC2A2;Gnu?36PLV;MqeN9*1Bd8E`HWW1?0SlD6Q~8!diV$qZd6#?VxGNxD
zjRc@vIMU{#UXU#L0d|-o0%K+FHa}t00<q60jwnuKy}MFdeCJ(3^mbhPUSz@Iy@v&m
zRU-)4tDLpEe<8o^Fdh{^;}}nQ*7eV~0}J3QFXO$Wh93$4kylYgWLLgwJ_GHbe^x*I
z^*6r!*An~R_v_lx5?&AxlU%<>{Y&5Gl?*Il(Xg~-iVDE~5P`>MWedozsXe!M{(HuL
z3Dulb#h#u3yxv$nBC4)$Tf%j1>*s%Ot{vb_`PbgMYkzzKu+6M}SrDuut)*lL0Io|i
z7pvdnpZmuzaIcY+iN_CNaVLrr70Dkr>1~nr9YC8i1$nwXOGcP7k<0JS1=#%g&f|g+
zH@f*zdrshuE<~j%=Jd+qERFMJX8~TpKP(^#bg%T4v?U%mC^7ey9-(%QCWl<>KTClT
z1Lh&cFn~HZxe(pT8~~kf5m)D*uc+=!$?KbBN<^uKPSKjO^erYXBLXGR?U4xGTTI+$
zhAtwUs<q3TG8#bBxxH=QCB#`t71o6qvSjjZGxM`!)aPVjrcbs$ALFXys&%?`xXo!N
zaFLLE0Xq^Yo&et6zv~{N0EHB30{Hh(AR#ne^HJ{|Jg1q+r)caoy;>jYV#-+jb?-$O
zHdIS*3{&GF7BJ3AFlh&`Ng=8lz}s^>RDb5cI*4WFqyJ|Nuj{=jdV51(!|Zfd=Ww^$
z)PFx@A}Buy6kLDxNIM1H8VvxZpx$4FlnpajkEm)2E>N=-eK?&Z!R2SjT?4qaf2J`Q
zl5$0CBRfu*=lB`8;K|FB{eC0N5%UZ>^1imIuQ{)lU!p3RV50f^E#MU3EvVsTdvy|N
z?YmQVj{PA?g;?S^4diVHKB~T^u)nWV{g0{Jc$1)`M(wEhjLr9QFn_I6vU2w5q3eIb
zeSI$>w0iv$sD&_zS|{@VjxPzucif5diHnAxf$Zu3{v5#2yA?%I3o-(=PKW;c^V`VJ
zH8`JK(YRqCd-gv+KZD#|PXx7a{#32gx&Qt=0e*e}!}&zp0PcSNzduLrt{{Y3cm;Pi
z{qN6jB0p#1d?ITAcbEOo&wpt^R<B-Tu-|{+Kks%5k#>_kidxcAcrwcKpI7rF9j=~b
zzoU)t2?nR|pQ|GxP!6gOs--~0_N3l_R)#53Cd<?zFC@&MTX*dG`^BPJ;l*3$AI!EX
zATN0o@t@rn4u`rqAG7ax8a#U6e;&OSIe&L0Qd_G-e~5VQe|FTbWnlx$rhw*r?=+sl
zzMBjEpMO6IcXvFw<L5)P*pGvZV~5?o1KjE#GlQpXCY0RAsO~66`uFWpk-Yd|8WK<8
zK;P{*$OrJvQ>2y9;0j&EH)t4!A9xoX=}-s<YwpsceZp@<@XfW*=3H#ByzO5?6nu&>
z=ohVdz~nwX3?$BE6Y;QHh^!(Am9j{`H<;pWmI4sUQ4KAbL75F4z=n%MMh+n6tf0HS
zRshj{=ryJE@(zO-l462XwtS^7v*{YJ#ydUSW=G5bIEs5BSqBQg|2z%jRn*MK-!|vB
z3DiKA&v^qG<$w6+>$m$$8SdAW_|CviKPbq_y$Ol_Wd(X+8T`w%uC;!SuGYuQdb1fP
ziig~%j7Y)BaVl`R?L6>&o9mMgLyzm999P*Vr14xsvi^qJ?VC*Wx9fvTw$Uqaos(18
zH8NQib7lWhjQhKa!O!mRO57)>bqTHG7na5<THIsZC8{b~76p)J^Ti<!+gFiK)PFuH
zcxT+$xY0w!e;NUJ0%}Achew&y{~y0~{jaOAPL%xs#Pa9o|8wd8BPRr#gUk1mfq;;Q
zBCBM81?njx(*Rn%lj|u$OIAe#{PQ;tAqj*;Wp<kJQXA{PPu_>UTk=}YaTIleq6uhq
znFF|*ykr<Pc!GMFD;tthNa5qi#`>a0<Y{_U4$x!KQ79mB9(SJ`qq*|Uf*B>XTPNQl
z;IWk~Y2%x6ps6HwoVyVG01-Ot&O!mAY~NLCG};&Vcw07hAhws{X*}ysb-AsK)XCV7
z&)W$vNd)5_jTGF0DEcH#YNo*I`Vf91q**y-kE$ie^+~z>8bun60{5P6RtDXN0(RQF
zmKQZQs3L*bb;Uh~8(fDq6r$1vk;c6-AKtAyXeZXsdvYDLdzicr>siv@aRq2{3IDb#
z8@>S}9~f9%v@#0>-ha3YXlz`=l}(%WYtNyPBmUXRXp!GK02_PJ2OIYO-#_-jrXca6
z)5*(XZ|tlNzR}~rI1^n#BmCEukPFwA!NYwfqUIm%<Wgum33=(s%2f9sUtbe}u<fbN
z!S_3-feiEu9D(%jOCELx0ztN;6$<%}i#rpRZmB`R7dn(AFJ-ea{CRmHEWDPj%DH|D
z%+eI4iHc#ONM(g`sOcYS6MS8+2*yzQn(oVWuWgQ`h6|5v4|wc7ap;eU*as64Nygny
zI2$(nk?6y~q4WoLDUK1Il8)?D_Z`PmrCtg$nyG4B(-0#@KsGZm^2WD6JZK+mEeQUc
z_3|T?-)H9U(6a5R>=w(+)XhuEY=GF&Q=R*-=Y{B<{;g|c4l_8LMFhL{T&}m{pB65)
z=b91M$ps)V;m96(`Cqr71Vd|HE#Jb{P0P&p)3#fWq8BQo!V=D7wLW>+U4O0yv?099
z;XuK@ZyK3;q@Ss>+H4qnGmy>3j&Cy->|>z1^BcHVbnXk(Swb2@huw914&2|}pKiVQ
zN_;1(W@j(qZ1y|tN~?6N3n=C+czN@EPGRr9^!k1{hzliMnjc$7*TTyVJd_Q63d^{|
zdWGc61U}dXo93Onx09lS$_|R&-@$>nJ>O++sat?2HD(!>J&fI{#>}hssu?y(hEC24
z&{~kt9ckG5YxR78yxzQxU%k}PvPVN9UF;|C_c{7)%us7bJedO=<Y7<h4FhHs^MI0f
ziSR}W=?}Fx&v@kgsb9;l*N$v{rAEa!%gvF-aZO6L3v_2QU_NwDomonOGr!P=Rn+>s
zCM$^c*s_S(T5Ax1(eg9RR%3JFBgmo#ZYaU{A;Qo3!lCaMThcTef{z-1wuX{=th-!y
zC(!zE`eL&qUTZ7!sS=FtzXY&+J!p77kb*g-XNUMshiZk!A*(wF*3sqEn=>UyuPv+#
zWCttp&sM)#w`N60pOu8veJJY`sr{MI#s=Lx4fWj!fNZ{{Ec5z-i);8`Rux4*;_I^p
zmsxR9R?~_H$k7X)YcCPxUGC*Rm$%IPS|SKZc(j!uh9um)TwX41?&>;}Ud+}0`j8~3
zCCh(ysBumvkEE^Sg?nL-fNjv>JK&iGsewW>ugK)lVyo6H8z2DTKYOovB0Y-Bl1nyD
zaCAQk%CWQkw|axcBz?CY0WC$_L`$3(F9be$dnTcb=v}38&kd?6?_rpp&S%tg_Z!~J
z#69yvs+Q`mPf_Fg74Kl@U_huoBV5eR$an5_<i?tnJ7)sWly$ZXQUErce%L9axCL)(
z3&ZwonnUW|E&Y&8{H_$8LXD#S3GQ2~@gQe^fn`fZIPvN-<Zcp|brw!xVVU}`eq#Z$
zkg6Xa?Z!S&-H9|RJ4lu_;GvdFYr)j^DN)f=%o-Td#KPyWDBkaR%!q|0gX*pl5c>L=
zhC9lnv&3x{Z(KCbvG((hs^|ccYCLfE_LI2!)+-O{ku6!nbRAe|M1UlClGKOA>X2SX
zGV(`QMJ#&L(@NM4rRkIBm>EN0$)1giyvX5)Kv8*qGyzH8qh|_gqaAF}hG$M;3OLe@
zQ1cmDI;nmdS^mBz)G6HBqea@VX&r|?)M)9YZ^4097I5r{@Q^miObY_TK|4*y0XOcg
z9qNv}#9**-Omc@Mo7JRfS-P`1q>D;mjVD17U-YybunCF5yy`>Uv(q&3w#e7FkV7v@
zZfNg-)K)WRp#vNWw=|=94Vtyfh$=;h$oHJyB^lT<n+Z4+Km9z&kefb-J5a{Vg7^Kg
z(2bO)>1*K`!l!uwR0I5Tg%C}<B&ci+0FJb%efQRnTlheaj~d;}>U;fkmL1<v(0yc+
zdjJP=0BM;B+WNEdnXCHrW(J>2ozgAavY!z~&YLk}tBqlcuAbzuOjUxGIz*&O$TWoi
zti*Uf_f#QhB_aJ^oremtgp47xZQfTK;H*OeYvT7|+sL|@1nw@1yyV^&{wK{5bFc3+
z3nMw<F(u2c5y<uD$Tlg^Gr<g_W3V~bkAC+kra&;2kt7|uY7YC_1HP8l3se36w~f|7
zKgqeGn}Tll-W8^!MqR5{JfL%8+p`N1ihapujHs;_&cl-ReL~CFci8z7#>0K+!?U?H
z-4$>B=jZOd`uy}wW-v{uX<eX9QzSz3#Y)1mx-N9U+ep_R<bmW7wI3?J4^GMFUxrIS
zeqBUXb`&)#UGZ3Iv9=sGuI(=O^lF{ccEDE2m3)80k7f+WdjD5@Ul|tV_Vz0XiiEfk
zL;(?06hT0dkPuKxQo2JVq*1y{#Xu?P?q=v_0F#!G0cit<76k^7koR7=_xbPl{dlf(
zo$Gu!`-7L8y_tDtp7pGC-@p48GKXI8y%wBCjz`56z)y%5ZToXvxJXv<X!w}JOWbFv
zSH2CUDU)pg!Z9d5eR`39cDS-2?vhC$Qq3)>;48FQ4jt4uV70^K`pLxAVQv=JkOalW
zND9<3pC&ay<mp5c@}B!Fi?Pok@D@+$4Re%VCUNe?!%8csclZq%TU*i?!0TN%ZZC(t
zPCn9l5mLV1cCSOq?;`t!BBc>1%id1JzUSgE)H@L~w!38|%J&$mCVyz14=g-nE3cE6
z=LhL%Iea_K2kLs6;S#9Xz_43r@NUpTX4~ZfOOh9uSOL5EB*9ZQ#d^mSZEhRG8g;EP
zDztOwb|6NuN}ehpkEd4NAFC-k1mA`ECS&Fj2?J>sYlklv*wmG3e$-$?kA2w7eQmjp
zeTRP8=<%ZbH$KFR`2~*rlH+>uk6|M<a{odKkE4wfjEde;Icm<sR@y)6x4Yw7)D7PH
zul=YFsl14Mn+;kWWanrBtAi+CNVBV;8G?r_V~*r&6~*t2-B%!cH+(CbIsoOwFOXKl
zv8=<%mbPr83(iZ}n^S`(IA50xBY1-R{dObFzYeiK6aXSao&`Wtu7`+ra?&!DVO8G5
zHiK^7ygc4B6vBJXfQj?|db1#Wlpkn*tJVz^ymQ%GfxcIS%%z0ZKAB=2rhEg5QF1Yo
z)*T&<?+WDjyx69x{46ohpz^Z1Z0^jB^`_qh9+5{@a5@PBAEco;M;;l+<G~|J5H8kU
zMLfQrk%*)ZQM_bq#hK-If2}W$3|i>)hxKQK2)BpgGL^|5%HRI-`jKv+0oSiA<`MTd
zq;#=V5c6m|-Y_I=IYbSGrO#PQMH<qE>-NA7a-?#6=|0oEAg@3%cX}Q9bFFY&Hrjv%
z$ny!Ws!Fwo@?eRz-&6^npwu6zvg#fel+}9u6tvo(p1;#OZnWYA>pYBsh;Y*Y)WqRQ
zHZR&i{X4)e>E7TtY?T960xtCFNZAe2OH*Tv!---l(NGECbvZg|4@fT}#pSuikMF(%
zQ4(c!vabtn-lkyd?vJ?9zI_W(^&DY*+NH!UX|(h)Cv@lQHYBtUnS1eO4)GlgsG!KD
zp8{Mlgb8JE2py=kf3u%jw8wdAd`vNp_njC&cycB1k72)Kaeb(AsFF;9UD_PH0nKB`
zQT9jgkGDEgfd5y{TUduA2g!1>42D@P%3|5kna7}4B=iopz896jtxoOF^0ONGs&y1%
zty0MQQ5EypLRMgPRJZhwwAqIR#FGlmV;zRWyn0Cpq`YN?FLvO*jWp@zC-_z*4d4oa
zF@|MH6>twW8l?9&r*xCH)9Q}q=4FDY>+iFzMUM0{*BEYKzC`Fb87N?wd{%ocKyMXd
z+j7#6?8qeEo0+egfC6Nb4ySPu1f}Wg9xUr>X>Fo&7Jsa*Hmd>QuiG%6iMbU>mupa4
zCIlnCE-)XpQ)BYU%YuDvnY7<)z8z(5Pk*axsU0(dgeQT|YDaQB9+$R%3l+DY9<4R`
z&^s~;Ttqx5xnB4VD;uk*6!0`5c)54kRBw*GZpPsasHwjV=x?c9!>!O1<aU8~Vn<a~
zCGV|JKyMnQtA`&~i^75zm>F<F1A8FaJoTcq%kVdt58(Kr2~}uFEhI23_^gL=ADhX6
z0!JX%t=mSo46!`BreA3C^&315ai)AG4r`=5<)9&D-@9{$g{s-xkUczH+o~sg^G#$W
zpK;hT?yFJks;?6vMe=&SfD0r{-$1^mQr=A`^-jbh*58)7nl6XaeC1F5ojht{U-}+~
zx4`_;%Z#^Ycbt4`@ZD0$kZbp^;?<>zi4V0ajhJQ)Y#qh`NM%*knl_d#9ONpn-_yoX
zGhtt>;UIs01@5xh7Gv6NXCSWWj;A;4&g;}S>yIFQp9Q%6-1mw|e_*iKL%Ih1xWrlL
z5~@sTx%BuHMED!FAd@h}eLB+emT|ISuaMdncBlw<#xPGU>*nnuihU>-yIo@>Lza47
z%br3`f}Rr(<_T!WvT>bq*K~GLuRK<tO-hny*=4sAa-p;z#^a~O8Nz>{_^0fd6)zBV
z3{&U|DZTt*Q||{BXZ4LWzt<u;G~0`DnMQE3LvDv*N>bSC!Vvu2jKU8(Z3B}grn!#f
zMo~35VTFjtSv!U^f{4!-u;0_Tb3%J`sVw?XZ;T<8hwJqjbX+V8bGth)-LEE>=g(B_
zV796$Wkyv5GK(uom?<3??dQFQv8<`^yl&Y0=n8>(o}!9(c<yw3_D#^^l>kqSY=m>Z
zDzp*DBx<guxn^a^)Kg+d$F}Gw-8}~OCgI6>5T-tCq+=egsp=3NXwCM$?sFi;LTJ?~
z5!XVW9Z;w8qqKVL(r;C)L0Ew>d#uX{JJ~=ZK`aV{+FgzBy{~tuvhT9D`hY5e9LlAy
z+??gyVKe6`rGE)j#0TOh4qUDwB=6thE|^&&hiH!@H#MLgRn4OuQa1jCshi{2BGq(S
z;G=&5J#O1yPngLInTO42Ny|`(xS2M&3d-bYH6^WQ;4PeR7|XZxGWQ6nG|CRba~yZg
z5axTUy7iGsNU%ZYmA_z*X3w;zolAJ<h-u+?nO2cr9MBnuKJ4OLhr7XDjP`^cQ{KYv
zcT4wG<lS(}5Yz)}T0$hF^hR87GSk||gYU;@6pY{Q;&N3Wvr@KCKar11SRBsTZ+*4W
z^OZrbfedM*aL90uYwlS5T_w3D|DaHtn{_fQY-JBs?D_krrJN)lXwhcQU(tnbY%7Dg
zJN01IxKpZAnMD^vf>oSZoY9Hzz*De?(yjK8`R93Zd+0_jizDVKifM*aU~KT2c#hyK
zed2>DX`H9wwbXdY>1T|-GtBc^OSpSVq-o^8M}L@6(S`6I)Lv4JIM4RTaRhAOhI5&L
zypT@eh%LVG!#N-|NX`=puxUMVB}BATPZuIbatm+VJa`9MN8cX8%05jq2%As-STru~
zy>U-uP{l8&;X1}mq_UT^F5V%*@@Lg`?7MN6!Fp=K^LhziV9Soj)TEpvT1p%>BdN`=
zD*L6jeLl5GyG|!t^z>et$O9k9qqjZPjv%f3X1!v3Ixi>Lx;)!LWyOkv_`+X5)?{o^
z%Wf(~?G#d38qt+o!=k;;pRx)Z)t<4~cOxG&*58pke<SdD@pxJprE<34T<~-vBz8Pp
zFj^ny=qTkRuP}{rZ_&P!2fj~dNAx3E_3V1@M^Sb#dV7Y(EWS{^C3U{XbB8dPQMIv^
zmWZ=3Z0M=IsErw&;Bn;1A|7XnadA|aH=Gd}B<0_+uZHsY@GYKIa#dK1WT+uXRo33*
zD0s^iADTlD!`WnBd>qdy3l!^Z3igpaPKNh|w|l#&oV)Aw&51*BFBz|G>yCKsmtpIh
z9Vki|Q3|OZU-u9!RLuxlH<|QpFuxhO`d**rny>Ti*|qSI_+;`+QT_5n*l3>Rwa9v(
z$?1XlG@&m03pt}}Sv+6$sN*6$p*HMi+?pke$E1cc%3p-HCL&oz2}Rnal*2|V$;i&}
zXlL$YsW;5nA-OA00<%wt(GrmskEiq8her0NnE8@TY%)IXJqLr)?BX??0Cen6Dy=Zx
z$ld;Ob+19w=(8a0`gFz0Ul)Fk$EZ7==i*xR+g<k?q8$A>)@MmBwxV~(Q0k`Hsh`8)
ztzJXt?xSk8=ygdJck1lv=!m|lRY6utyDX}`f`fRhzb43{xV<KSQbT?-y)$}lsLLuY
zq=zpth$W1eh{_FjMAV!`t2e%`lz6Oxd3Z#98&W<$uFengq#3fGapZl5|4^6ytCafM
zYM58-H@gP^VJ-sAjs-!h-b~L^DOpclr$ux!OmZPnk03L@q}kx{x_4_;OiRSTU(|83
zHFo78cBFo5e64=EMhOG4vyRtx8U>X)+@W~0OTUcAZ|>Ntij!;uDltFd^erL&!sX}q
zHbwBUIZ!`0^g?j7PcSO`G}(rR#2Zh;R1uG5v--m}*B@i+j1>8Gn+ks7Yvx?cFX6&w
z%)o#kNAPJV{@Ap$_euS0sSKBCc;X`0P83+mozXVSE^Gw@sA4Y;$N7`(3;4^jMx9?m
zn4-WnN3PJ(`f*6wEK_NO>(o6mqxQ#em0~@_zmJjD>TxS}oWv8Bq?!|y3E*h=l=xyq
zyj16<r8#h}5CkV%e!U9A4xzX4BFFJ-9|cGv*m|-Q=)awOJzAj6Qx1l@4_$YGQ}mXp
zY@Ukm?uSIHfnwuRPJRsbg_DkRJN+unP_A<0T|Iot9H-jDAG$DTE%GwA42MSZs|kF|
z*DM%zsW-Jdx`c>R=a(Zqi%gMVF{*1QDltM;diUhYfvWhFd8L}EwH$paItOSs4H@Lt
zo}N9t(wMUKlmhxVd%9>Y{zno|r<}tLo9^yC{!Q@s|BC8pUbniL>FI=H1P|<s<uDwk
z)JSb3qg9kRlvMFMhf}`Q`SqdmvB)Qq-RoXqI^|6KjnusTrR^ZSw0JC4KX~9U)A5yc
zC%`rY66!{jTa>hZT1YlREE+JWy<gf5wm;d1j5?`gaUk-ZSwC9)%`imnnpm%NKmqUj
z`Y$b>*E~H~iv;JMl}URAbK|Y9Ry>R5tA#cZ%QLb??pqA*N3AREdW+b5e|ZFmkV1ep
z*CmtBt|fLO0vWc`4EOu09VVJ_DdcDT{hLUR*DVyUeAyaa{XRCro$(Wlc1{!t<?)RF
z<SpH7ysU8~;08?(Ln&gPWy-2!Q^}K8k@>fJy{nymIpC`dJ5uLU)GlSS`pn6>NN)j5
zl;pDkrX44WxP)&v@5^v;iQzG}_cP{actckdH6JLvZgAb6?g{wweynm+zs5WFqIPqx
zV9~Rg0bd~Wd)&pEfAteiu=ks_ok`b6&0f*lIAa@Mn~f40uPz7(<6Zr<=mZ%4cMO(L
z_3IB`Em=L<!>1Ok!{+U1#Ow<9jFV*OfA#OjFi#=kj@m)HjLy}_Npwpm`j@==pB&v;
z?0a|C@FJ&Xe&j@f;tpU8({N+O#IO4n8UYBm7<Qc?KO0=oXz4k4kYJjWca+%os7iN6
z(2u#V;chbpG76I#k+Be3X5Utoo$2X97ki;b0~P6`%zsfC4;=Ywc8o2W0UR@Cp`EY`
z4O{`FK?u*vM-Pc5dFPD!@?sq}=i9D7I1oo~<jz`_*aN0UrmIH{_rgg)D6+K*zYsSy
z1%Qz7vW`pQ)%)o_g%k(tEC-=>t2Yae=27c?N3=vryfwM|yB1jE3lri?PNP=}JEivR
zV>FSIx~{=f0}tXWss}X(mAX{Ux$hjtSH0CaG9sG<58@6nrJ-_@2YW~^>*LM7ykGao
zU9oM(H+m%)0wfqn^gtegda3szClgH9%n4VmlefX&(Y6}5J)kp{9?V^a4f)9{Wb<SZ
z0tY$)z)IvwOLWt@*U&jPo_VR1rthRnzbY&abD6cZ#JajUA%eg4Qx*q@3ixJ6ye6>P
z)pMGf#f!isTJjNgY=S1p;#DB2#p{QMs?O{4SROAA^wsX?y!c>dv$oP7$Bo6Paxz3S
z70fi`t?SWWU~i*iX;s49RawhcWzjpT&v^~S;9k!YAOJp}LP7h>CbxAWn02$EOrzV6
z7|HTZ2Ike?hf#qbm<RnZX3JAmBX$13pQzsE8R;URDXrTG*ms=ydi=x?um>-8Kbj4|
z&sf_R`0{BN8Pqxo)TlSpah7!EEp8?mU_C39=F(F!5M3$L@%b&qX{TXrUt@K(LTKQI
zs%@^i7U@+!cMGxgT`yR_pzS}LL*l`Z%yq`0BzdH9DuYF*Ex+Pyc%dm>(m~SpITxsB
z>Rg^jk#G)pJy<q?7I<-~qM%sRJmLF|umt=2eoVLGKD)d!XnIgL(-LOxmq>3NAaNWX
zSEitHohNoZIBlwvOk&<Os_7?uN^zqNFioMoQ*@ONFvuPQbyC}(o-`|c9Hc&N!`en6
z^J7{>nZ9bO`5Jk5R$oq>K4qTrp=jn#2>vL&^P$|)ke9Q?M<*@t=OySP`1mF0m+e9E
z95~v!eZ0q-MD1L2&s1cu#fK1_ItqsxDlCS^r|6_!?)b64q14(#&^?}`{V>Q+u!17W
z6+4PxQLJcUPMj92N-jKGkJ?WqKd`y_6)8SmPFUVd#nXV&r@pY7F<D(v0U?~3==n_I
zJ0j8_NmS%kWYY2AAhiIr?v^ZbN^G!db>LNA2^%*S7x7-=T(N<f!NRDhxtL+O#fznR
zCDJ{>xQ|mJEWyY@{G^iHhJfv-+rZ1ayW+KxXWtLS_-hnVl<YK{5N*l6GSi#U1;ktP
za^5D`c3lsD*c(2{{u#!oxD(5Z@j;eQy`mQ27t4vZML(=Rgi~;&vNa)XQbtx^#kk;J
zeIEriRleC3?<RQ}7o-+J%qs-KTh8v|q9}u8Ny+pj%H4ZC-Z1!bC*pF+N>3|j7f^1a
zB%;zDP%|2<bGb5B@7DcX=A%4targigyphrUe7_CZN{Pvv1*Sy#D2_M)bb1h%77q%8
zax6A3O@edp(>4$T<-N?~BsgqaISE7E`EHAMUGm&#M>oqj4A7<Hfsrh3#+4}qt(h#e
zX5z_FM+RnNO+zjtdrhp@x`5t|DWH@Kc8jh&H;i~v@QYRS3m=LXZIw=iB@NmLOsTFL
zKwWlZu1-W0GaX=VBPiUYx3Y;`ncAnWqw1{>3HcA3hnSF7D32qE;+$gaZA{Hq^7;dK
z5U!X+mRLjEY60*t9`vhHdr>Pak|boVZ+z0}Mv;Eu@U2tA&bgo`)2v}soQKU=8*X)@
z@X0vUxYN?lVv$DW!vrou>flM|FzA9Eo|j#MClCjG&E`VmCb27*ERhy3><ckT{g>^1
z-ryc<F;J+R4Nor`^=6yA3HoBoUIcWF33A#MyRKXv8L}CQ5MQlgBfH3?U*PUqHgr5H
zywv^YR}08eFIS+`1eARN@Xc&XRZN%J+k+2e!1Y7sP|fXiP$ZSUG2q{nC8A+-CkAWl
zdR2=lzzpJH$BacAI?N?Wi&^`R6Oj<3c&82Z;~@ufwY9~Ogh#j0yH>a6UfK>Ce7x}0
z;9ISQ%Eh&H1r7FAGg@d#UY|}4r%LhN%o{2-eQec}8jz=7&E0`{I5pW9Vn7Nvfo*x`
zK!SiBFTy-e=8wX0-~?_=8;u_{AQ0&_)PaoSPK=+TPXcr;BMxDNc?fXq>M5*k;MB%c
zbkL$00Nvp#C>0Wxr#cb<m;6-lp8QEL9jC>9lCioi-SM{%yeVo3Vf-XXYWD)D#WtEp
z!Q|coya+8|E;8i$37ndc63z*ysXg`SP1dhPIE~9YTrK&edgOcNPw_c44aD3YEWUeg
z+H-LnJMx?A<1_M2@x7y(e$LUCz8LZ9@*r48sffpmClyHOmCt?oo-0WQG?$ro*zPOz
zXqY2%!Zu>2tNkBOE*>I}<Tez3#Wi#>+JE4sFAQuJmi0@G+`4^$iz*|I{GScu6|P}>
zrFA=_uv*wxdky@|QS6J9QwOoXeXFSMjFGuMa2Z(jTY@_Zpd;t*d$t&ZJMEDdjuy(O
z0FQz!3!Y*2TQ=^DSw<sh2J_opKktj$X1A8p1NJ=no>43LS0pH~z{_#W$Sj|w65u3P
zv)90Xzk9UeLaXLb&HW4z05~m_NEFE}asdkoOXY~|CY|+#@8j%zYJb%2!QH#oFb>0m
z9Xm=*DfO!cBBzCVldsRNb|M2TuJ>qy+nIORO_Ecq&~jDsItv-Gp7wIuW9DZXm0s)L
z@t`|<7FRL_G9Ny0uCEy3*uZT;uf*iR<939s0K&M4_8&lIui>ffvgdXEnxeT4t#giA
z<@b=%8{9^9#T(~ir;^U*1Mr~bPg~`E1Z}x|6qXxUPfN{w?PB|w@h34+t00u@p!xIx
zDy^<csOM>uC1ADLmee~<D>E+G50z#8;CT5(R*M=F#15>{F?I(TFsAQB&^Jop4IHK1
zwS}#2n?oWLW86(i0GPjr-1bumYhM+@OwTC>i~jX3B<LcFSZOQ+`HqN`@u3{PLKG|~
zfO2If4SBGu5lVVlJIuZtU+_@ARVnyb?x)()RxLqn{NZ0qfMS)NNUjX2%-&~th^Omm
zXk}y!O*jZ1Y+O$<+$oaMEqzudQ@|X5rSJkNgPkgPY1R9#K_9{7p>UcoNE!f@Y)hl;
zl4g3Hj<g(hRCH%;aYPnRc5c{rNo*`k5z5T~>-L8i3Gc)Pn5>q(YE!(4Nb5OP7tT!x
zgkPImH<HVaJfa3p5rPi(b*_sH>u#6BeafiN26Q;!v>%()(94kh_bJgxiH8`zoM)6n
zE=veGLu0YAH_hW5P8-k$9Lw89bGtcQ0lf#){0GQ&vfEdXs`@ILxED-cI}a3<R*8et
zrCnJ8jG$v-BcHH$oG+@52fvL^3sd9k-eA)p^CkzUkdf|EWLjjB-&_IeQddR-->t-}
zz2ob@uRynV#jG_f_C#i7(p*-h8ty^T%^kW+HgPmjAj3&OfYx~EZ&^Z$h#SVo6H2gf
zf4<(e!Ccihjc&lkz{60adfr3!lw>FmWJM=PTvTjxbvBXEZ8GeYrtvwjtyQs*7&wMl
z8NruJ#WH*goK)t^Z3emP4`XP=tO@%Kbx<#`h{e^d?X^y1#H%w0zFbs=a3-=S378Yw
zQm1)RYAOrQo@IQSH?5`n-jpuJV{|JX-|~)=O+K=9ErNJ#COui6ep0$Aday-LP3n4K
zb?&q__5(%{wtvjg!rJ#9MNK2Jz79?{yI>t^0f{7$=Ii8}{hSNYzI_6FX-9$%$%Zm)
z9q&bN=Iy&ye_@JW@2kSotcvsi-@AkjuxhFFO8^1%Rmtp~{>^eXK<pY}H&|90Lb*+0
z1$(hsUI#1F0@+EQ+Aqox648KGrp+A@`fIjOjo%yUscYzphqO%I!Kp|6TY0R_44BbU
z)0Ud8?hcB%zg7>_?gc-GdZyB`Tfdk9UiQ21YLYqptCyPy-3r({c8mo(45zQbShp1~
zk8whpS@W8K=JSNQb&x@=1=(zDK&5Tl?J`L<Di^%(mHl!oztk?^{?+QIRz>o|&ym}s
zA-fX{g%}B|)2r*}J@`#-p&_F+3{%v<wWA2x04CCY$@!qEy0;ph4+7bDEMu$WVPLpW
zCdt5?@SG<}ul?@l<^b0TU}~3Fw{6@7gURqQlvupScWxm+2KA}UG%3IqP%J3+1unD(
z(9|3)-eI6w^t99Tqx$J8J!Ga|X>AUGzd1S}>c^lK>O?sQ)6DgYHA<oWfrf_O$3dU1
zq>}a5UsD}1x4K{Lcc&#SMkBbS2gThrG=PXT;m7kW$=m_&Uz~_L6Dr!rO-I{op|<B(
zf!u-xXp#~6sPD|ZepT5ak1TTQyP$;MKRR=!AV)*I-QM#Jm7rNdH51myJ`^(L+eZc3
zR_6V-e?E?b(40x1pV8ud!U^)nv)*cejqMBDln)|5hl_&`V@B_{ocwMdtxs^aa<A2A
zm;*sB*N<PVj{BI;UQDGp%t7wzLUe;Zgk1Qpwqzfm_k$1TKMzv5n8=97w-l!e2&-3e
z$684}Rw&7ykVQrD5fmbJDOqzEB#Z%nUH$C8e<93>Y93IJA}a4FXIbMqL+T8Mj4iPC
zNm_h_4Az%%Vr-@DRz|jsuA74iktOeOnjLKKIur|kwT%1oHnnstZGEjYej{kOQ|VwH
zgS%CjURaLVxNP#WK>ysR=L17F8&1n^Owi83N?J~>&U3SUquy-ueMK~lRv`_)-_N++
zS_cvYWB)ld2edbJLy6^a?i;{dXqi|8xq}<Ks<xGp7N<vivn>=Hm1%@NLkZdmLRI?;
zJELG4&a8Ax`xMSLdE2SAn9dRR#n8j|fVU9cW6}f-{KG2KjQ&IHpEQ@U$}ZDq-qp^a
z(clx*|44!+)cc5E(X%sbpX-5OOH{~18Ck8X1h@cm_1Dei@j~4zS3FlOulQ`PUObeY
z40F($rH<K0pn2fQiE;Ab8n<?ax=eo(`-oq=Z{#CGo??HJ`{Me6IaP<YpC-N&G2fIu
zE_V~hmx}=4P!30x$X!9tT>E6VR>nQ9%MDhE(7&mr0jXq`hg01)QfY2Y9b=jc-bUg(
z-D-ilgS@>}obJk*Zz&kC(0+)=*Uvo*BFf2U&r@zlr!1?Df;Sv6UO6_#G&;K0;>cNh
zC~L{3>~c<b{29dixQ=LQ?Q!YJaOcU1NGs_$G@f)G!X@HXK8WxPR(U)o;cQbZ3wHbd
z9hk&rt1N&_eJXi1)&+nXO(8g>tuXTBHXv=kBNG*NpkWe9A&VjH)%{DpVBti!<2ZT0
zp-x($^@}n{dQKPT|0c{2hKhzXX0T%B+KOO+qqKH(M>v$?mjCVQAAhV%KFbD7`spv5
zasqe(_5T?nbA4BH0GwR86xOc}6cM+;kW9v?o$E-jZR}$ad82R70<k(%kP;Y7sJLL%
z;BF>I!{*Xw(555(DmR*5Ey#k6nghHGs=7uE<kkae=4)t*jpV;-hW#pw;?xcyS<BxI
z=4qTc9JjazR!6*gfmtvBl~JHm-DYdPWZ{=YZH;(_w1~`aGKcaFTW=CAZB~NQ!lOMC
z?R4qjJ9?JvB@t&XT#n-GZHs7-v|Je}&+9D2paS%}{iK#$sbf4Nl)Ry>RZx4mWwr<}
zPa3;{po7LEx6<0s!A_a5ewb4$_2gn^a|P5);XC;C4^ZmSc+4Y*h%0ik$r}?%D9L_6
za6lc*R`0By6I~0~_X<O;Ost7^6>pa29`jR)B>8Yh<JfwD29PkM`6{CwG7nH<)AI;6
z;L5dvq`K$8GsLuz*O@ZhlzHP*srUq9JgFCP1u<4|<{6&363*IdFCM>t!IPIxk;7e-
z(67GO;n46iNa@K`)eDyqFm_l0i>FV?I7BZ%gdPK!&$wch*$%386B31-7WDa@^+)O>
z)ORY+hWR>QmyljmuaC@~2glvZ&_#K75+08=9>rq8JEHUBEpfY-55f1>4G;F7<%Gfe
z&b}yu>7v3Jm`#5YK3@k#RPnwyu^;EVdeH+R4#d07pM@&t;UvcTdtQq&uv(sM3$Nlx
z?JF_bdCP&Kq;is6^{Qg@s%%djI;~x(W3rmzFT@jo8DP=;u?Eo1$PRJ@!pzQJ`vj%0
zDGUnnJ{*q5lFwV7pD>M)QA^EDdkbMtp$*@$E6w|<k1sWhkIohQ4-snGXMpgz{{6#Y
z5uif)U|zXL3rb&8DPD%SGGuRVTT&2A0U)STzt*l#uGt+R8@*V)*)sa%JcS`F?w(sj
zbm&fzYx<{kJZkTL1;BthY2@}SfC=T-B4B&1#76(TNc-06lYJqJ_EN9P%uJ;?zk{aq
z%3Z??S{Hv;D&b^T{0${$$20T^eAG-=d)Os39SWZbdPMnrE%xJE*TFXF1@mbkcN4o7
zXlE0rrV{6hsYJkQ$4edhq0bkWK0ZL*9+CwBbywDCrRqxW+I$!0cbW6z^M%!75zp@^
z>t-dGzc}WrA5qZdtz9NnNwo865=;M<cPu-vDRQ?7ZM?puC^OL$@TaLoBo?6ZEE7GE
zeo)VN9ST0d_`3z2?V9--m&w03NP0P0W1@FeZ{>TQ=bx_9$Ws%tV_xSc@D*sg11ZXt
ztb+IKbfx9f(_Bdvj#KjLKu8@qJciex$=ii{vk+!ZT8_@UxE_ui)5)ny*ghDkPHBf<
zN8Oj8ke1Qf)U=%ot2wlJ;T-wZhSOq-=0NbU0M^Oc_UsLX)uqUYR4eC4+<j8UeB>|p
z+o8svSB~ZP)Q+NBY7u-Ee`*riRBL5R-{9!EUf~i_uDK!?C)1UB9@4q?!%6BvGc!Kt
z)pmgbn{8)b811~wJx^<B%zRMR-=y=Ekaj;Unw3WHlZ<)E*5<pnX#)&UO-yOJwwRZ8
z2YpxasFr!%uDpkQz<+drZr@krOg}ONzL^Bu%AD=cKaL3-5VSma_x!^-bo5+6vAus?
ztp7)E;788>2{+s(zfTZM$TKRFHJ{O5_}Wdb&IxyDUw|k@*mI6<vxa{Xj03rQpqLqX
z2!ZG_18Ekl&r(w~=a3|Pn@b{ftD1EE?~7%5K+^Ea{s4bMHtKIrumWWE{Aw2;Qf{L4
zdIyfilVEx6lLwrUQh_f*dvCUnyeEUP`K3(M-yezvqWD}9(b6xXT6=BJ>_{aR%I0I)
zVlDsVs2Y>8_W;oXuHYdOt%*)YMDP9EfBx78MOa4*yy@SUw)eOH4Gy;zB^(^+jsE!~
zfIH>CzbHsNfW#Fv=iGPvzr7i9ViMc`zkct}E2P1Fpu){&Uj83HC#my3@b4>y%e?`g
zfHD80oaz7gIhpe>%6~5A@5_294_|^t&dc}z<L5_amQSEB{$H;^hOoL^<?}Dl;{Dgp
z|F4td|BrvM?8ov4df*n2z4)Utpn~=qxyA6=WwMfOT)zz|D7yKonK#2PhI;(Dp%O=c
zmHxneAomLK36>H#ENv0kZC*9)4pO5C&Jw`<CclSa|FFzeAX2$rkvxrNPl){F7`ryc
zU{zfdDBOHx2%Vvl{c_~bP0iGTyHVM=a1w^P7kbc`sLZU*yD^~}%*1?2g<Th7C#sjl
z{(Sk~XW+1%NF+y)t5E>2d3!7<uj4PMXUD=wz{B{f80a-5B<Q2TNyBDyeJLdnTdwfu
zH~TnMCW~vYD!!L17LxssZQ0)KzL^G_i29Qb%-@gHKexIM(6|KOe`THjSubHXpiHOz
zT4wX_1p(W@D#!w}C;mvE|FvE-Rp9eIs#*9~pbJ_`Fi`<gz2NkdfBmoM^)CQdNQsrZ
z{-0;}?>D)L0x~zdgiinG(L#TdAcuSM|1bUj^>003_n-{V!{OfX=b!h%KXTH_QkWaJ
G1OE$<ObZkM

literal 0
HcmV?d00001

diff --git a/docs_new/examples/images/knowledge-graph-extraction.png b/docs_new/examples/images/knowledge-graph-extraction.png
new file mode 100644
index 0000000000000000000000000000000000000000..a24b1c61a9c371e4fdad95dd5ec9fc31cd71c9cd
GIT binary patch
literal 33127
zcmbTebzD{3w?4dRC8Z@*x)CI06Dr-^p$G^fC6ZDW9Ri!~1}W*3kOm3q?ndd(cdm2p
z{oU_-|9RsNKj-M)d#)K{j5)^hJYxl@D$C$uQ({99geNB}sSZJ?s^E|FEllt)U&-rj
z!GBO3)n%SSMSav8;4kQ}o+v$mppr10%NH2n@3-w`^&BCHj1m5WBB#!<3qd}Ya*|Ip
zU5)?F;1~{TU5j5=*^)b5F|FdsE0Jgn??;gCj*#5p&W($W{(7bPuGjv;)_Akgs?I#O
z&bM-_J04@48$;n!DJu7^+4B!kX_YMTx1yO`4RZOL54YCDcUs~W9<|^;eE8^>hgk92
z5|2}#UwDs4pIiU?zB7+LtL(dXz_g%Y1IIIO_<wSl8NiAl$oVNMJm2U4zyA7Dn+uvJ
zBIhzp?TO(YielHvt*=k?x;}eV>wd(hrKve^v^BQT9zykQYm{?udbKxU=x^1WvdgR+
zbH$s92>mKsC6^f|ayK_OwL0g&xt<rt!+9FadxnrtwhuZsM4g_Q`B^y`AvfoFLPJVQ
zs^;LpbG6$-@NRor__6nOJ*)6r66MnEnl;+Z>6%AQe<!0nJU#Q?pKQOoxjNA(HvZAs
z(IF#73^nb2l|b>C38fK{_QR)1r<VyG_?;ocDlVSFWmL=1xJmo?-MYlp`BBF8)y0qx
z3Yz8aeDiRXoynON<kS8HY){2vc*fPm$<PNhEEZ;FKYUsQOZNPqwjtZZ;aw8LsY;tR
zB(+L!#p@Ay#gRQEO&A-`bsQqAL*xBekrA<mCt^iMTbn5x64qPVka)jG@?1g!!oa}z
zJ&^WnqojlO(Ul?uy}$nf)rYd->e!(B*RRZ(I+vkj(f5*c8ZiPkld3zjKiUrCQc_Z4
z)$4W|E^-xOxJ<wO4Hen`+W#)t#QRE15f6Mjy5Z*fJ&%p+`Jok0i0fkO?emSCl;uY@
zORbzjLb1}pcb~j?@#1j)6G3+|umk+tB+xBAJ-x%TwG<6^_s2<(o#bgNwrAh(kZPEI
z!o(vQdYR}7&P~Z{L^M&@QJUc+qBk}shB=$s`*Kg_Ol!;i_>bJnj?gTJrFJz-%UlFD
zq)JT00?|y>xo~?P{+2;TM>BDuR<+H|nJBv8QSeGCdhfMOCi~0exl-;2aW~u<iU6Ci
z&(6OmwN6M#c(@#?GPt|YvZ4n0u*s%E?|gHr)z#H44_12xWa3>P+YaZd@x5J^<sn+K
z%}@yanW2f;ue2WHnyhim>9rE4jf;Qm#3m&z9mpyVLBO;h?=AS<8+^a>Lq^2;&sUjW
zdlmA;Op14gbP^F8gv@G)kQ$pQ0SB8X{tTJ0?2VzEAq^E##xTf-_Q40#f$u@&+=J&w
zTdWTrJSpfh3)t~b;5847&BnfcJ6z_2`Kwn!8@-$cmHr|(9D9=*Wfr~Z!j7w|9icQ}
z_5(QJ>6EQod~F>aL$yv@%Y=@IK@w6@sYqs(Xqi+2gDTss<z+J-Ui04R8b?+pCd@5G
zl=WL+>+lvtdasYG8UqC1D<Gr<9aj~3t};ig;}b)Q;})$3GBX+SH9QiL*9!#RcU)Ld
zeJbIpLJdkcy@GR@1Y)NWw0Z3=eS3UEkNOexVLtS4Cd=g%!}e)Bqm#YPt@O(+!*y3t
zpz_<dNqn%_IXE&&*|q%%2^A1Zf_Afr5?KZz0&am?=fC>-L+=rbCmrI~Y8)Kc(BwyS
z>?X9U{T}Yab}G98lbdrLw@gYClIz7F?m$k3CMVaRgsNG01;cl?8!Is;**YekQC=5Q
zbKUgN{04A5*Vw8#uLR@wfB9?e7d|4?%@<n(4hI#uA2_@=O_VJ(tcei2K7HwyDW9X9
z)av#gxbre0v}*=T0D*2|U1rurG+AYrf>8f*^7j2ol+Ht~H4E3*c4zx5rrb{|UbOhC
zy?IkmEr16h$iWs0x*V)3@nDp@AH8<V#5U16SQEdYAor^iZWzBj8Z#ju%g^gg5jRn6
zo^_tG*}nz35`n|A{WGK@p7jNXe02Qn7h>jQeANEtwIE_v{#iOEF0N2o?@I?abO?a~
zkKdZ1T%UZ0+ZXKg=w7*MikRmz;Vwf**gf*0!f2O2dvuBQUgXjl5CRMSc1yxk!|qR<
z#okf>D#Hb*WQ$+155jzyr@p`Vq{}S-LkgF<&3d$eFhLN_=faxnh5)kliD8-#Y98q#
zF1GD8D7)`Ah%Xl7p?o$?xhK5l*lz<t)AyuPp=k*@xd2_x%d&p4NEv?+M>(Y#z6x1U
zg-P%EFOB5utPrN}Ep-HG^?7=DfZeng1iKkO-j^iOR&>4Amjr$z9Y)KkUA6r?!w-R~
z?+ijl$?HoHAeRZ#!M8M?K1B_o<d>&Kfm}YaB7B%y*>xOCGnJE1@$3FhRnfZ*5{iv^
z^qoypIV8+iHrt>>(*$5v@tp_#;woFCg^~Q}V5~t9tAY$BokC-ZVnyEz7}dL~(LRV5
zapAf;NV!4sL-tr;R)&A(Zmz4UU3Eo`TJi%9e4A!!og!r<`uh@v=>_Pti%t77)VjL5
z-kr?2^w2{C$cJCB=AU}3k=m`Kf$1#n{rU5TLFM^Rn-5dyRe1EakEWfg@lYg&o}psQ
z(}A0SlwRrV>}>zmZ<FhN@)F^PKK#=H)ggBsrWOQRZEi#HWZAor;e4cuSejfU)9*rq
zs%1i=A)fH?@KVg@2J-TO6Y*-j34%Gp!<z1k0nG4L@&`+x2vOKEi8w4uYV5b>2Ohj$
zl1)x+s4lcmSg35N=5$C{ti*tn(14c_Gcqv^mRk*(g>?4!tJD+ve*WA^4;{7G_)g(h
zJ)pNWm~Mxfj^HIR)H5{ic%6vo(NjKr4tZYgbyVw8Q-0YQSo)m<?yF9E$({OxP~CF~
zGQ@%h-pz%Ixoi*DGr|ICPDPwH8NwjGX<)&YKR@3-1YX!p&!Ej278)82wpjp}IDY$B
zR&LJUj~_p_(z@t2Vo?W-5if}zynXu?oOlS8;2<0_tf6^Da3s|JJC#ctr1Z3I-JP8h
z`{>Hea+<EL)f2KLD#T0M2xtx7-23$cDj@1_*ori)29S%lmeU9ktE}y@t+ZRGy9S^&
zjC-8!E~Mi_eIG#dKyh`lG@job%u-12bCEZ<uvq-)V1K*eJA#8F7#jyi>c=xbd{e5}
z2G5JY$B+;$5sQ!Xq7})kX-oCbYKNp8a4iQt_F<<Mb=M5`F|K(lHGFZjC3A87x8;z>
zC-ONB{B_dwk*Tsb6Qvkk#MRVQ!nW_Fr<q3bW6*qFf*@)6c`e0z_RY)Zv&k4iW=}O1
zDm!cj{HK_z)F68V*d4sXQ3Jc*AXb|dzJB@qO+Ttq$W*F3nyGEZv^Gjpi9H^P1lphJ
z8ja~QGrm`S)FqB$K6S1jP+5Ycp;VinlK=HEiyoR!J4mIT=R`yhxi5#ytb^bVRyzk@
z7Q8+p3#_6IRAElr({T;wo9z30dxTDqCAH!L%6NcGLU5M`xI0)`S>=!@(|A=%wi&@w
z992e;JuaA>v+GibU~46Zbp4Bmqwc6KZ8nipxr9|lwk^<rYwS<7TxR1bt5$!)Qt3kw
z?Bd`+tA}Q9GOY{}11CND$sg~lbRS_ul;Fab5S^}f(<mr&ygYSKTf^zBvYX@ThLWHE
zisi{>6g)rNpeLKH|64gFNdx`y1Agy*+H&vuxqLF5s9BE{k(xs)%{I|fTuYU|Y$*R7
zkw)z5P$MubEQUXw-+ELvj8@bPuj!F}!b+uet4;ir0K|t2T<mqTh6`fGVRM*OZLN?8
zq%;iRvFmZKl6eZXXUkrx@KAoqLqHcV(6Lj=xEn7^S=|<uKI#<eGlH)LPCcwLZ?{R9
z5`xAFfbHX%c<f;m8voEA^`G@PP}&^Mi!4BKWwLFVjt;2~aWue!E}ml(H2E&H_zeMz
zl5~SzqKi4jpVx;BvDZpm4(@$5T7frr|IXCuAW>4i9)eTD*s*@4m^{9}(qn#Gf@&Dd
z{PFQy^zQ^9h~qR~?gvev#??P(GLgkkWt`5gu>o;!t&#vb`UH#{)y~Ob;U|XL@xejM
z>!nn^>B^drN45v!s~D-*z;b-#U%W`Gu9&i1Y7begw1Wy4&4sg-5}2&DTvjO}*`V)u
z^r{dWyLQ1>#`O95=L{;TZqw~Gg;aSL5H#*f$O5r)bBEEouRqYE2bm$?6ymN#zN_QH
z_wNt&=+9RZ>?{XTpOj)kqtDT?CDe;U9)Yj~#-N%Wt?310phCKSV2n?|!G5<8h;vv=
z*3{IH+_`h#8k%(3oqvLji;Lt>FK#&RE>MAdgh6b(d->PSJCZS-RU`Yp8|uu0K)=o0
z`i}Qa9S4dJ|J@I$Ga#M13(|U((TC^g=jOuLQ7I`ZAajr6hg4f`k*phgc+_)f*-U&~
zzI_-tC3b>B?=%NNj^LL@b*_8q6PV4_2l$75`=JQq=Vc&&1ap^$KQw6c`oTd6@<jA5
z;&s{+lx-a8O)*qx<g3JoB1Q&=PNLr+3pKkfF*oe_WJU%`rOe)ys1m!}edZ3*T&Wt6
z5+8yE`tw3};6k7;S(Lr|p2seV=+Ur7IxeII4!QOSB)45ezkkI&GQSPQ)~e)QU_+YC
zV9tYGk<586JF}(Inudmo@D0Qd2|2Ffo}Hg}Q~nMhX3o_qHU<-g=P55I7oPPk3+F2W
zC3Z<H&3I5^<`pRSNY;lzd1D^H!p0^a4M`}?2o8H<LLjt1E_c~6#>U4dTXFt7`Fo`&
zMv4M@{d%rJ)CdGJ<jJ3jXDxmN<mS*1c=P&X?s+mOXkE;r_}FZUrEiNs?tbXl@0d9`
zm7@KVJ@)1FP$BDKPste<=n(kxy1ceFSriiK&s%&&LPSK@4LzqhA>3vUzQ8|u07;-?
zJU#%~eUJJG_`a&3lZhJmS5XkXYa^1AxfCRUMMg40xgoX1xfgeQk{7rT10i;eSJCGu
zJKWtiljUjq`)}1SeQH5{_FF~#ilJL3j5Z7Fo`)(D<q(@*HI*tZF0NFP4-S5)KIf_@
z6iz0n1MCdogF2U;XvXw#dYNt_XxOi|Klg&j$M^*{!MlagLWALNTI{u6SGBn>cSN9z
z+>`dYgFcas$qMV?dSS6opS;qJK5_k-Yn($O&Mq$K-B7<1pHOUz1YZ!Hu>L!H`}rmm
zJv#S4B^_qBp+9NcBAa3usWP&u66D}MGYhcn{Q9629XeJ$p>aw<9iTTv!>Apov@sYe
z#mG@f$=;r+3*v=_P1Mu~kJf)bL%QjGDlO$z!yKRwmiVwrBjWrEiI@R(gKRQH=?`yY
zyJz49nU68JM|wrxt$zhZN{i5qhEg94d&<sKL5G;PNi1i<TEUZHUqpxS(zc&%HejTJ
ztV`m+b?SBo3L0i75ybbomLLy|Gz&_--k$#c7{;I1Mo>uR|9p8!US573z9oNA-Fk`A
z%cQO+Uy~4{HjE}V{3pcFZ6MmTJNBNuSZ}=A3G3dw({Bh|+Yc9aP8i!ggD&p=Q@JRK
zxb0_h5FTyM7(}OPM@^Ksz9P#qiawHN5f6W|*B_R(&FRU)^7eMta!hyexc2n~8|8(z
zz}};vs=kbM7Xlqmlxg2u?mRGBGMJ^(#dh6}S-?Nmk6yt#m;AIxo|tv7%+xLO0UPU5
zl&XF#@)^b}>|CC?+tXqH@w=rjH^IR;&{La0)}t9owQk)=X}o(ER!<gs%2-~Qa}+8m
zq+;w_db);~NNfcC45%tRN~(`0dPPuhj#`Z;E-tReyYmuq2ZaY;VRg0pA2b}x(ZaVX
zbet1h>@tfiPAXXid{2YK_$*5~#)7!N6f5S?Nu{N8^6-7H09(>B7Be)(wkd|0$_x%*
z$m?_gOOLJ=)XiqM3)pjlLwZ^(WVGb1CTjV#Xi@|JW7Z-r+ld^6P8)+#(M{ieYRXUF
z?~yiGVxh2R{D|RXqJF{yi2^hr-ue2{4Gg%J{2mM_eCdENACS=BH7!T{#MhXPe)G0T
zlGf7ye}?S^We+NdDGJD`DmxXIX$L=+PoyzJEZ12rk}pkJEZi74kE1*A<aIJtebJ)%
zpuh7Ct)77n_m&zk|FNJlE|Vg!v7LdmPaGfk+fo?S>D)AaP0KA$(S2s>)tNgR9cRRo
z2paa3pBaM=QGqRv0vrY$no*r|DLOw*b~!(VBpR9ep)0q8*?cf#W^$+XbOh#cJ|aee
z)lO3FUW!HepHzfy&Re$GWLcb7zUr#|7Q+P;;vTswogDhr4?YF<tPE|Y)j#}1WD+lx
z_F-n_#7bI>;PYTX3=wO_K&V~ji%qetZEMIKq<Vb$_4Vf=MK)8FrRdRYk0M#OZ>gQ)
z5SmfCyuPO#?ycFoB>Y<L(`4sgV+)Fi=xC#lESoB<nyUQ<P|pFem3DvSu&4(GX;oO}
zmmM=TZD%${4s0TYM+Nc1(X|(TgAu<TN2KIqAce|ADrjoIma={|jb>KW_FXuNqj`!<
zQ3|l@OD_?<Vm(xu@O`HQZi@felYa3V4h2boHpOs!s4<oIvL=mk=g!8^$R9gi`10^s
zHb;SSuB+T9!N{bl^wSYseO|`sK^*zq<x)@atkW_#lItc8`8SQRg9?ExPUd`4=zglQ
z_UKZ@idh9G^xu+o<CS`(eTPav2@o~+Oq4Se5cIP7XBK$V<d`j;P!P-E;NI42{-FCz
zzD=I>B1Dz)@TqT6B#Tpu`UyYu1H|cgr=!<LT;+uZFZreOLo?Mbozp)YFssSpxc#z}
zKyM2UH?I^;9!#vre6nR9c=e96jmt<O@223Ub|nf6#D}Flq{0I7J5Tgatxp!x4b73V
z*{YZ|<z}|^Q5&76@*V=n=%j$m;ldIUbtlhqisA=BtY=B~dXFxHyd{Ss`VSMH=3O9b
zi&54=i91jb7>KP$_j=@86vvYa`R-YwSF%iWmJ6EJQ(33#!|83BNs-kSa-YbqKWW}p
z;be2P8kuJFMf!JMgyFD!b9W(hOuDlA`ncq=y55o<%S`=OivY>D52W5jV-@b=1*u=z
zW|1^<X;ocZ{^ujN#)^B0ekZxivMM}(%N4SygQScIyPF{QENEV?+}@|)ZuL+E=E@YY
zRRA?%5H>zt4@&uKG}iqj43fUv(N*G7*N+VO`MdBn_PhDiunEeeWxu^U$a+%B@}@t#
z<+0H>qnVsxPdo{3kV{z1R69J2_P1@ka@Ir5Xh~RN2|=}5Kz`Ma4D%=dmQy(WO74?-
z_qbuj29Fv#-hR~;`NO$A2c(z$gREsC#(o;h>l3n7HDATnXgvPeAFs9uB^ACUvyrNn
zC=>tZd72waM)KQ*X4_`-CaCM&V~Q{Nnyy8!IC`pVEP<n`@w~`*u<2X(60%Awnkuh;
z{-Jr#{$emM8++MWxDwCRviK;)e;G$p9kvgCl$}bNIpS^U3!;jO$AjtampUwk>S3kT
zrgEQZ4Y{1gD`cW{6oAW{8*uOMm+t&t6gUe*k<fvcyFgsZ>Bs6|PWRhKR==J%baC_G
z^kwE&G~agvVJ$wo6n3tH+UtOo4JWZ8l!rfY-2Jo(;=fq)735Dx$~$1WE&S3AG%EKY
z8(u3xj5;Y$j&mm9lk<59e;_N&*Jvqk=_qk{E4o1WY(NBD`+C(1o$+tG$E$<oS>w|?
z^2z%Og1D993umX>_`y$yCxbl9!rQ#xbac;uVJ)|Rd2|{VR<-HmN=(H(M!x+P;|EAh
z;zvr&6+m=#{u|G+VfccL(RSV`l8$w{+cdmlkCIp}NQ*pDa3ccK_?qvbV)TTsq&z2b
zMPEz(Sj4((?ZakfSKZqE5q+Hik}9sM*?@%w?_U|nN2TeH{v$-;urnI)G7!iquzdQ-
z>TF@7Kt4Y3Pb@8t97ZZoo(`t+Y|b%$isn$tJE$$_{*-Six5|L6Hs?0Z4GuBj#dZoE
zi@T%fb<s~r{nBYakGihfzUNuq|8Bu}e$USpTzNe;c|Q7hrV0IuI(#(d_Pc|{uPLn6
zS*ik7_b@{RZ;a#xrPYSnD|7a{2@2EYh-1F?E2ypFSiLnpRUhGU(UNCn<n?ba>dYQY
z*#C+#TQBr_og+Of{H-_Qs@XU))dk#ZWv670^u}c}b8M_GMjU&Ck%TXvXkK-XcX7Aw
z?wH5X=sk5XNU$X7=8iYjNyy=@VSm~sZbsJTov473wW4WktlU;VH@Ab^anD$mVrJ2@
zOY^9KOAE8lSgHi2ln$xh&YrfsBp$tb#Tq_%Y^G7U5MKG>b}!D3vdz6%Db^WNR_A!p
zv2d%PT9v#DN*`Mmh!WHR#rDK&evOTdh;J&SH6?)$M(z~1eKaE7w|s_V>59y62-{eo
z$zvez;WmiMX~k0&^;U~k-?Fn8Ql`i~zvfAa?&8+f$jt29Exews5WP%~8lLp(Uo>MJ
zoa$~YJI+0@8x`BuXdn9o9{5iV#m33Wk8yhZcd94j@j(jFimP-1QIRq(swwjGG+ovo
z@qOVV^HXCtR-SChUw;<J(^}1`b-FnoWl)iJ@14+-+Lf(gu?0N}L<Umr3%|Y*E@DtQ
z&=wury`@`_nNrBP!lqGbj8}3>u}ZKr-zGoGcy>5x*sV^6WpL@y<F{~>&rdjX8=lYS
z|C*1I^=y3&Yo|%0ml5qqKv|dk=gTP@Q95<`<w)z8vf=Z_s$>7&!gE1JRzmGY+hg5$
zjUzKgnShF8r!tw(5gyuNGl79+dh$QhEg3UY6n*NiqP|k#AV2?!n3So1ovSMUHZp3Q
z;(>f>%HFFBwhxgnkrjyy`U;otd@myQ>h15L4P;nc-h0Lj4P*jZW1z-yO|quCI=0qn
zXoFJ&!hg5=i(>z%u=YCNajoy&@gF^q@e&(K;>Wi)&)j-K+r%^MDL2yr9ioE_LLcxJ
z<^k&$JB=}nkR|Sa<USUO7S9o1u%HkfO-%-cRZm<A6%b{zopxr8qW!UPwkpQ3py7qX
zjiFMr+5Z}B^6F%cTS1apB^mj@1{=-S;nt&j?6CNm14ZIx?nOjx7|Qx>a0T(*yLWFB
zBs?rPheIlh$JUUg(Dt+Gn7b#O+aN9c4AP8Zu*;>Ahk#826~_N9vEG_%Y~&!M5q>NA
zzf*L;?3e|xaC58t?+@miKY`T(Hs(Kb_i-G8cWEGx0j|Q;UGIxGT7>fb9(U?Qh75t)
z@&11L|I`dz=;-L3L~6>)p`b(-9rl!&$$<`Uf$Hq3sw$`!Jonnj|Fawcff5Il<57(1
zfKuuqg2tcCxD0z@ON0TiVEre6Aak|ywclNxJE^UOe5rHVV(JF9o3X@DwZksBAmzZ(
zWUTh?QBWV^6yAIH0nO9i{(AwTxw(1QzonMF$<uvwcRyOFmdCX9mL46bt75?RdgfjP
z*K(t*lY+8Y&=m)Jsk;kQcs!ie5ahIEjuZTTo<`IiN$auu1lR>TsDy0on|3D3a##+i
zngJpV7p)zhZLtW4dti7#&b_nvGhPBt0In~=gQaJ5hCOdd^uyJ9FqNQkw11_|6q5p^
zO7&k=XA8e-s>sQ-Bf&rA>YU!c<*KTNv$Mb#U*EveurPuoac^T#S>l7rQUnk}VDR5E
z_rR)jGq&kB8_-f+VDw?4&ZXE&Dfj<HEsfuZa(AbFNzfN;X&^5YyVUSPF}b=QfEu~J
zVE$GBI4YGyCMI$yK&jc=RGVUG5J1zC9=q)`z~$5EcR^Yo#-mx6Kdk}8!?2~IL10_B
z_NbcIe`lD7tp7Nf1=aKOO|fQA06O{zgL5?)%2A171gs?ksME8zi8mWiAuX^awlE4Y
zTw$%FCiecA1MEA%`3^%bP)}~RdwvEWL@ZpB3s{eG=+WuJFbK+eLM?ah#Vts)MQS<K
znw;A>Q#t8z7p0no1yjgfo~NVG;j5GF84@xwCTmHUeFbQHHzjlz;Aihb=I*ZcC9zve
z0>Cbc5%N?APzn{OWeZ?@VYcLDM=tC-(#V7a1r!vNQq0tQSOm10O-&L2U+VAA0f-qG
zDHk3oZ0>$~db(N2dSTUU1Q<!J%5$3-jPwSK<kOt?%x@hqY|;F|Pd{AihKhM!9gXq1
z?mh?GqyP@;;k;mAPKp-*>9mUsSrwr3RNI4TasZkPgY$*djg4tV4+ox=`iksJh;!`J
zU;Hm5Yf&&b?$Ij%zWj^GO%2;5T|g1cKDA(C5?SBBe+BF#tF<JRpl#->WH;$isN7G7
zCPly|zJL%AnSqy=mu7g-8-VTjU^hC{U+(t+bo6`+B$++}c~e4Fo(wAgf_FdjiyL0+
zCnd^c!ni3geM;M@>2``LH`@zHRUFvE_eBt61Psy=$*2Ad<%9vGv0<qrETQnA+;--%
zJHUX!b2edy?EeGa_yz0vGeBB#gu7-;044~GDPY^vz8sJ@^%xq4AxA1-T#Ba$YyALJ
zl?oyYNZLR_P*lCJ?U9C&K{Ovv04u2mB?#IDTBn?j+O}VO+OwGhKxxd1tg<qhxpC>I
z;xD!T!nm6RV7Dvg1*5C0h4kq6+CSwXb~UG+L%N8UDnpiUr@jEVm?{C_STjyfUDa%7
zYWYFr{Oo?X{$LYIo(FN9?|Bmh$oy})9j&bcAeMkY@CQIrskE+a#ojm9rnR3!Xhq{I
z{|87ef7yYrwraHjXjRk0+o^=`TnR3v0Y^#+CLsM8L!ep1-hiJ;C3uP#94O4MtE!}>
zrIo+iP}j?!L|p=09KaG!T5U9c7~#OHh6aS-gBV&06IV|0JdR^=`7<8gaJjFrIa-*I
zi&h=^=vCi9Mwpm9@Dq|Fz<beeVA<b+k#jay)-)Mfx6dGeF>f|EHzSjhIJ=>31_p-D
z$A2d1yOTs*vH|uP=;7f3Aa6=64Ohr{Q84xMXLK-C%rBy?O*8fheDG8`BAB5LFx3Gl
zLpmfKFl;@PzX3nIQS5z{Ux32LX&W@n6jELIHUR_bXeObRphw(X9+;B9r#H@g$jhrf
z3K?=UGD;Q1+VQz;8=eE4o)wTSEP!mGl>5<u-_zYqG?oliQ<?<jG_>GvMWF}*Gw_i<
zKWV&CYVpO(xV-dGvl}Z)l1+DAP=a(8zVHaU)xKQ`%str}3j+y|jLV#7RW2Yq7ARpd
zL@*by!9GCMaDVd>^B558fd-u@h99OVtAt<fx7vhHF@WW)f#u9Y0J+TqOt$L5s+Q3=
z;8#q_iLxM~+;M_lEeN#RgiKK{33S50fZ!=XSQKI&LRs0_-IM`%e$0xo_p?u-#1B(s
zRXn!V($knVoTxsI04TFhWlnbeC}2CyV$Gn<x;By$F8Lt?CdS%qe)&()Ry~32#b8$I
z_}8QUSkCge$jE043k$J*XnqTotu}#E(S$IC1I<%{9ZJ0a_;LKK+Zv5tK4rk7LAS#T
zVLq~|2et~E;mo$y)2mRE6JVKzy8cX*^-8CoK8^y=xI>+=#4~$9@?`>cYU*wkV8J*G
z&@krB5(I40t%kC@^hX}6dBl*AO`*TvrHj+ZR^;K-YC4T)R-8iN!vizp+69ckK&D*e
zIsh#9Z&iAJI4_${{u0ABU-#F(Sx90C9I_A2E)`9&vaR8p$UQ1Lf&0F-_Ec)jD$~Bz
zoDbYkE%iYZv`+&F_itd)Hj~F+dgC7X&3awf=%M8vPdoO*_G4c7`aPNP3+f;<R6bGo
zvF%$YyZ#^D(G~#}0KmX>;E`SQsjob8W&mn!)t@Z7FFVrSg!miuR}Y#-%Le3S@Eqjk
zcL}mRK>5wG-19tCI|oVA3gG~|E!s5Ma2IhEsqG^WwWk3g?ZdR+?|mJqU2Mz=?hW8A
zQ*Y~!C$0Byt>dM<hrb?N9b#{Q0-gLr`c^uwDaZ3$lL0SLrp=2l(yMdkgb`C(6iyM(
zb3VX}_P^1u0@<W#NRC>DbQj?Jl|eWeXg$T-2Vs@yBA#N8Y6I5O02AK7Wl5UyQfzE<
zIbaI4>NbeK+c<#Cq1pvZ*jNLHQLQHJ131o9e~l%VZxwT&Y*|D0bHcv>W>w8;p8!E8
zUsw@<Qn$XW2bB6dKw-IMzCPL)ob!nMFIs&`EeC|lxI(ts9F=?eWfpiKvw&3yePCAe
z7N>Z~6GTuzK)8%-{qKEQ4DRGy<$|WN-I^uG&NcNU#9qy+W9ahjCPx5#+CQtBaZafA
z_W;_SVZcwhAtjT@)moF>ll}TBQbff3q&;kF`qtw6g&|jbaBjAL_B!s7XOO`$TjpE+
z_4(FfZ;T)t=1KdSe?Mu;BYv?Q58E?pG@rF*epQGTC?^_H85N=jz-_x6&Q>3I#$H_i
zlwLact4tHkpV=P?(cC7smG_yb087=@-ah#C!82GBAql(zfxDB?G4v$bWmPFrDBRe6
zgAx1It*<hvtkekCui&bcyXSpbf>6)W$WdOb-@A5(=FeaaMyiR|*_T<@CA<=i#w<YX
z41gK<OLn&jIk;~XR$2B$bMDJl#F%)+8G9aSodX7l9Y|qvV4}Af5VTfpfuyGNrV90p
zw~B2;@<i7b^Byl@+(e(+)`?&5g)OhGF@vvbgN+;Tsjo+rC=Uac;C`6qlC{n1uQ+-#
z8}yEapq8;>6Ye6VvO{2=K~Bk-c$9#&_8EnL^cAM3=B|$g<&~AAWxl}(Y_Pv#idYh$
zWSPP~iUKjfvz;F&$#?fe-X9kr^f}qs0<YJ+ul;DfPG10NMmaVxB}@R_2QMxySplqd
z-(NU+c1iF8@H##8stIrgm^nPjN0T7pQlyI;Ip&56c>>ri^#$w%LyGnJHw5CB_rtW8
zI89H}513OtBAz~dx~%qmVqyYLLIZ++j^z7N!Nih9Z#-i*EGc3ML-jF+Cm#TQs;VfS
z=W?a9#41J`M=5Aa{f~HJBnEiSAYh)kq~i;eewP5vS&-F0PO>|!s}KcTaq1&b6p@f4
z55BBb+0I1B^j266Ni(PczMtfx%<3qfwIMqINOM2}FAD@b8_Z{<j<6WPkJjBEgjU%f
z?ne<aC~(e=vi=Kj_^n6{H@BKxm3Y1Yz=PywW=akh?wy?EbkM$6oc>KtLLv!dWicju
z98ywJi*_p_YsWitD*sBb&Q4GL8~OS9wT6Gx*S8|qLC(Xq38mTrgL~)8?Z3K}99vvm
zEX#H`tabNL>%Y`u)MYDxP|IrmgcUS(h=qGs0$e(&Qz7bm>w`e?qaYJXO_*Wh;kOtj
ze$CSOnr25&(D{!pPe>QNU2tUE;-abAs&%^10y@l7r9MiewvT9y<}_eB{$17@AY);X
zt;-N`^0p4}M3;d!h`(cXE_OU^BcS>Ucbx{e-QwgoAQkf4u3o}FQ|vee+MbQevx7zb
ztewUiV+P^+?<^{6$wCf4SKlR0!noow(Q=z~%g>*Zbd~`pg-cqk6phaRBok;p4dB8W
z2nPgWcvV$Z^>LnFnFXvc44)qMP3ajj!osJ%Gh|Y49^i1^PYcv9Irub>f$(dx(nbLm
zFIL#j#OqhQW+b9Wrd+ZPxqG9S{Cc_dFG!W%9c^lzgZe11pg?h&u*16pkbiBCf18Kw
z>)Y^Ddhuu;Ybq*GdS`PcX2QAvA<cb91qFq+ITYv|6etoh+J*X3IOJR;=E>=cgxS9H
zPT8*V!8skFw4(Ar(?XZ$z?`^z2=0HFuFIxy>n=&#;@h)s>!JJ(@{^z+m?#o+uxZ>f
z!R;F|2;O0FTt4RVAbhNn;t9kwi?wXY>Pr<luE-!6hc`2OW2y+s){Iw($F0`9dq;){
z8&{yA46>4JxHYXd>wV*SE`GCi<Ik(2G3vZ046UPqTp&rv<Mg|gT_tebDpTwN5PH+M
zvEkf7JThnA75i(>vg_-j^z^5&CNhbqYjN2>=g{?av3XCl>9??)OIPgB9LVkQcQ{hj
z6shXzv3jin?w@Y+Qhz^}3|fcfTuJkylcNQoGI?b2>&`{ZldC=JcAW51@yi7QSh5ko
z3$6d7Is9&TVKvv_xrq8YxADjOlR%N<yCx=W<dI`eD32jaa(8hbd^0n%*a+qZ+gn|c
zSH5^u7PT+KxntM(%OHtHc$b=Pt{QK2qWvxUlQ|Wh6Br+V5F`*#Z_I{L;VNK2h}La!
zLnOKVrH3E&TnXm!k_b-7)cCH{c6|ko$v8fA{t2+`AxkUR!Ck~EGY{Ne3wix@6_;o)
zwtD-vlm*S_8jch~N1Bliob;ALN7k10>;%vuoN!XM2U+Yd#s!agEIP%y;;~o?8RG<w
zeO|0cyUYEpE65tvrvaw+TZo8>0Z0L|63P8#=#o-eQnHisH&_UuKo8MDl26%|?|#tU
z+%CGw`{)s@2T8{bDFgaQAWlKDyi{U=JfDVP4dj9#3AVww(O<9?qvIA1&LacjKw>s%
zfE?Unlvxd~9zX?RN<(jt>%|G&Lr<?IUKJoN;Nr#mIjPd;hjWN>2n!Y{51X;>#YW`#
z_&^LQ%1th_qWhin>Azf-U%*cSnPuve=pmPR>di5-Ay?=e<P%Uxvon3sr#yA&>L)16
z+Uo(g2*~4q<O<5I#eYBFCN>LU0TLL%o^OE@vAjjV^76c|V%VV09`(x;Psp<gq+Gt>
zb~GqZYyIAoVXe!h8y^)V_29t+K=4m~>BLTbu-R+lW`^Dsd9Y`Vu$ko3YiU~sv=n?a
zO-xlvSk_Nb)(OOtQ7v2j?s-|tC&RJ|1{JQ>;v<lyNXGRygJcgBIOT9=BAuHPI!rF`
z$@)3GA~@Z{eEv5lyMsLZpl#VhO?W5D1q6v9-xerU6l8h>(2;*fwZqa^1V2i{R=COg
z_7nwTA|gpJ8mza(%q8z{Jv16pQ(4N}@kfXL!UOsNLhv5mJuyS1+ww+t6uWLiYQSzZ
z7ceQiy`T8hLK4q>aS6?#0WwKFh~Smmo=`+icFS0b=)tc?rM^L3DCM}|Bxr-};?^os
zfaOHT#K@pw;ggtKQo=H4utZ&?8z+k~ap0IQDrD0vqXiMPZ{blyG8+fo>e3jZLb!gn
z!s8MXw6d~-cMz^(Kjr28>jDGA7JZ4V3bE~I18hL0<eCdo8x0^GY~wuem?5^K{#7;j
zYzQ`vNmyS(0EC#B;fNBRm+kl6-qL$;NrzIic6)@vkwXm=`U^t*d>U-QT^?5t><Pvv
zF(js7+1rP=j*hG>EV4+qzs$3>U+mJbPF}eAErUcvE)22n4;*I=1{<){i{FYmZE7;8
z*!stC`3D9@@E0iJr^0fS-G_H4cB${Oq7QHZ@rr9KK<wfQuRbq#d17E<Mu1C;<p4yz
zZAgL&*p%ztBh5Pk_|D$HfB!E14DTR1@IYGhEgU!6f-#I#C3wL5i1XGiVjLWtXZy=2
zr_^6QfA;51sRQ>b0Fhi_EgdZbY^4u9*W`xRHfLPs0Lrp!27<)yo$-}9N4;5Lx6|Dh
zK*r|-rvD1Ibw5=TpI_GE58U>tJ&KrAk~yq(iLoVc;H(tIcDnir@JW)YpJm|e20*4K
zNr7kqd7&k&pZoRS)&{vU9U4OfyA!Wc;(+_=fGD=+1_Tg_<W~nX$Q1AE8u(!#r<g|>
z9&!QE@+;#mL;auD0=+UbJm7)J_+X*7_}1c9GC)KR!?j2^1@hU+FIBE6P(K{J1Y;5t
z-{@ahjemb3n~b=t(g?0OP2^c^rBZ?bh_kM=aGI+u>M7`X{!R~7CCiu?+n^3|<pRc0
zT|&&P{1xE-EC+?V^~c1;C>Y5>V;>j!(_C#>v{hB_3w*(Z`ai-CIqnnL0a6|$tHxJ^
z6_nBfRF_@xXPf)byfC~S*{8M44X%mYbgw|>MM6!@(LFx_hqL9*@bqB*GzdZL&f${)
zWQh+MRXi4|9SEg&JwV$0(17?gz(s)(C&RMK1odKKKA@g1`HW(TmeTVo#Z!CqI}?cy
z^H1P(REt_2Zd+sxBT1m6Ai!Y$iO8_^(*?}`-UVbtgsFV;^+n?iRaYeDo-nLrmx|lz
zTsmri{;M?DWTiQ(eS-JD#2I&{pF|elb}CE6uIw19-`yU5Baa=`-2$*%c;kG>yv_ms
zBo&wAOh->IHDSEM%*%3pRx<n}ivUB4D-G#UFIt<Br;!uLX`~B~<k(TiFWrtc$bD&&
zqm9An^UwVmr+<siY9M{sOM7!=Qb3_T?z3h3pE8VFb|KqAC(ZAHQWb<!9v-#m#GLsV
zIX)}5$19}N*}Tic5T0*PC(86Kmkxj@#zk|Fm%PM+b5$f&CT{{+C2p2hsl0-ttQ)_o
zb`wozP7VN_mn7q21m=6wjT}47S+f-P(sBz+9}CCHS4u`y6`)G|g^x|zRrEdp7H~Ky
z$qn9oC-bx%Fw?`V;<Na5*P&?Q>|m|!V_wmMQUgI}dL6rBW=XXd!89(o-38MdzTUMK
z-#{ytC<5k0g=_FF^G_xt@_{w94TKCn3G+%8;y@d&M;|ZdS<CTZv^Sn#0U*n^Gi;L|
z*~=|ORimfJ<M;wiJ_J=XU4a|cp1A%v(!uW<guc5Xhg-K&wmzWcM^M<JB}&$O46CXu
z=GInZ_12fLD#e^v2JxW!_R%n&bjqR34<?dI=ND1P3uzK5o7Q#WYPcK5)@SVI%=>2N
zCV;mXXLz6@rOD9&wQcNK3Bc`xaJrB36H85Qr=^N7{B*E40h#i&gt~eJ2chWo>5{2u
zYIZx015*QE&{MR!prDDfg#!r|C3i2<ENYIdZ(Y*teOUtxXC4Uz`7W9(gS^)LHUa_a
zeO)=js>*GE=3e-vioI#&(Liu?J5CBm&`i~##v>kU-cii!(p7DBg@Xv(EH!z^wVdC)
zEnAZvG7Z^bz(7QgsPTx{6AhBdTQXkV`Jx;rlP|EQzM*txh_m8{d&AY5@>EB+JX{<@
z%lqD*u7xF!!1VISiO%nO*7muSdo!%Lp(%LJ$&InFq4+*fG9lpvd+a4~SijhXqE{QC
zJaV~9Yfc9dz1q1{QH{)KRXQf^KcOMcXwAxG*PGCje?5rGSejC`@!k!KTU9-1rLja)
zEqghumE}WwiuI-7P()NZVA?7T<#oth4-S4rb-L^$d8zt~S!7kyuW;weOfpOFvua(2
zZm>GYzq3Euk)3_ey=Y{<Z&KY@Ud}hPv%R&&9`;y8PG0iJnqPQgKwmd5;^?*iq7B0K
z_+ersl5UT~Ab9Od$~~c%g%QT8>By)kRz{4CY-dJVuix5}Dqqz1ouY+e^d|DU7auP5
z^CK*ZqdG#umVD>RvrCPFqV7DqD0x#yxbMPy{pa&X>@#kN5@@?5h<p0_1Tc<4JY=~6
zae=*Y@s=}_YFplZZ&+l^l5f=`1#|hnw<8Jp51s_{R$NJIB8jYyrK!W!kI0Pb5RA4y
z4|w7eY>)6F_7WJFUw!cYRk}~zvQy1BaQG;|BPtfRUpC$A>PV-pr>EzzZfCC1#H&sC
zHIu58{3fnRStg&8Qn{?GVt>Xh#<PM;%;><#DA^@Pyt!hFZ(5lQDjJN{FNq2e9i6kw
zjZcNI@v8)>Y)*XJ`UT2oD*RT)AW9%jlpub8Jn>5D$yFP<2~vipEmSb=W|^e~MXUe7
z7-`Pv#`{gRlq_$~pDxPGfL@K4f>&&8>g(?CS<Q`oBJWpBHiydDX=O&1h07n=S>1%h
z)`7^+JR{;U^)Hc!3i}GS+W@8fv2hX#)aEitEgT=>t91jC2#Z=Xbtxuz7^Df?*C=Qg
z<VJ^*6SKZKzsN77(KWnEoQ&WRA<hjn!jUYctw!`aQ(dS;Q_mp7e1x9iU|8M-?x(cA
zIg&3QM8>Iz@*_1>!mCYf^MXk;ZJARrbaFVN{;rX{j_I(SWcEf%d|<@S{1lnr@~;a6
zf*%N9WGpC+uXcXk+sV6d;cz#JqP|O8IeR$M3n&Z4#sfR=3xllm^gd)+^DiP6Y)w;w
zamu`<%$oBGhWw0A=JMx;BaUw>D#}*7bl(&qj6c!b*8K9|j6A^I=x@*0<saY9la~uQ
z`kZICa3<kk8b1Iun`YqRZ1O=-syZH8!4345GLG014S0&FNooURPrnbef*#t2WGvH{
zzG-&ZMAd&Y`J6MHNIc*C{mYh}@X~Z-`YRQ>e0CBVxon{lNp&HNkv);~V`?A-<vpfm
zQmKA$o#sc&TY+Y*`UsTh$Ryi+L?Er{*6J6s<qMw&e|c&RZC993xv|{@LVsT2N;Te+
zkA3s9zpbxfnL8*{JvmNMq54V*H|@jU{LqQTt>SYLTu!>Z=N~xG#z8NE1hIPYNro|2
z=SZOVBL;y|#b=&fGGx||9&!YJ)5vWhIf)B-pRWITfI+J?s)VU`_sv%8&ccsax(;1$
zdhYAT&#tG;Nz9yOqN8NyiS`sDy-76!IBM(G*DdwlrW6|K&obfsVN(0T@vY*PuEa(T
ziMn!@`a8}P4E^pK5oyL18vkN8FKVfo!U~L$7Un?@y}rRR#^oo1)*q>0v|Z9QDFq<Z
zAOPPGlaexa7wO3Ro#Z-GZmlXE=}V@aNMMIcX}ix0O;8$83i0^<G>;4A$UF@z{9*o&
z?v2m>ngQGD0YMyV&c~Ma5NC1?cRh-ziGy+C4EfXP1b3xefQZ!1#jD;UpAwJPVT~cX
z6-E~{Tp5nNZlBGeteCnq+UIE#w({V}dIur{7Y1K=YpZ0*ECH2>bAgqN*BE>D$B@#_
z>wHGNUK5UrUB;V-dKPh6m2;X`OlrRjhY+-k4GH@66W?GljZC^#7p{SoGGmlUwA5=k
zn;-S9lX6Gi!&}8~ai+<Bw&bfT+tLHX?~ZxDhFlhwhNYg_y?rYRNgQ5w)^(K<-@~zR
zw!A?hPmhGkDU52kUQ?31${t4!x$s}+H)P;&=9&KxuYIp}x<jTN2HAw=@ReC-ceB1+
z;!PKqR<G{0)(y-XFPAc;C7m)f)BqaykC8V0KXDR^(sQvqx?Y*mkzy5PwUA07Bg<py
z&^%~FzPJ>P4^U<n`U=nmL45#{NWHyV$yx7-v!v2oe_lPsQvBWBQz5DTOnpfo*FQCu
zr}WpDqD@y(!=0LkQAC2cnO`)XzCNSn%>A<QbC0i^{d)~rEt6vG!zbl<?KX_e%reNC
z**LzE2-lNcAvV0Aob)n1<nG0tedkdL6>IJh{`V8_>Q%{55^-T8RW(8MUCi!XyC`)e
z)c0mD;%hv4R`e}NXy81_5nQuY&PQ#(v)C6|m4fNC)SU;3y6+M{BdnO!TYDxZYX1BA
z$SSUq$z_j?Da`t=iUDLGdw-r68!u5Z{}`!0KgcI<Li40yGpIkQIloX3mGq{h(0~M@
zh6}-<y;2OowMFy0#5Npci9O3NR-u#6N^Cg}+m`gh7;)n%8jH0I)C!?%vk|DD7^!&o
zAwckqK=Ue@YLVsoxu<u33(1W~V29EU;^I%)KaBGm{r#VuS{53AS{ad?tKJz_3)4;K
zMhat=4a?$6WO(?+p|EC2o&4F9!Kt+Ekt~Z34V<j#>g&hr&iMgR`XGH6g05SM)xLi7
zoWYoi`B4t79oskSh}U}VO^QJhfga88#FH0#xuO;Y#~R*!i1(J@>nD*1f<V)EKV~_C
z=8T?}h&4DdSZ*`lzb{cv*lx25I-m&6T{PY9e~OOdN<Lsrc|7{^EwvrxUBj8|T)r>q
z*<2aZ`K7iRjON9`*{0$q<c#jA{2=EsR1_sWsGGY|j*cJqVbSL;e9X^kR1z4mRU5r5
zi%${bitEueI*7)978=GhBx$_C8-MZ?2(DtS2@wHgn4Q>5f;~+o<!{0YduDs`=4V*2
zdFy|bXxMQ4Szm>bR)uZR5FIZ)(FUET=9Q~Db^^PUb^52FVQF`d?r%0Tl8^|5sqfod
z5(z2*HS!(onY&UvOCvXTn($3U-$M2Q01H>M8ZC$$m;QL;4XX7s(pM!fMR1R#zx(=~
z@L4A`Q);mY7vmSW+e#Efyc)B;`?ij3X-9HF{wM2MmV0hz2r1QVotukqSUU}ky0$(&
zB4^C?lo_9UXilwX?)COMQ1Mj80o6Y{K#`FqR`uY18~Yyp*ERO6lDj0(Tw(FY`nN_~
z*IqoE-hF=V3!T!^`?{XnasktH-rv%$+9-P>CU#zx{80IQ5pFT{UDwvo=xA5_GKWib
zG&@tLA}H9*{BOE!n@PFj&SnbU00eKhB|ze{=V<})ANDx%%%VYai>aAEGO|fLo1IZ`
znOy6%SD{`Hqm^YtK#-`YZ9!<LKW_}Pg{+`aPv3hu`Ifuyz<}z|+`(AmwHB=nN6^(@
z%;AO-+utG&M*10B+^VK=``>y?wzRmsEr>{bA!|4|%Et53gLFSgn+SU_N_WJO<-We&
z<?C{yMIA?t*9V!s2DjaKp2+np>kt{^s(DzyVX~3kjG;VnEVy{V*^KrbZh|D=KjN<s
z=!6=qhM3%j<!m)sIzj?MHwbe%`dHN23K*oMNioPW(7gz|QI|TE^*@$A=e@!6RXHvr
zMlMwzCl_kOMLho0=q)P1Rhjl?%#^Wp5pQar@bp46D>u4vM5F%Z`Xaj#?%vq9q`x~U
zt=4WD6ppZC+)qMe{gT&iws|W0xGStFOOHR<z`&I1^ulcA)caJAEF=B$A%k{wSLALA
zhj|IL6yJOFRyN_q_j<hLXPK65WBX?o1a^!b6$?Aa8`RV^*a15M)UEWS-f3%;YubFS
zJ)?31(K4l==KDY|o-iFfezpl$PQ(5)Q5mmZm+;i-+z-0<C5tvZoG+^ai5Qs~wFB^R
z3cQ96aWv^xi?jxD$`8o8Y`#nd?@fCijp*#NALr{#eI0Vcvyw55w=LcCIQ>`SH9_`L
zDkmap-77Kkf^8uPH$clKXHz}2)Y;Z`<m6G})OwVjwv1G-;p3vr;s7<ek5@{!lUy3_
zFuLcxoaBP6+i!UIxFGvdKtm)0bc?wgkmZL>BUx#zy|Fwk<zBjqinxVFb=sg2$_lQa
zJ^)Dt+{m0)U%y#&e^&AoG^i7s`<#!1u5Ju0tRUWERj<9pHYQ^qQi-|m!6<+PJr`|X
z-T+L1)IPd1xW8g8MS`aWf%b!~8yet)_sz%sDtUc<^RFVbnGe8qQ_B8M;!0<f0bYO|
z2u)-u#B@$gB><+AaT72T*&|sff0`4p!GRhfLEKf&%E}r}dxoWk1&zUAO5_jtz?ZE+
zlLu(KY_9%`j)4)tTb1`OOi>M*2O(w!S=q0QF5^OX5lOXpKR|a0&|ek#0t}M5yQe1u
z@F%iJgL2CU*#~gv>?VZg0Fe4xfJe3^CH?6Pryp+g_6Dfc7%U$l&f`8z5^|u6V7BD5
zeFn>tq_Zp1pr#7A?q!I!7XLe5@<}>`5_GDClz3h^V-tX8o(BI!Zl<-MZj>gcfh8!B
zK~!`GiP%e+24hQ6q?S~3QtUJ=S-r3Y8UTIZ)rpd#o*jVreJlor?k3D#`X>K_yz<dX
zXuy~!@op!v${YpvAtwt<>g<pApq-FDQOM!r<<*r`T-CdULL=88a>yPJZcIffzA{rx
z#w4H(UcOBoIyKz%0Zl4yO8;LF+8RD!)FUMJs2A!hwJJ`<C6sEo0hSt9gMFS2p*7v$
z`Tl(F=E_8^LzBm63&74z&BN?rVN`<nUBp6oRfRx31SC2Cg3gI};BG|QyEgN<iX_b5
zX_^?e2_|M)Y6ybn4Z)hI{mF%vo!st*;ol_}LAQ?vvWx_2;hMjI7*Rl?G}QpR5c+!l
z)Yv%3uNcgV$@Zn1np&wv!}VPtG~l8G0*;=Fl!miaDoQ@fuPkUnL!hl(DopG=m*pUP
zB!=508)hB;=`RJa64pe$TN&sX6iNGv_>hkPP;_?P4jVgr2@{DA4!r0)M~6o5-j#Ep
zRQsNrz6I+Rx(BaY;>R)Zj)u0ny6NN}(k4<b@M43_!7Ln#XYQam2;8#(@e`s2jm^{&
zqUV{jtoNF<wtt)$Q}UR8E+7Q0y+C>>1c(qIxKOzSZiD16xI$Ozc&o&s_z)}tJb*G_
zY=#2H-vPa+U(nX#@)QMVJS6%uQb0c$=@CdiH9!w7K<-`wSyO5_K<)q;0^CT+cWM?u
zWM+8ek!V81U;>m201_oksM%-sJ_beHfcMRbcMz|H#(1N*I3AUNA`)Kd0XVQ!Lh#N6
z#V+Exz0b?H<0oH+i867qyqXrMdNTpK!{&9t$xuEKmEwKbj;jYG3Y$)y4@0o#sR4Rv
znF=Vm&DkGyEP~df!JvQ6`h1yS06A<Exksr6gnK~x<U<cX;X_13gjql<KeKFdlhN+G
zeE?Ykjo809s67ctOYxe}sy5NFt;IjoHIH&ZeSo(2!R!PzbYXFe%j3OCm3mAPpSt;q
zR97XIB_(_hmsuda1h6|LtX0BGy_OVr+c#rSnRl*MQce_78~ruL4ti^095{(%%<Jdt
z{`n&3?P)n6Fala9m@E1Dv=|D7QT|UbHM+RC*g7NcG*Ro64@?55<Jr@vk`zrdyQ@Fj
zLkflWRFsrRS0?T6%{Mf8qC#(A<(<n?J02Vu0E49lAZ{zTVHD6CoyDV>$qh?5PHwl+
zAeTefF|$=txU4Hwmf-f!eor4*!M9BXn(lz8@}$V9&cZ1_2<RRIfBZdtQO|e@6OwrC
z<2HeTfvucN(Do)!5-#t6-nzs;DIzX?<xC`YqUCM}7JDr0?8F_zuAnh4UmLU;z6SE{
z6PWPb-`|hm-=cM$M+ukRn{l@LaTF~?cIS=+=;2Jqo$mu~t~lp$5W;e>PUI~#pSc0v
z%@8AGq2T5w3^U22kOc72by^c$>w4u@TA&BaijBPs_Xd04T+D3{c%ObM7`f;De4dI4
zN@aq(<i-ZJ5U`{mUi5t>fT){j9Hy29&b!%*JL!*M;X)-W3)}(dS}F*ZDx;Lyz>KrE
zvbGy<yxd5djv6kuMTVS|V5P4N)hbXA=YTzd)r5dxoq;7IfVS+?U)qaTZ8yUvpa00d
z8*-;Tla-xaf!3z}=g+5`;B5rJ$|!*XSoFw9KJ;A&jf+3SCU}d90H?*Rd0D`^-{~Q+
zUp(9@ZtTto)lgNH0AwF%aaV=arSJP{`##NqM!zXv);so;aC3cPID!YP>p2xX;;RO^
zH`CSOaF??+6xwDJG!<J0V#9ZMkCMw~8351t4DN{Jht}qS@~M{sgs5>?DcVnAN5tK5
zN_TaA(ij6iU94=k>HGbEKK*S+RG%^qunKd$CjRPPhbcz}-~{lyIc6M}6IwqJh^-G>
zt3$6_Y{I5eGycyLGq#_<268+2Haj%AIWpP5<zwwI1&ghI6DB&j-18v=U!8Fr&Utfv
zI0r5yJv2+oX-pb^>_A}XQ(q-MYzu=P>0)s=812PNwQ?|c*Xra7{QDo19vc@7a64EP
zUcqPaX`dJ9vMJ;F?+F+4*b%|+ZSlTbx(9Gj<7%6?R5mRCJ!R|l1oWMh4)Q;!)sJj5
z@Mg9ML72CJ4U%%7>36ymjziiu@O92GsG1pWkFV*Sgo~vA{zkK%EP?3Z*8-GkLEC)1
znn!TM_CKHf@r*p(GP#jg5j)ESf=5i-vT5hPAHWR*7?srb$Mtdu5M$uif^<>(yjrO2
zv8itOKND`SBLbR;OptfMA1(^~vZ)b+KO9x$`D*SXT?hK>j2b-RiVSO>uB@yaf|pPX
z0R}uTJ^f>mNh25PKEN_t9pO|0$Pz?m+}4tYL61qvRjZN9jFhpl@o<rm2;7kCv^gw?
z82d*76BrT_#sAM6Zs2Bfr$6Irv?#lVAkEKw@XU7r!1T+6ThSbPPwndWd<%rbnA`Ed
zD`%d^fYyIlJ`D5gnwpxDZEbCEPckgUUk9-u`~DgWHOJ2X!|RI57Ooos=A3^t`E|<n
z|Eli2<Ej4t|Nmn~9D5%lNhqU%gX~q2y`rd;G7mz@9+_F`pvc}bO2$c&j0(x7agsf=
z2@${hQ?J+S{rY_0pUdy}{rA^DUFtl~^E}66-tV{D_5M=lkRCGW$&(IkB8tcR>fgY?
z&s(?$T6w)aE>Yg^ViARbtwDVrqBKK9xv<D^U!5{Q#Jr$sXeq0KWT2e>Yro%3JD12P
zX)|CEYXrISakJSQZ-OW)ao+di<5gka0nz2$r`kIwfuYlCp#$l@d>D;dEPhW3;!w~O
z1C>bTzrRjL(vq^WvhLs=KaR&q@G4FYHK|QQ){RwdR=#Nvti8v_KB_HjPsc5FjAh>8
z&3X+^9**+K%q<YiAIEp7B!#WbHLGoc_Lv0}DIomG1%1kg)srQ6hC|pLOApsMem>y^
zoR@28Wp9u{=>mx-QOfQ0?(YRV2>+LxK1ds-L@^2TZ2kP|EOPSXgIf!OI^I8jFp?CC
zBbH@keRRbIETqRJ{Qw(fkR)!|6t~H9o5-I&&GCBY*AtErCD35|g1W(aV;2Oj)^bBI
zEQ2OVspZC(w<I$ECo0MK5fDpW{r>s2+*AyNh!A;)D|-o>Q;mp-h;q~Dg-_3GT}fTZ
zGWIhGr?#}VW`Wc2ad%J8Y;C|Q0#j1^iolc_iQLMFnCbNlFG4mUmD6J%8y}A-3`Vx7
zetmLv2Oh8QgEa&<NY*TTSE#6|8OvA$tXwX^yqQ>e<Sjj8I7aHvjUy}ZfA~rh6(Q?w
zBUc>ip5nuvZDwifHKj}6kqeaAo|p~^3L@Q1NZznEOcNi{%hTMRez+^<e}b};hWFWU
zT?e}-W+|LeFP0zoaFhV+T7EVm9z`F5{C=-sa;<C`{-Jy6n+FXpq;uRk?+fT9YaNaq
z^>JJ}6Hkbe)3VSMxJ$c0Ry)gGr>Co1y4GL;V1tK@5TkSGN6vU(q^{-eRZ5}98GL6l
zBqyL3<g5>R$KLw!i9hz?r~@6ufewUb%2LfR?o&&P`BdeULeOkT28C1)yprj{{IvMf
zkcM~7#)V@pQ&Zz97qdLUBgj5*Wbw<9#Ep-MAYlt1zFt?guV_h%&j*e7`l1>=5I0=b
z@#y5fd$vJ&y{uRcy&Ee7jMxvZ#G$r>Rj~e<-+q3VkV$_gL!P6onu(9k+0C1Ar|Es;
z(jJ(po~!N{-Z^t(cV(5wzo8;@p&zt@Z@i8Vb<JR~8$qsCOaZ&}RZib(SY6f1WCBS$
zmO*oLG!*qN!a4dhv@*QX?q-LUtLrl#>0Uum(Q}DKiw^I~_+KaXTV3^=+vG-h#)mec
z+v-TF_;FJprKMj6YLhV7>MQy_%3gcVU;}f{@}tYKh<+*2eq;cD>4HO<Y&wmi5Nf{o
zxNCORpKro?lime58a5!TGd(p5stNpu59c<|xv=>~Z&p^_qT3)Y%96SWg=2K+UDaTo
zc=7CsS=kW;s?qlmYtzaih{c*IGG7e2<mNY5lFF=T=JOGi%rp|Zr9$jHd`KS$ladF5
zVzPIy0`S2>R1JyZJ|A(>4*K+yinmcZB(9CG_;I#0Ou|?<u;JT4x>!o?yo<}RP~ta+
zbY?|t;^e&!{&a1WnKE%vG#v9KfV@?Q2%gJFO8$P)bmNngVfPvpQ0W4SW{H!r9U|!$
zP&eR6$8by<ni`KmT0-~p8y#G!;m#2h9SxHomgSdUjJweeoq=-&5{4heDNFJ1++}2D
zHp(7dw8X$x^+5-zUgPA)FR56c(xpNsE*e+06b7t3JyXHqa_@Cqu3oz0b)H;kuv(Cx
z^P%*VLx@U!lMg#&)4c;im5H5-P3Vq5@*-+bq4pNLEe4wviiwFObf8Lx-0Yp!Zs5pJ
zKjA2PI0|ki<YT&0SrnhrV6hrl0Jze5Jpww8r#q~@n5?RIBF&Hv$NUN*3=hJCe-Ew<
zJoJ$ikGs&YkA{SVEOTaf_>^uwe)1JnO^w5#`VZhx@Z?oV$4TKONjq$B*QcKNWiuCp
z!3m)3!!a45gkirpW!cZbl4<AifqQjzI1(;h@1*ignoxo|)`Wga3GYL4beBVyROS;5
zlFqQ!%mO_D0D};{J07nNuk$aSk{WFNa}>OP!e<Ec3>aj{o)?tD%uVP58n_NyAW+e{
z|M+yA3wH7543AIVpaXi1096omH3g2DC>=#;M>N&szk&!PYw#{$5|;f018%d>TM($<
zxjQA{ze%$<R8n4!`AS*MD1bYKsG$QD58wh`F<_H<^A@?M{wHjyvFJarrH6IL?FVaR
zL0;a?tS1kkEH-|e^3?-4NuDTW+4lfVu?2Kk-hodrU$?flsxsX58m($TSW2UNqH#7`
z<`GIswh_%971D7OjLwpX1V~_cWrZ0ncj%BhFG&I4A3zkR>(>!!$}eymF+qeb$zve{
zX8;c%{(JuRYd}{4t)cTBpMTUn#Y}?@UCa*4ow4_faCaah%>4ZPMjMysi6*oQWfOWp
zm9&rQ@wX1hO81BJg7O+Pff!^J!AoLG0OejM^v<a^iP!kk;V^9o%gI^v*E93u0#5Sd
z2y{#Zlp~uqYm46c`V791o;x2##vM68!KLIUKRz}F@8}<IOZ@|Qd`A!92$b-hp_kV^
zANGbTsa(Hy?EqDfk;^e!dV1`ZdP|Q2q3!K+<Vb#;3?Gy5sU<E9*edc~Uf=jaVbw(2
zxG;21UbN#08dK&pbRl6gt82J1?4BF}+l02=ACCDLLf%SO+j>;O_||l{puyxRUeH~*
z4hs>;FCE}wW7CAzn>7ewm}jYKuz1kmAtKLOun9e>LTaKX&)up$jxpOWT{_T>#7_3P
zgg^_6ydQPctTfyUmL9h7orjb+jlwZQbPU+XWV6bB%w?-ZgV(Qs@|Y@U=q7Q3Fvsbd
zVqB(5J5S+CMkE;tbvqDP6b50dLC0Xd!RsXVzN!3++vFkvr_yD>axY-ZINzT*zk~ps
z%5B;J;1t|}h~c@Edpw5%_Z5KJumT1F110w<2G|sZl9G{W@FcU1P!a^1(C-?g^eaCB
zK#&mp#unz~Pdq%tR<{m?(=gX0RTk+bfY~En3VA$3^YB|_)OfU@py29zf!IBJW$Hp!
z%g2vwHmKVTYAZfw=vkmpBC))&(NTe!Aq)7H2m_Xjen9zieL7dP%2QaOL5~FV(R?Vv
zo_dR*lqAtj=tY)rOmrA|Yb18iQ_ZpLaudeix}D9%j41sgn#=X;xHN5P2!SMx+%<d2
zh4BT2RkqFBD9%5)C(cK*gjZItydrj2zKBPKO~6`D3&INQ9xmc=UUfKcutSD1Zm?FQ
zzfHz(n^3S4k^!nS;LYeA19?VI$%^l<RYsbrZ-B$gqWc&KKSSr5edSn4{&JYo_8S&*
zpcVc%`sp!^E15jch-Tw~!1(_U{*%nU5lUMUIP~)Vj)1VCJD-*T@tO5eo9Mjff|Kt8
z@Yq!g55kR(v=Jl0&@mo3bp;G+iX4F_JU7RabX4N7TNf+s6NJe;k06XE&T1#z=+Jo>
z9HqiBKf8QM*N}h+vi*vlLO6(fM!7$`gwNq%_406`Z9;b*Y(j5M`R#WGQB~dr88EIW
zr~52N2$R{qfssZ9RQ}B4fD-J2!AygbYad(!t!uWQy8>zRmJw(0feUp$FGNXfgKKo$
zF1$-~j1D7^ScDM_lxlP`{5Hqg4a@jjX}&K}WTGq@fE-oDh}g$EkHIDG3&*Twe<hDh
zDt-;@L*cE>%|{d(D6!$HcBjX#DerEf(QqoSK!O3*(u&OF{CQzqm*f|MYc|1`S^3Gv
z&tHG6Dm>C}hp5(q;mBZu!70la<&yKyR1?F4f`b{!RaOw@6&7561$=Oyn5O72(>U@Y
z|7J$zFbZhC=L)zr2gZVih6V`Mq&U}QssvNEaP?BUlCAD{ZIu2Ntz0-jcnE>cpaWBs
z)gHJN&6zq>r5X|SN|pFj$4o%~Z%JW)IHK8uyb*Zr7Qi{n!oeZzpnl1BEx|lJAejNH
zyug4h?&+iIMfCjU2a`|O?1xj*A&*28i96U(JF380)re1Zu>we>Iot2|uNE#vjUYdB
zzvK^pTM!HQaVTRIV!>HBXlZGw+_-X$6wbq1?K@8fmqs$dfJGG@!1d<bV=RCzLbB6p
z#3@_l1<{4#;gDrHkmozp!x0}~E!xve)r(kdA`mVRRZxaEaUVf~qi~A~PLDbkM!2HO
zPHn8J8YIdH2A$K1s8yYB?7hC^aB@`dU4ntmijgOu-v+<J@710xVy-s(FK|7AtTP-C
z1zON*S;rBtRhrj~h(Z6UCN%0R1GdQ4k9=m&9-4)i7@1JVJks^oS>c|6_T`!#nRhw^
zq1x`nF0w`->IOcxVv>3|_Wi(sS${Z0y;Kl!zm%^k5mf;Fj8!LAIX5Tgy<wl&d#WJR
z1VOLBS;usp6W+~Vp9|?DDN4btf&ySMp`<*%1C43#Q_M(cX#V2Q`ES~=9>^lX)LG-R
z35sN89Yo5GjL*w}j@=U*7GuLPsI~lcLgr`20@!t)Ni6zVMvD5n$D3M(mXfiv%8Qju
zQHF>-V6<Q!2Vdvq=$hDqt(ZaNX{8U2h8|`vC@vNS;~k@oRU>e_)FId_H!nam<5=gJ
zSAzyja+&YdJG9PWeMGb<YVu8k4CP_8&g#L}O#@GFYg!g(dm>Ov5Ao3gSr#!h?|F;Q
zNWkm-kv~bueDQJ-!Q{?>&s(?}Ms$eWLGW;9Vtiaf>G$dbVZOh8%>8mr=z!232Vy6Y
z$`}j<qKt^t4b0z)y}p{?k}Bxnj)qscu#!p2&^0b3h9ja?uq^Y<C}z)VLkB16bn%MD
zBquk33!4q-zL7{cVdhig{(8k3h0^)j6<ufpTvE=ZE!u)QaCuy(V_k#oW+aHO{a=ev
z{Q`Iu73rslE!uUSx4scuc-R=BBx86-DSM1bJ^mYEm1_czKRUmf@J2F08Ce1_bs?eE
zzX@0W`q%?)3?r6KH6PGNlWT4-0K-q;A5tg1h&O%W>jSV8s+&JSE=bQAF6cV>4Cwsw
z4tRov<{%ZRBm?p$qPAGp#Y&+4xX>yzqXpUzAsswkFbkpsZGSuJ6Or|)OpMh?=>6+V
z1qg<csVhSz7%quR^;=l4ueJmH2KQ+R*Typ^saYiWoDs7qpuj@)*ZDoZcXPq}eAlH-
zJl1`V6yOjN+Bqq{o~Xwf|HoIEsg2K`rM9ebKa0O3pf7qipE-#}c$9C&*0JNdq@7!#
zJXK7+Z}=9|XkD}KIrv4c(rtChW+kP>8f=QX$mE!cSb=CRvAs=rhJ?@lr)H_Vt@d%7
z5A=fR*V_xW@Sbu$CcOQ}6`IgeI*0W4AOu-3%C|A~y0~ykopk>hD(}NfmD_p$f$d=W
zh=`3INrGAH!n8fMt?;t9e7koo{iger+SN<p`BAEyQZB-KEQFHI)5HSFL#ICDmt#XD
z9j*zlEj^p+ImvD~@-D2fw_J(+uwj7>X{MI=mtV2DcV<t}irkfc$kdRi$injCuHo|R
zj`IIP-cp&Y;KlYz-d9G#N9F2!mh&JkQZ{CZtz2S3M%It>`_~iSMQ=BVO!UcEq{Qoj
z<Zk)C7WGl~o<rZ6<kB8#E;)b5E{Id?t+KhDysF8?N?0{ABN5c(BK7m5Hn9?84y1W}
z`dd=3@42HDJXl&jNkuEo2VPZhADJv`swc)P7W2c+;zcZf&W$B>v2?wN@7`h2RSeyc
z|Cy`aI<D*S{Gnl@c;=5HXXA18gF4Af+S6gNJ;QAlGj|g3OIpHW=UOv^PEsVFrAmqK
z9}VTSI@8nh3Y^!q90i_;BW_u^H+L$Pp=b6mueRHmnIi$8f9N$w7RPewjQ=QmNe)|E
z*~{)ia>0vh-s(B{_Ywiw$bM|~*G;}2u9COWkwMpNRGwCEf2~c3V}FoPtG#?r8lS3V
zHgb3?*@}MY;p5`-vnGJnD4Ns|Zn-heE^n!kSg!ngeSnod;h><e0b#_!vvgssQfttE
z>`H2n7g4{biHc<*4R<$Wai(KHNk4I>@V9d0#Z(R1h12nNi$$G_Es`$$HZQ$)zL(Kc
zv>cQPbG_}6H-FV4vzcJ9{aJqhQf!uqw-)+^kxa9%rLy@G0lUlIo8*|<?FPFyiSn%%
zc&xrvJ1zd~o4QFa_Nz0arTn5q9ZTJb)JT?)$f~B!`u^poAMF%y1$GAmkvZjB#1sV3
z9|#uoIEmT3QLO7>5}E4@EWfT_6PcBG!^p^zjpAeGy!MP@9;?MYOmj-M@YIrLQO!+&
zU@7V5QUt$JFY&9j9lBSDuQK*~DyM($OC34KW0hG~H63Q^-WP6t)q(DNXtO&O`a{o;
zvxqwsyK8XSn9YTgaQ%JWNK8UlqJi<X$myE;`o5(KnT(L)&yAhcm!k6tD7CJ|_F_L_
z`>r5kzI4<F+dr3%fY?PDzrm`xoLqMt8?O4HpC>B6QRo_@q{^ShCdrIFLf=#LUE;og
z9_1j7@U2@{4Ic-xJa;F#W5j84zAfX+mj2oybbxHhJG4}+UWWKZf-)q`n8v}_+=E2!
zT8A7J$9jy|az~hGRqUwO1?P>Twm8<MN;(aR!h^MU6S?uM!GSB;!a4T`Mk}c9WrdJM
z7Os}dqxAJGVlavFih(sgFKWVWPj#v!)^BPRYy^mE)hYhG=uDEPjF6$)d+da8+z2SG
z`~#(-@k4>bpC}ERRyb}1g?8B!pN=h2_x=m1rhEHfSb2M5a71R{n&mxx(f8Gn6FLSG
zKL%2RG(TpAU``ZdJo{ZDnPfQlP(wxG2|MlMHBGHPF6?FYJ*1OfegeVZg;2uCho6q<
z`~&-(rYBFGiaB_yP0KfdT|YT<A!J!=Fl!+?_s;Gf{%6)>h(e6{(uwV@`uP*$99J6;
zC9hvs*m@e>*VMLRWP1fv0CT)`aVI{$mylE)myMA!I2a>l6iwPclNY^#wc#!K^7z8S
zKx?X<6t4F`(#gg1Fm;?Pkv$Gd;yWxeg4q7mFA=9xXVW+mM5f)MXFnJ#P;%`!g=`l&
z0VM}dp%q>|{M~cw=>2<4_}(9)f=30+BT_C@rfK!9ub3R{ALI1Xa=M8>-`9aHB{NyI
zi()E}!|hNRTX#n`tz`HS^f)gm%{y3RCJ^pF@2%zgV)3#i^W)*8jCtwiujq_Fmo}UD
zUiiKBK5V_>T5f;ZditnV1MLes6Xl9>=~2MHo95vYxXrrFG#@sHPKou`8HE^{#+4RJ
z-4h56i?TBvRjs!eCu|fMM8wORuX$WpdYYx)ayG&wi|oTddt`gUrLvGEspY>F%Ap%c
z$rZ{Gi$=dMzP1(2%$T>G`B=e)6}_`f)%$Ne+zP&DxH?PND|IIOva)FNYmPIO-&c&z
zzBu~U@XXke*7>T3<vSj=_JU(W4Eqtb+s-3hjpi1Xxpy66c@^xGT^Wz;ZWYsByyA+f
zmU_X$$Ea?QQu#_cLSPSgc1O;sDcV}juipC11j%=~0oR+MBPGvmJD=N)DToEq@<j?C
zbrRFqXSRDd7n5z9+5Y>;J(2NEhUCh?EgNz{XI+rl?TpQhYgi5Lu+#}Blp97lbg&wY
z;P2pVt9*@QZE9IGv<)4L5BptX)lf*g5^lt16u++xz1yv8cG8x*?7fw~s$kMwgg}Dv
zB93J<_S~U`+;`KSZ!Ij2Nj+AjU6QcY<j!SdF<6$|^kO~ST|D%R#aX(1${e1PeOv_o
z9S9~;MS4<C-iVRkzh*5<y7taJxA;S=P&y`5?{=tOkgS~ptD3&#7S@_#@Jm?a-7Nf>
z7eRLj@h)1t6>qIWmXuQjE=&mM^Y<J~fQ^EUieZf5wQUUT>)QsnASP!0-?c@jd0vw}
zbV)x^ha(PDsA60NPMeGFQjT67nvvg>5-R5iw5XN0px*T@zgervEqsEni}S%{UzS#e
zup(<RC2O>Qg4BfnleyG=%Xk7wUq`mtD4sBMrYBtYbKrTwi!t$iQDSS{f$>Lfi52*j
z4jX)>Vo^%keOo_%r|R^~?t@%W{n#OjCe9tMj+{Qim6p-d8ApyB0kRsBm?cM-TZ>jY
zeFdHV6W;3ToGL{?X*Mjj#G1)me?ucITb*N1F<4W|qm-nnJ<%p*e_g%j(DBx5!IvbP
zOG~Zz71s5RSIk`grtN+`eK}V!?71*b^T4K8%d@9nE#jP&B}>}nb*bO#-3}Ty8cU|#
zQu3XNL*QD;`911H8MAW~<+fw9a{}6>@tU8p*&=sz4s8yElhv^t_YM@rI4Yf^FqO01
z%#HW@;V`Z3Yy9MHKI<2*-_HoAHlk%(Q(i`9d7CwpBjCT^juf1V!iFE8ctj@;l!ojc
zIB_W6<H?6y6N!=&+tFPWE<csODn0k0(n%I;`*E2&cYf)^ctj=AO?*f_fJy{xHW1X<
zP;*^7M=O}aiDi^l>_)#i9>ey5ir?RZuKWh8*h3zjm7V)4%^uedhGE|sMn=WmzZz6f
z>f<m}5fZ9@)%{^&t<mjr=B0+(<P-r+<<I;n|HBW4ds5F94zGVyp)?WtGf;*RhJ_g=
zI~FsWP1A8*Myc2E`Y#g1(rbgVg>kfcWt>c<IaK9!Z}8a7{`JGGl`T(f`>x%qzr1mI
z+-5s&oOJwAOU2#%(Ziv=FLGnbUdss{&20)!6e+*a`PY~ZmM)lcA&rzIXd72A94YO7
zGcBpi@2=6<0=xc2LpC0X!i@MIGLdltXL2_=S#}%F&&=~_X`gVqCXnzMU-Ro}F~i!l
z_0cv1J3iInlh+<Ooym)OD{kws;m-bCQr(n;ZKs(2pF25N-Li`@f7w{<aX{?7@Ieic
ztX|HY<qXy}IUTF9Z#rz|%$CfiCGpF2O=hf0F<JA8*;{P%>F+CwCbe>JpwH%udNY;0
zMQih>D|50pUTshD-cBE#u!y_8NV-D9Yi=hkhT+t0E7qwTpES&}^HIE8JG(({+I)|a
zvgko7Xm4$|^{W;dhp%7IA7dR*aLu^A$HPn0Wp>SeE@6(Fb?~-t<H@=I+p;uXi$1Tj
zX!4z5Z~9oKMP)|fuHZdfW&qud){!Q1IV)35o1gNDVZE}}3o*#x$v=wP`Zl3`fA?H=
z^Yz*GvF`YYXLzUkllZh&Bg^kY8e+4uH|KlWeU8_Uw9##F_qV49U1O|Yx$|H_-CXjB
zNKaFxZ{O_E%A5@L=Y=<?m?D2?Z)cq@%+VVC=3ZTyQ{{6b(~^EbM=Izy%;MeJd9$}H
z_I1}LiHc{6T58%=<t>vp3UzwvHdk(xRh5q^TRFA#C4KUeD~VaO7r<@A@9N((d8ef4
z_^z98J|76CrW!x_zl$^CJ*C6@{VY{9UQkzf+8@}g!s{&`nv4|DW3zL3b~+v7Z7YR4
zzr6AKoP=A+{QFZbhS4weM`C;Zx~_iq=aOa}Xjrz`zi-hlTJN?;*G#3VMFfT{IQv6h
zZlPpCxq|EOjFe`)v=V;XJ0_RV?Gh;=o>_d`S~%x<*m_sLRS~_@XLq9G?;~^FZBn7|
zKBvd+{i}Q17pc#)IW5^c*Lhmr@>82X!g(gxOK!-pvvvQG2O2wRb(Zo=Go%CEGvDl!
z;pYwtUyl-{2@{ch{woh1>mrqu7)_gX*4f6$PFXb9m+Lg+49y#bN1qFBOE>%fxH6d+
zxOy|`$vYu^u?wcLvsN!}=_J+*wmMij5BsKU3Td`g8q(dE_^cGAa3E5Z``PP(7uJnm
z?~N5K6uSP^PENTtCS_TfJ4)~O*vLJvW%tAK$&YjU+r0!+9=~KV5VVnq%@Dll`(i4|
zTKH+Du>IX(wFyacNts%?7{i&D`@WCUT5M>RKDgvR`B&#Np6%~@H8*2&^j=iJ%43;x
zh2`g)Q?kfJ-{#1L8R5Oo2F=;wEn~0z=(g@jAY<84y}K!?L6+c`dS!$5e6zuKquD*S
zt9)wWh~_<hZDHJfue;LGY_;*-gjpv~W83R;A=~0P2KRLO3r~cdFg5;uePlMf(%=dR
znz_Z>D04+vq&zss`#+D-MvS-G9{IWAF+aO-$Rc`gvDn?7b_$pO_{88Phq@q_devkW
z$Isth32$$n?;G*ttQbqJQS3`J8YFqa8+n_yIjZ_oW04A=N4Yi?$N-C2Bv6<C)6S#p
ztW$F|8_ROOWF$Jygly1Td?uP0`t80M?o3((8-ti$&Y2r}xt8Km+N)aPpLe1aJX9RM
zFNJdCeRQT5&&+sw2OsY|>io)Fi?i&me!(#J+bHKRCdFJ@Ph1=0;XV31$!9D}At|lL
zTz&h{S=jN+ed}Gus~*Pr%|jFw_I4ioO@2+pf2qD9`#>dPmN2Aq-yu=oQ=vcR$m@$=
zUzFt#vDUW3Dr#M);^VOrd4arsPDM{(vs_blN3Oj&x<9bO(>unTDPb%B0GFg-+D(Zz
zvNip*^fS+SW3omzTPkGgHHBo&&P}u_NUM4~8~WoTF5xzO#ZKR7r4hFA6V@n+^SGsj
z+r0@pM$?=^#rFwD<`JSNZWEL;d}YMWI}vHk<v8>V=r@eg--ocgmJKdnzh?CFCrhNi
zNaQc~rOanXBhk;C`9nOXhN4YJHPJWq<YOU#01@L(C?>=k@=MrDs0mrN@}%6@pR|l0
zV^=dy@n06ukJ#*;k<L7#p|vd8r`a@kU#sDD@ekFDiP;>3<g^`2rVVBT<x6=tFPb}4
zoqbx(w47Cm)@j4n_6?i7F1u7=XJk;ohX5TPZY7)AO5?UC3Gv4o2Ue?16nqlGb6d1b
zRUWPFG^OD9bWKL`ejOd$v!_)q(BbML=9LBKpKb?9<ul<K`WBd03dh&Qjk+h3o%MD7
z7_}U=YT^@8i;7b*r5jv+yHh!%TE7hY+-Q#r{pxGha(`OJ^e8JvAynq1!CEcJPsjHT
z&ls1oiY-QB-wew%F9vq4ZYTSvScJqsmwfumdtmA5tB9JvXE9?tyTWPpO6TuZydf5@
z^&id`-z~Vk!$yBe@R+ED`O_1+=0{1-C;H?ZfjGg@<D+Pj{vVq;8@e$jhO?=7yol^k
z0a}u-jRo}Cjvq3DvcdkEHSx74YG!UP)0kPGbeyXyy_pS5YKzNDw>fq~TaLF{H-2YW
zw~u0)7xItk8{0qA;M=<K<>sZH{Uz_B<9}3!jx`)SexLq(0E6J6bBez>(?_4&l_+`{
zYs`{=^j^CnA#OxLOr_=JH9<;_c-iOlI@?s?Cgq~(7akX@_lIjNGcsf!YTGc3XS%ZF
zp=O<{n|6ejFL5yTxVHTU;|44>r-12M#bG@H_)Z6(V;3JA<QNm<It-rwsunZt?&xQu
zK`RM}Sg`IED-<cpczG3c|C#Dp5HjcVHA{rd8Nzk*s!o|vFt1<W!70rTmZD6>wHCn|
zVp=z8TMVO^7>l*uZ+Lf{&p31qV4=kn#b1cRMlm;pj=H)D7&}u9M*9Dy@V1*;D0Cwx
zMNOt{J?x+2#~HIsiWHs=rB`~;*AoBC-*|2OZiReqPO11!Y4eC<zb{;f1p}X>Z#w45
z!|2;doJ{c=j9M8R^^IrKG8al)47GZnWn`T>n`X&+(&wG1e!PjmC)w-SMpe9QN=8MG
z9-7~Is5$9|Kksk}>-~P<O``U@gjK1DrH+@E!kJzfW1eZjHot^(_uXQcSH&Otr^)&p
z8#}vEHO|6ZojhY2iYJ8n_&zDspN1q=s)DyEz7j+$^-D<1Nq*<x&9_1=XWE2l>J#=b
zpcVgo2JtAs-k2nYBgMraw)a+lUC@KCXFn3#+Y6$oliF~LUz2B{|Az^B?kEx#)cX^7
zFa}PcO+>*{eW909q%xnO(3VY0ZLA%|2xY+NU%`(WkiY@jVyxu>$h4#X3PnIHG$wzI
zkeL@jUIZ$`a&>noX!gJV{`lyq8jqYi!+5&iud_r=Y83k2zvG1{Zf#IPKn#QhvU=&j
z<J_97{s;7#3rQqYOm-6CO3owlA^_UK4?(u~e-C0xA>qw~5N`(Vqb-#OpCLq0A&20?
z{i<m%f*QP!9zyUSX@aQrA13L)BB+OdHk`EoS|yf{lmsGrjm$MEVIZfmH_U?QQP>+Q
z;F+605_Vu6p;R877Q-EDk5YaHa1HpmmX`a1sHb5y3ot;NA|d3o!4V{c%*w)YNLwBB
zX(6J2V@E6YLr8mRl@0qakEsB(V69ADP&)^TUHmDM@IOMxy`YNw4`~#w`VVQ8s)Im+
z<huVNjY_S_K7m|!z}v2?#LLOQXg2SH4WNz7z|=JH4_1_RUC@$T1&+|%J7VkT=_n3<
zZFMN<2hhj@SOFWX3zk6Y!&c_iTeObO@Np}xK?12gkl4;u3y~>*+xU5P)XMaq(h`;b
zgypfM%|fM+VafU+<PNf2aQ9b^)>+dA5yMuQ$T{IbHBLdY;AYm%|0a?CCtJ`$ZV&2{
za&@HY3%u1Y1zV9d^wUW5`@PDImAs9{cPXqB)gY}4^b)GO2<!pX3!Fe5$em~XfR;7c
zk<M=O3SM)8667t~2OT&-E2XUtptbw(+@<!abRcOBeSfc5Ofpa47X(o1l=S}!DE-gr
zVw9I6BnM3zg*>5b7Tn<k!I>lY0OW2C>(Roy^kcwEK30APg2PrO0}%1UFDBPG1IpD3
z8bUAHRjtv;^9OXCUZ5kHH}aaH5_`SRBQyqmsD5}&CmlMXJtQhMdIa=+RN)Z=_6^}^
ziISj<K)HO>OD~NBp5q*nvXTq5{x5oI)@|q=pjR^P@*xF+sh$Bq7Bn^iv<Ee<4g)On
z)I1GQhy$rOq7O;?dby@(`((Y0PY84(Vy`w5d4}^$UB<is@^Qvt<m0+<;?zj<J*3X#
zUO}6`l9M|K_3Te_vLS?~_u19c=47FeE2494d@Qb+u|eLEdI~aQ{iT*ifhC$VPe#PX
z$@ks;&k>N}_`e~W{*zW@7toK=b2m1|7c;(E>%UM=o*^*<l5WzL2TpL$CJb2n?sh5`
zsQ1C3u)Ww=9;){sLn$%+ASehOxPY5d90&tP20Whr>FeZ?OVs{QgvKV?cDywK&O|>W
z-Qi23@)X2te-5!Hj|5?Amm0moAqT>9%GJhYx@70N%}|7a=Ae&(yzS8@^jCK&(&mk(
zj>ob_{o$Z|6xT+5$-~VO_c4&q(*Gw-`wS6E1DLm%(FxSdX?G?#&B;1>trR@JRM?ek
zx`Q+f1TPPy3P7&(cQ!cEl?+LxLxLP6Yd~~%GOTvRZBYgNMAEyEV@(xj7XCT&q%7H#
zb3!k%?A6a?NQ|SAJbTnv`D$NV4mcG}Q*_Pr&HWt+8ch0v-9z^F*?;3u2g?7!p>o08
z0l76upQkE61C7c5MG>Y<SI8T;F<@s-R~vb1K6?gW;jA}~pPs7pqvSEwHdh(0%~a7v
zLnpHcAsG1A=w<v*J}OV%zxk-apcq1wL!VBsL+1c{-Ut_EoyLPDMJ1oD!vXmtUuPdY
zc%bpf`PHNY7w+Ypw86f2%0v_l<WFPgPiJso<MvRgT>-}q(L?R2(&JPoa9^o{VvwAF
z5aEhalyc#R6l^Sj<}_Y_A{m)25Z!tKnP(96AEbOJ;#<$+WoA~}E1Bg&2SN<@a10t`
zl_6ncTL9|{75fYbVLK>b1}c7P&~m6D%66MHD><)KVMPr&7~4FdFNMBgN!yQo2xYb2
zb45^sdaND+r&6lCb>dX%++@JkKj6nn5}j!!c)e=0V49uAzTQO7eSUtr_f82r6m*1X
zGmCC01`%jH51CICx70#9xRi_Z-^vgXc??q9@Yj#$9qKPYoPrdNGla;yI!}EnvR*(W
zO*K%nsUVh5$%<Ul1(IlgOI1SuUFD3IoBJH_THt1M0h@PDfbV~BRWBl95sDGbQDBo|
z2>eJbC$5Ol&{qL_a8=}X*i@cQ!#fjFCc4mGBr~A_^{#zS9(oQcuJyhK3y}TiazWGU
z6xEA}ZBaf7!B?_11m}S!W}eoyG0<~Ay*(DNs%aCF3h@sfjGLj7DXUdHQpTkB?`et9
zBcZ@WE$}idd+?^byu69~M8K0JDJdA>%R@cFi14%tka~Wn212G?koY4696Y2eXTYk&
z{3edQlgx$HYA7an2yy`GuK=DrlDG+#4KMV*c?B1J3ND)NqPmqH$Ot!v$Pj9Ift#Ao
za*I%NNd~I49tf(Bk5k>eFeV*wV5*J*n^6ezERcVmTAllv1gcck_7t;Mj-akWNLWaL
z%mOkQ8ib}woZzTdL9nuKKyRsRq(XENKcs()>mU4Tv>Jk64V@wRANbWE5L2Se)QAXP
z6}{NRtk((EcWt7Z>>Fz%j|TC}a8-*S#)v4woGXUB0|aoTL4;wTta619wChVeO!@<2
z<w)H3FA!_-=aCoC+CDR2q4uD$fcojvJ*O6G1sqJ$g*W^4FCN$*-Ru<pHU1a#p=v3d
z+YlUAZw|-Oz_IOV3SWOP3wB_`LP8?GYLajTeTBk@Yn3#_w}g^FX?T9wpoQb@54;%%
z{=F`wRAPz!EB3mtB=;r)6+EB41f6f(QsULO1LBuAEJ4*-&j>O?N+#ibzkq6M+nsY3
zsVscwuCT1^BjD<Q1EUI$X8{dlycAP4_lo67!lDrfGBS?Cx9C887>*GwJK=ayGS;=x
z)R{<i8C}ij!7zrDf>hDpg5xsbIF4B5z^kBBJqpGBm{X*<_}nHw+`mL|?46U^H&eo#
z3x?DeRs-H3OH?>FEK6P1_$HmU7ZX9bZ=aE&ArlM<I=mzbqo=L|KkF<A)9;@ylHM4b
z#yCFD&<n@tvp1pF14ZP8hC@G`vP32|UjDA3|Dv|4#C_dg%}K8ETa_5_1(biB$LT0^
zE1{_A^Gs8}?adDV#7h*fhTWV$^Ae)LUV{tZLIk{|;|F0N(U-eD@7TWqvwaBw!dq_y
z^dn@`Qoy?&CODza!zdb#0ro1MCKY(=oJxMR=BI7H)l|lVGA*6Nq8W6tmA{B?7lx!s
zkeLatzy&an$@Nf{s$Gs;35ekM?ry_P+Rpt5(^RYjTTOyIjH!Jip*NIC`_w^?UjiEL
zcg1cM4g+-QN2Km{2&NmY$xgU^oaD{mXTVm_yEcNh&t4UaBkEgQTYr$-uLN@N4DM66
z&MDH!qY=Dn;$%XHRJvQhThQGe2HkBic`MrSYn9G4=r5yVf=}V(V=GVDY!H`aDP0aV
zq31wB43#U1<o*p05t*?r5J~gF>|>xvLuCC&YOJUb1o}u%82&m!$y?jGWn7dwr5(>l
z)3I`M7hk)6z4b=m%OSUh$+ycsOR{5b)I*^AOQUdA19I4lLU^wtKo&3RdIR94g0YE-
z)*GNlbgQ3iSl(Wg9d+B+$-;mog4(9)K;X@*F!Y*Ofi#ROHJT6L!Yay+q}$0W4$gCr
z7hjCHQFh|tSo)DrY5u@v6lkptp{>|et}Iz%II?al`bC#~<8P!_#QMabh}0lwRU>wS
zONMt0WLpMECEq8ht{`P1*!EIyNR1Ele{uLY8Pq`?a0<FT4AH)};Q4)Qym$anY`y+H
z=ZrA!V~uAKW255dPIxzkAnIEd{xk|mVZ@?f0L>QJ5%QL>L6zd&6{TG@K$<=U&&dvG
zH9<`srp`sd3+1veffWK5ttaz#A3cxr=W-22%uGXBLdozgzeR`kGT8tjxByXTG7VIC
z%8Rw0_xsB(=}}WtBOaLd*?mC7JPpqx^S|h-_xY*1VQd}ivImWJE*ON7N`D0~A1w<9
zOsWxWWtz~J|1uG<G`@!xj#P8je>L301QGlg#Csb%(*`>0X)ylegA>J^$NmOl4(#F7
zQ&_{xlV(_kV?1D~qYWpSfyFW)zEnXil?R@emh1^wHVFg`gK0)YQquVOfZ}2$HB`lh
zwkys^4y!r@x(b@<99V)IUb}W^ZKh9~mjbll(IC&(2MCAUQz-9uocK5~Q4{tM%eFc@
zJ3FhNuErwT9bsYNa73Q-dc+)tB&2e9-Np#b7w~Mt-tB!^FX%)ON3nIXHOLZ|g;57^
z2XM-Bamv%Q{#heerdC(oQ__z`?^#aeJ&&QeB*Q8HrmXsY8PkjS?zbg|;M&0^*p#XX
zq^#`iB9H9|SJ)3(%LOT^kRdC37}Y@5FZY(@(0|M_nMl2*vQKXt*Fl5Erq-w6waSn#
zz3_5D(KWgW*$yF9%6#&!ZUeH+9jcwVw!c~b?ym=aCS%Yj=5c{Q?TXuZ%Q=9#<p8J}
z46)=egGB_;q0q=qh=Q7h3Q6EM_9N`n#!pCr0VpN7+4su<u~|VHp7^E*3i8#=mcA^2
zj)fEiWQ;>fGJ*X2g{S!L2hcu;qkQJRNFjCPe{ZjJ7()T|dB+#-(@fy>GM69V1>Y7@
z#xjpNl#b^}9($A$Gm0B@o2vi|`GiQ=k5{xe5|K?MDAO4q6RG!o1DT=0?#}OKcM?4w
zn0(FrSWy?z@C_i1eF~8{Jbq;nG~l`y-CNW~YePSPhPWwC)?d#3&<^=~wQX3@usJ;0
z088-jE*Kjxo!htA{1`Pf2<z5+;Kft3B)F*1rmyJB(T}fm%F#cA@+jf_w|Tv=l-hJR
z<t01RRspikqV`kXy+dtU9^cXU3kDOW4hqy9NU-#vq6rFxs@El`((U~XE<X74<M03e
f|NlS5)pwP6V{e-~85G1Iw{})TSG`pAa?t++)dcs<

literal 0
HcmV?d00001

diff --git a/docs_new/examples/images/nvidia-income.png b/docs_new/examples/images/nvidia-income.png
new file mode 100644
index 0000000000000000000000000000000000000000..9727bc272fd7457b07f162c81dff8e987278bb48
GIT binary patch
literal 185563
zcmeFZc{rBs+ctVhQdCl=NTiuah6b4`5|YAg3W*Rx=6Ogd$<TnJP?@Jp$&fKqk(42o
z%w#HK=4tI$&-1*$-}`>=_pP<9Z>{y$vb8-Ab-V8SI?wYsj{Vs8{kU$QJtMb{j-8Gm
zh;{O(WK;-(2H#T0u3m-z3H2)6#s5~?9+y{JjW4&=mwfT_W;<C;J5?)VJI9MQMudr_
zm4y+%t)Y#Pk)`crE4vZuVoAJ+pM22?8>5SMrdF2RYNi%Ogqpn(x6l!8WkWk|AweM#
z?gIzJ4hV~h9^^iz#C=vqEikT1fgrdEd6{EsPEUTeI_qdqmZV0TZ%vl%;`+*1PQx0k
z#^kueQTUUBZF0sQe&+AI%{p|poLM?s4G*5#e49Qz_I~06Hi5j6#Qy%=ZGOI9bbGi*
zmljKdCKh$R-phQ}JTW$UVA}RWUSFi`@S>ZsQ(azPzl@Bu%s*e0`|q+W3C4fE7CqRA
z9Xx-3CB6CofB1iF4d%Eq46D}n+lk?i|CS@;T8H|SFV6iQFgvZDrni{#IQo*YagxLA
z%*=v7^CKZ6zx<PzE?o-yu-CHS+bfIYdaL%L#n&&VgV;7<b9mXQy|955zW)Ado+*QN
zVkw##&)?dl6^#4wQWJe2B2=T2HPQlq{`|>PS5sT-la-al>pJ}X$oC@m^X>=P)>gcV
z?T88pSZCIh<fEbIOmo3=QFN<>i(0=}Mbm7I;zI$`LCdBjit|my?&=V}M<S*hwX@A=
z<mBWuH8o4h%BU^Ay(~@9%DOu7qn56DQR8lSxL<m@V2Vy2?kZM6L0+Ex*N;W+u7}0M
z-hBVQ#iekoo}Ql5Z12Mi>+VXERvFyGbA<|mo00L{W~Ijm*Ci$<#)eKzPEty<T&IV(
ziHYUEwR!D0QrY_2k%~w?pK~)U%_^_u_h{#d@`{R!HKF`YeG!fw-QCym$+N#t64Fji
zxj!bbz*mRA=h-x8U3#~1W@e_NL0PH-e|h)*eebFuuI|sN=MUTTtd^YX-!5|bBO$dk
zW9aj3RwMml*{b(Xj%1xB($D8uo>EehH8bO7;n)7{uRUG->x~cn=;UOD9GClz>$D6!
zJ9p9?IdbG}b@iJ7jzb%mnSH3JS1l;NPhU$<uUc1@oK`faW-luzmz=MotIL<@qwP|t
zr`z5CDZ02u)YsQH;qqEyBO_zTC*0WBCpoT=#GfCZooE_!Y<JhVw7FaBq*d{vLs-iv
zEZxuVd7F+p4wuI%Jr=b6ak}w^E(fuJmGxGnglpH#Shvv&24Z}CJl(8`mzWi1x3I9F
zVp>lZYK(1j8Y!F}of$JEr1kZ=L-%QUPYyL34%El?*T>R(EtD9w7r9&i{E}hQbb-a$
z+4=3ecZt7Umft#H%i`YJ2@4CetzCsbnp#>CN)t<&ZJx{SlYZvK9*d(xX~kW0lS7AH
zr}q8$5J8WZ5+zd0v&ui)i<h+yW~<>|B^ZAE_@PXBn3|fJl&pWy*2sBb=77vfqEQrw
zk#~|Er&dj96>;0wxB98a;vC`CTl28{eVA};sOPWjW;HKk94TIIb?*K986F?BSaEk*
zZlR;}dd+th6B84=q0j4D^X<#{(h9tTB<EJ+oS#V3D@1bMym|GSHG6H{rI&t2&CECv
zCGAU7hbYa1P09BjKRzEPhO>QR&8A%mb*k}aD%uKN4><hPT)XA)2E$7K(igg~8{cPt
z|E{x7GhJF$&c#LSNA2TVq5S8c<IJbOd?}NvojqE~Vb#&|?m>9w?c28p?R&3L({?S)
zPF!xy7ya#0sLhayQ~hzkPEc6*#L1JZ;?6|SSy@?WYHPn~XxOA0clwyRy85#sonjA8
zyy3CytJdio8PRZXaLmlj8FZE3_W3Nwr+CzBp*a`(?awZ1R#II}w~k$;WVEvk7hXr%
zwQt{Av558aQ@^MQtTh+DpB!wWK0D6t#p5k%{`skS^NYc3b5>T?--Au`j}KW1G*;cc
zyMdx{<;uS2(b0@-YzZrlEzVDC)X8WC$9fX!9*Zvat^12+R%pD?DaDzkBSTwDYa2KB
z)3;hFto!w3u-*ECZ2IQrcS=g6%gf7yk80ZYf80iRorymoBeQbFiWMEby)+jtTyUG|
zy!Gg)qkzlr{%f6Ox6F%jUcGu%QFOBX*$LXF_~(`H%ga|XY!xp_I2YZ~*-2HX$3Xtl
zWRr$ZwXv~r*NEq`>$8y(FDh!f07ZFoBGj=!L7O`L`LshroKnL19D#lO{MI-VD$RsW
zjwPp<{ou;$rG%oQVq>xf!-HK)Z*W`mArdaX_pDmSej{Ae{LPy;E7q@Hf4QyT2)U2H
zfB)v0KP4mc`{6F~=*8F%eO^OQao0TD&swOrpNPSM2)Ef^ypYLj@^iBJg2q70D~pDn
zit_S=f@VEerP$p&b|l!jy3X^ojE~psXZ_mN({m{!D@(QaOq8UqK}6*$hOPUE@(&-*
zY?XA|!*R$`t?x=}e)j3^Y}1Ba_GcF7CU+h>l=CI?g8O91lf%}>5wU981_sYX%o<NM
zQAy+Gnlke8@(_9_LzER2Rh@CZ9~jq$C)S}L^YGMX8YzZ+j#XsM$jN#5%A&<DARs`2
zOWY|v&H{&FldG$1h?TMNuDH0k!K^!X?mUrjeWmjY>H59`f><TTQg`xeY3Z>SdWF0^
zJW=jbVY#ngCyUe5(~pgg-d)cr=Ii@*>Q@V;&6jbz8i$Rotr7O2B9N0<vu<6}Z`*~L
zF<z@5@52m`4Y8-;w&91YUgXJD>@{DOk?~j<lM6Xy{GQF)+L{q7iKypo{q{|}tGm1D
z2>!0Crzd7y>GWxtwB|cV?ggW7=zS_*@4&I;o6h3r84TMyc$8_whNpuC*4;OKeZ^gV
zpNgHGn^VATHBDR+ue{(k=99F0|9)jL$KjOmD;5^9Cj;0Ax2USBYUNl)@Jcz4U!r_I
zT*%Yi6sN=y5_$LT(|SG^s}m<qR8&_h2n$pi)~8=QQOkDo=FKE=qz=>inA>x0qNepS
z_4V~?U!y#iv;q*yw{!FIZkOtx?aR6HRYgchsM^g6nQ4jc5A5N+Y5non>ye@d0}b)X
zb@%SwOVY8XD(UaP9I`akVr4$=zx-u!wl9h@*RYb1l|f#NU!xtWW^xS$EV1LqW2>UQ
zJhP>Dyty@(mlkV8^te1`p5=d-A1S^NXX`&X&>$DVwPnk`y?giGfAC<t)WYa>&GYB!
ziBCh5yn>guoc=g8sgkBA@ioKn)&kzmW`4?o@zW|%H+T1)M~-O8$dG&b;>8Qzk+=7@
z^YGlicW;}erR54~JF$p`Zm#9KazR_=PoBK#wlG$e<c3J1=%>7NK<T`FCVmq^Uf6kN
z^qH^k`zWak5r-_l9~NC-InBq-EiHVqf`*Wq|0x_YJTkH$iGYdmUPHXH`3k*1XPb>S
zXy3klmwtWAzW?yy?uP=p^((jKSheeQx${{q`_Jp+q;yvXJV;5|C$FgZKtMM?q2S$v
z-O08$*Kx$vUvTQAkyla*`T6sTR_@hjyB?{B5*_}Vh1HxR&g_mj>KOVp%cKU`xCaTk
zB63=Us>IuMGZ)wXu&}W4f%vE-<hEo*W8*kq-z|r)O2;ZcIed=hld6h}nqW(FvxbB#
zJ~^L|o-7R!^`mWY(q#=D-Tks#3^n(*%k2;r&iX=hC`&EI)L-}ZZgl$jHB0XF>C-E{
zPKOD}hKziQzD0McvZ5mKHzip;vCj<mw^Nc9`NC85Pq}6;JKX=uvQ_Ne+qWB(m6fTu
zk#p6APy5jmKYDxR@#-e7*{@%}^4ym?pUhp-)!Q2zws-m^RS8*gY%X5Bxbw4?{_J=!
zuV+;tC+}ivPTNLq&s|!_I#PCw6ifsrnT&N+#D*Snm>i_k{*1!bnT`jTB-ch9)sUg1
z)tnk>zmTBU{j~By>9+2UcfSW3WMy`Ui|fQr4!6EO>%O$K#JgN_joQep3TwDy*RGQW
z1_nD09#oILkbCu5L_@Z@Mu^@$Lci2@m2@EI(YPa3PU`AGTP6H{E<ZhXU8L{y<AX}Q
z?n|?MK9!xOb0*(uaf6eE8yLUl*~VA94#YlG%a4wZR*>>sD)=$y;o;FsM;IF!$zQ&l
zV61LZDz$0Trm=~M2hI~e-XRl<_DNU}qu=vv!)<X=DSCExcF!_4i`gX@v$PbtPVbPA
z&{dT~5c&CQoTDm<&&!#bn({I3WM(BqA`A1a+9hj`02r|nT|a)PC!`|j*!=E4r!;%x
z#*K|8T$>mf7b({3qyZg;GGD!_jF0l5BXX^~R{@KhH^3pf)zP6(NaGAT{Q5>i%pjfL
zyNh0K_4@TUZ_D>CHfM~*Z8B=z(S^{YB64$cp91aiF5+dh#3du6i%7k(pA8FMJz7|%
zS*{sG@g2;yXnBQ53|~N+Ozz&a`wWv<`{&Q<s&RXONk6-J4CnPVj>RfIBo^A|t`Z;f
zZ2OMetHp4B`0!!%+O;z3>MRE>zTJ?2u<PQR8*9RK-2j*ruei!@+_cI1*EdnZD<Xn(
zVR5mh$(V4$28c${f8Mgv3m72SX)%7_!fS_LB9Aqg>|0}EcI?=_J9$YW&p}68TKe`c
z*ZV#`G(adfQN|2$c$`p9I|l~VaC=&OKtby)V~~<T&|9PY*fi;G0WJ`+{I2=HctKpR
zz==_M$qBH01)=^zXALba?Mw6K^#F`pr+$8=R#k0KNuz{yPl`+*6cpU>`SWL^{J^of
zNL=hhU8F=)@kC!8HDTFa6o93pBGA21b9>f)R8Q4@hN9KqUMvOJe1n!@tJ<IyFR(r3
z;#3L;dd=U(mqt5EXS(j}Psb+j)5+a%;?{b~>zgdSzjE3ucSpIcU@Qd$Cb4Z&t7ic9
zjgVfz?!hl*Wzmh}eRWY)2x6j#R>yBh(aGsBDgnhncX)VsaAv%>#>n)A%}ASs{m-5r
zwLR@lRcE_1X@I>I3o9QyKBShcuEfcDg?-fZ<@Rmcs3csc%7L)e%{chFOk36x01DxC
zl~iBqz4MeCiZgYZole|BmU*Zxrv)5Mk6c!U1BQ2=%&};>IXltU{Vm%(*JDv^mvW@e
zW!nJ1J*x*d3wvx877`)|oaHY7VQucyx?3b%n6$LC4%_u_K$*R<Q!%vU>(|X32d_xS
zpNa4*R5uE2Zus7k>9(7mEysDJt*|2WwQb*ev&MvL3{ndlknl7MoDS4CG%#{;1p&{x
zEiX>bbX;c`of}HqE-t>1=y-A$4^NW9jAsyFLP$e>tl~uj18<~Vx6xAC^o)$XudV>K
zegtqJN`j=8d<zQ1uYAp<kzPvSv$7gdFmJAXd}yp@e=+g4-Tr05<}1p)j}|q;4XXk>
z5xYA>2R=T#F4mT5bjSE$*Ml~fA%3^<ce~AgL|6;<Am{j0@7}d5sr$J%`%apRcaocD
zY>(ajIeobb_?SW%^?!VJ>dcvwdU}!J+0!p0U)3rPO%F$*9S}RCq{JT_6&$SgVoqk|
z*x1-9dHGK<%_&-&wx9HuwzGSw!1pC!^GxyP7qY+DbsKXzT*tMQ<2KWpJ{C2<xm_->
z1A9G#vbuqtJ;|71*Up{W_UuV0m(=*-BvD()U*nkRWi7bWRdMI;f{`EH;>5%R_1)$9
zgXW+2?Z8XL#MrY;>P|*Ux;I*lVGj@34>k&2i1PLAPxGQ8Gm6dl#6%g2Jw3P7N6NRK
zkDi^lwc_n^*->%vEkHBGxCyO#=q&~=gSU6Lz5VcE6+r@RQM1N9V^7dIr05m0OM4Io
zDD;P&MxXv@$H$<|j*gCQ92)wbmqmBEE!UcQq<GomURIUl;%`;A`CmCRGu=UqOiVYh
zH7`vYPWF<h?xVR6nqKuPR204p&WgRO16c&u4+AT%mR=gFqCsL8>75uK7s+~1{Vfp9
zXi_)YgVNGcDl`P~%2DbA6}6dY_xZCuG&D5+G>mscLvMCh1(}cY8U9ol7g^+^2N+_`
zJ*}vC;>wkjxVGIR=JKHex^zd69&N-<Y}>X?HuS(&Ti|PZ9l_T^!osSBesn!`_g~xe
zD%@@co-6lUn&46j9~=5nnMNz+HnZX3{tG9OImbSqx4QT>)9A3{@RqpKVb90!xQ1{f
zd>@-D9I?1Y0^07~p2|RP^fbD%D-T(<$pE(IJN)u9ZHU{(%S(r=s8fxhnV(E-kp)f<
z5hMWG)u?QLyg%o8DN4*fsXP<eOw}{k<M$^yY{$Uwk@n)_%|!?sho4`raUONx#%1VP
z_DQ2DP<vYS=@TO$q-^3O8r>7AvNQFH`=7bbPhD<#mHML{n5pUKlcr>ihirFok7YPn
zO=3pNGgVR_fhPB9zg*q!KFwJ)(?xGL(#DAkt{@(ThZ~@T3m!Od47)Ri%Tc|JQX)@)
zxYH<s;;&igA_`W6=&5r^?l*712QR(7X*V~|44{ihnga?5D<P;~LI8165rFbH(R6IY
z@10UqEJJVF=wE?GOT?<J;B>RSgM)e%P1@!50_RN+b}Or!hwMZbBJDSpZ^|m5Q}U}N
zNAZc$8Uo=;sX}0sXPY;3^B~vUHQp6$omZ{WwnULZQfV4P?jos=_wP@?e{|qPYLo^G
zdOn`5QXXrGFX<PlT_y*8&<|b2vGu)uyEN0N=3}zPex=zJG{+1G@_eFszVq^>$XYJR
zg46EwYt|f7QQ4T8m36>#$&Gn~oxMHf+Sf$Y_&1+Fvq(xxhD+<7ITQF}(xSWh*BuqD
z?n$}t=ZWwUJ55bCGzC;dMn=Zy)YMJAG!5Z!e7ZZxvt?+HaeTv@Z{cJ#E%ICVgCbe7
zS;}2JNq&<fo<~3W0LN4NGV7-<aAqp1mPjwMp?oz0B|8_BFBBI&5qHi!ZBIg^SUFi)
z=9Z~LZB8dbTJr79GkWRBv}s^sqG52<;>wjBLP9FGdVjZGS=*D*b^Ulk_5mULE#=M`
zRvvwfEz%%-eSCc4_$yxz`1$^9$x%+CB*%^O@7pI6-;h|>*4DO>l~v=11Dn)t^gH#r
z)`l4`Uk0OSDO|mp9@pmGsz216mbRIekB^UWuLo-fQbX<H%5g`lgG=HjAGRpH_G>w&
zE;C=C6e*tf#Z~v-5_2;x<FBl5WDDalP`|TQ;?sbgfu*HDXlSU=<&WGUTeoi2GdM3R
z!?S(6Y+CcfIbE}_nXxf(aY++^II-_?GBY;<Q<%0D<Qe}=(|wJiFfcH{TU%8n7wI~j
z#}icypqgL~;z3tjTwE>k`t|FLNfgDD)Ktnh^Rz<sL5Y;NPR!Z#SB=`v-@9MTS9Wt<
z!^`H13R(A&Eu5Ua<+nE_{RG|=ZLdRP?f3ZcV+9b2d@nf;GBSR;z-dA<S(LLR65Rdu
zK2yOeou`?4%c7#9KDTbs=GzT;GxMpFq={KX%Key_7&1Pkmny0E7mOayFrr<@9uuhO
zx{mE_)R<FmHS_y$QF<yWs;f3O-g$XPP`^K6QQk)!WkqKA+(_Ru6716RmhpY01P3r^
zdS<5g?b~YxKd0Vd+<vm7F;TT2Jwsyi{tu{x2MjCMHYS{-W?^9=`2#et*NtjIO{+m)
zP$5a4ynK02L1E#uh5?W;KFCblK(2rsl&cLFt#NDp6fI}i`Nrq+$EQ@Mm6Xu*T?2^M
z2LbU&(p^mZ<z*^cTic$Vo&?eN2=<T9W$D54kVMgC;Dy`GNDrmxbx2aqZutA=RoAI-
z%&ywo%Yt%a+_Y)pb?R;PdrKi2g)WonuG5xwzrHa8W2I+hoj_-R7hDHIPqLwTY#My&
z<I`il0Ra`*gU*o=25=q-5%RrAoYngJP|@rUjxFL&XS0e7(1vYbWV}`xz)=RO6`!N5
z#i}^qg2$|^u<9@<negS6#Gyln7QkPSP`I#A0<^;?d~~<zR!6fO0QT|mH+UT*JG&n~
zd>59$Hz??qUYZ=h$|&G*a`V)(@7Ub-VT)^^H|aV%I}wG{=*g{@m)x=Vv?v>gU4ADR
zSN;N#k?t~SW;akz12~jxKg6P<s(P;2!yVDPakp~hzP}Go#F?0lEW_Mb!?{gnEwA|3
zb4jk{UZzxdH@l(%N9-#croKylLw+UD84~}qAG+B|BK)r}3frF}`InSIx%j7b`@i$U
zIUf`aYSdKxL4MsQTDSu}utp!s{M@veO{{NVpt^-S2<S+A+kVfFAOlAiChE%C+O|44
zIP|yVjAU6NG8ie=e?EbP57L0l$&=476nkK$y|c4LkyUO2#9jp5^5vBU2S}kvd*Njj
z5<91-r%N^E#>d3ye@@Z5++HMsK)h&axgNaP{{8z4u!&!gcJB3ofnS5pGS{ZJS~aJ$
z?yx{{PGh6m4IiJ2z11Og`Fj<>L+W|V8X<`>fPV8rP=TS+yfWL8W0mZcRjNNjeO>{}
zM!Eg<kNop#IVm^y@b95!2E@YtV8_lkG&;FgsnGS8-`QW}yN=^v$>+~gn_rFxHK(3G
zgcP)bz$se(XujMhwOmTe<wkIOsz1B^E^B#NY`;#f*N2Fs-AJh6zn?tW@*U_0B!je?
z8q-OCwrePSD!qNZy%$lGa-Ao3iP`mg<=YRb=dSatw4U95<x4u1u*nA*X=&o-dd|6&
z$J7KF2!E7-2r>INU<j6krmy2%L~<^rtpj$WbDA05hGG`$?TJH2vftzx#Lel0!n$f_
zx01NhUc`Cg=33{_+Qkz<*QNS`CHPZaJ;+cvh19n6aVVq$%uJfMCqPD7TJe$}5_vaR
z0+QK$|9*UV`3rDQ)R$t<Vlk`f;Z}TX<JgzvvU}c;GAsrZNecwJf7`)@eP7%*Q+ya7
zkC*KENjyY6#bzJ2A7mmoxOnlG6`Rnd)hq%!zPQbjsw!Gyv^z+u3nX~B*!r#FzBo^B
zps^r-39_xd-28%*Wxt+3u)We(aJrN1`u{=du0DVMJcxQAAuhmapKp1fbxDr=%$YNS
zhYsC({FnoD=?#1mo_YJlIeSno;7CQhLPD6YT)EOQG;{|5g(bXH{mK2$M1j*74=5~J
zBpO_G<*HRYyLPP-i+J&6j%;A?k)EH45F_*M!-uSB`>umoNfrWFy?G|S3_-6Jjsphr
z$yg_F%MtIT#rd+@+I6^ttB#IUcekCm3iaW5v-@ZV6|&SVFx=8<&C$TFnniBTw=MWz
zED}!}2FN7l<wZC-JC`;$vw|+$hQ|I5sC)gH(N3p@nGD_*kQo#?tUrBf(9FI{BW%`S
zo5xrM>tW~d9V9gjPCYk2-=N%=(dl<TBcL5g^n>&~fwzOq6z;%`%vJ?8rt{aYbpW|)
zX?mPESXDT!qP=L#o9t=VZ-J<`wqSW_e)RY68?muF*$-av^;{gfkmLp=V<*OmQ%y_Y
zBuL?VV3_#o@A>xhz+$qfo=sV8PmfWP3$2l+$6ppv`!F&x5ZuT!RG<X)RCa{Sxny<z
zg0V^teB_w8l=%4F`HmwmB$p?mmT!Ps2m4&(R=7gB?GB&Di;gCZ3qH}&+f!0fHf-47
zMM_<7ZKIz*{}?UDK>3V=^X%C(K)g!TNd^LIx-#MnEn5Aa-ri1}r<=Gm3O$NPkmvGz
zJK^={5!*|P78cZ-tJc<~fSh+9JWx$20swpl&HzV^)H+a@1WoIAC~=42*i=j7KMCIv
zAiIc_UMbqyNndPGryih_(Nj@T5gE9{x}OCE5am{EY^>NSZg3)@fQuB5|Bg8pfYdbj
z1kI>sua>6fsQ`A-a?r)GJ~wZgeR~N~01UUrEEEB-%}J84b93K<4W!Hh>n0bV#s`OQ
zWo6xsgUwr4UM>U81$exget*H!da$33N6wu)X8>lfG6CWWiJu7IwRfQZWo3FmjdgW)
zKC51_fq{X6<h^g)Sjot!_1Yn|8cIbd%IR&8>cj?c=|}7bgYzu`V)1$Ogo|hln*oPI
z4k97l=VD?^)JTiuEyi-G)ZFoPa&k&|MYUn~?%m`iaaKFS0&8${3C<pCS6z>bAa?B7
zAwy~dpyr=94^c=C?MIlT<M5NiM~~7I2Si0f`XXJ=xPKn7%S3BMImpkiw3E{aK7_x#
z08I_(-4Qf*9zA-*ls^RxP)4RFt4?y0qoZT(#7Jl7Hcm0S4UG42GuF1Y4KHh(nkWcA
z_K`1NzNn0-{ta^*R&|eL{v$6~0XqITs&8nx9jLM%NXQ@!^w`ttmX;P)=lzxbo3{%H
z2;j3e*`B>B)!p4KI;j*PMrqLNc2z5Mu|EO{(r11*PuIvunupY`Gf(#5&cba;HhAcB
zs=>vJ=lYjGDk!UJ01sB%Iyi*xQVLIKkHLR0NA5<Q=iU5AvhMxe|7kmOLBoI@NSi{N
z2u~1v_C*gKJdlx-Q?#^v0RmIu?Ag0>*B$Y~!6#sk&6m9Nuv<9R@4&Q)TnogRxa0d4
zT`i@AS5WXQXnPr%9Y>FzKN+}XFHsKq|HXx30UjQRg1p?%8wPr~mS=8}`T|}ka*dQz
zlei6?sl5=cssjL{-Y8}Y4i2xtN}b)bQgA)Nva$}`a1RJ59v+*~j$@QpT3N>785UpC
zpH;ge4JjxqD5!LI_V(^X9}9jQdYj+FhYyLn=<1TjQ2>%60SFWIKayGHT65+~!sPv<
zXVpV?Vgn!p0AbGqyi>Zegbm79+qaP~a;`Je@gGTrPO?ucQwDH=Y*u0)hlSk^3JM}M
zdwld5<S$;U3P0wg7m%@&T|n32=%bho{@GK<kv8Ug3nNd)?2p>@bIZ%i-+%m=57Of~
zXCQ@hlN_`kAlZp@A8;Jj31MYsj;#ls#|Nyq3EVs2+s^MzZtiox&rPnRrBzf=@B_qs
z3o7bqB_*xd-1{mx$$jJV=bm(}@Iq=r%8^%B4~>b9l?V1zYXin_Is(;$G5}WPTwS2G
zbRKe^a+6f|{I8)QJ_rP%zkl1AlIL%;R=!=LukV%bx#wb=6V9Dznt()~`1|Qo?sxCs
zV=Y0@CA8P=RgDW4^FRR)4*H9n@e2yl@O;F+_NGnLGKhx63G~4U?-b9RQ5SYXEhBed
z27EiB-}FP+L5uL}c>pp5KTs`iDn5(N&}V+i5`gXfCr?rmL80k{j3DnnJIMNekct~Q
zf5W|TT7r|4lQIXAPjqsAc9t1Dw~2}3>C^sDKNHMZ_N-P94+}FjmSAD+*M4;+Od(3t
zJT&eF;v%8I+v^ncQnhd35k!02T}KLSXo#%L%pD>kYQm2*JE9;o&CJfeD{!9ZL1_%1
zfN&sLV|}0~2f}=F@bu{DS?9_u18A{{^}A0$;^vvnTBfT9JCc-Wg5I2W1WX+hBXIh=
zzTMlmpFqs5^dgDP*oRDn$dY8LX2x-dcNFC0`{1$VO`#1X4PHeE{w*c^!Wq+uUEWG1
zO>+C^$9hEdTn2WANM3roA!G^o-u#8I>(rGs>(__ow7Dokj^g_+)lGDuxKZZ(nEBN5
zb8>Qi%%)J1L#q{)m020@!Says5^Rd;i`GI{M=EYePqBd{^y%#uD8aV%UMg7i#5Z_;
z!CSPGcxv1CAzU0pGik^G$a(wrZArx)*3O}!<h>KvCUZ%SG;@c(51a&vOaKILLSjQ-
zzY%Hba$^GHx0j~Z`=UIz9y@l-$Z``w=4J~go38RTAg*7C>z)KBfsaf@0DGHRl&?S&
z38)pks}{Ww2^WEJ(oO2N%E-zlijD&jQ1y(&pp#8LSidK5Oxp-JuQ??k6p1h}h)XJv
z?0*pWys60ZKHu`4#=2DIbxI&9!*Ws<BmW>Bz>kE~paM_2xrrlGuMRfueIjO0hkk1(
zV)8mdKSZ-VYYVF(l$r24*j~&UHg=15FERb}Fn4POfgWvA7de<UHuq5WSNgA?KU17N
z$T~ed97tl$((APFmhjq74<IJsNFaE;afl7!5+Kb6U@mU~5Ix1(?~G)S&WgJT!0@bj
zNR%v;RVGpHtGJhsPPm!@G`_E`ZR&gOv5DyTl3_TQ3t=l<47o3E_DaEeL)rOeD9iR&
z7Ph>+{OS0_(ZSwcAG|yDThEgSCy~0hpd>{&=pv*%w-Q$&J(3WI<hsV^u)DuK|B^>h
z{B}PI3){APH|_LDyPS<p=C`q&eISVA6Nlun8?vGN9UUD{XRj>W%CnY63fl<0GP5*4
zLeffKzG#+LR+3W8HDD}Ksh@85ELQoZ7LHV5uLB&SGQW~Q?Q-f4IFxSLns0L<EoLZ6
z!j%Q9$DgTlEIlS>ilymuW&$Odg#?vIN>qg9_uOlcY%VS>%;q}$Vn-neL<QL{B$Qe1
z64{%}mX(v!?Di@>{cIIdAD>WR$E--vFks77)G(n7ISFI?hmW3p0wNCmngLJ@i54Mi
zW!UzeA?-L(ZsThf78cUctN;Koov#{MjMy8LxwMNdr%YckSkuJlciqh9y~DFkLTL#m
zmq=p&)dYg?FV7E^BNYf5-rdmOYg5wU9%4D~04&RP_$oJKq8Q_Yo0mIv1-DmgqI!!7
z@M@~72L=Xe7JEnln2s$@w;PW4yo0QC1I$NeW+sS!t`10Gb0@xCGB-~SB)Qez-8*+O
zLqcz{qTH=%Z|6crIR-Q&vl0xUK=Wh(^?E+_6tABB*H!w-ETv)QddJLM{(nZsXTRD}
zXQRV=XUv}DWew6^Batc;zQ{l`z#|~A9!Rk4{!T?fWRX9dCi(-?OBP!`v79hsZ6vh*
zo5ZZR<bc*RmL$(<%pjxAH!A8v?1jg`l)MIHL!Yps&AE@$Ufg}ou9h#F2Ak}Iv)&CB
z{sieZYISU<dg~eSLK->v#}fp5CyqZczMSAOo?SDJhTs4MX1GDT;3Fb;{L5>{{V2h4
z)l3EBS}Knbn<BzjtLNwT-(CJ@eC#eaKmR&#ki5z=qhvl6>w~XoaHjT9yA0BVKI9WP
zJ0_D%OJ`-@)z(s~KnJN-6)#Rbj?-~*$v1w*NWd~C&B;-`b)EF4biP_rb1^Zw1#rx>
zV@FJ*1IzSal69EL5rL4{m?2b-?V_UD$J3fwt8(jT0G97UiS+dJ^yfUvf;{Pa=Z<RH
z7p|iYrzgb5A0Wd;2ck(&&?^=oX(4-^dn4~ho?*EMzsww2;Wgdn$^h1DbYcRe{ZHYM
z3-WcR%*@PgPs2Fj`xiHYrF5H6%2sbfRU;w)x#*MIcz9Nre@-b4-mCiYrKze;zTFz>
zrF<7Ow_=1o2r|I~@0Qa#@Fo{=PB$=;O*$#cmHMGfmNhoMJpQYt$o(i%(<ihvq&I34
zaUinv$MEoaf=y80D|`{%#6dS$`k;rL)YfL-b^6iMInrvhh9_7Km;5jEGd0Wv`_$UL
ze%0b+U8xDyl5yk4=L133A8XJC;IH<egdG4so_$jjUcPWs02&FBeMcezDX1bGfvz)p
zFigr*5&?CK6tuvAU@47;CqPgneC})cUohJD=TIUhPFKZdl0s-IgX7dst$n)rOg!G8
zC;Yjjwn7L5^Rh=HeNw*Cz36J^3R2_5zY()T&`QO4fHPAiXbA=`No0mquxgnpYAdD#
z$2mbECwi7w!L~(*EtGa}I0!puYUC!IOJqN~>1ZKvC#N~<fQ@Oo0EGfc!y8CdltF1Y
z!6KcP{fFQkJb`LLa34I#0+Jbuv+N5ElAWoTS)5|UMkaj&jkXhd*T)Ypl$VgK9)ExB
zsBAk7ok<qVv=I2>a9eM0Z;{t7>?7Lwuqr6&;#I27kPlcG@8L!U8WSHV<YZ@upgzgN
zg+mDkw2ae*-R1#IpowlsN=h7sg@ro>1kQx)9}y@a%^i(t4Flz*Gmf9%2z(^31b~x_
zOdyw(02(#kLbBNsWGioIPz9NhP_L`2YlJZ5;Ta!(Jz`dQF0DD4Q>Lt}3}z~s-XmgS
zNv(!TN=ioHr;<`3eAK6C2`lN+7`5ZzhD-s~h0Hy{;`-^O?VmhCg2g{PH&`&$g#FO&
zJmR`lz85kx4t&Xou6_;E(2qRcQw;0Zs}G=+`79oUa{{hEUXp=ROf5vTjVAYE*-f9y
zFHR2)yIajr+tTVu^$OjAj0~A1<`Qsgt*yFdpHmVYI?#rh(RSS@n~&&|0V2(N5~$y{
zzncf_`21wT_2-I(OA?iRq)hkV5Y(pQb>3qm<*3wANuHIj3DP3=8{KutQqS;L-PaBj
zXEgtm0g_B#(1Xn&p?wGJbw{kbWD>jOu&J|usLcfLwX3LMD)A$`^O=_RrJI00Zx9Q$
z1stdd9!s+(3qVIc2T!vdvfKjF;g-e)cV-~1xA60p>PsX2jsd%E=jBz%N;Yv0gFc17
zdL(Mj0(80tT{#)hfHWt;T^~U6wgv28!wZl(9XNce*RAu($vFfD1Kh$#Yy^nIPMAs)
z(CPTJfEAJW*;dGbd+`BV3FGF+3`04<;HRKDl5|Sfrhv%O2O<LLO;Smu#svPo8;%J!
zVdJ%W&b`c}nQ-@Rxl{{Mfc-t($^kD2bm?PYY?@Z_L-YBPWx_!6lW2!Y%!3d=4vstn
z*{@)6swHV1^p!7pw!6u!Mv^ozP+tC?djP!H0i){R;}-nyU|!k4!9h}3Z&2yLw4)y-
z84Tp@Pp=)V+$L%dky@;AYc?)#2tt>X763KAx3+@(Swt=n61DiowEN5x(q>7rGQiHI
zKzeAbOLE#h?jTXlC8i<6T*KX#)YoqWWKccY4ps0D`QhiLM;HL7Vay<n1!$f7o71>F
zW_wxv*$zB!oYU3Ox%z)1Sj?*;Iq+*`uEgn-gO~N9Alu!?kLgL8cjCj*!?r*CjgNSl
zv=;xbIQPpwCfiA04D)~tuk|%%^x1!X&8c>-g{=M8*K=NfvW@!x>WAm;rSQf=U;(E$
z+WYW=<j{}c7E~d%`R&`cQ(S;pDRY3UR0NvWKU!4?zd%I4r~J7(H;OQ@KvNZzS*eCs
zNYHB8My1{~bual5Hk37g2!xQIKmp`2C!PYMvq`!gCcvS}EnajTY)mxyD)Zc?<)!H+
zxInKW(U`O{&22jW@-icSzWeZ@#<K~uMC(w8i4vGjyb+w$SGJO37Rz7PzJhCzWHQk)
zQ73il_Pgaa&(@nhKHEe^nbDk+HXA-Q2qPNyFnGMYfBuwb3<?oR_kqd1MBMe8j9Oj*
z7d}d5{NaffS|30)Nto#fY0?#}^jJMi39}V*HFNFrcfiZ`EAPY60mZ(O01PKPm%3b8
z=IG`joCMN*YiX(1=|VRMc__L3^CvEgzli4_S(Ycm#61?+c*av6k2Nl_&S!wcQb>$P
za9;lQaxbBe%##bm8r!G{u&z=0$7R2B4ybU9G=-l{TMFHA;J~>;7V)bc(wH&m0JgjZ
zwvbeXV47SQy0FaKklmbiA#HFDj-9LHJ@R&gjcX1bJP0ML0$$NkP&gP?Fd&1<(UBC2
zVAZ}lGkQ@;w<48Fb8BoWl@oXq(qN4q4t3QF`3i2Pbf7>4nQ8US8#fa24Gaw#IXMFW
zw!eJ)_C{ausps+%@b-SYfM~fO1C$#iBN(6ep^AE;KQL@7a6YB1oOZU@AAo&5Fq?O&
zz83--ZR#yptUA$IC%p$XLTRA}(;*!F;mHwdR5eUnG^U={utj&g5fH@-9o`$*s}ZoI
z=PCo(k;g<MB$l5beI?(r97psp#;}oJ`=u8t)L@8$v;lAW3Bs!XTXyo?ci7m3Qkk0}
z!r;ne`?GyJHTY99onisu!nzIif+@d_Aj6-e=8>|=C3yY%=*_j8y~){%YKhqbE1(y$
z$Bd0NN-V90o9ZG~)9ea_YTWsMFvwHl07@WU=|KMYV4cC53W8nbnYV@)n_OQIF4xdC
z7ZenHL>WBZo}D=rozksxAIvPNH<7GvcD5|W0N@=#IQmJ%?_2y*=!(oAp*=?sXc7^p
zBJDMA-<~1`OZ0eT6V5Jj*^5Nt=&O8QoFAPJi--t-ZR*u~cO2j62CI=2DdDVoPREGV
z1fc!%U1=z`yd&U&U_}iLjjM<H%cGlb2SsxeAosZe;u0KWv3m1WT%KyZq)As_kSPgu
z^S@H0pp+8x;jKy;$4182Ba)-*H*MSq?h~MtHx=)abj5mrhv#qNImF7;G||)qNJsS7
zW+{(?>c>ZH<G>92!EW^92@Z`SY)E2#DB~<xM=I`JyLQ1@xNFt*V<99PinAZw`-){R
zXqu+ppFj60g^MtvjiTblH=ka5LY_+c`CRr6*hQ2w81-npu;1_C3{sl~JMSAGA1_w*
z=FM?oy^*4MOP>~@i*Hue3Y%QKxD5xC1rD5$;a|V_A;|OakPNhQ;FBj$%FD`5fGobX
zu__*tK={y?mKH@ML{a;7Tg5ePKOnU=No4&+ODB#)%(4NyK$&{i*9zd`US!>UGUJ!R
zsZ$#u=hkM83=Z;&MVy)353^^;A&(!d{Ms?jKj@YMIFG8^1|%fxX<~rHVfyjuwMQr`
z1gr&_Bo#IpE#9s21Y(~xV72|@M~{5<(pYiZwIk}tvU&5?9VJvCwa*$+`QEy9izK?+
zZzH}E#jlrg<HS;;IJa!+LC}RyMM`<<^%kQZ#!dyH6H)j()4Z#xA)V<|+=wxs@jG+`
zJMBAY>9Oc8?gt0&Ku<4ZRLvC9jD)C?R_s3Spror9y6+w`>1~WN4Ukg--QC+s)(}FB
z(pAGhoM*c*0ZHc^^LrZ`oBJS$(4&O_SHk40(&xWbBLD3aMw8xR{dU4Lj=pdQ?18*i
zfVYEBK$s-xpU}~HqHqD+40|-1;ZU&`va>MeG>XCw40o`qj10J&=(&UVU=&IGjJ7Ur
z#bJ04LgDF568ljXm6lkH*bVnGA(>A&FeIt&@}GaDmcU<WXbMgps;S0wGEy+F6)M)?
z6KEC*ze!WGslR^|Ge#cd&zu1g$5fV?cYs{r7DmQ<uytm<dc{J(FZ|yz$#Vfwjxg4P
z_Od~0Jz~@2%YOYiTac@oTCjqY>r@KK03wXxjb-}E0NqC`!@%dsQO7fqm}&|^(h&eD
zqtS;k31ynE0y??S+DTt#FWhwMKRkjRb^L^{d=a#dAj6n!E<mA@D1MDGJ1tSuz`*#+
z>N`oXyI|eu(m&%WkdOE9@PNP4=&cIi&`79-%XNh`GUSUj!+;>?W)J+aY{{e3YU0Y0
zT|B0zO>FRbf-@3Q2j4VOF%>t-x=hYhUb_LQ(R`frq_jZ`SDV9-8D$QVl-k_LX2NYK
zHAig@EF68n#nQFgq3ON*@Ijz4@;|e?^w~{t;s$P&NIC)n`xv@MMb3t<wQPdYy0$=)
z;RPU~=m}CBiHX8zps_&!H>Fg8WH9ySSY90Qq_l#Y3ok%xLArOqd)G8y@YH?wiY=%Y
z17RY)0a4F@mJ@!*ufqWjiM&z<+$<||2_83I3v?yn<D@sundaJ_?pNPhS{^zdj33L^
zNYQ-ey#D_E`#bpflD~M6T>pb+G~XeP;42v!aUY>*6hH^xqo!SdxtjI8xCjp0boR?X
z8`9n1uh}yK4-|NkXRees4__uOqQJH<Fsa7Lq1T6=Rc#jHH;b7q5+22aWB?W+0X)AS
zb`8<Ti=nk7F^8*H_e{@XXTr0`{|8nP@Yg78aC~~Y*==fkTxIC?=00qQB4>M{YtDCz
z7rqG4#BOxg=HpwoY$1p1fNHo=-q5+<NJ~qLdk;{QHmif~!2U{smUO9@5t19(1BgpV
zkanGwD_3&w*|QpFk9oKZyj#;4S|+$`VkNNy|A|N>U@x1|e|LYU6D%m!3p3V)*PT0%
zTX;mw(_!U1%}LH<Jm`Z&4HIR&rQ$V|19M5K>F>6Kx#H$EA#wYf4-8k~5RG9ZEvatJ
znl*znXkRWvqKPX4qQEG{!$f|Bj%K<(mCQ<r)m1Rke1a{EbUR~ONE(^pjzkVfhVVMn
z)(TX&RUONb5^(pBScmKiVC%woU+ia^+V}5Q5)eeV4;^APHvYD}S5HkU(pZO%^pzet
zvYF6_#q6@u2T0f8ffXK1i7YzZ&&m>BK}}8K8(h5;Eax#)B4f@E-eLh{C)^zz%*3|#
z=EgM4Qpr^OaknJgCyQUV-%pqGY9G&{cQ)Kx75Ee-AiNGb`d}{lcG<Xx@k{a6|7Ldn
z>0iU*Wb?$2Bl`b}Z@A$2>n$9#0#n1S=da{J(z)zyNxf=q2^Mawqm-JEZhoOlPs0;1
zuM_GeC)=W;wpdwr(&);4mB=Bs3nMT?VD9hAJ$vTt?J3F0=3y-x+1Px68%X*D_Jdr{
zWl4wO@3)cjx*jM?UW<z2LU&8Yz!0R8W@u;_9Y`Np2r7Y;zBMtJ6xO1qLaw4;QZ#8p
z9KSR7%Q@UbeD7Z-ZMD++Od*dm{+<`$21;1Sr58kk*?C~S&R9@W*Jj%5e&O~IMiT<>
z(5VN90AvFI`yib{Lo`Gl?{A0?qWhMDPysMYJ=<D&?KS|z@$rY0y}#?~>XdEupi$B-
zQOzEmw)~nYt6Dj81S6}nt>0EKZT0M~V_svx8vo~jLHzLhRYfwT2}||HPbEw{O0I#r
zE5k@>IwnllZWfZEF;M=S_tcr({D*n{`2N2#uVdez&y9(6tH42{&dE`yCU6%f=)^^*
zdeSjHYV{wE>8*%v^)BbiqnaE+)WBFFTz_>X<xVV0S}P8DQb27}stz1D5MBT_H1Vl4
zc$_=#u^Ibh{wI#wW2d^C_ck;%+=C0j-``(l$ntYan3#u+x+;bwz#-E`WWgvyE))hQ
z%e{-hRISaiY*pzcC%poL;N-xLP0k_}gL}M=9jy%HJm2|>X$D=LNYwPNmiRypQ8Trk
zBbC4QCTHrz$9OP(e_W09=M7zYW`3S*yOEltQA1n`7obDz3_LFni%UWvN&1k^F(5Ly
z#nsa~sg0tn&u{MTf*#^MU}|a#%TLqG?sxO5n->Lf5Qf-6g5V1n_9}?LjY9{PINigS
zg}re3KWaph2v8kO=nIS>`=5t|2eIPOqn>5uyZ-~hdF}c?1ZQ)u>$Ct0*n0Hw$p7Zv
zJJ95r@e4y^2k#dIj;i#-hZk=(q@tL%Y)J_$d-G-k7Z(HRMk3KZObEvAWhm$dAXAC)
z<^X`tT}5+)eCP{ER^aeqO&V#~GH!q+gfang8~y+`Oa+jU0s-9@o-&dMG~=%te^v1X
z;~eS6wd_4Vesp1^R&(8a&p%!O2Fcl;O(E#Vq>y$6T2a@ideWBF)&G;lc*Ew+iQ{uq
zzv#J_C;Po8>A|~E?v8?$fiv<b7@Urtp5Kk;lvOKNmcSZ)?b<a=G2|<FVhW6IN$l|9
zVCTZx-A9fdP41>=6(HVv79Br+T;$k+Zf<RPUUMD)=KWi+{v?eJO?tDiaR-RUkoutK
zkq^Fq|IRa%4;L4S+CAtuZrn&mLq+c)2arg|Z!V4W)?4;h0}A|U=Fo!lFxtMnG`%^e
zzb+9Fcm^C7Wt0R?f0>IOhBLCk6_58&n_-9FgCvv)(mDz)gmllLjUv|+$neY8uNT3b
zGbc%*=Ah-JzFREt815u^N5{U~){#9seuun1RS&H+*?R&1lr%On0lVLTTZy!t!B9nD
z^gx7c(kaV8A)Q@aE4Zbcz~9s23!|LO(jI5~0}i$AaXcT{7_ZEQm|qDC&FI+JIjV<1
zuw76oD7~QE5(6=(!T{&*0<gvY2`8Y;^-uu8<-)F+^KJAZOi7pPW7c>rcr`#LCg(Mv
zbcru7OpwG>axx#l#!A3&x4D6M*%usITX8&YVax|7+8fp;(n3bwKMD)@*Yy~70x(t2
zT|(EJtl<oxN=lb+T3b1BljYtt^lEuvIYUj?@|o+9<L;B}7Pw6`AZJsxMalUaP|qs8
zbl$Ga1GmO(>$6Oh!Ps>|5pg1?u;*aef<KHNL(G^WI4@BJLQ@vhT?S+%V)oh_@2AFO
z)AG0zmaTtq-)(;rZRUk-iarpw$Vp{#6rK)!3Jx^c++%}hFulUa%uIP!d&D0s9X&$l
zBv@vBE2|*%zvO;F4zGeY-x@;K@z*=4_e|IqBc5*Dx)lv*f=0FiTHLw#Gn~M@9WZeX
zQreUyXIDbLCy3LS2DGh(cwbgiLodCwz6d3pxLa&E3LJ6(W2#QBt|#IfN?}ze`9173
zI)4{Y+az%d=Xy|yq+yVpv%rH20G`N)3#8*-YuD!fcjTdNob?JXcu@>sn!#A}Dv%Us
zQPQAo(}EOj2&P}Zz7ptR6o%U}jQx=I%LflOAvo_qo$1EJlfZ=Noe5AY1~6feLsP&{
z;OPC4G|8zn^zCGV6@-nAuDKmPU2=2><2>GYU;{$%BiK!lxoqU@Iws*);K%`-1UpAm
zRI@!qExibu2vh|h;2$Ya!BI?F`GBg_dg+XhsmE|-8P^_%K6BCBoF3S^G3^2u;)G4Y
zMTE2@!XZe;2+T8&1TU^QXw~MA>DOr4J1lS%%E20?9o;Z9ot3*e1(B`hAT}HQY5^E@
zj{el?(>D-&m=5Ud>$?TnL2_u<xg#Xn!tIyftVY_adK@ARDZ;M+=5AqKmj#+<J%6@3
zx+zUBmy9*n2A<x&CgV^>qWTS<1YJg9hX7s*pPGf)gKs<gbCMk7eG3gY@s(fM(B!Lm
zn>o|g!&BzNevO;oMQ%7raRrHQJ=Etm+*eeW7iD{R#O)+GvyTa&jkg?&^1fwk7)aE~
z6T(@0^X1DU3<;)QEI+<_+l`MAzX>LMlX(||csLfsMvYBMMPIJXq|?yQSpRsx83yym
zt<P4E$>Yfsy%7AQ4fa#ym)2~%v0*u42jR2fxbHr)W3(4q05btI)!kag+d1c$d?elH
zo~EVMA&N4OZrLWOMZ%-kuc`HojVr-}W**_e6?jbUu6(-T`09HVek}aez+hD5V-+GJ
zBfoCB=fn|kvJ5M9J1{U#C(q_9XW)jY6o{~t2(gP$rI`eD4nN+$5ue;<RT%YI+zGBj
z{>OLAX%~X-KYm<&@4Ebo|NPGzz5aUTpO^RO-~V&}fB!lUt+0~@v$;sPBPILq3z9E8
zqOJa-g-?RK-GAQcZRrlGJ^#Gv>TNRXr#{>yOpFf6{_Cx`$^1CQApLJY+X%}4?|ife
ze-`Vn>#-em*vB5Q3Tp%!BCB8tTt@biO00sB^Wt&em;buuayb)FSY|?rz|QdF#q5dR
zLx+ivJZ(#)pRXGRT2e0*>jfOxE`<rNubi11y0&j$^B*6XD*~EJ1EP;W^IU@27?t2z
z5T)?@tR+ChJqH;uB6@P$jhVA-mmG<ImPW>yoE}2EuUF_&0l)*bYC8`9st!aEDG?H2
z#PCFp`xe%fm6Zni`d%laoz1_3wh+dIx?N5G(2xn`@^N3~8-JZla*fB%@v5h=ZQ6C3
z;NG{-A2x@?`N|-!I*_DZ<$1sE*BxQqyqTN|#$ED)I{`%n@pm0O!%qg!b|IrSfKUAS
z>lZACv6yxyN6$#C69T#Py4Yz)Opjh^D+t@TQvtr>)2B|Ix*7PcvXXSC$Id?pKIJws
z0a+1KQ~bbD6-W&5d?2HQc9HFkHR#}P-`LSJk&_sBLQDP{1>Fx}&-c84d?*OU^ozLS
zv*SMCh<C&NaiULp^}Wd$TJq47`{*0=8GW{9mhn0e-%c(r^#C{oKK5F#>({CAXqIU0
zCFlT1PZv1Zb)L(t5n*443fl*1;fbVsI9_mqU&^NE6rev1N!6lLg5q@@Y&srHQGqqz
zgH<E2AcONLp>Lq^w42uR&GzXQdqjZF8e5pP1<eHSCpqRKVp@L#Z6h7azSYuKuL=Tp
zkOVs+AtCW0QbG}j$P^eaDJpvC$UMm!*{%Yy4Sv}cAj(BOYQ;90Rr3Ej>|FfR)YO<-
z@;e#5O;VDiEe>Fo2XFT2vIy?e0L@2e)M4xA;cvr*jOc*3yTDXpYiPVe_%t%%RpLIN
zbr7u<p%B7b0lJx-q7XKIPi)+D9$H?5-_Cb1;06K+z!QL_VCUd?yq1$Z4=^QrhD#y+
zD^5<3o^cq!QDES9+k&A#?gP#_S38eqe}{{yvF8GY+E&xkccWn9$v3JJ!-ocJKH$+g
zWiaQj@Va-85d&QXZnO7cLV^+=@ywR{##<M5*>k}^*Z*_WZ_Mn61{MSs^z$nuymYr&
zn^X(Pe6U)s#wo|7@v~i@y1Kd`%E8WfANG(8=+YqL%5j?nM2%>-*>Pe4F%`_Gv7Y4R
zGd-6*2#1k2B~&N^aIg#TwWm5{C^aMPLh)8SM`JB#AD|OLN!hQCPc2~s8ho=%TUg|v
z>a%%4uw&=q;v(=8Jg10_Ks&1%MEw-OPfjv;!2$IN{_%*qujpqsi(KA7pe9}dsxe!{
z&7$xUXEQI!TOt^q8L}K-a1>gp`FFl6m$O38fhz1XKRrTyTRyp`>?{=ka-%a96^Z<^
zrA*g@rujg1dt$69_54Gilq~qf)KfH%9Y4MTjohH?^XJc10^-pGY*tt@Vul=?8X5Sm
zPuvY!)7>wx1^3Q^HhvbeKuua5HhIb0NtnF5H46>6Z|@-M@k<Q<N}V(PTIbI<z?Tf{
zxvJz)yIb^&)lYhA%XPI+=snnI42PA7#kXK^5fR;ZOi9D61o`wNW55!z;{D2z;@ZU^
zd=t_htXkSBwREUX$;`fBHq!q|)^Ew}*>z0E3%G>-6`xzSIWy$;Omy$Wkek{C*o33^
zC?rWKM@iCbWF_PkmdNfCPkjR9p8><(#;#}<71c`G&YJN?@)=Uj%EdeKBgspf4j<cA
zKEL~4$u;W63K}{*T!y{`Yt^J@E{|g&gD!Dyo}`VMG1`iDLj1A_yMSIng5~m~xTYM)
z2(F!lPGjDHjY;!NvLsmD+lGjQ%hi8w4$Y*t(SsifxChUFRkg8qWcx#aA6mOr{1M0s
zZtHF9G2eh@m`%z<-0O37p}cL2&y5>`$RxjiZ{&)HIFTi?1_K8ta1bMg>(H+V>3WVI
zS%8`QG-l7?H23E?6j)MHqShW=bHx19b04w&9cbRZ6gUgxeRPW4!Z3k$Nz8Ivx#z!?
zlPG`8k)6#pqi`R%W`w~rc+T_btU$k6J37*S)DFu*N3WJ}_8J^jDk+)=;00Jk4tnm@
z%@+Z_eS3fA4AvBJQNQ@Lz}Yr#hzoLuSc)UgeoxIqlII7R+7D}lwt%Fho_lJ|5k5*1
z6FJ37&Rvo-5oTrzu?isxy02e}iHMPf(oy+eU_f#+^mb^P!N`o~G7MyYPfn8912y$@
zq<9!+pupnMl)y?4Vc;oJ73$Q)c(3A32Cfgdb+R5KRabrdco(Jk{rmTsvu}ms)Dl&;
zh}rofuJ7|{2mm7|Xryf|Dk>rwY5r;a_GGp|-&%dJ<)<GBUgMGV)4y~{20x+oTkE+n
zb}Rqmsqs0V4I_~!XCLj__b>3V;dsBy{}#&mcPIgP^XLzf`xmrX+eWEi5BQ-=;r};;
z6BxHgA+%}de?dWQS_~yA?^XT{acvWl`+xJZ{R6wock_BKEnF4}XOiOb1K6$y`-*wI
z10o_j5I68tWFgfs8G&~Ffa*pf1MDvw8ynpJCtz^Vgc$1@Ktm2rBJYt6SVqPHcydiJ
z-HvW6eh&vA1<0vHwZx6ERe)+cn{Q`Yw7lestXCf}gapb4zM;JpwCPk-(4=9<z(AN;
zOT=Mo1BY9y`j;Z~P-M{osl2>=&OaakG0lLk&=~#{t&MfSY>Al5grfu3IG3(}>|2gi
z5E$>I3@~XSpxiLP5eN|y(8UCOWa8)4nrvX}5~TG+-PZ?UbwW+J13*@fF&KdhMNf8C
z%^m2+tt9=L#<SGEa<UT`HhYSem5E=Q36=vqKq3fJG=SFN2jlnPsXM0;Hha%KUkUaJ
z+H-x8yDKcyb5Vw{K*yo`V&YR@i{C>kOVleAL$%q>%R4Al^}HtgnfX4hEwX=|l`(lV
zJE~sT0dKh0_qe#YBv#Vyf$U`V)z^1*tU-mJ_I$hB2ry2Y;y%CUdq{&kNy70D!4xa*
zI(3>K-B&X3jZ|$uWZ&SYPoJuF`(d<@my#4i@xbg@nB<2baO8?A5*Gmf-VT4XTRH<$
z?l~0BI4uu^%JSYwP?2$3IS4B(2ut=A<j{Jwq#`xNUyJPW4gz$9kV>l7XP_H&A+yR=
za~VhzhDm)T&CQ2ipHRiP)^n^UU+qXDcq#>0Ewv@VT<t~4Ofssit!4J~Ts~c815hEd
z9I=8)POfU1>5f42%@-MNzLNf)t$%j~Bdta5^TFUJ_o9|^sjU!%CP^y4glzPuJ$P=4
z8c644Xo-99K&HVd%pme%MPNofYnX<o&f)U9JCV1(V!qA*&n1w2FX24i6H$d}>v%Lw
zC#jggAR7UyR(F=&3LzgWbRdx9(3w<Ecmf8epdj%=%Bcq75sbW<)KWbA<FT3{#}S*D
z6??FZ^F15A9EZQ3BW+(-T3#vfg^h39LD(;zdGL+?DJ+J%N3M2wWnTWoQvtS<iV4`I
z7a~ZQUqyTS1wb~sr#-cgHMC&)C$U*9p2Ik30(7Orzl^Gv)Pk|U1oC-A<V4Z^-O5{O
z&cX%|GCF!y_MBbu!KImQhU{VJoqJ%#JckvIMN539RWC^`@qN@x&|BcIrYRhV(b63R
zW^O{T+eu?<VPV7cQ&6VefJ*i{q8bs?G>r#ND7H#Ejq3LT%2osKH(5Z9t+f+}%4#No
z?)5Brb$A67_Npb^&3A|UCCpE@+5AMarlic*<+Rs-Wl)~;6_u57kI9FrMW|pdH&*)E
z3O@)^DhP1$-l~h2my#(k@D-^JVYt*U6QfzNz;1pp>li@Ff$38nVd7yfjx%Q_foCB!
ze0wPVH4BeBlRd`-+QQ>p4&-zBWVi9iYgUzy8S^V`V1%$g4@YlfCaR}jY;5d`Js=b)
zIT)`}=J%W(H?p4xTh{ms^Wvu&n~^P;-QC^MY^a5S_}cxDIBf|j-gFl+@;xsabzF1r
z`By3AxECs)GE%~;Oow0EJD~w4U&*uSWjgiqS^OS9azA4MF#T`}Rh~*y8R3Bk3TkSN
z=es}k9#twzMRiheb1TsH0O%bo!cz>42fw2Z9ZNODWc1(^sF<C|u!BX=b;+X&Qpl`J
z0EPwrDDm5w1{HLFc6ZfUwx_}Sv{d$LWyaXd{Wi_Mg7^vlxiT12Sf@FMs;b|;+grP`
z19udHW|9v=wkjrzf{{M?NbMNQ8cVSx5hy3TX36A`N(JV)@bVzoHJ+2y+Yvlg@4O)f
zJM-r_?J+fEeqsEeAZ8k6&uuJz9n+SKU2foQkCES$?WSG$s9^$+SIZs-wz`~sr$@pI
z$l+0<Ch~w?K<^17ufJ<(%*CF!_sDsB!ra^(onamxp@fWCQd4sWAod*42Z0*iIE}%f
z+h~_JgOt=Oa$DmCdIx|T60<;S-m0O-#4?B~grJa6{Wn%FuC$^va|OTxLN32A<B@~N
zHP3LGnD(6Yf{JX5hZqu)%L}%2tO6+$V?*pUu<bQkpv@XIM`Xi;`3|OXc$U^9&tQ%^
z5~%Aw4AYoecFwoo4O>7Ad|TyJRl<1e8`62&gta>z`ck0fE)bx=iDat;C<&A0N)Ue?
z5Ua>>ceB};iG`kbn=ph{jyjl;p3W^M7XI<$Sv(YgG~A+{Qk=H6pFM+zuROI(KKpEi
zi_5Z<k|(AQ8k^s>gb1Jwu0Xg^35tqRa&QhR6FyzW_hhb$m`=fT2>ct#y;zHAW<|7G
z_C*Ih@H{CsL;u^ix4|Os1^0-7KDQ~I7cyh#;Glwz&KC6js7t{}ft$8$d5>P21dhPy
zt6q*Ug5x)-kC7MmTzWDy;{<~gc|VAzZGwV#!Rf}$hJi8fMYotZ>^eCZGM9hp-F*TI
zdnAx8fuWC7N)gT!Zl3nkliiv4cw(HzP{-g}UMUSuB3U=)bCF$u?884vK`nE&@83Cr
z0z)SGXaz3-K2lx;c&LD@jabdV(5!HChv%MfF?-%*_0%q;E3(PP0V6sv2q@_ri>H<l
zAOKJ+vp3S?c>xA!H3&e5Tgb#K&c>bIjI1P!PT`2>k~ms)vgCun#Zwu{M+kuSt!+H=
zQ8d(Zc~OJ^f!%xg`%+G{G&EFqff%!wb^uSq+o8QEF|bYQ9f;vIYuA>)dv_ZSoacKK
z;sW{O;X7l8&4E@Qqmr0ew!sw9q$}wW$7nn@e~N>ZwF(&JW^u7pU6d3FC(PcsppIaL
z#aru|dXLs3*7EuHYDMhW>oBuC^_&mKQIrMg$9^8|edCRj=>6Y?@Uh~V2WFX)+buMj
z&9~>_EM_1rE-g$blX?$&yN`GP-3b2*VBM40Vn6|%0;hY(q>lleiQDo?<b&Xh9<;kc
zN&l~1`8BdDk1zmRX$Vf8OhJn|PeJ37GZp#mBX-X|U2dET5kEE!@17_N+2o3;X7t-w
z^t0F8QuqhK`>KUKdUO?@J3MvmYrcK*T*v>x+INRz-S>Z=+IyEMX;BnHcBqt$WXo1Y
zRw%RVb_iu;h3pk7tIWzME7>7rZz6j>uTNdqeLuhFpXWH9<N4=0?)&PB^Zb54<NbcE
zcSn1B_;YdNuCK6`{Os<Ip3KHA;+HuBsEPkG5tI;x%j6DK*zuIi*5l3tTwHrj3f}Uk
z-_`Q5MS3bquOG*nmM>*+lCl3imFgvi(;N13DJf+r-047)S>+`sv$^eJIS8K#f(GeT
z1&i*dLr@267tkO!$cI<%$Hi+@kqNf{WK+98yG|@hQ3}L$GIl#~50MiuxLs^S!0N^g
z8M_<6%zS}e0pi-YX)o^wG<dcAUbc;T53uZh!_vI$%GIkMaYH&UwOF?e)f;-@X!#9~
zH2A#DmtOGRfrJ2?|0PJ@0O8krk%9M|Xzd3EN>L;oRZBlkNi(XIYbpi&ASbm*jiU*d
zY(hef2Q8OGbId?a21-ABY+oX~1{=Z-2nJM(2TJ3oTqdH_&<aEW=0!l(YG<5Q@%7h~
z3cTw6l!VNEYRCRI3u3+i>2WXXp~57%8m=_J!l1$Kt1>b(GuKW*A?&A`=f|ul;SgMt
zbua?ZKfxk!OZE2ku_Gh~FDPbs^O`jvC03zcm49!p2epX^7A>kHVQ4GhwqGSIER4$c
zmC(CubxqphLVE|Lis|d*_bj7Bga@7iq=2hlVlRWgQXJBlXfIq(;zD+uCvnrNJqnQu
zDc7X#-n-X;v@#$Xmz(<WNNPATQ(N&smuIH6W;n#m&dvfD@<v}yIe|cxG-`D@={ht5
z$Hlcj7ZiL0(&V_bV6*?GSe~cnW(El#%A#7+gAA=beK`5sP^H1hdLLz_{`gb#<LIgT
z>ys<+7WwR_Phit4F2z2pxrQYQy{9z0ZewMoY(mbwpPiXG$bu_if3j^ST+u}XL|+PR
zdcsOxZbCu=|Miygaw+3AOHM9bCVxuM*nb<M`(ld7b%EA`0U?$>Q;hVDYg`1!y>u<V
z7y*I-o^TD<It6bP%hbtW;`9pWN^lp@id}rE@Je~B*xqXmh_%4|RYa1}CMT61CG{~!
zyun|HOZM)LkQlN-i=9Zq7E`Y`c%hW1auF#e2SDO1HhmH9E7IK}#Zx5OqeGqp0ZB-@
z2xGit`E2J9&!JYP)414h{anEdAVwTqsV{<q0|5W+WL6O4+Nbg0(W7dpQ*a%uI*pQb
zbZjgf#fwouS$gvm%{|5M-@PNt2&fkB4SmFhAWQZxbm6W5Ll%2F7QPpL{#=bOKx+Np
zb}8XQjPK!L!G6V|$N^{rSVlSIRVPlIxB^>sfGKAtP_dmb{2l-`1e6oJekm@<G_TtM
z?hJ?iKHA8<cYm*k(8tgc5*iYY5hb?V$TnN&I(0FeYzp4wraTB#8s!qOs!*GK3E@Y_
z2OH$2nDJDfK`Ry$s30cvYN&kQo?`07TNWN^Hd`gDI;WrdpO#`po8bVFsNuv!0H;3e
zs@;UG#wNgWv$waGJM7AB`eV)EO(`|i)glp<FDngtLC2^@qpZM%Z?!4?1^l%>fIZD2
zP)-~wZZeZIi-i}P1F|+;oyEZdJkZ(U3yLnVwx&ICj1dQ0oDc92<plCGUtqCpMSsC6
z3W4#{K4qd9NAA}4y+B=hK;<!Q+Y3ucfA#|!ldMejJ+9$^e3Mu?a2$W}W1~mM7l~pC
zgis1uT(io;3Rw6nR<3-5V`7Kf^~gF$V&`RCLZ;st8JR8k!=tX@teN9UFrwZ>yJO{e
z5r;1G<)kw7wY3~Te^zqX<W<+AjRUi}8X^y9m{Pz&GD_*iu-yAs3lQjYu<dz(GKcN|
zJpb(v!o+hP^(B~$N{KBDp`Y1iI{S*Zv-}@On$zX=G6cQ?7D^?9>L2G@NcW9bor3%=
z7cViESFBiS6@du`zOpqCV{z07T=Y}+N`8(aRCZzlYmHD;Pv051Bqp*@;2KXgs&&O#
z10Lf8q{1}ND5~kq))2Z9`p?|?F-f@8<G6mXrA-d>^u#b$;1rNXR$T4XXCs*l{j<Tg
zc6OYwrXVjnK20#LoC+}fIR`o*3+b(qIu|8zl@i5^C$ehW5W;Yv_V&z4T6g|O4R}Q;
z4P>VnScIqySo*_C(_~TNkD;1F9YBiZx?D716INrj*DTiIqLO%i9Ke=T>MnkuT6MGe
z=j7(397l!#4FdGIAT8|)7&CE%A)zT5GI#(UAlUqI#2XDs4hh7AIfMX<vKO+nV<C~;
z5tRn1O6R1Fvc%Vajf59f5Rya6dD+()89gX_NU(L>^IEQ58c7>9bY^6Vw_N$lSqW{0
zVqK2j8ajkeShKH@UkWKOOEnQhrU49WRO^DAO(97=0Gb3O8MuUm?8A4w|L`G7o#&Iz
zBu55F#OiH!hplmI)@q)?M~^|-OANft!jAK|T&dYEz0oe@wV!_7=q3tDi&QwISs~b4
z%TB=ck6-Mi_XhD_>mh6a+);wlIS!E|aK6ohm^Vr;1&ALzGSHfr$M30tvZu1TS~6`)
z6PP1Rrx&@}IuL6s^lAxZHI(83xCUcxU!;#9htR3pW7r&*pfV@idTp&QHhu~W2>Wlg
zq(+W>FDeps7(s;#<&=n)Rw#%hLk5cX5&5^StgUglT>%4JL1lJYTRRGWhUfeQBpXF0
zQfw=v9JNziB&lCmc>hiRn^oEjaVk*s!F|1In_WIXA75KXhj+L-=sYIn*ZaYhTm!Lv
z)b%V*TLL3uiT2N511Hfh5tFI82YJc;;p;#RgIV8-7L8aIz;JIktQk}6eV*?;%2$8X
zjud>L{RrYCF9V9GSeG<x(S)4uUKy>Cx`qfbLZeYH;rQtOeO*J(4=@MC{=NjKaR?qs
zb*${6kW;tACq~Z6ZNqQyhob8(lwu&uXps4XZn*;xm64hObXb6YA>|E>L{LCGP9?)4
zWJaAqoPABl5OVs_3P?XY1*`H))ExW-d%BA5TmCrx@&CY2x6lIMlHa*fY$67=ehKz^
z%@`68Wu|CoX_@g}N$*<)^=T9e*8W@+PR!)C$Kr}J#U&hEr=P5rajBUH{kdfa9{cq`
zcE+W?`sC2?qpsVq2?41cX?_djNoGKLUk|=r%E?_|fKVeG18^4i+``gQ0l{=T@$Y~l
zZm8XO6zqdx(L02nrTxGos}>CLY(Uky(_PfNn-BHobJa;KT8U=|&(}Dh7mxrqxC6BT
zaRl~TL!cjpiV?kb3AL8}v+kK;e3DEvkT`}#{sAN7P#gg|(2Lgtu^6?N&-~l~r`PAc
zk&)10KD#EX(e8;rc=%JnFx7IP{w9GOsu@Z3qRPtta`-)#r2veEP0D{p)IwN((z6NL
z@c}!M$AZJ_<uDsK1Z@+O#2vVeiS|9CsiA=hoEyPZeEW06WOp#YB+xunNUfF4^%Zfh
zk#@m5k1NZ7X7o@cxe^k<>Ev0^%toL;5@7*;8aIp=E5YM|#iAeU;LD=M{ikf#WqiK*
z*#@dze^|8k?N^PAN|0?!)E9VRsILeJjzf%w`}%GiE{cz$ml2x>qH9WgnC|7~3bASB
z>XPFSWE84eIV7`e*tpTf%}vtcyTY}j{F{t9+1Uv-g6KRy67Ub%HNbo<1Y##f3|25Y
z@95c<jKrjPV-&>3_1%9#;HNxhnm{Lyf;E7!<SG=@{gLwladYx@2%-UaaRGHcG4N?=
z(E~Mr1gil5l7?OTzG~f%A1|P5Ayhd|;8K7Bq}Ik3jYgStG4SMC{P#P~AHa2}vP<V$
zZZ*yw3*0A=U!BH_0e)~0KnqF2b8>P@M4EU>|71d?0aDHmL4I6-a8tBoXimV7uLb0!
zD!A1uQ2-z}0yWr88ll;Vy#q%<ofY~`C1RQk?w|Ju%Sy1by>xV_xj@x8qhUo8RD^mz
zEw=|t5>T564h?orPOzgdQ1-R=^n8YdmZ*?%{v#+oPO$%N2_lI<<0+F?mGrK_&%vTR
zShsUGyY>nskh~OocF<@Bbs7Xq;=qDVBRG&i@c2HcvWR**^1bRz*1M_xZEma+JwA46
zqQy)X4*o+c(4+!R4CFQ`fw1E|&c75$MJzd>TfA20{}GDiw1;l%M%fWx)``LyDiDS&
zk(tQ<XetVh=mR)HxfJKNZ_Vf%l91Na6ol$QVociJ64(g*gBLRe)e5r?xH<nrZU`|w
z+C-;~-G-;F7`#REv1cI!+Bn@i;X&21esu8ur5)$PK=6UyiOHJwy2Ycr=Jh|I5&02(
z{{JrAKdblP>i-LqqPum=KbPMXT6e}Qku7)2F4D1?T()1sku`WHRcrLXml8dJ7L{ec
zX56Sp;*I|P$M?64JUi9u;^~{JabVkj?$19D&HRMBm+!(_?`OyVK{Nk+jsP9QtIzbT
z{x=35|N8%z+=M>vtEnjhP)}v$Z#k>)w`|$#x2cCgW@U+NJ*$tk9$iPRrE9bMY;R`{
z17#1tfR3JFK)+|A`$jk#v`;TBs)K5AcyRd7$A=yTHc!$(ywxjtjSF&ds>#fU?Xm}c
z1yCHJto*rHja*BYB#lg&Y-RSbriRhNp3vWJk5*&<GL;C~Pd~+FAd4*#WIuCM@$F$$
zp@&h^=Z6A-0}+2jt1eN}Q9uBJy`C2LR{bfN#&Dx(55M5UE`M})a<VfcB?9OUL@JMQ
zX7A`i{UWK)!^6zW`;^U0v&4s03jgN4`HXpyw8So3<h4ayaBu$|$YlugEyb5vso;$g
zOZV3R5d{T>^a9`Q_kZM+g5a2bCAx9&x(SoqyA$Mg1<w=reDnJC`9zKg6{X4bSmm%~
zknnW@tfN^hgWt6}#>R~Nw%qL7Olr{^gg5LJgDgVnJ@Bolfc0XFzaepmo>2je<HM}7
z?U<w_B3?t=cBfQzN&59Rq#PkncmCZ?JYDx~g>KcI3CX`E$4Z|2n=d;ri~kjF=*{SJ
zTy`0rQ><##x<gL$?B&8+dd-tBm7>WoM3Tz+b<{<ST0%w&$zRK8IID6t2lH7x3=DO6
z=o`fVfoRz;{qh&=%khh8`33Z+1xN0R0^Qb7n4d3PpV*4`Y=Hy;GV;VdG4W5aIggNX
z=)Q4QanDasBWCXw^;IGigMOznkjsJ*phOf7?NoyXFAV`3+%;0@{z$k7E~~*GX(SC-
z-zJtmIO2lGw{#OVz{CcvVavgSv^`B`LycV&CABB<Q4TKf>TeCCMpR8VI$yYMC+6N<
zMX`n*i@`~z%!)}bhI^P?&t@cPG$2_LjrmJhx^Z$5Q4CIL8eBAooMy#YmY{#3<uNJ%
z%SNg|l3s(bQ_v&sz<$)5()G|x7|jTyGy^Ayd+BF?zpKXLy!w0l0IWKkgC!wITt)d&
z7y9KSBp;9YFI@t4h@PGVTml5t@y`;#sn#KPnv$rYJ`L+gtJ+1rLbzBA8p2mX6ak&W
zjz{<cIX0W2GX;p|&#AvDXKq}!v^CEuRCMDW1cV8qWfLJaz3mXKnuGmO&1~LBo_jx5
zRdF;|jQ{=}0R#MYIy$L@<UFF2QwM%D)RZp!p(O;h5JCO-vv18oYjtu!b>=fY6pJ?&
zM)Kgfx|CK%loN~%+1WWz_B>09=@aHNpFV*C5fQ6ZNg4t~3WJZ@+u2>z$Yo3{!6|~L
zreUFNHCE?4(<<aJ4dcmcHsz;EENZ~&?~a6ttWeJubJ@ngYdg3bWUJ9_jp16u8GJw*
zL8BdLAl{{$NFp7fA<v8+hd_kmy;YE;M!;6{+|$)nyRjyY2*-e=f;KI>`(S-vWyEJS
z;enE_Qt;)TNXsD-<xP7}#a7R}Nqzk-&fd1gZrLv|d4`KR(@kSt&=<->BsI^b{@;Zs
z3t}hC;MqYarEiYaFu{Y@H`x8q;p1HpaW}@lY=w439Kz{QjLON%Ux-Ec0;wm$Bf@_P
zT0RDJy^LTEk0yuKu32+sDn!udsO&Q5JFQL?Al!(rM$mRrVSo?Gi{(LF9muBBX;rgj
zGHJ*T0BH29K|l13!{<%-om96b2Wcj9=e+>(Qs-625J>}Ndy%I;$UFP$1Zsfr@&mUQ
z=lRm;=)f#>Lkg+zO2GvU+};wu&w!U(MIP1g@4wmG9vkWyjsxPAN*1@|S1r+7Wg+Eg
zxoRz}nf5z_DcWK=#|>#VYQRA3qSs3ws6Z^tw6vG3eX-}{Pt>pTBU7!0^+~~*&8@9u
zfoU5W7MH7K0OduM*^q0p$GAQ*0%elX3}4>-C;Wu~P8(MAn$lC8#tx$^@gL~`$xfY}
zHMW|=O}m@jo|j078pgWJ+#M1;H>0)H>3*d`Vs-nTlU)V4sKJW@6prhWy#t@k;wdOF
zw1MQ!<2H|(pC`)-UmEON+sKF06RNw+P@4)hrkO}~F)Jd?CfRlv_pz}bE2-OPIQ6T5
zx}Bb~voeAO;DT?7FFQAP_q5A)BRbbVk&yQ3*RQsPpC^o_gEVAi-!p93pbbFg&|MUL
z*N#%{rl!6C-S=Y`4+LupKy7kBZM!pl&0c4R*+~fb0wMO=C+zqtZ3N3I)vVs5z0ma;
zxfOg_Bu6xJ-q}vXt1`x^b!ES|kWWO7Go!4c+Rkp5o;d`TsBMmF`BAOhQ!``TQCM>1
zCE&O*&f$Uj45uhUF2Cac_u=SfaHT$y7fk1oW)$^>V``{+HU1b{I`4xXi>(fR{kczy
z{#Hmrf~fM^peunqjJ0-kS*Dg_agAX$N6n!#22F4E)=!;1dxKs`-C_5NZ`IX09nn%N
zTHb{T$%vjmk7&O0fcyFui{4vXS{5`k1fX-yvpRaC>RVacAyET|P>2tnOpjgwJ+6XA
zC{sgr2c+@4M~hx~LS8_;_|C_YL*0j0t;VrUmrKtpb8KccneI{W_Ht78LV<2r)`0^Q
zg+wKmLLqPk&?=RcA^7;so7>4Z9R@*vL3%|r)2hY7&H!d))7ANqTyguCtpQ1$BP$KX
zXSKv4)do85CqJ`nS_9|TuUpp+=ohr)dGfoU@rDwj2$d-eE@BxW*bNB7Edk>ka;@S$
z;0mCQ4<RZyQP-$W9MM?!-i{h)2jBo-ZOmF=f$jccFgHtW*z`dY+@UoITA}DNjG9+-
z(QXX>@q_X%a8NU<G>4xBC-foUJ=n1ZwScxTcW@mg;c{Y?tu)SdUh8>Y2dlxh8B`!t
z)NCAx@&NvFX~cnAD0ama;HrO7YeTD1GMXz<HT^Izum9EI;p4Ltm-%h_J^@0xVfyW~
zRHT);1rWTST|>b}3lM0Yla}6`QNOVdoj}dpm?U9{u(wpk2a?}GP?Ym@3%_;m2ehj-
zQ?xWRR4z;YnW?5yT@t>^Db_5<mgcuREMLBA;n#AIKPBUD2DZ8VDN(1#A>{xo{17%J
zwEvQ6WsoZw4z-Vk9GaURYR%Y;&jFP5_jsQ?6l-#)QZ?S2KY&Q24chts<r>Xwc`Jw>
zrq8iEr~Y@OFOxiDVJ_Mn3hYLPR!tS&M^Y^6S=a4i_6fIFusrD;cq&0XYF$uv_HG!U
zO5P^-WZ1U4q@zh){xi5J1?Xz+G)Q!4P`}$f>F?0Le!VPnVn8M6xXKH4ONAY)R;^O*
z`MU9Ny487EgT>555$)t^T(TRE9Q7I&j!V<PqKa2a5QSv}jM2r6khLA3e}l*DE)Vq#
z(d>^~3rq|6+_cqb`*o?8J0SJNDmmFhZ)sTl5Y^q)XAg4HgjND@(uPwM^(IGQZg2fm
zntk!GV&;#8>TQ9S=2+w+*8phFhsNPpfWXyH_Z7VJ`cU|Bzz^FUq?r?YIWHG)`ZC`1
zPw>^ZXT}&VS@k0(YuN{ErknKKPuA@`X5QH=3pe{}9KHoOwlS=VRI2kQ%@0$s`U@*7
zpE<ixBMUNGsTqzV^>u4rg5v2H4itH`;Q-i~<*PSyz?wn+HrNhw=|D8YZrFYf5X!Qb
z+|dA7+0hR@e4hnptQY{iKd~P{)zhDkAF_1$i-XwPLKlnyBKy(Dt(q9%w5vOg#q863
zwQCL5R6d5!i7a1^-cS^EvFk3Tvu6NI4%AB<ADjh}zFg+8O2xBI*PNDpJ9hXC>oaZ~
zu(jyi{&MpEd(<mt?hBe?Yt*>SqHnZ)LWaatVi)z&>@f;3YiPD#)qr(^vC5yvhE;>#
z=348(>f58kgUwlfz@Q;}JqyH~QnS?(@_xM^-{0(Dk-LBJ!Fuo3uO66|bCYTR&%}x7
zdTXkSTEeBR2)Zffht%;4ZdRBVHxp7ItxSoYo*MvZUr%W53Uh0k?qLt~|KwK{{AyFp
zsXABZ=I*Z7fU%AP1J%gqvC3kd>Q7b94r%2`oHW+fK8qUCweY$nK<>{r3v=GWPL64p
z+a7)R>EQHp$#IDeBR>*Nq(9{oMLQX>1KJ(`Gj<Lp&SK0JGHeJx@47-+XwM$#kqXc`
z-$tS`GC(ZNEjmCI+_Jkp+A&bD{{y<$m8boZG^mg*UsO@C8B${)G=(U@Qu|F2M*`W^
zULnghe%vNCL*m}&7=Z)uQO;`G7}qOa)7ko`feU>e2Kz*5*%Ms5;X(=^NL-vuTca@=
z#B#7k-1epJSr8xCG6Zf$ot$#3cpCuzhW?Ixyx3n;Q$63>@}Y^^Hfoz>fk|Hq0~}5c
zi(1+2M`nYJ`1s$V(c(7w@dc<-SejXNkvj##W{zh55FrPN&zFL+bFSc4-%U#!t>%bh
z86(l?@f|W5f?E3>0MlL@$#t0L)m{O=tS(MVihb=ncA$chh|aPH=P|++{bFz|N^*R-
zcKzJll>yuinVEu2XUMt+oH7l)!f}@4yeN{Ol-r}_FtKyBg9DRRu6>}DpUn*<kh|iB
z*=jzOq?B9(yeBTFQ!|=##jW=HYjGSH2&NqIv+o^w^!oKf2-vW$u8sS%A6HA?AT~Jk
zJ@%>y&Zl!dtQtq<BE`5b1aQLRaKiG-79<oxk@+gnjH($W=uPCGZ=?36;lJ*$G$Icm
zb+{#G{ja(MgMBlF`S&ELsKCt{xSN)njtEk{YoBVBZ$_+xqp0UTsSzOuP#vN8Iw^pL
zEu)WQ-OTB0by@;(y9((Rj(gdcp>R(Oa;cA&Lij?ghNO3QHZqs?>$Uh(AA0Z9GM#W9
zSf89F3l}%{s{P#F9SfX${ynUbH17$?nECJzK3r@Py*YM7eu`X60_rFwYPutw+WOIq
zEQ%xKpWV%EbtD8;?GTu@Xq?)&Od1PNb%+yHk%Qqz9!XW5QIq^ibYvgB3f*SlvHAkm
z9_+#S)(;eX%DPvY?ocrCZ<IA!T%0rGw;Dvhc{D*OS?STfn4Q%)_yYuucR}TY;!?@@
zLFrVK@*C~>w3~;Jpaq8k*Asdk53pcR^_Rd0WZ0BMhb>a^{JXgq>cihAt+~*M`IW1s
z8(4jCadBU{&5m!VCFO#YW9m&SqC$de@IEv~Z^7%I9N;;5QWwn;O5o3hg%8Aqmo7rK
z1QbRW5Fgwdqu_r|HjMx&<-@aV9q*5t7z}*@-y@*EvdOT#;#%D1L{#&;Tk6_Z|BsNh
z5r=^O*B2l2^Y4Wok-i3guz{}o!VqvHEK4`Q;_X8AyoVg;WsRM#4cD5*L)A$1m;m|S
z;q3i+Km=<14itSCfuFhVVEROq3&RUMg7od*S`TAgOm{PXKpA1L6tN18n`K`kSK-F}
z?Gsu#aEtj27avo3d-&;hk>uNMyKpSr#27m`bqQesxzPB0_a3h8FUITFU{j%z`jt4r
zc)Llfmg7*$n?JDBV`*<t9hg%RDTY%vUeOJ&HK5gT&K-9iK?%Oj&6P^{@ME+pf$CtA
zjQl<Ykb*1#;I4qH>7OS*@{Fp@ADKzfrx?9YCA?YTXOhx1A*I?-|MBCUw1K+HRPpWL
zmA(7zYzDu7%y5{~)-x$jYR-y#Uh;|7=u|~$b%r!Ou^wPS`Ext)uUa*-z%C=1Yd?bu
z#(DI^vxp08-`n{i=$k58Dkv73#mx4YZ|>noFNO*Qt8nVeWm%gns;4m*{!W`22uT=f
z%#dau(X2OZCIMBSJ@z>a=p*69A3jYEvtjig{8r1hdAs6UvcRB%A0b#inN?H0-B~Ed
zE+(^8zjPNl_?z+%H9cC4CPHe31kGR=j&9ButcAZle`@)v6W7KKP_xFO1Gj!OEjx?p
z1!(TbykwR3zBpE8pBK0d#$Q2xIO{Y;2W5#owMIFJ%j#%p{pQBBdHnZ=>8<0)V2u}X
z!eEY9zvo)h26KKdNB^TY=EGz!e1&ok-*%)tC~fjcIolN-<Dj;vTmcLe$nc|XX=&l*
z<$cX=IBoX!Hmg^W!ya1N+$n|v9c}GY$BM7Yf8nmrY&h&U(4Kt!(`iW$xf#fZh}^Bj
zT&48t)=kI{<cAD(b&S`)+_b<lV|#GA$*^qH@8pFmqKr@Q_x2JQ7Y_SVw|)`yRtC2Q
ztQq2{h`Qux9=-><D~g<xz_lTr$R+twpQuV^Y4ub^^=m(-4~#r732>=Aq&bU$%6~05
zQQ~^Sit#ZG=P`WdTQge&h6ET)c}l(b;5gW17&$RBZ~iobiF^A7<e`9vh_kObf%<g(
z5D(19u$2&ctx&>OZV*`f=jP^MoW0+|Z(9sA1YH}zMmw#2B;4-6fzINf%Nhdvv&fwc
zoj!77{&Zq_W}$~EoF6MHJktR&D678JAJL3gN(?mO14__GgaH$QTmaJ~f4c7h0wD*{
z2Vnhx6OO0AL3cXl-0)!So*!F}B77{b)u{(7y`TR6e)co;+xL{BUf$sT<gH%e@6Fm<
zpMr3bi!s%Je&%U-!kheowP&Q~W};?WY8Zt~L7{>RyFM`rUhEEZ4)P%16Zseb{t-2o
z^x%N$w_*qpywUde=}7c5`KOvonJ;Ct2aIHS{8MesLwrVBPx*N*j`xvaj)qdEta6vJ
z_$e?-1!=c}=A#@16^G-X0v#n>zfrSJ<0uA}$$(wk;9w2a9f%GRr5Ze7gr1p#gQ$)<
z3-1&6$z8MpZ7Ej9nZ<datSaB0Uk|<c_e>MDbjGFO2lv0dGL$DY=CuGYjvz`3IyPKv
zXG_I!EYdG!=Vbd2wS9cmkWxL^6`-l+1lR$mcTr1=kVxnr0HnwYv+A#l!tcN%2EkPf
zG?Ga*^=OoQKudp3u{ZFcZztS6*vL~x!WoGfg%b@DqRUR#@-*##BM$Y?&3$a_S{mY8
z;HWRiJgKC*>b0+(S#&TwY22)=JDB9|1L%wjm|u>|*K9jIz4XVkf2Ii`k>CRi5-6Ux
z)6xdEw*<+GPc60N*czragDcxkLnDsG=fkFjIR}0hYnQYfwyZrZHow^&UMyg+RjMXY
z0P_4Virbv}R|^mjKyr!k!lNfbP|V6lnd_k*icV_=;ejtXia_NQOg0~G%?s7`O}P7D
zU0ZMO7Z3|7IYz#9xsVL1t7b&Y6}m!f!z}+<{^x`g=dY2xv~(>%&?vog>i^xw6)*%k
zEzUdwiV*+1i~;fSh21P_=(;1~$4^&^!&eIy=abIl^~6o9pgAdX$T@vArMvu2f$%uw
zUjd+j-Vk$`Ei~YzqaB|K;#PN;0<sVt!8eIbgz*J5y<y);u>=o%k(4VRmqC`Awr?9o
zynd8EbODO?L(f^X7VF|%W+(gdvBs41t}aK5RW<rVH*wh{)VlXyn+~ZW;YXf@IIjRE
z#zN3<=P}&D)8nW0{q1c|T$oM+cWKbW5Gw;jaRuO9(<aMr3HCVtt~a6E=IsV>zvlLU
zx1~XA2tE#zQdS7>)n#Bq^gc|r#>~z*CJRg|T#A~aKOI%#+wiMiD{k?y;1njQ25v+=
za?37vCU9q$S+$DrAW&Nd8U;jlUYVffYLUFicOK56=1l9tl1AFlL^b*2yXP3O@t^W4
z=0`bG5_iUFW#2l3ZY-{&^PH}3WTqhea<ak;_d`Nj)E;e#>FO!=uD+fJ($$AmgZ#Ea
zNIa2O;1ne*H`gk%>(-)ZM(M+^Utb8?aTo4&E{3wjJ%IBX$@Q!=`~K2Y3*Gth*{RRi
zkWlBzMjFTxfu}1D5}+%$ZrjE9_wxR1KTk}Yzz(oyx=U&km=?XrO&-W?_z9&>UrWSl
z+|d_bJhh%M0@?l5Cr*B*?Ge3m!OnMs>vy~a%CT3-{9Hmvq{Gf*iex_h>Gp!?iGkYu
zy1M6};$>oQTuoBXg`~`au!C3<daa8K{Zf*C|17eYBdn|*reMplq)|N>)B%&U9xkY=
z@>2EixIQ^a5UUOJ>~|p!)U$a<PX~PiE`UQT*X{UFShy0fJ91e+;w&+!3}0znm!$Y8
zSkPhChoHzPlL*iW=<S`u95+A%#Q(W2AbzP_gWI}nH%#weI+^5mkMp<r0q%@=xRDQ)
z3o1d#uY-uv({aHT>*+f}CAI1jV(Mh`GzGEHNDOUb!%wN&c%?!>^SG5ofRr+!t4@N=
z;C2}f0YGZ}$1b(10V8~ZWB;ftNqGSB1a^|czF-Zmv*Pr6L0a_sX-d93`!HI99bx-O
zP4hs${>w_q98YO{+L{nWly+f?rO#2`e*-kVZYOIQb|SBB=Ni&vPklQ8+I9W>Qi>@$
z&NVqsT8U-=m@WJMXf&$9;X@T$4#_O9_26z&^({Qy7&m!#5SJT%x;82U)gggHyN)Y=
zc=twUbZpXbL|z|vgD1)_Z#FF<m~)5+ju(Pi!MvG=BDuK;IdVnI{@SXETctpBbw`m0
z$>`+e<CEodzi(=4Z2WKoUHoEFX^t&m7|X$Lcc8xhg|Q;FZA&A?ARjw`Nr;PyeP}9m
z1K<cNv*+BeHlqdz?NgwHotR3|{~9}#eYVh*qU;~UX<zWz(lhP!;ymh&8$YesK0c8M
z+{JdxUS=(s#Qnmkz8lJ;<4Z$4i7s@WbOJ?pXMUrL-GDN+Gv&?4IBOGfVf6o?q5x<`
znZ1sLF|%h}@xpnw+Or{r@5ZnFp8JH53+xI!gIJI;<1OrYH5b(yGBtqI*R9ERRE|n1
zjKCkXM?wTI!lXneU}Lm#;}?VDf7;jfYIsKSGt-P!a%{Pf=i%A79AdJ&C5rLivMwiE
zZWlV}01{BU^3wyDYVH8aCvv1CJUsAI&Fv_Dg5v9hL*vQxk?$Y#S8x6TuM|F`ZW$m0
z+e3GJ$6xcA%v$D3*Kziv2QTv%3dY;m2AL-kOwxj1?4GEW?!Tw!_36R7IE$B2Q8>4c
zkB;jCV}61hk&JyKQKf!qw#>Q4L|TqdKy-uHd|PQ~T*AT*tX#h<a`dtmyZ!VHsOpzm
z+Kbn+u3VvO@NTA^K{8dePy_TDj!BEbURgAS9!N>R%IJxaji~9Q$AU{jJc*5W`RCOx
zf|@`ODtt@*pskBM=d#UElTEzw-8-U4-bBymH7rw6qpQDc%eF_$*e1q_-RK7>9(V;o
z4=^cao;1<Wp!u~hwR_K=ys39W*Mge?nCM){)Pr51tc%C?<+X9mU>^$fw0r|~$(JF-
zzYDSxa<p!1nJu*L5ApMUFMPNIpi6Eb1yY;1nFjs7A_HUNQuiR9YL?&P4F84~s3fEW
z?U`J;9(jz<n!_Gv7__{i(a5rrw9+e;mX&K~?wvGAyzf2zYp<aBeQd%TePdmNo4W43
zL2&7!`g<>V!&i#?h4v!LvcvSo2{*sv#ltcj49B)RH{`DShbTX2(U@-H_8>~n`q6=w
zc4Ro9=p`Yz*KCf`Z%`8uaHQpO1y!{`Z#PA!RXwx%#=J$!Apx<&q6M#14r!mf>%TLh
zK-2fTQN_3KcA_SCWok4HjXht=rHZa(F{)`fb62c$X5nLwFbTfU9-ni}86M~o>ah#?
z{lcOm$2@O}W+aQLohAKx!(Hjx_>n9cdLMX?u1r3^&09DZ<D=bFmuTsvlT^FJ-1naV
zhY^``=ZKm?EcHa#7Wzy`Iq$|u<&V_9@H%UMF-a?{2<1s%bBy%c{3>C~|C*vv;y;#F
zwq>MvYaSH)M_}-W?iBx<Yw<6cANoIYA4b>O>0A8oK@bO8g#Y_lu%8N!T>Mwn{^zIK
zfBKKcm*BkrFMqt41}N#-UlS@B&o^i4@mJMgm=0|gZkulgCkg*v_#JF0WNr^|3qtQf
z?6|lf_gH3Q&RG@Iwau;7-~gi0h(Y9YNET+F4Ja#OV8S5Kv!330VCMdCN<i!>Ugv~~
zYe*BmCmIn3wdJZbDf<`jhFpf=_dap~4j=B+`W#j8ZR)hq$Y<;wGJpkA@ZuMIoT&H-
zq6?-U;}sW9rYYLfqJ)!1vtEq;4b&n6>DwV{hse5GJx0zIS1*wJ-6suX!iRvq!r0@6
zcmA`t_c10tqM7DH$=5@|MA}iyNP(F}KlHJzYy%)1a=4R(h#y7c``3Q3>BO5xVr?jT
z$~J`VVs}Gs4tVGU<jf#7IA5`#h9{}NBay=>AE9N=2U@(F&-^fo+gnFI|3&lKHL4gH
zL1kPxd-juB2ebf^aGVl^0`yn}8Ujos4TgE0!)|g=f$ld>Z?KUk%vw$%!xkV&B;3V_
zk06vV8oO9G2XJJ75TRoR&3YCW+bHlp$ehQAns|zJOTZEU{aTX;t!Y2o{b>Vm(@(H*
z3EPdMYRT$H)9%@QVnYkNQHTTO>#IxLgR31$9InLa48oKEU4&o;c^=rrBBTlVZeew`
zH^3T#;Q|dHnN4tiU@9G|?XM?m1@KW(UHf*X<0OYpnT(`{vM~Z@CAK<L%0wK5jPf?1
z6PeTKeB@w)!Kn4PIduRyry!#WhYnzbXKn1c%pgvE4~(RdcP;_WhWF^(j{#8#Bnx>9
z>!dTY^6QuU0Nx&^>O+B4f@y`pH907xSH>O?d6?}DZ#21nzN~@5Qwdj_93H4FYBU9B
ze#kw&-jGC6QBl#7(o#vdDbHPY1fA|HXlLDpELsp>t+r+0keJ;?BgpWP%R(@f=^;P=
zIh$6}u|1gYuuK6jdt4ji=6o<NXITvjcdIT);=UEah9^%I4<a@$b60x@f-w#ewuB3M
zBQDb)yenhLt-{C3@Gy7vhWGTq>JllnfcMWeVZB69xG?nUUgI&Ljd)&)&iySpS(-C=
zQV8*LDG3(HfO~;6_A7Q66I5g5lTx@4Qk^r@XqSQZhZ!cn%OTMDVPoM(S@kPu+y=t9
z<0Te^pRORZCmgX3ed4y1=oA<)zCpPLr4C6L!Mf=H;(<bn4`U--z(J82dW*9IY%t6~
zWKk?&R*e#=MWwlM5<v$shD+F!zrp5=U-$ZZRII{vvk$iP)v!u98k4EGs9}D>CYfJV
zC3krkofZCo*SvEL{9R0SqZPtS(8a7mNi7M9{CGM9Aw)$_ii>0N5(uC#M>x>9Hpc(N
zYUK%0Py^KXs;R1ufp*};+e0dUi|wWMj?2(6z!M+}Hn)o5-$#n^nv|!Vc9l&{dqA(S
zMZX_v@4yMi2qFYKt?C#X;5gU|I{{RpPdHLL!4$P-&6*F8=^>Q%Fb&~Suq$c!EsxT=
zi(XEuG=LhF=GT(lB#Bl*afj_sa=>$Jk>Y$7R1A55XbLV_?Wy)YOsTE}r`%hcKn<QS
z8Gt{V>=T%^(bmQQIobhDq?zWk>FHa*cOazvfOr@v5ZCQEdOB3d;hoJ$h^8P^&A`}0
zK@vPe6m9dC4(z6zfPpE*Y$F;$CK{1Z1C<8jrr$dw)9FCzTViytGrDzXuQ3wU;q3(G
zsf$c8HeldOsKX=;I9pr#7Hrb#*;yP2yWn2@5+W!7o6&yEQKROB0RqnkLRPUcUQ}^s
zBPve-p4VuAzys{T5v&Tpu4s<Ma6-($ATlwjfsP4<1(w`a2%FJ5knl0+u5c<rh?Gp-
z3fVS(BwU-u;nu`6ByK8cqV@;%I#;BEym=hrajrP<j-eQh#0Bx8ls*kxi6S>Km7cyS
z%7^v#YJ`gxLMZM1N(aN8Af54sUhrHK<R7b$7UrFFeE)uDgt2fanrsbyNTi-GPv5b0
zlXM9ip%NMXjJO2KFR;Okjgx5e2MQlAJ%_3a>Km_4{WKIu6coU5NoJ6-ATHkxCq%!4
zV{Q`q1rp73_Q5)a#^~}MgH~y9WXpd-yCJ#57-2m)2uUCgimy0R9b2yzSNPW<BZg=N
z;^ZC>!`SXDSYJ_HoJE0Yr3YtX7|u(`OW3g$AzM6yAAs>|w5al{9bPVjGTs?}#H%_w
z`A|IvAhRET1NYiSa<gI&K80Nxl`^FBh|#_aQXBuL#dcqu$!A<2eI&R80t!r~!1>^*
zJHW-sxm<JcYBkhR2t@jV!Iw8~+@KKWAFikya1o>OQ+ape00=2AtRc)6m-iDhhV}w4
z5;rh|or(LCCkEBg8-I=Y$pFN|FC)WMk$0&3>*q(J9Rhm70Z+_16O7ywKqwKc+GN-u
znMr`JL0pW$9jnxmubOW#{D;T*Mes>{6&Z*rfVmOx4w0M!bjBgi4rR51`;E%Q4@1Ml
zmvEVbi-Dl^4z4&^B!Lj~Mf@cB#W=MgPZjlLR$Q!ozxEkz-^4yK4>;%$528czF<$UK
z!%}m@vG@jLF1fOxoF$P5{2|Y#m7z1g0onWE9>dznb5bTF-|*j&Njex@8aq(zN6V&}
zO^{TiK5^oZhpa#pqjf1j0x6KV94U6#Xrhy8XLiq1_zZ$^Wt;;~T)q2o4n4jmRX+wL
z)$F;P<Rb_AcAX6Vk;#DosILD0cumCnP-nrP(ls=ch@jW_W0?7PX=1PQ!*ueI*U4a`
z*%~(k9+7kAfO{4-b9Iw7=0)p4W-Z}!Gz<*hL^={9&ceAE3pGuD3EpZf1%fH3zCtjO
zqEzu<OETZ0U^<GszzTYkb3?=${T697=q%+Wn0D^k6;08@&`6RJLrub<R>mUTO+y7m
z&d!%0$hQTXVI^!g;f!SY+o>&x8E*Q3M=41a4^FoPzO#*cd1Vq+QbfxU)yy*`kN>>Y
z_6b1FW2f;pBygr*V!`Mm+$<7S(ACiF*@L*fV>@YR0<`H!zBZ)b3czKI@PUep&YwH?
zWv}xRdt7i68;b>Hu?f}P@Mn|>%~J#Q^tb|^jz}(Dh2~nxSrw3JxCzQ88RxOzzwM04
zsaS$B7J8T)fn!6GDXIp74X+o=e_4?;z1a<*sS%>zY%Q(`IW4L;G&MmrBRPR<rgvx#
z;k{j*oskq=!DTfe=9ymmocnqq&!)gbN|QZ;jZ1-bN-dnV8kg=$O%I!T4j!H!h=`CL
zrOJjuvX9_~%a}ZUG1+?eS+&q-&t${HK7nmygjpI4&kejafE96eq{m0eL^znNqQ({n
zv&m(MyW#s~KFyM??mh$6IW{&Go==V~h=5r6m`$YOOl_A&UfLI2?wTqPNZ;|xdmDXT
zwdo&`&tCX)SCf%6uaw{1MUBzX({llUBDqNsr9sAFnhE9l^kY-~M7BT}2ttz8g}{9e
zayX;!#>wK{yadA*V3jePh;Ly5=vz3#anUwswOo82c0NESZ-R#c;m063ete?>fTl?B
z6#s3*F?w?kj2XDJQM?hG4Ds|Kp_Zr(Ec0)sMa;@^;NKGe9DJ8vbhAM|SgTc-dKh@p
zkSHAq#WmFA2ZiP`A`XWK2H!afzH#FI#PLpitGG~iLcuQITx4@*n{(e6#Hj5r9*#AO
zkm^L4a=bDfb_OJ~<fFm2(!=qHL98CQyydY*qAvz~!r?C?5%p%m(D;HXl6?A2?OBGL
zHleBqeaKMRfmt=nzirGjL7fVjq02UB$X5%1y1|(vk=%mw>xjw-gzT#i58s`;Hnv%l
zQG*JlZIxw<p~*jo*bRdvk8&V$;q=WcEG$GFO94h8r6`CJ`tujDCF212r?JY5>)xQu
zD0$8z37asuaLLrikGG7-;X!+G;sSt8H6JEza{YlZjle-qtY+w-ED(MUFlZt9>lynO
z%nGp|a2m1m@WkMpg5hEf2Hp_SBZ5Ah{igIJH0ftC_d))ihZe+$xpTt;&`T784edkC
zfdvjPybN_UlyK0vzk|sU{mT)E(U1V<j|#evpCc#HeZ;lcj<5d+Gb&<=?RvXNCD29`
z=xrs;vrwy`mSM!8E$F2%5)~!cQ8Y0mAqyw=%3A^N)~#EIi9*6sJy#3_sy<a#vNv>0
zf~<x#gE7ktr_dcgodMjnL<NaS(UAZg_X^qj<5d$8M)P-DlN@You~Ff*CON7j2M>kT
zrT^tQH(vlq2*-h>^&29RaLz~?^T!3azi;Jsz#uno-1ru>T-_K<6bzK!IHgId#oKG;
zM!igHMw27!?e_BtFN|+@ZejkeieIoFQJUz)4oO+Ld1_(Jpkdhrw2}d0E+hqb9lEdy
z+|%Z4hdchr0)rA{&AN3}1LE;?Cdf(Gf&I(X-CfFcy$B98QXWN1hkgZ$0{JC|#z!M#
z#*1+CgWuVa-opgZEy|fpyIBVo?L%unSYi_I$`9xr@FOMIx$z582+ASH=VzDh^Izs~
zzhKiIcJ=h|C_;h<Tc1+;brmcbH#w41guSy1y&t)N@FQW~`0~?2Oj4lgp7q;{cqXI;
zig!>&DgYj?G90QXScnk$!_uRrT+QUTIK%h>DYnL>QTnt06kD@MV~7Htf|;D2xXF9B
z-Bv}dzg*KJL|hzyk&4%{cMHzb>ZV<&wt$==i?IG!D7(%oTy-z;)8Vh{7u~=1*|Vwv
z!KZtNZj|6MJ=cWG?F{Bx1&TJiNQ*${29g#c#7oKc_)!!l$9fM>Pko&Dl+2y4T2G>L
zzg7`Kbn_%W2(E3+F7-m+*YDM&ivA0z?UwD^BQ7t(g$=n-tor-6NrSk@pbFgy8JEVs
zqY=ehep(Vswz|4H0ck;uFcpC-34-owp!>%$tk7RewI4KX`l=YvhFe*D%Mp*Ll*ztz
zbJeE3u>)%%%<MlIoS$d-IWLUP+#bcu;SA%H(c}co&moEiwG0n{I`HgE;bJBXR=BMb
z_|U_Th8Z*N$`~0g^-A;$$B?Hd=u~g=w~40_$y#QJQhQ=CDGO>Dv>8R{AHaF`mLQAo
zD-P-jrH8K&4~1s0^AWuxrc0bnUPOOE>D@Tg45K=_o^Whj`=x~`ulemX2f;W3D>)Ka
zQSoic`2dD!t*1at5opkaQt$R?`vcZHtcz*{3<gUFjY*E?4D65%sKcO{@ofp*fI@<>
z3YmuKSdwekua_JnLJ<nIE{HKCu)>GZOA<h_zB#bwnPL+mEJQx+>`Dz4sNf*X6hFyv
z5+NE0NB)M@I{tpQ;eJ6%^)rNHkcJ$$gXQ4&X8@GZ^4tZ8!V0s#D{zF@prS}_35=_o
zPosoE3J`R-UkeZ-pDZweOXrT?@0O3HE99^P$>R!lxszr=3G|;|isE8tuUwEA>z+So
z#t})^7YE9PC=<X=k=U1dhDqXZ090D@O2r!`L4S!9B>zA-5yb(84^GZMEozP}jyxA6
zuOQ0h3N)>@rG>^vx@%aY0XOO0$V$u4t$4vFf(FI1ixB3>DxqoI^#^);#!bvV_*|Ji
zjeN;yFrNSpq7Xa?r}}EmuT2>sHc%Ci$xp-v1~`da!SHPXJ=%ed8l7N<HXF6-13)E}
zK*oRSRXl?^0K7GHri$dV0(H<vS&J@nHI{Pj!lar89Wj;TW#bSyfgw<bD<V)b|A}XW
zue<l*6(obvkw_R$L}F^#<B{pD89maD79*4I8-)O5KlQZF1wF<q3&Tr|b0xR)uebGb
z@YlV0`?gBTaK*zsf-B%nmZ40eK`c4gH)^C#p@+K(Um5(E_krKx{+7sdJ0e7jZ>x_6
zAJ_Y9RI7lwD}ZRL;A{c5mHFa99S*u!#F4>THb33r2`vD=w6|_^qt$BADhYp@M?x*(
zdW2H!CBC_TU?9oQzy^i>j&zmZ3ZW@^hK^e%*ZwvRFL!9QpcW$@b2M;r8Snb-Mm6j9
zEf%n7<*FmS@z`nD`C|cmC7_Zzid+`F$|#B=IDadQ+Ra%YjTS`_Io)aCLv~^+70H;S
zD1x?Z9;*VjSc1+bqX3Z^uMyK9LVN1gYjZ0rMUr$1#cQT*PEO9wmt@!p4F5HFvvSv+
zjGDrH<cKrW8C44!m{Pm|E33nLc&H@+@<`FS5mM?04<E(@T8hQHi^Y47LsfWg4zer-
zsuGOtq=Baqdxy~#lK;rJ85x&R5yS%$V}>)5BtGqARad4m!lTaoj#h4EEFLUl{DrJU
z$l5aBVoPh_2QXny4R~ATom;nt;qAp>7@w7-57Ku|Nd0^8=m4WxCpKD%5+KI`g-Q+D
z97Xt=eI7h`KwvRQPmHjtUHV6wGLn~(KizKPhPgoqLuJMop%Pp=Xg?(#F#oR~%gXj9
zE*divlC5Gjiqp`{T^K;storHE7C3&o6O%JiQY_*-Qd3*o5TLi6uKMYebJ$b|G*dOs
zRLd36XOUk_Iv7B|N@xoEv2PuDsAcVUkbMmQUObL%pMU_Bx1j+N3rhm_IQd#sM^L@*
zc!~4b-#`s6UraA}0U4!_IV`jA7nUH#0&nILE+kgjd1>K=NCW}Af?d%Uzl&KlH8v3O
zuz<oOo5W+bi&JtB5mt#7X3H^nQ&IOS0tIHa;re0p>R&Cu@Q-*LFh;nan6VWqg|JMF
zfVag1dME)vr2=^#Zj%IHT=6*Na=rr`OaS|)h_EL;GqYsCWIkD0S>*oZw;uG@ypF&U
z0zZ;o3oWfY-l5c+p@*r51h+d&Fr{HgS{KH5bLfA41*}B@8FC6ZMhQgd-5dkq+&US;
z6?1pvQ7NH?XGRVKc`th9fr?nyiYA!;&i3`&ikET{3)pjHJvV&+eiU(dWYteJndQad
z@iCtsYR=;McU^>T0``K<Fq<26;QmL{<B6a(7w~`11(LWl2VgLS?pWb2mt6GZ-#|t&
z8y>TkNJLDMD;ZTn&Pex~t4|?j#EqPfY3waF^7l_XmUG34_btOrt9i$!CB6R!pk|8z
z1k}e<#05r=IZG>0jRhJcSsf5s1=u74H+|-N*`<O1CK{7EFZ8vK-~F9`y3N1;u&?>Q
z@E8BXCE4x3b7oByhgJ5Etag#7b$XjS7kAR%9am592xg$r3p&mx#s9SqBRKp{=Uni#
za&gYF0?Co<JFndK{_8jU=k!nK=I&GSEBg=4gHQAS!kzrVwEq0#&)*J>MLU~mH6*dm
z#zy_E-l5~ipTJ6fRo~dnY^ZT0>-K-Y_`!R+x0ani2Z^pmKF`r%tf%zC^Rj@OrM_+)
z`}PrkENY}{mfk|*l77dpLLJsw7U2CdOynPuLm^Ss7d_BGA*;V|D9H4jnCE}r>S!C{
z`Y4mu*54Z%$8hsx6iM$E-Y1}@7C$>3IlCK|_i_A<5?|)F=f|k|EIMBT?IL|hY;5e|
z4kQDeeg9VPrnUJ&eUH63J06k#$nMW=8+lU)9>%_JxeoD7j6y=F=X4BYws>sXv}1NK
zMVI`Uygj5U#}z+A1h->o`LbzLU@wAS9EIyqik9cvQm%uGwnMyv_^3Z}S2e;Nh)V|-
z9HJCYTvADqOBLISCa4_u(Z;M3?Ga==V9BjpSM-fbfmwPzxmgiBmpEnS_ceH<P`aEe
zYC1esVy+xdPcv<vP5gXZHS#g5<M(eU4!I3~eE&HaMeY6M)ORpDGf+rj_n$|w_))E}
zWr$TfC+t|($EmNpjkd(=n+*7XvUj5$>>OU3Tn()-hus;4n#_BNof<Q<8~2R21B#)+
z5qw7EksPqS4JHwPA#eI{R1Mz}jW7Gi#kmUaq$x&n^ag$mh(B&MP}7FKdk2fs6V!NA
zuX#?r80{!}W7H)vNzdzyTahI0o;0g_0=z*M_m=HADqjEHs5UO>2n**1vEPGewV#em
zcH2P9`|YK*i)S@<d~f65+=J2FTE^$X@VoWbBnIhT#Aj-;&BmRQYudC6p(*GdFR=3P
zB+dWcj4DYNCp!tL-9#tQ-%c-NFNT)6w=q%8YvQ)8ZNFe1Mw39JLBcgiHb-YgXlS+m
z*Vnfy4^9p@h)m}?fV#^~3vSP!8f@S!@}S8_QAZqt?CEHM3vg_7rsIx#W7r}3_Wirv
zKOPpk?b@1aW9HRFTazq(XW2@xACu$7+tC2Hml}V5VmrNsvgAy`Nxpr^TU~?ZK>jvo
zC=HG{y3>y0=ZiksD~_WBO<=VfpY8Zjs>Ma()8ETy0YqU+cR&)90<?q>Q)drE=UoPp
zP*_z}^=m;hhiPzdNdCzjz=T_i^S1*V8UJp{7z3;aLZ|@`zJbDJhYp<WeLtes1B>qh
zn-?ckgvCXgAouxsWq_lI&A-*KJxZlI8vTbaW>{1WDSpxgvPocK!pfN~+H2TD`5oKa
zRrvG6Y5YtPgym6@T#SXeiHP7lVJ$9$)41cR%$W)<?1%B)s4n&l3PI%zN5Oq&Po`~y
zxv7+4ctU$#Wtj~oV0L`i9Vs8Eh$59o+cQn*1gyes(~J=8MvF&<YToLz)y<pY6YO&n
z17z~63V@m#N0W64aq`h8P~&5%h;~eU_m+qZMZNS|XU-a9Jju`&ARyA*=^#eMp92X&
z5{L*cqTZm37tJAr`2<)*67`Mu;6me<3Qy7E!KOSdVf?m$Q+lpR7w7g7j!RSfm&ntr
z8iUJLZ^Vp@he(Qhu9;)Y0ArR9v!bhF!sTj=`BZOxCle*<d(bJ@9<=X0Hsjh)4aF|N
zmHN`bX#`T7oYw6+`f`sFUV{UG(7;|^bf6Jx^+}mYCGVSm#KNW!{$Tw?J6uO=G_w8U
zg@ih9+I(g*R&w3Jda<5WV;aO%tL<0B>wWrK_Z+W3<#shA^X2$+-1_?J8tUMO9K(YW
za0BDOBO#NS)_r#q+|~<fO5SbBvhn<Q=MHf?$GlRxX>~KqD}x7tEg-I_Nnp(e9E<4t
zHlFSRgOQS^2RzxOn{vG}n-5!J<^J9#vp4v+_*Q0uY=5qe144?V(LkR#-rpPqSSp*#
zv81zj9x;RnajjwKlFJ8vo!peB^Yg<U+d0FNw6rrbR5$wC&uh#8vY<?J9~EO0LK%i3
zYAz@=;Jb)DmH~O|P^J~|XYb+?=Gvic^r1aBZxZVpz0<4<uH~rR2~wAMv!f!!B*6KK
z0q_J7TqC(i1!nlEfQ>2SSGT6)kN8;A&nfIf#AU+#?p;Md1LOSW@A<7AId(Z&Vr*o)
z7t~6d=EvMLJ!M6tg7|%fFTL20#g{R8&^}ns73bqsQ`4`|x^Li*o?LP@%e3hM*;ems
z+^Q3o)5<CLaGa72k{VQh(Z`Pm8d&pGp)K3r>3(jnb3gzMtcX|fl_Al0L4_;>6qJI+
zcy7UVf&dUYF_*x6JuL-3&wJ7*S}O2_;g~LeZe{>bbPs7?=G`<jZS-5n2w3Dg;lqmp
zt|yje+-obZ;F1BY`qJE&dBfh5Yk=y04HlGyIRW=-*OhbUT$>}xwHWw#UTIa}jM$CQ
z;C$xYouFtoz=PEYPh{@GTyd!IQmMsBu9a6NU#So|CCOff%#X6~)UDHs^z_Jzy9j`5
zG+y}`PPI2T^H;4}y~-RTWYL04z=o@mX3K@o@bkms!jnD1p!~*U=G`}6b6y0K-rg_q
z<?c%2=j68<z%$O>p+*1iT4@Z+1BVvvw6vIN1e4jVp>M6$i;D{J>J2Y-2p8ONxah&<
z;EiKBvmq~i+P}2r<BsPnULk$n%<NCEH@|!t1&oaQnW3%p$v{bE5sRx(m>ci{gn9>z
z`p|q~2y#d~KPt_#>9yN&*un*eSNY)Hy?dt9IuKw$%#f+W97VE2A+1wpc=n>@z0PV0
zCZoC8{t~RF$H<KFqH`(uJXFp%wE>2?^8`}8ym1e)I^UbrNWF2O&2EaM#ggi-e*qmw
zZ&S=>v<A8u#B19Wj_Eas1?KVkjf?hZl-J%(FDf|>Q{BStO#zhowEI972@C;Y>7$KB
zZkuO20ptd|^$p+gi)O>l+$K$Ob=BTsXU-76J3LqE^YfelGY-NkiD#Ah9X(3*8xbTj
zy$?U82467Zd=cDm#!ToF7CI+#ozQ0ZYA=&%Dk7$cR?mI43;G(s)S!^<t9f&%`rf?&
zKI2`^!}){g{%n6cP>59pcnt?cUA9ww5y#->Jp>`94xCS9s5_+WDDw*;SEC|+T@V;6
zG#RzEE@SqA2c6)$(|7K!jKkqJ#`aHKZ0I{Q4dzgST8}jRt~+<ny4MUB@^0W~c|w!;
zJ@2f3Gw$Ir(S~^i;^%XI1zFfd*e>GS#U0(<t3F_XELsT^#7G5EV2Fyu$v?k1H<Ep8
zbQwbMh_x)XG6E&i>yHzn7cM{sT^#%D(CyiACT~XR#U*z7Q<R^e|ByQTJW(yZcUK*g
zQZ_GmsOjA2g8~k-Cn-2#*UV0;k%f$8CgAw?ZC!=m*>$!}oZ7LZac}u@Bsf2z<1Bi)
z0J9c7E~Qwt<bAjd<#MJsX1~yG1A{_Fdb&hkcXM_9gf~6XvJzviqhQM#?``;<lW6no
zz=a%(Ic^H-SZNEe!mFZo;NG?^&I9CY!OY=>(J6-md28S@27fpPd+Fq0%SPvh1T8KC
zXEeW7F<F3&-9PWjousm|A1|;0M?q8c*A8gJZ4)+YI9q!;U%$26jV)JfTrzlaIqRuY
z>d6~_nLn8sfO{B%O$1_JoZCb<ULqwWWjHrcPnjRDmX0&>{9B}*ZH3Erjmm3hA&0sC
zjXe@yeTu#L5Ox3=Q0dD9`r-j6Z?>2Es;%Ft{b_oX*pM``YzA_b@vN60ZctnIjcy5D
z+fqDTdv%PI==C4(p24j|*jyB8%Uk)azJnpeT~W*MQ?9h~<DJ2zq?%W<%aPS&J=7vq
zT5nrn<arxgVV~%E_UqF(;7=utGM~-lah5<HeRPhQ)A+trCv?AT0UfuS&u9n0%~Z42
z5Uovn$wV`tD&4J`>qgh|A{bJ5X`T~W*@9YU`|1G^C#;xj-`FD}A_6#Wa{%EAH{E8D
z0CzY1QHOBPN&tXetMcwCx>($-yOEdFnysfWnB{XokRqkEVEg=OnbA20p`{mY$>nBY
zL?4l}^LpD-^Ut)@;`axWS+eFc?<PMN7KYfDLVS5C`sE}V2POa+K(XK=AKOGpxD**|
zhH_(HbC!H@QxL!(**J4O=G%7iaIoC#o)R<irAF6EGL)gAmte}7ZQ1(}YE}`@WiCNM
zYT*U@v(-QLO<0gAQ>fS?Lri<ko*jh7YAJDa=~}Ou+^G_uFf6qnxfYL5&Tt|;c@ef`
z0+mg$iVq7swywP_k{j`ODH?z{i8LDPsz{!q%Y?D*B8w)`bDX%|D^r(EEBl-lw$wJf
zOZ99^lAXZ1dd08iCp3^PQT+@Q6advi^a~>Z3g^(9Ke3${ke@4MquIL~!)A%7?3Ab5
za?P(g@1g3|FpEp3<&KmIJB!S5H++`YU`EE@9d19wtv99%E}<7Hr89FZs_&jUw6z`A
z%>Mo|ERTYtfCM-eC(R345TE!)E=`7lG@S59wUYKiK_26P>9YIMLql+S_V)L$`%tKx
z8R=B}`hd;Ik4nL%g<jc9=S3N5NI3q~DWp<TN-qX@J;|_DOwwR(gmNA0HU+0N_bV-F
zF|jq+d!Iq0VTJDIv3|vB@dE=8pccJ|s|u{hJuD*%$q-(JTp7rX<*gQrr814LUOj^|
z0v)2QNW}Kwjs&fF4h9R4#<3&GN&m#yADq}gGmFsRm_4%sNS3nXa|F^EI~LzTvaU(q
z19UE*%W8Z^w=<IIC5Kz{M{qh(GI<Yp5r1v--f^@3$#R}2G@K8?RMIH)_`lJyT0+d&
z-rO|fYIx8#(Ft8dl?vs;OC*4Z532z(1o85jba5!|kIF}Hs7uZjp5rooVGygEP?tyy
zZmTCpcw*h#t1TdMKsEdjp&@5-mIzPwTGo7Ej!}xK5VB_!yXz<V>!mrOaY2ed7vSfw
ziZ)SWyySm+x+P}>MyLZK&Vr{-VHW*QFtw*{aV;)c0Z7v+DhtSbbYMG=QC1cP0rM0O
zRVI6Bb0=GsR`@V&$DGcMS=J-nRb8d8VDu(8N72ST20tQQTE1O@h@$&z(C3-FrEdJh
zC%f72;@%sT$}qj&lFFoLVn53~;~#5MSxI!gR+{GqF6S(I30yGQHr~2P@bBr>o+m0W
z4maMUZn@Wkf%>GreHGd|RFHP8s!!GsVJfU)lArD<So_d(_s57wn<v_R<*-gWX48y+
z+)liA?*Phqnp3)K0DXew*hGOe0j!PJUapnBC3<IsN6j$zTn8#(@OzK#YS&lDpt$ZC
zupWBs;c;y)-6XE@NNK`8<;JjKKzpn9$1T3XO^!hC0qsJk`bI8e83~4<$C8(zy*k0p
ze)Xp_m2GHiW5poKTBu^9=dN~H0W9xEmj$Q5p6bhP?`Y39HT@d3%XH#AOa=9^Bm%m~
zO-A~Z`pmTLJ|W`QLhe?_T%uZ{UpLQ)S^(lRm2`VPK!J>Cq`pEGb@URL9aL_pA`<63
zZxOA&QAeoLf^X~~x@Y784+#B`(>6ACGoBp$kz?oAl-w<E-yC<jXIh1$=bw>mVvQ*O
zWF*+ZTSGh%Ikl!Wi~1Iyo-D9Tj8Na2rB%+d@8?M9(M15^q_Rh#S@;<a!H`tdlV)Fg
z3f+#(Oy$6Jgcd%NezvXat}DqRG;gFg3Cd<xx;UV>Woc^=E_`^DPl<<nm%xrgCYqvd
zVPfO~RDNd0yvu9BG?~sTiSB_7E(b`#yo1$b<}@?~%pCE{oc-JkbNj$I;0s*lDUq4C
zfq#~cLQfy6FapC=MeeT~q;NxmUrUb-@B+%&UB_SIG!kcJ4i~&;K5zyt!vkC#uOUGM
z(`CV9D*mcFWGje*6Cb|c?7$Q(;V?I99Yhp?4%t4TOl<O9n}xSz-eLzI_dSRGh?X3H
zXRV}bc7wZ7{PmZAB$3lZiG+zs%{w%*EU%i~;zc7QmbxKj@?J?n!F_NMm2dSV8`BN*
zKkz)as{5*Ipnd&Cyv0)22Py#8B6zlzdZk+Bmw$SfJUB<ob^Qy@0Maz0XiH2wg)v~~
z*HbTJdnC7~<a=iq%df#c`Fm4;VZwHudKh(3K6!Vh!*oHi@nClHh>K$)DJ6Tkn2Tkl
zC?-v*2cT1XvX4{LjCUAS!8^O2O{_^tz;{<Vzq)iueYWoinF<F;Yob4Z$Bay5M!wKe
z-f(grnbnDIJFqZ0myJq2%XVT(q{_?xspjMQp^cmMuZpc<hXVn!OpV+2_NP!WQ5j4(
zPL%-f#DT1>d;L`q=j$6`XJIaSjwmJa6mkv}Li?hRXG)PlB}zf<`sp<$ue@;(7~r}7
zVZQd$i(ecK#M<o&x(QIk%Af1m*a-CkvTv<HtESnT3gAUlGO$Hh=CHPRY{Of}If`0d
z{N|=SVAB@pD6bC<-mtL9M~EgG@gp>LZJ$=6PY=hjiGY@a#a~syZ!HtE-f09#=;4Na
zPr$m^O%5~QDP@{C!8o@0n9>zPL-9nmypun#S4%%+prq&)e~K;-LIe=q`tAzy`DNhr
zH194GSGPa41a}hFBuP0SwAM|7qbQ9pDmGW&y*HYu5d@s8GEI$#d<kF_Yf*4vkUQLo
zmcKh^>f-WohL0vGKZ`C8e8zcUQ^+ZYk~<4?s4LeKsvcRQsCVyPhwvDq^W<Zs<#FmG
z7GS!kWT(<`Ef0_f;X7}8%eT3)O}|xQX|(Ajts>ZbJ-i&~v>*uJRJ7W#oA)X98VBxR
zjAU~MCs+x~jCq%Y<J=Gr{?5xeXRdRAWeKn9`XGv-K|{lEocZ_PDMke`tWkWGsluty
z<7)7jNJ}_bAfpoVK`%hsG+u(RvtlgR-CVXOQP@*7lWt_{Pnqs|c%MY1W*<{3sHyR3
zbzBNwJ~yQU+bmQbR#wxqvtx?+53eidg*dTlMcca5p0b^OisFI@8>dR2_gTI6KV|4;
z>fcblhlfHU5OCX67m|1)zo$FslX2>i<mTqzz=#mp5jmw=pQuKSI|WnaDw2W))`oDq
zYaKcm`>T-pl&8BM$~`joFTBuoEAOq>rg>1@v?J#S-Q2sIzZoeF7M)jmUtis@ab0-0
zA2k52L;6YTKi)e80dM^cffONXB!AQ<UiJjN_SU3{`&6k<7+NcTHRT)QQ>`fOH$(DY
z79gl*a}l$Dy32wkoB6F`I|kfPACZn5H{2jnWH8ZocJuCENo)yUD9}FYfojg|$)n?2
zcKUmFx%&U2?5)G9+_tFkje&uRji}g&fS?G7fP^9v(xD(FQi6n1B5i>|7${PL(ygR4
zih@YDfP^634N~7&oOACv_xt_vd%p8L*XL4uzx#dHnrqH6#~9Nl+2y`_GWUG5ia&UX
z7cgdE_td2sx8GzK1(!+Xee&M6B|EuxO1~>Jh(%DLI>ztANM|wFuj;HTW}g1><HoI9
zVegR_DVKd(i`!TnrPjycsTM+7#%<Cs_J4$S)#CxjODB#}tv+@suL&nMk#WtZTM_Cm
zv1T(ic7rhxx@@HXUTs!R+47>ACl-OCzdpnJ(U!ySul+g;L4hZlq#)K8YrPk)d$2=u
zp#Wn*&O>Ea^thu4Y<PB9eSkj3akKJJHdwB$9Y1$z3oVFaIHJ(}^g+9+ei*oH2W2BO
zD+^3by9Cu19^kEHBoi`@+!!BD)Ln!%3Rt&S@4h0AtRfE#)j=wc$&CkvVpQ%Rz==NI
zs$JV(4_^<Knz}x#ZCP2Qz==`Ic-3lHMLcr<Gn&h0m7CVCGZ=IG<&Ku4Hg*N3?py(S
zj05onoK<1BBiiCP5ABp^^8&ihz>VJb&kp(Q+3l8=Hd7Q*>P!V?t>!UnU>%QK=mKH&
zhc15in^Y)<HaiD82Ri<FG!Y*bS&8vWhiw|Tnk-V))6o5dwD17~2Ec*hLCW^A&QeCS
zwol(48->BdsPxpJEvxKx`W%iwq{&HSeQ-!3Mc>KH-ug=)Z)g+UVc$s;%MWOk+UFS4
zqtgM+q+@wgA5P2ZRVzr-2^`V4mz+RO`jL3+7@Yickgt*vinzCE#eECP6$bd>zsFl9
z(*QxYE<-k1Q|o>W2Fu7V8wMp>xR*}Qc5Fd98qR<^dU`LfI>18jt=zBA#eqDYFvBBW
z<ET#}!3Alu8y8nBh#h|?-*7X&r(gaAtJ1u;al^-F5CD|NFA=JmZ(01^@EMXAIF=Kl
z+*gZW+vdO{x|*zhPw?i<Pn;NC`q~|_Ec#O1al2CgsU|P34-UykjBUR?6&3#b720yZ
zepe%NA#5%FtQq<611CpRR~%iN$h##WlKEox^S2_-e`nrpIh|-wO1(bCQE^;szn~?u
z4=mXSrS7|OhQd~Bzu#3Ck2)zJHSYU@*8cwv60~t8%86?pg6PTNY#;AwoO0k1dB}CM
z<c-F5x7+Uj+x}X0zVqMnZqJ51{?Gba%0{Q}XC>~>{ilO<_TiN5{|s{*S@XY!r~e;m
z)Bo>}uTXzPKjJUR8u?s*0cp5l4;Nprlcfm4N0C`EthK%U9>#kjttE`hC?#fS7(XLI
zai&Z3umGpVF=YXDk%kX6fQYwWs;b5WQl$QQu{k<JF9nw14coz82Wce@fDC#<C<Q@M
zI0**P(i~g#u#oN&B@6%reWk?88f1+iMgBr?$NtGgtMp%2cj|>3UpZ8#m>+l&1r%W<
z<JZX$COYCfi4FLbOwmqCB7=J9Ych$2(^<skb<IO)6b%J&|LeAfNV}}3f(s3C@zut-
z8<fLB@}so5;GyDt!}K>ol}>Y*O)*7vH1iE*4HkYliUc^80uJ=R*$CDpJbD?Uj=^0H
z1PWMe-vvmETNDmX7kmtRze0Pp#{ImAuI?e2Zs$S34jDyn3fcJ}`}`Ul^Z?HY$^vX!
zh$*8kd4e1RScw7>BAF|Ui1xyN`3<zwK?ngZNc7vN1V3YZ_^MT_5cBFmKzuj$2#gSv
zMBt2}8oYp$6y0ngKx1%8b3sgDR*n}nf)Lhj3W(Cz=4MT^Q=Aa_(}Bo>N$ml+Q=vT*
z@{*f2Z$68Z1C*CfpuxlJq8xMP(L4;NYXr{}l{D7<RS*x!SUSMNFhwVBtgsTPec}X%
zR$@0G?o7@0?LS<ACBsLM?nzP#ApH;@A79D9@es{_9w)8KcM2kwvh(*h0vX^L*j>-e
z$;nw(jZbf>2f*$kRz?(npeUShM!>3j!vh@^aH1$UuU);mJX+^(-pDrKebR4EtFmdN
zE5Y4Fg_n<y1s6!j9Z%*Vi`q(PD9T_h)jZsEuo{d(F9pc}Atz13UdDelxQE*|Z+?em
z@(%7AaTIQJdo%=qdn<z$Tm91Y!GlP^@lj~<9T8SWW0fug?5ULzXpKgqc^`#48I3^&
z$<Xk1bazLP$)os|)d&$jfeDMW{T%Xu3v8O1DpoK&ChrRSqm>b8`Q}VC>@?RvEggnO
zWmFz?N0R|r9g#~JVeJ_@IzSm0Pf-Q|mm9HviHEasV-TPZC7d8N^RrV{AHZ{xcLqlA
zY!uT%zz4kq#f1?DYe@!IM3f26O2L)lNT`>jxVRk5_{cXw$c{##8Abkwh<gOeD*_;g
zS?{<U|2LM07QO2Mk<CQabJ5udTQ{l=a9T8_1Je^&SrtMaE>nVXXW}Y!9!lsvdYuoJ
z$1=Dm$!zt15$oIFobZ-hE>ch{y@nNv<g!tRPP7p{79pFQ6%m>E4mZMZP(KKPWBfbP
zh$xB`>NO2jg*CZ^!gEM+96WTygKa^m;qW?#w8X_9YW9N(y)}rEtW&Vz<z|uiCkm1i
z-4!BKM!4CM%%OPUZ9y288iQ$)eTfdU)+-}0u<wvjJ9_o_0T_#{eT2EA1mTtvCa`H{
zVywuLyzC`b2MFm@QDC?{>U{$WOq6`EOFD+oefqzauY}7ws%%6x3bu`;awg^+Q}DT@
zlVX93*_MGwH#{*R%KQ}(it+=~wJItqU^wE86Owo)HRWN16hPY>=zlbA)Nwz*To^4g
z2z+Xyb5IB3GR1EFM6!jyK98kIe5A;Wn@`^~s)>-l2hBv^OK_b|Rm?yLw*@R>1iw(a
z#xx6b>ogRtj*7(7<brxdWQlO&=>otUf<`WJZa~~-y6sk-DTo0Bl-B6@?8@tTNKeVx
zjQ=2XVT@{A@(^y{8vhcpoyfv|0~xaYfE8zeDh|sluhn{o!oSzcA(8D=6{S<~E_fPc
zWAf9laE~*V-%)v=lXt8^7;V*iPJIQ2@r8$@l16X}q<5jMC4Ryb=_yfRp{sX@x2_Qj
z4t*@?$y3a?IcHWa%APdb@&rK_b1xqw9(AV}g1ZI$U}bHb#%!%lXh%9))MVf@B&#Fi
z<G-=(j-r=PKgRugA`%mv5==fqF`Y1xIaqT2B{<Gx@Cwr5LO-5k$oj*8qbfyTl=fDs
z1vAp6`tQ+6*aIW`3!kpI0-E%lkQaOs73{rFl{{HkS)<SCkqI299sX<;_71^b+lvMT
zE*5Z3ogJiiuaD@Xhe;;<Lczv)v+W$_ct@x)d;~rK#r=NnaIct<?IO1^dvD#-@idXj
z0U625@kIdPWw3sJq?+x;-VaC9vvBY_mMl<K(hAZQXwdYDzfh9AZ1wu}U+e0;Q4YQb
zWfB6(4Z`?4VwFjJ`V{#U)c*H*Y<AYv)I`e<y!{PWL}3?JGa8`#mRqEJr)3--ROhl<
zlOF9h8I|C7t3%g-ET6O{h;*(%2L{Ay%bq<RPo6w!o~)j{HkX=I*N0<G!W1UKlF9;x
zS?p-85_f(a1}G-2$wNSv>7yI7KVBB`9!aaX6~DrT;EHk#1q9pgEa)15sq}G_q4(yB
z0tC)Cz^45^4Tzg${c|eseZJ0Tc{yfx;5W!PPJju4SB-IeVy))l_yjlLaDyYrQ0Me;
za;zjZErlFN&S3EzzU@}m_3^?Z>uwVq*qa#_+0{P=^+is{fnfLmM8ccEEkUV%n<~u=
ziOjjEUWp<rD5*B1)Fi`%$&_FiKUU!I*DVt`s1;a!9O3|chbWfN03#U-h6^p?b46YH
zZ&K1G@*H`fGhzj**a8G8dTyhw5R=GmO>|mx#HlQbPd?UJl2=;FE?<mYEic*gMod)n
z1IipiIuM{nqWuI44hUVy3<TR*P?cL?<3=YWl#`kQe5n|V{04v_&||EKQF<^y^z=zX
z*ObpiqQ{z#&ZgpM*0LX2Nzxd+k!>U30F2C(jOyhWPbX#_z@+iFYp{G^V}d;bph5&u
z_IW>*k8d5xR4BwMIQ>`2WzzgdtpPwTyw|^(3&<UkevF&8d1umKZK!onq(NK!e;-zn
z3LKvkvBJm%0i;+yycUO{RywSfVvJ_wFgSEKp}GLE?Htf@1Q}T+*Ym-@a0U@u6EvzY
zTD1T|0iw%B(Rc`dMgjE6i2?=~jTqW3WN<c)s!qf^q8S7n1arNS+LM3{aH7)?nk_10
zymA*_6b3vbX@$QD`aA1(opi%;8O5-SZrik#D_6?LYYG4$5kpS6Ffjq*1qHwladIa8
z1q3TM%#tEo3uJDE<kb(~*6Cp<^ux9Y?To&QjYu!JNC3nhCwV7$cRg*2vwsLf3~#*!
z*%MB=9?WP!K1bFPILhD##$G6;2UYnpW-`gc#22m%(bsQbJtNc|1|M-rd+@)Zbd{*y
zd*o!1_|z=ulD_0LkRjqS1cN9^Z}wOWyN)B{hJQPvY?XJ0IsTnLNOF`Au3kn)_ALI$
z82T91yLhMln68=JrZX#e)|wV?V?Rn4_w(tg<BZZ$XDq_g#C#ufmM9OhZ73F~+}*LK
zrT8Ze-NU<UM`z~uPc3{@uN5sxm>L^tm3J)dvM=p&vl~ns8lxvtH6u){VPR&LFblxb
zI^yKyM8c{b((j)?e=1>u3E$1O`!Zzc(aH!Adb3WTjje<w`w|;H;r)9#1SBIMj4Zso
zs$jjIe~Fo;K7M}DNOM_3y~xaAq|5S<(Gp@R$>y4}9n`4E-D@xh7_u!0?!^-+k*bq0
zvgi$g06=qQdin^CzL#KUfK<XtRC|#~Sy_dVZL=1&2R5~v&rRPX7brO0x;#r~77$hk
z1_(jvQiHhYr2w1W-@kRaA*NS`6jBd^Q{cF!3|Xl%Cbr;;Y7719XDW^&84Nf%06t3;
z{$CX4Z9oPuS>MnQ1-3vW%HL@CMRk4L<Kp$NLqt|PL@8>)ckv47O`X{I7-%JVsfY_3
zMH0^+xKS7==|dqK%pL!(R1pOid=^qTC|~<j(SgUk(8lq|*Mp&`SHR`Lm|0J}T~=_z
zKLpE%PArB|6UXEM<Uz(M16TyALRN5-e`el(fdxbmO^Djy(Ct`P^LO`;gs#Qon<;bN
zM=*OD1a@7ts)#xRvE~ZgBY<ZR<Q9b4q4dZ9MCP`=-CCT~LH|t=?{EBk<HY}bJ;br*
z-;Nq1d+e_3Py{`m*r|!q{5R=eJ))VInCzgVCqX0p&o}os>ummaP247MnCk!a$3yDq
zs)3NXj{fwt#wA8EwCFsu0vNEV?PJZOE=JzP7ZMT@W)b?B2izd^-#5?PwV7gn4=1Ec
zR)B_!6Z_1VV%BADGE*LU-)1YT2OHyw7#t&wdo?bJkh?cT<|WU>T%h~z&FX5AZ4bA}
zigaaGF!^-kcriK<CcwL)CL2Z7+|cL;p6?9|1P!o%K(p=pcq-n7p-E^_-LEBHCO~RT
z#_!~S^ls{rV_RVT+hL{D(8#Z*hu!t6EZy?0dVNAenXq4NEbRKV(r2LDCK~xSjFL}V
z>gfX?Z-aL#9c*ePjt!x+iGc_wxvy96%Sq(KXK23^z5!>ZP3%0~qA6Bq*@@2Q8pzrY
z9^XeX!?kr5iacuj+EX35#h}bzg2%~vsAjk`dv7z>i(BavQ(j2>-e%eMgfKEOZE=`l
zhl1@_M&S#xEcvbbF<Wbbp+;f#2w#|t&t32=ab!ByDHF-gmzI7cG>3yU_~B==da&{V
zXs<V?N*NHR&!Lc@N9#SSa&C_}F>{)ZFx5bh@GDcGJrQyfV|s#zmkl=cH>J{|FGWET
ze|zWNKvQ1WPwVFWnfQI{W<?l$Ux~?yCzuMi0uax;u9SVv(HPS&!E!#AUbMNbog6~!
zk#WIpD8VZub;MACT!RJh;GR9|Io9q_>LmhrB{Gq8>%OoOA)E4j5T1De>X7z5`L(pv
z1HG{kRM4<VHA*=zN)k?rmjbbnLQI1%=~-+uB+4j1T;LRD-Z832pjiD452r6dw!X3z
ze;TGfGiqB@Qc|3tT>5z>O*a_iNF9lzd)M3yuU+T5zS=lOqrQtN^K+a;rQXT&TXQ}P
zx1V|qC3_|M;2@K8HR)lw-nZ<_2Lp$m$NFg1fddPj4Otq6g@uH-LPPW9md3XI&AzC;
zNR&enI1(34^sEnWUFm+h{}3bN2Y5e7dNT)0FD5zI!9Ty~bwx0|N7P=GeF2WqK`g5@
zQrtj*Fsx#d_5DKT451iJI<_nCb6yqOK@N^hs8kW!SIfR+LN6Safa{#R!2)=NF2sp@
zdh_t>M_<H@N&0(|TS&;E&ws{0;cMiBa$&}*0|}=4_uJ1^h!Jq773O>x6M!&CoJtW$
z6R%I1ngp%50%oVZfaMV~h@pP8BR6{gL(KAugrdvuq*;Dt<!0yMY==nY#KXklMzrV^
zENSK@Hn>dJJ(B2=s87+z*v<2W%r96@y^f@ZO?H#r>S6#@b<ro4+9C9Y@q)Ke)%vtc
zIgvxIsb|tDizS3O0B>swve8=AFt4k~BC03G@@l?5u$Pfhijli>X8Ed7a3=yXai5Q>
zzjsb|bll3U{u)0e`#jig%F<76@2MNkwXv7wKgz-NtWZQkRyKXbDhqU)vN8cnVNQ=f
z%S+bwi$RnbfB)e~m1ihMp208VMP2w+(Z_VG`KbL3e*{sL&W!eO6+uaT8uu&&^9oLN
z6}ifOee4G1BW$U%26p<rzR~+zLeQ|;|E{U&N4lfp_x?=4kM~N7=Ma#jYfVwx7-84r
z$3(kla4^ZZ;;xp{yfRY@O6(7qumea2$;v&n7vw{3Xpi#qE5;frV<(1On#7KsMgUt&
z$8sM5OetP70EwZFuCCX>A1df>?)TP2Ju54Gbl?F^72~PbYOs7(k+Ck-05FUfCLX;#
zuSm%2ALzIAxVOLVoGmDRxj}c9lZF94^9^u7FP8XXUiv+(PNT7=^xIqP`1#2^C@g+9
zwWQpvD`yZNM}R{oAXg1A=P~`^jL2w_v-k$d9H!-75M*M)<$dYezaK=v`zW^%A@93p
z@S9?YR!hVt1DU|u=yv3q)9=J+QObp8?Lp0O9x4|ZL6JCDcuhQStRNO3gsOm7UWm+I
zK@QrH1#H>1XtMAL4O09A0tk}}DDO|0cvPJ*{rT}z*f()vSul2QbC9gcjykiS-)d=p
z4jq>5gg2*@58pO0ZA@&D)I$nS-T!j>7(a9X;AV#!T#Jw?3MW>!m4KKD!mwp-c?4E6
zdd;Bqo^QNNN+|o<WY#<v+PXChR1MeX5WX6c@OhV)r)mkA4AbVq{oGN@;VZ&9G=8cc
zQ_6F8U%z~zWY%U4<WKaGUYBUu8{0F#b904Tm8Yyt9G4H(n6=IRGI_S$c_{bEmvgN(
z>yIWMvzu62*BIG?#>Z8C)6!>o<c@4Weyv^BoYg=x-}hF8z=qZVeB*nMQvjE-FN~~$
z`0@~H?R%~po(Y96&O&|HU6I9)_RReUuI=c0>~)&^G`H9B3d)RgrVU?K?X{<58rA<E
z8Y=5i48PWx!iJ{Dkx;CTRa=;(T7T9j9+kFAgki=(G@g+qp)v3k{VV3c%B>h9-9F7@
zbPg10Sg8n@uB0(X!#O+P)H4UWsm*DIFJ!7`!IkF%M~DQQ_wonH^?_~^emhZrH#&{z
zoO24q?W&CrNBfGfUeObHURLh|l<qjJjeWUHrK(e)GM>KHw?1BQ4zpnnLU@URMHnqo
zj54if+s?<Rx&W3g+lOXK{=x$BdzZbOlnZxUyNU`a|KoQsdq#iu=P?Ya(tP`tQ1&6k
ze9B~|xnW_H23`VZnM1vWGP>B|kTO3z`N6tmCwPP*iuu@m_2`G8unI-}N3FG`hjeqT
z=I5Z0Wg&QMHn$PfvCC`Y)e5j>$S`751CCZPB_&kTV{M|Sd<y6U>_(Lr7k++p48Rkt
zoJcH_ldOQ{2t`V<v;<gAaR7!Wp@KoaQ#`X^hXEbu6%;lDatsFv*nZ?I9cjbuzqe|v
zkSXv+XTNH(V#+OyiUK3xAxQ8VD0cASHZw4kcs`-4N|D}`@4i<St<Ck>dw`XGIL?Fg
zRFKTh1@#ux=Rsqw<s=iFo+2LG1P(%yeTtv|dKDZ)KcTM*@07BHT^+C~k(aFa)}rON
zeb=rhp`oFZ^@msy0Q3Q`+gH9@iS_g0<UGnt<ibWy{6|7z#j^+Hl8Tzzu7v0|p?w@2
z&xYS02X6As+1N6+1SW*-`JS@3RKW-!j#x0tn04!KZx5blcvXu~w^F#`DrU!}nzm4j
zr$#4)`6JoWRh1oCkl00$5*H^PmOyBqvtF8PNPWq6$)g7L-WfQ93@GSm7*?&zx|H(>
z=Yr@mXQxEKv{-(doKch-RpHG_aD%!>>kT04Q1!!HhI611Lh#!TucdZBvuvs>_!^x6
zV&xD(GmkT$fIozt0Pu%Y*fuIVnnN#;(YXOMNqSGV6M-$&{45#YK>-qZuKD!sX-xR0
z7t+3a(#*R|FfU$)0aHMWVf0I+Iuu~v1ZfK08eJDtVfb}>Z8$BN9S<0suVSd#!-L3+
zfIq=+Iz(D4)tXH$_+8Q&77A>RQ1s?Q)tLOr*Ug{T?iCj0aFZ7^`Ywi54)*rs%&^ce
z>zqxqph}vULkE%YL)a9TIS?-vF5_}%tLt?}%biJ+5MUXZ+<bk~*rd7i(<h}1$}Po*
zn3y<W`i%;zK2wk8t3vi~(jN>l%h5w^$Tzc|_ZH$A)9Isb7(dQoTf8G+jCkU1-m+x|
ze9`snJO|;_Ni-qkkQCI`ZWu=jSOG_6r0ZqG4`Xbss<s2w{@ih=?8(;`(YvK(xfEuf
zX6{BWaCsM{`nM3`twHG<M&KP3FLp(j62Frl(78P<<u4A5fr5L|0U#nqBhdwsfgCW)
z%HiLR3#d=L6+daa5(6SS^Wd<gKYar`%@xefGj;b|T~}o@Yd8h?`(Fd!+ND5ZDbEl8
z)nYHwc+zC`y43`{jMwC9&L{wMM)ot>fM35H8VkJat4a6Z^ywXZx~oY|k2DD`u!7?g
zdd1xpwz1KbJGpM`1E5c?o=KxhdJtd&5qU?%swE2<7kRN2fiFEGJE3!NF4Y0h5Ao(i
zjF1Tb8n7Qg#b-qA9lTdp;DnNzn{1RC6N1stJ^0M^>g8k!ELU8)XTaEm3dE;uO&uEP
z`M5A6jw1{FTvWLeeYGR1-s}ypg7H+zxHphAGcWe#!jPA8&58fsuc=<1k@?{iamp|o
zv^>1HS2Yrfk~dMQSPw2)*hXAVX35H)Q^>GamgM;c_tYD};{vSu9LG1|Q=T(vids+c
z_3^Vf#^*~!`ztbxYNJF;E+;1~HiK}2_!0UcC#ShM?>4k);1i<uSmM(EW!^#zflGqv
zt7wBz1mlC5sb`ss?vmp;apEkqsCVm?fS@Wsw9|mXA~mzpHLL2-9jt77A$#O`r@m(N
zMYe#}y}~F|?_6pUPdtHqh#>v*Q+NR4GBU)eo}!**t8B(l;(j0*%A;jBIBQPc3SE)M
zC4HA0pbVUeO%k}DG^8aOph(^TJN>sOr^74Q4&5h-n!kUHH0StFZWz2Ey&(Doc;UsB
zK$gk^w=hv*VPTK&(q%ZE$l$BQI1Q^xP@;Q&L~g-jp#h8sH!|6bVqk69gnhrENH55+
zG?WQAU5Ls9^MQk`iM3AhCn1WMm>48lx6wRftMcb8x%a8bi~cpV8l^z!N5;nTfH4el
z!nAHVbX_IQ0(RFH3Yi+l`!#8oVzF5?yf)NQOc_5~7(mz?BYn%#QWwrPd}Va><TZzT
z>1&IN-8LZsfPxR~`PZoG(HzmK8*FMYs82$-<ulHWX>dYrvdIY>_;iYmg8hX2Rr|;A
zuivEselM70i09(6<;!04UVl~UsXju0<1L3BX4%jg3z4V_<7GLR8XRoBdLQZp>%oT5
zc?R!Vywf$g9Km5Dn;4G(z6zex>RmhIF30X9xyMA5G9Heo!F&xSf5%@=d%U=(B+CSS
zJiueWfUiN3aJA#{wxi(m9(1K8H3BT?*0I~+H>3jiw&7jw@>!OL?Xrz`+|oj8_GAU6
z<u0+UWS6GM1%r=AJaB?GFj&6J_7%k4=Pr8xvzjUf$96d@m)B+P>k3#pMm2HLv6$4X
zm=YtiZ%|z_jVyf{$uKlzxlP`8f%PIaGswSTuV75p55)_T+}VETKX5s8am$a6Q|nt<
zTh|({jee%7S>#1)dQ~#Dr<ja8F@fz&tXctV4I)!z6cnB}^T9R9R%ggC{psPb?@j=o
zg;NdFT?{0hkFfflsH9$ORc<W=%8hL0TGcv%TU2-M#K71UOCFJz$Lwb-fGpw=-*BwP
zUwum?N?xp5hSQ+UO5f)9_C0*X8VZttamejLX2bIGngSG=WQjN;fO6#y7i|Isls%F(
z<M8VZQM(@Z!pkCT<}@&|C!2lNZLnvQmOGVEcaK0h3ynm72nrjhz*l}wkC2h$V3xjO
zS0z<04%RS0kv^~9(r^EUrf>v`!z^1t!-~5AqAFl^4-c>s=*GPU|GdbaC!$E=Qjw(>
zw{q=!bY8Hyt#+;=_GKWXRbWsI_4?i7z?hTD_}X#|q3$BCWsjJ%s?Fblo{vuY&6)4p
zh&y`?yT<G7o9=*Mtf>iqP%^a*Sj;KA!QDc|eg(kOUJE*;Po@eZrubl}C4txBoQ%bM
z)%faX1a;NXDp}+d1{r>tHaYeciA*SW!i<uQ_&6Srk(H@NUr{{0hU#4_%*goVw$QD3
zb$t*?G0gXaZjS<R`Q)PY|0D-ycLD+!>J(R+QznNR#ZUtwOjXh>d|J%;^m=T5%J$QG
z_v$;3AjBempP6YE_3#S;<g!Y0<?-y^7OlqQNeRg`XcgaG9B81gRjs<)i!7MS@aGPs
z-GDe|g3R?m7$zf#F=A$YUtM@cHcq9%1&p7%QeJ#Aes*}o@X9`-TtvEle`??f0B*1{
zPCL%M-cEn69?>UJIi+(hr0!5Ni(kp_!2E*c^m|_<nK5H!VM*l!(#NZk9tJ)c5m`wx
zH-z^ZnOsY=oqc9*%4r+@PQp`m>>_*cobOm%&#QOOD$*nzI7=UfdBDy9_s_s(rgBin
zI(`)a&E{^w$MZf6&d}+u&ob~VRmmku?fP9@SB8VmR7a_98LABBJX#PsuOjdtF2G^E
zquCc@)%x5uY6Oowge8OdmBj_iI@}W$Kx<_BSNz1@Oxu%Yufn<>oJWhF96o?1f1@Mf
z<1u@**-VnTGi+|b*jGZtD~8ZdXBUA9eaVqONVu@S4vK-JSyk8vr%q1CE5GyI$q|bh
z35)3hN({2N!C@wi9GHrro4U*yyu}ONY{6&d|D2$na#F%uR!Wzl$4VxU0^+(d^hTh5
z14R(GI4;luRpnNcHg5F%7mr=k`IhxZ+CeCj4G^kdb%g2*I2&Q<<tIfduA!l=Q;jo}
z=;-N@K-HI^td9-`DFG#!p{6TBo4|*7{^U$ALow`o@_HD9&?jasLBT)W!AFby`SjKT
z@PT6HLj0PK?g(0H;l;wQnvHYJ@<4eZ5*WSjuk^5s`OdsUIh<oX<D08>_Uv8F_hU#Z
z>RAsm4tt+0e<{!F@+JeZup-^+I!Oh=0F6}6IJqS3ER_lG`S7+QOz(w}l*jnByP=T=
zZ<jxm<-BLvFvn{c7O<(&tNtreJBuVwa>i_=pBebUL2Q48g`G7weg?P$`ov$hBaPlY
zw3$HnsIGXGo`8l5cwbUtL|C$I4U3lo?}moo(mfp`hdM$dr_K1_qK4jeDB_G+zo-qc
zzMkqdj6!g4?=oowNXS?<2jq<uON5{}hbCLYBO$nx7ozD*T<}Tik2Q>goH(nE&dpdz
z>X&nC&fTK7)q3LL5oY+}@EoB3;y0XP{JIw}yZnr9sh-Mfe4b^f@MRatK@mJlz~-NB
zSzHKDEQVpJP0jpC0UzltCc3)!;b$~HSmnyioLC>D0s4Ntm$|2IqI$_b-x|fk3gt*B
zI+3E$*Y0PNUYvqxyPJ(j%YZRedqhtiR+s3}R72|^Yx`ALk=YH5Lfb%@?yF?X(hHXf
zEY+lE=JB}D<FyH;R)E(?9~ZVHg^F2ctLcxA&p<)E#~ChL^-k|Xz(rSHW*%;Up6WDP
zTTv?$jy+qQWe%H9*!6KOZ+63i7g7T#{8p6vM6_Go4tINU8$NuN8J<G+({cuxqGV@2
z?3Wb3?K68ipv$+qNb9uI8$({-{b~qxVW~$k=MKb!_TFCK!E~Z<axS$gie$j7D-gY|
zxpqkO*-NHpWY~pSZ5`q%vGTGt|K9%tVGN=xJ!HN;?3W~|7F(7ZAxJ8Xj*vd1Pp7AB
zVf@Hw#|1bh!(`86GWBb|*GR8onJwGGF2L!;1K($MeyWFE$RO8Y08JjQ$*C@!%EoSP
z#mA>txitOtm<~|o=Ex(}C9*73I9*Tn2v{!Lh=@&1WTdt{*~!;y<=)xcL^FJ4=2|1v
zF&vIRPi2`zcrE&dtoQ`1<{=7sWEa0sdGXzbG*l`LbQAry8Y!^=>(6fJ^BAFLLZ9?o
zvILW<&@I35r<xh+xY$HvyurHN$DObDNPT~%$j-d?V^ZRorp%l73>Cw#Vgkne`;Pni
z{az?r<(cNOdmHXtOH_c701h@%wL8s?xRxf+SB3X=@QnZb6s<pk4cdV<trEC-1<29!
z*S-BN7HwyflN@pI^)o|(KISw{Wy5{=tP%40qZ$e4m<9kvsAt+dBaNMk$a0#GY4Y$M
zA+2WS5?<)=g8+AUF>72NN_?f^tN$zBLYg#x+FA+=3RYp$Sr7b-<f%}(diC8&VR^Qi
z5MxYk?Hq?8r@#@EAfec5F`8MvsEWJ2g6U;5k=@!h%1TONzM2zp?aN!!XsVu5`W|zO
zwOX=I%Z`ayH2+K^*OSokUGLQIC(~x-T!MwhDLrVC$prF+Iuz2jV!Q}G_$qvqGiE=B
z0HY?@*Poyatl0D5$<|*$b^*I;4$aJBd>A}?wyLSAnH+qMIW=;*H7|inRTkPI8$w)i
ziH?qL{(ZM?-auRX6R`AduyL$cbGakmFl;{<$z<gQ@hv`z?T9-fa6S%wmF%kzn$Lw8
z(woPZ7Ea~0NDw%ec!vY0I`!U32wO75;QO?0vQb?aayl}|IET-w8;LpT$U=@Bdf`>;
z;9%VP%?Q`8=0JmoJ}R-Rl-zO9;6GZ=^P%X4C%F_qe>8}AkbZ`VO$~Ax{UEr}Pl3}P
zK^8!Vf(h+QgVt7SDk=zB{mMknt{$kN3Ec`TstX&J=#QY@=RiwQj-^QT-C^lBZcf4H
zNfv~isE}LYP6V_#j}>9ZXmUN7=Pf7(#AzLvFz2;aVdCYj_1znY$%`G(lF*d~8sM2!
zM``fqrAE?|{xZa0MAJy7T{l?u4xc*tO7sbxVg}%Y7_E2*i<hvY-129#ub}nJy=wrO
z8ThP`T0OucxUuk)7o<U6yUCc8-6CZH#Q*Ld^#)zo^Wp;3&D6d9U->m$nxEYQ>O3lX
zMn1lspf$ev@|6tE!PVy*Xk_%3{D9YL2|zyHo1=%|^7Av>>IonnfX8HnTY|;Tbth&r
zhDZrOdJNS)RrbT5R{g}F1xBIyFJSe7-EZlU!H}`UqAdNT<wWLEs;H_}v6d`<Omfm@
zSrCMXMpPS9<`d7a=d&7g1KE;jJy7z&A^#0d+yH~O*mzquZTj?0J~*jt9jGu@!Qn<$
z^%__nsN8E)@2(`1>t?40^04iJs)nY{!;^BDEKxgIF2e>dlyIaI#Nsqe1u0AZ&(C*T
zoMzp@mTGTpCEW3rxlfxGS?CmZGG0qR6~?rGKNao<3X+cjl{%5me!H}2cjec?tt79`
z8w*;YyeR;{ArY@)4HK{q1h;j<a3oU6GTMf|ap>8Z(P<u2q7ZRfm|=o==>YI9i5E=Q
z3#j8U3lBIe@%|#MTXcLZ;FAuLPbpMmDsXs7*GYUb)q+KFttM*QKtoFCvm7BL7YE>M
zBmL|d&}&b`r@qWnGw6+9YYbbDTGK#JGA{sBv=3m$0xl)wm1H*Us$O(I2zK2RZ817O
zeCs)-mH#-hBEcBV0C_--u-qQ-OO31wlkd4ltCD#Q^`qN9US4|S;UJcDZ`mFeu=@)V
zfp5}&gnL9JSm@Xhi2-p=tc{Vlb}gb*%^#15z<5ZCfMAqHt11Qf1KJ73Q4Y-Jyk9D=
z3j`4Y)qc+V{H^czxfl8Fs!28+kWR;_X1o$vMZ0;|mKS=Ia?o)DGORg}yn+Yt73Pib
zP7F88aU36PnsqNdr4M~o6#8SQu24X3(m^-J3!A&%qI0-HY%e*T@Q@2&U-Ozn$K?~1
zFv1Z;yM(DypJ70Yw&q*-L=nms(7pui`9H?o?E^3?M|NH8$M4;<X%l=|6v);(e;)GQ
zWw}5Yl5}qDFdyV1_K<k^z)%Bz^`HIw@Fvq!2o#1^=TWuvt%%UrZ$2CrVVcn*C452)
z<6E?78NQeA7kTLE8Gb*U$ssS#5Ig-<K`G%zz#Amc3`{sc@O$BO*xbAane>e9kK@2(
zE!0y$6$VdE7wkK*mr!ajF15(pghFlw3SxZH*t|b)@%(}v1{7dkXntd(Uk|5#$n4P-
ze_r@FU<>v)C%WP++>E`AW*Yp+!muWZng^9)_sbEqyjp7#mCvI;Ha!v^NS-{op<p23
z=PO{Cb~nRvKSbeX=F}rCWB6@Y$tH3aYBroEDKmf)eP~eT{`AEY4J0@j$O@)`_^s(J
zKjX)P0enYQ)4foPB;5G4?4()47cgnyH5vr)4<1Y+2IilUv3@X}x6OBFz}u=#V;K|k
zJ^k3rIG|et+yV|S&>L096apfA4Qp>ytXI{Pw&FRF$6cY5Uye49W!(jbc_WwK>DB`~
z6|-mA^GvLIDp!Ijo%`O-N+xIT!iMuK(fSv7$EVGF=gzDiDVqcePwY>?9jSlysd3v<
z5;dg}3>U?{ThF!b>R);AON%xfm2iy*n?{-*Oo#8{)L+LFr)8l3{i)I7NneT6Ei34k
z%aL4}M6P!r3Y+&<Gn0w$q!;@Mi|ujXoh=VmkI2HAK~2d^J%>4j=I<dbXEQ~KX<$2(
z)Kk;cHZ7bhKB(A7Z-2KLKRf3~;wn0{MIjtZf6*5Eoc$kFz^`Uqx-0)nx1c?!^Kbga
z{sh3<OG*4+|6~?4U#7eA*L5yk?3WBCw*L&IA-`hdDM9(q2O-Ji)aw=hk3SyPiU69o
zgP#7Gm)AHxe&vX)kPyOl!MAU#wX|AttTN2{K9Xs*_>~W!@9mDw!MOrE227`%-W6Nd
zcoFUo__EoGd+7@C&m0W_C{JOO5-_u1U`HtHVd{Jr$1_-uL9qNl=o`|wB>!;AB{h`?
zQI83;-gqy!;KakuSA>`3?pTcAQ^n$Z5p0!6jBR(Hwvl6sMc2DJUTZPdbcxGHpW!&q
zM~Hw>dtfYI8(jGDhV}2Y@15x&XV<ap>19Cjz*0V<6tR&IY7uVvX2|1(Pqm8^WJ*4!
zV$5H~_Bz~$Xcg_~<2FuTXQ3U2A5DaOFf|VBJXn|FtRKPBHV3_Xw>Ll3j`+B!b&XPb
zpVmwe)*ia}cktl=;6N($26nomz{+>m$<NJ_L1G{V6FLG6T(QN&@M7pr)VJmzhB){g
zeun#6f#O<fOn*u<_5+N2q3p#Bcw&}=6NQNM?)6TE`D9kb=D60zsEQB+b}~d5m^tDc
zLYXuQU06wyNyPmP$uzXIp10K1_Akr>r2v1hWaQ7^dM;i~C(st8e7PQh2G{Y?tr+p&
zA6#wX36H-`?m<J;fwgf9*44RMA`Aqlj!Y-PHDWx{bAY4?aPSAf^%ZPTBHVh_@iW`O
z4;3JOV?B7f_h1JKy^%IVtwczVbWYpMj8+)b+2LNE^SG%K-(enJW~l~#q&CP=yg$@N
zve&F#dlg6_^kNklP=;qkI*|y)I*2I5?E{~2r&1^GDH)cYGG<^U-~l}<jC7C{%7Eei
zAt^~0l?_f7KfEEKfb3Kf8gYE+?j8!Vzd!#i5A>NwaExL!WiDW*HMF!{$lnR)SWORO
z9v<R@7_i3gpE7cWT>*SYui)Y5W8m{A($58V28Uh;QbZC>#AX*8ylixyKY#9(n3#Ct
z<jIQ&X_abk06yDJ_D51`0Hf|UD`;#~qhjIU7_o|CO&A7~Yj^Av*;DYHcm6Y+N4r=%
zuu$cBL8c~NOAD5s20q8a>&OI=twD!n{9gx85LMGir*fRSDwWX^mkgPSGSGV#G*7_e
zV=gi$qd>3*i7djhvcQc#7%HtARN6^~aN?13mbq+=0rjHrMU22_XBM6qYH4Xb@nu#U
zw{)!>wqW^fL1`J8i|F}Wa`ru;_xOw+UOoy%))6qE^b8Fn;idE3a3LnH0JK$R9Ei%;
z<LIGBOQQ}^z~WPK1}D<*4C+o-Viiusa_G=GCIR%zZ`;@$5r#AWfg*F@Vo4oX2O?Cx
zRkA`Q$8+{i_2MXmiLp5|QnGD(_r^;@gK~#OF31`0r;tLJG(F4as71xZEYP(8KfKCh
z5;Wp3KmlG0*rhb@`UKr9dh`zwG{6M_#kRGlrx0lY#;JU(ox}}(0<&>whQ9(r2lssA
zjve2S$c&E;{`jyQyho5?zz#+Y^#E}Mr>cklfc;}E>&k2gJAZ$FD3IX%>&kt-_C{+i
z74C0$U3{R-0yrIH$~5LP!Tx~oRd7!^kb7l?HDOfGCkO(-Z$;R0b#+|@zR*^-%U+n2
zxCxT<b{(W2cM(~zDc)jpVJTO}s__hrq9u-LPzEraa6^88Fr*Zx?t?5lyLRNn?@U)0
z7u2z2Z2An=KT7>|U{}Jty%N|E+_$4%EP}~~(YULjcLkisjM#=c;WxfXT~kr+pat4G
z__1F<e_jSt0}Qr+IMb!s#tIy_+;86=0`exD7lam9s5q^yt%0B3gfAS%`k+&qS5mSb
z@*40x^WYj{`CWpa`bw{Pyhg@(Kz=vP%|F9U1H@o5b17h=YS;4PIR4<&`3Aq-_WL(v
z2rHpjvBcm`<Xp>cfJKbk_#?$+7dlHd);TbDGp0vhtr+uOp}v72+A};nJY}x`%@iO8
zMU-?D+{kdnvT7a5U}GvR_@h7yUJvpT!#n()JrscrTq1&Y<mj7r0`Rb8-wwaf#xb#A
zl;gk|MgY^U%Xna@j`?%wW>yd?JJczO#<|W^NM5lfBBbrm3Z!^=cu?0|LgNZ7CyX>e
zdzj;z#>he%{0Q)O9ewdR`RL{Nnd<7E!!D<sa`=;UM$ZG)CvhN#+6Zl>(_fwj^xecB
zMQgkgiX?>iq#h*0U@@<A#`D;vBsStGCn*n87}D!{VQOY33%!VQzeWxoDV$Lr;2}d4
z9bl*_GE{@+BqMGo=8b7%3rd4+#{>6-F4W{O!7xxQAXKQFGq`NREk`}RFx^f|reS!K
z;hZRT1Cr3SR{&pG#{bkA;93a&qRgT=9AJvphA<R^X+c#c5DUD9Z<HdWf!IpoAl(xu
zd=xDDu$;j8abcGio>>UE@m@#d?}87FqpSeH4sig)7>X+h%@D-ow$4L(6OQbjFfc$*
zDn|(d^aKVTO-XyV!sN$f0dxd~;PdP8W@2bd5T6_XDDbDc19Ihi@a%Dz1(%rV>*u33
zHpJin=vwCx9$)+V5-lvD`Vra(Y<|owrMM>k3}y@whjUzIFzy_Mye}5<i`TEk(SRBr
zA3p@N4D^i(Y%FBc2~C!#xJAk`!aSfHCp}D*_{)$Gr<k>rqOu^#2~MCY>IZP`{*2Ix
zuyg`1eGm@-ijFaG;Po;*Tn1;6+jWDMjB6tg6&ngEgc#slC<44<>jPnGOtvMo4SDGP
zl;DT-7~y5okn|SB>#axRmgK>}#5Zo=PIeIU#h4vL_>o%byR@l*rFX+(3&eLYkWRqq
z2w5-~F3HWy?Tp9`gA;XIu%}?9;BF1Hn9=r3fEN*{VWJQ~VSvD)BS#)0CF@ynnA?iD
zAOTM!&evepY&_ndNsDS6QSdlCiy^_9`Tat+GAkY<x9&kKOW6O!;0+L-<*Q>F>08OL
z9V!>6Ns_~jEF!FMGIR}Lb-sT6T9-Zo{#GE)%vY?9*W7$rRki0d=k<{Fl>CB%Ep&9Q
zFJ35yo$<!+CSv|7%7F7&Qzv3y*ers+52l0HU;?g@YKLXOwe$s-`}gIrM2?ObblFzB
zIl$x7Buj6v{V)~K9)cs+#|7IsNxeS4KTpp^JqNZ#g_9@I)PeRw9#aAahHB72*ob+|
z_wMCo1b!I_NaDBml=eR*PxU=N|2*-a#!8LCHA~g;IxS@WL1nItDt-KwVLIIsM4gnR
z!3?9%2()g9IdKglu~n@QOsP-6a^29S39^fo5u`Kz+W|+@=){CFGOOb@IWHB6CGuK1
zh)4N);Fd0j19Pn$^eeStFnJQEGD3D^s|T~y%Tz|thBR;?{9F3X9)^$);j{zBMCeYx
zMto!z-o|!WDcmPbK{8f7O`#@ELyb-Sqa4#bVDs;vKc7uYOX~?SNR5tu+XYZso$5Q1
ziTJ|zfq{WC7F{?)n1u-eKfL~Cul`tl&obk0*+|C~U<!T&$Mi(b;=C-=A|l4NAPaTy
z_>&AH%MukNQRvA~MPYOB^&tExI1}wbk|CNSYvquL*RtM%2!gK<rr5GnQ8+#ul5l=7
z3zKoDen2w>muB{4+eaoYU@t3UK8G?0dnZfk>s8}vhEeTo3xbn#HI9L}#NzU?xcaUn
z4QBxX0p(t`X&lbmf?xs~{9*_kneR0<%Jud2hVd&!gqC9XH^Lia6<)bwg{}_|yuhi9
zkkLh%wPcg=I#d{_pfWQv6N!M^_Z#rXMQ4nZQiOt{LU0I!qasmOMghv~HEBnTA`4(v
z$_j+Eo@GZXConNVIm^}xJ-Qe&T|Cx`OT!5_v>LAktsqIeeyl`4RR?N59KLBM`VaR&
zVnCSy(Wrm$jZK+Gj-v+Aq~eSvuXV4sW?h<>Jlz$M!5Z4Vf=bS0(q2Q3Q%=<7<fs#v
z6y|JhZG91rM<x7y3(;PdRF3U`CI^ZDP@u<k5FJ9KYN$_OOhZR(^a184YF>yS%@H!S
z4W4Nr2IgU#5T{oY)+D5!1|Lg+Z^06M3!K_ZK~KZ(_m~7wZcl0qe7%K%Vg1}9<rBKF
zHO!=dg6c&p-p|yUIO3FhhIzAI^1yEF?a)fMWCouEJ6|Nc6u(!|>51G??U+@aI$<6W
zax_wOoX{)R2_5zow<pdjzCO!|Z6j(XGMR*ejMoMu(NW+)Ib?xsAIzNff||SnLl`+=
zEA5h%nMpyGMol4I3JStsg`qXJSTWM*X9IjY6NYdW)Nb(bO?i14$`Vd{&H;#S>)Mil
zb>~KpJ=3(ks9{9G=?OYwtjy(xcCSqUBhbPs!-ux4oS&Qffxh*t#37@&a9Y7>;rP(D
zKF4OxcRdLoP>}?T<1#J!YT1ZRCu_U;pZcW|CSL|df|#iTuYMF<yhU@q^%rp}7J-!V
z`ndKo@a%(-Sob$&&><Ve8i_86H>AQ@coy09C*X~+Rai?*E?}~Y*%j9UADl^z?f~Z9
zJly<vO2XXy1bL7|iGfy2gF$Bu0j81P;sg`Y<u!b`zfMP8eK(GIdg~X}#(kvO7Zq-k
zgL3sDaL2zTR1i)&1b6H*kIx3DZn$?!AKNoRIeI%FPhIR9D9XHnr_xvd8J2#*o@;gn
zi3r9yE?=B&StQy+*bEUrE+Wo@4ulLR$9DBdO-;29KxjGv_zgk+^<Z_MkF-H-Pdf_s
z8v!?qD<O+ri{M8LxUt7KZr=QvR%nj+@{+Mj7-rCOQoAG$niQISdF+oPmv2Gc@e8C^
z3UW|=6@A+~Ptif_2{pXy8+lLu>AcRnXPQmFly9|>JP=SHS+ZFjSg6Gn?uVJW1o&s2
zpe5ZmKVLXVmww)QN+8jsp}ziCXJ<Z;B@&tQ!@}rYKb4fQHI4VLP{$@6_f-iOkfEoC
zFCPwwugJ;4?^8%M)j&ZC-^t7~B`KWgZ#28c-aJD!O*#S4QRQI~03V~)bQfE_*fKw5
z^;ecC-hzx0HRb_`4viaA+|h0_NGnBl>U`-~B4Cbq2h|1yeIR}_Aok4w3QNv2N1WcY
zW5+{K8DwT~Hc!tkf$O-<$;Za9NlOrTRI=^UEa$h;V-^l0II5W7a2!)9$iV2>3G>Q{
zx~F{{i3X<9_k3?P2PkA(fbSBH#E!tks4wK8%IC>mk67ZHi7ezG3yWm<FL8Wxcydx7
zXecK7VMLD)k7ivAK~hG?$KliP0`IyES;flp9=i)c@-u|01b>`|S_ilsW?#8tJ`n~P
zLff_th$)h|*Ib61k)V`X^7{r!LHEg-HDh#H_9emxGA@F`BPgM&Ci&rKaF~%oo_E{v
zcD>h*vt#)jS!aHoc&zy!E&vyfo&eJT9G7KK{+&3=IiR91oO0L8WzP$@-ZjTD8xEtf
z&Y?-{gRO;K(?Df`i9?DetyOXe67}^?C28VKe`F6%T**=8&~4#t)Yoq;IeOy{pP88^
z)9WTo+k{N61b4U|a}NP=uGCzqg~cuj5<tp#i6L9trZPKyKFjW0)X8MH50q>~IZWgn
zkYmds(zj0h^}x$Ry{=+M5;i`z?R&^UP}(IV$3gKz!3YLh0Y)T4J@tv<*3UU90JtoR
z3c(;CsUHwr2wW;Lt%P_$U(#~Gfyp**dv5hxQXi2hj0KA{_-#s^>_r4XqQi-ci}Ud)
z4AbtOxMpTnh9xK=Df#)!IU(Im;L<*O_Dr*kZKNduiVm>o384W`K`Bb9C)T*&sa^76
zGo)YRw(hEzFK>o*4~!0eBpP^v9(4<PVPIN}n2_d!{`*ZhH>1Nuq87LYM^)ZCBjN6j
z2~Kl95u7%{%g=9MH@X2p7vO_$Umk5q1CmPE5J*^%N!<aP9;NuYW+BA-)|ys7L$iE1
z3HQ|UBBvGG;-SmPRB%WD+g11T{WXvfCt^SYrczkGLay8OvA<4W$fsh(uS&l&e5g~^
z^me2V`IuBa9avCA$^HN|RhaXlwjhq+WUv}YvRBN^0!>pV`)&NXvVffsV1^7-g##)K
ziO5NV#P;RMvc`pe(iL6YhK~dy8EIAlu<|*5+dTc#V>j5rt`N`J_5wyfD0pnk%IPbT
zBH%W$Tr;bf3yyk-f_x%@*IaTQq2Y)W1f}&=yzc<tlPThdX!zs(ew4PmPu(EaaK`bt
z9V*b;m15gK?Nf8R^pYgjDh-qaci{bjSjahT3p2`77y{IaNC$DUltCjd3`OL6;AwTX
zzm6?5;Vu&1-FnB7=UYHmpxJr^?#`+%v5MSTNnBICIeu<b)6|#+=m*@X$352jRhaVA
zFh-BCZ6sx~8igv$Ky`bN`|j)d7bGuO=#)G}e+<Vbkj;(g!eGp=Psd44mQ~`ZqUsH*
zzGzty8vq>39o1;4&cH8R5{tm2mnz;#iVG%p;<%N_I;&`$F~33!)Jg6FEBmIgH$F)3
zm)`)@EK3iG>bHyZE`xUZjhW~trQ{FHG9g(ZLxf?RqSd(hczEC@n~bieii%CZM;vE5
z=+IWoLpFsy;caa%Y%?1ufQQtUgJzNss5uv7RL)@HH)=r6Bh#zmUA|GF?0Dat09=Nc
zjxcQKKp8-eBYJc()L!XN`{AenuuO*LJ$&>CBJ;Im%<F8|&=74deh<vbeW=ecwzqAb
zcq(94k|$skoXgyJ{Y9Z>8|YI@CzO%<1E1h^oI6DZ1fx^So4$v;tPt#~>c&Zca{z$E
zZ@F9<Y1z1G(-|cEm|1%TJTo)yCB_1{R*|44A==CGu)mv$^|y|Q&LK3!_h{c%!D40I
zauFmwgbM=K=xcrv+mFam%J*7aCnvkO`T5u=3Ypl42m<O~HT{ttn(iwvC0&yz>U*|u
zc%7gi7LH%(4WtM1BR<+=m=f2Cjaa{vR%+~1xf^fTO&m~`8Qj#=)XU<V_l4HLy9pD5
zik=-5OZaJOyN>s;52kw20zD#tD$a`QRe^Aj0HBp@CdJ7wM1LF4l^#UO)|&epBk%EI
zWi}KtC0y;!CP^N6*A~OAXNYiK@uD^L=Y?=mCxgp~lJ~5ckACK}>Oj(7euF0LKy8rA
zn_DF*33hB`FmC8p6e~J-oJ+`xiBV=r*&9*-YLSReT6ZWV5GxH{&6q*+2SQWQ@-NG)
zqm*f6Wo40D<|7(gjyhpn4uv>W+tgI)LmCFlImOwt3hf}oHgvs0i-))n>+0&JJGD4W
zF;Jl1IDmxN%v@k(!H&>q)l$e{Q;4Z_c{TNrS7FtbQ~Uf<_I%|Q$AO>)An*}tU2H$G
zx^}q9JjOYPM623x<t+9vaTueZ$=N*qcGKcA&}G8WuEYaz1_6YlXn*j!M+UlIp4xHG
zEhf|N*%QY(IfwdQ_5Bqp2==O@FC8nGxXn<xa(X=sr*`4Qxr(?zfDdFGZ9P4Yz#+u4
zAs7tdMS%L}I5#3TpmG~^pm9rf0NU%7c$QuYgt37{oOC?#m#{_U1VCU#zk<YkbT0@Z
zhSjoZ=T1rN0T^5yAkzr3r?8Z%;?oTs$1B7LFb_b3h(CyNJ<(<XRU>ncOHXaU(hQb(
zaTu9MBGL${sT|lTlF1eL0*_V%fCO=Bf{hF@&qprU{d+HjCOCkAbNi!=LQZ?;>eT~S
zzobzKUbq~3@4EwNL_8$)2|pTnGS2pJ^b|1cY9n}0z&+6V%>&&Pxa0}rAIFhg7r}q*
z^#EV?;O#OgFy+W}bG!rWINZrYhC@C=XEGaiRq6Tp!;`3=Kao5J*q0#2alqfC9%?Ou
zgu=D~ZhU-21KHPsg%F=K<hV;`92y@<NxyLuj;ep$2n-3X0dGsRu+!j)Oom^;FC+x~
z1fN%C-M69eIYJSd8(vPpQ54N39{!h+$b5rO2r-@|fTPU9vxdgGrDJ9&gFGEfZU#?|
z2m#RR5$Dmu;!kFaJ}mpD7(M#eK&|oOD>VhKqauGwQchrL81snUwE<c`gg1iF+9&nx
zxK92$<Q7C5OOi06+=iVhKtmA5q>(2j3!TM{_BSQVgF4gJvG9neO%94V5AaKIKT6hV
zRT76?0EQGqwMe<Q%Yc3~>@oKJ5L%@J3qBasmrm$)Pjt7nEsdf8zI8+`C1x_U7#DR4
z+&XP-ZDe^h?-}TZ9-x8q{KbnxeC&JF>m_-B%Fz-U)(H;-_;y}yTt5ZCH#ho?dE`(R
zl0!&n=<VY-AUtBpkn9PMtO`@u;Wd%0(a9w3>lORxUGrMT=8ptOL9ZtgADI=KkQCDB
zGQ7m=|LKj6uf`^_8bykLNgLrThz7*%fLN#C><a~K7xPBYAqrb68c32dC=YONx@~92
zDvTMEy<Pwt_jC*m#iLRP268}V#IIeqPKoUM=ypW2mq@&cHyasm562lfG^Z3{+E-pk
zhcRN5nk2(2<AGTK`${-EUdG18CZwcXB;#;#94MizlXL!qG~Q}d$Z<|(J^hJue>)bi
zf~SXJV4~c&M|HF2$3{lvosm#fHOA@JgGmgG5k>`={Cn`=#CLe8C?dmDLg`2_ncKH-
zM*@bA0H4aLvzXqb32Q0V>-+Rvh+g;YvKS9vJuIZEFfeS-!N6ZFEj^tDU`z*COVsOj
z1fd7ug$ebeBO`@Lj=c4^V4Cpop&ags0vh2gC`+PH<n#jnCR4chx{k54%TV>ylCgwq
z<!}(zNC;VWKV7B-nED^crZhpl1QN{ehE<5dVH9rc&j5R5ipuCn$b%<BuA0nLOEzoK
zWDs&xkC#DCPKJ3Q6O)>UtzJW!w2jS2O@?&>Lz2uj$=5IGUB4bphSniz4E9nWW3}`w
zX#fC*FZ053gcX%IkLUiSWZ>*v4nP#veMLDqi?7vHoQ{@$b6Pb%6O54v92vx3%P6p(
z^utI>;5`ND4a<=u;^cB>=QR)xCW!7TdXRLroO@exoIIwq<}y$f9WA<EmzO8vr!_Wb
zY+&FWJBmlE4Cl0jmpsvZXfe7+p>0E89#vJ<NHk$0fv_7y8;>|*pj_GLpMf7Af=@b~
zA6aK$&~>zkz<;HDNYile)!-78QO+-oflE$S#(P!IWG%4oV4k2g^ElX?M&b|lgOtI#
z5=<<4ffgYXP8+qRLw|L&Z31!RJ!x%iH7-kH`6Bccp9|wU9b-+s&StWj9e6X6{gi_u
z8s?>dxxKt<4%K9=A{j(NOIs(gC^R|9AJ;kWxqQ0sTC|Lv?1OLJ=T&*=>u;9yiyx`~
z{>?+_;kob42mKC3>xN!1?|XjJeKb+5l6_}j%ii%UhH$IRp|g&Hd%AAM4^55D+qbk>
z@pU-}9uyMVi$%DSvNPr-;V|5v>%wiPhH)>M>}}<G_Te`q)6piyC(dTO9+`1lT()!h
zQ?DLCaOi36t#i#j3jES$W|R@F+IzreKIi9C!={N~$UqH`8&w~{c|Oi`PI`$WAzm_q
z!LKnI8LxId@p-Bxs;#}hy1JT$%aLzCxBJqo_b!u)s)ms9*T8@~%1EMgKfQ+*??K8-
zLfz)z_|UuI+0vzwe|`%+*zEitx&Ypz`YIRozJJjil6Npt{xQs09r@Mq$Dw^a^z=ee
zM&#4}{W?b6*_rk)Yed=p_jiv_iU0FE;%w*F2><(UI3y{X|M^<siTtvs|GljHPcOUu
zU)SYDIdsbQ_v)?F_K>AM%tXRy_BOXGFXd9ad67(@Aj4XF!FO?*u~M%WdAGBvoxM{Q
zKX}|oYWH84_%w=`hN)au5x!GafE4&#^8(fmT<4D#`|`NDr1SZ{_PzQ(fB`1R2b5AQ
z!e!)I$1XS^^2!9w;i<gW8`Fn_0TuP1Kjl?5qDWDRQ4Vgs<X+DGi+XJ}48kTg$rtF*
z)<S1*wbR$HcM95-Z;6LimRH|Lwh9Q~T=PXq$+f85;H40mJIdzA6P)HlPiRnV*I@(*
zj4mi7WW0&P$+N%T7*BXRf=78}WmK)o_e!+myZibUljdJqYdTu*=B@`A_w>!14gCBW
zy&L|r-h3)qSW@yJ)tjz1UW=BOH%+nzDb8Bgis1cFae^tm{0}v?%|bstI#PFlAZUZL
zvt#J@mzS~^E{K4t0f*R-&3)HFj+<IUeRDRjIg3yUxJ+u-f&RJzJsx}dA_w~$!?{%I
z?vL+gX4;2dW%`}-xL}8l(7U#AH@Cg=1IK<%wm1dj;3nV|@Zp>N(>GzNsB}w3s08;g
zC_4DwgD2ayk(bsmcszyaXb|$lkCrnha56>n`|rG0dSS+PqJIr9FYj(X`;X}&e=V;N
z1)O%992ZU_`+Vl(gOETQL$f*;i6Co|gQKJ1P9CHCNl9E#{!x|q@^sZg?ep%sH+$>Y
zjJTBKUDQLA+#laC<#_EhD|_6}KbR%xrpjq>wKeDrX@}OmcHwY&3l0O|=HbbaPaz7=
zAKkA$JjWtssYutlmo(WZ&PcbHQBq9n6#{<ES%M>+kHR3D-%SB;weZx|!ra_ZJWHs^
zlkZrv=$SKAZ7jog{(dOFF=#+POt+Q}dhq55hJYQv(JBlW^U}VVRTR`1$GN!985yzQ
z2P;zKmEfK*FDWT`V+SWSa?(xR5(Vue$8VJHqvTeFgh+ms`Q}8=NQovxAIXGNbTf2*
zy%UAm*H^3tLd#T0C2W7pzkC}Vyqv+rL?zYC6F5kFW4;ey23X@gcolZoGAUUb#`DOE
zazW(Bp25~xPeBEHe_5a>1hzB-L}q`sPjcsSG}$-u-|BZpB6U3K5^b-M(Y32aj~}tK
z+l>BRb7`=V111o=T-deL_}SIp3kNk<=2K4FO&$f>PP5#>AtEXwH$+?H6dzw3dVb3~
zWL)l(z4xlbZ_~DI4>Ser&Wfq+`eSR`T_3y(gzXDC$F+KTdU5hEr<7xo&RZMIi@L&d
zZDTL}&a=v*E8V4}j;GvL@vs*#XiPQ918`d&<mdoPkh4chk$9=7s-Dr1UO%yJJ;$ed
zdLJJjBFe<N(GH5I)nH)wG1>dzKL4#1lOqE^>_FVW?&d&fCD9SvjC?xZojbw%lz!B1
z4e(Lub$8oceQCe5A{q>reSNzzp$GlMOTWexDt9=e4}#TncGdOk*9?u$;!W9#OE^Hp
z3<(KY?G*q$6(STkPkV9egdu(ndK<Ulh0A*Kc3}BrW)|f~b|;Sh%3B47Bwc6DJhc@!
z_qFSWZ|@~@gCQU<C$~@TBscdq@EM8J*wV@}C^mHE1sWO~8=HNstTYr@ORvylWYje<
zAPL&GJyXw@g630Sebh^2%vfmZC3*QzquW@I=hY}*)3CNa*_`b_!^p^p6|<a@kdW}y
z+q-~~La$>pHOMffI;#eFMkT|V4a~*+wTzDQ?3Z)S+><Pj{pPgIol4~^hIG37`|n0M
zE`0g&&V%PYzQ9`g19-ncX;dQSCQTYi5{|xj-tEsTz7-W69RDpL<#U>TYN+X6*=PqP
zd)@m#Vu6C+%F0}yJfYdiZR+L(`ok4|{@3%9yC`_g^zR-z6`vmfC=f1?2{>NR!~kH#
zW_)<R_2AD(NKg;_DHD1N^eL$+ratj)XskO%2*K!K4;+%2$2^OGoVE%YY=AV2143)e
zhio-H3yNg3Ta~u<YJP{EC~8+w#Kk3-QIJp)ELjzeexSBCbJHB1=^4Rb?mx`3iY^}M
zg#z<|a&|W6O1R*{p(ncwWemj`QY|UVn|Cl|TD|FKP*|7?8eLasQ%o?iTBJKt5nL6#
zzN9#1y3%M6m^R3Grz|b4R<_}3pKi`{paC3sWO(LB(rr$_ozDUTJwan}v%ZmwLgxl1
z1w)9w7pSq==cbIO&!4k!SQo}mWZNw4Chjm0Q~v7jcN-tKgv0ICUN7dSFJFEQiqCnB
zr}-8!4!1iGP7v}rXO@SAhwGp;A`2J6Fz%(1YFFr?4Z&guxwy90GVsBccNOvg?|Ahy
zC@@K`al!kT2x{kjU_zMLi_7F^&WzzMKLS(==sSPjZfc{&n0{Kj?!|G{ndy;(#(cN(
z7Z;|4Nb~;UKw~@FUZ-x(o$4$JAhF~M01`^J{d9L->M;~jWU<HfjiJzL&Cy@~E?y&V
zX7~nx!Lyeo*Mb*CNZokO0kgtOeo$w>IcezYA4YcGnr6vtZf@S1VYDrkeWG2AB5Z5R
z2U47SqCx3C9v(VMdCYAK#7{aj(L!rA<m8-F&6-=!iLb_ZJa?4RSI}7#nsi<r`QQkZ
z$^#rhA58i#Qk=nFrZ_{q@#$8FWPg3uE}VZI&-dqgZ`S}_O$6PR8xT&KZmdFCPqyJn
z;~x|z0|P)3v}nP6D=hrv;=YRm9h*3hK>^#KOJ#2;&V0(JD!!qA6%-o!06BQDOn}p_
z$0*Xap<h^lpnE}z;_K&k5OW(q5tNoHM_r>ZH8MG%25s&}UtiycnKtr45;g|Ac8w3!
z(GFU4?mK;88Rg@O3e@;3Hann_CWkdhz@RG-;~<48Lmi!a0LYe6Ncqobub`;-tFssQ
zfggczTKU%RK2AaHdLxi?72Sb@RC9j6`H?~biE{4L*H@o;cM9%TPj5`ID=8^aDAIo1
zB0nM(Ns@W-&!5lr_4E{Cb4dE~r>w`1padPkI79CpcJf)E2z*BDOY0n>>^pYM^NoOQ
z>+}a5^<~3D(>7e!n|B{P@3!%y0lF1;&L2LEJ>0t4IW)lg`F$6c16X;aX#{oQ$B*av
zPvz$3QXHc%L9G~g!G<o@2|GMaRLr8?nGi;el(xpHXS-#rS#zx+Mm2JFYlo`pVEa9|
zA|dXro@0T6=6<%*+?}Uem^Q-s65VeCYJn4(mkYyc_I=K*{P`7|+@+;&V&+7q?aXEX
zeD94~_E4~4FLpZol|6rRdX(^db6rsw7Qn}i^o0{1%W_K^?2s7ZA!~ndUFG%i<;X4j
z(_p4}{K|9Le=U!i0Cy3lEiYL=7(=Dw{O;X>&ayz-t~Fcs9XvR;#l!7s1JY;ZZbV|D
zj&p@0a1a(6g`%OMfz*dMnW4A8EYbB1uLU*on)ka-t6ZTzNZ2<mAC}`F`stSTjE-Fx
z+9nXkw*Kfs=gtEHm-VP0lx>ebS5UB79Ap3hML`*Bnr7v?;r(FwTgl0ro#;a>%+aH_
zxP2$>JA9sqOYwU^wt4FBzfM_M`Hodf$|x&eOT1=wjM7mgU*|7g+-}?0(AHk374bMC
zD)KE@MIdH=1xoieEG(Zf8L-s;r*o{OVAWw{qgH-`oPxq<M7sNN!EijGv<~?3bmX&O
z1yHpZYgx`6d&1z9voK!~uX1T<ZqEL9SC`27^VI3-=_IU<j*jjdN#@oO4LIevn{vFl
zP$HtHdHQC1YpN}`jO&9n$u<mzW@U_vA3sV<-GKjBru{@z@rs}$W#5r#?SnP?&GrJC
zYTURx6&p5uL=Oc+UCKcwic9Xu%DDB+*O%qSUKqcvM{X9~ZMc*r&isPbRHZ}5gq4|j
z&D^x~!w7Uvo<4haCnQ{!7?;kO^{XIjbOvBa5mpmh1+s+;TIspB3IUw9d1gCQL>tJG
z6B*FW=HB&s+J<Y;3COft+=JAKaOK_zIBftWQ*zn!z{5C(kzpub4#&ButEU&8R66>9
zXnXT$F59+k_(VuFt2C-ahGb5u5J{oIlzB?#sFcX8A|z5MijaAhP|8%M$ULQzDMJ(@
zA=A4Zb=|}BtZ#kmTkBol^T)I9`@Syxe&=}%`@ZelaEdl#iFMnn^+koL3%-6AFCT{0
zI10w?cc@njcDgSZIQ<Mx1=Z0M%toQ$z7Ec_PrEG6P2FB5Y1|J=;llKrQ&Q`rWEn3H
zykANFW6ITW+n1K7oA8FI$fmS7Vwi(ff{2YBcc@*7x6;3~j^Dk1Un5a-xILelg5~>4
zHs&8XnCV{+3)?U`@-$C;XkcJ5f3i;QZlE{x{*2Vwp%!}TOK<=EjmbFI>cqJ71Pa<E
z_wHpwrBZ}>o5CjJCy|lTUlw&<cbli7M0vXMEQSW8Pn=i=6n(9>*{Rod_mVp8k_WQx
zhJ<*bj>e9b-l{xh-39=GhGJQ|R0!w?RKjm?N?#soS^7OSwLVdG=fKym!RuF(*Hlrv
z$F^kgVL3T<i*O)kx+fWBc;8fgd^od!-%y;RAprA1Ht{oOB9oIz!SeHaYkCO=SJt<%
zueL%>MdW#2y1Wher9fAiitdl>R)nY@Ol@4ihR80J8jf7dLtJt-^^ZhE3V-UaDaF?O
z=LPu^B?e$pEQ>=p_yZq2s7STKc#3t+nu{>v-+_5c_T|diuseY~X5R|FMcc;CzTP%2
zc2zAHXa_wu;JNmv^g!-_CNPe2PK!F#Y=YUU+oYuLDiu4$eQ9|HSzX}N&{nyX2n5Ns
zo<XBCP?u?(*z2o3xQ55l?)2S;RaPB2*H){`t-gYjkQlJT)y)fB?G2*lMeTgBb}7X?
zm@j)FCB4u&=!z6<GPjF}tVTn&>Zqoob2)kTDy%x~DiJ=>y=|p;qc-lbIgG0}6C6IA
zxZU9RPz9Hje~tY*0vwEQvp9L^N7c0lzw$Kc^g4-Ie0ZiX#Iaxv_8_?wH8Y*})jZzM
zSe|KI)qZ5citg5!wk7~FqtCO`?=y_(K<`#yk{)Y>nCvk3zPWsKu}RMgd|^AcrtRrf
z5jJBSn}=TJSSt$T+VR0unwh#Y6V|CvTfv=W^EKTk<os$QIp34%`dgIqSiqDl0uR<t
z=+jH5b>tzku#|i`b}zE1zX!XIPgmm@azQ8xQyb1S0fj!+;F-WF1D#umiOIyUSIXH<
zN*k>elswxwIXxTAY8Kya#jn)-n%7$pkqkJ{V|R@l);SIh<-xOD!5_@%9|w152FGxJ
z)t!y#{^2knWk|ti&>d)W7jW3U>sMXF%8GI;bbY>ZwQMR4!8lYnSdaV5AGKSue0h1a
zCv$zi{Bm;pq?-BBP)`e7-SUz80PMgE=pYk|fO<?q?n&j7>9{GnfF&#r@OH=fC*3At
zHyDZ5scHQUrs<XBk#QPnG!F(D+z8safB!~|59k8QWs81Kyw7*at+VQhJ5FNkqpmXs
zqM@hEo1P0%Tf1sDM@hg32b{ed;I&9=X{|?DL{K7eo%o)KqJMD#R4%%?HO}QGc)pKc
zb=M`>VxqL?M_r7Cs?uX2I_xNVYSWCwvQLxdvlG<|pXX*c(+WXuaX6sw(^#mH9c@z~
zByJP*97DTFsRLd~YOg*{<-Yxs$OAYpIPK%lAs3yWH}jr@XR!@O9#?5z?`cn8Nj_&u
z_74&g5O$a06unXTdG15=OJ$xP7==aQzK6_|{y@mCL%49TwlqBD|3W+!=}*v3+S_#W
zNy^h16f8gBJhcLiT#r`^56<Chbj8I~(L~JEAG7<`3TWRzgN`W6leE3LHf`7NO`-sm
zw0=}LuD2QQuM|^J&+fKo7OTg`#}n@_+ZHU%RK!c52^7IkKv^JA7ZjX3e}b~iHb3xj
z9d;$j6Qd#_qrVdpQij59(cLv5MUryqQhPfvl4V;0(9v1pQVn$y;a|0OEtlwL)M*=_
z{roXi&k+@(k5iI+(gC2Hp{1qo8qQZ;PtzCH2t3)C{9x+7SL};@`%H$*kt~5`e(&Yc
zttu)ii&|hrx)vw0EC!&ST#Kjm<udzOZ>8gTk~$uu&x;EWZ=A)c4h$;&u1&4ft*1|g
zV`5`}^i^;4ka`LN;sYTUi798v$&u}xD5_YGOFx~qN^b3j0_<n3Qs!IvBpJEKYdWh^
zs{Mq{%}a2Z1>!C~*O9Z7s&m64DME8*Xl{6_TT?=={IuHndbdjoM_&_4){PZoeB=cJ
zHF%+cF*3CkU7JlCuYo}tj2f3=;$2r-=A{$&B>5J^u^jyTILlKyMO_I7BV<>}`b<!~
z9IY8i1SMKgs;)mkstz$1INGCvV>$b3AKrxGI2yebNO#v!HSJkMe{ljVady;Y?{n<<
z@E3@9ahLux5{;!P=q8Enp8@&P<I81>6AvaZUs!{r8D@8Zp7LN59He_)zp>?5xBD`!
z<|4ZxqAJ9w<y|OY?1F>CWWmFOIo1YTNhJV{UmWB5WLiURV2rv}u44~CXV6*y<|YYH
z2dlU8c0D77w?wS5QCJANd=O<}7ox?DzW8(DW5pVmhtGM^z`d+%s&j|DVfgdQn;5Fe
z9MRrT8zEIb?=rI$b(`oL%Zo=Q25Bg?$4lX>KVWWU3`9w$7xi`L_wP#~_q_bQfxajF
zQDk)VH9!v;2)*Fu;W0AFl#*1B{oES&QokxB$}C+U{o`RA)|nq_m#@~-twR~IcmICb
zjv@+NuVUf|r>W_UZ>@H9my|-dgq@+8*-Z>Y41@9yGJv~<z571jY8}LcQRqZ~y%m3~
zVy0tQb{G~-3I_!`2J4?NGA!G0<B7a~B(MQCQS%kRC@%;Jufl}&HKNkJeC~0nfBoS-
zmID6=Ml8+JLG|R(;r9JNRJMt5peo%Y4Yx)Z+|_Q@xDSyJ>R^_ftT&@8IqC+itQwsV
zSk&=3zpkaeD=*K=!r~A>P6+(UZQ0qIf`j=dLwEiGU}Ly&Ja{aGiOE#m3YY;`$3Ha7
zlqqH5%BFD)PNOjGBo_koXBT0`di2<_V#I=?wQ!$HP}bE#qTIK(m+M~Ie<!XwzeYL&
zATl?%uqeLxO%nshLmC?Q`O@(#6RcV@(C}@Na(%>?j$Rf+{nyrg@3iYT-@Ehp_A2*|
zuTUL5M=BOZU6tVnlff!7M~WQKr;DgUHSGsT_T<3e;Pwl1j*HR^mY1>NUUrXMx=|2y
zjg1awZ)EfgWM;2JCxu0MomA>Km$8?@9@6vInrB9IA-lA$jrZ4VLs6xzt<5HJjt#a8
zo^|(j{&?Z1r2#9-6^?_fxWrao8D!n9_sScm*&$h3VtG!)z+qzN^#^~UY7+WTOd}t#
zi99bhiOc3bOwttFew$4{e@<k=VBLXV#|s#-xw+*Tn5{h!bGqUqGVdyY&kolMj$eJp
zcW`kQH#V*YdR+#BePn$6O*^aBGpLpgOo!A}Wib~Oa`}DPW-P>@)Hj%K8wbaARU;#!
z&PNH?C1}dRrS!lZ*G?xUVZ=5C3kKccLzP;^W8e*4!(@p>7d!yc#byiyvt$zr*vFFj
zO-wQc@;Htp9|w$jFHX_VUIH4O4AQM2|Ca`Regzxr4ZIOslZlCml^45cQ83&Z9q-Sy
zuiXMd${~4q#?`vIg2;WaiN<tsJ<K{^<4b@jMzYx#{hdin$BLeNzrw=cW=zsz%%0%V
zxCHzIR_#{E#h>vT&(3rMg+s%^gpJ3m7E#-8>;<o%=Xt01o|ZY<=8%#SXV?+-s`&S@
zW@2)jbMCVrW+azqCFCSXrS#`q%qNgW%C~bi!iuO@RYi>~)FapaXF{@4MN@MzIxMA(
ztmI+nJv)5<YfQ0{%&DK}y->GjyK#qCI*a;y*8cg$D!1y{{_|@vXc1K3$FMb<`e(95
z{(8*XCqn-WP!fw-r)CG2d8~+ujlcD3Y5U6<u7r2_%yn~$mR?EUp3S_;``7O-U=Juy
z9$v9HvFZM)Jo{tEPqhUxwim{5MgH^1{(i9-8FQBZ%i+sE59sgb{2#ykzh*=+4{*gs
z0oU*R;lQ@f_H1IW?WIdXNZ&!LE{j9)%bfG?pG?ST!m)P)w;gJ4BU4ibVEWFpQ_2a>
zs2V_=B?9QtY3GW{O3=2~g8d-t>RNDW8Wq74ur1J;UJD9h1zw2@L*+|fpYJBVX*=CQ
z*UbQ&7y`=QpuQ`7sjGGPDg&exo~YPlV2}w?fG04#GXKrDP*5-hoA`BsuuI%io0^+R
zQ3-acF{G2Y7Sq&HzO=?CB`w(!pwU-VTzu$@<9g$PhFE5Y16$w(;MsEPxBgrzF1Tyu
zJy9^j+g^S;xFp$bB<6J3fvdtmaaSVz1N=KR`@LeM)*#0R&{8%CvYIz9`}0D0it(`o
zkv?Qw1%CTx<1#-Yax7+BeN~E#U$8~NU|0&`1-Q%%6ych}R(KHxjito5Q}Xe0>L>;`
zm_uEF&jIRSLr?|@QUxttXJ4NerXFD2xhK1U?}pSW`p329cecqAunM;}A=VHtRWPlY
z(cK32aEWQy*z5+AaS@13z!cr=UyON=vEXPeuB}~;Du@jX7FJePBKN@9jRU6}6zLw2
zm}!7<u>~dV^|Z9zkmfD1u&{u*-VhBKL|UTr82*p=hR=pU?uUv;y*tlsewq=n4Ia4j
zD<*6{JYxdQkNaYH!bo#Rp3BT0jL^Akm82GN6b7NSW{7+(fSGaNAfloTaq4!g=>cZS
zMj;SwUT<Qy5-+zD(A3V1Zy*z<;Mtr*t<}u>M~1$bM-l2=x}rzn;aaF~=!$?)ZByF~
zBVF`2LMZe?K!w9u@xb5zMC>oI%LAuQYid^T3+&o;79g7`^#G|Dxz6B<>7FZ%jf)ee
zqoZSR2me{5jLqpsH~qb!ptH`BkHMi>(sU4PV+MEY_B{0%hhx>^nDCqiK{^Kqbze`<
z8!7_;v|8Su)d*EqwTn17aBDPkAtq*UC-eVLPqC~;sCQ5|o0ynftimm$Rp>g$M)i%1
z2&lY@xiT|5Eg~W=Zua~4?_CEE>cv=NfmEVedVuFoR|IZ4M}~gW<X|^Ymk02j7ZDRP
z0f3%@^&ZWxyC!;fQ6U>HP`pg0SK!yMP<N0ksog&?z^yWe8sJ)P?qtB?IVlqel2IcU
z7Cj|KY#4@_T;ce8bTrEY*#U%92689*03FL0sN>#Xcmx5^ji@N2*k7?Q8N}^F9Age_
z3DBv**a{Dnk_OEw_mL0-o;E^?3?jFn;D*ECI*R!DQ?SxB8wCa2Z=?7!&`MS(d~9u9
zM`fUdQ7AobnFd&YJF0Tr8%cL<O;+}<C#xGwzOJ4gI{Y3FI{?s(TOHEzcaaOn07O{^
ztL{ZFKyD2D0Z-<iSWJO(wF%$-{qu`Yj`GW{H)h*^Wfxw01{l4os=R;e>pen&hl)M}
zF0`1|&wUjiq@BIB5}_*Kdo2pVi$4Q8@~Km&R%q`e0cOsvzp~$A_O}i5blX<%Yu9K$
z<=HWVFE=aXLG1f)++jZho$VT4GYbnQMa8O-53PXkUwkRg1rdj(Q!zTO0xTcJ)MRrd
zDLXqmst9|Yh61#+ir32Ae6g6A7{q7W!6@kN{u!K+lVC3EpJYC|tx}{8=^QAulS=G3
z+lL^yg<-Hj=4|ey#5Pok9{3UvHTuCc%J<oKyrqjcE&#*acRcBX?--LXln!%#444o5
z4#_9Zv+kJg0#B)HaL^Y_aWd4w${~VU_<!T8QWVTJ7;%b{88h0HrIA`UguH<Cfb7Fy
zX=!;UDCma2|1$WMjbhp{;+t+RsELJpR6&89vfu&9!c$H91$GfmM_7VP?1f?!sUW0<
z1byc+FmiZg%GjBmJ!=TNO>@k?(eiBDxswqqAFR4d<?s+QDC`rS0?$SU?-EEW!GQz)
zJCLe{mxWtFMpKgwXYu{T2ViFFa3G>+iX3zV_(Gd32>uEi8yopz5P&`qvs(K!D~sO0
zAKQ#rc!JZ$hyos3|Kh4DW(@OCo+OO2clk!F!z5(X;Q%DDItj}lP{e-*1yAkwZ)i$i
zgO}^*=t#aiZY`n>ghrGpXlAsF4j%-1nNi4vfRijDPbfH84%{+@P0#jj-%g984l`O$
zTv*SZJ*xxF5eKB2#Inw+_p89CCist^9}Bt$0>H!kCUbW6Nm7TX_+oXD`6PlwFpxQd
zU&tGD#~Ei2tTkD191#GFdY+l^xzNp^dk=Q}6wZnJ+&jLQDWb8%JwUi4+`PQxwC}aK
zcmMuPRM)afN`8*&EB)bsh>wk_34g;C(oti-3mFzsX`BXfYG3jtumTDDprxfxTd#F-
z^9PzxH6LJr!W5_f3oK|xQwWMs|Gt5dR~w*@P#l(4urH@8!h$McGsfG#4F#aEoSYmN
zKYum&FIwj2+#0p$&DK-DFj)`*#W}~MYV972h8<knwiW;U`4dBc4H~C|gs8c>x%E^F
zTw8NUegyoV!456Mrq#suZo>ut{3+pvQkuARw7^8tLPvmWaDxsem;~#*7aVK?eRXH0
znW-u8AQAHGaS62HqR_(R3!T<Rd`uR$qG0ImO({*FXQ2fJlGxmvbTZG-<JqnKI)z)O
zjVLlGQwxi?<l|qyENbkRiAf-XJAw!wd_0fy<^kSYR(7@tK7=s<EpiE+tGKg`g#sU7
z)m{Va1Y=4QI{#28;h9-j&f>G1;+{96Fl_x^d9XkF(S5vPQ<%D=z}7^Ml{<^I=~v#9
ziAQ>=XREiP3%M|Yf`T3-CGqM&^ot-#@$rcX6LMf|XyrI~Fr3B^9NsfHjo`}nP%qyx
z7#*VT9Q=S;sE+_VH0JTh@f(Hkg>pda&q0tMN_gSZg|066p)$+?RTK9j)T<b``-8i8
zS3QKO=SS#1#wRDuHcMPKB*A;*l?N9{?Kdq(=0mamG*hXSel(ho09%TC<>H5s@42rE
z`CRrdvB*t9J_Tl?g^Zru1Ay{r452GD!--A%5fuL0ABhW6tTC6Ej$HY#<Zo>59Z3Ts
z7ht_QJP!e@8H5nu06z8FXq-A#gw?g4>ihb2S}*hN2M!5|Fd9Sbf>jnTL8)uOI3O-M
z`ZY$|RcyxJ(5Z0x(`f3-btK>4rS%$X`<AzNXUN;i%F{D!?m<)veWe10dLfv`5SClC
zRZ?#A@6a=^r!ZsQXY<0#ivjCq4fjJ##z}59&dT>V=IL8-swkST=<?;F&^%Wo5+pPv
zWUb<{V=FbG-rL9Eo^AaxMCS_9wHe$YH+hO<%3}<m80rVORONz%c9n(N?=Tu!SMdGY
zH#TbbO0^)vyLavwS`mYro<a!R>54$upsx%Gq96q@@wvXQA2<lerDXHiXvxUXiaXf+
zD8A6x(4dL4^u;2RtB3aq_z98n1;7v%xG@gZUE<>RLA&%$Oi0K;{e7|baK0p1bjCOh
zzd`OsYtq=*7z$4Oy}&>tXl@Ph^MQfI`SlEpjLboW(4f%)j~puJ-*~`5kZ$Xp3unJ6
z%6x()EOfC?RW3pkB1E0TGN(1c5#2__F|h6Mtl=T`f$MQ85gL#!uI_;m!yb5c#Kp&#
z!_)V{ty_wr$BIyuW#U(hprKp4YSs2<1wa^Bs38QpkQfOkM2~y~%%iHIVdmxK#X{kK
zF#5APjvuG6K%udY6LNBL8ooJ;aykP=NJ!ps5N3v&lh@64p1462Es&jPaQo=$)=%aE
zthi9cdzVV7aGL0FTd2mzc#jOnMLB366IYpVQrXX*L6Q7GnmK<vysM0;_e7%uG6hwL
zys8RxDp~-)?!OO(qkC>aG*c03dwg@0Se$9%d-j-8&_0f^882`>zy-jowg<-?jUhhw
zCt+DxSuO#AT5!lq^$T73M<p(Fx))`x<9t94t|Ldv>^Hlata8UJb=8NH=)>1jA)%pf
z*^JL&eX&sEqoXT$b(#h<wlhg9*C+-?_xv0iI|CB983hNH@<o;2wk{y?R#GTa@|a3T
ze@|HLzUTstM%nwq)|<8c%<o>6ye~`@=ijyKxe1B`?EPmJs-gWdO?v@cl~ICtf+3F|
z8w|hT!V9ZNA2Q>jZdKLR_Pd^q*ki5jV**ZC@t5A-(1?gFB%9U81n8LN0@8cgl}Yn0
zmIIb>3z(@4?lXb+nk2=<a0FnW4WYIWw)^fT?chuSpJ5!B!l?gqFpM1W4Q$$tLK{a9
zljx#EZ|n;$6l!(jw=;b?sat$s^YhI2oWUg))#hy_Lwo0lb80CD+CT8kFm(^u$QZL1
z;}*)GOq8z{&mj|S9ti9`I2pKi?PAr01Y5w?87DX0sL7^RcVSb|4bjFF%uO-vE#nuk
z5Hv=IgaH!hNGyIr7}0eBY^5k`>+dSY+#7n*y>oK*pn|?pIp^IYkQNh~h1WzmuOFhF
zvNw&o%fJTRu8Zfdd(admjzi280D{pjDDyo%Juh6od>Btg@l($B)_C)lG<-T)MaAUT
z7g2F6_B+nB`0>ups=X^PhPZI$N*Nx19c)I(RSm%|iia<iJ{xXA$pFKJ>p=0g^71Z0
zeL!GL3HxCdPR>}B+&_3(A~&8b`IBgOOVL4gq8G@J+`5&*;*rJ8cM$=v9<^^ve@r`X
z+JA)w1$j9Jg^`hw6gp=xgbi`B*EKbvRvD|^yF*4n43|5x@PS3r5(t*KNA=)*w*jAE
z(c@Zq_);KzQtNi$-7qe+c78%EoA#cH`HoJR8O1XZ0KloDGo;XIa#_*+A7jX@puw3F
zrIk!x<OdK-NQsTx2CIgE>%nGzU3e@I6w~9tOg7oRQGseX6B-&yZ56b0ya)mk<qr5)
zcxW2M6Y(m7wFbgmneb_D;G!e&BqUQvD;$Y0(|9(kgI_xZ0t&1uN}5xF>Y>~Y+Hl?1
z*SDxp@%M_KPy7p?cOOe^U=6={*S|2D{WGtVK2-!t0}!l5r+%?qjn;jPePaaeM{G+o
z>Ix4xjV}jiiU{@zn)+5eRJ<)R;zjq#fQt(7E3s_@8E4hLW%#1H=+A+szQ&*!+^ZX;
zio&ss!}vPNp38Q2e5`*OrTLFI=WdeUh$oBxqa7j$A+D;Z+G2oM4<zh(5sK!IIV*0!
z=yxu-aive5T#lmh(^F`*S3@iH8ZLWD!-k7_-oTPG1L{B6Z_%P5TBBXdMz?h;e*^}$
zWZXxR7~$i%1)aGRzy6pN>9ySjI(m6|QMh5iNnsU?ud2eiy^WiDF_z4A<U$`EX~vbt
z=&oY66r}p=*y}(+(a{@pJx^Z3Nm&Wut5UJG^Xfsf+`JlL_?+EP77$nxaLUe%wC$;k
z$r1(U&C2-o^`UpeThYB$juHrNOivnwXr+dy<Tx=Wpem}X*J$4l3)7C0z?&fG4~jf<
zOH0z$)YQ~~bkM8av1VuG1QMO5hKE!)xVX3wtPkBE8I_<}2ZJ>LJ_*|~EyIHg_LT)F
z1jHCPh8)g*gg5$O%+!TClq6T-)zX8~%x<gnw?+Y*X9+?+uA@b1J#!M+DtWn2&oV*v
zxu}wH(IQywv*T*)2|=~A1}y=wpo1AW-_ZWWH_gRjfCLuMDK)jj_>NFYQ50Akcr6U%
zWkPu{>(0TMeH^rFlx9oucU7Ehg3I5(XIq#ZUyj*GJ>;_<40Kr0<ya@h{p0mGi7A5(
zhS*Xouw=cxy%R!0RANkV{Fb9Iib_q@>zl*m&mZ|I+}zxexIr)ulr=Xu_c39Yas~Qh
zY6y^*4Qb>UB~sqM^z(rn;J6CXEW|m&!ore)woq7S?fg+xIi*-sn4RB(`%E{08kx4i
z=-~i4Jr70IrirExAGFQeq15HuY*4?^8H><o2&tPzm>534XzbA|r+4!fs4z>>NjwkH
zCTJN7254(9f}<vS{<Cf*!R9bKOiW#VYeu<TfEVx0!@(D|thVQ1!ndm=KMy_u0RU9N
zO*F>*7;2qz^prP#DF+Uz{_NVeXU|%!F_}Vrx2aJY5M51NGuC*b_C}pY*i~#BHjn`}
zI7Cy!dw~nqL(Vm9H}>TiMun#aA~^rtU8HA2EkPCR_)cw=ltgUTu<m+sad9+No#4*&
z!9Tbkg|S|wE>3A;;Z3<?m!i>G3Tpv4*sP|WcbS0T<xaL7isfr?i~?=u0(9{JzKo&b
z)OEpH!!?7m$TIQZDV&T<kgEWVB;6e}++KiPi1QV8HqncsE%G8Xj8ub)^sku4XhQut
zIRbxiItrR2FOYTVrW!I{zkQo1(@UsBomBxd=xER*_HHnz4Dt+4|6EUuYQYTzVViX1
zWKMJH^l6fVbP={qxR|kvy{&XfNboDyxeXW=01@T;;Sd}YBnTB(<zvAyMMWj05^Qgy
zbLVtTU%;~<5?vh_zol5!<PD>4SyhE@n0D%J#%}#1L4b>-J~-otSg<H#hZo`#T2?MT
z7R+t6Y6Tu9njFmxX~V1f`dh5v21Hi`tg?*1wZ7g6?Kdr5h9f+rDO!<u;a?^Un}CHg
zxF1Jpt)pLmOd8#2E!z1|bT*`w(p<M;1FvY*;rbm|YO8i_+jc1O70?R0B8=op_;sND
zqbmY#$op&rBQ8O_*ef8NqpZ271fM&Vgp`DGSg3F?xZ4eX(u#D3SA3A@kE8W@q!%q?
zBW>S0;~#C_BErI&019{yFnM$Ldu9rV2#-4)Fog=7pQKFp{BZaNq~epXa(VzjNFR|w
zQ!}$Hpq+wP!8Wu(Pt#7;OQ4Xpal-#=!4Nab;|6ZgP@oBoEiGwko6rg?L02{4YT}oP
zTd3}wBUy?9Cs6bmHu&Auhb+-d){BKF_Y+E>=Z?1|@6Ak4Gkp^f6AMPbY6kgB07yt9
zitYLyhsk>sPm3xc#R+;@i^z~my(p_mmq@C4l<Z3S`Vo)|MQny_>Slzj%0XYQ?Yo9z
zqoS0%Y{91}(|D9nD-GiN$6SFc>#G071vt+>M9hsF8qVM)L0fkl0;q=wYF^yb4%&?E
z&>lg-si=M!U?4}GWlFf}s+RBA5;#%Izpx7t!-E@Xj!k5`=MEy%T5diX)KrxK*fp_5
z-h*M<1`0Rnh)L_J@aTIpY_4bw0biWMhWQ39%Qt)tB?NcyaB~Y{qvhc4>ihQX9Rw2Z
ziDfYrm!4irk|_vDt839#<AoE2?XaoJ!Qlb2o#Y-22*9eU0{^ZJd+5beIVLXT#Kong
zyj#0Vv*^mV_r*j}kG4(~Slk0i^gSrfLWQUlv=ishWADNaL45sn8bd(pU6mohhN~u~
zU<??Fj`ZH$yJvvBzeD)9(62eBqVCps+^M@@6+zFy@DzhB+x}O7D(@=L-5#JaB+rmg
z^Tu$Di{Q=0rKVPMzClZcp3eyTMM+op(Q(N6zkhmhZ~FId;@;2Tjy^m|rnUbg`Esrm
zxRu06Tl^8Z^^;z6*1{S=dr6gz;^u*e2hBq?7^u-#>tvZOo!l0SmaR9!jo^o~;_$<1
z+!gobKI{7RZ!wNb-p?Y#YXT#Rcg4j~d2bR{g4oI&6ilH*TUB0NeS7`JjUt$=qPi|8
zyih#M-D<y3CY(k0q;%wn7n}_rTUE2dVFIGpQdGjBD6vR(^9-tGHe)|1<b*Lui%U#=
zi?d!%ff6pmG=VJ(DK=0{kZ=0+>(@+lA%A*XTREmUNX;@JFazo8gAjg_{AIMOWmt4K
z22Y$kX~w*EmmvU(Gax|t)%O3{XP7F4ptBC<<15}V_(P=ZJ=*oqygUhjQf46GnE0t5
zrxnBm1_+6NVsf$w9zO9U#vr8)o`67MtgZE;F6q~gpD{Blr!wGdvZR@~GBdb8l*|`y
z?MLzduD(9->)@a;n3YKWDkBX%xP*%;eZr`-ewl#a1!4XamzGvTdU(_Z1P5YfVoJQc
z$xrAC;L;XcHgQj$ydnP>PbdpkmiSkULBWW9bn(xdxu5Sk7lwNK7o45ap-nk_7EESC
zSZz0NX4tl!4}$~zm9yYerVpR8T(rg5ApX;~7_a@vZU8z+a)5S0IGXn+Q5jPa0i14x
z@L7U+Xc8wwo+PPVwsFGS>ov&4U7ek$+M`x%zl$q|RDeK|t^;ScyD95^6j+T1O0(DP
z?VG#1Md=yBX=GeiVjw^qU2$`-fXOnJgH1aKvNRMb<&=m{f+orwO+?9!cw+T-qZQkq
zf}&&ydjayvgQHO|;GL@3vxzZ4>NPEg>H8sd^)=WGN3h$;><SG<C(y0E5dGs|fp+)7
z;Ieyv$EEn#Vl3pxJmG1#1uw7+cLls4eK5ZP@6QM#M$KpDE3xpB_a7$D@z0Mn3I1=9
z*}vwXF$_4an>W*LOHg{b^6xlv&&kURcf%riKYp9bSY+sUq%0sQ?4Lhgaq@7-NJYBu
zu%1jof@(X>zdsZHjO!!zJq%k-+WuR-w(0*Dm6}xXzolpbzJFqY{(j3@HK~I%55Pf%
znKzJ%jT5(TM<qqm8gp<wJKwPdtTtjK?RMp;Ixi|+z)}u#zr}Ku{t<}*3=%tX`cn7S
zKffy`?&`w)!L|IlFpbncq$*3dbg)&vb+;aA62QG>4tAUFrU%=_sQxLZXj5}kdi+h?
zR9koLN_jnlv#|(60i*VUgQ8-hk+F$jPfL*5r;N!Ev^5^J0^_~HM%Rk=0G4NyJjVtl
z91+|gkjDM`b-H1$9p0|jj;2lk_H7d1;{}xvir6LRcMgh-9-K5bG2T(=k4^fz;WaJ9
zL^u@GEnlGaC_X=7`OKny!-l~@BA7!(PMli;e@!Af!MJ3!i%4!PlakCor>utv0|cYg
zPu6_4QmBm1Knrr}w2Dm6k=uuTzss*&2;EQi1sfwHj5C%2UN>*d;TIJZy%QYjGd1Nv
zcm_xVE`=A4pi7zG<MrCXFV<|l)hj89zJw8wJ4NBh-g{-%K}`NXkCW?|lG5Q^$k^xS
zoWRqLyr0V7(Ac=racl?I@kibWVD}J+Arek&t6QL&pv9qc`}zG=Yr1|H5Siz;)57pS
z61?*1x$xv$wl@RHPgX!{$HBn?v6Bp}rqEM(HctI;2n)Kg5B2FZ`i8B1`(KvifxUYJ
z4ED(Bi5W)f+ow#DMQ{^x4dBsG2^{}){Ake+<MIo?-mQPM-))PjscFa6zN6N}Ht?0(
zf_{wkVHbviN!Ju5zb~zK9zJA75H;C#C>cy}1@0=zh>|6vA9A>;#vhlxa(XQooJ(H*
zA;?pxId<DW)=1YacBi2unVpBOT&Zhmka2cCKpb(<a^H|YEUm9!(Dc3PB|F5>K1As?
zF}Q}!xVLOPYxMncU#^23vYj1F_QR1_yky+4*2w{;0rVj7(V7tYKCs3fbTyipwwzqf
z<IF9!LHF*_p{fha)#>SH#E_rJ2=R`kwhI3}3y!KoF<k!k&I$<N8T^gQZL4sZ67PRP
z;m5ToH=fSqOO`8o(Dykm`^kJEH~>_FbsIOLBtPIfMymJN;qj6mJ)}p1Y4HW#U4Ape
zfnX>vYiS`?X*MM#k+4}dDT}}i1m_}0?;FNX-aZ~Up>RhTet0efy!zsg22KwNJP5(o
z_RW}Gl@-j*@Yxj{{eDq+$0A^A@FkVUoAuwa-wQ}&(93Y{+M>zN0l~KeXE*$Bd18Rd
z>Z_nlxo6#54Hv3w#~+GcFYlqDFmtQ_T)71I>2sI<WtjLVv>Tw2+@+mUf-&Gqs2U*V
zupOE`on8hvy3WBqFEnfSpSxV0G=3evABNROfI?(kY~<2>c8+Pynl+!?Mk7`FZyFrg
zuA(ND$~=lkh|w-xtx1(2JW->*uv1_(xkJ28C|S(wcCDyLm^U@`a`f^da7vs~gr672
zf5+9=R3o+<fCFS<JJk#T27O~#Dsxa)76QJ=KoBPWCE!D)#6T~}4*cQG;dSdESu=tl
z>C<mA(kH|<WKfzXB+DwQqOH180vqi`gY}Kgo;PnYQjx|tmqBs2(r&`i$9N>1g~<b_
z2>(NI_&hDicl>q3KWER+iJx^@^2<SBV%~fbZ3*rTI%|IZ0nqBC7v`ToyhA~Q0l6>H
zyqV}<zuQgdNB?g^JKcl`+{MP&x`KOzuR-9eonuaeD*mkVy`R;!weqMVB{X%&#~s9x
zVr1ig^-}+KPENVVbIK$z$+Kgy*}UZ$Pm`K@eMu8*va<_~^V<eEG%nQ+3nSqiWdR#;
zLrN|#)u<mj@%LVX^Gd^_WyWQOP2vW7@q+3nhSt3VW=%WQFsB400!SaN&vFV&>E^K~
zM+nL7qxkvCPiD&EZwVPr^Z=%>MJa7r&;UYo)@f^O(9?hOquavLg>0)AEg}LDQp8sb
zfhfCzq&GojMs;FP)Z230G8Z2Nq4H}A7v}ea<9w>?J2@!&T~p5CpisuF=P*diKTe$J
z4QXExJNC-3OJ+Cpr%6dkz$lNHnsUjUc*LFMj}cfWDk*S1+@;5sQ}{b9K*PRu<ffN|
zIIVfXF4f;_RNq~?cg<P`7neEPfUjo1dGy>LAGmrAiFf8!R&UzcUc7rsGT?Eyz@L5#
zX^&B>F^rnb<^d=IYw$WemXYz`0ruLt56@Q?@M&~^e<;3KvoIogCy+-hBf~RiXf?#e
z6A}}<pznEKV7(C(O+uQ(Oc7gX)hfz8vPc>m)(FVqr`u#6o2HisWzurorSfwOOCgK)
zM+;#beTx|o6`_++MKyXaGb&rxq462jD%b?~VnOSMn369sC9R9hF{eVENu-0jrX?7<
zb?xnc4t-pgnQL=VTnXE5pz)<RSjJrdixI-tm0mb61C**hQ#D?oTv<jX`+(d>2PdZ<
zSzF=0>c`7L?a=k}yMYdOxZ~x`2M@(x+=O54vYdR!C=8##huU#@;^zhEjZ(%YN`tHJ
zuf!Kaw@gF*e-Y~W$0~+M$JywSBfo9mzERCrY_O654d@NtS|(oVz`#H!&cXL+EC8}*
z6uxKVP)^(bx8$N0<$ZAF6XIc_49}fo>8zT(+eyc$z`wvFz7Cc9DK!<|OOMkF{MUa*
zR~BU_XMCl+q$Eg79^7OitwW4)B#`J&V|}Kkdu@N3`+NlJik3oqSDEs#$ra`t74P01
zHaxWgvr@iS;R!e!Km9oyB^0bICi%<QY|2UicBYq<XggV1f#^NdO0R9}HiWW-qLP1o
z_iQ2LaY6zG^B#910XRJ-u;z50nR}prRvUQWn)CQB1^7dPiiEqyigf~R_c0KD1J1*u
zd2r1R046DZc5Jo7%+!S{fUth3lZ^1D9`p<^gBB|$F79=8^BOEpIpk7FpE-wodz`3V
zGmrK`2*%Xe@kJE}U)m8lrk@k@`z5zPNM4RP*K>oNlp);U%+H>=h9Yh^jro6z@a$+@
zXMUyD39aR9g3fO1odI+WCF5qboxQ!p3kK=N6a{(Rc5(5^g>UWlnyjs@MLz@-?#Erz
zJjLqN%H8r_TDinmlOrthjOedbif4|U%iMPQno#B$wG{4Ov<XvhHaH1n8XZ!L@m?}M
zziqz&->v8~OcAx?Q#(GtxHMVw?cH;yU!O099yD79cV6Y0Q+e=_UcP1aX|;Gn+cVL*
z=MG)8GjHBBTP#7N{;fSXAfxUSLa5|kjP=s6AypUi8E8h5Oei|rzO%>?BHCEYba4!v
z#jH;h<HVxo<mT3|zrW`~=F%hLA_3LVV0DD!Kw6)R$9<o`!mmv}kdcAxmd0Re)s%1u
zXDaO7I&p2mt)OsXX7}GZm`s9*eH+@qUtXSFU%!T$#(3{W&D@Ws0s;&0>3@I73$xHu
zR#vvWpe!r0eY7J;`R3R42<`<ujwtc2+b)d{#O-&Fj=j7wWlL}Uz9=u-Dpc6cGHh%4
zjKf1GO5g3EYl5WjLL33Q;4^1)p&G<AhtS3eyR^931mnjw0|#badp2KDF|gO_a-8K>
zbsy<?$>^?Ed~JP_fne|L=PpC1!>kjpU!Xv<61DlnCZ`O2FCB?$$6-_duKG4kds}$n
zPXuJ&UJAs61f34D8Xq4`ynRzX>~KJ;b}oh|nl9~NMImWGm9P`DvU+}{>PW~>nV=3*
z4I8ja3|ezWh25<p;vncJh?;Op&C_qsOT2aaRX|NdgoSNT5U<uMcb(kxbQd5<BJvvq
zMm)RbMu7@#NM6d(^|zl(nq&rU`TSX1_7aU_%#Q2rDaLjQIp056%;}9AAuLfvn>`P=
z`H`;Io>zZ;3zZ0;vhJ||dkE+^N6Rsfy?=kIx}N03KQ*u4-WRioN58-6?Y+z6W82+4
zXtn9qHmE)^Jx-lwWQc2v<GF<~uKfaLOHVNvgf-_|gUBKTH(_ex4*nJ~nm4PAr2Wx*
z?BMPC{dA4HMFy=$JX-Yv*pmXf3;BTfx^LQNJMby(n8&E4<x|n`G3ROnYp!uv?iUiW
z)H?lQi&W)e>FUd^7{;+Qw>C8fi)ys4n3mxQ{5-TU&s-E5G)n^p*qH{aEe8O`v~IA}
zP0<tzpO~MYIge(j0tcVnJjWwk0*5L^yK`f$29lxzSGsS|bjD-=AYmNRiP&P#HCBZ1
zZ){G|aO*iRDPF&3ZjWOud)Ow&yj3ed!mJQ>JM(j&rpGBn;p{oyK1;vE5Kib|cQ%~k
z8?C&#uDUuLbOBL2u|n?rx}PJP|MfIQQBfR%>+SQ^>k-<-jZ3E=eutMw5Ow1Mz1j`?
zfGs2i*lwuR%4PS7lf*RZp?CmVZLD_%lyVhm>9r&ULo){snYOR?l-B;VXsl~(t*v|Q
z;~On#^m<X;iO_2x>(Nx*FTCQhDlmH~EMy@b@2h!u-!!A_rQ<`VtsJdAHwT_nY-o77
zMRebcuw#CV&Pn<GTD@4}G)1#K98I4^Z|L=uO$-T4HvUiyNi!-b<MBwkf#;gZb0`ny
z6w2wYbC(VJo1=d1v-|S3E`?SVCSN$0<v|+~I`&m|(x$!XY|#iW$JLo`RsV9)@T0kD
zu7&*9S$=Zt=Il;;`{RCK-q+L^M8_|KY#41-(-O<Lbw{g@x=-l+`$^@+zB}!4pW~Rs
zv-7p2IDnZhARwz|byiPITZBx;&SHpBdN)=vGbg8}ZV1wYOxHIIP(P1OOq60U<c^?&
zjM<Swi-zYya7$}=0jPhz3e#1nOPqQJ3PE>uygX8aCIJ?CJ6G>KUU^EJD^ScDg$0=J
z!ge2?DTS6z<xEN+Nh^OhjK_sO!TqQFRB4Rg`^F~4-0&c~#VP@#ir+Ki!h;;OzT)Yf
zoZld3br&mw7~0}W>R^-pfH)LtLgnt>xDas0Oj#dsjo#zn6pVixZ**^GtrP$92rR_(
zbO96?=RQC6B<@SrwFBN|lO{p71*=qPAEi4Yb5j1Ryg$pbM0!ER;KswPt3Hvgn~g2z
zsz8UlY5vIHwuplJrwB$On4XkjiDc8dhm}sZr0CDpcw{460Vi$^b{|nPz%Y}<`Vb3#
zOpp|Ucc)$|Z1(Wr_LaF6!f)_#9<7Xd{i}VT>vq|@xglzP>h#=dEk*E|u&3i--Sk|{
zIw{yzVf6u$d7XQ{eY?i+r}ML)=l}DtBw!A99Y9#vn-TS}v7e?_EX^4f>Ai4N*W0@z
z|H{)36I05x?)V(2|3ZN{5jO~sZ3AuEff6F)!5g-o!`2^S+FZeyK|D{CdR(Z2c1baf
zyD(4ViRE!A!(&OP)XmJ7B$y6bWANIzXo1WdfF*r?f5fx`*$cojF@y~Viy*bD-Z%Qy
zMZ1i5DiXbP6>nAJ8_vB1aTqv%JRSo))s*8979O#;xJ|^2JM^yNsUA#aae~Lb<(MiE
z9&%at=K&k|<DTg5>heH6N`x|5dLpU2B-ME%AtWSVrq{$!sD$+gr`{VFQ$@oi4>6xd
z_?bUs&`4XEbTk2FUH0nT&Ir<BG~#Ngs5p9UkB!p!W(DV~CpJmd5C|KQ@+}h~VfK}3
zlkUA$FsIYXv(Ep1@J{`ZkjM=8(+I$^#6)8^*koK{ZNovOl6I-X54ACxq*=UlX)LNO
z?T@I;_Oh)cc}6%jcATw^*kyOnHAqgHikq1-MC{|24p(EiWD(Pq<KX};K0JSk5Qd6(
zT3MHuwd;<SvkywW9<}|cGa7dsKOCONQ?COGzU`NaV(h+sh%s!LM~@wQ#&ZemB?>ZY
zv9(A>Ey|}m4<00hNFg!FgN`9#&MD0DY=o=Ah27h>b>04M7KGU`*hU0HK(lXUt@G4D
zDA(rmyJxMAn?W@}Y~Q=Lt9LGL%aXVHIO6Jxcq-lMHwjFiYd``XnxlO%GP3aW#86A<
zhr)M`9e<2+O76dDX$iS8<$(Gi^GfxGRBP0OQ}NDMQjuP-J&e+0s5O=Ja_g7O%N;pl
zJ66I3cbKN9!10y0y1D+I9Ew+;1bGhOzqc^zX#{&$)H>hc$2EW+cVVd8I{&==i4*dw
z3*JHS!a22*;x>q)<<m%Qn3f+uXn{z$6y0oJoIx}KbS^tjKI*)zP+)gEH=j}1@nC@>
zd+GfrL%AKag0PfKT_1|C`Xyj`97k0g#0i$uXmR#l?OwDHSHeT#yDNG6u4!XqEx~~C
zN%MTr+j~9%0p*xlRyely$Scmnleg9@a%;i&55Z*Q1zflwO_lNEfvaHM_77wlKa#i#
zAbz!JZRquzC@l7X8u^@kji>y>J@EV<n=NIAJhBA4dZ0P^ewcYvL&M##l9E4%n{NT!
zhTZ2qnP0z%SFxwtoE&f=ue-gbRUyO_@+mkpc@B=<^*Q`xUiG%Sv?*2RJ{(h5u2@l2
z!nFFr`9?1|DJDEwvD1$|z^XA0SeD9Oll5OG2{9CP{nz^Z{rQfQaff|YZ8~Us1KYFV
zse1CH{-PaEupB$lD}yjdx&+pYtRC>g&f)OID`^xm$^L?AUtQUo>te|ocr|<n4je!Q
zwBPB<h$&pCV8ZP4S!(u9qDxiEq~4!S+F8rACQNRly*+LKjDsVcJ(uh#DbE5I6x-z}
z;(tbm+g3Y-cAKIH#MwED5h2(ePqM0U0jMUba8i9Wp>IHGT`%r>U`~bBm85Qg@(n&s
z5Y(Z{2(c%I_6V19`2CYdF=SVL=*bxzv)D)9vR@<+&J@}T5>$p7RYRVoz_PQ*19wt>
z^NUprdZ0ze;nI@GMGgp!JFj}OWvnVp`8r8TO;b})#eRGD+(O3BeU8reQIMu3BygM7
zMx@TqjY6A!j%7(V2;ey0Lxs+QArD=I`S;-(6tgepM+!x4FK+AmVT^W^_+lbnTUFx+
zJt5Y7|Cmjpzfu&kyMF`axd51<mFu{9Yr(<ku|7^j8o+RbMD{pN3@?54@n9B+vvAx`
zoS#D_j){|b`%diBcr5RcD-ECD$XCz*Zq_hZ>DozoB%jW_(k=mO`lC&AGq`k&F<?R8
zGw_Z0t%ldkt&uZ=NZpTJ+#dw0OAGTZ2nwo2Vtq;HHrZ!`n9Btc4|1zMC1~1U=Un=s
zi~&3^z?pZtzQ*5+SHi&S#_{6f8^{`Ye!-|<Zg_ve!n_kexN@**$Y7u?%Pb!1di2n%
zajA95-h6I5!HY__Du^;fH?O&NMG>}8=zg<h80$2pY6Ms#<`sCT^kmZ8&{W)C-#+9&
zL>ai6jqNBXwD3YbQ&AVi!X_xq7vBNj*+Y2uSVD#acede9L`}SYkMmw)Ico25XEIQ0
z<mI1{!IW)Nn%=_&xx<1OMB)r0Esg7yX6Kq_=S06(E;`}gpG<E`Q$#{V6BT)6uyzRO
zym9;QiP=3Cs{nqRf@|;~G*mfP|KJf8_7iQky_a(~Lca4#vVIP&g2(+tduOwEM|1T-
z0Ng=w3-6f7p2?YX*A3Wy5SWBSI-xWrSjmE`E5DSYC9dK)z@DD_s$v?K(sl~z*k1Ef
z9Dl45CDO|j$oU#^X9WhLN9(+nLNGw4Vs-3pXJPX?l70FRjBdl~zo4wpF1Ij9sN2az
z4f3Sd)V|vxv|f|D+qx-1KRnB<c5JJJ_UTkdxtqVf<I6BJ&QBIz?vG{u+bBqe<21Sa
z>t~E=rQiB(^L%(D^@x24f6?hV^$8&_fe*!P1Z!(YRPB4i&DZX)W?|=NjC7D9I6P*O
zMQU$O)7W(6$sIkq7=CvaYd`)omvMFkvepaErDcunws}P&CcKw59zX20;{0)CR>*sK
zThA|9uQ183Tp&YFF4E=sPL~Ec4-rH9?-fUJ@*M{Asxj{50{XYV4r?pU_hUmTMF?}Y
ztI5karvYTe!Q2OXCMi3O8xB^1x#t^Gv_!(0d0l+YE*W*J!1f&tl?>0H`L)t+K@ydP
zT8f$vnd&1n<77Lr*#2iq6?jbe5`KV-;{Nylw5{a6jP3sji4;Ce``;mvXN}m;E%{eE
zB{|Bv8}F_BpZ~%tiuIpAfPcK5`5qPduRn<YuJu1NBP(8GwfyI=A70BvBmLv8^TOOD
zMr$iqtr|;wx&7b&Sl%FT=3iWZK$ByX;63#uH710=0I&X8V~rX}2u{+VErt6WoY@Py
ziB^I0Rj)yi-a=4o-2-Q2`6s@=UtdxR$cZt?%fwI^fUB{E#WCzGHe`2_^*Tx1zd^h~
zmgGH@{gom7Rd}aYi8Fs5n1Y6gEJG``7PlB|OcgPmN7(o1O-a!o^dy)BcUK%rHPCSH
zh?t#t1ZqA~B5QuH&)<T)iMqx{r|*wpbnAA5gSDGePFWMNe|CdSMCO8)SsC3Q_}M1`
zo`?Mg^`#3@VcU?y0_xCRDL2<V`;qy##s?gIqq!eHW7JZDvn!cdq9cz%BRJATj`a(+
zwuEH_(_XQAPE$spxGe|S1O3|1e?80+vmu<DU;(a1QWOa1gm-3O@EElQu1aZ`MiU)5
zNH~L%xG#zUyW)#tpn4bsRroUQ1o9t=p>~RPwnFk8u&WH{9ALkzNZ8vb*a7wL<l*tD
ztE*f1WWV&s^j-{zF+_$n$qh(_mGMr%9Vo?`zKXjE=))6KXqI!iFScl;XcmLbbGg6<
z6B{5gBtB|%bTq&=MTU4<1D+zOE#Mu1=JO5A1OWf!aREf|g({4KLV-X!fTO#S)(8vM
zy>^4da{P(32QA1`chKUVytLJ~LJhvOYE&<b0Y*NRL2kz?y$@P0u3t*Zs78ge^EFUb
z$-E8ZxY^y;_v?^>{GSWkdm#_t9glkhDZD;nNE1U<3KY&phd?pM2YWdZqy>-MHQ5Jv
zc#d7Pvr{MjwiBDrqAFeziTGg6<nO4N`S8gVFw2%KQuYy$!@<oejWzn>^fc}YR&aGV
z!M~_PUP#l}ySyzkaK{D8#0NFu)rIRTw$s7#Ern8miY@yLFI+?vl250yO8-D?R25cW
zGF6FQ>e&a7)N*LKF+iU;cX10=qM2U|x58VfXYpK2=Vzx8=(B~a+y2^<u$`I4u(CN1
z%w1x7M#;;Ui|!jkC3Nx9r8kJe1pQ|TE4MU6&JD9NWB6B}fccKbv<8*OTbMv2IYKR$
z&-XQMNRnWRu@_A8RHF|wks%hkuV)_|fl{DCjRW%^@)y{4pFsOVQp<37Q{Y5Kr>CF9
z&<V)|*zdsc=UOAy>FuMigaY6x3aSIkW(ihQ-0;5Rfd@BmQBFxo{p8zdlrW<|gyj$R
zb}7_x5Hp2er9X+ZGJM*K$5K)tt$(QuzJ;4Nc75s;Vti3%BJqbGd#V@%J?scFNyApb
zcR>=P2qcb##Q-5TD#k?&3~^Vhcb<$O3L~7&w_y;4StYjA8gPFw4AggW{c|iGZn48l
zr@|kHVxchuP1I<93LzOR#NVPo<WY(7){a1jjk5@@@=akLUxtPhk;8#>jccbr;{9)j
zToVO?Sznj8z|k{jq9t5%v8uH{CY?@IzzTLi#|pNq<yBMN0A$0Ecpv~U<EU*+YKj;C
zU;vzmT@34TbFdbgZE+MLVUD&4)dCF>yW^GjC8rmFN}-n=4&Ehb7t&A&;iQ>%V7%iy
zzw>IdfL~P9HVUGhuKBqc63+ak>V5#}abaH$K))1CI8H2MlI03M5KcB>G{2}TkDwi<
zA*@r_?1yh~u0o;!6umuIBW5@Th<!OqZS#CnkX=ze#F3&ISq$IrVQ>KYOeFD@4;7}z
zKpA$JgjWflzD*LBy%E3{&AkK~8Q?`Xa7-k=d5Uay1a%0*ObMA&k8zk2hksCs!Banr
zehC27On8kt77>mlPMnlP!h@2zv%kL_dOqmxi#s|DQahlgA%PP3aO>fng86+fCLQpF
zBS{kzag8i*!~_Kir7&o>UKlu#!f0Da3JX>T>cW7Q7c=;37DNHYxQzT+XdtgcryLAA
zfO5>ymDnalfF#NLg1uHWPEy2r_~DGB6S%amB~kSzkOF|_r2$g&M{&W!VB8QDH}atF
zqv&}xvG1=-@PZ4n(C9mHA6ew+6*y}pFDpX5Ph;Nplyh|K<|r^A%!uBG9Cy+!2j`41
z-7y})r9ra-0WjF&`)t2p1|cxM{HU`KxU>z`<BSBT{W&r5HdJKioqer6V6A_JG6mya
z@NT&-Ej#SZf;2+}{;2lnkU8F+ohP?Yc<%Jl=s_Bq5HPYIcWf8F2uS!!b5Z?IexEg7
zp2e-FO6(JG8lkev_BzDp8x3HB$9)XEtOOJAa!-oTqr1`;f~h)UTVi%_aam>)-wRwb
z28pEOUCYWEL-}G&+rsEI^*3olD8dr6_$`IFPx@0};>B_MMq@o3p$URsHzsjE)HkK2
zrE(-;G@9Gj7Lm>L6{D=3xbb-1k@~kKI|aq^l1P#Wk!%Bn<{3-NB=j_iCOBP|L_V`>
zJ#h><v{lF#jl*rqi)UbY6?Z->b`2-e?~>aAa&Q7h@=Z)k97BA$oHA=KcG8ka3irRx
zvq7uR1~4*9wUv@86m1lRDN7<%i3iiT$q5IkU5>z?1EGM0oXo^49xI0rjq_==0`a=}
zj_O)@mlN-z9*Bhni~>Z8+Q`xy^Tj!63m`leIvnyeXu0tYk*)PL8vd#VsV_~dH_)F#
z9h^Yq#fU>cCDjoEphgQvNBrKiKj`b6<%eKJt{t+z^$FbaYb=6goOw~&sG=2$1tOxi
z4@sirY2hi;pTd@mx%3#-61O{$@#ugyb#g&?qX$K|`tr0EP8{kvd||96mm<Up5tlH+
zfJna@w@DLn3%^7GnZr#=1o_m22CKZhyxrk2FVo=9Sfy6kP{nNCAa3LLS8KTFlv~uk
zt7AJbxNWjDLxc_jv6jNk*?!;V^xp5^Qzz`<{sf7w0~$f7D=5pmJ$RFrbCzcTA7JXM
zm3f995D7ewJ>Y)<3;Q*QOiyga$$FLKcY?Y?;(o1kAOo>zi19hC1vHlif9874Sks$N
zpojaTCO}<v>{0}Hh;)$=W`8QbA13vUjga?0V0WGJFnZ&qjA&AX(p|$?2oh~$sA4Qt
zt~%^nT3u8`!5$C0h-8sM3kiO+5AM5DnA?fz`J1X&;;a|3`Lx$@au^aPE0ULl%bVCk
zK_ot)x+$mPH9~++;5NdL<V|_GJilAV&=0I9rAK>Ll8d&H*K>*$#XQN~h2ygszK*4C
zJOYaeHVnon@`9qWYjJrt1%3)svt)lrUe#rO^)#l8gKqP)*)x`XK{%;Hz+8;e89{SD
zII8N7e5#m_A(&|7eK#Mw4M*EL(Dh;2A_IOUtkHO2ix!D*jcHe5PVEQ>N58_oNkKas
z+1vQ}<OMeb4p4Ty6Un0DLr@}RSE`;2ZNWXpC><6VAE(=xi*OGBmkgsH*0`$l+x^jl
zuQ2cjLzM)fqSSd~-{C?$aB=yP(;S@$<|iefRW#LxV^#Lx>(3&y4(TJutmAPfTt<71
zYHSotO%iQ_1;FrXHU)K|1q`s_Th7-<(?cd<3Z^~aySVj{TT5PQ7P=n9sfWru3*aKD
zvTU3>hCscVhBMQ#&k8@Z9N^CgQtug*mL!<#vD>^>yQ!gJ5xRahxMksTAz>!K+b;mF
zMSb82_!g88)kc4pbE~LX@6Y|mTMm2Q1g|A4>r&o?C`Ubs-wn3W0gh6PL;$n>iSkVd
z^Ii-d`C0+}cX)J6ix;D7zP$k^G2wtc^V^HhN*}2S?%RtQnXsCG-Rqhr(c=Bp6*5cd
z=&oP>=-i&ulpZR6UVrUw-3K=L_HXa~?SbZJ5W2J&LRx_6vp*Y8TE0MP6P6`OMu9{s
ziMbk-U}O%{6-@U%6tzakMHV6np>r4jQ5y9Lg+thaUZ=BnaWFo`%C&3DaD6w-1?qbH
zonrOGLJ`J-C8}nES~-_HKORz6X6>s9zjgceZHZe7-Y0u;XhDSU3$Qs0qcdm#!{Tbt
zO91#V0Y+|Fa~hi##FguKWT;c-&R(RLsCjwtX(>Z1tB0oAgD-Q`kql_39MFn=_5`~D
z=Pprx0D!5$HUX`l;5=}(8NBY#pcK%@eiye|xW|d5)lELYnuk86F}1`-ocU28+CPQW
zGBW4-hu!F%Lf8`|H#WAdue<vij_A|A8yFcG=jSF3!kM|?yi1$Frzo*^FBS9h`Rn8f
zPY3iwuZoalhxJv5wS^~J*Lrf}8HdM1KZ-o)a3Tlep3A?u!fOS)3XHTMiNM#$z{W#9
zcytz)*cPM`@}Y|D10jxp%ZMyDL{_roh^oNoPoZPWiWU}E3v5HzA(3lkyWldAgTvyV
zP{Ud1lL1>`qS(h$c*V5glPFXYC$p?^QlRJs`Y8(_+I``%@GNh2ka@LQ=R4xK1}-8_
z!bgaPss4o_n3(*nztGJXRGt*@>IbGeU@KtMto;!$kxm*%aoBHs%|-K4$B!esLQqJk
z3%eBr-~7yi73cX7`GW_R2{jGyxoW^_ml+m1=hEb>?W@rbo1Gc)bl6XNIlK@Y2$yhp
zUcT<O0f#P1bro=Gh#&~#101y^ac#St?|_o=FCl~kio{&KU@6H;K{T@ojDB#+pwD`C
zdJbhfnXLi+6mBZu8ax0EuxO(`uGC{l>3sk@B8u>La4K+t-FSSzZ?yS@hDM=O42mck
zGE_-*JBk+bI&w>4Jx~g}5P0LL035;@{|4vIvk%Y7AfzSr4TQVcZ6=lrqW$~q{MX*z
zChKKrZC#DpSY$%%FW&BUoy%+jEVG7lI?$`&OIg+}??lpaX<gkNz|!V(h6^Z-TxY(E
zcbZ6w0cs_dxnxy9k7S+=CbZ-HbU!=q>6D}BhlS^!m)tjMh&hJFKp107Y=Q9)c_Wt&
zS~8>p-Q^jV5Gc_h#UIj0i}B%bVOsX)HpPNF0LbwYq&KN!D>;sgqN)ILIJ)iSjJ7cS
zl!zK07KJYYke#12P`K64-~l*Kw%ZetC>Ulcz!ZT9dqX;by5i?htAOlxXmyYyW)Inb
zG|q;v;DBQ5V=ZK0$0FrDg&`dQVz2T0%#jy_9jGq<7RRA4Dn`I)@qK5wPeoD~;E`(D
zz;)Ual{QT7?kvCHMWk$~wy)!6Q$<!Xe1LH0GvU;Ot|<g}U&5PrbF;JNb&=$28X@95
zXqVPqkSRz&J0_zz1W#R}F%1E$DeGd-5mej7U@Tp<x9`VYLQ?naKtK>be*)i9_}_2o
z4HhgCGtc1J61^XwRY(Fz0=wJX#5&{$D1}R?ry8Kbjyn;q01z?}@wq7UOA#)QIMEA0
zWFTKk?0ZQ20NVyaIP{A_&}X54sszNL1#AT^H~OYYawF3cP#i#xkf2~EJ|zpgq^?;$
z!0;rF*v}J=2M{eVy3pwDIrW|_3wWs~>g1-i47Mcpz6-sFTa!lM(o-AcsN)<%T-PxK
zHxlnKRBf1BG+^8o(B@+O-|&KE8z#YPq}w1q3;P{rjfJaPg~}$LB)CERs*HX!2;gb*
zN1WiSxJ*d*Pl`bDnjok>hDOHi*@t#myU=pu9BOH+#qiFg_!_+e)CVN?9JUIFdGudB
z7e@2i3;l|=9nQQ2AkBOb^pG+Z4R;zE8f~PTK3Ii3BXxV+lL}}+S<(N-0*y&Or~y%`
z8fAjp*-{H<2&**S`7!cwQMjp9A;v9D>EWK^JO%}?(08Ta+U3I^R{>?M;TtIzVsQ=l
zB0C3-d^%b&9Wn+C%R`5$OrbgFwfKNQ^JH3X7+N&&N<lJC%MDIbBX59@kFx9hv{@{0
zGGBDKXN-)ZoF|$!d{M-XVZPNG!TzUQ3(P*~Fe*I_6QN)*X$XAc3ui5|D`GL{sKnE=
z8(xi80^8-Jkhr)8j^H3%)}SLR)latxBkS~QM`0lZ&gp0gxdCYrT0#&Zr4Pui)*{+-
z*3R2*xP}w5J6`Grpl<OE4Q*qT+`g@$75$vJKCz<Fo`Ka29~fO4_|>t1I0}F5n-luV
zqfmL^#9uh^FBTBOv5)!(vn2rN{}Z<PU;p}3L8ku(S_3_gQ7~Kk7ue~&OX+{%uD{un
zx*qI2DXG@@A9Pjbti{^CzR;EszyC|j_1}MY@wWf^TqTUpu#r=C@uQ5Wi68b2!Fgt6
zV(~VtIT_e<iP^cS3?sns+qQ4-R<)N4e)LGGeA(s<0X1)gnmZ2;d)GE0#FYEo`IGVf
zo6&pCH?=Fn^FcYv`RZV6vb?kNth|#R6J`1o6_l}sQ87xr&o6F+@s45IVJK6%etqj9
z0Va6kkMlPE{npNF;pwG#^89)2moLFtV9!1P1ZBOXq%(3USt}kteueC5wA)CfLtY4T
z>Yg&cEw^Jm<?`G<WLqb<5SokCkmWp?wSyT_|5Tli)bC$a<0U`rTe+~=ZDCH_)z$T%
zbi8DEvwl7cZK_(T3O(kR+-ezHF=j#~qG`4t4lITkM__F3bzsk)AU>TCRA{!qKt%0C
zY|o#?NK$I;t|Ht8{+lIMtXnr}bi#eA7Rx(cwcH;S{*Ur&^k=Jn3D3rIFR^M(i8|I(
z#%gz=sO&?S!Wur|Z{_SDD;x^DLJlmD?-TS>H(a&0=Fv}KyS-u0Sv1`i5zQMmUcYAk
zty{)pcACfBO#X@2Qmx!8sd?>Ct^q1V=*jU8wT0cz<^(KekN&fE4re{i$HTrf_~n@Y
z+&EIsmjN>VoFRx4?Z{DcZoGh+FI!wSkL&57b2koNMDzD9U3Ws66##G5?V;9vkY0Ux
z7rNRVW!na^b7a_%fFl`VT22hhF*3z5<Ky&j4>a|stmEAl=1Cerx=nipes@?ak8<=P
zpaYDZpq2p=<ZPGZ*MH`XuA(Mf3d+Sbz~0r;)R<mgqi14f<^cE{>b$nZO&tu@U_SL|
zh2|t}k9@cJ>6`FmKzW5bV|8T^zv7qgA4y&m0e*XWWOIkxn}3s5EGkO`vcNceG*$uq
zM<Qmmt6*P=K=*B;yI3Gc?EF4*8+<`>BjNA^)o!qS|Gw@lap(7fkUlJ4wybYR!mep-
zOZ&?#%VyrOb==b9rqwVrGCg^8@lJpCa>Pht#6qn7ityboSTuQ~e(m}8O{r8N7^stx
z@u?M<1;wflwU8K*A-njK56b{Kkt&a0Cwu!o9v^HDTZlp9PCh;{x#eI}-fAu-Ly#Sy
zqVnrQ=I()I%$LTM_U~Lpy*nGal6utD^9Bc&$U*ACojdM6KF8v+CzF7Kep|vCi8AkB
zQx}6*xV{^!t8T$zz7g@l+0O3<>QQu#h2LKXOx^M-sOQ<1EniRpk_y1Kw~DR~#uqK=
zI^{3M``_Ld=-5BG6R`l0dWRk`Spt)heC@R9`L4;C5!^{%!#}Rmy2Um6YUhp{=>@+P
z`0Xw0+!3EN+Hc^7N;oQneg73dubLW^(igstnTS0AmSwib`cuVN;X)ihKTDU#UmHUq
zHkTAJt=nV5I`r}LMrt@!2in=!fluTCm{u_EXYVe=d~~C;`U_$7t98D|(HheQ?6FQb
zmARb&MLit8(RI@g&l4JO;qeFp22wz=cHmP36!y@_wXxg}&3iQNw@(Fjw*XqIM4geN
zPJ|yghC(PoKVM>~C0*kqG}_znBuc=`(0gg0Jb46D2NQ++W!SSE@&SXhG+ZKd?kLkZ
z*d)#_4q3bQ(LtvWIp%dLSx1(h8~+k@)@K!v98c6g33kjM+Q{gM!m)NU){}nb*RXo-
z$j97(oGn{ed34!^KE2>kf4Zan)vLXi`utj~()9B;C^9q8{~kYysWNu;gBDpGWmYtk
zeo$0QkN0u}++J#+p7yR{B?q!vS}cc%_!Ri-U~`g@MkIxQ<4vZeI_xWwmt{x5@#}?c
z-wvFED2$9Oj68IbZaO)lAX_ixdca>j`EGfcm-E$em2?|DkYYf1zN>xy_cIG5g$X(~
zC5T!R;XXbMJ_!w(Md6zQY<dh&*aU<0EjV@FJ{Dky@b3#2pT!4~RUq|q_&8wv%$o3x
zGcuk8oy1+-oV0H9W*oKjVCr_^?>NtV&)v()BcrdQW7(1r$n|^qil=7Ftq&gH)=OgD
z)6x>et-lHH6daoV@R`5YkEEMqW4y7sc|~rAYqEM8_Y!OnQ;fB>dr=6&-H+yiM05WP
zq(?N6`AyWg`1t#q0!95lq<wch)_?!+g`%OfGt(eR%ZMncL}f&XB2>sq60%3B5E@1z
zGn>pZvPEQMWRnqPW@Yd5e1FG%?%(<6JRaxV|J~n;>-v1&<Mn#37bv`S$Tf>M>Of=%
zT(}VAjC{}iqKZ<hBqc&R2Da6P=`Sdi5JLoFqjt>_l@jJ3dqQOfY^%*XCI_j?+O@?V
zDB}!PE>jRDT;%aasJ@dudohQqy0G;PcG<Rtv66ZAj4zvRRr!AgxeR@VV%!26j=DI_
zVAO{C>z=`06((Ntj1nMhr+V<PO4@-A`)M}y%=`5QC8S?j$3lHLuh4L2cJ^gJKn3oN
z4=q(dWpyiqI5`BAc)sYu4=U7QTZrx3jLLtWeTJpaVN_)Uuu5to^51b*g7;5UFqt_@
zbR}Iyt^uN^Pn?|Bhk}%O``OJ;(0o6(pv>Hm5MZ@^ifIQ%`IgU&Xyoj>n!@2k%YfSH
z3GU_5#$bm{tgIEqvTuN(+s^va(&m?QQ?Re1&~!oBqBvidrzu9(B7hw1!@}w>E>0Ua
zBDd~61hQw&h$p9{oVm0`HT8BW1VG}5u&dkXiea!bm9J0FEG?3$AE;&BJ=__O-Q9`t
z+gl5tFf<B_?n=eQ4tTI)*0dMmiwnzCG?xA^P9fWD&l3i&6DI-yeN!D?Ho+>R!)&mE
z99b41$1BjP(_NWq)CA2D$FA+JP&u8MYDk29fLIF?uAfE%L2(nWmfpaCR8xBn4tiEc
zlm)eQ^#p&djkmFQy6s>YPJk<!+N&WlshwMSYFwi|+r&xvpc)ynfYFgqbkgH{k}*k;
zNr^9(+y$Lx_PhGQ&CgG3FiHHK(QP7tMc5`7;dg83!*7&1bsNq7vu2I4cz$9ixvhHC
zGm)H2-Hj&Q-`Ym>Z|<f*7+}yG=FGs$$M=m{_U*nh7BS-}!CMo0J5JiVj8ym;96GQL
zmR)2+2qWVUH`Xp#(wOe3Xv;#LMh4D<%{(9X9i2{|%DeLxc|}ojHI0oW`*kwsE7^_%
z(23+OIPM5wK?}r@eY(_;(a1VSXaTjfwTsZyyvCw=!*EgC3Y!y+!Xw;0s&Uz1&@hYn
zbni)5?nYnp*g?al(I5S}#XfN=RQpsvM5`j(=;E5D=9oRyWGFPK3e4#NHP?T;1fz3N
zRV4v*gA0@YuGz>%YDe}5Js!FLi11w!5eQ~8J+$Y5fOl=w#Y+jcR-nG?LsS5I)H}8U
z$YeaBw?U`eM+;z^Dnq&8@G#M5{V05rY&d;~La&fU2eUd9P?OXvFq)nLG-#yS^e`qN
zM*jjDLqt^YxbXreAfv+rb|)GBs|A2HAs>~zUiXiJcg^qXGjR@F%9LR`Gxht&6ijuk
zt#RW6Mz?Qcgm5WKRXIqZW9EN2&~m>`&B{`3#|Q*-|3aYjWR5#gy)3ai$j9B1wV(Mx
z2PzbyXC&kuHIJ8-z5Rey!@@L*7gk|wZ_F2`lFzX=hwZqpf9R`>re;q?;M)9rENo0>
z+Z7tBDqftEogJ+nw=K_|O&TrG7eD@}mMJ)*XCS50^BPb&j5h4r8-W{bEvW{Sy8s#Y
zJ^j?dqa00j&hbuLRxO6ap^S^CrF2BH|5g5Wji(v*UZeyB2JmH94XLGH$-Bb_;bi#5
z*nrEp@jhX#ToPNZ0PIn>@d3K-C_u9lgB@bB@`qE=^t^{c4hnbfH+&hLmqE#s09^nC
zV_$_jv@@!98@3>4opQ0VlBr!62~1Gjl1YSM7OCp|3?2^+r6zP>Qz_05Ci)$_LH&zI
z$x^Gezm0{&m?<S*azZhcpqZCozr0y95=hWEh+?9iPfj*@G6m^`@c80lDQE_Nxy!yd
zjDhDoL2ijn;U`?(;kQ5D%P;SpAVh<1e{`@z51uj9Oz2cehQ0Wc^|Jh5(<Q(90|ZMl
zEqP`k)<^HjozH$rSFTFH_<oEhCsl^|{FLEz8`21Xp$V9plHp01l&rF-WuzqzvJwi7
zBaj8sEs3p~0)MKHgPvL?o;m~&o$3(&;`ZJr>#)6a4W+uSgO=?4b>zxdx^*64Mc?09
zv_|<idML^%|8m?Be*Umq@v8cBHGG^@w+2=e6G2wXM1O#Yl_z2r!D<3%IdlYWtH3-3
z1b`p^TByl?`vfLMS^BZ;gpv)|s}`84dS^g<0BX?tQ-|)_r8)$lsJlJVL-{J0H%f;2
z;49TgSRoubPR3_q-)Rq~8BL6IEa&`t?#O}Spsk+UIAadW?b-SDY-H9EvW%U)nMD1K
z051^C4);h(XO$RXJOLkp@_TX;Q><IkN}e$}fzlRR&+_U{AdWez!(Q!c$I=={V!XKe
zdwxkQQLlV`Oy5&A;p*i{3_<V0DZC(Zhn4IZ#NEL##{j+^jo5rknx=)ZUUZ|eRNiOb
zL8HVrAQZIg8Bc-sg`q<}uQ4wc;1+tsQ2($!D2qtW!$8PxM`#8q5bDUHUyMfQMxvwH
z+1b%PeFZ#YelpI(-F@Gcq@6-{zIS378{1?UOGr}e&rXThyH>AgYM+oa><N{_4D$;y
zJE9XI^3QB|q4(A4NNO#&QK~^RU)z0x3sZoG$zU+WEb=&Gu7pwp`LByn^MyD;KIwWv
z<J7D<1Roj1aK$O@eLjyprl+>Bd{-9OJS~9T99&%8KzYC-m=CL+IhO@>1Nt@-SlhNp
zT&f~-dAY~@>55Xl$3rhg%ZCbxTh>qW)hl!&o&(|dugSvEF5MIm5E!Etwf?9<Lml(y
z*@VTvrBK0KUwuyxszmY(+yjCI&ww0S9LOQVF1BT=w)R#85;hkfS2cK7PNsR-Ee~eI
zv{vJ!ioa&Wj1ylMN2qP@ug9sW=9zZYw$9b``!;B%VKShvq80xOi+c#fvYhTa4J`3$
zy1gh|nFDFM+#RNX`I|G1-iRoqre+<7{o@ueJd~BE16NMR3bnB?4@|zf*Wr5Dx2Ji2
zK*>ZMzT>{0UnGE<>4f$wag4Zc%LEY`0;SP+S9cBp0dIWCanKRwmpgFKP!T?*AwkXA
zG`8k_VD9V_RIABI^qHF&dUJ~-ZEud6l2V<`!HJ2zgk_;O`+4k3bd&dIL0nnqa^@y$
z<9)4{ghNa-qi_9KzYK}<;)WL@8NbkXz^wEdNY7uOVubrnGZ6ouid>Zvkx>3@rN*Hp
zDsT_w*^L}54L*%UgIOb2vTm)T1gU3+XC>-CV-+dK*CujP0Ov&!6Ubq6WuKe4kRGp6
z6_=nHXvPhl^s<{v{_=DO1})!$<Cn~;LW2lelySCJ4*YtKsn}n*Oi^z?{NEJ)Kg&7$
zK}U@25Ka1!a<jZ4-e65g-M9II<AA!&`dfUsq1K!$lZt0v%3EH4YFdfs%D_!92ld8d
zIyNAuu#)<F4m$a+y-$D!{AESxbzdD&_wSjHxLRhx_dLz0Rz`8;cYi4o%|C3*kvX_$
z`pFaP#auo^SzD_*UE{K&FN473*T!g_10U)ay0tpXd?aCM8FX`Bb*?C|!eT*V4Rp58
zz|)NW`S^RGtj>yPxXgZkA&Ppioch#`F~NCg8V9~r30nP$7Vd8Mv64~)%4<6$t|Tre
z_q<_v_``h*E~KZUzuSP;8I3nrJUZ8|&VuwE>!N2dP^Hu>T_gy7cW!p+Z6hnh;nB=Z
zbD^4Lv-4FwW1C$mh%N+Y{&CcvucIak^Yc4%stm1xdon_|++F;<7mWh06ld)5ZYX2W
zfWg&2zwe)2!Y(ICZGdoK`uVXEGbb*lP^MphIDXOnsv}X65D4s32w(N9%-6x_xWAd$
zE904O2Xg6Au6##ewZtVeS}Mf&k>F@U<-|xVPniIL)pm;?emsWujwXc1UGMsJ%u4gA
zwE2m;X>!i)xlf8j!;6>K@8awN_RX7~biH+IXkd336UNO~;ywbj$pp?a0yhRkG7RG4
zg=S7YRtiD&$#OyA%~Q!id7d<*VrSlO8ZB^H71-KwD>`>?3+vc=Bg6w-!W*`KbHXn<
zZ-K-oU@TDisG%WlGa?!WY9VYlpyWRUrUg__I5)ef!4^bam00x10QN2e`o?(tmV0I6
zF$;)NQiuVNg(YrDTq#Y4i9~Qknlj&q-32GSv8i674#uZ|&4x{p2z}nEr4M^e{5^!y
z_KJW0cPFz*j=A~y@$TwS7!#8aywyf20d50kuC~l0(0{LwdxDwx^6|TlfMo*sM`^}e
zW!`MQ8VorWNEH1)`Gw^HAmGilhWwAp1PXd^D@F~_FHvD=`C7}5r@G@111<xjDc)$~
z{;Iy5m;P9{;v3Yp5Blp~Q4*<Y$Kw2!d6@@f+F69IWESM%>yViefS@8xv)=Bb6h&1e
z2^=|+kGLC1OzsZ(DBT)e{<zvaK6C@7>IMb1A75NGA;>|QVfrR_bsvX0r2_0INDPdh
z=@Gb~05Cx6^O~(FDk|jX(Gaifv=9^P(q!UHhwDOW%i(95WgouhEiHI`nwt7;qqA}C
z%<@`|@ubEDAtY$JPll-<r$k1nx?btVz4!L~k5Bl?XAk}rOWa!(h3Xlz%Dn}5><}<)
zpeJrm_z@yK7U!oTgjJ}xKXMAhUEP}WK8zp_Sr*TAq<szQWzU}vxm^Cr`>@KPuU2I)
zNMkc-Oz^N7Y|wvb0q-ZSnYiBH;#2>;BHQ)cF={AiU))jyDOu>eE)30O#d7!a#z<e3
zC%sWG_w=_$G2mjXZn@tzW3r@?%Zi=tpOG2atTOK8@L0%wiH2yuc5>?$NlR#&F3!mW
zSI(IaG9o*oy}dorC~B%vQ!(4p1g>X7n5CvIzVd9%CD=@J>vxuzn8w6stTSEZe=-e)
z5c6178JLVSPF((la&DPQb4!co2KFi-hDTTWNrH^T?xd)4*W`}>q3*;-OQoIqo?+sI
zR_=zj_M=)Jo<c6f$OGeAv+cm^AA^hPrCU0y<TujMQTG0C3kFJP*ej~DR1W)tM?u52
zrn%J=0}Zz=O>W1UCSqVFgFKDJIRvZwpe^dI3!zfaTYQB#a~EN{ujp3DN6YUzdcG;?
z(w<__tXXMB{S#k>24X*x-uikoAHnY2yg#3Xrob7Ae;b$Q+4E%mAK`8wgj$<DJ+pud
z0FI<vlLj|;h(%htZOgV?3INjk9?cM%9Ta41D4;{aEfO_AH1k=NZAo1h%*0QO3pu&V
zL8CNuAMVjM#HzEvv(VQbMhB4b2eMffK3=B+axpNF<uB+jaHj&N`7x6}9Plbbs<1KX
zHWScdt%XOc2l%Fp3(=v`B1IVLe88&bQFRc>7VuDP;Ag0o?Ae5dou>a?B<mFI?}G0U
za&o4xX(0m#p(81UjVy2Bk6Yy;k4o&>xQ2LOdF;rHBXRw8u~)rtp8td^k;nJ&`x}L{
zBoX_!u&J5-%X(X?Lx9~_=CjayK_vO@XTm1Uy{LvEGTMsG*|6p=J}xf;H<0Z{)H+(<
z{5<NBJBarDEClYO^NSY#LWYw3LP8*ZlX$MYfKLl@`Wql!Qb#{wO)*2qd3IsM!F%7O
zuAFVLTdq0*14WhGgR=63-P{&v3d~#S+1XL9h{GL*<L+7IqN(at-s_#GDF1h#jAC0O
zNXw9roeu?XlNc1Z(n-K%XY6G-Lyp|&DiWv~p!x9wCu?=mr7(uOf*&q`!u~llB#Ry9
z%)r(KDGx?8JR_orWUX^ps$bsu@sCYUbHUte$ifm35P-ype3o(Ra6!|C1F<3^DcWN<
zix=m3usVNM1qVa6!iOQ|=6(90ikm(fZU<YUT{1OvKrPdoG)y6<sB%Z@-SEQV)_UN~
zKVm+h1bnjue+Hte!jQ%h&JQAwf;dj_on}?_E^#t+2mmuCSJm*>dkDav1Um9+^1X-#
zDJ7J13xZaBqB%MJjfuex1D9Yi3PYkIMpFBIH0{AKq;t?9&P+CHrY_ysrx;0p1FB7<
z0j0D%ZDrp$r7@O<v|H{w_-TcN(&>X6RXl=^Jm}(-8lC<FqLbg5APPoCvRL&qHlNqT
zYi6C){-^vYxtw4547Av$jT=ux!Pandr+>_K$m(S*mVrRWoPPGeiI#NUJ-~|WS_NTX
zOG-(33wZ&?Tpn(dn8g^X2SQwh4p>C+nn!|7OJv))CzlnWsYGBg)SAY7vuIW7i9b_D
z=8-M<EVXTUFk=nWNkM8h_2|(N<gDUbKs`r~hX9_F-?nHKnuhq{>9lFuQ!rzo#>*IB
zT*;FwgDb5Vb9j+bfHA3c_r6Z1VuB<FYbBa?K&~81l@A_dJkMnnkw)M=gADc37E&7U
z!?gsCJ_KduVcofdhz3TIKlQ7FTEhXtP-815-;}Uk9(AhhFTxE=nc3Hv81&Y!#IN+G
z8HeDOE`);Givj_?0gH}1^U(oCi@)XLwxPt@0e$m~`S)Y-O#6aafP>He^3{a>=f&{T
z!mje#95o>l2j^9}rZ(6O*$yV1`BVtwd9^n<WQGq;ECT-4AiIKEB@sm01r-!aK&vAi
z1KH54*M4+#xplg!V&{Jt$S#_hNK;)H6$-Jl%mvze_u@3i0)x6`cY#=_PF-IOS&9R_
z!JzWKZQaIPhK^=zYU)*Iu)NQ{OAn3n<}sQr{k_nE--;Go>Im;TVLS5z@Jo1>_1#;Z
z|1ovgk~jf?gPYmetFZe{np>DLBYY3F!a|T|wZ^J-P;cW@u0kt1V_NV~t}RdiMPWY#
zODNrRi@v-J&OF|QXv*)$Aj3O@%G6c*YRt6hJ=|@#=H|Xlwi>5Kayy*=m)<zz!VrGu
z^V5nzq34r>4H&!hTds}mAUq7af&TX_kI`{TdC%0fZ43^@AnVZ~iw?4)%rEy>oxXRE
zPc+k_7X|zQxvQ(yT<Q=GOk$kK!G3Of827Nlz@ZqdS&Ta7hiFI});sngb56Sdwyy;H
zEo+VB7QW5Mq^w`wwlUl(hid9!ThrS8(S5oF3CU+aZ&OQ}-lq1pX=_&WM%mvhZ9LBW
z=5(!j-IP>xWdn<Zd`1U*3X8A0|J%1g(^j^dbSv?@z8AFdr+;DnZxfA+_RAIj<JTrx
z2(GyDzuDncV?Qo1ocb@)_T&xE|7L)@Ht${i|N6&8Y6v5H3#BJH+Ta4WxF3W{5Oh%K
zYu-ITd#!bQH+CuRUY-9uADv;1o&c`TK5-R^zdzkeDQ^Wr#^EsaI%=p~@%1Y=<R&OQ
zB)zzZD-+`*pp2#m?g`26MF<YAAVU)V)w<8PZy7u!y?zEG7--ruwac^`r2z5K;ixPP
z5Po077_kL}fMi$=3G8!PBm^w^X@Gn&eg~(LBz_vI*lv^v6fA87ia*Ca1L*M4<HtF*
zgPVdQfvZ+y2MI8*!^)NuxQ{XxT_hU>HPBUat7i%T*aw&;xqZBZ-GO7!Vgy3>_AwtO
z-MQ1S?|>2{1`1^Uyg(ZO1KmXkD~Z(r0B4xp{FHW?cWGw^z@H=tUEmKNqcRA*UuyTa
zN;23VBIYF67I9ob>uZEhPD4E5kTwx33w|+n1r1>mqg7M)<4Khi7SbTh2?*FH=qE=J
zmydV_6a(uWQO5AY;k=?M3-|(r7|@s69B$-3$)J{Y`v#CMcZfnEmf&JchahM@>#^%*
z9eB1eSCV%_TViQtRg>r7uzUAzh$?BwNrh4a-~#2kmR2EfLnIpOA`S*;Z*Sl=-x&eS
zNALtw&dGBhY%5N*HBjilKTaZoasH7YUyP;nhc$-k1(fV(9RN&Z`GQDzi91l~<C86l
zyXzq$SBsDf)}V@&4?Po$2NesQH*7S@`|<79$OCd!#ETI)cE92^FUA2&8LOQ-<w#wJ
z;-I6EGhDcM_-`592KG3>gZoeUK>v))t--XC;^J5+1@@i7^OsJ?<l3G!^6(Z^^Kc0W
zDE)#Kg=zpd?1dl;Pz7t`@q{ntxNQv10dBS*=#>wr^tPNSR?WM2lTy;sl!3h}Ix>rx
zMYk`(D8cl>F2eh%t7{FLDJskyJA|w5$FmxM+&jkNbeET)#qhP2(E#3}iK*ulmSv^h
zZxA<^Aq>0xa<@bsI&rgBm=R(I8WR&Ksc1hrs;r>%*3r@m^A|MoH)wpRz_;R<&p`y>
zgS$dU+>~o~oM@h4>OoMrFX4s(H4(xN$s4l%0zbepSeFDJGwGtKUl4<A{Di~Og1Cyf
zoNC75zoil3$ZXgE(G~=3qz%p-V)l3fj&_YBpn}0PUud=PZfe2x=G+58M5d`zu%x6X
z0>}x7Pe@2H*aX8HB32bZ(EZVJY`2DnhV{^_xfC}P=?|f&?y#SauVFM|Nx1qxvi&~#
z4!lNMl65pWMY4FI2|xh!@+*~HCy928ENr4W)6wzK`;GpN<mID-`rOc<oK7xDEYxij
zsOdSXn(?WyWT#KJ#&A#+YXDU<fjmN0gxVFg3^QX<xU}yx6j1zl@TelpmmQbkRnZe*
zMgt~pX_;~7_RJuFYGFvY@UBRh53DXw$E~KP7dTdhcTd_MAd#F%F@}aoG2xmpdb9{3
zNwyG7o+HHvmS895MPXf|r{T_=LjCZzMi1#6ztH-_iAJNQriRA`tK3o5672sUh(H42
z0qLD~#p{>Qvtq8-eRY|8|5>eX$c9JF0gU%%lH5GsNvI(+jo`Go0SxrDKXYZ58HU18
zNN3b1401pU@0sY#i_Ye*d0H37_DT)q!U`>DlfHlh!(68zjaGYgq_I%k5B&}`>TO{>
z2s;7}*k*7aEcNQP_IOF2dTMrUAd9IA!v(Va?+`5_G=7{ld|-D8%Y=2g7k4Iig0aEN
zj=2~;*kMqCd*fMLLhX81E6aRP@&WbAnW6NKW5`^<X|e~?18{}&X@5C|Z8_{4-GC=U
zWUk;Hv_V25=raiK3tOHoXSFTlgRpmqKz<bpT_f@N!qDf*t+<zX8nLt|%8v5$@4?I{
zEPC)Jj9{qIEXvlTzVY*;N-X|h0Y)8X{(FRelEMtDhkk&JUctsY2LcXn=@!llnAIMF
zu&j-~v<H7{-HUhgf15;$qG0OVj=2iY!OU}taN7ZhAgC3LI!Ez@Kb4oe;fUMI=m75s
zG#&^T*p70+a{J}&!O7qVZ=&kO9pXL=VazQY(CG2qVVmT*Qt3!V1F|$6ryY@HlNJ~8
zkh};>pr)oK^Hp(r)E+NJQUUhEpGHlG(ocfB>q-1Rwg~vWbyI%0kB}CPYsdBJQ$A~3
zU`$wvp-{iUOVq8dB4omTdvuH|8fv;V@=zd^J{^JUh`J65D`prUQ~s)x)e$$-{(bw-
z8+V}0qVGA+$b0h<bsehMb(MG|(kr9E<n*Wc98^=*7htZ>wldwy!t&-_W=T|h7@|!G
zM&6!_;)kU!8M?f(mX?<5K~DhP#lnTc4YYF-+SIoDTGM~u<1TVwMcpwqO~KLs$c-}(
zO+bCFk$2pGEhDL>YH_SSUQOkyswzn}iMr+USRVB68hJd*Bbj1?g39=HG2lUBKoxLD
zcw=jL*9cR56bA5Ot&LdaAstWzq^JarLb7TMZ3@pRJeR|hhh%KG0nK{V$La_Z%z1(B
zmr!f3<e14qLVoEQd0aQ=edco4^;5lmii3>K5rI=FUGVTzK^1&IrK_#2jZppVX}BL*
zyp!RaK}0eUdq9O#KVORnLdnrEWA>np*n;&ism73q>9L2PcC2ZIqWrB&36vanVUt0D
z{1CXH#v)n*LJ2~hxH^HJWb?1uRTR|&Q95|pRlt65#TP(jyUE?CJzE$mO1P-JNQD8W
z^K7q=2(}L%Km-mv1T*|nSJJdChMJ4vMQ2hwmi0ZzyKL?3HoMnkV&po`tbh|%o|nQT
zlfkiLuM(#joONtyhe<s{TcTfH-yq>q=Y1vUEVs|~3LNQVyeoi)3sWz;ElTfFuO2J@
zP>4%h%WG?*2Ue$gN*79z+exy=3u32yuV1<HI-%9{rnEG!kgXxOh)ZM7-R$C8rr-dX
z9Ec9Autb0@=<hue=6}1lyA;QXD+3ZMVrE1kgb^-UZ^!UG>lT+mvFl^19rny#xKja%
z)$K&~%}3gj)h|`a96(3BmV(?y=OuJ&BBd#CEmr~%#9>@_aE_`Pq}aWlic&}*qLt&B
z@k1JpSGKonBpx~wz74e)UTz8aLyyXWxu2z*1(-eBsduU3#Qg=d3Wg3o*>^fSSLZ8v
z^Qau!IqMpS*G74P=dS(PaI?NEbYnC-al&CP;)#L`tU{*S<;bH?8F<NPv`K)sMFcmS
zFK7!cM&CM);(s)Szex@mj>6>BR8GXKAI_cGEm4BtEIw#@5MY5w`1sS*nB>@t>2)um
zPF2Jjse6~X;Z#>?I^0yy?w0}XnUsvzw;8|#a>v<(7vY4v1RJEJwDiK^*{FIvNq8PV
zp%0?p<Os6%Tb$d5n275I4@t5S=B~j)7?M5$p5uGA)n-_a!alx0ML|;y-Tt&fWB7UM
zBU=$=o^ri_8sxqt>a5M+H9UxLmjdvWr+y&htB4qWx%K2ggzCqb3u+~(7^0d**8J5b
zj3vO-Ne?;~j+t^?q?@m?{lS9<DzjYJ-W~*0mic$%`qOI}7=)U(Z`o3WD-(VI@d0HG
zmt!qI(ot+zw=VtpC9Q6AP%+ZIMhozC6`lkdMxsOk(87`N`V2@c$C||!Q?8>$S6saB
znVWMWL=A>XMh*%lA;AGFka>o{>uCOqVdd+i-p&570^9kY=7Lm#a}48yaN8B*tHVWo
zROG|7EDjnZ+?>RAoL!vmSU2VD?7Vs7#+L@YHk(^6mw+H&;PzksR|{Zdw)6x)ey<H<
z>fclwI}{k~-?00UVcU$#_hG4+<<U$o5V6>q$w=qVvBN=>xhuC=`MM}{=M?Z_7eBYG
z9b?C#2!!G@DlwwLGLrK>#D-xllFJpCvEc~Y`v!RB@ft`<4t6-fd$E+619|}*Q-|O!
z(9V+(w^NcYUF%S1kro5^e$C)7W~h#^Bq{C*!!~d_p`59$8_^)g8XU=2GR?S2KLMpJ
zltN5qlW;!k;KZ~Y|0;wSXEF{rXAy-FK@eg3bS-`JOBaMCi6@vH&KjA4QhIU<syP>Q
z5`V_4#1zT3UJjM!dh`@r;HP3lt|p;w0cNxZlXGy4<v4Tl@p+=A$M}j*IBOv9p;qY4
z{imH%Ch+S52vmrpi1jy&7ot&O%67zGRZP?n&0Y8-O^mhpjIJ&&*RdQIaHIo+tA;w)
zi4WUj#VK@69c`aos)C%I>2SHRidri*vPgy1kY+4w43`kCMrwJZ7N(GiW4H~`17N#*
z@4x&Mj!ZIB0evQ9?#IwN)bsG*%@NxIv7>Bj*|M7#>3LnSQ0K>G)8#|sQigfpX1rK*
z95626Kyw1JU4xmFQm+nNfC3*Pab0vg*Rm~NQqj=LBTC1~H2iZ%jweM%2Ys0irah36
zYpA;w0V<Qb05fQ?`gdZO0)#X#;g7Y)L<uJ$!f@b#Tnfd1-2xjIK4VnIdbAoSm>h%c
z=Um|)$KK^9Kr1$E6VO@+=cV{&>4?(>n$Ks?pEIDgl`;|C^d6fUO%ax|9f9p)lwvlK
z?iSS-`JDIy{hkij7jik$zPSFqsP&t1PEe6x1suP;%ut=alcr=MJqdj)MlPG7^gIir
zH@XC3=LUg?@@)gIL;5l@Q0Tht1KM2*()i_Qm$d*wOFE11Rl%ntQ<hM=-N17qlWq}I
zO?=Hb+L$oD1>$+@5~3IZ8Vw`q#9viDJTt_2VBJbIIJnl#97gJ5yb|^(VI2@a8n6+R
zz%KwG0ENSE3&TH<ta`c4VOolMlnqzP9^!eyLq`jm1hk#7M{p2ugZqbmAOt0i<nn_c
zOlX2|vx_w4SizDiV^U8c7KSVw?hqf?K*&=22Wh~>f+Ln%p58wb6aqG3iT0wCB`hm}
zvM?4BhX6O^a43U|;dwrM^60<iy8dUJItsQjFe;(um#yJ#u_}3N=}Y{qm@JFyNEcNf
zd{KLuHPJh?efJpyMC1+=>8|EpqQFGE13WS+y0UBB9EZXdb`+;!TR8o2_z><;(Kg89
zLf9tjfk&P7ER2T_A0|0<p>h9ao9y;b#c)sJszJj8EP6O60KB?WMphO-nKyVba89F=
zyQoBJljH8wL{yBP<!`r46oWe3zc7&g6}&f;f@`@h23YpGu_Nrs9ZeN(qC#{8Ucb%v
z74=>uKLP)P&Ga!KTf4s4e?g`5-GCDCnQ0L!*@Jl~F`2J_4g!-Pu2K90(&eKV&Gi~z
zHOBpafKx}j7}Tl${S>51%G&dPpihzPnbd#ZwJwkJ1^gM*SMIX(@(T+editSLR#(((
zis#$Ix`5P?Ku$_9GDi}gZu|RhuNvD?I(>XB*)xr&h-c*X#tl|q{xwM8b6fJK2@1w2
zzp9}6uLzkUE}2dL4W-$B;`o35V@pE>EM#PTycDZ_VaG|^naaw|EG*@|MlGGce-=R*
zy;xbf=j0vO%YT~Oj~w#V1T8xs8W#pJ^O@33oY7aqXi*rC3}ycZn{#2n=rsQb*oYx8
z`*ZBL*Kw6Yu}TRQAkv{wxa3(vw{qnTV_oOK_=6aDC)Biz8yCHq1PIM9ua5GpT)U;S
zT+rwkw1Qa~th*Si^ThICs6)d_uqQotg{Q#Ev~&&2QL3$*60;n2<PAHgn0#hsif*}N
z0rkpg#VkSm<!sy8O%!w{kQ_Wk@}6dzK3eFR<2i@M#&Cp(;ETX8Q{@>U&C6RO=Rmw;
ze}0wSz>h^2w)p_p{{2sxYNHfmZc3!+e0{1l{9WL7(+hI-dE}94pS1KLuie&O<<pJC
zxle3Nhc5*Ro8BUHBMykZ)+o7wYj^q1Q_J@~a(2dtm0qXP$e_M8@SdjagTVM9=Y-;O
zd*X9{%rt#rksymI+Ys<zxN8&_In&WY1B7~t!6yoc8L^GL_&L>j@~n+KrT`HF(OY5l
zD=7XAi=Q!xqu3f9ZhTI=js;-ZMkuqnE}VKffMCOCFw@b{t`9$3>bbi3;PiO+Xq6aC
zo5fJe(fum1MSzS-eR#Hu+DyJp#`xDe-$TIqp`Dc2ZU~1V<m_;6uIYHj;#;P&=j8ov
z@Q_6A1qU43S+mw)F^9_6HDP1XYQ8A`IaQxb+5CW-*$p<z4oe#n++p(Tj}Bs0Z@f5$
zVGoCh{@l2k4=#ps;9W7Ac_*_hhTbZF^$%EHv{1k8>dD#Pm?ws~h@ID4)9P|&R_H06
zS-zPa5mgnG<91Wx1ddrU(F-5iS=wQ#6peWD`}ccv6Rr^x46Ypoa8FL%K>J=JV*_t;
zH|~r4xE0qcS%voj?{t37YSrO;vrfov{vL*)2^zI8gvNRH;AzUYwgtMltg;ujRvPgQ
zIR6nmO004We7D@^sA*~^76qnGkR+|$h~2Z|M`;-jn);o_dr9LCXx7OPYLe&Mmvlq&
zIXhIJJj(AvNTmxF!)AKKYnM1}+Y-<iUhUEANT(?;y5MEjDi?6<G$vxq48$dl`UwuT
zT??4`jk;1k;+_sPax>kZR-!y|F9v4o_H^rhLiwVfJeRk$I196_`TTLSzWlU!HPRJn
zBy!G^iv(TyN0dPDY$~KEaku5(l6^x;BGaDVS5XDe4<Ayn6@Uj=qM1)8`f}Vhm;xGG
z8$-qEriR+eF;o@l$XYn`EBuc8qAg|i9<P-Xt3k%1YWkh8&|3rf&X3y#x5edd)Y%L6
zOW6tOhQaMQgf@kMIaDj^V&*;k!{X(EMq=y#-I}MFE(++w_pL;7dU|g~^Lv0=l_BfN
zA!Z^?>8F~xw06y!x3~TH+?j<v@j5R=l=<{G#cW23!4C;Df!Zk9Fa=N_{=&7LP5e=E
z<BJv@Cb(-D<i~%$;kQWOSoY)&6=LoBYAa*BMeJ4YRdTd1L(@bXqDrdKSf%(i`2cU+
zcd($q2N2SSrG?5+keR=u5%6Hj5$g)ZoFUw6KZkQ{JQp^{81z14Bi+dqn+V721w|>6
zkcNAfiT7T3&Fe0xqhm~UbD#RU<Lg9S9BRpTsCuTHKHLTp?nb{Sr?JDXF0UgNm8__6
z``g2u3v#JfuGuhAe%NfVU9|LGLs3ou(^2aWhi2&#pNvjnv=!D^+MQ<V?18ID(UOFC
zcwLJz-XS#m-8dkv_f(*S)(6^xoMgGX=S>VeS9G2mcs|~%kN8($nA)JyuPyv#sQ`{C
zvUdByt{K<OC@fw8e}Y${9PHlYGemw)(8Yet{qp4tgZg8+PXGbPWK4A6l456{3U#Z7
z(E{rrL5eWJJM;~3?2A89MQ)?>xw|FYs2KzAoSpwD&N41CRILdZ8FA#dKl}y<x`CC|
z9d3PSyYk~|PPUZxXfE&OIcw0A3P~-Ap*T{EeZGEqfgjquJt$2jwVteJDaOX-#rWAK
zi-W@-K_%q1PXQ-|YZKZ^oAwWLU7O!y2o%;m8fE{2TmXH{Y@AHN3QrZe<huI$T)e!z
z3YG>;6r(Umjnxqc(FQQHDD6F#o_Vk*UQWNG?l}F;o|9_kB{Sw81IdtieQYD3H&9>n
zH?JAiUB`zZLA=&=`*xhLq&VU*cXu1CQG&(g_idxqYv}9Ph%Etg`ugkNbFKW-%Vm>x
z8xAd3A0B1K0CDcmxDiY&O!w$MW7YoZ`8Y+V?8wcTyHHI}-pt9#Nq(P7Qnn@rn!N&q
z?~TfK<hp9;E(lW)D@FbY$wXbKnsG(A!k)6Ea&C{8#YyW4mmf&wcx!2rqFqTG2;l65
z{HvQ$K7+|Qc3k+Su;Sh|+xI_$NgUSqI=9v|bDsNYX$P)mzs4~nkI=J}Tcu~(Pgu=}
z49n)@?V^pPsl0gcXQhuDEtDS)0b<5?E75OTPY*}{i7gG_-i$v+zC;R6?6V_Z2RGRi
zzWg3sgPO@CDLpDJE307P!5hIlMU@S&@NsGGqUuk2pGtt8Yct&pkUl)|3fpt_y>_TU
zeJqh%c??p?e^lW7^XscL8se~X9&gEgh(A{<^>p8Hv}~sdAU7TI>!DjN#kCHdJRV{p
z)`ycdxGG+lJ<5D}&NYBtdMSGR=rwgRtr<-!?O4p1<k;lYCTw3EDqJd+0mp<z6!J7g
zfnUrOPs$8fk-F(@%g-M_PJdNRz4GzqN-(2c0K|YPc%hN);y2%qOSrhayu335T<%>Q
z`r8g%)0;R>$qzOOR>umeDBH@j4`wA{)1N!fP5oASC;#9e-zUlN6hH((@+ENFcns$O
zvjWT&d8xZtz_@02^DjvWi80`wH87k}KW9an+4?O!8rKdNfBN+Bsn7oYpj~5DS=N*L
z$#sgJ+PZdkcpWMRj9J1M+vYy|p(!+cRLeOTCQ>_3%4J(3_L4^n@xM!U^X(Z4rJNue
ztAPj*UpTbT;m$7~RSiDfdiWv`W%TB^fRa((UHRw<MG9rKZMx$<5<jm!-CD~Q=Kc0k
z;gb_~Q|?qWbX{$U>}abxk*J5uYMu2j8;9|A9TP*+NLKYjd`qovMF?IY>3q^vQ;{N)
ze5m1oopopZ-ACwJehdwH=z8ttMka=HKx_J)nxa}SPfn>M>#xkpOQSUXGP1Lg#+o@I
zU{Jh8QVmH<M`A2r!!*?tLWL$VA}#AUh_Cw3hk%L`et)n>mX?9((|4<JVxelKp}re;
z*$uQ=-_6>RdM&_lr1hp}5p>6yt&pmEl=Pi*RyMWx#dxFSI~KdMl>=YbLy3+;HS->t
zuv6<)R1l{i0Zo8k?_hcUX{QWac^~f7KLuXTF__CS^<>XSu`oQT>O=<zv;d!xv0zq$
zIgGFI);vXR?xPDIAL!jMy>}M&YM3RSh?oziAgi3@4Om`bY=faGHQl>`W+x6(qz+V%
z#U)xICIH^T+b!9CDVnwg2yJ4VYPm+r)Pi{BgDAN!zVG=u<|pV8TKME8E`KtB9xaA>
zGRk9!5RV{THC8#U*xiqxya(`C+AtPN_S&8I5PI4&MWE6-SkAIYj;A?~V&uhNh?zn~
z7}f6V<6Xxe^Y=}s?<E;_10bH+V1q}pR<ZoB+u?nevX8Z8nTH_^<#%(|cF@E*S_rEk
zULaBc))|_He2!8-0kiHw89x!9gtp8dBO_J7W-=^BcHlJ827@#*F;~>P2M7tueP3i4
zzBoC2ykk&<eeQVemkFiS1O_F0`wT+kR??;*j-_XIj@hGrsxNb@1+9zKH~W&qSCTr_
ze2}$B`=WC~wWXt~s;ePymD{}W!!Z-nSy8uijVIvNAPa+X6|NVJ?8|{(H=7cb!>w8l
zg&~IeNXt=?TmUpF@S!60cC{n!8I<w)amjr;W!_ZP2LtWl#7F6W4yGzI9VAn}g1afj
zs&+4Ol#>x|xB>T5xO5{{Bd_@OR#Z47ZWNpbA~_=OD5x0yIRw-*KV;|a_UnAoMk*~>
zgosaznx6Us&)D4LK!3f7|8oWqr?<dwnhiC`ZdU^|4}J7&5uG(|+Ya{t0&s`z0JoXx
zX!R~M+2+Z2hKDgM=+O;v=YhSy>ly$yxriOQRxLYsHT&%9`?otPA^Y|l=nj*vR)(_&
zyhpNrw<Iww*JoI%BksBcPhw^k<m+m~9GGZSBS_JWjA$c-zn&ir((AYjM<7%7kMsRi
zdLXP>Puz7J>#!%F215OAV*c6*v_G<MguHmQBy$GyKxURNESODa0c`M2lz;FoLW<Yw
z8wYB+Maju(J%O{5lAio3hX^U?c335WHS7Mc_IRXhMUW_qqxx}+KgbFpjaYAEyC=wl
zMzfIy4;%vhX~2v;#43YzaV!cLHOQpr7^dGLsYVB(9t8PkQ1bOK^>WUo6ZdA0z$hlr
zvUr@kW!@apo{Pc4VLV~D3BG4qc_D2Xcl&yn(fqos_fGs%TDv`a7Kwx7c59{&E}rh5
zo+prVL(`~(Xlv4QK>+g-pSd$NH1G3M&SMkU`QLrN5E`TF=LN_s5?WQ39hSqaQw`kN
z_@98uVhl9XaPeNXdNtKfQ|dzN#a9Qrzx*it3n~xIE4D<yJGj3LfPJhdZ5dG}py!RA
z&bcV9Htdv<=~CsJCK?$Y&$&00d-I3E{^E|{Z<dyq8_|KjAu^H7=&%XO!Ya`W2X@Er
z&z_Y&wQ!df{EiF@?ayjgu2gfiSAPeQMLk+&sPpbQOHVV%I{?s8(<M3gjX>Qtvk7fq
zvCvrkCS7k#vNW%;xv8o8ot0^h5jYOs=x%5ri1WtIdA)VAhp1?Cw$55KR5Xd6=b@2A
zWRcY2S-Q@i9!bQ5A$2T#*w^({R#p~jOQ=agq#KStR}mY!wspE|n*ppf+klXSU?Bj{
z8EDIcToG!ObGWOW9hQV^Y+g38Jo8?1{}v@z^-ZT-Uh2o<SD^*lz70qPjD-1=EIV)G
z8#hjSS>_xI3<%h#oP7)s!1t%pxe*_BsDh$wn76@HG-A|*e52;;jj$mP+$JVDp~+z6
zIi-D5W>Q155ciIVP5-jXv5a+v?@hEi^Jb+rd{vFNw0-|hzsaTLz71FV`^mkNgpDaG
zDYR%mk#@Vi0tC%ipja^M;nmS5T0}%x#%|w<8W^F^8HY54CM*gbu5ey5(6<T6XowE=
zU7)#ZViJn4N_GQSp@^l&u2q#@881(yCCaviPtH|$%ye{kz8yM|)KkL#aiUQZBF%^3
zn>TIPaD9Uzf=E=hkIh=+e5a(D7<I|&x0){Ouu(-$V!fL*pN87uJyos=i^ZR_dA`w%
z?g?kR$mkH1_DHYWbN>EexEX2H(+^^*mF3u<h&!z)JSciy{eIW47Aqf;u6*?7f(bNP
zUjhudjx{v!!U2l4{t`gHQ;-Q7szzx&eX#`&0kf3zFL#r?24HYrz!BP}u9+BXYw9Af
z7n~OCo|m(2qH`P|6%Uu+-R9-A`O2a!PqGlJ@gM&EKoQAo6}VgoyM~MI&yZoEo+b{m
zlVLgXfq?-ti|HLCu@`5r@ponS$+TMg&ieb$@8Cimx3in=j|x6^|2HFg10oY@WIj6K
z`4JY|%o}Ns*KU={*fFNO)v3-VOmD*a!;s*xb$etGX%uEG^`f@xH^=%EoP-e;#rrX{
z(Ve3m)g91_yx=swm9yQssyot`M6dD+@$WfT8G!DF?tBK9)T?(_A1pt~SUo2AyX&MW
zi?jxrsR<PJ$$F-@jdxPVYn>lIHX-F_S~9OcV@(Q1*h0lajshG*HP}!&i!J?+es`Jo
zYzPjnlcLrA>Q6}E72BQa?lkrY{V7?Tw$2t_+Vs>RFkvxQ=NlilCwA;F_rXp6-&f?v
zXK0_fxTkc#mZx=CbOak;w%*uHCvD$X$qa0+>vUxgX_-nFMJ=B4s^*(WtkrPSp-z5f
zVtgqzr_wX_+&$0uA10G<Glp3NpIlx1S@YNa+@kio+5J_~mj@f%cZm9<xkFNqU+2%5
zmZH11E!w|9VRppsDAjiZu*`bQas#LcLF1MpAeL9NEqSp5qD~h+u@$o76E^+zE!7!u
z5xinAKW<Ge4AAsdU3TH<{jcY}{iuzdy<R-@Cqx_~UJU&Ekpf9PO1b;vOxQ1<XKyx-
z_Qb1!aij&SV=rp)41d+BDzarGN3q7OwJ#M~s-_vba}GGn{^2~go}KoHh>CDNw6dg9
z0HGJY>`3)NuZ9}rB#N0k8{+JC?C#1{%f8siz5BrkbJuw2Q1(WLiS{?qP>O?m{DJz6
zcF*wo;+&8e<s6aGSc_EPY(P;FS^-nl%@gfk-eZpKL&>lUqgnS1UZRhUR`^v0(c0a<
z>Rrg(3Hn(kINzo06|*t@x0ylB`flh@@A?N=$Q<X6N;AB3@8;$?&QRHg&%rLUMomRI
zKaeS$(;i^T?Q`YcfUJY0xHJ7OR4kxKI=j1_P_;^SD4i`*^H7oYmG_X|kUN+LJJJQP
z4Slsy3^=36^mX>ecQ?vGi~tx0Ip1eqda-}eKG=9m$G?l37;s7ADI(`z;<6eTYRxp~
zL06JdY-ev=8%ZWc38_|kkL2~D446sOkVUNkF6k2{1}Mj?y}$Q)4ft;4))||PfgkO~
ztQ$pC0+7Qk4j+=Y6~sP)J(df-J(q}x>h!nc5D2~Vc(Z0-E?NU#gmRpCBM=Hd2lfhC
z3mcg>7L4RA7DsxpOg^E<D7U_C`vuazje_=aT(-X9ipz@TD}iVmbj*L(x#A@f5B;md
znhDYai>C8}(p@6y0?*#O<wh0z!Kn2W*qrwq2_<O_C}oH)7ix;Mg6Jf09YBa6^wCNY
z5y@G1fnxs%XyOADU?{K&VMXs?bmzqkBD%hyo&gnjaV%i%1#abcj1W|j9C%4D5%e7A
zQ(c8R9)Ys}+|mQm=m(lhYf%Ot1!8=ZmX;PV>pOL6`*zBqrP-*O6cYF)85RnVKSnw4
zrb#aerb1PC#IlKW&TuqYnr;9@d#wGZ?v^AY7IH7)Kt^5WQRcJ1Cf-nGYN*5b$oJ`%
z`Z(+5FNXTdR1Jr2|2f(E6jT-ofP_ZgwmEVg=vE{o^APe1buI9VipOG0Ay>w_tV;j@
zhScb}PchnI1z_%Kd&WN}8$GLl3vM_*tvNG{8j(S>0VrW+GQDF|Gj$E-nIo!nRkr_I
z&d{v~QC-lGU=nLK8h|g-^4E8w72dt?$O}H5@_UnITTzdk#WndF4O3R$`LSjw4;@lR
zf4XkY+;Iu_BnVU{4>1zSH8B5akoSFWMX1&LTaYg%LqsX%ec*!j*%b({>%>;u!maRV
zIFVDi^RjWvW=N3$4YGO(%jM3PVNE{Acf0$$iPvepBI3c21g0`Q4SPCOVlIuQ8%w64
zmeJRcjK{B-1@unU-ul{<t-o2w_<P93w#*ZDe;z^EHa;_^r93-7XugSA<TfLD>wPbF
zt=J)a2BQ*)ca&hWfcs?l|JY17)r6srRLHs*XmL+WgUij)HaxQG^@I5MgNm?;#3)8?
zU}H1dU<#z47$zW2!ndk<qgWtthw+p)wP=QC7?eJg1J{!Buj~EB%H~Gk(ArH*pJq-H
zDH2FD+wOzg^AJ(^gnmawmGgS4#2vrsnfQ|K45|Kwxquq_N~E4p=M_-zGTx&y{GEvU
zAVTQQGw^8#=5^lR(pmxFijxfsSwRYO{Lp6XXm4-dvQOUe5?CUHQWigBDFLBM=w$k`
z=kD&>UkB=~ThVxw;&E~#B;d}<FpFko>&f4)n4MNowQh-JZ;0$QgWe{;r533oRU982
z#z8001r}cqP8NC!jJg}Uoh2}9)F|RZP|Vn0<!#p@^2&JtCd0(Q?yx+$omf>!CUQqQ
zVnJF6PPpapr7n;1DsS%1^9-yZ*ieKG?XQhAK|{zaaQh4NAfQSW_G@^|e7CZ+q$FC4
z<b(%ZIQFj=zyoTmdkaOYGtGv|2+a54?k+S!ML3X7JJG#g&Zly^bQKX6{OFzQ5;d3O
zD-WJz!(9(yC1!m+>oa0lw@#;_!Y@+{CwG3_g5&)+X;Co^xS;$5toQ(?RRkS%gXmQr
z0f6_1TO@!fOx_%V${^iA4C6|Ae}6PZc)?wKSX_LiKlYq6W7F1%VoRPq*^-D#`MmJ)
z<L}K|1x3t%e?bcU6<m54;Krra@Qy_6PJI_%I_9tgw-sigor5lL;#t+3!|W8Vy%)om
zmY2?f8|L1T7<Ea(pzhrTc$`Vn-JMhejP2wC5kZJt*dB`^J@K8Hm8V^|qXHeo7<yb(
zQHr@p;9o)4Ju)H(j)YP47X9b|%{QE@OwjF{8#I#C1@K9(q^&`yE<$9F)R%D84+?4Z
z0AaE`g0MMzTv6C~B(=xD!2uk95g`0c0lIRe^5R}(ngryBQhBUfPORVh^(3%e^h5ln
z#q_Yg$pX?fA49*?oi&o%-PmE@?RGe~v*Wl0VGXeHUxrEMn20L8#qCeTb{Wcv7YEVx
zRRq3q!4N8H8fZngn-4bBb6%jMXUs>8KvTvI8oA&Tp#Y<@ZO0zscuGO%1>tSgc1_z*
z75g*BTO5!Wp_*<0*)SbYeuuD3E{=#vP#7#M%!Fcc`S2NL8{?Aaa(4?-f1dwsdLEF_
ztbN&*++e<69_8D23lA)hI3Uov3-G|`#yy^_@d}YLOhFt+E-ONytxzX!KK51unMb4M
zRYBab?`O=+aTvo(n*^f}g0tdUHQ*3m(W-t1jgEP(ET^-e#fyN2(~!cD!`sqzQ=Juq
zC<Qp}Mjo}a1LvNzX)R`&6F3D#^`<|aREQG!i9vgaG^ouLrwcQ+F&k+Q1ORt$u8uoF
zJ+BgMg+c)3k&60<a-YLh3_L1)Hx?{G-VzTPL@gnxUq);<m4`PqHz#A=M_%;%1ZJY?
zqZS=>CZC~O`>{t7`owo`L*G()o?X01B$O96a}l3U+V5t4#PGst#+&(Yu=Spa$&d5b
zch=h45=BFt{R|a~B0zDK1n5FkT1wg2=;(Fes>C-rvYnM<S@#1T0*p7fqU<I-&+Xm2
zccwB}E;aUYe|tO&zH05z`_xn&pFA-w$>+BwqAy|f%SYe6LA#DsFNv=)%YX-0)^#8>
zL`QHmP6cc3A-t%t49QH6g+g|kc?B0J8DR&ZR8ywTS|_@7pDHTc(OL7(W{phr=vPY(
zc-*|F(s<F;yP>ve{J=ll^jPj|e}6t`I7KXmEAVeO1xD^a$|h|48bD_k%wbUGca#fD
zw5Z@Z5iuWj??}-h$Nr>}2E(pYASq5*;K`Pyo-0(0<Ybv6*!K9A1%bOh^mOYP#xd0P
zgm=Yq|2nshrG0i5t@QcaY%3z|uxU?u<DlAzQLcm?YBueqKdD3#iGe3@W~wIslMvvB
zt20?^{yu_p&w~c+=>JIgdB7kIf`SO{((L)N&6@ld<<$L@JXTQ%BI@Z9UgmZ<%&&ti
zIQMqzD<GvDjJho8A-4=(z!n7Vu;VUCoX@u$XLS*ZvJ1l-v~c<fHbFGYqPu5!>45>)
zU+ntXmM)mF@ecJna&u}V)+kD2?htT4l7@~4CF$w#$uTwcs93yuIo4obLn;r1ChVC~
zBDi3Q92nja*abgK--oDaUlCWA9oSq}Ys%h9$+1eYWnzNB2<PibA}N1=)ibNc(!$}g
z`E*LvzHlu0LM6-RfRFchEx&BIu=H2>FCkyX$=@K|)jt9l8G|W`cLLAf|0Vne7y$On
zHD^YD_SbG<-(#c_zS8*H{nIUp=UEiqSG6A>{8s{CX~XQ%Bd&7VMQCo*{=fd3|DP9s
zsSjkDOaF%j$M)0z{Efj!@MZq_?W86D|Cbr)BQ4qv&=cO~a-cUm%+265uBK=g02aV4
zp64~5?E(KKROr!H-}g8tJpIr2FFH4p!%3N+9CQZc3E|bqp!X*i=+;t@wgB494^gN{
zC7G+f%y<3Hoo6`ow(eJ&slI$+92Gb)^gic=C*)`0$3O{*0^$(>f*Zj0v5Sb;6?a50
zu6o)ri2R};!GRHoh~e;)i5`|n)u)8R!omnEfgijRiXogUmr}GZkUCFKPw!y~TqB$4
z=;&ayA>qGA?m%Ni2x&AIH_&)sMioBvS`_CNDTDh?+F{uMf?ek!SqMN#(TsG4L~sO^
z8xFnWHBCbL&*4doLSntA^ac@_Sdduu0$d`#+(jJJICiKYwvxFd+%nbI+35%@6`kKL
zWN70iKT)dXHC_?^v)NKJoUT*12V^`=ClIbEXt`lT`c2dL<po?qc+R&Fcg&u-Utbk<
z8AJ>({x)D0sbzxihqO7rq=XJGM8xOlL+*<<1rzVwz_1T|2F~FJWGh9#b}vEv!Q(@@
zsI$DZ2&Cx8;NY8|!Z3o6<hedfUVuJF>(isvC=8JDua9#13ofbC=g+&e^8zjpy;1mt
zjJko+?`%0chH2yM>qJL_X<rn0_n?!+C?OgItM1;nZ{IH=DO*o8S%vVRDB6C9wi|9z
zdQdn}2<zRJY7G;#QUDz`j9DCAnXYn*ut(x`UB?q6vN$vWHIR+2XFn+e36`g*CnULk
znc+8Z(jqOEkxYny78@e|jhi=n^^IK+<>$5%Z0DH_BJr}P4qi?9X&3k>tPYIu%4i7l
za=iYurW!^OsxwCEAaMd;F5i3sn(f!r-p`*`{j++D&c*-7rqjBEnK_o2H!0?j<41Ta
zpvC#i3<L^w06GinF}!pYTo+{p1*datXZdPO6V%c#QpEx+rJW}hP0A!}3Ij$c5WO%6
zPyq{50p;5odEh&O$Cl9iUB<uOkH@<IzyTTK4!o=_gy8YfTP*6ugN7*xrHG;<B=jmN
z&*mY#Lv(q+cU)W?5zF%-O%_rQKIZwoLl9;6lzscAfL$qPhj3;rLn-bEbSM#&dPIxo
z452o?jODr?z4QP?XRfbIK&<K9x)q7Yxq9$!X01?RT}BnRCFB+6i*)|`IMW$gXzn6=
z3`Q2RH4#w+xIe$4VH-9NjOTy3SCExWa#l!X4(=V0Sr;KrfmqTelh8I?6EL7`L>e&B
zRQqb$m!d%r-QI}Lh3$tO+>0`+)@KOr15}w5GT8|PF@YY@rq|}$**FQ^qd=t=i3Oou
z8Nft-(62HHXiC>r`fVAqLz)uOwdtX7eGF8E(Qe)ZLK#5+@Iw^gnnk7%NpAowlHi6}
z6HseEA*yA0)R_47f3XxoFF-=R#omYUhDbjU{ZIt*0ZPQs_F0@p4<y5WU@l4sNH(nd
zYFtWfOL(BXm}-OYHYxOom}4G^e~;OhJf$)sjKMUcvMpuU<9hY6TS)-}mXSDjF{=m3
zAhFwhJq_h}j3Tg3(ETtuqD=^^(PI?46L4o42Mjqb@M6EEx}swrFF%~uchegI@Rn$J
z5$7CLAj*9?b_;Ah=wJ4h&i!@c*j5nrD5Mts2s3WKkC%hTPnSJcv2GOIA$+2R01=_5
z`JiEox|aCvq3`a(hI9>5R0~SBN>o=@r$V!s***oxb{xkKm#_+P7$LJMf;^b3S2^Qu
zkIlj-9}0?)#7U!d*}{eidtXtP69$K*G7C(}wrD}m3u28G?CHfzIYm(<RT`^jugvR^
z^c=|DN!&glZ;q%NM0o&goP;JoxHphyh&5(wOMwWJ-}?5wAV2RDEF*36Wdh8DRMB6j
zz|3CxRTZpxP&4-$m@gTe;0ihwuP^z%c@qaA#9$Y@nEgAFE{->BfU<y?*alkD6H}6s
z<Z)+R!eO3%=su*IYve&RD#0M2f?i7rnk^;lXR>($kC~j=9%onB0MRd2AP@0lry@=6
z7t9xwqi8C80LwznWnd1u7TH!S_rd>|V~FPy!H2YLJXuA^6%-L*<-`o{OWAAQF!I~c
z{7C;fsc%C;*Qi9kf(N^p2Jlc>{1R!ZScDbr2cWcl8Bd)DZ_Wo8@=+|U$wCIPeaM`;
zjBN3hxL|Mf539rA6pP-Ox(=ez%lJQ;cJpZpQZe8j_MalA7D$$2u)X8(oA#aBx=;QT
zYU}eBl|V4|owBmBih<Z6&Ny#T8HSLn+!0E^r@^y&6L4;zCd~QI__!ii(pUgWYvhS%
z@y`w-tX?CJJ#X2ac5k4LTgS*K9!G{xdV0G5NxMvg=9CC(B^**nzzshn$k`aA+)@U@
z*UVT9J5ULq5GJ@hZw0A%%{&dbete)#J_?U~DwM8A;awmeM{)`^PK|v37>0=#;V>iB
z<K39If&q!99L1{QaM1XxhzjN#>K%o@?_g;Ypjz@l_z<S5eH*`ox(?L7QIaw4nhVb+
z&cCC(j{AxaqkWK2wI6;T;|@eNYzYB`@<N;Kb5uO6o#ch9f4nC#J2Mj=O?Tbab@HBd
z>(o9`%Hvm(VV6*u3S?F_Ha6;2_0M2OL{U;L$8Ltrr*^#YnB=ha(%pM-lClu8`qYq^
z^_Z8C6gRkU%bXRj6r?a8{0dcBW24I9DX4*!90xG)MPcAFb!kO~%rAsDK7R6K;PWu%
z^7^j$#(eB<2(=QH&E!o`Z|^8h4x>m)R6<=p$t<C#7fo!RD4LYf{#*Yd6C29v>gqyl
zwLYxQ$MrMF83d$O33~J`Au!~ax5i;C#Jl+TdLWDwWma<&F=f7ogUJGf^r+5ANN|NE
zR_7XdKt@6{kKvWSyE6JmA#AGpUe^IGst_v3jr~8KtpwROj?x2Eyss!S`f4Oyo$u(;
z%NRHjr*Ru+2d;^OYEzPpL9bfMkXGPse*~KO53#T}i8w4}w@+b*?mLxg+@WEdHML`6
zpkA6gq7Glh8;dknBPSEi)e9**{(i0gxB}w9r4F1tZf#}7$xNmmY(-)^nNyAFBz0)x
zy-u7%w0kTNPw>Kl{(L%5iIb3Af*{U)?qTW>7Jj|WX+TD+Pem-vyDvb;7z2k$EEt;@
zoJ6q@lkJB;(i|lO2P!)CmK*NC#4bt2Sol+v28LR~IPl_;5|2w&Xil^q16HYG(Ym?5
zf(C_tWi(D|rnZrvKf@_oSy|tuFUkh{eekBpY<6=*LS$ozVK{U^6l6I!7~vPP;cNyB
z*#8;h@B#>~3xzZVce^AakRSvp1Q*wNjfnFhf(7e!%2;!u)DFJj=s4+8PfqhzSs9rx
zIH7SgX6|~6Yi<uzg|HECU}i1@YGdw@a_g%U%2Q#$#VK4WLon#07P9VznU$Xcz>%V1
z9s!loK8(eJyHaW-0E7@>V1TWu+n#Ai)MPjX=?27s*7GP3R%Q<v#6@^O!oLoQ71m?W
zuVRD@lF{M@>w9q-kqH`(_@8zgjh1JvGUmi#X3-WbF)~2XIvJ%wH|j}@Zk_i?Q0F<A
z>N7(UZsg@R;25IQFh6a-wD9pxf$vP)Aeix|=pXP!iB$n-%nhi>aElz_y|6j&Z=HGj
z8<aZa28I@Y2E^=DXwFgYNgrvitJ{LthrdmP2&C4=%efGAF03Yru+m6KhZP9f%&Slb
zML^p_1~VdW1%`t584=0nJHDd)ZOSm&1N8DUVw;GUgoQ;K1_b;EyY7_FB}0=p+yS&(
z9q#p>+NU6ng45;cQ!=23cHVxjpNoP<*7E?(b-X@WH*PjjYj4bqgWz&kke@}tgt!Xj
zh{&Oq@84M+A&#IV-dV!UNrqUUNvL*{4L&JK;Go4toW2yWm!JA(@%=KOHX?Rv+)Xt`
z7G`F3Znc$i*Ke2J8aipTh%wy^*-}SxPHtDT#d6WcVO+4Nn#lF?!-ordji}GrRANzk
z0s~x8E@fc<yl5Ce>PPgryqN+vxPK>>AorI<axDc1nQiwR$6jG0iQ~>1=xf&9BtCQW
zQRWT~wT5!iMw|q6#l3q43zyuH!`Z5kGYcHMg};+2yc!{^#SCH#j+iv`wCW&fFUOJ%
z(O;l_>q75CvIf8!_T?@xbq@`F#mW5<`O5_MM(qN{<|7y}IIM&E`t(}TjjM2y_G6Xe
zX?#XyhF6I0!|lx*7L<uXbF+NHxW|g&jv=Ny93ULnp=6FWC<zR@EJmK1XMnW4tv_Bj
z3C4j>hc<9$`&RwuyN|+k2W8L{nI#38Ao1I;CeY}+tGJt2Quz7bpDo`9za5S~T{xW}
zr6YC|2oNf905YMA!8L<O#4Cvb3}Y49YxQ?iy{=<ewN##~dgTho@rG~TzV+D~o!N^S
zY{oD|^?MX^eQ0YErd!ASC@ah6p-BsehC<i$JE!FVHjJ>asKoO6mglf6bW-=;a3dZ!
zmeM!QRtl~wy$=B>YTpC;SyAPSX6Wpo*K39vWp{NORRf-}qe$)&wOOiaS3f6_nqx;<
zULWWb?)brDCGAa(m1nnXJiFmY!sCOqX&NkR&u$#GKJ9S-0h{BM=;@ZXubV=eo=OSL
zFN*hM)AR4#`Ebm(;pRb6n_#;j;onTRf|@t31NT9+A_(+)@$%(qDXBH-F6M}il|ph5
z1zxcgmoHyFcKo=?K+9hcf>;(2-z{I1m6gC@!He*dzP`Q?AY{8I7Z(>lIzYNl0sr*?
zoheoxymL+pi6n&{-;NHJn%Dfg>n*-FH#0dbjM0$_8o-Pxzw-VrY=PH;M$C9ZPp~FJ
z8Q!5EA_FpUhK#{W89JTuvoRgsYAElC`wJ^2{*|GE1IOhEv~a5!7)prtsnKA^b7zDg
zU~VEiHZ8YG?hTZ_jOfS_CXsIx6dM!ciYuPDgwD&z{A`JF?X8&pBhEo77r$^}HRbI6
zyIap4QOf<fMU(0@b&uMUI$w_SP4xh3>o}w(v~S&dWi9h!)s}0Ds;YBu60R=T>+9*c
zct=G=iQ_-S9WH)oi3Jh+bBB_$&U{(=iK(a21Gx&<fIxxTMcV!oNQF)#!<+{^>(!Xv
zwmb)REL9?bL!;}4n}le<E;F+br07f4ld4`VxX%znbOzN}(~9dRunQq{W=}G*F~HT_
zg#9|A!NbLcvNQ7o{phc?_A6Je!~~|d3MS_yC8=jTuW<{jN}M}K2j)=%7Yw5MB<}O=
z%}YSeDSSHXVf#TNv_<s}4=3l!eBYR7Xkr~PMdV9GR@vL7^Rb>V58)bkj1n55vNZVR
zkC5yMvWCi0Z7%kJ?MZf!Gl*p0M6AJEw{C?P7mtA%E+)scG(prO2ET>Ja~1^F+H!Mq
z7v$%YT%DPju`qciEqbsL-<7{UOs!g>cv?CW4nNBiJK$HkrZcuW@`Xmm-P5<NBpOZK
zo<H9%40ZJ$N@q&tLqm<M`?A{pT%*5M=Kc)0Z&H@23AOflIl113W*)bP52;ZSU;#hC
zJ?rYNtF7%Kdt3Qv!R96S;dblb;5rKvKE~Ru4;k2DBcWuj$vq6Bp5lnka24k=#XFPD
zT0>(8?RwT#`uh6h@d9ZeIU@iuuk;`LtCzCh2E61nt{nhQoM{2Pw4KYk444WMiW2-a
z5@S9N_(~p$L(0gx@OL306k%cEo8@I;@6yT@FWSqZOUVaRgQVoGnCEf|GmGcpsidMH
zf@T$Bh#=Ou;-l_6w<kvG@#Dwjd%$;rn(d>`CeR;rlwUiv73JkA7_<wK;5xj!7q4Cw
z<0c&&sOP~#pgHwbR5#Z*crVo|3Ko(0jT^h*S~+NTV*ma%n9Y!nT}DNLROE%1S0SL=
z43F(A$n0T+8KZA?xxL}fK?FpEgoi(ZUb-8X`xzY+eWzhq_|es+Fgyh>^!|y+2fR<f
zMNw1I+e0xZ5jHRv6gKxt<sQ6ycL3pXpPHIx;*3wibxffURT0a!2hb{0G>7f1R^=94
zgM#9G^Z#P+&7-;O+xF2*BtwcsBAH5M)}Yd0j!L48naNOOEDbVdh>T^7h$JC%qQRIk
zp$L^i6bVU#%=`Gb@8^Ag@7`;#wb$PJkNwAPt>;;Hciq0%_qwjnIUL7v9w(qQfM7o8
zucERZG;WoVIe??cwq{M;Z$=;RQDzvI?>^zLGu8DXD|0$Hsy&wGsjpldrHMYBO8q7#
zym4bCjDq*BKbo<B+)5musGp661h{|b4X^n4&|t9L2NkpSu=L_u(pj;hhYus;<Ab1?
zfIOKo8s#HDq6c1gn~GS;i0?h+;c!`i0%tF4W^!U;2+j)Qn`PeUWCbK7G;9M1GDCU0
zhnsc-%F@!*^af_3Gc3*c?qiw9lX`D&41j(D5#9T9%A%8ZQ@I2n2*MtWgqz(VBSW8X
zT#3now`4;MJd+)Ge>UF6(P~8HDj+G@@~oK&2r3_{7FFP>ejNoz5mk+h5?crR+uOBa
z*;S3+V5rbpSC>=xXwq`@*r*<9GP+&AsQzhFr=c$s`-{z83P9vAakmWaXk-IBIy^cS
z78VKGEZk@k>@b&S+f%j=DRvqDJsHu4EW!WgO>TI0Wo<7;p=^%og-iDA=6Q(1@d=gn
z^$e)w6@j`jF*8e>SNY`@3#lmr99fC`Q-jY+$;=E!F-4pL0c1qCwqHR!CX5Dg#eg$0
zmdyoWXx9>L1<|AG<`^{`Su1e_2A92KV_~ROs{s@tz(+(yL1?gvlxh&dWj5`gzm@>(
zB5d7mU;HH@1SXIr^)FsjBO5z|2?wE9QB{>{bjITvps7@R4}>s3ScYOxtTRp)ID$18
zm#o7j(rAxbO6Fz`i~2_1UYO}w;FpwKiFK*JM&GYNC{qSf3Op2>d42x=y)X-%F(&23
zQDp?+jDCXscK}!djLmstq@<#`In15rxUOz3DwkKlc*qPfI8?dciuV>%qA5E!Hxsxd
zATeer3?cSYvbGjR%qRN}<8Iz`*&{dn?VBYkb0`L6frPyRM1b80pKz(fh;hH?>{aO2
zm^`?QFo1)Ot#5EJ5WVse3Kk)uIFXisUdf6!(4+i9LRS!xEm2<MrNdj_|E}cnR6%QV
ztZ(Qr?sE_xCk=I?$d(2J^Vq4F0yyewNC;m3r2sZ>=j4O|Z|2{yfez6LU-?XFN0V^*
z3cMGj6|;W}JeGz+bWs$_K$na<d<BVb##$IL+qk`7L{5$mtGS@P6+WMhl#iv<XZ&aS
zkq7)cI-bJ0(_8Z3`pN?<VTGh<XLG&h3h;Y!O?Z0!IQbcv5kqVVM3ath?{eh=9qPno
z_yyA&SG&0%Q<nidlhlnUa{zW~TKf7|@pt!n&T?Vu#0T}gcjG=Ak%{gMx9pqhiMt|c
z%y26(=!1pJG-kl^+3rhjPCy21{ehRSkda+j3aS%fA)#v6KXyU8(g2K2<n;k?mxaZ}
z)*>Pz>W2;mfBp8Y5Eobochi(hWQ3UN!ql-}*rp8|q^PEkA5GFmanai-F>FU4`U|)q
z4uYuTdU`fnw{BI3+3@f0J;`tc@c6x6x)vJ=s@GC(-#WyLQSuPDh<rLW26mwfd5im`
z(u{(V5yA_uHms&bwV;0d_zeRr1O;;!+|G-Ni?#wQRz$e?L^=PX1vtA`PA>U@L16#@
zt!=<+gg0%fBdSwCWBMrCO2=a5PoEZnwSBohga~m6#{EC1veCcT14a^)l>DAG*^Byi
zIW^NCE>~h1{Z;gl!v_npOH@heN<(8~0qW-h^zYa1rl;#pS>^_nRdHF888|#f@>h<4
z;KNtLfL=U`I`Yl&2iZ96Vi>2yAK>o9r>6_!fLJ5<S^>CL$4j$8rk0C*PudEEnHRh(
z|0!yQk%;UOkh0;dsIH|I+3?~;Av)MRbR&6DQL7`T7v^UqUESQQVY{q;;6MQ0i4Fk0
zoak_wn3(*AnLaAzoOpO>=mO-I2{=r4(6D2Yh}_90vR999f8xH26UrH5s3W<&w|XHN
zl}zKFi9sd?Gls2JR#w7WwlowK73rgTH9*Pm=EJUu!#GElR7nfV66F5n6mFkl9ki@v
z@Ax_&7}mYXRr3l74_`P9mY{RqVbXWD3-#Iy{OGu_uobcF+;W~zyNrVP^;*Fwr1T?w
zi(!ubq3tVQ6h6+$2`ks_(j41$$h}$nSE<&m3n9^QlWPC6foX^4;Cet-yY6K7k4RR5
zJ0u>4a6uERfZP=T0sZ(NRs|hD5_qiNGjj*^FQ+&_d;oZ!*VLpS0`A~t65wzH@aVvM
zNzqd{M4GVB3*r;UAJF{gW8@N$Vl2`7NJ&X$eAQ&QA5m4MXnXo}j9sXUqUUxLM}9YM
zumiQXla1nAa)d^BF{~~A-@uv!$G!hUWunKM^4D4tK^VAy{z`-?*T(-`75O(){vMBx
z$RBHaU8dLe-#^4}_VwS!Mx74MP-??}|6BZMM*08xNvw*V=28E*|M=<-C^T$OohAcA
zxFPrQ@>ti8jY(~nszCU(lU=9i$%G%CMvraYRrRf`bNlvcBDg3mTdC;ZQis;VPPXXK
zZg%!7ZZR<#5AH6_$}y&)U>NSplx+ItA{=RaW736@FS3Nx$puGzx2~D^tl9U|{EU@O
z%Fh)eJu9D#eMr%}w*naU0hA1R@Qzu%dNsqFJ<gx#wExZ?^EzAfP~{y0{Vu?muQJS)
zP!)t?(!pZ3#2EtXA3OKyPRtP48y+u#IJ!hGsAz=&qqv>tCEB%hUX^IwIrhH(<;$7?
z<e-O#*Kn=nR?OAlEpqH!0c~^*>JYzSZl1N#Z(DT!UdBKJH!Lyq%gt$#RhO<je;zC8
zF2~j#^*obnT7E!IPcLJ9imUId!Rcqgzs#u;%s%f8T&a3x`gcl7PFMg7cZJ>Tg{Sb&
zbKjqsdp$FU`DLOsEebUyGfh@Z4H_B8&TV<Hl11ewr?|3-SwpVCwr%X>mcl7KyC;uk
z)hfC3>yxDEfq}!C(h`REB=S?Y(Wv1CWX<H^;|s69%Ag)E%R!M!0q~y%+&*nTX0Huc
z&U<3?2RpgRm$ROnAFF6-$*mfE@atza9ul?%1%l9#*DzTR2rO>5CB1`Dd06Z3Z>5iM
z(kSd5d7Ofc$ZCcKTem7Mm=@Wx&FbYSLq;v-Fnb{*GxJm5m1P9$0e*4i3YHInALa;)
zpWSa#?it!}Lzs-ca+wJ3c7eAUrVi*)tt_$3ZkGZ0-8V3F71VBZbMp%DtKNuI5IQo*
znVOEgD`m%6ga6~lvQ)*37tCwp<1OB!tz}!myO-e%FKWl@*H>X~2fT*2KQm{=qg5>m
z6Ep9yd(*l(?#G);9HZ5s`~omCRo~FyqxAiwZF>eA$&q;eeFOK86E)P?;Ay9GS;chx
zTTr2{BdxmY;LAI7y^^P2Rr!}-5*stpw=**I+xjjSqW0}Y4o4~Y9JEqtzmI}~;5h^!
z@7opf66g(*gbtX2-X$5FblS3)F1=Z`lbmHWo%)uRx*=&xO-x?Gej&SiaHFuW1+-dZ
zX{tuub9n=UwUm@yhgb=?dt~Jj_X`!7w#t2jsanDqn1Q*Rc;P7e0gK<6I;{_mFWYqV
z7QXiqy^b6YclZ#P1Hjw`?m~r6Vk#748|w@V3<L|K#=6Tw$BK)|^_I=eyn3vYb?p7F
zO9Kpmxx&%Mo7Xkp>i!|16o|Hy7Ci;AsI_iNWTeKEwyN0gL10IF?2EQEeRWpbRJa^A
z>4u`ZGi&O~$GW@um_T(;cJv!@*e`{V)qwpP8X&UI-#;h%E^w1ZoHtUXeNP$hikf#E
zMy~TuFDjajY2QnBA9^J@gmZ|~_TZ~j{rT%rQGN?^-@V8>7g@Jq?Y2JcApDn<cW`i{
z-Zzscmvlyc_+R4`5EW%cnR5Si*J5xi+qr-NQw+*sft_z%fAq}LeSLkTluc{6Q98g!
zsU13G@g>`U^2U>PuETQ2<rF5lJM`Xb96F?k_U**Y1ZMQ5w^3$O-Znu&pTnYOJ<)^&
zj=nEWLA8y+ySthO4kqiryi5+i!^S4K&GXP9v0<A}`)7*Q{Hk%;(5#nx!rySGqoCPK
z?#)}aFh+-R?Op|lpi5KfAjibdNOXc9J!aqpMEU0~Z@jO@=Lx_BAOXZP%qh6Q<$U|s
zVb|`i3*-D)50=%drEL@{3@>q6I*$E(-;|VofTd2c316RaCySZZ!{b+8VKq7)<>skh
zU)il+CI(p_Nl-klL<?@2Ylq7Qgd^qQ_3&{*6N<3h`^9)AkIZs_Gwhi=VepK2MBZem
zRb9{Ql-^O(m}>rfA!$z_-i_8*9S7zv{gz6#ZL~ayu|hs0V<R*BlXNWS&TT*4#adls
zbUbl>{rbH8VtT4+?wcfj6q$<Fn@90Yj`7QzGAd`p#fNO|?%2gj09jwA7bs0$1;lS>
z%(kTb`T6tbD-;>F8CgOS5@T6E((c{E*862=PCN=g*hl34p*q<77kpRgj)NkX_T~QQ
zPpKO0ZytTG|8;Pg>$YvkJ5|rSxS$cP0$!=|3DDB>l6=coSe}cYS>0MeI&U=iahlKI
zP(m^eYSAD7!o&ntN?IlWQ;)Vbr-ESQeznL$SeQOXLc+E?gKhhc?X9;xJxYF!UfU)u
z9UXQ@Ou@&8<bZmuJI`YdtEu&(UU>ZeouZ!Vvd6NrM5BQ%MkWs?{iOr&vnt*-GT~ph
zzHV1O?CJ`|Co_o9f#Kh-=ln0eS{((H4Yx|l?$r|Sac=KNjvtWhh!-M)`3=KzrQ6^k
zAzN7r@2Z-d6nXMaPYA72pW}s;cz)>hUBy8r8SSABUzCuo#8YnHCL=n|b%%th!h(YF
zu@8Gzb8-eJI$BsPS-W;EDtP^qm(qJ(kE3l7T`f8F3_~rerlzKF%o5ZL@B^ezy-srx
zw`$MGNDg|Rq13vD$K$S~@P-W)1jtLgRa4sr0JpxeQBYQPI_n4P5(-`Wy{@ti8Y<J{
z&>>P3)F!O=Me&+%b9EgU80ZD+6CFHb_{MgL=40Di!GNDFk!rc5ocLF)m-XW)E4vyV
z9*E&*Oe{3NA#L|g9v)&*OK6y<Q$Jt5__!W<+&A>$4yYun-WUS|?*^{&`sZMrjEAhR
z@0C6{Zr)K1tH`v|!)S!dIji7LM5XKSp)3@rhy90Q7C@*(8T$ph0Kj~#Y+)11b6oI*
zv)MB6F5}#CCPP0)-aTYq@$8ZSuKVNhbp8vYCw}osQ4mJ>>U8FmoM;@{<a4U+UbX9U
z2CCw=!WCA|_025bkE*H3<7(k&L;>UU>1m9P``Kw)-|(`gOWJZA@;zD#a9YRQ&y@h-
zi0^nxlS{?$n-cfI6~uCKoAkGU)v>@tkY<F;exE)&)Wwzs!7WL!>FEIm+iYCE?b7{q
zX=_0N<qe3eAFAQ;-#@D%<mkg3AChLuqbpZh8R|)=9Nv734eBp+uY$OElF@Tb%KTve
z3DV$J{r4X#$5th-<9+qiVB1JnDZ62@gMV7oYT*kP%~5?D&0$0VEh;orG-zC^hP^4E
z^{W6p`2c!^Xxrfau0|Jh{fL@(0G(RAaO`>Pj{<<ansMc3G$^H<$3_ggCP%gu>HnO3
zDp?V0{(}AK)7$tOfX1r5-iTmf91VVc-tfC<cUoCk;KNa4$=`_s{9=38DOjY>bFbwQ
z!at%F1Tv<ie_>ZuRki<cm)rNSr%(4L?mw4lG<ls=yJ^CV>GP-U><}03RSlvwln2rR
z!4TV)8`x+H;K}6K<sf8dY9X20d!d&)vE7lpN?I6j=NkG(WK%}Fsf_#g&7rs;<rdlP
zi`e|>&5KwpT&)CleLFilWVhFxPJ^n3=Q7{P0bLbZZILX@^$$VUiut(^EXLh?=@$dc
zbyW5Bxn?hW5Jz>)SZ6oMV6o#;0N;1-HgR5$ja4-~d^qQfV&*ZGX{mDC?UPW*JTKlT
zxoO$$%Xdu51G$Z9HKmm2(T|sK^1NW-a{$42JxQ((L?|p6e1Ps#e+$&cAdiF%b<4&^
z>NYYIX;CigWy@pJE|9Nl7hMMx4GnqjSz7o7G#DBf-U+|Kj)KOAU4?PI!aBbFmn-hK
z-79^JY(tEG&;UC3RW03q>fP#!*idHZwnEr9Y48I~ICkvV-`S~c=@;DZ9BfM`#ZK&5
z8t<~^W@2QHA1Ki7c@P?m60Pm8vlOZWU>Dur=ulgeV({||VJFBiHZLFk0edH|g4KiU
zi<7$j;6W$`f|sw8qENC0!Z8Ol6dfdJ(08uCA$)Xx_`5Z92_=}M0Sz0nzqz^F8E2Dl
zv6#fA0FU^Eh92?H*;LYrB?-GO4=|*qm7T2&yCo|iF3tkPWAP-6jjf>K_MTb=pHU~6
zvZVZ8hsG^n#prv>KHsI72MNr~6zOyW^e^MkeYH$N!pH-yC&`gMI7Y|L?iCYT9wI&R
z?peb7k|k7t`=zSt54@Ox@FySLzxgG^)8iv=w?SQ_kTVoydJOoz#;yJ0+ndTz73rh0
zUDBD|*YRd1X%bZ$J>0V^osJ&OKJhR(YY*foaq;n$@86T@np#wWuVAq{Hgy;NH|;jp
z(4_}pmgc0DRb*`Jrs5<F_TW`=hD0ZYYq8DqRGt^r`}F65A>>vO^-<3!fUpPR&EBwa
zV@(Y6JN3hdub}ve?R}3#3VUNe;Ufv35M3EoGFQjEOdu9PiD5x~>9ZQi(zLRVAII%K
z!;SO?E5g&KgOPH^3Rwg`G0=CeUP#nlyLbIcP$8ArX;U?O@u!s6$Gth(fL<9)o)k9?
zwtz}z#YSe}gNl0U1xB~=&3B#uBbK!P;>4R%4Sw3nD$`Du$B$!mST91<g{e>et<!(d
z2z2w`3F&cW$|Eb1>TeWjy^{HapfnIR3aAyB4c9K)2+a>4k&&Io`GwUJ8K|qlA<7O8
z5<vGh6&wh_9f5HnK@%ghv+c?~6&G@0wz}r<54i{wF;ZwP@OO9e;wNEW^x7dS$8aEF
zeSq@1!voTV;eir>ro0bDI7gWr)!Tk=CCc-X0aH)U{RM@EKz#`c`SjA%Dip6q9ss`x
z0kH5sk7kX8Q3ai(B+Om-&(2M|u)O)mzJA5VjT?*XdQ67$pL)!2hN-x|y_1?M;_(cv
znRA1jPtu=1l((x*%aXqZZe+scRq+`^s4E$2z#kBCJDzgiNc=X%(al(AQ|f4%V(B~o
zt3cmt)ku^Z_IvpaXQC<6Kf7I+^Y^{RTCHOj=dYePaUwW7|A);h+ZWqZM!mkz6x_2t
znS`w;_1{0ftEI<*-K9Hn-c*OE$a-HqW?XFN<q90*rwxq^?C6=&$C!pgc;wat399)l
zs+ZSQw+x;%T=4bA`%9?C=XG_K9{D?35fr4ItLTdC->$M-nCE#?U6pt8>oJsGckl-M
z{odmx%DZ#wx5wXc8hqyG6RrhfP>8@dm=3Lgv{SD<{7l(UZPlH0bcEcE9%7eO(@mz=
z7aF4@u3xVLF#Y0bSbDlJ)$6fED3{@WHX2=kI~R_|`{MN_SqIM#53UCDfM&1^JF`UO
zxJBom;R&UYC1Mgvl5rv;YNxGEp5#Yr2BIZ+=U&>~yQLrUsb1tkifR)+^4ugBrnguG
zRe?}>J0+!(S915P$@q&`Z{Bd6QW}4G!c*(=<k-ih6qaJE+kPlRjP+`Mwo3<!h$ne4
zIOs0(;K2(d_!&_WWX_lQ@~X~Q!y8ij_VEg&q_lQ+E)1W;OIDK7#!}N{yk>K$laor$
zCI^kD=2~(WbkB}HQJMUOV!~GQz-zJm=i>&TS|fKJFYuj@N^arh0S=G%jtmake|XHH
z+tD%9eV<+I{X_Okin=mk*XbAs3p^J-l?A<NZ}%%G*otdgQQLLCT+;W?t#!PI&&;?l
zH_+A73-t3-zI$U}@gcv#yBnS`ZRm2oJs8mQg~>7Aop&<QKvaxh*z0g50;fFaj4IM1
zV*7Kze9nKG)&itugN&M-a7EDT&b_?~DU;rS9`kJ5=m9axcfU;No>5Q?K78<?PeFl;
zzf3Wk%^4@D73+6-gWR_mE;`li90*McmfQR~=T_TfP)IfuM7lzJKxkqB$FaMoSN>gK
zVLTpu#D7zoh<4FRxN))J#xm$X<6f__jpy^gfD}BLH;af^07ln5empC(Zjti6F}FHf
ze3PW)!kfpw196I5&MkfN6dW%TuVm6*(H7^Fzri`X!F_BkU%EO(Bs)WAeqO)CNdPl3
z*HB#C!4Zs*JfwJO1z+>1`tiinRKe-eQggsU=;Cza4Gq^Ie*iD6?Xk)}qgT+TJ8<cf
z+q+2dLnGE}Mny!qIPI*g{GB&;7>eZ#>=SSK&P?6h;TU!}R#Ipn*$a7?2H?WqscB!3
zz+fTc_Svk!VHV|8JxGua%5&|L&zwCpIlBO~;=L5uv}y23&zj1gWP#mg1(WVaWo)d3
znhg#P?)F$z&`hyCn3$}4kg)Sdc_>fh&FjAH7T=AzEbCvssLqO&6c86W{$))$kzW8W
z`!PG^ih*8+=+A?L#KIZxqGYGd#?aJn3?J*5Wd>cH&v9JrNk^VVmrQ{D3x9iQaQP->
z%QQSiQA=IGEHam6&##ZBdD5y4Z`~?lTecxlUfp69pZ(1JkGFKdN2}%PtgPyhdQ-J*
z*ZiWiU+r(|{~=}5ybknJU&l+hGDInZJgOZ0DgD&;K}GJF%b=2387n4Oo9;t;Ab$SC
zYfh8Z8#F}<byZK%UE6p1^eN|WwK~y29N?NI55g=`ORE(mlgOG!Ki7;~#onygVF$D0
zV6Z(+iPBdLm2$*pQ$HBAy?!n7OJBt0+4Tm!+xnl1H}B3cS{+zabT+!@I1K`|*q<Mj
z7JPh9FZ@`7d0_??78S6_U`F~LxmZQlXeBw`EqLC&ruW4h0SOWlyU7XKa1sJ-P$wpz
zM`P2$17ihoD`Wx1c8ipxcKg@Vm6xgyzb)y-_p}9meN)CR0F_GYI}NNtC5z$;ZJMuc
zpd?L72teTwi0LMKf-|9C;(~pd8fM%+(2g%`%c3*Q800*|f9#l>n+(N&uj`Au-|Bl%
zI6yYnj~{52<2t{?K*d8^RoxW$x4eTxPQAvPCBDjTKi+!Y?XxzF-s<fW(U_^u2M-l0
zxka_k;>lARqzveGvBYWi%flU_-vae%^Y*Uh;2_u=kSdbu6|F*~{rvpk>UV}J6MM=)
zO@8XHDL(i8<30oe=*jXLKe4OtEOVJE=$deh@A&(*ZBtrWS_Qb~4O6jkaaXaDJkk}f
zgH%-;J7Rm+lk<{*{esc4_9ASXK1)LB9TE}}fffY_;MFJ_YipOIzKtf|6<%#0{j^^0
z5*^i@&Uqx*GaQ^c>xmxCUQiQvh2=!C`jObuR-t0JefO?E5%8W`=j-Tw_^>){1qc@9
zFp^F{U?GWTzP@m%EiFURbbf_LsTc}9vk;w2!Fv~>uyguF7o^b$)XW2vOTj^BU|t!R
zoXmH=8FZ8LOow>Nqz_2;E4WSwiG;S(3hvyw=5Vmv9^MmCC!>I?09qr9vLMD#+?<3Y
zL&5@8@nEvM<MpbF*dyN$l}UnmP)2na6&+nV>ERc(A@0a8a7T`(>k2X?IeV@i=mlYf
zRL;8R**?_CnAr#n3)3jw&qv5GO7oGev9n_4A(a^?V7cwD-f$Sa_pF?qBDFL|U?)9j
zC903!py()gaib5nKqsPDhp^~_3DiDd22{nzkLxjafTFEm<|`{Pia-Jg!!YFKpT6le
z7<!~d<N$4`d^za3B|<vdnJ|ToGLmen1-5hHBNmJ@aDKzh>BCroqQM|uQ1ztkK|J}J
z{tB=BTfk|KJ&`fureS~=$AA3r0X~P8kCaHJp-NIu-;{Sr<Sl`O%O^jQ4U>MN9?#HJ
zIv-yR_OKfxkV~lSYh``+@7xPSO6db$*JTpb&4K3M7ybIp9cM(`w)b@+8m0gma~f+?
z9XP@kUPlxUk2nwO;7rP;f<($yTczn_nw|Y1YkSgbsM42E^9GMJ(#XEU6ot8IeZAWD
z@dCAj2L&)IVefl==8kd0sfmB`9KqhvX|T@-p)rN91rtIkfUdiD+g-Ur6496k@3@u7
z82Tl6$9Ur!&Bsi=%rR~{)SfRy2H>Hdzk?Rl$ivIa3pc6q!%G(OTB1k=%(0>9)CX?p
zZl6QgL{_v@rf%sFDbxT2Zw0PT2NzYy!pUuW7|nralxIIR;Ub=^?0K|9Wm~AmwfH3T
z-6l=Ra>+9dEiI~8$K0Baf|G_ySHH!C8Xy~C9zQyN|Mt~}GLE%4%hK4}P71)fI@JMg
zNM?W_eWnkL+6cTdR?=7zm)7|3%RT|ox!IDk$0@cy(@<71|2D5&BLj3A`>RznG*;MW
zKah6s5JN_)DziCC_;NruJQ(V+8<3`zd$rn(ijzIV4M_<T@)l`RF+^8c_*|L5tHZ;?
zafU^_g}}%$LBa%cp0kDdl~*H$Qou&<dE_XB1=A=*$Ho7I8tG_iY56Q<>R@T;@=Je)
zd{ALkR<yDL6R1oN<pCl5_sy+qR7Kj*QU~x65jSF0d5w*Kopbid$`XZ;Hb8C}i|@58
z(U0X`;SUUp$y`S3t5=m!yyTs2ST$x??)@gq9Hzn>d84DESXZwL0Srh!>d)AFy)P$L
zV3-_=ysENi?#Pf!P<KQo#|A<KNzf|j<>leoMp0mcGRJ>IZ~*ehf1Z~9c1$o6(bAx0
z!|>+vXY1CLTmqgFU_3b%CwoqW078&t^dhfz<$LTQVyWbfhf7!tL=;U-lH9kozpSN)
zFS(|UP7O#pvN`QCg4DHO@nItlXJp|Dh%o>aD!aKAw_3~i3RFW}r?3)YgD@@{u-h;_
ztVZL4L|JMaR6giI<1P32Y$fy$`BpDD_0Nq4SAsCn(qcpPRo%?H$IQ&EVk4tyV^<d!
z=#47$8|_)9d=3r{xJ@)vqU<>)s_gw)$3>FsCYaJKzZUwEk_evfW7pB^1rtIjLWo^U
z_S}pkX4y0|9PTyWa)uW?fEA+QySm?PIjUA$2tnb|%LLJMjzoA}b<T$;`7p-x3-mXi
zDzGAF8Tu3WW=KIlm~#J=U)0GnXULM)b>U&PnA7xYMDd!31>z}|$eyU=ZqAga_|U-e
ziUp=Ti8GU?blE1MfR7S)d9FkKjH@Tc7B~WpFVBi-NJXehb=?)ijwS`XyP#vVf>aNA
zssac?)q1NnA4Nb)r4eCr0s;>B&p2VspN(^U7!e?0+>cycP!J=hxJd|D>Yxw5?mwQ@
zW|T)HNZfmlsZ3NCtI(H3vaunl`;s*}FOyjCAacggfe-O;@$cHjg>Aqjdtp%tEQ~6P
zzP8nj?VWu<^cT6g;V48{F$nzY*I8TJ(^ZX)OdzNf;ong9?f`BjnH?Q@|6~nKmR|LT
z54;}6MtS>VA7B2D7NDwg4Kn90ZZ;nnosm}q=d%V`(|KXu!w5@TAU-~TNT`vxiw313
zU9`Se3qFBFADj;8#Dj65qPYW6CPALU#>(mq3e@6V`+Xuf!*`C6Fs!|mfPdCfdGc0!
z;1Vva>e=55_U}sAaF4El%Ys1es>-!B%+I8u`mZVVTBtcBC$|o`{(E41!NE*7x1R~V
z_;V188On^xd7LlDh$+m`v*g(gYS1lPMg%?F+}woYavS~l(&NgBXlIOr^Wwpf!61x}
zl=l1n*0X<dzAod{oqUTa7qZ<RC;{I2N_ZgomA#x!np{eC(9x%!OOPQW)7VYu2QqtT
zc-R8c?zoXr<KZv-riD(g1PKWW_{<+1{kq_d?nl?jtw2}ckS0WOe0|H8-w8V{!e_(`
ze38_z86SRtndr|NMg*in`}H{!sz`t6KrGVYcF;0TaUVHy1R_{V+y&yCL0l1{5#ksj
zT^BiTi0GwXV#~%UaYB2{6>oLKv9R9@H2m8)(BWahr9i#({igCqsw#yojajK4U05PD
z^m9Jkx9*Uk^G}|;>32QCts9OyVEj;@kDKVKQVtVZU*n-W9Pi(q(66YirFZ``_Ta0N
zCl{Jye}K|Zu7?7|-od+Kv0{n7H1kLN#6$ZaG!UqQa?`5?(v0B48c@v1NhZw%1%*I@
z=QrQl#|*>)U740G{ThLCrpkaqoxXb~Xd@z$jq#F5VS(g2*s+3n(QE6QmHP7X7BgNW
z>`H!+u0%eF;l2y$nVC#Y56wp=2dAV6Kp#F1BB;-cN2dMkE<rFwWV=-Q%bVsMBX7y&
z$?ZKvtvdFSQ_^HP1_@E#3kV6_Nw2=|3<`;$g4vAA(|8vZ1`~%dbIyzDNX<5L)??;`
zs>CWv2rD_^*J9Mx=S0M4s2LHJFzvSKjIEN|Zp|eT5vS9kTl`>d{ujj4feV*{Kd?=n
zcg7MWOhPRK-;njI_Fu31i0^Qf+HqlWe&0I)C2c8A*sGdnH|elpzq@eJ8cHDTY!N;G
z;@AA=|KzxEXZ~C3z<NyL*bZLmpFc}GD{|TMNT58!v17+YFR%D*JRy{!-m^4g;X0p^
zj>fw=hyPMd^n}&*zd6g`r#-B+dGW8=H(r^RoMOYJ|Bs*jKR>%%3b&1f;eY<l1oIub
z|Mk-TFaPlbnJ=w^W2K$!azzm_35gXrP5@{ZWgvOLl(7M)D7~6LO>y(Ta*_I@1E@p^
zubPu1L9kkMg$k&wH8eCZnQ;0@;AC54eSHOl7DOJlKba_C35tW{sOafgrZ$H|XAa*o
z5|cwhmY+R$uJ7yDAiTHH-!7r^@BMMXYMu+YUgCKyiNl#f>?GBu?dvpzALC#Y)sc@;
z?-E7i?fdr`!4o435@9H+TUY<@M|Cdz{K7)Cd+6pM-lZs@b+53kf{@A(Ozc?I`F#w@
z7A{CKph1xOb;o`>^pu5SXU1(P1R`Q%1M%JxWiv#f`M*K_ky!(XkmRBNB_SKrp=jKd
zhXFh<Dghy#Yk%`5B{kI-q!N*2L&a=)m+#x75?IMX1!QJn;Y)ZD@F}hB?aHX$*f=?L
zg{0@gh_DY@eu7*^L`M_JCHy6UEJnNB;#;?F_D!b=P<!-6l)RhUAH8oeBh^TOTzlNa
zFdIP!d=Q$2s(0_WK}2N^zt<AjTg#R8is5t@;yoKO2reJy?ssw47hjSzZH!9QLKM~t
z-Vp#-AD$wr3KdW*d6-oMP}K_!zAd`yBZ1S$@j(i31ad~#R^;%38wI!unr3)igyTt7
zzI@4oD)Y;*X^9gJM8HaDFfl?;?l_#6;B-Svwy5D$E;ly*ur;?7M$Z^KQ?)mM>=A3l
zh^7Ox+)j0vS;I=Hm4v_XTA0_=)2k!6(2q|lD98mdGm())py-n)d;9u)uxpDSON(Lh
zu7Ioq1F5CsQ?rgwT|cAS-2*^dVE68|IXOAx8)19}hL-+F|Bygc*3>Koh<v{|u6_-W
zMx!MFAeh|T+)Vck;yQ=m(1Yf)l?X1^uI+<2P*<VNN*oE|d5!GuOrAgS2FFsG6jX~K
zHkk>*fi@%+tvke%DyTC{AH(R4WGM31fK8CckOX0(-A1}-P;!&Ri?r4=7;u$TLg*J1
zR8*Ei-imuJk0x&(&ffMPpRRT6DO17xQ)vBsGo+;yygy*@M2C46WmFxJP#>d15<)8p
zZE0S9{+7bf7qOE4P!QjZfof+x3L&zEP+Ldm$MkPIBs|!Ytp)+e=H3iayR>r_VrcLM
zgroo%s))hmajal~@Ka#V9v;l^?8Nl^@u`Gyas3^qu%e5$0zy;+Y)QQT8*0%?wD{#;
zPap?cVeoyk-iL|``3^{3-5@k`d@TIW8`g6i)zooq?URTIR`K!iA=0&=;ND_@n9DNc
z7Pf5An4w$CZIJRh1em~v>cUPAzc8=)+4!~G+*>IGpRhe2u}g~H8(;Kjh9~X?Z3PO7
zzej$A@X00NmCVO?7}LjFMAVVEr6V|3-9J#Rbl^1t;PetV|2c0K)I<>du}>uyVy6(j
zH-h@N1dD7k5=-xmG0LsATu<PPNLzt6patq30~EhdWuikn9s2D&86u(ge(3N4my7vH
z(|v#d!!hHLK%|gxAZ3U~wGnhsO-%}eFEPmgQ&%ouev;~f1E1e8mLwzrEXYTT=UNA9
zd@c25U|@sG88kKY-f+6hOI`Up0g`$tY6OaG+#s?$0&5hZs2oAB4EE$2M48vz&j|O_
z;i=(2|NMF5*;!QZM2d>~Bs$)4n!EC+IlfbOKF;7)${rT6pHEj~P@KjLr*E~aC-g}W
z!q*@u`6uzbx&Q7S1t&;ajFR;H=EjsRlA}%OFeWDZ@Chiyr_P7|vwJ7P@yCyTn1!CK
zz?*~0MDR0?y*~&t(#P>XPUf*(j`zM6aJpYuRz^m}t5<9lv0uX|GAT|9F(5R4g@uKh
zbKnR8=g{HUDS$7yYq1*(<2rJTuaQ^>IY2jAwbl#ZkZf*1bS~`&aY78UQE6#m?-FHt
zU^C_r8PIkJCc7e(Ye6_h4oEilrYm4PpkID?dgC#)AB>>w&<@b!Q{Hg4Eyi^c9$|YE
z0$2u`8nB_HOvdP-H(p3h17K3-2*)T=7?12&$-v+p6T^vBw}RT`=-}&{nw0U*ohNS<
z#qSnif@4QRK~?`b?2!ogEd@xNh>l=O7IFvX8R4ib=_a!OzAOUp5;QSrZ~Vg&5)v$t
z-JtOhJI6HLNd#_i)PbDuR;yYwK9`2V2`ztj07qr#!w=QfN*UQWE|_Fefn*g*GvoVQ
z<vI_AsfR^HBv8W|h!xWZe<eDB_8?e{oIjPYnDQfT9)mm^!ks=)cmP!IcFCeWML%)>
zOFhzTSs7V!i7r&7k?z}T##P6C-FwiMMLxcM{W_`m5aNhC15+JnzCdssRjsY8zy#Nb
zuxGy%R#W3#y}B1?bhDTkU3eXPrf8gun1qj@sOU8Uxj-yzlzLYd1FUN(zw|9@Iv=02
zQ;(CP0S`i+f&(^7fE295K!M7%rUBH=5D|Gp7#$o1!lW0kIY9O+L=o2uUG)X1ieS{X
z0(qR0zjR3!%^@W}@D6KXX{hdUFVnO+Fs=-5?n~e<so*)^x7&biP!<5a-^f0gKMz^D
zC<e^r3DDhCRkeh2!ZAMda36+mFVFsEL+20?@pnPnpn=W|>6<9|z*R=d03LFNA0>tm
zHAi+3-fiPXot%@OCN&>0R-nQFNJ)eI%C-OUE)<n7JTP``WA5k(*EAOJZ?`ezG1OCD
z-uiegwx)qjQao^g9t~+l|BtfAIXy_fmHj`Q;9E_Z7L=i!LR!M2&)p~mNWp?Dh2<TH
zd3)AVcaTHC!V>di$U(`UVpcWh<tI$WT>+1zq^GwA_XD>}36gBG!7=4hO+T5KKr57z
zoxL2@71m9IG82Nr5|6*Nt?de;AxK7Hx0fMBMcZKtxCSGo0r=8UQBkdUP=LR0-@e`Z
zZD;c0Yt+6AB?i>h{79du^n9?n3nDVTHuJgHSnuV7gIa+%194y}HAXuRg?der5Z5h(
zc(y~3+mS25ID*Q7iyuo}m*JHHPZfZHX7XLq)9Lpoha<=_kEV;NUBCKXnU31&JC4A}
z5td(oh>`++0Mip>G!UQ);BR&<6dYLBvkO+1%^PjX@wB<gI4f4Pf_ACG#(Por%hrxS
z#4(B-4h&Th=Mf?rRE9zIHv0H$2&`rRUa{q>25@c!uw*0>BGtkzs710V+x;b&^b0u4
zudr7O&4#m(@4?^4bafB7xD<35y?*rc4_*;oMEU24jihQ!*^2dupfSjV%)sZ*DOp*}
zgs8_eZhvTWI5;HaIciC0eVA0R$*4|$W2Rk>GlEP!o`PkFG`3;UdoQJ@4{m3FMBj_S
z_c%-u*rAd&gr^-b8BIV0gbS`EqHVV<8*YUa$_YzKAAKX4GU$S0t}(<8o$GSy8di^l
zNXOi~*+@yjvWN^16eGR;!7xez7)Fo|+wxsYu_|;e)d4XS`7`Jj_NL#vSBE~vuVd+M
zSO#H6%nSsKJ)8pUakuSpSky6UMq`GFJ0oL-b$A8Lp~$nxT^vE%W`*>Qt3gdjr2RZt
zG)%4n1-unKJ$*=Be?B6%>nKRsLQe4$Qs7TzU{q2JAg37H2d(YwUQkjtHiIs+e-;~+
z`fx$aZ+{&clA@MV<Vt{<nE{)$LdUj)_oHM#aQde1?nIL45qw<hV7hDY;Kai?oJjZi
z+0xeCsQE1r1C~?FD^}#R{yuj6czi$ry<J;+XqO8wFK>Ewb`vqjBf1aNUX}R%O_*Eb
zqO7s$0|yCZ2+S0bd%;dyyuT@Ji5dV&T6A^)BjN*H_~?-hGLg~38Aw`xPE1S>56T=n
zM~Vb84}fG~0O$ly6?>gsa%uFhgQ%FC0XAy|3X@G_B_pFbr4OxIJci#EpYJ0_KEr~O
zjO6)JU6>v6Yes2cj&Zx*cO0{*!1EzQJ0$nVk*F&=psJMM#5cD-b*cgT2r6(Pr#Gbc
z908)RmMTU~61i0&th%P=Zq+E7kgWi<us(24WC1`aDT)*a@Q;Y|_w&m`bl#??r$_IN
zyGm?yV&8fKZa=Q2WsYHx5&S7fSZ=1a-4ab*-5BXV8@E{5+Sa4GGqA8o)qHjx>nQl-
zwj#;Gil7!J+Hd$H_G^$Q0)WV+4*Jivl(nOyX-)~Gd|<~)HvIx$?KhmX=e)*A>DSn+
zMDLBy2zUQ+G9(|Z%i?qm3rlG~!oCgW@&>}+<DH2|t3zG@JkM1@cLuKI^{U)EIiV3T
zD^|u&Hk+>{Ge*J_T}ye3w@Us4yQh6YRLMgNLtgas`}Z4g^SG#V@akA7W7yU=_Y3`1
z9Ug2Qj_F%q|H_Se+4_}K9(<niftlnJPdm)LWBO{FwS|RX(OZ>^#0h4^hLz^rxpSxz
z6L%j^Q<vgbkg8COpq{#o-s~!-OGv`i+*6d+jfVI}8r259_gXG4r)2l;4<GF8oo9Ov
zVv}_o$X)l@ztZPN3|Dz03OB)Z5a~=BcTKn4(*tWK*5(-H?GA(F$Nl$kbZaSO(AHqK
zNVPZfVG#T?o|mwIKxM!R-bN(Y8ccN7SvMy(fc-Imvywjk?nb`3NA4K$;(?k8cdj=n
z79%b8SmLk)8TS*`m8xTK_<lwP=<g1KU%JiDUi>mV-1zJddbV=KGnkhN{y=w#s1xk*
zN{t|-30nCciH0w~!RZ-{-J2njk`bBr%IWFtYD*%dYk>eFlsqupnR00(yEl+SYpM&5
z9bI^SDxP#tIG%aT#7OI&W0c>YEE&~m8Ilj&z-iA0#^rbsY0Mx}Ty<)ud|}=a==4b>
zQ&R8>4sz~<eMpL9ADuIr*WmVZ&dzKzp&!upt)*lj{~n8rjASE|LO`Eam7Yg|iw{L8
zesM1M-w>jb$LxeM0d1Kjjz~6AakrY(F=vcuo+R@~rl#fXY-cfJV~-q4CWH=~ka`@`
zas@E&x%xq%L=1n})YTPt>h$S??rx*}b97JGuJN1_?4mg#C2xkvzTCUDzUC8MvJa+#
zO~$ygPW&+TMuXLiqO*nqB;w6|?d$nGo0j#FZ*4+64434UAT4e?bLI>QKLFKgxh(Aw
zF_%-{U?SAzN<0`TC&m~QzS2|^NR~7QZEfwlsuy)OR!UTWv>gGJr*3R~3oQl1X!nt_
zaFn9XLJw%9)`!<2dUZiO`FxCZb_cAg*xr79_7vqxJSGfChqFagOzb3y(R<g2pBw8o
z7#XS>)*XF9F=O1;V!XF<7sgA7chKp5mEF_${_?PS8yJLD2IY<YV<Mp~t*x3TrH@tm
z-mN+RhplPSp(`&%jfYqEtT(2M(HA?R{ij&tk4k_VXQBiX?p!szsxM_~gvpH&XNAjX
zIZ@{-BToaxQ5m}hh~0a*N8z7tLR)+h8kFULG_wYsDyDe#+63mJp!z3EEin!*gnk~Q
ze4K{ib^T;>JdupRn8s<WdpBSTdNeHEGgHC#$57I#Acft=5k!H(X{egFJD}txH_$*9
zs=K)w;bz(Rf*noFhFua4s%<xkF(6182UW`n4^n=k9L>wi^9B@;il%Z7Ukd`pZQ}?f
zB`+OCGR><4xM1Xf;{-Js<L!|$=Bs~975GSjr)G?D40U%ehESt}b2H(3kcx|7U}RLl
zWa#|_zlcoiCk;U!fPF6;>eS~bp`HJZn>!ro4u1N_+Q!BbV~QD&hoObm>eDBrBw!-Y
zp(%T{AoxO_3WxuUIR<=o!5d}!jvb=io;o)&8?NE?=LS%81&-JwBO|wQ?Z-Aj_(Kue
z7+zzs?_zqxjdTD6Y|-1NaCzgJui_7qaAUh%6ecwC*ssf<a-_<1b!TrPRagldEPb+#
zZIQC!i_OiI#L77*EH*xN{?MC_4tmXLdXBWSNdIxzPeU#Uf!gc$GdBeV8oZ`{GQ4oX
zeojPC04m(P<xkNCT}PmeJwbd_h%^<lGL`+kq+P36SjkFsnGm(zKaNeet1}allEQGC
zF^0_r<ABd$KMaXb0@AlYhz;kb2Kn;Op-BbctK>1c=P_DxGQWTu3^xFI`&)2rqoLr-
zKw)A&|6FQnsub40qTXjlj%D9s6GHQMp`{q*fiKDf(8QR&CKduI(;H_T0aZ~~KUh6P
zd5t+%j(3{B4B>QO34~+(&!1YT>$eoIc=ks(8O#9c`3g8X4gJOTYKj6d^x5GXq870#
z==@&u7C9VWcNzNk$d`MN3a`K&0oaO**ET|Qm3w)Skz(9E3^rdb?VtL1+MbSo+=}Nj
zG~SozVQo?yI<qRm&q>YKxR|`Jd-gmz=5_isj}<a3g|W=j@xP;@7s8+I3D4R+vp~&1
zPJ33g#l7HrYHB@afp*$S(`)(9mm63{<a#*L8`Mg<-B@k4>Fwm*nldX!1CgSu1GDo|
zlk-N8qK9Q(ew(`R>w@}VV$1IDEzkINOnxvpOP!pjRk)dbqxc2$>rM#>R9<v8H#Um2
z)NVkxk~(W{Zy$oVS@Hh8;UAhAI|v^TX_sLidH{0f7J(|STyU1eMuO}aLl}uVGtq*c
zUshI@>E6&KcXz^nqw<Uq_4+fi7IoCNIkCmA2R4-#(=GAjSV4TC$1m>w+Cs*HVS-E%
z%9SF}7K9+ISaVx~;bbE@kX6bkHhhv&PvjWwMrbmFM5chlsApnw5)`(GT*815ib1GK
zPX7J8D6K{9hW8jt;c0shZxsSoIM$AO`r9YuD}6XSRWTJ~p57h|4AqMHNXk_B37PE1
zSQxfY-#3~;LnE+v?>cnz5gRuq?69M!F!*(+Y9G2Iew;I$ftG8r8&t8mn1m8UbwfA7
zkw<Cy6H_lu5pS>HbskxJgK7El=cqJV;j&mNXOE?qE1=wu&$-mp(DD(HRXst51>Y=6
z6vkALoj;{06uc49E$GPrqE(=>VAh#fsX4J+o+4~HtbJIyxUK^4+uW3PLn1!9OuFY1
zlu|%Qe(q~-!l7CQg$<cbhV|M8y+_z7Oz=T~nk2$dCod@%Vg29PElwr`W(JB(#>?SU
zkqRip<u|fmgxZGq=0l6mzUmIZMg}G(U!Xha*L^T(-H+1X!rb%@Al-BjCMe^)VCZhG
zgOE}@YT+t~hRjc%^h2h$>B8=_B>-nAvJ9FL$uAJD#ek@m0qcaj0UhxxK=6{bc+OC2
zhwY2LfO;$eQxKm*#U6CBOT5)>+?D{5S<5M5bQLSFmlF37*!f@}A&-E?kUkO`8&vwq
z(avmxE9jh3kw3{!hsGFjvPq2SUjTw?sU7>WDM;BDIp7Y8yu61GRp3l;``*1}6zKv1
zSCLx^WrZc!v!$G!UZJQj*75Rg%f7My3cA9W1cm79*L|VQR>V6gbLy_5KNGQ_f$5+(
z{$&-knrMdzt_nRW8|D%bB^s|^MO)Lnj<A~OSn5b{1fq>vWswd5uz{uF0(N^YPMid%
ziM@lRB_1;}B#fitfwP4f5cAeIZ)$+s{OsS@2qt(fh=V|=l2UQ*$dnG@UitV|0<R>u
z6SJEK5b{2+Qiz%)05dX1Wpy<z=GKVa5qWxuS3I~({j$rYMl+CCRx~s)LNfp;J6zH0
z^<dedXp#7^DGY#r0`VUw=d~YIlRr8(B03<hnW*3cpw%Wmuobb4rjXt!U<CkCufTN1
zs_Pz}3|PyupE`eOmOgyj`zX-+@+c;c6;TFo@$fJMf+$szM>XaLBe2aTPfJQrJa>~_
z(d@`SE(t*dQJ`26TXzSG7Anc(XRUsB=o?J+_(k928YzD8rCZKhyg~N#2UxP%usC}}
zitVe<MxgzX-zEXaDWXCi0!@VEZQmcUe(%I+`g4Hq92D$+0<%Y8;un1@<M#vPa@aSq
zID>>4F%d?P6BMBXl+I6NSg`c+gSrKWKv6LAD%3f_x(`l}<=LWpo_pQJ?;B;JT1M=E
z-MMrksOyo<OK@UgKy#plna6x=y^D;ANpD?1kX%aT!!>kyot};k2fpI&;2rRQ0?Bm@
zP!z+C9S_B@`DyefW*#{ZC=`JcYd<)lGYdz<Lwh!S!AYt5@WEI+OZNyIh3hqrq75lP
zzSQ0&5W}QljhpMC_T>@gTect~_z~C=p90J)6;^8&z<mYx?K4i9Lls^lSw1U>Nq-eQ
zEI-r&_`LX^%B%jP1>nF6I;J!IpFTZ@hq^Knr5}FCX024e4DI{Z$H&K6VS>=|gr|iF
z;TS;sYGLsi$`e5fGDf0Ll=#woOB5}3^r}v;mpEYLWGTmQwemw$&(P0LMb(sc0sV~P
z`-liu8hM!e?yfEafmOIy+L(AJVmJ9gh?xy-?d+KG6SdF^NxObgFHOGpx;nIA<^(#Y
zmCxvV4ht&Ne_J-;149jxLq4dV#reSHQB-NU1ZA4uw&EWL9IhRwS0nKG;hb5Z?}0X}
z5)sR4#^|Znf=Tf@NPP{91zfDjm<60jV12h7Us(|%*RyBOnwguYNbNH|(Ro{ewlp*v
zv-<>+M5L<#JA{{lO#PzrM6B36bGVyGt$>uRmNZ$1^oBFJV;l}sf2XGU5dFFetgeli
zg=K%@+U3MAW8dYC0;lne`el53z)ux8m_Rg^j$6f;f?**8Ghi62@F@UC7*?#PMKRwy
zFyM!IU$Kw77hlk9FlSoul=U}Qsr{D_r9=-eW#(O=v}HAJ2$|w}dg(V4MlBmx1QxJj
z)gDaH*RP-AAlNs6=r#^d?F;bR=0)S)JYnRwz;suVCdTFb`8u?b+xEV^=e00TjIt1J
z@?oAyCfrev!T1japKtdmanLAa9@2B}x1JbQwxEA}xs@hv4cdKrY!xI(D6p77ST=FX
z*k46rd@*|DUjokzo`^`c958gzghe(%-_Z9=6@>esO2_*fgpPxspPxe4?T3axQ};nI
z76+IC&APZ>7%wLqM+n-b#y=kERibDGkUMQZJbsApy&Yzap*Z!fds4l8@N(hVitY(f
zta<lNwr*MA^$7-nJOCn?=FxqdvYKhB3~dD1gR57E%7@6*5<p3iMj0cTS6x0MqL6rp
zuz5FcaG#d>8qDy+%7PgbqHpk_l=y}8?dlYE0sQle!@MSnm-^4|*go?KQ`j2*&#z#q
z|M8nKUp4qBC~IFmxM7u5puFGrx3Wns5*+kB?jtnGrXtHHCc45dAEwF<t2gE5{PROH
zs2@Lh=L~<m-8k)g*!zF^!9n#eru<CJW3ZL_pI<{WL&HM@2BwKLiyyN1L4MNz|8M@k
zb`K7MSbwX4L;lY%t@+;N^5jgDU#Ws2jQ<aRGa{Evt;1m5hE<FRBIzTVoB4<_+o~YK
ztZBJzChrCLbx6yMQ66ViZNsYVr`s6-sx=nff*TH?NiSOn>L)D#D(^mR2$L)_;|Azf
zZu{c-BbTjZuHCzhLcd9NihcRU1*)vxd5IGa=`N>U3TSF_uJk`&5C6YK$w!z94GkrP
z99S({09nnk(h&7H!wBH=*94|ol!=iyZ)T22mAyb27@3rmGa~h$*ZL&WcA7Xa1W->1
zf`tYFP9z&)x%-Ii#kzH7Cr=VD!6;F^6}aYF{H>E`U>#06W;9)-&%!rHOm8gcah#+=
zu3fvLEa-#MWWAgxwy<5g4O>x2?i4|8=K+28-;c}62MA7l=Z+s|hqJuzvu7RwGfeP_
zW5EX;z_bT^s)!~Qt{fWX;g56s$-+Eju#`J@XpN1HLDcu5GDD{Ef&C~vaaTg+)ee^{
zRPUNx7`bLc53GVKf^E_Y;6K>8Sk>9NQ?(%z(hEEc;>3ouJk*sh1R@S<))nYsu;A-+
zU~RV69eNVvP{X94o@;Gwt-vQ0hI~cL#&n=+w1^HyhKPU@-M%F#;kCtjP`;vCC(;jG
z1~#E`q6EM2Yba9jJra0DWhD(t{u|F!(LX##?EL(0pV$04+%Zr%q<_~ev?k>SKMbQg
zgg3Q3&_<ROh*aB}E%eX(DmZPj^d>bs-bV}nYbI27iTgdcvE07?)j7~1eh^~d9@C)@
zo`t<C$$c=sy9$c0AIc8O8`5wyd^|;=79`q#6uqUvTButV$zjCrYIvNYO&*_|41@|0
z#^=6JQ8yU3L3jZgM#0pS5BIX>Dw~iKR6khFOOM{Lz>O$05vudB7pmmR6HCJFpcO&Q
z)`uJl<U#cL57_^~6)FXs1hL9Ohfj8`6VV{~-snwWaBofeKnxTsp@4A$Ic^|_W{4<I
zzqBNWqqVTWmFwr3(JW_Sk+yE)!hgK)eW&t~BP<oMX@=-QS|REtuOBg81$|1yBBg%D
z4s{@MRiO0n#0DPQQ>Xff`V~(n>eOEFI4b0_?K&lh_a4BnI(#K!t_8xAbX829Xf%l*
z2v#Rl0aG=@un)9WWExqKq1281TQn!VD0nd})Q7c*wfM5;Ha3Bf&1`<E3Y)=7ZyMUx
zrx#rb!3TRg%4NR$KWXf<sFpk=1%ha%5TC$x5e6a!fF2gLZ0^hftU`L7KOUDaKP3||
zM8<?Zz~u~H1iycTRzcY#cpH~^V;^n3mJ#|Y4oVpwlh`H%M4v{UVg<u>!_NKFb4D8Z
zU3>PtXap4@N?ip|at(z0i$>_jHsj{_qv_8kHZi2&M$cCNdj{v-9Ok(|x>OVueHGW?
z0qI=`r5LbozH|ZuolKMk9$`TN3N2=ev{Skz=XdUWw0vfv13c-;d<?7U<VfHhMsa2Y
z3;d}kb@uHUktpiXs_BgFXbMSAPNw)#*>BXFpZaM-Xg=bmX7F~gYS${ZHz4Ym(mydV
zF-yE*3mO}J1de?mWYwYYX?Pf2)?zw22V#Xr9)7!uV#bfPb37c94NrMN27tmFFNXKO
z;cW9grdgO{AmmhmZ;gOFM?fjy<Xai&djUf^xw##<e3{@<R;Iw-rFQPDL&x+3s<^$O
zTA+0a5+F)+1m2Q|E$)IurJ-=2{gnkHqZ4l4`~-fntIT5$iL8UoylCQKBfJEwl3eDX
zfm>8#;l+daPE79}B*T!o)kot^VWC=K8oT@jR45XngkD7HOJ0)FP6^ak=&UGjG<EBq
z38Ut7a&^5S3ghop2=U=Sf$C%0zPPCeDK8Q5v)}FJmbQ&dKZ-v})EA7bw^s6e(bmR}
zUdF{rS67!B=xHH_Cj_&^xEpitizH7xYNGj~>cziD*Ph7Mmj(-<pY#3uMTc`|&!Qg>
zw5W-iXu1CHy$jyVVZ~za=ok*s9f5lE%RTp@0gW<$0?`)b4H6o<(XFH3aa`cj(p>l#
zT@J@}t=G@tta>wB{jsSbit^RdwjJ5=BjsR(X{Zdy(gOMq-niDYKI~YE;!^)&@Pw2%
z?zfG;M5rXBB48)zD1=R!YG<0MG+kYkqf8teHLB^@=>y@S2u=rsl2Y~e^@XSwcL!HP
z<3^m5z}q>&z8=nkw{eFt_i;lUHuwj=hm+yasmY%}Gs)N=o*;QrrLq)crZ*JjfL8dU
z8#anBNHY94z@|EG5r@SN`Jto>fD9JT?KT!0Fr7hEp(tR5O5lCBjvkhs5zkedJpUqW
z`+gRd&~*=Q2;VB(c;w^8O`Bds$8OdVY};R$IdiFrz#0gN@x?Q4)<)K8(w>Knjd?3#
z2i}H<oH8?uxU&n7pMdGa0u$>hU}9E{`=OPp%!os43Bw(x2CY*QGJ61q)qtEK)4^EQ
zv=*U;Os7NUmL<i$EvqE@_kV>JkHhyF<>4_(w3Ns`Xj+xc%x+grs3KcZ-mqp*XXLWF
z-aHGVW#bq|F;KEp1<vi(>{#KDw9yIQUk?wft2f=~lmPr$?Q&`a&*<WQYirP}xE)Ku
zmJ{3`bweGxIAZ#U+_H#yfu?qSGl&Z#`|*&_iF3G1r_I?DPP$Apu803AgOnF|5B1o?
zq{9I!7zGG<Kmn&n-OzY^@(-lGYfX|n-4{N`oOLV#N$LL>&=8XK8l;6Aqgyr^Z=DtY
zdR^=5vEHXpVFbalQwV|xByT1HoW%?O-s*y^07Z#b$}F%l${U^Nz*O9BGWLet2VZhl
z{6JuHeq!m}#EKEe`ce6j^#T6=pEGl-{wp}Y@(O&-?oH)c)Gz_gFo%aMlY0&zY6Z9@
z5J3fu>5V?%mD0L<hy&%r5nbUAcgFDhDZX9a2(Pq%3lgWk(QZLl>4Q-wDmFzJ_`<K9
zxWN)62&iM}ej@CK+W|g+SXf=;HpQ2up#Z$<WFVEV!|WX7lm6>|6XXlc(BYgBK#>Ha
z0!AkE{->m5T~j^WL^Od1KtQ>rol$_|uYSf<cpTU)6+FX^0ed<mB0puph99oc?*}>m
zZnc|0XGp5K9Lr7rzV3}7$6~3&(Fj-QkOKk(mzk7Z-twq35z}74$9#~6QgBHAcU?Vx
z#5T|QDI0?sY-&+J3I_Ovgs_dw>p_nB*>D8776D5kBW;B|qokySA9WbMmNGhhcqwe&
zqka<U9F61-_|=g`E2v~?iDw7~^l;4b@R{n9Z)Z%9pNWAfw!4I(iv2%!=qPOlGjb6M
zDuIfD;8(!${DB1nuqLNNRfTlc6f3>$6X+j|a#bQagSAt^Mute-R4`rwu+T&rpU!#q
z8?Kz1cMh-+Cs)WR$wa5a$ES-n)A;M)4fev6X9ZRSqeM6mBh{pabVC8!tOVr3#=)^M
zLHf|Y@At22Wg7V41VIn<qTt9Aw^2F@Y5lI|d3d1o_4e|ke(?u31Hba8KoK7yyXo0J
zv3|)7oL>x~fY?^VSR&?VQ>>t`KnSg3XJ1O;A5&U&|A9s(AQ>X|(L{xZdQ}3iiZVWj
zc(AUXMsXZ4^d>7-3O5`Y5-7$rcgf0%6ulG)Ll|eEK0*KH^7+3|IppT^-=F7}&x@Sy
z$lgMV5hNfyjY$21lR;{S55JiB&8B!2VVjYv!o&j+2qP>pqa26Lj4;HfP#%_7?I>N_
z29}0m1xoX_RuDQ`5o~Y}!~xwzDh!-ZDA88|0DYo%<Vf^Go;_~a6hD}O)zc9j&`2~i
z;?!(POic6xTnfcG8(z~jC_2_CGUR*AxM^X(C5n9^MDN1O;b|^EyF2b5SEX2?``qox
z`=7q0K#1*A@{SwI@dO?Rmg*06aUYNd;4%OCD=sOTBmdo>D^FedKkxnA|5v=>e{}fw
z2lVQ`uVVk1W%ut9LKAsJPjyqv^r!z<`^tYC(^RkfA7P;edbDxHmCc>0f<juy^wwZ0
zQsWB$&7HHevt}RCf(okn8>1IaE*S@gzxjR^bD+kpC|)Mg5NhR&ynnP3&Q%#&&t%cI
z1j=>hs64p2<JA*ANtoxWyzQI2mzJJRriY|jJ5})~h4$_kyZANsh8e3k10qudGWD_o
zkP1IdKeXGu{yU0L{elb9*$LZyg7T$Bq9j*>tzIX4RxRa$@wX0qc*56CRD?<2X6Cfu
zMP31Chwa4$ig-UM>i2L58GDJ5F!05jwj?~rh0OnD=%*|>G(7wX$!&L;1E5ob{nuXS
zBEP&aM!`vBkExJbNJ;20o%_~KOm<kYQoQp~j8G@YMtDfH=jyF&OV_BuOQZF+X*m|4
zvX!{rJ7$qOVhokom3@1O{Z<?o&mE1?cSB=e0&Z;57$^zH<QAM(n8a`HR6WJc9<{}o
z+YMvzxbo&-H@jy&H*XeMx<iv^&mN=4XAkD63r5N#_FG4bVkByPQc@Be5bG!P@g-W*
zkMKZZxwEHk@oi`nW&YAqhd_-9EtX$nZOdjRN3S535{e29Q9r2|v7HqSNZN?zwWo-=
z7SG4~E`XWyV>KwAvzxnwCYc+vmPn312fw~9@f_vG%28*t1<!SR&oY1st1k#g^?paM
zB?$m9L=sCq=RdDW+px(J$|onZvgMM70XXRO@1T|Cym|8`3?k#rz0jvL&mCDHme?sO
zKfe&$EVwfP&<`TxazlCMyRiBT)tu9gz3ULB#ac@?yW?Z(B;^)DQ1K%1?Lp5B*rLBS
zyrsyq+SH4<T>+^sD1g|z=u8_u<SI%~Xlr5NTX)Gt_HUgJs`K-t*2y_5=6HDSah}};
znEZBD)-?=okmb_|d#Wua&+^W_;3U>5x&5EU+m6a{=&MIP1C&T?e-OL~Pc4EvyP?01
zxN4GKmJso3GT-A_m(Wg6sv<RD<K1#Vt?-W_Q-6O?)wAvm39H36*nymem)`?@H?U2u
ze);6Na_RTGj1{;sx1;i<y(JCW^(0LNF|vv;83<}eBXO_6?S+U%xNn{X6CVt1!5l<`
zTsYXQprl0m%>B3GTbIvyh`Nb;otNM|VFZI^$raxyUWZ6GaS<l*6GdbtrS1b0mcz`<
zFh>4_b??CS`}?(?=!)a&_!0gInV>s3IJoTl!5h6pLs!KOA6f7If>SB|>s~UMvB)+l
zq(|DK#?k$3fAp{hv6*!($5H?clzen!G+A+IRC;O0%I~u?;wMU2k=~x;jYVKzCMdFX
z6)M>ath_Afe=t$9bkotqipW+?3=?eGV>BIXaQ6@h$LEk<ffQNg7ZC92^o3oxd9&3X
z%f5dMMgM0$@4FNZQE}N%!8&%aukw8WkHR)R(a9JwlMv$W+3@-<`jMHwAFi1wk9To$
z2E5SAAq(wD&05RC<1{vV@#*EMRY%U)LqV$AUFNzBEzDBRHCAH)r!#a4g~?C`5)!6W
zO?A&B$`skX=fq2ds}4n4vU%5%1Is6@8bsgMRd_ykWDe<C{6w*>528W;tUq%>@@VPQ
zTm(A|@p{e~A7Hx7I36LOMC2Cgw`4i6m_omAFz{-<8kwMwmF4!rf}&P|MO=*I5?VuG
zL<)#6+h)DrK2>3EO4_FCz{&MFJp9xtVh8tU<ij=4>eg>%zBI5CCw$idHRACI$r?{Z
z?8eh?-%cO(9q+t<+^Rv2g84ud$jwL?MtpeU*3bwA3G=>v&(`%E_q=v-dNpPCF)u~Y
z7!@%gQ<_ZkC?H!RL#>;62x{=wb<phn(~C)eh=(d<__t9sxojH*w~@DL0~sZWIy_HP
z!6hlVHz;UJD=uMgXpj%EeKbRvnl{h_kW=s{aW|Tn<36=2@$R@wsH6HuN5x}jX$fzy
zovF=8@CPUh3rB2FSyJB4{dF(q#;`YWL+#NxZ-hVl_r*talM;GXE`5Nsm;ouU>ZVOp
zMd4t}=V3qAzm_k#i}DMVthQ}~OKqxkC^y%Bd#<F~=k*WgQ2x!IC^zZoDAC^1e&3i!
zC}3qB%3rI|t?y}x^9=wgdGIQl@1D~|zijI5zRRp}Yj6C~^BSZ?irR)woAl$P(?+CR
z{+{^787JOi-c_)W1Rlp6Xv}6^v6=AWzj?OHS2W*(8bmka#@NMJ<JX57e6FpLxDEby
zt><m~kEu(k-bG{!&?6jF8diBLGLSbA-N4)01XGA&C<+kO+K!#jF`$E0!c$<95@`(L
z@bsK2IOI58bPCfw-rpU|8|DmUUH(e>u?`LN@{5uiiXgB#c2+_S5=*#zRzc^xHT(S7
zI}Y@6S2Wf)cck)JVs8khWGzEtV-=0j=|i|g<KZ!9|0w0gmkdPw98DW@a~i6|sn0iC
zY-c$i(}|RzAvDXBH$-X;(MP%D)}vg#0;G*Rk($FX6|Wi?fm+J%ad+R~`LO*0lJeir
zv8J~l+_@5jy&|tZ)sUVELGRteGYO;CDbSi#BEOOz1>SUAdrvYr^pw&<h~r1MY;bBH
zJ9+azM=1+dVsens;vEl8n8VnN(18WjseHYUES6G4*=<nl4(=TKMF_}lnm;Qtx+GPd
z;73Q1RTolb6;zarw8Fzo9j#*ZwKx}z)<vt+5J}1KNZH}o)T7@&B6~Rb)XDp+LiPX&
ziXT$>9kACf=iu!@QQBK~)U=y=`Jbu@Ud;6gafJHTii}*kKDV<gA6)dV>0e`^Kl*F?
z4y>>qJ=S*vrwWZwCGsk0!}VBF-|&a~>#A-H6pFSVw4vKSy(w9zi=#_`f9*kxFkI`>
zAA!aL5vUb&NoY3ehvht`{Qz>`vC7gVQ>KTaHVb66wYOGaIUZ@cetqxZtBCcG2FP~>
zzb&=5LwzCz9kvlF286n1ssD|=H;=}$f7^I38l*JOb7@YaG^vm@X+n}T4~izugG{BO
zG)J>U2}x2(6OuG&5QU^@QZy-<BJ9uUdEWj0-nG|W>s@Q_KlUH{S?hV$qi*+o-PiT~
zp5OC4&f_@FL!5eZ7a1rwZd?<x`laX|`Cn7(?rt$)D`c-$-X}Z@o2to)@7H%0M2@68
z2jV>XVymkdFp{9nsL?`m$WFZ6Wzng>`0-z=7F|}@5^lk>T%Uw~M<J-GDqxY;mc54^
z8Lm6{J(-IVwbUpSu6>T49=U~q(S|{AWbiJqkFm>&U)pf*M!slWw>oX_ZT<Bj44{>b
zqo&NsywDmeLRX6XYEFIByX{TL&k=pz)*THSf1%%}C^+G71|=vTxVrP2wQE*DBc&PH
zgtp?K7ER9f+1OP-?d3}urExG|d<^)Q;xO$-#?=WeuPjS%5?PwxZOx-$V^4055|Ca+
zSXVXRmXB6jI4%9<90FdAmtj><fb0*h#s4Utx;J$P)*PxrBRSZF+EK8jX8jgD*|Iaf
z!tD(0fhW_iK8R{+C_XRs3{Ms6J9iRX{QRsb6Kdn^b{CGJ30j$H8$Ay7OoT_$<3r7o
zMcr?do*KStY!o9q+bu|FBh6mujH`bA6VtCi>zRl9D)c06!IVEIoD|y|$z^DIx=uJg
z@H9hR;_DpLZogHCJYg$hRB0E?eUuV1ZMi{3_~XtU#w4uuOgY?hY8`>k(c`@Se2T){
zZ}s6`;sero)27oP6_j$NF+<|76T)Mh@7Dn=Z6RiTSAr(fMcGLodp@o172i{8GY)C=
zp4oLv*6&0oMV8a|!JF2!gliVGeV(K!q;<H}om#qT($uC+U!0p#i@x=}!iT?F{TCNN
zE|Cdms=IHpZpBBDvdHbv?@v;uLIb!@b}x0`U%OA^cQxzlXE5LIL`%dk6o~c9ezu1Q
zyJs}r8ppIt8mrA`3kTVv<c~}R;o#qjwZ_iiK!HFo6xmMem%^-8b}y&#n|rIY`S@ec
z*o=%zp)fPVmADvZ5OC~(KHf#P<29Vx^q0(gGwWd2IIk^a{oBhgkIha!zI5U-9d$b9
zHfes<Zd|`1#$4{Z+#e){Jxa2@HzhjlNK=K`H;mbd``A_<SmT<M(ylo3b+-s~!5!D#
zshAYz33Z@q^-UYZ@jAZFer>Ux&Ht8IrB>4(w>e?MI<!+rW7B4JdE@;3>-W7;-;2Jk
z153|-|Lw%Pjy3S2qxT}bb$9bSX2x$H#dIQ%ucDC9?GP^y8X7~Ys4lo19GGntxfrMN
zi*BR07rd%?yyp9-PiHUgwaRoVI|Xa;g<}cQ=bhVkx9}CDaeal})n%D=fX2oj9on30
zH}>$jG+@)yR)3?GHWzAU`JHND_P(Qwez^bnv~-6Zt8#AYk2ir+ZVsRumI$1y6CM_W
z2T9o4^Z^d2r_vp(*78tIu-SO+rq+qust$j&4DQ!yEyHGKT-zXQJ#zBV0E-ki9-Bd3
zz=W`GG9rUmnRi#6Hr|K}d&|p8Pg4q6mfmm68o#)7_73uuc{sBpkez*Y<=B9wc@LIL
zR4R!`B-cwIdGGSi$N*e`%q`Cu#TOtZU9)c09fUU*7Z)rRUDo~77A;$j0HA_)*F0t1
zqkg)+Ivqw;w9?g_X?1>SlB=22f39kE<`n;Y7hh*b-dLD^>vA({!_a>bGPQT?fF##B
z+Q)LO04zC(tT(M%GQ1lR9VPHqrg_#Tc*gFJbxVZPVeK|DO~_Ey>$KR)s($w@#&tzP
z$L2k!(DXvCSe?V-HWiV{+40R^oS)VZW)j6_Ta(Mm;nNlFhO29g!Q9~4*=5r)HrD`g
zs{Ye@_AYVB3h;5-0%Mlu<bun;j1PA^I9Hzp(lXq>|BpA(dr6XI<TSuT9D28^xf$L$
z#Yts4+NHrQWKr(d{!>!ohndi&*x%Y@V5;ptJc%E>5wUl*(Q0Z#9ZMR;v~PdixO^<r
z+Zs{i#3GJu`~|C+n##px8Le&~j;!E;TL+(rTKn54Z_BI8-_o83WWV&Sg1LjYZA*X>
z@vJY<FO4=$T!YW8teOewSC3w`Wn!2jo|iS(l5fYs6r%5u=HT&bIv@y`d2_fOhiqHU
zFaOqa4Lt;nmP?i}B4~kS@0s0>oj7pdbU8uUlz=J1Z;H68%CtQ}EE@ebD^$Abs2eE7
ze-pOvN?IC~@4Q_5RqV%)I~q<a?6iN3Nx<@%yV~@tjtI4?Qnh+D>sPaGY_>5s_cWZz
z&XW-ut15XKo-zP(1J)IS@xOlGxqtu0!!ZqJnr)xY8lHyM58lRz1@2WO$!zk=ONB)o
z_+ZSok8fNyS5Z+&|AdTaoi%Cxyr1`v^(S98q&`zo`c67IhqSAi{_<AB`ASWArC)xV
zdukdzyb7P#sqT+no2=sx#ZfVYE&FPLD!YH|0D6)<zbA=j|C(>0DO`u#VFRY$+Y9&z
zHRVOZg2J|n&gJlt%yksX5U}Ti<R{06G*zA<DHytQ&?cG!aRp0KFWS5_Oj8+>d7!ge
z%W0O4EN6IzzY1Uc4y@R+=(A5v=RrI6?!8~=oN+mH(W6YXvV4vJR$e6SHM6)V%HbZV
zqp@*yVsyMTcNuK-XG52b0ne5$ZVHQt+J>3#;|s@yebc}Ew~f=<XKD)C?ROCR6qgsy
zg#)kC!eGLnt4dkv{P$<ZuKMm@vFXf#xP_NyViAMBzc|Elv-F7uoOC23V-A)I9qWlB
zE74m2oW#>=g<H3_>)NLtM&E|_CwkXw-Fi0+THzI8D$a3Q|K9Gldsk&W(i-wwvuC?@
z2Fu^yW~WF<<ux-ckU3YSIzUh<*dp{WO+IWZxr+76&*EMTPR{HBng0#_z`cC&)c^!#
zq(JhKd)p2(Br(0UskDjkVo{o0wEev>&Gz<0?;)qD)5K!F#q?=0z;}zke7Uyjk2c*f
zc}FjD?ceVg7-;UlZI@+y?Z&lJjNSna5H=FzR3D*kyX|{BM+_p~Z&jCiF_&!te}BKD
zvq&3mQ{(p9RQGFgup8)sigIU9n~{t^NW4C-DLvz{zR|2@MIqVOZhe{cX?cBd;}fk~
zzsr7GIZMJ}Tg^eMKWR)nIY@;L>#p9nKi@?-t=Dg%EW?X7FGRpBcMfJ`pY=axw5nAr
z7SqZ((BYakJ*-ODx47W2ar|+;)?iE{S5Kb~gRVIz*xg{n64(zQ6{KDn@gx72lVj<7
z8nS~n{W$S$P8uC4A&S1dV%v~;MLYTNJ&A&}3nA-?i_-OwkE?!{`koqVG;!jC;ySre
zM|-B;ds>H1wjsHd^`?o+>_TsMr=JQW-4SSM&Q9{F6B;&OML}q6aPZ!4@!k5>u=6Qu
zeffTdTU^7O%Ip!W`7(a;$d3=M#1FsY!CDwR?CSOODOmO9?KgDW=x^MDr&c>GMB1Yo
z#g_t*&Y)m%GGEvuvb6BwyqwHv{dKvos<BQ6t^FagiKXc`8cXE+)N?wFxP5JV3um#u
z=^h89yBG$=@uh$Ld{@bTQxHM8mOxU)2)WmV3&wjcWw+F^si#^o&D!%QiS?Z~u_AH*
zS-x<K;V8cgGe?jSAUSc1$FWo0;O!%|hwRk3zt2u^-?L}Cm`nBS+1Ke#RY+@;ov<DT
zEPpAZ7NoeRecN^NS&dg$mU#$7c&ovIyO5Ue<XJxwdb^wvNUBCdx-_97a^-TweUNbz
zP-)e@9@-H<67(Sv9O;pM?b?lb98ro3RlN04G>6}Ik(IqK-qDa|aKfcbmiatr5`)&}
zb%Y<Q*05p2HD6w+AG{Sg*T$x}Y+!*mMz&%0$=#)JU_h(5*C3@f3AgIcLZ7oWVA1B2
z=OzqZEs3{fnD|YJ#u(gCQe82TFDofJJjCYSjj;{4i6p`6%dP9HCBGjEEd<W0&8zdx
zZ*`eq=zU{t*1bpIStp()Cq9gOvoPd#Lzm3%rhZk)Nvj3l+ue7;&pA4FiG~--i^Jx#
zm8;Xe5R$FVvJ?$)kwB~P?q|oJtaO3lsCV1jYV-d6^*Ol0)+e-q$uAsZItD`TWYKmS
z;?VJO)we{Bi>tDX<?)k!x{)(V7h1hxm-xc=xMxAz?O|W^W!t5!c+mw39z7L!dM*oz
zT*7$lr+Q^IXqYU0eNp9e`Lv|*p0mrn{sr9Qr<0VOV&uf8O`ELU3_Z03DHgv)IziX7
zRvT4R7qc{qpz3v-`+YMBbf2&K4|xzIKd{E5M^)$e>Q+1QU$DM=UXz9{|8`%Dj7B#5
z&mWHLyi}#~TKyh-y-1$2u7=Tu(jPM~guc5xZd**z(3=W0z*dzsz#&z)R=%Or2_sEY
zUYA+W-m{V`cs%dz?{V8y{^MOeJG}mv%vgCn|KsBRAN-I#JD!oM@G#BSH8Z--WW73Y
z=B4Hx(#UVdZ%hIp19`QsZhE0nNmUX0z&I0&1+7{*Bv9nj#5s5C_NBhq<U(tb0fuS5
z!g@2U-=I}14UDAaz@mL=iK<y>QrJo&h%%)E(1osO9nQ`zybH3ke#?q*EvIV)Bjgmh
z$@Ml}^z_=$E*lBOiMCI5k{UYgZ{eyoznX~?3%}#W1ZG1BT#S-;p^M09t>a(Aej9R?
zNGfEpi<(}U!}wI1b#syEaIQ`<p5fUKg}n<>T)qp(z~K*Hk(Kkc4}4E$MfUstf%clL
zOD>ESMBdxwt9oPGCN6%o51DCLwX$G`G!_!lv%I|2&i}4an#azahOp~QXwE{V5|<3%
z+8P)HtZd8ZjIA6v3@OzBQ}ygItpV?UpM3~A(mQn>daG%ApLz90M~}g0%}lB=)9nMQ
z%~jTvZc>&LDP+b7mU;D%>8@HP4FG<A5jtCc(ueC<ijf#Igs97B+vv*m_{W+$R0iS#
zl5XP*XhJ)ob_twzxwF6|9K7#<X6TKXtqZJB7xfLV#pyz|b!+b)>FE4Kxl{fl47q!4
zsDn5yv<2j)iJ(Sxe-X}<C<r+%7~vOkz6|n=V7s0)bFpaZ3epENIgBAzAbqvFOgw>H
zFl+t&gVSY<Dt<!snl@F#*-vO(!hJ${f6=+ZX~U*XAK)MGUZXe184do;O8A@qe9}wk
zIad=lj-pN4$2#6~{MQPmOiC^b!~W=1nXydc_$ZG&p0YR5cZ8bUTyd}=S&|co(&YmS
z)R1#ThMhir&})H63@LxW#&09wfJ=9WFDrZkVR3OqZUF&{4mnagsZ+wn(%cltoSSHf
zEf2cYJ6MWb<Qx@IA2qh(Lgu0<P!g*1Xt$Nq&OBN)@YrNO7aiLqBc&b{PZJp~DF4#n
z^@ED$2dhKXh38t2{4%XDoR3y9;LoydemFW{*)cH6k{@3$f5^${rns{jKc_d7hnk=H
z>s_pxFtyH^2WbMDLEqvG9?8_{n(enT9Vt+0$1v>B$9*a$1IljuSMqUW$odOPJDJV7
zZJkKtl)DtSaiEG-73fbsGj{h!h@)b#t3`j~3x8>x@-0V#;*JH@#L=Jnt&w5`o#rkF
z(#Kc5Eic0FV%I*+dQ*M`j6+=!%>m6~89S&*zjg<>-VQJ%2wK1j)mk*Q78$Aec}r6g
z(jD|1_zQO2#-#NI1};iFr_%2cOVi;-!tElr+Niy5IUtaQLZ+G7gf=V`O;ebuSO{^-
z$;4J_y@yT%4(8F?R_9cdZm@jz)b&tVOhJEqsmSykOFXUhA+<nKTer?45QS0$<t{{l
z1=b{8sxG*S*k-&Q)292+PPb{QiRpt_LIHC~BkmscyKqFsNt4JI%lZ3z*dDt@9V#hw
zjTbU_cg)0zqgWgiOui<WhyjEe&x5uwTvIWDZ0kP%UiKdD=j*$f1a&UaOP=Y4FezUN
zt(cv}dkHd(|56MO*m~S6u^g?BWgSO;&CBN?$kbNgiNvgr`xVIq8C3;|s!5<0<t^fW
z7b$PSvgO`}0$1{kJkt+p^@fcxms**G>*1PDuc_-qVs*tPQI#qu;rXxC9|#aMxrc)p
zHU5Hw@I{9|tK`B#QjFdAXy-8$F#lUxLr!f{ki)@ZL_wcM63w(-k&nkXTP)1~8^PNt
zWdA+27$Br544JJMQ~bvd9N+TRhLsyOY><&VU=nzvTw7Y<wyRKiJcc*{l~kxk+itLp
z;=i0Z{-h0+0ATs0wJ&E6`M=}$HY(A>q()pjsYS%Giuq}*Sc~j~a$aVSCU2MKC@jzz
zine}BKfswHPj9?pLB&i&N)P+A|Jjo<rfa^v{N;8JC3|yF*Msl0AHv<JQk<7&R8j_t
zA?mzsSCb;sS2b8a>ztW?O@(iuA8St*`WK_xKzh}~w~rjmNWL3A4u5tEB^+pSC_!5q
z6*jMVKxeE<q!$Bt)vci4I)VwlB1dw{1sat(9v3bwR~bF(XkTq@K?;zRoH02J7m#&~
zcxb`3@kL7*1A~>#jkV)^e-NuyEpzFuZh$><TH7zUDUhNTy(!v%xaLSzlXFmDrTNEt
z*uq--^r#t(NG1r>$j{Fg@+j>~xU~<yfM(R-&>f2AHX|cLAT=3yAf=Zm+Q#)6g;yD*
z6oL7+*KNB8ACK0s_VBG^s;zm{-ANK_DkjhkBw`K68$u#42($l>9V4raBZq?m01xM~
zp?ALPr_{?PQ3do2v2XQ5NP@vkEQi|fo+-NogeC93$0?S`Er#9_ZE!!}4yUIU!+l!I
zq-73d>z4~D&Y}I@huXD3eS!6_!opENPx327q@6xR75<=%;)}+?-wp#%aGIxuE!5Q1
z1S*y(@UkK3Hhp*(>-O^2#(uJoNlxY>O1)G_>rG&3jyh}gQH&w?>^-{;nvaB^i#81J
zc5#Hks&kx#w6o*bk<Yl=p}DgmUvdtMrw&}g*XliZ3*oVG03#a{OTG9nIKrx6u|mV#
z>QL1a#17fKoaghgw!3syh+b$we-B+(35eMp*RgE?gH1|2=@s5Se6U;Fp&Jy)3rJ26
z_cPhqHD`ER&H?0EUcu><X$nIl#=g73(|)LTT-wV3%}0iA)FyTEsTl4&5;8Qq;GTeN
z$cSYIlp<Uw;qG(pZGpzK`JPeyhwtCX!fU4`gxNVp=125tbzbsg*3G<cFT;=m4tD9S
z)qL5~r5pxo()ncNnl0bw*$JOo3|p!wVl_xGYSsD(*iEILrmcP(nau#($9gUBgDkA9
zI`^?9xXtYk!|*`n=g_pGg1X1X>VsJ|%eoeED513F<IRTstJG7vn+Gb2@t-9NG2w%5
zg*W{!+$flU6(5C;6;;_D9?zqBLr@(86DI~w?S0*e_Q0Wc8BJta8tE^FuO+&2aa_Vb
z5<6c`l^EEku-szUo1U#Tvs^I}dvJ9ktDrHvb{ja{DGuFB){(O(&DlppYbBIy>8h}4
z<5-V<nR&x*2|Q(|!j8HJ3@=zH>LjIj)qy<nOVJ3+fa5|H@0}NrXo|X2tN|&3yymw2
zXHKP-V@__W^hAFCTp<EgLi=XwNb=EI2sO<^XSD|82WGzX^HJ~UIfQn&q9bf<x(X;m
zV`9;!H?*3Ci!9_x@J4amy@-YvnVdSOjJ$gM_+97AWcC}KBksi<!kNfH&yGL31zs!$
zbJc}{j)nLUpNL~YhKH5$W5=qvXbc?i)Go>P7Ga&Tb=RF#*15tUK0PA{$Rb(iSs5Z3
z@C_xKZff&QZgYO7oTT8}u_Upa9`|{OVJqGlVkLDN0sf_mfU|z5(Hjb1pD!t4gu5LB
zd_yXE$$Dw?@7C4}7gk4tCUaIu3gSSzu(jNdl(hS@ZdO5N{?9JV9HQJ7&L*RBEG;eP
z;5T$BZHoCCY~nPOKAgQ8U#qC9)_jtbxl8UIz$)G>v21^l0E=NS>-gHLteux)Pmh;V
zOkx2w@yfE2wIX7p(O0Q>yCIq~kc{HYigGs|xJ<OBnHfX(7K66EkUdEYMx{g{%ac^0
z*mW0o``f+?XfAF=$F%CYM%^GsT`TFROSQ4O)=sdY&=EMFoz0Ox0!KEN#|^z`5>#t4
zP>_kT2#!)=Ov#`b0bf`P)^0|BL`Enc<Sg2iOxhI_e#)Usjg4p98Tm|UO9B;3!T>?o
z`){4WqeVzZ)*y%{!wo_&#risU((NIx__5%ar@7hCk+96cBO?NZae;LIQYiTngM5Y#
z>fCwMx)p)ijp<nNEBD}CU?5Oh$K(Ctbj7K6k15B@P46nEjZDJUhiP^%Tz>xL^r%~f
z4aU8Axw~UwgX!O@HU4IK`lZ>o$f-*M>-QM5F-m_;mkT%Bd33Fwo4IylWN%K}!TnY*
zd@%F`vYU^MnyULIc+i;?PdyGbZ1@Hz-#WJbz`|+;5jb1jNlI$4fB$}4-3tr7e)IQL
zmYnFTcXP0UAE9FK$HQb4E#WE0@@zki<5Yo^YjSR5|C!Hc-`e22{^qAFS&BxU^+j#X
zt?PhAQ=FIC>W>dpDs!WN^DP`T$sxoX4<pm?Bv?STl=y*<DO02z!4z^<?yWDHs)}*I
z_PyUpNf~<C2~=55v&fAiDZ=;Z>hU<4>_l=Y>cWig_yEPx5E4ghG$mi3oK9O$4#))B
z^3lP;_t=!umlkFQURa68-EuP$63~b|V^N5jn>DGM{=B5ki@Oe--fOQyWNxpgH!Ej?
zBg^tUHk@Re5aIVsbb1o_{PD5gd6&Pa48n{~x)7wRqK+P3dhEy%<(czjoyeM*qc4{{
zBW228?!0N1A%~u{oJe2n_dbliA=#+us8#FM-6LxCA%<neW-X({5~_`8@o~IfXjr>j
z2Z|-Ja)gfWX(234)UKP!ASY-ILj+6c9!<2_*JV;0#RVCYOiqPDmgDfcjqaKnl`{-?
z*M{t;_v7KM{aRCdNksQ(($nnh6#LY3Q-bjbeWWuXev(h9bNL@NwplE?to-Gv4ogZ)
zOTWJP@y`y6te|tHDEfbYR^d$4OY`fC9!v_WpNtfLkLR2HZ`r}hODFvffOKM&D#tiS
z!IEV-ikW8kL(83#1Z8S3V1R8wubu>P7w6nx=_t2d0Sc{sJ5dS<tObYV#nDBKpg1jK
z&XAJlS69>Dmf)v0hi0F|eJP<iq#)({##n!h_o}5xa~i!(*Bk=yed?k$dBMr}M=n{A
zuUp58(nB~sm_cynrhGAU{*z~2Tv|#M+rskl!rJED8%v7JJ{K*>#^x%a9i9+i`WLsV
zgmlc3&P~yhxsbH@?vP{1ble$H4$X<aGQc5>DevO`3Z3GiXU>`^9S~NNPZ?EasinB2
zstkVO*9zm<X4(sCIMLp<evrz3R&M?1Biuvl3Thl-GZ5rP?bjR$KI%fxs(bqMD>j>v
z<<p?9BLoDJdhtUL_0@xeUG}ED-YC(Ar^e#hOwJ|l4;LdM?!y(o2#Sl3j`XbTKWkSs
z<GV?2#bG096Tjw)%1dbnca2$N29}o1R>4nyf9|5^;D2qNyXi04zIzg~?(1&6okf#-
z!ViW>{Y(u1r(GTuaEoF^#bD>^(!sjAAQG0j1D-??o*i=CBX@no#1}S_SU*F%5l4fS
zga%1s8T51;fuoGJ5yR;vDc=qzdzc*_vVJ*{yE=PyG~Zu7gATkJy}J=y8e+7qsC;az
z9cbBq@m8vX5Bz0(+cAuvmWL&o3LPBL(Gdf7Mq`k`ViTW=&J&sF_TXs0E@F``By_m&
zHZ;sAsKpdq+(e-o<5hApX@35}gOkqCI#M;D?C#AxfPIit#=v><r<V@(yN5zWHd15;
z;j`DQxuWdj@;fb~`cgGO;pTUytRQvvyafw7LVUgi+&Or|&r@P0g3<pOu3gmti<DCz
zLy^)e$a3m6XHGnc#dIShtQk52JiLd~&Ar*jy^e>Bb~iE66x7<9&o^$)#wBfT*muAH
z1VJ|*z(&qWw~O|oARo74#YJTwlu+k0T9wT4*~f~Dy;{R8Q`#wGjfR_VV4dzEnHWW<
z#K_-lhp~SHY>t931pr9ayc}E{hjr+AWlVvOhlhu|JFJ&m19J(_BaJ@7zEL$GuCCd$
zk%ZkAAOxr8oumXr(s2MVS-e1ZBiCh3ylzmhUOT?IpG9!o!-r$>OI;gvUC)b0MPcdw
z@yYS2Y~HFa6jSa20Rferj@Q=@b}R7n#EI>9<gy6B2V<uv5%z4w{6kHV<|`jbiFK~^
zB1vtekhL9J<CEa_^2XX=aT&mnZDfc=EhUW6?%0<!hX@W|OMCK{kwJt&AM9pKxbHM*
z%rSyE19L_j3lI1*N%3F5e$`SwvV5#^IA@ax^xaHOxu7y3GVO3ENS>NR27Mi$nx}n{
z$ekfD9O$=bA}_5E&S}eOjdINK#cId4UhXG<RcSk%uPz=?Av#{h)p!NyjUNG59AKZu
zPa(P524MpEf;fmNp*N=uuuZBebSom7*Dc!gcW`v}%2*Gd><+P9pg6ojHAL}iuO=Z|
zAS>#;eZ*=G<L0cyYKq`mlDZQV$ZTuzslYMy**U341J{ssP^=FU5EHkK)0mmaxV?2h
zUhm-LQo_VV&k;1YzN0(V6zHCIYXu}ZKNZdBnKAW;ZulwViI2`~t?3H4a~=t$?uddN
zdW(E|DF3L_n=hf&0wysWC9L+%?nc$GGpSj^JJoJ%Qj+%N;1LV<2_OmbGK%fnkJs4V
zQ`*T#%0H$}3a5prj<3uG<Uh?lRQ_8y-DB5({xIG<t?@sYyJwL3xyKh&>nHVFJ2btQ
zW1Xa%-&E3jJJ!)&l~$)v5s#b7f8y`xS=GHA5Ej-p*yf)<DcM@=IJ#oW(wj!1K^;4k
zHjLU<S6CR8|CpA%w4A=ka$2kZz8cSMegFNMlBxXvKm3319t0T~Y2R${f3BYVl^50i
zV`=@5b@spf;ZRl;7ul!A{)X(rq?v+O_?iWQY9C(NxKu${yw%wM4ks3?J%3L^NQlG3
zq>>X0?herm-hnEeU*i{8<@rY+`snMMcb~Fws%vGoppl<g#RWd#7G6S>nEd!TqgMmX
zyE|YOZy(X6QzwcB6#_p6_JHhSk{FQl>c)q9*gkV=`<~>7efzk}kT&)GXD__f!CjBe
z?|`{bquQF^L-yGREKo!=T9I&A`Nex`UK1+siciMqx(c+p`q1o($;T*7t9XB>$@%i?
zvX;smpbYy*;sPW`>b$s}G0Dlvr#_J5w3Zp2SjT9DI_R>scd8ipRk%OiFuI=Ocsxvt
z-zsm(rGCcN)z7Y~5Mw-`**Te{ReofCj$^)Z$|__ONrf2KQ}SgmO|4o~vq=wk61;c7
z!@3)FN|QsT7tNnKa6~J@)8*iwX}DuH1~kyV*`i`2xnRP#?#~u|jDFSTz`=u}oPB&@
z)S-PqacI`Qx;}cB+gEn9-}$VAk5vNIUz4tnU39#2$n0pRdi5iR8%{{L9Z|(alv^NZ
zt3E%iO^+TS`e{HZ@%l=M>h>geZE4{$e12*&y8`^xAZ%Woz`~-`>h4t&MCkdZUYtH#
z0s^|AN0PxMBvUdZKyl$)4EHyiix65Q7Utc(Qf7?R+(Iz+rCz&u_3C!~7r112mYX_I
zuzj{fVk@3paMr59EF-C?hRCS;ljx3`NAw(Z@dBs~yY|SjW13tsXGLqetU(`J()}Py
zV?20h#0k=ALw~b(_uD#~Cned}#UzSFDe_(peKjc0Zb&u&$`_lo&W!3svwIsYSI)Y6
zjT&v`Y!GKllOS{5?>#~a#y?^4O6v^ed<-+GIHhGO<+<OF{8KD+GzJ$@wU6N2k$fLv
zL=|bE3c~-8OG&Sy#B2dBUkuVo`bB)vQJYn(QJb97c`|ztG*?zL9u*T8rPIP&d$swh
zTa(I8WN#WR+)!>QhyQ)<2!n+bZo&fA7!J^P@zhE(FT}&CHkg?tgyi_bDuj0-lON=i
zYTthNA|LoUqSpb)G=DWh@<(*JAr2UYZDu<{$B~8^_0wsR(4dqq=r_TLO<x_CC-L#2
zD~*W{NC9;JBrood*4Le%vD2-_{uvRorCMLmJMUYp`jNmCIME3HM_nLID5)eS&f*4;
z4OgKFF`VNLj>~2e+pg*T-XEO+hnsYYY1nmz*}&*n`aYc3i4Qk%f#T?czxMco4P*>F
z0GVS#h3$&PVb(BnI*BKSZEC_|A5B`v{PgO)MWY^;^?;vKx(Bez_?1E6=Tab$Gc`r6
zzhc!YL)Kk*dA)TctlV&B!sWau1>CVp2OsgZ2!{OLPxvbVGXo0RF!N;Yvs1$rmj<1T
z4gAfzrB29fg;aNk467tgu^_#^^h*bb<DE|EIkph@FK&o@LyA^EN)1t^T9m^C<JWyj
z;dMV(LdZtd_f2|B#{p`aCZXlPzI~^{Lri=P-!rm+FVqwopQsg63K_&|8f5<d$BzqV
zycv0TX;DHiG79ZA*(TF&Z3vTTLwTPQdMh1ubj+7HE2NF5&JbhA)84J>B6*wmP|>??
zL`&b6EK1o&mZ|l}qp?=!?q0?eo@b19G|*o6&GM*@?SV$UW>_j?2uEbAAZh!7f5Mp6
zhw%mq96J3q{7<m<Ybrrgf)5GxQEZQLp82{4CNKV#mv;bgz}EXT;wRElX=_+>CnO`}
zJr0uGtF!~_$c-%I@){{-us<L|c){s0##sWcgArk-XT{&&fKH--BPulNW@cnJn8IJm
z0C4+0#chT<6q_hxq~V4Mm@dOK00$3K*|jhqU4K6hd<S!kxtc#eZssnKPp6S7pAINc
zOG#se-+0A(i;W#g<{l9dA)5^wi{3Oi577m+0%RpI?rEHH<!a**pkEIFSF5^+!3j1~
z{=~*S(8;dp&tF9mRQ<^X{mg>S@TuBBCP%?xMTE+A)6g%BOkDsztZHy?PwU{z%eWM)
zbd6yd0ZtS1BRg;b)m%`gP30lURjxvpJ?)Ckegdoqj%C!m1(B%_F}tM;b%w@gUUWs-
zq{44$9Y}BwP;jcW!({eZd^DV6IwuXl5#`~55m-{d^?*e_536lLrg`4tR}(20FyDGu
zSO0H2M1AAJDU9>$D`x1y{-9x4Kt`GbtQr;`-kA8o)7azg4&Yo<={D`s8TOJ1cpJdz
z`292xJeW_R?Bfn~%bJD!4&r=A_|{2u*^g53p?;qRAH1Wo0OZxG0Lo%RdgHdQuZ6M2
z9+HiJ<Q6yBziZ>dkw(V`+VxV-!06S+GNd`;CSvIhRtJ&J;$c&CyWpAgmMqaCwck&q
zc)*F=k>H_faCV~ixsQB<AY~sr5^nC^$KFGx5{<ebBWb9-|NOZN9i#mKsiIxA<^*Nu
z^kSt~vJg)2XSI|dYDS2xlQB5&fs*1-ZdB<+McxYh*2Db)ZBDPE&o?;?&V_#m|JEk{
z`0pXbUB&o7*W5&qGBG+dwnwj?7scH^w7%lGFU7E`fpnfpCJqs@x~5iO8F&cLL~A4P
zdQasHWsDi3aP?ZX4$^0cr%+)Pj>T-R)8N6=9!4|$sgr(XBJH2qMP1qH&M7xic|gGT
zx<W3YuH<Ay-gI{d?48+VOrUyasrklgVfII$eF$iMWseA~f8iWU@FhZlx3(?9QLLeo
zL<cscY;Wr!ofn5&DYQ#Qc;=I;bW_>Ert36}0~UjbxTjAi2Dd`k`S4KB;});Dqt#qQ
zL;v}eY;@&;i%J#GE96Wym1!5}Y_`atJn}I1H=aBPAJgGtM#vH91(a>O3^3eB*~ekt
z&GuN`C_$;R*zxC;eE@G-w%cD<c{2Jk*Eg}<?Pt1+w>a1Z>~I$nq>??O>4+~!8$5{w
z;5uPp51bFMG7EI`)S2c<78=9WPcVt60M$CuZk}bS?&?pwlzlMVzJBz^iH!zZJY4h1
zk)5|!Rca5OwQy*ck_@?-ss?ta<mj=jV+5c0OYOg%J9Zos<?x5n{UJRm;vBq)dDhmQ
zh%&}MM}MFfM#=5Zy*EDAdEfi2EdRz`C+=q5e|&s&GTGIYB$i<}&RtIuww({6RSX=9
zHh?**8o<E{Jwv8vEdHV}U0=kvY}l{^C(e5a?OwC6FeZ#Afy0E%**oUm(97)Q21*jG
zlaRd7nKwH_M>qh=*eQ}TfT-eWaC8))h(rWB`3-K{m)^M5y0~fmP7gA%v<q1?!Qh8m
zTPKqwnr*&SOESt%Te8oUeXLy%_qi$t`<*joK6Xm{lI&r^xKQ@#mlu!FDtn{M!HP!N
zM_l{z`VnhZU1_dkv+uxJb;Ht<lEsl7MwXZ5ZY831v&Gcv0IT;A>v_#npUMn8>Bm70
zmFoF*tE6ZYC7<tViI!G8REab%9-dCCp#BxtEAbcwA&@jRKAN1MJ>(kUk<sIre105}
zTscKOYn1~qGm8As?6n<B5~C6er>hYWZh)+GaQ*sS`91TPdKx6fQ&T>(Nrkg%5}eFs
znB+lyYR&DkN%Y%hF_^4y1evGVutVF)n8CJWY!n!Zdkqj~^4`jPK=GV=O%X}3a+?hu
z9zX5Inr_OenK#!d8qsRU-p!+Gz-Q`%TJlOTTKXs>03@%(=e!EnTJ`C1kk_I5%c^aN
zH16@!-sV#}oj)N0M#jcHuUwT+*7ktzx3f_<zqLb0ygK~JLA?98aZ9|O6Bvc!L$5n%
zR>Kj^R^58MVWL?>ELB6JN{qFhV2z+44Gz&c{QT;&=^;L(R^oD8KZZM}qL4<pdw7f>
zj~17vkk#wgt=o=+m4Jyz@T(L@4AZg8Xoahd?Kx4(N>T=rY2!bY<d}eyJnc;~-33Br
zqOleisC@rzNAd*4BILv(at>%g>SRkmPag<qR<Yhwx;Hke$}LH5G4Ysb!Z*-s<?oR3
zRvqZfyrlCbt`lO_&CD_=OOou19w0=U^ma4mGjB<u<8w83lTClsc*s`l?UGtFV)l)w
z#^0j<d}_&&9CC%)0kzWiS6*KRw}^kX$<yN4yC+b_o;7d5{G;{CZ8$<}3TvsWC)$rz
zE%E8eLqrr5gSGhE8<lY}G5SSr!hyarBc7052gZ%Xr=#dK`w_mx@<R!&;zDOgc(10I
z*Z}Oe=X3-kjvqP-d-1A2OJeoI8*k36bK=%lmVx2|Avb4fu!--MiOTp7-~XiO=(O#W
zTe>3Y#Sq9xaYK*k>XS}OXNv4Pr0S%MPai(G>R0^jm;Ax@^{tH!$*PQ4&f-Ub&N1XW
zK`p+*4m6Z6x{ap8JeusrCG-z5TBeE6!p7Ufx2PQ*B8X+i>s$9v?KC^eK{qvyL8F|h
zaqLOYE4ys8lxIv#6X!^o%gK?slVlgOrb=fSz`_j2q*oRi*Y#VJyb5XmU7Ure-D1ei
zrNW}k=!^G<s5!{m556zYv3~GN`|IGb4{yfkzs+<x+dk$1l!E}vLZXB6che6+kMRvN
z>+zKgSUP-8DXDGDw9yR;pO##G1g$*N@WoT4<|0Pge;yB?Labmdtc-JS5pU*`kCR|u
z%h=fQ;DG~c;_V}TG%t9(_|v!-MXe5LO`Z6s;+R2T=eOm(CK!C-qpfntF;X5L$4W?f
zrhalsQbG9PD_1;rd7j*quNO3P1k5+-;!ptTrfYvmtAw5WuKZb`VgOgAxTNI0m{ejk
z2jUz?{};cziRv|(10YlNMEpf7NAN>wB7*Z-sP;Uz_YgPmZ?1F2xL+HN&1za=V?W1M
z?{NlmIDFnam2A<l^$#kN+xPL~nSpyhRhcnap93`VPtvP~nw>jW=VT=i>qrBS2XqG(
zJ=CI?k(&mRYMuAWWOyiubu7zC>M3zK<y^nECZ_?vaQ;P?$RdNx@cw7^1icyPn6=1g
zx{3n2Dg%$juZ{IC#9DC{5cVL$@kZ@w74&>S5^*I`zngcLIhLiWvwFV}HAxSvD>W!8
zjC&`DZxPS7>-4MNHhr~oztX06zmo@FoN4s7zh~5dB{cmNF<f?WPKYE<!Y^LD)7n4l
zy75>r@3)^nyETVkFDY)_uLPl*1mck#92VtQqJER9dQcm5K%Xpb_Dl}*`daa1U}jh6
zH5*g!+b`^&b>_u@j3<Z9rYF9RSifmQK$DLBIx+%m-G0-gO?xxC^h_m)H-Y_S^}XBa
zlRYb+obmBVO_Zv$ic}Ho=m46k<5I-`nlEZLg&e+*bGd_XCrNLn6h42Zu@+ar#GBpc
zS=s1aBt8Z8Tl0eoDX05yN1gki%F52By-zNl^<?xC<F+mo3X_iQxlY#|*hY<v`3`=*
ze&OHWwCUM`qxl|BB#f>BPo&!j?f$V$J3d_Zx>`=qhxDQeKI?O5j!JpXXfuI*VQf&A
z)uuODg>)$v%;qOnJdjb6-)cW!?QAx03{OEV5o1453kbUjV`bsviiBMJaUeY!ueZBR
z^5b8+;pndu4mBy@Qq|Aes&+ZnZr`WZGiS`uyB2v?ggafQoNHuMRc0N~6@{?bNKET^
z_(2n0Pffyuhipfo@&l2c8t^&eW~9~RgQf}RKkYS!H8`Qa!D8k(OLKQ~#`j~J&OAJv
z$YA-C@B2<J8F4>7f7p_e1s{jmr>3XEVl!Q|i6%JpgI;!Bn}1NHcY0~9Ty95iz6BC6
zk9LFN!h&9$rCzh9*{3+ZPE)EMv8J@Xx$sCL)nCy1BPRGwZPH&zcR<C!Zm%0`n=3y1
zRCEgTR=joL5;jIE;+#l?l^rP^-MlsokehJ-#fjO3j<r{9KegU8l?VOy{rfEh98FLK
zPS!0{UMYVqo3E^!G?|)XH0;Lr<PNC0lU{Y!v|2rS;hJHYpKxfRt<!4I3I)<^yu<wM
z#tCI(ceMZhb*4ajCcdgv>wTBM9wf9rP`IErlb$~sJG<0j_9VX-dk!~Gco^{W$B9jx
z{QIGmDth1AoOJ^@^flZj9cuqBbL?N`rH=^9tleYg^;Okib5B}5jkpC1SXDu>dExRh
zS_Q&=0(E(ZHN@M@_OZE18e>y5sy#&`+_Y`pd0V;NQx~WH<mPcym)fLKy7CWcQgvnA
zp&sgF)6!dz!CV(+Twn94_PJS+x-S;@r%k?{e5?vlf19hT>!o4jl=uz}<=jVB`gzIs
z()V$-NNK4PJ0y=BT*(TujB>bcQGzvP_@{m2dWT!>h8Pr5H<@Lmsz~G3`vykX4Zxy`
zahO!ORo*9b#&vDfa{d}#FK}J1#hjnxXYHk<>+-{48l_hJydJrQC(VMB$)l?Aw;t44
z;%^yT)OKIrqISd9ng^M8rIE}M+e1$<<S#clmV>?Dt&?PPE~Fw5h1aOo1(<?mpnI@5
zV2b^ZNyZwsV%<!?%Jr#V=i6Tmy1NveJb3m;?$tP?19PxVv$EaZN<!A8xpY$Fud9Y`
z3dX{z4N*0|RUoz7W`wAGapAOul+mx(AY|z9oZqY!N|hLQ0i1qWI_|4u>U|a;2{wRl
zN=>s!`~{#QxWYJhxe}hW%RYkPyNwjtWFh)Z$S|6=Z{PmNpp}});^e3#<bS}~Ot+SY
z1Q60N>EGfGP@Zuk;68s@zh%qdzdbaiPwwLUIc5#UFIFM<csy$}hJ*4XcALDqZe~<Q
z&h^nH@rePn+Y}clKepj3C$#vPTQ+7dyik&3BRjjXm61^UcEGH4gF1FJQrtPj+4jTj
zPI6N3;0I+N;N+L3`Wy4kknrrV3OOX@D<_~e#CLX82W3IhnX$$hAVDhan3-q%isCCS
zCT6%Yh9Y7C6{I%D2n><^G&5@`&9xV}ABct?qiTTD+1f9DX@!iK>=Z)Dr)mJ;vw(pq
ztAK9n*Uxzwc3P<id>}v!1>0C6nE)|xXR>HN{jd6@0rnv*hpyz~Qm+$FJy}O%P)psV
zZArxMrU+64ciwMG1&!qC3OmK4S@Y%(3DSDmZT^88EY81FO9jd>qa}V9!zz-cj?gF$
zz%?@_z3?YKs~Tu_=<s|>f+tA+Xk`o(sRb~Jqz&1eW@V$ER_4kBLbvCw?N=I)NEoK6
zX{d}rZIGYRFXa1ID$n=X*<ImrCgJ9CRL}8-RGbJ~I{#Cjbp~AwnstyBI%ZF+!A2n#
zk4T=fps`(|D(_}-{zxTP{Q>%&q_zKIvS-$X^iMC(ttjuEylSKoiXAED(Q0(;+I0+)
zuwKxl-hcfX93h?{f^5Lgd_bbL0Pi^M1rO@2T(hPT>313orUO%-_o6kLF}-g6{oH)P
zU6g$QR^x1J0;S<Cm=1tKBM|tsXy<76Gv$vLUu_IhH}?6hRQ8#4i?vzhzX}Sx*zjX%
zkyLdNK3+Hd@XsAPoSatQG((M6%mJ7rkBTy#dt;F<fEsoB)V#7^_uESO%NZm|-?Xbs
zj|wq}1fv5*(IqI*-m6pB&t#-GUEaO=<<E@u>BB3}N*q(UlPc{XW;KPtca$2O%`_T4
zg}<lIkt2a0L#KwGRQ7=u#<P+HR7+w;Th_E%8XD$#dd}rDWH8R-$KxTo`u?rcKfet#
zEXHr_)p^p%9w7Ugn(y&U$t@2ULC~q@BK|xAHBJLdcKH2dVElDtSTi0Tx4N0Ul-$Al
zz*QcTN;?jkr1i%uy@I;#*4wD8pmtI<P#*B~3qCc+TlT<b5>Wf!P)w$=5AHsBGJyk9
z)~Oh{6wK}7SMgz%(MJ+YGnCI^SFawTj2*kJX5$4QqC!DXX&1YFkyF3=|HTEUJvpyM
z3xyu<DbyJr;GPa}7FID_xqkhz6DJ;$9mTVWi;p*^S>KVoDv8WK<kWUDts!T?I0dF-
zO`0LE9n6Wk$}uVLT{LP3Spo|l9h$(X4S(7N@%_JJNGE?lvDKaqN8+o{IWb#Q4Pblj
zpO<DEsTSm@6A-Y62YQ%Iz5@sm++ZB21jVsd5^fzT?N}7XQ-@9N2T<P)hr2;ThIkNp
zI1a`!n!_6G%<7NEGOg_=N=sE;wqFT_M`+BKj(h%mGNcKt<E?5o>e@~?9L#9YnR$>(
za}4LBP!s6w_>bAIc?oy?KV?Ok)m*SZ(42X{^hLtM`*oDXKQ@wTbmF1~=*F~~f^kep
zyS+jqJwq)hEDd4@7)fW^T%I5gwpsmjl&*Ot(=aa!6aW&K@h+pr;2)d;V_L-=&kzSt
z9_HQqgD%K#$|fdEvgHn(IMIV3CN+*AsSZVN)@vRpS*>}XZ^BY6b7iOkU9_(93&aB!
z?_v1eGrQzM>NvwjMZKd5Mg~WtBmA2#cB9l3Xwm3uy2Tf;V?9B|9)aCmi5G{AHQ!%N
zAv1~tlj7PIFjF-T1&we@;33*(`M?vA+Z-MqE_9Kgo9&CT*$QUYS6K_of;vYwn|pR$
zrE}-b@o+PGjNI9%KBb5MU!M8c88c2;%&Pv6gTA(#@kj5zLuOx-i7{04F)YzUaGA#!
z>NJ<B>*O4_NtC@d2Dxh=L`OBzM{hS=_ce7J2b>S!9!Zak=!pkdIKNU^OFOFQ<+O6~
z@t_Xzg-ye3n7tPE3^CMp#*oTO+jn8U9+{dU3&<;ZM}Of7!>+o646a(uMi0)ss(cUr
z<LDIipjWqyj0`crL?*ulo~SQDz<u}bIJVtN9u}DR_F)h7CR02j!s*J!TfyvDspPnu
zsBd~ng2Bld^YSHa%(4HrqVv;=HLJLAXJn$dc!&@mq;8?1BSWaTqt9saJiegm-;N?$
zU?(YGAh6B_C%<^*iV@1kzI4<ew+y)){14Rv;_kF}tz7>71S6wqFPjej?{0VE;Vu73
zmiTu1iAu4~fWrBW|LFtsTb}-Dl>~?X@GAU&fz#l)U*$G<O-#zY>fuY0<G-D&WRoO1
zKl?sCLXdz)y)87*39m@-Zb^0&LkW^JbK&^$8c~ILTS9i%7XH-0grI|662Ba`d;2SS
z>VtRhx0rPrH!trD-19Nd9^S;MQEb<LTwfjgFtL8z&L(liD?bP`aZT#}@=~L!b4knt
zN7df2(f$s9U%su@e1nQ|NaSPJU;l7_w=17<&LYTf{mrw(J{1*?s?)S-BQ2{&!#3sL
z7SK5_9Gioy0|o?@|NdPtVg0b2XqjzuB=7koNoZGPJ>Jo(S1-xdx$f=c7Hhn^Y0AK*
zrOTE=VqQ_%Nf&*QSdZ|^6;Jk8nbLi#RBw}e^wS*Ls=sQL{aTAAeq6VD1Y+omy?XSR
zS@5Xqmq}|+ottv*ExYE0@#G+L0F^Cn0bK}Xpt*Gke$M$fm})O8kv2qW_Wgfgg!5UT
z12rzpY)P)f_$I?Q_67AEKi5GI|FbkBvN#8`H{bcAT^~XJ>Wu<t-}^#GT--n&W3Vfk
zaBRRf1oDO7GIo=IzQ6nMAgSU01M6X^HydhGkE}BBA|yMyoI^A_h!c9NL6vHLisFf>
zf1W!}|2)zAB8<k5kG;m(PP$q2Ek=ZLJe+Uo(QP>es(WoLdC`r2Uf;OyAWL^Y;zjrM
zt4Et*n>1;MURzN<R1DV%S@X-b*A|7fa>^;v{sF3oj8(ISI`5&dN3WNBs>C$qCyfB}
zP>b3`@DW1(gMlTZ<Jhmwv^Vsj|KGcZ<;B}?S2ca?vo~^6VfLHOnl|ZYYjcytG@Yso
zxK4V`-b2>j2@jsycg|ab5Xa7wj&0?O-vMy))?Ghpm01Jso6yvGs;A}$Ow1WvR`+a!
z`hS{L)8D;2xnx>H1?;oui4)!`xAyqyU2|@W8-Ni=3bFEMX_O<Br$5c6R6ZBYk~#9s
zXFHc|oBcUqH0PFAJ{>tftK@UMD#riO;o3DXZnJyo0#JFHqej9=%Jyx(U>os;<ZP#?
zVBDSmG~k{N2Rl|2>Jgh)XVfR27_=F$Jdr<0<N2N8Hqp5TR#Ef3rMZO~Y+h_h!4chM
zEs!M~96ohxxfU@p@tSFeM%`yLgY_ni0GHHdr;`U}&f%Jx4Rjhx#(3zHHo&AQQT*1<
zz4vJ$yyGEx-(1%%J@?Jcq8q6`!CG>IscE~XOF;be<e0Bv+)hXnA2#hx4z-*8*hjzZ
z&2_n>NCaySS{+|=d)=1XQ2<`4t3I5B??5@v&a>y_>Zm3W95wn|v&s0?bk{PBf(~@r
z-+Q)Rt6}zyGO-UYe&^QK^E-BqBj{$EA}(9{`pVdi3uP`~RrB-i*e1RGxxd4MN+34!
zik5kx3*EN4Z%PV`%K@iPFFmR)H2<eFZzlaMjk)(q_sQoxJs*b}D>yy-&|B7Mt=mYj
z!Tjd^R{XuIj-9)z=oI1`Yw4}^s;qL|lR5&0P@4zNym{5)XI%vdTs8OaA3>{M-`q6t
ziS<1e223C+B3tMz4IG$2k-zeDA^Q6Csp{D(E-2kCLZ`XbV#Xnl==Rrk354`#0)^kA
zsAL6HS9<*U9-))E_tG9_LR2+Bw-p9y*Ujp0*QVQ4ddxk}d0yRX@W!N$Ck8pSf>z^V
z?4Ok8#0&%`SDv=nuW1#dugJ$MZ%e|L<PLkv`72a~{BOw~G10v}-A|E(Wv51jTHM&Z
zyLRP7%t4Nyb`8&hAYBEN0p~XNjG@PDTbC4m4JdewsMl|HMY*`n5k0-~Ng4#v;B%kF
z`wZK(?!@3tO*R$&cyc8p)hDi%3vW>4!ek)+F1l7rj=RtKabj?wm4D5l!-l@a0^@H+
zMs`ZL*7D37{$2xl7i^VG@^ps`j!C_*_U7i=SV$yZ4&9!p9D%Y!)qk2RC9<nds#{ll
zRk1HE9kZy?x6i0TrVU$0&8v1OvLL4DYVqOTkQnA39=7S&qi>9SBd6)(cuylzrIGTo
zesgste)MvHCQ-=&T6;VC<-S%?B4fKU`uZI@UhewYO;J5cau*|2pbW=K<NW2GJZR-@
z_D9c8c{(GU)7`eVJ7v}>awBEpT<F1gPQH7iBHGq!N~5kO?UFypxKZ5a9(WHKSUsxW
z>jP1(I*i_Z2b+q0kDg3el>TVo_tb|jg+nS*XN%1NNm;kJGM&EN9dn<PzN#pcV18%5
z1IJx|{n{jZ!oCuMUNK`Fly-4Nr2%=<swygPUmbU^xLMb!!l_IKN)WH|zh<^Q=v<;6
zv-Zmz5ZS;s&D>u`Y;RGwrEa5&BIgP<ttCq|nhmOVu=Au*WV?jI&0mtwz5B7Lye7_(
zT{PP0)PRM-*iwDnbzipr#>)_&Jog%P6MnAlEJ4ce{Cr>>0d}R&_xyRcPVY!aW?D9L
zYBsn5X>GSpLl9$<3(xhlI87QNr|{4pKOT7X>vCgVj$il&*Mj5SylBz{M)ZJAAv38G
zbJs+j>OB?W{inw#Ym?5x-R?1=6m^c@l8EKN$ZkoO&D%oT5G|I3@`wvFSH3S^`TDwJ
zOFCksaMhTzY}xMina)06!o*h%=FVK2wN{bm+?BUazo%P9FK-Q}VmD3sGqyfNggG=M
z0w+iGY7zPTc<HiZ?XQ(&1RT2>m=jaNgf2lyGXCVf!xJaaBE0(v?-OrN4jGu|RM!9A
z-j;5(eySw&`rvSBq(MHs-#d{v*M|-0q2bZAr=h5oeunhQr6&LqGs?i<*z)GOo>W$z
z%MLYZB5;&w-K~Gn2(b`Fxj-%Z6QjePkRZx<Sy5V2*!V{DBIw?K)qlyeQ*%dfKpE_K
ze3j&4q(`F|{M75s2$*<w^wP@eF8lZDFCWLQsi8jn>5S`r7JhrF^Zj?>vrHO{(E~&2
z0%8kfLcjI#k--hoC)UWks3JmoQiHHXxw8;cb@k|WzILtH3j3mzk^#E7(k+i{KDIFg
z#iy73&v#@qR+Ro2(n`;v8O6ac*fl<9%jEPs2oH?4lRH!AOofZeiYG@KmX#JfZ_z7<
zwh-ngF$2A-=`t{`5Ag^*qS4wfDHGDiCN>pau}#XY!D0=7?k41#sc}Y|k|9(T=D!d1
znp~Ye*1coN!Q!SJvTTV(Xmv$QQf@~Nzr%@n3Rhs8^b4(}r3B4IJWqsw3_!H~=;ej}
zpUw4OcB!ji&M>RX6Z%Lpz9um7w68_Q^iE)eoyHyRuJ2I;RG?b@%?Zb?qnX5fVd=pB
z{bF6w@Sr%xxvFbU<0`5qecQE5wztLL+%|KT)&o%w7%hEWgGEUf%{KWTT+@bAF#B~-
zL&~Fz%ggMcwx6+$hEDPpPG3-+vSh?cGR^E+(CS6hfQDo3&=ZmFUtIU3h3HpgCM8+9
zX{7yHBj#%I#fqbIqg*{aYB6McuPLbofIKW8m!O~=w-W=bRlvMQy6vmCly<LUN#V%o
z^0MbwW?OA9kNSRiwo{DzyI4~?kXLNd8sKU#R)Y%@j%`Eo`L{6LuB(q>LWm4dqk>4D
zvOVX~8q}#eb}1T|sTD7K*uXr{vhRe_Clv)#|3<CvjgI4IM*CQvVcg^9{lssQ<6CIw
zeJp}Smgf&LN^dFtW<HTfabcgeU;0K@rV@4w$Qqb=${d;-L*VQSzf{W{*f(TS&cV@_
zU%73yDKm7)EAETp(4E|-u_<0{h3BVrwG7;MsjM($dNa)Hsere!>KT_XE%;W%=MWhU
z6!ks1mP%tSO=iETcy(`oYTn33J<PKsqfHWS9qpR;b)Z(ntq^FEIl5CTv#0nQMpY@B
z<1nC<0WCeszYoMB<g9IGZ6vn@sLm6Qd472|d$+XY*#f9Ff}UP2{Qhe0_hbEWk7z5_
zJBsDS742`V{f<3i4Ti)tCfH}Y_nm)#-k(d>0`1fDu03da{9=(m9b)n&iP(|BOPb(Q
zee`S@22Gy6(RA|d?aE19=VTXUo(9>pjBtufiP*@RF}-A@F5SfYd)m^eKAIABtiKA-
zhH1pCVX+hUUAh7h*LQWm#EapVV-JN)H$5|Ezs~m}_Vf1LEg}a-gUY6m`|TU3Rd<O+
z+rnSJVlE-iF+aUMuUB&Z*6@gk)SBfgp>DMX+9j$|uf%f3NAFVRuy8^YpL<|lrv19s
z;3MMp>z9iav=DV^Zi~r+O={FI5Z^i|$KdwQ&&-Wp(({RiLO*pgLsULJ@%CVRtx?p`
zuBoidGBphYLf7Q;htEFh{JG&jMEat!Oz<W)hq1;*YB>Mio(aEgd?CX5vU<B(oNu-`
z_1v~05q4VX2p(&`E!daI53w1QxOH)SY!x!Q7wKU!57U|}^@>vdX46ZtytJkpX#w<z
z-gjs(f<JB9u|u6Su{!;P1q0{!Hi)+p7QiSbtiHNeJ^iYkTXeMZv7$2Dyakd6ML+e~
z8&L7rHNdXPASGkht8rs)d#6Xtowv9q99h{+%Cl=@8+sQW?=ZTK7<Uq7?jj~zJFlAb
zwa91_snv99e?QYZmTrFEQ_jw=6+cyF`mPHAxJ!9ict#{jEaj*DgFo`d*2`)iANV_n
z>ZGk&omRKWie&IP(iTnU^1pIog3au&cw!|ZbPyv>2&;1J_wkJFcG1E5{s{|*E+G*%
z#g}O-4;h^<T(P-;W^$4MBKLXc{UeGL|Hbr)uq`Fcb+7UCgx+qRi<qlXg}Q|7p0Lxo
z|ARw4TPp3qUZQLK0|1Vs2&kOAe17&`OnM{Nmq)Gl=I(Iiw*|sE+sj@mGNvgX?l}Uv
zwXjyC+M~ZedC|MO92iv;R<Xa=NfNLb%a>beEo{R$S>lIH@;R^JYtNqlz!7a@60H|i
z@<q#a`}rjzHNXA!>97)c2*4yRvtzcn@7#IkQPbXv5}I)yhYKbUt19m^HGCY7civFE
z+^&1cswXFhwd5EXfe@_E>XaU?s<E7-#g4hNloE92E*n>Wo|p0Zx(w~WLN4@-(A*&g
ze)Vo0HMfdO>eq=fI7S^G-Ob#)8O&yz?O5Ptk#=nT+NbL?l6~SAHB_p)d>k0;YjbE!
z%6W|$ur9rCl{!@nOTGj(+M(?4+pnM9U@V=2(N=@HJWiPdC3yv==+dPIDYi<#GjQqc
zko1T|SY(#b(^R%sX8%#+^(UO%RRG?CrG{<I@{yxQe|YDrCWj*IRAO8`iwjo%19R`W
z*N}MJZ0LqQHPHo9sVOdYQGcpW2FW_ud7_bt$v$nnWbSs~tN<TH>Axgd_2-SrBj^*&
zU9`kw=FRn5r%&MDym@#PwL3iY=H187J|w$qEz^XE7u$dSEPCpE2$KBsfQc~=URpN4
zG-%a}u5|w`?aHq%d)1D@uUXlLy~ep@DV1Z4#bxPLxLs&?*C%u&f5PmD0o}~q_W+}j
zDlQqOTl2HPVwLK3nz?`;%zgU$y4A9=K4^1i+_Y~z;ObRmf^}(>PIf4oMc>f1dL5%k
z%!g1SblMPsO^?o!SB;nkaQjNmbFaTrG0~OYB1~Yo@5c6}`|2Tgnu|xH<u#lA4Njpj
z>@2oIw0vm}XH^L`DsdQ~avM&oj19n>(d^UfM>$_=D%T2A4BILq?G{3j;u+n!vJQ0k
zUHjElTUV7TYKyZD53C3VYWP)LyTyRJ)F*xCm#kqk#S~A-0A7+2Z=fWGB#nB-L;uO5
zqQAoFq_xq2Tuo-QYP8jxVHZX3rE6^WfJw)<LkgewB0I5|=f{8)Q94GJ&V>J?n4I6R
zQZV_k3EYfPyW8x>Hu1FKpFCv1df@sOZJ&1*_aE-BxWK5~YEzybICq5DJ9w$8nh^FW
z&pIsqZbeN~jf851@11S^@-97dzU%4ZbO$}XDXWz6-xKUkenWKe!?~i$PhA^O`;8pS
znk7GGD$k~=7c8Sc!Nlt1#hK_nYTz>soM?L8tdq|2q^giIWhhk>@D19e<BtuUmb+#~
z)6AiDa|_S-^?v?#xGNLKGzk0N-QMZ!EG|9m<XT;k=p&D>|Kb<7&m{*V4={5m;<RX_
zz@zu|3g7@0=uxbyIkHdlql_vNUKHDU9^uJqSp<zLnV?X-y>=k&SMVs|WS*ZDUX#-x
zZtwhi&0c>UaM~eHc?<UQwm4;WbMmDX1@VnVgn*_*lxDPEs=Qrm!aSmtM|4(dvP*?x
z#gM{wz+vzTjmU1rUn%SGe6Tptcm^b#7|vwjIdhlqX^AafT9UtLzV8owlU)#m1m;U}
z=w5Px4j1j$z6hEwBT6hW%LRYmp0@hat@Z8e|C2_%q7sbp80cM(2|}M6{ign$v}jiU
z59YcZ&x}Mmz8`Yp_~W{b#W9-`Uj}gPKRrrLI;%{r)1&r|>NRknO!s*nFhMjt(q~_s
zZ-eOD*<@ID^tgUi`*!bMlh%&??~hNtr>iTA9mb?j2{M<uRdB5-(4}6#3=`@K$%YJ6
zfZy67uCYs7j%#kU<Il?@rn#OiqmQl~H<VU^xvEQxwn<Jj3?&!c)pAG^^5{_@CNeOn
zZ##vZxi!?s$HK|WcpHh}9AII~l22v97Qv?1R?J<Ko7iD;W_0)Z`->L!d;NTU=l<s@
z&CB7Gcsr0Xib!C`fli4*^GADIA20fm=GxY&SOp}V9#6ISLt1Wl5xs@pdk->wUq~!4
zmO8AuUx&O)W>*{^uv}cT>oJ6!6I*P*pr@GKsGWIpeceBKpGKnY4#T1|%aTZcWbo?F
zhZqe)1nfYH+0{I-UP;|%F6c9zwg&%$RL|aG)qjcKl&>kavuwSxSBK>mWhTOxJ27-^
z_shYC3obIK;BCs2<45|gUbt{-r<qH&SG2eR>*gO7Ur7V@Q(XalePi9;6{c)yWmHkk
zRju}&%byb_fB$8iKe$1C^O`jbLhYhQlB)H)she{EKssSz9fQhc;dee8EVh~cTvb=+
ze4m5t8VDO}mL+wp{OD{Z!~+n7kGX$%*RX$j6>>W0^=qD1rB6ZsgOzbc{v=y}_kV)x
z|3j9z%fsa#0p&eFVeOkWjfy|#Om;1ro9>@5-2PRs+lAwN+;+{l`r&j>Rl~|V(srW1
znf1EH?nwpp<FAhMan0X8^rchI>f8S<imdOi{`o&;jLrV9KloMH<wnK&=@wROj-EP4
zrCaw2!>cyFZrP!`rbVMAcLUuU@2<9_#}32ses@PL4O?|Oa7XR_ox_?o>Xc%mY8u_S
zg=uJ~b#a!8f7k8~PR_;Y4O3?>Xd1Tl*6f7X-!m31{QP+EkVS<jCoTLW-^{fBe@f^6
z7eDOZ5BYp2${jy45NT#eiH%f=ec*q4Z3c7&|8a-6Bio_kU7I)A%&cCUHhX9~2}~_+
z@B~95`|ptHM$c%vibSq3ZBe!59m;<1s3DdBG?!!1#?u93n0W)elI#vpAnu>i)ho-N
z9B6)SXa$6pfZs>OM@yV{7=DEcsJq2^cS@r6J$p_Fp26uf+cJWE&RVYZ!R(p7$-;%5
zX?y>d&k1-ou;2heB_KOBGat<knDsImU90~Ty&*$dl151vfYzS4&5Elu%mJgti_cl4
zZF$`?)FY?mkV~7zDibj-eggBcK2uysO-}<hv7Bzxp3hub<+rpDj7E>1PkbV?h&#Ug
zpXOHM1Hf2PKjE|`YB1O+;(H?MbdJF|Muz^3QES_5X70~-u`9o(^o-m{njj<2Bml?|
zV0a|rC;jVtM5dI1P!=}5TYaBgVK+?nW^X!ox6g2{aFoe_<h0otkwfV3%Hb?}XI>o3
z8Bv{KNKI3G)nC<at}W-0?jIPTDJ4b7wS<T)h!r5C@IEQZm{hh6>MaBnA%iI=t1CVg
zw?<i5G{FHK?v&ZS$pOyS<P?XoEpsEqr2`UysEdJhj5cf-Bn1I6s(q$CeAzh+L=}?E
z6UzOlylB|#XvSP1q{xk?DUN*+=M5R>orp{)4>7yQiN2T7GSur<XP~o&ij{MdCY8NV
zDxv)%(BZ*6{fwpF8IUdEnsi#m9fFaesaf;&NuVK4ALGsXK0Y)q)s-~PeXSwYS@`wq
zM>=8aqiPTZo)VHzaVXpu+B1)^klS@`b!C5^?^4wT^4$n6XYzGb17QSsl_SR&YK$lV
z5Y3&B_`!qUK`Yg)2v_mN=IMdi+~<ah8GN;ej~|;SfvNdZ1*;lx$MQk~c(tklw4#%C
zX(vv!`d1V&7?!F<2E9{y)Xn?dJ5`9!#0<E@S%}B0dj<Gq9h{_BU(w$1b52M;quf@m
zU*A+&#g#dH6}ARGMy`7XgAJ&(V`&_wBi(8I4dW62#RV8PaNtz7O&4VWgcai(eT(jW
zEqrxYsf%|M_d(VyZJ0#p|E=_$T8S*u%}osf%K0f#WPV~rf$gnn;CF$}(70FyuB~*_
zGfh0??Y$dzhfwCZOgD>@RHiZu9t05ZIqj`(?(6HW5-O0IWxom!&d}x<5lLUu*$%#5
zv$#KET1P2$hn2TRuZyN&oZysw@t<-{HyZhzzOKX;SRZX*^G;2I2VUu~H)!R1qo@L#
zL&FYrpVB?^5<r?5kI(_xzCL5)XszD3qrmhrq7!NKaa)0SQA*9mQUpi$&RrRlUB5o8
znW}twtDH{8o(Awwm1Zf<&y$igzT8yiFpihSJQpq!EnVXqgc8K<Dhl2iCP}-m51bW|
z8dY@zJ#|v96yrw^9}ee(BvOgZGv9FdoH=i`mfh!z&$2D6JQd9IF&&6r;S&u^b_WB{
zK(}a?WBU#rs?au$rKHV2bR8%IA(zbX^d1WBNvwZ4gYmubx5&T`Moa}nP0aDHeQkod
zPGm)S=>~o-_%zI+xSpc%r3*_E^FKFCCS3wbC+~F6=(Lv`qqy+)Up^i^e*83PK0QXY
zUNf#K@RcL)Z~9x@cYB@IH5r99{z#wpnT4i_`F=a*ZZFK-wM#?Fl;)>LRwXIl0_%+y
z;@WsOjKpDy+0Cr9Q2+Z`zrr+Wm)wtt_nidIlE;^FVWtPm+JPt*e|SCZ4nlmSYn+T;
ztwy>7*fga}DC}S33~UHap}l6t3|t%Q4)+$7@timxw&EWS5mAa>+I?VrIWN4ik{~`)
z4G9~$MvWS`g-|=Wi)`cijhB%&Vr<bv&?2T(yXsw@8(Up^gwP`s`W34v2oZ=-y;iMW
zy*5v$?lGy^jiV|4_5OSmYv!>>gAC6kP}oSiggo#Rwb7qLlO{}%j0!W6+F0@t7-C`c
z35OA#U3f_jBvf@}W7M_N6sJmgowM2f_~jFLzQcX6C>gR}rkvwEsEdd}28#6V5q#8_
zhSM;Z8I4)Xqa`VZdQYD)t5h?9J0=A(IjG2T;m&LZDq50kmmT@*Mr30ZgWa?Q{ow3K
z<p5+c`(a->@~Hg8GL?V`r(xvMrRNVs!vw+*=PfmsK(RQ8>||>RFPK<orE-sx>=bV!
z6N2QJc^e!WWk|S8q*dyA*8AUWy#kwjC>nx?LniE--7le6S!;~<Ydspk`Ge}zeo>9Z
z10oY1JTOT*IkV&@Jm(Hix0WyC{`MvEixCavK%sX>`HV%q1*cbh>bJVN$;=b+!{e2|
z(iZvHRx^?*F$THcx-0RuM*ql+4jrh^EGUQ{;#-_Z^JD&`y*sbLViAixPE5-vhrv{F
zc|GeYXvMWSPx>r-rE?y3skEgl%8T@|jI0l1&R&1W#=HfnuKg(FaM=_w1itlW!cY7k
z<(>Ik&-wrFUu7*BA7h5nAeqc4B!k8tvSrB{nvvyQXbi>@()wXa2%$ts6xl+nr4^%6
zQQ1O>Q6y5{rKr@oJu=^O&H3T{0p~h%UGsyvG`-)i*K>J1?%U((K9?6xE^nS{KIW>E
zyM|4Y(ed3?douQbJf%({nCJe6W~&)mfec=blopbvWE)|lmLrBK74xCet^V@|?U1xt
zAly56%Ek_~UOTB@I&8@Q(!`cVlkt(jMrGW#g{$ijv^uJsSMvX4#rOOru4lowmMmbI
z;z1Fy65XT7LVNV+0W>ph(d41Eby3(TMW!qHOM&&%f<=YPrVw2*hyO=XE?d_;Ms+WK
zaUmz&&3OT$`{oho4qEdgew5RNMfcMBGaxrW1oO`*%OnElNciZ6ugOIk19K$x2cC|u
z1rI%xXC)<Dxk#^&*C}ewn%c7Do#58gIo&u^1!u#=dVjIx>Pw#?@w{lK_|lu{5|-Ya
z*ejhCn=tbyMML;Y#S;l!VwGwIxUHU^9zLdS+|)7%z9^d_4{Z+d1b(;Ux_H!0b>99d
zTJqC)ya`oRjdQ4S-*ht+xxdT>;r@C-No4CXA~7AVwzX}|X_*fq+nieQ<Fid|reL#?
z0!b2@c&{234^<@(k0rB^Og?H&+qD+aF3O&TxUnUL5S5Nj{qM{Y#;AMZ+h<B})fEU_
zMWGjKqGTimINvPi8Vk0$w0KKP4Ee%8@*Fnhj~!iru6YV$RP7fpenAq{l7ghf5r?T`
zMp1aiZuIDg)NdrnfsFZS6mTU|NPR3heeOI1TLpTMm5P8?WEjVM8}H)#&6Mb<DCODq
zGjH;G=HsL`<R);c7X@dVlsjGS`ul%Fq%SU#Nk16tq=Y!$X}nY_vnv5c*r*0Li9m?b
zXq&;tm*mLowr$)mV&%t-o5D%tKYUUjdfUMl*OmrFnGaF~^5&Eq&AXHMX@U{wgCA!!
z^#WJYP-?2n+Q{4k9LT4aPrAh1w+K~kvNdcVabkV}0bU$_dT70R#9Z$G)8zgWJQyFS
zQJ2`^Sw+9sH5*@XZCzad-jHOa!)3Q8*Hp9KI-|#)!hJ8u5NU>9HjLLcH|5G;YyQc`
zid*pGnalG6u#%pC`Qb*;e$R7z7G?XQ>C{ElogICU9HD6g&RG;iFHLFOS|^IrZ{EZm
z((?Yfc{@4KisC&ADrCUMV?xg4$~ti1z|+1#=CR8k#b`l1&DcS6jZ{8-=2~8j61+TI
zo}KJ0MLt%d6Y0oGtM)xp#&&W06Rail(c}hKxqg{8jKtlO1)BTJ=uJn^lvJ^B;gwCO
ze(EylQ_z<FL4od4Or+eIV;!V7r6cQn9Kbu!_%nR*(`R9IP(s*JV`6u0J<CUZPOgR}
zpM2rMg}z_uG}OE~Ckin{tlQum-xMyjt3UNRS-3KU*MvuTvb&cC^a8Et(A>3uzgEh}
z#`8<llj^#uawsn^*%zQgZ^kVm92JHYI3j9$>E*C`)hePQI}gb@klhf(_-aeb8U0`y
zTJxH(QD(N*u77fIWfLCNNOBTb-7O^tqC?wZOYxZqVMN}o>^YNK%SnLQ=m=7F8%^+k
zW>8-F=`}$WfTRCGVA$2r`Mw+U6@jnX6l-HpH8WiLy5_;5pZ7e0v<3e9kxX&c*rBp|
zKaLKGDVfiVOs1JOj^Cg={0bTwLUXpsAJv>TDv2D5iJ75kz&atjqCKOZh57$&S>T83
z9p)MQ^iw+^^7}4zv<!;pg$vaa4w--38k^@|K;Eze`ZA&Q`qsqtQ{SLOUy0xqeaJ}X
z{7PD!a#U1?OtBCjkjbD9J!gBO6%9dRn2~i)tV<MQ0y#pth0JUK{}kY{-;x}~^SxBe
zHq8|I5iX}I`T6<9Jq`w)Sh03(+~nXzHVfSQH3?ee=lm3C#b4t#^IoOoSXHDQ6&>xg
zSEI||wOP;Nz8q=8uc6>FWBMc-u&xZqIqX)o|LH3D$Eli~xNWa+a*21xweEcyw?@te
zz30*O)yWMS3Nr54?ncs>;b9obeIdE6L{N$8O!Dk_8&bb0*7L&rW<3e&wH+~&y633T
z`H0dbsf8C>_$jUgrOk7urZy{Iv+XQ5^c4iabDF)v0|OJRf}uY_&iv)IrP?&N2Zwuf
zMz@<Lnc6}Q@AoPI)McWwq;GQA$Y^aY&yUEorzX3*`F=NW+KONLRX5pmuc^$QW{2V+
zX@d6g^ab!$r<`LGIw%~NEy|y|lu42X;-DtwYfQYbzzOw(fOa$%SBa+(s*b){97$ZR
zXYY2<)7wH1x&S&x_7)F7)KZ*z(tL{!3({39WPWw(QloR@Y3G}I*45XR4RR*DVK>E>
zbZZpgnN*zq2`Ec648&3Wh|9dGC@z0=<Pz4jXy5zS6Emy%JNU1)6v(p~w^<M-Vt(GU
zCp7HX@#9-*@+3ZD+O*v^L#cr~kRhP;)mJvRB}7iKhZ?eVfMjp#srd0}_11d{yyR27
zOxCqY!@P#k9$b>MhPCV5xp!kL_UZm`UbTh+vvG26P_KlEshy*rxU+2>V1jusFGyN8
zC4c940}YizUR(c<t|ym}YB4Iw4c>2VkHf*iy_LThvK@YwV$hKoOqg#Uh)lvt#=~FK
zQLJv*dC1@Ydv@Z&6)Q|p5Q_(#1N@TxqD3zs+Qr)Tg4Z*3dpe^Z8^y4oK0n;L%Q07q
zNyQJ;dg;9i#<c7o?vcdZY5X~oYF8`*jiG5P>6h#gs7=EIC{Y|R1@xtXv>7*O%-3Y-
zyR1m)ZM>_Y;a2Iatn6$?5q27x?Ecw*&Ke3VN(OVuCYA#m#7M$UW2*|{Asu)CvcTBA
zP1e_>p20U_!2AXlVRJ?38o_1=)!EJ|O)$fA?uZmtxOi--iUg$qifN%xu*l^s^Nw+C
zOY;HTDRm-!{U{(S$yAbTMDFZsf}NZ>(^LBVhu0(^SSGnkF^d^X@@wVqz{=)eZWo3F
zca-W#@5NuFHy)iaO_HA0z9h4~kKJZ0zFj3{(bUfM&ZMq=AL07=gOJ>aIc^O$=iIc)
zqdUw!Hth<f53-z(g?p?;AXIekL+U92DFh{3p^+R^xqRFOlr5kLqmWwPxAUCm{>iAO
zs&pGUYn`bkMsxToB3~n>JPm7uh>1yl7TgRO_0=B^yKcljXY3@>TA91Vr;wp6EZr1*
z-BhVkDo9D5IAQmfO10nxx7yEZ-|F&nRmTsEw2xsqk)KpD$$V4#62pz0S>1*Vsheb8
z6jA<Y(LZy}KkI*{0R7pceh8nfR({GK<iVBb4$y!!eV5ferjB4Aw5AsQ_%=Fw{%<ul
ziJC9A&sY96m&N7%;C~^pUH*;48vYxJHS~=)Xlzdsug@^PcVM|@TkR;PA+@~|TWi-m
zYEm_CgZ?)!mk$2<{}{3I@82_h{DNCQ{O|C%qZhOn|9e=w&i}iANHh{ClFphoKk4=H
z9245ZCj1fd!TgE8p$ZBV5JODC%8Rm4_6G4!&oQfcctyAlOpTMEa0WDQqSHlDx9#o&
z@c@9cYI@Xvst3}pTyX{HI(LE$p#rMJ;`S9b0PZfC*~K8w5wIHQ_0-1?jB7BU5SN09
zo2p>Vbi0!@IY@@vBAZzsTKcoI`g_2mnv9a2KrtgnUNt#=xYDIvb#DIca1CP9wK(_J
zGdfOmu532L{&5)Vbh%I6@7&Cps#{%qy7Qi=-M{Oy#m8>ASP!&+q#u#Q#4o8+o}Al;
z)WIBc0tKLp_M(|HYdNmD373dOBBDV~czrx!8W2os*!Vq9*PplN$Irc8KzFatFLzCG
zYx49jx5~ImM@6a>1)8RH<6c5h(&1#J11uU?*sBA}z4u<dTrjU_U*&-3AIDi%F4~v$
zy7_<Fer|UNRvP_{a40LW5pV=M7G>`<YXVX^g6@wxLsG^PkL>+A(T!fT^7sMe*}b?I
z*q+iTaO~!Mu3YQ<OHDt;-oZgbAsp_L_b=+Qsk5tV^Y8AoGupA!(sD&}S+?JLF7h7D
z<>|E1Ly-|&9G?p@LoC5@V`H~r!yxI#IOJ2VUe)AiFMt1DUec;ntJaj%Pyqc$SGOhX
zA|NkOK<zm&(Tw)Jm?*s<hcgKu&0sRGg0z4b&?x2pQV*aFdS`O*;KBKhj^dJbg@t2~
zUy{uwa0@bSI?%|GfBZAT8Qz@<DQok>Zvnjuh(L|-TWm*=e@So##W+6m9Zf7tQat78
z7`=EbSCEqYgX<tC-9>|O{(K)q8b_hrBq)y;5I!@Tlo#+K0VjwZYDO9}!ET<Rq1_~g
z9e(Y%fDUcJ+plra!18CrzC$Q4IXR6LKfezIFx7ha=cb(3B9;Z&l)x_L!=Lcy1@(Z-
z6m4g<?K%J&B);TW<i%SDnx)<5A;<)9sY|3tB;{6Sf@1;X)A;S*fAT^v)74oBRSbIe
zoOtOLg&$8`GRk9XoRA7go2OWCng^`T>vKGr)=Z!2(o78Sg16wR%jKE4R#(x5N@}mj
zQK<0)td2#86XKVj19^|{azuoUB%0BF_bE<#Ll=$EvyGz5WL<uS+}jh=-)%EZviK5T
zrUAGo0}jg8P1jXfFk@Oa13JqTkPYHKil4-rH<F1_-odrF{^Xge3&)FhO`bKDs^?UL
z+`0lH*%r|Unz`4lhf;41QtF8onINH!@Os5>w|t~NMl*ouqXi3%*OB5;U(rQ|7xIct
zhvX$x#m6(?i1{5_tT@rF0piW3MG3CGp*LuG`$dNb#BI-xj*Cm>s@AA`Qr%ofcghk~
zCXH%al1LyLjbD6>FH73S{Wtpb<LDP3DdKbRj6r5z&B)LeHk__xP04`DYRiJoo~gbC
zsf}~(Mvao7H$032+p!=d_Bx|*U9$e|ojQ7ZbV|nt)vD>Qn4A=}Dmy+tUIj9vNIu6A
z7!`@W1D(UeW6E0``>L+>(Bo!quSe%)?lary<D+fi<CDIG<<UV=-x^#tZp;|(D~B`F
z(<QfNE0q`1GiHMjbcaTg2x$!iO5fYY$qTP*i&;~o2|NTNI&OcCtgVL*9pbrI(F`V~
zVH`{V&zlBIY)_pZMa{i`s7qWz#jU}b7DT0;o>@<ivvAq6UThY1?$|CjXcpM~&<U&u
zQ<}Xg8FX56m2vcN4kdR__T|N`WHtrL<r|<*{;QJf)Z36%{VKDqsd?3KNEu-9T{%74
zDSfFTj=iPq`jY<zkZ+9DZZF+d2FXJ>9Z7ehavv%l-naP|tZBVXSx*E46++caFVVvj
zr})ukr1Q#5z+|I{qG8Gtj!_PDD_W&#?%YP27LwWp(iye>{X&V?;xZY3X;o1n2f5Wl
z4t#CJ5lDr3H0dlrBQvw9$QZhEfGk|Tyf=l_H*I$jScjGP=RZPjty@jldt^+EFd#Q>
z-VFGk;=b9Sy*}yGb-fg^P>OStQhpB`YAgPJ#wbrv(&fl~u-JF#OBj>u*RG910^A*!
zo_HxGMw$6s^BUvk0;1U4{+y{MZl(kQabI>bO#M*J2i;1(GN++fb^-Gy%5Nc+TEd$j
zG6^gAmB77Qwr>w$iEaf-#!hS{7;P6K#FW2qFD}Xs?#0-9KOF34Ozpwfohh8b`p<F@
zRVHJ+ke{YS_3;_UBBCuBj|)Sx*|_Vw(**5i$!Rx+B-zaMGHVV-v?VYw5WxNQz0-kN
zLH#GdS9K+=O#Flh1bXF=?(L<e(6a26*Ti>{$IFc?{$SRC5j>HlC6(`WWL#v$F<it#
z6061T%-bPTYqvzj#16zqtc+s@lz2+j62tLg63)Eu`}d#OWAN_}`ZQWPziylB(#65S
zVGG!e;wP?go`b}%O+zf>2(2OIS<7PyW-k^k_~59@Kaq{|)c!A@wzh6%tT(T|aip(L
zmc%-EHU?!Fvd{yx+I9B{4Grb;iby)5m*(c?c0Dt50(FHk2!KB#SxL@Q{3JE)bg6fJ
z@Oo$G8x=eIa4=+72POcqPm-lChGR~!bUx7c#>Q53w6JcS2HRNLxhCsB+Nsvx<NKN;
zB8uYmHq&e#Cc0cec{)>egi=Upu$Q%xLBmqSv1R2UA%pin=@0I#|0qy>*)DZQTT4s(
zA&oVP!T%6%;xp$t%i}f)3Co6tY32##OY~1W`#Vj01B(8I(u-IVr;?#Cf#B9ww6W@C
zl6-Rv10stmG{C+3-CVI}&YCqVqLSuS;<>*Yymq3A8^N!d7J&W6hK5x4vpUb{-{zB{
zA;j9(w08*QtZD|AC3GWc4s}dS$;8Q%+f(!~arvU!qOpUFIqT)!eVp1gWDOm*rUi>5
zEX=X9TE4Jg;l^D$b~IM(Ss{joYW(`da_p9Pv!-dVBfS)oG>xomrDM!-wrJmWi_9JS
znniACDE&7Cp$~mC{mIgR?L@yRYRCuUyceG~uTkyzzzx!4?&TQHfE$t@kwMWm^U~Su
zc{^X}t)AF+7s-HAlynfSn2cc?D#r!&b2L=2_SlUtyB~2t@dGFL(J{4Z-OW#up>xL=
z8hWlHWe_;V3$Ir{8cyS`iHwT6Z8d?f%%foT4LNB?1F(gkYkIXJ^tG6U?^>N>tb^jm
z>5xH}8S^NDkTt%OmF<G<8bQM-Y9a#aaSyjroZ;JAt$D}YLY&^r)5o298ybR=Yg*`y
zrIImH!ub7zN)MQ+ylI`RC_Z=}pmg6TY9b$h+fM9^D%BSjmy*Efrl0Z*vL81*nE|Jh
zIAeUkkSG6f^TEpKJ~c6s%d#>#5S(hBX71qUkiYEV1kV5yjLG=Ty%inev9O_ocX;rL
zY>V1__p^Z#2E{zcNmuQnc2#m7ywA^3LSEL24bW!8^5thCTBpBpt+-sUy>x6%>i4SJ
zT`!Yd-G6_8y-a-w+V|J-0P>8i>M-^6y^Ix`BGdZCnZAy<<KqGx_W70A>=_DGk)Vf&
zt_1@|y)29(GB+>O3DE8S4p*~l_@Y8<akNNBY3o5fu6WW*Dk&6kL=F6xO>SspH2U<J
zJ$u@*-lb3m0(j%Ktdx2Xz+p3@eP#XSn0Kf|q^5lQ`0*d5kNQhylJW<ssLZHrN*6Ph
ziY&HdKTtV<{U#16erC=<MK0f^CwIrk53YLmWSa=11`Zr34S@{N11K9irz%vG`G^m@
znwYf0d44-X8U#>`(14e(Uu!8SCgPt~iv6g9X#5{0PKv#dP?~e3i87zkO;IR^WG)WW
zfH%QrmQ7*9q-%k((@+4tk4}C<2QJE!*xT$L#F1-DJ1Qw8@p7+@9W$mm5|_}KYoWm;
z$&DCNr;^_3kzxjf&T55^SdxTeGbUp@Do}k>xK9xZk-lBrF}4mjv$Fh)d|;TJpGW$Z
zmeFo0!M&%tK*M*Uj`cyTq3@Oyv{r(4(SnFFCq@m%*Q{#vWM((D>nSq=C=PM)h`+hn
zsd+>@Xad69mR6OQTQD>$O6TJFR8RPVkr6iC2yHbPN<JQFX-2GZ@ukh)-oG7c6_Rh=
zuOLsb>7XELYw8r}tIcHkDnDj^+T2?Ic(l&6J89}S*4ee-f*KZjE&3yx!x3(MNS*3?
zz}c*-_NTb*Ue|MKhtRG`VwtGi6`7_iF>%5_<P0Cf7~7*zN1yI5D=ynItXZdI_G+|L
zFZUzS03>~``URNRxM0vz%i*^f`kY&H;pMt0es=w<L+cs?B#%(@h!H+2At@xSUySK%
zRhI4uOUq{0=w_#C%$Yq~NE2k(z=?l>puA_@{tYS9$iCh`vxV)9n5ebl$BO0N=?W={
zn~)5$`!#o(#`bGtOwOEA-mkpMbq+vjT|vX=E}r2d;_H@F8OAU)nWaG;y};3N3v95a
zr}~f3hk0pSRN*@~4Gjzo6zZl=dzZnvL8!<W3qWNJPqjhSkek=8P<_PQ&I~XKpMIYj
zS6S0p6IP{jSa6#^HKuf)5}deJq`TZaQ@KaEC~&%Ja>q+quN1o-sHkW&D5ao{S$3QQ
z*}%*}mH4%Q7+GGNPNTu#D&KXR<ELsA%a$K*s2dypJ8KRRtzX$n-=T|aZ94*YN?tLe
zBMzyZ-IB9^ic6+$kP%VTf@8Rs`Ve^ocT#=Ke6Q&ekN$g;YLC;YDfLNlR3Y~E_R=|g
z0TsM#f_WCK!D9!Pj>vxE=PoIuGJ*%?6?||D&M_}4*pj#H_RSkiY}x{qD%5wb_wBb%
zC(54Y9L%fG>11B9@NM-hzX9!=k4lT=mX{ZY)n&77gd}nEm;<uk=y$^+=(O=-jdTLv
zn>N!+uDAo*Ea07FI+5}%L+jXH#ZAN26_;ywrB);#asKQ+>_BytYUdHR@m*A@Ee#Eg
zb(5G7Tx_%-pp348c{U$G0(LPkLI^M6?I^s&*G3y+tUo)Dv0VVV+0nybFqA17jIl~v
zLS}7nQU8XS!%Zcyi;Mm_wq@}@^ZK>=$uX?eaj4702@^71vuW>wRt2esSyw!JUzgcN
zi0v~Kqn(}aHT0ZhAR{>-s%DoFjK*=K)3IMm%9`F7oP<B2R}lty!yB8^_nOB_z#LsA
z6EjS>-=^SVau}0|F-_;xH9wW(*&|7-n+M<Od*?!KZRU5;Aq6Y!KIlFwNaXbW`!B7m
z1LkcwQm)YDn?ae2n^s=4y?vPanWOL3L5J&fZ^vELZKA!(THocJ_tO@@d7|Q@1`?^S
z2+b4;Z^LL6ifX3$g>GyXj>YC4X+Z<t#D7(@LzQ)Le+O0aa8>^7**c%=p#nwp!I?du
zZy=p8)(tcQYZfil|K)<&=SpYU@7TLGi8G0R`-`STbh4xg$*~DZU`A4*>*BD)E$(5v
z^2)-xCLeA#^!Sm2b#^+Px-sr~zKAM~?Ij`*C7Fz<1&om-E+r&uhn2N;cEy6%bAr^Y
zya6W?hH*v6^lOAi?}6B_JbXCEWH^P}n*y3zC8Q#5o?CFQrHdB_E*?m<xsty6M%c>{
zqo$udPjOr1JaLi*G4Ra+xg@Kw{NAj2jsb2D_uF;s;##gDkS=RwAt4d76E_Hx7}E1(
zZ=Il&mv9#!UM^h}+Wz>}8uwPW3!XfF+y^co>)^n4<dI0qrjR+*F-_*one&CPKeGPL
z=MQt(VY;iWlS@|50zv_mfveg2_U-TD2xKgB3TK#@AyCa+q2xT>sb*2cB71u-psW+h
zx$ft~v}%UGp_I367_jzn*0#uk!)D7|pX}&7bSqkPf?`D4Z#yBWmCAeXk>nq05;ETB
za|Qu@FbkQfiXGi+UjHD<B&h><;gX?osHm*tic?*_^)_EQF{Qk*L-wBF2|)KObmy><
zbSZYaQ^>24%^DRI<+o?gCJv-fj_35Oth@2UA##q}oVSXu>qS=YiqD@)Ynv%3AjVCY
zFp4sLP3K|Al|m|>9e<elnyz4w(_Dt8(>!Jjn`vqskIo1D;W=8F4EL!)Mt0eEdiLKi
zbHd|(nOcXB&Dl@{2=J`zSTs0gZjX3GM<X4_^2vq06)BaNJ888x!uuEZ-07khRP?ZB
zQ2?yh(~x1RGekIx0*2Xa;{HFjXGL*!ZPe!{>@>PW(km>Wdjf`D_YIimK~B63exUi&
zo{?;(t<nKP@IG)rUzD<N_Yp(dUV%+KPYR84uKMxKn>S_po0u(tuF})fEg096)ac)3
zOqy?J)5c6S2^yX(8mXB2WpDr4KaW0Mc1-oB4lHS!psn)lS6N*+dYMbW3v)$eKtlCX
zcge763c5P43AqI`h8ydw_<LWrOYXlezdB|YF$r%tU`xolIB%c%d<xLKVN9;^Z_)kc
z@y8rocRpo7@`?*V7M3E6ortfvlCQ%P#%2Hj9xkV5V7bhDFqz{I=c>WLfqLwFs~Xn!
zd&ET{gtsg8!7HZ$DdOdP##smqeZgjK0}%mczuo=OA^R|Gp{63xM(pNXnEtjMJ8tFr
z6Kfd+iQs7XJVf~xF=)l(i~kif9)&qbVo|H?V<9$Yf8ByMp{T%cBlcYizy$*p!8ASJ
zjd@RHca<7Pqxf)uN(VTc!}JBcluYD8Rhf(UH1Je~Gq|O&hj5bQF*!vf&|ZF15*f_&
z3Ui;9NdM0T0DiWUe9M#qdc>%}yy(2U`|9hNv%zKw<?{@RULr1t)vM=SocAEQH2v1C
z`=>(DLkbv1hZxZMePf~?HN+7zd<L1`$eC4L@oLE#4ph(ZykX|%9YBolSrwqm8&+37
zs>sgt%<8<wSW%hmmFMi(s=P9|X?@b7pujO?Gt9ejC#N!UaqXV2yeq{*Ywd3sF`6JU
zNwf)yeDN}RnduTi)Xx24rz+%%6doBUKd|Htov`kmxWG=ywIANd{O&_88abuWHAxK{
zt3qF;g<b8~!uS2DohQpSFi%mjUc63{60`NnEG@&fHEv}&c?DP-@cRQ#D$5YQxI`DA
z(vyH$02CSjEukvHp}M|{&K@?3DemqcpSO67!XND7D(bpOr#CilA75%$mRQ{xia=w?
zp=C@X-cAp3g~7uFI#>vBj`3Ge1<2#;gh+q9y0*QEjSeWaXc8po6w`}XJZDGQYw-{y
zypi|?<cl)yc@)}?w)`&v*Z@+#((N-}!62|%cw%Tj8Na9=a1FLubea&`WBKc%F{3p#
zsqKV&W7_^FafsPu>^yyoE+aY!!T*C<W;nhTbRC@GPQkVJ6MLeVB)szoCyJ;WkjnZZ
zemF|eNOFs2dFAufWa91X>!8ij0ODOqguXNlOYff<!5tKdC=4>yuuq@!Da*7-apYWE
zNqoEH?k%=S&+Kb%o(9hM9<@UVDjtz>0llB+taG^UpQN|TbSZCLCcFT9Y?ZnWGJBl0
zbs8P1q!*EnF=J_2_nTRn7=DgWCy+8T3MxbAT^VH}QgpAVA7(%xqHq6$c3DQ;$r*?0
zSiS55B^k=(9c+lJnVGF9^)_OGMkyZ}<BZ-FK|%^`ACFihk%t_-o$s~AquARE%_+s3
z(^I=chZ`66iti|cegY8lrN5R^3fxY7Ut-nb{Ai+3#b_5j{s6Mknj>p7UB7rzGjx(*
z4P=TcI?Id$Ir*_mH**udU?Esg$K{ZpBfSvH0J%*8I5z>;>N(YM{!7yDY^KJ7q=>IT
zmK`aVg3`-uJfrj}ek&^+YH6!opP-3r1#Cy{(S(bw{KJRqAP<oU?L@6lKO;;P5)e=N
zSCbI~N9k3#s1s)OcVCCFP-3S<$pADA_beSGIPh4OI`W;<Z~q&Z_u0<RJ#589W8al2
zx?cUO3P%@w=)1^Q2~N6RM}38C`Om&f_xbp!yV6w~a99d>HjXRlIz<O}w4}~~Hwfa>
z*QM(B@z&OPpB~dW7D3nrvXr->pn<%wg1B<v3kw!=NX<tGWKl<6xX=NJgC_^bV@Qqk
z(}_bUCM7u!4ES4EyOXxDr?_>-0|!ta^eAP4hpG2V_&OV_Cq{KaSP3LmbGT>nTVqSy
z>pn&bj!1jkRb=-BqDc}OhOgq77qruoSl9^Ppmn31lwlO%zi-%ZS@A=%SA3Tt^<6X!
zaPH<>Sx=n!jUw|ljOcn6Xj<^bjX3#c-nv6pBznFdScCUZsnA)IOlX2dLw*;Jnt^Nr
ze-aGOD1q_7@`Y45U+e3eDBlkm@@x3^poE>nYQ~jS=U!U9Si=DDEANJ2J<8u@W#^_J
zK9ZJk<Hj!T5GzW-fcm~%Iw<n}A;(%u5KnG?+{Vt+Eely=<Ye^b%8`+F0C57R0~p^q
zDf#2Knb0f{Gq9VC?(iR?sv!e)+cDPO--U(Wqo_ctY(m?=n`e_yUe^z>PEbi&lVhlD
zR#m;7W3&R9<!zMgJ*FQ0Q@}m&?l6yr`c?1$9=)lgeOtk@A3S*ALt-LIDrWn!w!0~n
zJ6Z=NpnZ*~4D8gz(h@01FHXcGD2lh)MzrS`e@*lrq30vQ&m4KhhwrU?4?rFM)ugAi
zDgH4$-W&7)VU$|aFefu6br)Ye9gl)B-`)1Ghq`b}9nzuz+OsVL`S)GBopeP?5U@6v
zbK8oOAsmG$61Z3(Y}Ovi(ow=~NUHwpveX?1yzYf>cU0N6H?`}j*mF*RnfdlY854*&
z&YHgedTQ#Jj^@jIqC)sj+g*J65iD={kCgPaS6BYJXBKIO5o<<ySaZ@Aej2N*N-cX;
zo4E~QW9*<c#RiBSzGYFfHm5Ov^Ajq{LU18lhYwH6QmaBcY#;Ak6`Jra6P`j_(bCbm
zJ20CU)=Igw;lnAL%SY0Brmyf$dgG9#OT1iv4Fj6?oPpiR77hQ}c~nCWM#vO@+YwtM
z`p!tg;YPaUre+EdL3j)?H7j3c>3w5TwPD@gf^peK3@DkD@{jN`t<g|sw+FcfD%WOc
zxE`9csz=fC70^_k^De2wb_ECT5Keygi$>i-ZvHkGQ5htT<ng3Ynl#=DB&h>rnN$XP
z3^~nfkUK-NoH>7wLUKx)4=Zf}mKgw-q$t%F30eLY@v9x3RZNngE;qF}CYUQq9%pSp
z@6V-JetuuJOl&*Y^ugMi@$|Ttu5!%;Y`^!9-eXhahk*m)pKljD{=e!bRR0!H{QvWZ
bT95Dcu5OxQcl5HtKjTMFwmds>{>J|W@-@wt

literal 0
HcmV?d00001

diff --git a/docs_new/examples/images/simtom.png b/docs_new/examples/images/simtom.png
new file mode 100644
index 0000000000000000000000000000000000000000..5f6c926d86b89493990cff0e14036545998aabdc
GIT binary patch
literal 109657
zcmeF1gLh@mw(sNY*tU%w+w8bwbj*%z8y&M_TOD_7yF2RGw)uAd?z#7zci+G8GR7Kf
z?P6-KxvFN>_Y<kCD2)V<4-W<gh9oN^@fi#Z5)KRuyaxaSN;%r0Tm}O}6t)r<SC$nQ
zCslT`H@C7e0|S$ZOwokZQXR$4(M^mOvCNl)>x3i!3?T!TAG(fa6GACM3#5iZ^VU{#
zI4gkQa#5G4hcuQxWsK@%iK3-FG48GDkOlwpka98kFq!@F?z}H-F_qo+fDDFD!%MD0
zTmdK2RVPA7f-ggo^j(U^6%z3fl4%p#g;y^!0*`<I&JfS$Lw|h@Dx#`$pK|`$;=`Xw
zN1!VN0Sq0%5k{$RGW?Mg%mYi4k{K+3c894HGr|;CBab2sT&zPmtO2w2Qy4p@N9`&#
zFbCQZ0gV1DVJr^<*d0TBDUMS^UYHlj0JB(*2p|B4$hiCURaPKko~${Ti#)>YIVOh0
z<;QzT2C!&!%M{6IP;Xa~N`@3|&#~k^_vmqhN?3{SupdUF$;MHu!)A)rzQvhFa-<G_
zptCRL5I3TU?45LT^o&f3Zh5YbGF0&EL`t3EQ~>b+vw)&9)W8sGIN+v$BE?6lgfs*R
zs7Gd*kQa~*s0SZFcuL%aDxRdiD)K01X%teOFs6UHA*^4momdpNvO`xb|KQ7<);Kmf
z9$b#3LscbhrpKg|7Oz6CZ=mER6SHFf$&-mCXi1SeG``&u#FN8<h@^IC5}L6c?@{b)
z)n)3?wNzpQ&aV^sfMZG%Mt2?D<iLCVxTROe;6ui4r5w(Jhr!E(Lo7_I1$3{<WRd@*
zeC&gY1vSJlw2i5vUo}VEvUw0X=7A8gMi~k>R`1~it``sQE1;@|npqZs!7{ZKGEBeS
zn-(b310foKTvTTsAWI4hSn(q8+R84VfpQ4IH3b)P_{<he@p)HJ%tHZea8oJbQG|_y
zS$TaNHZ=Z&Z-Bl=kyLeEmzE>}AxA6=g<OQVBW+9?F1&*(m?j?WwVxkWgE@v9MML|e
z`v7+N0uh5P;gb~KVL}biQWlXK2CYAWOu~B=@X+sjDdT|@@EglwQ$P6T(N4+ZlO>G+
z^(v_g>bX4T1l9+p8RlWYGufD)>1#T|D$LiPJAth%YT;w8#CA|w4%~;~kuT!Y=kaEA
zJE>-&?@9L?=9m5R$}0(9MrgHj2}PP;Jd1t?|9pv$Vq6s7mqxQ5Qtr|g7ZiT6c#HQX
zwxRHA9>6w+t#e+YJB4t=v<7oIg<6PcG9EVduyhqnsI3vTzq<Bt%8%V7d56iZVoX9$
z(@7&vV~Og5@Jl!)Leff-P#p}aI)&iF@I!laX^uY;$T~mXK86W6eJSwp?Tk$F=SJ;Z
zxLEBF2D=mlva&kt^5ghH1o-ASypX(4Edw`8#jyfz!rMMO<_Gng;XwGvS@vX`1LOm#
z>Vj_^p^1B7p#lt1k&Q*jWuP$(Z<igiqkAz~i6+3Q@(>vzD@CD~0L=kJhTKai_&q8P
z@LAAlJy;HqtHG6}2pwR)n?#Q>qyV^la*hF1pm-IU^Z=F-sh<ohdgK5_xeQ-oC;*L)
zjVL2LnT#PG-jw}w@ToX5MQJ>O48IesXUHvSZGsYQSVcZx8Qj<4SP?v0jBg?svsaeb
ze0WS!2ZeGoO_rQL5vrw?^3!GrEMcD^&W6ECqK=IA(=mp6w6qXY!_170ZKyS(QcQkp
z5iLOF-f&hyo(u|YOJs%3JJPfU^<PqEMJo+iY)4<g`@rhQwQMe4N_&vJe-oj?#UO-R
zhvG6qB#UMZRTiVAv7sO&&kFP|kG_nRp*%*jjm&XiDM5FRn2(5#Jc;y<pou&iP&5{f
zlwhOeP4tlkQX-cJsEO%FuSoKV{`_=I?yIb#4penihf|SNMXQ)lr_)?6chcLd7-LVF
zHdXk+xtvKTzMoK<n3mX*m^qR*`ek<-lPQX6FfCDLG=Cz$s`!`6l=8XidRaw<PMKHv
zP&vGsMP-fpTBVo%UBa<6r!6NeXX*+MXZZ?x)7e*=-{)nW`Jc<Vet~_(ZwmMsq9ZP-
z;4AZ%r>N?#VNrSW%c5MfVzJUd-M?IPp>DzOtNaY@0_`GCo}+f5W>S?^;pp71<<zgL
zUjkL0at#L2C7hobq!<jW>lYNO5p>p8C~XPO3~h7nea@IxN}Kqta3>Ro3LLF&EOo6c
zEWhwc5a|*%E5<5jOxa~~Wea_={ldLwx<<bCY0cy(kym=oH)35DCHwMqzs^jr+*^g)
zjAx%W$!FJR&ASZ(M0Nn{CP6a(DX`V(ax~r!*CyS%b+xYdG7!yf*AUZ!W6&~q7O!yK
zh_Bx#%rQI#`H%2v4e;x!A)&p%q(`)4rDJzwxolbFh-vnyRg%7{;5^to;k<guu!8r5
zrMaIuM1#6H%|y<;>yf+{u`g3+=(>fS4Y4^h=Z4L?`TKI;0{N(M%^-0!Y$~Q)m|4jl
z{kZwK5CeN9ZRL?hlZJcs=mO%2%?aniLDj~erIn&3#yYFIqh`)Jvp?8>7FO0<3Rj-~
zC^T0#x7#T-S}Yx`)V6eQ2k^D>oARsi>A6V|2@*obo)G;B#|ZOA!a=Sj_8XU-j6701
zR6DsH(Vpg=^Q>_Wz8Oc;pqP_lByqGG;Kp%J6G*FOevtYahe*QfZaP1*Sc6=nxA<xC
zq_C_|OllJUH_vfOW+B0bZSWsk_wU}xFAxvr54FTO#Kv)aal}dt(=OBW)4GU6Qj8|R
zoerE{h8gSBMU~pPb>emC^_R|%Ysc%+yXB*K!4)?S|J#Mx>@Mss!i~u;B0raImu>|=
zqWArG!jG>X!{Djl9RarB)9LpUg!|0<<05Gyj+^;C4~+H;hzp~njI<nO9A!*e{5*86
z<c~%Xk}?`HwlW4~-iM7dNk$=c5p{b;Elbok%<W-oIOljm6dhDzOypz*<lBx#;zc|^
z=-|J>$LZj?3a{|#;I5%U8ZQAEOo#_oh7@B3(WPUezg5O~;|sI-dA+tl>GTbRWW!yc
zLt>;+ugm?AOQhzcv8IfZtC97~MNj@gUTiWE0c0Q2`^5PPtB9*4VJ>mLx%e&)iMZX%
zW|8N?k!r^-wuQ)m<Gl~65Y7`LR^eCrNQMV#c6?4^7yBCmO_=KB(X^q3n%S+R_A2Kp
zP6yaJy*|oDT6fR~<2<*sq`7;l{lZvPj?KsE^{GC%oHCh53Px%J@W9tj{``J5up_xc
z@QMEE=qWsOrBBajoN``PGHFFJwg6pa*z{twlI9_jV&D&DglR`ny$J04NXw_QqHY-y
ztY<6@Y`^qGfjyA~jfA=~*NXmMlB&5_;g}y>NV!AV3j%9C&sj%t=BySrIgt|+>B|D+
zP7ANvUmC<MF4Dy_lK2lE+L)a+Cw%!?%=G7dlAan4+~3ik<!mFpB3pMCAabDzaJCVk
z5&mF*(`mGuZTYN!G{~0r<N60nbXzpbJKyzUQ#Uf>sNP)jFO#D%vjMX{Gw!q(mW_Hb
zYsssqsT3846U|x8lp3#gukB-mJ-a=EMdbyVMVG2?dVgA{zOManHK6~&V5bJ7abC_=
zqoa1`Cusln8ws4?nCrU@<DW;XB*Xc$$^6WdO52L@7St9@JI|k-o+oV^xZS^P*jF@~
z&raO7?qi9r_;tPtSbufHztGc(Uif6~?e^`DtC{=E`4C^}&q?pw`{Xld?_%%V3%Lz<
zXa9lseaIL{SJ-M44Wa@f`fBOp(uD_3Q*IZI60T+vWZ~7--FcV8h?|5n>B%?^;(5X5
zTb{kg*BwZjD5cw6%<NOa6xWIdPve`;c@sqg`$%gmehkO`*X8(z(rMsyHqV9k`yuoN
zLO2T>Gnd|VtJ=o#y4Rum!v;!|XX9^!`;JwQ!IMv_FNgO#ns2o^Ke1iTPF?Ts<hTp$
z8#m;CdcG{Q>JqhhwzOaDu6eq)T%I3zH(X$xrwPsYD8J7<8r_lH`=1_N9`fAN9;%$|
zgbY9MwPmy9$mIAOeg8;)ZhXyHZuoAucwQ(-Ca9Sc=evF<cDwCtKbouKuhH4)?Ri-n
z(5>?_bnDuQ-i`0k`S@)s^UQv2o$F!vUf3t?<NW4j4ke7l&BxS7;DK&$zF}ImE557h
z?dwO=k?ugt3*-|dBzbH)#Rpc8p(HyyxJM@u*jIWm4iBNi81<}=U1*go?LELNO%)Ht
zSTrn{>*a@F4bJ=8Dum&MhR18rGyV=(L@`ACA@qz&Vc7Aewt`<xG8VZ2tWDJW_k>*V
zcLkL!UjK-f1MCsCwVvs-Zm{!5pJ09;%QeuBLc>f;)?7gWj2`q100R%V0)qlQfrBo5
z&;{CjB!+>(g6`;`OCk^AzjGns@*w~F8N3IS2PUd2E-MSVtC~2Onb|p8+PkFnq7{Ie
zTCh^na?w(d=QFXlWi~RkH#TGTuyy!b1x&z$5A<kj=3+$ZVQXXO%;zCU{!a-$(DUD9
z7IM;minv${l4~g_lZx9rnUQiavof=i3&E3;k_tGPn)7{@kos43&@Vx9OBWXhJ{A^t
zcXwuYc4m7g3l<<RFE0x#8w(p76Q~4}v!|Vlkq48VGsS;3^55-9m^qs`Svj~^+1ru+
zZP&=y-ql5rocwP`|MT}>|EHOU)xUePbN<(`Km%m?o5KQRW@Y)Gwn0?|{-*LNTX~q-
zXiHexg3JTzLx_WeN8q3G|4+`pd;E`@TK}%e&CUIvHUA^$f7euZHggiUw*__SBJ}S!
z^RLSPnfb4Z0xW+={vVe3FEjs>3bL~hya3DpyfYzqinCSF-c$rkRzg(G1N<}_HVaeo
zewc(;R!S5~4+;$mS`tkS)0xj%1JiloCgw0>KKp7S2d&{lbwXRD2Z}t15)BHRN@mIm
z?xxm$RM#srr?sJuy`v>}Ipe(M@#MTB?O44nodU=Pg$OR*gUb3JH$x(bJtVmQmH4+v
z5FkAsRifXI03s}a%n*S4?^}AjAxhx?odo)YDc%zjV(hT=|K@<oPyeqb{(t%ZFv$P6
z&tGHwFaHr~M%8Ninlh2iuW4Mb)g0C6^e@-tLWiKHw4uza#{b7@`v8nubs{h*#Ekvn
zC~0Qnsi7>TF0XrO24mK>oz&KgH41c{)MFgE`$~K57SlO%d7_Z{HKu2}9Uk?=Mm+)G
zY-VH0YEdm5ASYK2LroX<?_w>f)@z%?grsU^iAkx{ti8`Q-7$x&kfpPpLV&X@OJ&q1
z7Y{>Bcz=5tzdf3VK2z6T)SLXyDW~tT1^skXn)@0YKezTn0@PqU4#4MjR(h^L5}{nT
zrN-rGPSS17RkIwlxSTdE>R*qubQi5kQzt?q^N9R(zhD;rcy%7z(P!N#Qz`sBmFCpH
zK=ljG?)CBNhvjTR>+(g{i(R>TW$Lg|_uG9^P!PnjNt~eR{zQhP=g7w}i9aUM9~6*H
zXz?IoATg-rFMU*vjJ9Bc1T=aacZNpiisdMtl4X;qM=!cR&O?Ki8!Y5Hz3z55gHciw
zD5ReSsI0%duCkj?#Bcb$OfFWMAeO5%vM!L8?Lv^?o{~KjXx8dGjgW;uUfB;5xhI3V
z)>!xd@ZHGev|fNp0)+>>43wl$J%;Ybzhu8IR2|VIzddXiG%uSIpj=kgb{&(<DKIf!
z)?K5Akt|u{`f5xGKIpaWf42^S^;2zeE);asNFb}rpm6wO+cKIUO?*Vt=KHXwImLTG
z$a~(pQ{;KGf4%0uA)Co%TV&AfNB0d5i(V~$UP<`N{jvp#D8{rzE_}j8Rl`)G-P+I5
zT>p2v@iZ1CD%}+;7rjDcTqHqc?Az^VmN5nHHOjk_#k6SVpEAttr)(P}n8F6WcPS+N
zvM7q6*O&w}8InlFV4VLu@rgJ=DWlqetexYJ&*xG$k;z@FMg3U0P^~+D*0RBDG8~(N
z#h`)tCA@aouKjE`FBtiHL3wgq3j$JUrR<@zECBuWa3()k^9QRCYqK)SC#b25x_;#A
z8F7>+t{(S2c;A~zZoin9=abrE#caMG4qIZP83U`AOD1uiLXN5ccr0h=9z%Xl(sj=R
zBHpbKjMpQLY8_7`wY{+vv)jX2Hk&0@-kS;bPV=)N0{dGr3D%FHX9a$%;cyZ^T966a
za^tNiLWWS0$XD1XK%TKAfGYB5Is^g8b=Ekq_+t=T7qA1NQ}^+DAe;m2x*lUNo5>S>
z*I_fA>0>`vuhd}Do-<y&XulA`0~&{@Rj8FOJ<aim{K1Jc@_0FG@l62$HhZC-TDsp9
z6osBl%XOM6hKt?KRwTd6v)ZgQmbVBbOX*h2iyJzN=6z{*%Mt$0=UVYFv?(aA#iCi0
z)w0riD&C6F*TC`hmkg-!?b;n|KyPy4;i}U((;M)j&F?kx^JA$A+?n-UQKU`NZ;iP^
zX_TkCMFX>zHMgW@+drS4uJ_(bF)LV7$a~S`AjI%A)b~R{AA;U__lKY7^(c+t`zlHD
zHbIipXBiO+*Q4SzuPjCWGmfVN;SV|_0<K#OrbLRKb+41^>&t$W@mXmSruK`j_B`Hs
z*fg8~rTAZ_gTYvo*JZa-#5LhJiD9oktx{inoU6VIPVB$dXw?^VhwWd0T<~t6jX1&~
zLx<RQmM!>fyJexY+e*5lajV142DEMxkOX?4w+(BKwT3K2kH*v^ND(=+V-W9P$$!9-
z_>h)nIYn6{J_#L{6?S;feQ{sg8)mCYpv5EhxfB)VH5OE-g4%zFCHCmi)x~peUUSnG
zzE|emp>n0rGyHC09E3M)t9p{ZDox_6He(000<LdM<qf1iF6i2Onjr<jm?*v6pY5$!
z*WN9BQAiu(u1Q0EgGH%=Mnyzik<p8p68`Y9H`{v6c3-Ef?YNE_hC_fdA-;2t1>S+|
zm^}22U>x+3kS~65&$j8jTQI!|`5olO&b{h5Ow=_2i6St=E^PT3ZBChw4mf1oYdH0^
z@$sIfDj$B|+l3V90bo5R^n2WTUmp5AQ*R>uLF|;{ZV-6|s0B3M(M#N({aG{^x{frA
zYd1~zU`g@++(k?O?V7)1e9YvEA1@(}@BQt<f86_`^I#Zx2)>HKBPHZXXl(wT5}mwJ
zFuD4uQ7~%<(osQnI!STfiRoBiZytHh>M2ImBFizXC5z2`-fifxG{<X9R}OUmw-Aju
zXwUks`2LJ9-g%dr<y?6p<EQb|Eur&-`$XXROK%GXrm<b(jnyh7hO@bc|J!N9bS#nA
zVLouU4yTwVX?|sJRRYcV-4)2Kf|atv2nIaB*gtW^aGy{}2DaqY&?3RnDCH13o4=%c
zs~FokMYlx??CZf8pBG2*+G<Ta5x%&!!g1_RIMw^V?vMc83r(hyiJi|?fHX6Y_fMbJ
zgjc*klfm#$*OUG6la(gxi8ji?5`|0+H5T8F$L(knMUg08f6)zQ?)pJX3eUCe{B&=y
zCYcv5D8xXvAQ4f~KoXSm&WClRA?ELg*sp_GU|aK->+Bk=`oQ%z*(Lv*%xMfIfvtId
z>U9!+^Of$0SSyxK!?-)IA<?nr8^BTEu+IxH+9~T#0EuoRaVZnyrwgd`f(-+oE6Lf`
z(N;-)Zll4hFkp%8W|nTqm*n#c|BrU1gY+ldvw2584}M(D(S!6jtz04BDCUIJtEIM-
zp{&#&Wkia3S3AeRVL}XNj8c-1IN|rlf)mA=2=!Y1HIjLuYwOMx;^P8Ep$UpKJ*UWz
zNp6$6#D@LH8%6%3VZVKR3USh0R1eT+`^tWfnFbuT9L~y&z98<vxCoCWnYzY%*$O9l
zeHrpT;#hhAsj4xM^&mA-4SP4-;Xm~WkN;`_!(=LS?uP9H<VJ!h7(V_UG~Rc|SkC7h
zz#%G{yn(C}(+M~lU<-CZ#71PF&M;)DQJf0b*>1MGom9Pj_gizs!H#eEg)@A>5-6TH
z9~O9|EES00WssQAwvQfyId)t?o1UJ96}og*<lP)LTsdDz7|QxVc@LV4AFt$}B!`?^
zQb~t!N7IWAvVSk#_-=ApOl9HjMBb+X=RLm<WR0iUOREX<mioHF$ENFks8n@cU7Owr
zd{&%FNNRo+cO1mh^a$grfaHs8lXL(cH0;wTWxsZPynD0A#K*z@wyN^+ocdtc=aA~+
zq3L~co(dre?p=f%%xE*2D%vA~pvItid$Lw!Kge*IyUM$%D+}o1rRK5#M$8i9zH@K*
z%pW*YyMENEl(RoQ!O6|Dds=nAx+Qkmb3=%Xr!u{9o+YujU=?>>_8|x|^8ufcmxcc>
zAw>XEu)~+rRVoujh89t84~yFFcdy*3Hr3}*SsH_DHf+I<*Bc<UG_}hCETOXwdv4BX
zLgZ->0`ui?%c_&ygf@;b^R>!s>-K{9fU;lwbViFW-;iJk%#f^Y?u}$@=lk;surxc4
z$Z-JeT_CBD4Rz{`bidKv526@(t^kuvcp|h!wL~BxKA>PwQiQ&OU>`!1LI!|a@tAKQ
zpDzC^X786$XH2A$@SDpi=%wA04Czc2I?Ot*RlCzx3~Q?yv&u|uET&28;F43@Vk*r1
z)-*Vqrx0lE!Ij*D`eWXW6sRW!<`-f?q96Y71y9A|_w7Y6LeysMjXwE6VXFu&X<RHu
zM{b4bW{00z&*qXgnBc`gJ2VPpL)Foy{bmBse&z$NA8pPLRXmF!cJ&;d3$zWPA>ut>
z<@L`c*V3ZJ!{wzp$5?<E-}nY{>?T+trJHz?J?2*&2()2ZpS7Uv^oE*8u8JNg;o&8K
zy-ZtCv2RS=-?&fZY3lbX0-2tLfNBogD`CC^HPR=fgVdv8%QIabTPvd4mXy}iT?k&%
z4y){DbfN)4<%G$7e2GR1_lBqm@hs=h#0CWM-lV>K*t>>_?kKITaLbv$B6<4A#UI!r
z<~N@tgVJ|c!HvTF&)W`O>0RX41hec3nIY3JR-yJzda1xjOlWw$)4;2<_?|@4yRYi|
zv5u0IgwQ5DL!<_>hCRm&9pk-aqMpUefUC%tC@w6iEW-*p>jIK|f8!2X0zjEAk_ooT
zw}K(&iIMbkGHkGGqlsXdi$>%1m{@<-bDmyT0H0rd9ZsL?s8xT6rRY@9@pRkh_zNoR
z7f6TMUwkql8$eMk+z$s~hcVpTG~fsj)9=+N#`ucUPVf=T`IP!gu3ciZhnRsXk-q~1
zX%MW+chc2B2do19A?}Fg%~ZHqi|0Yyabx?{7*gy<73>2%cgOJ{Tmraq{8I{B3q99~
z$l>rjoXye4s1HtO7>}>Zt)I70s?PPg3|}k&yWkMV--or>9k(UOKA{Pqo?vQmk2avx
zkKnme=v{b}lCG=z?BX>9F|&5$O4z@GmW}~BCSM&ra6EA>??FDwg_P^}ZTo5Op;k#$
zm~}x9yx4LEG{Nf;3L&KP_!#ky{zjT5>}&R+dYDT0i}jq+omJ54%NV168+p|W2eDm+
z<~ND!gD*<yARMv@na$a=*bD5Ae3cl{TxG)3BzqK|a+deAoEChzJuhOE3Cr3ZSOkxQ
zML9}l9n_21RLXXh$TIFbphB_BYb~~O(WKCHpMrw}8~W--`)c@{bn*^P9fE?B2TK#-
zD!c7>B*sULuJLe#e#OczD$W?Q-#Z5f*{xsNdxL@3)g?bBZ>hL)7wt3BU(`{3gD|wg
z#Z&K$ME=1(EwJOy&U<~G;CGPij(umcPiAtho*y(YHL*v+=5uLDf`V>K@kPLBMAP^K
zw3Ra~BoAO=8K5llD^o#o_|EFDgL87cG`Iw=I*u#uAz)F!f*;&SIF|?`*4u*{B+{1-
z4NP^cGD#E0dmdg2^=CARml7g_!{ACH--&wOMkeH$i4&dj+M<Yxk3A_;P?zcXz?KSu
z!`lZlLzIv}DJ8ny1A0~}hjw$k<$av?g55yWr-BWUK%+sR%R>&MCR`d~CTm%2F`^L5
z`Kji_UdXSnLRw^8?HsOqUz`oQ9rr0Mji8QjLv6Z}V!xA9@>FIBn**6HV5fzmKI0lL
zP(v+Ju+|*4%ThzR9<1zOAp#Jp5Q0#9wiHmMP)SKGvM)a&%v}jJAj2DXh>?l<^1VMH
z<om}+qY5N2A0r_E^en0|Ay5L*vb<aEDQxpuE+eKIf=&H=VRjgNa0EyYeGjIxecSOY
zaR2a7lR%_6lU*kq`j|hUY(;Ug`;n50#@NP`16WYSpmzXg#f+H1%P77Q^v?@zNiR(R
zfZ$Z<Dgs>0ph`y}OC#pTUQv8r5)(Hs@lk*OF2th>ilUl6vCKAv_>{n|$-;GyL&L4n
z0aZ}Bg(Jz;WOm!CoR1dj5^f}08H-jaG3+$-iS*ZS4;2p{&%ERW5%kcyDAIhGP$*);
zMjQD?pNMDhK!t*|ahf0gpbc={EGj8I;DUYt3ism86g@CK-F162y*?N&{YnX^S@Ii=
z<sBJz2VLQG>p;lBBRb<e4dPp&JT()MW-|DuKn&M}a6W9tk~Yso$P_P@z@er?3_KQz
z=Lipy`1S&xhtW>Q&D70N7Jqzhgsc4TiAKqn{2Q<{Xo3(x%?*+wYK+K?u~;r@p7J`Y
z(G3{@CS)@>G(B_-wUtZ@Sgu-=Ko)NguvCU<e7IE9nkR*_8Y0`i6ZeqLE0T;iXB49=
z-$_u~e9!L)rQ64suZBVH3F_eiFRTvo4J3%8&3%Y}$Z*G{CK8(+ZO@ij24hFc_q13I
zv5NbobW3q8>M1YS@q!9Zumb6fGE4j{P9ih>a=sAyY@LksQ4Em+uOLxV^rpo|HWf_v
zN-@wX)g9p?`=({ogW-oxSeFVjj78M2`J#+k7+^$*G7Z~x`R$VBAOx!C7eP!p?Yil}
zSDy&jj8R^7W41Y2*f4@K%orXL4pj*Vhv?l7?I@g{YQgx*nxA9BTJrR*;D!)8TXNxz
ziqtwtz>=rnv(j$SeK8VKg>fLJeBt8hAJAbR7ZU_6A!(z$iV4mb#KbkK3Ce_x14Q7}
zZ}(Li_%{ls6_0eCA$n?oDrCz6)|dX*(gKjc#h~*|vB$E7RF5*_R5>2i7jDe-G=D9|
zSAO4e-GL-vW@)RS2Q*CE&Get}s{_cu$P=h>pcuF_)Ibr*g!G3D!2oti*i2>2+5nBa
zHNh(nieLHmK0||lm`E&HKp(~+B{ny%Udw&YjbaQ%44KU(G@2i7ixpri<0{gfLB2zi
zPZ7G>zN6MB=oR_G_fqiJMiIgd|CU!BBI~_KD~6GvWVF&csg-x5jcsPLbjnyv?bf}e
zPM1~OPdju`WKoz(_(vnOfF6J-cqJlkAOkL&q9d`U;t;ifbe)~yV9!{A?E)~yjZ}@5
zx&jsJbbjbL+gS{v=h9F_rol1-Q_p|yZAWUm6y-><M1YWw7v?5j^G#n?>1U8@5;8t-
zqhkp6uQ*+(LmEjtLKGzK4}*{Av^3*LZ)C@q64ntYC>q*a+FO(xo{;3&0ix5SPi>Fz
zG$!a`gDd%<v#?||C|q6EesR{kkW+-N19_e?%l^$mM5Dx+NTw7!W#W7RBNsFohkPSp
zJZc6^qE_TGGt!jn?P8-J!RcU*7ovYkcFGhhuC8IG3o}eQ1|({sRC}eqqH3tmky5g%
zh}i`eee9Dg<#hx)U<~$E-WDLZk7>oth^~`16j%jqB)g_MO9mM7j6|Pc_)0{VCbcLT
zejo-4hy_>>faCQ|B9~yT!%1P0f#G}wRFb_vqNYkP!f1idQ!!$qUa;+iT}!i0k6rZW
zs`&r73Tc9}T+wEUYEP^&1H@L3u9Px=-@1=i5|pz<dBtIliID?<)3a__K^Tnj^zuOp
zDz)Q!_e5Sn7jyQ3dn)lAQ3@;Wc!iGtq)+Y5yaNEtwMY~1gta$5c?r^lLQw*V<*Df>
z)?|uVcLglLCPm_4A_;<@@V5d8ip*JX(E3X$1PKTn`;%o#(aFpyi4CdY><CGY_d<>P
zwgQcvcBntQP3H){xu*M2<BlyxA_=oo^G!(#b_gIsEAr&C$h#fS%MnvE0@w*{Ww@5D
z4F364#*iUmm<Dz=6)>Y`XGDNxbspa$F%)q3tzxf|GzX>lIz<rTHdyNW(LS+iHk9Fh
zBft<iqlE+d)oUP-8$#B=F_R~)y8W?z*W>#zlv1u?Z6lx9<d0d_$KfaEAp6C-ABZqe
zAa?)BK}Gac3qA=9sL#Icj5Obs8o0rZuR{H3>MPA-G65qTAxpg9qxVgK@SY_Ojpwwk
z-{(R(M2vNcmd=A@Qwg`yp=p6LlXa{^Q&b5y7Z~b)EcR^>g80*y1dkRKc+xy&?pH-N
z>Wqi)hD@l|bb6Fz-TERa*mNAFGP42dU@dl!tR9?t>B{>(5B1bB_dOp#H}OWCAa|22
zkr~j+d>H^li0MHk3v+}P1^6WZF~sUceJeKey0viAQR)n-2^xIJiXSgn!Y4mh<#9y8
z7*4EluQ~*`;2P#w%mw93MVV{|dd#F-szVOINLBF+Pa{}y(!#;~iR{MxO!jUpk5A~_
z5PyHkUa0T0we(C74UK}R^V9F``3W|pnYG@rVRlutm9#!6CtMPT9QK5g2?7CwH#)&F
zP0ZM=o=psig-)nJi&b6*KyCpF=vPGpM#G@nVu&Pe1FbYC2~#0ozo1dh@Hke%!+GIe
zbS~b=Q^y4Ul)>A_MW9MTF~PxL>x11vtx4|n>Ak+|D5ywk^ixHF0GQC~@Pj<XMK8{9
zHMq#czExg9oCNQoM<N5I2AK0>hs`7}cB@LYrx>H#lCdLZQp=#uv}Z|~+nVZ8#FDSR
z2(uNAP(Zae(J>R49>E}gzB?F{iF6WepZvxSHAV{&!!P=|1lLw*U<sTB!l+n+XGF~^
zGMZe49G2(9=aDnr7z&k~SQNh;v5N+c4iZUx5?929atJ2fIeB1I=B9tg2r1w~><A1y
z2InU;T(z(XT?fqp8cbb4L9%()W^e&yjJ^xD{1m%eo?d{vM0`FX6zPILYe^Zal&%WQ
zkyhkfsKRd^SbI6V%Ws)cpz|^?c{|EP=2@0H+QUABfDqYXgBaNONc56+-4?r?dKw8V
zA;a_+yvOD2IN78o=&vi=9nn<G%q7ha3T6kNsqAQzus>59-NfU1(x%jjp$eshM@FJC
zkp)3x5SUX@TKeqY25xT6!eNqu(P4r|T<ZCdQhGc`kAtBg7Cqf%ItyI#WgJCMdj4zo
zaM{(D*N#LuIJk1}cl)@;o_{sPM%**A4Ldym2~^n@8{jE53B3<<&v6V29qo8&Ju*Bl
zv7u(8rahMM=z&J}*A9WECTlb=l608$QjTGD1Ff%q6sCEG`QLtx@{I6ASwmouzrO;|
zLU1L5OV3P)l{u1QJn2m>|J)H+!<rD7`wj`?q8#Uc`PIKGuTQNpWVSMCI)1e~nk3jS
zX3&e&3;kitT8IQ-F_hI=`9>N{70%^ENIjL07NTWPlH}9T9Dehso-Do+(9_UkO$uO@
zx|%xlmdG^>P=^bENPS3t5Z|EiJ;3XKyqt1S*dQ80TeqIOs`>E_PDXq}XM!B!3SkB&
z26<sCAV<F^662XHCwQp@NjM$8jh%zQEnZC!DCKW0)YT-|2)9poLgBfELiX&RdL!R^
z7k-UPP=-K^5z6!lzRvKKx;o5}nk7HIH$vdoH&=|Uh_&^U9?TP(H;r|LVc#(|7CeP+
zx~_5FdB-a(;>-dpyoJbUG*YKR2bKZ`*2`5!?DqV#aNh8lRmTaWp*vUx2o~r$k|Z=U
z2XWWWR46563xN34jSdb0+~ha}^Z}?iE+nh;AGAquNo?iOTw7it%TX5)G+zn?ceVyz
zB}hgJF*t|i%R&|l6n>ST8TaiN70S!Iqz{f%qL7`Du}pv+SjKg=>al}F%fKIjU>CeG
zHBu9s>A*m^AosM<ge%BxUyH1*?1OA%#R<J(&IHFOS!e~^F@zynAA{Kl;1elgk4zHw
zAdH4OwUZt-9tAFUc^ub~mJGiOA_d7#6x-x(ub60OjBjH?vvYRyb#)SvE34met4wBa
zNMC4u!aLB<D7N(D*R(Vq#I-Wb86@|R*w`<}SgwL^MSdaVLuxg%$j|VOIj_m$xxr>9
zu2ZVW5Xjc(S~u~La-p>|?|2ZrAz&Be7YspM(*m#3%F&bjQoAhcB<q!=-4fhe!!81w
zf^!crTe<G3j?6sBmqKh|J*6rr7oFGqgeSPTVewZIYss{6^-A;C=U-o}C~U<Nz%O7n
zr1*w#=Nxl{XoaOeMk{C2XmPrtJ6@6D@mJi2n1}E^fJ1m=MM{kc9foB)C{}2(k0rUu
zvM*0WyT5zV>shtzaD%_OX3dL1kOFz3g$2M?WjdE1C)ntW!^BJm`$fEnH6j{!w+J_U
zf(AUt;Pp|2XtNeMblh7J#|jaJc7MNq*Nh9k`|33XeG@V-(Uj4lauDo!)&dGyY`ma6
zG+ha0IQ*dI2M7#!rUxYS{2QH7lQ0jA-pL9Gjm!@B87t-cw(80C1tX_aVFq|4zg&&Q
z;2-!5xqi&`lIxYGiu3oIc;m91$*Flq@P+OLk9=-v<HXkTb8H8s<=-v0{0-O-UqrhL
z*3mkr_|fk45APiGr^L4ezUvPoU5Cy;#^n2dXQ?3Sy&}{9>2XjV_S(zV=L^S0m998V
z!@%z$sP9k5U16^?JoG%}Nk)dIyb7R{D&hi@)<O6_i?P>a{g2YUiw_awsim?^hKWNq
zB;06a*~hv;fxB55N1&waoMQdPSbe|kBtpqw)!C{~>Of|Rhxb<+uj3_;UlbF&3}f>w
zA<u@nS+WNme6T_}A%W|GJDBYTLxWH9jiFVQJwDv1a0lp}=vuYlRB0Uqw9^DZAdoAh
z)7$>4_W65OQVFo(5);$z1YcnOr-dKiz*1Fx%<Cz89*yUQK8QE8R8uLX2*?mtF>43P
z&niH+$Ax%`4uFGu^0u$G_-6?J7VAZ%0K29DwKKYX7qRc?I9U<cQeh&@WFoD4eWgdp
z+ikzxtgVj$dB17U@kk0Z@`}0sK?F$Ep|n&b8cBul;efw8Qm0U9;UIXeWJw!I2rgNV
zjM2kp$zxJL`zHoK4=wp7de}G7k1Xe@gug;kbDfZ;onesM_Wlh>Pr*8K4Kg$|{Ywvs
zaG{c-f_r<Q&S^csj!Ly@gpr|!iYUUtX(8AAJf`n9ng)pwNA9Z<BCKpsPFUdNvjT6o
z^Vffcj6fXGK-3tiws0_l43Gxst1n{Mz)ZDmh#^kEM7nXs1(>_NxBZ7uAZK;h6zPU+
z(;<IC)}e^(8KgE`Yj<B{1ZA5d1!ke4qoO*DllqVl6r$x!fj0CwP)jypUz+L@I}e*(
zFJ=p+yIhzqUIO#nL5C@ib%idemOK@EJDHO%G5@H~BiImH^mqN#;mV6OU-tfLZv0Ue
zmBb9gs3Uk}TLTzMx95@|^|X48Q=HTut@UX?b8nO;_m*F%z>|yq55RD?V=q<Lhho_*
zUG_POq2{ukFzqZ=_XN~7DG92_Mdw3D;x<UTyIkYlPdC{GonJDqNb-<8Tly}!0Cq#-
z1KhBjS2lZtCxhyaa<D2rDadBpoGs{RY2!a$x<CHNaO!YokUD_$Q140<&)Uv=9Tw6$
zqr!xLJcNKC@F2%q@%^sW|7jBQ>Zgb@NSW*(vJ3m220GsgiLaaF`ZF?!W5CG0=AxR@
zYri&%1GVsT@;S9epP6A1^HWoldXd!NS3LJKkVJNaF#mWC5|oGbw1(zH4abt{oQCi$
zMiveH&$w5;_EI!E%DY{@Hhf0<OPk0b1s2Ptjb&Km&|P%g#BO{c65VRuiIrn&C}x8q
zN2R?HgRn;X&S||*(Yr*x7r}2?%Eb%8o;sPZp7cCwkKmuCZoNt^k3}z+`CPplQR4k+
z=NO6Y>jo%>rqML;ZPN2PD#qJukuMJJDUgir0OR($wOsc9c<XFNrV9J7v714IGiIU)
zTn5-rfvkJDD3g-Xtj+I)5Xjc81YtzZIbmfdYF#~P0ktr&FqXA<vlmy~uop-|oR$(A
zZFaW8DfuOb!@4o%yP(wj@|BpCR<-`@lbU~BIK*_9%CZ<36A?t;)iB9LVO2`?f^Ao|
z5evLqV)I%N_l8A}R~K>M%SE@rcP?9c|BsiAPP4$KEZdbv?c5$eJP0~+WuZZrOh!C{
z;tXh`q+WZhVbRS(;jV<}7d;B47!MU=W`zkLov2Q`Ta$bSr&QPTP3h)Qvj5G*%_%_&
z5fq^N_5J0-Ti9UYu}u;9u)9spZNEpZ-Rvbf<!15z`sCNSl1L@z)Qqy=AWr%RfGQG*
zasfN`%}{IKX3=bKj|IYPJ+Vh|tXW0pT_~V!r9sK>WzLF;VWeyF=h$irg4tM~u<1(2
z)eo<`=f}re+sCZjy@CCg(lrIQr7UiH{go$1@gve6A?#Bl6_8qNge1o0CTNz_1k!o5
z!jOrMDO&Q8EJ#A{&;EEnfMM6BVNdD^DS0WKvtaBiCH}Vla7wV&?QSAmC~DxS(C==n
znEcw-U_O~huU_#O`u*+YAq^z7%S}`xkyBDK8=9++(64G}A(LN@P{fT4L}I-01fS4H
zXErF%;$mYnC&2HG$yJZ(%BUaMStknvl0g%^e_nIWE(XN7CR0N5H}|BiscOriKDxOo
zeS~F;c1WUmPHl5WhnU3*v-Q3@WA*c|mN!R!DUB?r!C+`s)gZ;1ib=7i{hY5mave>Y
zknH8OA#Ot??hOl-P^-=l#6XO|(PV8xW8iMt%;;z{+-NW+AITkz32q4gn3A+}Rhh-G
zh)v8jk4yDG_v*4q35?cL5hd7?HZsZ=c)jfZi&DtEJ>Qv~ueSa0xH86LH;?(Nh)RDS
z!?q*H=hp}&Q~nANxL-24p5)$m^wo<%Z+kg!w<xZGo*Hc4h#}dNwg4J*7$+pVi;1m$
zbuHS_u-&_){*pDLOvZ4hMV5&Tj)Bg_^seC@OFr}C9r?#XwOso40@1+gLx`~S$S!ZU
zOl8>F-s2i`qS%4m?ORpC51-%HglGAjx@aE9<)yo&xqjCODM}z=Uod-?8{d94wGO)u
zjsVdv{jX@;T}YpKLKbh>YpB|sI7j2}L&Ji2Le=z@!vW${=YYJ$4++U^hn#yE4jwW)
zT9106HnEu{`xOAiZZUN$%2dbEP<dA*oAg?0KKb2GyxnE2p)pLPi+rZT)46RcjGzca
z>i7wVf~NWwi7$ijyi?j!FG?SxQi?C@bORc#S5T$833PT}W)6>p$3}R&$AsN^+a>Ob
z5<TyZIoXUSJUl<<-`T9L)s~yBOn6`OI*U!hn?Dyj6{|(namb>wl99oQxGkebh~;@!
zm$2C$$+M6;5Q!S4U;u{XiB5&>;EG-B9P@sat1!^FXOE*nEmUdm&}e)gm$ITF>qAE$
z%OKhT2~4*klIj&|vc&$cjz)z=AeKkZd^(5ebX*zz`LLyH9lBJv!FQo;M8T2S34_Za
zfnXnM;|DY4uWH@Ji;Q6Y!RnTEdL(>~OJ*r1-R5M_t5u?wPv2aN`xKP6#|d=}9e7gP
zy{;>YrzG@tt7_jiQDjfC=Hd8f(Vd^H<HxTu9qENuOFw#2W#V(c$}BA3i0FInpN1q>
znk|ylzbPjhjQ2hr@f+6pyzjZQ?Zq8^Zarp+K0Ix^F*{T;G5zqpkC&Sk)c&eAEr97c
zCuKtoH}=}MH}@08M(qx?tPL>v-5vj?W4e{qP^^QX_QFNtFjQ$YmfWwQd#v3lxID=9
zm%_@P<9)4}b=v-nRsvcQbV0)8%jJ=R+k~OCH#b&OO=mTud$bT9BgJTjKEfK<aoFU@
zxEmWc>Ly+$2m^`3bm{?zHC?sj*wBLUx;su$69H)SotM{kJy-a%Ts5)4F)`@T^K+da
z>O~EW#0z(&RA5zp(mWtvEE2UYKlEd$My#%!y)ip#_j$Tr_o}m5X6tTXtl!k`@6{o$
zStg(S;nGsYWCjji1vdi%!BQQa*K^sF7izSbzMXKSndrVLtZ*Ie;$9uHZU|4jQ|)*q
zn!TD|vU;AUxW7)_^Peq^X(L5V_Z#@#b^5#x_LvSX-C?4Rc#>_E6~<m;Sk~3rIQ;@)
zT~8pSD5!KVk|rk}1n@jLTi&4qGhm5a;xA>TM#tdx!vig(2ae`TVU=Ds{!&J2l^O~#
z$P=f%QkNoyLtW?@Gazt-k|x_F5y!9~BoM?pXuAA)Fx|sYAQO;n7>N(fSSCMTlVj;j
zuHK$wE;Zjf7E(H6!k5~wR_V;CimXnc+PA}uRI16^?F?{H5Gy&8dJIbt1WhmV%*LPq
z;9BJqZ6W2azOyxz^)^U(w-ZZ^)+V?e3&b>g>btaOyTtAv^JY#erE<mn&0pSxNT|v!
zO!n45a7QPC<X{vQ^f<%<jt~f;s0j65)#C0w9>s%F6u6Su&>)&;vwB=LyD_cw`Xk%V
zgg>flvXSCBzW>R(1;(glf2~L~JIHt}Y|ola(~H{rPrQ+^+>0Ck$ig@LR)w94aX4ub
z#=GNR%&BMDC@;J@{mCwaBBPt62cb{}T6M-GEnrmSVVf7x&Ac*N$oQOJF$J(59@-9a
z7(kSO?CQyK*Gz>vL)+W^AG2^Gw-L8lx|HEQ>Xm@ht*3)rB|4=J2?;zSm3;JBtV=v%
zHq(flv8RjQQ@P%W?49Rprqjld6hxKE`H82$U05=jEh}_012}tWHPH+i13&$s&I7?^
zY?d>m7v{-y$T3LWAMfoQpq?_Y$tJD29NPmm5!yzWlF+*SDNuYK0t5tiW!*a~Y%ruR
z71V%H&^|f#ZF|_-T|R9XVhikmjtBR3PYad%L@g>h94aUhy%<#F+jRSaEnsZ@4dbwl
zpC*0?GzD5pQC#F0)A*uP3@+Kn`$M<f>cHhqqFvx9<NZ-;8l&%}HemfB@t#7v^GLt;
zOMA1yVK)dBdjiRfbCc)u_FtSCc)RwtuIC%ces~-)W(Zki#e5f%r=GxL`S#V03ATk#
zOmVfXy>Z$6p}>*4Fe=eJYLuD3CDjiS7C|Fa=&d_xd|LQ-we%tV<adr#9Zu_!Yb*wC
zWesYk1$3JLk*Qi)(w~E%;Kcxs*Qx0A$2fCWu<!Hj)y65u<8G4D^M*XO4XxM;>0}1G
z^vw2Rx$3Yqo+tyf>)}lA%OdnVdW#>3V4Q1pG+73@cc->)?J$FMt+X#jLNHRp3?{mc
z<6q%&qlk#}Ych8+5KgKojzdB$@QGC67lh0l)DbAC#Dcj7dBF$>vD3iR!9HCgg)d;P
zVjXo$kOUN-wVI(Zfg2byLK`9`iF`>8RW9iRKZ^V$=2b<#yN5r4g5h28#!Bmbx!Jq^
z8A4G7v)2E6s?asr#4fS->DJY6YY@bD7p>Zca#|$!f<9aJgkiR|EnK`&)aA5QgELb0
z)ICO6YPMYPNnzRlBzDf)#y;b$Pmaw2e~%jCFJfoGhi8II!bS!4O;wS%cz}#o=`Uag
zvdgQKT}ET^xA|KIJ&r+y`bTtok*P7|))M6kSiLwd84O1V1S^oLR}4Z_-`DyoMvY~m
zf-~)Yj>cgCP(XA>!}K?G^srEO)-Ph4@C^RwDZM<Ph_{96Q{p}6v!p+&1RxY4DIzfn
zdj1pDQDJKR*WdIkyG3lx);S7KlUg+hrgC%0`;nU<H`4xEFYO%bRKY>_$_RDgE&HN^
ze<rou+Zx8X1p*Nj+M{eMxRa?ln_VUz4Engep2cwXBBp}|=l58c#flE}7u$IqjaT+c
zzi&}m@U~vnNUwQ6q`CK1@!srC!RB5eq&+M(RogU~30nOw=$YI(S))~D`ZI?m4f>ix
zKq%%DvG3Dfn!ZdAH;Fm%he1J53<39)^?{Lo{$G5|{|4nD<E;756v+<zMBep&rv0<Y
z>x+@GBTbMN2MBv&@ygZIvj9P$S>)fm3!HrRBm6Df0KO6CF+4|)f;;OHf6&};eP(+4
zAH}xrds#_*E_>ev7Be}l`ZV>##)rYA{AU|QZjl6V<Ad>4^%y`b+mBW=LD0U^X;u<v
zbc+At)Z}U24eAF7HC+t1EvnX{=t*3fX!R65#(>iY=!E!Fenph$@55QhEaRDC4C@OG
zxK59tBiYz7q-~LSE40(&^dNRgV(aoWH`HjYT{TE2BrocT2~-#8`({}bZK^g12h~*Z
z`)(n~S%>IoHG1J$Yj!?uGJ(n3h6-WTZo;2AXKQ<=HZ^L``CUC4R(d^0S~(mQUrvm*
zITBlN=lk$u2m!em=LMP_yOEgVRSzUeh*GJKo;4_>s>Vk$>uPWcE&qw&J=W04p@7r2
z)gS&+pp`r5J^3<*i%w|W?`q~X@i<kms|F{DXhN_w+iY{1-NU2Ms-sTHI+R~Gs>V-H
z%c9lgeS26R_=8^Gi@*B{4Iqp_&<%a`yNvl4EdqP~TFa_R*wni=eft|K{MtG|MXlZ#
zOS{>;7$p+lRgmByERK*N8k>bl12bwsR<-0A1qVbiV=f!L1Q}bMlKMi8mF5=P?1}V;
z=LNVRqL~l|S%nxv<%2`Qc-V64I8mkRrfXLrN)Yx=`(E#kdbJ&9_L4{?T2jdGGLouo
zg-DdA7D5+z@_|T+&VYbsA~_LEGS&g>4bDZqohH{)wx}nE(rmXV`o7#k<aGTpu}hJ8
zh+q_x7#PG!LV00qq&gW&Qe{#xM0r6-B#(SapRG_F0mwLPG#KIrcC&gI@eF{&E+MnJ
z1eK_WNW@P6GsH@%M!2CxYd8P^!kT_g$B6{xl_XdB>q(3B^o6*=_9ED*ljffRXp6*U
z3>jV|)D*|Yt(Z`f6esT!L8$o13x;r-byufS&BwXl$J?b6%UBf^dvPawTh98k)^Pgn
zt92?MwxhGeIvmwlPAs_rbXFjV9T+fErOj>>cwHDLbc_W}i=wO?0yoJ&VOi3+0#~1O
zppaZMg~i;eoN+ytR!Qs4^+(m5C*Q5Ofx76E7P#w#x*SmqkBKasb<si~7D^bdIWdDN
z7u*9yGA6AUZaAu8-+;%&Gv^RB)!h5*&9v9yu-2S|L94-SoF5wu8%dsr(d#4Vpdsq|
z5;72GOLn-&7zYy82oYDa$#0=NAM<lTc4IM!Nm;Mfn7*038Sof2cE6AA!Ostw8DAP+
z8r75L(Z^+&%hP@bapXS2oPry-X)kglxa|EB{%xr-M@1;ACGvD*Q}%j#DXlDk`HG%x
z0eFt*QIBa~L=0bW$bvhlo6vd3EI@<dvkMSj2%*=h!0(1J;HiX?Xxzm{u7!!|D_3fu
zDy-~q$(Pl16{Fx&i973^APS#tY#vSK?)(CiU_`u}rOKNcI0~HKwYt*xr?+VqvSI34
z&8~o5=!YOUB$)W#oHSG`JSi-sJa`n=GEm6nQco~l&%CcPPk`u>?r;=nSTReQpTM^h
z!yc7btP^YAa~A*}D@!#ifnTc!<H`}odf$q?MV=(nKVljP^A4iSuN8VE6tyHnGUZ_q
zu>HAhSAHq^Y<`Q!FkNZNjqJjE)o$<E?WHUfnmE44oIrSQG%)$)+^ua|4RR(x1S+%9
zuh2c1@n}VxfdXlfUQQ$&h;w^Pv}4=ZisI|t>CU}<@Rsq#Za?Z9Y{eAq8~-F;1nTx_
z%#TT6E?M_oQaG$2Wn9dYa5se+tD2tLlUt1qS>A<I#{t>z1@eJmXH8>)tv2zUZvip<
zXozH)L80Y8qBnbC<#ebyRx&<tc*r)#M-9fH+jD5Vn*^3#klh&%sWn~OK1)p#Gq7i)
z*7G&z2cXvqg*>#M{~R|kzu{)gL%k@|F?x&S`B3hv1Iy^7#hw0fh9A-n2Tzf%d<=kx
zSfO0^)EiS46Dcwi>hUUvnpeOdT7TH^@0O93u$0{=#<hHoVG}!Aw#ZEZk@Nb@t(y?`
zy=se^oi!UOxHH&I<^{@g@oh12$&sHS5nw*CT#JuFDV`%aqtxW@TKE;&0tB>_GVF2x
zs4MUAQXpxeVitndZ6^>H|Mc@w+JB|dYG=IXbMT!Y@fQ#_=?5v=8B2hls+;2t*e4vv
zfTy5Z8Dy7qKu<R8kp8r)s;AbQ>j_Sn4qHHrX1;ZMWQ^&9-Dd_3g6>=G1%fjZH?bZM
zms0I?mq=}!GY?YRM*OO?S(Swh$yg3*nwEYB|0>h^Lfy6rNEx%Il_=6R=(yG8wjssI
zaSb9s;l`Ax&?slb$@VSyRo=P&#-h*bCd3QjNk&-)hh&a!6cC6+o(Nz8>L9$H<Lw4Z
zlnE49sDE64iJsfcr!yuGZi_8Y8=vh+hE%PFEXqOgA7;Qw;vb+E`xH5GhzNKDz>cd5
zFP6!wj>kM8&3{83+6^oQJM;MoRbnY=MES_tLs2Kmv??VR0z<jv_l5ueP<56;adhFf
z4ls}*xVtB~4DPPMA-GHM;O_34;O-LK-3b!h31pDq?iwVw`OZCc>;8nInyH@N-M!zn
z*0aRAe*P_`uxMiL?r=HSq@lZ92PR{F^~Rf`_o9WkCF(YsW2tj~Npg?tl)gh<hKyV;
zJht~98}~}v-(I-Kr2Tyu7?w(^i&0Wdhl%iD7FH^5ndVl0lG&^7@(R7&Z~3aR7SvOW
zh^d&ut*C`W<(+w#+nVG**8--4Op#dnCa80(EVF|~_0dZ12hB!)?L(fWHm`-F4&@+D
zv!Q|mMP;o-)fp`g&Lobr<<}SAq)V#;mnx48mIeFsAcB*%u8v_1{gI#t0%>Cz&O;<Z
zpCOA~<RLmR!&~{rc6=+p-OPrp%h0yaI&4k(57>%HeZEvIre@fZ#sS_LQuc*<jJ@ci
zBuRwCb6jaY_zp>o-@#ds3NTL6&k|o+Y30RfLu@Xsh?#xyNMM#fMp;DutFJShiMBhS
zSD4b9Lb;wA<p}<YttdutR3OKua#mJr-2=@*Nj+7gI$uxjGr`i+-)u0u0@jn_+d`vm
zkDfa64bNWofZDe2krxqrAFN|OE;i2~`8nU&H^uzj^8UUFjcPx0N@)b#am@<+V+S4A
z3ARwe&io$#Edgu|&%58Bcg(y<{KC7??K&^@ELiu+t8{QjP?<92mIV1nnU?r!A_Vz9
z4>m#OZg)O)M?FjvG$KQ)Is=f;I};NA{;<9mqYjb>+Y0AP&djTQSN)Jk8MR}qYUCgc
z43PwbNtDL47(Z>~8FYEQ(C%lB<&nf2`J^LHqaVC4B_ygYmaAmy{SXwG6B#MW^F{G(
z1%y?UHvd7CVf4q>nAR3~1LQCC;}kpzWU`CG)$A}AK9)H~*;=L!@g^oxb3<OZjhdnv
zyNh_8kiVSTo89abmtA~^D9g=KRLgA!WAah80z?EbI76N<vSOC#bviiM*k6>9Gpy3&
zIfg6iX7<=enGo&-8cWlaJ@+o%+M@;7^325vPTk)Ab1)(E^zYxRriI#P!L9P8w#H5H
z$?Y+@f4e+|4w2&<G*(n%i0K-rRw$@t8q;bW1tO@!_c-u@I2W0W)kP3tCu4ZE5U0(z
zW((xKdF(q)hp=4X%${+!KHPCuPJhIK8xmd_^5VfJLEoYv^v4E{u>TgB(=Jg3mL`TC
zng2%yS0#x=K_PLFqezs>^vam$Vfgg;kv-*`S+YZO<G-PJ2K#d%|7QFvC1@|x&MevQ
zh$)cd-Y`ox3YI;o9Px;9;rFOdOVw_q-65!s@tC%hT;M@_)F5+>f3s1gGH7PF`%(M2
z(vGg56_dIfYbdbU{|&I0)kBi`w4Lt;=g+b;(K>$c4^wiRH&!pm-&)odPEFoOGNj5X
z;Hi4H>5d4PJ`igk<+%4Vry{6VDvP@PF3vulZ%!UJto=oYG3I#$vlyBuR>yI;mNhz#
zL=h+&6{tjo$9gY<JdI5K1$8L82HCuw8=ncIL@mt}L6n0C+4{hGqR{|>6Ty>%2l@UW
zjPtV!RWNgjEr=5TGVd{Vv2Z=Df*>y)1Pp4dh4ZJ$<j3W+4lu`sGH5u=S|uT#^+TeO
zi0Oz36cLD`OM255epKWRJ->W!U(MLKWYj3spDPhF48{gjLp%w4OhOpa-pvNV08d4&
zq*nJO1MezsPj-wyXk(Oii<*$OK=^7zXeLwhFR=>dzd3b2c0;g1xSOVPEPq!+K=7a0
zABIrinK(_A(;Sw>MFKp<qTdDM#;ZtWR!#Y|W_b*^wlX>L=e<m|b^Bgl>3$V@=<96#
ziIOobE!85VGx_Vq;k)*CzGD%{M-6||K~v=ZsNuJwLX3H-jbixDji(UF`vhqWQ_vQb
zj&<V-7@;*}^Q9P9*Y#uUD48oNeAG93i|K;5Q_h0w$ZD37TAw@BOpYza<s~<e*_6u9
zRr4iltauzDrj?q3A%EvrDi*c-r|GSDy@;R86{UVj{W8gtBoF(!m(w85rNMC?FAYjl
zX7FblbF7W|oCZI4Xg?UZ^%>tPM`B%I9c(*3XC^8mGruUYWOR$>Zi7MJ`ey@euelWp
z&*QP!wpQ&>Ev>AMXV8|?Aj4I?Dp9UhM#(?F?3O(!*s(QC-_mgzlkV*EYp$+GF-6^O
zRsKuWC%0B>ODg;po33o)T6;XJR(0R+TE<<&G6gToO_rcbF_}CY`5$0J`5%8NFY62S
zMRfc7d&ybz;G1yyAp7p;d^&Fs>W`{BCfRqvq{zq05R%~x!yew!qP?{D)19>*F`{@V
za4YX+Xc9i}n7lJ*8?+6(MVHPNP4O`{2^jL~3(~_W>J1J3CEKX*yL7LA-HU|mbEEM6
z4f{D&83->WKhokEAqb9toQ+@u?h(ktJ)jZt`REexM6Tsacr7>`@OED$7W>~12Jzph
zlw`;y-o!f@dhX!#oWVznQ>;|=6P&6ZYEF2b%B3yxxW|4$x092~#DneJ66g^)Dz8!p
z86A`3tK$Fif4Rzi;dvt6Y7SB*AJreSh`1kaxIG#FIthClhP1>_MJahr_XWxD>9QL+
zgHKD*#yN<%cEV1z_$jQrXQpkAKQju4r3*8<H0;lyFM=89zTCXWgZ^TRbrEg|ge{_J
z;8}b}VL>cD6Nr;hM?qRg%KS+-3?_!pIb$(?@IzF-6l-F6Z@R25mbSO?A;pk`>+swK
zPx+T!)R1&w_`SpsiSAbSd8584#joi}dIv2>&YTyOIYBW@yB0e8cY=Lz1!Y%Y@ki+T
z$!s}y<_@z%O5haK<2<?H{kS3DCyh$I<JGu=LnBdVoV{5d=byRtMGkFvjE6kjL*Hgc
zlSA1&_GLW(7PpwKJ)z*$za1mvHP7mNxh<`qSd86{LQ+a5asQ=A&%2YS@)x{~VvWp~
zhC7fCcYzVgimZ##ubXJ!anig-B0U#cQy3cdAQ>EI{_zO06e$&}^Mmd0<{MkJ6VM8M
zBT~{}uju2APQ4RB@64vDs#J7{L*Rkt)!U;T?nIIp&J&uhcd8kvF$H?W*L*jYxoG^y
z5VF76sbrN1+v4IS5`yyJv{Q8?>Gu@)KGaB!V(5Ow^5~Yu=q~p^s4Bsh*|dqbpA2g!
zKR*x9q@z8TqKKm4uZT6qY{aL?h30$M)?I~(@PRj~!NF7afjDe3T0d(;vORaxUpR1?
zXrGY3rGmdeh`wXyd*l*5l&@dYE(r9wMC*rCr%H8EE9mpAXCF3!4Mtba)G9qbe2=Mv
zAES~bs8x^kQ%jE-9<4Q5pg;@xM%PNtn}wH(f-j1jGjPs6St@=3G0F;7rZjdPItg=3
zc1)D~K`ZPKJoqj&lwUoNeWCe7{jI9-7N@QB@(*(_!c9Jd2Z6mJ$&Wv@CSi41xX9l8
zUlxFMWm_x1BhzRQk-7$OZ-Agh^-2JDhm=M8Q<5V$EpYs|H+N9I0^yYLZ~kcBQY8~4
zXS;uIu6#O$DqWA#Q)<#!=Ae5MZ9Ss_zy$rsP0&zh(HfIQZuZJ_Sn1o6mp`NM3&&6s
zh1F#NXwLkJj(hE)%;ls$I?LsnqF!Hg*Or#vh5hlXrW-!~_WAa^tEw3P^EEaP5EG&v
zQb~0CB(EM!k~X!HTDuW%5u$2ZWVp8GgxI-G`A+|F`3*&yuvY0y32E9!Z9uQ<Xx^j;
z_7@r}nP`MawmyL@ycQB1^hSAVK^*jJ;+r5VN*K|(S%hq!!bACaVJ5f<Rb`Rst0X!9
z{;7C_7JKo$_h53K0hh2lxs7!5d;vQQMrujkfX%$jh7`IwgRr|u`0|5kWSaN24Avhl
z2czKG7HxY|B3zzk9OdT@JASy|KQG#7U4Y-g6vny{+L&@hvPOV;v=EHe25gZRi|rI?
zMSF?<DC%q0sWF5;7JfYit5o=17cB@~@{>yGMThsjFu!^GOS!u}-B`4QGzV?5q{)7L
ztwTmdQF){2q#Z&2x9EECR%6bR{b)dAm1TEbjwuYBG`?5=Pm^pIGN>(vmlc&tVC^+E
z<zn4|a~{!wu~{(sbdqI!$(*x~4gZ4Ect^h%86&z;af+it>9b6TcCC+JX|#%(Zqv&K
zC*vLXZRdcfoOJ6;z1w&<UCY1oG<nxY*kVfZkl#wAlqIX1ZZ3w+)D7oDKyoWJJl9Vl
z3{^=<WV-ph6ZPJ_j0JA#ec?4a?OtQA)ri>vPEPE6r`nI0xG0{)*vl$padI3G!UW-m
zJ2V}#HAu36ij|_3e|wgigUEcb$)1*FT=YA~{a7v(*w&I-8LVjSB%^e@3fi}t82C*<
zfEDsx{S6bV-G|Cg9y5|fqE{KHzE^uPZT~tRXMbX*-)`Spj9M|F#&r&TEJ4Sg&HEN)
z+LYi#!qh@f6SOpv8;66|BC1F8sd#rHvox6$CibxqnHm{^%$ENq>~VniX{W7+mSTkH
z)Krc1l5TJyz(-41r5cp1q$LR0oLk@JlnRU%pg~p|#eY`|Jp(Dom!S^(6|TQzQ!CM&
zcnl5gH=v`rr@k`>qJ1)3@GZqVh8!{nb%ZQ793pIi73nGlD9a>Gzx-9`@8Jgr?LKcJ
zJDUQVBuSzm-yP@>b%AdkH5%UVXVcNjovZC9rMXXvKaQrD9-)jRL{Dbf(W%sjPxpw4
zJ5fZN_I8BHG;3`J@#}OUN}RGJf%oNG^JBG1VIpf((`FUf+u4PmEEF$E(T|K)z<IHQ
z_QbaAu;3tAW|BW(Q%Gp5X{`}9cVQVPo0X<P6v1>i0)ayLJ2qQAG}5>Ji5gYGPl{5a
z2;AB_)uow2P&E8cIQWg-4*rd#Hxqk2(BU!x9n$p+2szo}I;79igejuHNW0HX!WiS!
z9@3Sf@NbK$B-~09Ch%J2p^3PAz&6j{xP<pzMc$3USadkb(!lek9_4r|DsUFsfU_VJ
z_B&&r&@A+Ew5?vAVd=BdRri-0!<o9-6MN%7hDXh_6L`r6p}l1`Zp?&$$Xwu9*YKa~
zDCr&SF;j1WLI`}NBqR`$1~Jt_(L#>$MzS#jXBah-!ssbXhm`OwSLlxY#XL^^&z2G<
zQVwuN66jR{lM#R=31p<`E%-!vD)Ks&?}v$DA^iQh`|3Mae$d>k#)z0&!Mh!XA|12u
zrE~gjpD-||FQp!uQ@yCFVE?m9U4np*bGa?0!~N>0;l?00QQ=dGRy{FG4V#oaO*8mT
zJr(H!d67V8d-W8~O6S#oLZ5f*ptT4t*R0G?VePz1rZ)>$FJ=T5%LwY_!9x1NJR>fZ
z0Ezb-i3dCCikQ|xMj=W80ySJ*+z&R+LI(bTW#?&AD%On^NQRN|2B#9@!4+uX|7QsM
z51G<CB${WppyPC6hi6t?zf16r$o6+*5}M(Z-F!o(izPeVdqDGYdwyQ(dD-FqFeV{m
z<Vap(69RJk4F#V6W&oe!#hcv|VN`-!JF5-QeEiA!a;uIJtE9J)ah>Yp#j6KXnX!6x
zsQz;&y`Dd0^s-|pj(fQ-pLfycY2vYZM0DC7#6!jg_06+Uz$hS5I?ly0MKrX_-t%-B
zz0PR&_xivoC7M-<R&R5J=>L82S0YD}5G=3#N3dAPhc_klpGp<|wo*4ZH?yxOqQDp|
z1!7+p%GJlDvHs{H4!Hp8N{;B@NhpdLxm((MH%{a#On!3HWTRjr+GM`BsKm7J@bLYL
z6X--$=w#X11QdGt-A|WtJss2(*#3_v+&vGdH^<$Nr-Pfsrs@y3zwMtDe`rYr<sbMM
zAB_Ui_?Zs~$0PzvQ1bbD^Tu_#-b^fqS-*V*uqYf3+yMNE@nLv0DMJ|VG;glvKp-`}
z>rU~3k(fUXb51w%eq}?}Fxbo?TR-bTM8c<UR6Ohc`=?T`qoAJGR0CT;@2futLC09(
zKP*lj59~R8hz{U`lMNDC10x<=$^iE5i_O8H_(#Cdb`rmw5aEghc^C*oxmt&;{aN+D
z2)0;~oqg5Ca()VL1i<+mTX22sJ1GWu83qlp(O4=Uff)VXQ9!!Iu%Z}|(9QfOKdd6P
zLkaP?yuQ=q=~p1dtLp9E5cr)G(7@JJzOBkQ#<uSlM0fV~qE<mB$o%f}(NbmjC`--*
ze-$TZ3k<jbiik%TkyOqq+cI;`ssM)N=8d{z{xQJukWHjkcv8%2BoP%cwR|;6&DIzS
zQwW}lzlzKjhyE`Q7G!3iK(L-G%1EYt_ra8N54t6~SHb?P-=nlCr;NjG68#r8n4IzT
zY7lv$>g_Ck*YtE&bu*2w6&~PyKE}Z-eEka`bP%+}##xyS@28obnj+ycMzn<vvz9i+
z@zU|_HZHIZr~N&8RSC^n^%0)yqE$}gSH+fp)eVy#`P*nK%A60cw!232n)B_t>BY2>
zkeTA?If#yj&POvDb_?v{>lsc)U{Lw2<co*@-2zA}G*xpJT6a^di$Y|st=UYz57)y?
zWw&&|l_BIG_c!~7lFRSO!+%X1(?_`JWKw~CH_d(B0<$&S%eQ?)ctv5^Zb{XDr3&Ue
z%`u^@CMp`M-b9fnk~dm>9FexWk|xKaKNbGeNSq%Um-}+Z)B;sii^qbPDB@oX`##>@
zcpCaYyZRr4Nc=08o<H8jy8)Juix)y3Tg7Xr^VA@eS~v4k>X@G4?<SkQTtPhI{PCs<
z!bU3$BRk_xt`z}vqiPoorzyj2ulVfLAhM;6&=};v&*1He3=qej7Q!!*3*aa>cD!Gd
zrYN-{;A<t=oW?tsLil&ra^vsfX{EOZalj~kXKP(u(U>~+dO_3O;;-TKf4h7u4qSgf
zmMSt?0N9Q-Ymo;n0inlZc0=YA*kRhoY;wm!6Zb_x6pF@@>imtih!Ww7t%dumEb=zg
z-SGnovql;@;Kx(!Qz!uZ7o-iJeLt{V0~m})zDU7IyId`O;=;2U<MjL!np~gyyobFz
zT-`5!SS>!$t_1JE7Stq(>O{mA*7Ai!K87OU;C|J$@8a}Xj{DE?_v<|vlDpv)#ZaR1
zLMZ}4Db2$gRvrV!?%RcAv$~}T<LipLQ^<jDU_H0&usPrko39m}nO~hjtD!TR{Vk;C
zLKuH0;phe`tc=a+$PkE@Nc4iUhM@k=F=)Pgu9WEHdK*VdF4TH<lTJvk-nYkelTqZY
z1b>*}DSS3&)v9zj9d#7XD2wt%t?=LoVSB~V1mC|Zs;|_FmE!_?3ze@#cCDguyu0`;
zklx;bTJej&<<Ef!XtM+sSrisIo+Ws)4kQI9V20NBqOGI{e^G(md_ZkeB?_IPUssGy
z-DX%kr47OPcM#QUJ^@J2al%%1m^@k7;>)7xL9(eQYeK~xO#l;Go`CmB(^4l=T?3)#
zAf643<B8<*hyckp|7GdkB9^$3AS{uS`BBZzJPZEk4@G_vq1~3X#zDrr#(bYX=wEyP
zJt)OvJ^0fJXj~bcKo0*Bkd?idx=)(`vk7Drat{Q<s~at6`gLiwmK*>~!l>XKbgM$(
zych17d;A%=7PhM1fT==2gXM2nwRqiWSz4j4OZq~xbFX77&KIG+c7H8Qt$!lRjQD=r
zU!{FKRG5d4D&8UXIo-$h!Ui(o;w)N;@M%-lN>s7G(cR-x|DyB`6n_5Pw*5`6ld~Vs
z`ZX0M9>A$a=KoY@!DA$(Jsy89zX4Ziqs1y0ZR95OvT1X`?oi99+pWrXvZ4{k=OSzc
zEfcj;PruqRA(3<pb^;Y6;Og#@z?zdo^3qsKSoiFmRo;5gAXd+GdArwO&$~uKZ9w83
zOOndV@d8ptUG&_;{E;ach^d2hklxn)UKdgTFk%4b0G#H(lFg7#NaLv9@DW$yoAF10
zhH*6Z!xa(b;gu4UBuNFX3rJ0SX}9EA<y*}_n&L23F!B#bVbb+t5<V5w`-IYss@l5Y
zb1{f_IV;D4)3np`@42Rh&*){5<fCKPJIurk-RM6}2G?jlwX#LCOGdnZiSC$jV;X*C
zAYO-HDW0eDx$JLLQN4On4WTG*&{Q;``(DFg1;l<>cH@2!1qp5(<PT;sQv1Nt@h|id
zGtjHIkgWC%x8NF{-g`vZ06B_NS0AeP9Zj;@N7OMBz{qC%yMg`aa_Z`456#nUhkP#U
zr?tm>xwg~A@a@dAaBt%eHp#mxqA?vYW-(J>hnNQEy~!Jll?cV~1dG?JZ`|b6%s?uA
z_<3ibi@Fu@ZVhleyMvqzWRz2J8aFK00oVxj>qh<M3Mzo_-tuNq`U53BT0UpGPb#zi
z;ThA)o3zg|V`6e>`DQeF^k$0{no~CS(}Z+i(eSAD0WooD?&A@oZ0qUUMHttId?6Wk
zQjMWZXDu+8RaSV~e1_SUUFcg9+!iX_<^5K?Trk35Q&2L!`sLTu>1+0%`As3B;g~n-
zN6LCWVXprixP#EmUD)~cqJgln1V&D8Tr=h=b`Xq-a1`)-UOy`u4~Ra{IJ$_03inqZ
z@*xOXtOG`Ns$LyRN}swEa4OE?H90YQWb8$>V+hz<0%gm$w+h{G|I4(K1*^wV*Ob{t
zi+HHMysrJa9(^lc((ZUBOPBCIL&IK1Pm>Aelll0M>`UP3gauW?Kf`Zv)Hm}DLc+z6
zx9f2xh5>;nU-F?QiUZC87QOK77w?*XCerMApgH$ERIGaQ_-wN{9d#-FOvn58$?jzd
zi)}v~LshGl{*GT$%1b#edBUuS>Fb1u+|*c^&VHGM<*%8j!U^-Q81h!wtgP%82<aB%
zrwQ-gS)TS2diY=&dmw%!Z0dnsDxCaer<R?C_NkQwio1fzoa)lLOMaYyFcH>F!T_7W
z%s~zYU4UcpJtY5N*VL%AT0ZC%c0&@UT*ogOqq#yukxz;G35B}Li)XqMQjgGWSE2iW
zKk6<hiIp_-bh3doY`VA8zK@k^4!4f^5U^U@+Jq@v7m0;9y$lP#{Hd}7AYE?s+M6Zm
zTWS84V3d|h8Y~)x<RM-A1N~S8G?G6<89i%FQHr8Y>$aLd<EmPsbtiXy{s1=b3k}1{
zg&14c!}$w48rqz+3iQL6%eOGl$n+Mk>wT4Zx?O%fD~wZ>Ups^p^M0=gELJX!I=vzs
zyUs&?qXeHB|62JM(-k23|MGZDk(%QjHD6}HiPxc7vXX^ogP9bmX6qvmVyrfR{aLmR
z7Z;u9ZDF<FEpa&A_{iHD=>>w$>Yjm<-hK>q=7ZWdB1MR7yO#X-@2OVH80qypyQTgV
zJYeY$)c4Uuwc2}~?Mv6P6fN1Pc%EcYB*=mDdIOe+Wdrz4$|V9|7B!5`jH*-%9BHl6
zcO=}7rltP+aY_uuWRnZW=hAw+pVIE-_*Ei5*`<wVuxq9iSZcM>TB&5(zy{G8ahiSE
z>=vQ5(TWdC?xVfK(5j=Ru2}SpJMlg@X1<rQ)$XLvlVdW}p5(cM{l4lz<li0ST~n^<
z@-K#q5Oy7zlzOV`GVaTIfry)RLstyxV?`yh7Fzeyp|{mqV_lZVcpp)f!h?=VIv%DL
zC}tf9v)A%c^oF@Kt7yJJ9VwGY+@X#{VDKxbZc6wcl*KD{>37%d%9zb}y2AH1E{F3g
zV$;3PT%P4M)(hnev;U6H%+Aq-KMtdw6kXY!e9o+D`KIq?hgCyIbR$5KkPNe4h7(Rt
zB}<BHT?T-`UNTn?g69FG69oww$5cBrV_M1kp1L7w)i=pFm24-b(}ZNlNE!lMAD$oA
z{~2sgW>}GY-}QFwXCn}jTU9S_`=Yh{Qr}p`QkXI_-xE~KQKc>F<#x{?w(<Aqwt@o#
zpY2ObqkT(o!}fyH8B~_{Vn5iIJS;D0>jdp_^g^0sJ1L&JH+kCt@OZ;apa|W{r7+Uz
z^y+>?)in?&$3r*DMu-VlIi(VzHClc90ZN2R*AS39!V?+6DG*{uy$>r&z(xTIxxo78
zka!qnvYc9@MYoRpGUM-N+f#VFu}H60p(C&_T8B(@@aB#Y%D`z2l`3}&wvG_MYw~_Z
zKwQeUT<dJpb6YmMU@r9~UBb=5&|}BIM#tB}!f4F0m#-{1RO<K?h+KI^1s(~2fqQbV
zWoaG4<c#O^YTso_Xj1c2)b5ljnbe@5@Weq{B`pp%$CItl!727jzOnK7w%3TAP8o1%
zq**%dvAfgKMQJ>SaPvBiO{5TxOPT4BPlBH$Y#(n6+#=Md%4j1ti(-Fp0><_E5Cz6@
zPBswNTad>ZH>YzW??xx<-w{f`f`DX>&y+|s%S-P+QRKI0GP-|RfHECF{tN}CJ=T~<
z7%R;yiq;F2b*x4|)A_Wph9A@DF;-F{)Ik%|z9y;@M9FcM4q7x35^pg`Gx=NwCIHFW
zllk|JX|^bsS;3ttWTIBiz8RuZ@6sf2GlywqQ1dOT&X?2bPYC>+z6H|izLO&Ft#TXS
z9P6x=k4kUY9mz%$po0HxN2Tw!{;}Cwd8vakJpet{kk4YQ^Ve8`27<1SXDo8y1Pn}p
zqp^95!uI8so$1GE1+=Xjusf2QCXI-A=X4Y#2Y5G+)s&dMs(j$J>|vgr(MBpL%Z~PN
z842|MSnN8pRxta6Uz&s3WE86`(D#$pIZ9s^MlmwpH%Os^jFlMqRi!gYmcvq{^cM$2
zsV@$pnHp)-3K{Q@ed3i%W{zYM%TO%e1L!~9b@OgsW;zV>%;{zBHIG6x7!q|-ihRV6
zRX^-;6X|DA&)*P1Q5|{3M~DGRUcAqsC&#%&2NEW~Xkp~AWU3zx<t-L0i<pn9LBghS
zcBS9wv;|BwoLR$t&N})OEVFZ805VwhcsVgjWUxc@{%qs<2A<|YI?YZjqX9#w`Mb`-
zyV-6Pvj;O^>dEi+y|$#U%mw4#@dv(ktL4W!S+L*b?;_#tiM#HE;oQgEC%%h2lcf`K
zAW_`;A<*tmq6{q?2fgJSu;cKNd~?Nzq-P9?s5VV{1lQ7GJ?6fr<Wi=JVU_3VLp9PG
z_$m#Kk3ndAuht}M*7j%AdFN)0tRr+R5h}KWGuc#D&tbo)HOccpyIe42xN_)nB+}Xj
z(3-lMGlgo;;I?jojKm*zCz@ZdM6^yXfLr%!5NkHQi5pU3`Sa%zkW-r4C|}h3-Pg?p
zo{&hSpeKSwxOu%j+<uHea084SIBgr~j4(q)=4>iNVB=@tf%LaFB}CV6E0){GtBO0(
zenlqRzUwBX)_(%XfOQ2{Fn&$YX|dC990F0}{&`28OX~$`_)_r72Yg(YP!e;<n`$pb
zj$W?gI$RTM%$;DPz#ZKxxB@JOQf~P`zQ7=XqU#+Pj5nVSJ_0O7l7AUD+(I{`bbqN~
zQQQGOm7BW=XQ=oPPY%qkTZl|Q+%Og8-5-Sjk%_qBK3B5VWb{Z{*^yuiBeD;R*Le;S
zTD`u}$_LAN^vR<?cWiA&D%3}R4x6U_tU=Noo182+Qu6{IuX<`etas#gTrlW-SFW)m
zS$2*`EnW(kDBEVU|F~g*o|@K&xQn=M0C3L?F*XP+HpA*5qi7%wHS>bC4Y3dijA=C2
zbKhGsD0!O{TyP&fqI#LKfuhi<^1(q!=ur*NZ$y;6pS;ZdDjx^nXPFCF>qb&z$(Ox%
zfUW&E00H`dmMPpn{<i83BrHq%ancW6i|UsItGooJ*yE%DgJnZH+VrA0qzw!{CJDyP
ztgmj{C9LsOnny@e0D&0YEl)}nv*`f~uY2oZV|<&^v3;-zC{e)e_!r8k5e7f9fN{=D
zfOYPI>JJmQMPhOectpIy251gjII&Yit%nXgXm;A~eCG++jBSS|m1(_bk@v<mLiTB?
zjdBp{#&p6Zv}nqOe<|jK4xsYO*UAL2z#<3L2yI=*UVX<*0T;+RPFV_*B8YDCc~GW>
z3U~U0$&OXTh%AZntr5kEUG0N*O8$1#>k@a!`Et=zkq_*ZbmvekTr%l9f|~7qi7@Z~
z;To@M%!g<!RF$^hA7d5U+3Lz05I2p-IrV=NS(?VL^N4|0L=ITi!%rb_Nf|>QA%%sA
zfWh(wjEbD2WT5E+XG5fS?>60{S|8D?KS0D=%|a06;JOd`;4A)cM8;|7JvHdgFIj9D
z?G5m2F7oVRpAqg2ZgQ(vXtp#TuP<2nxLcoq!p0fjf#2TyHDKV_QCN<7HdEjWfHmK9
zsfGNKI0a<Rn;t}zXdHMZRvVt65VxH3n%1R&VoWni4iU&s6v~W4-e+48atmX~!}N>A
zm_5l7v)k_AVPA=-<!rEoK0w9H)by%bf{kHF!Ax^#Tx(UQ;3=6St({lkK@+@IuHlt_
z0~U4%Q(og4!<n#N1CsnH>swvQcS8ew^o%Vm?>BRGJkMbRcoxrBTk$BSi)97Sz-r)R
z72^O5LUh<~7&rCR;f=IpeX#j#M2+KZ1Z|qK<-8P7;D*$Da~j~w-?8U`bg*?Rm&Wd%
zPd5AuZg(2)HbviUrH=towgo3s`hb_}y#qg_VGWpV`iX3DQ~Mxj2+7O-mq-n0Bk4<x
zj3=q@a6dau&-n5Mnk>h&?X9aF|Avald~x%J$Q`SoEfle;Gi(FdP|ii3qmTvgqbhaX
zYp4p}*!Pf~-ZAjjJUTj+>E<d=oM3G2ZY9y}Q-iR;rr|`<nDSp70<9hFv8ny=&2J2s
z1bhZM3(}?7<9|n0|L*MqN#UEg-$6duz+H`?a6>nk$ic{vgudI*F}XebC5YbVf;637
zMi8aXZRAQYI?P-g5xO(6omA^^k+>*ha(cW@>YJ2SSopcCaZxP_Oyg`tZ;q<dLaa(O
z8#%S#97^{@q6l(`cY^PfatI2@jgt}Qk~Iyq8%yDXZa*mr?#xkolsk}4CZ1k27TXj}
z%_l3KyQ#PnZe!~v+i8x;d%-fftN^wSXmhgjg5Gv$P`j&jhoC$?u6Og8e6VvP1@9dP
z7Anp^Nd#2`w<kS{1@fJJ-&%8fb1+17Qv<e8r0kO(^Q6Gl_ylt3L}AB$+?H%`2{-8<
z{u-f0%dx`S5hGD%!hJQXCT9`?&RtE<#Zbg2+tGQo22ZJ@^wia$MZfh??NU;@Y`4{)
zCuKP=YBNCarRP9PpvQwS1g#r0hIEV9PTbg}5-`b<AXTv<*{R!9XNo&1aP!PP{agt!
z>J`&TlWZJsr4a;t@fm`J_Ry;Hqtf+8&e18-FoJcD{)$<tQ-;hP3|<)m&BeDr3TZ{@
z4OY%Q8R1(ASzpSO!{zXQGy$7HVdh00jgN;*Y_kvMU48I)dK*=Wv>NEt!6BL0cfwJv
zWMCUwlYJuJq3rkz4fwP&pYJ)k^**ssL6^F5?Ru$O<}%8O=CTB94Y^bp*a|3=cME7K
zao->(I9gQ4sn%g)rOCojE%#}=Jk8Pg=2oTaAvd5gSswr~PU3|<h<{37N^Z(RZfAr9
zdpr<uobz)4F=cp|#u9;H%j=j+BA7I7F@f9t8LH|jexY3mV1s^YXi4jz1|e2pAqCDw
zPO{nD`gm@J{`)y^<!=QH(TXo0<E_}Yd%C<j9D;82dHT%1f2K=8=y{B0>DDsvdob&X
zBxdHcJna+r9i2&QpuH=IX1<Kpq1yl=Nc0-0Zy{dW(wQ+~q*bgxyhDyKWeb3ulLo2p
zwMvf^Oss$zjWwK(^V5>{uP9x#6<f0m-#-K|!P=7PgZ<7>k4@X4FkON>g1(5`Z>0I~
zsZ7;_4RM7x!ympjk$_VUn0t%A=w;Wc>G|F)<g|5f#qkTEO5n?ea3&-*zQ`NFi!-XW
z*3|vVdT~BWZv7`<aRK&;Oo(v8h>q3K*(_GtdpcZT)!udcEL7s^%3ENiZtNSRa*LsL
zaO+9H_9aPR;fN`nTbQ+n+vJ+Zv+9m!w#bpHZ8K=DXaaSDJ!&Q8*<jlsl9%p1$k@F8
zW050bNz?j&0@If-Acy#XK8p!NB%a_9DS{#N=x@xbp^?TEb_pY6I+a|uD<CDBCqeJY
z&KP(QlgBO8lWisxN1poqbbKvms!-muG>@uDAY)7iCbJkL^`PpTK8{3$AVjrXpAc~4
zsTazU0Q6c0&J+pmr<hvT<X=$Qi-c|2Hm;y5OFw+b7@EBEM?`MiQ;K(Qghb7`w79-{
z^&<ZXA5lk##g2~BzOTd{!j&6)1>s#OqmA36GhBY9xj#W$d>&XpoDewTis%OYxvBxA
z?#hou>cV@;{U-}AHkmTR_FXMrIX7I&wJd!$t$`HtffyQui*|s!o5BuY?pA^W|L_ql
zR=rnpXRuoam)324qc|U{Nfek9cDc9giKAbbEl~sp51m&S%N=9YPM56VGR`gWK`LJ0
zBxj-QNk!BD;`ad*=n?_vF^u&eycD?>BYS^sN)?^HpASXjtj2^?&Ig<GRC=2C(Zp~>
zYtQ8yQ~6(DedhBlm^JV^rE}WOf0WXv=Nsk-LLeLxkiU)}%!m9C6@_&tf}T#lGbIsG
zRvr4vHgBOnrSf(s?<WGwB7ew{sV}T+*-=XK2Lqy3Mbc5@ua4M`J+%QvvI5*%9IE+c
zLYR~6We0KnDLw)$MhMO%%;D|Y^CXNkP}0#4K%ZVSB<^P|>Jk|ve|gt6-Tn%eo&C92
zfzAAY@7_zrIYuQvQI9=oyNuLJ_aF1%i&4An{_Ndv3DZqBnmsgc^TP*>Q?_m#Kjj$i
zE~@FfUcs?>${r%E7OspbUZO!AleMu6dOtc$ru?CKGtB&->XpC*s-;G63#Sz#^ukb}
zLq-b`_Ei=khtlYLT4YaXVZ=h#O`Dm&?f=1L2W?FY$XmLnW?53lpSB@>JM}g0=53Uw
zny(%j5;!Tf3}ri#hCD&DY57SpboL|J!Z5^?ER^XX0=sDC&Vjd)T~D#~Bx$VRfR;GX
zn50@*%T3KY7#7xQ-~`1$$tAY4F_lj?x?Dp(o~eBI_9KK|5af``_&q@6Xo%P5M^+n>
zY)!tWj?-iZ(awQyo!zTQlS2`JE+RAe3W>I-93%wk*=sQpzHjWfSKbl}&ctBGeUQ;_
zcZrOV`n33WR0?>LQy<7K^M%w{TE{x!1&Xh>2gxcbYIPas6`}B{ck9XDTC14_U%63w
zgRybV?bjpK*8Utf2?)!wQBVk<7T)x~axFkGTz@IQq4dK7=#1TR4g40Or~AkV2_8kt
z)6cU0RxaGoFSM&bVqsQH9+am?j+=}X$cTdg#{kPLBGne%TYbX-?0Grgs|z1zn138u
z&9r4wiUJrGA1UfV|E=}UcrduXtT}M?;?Yr!#2h;)bQHLtbib7MgO^mVVyEoRE-N@p
zpRSJlsE<%4#fR=`#D7-XGXq+3Z5!0;sPt97{0NHBN>ZFJU|UACN%~K1E)IC}E>Zs0
zD$-h(D~UyAXZMO15R4yAb=Ln^Kz#)kjHrNCnyu)_x?i;7nuYE%v)9;Opl3BV|9^4D
z>rdavjl{$-1CHzeN6UT{P5(o0)V5I=ol0f?CTV#g1iqg|bdTe@;m<%IUz=g--Tx3a
zh`_|&$8|#UPF$n)Y+QyQd6{26XI(c?;Ha_@y~U{pI{lc89shekWiAW|MgEsHJey?T
zSKDu3500xd;a2OYUl0GCV#CcKqK#x_^v{lM2S+O5xQ2rU@b?rI4m!sFoy=_n*fEWo
z1j$gj;!Y%1t%^%r)h4kWsAX%@Ec%Ej$b`XSWJPja?t$-7woUfNsrK!Rf<d==F{9g-
z3@^>oe7Tx<xf<;QLfj1J^0Go8@HJ(^x>n2q9~C)JA*8ooU9RX~44K(}<xIWhAf<5I
zQk8}#pIuua@mf)|8m*==R?~Pn5NSpa>=*u^QC(iC&t;%7^-qU>>EMu`K$lf=RDG4p
zg1^&YL6u-*=qnkNr`}`7#Da*9=B9ihakcpiLpaC5=PjAZwdG9|Zi;*!QcwRzAn>aV
z^h3;7>+_#IT9a^qoM2$wbaRA!+bgvjhR#}{K)hVl>h^f&%Iv8?czf(XcqHQc`q-iR
z*oWPhJi5sYh}ViT0k1_T^w*!`%Hw-I`s`XUfc`6Hbn;I7;%dEkma8;E&5E*ahy8;F
zkMtrV@%svk&JXIpg&0)3WtneecB{2$xA58>KM$c3ybN!cY>j;8j*H2<4UWm;Ht=db
z`}*m*en^Tcfm^H}C|qC;4GmRQRwl<`*i~pLIMeyo07#|@z^3eDEFtG-Ajay@fm6->
z&-E0lEfq`nK`xa!`Ea3vhK{ajtifU`1(@OK(RKjs7T?Y4)xvtsV|oGj)DEx}OnPN4
zq_7w=1NvK&TA8vaI{sp{K7;?`@AYkfFp>b^_|4#pS%C<NZg=-@)_*v~QVT1J<aCUf
z=w;-%(Y!l-Yy@lvas&&N6~}u1kCpQ_q4=<w?67s5avp72fasM}v!*(#ZbMtA?kw?c
zk1w@w<y19)u4gYUq1KQ1JNd#^9F?5e^baAnp;#*0BfA#f^_R|lZ-e8+^R0N{(dOTF
z48QGrx}d<m(bLhC?LeZ(@3~W18Q8Z^)t&(*1gdRLW(Ug+rIkSC20JW+M&-F=#V%W-
zSHEYD|BELt5XjjEcI>wrQRlAwudMSMumHee$ax?W_Ui<~es2vR_*`GofWR{vo9m$(
z0gpo-xd)(XPG08zd$RBy!hG_w9~IGVXjdj&VoKO@=TTkZa}Qv$=7?Z)+=n`T-8(4)
z)>BdV8yhrwyQUs&MGukDDkhg8F$N1BOG+Ff;?vOZCOPR+%?)bB=gIAtqoJ=SP#&kG
z<JO5&)k3f55e<}dL9WZr>~W@7&#cGIf@ZX`KY&jFrDjaT0@{=@FQCT34;2FT2`Av!
zqpu)|*THzur&mEY43p_YBVg-b;rI+!y6W~@;U19l#C9jMGn<?mX1Evz&}kXKMqRHc
z^BsWAf(`Pjv{y-iO(J<!{#x#R0fF#EP^ydtaY9!}sk!pEUUPL@Q|H5&H=F&k-m-Cp
z+iYE4CP&$)jY|BZrt~pQ7M73dDNDe*jY6kAgC(qTl%C8`Zp&A*cKX^tjorGg`|-;l
zqa(4q%BbEw>ryOH3{~tZ&+PA5Dx@qV)66jbv1<57+2@rBV1)|Z_g!-O)4g}jH(VUL
z2XA*8qQ2Y!N#3lxx?x{)Adtgm)cb}P;1Y5X^{l+MjtIY8p!uh=0F@id!1k{g06%p9
zd(~Py@7BE6b}RwjfF!cODf}`eG{zx2&qt5?6}e+14*Cxcw7UV|5?0Plexw7n52JX7
zJ`?~<JUZd--^<xQO4VYwa^u*l&;Bj655-?=_$oFjwd+5Zj2m$1mur5fQ#i6?@a9-k
z06aeQg}X5fgjA{w%c`%>O-c~`qVO+mthjaZkHxr4GNC%zTs(#;N*pT%`HeM8>Rz!v
zgYi_+spH?4qf6zQMmf!NEsBq4Kcjuu>8_7SV_%G=0?xpglYgb{PmB>yIWd5L4-IMg
z(XmAA&<oQ&3?%R~3_<wELM1++aNsMOWS;;;@Wx|+e)C#okq$8ygM8<HX!tk_NP+ZK
zI!&1Pwmwf3X%|%S0<#@(melyCbBg>EHi#HHNk$p3G)!Iqt{zRjCK-+U@UaIV@pZqP
zu6D@%dwJ%J{znQll-M5uacQQOb-xb#0R!)oou+?-nAQBi|KopRnVoJ124|~VFjw8c
zDIWt0Xc$QNTxe60-Spg7UDt|VTV=4^;$kt0E#XeyB8>{}WjS~1O4QcYnGTP>7DQA6
z;XIW|CWD|bG_uo+LEE6Gd2XAOxf~Ff8jReE*SH0+nr`mc6Mn46%Q3eV?fW@QT>~_*
zWx%~zNg41V-klEXj$+Mz5%Q{K6S&)F9_o%Z>vCLT@rUM7(lY8XB0)jIhj*fv7gJru
zcV(J8G8WopYJSgzB{2;0K;NV1RmjUTAYpyw{iYv3p0x4alnrp{USL3CHcb-`lW`_J
zYAw~r{c$y|>7vKD8Tl5k<SI@KT<q%PGLhzsq{OXuLGaip=z(bWT}Bri1Q(AjmosTC
z+b1eDKM|R-jJ6kr33dR>lRw_V0@xg%%qNa@LHO@}p(+5lH{@+R$@RTWcOQNL;%iR@
z22VGJzz)q5KwW<Z=trtG27*0#c)NtrAhUw6{r~{UNBeHoJvVn;=<hz2Cs3BM0}$}h
zxjvatoD<o1{>cq{t)EfTA^_+Fd7Mx{Y5NQ~jMh-Bfdz!U5|n}-2pbrN#ta5iL4Gxc
zHutM-6ufx`T4G)VP7(W1vK+|z!wL9+n_kzW7~n&S=J?y#BO2V}6CbSO2PI7lXaP!h
z*8M73uJ)$xj@~8x91i`+gv}^36pI8?0=SY$xzoI7DT!ZcGwgrCl8i}1ON64x)_|yS
zHVkz3uuw_GBmmr(A1!D{1FuQjGx#8o?MCnrHkmE1!Q8*|=Zzdkdu@MAy`+=>XuSmD
zZ<d4@VWtGG<m=qIv(1IBr^R2Mh4%A;i>&5K4nE&l-TkW1_Tjck(k)b*Lwc(MpX9Eh
z@ouSlv^7@6jGCX@c~`y$UnFo2s69gYM$^$P8up&1H=b4%AZDn4OUgXk(d&(9i+CJ)
z(!f0!Sl1lMBlzC&TdhJ(@pP^0Seh7GB}zoFyU3{S_E?UM+b+3`p|;k>T>n-s^m11n
z9Y$#~d(?^|@F(6b@vdLI;O%(~_xSF}S1Q58s3C0vYyH~j0Ebb!QV#_|WUiQ9mY%20
z^fiW>PF0{&>KGn`Z&wKi7~QVM`NwH-ks5Cxb*<-l>eAD_8TlarLCJ5zh>c;z$c>Z-
z=mP@YIQ#DBhsy$g{x7%dDj>1E{pH^HQ8mRtLIhnxdD`gA<m^`mtAs+#kX;Pt(67X_
z9zySE!j8JTI1^E97|D!lR5|^_+lSmU{nJagwDjy{z6S3}KXZ%SE?xh-iLTPMqcroz
z^EENi>ADuFUh)CUt4(%7QfjGnn$Z%&B*gqGX>#y8ILJjK#wD;L-{sSp09<lBvHyTl
znGaxi$Vsbhep``(&k$@tLT|MS<N{6jsrD70<@C!q{3F03(C?s4G?Ie^O+SfivARJW
zrI$)n3Z;J5n~mu~VLg=b9%kP6UmH5~!8-xI$9IAIU{Ap8;F^3v&Xds%sWy5`t;PIm
zZIapm7OA^4Y|ldm<|up;m=^n21i8v#;*$Og?<wnY@gC{r$@o-}KuGR2wMqO{;A;4Y
zx3kpfh9Vr-s-;OTPz(h%?|IaB70-BwZ+u`u14*)q6|Eesb5qHxYnL!*T-#Px)+d@{
z*BKf4FvwQsTjDYzyEd2hxMNG_{_no%xaewRKb>m+k$hI5K5>%Da0J!sG)m==Dq&U(
z7Mb@4Y!vQ0gh#-U;4TJ>#NPi#-Y>!E(wMDf&C6!kVYAr@Skt2bUW%mQsA(TC(>W3H
zLyaIxSEG}YDZuc09w(goT`EbME-bp>AE5z_$|7H2d9Rus7Lh=fItLgxUf_b`kkITv
zTFkWq=sw4*eh#jDKt~fvaNN*)7M7Z*HzI$ob|<dXJ4kX&6!el2Zw7!n@ljALP)+gy
zkZT<mVS?h8^+Ub-(8MYv9uZ|rPPA;$vC%8{8So!(DMQW)fj6HcG^@G-WzrMr0P$KX
z6fs;02OzHu2c1+i3fo1efWfN-Mo6uy<o(5w&szj&AY^K*S-{QrGw#jrEXs|=l=~_u
z(0Rm(i8-qV*3yA!wwH~~yonV~D+nh~PSpe{IH_8bp$L5`)Cl{DB%GSv{>8AOgiX`$
z`DW?#`0cKlNl_54a2h{0!uXP?O-)T>`{OwLIO}*gu>agJckX0)yMySAc!e89#F#kV
z;n!_K+V`B~t;JloQGWo(zvJ028-0Vzz@VYloMhg8ja^Zrm8N12^jYlW^}--w3sw@0
z2m7GqXb%WNIqHak==o|DnqP(WB6!z3iCHWCyZ5^9U2HjRe(p6IaqVusVtw}v?yiaO
z5}R-Xs^>eC?V3QeOp|G9#$yAOLBa@=Sob8g%8D6qF)BV=H*GecG+8W>&U#mRP;2R`
zywti$=uFj01;ZcVSZPaZi<3KS37Jlq6Vm1js4v;)R=2z7z#n!Tlw@uLEmu`UTZflw
z?md+p@Bh6hK16mRKWuF*vbBnby_h7%^H{R_4}Lk=-Tj*V`LMt-bArlqxytOipS9jO
zweRN=jE%}Bcwge%w<4@BGI(I(2eCs0Y}MSw1jX8XfFrW3ztuiO7$Wob6;5&5&elZX
z8jnbqcwlhYzFD2?(2q#5S|d;-|3!vodX#P6m$nEtL{3iokPBP2BDAB1S(-xM5^9m1
zAU4|DKE?N^jn6WeZHXe=NZh)wj4{#1S|3b@MKTLtcG*4z-+GxO>b|75fp}oDUp>RU
zp1KimAYquE+xT_%{yHf?u~n`H*-wUsKb*k#+;?$sSpQSWQ1K^Y%A&;f4yomKNa*Y1
z|Ma^Qdl8nuOcn3x<7=hKZ23{rqY%(X7Ea<hQh|#ZPPWcD*;hOhDpZ73Jpd!%g5k8(
z7w&a4{x%#->r&N5Jo)0>x4~UJuNvdLN1-WUq1@7JS`28Pi_k{siNv6i3b{XF_m`k-
z>JLqdjB=@1u}RdJe23zFGd}HD^N-qPD=l37UnVD|%d<OLNcGF@+H^F#FzPD%pNDBg
z`<%U50rrz;>#^t*5vEKEvlX#&Y%tx{fnUQR^S<BXw#uL0)vF#0oF{!k^@LW+ijJD<
zIOZ<4VT#N0WOkEJF6>CDac<|7Ojo(>+F@y$6iqxG9M%|B@^T$Mx-#MhJ(K6@j{r{a
zO^evVO>%pZqnp`%XL+%E>;2hiUdKcBWzr1lch#YVKG(xd{ZKr+@YgCY%@mfz$LnZb
zHp~aljN4DSI?eiQLcwGY1Pp5nB-Qm1<WhRqJNJ#&qV86nifvLvPl7`<m~~;_;@Y1G
zi!Pj#rzGE3w&F%Ud8pv}s<arU$}>$R?>J|t>O3$+msPL7-bVlg+3`?%9cuSJFp}~D
zzUPA$`hU;{^l|^;`-Q){qw$Zf$2M}}@k-iwe9zB2`OZT67&-=OeJ!A|%8DuaZQd-k
zJlR$mL7Kh^jNu+uf$J7BQj{)^qq%8LzHmLx!_#RbcA;{%cme*C;E=Hv<e%Xp886eS
z-=;nM0p?`&N$WsfUH=3^RXoim%=O!oo_2Sr`m|~Su5HB9EG^a%hrDHW(oa`r(N@;{
z=VASQV8-F36Q9Q!c!D?+A!?aA-70<$bw=w)k)^@hBOGw&kb`@R>cY~9(dj_b-=3Ff
zS$u?QWyRq6Q(U(zIoLbLvd6tGy=mrTWIg&uF$X{eb~w#kdd8v<dexE}z=$;rj4dhY
z#HX`yC0DETwNIam7)csjPP-lEDRw(=5QQu<N4&jwYZNF6-@<UXNsPcqZ%MMexfmbL
z&<n78KgPYu_=GAKEoDhPvyIt0r56St$5ZC8`LMUaX%MO0cFK17=Sjs*_zIfUtSHW_
z+`{lQ;6^hh8LXHn$YGH0YHnjy_?W%7{~UUBC227rQC=oL)>%GYId^G7V8$#l7gLPe
zKZst%Wf!T1b6}HD-Vwim<BrcUGVb0g7=~pR4?Ki&8rJe0rA!suzV0erny46ci_1;<
zJOc1TH9A!srAfYttU*0r07KaN2wa5{mK(XVHy{OmA`wU`&=nO3mtos@^(7qXk9Fh`
z{b{{ct7GiWa-YHiR-LT?&3N659*TY4t(C_xO0t({9Bu9mCQ<;+=&%kdSF0=LRnUhI
zSq^*oS+CAJ?XX^~F{F^<)Ef<Wqtay>sd%_I-}$o~(E2aoCW~myW8EILgl2<r?gMa6
zjGuF=>}vi<WWL@ap!y%jJT3$__~iXYY0yb+5HnuL(Gp~Oz6(D|E?0&g^9)@%0dLBD
zsXC08S^hzIq6%e_(Fw?B>#B<D)yb91%FKpL<=VWQ-}%=B#;MMx9;I}4-4e2n1+po`
z*vj<D9=7u4cE0sljOJMLS|%^c?tMjVzuo)Uk0Q5FRcC%B(`j4pauY8y+D_EhD4xco
z?P4UO4N^mOP9(rW<&UcUKpXR&W)_5p^e^3~)5s|OM-5=d+-v{wxkLjD@a+JzFmD)1
zgRmn=et%7`8(^pMg{4KpauXZwy;v)TQQXCF?hKs&eeBkk7XTv)`2`#`POzT|>EH7|
z7-g`Y%_%BcS24%4{7&6r-Qb6XmCz-}#W)0i%Wc(G_WirzJB41SC2bRp*8SssC2sh{
zY#>@%Al|bn1)M)FE^I0yXs2VCQ>(|93-k>U1{)1h&?!v``_~v*g@^q8fZkRHh9ZO?
zLNPOnsC3qI^(Z9YB0T78e0m74T2#s;EcC!h$5_RqL`@^8dFiwwbJr~RUN;!EWx*Ls
zY!p4B9rwNL)0z^%aT`wR|A7?@>YJy$W>=z+*g%EFK-$5fUs$CPl&6b=)zeLrcJxQ2
zfhv%PATE%0OnnVObQk}J5gp0~C&mjSPV#;cmSr4r*8^~F?W!UcG;_)F#j55{8GsQV
z+I;xtF#8D@7TwP3N8dxc)>4EaQhUQU)N%4%#lLt5$S9-3H{q^P9F6=ucylyK?C3EZ
z$XruK(+g4F28dFGA(ZDsuBf7KT0U)F{<?=NPz=EjL4no_Trei4^!moLcbv}%LGwEq
zNVUG-^qPsQ{2)eaSnqJRH&mN<=bFsR1*4%K!HOZD)^2PqWqLfu)~b;Fp$=1ye-nbl
zE{1uOY%lKmFmP6Sgw|+i(4G^KL;VRU!eRtfX|`-V+4&olmqu$P)AUK=ssP)8dbTuu
zNI&taSepsyLGhJL$t0C7uc#Q!H!0f|bpw8os;Uf-?Fc7{Ktx~?!b+oQL?WgR6Q=LJ
zx)&OfgEY%Sh%G@4V;vc!*>1r<!y;pC?i~x@#RmF6_x+eu6a}Lgu!FD^R>A*=s&j6v
z;|tez>@;ee#<tPePSe<pZKttq+h}atww*L+Z0nu;&N+X)U!dK4X0m6k^<4LLGs;ZE
zuv#{a=5_Dk7<SHV*V!_UkO~!0NTdv4@_S?Gl8aA0<`){~>JE`PBZnqmx(BL(=uUUo
zP;};-LaQUI8T!Ekj~fFYEyAhCo{Or`O|Ova`_mzJm&WX0P4_-T4yB|iHVC9@P$YhQ
z80&5Lds7!IgL*q!^UkV002_-#1jVB9%7*ESg`xW=zc5KmDDXHSPW*b=0fK+zRbT;0
zX0Z@<nd!k~lTfbKBm(EuCPWa9oW(~+CtS-$uQ#$fP+M6bA9dPNTt9aaiX6cwgUTeU
zCf+{l&$XsW@unrz(z9jl^dK265Ga<mZKZQ+j2Tt7C71YE=t{W^1;O}?T(G^m3(0yw
zsWbC@d;@*xx(GVFba>$=%h3yK2QH|8Zuliatem&}TPzG2MvV!jqFPf*6YmZFX?%33
zKnx-UE%Y{4WyF=i086OucId`gJK=WK(|z5EPK+9XCzsZP>t*U4WY{8y3(IA619mR8
zT9E2?jf!hn)7G*6*Po9<F|hA^UC+lfO;=EAB-<!`s<;{&YV^2a!m|xV$^%+fqvhf4
zWzvfy!SG_LlO(us22&(lzu%=gvvTE2;6nQv*R@wm3uh_WP$VRYKqH}x6v(l`niy%I
zQkBWwSRYdyYB*hkII@U(_reFk#ganNi!g5Z(7?&<F}MNmr*!~CJ?ZhyQ8}bcBkjFq
zG@PEcuHdaz-`)hqA5#fr9c1g2Ze(hh=H0Yk0=s53W+}J-XyP$(=vj?3+<el}4m`U0
zP-UFk(dg0W;Ndkp^AP(Z;UW><#L=K?u)1Ge$-A}JACW-AM&d%z&agEoZF`;Ib79Z0
z;JM*34u0|FqxW}&=ZIh-)EmM0oZ1I)rC#Zx=;j*m{3E!C5YJ12Tr<x`^nDid0g=hN
zF%39au#nF$-1#nr<aKg2+v_B~e_s<l_G+Yf>!Pr5#;kx1gM<sGItu35(e9Dv!FOTN
zVm|CvE4!8z8z<ggWQTtKa8f}`Dbn7Uy~XjBCmD(?sQCCi{>uO;u51p9v7A11qb(8W
zk|K<jKUip+TJVSdT?6?-&gG@GIv@WDo`Z{Q$X)o|1&|VR>4mnyl_I4a&Um>BA@#{i
z5p40d_=9ObJhckXWK)+lK+J$w-T740ODy(Foxws%Fc(T1bk%zif?@b+E2D{6{zv<p
z7$;m4W$c#bqrA@!)FAg1k!PqXWnqTI?Od3G+fnkJ+=Z7~^j-}ytyY)JIo3<cxF*gA
zPp!@m{e!#L>tBZb8DVaQ78`mKly}IVH~s)k|H=qd-wu{J%+&R9hnSQx)&S=YCm^TD
zm%065bE|l~cpf@lTZE_Q)IhF3L(=Pdd(nRdxA!#MIb8B<T_q+e<(P#TB|#kv4$?JI
zh^zjNLG3_8u;dCn`i|Azk^e=3ph0iI0I|RO*Z9ut#}+Zol^QT2B8?R<&`Zu7L?OcN
zhd4+A3yA(*kh$m>5zQXrb{PohNda9@agbe>0J1<Y=AU&j4hr|b-mft^W@`Vr=T$^d
z!K0D8m5cRU3H%M4V(QhfUH9~I=lFc}czIjij5&A5hHRC5bcnTHRPFGE9(>2AR?OWl
zy0}OtqcQG}_o!oS#*KZb%0G`A`qHVWVI<<52sibw;gXW~a;_5~1hrAUipSB%qb)lS
z9ELA&jr>1vchnPDBwYkW*N1JB>}<}oUL}-G0)lC6&3D~~a_O*zrL_5cTz<}MuLLSf
zhKcnNja_t>(1cK*fC2RKl<d@a-WwS%JdKgoo>w^b@e@JBma1^)%TxVWfuF*75cJtF
z1QFk?r6|4R=ZR#hRbuER=EG6LxiVV5?LhiM!Uwpp!4*XxvRFamYTc;u+x=<<c6b@x
zLKz@$#T&QT4J(Skb5R{8)?b8-DfCfO4*i!ffB(;UYb~w)%~Ba@l~HR3Z%Lx#s9Y7j
zwEw!x&9KYEf;ql=iC=Hz)+SjrzH1xzLWdxZb>cjr_@*N5ChWECU1nAfU<qTKlxfz<
z4{!!m#a$0l)AAeVK%j_6M1$QYL}motT?(`9al*i>UpCc{04IqgPp31FY<c)U7l4|b
zTTe&OK-h_?RJ%JWS{}z-9MT{rzxs9FpYE`32&r+fyBzdj&V1imbyk0?NcQ6t<C#r^
z!tPSv4+fm?t0FD8saW8n{xbq6f*peEy30p(OYOdNy+}BoeholA%xD&A7eY<Vl;F&0
z=W^Kx{H%u!<P630&LB_)n#T;@YoOrD*EnSZMXsdJqJgoij)SMipAr2w81wHBO@Ggj
zplEFl2plv}F9L{|-tFId+{^e_cvx)uRw=lg!GUfhluDP}I~?G1Xh;>58j61ddDY%q
z@5YRFM^|b!58*!UwmT%UYY1lvCLMh|+~0Vh9ER~mc*b~q?7?O0KfbBD5Lkq%VljSn
z!vv$<fLq5=lIrRZ9cW`29mA@cyAHq{Jt9mBURympxD{s{KjV~#^K{~UOtK9xMgQRL
z8O^}_3+oYN^A%wP+IZ)y4*tUt_$cKE!p<|~&1q<3-(z2iKZ!+-%N1;6`V(e2bLRt$
z%luRw43kQ|n`t*rw`(vwY{7t^Z#bP0v`A*Toz4q>dHhS~kM~RqL7`@WnZXN44|!6Q
z`6#k)MDRjCL%3~Y^zXP+=od{Gb+vCQHEU)E-5G?9{OTM}=B(EKXx3W$OL+(}Pha=?
zeNPo%EIFA;P<AjZmRlniZYoEFC7c$*o=bq|A5wxP8Q+8JJEnNk?V3r;uMu1GC+sm4
z#wY)Q^^>2e&owTJSpr&5Az);%BgjJrZXRvE|M6N9K0bQDr(TOfEt=gvvt7m~OE<vt
zDMh+vXzw)j*Qy}SsIF-VN1Svt1H+@-hRibz0SssT5IkIc=EeFDV(7hn^z)x!_FST!
z?^Z$0a$LHA5r;M5FSc7W(HW%Xrdd^!ln<TTpXl0>=ypMP5VMkGb{3Tgq@(UHTEjz}
z@=-(K><VA*AZifx^?-NADzQKYN-q^yZFJcmv((3te7gW_co*#5mY|yJ;vtgh{F2M}
zqesvD{X#mJtt(`DFia0ybsotwS%(1~)zSA2_Dn>1D=69Si^P_QmVg#Uk1qyZH#dw9
zKG`fDD>dhe>^BO{y+enHo&MZb0&56N;zC{%FjeVX5nMxC#b@J^URgW%H%is~GtW3W
zn0G--m^yrB!$T~Xn_Uu4a~KDD2lo;pA7(T+wG@jutow|Q<7+xZe=o$%IhN!@a#KK~
z37Iew`ks#<Be&nSO|qO8ajGJ&ABYzuTWUscIEic@8(<<#DW+6xBT9-}y<QsU2VOGy
zD`S7MP=o#cOT|D#Bdf!wfu40mT4{u%RUcZ|y&7CghcTge@SYc~-xfbGO3?mL+))|l
z*{0n89ja>Wa8Hboj1Knou_Lkj$L0<r%g9UUaB*lt@Br28Z`+@_TGubk9=x3%DZ$$i
z6HeqV<Yn7V+0i2rBVpflTZv{N$0!id2)?$T)Av7$Z^%S3F3%$cHf8?S>wV(|{unT3
zbv-@AD0TCoNJjTGT)RwdI2RM(8+B1mrK>%%gI@L?_TvYO|MG?;cO70R0^JEQr!bc<
zf@ovxDa~8CRu2Kw5rw|XxDe*E&xh$&Su^$V96F#q$!jSW<HA|Jsce<dft9wPxveif
zwVvN*Hs99IFIdE(sM@WkET36-fRfeTD#$goc1#^iBj8TRT%3^p;`}?=A>kCva@vXT
zfxqePo#}EMrSzk~*<oTR31V!U<pjL4_zHRIdzEdNk_j!S?`y0r7|CLW+SiJ$nrqVE
z4BBh4FtI!}c%cc_W{AFNB=Sy!<!)SvS#u0|HPl~zpuKM#Mmq9lpxm4JU+je^aID^4
zw(r0zJicu|i14YZdKa)OdX=BdoIg2>woTocKKQx)l|+MQ_?6Uw3pNr4LXQe_)GSgh
z%k16*o!V{-1?PgSMTv~uz9v_DKe3wl9j02;iW)Y<DbP~*-d&<tCbgzkX_TG%ME_1u
zRRVjO$w<9MO}oxr7Q~nQ+3+_Kb^JtnH;cJ8BSabk*eT9n)GsyY+dU`D;qVs}9PI#I
zXK?*CC1>TnEgVb|n?h-bfbn`qvo-Q(Q0vfy1?CQY?op9zwbeb*ZX1HEcDnNHBsRk?
zq-A31XA4WgR1;BsM_Bqo3u{A{8mtwZJ(U3-g&IvYD;mo`Ce>Jl_+YXM5|qDKRHi^@
z<3%;^r8`h+FmQn(XuKFmIA~9z<wH1Q_Y*k)C-;%iR_}m<wzMxLaBgjvW`4-eoho17
zSu1kL2qg-gg%eE&pmcAPsGjUC=a#TDx(tuR^DoTs%oayl`D&!sn3lK8ic3&3C}iI&
zn8hZM40NqiWAn--BcP)S8Cl$=NmbL8LTR#LY^zlLN4F4mSDL<W_q2+UDk6#DuIJil
zV_NKgPG?z!2x;gPo&V{ME*i7$r8}{4S!t!S)wRMCQPDg~jb8s!%+>9X2tB8kuY|Tr
ztc;r#{YbpNV_TzXZ`C;K{6BK$e~FywuPCNCNdebFR1A-Tk&tH#6O`i2&lz-u()@At
zi(PTo4e?|h8}G7qr{zBYWM+C8rkbt#f7(MvLO^iI0H;HPV2&9KEZuEP#(1N-?4uy_
zcci(gOp#G`iExBB-x8x-kF#iMM$2MObwUR#e2g&*fCW-Uah3XEVp8S)KS5z&Ma3?$
zU5-g2oWl;oib5oa6mQlPok0JgpdG>4TqeTY$fKySXH9pM`6cPvN%7W@kvBd^^cWg-
z!MK|4^K#A8fdRvByA2G}!vCnX&u=H`CoqSM`c+0BMZ>TRQV=|u<vYGQtWAmQ9}SFK
zQX@GUrvHeePhLsk3+RS)JnzHCcVi3nQN(PvWn_t;5NQ!HF<En(v%j_T{Dc-ddgj&$
z3YIt#Ywh&wKl|TZ;x3Phvvu;t3mTu${v;TjuLQ*^3VtR;zu4_=Q^-|Y4m(dD=xU-K
z2|Gw>s;T{x<F=CXv^Mw51U&x{-?$s?;g#eXzOkCdd=9?vHWy+73aZQ1W;F^Vlg@+s
zbLGTFLIp|;3(p6blxH;%Rs3B~_ACn;MuE?ph6hGPN7~)Y2MB=ekl>1+@tv3TJ`asr
z{liPS_c_k}3`jkI=r3NMCJ=3Ed@^9Qs?+=o9H60!Md49}V=^RhpEnE30VNg`q^ySA
zrfSu6>1^}#!_Rjog+PwYevGc^6ZqB*DZm2D@V5m7+auTeV@ANJ_YaSZ72qwu2Fx4_
znf*Yq(TBPh0M#tIO(7!;GJJ@7sj;i8U{obJI`y4y)`p+17*$)&H3q}NJ4Lxpz#9hh
z&7`zlJH;0|j*P`Ev^9mm!RE#pq1U*uF|69-fI0aIq_*1IP|_)psKSrTfX~e;AL-6J
zB1@PN<&yXL9GnA(qj5ySL7(|CQo0Fz)9AmS{(Effkt+#OMNJ;Qk=v~FbehjWKOTU=
zjLg$oNVCC|Yqu?CHje)p{dn;HI84BZt^NGmcF>icuUA8pdBJQ|jY=9i<Yi$?!<r6V
zNk2?H^ZX9rsE`u5UIc%OHF~hWIB?lqV%OR<r4j6^tg$b$dh7kHF*GsyE>>7%M5@gf
zSwWNf;8W0K;=<>|My*|DGti1M#7U)jA8TmsR0xMW!t{&coAYAMZ*dT#zGWlD{kB5>
zvC%x>G+~9TDgSiUEa!Ay(ya{2`HFs)i;#t-&$abC{*y6VU%-5^1EBS|24>u2z%-n^
z3+O2_0t$*yi0Bv?q93d2)T#xFWKxNNY6W?tPM~~4sd3B8{b;db*Kk(7&Y0`L7YH|&
zF(cmP<=QM&W}X5UMmgom>11&vg~>oG3#g9#lzrnl)Vu(r$w18Fx^uub@>5%`^Klpc
zuvK8AnRR=x+ZviiwK4%1ipsY-9w`*kWn99ET}}U+$%hnk0tl^;lvlveZ2w5FUQ4?B
zwk^7P4;;9|RAQJaqpbT`30k}ZJuS^O#Ag{?4u(J!Bfg?>brfB3!xObatwu<9T%}5f
zZq2eTa_nL)hEi@XoXr@GT*lU5w$t;a#Nhnoz!R9{Ujv_xyW$$ql<x+pZWV)n;q1Q4
zJhWcYc1i5?r1>%|WKs8tZX{{$N~sE-)`H7rY>siVY-SCA0Az=xib@iI8RE~@<PRQU
zu~CBYzdz=GLWcIhStjh11TEwM;87<5(y{NqIg+Z{uftv{7VCT9ml)#9#5TzT&AEJ}
zq{FGhcQL0Plr1nuH}XH|+VI|D_DaPVSN@bVi*$aw$#1`2r1)M{l_2+RZR}_8Imyde
zLwNo8gyj8!>Q-kQx<aFUxuNq%WF>T}LTTGwBIcwj(GEoeY5DwX(RL}mi1g1Tn6HVL
zW3`Bo*E4=I*P+z7i+^f~`Ue@M%#Ir$7uFf}Q>>!m<{B%#Ayp*JQX}5%;@WSYgKJIg
zI_s^{DS`=_i}83I8O%1Qn=0EjX~!&7guqA;Zro-iDMmio1>o{j8Q60>{*yPlJe)|U
z==gmX_t~vN5BOB59{^VXbklU*a9@C4sz7=;{2VYD-=EcST2PKc5qI?gDoos0fXV2F
z$-VMW;31HYao(bkP1~yv1xN}PGvmn&sv$gwi0w@@$ujQ&bXCrYT{-TvJuL4{6xY>8
ztCLaq_k28vFIe*T(naT8Z)Y}%XL~P~JAI!eV2&%+olnMxIHDUh|A4Zn<gK@Jmuo=i
zE`p9478ZtggIk7oz9ZMe9tT7F^z;<P^m{jyf^<0vs0HGDAki8?w)Xg?`&%!gtc~A#
zofEVepZ7P6!5bhp7+FyLMF+G@6kF)zaVm2dH-pR+0n|myb>GnX=b?|De<RV?+oQQ5
zSQQvn9;9ZiZ?4<E)*hbO*y#YEx491hD*2ggz9wILkbILnXMS9NJam0H?eDDG15Cv5
zXUSH5W-W3=brCij9@wjMDh28a3T$!#;AG)G0S^mEn<a_=H8bGx=k{NFCf=euznj4-
zbgg~^bH;w5au}`R+4ivL-~DR*Hr19anO}=GlnESLxI_d|ScSbka8wJN)TTx~<}HUi
zT-2O0F*F~KxMp*$M$By7SB-}Cio$}G3AC<KwZC8GkYE-noaY=Vx{^E#4Ub@8(I;oV
zy*`ubRPHlEn1=|w2PN7+4W}1Yn^wS8nkgvMR!T{ZQ&=p~C1QGeh~1yG<~!Ofw2Z#B
zJ}Z>N`91JINo@j5bbc~zJZfZdfC@%=Y;g8((N4U#d@Yx!mf4!O1CO8D4Y@4|bEiru
zIxmR%kp4l=Tp81iwT$_9NEEDX$m7%y;9vAIeRQKfDV1V6N6cFS1l$M)4NENq-Bq9o
z=*;@#D%v}ty`87{D5%Ecqyiv~nzoWKbzFN)syv2J?EpXs9zc9(7;{Y*^;YvyWOe|i
z^BcHa_N7Y@7d<7V?fWU?KgPnV*-v5!BEb9gFM&NEu(wV^ln(H=V|nD_xdQZS+Yk^C
zje+o(aI4l>8>SE>vYIHn$ke@mf97pmthdw|gpCBVV}kBncz?JjnxpnmWdXm3d690_
z`VbQ3lo50cj4*H>M@L6;Al*0t``bUfpA|(c0NV8F2L)jS4%_ICe<8G<9zeACoh(Lj
zY*M08$T|LvV!Tf7dL>&%c+r}IxlFeJez~>4z#JgJs*AT?4>)RpLWfa6G;+v5b~|Zf
zJ2|?d;Z1Rr+EnQA{Q>uAqchsFX}@wd*XoC|*dAvx$*+IAh+}U<1s)%-Wf%MV9DE7q
z8{fz1g`upB*&dz8e_TIvXOkMOp_wabG#Xmk)g~Nz(-B2V5C3xxgM(R<`W-BMx1L?G
zP|C=XyGvU`B(DjPmHM8A2KL?Jbhf1nhn{!7f|A#7fvLbcBAf0xw{byq;$veO_IeNA
zMvA)u@3eF5%j|a<B1XGwR=3?14&uthpYRR1C@7}RLVn_)R@4VxSYx)mADHkx>L!`!
zUSno@Gdx{4licU{y<n@Bv><cEKU~i7*ExYT3eC-~zRw-VCz`dv?b`!G1%sWe?VP%u
z2KJL(1$teqcLbi4F#^IWUp)XN;3`rKq5;)CHa?yNuhP#Qd7Ss0+VoGZtCpWcut1<~
zU=o3aB&RP~%5~n9=``@$1_-iuMj(SQdecByBHB>sb!Yx%xu4r{_J|mC2tL9o)W7J~
z1CZf&4yJPYtmSXRitU^O+{QD5>r01Qddox*NDxqrAS^!^lVQqWD!2O`f))rfkQswY
zeFAAv0=O~2=zOaqgijVLHdt;p-7h48W^8Tk9i!!kXF;E>L%>qC_Z8x5USRdpRVAeW
z5j=$NJ`NkFY~6oT7$_lU9ECUcL7CdAA2hCW9Uq}6n442A&{}^??*70ffB(-t<aR+V
z;PjM-2)Tv{1`@;p-obzt=Vk8e%9H9Ejej@ZrjQ)hIpa#6g>vNn8i7}?O5EU)czGj*
z#}Dl&4M(gT!O=4|cD)h}4C=*D<RzA}d#;^YI)9=VLa-YZ2Kn4}|69v7>CN=F7DuxU
zQK*My<yNQDVhbe;6YWdUO1h1Vrt1gu7e2FN8En>^Y{#v<It>=7{R9J~#LW)&mQ98C
z<ov;Zg2`kJ6{RbOQ>IZc-T!)GYBnek9P<KxFs-VmSkS`2Ae`z11QB?<t$%O7;533x
z<x6o2q2pUI{tAaAB_fS4*>;iol*2s@+>|3MT~+x2@1AcN%d4#adiz~q@OFu?lwl@Z
zNQD%S?FKR-Bj11}%1In5KZ=~`L{FH<6_EZ>3y_j3l>?bMHz?%X4Iv3oLs_3#{s|@2
zcTAPSfgyJlrC!hUaQEc0OZ{0C^eIa=I5neqKg{wwyT*Xh$Lb9|oP^KoaqsMC?$oRC
z>wcM6{DDYbH+2^RlGFoF-`qXWJ714;0JA<y8J<!k5>g@CbTs{|^`JiaHIYSov9^{t
zwaGDV4s3GFx^PKD!r8bogX@u1T}KV2DMsR(Q*6=`qBKPpD8Ux`pY@Ck+M~I;w%5ZJ
z4%W->q-f-C#D_@y>IV#?$#&EN<r)oQVPWC(rJCh-is(011@>`w8a1QrWc%n2iX~?~
zvw44ETXihc%Cy?tb*Kut-kgYFxoG6a5E3B_ki#Io9=s-_V_A=^nWhW^n>b4c3DVeC
ztPEc&km4>d-U+2uC*z*R<_e`>HfPDpB5%(MgO`rw%xJ>l3=e956_n!#-m1_!7&T=!
zCW~izEl`1by}FfVxG&n3-qIyp49X^;)4<Pfv`n`UI-lennqV(r`5IG@=7f?tgE;W5
z8l}vOjA(*iVg}uBd3$s11fcB!iz_F30h0_nF)h*GJ3a#c^!kEYcp$QOG5l0GGp0cd
zK-~ZUBGZ`b)yva+NZ-#G7aqQn!TYg~fW0iPTQlSlY*k9viIw;t1}33a{jtN<ZvYd_
z@Btx-9DpL{K^6X1k=b(sM2Hz}>Zc;uRM7o8Lfd>$Zl}o3g|gahgOIW1U~tuHdq<$z
z(I|Qzu?M1kqLp{m&*%Rfz$4y(J&KcD+=m#2w~*rKW%l2}EY*tq2{wOSHQDD1Vji|8
zrY+a}h0<R>^iv92R#Cp5er6Y!^{u)ucr@FlwJi=-z0SGg^~Co>$^=mhzkY+g_HYuX
zzUtx+B_YUOx<i<uV`r1S3d}w6jvsEip_0~^qzzHi`R<%3K3sxce}7K1BwO!Y(ji~T
zykr9(siJ$w?yx%p>1?A;McWvDCiDn5+Q(sSAS;yK)%8(a73kQFvb@Ax3bsCa&<VX6
z3+ut3PScMLmqd$q=O2ZBb@FZ@&{?jMt4$U({&!7XH&Nh5FQA?S5CN;v0(EGpx&^uz
zE157H+2iBm6Y_iF5Ob3SIH9TiWDr5&-NslvT3pj!na-<bm9$_|e~IdMnhQTcH@UvZ
zbd;X?=3o)~HeX4BzY%3f?j;83-E_Vli0A>o4^mB9zt2RVe$~buL<_=t0DE@)oSV`>
zATVprFOubkf+?=YFFys}5ai?}@ohe%eyckk6=m!;hrrm(6Z)3rTD1Hno&^PY+7+LK
zx32k*w!;J)y&858f(z?tUnFd<pO3me=kzq{x{OR1ri`cdy$(94&D(D3m5}Db*DFt9
z<S64B!%97I1_*%w^Kwz1vRKpYZeM1X^L2@)(vHUDEZ@sx+Kp|(s<v{C4o3aw%5Tr3
zs^LS@UCtfEo6(N3CLF7kdRM2{kc}|IjI>wgVcuzv;a@SVS2IJMdjt+FkZ(*##M_K@
zyn57`O?A5@WGn3vm&jM6El|NU<Uyhv!|^cW2NT?qw-~rX^&odZc&bEl;s6X*;rGM&
z+4@U0UVe)HYY3u2ou4UdevVoZ-8A7oCzWCa3_mSHqL<Zdp_PhqIM`dBw;DmHULRfT
zb4Ov2MROlhSI#z(7Y(W}gPlr$6H`_7xf{`Uh2Q0k{S!>du}RlG5o$aM4q4qW$<kUv
zM59BZsAUwuxL&FqE5HfID9+`NNOM#pO%4Icy2VH^F`SF$Y7`BFRS(=06j_#jWUO}*
zRq*{!l}q(2qk>-UE-oIN-M0?Nm-@&-M!K04vhsMm!GEDX`-p+)*4674Rh8n{=E@}u
zY3vS-w3;<7YiFa%kj&TVOw0;2_+HP9D|XY3cMAc=iq>_gsJd{eSoDA%GbK&X<HRmr
zk2qB96vN+Q;-~nx{kTWEpQ)UF2;5Fk+8*t=Vcy<A2xTLRk1-Va$xiQDMv{>qZSyk%
zst@=6cHymBL?Qnj=jnGdfhF|T8}BY?j;?mi`BS$PmYQD$35siVNk4I82pI^DZ?KPL
z3TYKt<O0TM!9<q&7aNO5Pg;CkI4|TR<W;e6SsvIPRbPrl1|zOLR)D!fR<l@pkO@LC
zz}f7otI-nCvg`aaw1X%nd}3Z<f-ZuJ#O==uuO7$?Ax}SfPu=C~*%hln5x@r7sZ_z$
z|9KB4?#gwF9OUD}hJq>j73x#~BUE1Oa@{~Xtp1Ns0nfVz#~lei?*;Bk;SfG^k)eP2
zlD3rJe5*3WO6QHhll|ght@PoUe^Z`UrS($vvGbFDRN%_>RHw<;e4w3ku+#mV_QU;#
zgi5W0U4PaZC%bxxo?EOPvmNH|P{#HlWPmM}Cgmc%y(PV`7(Nsh4O)$m8pYGuvm3VP
zS8F|(6G|5~3SD})FIYC~$_7@PD3&Hp+0tV|_FK(tv&F(e88aNJ&kg&xr*M9;#vJ^N
z#w3u=3xAtFAvEqnFkDW;Q`a-*%JSGrou2P<zvSVr%7xV0YpjFO?lORlk~QBUz8{0D
zJjWD!rEpevAQs-GF&bjENQl62j@<RLEV+CDVSn`jA!oI5-ed0DP8--WlMqNQ?emW{
zI}WnO;D#UzN*UzqM?8xN*CdppcD7_EPKEKj3)R|Ae_jU0GT5_CBli}ESH)RH7s>P{
zM|}aAWhD)6Ws!){Y(h6`%6Uwb_<Ey<%39r>xDe&N1~amk3!|i)kZsc6KiMFtq|Jj6
z{v;8fCqvl>CBR0Ob$xFMQi4ZgsenUMJ})bimuOg2uDs%XO*3~YE!cs-JMybrxs!O3
zEWk;MkvmN<MADJ35KrVQDKe~UED0AjO1f;--<abJLf&73{8?r0CA57=rJ^Z;wwbJ}
zNwQ%=o2Xt^PO^M2Vs~vNV>nJ6*}@Rqd905)>)-fV7(=g9VkSgA9^j~eWQHfY&-+2t
zzM)ktJ36m|?UJpND(!-F8c{0l)8fY@PhR-k&!kYYxVVE)1$&dM5nQ^i07ZUV<^5JR
z%GSxaRRU#OcXFkb(!|d1cv@x>=*7P~IIYdkTYTY-nEANq_WZ2<&yT_Se-KW$?6-kJ
z>=suK9?>ABP*epvD<ch~jBh&cU~gdjMqf6_l<Au<M(bNHs6m1On02Z{qfy8_(&<6@
z?|FF0<D`w-9tVFw6z@&kv>wG<?&wKC9Rrdl|GkOlJBe|H{ZH)07&!98F7$=#wwK#b
zXN!o`m&#S=2ki}|4Hsq9<_qfewkIX*@tDEz04xr#u^6};dc-bmBgPG0ujFZ9&wX}&
z^u_Sp{jrOCu;L4%I~)J`Qccx<u=N5@aeEt&<I@t2@CjQEQHIjfR|TSy>)MNa=G$|e
z8>qja_^}7jk!|pQ7KRIrGQkMY*{R2F2wr&`0@~cdO;k7%Y_a?aupp@(%gbz$6?R58
z?k&K6@#yq;yjm+A{ahhOhto=TTr)`?k4BjyHKJ0yIFApu%VC(WP;pD{Y-PHhV%}HO
zDSzTwo-}+Z2g~XctVm`v*KjN0aqLY$3_O2(8knx)Zw36#eLB2~*$ZJyM;E_#58~!F
z+bD0>Z;{NE#M@H`Wm9mzn?0f{dwq;UkHcq2;b}j$MK$qAV`BRzG?Kc~mCC8vp``tF
zy~**di1M>0D!5BK)9tjaV14kov}gQwCtAv);B`G<aD3}K*okk$bh+h*Gzb>3gnok0
z8I(*w%Mm*@f6~y%_p7`yhP<7jr%|kRYS|r)`@!D|$tx-$*<xqkfi9LQob@09bVikc
zTi&kl*jszNRA>2-4tV^c7Rk?Z6aN~0B>*$UB69BUT~$^Ta*w!YiDLt$5xtylQ*+v;
z#{Ltb=GGjn*=<qB?{p!}y3DBk7Ua$U&+%xbG+blO(S@afnuq<!yVh~RGTHdOb$GSq
z^7r)X&3c}C5kXc+#pRUyBmFPa2)ipH0&7OKeOpoA5OnvCw+BzApIc9K7ac>$PLgJH
z1Rl?kwAzh$EG6f*^O8|KrUcm2UC}41^I+_ox|-^&GnO<SBV@@+_>)+j;dmoQ7LWAy
zr$vJ{nD9ursyVq73~7%5YPi`FalJM`2J6DIu0wd~Tm4!CuWl8})NUTkel1jXcfDBP
zbln$*x>78yGOo}6w1-wKJM;|7WTP9||9xC7NE2qd7VUL>akRbey?RWC(=DPp;nD2`
za9AS#z_{Y>0^c379q6rK^N_tvRm!pB{LD0=xIiMb9_xSeYcI^cc;^Tp!)+p^%wp<&
z8%h4w$=4#JuuYI&CeS>%5C9jpWR52g8==iA=RCbJebA&u3GcM2n{>2pEPYhfh9|4>
zU43Is<;aGyf?T;j^HA0Qd5msJhxFpu4K_T9wh(r<MbVw>)ThHR`0T~vZ8HnQB{qo>
z?DJe-x72vc>$U7Jkq6})7xYy@6Kjxw0a16|=h4@oW4J|mBU|!sJCp8;V5=yL-QB--
zem(lotolg3um|yHNxVD9At%D5y;kFGS)#23XRE7mcL0~8bRZKzyD|5g!WS$tHg2h&
z6I@pK$3RL>EGc940>(`eUaf}ym$7*_ksoB|Kn;v>oEmKdg2)CNZArqFpTBu2jf)NZ
zb2mk8$y_by+iz)tjr(c|4tEK~AW|9{#o9Y`h8^bPsqmh_xH{pG*j^<3QfthXoH!U&
zfx$QyTU)yx^o6#_o{(B1(P|$3aAbw!u-XKIwL-<D2-8BrDe16MbW|+vRct(<Ek24{
zit~@r7_4GHOSL4JwMwZA!Vi@ePpjz$yE>7d^xNiAv|u&L?aMhN3hN0jk#K|F@}z$7
zOw6&vG;V%xq9eU^Fas=NvVuLkP1aQ-JiV@6tT$-i9_)EDJ$+?nboXzdv#C5+EkV8M
z%fgpN&=h|qzbZSdzy8dDu3Vs)$*ORFh}p{a^qT~YFetJA!uXVP9`ggpNkQU8M#d6N
z&m#-W;jF_o{JCRLeg^md035|!<XuI4!?eTg%alZ>6JX%43p0j_(5t+xXR$Clpc7PS
ziy{)=eiSs9D>*N9?_0NI8r|kGUH?OhwsfAg|E*Evc+nh**LX+N`p+eqjasdK#1dOk
zzGR|OkMAE_Gl^=!4S(-Ts~~ui$#@#Yh4*s<?L6I&vC&Z{T%`>bCC$-PST!v7hjpPS
z`Ea2NosDmE$T~qEQSw4O)4;K|A~t+46dc=%=$8YuGp^pCsSD4SP&Qwp0g5iZKO%ba
z95m<<B=9cBS!(LMR;128KE@2Eu6k_iVLsENEK-&}O(mjewy!s`2tx881^UOiv{(+T
zS85P`U^QDAzQW)&rpVk_Ur~|kyJz5`)v8}T&qdeJt83?&_*mz$226!IwuA`9c6_V<
zx|qro%rf&5(n%kNPM$qWi2V-2Yq%4Pi_apfTcouIGpMA-l~aK1AH$*Ta-Hx<?f9LQ
zkd!PLq(OfY7j}E<&*gwQ3Z3mqS9hC9x9)b1g8KNa=)?k^*YqmH|31m;*_w=NrKZB%
zX{E79O+^YGR@P47zky{84@~!ZUw2>gypn`&6IIHOl&s76`4uvRCfr}_TU<jVT4i!N
zU>t8y&{u>?N-DI!=-+?xrSeM^9e$J_hLuOBRg|~)XQp?rz~YrOzF`ROP18NZXtN10
ztrAytu0)NLja2G28x#AWgzy7r>rF2RX>X?WIuAS_5~+SB9}ksTkF7)+%r2hS^Vrv&
zt%Q!#XUGhkhlVS%O(GL22h16tQRo$~m$OXU1MRG7&Lu}ouGQ!}kZ_2<`x`%2@YC&-
zy`5FrBbclS)X>ap5sh5=A9dvM;!wij3ad#;+nh!=X<^5?>DlmtLhU0>09=b2ENJBk
zp$MM_uBZ2%2mM#bQe#~#v<gPRH!#<}?4PrL`2y2XxGz!qBNqwEK-#d}dU~T<FgZPV
zx5eIIK%4n@1|i)dN-&$~ob{v(RRw-|l|KYw<*VIs^z&9ZmWHB(8udvzv8|RmaZqj9
zb37UMn7ZkY=TDX*TXHWBIlE~NXYt-AbXxr$4$E7Zr7F(UcqWFg>}g-{q@Q>A(TR8G
z&+|TR@soq@(n#jrsA%md0_0#T%dQ+@!-QjGV&&CkXKu~i5hAT92Dy}x==dsO#@*WG
zKM6$%6et!K7Csi~mx7zli3P*8Q}#BDojAXHjL15Q6ry!`+be;#QJ0k398HLDZW)ZW
zD^j7UiAM>(wD;WS?kf9(57!%vqh3Di$z*0W?>szy-IR>=m+lbtvNbjYa^`<+XD@qO
zE9cm<jNFqwYQ4G!lCUOx-WiO_Sx;Ns*Ef4dC~GKZyKoy{uI}5iW3+=2;^uzv2tk>b
z3&q$H!Q;z_%YYv0U4}J;<DEUc-}hYG*62Zh(~8t}=B=4SqF47iWU8#oM~yJkj=8q6
zpG#Rd#?>KTa~g~5Ib;9zfUL3174kdV<65!_4Na2<jlriHBDVwD-_}p^@M&M^_xYUo
zC@cZRHIeQm51OI?PRQXmc#z7@$6NOupFU7Kr0XegZt0&$0Vl|nn#eUEy3#t8Bqd5s
zeOixLPmRy5l*v=?PvzDTz1c?B#W)+l^Hw<$Axo@O4i^v`J~%dX)9E}KG&>uXk|tra
zUDT}5q~>KaTNY*&b-S<;^F_f|&s+KXUP+mOtl4q>%Jo+AW)gu{MuI#_*;0ADJAs>y
zm<OBKB<GdH)53PE!<BRi5miPr<kdLU`g@2hgZ6EG{qW4($EwY0G?%_mo9U+uM1T!u
zTg3r5w*PJnDVST-?*?M}2Nqe>0vBm#*A7q{jsgz;)R^#Ydvt_J&ku!#+r$Ay*?Cs8
zW0%+0nP>@Xu|3GFeefbgZ)euSO|k&64yaWYeJyw>wwrInAQ)m_Rf)9p+dnuTs&TaJ
z&$bpCC_8n{uSN1UHSc2FZ9Z;{nc$o4o&>b2o{$cP9`KEsT~qKq2gAVr;>u-63v`ie
z^wEKpBQp}iVt1Bv4^;u=Qa+15pY|m$1%>wgCbhKOc|6UA#<8zTPdYCDoT`?GpS8ek
zAfF!q&OnC~@iSECL^DqA;7UNuq2<W&cO(vO2@{eNEcL`rWx1Kqx{6hkKU(bFob@XY
z4mH~HDr9>+UoydYHxlz^&kRy3XdFwT1L?p#2Vg+ia@LJz_TW%z>gvAAtx$rCM><M-
zjb~g&rq{LkGDgtE%@u1-Gne$Q&UwY@Ly4oWC2`68+P5P1qUDXc!BJ!rrhwdtJ($N#
zwlF?d!P5u+uuRdoCSlx5$0QMTsX=~63pTOBf_kIg>+`07llYXuT!Ee1E2aCtPp8!7
zBX(}7p>xht;h+>m{NTobxF$Uo=9znfU6K>XvBHb1H+?&3TYhxzqud?c#Xcp>548m%
znZsy?mmb)RnOQ3ca4_rHRWCV9DgrqfD-H)^*XZdsw!Zs8j|5v@vrw=TPVxfV_?}<x
zSyJ!^*;jd=hog9FAOdvNue>J5Er}Sutq9VubT(L`hzjJ;KI42vZWjBcJDlH;%9=HN
zQKPH%mh$wnbHRhT189P(*&k1~@xMYLt1Np`x<>MJX6{V=DByd#-SPbYt}JA4*gg{S
zfE6<ewNmncDtUWO4NWLK%)Gz6FWJrCKQ?2pPOacS5kqd|&1neb&-6Dy%_cJf*-Qm*
zCTm-?nbEK0AM$1cr=&_;Lk;iW>*`Gx)JK!L7|ggdC1=Lh$m)h#;x?_TlulC{Rsd2X
z4#fAQs!QjRi((p`|K7EU4$<a20+k<>wRp;#UefX%%mK=v#MHIznoR8r0<`l$o2Q{G
zZ>afe?+tX<l976k{of}ojoKL})~x=80zv8Zudq<7!^j~)AW71GNp>&2%CsmNDrz~n
z;PdYs!%-5zyEFlRG@VS!QchkqK)IKa2_h|s3^e{bl6`BIRgUJ4(S57L=+Z>I{1X(S
z^zec^L@V2YLh^EsnmHi&k)VeNk?pR+RfYA)w{s{#tkKQi>0`)H)U`9wEe6h6R*0a9
za*M!(tX7TOa0z@fBKlLjXkh_OAt^+PaJ3Zb>BP}_D5|xN*aWU`k)EW)SW^AHyW|Mo
z8_L(i@Dlf*VomvE75nUQo5tc)nrs?$u)@kdC2JMUX%G1Z9UKs8m|%<-K;41k6)3<u
zIA>b^9J6M0vddAq!3izeSOaf8Y~BbP-QAmVK1KdmL6*hkI?+Z3bxz&|#(dlsfo+<Q
zbE_u8gg9Wb!V)<yIN{rokVoH;N>6RvW{5lFJp9ZwsRe700E$VT1VYUcn26iB0bY^~
zIdPAnK*|-$jc@cSZ-1_?>eD@kaKknJtw66h)la#VHt!=?XA|){y-d||);E`%pE2Xw
zc;KA0RR1<0>5j6@(EY+Bj)=#}=p01a8P1`tJpsB?Lod00R$!(Pzp_E8q(MkGdRLt?
z9;HkR1wk)uOu<GxKjD|kN@9Y@1ejaI1buk-?YM|bQszun=pBF*uT?^$TH$|wB3Fpz
z4mcW$wQ1Cg03Z4E3TMpYBFJPfXMcuHl?&B#=h)L3Bl(Ka>T$$q0btQ1iW64Nr;}c;
zAKpnO%^PK(pCPewd8ru*{97@lFx3~YpH{`R=!2+CVa}4#Vc_vZ@C_2%2FC6#ZCE>~
zT&|b86okkZhK4ZnP`(KmsGHfUq!*J>p3ladoNdQ?EL|%_N9yP>1%}+5{{VxH!y1MC
z><J((IgHXjWE~62)78W&om46fCRBFTU?P}N!f6<`>ex@xT<3yXH$rlO2n6&Oq`NBF
zvC`FL|9!uCJYPT^EX+w6Df~(}Db(#JzodD#*dCa~l87Q<W-=&>D`m7f8yBFHkAX3!
zDjjB|ZI<y<?~bt*538?--bs{^ItAk`iUu!ZmjXeqW{<HbDDSe7RjK2T(iBlJ9&vkw
z0s&~i1;7e4pFyq|$5=^gmhp4Jv!(HVN%*Wf0u|1oM{xxrg#l<&K-oLd5P^5Q=7pV`
z2e6^t&@&?^oF`kIXe%#5YsNd|q}TaL{NUB3(Kx+b-F2Eqky(eUBZtGN>!SrUD@Wyq
z)U7bZ_t@IQ$XHieKdcTX$XJ;fQs${l-6!t^%ZSQaa>U{Le#%;4Au@6=qrCx@KsOgr
zDrUc)G9L1PI!rr7-v#TDU;`9juJRGy27>A_R*frzR^j7Z)RxiXX0utGCi_6QKyuyW
z_px8KKtgyD8JNThJw&C>HhH#2WGT`vMN`MTJS_bEmwy^Dir+Ki+~2KjxZZ5XY_-bT
z^ynZF>hhrHcnL2O9&G{p0FYkA#mPxyfEVTLUYKGF5I+=pf4raf&u)j%15^>vJ4V5M
zTJ6s$sO!WDh%2<8GG<44<!sIt%96(hxw$JvEG()|kLHToUZL@91upFqr)0i=rZW);
z_}WMBE*Va5<OuYB2YvZ=T#mhCmHURr6L@Dhp2GCs6g-AVm^wU-QP0m|LN}fLvb)$F
zkgSv$=3)DPN>zY)zxKzUV&UK~^xk=#Pf|Vaz@2eCGx{*~hNDWtx-9&1Oe&CgMg=Pu
zDN-MKyFFd}gJ1=)Jg#HC-p>7~tY`+_r$@z=9_MZc$ow;xZ^eHul7D&GO`$%%Xk#rW
ze*oqsr<a>g%@UZfc|r!vWqJGcD9t%XbWZpCUc1_IGOarC-=XLqlR=tgz(Yd#2SqpY
zwqIl9xA*Ogc$+HodF8>H`Ng0AMlatRW#Y%68e+Yh)M$RmtP%cVpi2U0l_SdFEd0a+
zlhA=cKXW<XL>imLUO(E8PcPLn(-Wf$<BLbI<Y&7`1{E@S_A3C=Bm<LA&HrX=M+NZ#
zuNlqH*JBK0z<~S0ndH*w<{hxeaM6HAP#{nqp|I-G6S$u;9a;>trThd-m{!wq-3d%*
z%IaZ$LLg}chVbyv!UL$Nl<EGN`Y31OF_*@&So$f)myLK1<4l#G$`*B_Hm>==d!c5E
zol4NZUw(3&f#d#*V3~{oILetsCZo}>E>^q!p|gNxE5MHU2JoGbG#m6-GH7eq$?$Ls
ze$sW7-_(S@0Hu_eLxNkTNOs?1<w|(~H;&f!u`e|Kzwr)+_DerJr}>}u0OO%@58Fe?
z&;JXkTt)%~@7(gZnTU!l&>4pAc0QUNC@j>v{505Xu&1l5t1D<}ot}R>Q5WEBD&=36
zOks4%dN)Ea1MET02pAxOaxkiAaLmv@A>C2J?o}VMI8}M<ROWBhK}Ganr86?>zx9t-
zr>j+O(0*@RditJRP{4-^4RzJ9-aaNYpuAK?=tyAq#b|)Bw^c>Zfg%za&2I2E-~=S(
zB2YOTc1Gj#elPIzf06MrLKZ&eP*ab3-xdbW$4&MhE|Gxhk{p^}Z|TCL@&S(Fq7T|X
z3=paEQ51Um7K*t`o6Gq+7=jx=E6SA5snE`^KPfY!2ARhK<;}!Y*eit{@`8x=m+=Q%
zg`%Yf@g`q{KSkUHIFHkx{WP=U=J0QjG!2~rCBNP%X~LC!yAunAeA|kkRe<wcyq1g7
zSf{uB@G4bE8keahbOu08Z)-#Yetv!8Y)7;(+<ea{S08Jqyd|f4SgKqmRIk0^ul2mm
zkDjNcF$ksLeT#fVv_*jqw(jyQ&_XSuwD~aQ{~XimejCbO;zU{?XhNT<!74=U(|ond
zCGUd45j1c0zkbaBNMQj_;+Uk?UD-1YJ|m-=x%LNB^pzBQz<j}t$lUq9695m2V_pA<
zjN|<Z3J6{u9=<h!vu-q#L}D@ZX1a#3bNy5h>{Vy_^8FuSa^l4!tATaDC!|ZdE9@Ak
z+w^pP>KBUowdUr_UZE?XW25Gp8^e!qrxbigGrIBH8Rt-!xPy!S5N36rCF_}*%582X
ztpX&p3Ga6APB6iF<%`dcSz&5dt|f9-YM|!PVd=HY)$SRn1k!Tvz(vQwrWLD)rmLbr
zcf}i26R&2p<<GFUx6fb$lZIJD8zt_bcoq@Bb_D(q;m!s(s51b|+&;aY?6$hFVn6;X
zV5d8GJKkt(TD!sFz&P%&Md8EWF<Ny27E2*WzO&b}kEj}@<dNN_er)C&3%m=FpYIen
zuvo4=YRR<4WT%1s>)mmoCwu3bdW(1>;WnE|M%5*Q6fvxGLcn(RXVw!^Pj9(eFUoC@
zfS?HlEZSVkt4!viSdNSFxzsJHZ5W`W4qHV$gh`2ZT)JK|9qMugAPBLv*1dFJ`#kY^
z!K;!Yn~#y;G&UIsFcG|Yml?41h9Ay;eLN?>9{upPKiBGqBD92rjL-!y?SmBr@ugS$
zH5B#xF$=ZxBg?P%%UR?BzMuQbr)Xte%J$c9KwqOZ4!I9;lEb4<_j<0_iNA?*+s(8d
z5Rr}(K)~(iTdXL@H2}-_-Uj3Vf&?azQ<~_1RpPjX<1HT&KE3(CEmfx_8n!54`wl#f
z8A-k~k66rUJX<Du6c-mWH!j}Ki|4GUqZ-pw#V(Oo#amm=?WJj7lXA;Wy~@Sts$7L5
z%>s790PS!>EO9>@Jt<hNz#91C)^#wp_EU+iPo4}DOBzp*lpGfx@{sycAASgcL|doS
zM0;I!0p>NmEU%Mom2n6LrXN`*^Skn=V~T=S<oFlQQuVnm8SJ2@JbizM2yD+bKoO(*
z^(+BGz|U2n5sIJ%bgwUxRHaPy6*w4{@4uZX@FeJe<1AB-iv+X+?IlxNYia=<pYXjO
z%-7`P;fbw{Du*5^H<wu&`?W_3G}@lH;_qO`5;P76M@3-NgaW|EI59AsW_Mnp!)DP7
zuz7sGtHIiEwW}3zTbvJhj<%c~REd$wp9kA=emXP36$2}?1;|D|0FaX7p7Q#~xMS%H
z^F*k~dd}DgApi{vOe;^-K>U}H13-jz=_1H`W*dkgpw}Zfjgt6~W5yb6#hTciVwN<~
zx<fYfSSNd#)&ym}iJg!|5NqCOBY45^tX0(Hs_RLGU?c#iSuz>~N?!G(D-G<v$J2f?
zJtuB%K*tSHx%7@m3~Mj49~szK-4q`U&##0~+fbkWFF#Meh^j*j{8xN;hTLp@>lxQ&
zZ)`-gHSwxciFe09MN3oJZcLCy4~7dDZulJ0Qd`=s8AFp0D4kppvrhw`W<6A4@_Fr8
zX^4Eyax6|9xpc#6kJ94FYb;?-8^LdIB~PLlZC8wvM?US<ud{XwPitH6<eN*G8&=Dq
z!TxfW3<TdDP0=6X0knACDsc2%oUwE#pbSo?nPTcIw6@ygDckGL(a7N()Ws*~FbSQF
zre~lIx~vV0=*;SuYixpFPI2p-0=ub~;D7h@+DYf}Krhz$qy5FpcB^<la~Zx`q(rcK
zmQYVMdJ0!Z_42!sV8PXXUEn{U2%Es|yyY<C1@j6D`gl@A$Mo9ig*`5NDm69Lqg+VK
z{S#TC!LS2Y)2{?p5u>cl$Tb^hQQYVMMvK2N`uTYdT19r;0sNV@NvQr3#VUkeYIiM*
zQZ_-OXCZB4D3CBm0P;Duz-nWxUerJvR9!v&VC5)$a76Uji`C^UE)CO~D<0oB5yzWR
z7BT=4=K->?_^(q3m6-Lr-$1Rk=jq0qme`943@bQ<1|?ei(Ne7bxA%+NyX)g0C%e`S
z*ZH;57kDjt!$pBa8pgCRk9qC{PZCGdj&+gb!DMbwbHz6gy4Jpc)orRg4hr+@hIX-i
zv{!bt3oxZ+bOxd8cxa)BT&h(xcKuC@%~->jzw<CGEpJ)w6-%zutKlcm7dj>H>e<sM
z(n@d|Oe63S*o0K>it-@%8;B6%>tH?^?CgzE5Y!;T5D7+*>>qLwdfJ|PaJA2Qk6xN^
z*doG%ob|W%CnYySL*9rVvn-jSnXm`IlWo^{JlPX@<y8&>t;E=16NtVPMV)juvK}K&
zxS+Ysu8Qr^L&O}#9a=3XfwdNgzVC4z94HVTuf7*wP9#Wed@VEQUhdAwuzzn2Bb!db
z6x7FDe#VwPW;LkjqHSR6l=W=;Ra)%-C9PZ;RVxJvAR=E1`SC;NLW!zj06*;|1`wNR
z4~0Beo`G#Tzkeq&2hV7SYm{U@xwPpJir<yD3W|Q?7zwi=gpi+|P=p5sPEoI=SU-4Q
zH^5T+P--qFP<#_MH%x=SVB7mc5hN!KnIK61Ks!pXO7lH<%;7N1=0Z#t9r@gUnMZdw
z?{^J5bhXnpoq2^?v(~hn{NiErNawFn5Fqq)W7EtInt$7yxgba=3@kHzWEUS)3@nsX
zfB=YLM=+t>K_US&v!n0t=oq_dviH8@8*NTeiL?j2`ULj1bs2Zu#zJjoFr-vqR92D_
z3NjVy|LwZ4z?=fQ)8d@$7uOD@5AQwqs@il&c{N>ROZHYQFvk{fEacVHP-4<zW6REc
zu3j=O>LzyV7q;tAOXpB2puuQVE1bc===0@SAM(^H&T5P}j+VXVusLgHp)T|qQpy&?
zb$DAP+uDz{GyL7o>WE@!VM?qw{@ZPST7)I!AXW_<<C0p=jE+Xt?H7`pMjxaphBl@V
zjYRFkt(}e`15Pa)cqj-7<!4mq_+l&^IY{&8g>+})jF?^Ci7^$$#w>^lW2(q~$NTxl
zQ=8<4%zs+9T%8h(0$Vx-M4;(b!K%N!<G*jp$^!Ad;pqJLcnegMr&C)y$3=QGa<fkA
zi&2?zTRj9dCi)W{!{1SzRMTq!x({ZsP9noqQPQbH8VLod%&I%2Pk1dH14pz(VG`hq
zAv_pzyTKev<bOR4MA+b>qA4?`w`amN<UsU=66y{JfC5`B1vd1b-?0Q6Ow;-+>Tu-A
zFbNLB<!$|gKE@!Z84E_Vi&-_H$m$5CF{mB~rkQ0wI*CmpERAu%Mora-AIKH-obhC$
z-vT=5Xg`6HyoX)*%=6CzZYW~P{=A7P9psHa^o4zV22M-oY-oTP*_0gIYJ>Ijza&&f
z`4```Z!GI=)q8V21v_Q!{QvlR=l{I6FKV>$6DJKC+jbhOv28WBZML!1u#IioMq{f<
zW7~J<ob$c+e%>G6KVU!4+Iy`z*IHxFF-9hQjGTwaXCg#l!J=ZQZVZ(BMpxaAXI0%S
ze$8{A%@Q9OXh=2lXE1PKcYjCS`tq*KAsq*9C-*cAGsdxEUa#&AyP1NoR`zde>xcm=
zBESo(;4)kN%S{_c2;c`iqN93RW!t*yH&mlOx%-@cx;)KO*O`utJmyA2@m@1AA$Q!^
z_u75HCVY^-T^|CW_%X@<L7pAk(b{%nDmgNx*KN>ID=Ry!X)qslOp|^VGFeMsELqiA
zv2Cvy>>9DsfaKsCN7SyIGkph)-!-KBLIgKB(zQQ&4@h2#k3!)v##UZ{Z_uR?$wzsN
zZG?MZ3Y+}G@pV5Kd>iTHBp6fcC85)%N2QAR=+}i=cq5w8Smo>UbLDRf$*p06OF-?X
zz!iP%DMz2Ez@!xg&?06LVo2Sb(Z7GDU<4`Wao3Zo7LeSGpO;rBAz!M$cQ`aQHaB@;
z=b|+e`eIq?vt!5e_OoD#ULO(Fevi;RL(ouh?h4sbLq~Y2hK2Q}Eu=2hr8a#!C*b8H
z&B8~3IT`A&`7z$+OBvsPahOg#RXmP)ic?4|{_Dn4@*G@POiV5a1gbN7&Zykkt~`xZ
z?N67xikjtbRci!|={M!YOWOS=8OM)RYh?ezSl*PES95A!A=}LL*Ag9Hs5OD+AJb@3
zlaqf|MGiLU6kq75B1&>u*}NyFu;IB<yI)o*CHKbtTF{MNjA|8ZfS%_N_Odq>x@891
zsSzwCz4l*Q%Q7DzlF;8H$$7JE>EB4Ag!Hsqg=QzKZmgwa&|BH8B1+Za+sS~gY~)f;
z&HeG1<@v+GNQrK%FEAO`$++ya<vk=?gr`UT7ZRM1kB}2mMU7IH3QWs8R(DOp4c2qb
z(9QJHHYnyP3t}xUCo1VAhxC<D^b5FkYW5H$fQC3kppEK(guYpX0ACM2%jD`Tr_9!?
z2f%pw^+qQGgAj2xiCgD?S^xnRnLJWYt{%T_IY(RqlTY$?#RAr8%8h_vDKZi){|v<D
zJLuISo>O|Nmzb8O^Gz9_|J<FcU}4>5Lz9-77}C?yTiDiCPT1p+wHg=+Ws91t$ooh{
ze%ZMEZzNa&NH#RyQ@JPhDt`j4VjaZS9<KufDCda^O^U4`KWFDGBx|ccvn;qgge98^
zjfRn!9wcA@pEKUYxJ>uIsaHkW&-(PDP)`GU^dLPsu`)dF$|1~`o1t;R#ifZeq*E<x
z`<3g6zaAI^UxC`!Kktg_e>QFF-{B!36v)1Jsg=_P2OqkAma19|j&Gd%aQJlB&hV2r
zxykVTdva5Qr_q}?8xE{k7W=)wzF%ZH^nb6<EtoSo05Nx%Q-p&NmzhcVW(F=CRb^Eq
z6NNuyEi{VAKCLD%^8SY+Io2-!#ojDVz`*Jj6(DP4yZl6RkT_E@|562h_z4U=O`AiB
zs1_LK*`n3~Vg#Uqju_GSzo1nRf%L39B*DgBVjt6Owf=PcD85voo4t(jwHX%@cs5L4
zLR#7`_AxL?W6BW)@hRi*eFYzqnxAm~gU?|TPgS_|!W1ImI+x(J>{g6}c4yBl<v0Sy
z5dBDiKP`N%HR4+gewSXas(NRr{=@u??!xp~(}n+y$q9Kfz4w8u@r@Z#F=<Y55TpXh
z!g(yYh*_sI7-ZhhP6JLs#zyn32I$iNQ<WkCS<jjNK@$I|K(b_+*dV`vzr!dpQLd;r
znurH-K3Nr#H@00NimJ}pg$`Jvu@HPu!h%)hH2;0rW#Xw;mk{!wuK6n2ts*OWj+5(t
z75qQ*B|IUQ>upF84LwH2C9{;X6MPlQ0;t~w<Efct<|Y1p8VqHBK@g^-e!`N}N@^Z=
zpE_FbwI+$hYQh`VrjEDXT+z>3el(eaEbso$*KuT#oL66ZLyRT${~i{}X#idS?j2i<
zuIsQkoep!n_{2F{H+OZ|N<-V31zgk&s_C>c^ff&r>ew(aM`$}2ZeZZGDy+Vgw+)B^
zp*lbSR=V*X9KdEm4CwsZrs`U$Rb(+PM77Cw=<xFYyw%(u5Y4G7(kK|u08}b5n?l`>
zgo@f;At7mMa&`ys-Ada}0df7Gr_A*q|8YBJf;>bGu(g!+$LX1DYjQ7Ny&2cN963h4
zStO-xsX)=F_biu5XI7uD+}I2!OKsvWhv`_utUHdYx+{2}Pgx7La>joq6gb*?+C)m9
z`~F4zy=ykWPmC<*TTfh4!)^J!D?WeJWMUOcNl)*ul)nQbZRx#EFUg+?Q=w4@5ffNt
z(0dXyuPQ74Z)K1m*7quYo+p&^)qUl-tv(;UZI440)W~49i=)%4u~wG8#)}FCB;aQQ
zlR1U#vVS1GuaC^?x~zYqV2Tc9e?^c7PZZqg_?U5Ly7G+yiu{OVt-%U0vMUr(bfVo{
zwfnGyowJJzryGvmoa=k2kL`O(o>CE;el+}Ng|Q0d{A>`4g5EnON83?ad)Q9_UwNRy
z5kqvz#E%S@mU14XT}%t*3gJ=Opfl(CNaIRBhid-QwB=Etfn`py<`%ej{}cZh;EgeU
zQc@!MH9w7mRv<cc8XFJ}vb`2e%60S~{bu;u^)YD@;%h#h3>G-QvYyd>JKg0^P`|Ri
zwyV^jp*=-o8jt4kpywI+1G5^h(!_Q*pc4Jl4|;@?A3oA%ckeQMFZa^xJgZ<qH?|}(
z_kgDJ$Se!?&fA6#Z4Bo6Y-aUm<144o`|1aJgLMCa(*Mb<?}s1YR2$^qj5`O-_Yc+k
zzc2sqO#a^<frYIXn!Nlkzx|hWu)t+efGPa3nWL5R&&lea1^RCiUw+R}HdD22|Nnmg
zo)bj-0<33>&qed(|BH?P#s-kfQ)3lyuj&3h#r%KIO+mh|Tv((Mm*Iay;QxJ@B1B*X
z%>Ey*B8VpDk5)8#lV_3p_nV;D)ltpO;(wuojk>6yMpYUyFBySpF-_bWWt_M4F)8wo
zv5h4GC^0d4q-GaN3E+ZrXJlrQ<1g1IWn`571@pgA(h5#Qfcba40kdBW0P0wc?~CWb
z@7!;B`S}G7O>`<0b`vj39eVdcWwd&w!ou(eu`5DIXU#D{4Jg)Fjg$@@17jG_0?z}K
ztmBhR5M(U*hvyHCh2BH<GKC?<<%mN9V2%b0?^akmGoYe3tp19qvz=qn{9$!cBo3+M
zBw8#kSY{3k!JHq2{^`>wfWn*feMe<+*#1IIODml=!KhUyal7gye*p03tn2609078v
zWU=L|QrLWAS&GAG!7#lMVT#!RANkCus07i`!r>m1&Y?b!+}#Z$g1E}$XBC?i+s+r)
zCDU}99m2Q!Z!MxKOl)(eHINRZU3NPFxh{qdaJfI_O=e<8H9*pd@6=JlC`-DIbjDab
z`1VxTBP>x4Cu*syk?$mXIF?k*^>O&OC9ta+;gw+kb2-%oYP*@&1`uocCEM1M9e_ld
zC=>+EW$Lvv6lcqv$3+)zUbIlAInz~M+R?&u)~4J6xLCf|4gL|a-Pzq8UTt<#7Mr4`
zXH2r{Y!|nE`Fp0M(4@t};#lI$O-4-3phsFM`=&6$VaZZT2J)r`xQH8=bQ(mjPMZxr
zk!S(r(#b>9`O=)yKJTXE@5FR~O)Rx*XH@&e^`|6}f|}at-&mxRtBm>_r$VIo^n{Z#
zuZ|$Ag>rRx5ld<^@0AliX2`KER^~RkdSF9h1B0B3Bmf}f3QC|J+TeV=sAxKtO8jmr
z-gZ4LEoWf9(a|oG$|UqINzDc5<L&~Avr;sGqH5>!bp2bRvRs_)T8kt=U7O=G$ehUo
zKsd@5DnrrKPV4aNlZM>ui6d%!ZI{>ws4J7<5zXaqvIs-lfn3We`+C*p{6Ha6cVXk5
zAs&i|8(hsQg$I^JN~a9}i9zLOc_3dGi_;MCw;{lcBOkEe2((g{lT!^~xcEUQ&O#cS
zD=ZY(f>*9rD_2v?lAqLqiTX`JM`xQ31j+o3E|x3pTh+v}sjzUS&=M2WOHsjn+7aeR
z4PvRW<`xaLy8gS*;}WWQQXDP5)3WqlhHH&gVZdXvO)kd+Pz`ar$#xjPjs~-wDTu-Y
zh|7-47KL1IuMbyy(kl%%w7TkQWg5iq-XQ!wj}oS)rnwDk>T<ZmSla+FB+2b^xD$*u
zx%%Kd{HO;|{yoZDrBc)~cOufVCDcr#uTK~CL!Ghd1n4u#M(6323XIDW(JP&jgRe2;
zpt3nPyp;>&vl&3r5YPy*tE;QVCyq*uqpcQOHKFW3trjYhm>CROUv5`g&BlAM5<Ic2
z);^;n^l_AHcM6TdNyfDpf7KltnMg!vPstaCa56A0sE(<J=7EB%Qj`l2`qAOp(Z+UV
zl-A_l_G&*)$p#fV_5%wd;M?_ja%MvkU+Wo7=c$`}Wa5pR)$T}Ag*~`0K<6fUm&X3#
zUYA#@R_+){amt(R`&cZ^X*o;vZi%3)^}GR4?zP?UZC(NL1*yREd62dS3^D=TI|<7e
z;Ln${O~zoeirKbZh&A6XTkQh4U&4FqE%Wu=s#DDKO{HHF!1B_41&FOGEXL^1zr_h*
z?--jl!MT8LS=!O<b%1J1^sbqiuRYX@EI{{e?J%`Y-?*w2&Gi@S?w39wA71K&>;V+c
zboq7)+d-0b&0cq>6=&+YE=kta{g_EN2J8<2hjZ`H1E|^U0$f^qoE?BiHvCSM)$z1#
zIiLV4LGpQk;PvH^{+rgjVc*E}tTKN6$XJrKkE}kwuGDNvm(}*8ok={m(A{BMdj|hK
z`a$ZNT2|(>=h&}((hJaa`XT}yO|E!M^4nAATLNG&(|iU<Cd97V_ae%jN2p-E-d1+L
zwoKVo_5XlmeRtkyb$}=&l1=B31q2anA6}o%)^>GWmqfelfm&9D%wJ|>JUf84{s8DQ
zWD1VBZ(+Isg0rB>x#fB*U2E8;$Gt<KcIT#{L@|jwiHP$Wgv_t;3TRHKSDSaQvUWsL
zW3oV!S9S4G@y2E{`&{&Y$_zmu-UMFLb@J#Np)b`NK#|d1d9c~+fQ~lHbV;slav+p3
zXO$t_ZM2*l;gc5niM_y^_2zTCB>wJ3x&Tx4mF^xN5b`BNX^M(R4sdii(XIp+X5ufc
zD^rnLNNKq+-m3=I;o;tnRcn3Gf9JIF+K=j(<;mg&Z49}l-gwnxm?fIh(h+645vkn_
zIN(vif|{BRNk2Z(rPEaIAN|qH9+89y9j?lHQft=V&_{tIO4YABZ&L0$?!xj=PXc@&
z^5bFq=D<VIpu>*(Sa^X_pCbV2Hai-e8r&KLX93M^ZpdO0*GzIu_I=!&NRW!qLOD8n
zP_@vxN7xKl3c+A(O?g3EcL@5vz8&5G)vN;V71f>(u&*hU6_1EMLcH_Q%?rRj+34(i
ze!c5V0W|iLq~%x<5}p7Hn5J7_pqFE!J{(`h_wj(z<Gqs6)(co3W$?^w_ebDE1NB_D
zm*C$}FpxNUOc)^ej?9X9thbt(+B~;beipa6W13=viM^dSI{Cv=<<|9SIeLYIk<>3F
z(>+9gXrO0CPsISu`*^<F#a{A3?!Hs`e_-@gxzAg{Qk*{wJg)BmCy=;zs~SKJ;Q&kH
zaXb~RwWp~kPh^uFXbine>$;#;E@*n1<bNsyYM-1Fs)z`K(fCJ9Ukon5Izg%${zEW8
zTV7)}nvTI#H6-!=1p?D;VuYah)6?zA(P6&rhzUY51lSg!6q5kdiC1p$5$?w#nD?-E
zv4ahk0Sa_y>*TvE^e)q1-p0OKl(w@l4%TR3ldHvZ=tDjA5L6)kyQ|*$m^7BgItYsc
zLKVk%Nwk9ZK=t>qbx&U6xN1G?hzQ3xAw?RZYDSS*a2V85KrPdm8;{w*<-xD+jq|IB
zFTI)n(ei#*Xc4Q69_cOL&lcdNvzliprgMmhUKwIobB)g`ZyA~>2|LNFzy;cZXowVw
zgnN`+;?g@jvrg4m^?XxbLbR>nX7hK#&IseeXiZ1zd0q>;TLQ(N{UMO-BT=<{gQ3=c
z$3FI-{rpDok!`B;VcZVy>bogswlzo$9SrOz_z6vgB$KLn#PD6Az{@ampt*~t265^w
znmq6C6X7rZRBG^LkW7V4PUbCPQC)Ko6sU$!Vki`-#NB8(UXZwTt02Z%c^(P)UKqts
z2v|7U6~w%}K&Fds4i0GD4yVnaNUbw(GyK5VasPDz6HDyO6Fk=&5X1Ql1|jwcw7j6L
zdYL2lD7d;d<d_UBp9AUv+R?n15`)Ln$%o(3`LtY;b}jMA+XsjDtbX#8>>S{!>nef-
zu92a!@vx-7cbrxABFRy84Q5?{RT3q4Fm%$QgM+_>P!EoL@$23MvW(k2!L-p@z|=V@
znIePBuYB>VNi+Kk6B43#$4PJ_xevU7;Naj!?@mhGr5}u>2sZe$KW+FO{(T@pp2wvl
z;5CaB2<QVr0@5Cp@1!odt{i?kl}KcnXW+-r=nz04US-}rUNevjsEGI5Hv-Qhv`BNb
z3OA5yT_NDDg#^<n&;L1|pgX`FKNuI1S!WeXz%J_C=SIU&5={B~?Tl?OCY?cpNqG>t
z#p?xc<ERy>)q;(^5&nAB>Z&zR>>GJ8?LAw%EBFRn;3Pj6oJ|sv$9TWR8vQaSsK7R}
z_9{D`ars7OUCus?QbZ7dH-~d)z1H83wBJ!19GXEJBkLi;|GaBsQV_nQU7xI50Di&e
z_165@b4Uag_`&v5(?i!{PNpbngy|<!oJRaqvIzozjnA-$!NfPVm^zfmk)Lp^^_Xm#
z?3renW|%6n9{QI*IL!f0XpKPQExphT2-bYtu*krGFCoHT7^ea%&30M%$@>^=7gYz5
z^x!D4`Abv4)*w{)*(Y6RR9V+RI4y_eUY~Ub58={CQE=I6$}(a|h#Ah+F9y7Z=u0P|
zZS7unuBY@MSf-ga^O(l(_l_j(Glj`G-=>Hg+v!vc4SGZ2J~3r;NFysBElUNNFVJ15
zaCj9Wq-q5-+i$g2;?i^cnbBSuQfp*-0dyFyQ`Rc%Lkx9LrwmLD3l6TVqolj@VWgp&
zVv{er$6<-VPCjeTT&~4HMRG1WM45fNhusIZYF85$ex@y^jjfYhxdE7G>X=K9iYhXL
zn=VhV?wvmjV(IR-4nbZ(nNPR1826+@9;Q}~d%4FcNO0KcZ~ddekdIE8M3BMpQ{Z`z
z9gNPBZL-+{JO?iiX!1It#vZU9J1iu4Yg1$p2A{;O@ZpV^wMgUz)mbLduC=(Tz-t6p
zPnjj8LJqX6+D5~rp5&=YM?6Vp{Vv%^25wrN_Y^vhleA_dxh<Sc0&P2+Hs6N(SF+>n
zxstn4BupU<V0{HPKXbW{uI)4c6%La_M;7V3-cG}@AjczyPxD?ghEfKnQ)4xwsJ8eZ
z#3snCM~E2$XTCtN5ZG^;L0h4k!3JaYXv9Tmf3a{%DyV>Y2x@}%DCd>==+cA!>u0{q
z$RC_B3>hOk6LWp{Ehu%QF$`hoG56T8?b=42*w|a5u`asaSz#ti&_||fZ_|KphCxL^
zw>@+u=2I{oe;R^av4n<LL#`1+)1H_j%qLJk8sr#`Q3v@79@7L1@I>aB&8|gJy)jjd
z(WO!<C(9+0KH?K+PY@6v!P7Rdj4Of}zUe&;?}yx18-;~Ud<lekx%zU><p9^D)16~<
z(JfEF>#gmxeC~E`nz8>0;mPnD{@-9oopJ+nLvLc8itof9ybkZ&yO~JI6jCS${8<d!
zx<i-;d%9Tz({X!<!SR}4o`igvLc9NBZ*ld=0?WgV@ceIc0AMx(M!D*_iaJlEvu2)P
z$2kd}O!k!$%j+@1KQI=hFVta=>jD#8eR04qFIPaL7Nx?$uJ>c00=VH^jt(TyS&jn_
z?+#u4wN$H&K!d&K>4asuQ%uC)q31_zDY%)ULzves)?LOi9lp3uT;vaU;Cf8<Tf&Zt
zHr_ANL&7f>W999nX^tNj76d>eWcu|E54dWeQ+!<&kp<NkkJkO+d|+c>A*pgnKfN59
zO5dpgK$NRwxnx+U&N!GksNEg_NA+?t^f#C{v@{q9rEx-L9bSR}!DnQm$k8RX8DJ6f
z_wEoWoMT@u8>DitO7lVt_rPBT|D9!Z+6m}x0Sc`dE1;7uPP_rUP`vU|#z~Bh3qli*
zZPwlU#}^kYkH@{MUOHhAeqC>4fLoQn0q<{fuWfDwQgHCyCtju`4Yet)r`I@`C?BTH
z-zjJZJ|2nz&t-vI%k|Q`5zYD|YftM$@bipGAJ=ZopKjZC#exR|o0nhJdR7;;+=jxx
z83Y^ZI8++qIaCTAMTNkC(PRE59QfvDpP!B5GiWoK-Th%EdNVk7a~NJ`^Hbl4W0bx&
zK_TGC{{ozC9c@jT3NLyB`_)>7ei1+lr<yN!IYB2SCN?=-o@_N6X7JXIyU{^{up+|*
zp8Olmb(o^BV|+dT@-3Ra$y^ZN>2GiV8GZo-3xesJnDPa|V|Bs;p-ZUououh5yRj)G
zK$0YnlRF^l?}UPa5{~+j4Jkka>P^Sk*!trLZex;~mk%ibsrZm)2tEL*So!>SDAK-P
zq7FyVVIXQTRdAr&o)75;BLOXBkl3$z%NoMf6Lmlnjku<<nNBotsDhmY9NPlhF;@$j
z;c45h#cP4(LBXvSz}7n$ZVNinp_D_Q3I4*H;v=#J<iC1D>UTCm-=8sYp;SX<ZC&2A
z@xaPx$HIOMAh(H4+rzOH0>K{V1si-oWfV%MB28UiK%L}4=rF@E>@#P55%$70UNk2S
z=tM!#B{OAi<6l;paeb&s+>JHlp4_#P0apuAH}#=-2%){UgnuwxGG3}qXnCnSAun~o
zoyD*;G+*Q9Cl27xbr6@h2ijlPY(+uv_XlViN7#NF(VIM<p`LcxFW0|q#2}wB4KOWj
zL56m-+>dE6&7+)%pvAamVqrnz&}sICZ4RKV^JmAjX`m1Z$VF1mle+Tt2yYHPy)E$|
zcYO{ojS;x9T#7dtU326-g|g4Z5=Q<>MivWsPjvtb8=Ndil&2spEDIqF3Wfv^nDSy>
zG{AM!v9GML9&Y$CcXa3LMBvfh!IPPn;2eLqRM@_fnVII(%5ydH*8US=p{G{X2a{v4
zy%CPsgvjjWS*+MR+&wz1GN4c}TA>RxAPy@+EEra`j?PPY9`2L@d6qF!f0cUI{=(J8
zGsLQ_Q=@;2fiB&QgK&RBQG=ULgGXI!*0L_ZPJriLv9{Z`p&JB|6^}xE=a2#?en!iW
zJ2D$dAWIEmKeV}DX-V11bg5+XBmlgU^2{s=4=UD!f*0MX%c$aCR6~?<;L+oV!zRCX
zDDSX@ctlDQ;t;E!y7d0>$EsXtq|hNjrfZt4`cv(~IKo)yT{x~Gq9v6>rlErxe@}$6
zTn|&V9%$GoOeB;mM{)!oY=9iSMuQ64>O(iE?*BFZU`srviM$a8CRhc2D$+&u<wc_o
zwV_AoI-m*S5k@EEhS+*7rc@P|mg_|@#~D6u%X~xk8>*Pf{hRyV*QH{VQO_HK*@aS7
z%|-<kwqz&t60C!{@=ni>B~e||=!{$`2)v+v$jPtWdjvUT9}>1-!RsC1W-^SzREj`#
zR8XrkozJ_Be^C^O(P?u;P?r1@HYHq+$a2&`f91%>JTlHSZanz+3A0Foy_g=|q4r4w
zJITR)VNXT7phrxMP(FWq>T?nfrD@aeYt)d3*vox(5U5^UafkRy3)*k^!a>pYwQoew
zok0Vw&vV3@>khLYMsyhZO&@46E&ws}oo947_@K0Ha~}j<y>{!n@2V7y!LDb!?g`r1
zWa3as-z?EF9g#pDeNfnc$v63%P*x3FHnb|66e@V8Qq|IC6gLSbFZn&z>$-=drv1L|
z7M>l67|W;P{$zuiuDGg?vI&pp8vP;gZ$|dISJ}o)gO%Vu-VIw@^9ABkl~wU-S$3sd
zlLo_S4&IF~&%p`rHkW-3G+So|yP^@A+R$`4q2$(L1eq?zMgvg}!OpL2KCg!EZ@DPE
z08~&wO+*M%1EmOB5>>5%tU>k$tALu8cdkKfL3~~I<|gWhH9knT%d_fU>!`ALXp%5E
zZ<L{ROd~nqNdtjf;io;^u|~-O;gXAgILdMO%>rCj9!!0*Ss~M2;VAvOY9aW5+D-DS
z#oSDd=7JCG7eq_Q>Z;Mqr}<&Tvo*5b>!thkCU_GKEkD^X9IR9SSmYg;NtC+}RLH>w
zC&7LXR=DKgLMdh+@C2%<LCb>Xn}|IOcbF!DX8vX%lh~`FWf7(e0NpsQJRI;b4)}#I
z)o^@ljixi^v!|2DUTW_Mo|fD7svQGyfkf3n1;d*Q$x(X>KJT3+P$PZ8OPuPYe_?qs
z7B3n`<l~!*H{<qIML}?=FQh8wtB4LH9E{JOA7&F77QE0=nQq<zjP7XugG*)^W+R?T
z3bP7g7wc1u>$sGa3R>ygNx58SR=5ryCJooKe3TS4eobb{OCr}7UtYouFEwj>nqJ4Q
zx3eISQO`QHJ7)eDzZSz4CIysbf}TI1ouGd;y7|%|)B8#ggUt=lo}Vy(8lrL;hRGGu
z+RMkm8rnk*VQX*r+j+S$`I>*f;oa=eliw|O6YYU#u8!nwF(i4JLjyYnQ?lsmac+IZ
zx9W$2Cn?P(9;E#ssG*2TS6yU;NREO%O6TrcR+$GIf)a^x>K_*C`jI${1x2bK=rgn`
zo7;Rvy`~={iS0)Zz%zuqhYeQvE*M@lr(?LLb#*+i;=YMIZyzjN5i-W&O+?p5MSYFm
zJ#ipwzhG3^-!7oC2!iW5iFl=ZVw=xtu9M7;*f>medHKDOaB<KJo)zscuTZ^93H6Rp
zkzRl8MjlWG&#WFp10kTA2_9;na^3|UL4OT0=->XDCsF?eyKJw^?;Ko{p!tj^oLcjj
zM7?ELnMNLM$lU~Ry$QUbbVB)}M$QWjTA_lila%@3@QDiDn=bAL`Wa-vFt0onBIpp3
zXT{$PodkU60`I)+*R?mGg~~xd*(T>M{dq$VUVB9VU8A*$-~xs&CsbItlPK{RAhs=0
z@wr#7wvo8w=8xm2+C81wtbBvONK!2!LMQx{!BJ?k9^wNfI?YF~8lQkPu+LVARB65A
z*roiDWOMzPYMgUPcYh?fBLU(TgdC`fB8TNrpu_W;4mJwSeV+My4@Y`cKZTbr+uB>3
zAndF{({ut+Wp(AuAxD?XKE7w`6%VaYSR;yVvA%6Q5wfhfOHLv|QU|ODI}jZlzlSdz
zyc6~N5Py)Mh!`#;4$}&V)1W-yuxb+%Qwvgoh&Mw7-yO0DqB~(O3Uv(J1Dfs{#?*;G
z*i;#u{soL7aMKow6FswByYa^u3n1p9%(&ZC5#$H63|SU@nJALeD;#wmMkHYkJ+IGn
zEJvI&n_dT)lB`bF|8#Cz$<w8zpOq1#S46SK@3ZzvZDlKp_g9D!#gpb*=;1ROt>1#(
z^dbV6`w|JtB6aU^0*m=hTm?jFcwNz|`lgUzq(VHEG&bEMN+3_Ndn1ZW8zeZ7@MjY)
zIc;8fUe1YE2jr@Wt4O72v%oz}!;B}`&f(ZTkG>>(`6XnXXnJt|EZD;@sD^w<ZTgA1
z2sc0Rh!(HPHu-YwsK~Ej`#)A9EJO%{A=4{n?V%&XwS_XynmoyG>{E{5WYHR^=`gkh
z$|rtvBXFUEDJi{Jo@*>Yi-ZwO<~{VpJ5LM-H5KxVT;g~+ILi}8*E$RaF!m9BVD1UB
z_lC!S)ZY8_h)+-bgL-Ge!wsjh=L4cOztyIU*W!vN-aZr37b^vwjmWL|+_dORX&<3y
zo8Q=wRFFhfd!U{!J`_*XzN##(kbnl%ysAwSBE@JUcWjhE(8QNbp)P2euSbIJHB!(|
zD986UKe0}&YD$R`EW-GSt)Y7BNppzX0>cV@A-TW5fFcFeQJ#-|R|vpyq(CoX3SZ1~
z@fKu!&O^A((?7#jguPZF*X)s^6QPUX{qYQlvF~V=Ro?H)j>p;fYB%9C{-Re_jnF-`
z&gd=|O8EFq5JA3Kz5BZr_(+h_cbANfjJ-Cyb7*7;Z{mLXz|S@coV(VV%U_5$U#BRg
z&(pp9d%*pE3ciRqY&Me;fMs2+{wCDUZs%wNOY9}>;@jm{pu9`0Y49s!n`ke@J69XA
zKWnPIPIzpN0o~2^vO7jTHepu8?X7=o^H-mh<k4wO*d&}pg#$aohT}L<pmb@)pxSF*
zzq{S?ulR{q5Za?EpW5<A6%7*!&wHg;DN|ii`g2v`NQzvXlb6M^lHZqD7G@A6f&KzG
za$u|6_9uok$9&%A>$A&9=rB*Iz;qGuiL;NgnYe}6q48glCTFKnbtD$th{SOjO0V;e
ztxNfdHi_(siB{Z+AhG!6wmEwc0ugubqo$w?*+o%An&4EvVeFI%8c(`37&=^mUO@67
z?bi{mNuduS--9!J#3Uxwu)@6rI*Svo{4t30rp<1-Du=UR>`;ifY@Y(FhCCV~AAb3v
zc-KPIS&Y^#BwKCy5`EByFK*}e!Vh@taVQ;w+D6>#fyu%>Rn6gOLY}`oZ4(H1G#g#q
z`|gS|r1*H;{wtDx(BV73lRPZh-;QP9+6~R*4ba2&u&IUP!r;{y-(E}6AVH%Z*x{y8
z*q_8*eE*D-C6;rrdx8bS%JF%tQFmacBY+`Ect0M3-I?g`3X%V8tl&MQn9hjr2a!M%
z>o^PDN?saHfG=frYQ5H9jOPT5LD5vraon$lF<YLT#Kq#UxijhbL|%Jtj;=mquuvNX
zBg~H}oe1sLCn;$uqt;PRPn$e?B(G9psGg;~dEGo&wi6QjWw<1uN@WQ9(+ddrCFP7|
z@foP47q*Rlo;tHU!6jfrPi-KnnCl@8tOt<GDx(Iuk#pEDT}s~(Kg<Wt&1v~TBsD(0
zz?#U`_cxax)OK<dvrhR&*K=wT1TRzOuywhh7;~juEkfriEQ4l!mG?g0^?bDg?IqeA
zV%Oh59aZv)WrN>Du)HqAtN8)HF6-yAY_vZA$A2(jXjgD$yd%zQ9Cd-UFkwP}9%){P
zXa+_z@&}xf?P<R*?<HQdNK)_}TdPuzwfNbw4w;|UeV3ss@<RQmifKB@r4uqd!ubH#
z)oOJNu1qa*(u*l;p0^baaXrnKvdB}X$W@L9o~Xt3hU4W*wk+fF^@M_b#==!nA7w8n
z4A9|`kG#I-l5fHRG$<OEL8)+LOZV!OWjsn^pE8%FHfVlHrSk_jL8;V!+!qb?FfaoX
ztxZaepv>6liaz-DpZi2<VZzPx4MU@H-e4oxJ#>2D4fv8}T{6cX)?*F8eZ+5nTeVu*
z6Ukxxo=pt0Z%WovRSvh7KDwDW(m-;T5+1x1$eY)sLL?ctp!y=`HjU4q*Vg0^mnXO<
ztGtMcodg%|HiU@?hc{h*W+)m_a+~f$UEOXXD5GCI@5f{1IwU4*ct7*Ze%W=*V=U;F
z4Ru<T&^i=NtN#|aUB@mmRD6fq)__4magD_QuT)F!;+BY71MB@|$XuFRF*_(Rl#E(U
z$Dw^Yu~F47z9m#@I2@OR`nPLUOgv)gyyU7qq2B84!Wv_QcsWfadKT^`tt~lNzl-pj
zW_)_1h$H_7%?qU&&V5uX!<pOY=dN}zXmB}v@D{Xv-}<tUA+95y7BTp23DQmO&Bv|Q
zrw6;Uj%gTlOd_*L@~rc@b+^qS%b1Yg@gI@1@6t!1&bVwC2D!yOfFt(#lS-MpgPM}*
z&pN>)qR)(5?^Lie-@8_dr*J2~sxvfa^1mK%oR8*rES*x|U``5$uViV-pJzeL7>eS<
zwSY8F68doNXjV$LZ09fLSY{3F%m*xRE=_@|hLx^vYCiE1mmmgP8O23{1-HNc*et}B
zv1RnSv`KJ$p4^r(i6!w&p9Mrh^RL;#DFelPOv+(h5$feJ8ILQU>vKC8BBU@Sg;lIc
zD;D*+?DR)J``0h>E7O;D^OUmVZTf_fz5bCrELqQ+U!^=-J$K5XzSq3hzP3D;`DZKZ
z)b?T{0Q*Y5t6={6{BQ|y+EP`;hr{sU5Ks68|5|jnCv`M{XZ9Kh6WEvO7<Vi-RD9uI
zeyXruZ|u)&B*m&Is}Ls8fBR83f6SJzO3qRw-jGq=d}?MKWmV>CvPI$HeB%T@Yw>%X
z{c<=uhMbFE^)$^Tw{M_tM1Zr6Sb9#nTA!G|UFm5As#h-M&yyU>o<QY!mIqhF&U|hs
zEu;05#9wN3_^vTf|ExaOq0X?>E!5P~z~56nV!OX%3HTw}WU4dm(ZhK3DKWg-O55IP
zn%2$iqS`was)?@VKQxmw30JH#s2hJ7<hC{YQ~;7->I`?bzEaZ>EIVaCYj+c02Z#2^
zS9Z4A-i94|3x+m@mK=M3!H^MC==nNb-;v8HduV+XJ=vi5k{P(gBUg`pjt@Ywre7<s
zCaMI=<l_Yzt}(=yexwu0LTfe07QT%jHprxp64FyUrK2el7K$Ag_6Wbr$c**2^iJmI
z8nTzKI(jhYCDYa;R8-kwPFY162pmGm2B9a+5bF;0bj5Rp{}A{Va~-rEwR@N!N!PIQ
ztJ`8<(dP!Ed~%rSfiV&5q-J6E3Fd7-kX_0fv9@d1-$UlBzwljx*HCYheQkbKT`GRK
z$em@4i<?z8q~zE5jgdQ<k~>{%qiA>J&;?{Su=ixoi}d9K3@<q`xUFy%!;UI?YVR7_
ziy5C|_NZmn6a+FJfygF<mEWDzzF`OEXYisE<~__U^zPY$<|F02@KHTP=hA2v>YHtr
z_}yiX&LkF$?=$g30lM_1Y)n5U$^S5>Fe38|gDP7`?uO@|lN<}-l3Z@)3c_AY9%?Mo
z;SQhH6Y85I^1K`c4p5Vbxx7?=F22>t$Y5I%G2p^A4x8$#9Kf8dlkBQKk#YSjTw+k!
za_XNTzbstpRO|04q<R)fO8VqtvVb+dJi}U?UFArdE#Kfn=f`)k5TATJ8uX@j`qjg}
z`=x+z_ySu=S)JS|VngAlTkRN*eWdkLZ9^Plzsp07zOr~l*a>aITxq;^wYuxEHYx5e
zuMuh=XGN0shQN6IOQmNjF00v#sQF>eL>TQ${)Ba%go=q4z6J%~&ds}m_4Tz-2_cUx
zIyEShrQ4X@<lM@Xuop3?ev*W=Z^-CgAJ)-x!vkb{vGCF8Q{K)7|GH~A>PyC}EGgHC
z=+Umy;oe%utrWghU3kBrylR%=nh$24WEq}Nu@my`7xW~L(I`uAe3Ocll(qC|4%WOR
z@(Z?iJsnlMJq`u6M(-EVmsA>rUvus#9~X~IOy6LkI|2gjd%ZZ*F(31gQrdT$;p|py
zeJT43N9T1{<{+jg?ff!)t~$K%&OdKD<j@18Kq9eH`jzv;Xm6yxFK*T{N@Na0z?I~{
z_Wl6A2P+Zf2o3atZgErbrQTaawO*BYA!KlK+~(}ZCErmCw#I1g4=D_&t)hEsmJXs(
zP;)w&J|;O`e4*NlxZF@GQ9BC@hWMFw6BLx8!Wb1-gY)+frOWMn>hqY4@p(<oT<r$G
z%=xh_z7sG!l}MQYM&T^J&-Vk)K~$i^f=r5b_>cKz1v+{W+C)ikKAzSWZ#$GTjLAMO
zPf-Bmy&yYblS*GxsAl<9Y-EL5n!Wxnfrs&h(`By_L;QZQ#=}ZD<(LZ|q2DiK3>pHR
zkdu1FW+fE`1~jjml#8Ga#zx!6_AhgT!v|h5?16qdqGdY;f2DMPAcxmmnN=K1tP$5f
z;?Y(X#h*V8aIqAxFeMNYmcEL^w4w4j=9-AiL3^}4y3G7-=9`GiYvV3X$u$-l6H_Ia
zA0<rTr*0h7F<~jUx0MF1XYxK<V=Mh&z~BiV`z4yMD6)IgcW3glPF~~s!p&tlyDw5w
z5y#zZ^PnFU$V6olDYHJO?)v2{b}TFH7Z#~9W|L~A^NUZHq*6Ydu$yY?L=lhJp>)Y^
zZz_9yc9rgH`z1M8IMn=;K#Msco2=<j&hzQV{JX7?#b!@|+tHms)c9B(T8tN3Uh;1H
z;io@9><fC?=adxaNnyk{cb2GG9^p76T}3&Z{BBqf+i@bIppc3~{P9uKBzX3Vu02bI
ziIc;QzQs9#r<Ab$tMmQub(ncssxm6e+KywcO`#zPZuF8rw)S>o{10~%Zh4;a?Pcrr
zDkW^9Cc-MnX^Tu6N=nMc7b9G7#9;})q4-SpzyGB8bVnD`Tc**XR&sPVT6mN)TC6N*
zUdB;DAsbCUsg^oHl;Fh6ub7myTT(WAoYK=L#bq@uN%0ymXM+sw^ay=v_#`f6>e7aQ
z$2+xiQ?ZmpR(awfntY~dqDJ{r<QZjQCgIZ1H_J9;t~&+uU2eAG^s>`ta)jmv=@@|f
z!)H_kRVCE5tO*aaGw4J@5t-p3WM@SET+h~R5<fb6k5VW>JYH5pH235R+n%q!9Leyw
zxbQ!}SuD-$B;8C(3prO<k1m<A%pF(Ivsmfm^_fUn8kV8oR3^vw@E&BcseE-}+tjFC
zsyd&!4;^+Kty#&hGHulCPVqjen^9_U++U(dZLu{E-C`7q&Gv3k3#V~1c?eBnse1bm
zN5*r`K6I7jl%He$t-ukVR(uaRHY*LA<!j$}h;<agjes;Thng7+8Qu5{4bQ~u(-dNe
zL4?GP{Gi=#8|2aTY2Ml?^nfWY9CT!pWXd0G**k{%jTM4w==8BPQMJ${7?SSi$(JiU
zlm8rU_XIUnGmlsF=`rT`A!1G(>a=1|;rK2;8zhOL57Hz--5y@WqZ5+ywf0L(32$Po
zw36%rJ4Mf#<vdvGZ{PluMvV4s)(w@+CCI#D?g9<W<HxQ<C@mIV-!>Y^G?pBFzbMPw
zwB#dP%bogJs7t*IsW65c*c>X*RI(43X}{K$$eZzOUTGlHV%cJtpv>lR-SbinlUsBf
zDdA8kzSoWw>15(Tug?X7T&bs)N=+ml*L0b41i0b?={yyzDx!&)<RVV(3|no5VY#*J
zGBNCfG?}G*A`Q{5sh_Aw^YL1%x~r)AkB#D_VX-AaFrpJ8pK5q4cDO2TzYlxFqWQYV
zG$q$lk|Bw?K3<Wrx}vzo9j3lG`Yd|u3>xOkSULh-#&X3n7amHtYSt+!%9rGd<h!Ye
zF#7$EMV7-Gt>Wv8{Q7`wVye?g_Cb)@X#Jl~RP=D;(O=FdBSo>3a0}AirzB`QZJnms
zTsm5Mwx^=P@pe$lH0_P5F8v>IGs##ZLM869*2aB=s^OCBNg9atpp9=*BoipGv#u*t
zK+7DcH!D~m9NKDAGUFlvi&|@G+IV<XVhI`9_|^JWSpm=fytMw(=wPDpey@d*hMG0u
z3`HT6>ULS+>LMtJlxKO2ew4|?L72i$Y)+@xCDjHI3UR|-a7I%SKOL@hg?n!;pV_EX
z^>MI<A#{)`1_Fl5|EjH|K`#!bzNWHnfC}))p&?_ZR@=LWWdBZpjXWkayhCKyW{27r
zux_~~FME)WW!qm}>f#z(K#Ry3M+}3L!p~+t$2amcT=$3?k69^fqgsbbPz!B_E#%|Z
zhKz|ZheCP~Pfe{@Z2X1u=0?Fz!b7sySS&x)-z;4eraP9%937JO`>Vk<)JJs%MtJFz
zk$IoeozsU><HIhr-A~Guym@w;7pQ{eAF@+7;rr%HVr42Gru$fw`5P!CS-yHvQwT_c
zeSTCbWl&z5Qdffi8X2oora&QL$t-v-4qG(>+IT@7w86e}pvY2&#;24oiLYw#Ers3B
zYMuFX&s&w(YJ5E$%LmI3ujcI2mN#c2p{4SpZ(o3lR8*atzW8_3cgx`^w!)y+nD~*)
z@PSC-3sD!b=4#rlmX*d%tD~bGvuu8<bhZx}sgZir5TdzQ!A}r#oU@$5jraK4*}t9+
zRhlYeAK<)N9JVo_PmRyuO`boZl~**4z%?eyhTpBhjHs1Dg?jOI3`%IGH)^QOF(jP%
zct>~=<Kzhk?I`0qlP@hI^h)2)V^}0q#J?s&yHw6U`45he#z4^Tc4go65o@CD)(O?_
zlm)PAYmK9N=+(n{HHRcwl9h0-2aego;@tfW6A<6&McEc*B`2rpqhe-QcSi{ywVKP_
zRAy;Bx|(I=Q`}Z<K0e%OC(qnOlHcdi<X6<W!_#Hln%**lKDmg-=gAX<uv!wPRV!m^
zZW!J;i)kp{j>}E^t^r|SI`8f$igQOUHaNT(K#DjPs<UljVIlT;$4km3q*6zuZj1__
zVhu-Wj^z)^^R?tr=F-aML@tf@6Ib-FO*-++6d5k@f(mSg>>Xc<h=k<fcznW<_-hNt
z>#Cp^h|_KKSvT`dR&@1j^WcqjoGF<mhm8@LG&N3{cy!D^dHX7a&P6QD-Ny`3w}#u9
zF5II}9fXHj{Lf8xlNLb*qY2qBBDaHk!@n|1WxMReetDw8<wET%eJz=cPf|&zVU_+S
zb!Wa1D9L^5EPuve#2PB8X2ZGE<Yr_<Gt;LP)lEgqdQMEavFfAQcIT8!lt+1!!l*_-
zI_IGpy1>|f8n;~JNZO`t66Iusvu*6z@19nums752W%9bN>exYJ|5Je|Yw2-ggI!*I
zPcAX?Ql>b(T-D8n6C1nIzOKzSp1r{>_tz&W?aX``QQdsrqKM;*b*;k$W*srh4Cv&o
zws0R7Vxv619UDHR`(456f#+k@UB_<~hP6Uy;MTt<dmlY)MxQQa$N$dHrk*rTlUkca
z!8xzos_nN;<>&=rk*?xXt8J1Vt2G|%{!($34rd-Q)y%Ju&zn?nHI46caS-iR{<wTy
z4JW@lXd=tfGmwtEVXh?EbXZcl12F0>t+B8`#!w9d5HKCQt-JVFHW>vq*5SW$jf~GX
zyG^_4Ayo=vrOutCyydA&*C|7M4tFAOyf$@IBByl9#vp5?KIDmpm}1aWDn|OMB`XSs
zp5ZfS3#`z3l{Ai+S*eetcJy9zd$;T0v}!3gU`jK-e8YY4OZVx~Xc^$jagRJ?*5F^J
zh)>U;6ses|RO}V!puxaY%#Kf_JDBUxh>34ni&qEOyVUChBqj&q9;3TfOZrR;VINh?
znXz_TSiUGHVrG+&))ovsw&h=ryxy^;3X!$8WQQ?+=$PS|yIVeqnM`SlEM{?99ut8y
zaZkcZG1z9Rno7``Ul)LMY@-Z-bXsZ5xa9k5W!+psP|BOm>rumI-pW{(QOK{dO^XM&
z8jv7%gGE-2qs>yd&nG+nn22!3uP}g&kYzI5@X5Nh(zx}}?W>8sGoh*{3t0t)mXc9|
zaaXD@G|T+&2z96<FX`y=nK4;6`qxs;929;(!`YOb(nq`-$<w9y!!{>gBsNEXI*Z8)
zFUi`D@}!M}y`Hd?5zf%edH2;P$Imd15j0VOQ+(udkIr}4AxMlE*j`wBFHz%hzj-P@
zQKdkz#2dRgB%!CwfhOlSWm^3$7ItC^$(#e9xTUL$NFU|m5OL_9hn7>E`=^~CZXbc3
zO9b~ZAA<^EqQ5roGsmV;0Sch4M4>Uh*y^`7Ftg!KP62AdWIv*QZOfe^y<YY!g%G+)
zee$P~%bhSr^H3szo?;uf;({;jPk&lX4mX!r3ig<Wmn!dVR8fAB3vIKP{j8aWiTLGh
z<DB;V^<a&K%kKx=-9wGfq$L^~UqMNwB2&SQfK-;EUeb`QhyC4*<JU{r__$?}2AXFh
z$qC_yzl#w&$90-4RK(Jumh<`JP}##Gt#9<w{3|r|%r`aeqcHJl92#=+0?S=LS(KkF
zS|(jCJC~77)*ivR3O3xijJljxZh9Nc5V+T=VCxn67iYR^*bZ28Vh(VHFREYU<K^9a
zi5N##YIq8@YP=9xj#TKTuPc`hdUyW9rr_Qq!Nn++Az2aMNB>@SGnb!LdthsBJuZ%k
zM^+z?3n!|k7i?m{^J%fxi=$nk<xgo%79c}*ZWXP?uGeWYnNH7OmrmaGp^HGneBzGs
zRVuu|7j56+!AKbjgS$GUxjx8ZzifTDv*ak2!mc3bI-+>ZzLuT0Rv_@e13|z{%rbwQ
z`~&D1guGFEnDz^pG<xRya-SX<7PTL#1~92>>Q)xl7?J1+5m7_xeWiSEI9iLVa1f<R
zcP|=Y2W6SjO(#vx+qWM<lxX#4g1GN$e($BNw*h5AVZFC`rOfTg?E7aA<#9>UPAUi`
zPna^6+pb^g&@+TN=u$lgA_b$zc?sk#3_{6g8I<yjemZ)3FOe0pRYGFqkt`@h%gb6J
zS@#4yhfXwzq%9U=nX^JQUp~orIUa261|e1!yOAhI16b<<n*?7U>+e!$e>SQP=id#;
zk%6&zDa5r>F>&i_o<BZ+3%-YJcIiPT6A22TJ%rmCM`RpP9wkVT=V&J##-YPask|3P
z)Csly;hcXRg-zjtm{eo%<RvB~Avu(IfcVFFra+d$XWxW$f-^R4Bve?l=8ANw&MJb7
zkd|-f?>OuWMJev|r<qu-CF!q6zht{u13xc3|8dYykSu16G}}qYJR4RXeIvH3^ieD+
zp)pZ#kp~z%5x#cIRzfSkTRvl>2aRrjh&qU)r@RyBeb!R{2|rTqC*(<X9Zx*mL1e#T
z*P4Uvvn|4rpDo`m{4tk3I!~@vv)p^`ItXft$SZT=YAgW4uMBtNHh9lk+#{NcJI69r
zCYno)Vn9A=y;>x%>}$+w<GG&8#$R?%sSFpFhO<_esK_Tw0b}%F&0hHwFH@Z(YEb40
zg=_q)9>+G_!|%H6B5`Ip{@ELYtfgNB(lJO|0Nq@}t6KU-E<y(xe1H)>I6NGVnt{O<
z5B}jqamSHJI-5F+K8rc4zCwz*k=kW0;)rgc-g?7RGo^Fc`EC_e|2n-pP^HMlVYo)@
zQ<YXp>a38Z4HCs7OsgwoW5h<ASUASsk2SBhXDRhLL@|;0N2$*3_32W!rLL&&xb|>&
zw>prU*1Fg&8y(IE%B#}Q^GLDGE~Li7bM^<ybK}N{BV%K;EDarG*q;q@mgIdTaSV8<
zU7MUoW#=^l>9F++mm)^kdE~4)Yo}S4LjM-u8JS&RVyp7WOdNAaKP6I=($<<7TL_H*
zni)6V0m}-4>hKDGorjvNxtz{-vtYzl6Zk?g7oEIiqWQ7ZVIK0$k)Aqin1z8@yXKOW
zEcdgn*c?I<GX?~CJ;%E7=U+OVvohF}y=26cdZGK(1R)O4x2!Jb8rYxsq&_-YMLB}2
zz=ec`jLV(51zwLGF{C}klSvzj8OxENp`llHE1+TIX6a6=Ps4i-ZQ_L7?%zZg<B`FV
zBMx{8w#p49@78*Gm1tp#FCZmZ>MjJN!nn&24&<4P=EU>I|Hvjmt2;%-(2Unjp*AuO
zh6b)|UQ4Co@?0Hz2$D-Q@daQz=M1^xEn~;bk}YR4$&qj`<|PY^h<uu!J9*~30qd;v
zGrFICrDO4kd7jF#EkH+xb10Hf^x=^oj1;c7drI5tgtfIFup?>_(y@+&4i}vU8B2H*
zVpG2uySPLTh^t7&lw`q8POH@8+#}G>vNdR5XZm*(j8v=Aj7NiBoqC_WoQd1<(%A0Z
zANqG_NeT+xo#=GNH@t9}@o%`N>5velj~Mx!+jQHmlr@Y0Y%i{(A!$2wXkxfl_jr)2
z3U@gc6qFwqU|VQOiHeGvg2^bE0v2sEeb2;W>0I#l*0kT9r?nex*)`9w1T5g?g(W4!
zj#nBd`<^|{tZA`ULknx7!BzCiVFjEcWv%*I=R?=5UFPWhoV3>J)P<b5>@Ut)jSGdX
zK%s3;>zO#4Uimt)K1#JEX{}x%o+w!fj=ICm>84UGC#h$g9{fFTi>Ts~nw>&o1_*5C
z%tOR^@fcJB`sx((H;5>5{tN!);)`G0zYQP|(&zikI_0D}So_>QCzxdzKugdaWJF9!
zbCCu=u1h2OM<+<4QH(S#R}VY}`iCiu4>okW4#g4|obY+wN;DU|)Isq0Y>8x|1Tv~Q
zYfm&B9bcXQ>?K#_!;wL$6L%$dXU%&v$X#6G{{eR+&#9^uo%UHZ9QxBkBC2=;*`p9g
zzEKRHTIezMR|=Rmx><2JwZj^?kaET;4i0X$P^8%-nTt%bhhbx<>%3x?9C<${#v)uj
zM?Aenwu-bh)}7R1@$}m!+vvN4O!7hbbd%MM^N`Qqb1CJRpKD%lV&~*P77pOPYVkXs
z3feGkrG8d^YdO@JISG7Rn|8Vnb6m~C78BBHaciHYQ=&@7G1!(qL`;l)by4$_{4}~!
zY~L(%v-xa!=qA#E00|b191M|YkkzYz-W!WM73e<3VZBHM80N8^nL16}jEHf-`15u&
zpODr*KKyA(-EC*Bj5tPWR6cWFW-gX<`5b=@anPK;;%sR6)58vH<`^%yKidKBNMKzL
zGsT!5JGwC#>*|5&ltrJp;SvQa())i}0O_vByz}v%?M}<zAb7Q(q)NwY=cj@+I;WOJ
z8A{4Z=AU#*@3_Q(4hvH#g!0f^+Ffeclfk2^I_E#Z(z|i@r{KpRiHUDGL#`J?PR3ei
z)j~%b1^2IJjU6jkLH;mcP{e%1^!pxoUx)^TCxXWPu_Hf6$42;vC;dM(ods7^@7uOv
zX6WwjE+u|Q*C35FN;5P_mkiC2(w&kbf^<lCmw+JM-Ho*1v;Dv8{RFJV%<O&N*L9x9
z0lCDul@yMaAR83`N$li!^6P!J6<(^_TXOd|ycRDZK{H!4P<l!&Ji}f`e;j|-E0~=f
z4su`vjZCt6`F~efZ*YunVW^z*d6zxcb8mz;ci5peZb64h0ut^i*r{qY9(g6lZuYx(
z0e9_y4BBy(RTAC6jkhi`l5%BK8T$x(?Bjbtzo4o01jaXFkIjGm^7K-f?J)eme0kp}
zp7mG#e$aO!;|X?C;5tLZQlX$+NY;IP92%HA2h-aBjBnu!R3~3~1Mc7-%A4nzkLjO2
z*>k7<s~;M<4YRnddP{yZQ(2JlUyiPL|IMh$rQo7glPq)MxZS~SWz{>e7x8GbLcivB
zM43G15I0aUHhWjkl%uCaVU7dALO9^62W5V4D@5WAe`MMPl?5ehCe(g#`2B8mO%fXv
zGtrhjJt*E6QIqXY?hY#2ym^<`SmiI$Ro5>tut+8D{>HmJLtSNbq^P>&wen3mO#;hj
zX8pPUD21!@D1S1lGsm~l)}Kk7rc@9#iA&F^!R2`rmPo#OL<CGt12V}8ONvGRpx?^-
z43<>pELNn-7BrBlYy(ErjYDhH77-K~L-lGNPzee2DHZE2-C#M=sic?NfbH-~Di_6!
zWLseMtL(j3Gjt=1k|<@r*7$sL>m(i}=J{+grRVHN|EtHDjq9{8ZYz@BYq=GhsM&kf
zAOEH4b+g@F<973ovyzEw;zv~cT4^02rNn2Jl`o%Y2bn@XPw5{dx20mPEb)MSX;_MH
zq*(W@+WtO2UjE8^p5~Ukjga~yqb7dQCp*IA(Drax=GKcNGz64-m;gs+UX2m07JDci
zL|W@FAY=Btdf~X%u>$9k1Vk8_Qh^=+3}0iUwouzyt|=L6@>~&vwDgn2TX(7J6i7ST
z2f>}`(A{WZ0GKDRHTm7{SJv-CeK2uuOMf)kDBmC?@b5}IngwSrMhbEY7+d)V%CSQJ
z_Lioldyow_3&Q3_Mp9S%kKwUob7O%h-yhEQP24l}2_LqPr~0iP4@7Atw=BLGd0wsM
zD<0evtm2XW+PJEUbsq6ulKk(0bnoES8~F~3ss1l7P*PNbja=$9x59Rv$*7MKsg<HL
z&A9Q+SnNp98nu@pq;BW)ejRpY<M;16hg=>qOgZ?i*QeG~2j)t9UN%L|-lM8>@tQ0W
z<WE4oF8zNlP@wuXBGhq#(<yfWH}*Lw`X>h+`NyvWDq!i4XPu@%g;2F~2Ydo~g=S=a
zcfKLNj$|-!w$kcFRTMM)tv<>wzoGr<WCD>_GP!r@=)QAh%w;eQS9+6K&auoXm$fS*
z@=7M$o-i`t{l~*{or-7+;68vB%TQo6+Tgep<I4531^Rs0rNIxBF!kGg>MyWyX+)<@
zYl0rnZ#I-(^SQ)@#6;(k0TB_#-5~#gV*#=fiy2Iw(#p!zQ#<&AV7F<GwL)phv_EjN
z3t>fPXw?m)Rmahhi#R@O6L62H=DMV0QA13mJvCn@qE?;gp2;98tqpYzSq@qZ++P|u
z*=3`Oc%{mDRSU<{h|O}pm`$3_PxL}t>QpG_()?zxr87529<TiwHKvgTQ75POPe9(i
z#F)=tV=iN%^?R?tcJw*4^>oI&Of|I|MjHqO+O~o0V@fL5JIV}WZU1$VzoVwlj6y3O
z$$$10q>=z#xF8qFDInzjp}fydo{je%4}Cs{LSP$!fZ~T1>|zSMxn2-kb)_ECnBPab
zw%p(|(XUn-ID*C$;9sE}jfppJkh~$SmMJZ*abs1KimhmVjianBpb)0@rZ4rI7?Njf
zEUvK<eR!NOc(cy{+7%mgU662vA@Aww8EXQf3VF>PoSd=7ZZrH%dfwl$ndJotkNIa`
zT3>Yi&7WP9PdV>|U%q@fk&UZUUh!y_$M^UH<xuG9C^wDK7c1&oqAU%PZ_S}ZSY%H$
zIjDF|?X3*W7h4>@!8@2$zw_{f<{$s~nG)9_b>HH1WGF^9vYSMsW&DmusikBgBc{L&
zdZv_=`6cn4<LjFa&wk%`l~7U+b2R&}6oH}AxPALTbrwfS#%0zOXoqIxyN~|`QvQeH
z#J|NrC_lq?M$@VMu7FOSZe71`&^j{paA9RX!sPwEg3^#S(`0x30oc?cm$j-%VNRcZ
z$++MJv;J2OzXAh6UdJ(0j+bm-zt!^TYzmIOw_!3iTF~$iNC}Ol5~8#~<um7DfY=eE
z9NJ-iu$1sGXb53M6ckBB_A$g5$1w3|5*qjhxL2n3;Q@`WRG^jkA1kNKzPys{QNX;-
z4@q~#Lrw<%%hs!#S6^Q#XqIU0bQKT0e{j_$Ph#W^TF@p0t!O%T4kBLkARfYfILfE2
zb~v%N5j*NblExb2sJQNtT#5zr&h17;SU=1Xf3thwRx86}p@K<*a<25yF17)r7<UzP
z_*9eY)^-?~Ny62Hl+-q72CurrwT?iySX~v86{Uoa$Ivw#MGOtir*w{du}~Sx-iLYh
z=wWrQq`wt5UqJhSeWvm9wWDL0f#6Y*bUc~vN&4zx;^|>1OM1B{P}F=sQ*;Bko!K$U
zUGabXdgD9ZFZ1dBcLNmV;4cW)`qwb$(&zC2>i)QGDjwZ)(c=f*#EXNvhj)Jm?Y<T&
zG64aO?_!(RgIm%WBD^+0jbSuN;Xik1FzOJ=p`J(Aj}Q4d@59KsmtLu-S-(ls&O&d)
zai0_Q_e_278MY}%zC?#i#E5@+zvqud_Q3UPTC)7(Kk4-j(4BVFQj!)1IGh>(F}PKV
zVeBB>{DS-wjr$A7ClD#%XrZ^$hm5=o8Yu<~#lJJzqdi>58I1+ljOUEWr`?-UO3l06
z6BDxw68b`1((y}=zsps>PUd(?{+mFUkXE?OwGUq}BjNG$w}^*e4vbUxtOp#OFhR<5
z>4be`s{)qSok4+%|IM;<`Zdw2^u1^vgbd)1Ojb(FrLpa2=?UERTuA#Bh$nC!nq&Rp
z!;i0izG7evk#v9Rv!6sAq4sA0r<`Ct-j6kKnEP#;gd22xdw@!e*QG9KlL5YJh|u?#
zfZHF(W~;3lNA^{2xk*1wvTqhIgn!CV-<=K25u0nVn9>F&9&2M?Mi}mu1eJiRda$|8
zM6w%wzZEm<g@@XyWc_*>a1o7c-^&vB2_Rid3B9desI&Pr{L>%gxsD9u|De7;UJEoc
zKX=QC-J(3kb-4okSv0Px-9gEx-`d{)OwwtFPIj7x6%w7zv5ds}(K~=y`bBZJThs+%
zy_ZzhlC#(S@4Nq!L9IH2pe+&hRfe#K$T>NVU)nxg*sh^{^7pfeu6au{0Za*~bw7zx
zu*Yc_=AqZ;*ai$U!+7+m(`0H;yV3FSS5cr5&_rIWrnIfKkqZ$>IJiD}w#w9byk7vr
zakXuMSwQ&tZY$PBU4mSYmwEmH0FBy*zK7`NM=8+xdlV)?OZDiLzK1n}85e?Ejc<B|
z4(0W%mkY(QVhHn3Sld{;8cN74P)~Y%2PCz`mOAjIdL&1XDj>UImjoa2$0?jOMYDoK
zQA!I1A6tu9Rc`X1OOH<kMnJnQU-}@!hcnt|DwdV0>l_#r&0DL@v}mHpXqYNjR{38Q
zjmp1?UX6Yj-3}O3G||u9wyXBd+)O$;^h?{j%(_%_5zHxz=5Fu+v=b=D{{i6RtGvG%
z7}3MzP<q!Yiy)1z?a?nkKHDNbE*}D;eRPkz>L+w+h>m&&ZvkKLh*{Vod@zYcMOAq=
zjS~a-3`EjzKdt>4_)UU{@V#tZdm7hcCLmv`qZ>PL&#!G4ME&M#RQlB@M*A$eiWsVr
z>O-h3*h_F1(KkQd+GUo1bkj(;!IyzTO-?`&gP&dVdLl<`t}9Rxf1Pzi53%51F4$S)
z$7FOBS4h5I6I30wW#&S(iriSO1_B4y5hM}Jo_K#TLG7F%Mb2(0=kGfsQWNgh>4bHI
zn7%MoTWl!ER0m52iwBFM#%BuHw6rp@v%^uzJ8eJJB>I1kCfsH{+@e&f4hn9t9-}Pe
z=FQc52DI?6Fx5(GD2)Uxm<)>vx1Aa*0*g!d6Zo3~4TXpr9wb;89Lj%lUJB!9QV%Kr
z@(8zkvb?yt7xg|#S0}MwE`EejB=H$t3`i8|@C_|RH5`7Hebf4$=?asn@tt1#AN>)B
z_<z0Q<o=Hi?5gt+jE+B-qnEvO&p%1+mbLJ%QM3suu_ErmNb&;zKCV#}ZKFxuAtL=@
zpxYrSw2v3kwwRFQ^T4;3r?E4)ProF}w&!I8tTt}cJsi3yX;>mLzVTahsp_Y7%6I(7
zTOCZ5`EI7x#Oz67ALliC89Sl0l?LItsGeTlBXt}4vpDvt8_)e4DNXC3B|WI-MREE3
z);Ld%5<7+w+6Z>AaGF;cGpySN+bs?mVgaBAwo)9IIEmRc@Xt`5x`sAxzR+iYEzGg&
zqA-FbPqg!{e~3Je+?w_J>9>^E_BtP3$RxTpZb`1)Q|+@UlhTIXsO5KG*Y)rpsBv+Y
zO<!p{@l(qkJd)DU6_04w4%D+`oI!*iwX?G4rgGOLSPCSXSH>L~G!jnn?dJdqycwzt
z>AUwBVVlZcEgRr^0%Zxr_(}OAhIs$O<kYlJMf<5{DR#<i8Wjqn1t0apu|8AlW|Eb*
zUoC|&+W{pdMh<VZlD+q@**$^#4NSb(T^}>Vm2!sCFtQ>S=`_{zho?>NHEzuM9~b93
ztlKE?Bz`9>18@qNmFt4(VPRh9Rq;$oe;+39?iU(XR}&T(J6Y<$8GL|~fb<`LSl|L2
z=(72RfVCt<t_*<g+ps>s+PY+W^g^U%q{?UNYihuzib~Pa5nTuc6ox32Q(HWA;$%SF
zikmp#HCwJ9fv$;GD(V0>0+pR9vE_Ko*C@`*e691(nzZ#iC2Y3b?1$0|a1G%q4ksz3
zORzo(#=Z|5SY{rSXT@gB7T>0*PIC{uPQpxi92QVmfaQV?T_sze5A6>mL(Nl&iV1^1
zPdsk3mfy6kse!V3#kxftF3{<l+#U6{HB**NRQK-LK{tyFpSi_IFppw4o?Mf3OM(PM
z;`vtfQTp^(pH>>o8mSc$<oEGX_r)xT{O#0P3RKCIx!83SGb$0Vxz{c?&2P!fxrE$9
z)y))j!x)$(-e@H~4K|BRIL4nvJi-?Xb*o}OY;@gr+)7+&<#PwwSo1ak{QB7Hk!_nu
zWYl+@cLr>Hk%i4**Xl}-J=Z<2<#TN1w$Sd9Z$7~}OSywhxXVo&`Abf%W9V--O3GL~
zu@7ez*dCSZ^{JVI(}zd2OUu5e*@Y*I(l~5+raHy<j3V1V;prR!$aM>PWT2J7pP83O
z=d!saAuo-~#^;&~wB|2QoCBA*Kin?Z<e`*5cm92<y6~wE2)P1dcPXJO73T%Ch+j`@
zTMH50Q`lsiy$x*Gc?Li!nf!0zFsYtf0AWlIVDRG*Z}qz)ZWH>qtC22WjO9ugva07g
zxHT<3WE(z2%3YCl-i2iY1P|6ADwJM2J||<3NjZ7x7C=N0J9U|J<~R5og?%1i2-y#)
zl<l_zyq~ZBfu16_2;TwFQug=y@vEgRUq6rX>~fRPny0TLpa1A4X}73+*te;n#An6H
zW)<+1N*iZFR42H{pZI?;5rFbQ|8r-+WTQv`ag?B6pf4g)v~1ePp5m_a`tiMOa+J}}
z@G%=6w{YW%Btjh^?&bGz&_3H+pGsE%AoNvb_Puie<US-n_UBm&%kiHx(w>TIx}q>R
z`raTd)|8GOE^;(fzMFSFUdYU@PghAsGtyzLW|SR6a)&cRYb*MfbisOdI|@R=u+n<*
zcX2WEcMe~%%1eOZp=;v3Ljo*bcRxA5;=s!^8iv=}PBPXBC!ega<@uRcBjP1)GFGe_
z0N%=13f1adLAj<4P5zc(s$6@NQyHiGO6kHWRFxdy5HFs<IO+f9;)3$7=mLEQjsLnS
z-nY|m99dk^x$~)E4HG}0s`XTK>j0dm(wxYo@W6vLPKFHW7FZ5l57LNrk*2qxHy4*i
z@(Sc-6vkK<07+TkhY}(#LqNe{bd32_$jc%oFxSDJ;k%UNWQ&29Bbzw>fbinS{duJ1
zDG(S6xGpnwy~Bq}%g%QK94ee+WVsN&mwIhr&=c_LaH1z?SNZqg9)m08UAzeT6ad^)
zaqZaV@gT|r29<bL?1bM`8Nw$VAEk3mACt-_k8R4_B~wY3q|!*Tc7Q@?yrtYYqqo)5
z(dF(qidNClH)@R6_??}5YX)WBfAN3AUsdoP?_veNit=mPU_{^i9qUu`yO3QEbQ%&d
z{<PuqhhUZeVWKtMtxs$St%__g+w)Ou?X-|P)C+G1Q=NS-o^kyKS8oRrkA21t6~zpv
z!E?xQY!374^PGG)@T-9}_tUcRy;g;?(UoE<iiFMFX?p}aY^;pQuSlXv^A1j-Daddq
ztD3;Q8ES=Bd+JYBJ|x*8iw)#~M9`q@UBNcSACIvh80UN8#V^Fsu%%VAY%kJ4GZyh;
zjnDaDZ2q&(za8B+V#cMA6e}B>PlyzLX<YIQ{0{WDwS9sXV=$Ou8@%S?3oIuT-Rbii
zRLylfJ(D4>iFJOPaq7A}Yyf#PW}(DV5@io5U1u$S^+txROZ~E&@Z`tm?ylbU`rAk(
zzGAG|71B<wqp~7ta7x52twku};UR>}w-z0+pnj`=9YR7Bhopy)dj7s6$-U4=!^>ed
z?3Pq-%_l;XaeG`hL5Y}NdwM;OoyIb(@2t{cQLt@{9_}a)$4%|HOsygZqnv@xu+Qjv
zUiC<>g4W{!C{iE8%+m&5AaAF4W0%sF2ttFs`cR--Hb(uNSUBMmdiLvk&|4X4;ZTjR
zP2ga+`>$F0skb=iLsqruD=uB|LAA^tJ4*rK2<{onZC!i$mx7yEx8_5UHnYkXQ}5CO
z=k4Z2s670Tl7{vuj2%we9qC4l1*+0bkYlhTbf=WO@aad5y4tOm>qEOn`F0hEA)l;(
zUWZ}%Rhul2F<$@dmUHy_xcd*I!%)h5uBB<oHvE6%w|eOxt{*o3^X;ICIG%0K^Ge~1
zy<Nn7{L!CJNbj4w!jGDeja;-$xC7=8P_ycvHJhk2Y+Y;LY4WA(yg=eR6KLe$6}3V$
zv}tPiG4s%=t7E#BA7xfnYI{8l+zH;3%3GtnGja_JItt5(IT`JO1uf-b+?1A8@(*S<
zRdqbP7vKv3%;DCA@=I*V-M5{8{j~?4yP+R<f+EQg!Ra_{FPz{Yx{4VT#HQie!&|0~
zOnz+W|Gr2|Ja1Kaf9WYzEAsQ|Bh#5})sDVM2Iy2jiHiejdFiRMJ!kdhOQ5!}E*tvZ
z9>q%UeHGJ(qz!Nss;8R-fO*dw6I_OG#mZ+CP(x{5sikE!_^Z3<zx*3}4j4zX3@E7*
zqmzDDXg3n_G~5&e5sNnl6Bz~=#^Xdg1J`{2uM(&_i$!_U{r>%jk+l))SAdB|h(r8z
zh6Le7;*{954;OhXsTx|Qp;KvoQJI#{hcSpJs`iS^IXyc#c8aUthN9O0x}jnZX#xq+
zJ!j+|tJ#txczIm?j9i@|^>m=V>&-lxkhAQd4?LUSt=79F>+0os?@}=FLR7GZ9?ej^
zY^1oE2b<$><M+r}KGlSKrMhli&zU+Kc~YWl(QoQEDia51iWy?QRi#2YDDOS(y>v#Q
zhk)195JPs?^}HCqc*lHq#GI2al*QBb6m<xn8*WfKF`@ZKONQUy<j$X$8VNmbO3StU
z2oU!qkHqiGv1k<(H=7)^wcZo)-C*w(Uqc_dS*RK8so4?nxG5I&)Jh+FYx#(o66HN6
zc{Uxd>D)khn+@2PVM{hnJ0TG$RwMee{*q^aa!(w>h(edM;|K`9sF$O%T3^oEN_%@5
zS`5ZnH8R}tS<-tMH<UL@WP@qS3#nHboYmpN()ffXNa%>E`>CM&e$j{`DtnP&_O$;X
zUTtXjX)YhB*VQl~7%pjMpXFtw69#&M25`%Ja1~Q<igIB;kx?`LM*$nug*Eg*xTc`!
z?9#*<_(5#vGMZ4=?m8Go0Tw#sgnl2t3b`Acg4gzPU78NibTTC{-7)<>-)O<NTn?p|
z6X9lsNouNQPZN6tw`k&PP@ebjUbqBmI5n9Xoh7SF+O6ZEtItjyIP~7<%!i~4bry@4
zrd0U8RRP@WEoXP1wqKey$<hj6yd6IMDq|m|P^sCZS;I+08<qz~9~FO`)jD-RW~T?C
zo?%b0-e&u&LA%#8VQyGL8#2-y^85D}@qyI%_HSCnuiz5PbPs2&p%^Kqq>130yResj
z-EHaYPvpep_R#j+?hrJp{e(IkY}99Q?^#Zza^y6>$C_}64UOY<j=eQD4BG=M*>T3&
z{g7wMlORjw@y%}GZtl$eh``rEd|t2fY%z?{hU?NR$)$VjehNGYZ(+knxqQ3^9-Bae
ztCfX7!+0sgO8dl`cXm8z{l)0B^rObW6U`R2%IVhW+G&SyJervN(|B!zg!54zIG3$k
z%5>55EFucoQN5=KSDvf@f=tm9{L$sj2cl3X?m641Q)wlx{U!$F@#iJ+6Z9cHt%AHi
z+ic^Eu^{bgpR8b9*v23zx`@OoTv~~kXc4j&i*Ht-gFcTRg&5=a`1M9e5K128M&HK!
zO`xj@gA40*IY@6`n)KZwQ(Cuy5&HK&EVFMM`;a+M;>JJ^%lW+C5vh|iguXISvb82}
z*K&~Vyzfcy$xcZ(F=kv`S&fd@_*KfbZ49)%@3n%;-)$4OjXSKJ3LbZC=5O)^sIHVh
z_ch@nK?(6Y(T`zs<4Srhj`Uxy0zk>KGW0tKLY?<-_BhNQ<01&CB`3TZFcrF4C&W&h
z3iVGmC93!E&+XPQ+8EE7bb7FYK(aS1#9nO>^v&rl)?j5#$;$Q{gTm#IxdD7>2hc#`
z4gCO(xL=Gq`^Buq60M`J<M&)J+#_;=p#dU`*ZHJ>;r&&E`J5bCw?4N_!>8+ROq{Be
zFj8>2aC{H}dE;zgO84t?oB-(~ft-QPJA1XuFNj%}m)LtIo$Hbut7pBf);6)}xsefu
z#l1g&Fz5pQlo|9)c97l(hUh?hRDLW8Q5m{S#G4`&PberG2Y3%Vw8-x91+IZ-ei3^h
zb%T*9kC>sj=TuZbKF$!4?f)Z^4Y20)h>7K7C{*X-S8D79)M@{PjX|QWfvLhWI|bY>
z*Y~tpeq%hXw-mC=hcAed({%(Ad6cjih>_nXr$v2`mZu~}U8j0`=hmoS%*DB9d-IvW
z?y0D$RFBl*Nn_XXw|7*wm4zUwnND6Lzo31#wa{>KcNPxEfT2M5@aMOI3NFmJC{-X#
z?x;d3l_p2@Dl$wEt#Gi*QWU;J^<^UpLrNj@d8M8l`aX3oAIvW;1G|E+`_#gogtQh1
zf#wCJ^Paf4S#Z1YTa+Of9Nni5<Ym2M!Un1J|7fslyrC9jy*n!7ITDv2U^b7?a9Rg%
zDSXjT_KS+U9s6XHwB`421tM#B8<IgvtVZk!x{6e=i0d*RZwodLdWWEYf^GTd2N05V
zwGR6IBEE=O+ygQ>s~1BXNL$@Q8;Mt|jVS`a3sY-6Y@;nN@IgVpN~E*^nNr;6?8i}$
zhW#&H-gSCQMi__$1`W&T%5?Y(JP^peL}(+@5{!+)$dOSe(~^ip*C*U2R6aI>5YhQB
z9-KBbKTMLWqw=tLp++x6O=JUGB7)0mTTr2kMUJLHkdXdoTpLxW_D>978c69wuRtMr
z5S_vus2ImD(j&!*^Jhj#dgEBCx|>vLTkxCH-iSY_QQt*v3hvuh+?#&<B*~nh6;|PI
zh39WFsM<!*p%byA!?*5m`|!+=cSz(igOtiw(VNjxVfUK&pJbogVBO=+DjOk4YT|G;
z2I@1GyMVpASV9+oR7E%SCIz%4l#c<Y{dm&l5_H>(xtqN$Nq5GKC2DrC^VC}hI}-gu
z@N@z2*B%p$;d}dav2WJ%Nzm_4ov`Bb!t$HqkzHE=2AUy_t#a$e7rOXX#gU`ZEM2Sj
z$B9MEdb{g#>%9V9Xefxsv8%~NB=cOeJ$Hwt-wi;kH4<pG{*>|Bnq|QorX=~1-yxwe
zM5iNTtA51uld<|F0ms~&pDq73QGu2Z)EKst0Uf*R^+rDiG<xkAoYv-l7bXI%(`udN
z1qYIVA4X8Vb$L`*adq1w1{E^)=oY}ZRR<SBH3l)`rFSPhk59?wwdirs1)F>>RM&mv
zMXG<|ANjlFVwu)FGx{MBb-NvZ`I=9Z9bb&Jx&p8&<-5$yI8(qz>+m;2zkAxDKQHGZ
zHnX(ukM<@;wcgl$1pTC5Y<Dhf)5>NtBu?@eyK`2IhtnCAi+IAVLfnJ)LOcmHP|wvE
z6hV*R#R#`(5WLnRv1YxV#)tSL{n;n?CxomD^`EeysVT2NFrMyCkuXPP#`0E_55!c(
zQ_>=jv{x&H6Kw(+lc%>{l7kv;cMzsxki}^8<MHvQ)(Y5QNmYIDmlJK_#`a|X9{m*p
zO|ps5$7W&17t5_fOu7I5v&ivXRp}D%FExL8RQ+bS6w}2|GACTDXCpNa;>G(jL<r@a
z4+wSGgG5Qo=nLtqe7q8N(pl$UFCj3X=YCc#8Fw3RagZ4IFOWtHB3`#bdLMf35ScH-
zAEWmY<M9A%&T4q!D~VA#OWHhY3QKuG7iAMEci1dfmn3|Lm6#~FLunHI;?uf|;Y&|Q
zh}utu6d``O1(*sk3o$8o3HNv?2-$T7R<-io@voEq_Z@J5>#LKI4|$``#L*;6d&8yA
zzjlPf7`%xCa9z}Ea$9-!D;zP-=+b`orb65<XejM_ubka;8XJty{Qh5O4o<!AL;Z28
zi^3W&elZg4#T{rw>J%{dJ#ErI|4+bdEDh>WwLT(y1c|2Z^aIy?ZTujlyVp)+OQ=V%
zC%khjTfq-B5{A=u`EFbvZ||-UIJ&sx7)94PH{j;utp7O((H^dGvha<4evr%2v*IyA
ztWBf5?lXyZ9BgNq1$TVjccB$`<78=Y67uNL$BP!ZC)!@;n4`3{`XSPqHsXJUhpHlW
z<S>4#QtYSFktES7QXE=z98&g|>8!$cX8L;BJ6^$q-m{>m<%RDF+LUZaZ=UeZX|+*c
z6PH~HPg-~s4q>Xe!Dn-=UgP$6rlBtDHP=<cDrLE8?t}iumE`yR=m)xk9u}jyZU95v
zB>Yxs5`q)KIY{%1scNQj1R-iX3fns6kBpB_V`xumUAFSc4gJto<P~^hSs47&SX+IS
zZ)*`(8^6h+(G8}*F(DVndqUcilYkUB_-$f<ZNB9x)qNKD?lpGjgVCo>I~J}q&6g>%
zgm+(@Rhga-VmPli#wJ%^qb-UG7~gHxwK@e}=V#wB38_D&UfcjYlM)^s);?wbd&+NP
z{>qyX%O!9+UBXg9VM)nRi-dozM}j_NY=u4vBnD5)6vmQhlXr2_xE`gRgw>eIG~v_>
z{}M^0XX9=Yr>56*_wX&ludUxW+IJcSDw8{gGjoa_O{64n;ncQZTqHRro|mp{T|aRn
zFzTx(!wtTTc`F{?zWkM9++rZ__|2)TsKX;wXV3Of<LNZ<-=C(LK;EWo+pyGN#Lr#B
zR}N7ChWJ_FoPp$^@HeDnQy&OlW2TX|36neX_q^tR@*Z=)?DnDaV-(>q4iMZb=4YR&
z+m^eOxO&+ASKM=1a%{}(vLXhz=}ONl=8ewF`)HV(4@u8zy)RjL=^3WN{OX9&lhq@+
z<PQ5=)iNW_&ZjR02HM05o>>K*Mcn7(SHk#jap$LtrL}uWhE$Ea6e{*8`h)x}X&|*x
zbL1}@6^x_7*FN!FoSY<t#Ycqw1B%LST79c~YJA%HRz&&3@9V~AB=nm~ffathKRik#
z!C8sIabi6Xo_#<70IBV_VdOaKR6jLWE*2#~!d>RJ7Z(+SxomwNQ5yx?E1EIXmTin^
zV&63WU8!Xs)KPh6*%|6c7Mc{F4#Ed$+@L{pn(};#=~>xQDt}-&9#wPP?=)LR>=uc?
z4g0DvC{i(tTImGUwXY~pqUpVMRnYI)K5XH9KbMA`Q-)f6lj!8iQR#Ba6fMxVSMwG`
zE5-fiRWAW1xJP_IIVR;ze+kMaWv*-rG8PYtb)V!m<>D1_sC4#G_G6+|w2E?33A{5U
zO?O}Jj!?OJe{o&LpGC$3hbdMxj99vwy=+621VTO>t=vFX+a@fq^cpjnPW!Ce{!lJx
z)1&&S|99SqrbO+S^I7~LqG7YEyc~o7=s04r)Te5d)GcVbp<Je?g%&Zu6+EdFuUu-R
z^-!yP1u(FkNtx~6)*W(uJbBdprXQI1ANHHXrss4OohfurdXMysQw1g2nO6krO17jb
zt>zo0EtJDNYfLIP6{)uk6Ui$mpLFjk(sjam)aGUuI{A)0hHZCy?2W_J;Wt1u3pTZ%
z<T+k}UX6-|Rl#G_q#++eKFRnRTw~WahAaTxkiJ+CF$?-WG$hQ!N9d92`!lsN<gvxV
zCinOyk9`c9oLvwn=aHrFxDy3`NhfV|WXb3$kP8MDL%@%jn+(B_n3hC^uViqW+){g@
zLK%V7Z|n05sJY9t?n8Bf{PMZ5%eL#fbKv8r^?RdcH!DRfZ1Wn3i8pYxAC3?B%|z6=
z?))YOmhDh7lDu!Cttwww;-nx%>R7_@wm5ii%)xSq-~nZ*cX?%WXQqNK1KZp;%8UKg
zYDCCdl}Q*LkPhDR#K|LiP}OqoNd$X$2mb}Uqo3$Ehm+ZaporyojLajAO+J&)s_Q70
zB%Vi(F6p<oOV(9j>!F^)@(I~Z{I`Crg><>t>PqK>&Et(#EiD7^2p^BLua~8wUXiks
zkA}Lsg=B>bOxJLI?=TN_dkIYX?BxSu#OV691Fi3Ee<3+Uboa%QN$20X9b@m)y1>lc
z&xPjO4+rl?C)?#4-(7vBAw_;1a}M}b>DbUv^B2gqTmrL%#)d5GoOrikN-XY{0nCvv
z;+}-Frlf5>=D)Y-{?0e$<P5f-(V}H-B5}oo`9Px6%Opekda>8S%h8vDGpvIa{1Yin
z-9prF|C4<538Rb^ap-K^8rqaLmd)W_uG8YQ9|fxR`f)UVZPv3&&X!PMwf>?sqs;Y8
z;cc1+Bmm+*i&t9%idDV1Lcs%{%V;?`FE>r9N3OSil+CAU!?e7pTR7Kvdwjbt2v-mT
zqru~+bdcvw4PKyxwNSPVFf)IH&T|PEPx*eaYxzB7G}ERHCxjz=d%Fxb+X05+;08Jl
zlhD1!vBctZiPibBG8-cvf!YC~XQ!%{a8sxD>$i1;uD<mPiy!oYVv(<a6orp{875rP
z9&dfNo?}<zOThi8=|~EPc8uUhqBZyyPNtxJ+J)K*u%@v_LSuRWq@t7DZsGNVKntqC
z4GB4q98AdbJv4}oZ@m4y2g?nBO`}zS?*7rwslLZt(L>R>q>{s%3I{ZZ-HA*7-9N3n
zrEUH~pZ+15M>&eKr_5<17S&Wn%?NJN4O)t>!;acaFetQgjb4<wr+-VWQywSx&pn<a
zSrXznE>Q$FAFNojiG|F(Qm}5Xs@H1ENV&ns$HzhCfYq>s+oVnJ!^xtj;-Nzzn{5wY
zi2R;j<M-1kro!Us=K1U*3>?CwB&yp%YLRj3K<=a8kAy^<R&%EksL+9aB$l$-J0f30
zqOsD|M6%?c9O4+Z`~T?Sc>35_tXroRa>gki74y@*exwGz@cv%I_3j(mMxSIdRiLk_
zXc`y})@#Fl8O`z*CV&6>+hUDHM0e^cFa+D>J;&eAb|2@H`LeFno$IshO0R>Pu<8M1
z+$*$Z3D%Xy&-V+_2GcZ_oJ*-^mD_CBOS|>FAc-qG<6>w3E#w$1J_E1|u`L3JgL_Bs
zRbcCF>9)3X@r7sUz1qs8<|Mz~+!04r<xkiGKUmPm*kXUz(a%ts6U)l&uZ1OSBr!P*
z4@{C9w9y+u_mvlH$4aEH-S;Pm4OGNSeo|4IKK5uo##SsU-;gTw-*dyG*UW&>>C)Rn
zm1xTH@o~Q&=2y+A<{J0X5LW?bCJir?s1G)!T4x|zRLMTZK6@nhFPg;~gw-w^JoM>J
z$$Ilq<qC;NLOx73h(G?5y|Dn{vqmaO`+=qpS%`K&|0Q-nmo@!)JQgW8uXpBJz@sC$
zjzMum2^~*QF6b|Bvp#Y#&mK&_CKSJKktD6UPa|Ar?oD4p!p6=&3u_y22?@CBu^2=4
zc`b4a!+K7BcFBIeaML%aF%K&@X<vMHXp<b@c-WoDbB-&(=XFg^r^Kqb#$K<0+T8&&
zXH}6lYyjZ@32--1y#&b=C!UuJn&jWIY>EV>K1Z86JRLSpBO<Wk-wH?GM%+fO+pd2w
zGc0Dv&N<%pjHMRQ`nQu9F40U(eS?XxQfv$uD5I>PnB*pCGBVL*Z&kTsk>9AzA|0hY
zoHB%HYY{Bn=d+NFQbQNDxVBMC3%`Gr(IAFC9ha3RPc4SaS3&cAmMo-dk@H*#+^@1l
z+<#EN$$^t5BGYf9tY<lhi7^b`9HyGJHJsXa7h)an^Ro09SAzrSs3}DAFLqxYe28cb
z6l&~nNevY8suBNvkVje8dYN%Hl9@%psAYRFRfKAKKT{Q;t0sED><wH4QS6)RePApU
zfMXpa&N5-^_Fcy%iG_qsuwiFFdo>FSQ+M*eUEDC|b^DqB1p0wZwre;AIl&YIF;Xy(
zCVO^<vdM2i079{mjr7tS`JcJLeQ(^6`xXFxQmNoR!8XKUQtU)chnlZ#KRq>ls5$mK
zP%$NhDSvkd0PR$Ii@@T~`4=835X{I0hnCU(Sx2*$az({I5o2?Luo9y`p!x&r93c}3
zDi4H%UI}=0+2W$jRv5AYVK1ZP;`6x6+!|5g9UkiOT-y-ln$kDf!jq@Cqc=cUvtAsK
zGh9w<t7odZwi#BifWGq)!j=)doUjZ4C{ILbCS+76W`RwMqooGAB^QUEGptAh<Hev2
zBwE-7&g<`EzUsy?#&6nN(K*S%vz5laLmBIpbX8~;#A)hNb4e{puBX5RlFx$6FoxH1
znd+58h6qjDJ%E14>uv-aciVP1$GDE`Zd^!Sm_1?Bzjs8@uI(PWm~>659D&Lx#l`Ji
z4@9lXN?>0<<V=?+gh$a26fRG{zghM=rRk#%OP_etrJw1$JW<tZ<Q@Q)TWsr8)3PZ1
zV(Rvt`2+pUN7txkaU*oxfd&TIaB@pUclZ>=Xe}-DCO@FIDqqoxFmLY_3yTuJ?^hK!
zE%VnVTE*wgubKC7Jq~+UMz`a!PMwc8i?lmg_OQh}F1!>1yF++=ZvYnNvU}{g(_;u@
zP>=MpO{f*bnL83dkBGyFLEMMvBIsM|+%@3@@Zm4=FTBvZX)5w+RpanywHMuOEV?n(
z2S^t!L4iPCD0Y4Q`?L1kzX5H&6uW@x(94c}Lz_(?X<LwKOYwyuO;{~Y7URRN!3A9U
z-^!7|AR7~OK?Mlt+iGSR|C1pzazS9gdM(0Swh{*syKX`MMWOq4?*qCAA!YFlKL~)$
zcw7pI*2cRI`&3(3_a;iWykvH1zX}HeTQ$8!V_AJuZr59{7tQ3NN&EpD2OlcFMK}^!
z<m`WBIAH6tlll(R-`uD+lmQ7j4$hL|k4-Hd^M}A<X_zgKoBZWRAx5OsmXO3SF-VAD
z{+$aRIbBd-j`wcnSLK1vfI?VTP3+`90k!HbHcV_j<^+jxJxo_)Bf98aGgUJ`84<)s
zQwEE1U})moOE+L3Ugs!->6WlU;%l{)J5l6pAeH%_ZJMPA9Xk=~cuE09FLN<0S%+4J
z%e}B#kfh{CTYtNGhipMV)d}nR<i^e2q<j=7I-K}J$~Tt3`QVimzoE`%lU>7Ii$=7>
zwJsOUsGJhyN-Pw_7W?e<25X!<?wO>gg=O6q(ggoV2N6S%QBV&_RC?}*%)gHQrwwAg
z&M64Q)8o*7qU?5wrmWS^pbmmE`Yv_g5ZkXUDCJzVEcG`(`22b}efS;fgp+10*5$il
z=d`DGxjj@dj16)w((ag3o%h_kp)A|`h4HU0Xx}2q46((FC6k1v1p<x@s&x9!3w{k$
zay%y}w<kUbDFnS59VX_~vpCU=O$f(xpce&4vZP>1;wQ7GiIzt+RFqY9Eb9Fcjw3@Z
zW9w0|$FQB8ud#@QhPrG3AH9nU5?7{_OwVj=Pn7f%AS3wZ+HVzV6Zl$UdX6*{I&s(K
zOZrixmf&WEF7S-*oITW2+#Ua1_G1n5X3lOC$@hhbcgV;=^Yp8afbk<^1&~6qryGCB
zyg$#m*aEa7>3L7bHUV|Rux-5#zh*uc`V6UyKB7xtBleV_$E<yUkb>h^Bz1X7KBE`V
z#dCRX3Dyr_8Im$YJo8mDESmy)ih!e@`R<IUdI|<?CbU};bY|q`n8I@$2eH*OV1i#T
zQGe}uxlvj4_3;4X;PR>PMS^eqWAh)6o_<VO%6!@|+XcP06*%#+;D;Ta9!Kp1m5Rg|
z5fT{^_0&%p+TOXx0y~ncNo$jb;}l7hg(jioy2OV?ss##a(D^~dy=RC|o~F^oaZSH{
zznFsrb6fh4i8e7Kx--rl`j|L?uXq;Jd9-(LQ@&2|)hWe3pwOr$p65hgF3L65Yb38{
zqfH(gs}1ir*H5uSZaSqk@3)H11{yl%v$A@1PIZ0_J|X3L^(K;YXXS|+8hYj<(=+$8
zY9CXc6|k2-l3|`CBN8h*BVa(bJ>onzVkD0Qi<~f-_SD2Eu)ww?LWjC1zTbCt<ZLEH
zqsneOAIk&QZac)mknLSPBuW#sHj?=g%boP_&{q=M<o?cC+0Ef@SxrxVoeG?9s3bg3
zly{Vpj^9((MZeE{_fZb1mV+X5de!HM$HH~5V}qiagZ2qyQ4cxen5Vq^X3)DThZpOV
zY4*)&o#Ur23pNJT`R=jiPL?&zl8iP0qJjSc_dlzrMpu3~5qWV?&cAJAz$b-BuO6*G
zr0=`cSb6Co>_>pBL#K-p-|0RSD*d&(gkKWrT1EsZIAm{qBJjCRlM?QQ;HFoU?vq2h
zM-1uo2}2n4=u;9HM&fqh#?=i>T}@^R?99@r2z2P|dnK{cazQvc!AEX=cCrzqk*MTz
z@hP^@K)_YD!E6b&XbW!2gTK_E%<j_*XG3-lmH7KP`USt71}Q@*w|OP~gPB>jwL-@8
zx*U52NBS?c9|@~U6#UMQ;+c&HJULznDmj(L6ce}0YHiE+yP8p9bIT_!!M-WaM<ad9
zy{+rnA<J|243^Se8AMOs-Wd~jkg`EzXZB7_VxK;R(bW-D4lOwbo)=#){#!ME%oaMG
ze#tK3q)11L<nr8h6r*k=02Q^D<vkO4c{RRhLxF<0G!i>2GnQo}U;Zc>dD8VULJMy*
z$e&ah1B!&Q@N_Erc={*KZ^*YRtwY4@$GBVK4UKN44w*Wa&Q8#D@!J8nn`CT-8M%Ig
zuk%Ll8mEchL@vJ3nz8-Ns1c$4njOP7NzbktSKs~vd$w8Ez!=rXj4%J%F-#e?Np?;9
zb^cU4l_<9dvZhw0;$F*gZo}t4%O~zhP@oA{;!oRTw7EPCowbU!+K-`SW!75Oys#eg
zO$?T}kl|4b<j(?a+oWX^mXWtX7<2%o(SP}`Y792F-XuQKG9(2V1-bj99)w=isbEm#
z8-D>~8~uO!J4jrRg>d?O<lnyxOlk&1+6fUUXn_f!;3S(L85^gyx2<5|>dt|^9<~hr
zaRQA6WMezeWYs9WC`9TV6L%o6R<WCsg|S($*0~T(NGgHw6$!S%1i>9*-Vmz`VFQ`D
z+~|DQRqs6td$Cne=pSG?GuY*dKWpl9z;eGU)agA0-=_Y=dkNDC!e^!5%P=2w4}?;R
z@6e!qwv2>x5AX=p4ADS|ws4;yhv@N7-VPl%^;HxF_*G1HcFxT^q)f+4qB&1!lE}^N
zis-5d*Hc+^XwUN&;{gJ!$RXHlGv64`T~}=`Ka8{0`V98S%ceVm<qE$1r@s2<4~n<g
zAxS2$ZFu(d8M(*{`_Fyybi}HT_m#DB`MciGJG`)Sa{9X#46J6spyv4co5nXeQ9bKl
zU?c5y)N7e!nhlPBP}aO}GgbIlKO=Qx?V1J{LZzMKHZ4dvA1g1HF2inn^&1_Tex~t(
z`=c+uaX72zi;9{{)u7$qqiAU#QH^FDf|Vg7pbJc(x|dLHvXN}~>Al#`3*2{vO&74y
z?#u_ObFim)%9E-pOo8IOyZHf&AG3(oCUYXXVZ@nixr``dg6D5Pz*;6@n)pB}a_sBr
zAIph)wF9O}XxzRJ_RR^?N)nhE+cpG)SQxXFLBSGGOg$4wT3Kbw$inxvGcpr+K~0Lp
z@1L@$3V&98&nUuzFlGqh{-i#^_;)7|RSKis=OMKRI1M_C^(VQTT4h%{I{B`}$}u|%
zI!;$YQ=T4gwFVN`I<4U-R}ZNxkscw%PsR8nX8R&b!d<ZtQZB>IBspPN^><H^iK~vr
zmuQ%Rb(kqB`$XRt8?VnBP%UvwXrUdQk<{X%r1q>oV99jDNIi)x!)r5U;_FZSqRtf&
z&m#pvtrbTpTa<qYT%=dIw93teVSi+P1|HW8VmlzIOS4<zV8l?py24&#7$6YRH)Bq@
zW;|}P!oh?S{=qo^-YqQG)=l!x$ZUXNAnP9~F?OubL9eF?*|ojFDZtPoVu@?K1_4H+
zJi`LwDB8z(LKRbbAqwxF4Wn+)-q#*Jgd6C0IR3*7uELw*PXxaXyNlW*=lVmoV4+&D
zcKwuVGM({k;$sQSfAr2i2I3O50{Oj$|K63VVt;*(F=ZD&6~-Fj%M;x*=Av@-kCD?m
z(RAVazUVEcw-!uOcQVgorTHr13CaS;?L1JacIBUbSOgfkjrO-DsdBjQm_NOzD`E^(
zM_5|_{<&9SQ3w)k@Z2E=vxhVb2!G{3eYGb{%80t+!cI>m_}E*uw^sO4GxDIvuI2oP
z*KeYayX>o1Y5jqz(I)qC;th9+vc#d5o)@%Io2vIGN}WkZ@gn;@YnrDaG$^G4(@$-o
zZR_}NC4oiIb`ws{VD_~>7cvpUDl)_~)D^^$36wXEYX>%Gv*NFDSg>Wft?^}eFBRba
z_|z}je`{&4;=XS%MkRnSz*JRdxlINlt}*CAtgaD+NRjpNd#HHCBgn4<55R*5Wap5f
z*?2jyRidcof9ICcO?EY347z)de?NPWqrKXDoSOg2`jAgWGn>f!-w&~mynbVXb9WW7
zHY5eXF@`WliZYF<oY2jf%P}!>r_Gi1c1G470|2Lws14~ld!8>hJH0ATPrt>vV>lDy
zQ|gLZ$5NJMe;rO2>NxN+lq^`@j_8ODUq*-kPwoft6Qm_X`gw>co*MPZHE(X(Ifhj_
z8UHPJXfH=K@A%Y<TdaM8RV*&Qg%%2~<~{0SYCT76;c2XxWE<po(N%nLEVw-fk(z<=
zcl{Au_RWG>!%Esy-2m~mg4{KV5eAKr6Fm7Yw*=h`u>@o8wpnjn&(OnImt3-AG1^cZ
zmik(S)KKg&YJ$FFW{B@D-|b*tafl76Gg|n6DsM$L$tP$rK~WM{-=xqhuZ#>844wF@
z&&L<a?Hm{0HXS;}5fYb-xl9w3oNC;9HL)ST<DdQ3V2O$>irrD+?b=xOc%6QACU{?H
zcB*tJaaZ6&u>9S84b|Iz52W96rR_iIN2+Pa%F1dofS18)x-a!cZC_yc^sR?&$ouGv
z!<O>2`(_nG>OaeR1pzWY(Vi_1y3y!T0GZd4p5*c%yDTb`J%d?(7gdmpEH~2ZOD@}`
zhm0nQgzR}~ekopdl-l4Lx+;PwsyzmFB+u#{CF3ai?ub25cbC2tsT@dx;%D`9E1K;C
zMnI<I7RR~~jCMSd`V2DHf32s8G^=!YWkXUscl3{5-eyBOAXK_HbYu5D){wyhr@Mq1
z6KUd<ImKjEh1g(Kq)bP6G#};?bIY*Y=L@Si*r=qk=Nu8`Whvu@L14PJDJYMtOEvv*
zQolpa-ge0gBjs)z2@}mDTOh7->aUS1fZCi7A>V&@Dqmq$nHL~ub>5i&azQp}0>oUq
z#P$x-CAFx8ZDq5C?;owI@QSd*dN_r<B6)p~{&rmjgZkh&t`j*F`3`}Gbrp69tK|FU
zx9yn5nHgvqUKv<?am^S<ViVl1y!h~dqn%nSjBTsuT3V*o&U!<K(?|QmBCpRqq5q-0
z|4|?cO~&k~^Uu{^#XFj`p<Qe7EISZzXT8nx&VDk&3YTNMF!_vLz<r@=7`C9#y1eEc
zaemG)dXbBIc@x%ZTqT(ZyfNV9Pj;;vZ6Xzp7U@t#0N-D{*zW$i@VpoA?5TrUAfJRt
z)c!L~4GG?)Zf$iHQzZ#r02hzyQ!<m<lNQ^FTXx{%zxq7!TEEpj(Lt-69bx{hMk#{y
z{q9Q>CozmA?UT^Mvy7YsNBstDa_msV5Qm?3IC8IAC}P~KGwkVnWkI14+T|J;Dj26u
z@5YIDD(Mzi=WnP`{op|1ymUhi*D%9Ri#vmsr84pI*IdMG6}c0pSr8+$n&cWkzF>5z
zrmLd7dupKP>S~Z%nN}*bWP6Rog1pgd!elkXIUu>$8byEdA2ox=mYpjlx$qnzE!dDv
zoK=F;MX~&X-NY}Lz~XakGp0_QgECxTAKcUGIFZICk5E-Ms_Hj$|KabCo=>h*5@GL=
z)^FGwVT17fMcJEnlHQ1u17^4X@@gknm_e2u><3qC>RhI}HHz@er8mX4-S{eu1I@@H
zC_ZWIuwF&szLMTNr<7(FoYh0(M3SYNU=O{FN}M6=6v4B?xbLEinHr{o@5Ubo+@=9R
z)j!s8))(5FMDVKKijVh4g^e#kDWNbS1X>6(9fMScZ!}q&TU*<BB#Yik0qs(m)9fc=
zbIXfQzru4Fg;bkfEC0ueK-!Mi5gD3PRd5=a68|+=rsXmabTGvQneU6~NRCa?sEVEw
zZ@H1vg?5})PCcHu=Bv2pYf0EGVZF+-S5+EMs0<C1M?!fO{0@;p0W!^ja%`pl;#3+@
zBZi3Mm2VJmBI`4R5A~o0=M3IONiU}L%^wp#+l@Xgtw|OqY#irFq2z6;S%mPY)2QVx
zK2<EdvdTyQ5_8wOD?JDOyVzrIS>+3h!VL9iQ}T4IUzyo{^@3H+Vh2POlU1IdOuQ}v
z)ufj}HDM|HqF<0Y>cSOqEW6RmKhf-H=F*cuE<OA8-kPXKf{wm3Mxw1ajhFRvihGRx
z&LTyb1I6auc?pB_{T7OH_%C1?2S`_)#PdF6=OT#{jBl~eq%9A^UuIi5HV-Y!D=E8~
zYb*%rb#63XHDkSP;EDHaVq^H(e{pX7l^ni2-Q*2qv~PNsO|n1b<nWaL!OBge?r1vo
zHtUt?FGc^ON)j5pN%OVko2>Px6c?cveqHFlPhU-G)mdh7q7)z5yk3;P-N^5ppz;WA
zYdA1YzSs}BF1iV9`u{9|s$qR_d~#w-bu6$-jU&c2k~kY5fakd@<SgyHpyDFO1XVM2
zfCIY?Fh0=iQ&p4jHtMZqQY+FeF2@*JvI0lpy%y>-#{1xEQE*-@$cp(kwPWm5OioRQ
zrJ$MEGnBh5w1^sos35Ni4EYP|p<D&88jpm%U-{4*meoR5i68nkD#C9$&&^T&18p5<
zumq}9_TWiCt7p2`-w#*6yv67F4!zRqPvkaV8MI5u)}g&NpUavl{-{F;NAId2>>3=6
zl2@rFTPznSqJGIt#Av`#DfzbT^{>91Gmm(|XeKnwcfbyfO6or3m}A-eCLgJMxmhpG
zA=6#a^CC;iCux8p`?D+RFN4hcjnA5ULm_ByojcrXDjh^0UnxZ?34f6|ix@;y4k7+J
zeqw6<FBfg(m{0Eo^<`KkJaaee92dV$J%P&(!RG)&?>>yJ0cZReZZWS`0Q@3)YGTRe
zq!pmmzl~lr=bF0uf4|()LKUI(zgNqY39v9<b=BUSw%n*2dq!K$7#n+f8n>PrOd3=5
z10%&?OmxLFiP(fyU;efGwGG^ty{i05tivhp0hw(=XDbh#&er42#Wk)azClZJg`o>{
zp;8XwjFv)wNa%ms$Kt~YA+m@}Du`bO`NE=ZqtKsudeb;9%d^wSIVP}r8|ym=VqCh5
zB@6=pCF_XG#%Ku+Z|7t*SVtpy&-=!OnNsXSc@%aq`7%Cbf;8G9nnk&t@#>@nm`$<O
zV#oB;HlOseEfbC!t-8Q`_WqA;jDgV;Q7IlfYR7#q)~D}hhDGg7d37Rxt^S;`v$EG3
zpD0_&bfps9aa)kI*KMr5|2Qw#;<znXDQ;98gY$_3og0SzVm&3ih=5E6iJCscb?IjQ
z1-~MN{QoiamO*hfU9>PVz~F<s2TgEycL@;O2{veOcMT2+!QI{6B>{rFYe;a{p!p8Z
zeZRW*4^&On!0A5STh?BCZBDRB2)6qJ*e2+_2LgsjgXzX0_$KZl=nACzircUvxML@R
z3RUw$d@I#0-v(B6ZJ9dJLl!0$fpsmp{enl)q~HSS=y>jskg2n)buA_AH@R?dp?^k=
z@v*;o-<rV<ZpO|kawSokU4p$o{{0gd%Vg83xhdM@UBKq&iv80|gO!|yNQ^y)qSb)E
zsOvn30_jPPYbujEMiU6I9x_|svQYV~XD<r17h@lpW8+->``J7O<pxhJ!E?Lneg`&}
z(E@3{aHNW3e=Iq#%eVVkHkPq`DFqNoJ>buoggTt-A=kCZWZS$m->wdayA9Rl#N;$e
zHxuUxy1|rCNF`2?I9nhDc25wD&k$d0FeT8aDb^a{2f-j)x9Jl~euh)|ND<QTg<kDF
z5C0*WxA*1x^`0B+blg}|=Ex(1a9{IEh_0EF(`ExysnaSn>==!-2y;;?x3l|8K~Pmu
zScmA&XH&0aaU;%C8G(2|r%({Ahg7e|*YS)G{&T8r5yZ|Z3C7>~xY0lS@xMeq8Q9C$
zbR2T~#@@jTH~bOJA3`#@yu6&upwgV#yzWy>#_64ci=8VjHY~byOpVGwO(F~ca@Rn(
z#U&0qApsV=TQZ9;3)P1ZZhc3i%42r@m%eAcnHZZH%N9OZpVK#1D9=S*ZLFVeqfI7C
zVeI{}|Ku7@Ht%m75#yA~M8`RSf7s>HLxUDo91YzMq#vp<vKSC-kO5iHOghR)4NLxU
zJ!ugYXB=;E*jb0nrrb_DI)~A1luIi6mXFfo8_FXGTf2-dxHlfDg=KLz?whhN>sDVs
zEva(MYbTWP#1p^0Q#X9D#pSxzyu65ajz}9OM1YMKhv}s0M-DNGi#yXgPdVomC)AcQ
zJw&OJps^aN_{jOt4v6SlZPRkS8MYv+H~caXIu}9M$t;Z`24*oo`(DwNq?@Q1^b+KH
z_uGf>a$x_KXD{iY1+y!4!rb&vm0|R^gg}|M{Nk&-1EK;&e%4ZueFf0gwWw`xo9jv7
zvLSyA%fTZ+q*KeFlv@j>VB(Yhn1U*5AUEs=z~D1L8`Tu8*;kd-K$yomEczeU7wB|3
z30#-x%K3xolD|W~TXe(vR&=Asx*L7ti2Lo4%ax&EGs^fG$4A0;gN7^4jUxYfsv<DU
zs^ERF`^Sa5Kl|R`z?9J3K(<OYaisT=u|%O1gH6qPnT>A6=t9Xe^(YvXA0|x8E9RFV
zrSuF8316@~%FOgvPfNX77~i<gnP=GY(!L8KC%Nm`jdK0j6{Oa+|1;8K>SgBL1C<p`
z)#&}VONDwq87JwK7+HYzu@V)CX*(PoQUHaq?N>PzNmsEjB}I$moxc1ts#zVXRv=TR
zS_-I;*>d<DJ5zy}IM`?uIlt6$6WG+sTX@a7f&U_^iN&{Rc<d{qM5T#$Z_^C3+JWdD
z<tTIVHRp7igPuN49=BgCAGGDB(_GIbP4!D|&L-24c;b|&rj<Yr2lTq$KI9CfyPn-*
zN=)It;Rcx@Us*B}ikZNKMAo)Uz3{JE^-Vbc{G&w;gF};=FPQ<;2Xj_usUr(UKDP>^
z6Mt-JuD{p9lU+EjJT_orVf?TX(*7Wz_?*5XK+ATmqP)xAp4_02f{L6~$EOoHaX3_e
zs3LMelBVQo$h$>4&mqycdqLnGQikU=mnlSNtpahx4LT1rQDGw7fjY&%dlfecdc{q@
z$(;bL8V&rirF~fBc-uLNO6b!cav6mmj~nchQ=+SbVUR{j-&%Z{W3$luo#QDa*<6<X
z?BFY5V3C{bgzK0;#p}+xtD5PSWR$6b_bu#@oG16TcKPI2E(0;)dIo~mLZ~N|cteB%
zGT2D~5J<)W!M=GX6$dhh%EW=ds1nr?FroZlq@}t=-NvSH*j?GIA0!KPl&S;$7QO~t
zs0{w3O${>Ec*MlL18F*nnxi7#-*EAkAE3Hqk>0@K6M9#%B`UddDc5p0V4=#w+APse
z$Yoi-<xWQUhL-?~Ben(1G*{hDOMEvjppa2@vGaKNC&jo}^yMPn6<}sf1DYXA+9NlF
z%KBz^3$`y84a1@@QnPKjFY)y|&gr9l&V}U(?t5qXb(scLsR+Rg3&&)`>w|*3Zk>u7
za<Zc?rLDzD?G4L`yjWa_f*z+D#nQl>bFFD?`*8}vDabD@YE610Nf!F|JKT2T>si%7
zJ)j)q@gO=dBAYOW(d?bT;paUjYN42r6OG!z>huQo2vu__Lj5o<Rrn0rhx+1myaPTP
zL$3C@IgP{h4&Exgrt*Q~fdq4OQ?)P<thg9t-wuR9AyTOn)BSiV7y$z9{7uk7kE8+y
zaXNf6nelu-aY7SV>iJrzRdBZG(Re)Tyx+~Z20*N)@;q9_WCoe~26wg-bc~KN9PgL{
zH#<uWoiF0+I}bBN$25&4`!IMKze{cAD=p4>;07>pM9rl=LVE-IeIYN>a@{<@Xl>bY
z`7{iv+#q|OzZX1TSs!W6iK;KcyTW{zk??%}9@FK1{)2cp?28K<1RE-+h<6+v7$jy3
z8BUKe*?@fHj+k)=fn50&9S{f3dq$-R<eZn7GuZmP<bPg2<r+zF;J$f>O(kAHE&IAs
zB~b6yzz~eCu719BH2ySD6I?4?0yraz=N3Axj;3Fan7hRJpynPj{qny5vfNj!ua~WM
zq1MdvO&%0|)FwG6hu1*;CJU$$?|m6~0_t!TgNQEQstX+_Zg55{1B<@hkJYC#l$hj<
z=zi%-o7@?|wr1J`NfZiyqF$;VP$AoutSS=mdw^VWQR$^6*n`fWVR4xMy1KskquIby
zTDY;j=)j_{8nOjXChI>Iucd9aF+}@KZ?GV#z*)ishDZEeW;=0P=n|ZaKQOstL>_Zm
zIF4MnJd<w)9^Y;f5tF=<IxN%su<XQI;SdEbIFNFR4rD^t=(HpA01%MK$}wSKl}4lq
z49yuaJC=dPXTAVLlUaZV@NO|fP@Dt+nDaV2g&u$pYmSQ&V_yIb2W?Rqd(DQBzAiw#
zNKol$buClq4>{>f&;2e5UYE^Ar<WoAB@a7Sczy_<6Qis#$06$0*91f7RAL-ibxLYZ
zFJ#jcf^?X3nmDM}))bn^Kr1Zr_D4Yde?*>#k8g>^83Ms16DSlpVv`gXtSyDcA8@b0
z_5o_T$kTbbx?#wkvWVuT77QatWOs|9#eFq{DNlaNxIdO_1tkgst(fQpg1M13*H)dx
z`#QBmfDy=V`m5=aJkk>aGm*fyE2Mo|1n&MnfBqcDbJ-^+7vm#+2KM35qGATn!?8Q6
zH^2b_H=vghyyf%h_88}6PJfkeC6JTTxsR`3PV5opD#AG6`J0J0di!p=ZPEfh92~Y*
zuzVVaMU({t+r5&aVnI?8t{b}hat4pRy3!P&nKQRL4u~hu34=*%yJWs7LXPV%p5qMX
z{GT5fPr@u4w<2J}kvE4*Blf93AA*Vz$2cT#e+i%3!VTPxvi6h?bI5%`4wA)L31kEh
zhC@}sc($x9pj43<)R>$>*poorzEAL(A8>GRH2<}ur*1z1@5l9gOFV1{=qvvXnJpb!
zoVfrrN@$@ml#ZrGsm9u>Y>!PAwVQog-?18LU1Be#d-<N0)Vv5E=}DPJi?_0?TN;+t
zbWB~a$Gi<P%|r+v?%7a~J;N`A@~Vuwyq8It#*p&<z}3K8S?LJ<{{5Sw2XTkfroow$
z;O=<2Alo)o1i1+Vw)Xf(6KQs1bz=XspS4(_C@NZ&?Lgv;Q{D7&r$GrdDN&ffUpm)l
zm+@ZmEV?&a(ZqtL^78U=Cd+viwR}t4I8=F(rpbc{ur=Z^F-6QPT|T^7&QZ_rW=s+m
zD|HJ}-PsCm<pkv=G)0|%%lZ6xx0sg(w0^r}uno4BjWfJ<=sy5*M)@3bM+y)ra<AW5
zfBv1{u6}yUsSL(435UcYqd~?z#h@e^0)&{L2tX=A-O`e-?>AgO@+qrMtr$Nd^)l^9
z-vSAU?f|ITYRj{FoU9aWgs36TV(!_k4e6Yd^S|kR$`$e#1^!5Xj7oasg(<;8`qsNw
zDp>{OGxQ3OPR`eNXp?4FKMxDB29qlH)rv0#C4gWL+mEVg#wdcZPJ_}s!gEW8BI>9K
z25(PRk^Pw_u(A25J5T%b%(!KTM;Qbqh@>A6TqiW^8%!`4uksSt!>2AoMj0aqKG}*-
z<{_?ks`~qPe>`6mL<jdldQXs%J+rAy;z@WXe(|ApgC}XgB(S}P5t=_dKwp@DU6nL}
zrf~CxL|9wn1R%bq^}+Q^Z+>_nf{TScU#2m52xDL&s`$^j<Re?87L<hpm5>uLM-CCB
zR-_Lyq*z88ElAxcJN=%j1QNXcaVs>9L68PWeT*JK48lpS-&h|*s3`^@^k~!BTD=FY
z!IZ6${0gD?tq9vKk~CVZ;G=M4?bx`S(l^aR${gVf{%iump?R_kjR-p*?bM2Eo=*EC
zpe)y0SSuMuA*D?zcP0T?DoSu8Qc!BHT%5wT?6-N+ml;IU65^O@=3o=2gQ65{LieYo
zU$VqDe=l~1R+*|-#C@2NkxlS>OKGU|^^q~!(%FaNx$T<zhPMPscV+$tExH0?GUXJ`
z$A=+|e}QXq(0MiSVXTQ9o{N~REnU!qWgX}=10j;%X;mAPaUo8<d8gm3#6Ag3H0{KO
zP@2GY01ctqH<8{e9_YcaV*(GiC$^^Q*Pho{9Sx8-TXIqUIi5fz@J-;f#BYWetQa)X
zNUQPmqEvS%DJ5ku(S6k0YSpVodYD816=#qcf(+JJRs)!5T>OG64&h*F=-&O3KN9LE
zDXjdNLP7ghP@g?SmI-zLQF65Ga{-0Lx4r3H=&Ina31Sb~#+@N0p3=O;0b`DxJ#Zgr
zcsN0CaaxIuP?K%q7Mx?@bA(b%h{hJ`X8Toi+)grgQ1}D}-4q6nGZZBTlUM|Le6BmN
z2hrn(B3|6XZ@UBy3qt};9l*-u*<PKLY+IZK2!u(}877@h6fU7K7$2A*8#|ofS$haq
zI9pEWRNy4S%7Adv^iF6((EZLmpo;NTQ}`&r+ePfIY%lyw47^(!JxRCHr^9X;gK*~v
z(pt()6>vL5=F<bB^+z#(74`n7&lE%gBf%2VU6D;F%%XyxgJKV94T}`#<vV9HmzPX<
z>)=T&^4VWs5Q_{yn~1~4jmiX#8R=(vb$tj6oybhRrUirHg2k*T13^yT##E_1kR@<P
zkOd9BBc?pb_8eux1A%U13I<bH`MFv}hxn;D8bPZSDtysKds;T~n$F88j)*ouhULU2
zu#c-!r!Lqi;Mr}onras&7EZ~%F#*OH%yvjAl@XE<_=LG6K9`Pva()kp4lyAjSxQr~
zX;DH0ni3rUJ>OL<-5iEz1%N*XLB3H!;qY0FDhWy0OwJJkcj5aH-4n!@p1*stxkE*J
zkMpv*IZIGlb|_Ne*@8{BGSmL$*N(h<<BHli*?VZp>lnVj-wST{oifl%ZVG?7z2Y46
zoiAO5JsuM(9MVZo6iTFCZX;^*TpJHsK1)^e5{_s@c1;)!a<o6=3|T)xC*fNrMa4F;
zyHE<$ZIU)|s2lXu{Umc}wT$&uT;IMx*;geTjVQm79I_?OmN;Xo-_!=!pF~7>%plk?
z`mwX~V^dBERNB92l3&BEDCqqEDv9ZYFW*rud^Q<?r3+0ZXr!FtF)4sM5l<N62GgcH
zl8yPejIppu5kYw3>h57p5z<jpx2&nq%m9vOC<rXo8*usir_HerYvYbmun-X<FX_uj
znR!mQ<Al`|_BIjxfz?F2P-w_y6w?SR!$<q~%&31zFRyhRKrXtsK_<qbLfMKCE2C3<
zih}+Eo`@Vb3bOl+zWDdD7~(7V|M_P>tbP1MsR&&;os=O#{YcBl0AZ3Ru^pr-e85ZT
zvFD$2!}Njv{EBAP>TqiwDIk*_<E|S3BNB+#+7|&~tKKtI^t=~7lwd5S_R2xo#8q5|
z4)&}sbU#bElb1s=9cxxJ_#xY}9nuLWn8c<%1Cn)MV@WrenZhO_0*^zicTb3KCj%W3
ztI!E?5*7%1>j)#ywvJ$ewgGmL&5VcQbF=F|Af70!>3Pw+w_IAawl721zcb|lDWJb^
zbX#?<N*+jOQ8AWQ#C#TO3q#T{rx{fiDXXX`9zPJBPL7#RA{9+gIdel9SK<Fzm6-}3
zhm71bwjYKF{4jm-#JE5^+?)Q43{qe#H7UM$7vzP=X&-$~x@}QGHa@r7ZUuBO+Lo3)
zM&!b6iv`40js}+v`kLsKjP@eI>A|hd=0t9E`RMadB#dGCL3>o~SwbgiMBY#FZ3!v|
zoT}mbLAR02YKv&=$TRBD9;R+C)&mJ@!y&oIIr_pzVA?5nu9BPS8$BOThq3!uwCI?4
z2}sC{@_DR`cj$$?KAGxhSdHl&6T4Bf052n25Y3mWZ^_?2*eT)C-pSS8qv>j%p0|8-
zncapcA^}Ez>nB+0BmMiS^2~91C7h1mOGGWwfAfA|RwtU)w%AsAAUC}as=AF?1el&G
z|L!LP-ihu$ko)_J|LqI2G3I1mQvU9vyu|E#`ICO$L2x(lcv1##7G|a&b_|JYoBa5s
z&!i=+*7kFj0#rAK)rx&?GvHqrV9g6n_vQqCzZ*oCA@8VW^9a2E%89(IWD<JXe#An{
zz*lf$e05hE^K;$oQQcXm#WLfG=ZfHAGx_cSy8QwKh-T6_wh)dh_nF6xz1bP1MKpR_
z(wYmQ<J5=T&cV)Ehfb?AQd_tWl=`4mKXCWd@1SkOuQs?6v2N#M(7su~My=d#WV=td
zcUIj^;fmqFeh@hBKbmZG(|so5jzd_1NqAY)U+nWp#Ak4v{JRdaM*&|Acn}!v!?zOl
zv>b2eGp+uOxYQGiFbSM|GZq!2C!zuG?sz=Nx3z=vfst)tB45vPpL2wyG(j>?2I)aP
z6~0(zUtiShA&oPRUieB56)w2$`oaZ_7ctYo+)COrdv8Vk##Ll@rU3A{o)EZM?D0eo
z9?Pfi9jeM^o05WoU|Ew(K#I9~32(Lo+FIkWUpe}bzK46F!BGJ*s1FGTZ8qmsrE^3I
z7A|PxTL6Hp&tY;5S{DLJ4XUw3V<18rd-6OF{Jzee2;zQ3rASCVUO$BJ2!P;akx5tP
zhL@^o_Sh8REYGC_86l|%shp++vqXhyWPIw9g76J*AgMH|JVIAvoQDF06l^-)NV7zK
z1?W&~BcBuQ=&TmbmFIVUVK$bP-{?lQRz)e+hn0R^80zU;#4H{jmP74N@poubL@`ls
z&5Wri%1u_C?J;I%a!r4IjNeAQvN3PYutn`}=1HkziLgReVsEraXJ4@38i;;;Ao?L}
z8%}QDOA4{%ErkyNDK9xKg7T2eapkU`8f)`M1)S%uh6Z6Cc)PEKWml}L(H#3c>!@HF
zXuxJN{dbs81e6SYLaG@k#G{RWNQpzA(q%>}*wvJBDQ+Sil^#Q_P$vnqxrJP0g7Nbz
zI}gr+F8meu<k~F_CX+`z=KQd*(S{eU@==ig%i4%RwJw6?i?}bQHTjDk%r@$HmZByo
zRUW_8_G36FR4pjH_o+}_&_58^29Jx{{pH^I<hQxtbKFYxoj%nkm2F%0uB;?m9cw+8
zR_v6J?LzJnQ83_XE1r<OzupD>XuBjlkO`|6VC1=tFpp0CvL$e6lZT+v`*|!BEIZ{Y
z7AW?MHwixIp68V7a9!R{BkXU^Mo{N=ILTb=7G*40VQE>`oO)a;uhY$NJ^TXtu3&Xg
zCs6Q-uDbn@$wDj7!$6H(u<HCrV4L(Oo`Jw(B#C%VWVE=AX*Zb;eU!WDLCLp7JCFzE
z%~^dr?6{GtA_3m9PEw+e&BZL0Y_?r`>74o4ed3$=);NTIn3MSzUUQ31QY6q({B8cw
zVl-k@KHQM$uyfQPyosNbz9OdO5d^0NnRIpxy#`<j)NG5+>xNh>6gU*6$Xcvi`=>gr
zp!I;xEy=((C@>=HXc)`S)C~SvvH@-z!9KXo17F+`t~!-|7bedT=Ufb`dtD*A>V4X-
zq#rM0z#dmXaua`Dz=aRl`hE<s(-lQ+m5&;8a?JVKGYtmmVX1C60{;2^nX`P#`;LK~
zZ%0Uh_;fB2{2g=BzGc0*doEV4KO7gsbV+;wv#{WLa7L1={XMSoA;l3CkCFa0s(~>S
zJ%?+5K4;QvyL(GaAestb)Z%bl!>uuAT*s<3@bz?TdnQj<QAEKYrO@iOUE&{v`y6}{
zy}faw{<->+CTo4iW>y%iY>0pJkhW#<1#e!p6(1ZB%!1bypM|5(RM}Yn)(uYNy{O6X
z^FczeX8d8#zXrw%yE9%4U<_`^{Wapn4n=LO@DOmqREoFEyy%Kkt$5Xo#r+U?=F`cw
z-C>+EzgQW+VBvq20VuRujKOzg4-3WK=dm_EFX(^Lb&uK}4H}kXKq5KWmD^c1Alm6#
zNakH=@_3m^zCT=8hM)W+rMi6^<+Xm(`dN&ry0ReuQUJx`s=%=#%g9!Y5@NQ#U21F3
z6yY=VWm_bhfNWy19i44T$IL$5%8%j<iwbR$ip!iW4OTog;v6P8FltX572T0oW)n)l
z%J#<-Gnf)a6drQPxkb+93vs}50|$#{rkM7_OL&3H9<E|+2d!p2<9rrE9lO|=r5ZVu
zQ{N6&9vv2Re-b+Mc_=+QxPv0CXb2mP{e&V$b~ibO^M5qV6#uJ*W-6L2r&74fARZEE
zg&Jx#{^4f?ZHr2x=DFEt3Tz)LRV)V#(>=(2Kf@bJ9-d!XBosEja#YMAltcqA#yIXZ
z=QQCFDiO`IcQLY!B@{e<e&ZVw-e-?!#Iw9nyUGg1&_BD?XCrD52oT;pf;-3I`unRW
z*j$KITH+t=ap*0dTB&9*LSdKkfk1cJq`L3B?T{W@jR_wm<<{dV5S;R8CAt80DBI8*
zu5?!}UanNjb>FM;aXjn!E>S?nthxA~k}!WuU-XAD%MZ-l=ewtq);`<W$x+;GqCyT3
z!+|#3wb;v1o6<mGwGcR8MsAU$af$RCSdnPk#cItQ&4Qm0PXxBr6;jC9J5LB(UAp{G
zwSe6Ly-Ft33FOet+rWtBiN~BtGAaojx`R30@#!0eH_dEyV15^pB`It$<fpvk*D@dP
z){#!ewa~;8!sDcTXGWS*Rqu`V05u%c%1O=14yh9w`9<zn7nXKPQ;;9T@s{jPJ5XL(
zp_JnBii?|Q=0umm)b&gD<ivcYFEa~u;4hl7V#7ga!n?S2B%-%NL#CCO21qyWciIGa
z=*$eQvO8kv@|eH&hhVO*9CX4XeG>v3FX@m9^;XeBA@!&g(B)m<m24v!7drasB=Lzv
z4Shj;Mpobso`&^uKU=Bc4`u!P)9Z2F|FBr<bgazPHAKjbY-di{&DV%hyDlqZ-qjXF
zW@|qiO@Y|C<%$Lin~s!9gWN_3_nFizrhiVipt7_kKVqRv=`qgkw`1hR^;GL7|Juzw
zyrq-QYy&AveDIS-b9rY4F5;Eecnqn#Up%h%2X$9A&e2=hh)F7?UCKE#B;yS9AgZv0
zPv*_sF3}5(KOu(#*u^^WFk{ui94Ig{GNgZQ?!w#*+3uQ8myd?p4GSD|SDRTvj#@3}
zKs?%7o3pY<w{|zynXh}ha;V)>6|Z*e86$)g=MY4bpaxoW3Z0_M2GXf>$y!><T(TIv
zZKc@BNSc!)j?%N$FB%o79$V8dI4eO}0u|PwwITNJ+%16wsfmLNYlXe0$!|O>RY^9=
z28x7={7*kke)H4Mg3YXNGV9dulW_G{o0ABD(>y7&^}e7wug7^T)fnSl!Jd|=BFRIJ
zGA~AZ;QW&5$SDOJ+*sTeJPeZIu@-eFy^;syA1RjNR%0~9XP{R@hM(S>=OahQ0(Ibd
zQO>E660HS4E>4;RW^G^s1bfGk(?9bGll4;``8&cw0u41UP2nP54JL7_+_dxLDGV`&
zKzXd=t-ilIS6~i~)d%9S+4E)Tw%xIqPJ~a|sq|yv5;RX4u^Seo^b0eICNgfMr(78b
zp@%CnDCQ@)*~~tnJRKgk8wS%xhT83?3lea44_lc?cq|&1j-CS-^V-ImJe~Fk4^d@h
z*Vp~R8X9V{veCBtb@A?RtNP0%?FDI?nkdR%?Rs`ZZ;p(i;#sI$34)z>qw^WIo*xt;
z_SB0*S)NpWXZamy1(t7If>4Z4g%5WC3)1{ts~FDwg60mv#|Cl1`6zJgM`Xrcb7!E;
zH{iq0<uqx;$Mu9sJfCnfwRiJUBs%ePvq+u1NEA3y)_nOju)G;+JmmQgXDi4U{YlN_
zF<!N;)-OV=#-q;k5^m4zFNRoeL}a4V-oH@9KN2N0Cdu#_Y9vh`rKZSnlNZ%RdUP&T
zi9uj+oaL2+v^I<OTvZq-7r9BO|7qQMlzHR;bGtV|7bni&zNl0+fEKs&e%*8)G~*C(
z`2jzJ^4m<NynGTVjq!;653NIFxQ)Qxc{^{ef+4<#$!uw+OEjf~b2x-rE!bsy8XvlZ
zNzKS4DbKXBK+H2Pf8{gZyPLYV&gYgO8=JSwKRXJ4#z9+Bx!TQpy<BB3JwJsPYq}tD
zreKpuw9#_gT(S04CZl-aKz6CI#9wg7I{JrGqg9jBbCm<7;hw@=Tb22i27E#NaG9~D
z!G#0xN8Zk^0U%<pSQ5TYMy2@?zD2}PMSunNqqfFIBI!;ixwR_+CPD-Zl6!*pzMhty
zuj~3ZcpF*#<)_>A!J(bv2SLsesMVlDyU9mP0W<R>>R27(L>!=saE=1<4yT`@F>}bT
z404%cIoI1jo?!mNw`KC6gsx%W{nYW}Eli?(GjdRGFWAzu0E%myud*!!-uZ>pR7O_z
zJ*_vbO;QXSjBH`&uc2k_Z*3PN5!_uI2kV4OILbO8?I}#oxD#N37hd6fhzu5!t1Sj&
zl#xuT#lU5-Yc^ExAfFkk$h87`lA22_ffY1Vea)kU6~s2)G*oj3K&7fQ^BdOtRb4JK
zPu>xx@_g=&bh-=O##^wxSx#8n@zUB`k?*?e@`A7FSP+;pzTd)cQiD~vE1%fKmFL+g
zE;x`G1UOK}-o-!u#rEj)_SHc@6+hYrhBM%NyTV}S*j?b|DlfDpy^@lHPzEe8r5ShQ
zPNdU?PTnBSQtK<8acqhRkUcal8rc%j+qt-ve&BnwmRR}#$tMij3V{g-1wxemFK7sN
z*M6F%vHz;g|Aih06DwE=%0S5Red~?|)m7dzFG)5G(auz|udgn(YrJCgpQI@17-mRH
zm|gHe28gh1er(KxigJMew4>SL@WQsktl6oiu1t2z2KjVZ;AzfsqTZ=9`!t8hIQKV=
zBPqQ5;X@zFR3))_YMN|I=H;A#kgTrtm0C?n*?#rox%=<Gh4$jP{Gh)rY56*TDDa5o
zgqQQ!zV}VJj5jgf{Z}J<_0;0MZQ#$~6PRpAjc#A>_ufaTG%hNp+=c-%tPdXz<D&ds
z8Ct$%8+OJ|kY<VP#h-19MOU%bzs<mDI%X_9IVOPBSD+H_{Qx0c;|vbJi7GsheB+&3
zQdu@8`ClL)-usLNq3Y}7CLM|!(=DjjUyW^3VEgM9z%J~0L8(_u7oGg+r*_rW28QLY
zJ!ep1ko-1A^O|oRTHM*wcyU3a(?6$i=_~Hv)s$OC;8b|Hr0IA;6eu0)%Fd$zAp6o~
z$y9*%x(YGizpWv`VqxpmX-#31n$8wfMJQO)WRUnZQANB)CwXORje!M}7VMHaX~xq3
z9zKeW6JphzHs#;rL{+aEOP%@Uh=cpetyNLZ+KXk>IWnxJ>X|Kca_`>!m?j*a?!qeF
z+aoG=Cc(t50Lg-x(L{4-D)W4NblWfg_3bIVzP0~7MLWjpQ$#+0n^iw3>laQk)*F@R
zS5x?C>F0XJ@UYVf+X!99(i>R;K^jp7IYu;%!c6T{ruFpN`BNa$^0Q0+)19L$hR$42
zvETa(V9Etoy%tq&B&iJj`U?iXHWAsbAuIZLQDfemwA}@lq&^@Me#MyqpC1#*$I@bY
zbGxfCB`UgFVuN_4@^ZF!UGm-RC-*H`m8xFYz;Q{g^<gbvIxo9!*xbxwx}bxQhQz;@
z+k~RwKeOI5^6=}TT`d+8&eY%~G0KM8s14n0@nv6-^3LS=5Bj1d2F8>%^q~2pW&yY1
zuS1v~Ha;_{lw0=Tq$B~>9!U-JJ&^rHGq-cty?c4W`FEW65DxryxF)*x#Hx#z-rHG<
z=hQNXA}5UkpUJbu!X@pV-Sdfkwb7=(!(v66gg^8mcEM#m6wgrJd9vt2*S<F@YdjLz
zE!+`lYgeQoCR^ftW8>2jG@ey>LjmqE6(Za{ew^!>Gy~W59LAd37zxRhl&bFX_GC$Y
z-YLxJe}9VPMj8=<ZfRg(O&2#LaY@%+#`@;xx#vneP~4(zvs<kHl@^6ni{;J&a34}y
znq_hOTRNMH$!@>LmG<NFCQSda+#Gk@@=JQm$f8k|gXO|gDh$nbs6EfgxCl#K4i}%6
zHD2&ZsGo0uCgOAF?Aq?9;=-w<zJ5w>Sj`PjA-Lrq^Gtd=c&KPjSg_aoMUq1P6WJo-
zVB|#MK~;Z?$IO|E(JKa@JC8=UKS$XZX=Cpp`)nY}H>4wc5+dR!d#vT7myC?;aq8Zm
zn<OKcbqILYtVNK;#l`20iaSxA>X*4ggK5IJT?5DCDc`;M5zd9Yvy@?9eI`&s;Wzq%
z5FLL{AO^G4x17G`*la1K#Gx)AC}In1DrY0r0ScXo5{NSI_ypYCb%<OPMzO9L&cSR>
zLuqcMiu7GNzXCRX`>Mc*C&K?O7D<o_4)z)e-D5$|muG}*y~dQ<BS7j;H}h0D6!98r
zljE^F3?iQ4{DNPvWl9hs$83p6&QH*c&Y-3j_8D@;VP|0sQnqRiW>l~e{fsVnWi5=#
zB&*}2rlBqydyIZ~b|%Sgw>!tl9y3^1Y__)Lc{mXqlR&w32p%KEuW58dN2kM+R|*Ab
z8?zJT{CC)c152nVB_ss3xq`@&YfQwqW6k0Que1wi{ag@LJsoGK3K{GATpySH_8;x<
z8y%IKQEb~C-m{+w6>Fx>i6%FbvuG9^kdsM0r`3Imba&YwMtEmexok5rn>#8S?oxh#
zF2?i=%<ffr(0infPia)*$7Lo#ZRn}Ts<8WR8mR2NrD4k=OG&Yx(0)PoD&n4Fu7AF5
zs1`CN%J|>oQEn-KKPPO3&IfW{9y_gnD{vOqWga(9bKv#eDeIIJNRrHFF{y&PWTq4m
za;CD|b~&l!uhbr;`tWajC3-SoSfk+mXVl`avhD?SnRV_Et7`P~yQZ=7;^xF=a)x~L
zV`$gmVnO3m3h7O|uRF1!ulFZKets;F0eaAOS3W>+aw6M;no_@MT|xQj`9H%XL<AV7
z%*OVH!8htW^m>w(=4}6>`K{o-)GSj&DIoCve<tAczUu+hP?;6q^VTvogTiMp<vF(}
z(BGlJ2_gO^OOQPsy_>Z%s>OS#2N9Hu-`QvLG!*V&_3J!__|<b|ZT!2WwBX!vTeSFv
zYeOK^HeCdML&$|h`?QAo9|ZJjrh>)xs2qJq+pZq2m$Iuf(-k7wEfN3etCWYw{2ZUw
zh%>m}l7_c-%_+_s$d^13Bo)1x;Gw0wHXGXQ*<^3lh3#uU+oBY0ch)j~II=I`H_(On
z^_((1o4vRneY`Y+I3k1c`3hPm>k^YR7TUk3nnV6)9)km0C@Cd^akK>(dQ%eqD$PFH
zVI=A5RBf&Z4c4CZqTAiN{J74>^&X3-`jSSYfBh8TF|*M1gRAiquJ8{yiE?;<)aDhU
z1>W=*-)Hcjlv-y=maeNG{`x-ze(NdQ3l~`td=OV7nqOG>f_&f(P*4QHe<?V#Dy)z!
zss3|L|Gk8LA&``uBgcZ@64l>&vtT&v{H}kJi2VH<O5~7Hy)bdE?f|YIOH*$u*Pyrb
z`}ibM9>(mUPfyq6wyShg9kP7uKTbgPCJiRK0w-*#g3C5oE<cAEi|Q{Ac{p#4tA&&Z
zHhZER0=#Jq@AkbPb|aKEG%ZxpEh^r((Y@~c|M3&RuPJWeY)-+p<_NJR7i03$Gs^EJ
z6rMSKr;%m&BAI72@PaVSb4lXcd0lJsYvt-eO=u|50}<T>>G17x5YhE^3sQ!?^D=9i
z>*^-&lr2S7NmSI%&+YUo^H$yAX=@{6PccKHbPb7+ueN3F`5M+ZXg_{y{9peFY(y^t
zq}G1%7aY9H9hCtWe^@>L^?r7xvtCEZ=Px@F`;FV{D_m80B+d-aatN`+xC3f2*ei44
zT?%{3ibX}XF3{O(>J|Uo&~Uwb*?n)fo14SN*49Ghm}eu5MU=x7RRsbiMfJ!VUq3%a
zAt3|9lz{)c^4Aks9TEt|7I0)vfGX1~=+_ElxSM}#Dr+%r&bXq?6q$B!d}CH%yHJ_u
zGQl4~O+!;=HJR-S47U1UVNp~!jJ!8Hl+m#KQqO!A>O*hD#?8)aV`tacIpigbisMP-
zhwf-(V9E0q1=FT5HI*<bDoUL-AfO>LiUS=TT|-;DByRA55PsvR+p+qQJ+7vRM^&pO
z$mKf1-LX-2uv-=5tm-G<r_BH7wgB<$Dfly89=tYNa@H5toNR>}7WRzg&K_z>T0O}*
zeA%+PX{n<Yf&|PCN+rUXdb(c#h3S=zGX+pDlz>*_g1u}X74{uW(WlyV7J9T3nKOU)
zKu#V4$XFU_(FqdDe%IvPOy$>KWkGNU9o)LEjkU?@zn%~JXFpi0Zjv7hZp;4j{Pgxj
zkvGCqIDtT>hIOJ)<mV69pJUy)WMt#dyVEoCM6L`SU$R;vXS{*QWgk8;GS-nGfKq9M
zyKWCym>8K$W(Vn~07e)?Pic5@95@eklC-iBtH(}MK{0@*1q<6QUqm=q1-kPs-lO)t
zG_4^QO=S0d(Py3$mcm>ZoY;LF34+Xpl#ib8rW~Mpbt2JPm81FB1@pKgUY#;G=*2E2
zC29tCXjTokB_rkDJhLmZ`geHh)?Q??&6h<#2MYEJ)r9eO2b*0poN@pE813~8l0h6Z
z_1z+tTek=!8rJ{a<KS4cXI$gDRNmeB%ayG1UDX#PjkXaaK&nLP<s~OVlJ|kin7SS)
z3Fu-Q1)phG8!X#3Wgpv-kr9cd$-dR3gNkk@XXV!w`2U77o(O43>$XLdHO{1bZRN%;
zcX;PeXdiJqv*}=yD`L&q$XXC{sR-HIh)}42FQscqJhZuD5EKmU);E?rWBs|+P8Wq*
zktfOD_f(!A>*GEJuU$A3_~M7RgAx8B=g(ZV`BN?NaC~RAjx{A}kbY*-_G-V26{C#j
z*MRLlTZ_|@3`xfFF{-`Ya!d3GlOMyo)*z<x=hhx)f%Kkihju91j~_qk#C9b=v1}+u
zYn#7a%a^oeEa}e?nY7<Jo+mk2L6uQ8k>gswt3iM2PX0J>q$9a4sgL<^RS}3$v^Ue4
zHo!ndl@~cReM~*9{C}+*1Y`gPT#SGz`vmE7)-z7IHCY7f`cJg0lFl8GPwMrO!URdF
zI|D;92m%`9mNGID*OaMujZ$@)t>vwaaa$&9<Kjb5dvADHQz;wTEZ!odu&}V{Xw-U!
zx0?~&_Xm`QDRZ&e&pp$P(0h>wM_pL-_e{CE^z>?Xr(I=zyfFx&S)GcTas)9o`Rp`^
zUjj@ByV}nwPXr7UbNqVZ9rp2M>+0(4b_=Yg<Xi?dXA)q4{-&79Dkp){6<e~hHoHED
z#U&&auX)FpyYlxb0^U_pqO)JDMQt)MeOI|WZ(35_IzR=!>qZmHUT8}(0+T|EIIyzQ
zhL#e9z%^5uSj6QnC5nQKUQ(k?Q6v_}6gE|rQ@K<F3f-2Mc76D(qX6)#o|l`hpJdan
z2iz%$H>^aZrKA{yu803H{J+MzgL3L}p|NA^`;Q?n1VQkC7g3|HqDa-U{{GbG=+SYB
z{>09O2$W}_E?in!8M8Z{vG~y#ftI^m$w~K!bM-c=BKNK3O##+G#7r;j^L|XEe0kN{
zFD77$TX8}H#1gnS+yVb!4yOx{Zcf)T+{t<N5(wwIlETpyURhrXr&#Z;loZq~I4_AA
zL`9A7u}V>&K7G`K%@Q(YizQ``k-?ylTh;vvOFWD*Es1ztzbs>FYKmhV+#r)Q9rIOo
zCRaQJdV8IRi+4J@f-E{R{#HW->!^9Joab>6Z+73*584|DTZwfb-P2|M&XB{F%a-KV
z^0J_|HB0!NO)kJEtDBn5QXyK6#^#qdR3<|u)eU2Q2rgCm{r^XO|3z0UpI;5Qwc67%
z1`_gYwr;C-+I6pJo~#o1CW;po<2D>?Nbo85)Ts*9C;jVJN!|@l8s#*OFqXF8*goU>
z)kBV=wC^atOA8d&)oT3yx(}vMP*N()rM%L?=}4o&pPx%7i&4X;6SBNi+w;V{qMN#<
zNlbciV#(M|C2Oo`BQLFA&<@j|Cq^|m%E6WaczDU(^>OI~b*ICTg)#51+Rz)$H8suL
zDXU8_)W+JU^Z43)aNUk9Z#~3Tou-BEE}6<Jc?6ZMT$c1y4a^HTL{ZStG|a-|VM(h|
zb^gN6*DcR#XB0pWCj5LiGZR=^hSxqhNr%wtu3fZM*j&E<e;sLxtXOEMib1}{!W+uz
z6`#!#oJX2jUwqutoY|bG4EY#*pBQUaCT?3?QX&U0Fa9iGh6rf)7MS|rKFrdcz>U1M
zV^%O0(@pcd&mls`xl^<Gcu%ISN<`%AXG~FHGiItu6SDLs%7WI|SoDfn<P6IqpM09`
zkFlwF{Wcq#6;%1#{4Ni5YphNMkCK5XZE})X!*nF!-XM=(JTe>MTO*5OLAc93FBnPN
za!U%VgdPf68<eu&95>0gY>1-9I5&!>6xjIbG0!?JG-v<IH2#Z{95sO`Y5(`Pv^ml6
zXB;ML0Oc0Zn)Z1+d!VP1YZ;IIF@%wNET;he8{;l_zf3;2nPKYPSe9T4pjIXghzhA%
zvJE~qPuy7~m+cGrsSEI0iNj^=2F_p~?&qgT&YyNW`I_vSmYbWO$KE+@WL?1CZ1e19
zvHuOd&v$@Ap%tk&o(t~+Ax(HMNHW2YyfaCB0sBIVSTaF{*bgZoCNQRmW(3$EA+jKn
z(<hRQgL#)X9FL;(9e2gEf62`<;iRR#kn`UAgnoAHu59x=Sw`}d-2NLL)^+J@2swWP
ziX&ytEoei)C3XFUn38Q}*!5hy*{)<8=XH;uVPrWGW?o^W%&gB?`_l>KkpZQR0PePD
ze%+SbfqSJ5=j#DD;(V<y=AXB_rAicXaauw9YSzHtyo=XF;K4#xBm%}>M7tjVb(Mh`
zdPP8cevpZCC;_xT6d50f>?m%*sU!DdU`#i^>da&A(O<Zr^}ue$gDc7Ah<GlpXCh;2
z%4DJ~S2A~nM6$<7c{OmSvASTD&4&EAsh}pL@9Zmf-zOJqV+X>3<=qdyzMV4+$9$?%
z_dot{IjEA-(u15dN{Cld%<`QKkT7;2a+FdgEL+OyI1a7>fahH!gAKsP;IS8!GNS|V
zgh+ek-*dA_s$MkRJLxFvRaricp1g;G&U6SvJm^!otJZdQx!j`FxgZQrO1NMd>9~zh
zF%5G*3v)7BiXO3zQj{kC8_KwM6QXq*R><Iw4XKg@&%yngzom@@@v*q)?MC!9C562j
zRED1+5URZBh2#GK?tEd5?-w_sX0W+Pt7S2AP0s_orm4Dee;@Xa@xwv^+IK0@IzWI_
zwrqAg(jw-u)A?%aIosRepvXT>0_}wNBGeG4f6m~l6uePbo7`52%zkEzz7k9NVCTwb
zvFGxpG^>cq#ei*G*bMp=EWWTSILPyIe}R?G)*AfLjQX8@CbvlFZl*|~JP9*;z{=8_
z^ndf>Oe~oB1O&U&=ggX<-3t*sXBGKQ&m@f8%|R@J(Tiv6U5y`QL3kRcx#E7#W!=*$
zlKonx23P^@`A9{@#VdGYdc+}X1fVs9AnlyCc=Crw!AA3C8wFCnXy4}(>!}rbN~n-T
z!7M|8PLH5JR~j*Nb5J^Ip@O&T@9(OiPs3@HThidraL{yZpP%(=HGF2zu(34S@$dl>
z2oWd7-=2})#+IqHxjs{!zs}aqRYw*>a^;i~Bt@_)-rGF)4!hiN0xO=&ulEHR!J;`p
ziXQV!CS)cbI+>>)S8isx?<Ra?W^G-zGQMzo^8i&86I;hjAvc#@(-ZJ=<vpC_Cp*SE
zz7qF;=}snd?)tDtFf})5yr+A9cPD6ct8*F#!xByA&i2TL?0FuJ#no;MGr1n0T*+K=
z*Yh*)j2HJqtuY<~559gq)g#$zQJ{Q?nyMHH<|fynCrOV6mo?3>hI1#naJZRn_KgAj
z)40m~ia0;RR1SHM#hW7cyP|&QfCp2TCTxhel;%wDcRdRkE-xv}_AA~l#xFg0&E%Iw
z@hJo|{do6{q7A|7)X`;yg@y7`wCgFP2(a&^wx%CzigiokOEL|O)&0m;58OD&2q4}}
zuvE-tF3R+RUhOkklvD5EAh_X}hmz>r78=I37^l;AVHmB}_z(phF}b`XDd<z!8Wt>t
zC3RM=qw`Q%$6#v8r_qx40Tb#WL<yuS9uudwE+!=IcvA&6_NIc8q7SPlW~Hnt_N-N#
zx?=HN-O}7wPkuP19+;0`{0$AsEX;S=LN>&H%FB;D6&DxF7{Y(e%FHZ0KA+a{^Ajc$
zO8O}g82QEf4O203v8b+WoVvX-Ev<9w)cN_&j;vs<wNO$bfI!&tzp<vY62;KI$4W>f
zk`UL`MIbYsE&F_~iYNbbC243$2CssrfF@XjL@qA;vlN5z2vLSh9C^|3=UTeR2_~ir
z24Uf^s~BqbwqNQ0dYIR*@CXa$1nUZuP>|$3(V6IhaOiwpnV6Vbl~^qH@SUjJ5Tjzz
zw4U>fQ$>gn(r>F%co%w(#;ZuB%M(?IuoY9XXJ$#L94q(%z0r>AA1$_JkG;#8L*KJy
z`r1BV!}hg#D%fU+h-tc^+pl^BJ_S<?7!;UDDT6>X#HW5N62YjIo4)QwMqL$hvF1#V
z2nNny_}>TKOl>^aQ7}J<gm}}1PY?8vZ9C3z9=V_U2WasNKaI$^hgX-_PM#?eD2Z0j
z*J_e4Hfr?`=h8GX5SrRrE@|odjT?D2hH*S}*HK)5s$pG&&vVdUtzB%cQ;r$(R%(8b
zCbAWpgZX1z0f66m^9gQV2+Rz<*Tby2W+w^KV*8hWl3`o<;4H<rDN#|jv#1$92X6hs
zlzJOMzbjJ(s-%e;5Y5vLz8`Y^(+s!Lht|p5e%tBqdRgj_l=wmXPP8Y8M=&*EyuC^@
zBW?!XfC8?c|HD@@Y{(QMp|LI9GP$H(MP;{TbkHF8@q`%xh3t|4R8Lp8Bj#K-&vBEz
z-RwfD-HolSzfB+zPMpJm$nT#uT_?sOlNui1+LUtcksf@ujt-o{DpknH4i}SS^TnN6
zx-RV;P>z=pJ8vwBxM!Kn_~J9MU)UEXjo*HDXquVW8gWVEx&LSUdJw1~>ex&qug#2c
zI=?-*|94twwXxiX<w<OPYF*~KlL^bL`StO?guD?YNcOZVY9fojYygCS1<`9FRbLv#
zVnanaR}0KF@xiI@f<y*R8G#HNx}|uX6Wt;$1bt->niUb%;zJ?n|L(;fK3d>2Z<oq`
z4Ii3qrE=n8<BmG{6~gdlDSqVbQCV$_WR>&&#El?|HJL4=*l>sAxR>9-HL`r1JL8}K
z0u3l09OkAEZN=SZ5#vi*T8!X>M8BMp1+I5+61G!c>VpG`f)cjG78tJ8L{M6PQFyMg
z)^-v8pGRux1|wH@@QlaVk<*)Gh|nd9=(?wN>wS|JwIBPyYRJeim0}ZTfyiSdn8GPy
z?5fB52d1iK^gnV$P@LG-wokQ&WngsAgF=|>4W?aAE+w3VY~`2617@%e2mvD=#W@`K
zDA-C=7{Kr@9tNcmra$dr<W}%g&pc$^H=4@Ak};NQ?aCHq2IL_<@_&~BG05)KRay`Z
z(J6P^J+i*}4{5WaI-gqC7RwWhUSR}|*p%*Dx@S)gz$8J)-70Wi_gv^+o_$g|6ojc`
z7Ah?q#>ulo``=dsE2z$aqgkn*D{b7YTmU>uMOA6cP!<;0J^1y(1Vv%v;tfet$c7XY
z{_DvBnC@6XNdL~9(_JQh?uOb{$(_G}lOTaju?ZIA1T%;S`QZ}ZH!SH{Sz<xbVHXA*
zA``YuQJH43{wqOgV*R}`OqpGCDdBtfdKb#yGVWZ0K*1qn6Ju3axWKUs{pR6|LH&j=
z?pxCTvUn&xTtNE~2Q35X@{L7<)2g%>lzvLx5aI|Ev;an-rsRVujljdA>Knjs?C#Ef
zxh@e^b9d)CS#60Y_q(3vnPg*Sjeq$&Bsw!)AiKII<ASGe(^Fr65}AsEiaNjM;_RGG
z61?uCsi}G0kIFg)#Ln^nnj)CD2I!}1>vg!FATRz22GWtn7#L6(>STrl1lNY&y@NY4
z&?NSACjm$R16KBRhXpVKv>kK-?Aj=cC*}nBG6Az*<L7p_BN<?t(sj9yI1CJ#yu6gS
zMy$;?ppQ}jlpkgQZfi!PComs;rc$>d-2mQ91ZXIGI3K`7M!Us%FO&z0qYEST4QU=3
z897dNsD2(I_lfE6?=P(ih6821%Jy+^a2(%muA$F$?ZgR|wzsGA*sr)sOamf)XN?Z)
z?Kst|?Fktfim|b=$_~rz3x$E5SRf(NAkhtEaZ@I;ChXT8KMzYeXnO-Ugb+4@;|qM?
z_>=Z1NnGBwKqaWnMS$P=H+bD;QuwuFZW@?g{$BL?`YSeJ?OXq?URb?$SH_#OO=I-p
z($debLvY=)zkQPiGNnV`@pSG+vmp#xjYf~TmJRQ>rb<9Fl3PWFz)lS3@Z@B?goK2i
z;UW=Uu5jx~^J)@Lp76=ZW={?<p;ER&yQWzFD_vQ!ouTb}3UzIWqc-TpbFT>OEk(il
z@TDCyT^Tlrs`1kfwZhT-S0mp`Ed@nqqkejiA991tU|0NCSKPpPwUC8Wy7Ao$H@IJS
zR$IL38*S1GXP7#6fq*T;Onme!MjZVvlry}}>&kL|Y^yKi0I2%D=EhO+@!E?32VDXX
z+hu+w5ga}PjTLqWm%!>v-d1lx!j$;pKYz{zj?-IHae`oLWp8f*<y5vH@0-IS)(^_c
zvDh*u8%e3DazJ-&L0}+EzYe7@E-voW$hWtQR8&ET=)~q_R!1NUUb`hRV8kE?5Z{#-
zOxAT`h(4Xeqm%H&P_9)P_hchtkd_WY?U3M*Q!id2#^`9yje0MPag;&i*dM_3>ZIEr
z<{Z=UqAzl3X=xhj>iL5rmScn{n4hwBgCOB`18cTVe=ii+7+6@ecEtBuwe|HC;StgJ
zwuwh12_@Ez{qLo>W1ROF({pwE{!%(h$;gzilJGhd=H^oN6Nj)lR4xJyVM)0+M>Rb&
z6Ip_cTwF=7#AX0mDJOl|7T}_VqW<YnU}!*-Q|ZJA5?ym*409yh``6i%O7s;*K~5>}
zo^@TC9ZTFOB1}mKvTp+l4u*sNg}7Mm0&?&+TPa|$VHcC(k|?LN1WRD=?HI=_bB(<^
zYi3JWY7h@_ab@W&-zdKhIbb&K2AmAV%Q?~p0VNJ+`QeJ+-MZDs@agX{r3WK#YZ{Zx
z!sowRDjYs1GwW-RC5ap2lg}j*>AK}8J1LDlRUDN@aEx`PA08emaav7ioVj&nIJEzq
zi4d7S;l#21NCAh#*;Whysijvb#{RnVix$(ooOOrk<U@+AU4@@AbogpxF#YNS3r7y?
z^iaqJ+y1>g&ih?Q+@60_j*e_%u~C6a#;<FUFieV|mN&f)l>A1<{xP&@6F4*m+ciVw
zkDkDk#&<h&i$L0H!NknW4PX?~>t?XxO&)+AtUl)I>$7jXt?A4^tt0wM%=)l!4S+u7
z9G)1jw8c`F%J|dxFk#K>s+JA)yD8T5e$H(h*S&z|(#h{g6=P=ZO&32sGb4IkYW&Rw
z{;iZKg%!<D*lOrugFrz%RH}v_Z1sAqQy6#ToH7m-X(z=3h0KUgMNzH5$m4mvf)`O#
z<~VOO(UktY&WQZD>hnYBlBGG)<5H6dOiVfcJtPu@Gby+eMKwS6HZ%wrEc>0>)Awdx
zyIkZ<sp0{`L-}rrW?kegCPT+37X;eq)Ba(Pj0}-eEy>I4a|LA0-wtZE#6AFmonCE`
z9y7{TIeyK=(aea(W*0gCo@6y^N=sFrfo3%fN-$>cAbHR>z*{<7$+!jj^P=)aKt=p6
zyQO!>LK`3tgGRSlKejo6S{a%75H?sU1bcm)T>(It*Uf9S*}j}~#O9dIV+;NP*t)ru
zZ39$72(_4w_KsftuB@R)Q2^jwXKaRb{)+}6pibofue~dMgz|m&W{iw|kYuuqB}3Ni
zA;wN2LQ1KjNS2gQ_6){WLr8YnciFOK9TF;gWDiC5eT~w&r}}-*_nbH9A2{=7Ud(gP
zb3ga;+}G#2KA-D?+`neK^8SSG=$ls${Wdl?L*t!nJYDwZb3|sX<HkuuD%X)aIkJA!
zpqDvPQ=yL^^F;>>Gqz7mqySG`PC5_JA)l*wvM71No?-oNl}momm8hMt5;LC2m>?z|
z!-j^27X30LweRK7KsK8VSCG$H<Rfof7nvAdnP4<H+1{zIQi{m<{DyuOi2cl$n;i2i
zV625N&U&$QJi-=H;m`_#RY6$VBRJ7GGazXFWy1I|w_BY=(@6n{L_eY~SEsRwToPgx
zJnG*~L#J1M>m3V_g%#yfCt)O2{ok>6G?W~o;+#uUHAN%xDx;#J`X7A%&Ds4=XlPHs
z{hc)8_;Fo!OV>yjz9Sh7{#*@}--9#6<9zohk2`%bHjD+uct15o2i${Y{3L6Xl)Sef
z(6677&}ZQdAew^~AVpqzR++5Rk0K~h@uoi${2I3{>q9_A#KdG17<>#4Buj>$HcYQE
z6Zbu4lXFTSUm04jD}^YHz<<b3zA6Egu+Os<vU^Ju(lF<(0JZjv#CoUv1R~wEsnK8d
z%rW{d#YIuGWu9Mq)Z<>O_2(k+5;O$%dKGX0m2YUdPx?{Ai@$Hq7I<9VMgPmIL^=R|
zfTky(k~)p%$ofUOEM7Uvi(N%=Dv7N^QxGDV+p{RiHzXU3Pvzew-5>IlABjX{1Nxp|
zX}3R;`eAKGMg|u*C9Qqh5f5IOrlw(Tt3e*6xsd(Lz_9UJw}LO#gBI4lObLn)PKklc
z`DbV~L?AEyBT*EWpRFU+`_<d6O(Yhr(RIO-6m7PeIo1=@pc}Q(M*bC)4-s3br!jk1
z;Bffh`@<voTp2x(;AL<GT%2jf%gz2eT>Yh~s{nC&{X2e2o+@`V)cWqU)?#=`I*Onx
zn{ZUk9nZlsIv_IDWUO1R4XGU(N)A4Q&|Up<(~Ub4Hza|tbe_0db;XXQBp@8FhOpi;
ztQx1iB#$tlv6byUyUWQ-5Yc~#+8fFMN4oiV^Ic^aot<%n>k2sof+l^L`QWDqk_SU+
zI-jZI7)n+BvX1LN#l-RG*I*fMEh0Mtu<BbYwX3IasR7hXec)s$N<VW2DV>DzqEMrT
znMl=FI@hfS^@&a@rl0hS5P^E|M{qJa6b{aU^<&x_X2j^wa}kCfi2LzdxlE?=z#zpK
z3;|Yr*>+QanZ+!4K(qS1!J0IM)vm{Md*puETh2DSZvi;T8I)u;wz`<i(HG2gYn_s&
z_>!x)pm6_-W}*tThI6EaaTh4U_37BfO>5_?sT(2u^BdLKZJ8XOXH$Hy2YMwddz2wr
zJ8on48$5AKJd*ew=xYJ&F~WCBriXJCM*|uIOI1HGya{dSW&~>E$ENL9Iz0Bb=3{Dl
zTuN{y);&65$gB<Rp_+$p802%u4?%>;BA#At^i6MKNOC#DS#H%p4^}vo`?z=9o%xEh
zp!_lrQS%=^@hxzz9KksJdeon$Tn?T{YsIK?62V9K`x^TFMFCZ*a~M^mbZL62ff$N3
zGnDXoeWSu>aj8K@;Pv<FA5L7bIKw4Q{-TCR(qzCq70yv$T1_%W%STzN)kSi-$xCwE
zPUk={m22yHY-|n;(K#ePsIuU|3j^M^n1-VGN5m%g8&xzp*CgBKK6ozG5LMC{>?n+`
z?`Eu~IhChTL3}KEUbbb-(%5n?H=8#j`OB$hoefBx`^dGoMXDv`$D&-%`n3G$32tS3
zuwM3pPQPJIy^Eg6dK@TRD9g^dt$rR~x%R-<@_rq_@}}k!x;gZrS?4dz53^}RsQlyY
zE!IpG)|q>ba39QU;<P`~tAKN8y*MFF3$L~<D3VKQ@>zUHYT23VrL1y{dT`LXzth!=
zIBxmXXS~{wJ9zh#)1uUVh#ZM?lus;os3zFqlh29S(1$Mr>G-v446Y91iz{vfi#z+2
zzeuUC2+W2^L>S*|^16DH^x$7H`L(;Sma+EQ!5g+@=JdZq#YuQPoCb~IWB>QOkk{s_
zHdvdh*L~yb|HJn9yEE@Wp?m^K6rtDtVS|CDq*`VSf|2)sH7fGM=S8;h>W4ar1`(S_
znK3X`%@K`0W8PIQ@}FBC7$$mHS*4x36m;t*im>tn?5{aMlD`z%QP@a`XgWTew7WWS
z!ENk5HQUjnt|OPT_(33U4hpl!<%w^y*$e$BqEn0&2LnwCp^$+GGjd(zZDv-v+U`ZB
z<9Q7>e6$T84*^)=vdy7Z4v=)vziE#kYTD{b1zn>>d7n88M6suE8$kAN>M!AQfh~d@
zM4z;R$K|G3OAq8oWCx|Qam&fahodGR!BO2z1WLkwjO6f2)uS%XQaDYQrpbs4+%M;a
z=i`ViN01qz95_@*Ufq6ra&#SpqEkmO+EZEQWAo{o+8>LO5hphRpZ6lSGzi`DO-s-a
zdb#>s@&UTTNl`EQ^nhHw;3|0>^WYevp<1_M8~*O=UlE)@ajwJ0#<ugg{Q;8(CMwTk
z$iF#r7Dyv6a`@&LBxXx-EehWqtqfaJcX<`cioSU7UP;aK=g+emIKd9q09O3Vk>R#j
ziZV(jq>Qe44nT$l#Cm?6aMQu1ir_?@aL%-W0`mgPkm_oM^CuMIW$r}H{4BA0)9>ZI
zj_3ATI_)_fsbgACcCwSJ6G{6eW<dZ02L{<y$8-FuA;#vGE}*EgfBJbtSrwGhJsR^%
zcZBAXMNABi4RAk-ai%iDHloKJ!lf+KNMQ(4wk5EIPU;Co`SC%}1n6a`fS_PLEPF3m
zyR2v5^VmjU9Tdb?uic8*fNo#lH`aP!nW!CT`JI_xJC_}aA7+em>;d(DUv_x45;-&)
z5BEi;s=dFvoP>Mt4BvJIN?#LQ-H<Z1y2Jg&0mEOwaKE#=GM=x}fiJ1`{T6?_57?P~
zj$VNFU_1j*rRg|11w{FFF{q72!}Brcf_atiZtX<gBdj&-CA3v|IvL(SAkT6=H3IcL
ze<M1FWSXqE9e=(ZosYioF8J4vcM`U=veSjx`$V^Qs&``&>@^PA`{Swty}t&#zwfu<
z_Q#*+qi9r>{53T-W$&%9I#9sgG|k@F3qFrkK@7%7SwmBkaK209$c#5E9GBhYeAU8{
z3VJ#X64RIU-04{#yZAL-Y`eVENT$BWF)@qG5m1sTiv@vgpiG0<Yr$#Bbw0&>{L>mI
z{poUxioy?p?y?W$w<+qtS2_--zfb)Ea`w(eofl7w8L8d<>?JHFX0WUu9Ua}XHuXmA
z=N#*j;TQ-6$arqqZO`Dys{5<tWXB_&x`u|?c_ujvUtG4-VUI<g(tYT)YE$L84iD5Z
zcwMTPJbRO`ac;d(Yp&?qw-cws1O29Mo$AhM-B?~o$n!HmDts}<7JHqwKa}E7BMl(%
zs}MOTvnEW($jB&yg)_gPfb9m}0l|&fh7S_zD1T3QGQY;pcd_yVnBdKkV+Pnsl3#Ru
zydJiXiAP2Mte168xZT^wYzy$`U%)e~T7VKHEP|7)SOGZo;o+gZycMvS7e0YaksqGo
zy%8#9nIL6z?$xqdJO1@YU{>ja@a<6!4!v&WO=9VZ2jT1WdtaWM<A*qD;YNa3Pko>&
z#!N;`;Se^6^m^+JZe_g`l9278fhfHrzll#;Lf%tff$v9JH_zNC9`SYD+4Hde7*rS7
z_W6SvIKvg8ewmp91CGYV%xvuJ1-Q7QYFPX+HHYI^MS2jC^If}%*zhLvp(15>Y-oJ!
zih&`sJfG!+9C*uqL%+Cw2T6u|{k3k+o^xrzrb<H<)MA%QNE->t9#Vc&@+=);GELm;
zz>3NM_8zM#QQ=E{N=izZvF4ep%e-U4xd0?Cg!m5mlw{*6HoMk9J+0yfLgDHMFE#9F
zX*YyB#Y#B+^G-Cn8GQ@&UDGwE^~ZR>yJ$jW-h!lT^j-dqZS=3Zvw3;@+C;&WR7AdN
zT^#a52h!j|6XqK)yqX2Nw@lql(L|8|MLepSs)9oxDzvCQK|D2SwN5<eZoBGJEeySn
zrQMz2EbYEe8d;{6qq?dqr&3Lh$UfL}x|6$SGy$?TJdXGMvEDW>;rpHL=T23l8Io!U
z;olo<(Q8}-XeM)DiZaRY^NtWXN3t0klGbZk?HNU1c)LQ|3PE;?ZXN{tvi97UxPi#h
zfLTbiIBVOhQ^`W(JC1o$;()_|*=52_<x$#nip$A?WzX=1G8KVF0^Jn_IBH@IGJV3$
z`)jj=EpkA|$1r=$AW37&j28K$?7+}MTO1A?)caLeofp)OgHkW|o*94TF8Z4LK~@;$
ztjq}cKJy^$Wb%+|44-<3%V^b!&e~ruIk@9i%nB~$g}@}YAx%a1J{4Ypr36fXbxF|G
z?hz#s?Y_2#b;=cI@bjh(NAPNhGWK>>P%}XEpGQza7(4E%yZ}I)<|nCfs($yg$s;E8
zSTx=BrzDm)o4yxC0&R+^ySXSPOL8Q|_<UdO^r3~~p`POghkAo@T=g`GjMuulIYVF0
z8+ecbO&5m3Rx5-Ee+0P%+kp#e_n=5Hy{nW%A|iule{U;0F3KO?a~-|5It~71fEFWF
zZ<Y}t#+&5BT(bo-Mw!U{8p_jY;g_6tuHTL+`x*RxE#~Uo(iX*_Q@xkUp?hdj3*7Z<
zn7l=6-$gHSevLi^DRX`d*PTVp?T#6cT~VGhdinbG>zIc#ng|N0!_Kk3Zlz9J2Mtf{
zp7xuz?*8GyC=%2^&pS9lAJd&E_H1MQYx^mcBL?W%i;@Eg6%lInzRtpEH2KiI23s2?
z+G)GbGiYBui%08zO%<3FDW{2g>EhA}N_SY%^ew$#IfHXKh*v&%t1^<(8F@DDh(Pei
z<W~oZ%IFr^Add^`>i!0LJo;x;#<Zn*l%YnlBfz~PK$&dP<cX>r`DimK6y_Labm3Ef
zf379150j3aydw>|Rmx?>67JY3lwlsrS)#wcvUy-Kv9jY<DM~fd5{AIe-r6}WX|`;$
znf~fsTX*4hxvXE%*wglD_P18Xw?}Hk2LGB(iU=|(2r*VEXW6nd)Ghj1=PjG<!66N$
z^<1CHMIoc5vQgT+ejFw2(o%cQQ{Rp<;#NLKo=oWt;!J(ge_LdLHwqZPvsUBzFUS;k
zuL&N(I4-~DfS=LAebbQAqsZyO{L0J{)Z}}(_p5m!r}GjSls`Hg66DW>+S03X^PayE
zo8iTaUu%U2FFE?_%~-mPYk`x^?cvA17pTczI7-#VEMSvCmlOkAP*~v(OLa3aOh55@
z&U5zyeTG*T561abY6-d^>3zWoyQT%lZDbsa;aO{cp;IzOup{i2svI#su!Zl`w4r@+
zD-z#yuQMzRf64+v#Q6n|`W7ZNN<N)LI7yR*F&f@YkJBa_p!ewg$~;-KV&#o8p3M-Z
zC3zNF6XQ8cE!qRfST&Gc`Y~%^6VFu6c;FU;j+d~IY@5XA<ST~mG&Il`1QL;X-F0V9
zliB<19b+XsaossG+J8>P9XZ$<c4E-cG=>owBqIW^=7KVaV$Ee`_-x}$@*0eRd<(k3
z7Y5w#e005Il7`&J&$~M*nl{^VoGcG8kUn>6F|mLT$U|avP=xJ|cWBaNh#Bk*v)ACI
zQ$g)=;;Nn;SaUu+h7Kf$N*B9YnR}1;^I$gzV7?c?M~hw`t#z@=bS~CPjgmcT^FWYx
zbC>rrSEzQAG&dD9%a|RP&UQFoos61MZk|TneO@(^4`R~J>nIlzMbkHS``xvNnIWU$
zOT|F3nFaiUHNdC#Jaw{+GRBVv%KaE~-GG1UtrYZR$*^%1_6F^gJfCRwuK$yW$<{?`
zoIscMbgOhk+HXNbk0&f*YraFe+th;aT*UmMr)SOkjqt(^UF7h!!rf21g0f_T(inC*
z4WpBf%#E&~t#AG*B!y{#>6GH$x<u)BvYBhO{t1Pku<Vd&&-u}+;&VHDv#P)1$tlU+
ztApxaQ4zrYcNK^BaBqdKWe72<KzxH;7&BZ;3jGiUdhI@BQ`=|I5WGvt%XpBVLX|*9
zh$n3|`!j&QqJQx;GX=;8tBrs8M9A-F#}Una5gh*<)2U%K#aHV3=}KN=Ny90BjA`Ju
zBXjZRTw#5EeQ*f`XF}wZC0TW%$N`C=kJyww-4d(*vt!9d{6M%rR4(lqd$Y$lm}F(q
ztdpP}z$1R*7!(SHkH9hz#%MsYgeE@SSbx%KnEJ5Q8SvTzg0qLkHIz|RG(*V&0QV#Z
zyxQXX1BiOm;3xrmW8+ABE+{2uB00ZassjRlZ!*}ot$jFj)V0=S$(>BuaVKU`|GgQb
zCPi-;IRT&Nagfp$PQw*aAtMGaCs1(4GSBET)6unwJ`J~w2v7o>+F$8q4AJa2>Q_sp
zfr8g2kby}UU%3*0E>q6=bDiMsOc;5fxuFpZ=Nt?c0&%K_L{@845!cT~nP@Z$rM8ny
zkuJmgx3?Z}QWwWJttadf+2+?h-hK&?^x1bMlR0J?G(E}VNC%Y^$;Yy(KeVAV!=-B&
zNhzIvLd5#DFB<}5r1s=9Z~^(fI}Qhg-H%GEb-*V(YfudgBVmx-&BfA?{QNkC(A4xD
zvm|ziI#fNy-<r}#n75@EcLY`M3I`a_6<*kv`v)9iboeLVWYedmrNJ1QL9Mepz=Go%
z&nhRQO*w?nj6%)`1ElKaWeld(rk8Js;zO7?m^bYwCpG>Gn^1rtmU;hcYcI%$rn$bp
zE(3x=v`kFwcI1B0G2@>Vq!crqEDbew`-B2PzQ89_nMGQ<_gEtOh<|%9i$bulr~@!|
zsx$6%0M=fBys#~wV3coPHN;M#h&X|edp#8nQyrf{JHdp$n7{3EP?M{x5>WLVlvpwx
zD#nj^KX$976lO0L8^f~Vk9TaU{&4P{!ardC@6;QrlUZp{i<;lDy+7WZ-#?8}frDVo
zChJxb7+w^-nghDVH3z4qrA?okn+vzI7y+5Zk&qkV@K(KuWJ9jr?2pvK43$XG$buo>
z^e?2o2lE3tA7`!+jk3BjSH^1#LbSUaO-5K?br3%z!>b79$@AF5wLHgFx@E$C=^Lr4
z`LPSH<qLxryL*GB&5;_THHRTKgx?+b=WKCOPp$D|O_43n%kMmw#Nwx}kjbp0%C;d1
z7WG;Q)${p(uKe3>Sqsw<Hu5C3pR5l4^VomBl<7fll^xY=6#wTP{4r$QomLM&=kxE&
zL1WSz^2`3sG1F}RKkN4&Cto=`_^<!J$-frR|1n&Czo;8W=9`}z^6686kDivHW|2D9
G|Gxm3%Noc4

literal 0
HcmV?d00001

diff --git a/docs_new/examples/images/trader-joes-receipt.jpg b/docs_new/examples/images/trader-joes-receipt.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..6742d39462449d7264cbc2724668505f6ceef7f9
GIT binary patch
literal 430638
zcmeFZcUTlj*Ed`}ff)vL1{@I-p$AX~6eSBtXa<m=!Vpvps}cnaL={k^NsF>52$&FI
zM6x-py1I%HOn_ictAc=;EFj2tdca-p`?{a!`rh~7_dZ-beY(2p)H%O8bt-f>Lyw1E
zA)d=D=UE8DFysV(Xz2Cir4G@-fe1M}>moKnNDgr^8e##L1qTm!Dm51lMU0LxI8(=X
zj43@A0zYAduK->+l7|Mo0hApD=M!)k0?&tYB^+aD|HuhWWk}`RhxAFChxv3mLKoq1
z3=LheDgYsIaCE>b+?-NZn)k=q(Rr>IH=JgmKg~!VYS1?{wlFZT&^N^mOf3v%SeO_g
zU6h+RtY<wav~Pre3cSYVVI3$vwMY29z@tnSV!%n?(8!RUIjr*=$ix`odw}NwPvOG>
z9Ti4+Mc@@j`0>DF+lS>~l&x7DsoolpPRknBM;CbdNV*~Lj1k@fc;-kR5%4l2ybJKG
ztbcTt9pQPD9wT`?AWvo%O{%|<u_@aA6QA`Hk0c{D8tCg|l7HHQCK<*b$%Aeqd>a>`
zafScXiydG|ZISvx(NASZcu)t?f8rTM!}czs>>5c|hCDJq<zfHC%l^c3e&Xdvc&c8b
zqG9}fL8dYA0!Rx1e)&iqV5OvDix45=qB6MhhvR1tF|sf;vd}m9k7;`PW)}M3QU88+
zgSMyS>>QDIeB=x#Bm>8f^pn81f)ZWhhtJe~O7DO3XWZAvZzb;MvpOKc2REH&zIrtt
zINe~HsTm+BEkXGqjq*1{gFl4k!l8hM@%(+nNH=Ltgk;foXm_d>>Oj~?OTq;<!oP=P
z8uU)8pgxGEkMvYZbD5v`u_L@x2imY4Q^-#n;ng*u01#XHH}rE%p(=+Ohn^uj#Fmj^
z$uQY07E4x^&5`Gil9!W{S5X|p<xd>1t~PPJsw%EEO&iyks;R0foS`#S-@w?|SY6xP
z%FNJmnvt;~B?Oa|m6ey1R~|J=*-)S=F#JEJp-QB{M(;5&Rs+!#FuDRZbPcJ4o=k{<
zR1h2?=%x&&42vzxk%J7U0db5@r!nYECId{z;;3%SP+%$w3~Xh_xcjg)!uf^?+xN3I
zMaM6X^|;rjW#k)?D9ceAr>ru5(qwI+j;^tZ=?pV-3p;y<S&mN5E}mZV<}dJGxX8~x
zAaF%ca7bj-+I8!rV`7t%Q+`d|yd^CwJ7>qvUAy<>79J=nE-5W5uQ+k?)al>PoIUr)
zm8;jT->9m-d8@wR{)2~)9zS{d`b}fg+vb+mckkOfI=i}izV`O@Q+82y{-9yQ{>?50
zu#3iE&>1YsE{wJw4!Qz^DKL;xv~_3sgpbiMOkndx+xH*8EURhc(KgmM;vPpy%lP%A
zcFHuVW&i&SOZ>mG>~F*VvFj<4r(@7~bOmICZUm3Exu0=uo=Q=cW7STZWo_kcj`MBy
z>F4Da4UURtognbD8@H!aeRH%qY++;K%2xMGi`3El-1z>1(v`R&B>w;lnsfC~Wt@(~
zm2D?E6`FqcYzK9<XDvOTZkIcC%Fdz-o9W-KxSd{a!+qi3Hc(TP_hQDfuk$ZFb-HZV
z>Gyt0{l$q@C$F;1nqSw%W50NNABmK=+nDN}#ti(RrqOU}#@0~(wJ$9yl;{m+@AAXD
zjK6&PWH#x1%;DcMZ%h?TI(|RQBCYFVO!L->6^AUg?z_J0neSP<RYrR(u%x7i`DfPE
zFODlTZwz8tFJHc(v!n<67$ceH8p00Mn>OdyDZBG#8Gc(b@zNrV`-fJ9tIh~>YBBQ)
zbFNzS+cJ;e_&er=^ha9UICE@V;NuEa|1+fzkBs+~7l#xrI-UN=bFpJsNsvr{>n69O
z#iq{|HiY``&^zk0qwtrSvL|?KgF@t*xy+D#zn?By>T-PXO9!oOvhzG|(vRq*E@{|Q
zJMrpL`|Z^(``0}?o0N3<YsF&OT}1P_jLov=uLjMEIKz(aS8qSE?P;@IQ767<(wxQ!
z(^)IWRAFwv<{ht8++W+c&}-?3X$99ZW+>R4Emk<z!usBS(Dsd7fxh2+9iuUwdEZT%
zo_F1P%Uxpg!O3$lyDxXIzM7%Umx^V(dKYw;uG9FvZZsz|&M<g%&9#;~uM-XeO}g3(
z9i@k_hY-n)yF!lNJK@}bgG-ss1sh$Zdcs*ZV&?LOfin}G`h7{h<M&4$#imcxqsPxJ
z>Sk=XpCFD`?;9waxaI5S`=o4W@q)mYG0e!ntU|u-bd0_(NZ<bbcsxVqn6YrrlQ9Z6
zbZKi|{5nqV)AbN`<p)i5$@R(i2EUd@Y3gKsx~i_<XesYLHs{6S8118D%XhNUziSVn
zqfY)C*YZD4QK^4+cJn|+Z*EsSnp`VaRyFW@{JSO0t4@Ba>z6OsA3f%~L-G(ha&+Ls
zQC9oOg0XF*+Kd0}uG!15h}(WP@T{}e4H;Tq#!-BeO3}lQ&3``VTAQMs;+Xfv(r(aQ
zCn0}k?z_1&*3<qpB42-&{IQJb9rUngTlNx{-%n;G>!uHUj@$cau%f^r{<!J6!e!UT
z+wo#viw#RAt7cU8d_7XNexlA<Mg(&*E4Me|)AXZJ+~ZG;Jr+MX5n{ONDJQv-^+-GS
zLEF}v{NH<&t`$9+u=wUPnTIK}K7L-)H}+{oMVHuyW17;6`aS$C`)i(9KVvo$^@}c^
zG|u-H&6(qs5?_kGn>=1QW5&bS4Q_s`rumEHW^wN<u>3T!W{REhy5jIWf74wScXwWj
zHvW2M+{Hyzs!4vgHp|#{)oK(TOI{XT@v!5z<6gJKdRfJ$3$`<|5C?t8SoC?OhhJ`?
z>+CDFCVyTm2(U_N&t5r3t1e}Bm{Fw}E1wm6xN&x{!;8}|_pDggzawiGd3|*6-tfL1
z=QliB#WFrCk$-_|CvRD)_oCwI^Xk2Cw=7WhyV=-XvD75pCZ&ZzPqj<>-gbnXKQ6uQ
zQp%djZo46rk>=U!{kmPVmU%Ad1|!7aX{Kmg60W}0QGVuryIPk_!tv`Riw>?h%)!c^
z)f)6f`MaR+UelRX84q)A+#Gz5A}m)nWPj9MX7vZdrsm+M+Q=ak7vKF(boA>b)4TZO
z>E|oP3?a-!g&m)qv5I3me$3-RotGXNokQrMgfUn@gjkK|9J`AZOOFjWOc?jBuYR@K
zxdEI0F3f)qE6lod2px9M+1(X#`c&hYSMmM#ed4CjMf($K7dj0DUkdErlhN1z;A%pB
zA>I?R+_uO2<ju(5-0(!ffL?DyT7FZ(e&ZBY`G9M1!qVsi9nZe{{wY^^Yur@@q;FaE
zeQnDb1+Ly7o94A{Vq+T(*b`@5>O68sy{g0f_%iI`i{Mn_p5+(vbLwtAPg=UA5{1ta
zE|{;%i@83{@vwnxecGXp=W)j*8N5%*t5#@(Ic8sr^4|nqIC1<-bswo(?>BY5gzhf;
z%d0^4yrku$&lLr)8M{D%7Mb+uKJ@5;_Z}0bDqL4CTm3Tdd0?d~Pk)x~{49OhX%Z}d
zt~iC3D4eyR<{?Zg_-~j9KNPQ2@<ndh=uPhn2BPOze^~fZ<CpJhml!L)+Sl*8Ql-8{
zZD9y2({OQT`MRT*2D|5++Ula(-#`BMa%C&-+adJmsNi$#O5cg)gZE~4a9=OGRXWpR
zb!aU=!SR=K{mrCc>Yf|-3a;dn`3aXW4}XPi1-8Vz+V<@4?GlFY@wbdVFRg8PE$`-^
z7)O@2n@i3&rf1ZXQy78YB)Q8brOrd&j+Mtp$IYGTU;6nleX_xz4c+-|!h_19jwH2(
zC%a1rGu50!uG~o8;8Q!ztsyPA<GlP3s;roQ#`Z>VKJjrvXN7|L%_qcwbBQVYQ=ive
z#cdMx@}hZmgMVEib(Uan^LMl^l~ZoJru_Ze5Ne%uc6PC?Lbb`uUK{nct&2asJ2!;7
zG=GV9+nV>dfPA`Eo?WK+Ky&r0<1^1h#2p_Dxm;GX!&h9JH)h+EduA@#g=;#(h^M6y
zzjI%f96GKk<n(TT;3&duzE_+$lX*^TbbB@H<-VTXXZ~2xEW*^6=CErGr;amp+%<Vp
z&0=M%-XUat+Nj`^^5s`EZn+xWr&qOj?tK;(QNvm@%c^45PGeLs2UcHNCpP236SVMw
z)`ef4#N`Zn#-?yL!^s`V>5XP(cklj@bv<XfYSzrrrz#zG&pDF9ZYr;BoOf-ditzjv
z6!D_%(qMf3;nWl_{ipn`Pw43$yVCu$%svOpZ>vLr?)mBy<5IjQ2FtK)ALgu;usFqs
zM?Ia`W@Wka`@E=mbC;~S7n(Y#A3w;b-+6grt-;AO8@-FJ-*>F^9sO7Fyi>`$pY8nc
z_|2iXy!C+ti!PL}iCyWl-mx`acS+TI$JfqHLbXd<wX?@<+!Xce%qY9_o4j7#8y$A_
z`Tb3E?2g%_2(}pv@X73&%~3wRb!&W8cd2KVjjk<iObe!~PhDvJXOP?&tB)>5Eqb#|
z*hA=|+iUTOb4?<fFS&MXf7b0K4RqW5`Hc*&PuiNs7Y3i_I{lIn-s=^RarA@tPW!hv
z`@9<;*9(`&491kpc9$6Rdt&MZZ>swGy`~;~k@sDrXsYAk$&IbS2_8+%jQvL)+v3&6
zo&Gv~2<;!=dH(RrWe|;9FLG-KG?mVM=!|cj^~bRSg}YmN=Vs6Gch;Y1hawDT-cOEn
za)@Qgs%NWxdL|RWeQRf@F!g$jm1}(MVe3T0qS!j7C-Y$E>Cx6ccb#reC{f*7<aflZ
z;$V@+7mXEqcz|yl+vc{3Wsh&P#@4*8sbybo=p;;;y6RhvQTht2Kkmo-C&_bK!tkb+
zql-!hw{}jEOP#bhtk#wEQ|VD}y1za6=!D}&JDCs8EuA$Xoc`=ETDT#4-`LUX64L40
zbgqnTQq9~j{=noNQ>u?lTWEUo(%Xr%A~Q<N23ET9*<8Ljs8vUP$(w^A1s$x5FaXEc
zbd8zMDlc1$op-sQcFb%DRn65t_OK=ANzKcJm0v#0-}tC!;Qir3?$m@+c|FU%|G9{l
z|ECS7Bfq8S_c3>DK7Nl<QGeD^WMCbgxk)4c-tTq1<9^REqWbpn&(L5pjSbqZr`qLT
zJR!f@ymswae167YR}fAA@b^b=T^|oV7(y=DHz)QDv??4O=v3c<jBc-rFzEIn(qoSf
zc6&1Dg;=l0C$)m&$;btJa7j0CM$`hElqH-4g?{a2&+snV31htD>|5oy_(8Adlf&$|
zW#!xA;*H!-R^@xsKL>Ai@UD#w-%8YWm&9y&(%E(Mz(&8Us&<*(X=gTUyvUFLYvED*
zwoLj}(>%Xlhfv+tm``^qe9g4Jk8Rz=H*b5ezbAB%{b;_zhC40S1|}bvJN~#wrJ9UQ
z?ZX4(b~aV?B~2yN`f6;pmmR^U;}Ho%DAtBqwd}B`(lwQYGf&6<c4L;St1EX=*XS{x
zcji{C54)<VmLuyI_;f=4=|xk*dOoQBMxH0>+eXcucq+(?+~l5evqne@nd5PfkWE7^
z#_wIws7X3S*Y)XXYr0v%J_U2?Gb6<l=FJ#o5cr%>{&e{+KI!H8-#@FoOK+HdscB-A
z&AaR&q-h_fdoJSIaXhc_h{5b=Bfl_6j@NWr8F@=DAmirSy+`M~DOId`Y%qipyw=s7
zP3DD!(+l31ytS&GyGlFPpi(1e&Mh;Q{`$c^8SUyd3Bex(1Mg6NR04s*pAVt_*3KrY
z>y`_3Y@!>kXRV(aZ-IWh8~-Jvv)3agJ^roaAnl>xS*LM~0{3$rGv3IiqoLsNgiG)C
z4~A54v$5^<OeXSK@0`;s?oN*J-Bgn6UEo%8WN;s=tvR#Oiq*34`H5|AVV7#QR{b82
zE!en!@Lr$Cs;~Ud8u^Rw4Iy2-ODATu%&)opAhlWeLj7y%>p8OmZr$>Im)4abn`mhe
z$Xznyd-BSpKcWW?WW{U#@+GC(dcrBZ$>$Qpb?*gVj0(oIWj){CQ^>eHA)<ZTpWy1&
zr{^1~e{P@VxaVqk?dip5=YLzk>q`oQakb67Ou=8qvZ^yCk91{SQe7YkR~omOtGVWC
zp4v0VVr%!!J}nGk(~515#?_C$+NHiY)v?}hri0bx4ndz$N5@&qu0|LyiL_g$v*hj+
zDWX6v|CV>3S+70IV)!$5kJrEMq+;H8vahEMb<+AWB{xQ^IA7Lgbu|r!ROPua9iBRy
zn_2SWuNupXu05_ew0pVHdF{H+hW6_kj$!Kz^`AA%TweF>f`7bCO(Nf$Hs<&*<oA8C
z^~Lvy&1I&235FHnN!#L$`aRBiUx;<jwR6s$MtdN)Gjkczy0*b<hkS(f8n=eizq}9G
zS~WiH`^W6JiOMZ1FB%$Nl1d&!D7e{e`qwQt-q=nYLRXF~Y5W)omY9B@H{WN^#d)1J
zF>1E&%|ARkv*=s|I$7}j+?kG2AGL*y+f7Hq;@&Si<v94zMD&8aHO!!Q-gQ^=<x{FQ
z<X1e-T*B>Z4&GGHYV*{;^w&Vj<T3ZQBIlmMSUt533SZ+Fty%Cb+2adzbfgw;HKyA*
z(6y2)Deh$LFY4S>uw@8+s55B1+iiR<`%$OyJ<WB``aG|te!JSi*H^8#w(c9$HC<u#
z%OxYXndd*Ly9C#Mz2s7T!m#M8_J_MOf<M~mX6A2dA1IeqEBbCCQaSdO)milHd-XtY
zAFDTbM*N8&g{AM}qhh&N{$yRb)S<U9BTyO(25(xs_pQ9tc2d}OcaJaYrRos6u=1c>
zJLN`&F}v=ZUGUP&__F?mri^8`4qh9hh?>|-jz5plQk<zCaxPr!W|)%U?rAb|h^PwP
zbec2Udd$X(h+gd}iBswv<tC?$_jYUR7AbG=d-Sv<>qYKZhvh|rUmm~9e^cnZ_qMv(
z&503KHdV$aEHAAPW&A31jc)4sehM4cv$2TVwUcG8CBOcz_nCy(OB5&UH1E$%z8hU(
zim$7#`H0Na2jds*SrjVRQ~jdg^^#ZNPf3lH&U@Bv;Z3{a$NIA3i^pKvGJXBM2M#Ja
zYK<<;oEw4P=Du8AQwPhd0+=KpIZQ0MR<dM`b<VF&4mfRRN|QKw`eCp9ht6v(ukQ-^
zD0e3RTutnQigL?_;yU|fFIIb=54QZ=qTQhO*A8*2)k(`|SDAOxTK*cV2R+tK>K<iU
zJHt&iReot@2WTceVqNMR2t3<dAK%~n*R9_Y-j=m2S4%opj&*ycMbD%6##?&4jlXDP
z(O&Bva-xUTQ9RWkZ&P66dzO25iSfOo%W@OsnuRKTn-X`ei=Ez4Hc(w%t38p_%FVGR
zGz_gv`Dk@*TUx{oHwRjdlcSgVKASt$t*l0(df=!>@D+n6!Lzv?!87u&)zNRdUOAu@
zALUtNWn(wTWV_$`SdQUf>N{Is3#_+cw#{h+xzuinOi#(@$d{Q}7RS|Hd{rW3Q_hZk
z5xsKC>)$L<Om|g#?yG4oeWS7qHa?1ftuf)M((ZS|6NbMOYB%@J&OQ}5$k<EHpAepB
z<ag;+$mEnITSmVXcpcX))6A$!+gB*K(dRq8<@0U7nH#uAHF7iW95#KTIWyr&jcIz8
z1vmVc);9_p%d*VR7;N@$YT{vd%cH{<F@l}Ei$0vP2vN_QeJh(bd(6mEVf&72W*znM
zP(Jh^W_B1?HKl6^aqCIl!++e=UA3e4vifuu=GL@n*{#EayG-xO+03tb`>Ok{kXJUh
zP~W1|&GoMoCx7v|spxYg!M~g2zullo;nyYGV!oG@>MLzCezEn?goH<9@#VAZXxQcE
zU9oEPY2o>!&`e{=D^J_zoNYnKMAUfu`@Ng1)P(_}4q{bzS7Y(gvZZG7uV46eh9O^>
z@ce>(v?;Ed-<I8-GVX<%m!DiM9b(Fd`JcJ{T0^M&?xh_j7s$(tZ$D>OnpIm}I&86N
zb+P)#86738owAMEU%1uveeYVI455gF$<=SS3<TS|#Q#}PepJ)$f!>tQt)*S1o0JDa
z%949L-5&S02GKV9o%dSh^zM>c6KA<&_w@<v_t(A~FJV4+Y&-MjzU<;&&(0zR8LJ^w
z``Y7>L1V#NVwB?V=DIeoye<ZAXnAksrr_Ax(pIj-|Ncp^{dDETEdv#r5iJt2_1R*7
z@TR)7&|O^v4z{A0+UIG0&f0-cw-5S!C5hHORp+^b@3rn-){pN_I9?ezP~oc(tzO;W
zv?L<j^^NW?$k*@O$?l_0Z7T-SevckPp+iVy(lSFn-*e+tn4jIdFy3)M<Kd4R4Zrqo
z+vs>D=nwqQQC}DC)>$6&ir&zWmL@BHeP2m2GkxK)=*AMO`23K?iQl8yf0=$+YOu{_
zX~;}-<?I$G<*e|*nLW+tYXrv^C&)Qzt;Iv99x;sHyln?Eczs2NF3-Jt1Nl7Pq)r|b
zI~!IQJUhNSp)+C@cUjd~y(7)G&7`Zrqr^DJzAuD)c+ciM#lEd!S7)woTG3E?=~4ae
ziSnZv1(B>Qb4&G>1-VX{JG8>gl|R`UY(4j^_JxGAuBZKGsG&~4C?kW*t<jcV#}`hW
znlkaCd&2pG&iX?EV-iJQix<o{W2hMs9OkHp`{o))VKe8ie!Bm@;d&)rQhNGwgO-xW
zcZ=dv+|R8aR~aha%etNK{34+@LP(gtw>Q$Ac*lNwmTgr#MB3^7du(3V+^SeI^YE>d
zZ1l%l*9`@s*S>3-M9kS;FxWt-Eu!;Gzk458>376NFh_2Ru-bKB#9U#@5NeUAJwB?{
z&1%fpv!y>*Oh&Mp=HheK`V_1!U9C5S^pxX-bu6saW{yYR*OI=K&y|KybUpdoGxa%j
zw5Is}yFxs!d0<<xoWp<%_fc*vU$vsFpW7y3JkP(hzgX>wjcsoHo^3_q$rDfZiEk{J
zY4JL;*AsT^8a$L6KYw&PJ-NDlU~ad`j$0~qVc{Qjl3QTTlDk(nopIfml(9U0UA#F-
z*d(mIG<Jed?XAi|^QsS?pD*{#;@0QZytprB>b%m>{#50+bnB~o8FC9@Axte_m&QbG
z^Vx{i|0#Ew%-}4dKWd?dvb8>&CP(^iQtZBc<BW~9?dT$&<FR+|w#9QouFFk2KkLBc
zHC2`=<{!GsTNY;+?<O^qZdQK!&e&wE>*_Ie-+BMrk@uHFCp~z!Q6rIE@~ak(jrp=P
zmC-pEa%(ZWXZ7~BKaW`%Y|9}|=<a*fvaR&l`w%~3Y}1Kt&Wq;WIdqjl+LV=#`r{9Y
zR^slbSIeq~P(|@Xok}soaPgtX&lf9Srl;Q<Tj8KvExSVYZQ|p1dG}w8lG&Y@J6ETq
z+xS=XFz%}G{w|%J9VgFQZ_V4}d&*#*cH;U!7R*u;i#R(q^V)($D&ePI$iKcJ&UEUF
zF{^6K{bLA)U;MRVF5|?KEi!cn(>H!^DXS{6S#<8c!h4;Bw`tX=v7^{f`>S(*+Nq09
z?pNrpMzhSS2QZVS{QcsYT?(5VVXw$lEi(SLQE%EY?ewS@LulihylLYGwi7&diRBJ@
zJ4{F#k`?2_ZNHYM5224qz4v;4t>Or=#Fwt;bK-P1RWJVHR0WfdPex6Rx9RPrvVm6q
zHYA^9rj2$~Cv@kXT>3Vn(Y^2Ji`tg7H9c0pl&$i|I?PxdB?U&OV#;TnRiC=SM&Y~h
zgQ=2TKDDbJ`e|CP)8}Vg+57gn`sa)rB}S8D>y4{zQrk-y*7hxjcJ~M7uG&4=$9<c-
zyT+id;7Z{Py_)Ov57({i9j$hKUi>vxxqpEq+RE~h+|y~1U!{(+J~FIX={5>~Ci5r#
z%A|rTJ2!Ujyvb6(F+9UjXxm$LSh*r)O^3H>w5p<2%%|B}kvq~gUu~JK7JJk(UiY)v
z?S&;S&BiYtuWdYSs~Y+F`IqeN+s~)d;`j3}E_vy-+39jd$nwrSzrK3qZyUDj-?*`q
zwKOli`^)^0M8{|qGD_~$rwJLXGoe?C^lRG3uKVS2r6A+0)Ay*l-K|H|oK?p@E&EvX
zs!RR*9~x(~W?=^^pSm@^>wl0Yauq5p{$!{uh*ZDxnV%lT`qSido&kH0*`$cl2FnN8
z^(jS$atoXCZ1$+kQApn=A>;f1N-=DH`}*8AldWlnS>8Hkm(>auoyjeZ3|up<aJLzz
zkdt_>#_4NMiD1RKEOU{r|JZ6XtH|_t-Sf)NuawmW1^nThUa(e$wu2G*SC{&ZAcw1-
zM6I5p^~+HL^Y@Y6MLXtc@31+qIXeH>T<gLMC8w>7yR?{&0qQR6J~m)gt2`2&U&p%;
zHkx$}3CcI_IgR~ubX>~J+u7fW27~;@6vQIyQ|)8E@(XR&7KcqZ^Wb>bf!i=mH6B7!
zhtP%&l~Q)?a<|uM%6B@d&U1XWJGOHxJ;U7E3`%x>I@R)0;+HNNLSxj+u1{{WpwGoF
zwd^ih{$huN0jI_a>(AY~FZ@|sDcgU|re)RcJu#_x<Gbr<-r1a(=~Zja9-Y;;ad+F+
zZ2t)F$j?m+3YI*(vwZcns3V<Mo!WaX?z&&m*Q*J$?p!c`+?y5F+OKC_eXpa!xp&;x
za9v*G(cXC}yPY}J2M+pmU-x=p?xo><`^(kHfti2AyUch#HSO1c3H}p)TfZbqb?uSH
z^xseHOxIU#VW@3*Q#d1iV$QXQgJ)m$G!6upEP0+WZJr=4{QMc4SHT^{w`b=pA7eT2
zypc~Tw(q+&)l6V}TaA0hCFkyBn+Hvw(ek|IjRkkU@0hwgSzM*J&FxR;r{%IcwD#RE
zxv#0I{Z@OT&7uOWZNc81&dy{*eZs?CX{;sn%AOG{Mq13FwH@)NreWKYR2F}YRt|lU
zp6hGkUC|hcb(<J(9Dnyd`*HD}RLu#IwI3GEews_ad*rpl{WUipRPXE0i&rRNEPZW5
zFL>t}q89RDQR0z0%deUl@rM`OdD3j#%jsxoH>%ILvPFa<kc~rWLjI>-Y%m0AS^0-n
zdL<eS*cTl1%5jo;d2YZ-{noB2FLt$uEZ?$$$$ve5TT2@sZPn83-PNWdnUlKTuL}}m
zy^=D9kb4oscg_%!`#yOW>&U&n`xm~YiPPie3U)U?p{D$uM8cT;K3=bPzI9#dar?7@
z8<yAPQuRFB(`V<LkPXNC-{dgdZ?FdB#&WJK{-STQ#ik~2x$O<ix~PFBYGp08KCO@o
z4mLWtnha0MSmm3<_aFSL!I*t?J^pQGm__V@xYrJf(erD=s+$^Ho6pYP-Sg}if=TYA
zXRm)V`1n`9!jZh`*J?JpUHR>Bzyz~Z)%vfG89XvCoxM`EW6Ifxc}d$GS;_}Te^6#V
zQ!$yar1E^E#@4E!uO=_Uw{Cl(86a4OCuKRpe(RI1-mJlF_es}Qg-y^lyD`fAvu29-
zxyp&{p9-3Ugo~`zR%S-AXqunF`J^jbuYL2hexefKo^=PKKeFV$onzlr;Px;xyVb@b
z`K809Q1d3g=U({{rvfUqWTy0~dMF<-hg(>vZQp~)cr^D~rs?Maoez-{^lkTDqn|m;
zvUwFm6T4b)Yjz*gJ%8e=@`{<eTgq=(>6Cm@GVx87ADi0fIqSpJ4Rr&h#kBE03rDMI
zX_~j>9xL4IQnFiBrdY+|#aB(W;Ic$)z_FwF!559+mR%GTZPzxdxb##fJvsEog+G4#
zYeg|PMX4vl&V|2Dy{3GkZH7~&dds`H-!^1?&bSJjZyrWYTpw4t4+|&v``xSTSO0!m
zJtU`l`=v!wWh#+pQd!r)Twmjw*puov$fh3c_rczQ_wr5Nx7v*NxkReQkOj-<`XAE@
zBj>Ffm8E~-tCsunNu~-)FKU8S){Q#YpLVSI;LT?r#O@OgCC7CbHKfJb?~jRadszO<
z@i91qbgI6)G?_a!Oy{v(Px-`1SJkM-Hc6L>oY9K%fmUqVGv185u!S;uqe*ZVwGFms
z+4&glW}AJnj!CW?bpoVm^T$YIgXyHL{zC5+E7zzm@$0{D<D2xhZ__)5f0w4sn~=zd
z+gF}#)J%J6R^z4GezRn$eTsL^OPB1VETgKiEB-Vad>Z6{?l2NG``ncAf?cCBnshZM
zUA=q9?D>IU)N=IQ{fDKYS8^@VP_c2mmcRDhgni|Ck+luwrO&ps|4z^h0{bKMEjAR~
zI{ejG?G%}DbMrPo&yG(`>$-}0H;{%&V=2GgRU#8(z%v@3=U3MC&O^x2mtnGsC+YLk
z!ag}W#-h$qYn|IC?@zRO7*%`P<Hk#8`n&n3KW}s^ORDa)*hD+|=3CRcqlTq7YG<i)
zCReLw*VdD_KW|F*;&$YlwVbvJahdddrhI*9tunpst_gV~+|}(o!S!&9ulCyOJ%m(B
zXJMZ_^wDS?`6G2x)c0*>9g!_$o;GNi`AqxygI*H%=bRu%licFTUG$+R@P3*6`k<(&
zFpKHa*F;Y9@elP4nC2I{dU~`^7`*GEKOI@xM2Eq9Fac4xZ@`M+HC7s5&R^2NgZ-^E
z7MRS{pBrWu5ESeb8xi0c>+0nfyVB3xU&F>4w~V%kULCeNAj$`iUcG8fq(!uqhE%u(
z@Ko}24P2TfYNeHiH@qfh7a9?O8&5NyrmqLDyR9?S02vYfffgS2vqmJqm6gVbs_WOU
zpSIp;T4=<I=?3QJ=F{~JryCmTK@Pphm^D#8(RynlHGfF34~X=O2o8%14qby&5`BC_
z*G5@sK>fpttqz+z_iw@f5rfsMDQ%^>BcmMFLE--mu}H6&uz=|v0g<6=Bm4p!)&;DI
z()>}}-|z3TVQV8+Ne%G#n;x(#U^Qfm1jG&gu8Mkl?(f``!7GAShe=BSvVZ3S{_jlx
zQF*Fru+YLj)Nd_SiL<?x1|`YDKh!VS-$Kfo8^E8Rk-wgaNuZ&gvA@2do^N14fZhy$
zvl#}y20jLU<_16HIIoF}@>$~-K*<5g(}F>WiK)5Y3}Yi7y+9LVb3J1peIq?{gFtgV
zKSN(rb3<Q0fBityALZR5g29b^R{g9GCC4A+_?a2_LKy=+Q!{;2J>x)gGmv9!sOJO1
zj04Tg4E+MkHE@4Fi&>!&t9_syf>-;j2$&uc7O+ACr|`0{6+2sL7*5k4xfHMRi3)_g
zRvL4H*Q|{mN%aa|9pD+|Lm_7{!${xU$jo$xsgbe1k<pBi9P<MrBEerMQPQ^;rOI2_
zMFjXng+_RVhOV;G`1|EX%Fl)&b_$M!+c6`a>lqL}!mSF%rG8`K<41X+l}4n`x&VKT
zkwm{BpEWB2pg%#I4rdGg4-osB1^W31`kLwK>j%uxGxjwy*7GqoG1v3)pW)|c5&+kf
z^26mLLj$AM`$PoTu7LIiO`)}=uVLasDGEA2@~saFpiG4KGW84$^z===M&8Rb9e%M+
zV>;+Ro$4~frcD2LE2Z~->RvyxL!YJCze9jE{$Khp0{=zezX<#nf&U`#Uj+W&L*So9
zM8F!D>8yt(!q6MqF=sowWnvGvS<Z7D;5B%7g<Z1BZ*?RCUhQAKCMv?iQG~-A**L2e
zzIma(=inlyk6&b%tynx~`0eTc^xN}r_zn98JxbPp%JILXjq-=Z0=!R-Lmt8}EFuc{
z<-nV+j|!vs-N5r<!sZYBA>gM*fP%n3rqX?f`HvL8LdsJg1t5NemxmqLMtxm_ulR}g
z{fYMridYSKj3JMf|LQd$pXNQx`&0ZzieJ5IE$AjC>b-dRfHm_x;j9IRDsn{5@cn@p
z;m8-QMZstlib8sDUIQtS$OG<+{)h7L5d@SMLn&V<6^#6#ggpvHtC5eiJfhkF@&NsV
zZDgR4p)?7T7r}=kFNcOYC&33W*$5404h?<VKQuI02%j)KhYyO@{8K(uf{<A^q-Xw<
zrd0q9n~c!!*ZxWK-G$JZRD_hDg!x4HNb8})FD(#0gX$<nXw*XZNU07!tXVva8{DJr
zqXUl+@&c^p+(l>)d=WKy1;}3VKiExuHT5@c|7)HR{G^{v@JDkMxuf`s3X0>D_{!==
z0(DhYbsN3ulZ;$!UFSO6I?eJ}vd-UQL8$jEr@-`}(Ac=-lw>!*?RzuG+;s`b@zT#G
zl*WxypP+6j5Lm{0IeEqZAJfnk_-sNB$qmpk_+sG4XA|S8uf`~%pcp2ML8mcg;M0R~
z|8@gz!<Pj^Bk%6>=sdW7d-X7myR#V{Ph2pr7S61|a5mSWoRiEzg*rhViIp5t3XTYb
zBq$TIGh^`I%Dc{F%%`fNk<{-`1Rt7Ugd9@f2o<x)BE;t*4h8tXQ5ugwn}L|nHE_%=
z+vK)sPPZPSB~o1l%0O*!1HNPt!`TClVIsld{;rTf=&O)+<&aoT?3lG6%bu3Z6p|IV
zu98j&yRaPYGN`PCm;qW7n24a?jZu&*?_QrVAMPO`9RErLAi<$A^f;t0pN7N)65{`w
z0j_vV!hNcOEeA@ok>}gFk$3ss!XgCn5fdL4i4mws2Qe5-$iiWB;1WneRY)T-4qVXy
zNMjcoABV4Yl2Hy)vu&OvW(Y-<#d>0nE#h$CvpN!^@o8C^lFbN_yky)eB<`0tPn-aH
zDF1In&_Hb-sGUZmks?xqLrWaV!THA>NG98zx!qQm1QC?p;dfK4ZWF`#Kv8K%reL@L
zm;u&+whSSh0C5UH>Qg6Tl*%z+FtI(_fn$V#>A_1D5IV~_Vli8n5&_w0DLh;cX<@P+
z0=Gp4o+YxjP=J7Y|3cD_rZ_Z#gBJ&s2GL$4SYu9992+J^5tN*?osht#f|F<S+_0Sv
zR<j8ytRA#XNERZJKMz_Bh+s=k(sl&^DI(wimi)38JV=O=j~Z}I*qECLpJXdeA$45D
z8kt}mpOk+5$l=qJaBx+xEgrbl`?#gMH|t1CQ-XdX8t96|G};yi1{=VH3!xM{@L1|X
z>Z=%&f!XqD3;C2Dw8|7~|A#R?e$P)8F;S6_0u=GU1u*Ka7{-VcCZw?`_1R(q5=aTc
zoIqcSJRcetLUy$vlp+?<_(hCkvU)mf=w3n{TYQlsRg%DLgD{2%T9nD0<g|mqT~x53
zL<5Qr3y?|~`5{0-j9~9@Nn`V9b%;S-5ujlvv;Yp)BC7du36IGT%leC{^4>IlU+2Q&
zOsiX99H=G%Jaxb)Qb1R!P#X2imZl*#)y(4I1|%gU{fI8fLn0o}i-DtJ+?FY)hl-Je
zOwn2-K{!VTD#Hi~f-;ELDHi?tq}#$5ho1C5>OIeafgwG@oeCxr1U2hGUa109{RE;h
z85~&=6^)prz^%Y!ud&_zizH&t!$C!jI;agSCcs1(jyYfiGzXuSNFzZU2{%Cx3J@4I
z7D{9Zf{?{TYS0Xr9B*D}VJ0n;M>^<nkWdSNNr1{oCP<^Hh4APvz{@qj8mi`hCD0ez
z7s&?U2-jnK18QI#4eHDUFmY*K59EaqE@n&3=BJS8vPrfv*P|bRlIjXYq<}~uH;p8u
zWgs?D2tW}kSi+fx<6<q0<3jKY!Cr4KM9V?q7)wAurx<J`VLAaZ0rTJ|FfoL05hn;-
zcXY=B+W*PLaZs9xkgkw{7;uJKd6E=?6L3Ft<^bq4CO9z(fH}F-_?JyGjJfV!Vx$8;
zEkQ^?KyE7Tv%$#)xJMeqG$+LinKXWBAxS|@c{!Lz2!%YvccGCqOwCOL-rHyDuwxOw
zD>lI?rc8wKfyAUDj>KRmj(C*3=F4o)KMOo#HUmIZ2l^LcKrX<9=>U51pbSGOfkyHG
z+#oABKM+$1A$w4Plu($^D(U=(mUULMT?9_^pd$e2BFF^Y6Nt)HEVRsb@kq?H=QDNS
z8(&(^aVI%KM1qxkBo<S~A;km&hxkjQ5%Y9x5v%YzN-pCNlqzD9cZ~XO7f8#U#}x5W
zbX<Rr`=uF&#*j8Db>|21Pznw;a?;QU?tIXg!U_m}`6R*|NJPUq2&c*xF=?pE<TqpP
zY##*g0#GeBOk&IyN>NZ@SQA+!M8!fR;YlRYMx|PZLQO80QNbWEA%j4)a$i0v=9AQ*
zL=}S&#1@g56Pv`EIDCF-mfFgjqm<&}0&zZ#1CINz#D-E(onjio*-{-q>)|n+8hxoa
zhQ>leXdu2glTurP81j+cokaKa{V>EziSl8OvCt|o_L3M^d%If3JX?0OjtdN(TrHi}
zE-DvOK)^3Go{PmuMWVYI<{*fMI3znog20ui$~Z`t!}g%M;Us7kiGYpJz4C68-HD`-
zDhaAHaS$?0JSYWjI7tH03%r9N6a*pO{t=JDz&A)j>PLx8hP#N6Pzi_vT#0|8KTSyj
zjRA8ls8reu#0Umyz*v&Z#93L1i8+uLj3*#GQPxo;mdlkOf(HoTI!FfwRtA7g3D^;&
zD<VM?lH6=bV#I!5FU0#IRXc$Iv3$_)AHt|BCIvSl833lJE(Ni9RQI5WN4YvgVF!d#
z5|FzSl_rG6LWB2l0<=a9J#Y=FI3|K=5!`ixVSpfFkfckLnJmCu*dl(Z1VEsXBQrJ#
zHF8sFW1Xm|4jL0;f}AIvRe@Y0Q<6Yz7Na7z7yv_zqgs<jna0ia&y<K^?gHlH;CPe{
zQX-_DGD#ve8e!B##9jh=NQvZ=5&(>kXlx1C1_Vmbjn5iLtY)_sf;vn>gQq}9AiBXl
z4?gmukwUwhY?vzHI;|vh6ek4ZIfo$FP!K?qhFv{@EhQ=i7fL5#Y!L!0FoLNe2D}CK
zIy?s&N<?fI9xbP#ylZV1k|2-|aZfrXP(?u}_=5>haT)NqR4p(ZJpLzwh{C1;9zbA^
zeEq;cklb_$N>Cz6cUwNfY&lFEoRtp|iAIC?;*p7v)q>q5jilULLgN)733wNS8iA>3
z?}7_|1N%G>ArOg@wHEQfM>zyo1XGL@0UKkJ5=EwD5-N5_vQ0bKE)P~TJoKf8Yv>Q4
z>`$1p$s)wWl}L~aH(+{9*#|#AkvBs~!<JJCv`m1_=yvLOvg=~uW{`tu1X@I+ZUI4g
zYA~np?}B@L2*w=DmH=TMR9X>Juv}D@0!qk2384;?gEc|==0ymDyO5QD9!4<0g=m;R
z&N?u^wC<s=J3;A8Igf;dkT^^Xj04H1q!-XV#3=#-htu%Q;a>#;B~aZF`VCZw%w~7C
zORT1sNhlQ&Xv70QkuDk72xqVob0iRwkVih&T7;5y+z5I#<dA@e!T3p4t96RT2l%lN
zf|gYzWW%5>;Rz)$IRI;6T86{)o|+6Jfvi8nBqVO4Kh-f2+XL?ZLok&GyCqZt*aM-T
zk0~bdnG8?|2qFwc-Y8MGQp$lFB6|8+gEH>y_fY$=qe>9C0$6}k&4g&YWU$VI(vyk6
zcR&=5KvBw<MP#>-$0sqwog`%F5lBlY_8<w2t>Gobn#u=G1B#F|t^J%)u?2PuHAoQM
z@ssBYArMhoV-9dbS|fy0K14$u4mc#0^h0ZgB1v5W0Se}`J5QF=?_W$FPGPdeY%$a`
zd_^W(CXx_@JTi-M0o(IltE2;h3-kd38X5_C70~-K<eaD?Bo3_w1JEKKVt^wEamYb+
zUOq{JGXUCCN#3LKi*J2FG{bt8%V*D=Sc?VO3vFL@aQ&G=Rw2|#%Pc|!6=^C6s(z|H
zz;p^cn2Cr|AOR`{#E=r>Lap}6x*y?Ki0C<&Y%DUTUi=Mnf{uxF!SyK(DWZyjL$k6#
zh#Qe2;A}uVQhFc0&o3)m#7O|W-9;{tNy`KJb0+ergWI@3AjUZ0t$-aYhf-ibAi;82
zYNr^#U;U6;DPwtkvzvN0xUM+heC^>X89$*_el62Nb#t(BCNv~WNRd5+T|P%bB2gm7
zhhPEG&6dEKh*Y(0dK1N%w9<-xv#H7MUXAgi!bH^!rd8hM(Wxi5KbN#^Y}j-5K8DP5
zdcavJ^M@OX*A<%s6J|^dGXI5iIDk1*$i#8ME3gilkd~N;2*|^yiIoW0y`d$OIP?%q
zM6+iUapVZZfYl2rVM-_Q$+%^v1Ue76h%L5qX~|t5C`!Lc-uiQlN*(!rgN3?Nsb=$Z
z!^ug`c^%R253I`(s^o02N;*f7W@R1w#$|0VQ>&?RDH2@aTUHtEuWK3|`2LuYnj0%C
zY0vYf3oC<*6$q%FEY-v^Th|>4e}8eyW3@T{hPiXd);Ric+q}N2B<ISW7{TM3xTMEs
z%5JR4@Q<nV?RW01Gs*0Uw3t{4coV4klF|6>T*vYtz>w<Y)Pw~5iE1M@;&IsSwl1WY
z0t^6vB_u%*2!b(~M#4Nub}!PiMKEJ#A~B=@vzaLtd()7t7mY7T7Pue*<#cSZNU>5d
z_gT$de-JWyGX321K1ap;yDU%hu5H(?7iQ_Lvff@<8M#dM<2QT6@B1|@+Cr_iWCIUf
zJLfhz$s+1di^J#KnpY*twM$$)C6-6CZ}xcJ9}ORGj{fzak^PeHcNRzQTivAKNI`ct
zCW_VPJoQjcX9~EeYpwO3=c!|w)v&ty7{h$7W1VSc$HuGX+*2E@eeyasu>AHog`22*
zfnI#XmFqZu)PUo0i^&vVG^HGVF@l{1=+m%f<3KRC^+m_D#7M$`iHum33}tW}k$@C+
zrpXe>%LPfmrP(@)lBsMqv8c;~lm)>F2%UC<JPaMSJ(5tP8<R1~o=4x*c@*~Q5MB0u
zy}UxlhjK~3_DR9Vdim69<9)Ryb}h<oj-DUE^D_f15_3nj*djagj^D&}8w}}}_ixHt
zFXP#qx9zw67HY05n>ym`zwk=NUS8^16qmiHbI$#x@49QlKsl11ShO+sT9lP~SdV<^
zEz60d;7(hVae){0(Ns^Ir~92Z{+4PP6<Bp&V0LR>ZF016{@lq8+vT;B-j64wF4A^M
zbKI4}{KCUl4u~g)#W)k(js&|ggh;khB1Z7bJj7x#=0J4;*o<Ky7(56JmWDwtV(>BE
zIf)COKx{&^ljPacQYB(UvrL5II<pA@codtRmT4tSMtlg#E?b(CX%}~%lp>>seWc5F
zvg5|IW~yw(Z*0sAwchip?}kzN+2;gyt<RC}WSL)ExSogWw12v_&+W(t!vf3H)1M2(
zyk6$a=?$(`MhQ91!Pd#5-H%!p@VnxAOZ?t0u)EhiI#Y<lEJl31^FuRzPFw}OaSt~z
zJ6t|hbNk1RI8P_*g~^ZO#GDv&)w!Ab=aeMf+ug@9<PU;2EqNIg&}3j#$HqgUcSJiE
z^JzIbFs6VJ;2;Ep@)oM+P^(JV1H&OR1M=Ha1EmlyAcny#<=ipagQ-}MhMJQIhAL5h
z1&BxhIAKH+;>`UR+g_1W(ybFgctkYtk^TuahpE1U&~<g)@-8<CfeZ@plFsln{Rh@t
zPDzlzwfn1nnY6w->+PUl;L&ZKU(^zkE7utotoq1JjIiF)Md&m&M;lkFakqvVtFP19
z+_yvWGK9y3aZQN5>Mb4fG57MF-qh^iE;k3uJ#JYX0G+PT$<=;jrM9h9;-r*}c)cq$
zFS{LEiBgS@fhI8It52WMNy~pLK~y7(!9G3@rV?PWm#(B3)?X6Pz6`QT2MkCDARrS2
z3MUWRYz?JydohL)WdMyw4ouTwTuH_!iD1lff)y`f;0KX+v2Gfavbzf_<G(Rr7&z7(
zWz}1gIMYCVYO81Y>&Q~S4-+>dL4tK`m+9VvO|$&Y^%Uc@J+&!IYcIw!tQYQ1UEBZ8
zuW<qWUUy;A*&>8`d)l1h&c5+6U)(jSj06;jn4=D?>{qI_XM<(z-X?uktH#_CBV`)j
z@!K|{byocTQV&HvF@daYU)>ostKsIME-D~`AE+CRZx^Z+OJFpE0bFV+tjeiyi77xZ
z!zGe=ARY1o7Cato3BaNlJOT*_l$GJYoC<p=POxOd*&L4?*y`gYgC~eCJ2=Jcw~vwF
zOs2y89FPptcnw|T*&1rGHj{}%qj7T_?&NZhh($>{?KL+}9#4oV&HlnKLTww3E=&5W
z{^%=F?ufg&K_XG*ZV`X0BIH+83qB5Fev>wag|M<e=1UOjGi{MQ^i?(e;b2*7BFr)|
z`NH2(xU~xRL<nol+kEa>59IkeZ}WQVb9K!dH7_5!zruR^{bZPV(~``Sn?3bql3K@e
zsF^L8+G%torZB&_82le*7qA!wB}uRn_BA*-M$ky8k3n6-7y@M^(ALsP4@?pEBQaqM
z{>8u<zd4;IQotR+o3i%1#7-64U5{bH#7DwslX^PR?v<Lu;s14^^&wFNvII@Nzn)BZ
z^-cRkCS9T9F6#+2U1tA{6NsC?u`bw@U*jX_8u#udvHyzoDDLmUsIR?q<>m5C^`6EX
zV}EVDcc(uUywiH(0ns{h71x#Be!kF01yd5zU!BhmV%^bKt349T%5GG1JG2hbU*)^W
zT>VJj)8@u{ITLx_v3mfsnbF!O6FToS>CDxBU`06t%qu7)jukJawigKK@X#$FV8XmA
z84)mTORR%X0okO)hkgTC01(p4*#uL89V0-@GLOU@`pss1KVlI3Z8hqW5&ixKGcQ=_
z!=5N+Uzk)UWC$>X%L?6<F8<nCx{!qOG-Y{|>KA^Dq&F$oq`nkd+g@L*T`y!AwWx5l
zBV%^IiniQ0?o;Vmv8r2b&hxnCG1kT__DjAw7&_iMlKRks_Z3_-V%avA=LI73g7IgM
zwDxatLPE<uQv|zv&TrMZ|7!V?n@!(lWpdN)F2r%Vh_H;D$Rj1$@97C`313}br6%P>
zn${svQnR`}3|(rj%hkg^!a5X)v+>S{ok40luOpIL%n(f2sFE)GAw*K02!aO;lKd2C
zXA(dJcN!sFXn7l7Bw<rq8Gs~0WP&Bh7JG=vLLD#ruEIpEG7;@|Yq1VE4EPya1pf@`
z=z$CG9xkT@bZksAWBJ*3RU>Qn<gQI*mtFIoaphf}wpd>8HH!%!f}5=u&MCG&SAiCc
zd6n}rPC+FjC%SBn^|6fHz%vo6eiNx-H%w5#vRzBZv{u#0aQaf`eN(sxnnWI|x@)Pn
z?RMF#J)BtUn~&*l_pn!PHMvtVCgEdiT;y@L+>KZ7_0&{l8|*ZiXWfQ)s&46Vk>_^z
zX=kZr1U9ykly<1}PR-fCD?3(sR2EC%l7Ui~NJ4QG&x2p;LO`7)NJOQ2i;GYSbdnSR
z;wyMo0*pAiwEP5u0b!TiqC{r$P^^_u&3;j_W&Y-vNjWFdFqFoGZCBXq(ZE(xv12zS
z$f!VAnso1DT^x;!wzkcKNx;I<D{fdFt(!}z>sQ9iQ1{k2F+2HHtg-rhD{Kij*=~QA
zZDmcQRe}9qiWdb+rB8H5&zjd9Wt~p?E!!*ZR}7VjeDh$&>@JTS*vsl1Efaa~fpcZo
z8tY8dvdr30&9&5X3s?JOa2LeR<kmGa_oSB^J-PpFIU+V@>pZRxG|G4JeA0DSLa7Gb
z=j0J{N$jqB1UZ}_eFy-EK=Xsm7)%$SKT%gWiTFy}Q5o)%%zzC%J`6a5*;BP*aM(*P
zLP9<&0<0h$QTt&+Ru;clz_k~!Ik9NF3kGX4#K4i#-EXyV{#3P=>XtorH=F9Bj8TAo
zK6|D0cCPj<%e@(aiRx2tMVqJx;N{cDG}8BXjXEaioPss=thd}JUwR<?Y%4cCK&NLx
zeV|}VaM#={_QsPM-G|!`>*V!zw|@Mi(HHz%^&pm`#&ta4+%=j+0S0*-=!o3FlfkWO
z+|71RMmI+QLUpEzh^u|gGQVayk8aytOHfK7og|CbdpWVo#0msMLM6~()Wu<pft?5v
zh6+B?AQzDkQ8APP*kA<a0^I<@D<Tjg&q?5yhj}G$5fNgTu}ii)F)<%61PifN6k`Q@
zI@_CIpC8sOq=0(t0}e9YBohil_vED(=A@3f`6|_p5Oh|1J~`FAoMl+B^qS7JNSI1X
z`meBV9n$ttOK3{<dp92AX8Fq$m%rk*#U-5@t$oKbf6A)!*-u{eIaZ8+TbEyn(3H!Y
z!GChMmP!!AT;)!SVg6m2D*#Hq=e?%ZXu-#>b;dj-|0_)GGkapqGyAoVQEDLC?#M%Z
z$JBP+2NXwQB()Y5!Igx%O^_TWBB@oH4i##F2(TO0l{_?hj|8?B7%{j`5Mj&df+2}P
zV@eS0M7hu$E8qzOiGCx%31}!ML3XLVtzBzW)vzlFoFEwSikJ*m7AcNoWnW)!<>X%7
z_o{7!^`4LI>J}&SFW-N)!D9SeLZxw3gvCTR2J7U_E<cH3UXa}r$tSfrZ>{+6L17Mk
z`>UQPP>UF?ZFH}$6-I{?22$IMP#bsokDUX=@y)%R#<aR^ODRvy5fPt+=>x3GVg1iU
zr<kzV0t=~|pe+NDMN~s0=mBD66-YuUR9M86<s_4OFwj8e%@kHZkdFgvPaGw05pQS4
z&iG*-1IiOPWPnXp__O49c;0i0u*@kIj7~j<y)joyiG>~gYq78c1lwv9l}XXYfde&$
zyCXGtONq9-<oKIPt_Mjy^Mbu1k^^3ds0|0IdML=_gT^!-cx$Ev%%_ZjJ0OY2rr(z^
zafIVSrXK7?foz%*jT$b2z&wf+m<Z1>!6?t1csi3x;lY}p#4uaP!K1E+2?x1SPFh+h
zq8`_Qol=6?B&9YbUrNf&$D1Xd)MGCok|xEKf&?A|-~g{ZHG`E<)=*kVer&;mW6`Lf
zixGNM3&FDtX@$@vlo*KwaYUpo5)vYIa+%0OcICTJt8PkRfh~Ydb=bt8RDyH@4j|YC
zrhw#7NDuFBLS%ynh}16S2*U9HS63>9Ab9dm`2@AK3lWXYfh7h=0o<hRITfC8LQ{)K
zD$DjDxFxmM!iC8o$6^984}&ftcK4!0Fqi^cAHstgd7+&c@UJA<xC@D5Pj2e`i&~A+
zV0j{aAK{;ehm<^8Ce#Rl7Pj@IZHUHFaRw?KCJA_@lXL<^U8_t!)?Cd$QJ+JDKE~mQ
zK_Eb;lnhfj@B^@b+GPNFI77NT;8lncNhk)P9Y_2Vmw6l>xHt!gD~T2pv&T4cqRkR9
z*vHTXB-zO*1s+5QQ3q30Jfo{>B~>M*3bj{>B;Z2?AR#3VfrHaX;3jAeL^UfU0E1JF
z4u=#dDiH<jl8H)#2a7^4wt^GMPvL{WYx`j{k@9S+HE2i!ijdU9HUMvghWAUfP9Z`J
zy5n${kep)D#@Z85Pf-!IJz50MQU!zr2TEbjR;*iWoiS(KF==`5BESSX1OxV&hE;^F
z3#k+d2+Donrj)2og#!n<Fz_~M8vO6mm`$5Z5Ih5LfVBx?ibb0h%cRg!Pjg_IUIZ3{
zvLpSKssSb!l>AI4->Nx9mIrJxU?$Z8ela*;%mhc#NP{LLF*bp%>Yr%vw350(wONsr
z0Wp$c1_2^KFtd~HoV!S<9X=pH0y6!C6r^ESmZ;oMfE`^S30D$0Qs98a5Bv@j6E0!n
zI0c8I1l0+o*7J0?@Rbso0tftu;W|`%iz)31_!SXF0taecC1A7v72OfB<d5DYl?wap
z5aOx!|M6r9nhpd)<^K>204iZCmii{<7Z*~cBw)R_cqAr5KctbAjX(sb$Oqh*WuO5?
z_Ap(S03wMakEU=0Z5Jd9ZxPTSM7VP#5=E$updQ}g%#-h;suoL&5z=31c6%WKP4<IG
zU~oVk#svzGVWRkdQx|9HY}&N88I_g?0W54G_?(j%UIKv*2O@Adp!G#%JRv{|dOAQa
zfbaoUFfcxwy@Pm94NnMK9*BT;mb#l#qO_MFC(1IC>|oPkj~-~4xQfa|<&hE_6aCN&
zu0aT(1rd}A7N~5*5-KuNO8-RIt!4Z6IiU?ZNotlRZD>#ko=bDoxZ3XU+>PorB&5J|
zq0L}98*G86Yzbl|F@bZiAa0C=Qh@;Islv2O*czs)q=;%^P!Rlqv)FpqoTy_|B2N8K
zB-Lz~rLLeXL=Nx<;Kx$G|A{z6^seTXrOP?7yFfF_8VLyn(GCK<h7DUjl+)q~AcjG^
z=Z^RcStk%fJ&G8n?py#|inp|rQNAaE;$l(aNCq~zgRC&tx?osaVkG(BPBge?z}B=>
zF6Dy%k-&F%^qqaQ8~uNTy$e_rSGGU8tEiTyG0<vb0+CKNN^FQG0-69q{+mW35`zu+
z9-~Gh_(~>%sKA(SgS4ViLB%Ktq`+3tn5a=_@;>Mod<-HQ;_I0F86`*z;~0bHn3?l^
z_qVFiJkGiIE~Ot;yLPSjUVH7et94|5J*G<VxeAuh%9!jlPLgHKV1WJWOz`yN5UB0|
zdHO;Yi6EIYpOr9Hu3-L3W0~ZDvzLaMt?%Caqpk{s)c*!VFxm_9zL3yE+xbmfC$oL~
z=If(6nRQnW#=S65CYaHPRfCXL*&={~3O-W$HCaas(o@N20wjssczh>t1JKP7??|ix
z*PBu9lFT3(ZazY)YfP~%xOsNhr=$q~Z`d-9h7ctb0u>S@8~Xk`fCBQQ``a&WH}9H-
zMHwLk8o+S)VG?!%f1rR2TIn{Rrw4a4lG6QzvP%!9<=)#d$ia_8-F3~xAjve6gpJ&S
z_b2~9dO*|?vRKXs!zj+v^drB1krIHd^zEB7M_p6HUPRymswn@2;n;1M`amEGbL;x-
zA)k%$Sb#vc@tGxHrZPkVKhV3Phz1i+FFUZ88O#PKugt|j!F)}cS%`sWcYSH?#w{Mw
z`&;fkI4lSW{1^WVcX5I#v$k!>5r$by_}i@<OG6|)#x&0xVJew89fALcLlpE8fF*tQ
zL#2ioj=l+j^AZ<(6^R8XM&Klz$RcRW?hxEz7S??~$4A_1L`xNah$`sy*PwX5K!k)P
z+zMg#J0S8%5lTeB%A|7g%ph?RP*KD6=!Tk8*_RGLUlTYFJwfb>e0hAP3wV~v#pVDd
zym69jQ4}U5(*dSwmcugQOb2K~tj5fD{6oFpg$pY<fSw2=1g(PJegkG=44RWNFoVi0
zHoX59=qkhgyo>9_@L~DozUfAwgd-YrsbVi@x=CTZQ#CM?rVGcxbfC}&^J^h0>dacX
zc`pj-a3hN}ne75w$YVQ!7AS!h;=`DRG3I9o;iVtY<s5%xPrnDCYAAs-!Euew(Tx1}
zKv&(b_`h+);eHEP3Rw=yGl)d$*ueVDCnKY8!VF>*yW+I#K?moEA4{~5qB3OO6S$6s
z&NV0lU|_MwxXtF`WclYq-}Rh$mktv$=!9N>1E5NAhU<#wVg3Ik2go3in7RC{YMauQ
zkB&f^f*b^6Dl#w}O2mO3rZWn(z$7JsSV?wz##y>oe5ZhIV)fm{FzLVnbb&|ZCon$U
z^F?K4vS#aVuy+6avPXBp!3SUw4?fX5XL6&d3_K^o2{ipxe{uJIe85`GJ}WInWs1gl
zouGCO0j)&@Q_i>sW+PeOhHoYzmvIl@6$8Ry^G64Ug=6R|*A>MhIt-ysGe%%;;!H8i
z$mjRpV|1{Bwx9ptVXx~Kj16ip9{L+-J-~`Yl2-VD_h8R&LBIF&ujih4)?&cpIT!^)
z2yp-^y(HO^;FDBYaAcrVRC?GFpmp4AnN*mNl>P-~p)&KTI!si9Jrr#KROSj$bxgtw
zJ^n>eY|Q(;3A;YcKY;V#!~Y6U2VYFlTugp=`T^b#4tUUg6X>!>+nlPM4I?4^dWJNH
z)e*A8$wLzjSUoctCvhP|=qX^Gut_Wpxl)AigfRg|eG_L?#v<^|jFke#`?Cr49%PwH
zcKW{nm|G{8HIR+MSUmpd_w3bh%In6B8ByG;?a$EKA~sM%?1iH`kUTMv6`^bk>`h>-
z*t;yjBn>QdYGDr8n_`~IhIzEZ!!&3O;^S5@8r3cwWGfuir#54P6YBekM;koM`>i7A
zcf+!rVmIm&0~&+@zvr%w{)nt;KhDOqc$mTDW1%>JS{fMR9H+G;mhMz-+>pRg`+-GD
z)OsC;hm*}_d^IsFM*a%w1l$C|1%d#k4zY90AX#XFTyyvNnjda9(s47r{0iK?n41z{
zXe7V~02F=$9=;r{&N?G3@6K^?oRo#jZ4XB+jH0Ug5T~-~rtsoACNOeX)?wnHX?oo;
zFqgQSsheAImGNP>)Caf<g-RCMK(Yi1bftf0eD_c5K57N$r_%c`kghvwy+2)h=!```
zk-;ng#RX;hC!z)=TGK8pzQlaztmU|e(_|_tOfdH^HiuGfaTor<aiV}w#cl+Y2bjuD
zb!;L-87hVe*??#$fXj{gBoi}zrl3EJW5Ae`5m+$sK?rG<oaQ2}@Shns|LDOPZr$|@
ziMR%(qqrNaOs3$UBTUg*>vam_02P1$bl$4?$~`*bWzF3Ji#vC7Sh&#BpcuJJFo#8L
zC2}^!z{JEvKnemNfZER#ZXSc$o*~d}!MV|r(JJ;bp_xgSrW=LF?kvnz0(Ahl;RJ<K
zF&G{FHH2rqih;3>I|cPe(H6ZUc_SU~%GlJ^<u}&J;>xV;$lRNyzpv2Fudubk2`pq8
zBXeWHbWH_k?4BVE{p;$B*S-}O#-A7QW9yD>LcJe3060}U!ZI^sgqq4siVugG%1m`w
zFe$QVFcX8ySQ?+29{0ETAdkAE5<@aUWFX*Qd^ya5qyS+EL8`scquSWN!c6EI0Po-?
zLNGIZ(ra4By%nR&?w{+~TjDh;KYEAC3{xk%!?qUG4?ftuCh=Ueb=@0b-+WjzxAxY7
z6T8}w=CwFs_WB%*ln6tHM?lQ$+1JP4>^Hk#S3(j*XqXymu)r2E_)*(UoE=umS-1(h
zW9g^kh+qRmBM%0(WJz*SiI7pN)Q?l`G!WHCXq5sE*)nk*23iYvDEX_xdiVLae^`uL
z%ghvFnMn6|2u_|@*xr@>Y*j+rsUT1kyeKkg+@OtX5|8za@7yd`q<#?}Sa|i`+|icn
zFHi8EdU9dZ<KycS4=}lqhagn|o25ZRcSm3^;7s-0GwayF?gP_}j5Q$gL$-u7UOP_(
zjaYF|1ThmB<3HmQl_7SgFL8Nbu~3*~cNs8_N)yL;M1|uJ3-TbKM7ET{F3DZ$gJ;1P
z2dD!RD@pU*nA!p~bN#A_?m;izkgLZZtow|61>{COUaCYjUAHx~EK<E!ZQkJ1+&RBb
z;<WJrt-<5d_I}j#*TSxwxX{F*5)ch_$Ehpa6#oC&+|uWoKiK1IeQUUZX`0HoK|wTN
zM<_tz#L(^#Mh$a61M~-@zg<-u*hG^^r>hzJ4tp7(bx<+bIHFooI&jS*@J0q|)(Tkz
zNLq?`Kqo1|q>b?m#ArY=#iE9}esDx{*CX{I!RvLJf)~XNLfz7kl>=<&*Q~&z#UQ`>
zEl02Jj+z>Ap>FKn6G7h}&y6m>Nx`dLkt$7e;YNs~0J<Y%QCU+Dr5I7hW``8v4w9x~
zC<Ez6djqCUz90|9!lrO=pYeeLn{FE5zla9aE{g*w<E!O3MGO-lZF&*uAQ(r-Xu#!8
zNwJge1&`|0z5ejDTRBbx5nQB;-^~Ukm{x4O+jTbhG>e#g=xE2cLF+TGAKm;HNv>#F
zv+7_+`||jTm(**Cr>-_Hjoz_wY|yO&K|~dy9}J`tvI?+i1h)iU&*zptE3-cjqwO=8
zI>2K2RI{VZY}!TDrJEZJs*9C7G;%12Cpge#>@bL;NR+3b2Niz=b;Tr9WTkXufwDAE
zn3-gBkblr^NmdLE>QkFK<0_LViD;b=F(C;T!eLH?$Q;&<Lw9N$vpx`g?{?e`es@#H
z4?(5gQ7^H=_Vx^CRJGrls9MvKc<af$cxy>sJhJ2#miSe&%e!({VaKij#6A)qVh1yw
z8ll6YI<K{S>SFopLYc{FvX{u|iW?CRw-`_-3-mD2vrY6#@=2m#nomW>5t!hF1T#&S
zxru+!2o?z^{3agP&scw#2xS5Tq=#{635Gb#52b~b(K7ka3tI|XkN7W<{3kMNYFerb
zeXF$zrm0(x-*hTQ6Q9npKv-ry+TP<gG~+8~7y>p5a8qS+)miUD=O%c2O>4iM8#l<b
zJlaw}rxmM5b_ESsZ!&3c@rQ@@+md~6#``RQg$P-Li<P;=X%3EApdACLK@plAC7C2}
z9IH*Hw|>b^He6*xG!7#hjy;W_7ln@|BZ_fAiATfH+6@JDn8~yXjx`Fn{c^MXVesJA
zn3{1f&pm1TM)MxrR^UA%Z}gxEL-TLv$5baY*C(%D*ycYZo)_;Q!|tNDjwQB+p1Gr}
z0t{6>(IV{j)schzZtNVMU8-eAKknx}cawDK<ig>KJq=cdizN6Z!X*0rz6)jcc^$<I
z3EW7iO_$A(0u`~7z0mBGD92ax7`PXG;}?zXtD5H1xML9t+$Q`7TH%kBn)1n>L%)zF
zmnjy(2DS9rS&<rWA%05o=|v4vWnA^f`u@QwIWb@LTXe1^{;Qbk(j~>0lRNYFF5Q^5
z?xx?XBj3iTy1I78*(DnfkDU$AAoYerOumrZ_D1}=8-vbhO^dv@j9Rv_KB{nEy2p({
z#2QdwZ1B4FrdotUZ<7)_%NbDB&5F*EKrE6p5z@l=Z^tUmUX|_joMA{x=*Fi@3gn>9
zl!HhYdU2XKQ&fYE$)TJ}Um|w%m34Bx)?WJi%Y$RwXI~!t)zbC5KNf*=4|dmk)pp;`
zbU#&lY4_Yy2gL3b|Ck(Y`Dw<^Bl$5^kF4nzoWR-1h4z>!+dCJLKgUKd4+de|dzOr@
zdL%Pr+W6I4Vtm2I<B>ssOV-^A8X$9{;mxb~!-M-(_W3oi$?%J%K(P>tgIi`UO(fB1
zR0$2Q<c1MJ#&9kvdICVjIoKvZT7a#U0w1`fuVo@&K~)@x`y-8g=)>~u%(-WlN1n*o
zu)K8aA=?LTd*;Ry2eRX>XT5`NWyV*RzTgNt8(KQ4Cibh+p<6Z-Y&_Bx_(;8dO}bUs
z#;@MIYE;3-KJDNh>!pu{b=+AJ8`!dX?st}x3sJ}D9kr*cWxoBvOa}w0gYfZ-yw;?#
zXBu`15P>YCK?OA)pn+|je5}FD$^{uLa%o|7Mupb|GfY43NcJ>@T9&y`rvirrN(6>G
z+9N1omWiYpmPx+M==vw21xiC5*0T}*0q4DUR<yG0_?p7zV8g4UD_TyZSCnRL4&HgW
z@w=HjW4dp=ymM<oT-BMOQEf}3gHIbXBHHR>J9F1}{+0R#mp$N)3*0b!choE6SMMzB
z>df6gBAQTg(@Cr`07rxeUghvH5y{fcSsbGiF`L8KfTspkmsG<ft_~ES4HIL4iYXCP
z*6Bm-7~Gvbol|2h36yEOa3PghAmUaq0fA226?9tMz`S3cu=CSl_{>D_{Ta(Mgmv_U
zTgG5tni4RfAL<;o`W45N-%LKeFie^daD4TuP2a>=si2`2!MBYrW^#?6cg7y+<W0Y!
zvsc6gpH;jMFh~Tx1Lapqc!#2FeDST=O)P9)kI9HqO{YGb6L(2AR8y*h3XC0CP-NKG
zf^Ly~13N%GOijkF28&?fW#~<{sgyMmbppZ;lN^m$+bhXy;HF%7%;N{CD_rh@HF8-i
z^zkUBl*zscs9Z?#H^iUVQh2q;VSV?M)HRSuFyzI{ma%2&;9|0r;6H|Gx6^Bq^P!Bn
zF;!>BIy{?$oZQ4QT?ak}zw~P@Zf?qdTSzLi*P;txDCo5p1ikCd6g1(BjY~-)DIKFV
zCMTHRVQR$;*g1k=-l9S=^mv{@kO-swRhE0<&9{HVjfnDIcy-jE(kwaXWR7>b=?ClC
zA@SA*CFS(eczMvyv_|v!oao7iHZ_GV->9LzJjLsyRPQ-s2b^o|VKG&wmqnNNERCyr
zWDKj{f?**m-s@m_P;=JY!iJvSA&OQY^!bG)Ci8mBr$rX)u3d<purC~$ID}fN{Xqn#
z^OkM`0OAG}c?|ljvJcNUGSP1w_BcY-NTmpzW1z4)BcKep!d%RPkG!+<)8jaY%X=a4
z?ASw>w+H!UZC<tOjd<(zb^nT+*1&Q`)SMYQcv@3@#rLnydDeV(`ua^*k5rBlm`bpY
zSKn)IS`=5~cQ8d7e<CML>sS`nb$i~kWpd5<%nt^g>64h&clDP?yG|}Zn4)trjMdK8
zXJ737?7ugCD&(Knr61)g_ER(zC2(C0NGpA?X+@8pCPKxOemGJKDh!w?8o3XoW{>9@
z{H#lb=L88wh%Z5iMC=LY5E&v$%MfI&jLiGBtge4X5QTzX2-LWW4QtHH_Wf{WU*<@7
zfyQMqJCY>-s->eUUOISk&AKlGA4!Uwe6l&?-0=%64*8kHh^NMNpI+7{<J0_Dv&9OR
zVQagGYTqjkb5BOt)!T_5t-sJWvoR-P$LU^4VqY{ikdyAP&JUODxAB4TGW<Fs;r<q1
zzIL_>Y)Bp$*pX(1j3?1|J5n^VvslG(lWo!~T-cIV9o8OTuBr-#P+dPTC{A~&W~c>X
z%rea!ih}3rzr3?^-$Q}w+?LAE&l^>hv@EJRC2K>U9B1T|oJ|d4P_ud^^SQD82Yo-|
z)lrr&{;>1Z(K#!gW&Yl)&cUz7O&OeUsby*Wpwi6wmvUk%rS_)R<AVFH{$Trt`lwUw
z=M!Vv?j#;8yObDs_zqla7p;;Iq+2`!_FMGh51`k&^W5yi7?bTqNFJpROv8dd%7l<`
z9^yg?g3BeOiOxb3oIfSl5gl^_s1s)F+o{(aUZ2lLk6<72F>Y9}5ucJ2-{hh_*pF^h
zaZWOYOU%b9%=NF`==$CxUg8Xey)3L|&o$VocS=3WEzR;u7@ynP@tWH2+M$}4R_D00
zM+bgCF>}-^;GY*~eQG@Q*4**wzFYI7ts9HO$}U?Ix;pcts!Nk1AD2q5aFd@m&mLXz
zh&cXQRQ329IhU9wa1D&klil$;*4~8hx?r8tV@sH_3iw=|sQ!h9w-FGazQ$}e$P!E#
z2@S;SlEZDI@Y#Bj``y<tWv3w@`}dK)GHYumt2AU9P01$gVAZYGHlODyY|YAtv{JBP
zP;nVRF(_dZhV0EBVnShW?D6;Ch^<K-d#E9*(toIYF{6!|c$;n(M3;AE#aC@u0z*wK
zba!B4E0wo#7H2(L)sr7<IlVBt>SzP<3t%_^W=G)fyDI38#bvE4=gbaN1Z+&T%@gEl
zG~kH&wPBOw>nPtdxvatdHWi6sTgO{tw>N}hP{BH<#71|P5Pclp3;GK?orJKIuC;MT
zqvt-*;$j;M`@eZV2Q_oFRq*Eln!*AkuF{)D&?`2SW}He{VVmG&Yl?y=u=={HmSR0{
zhx>4Qf)o_<^iEH_j{UoPQS$+db=@>5Rbkp}o5~Pm3A{kPB!khF6h2V65IY(r2vA5E
z42<%BO7m>SG>c_xJ7%L9yXRw)MB7y`f&hslszK#OU#^P=C$L}91<n`~Pb$CA0QvtG
zG^i(yv%O?EEtuJxxPCx?p~5kslmM=Ij{5t;lERI?b6XXK88)T3TsTjYPg**&s7{T`
zaCZja1P8(l#z|AxCaGv5AG3sNL%8v&wuU44t`|as-T*phhnm6XKx61`KnZKYvL*PZ
z>YsJSGl=Oo;MTuU4o+8gWc<LSUxSV_PFuhy9QS_;I74x#%{M6qY^A~>0}DV}UenpZ
zeW5?td2ntZ>Y=0r+x(=d{W$8_P-<(@ENZ{m#`qKxHu9YzG`~d1iExqh4kLgLBqOsB
zF#9(Tc{u!gKI_d4s;4<!COjX1`-%l}jw5mV&oq7dVzSh=tDGwRRHO!-OnunY3^WD4
zI#Cm<8YpZB)9x^vG$Fya!z{i>Y7WbbG_gZsLo%5i!gDD~Q7A_MGE@e44s4fsuvI-8
z;A!P~C124C5{Cjj`10#_=qf?n{<BS=8d!cQ*YxkmGkD(N4<CxIkh63}WBdTaUV7+<
z0l|Uf{{}Xbar7_P5&aU<A_0nItS=^TI)wBy`}6X)|J272cpv@XZ_q{a2cRP)9(?=d
zoyyw?{GVDCi_mG>O6E*HjLe}%w}TSGi;(RBarkuH6N`;5U{2P=lVN12BzCFl=td-a
zh$M5Eff<k#R}7jR*F*75A!K#b1*yT2|2ypgzL?Czr<kzd<9x^!W%%zf{zX^Ns|S$k
z75b+;A4csrqh10L7c3g-ZM(2wP=?51U3Q~NjZ|7WC}u)qcw!FZ=0*WWLb77=wae!%
zG||z}NTy;wvr$k*kQ|`EI||Y%Aw;pN4qtF#`67M1@|ap2_Rw`^Tuu$$zr_E=j8omR
zbpq`eEw*(Rd)Jw<KYooSQ9%NNVlEeBGqY;%CBnvWFAtD8d-!hrkI}I|gc%ImS_KVH
z$l#!bF`YZz!%{R3n}>QCWJFsa%YyVh<3_517pm>WnVXBa`P~6+do4`$#U5~>_iYx?
zk7S^a3sj+Cl?C^etqDfwf^U|2240;I$#3l3!tUG4qJ!6WmgUTR(%rF)t<KqV@;r;F
zNzRQk-=5!RZd%8hoiT2?Ds?O>v}d=nxEkZqc;-K<qX+&x+37AFgQw*nUn5z{@8qMR
zl5i)WYFFGI5c2PTe}CL9zgL<h=&K3T0N{IsRwNQ;gZCC}4=a-7BA%q_*H1*T6r}iL
zK)jO|Pakl#4r6dyIy+44Qt<QL3c`%q1de;K1XB?Tn{gcSgHNh_uDa6W6lEM^i5uo`
zi_M?qbG3N^;({~TI|fu-k^>)ka_$1Zp*azk>b)wC<kXao+nPHha>tY#L(elWo4aT3
zvj`;(QP!3<13ohKJ)(8wMh9=qTD7OWJ8Q$F?dL3Bh5lKG&NX&toI09cf8^S7pOamQ
zr&{TD0P?!MWVg3`n!<f=+TYgn<FHvuTN&t^f)C6Y0z#mh;VjgL`HbU2t*cF*R46Oc
zoi%ScxSJgs1FcS*C<`B4Ql#kyMAw@8!@_9w8OkDIn~dZ;{YdqdUy~iPE4%(kXZ3`k
za|0XhWykE-Dw%EL$76$jHb+*jZz>v;n)%m4<D;i@_oW?N^3FKs6>$|*S3DAGh^jf8
z8MUObCpY54Uu_p`-yF2G<;83-b<chFZT7e6bEZ7jI(KBv=~rJK*SAT_wsqWc0<#C8
zlWl}S_Ai-8`2{+TunMo-0CJHq0PfNPgOE^(VgU;*vP-^%1PNz_MwH<)Q+%UI&vjvA
zxzT`Pc94oql6cC4r4Skd=As)JV%DbZ&v@PAn{{7dKD{Wi+JDqHVeKso!y0>D9XYu~
z4tyyyYJo97qO!2@_M#YTYM)iM105@-zRnGimHsabUQm83p9L=3m-*fg&Gm-@7Domr
zFORg&P2T@x)VSsg`66m^$?n8)jc+yVS-VVWT5O2i_YwyN4?uUN^3#34uhLBU>!#9S
zp9I+n|A+crFhxets6f%{1WF}|2k>$5jG7B&U9yJ7P5H~KaH@);)M5lf97Bu8C9IbG
z6Ru3)oRPCbs1suO^Mba^9{;Xejy)pCZ>%G@uy5w(8vm^Rm8{QzSGHeD9JO%$i4Usw
zzOZFeSM&TCYY*JckJ`am%kucZM_$M|*yZ-FJXm?;L|J~sg+pPN?ktP6zQk_Mi8Z$t
zl=S`ToA}3CM8_v7+qcagpV{d5;((8U#RJgbI6;t3tDOIKiDrg3s^!daju}hrDg?ze
zO#RYL7LbVI4h{owHBbylxj;vekvWr?pU@=p9cE}yeYnXfg~%)gV>#AS;bc{FfQfQs
zFn!G|N}R(hk5pHdskkXaUrf@s6?A#UwC*y*1v76o@x!Re+t+xVYh4p{W_k39oWkxq
zxsjEPtYuj|a!LczQh2sDdaQX_<dhX>r>72`J7Sb&chtC(Hw%XJ{p!-e%FnDDBfpP-
zGW62KlM4+o*3#LyM=$6?@;Iz24b`qas$r^tE*9+XH__<Qu>J!bw!pI}n8OK2g9^_~
zdPySmff}<itZE6rGFA-18iwa1;uD*&xMIhtUHIHzQ30r$77XgH*ciK;`CeA08){9d
zxL*TR8^ST($&V|)Q^2s)n-_jRUc-|H%FTBQLRXZI-MM}L*ezOEdE1isnqogoT=j&0
zSFaU>ZOXNN8Tj?WrFx<XL!t)^v|<6kgA8ui&136IP=zQ@H(MNKwOFfF45}R{j0poR
z7`AX<L2iaXg>Jeb@aaC>YoDWEvW@SNO+8L04}l%Jup&&`P2~Qn0VIiXy&?uf55iNq
z@~7(jX7u$Nvfv%<L>Riqy>KHijNV)&Dehr3Yrm7PdXMVLIF>)X{Ejo-llZ0<4+$6@
zdt;b=SpG0PFa-*OjI1VrCB&fv<RZetk82c%D6AHDG9Gpo`EqGCqCp`yB50iip~&Vk
zD4XanEEJt!1AvYsW;$a)b%Ik^j=VeVt9^tU2>1n%jp2I4(u=+43~1^Fuu(BEfp5?e
zsDmJ$%llheK1xx5@+6kv*bHmHE@ucIE`eIgav=@tG98n~4TMrL4KcL4zdmh&Molkb
zz7yp!Dzogu+K+>!pa%n@KXicDi!{%P%$@^d-{t)olXa)a(eFd}1!+B&dw85_!t<GS
z_;`?Z;U+LS8pGvxo*Vt`Fm|x(gH~v!<N$Uu8NRL>@XJiG^qUMdF(KpQV`9bzX?9X*
zsN0~yI5?i<<-ddtcRI$Pr#tSR^umeYb2xaZU;votD#T7JC%VGDRK<oZ&zBEP`>KJH
z5&VYV$S8Vvi($}%tNuy`sDsl{W)bv>qH)5`4yOL;>Oad5NY=kW!#UPAZ3z^5;yCUH
zK6}cD0){}#o%pT+iS$rV4@JZ}M7mMvDu;<^N&@KwHo4H{R4D*0gY(9mms##+ewLM5
zizB%D7eJm1hNOSzo_qG5^Qg=(4GA&3s&f)-j{SC+6Uw2P9EfQ>{b5ZE_jKa<D%kCb
z<kF@EA+4}W#g`2Wvj}L+M6HZjrW(p+i==RL%(0v)<V`79vJeg6seUgb&=c?~HY4Z$
zlNVpeT{wo0bN!@eR!jR-baRAC7Gc)({|#=Pi^SpY%MWs>Mj`cu+t5>o07<YVJn~SP
z4vz=@oX(sHY~71oc~`c$XiEZGS&X>L?FK9y7^h>&896|;fwUN2lRDDvg>0A9g2hk!
z%d{`|&{>>%2#`M3j>wIGHe?KK$<Bx8{^tkM487NPM#o+cBcuhal&PdD(0dhk=FEpX
zA$KUVuKlQ#qJF4ghnbPfVVK=QbHO<Y*J;d`!#?$Iq)~)*5u0cuSAGUsq2d%1^g@6e
zxFH3exHjR^oljd%ovi0Mv#@zNpB2ZIo1Zz{lQV1F1mAze)s%+Xx`bPaM-Csct_kJ_
z2%{GQ^gxD=ow6-0AHKw(E_H$qBj0LlfUd}5Aa`R48^}mJjN_Dx+Utr3tl6DO7PQbG
z%m|vmeFG@L#0~2S30NiozxAmIE)>OzhY+m<y)faX*9Tx%yjKN&dh`$1BfovVa^CdH
z1Xq6b>2lec>YWqT>NRd{!Nt7iv%90~=e1WqdU`>0)tSt>wa0H|RCL{25?6in<h^{P
zt^8+GZ+iV6-a#SkR@Zm=o&OVic9deu@861r39do&O)3;7UDe2!(j{n&Yq$wlkOq)C
zhLk`s7B8TwyAFXT5>Z}dFJcR9LlB5v&=}X@80`}$mHKA`*z_wI(HCUq@*;vxE`2`l
z?{99-oL*V<@9Sxo?aVsfw`a?Zplz4eo%x}#An40)_dAAW2W<;l{CxgjyXw!(s;jyg
zT)OQE=b+|AJ;TmeOA-q&uM8M1sK?&aK7GO;P>0?8`}??7`HNwP@x>Qa=O`cnV8o!)
zx>G|LQ;K<*A%rR6M@<b9FS747v31jphb2|8yLN_XizyI0L%|HkwIGD{6cE5n$YzG`
zI8q%{2A)u>-+<SqGkK5uZ_UbEI@I;#fj@oKU312OB8JNUm^8xLGD`a{@X@t*w{&j4
zQrQwwyK?A}`?G%z`*GXNklL>Mk^lYbnpORJCR8r&Gho-J*Y{i=sdo$)MquLlo14S|
z3X5Ci6LnjYu`k6l++fd*?w)04luLwS35S<<F>@^smp7N-!E8<rPe=O7;<@|~XB5CZ
zw0ek-02u-r<Nh-3+#cOykh|pinckPY=!+}vH+^X+ShqZ8N5YRUjm<fShe1udyZ+s-
zqq9;|y;p_xG>=`B{p5f3WW?0InZN8WZ57|&9OJ$7`V+6N+0yyM)cbdL$3OA>$1g0r
z_sOemde0sT;jzG646`(5?H=@foM!sGPQbQoXe7xcW*?R?iOa_oOCj4^E|(dMhoNsp
z!dGFY1RYg_ruz?g_`5JVbOm`kbz12QRM8^zcXWrcAoA5q7cx$VNPQv7GWraV{NMS*
z;DE(()ulBH<6hXj-S5WSOU!&f>q12JyK8SnPsxeU{vPyRb9Kv-efM{N+`M$=g65x>
zw6DqOcxURGE#JN~^~B3NznUFi?O(Ifc!g6Uq_<~-P9gZ&OJSY7zoC{7Ns|XA(QV<7
zj3%rhzU4o(EPr_NkqOCD`@1at6_6?5hhbspj4J%00|3%K9h{!cQuypr^#jxnV`*HO
z+3Twvxz#~}sH0$t-ZepA02zgohdOH3pU61cRo{QghuzL}ui)3#Z^dlA>h&1wQ%4*A
z7Tlb<s_@dir4dthUhep3aOvACN963^zH;i{Cl7xk_QW6F%wDvC|IU5tOlK^-ecQ8(
zohhGdkeJZ@p2CDQrhaE@KY#vKyWCn7mds^~b2>6N3BX0eAJey1lZpaYto0#r!BAuj
zp?K(}<9gUkq*HC>-OR7EUeFj?5kST|n~~&3gsG1Sc(F~uUg-p>s~C^%p--laqY3`A
z+n0y6pP4>8dw4;~QmW=6$E164n7*!!NKT@QK4JSVA6zK^T$+^bR5X_?)1w-F5*)%X
zftSr1qJ(uJOgYS_q8xZ4cE$tI;c7V=k++bGBm)H+*c@4606u~tPF8(6x`m}dD-{Di
z651;K2|3Jj4NTP!F^^KBA$r0Qj8x!|OOGHt4JcmI`PADU*>{kS_BIYS&aFIKiCD)_
zj)(-#Z@3#X#87O$KjKu91S%VAnRyp-i2YbtLK4HKF`>fIVT5j+Xi!P3Oo}~(g>NPn
z>YP__7zjTA4o&rN|3(rB^Z<e~{lf#4vHM!j(e$3~RXfwcULoPI+jEM_+X&?|I|57x
zU$j#|A3jNX#i_9plhF?OMgCXA*2Ruhrxx$R!+tFNoEU}6Co-QgSP(<@o|K1rKtVrT
z8McrL`V3}BI!(kC0u=|0+~oqF^Zq&e0X3);(a*xS6OCA=6Plpl<hKxBrj>>L`cQAQ
z6Si^D(~F7G6AjL^00<rP1yd-DT*&-xT3V9HjK?EL9qWYMpw>CYq&X(Rz$puB4=J2z
zme8YF^J>LPr3x;F14oT+qS5_{uaG53LiW;zIWeN18=naxj6fx90VWK;c!N4XYc+j)
zrG6X-Jicb28$q0hNc{w}2x8Z}n`cz)!T@rVPa9*L<TB#9S;;CAXSkcx^?_@1T7r}i
zuGY?1jF^7!a{?bW$;pbaDkHb8vj<RD3Wp%@X4g|s3PT103WIkYG^m`B8#BJz@m$S4
zo&9uJpG1a{^1{nQuU$YIb3<KG+|&S|QAZ|D1bF}$129L0>bxJ5o#n8Mtvfm&)w@o%
z2Zh>36;JwsBFq^+a9ol>Nyc+UX-+zjE^_G@3QSQX8OBKUdFX|7gTRbG;+bcwH67s?
z`rtADQq%jJ6?6<ib}uhA24@9~Jyp)V{(^c$wF=Y&Zk^i033)Kqy;h%T`#wSp-+ci0
zHNp>KsRX!!0S7zw4p*H;W>b<z7Xf3S;!$LL@#TtK<R2-ht`m$g`j`-nJ4&3;2OYRh
zhJwAINC9~5`Yp!j3)9>(xrpk}fi86mb$S4>P{ejPkEs~fW$}>LcC;akp&GTwkRJxW
zTn%r_t>JVnOcT$yL`DN0K;|AE$}Sa(27=|2wF)x&Ttl*=2n(MMaX_Q1Fw5Y?6O7cv
zcG-xD)kLP-;Xh{UBG6kgG{P<5Zv^B-Ox87oTo8_vcTy0#!4Z<)0BDObRjGIcI<N9?
zJ1o8f!wbnPPBPlXRNvVY=-K6Xmb7NG_4M0duG9U%dJJT2gbaEm7k3vQ_98mK0gRF0
z4uclz+hkSr4_3>ZCLTY*1m6>Mwii^gJe|27!3ocCa09xHKk|X7$N`r6A<wYnaaVHZ
zHY;35`e}5*hogieA*l%I6%!s|=3QrBv+7*eRI@=v4X)FNyHo|pVdV#10-rQT9lh^$
zaaF>IiG?f*476cX5i$gl_vi>E7w}A#G8Vt^wF~)|m`d05f=c|vg`f-|^jD}EkR;t;
zsvQl421h~s#c~ZDB$hDt^n&cnFTC2OzRPi)h1ndYflS7>94-`RopnOeaD_Y20a(as
zashu0KTd+0NL|sE9nGwUq97Vi?!$M94iaekD?F&VSjDjZ-#2!Fs9)eg@J&asc8ynX
zhH&(RL}bdO$fOk11Qts`i~z!&Ux>Tbx&|!fQAWpz(7{q(nB>@lA9R7YBV?V9LPt@N
zhCT2uB&%>9n}2(Y31c~iZTvBYhc9Br0}p__7JUc6HK1AIDM_9F6L5a7ywG#}f4pLf
zM7pP6hj)z{ILg3Z6g+V72Ol}c9y-hnprdn}KCY~T4!aGZ#A#s7IB?aNDNJ6cXfivZ
zno5Wh0HYIDA~#@38Q1!!zlE9-qth&kSdoAY*wG_6<Uv}6QlKLnY4bmB>?$vzbN?II
zI2k#eQ9K$;0}=QReOD+TK{Zw~o%L`zFFW&o+KZMCif|WMZ{}k1I($Gk36d@k$P-v<
za;s)x<t`O;DjW?~wH=@f(p5Zafq^_dKdWgB9u71K<Tw!^sc}VSuid-<J8PE^XzUOY
z9_L3j&*;8aH^^||_#e<oJR<<|6lwI1>2Ou_ovc+iz_bZzT`4dwl-V;un?GPRkiUMJ
zJe>_pG6W<z@ZFH6Or#5ekp8y0wI7H_?j5)R_j0Jbx^^Od3Bq7z0~^o;M=7i`W_+{z
z>#SYHq><4zEiV*VQK*PDt`E$3^i$XErE%7E%)7{ZFXQyTlJji;_*1>6XU1XdHqvhb
z<csJH1s23_C<JZ#rz0Nkrw=l|l{2P&KDp?NKfnx<s4hS=?=g+zGS>S9VIRI38!F@w
z3j#vnq-!4nhTbvIB_>efq3Og?1sN4@c(eu$g`l->uWxq$g9rhGTb4%!rM&vGPj}PD
zrd8*}&qGE$!J8LEKU035^~o;3mml$T$LfsQGb^hzuG}8&CM$)VLIC=_PPTXi1K(pK
zdSFm^1GvJBpKoST$1ZwQ0Osu=3FGE>X(^~xL%}-q|1f$A#bO-Mj|f~g%Ed@q;7TX~
z16d}43P#c;{8+?Z8Do%cs)wPEXatAkD8Wc**AtU80j61e?84VC<j)W|9Z~6*HEB=9
z%Af(^_eX}>s+ZOu>U@G<TM#vDl*Te+_aClbRr%$W?3aBiRtChrpEw7XP`;KUE7qH(
z;y8qZpb3TQ5MTm*^&VWI?Vi`QESW`>-^`D$c?lYhCu7~%HgNIOkgaJP>Ir!Tj7gtl
zb710K2KJ)hTNwtV^%{(UFk(;zql5UJ3I<4+(B0ui)9{)(WdYox;vP;g#0HJi-4;5)
zA##g)W)BZJ*6EY_)_;<H>Z3W?vz|PDI&ap3b9qCPIxD9{$O~pPp05#o63=yazY+Gu
z()t0(QO^|LeB(KELO^2Tpb9squeu8Wn(kg7>*u|-V}5E=&bg~%(8Y?c4S0CzYz<r{
z0{^%*DeOBuuRtT7nD&@hKReQGSgS(-4?nCLWW~-ygk7<$GjO2GQ*Z_z47RUm#`1t9
zu7ot2O#KDZsbodi-)nXw2E{rE<d3N9p?#f7hHLT)d~Z%>d{t^zYh>l4H+Fy9T)l1c
zYctk<JLk2CYTxOz@x<cK2AnIo_0fh?)wx5W%DUz~FNh?784618<fQ=|f0WGcjfz_G
zrZ1b26?OC~TQSqSyme`x#Hk2UEbOCiPwe<MX;^-L0WSa@)dNly*8)92=wWb94#Du}
zB8e~!Rooa^>0^pSG1OFfof%VmfW$eRoq|3HU@(Z@%x13(@ONLIO17&Y%{@rc+A~h=
zdNcdA=xP*e+uZkeUq3VT#vfE^d)H^?xv@{@EGfTKWy+tqd+<WN8>pwF$m6acK^dP^
z&v6Ff;Vx^p<^`pGhwWNy(8bF6GxjuKhZA9!Rx=~-^MuJKYCU=V!!}-KaO-g|kko<8
zRDh!5H!V1f52fh{Ls3d$gtGX*fl41U)e0!&k8uTPl9-)nlb}RX4}&^UY{TfY=aXsQ
zWvA2Z<=T??N>mA)I=OmPVQa_onBehEI|HxQ9~!k#^vPM<eLic{ApgusH}g^(Z_XP<
z80dxvsnKTu!fIV;Zl($P&R9zxp4Xh86J43m@l5&c<ueu>AmV_LVNmE7J38!lEgv*G
zjVauWL^O!9y_)8LkHMqBSidAl>Hv)p26#R=N1B~#_$y3?hPA=A6~wQC25+`@qJA@d
z;tLhmbcC=^nfvGOI8w7MG*n?;NB!rqJBgPRTi2Q`=aMr{`Hu<t7q`jBdlJvFkn?VR
z?#rkLw;inX0LcpX#`uN^9`^yv8hSQ$QqRsk9UaTV#1+)n-mW}ac0nH+(%;zTm-{Wh
zYk83~QxX+})G8h)py7E5RB)?FbDZJ58<Kn`>q?f+Xe<ZZHB&LaYF5c%uu$n0?zvV!
zNiVXD8IeASodO5^ve+Uey6Ta!2NGl2FU*VEAz}g_{kS0N>GE!ly;gc~fP^ggMGl{~
zp7eG^yirL$G^!KG_v7lh(TSpVx!gK|6U}(`(T$moFWb`W3|e79ppVPVWDpIAViNF>
zh;<U0g$)$Mo(BML!}_<$Md==m^L26=o5XP*0)$k6lOBDLx6lQ%@y`Q8RVEKQ+eakE
z{C(wa;lF*^`#&_Vt}C*R@7!|w_DjXoGw$QjSq@Y04V~yXQ2}>r@Yp!o>GIJ>HYPRA
zh`G4cayP&kX2RCeFjhkCjTSucsfeXv6C;9Ol@4K1CXLV&a2a@`7{aR!Bx{nhKf_)+
z>Kw8k&h^H4(cpO*b6^<$Y0rO5{cg5^&Tc>NOf0livWS{RufAb(YlF@P)J1whm}qWb
zf>;lxBETf;cmTvKm;pH`xD{r=AEG1RfJa8|`fGJw+MKNpe74g8=#D^NE>y&nc|4PT
z6fbb$`X%&{1SbmtO9!(xyx>t%KmvC_6C4siTZr%W{KpDB@0YRF8B>+0szskJXHD7f
zZK~>8?tSQJ)%m>m$Ijr|WQhaoBJ>CVnh$`yrcMts#B?_!aKj&RKe~R=GqcjhHuOy6
zq%Z)B(C-cmRA3^21u?^7)YpOajjlM(@N;M!u1gS>xSe3)u#4D7^L;oeh6xM?c%$T3
zx#OpeYjI54ex613|Lnx<jKFJu-hORI-xH4Q+tU5UCUTqv?GHfw@)3t*@-V`}9D-}$
z3gJcyfM+id&%XfA=boDf>!%x%-$C$(cU51K1>05ER9sIpQ>~mSuq^C85oqM<mw;wW
z(O6O~)q+gNERB=e$YvY{RUl1O8XU~8@}0yLT>=Mjzm0BIzpB-iY|8WAd9)i`QXb$9
zIZ7VxQ9}x8Fxl4Nv^is3GdU2_2s(p#Jzk}j8PhIq9XM<fZq_>;(uG0yhPu>_LFJzT
zL_<$GXut)e4um+hAjQ^OBHg5@I1BD+*!P7{I|PYt$(U~XF7qGP>o}c3OJWs)$&=Ef
zk1oA8`Xoas2#iRD>BnhfCL5w3(;0z>=fXpFBGjKOdVA9Isd@_omQq=N9OoL!wXwLX
zAJt`mr5_r&WDPSW;6NTtw%|PNqSj@FrdUQW1K0n)UPoG^*!+`h{$n>=+^St#m?XR7
z@%uw=*q54?TyCeDBo($tkCUXkDCh>jF_GdJ%*<4Mg}FS;07nff(-1wzEXkTV&#P^c
z$Nu!_3%qinA#JsVUO>BG3CJ$eHy!lNaG0x#VwKbh5~lecJ*`whr@d?^%sGE**N&e`
z=i?|#91;Bl4}3{ISrGSl-!*l?8-3f`YSypWdHLM9)<J$J>IS7IjVjYt&$~T8u738X
zscQxtJ<zda0yeQWcP)>p#uGhzvSY089$sNf)$m&wJ^8i{Ysu%EEyc7%N(3fAf<i-7
zD~!S-Cs|a+Cz(TkA<=S|YEg+1y(Y{cO_}wAfu0n>K?JE^QIUXHnK<x+-K63t)g;LW
zd@#85r%Qr+xQ1k*!bJ-X`N&?o>s!Bx*C0Bn?~8|9swccKc#n2+akTYf%;BK{x0gjd
zxwU1HPh)UlRrlwcw?EQ}t&d)te|*Weq^EM-hovtZD(q@`ZTyM3k=-M9G-n?7iu|u<
zhpr0Ly622CcRBmkZ?49ZOn3Bs_h3}@l)&!C!C#<Fl2r{@Rh32Aplxd#ZdXjmTy@D1
zv322s8QzB+OVgcbkh5)I?zQRZf8+Q}pvxH9dPZ8u1ZY(iR8@FRXqtYMShBg%2*n&o
z^B8^fkr^jvh#P}5Bus8JVmrildwQNLogaqdiT$RW`Y5jI(bJ1YRBjxb+q7uH*wB`m
zi?wa17fzt<SAiS9|3{y}39S~dqPFh5$e_{}OLn()ewv!uaVM#H@%R^yL_FQ*6>|Ss
zt!Y90JMnK8F?RB^FKb<M{uuXkPQ_=DH5V)PonFXsBim#c3rb5QM<WJ1CGMU0SG0ST
z{(a_MDf3$U^_%(g!+#pF`>Rm2^p3$$KMz0B37e)XS*`x2lAG7eh=x!a{K|Y&2uvj{
zS%a;xV)zM6Rwcg9WI`S<T##pRflVAIsTT;GK>w;M_j}~F2fCqAeKU`Au1L<<wD0)R
z*x-w%+*$LY>vLNcji}!8QG0Mgd24|ZwSUWW7Cn5lSI~+y{}--TG~RyQ=g~kS{MaXD
zi6MWTj`GF{nF068FYvzaUXI;9a7V_m8*z_siTqb|<x$KUG_%p-EP`F5j*SF1X7!}i
zTP|nF-;Ex^td9onv3>Q%?mGhLtstSNO-NsdxJ;KUC~8ogVYD9}|NFrY>CO&Q3MpNg
z>cB)bUH|bO(ip$l%%oQy3m3i-K*jaM5iR3#D2+xLKjTyH!R6m`so-104r6A#b@<;O
zPxU!)a?y;{2bP2^kEm?!)ZR$D99Gs^U^`Xt<h<^h#^_B2^{ZlPS3Z{VdR$=Xlh;r8
zUG!1J(|_&v;m|0#XXMbF@SjrGoPF;=JnQ@JSEHYZU+|Gf_WKWQx^#SHWGE?TZx{|l
zol^gbZDH>7*-?+5XY#+VcLhXlUGaeaKIDqiTR|+Gm|*08k-<qu=^_adnhnFnhBY(P
zTIYlTaz?t4h~wq`?5K8OHW)cJMBrKgCX)1qfe+tp!-@>=ANh|J_hj(Ka_lqb(&);T
zv8zrsoLo3!|EoP25m#?77!kB_&8njB;-8#*x%FPw=x3DHeAfBtBQL!A?xmhN@s$(2
zqhfme#&w<kWPIkA(uYsZt@|?ONaXv)J+q$^3{K<9-06!a_%?RFKK|YA&PC6s-AxuI
z4NqXewpfS&C?k+g&Dwem&om$Tb109i*%t9gvSEhLr#O@By%r8V6yA<drpC6}P+5to
z_@?LSsO-RuFCK@F4*-M84kjBhJHgPHK+K>F99>KUiQXJv(`75S#9sE0&ithXU3Y3q
z9M+OYPss%_HKp^SE%!1u{bQlO_m(~W>=yGazvo;UAGon|ZpDWkIpGT<8Xh0eSzV~r
z?>fD3{NKKge{5Yw8jq@)a(?-&p;5;c&g9o#j|lFYb-`#$KCB2&{9?rI5s3AF>|%^|
zt73{{X>7&1`OnW=_;!8H+Ee*KM22*2M&QMYSGqnpCpiSrR+tdL91^3SEQLvXK<j<S
zn?7REqjzA_So45CP?$!80*s$XhL#|&tb##C2TOc^_kUlf)VAdBJ?dU{GJe8m%I1vA
z-{qO>4>gV+y5;DEG4U+1)f>BJ<wklQky9QRI(4Fq0OT!XF*Ix^NElmIbHQpVh+_qG
zFCq<?K|3**k<5?U_T8zvc&_VSO+weV3w_!;^5aPrVaA9kre?I&5s?5WBQ%%1&qJUo
z0yU+--V~S2SxSH4fFX>+S+wpQkkQ2Gq;NY7o=fukc*os$CNr#jn_HGf?J#|q@yhtj
zt?iW;gO>aH`#f6PdU`>KqdNJG(bkeoTgMd^(t&-5etjC}_1Rf_@@C@NOYKX&$FzLh
z-?C;^^s|8}841T-qs^V4VdHB3>^e*<*WyQ%7?wRS!4d2*CNnm!@LAp;RF#IcY4V`b
zv56HOUHHXMxaqZTF3bHpxhqfs)_c^F8$tw`ErJh{B)Wnhg0#pS5gwA57{+8xzXG2H
zS+<5(G()mWC1UZM?L3Bz?mx1=!U@?ca9?w!+!<e=+q7f+jY-X~*&2gUP`p;#(-?d*
zbELI!{iUqfn#P#@t3E2Zhf3>IS8VwcBfeRRs+#2cO+Uu$ihTOG*gdMPYC`6uD-|2Y
zZm}JBtGOOyTQD|CGa11)H*uhEW)@1br^}nds#jE>&K#|@O5^J@Hg`o<Z(QKUae-%7
zF3Y|CY)xr;E9jxP4`3&~Bzf7m9vVWAHGHQ(UB2|Ly3ig?dI4dM@aU_j87uLbnSgxb
zA1|!L2*K15H7(xfV$2<tXRAK;fAX(4f=U;3G#~C>78%^KwmEp+t0S#PHxvXrO@1%>
zomdrfwI?&y>)?c;gEt-cuM-!ZkGS_`|E{>o{+7q4<mcJeF5f+OPwVkdGKx0d@GAUp
z_1rWt*@r8*4fe5_j56geG@-7ZcPDXL*J`gWw;`sc`Nhj2U6D_IP=*z6wgOMZU2!JZ
zSC7tN)zDqbBx9ON3~)lR;ex3zr8m?ET^bGu89!e&LfMPC3_~ZF!vNgS5ya`gfKhDq
z<XMSsiACa(9;+H-cK;aO6Ca%6b8x}&`niGcMLje3%8Iboz|<Qb7&@z7N@5+Kr*_S#
z4|&|bPtN6rmZg#J@cY^mNm1|TbcM3Vwq^4_ebHJqrAs^hNqXRv%WX{yV^(o!b112t
zv_NJC)nrHX1kq>Cs6QY6sMOG1#&Tw4wT@qO_1HV($-54^cmuKZK-B8#S_!3uH4;G8
z1=2#CK7$bg(iacbpAc~QkI*QnqH&0gK}#n#JCWpjfp!k!G5(l#?Knb4%((01qPU@<
zca}v}X2iUg9rR_t!-JAPTfcT+`^WvON}u|=PsT^xgC6}jE54dX@89Cco;5Tv>_B_Q
zfk;cs$|#@K)$Vsot_W$vvih>wFE2Q{Og;9=X2a@q>*)p1K~gym=eP%aHxX8h4Rtl;
zjv8vX_BskU{(U{WAf{i=4m@B1w!HQJ{fmJ~rP(D2D5xbwrO_M5$OAc3i6^KfSEW-6
zi2_K}H}FUmOlVXC=0~{1U<@g9>nA--q6vu1w;DM}zWWYEYaqEGWd6)?SMmFCTOzdM
zxwh`+&r&M;zIZvly!o>&wa2Y3O}SlFFAdfFpQ^f5k-BEiGf{g^c4tIr*OuZLSPS;;
z9(74bYnSzlu9}iPdzMyq?2|3)zN}rp`kIVIZDukvB*NiV>>=wo1K7&$tayI&pYC2Y
zmEF#bvfnE2x8(8jB;$T>zMk8Xc9ZH#b<|03DGiBJH0z=O8vg{TE2fV$_ojTH!G;Dm
zP|?F#8p>(}gdH%Ba#=_@qa`$i>xRKtv8vDt!pacAScvg(gKe_9rVjJw+J%WH?Y5M>
zxbmJ!oi(ZL?2WMGg82PjwcERv#s#-z?c9tdl!V4x%sVP}%e1Yz6&rj`Hhto^TU{Rg
zIEr4-znw!Q8noF>0|n5pI=*=16LIJ9D700Z+(*Z)+`Y5#(Y?7iF<-g{>&-c1eG92q
z{g-ZVm5LKY4YXHZj-bJm8`FyzWDT>xVZDVq4D!`mdmJ2n0W&b~C%Fquq_CHb#oq*p
zbplSk@An<|<|lu;@zl53^@q+~YH{YqY%Q=3i1=#d(7Kw!_8w;*)!Z^=YJNz%ongSm
z6Vo)XHpw0C|5O)Mr_k?nLVm1}s}EhLAsRmr=fGnKonx>eQ{pu0jJlSFrw40T{VJ0Q
z835(a;e7?JwnNul<dk5mkPh6jM~lUGv`JE@jjNPpaG<s07?#thod?XtDwaOKzH!BX
z?`2J&^j&@UEper0TjZjM%JrQhTYBZI$fw;spZ9;}RZNV&vsLwEzq}whJ#ez+R>qaA
znb;0<sWUrlUpxM;E#o0S#?}vx{CTALc3$d@1ICa1@VJ!}Oo=yU?W{O?GuL30UU6{5
zDmtQsK;IKN=hk;Eja4ljnKM_Hw||~-?y3gWf)wA`u%bM_2&Se-#LNKl$s+N>;gaBt
zR2p+J6Q=o}st-h@s`g=Mp(`Z}5gK3`C>I4Jn7Dq_^CEsj^ZoA|n*W$OcI(i2CwDv=
zrZra{tsZ;mbZ*r9?)rXzT+k=u<9{U-U&UsZb$*LZp3m}5yj&hsI+QJpu{6IH8Pxnj
zn^)xzqtxclTbeRH%4nUMQPJLmJ@uty2Ui@pmJ#vc$wffTrRYo2y(mZ}Ez6#@`hB-G
zSM{<rbQDBa_HEiMJd%Dr_xi~Ea1Av$c%}-CMLJ+!!x4Zv1jOzFA2{dW2lNr@LrTE7
z)9H(rROP8ccheHEH3A=WW5Mj-16UvOU-sHJD>Yo=(|2fY^XpO7sbPEGTyV9kvafgc
zs-pX&g9ZfkyyRW<%+kP(fBQ$D+}q1jiw!Z)M5QD~-c9+)|An@<)<2%S%JfgD#tS`O
zv3H&gjoN=`PsyeJE?2ardv<)~(e;m~D$$Rh9jamK=kX4ndz)T~3wlY4dEe_hO|~3g
z95uP&#2-{P91CD8X6?XWYhBlX<OTh*2HKN?b66ax!A4}J;xp`q072RVi07U*6#W8(
zh9shl-T2Iokr&Ki$W>8>m~@-eCNX63OeMyvum`_*bbEee(9y=8&(>$UPk-E+6LK$M
zd-vz9{!fm&Qtnmx<H#X1Zfve-X*#|-v}bWhhUr#hP~XhktK^?zhbG4UJo>TZWh09$
z?S7qJmm7NKgdOO}iw<h;tP@bTUqM*aG5q@3{m#QTvFKTN_=}m%_2<`_5*&prN7w9v
zVH&&{<<4|uA<S@^`3M{!slmAdR189P`as9zM)Kh}sK0_d_-@YdiyrhHk(oqKEmz-v
zd*vfp9)bfK+LxwgZaujyHFHb951STxo0mmAeZ2YO*6fgbDOvIRC*r|kUYA?Fj-Fcj
z<h-7O)VIX$_U&&!UTUB5!jB>IXRK~7yZ^V;PIt@a8=D5qx7$+LtvO+MKp6~uMf8r+
znw6AwhUe)83?t#0GrbSZS(5fOV|a{NsbODcmt5u?NU_D-ne*y*aY&Z574j8An8ieP
zfN7du2}LybjwcYG4kT;;H|Qe*D=-(TPHix)6bzkE7LsFV!0cBKBQX*))?9!|$!A+C
zx;mCeRrbwFw>H;fT6^qc^U{d_+?`rKcwN_oEVkr1Va%L)sv^JIZ|&9vUG=Lf8t!C-
zAs;YeM`8QzWsy4u6n%M6T*|kz_QXBqUAbY69j`@~H^2UT%smdI+=Pb*<l_3XGshn4
zdhC&gcsxcUCpdrOjYB7+g1%7fipfz$WQXo5H{my-@Ni)eO}-oBO)rp$DHq7$?l>JY
zaXnp>z&(%hRp}TkUZ&TH0>7n&d=iE(s6vs~DO$e4PbS|J;E4H7Mr*Pu$rbJ&s$eh(
zqjB6ariXdE_831e9d{~!P(s_CoUp6c^7?GCCA8KbD!htq<lAo-jI@^c_sKZe7Fe<%
zZq-ruwG~)7(Qik$1lT#V_I+z#&z#trFAsKf%#WN>a&uXnT#Z1GO)C65e)#0tbUaZ>
zZ%bj43K=kiheW*}az1cUa91?1g4hUR$LMeL;`RYGp;8Do?qF?(!_~v;sq7LzENJ$^
zd=|P4rX~8c-B}!0F<`;BORD|HE<Lt1?NW>D!kA;9ZCo>E%?DK%ub6UXdMAF&yhmA2
z^R2IRSdW_e54WY;15_xCi@BJ<z3@TTA-XR-g1`zUlKz;8=V3(`EAwVnKkEK}U94DX
z-9>&?XS_<$0Ys+*B&X{k2@YmR=IGL`&=AXe?b1{hAW255%EqvzkrUEyN%Jk5pi|9`
zhoHa%Cl|*(_Q;8y;YdTdPrl~OKJ)8SeEQ>~XX5t1I_S~bbA6I-&3$gDCSMy}wIOR&
z*x{y(%dK5=Mp>I*_%h;ZQ$c*t_*1)Pvq8-_c2141E-YKamxnd(o_o&SIX@~0N<~Ru
zfZgLRr1sMv`!z)EFTNm+54aT{w0=$0K@DVxurqI5pE>Vtz!;J!{geU7)-NQyLAyn0
z%tJnO*O_i|n;~KW!!t3WSwUW1L0OgPois=ibwdWOgAI@6#99|ADb~RomqtG^zCH1j
z*7$nllYf<O#Sb;KXY9GuyeK>yGzVVXbFPo!YMxKG*XbWdJ~^*#u5I73*ZU;C^5cdl
zS6zQ+_I>ZR$j3`Ed_H83ibpbYYP%a>j~nz-=HPQ~q^YV)=OK^v<f==Gfw@#@3=sFA
zFx&K;cI>l^WW$KZ3;Q-53fu>(h-x^bp8_}(Q_%_TC0f6x*QbGOI$EyU5uK?92;bzN
zW<){)MK)nhrqSa@ioi3o<}3oQRLp?ThQm=6t4KG0LR2%S`S{{UEa%S+JMw&X&&c46
z8$o`>v(%17&o8-pzU<Ctt(h~|jIs9pzPqi=>*(?Gj_r>gY->(0Z_SK*y!7q)Bi@;K
zdQrS}i|Llib`Mx$dc%9mLGd$dT2Pv4>dqf+ZB7c)dT0rXl+jqOV~?L~Gk)B%c2j=+
z;Hie^=M`*h+G6Xv_D8G^$|7gRV{O~o4x^obln6ZIsy+pPO~7qvf(}E?>jV;#$B7`j
zAua`O+Z>9LVZ9m*4K!-S>jbm{w+E=U4xuM7px(@$XxrD3M}uufL~YLF0|E~Z8dyDk
zwY%lxqU@>561L`7oAct|pZh@pyZz})VWvOby_I9Tbb96Bn7%h!N7VQ=ZSlF<wQT%b
z`^@gJlK(V4wqf<)B^}q0hPjnBdXQh|PAOsEJ^h{r5Z;UXpo`L_OWQA|RgV&Q>R(iF
z-`zVSh)}ueVxLi+v$BQ01W(1QM7fzRha(mA4nT0l$p_>29KjomNvQcpV4>(InMFZK
z3^h0J<tiKn0p;SP4&jQwsgJirv($EH!T8R<#{MgM_aUDnZ|1ohVxGx~*>`Ml{LoR?
z7RCqqt(mhs;7V5PP^Z7|_Dgseq5s&yd-gRii?;e@UJq;N%#C;_!8WDek&sr$_D61r
zKmAyH@@FOXv2}~$pNMW<8f(EOqZ>~+HU0JyJdZ`tN9k0y*OJv|UpzKpFJpgu*}tE=
zX)zs#FT2#X?P{Gs<d`m?%Q6nq8m3D(XewdiBu7kYa2C@LT26$RAyS1Q9j2m?H5r%S
zGe@&JG%?^Cc?goWUdyLDGnQiM#4`tv71)dgan_cn%(>C~58uncPYQmP(f{Ys<;{<T
zeVbGK4eMj*$v80jv2_b4Jk@yTwYb2R%pW6Ko94$?ukW+vT(`VE_1x(rkDcj^8drQ_
zUTn~Wth1xdKLs!9*MnU{`aPMs0l0f$x8dx}j-QL;20b!%z|n@rWzkP}G%byJT;QxX
z=srIyMSx(WplNW$u3FDk5H3j2>vR}IFPWg*Fj*)OKyK7OVCxipgF95gU#JR&Ax37#
z^I>(vra~PuY;$|gd$YX?PhKf9+{sV=$nVCh=G$+)9G2=^i#yQ$4!FZ4=K*=u(e682
z^NNBu@_IbdH+AUFgC}q1#12ZuGh6BhUp`)+ajLa*Dcup1A^`#64-hf=l5FnD^RB}K
z1P@hoe*6U;tPeQ4VcKGi^bm*<CG<&G*<Bz5;gf>Yj0O=QMu3;2YcP99mQ_-wbMj22
z=D3d0Vv(SC>IFd0b$V#LeeZVPUd~!pj12B?ozjV4YU%ey9`Xk*D`t6*d8^|JmR4Z#
z(ikMTLhw*hd6K|dxEXwUC>-ih(918NyBWD^+l}kvR;GZXnv9>$v}hg*D2UWFJYNQr
zYv7oWz<V#W;N6XXLN2t2s!&ZDZ~&<OLx5!<RfNN0C@j$!PnT=4ZHVUPIKsAW?=E1?
z*`tEf`@Q+fk<pu$>~5FZo0HDJ`kdRq1a54q!=y^!Sza!c=4W6$-K`glG$;QZxNzU>
z8PVtefDIl{2zwaNG|RXz=^0`K#P5ofg0zSp{x}3xTLMrmYSA><l)gBL%`TYJpb(C%
zvK+e+(LynRC<&h+1v6@kJVxao(AdWN&Vvy8tq44;RbYK`ZSuUAeYzSypWucJ0CEZT
zp_;HRidg~_!Nig1q~jjo4?g_r9TMm>a_caLpX)W5P|Aovi|s<d2uxA#HP9uqVjDf1
zU!`$@!f}fy_{3>hC|K2lJn%f62LS8nbqJV&g9c+BC4i99CH^KpvUFrkNng=gjT~~?
zqOPW>X_up_w^eMvI-sKb8?IlAja9gpMbG|&rKDI7K){avCz!G^d78HPfJN-Wh=*q#
zrQ;g#PPSOY9fzeSPX#mYk@->ig9^HI#$uvGI%W}@g^_A1W&Q37J3@oM&PxpGl8NIw
zF6nhqci)+hFVPpb?CBQmD@N99elKQR>FjeT7SEV|wD#+Wr+3!jJnRQW*C0Xj+zYuX
z)IjT<(@C?Y(%bLAv}DF(SN;H{fP0}^LU*YLOR-JrvY01@B`2@J0u8tfXG{vOL0QOP
zJw~=!pkEne;Dvjha;z|@A<;~<qN%_7P>T$R=+af)Vr^K!0e)yC!{JkVk^y@84aPLX
z_#c4smcXyixFKtSO9Vzfj;X5()XWxX<~Zg|FWVK0LBn$)-Y6jh>PZb0jgz}k;d>wy
zhmi4{F&vSL9KsFyNY`YBUkr}Fsp}0HH%Bv=HOd-+FA|YX3;AsY{gSvisTyN0dI5j{
zv*%Oqt7r02A9{4d&u3^E3V`iMWQ_t_{$%r={CF$;EDFkmeuYNqC`9Mz3iqb_4>_#D
zo8n3?q40<cHc?(@oQUxR4DSWAHQA__HJ0Fj5@mhrqf2YN5;{&|D%Ff%+4W;>n^x@l
zf4KS<sHl!??K=IS+ZLhG#!(b<T6}a9oht$g3ar!JAT3d8LE|GY6$AK)5fLLA?>f8$
z8I)Eq0}6OR1T-c-LUJ{enE`_c4q|H<ANffxqXY=yW(*SZn|1%S{%@Z)$=q9s&FRy1
z>ePPk+O?}HD&TIwQME6RTsLfqD}hP%ifrOb0tK)8Md5%J0^RtvV7v<qK~@)h7fM}{
zi$NrgfNvR9U7}K(Es)-LM7J<)=bmg+$h9C%7#Fjs8-gJltMsQKBMBSP=dVcw0b*(F
zS*KRH+ruYv069S6=5^J%#h(wHv={s>`pD7)ANe%3Z8f(()@}**i}y-#R!*!s`fsJ?
z_MbIU{-hF(4bx7$5r4&@!3|*zO`V@6u*bV2{B>-evpYR9cS&pCzR;=5Q#V#7#H|O%
z;7#cHb3#$PCMlxl_b~@fx!A+hMSh+Gfh&el2LMl8kRecYDrC68YS!u^hgEOL<B_r>
z-$%j`w2_lhYzUed+28^@DuRnd{EQS<@Q0c<%%Nx}J}3he1$g_7#q=$&1ctp8t}jS7
zUufGJJmqTY8#8m~ZGN@Bx$=)k8fRvX?zZ+uXLt6$qg}H;I=gr%EA(XK*Bci&CzUMz
zyzhmTZJEdpde;A#t9+apUb*bL^33qF>#nGik@r&;1uQR6rqK)KLfi*=Gw|%)M+=7*
zRo?VS-ohmu|5N9I+<9AHJ=i4S_l6mtgHvVJ6Yy0Bj|IxeA{gz7`Z}WkEP#6eokJiu
z27secxN%@cEy)(oY7il%T9A%99J$S;U%82ATK(#y>^Ble@6U{w(pg_Rx#?Mf?MmIV
zf~3pyRyPiBj$SKuPdNJRn3lL*zt1?>ILL8%Uwh4(up;m44fCtB3lBVbG-c_5hyF`p
zM3E#BG3i!qM#u?IjVrQz*}#Mj*WmgEkk{wM4KqmL5qS06MY%_gEf86B*15-Nb3eR8
zl^VjX9qMs0H@ng^60;&eu-B|jER#f0As(~yW=K4Uk8PUdKOi46$MD+?^%xv?0hmz~
zhMXLnz~TnhXDl}46mVCt=;XA}(J)zotAg*|zqPV6>B#8P>l^)xQ(UPYx!mQe+ckL)
zKg~&B@qk}Z_Wdq4X<TpLrpOnB(RV+(Qa9C7*POBN`10#Fq7QZ}CU-Rm?|y!!uV*h-
zbsWfO>F(bdRd$!&j{>eHVKRZptbRuO{t1_U_sz*A$ooCA`E+!at8c%aGGXWJ{NwMU
z9;^q`=7>Z=as+sqYYDK}y<@QV5FH+am;nqK7oS*0aV-ZNb`=vPurBw~Z<xY!r1`B*
zyb{>a(&p!=0UFC$cNAGbcF*D)eS1TWcQzzdG`7AKJjK#{Ag}FD6a4;bOlR+>S9h%2
zZS4=6dU&RU-D=a#ougTyle{I*e#bUM4f^i9`()6mn2aD+N8#7&7vv6{j6t4u-jJ>4
zKANUN1_XadD+$JqjDSOD4(&f~Wyo#%cW1(k;NQDM3`fVldiwgU&9Q_VUS>^bIH)`=
zY*Rd7<oX^901)ylhgrbNT4Hc=0I)ZUK>1AWNekRfBy8cnM6Z=Ct6obiz5DI2_wpht
zTy(lJ!>bO*ooc!NFmvhH!Kd#eoZMbMv^$~t&t3(|169kfg#;X$m3ySUXCLUS30%>9
zXZ?bcrk=gwQ?zeJoD?2rgq9y_WmCOp<{I07o@rY$*t*?V_~rUV)m;+$PlZDsF8&bA
zTH}+|EyLKMV)eHM%07SN__pN*rwalXZif@5b*y;x^fds^BvGzSKwIphD6&NTvB2Ay
zXf}ImiZ0@GF>^eeb=q_r49vr>;txbvL`FRq17y&fVDn^^h6G9Z#&pCN&=>L%zdG?#
zbt8{&lD-l7PAb-<@HMW~OIUEgk6Oj7$m$z=qAI)U6RX>IYf2AFBUuY4-^5%?YUzsR
zyL%$4i)WUuEgZF79DGVmQXyFbqGb>@Va@v$Gku??Ey5PgJ)duhsy}@r=~Ft5JYVg5
z<JDJxdQ*8@W<WfKq%0US*efAE&%Du@XC7S1t4rc)MS_Y7FyH_{`8<75b}O3kCkMV5
zmgi-|ri*jf^syL#VwWaSC*U?!rLiR~Ixb}W>6&DT^`(4xpmg%!(C*;s<=Aztr*Gf9
zvhIu!<%Xn%KQ9b8tXfo^fPbQ<v}3p-<aRtfau{1G=-O|E?8O;;l^}dt_IMxJAwVFB
z0ITN4tMC5wCXMI`6*%h}r`uwF5DUii`D~(i0pns&2s|p_lZGq#WY1-GR7p&&Qo~E3
z2LV+fDj+x}hZo&rZUwWo^HC4}lr<=FodfH5HqY<<?BR!7LXLB--`l=9xwL(EWYyB^
zNw;0?+n0~+ypx4<2(m)_y#^#s9I`DD?_|u^x4jind1!IS#q#>kvDZKLN8h9QD7daK
zsiyM#J?7?-eW4Y)v?we`IO^X_J#=Di5KrU;WyIJUo9#PZZ%f`HKsP*5d~vadhsThX
z{zvbq(}p~9Yd}qWs8mMm76;==eh*DABD~MTZX;|;N|q3SD`uKrE1Gcw9w<nQ9m7CA
z^)z{LBakT%eSrvLA5Fddzkyn&$-TB)WUZS&Jks~KR|03yDLb2g_s#JVj~xF_kDRKd
zi~IUMI~1qzLvoSylzCu-#G&-C@|*R^NIOeWep_z2{5Xqt6r9|Jw2&8;wkB=J8{KU#
z@7WMmiF7Ro>rBHemWYliV)+_DAS$7s{O^@V?)P|UO5=r9Gh-jV72$_@bnsbX(bPAg
zn>Le90l6)J3u6a*+91LqCMT6pK@yCn3}kx2#5AR<?5Sle$iOMEf|+hIcyd|=m<$|+
z#_#|h#a4p}tA3+&WTm&qse#oyr`R{mpZ;nr_CHLU>wj32RP$BD`3WnU+cOc)ZNMJk
zlb0<{s%ds@iq6g(&NMzbG{CkSmwJ5eVpBa+N{v4*jZ40g*SlZY(cH5q(lE3sEXUBA
z7VcLf;Fai)K?RlrjU3pK@pj{=r^jPw&AWJWZiUpk4T&%6IarDj`Sd+eu;ySwl%7I-
z9YEm%s4`#y2ijr03ZUzd6ej1Gi8(p4KKRLm9~s~T8hBzVE7%1HvokG5f=#^M;FHD~
zUxz!qdplGz%ZPZnoN1G?y?ielkb2{piZHH!XR7UX<^7e9KRYyQynhkqd|lAR{mHk>
zhu?`R4}2|sQQO{;ERQDxQ;JiQU}|=RRLAdE29NUyaV@M|w%B%MaCpP~9Hnu&<~A_R
zL{?@Z4wYSK#g11!POYCi#idOe|6TBjV!m(xbZE>KE3ug1)@Ff7>n<Vyz!jE3T9{UV
z*GS4Y5$+kyo})C0vPX_-MYK&ti+{8Oi$YQWQ@B~14;_t{&d7?)k#%q(3-DSc3Ce@R
z->mH0rZSdyd#1oaw2$lx@h{pN?$?#7T-*LQO~BN|eQco>eXwwFXK2OE)UJgA%ld5{
zolo~GBl<?RGPc7AYK;L#mW<r0-7&v>nR*d?lKgpoT|9ZlJckX|$UJJvpTKnDxQ39N
ztt|hI6R#x)k0zZhp4mrnh*cVMdd62}cmUl5#egfB*h~}|t6|geFvW%3h3u+a3CsQI
zj#kSY@xYLXhrtN+W42WkXf!1$IK|3H(d)hKG<Td8aW9Q|#rdo|>>+w-CS2^fu_gGp
z=XKke-lw}TAFe6Q>U;dov{jurUr)CoYH~^44tURZ$;zfAUo7`8t}o60(0I2|!)FE`
z_g>#_TPxm4UGT!}pJ#<w5Zcb4JEd6OTu9RDv5ThyIt<Yz=qs^vU-a2GUq5)iV8&pX
zx^?^Vms$_BB0kK$B#?h~8bblAh|mRf7{VYDOr)vZeg(}Ai*>*UG?M@o2N{Qz@p6&1
zB^D8B1TD}bz<?fMZIP~dIB9uC7WLq-Sx@PPRyPKF4`PKyWOc%NNu{2yBkv|4-gdk9
z>7FS4XvV?-@6-urnuqon`x<tNIBT#kYmuRA-@>UkG9m+()rUOjeu@JE=7QpkN3J3M
zFcMoR>}TIyiJJPyvxxz#?--J@zY71iQ#E83-E6!QF=r9omww}B`BCJQ$ONOc`QUBL
zGs0`&0gT9Idx@x`*Tcmic7{KNgTw0xR(Xy`e%y#ihb*+p?B*X|>$;$tcdzupS!0^_
ztia2WN>_MA0yLYT@$K5YP!2dD)ParE`ru*E<A6M${*z@wW%7aTquUBXoq`|vi*waT
zbv9>DY8n2wko#Iz$C}JHUXA=|2Bwh+VmFCkMKS*dS;jugR%VUEif3@BqD&N-Ks&<-
z7`Vgs8sHeIxb@6vNB`YVGj;}HK!KM7P$tT5<B?|O{qozyqdOyCE@U=y<KNaD9Dio{
zimM-27Y~$P89tC~H^%F^L%S*u7YNh^rW|vbJR`Zl(Lywl)>goUwPx!LIc|U`KA=^s
z>U<|C<JJDsS~k}E`2LXaDE}OYhc2k{o6m1;PNN27CmusLbXb%mpg{%vJM;n!1;kLH
zrz#@F;$;9-QO#-|7F3|bTalh8Ya~4vSv3qdWH8bx>dA_J38BB1CaQw{UNnE!wmbZU
zXAH*8vvODTOgePElyePk)_VAUc`V(pxOM*Il3Ax#wBFc^vyM`2mew2ll}DZ|^Z4BD
zz;g)9pnC=pGNLNtxFv2{Vj(ehq2bEAc7-~JQbQ&E`^{b3=V2D3Ny_J0?U^o^4=lIh
zjo>9k&U;Q0F+`_C5q&f<CX8K4Ds`GpJYun6ryNT>@f^k3MN!!!kup(F!v_->l!?Is
zsskhV;!8KD2Z<E_SxZ93(E?oY*^d%(Q>KtlRDRbBN)*G<)R5y#uiv&SdgoUc&&ti}
z>rai&#(Za1v1>#0!OA-}gf2$jeX_P&q9%cQ3BxVKoPg^XlLS7+J@ga<3!{Bw#FP@v
zsf&3%If~yWW6~SREO&AMNKMxpCk~xfGy_Lg3PZ=PQIyG;q81hSN4s4G8H^T~4M6}X
zB#wFbr{)&|V*S}MxOTHAB5pGCfNOPHH93nAm_AYbg_s>Z3hXTwUhRebuwC^TYvA3n
zNz3|^vwQrGvdhZ~PH~-2lg{*gxifO=A#5H{_IPuca^R#e=}_DSv?Lc%(Tt%m@(E@O
z>j<tf2Cyweh7*jI_Gk6ESB8gZb&9a6ZS&mvkNb{ydi3f9tgU|tAUoEJYIE!wgD#d=
zp)q+07vrJrWCf$v+Hn(u7N}hTBt)nT_tkLUrJTltcE~CkLeWX=m!K_pDM*x&q@N*s
zyD~9h`<Qs`70&^ixv6urxv(dxqPg3#enI77kB^K^2X&VZt-rBnfwMmu6dc_#zv9&C
zone#4H9s4;?P0T!;e)b*^x=?59f>BQ{uJ<p#sLr0I=6f#PRj0cjq?cH^1F_~);%Fp
zu5Ni16T}d>*gb#rn12(I1rgzSa>z{>ZK+5#LahQgTtF&Kcm>acpkPY`R=~P6P`xN{
zG1eT+mB^~n@tl%G*c=*JV4n?DwwC$!<*^@AZ{OX!h6Rjy4PO1~+$%j;yX}`<IJ(tX
zIJ8+a;lf<KD~%nj+T%YiHN<CJTS!gs&mMs-oj(`+ZPC<{I!~Wb)0-hDO4dv02JDYE
zQIN1Z8eYMv)KH2zy$1?|dVvv5L)Se2Ta6o|{110jWgT}lUERJkHgPMYk&S(Q#<Tl9
zH^Gfsl1{xoADoV%rp3b4JTWavD?q1N8d_6|cR>ZnAJR`yNjxSlNQ*kuV|9A)J6;SQ
zP(}V=B2Jj8Q-gCybQp2p-H%QGSmM<08nKC7=@}qs4>o*j2)rLr<sC!QKa<C;Cu@Th
zZ4%x~WB58Vn@#6YksY5dZ>3H-P4Z8m-}P6f=|^^rndpn{&0-s&>Q!pr{x_-)eH4Se
z8Rey$iOmLVk!FskUgLlC3%Xc<UBETem^?HFjs*C#Oj%cJjA6!Dfi)GX0id3MRjt$G
zJWd!H^p^Q<!~%D#BJRlRb|LAjlW}XmT@ReiTe--;-LoKftxMl*%gUS5Y!pX#hfMKm
z1}qr<NP3t$Py_l|;2BoCI}K2pU>?4R*oUKB3kF039^lPM9m8u)$S*GMd2hU;225r+
z9a%Qfq;`V<8bO|*ogi!iDqEdIW(+E1eVGS>=n}KB3k11rg(OlA=DP&$nc}DjHq{B(
zo7e8hW8B0>wb6SzSkOf*$OO-@<!z#8Tk_hTyy*1%uJp($-a?mWTkhn`-=h_S2*rzC
zWxGiMgXt+9#}1Wcz3P<U2+QIHjDkTJgPVX?fqE!}>DXs$dTySUXG%@JU=_GyOa)T~
ze@jcuB4A^6pav?!M0|uNBmjYz-q7+)g!ZFqSg@<7#<W3&V%`cCUnie109o|m`fIS=
z-(QIs91FhZNHZ>NICcBJ(%5-_N5qMEFIC8tqKykk1@Sa$F*D+CLMHW<sdSc=*xUzU
zZ8NAPV-zo=&1_`|Ko}s9GN0$)1c6BQKVG7TZUg_D)V$H|5ClAzkJ^9;8i8~KsTzUq
z;L+(4d`Gm%;~V@Eyp@|sza-`{VIBI!6xdQQzz$F9ZJ7?7XV|5XyO#`Yd&2xJo*JBK
zaNvuM!M>g0(B9sXn~{K<9L2eK62a$N)Q#mF68a7<#;eh$2#ir5i6rs`K>J;Paz-dU
z0{WEAhJ8XkvTAB0cFBYyRx?QwaR&g1P<Rjz=D_?IQ!z!0KtBY@?-Q(QFrNXb$e4kp
z-3~xeD#kuh{Dqk?@uggASz4e>cPS_Le!bCnxjl2y3v-9DSZv%T28=+u5gut}brNtS
z(BlZ*DICPnz_=WL;epEMzX{J_fy4IiDGNeksfQRn>u4ATFpL&~AOVPg22a!c629PU
z0@7RHFM$($7pKV~6}?F$P!KwTMgf{~L_9^23UC(k9}#o-uhb84kF#3O;fO$L3f!Ww
z=Bp?^$9NMzThIaX#Qkb8z6jd_g{#Fp3VL9)#!6rtbOJT-ftVOP?0g1pM$hNp1l)KN
zF%kX&Jg8=1nC9XYdOQMrqE~2CpNHQZT>QGzq6(R93KS=#CK36_*uRWt*faw#u)u)I
z#XY`)?~0Ad&|5qz2bN)}c-gGxLw@xabwSjr*#fpEB!L1I$Wl-PCRIE*QIy0wV2%2v
zLn4pdZzz#}-@af*3^UYWzMe;2H9i0=Cg)@(Bh@2g-pKF+Qt&YQuI6zmIL4Y{)bK*<
zRygp4h%_e#+lyR7K_@8y%5$<Gd9efgsE>wKe?mP_tim)SS|LL!5}ug$W3;6MFKWkP
zBT$;|WAfTSx`0kr0RJ59wFPKX04tnDubaq&V)*X+@2@WiNrT=%Gbr%Sb8>RPBhX#=
zs)jb;(NnEXl7I(`fNN9{0S!0DC<qLVq2rW2JO@3)6FmOW9tIX_LItB1uYU?7@L%hR
zR-UUaKiuOnVaesTZAQ-nADJ)cl>^F}FN5zT*GSKr@1})2zu%!ryZM@@#x__HSe2B{
z%!QA)EDG>SO|D4T5&m+H;YNCBRcC+h!EUy5QGhOkGL>vC=aeo6s$i?(v2fMMTmz6?
zbK~UBwo_O1tJ-!(PU)HzSaS}&Rjc3l<Dn1<3{RU3sUZhYu9!0r9E3I%9ww21O#+7q
z(BMp>w!oUq62tPr5rUSI(kKPm0Peu6$O407wE~obD)J6|Y&V^-KJdsJOItUGmM=R!
zcS^f$^6>C3GuIY*x0LG{-Cc6>wdDggpQhF<%zd@*w@tp>{I;eux}x~bUe>oK^kn$Y
zGec-T<y4br*k|Ii>$m5=5c7ZIdgh)e)Fp=oaFt(g2%Cz~%@VI><db2Ga*~yN6*@xp
z=H1V;yMLaUw1eYUB$V%0`XE}>s{k{D)?p!Mp#cTZjJ!&M9<UhkL3)6S2q}^ya<bZW
z0tqlNzzRd$RfG;<<JHDA4kQLSIc7vFveO3@5)dos+>{1NpB<CBwuY9E>qVGb-@JFe
z=twW|t(|)>x#fQEzD3n73Dw?G^1K&fy55>QrRd#9gQDOyJv8R*ric@r-z>6ToKlF1
zujs7Ngrc`20P+TpoXI7%?>-S0m5-a1D}dON!WY>;k^($Qo2{sV%5`ZTR==)6)e6s)
zFUEY<wiRU2ajW~2mkt6j1w>H>R>0t0KL^tzi2Os9Gf<QYS)%8c__o3XN<%S>m+y1>
ziBJMy4LJUi0Zs)CKWJ8$3*wcxM?c2CooK*wRo~Fw&`HaBuWnaGFX{ey25-I*akuc<
zkLMnyMproRY{G*5fsbzYO^x3^cS=dztp%dh`7kST>dpQMmraWdSKmVJ<gM!c%7L5{
z?Yi|i>bamZwBlBs&60O#XBhe6A|jBAX1oeX6n4z;DaX`(zay;tP&>Dw6({-xc+L8|
z88Cpz+e5qZh#%D8a-7DqPJ$s31S?L8rDP!7Mj_KJvkN#ISfn!hf@-298btVi7)gSy
z6bf~X!ACd?oWXCx0UlrmGv-x{{gU0AJR%HtIko5FOH-FER_>WQrNwh*>5+5#hDjr7
z(fiY)^iNZepI*LgUiIMyYx&{&N7+4-mR+A$oxOZk?vf|o(z~T+q#>`~t8ba+o+$Qs
zGUk5s{+(=nXvMPnz`^oyn$jg=Hy~GtF#U`bp@Tk4Rzg&9Sa(Fg;nrm@ieKlm_4BL6
z=&aAMzaw+3dUpAXg|vH-0((K7W_3IhMG@;AbXbB!ybTcCM20RTR~6p?JN_wR%xc*3
zeCtuBSLdl%?a@^;Bp;@#v$_}^@`&l;L|{{ep}E##RerhOKlNh~kLlXCMO^p$cGA`*
z=bp9Uq|Uat=S~q-(OEtHY3A;>G{CJ%-Z1>E&6yQ`yfY=&f*jlAEK9=7z%$~{vyvV>
zITSYmXUE^zH`lLo))&!PP0!-)@jgdCt?;hH4B@0@8cYq}O%E-rQ>&PSrS6hWV0vGH
zSM#jlW=K^+{f4l~-nAJKeipGO=}6y={n#fLBe}nnoH(>}5ZTeBg3)CV?AKX+AOY}#
z`tm2f33LHgV2Wl(jv2d{;CGnU@yx_=90Vd?Uh@uaXVX^*NM5wt+WGS@?;k7kymrjW
zm>TV_h@9$@S(l)C`@LhLaHdT}b+-5V$BAYANk^6sgtQD|ZH6NE!C=FL2hC6S@mPu$
zH=#!AeJ9+%gQb6XJ@87|o%Dqj2{Usa6p!v+gpmNHFifjKB8{6<RgjA~g?~b7X^X2N
zM2ZYp4x8MlQjbljI+WGbXe1qku}eG@8Yv~r7Z6Njlz=w6a0BV)ah_)e&IG$l_Yqf^
zk%J4_p21CJyjl&tkD4Y{P}+WR{GU(!=oW6S>x>I0*62sEl>PSb$XG?d*4u;q-W%18
zYX)8~iF4W>c=M;R5ewX7oDU5ygK@DZf^kJSTTGPjLck=Em0iW-jjW*7Faowr>KGi^
z6BW?R2Vwt;FZxqo4|(5!{FO4QLWTFWuaN<H4g`rNV*(&jkbmJ;+Q_$p)1lAfd8443
zqk#3oMxxw|JgrnTb|omytL3%?`!Vn77rf(c-`$6i4CBLB=e9KVrI|beH!Pi%k=@jq
z7FqS^+WG|nodvd*3;CK5^YAG8Hck`%gKbIq(4J7(n~WvQLs50pBLnO3-cfr@NN;c6
zvao6i#7NGLBenvgHY$ol!Ej9Af#bfbTNh5|mSJVkH!rMd-G`ixI@ZX8cqgm&be)qs
z29t`!AvL5GU)@@ocoHP41G#7`3c!mLOrm2_Ypou^!_$p|um-dCt9aGeX)IrAOr*d_
zK5AjE0d^f@6U+x*@Lv4?Qa>PM>aV?05`5!|-wkh{qNb`x<F<!a7q@*8EKM3Wz|xA3
zonxts0*X^}vpi>AzWo{}e$C|#W-i43A%ocbZBqMN!N*I40~1y(UtBuovJ>vOi8n)o
zpj{@g**<nGL(#57@}Wt$KBD?&K`DY>R%d;NakSIpiMi*gpvz&a3|I(T{C`XeJOQTz
zECx49lBfw_j2KR@fU>s4L&OMpRDuGB!dJG|!Un@*6pS!g=GZbxf>x6d9pdma{0|Xg
zg3K$<ds++8cJF-?nhSpUj-hR<DLm#wk0-`uno~Z@V)Hl6pK5<clsp30TJpNn&HZyQ
z*a+`Vy7;7I@rc6t!=(2&509=lj(RRGy}ev>DnjbpGSBZJ9MB-R19CxcT5y4M2<@7j
ze2)nSKO6FlE6rlYvXQrfXI^{O|KtG}f6U6Hfz(%@fR6#03V~2K8&Kjhl#oY30QZ_`
z3@ac&K|s<TQ?%747&^DJNvfl1K!ahepi=Rug_W5{4YCie6BytX8F)P}@(=tQOW<bR
zB{}w7)r6(|($y_<<GRlW_g`9kV>BaV3cullf${#WCa$>q&4@Br`uy@oH`5kP@)|gG
zab%pwKNQZ7y;4hSn*F;p8TX4nuDiN5{CM$!OM*c^F^@@*SQI_ODG!*LW^6hbyzf~<
zQo0?fN(_qK<5GZU|C8IO<VYGA0JwlWQ6$<@5ePCpI;%qwwDLXhGY`cen4o*iF%VG3
zD$XjLzN`#nlo<fzS8-{Gyx;*i;R<IyKxL}<0DH{D|6KCdwf|ht5P$J~D>A@qacwT=
z>f9gL(tTrhq<?39ZuiCy8pZ>b9bDf1q=7O7z}CLE=pHiMd+U`~db{}-Tf7d0JZNr6
z?vS3Yd6wDp@a=HF;*_fZw@DtZqh)j`HdaO8;<z`qV$Jf@j0+3PZ{i4p!*LVL7rH0-
z4MkB<(vjqmd22pgLZ^hMa0dfSbQr#cRHhQ4UqLfsX#z{F(glG-Iqev|gy>prPHvf1
zhmZ!$sk^9<^6gIGbQ4=e0}e!78uOoG*_9jOIawaLq3ecsh4^*;F!WJXXWSPdhKFy5
zv~BrvZ}f)CI7V8dJhEJ4%W8gX|7Kmu;*bX}*M<d?#`OwM)^z1Xp3o&n1uVY~2fOW^
z5dY;V_j4Lmv*Rz|;7+**(4%?+Iv-Q`2#h&9_tFGnN_0hE>)Sl!UZa>&h*WQ;o(tu?
zO@jUfTu?@ED#Rdih(rL4yK>wUreroeB^^wX727-_2gMC&gd!X0C85Qj2zmk^PGC}}
zC_|V5bVg5jp6PSyykEvThmU2%78Esn)EYm!w?y-)C>lm|drQK4NB5PmAF-L@W1O{9
zB0Qevc#6%h3UVLxM<uG}?9^NeXz3LjP9YaQsbYAP56UqJ`6OUeFu~v^uAstVEv&)o
zQ@SFn#mInmcj%KPd1c-TGl`^`U=K-&N@^wv5^kt?z7erqwHgy$@Oz27Pn#`+aK+pu
zsSLG^0uPBFL+@(>SJ)Lrv2;Urx-gbTM8|u2-HUhLzPs-MDe|DW@%J`P^(wGcB*cd2
z)(Ae!S9K2W%kKK~wezziZC2Fvpn#H%@ZjsvB_b8L)8JIs(4+1G!eOqJO9MKgIBgZK
z0!8i>=CLC!a%w{A4#q^u@OjnFo0qPN0!Rh{gU&(Z#AH<fA~jMl^lYVyF9a~=_{1Cv
zeEn9Gt*|`?qaE>CX*xVZqc8)MGpXPjqajca8|^~>Vnv`88~6WM{S-3Ilo1bWq`ttG
zvlsoAr<@wV$5rjbzogekynO%3+Cw|e;poL#%5BLXHiTUH9w9a$7I@^<Duuc+5$RYs
z3>JbxI+2`I(-DDp>k@TF^oGx{@mD~}CN=nXmhE62Rx430Q0(&s-4c9r!BAv%S@T$)
zios!1Nbd`J$*8sKMfkpDr$I&n`FVTvB#$uG4-AUNJpRa*Y^*N;LpuZ<lm#e7Q;h(8
zfr%5fWw4~Q$pQsU_R-uHwszXu(%JQP1PlxIlxW+AA)1zt|JxMLze|t&w|2ki)1)-^
z_H3Mg6e)qo`Ilr>i1?o>h$RZOH8IbH&PlysB5#6skBMNG98uMUJchwhA<3ud@s5Zo
zU8%KYKnDvRcr|02gvk8(8V=jHfLF_4SR8^?e*~xk<;o<vsoy>@f&no2?yYy2>{?dC
zW6+X7tf*ESj7*#Z9}Sp=a@Nv`y4V2FF@nhGpD03s+t**Q1de+<;(Ac~?(mZ<k-*t9
zx`ZVe2fuqQ(0BLrw7D-o_H4*qTllr-fy?>e2I8zSI2UB1q~?o)6jw#+^z)*{-$c^v
z#!74#v!wX@o%8jEPKYknuvRn<675ZM2iAVHt%)LfW<i1~Q0Qa824^hEw*@BQc?GmA
z0{b#|k|+UR=slDbrY)PZW2n*?yC_Mdba|C}25%w;67Y|y{$R(}o~jz}FcyRn3y$%C
zhz8%iqx8A_&D8I*KWoPt+|HC!-JCt^T>Tf{oan;(<tF{rOikdP;#Hr3!6|Oa@B$8j
zCAd~}LG8)~a<0WU9TWD<LlV^`7#gwQ=u`~Ov!x*bZct9l{CwdB#4D|e@jT4g-_`2b
zxQ&OYQPL?ih5Jh(LnjwyE4Oky@G5^-*DwhzET|dplV)Fup@A{Yq2sgyWH8bxOWr34
zLIA3lA9eKhZi|Rp*#B0zGI-RpA*s$`%TFuz`1DzSY4+%&fz>X}1fy$oXLNR7+x9|-
zwMeEMNV8;OjllkeXy^q;vk_pqiK4#YR;2$p&8ahaJzl9e=B@XRcjbPZoq~m1#-(fT
zZX0SA$vve-UGQ=w1s;Yqh3}8nA$Cv~h~SdSB358(`e5GUMJ*2%%&_-T9_^AN5E_Zn
zB0M9OCmQipntpz|io*5(wVwFdb62Ea*NpHJOFc>}8k;jCPPFP?9Psy^dFsg7;Fkyg
z?~hHx>5Kf6$LTV{{9_W4BnB(WFed^#C(8ncAiow4R0OkeLJimXZ1Lymd8g+qeVRr#
zEv)L&OsHuDaZK!bRchCt(JdHtMNter3JoiuO)_9|ffoP@A~&9LhusC+@{AW{NMSy5
zObBLjX-MT@NT`I(u;|~%kkSNsS19yP97?hS7=<|4j4aJ1eZb*VX)iL{Ri-g|L&2#Q
z*TC!9-j2wk)2eKNY!O4yMYQIrF^nRv`>hBfItG2Lcl2dqj}fG)vp)zbAk?$s?z{h)
zSmds$1!C2#j)9j56nrzGD$C>10faPvZH_l;y>}K_osxpjLpT7!=0N-ee>js>KYp5v
zix*bn36mD23`p?pQ{ExD!YqhF^h!P6kW{(n?5+s?;I0tE=<bMWOE(s_yJF1!SFi#5
z!#(DnCjs9ESO)dAXW9e2et7tqerRV@m5P0sY6(c&i_xN?YnR5JU(?r@88x-5pw#tC
zw^Zn09r|nBJl=V(zWhf=`&JWs{HDi8M_Lc`Vk7%RgYx&8Ct(OM<g{W$NV{49NC$FO
zfgHINtlvTY9kdLGK!yOhD*~0_pfa%S^jgHs9h{>{Z(MH#GEG4`j4W5G;I4?VbF}t3
z7nm_&+8mDYMjfIbxkS0OoB;f_rhPaosdDf6xFqxaUM#h}(20Rk%hH16BNt*cfj7N9
z%CD~zBdeqxQB|BSzCLM5xho}9d#rsh3x~-spP{jt&mpcg&o6$>wbu@5ln>fn>5C@c
zoG$kmyf@L!Zu;81*>9dp{b;ZnvS)L#8)*^FuI(ZIx7LFv*gWXZ_-qRZs$wGaAWfu_
zfoG>dTO{mY!Vor5ZH7MvAhEe)87BDz7!g!*c<n60R42X9jB%F&vwu!zl2nM81a@d^
zPt%J3#OeeAgV@K`GUd$}^{5=)gc4KDJ@H*f`S<%~dxm`)KJlX|U7Azb7XmyFJQ~v^
zG$>XlMVy^?_flL)Ra#L>6cV{isi9N74kWJ+*Sk^<#Te7x#5~p2?eHR&dOS|j<7n=<
zc@;NP0@I&^&hkZGRti{N|L&6q(&)b6<2M*<-5oxqxIWhfK6ZFMxV!B?;arNe*DJ`Z
zQsEP<3<#*G)Cgw)G6@EVOr!w5UkU1(g(5aRHjk~HR%TEuFr;Q?>}{o&ICw?k7^MFr
z%#S3ufDKmBhT)`?h^T@b&%O(t%arPE5drqZwp{gF#UJ;yZJ&qL3F|jUr`PqpJ>9Y6
zeDym~Cj!5@d&&}cEG@h8+t;RvK_~gUSXzz&=<Uw#HxI?AJwm=p3qFC#izJ+La5}j9
z=Hig`ny%1Ex6Yx5R5+D9<KbY6y!;)({)ev7(3|Z+%Hj3giVH)~tHkXmh}$8TL|_q$
z7L34SJdI&B7?Xp>HA4v)XnZYT|7w-QgT4+-lB(ECR_v|I^M#I8JhocWQ}U6P@jONQ
zIXjpH0W8IPHCSV(QW=CdHjidj-=#RYa-eF9-_n7k2Sd?sZT2|TH^tKUtih$8TYdQY
zo_vp;_mT%rTBG;BJ$p|C4i-=TDzk|BKDn71xI*mlipyQPVS$sS)=0EsIB8(Y4VzDU
zPs##4GEtO+Zes|7aWt-lq1Wl_kq_RG>~n6E8K2y|e!+Io^;8CX{`PG{f3s*ABoBZw
zS{a5s3t~oaK89Ds2(qsVb(@8gEi79^9#)L7YRwU_@)BZUFeB2Wu7NZIZUoD$Ii&BI
zAlOC5&T>Xu2PK97R6WQ<iFtbboR5R~29vnYw9kGVHQEMVuoslKZ#$QATLx!KaptF*
zz^ctYl|Q~FG4qq}vc#V?H=~bu%__A#NjS~CRA+YupMVa2)^NvTeNx3Z>=;-6W%7JC
zi&DRN;&P+KlsMTl?$l5~@qzN6FeTFU=6rwqo{;0W1X9%@DSch%YRf=k>ata&{xU!;
z9*IS`gsx#Y6?!N$%wQk}NCZ^_BnEeb3a1NlDzr^TEjCsl!%ok`zko0pc`j|5BZe$3
zau2_k9h&C1&nV3|{)7a|e<;Uo*PpN4YIJpOD_(Wsg2t~iGpg#8Z>HJRl^GS68<lzZ
zdZM9T_toV0El&<3@y+@DzD2F-5Wg)`538|NLaaJ^bNBGhuz*uD_C;I^dpTr-^S&VW
zh-{Duor9^>JWc7CbHms@@$|go%TpCqEqx8x91r%y^TCym{^}qwm;@V+=d_#(+56;b
z7-{EWoeJt*_Hzk~?gC@1o-Q%F7JevBM-WNn;RKAJ@uoT4Mw)~+_K6BX(XDB$&GU&5
z(m#e{zkl4oB@As?eq_M6y2rZ2fIP2EmoV^pOTw(iV=-Kx!O*{JhIX$HSx#wn`6!2c
zk1eqhUT74=8EZI49q=g^Ep--J#6D3J`*-~TN6j#<v*A<+H_VVT#Vx&Z<>IF_i3<_{
zd{`y`W>u^dXFyrgLCA8UAF0UGr5KjU?h<qxPFctM=~;z&>`Fm_3^8I3#xi(5JD#&}
zV;s8CleVm35f6LKDjLizw7Nzz7RFtg=nPq%6nw5fqr~?@U*LnmM3_>9?*y9I7W8$J
z&lZT#|0*Khh=m~%#$@_u0~sqS`^SmOit+=w&L}?&{D6yFzu&%3LM0LZPD5_>5Wz(F
zLnic7Lb}9nM3D_LRl_YwS@|(%uQNf~iT(;mAIqEmNz2AFaPy6V3N<7>Bv&g;Q_MES
zT6H*-D)kb*PwDgJmBmHhPFm*M-IY0gZ<Od=^@?Tq>bA)1Ni#NQ55h;Z<`7O1Jv1R5
zC!gih8~>#!WW3|8V;%J&M}}P~3(A)@D~}A1?C=ox@ieAYhF-Zc%!*9lF=pg~;I-4_
z2~$Ia1W6C|Dz~Oq6WutAyJ_@{dF_zd6t&J^>RDhoHt9XJx`y+fJHwh&cABzDBCEAr
zOw#Q%9V}_qjC@Lf*e8LU0+vzz!+&c}^4V>{Qxbx`W7*a1buU8Q+*-#|QW{W(lp2QV
zc_=G-*RKU|;#V!^wCO_46?Se9oph)zaDxCdIj?LAjE3OgmOi-h!>4cqta=dx(L~WI
zLO}o^cOgeFM}*2$L!E%bkY58R#X15tsZu%;i&=I|Jt0X5bP7_4Mu&t%x6>x$7`+bR
zSFL~*xzJ*f_m1i*MU5cJ=UHp~ftJZzZ(nHMet6aQQ8|X|D@Qk%99!pgr14p%G_)Jp
z%sjvXy#ZE^8>hc?i2mYaL!)3Mpr9KAJT;|*732Ct)(RSB#O25PBJ|%w00CIVm5d7_
znK;TfbGZ3XeuN?61_sfI=mHFv8&)m-f=i5uwbMkT#bEy5q>v=cNkFP|&>I){u8ucp
zYV{KtJfxLKiKX+<E|IDNjYi%KUA$%Ft$8PkX<+3U)-*AC^Jvhuxi1e&iir}UL?Xds
zJu2dC5#rYWQM8FjdUw-_i(Rj=-ERFlIjrpQKFzHKW&PMDczG{D5c1L=r(94GZwVMP
zKQEYV1%)8<_zqOb15z5yfI-3MxaBoZG-Eo!3h>l$%XN87an`&|uZWePFtu5jqwO<{
za^G&-E0`ocJ4i1>br$F{h;3WzHr1^QI6TW0TzzP+DSH2YMJz=#_C{y!XM9Si3(;Q;
zdn`E2L?`8=1OKydI)Rt(K#0e#UHWEpcKJ~H0>jnq;U`1yW@3z=@_8fWaCwe=({@`>
z$yI46h`|xe6z6UXgrfpNK^b9^6@MW{SOpU)!H2QLgS{G!f_9o=Vm?hsjV~Hs-KfRl
zOG%Of;Pc}2pOq>f?!}fd)`r_(Lmy!@kmLjF<+TDql=HL1U17q+3nva5o(-IRx;1)N
zX3d4gy05jEEeDEa^Fsp*`}1D7`+`L)FfH;msizM$NiL&hd8)0Wx1rWqAdx|V@>{xj
z4!TX0hPHsNCiu{IcaiVY5dJr`>9k{Y->_OHV6bdK9wD-#Cg$@h;|!@r-I#^l5LDby
zejyzqqhg||kd3eDZ=8d`+8{{JC4>#l@D3i%JIhTI*NyDOs<OhQj-edYS&RmGoxqsH
zOi<_ltEh0)6SRl>u4BYo=-s!_si1ult2uBPRBGgv;P*|)qLM%p^cGz83kcvh$WV?U
zpNG_dXmTsErV%)AoPmM7CN4-34+s^<Op=kQiy+l2F!3m$U0L^QwL;eUS;q^bUWm(b
zQR~WM`y#43Fw%>tycL&b(Dm-YkX}<-_s-ma!EM<l1Tp9(CT!G<&d7xzdtS!;0tb@0
zd!G#?uWcPnoqtsduUuY$Xm<lDLKMb-({`>2gC0CXDh)-H=##8FiUgS4Y=|!c!q~tL
zTl2$;7$c9zfrDejh-X)Ti2$)&@Phyi%Y1P%bUazt??uo@8T|d_<0UD!+x6kq@3#ys
zbk_bQ>vBNx%=;tzLi~aPBonkK5RU<Sw%~S9+^zl6WImc@YWiTSCpZ^Qy;(5f!LTSk
zPSIrC?xsKkd-VHPx7>gaKr{tQ<5Tw>2wW2B+(py_Hu&z)cbr}<iZmdcj%b|4f)3>1
zLC_NbyEP9O9@m`7*3mDHp7@U{U`fj@E!f+aWGS4oH2Blc#z%=s1NU6h*ER3gip~3x
z+m+{&1|JKUcM$IWXCe6oXHwTt)H~<-k4v#_E+5*(o96l>?Y-mv5D+J@Z+`#k)@}(e
zqx=mwZSbIoX8;>eB;X3nA$zYNk)#zxNG6vMWcno;SC1qY-t7TWU8Wb0sW5B9&alLl
zk==IhRil{z27Or?clIa$Efs}Zf=?uuFJwvmt8FzO&Ru=!L}1@NSJd^hVl8SR0TzeY
z$qoJM2Rg`rI!f20vN|xk7t4hzN>bNz?VWG2k)4aKNj$ulvMcHDn<OavI^dO|9S3S#
zMSPG$0I6Zv43Y=-VOD9NLMj@{4GF{e5ncl-kcFi`YOP)TgI-rh`VcSI7cCL2EA7^M
zx9{%aK^h_W_=~rT?;yEhX@M(5a50Rod=e`6`^A!aPQ>P5)*QH%PC9nwe-%ABiZJs$
z1r^~o2%FBULMsw#Qx;ZXhp>*q`>;9fJ3_4_+{#MQ0Sd4S8GoRF3OGzfe=*SvvnIkF
zCu=HCQ3C`M&m*dNyhcV9AR^o`UbRLDYMz1JMfG~4h1bPn9a|n_wvKORU=Dh~3Svu2
zW9^?O-HI^|KHgruE36qx*~MeIyVB>A8cX}&G?DQpu*Cm{hj1qQ70AkhYXWyjtW%w@
z&Wb+R`FNagZHl+=ljR3W&y@Qm-$xpu=(Y1v+qqg1qkifZazF%`Ea)Z3B3g4q0p1X1
zp0TwH$Oo}fk*tuU9W6o?0tZ+yYU>Q@8ugk)y-J@C?_45$&@>%Z*>?2lM&%JA2=BS^
zN>z!+r8e7F>4yu>-eyT-nhedmaGHxIekjf7c$4!2^#KNbtbhjor4UqjjUA4=9r_Y^
z8(4z=jboTGGHm(41ZdNG`Qmw+KZ6oMU#WKj@F+^;v<ZL%Uogan;Y2nx7Q71pLLs6G
zXF_e2X!rw0N_4(~n8&H1BOHVT9^lFL3EqKl9;EpcfU%^Om+blG-qV^=ml`u-+Qi(B
zL2T$!L^bsoIfFz0wYd2N?}s3uXBb91>wtXB*m64k#WWK9AAUa-f{)(At49?<YilP-
z!SF2}V^5Rauu3GG4?G|<jg$ceZb(rh7n(VEBc#UcFbEv-R|V1+=(q^gsFoR1QZOV#
zzj$Xm*?eegp49<rJ4lYH0lZGcUH828zw}zg)I7eT0ie1LZvtT)K`=Rdk-(7C0Jis@
z_m8%75^iIRskH(e;MX8?3f>K%MCvH6@F{Cyn*NYP!p0K>b`vFa9?w)rypqs3Rm2k#
zXr^L$xR4Csoqp%}C4V9_f?t^#uZ8;|HwK?q;L$;Og)@GY`4^f6-K7F>pgk#80KY|R
zf|F^HBV9V+EAnpzewVU!7KfnEga4?PbLfdX;8x1Mz1#wD3t^^iL)k*-=+NZRY(*>^
z|AED>gYQUhGEr4Mvtn`<7hp~czK&$o9ofJ3*IwfXIsZ;>3X3Mm46G1v7lnk*3iT+^
z7toXlf`YI)=GHj80iA~5MSXtqQQcYrPh&bOA2O%_i-0v?h~4$^$suCFAPH2!+q{hP
zqeY^Xczh*($DiMNEkax2{~UPlwhps#KNfmDNAoYt0aunNmrtQ)k)g{fxpUM6^Z?qT
z0t$%u1MEpFu?B$0HKRmxNfCFeYj|xex`MiSq>|8A3DA<m$e_5Ma|Au$E^f}ah%ewB
zv>@QP85NW86)lJ&RU)XM7(4pgZGXqax0UYk^dWOI7}4sK7zTnspc-hD<ew1HA(T?V
zG8Kd)1lSZp^$eZGT>&-@qV?qB`*X@TxvCjMCa@cI6czCgjG<D4B}94v)#;BL1^8@&
zwig_bM_?S<!rddNUIf+A9Oy!6RIt17PEn{q6yO7WM;wJQfgz)tirg+Sp8So0!)aY0
zqB{bl4&3i9$a)tvU2R8y#@fH<x1iI|B8i&{B0_Dt3-{~jO5g;n=3^?s{R3vm+;2e#
z@iuom7@}IovX7A_BX<;hiOO_-F<=19P^+l-FPa%W1#PI)dI<=8exa-L5oS<>YUICz
z51ENNi{Lonf>4A|3L2y9v(v!ZOrl@F4g%f6H-YhzsAdL1zGGsAc#WtaS)o>W)WE|E
zkyPB1*B!qi7Ho>9;XEJ@A8!S%@fiNr(gzhUp<n~&=+9bId27~3_p`^o+7WO|P!_)L
zZ)Xt!%LnEehLX>xt{F}Kpuse()tRhU#fP>F#SL42GH|c9+b))!4n-r1vXOU(x2L%9
zE?J(SW-!3U4@lCXmr#NICur2vU1OR~?gjEUsM{DwV?+cMK+lL0fuj(qB{4?LB<Wb6
z{zvSLeTiDZ(*y`YEno=nD`#o%`s|Cf-9RXwqY_ea$fLBybJiWbZ}U+4qDt(eBF*+%
z)!AUHV<CPoM%43~dDS;kLZ_CjD*r#qh@(yZ%P+86$)qW*$aC$Ax)$Nb=B%IpUgW*N
zD}`TzGU91Xr_5H=Rki#I(&Fwfz$OZ8KZpg|8j1upPt=0n7THXa#=GI9H4-B3;<R#w
zE0*U?z5)eSk<Ag-XZZL}dxk~&29*c}jw>pt?u86ruNNm+I<aE=J;@~BP8!pDXScc1
zPsLS#I<>1lW397$jWW!8Rz@A?`LWcwt&qc>6d$B}$8~+sX7PW&rgCKabe***AItwW
zfoq$;Fg)DNJ|5|+&%N0C7@JF@v%X`WWSlwirXdfgrST>TG`v8ZB1s}@xHYoEr6nlA
zVsb|mI8B38GFEA7c`!7OAEu!)LMcruf(^(bTH9_~6?eaS1P4LJ5e1+KL_x<vld_1Z
zH)oZ0IPdRVSn>Y)dDpiLq8c3NrS&0M_a_~?s58A8S-Gq~qo(X<wHUnGR?+wq^9-w<
zch6PG6@3y{G~nm?<G8p9=Qe!MpEP;Y>!RHeZ%`=B&O5eEIQ6S}&c^wZ17kPMFTW{)
zKh&{*UEo*om(zFyB{-5a$VHK(AB`7*a40s=mncc}l|kj{?-Pc_>Tpi6H<`c#MXXE@
z%t+Kx0s8OxGq#&q>=!nb+W>gbQ&2x(Yl8I9;>_O4-+kNg(Kj!DZg=mH!%QI+<7VcR
zAF3-dVX3KQxIS?65WBiB@?LV2&NK587CZpX*pRB?%*)zITX${xaONq?G%;R*!^qjg
z-Jx-3+qUJ~Zs&cO8dZ67W?&k$ux|FH|Lz=0`MFFd2<ST?L2?w-jKE+VgR(h0=)~g(
z^cqDckh=su3rczGb*vFQfS>EY_gDrZLnaIGSuIbsnQBHBKj*|d)!)*?xJ8;&Wx@ri
zL`i(SGqn77dy!?cbX7vFx8?%fXNG4Dxt7MsH)r^?{b|y;1{}grKVfqWxY6Pz8lL_A
zlVGeiyP2~0yQ!v(DCDnT<Cvd4l8=mgjMoXa$(8r~zcY5%7yiR>XVb!|<5DJEY3~7-
zDQb_p!6fILC$2={nNA-oF?ubj2N44xXnm1@m1Gb>bG`HfqDR_Q(@kS_4$SERa)(6c
zM{<I@5iFn_1LmcGr1J0Ezh1;M6aip>Gt|@=SX>hOu}OsG>VL!qu07s_QTr(g_c1S(
zjXg0oBWxv29+!$elD|%m@L%4mEl_sUv81SJRYf@XY569sr`5y{W*NO23Xk>u6YZ(Y
z+Kq=!pPyz<m@&r^kdWC2|AiBR%>Q<?>$Co!%jl#50X^44amKQuUl|uv!oy7XWjcVq
z(k?gb2Ze8}$)v?Dk&Gu9gv1#N-Q#aKT{^5W#hp5>Bc2kk-uewh<`CQ<s=+K;+k?Kx
zwg>Unzef_9P&|_DKr0+KM4qTB6@*6-^3|@Rm}N#(Zu!Eib*!e+g?T6M^<bs=puar{
zIUn(t(=_;lh#}7-2Y-QSHIL2;xW;G-ZUr^WF*4I~xJbm8$ZHODP@+?zfK!ZmF!&ss
z=ETy{v4X(yXe1{asW~w^%b3?Ofd7v@-;0d24pUUe(_jfI7}hN{*8-DcZXL;tavE^Z
z1hyn8?Ql(9-i#kn6tO^$Y^g}eB{EFmBJH<wSwT{0{I~wiiYqUK|J)XGu@zZclB)LH
zUqcrDaAIg3T9)L;NpgTHc`F{eL?4O6p!hkrmc%^+R=cYeyd>ru3ds`&9=IhI93hV3
z0fi^<1RPfp4A+fG;{U+hi#Amq%CuR?g_-(?*ETZGg7_c(vzbPD+1NRx-4G2KPE?d1
z{zGL?b#@va21Eyug567(Xq!+V7sP6D{40wmfDRwX_J7~7jp=Ueh%Vmn!R_il1&Gl6
zigSNmuy6nn(U5?qT;Lj!Bm)+cj7$9G$=}Ndnh2#J1Y`-Kz;Vl0#Mc_0iUJ}O7K8N|
zG$J57LkPfMxsYp!CH5VCh!_7ChKQQd8E<8)`8PjoNJR9mQBS98V5fwcS1U{6YzGhg
zM_f|=j<D+&HZD*eOo)>>OoIkyO9PHOdm;jsr6k`Tewu2=sej>9x)v+bDUA*Vz2)T#
zn1A?!MKsNTgDjH1t@b){AB&e(=Vp&fFZ!fy%g`>c+e+P~|L*y&`t37>Hk|bY7-g1|
zm;kW=@AEMCfD`0}Z-M`yf}q?)LaDNwQjm>X#z#8ePgoL9Lw<3h(}X?2h~gLyV0Mn7
zYLJjugPD??KhJW}xY9ObTglA5PrWW4K67sJtvLHr&to4V`#7^W^T*`L_eXZ04?ggh
zyuINS-)%}PUv|AT8>gG{9?HRhr31FpJEAI=9mqISe*euym4};0JyV)DETA4k;F*B`
z+<F(eoh)C37Q^^ktS{}}J@<rt?3#pGf$5e1Ir;sXZn#Je-Mjz&&-dPXF9($%a9G4G
zOCex|hCo5^k5|Je65L=!Y8`qLWOi5uc(S&e(`H+SM>)k+#W64?=ICl0QCZ8yvgul_
zfXdjy2-`_zA=hej0o-JL$GS((59^axlnw7H32W_pXJPZrb?aGldeg0So<CgZ+V|l_
z=Us1&(qVpbt2TA+y^P}hR<?10{zm#7+uF*WO$#f>)rYKY9?4t?Zw1)E@PS4iI-Y^X
ztWM42jo1%guT8{P7T=s@zC7Uno#*z*dxb+g&d*GKHFd$=2`hKiV0I&(%U$%36N8sr
zpfG?3QAIpL?WpN5uAu9njf{4J8x=&th+tvfg0iERaf2UkTwN}n2H1IUzD39@gcXw_
z2tg1gph&sCJSpl|VjtF=9o#g3ve$L{BU{VhKb8L5M0WM<@ah*UE`?NoSC<mze4IL8
zKl$cj+k^Z4nY`I~cV|fTxPf_hmETTYHq*8_&xKWskw3GpG~@s{-2(uVDBEzn4AKXM
zi%ughaS_#KmQKyAsg1Z-_&rzV%8D5OF~V_X-FR{1@mq)X&Xxp17OGPr^Tq1OhC|Fr
zM)4K*1Taw@mgF&=NjJ?kFh^_NGM=q0%rYjVN*EPt;Ru=x7;xrMZU>K*@M0eD1Pvs5
z&nnU*goUsv1UdOo)}qPdF246^Rbbe@h=7~=c`vt%%7NqGMbw8J-|Ex!)71Fg5#_5l
z?l!{L3%<8vidMJL+}E9=`k>%Ow796;t081_-e8hr-tp|>(Jc@tyiA=f!-~#AbkXxf
z0$VqE*$kjV3P^}yO@J*&Og?_|P*S%<#Fa<Sl1LBWTaHNTRs=JHa=1Ye0`i0CTG02^
z8a`RFL12p0_zK`0P<q&JnfY?85YMW?UwI(7UO!^7-NE$!r`S68<;G*hqq`Pe_v?RW
z_RjOuL!04CD~;Wwnc)H0B5rcm^-EK|u4yJ*jwl;t=JrACD}B4@6DQ^be;IsgbryPy
zyQqBupPE<To9cahK&0ki42TmCk8YhAHu$57G_Ey=JQQdvflmF%=Rr7tCPF>abRRk>
zK~QlEMPOcIv1-a-IBgN+8)~g~ngY>RaV%^s*9RxZBG6nbQ+9B)nG&X2>(na7z0Awi
zpPNCv5580N*4w1^=?8ppI=e*u#(<WvhZ$T#tXz{m_&FG0TB+ajL!(}@!%D!p7i47P
zDH-dsl=0vp&`!R_6$+0Z#~`H+%!q$HZzxo|r(1Dfz&o&hTM`YNUI*<0@0tRQDw8yw
z9{sJ<(Ixx>3H>DZNA55h$mfY^P}IZ-^a^*$XIS~|aN{{N>Th-|^deb6N8QMkMVqiI
zp*}3Z^T#MwN`X+r54ZsY_)c>&tXIOT8ONhrH9B%RL<)2Ru?(mA+K>bU`N(H0vn`m&
zKqr1F?l}Skx=ABFe}Vpj{C<OoN{VhZj}FPb$5%L~1~tYV+}ezDd;T8vk^n>s7?QYk
z7gRS1jM^;lGINXIHxW2RZ%DxLX+oYl#DJG=x+sd+&q9E2I2~=GuKYC5Gt4^B79H{w
zD7390>kB+he-$}0Ditmu7x5xDmKi%Ds7M8);1={P5sivNzzs~tA!zaPtKZEGd&vZZ
z!S4p+qP|F-wIDKqLh6=7$B|=((S+Pv`6g1BQBi&hClllqSsgL7W(r88x_Ai1qEmKI
z(z$zXQ0}Dr2YkY5W&}bL-T4*CA`$L+Y2trrhu&|ruS4&~2p*fyt#SQmLTnS&XaGT}
z$+#qk4un#p^XRF28iSc}6N3S26AJ-k&gE#T4+t1z)jHgK4ZMT1&>qho(?SQz3(MzG
zM@=8@LYKgqySKO{^8~)*w!3g!s8zX0NJ=s|Ft5tWSjzE^D*I0+1(BFW@93fce-vCw
zh2ylKx||_SXoj?@;Jt}p7L46dM0X!d(*QX+5ZDX>uNJ{P*(QM(tgD^2ujKhkfoeY|
zG+m;g$8~C%bLGB*Es#f|rC;z4{A~QS!(dW49WFO^v(ByQ#<`u6BvF@%ZSg|H(VXZk
zAXJn|5-L)G=8~JooCGvhr6ktL-Zjw715{K%yK)6|8?~Fo@ilNXW%d<+QDtBWsv?Q_
z!22+IdR}mPaBO5*Eg%)?4)HT%HB0EzDiZ(kj6aJvurftO|HVf(`5YSjNPPo}(O2*T
zcv!$aCgES63ir{BeD)6Nh{{SzL8#-wa8QO6UPY(YF=0&^^+_OAffL}k7dsH#LZ^4}
z`Kq&HWR68e$UDX^Z=Uf-%c$p^C?XqcJ(@@7^zFNiJzcvZj;lZ}8Bllc<*M{S{Go&I
z##p0f%(;s0634pd9T(xa2m+LtyW?OP#zYYY0)JpG@r{@irJuLrQ_837CikGG5dex1
zKdg<O?qGFVOun$JM)0$nL3Y%*p6)4GHKaOg!7~PN@;Kk`!GK<=e*g3*&Z)zfG(-e4
zLXMZjl^(3ii@=f38bR>Z*o^lJlDW?WBrzN9cvgm;UhFOs6IeX)gj%bac`-RLY9clq
z!9ApBQgR1TV1<P#=1iQyGq5*LKoJoLDDa}pv^*b48XLEW8}xMoKIPd5o}X5jQ-%pA
zEdCG;EPf=*cf#$_cd7?fs&n<9cT|qPGkeeZS+;idY->QaUAyUnS@wa~xgIC0PqIe_
z1TD^;e1C9H<h{~!dm{q2#$>_G+fX@TTT=G$t%a0loz&7B;H^pO7#`ZRsETXt%QSam
zx>65hJJEyZTt`<_APwL<$f<#!di>fU`{*Y;<Q3F}NYL*N#n@PVok+LvO$170nPws+
zK?RIJ2n@<rx~DZ0hY`-P_M%QSa`ZaM00&rW&?)p{t$B<!AkF#oZD@Lh?#7nLwxFTS
z`RV52E?dX_Z~eXEKDy6)B%jVKse0vbYEm~di#?u;Ypc6v1&!K=b_JLH^YtQI1ooev
zu(^E@2i_X6twxA{N$Rbr6U$QrEq$Zf^$V~SIr3cvJ$i#8%Eb6ZD5?J{lkAQ+Chpui
z@aXl8)u$z7gi%%e1{NSLl&u3)8Vx@o^i;_6279R03@LzggK%MDMpmfiSq-lni|Dj_
zewUtr=!r3nFR>rrh+B)a)cDRVJ{a~s%HOGutcu_1v$bwFV>^BC2H$P(&BQ+FPi{$`
zVSh2U-`I6Nby81~@9x7Z!zLa6;pks#o|H(|>ZSF~N&P47?P=IMzy1mKhO1ufnpfR5
zz}}gALLy8299HzIj=JU+Is8#Wo#fBTzjo+Z(b_byqpY)NV$H1tX!-|r@`<Mz@;&tM
zCJ;<DzXYER_mxl=A~;<Tm~o8GpjG6!BxY4KgI^l)5RXwT{;_;M@@Tv5tNn9QX_!4{
z+w{~AuQy(Q9cRZ^-I^5=lT@)>V(C$<?blxk4DKGp-n;1uC*QCd-}_T_a@Ed5*XFLa
zt!?j1jXp9`a_-$P1}m0-6LLIz>C8vwWi!W5=}Jl7T<l8f_gmhOWKLLIo)4}08?bIx
z1_w&As4+le(&SDI>t8Xn4s#liNnQ?<@*wLV^b08UyudmkyM#0y87?a@fTTx!aO3j|
z1qZ;6Wt<fo+94@B-^gQcG*}iM8`8963X1E?Y4Og)<WUD_>FszWCgnp!Y}SpYM;@QH
zI<R|R#63F(d`PH1HDg<G!QP)3m)AqYZN2$=LaP0)r|+GJRbdrH$<Y^`4W;SWw^$1I
z^PZWpqPH*2*gCv(p`l;5&)7GLr8P`J|74nB!WY+Uok;W&Y5rHjm8R4yY=-X&`h^PR
zdSXXAKp{(~(ZYD%NGc6O?i`2#EHjuGScnu9IQT8UFp;Ar_yQBl#Iw*%hVi}@ay0@h
z7CS=aOqf_u+A>@G@r{?(4vUGCw^p3{Y3g?^-AAfFJN_uU8(a6h@26v$kO#|szmBNF
z$=F@_5mlI3^9^i_49G^#wgfGMvqlwnFqFYGaST{R*0S#2VrG!~j8o~=b5B@0{RM1W
zT-tK(M$(yX+&2UyoILa_8_^;3nNEci=qcG;Fe@n>GY2~_fTPI*omR_YDq-s=h4T@@
zPDtOCRN!i}0y8#}JpBV5B}#cXMCao{Xs9vck7(obK5CCby2?HtF7I@vS?Ir!ULqI$
zX`cRa0`U>VnY0K^S&K|X_uW*{p-rsfdpYTd)Sc9Sq9kRbUki)G?h6~Mmex!7$zh`d
zJqteSt_Aj)NaPX8&Z*#CB5Ms*FgeCqpjhmbg(v&Ac&rZp_y}QQ9eEc`kTg+UBY=1q
zWI{rLih%4G#Ciby3c;$6JN{&aF+pZX7P)%Dk;Xw4cf1A!m5X9(mqdvt48CC+W<F$S
zh^G*CM%>6*hsD~0yDB&f^;qOR<&krPZ$<?iZpb~@C@PH?u5JmgvYbODN$G(TmIp7P
zmb(Z<9&ED%z!+2q&zrEW7&je+8FZ8i_7BElIk+NrE_N&gNdgX_Rr05~Ko}uMGKpB*
z13yfy<E^WYA9RZri@0~RdYsCX?fLP!I7CfHgnh=|hD%d#>|*-QUtf07)xYxtpTT@h
z;4xd%zVPu)ShtAHMF=h)Z>P>cc*wQ;rI0@iE%C9l?X6H)h}<iEy_=b#b#M5I<(lNl
zqXKm|`rk7LLNT6~pAkj;fnost!Ce9bgFF`!NfBcSpb-UOaa=qy=&;}rM572M{8Ef5
z;pzz194!qR&>^IDi^$y1AuZ{j`YCn|da-42c!#n3TYbkXffb!UtUEPuJbqsU^3H4r
z?_BX4%WQ|z=U*>9(*1C=(c*M@`C<uj(uU5#bZomsZx&_T5`)81xpg1zk~1s7U%k0A
zqzZclTxtJ#dWc`UPQY5qmb&KcAQ3as8u-|#cqWQ^5qY{IXoj(c3=p7XMzRCcxa6Ka
z#?(C^V2b9Zi1G3ufzZShdTiDP?g5DMFar__^r)C2FY;*f!qwPTNUDpN>K*2~vnkx~
zP_XB!zeG&hS~2=)@@M^6Ueo?~Z}gkn=Dx7z0_)58Ah^@tbMZqu2e&xvPUPyqJ%_{E
zzm^d%CX&>_QANL)8lUKjouEXY2+B?2{>uhRE86ecf*YrLHGI^9UHg5c-nR%&b3zqX
zj*FM>Q1-DEH!zYaBk@CvDzK!0cHn%8_(5(Bf9Z=NHxFtnKO0|>bFQVSY6zSKgqA>R
znhT-cB(ki?A|F|EOgDS^4nN(rsM<2l$Me`FZ1CNG%G}hOb>8Fhi1&$C9{O)-Rvvuu
zvmIvajHFOhY`roRy!zBy=Wu4RuM1&hy~IR^0Inq{2}Bgk4&uc?e-W71#Pz|$iBfj7
zfA6K)S($)C?s^gOsrLT=RDB0P%>V!Ym5XzHa&Z}bD%a8=E1{*kqj8n?AgfeTWTc`p
z(ojajHHb>;+DTGGR(2sxE0Gixg(N;1pa0|aK7OC?|K(iodym)q^?L5{8s0!1nTzDT
z(yMevCho^56rcgd01OmdL33#sPz_j6Jb+&s;`5{nEIp7hA!-b}BEd{z6Y>+Hl(;wt
z8H9N%wpv`t0~3|T$zoA&gjMg-7rJCOq;>O0SS{A|J!E?9_Qtqr+54qs%XH+%Cg;DB
zJAfQxkz?>&1#8oKTkjI@qy5`Tydwmn+df7hN*ayBW}$9cI09i%OQcAKaq-QR7{jN8
zlEa)#8b9e{2->R{6KtdyhKo|*$J;B|U$nL91YuwzGS`#CA*doqhD3zxW{xAuUc+Bx
z@*y%cxeKpBR1ydvP7VyhRN>fhsdytcJ_2w^b|5S)QVpw7BD_|}T)ba+@fgG3@R-e3
zY1yoW^6?|H6yiLwfA9VF-cTB#|K0&<fl^y911x>zdD>fNe`yq62U{r|4N)@Zz}EOb
zHpF>$7$%9Jk7@TuHE3EzXA9EaY@m(U+u0j1S22FtHQXU#wfwRClW~@Ad<bTA%L1(;
z_(OH#_!XW;D2M7pChjK&2>Bini7zg)KoZ~)5Du|nBpFGpOmOkx^&ssq5yXwen@b75
zSKg08y$R7-Al@jth7#9M@!BZt{;DB0YD6i`HczN{0QCVvv`jpTNGuIi{4T^xktNWT
z2<uosKjC($ZpPT$YZ6xbviEj(Y?v!oOD<XIJ=$46<4TyOg)M$ExBxjDpdr~Bass`=
z|KIqcjx))L%vXSsjPDZM$knj5<>FQr98wcJ{kuS*?8S<S)Vk{;DZQkjV2zSFB=HhO
z`{(WAHs2VD&aPGM5=WaMFfq{?mW+6sHo8LUU~wU*IKS(n{$!HxyoY)xEVR9i<d<Uo
z{4gxnc0bx<?~~fifI3ZMuJ=kE?%;Af2`LUB0~eXUJ3(+v><n~-M9+vsM*-+*bZHI~
zW45pllU5SC&xJ-M?l0Cv;lSL$O_>x-BQYI=+UA2Ry`&^}vFA2JmmP{5?gxw#x>T|<
zZ~D!3&b}XPl<>fHjAGq7*O!XpPOOTRsmG?^BA^(xNYbP@c)tX&gJ{^F!UnP_jUHCG
z4qHb(^F|yz2V)PU<|cIdV%)9iM`BB`@1=V`kW6u<(K&dKbd@y!x6=SWo=bR<i|&yb
zxCqIih-eJSh%k*p3GQAEKKAMmue^bGBT7dWqyQYy7|48_K^~XvvlBxb*i!O#gw@uC
zkM6!+E%CxozRu~K1|6~Z_JHC%cLgim^12Q2>JLsADOy$3_aPrj6bSkxhBSpTbdvo3
zQ!)WwTK-lY#Ruwj`ttFGA;1|;gdJZ--q6ekT)CJ;^JwQ2`-~hs5e6ZE+6ZD;R4=Kj
z>9Ew;sD~8FabQcvut$U`F--R3M1=*xkAwus7I6vuK$~!kq7lQe#qfX5@2>5$cpsv_
zWbUKNy8(rPF&-smO)T226&`aQl^z>6GXJ6LYqbqB;a4ITYX=x`;YSm|py1_UVw{m*
z{}r62)3kRs?SC??E<8Fe(_u<!bVT*@k2}bMKG<YE6St0-A9!LOKZV2_sDX}RuZIlg
z0CGs7x<_8X!7I4b`?w><*C)Z1jg*4(z>$=2l)P0ah60Bkh1rz$jntD8SpAQP2dvK1
z<0d&J%I5$Qg>l|m(d#<`bRNB;Axo&EFnx6CCxtp$EKblcb<veWi**C^uaO7=hk6rC
zAXLRq;F1de2w!pxwN-!_Ag7OQ)|X7JYkw1<*%^H{wQ~^08+N_>LyTcLDPe?p#09ZS
z9FUHZ-$<Af(TNk{H{pmgwibN^`ATU`%qXV8gP}=+&skEkJZ=~ekHQb5X)-JZxPSKE
z(0NoDiIhpL&*@PyuUzj>``k>M?c5}B1#2x_+gO}?_xi5!lVfk{G|w~3La)Rv01ZKO
zLH)rnMwS2DEf77ICRL(NWiZJ4t5Ikilk0$a%=5(LM;h_F{MT>>&cMGUtVT359)U7e
z(Z9y!+J`6Oq%D_5w^imRWaem^a3pA0eih<PJXVfI2l=UDJEW_@*bb@v&35|2^g1)<
zzQmi}hR#jJGz>@*1?sRB;y8obm9Et<PQ6<n!pS6IvnZ+=1VK@B7A-RjAmvb}#9x{;
zt!preaGC&|GzyIH01_fh`v9zRih=OR@4tjO34M{<2<pXDhaZw74wd5%G`dNcs>q4W
zE<g`~mI95F=HxsnSgsd~FH%fYk#Ua6#TY}l3J^3LC?awLJB);qk^Dk)#5j=TlA%0e
zrA{M_XW~XAR!Y$~>H&0%dRA)gn>TftM%rN@IS^JR04y#XN`_@pKS-<5XcPuogGcpX
z_(IJ(!c(Tu3-jZTo<c7rIJi?nD%dcAP0~Gu<lX;&kvfU`!ibEd7=R|jv^c}mgmb{7
z?j`Fm0oo!B;E-+suM1QA5LSUkAVrGIgAOa|C2n{aAWRBmC@3wAOOC|`8ZObvuP_)h
zTr^yIZUWKqPB?+{Ebx#NyhU4cCn*jmhDK+jwy-P=NmN^WjiQ#`@C5Ss{}e!3L^y`P
z81rnHTzg5ycr~S7#15MgJ3Qb}GtSeH-5qQ=;$&TZyti^jb=}EFN}WRvHSL1|hAt6L
z-*-qy=`ng2Un_}DO#N#V)`RTcs#HJZulYVYqUL!fUWM7t|6JIXZ=d4&Kf|@fBn0aj
z;6nLLI>g#Tf5Kp+GwF+uW|B+5zd6ZivBQ`qvgRE4$FOGX5l@vA*)zF_Szw|}0;>l&
z$s%wT!x{&g<XHe0`?;xUp!uC|LXFb)b`Qbajvk+GzD?=*M}K_<X;01ix8pCHoju$=
zI8MqaA#9M35ka84O!AM(4g=v9GEfs-eWK(ECSDyst`8}d05YKn3ri>i@Pxfk07~=#
z5(q_%M(R*v1cD$ahj|hbl3^A&0=#h@N|FOUBFjV5qi9$ejn;#pdx)y0#Qo^A4ec)Z
ziW;FZcNJo`X$KmbX$=L$Bc#$;8BrXvFk{2>fQ1Uo5WV^#htzJ7Qs?_%!-ywJ7JN1)
zAWLyDP7W_&;s5^>pqWF{Mt5NENdb;1px1|_KZ-e-ndCQliu98jJ&c}Eg}PIGNWSeQ
z5#^Lj0Le*)gR@~UVtNEfFpOr;(Kg3k0bDLo22z;RSknnzViw#R#=t(@%;>Boh4xDd
zLkzEFi749ym65*?vWTW~mx23fnKTaf&!87v{#~dRVDUKEpkk;3@)REO3b>RQqHsee
z2p3-`xF;Nk+#U@hV3=G00~U&5!hl2O3~&q3go|Vr85=l|04_)KB*+{CsgrOLEJ8GB
zW<tdcH4@S(yA<9QqFpq!3?0UQQiz2@wVIirfm)z?2_6YO<IjI3lN^$G9G7Yz-l}rw
zf=)g&<oA#lF78Ly{=k`-A?P7h2RTjSaA=Y&1jsbIMXIDK2pG(7AsKgCqzu+-2E>zC
zOXv~;4dT#o5YZgR2_w8vNCX1aVhTDq2Ed?eFxy2$cmjfwYOuMOPjFpafJO*CemMS$
zi-xMhfv($`qX1?xWGNX;RYh(<%Lzz`uSJ9Sq+v`tenuIwBQ?h%u)}0XO7OYKTn$ZJ
zSdytRgQgrlj*}HRDk}+N9rwB^ZHxrRBn{eQz#(NTk_B1Oateq4!J51-GfXiCttQ~2
z#o&$57wj+`|0z-&v{{0SkQo=J$ZgA5?2Zd_@|nxmlV^ycC7CY*#QzHXcybDjkR%KB
zg=L1ZS;;oRG+B<tM9C;REs_H|2BTu(F6F%}EJi;FL`H_G0o*vrU}D;?lGslKVIoyY
zn)F}0;aB0USQ{V1gaCi2Jkm2lc;Z5A30Mr0w%~yO(=`%qT!6ezB+*L@JQOAu5TzhO
z{iH#qFc@5PlW0Q%2~>*B6ZjE#PY0io4A%~!MXK(=ns#8}C}s$DbqXsalm0mzO{^IR
zn*;Ah;DSt9q9}!rd1Itm|49p5A^}Ldv#0|E_i+*8@K49cy@b6WwZH>^z&VjNLfA=~
zO{f#s!w3zrps7d`VOL7=xltTEhGyXr(teZ?aO1p7`HI%+2~67%qofh{O$L@^=E(Bg
zFy@zowTv8^U9s38cF5EaxngcUhensoRJBjCmmNo^eI=rj17uUw`Uk91q%*M4?L}iq
z+aPDda1R6w1K;AnG6hm(qd~Hn;^ZwZ(QWw=;tn9L3|0oh&|a#LWf3K$IO0vWFn$>|
zVO|g6DJ4u7;j+S@4?901V8RX5P_?ylq?yBzph=xvf|@vCY+5F20pv5WR4kq5l8CVZ
zBybV~=unrWIAPRl035jI0F0_FS%4N{N;qAphlCHnVU$%u-I$38Jq}PaRV2WZofmWp
zXb>x5IE|C|uYnc37+e|^|0ctAxZKfPCX9dBSCDuK2)!bV#bw$qUO1|h5nvRK<)vrP
zC8V_JOk0a6<mOc|GZ+q_k_D<mF{eo>yMWpu5dV6cg{TQ3JQNQ3LahYqT!bsQ5NY59
z*lJ8jRbt`%vDm~=ccq#xNhpgHI+lg~59(ZU$}pk^xXjE!g&-hFN#a^yB!T#K)Iwi?
zF093fn$GUUcd_^HU@+FpbmMU`#DI(Fp3@TG2e)V?0X|ARuojzz1Y<b0<zNowjuMt3
zg^&aa=$L?8L&`9W2xPh><gBfv5cB~_7>*u@@0w`XA}K4;J%B}g34{EBKITJMCh8w%
zE;mu)(nPdKB?jb_IFrqx(@CBPOX$zwg4|$pgOS6+sAds#hOx?3$8pBfC3Yz`w$R2C
ziYf%B#*p@hNl~$V%1l7uU`giw)i%WEJfDFoe~-7La19CpxDYNRc@OYbUc$Ihwpf)*
z+MAD!h0dtl+F_Kg5)~S+?#A7rH?hi0&MZ5B;o}`siUhU#F0>H+I|uUeKdl~T=CJo>
zeL!5`-QGZ*jEAEf_7>Orpf(C!Fv9RoLS(0BkILajVy$?Zh7no|Mqy}Svm|V#vL3q(
zW`fH=z?yVQtYS+WP$9BTN*L1356g*D&u-u;Cy8XaN^hjJZLyXcB@8oTBz#yjzCB0N
z7e<7nEEj5=k=#l4KJ9}dqlHO0f>A(ekfMM*)?>71zZvHiow=3ve#_{d5VF~FUPQkB
z6dNxk8#N>Lo9%iD@7)$m)OqZ1SoF`EZp^%hbtqnUN_|54*uX-g#}5}wvF_AQeZ6$+
z!s9A)GCwZP@zlLu5c6c4Vfua3vu{pd1vc_JiWBjNF|qPqg#IVitgI9aS%JjE=J1g;
zN1B?Q1>N@Wde~8PHj!&llI+MZS|-p64$j1bVUpq4hdM`7ddq3EOvbYD(MJcyd2tGv
zwqzAYG4){cZMw{Q2nR{gXnqo#CBce57WO-?0J^uuFof*=Px2;|E=<o(?e18w<FcbJ
zRPzZE19P5qt&AOIPusCDeQ*8U0E43syd~3?<}SA5VME+?MiFN|*S+;<NEq~0y4$f)
zPd@IB^0Be&&Q>>Lll2%TLt59mA+92AoC%gr-ud~={brB(NrN8xa%0zB#{yLwEie({
z{=oJ**zYXjaCsE^f+9A&*so{KcvnMrDg|KdZO3E(vK$Zw6q%$YcZsI28r>CEOAIF9
z0V0qMnV=IQ?oluSa&giXdwW^z$vzZnw8RiQAd@f)4<WJx)E(3pmm(@U#?Dg?JLJ0*
zbOjyb_SGlwVj2T<j!DZrxwRs@>v6-OH>Kl}*O<SOd#Ic6gf`N_cHyzH>JGff7yf#R
zf9>1$w3)TfV|iq~^z9LqQVWlpkDYd7PCWudXAfid_OA2?qbfzTuEpg$Yd<iXw;0I(
zWjdi|X+!55SERSn@88a3VwqD+)JcmU86lJ39&GscT`)Z<T<Wi$<>$|Y86;uSB(a&?
zOk9b?-ck&7TMctY;Y3i33DHdi1tlzCc6A~gNz=N;5m`hZWwVCJB6d<(k~J(Jv1tuW
zjs+EvVj;Vn39$(s4Zp1_xW^43fK062pqFO-c0^5a*LbWf;hZ>n&?9zxOz0xN4SIij
z)YMkdBoZD;j1|p4F}prkcS>yZh=%47sh=NsCzk4;FzgM{J61O>Y_Uvfm!e?xp}a;z
ztjN{$<^&hxGv^tY%C$~@kZUox72D_bx-H5#x~<>8^K(IFr=L=LxBL8S6^EW6i9L2*
zw&wfx%~I$PEs7=<;u6+#fBblNpBb`(s6+{_8W#yxpc{b5gy;mYi2BDsZaR4c0+x^i
zN2wFbISlY4n)xzoHQ8Zb%r%YJVPv|ZCYEKvvaU*wG<;~pP#62U%Gf=1T#(-j?9;`y
zg?2sU)QmVLBBF`+x0Ez=SFi4HuDrJ@`)U2Pz<dmC_r>s#m?!q~aib!(etz&~ad6mk
zZ!A=3Gt$A$kk#v^kKN8aYMP@*?5}()A1_<q>8tbi*lDk2N+)t?ubd5zZV2$0KY6;R
z=c#=<KA#>IZZ$YscW3d_gdv}Wigmd<%k)y0dY?r;X#>bIW0x{PE*w7+B0+gOb1cBI
zeEmrd389p;=rn|cFy}>lmE1w&00t0cfD`1vIJ#1QG2c<9JCY748JQ(kca|$HVTPWz
zTW_8f21|s)wUy9hCSWkDB%4mO6>LvZu;M=lk>kF%IbpQ@eCXM+>+VKpYj0C3|1$YO
z<eX#HW*&R<M@SS`udz2g`ffcfSH<1xZFmCqBvJ8l$7)K;yV)PN%u9OXJul5>q)@)v
zW|YUA@-8GeYU={$vj2W~G)P~L!<BaH-hxSqW+Mc*FQ%yogcW8Zo*1V4rp0?_A8yW@
zkZ&j-*U*0<BPPT!qt1-t4F)j-2o2EmEFn21Gm?o&gehpx%;bVb#03q8)j;VG$VxUl
zERv4hfoEXs7Aq&-Oo+MD+$c3}2u(zf=jRhc$ehM!4lcgWKajK6gu}!@9a0axOme|R
zDBjUMg(k@XHWn+DC(LoKd^5e=eF66UCiB!@=qqB~+eKcx>`nEbNbr&d{SDG$y(1Rt
zrb!&_4>C%SRFsZVx<BYpU9V^nY&6kO0fQB~qiE~_TU&DxCPBl1hae-YFDYb8(aHD%
z6j;e(fn{w^F@Y0TOz;e3VWc=9U#dvaGJ$i_R}KsW^i`6@%scwgf-O5{NeC@R<47$-
zim!Ng$EfV)aicB{5}Ba%2EDku!r8)=OOsen07b^#aA;f_)>P$i^f4jlof}q<2a7Dc
zvt{xA=O4#;kGLT1?Xb5Q3ypg0e^TdjdQQhh8Z+1s_K{ISq!C7xG?4W&_~hW`Af~!Z
z+e;fKdtXl4Y(zu;^$>%EUWY@0bTpO8#p_6c5zPi&lS7K`a7G&ad^r~*hQtyLZi|F3
z8nu_GJQ6b?PjdkyARBoJ2LD=KW-v$AVw|dyrD849Tq8wU!j_#SyOuSmAt9ksW9b+f
z@1&ezZW3h^5QFJT#CM=Xr344y;)1{_izfx$&!o|KFm&NjAy3GK|4LXxT77ZD{ocjZ
zm=R`ot;5F%Yi+7DNZ+2|p3sZ^qs>DMxFZzSLoMt(D^uL=Gxu1W_u@GvltL1B3v?+-
zNopF3*cCt*+aG+Mf$@Au=McJsu+d<{EZNJY(3uJOL^vIY#5ftSA~Zhw(jA>3eWp7s
zO46{Njf@a;JdHJ86l|oe%-~~@QMQPqQKT5A0j~-FNIi(QnX|-J&QHR?G<;$Rc~J#6
zFF$OZkz)KPm96dVJ9N{eZ_PPwGtFW7p?<FgQ#FQjJ!~-t(_{YL5mKA=GVbTzka_;m
zXU^Y9Cq54(6`*~jkmxC#eX#~67UF*xbmEeMjXW++X-IMK4HJ?EH;t>I1Tw?6c9}Ui
zhapX9ZFtmykWq;bu)N+t2xkVD21(751nI*Z3sF<9kBnC}QtYPVG1NsES_7^w^9&a4
z8xSYyo<JSo6NT$kk)k2)G}417z0RXs9$R<W*?7+RJAR~x`=R$cUvItk!b3m(klD8V
zPx|T8r<7tr>+3-}3U#KR+1pN^q@D+bU@Q+MIvA9kL5HcoQnapFbq_vk<gCDb7-~TY
zT7Z^}Wrur)v^^43A`K3}fGngW`hjjUCrW8DcIYEvhQ;6|D10kn(20FQByB{SLya=r
z2<R?6A4b!{1F+IzoG={PFx~VVo@YShphM)yo@B0g5<F}Wa`Zu-&few$f8C^;euh&X
zteYT6>vRbI92@N*k$-m=i#~mH<p_zvvFaXkN)@cP#+|=LtOSZCMfd}pal|~x52LD<
zj7Yj_yx@LTMouQ7H*o>>I2<J&Zq?AxW^l*Q2p%Od)fP?+{@OVvX!1Jq5}$UnKC^92
zfK*T&2LmxUQX;qtojHYc)PpHAm~$$wkS;-bM&(t1qfE>a5=`zGE@?^L!$o0+sY|01
z_cjk=%IN)RuT#^WV8`J5L!Jv#{|dXd`H)lo5Dm-Tbd2?GToj!D#)q~~XMGs@LZOGr
z1!;f^AXdaLq`o!u?nN)SnU!Ko*C1k+VlGe@TsI3E%@L6T9#fJ_s5Fz{2S;SLN>&&<
z(nyi4VVe+1W_82(qH~PG6?qnjkJ(DulCJ$98_*;;jNttL?HKpB_r^i+98}KS`FV#<
zTI`(()$NU*y4XA>yC$vQO*dod$PvK<54_LrEq~2?4`O`YfrXHZ>#&&!ccg<O;VY_N
zsDfIl6(J!K?E6zbQHH=d4FH~wIWNQs$Hvu+(R_knu070{Fk37@Nbp6VLJD0<O6h)%
z(ZrMHi)%xS-o#KRS&<yAdX{}X)+@dS!a(Ak$)Twca>3gX{!@B$lBSQ@2JhTxM5%@$
z5=9ZDB!(D)zNc7h!&YP76MU{I{^fQb4ndTLKK9jfIQf1cP)VhQGuk}){hJUeY!^X4
z0RE@hl9xie5)}s-0ZJl37ngD4+v_fI%Z3ro0O2F9BteFuq5arjqtF&vS0Fc2l0<i4
z=Almry-G;%q!wT2hKv#>l{I0CG>8*!DmEUI48v-}VO%Chic6=-E*V2yaJ&pDLbYHe
z2^+GjI@v219|XD}`SB==U`uKgHe~8#4N5Q_s=IMo0=ojzSowBj7A+n|Tks);;08R2
zumt$&cNuZ>s<Hkj#;mOHFbN)MJ3EXfg~jI!k&B~4UfQBVbYhrgQq|x9?jFl=znx)*
zbaYnOavvL8S{Py}=rEJZN<v`PFpMvmG*OC;paFRyDM-biR)<CMA{QGJg`r+V{@@s!
zlU7Jbk#AcrzNjJ=TsWjC_77iV;zSsfh4>ljMl6`5b9#>C6A@TAimi(K**P>xM;Q}L
z5XVNO99cIEqO(AFhXkep0lWhAn1HQ+4I4Cj?RY$oqne~54#xjv6>vu-UNa1HXs|$E
zNnjKg01O(J&cMnK*f70-Y`;U=L<h4*L7aq%;fY-sCjtdjV}I)vJO&j>-66$6HC$4a
z2AU5wjq_9)H~z=_cQ={KeSmsQPcY#ya9ToOj#X%+uxA6n%>u)7vFHqlMqC$cujs5K
zFsE|FQR)T6FOZB>>d+p?845vsh<b}8JrnaSvFQto78a?bMUOJ-!RXs_;+9MBNL5IO
zkg74-8SjAFCPl0faUgCvkN}{GcYT7L%7sw;g9)0H9#-=St~HwIYM+3p5=AFrWHstA
z@CfV!7EP0K+A@KDfSn6@#p6oRpm{ko<9~0`W&yk-Mo3~O5Xlh|5(LN~2bmG$l~{8*
z%CqO_&?j(hRor-6<}F^rS7rbE3`oZ+{!)^!go5_)g(9Iq_0OkKFjLJcl~YbiIAeI9
z-><?wyUS&YWvU}aj-9MFdg<>wPaOGu(4(68C3r5+ht4vSPb_Bf==e-bK{@<LxI1TL
z-><^med1fy1;3(e5>*QlrwQzL2jGS8$oa$;Ue%t$!cD)5g}YPftv{Vywl`3fkgE8s
zePbujcDrM`NARbCGg0tdd-xbjOY=tp^>?ga#)XCb9pMZPKCzt0*>2Ubp5*S14$Fi4
z$t!9SrvcVc<T|-!&afxT)s@F^i&R1#sD%?gc5)+HGA$r{(Pe{wX-+L|qCc;jW|gXh
zhbt@{2}B~5NOP)VNv*=&hCTW<szRH`53L1gQ}SHzILBl)E1~G;yHHf}$a+1eI)Z*u
z!(u#c7HC<d&h>nF`**#800c-OAxB(9EJXD9lZRtS1LFZ&o3RrU%v#T1{j$pMr6hS=
zeI;o}i#9&u!e^15M#u^D1;!#9Y2b`>k42#g=nHRTJPHVJTokwaZ_=a3Z6#J1YRd#n
z@g_Vsa-L8y*AtN0K|QBGSD<<X?d(w%ddD$T;snE!RXS=6XY=SX)qqQ)>Q>XkL-r0I
ze!8r=D!&1^wjhnd=Up|O<yJk-DuE|gO`mwU7$+0!!xwszi;k!s7eEnHwTvy&=knwn
z+DHX-q0PkK?bWX*mnoV_;3+KQICk4u00M1FKnvftvNBzB)32h_FHeeJruxH2e4uXi
zu{}vW6TlNqp6y|I@ZkP6sp2cf`KXOQS#1o?IjG9q5a?YkeD{@50M$=MPiIXAp(&3X
zv@LZmKf#qp-%_(xHSwdTFr*=O5k2uMI?XaOMf-P4-BNq|tLW6Hb^_W(2MEQ@c1xPc
z@921K7>xQB(}jX*JF-ua_U~k<bXL<feZUAHt!jdG)H>+6oshyee*!D&V^bZ@=)iHt
z6m=_xkf)%jUhw-;;^1MC7#4u22Dy9MZZ>rSoYyE&KY)iOb0VY^UbY18SFNl}Ruedg
z$&*1yg6iL*aK;>wFzP0#Zt`rE!%|ySKXryTj-3pqnd`kw<&|gran;1e@I^Dwf<f<t
zs+zvw)PzWNi@1Mjy%*cA^b6Dl>Q*}igq`BX?v5f&``B5-bfxGtpHHZtG|eLQl`tUP
zpC^?=T2r;F8SMmFfLw|b?+DrDp!=;`!Rq?FsqU&j06mH}DZb1&M<~(9$Hy0>7(o{b
z4A}%mEv;Wm0eH{43%FTjuv-P)B_P`YDkWrV1|ZC)SSi;k^<6_-i*Qe#@B8ls>Q<>b
zkHL#Fx5(?opPe=82f**(0=$AqEj}s=c<79-rRs<vRmh7^>Vw?PwmWAhd9;{n#qOwZ
z^b@T3)Xyk8CJ5S$2TrsmTPD7BY#pXU#1+fjkP>aNu?M50s9AkcJ#n18Y)g%5Vgs6)
zKG)l-2Pfq1LH{>!Fmm1<1%)<?xJV~EwVYa&Gs@dT)^6oBo#Guh{9}5utSmkCf?cDw
z$@%Awnu{FXi-dx2lbC}`EQF!a7qkO;t91ur`wENII*gg$91CV<$OcRjy$}|~Z}~cY
zSB?L|!)wx4mO49K_F`J6e41Wwru@V8_@sUc$P7NKn!yjqBjhaxvk!d(0GVi{Nx+$L
ziDgD7U=<in0rB(bljjV41XQ&=YtmIB6yyvpZiw`M-hD4I$>&F5s@?L@Chk86Lublz
zd=Ivjy?U^0`*uB(<i_~jT1C=d=T^*3)?4|_W{XuG^8!5}+gDkrJ1{-d^-_VoY4OKd
zo3~s|`Q_YP`PKi~k@2fT^D4{I%k$qkf70?=BY8F1S1aqnrXa<~F<p}{wRxX={V&a;
z^!&Ctc9z1<nx&=3PFee0mhag-{K3v(jmo4Wu35#Sg`GqH(vqLAKD+arsb6D$)${x!
zex8=^TaV|7&ks9_x=n)nm3A97%XS}pTySRhF@d^RXxI=m?d)l<D;0}Bo{x`Ij)-Me
z4VfCDO6Y|@`IT_DXX3}%9yl{?hsO4vtLq$xZhEVkx=tJSw(`+|qS4Xz%Qq#j@IAf5
z$aq^TuV?*nhYIcIPG8T|U;XZ+w#4rE4&QPYZr!h**%b;=tDOf972SXQ{Nm%u(@)B{
z*Y)3e<FeFjmO)iT@bU8&v4`(59`i5Mb_wjCn(5jWb~z@?_4_zqNcDCtREwAKFn^e5
z<=Y$8^|AHrmW!+QHHV+@)5^K=QTLHBP9|%Pd(B@B_Dq}GanCxV_7<efZ?-?R^810v
zL(7&YKYMy#qhy7J&(<G-C(Z;2iA+=#hES+Bl85SvQleB?1qkGhyil-J?PlVDh`QVZ
z(LZUtK;N?_;}C%dMIea-p2FQ@aUE)hQyp%b1d?$uVk!)8O5QbpL+`o#{r*iGi&l(g
zEXr8l*VMe@Uc%t_^5RSPl51nFH_dYBoTS`UTzi3)TUIA`YiX0RyK8C)w_tzbnWgpR
zZQ6}hUNWDu9epjHl)dRbkW$ER3er8j*I~@w3R(L5p_N8Oh06I4E_(?q`l8g?1E>5h
zKH7E4q&muT;_EUkH|yr!=}R7EWv?z$vCXrdx6Ny!kbiEJXU<iZxh9)@ni{@O(#p|D
zcD$IZH!&@*+-cJ!iy7zJjo0S7g^wXR;2VT1x<j~FU=IytxR4SL$q+L_c6eiBBh=5y
z{}h+Bpc_Ih7Xla{$SD8u)B#okV=+;mfQWiT(_ovj`ly1ZFKujII{D^a$-8|l#P52s
ztYTb4zJuo{$G}skjz!IX{#a?L`@`j~ix#P$SDxlmCev5BYlarLbMO8G(k=e+ZSqOC
z=eEf|Rj^5^cC~Lio&Ub8%Pu=8IjiNvroUC@2fmxX`l7wHpKsS|TMt!NpV8J*8?`di
z57%r=O5YZ6y6W3t!*;2BqqO4O{7Yl3SMQf|Ogwu;OFf{yFrnD<^tGUtZg2E_;oFn+
zDMaOi=s^ULN~5dM1M>NCSEYc+2h!_Q^q&BUqJtj-^=xuOV!crCiimKA%7N%>Uks!^
z^nCvmazZiV#N}?st;(uNYvoe%#^rzd+SJrv_|#b#rLMZw+1lnug@4;cN4Jh6o*$M5
z2kd^`bp3qzsB?CQqJ0OG@)O>9u5XJ9Oi`%a7$5$tWxTA7^@(`1i_JlueS$LGbMHp$
z`ow?uA>i2s7FmU)9|%;-=$ItZ_0qR)m4DW_Gq*%w;TIoX<-~Ybqe>W^GOudVJ*<e5
zhy*=DOHcxFQ3*rk-4>6I8gXY0?QR%OAw3)~P#d?mH;9{;_u*FWPEYsQ2g5rnCXPI`
zaqHuNGlg!MY5g+l7Bk;;oO=Ct+l5UBgf5;_|H->>`oR7t`Kgf=MWaK^Dx-Rgn5{Xs
zR*XxhcWPT5>aShpwXA*h!{@f2dRD11eP)?vZFOq<F>RdnrDw)#zuHYo8FL_1-}0&J
z^sPC%OPoS%uAG?gZFS^er;sAVqcu!Ev6~a1oJQ&s%vh%mPeH4HD?lQvMM8moXHB|W
zxNckutQY89n2IpPaAwiiodn7Gfz6@)uKT0eYkdEu$?G;Q(`8#&j<#W>waBn|RXd#A
z{A2cLoAa|m*XAE}-CT9xbZpQ`n>(ihWW7zlZUP?<E=`Q=Z@U!uIQ9A@?Nfg5dbc~=
zcy_i~nBDIC{a{fWgV*!J)lpzJkaQ`{X?xYhvCFp2Pd40BBS<Z<Iy>)NK=0gF_Vms3
zbe-A;Eeh?HdtYA-ix9F#NK8#ABnaEt{=!IL%t8nZiR~|$;WVLjq6jw2gqFC`su>KG
z{ej@EWh&u<9y9q3at-fJd-Z>{bW)0)ayGj8wED+8K*C*<>q%?;w+tswT2i2KAU}EP
z`m0I){A!a&Rr}5#;wLsZtg)|LH92O1&zjrH>_)S5jC(<!S_T$e)Rn*Tq|?>Q<S={i
zyZ-79=_1VqiYFqP4OHT7IvOkv&Jvh-lynqa9rc-KJTuvFLk|0guaG}FE_>D1(X;ZR
zIR%pIFLz&@^vWO(yg~Sdcn3smG%GQb7`-*BKQ|^0euN7I!%|@iu?x)64MKZ|CoKbh
z)K9D3x^)wMe<wdL<jYIB)n!FmFPcx>S_Swoy%-X&ygj>iaQV{JrO7F8ZXXh;=ATzc
zzN<8M!Dgv7lcV!qW(^gVnYu6==4WKzuIo<qaC1EDn$C=?n55NZG(X<&?qH3{$MjJ_
z{zWeT$jz~fzdTd^n;Ac@F*@F5b77CBk&OD@)roe&4Qi_suB01Ji{Em&c9HhV($Lsh
zCfj+*+PjngzWYUcg6js&kE_7)Xmv+ThT-@~V#^6d1q6eXi4_Psgu6s_o}sdmGO*MS
zED8^4#OaUfl%xQU^$PTPbf0kBN|=H4!2F@!g5HU%c_UcsrrvPnYz~6jSdeC=FC5sg
ze9zx0f*a#aXIG42Ip4@mvOkz}Np|NJ9)D(I_AB|br`XAjRsXC%_jbCRx}Q_MVCPlW
z?=!C}9(TQe?LMQY`$9}dHpibgd)4uqr|wA>{Lod+IqPy>OUb%;dLggBYju9joq_F|
zWo^6ooz{a#J|!MYR%$6UkGrvbYfIoMxxrOhRmuG>Gtb7`RJC8dI(G7FN(_RxC?}CG
zGvJw_!PnnH1fsn_y@;Ta`hciy=mD9KiCuSGAs83wDz>?{au)xi+3y&g^-!`{cGcDq
z7UPl{zMFal<yN$5U9KHH{gfyBD97Ve$QEPY?v|+BPp6L<ONYAG>W;H{X;Ia)sKv#y
zY15HydXM%0(d>Hmc%`29;-(F2?&M6`7Ep9P)lO>hg~*vI7Mc&P4W_jC>RQ)kJ1Hrs
zgx)N&n9+W^OXaVk+IT0eMN^Xv4%gm%y;9BfKxbNGU`gOcy$+w}PBL?Nhd&l>`cA}n
zcMa)74^a522A)dPt(%FBLdv!U#S(iP<~ilb$or$W7C#v61gDF}kGY}RLznrO8S7zT
zbn&!c^Eut{rU&7wdi}y9yL=b=uVZMu8aCLFHKn#IWcso`Uy;6)#^%SBZ_0ySwdP+9
zzgRf+AC_5urOP7Mi@KE?v>z3EroD)1{rJ{Zw|-ws)(hst<mGY>mb#~3|8m=`x%6B2
z<+2vHg6a{=RaT6?VRuCPg5!Pjf-yg)JyKd4t~<e{Sl7QW+EZqunO8kG#<ullUrFsV
z5_>p&^q%J3=6yfps;)fNuci~ndJd67Tm6a1A<jNjMAY4tD_0U3^xH_Lx@*?#Swr6`
zdbQLI9{vvW8BWNnx}?yZ45cx)rkVTY6`wI{)~p|j-?(&AMbrTm^SxERV-7n=``9G+
zWH4K2G=w&OIGz7|I=w&Q7RNMc#l2G-D-WICkY;sQbMm3aPu*Xi&2$Rb@bP*1?q=&f
zKlk*>vV0#3Rwy?)%0GXmn|EG%$&&76!l7*@h5hamG@fcU$*Ao8$L~Psj-jNXTWb$m
z4K)wNZ%z^Nw>qYp3i>q4n};6yEy-0|xcyw-ITiQava4PqI%9pIY1)o6w^SiyiK8a&
zV@{zfRU(z>pl6DF5sbJ_4tB>a1VN~DHuMe36mM8-5)-;!ZBBu_?0oK<m?zp3N{ghd
ziW!j;{hzt2={c88RB@?dKGD{W3W0!I=IZb?$GJ9(d&=mH=BuJ?p+oD#^T~PB*V@?K
zH*w01Shl}8bf(eSR=HM(*V8?H`8zLGU4cnS2|!|pXiAA#Jcacf63VeOr#x7db{pb(
z!~4oZqbNbr3;XNKVAZZ&`P1zzSrv5i274R7aT_hxtBiL#I#fS&``kjOve4k6`ZHr#
z6;i&MYnj5s4e^hrZ=$;C04W76L%FYrvg{HOLfA@?{>Rq`rv*}Uy%V>7tOkRVcOx>T
z-rE!OtN2tEF&aQ70ur;b8vHC4iK>Y4z@Dxlj>Bw^2-FYwapEhf*Qus^N5j67YZnUs
zX(X;-x<((VX*uZ=ENI*f0}J7oh1;s>GPhI{sjv*46|EVORIPzG5lfxa7rw}PG#+T0
zX7%yTO^P*DGeodMVcU!sBpR~PAdq%069<?Ch5=Q3oSi?wN>#%{j(=>~G3M_mTldc?
z!ug!|MA8(}0U`OV(uE&74RxiwkqiC35##fLE6C&lGiHcYQNAwKzqn-Wt1ZbfoVZAF
zKZQ0U2^2_68UG~SAh9#lT_hAdLf{}{>||9T(GHyQ@I|nc4JqjfuY-DT23yk)Fd_c>
zR4urq^6%&}LyZJH#?r(?rygT`U&x6?<j*zbWpC}}s}5A)%!a>6mBd%*GG7-`5grn%
zL?2FBZvNu6>gvfF(zSOd9(UqT4N*EWkA*u#LIEr1{BT*x%hGgQe{5(cB2=LI6Je;P
zYJ~vOn4+`qnr(2h_fOyj6T3*yzBSs&<8`|l+2{5+J3hI(obE&4N${eY?oYrZ2l%?K
zjDkil0N99SO_v!&XLrY{#^h+9Xo-Jv)x~f;=^DH%xEBm@G@_I8K0pi!gey_MA<L}W
zpWFue(76#Yl5U^!u=mc!H77JctrSfFD3P=MkCF(8mPOi~IYM?6Vm?ZM`E$}7=qz)4
z27)uh)*v7hM}#I=gc=tqRD_g~x<SJ|bP~V`w-sBQU_cT<W0}224a5vpx9Z7|pWH#Q
zCaz$VL}UOSe6^RAcXs$pt7AGvA&5ocedsIvXeEZ(d#Z9`trFK^(L`e9@#sqQ08~iY
z@|#Gyv8vvewSB7zSkY*I;*yXD^b;tPf)%OmDj;DRxMYZois2&j$(Fo0bk=)^LA+Eb
zFq%PZjgAfhVL`<106Ugh<4M%){qgE~!ogz+!-8eeSyYVPA`IL?_%1;-k+QDMo@G8b
z(Ngl0qTI)FHK<i}&@FBT2nkjk+&>bWJ8_A3+_-4fmjts~gqnf5SeS04KPU1iPe^5u
zKIQCi#Y-&+&KlK3gh<gF2Blg#1Os~<$n$Gn9s~_iVgl!^evF1IiC@F}@lN%Xe;O1S
zm>J)au<l{+<F7a^&>41Q@kmr|;>3w5C&e)`n9JV#Ky!#)!3H6*x%dWk0gHO(JE|sJ
z18|5mVRo8b2MXaRK(i64&!caxOVz24Z`tBudg$rT6%N~5c!)dy)&>Yv#_8dC2V+SJ
zFX8jz(gxka83-mCn+PeVpMo^J5}q7^Tg8H1h5$n<=BiRwAFd))YiX`Rd@w4YZ)|4l
zO^zu~B~kS^&=(VQ$hlohSm6D|Z&h3HQuH59Be#L%Np!;8o&rD2=+_8d?1y?9Id(j`
z2+9xFZ^WHMVyV2UF`7I~wKYB(F+BfC<0%aX?CApJ8=T_n3;PKL5$ny>HHr}E4bqwr
zf2v7@NfCD^l#r5OcBeWb34sleoc<kW0QD@@ZnWu-bnJaCkTHG#{f_>~;>^=>JuR!I
z$~uqPchxM&d($buL#@M7tA9?x?If&tBiupIM9^(^Qy6mL+iV^L$5ds;a4O*eo!^NO
znU4fhg!CEhysF)OTMUloybg)EujQ)lE$!qo?Zv1PyT~U6t-0^7utG*ZeY#<G^D1?%
zQsui@S2cdhcdr;YFmrr%mt0pxk#VNe2N4)zte6jgXCjvZC?PZ3?KVhKm1Xt}Vga5M
zl#%L*viOvUpr5VHAtw3;3Y}VhIi-cEOE1Zum)*C~zk0V^&!n@?0aXKA491x>ecR?P
zcZ6{`t9a4V$q7y?pBX8gT5&Axiytpo(Z$*|)R<fJvPkvE6x9qvqDH9TY)dD+*92a~
zNQ7F|+wp?{qNjVo<~9RHomWxWzB|DOj{9t!KD6ltTxP$W_nhr&rYX(-@apJ}I^Q&D
zfox&aGsmZ9>rV7Ho)&(wd#9aqrZE4!#{0@sWf>3awZ3*&Ph~QOeSk-Mfsr8z;+pm4
zqjtw<Pop1{d&&G-n_Ihr`SbYM@bg>0uOIS1)qP#N(8=6ombU60dT0Iuzokh*J-_<2
z)Vm^OS(|-B?`R2JJm&A-yDVIPfXF^ry%d8t5kbL1KlEjODu(-guPvw%==qj!IOCu)
zUe5Xb@g39q<u8>74HfFG%YUC`c_eD5aG`Zw>r!_aZ<B!@yDUSOv3(O4ecpE9xLy6D
zw}Z?}_K-wk$xN&<L|HvfLCF#!PSF8!J->d0`+b|aa^1}6Kz8L}odK7#TcW}jxQMcX
z+6$k`$h+z)DK@f)Zocm;ugnm-=A<@OpNiFf^XPhWgS{ndNvyjuujr0g{l=ZM`tg*~
zOaQ(q<JVr^{2%NA*$IC=vk~~Kc+HSeX)iii)7oJ(e#rUN_z7<>xK(MDUNop!uu`=D
zx&JkxMLF-`gOc%fyK_%JtEQ9G8;P3NfTp52@e!!=UH7z?;cq@?=ULw<n{H|_^ZoQK
zceY%1|NFez4Ap_`vgrqJ%6eE<R92bD+V^?4Uv`%}kacfKt*pX>!Oizx%kJwMReLe+
z(+ZWH5sfPyTP=MT94%JgoL|&>l9)lMoVYZZ0Zx2(4N_gi`{K!Q(yz#7Cp%c~zomOD
zdHxx%yze`v`)x>hTH~-`Lh=fIzGbGUXS&~7OQ%Yap5>@>BWGlFDYlySt!_PNI@2-A
z(b3U#Mpl8adv%n%!A!gG1Fb11HfQ}y+q%NbD1MoRxx;|ZuRi%wry$K&;TI=vD)^!s
zx}_cZv(b?RSQtO$7{kk7)g|zz+Mz8-gVDFNlOU*SU24a*rf=UgE_GSglpSM67*+%|
zJ=t?U{QSDS89pw4MiHg4N2K!(DD)0(YRq{1&%7nET0@&|&5XKCkD4B|t~xeqji#Mi
z`M?YPwx)B*^P5_2q=&VRuJ~m4y!@Qgy+kRaXM3tl1X>H&k5*(iK2~pT=^Of|r*@Yo
zvNnw2^$;BR{4<x5nk0VD;>PTmD)nSVbeC~tm+X_gi@CQq8n;!i*f8J6VZ;3Rr)FhI
zp^x6T<ru5kXk<Iv7F8T)Ur*uGxh4;tb&L3sJ=EBGXPNag=l3m3HXIq)HC~i*a)sBd
zzt68Mz4hamdXm%KpXEVy7MByxH#_wqKche-*Uy9SssIv5y2anVO&2z+m<VJ_&u!CF
zJ7vz-`=+Iuy`^nx%2}TuuBTj7&mGg!UR_hZc3U%V60>ZM60>Ug8UL8>nd6<ljCZ}`
zzS3SR&*;Wu_Q!_0s8_`c%98GMJ@UVt5_@lVlt8geSLxhI#vJ2mC;GIun<&0_FyGWr
zlDu8#N^aif2-Tm82+HoJvP9~=5QUFb37SYuTfM9N*P>H9++tnypSdcVrih*hkIJfU
zFxeEayR+|FXHtdja=U`jn>)2sD=IhFUr?JNbNE_pw9nK;OUnsnstMt$yPTF@b;{5{
zUx)F<g0)_T;P}3=K9v0mL1)7kJO*2jLGCzyKNUTP`3LQU=%X8@4%*}Qz0CFh%s<p`
ze0JiBkL%_wohcLO>)oGZk}m63X=c;n`2K2cHC$;?M%=uNGy=^?bvCFgai9}I`*+sV
z=&IF2&fvp!mxFz-Ot3t=o&HXFmaJW?=6ffZqXB33L~DAgJ%61-4>BuOR?g4Qp6F0m
zGNJS>H=<PKte&;(GBwNi4(A@bv&#9m-oEm`_SWli>~)dK#DD@d1%>1@qXsU_Jz?i?
z{zId{5UMI?K~M6f0p+@6hm_-9F?wzrOx%}lxqR%C>w>do@0J!_`=QdNl77!}!;h$?
zr&?`hYW4q;Red&aeqG<n9O<8CGi1k)>|5P}uY$4@?hDthcD|PCrJKP>_C6A2Z}DJn
z)tYlD4Tj;l8r}Pu&URJ1vvOO8N549;7Yt=I<^M9Er2kF$s-y}~3VkXw)b#oIV6SkN
z3Q-_h?VB3o1N+RDN59b9G~xWH?}hD~9uw8#mseS=?*5Bk=3a8WM5DiV)1Lfi4{~;u
z4>>-$vv`wSS(;X<)ABW6&bVDLE$)3dOLo_(Ru?br@ZhLEGZ(9>nlHNZPeXJ4HbsDV
zmF!s6Bs7Hb*@$HY6a{pRCOPR!VlV*vjR^Q5nL?8PB>PV^;h)U9n$<^}d+12U!p9@W
zZczPcx%=3$&&i>AzE#kHew(hRybU^8WWUCj*=#+jSY><FthF@`ccK5hjJn*b%z~q8
zL)V^HJaKh|(1cfz!92*gHN8Z0<MfSrXP$0YanRKL&ZnRfflE%wFPT!_x!l;#(#FS~
z15Q^Q_kEvJ`Y$agDfHcnv-5wNEG-)FC#{l#6+{L_fUX7sa2UfM#Ulqi>r&gp7lHJ)
zwqvA%&>Ft6>W^;Xo{q*C3nB;OOE)*^U8*xV?scQlHKw$|-mWaXU@2#HbKv&rC3E{H
zbtT#QTGkdM1!-w6RVu3eXX^bA?R$<0<hwOQfr`RVuXRa6R#92TgFS5`AHC^)>mE$)
zYPhmVt7^r*D(&2hI=b_+E31y4dU|}($?uMkYqN<|S*^dUr-pL8Dc?I)2g+ft89lLD
z$mpr2@4oc%bgZ3<lYkR9O_lLv$AuX($Fv0NV`Epo*c$4(Jo0*=Z`a@Dc;~IYfJtNT
zeVbIHS8#vHoZ05Wig`<ZJh$n~DJ?5Jv~h!AuAWHKv#NJZ#cr6^1XufrZ~h9I`$?0=
ziwf?0M-$hNWm-mlTM$39L%7Eh-2HaJmoK%`SdPMm@e7^CY};hAF(u~Modx3uC%ijx
z235h^LWp7ORw5GEzXy%<!cv<|90}sJ@tcX1P)*jUMsE=ip@vyBXDqaDgpI##@B&PE
zyqL<ht0sP4@pAaW_enX#fn;zZUm;^P+l+WoW6YHpiy@HX_tIq=!u8MA3$1=SLXq|m
zQr$HHLr`1{d<ni)PhcSi0@#LsX;UXseHHG&@EF_b_$O0H<?I?<CsDF`qFszlh<~v+
z4#8`k;Tz&=dg!~)BGZ6T8}N?iRKwK$>6QL3tth=>_jShbLSzszNC4!OiXPyF<eiU5
z)vYdrBn(1gig^ExNaB?u&J+3S;nRnFD#t-OZ*%p2qROh|_m7#Sx}NqOOYFXxYTwoM
z-e<Y5_>(}t#`uDP@2^y`_Z;Afx=D59Wpllk_dJKMXMB<X(*DGrqp4duqk?LeFRm~P
z$}&x!Z)az*Q}g~i+h_fw+}4>`E~(XX+plo9Pd50LOhDC%y-l;0t-KzTsB92r8MSiS
z4o4GpL5twknwX1uj}3f(_LJC9TzL=WOz?NEFz+RAxkYvFyXp3SlD^6s%!)hW{=!7V
z_vhurZ+S2G#zh@=UX#5_%jtbvQBuy+U;L71p=@NUcmFMTe#2$p<Kw@}y%P!(3R+$7
zF>Y21wpK6Dm#%Y)^qyO|_E?RR``w0baw->U>mse6wJck5^zgip>pde<R_;&kTk7r7
z-8H54-KvV+a9NF+3{C_**Tj!t?{v?MD*y#=>Li$zkv}y5z<)nENcy_7c2({5Zv}qI
za>9~6!MM(XbJG$hrNoc5eE;`>V=BQ1!h`G+K4`jlDI|~FnAM}T(Rb~Z=ID=){w~k!
znjYJF$}{Dx`{D~uZn<Sqie2u%^bJOKobtPI&aPJL^T}nK+BWQeTBC96?vKKth74X%
z!=o(AWtI6^LBlQ2u3D1!14SLbLn=mMPj14iXJ(Cr7_EOhtq2uFC@ja#5Wa{9#Oa4#
z@sfQrUHX$<tDHf_U%q{LueWP8v=$DXOxIF3{&Z^N=yH+4ePciE<OQ`RGaL)fq>64P
zZ$GjRnn*L);%Dkvr}x69e`)iOppiSAquu47cOdS@&tKV>pQPDanY752KC4=O-nAo0
zvw!z9J=2*1o0EQNd|V)H?fuigZw<n}HhhN4Y0~xrEpIq7gy&|1qHP}6sFLupPRZ^P
z#$$}S@_a}S(ER#OY-hoiA4N^VxHg4naie8Ci_T@#7923xG3D+*3P-vhf0h-x-b%UP
zt~qTDe6Dpj--MK(yJO%r^KR-Ms>zenx>glGy(Lh(zCYVA%Ke3YsLj`qxbEk!>g$4&
zFMSz0yRPE<0w1%$UFD@MT?JXe@g32-*3>37y?b2tsm4eH$xkY@lmWigFNK4RFqv_v
zybhdyYA6b6@t+`_J(O!p-pe;UpG_`j?I$?qx+-(~A6%d5dNZ;;)XOPrf&TLQdt{ge
z%5IKA#$f}UtxmJA=!crAcfC6Gv})<;%I$M=CxmyH`Z!b=cJ~d;(s`a%Y`pJH&hDPp
zCAYlz=N@f*RM}P(ndyA@_N$)z$M#eLtWHlJTwQtjNIPlek10CUQ<eXW<2!+oMV<&c
zDrgc=gPAE6%V(vN-yPv8AFjksQvZ0gQRjMRlJ#}b;e$J}r{0ben5uFL)I^@=nQ8e6
zM(4Mfb!qjjYun(rp|neQj6W+OKdrb`{ku#+-d;zO#**89!yjIeMB>mlL~q;>_eXRz
zBaH-rDLXC^K@XMFlxQTWx}29odpcLk3YKh?5qYl6av!53r{b~K|HPUfv)*q3#ljSM
zgto#Wz?ioUgcv@NP&Q(JKUG;~e=Yi#*3$pek~n)%6`Xhs585|2fQ&*U0jS<;gxW-u
z6kfffjD$_&_u5?j5_**g+i?ADpF7ErM&uh1!^1d(OG2A!%7^Anox@xOLEAPgk9P|x
zqv8*=;GeY%U5N79fwVV?{DM5IW9J|k3-tTY14S9RI2k}iLRqHeRnQ1kUvN9lAP9mk
zp$0rgOc!)-G^9ygB^A*IGN|U7xe9uykfJ!Ef*1o`bNjE;+q+3yh;}2P>cyipkwT)<
zx!~XM30)Mds4)iVKENj2xkjS6(fOU7ntPTRyceG&0V@P)_>=Xl?MbAGGOAFY${>J~
zY&2P=NI`$iwk<^|`T=$&$%>_OHZ6ted>BIu{fne-?$63UW_@tD>X##-9}<8=8Wy~Y
z04o^)fB^v{)I$Kyo4O*=Sc}oPh%K}+WvCR;cY12>x-r!B!Y|)xwX?y1Y|7teu3f5K
z;S1S&=bn3W@!IWs@=q+L*E+8DwVa_?I;3xLZ-ny-trCBYmb+iV6Z0(MBL1b#Ejy^q
zTQx>GXj`UPjGnt(k2*uIK7Xc+&e!-4)xeV}Ap+7CH~gY9ZB(X#F?<2`!2lv<M1OJO
zMF<*dt=ZC<a>j4DcCywiU27NRT9>Ci-+g+w+kg6G;3rkn8`}7(szRY|*%Hf_mU2#O
zYIyFuY=XWmuZ-K%k*B+~!t9?3XKO?B_RNI3d*Qs&-pbL+-l2Jus7I@Ke%$h}n@Lun
zQLHoJ(Zo|RaA;8P_Ub=bKAG=w>??1Y?VmUs&wQ5l@C(Cs`R4pqaT}f$mZeuJoNxMI
zWEM0_xw%UewX}SJa;t5@o0!0jMH4<T-*?|{Id%D<c0uPa-8D#<2e=&l@^`_v-futW
z|4Vatas05tnu>8#+pc2V*)tFg1qzI6B%w+YRGjKGRn+)B>iwO2(VN#yVC5GU^c>M0
z3fZ;Gx>qkx-m<mnA7*UwTK~)EE*#&$-r_vqusqM@p37gmorS*^@*W6}eQtir(<`d(
zIg(QLInQrZS3|2tu#Me|BXN20O~FPs0anfdd3m39wSNEc4|zEijs|1NFP!ZWjpPrN
zcp@}+3PUxwi4uoH*jHZpAG(qJ<avqB9l@o-h761E>w^xDvrTK>YQ5u#-#De^_2uVW
z_y=m8R#qG>ysI%w+w-k+n|154`}U%;*A=csV{@My9UZ*)V8z|^jM78>dd4&3o?@Ow
z+~7H5@4SJK^r58-GLGB<UWlWE59LaLTBzZ?h|F8=cOcRb<{vT;5s*vBB$W|@f!2rd
zr9OC5bJv5^(hcr)6P9@D^=K|FbyZ$>*2_-SbpMSXb4okPESk(cgewF(c9(jdmDEZt
z?Q5NQId!daMV*t0hu4>c>@!dORDUENzkh3A^BYZ(l0(gpajPwMo;I1Tn)Fy$I8blx
z`ye^@L|pc^o}pcB)<22<^EgQvK@?+8Ze(;nb9JlUdS4Xq8c|<s620XhrWKIgRXw%v
zq~&1q(n<QcQi2Bmm;^0PfAjS0*~;9`Br8sAl)u22*|xG~m3ERvu<7+GjYs}PBj-17
z573%#AsGENaDfjI5=aRWr3KLf_qw+;LV-czmtxZy9Hlk}v0);2etu!DsXYC%xvd$O
zGe%>u0ZzvqN^GFwYlOXk*I@<bnn$t*AK#i|5bSJ^eT<%V82ICD_&e11vEwHH<1bLf
zP|jzzIX~SUwv&jhBoW8dAROY6*N_YxBswf@4CJPQE>HvCec4xwwYCr6?X5f09I2$g
z^?M`xUs|zNKgLLXE5$HFDipsvkG^Y0kFi^n@jGVK6=%|?hs*c4Q-;nXhDx7pUhD9c
z+z&5Ee4AArE0*8wXmj6d5qT#47aLij0Z2D8z!V>`94A_H>NXfqH~A+v_YPkkuIDNE
zsFNFd^Uz#~m$~Ez`Y(1P%;4Tn5!(QZ*fN6*hrXX{T&~u)-?@|SDGXUg2ER4oX0(C~
z`2SXBBOM4>LzI$H?!P0~yThXvy=?gs`jfshSlkk#KS;`42E(QyTwsR^8E|!horKED
z3sjy{X-8&!GVV5F42v{}ew{9Q*-f5>o<mTLLNpnJiQgo{0F6jV?nz2e&d8;`0gj2M
z0B|6|g4+^vt&pLW`JzX_g#kj$T~~{GzkVmpAznMtCGctKAV@OKo{ar1uK+3+1&~*W
zRRO{{RMv?MP9I*bUM%ht$$KFSjt+xvC@l&_q)M^}SlRl?AMFQq20L~o{9|HL0ZSYz
zZ!*_2E=lGQeql((Nt^R_b>*q1j_FPwP66wSB!krTm1`5TiWwyMAlPzH-@45FpmOo<
z?=BZVl1ys=nn!GQ5EjOgpjvSTit%&A@4?d)<zYZd_@em%HRIvbDYL%^YB8cruRSj>
z%%14#tXX`fEw%je=Em+qgJ3&%O#^vd0~N1xf0;LCeLk>h&Bk?85eJN2z2}AR*X2~!
zfy$#$>1+Z;eBR|z*%86;A;fy&@Z*%qYox)OjBTS9wOL%BHU06}XW79mX{V>NlU?7s
z|6Os9QQWomo>`v3MAr|>akW*4pX-J$*mR)OD8ExwcE>OKNyxTGPi=1Rx%%b&MY%ie
z@8fso7+qRm@nl;3Zq4E^g$G4FkQqd3P+1UwKp22wr)Dh_BO*6#qeun6%o+es`vPyR
z3fDPaCCe=Mrz)7~z3wfOl8<X?{<=6lOZ7t0xRB5JsRhe6t*BmccGX{3%}S=E*!K4~
zhx)AWQeV88S90<EJEv8Q&I5j%)pV0qKlD}6>}$_2trf&N?PrIyn6$?~2K789oI(t;
zNTr1Xo^=U)B&9Hddv|1nL7>)G>stFZmkq1_+Sa+WBJ_gVdH<%T4tpvST7DGj+*>ub
z>q6}uKbI=uz8B24V=q=XShPLsjB81Mwy)yc9AWGD$%j{bDXh`>V&bKJ%DgczE~Rc%
z*LjblzIEkQUwjV^6VikoKp8H`w9Xlv4YmO6)!WS;E1~nH>7v}Ux;Ec&YgbtME&i7O
zB)=`zGJUPIU&-HH{wrhK?A+yQOmd@Ux@>WN|KQl>?d6WX6|Jc~gCWU5_D=x+ocBZ5
zZU#66oOHNjwpnUv(z%#R!}cFq3TM5nLBM?n#%=waFl~oNUwvL{fnC|MS8`dZ(wi4v
zE>KllVtgPrY11<wX)UE6y2hM0H+ucnRU2NCO-w%RXgSbw_-LUs@<6h(@h-7%0+%KY
zP3`~bTlVVRy!_Lz2ZxQl6fSCAix_fvqJF|+)o_K$JMfqF1s&1O<pB3@efKMwe_~To
zwf}_caT$@HY)n~^tV0LW-F2;eru_7Ia6$gyUl%pYZcZ(WYt#B*T`N4)<rHjkv5_>M
z53&WYJ`GSxK#5_~%hnv~F)GWbzi;eP&{sG%u%_sQl>XuDF6ZOPB`x2~{ga+Mt}v<0
z{eE?mzopN)M?KG%#5#=+h;)i<UlLcb`W`jABVuY1CWc%Ih|~Ki!Q6FmQ{U2s#zSv4
z&CZ<?6!hJ5@KP>49^ja?B=TRHN9)$vI`@7~aw!|k#MG%1godE8Q*{O#Fr4FyhC^Rd
z{<O9|gBkip6~))1cy0M5TCuG~DdUf=ANDQDiTZW0=qa2MB6Ay!MeXq={EQbTwtw7n
zHZ^C)mic)`1#8!(OMh9?s;{|Vf233FlsDvoJ*SA2IuRK$fAC-0(`u(Noq2}ia_%3R
zFxsv#Uf`c>5xS<L$hTs^+1k-ROZU>thCqW^efC7~d-s6F=tRT`j1l|En5WtIZNB<=
z)c|e&1WSkE`ME9Kh@W_OwL{!z$Rn{P^JOj=0K-y{J^%KRP^8@0)}Lz}d`}Wny10`F
zZ>6b}cmKr=BQ%mI?+sdUaJ|lE=RNs_eruk3BECdCJi%5(t>9?LG$cX2C5qh=nJ@V|
zo9*)gJ(AU$pQrK*ueGG4x;*<N|K^hasrmuMnawU5U7@D>gLWnUX|FdOd8Ju=-s0f;
z=jY8XHhvvjH2p^Z7)Or}7ICG&MlR}d+N-75Z>Hb${BI{&vqep9npgVddRDD0ja`wj
zts(DX({o;tN?PoS`pXNQ+%AlMB=~8X+k!EggJ`%%9|7tActYRG1h8^2@MYis&0c~U
zfYhxW+uB@s>E*3e<wMeb{n@coy0n7U-NAt7X9kYhZ5DFPt!C?z&z)Vj|AB2`g09!a
z%YGJ`F}$mBdHL_#qNb{uyp7}+d6lPc+urj3n0ga<sMq&>yx;kDPEwr{rPY!xjO{2C
zg^zU@S!T10tw@E2va}HXN0i8#p$J)qnPDgy8kCPpLK0@kSSnGLC}qpm|9-~#{{KHO
z=Nz4s_w#<1`@XOHx~?Z0qq9e@b@xvxkt5SQ<*(fp-`}`NFz6Qw@HeT9WZL`^HHlh$
zvWIupTfspu?s?%v7Ws{$q+VP@yo%i`M<=ffn3{r>;_ZC=JJNiCMLqUOM_Fj<u$cE@
z=~+iP(yo!z&yP=x0!NyrTfW@~ksqL}LMv?yX0f{&4jX-FZ~^%PMO-wq&*2RqPRzgD
zZlD6K5MZ$E!C4Jw+g3p`^yk!Fl0y?L{iX@S3e1XId1aC6QY<_nm|2lCph^Wew;5k<
ze&kzsFZP75op8LW%g^TQ1?}H|x3X=US&3WuwJy%k<|3K-a>c-+`z$z*#PiM;e^mbI
zcR4Mh-KHw{?(C6m%aL}K48G%vb@!f|j<N5~JC%f3l-#}XKXdlZ?^l7jiBTh$TqC?a
z*vw;c;L(MV!|qX(h{@siaBwI6ql_pxz^X+Q)k|xEZPf1NS_6k?oE1tt^q;vF0)h3E
zNUA7~9MWKnU!beqo5{?1b2Elnbqh~;zIS1euG>7h<*bBmXD^*v+FN{{*Gj9?slrc>
zjq7sUM9U5IS8*j0*e^3HcUCRp`?@X|{3q+o&tO+YLtk1KyrtQ_&L0gdzMi+>)85bS
zRat0>2pv4JWS}dE2;yDp`!F-`gx##}l~;Mr#5N~yWWV6`PJLcshGAK^rrOOm|AM)4
ziq*;<)cSxh44_j|5^SJbjnH>gSV4he{HpFOCZIFp;_w`nNZM7=ec4gb&@q0bvq*4c
z!(~zd+b!*a51{^MdRpW9oF7tZmA`1TMYIX<Kgiz`p_n_goYK0!HC#96o=xSI8d9%S
zHyztAkUmg<>a6Nli~Jox-|w}4PW;jRJ*juBmiTMkzFBFg;F88_NxqLlUR~s?Ym>~#
zK~gDSA1GIi3<aeZbnIJIJ9jhirez5)q7khipie|>zg4)REyfpeEQK(-uyDZg+TH%3
zjyoJHXf5<R{%hTtmCCB>7Lw(~;0Vl8%*e5dTWss+pTcSh!8>EW@Oo|D@PcD29KSRL
z_GeO=issJ<d)a2`c-L(@s+0bm8JVmgxtAl2>z{si#85p0@gs!t1I50ERMknZy`EjZ
zX@{f>{B)%YA|<G4%ifxbA)llkc1@oCxq<Qa-S1oISU1OU-%Dd(=rxgxN92p>?)_ws
z-~V`eZol8G3<qRGupQtNacmOKm%`V#=23GRkd35{t)~X_^HoITEK;(?_9dPmm^}qX
zqCMxY54%!){U-A~MR?|cI7Mya-M=aPNqw2tSe#z?C$%xz`FZ8fyNRTtJ{(hdpuj~!
zLgB$n^+U=xHj*<E{E}mW;?lZ@vH8Rz20cEcJPm8zNy)<b{H0pRsTKVZBt0}6J;@Hg
z$=n!}L!>gvio2E#3+dX|NWSl;?)?nQ-{x%>M05FTk!E%5fd+}9-O{?bXxKL_R;S0u
ziC+0@oq%MrA~9<4JuS;O1jYwLgIp7_(GdOvEv-T{+(<KE*w_Jy`O6y^BJ)yD0jEai
z@uoL-Gj={5j4IPKmd@;tc|PJ`$Pc94-I3?(=Vj)*KaPjr6RXFlcebhfn00<qdrlya
z*CwTiZ=LH@GgGrsGZx!cx>E8Cx}(DW2_)r6)CpO>){+|J1H$SgJPn@V%KF{ajMce)
zp9GAG7;>uvMMon%Rzq6zR-b8~@q*XhAUC`1+2<ZJ>CHX?IyRMaK8_DBjNz?a<w|_d
zaUCwV*rt|ylR4y3tsb2|yle5_&Hqenq#IyrI{H*fD;*<jH`MUvcZipnM+Bsl(<Dbd
z&dVm%k^}c$9XLbVip?+YZP(4n3q6l%2vv^PSIrBseK>~^ru)zlLnu;C+L%xz1o119
z0LvnJNf|<_FPxKst0wsgf_yoWiZTGr?rpuR1DG!b_vr%zj0r&0GaUp2nM%o}w@*2j
z%ys!^VI04-#iteCAAjVnDb11SAv*h*zZqh!TBZ{(>e*#4OJpb~)11;foxQ~U8u033
z-V%(0E3rKDP(fNuQ5?aE(xIzZB`5N5&NjmT`h!k`Zo2-(O_oJd5*Y@&VlhsPM@6Zu
z=hsR#@k?Y7C<%l9t)kG&LeiL#KlHMr_f|8Bp#?E%q!xn{DwF;gW&ZW5!gvk$mtV-Z
zRC&7Zm$UgKGAVxyx+kNW#wF*`ARGm4nIvi%wp{`;NKqbncCS-is+N&DHLWgG^OU#p
z*S&x%$}HNpHd9#7l#4Zg_fY0q(3y6x9iR$0v>a8R825e4>ET(a#!0M^=Od~@tLH*o
z;#tUTqVH9#ZVGgaaBQ?b^K0E9{L)Sm0ggHdPMsq1b5@I`J_;*=Tl#pB$)Py>q$@5t
zqczUz)!g=ON@NvJPPA6#?6%Z`TH+}+WAP1>r(ye_Y9n8Bl56oJqQ~yDDRU?s<VBV3
zwvjBgj-GVxrmLrxvgqqSabJ?pZmfFfUBQY?zAJI1rj7Hszw~j~K8RUgx>eVQwz{1k
z4;#EX9%z5|(3hRccjJ<-Pmp`m_d5;lN#wApd<n%}qnakCs^Nj-p_OVZ;{~GHR69Yl
z{xHHNmK63@GoBcS5he$^_sz*|Oz-GFl%9X%Q)gKVoYYc+Br1@vs&Pv$56v#>IijVY
zi+`({=SMxY8?VN2nC_S!^N@FaK)XCPuNtYz_>vu`$9zU?%U8Q;RgSN#lww8RD&79=
z*E;-wigZWcmlSWxEpkigdGeby=NBXM(mYQ7gY9!~7%4=?WZ8RQGe_EkO1!-#DIVo-
zS5O#{8)(xliMBMeJwJABlHMN{m`yX1)3Z+Z{o=_|9M`IA(IIboQR7h5;{T7dPA6wh
z>({yjj9hF5#s%*hOLE?mJm8utuwLJMhzC7q3^l(Xm5{z$U&HqWrdV}{vnh9cy#W9H
zT#xPLlrX&DRp?c3>bB*m^2D5mCz1HtFFjNx?;J-q9~;EuJFIu^ESXfP8kCw11+FYL
z8^B+{z+r!_LWLc-x$pataqfp`S&^N}31+e)k{NPDGc_x-1U1t>TgB5laa!p#0;T_N
zKiMfk%OwlUW=r|+Nv*q*1wNt<XLB2DWczjY20341e0XG0{25be+fs6%n&aSD;oAAx
ze7_(xx_x@i&rX*;Gxn5wf{|DwsXHlCm8wC*lP4cpUz-c*F}+(#eeC$fgV{7jK#|}Z
zwh6L{MAsjiBZ3_T`LQD!&R(Cb)5%YbFj~*D{o{Kvu|LIn$^RT=m2QMXUF>vXut&!7
zH<R$815!U^9zbehXt>!Ah{-bhMHYpT2GBo1AuX5Az;7k|abY}?gFW$UT?0|$6kmCC
z)$yXm$Ttc=u3fY*j|#`61aYoTQ3`+cSwp^A?H{*HH}u&3?w@1aRY98)MsLAla!)<T
zOH&Kz41W6KH<XcMoR0$=4mX)3cr-WIsmmxmV#xPv^~k47_LQo2caPLb4>2!ghDDLK
zsa!lfd#zeM?B4tir->&I`#+m^e09kR)`84&RIFW0U@&n=xiP0?PNM>}qT;qtCZP5-
z?tFuc;lI!_hzRrvR{HIue;dmjheam&z_H)BHR#v6>yxJh;k<iX%neS-Pi=O^Ev`iK
z_T-BJc73=km9w4DH|pX9=3Bc|JAO#hjq5G(@HX0&6Iz?UL80mzzs-2;np#!v!MI0}
z({}A8bDbAB4{Xz)-scTRe~4(IUgK7zrAv{!3&i)ZE%Phd7e$}wN9aeUjM!qs$>xt+
z_}tWnE^qDs@;|@7sAFb{Ic`(={Xb*4Z0r~JXsGd(x*7oT=>)cC`{wYY;R;)^gZtVZ
zV-@&w$w6v+Dg-PoS@uZiP4hIa#NOg1y}W8SN3UJ?7jgDqrg+xflnO*xM2SCb_qiUH
z$S~<Wv*Ox-fklg-8uyRWJjeb>>D#~76_{D#q)mAjPQ8dsVb$i;V}7msJwQrRZ2Yf}
z*QyiKeR5|)|2ZQ3&O-K3!O+Y=qnSrZaVx=#XYi1XrL>5b53%Zgt!va6u~UK*XJTJc
zJOptg)jxD{^WaP${i%E~fVe`psjT9+{3xKz&hO;>IlC=Ao?dXJhTLm?LQqm`-(4hd
z(Pnev)%f<IPWy?DU-BN}mO%H`ft4tk+F4z+Uv<LtWoSa_^uy%@XHBM?VyUx9ePTxJ
ztf?l+rR+WXW5JNo*sN5S&)Cwhb)BT?H`9yZ?IFe7w8>cw@7F;mMnB6}MjGI#59)Mw
z?6!GW)HL`YH;75}TsaBZ1&aTW0iaNtE&oy26LymS1RQy=xqw9EB`g-B0r`PApB7#Q
zjdG~zKX|6Z#<(szH7z4Ki*<zar;r#*t0GR1#=j=FIzM!z7F34OWSkY#mc1ZRE#Ucg
zUYoz%_*Ca)-U!qfm-<#M;)rDPpnmqulIWnmyFBkk^|k7_q?_n_H&)(Do{vl>Puou~
z`cErfGPkKGG+F0<F1usfe(nXO&YAJ4a39-zU&+Y0ktaw5)}D8*rdYiWzXVLQ8^Y^h
z*aC8p0|Y6y^pZ|2>)ih^rhFrgnjqbb_hHraSv&YDJSfHE$&^-dn3m)<{O*H9Z`EYl
zHqOcPe6CoDoW!<6%H?E_kK`7jwfkp^(vgWT#UG8{IalcP$E{LS;+TF@mSnC9oo$PY
zWzfC%KCEV>DL@fqYpF7n=h9^TA~dh|lHVo9LjSPwUB2I(A6?>2nmx0Xf@^IyH<Hr8
zvKok(tcW6sq}uO#<cVJPMZG}FskV^Us-)`&*S}81Jk`;>NGNLj2pLV3QSmSunGpZm
ze7<4*`DTT0uoJA560IhR4)^F&?&4puu_DO^8omnJ7C~jiK7$A1Vq@|?4X2V$-mu-#
ze6Dh&`Pc*84Rue`^#LOo#aBhiPiy`6H062a)rK%Kv~HgFen#L0y56v{pX-M6`R3z2
zZ|wrG+77DebUiazMOKpUUM84HroQvb?Ad(s%y3#!UvSugG@hSi$0KtSiDxbW<fWaK
zvf8;FsE;dDL7Z%VV+W`qJoq<27<zKA!9v=pIrqH@F68?evlk=3y=bI^9PmkeSca2G
zwhF|Yym;zy<G5A}Yb4sUKU437n74Ki+n|FZc8T`JJ?gY`_pqYQ?G?LU>pTt=HLm;_
z+?=Dl;M5u1P7skFEku1>goX$2OJYC_$eqIYD>zuT#o14hpB8n0szR9CV@7F7ZD^o2
zvnsI-$XXve`r9oO$lfeIw{GLBU+bvrBZnoc=o?z&#u<u7UDEX@RckjpA8%Z4bTrv7
zNaaw^F~*c@h&Nh)8`w#yiJUNFu!3Z5zLTA{J+L~-<iyUZ2X~os;x$y&HujCX-q$ZA
z`E>}oUG1H1H?Rh~>nv(;iJx3v=!bH2HD5qO*Xt{+TKPO-Hv|MsM|WvKmsF?(0m=-@
zYd7x#>@yhvj#OC3<mIx^dx_je$gc_R-pVySIOSPxn9dH>{!H@QmUJyX_l%Nitv!F~
z%$LrpmIwpMD#w&bZmG2kYp&pJnBs_cDz9j!dE87u|NWE1lM&pNnj;c_+1IDUYyVm|
z*3)%yh`c=F(CW+y3!)5g98?y=3nM}&exn-b=M-!y{@9sBb88>heR!Hz%45v^z^Lo&
z7Z1Q$NveGAxX@|yH?4-ItNCA;X`oL}7UpJMl@f9UQOj5XVE=?bAr@!~%$Pt}1V{MM
z(r<TF>lUq#HftpBYK<{3x@ks|c<3nD)8|msts~i(|F*D-+vmhrJJ*akTksIqr?M~;
zgl~-N;9T;kNTPaV4i-?2WSCzBOfS~Z)e};Je}1jA%%PCQF<Q6?qs0JYiJf*f@8Vap
z{nWI!ed}=5)1V$w*U2_;JjGS#{$*3o`{I?dTZOK9FyqH!LTT^p5lm@V9Pa{+S0x}F
zIj6Rio2tSM&XSpGu+A(d6!BLAVGlw86_FP*(S+^z<1e_VjREa-nZkQ*kL~6Gm=H{b
zQ2ZSPrS4;<wVi~xrwKoN5Qp(+w%|GybG9&fK=lj#V<(<edc$vrOUc}uNw(A!gC9`T
z5}K0M+2SLC8`M`_SixSHJe}Q4r6k0iefw3vGtEoAGwS@WbzSVUJhSMc`r9OPwNHL|
ztHm6BLhMTezyR<csfvMNzOvX?-I$sxr?y-|DOQ#4q2YC{kRQ>qskOn^l{B({SNGto
zr}7{r8<*u+uCR(=rL|rR?AP~@<^-b-kDn0}&K2{<fM7?}Yv3WA%(~0&HN^(<r))Tn
zyK{ylE)YZI{msQa@B<9PP03g}ny&ptg`%7SZ^h-%BdL|NjMR)IN9xQXH0R^}e7fo=
zBKTqExW!~?XI2?CWUsmR3|Z8l3w7_y5beO8UYk>yzt-I<2Z=vS;4G&!q>WL{%Z$Bg
zUNm2`+ongG1dDT4j%r|(!MikeX@)$<B08TOWJWKhmqm`wZhBoy$m$?V-msB6N?y{8
zOYDezIZhH3j=`$)nmIcr1qmJSN<{2>?V)VuZ0F?5?T?Mk)G)KO;Os3(k;?D#EwAiQ
z*~e6w+;`JVOiX)j>)f>i{6N=tC+v=UNzg3@yE{k*#hWd0rH>7&pAv(=J&x<?|DLIu
z4IJ!U%wOv|#{ZZM^QUf>I9gD(7+wjqrhDZT83_{-D?IUhjWh>oj+jK5`3H8(46Czb
zM5U@5@Y8!FZrVGqa?9eMn9!WIxAxx`6Vg};>JIsw9_fNhUmO=%3?|;4QF*)kLK5`O
zaV1*{{#L6huIZSVGSqXLu)9px2+hXTlFf+wL(tAjyt_R1FC+ijW?}nJX@2|e>caW|
z2)JJx@ZANfr~h%>m`47`?>|k0nXHDd`}o}7S+Vs}m*m|WndgSNVzQDixMzJPtIdjH
zS<eA8s1r>MQy0&azxH5HL0?2-QKAcBPan48N5vLTotD%v8-t<*Ghe;fNr~99#oGN<
ztkeW+d+2viYcN)m+J0g+s$zCN`3g#eGJdBBG7|vp4YYpckL~I_wyX!vP-yEQh!A;N
z{a=kxFueG2FX@I%9&9J2gV`@r5V#+5CN50-qW|g5j#_T7UJ$Ka(~T%?>2w8$hpBXL
zl0lk7CmH)|ojG3s39lb`DZPHARk(b27qUzNPAFI^0iFSRuADT;A3ZnsbukLW_m(e?
zAJ}VmDlYX-A=~0{RBQ(CJvq84qn@|f<3+pCjGIl#8zj`xZb+<y1ZY^OHuxy)xti`7
zew{K0UI?@2uzYw-MW?;5R)6oszx>qITfz==DzU@G0SMGUKIuu}Tc`Nwi5hy0)?|^=
zpgk@Egml!yf=G_qS*B?(DvCn8I+Ius-~ro*IwPqE04$oLH)l^QAA3&ca$f8~c*q;-
z2R`(ddT~<=9I+%0rCJf|B+xJkcFf<N3JwzHJV7CB<Xhe7eA;$Zzlng@OPlX>Mw;gw
z1~v)&o{-!rR-`X-Fu;n};WoK1!QvaK_?*B*VSkyGpUW0f+MN9y5!=P3q*oPa2HaOX
z95}Kg_J!cxXie?kUZ2_<gaYKnFB7O2nhL2nv|XFm6d_T-hH7dw&WbD<KMMvv5awqO
zAk{YD)D{~Ra-X1~U6s^>rp5O+r{0Y(W*I)-Hg+<%0Q%VIp?lhQ7-j?Rs827Oe?h{I
zmtPkRYYBUXk*}oOf3iSp(&x=5<Jz(fx^q67g<{IqlhDhIs9JBU4Cn5?BswV2cEO87
z3R2W406}ADZzpPVi`(_WF=ji<Hg0>ch~c_6a#{AW8|aBs32lD%9LQWV6TNX((%CIX
zIQ!+>ee#KBlq_W`#7%btslMKGjByjS;@=(B`pBXIA+vwKU<1j{B%c3t&THJ@RB4wZ
zE6IhzPtW=MNwSI)ALO)@7FKH3axQGreE!X`B-Txj%lQtzjKx?u(`K}BFUw@er5@o0
zl8M-w7gZH1Q<dhq52@C<1ts~KPomVm@fbG^rfC`-vnj>(Y0l%kK3H2OLnzxzjmQ*{
zr-Q2q;tXF=`(FlGAwY8B6_66#E)V?kGqjaUd(EI+qo2LO$q84zG^+WLY;yIu`(Unc
zOx1}=NmzZ=zpSV`N?aPoZYc(1E`=N2yr_P9a;8p#{yHyE#ksr3w>8{4?@8|S`a~u%
zv-<bblosvI9)d4@!>JmZ04g14wLLeC%ouuz$1G9J$<dx+9D^!prJm_duW(M`+x+`q
zAB1BKn!|bH4V3U2QtFN@QvR23I`ITLTV|pEv#Iif?VLXXdE=zMp4=o+V^Cn`-Uq=_
z@I93LL<`RV;fx{#w;<Jb={Jr~c`&Qc)Ma7t@tyM;CRy6f^QZD`NbHZ^I_D~L>@Vzm
zD9vD!E%ZMuouPJ?!w_+=xSBzKJeIs`nDvBqC0H`2KW~^dHF(F$ygP$7A<;6X^H`tF
znR2wBb`<b4gD>YcjO!jeE;y6K{w(vbt60U&NSTY<<eTK(3ia)7yXB<&Zk2LAKi}DU
z(6N+u+tfN@xNnX=;JDDcS=qiz;SmJPgp=7y*B~<xksq=JSx73%KZPb|5lIT)fU<W6
zRZjg-<9Sn8C-Tb3VO5KP`8a~}nRkY?y5K8b%dF7wH&MqU+U=@wwaN>#q_hxa3$^O2
zT^C_US2d^>Z`r=u?mXwvB~gSdQ%@Hkb3CTwUG{6;{u)*3Sf<JK8w7%DXBIW!_sG3R
zLql`rcnn2;3UiBU`EbWfhZ**7Z?p5`SI(BjHkPHo@i%Oo>Hbub@+#!z1*xAb4x{^9
z&lSZTCK#L?H-3thkQ>G%1_0R|pl8G~SkUTDaD)NzKxB?KmrufRV4S0kmUDBpBpYeJ
z!D!9^A`Nr(C&2usDqo{2PCU`yb4bFX{Bvh@V$`Zua|qeMC>-AJB>Wndq2@69`u&W6
z6jHDpk=sPBJ`g{$vzq>u?mg^j5Vyq3XOv$2slyzki7SUj?sjZ@29xvxXU7;kG|lg<
z&#>T175rK!WrzPrx2X!FE-6l0&2Rp}cVdNJ_e?mlEvD7a=TLt{vWd)02!@JGpg;~H
z#48V$ihYBI+LarKrUWwL&F2t96j&+g$siia?Y@5;bac{(m5N}-q-@Bmg4ox1c8qji
ziui7us?oWZ+3BezE7c-Re^J|Twql9bD)@UE&0UhsgSm=qF1>xv2)X2=U+L*bqh4hg
z9TF~CZz@!)>fzN!vTwd>imx6q^;$!2Vnv$tO^=WQU-=B)8T$WuH!N`_BKHO9$G2`P
z@121-IpU`hj}}Mc5_QbwnSz!Nzt*LeGp0MIS5oKv(zCMg;`we41@{H6Q)x<Awe15J
zXb$7CY){ur%f(n8{naYnt#B)r0*g@_#39ftFtO^~WdEW9mr94e8)_zs@pAhh<c<>i
zW|O0p#Pbh3=va0zMO4LH6?f#cnt96cycY}boAxt?)4Ql_$!6yB%JlGJ)4iS!p4?|Z
zS7Y%)mB#~WGdt-zj~T-%#28lGr$d{U4rxj1W~vH!z9jP?LR3y`TCpu(x-az>^-QJr
z;D0N5$2M8n`K-6JrMuOMO`~%fG5m~yTO<1?0ZDbHEK=>JAaGpa?XLdkttMWYmEim#
z@Tq)`&NRJ+m4P$59p2fbN4{0y_FC^a{Qv%Z63KzxjHd?F%o=>&9P;vUmIIl;&#TOQ
zaw=yO3|Z#NYp&O%^0|j4sA8iv1p@lZi~JpzPS5ToM0ndgMk^R~_79f&AT|+dd4sNz
z2KpP+ZTY=WY>6Z`Yk4qWQ!Qcg>B<Ocq^Lp=VN$gIi-|`;&-eTmhXLSfcx44@oec+k
z{aiA?>fVeQb$io^C8y^O1^D%r60sw!ZVua1m6BfkTlu43>%2Dva{E*&hUXsGP>Boo
zYK*)k8myb~Z_>Gxk)ZXHZYS`VE!Zu6CEeW|@BAY_BzANYollUvV=HJo_B5vRnZDd9
zlUtv$WS4Rm=p#!`m|s-&$YtFYTOO@ABB#@HWOLz2&?DFT0z2}#&#L@gLz%O_lZX~p
zCd}%=KN2yGL1YcmeQ5OoB@84{+0I}abTix#QvQkNqvx=r_WRoEG$ySBw%9t#ci11>
zUQMedzP-pU=+p60Q;+yM-SL+E9Y{kb2D!whu{nt=!B;{T6DpUlXkUprf5kA@$K^8l
zO<X&5pe0A2tk!uwJ>Ta1SddTQA>PHIS!vB<`5!8e7BgrDM;pd#L%oMfJ{AwrUr#TW
zhd%S!zt3~kc7s|S3z{&qGEd#`-Rwzjd1pnVzIyZr=Ic+og7RMMncHk$NwpIs^TTKb
zS$$~VEY>Yd$bk2U6wL~`)zA_|=a!M3an>cbs>DGF-@_pX$Wh*Vl|fP3Suqb<(rvG&
zaJ0A_{Uf4J09|v9WsMcJlHl~MEcximjipH*Qim4B>ph$3;q_yM6V4Gq_>p7$visfM
zNz)u735m?$qa4Q{e&g1uhy4~9f#r-|Q;8No9hzP{nUi~y{gM7*MxcssMc9whPf3Z(
zGGSOA*X7y@cG~7_nCGljcSZlIgcC0O^N;!_?D4pG52oW;=BuEsLqO(3Lbd39Nhu0>
z$YTWfb*OMQ7K(v{WN@L{$jfR0QYldRG7eU6s_K|*B2v}UI(?fqEL8^Nlz2-{bc2+l
zVeF^sS$YM_R8Y~eq)f#ujG1Tld>N?I*0L|JICZ_)I%YWNguQ1Q_Gxw3Bn-YHR~;wq
zJLvtTdS7GuV%m-jc|88G>Lqrro%MHvZNJtr@;j>}N`k^2ec8(aA%i(d*Skck^2shG
zmzbBH|5_)1m3MLuThVCFHKhCG$M!ntowoQ?zs1oy8EPSy%+b&iLdyWIHX?OvLlW>6
z5$zFzi;$B3Z`TpR=hW#upr_z_QbUB3m*g+uVx1~I9Otu{KHP2AM&0&0|HD_OVys`b
zuiR2nNmWL|m^$pF)5QXEFSd#8eeL1tEizi;7Y;2)+6eNk4fNxqy)?;1RUIzwlTWe&
z7(Qclw7L;*v(v>{2FA<hdZzK;I=rJUkDv8@f^i+nP2y8;6pq9NA2rOa4}2o!(rEo6
zlK)>Q@7|zhPq>ZNjh`i&!_xO5PRD!+X}-ML613;Yaw6TWUL0259fcXK!aE1hJR(XX
z%;Ta8I%SZ>0k#oFH?DLQ$s~Kw*aVp-=N?yYvZ2oLR(HLPDEDrv&`HTZ^h08|MyU0!
zCh38eZUNcKPkfjB&H;>kKfy1>o>`ID<_Fm02S#j9ScwLiQF6^*aF$JMGCf{Dd?cWF
z5~Aq!TfF@)&g!Kjm&dw&7Ry$LT&A>-6>*LR*Lf=aT4$OK{x!wU9^~Fy18Y0G5A=n>
zGBaE`mmh??^+09&&xvoOVYafvMOE?caO^i;klPi51jk2Z1j|#j_3wYJyBxZ2{B6-)
zn8iycf&C2V2XdQ*BNN%{!1!j+p^8eH_lR-{T8c4jPs92v&v#dUvI-6ElH8P|V@B8-
zz1@#$jF%uQ?n%${bk|`F?0FlOidT`_L9e{yNiM!T85m|MXW~1<P3!Zsj#PB$SMt22
zhB+|vg(m)qNWIBuZ9l`rIWEl6+fHtw*6BD0hWS?u+WG%UFlXKHr|k`tGdVMyvo~^I
zWrvxk5~b`VQn3NQMne;9B%xA{GMs{wW5^2TN$72hj5X5WFhMy06As1WO%s6A|88aG
zsIT~@?-BXU)XCk%qpE_`*JV$2{B5LLuPc^~@<Q12Y1ratTpe-}Z(6_IWVn0So#Q6t
z>O{-R0QVGGn;`d>wKvp)#05Bjqr2dc0(_Fx{7U%duKpnub-=rut=wH51AYh)?$0dv
zEFass4_e%*%4*`72BQgdysTt=$+t-jtF)o=nCR>GJzFFGs(SK?<{3#irccD;()Gj=
zxrZa;+>KjUU(b5cAve9^qqq^tT~S~=D3lB8b7et@=+k0to&1Pvokr73^n?@$e5lt*
ztZ)FKxeUBp2qtoH?tn(tSmq(b1I<2cfanud_vDhP!=bF2m*PIPkH!^sl?A?uu5_H=
zcIL|$lp2RYOv9gy3F(-fm~V?<QQUxS`Ao^c)b<C8R3tL}zLH7IR|az#byS-j_h>tO
zo&*Q6_mX>PQeJ+f$Ij%lp8fB-I>_xYl;(xO*=t?L%;*|QYB#Bz$T&SB)tzp0H&Ce5
zRQUIp1OpPD(4FvK^=8gZU>zf7bmA_c=JM2G?|pzdg;E*{YBW$up^-@5#zs)9B*Oeh
zI>ILOfatJOp__6zMCW*qYuYlfi2xZ^n;@Pp?qgF&BBd~lgn7y_{(AtPlz%=p`N1Jb
zi3<MYmbV460ts2$G9NhKzv)~&)}LXr$;Mv}_o(!ufhqov2UYvG&CHco<E4Jm7@r+I
znyinvANq~WG~W;jXQ527z^I+ae5j=L$~2yY4hrFnJl@pWnwBBir1eU;3{h~#@{A?V
zA=N3!2Vr9=&8jU!k`0UG3oMe$&^oWIqmxI99}GM;qRLcV8}JKIT^Q@c9(rr65_B%~
zpIV$@Bykvrxn41JRidE&CFas7;CN}X()xx<_H7&sQpT7|YccanyUh5XUhd91drEBh
z%fWM-e|$46WMM?>SN<YV+;f2gWJD-xjpuP~y>4@?tvA^VeQ7OF^hFOfR+t*GV{^pl
zRo!p(RH?cHIWu|W{>r{Bx&HK*<Uov;Y)M~NnfFfY<jh1XJG{}NzeUqB&7k?>28PpK
z&TEcxbrvr`;<(q8^n_ot^JsU6rI+ImiRIurlCOI)<Jm2r@uQic**B+Gb85r&M-Gmw
z{#0EU{j&FS=V>gS+id*o_LCuERkl<Cw`3M`_Jh!lCP?!6VNN5mQG-<=8wLNsyZqN^
z9%z+-<m^B8Yn@3mJUya3`6d(lK)U8dwcXjl4gW_)v0!XaQLM7S!|P5O8y}eS!l#Sm
zCf&*j`JcDvZGmTYyIh?W<{v00r)c5Mn-^ozB#G373*n{=>6UiS+U>X`z44mAE4wr4
zr_<TE?8>s-Te!*e0~G~M6J0a&a-AVlHQ73$zt(N>@ChHoQ4>_p-h24p;a}@U588x(
z!F$enJVd$Kzq=6G;DPy2vT>9>cT+Mfk+l^9>jq}R9tMUYD01dPa}ntS{|1Oh$j=8H
z7t)43fPH8)_#ep8eu9w?$L@WWHzQr#l#?Kae_3W`7ZRl>#S77yk65oX?%p%ff!}VS
zqeDs&+uN{FDG(SznLr-}|6OF?6`uy4`L3v&6N3!2f_9_qWBV0&ooQk-Dc`-u&=Wu&
z25f%Npg1E?FF~tN*iS^@+etG*&fqP+KgFI9(#%@L)?{{BsJHR?_gkKS7jRsAi-`*z
z4&qHD3a&JgG2KNE^67fGsI>%6Lx7G!ak(GXUW_{7DcaZ#QqGH3P|Zb<p!bz}T1Y|2
zf#IX=5AqLTm-uuZa(V#f`kGBE)>t$wl&r|vQG>ysJw$TBkIzzNq~is5Cfg2q<`z$m
z|M3ON((8%!w|JNC;3uj28Icx8Jo8SR0ORv(`W042`a19s$cYZ}Wcd$kX559h<|+^-
zxZVf#I8f7xBrGXmLUzNCvunc=&*GhJBBvdjtee=rfq16K<dfWL#x{Q_`HzMyCsJP~
zSmYOfUX6FF-(fW_^i@J2*?Och$2KLZbVgh=1cND|;&x5i0)#9$<$f}VeVrHY3^305
z%TB2j9wf2l@5b?_?VVK14_qLd9id%`$iA%VJ$_%V#i(s%N3G*WB)Nb}Yi;g8l>$f|
z*hisY{S%r>SKjVREQIMBLj|#jspRg1K#lI}3_p@Y?Tk%$&*1xn<q)$%!q=;GG!QZe
zY%>QPeOUx5TgIj>MD(0NwJt{PKu%;n$Xs7TfI$iw6i}d#rH|!-j`UxRE_hXeq(T3)
zWfvIg!Qy!<YZuU9X11EMilm7kslx5tR&;Fn#?RZJbAe135~VBfgy1i=*2hKQ$~7w`
zy21pQ&Rz!qHL2Dl=iT_PbsvtcoP-3+0w!ZER~FhULGl6V3All20WtF#jC5`8%GZZ0
zfHq?e^p(HXZJN;oV+bVc0NqJ@6M{Do$A6P0IxYg?PXn<9N$VSAoE27XqAg~bO6cR&
zsM2=qHT9<WPEQ7rmCKSmn)Byoh7vwUik1w1J(XwP5_49oG%UJsM)?U=P07I68BgEY
zkLnmzHC43(x61sg-UM<n?s|MYUR8rO$TrfhC@K38j&++NX9hgp0N+x6g51)r<$sFi
zUi!X-g1|!+TlylA8Je-`z6toA^OSCOThY=Clh7vRCPw6K3Dg7pu_P$sf$yrDjPtIr
zb`{mTcmE2pE-%&Rn5nS1Gz?4MH;tv5!CcmC6t=2Rt#sftfIEROtTl=uxC#Jfh20JV
zU5@?9?G`E^j|o#4CrY$z3#3pQVvPomaZf1z>7m!<*jr*EAkI6s!tfitp{4PSJpIPj
zq&&SoF)7yLyMe91XZs5s0+(#i(-lxnnl;qKEd<Slx;HVWvg3~OabI>nq&zA%*V{r%
z!fTmJwp?dS<EMSbIjhRWj}|Zyn1UVB(BXC5u%a3IH+v|*KE88o9GA#Rb4u~T`FT?Y
zF{%qs*i#``ZhWXJqa@o;g=jv?TjqG~Hr2p<=Z%-+gv%Kuuh2qme+i>PlZ#HwpIUSF
z0Gw(gdGY!Eww{uZih+ZkB+7e)&Z9z2AEb<8IG0e`-jGvzDlGbhsgK9SZ{+WOV?Ao0
zna>=XOi#F7o>&+o`4Vw#eBG=caN2w>PyGCg__0bwu1&_~?X1ZK1EZ|C>bUDkWRJG-
zb3^rUN%3~X^64?{S-0ojukTG{35Vyt#wK~(P9T5ln3kD!0K&tHOQcl5qI7G^6NS=*
zfG0!Q)?G0Uy!^ZG8aJ!B5kgsTt|ecGk)d5?Eb}MKVTjPxYrP*LSU<nhWyEPlN8rg%
z$ny<C1)SXfny4B}iVIUIVTQMUYm~8d0R9R;tL-d4ud@8z0_H0}fgH&3G1x(jDo#v%
zTAy0=qQl(Wn`TF>%1!tZ&O1H8{5aD=y;PBu1f4jAH$`<h84sYFjQ=t1FdZ5x<!B&D
zlL>sE_j-IxdsNT0We{ViZeX?SGgh0!u;m>s%?zm@hgyQg;Y-hEZa;eL$!IC*5&J{l
zkkNR_$C*{5l5p)nmU55Yj^f0t;`;*B{r9vkj1~SJuERCpVe$)VN~TXj+7Sh%GO}6B
z02SJK3@){%0mObsA?rpl9eMKQL0*r33v@z4XJes_JRnXeB7j$4wwcQ0xaOTU?4P|b
zvGv~^-jx>*kgIj$ANfLlp?X>MOTkA*Ye&4wc-TSDQpH1V<Ek~?x~YlzAG^460Aa?}
zQa!BSvRXTiY_Ds?Qo2j4DTA7JA0_KoBqYSxRO9d-r6PCADjK`vNkOD^mK{mScjHgY
z2NnJCWr@qGgpg;0DK=GIGkxu3LHX6L1=!^p<w;!&t7Ta<-`eF|0_mImbo{E_Ukuk`
zYPiXd5Jx6xa{wQME;`KdNDHV>kBXH+Kw7~e0x7Iylqw(?Q#|HhjoN19=v-GPRQZ7k
zPUqHWP*gxr78hm}u`%k8&A%u0nfkr+aVRYD^YIAtlgk-D))E@tX!c;VB9nz9Mpivp
z+-x2;ahCpZcQc{DpNdJRW7SP=kvS`N7bcZ2mn@84D&e@cTN=zs_*v2j4e#Qr>pBZo
zn&c=0t{J*n9amW&#KHw*YNwv28<<B*>`A~~tn^E1V%l18<%XkoJ?Z=EgCCq|k#?l!
zr?_0@P3H5cM)q$>lVkHaB@a`_4YSV&?rU#$FHekmreV5imb|AlvZh*2{i0b_)|VY6
z_VjiVxEO^WSxk_@?=fWQsBqj@Src#x2t<VABZ~Bw?yiN*?DGxTUNdLpkpUH=RYRNr
z_eH2#q38oV7`;w_MgeL;f#}-pHL8UyU=$G3he$NN*z;=(%4FAQo^>_hU;#l~%o=Yk
zrtMG=?PU-AYsP@wB^o3D0ZZ;bpZa5yqZQJ1@r=#`p9(k@`KqbIX3@J^H|R))spa}G
z3o;X-X<F1ht$)jvU6epBikqi5L)VEh!+r8py3=&mO{jD&<diC-PRQ6uNB(?MY_+Sv
zr_UZDmgeuL26>i#y%A|+Z@-rWm_1dO4)Aoy5{n&9r~ht{nx(c#w~}su6`Rn(6O7DJ
zbgJ$#m0QjXP^~K+^3#>v20&0UtZDLnMDfL?<51lBtdB^6ZXpV5uq{qTZA5MZDCAY5
zV^CNCBFX>XUjMskgtAKH^8-5!^rW&PULR$x104I$B$;<O96n7Bl%3j_Xx>5InnpNn
zB{2!i3+t3kZ@xB$;k|d7J875<cRdI+;chk0)-d^$`{Gv4ZZi$Lt_S-4*CTWS_s89b
z&R*b=(&09Gp;+nG=m?<B*|}SN5T;Z>GENLh6sD8N?gte}kRTUrL|GBCbKmTP%vaGf
zLy`!-pr$f=tdt62d1D!A*aIE$>a6=Wsws(|?{o%~OjeQAAPNpgR!5UOffz47&i2%X
zrf*-8-}av#|6O6z1JzZ*)m*iVki6(rG)BEhd0re8erH!7p`sB>d^C5BUV5)0zV}AZ
z0-&yZ2_KAZJYDp)MMY|>Xw!5_27q1A0g(PK`9f=cWg^?xLx=J+D7{-S9@Rm}(A@d0
z<dt4PXh+7STz+IE4XJnE!}|MQx$?I{1WKbQg-m1CI^wY7P0qhgwfisRLh&~g!7CK1
zbg&>s@>XP)%z$opD*7A#bxsDOvebz6VlQG09%C<?w-BgISDPB}J!tLDD#kf6NLFTG
zT0p~-Rz)K`>l3fv4BFZMo^{B_ibeDLYLsOCj!!+EWJi4GVKZ3!z^3Y9RMl>~<GkgM
z9T-w!&VVUz(xwJefF(Mzb<&Lh0p8f^@>NO)!#hi6^;y<r>k2wnBlg5+j>=kVf&yf)
zU_My5tkzCzp*<6NYy?Eh{12R3`fxl$#(}_+|Nr1d)dWwrSS1P(%u)P;@P%BMQh~%F
zBA<6uYx+1G{=$YKRQ)sdbRcn4{pM_dM2YoQjJwKYam!y7C3%(MSUBHzdM-`arbeJ=
zq}ma=(#hkgX}f(x*=gfdjVxXl|030a3j~dvY>OcNhfvp+i}(GCDej?9ym*tRlW>E&
z3*m=+pBUMElz!JbAFp2e+2)vmHWS}e-0}_cl<>)eAUniVc3_?x9Kg1{DLf2SnYI13
zyTvMj!b27T5>bYdNTLS``1*4&1mQfP+LRKlv{^`6oQBJ|AX<4j!}!p+5;r1KwH&=p
zs|n#@A7y=mKx)p}wV+Ix<Xz073{gAlg9Bag#QqKn*jwyP@jp|F-LwF{fqIU@k?#0o
zGW?J#<L6`BuRaochYcJPX$llSoBN@8X(^1f4(-CeyTfyXcWjjor6<o`vsaz-Yz?=4
zakjHMqPdSpnqCap^~GW7m5oCqwyN=Kp_sS7=n7L6ugX<S!b<E7)W6F<MM;?CF%O~c
zV6?1ZoY0J3V@)(7tV%bBOTuhHwFIMw`2zlh-S({(xBp(-qmXoJeMj<LV;MbBx0>9q
zQi9A@zm3&EGN0pPi4XAJdugg*CFI#cTRL>S9R_9k0<&uS7qPci;)L?4YUUt$B+TK{
zVy=|tjvRVWd6BB_C;N+<|C26u2MBtrc_-3t4@+6${!CxYPUxkR9iQ~P@Oj)@Jhtrh
z!{e)1heOS3fIZXCK}D-)-|Vu$HI8oK8g`U_$G4-{pVMsy1FIcDTMzI+p{^Y)z<>;;
z57{I1b>S8ns7v7Ps6bfCYjITQ*B-6(JBlD{BVLiV;={mb=M+zTO-bN9DX4H5!Nyh`
z&;&I8<3&Z%r4dTvmm4kfByr!Pl;qN9f{E*jFPlns_Ky_5S`}q;)5c9D97@8C((991
zeRqs6hubA)pYdMsZ5}^TRkeeKogSP2)6T#Ax55Gp<=q9st)Owc_j&Jl3%xGVLl~_d
zMI)_oBJGNHPJ$;dH4an%RrT+S7rKJdwO%!-dd8`;{_Hc?<IKumc-+!~F(6m=l&%P`
ztwU>1b~9X(g=_%i-=qJoA@oQ-FEsgkL^}|)<Iw}_O<>+GVp|shwmH$~Wv#36)_W+K
zdW@mxu{QFzU1O8uV(APD7`=&Dz*jd!csA<Bdt&z=W?P#5Q)(7iINmMLV+=po`d}wk
zN7AYD#b`o;gwhQszROm#=pek{LDgT>X3rh_Zi_ME^fW9-sJfLF`I>NH58G{Dw3s}}
z%yxrf^EY4r?$C3_QXA3Xl8d?t!s$XzYw)<>Y`_J~;s){dzc(p?!$4C1R(fB_R3+|)
zT&eVg=WvR9EMu5#o-KvnqiDg1KWZDQO?j!)<C~aqCN93Ach|(5ut|^e1+Iy9ZzV^&
z7Uq-Z^?Z2skb}<$(!4><$v0zmmvDu=C1k6C_HE^wU<(!j0l6}7OGFz(6J0?DhZyPp
z{re9#*+*;50_)yTWNEW_6lQyE2u6Q>zQKJh?{a4n{krZB%Nly@p}%%z@aB>Qby`!r
zxs<>5HmNfjKXqOY_pRA${4z5%vDrpFjH*)eX|?XN33qC@{RP3sK&Rvcvk#@C7S$3i
zan?bB<LUQiIL&F}<ag+PGbx|7`2!$$0-`q78hJ6xUPfv2-RF=igL^e3#L*Ekpfg)D
z#T@&brwp<$VLlYC#e(OQ8Ib~>FLCALT$i)o(Gk~2`FEc!_x}lMMF|(xWwN`+M>|_C
zf0-{Kp4E<rOnPxXbhfVz|Hy7<v8!sw4*VrQ1s>}72iy&Hz0t>Xyof`97P=S@kR2p&
z1B_0r49*Mibr1qwt*t!prVxTYPoF+Vc<zcELVds>T)*iBIz0XW13=^njovqe7ui3F
z>x;PJIJcU_?-Q%jOUje9`4m@gj;Jju?Cyt0CC@wu6DzTDWBC5rTdDE%;^Oaq%A9b!
zQ>E3s`BhPaU+df>Raso}7q<F^mf-k5)XU$q|KTWN>Th+;F`J3+ObSLyk+yh#=S~&W
zse`*6O}38c9rJ;=3I&ENKpt@B|C?GvdW1e?FcF{NekLLI622Yovb(Ml1uO(2{D~S!
zP748cC^{C4N2j{SJkTcv?Rw61wTh(Vva1)qmDqc{m}zmpL78KkZ?m(jHSaTSmdNP~
zH+)BYx1c;!KG74@WA>y3&pzc$Ke>EYG?kJn#Fjr5BvI*CbaP(Q-M+86&m)K*>Hzpa
zZ}j>OCQLb+aM_JBfp(({$X4p$%uf*cA-q)S4}_zsL(e2}m+5c9ncsfu?Q)|{ny%7a
zc12=~A))4(%EGKW&++#wtWqLXHRG+Zkrc5@(~w5Sr;S-w+S%R@WxspM(w<qWcMqef
z@);T3teyiIH_EaxpNcXRCh+}f+cI!dj;e&cJa!FAYKTgxO%f9WHF(j#CRbuVM$lLA
zW@U7(7I(uQ63U=^-wW?_1at?zMSzh+;Sp73r#`%P+-&L9;a~qJ`4TUhOW=&Nev*7m
zeTGU{r7U%0@ooS3d)u9VCca2J8j*W^X8D<f3t^tZCmWdcTE8Iv2u<5drUrS~7q;$Q
z{E?;Tdx4uKrX3`iXckQLd`H*!O_ESlCy`FdU10IgdtK`^b7xcEAkqtz-zdl#N+OpG
zfCV8%hO_9Fkug+RyAh`Z%ZIBR$GkV1pnC)eq)&RW!ZJ*&(ZEdQ!nhe!qJAPKR=#e_
zFVYn|-bC$lZ?iNe%IPSXTiM8Kn0dT?XZ!hzpf*p$?gsjM0j=JV94!l^DR4ozg-j8u
zzwu^N_@SII%03swZO}AE&J-FNa6<rO^XOoT23s1aWJVx&gcpodP0*ARk%d>PkQNpq
z@h0qs^w62ll(1NJGcj+?PxJQJeCE`TFt{af^mBpTxw=(@X9KNnD*r9n7?c#Ge}LXp
zbAaBW10G*gvzGJnbk~Kwgb@k5pp1UW`|-p(Qi(hD`5im3YQYBmAYs~9B9a~BVizRx
zXxBugYk2r+76x;s<agZN=DvSkSAHVX{8GFzH9V|7FDPYzQd>h<>%^^9@F*LOfWw`(
zR<^_LLF+&gz6S~y)PMx%8ChfmDm;TxD@&3fEaqTi36n38(FVY9@BR(fZ|%QS>^wG1
z(;$5RJ~Z?}5|SxK-S~6qr>U)e7rbPv_GRaf=Lk%*&fKw3<l|#wEI{H)y8;trGD&|O
zM%_7G{yRA?n9JnGCd^DIPgWjCm-Ut7^@6^uTcxY@Ftg<mBYlgrxii~e-6a2+iUQ|V
zTG7Y{*4@g1AlIqjNso(C>K*0GE8BB}Y`=0gDAo}B9h~gg@Y2c;nKBccaS2>ZMWb!M
zL?(+Vfaa2OK<LOxU{tvTXNUtlCqykoP!bHh`u8me+CGVVz7ZAx_&Ulh|F-}F3Te<A
zNR=z2^r4i!gajDEkOav1s1>Xt190I<DZSN2;varr83Rtb38Dryw4?cR_w?}F7CMfO
zxTRHf{-A+V_OEriphvF+HIY9DSZV<zi#NwK1MIGiLC5(NZJBh%l{7C=(S0_zogN_Z
z*RIcvwto#}hMgG<{me8<#7NuV1Iq4ln&Uq91zo7PMw<_(npbd6_3LSR5xR2<DlQDo
zisH}e932U<w1tlHee|40JCS$O8ZvdorwjI0;znpc+#sSt5ArZ1O+t4+=vS;&4EnIr
zVO1AywLsY#WZQ7iA)42O@XesxA}XUhtqy*JZQ!n~X#Xe;sA?h<<Mkpj8Uwgi2~FRm
zP#0q@kB2fUQ={+nxm!z@@s{3)%>6q3^eqZ?D;fhXQJozp3zzU5#S^ZpqO1GenfUmC
zVpmOH9d+@-61PAj?=6{*z|A`E&LsPmwy+Q6H-?0zI3X6;%&OMQocS$oGx&Y88>Sm=
z{kM`i%AeTol<fA!;8hi_y(42DQ#;|?_1HJRvZ&1eNnwEhPJ=y{-EU)bysK#`xG;TM
z?d&(4aIWEE*pR~9>xwALT<a(cizAd@7b^<yM?ykBfXsbhhT&}ikdc@M&YjlrH}5QH
z6<HBr{|{ZdhkT<BoO~S%D0W$IBq!;lFSv_W7(B9(|FJ}P(rr;S{^Vs^9&XSDFJc^q
zx1vmfqX)=>lWz98^@U?jX@@>jJ_|R`ja_=gEL@CZb1Q5AvRjSX#~y;g{#rNutx#|5
z<qw`(u=3^L3wz-Wa$MrvH5Y?(JZ$J!j!Xo2MDk9wKU2BX4!QiLBY%bS_atZXt^80N
zc!w@}G+Aq*h?wYqN5Vy+iKB4qCto{!FR?O17++Dz-zl;pYG?~U*GpGBH<W8v`?=})
zEX)MtQVgknmn$#q9eX_#*!{kIDDR`hE%H)0#tWBaVAy7W1qe1S_var1sGMGQR>ws}
zXyvBX|9;w3xmX<^ml(j>7O=C}qq&&Ts7o}X;?97_9j%|!GBx6BZe&-?a9yD~{C6=!
zbe^;;y>iwdtJ?b4I$5_jEw=5CGWCvQU(VJYQzBGDnY!&gYIUGAT6od6*fdOwBJveX
zdQ;eZ%jm!+VO0xPRaI72v(RGfn=R)g1?tXkA>+6Gn6NR@<<wQ^LZv>#fP<1UsDDtD
zMWnJGGXw9`WG20BbfD`++xm`aj}6Kl_0-&Aq({!bz=;1!e_9YXxL;=^j=taNt^|SP
zO5ZPdM@ZEwhvu;ExiB0}x(QQtnx}9Bq;(|k)<5yP&pm#uZ$V3Sn`LE4^!*PdlTU3f
z@iA?x8lEGB=ZSAnh<$~nR7dCu>z*qrsJ)2V2cs~qd01{@9l0sUgn8$-XK>2|ggA(C
zYj+(d*Z&`wm&CK*^cZwz`45(^R7sY?en!d*rt<Cg7B5PA#r3CiZ{N0E{~~_&s|9n}
zN2Mun=l-~1pRiNI6XdVol~7;ra3SjdK<tOLFo%8|uZh<Hor?&#xtj^@W3%@3KE`6Y
z^mmb;{PQf9>=eDtd5B5i*Gez&)4AUi$s|gglHU!I(j3PLH!B)~7m&_RJ}O-G-%#`1
zD?kdo3e+$7(hoG|u-$*bjqfn9kExKCqVpygl-R?~qR_!Y8MHFcHqd2<;~(Hes)HDN
z6vE8&)N6kd)1vn!rlK6=hK+hC<)9Dw8yxr>f(-j7!N^>q(^p<>)1JsppQ|>B+s$Ld
zmfS*A>7_U)dPRng3Cn3!G~tn1G;FF1<SI^^*vrSh12b@M^#49aYkXS9Zsh6b`t5i_
z5*=0#3ik>tJd|q`o(D_Q=S;*QK!pVdeOh*-bp;=mW43po-T~x?Lzd05CKP17oE|(8
zc5usk_AvkL?4IIVb1e%Yz8TvGa|8;CTRt<kRc~=T6=#b1NythOlgL+_^_`C%O0|YM
zs`UoJ6__%RcZkitfx7{R(LISmJItSlc4mmz*#iP)dJNs;3CXYR1NO@`gO^*~%_#~6
z@C0UoRr#t8S!I6dyu5)`;%W@I6&%$`A&<}R>tQj;p0)0?W<g~Vuc>B1eCLBp#P_qC
zV=B6dA(MG^vj(@Sf5!^a1l6Lw1h_OrpO$e*yciKA5Xuer`5DHE$Xmi2nFfh!0lKDR
z?Y11i`xWx8u9x+NG<QcH;$g`q)@-kuRlUJ!Rxu4BUn4NO3s|nDy?wktsdU4BiHz?S
z9HW;Bk?)qg$#H{pZ%Ia4xKU!RI-#QQ*Shtu%dA;v%FGB5)z6MQfC3p>N5aL3xK03H
zwZI+fxirZ5#Q#5V2i-Qk-}1aDrSOWAI=Tp+pcBQ1@RaobpuPU_f`+K|PgS@ASiC^X
z*t@JTu=d`)ib2sqH;>Of)r|M^MrTS}WB%K%nw~}V6AXvBvs}V$w!I`fc^>skD_n|$
z+j=|j-)~w55UM*68nOP$VRWH+_-53|qX+>B$~AN&NGM<~;J6AG0|OnxcuzBweeekz
z=hm~|875m}9$$uINx@usZroAKLo|_ai;Ahf?amVye|QppLWWXtUjx}lgR|{Nkf60-
zAyl)LCdscgE2>`Yz*opwJjkY6NhqqByhzz|+@Y9K{Sqxu!y1vrSK^RooP3?q1ECr&
z2^xhPR8osB9e_*dVi=Sd00$`1S1i{Y^k8a4jQ17vF&OdxW$Nc+%?X#_!jywce(CEc
ziyvwAC~acpYt~#POC|<u)`Z<2?;aPzF9+6u!%1N$3nk&9@FPLvGjTjhL=4iIBquXx
zyPzzrn!E$Wpu51=tAB{GM8;T6J@RdJUvisL=5J+ub_=m4zqS($R#kmgv3ir>l3MVA
zYD^IHOH|=cNm*^bN5`6W#70$x^3k~s_n`+P`HvAHHS}Ods<51b;?;9I$9fZDVgOnh
zY7g6^V0<gAdKZrFGwGMRD@#kZU!WbK#F#xha?Ww^#(E!eq=NAt%qqdikI|Phw#8x<
zvvCWsQt=A@17w9`5<7OAYR+zIpHdji_I+H5(D*1Lmw)Yst(06g39&_C*?}$ugYsvT
zfgkS?w9zusuXVgj8B@EAkKD&p<|WGD>nY5iAt0lkYpL~P<>87|9q>DU(B`^&pE>)7
z>~Nw9Sijc!Y`9-kF-W(4ryKDJrG`BUu$aJ9@7q&!y*=_37W?*jY#10IP^!a+@MwoI
zBUt4+upDD#&CucrO?)Nj4y+b=gtN+g6sShG!(`m+aeFpit<`SkmoECmJ21augpkj2
zF#Wm0Dws6D1rkqxD1aRYRGx(ejDB~Qbp~8NlHm_e@|zXhPFiKv4UYpE3u!I$h28Fh
zR_cuW*Wpy*jckA#Kq2c1*ah0<TaYg?;0f?q_kVpn!fp=>QJp*WG;<XMFf3=cMxAuh
z59Lo!?fV+W4;51}KMNcc7GqZ`>O&Ksq>|#a(<!Ssx<9Y8JL8Aw{TySqnei0dm!o^@
z1kmXwqO2CVO6;Td!V5Q1F9fPTV7~y`9@5T3+Ie&|QHC1T)BrIoRzXdrk;WG5M1*eu
z+zXW~G}oO)4Jb>__vz9jA$yO7C=jBOe$WPz?E0?q{JX0&Jm$Z-Et$|%p7~S^n?)7)
z>*Y6QOY*r#NUeLCHrIJS%L=3nI|bgfCW1bLcRGGO$j_eb0S!3X4Ua$@bO(}t;Bo7J
z9v_CYNqCiFA(Vx1h$H?EDwbe_#-Mb6?m<femWw+UQ$C=ZR_$;$U-h}A^4xpq|8`cW
zFTW;Z_x0<RNJZ-0pLA^(`09QlP)kRfo<$69R2nhf15kDDbu=Gq5Fz@RhzOHg0wo!2
z3Diz6lx@Kwj0sYCIG|*HL*EP95d@7v5L^3vZ<0Z7OBLw|@%>LtzR5X^-EW28Fu*8S
zo|UN{ZiZABt~%{bJSz0sgv9B3w%14#G(rCT5dandRD{3-l<mURP>^q-_z3YDO9m%d
z&ECRWMIlj#*SdI6M}`R1Ylgt3JAD-u&WlIT-^5pk`Pvhh+BppLP$46Eu@}K*T=c}y
zL(-hs2`1P>RtKBFlT`Q$0U5a|?6=$*J^A_ps`vFz7bDZtI}6~I1djmX=A#E-@>@`0
z2EMDa-_V%_q#$^xx2%|#!V9H@kqX8TQkabh`Y5qV$`+c!-!j_vhj#8&nQ&NA@R5;`
zu~P+un}uU#v7ygV2+YD;h0YOp?!pz322P7`ibn{nl$hwJ!0A_Cy3=>|9$2m!!ZkM^
zY`H6|>rIjZCN@w#27n}Hn;5q}ED~?$+_#enOZd*g#TnT2Vv7;MiHGs<LnakBnk$Wp
z7warH6CAItdIV}SCh(H^`j^TcIh&cMy${N38A<1!j&W1f*!D2{H&%aY<d)EqgIcdf
zI2!*iYi|M$_51#hDv@kawj^sZ#uB1X$U2OKF*7tHTPemaD#?=UWNU_!>=R?MW=o{(
zQI@e3l4M_#t*o8rJ$=8||972p{?|Fzbxzm&)1^<-d!G06exCce@7MjhUpLx8wjxHR
z&p+$*=hMbnP?5rn710I5*%n=dArPPU2R4s?0P}$|{S19$tgl@NFd-mld&nd|dSm0O
zMv(!Ekl)h!kz#SKj#oJUj<FdwOIvb>*Iye`Tlk@l=QjVJft*4@;8`^wa}!1IRP^w(
zV0#b5HNOn>|BjBU;J%UZVv!tlw3hSE*o%bDXO({$ic)PXyg#|UG{#LN)5daB&LiCN
zvNyM4K%>Rb$E&L^x!0Bv^MJMl1r=QH>C{-LY3R64%Fw{CdjY97iod0ABWgyj>nHCp
z!~H9?K2S!^^9~%-7tdg_=C}vP`lvi}x}%{pcHYJI=3YTA$LRZUe5iL#9+UTEyqCK^
zk(T{!!tk2HaJu5kT+}V$u9mNPpk2ZYS>HT9w5Ixb_?_<#RO>>Y(n0C?ds&F5Bl-)q
zS0Frj7TQvP!-oJ$!E+w~BMg|t$IXHW8d%^UD+{0i$S(#!j-aiJW$WyYLvGG-ra<#&
z@E6=(!iy&Gf=^l^@6bCHCpKy1PFm%j;9PQL>d;03&}5n{U8LTk*61x3Qx?9~)Ia{q
zaG&3w#QoH^)#72F?CCuWg~z900p%X+3Y!{67q+*knYm0;c4TfmKWkk+@OdloJdfqG
zkfd9@B|$&pz3`j&Vg~OG0t^E5GJL%Dczv}WD!8I6Go~j)8$Lc>L@yGxjZ4=uEh*U6
zbLpnsi~SY-BYcEcYE0!f`j*xfTr%5DFlAt@coK*Y#Wfpo=SNTs4O~C>-+qGohSi*2
zHoKIb7tt5>C1<dMW%L6W=5GzXxK&yI5%ni|`Zu0S>Rrh&pF6K&{%L{ey~uqN>9UC&
zK0Zz+TVv*e*ji5lr1g4N3y{4r{fcM{Q3g>noy%y-&i&gEwYUL+f-TsU^t@LWd`|!#
z5rW6>Dv<v0-{C7HfykiC{I|qyHT&}O&O0irWLGpoBI#99yRh0L_^mpH)B(3oFY>dq
zc{Y|UY~5xiylkHiDbjwta(5l;N5O6vEfFpo2Q*z3ENJ!5zq*80DlTl_kXx@IBt6=3
zIfu?|6CO_O-{`URM!7oU-GrPdn`yu1{RFPKVe+$$Gx}^g8c#?A)nu?s0I%+mR1Vbq
zHb8dw5-_Ka0bUL=8zEY{V6pc7+6p>$RqZJbTSn3UtT?Z7w8EMLpL#D=JtO`*`Z@`}
z8IX%OY;^6o3&0*KJm9@`8r3i6kf)ANF>S<t=H9R|f5~s%QHS1b-BFWns7QWeP^k>c
zfxDa!7@PO?bZ#VUbYGHq=55vU(x70&K}SRW*yV38d7k9&%86cbnn*mNDrdyYZ8%?S
zK`+7~jlBmTC9wBCA(};+1(3nt)n~OO)TJi$cw6;p0R3B>fxeO60^IP4X~>j`M}4}R
z(4W&r<G=X;5Qr;LX`7fUHid50ONZA38v(Tx@}D!Hb7U3(p{x*}5hCs(pZ-z0{Q8GZ
zN=ruEZ=8Rfl!oBjc?XLzHT(Se3d=py?hH$g{2bvp{#R;4bDl%Sa%295#w6kMspQDh
z<3X4a7KtCN?0Xc@y{C+x?*lN|0AR*JX9*}B2g|03+<YtN>&MrYVI3@AtD>1}mf$g6
zA5~Ur(a&#FhmtycqTnm_Iaeg-jdBKEQp-kzXQFS9WETU5@m@$>ezvr}+9DQHP}1&?
z47A}z`_G?yLQWxt0?@xDV*u1Cpg}y%6aN5GL<es;#7GJpH@<C2sBuEz)tk7^i5E*I
zfJ-4114_jjwTOirN1ZCTnQIp1n-A)YijVQ-$w@Ux<l!hP<;cjfr3X}RPwXo1kt?%c
zmxrJc1y0p!;cD)}ucty=BBOtNRdhU9i9E!6Pz7zE?}F9RHX`JZimvrc13}4_Gac_H
zE2Nc+zmjW#R+xI+&-X%S1~Pglk^+U+Zd%Dn<N<rcsx1PYi$In=jGpb}X0>)*i7s)R
zeEr0mg#Ta!dVRo#{K~jfX)B(uN1u%}tfGYqrv2SS+Qr-9Vza%9WWC9?*~m^(vM|_B
z_;!lKRo-fu8zpsh=k3p_7FH&898G<V-+$vd2ORVrG71vPj!Z1PFwpa*c<TsU?b<5b
z-=nw;h?4~#He0AH=Ox&nC|(F4hg81dD>6#fauK--PX8tlj{#>LI9>ld1jzM)%FILm
zWoV!dQU81+sM-OL8g+gcoiu`N{K%VxyXBSJiUDa(){^KQ|B~_%Z>PTKTKHp|7E<?%
z`P42o+g3!2nf1(W`g6yc&nvvOKHsKNYO_E4*YaZF8LclC$|`ya)Z)sbaqU1yYQf`S
zFT@!nZU%sgs{X==nZFDxTH0633v?6zGCW_Z$)VwmdUJTO`BawAS94}_QigS2ag$Fq
z-uFe$C)?K{pVeQZ_hg3LCYlO(Bqrd^0kc0Vau~p*bY&QN*D9#i=x7Ll`~gmW>)2<o
zZOp(wfe!QN3?ij6y=M>gb|_G(Xy6?0@ebV^!~qx2*1rte7LJEcXemNPYUU1AmJ4g!
zzWHgb`M-Kh0crCt{A9oIn!~;8O|2BTtkSb0NA=_yvw{h8&n}ZMg(>~ba8shdJay)?
ziHUIO9L8czibgWn!}f;Rv^OK#3wj+3(1RutAvK00MMk_u8gDC1Mk~RdihBupjh9Os
zg^rju04x0cJ5U?efljjj^70_|T60q*S@`RvRZyue+4O9Vx3bhe%)M{a8xsgB^D|Py
zurk|`=ztGK;?Ks4>%_^%;jV%+#%q-EJad7_&8zTYHFJ4KmGa+w?a~)iuDa~D=5J7+
znaDZbFl<%B&8Mlzxt$s@mrfrtVkN1OV{N@q6C26q#nKiPLlzE|MI>kI^%r@+whmWq
z*0?MeEbZ<YfsQEmDUjg-xxRlrbOHH<-rM+J5-L=AQt@4q`C?N`<65aP&x?K&VAKaL
z_B}x9IT$~UmV0E?akZhGy!Xx3P`D#TiP%HUx$?{c+p8yGDW#d5&E2#y^6qu-rnLBW
zo$2(N$MjVH$ny1!jFY>10nES-0kl&BEK&N*y@s-2z<Gk)PB0iKUkXBe5CB8=3K&6;
z`pbVAs{2QPzV#~b3h&=H{L|Yias>SAyTBs^-ajW)Wsm#^7IQCm=j=M-@8DRr{J<IW
zy|vFs%j<#~*ez^^62N{#k4sagL+ZU6#{f?7m*EO_1jHL%oq$aV(fy%v#Vx-XL_biW
z0)7b?n?dzI`DidthQO3U$wgpHB@F4snUIGK9zg7&V3zS#dsPtVnGNObp}UyoL!?d?
ziP`Ib$qH4SgS1srD9Cw1Yah51>Iw*U$j1heC#Wz0X|`D?i{(kbPhc!JEy)M^Bnyzr
zh8j$QJpv2?y{%swaEdu0>i8cs`W>LZc(0SPPO>Qn;{_zPxd|w;1*4$RG_9do1?dF4
zR=yzUhy_@>dP_cParYWN92#UHghyI{F9H(;B$(;MX>bmKtOS&i)(4;0vOK`uuM<DF
z%Je}U{$7iU*`k1O0>~e+g18YZBQPPrtCI$0W<+`yt*t}ACpQF9^kzYRi0oQYHvC)I
zO2(<wuKnx&QfERPM8B#!d|hAEr)3miHfd0j2O1!td4}Hq@DGipbECkR04CR2f|!j&
zEjG$Xr<@v_=kV|d!@oIlyc<m{Wj<L~Xpq<ZnsZBIX29$=x1+wd^6y&(b=)IelS>uc
zUx(n?z`2pozcX1-OSDM1xhYbBUIYyz=-FB56@W!(`i#!$k%JiX^d-c-21+B`z(WHL
z@(Gy4%P_EMSCY`)R-it*GrGDSD9u<W6zS<*>0z?6;k5!-xdfk?ORnXr%4*A}Etw?D
zxNXujjLH%t(JyHGu`K$tAvN9k9@Pi);FEVn5q1{iwU}Q}-h_X$VFG6<68vJ0i`K&m
zl??gIN~<GRWpBz1A2M-dorQl&#RTNwyD;{|+G{~=yQml6z@+{&hyRIzK@J&A7BE22
zV*VH9&4|WRp*IXEsl}%-k;4F~Zojs174OMu-jgrCNWV%v0yIXGIb`k>WW+slM_&C(
z$?10gCiuwM*;OC~ewO=_LC)=E8qjrW8yGS$PNls{8agfm8+cBOPkPl~gdkUYMlq&6
z;;YnX!U$AZ3J-@ioWJ>EvCvk{_S$Wp3l;YpcYMxrKj&xpEtB@V4gn<mJT|oo&|m&}
zAs|wZe-4gWX#E2O82s>Wi6Ov)U9eaO8~SlGps@l8bc8T?*g$sQ1Yqrxcj37?w@D%{
zs3MKAxKz@czF)LKgfGyLH5Z(<frF-;u%Dm}SEryRz&^?g=B?VjHJ)T9bU};)kAgn%
z2NyIUCz`6FQEo6WK?`>wET%uJtO#lYT73&Jzj?p*bgj#^*{_e{ib5XSlRZxG?l@`^
zMhmn&J^{P|OfOiAMCcR(KgQ>P*#)%|Ax5A%4&dCEfrcDU0R@%*xwzP6=p#`l*So>6
zQywGVTuJ8LR*CYcR4kd>opa48i(XZ;tB}MqsU!o%awEqu&dC6JmH7~Qb$0Ij88=Kt
z`m@%#b3q65Z0^0;CY^8z&;$(Xlk1%N`OhgU7#U6OXD2Ej72c?*&C7Q!dt5K_OPd<t
zlM2Kpo;hD-mMtB|eLiGvFVea6$<)>IgMQ^+!EXJ2sMZNeD${8wbRq(<L-D!bLBJw`
zDsygHiGt$g#~SxNAzT9#OyutNK^qQ+LcjGXtV~@YS+G+&mQI?JeBJtQqXu;YEOjb#
zNpPMUsxGcM<FLu*e@{`PrLvZv6ibB5eOznD{v;nj2lyPnm;AcDsiK0?Rwaf?&3<)l
zw(t`IUY7bbh(nt~eWs${E~z1?@z`un`lNzCcY73zT3PZ*T<TEN*2x_yx6hDAK<9k}
zE&zZ=)Qo^zV*!!X5Iv?yNQ3Pk03^ol^}Q7|a{<Oj6qxv=yJudygg_l~CBg&a?)d?9
z+lcjQH74woZfdHUWce^lNN!%$r$ty_X#OA$5)HH(#w?OHGB-jSJnQ838hD<%#~WS#
z&@eD<>{n4CW-eA(?(g1D$#}T47jH7y|8pnm%dI4V`ReDd@f}9<GJ}#06$bL^aYFzE
zhP<<ce{>jdqOGC+dGq7YIO&b*!I9DXUq>If{2&dZ3Y`n_P>W2+Z-ds^+zLH2{z&RO
zpdw?s3?kPsySwu)$Kfqf)#lHw-#>NQkgA&9_IyF|+<JDb++<gOh;BRsz&--UCYT@Y
zWk{xwo@9nv3UIG^Y6C706jywYD~FgAv!=}S_QCX}ogT{#{yq4WV33a*em5EFp->cd
zt>`kp5;o2Dm*pXo&r*w=MsDZC&-50nycV9tMSU=~+a=z$tl+<1Soh(U^xGbSLlP)r
z7z|v3UJ6}8o(GEJ)K^T~Gh#%JLx&4U;OIi{6X++sxaSQGRO7m|+xNgB=e9zfxW^4$
z^3gMb6IZ3=eY0g}6(n+@PQ|qd)5F&TFwq_lPN=JO+lVjMzah0h@{aXwt2X%Hqrqps
zG!0Cr`|J7w^%OZZOO$yBM;>d%sdhZmuyH+WFZx!ZzM)Vh=zpUY`6FKLa^5m1N;{dZ
zb~5A90nlLw%u`Rsi!LZ4+Y%6c<VkN=U~%g1^o*TO@;BCJ&c!kXxu^wkcyxlr6%*-A
z#L1j15i5r(mTY~G0k}K0{j}1M7R#qs{-nt4Jw>H&tdtM4&mMGeya?q4g16Qt1y3AS
zQ?li5iW6zaFs%ME6Arl1zR_^h6-RdTBCc}CQW6<GLxyKBW|!6aK32Klb35bbT(JRI
zzyKqD?s?5Rtp%L^x`IX!vbPB(VxE3xR!;H(>6wq<_Yy=m_^8pTKtRl_hP&xigF-~>
zBBp%!Yixv@v>&3T&ldrIjlsO(?*#2CUw$%2U2bb_tm{N**c1-G2k^?qVJjD<hdW<|
zGu_{(dzh7sx+6b^q<(G9t?JzfBn0M9ynbox^jqwsYn`{aHuU+JN+2GshC3%G-|?LB
zbphPi*<1Jr`&1C7{a>-c<CEO)fvn?6fT=^#JuvrAS+yar@?Z8|GaO{N13}V}bY=jW
z;5X6>T7wp_0m&i)Ubm9?IE|WuXF@mL{>xAcQ~`zpwXuEr#O&epm-@Vsr@jGgU16oF
zxHv1!swY-3zwn9ljjH^GX8{O;jo4p?*zD1Mk|a!OBC`L(E$6d+KsIO~$NhMy&>;6~
zgO=o5{@3sxS6U!H9AX=Q22cOcF&1EApx$PJfMEgI$Z{~zUVsxDILERHY~XuKN0o`!
z2RaHCU+m$En>mLpzM)qn$nYA>Gl><_ZSybRa0L@O)RtVO42k}kha_uCRmh9>%M&LC
zi%O{&x$LAR*4Wz0h8*j{>|y+sg0VO>p}jJ^p~jx#5`=ZZ#ecsvm)Hsu?|jTOb7x-T
z4Aju7@nqJ2XrEjFji<ZA0PZ!K)fQ4R1#W5pk^2S68$A9DBKezE+^qP!aK}QI@Lr%`
zNB(eKE=fg9+2K~aYh*KDvh-NwH?LW_7YU!sFwRk?5>Mrk4k|~ZykoCRt@t}xCG^D!
zc9y%3xM_b%u8`BefBrOe>s9(Afj<;j&43S&cTc|@?DWwQ(hYx+hJx2^ZO7W`EPsRE
zS@a4NYwO(VDJ5;DTwl=s0@UgP%exB)4(YvakS+oSZ8po)<H`|r@3gClMQ_6IVUBzM
zWe`G!oIE{89n^wFso<hNvxUET3OmK5he}%Yd_mxS6I2jAvmX82vd9tm(-4<jhr1s-
zq(H@G&j)qr90p>;3ZTTT9iUR@U!8sYTt3AYgX!lkh$Qjt9lV`gmUj5HZ$U}RWPMo>
zyRGx0x8zlLHG0P;19$!pKDybrQK@CD%QwHjrWmtbLyt6oa=?GR`I!<RKT-xzZ!3C}
zd#KJnBT?N*>tdKj?comR)3D~6%=J)*d2jv~1rypu2rtwphwO`AE35&(OkY27+o?c>
z8+0#^_sy@!L&!?rY-=riiLo8TRPlW>C@7sPN%VXZpyhUKVyvL25S4=+xcg-O`3@Yn
zS7<UKOZPmXh(R>I@}J8Qq6z*0E~uOp${3MA!#2<q7Vv}UyeRJF34ov4>4Sb!sK}*0
zA1tzW<*<+NH=!?rW$}v+)iwwzcUBvn;v23pbN){7p`u}ZNpsQ5i0CAZ%6jB%H2&y@
z{-V+K$aR$}4ar9c<DzKq?Ay|tcjp&x#>J8R*9q<*|NOXF#GbCwf}vyfAQklDUo}Rq
zFXTc;(J4y6gnkB71(Ez!uR}_Ffb|Nh?@T?|{0Nv~Q+FUgaCz1YArZPjWh{6pQ55v+
zJ)#Z!rRz9Bw{Bq+zL=hY9Qn%-Qn?FBhwhTht$uoWc%1L>ro@|$8jvfL?-1PIAY_HV
zQ~hyil=h$A2k;sFls8xYRXvKa2WFhyc>*%gvVgb>A!ktYHZW%TAxw`?Kt=*e?}IZ5
zxC&qm#ih|ZUO&EI2-U|&V9bvkHSj}R$9cmqHaF`}l@ufZLt~@!wmnc?t|qUt=L+M-
zy})LuBGmQ+Ie=IjRMwRI>AHgy<8k-WPXNcsdFJ^@!cb0M&(*iUT`?CNu)7A3TL?u%
zpq%Fk{Rn>Y0c0IP+VKT2gwT41&SG#5{(F}ol0$+u1{XJ8ze&u0ou<}C%edoE*oN_`
zqso2HtxuYOt4}wb&5$(GL&m?h<e7QNpIO`QQ#cA#W9zsdXAHVcZwbDsHwn>9|Hp=e
zj`9S=f%*SA1W;>{1RcOowO9r@2`XIYT{IF?A>4{Lx<qvYG*d*@wTRx3B;C6Lo!9dY
zR;KXXqB8e(ls$3~Tbw+`*|YH4=N|HGs?NzEJ@iO35Iwslqy#-A=;8Wb8Yu7`K|c?O
z;u22<J*84scTFvQAXNVv2wsB>!73rq*<9QksLP;1Dd@|ZkPXVE*AMK#>%N1`72x4O
zCDb#^V2dY%nW77u00clC?0LH4vifx+>BU_nh%v=q9toY5Y!mN{{(}#_LR%R9DD8LW
zZ1j@L>o6a=FNDs>S>6Um{xVg4bEP<5Uy$gL_t8IUWRB6}`S2X;YqNP=>H0?%2}AR?
zf-|LTYDdd*qoZXsc>yS{e;{xSI#TK3GSrEk-d8pOJXa3nR|4s6b+B+YTpsWlnea0;
z%@sYc-2ak)dQ+$X*>EYXp)+RC-0+I_NFnMgr;)7Olu!~5|54_8i9thhmf-L+-w>4$
zRf#qA=X>`bZ|nX08i@J|d~hh*JixtrG9wtQg>(Fld_I^EJ)UYy$Mf*_eu2>?;x9Y(
zQE2zaL`-PU4Xp1KrE}H!Ws#?%ldkJMlF(~74}0|B3LI`2iGS&dSi+p&aUP2w0KJAG
zG(Ci?<yEtYafn&^v3W4x?A-4Kr4P6q2r@(eqyZ`_sI+}aub<@SKenYxCwDcUOrj%*
zP^KD(!_ywwfui(zc$z1O4Yq*Nga%B)P+|`7f9b0kKqLszxe}5as|GW4=4A^1(H_tn
zavuI%TIX)gx-|ITVEM5q7^NRNS)CDfK><|FKzUR@Z+a@x1&RiNlOQh-dSm{todI0n
zL>m1zX=d8-^%*JPuOy+9JA~!4B-l>*I6N39po{{%gPslU=m2@s0jSe2AQVI1Y}N^g
zbp$SnF~cy8`T>lV0SFO5Rud92JmCtVa|sAx)syKU@kIY(D406~fl#2d8blYsX9wNQ
zlkOb_b}v`@0Dygf>KscytAEhTf`G68E$`w&AX&yC5@XZTaq#$(#!(i}vi=PqLC5nu
z)a?@x`mC*A09Q^T8t?)Cai&2MRS5bn2(|@dP3(cDSZZTy_Frn3t}Km!B&RbT1H#06
zNH%Mx;4=M7!PyIs%wVP;5Fwg9^zqyV?HnK@2hB79u;{OGy&F^z8zksNehY<6G?ZQi
zF^J)S5tMKN`rC)Lc%&Ei$|%E=^ghy1c`aD!|Ll3-6iBegfL}!Yt32gZ5fC1967bWY
z)Hh#Z<H_akL@0r5p#^MLeK)@pt8C<}+$rCmYyIZ|g^Aq-N^}E13pzhQFao9&z+eEV
zI}0&)0Tc@E1bi+QIXRI3OOL((1nhF~L9vHIiYr`sRcmTL(q7vUPn%$Vyb&+bZ;-O@
z4D0E9V58ETU9C@Y9eKa=`fGkc{=>AY8}9vAyU5+YEYqIeHoFE6UUAvXy`EeCswMYF
z1@SU}87eE4FBMereARN~cQ(1T=6fDE`JjG<o)(AP7wBmr)bBq7m4bpsdQc2DWa~TV
z1ZDx^9}+PqixR-6rG!$BZCwEEVMODk>1Uk?6RUsV_1WQ&yJU6#8;h2Qaz>nrg#LKl
zIj59zYgvAvy+G=W-&G5jvvIavwyxqiPooJKABkfu`HGHOU(vs4Knl-smUukn-GXXP
z%CCOVu?8_EGHfVV)?@jc5A_1x57=2pc@Ii9%fi;DMJm&2d4NtqIt$t>)(w02>m>io
zp;n+Fm=8>@It}719sx58P5OOlAes!RSc3P22B8AYptSMi+VyIZnDHgCJD{071;Qie
zft&U&-DCRPrJ3Ib`L&Hri$>nq+sDg-<O(D_RR)D>myCGHDwZ;Mq0_$kErwHV&9oX~
zvKh5De-L>&xvCRw$nAca)jzsg2GD0)gA#3B&ayY-YEBcJ5ACgh7RU?t8U;sN5hbmw
ztb<;<A6g+@>A2`G@S5YMjwX+mX}!DUmMS~(MvmuX%=s)frC;{hivBzs1;oAUf5KhK
zC4U*b7xi8fuSVOL=1@S}LNkc3M1O_Q?gURA9|pX4a14P10x;ENegi{#%BQn3dDaV_
zEAS7_4HJ#81Dh*&u>mN>f7>WR#t>p^Z>(f@43}QmDp;*>6p79z9=)Ivx}f-4{P*JB
zp9;Ig^JhT&AmdhQ$q<`520sc)5L?dgL~omqMiJQC2W$eaPTxL4O>URwnJbn#i)x5$
z;xo6}L}ffq#gB#;$ECyuJzt5{tBV=}9e!db^Uv)>tZ4Olh=w5)y@A{{Q{bnC$(^sf
zzvuT>Kw|j)w>z#`!;U!ch}4Qp^q`%7TpNnUKLi`+Kh{wP*cC|T>zgW2qLKgJ%9G{g
z{Zs;P2>Bdf<0kfjEuJoYGW}(eX*2Y+yNz$m7wRa7ri2Cs#?;a*tUtX(H=b1qlKa3i
z+e6tLs_nlMZSmu!lpE*>z+PxnOP#W;!L+YA;9C6Jhou+#w_$DhImZrD$I0U&dkD&~
ziIx!V*J6s((-TvuY-5<)%O}W4<aUVcthw0a$><$lNbnWp+_O-qWFbs@slC)n#v1mr
zrR?ht4nl{)W`V3?J`k(qP1gDt#J*rTTD%?oR{cEFDrgXGKm@S?cyu$!Z2iYhK}kbU
z7KQ$h0ZPSB0;iJ2uKM{+%E?&~!GF;GAF#x^5G~+IA7K3Y+g}*ilPuXIeI&w!o?f$O
zJ)Z%uI5KeV=JtZ%7h1;CS<d@1%`n+tmPadrb}n|H?l(sVL2Am|Ap=g_nFP&Hgo{DX
zIcE#7IA5;|1L^$}dfeA<kN?TdsT~I`0Iz@B=xuftcO+tY3aT#TO^wt0k$gi>dVV96
zV2usjKy@@4EMhk)PP{qjg|SGm%S3x9{Eju$dArNIQBXYj<|)cN>9Y(mFQsTm<9TxT
zO2CQB9X;TUftL%Z`#_#6eM4$4`G7+*>F)Oie&7%=Lwb->;LO=nKw!a1kOe)ir4KjR
zIyP}+_zQIuc|5ZKJY@xW>YmYGhEh}0(eUbLNSD(^ImcKwc=!;TeFe$X+f`jKEauje
zgip4hJq-M0X&m@{V9(Z~;+aS?P3HD*$$|n!QMBz{&^_k3iI((LN#b&=iuqiec`UvS
zpI8*8Z@JtD=l}@~onEuBPcI#d;b*6+4uLisN;X$I^HoFv>F+#;TTYTsxe5I2BBo46
zEyu7e>fmyepoIM>!je!#&xR<zSHIphvkX<to&+eJI(MrOH&n<6O2R3CN+mCakNsEN
zB0&bL9=dj}gI=M$=YSfHEa#%Jvt=Oh9IyX5m*k;uW4Bo~G5jC$@_se~!M%C&B_B}X
zaH)^Kg8ALlCscv4(_%$H2Zrs|yVnM1zp>rg&wFIzawCdKFu@>Wiyzh(q9kAX^5(kw
zGs`vU87Th?fHyj3E)i!2lz>k@ZVBZ=;DHrYxHAu|zJR=>FO)U<4p~`nskD9qeo63D
zM@3bdbu6Isbd+DSM?TdXdYNJsma14XcICOimmeq=uGpEVL&rXEN~t3~v)iim+K+y$
zt80p(nbV5!=I}#im%p->*x2R_9Qn;N3YY1}oZ2W>(GSnF=U)kYW<l7kO)@>;EJ$}C
zfI~vZg8=*kW$4f#2!9U}MTDCCs9f&%C!A)bP%J=l8Z3|UT79C_3foZXSB>u126^%c
zd3!$&ES|w|6j91jJM`~N-yz&@24Q$U>Z@AG(B;>vLRD}3qU%!2T5i4MRpXuilYdI|
zj+|v~Won$28|Pn!7kxwZUSKO4L42;ypw{x?L5Ktho$SE(2UVCqKmr9wQ{+Adk`r1E
zz!|uC#j~Xm!+JTfV>7}nvl^W`dG%2TVpG4-$r1Sl_g39hs50l)m0_oor_%EPPybvX
z^IwJp4auV%S$X+F_2J>PJmO_sMM`q2ZjcG3;_l6<F3^UY{c$VMF6xB*Mo_9q-!<S>
zY%0F1ll}v^i+~Lc&^!n{fn6^T1=%R5!S>1Y1JWDY>ZR0UPCr!_%Z~`|;Xqs<W$70I
zvckwobgJtmjm9LI;v4FRJ@p^=c_7+;e=sWfcHmQhlZS@oP0)MNAn$Z__N{)IT)5oN
zUG*3JAnOR>GoXA0@QvvzX3%#oPXPS1*Cuo)5&{jXfy<1~ofWa8PlaEFt-@}4hD#FS
z)Csw(C#FvTQBn=Wox44!NV2c@ZZ2l0$b|^!U$dulScatsXvLd+UDwFi)5>#~(=H4H
zrKN!x7KCMB=?tEXb%FdS(6U89Xc|D*ARj~xN@gZNp7*~;w%Sa<X&-9TB0%4b|Bx*k
zx5_!JOOUOmg4RnSQ*hU{VN>&(ytw#Bg?hfG=Qp<8hwi1%4i#K%PWA+y@(NaZe{K0j
zq?7fRw|pl0hc(X4uAikq&8VTa3y=%{PoWlgKJ>8f-!Nz_e-a=cZ`ug-S>B8L_thi3
zcDY)-{v@H~9zzd8S=ddaidgnb48OE7>$HwXAs&c({$)7-#ylXe#v-6Kf9ZJv#FS-y
zG({+gk-t0DY`yLDvA$K8<MW@)#H|1mYA*;>E<j0PC?N|;D?mE}#3rG@^@o5vaCAh2
zfRb)NAda}v)}hU#>%Y3wlXy0EpJUqwPrx4;B0nYdmoAVlNemTK@)?(-cT%-IWbka}
z@@~rE+?$paXK_|4qW3OSdSd6&K!X_2CgHyT91FeP-{^nDyU+<upAKj(LL@t=hdS_%
zJ4@H-S*^qk@Y=Xn6Y{DSx#BWbw;!2MUxMv&Ob2kRA4@3$h66pTqywNGdQc20eL)l;
zE^R250oV#?&J5*hfe`t1D9f!`3c2|i6?eh2LD(t;rW2EsfGZ8Y6f#PnsssXA|BVIO
zpm30&CW~Bw2-RRDzvz~WWxzP<M|_!1GKl7$VL(^`)hB`K7U)^x)ac<Hw5`M0AwK=T
z+#Ivf4B`{yXi?3NE^(2ow%T7bgL{G0GmzS%zXM1`3)&D~pjH7S^`YvZ#sj{YV0*68
z{x&fPaqru=k6}OKK8FAP+86AZS(5Os`}^M^29|w{`}S$=n=aYF3g#sLkL#i`<Nw2@
z=1h6*Oez20d`v_1Fnqkv{O<qtCKLNI1^>q-%ysxZgA;QwxPnCO%$OLU>mUnb!GQ_r
zN1tGxeHZE1EW>|)`uFGm)9-n2T}?DTA{jDS85X*r`yC90hBkQpKeznPjhi26KCA-M
za^l~#>}UM%tT2FSNyseewtM$)UiL%t^5=cBde;BEE*`s9@PEC8g8y%acFcwH^nZU5
z|MOhm<-urukDmU2bsY(Y?f?IFG5)fhW;V@*ru?2(oha^qy^5a<-5me_U5C3L`2FX-
zLT}@LbNMmcFOmCy_|mO<5Aed|xfu2x*tehQ09deKL4)_UkAaI@QFOo95d||l&x_jj
zUgp6M#1oV<kD{{ZD|;UUv)2B_#BByGyMqf1zpel`@nE*)Jg+^n{zfqgNeX;9Y%5NJ
z?`9Mqj{mZ6o8dwVVbk{q^!YKnmCRqjLPkH_yE{~WMcyXJW)dl{bbe`tOwK9&ZNEZB
za*e7j1)I0UN<3z{L%waY!7^oxm<-l7PB%vdTnJNZ@g|34I8yF!F~l%6kdXp&aGt&W
zPe<;Typz~;G2`T3dTZ8z>_gj^I(ha^i~k~CQ`0tIia({1$|G#>Z9dH1oI@zmmX+_5
znDNrX_V6%H0U2GI&{=#DkEYT{6YMv^e2IeH*8%>Ia&`n+;|K}fzT<>l5*9X}Xhgtx
z-`t-P%lggVMiMYS=BdWW6~Ot4RRLDj(Zc;;kj$9S#&`Oy*Nmy|%R*emT@)Eva~1y6
zqGR}Jm6CUw=t`6wnduej?ZG4k7S)CTZm-`D?l$E&m3BB@h>ynxIKt6X8EkQL-shjD
z2VhekgfsF&$*Rd`I%1iX0xskb3rjX{QM;2bJe)Btt_W1ttnKKMn(UfRvJ#8(h)`{X
z3&7ZBtJZai))Q(sPwXnL@le13NO%8i?VCAkpLR_m$=Xgq=t0}2C;5QdFl^)9G}~pz
z4ScERc!s|mm19f)CZV6&PJ;hhV0i0ipDXu;{bAGT%EQM9o(ini*H{I-Sw~#AS7U_4
zT8?MllDRR`5w9x&%dka~;F#xBY;mhp#=gtkAA~3ux5rFgg(|8=d4`Lwe7H7~r4vCC
zf5qT`I$h6Bcee5NaFXYUX{u?iP{Qid36I4#HKDd?9ioljn`<?X+ay5~#AKy7uT9VK
zKqq!Yeg1_4!UV2@uyZ#+bfPhZk>ywfQA;50vSbJHFx#wW*3*uli$Ow(541v!0)DA7
zr=;6<-nWU|ytH*OUN%!>;Lk3_KDkX<NX3CQVpxG3<>?vUFoKv?%6Y~i^!`EGP415u
zN@wD&DhRv0;1LAT)Oypx;RIm;2hUN)PlJ=;jECOYX=(&6*27f1W2M(4N?qQ!<|T8$
z)Y{|{TC9Xxcth_r|M|EV0=?|wK32@5H8yOajWWgZf$5XeiO4t6rEfFuyC7?r90~n<
z(>5yYWMu7kuxH`{e)IG^{TtUZ*9n+M)Cn5-itTLTOvNMmKq!_Wr`0jzujb^yt!ZRr
zX+<D=v1GdbB(hctN%HgDdndBZ5s(UQ_3USZ#0EoPh;86ue&L}C!ZI0|-oJI+a#**y
zwavFnSYE@RzYdNWEB!J4sp77s$xAugi$y<*?0ywe=lO~Ua8=~16q}OU-}WMJS5Pg-
zOFtVq;R{Oy!sA>O+SNQG9BE{$QEKr~&bNw8?79ce#NJFmCy#0Gm7W+SYz5rOoKK3g
zXV+mhZWye_c@K7-I_`>ybhEu1_vQ_=JXe#V{>28}%igVuA4IaR6?clU>k2VGXjkiU
zu|2!_`4L#}Y?fnVvl3Kxsr`A+HBv9+YVatp3Cxc_eRFa4Ar)885w9-+YfyhOlO}h1
zAAdNz=Lwjfp^sRwpOf1_TWu^ObL{VlSfiP-+?+JgaWba~95eZ(Z3Q9vT_9BH81seq
z4`#~77+MSBl`fx`o@9>`i$2CK_AvZH!oyI$5drC@@6i2fypJu6F~fu&>Kg07GW{Nm
z>5w#z@OU%o^2HIg^dpq-9(i-Nj_FRCVg_E*v7VtdTOd|K+G4)yU01d$<KmWn{V74k
zQP@R=HV3Y2of++%f__S^N8315)5H$bbjCOhKPoshKa4e16@7oe@8anxgu+8+S@Q<}
z24GlruTon`@MXnh76*aS$xg|b0|E|y?+>!nbJ|T=+i`vTabA2^|C%3)Y;%Tm;Z-{<
zo7OFe6LRqD4W2s4>9;TEXsFni&s(BgUZIm5hczRAyTKd!so=&_mZ{T~i^YPNnbY(+
z-XbIGaq$cLT7YUw*H1Y(=22j@@BrELu4QP>W6gsa8DRb2xx2pl0WL6gCWZ8?jj<-)
z)iyjRPGC&dE2L4hReCa1Ae^X{pe9;=i<wK{yL!~pJv0SXG$mYKV%y}W5U;GL9mf@&
zeo;H-fTF-b$?yZB%`vfzE&9=_?`EglB?gk&HltcyW7u`r&8|fO9l`3h2`dMoY#xPo
zYWlna#Cur_^S`*(dWaKjU80I!A7vF;B!YDr9tvwC4q5be%?M3a-h9F}Wt=9a+kWhT
z0*m5BZ%&E4j2`Fp(hmacjG^e2gNgI&8>FBv+^#Lz3Rf!)nZA1+yL#Xc6@I9f#eL8g
zEeZ?D<twy7Q?O@?*8}^h6Q2{un=6akb;UY3)I<gZTKGt^*4}KP0*YqMTp^q|E+=Qi
z(a=5I1re*EDza(d>5=!|TYgL}O!d{yW_e<Xw}SO~<<IZ+uToo<n}c}r^%T9jSV~ak
zEhoiCVfn3F(+n}~w20gEEo@^Yug1i_uK^t>EKnLj)RD(;O*XJp?;5FI7G3cvd6yOJ
z<PpvedY&@1FjiB^L-i_%5I5Vu4BVgFUV5y*_{(5gJTUr~0YTaty<)rI2%o^!QY`0B
zP$y8}tf}vh#)N?%!Cp15?@ev!U$2?|?A)q;LB$VMeD!M9m7%wT91>#fhmSoBZV^xv
zDHvlvr>9^ob*M>M>`7bp`v{Ntj)idZ@Na-B$7bB9n_%Y5A_Sfyj(d1`=pacoeT|H)
zll;wcD_A=&t%aPsqQiOoh8n`1iSyQ=$AcyHmVRo>n`O~v14fQ7T+1u1m$k>LpZw+t
zHN2JBwoylIsg+gG3=yhE9ly7@JQ;WGX!m?Jt|BOR!XIWUduzlsE_*udLbIB-6I<wH
zBjxwotzx|kZ=_C?ic;9kbdNbLJN<+c8nzR}9&B~QAjeeM9mm(R75C(MJVVIQEcU`&
z!uEnsI*$nJ#%6ilRk-5|-@RY_*~M~ugST**gX7J`Mn~132@OP{`6GfQ<in-{$YHsm
zm3n^EOE=uxU4ezT#~<XaUN>Ff`@!l-l4}Uzpm3-n#=^;PWG<VK#7Z?TA@n#h_Py;y
zIU6=t(G$&ztu{5@w4JFfTzMqNREhsB2XENf&nx%!?V>>PpfSH{L}ELU%gs<xpMQ{#
zHdMm69IA>7e0KTTqO${EGA#Rj5Vzeubdh-Un2ckYQk<v2Jv_F!J!|Yei+G;+@<uCn
z=q!E(+9+$$ssuNfL<CnZYck7O60(fe?CfS(Xu5-Vp%W5?+L2I0nA^2}(r%B<EV&g0
z-p#nuatOPbL!4v?o5kD~CKzW0$0R4q7=b&T-qCjrPJdce_Ryp0R`9mm&VZ7tm&2`4
zPXS{CNnIAZW)7qkMU~5nA*4Te5G64f!jYFCkOIs8%g|(X-zR&BoR_%IXf<sNxL-Le
zEvpH%_c5+j7d6*t+$gI?W_G}{bC0_ymuk6X#)T5pAf67Z$f9MThDW^kq(&Kb5CXKS
z)sO3Q^TU4MY!pReH*U$neuy~A+^=-<+i1Pt!W~<fyc;3>b_MKOdqV%Fg0g6fB^$p0
zzpV8SMWO=SN;EVdaT)O|7KSkPh@{b)tBF-?8I7KqA&Gt8?TqlSrg93Z5<i+<EXzE~
z@g+?%i}#f-co9F2rI2>Dqg9EzuEUu$o;|PL2VPJ6%3se(8XqLydJw$UKk3l8zj9L8
zu+cS+aS``DS;kfK^Du8id1l^-=TNNAcZ{@I^Q3zv4)3Uyp2JEUYvAwnM#G#pr!o{8
zjrn4Y!{xmWb1JBYHMKIlO<u4P9V$Ee0FR4p4<qAFpb!2~Yhc{>Nz+kIeDnAZw6cPq
z*8`T>(_log%uMB(LuG<#Ic<mY3V6)%-gfuSOtq~a;T~y3$|7DF^T`UOo7?GEwq0Nn
zA=UKuXG)(X8@o^0j`&x8UG0_W$6Ai5s=0Ps#i8$nu0LC7$9x<0<74+@>J<B9+3By$
z$F3WMol$NNAFMX!j!hnP|1REmqKw1=zjt4Au+hDTd>9Wrg{$(lI=C<4EZ5Q5F=@I#
zpEjuqc$UgBJ-8Ns=$!)0ThsE;$IRGm1&LAeauLss<aVoA`?uJcJ*&f(ffli6@mZ9Z
zWFpu(ngdmBIl<T=>mOF4aY8ai0?fzK_H$4=2yR<f(tu{!*%!jDszwey?ebm%i%~5>
z@IJZI6a%J+ArZqdc#r^d@6$#k^zqs5;J#mrPcN3DxH8V_f~Ro=uQ0Y<mhm8ME8$(q
zHqN}Q+6`u)e~{&Du0Z7rt_UP7AhSD9RV8vxnc8~v!MpNsN!QjRLM!`pl{88Y%lIHO
z+6A9Hh+sXI(9BE=*JrBvp-nzg{Ultpe9vswmMt=W!`3>I?N(@!#Fm82o$wzGN_E8K
zJ`B~420ouv^)CD`g9!&gcu40+FpEz$yN*C|a@A<1n{DX#>VQ->xAu{|-qCnDTY|0q
z!Gy^Nx=PJp_Hp{sO?wK&n<Ed68`!O3+HIX`Ovq57^2(VKT{6>h<-75HRqlLoj{=?{
zNsfeXBp>yK=gVTT<*;{^?dJNOVv|`0CriQwu}MQ_4nrwyB2@=ZTxm3JXZg<MaNKy#
zETorKdGUT>26|hf4VJ!DD?=M~Yfm%Y-ztA3daw$?wOv@3{5uN4@SgdLxak{yI0prX
z`?`u0^SCUOwNd^d;1bvDt(9@4mf;V3q-Z&k+;o2*@E2E#t2Ufv@u4e>V8+8U?ju!?
zmEX~Kwk0Sq9k3jVCn{bj%^TDa)yT=stU9k~hXu8x;^kR1vS(J)ZI6%QaY{5<inT$i
zmc~y)e+3*f(pTA5UaG2%eVeGqOX$I;F?PK&kR;fVPv}>hX?saqtMbY!_J1JB?3Abz
zDFLvv<X!E(zAQ2VDZ78L3Xio9(m0_(U8!?qbCn#IK}EXQvhx0rk{A7||1gwCg3Jud
z9=m){vCusehI@t#YjDmjgq_XR>=s@uqcxH{y!iX-i{CP^npL@<`O+UowHzDywP`z_
zcm<Wv(=C~qkHTD-eq($zRxTk$(tTTQIOYZRi%xJ)+_&U!B=1E$t=IxM^X4`j#M>>%
zP2K&DOdEV-oI-c3Wv;W1E&Hx6&RSlE+-#d852ChhRdpm*B3h-Q08Ch?LA^Kg`G--{
zZppjt29c1j67pO<V}yoX4&Z-}r}MJ8az4Mp73?mKsiol4RO_!yRyG6-{2}t#!+DfD
z+{Rcje;IDuW4YESZDB~_G)L7lk9wIysAkjw(P*+^VZ&br`0?9B^9dZY_ZaP~dZL)?
z@b=kHd#Pq}@0uuxl!F{Y3WG*-Ur#go8oLH!v?eWsW!}gc=!XdKOAW>l{Ukl2WLa!m
z2YIQSTyf8BK8fuQ-YGwPThMD870@q{#f0?PK25G+kPbzjq>(oo9SvvZ%*k#*IEhFy
ze)ubX3!UpvvDUA?|JjMROb`w8SgEqskHx#kECcHzkUhPRl{KwGO&>SNOTiNbvJaPO
z1&5GebE@$b<m%*PScwBS&5ayncC8P2#wpWN{na1DyX4}U7$zhVR{Sz1T;BXv@*sa-
z^nUC|OsSZcKT*X$6Q^B%b0o7!XssI0v=;0pZ&(=~?^{tM5nvpH9QxdWj2gVyzR%jA
z4`#N1w3k-&{_X19Se(G?fG4lF^gH)Ot0uGTo@<Z4-S=w~4}<UL5aB#S>n<<t-3&PW
z%5CXgLs71|K0`Op_9X0a_vzA~tji}UW;h#ew<lBwHKvQ+eZw%b9FcI5FqYl#j%%;A
zqxUM)D$W0(;<Aj5_ix!^le#sYnOeyBHihS+=TxGcHq2F@BT_GHUs<9aYtT7d_QQck
zL5;-9mmKQoN_I|8sHgpa?`JG%7`!-fUm>8+XAZqlObR?TnxvX4c>KC!N4w0fg(05F
zRG?Q!?PnOZ(S0EODY6VWufVBOC*3Vjkmb}bua!TtmZ?l}$bGxOX?7e5choC7GNBlN
z?1m|rRpH0vb&ou-qmeH<{dS_|C7w7lR%y#pJf&ASyi@vT?KTJqNYoYXPBZWK>_~Mp
z+ty<|R+D++Lzchzm~uuqOepJO;c4w+{DL~_#s_p{effq*s_WXr8W$A5I|KVidb2y-
z5F(FI_C1a*Zn~+_7tYei@rsF^5!>D)87*sjo5g$op0eB*1>D=@7k%YpWR4XWL38e4
z%}8`ItvsV$j#6cxahyYf{2YFBU!IcH79Lw@s{_Th0Y`Qi%qN8}ZD+mR_a_0I`I_Br
z5qlEwK>nldRr-W2eF7Vb<BhYnjU)?xTdnWG#X=ZiM_QYJFP!B$z8%@S6moFTaQ41f
zP+nT5otREs9<t7tYS~uINc2Mxt#&E40$f(vtzwDx8}%_L*eoh)6C1gcT_Jy#HoRS#
z_|&03zc7JBb~azG4m86J8|kw*@~E5Rk`tPRqW8I1EzF@Q7DE+*0xUP3l*=Q&r7{2U
z`g1LzO(*xP!>CiWKEW(nFnOagbI44MN5~l5XX|2SB-|`K{6}k4br4(~yB3#84&hW2
znKW8kqq1R(JF4dQhjBDNfD3YK8<=H<=Zc`%bwYVAcbKw<EyqsrJnn5h1<)YO;O#=&
zLh{u&uYN|MVy6O5+&$I5aRP*R(>8JaR1kkH+Ipv(1w2^-XLrjzjI?n4&E6sZT25{4
zOzGK*ZaEtO0D)-#6y#CW!CSVQa?(1M!<+&Ib@4~Y#t{jH+O0@-JONGS{lj*z?SZR}
zpc$UBc$mCvhHVdmb00e)@2h;T24E92zh?^Z00*!Q6225JFYp?{(5&rNN_$AaVcYU|
zU;zVrpPG1bG#Ay4<5RdIoD$I6qE?D6mfXJ`VmN()Z=C5T8OOvA`z}CnZUqPZfpZ8S
zM`l@TX^sY4<FLH>@@h1}u2q=7FMjmG%4yaW^9dWiEQaQGRc9w_S@BX@$jAvp{?j6$
zh@slRN{55o&^yG$$1d;#lMjm+PWAJJ5R+V59sr4s5_Z|C_^S4bHw&0LioCEkHgA-h
z_4AgDBP-v}Qhc&t$cwNh{yp@zY7m(#uUrb5;VT&lCuZDHyAQWUq}udT3216J>g6J?
zB2bKM^UB>|-t`V%8b0u~Q-1u%MnJ#jysPAVnqT=wA;3+LW%aB>wbGV4V<=9ATwYZH
z78X?ApobMwE3A?7!hG4@1zR`^b}z2N1At_6m@&=M*Kgb2C1}A-t`%UP*fwE_LM2Z7
zJObG@+)4I<FSc)k*EGWw_hoLoT|Xa|U7k4^5>mPBJZ+l`Zc@*PDoQnP#Af(188Dme
z3*u&C2pNeE6>EKXSigd>{o@<S_Zi;4nXEH<2!->Z$nq_!qe&di)b|l2WNJTk5`;R%
z%IQr29Rf~GmlaqT#Sm!KJBR6f`v5x5SVlRbLog-ps<_(sVRI}l74M3(U80cQ2x%r7
zn0?1r@2mE86y!afENSG-!p|MbX1i&=)LgAN=8TGOJe)l2iv+mIxv=pO)N`vR=&FmP
zMl$l@C(Z0%FN?EL#jHf0Kq1w4`N#O#g+xc$8Q_v_J2`6=XNdwGhUEbK8OgM8W*qTM
zIh@BxwJ`i`|2s{t$IXO{{IOGDE!>vXtEw|*=IkYZ=0kAhj<_+)3=g$Q8La=a{ae>^
z3vr2d<NI?bY>nLuO~(!>z%-xh^kp>k4Vk%JNz<X}M$je+ZnB6m#JC2m@&)aEmx&T6
z;9<k0<xqIE?Ps8UvS@2Wwvi&5+l@1_1`vN(s|tVQBid{03mqub<^v1;W7VdaIc!d5
z);;<#`C#7=9;I&J$^8D9$4X<=!#Ue;Pc`TNGPEv*KQ}CraBGlJ_=>YK#ga`3{l!5n
zw#6enR-%pjyS?NEMBgT8lM$27VkbVcF-eRN>>d)FFMhY(iY;Er%pvoLAP-Z@ORWus
z49NL~_X5%4LtcILbqXOveit3oF~GgRjD8+?$b%y(IH6*^q#L-INz$?c@f`+aK|I)4
z8@MN)6lb2V;t2v=kup0&LY8uQvyuARZVB4PT$rrfv$@zb9TAvF2OlkG>;$~0{|_$F
zlJ+O$iIX0mddxR}+>icw3aH)yAmuU0WULJovts7&pedCTv_Wk%S}&WU%pDS|h_7kA
zIJ!tym8dM2%HKGYEVc60Y9UxB;7JLKG|e8*RR!-~`xJg$+5U7*u?d)j3iz@caDhI1
z&=%XHTE02r1{aN6?hIMBkKl$I*2AdSHDF)sjjw6~z?Pk%!Lor=7Iv8JhSXqOT9sam
z`BF$l8QY>w`>?SqT1!}uRQB+0Y<v29(QI^kh2ZP+{^wd%*9e%fjY)XJ_NB~lAzN6u
z(y&~Qy+BGDh@t}w0}4;)aTteQ`KV}5Ca`n)$vk4Bh*UnfdYdauEJ(<)fiGE>YbxFC
z-tk3h!<%3`F7FoHz>=Zu%ymCOvj7=iLG(6JbGbL1d~!l(yAj3Cm+?|7ve0<z-TFuh
z|JcF546vu8$Mi_#x5L-f$cwnzSu9^Md4PWyO)X(&5ZyAv+p(tn@or5dGl}MFGYLKk
zD6XPCB;``Nu4|VgrAAC^n3(cr-|ghP8E=2#(inKW7|L$1ms#ne&FSoK8Fp8^&TUPi
zOP0%PnMTHkQ5yL!Jc0#?qYLGl$cHP)Yh#qm2{wZHCMs@Di$(6VppFQsnIe)KIBaUe
za^P@dL^!Lq?xg8tOJb*{>4k4%1fl}^FT+=<L4K{UvOm|Bt7CNjpz`E;#_)x5Y!S=d
zRc<n%@G1r-vs~Ry6DppI^nO=7Za`|fHxpb{&RJmA?4{TwT4|QM=hyn$qb6RRbm78S
zz08dwbJh3gDz|=kW1ybsr0&9UR$g)3^?MQ%t23=yt`do|=TnI-D9eV(fP|gZcvL;_
zQtmkxJjFqva{l^(pD8Uz3P!OaE@2^|hD$64=;Po>t!&|y=BQI%jn}JZ2iO~kQK%nR
zb%j`oR=RuwM3C|j;E*wPzV%g!?=M5Bfqk@=(-z~5G{E%)fm_vBPQe14PvyayW9;cW
z@Bory6W@}c#xRy4GlVu>^_~^G5uMcS#l$SnDn#pp4tEfU!Tmw*MVbDGO};{5r?#r~
z>Jq_(PBR%_r-Y#xQeFOUeP{3UL19c{2L+bMJlF-kbYHl?<#%;V&pexD$6I}B)2abN
zeEi5e=|6g1H5DAxL;76hnN_$q!MqHaKZNYA-4{D-3c$$I*%qUrU1!bga+Ng{^j#mo
z*l)X^G`AZTwLY86E4?H)q(f@5@#f0h!eQoJ*yi|(s}^l|XX0~`)1;nXh{FFC(6A|~
zXSM*-VpF}Er`jMUM5bKE5U)vIcFc-Fkyz3qx-!#CUoBO3qDgXwhp!-OnkP*Iuywv`
zgtpaSRQFr!$}dQwQg0h9dz$;HtjYn_Fm@ynE!2|S+a|%u!z^SgS0v!G#J%v)RKoR~
zT|`!1MF4a9?$s^idLKaS!=RaBU6A<1>9zOWlN0-b7A_X0&o8qb++{hIGsz^pWbJ&+
z2sPv<SfYj0NEPJo#>F_#(jKt#DCjEi563-4z6pnXlJTqIep&dib&%f5(qVx+M30zu
zO1hb_WKRkyPX+e7l4I3uzg-INZBz7s%~@NKN^o{^Tb>t@AzNk>ZFXieklDTj&fm}m
zPR-@Lv0eMb!Tmh!NRC~7e%06x-GlN<eXQ{v&J+Af(eaLXhj7n+=6M!MK6KWVD`nM?
zgZt0^9@G6K5Zk%0y{}v5$VznIq)08ZW6ofGk&V|TDqz@=%-*pJ7z<fJd@mUDykanh
z2N<2T-`yJ(5J?Z9F4y^9`y|k?iC^_q4porb$JfvHr&&+#+mlYo+_N2v7xkTYU|O5D
z_n#~DJcR}F5s#yYV(yW+XQrlHFNEynY7(c*{C>?zmH(c?dwixgNzs;a6jtIJUa%By
zSMCoV!&aM2O1gFruj8?O)Y_l<=}x*Vo&pnsZ@NPLGkJCtWm!Jft=&U!6wX5H09nfx
zz_6;rDwNR>e)z3Kg6Zu?h6j$Dxv9PL$8QyY9r1XHVi`;YcX4*s=fpliuQcG%f%w3A
z)*u)GRdF_LqP%={)&jUaSlUA#w1VHUeQg>;d?exJCO#3NSDl9mJB}WNEiUcoF*o7`
zq-rd)$hV~<Z+YY;h3Bm8rBF)M#Aq{+D!IHhHZ`m5nX`Fu+fM9&uS8th{s@K8{1abT
zHYRB=WzsI}3#=sh9__waZ>Jo#+}|lINWN;eCEa)*Byw@s1^bBG0dfOv&brr?k}9p?
zTAIOk<==3vBE;1gIfcZ}SL+-N)>SfFV?6#d6*F`&O(B9YF|RIo<D!Go7(TPNQ}|+$
z-&9!TflMoJ<-*^dqwHp9Msmbb_?SfC12<E-ob(?uOUbEn{$iW_^TgS#o0jb*^RtIm
zRPnQ`XrSW$_PSImmmBePc&H$E>t#Vt)%01C)@deuQZuh`$32J4&jc{j#ruSv@TD-(
zm8NG1?qIR|FcxtgmOIz)aNNFXA8-T!I{j2^VWBN-^Tn>?y79uTaok#E;d~3C%{Pp?
zrV>W|#r-P^_?X8nXMeI{#tNo01Tv)l$A8+$+WY=CJl^LHzrDdjX@Z-12M4z^Wzrs?
z;>Wv(Mq&8z#!s528fNaTzET68vlR_yDd~FO6}N4llC?+zU1Q~G<h|pQrVa8~Qr<KW
zIu;3Q{GclAx55j@gi+hpQT4@TqluakK78M1&aK}c*8&RkivVK2mU~}RBl%?@*QR+S
z`Ylgwyd#@AUlpk#ivz~#jL#u^*?Eb|^(h)~Pi0sjOX}GmXM0y{t|nbTPCjVYfU&!)
zj3}~6^v`tUgzIqIvWNxaG!s~M9bti9w25=9BQdV_Z$Gy`yg%|R?4WAWyHdF-<;@H0
zR}D*qC%yc$n##*fc&@F^8k&)r)tF_rwy+s5l?xA!?dNXAF6Nz($B&<o)nXC*D67l#
ziqKU5i%+Px1Gum8U6{W5!rZeYa%`dQFl{%}L@r@VuQ-j71BtdjTAryesEyyA&}C5E
z@<xkq_cDis=UaRXGj=@Hr+mPAoTDsBEw2a5KmNGiSMxcNr|J-1I!u;p%`9RpPV8;M
zP9~4&{-}>!g$8G2gN}vmyR2M!clu|qvn5E>@0)e+os%k+NUVC~LQXIFKUlgFc&7jV
zuX4<l&3%+G%u()Ca%8q)t{f4$a^&cebU4clbLCDHGN$CF1Ldq_shpKcUlY|xwKQkw
z|60HQ9^c3J`z^M8KJU-#c)p&m=j;96#LgibpO=5r;Sx=K;QTyOdt24k1IFL;)s{YQ
zi->=ISA2EXn~&+PH}CsC3G7Lbly{>Y-eDh}gbMb%q}DOH&ESz*zl}uIi9t^Z9&)tg
z$6K4zwvV2d@$@MW3eVdjiHLM+Htvf&s~OmdnIV2`IWlRlBGaX-aU>b}RBXl4o+ti@
z()%#&+nq}d&l4N(3!O<9<TkeU^89kRy}`WOFLHmwz5+Fdj@}-Ficg9Q+*reo5AK{Y
z*jeP(*iUlHcCt&fHd$6b-&X1|cH79*EZB(~eP79eFvqTW!EM%BaoJsa{VCzDb1u}b
zX!H*9rCrUI4^K?T$w!L>iZpNQ5e{|OJ6|pM3qQqS?pKoYg&TYmc&`L-i~qSl@Gu{F
zx&O0h#pi~^`O8in;%YceAK`n%ch+mScL`p4oHH*zaP(K?c}_Bz7tuU6p)YOk^uV#u
zIG(AJo_TusI;0skW+Aox%=%^1fDSgz^EvxMpQ-!3*zck1hd`@v)*<(#N2he@_77l@
zv9%7deK7&L9b>m1RXI<D%a3ijpByrLt7FNu^HfIl;PZ#4?~kS&jCMJHa+&+9{Ef3B
zck&{UgM)vjnoJt@Z=J8-Tym(4+~E8IvEF#ZX(C^6t@cB}Hg1Ren>JkMxO{KTcVu_*
zs%ht8vpQc_4X2#`wAJ*WnULHc#+^fx2{&5yo=W4<a4;1w%RX~$t!l}C*+;-9!dHD4
zyBn+ND(h)Tquo1Zy2Ih*geA=((k;n8ut$3R=4awWK~*oFJ=vD$TJTzYP%f~iEM(I?
zU)P!~+;UFC?>n+jP6W2M_*eg82=x59(l+^mA+4S~JTdBRsn*i(oGgGksdezG%7c^b
z*ZL9W67ylh6E^3SZ^HXR%y-b#v=b4v*(R?&+YXJM^l^p){^$*QkF47WaqGcg#DMbE
zO+snu0;>lT^5@4@u-PY&o<zrj`0qEO_wM~sR)*c)6e$i1sZ_FBs{zq!V4uXp8>SKs
zXJ2Z|*!&7?+8Q04zGVXa+5A%7nQ6}Y`~9xb?#^cL<9CoP243-HS4yAl_p$F-t;=U!
z2`}%`xyQ4s=l;KD47Uqf^IK8O1!BX!&{d%*0bF>(sn7B%YP=<H-o;P_Mem2-vv-p0
zKfmH<{Isdy#*#Vlojt9YfE>Ohni9EFS40?rh1r#v1Lw<5+BHi&_?-HBWY>%nT!QKv
zTBL2?3G&U)S9olMAWvaO7A|WXrrz&(t#~R0a#iaS17oyiplQ=<_CvQsljXzI&t;Eq
z1fH4_%X((ps-|$+JoEXd?`Pu$wk@91V0FJ5U2727tmSbSE4ca$5$tP{^`xAfXw)be
zB=~u|y_wq9w;yi7tw*|%zg7Gg?9kn3Lx`Ra%B@89itp+APKxAHi#)=cFtf{BV(i?f
zUmrf7s(M~*HE*?33WC^FV)Z=Pbo|rioCi<J>DyoO*z{zQJ?AaUj8}Ws;)PAgcPE?7
ztS0wzm7N}MD-g)AQ!A%uJ<uR-nEuz)(5Y~Bmx{$&$IB*}zg%HE$gbZPe|=Pu6c%nS
zIpFgo`q-b@hK_(B%?Q_*8us{}0yC@oH=dm(=I`27wZYo^ScZLXBEP?`H*a(L(fbF>
zNYVwuJL9rXYUykr_+^~3u#mV^yc&9_zpi$#<W|4D2g5Qw<ilK@g+hF12hw(VLt&FL
z5P{hw7!wEwhPx<@zzFI3g&SwfED}YwWi(6_d%iQp)tqii^3UDtQ1r98`s$@Gfg0;U
z^|~TRpwYaxza(}NEK}9pZr!_7^Ln+^r{UbJ_M&}f%v7#)x?yaq;!S@MA-Mp1AI&(+
ztvYF?!({?;&IFZM9cf-+0m<cp5APOyBNcCL#}9l6Iht}+$TK{xbnxQY_+WGFgS|JV
z8=n|7o>JL68`AHf+}JgFJKO>l?<bx4T#@^XmhQIFS+(0smbBp-?Q)Mt)%UdTr7FfB
zjaT!t?B~;m+C2S}Bfgg>^k{|BB#BEYSxLHouI!%fi-65!3kk)Lz2Y{!(&=e`xg0IO
z$CeMPyIFxUW2aBV!JWj9Ruw4cctki>L~YjTgEp-R@g*)hDbEwOCec%?@(tt3R5n%%
zNA?Kh??O-$gb0?KQ$KA#|0rGZh2Ed=@^>K!KbQ2a9tXoe=&azw8qdVzpG${0J?44}
zji$K6NnD0?R;RyN`E`+X5`}k4leZe3N!&_QYy6f!efkTRQitchql-z7@%==hgCRoF
z;<7H>ygb`zJhwz{5Y2)M1T3um9;+Q3iSIcqS$aT3HA-Qg)z=aZJ(TV4-S`AaKIwU3
zjpMHM-H-X7Y$F-x4x(i~^Y<T{WWCgY#(pTk$EVkg_Z`sH?)g>NrEtGLk(77j^f|Tr
zc`izs3H1WKuO~rfHrQ&J<w|`lGON~6V4MS-@{aq$^1Ps$2wh9}q~6kJr-i6h9Z*C-
z^-DtMSKY1ul)g7>{h$o6yZ5_v@uUr34>=o|zR4=m20GrdL%V+q*Hdu|B2NPR!d>bx
zNv-4RCyN{luFh-u)e8sq?NXb(r=Y8G5Z|NLk*EJj@Eq%dcD{=Enb+fcW{*R6bn49I
zkG#UaZWySAu=U5!%^VkG9!lJxIo*2n?G2C(uLXPZqmI1lJT&7*rTNt5Cob&(A>>XZ
zh#xm?;oc0I-v7P0HlY-ex{eAUfz*jVP<Al8(KoDni1lmo|7DPy%oN{H>yq7bOss3Q
zU$x8+w%wy;^ZXuEAi+>S!KjMpR8X1MNq+E`>-q4Wa~?g&A}q2ft0;>?DN5D4Sxwq&
zBH}!_Fr%TVsqUU@qEmzwb02YZbmWdfOW}iS=<YD`HGoq@*eUXl@Q?i486ANQCbbnB
z1wWN2RBP&2qjqpf!Tf@@*C=gxLrBq-e}w2n%_v=~MTd1C%;J2$eLOx;BEN8YKxKn>
zEeuW(s##NC8Kr46u*l+5Oh2X?m8nKAx#OKOKTM_8QJvPfG&SEE!6xhqg=RL;b0Tb1
z-!ck-eZR3Nu}PYG)M@-O+*TfKxWKbek3b+`FY$0LY(t}fT5xB^X5^eEOrvx5a5o^U
zW|1+BA7;+*DPH{I?cy}`OPlq?x@C57$A~)y7Q8AwJHBK>pDuAwO-I^^SPb6gyl7gf
zPOvrj?-0(HsCyUc51DGf|7qj3s=qCEY%`H4M61<0_#sj$Z)yg*ZQk0f?0?$)cEcrO
z-APeK1~2Yr9gRYjagX}mCh;$~8aU#`9hp&L$p<8TROa0<tv{kc(~m=O24gcq5_UWv
zNK+0x9xX=O&3(LEHXysoK!bVS3-Nr2e=Xv^&&sjeov4lHj5-m5J@@1wjC0IO(B~|b
z!z1rvELRKS9k&d!O+>IQYBlg0cuaJfAv!VJ)T83z+S|T*7S0UVM;*(BRRcS~Yc6YJ
zscbcBv#M=&q2x_39+?0?f{=u?(uSKq3<A+37BM*?{~3w!l39e*Wo^>(s2X+NBz@k&
z2&<%|gx)Sa4C_I>YuU5Oy+i~XhS}0RcnGd08bp`bbYu2qb1B}vx^N7_Z~@T?Oo8)I
zmbtLVQf?)<R}#F}J*VL2E*U*b8bYD>E<6jp<NPf6OzjPKbOc$UL)Iz(EEU0&PmGaZ
zMdfB%lvwn^l0}z1;h&zDKPhu-HcCFxgS4zkYSGjLeSC@Sd@MX)J1zKR8`^Ri|8FPw
zV$L6i5Oxyz`T0ze0CS!|W(uH1&AHRuP>JQSI}Hi8>;V!?wVa!>6{o33IOWmZId2+f
zdrAG}y1E<odN$_gAH=TgjGi%0&MSglIawN`65aUGX!$2G+4GjKGW}pA06|HI2YCB7
z34tkF&F}V3)g(AH&;1#WWTKD>wgWsd@Qz~<oMq4JjNeTjB`-I=CVCOx<>=|PVP*8I
z8ft10DuaVh;7Sz{_UUjXM4!~Ww&A>X@%BA7<w0<i&V1)D*y1@GLpRQ)8<#x6!IvtH
z@Cr%53u$d)HtUEFJZ1~Ie(ukR5nMEiW!ZbHaT~LY=`E{SdE!q#{f9PM{*!a-6|1ke
znC*`FR8%-vq%~s!9;`7B!wZ2?YE@gDft?_!$l`ta;M`*1==LeYZK~DO0xGgp-c*z0
zFU$&t`xji&M)zN?&$F?c!7e(4V8|?7GiSwQpb<8;(h@2E#_=~arf<tvT#mU^?~u|!
zIJgz1+mYxc(QGU;LmuU<Scna0LDoh)Rp37h91rDZeGK8OzV;+m%=-1ehi5f$Qe*nc
zT>Pe+PxX}mTEIdqpm_ySDRBR@JnA;{>v=Q{bPhtfBy-k{C#XzD2G-|$#d1EgO|-8z
zzX}aZ4kK}FL5_8}rmtfg+#j+Sme*?8wkl3;S)8VIx6P-A#_SY+!-nr1RG1G{U~7_!
zH={yCekA|p${)4gtzeVwC2!4n$o8qOuCB2L(K6Ems*$E_b1CI%dyg#tgl*D{)giW9
zw+~N@RD6_txN+i|ur>OY1#?xu+ZcKB+%Q|f2_{Yg6#1vg3KTeaR9crcJnK}z;4DDF
z=eFT*|L*l1P5m5Q1|LD5r7^`FbB$rY0)*YEYZ{g}zs~L0aD#Xy_9m~B8$2n3j5J66
zdm-$LQd2or9=A6j)Y>4BHXD`EqIm&hZ>4EFtH3r9V3fn)QV|3{2u6rLY=nrP0W9?!
zc8{17vS@h5rgpiB;OOBJeSAQ9HiQ*yUy6Osi5@g<&o@Z2YT+=9s7MIsbvyon6K|gX
ziIpN^=7+N^`=5YE!LT76?uzrm9Dt9^kSNr1t8yjt87b~AmIx-nih97bT#keDC|KEZ
zO-=n}@T@%8^BB`3%q?~>!<5Ey$HSYjND75gtV52qEU}u-AZnnUn(}C~LWO8;Z$+`W
zODo=QFbSh<v9ODon;6>Lbu2y`Kb`k6GP$SB84N<ed(-1wMwGF2{KFZc#RHY(&2n$r
zK6?1Et$DB_PAzJS=;_R2B#skciB3%r?h4TgR6RX0qL;QUCoB~)dh^;h=h2A^WT`%g
z_4e0C1)f8wfqxFs9eLD=Zrx00Y^LA7H%hvkaI#~=wcw^t|3{>`BX^iDi4zN*>(D$N
zoN2k|%rKm9Z~8woW`ae(MPQE;z0`@`6!;!H3|{*&q|@yOYHH{?{CCtk9Q-1Ui_ZF7
zl;3>O#D-C4pf-Scw|J;30=DLpv{X`ur`}}ZHqU0$xlyPX9wiMAI=hW9?2UC&C5O2M
z`0!E6$I@$&<du3Hrl0?9hPqAFxlUf&?CkuJc|!SMHL1>sC(qHn#GxIo3Jy2h;SnX~
z=!R)DF0l*ROqp^@eF0uP5*$6-77`V5DV-b`?)bB^6@wn5%j}aF=#^oT-+Y6NHbs2#
zFSXipEKS>PxeTm{pM?$e?ai0^#O1WC-_5k=9(43D$@D{^A_R=NSy3f4)<wbF=_S_F
z&r}8AZD1&VQ@?Mv`K%pZ=ex7vk?!boAf28wlM{k}<0Y!d>va@i4@;DuPNiVQ=myaX
zaWI&QKvzOkFxy<<lvh)EPEa#<Y{~ee;3J1kpG+K_An>QMgK|u*r^%3xAg5GYmNDJB
zBNQ_6aC>J*ABk+@@yH8kU2)I45-`_F-qp0!YrCm-UBvX=*b$#&>cYW`gZa<Sa3k?~
zq?t=o^^A$^5ZcF7SB8B}-1!Bc=hzVb(VFtXI)N7@4!GfqXu0;iGo&mR-|j1I6y=N~
znpha3McAi$aa?y8955;fuT@~x_gvGVRT(9%p)rM*pf($qMlre_w*}V;FzN)BN7G4q
znf@P~Q1Ekwk_>kU|EX$ACMkVsguS{W&GQfC=cu|3TV1CEC-_366|bt1I-AwJGLmBh
zG4uci=1g4rt;W2cv3K3Hvv<cP*nh$Mn>FtHFKWhS<-!ysFH0rMhWt6twPilOt_nfH
zO>}L=F$Xj{`^eIpG=GV&jcq^1rg?V?Ba7YK8pYbv5B2@B6DvbIbYE_j+$0fJ*pkS=
zocpr{Q6+dGgRWJbs`yYE_m@j*J&7LSlW()!m_-f<3|=%DI6GeApg66X(1q`o;`T)L
z=Fx`SV@d-S9k^Ko<;<EW`J$u0<^5VsC<-}367tDao)X;=urDAOcI>rj$dLf70HipO
z^614u2f9gA{%`Zms#dC`fmpR^@a7h!Dci!4pVp9pq(^uWSS>`VF<%+;qF@>Kd^9Q?
zhci$3R=;LyS+4rfJaE7M(ON^ZoA$fY%ed`M40)5$;~(v(qS`dJ2+cVvrV&Uj+u|ti
zmfoykYmQr8RuOf^iR4ykeF}Dz|I5X5Izj2r9usDxKC>q&?M!v$b?mMEQe?ya@Hg_G
zZ{_U#Njp~WdQ^P?t?S?ZIA=*0Pt(ud+<UuUxwHG7@wI2#nWesS#N=iRM+UvP@a$a5
zu0zi@2NXYPpKxXMV+41WAVLiVVPJh#MsChGu;}Q+WzDW#XY$8;1@AvRb<{0V=g;M(
zrxN#@5kJP<azl7EUdtWYMLCjm?TP<p#b*x)pVZxvjkT?k(Xt;{C*-n&AC`{VEh@@l
ze?0yD<N7_H-JHzP&Gt#kmykeJ5nr%^NKLESxnozXG2Sg|^c=l55G_Kgz~Qifl84;g
zz0<SVn=BhWOI8Q1xd8wH(vyHR^j{W)A|6TaM@;hs9+r>n|MClx<mE!sqD48)khFxt
zVncGlj(w@AEr2aM1eW!y;Lpdt*26B;6?SP|jq}e(&iv^uKS3^NoJi{G`Y~9s@asL=
z{>gV|rI)X4{%jdfiR5?rBVzP2^}C3Y@bsfAr%Kd5$k@I!T+L5xP5WSS<%yVL?R2N#
zSSz9ProRIZ{zw2Z;6LdWtz_Z}TE%r?tg&!2;`LIb^!;meDYgA=i@IHpkJgtTo(PxY
zF<VEK-)G=<{mN)o+h@0qsN8>O@t91gTQ>9SSC6Z_HGfDa+nymarrgMOhU}LeUQZh@
z=mvDvylr`iwf)G8wRlkU)%E77_4F6*{oCksiI4PjOxk;QnIyPok0zO3P^<iX$H)A^
z=f+6~v<g}{c*wEEbgQ6W(~rY<gmR8#?7bi-S^H$W>x{QdoV@ENM{PzB-H|)oe7h~}
z7;Mio;#v`rXgV<WEEfW=egzvNx?|th!D#vEvx7KP7`{?SB1ZT`4b)Rc^$_1j7P{SY
zjj)j;bJaQf6mpEVOB6WDzzZ^Lpj5qb7S6Z2rhmjeR|!IW6c4JiKWe)~_e@z#$&=4j
zEV1rcAHK`ws!<{B-==Zg&ZDK{r}R7`!;J|ZCr%pARyD_xCyx5n#4Maw4%F4L-ulsQ
zRZ^ocL#|L*FWua$ggg1QW3Jk<|9<S;N(ti0QGCO<<omGwMYGJghP?Z~>+2QoOdNe=
zenUk0hK%x0bcucMCd$F&@mCqQ=-e?WAy>|fzx+?gb57;7zf#v5x$9;|%1>8&Uxaw|
zSoUG}-HEhZPzv48R!+Nq^H<X0+SGSabnyt1Zr@B}aB^<V4W)0wWQFeyDJF`Ujh_rn
zqjcZT1xb2)9U~+LoZ9>Q=$l3H#O}1)J+b|Vt=!`*v}*ryZDBtPed|TR0&GYAi~VSl
zK;rN+g~8U$z$BEdr!#$hy*=bNM+a9Qc(nTqX_Y$%N~{_3O2{JM;}UlOPFHqoG9C%u
zzL7e;XmS^ZcHwbAF#E-F$`7%rvi*jzkLSjXx<nc%u=TrhV%+e6g)j`rpv(>taH1WU
zRwJ=^rVAVtma|*k_IkUiN#*{5xQtGJ{?f05U*z8#{}D1_AHLVuuemM7drGXrrq;4=
zS%Ilo)f7MOwuSM2_M?1-CuMxx;~}~Iy!bW0a;d(rOwl@8m;aU)``{Z7nKY@K@prbP
z8wFUWroA%lFQ3hX>`b8;zEC=8SG~M-fxE9Mpul`-*&>3Tr+ag*P+v;rbr<>Cpwy4;
z-_>wm5BrPujqep3W{%C18HJ)z?yp{8@zk#dyRvG!d93|i?`S_wN@mRM>fFM9`4o@m
zR8QQ260a(hbo_0od6hAx?~L_=)ehW1asR?7+v9^PBHI*LutYsNEefRx3*!Gu%eAl~
zIKJIT9YC)PB>qTDP9!AZ+qbI|2I~a79vN|?=6x^^9ENB3e-T{M-jL8dtf-j5Vkt<M
z%uX%=Komhl!aemIz0{dV1xoMTzg!|+&*qd2<xJb&tS7dr|F>O5#v?Szdg?&No79;0
zOku<9$j1>s$J`$bQkNFKuL<u=)=xDUj}>Vb(?279(NTEc&s#@Nr=*M<VzQ)Uonw0+
z@e{3{pBX-}hew657+04<X?oGt<{xObNumRpZ?`1liAcV)b1d}h_3ZoRY9r%VwdVTZ
zSMiFy0Vk}k?_s93+!sU2JsFRgYW+yDi@8)|@~ly+va#A#if2gM8QmNwGAn~n5?!tW
zb|l!E8CB5DFy@)8lzDN3@<jg0e6;-2ujQXGuofby2uxQz0viLx>8BnsTrh>w%_#wN
zT$~D~Xm6U$c<@>+y7VhV?ag%trMuN_SfqcoTrN&Y=MXi&P@yhLS#dWFua#FGh{vJI
z&AI~8`kIs&KQDg!%XJh{zeRWSTVPJ%H8mB3H!@}g&Qi%aLPYM4BrE$(I`l8A@1uJC
z2GB!yP`Ab(A8)zi++E~2kZ$7kD#QMGYb%_bcx&PmUh{<A51Z?g)jPPnv9*jSbav^9
zgq^7I?Hw*ZBoDof9$fhH9%tv7(rYz*Qpw_~dP>Cej0yjSWZ*&hKT3YEJTG{9Rm|mj
z*T|`Uzp$9dsPtEbKa*^;d`>fn>XTw!o%|bhwVcwrJ5zP9j)sKb%`*Sqi^Nj}t=?a+
zEt*iKuQ(Wzk*m^EHjLF-8V)tYtS4CYK^<D~OFe2wm8Vov%FQ~P>7|h$=G=p~p%Bm{
zbqK6-2Ly(?5*!isH?o010p7tJl}JrhtG|=izBRpY6i~U?q3j-NnGoc{U`p=;Y91{L
z!Ue_(hXW}5;OmKP<<r=w{O1_~B`@sDzwsJB5^_;y+Y~g*-YF+!zM1t1I3JVIV=srz
zGD$zhxbk=!x9)R?;Kkm^#A_eTyowPnt9}%mdTs4wd0XGqi}TOE#qfV<KC?4|ZeDAW
z-rtz6dvGEEJ2f4guCeHSve3{mw7S&q%N2RMj5FA_;uj{xA!&U}sGKu-OT$wS9bWIb
zYSQoSo9lQ=wMS0N)Jj~Ay#MKyDZl%CF4u+AfBZ@DqwJ41U;Mc2b*){8jKe|lMp`Q$
z_u*k3WSc}h-Y1WJ5{zu`FZ(6-F2p;25^AJZmRTKeJ`#e`vDcxGtQYhNh&}-A^G<69
zB!q@S^-aW-ZFVlB2wXOU_#!(`k*C6N$noG^@5EyffCO+MHFQpN9fE%qDWbRu!u4NI
zBxN!@F4ufWUDd_E7TVv)Fj9`YdX4X>oI<F%@#fi3Yr>dk)?u}3hM~gCkUmGO-{T9C
zAI&QE#FvZKe#JRYTjg9BdYY|d9DHv`^K2sl|K<FYteFiX`sa3)Ir*KBv>xx=Yv|3~
zCk$Ia%KDv75dGMwZ;G5a8IsMgOUkHQS9p6~t0$7bc&d7TxM2rQy59}<&5@WUeA-T?
z+-sHUKFjhb+{OE<Tm6xXHR5Cn&$8Ob5t~8vg<%;P27B}dz5nJW|FJ7pqqh>eEFl}T
z4Hw*WYBuWO&;i<${-4@n1zZE@kZ+CZ@@Q+;Z+42Bqjm?vOWe`j!q3DiUFy@0!#~K9
zH*~yY&~Wa@?KZ3!kpdOA;BcD^dI({TKK<(WZ%N^<)vdMT*M)}{rLRv4f7Biwx|427
za(O6&c&U{d!+m#qG3!(;)v49ug3zFz+xW4>&{*e%b0=rt2@{P>;&l-6s|FuO1??HF
z<#xW`Yng$fv`0T)<X-t=*D_>!^1SS+T^$0gts)n;-|Kr=_Ti*qPyw>ApXxE7r^LG=
ziNW4kR5?fq(pC(m4k|w2mgu`z`DOCgot#e=-ZzQUn0=PHCw8^{L0HdWTyt>`w>I+e
zv*fyF&riFB@0wycwzGHX!VVTju`*1E=@Kkg*zIBkjaFRr&zUzaI!L(AV99TJ>me|^
z2DS9F$3zOJvLAdT*tMXV(C12-PRr49@M9Wtheb(r`*4a-fa@sf;hBg|AxyugM1Fw{
zZMOXY6wt4Yv{Hj^Gcr)M22skpD-?xL<toRYxV&+mbCjnjq}|!r8GNEWyI4Hc7uOUK
zYw-A7)v8A$<B68@oT<ZhL;36m@=Srz)-_R7dpv9UxOG#K-DADQnRUl>dD|}~PiYuQ
zRzoy;S9YkOGci(iQ|bk)v8gh-LcLV(`^+OT(tctOscDK@LMe;XBJIS)AGRfviTW;q
z4lUV%7{-#fP;_<p^%Ftc_3mBOzXN*$-TiVSi6zl`MdTYJknu+_Y)|m>=c0F?D8+th
zy2>P;@JchZlfB)#SD{#C(fZj*o(DeT#V;lcslLx;P=;=2kL`TZM|@bDTKTb><@$NO
z*J_26CO>CSm*6~6VZN~(+Z<wJ;`fPVZ%zz1dDc2-Zse`BS83{3hz(W1%MVr)-@Wh_
zxJ`hKIo8~`<xmBZfKx$T!Dp_0%<}={A)_V&!Ob3MLu`nnV-zr}j!}@#WpKk0yQ0&M
z=Gtb{w6aJlHER;TlSAIixP0<=7n9L3zrPdlZb*mTD}b`3^DKE#BN&INexmcmGFwiv
zALg7^wz=Q(o<H=Md0<cekv&K2EnY7V9r<As;9T{SWSQ;LvJh~7&|&uNtkd;`6FXK7
zDo?p-A}=~ll@#UFSr?qQ^)Cy4@d}^m|Fg^f>tsrymao^&oWL{dTh*HbjgxxpR;8r7
zKa>mY+%JY4D?dwrMNK-sf8gloFp1l^BgsLvx96cxz)+ai`9R(!#_3^|oa$Eh%^#{}
zmz8CoikIt&Htn@sMIG)xj;{D7M?WI&BpK)ERO8TnBxRkfnznpn1_2DTa&|Wgl>iJf
ziq0Lnp310C7<^=4F0(^<9#6kwiyFW?4r+JDqBvygwk$$zm|OP@HCQ_*^JFj_(>q$!
z)YO!!Mr|<$Xruo&>KFwp|8W=hvYW+FNGnCFTt)DRlFv@GCQdcwcdAuQemLSu=tMmI
z>{025rj(Zy{Ri8V=z{uUhdMG4R(HFDG1-BRh2rv?^!EQTlC{o$HI#bWwQNz2>2C5W
zDU9_@ajEhdX6%#v*r6MFOA*I|caho`{^X|w9LU$|vs&i9&^9#1Q_xo&_p%Qk9<k3j
zQTlQ7U#@uK2fyb-`l1r9dv;|nSx`+?_RA<Yk5wBU{&MAkoP1ahHt34*&3k8G2QDW2
zogAMK*(BGvN4QDgNQYpx&IzOG;>6Q&#8uI7Rp#0KQ=9kf`IxG8!)IMe^WHbgMMEXI
zZC$MWhnwWF#al|{vbJe=V<T*q?>Mo);$o;K+~FogWJZ)kn2Wj5YXCigrqk1|sW_3?
z)5+$t0*qhsn`c%;bUjPHwkl^?{zKl1M*rwA^-Eg|^{E1CnM9u?$VI?nnv%w2m9uxr
z(~jFPIK(FcUps_9W}BU*{Fh76;J)PCMQ-^X!{cpoGANU*ndU8s=lEj%OE~NCfY(We
zfj4U`vXMCxmI}Sr&P=OUpJ<k$)CZSW!X@92?>|43uhNw}*n2%^Z`oNr<nijTz=0pT
zx#i4tjehaqV$&&44Ib<>Qrd%fc@nGdJID@Bx%UD0!FT%2X9J3<vQIW``@=J-t$J#@
zNhi{3D_-e-*eTMi{z&kmE}x23c^SVHoXT+8*LG1bTzFZh`RH37QQ;Qj$Zeym%=(76
z!K&xg#yXf?ER#pK&l8icZw_r7cc!j|b@FcTXnei73Y~~oDSeqxiC9nfrrx7@X4AKC
za^S0)jmk6TOR$eJ#jcpJ>3o-ep};5svDk4+eGXUdg&NgRYm>n|HqvwT5~~NkJ5U>k
znX*rzA~SD7h2ygg+Hyok<BiZ*Jsi%GFW8{u%X8rr^@g!hrJ4Z!vXBQox(wYrVhVqX
z*{+kN(jOWp7<a}e95u(rwnzpaEV_s5-&W$GFHClMc<P9Y^VoWFM&M8Gq8BcP_8AWZ
zu1X%G(?8hCWH>(x;JzbzJti+~E=e-xjHsHr@<DtQqp`+j3Tvjizg9&?>eJ~6ts<ua
z52q|k<#}t_5?YuyVB@pKiJATrwplnl?1`GAm&W|>StECrIr!=}B+wLhfJ6h72nhmq
z!Wx1OWK<j*Dz4RGJ$S)7yfnh1vhy7A;f^>k9<U%zQ&5q=g4ua}H_3fN`@&2lftSpW
z-Fm&m%HXO9-(09l%RLkFUIt4zIPE&SOfTpvenKu!C*Vo;JE2ieen*w~lY_$n9*^qR
z9tVD-)iif`^GsDr$0T2?bMV`zK#C|&*mg-Qf;L+rMps6aj{<>bvKynM%gehjP@2(I
zZY(oO-*Fh4T2(eI`x}LtO!?%p2>K3Z#eb1!VKX~Xn;sVmfQzQy3B9CxCJu+Et_eWe
zv{GPelkN(3u}yC^>FLuKWAkXN=fI+t7!a@#PEh<?bNEeTzRQo!t@laXOqf>9EQ`nc
z53+~mfV_o+AD|XvK`RNE)l^`Ce4g@Gz=}wZ@Mw2!JlBMNWrq;Uo(Bp#f&j*$%7^Ee
zL-WtNI36tuhC_cxDF56*NwFpvCK?JP{*hTgkrzemtot%^T8#_0WkKU}n^bp<#IJH?
zXmCY`a`5u)3{DsB7xkjN^NZ-*@ksuDk5bgOki&I?&m3B(wR0zIVFx|PRHlMMB|@+-
z@mn4|Y&U2ku*^iw6R^Q#+%nL+sdvqlgV`hC;PbKH0PE{_%V+v2g^E$1vb@w*6w)3V
zlL>oAfI=T7#YtMRqs6l4ypCe3VxYl_7X8cB-!#QMuI!O_k$;N#>@ut1%C#VLpN~<4
z0~o&4yDK(*kJ4^qGPE}_wDP_jDEpyY%CRd{H_A+84!XD`1L(^^DN!+Pj}MR5ABWBX
zr8Ivi0#@HYmlrrLNW^F~rx>x)1zL^w;%H`0q%M6fw!aHX{)?eII~;;4tuUXU<}XPu
z^9*+W;fCpezL(Y=NB!uD4Uc>E(CR{#CGg<!IOxZTeXty32XeZHQb`mbco)$**08}X
zgF-1WmG*t6Ds=8+sD=U!Si5%lXTT>k&T&Wx)R~b>8s|pgJ7l1zj8IZCP1R)3ITUw^
z!%3-lS%EpNf~*dQaDD~Kvq>km0H<)5k@08%-}e~x&W5skVFb!<2csW*)`8f6Txr#(
z0<C`zJ&xvk%Wx{F;It^vXb$P&(B0VuV8M|aH4Gh?fPZLTvG@$#5x7>jUdnqZ`j2Q)
zYi^%Hj;pOqY37bEDUDL0idvTZ@|UY~?ZHJ;vUaN2X9(n!^fR^U?lvpUR`9F3l|te!
zmNt&=pcSxML4O5brKGUn2dXJmKt2YLv>72ppCTY`biBAxy@`t+@=vCtXL1m#xt|<V
z|9x`M*bqN;O>S-5a0CPR1j~9$Dztcqa`eRLz-QuraoirRib8$^!2;;_PVV729(9}L
zhzeA#rxZ3E{nfIO8(id<Q(lHRO+yZcG~c3e`VW{;LFbo&o(hQw0ZsjX){TuEf>oZH
z;(T7r|9_YSv=jMFBAk%L5%B3gInWx)hwcZbFABW~Q86cI_rTR-umK;Jz^!4csihg8
zTr<Iz-Uc#seBi=LP1GX@|C}%=K{)^bpDsO8>M9w+1JjKs%!?m9&D6nv9JYEZ7Vq4x
zMf>*{zYl^z781QqPyH|h1=B9#M?Cb7Vq735b_yx-SMgz8uo3=8IB5XH2BYBNQNam)
zZbm4C{O93|R)Ysw@}e_;a<gMfrh)D96q|>Yfy=gpk4@X)&AXkk&w>4y>qU2j_jT4l
zNghpei0Ff9+?ES_$n_@rv4%vl+W8u%P5Nn&!1oT;iaowHexrZ748#SZmf+gYekmzw
zPo07_viSx5KL_`V@<rLuiO!aPuZI8dm&@RDi0N_YrdH?F5$`Hr5y392dCLar_uobz
zxUq10J9pZkvC80$0mUys6&c(jSuxT`TA(5M%ca8UpTIgNQK|eU4%ilLVdN{zZ24*a
ze!RFPUuB{IqwO;X{35Mo1)SGA=O>0|>x!JR>B=+gz_5M@q2y@_8>wLLv2-Qj3Dmw?
zH@;SIya{X<GdTgMC9ho_Y*9HNK=BH(Re=QR2+b)%ge3?1NhuF@vtuP#0h~((alzTj
zbJ{3wsIJ9nX!m5OV3>WE9&8+0^=H<TIt;XxOs#b3`$D~WJfNE+SiVn!9s?+cJLDqe
z-`)HkNSYz~Og2GNRMpBs8S`nF38!;JG0L@Zff7~R1Ov`(8Zo-B?A9|u7$=5DmrTz`
zLcK%U;qjH4Wo;sxRt7xA|DGQXV?PBVj9u&47uoZKXasCbX^t|gvRS^sV)<A8<=X8Y
zfmUl8IH?0+L#&?Dc9ehCsvL@vo^mK>uEI$-)73c!$|WL@MIiC+fMo}k6YK)R7|zK3
z_rbgNfI5V;*XmdB+S1Yp+QwG!(Q7&`E-qf;GzbW}X8A<O-&Pz?=B%D^Zgynope0{b
z3?RJ6U$k%1;q-?XHW<Y3ABLCa-##3y*wETg{`SuWiVZ4lv*$T|-R=x+I{zqtr2us7
zcbf=kQNtmWlvpct-L~dL8otYIVJm{ubq?Xy3OR`MOH!tiZ#Ry-TPXN7+P1y}1WJ(M
z!H?T~(%>2b!NaA8935p%B8Ko19I*o#r-(q4v<x7b<RH{KB1}@0!M5OAp~<QUT}!Cc
zv6m!>UNYw{t-Y1{;#A-{BZC0x2B%-x{Xfh5pZS?=&Kuxn*zt%takLy78s9eOb#`$m
zf|(t_82NiR{pEtQj{7HRc)YqT%d#7I0&Yn3Bu+#eW=~P`+R#q<5bp;`0@I6wOA2Wq
z3YhZv_u25THDW4cYR=D%CCVpk#lX~U7#+ZYLV1nAqIZ;nQ+>1fZ5Ol=+JBZ;-2t!#
z!<uC-;|a12d$N2Y<4`vfs%q$!PY;+$07@!k4Ded@4fu~OVTut-jAwj95g^VGIQOck
z6@6Qr&eza%HyTOEgOq|g4h`oh1yamicv7!Hx>P{)c}yM-F=OlCCYp?R{QOd1aBLuk
z6D<Dc6ga|O)67~;^`(`t;(wu5+!5HMi1tkXv69Q}e1jdb^0&%7rQ&lj?_#p)viZNL
zdJ9w~X1jPBCs0DfAt|9{z0_^i+SP5S%u8zz(MxRw-9+a<Z<%l;N9=|dzz$GZFxV#i
zx1p0CJ=-nS4?3m#oOU4*NOP`b&ilUcH7aMB8>vUQ^;X1W`9|xj+Rm6UW{8|MBmo?Z
z#N=+9b<yiwIdA#nTx@BSgYsG*2OUO==*HyJ=Ny#J4%(U)X@J1OX7DtpQ_DX@^iHW#
zhJ8N6EV(^{A+`^>bC_#1yr%1jkP-XfAjNcWQKt=I&-`!#%~-z8t$RxAhAFf%Q>MW-
z#r|>?Wd65SBdx($)lT?UmoIi}v_pXLNWdh!<Xl1a+~9#mhPG&kX0iN}ERMxwIaoO@
zh&DAVEArQ>Uy`I!1yVaVgDpHyrJ&1(@ZzqnUN~`AU|gUVCr*Rq7~%<DR08XnS}>cR
z{#M0?J2m5;PZtts4Qi;)5LnhT$A@B-&^!tynU-bHac9^Y-2$xYKVD*G%ts+<A@0(&
zQWV&Msq2|l`%o0!SE=ZKTs&Jfb%8-%LFeql*oB)QeBmu)X}R0-B{+UgCJ4a|L7Igg
zJAiV%U^2#QCFI<zhEgsNytD<pYAA7N-d2JizKCEfs9n4pvIIgQlRhOSi5+Esxs0Cv
zzCgOL!FSP*|7Mw)3e$u#n4yi@hB_XnUO^B-TH@*ip1&xZp$)+S_t3EK;ON9QizpeV
z9gf!HIJ`p_?X7IaWJNXvl~Vt)0p)pfloVc_7(m~la&{10KgfOMVDULvBc!EuHE9;U
zdo;&N3G7H2Fs1~mw_x^?vEw-^GNVmbIQ!)AI~TD|jV0lc*r8nE|1#8~FX=ytBh>wZ
zX73wUIAhc1utEt3TXz8V_5f7~wk$#-xIP1w7yS+NG?`ZbC-DkUhHSvQYzs}<w`s>X
zi6V<b5fUN+Ho*xWMS%H)&MbpwG6@P*km{&Mw1J0f_uCArwNd&u^fn`b-Z$cOLTE!E
z2RerW3=Ib^7{opJoguNf{Z?RzgYeTxL1*FMyA1(Bvk`QQW-p85ytg;-S){BCExW5m
z0XiXe;ZO2{35u3Kt!*CeN7~3K1{RdzRO0)7u1Q6Cl)k%S#Cy@ip~@g_dU1$5yW1GR
z&iqFQr<TB&OzcH@EcD)|7e_AU0~1%M2RPc3Nz$GSA*QC?8p<`YPlM$37&jf1i>U+E
z0s-8g6NHeZP(98M12BNt50p81SPV8gM~{6HotXQ}<#m3hk(wpX&g5%YUNp1B|3|u7
z+`X?>PWk)HM{eDUCDE~@ix0Ctki8<F@@O&MSIYYLU3L8w{7}z-`$)qVf(S1YclTjx
zC&S^l{59Q5BcQ5e)#E@MG(>b-)&`Iqf9~wngz@MmYXCt{d}PIvlCO1GPZvp<^HqVS
z!`&UC8_~OHdB{M1b>3X6+lIEL&hmXWISjfIRV(`MOExR_`)uSIK5taMeRKUwZ)ro^
z1AFg;_Y!kBmD;qSD?=HFWowd8&5dX#+sovOQwve6&;A_$OqIu^r!|?}>}qXbNhbYK
zJVEX}y}WC~f6*zAdJI7KIS3mztR4ukfL4qeoFLeDs(^sWpaNAsSf8L8Reyl?4I1KV
ziwXpgJHRiQWr7@ngGt-NIBYvs%r<ThyYo8nlcutv&fxsh_3cXQ2@6Jkv9J1UDEIAD
zKZ)i(OvU}K5Y4E4XKFbzHo_x!-3`Cd?h(7$@@QgqK-nShCrBcCJ^c~`R~scwQi_SF
z=74xOo^GxJEEfs64DbrbN#L9xx5p+TP9lnsH0JC*t}!|Uf%Y=OA~-_3SqkM@Kfp;A
z8G}<aJ-(bp-f&x;pmb8K#r<}Kr(O_!^C*K~YkD{J($Ur8y#@{lEO+1IBWjfW!S}FX
zVu2%>qgK!Clvi{t7Rb91@2GLBBVxQ4c8-e1KC|4?Q*LxTZmO*7yZ&FUO&^$1!%U8S
zncDGB-?;lx8KgjoVcte2lohd86iQ01R-vAQGAf+jmpod=_{=O3T`qZoSb-vl+b5eu
zHSm5&`^$B2<JP3Z_U9>QaX(Mzmu*WqOxIme_qCO^>lXfVeDC3ET0L*q{i%xtuJVj0
zyQ)Hhe^ERQ|69Pje#X5N8-~+TZ?tcZ2-&0W$REm<dO)~HuAC4Pn{C^rekrXauhKRR
zDg=o4bg0K#mgj?Ts4RdiP*+r%Mx|1wGo&6vC{`e~g4*bcRr)y_C>7Jsz(qZ1`>0hq
zx~5w9@k;NzmIVT1D46B6urqjh2~$euy$w6#Q9RRZ48{}wn{@K&vcy|?z1Sj|abx^M
z49Esb$gyb;rRVo2-KGj|JU%n#^JGq>Dh>}|B5wakj$Z3<iDg#u=pzKQut^=L7@_%~
zoJdTQQjAzNbeuV8fUSD7e5}NZ15GqIPZs`J)eA_^%>8f^8w1_}%4iu&GYeMZLz3D@
zf(i;TDRS+SMR9Wa(||eMyW^Ed>TWbFf4e66+t9(f<+N=1BaOUlqq{{<GMGcII(2C$
z<;a3W$K!F^8tx2Pe`1sBc-G{0b5}`i#wFmHAJh)l%X<HAO%@$`ggcRgb7z@2MfB?1
z3uk^xn5B)f)qVIl$Q82gJZS8~L6Zd)3MD;^gMLa7eehaQG%Di;C}BWWdMoSG^Qsww
z>c2LYwC?N7Zdm@5<xvQvud+%3bAhsp^8o&}<}H<RrL^UfCw1a3$UViF@|-^?Z<@z8
z&18#qdH$+g@eq(xy!_*#>aj4hL6Yb-{l`scl5EW>0x6HM-{P(G;0Qsi_T}KkYq#sB
zLVgatKXT)_mAlWDjK<)?aNRPYvDinKxUYN{%^VfGcJ^lTk0C)%gNprS29n5mlg_lv
z!?**93<9Z<F*IjMhY}sMts%R~!yL(hX{Tx=#LVNFJ!A~WOehWmG<7D_TL9Tgo;9M(
zjPf`TTE@#43-uzPp%jcA%Z-N-gbSF*qRvuzn%_?DvDwkk5~Y3m=w`;Ah-+!g^)lz_
zj6HT4vNcb`W6Qz~*JrN%<?{FZ9WQ6#(ZRkh(AQv~mzt4YT5~E1^@*8uamR&8In%oq
z*#GpLH-!C)`;ttJX9`zDMHPRwuim!3&bA)bunMMMlGx$2Aq&8%^c|Zc3(kUDQI3Ja
zttry<MYzkMG7b(6N>H`x>%+#htO+>HIu%Ubq)>YEI)&9Q@o_+Rj46_n<{+ru`tO>K
z?=xNy@pI}l>zja^mPU+da5D$aTpJJ3dO@`MuC1hbd8PHboW~_7%^pwK-;Y?g+g|>e
z`EGIN!GT5on?ugKZ~e(Mr&6)w0m7L<%=Cywv!```xni|lf2}@7j_o^!)wz_QuIHO|
zXg%G8XEkHsOzOm$lhm1>-ADd=dZK0I;>pzwHB;J3E5x+!D?(xZB#s7|12gpD*5<(o
z9d?qDoE9IALmh#q18(KCgMXDF@M{Ll?#!Nc8Ax&fBh%h4jVL35-|Bq;O+WRB+LBIw
zQG%~kKGn%wF~Ri3O^PP|I<0xp&-~7iqF=3SO{;RU)neo6hIKpnYk~I6Q*7O1?=$~g
z+w3TwRF%3p7MD$~S3X@gB^9h3`*gCe^FyWb5%0Vz;aCghrv2xyt?^mkl-<@z>PGtP
zS3;JL{gRLiZdg5)?UbJUA)s)_Ss@?M>*{N+m*UT<6Jvp{>m*zRP7oZ3(=-BwK!hV6
z2h5yQ_rLlQk_&4F_#33PQ7mY(IDg4$n!_s(8gBF`Y@tH=#4SZL$vkMfapXvEiC+2@
ziu|K<8xnaQCQd&`xDQRaSZ5!#lifq8Kk($vp3MXvRb1_J%;uh9pC5)%XH>hz2)@a!
zH9LK}{cMnQwtLjl-Ow^6y$~VMVysN(r~ApH9Wu)obwB_e%#nrlrWfh~Tq=2=WiG{y
znf|ENj6;=e6N}#~&3MM+DFyOqDaRaBP76#)X&12tw8@VQOss1(b6PYRHS`jPraEFP
zn$CQ#IfawQuFULJQd*fI0%y3OozPtk)i#V4gxlefmV9NN5-cyFc}qG-nE{)DY33K|
zmp5&oGP+C$zoRWIkJ+%U?cuE^P%MU0PrG>NeI+>EL3d=bcF?0vA32jom~^d`Y{KGn
zI#b@QxI-9C2eMDmEFS_Cr=I>y)5^p`5P}SUt{d+upU86S&Q&=UL0_VUAB4m)$bz;e
zm3ajKb7>OTO_b_2-EL6t0;n7kn)ZR24r{O~e->mFY>Za_GGTXgCHIdL%Xk**W`LSf
z+R<W4GhtH2=<s({i=QkLx&Bt$0i>4dd#@*Bd&HfC>Zq?G7B^Q?sSYn5J!{O-otd`^
zroEF%Zg<Z~-ZHL@b1CZCdO`Z5>FwZ}nH=4@z^_MZiz&a$cV8EJUn}!6M4d0;>vD19
zg7#<d`aIeOFT`ClzHQeuIBM`^J<CBV29N-fQo39FlTXjbHU03zV$Zs;8Hc!Q!Zb+O
zh3Pwwu5J5qcu%Um#r$CHF$2*(w^ctkeDqF=u<-PhKJW@Lc7z!~9l>4kdfGskw0gO4
zZS#=arnTJH9oN=3o#LK(RQAnSx)~Ko$xgO8#=>_uln2HRS+I>q@<(au#M^N<f3fAR
zwH*nr{Iag@uZ1^wc)L;FEM;s73TgE?n_57i@I0B|jSx-)2E0Jl3O!@!fWW<Qeii}K
zE^uItrvUsJFk!rMxTSJ%!r%gD41_^v25=HPsEIm@IXAEUGy>Rf(0eN)9G-k!(*&+r
z>BM=-Zw|uK;n8xHto#SP)#$zlqh1W(cA@-&9Cy*aM!w5zyR#C?7e*w%3|LU_jllsj
z$qiGq7DG(b;k=gQIbp9UhcPo-;$i=XLnX5oc3MP-i0#2Y^iNXFUYDy!VZO<&D*KnM
z64o#3AJsX9UvpSxSK;~xL2o8vhl~_a1cC>G6hs{yekel%MKdbs_LLVs3~h%FhqGfq
z6z>DNmXVh5t(@0L0Mstp2ivWHopLjBrhhRm5{rkB7rVoQ!J%%m%Ljp+j=-S%BAH^>
z-?G&X-2@4ofaY|E&Gh1>t)ROur~@OIHJ4QfTAsuEg6oB9Xbdr(Kn4Goi@*Jnz-z*m
zmNtiBowLIH2QSUYjhtFF44FG+c{$;8ahsk><AvOY_1=j$<!9d|8(z1ZiNI40PX(^d
zN(i4(Ow%e(IAa|+{WMZ`b)gey#q$H__-?=O-2u|F<a<IOoQCp8;Mjrxwj(E+iBDvY
zT71BjtseW??2G-&6?OjDK$%OkncFWITmiVyJoii~N$9w3p<Zkj2jp-pAR1zT@sse#
zHhC5x9!|#DoT?8|0)AIjfMOzI$1_O<AV<=&?zj8I;p8M(-aIlx|FRp1fl#JuSQxtk
zY^*4y{F$281w$Z1?iL0>dWLXlkt+h=)T$sJCkTba?VE;m4<CKR51fPzgC%8Q44sc)
z@SF7ilGikKrTxVe+jn)_TAmJnuNZvRnQ(7iP?C&(*60{1=eIvXdr&vMI$Qa^NfrJC
z)$8RZ&kw?amcDuT-c+F|@sf?hkCEh6VOB?)>5XpL^9Ju0<C3aYcmL%I@_yUs%i6_l
z9%gv%esM|d<4VU7``Mkq(u!?-3N=R|b5=NYSrKGxBY7as5glsq7f>I&BJTP^K$YMC
zEUk)gFed1=7`($zQk$BX<~(AcC2K{4s?Xe$nRZJ#;FuLu&vUy}VH%D@yi=gS*QOJJ
z(Aq*YJqRZsV22fN9VC1@!?<E}!A>SHnW~&ll1n?fzdx3>tM0z5uBLENlDlBS7cbW%
zi$kX*#N$tVY(rUD87SYUc7+)V{!wz&IL`a+IX88`M1-o&pSH(D`rT;4^V6q4bxzcD
zYSdXxllYncLH{V91>-YrgU~picTn92q<E-zQSRgN@pN-!TJD^G2J&DX%3H(_q2pqS
zxt1_gP`0R9d{J@t?F<R=HY}3v!!s=A%1U)wcw&DC?wkWjK(Uj9EKMYu%NBIUgaA#c
zXq;7=+%B?l^3Ij_TALK^T7?dFcG?Nerbf6^Wj7vpxU-Z$EPtPD!=~hQ@}P4f6*7qM
zxb4pxKNa!gyEdP-Dj|@}X<@y|YlHa~eY$jsRqGJbKVgdzA{LWXX7Vh%WcZAU6=T7~
zX0ROk8zw0VB(%H`NU-sN$87Tl4v@4c^x<9@Ks!d17_f46gI@y5W=ArNUIN`DPabrP
zDOnA|20P-yV0;ZUHc)H-3(&!rR?aivfKpkdyf%mx2IihK#fX;C0iHq5_zF}5IS7ut
zk`oGnWOEEG`Pz~429A}<1N}DKI1?qUTMIwSqvx1dDJkwIt5zs+Iy2JD3Yh3)lptDk
zyIoTm<-B`g-V4$#sE+wpK!fPQ5$o_oL3w6UIM)Z$H28VoT$cf<t6sPG%k}b%fs3WG
zSUnRuVtRxipUwp3K#5z>!vzP+(0*(t8G9FSypIXYXjv;S{BJ6wMV}TZUvcJ^om6F<
zZ2&OGAWZ=~CJ)Ete(g+YFk5@%&LjWcxv%xE!i|68a-gqgpbewDa`sDyU}cn|(ab+;
zI#R!XGy@449$CD@a_EYI>fL9m`((u`AB^3InnXAxiB-mNzu^$jup(<-Kt_Gxm#<LB
zMj2@zS}XElRAv=ncf<l$Ff3kA`X}gsp)Cj?#FZgqRQE-RutVJ$c+cSa<tQ*Fz*$1s
z>|t;az9Cr60L;{!$X_n4@o(B!#HKp6Ie#~AXN8SBgy&!ZWe{C^ML<sxe!E0qMO@7r
zXo{#{j)=r4Wjhu)zDVeJb!VnR=j<p{-RyE-GsY<xx<ws&d#=-f<JIKZHm3qsNYt$y
z7#!>nq%t7@#pKTq-r(=I7|T=+!(qq)^$rlFXG$VE;Ia1Ri2|swPeQyXWZV#91p&3(
z3Ed)230O(31ym>l3JMtmP8YqCixcbBys>^}FCy)2?BwDf9XT_I;VIw8leVq*))ozY
z&1W_wMLy^FU2h{KD}EcDJ~Xhd5l3VFdrYNi{L2q2VMQM6cYiC3QvZq#GfjI9JTLbS
zzm~u-SMAL#6C-Y<olrq<+H~7Ay;v3Mdp91I<2mx$b=EMu2T4ie446Xov*Q2>kq*-X
z5Y_`o*tmcNor5PzNjzn7&0#Jxo1M)yM;-juw)D9_)NgoNj@Nl}jE_I<^1J<s`4Lxs
z1<Kio;9|F?@h;o&dQZiOH5){;@5=gqRQ$xgt97l~lIJOMh}oL##~kz$;bF!K=gncx
z1uJG;%Cv&D6~(YPNP)I|w!4JCD?J?u&v$@Hc?IA}zbT+c1SLfwBMsX72GbH|q<|Ra
z5P0fha|cRuB9`8m0^okG4bzNw?ZOETR|{@$+$xKafM1A<erqNpcj?-T>I_-O5(}x3
zu#^5lu6EX$eZ`i;-cNTuH;{b5cXZC*-~=*DGR-;%-PmE`Sk~fkb2LUc{8{6P=4|$y
zpovGoBtQShOs}c3wBjdU``!qITtxPII3?zMF1&i-2v&L#6X6$=m2@yP&1@60)o@is
zu7hZUK2-PSz2QCK14bXr4)^`4`!bKQLPxJ4X<BK^po34~72u>ui$UN4Mr`V#Q>q8;
z9%*eWC-yq9yd5&^7c7_%Nif!bH^_^RF+s6je9LL-NZBXxisF#EfYq`}DT_dL$Ha`%
zm~eC!;4L3IWZ;~0Ls260|2+u0UpS^UiEj=&Tf4OaI~LqiQLjhnWghdCn>YLk(&D?{
z&8)^w$SC+hHZgCM{3I=R-pnQBUQo<*?Ect$KC>f1jjJD#K}S1p&KXAOqH~ql4sS-*
z(--8?5hjT@U*3(nkbS3oE-slc^5CX>i^x%V(-$$Sl#?&b4t)4tKPb5%=gMewn{Vjc
z!-GRw<_>ZC_+6{=y^-k8z}#>~(V(iCPQwq`cUF0dyE+w=&)r@P8o_M7C~q;ggtpkW
zOADSJ_z+DiJIs7;q0MQ7=gpc<5gLcYaz-=R<_2#6+AK+XVHG*06zIg@wECGHPO0pF
zJs4<$0?8(FCTiHMKU!K?Y*CMgo*|w6@Y~G^eNsu!^a^ua?rf}S*W6I><Z8lL><#JQ
zxQ48gFVTn22QkwE{0?WwC6D|`W`=BJ==lUO*(w16${K?&*QHGPCyI5vuO~0d8SNC3
zCj{(aTi+<@dos2d79#@ceZ%$b*Vsi->5sp<5q>@>U$^+`&3u8qbP4$Es<%Fdz*Iw5
z2(&c0#)Z=*J_kw=4p$)C=7<oWt;Uwh_ECu`oc;qe(nZW6({z?Sx7gL>W{d^XTz1r_
zyEC8Y#mRl~GgP6%p<J_g#66leP7x$n|9n^EBq|(Ftpl5?Cou!ki<A^m8P0~6cB`NF
zw-Hhrqk9X#ybvpVny$p&&%5>?bIv*UUG~w`0{<r2+vn{gt}Uxx+}W&>CI;YtdTf8m
zQ&&FgzJ(&jgz8jKwg0=-&B{NukFcZUpX?WkdfjYm{9Z+KgD3YZ;lYY%&3}KSkB%Rv
z*k;rRW$WxHAM9S`3%V7YazQsHD?TJ4I?8NlI9#k%<n+{EE}Lp11vqvPtH6E&e=s@n
zU<)iD-Z_m%Mm=dc7*c16yYVamW(uhnVE|UfvT+ViH_(78GfEF0wI(Y7S`K|U{92xB
z3QgTpV%GKlh`J7Ns`vljkgTq~vk$U4M48v#os3lW$|w;jGRxk3MaVHqA}bjg2if6J
zIf|%MhfYX`?Edes`+J`M^W5h?w|k$<Ip6PRyx*_&e)%r9<X!X|P3Sjmi)Mj=fD1^;
zC?O#sP84gNy?g4z?z`#<DJF5YdfE5OB_e~;Oapnk*rm9UY1f}mJY2YH9S#MdQ0#)4
z*Xk~HbQ`c6#fcJqJatsa`@go|D`7;26wasF(O;)$6CMi<NM<eSoz@M}N{OzR<j#_4
z&}^u|WW1L1fW6SOebg*I2Yr-D=x}@Nlg02*m3IHXhlpV>;N_QH=A_c14zY|l%@<ea
z^~^%#j%XnxBa{)!q9*U)KB>|bGa4EKZznBqeT4x}cVr@(Wd^CV!GXaWeZ!7wU`s94
zO;g#^$i8G+_stdtdXn#T3*%s<O-xC;b-V;2e4U+S_d~T?PtBt|eyz)W*0mOh%475~
zx21NxG<nDhFLhb)f-z=UHlm3|z)AR7fKIzY?!P@M9*ft@8{K)M_q&q;+X-^Jl&GdL
zZn(6sw!B_LM*3(VlS%NFCR>!<(h50b%{QS`+UMH0(-k{V1o1_nThvm0?rKsVzYFX&
zv~**9$M6(OLD@L%aqJS>{Zsh_+B1TV0|7vx@&O55kwILXRsvXQL%~LZO9p)em7>P_
z32(f3(ocLGgP<vo5dH=&%W0ns;dtg&vAB4wBO42RAhH;h7Bko;4_xTETaUH;EnSR2
z29$(FkjH_qX`reRJgBL1%?<G=2FLz>D6QmhDGH|o0pE5F$sEmQFr;9t6qf1Xq{5(d
z;E3=<P8nqXWTJ?04Jm?KeShp-0>m&lY?{D9d;sM}ct~Wxm-GYT`V3IFh`bLvRTI%n
z5-viB!rI-4UJJVk1gs}TI#d<c%*hh~G;msvYXXM`{0I`AfE;LM$Ha^_K~MSJoJvq`
z6@CQplOxh&TmGB-NeDrW0Ff}r5?oDF5ztN>ng0MN3B0xJB6ObX)f415dXESloYYtY
zA=~>xV9$R*-lUo@u_0;scSC!`0VZs~$`?06v;8bBfp)!!%YfeX2PokfLllr72iQYe
z>~!Q&ncZ&E-mTnkP&R@ewi)m@Xl=oE=s%HbF<B;Nv_6JQR!=Q24|1=-Qi+=07#=J+
zQ<OP+K7B!?<({r{D@|@?R;wu^pBC|vKCsF$Wg(e^_ehQ%x_}skfCug@D3LVbw=()e
zd1hy10o;d>fa+odXzu|lN0==A!Jk+KO@`ck$J?P-wk%uW@rwirXmI6LiD2r%>ki72
zh0Hh4(zV~jPCl$=S-W#tVNZegNVBTw8z3`dTPS#sNUN)IsHc{yUOlJ7NX49!00|oQ
zTuy66&QJzuqS6qx9ULprG9rESLc|?Gye<P&&Vjs92;0^O_PrYzfP&x+Ofavyr4>i2
z(*WVWDU&Q;f$4}lmjodVnxZAPyiuPqJePxHZVqS@1O1DBibjFTCa%BvkL%i?Dupd$
z$PIXgKpJt*r$!{4YF7shEisV(Fffzw=t?f)?BftTGY+tImt-+0s}MEdAQbJ*|KYRn
z%{i&Sr3NcQUg__lE_+BYuu5|Hgh}`j7KB7502&ekG9>zRUY+xGN6xB7fjxDw<df?0
z1j#_!etVE+fRYdRaquc3p94{38rdg+{KEb7f3N(GT2Y*61T)OcgtrZ10$5m?A|~mu
z)L>b0s3*>YE(y3>)Nd>!LtNphx!oQS+!Hs7HhhVGn}NpzuIi{pqrP@}E~a={mZ3jF
z>5t#q@t?N#W*8@#)v0kb`T&i=mON&>I+1G)RhhE(IBfH=<0WU9&m++vnS%@C`2q2t
zRP`&D;EH>B`nzCk?H|`I7~>2g5Dh@OejpzcJ~eD=;~pJYNvfIC8Z->~nZSBk+uuW0
zPkI_~raIOR#}J}ZHB*P7(6UoL;P(m6`(L{zT(?ttxfs_x`QOheudvmL`=6gHihg`G
zQ-HzfX^R1Up|A(1MosNzw6qEf6Q}0wG9!B2Vsb`BQP!y|pJt*O_!FGdT9kUk&uH1M
zokkxaT}yjwX6|CPM$f+Wyyc~NR+&EyvN#Q(zyv*=+TeFB51_~9HC1`bfgZSFWk>I<
zZdeIgpOcKttDm4vC_LC|S>TE)B|+n%9E#mi-(E2=Z|99-03qW=l;}KsUF10M2g(|;
zT!3EF<x|um(kT9~xfGq3`DPidEL`iUNpjDwykUGUPl2bVb6oMk;<nr2%<TCJO_F&^
z`j#iXxHNAcTa8V@pALVY4W6RdXpS4&F<t2bZFgh9Y9@w74c6u^2~*XtVfsW?>MSFF
zt{Sa4e=ot#{l4aVfkSTm0AhrTcU3xy5wcO>>5j(n$3@AlL9;?f?8_7X@8uZTvFrG0
zqKZIvrXF)g|6}}u%U}v;inV%rF_%onGvSR;w9<5LUv?HOZz#PoG#OJ8ci~`10Kq5_
z6&0_2<tlLXp6=STjZNCrZfl^~$kc^VV#(W)&y|JVhH5<ZBE~1#OT{+Z)Ge9)qE6S_
zpglD=PZ|l2;IDp4We|O8HRe$RyCj{h4Q1wKCZpHr7oyiSbc`>|O7ZRTRXqyfJr<{W
zWj;hL{A|wW1w@!hEt3F8bR_U_h%f*V%D}tnA4Ha#bQxo%ExLo=)MTyE(Iv-a@E{FD
zCSqG+;n`p@YIR^!h_zddi+ymV5(HQp_bQP8A0)AGJ~3~;@!ogg5{p6UAARpE7t9e!
zq3jt}d6>uMi`Rrh$!$YU1&_Vn+K(P}t7t3q(Zi&(<;<L|mpJvWAK_8&Mn5(?DR*JP
zi%yGI{pp@4aP;iV>tj<&p`85`=l3pM+a)CLmxJGn1=fb<$qYl#`64Rp>f>o3Z>AFD
z_*yiAU1hfB<os0%aSvC>u@FTBItyDy-(MgtwNC<}6o5~V)PqQz5IvZPf4LYWDnQu;
ze<~EsfCjfHt8-8(T=u!P${zKA)ys?vpQ)Y*l(^40%<M#ICI-dKEEGL^!z8aKeIvXi
zS|#rf;pM8S>SD?*y87c!d2!VCJ$XJ{DqraX`z?$^n9ul8eB>=D=G|_5nesmD@!ZEn
zP6dY=>9DAtcQGww@4)carQzlNAgTk+GHe!muGKA?1yI96f(CQxE@Vy+SFfG}p%TV-
zjh=8*cmCqB_&fX`u7H}O-0t6Jw!3GyD$Tg|qeD3H6s}O-zsqV5>xGL>)>0GERuwU3
zY7BNW6vxf(4(q$lGCUVzRoG!n6>y`+#cT?lq}L||Uk;{Lw!h43@8A51i9KV2BYT5`
znswf&TL+&z-aoE-Y{sarK00^C)#0qhzlML9=qV;~L43zQIF)Ei{|OS>IM$$3;g>Y7
zEn<>!ncZ$$$xR=3SLAcQeRzQ$aT*K|04snSC4IDX{C{kTn+Y+WW{Iz586fX-I<{)`
zI>Pu{o(<9q0j}nipA+X*2^qTRP`IUGPXLG2*kQn*3{go8qt*#$MSus&#bwANgcu{o
zyM3i5z=9S#2VlkSR;+wiB09fah!3I0=+e^a+~Z<C{GK|X+|odJ93SVtU=r2Uc!81n
zF#QEb6)7&+fc_gykM7LWht8<=$Xz5DBoAWdei!tZBw<@53h%(}6L&7(LgW^I{=wUz
zLM5gqzaS*fq>vs^<bn+Ba_&sBCxvGXy%wKz=77HPhYAqC|C(HoDU?MLqIL;dXoeyK
zj7Go1JX|(SD?2QYwb?WZ^yc@u%(wU>4Ap4o%Rz-<zH||n6ASlYxfsUoEEr<ZchJ;b
zWFcKx2*B`7BLZ}2{Q>BIT`q=bd6B*#q+xZ@cozX&!oRx_r;S`KBh-zcVqC}vf(Imn
z@`MN}Pa_9mJvxup$Z;}w4*W}9Sak757Rv0i=~8YWSon=xCr0MvXapjHTng)i=VAyL
z=88~8jO@FR?E+9Lk#cnrF2ef@shI=_W)c+bKagy)Q-FfR#1A84H=xSJbgnKU_ft7B
z3cdnGQ4&~^A_vP=+g*NC2GvJ@syxf0mI@+>6Jo}2L_ptSpmoi^T;>}{fZJi?!6$+O
z>wvk@Zu@&kc8+@tXXNk-`DnS9v^)%D5;*kZGdzkZ4HqH_fgo&fqHq`m0pdP%vSD`b
zmR6uN5sxWGB>foF@5iYqTju^TzzNjTWaPd{1~W3x2mTlq$tso<?blFVg!(@ub8NL-
zSBKRDZs79WQ<$Oa;af5D+l=a}0FEN9lf)@3nbKs}WA^VMRz%%K^$t2kJ-uL$2tk=%
zxPBq)N6h$HFnrdJoc`D=@IenAB>XL0y8G`T2{SYJ=(fAZc)k9pMom;{`EMgRAqdmn
z1X7ySN+Ubz7O>rKE<)uBpC<^=%+?PM`%WF|(a@s8MUGtVIt;qF(*S%G_g-PU)xXLD
zzY?;0hRffjzX>|k%Aj}AAQ<%UC_ls<b3Y}V5`I|jr6H)zL8hus`F-EHqtyaA)hS%B
zfCPo=9Ds7>Q22YaUo2S{(D(=Bukec&H&W|s41@;N4L=pNLeX~>G1p>3(IDJAvn$E+
zxO_E~6O!k&JU1Xwiv~(84~WAmG{94c!ZiVkJ(*GO@NW`xC4g1_=*yGU301H(-}M2E
zjia0h0xgIx15%?ZV@)mgE>Lu@=7Wf_1DF1&fl&ndgTz!^xef!Odram;Ybro7j6r2;
zVWBsY#;O|xGr{$A!xZ^N%qGxI<<3X#J~2=F@Ark7KphT+bBJZR{IlthY`+VGyapf%
zgt~ck@+=akhWHLp&?E3kiJ6^DGZZZ$T1h~C2x1P~%)#X2b*Ad_xN?`SK;uVeGw*+e
zPfCrYio^7P2-Z#j0B;Zt3}Tso|MzNxROYr4{tz7^{eck`3p>ODg1fk(Q;2gLN%8@C
zA|$(DbB*!-WkA`j84|}1U<x$_2pD*jIbf<BqEXtV2VrFTQLS$#;B8QVSd%eK7&N5R
z;9?13#o1Q?3J!lMQ><!*AJ0_f6hTBHc(|YaUugVhBn!eFJqT68Bw+6FwHY~fxy-5f
zGr=&vaDH9%-q$ybgQo*@Rn^4DWq9Q8FGvQ<Pp5y^3KWevBNU~(>H-V2buI(>U6{H+
z-d2zrPNW!?{MUy3!cuL4APDeGq>|@?_Il*Ld>IfarauZ5fGWsEE=f0&#J`8~I_H6v
zwo_>3PcabEFM*>%LjKXm+?s39Zrl06sR6>=bJm`%)bqb|>8V37l}BP*HiQO`$ALQ}
zYU0*^n`GZ^!V`be^Wrf)BvkwnElL+kR*z2nqD4fa$52Gxd{I-p@;WZi!(;%1n1C!$
zL21P-qUS+*D|4@94UJo<?mg`<H9q8b8&zt}(bVYNHDaG;W@TsBl@&E#)|d7ms97}K
z&}@eK?fKrnhn71B+!4!eJotT?2hTF+B>7^gfGS4zVwsMP^+YZdEl|#-`vtR~52O}N
zrUl?g?<Ul+@bsB+6e5}M?O_$<eu@(Pf_wwxDgC=C3NjIeiI2_}X*Jqwd4Ng-Bnl|^
zIuh`nxD=@BDdDs%@<##|R{^Dm=Om9+*GYe`e6a!{I*=ciRsO$nev`XcNa#jXTAeC4
ze(*8oS9!%i@z6W_?)6iJhDi?7_y-JQ4U9aq4y2ohi@Xg!rhir%ygS)b`^BlSKz{Gm
zW#iX>4;8l7sZ8lsZ*>k_*9plUjS~VXodTE@3_v(%@dmN@up702k;xvh>9kfNzg)@H
z=lkPw9sg(+AZgSAXfRwr<m_VL2|2i24{ia3Qy4Aif<U9g{zz}uICv6d^qH_9WfB-7
z1aH}aamoYHrG(~ve}n>f%KowLqbWT&;e{Axx2)-xj(s*CA-E=KJH_ie(uKZ`POK23
zZ}>rZR-;}?xwL0QXJ}#8zA!bE`^FGcz^OfL-wI!j<E=bJ`hO289j+!+#H=xV)Vs{-
zKl}KX?wJqL?glr}sw=;}?PH`9C)jmwAZ%L|Le^-og<N1|uwyyEHxV`iYCR;Mu`jFq
zN_ZwCh;GHf4F;mJ$&-9~foSOBFIj>#6cny-TR}fYjuyLYe;<F$53Cqzn$(+@cDqLi
zFCyi~obMC`)8>%Q$~5qyT$xh3c{VD>^CGNNnW~$I*h5eKn}1@4p?Tg#d&>CL8tVLQ
zmU`|Rfj;lvX@?lu1f4#sD*(LBYv&ezbbZ15ORK`JW+*lLTyy}A*1PW&mVR${d7TJV
zC|tC^H1rI#VDn1@Jq)Of@zy$6dyA4fI31*@m?K2vRSghSWpDy=N?fALrE|~u@u&DB
zJo-tNR`A(89po(&9ZbmwEF-Jk@?f#fmrtSuG{o8|$4XJ^M}<;i)Q{KvD(x%iBNL?k
z`7E?uW6CHVDpFmaPQ1Qwb$>fN(Be*h0pIxtnauXMNNxpuxYI3{KNPR)!+-no9kIzl
zhpF{S<d1DEI;^u|_hUF8>j*WF!*jpQ0LlK}Z`{8EXBBo+Vj+mP?3ZelW{`|x?A-@+
zbiNRUnJI$lEM9RaC4#Ph2-eXcDgqpoAd*LP{(}oUz|?`rMI;^svklNcVo*SAiy4jF
ze;39Ip^Pb}O#EMKH)EDxC3&gb#J!ZN7@WBzxjd!Ws`Z^@bIB_G4o5|_twO#`In%Sc
z(WLN8=X*5-UcM|@E$7IOadF4&_k*QGn$K7KZItco$c3Rfv;B6p+5}rcx5-@ord|4$
zphcOSgDnn6Y=0!rS^2ezT%D`114NIY)eRpe=I9QKQl6s(I*jhV{9vIIv<?uMqjxS2
z5U1}DYq+hRitV6*Tvvm^@4y$L4kBb%XMw(giVZ}KVj$H*>u=_;Qbf;$e#^imbK$RH
zqj_g&or-Kqd@;URo;vw#=dZRa^1NWU?vc4NiD`~Lw!sjOeKe;_aN%mBTB6NTBs;-c
zk~;!%uDo05aJ%rGrJ64SUpyJ6c6d_8rp8p%LK5I|M{J2;z!8$jxU(q_lY<$rAqoj)
ziI^Y+ojE%ASVwS8cLB7#^pKo*e*r$9mZv&$9xNjNDANNVPtmhx5wG6Vk~qS)N8=62
zn%I_w+j!$(@~2OQ#8f5$FFgxoL4#gwoTu9OEWuDJHcQZ;F{vW_+uH>#4OjECdOK~Q
z8YU*~KQ7E)>WDfm_xK6(CB-|@Z^N#6cqhTZfF}iKK0f!~=G^qoDxVYyRHJ@~bULEi
z5V}SV<&=L4ABFjs-$B9KSqL8~s#%~u(7CxIMmTxs(-vB7-T5Om=F}G7hERIx+u!My
zgq<0|4OB%X1p(Hl(!^|@X#3m8sZO+P)fU!mCwOIRv2VHAM>$zppi&FNl&UhOgNhij
zUBM|qU+w4qDQeTxzx_P&2$(k$pvFN$H3k*>AWpc#0C|>za)S{yGzc9rnu*O_UD3N$
z+-s#6lCwa(`W<IsXx3hJ!9&+iXZWLgT0!cYw-#=acPxkaBTQc>O>g}@)B|7>8V#Zc
zaMV=8MY86EtRMl6IOKki!I>e&f>76avU<MzamY|C6ack;Mizp1=?Ea~yAgohVqQU%
zpG1Xm?bY1hmIb7LD$$@7OUbSBy;C_UC)5Ump#+Zc@_76hJP|aT(y;(7jK*n&QNTB1
zC?4q+l=-&iT69ZLa1#vI;qghquj5&WEfwfrthl9j$F0VRrjs)KcuHf!p0o7R-LxlE
z#tV%zU;vDP3zB`98V(mwiZc`|?B8A_yGHGD=&sHpRVrxm00>2VDDRHJ%`r51?twLB
z-UzwB_8(6ie@ai((^W^Fm(c?^xolE!PqVr%MfQe&YN65;3}Hd!P6x@h$fMX-M5=!y
zhX|yKqu@<?mVRmMcm=?Lh?lTo#b(Paz34p%pAh}Nd!afR7qNSsh$sVKr#VrXdIAex
z?q)FXYFD7_jjQ6)Yh(Sw-i&FFiQsrt?P*I@%Im4_k$(@JAp9BEMZ9C!18SQNiRO&j
zH_ECAMEsJOgODGS)h<7nax0Xkd5=MjW@res1TX2qB?u5~m=s`v{)NclYX1<BQUy)@
zx|3;|ktf;V;&(-u6y~R}dcZ6A!G8M)_-_-TYS`ueJ#moz3K6_ef$iS;p9?9H@{D@I
zrWzwjPa=JJ@OL?dEytJ7$m$C_0b2K4M8N+4Ou~=7;*pqnB=xW=Z}fP?^98unWuvy>
z_Rd|E7rR^dR*loh8=ZwQ5(Qz_Gt*6(y^noQ;cf<q43;QV7>m#Wp+U);1PVpIp1J%j
zadC(s6c?91fHl|cJVAc=<Fb7m;?~R1R8}8rIGKIb0tfKJCoGK`*^naa`tD5MReBwk
zk+g9at_0Pb9(l&J+i5m@P9$NcL_a#W>Xl>HNs8Z&O?4!kAFlmP;;O#Q?OtW3aQftr
zjmBms|GMFLMY({4Cv%as-0^Qhl78_wKK)AVdsXnQZiVr;OL8RV_uiXAX$1?xRGB6J
zTu;mQaN>B27X8CiT_T|P<vU)oYCO6BE-I*JIV!ayeyS8-pSZbg-)hGh=0hgRL}W8o
z!cIl*5D<}O5;g=yVrUYYWg;MaYxBLOrfNn|_CN^{!#s3{X1r^SS$jM&LhxE7B|W?u
z$VR>53!xY~OR_FTdiMcy;R0{1L{TCNnnmOSCC>rk4Jb4~!LVh5oQ`d|{X5xZZ1?(J
zv$>b+3|&Y&v2xB2Tfp^YNogd@O`LSy&ALuzeQJ+>a!R9iAn8Q6jsK!lar5jnXJvys
ziLa{Ec_Xlg-9swXGb0DZeVf!reVoer*7f)DBsNS~ztf7ox@PoVqr1Jfm`aYR)p(_l
zu-_RWr>&?Arl0r*Qmqmv??2y6muodm`W3vzSi7x$ipezjyGfIR=jo(-V09D!9!d(f
zYA;~)VJ1Z_F+5p*ot&y2dKxG2dlEP4e`Yr2YkUVyd&Od{t0Ma-T3inUL8)n)3NV3Z
z5qw_HB9;!s#LnPSWCaI5=tGm4)1k6EXYIq>4fze$$3CZ4l0YB|?Q7&oxK+R?0-cf$
zq>N-3f_4|g$O5{o;}0VQ!%m3#mx~iwo<*YEg6j2PkAW|9pV#@k#l1SKKE0kB?>R1!
zsb_~iWgOv^eB_TJ?3u~f4YidGgv7j)AQ%o5PDoDe$1X0_T7BWjn=p9Glpvivx##cI
zpZTYMKO=r8^>@;qo#?iLZQjJyPk49t(!IhHGnYzMF1?@n^Bgn2Da2<c?nxS9BJ>dP
z+oB{|c9Q!~&Xk?e#Lr!S55>Do`hC!ED(VjYB-Eu<9w9o;Q`|I5?>{r*b6e2e&piDs
zljF>sZ^OkSTN>l;K!HRsr}IV<^d%1hL6L21q0tYtYDm!tlC*SS=?A{i&k&)zzF^^0
zKaB%{8Wd_cqF$;+7sSMlfdU9}PzU&v;lD;gkZzd7FQ67epsO|`L7~dHz8PQ)3GuG5
zw-+-|wKBZKVAEK!64&<`g|w2ZVw8t^O+%OSGI_r+Ez?hmU*=98Q@H;;w}nvF`%Kxr
z<z|i8N@?Q@LDh(UyX^oB`K?Ut1zM*U=6l?+HxJJQU0f+%f{?TKC-?|^M<yH!ty81a
z{#_PlW-D$DONn1r#gKZ{#vZxD(I~X{eKRI3eMfu$r=nVc&_;g*U>^ojdr^`N9o<%{
zOR+m~o5?G0s^W)_TwW9BnUEbjdcJ9(1G7$)PC%qvw;<oS(207NXIB{(2PJ)7h?;)g
z&ngOA!CSBfKt=o%)KX^=z+w%WBvnqt9ct<cUUCGhLaSr01I+h0nSC=)o01Aq^Gcu{
zhUNO}dEbY8Cs=s*`cST>bN)QaeUy8W{`^)j{)S>h`?$vBg^r)W1btD=x}uB`=<EfK
zK#X?7TP&v5$ZOA3;@t+CYam$fr`qrJw5jE(jYzxJyvxikPAj_mzsZezzUVH)Tf&PV
zTD+V3(M|m>+2_5+RK-+-+Hro{VvSd=ro3`0WSR6;Wpx$$M^4ndQIZC;>6AZ?7~th8
zWHR>>(0Zr=G8o{HOpJ-#U!MN0=DTo1fWmzQDjgvqI5}=fCpZ)_e}~ITjY6dH0A@#|
zs`#@Wtse7aRsBeam;ldk234mBL-7f)%;>tAEK*_m?o(WlWRK8U$vItO=*Om*s_*!q
zSma9`eFGD9bXU%!L4}L%gu%r3!Dx>dz9-UNT5p6LJ7=UX`d&9vR>&TRDAF<_Ive|X
z76-{$jb(x;VO#W%+yM0@f_2c&Kei@eXX})o@R%?63!7Z&tVy1wGN@sUovv!E3~>85
zhw9jqc7?r2GPz5}a%CY&mF>y~@9R!Ks&+LE%l`MM&O*h|yocGONw($USHQvPZbv@0
zao9icq>U|ubSnLNm9Oi$#p0J}5eF@Kk?F1QDP4TX6d+!jJOPk@184P+gnkpS@gRh>
z0vl}&9V@#oP@xCDC1@dhHx)qH43``>*JfAsqi(NofgQP6G5NwVNu4_iT6!W!(XZ5E
z78xc&xX|Y_tR<F(83l`lLKy3E&Df}HUS^jK^oj&9m)doYsoavIWItxwmTF7tjawC1
zxy@Ebq&xM?>3Eiq)}_f49O_$vXHJxwoKQX`5$xwS6%yyiCe#)C>36B%!lm54gz@>`
zf}@hD7uvTK_fit;Xbnjv0d5NQarHCuXDaK2g3~mbFU!3kxw4>O`0~Qb{=lQ8+3%ZW
zYHb<1EM8|QE8@rP`oH+5Buqcrq@5ehSw_n?kxw(E%|!DK3~x1w=}bdCk>7o5;w*@w
zQOUtbgffu^A?Fd$8G?f4fD4Bx$>B}e;$o;Quvs1S)i=w4-BWK}G)j1mMANSV;sU*U
zA*T>T0!n!rPeS%OAdqZu<bEE=a8__J-3Hs*fC*DGCpE5=S%B&WuO`Hz3d^@pKNB=j
zDY$pV&Wl*i+-})wC!JLOZWmMD-|<X@k=+QDOsy7S==B0mQd*8`Np8Gz-d2~Rp6p^>
z8kD0a{xo&7dQ$6Z*)MFi=m+iY%U`j7;vy!msE<x_TI2`5&RxC0A`Yk+1Rj7v|4W5u
z$Sc2K=D`J#wv&ADrthndffphH)P(@bAb9O287^F)UB(8f&J{m;3UViytAH?=?%0sG
z{^0XV24%SmY!Og-202mxo&$>uoSn-6#>(U{IApnaEj#P@2lBVeHPcvVv-HS5H61Uw
z!%au&9>tH^!atrQ`7j->QKez84K%f|id|p7`XFX{n>xOK;!V;W-si2HWd#!q<)=AI
zzQ}3ciMU!2fGd3Dsv<yAx!v06tUY|i?L{Q|$}C4_V<^?*>$s*bOh!qUwJom4))ome
zXL31xJF7&6)to<9@P+gAy_*ZpKIxt^UzsMZPTs7&FFa6e)cwsPtx;%3G|X<rrDZ$?
zBXD+c{ghrr?A>+sR?hJqH`P4Z1D6jX#{v+_u@&MauG}yq%MVh&y#)f=J0P323rhlp
zn@E@u@a{jTF!SxHes@r-jQ)1u%-y;pBmp+L)u`RTK^~TgWm%DgY%0z6;}ah?PSfs?
zDfjxmeI7k}6Ti>Hym#w8WQEq`+uP?|Yvru{LnKjs?MT)sqO)75v1Bx>6Icl1w%_Le
zO*d#p8gU7(OY{uXPVD+6@JER|gHJz3PBzOWo&LhlNP(~ET{#=wERD<FM4j1Ozk8=d
zp5e_pqL1vX=ozVZ7g9N8iOKC|O-Yo#A8rsrZE@F`)1rs-iCX;4s`Y{!MR%{Xc_gvT
zcbMXBiT~yvJ+i5Rh3Fp0R6sg^R@OlNlzX<US+KYna1o0p#E65-!3Dbsa#0-+VI(j>
z3F2>_yNLuB96#to*7?B$x^_7+l#<sw)}vACMRN;%=<!qH$(m_>VzL4S%nht9de=`J
zts)3k6sjGXB@3Lo%9QOPyK}9Mw8wKPd6IcB<|U0V+pMGcvL2yymmyu`4pZj59j`pA
z^29J_&0AY~C!2pi48KJGNZ;<;NiUo)*3#dK5oQ;hPyMsF-={296d%@YE&RdYPM-R&
zY3|6cq`KWid^f{~;I@RBnUlivTHkBqso4lFvyY4tZ9EoUYW-PF{e#D<CN6xOG=2;v
zSPc#DkxiJpQ8a?=f53qr;n&NM{teN@KzH@3e6g4X%oE#)aGuW%wmLM+w<dCR#WFQ=
z4}^k(vSwD$Pa7OuxBIlu*W^)n=Q(lB(+UbU2Ei=CMhQ>j*6FhM?l&==*4xO`d*wB9
z*tB4<*vXiY^|r`&Wm2#PrsRZL<kbgp+IAXy>nlwhCPIvD_3BgUOB~~d(XVIx%$Fti
zaShj;YG0K31ZF(DxlVW35%p$`rE&4^A%^v_l15>jz*iB5)b)6c2A!IewSYP&R7BO?
zK|hXwpE}~NU$SVVGd)OjCe-Vty8T-EV=k?-9!aXz2(ii0Rg9}Lw(9DzO|{fkR-@NO
zxw<2_1U!j@*sO3-U^4!*z}e>EQ|JL{0#Z?Qno0)~!E9g(;_(Q7ynR|RFCEA38F&CN
zoWQuUh%QspwrBy;r6CUE?;*MSO^QB2Y;rHR20zOt(KidIrbai#;eKPsm6qi-9uKAw
zPInb+1kI-otaqI5Ncj3jT_aSW#$a|J&-K`{=!J>!re;<j(`c`>o4zhXat@KV<YK}G
zx2;`^y&tCx$h!(TCNw=QZBLF?m<wJ1h;NS0G7ZyzLLxL)*$l>8ohw6S2(z!LMYr{^
zeSIZHQ^2SunxnW*<@!OWk1O!r^BQ(1`L*aZ=>|csYIa@L?0uaHX3#-*C?H3`V!LcH
z1wb^^tD)(By3iuWoRNX_ADK_x%$3wIO9bJxLXQA7x9_8ICV%|Fsr&#+)#0E^j}5P{
z2j$Q9PhR2(;Ype8t#SOuxYXBo+S~gg2Pf~y9^-Hf&5I-!$G?Z1?dd#eOqdvbuqGMP
zX@$IA3{S71&H8KaTex1QOT9HrNA=Oz3mu!mQJ<t`BPST=!!+a?BXiASUE%#Lx!d}_
z+=-peYktmWA4NFZs98B&D}OWP{k*BCo79=PXC{Ve-r8|Z?;`ZsJ!NC`&z4VB2;e_w
zH(Z@i*P<J}DcH#cTJlAp%of4MiE@Qd0fypuV8+O<qn+;4dzsU4|5d;{rP76gOIgp%
zuVt`8@=w&h6@Vm3srtb40AezfM0nMD+TYV#R<|pjn;Ml7WyhXmP^e|5{<)|3l(REA
z_JNo)2LW|%?*hRsfjU~=!FO{<&T-ZIA*uV9&v9iRy1Pak>G4x+3FBHUX&>2AE-y^8
zQ|XH-y^R-sUqV^%JpS0sh=%gzf|~)7Wdzdze|Z@v2?03bXS#;|xibA@OSe@&%^X)v
zb^pY(m-DYnyidlfd3jZloNJ<d^&VZ}7A7Q)jqqKV0v#)Ww5_3xt9sg`9l6mYU$8(6
z0rxWzvI$V3ypSsA>cQlP#PpQQFx3J8YiKyu2EV35khjZMq&bD*IWUAM!%)O<w^~Ne
zU;x}adZ){G34M>Xug<@wm7_O}Qcy%m21g1DM_zP({P34qs=sATAMsVv*~@I>wVL{g
zPoF0ruevEBp+gF_++edhabLwFkK7t?s*pH)HEUC8Lb8px_WZT;I|(Ni!X++D*xV3y
zwD6r?;#+xr4cg#2T@eS-ruaA8l_3XjAWM^7S;iSE650}k;EZcl?|^d)%!?9eJ*zmE
zPsOPaPL$z4B?wF~3BH?=#AvFH?)-jZ)V>IU02&NIrb7fciQ!Do==-H5p|gIhLGOBh
z$p@R3S|ehVY;KS!`#?fn6_X16$>n3B>3JOkvt&`hGdf<{{hXoCJ&O7|sy!B4>YU!S
z;m@?6VAIX5W1N4%A(v{kLvJ_wc#Qj_cKB0kj#>k1qp?uVS`|O~<>W=HZSI65GNox`
ztK1E{?3AASR{cdWC#Omj`Yw^x_fEVruI<XGds51+&yJ11a6H2+{P*JZ<9A&L#4Rpx
zPSJ0+r|l$}(Ysf0H293Pai(NB-ifv``!md-RfuOD3D%Rzys1aiarDCKm=>^Wx(kzm
z#8b~eWP`>i^CzZuBQ#QwbqDZ~i!iO_j`|>^pt($^FGSojhg_a4L}mt3L+F#PUH&r}
z<fNJhMG+s105weY5%Wjf2M278@z$Gi8B6A`ocHHL%5;iUXsKdT*s2G6V!6=gdfd!r
z_FR+*<hfJXf0SaJ6t?)PSE<80^~tQ0URLPOWHm|?^4(4S?W-Z3+AoG#_#cV*xCTTu
z{?=Ah36?Fy3wXcE;@WHH{x&4^?l#3mCPPPuhdPg3C}+>s*Ity?spsFY&J-cV`^hy4
z?PP@isR>?^=Pf<S6Su4)^Eof;eL-sT>mv;Tv~gp;PwpSP-7ZmEq=VAYwz+@xj}~X~
zjZ9xWM~CT35CmNu)77<Q?#mw<S<@tufz*iJ`u{!&jB9uz<O2aFRhhvMbzq1E0WYZH
zMTh;jk%?%*3{kR3Ak8jI`KG5493NMB--E|c4sW~#*Cr(WBC<g(pi3Rc8aXVix>@k7
zIosz<UMr@^|I_{6?kLZB?puB5zFe2Lx{vy2eusNM_FC7xj$7b_Pw*6K!!g+%<Jfi5
zll=XMMzC)4BuP}Xc8Z{=vr51Z->m0du|B)s!&O-!Nb5rWSn&6d=wdO6*u!(zdHRR$
zvoF>kF5Z?ZYONEE46>CGDxTn7vEvWlkl6fSu=AA5`*Wh@x;Fj20FTz-S+vx-zlU^C
zqB=g!#nOMyF>lf!=9eOjGxZdhyM<2j*;&GXwRh#AXJi4ZzY>a^g^rMzbOXW(94MjN
zl;89(pCp1-;3sg{op+H#0%1~28H&4)S4-!_Tlh`X|4Y|M*ViT_OOlh=nF=IuJk1%I
zVJcx^vEeIosdCyn#izv;C~X({>No!Vs!uT9JG^<1gyN~J#axP-_Mahpb0)p;2>N?S
zk9<XDL(RS$k57GOY&sS+d#|lOg1TsP>4QCaZ2?RY+uLfIP1fQb&X{FXN-y>6;pd8a
z24KZ&%e&86e^!##WT8z_ws5;S-6D{olqUhL$fbAFLHc{UtV02nHV>?CW5_AwPca0r
z@(o!Hk?|k&hR92u3PS6hOF(8JM5H3PtZD-Z+}O4<xF{s$g%JUX6F0dKZSQ^fv&>Mr
z4)Zfx)?bf1!t4qt6Y1>f@QLc*?=l+r<a`@1h<f~LwU57tV9Fs|VI%jRu;<iA{1rJH
zT*#QP!o~m7-(K+gcb$e-jb=9CMbV8=q5AblMD8CmE(=1}Z@ZcPsR_ILdt}`suWxgf
z&N!1~r_}z#`D5>@06R|chi>KA_yoh1%;YP74+UU7BlS`8>l*Rej{}Y`(LEEHDn)C0
zS8o?2B;P(-8(n>J!JtX8D0<B&_+5IP@l-@Ea@9rhD3mb-5zoJ*EdgqF$76j8G8Uo6
zeN{yFLp5_49>y+-LtI?F@j!981#t&16}f9-aEBuA9YFs~(B|7iDiIN(wxRio#Y5y0
z)zJ4<7DCb^U&fE0%^1EFm5QBQFDtIiGd4`%u<9+lX&~Lqe$hi3I~5TnHA1JX6-*r}
z=tWSny+UuE+%ugh@3L~jqxYFhsFIw3K_RVW((t|b;DT@A58tMm=%W%_I4}YC7;cMr
z3K$V;{e0R-TJo<jOHWWU&zs7%KQ5dp(koq-x+thi2=|$Q&`|9cp$V;#Snagx$tZbS
zyIyCOy`K8^5pSG?VH_%#^!)X!s_WUE1ACay=LzaM8nQm;K82s3>DcR?z$j9u75RI*
zv3lhc9h$k7d@8-1QPGU;q{@e{e-D|;UXuRb)WTLVV<1!KpN`fsUJFkF;6JxVJ#G_#
zC+dYl$cF;**T~AseXSBiNoi36jMSU!44SCLn*XugaSM^h@nrDN*I=$AnsKq9V`5=m
zgQz*1at6cday;40)ZWq=m3XnqpjmK6OGn^D?!{{x2KRz>yRr%!BWts?F1Ju?S>I-A
zQQ|8N9{p@4+!Y_%C+O%>T=z)wHdooe7)jY92mj(j*#t-CoZcW+u4CFzR%TntlK>W5
ztu-e(cC$LSl9ea{W;Dm)hMp&{vux;mS_aKNc?;E3B>dX7ykUGn=Y^#1%-Ug7Hj}*k
zw(O}=><iwv<+B&l+X_c|gP2Z5`zmju3eFAgcK*cDYDRz14f?)T(0-$sjSjv(FyBbr
zm<C1265}v`O!09dStcfscHZGAIBs?{I-Wo|HU?zbrqHVX5Y#7vU-Q<GfDAZWh~@7f
z6z<?;4bz2bPes@UP(OempHU$x37J{~8V4YAbJup*Dp@l*%jAz$^tru@=_hbrmlGA#
zZ1vG%{T|8vKAG2qZ+zIUcPc`FRwgCdjatUi`{WF}w{rSR+frS4a!Twj%`?m@)Msyb
zwpqUYU3$|fbHhC8vbA*8FSUy{yMvo2B-_XzChLr=nYY5nwWvKc1hLrN%W#`U?OG=Q
zUuQ-}3o#FaO9(L|a7~b1Q~$GACw5gu;vfZ91XI5Slu+SKfFL1B15(~i{2aI6d8UHM
z%ij=;w}I>_aa;Twf`XI8DAxmlha=EX(2nJY**eoj_0;0M)e4sNAA7ltnTWP*Q)1e!
z74s|jG~0;hKA({dOOFhG`{=_%qXK`C`0AAD_(DmUo4lut>{aNDZJd>EEGt<4kt>`N
za@~u&*Rg$FMs4_XU%{6?J>vj{kAVV@oG!dDxcaV#@Qw9dr{n7gU%|3>)^D3~W4_gJ
zid&v>GxnD6@$>Se-IA-Sjg-;#Tp?*^*WH=^qE}U;SF<lBV>c4(5|OiWIs%t_b3Aq}
zV8J?g)a1Jq*Qeqm^2>F#8&5I0dz7aVH#PJEMeo@XHG!1H8NUq!`G$Z7GjSX3j)S8o
z3Y-H-*6G(`X9G~+cv5MfEPAEj0IcTu4*DZ9_d@7?SCWGGe24>M$FZXKRH^n=CYURi
z5Elt}7XZ-EB)TMnpc~}Q6rkZccy@ze=BW%nj%M$gq{*<Ktm_c=aZf6BVISZr6?!Ik
zem;?Rj6XkEO@U&{*K{W;n2u-u7JBg3+~Ttx|C6O|DBqf|a)th?0o~{YwUaBizsT}@
z;3~b+jK8B4n)-%O0k5;k%ji+WF#Pw>in4r?<y(o^qxWw}*`-eWsK|ZHEu<g9ov=F6
zeoT<jMvni!;IxqM^yIoi)n}dxKPMvI<H5s=gs|LSr@VVJmxV8P;1X5^%6VU@q-BSY
zrADS#r;WlM=6uL?jCsMc??4{U?J=GoeeS0Tu^lX-R9j$Vj^Dbu1|b9pi;Q-!A<qMZ
zktqU^ex$B`Wi^HW{&nVxfmRw=+k@C~^Aa8p8ig`E?2my9g{ZlJM2Pa^tYCL;vb1yl
z7hcxghxw_h8-H8dLor|AX9f@Nx7RBE-uhdd94#uxH=Gg^E_n5Ev9%vx`Ul+NrwL*4
zJ)@l@_s4j-9D<{9*`5Wf=bT>GD&)@!%^B`27+=-8L+?}&r*K<@=a1xYA=)9RS0;;p
zlMq(t7v^sx@yNSAHtSdEP=08<zm4PbHuR5~bbaTGWL>Yp7Lxug0})AiGWl4Jfr@cD
z-%p16)v!<7;k_kps29n-AIhzRGh#zuz5PA@Lge*Qe21R$9|^;sC+UTKC@pTT{XJ9@
z{maf-uD^#0^SvGkcp_|s9$|?6@DLif2rOhnW3-~_frryPNyOIB0{Lq0DH)tu0vRhv
zo@M9jiG2p=dj%CM^<N0W68ssFckWml5nGz&e(X%$comvta7hX4Wrrr`>@Cf`(7qgF
z73WB26HE@vn_7CCluJrt;G1-Z0)?dycj9_>zBJ~ZZ-&ThF*X`AGuCN--SwQ*!)MEV
zT3_2rT&s$jdgiRW^7IeIy!lx0gnOa3i5*ko^YW^FH}2^yqYRA(L6dWqFi+ALzi`Kf
zJ|HNp_?Oj*5&d)@s!LT2eeWsvGm1tZq*%LM;_x|(b7<zdp7C+`Yj{WWDoAw7b$FxQ
zf*KHThxvPP-^6;Vtv|)rk|Qi@f(xM|1vown!h^Ui_XhrC>MoqWmBZ#sL^5l*;85T-
zVmQ!C5@EHTZ3s(v?nuaAmU#Ub$Ghd``ZEGwyLVVY^mvkNJT1nSYQg^ded<RI8gFcr
zi$i6LZ=^qo-r`6WChqyJrka)v#6DL-x%)}=dxw4In-oY9$h=8uuq>0AC!w&U|8D;8
zAx}4wbL!x~Zbk3flXGppk7l3o^s&|JxjEB&OH3@+({-QRc1~f>&5M2#Pv>tLd^ng3
zs|$61Eu90YEAjDQ;|qzYqp5ix$~hWwJ)JA1PgQq2m(lF^9E9FHNY|p^kef_-f*b`w
zpVTF)cf}I`=K$L`wEnJ)*`O}FwE!KCniw*DSph0}1Q>vku?)rkge!tmsa&kMAE1x#
z>rd~;3^aP0kH%%lofIM{7;<D8-jGGl{Gm#0aQ1rKChw@XSfSn{cZvJqS_HeC=I~Q>
zsXn_1?fBGNGaILF==3$1X4t7d7|4fOB+^J$IMj`sY@<&eaw<Go{GD_AbGRqH3qCxQ
zOVy?Cbu&qrmy|Z}PCEO6$&)iCL-*;#o8p`QESRq-Ng2g$dKHX`gr#%H5SorW5kX6N
z=xgb!Xo?h$L)H7QvC2O`ea;OA|JbIyHoh=wl9uF@@4h`=QuPV_4p-tFH@%k3KK+Rk
zxW<K$^Z@rdU2rpxY0-!tBtkzgw-$qQM85*;FPVdMVaIeB7dkkhKaFC2cyHk#Vt`Lq
zX5JD~waMa}Xvi>l1(FhIGvyymZ(uZprE{M!+j^QABOV7RkQq6);P!^|2rxn@CO>_<
z9h{6~(()7!Bu*cF&}}E{2D;PRlCK0p=u@JkT-&0E$v;isIz-iByYW83aYfr@IlVUW
zPmG<J1RJm_hoVJ#4bvr#;>y~5_g>3P8NK80-<8)b4Zxb&cXu1!X5`j?$XItw&`ZmZ
z+N)7>C&57eLgeXN>C}{B%hvo&KQ>?GEfuI;(`<6~*BKWO#9!m6s(syv-!Ktm$(fTL
zYZy1o9*z3GBYQ?eD|$YZZY^}tA4vfeJf%CQm*2>K?{ZPI76-&{LZo-nTH}{YQjnZx
zIx!R?sNr0JBuhvpHLj${d^BFNq<t2#OM?;c_f?3K#E~1zcMbFq4xH~Fb-J`D2FMS@
z^!z8+k}H?_tT8C}$f<#wyZs+G*IFfBZ9BJ=T^hBVF9JUan9zT;So&Ck8u;8KTT3tS
zax*;1#-X{N`UU{P_^h*_j#SnBGOFl^G~HG3G^wWbAN$1K^)fwuBeYQ*w!H}?1m-h4
z)Kho$bD94zvNL?U%A4<w(&N9kAV3xz-v-vBC=P=w<O4zAflm?2c`ZhmaZ3d=h5gHS
zpWYXV!=P*-%J>2<ga!H>VxfPlee(d6{84&zA>9y(xg585&p7zqw~tL*oLt@Bfwo>)
zlEQ_Q*Dp+Cak{@~jpI$)V>-Trxaazg60J=&iVVS;raD@DreF_^+7t@`sWb?ZSA(vC
zL{=Cik-dHhetNO*QCf*=krv&JAOkHw(VWn@_<1M><se-ww+2zN8&5_s;~=P?KVn}Q
zB(?KN$b8g;P(SdUTY+5*va_It+dL1wD+~faaG&r)QrFmBQgS_qYy0hk<ZnuLkm&qY
zc`7e7&mFG|RgmfyUzxG7X^Av#Y_srQ315BP8I;QO!kK0v^XSCn@<rd3@_E7ecDcdh
zRT7azKzg-@O9_rs4>#n;M%uwC92e)iI<b4c@D8MdnKz)Wu)ypT1#zT#-1GoMO_NbQ
z1ld*4Nt>A=F6#rq9wU{D%_Nnn@2b1uFW58tG=R1(TwKE(OR7b+Q+cJLs$pE1*C?g!
zAGT{vJ<5L!v<@#P*)p@`l{~l74o;=g^z>wt+2?#gO3+l!hQZj-jNKZ-WEcSL9gJ20
z_+Gv*4lK%XIv5^?{Any%k*ui;1{h9emMYkk;2J)tFT>!aUxQHJUr01Nl4N3D6oV`s
z)bC0PxpE{S7Sdq;w+Nn&rbUT<nn=@oRN?@tDMTq6*TRpx2oo=l!Pm-Nv()sjeSb6u
z&tON}HBeTetJaqIl92YZFMdQMwU+vhh$7CTg-7ek?16?Xm}^1iC{OHos<nAGOS?jA
z=sLW8tAvYq6VO4oni~3>GH9x%M)J&esv;A&Afr^xK`&4qM*)`M_;D`jsF3uu!XpqP
z&(gXuTI{+80|;mm=K%>sKuioufPD;dtswTa=rlhP)e9nYz;|G|!DfLfDT4u2<|^T&
zv$vVy-`K|uzKPSOgmH4PTx4+YqR8$w%tk(VDzZdm7mRjrIKs%hfHX3&`ZqzNmkK`W
z2IyX*1dyCv^F{bSz~Kc$B#>gv6&4-_!r?*H0|_FaI0PRN6dBJV+9B!&8ScyR_t1ps
z&I$N|NU=ddL?+pJ8du!HMdV^(60FxFf!gUXF7y49Q(UlEOg)1{O?Ba{WkjOlcgw(|
z4yj}Rl3^Sv@Ddw{Qx$mu{|jwWO~3hHJdKS0+%L}p!4#ELj?K42Fh+-UqQ;%Jfg?9L
znN=xAUMjY?uT*5IByHKbQAa;L<6>>j^S3R{g%|uUz+jx8JG=|9dqQr+z%y~|*9A&I
z0d7&WQ0W8%HuX9S3=g4xx>@U3E@N*(cEo`?Dni{ra3P;X|Ja7^Jh(-)KyXnFrXv2D
zFd8tzZNh6QGJsb^k7#rK^<cBF=~hB|4?B$OL{cDickA7+q(=Rldb4(f@5NE`4+WPh
z7ut9QUUzig@{2dC%Is4$OGsx+c206qniirAqd7@W%V<%zExK~8Rn648`ws8tky~lT
zx5<q)W04R39;&114~}3R8LFg^^5*^>XrS1m?UQ;0<Cc2&=If8y^OJq~f_~S_`Y98?
z`Zr0aznU{~jjzsorai+i#gOYe?<(?LhhlA)SL6c4;o*{pT9zwo{zBf{UR8wjlL4n*
z(AqRSqU}7Y{!2#DnM0?uw(9SpH`bQ>cAf6wy~Qj-6WII2)_xp#ZA`CPca*QfOwnh3
zFy}0BDue}GD(wb{AwC1m+shD?4daZFO9URzjygFGh&DbQS8xqMKjYC#1wpis5NWW5
z2e3_am&yIk!1vmo>3;^Uq*V#Nztc~-Qc~|huDZ#0e)^fr!$&(*l6A>5UOquY4IPJN
z0N(yo3B9itcu9A%SbI8aB`5jGcU)Tl-(m&&)!DjjHB@NJW%A+DwurNwB?F<!saFK@
zc=|%i)pI}39hUl_MJ*VTbZ(;s6(XpH%@l4XF*|OZ#<IlYP$uc0O)_%0OMH#5h!gB-
zGfF017)sWH&AxB(WHb3DWk~k~q)({im95uN_M%&j|G|m<<as7~IQ2qivr_H*BG%A;
z?~l)hUyxRisdzCEV3lH&&<G~)2xN^678Q3Cl*{+UX37dZG+)=IgjE;*4`kMHOk}lc
zc@$ng=u@KJ6e?MR`QO<O@6X&<aH=%{Nj};?2YWZ7LF86=^z<D+^&?%(mRH>un7S+i
zECQ`cnyO2lA7xtpl-SPqYF2RQfjk-(NiIRDoo(<vdz241Uj<Ce(}8EE@`~B>wDdx0
zIjMg?nREM_<HnWjxPTqHJuP!Rsd{eWpJmR%*yAnyFH{CS%$GkcI*ko7^0^eV8%>1O
z#-L_Xw<%RSkLJG6>(t^XGNby~kF}aQIy`Q+KiRp)p&cmhcx7g=V*qpAM}^O+I`~UD
zs%iOp&h?LGUrb+AUthbvqq2KR7Ic+1zq%PB|Dn=m?~yqHG!@#Pm57}X1Y1o@5C~Y|
zs*p~E+}w%WCs~D8Kjp?0mlRU92+Cg6hnQUs7*tV?7{0)8>V8QJ2H7lGhy?OSz;sd8
zo4>>N7SqC82yToN|NKD5a-AJ>A=aXQJPdEj^L1QdroZRf_Zbz>haH<H+VHMpIqn<%
znz8=w<$qhl!C`{g;pPnkPYZu<ZY)aJ%bq+tFgu=9aSU}^{&hsy7fIT?lfSu{)BR)>
z?HZfezXg+DV%(RmMP58YGfzH$fu2%|6Q!p3=CZN3o@^1#{^v;1V-I57AczB%DX)V2
z<^8fChPf%ZtU7IO{Fqcq%`(L!Au5*u4O}9d@tXCYkhZYVJ(dV6HBS488~g}VW&m!@
z+=?G?27;`0snA5YoHfhgsPw{GDA|^!Tns!ZNMl!d)<btR6&v`b5E;scgk{1_XKlv<
z6KgU6?*<hVWa#dzg9W(-kzndhARPk4nzoE50woa0dlZ>n4r+XG64I`bOl9iWEJFRx
zVLEi4ZRj5n5zBPDAx-bpoRE~v=U1ac{t@n+q2#%z7Jg+?Z>-(6xvmuQRrq!wqa}E;
zA;8Ut(SA7H(QU+gn@Cf$8rLM>N0EYytz)Vn4yaI$eT?7k?=*5D?U}NzXP{i1uwV6>
zm(ExHL;mQTwnv&d8gwNp<HHtP$|t&@JlCDdBd_vZ+I(5Pmu&Lp@4xq$AS4y=?d`UO
zjkINsMyWq0dm1hce=0hXdZmO<vgk6e0gPxtq6iQrD_GNhVnNsdMzSi8R;{ZaVDsG;
z<>5vG{0!$U6I_!vCQ-W9MNv0LU={#<MXC-UEI*9#RG3emVb<nQbwNnQ6FbDNcSnF-
zM8O_ViT{l{!k~0?IAG)|sGXtA0ej~z<yrw>;7&~{-Vsxl{Cdu4at5W-*EF;&Ok%Pq
z977A-o{)3cL*GvF9>(D6pXTJdrbc?bOy!EIy2KXwuzt?_SeE=!nt2;>p=Ksei>a-R
z{Gu+FL3+e9+x!>Zc_n$~X_A+_Sn65H7fSrKxuqz{u-L7bu=_0~Ihdo1K1r`S&d<nq
z^E_nFc`t>2_Rq3bkg}(X%ISf1&PmASHD3CAC~I2m##yPtp@`6t!wG6OjyiMA!n4%e
z@r!(a59wE_>zL^MIFcl*%6XabWR^{xbxv?dXRXN!mg>NV_cR1F{-p#YPD1p^4K+ma
zi(KmE{BV^gX9zQwx@q4IRJ#;(;m3!cFtno(tw~+|j-me;-oJPimfI^T-9gBm0xdCE
zLD8=HPy&Gn1B!Pc0g!Bh!rBOCO$ohRq|(%gxL6z^^@*MGePC3Raiyy!%9+W6S*JzX
z-#irG!gkSj<5&mI!tS}>+>NhKrBw4$8$63{#7>Gw)_ZbSowx1fbfRlyE7nXc@w+Wl
zn&a=JM5t>WW$2&xsT-n{aBAH*>oCq4$RE#mVT2y<Z(3Gx&Tz3U7Ubz9i9AzJS<JpC
zdonn28(a2sjRbTReA~Ei{q;xFe^$Y{3kJxO2u&PR)E!%SHhbpZ2dUa8JtMLz9Q;!>
z@2RCLdps3z;*zeTM#`(sbSe;00O;A-H<|$LIL*+=4`tbdR&h)1Ie(Oc<OD<U-TM{P
zXVS`zg>pIy8+@OJYdf_R#(X`?Y1US<EdCKM)ungGhs)97`l$<9n=~Klz?B=}5%c1P
zU6(W++D&rGSxB%q$LO5F4W<WIH5~}^RNCHSDwOz&D7=5Q1oe`$XDUqr|C3sgj7LwH
zB}#8fnyNJ#Yy@2xo7eb0!QU|YVxd!Ut}_M;f;q#<OC^Eq+C=vZH459zw4CC~TSr!9
zn2G0f19*F_C$_{QDc1BQ|L}_FOG(UkFxu1NQR$2!<A~vZ*B`>6KzRmR1ygVgszkWl
z477lHKESgTAc6ANurI4(6T9fdNILIlLsYu;yqg)DjIwrs$GY`|(M~(3I*mBdgR^Z_
zI;_|7U-}I;e!^gxyHlp8$26z$kq;f0G6*qAsMZhYaouxYA3k|sw{@XT)=u&lPd}Za
zOLo@l#E1<_*ZHYW#>#4ayT42OR%1It$=5!cXumrVPuYJu6Xn88;xBb-ot}TZy!a-!
zv@p=vu;QCZtNZe2ub2^I?=Qh=gFe}~wpkiwN%nl<j{B6Pk6_A@!oFMeDp(MZmC$K9
znMzopbb6ybN~Jv7&3?h`<qOK@a53x--#uG)?P&(xnNgjuk50lylC_dKx`bgKxI@eP
z$znFM{4+oJ7GSy#%oAI(gt;{eFOT0%O!f>?A71DH4*wuZa8;86g1(tSt_7+->faN)
z425SViQALP2k|X}(4IsQ5t9YvAtA2+(KuNbnB6Yx^1IS_3RF3zaQ<wt*X#)eSAIOR
z(Bj4wfr<;Cg9G_?rPpku>x;}8F4f^Z@*blP;lt!rFS+oty<}e6{XK5=vGFAbabwfu
zg{vd``|;{|15|7B>lcL+2e&C+pZ0v8G)FK{Ki-vNFYC-3TmR#n{}-o8dJna+fKqt^
z4}<DWtoyvpGBfYd`CU0HBRnhX?8Db3;mrb1WzVXS6D|C7GHki8oeJVz+01SAw)ZUO
zytcfoV8dJGw=BOB@Me_#-b=bvUg<sVq;Oftp9EhqB~6R?0tDyKht(j>i@|}doeDh7
zi9p0stg6OUvp4@%bEuq$>_{WXfX-Zb0>X_&BRd%!==ehU9R7jaO!G9o#Q6$i5IG{!
zFW@h65%-a_1bDz}@Lyq3Lf2yITfKW6GWw2fM_SpL^P7f`ORe}mSyig1=9FY0TyqN2
z{b*n~quG9j{m;Aifjf{=ucya;ub^acyPdYLI=14t|19>+9VVH6jB~c<PK5u&gDdsU
z=-w$CX1udK6@3gkHd#pag{WVx*FdQD+{FmG;MR=dLke6KUo^SczE*_4&-i@Pc-4&i
z?;(tS!-mgz?auMR)$zCQ9rsSXX&*6af90jroaWH`vVPU<><q~>oiX^@Ns_R#<;JTI
zv@+}EIQ3yx&irucxw;KTn5@Oi7<B3XSUS(Jrn0sRJBrc`5Q?CL-UXzIKmZA$mq<~R
zDn$XMC?e7ss`TEg1VR}Vr6^75RZ$3_0)nDO1Vozj=Jy=mZ+^@*Gh8zhPR==d@Aa&8
zFSY9dNw>fLHO<c;Az7Q?0p|Xh;UtlUm*@^#(VapH9VMVf7l}CjToyB-LO+2?_c=7;
z=1SsIa(O`qfbB~NKw008;Ys)i9=UXJ|80$crIzptUiHZM942~6xj;gd?pNd3De#(o
zmOnPQaQ#%xIez1TpC2Dys%SgwcKa}gG|Z-L<Ot&QI0sdm@+|rN+m5@V5_dMV+Mj2{
z-?O>cKE)+cdxs)ikj-I#!8^s1_;AL~C`vgn*Xdb{p3A<#zBZ4^OT~!C@2Rh37ANR^
zchgoY^LYC`C09k3@6Vyl(1twT@kzIyA35IHKC?3SrOgyYo2VaftZ`>2yv^*M6uc?z
zSWm@XP`+oj@%!>yrRVmRKEvHwP1i);Wr&Y;vk5arGr8u}6;dWZpLHI;QmFSmIU35;
zXp;UpO#R`jZCVw|NG?Tlk{~e97lfky3i$foh$gcX>p94C--ZMrbMR5SB+y&HN2V4^
zJ7NHKvGkV;Y^*y#%|kNpXsZul#m_4BU;V6EqaOtfNc3y5O*PPf?KHKTg%ZE8sqF~v
z(q5?clfzKswKP);6<vP?Wu%u(IJ%2SANlN2RHowR6GUs4b&jjG*X&xb_tlURqZ9Q4
zHnXGC+P9vg9*w=e!JOuzxWUxc<UYeg!F;n^<K)O^PPS6L8&Ug4@hv4FX#^c{o7o37
zs)#XE)7UQA?{VYOJCTf9rI&Br{c~t~M=}fF+O6}nfA}8%LXlt)BO68F_NTStXvqZP
zR8O8QgxmaF<?!v>+de@D+|W^|aTMZdKplXyKUAjQW%GiE41njl5OGSklceL&qgd(>
zf%g~Wqqw9uNT1f>*mbt@F)$P(I2b4#Wmcnu%XF{b<bTjqIQr3Ol_|hb6R2czAMs2^
z!F^C|C9nO-L#h5hCk@Hnf^TE`{G!i>dOF*~FUZt<)NWy1^EI~!Jzb7IdA4JNA(nC{
zVNGLVmDAGwEOxCg!SWJ@D(CmmGu5J!(YIIbi5*`5H|BMX=Ipga`}mt}X-idBb522>
zn7oWU?+4ofj3fRuH`rIk6SzDkX<y&Y;Lv>I)1BobHqsDTf4fonzQDL!@cZN+M<bhM
zcaQy;(QWI$%%0Yk-w@^4%>MS>P^{I$AU|8^_n4xt+((h8GiaN?)V$sBciYXA2%KLD
z$dh|=d)cN-{$v2%Z&aYxa#cc+<NFNeA%rTHZ}NS}?*zr~Q!}b^)-j9;03dq-BfwMK
zn+M%dl?LffM8M6=;YX#r-e&dW$Z8uwi^Oab?6#1_wcwUHbY37b15P?g7@99#kp2ZY
zKGG#E->ll>-kFdb%d5Hj>*1Jg0Zpo0-F)7qo}02x#+<^D8ZK)A#RQM9!v%G2Gbcjd
zeib<>!y)dx8*}bfz@hv>2m5+C$HS~yT5oUxGt8%aSol{+Z^jq+bjI{)-O@+&%I2iG
z+m*ubbth}ig^%am_V^K8_%7W^$`oz%wxBgOR%7FFoS~%Zt*{2mFIy~e#geY;_dWRu
z-^Vr-gVsfG^wWlE<b+QZl?Z217fC&ZFoOjNjHTYCfUCujp!fybL^q@(lAn9;-y(!n
zy!)Q01o&scxfk-l9h~W-{ADhKJcY<t-34lctsN0RTeD~gk!R3~)-j_GIg5@*c}k55
zWW1#PSbpb&hQgKnv`aN_mH)Z>J@aeWB3H~J#1_?CDz5=)q@Q2B-(P^!?(m(M`NEjb
zd5+@?=Ql75MIzBF%)RC!_+dmk%re*-17YasGJwspO#Ofl@t9?q_lG}023zQbGF=g=
z8E%o6g<xgcN4%v}th_(w2$<jv(4V!D#u`yy`L^OkJt}R0pB+$v&=f)qVnD=)qsT-7
zVI=IuNI$UN$t9>piQK#$Tui*=hnIBcSw1TJ+^1sMIlevrlQy%ejoIhRCX`jFvWHLe
z;{!t92kd!PJ^rp!BDAd%()YM`wXi#%psc@)I<#8j<ewKGpiwE1jTcW;KPJ(%9rE!D
z>1We$dLlsvz~2x>1V(2VO@Eaxq(=S{vozp7Lw52l49Rc=dW+2s<dWvT;OlK{NzeQ?
zedr-cq|%?GSi$*h-3x{8TfW}A`C$R|Br0ZC_;*P2MawzEATmM^7T_F=Ti2RO6eaKX
z&^#R%YAYkFb>a9u<!JQIUcy#&9PzajV$N4#Iy=tzcz(>d&-DFte(k~y>n~<cn=W(A
zk2o$|JdyP*FzTx(odM;%YVy4JsW2_#)5UH|znkisomcp_d2>yAz|^T%;^X;}W4&!v
zhH|t#kDmyBetYjgr#8Hxty?idBAvUhL>3#B=<!W(c(>(|LEDoW##bz9Rh@0+%`h!}
zFs=`^OtXb1u#XpcD4iKMl#3DzT0{49p$$5GQ@AqPr^T9>e}A}(bjsu>@j(SCAC0z!
z5Xwg<ZM^s^mL?8h0WlD6JZ4j;%@4BHjf-DO`pa4%GJ--LSYQE}&J_rk^1)drT12OK
zZVztnRXf-qWR<{%*rRogaWV4(?$oDmk(ZlepPZ1}l{d6w7*M1ps)p#41{aSr(gxhB
z_hOzH+fhV)7@-OfYQwd?Jtuznn!5>07yI}?@kg2G{c`-%7XDl<v!@GoQHNQ&%xbut
z-gk&rAK}vDs<7U2%E{1R<=6`t&wsEeY<0)6cr5Bvj$ML^o$6U%#=D`*25IYq6C*qi
zzkl9Pw4{Bonl(ufNxpw{d1$#1GyoJWZdT(5_I~53l*byUPP|`x19xq(Ap8k>tqPbR
zln3F8^~13PKR2I{d8qi{m&V^P(ee*Eklc1&HY+&>L>pKILA>q<0?+5Z*2|Nd!-EHP
z`(`2I2MK%-DIM(`m37JaAhgA?+r++O+y|WNTa+(E)A!UvkK1?r700z&LhnfS9RCti
zj}{#yi#`#ovINS#5hl3r@#kHxQ6r0hhBfOw^=)4Jg2AJim!WUWe(8}F_yo1M^M?cF
zg+AXn&*eLMcK=C^qD|7V1P|TGLeuS`-!5BgCMp7_B;jVy0HlnUPRBp`&Q%9CcrR#Q
ztrc<}-w67Nb;Yz^S3tz^RXg=la|LILUIEQyZUge>dxOM;4+kl&!+lOOgcI8I@}G9$
z&NLdQp?_57;pu-SDbyWPR3(ftyz`Ly+U%VRCV6)og<K}h=8m<5vXZ@DN$Ne0kn~_J
z9UFK-qH$Ge?9)gO)+tjg{qPcPvhY#Y`SuQ1aPpd#PJdC!f#%w==A8Ox+>W1*>Ar|3
zdDVTnQ}z*(zYN#EBs-32DCWt%bnDJSX(#UOwbmJlLF^b4xoqo5PK&Dk!d4PGGqjAp
ze;#wB&6v74k=dUCjaZw3LW4Z8524ntfBqG#T2vtOY*0Kix}AsWuMo=tNs(KCRtaP!
ztp;VVlIG#g{oQew0KjwNUxeDUk&p{NC2l(RXUd4@e+7*1yYW_!$$AIzN^F-t8VyL%
z+^`xa3X4n{8Q~U8azAalTiH?;Sg+qJP5bwTJ!d<LSDgR$3D)cA*4wN+x6?<=WN~^d
zR{V)woliJI+b>bYTH&@DMV}l_m+y*bATLThUGvIaMK=q~n{6v@=fzjlJK)E5xE|@L
zEc+-i7|&SRe$pZN%o2B17+16JaA=)e+#9dzvTL;IawQuSztCE#m{J!DEbMJORoE3U
z|LI64euX138{3<|f)X4Zv_FL|L-8qreuGR(VBd@69<*So4JvZtqn*yRVbZxEN&;>I
zZIwK1>tK+z%Lny&t!bcjgB7rC6e#!Q(CIc52IYo7a4zr0SuaT+E%Q2gr%ab$fKu^Z
zBxOx%(BYrj;mJqWv`UR}15I|N!<meiQn6<Kb}1P)|IJ1B;`H1<eAWKEDiI`mcKEd0
znM#j}E2-@G0@PIt?bK^#0=I*G=6FP^xwR^$*Ci6DBX$aPow}U)HdXC#q+7mgE}KJ1
znL$a_vjl2#(J?=fObw7k;rK+7EA!q}TPI1MCrLd;!2UA7KSO~sqC&&I%wYPZ0_$G{
z6|6q95XF`@+S50sg~BOYuB5Z+C_Yr>1!4mXodD8FWeJcMJ*fQV{zraree09?ZS?b7
ziqamDOQx3!LUrC%b=rs9Z{M=%Y2n0A6uUhEp;<Ypz(l*R%){rbf*f}Oxl)&;WcBAz
z$mbc)nzxS1!z*M0%PyU51SU9|pn`SF@8bMQIP9yl=A7#1#793%A{zn_qR4q1vN*7%
z*(wHBH;s`vHrJ#R(4OB6_iD?*K&V4xiedn+5_p(75ukyt@dIR9H3yY{rs-g^nnoCl
z98o3EJ>E~@TmL(sOaf9SB?7Mved-C85@dN``tSlTi~oH_jx@!|7e?yw0YMDTPaWH>
zcz!)L7n1LO`?ZnPvg*jbBPBTL{LL(3Ov#<u%f$lj_YY3&n><brK)(eV3#4WsRLAUU
zqN~b@G!Ld>4~_wMMyASF;4kHZJS({F{PdB(z^pMoFzGrW^_)OhFjb(7MHmFv^olLG
z(n&$kSOAY2<S^FZ`2-U?6onZeHHN7)Kun5DMd6T<TDUM^`=D-;M#O5UN<q#jaMn2N
z)*0TUo+AH?Ky7Sy8a!<J>bU!_dL1{BZT6rhzF)^E&esjRCYYsWzU&lG?<bL2orouj
zH1{n29CF|J_PK0F%p@oasQsyO{6MBlfQ+pHEM7U3^UIj<zPguPUVXs$7YTot1tV8K
zIWPu~Cu?=cOp~^e<Xo7nnhD&R(2K&rp+z=$hiqyf&jW*9a4V8&N07gtCqa21)Vsog
z({OFNCq2wx$hjCHlKIgNp^PxSNDKvw?tzY4k#S|$5WyFO;txY=$`VHxz4M+cCr-SK
zH~4z>bA4m(G~aUG0$=G*9a8vRLE-1hlKSY2xw2Nmk&*#4L|^rn)UA7|2hMABvO%!s
zK)Lu<@T*2~=MgORgD2FDfRykpX-mR2ILB$u`!mE)bWRuIr~fj4tk-)xkx(&2M5Ckm
z!4RdveK@^ouq$E?tO0<_L5#lybw*^*&2t`DvC_IGw!}e)fF$93I*D&O)rEQs5zDwh
zdC(Db`-j#^Cn*;T2YQTC;*F!!KVf-e2K{AC+KuFio~na%2Ww%84pU{`I&A~8V`drA
zjz(STd(vwOQ@6|(&aZKF>g%MW4dfHd4V>x#oRB({)Xd|+FNOLEh^&iZ`p`H=HlQHa
z6WCIS8-L`d-Zz8qCn&wb41lnJtSUr3hLPg|04P3y6el2noVj3b0CImFX_obE^q_+j
z!Tki%s^wgbU85mD6gJ7=ojUJ}iZ?07Ss7HsqOHz-oZ8VgJJC&Sk|w5a1rxa~F$&*}
zYRZY}9m}Bg4yFj<41hwxXV@&=`XV1O%tN*ieUT)f+U4U=?Te~h&{oY+(}nU&rd*i2
zo43*w+_K<87>6HieMJ&{f!o6y1{?_3M(051zzt0VUL;%vFRk|g_JI9k@0ZU}fyY9=
zC#^9ob}!5IZy*x89m70n4zO8j1CYc8{<;o%Er@xg*cn>0p#RZdwt5!YYtiR7ol-T_
zED=P7b`mm6yz=@2q`ItSX#~E20cxQS479fQwT2-0=s?381>Y!*CiVLqSxsPE!2ew=
zL`Gfd;N-)LoTi9(RlN3lukyELk^Rs2s!D;C9Ow#(F_BEXaU5lLZjfK*9P^sU(3)!y
zP-c!|dUVEVq8nsZ-o`@p$<HfV%xh25UgKBi*2P(EsfWV6RC5F&E(e9V4H9Ii+`wzC
z!iW@VH^T2qub7jh(gC?sE;zR?5!w*=kju%QJK}lcCB}|pG(sLn9Z#F~kNJC|Y+5?C
zd=;ZOjP#rwJ4z;@=Wpg$Bloj%Zgl71bzRx0kE|aZEOf(svbP+$C+kX$<z9U7zH64l
zA^&qy&e*E-@+G=<EV)L@!)3&$XO)lE>$ASf^l8-cg)c3skB50Cr3&#V$L0_2H9qgT
zp4=)N7Q6oH*07geIeMu)e3PV9+G|?S<kJ0`=y%w*AnP1jDTK3{F{i&JJ&kV=ve2tZ
zHe!6YZ-;{3xOHYWZOtQ^eCT>o4X2_-NX9{eIdsTki=5tPJ(*2Pp!=ck=g(xz1?^0t
zJDilh!Ye={gz7rH7Z_ybZMj}i5JS-h`nBd^IN$<r2a&VhXF)?X;?7VLHR7Wsf~nxl
z#VrK{qv)K#oV<Zp_>XEGuxv}G9xJc3N%1X=z<9yP^lUzEYfgju-@)&Vd|Kl)G^>g)
z=id)n2LHP@spy{_@Frz+F!-UwZ{k5(z{IGqgj-SF`7Lr|e@Ttcj7_*@$kY4(D3}E-
zO!<!Fy*Zs)&+)e4_yn%ZYpFCqLSD5e<QVfxd4c*!$)al3V+?0i->2XBTPo9F0SaFw
zP6`)Z>X>E5Q48!va&(_G`gBBO(qq0FKk9VH;`G3=qVDKl?lKBZpS{iyySU#yp(#8*
zb+hAfujxXYP2c+%TaJ*d+VvHIp_dTw<|}CnsN?9$j{)~Jl0JYN`x?Tv7K9RC*R%y-
z*|#80dkFF3wYGf45nyu7-ZNGbcry^^0h1Qo+5a}B0<o4|%>Rb?qDWT7zWh}v$ZPZU
ze(R5t!Om&S+A&fEJ`}Ce)_a<<E1C3Zbi;pSU8iL1AbyGkRk5@?Lgl$nqgPx_)iSi2
zUr@x&u<U;2{wcEOWX9u<1<Q-Gy>qVj$Co)CF6lnLseC0+HtaNM_S5FjrSWlp@&Viv
ziKJ`NIxl<1CqmZ$9BN-v|8uBF`}YB#`Tp$H-@7L_+Yk0x`uajXZ`J8@e?(!=(pv`5
zhYyvtR)4|F-L(nsv~%d)N$gP-+>N9B7)sKwJN0C<q;m&*xq4csX{ZqI2cSr`bH=gI
zk^SJ3)0%(eTq*+_#vO*I*0F5M;BbQ(s3B`>II?MD6bp7SelaH>9cQ(**2*Y$3a%q2
zazpE@#&@J+-XAqB0<0$#dl_)ep#f5W3f-?9q(8mFLFlv1!G<M#PRWt3la!Vbk*P}9
zBv=|Zk6FpHC{c->4O0Ivq@}1#8`Y>kc!nmklVHDd<jKTdLW+9{N{?FgBL)D)lN77@
zu8J(LrI*PQtv{pnHttTV&)*irTpC!wE<4RomL-!~bc@~d)%${HTK6M&<QMjD-YZDb
znUg(KS!eg>P)n01n2Fv{N!}HBI=;&uv+9@qA2d5m?F<QcLQ|0Z$D$tb!K>-UUqlH4
zrN6pLA8DF$+Vq;Dg;&?p76%e0+}zbJKi|GG0?#7eng=NZnFG0Znf=NlK_okB%LR@E
zWL4*XF9lCv$}4AMenS1W1o8$3WURrK?-t<DO|w@1<AvSGmNcVhGW5uMvSdT3xXVTq
zd20Lk)$S+%W-e>MYJDragZ++j>@Nl>un@AgjIg_44r5$OS-9%NTPaXu!|~+Z$fw?U
z*1&7*gl8i=!ftzWJ74Wu;+Q%OtvNy`SDk-~wlHWmJiq54oIUBcDA0FF>&cStu8GHc
z&lm=_4(7c5IYYnNQm{V9xt=>IL^%AaBwqW%SLVKYl6l|t&s_KXIh56R^Dxnhoo9_^
zd-Fc!XI&Q?ZqBe7^U8kxq12CYQ1|9M94H9~JdZp>*fq_^d|TcW%dSOr8s7R*Noc^~
zXKR?AZ{GuwF5i{ZhSRlRsp@;NzNCFiW?EH<{4G<9vcc5bp$1#zfnAtd&>OCTM*8=3
zCB5v2jEW&(P6QfU3SjNMKZgO2RHrzQ(?>WLOlm*$PZASRW?eS4eZ>H3J|PvWOAf>w
z=`2&4fgfIu3r7}32wULs!-(>w2lHmk;hYDxV+7$JuUabtipqk&geUY!I7u)sSUQIt
zzNvlLaozhyfGa9XOL$R%Hi=jBRlVZ+n<?C6k$dd>o(`eUla~qgch2mqvIm80Z@2|u
zT??rAJsTouCaLBLDS3*=6k2oyKk?13hTe6Eo3Y1i#V>IW=tf1>dTK}X4bV=?l-NXa
z`Z&@rG0b;xJ-Wx)fr3c{R>odXjaht;vD&-JPT(Iu83}Ack<QQglfu8K$`uU*N+qS3
zEcG=SektB+E9`9TDt<S1QYvpa%PFdEL1=k+OuM#;s5VTN5NfD?-1s~?=vUPv(|~lP
zvt2?-87J3X9<Ehp4=AKi0tAM93IB8@UGa@CYQ#o2GU>GKFt6&oCUa0Nw<ZJP2X})Y
z%z%_n#3JJ$P$?Y0RwRbnMgtxw<_8XiJBOu{iqiXm2jeF;m$`gs1;edo+O;|KBI2u~
z=1?IO1ws^DOpfh>$#9yWj?-MQDmeKalR(|9ed~;L=#{i3!xQA@)gCit1;{1cZ!iy_
zM5O1cRmfLde=3+7UCc5&;#KxaDs#QWi^h=mU#8M=*NNH^$FqhLRDQI33OT3aDolr_
z=yX)#EdHS!eXiYS5c;ZO{aLFi6?3TGCeG<iI@9_myBWb%LAK*~1t%6<$?IhnnzwG-
zVxR8WZ;<r^<O@D!7W>F`(^$#5R4;6dba2X8^Y`$?aity%im==E-~4z-;_iVMJr{A~
z^hJCMy+L_vc(V2SmAHWVM_gX(A4d)D%}93j+2SdVC2g?_u}K=f^50T8GPOwg%z)Hp
z(V;224<1<Bk?@~05pc9c8@ifU;7HdHvj`&X*ys?UB9N^gD2UUKXT!Et`zT=oKxn&i
zrNADp`*4^%`KvUG>&>)}dP;>P`nKYq<$VFnmH6ov>g+I38e8ZqEez11z#x=$h9(!h
za1pq^c*%&?z~B-1+~uqYk)`&oKbLTp^3rJ|SCNJuk+Rt?65%kua*o~1`}%y7Vu?fW
zfRS5}aVMWUCMIsuXuP0-T$i!nc$Q%@8TVa#*j2G1ccPgyH=eefF=FJS?&P4^<T>Gm
z+<Mur`G-0QJGbB7eVFm*(4h&ZcmBU7U8ksqFV@?LZ;;u^>DN{t_nw8#;i?{$NM?Yl
z*mPWk;}%UH4zY)V#;z8BKqaCjeV&MrG{8KgEl-kyd2JdN9SX;o1*GVs1L!c`rn9*}
z{v#4~1)+lGBZ@W!<b9X4%-9Pz6=OvbA8d>g*<AIwGAe-kxJXb&hlZ|!o)EPDzZzMY
zeh>{uiCNQ{cfSAHbi3M@ggShYJM3b@;OJ~D%AWsSNvLG<RAV3~&Gf@|D)&eHXaUj8
zJ0*r$MK^j_#U5|SoljA2<l6jTpnb_?lvR`Ss#o^*Wgi=2<}*?%epx{6J=3^cHU6qL
zqgzS1H5n5T%Q@yU7*6H;Qj$D|Wng1WSNPRFcWBe$;$RS$l>8N)Z$t@O`Gr?(3{}6d
zvDzOzf84RE$%t%%-`Mxig*pN6NB0T}UspB^hAvHN67@;FhY70m5)`YVI_XY?y}11P
zOV0G#P}N;N2Cul+`sbV*z*W$;!5bjApm7S_+l^Qv-~qh?v7En!>AdYGRI7T*H^Y`y
z<uxo*ilHV@%-aTXZvVlVm_{&6A&?80%|gxZU+7^3+RN??`_=f363=KTzrPecaj_rX
zpwOvf4QDmLxU9n_JM7#jNrLk7o^?tF5ncImwf5#$221H9qD$pFO~al~H{3bydBnve
zJx`c99aA4=D1R%;<+*F?ui5T{+;f%BG-p3OytBsG8~0J*t#ED0q?bHlbk-)bpfF6V
ziDl$-tAWNgW@qh&!$$)F_b!>>I@cw2JM;AD2(3#U!U3nVca8}*ocpM-_f=G?tK|p1
zWagVBzM;U`jiB<277M%++T^-7FW2lA)HUIHHApoz0;WiyKTvMWSmZ&K$pS$zdBL%r
zA1GgPUA`vk&@|Jm)-|UAR)H__?Re&%-04S6oXAl%9&S&K#Q<+Q=_4O>`T`0k(x)B&
z0!>VjBbo@1p4N`^6E}!j8wd+)25Li0bD0P_5GcG>vrYl;fxZoblDmd<`zW0OBf2dd
zc0m^!qylKRC2#{km;fF#3fDp>&EI19ds$J<?l<b<LFf9quSIJLN7;_jvoQr(v8mtX
zSNlv}cBHAy&$q1eEPV*TX4{3aFX&DhwnykhQTHWmUVF|sbj8B^j6sTWn$3W3yiN^|
zX070dhPYdj_uoa){pu^?&!`h<SiSmv!nOJL5c*zqE`!`mQK^SZcj2$AFrqn^RrUQ`
zC$S-DUF)hQo%@dY#x0}uADUP9qx~Z7t7-S`_$PmHcX#+F(X`BfGZr>V<wGyn??05N
zb(KcX8yjdWiu6N$BL~kZB;IXu3h>HoD?}G$#3qPH^psRRRsC<c+vHVt(D5|8fHf`O
z1WWmWDnf>bLx?u+Zj44)RVDx=IkkSaQ{lDnEZVVvmKgw*%EaVwk`Y_R?^3)qGg?=t
zR5;1;BYvoiK;@fstP~Nzb*N^G*haVaTBk9O_Iof1c1%@-V_!E^Ph7!uzY+Et)>=?`
zG!=f=Z$$dVq9khL=YXHOq3>^M?}0$Y>RH|U9VWDfE_4Ztr?k$pvslw>7JJH-czH%W
zAiX}KlREDz`{=2o8gpSR-)c-@CpKH#!=FWRTCjL5`~B#?<{fRxWz|22@?UeLJ*PX)
z?SqckCYM}zoS~L=r`klwHR=19PRhCEM5^;i!`n<_)-hX!vymq@{4t05)^yOly|iUW
zp!K@pPZ}J*tG6E2O>hV{slmYt?xtmQ=>P7woLRu)`@_H-%)KCl))_~OT*R|zK-a{R
z%SO8VWPGxAZcx%c+6a`&t54*}kvZ0;F5^dRFe;xzGBqx6H}}>G!iwT*uvNR9^5+n#
zq}TV<%-BjmQRPGVvRM(iNSDPgrYCL;mgQjvRFC}ki})_DWZgS^?uPK-OX2@^c(gz2
z>zugeE~lC*ssB{lc1IhNaOI1Nw588=!r}BOH_L_c#QZS|KV5SMuf7HK$(Q5!;N_x;
z(>e(r-8jp~#@}u@dk;$O6yL6Wx_RJE7ZBj)rdpgX>vi~e^N9OrF1JXg-$bEZNqMxx
zfAtNx>&x%+quh74nRjtN2kppfHo<G{;Sf2|MNVeias=zbAB~Hky@!D<a4k~W90YXp
zi^IXlaZH0V*k8RW2H0%OBxtvw^HJ3eEJtbqQ2(&k_-Bl}+A6>6Ujml-o-dr`;Mj%%
zfQy)gAd~49=gRMEy)FZ^*ZNuS;fn+aE_1X2B~Rrn{K*n|M|2u2JUWbJ+A@#U$Qtzs
zq*vUyB!9l`!=jUA%uToc^5n_~^-Xh^dM@>dKNUU0r93HlyjyUx@+6M~(`zv_itgHS
z-MFWtr?7_C>G?#w{maBDdVhzau`zv0R&`MK*saS1ytd~;`q^XjYE1dhniia>Jr2Cf
zbBY}{LN<;_bsgM|*R%Iwcajgpb*j|zzV|&bZqLwpv^{svTfir2^pyQr8(FricQs`<
z49};02FxX<>0t^IVKopbP#yHr#ZpZ>M%ZadUuFYk3#4j3H5igL2^5>m5uqLAF6=dq
z0lms46)6I<ppJermM@QujQ(+9g&9=7>A1_d;G;iVCAY38gP(Q#8whz$+rx=KHQU0+
zRCKqgKQf_mFlOM)Jj;@YrjS{*BQ@OIOWHKb+2wJR71~P^g0I-+Ec;u%CAy>d+r{2J
z4?8O=qEans`(NecFJ3*jp3WO(T9M3W&$0}gB;X3GK)We0GAeG_7UFF$ayN}`j<s@k
zt@+_1_<ZIoDUbSCrXN{>QLVweyY!#y^xLz<wx@MBK*fAdJN}R^{UvRQqy2w<4cp6`
zbRAh*QaOP6uLa#2>`6pgwcZjC6o<e?ib%m^?<P=!$&j-eg%WTrFFWO{sooIqQNs47
z>HDhc2gPov$_Mr)Z_I*m%&vbh>6ntZxb)wXLmw{WAMMWufy12~e-0fpxl_vOmlCU^
zELMi+V^eWGyh&z%c2^<(Up|?~{jc7Y7}RUJCM-mB6Vo9inMJ?a9MZY$?R#DR98Y2-
zqu%hp=aLu7qU{7mOqz)Ks)L>d@;_m@_NJ2{Q&C`d?#G)5lM-jJ+Kxgqa@uGWeW#%s
zuD6jY6$(Y(&*_S!D)FX4(|Uq9Ma`@a*=S<~x{PUhm(5AT=m$9Ta*13!ZIw4<nh+7E
zC%sirxOS%CR=6e@mgkRouGW|?T1lrYZX7M?;JG;#`N=VBB2xDveM7a{jLbK@lAEUP
z<v)jbH|L(KwY%Ss+_*lcre8c?RBPq&i|lTE1SQf~y<2*I>IoIc=Yt?@W<>72%!2A2
z2N#}yrzgxvH-0okJUo(wb3Yk(=jl`53xzL|s0Rnw;PhsM=>h^iN4Ejd4-aF)RDanm
zi4C;~3WU}Z3HH+<IG{=(Rw<HtLR!%8nW(D4H~ms~bs6izt~-x>qo)x{v9$RCl#78L
zD<St-%n)LjZFb>sh<vRzt^JZMXBK`gT)QI9`?nftSF|F4=1W%JBhz(!@J^T##z-4g
zGBVR|G3%f;T^mre@7gk5_wo7FC;om)<LKZV0~J(o+zqStr<?EMXqNT;WEj60cPq1Q
ztSL2{`EK(_l)Q3zQgRTab1{5PyVA=dU#e=Hm4kmmt!S3(9@d3(ODZO=ErnwstJ6aC
zw?IEh<RSJAM_J6kZ+2{No<Wxm6?4qAMFQ&fsWeS;g${LP;N4@clh6RAG)fv<1|9g|
zMtUnF{-&vaR!=05M@4G+nAcZ(+c`ODoblI?Xd7Nz4PN)oV{0N|$j_hG{~L<Ss!Ga!
z$-0TL;fR?QZBaiYQB?d>MTv8#WJaa(4uf}x-^V0-p|QJlMwfs2NZYT~Jw2P%s47yx
zI$S$erc%<*bv3eH=k_j*xczMs>`?WtG=hj!S5CR^M%U&jt@~<hIFPfI_9}5d%ec3R
zX~Zh7w7DV3|Kbq&jM7o_ef3O@`O6g~RGL=M;MoG|FzG*szM5&i<1syovOo1wYxzAu
zJ+NA{6WhbjhV+;@ASDY*8Cn0;pF_vfTl4M~dIJvyza<7B9IVTg#Q-=cUVZq#<?nzI
z2|iJ(2zTE6SKg!BVkyzhW9rH|h{p(wdbWRg?}1IqRt*^$L0xHHnB3rtQaL`q&1@M!
ziF083?1=qP8L`svGX7*AZJf2RF00l31D7~@yHVA5p;43SzLr6Sb@-Uckioj5e@YGO
z@Z((NQ?zsSOT&>dCLp@v$}f)XhlYg<72ztjxprDC^s1kDQU%F3Qa20e`?8%I4nvuZ
zE@RLO0VuJ7LB}@_w^09Za5F{hwbmeeP9AOlr5_;Z2*Z3ZTPjQoS)+4)=t43M-alYf
z$%67CIJ0q1?_hqK=qTGBX@l4F(Lj!<nAaMCwF~}oL+d(2h1wC;(T)D-f8&y}&u%7g
z2|JsQlYcy#-0&^)c8hwnms9&G8n^h_SijHFz(<u^aEd7){0{Hbr1{{HF`|l*<pk?9
z3vbii4PK8m7l}1`2~MTccm5p0#)@6<jPosX$dfb+Uz+>j{}o+t?G<uf_V~7=tH-O<
z$I}`{%}n{y86~ODBod@eUG^eTj#H$gwp}3y@-L1qt43o6cA20Af{>Zvjbn<C>9<fu
zcMdP}nBH$5PK0u!QPR(dz0-3WriJ(aZMu^J4iBfWz)ge@qkPMe0;dkmqZoHusBEX3
zeo3cB8G`SeQOb~wqwSNLOI>we5sS=h@}jVtMS*PBlcO9`dbAAn$8oxW@g5FuC*@G=
z-X-WL%WsaNKO8-$_up`PM=8BvbL<c{>a@Qi{DGfvIWqrZRtv3XoyJXaC~R;CA1+Ns
zWq3#bIaERHb(FsvC;R%^V7Z=|b-Kr(p~=bSlZ{yosSL$b>xTS4XAIU^ZW+_GaG^U@
zb5Z62M3^n8=?R2$n}bOA&V&jR=}1Pmt-2aVuxf}vw=t4VLst=r2W<GXlA*_Axz|u$
zG_4UpiI*TlH>oj9OG(m$v>y@){MZQ=|H31kL@#XjJtph~98v?3VtL5U!bDR-zP(WP
zIH}~7(AY071B_lnTJ=KTv;nt~lgloxWw8Vwi@vL^5!3ZdN*9Y>cCo#Y$W6f${Fk%;
z93pCdnu1NJkc-w={%^~!CRBwal;*r2J^HnPK9>LiJ|nQp2o6$Bv2nx(WTb|`&Hx_*
zGUEn6s0w5EgRnHv|JC}c0t(~TM*LIynZvnRE5dXkF$N9Rf~<Bz8N*Bv4Ahc>UrTNT
zvGr6HZ;E=+_N@;=g@kJSs&S|q*Tu<v)vIHU3Wd@+lHb&8qRbmdjvV<#qnCC@$pz}K
zjy_LdVo}XwVDW7_6K_rtzEm%&bd%b1R?lN!iJ0d@&vNt5o$8dtncnz~)$3%`v3nPN
z{waQ+92B4Zh$FHDts(4rTW;vODBoG1>+@AP2`C2*xq!xHd!`H?M?PZVRQyA&?x}^-
z9&BZ*IVS^M$RSC%>Uir}QZrTeLVU_vJH<oC$KxM~lnATr*qNl`zkba8Im8#Jzzhy0
ziOlqG<*>B4oQGK#Tl(L2Dt>+#MVjO76!6G*rq-KZ_3blGnhIsQOW(E08<4C;hDBQ6
zE-_gTkQvk0{gND%;KsjQX6R)@U1@jp%U^x_H`+MTn!}p;Z3I;O+EIra$`wZbenJJ^
zZy7lPKlGL0ARuW^bxZFUT@sq#`>Qo$Y|~<LzU*)EGOH@BV9JrtnO^1%T-Mar#`G<Y
zzPfkaeo-kR*`Zm$i9CO-?>Q6Ycf)w9MXDX&{HI<A#^{DK-VKUN*N)8?n`JrHYi19<
z$yf?2_lYu<4++PJkEiguo@P9Cf8p)jgRUD*$&Zc2eUc6AomQ_7JaQAfOr!f8ds!^8
z-htSCs$FKp-)i`{S3OH}F~2CTq}P)9Z03cmP_Ytb>A-56c1iBIaW~_uyH93?b1Nbe
zvPWomR-U+BojfOl*=%Icv2Z{!+(;r>o)N_J7xt*cv%fwaBPu3hMgG;WeI*2^tE#_C
z+qL>r><3N=3dS<p15scbfCiH7|Ax}Y834#K{zbcN1oj{(*mX`7BSf!ZxvmDJ>Tz&g
z?8heSfx9libv0uutzXIIVFKioZ{CIY;~IfVfXfD8M}h3+#Z-aGS8XlPJsI@9L@^AG
z2PR#z&X&>h-;qWWr=+r!{-d7N^8DiXbf<_dLCKb5G)}9k)doRlE|%~yQ`yNbHgLXh
zXJ@cdxb#BaQ;qtGlKOb*%&we5AX&X4U~Z>piu3%E@9gKVsW0Mhc-sGb8bA-NApujq
zstNmc3-+VMAG@2m1pOE;;avKQ=ML`d%g=<H2cd2gn&gVy70*5QpV#*Ds4z{cqmPSL
zNjy#<G@<DjKFf_Th4c_3HnF;O5udn%($<@7lTuc3!95K7v&6tu^5d3^c0(rP;C!qC
zxE7%X+C9kru6N<x>xVIt_6b|jX{2Ds1(|P;q@te)LMe~o3B;7~4NrkWGWHB+5)_&~
z*ZQhCqdF*@RCx+h{_mzl0h4yjG6eSdnEM515e%X-A4eNp?!d@Rf56`yzQwodBHfU}
z(NaW<b+wHIBZw)DN2(s#|C@BGO3hK-LV{^Ek02y>gK0kPoM`ceK0f&f9_j}Mlday)
z?O44x82fF^kH%lWbJ&(8!)WUq(^_;2zdVkkaduEa6$JH=UqlDI()pc&F^PF%Mcbjr
zmM<tAu*j)6xqntd5zCm!ko(a)_i$x){8Fu0QSrAZ#k-i-5$_rP%^E(var0o`{)EC*
z=;<6U7X7+x-3KmR%GI>>&JK5e66+U&PG=Gr3Aj-kmR1e7_0dQJEmsMzdo}luIBPL5
z&S}VJxOF}BeRmM5_Cqx3#U#$4@Y{i6n5N2U>{H>~o{STV4fAPU<1c-1IBU+}tG<uF
zpoA9(N47({>`n=coXX0KSSh~TLfv>|Qs;E)EkIsp5z&^=&4iTKRxvZ>j_ScW5x^!n
z`J4G|p=$KPX<w9=*N0SjgojPS|1Jhn0Y<jSLG5mk_YN>axi;IUN0cy>wBl3DC$=uY
zUdYD$ZE}n%eOrFC)+3Q_Isr1D*Zk{80%6MBP;CJ3TMOaMSA8nQzD#<#V-3XfPt6$g
zFY5xA3*+eLrvLc6S+(O>#}9tYXRXSBQqeO)K8issjAs%u7EGCI=Jd_#u5YPelio2+
zY;i3Mu^(02Qr!P}Q1XdHvpi>X8m*eTKbs}Zm|!COhG%Ea2y=fpc7i|0aczWC*2lIf
zxjLA(MEI-*Yt`CV^<?ma{JRuKSMd$mD08_pzEbonMHJPMBXaj@8dOJliiamJ$=n<@
z%&b=H(#qkzwe6!y3n5)>ZZLJllbEFkEmxFH0IH*TjIPd~ZKaY&89Ab4O;;oYV7Vq1
z1q#I4z!H4PDXBo0keCAaSkm(4K^$~HgHe0$=GI+=BOwQ?FMive_R(DXn$(K~j<5cM
z1w}mJHR}&^TQG$#<KONLkxzY!Py2bPBtpfzCsi9;&<umbcuTk99VgE6lJ%}pkAO4t
zCI_`?v&Uy7W>wb~u}r&c`c50=Izutm&jr}E1gkGpw|(AFH+xLC$0oI-I7^$9cJhjD
z33s&BJ%_#w>~kfFhRX*{ZqRgeGALZ=W$0I#vBlDN!Ott}Rb~flx)nOl7YGnd+w)7=
z={{#Usz!898mDV3dpYakPC&sI(kI0Qr)uPPum4&RrtnbKBqS-Dw?LuTPS|f={y-RZ
zO(3Jif;V~(6G__@rV5qo5!)3TBB_CeM|ymt%=#Bi7okxG7mNw69QXg=ITpEpEHb08
zBdVV7{fso8XQ}v<LR+zwHRhzF)zXQ_E;np;Y~6q0E}~zN#yl+gkAAyzhaBFp!G7au
z4L_w*-8$py>NOQ^H%DhdnG2uP@F%^Q5*i^kE;%8Q*LaO2Ufr^El=B+%v&BB%(@#+2
zR)uLd#lyzw37WMYy<beU2z0dWVR%9@T}HQJbTd6wT&9vhUo}QPNJTaNp;JNz|4D*>
z+cV=?hfCS8BTKMluzF23C-C#K@zIFPib1k0i*}vRv*yx>h#0e4SfrWx;d%NSgg}y7
z0^m0}@eLG?m%aeD^XFZ~>@4G9lSTXfQGWrSvY7YVr?9!cVRL7McIE8@oU4AUDfH4j
z8dO)64j1-@ZCg>bHQSuNh7cSWg+|ob-mOa~{&%b;JFxV+zHL4A$mea7bF-Q!8e|O@
z_&sLrZU@l>mb4jiN188{zMT?(svFT)uZa6;$S1*XWY?Bz^ahil%;5fhhl>yr#-USS
zqJb*rmqd#VJoz?eJ1T~DS21%`S>ImQKQ^4zz1=!sjzn0&2j*(9JZcU&Iw;mDnFN<U
zeDjC_pSe9S=^sKhl>O-lGKZk_ZKL~M&56dJ7@31`RVuna(cAoDnfh6Nes42n-k7nd
z3F$$am{R<#bB}NAw_>}{1GKml;ZUgsSDlG_fuecSF+yVb>G-7=Cr{&e_ze2aXy_-s
z;A&3C)+{l*(Pshpg``FIYy+*5I_!(B{+h#S^TF6;^JLCBuA)Y!58jR*C*5C3x__RT
z0KgvDci<Q%`q+;*u8gRJonV5lZwQ{axLj--q}QpYh!*NLo^cGH-2ufVd#|r=XCOjE
zTUT=|PN#U*pY_t;9VjryZbgW6ErN;}y<K?<%6p`>B4#zPg(dUgPdr610|eyGEuA-+
z`3;kj$-h2+nqAv?RAiFC(LvF#ELHuUH+Hwkz+w9JJ#MKlqA~Od<1r^Z9}B-7HSi8u
zcYir^hUHaX<e9Z=FYB~tuKy$|STjXD*K!_Of6AY*DB{CN^xJ-}%GxBBCV0&o(zceS
zY5g<U2xzODNLloALHFg$Q#5!4nQ4M-ORx;EUnt<0*}f35bqH^qjdM-ZkA|bl6W{^B
z`E7Nn)9&bX!#(qLB<5_{<T8>9TjhVU9J)9CZH)_DD4i{h47*vwEJ6w-G3}l;0Hg_U
ze6P>6J6|usbFyUp>OH#75r&V|i=H&w-10*Wv%Y@e?^u3|?BBraJoHesm^NqcG}5Xb
zo!PJIq8erCzc@AK@LYr5?BVz9HQv7@L+I-ClP~EHs`LEkS4BzMB!RkEmB(dQ{ChT!
zJLeY!#O&v^f!I}8Nn7?b#-Q9;3uhy*n*ZPZ^Y4lO55CO-R?zs|WIX;VSlHSPR?TB{
zimj9WDG)Qv#E$<MjvHG;6Crt^P?>2$*NDwECD;%`;~$kz8hoBH9+nWh*B1>|54YyF
z`fe)M{ldKcsF><sr>B!Xn9@{e_Lm4^#Cfv_bG$@X(y>2>9yU`;wvqU`e<}J49Q7S_
z-<i9^GL<EU7sAS0Hw2Dmk!|X3OhsetDP41e+$W@?5DE!+(EzWK1@UMQdZe;eifM-c
zBm^fBJQH9NfzAVKMwoOvNvx0oiq^qwn*z=yP&cQM;Is4_6Ziuvz$E#Ula6Nxb_0cm
zMUP8kMp~+A_ZwC@*fdaEEpzjA!|64;8!nB$5<*f2{26BkUOKp^HFM#}52gawo(axS
z#hkvGYDdK{UUZk3U|kdJ1%8_c3&o*syv-U1_my9s(m5S>J;sXbtGi8if^bgL(4fio
z_Prh2EzE12ANn=_s?M=<2XssYR}mU`y9srPDt#v$!jzLE_PEUczgv)>65S4ZEZ`;z
zF2EUiPCr^Oe8h;=pUDab4+|w#-pGSAec^W@?t{&?thQp%jtoVW3LNAW-jbqm(=E#O
zG+K{EdzM~Puj%O7c9%TAytUvbcl?sjw85W4B`2othp$En%a|?NsF=(*W`$-4tr?wh
zOgr^+qIpbx$4BU5<I`%zt0iU|LXi=!StpM?y?XhAW|y76!tdM7Zl`NGj_6&MVeQ~*
z@GjiG=kGlXogQ>{9iXAUr(%Y3)Fw)Q8a4$;-^m*4$I?k^@qcUTknbTQ<8Kkwpg#mj
zU&w#Z8L$&+#HNU_#0rFuXkGUg^B3nx!DAwC6VppL+KU`|^IHtG(7M=yLmE^+P3ISu
zLCQRzh5tzsrzC3UCqSlI{^sh3YR2L#PpipIyuA~3)-R7=TU5@A*EiP8_hox>&*-x1
zm3_^8{^^_cRG30z-?SPZa&Lph0cq$kn#(@sBHr+4OU?+hXmhln5A9h|g$t=-XJpj>
zH7k>CqJkcub`wZo!m^BBsnkdAzG>jOA!&4^p2TjsP8$raq9tST0Zj0|o{n4%V17%Y
znh%a0!E#|HDDh@fDQ~l)SI{J7bi4BZhT9OthZ;#B`^3kWOhS7wC5)OZT69MDzd1QF
zZ$bn=SB1W4lN%v1>Pebj0hg23pF>t$tI9V%1fA`Ba)yvsUftBE<cPvE2A{#{x<5x(
zs%Pyf11g74Ig|=<S&slFp%OXG6Z+w@=Kzfzx$0Y>jM8Nh=k!ejpRwpcZ!&!`gb$qF
z&lV_Dv_lFU1c?!0#CiZko<>=CGi%?ks4~opQ!~}c6pks#sCl1s*sSS#&xlP_txI=$
zks*h`?JMlNOeY-W#ySmdH>+v?7ei{!mXbl+kjF#bnO~Q%3D3JN89AmUovkRiz<dln
z_slgXsQt)TPx;OM>Au7_jSK;_T52}M{N6OYAa28akV6yh7^Dehx>%hnz4g96jH4mI
z{X|GAc3_SM$L6IY0@emI#7e@36ulKXzV(y074(5;-nm3jI7Tp0dCuUG6PXsN0+$}>
z-O||YLJ63b^k0wFA1AKtlHXrz$VlvVdBNk6)jdsccAR5;dQs&R3yt5%)k)|%cA2;p
zvTjye&6(=UPHEu!+25ott()iU@4z3!>}H#!8f%|Lb1RSidS2kyExDeUzT$;7sH;BU
zhg=)7_h4;O%auwi<pcWT?3(DKL6Pc#K^f}eX)K6*hBJQjlnp475g9Zn2MaF5dT6)e
zUosX+P;EH>lpXH3D1U$x%pBBNVdadL<1k%#M!XX`zp}}*@2L7B`h2V}Q&e*vN?P01
zS;+ZY`BMt-%%&{e)3euJWP2?qY~D@EmF~@*u4sbojNu&ZY6tX|8C3cI2ScB%ez0K!
z9ijgyW&-JsF!4V+BIwUb5UJPIpi0I&E(5y-)~{`mR5bAb?jFi-@bC$QkFKkt*uy)_
zydO^f#{A-8ul+)PbcK+$(REXyoi^ZQcX!b7ie2`DiUGIw1G6JXmapF#exr2D@qU|u
z@<)Hgu^5;Ac;Y=tL1VU0!ckwPck1*rqsvpDO=ZY;F0NT74VoEXb4F^l9}XXBdqrQA
z1@*m6u#!INM<{FF(D`?8P3-QVHr+EZf)nKO!P&f~<>Fj21IC2}v0{tMut0&;D-4X+
ze-8b_LFt8sk5<)SdL=9LoU7j;-y%U(Fs<zc^A{{Pvj){If6V>L`OBbL4XsIs)&;D_
z?SB=T!wdwJy_*ZEQwiJ@sogtlg5jO~JzPH|NWfdW#;Gz_@%<1m^i}zD=-tC6b3#Oa
zth2k^Zi%<5rcV#4&R;<_p}!iROQv}dQ^t3{-Tve1^Jl$66zxcd7P^5|=TxR5DK!_|
zO{5_cxWdo@1c6G@YLWxk_uF~G)bzg>E^P3o%9qEOFJr!LVQF5Z*a<=J@EL^66j<2B
zjQ;TuDu?LcCX3z{ppXuZ!&j(Ex1mad%|*#j{AJHfugrnOM15aB$>Gj0NdGLx5llK%
z00RVf7S6j4$Ipja-VV9?`aRaeNqs?%P3*-M4f~bmE`zm`l>ASZ)5>rQ&ybRnj7fR5
zgac|IdmCq*c)gT*SoJyGM2KO<es7eO`t#__ldQVQ9UaDBmHV#SgqdDvt5%hr>;U|w
zmTq+XT^v&(0EWT&@jtEz&K217OufR$L=67XH@p*UUbW>lVDj%=9hkGKt!)ve26VE*
z91$%`q&+>ZpRMS5TpO`PK}6s)9!%u=<{BGNynq5BqxKk0gIZEm&h|Sd<JK~gPBqu(
z;}N&~3k7;?l<)noaDA_r+&sh6Bq5mPNS&pF9kUc;p1$@;ChhPP)0Rs8VQufnY_;zn
zKNvr5zAUHN!B)C?;<$L><@Dm|1-^{l`;FrvqcZb)HV2pvt+A9~j@^P+U}Z%bXPpL>
zacD|OF!GlOH;bPf_E=v6DRPZJu_Zo7OG))rrQ?Xo>1(+{l?)$R<0v$1${XWX5zNZD
z5oJ)m0mco;G2J!N_B^mo$@3SkeKg3KBPjd&U`1D4TIe^#k|vQ}dL625S6#tWf(`ul
ziYlRiJCcsALbXoyel;|@_SPGCzQTX%&=s_DiCKEYq#s?1(bDj)^7M%C5N59nS)Xb>
zwy+&Cc~kHwBDKHv)}TUH&c24gipY3iI(W1Znp-l=ah1(dG(Vf_2^P|>e0o%*i6zP`
zz()lu83<pHE=ev#J(j-sw}?TcnnQUQC<7o`!w4hS*0IkoCVN>oL81NYh!V%!9NxHZ
zBq<>C9rJenOQk#~66Py{hb7~#EI&W9Yl3ZZ?%4yq5c-b?gSXt-m^+7_gvGByM+9tR
zxGh>Ztz|(L&df(O&RzvqnzqhEL1v`d8<=*5409VcY)FVFZBW~P&KuTQwprlDs>UEq
zsbFLW6FQBN!{54bZ<00qb`W)R4i_m71UX9F&L;qa)23<`hyhcvrT1Gch=M^DJ~RF0
znFx_L@M<Q(G;~cOfnGWsAp9s?mO;s!I{;m*!d*bl)}JoHf};ei5LqcY#fkFH337$8
z&9fSU%qQkcy$FJsN5!*2C`aWJ6a+))oB(Mcns`?#CV!+b{B7f~AD{fg_EX70P=ZP$
z!Y|9zp5O^3o8r7laGFeDwGequtOCM={H05mEXuK%`_U^I{R1uF>p*{n&<LDv;C|Vz
zgf0MWBP1XJA0triakjYliH}fG-yF*EH2honmUk!RbO6ekG>z4bj{8>rQ7af867K-l
zHD2B)3HlJ}j@DY~kfQzebw}2zO1HVj^>59@5Ks#H2&t|#jeu~bF4ljwRwQ^!>$hDe
zN!prEO&Pf^sWbwFVn}nXJXpaTuBoX>C;d6}_Mh`*tv2^S5QT6AQhJ8s3oQImO_i!c
zWgV=bWeEfH5oB*?#LOI+-g*iqS$xXk0#GD-1$Yd$(Kz&FT;K?R2LUk;Pk~-3A1$=N
z0WA}LUE7$lV`P1^dk5FlFHZ_XZaKG>1nSltX;0ktQI1G~uE?|v;1f%r?wbyQ7l3lM
z4O<g1+DE`4@|`j)XzLxW3M8rITr$)%4BVe{vGF})ZVF5$C-A)ir|>7_J62gsnfGKN
z#O49%78jfg+<t+a-z=jB56+8g*%W4-ajADml9*|ZT>h3?<z2ddjgA!oI^xh@0Yw28
zVrY*4b4V=%TsYuDvcN-yRV(#T2=ho~Emy*dS{4lLhfflie-j0M>KLZ1JXkc!TFthK
zi6y&nIq1~xZANq{Gn!Q1ym4>Vo$tS@C8w+AO)^H3W4}FXtGnlIkJ_Kyp=7--gNfS=
ze`vqlJF_6h!N$Inqor2@Sig2~fArkJENN4i(_UJ~yi<%8dL&^hDd=T>bFzNhJ%(lV
z4Y{WDc5!%hGLNCt#T*mS-ay8(vwS`UJ-@EgO$846d<}+>Evz$2FieN_fF}|@@%P-T
zNSiY0Cu#c&?C-V(U^f-mxb<Wj>UZtcxc&zik;esIsG|QyET0KLXGU~}h^kX*#L3<=
zEwjDfyw@R?8ucru)bhvmTbnquO3u@n0SP75Qn@ix?c&)VX~ORvCx6nOY{iL$JH=>=
znEW1!4|A)zx+I5J%d*`Uz4@q&#`LJpP!DOr)XRx1lSi&!Jeic)JKj{c!T7ANo0<##
z;s>R@kZ~%ibm?jAyixB>W*6Ml01u<ulme>S)atR~ZglJOgfmO42M?kz^m3b7dtEH%
zk`7lt&6jBCWo~ZAfH0Q<hY#<x!XCXJrs!pnUT`Mb9sM9UTHuXT<o^9)!O;h1w4^Ae
z0(dhFBI^}ECjm52rn@CfEE!vM&dz^zKV0L>fARsGjK{JB!LV<6HsrKQ(<%Lzu|5)*
z?x^G(wO4UsYKBzx4tHI?Ea-pVqQCvh7+<Izd1oq;MIy5NV@$@i;rI1Fj_^a%Z?S$J
zdx*o?q#(Wpb%PR@n8}%8=$@Su8d@#q_f#98^<Z#v{e&u^Kj?OBm(uP|JmEUIL@LN<
zoO59RAs{q%QhR^gEV(qxu|Q`>B`|h^<8W))8B|cjoZO8sC%I1RaH~#4k+0F+;%6wz
z;KDsl9##GDddJel;F~r?1>{l|G!R*=stO|a09v+_L=|Jf`+h1Juhshn=Nd+&nT7#h
ztvW>KXD=ZK{(^g7VE+$^sS4!L+pLlYhb?JzgM2qW2~Spl-PwO|^Fsg>kk?@N=H%3x
zg$6*h7LeZ`m<19v3QjaSf#Hv8rOY3&4b(&Xtd9oyH{1^{ayD@7yp%JFkid<INIL~r
z>$54+JDt6HR*%MvJGC{^P9}`y+c?qWwCAH|MqAuXrmIiy{re7GaxI_c%Nc>#EndMF
zaVP5?r7g5_&u+<&E+(h{_D*wK(6}jYP-sPJkL7Stt+yw>%oS3<a}vGa?|XCj(u&D~
z*~I4~M;P<4YTDmSj3ys!T7^5k$V+@1AHGx=Az4kTq1tSr@~EP46@00OYsS%A$gm8w
zWxzx!iA2Q$kq6Or)P?ElcGT)}KOgl+3j895iA1`;Z|vwEi26u()4l`mCS?97yNF|3
zi~KjDoEuQM!=wQOk;|;`#<P)YKj&g%4Ut}2BnDjIhY;)1#=z2iPOV9nATkcU9D&}>
za2Cuxsh$yB-L_O3P35t=a6!GoBbQ?8GsLa-(NyuI=ySV&6D72#B34baxbxM`^fznf
z%L>QNn1#yTQ#|}xh2~8C1w)BScJiv2{70|%KE3*4jSjAjiLZ&g?Lpi{#M`UFQ?E)c
z8rAi>qh-cv>nH7(G<)iKdfuc9J!~O7H8)jwricq4Q%nn;&pn&Dp#8RkH__Yn;FfaU
zQ{Pn0hS4VwD3d6w(*2?D>kq+glq0_%dYpE82SV*t`gYe~EmmVlK_X#@#3Dv;M+}U~
z5dbV8?j?iK;qR6Z8U)gt0O*I4b%}s6Rj}BIf^aDlUh5&@Uw}#tq@<a`(}*;RfjbAr
z-|#f=^1(J5U6zpEI&g%b{qXFm5vTg{c^YSF8mpqvMXbWVfwbLr|Hi&q9I|Q?VCh=7
zov;XJ=ID5AyQJ0PA7ag#l3}VWXm{F`qaimycJF+|a;&#phm?ehIJR&)Mo!;0Q^~49
zEqamCK8=s;_&=Ji1Rm<PYY!n4VyvO8V<aIGPuA=PL!^xm4Vf~uXt8HWV+%7OJjmLH
zvNX032}8w<6o#l&DqE60edq6azutLz`(>Ir|9d(2xzD-IbuoQCcXd}v2bP~xy>lk~
zjeD_NLGpR;Lp(%J@`8+p|JK&-GbecO=?dR8R>);*WlG*p^;6fj<2ttQ!4;YH54m3n
zi9}1vGV{L6wxSYYd~I9qPAxvhrsPT;30Pm^z#iNQgvsAmca29wJ=5uGu4$djWR76I
zg}Z~L2Y2|Em%qv^PNY7>5<|b67;hDEiGP~$9HA-!{f}6FQlQg3(%ljR3g1raSP1v9
zP=PL<J<9D(6k`BUAT~-ty#eIdT4PwyD|#Cm&zW$ROGl)s{g(Yva;~DHa?i1x&y%&v
zBTQAUX;W+YEB=$2`Pq2C_=ht~J74(^bclUGZ>}I0aSp%N^>1`PepUEXn*-s|xjT#7
z4q=q^w?C7v$Vla#opJCzcP&H}NUz8Cym(J9(h%uQYR(R9!*oBMY1lH!8T}{T=2WVN
zO0Y2RiQ6N!_swu8(#k*fa~m0*cre~R5+_;9To5j!nqm{I<*|K@`c$9v6@NY~18wBq
z^EM5dnK2iF8jS>OYOU@(g6>7X_ISl)=0w+rjL3-%ktAp*1)NvLL%kjv$ejccE%}j^
zmMO0dV*_%VaANOWLnR5BpE`~x;)L6Hv;jZ^Nvq##Ry3dRTR+@fMtOWDt^*^Wz>wd`
znzm%!J@4d$DLA4jml4onD*R*9sC{c|&W`Ei=w_hDK!@uMhMu``;)-m`8)6j3XMe@5
zHl8N;!18z?E&x%pFu{_Tb3LX_f+caCi+xy;f{c%Fpn$#uq>4N2A##DPY)FYMWg5-L
z>gA(L^a7}`J!rT~rM7o0emJ}=R-^*KL$Fw~sh~XEZeP(-gU+~@{#K3db<71U7ZD~C
zco6n))Po)+Nsga|aJ_|RwB1Uf=XF*{U;w>h_Yt$W-N_$j9tS+_bn($$a<1KS=JM-W
zgNQM@^F_D(wSdk{*Uh<wN*B}>%cMW`9Bq;zz@5TSuJ$q3r{_ycg&~q60=+gLU_HXz
zY=9Z?MLPVE033n{<UVelxQ#X63Wn^1mS&aQ2OiH>f7}?nX}67)am@ZH1ZLm~du*YJ
zJ`LdA4>xY;QA>tfb^_JV!s!NV6ffZ%q#YbLBVi^)qzE`WF;pX8T944CB|E=hxqS11
zM&#2QZY6^k+z8ewgACrgWqaLz|2b9rompurkvcP5d}KR2b(~7*u>M%F<&gdIa@g(u
z@*Go#;8QCiKP+u$)_JD=4;sD*zKwm&U-OSb=O_=_uRiy<r{anxxzwCgk+rgXMN_nx
zwYDcl9-hs4<KYnE3BzxIA?_mfIR5}@RxQH0Qx4EO28z}3Wk%IKmi%XJf5NhQo8bgF
z%X?*9ZA<0V_aIyD7zz$P7}N5tpt89Yn8<?Q%yf8OpQtMN$?6XKjZg5WBP<0+{Q{jU
zqD;4>1!goMd(%WuOK_I9cd<Ty_`O+NXuh5duC>>#XrCl?NS+b9VEW__TF?qpmy%8M
za0qJM7N6esb!+miv?{TYr}Et+Q-#$f0`d1$bynO2IxmM*Rn#TDAdG$xDAv86{<=r|
z^->Fpp5^7}U=k9~|7CP(w?vg#WV)_pN7cMV?P&8|y}cR@%#>6Y5I1R77(NkKgrh*7
zHw`8kVJb%il0ebCc!Ns-R`3JJTNEroCdr}c!6Ch{ys(Q5U}rzMvN3$xF>edXBLU(9
ziT8Ln_i8UDdT<gijAYmQ8IH*3A+j$F1tcc_&>VLO8Z8Wl+|>jzfnT-iq&sVW1cU^e
z*zcBaI<h269=45)$9U*Om29b*@Ht|6zO4I%WXN~G+^hOhE@zAm3@Uhbrqvf<AMWU=
z7tv9BGDZD|`8wpeUd%3+PX=dN3odkSi&dO8ia9qEnOC}RY*=5<`Ed#FrOWznUkkp(
z`!aMc@{YZfJUuuuRo;E^s@+k!)b>N-KS#|C2u}i@0Cj<ILZqtIu_GeyKke)eIIl3u
z3H{wiP50;%N)Os9SD9;;Kaa4_s}9eD>K@dT84IQg>ik!s=?D=u+jDA+kz2aWD4`F{
zywr1o7z`xScBD|z=Iw7G``s+2UO@xmwWwQ-LNVV0Ac1)u2<O*}Lxq+>?U>b(^v@@=
zmcnVIQP`oS<*0JXFJE8|wi#;@<hc6X;KhXyA<K+w&!b=N@eNLEF$?8bb+EkNIAh^>
zVKO$H$@Y!a@sjlokD2z`y3*+`Eo+lYcXCU*QIR556FWb=u+MSw#lEs99dC78ChmTh
ztQ&8^4hi4CmXcICup&E}WhOIYrl+sTl*U&aJEJd|^utR-n<4PzCCmSEnDzpfy?bx8
z0^fuL!TK$ZGLk;xey`zy@zVA%9_K<dWAWrr_CKz#2etAZ|8$tw^U1u}uARS&T<}4=
z#&md^`KL^3$s(bnGchbn<3&v0%qMWcfcFD(2yho4GFn0v@=h<_eFti=V-Om_wo<Pt
z5Q=#N@{ds&EjEB>ZWIkZ++HLuk$yY%$P=Phr<|j~*+&XSfe@1xtLFuuvN>Hi*`f6L
zz7_ZERld90hhN`TbehGS1HoKi0RglCfQ5k@s)G5^>|?^cy?NKd`<NI1S(dDr-+Ox0
zL}=LU`oPHI!`tD(X{S41tXKUJ{7rhg+wrqt$aQAV;SN*ja?8Ee?-UG*UMbRM`X~>+
zq-!*7R&J;m_qz8~A|@xM-nBxn{fg23iEUe{K|(%h?SeFl8hP9=cK>9s;_%%^gF>-0
z>u8DAr&O!S7mlAxytm)tF8!ED{zl6-x>1^d8B5)HK%}l#BTr529gkoM@pKG$SjSX@
z%8&`hV!sU(u5u}+B*0*o+d%ndi#x^s-bV8@+~CT!kdxq7gttbto}yv32^t7PFc^jM
zJBC}Z@==)})s!f9#+?E^*B?y~C|L%yBAHD-d-}uyqu6$iufy(Zso`%y4g<kE%e;ns
zW2d<<@xTAKQ7zwMds6k~Z9VTXhX(8?4HkHl1LdFle^PauF3+|Q4Liynzhavj7B#-W
zpDeYgIlenvHOgFnTWn`qb;(LNrMHy5r)1^6l3T1&+|48DuJgyAj7_QsyQgQ6+q<55
zoefFm&&;4dcNrA_SX#z6@GwZhx5_7HTEM5QW8t6*o8R{GV~RUt_+su9?@4|3!?cl+
zq*ov<C&9PQy)EMOvwGo#*(-tW^D9P8QXWz8{NR_sF(EAsG^pv3B<w)^uILA>B#&5R
zbeOiqeLo)VkkkUF{S$(15uoUvaE2dKn1L-(#RC8gQJfI@2@CJLeP$mtJO_T=`!a+S
zg&&KaaJT*?WJMdf^VH$~{xa^l{Wns)W_Q6HEG?lr-aTQC7Y!J751f4ewL~tY^utyw
zH=#`7)1CB0o`wLI#Fg8>Pg(Xv4)9~CLLXg96sz7DtO(19Cmr6L)S_^keVD!|^W|f3
zaq2@Sc`V&cTY=!DD0<9%!B1Y?>|VpS(REGT(1){Es_h_S;t5@P&Iqg_stP<WaS2JS
z{ke@0#LPNSAq|q6nvx3#FVY?5pr@eulLdd52?~Ie4tu*yxq|NRSaXkPM=oIptIGJ7
zDJB*fub%><K5AZXteo%J3N6)To6t%|?Dg5JTMXkm%Z~#+9wuDD;Qlh7tkiO!^5}Zx
zy0DK4;7@izmMs2dUNAvkFCT6#S*%;YWKyj91LAW+cs=Yl<hmVY2zNn=7Q0Ls@El1k
zs{&~Ch174R)@gj7jR*L=u9rYetN=VmFKk)i*S&r0Qq!0qgh`>)PiIa*zTDjf(|+Sx
z6Bm%r0~8pcmEWaa&B%>)KHNL_7nP~InOm^%$B7H4sD+UH0k)ih1){<?WV$!ffu>g>
zG}=O8dhpLHM}oqv#rVJc=i(uk0Z@<K+aRAwg<RatcsQBT5DR0L;gEv(g#v$6h6wEk
z>?#XHQ`OO`0{BJoW8j1k@&buSpr3=xBPH%0!r2!p3O7GE;cpm-T`*A~<So<Ic-t}!
zQh<<D6pldTBW?TtOC*3^%&uMt?w>>dTM*y0;h-L-W!|HmL4{hmx*ci5-^HJXYZ0<G
zKgE253<T^_V18HJSx~2%TV%#p7evAvXuSYJd0Of?ki~rewkT7a%()RtDO1%f`T*y+
znNokgA)|Bg4d}cGj6sqxQ~LMLOn4ZGJbCRYgmZ@(<Ji@|D2sF5Fyoo0h8+Mz{=+Vl
zh1sULB&ija65}CYeufEh;!eB{ZN(ZLN0p##4xJWnlBxBPy!MG0pdADbr@XMuDY=&r
zaqcK2C;nf-osuYq=)7H(DKc`~obpQ~H%swG{;{VR%OJ(^RblA48%FY~*)r7*1>n~q
zkc5a_qj2D+?PR&+)o{kAJ55#%NUPuPpHMG@d3XUmmZa^_Z8PheSRTm0ny<xMVb2@|
z97X0v27QGK6ADc5kfq*Gl24Rqj|Xl9UI%Irlgknm{z3sld&C{;=7w-_ygu)k3WN+=
z0q(KY?RhSvI}6H>&vz<{wW_w5jg-CyvBzL}z4FYxOu&{2K8jis6Shn?YjelyGA6Nd
z2>_n@ww*RU?+qKpSG7SdOAHigAqqy)Nx$;Z%Y6l*7zS7FT>$9`0OT=8C~vvIe}2c<
z(I_n=2Ty~lHfX&;=*lb|unIud+X`A6?bVHk#f0tQOWcg*Lno;Ce&a36iqyc~rw-*&
z%fva0u?G%<w%6QuE@&BM!jo!46uNh5kJW+FM@xEc7eJgw;@OI>wDIRxy6Da;B7{1&
zEp;T-33xO=H+O~|dwZ){7Kn6Ag%Uo&`N#ZzReRYM5MxyIGE_?SoI(=v|D8`AnFtGn
zs>a3=B4W<>qvv|!3wYnaH-SD)%v#Vp+c~FsR>u7w<^mD_j2w$;Mt^YRxZoYfRC8L(
zec2PVCK5PCd6u=hfT`Ovo4q@iA+N}%a`M5;k><~rH_whhPd0WgEYF93n@aF~JT?g-
z!_x~b2GR4YabuJ`+EMHU2U_nVNU76a{Rj}g3j8OB_KUry$Qx@xj}Cz0eWTPo(f#|t
zegE`m?Xgg??Sqd3CdXizC;X4xuO1|~A52n~(xMEmY5VhNe0W@^<<4K|*Q!{y=CAf5
z=VBL*?%VQ5V6j2Ct%3CPyfsODTF_pQ7hQBOU`0i~bi7`>OV`D`tnGVYx4P}^RPV7$
zvrZF_%l<~;qifZ{8pJtZOMhP~CWO~}(lVwYK)w)?VjpAfk!<6X3P+8ANM>mAsh1Ya
zTx|oSzhRllr96vh*)}IbhWf_6s6b|*qG6|I-|-y9H!(}w;a-m0FDcM(_GHG2H|UO&
z1*XcpBoLospYNAkd($*3#j$pub7V_>qqgj(D*~w|`GV=sH60Wyul%@lw&{}7dbe_I
z^+^M#FYoxx0>Wa*#4d(!`nfu{&vKMUH#ZOZw|&3s_i*PulFlQxzMyNgXxQ}roc!;c
z&#TT`d1m}9{5P3xRp&ic^2I*r%I?XdT+XL`@iOjKB-|zWY&Kud)3>6o*Z-mOReiO*
zIhwIn-aBvcCMpOU&CF;S+sphV(^0XR0v&v7o#WKdi6e&Mlw||T_=0-EoAND&&<QRV
z6~STKzu2e81y5dd0m?~tuN4je3N!9gNocw#w4k>{2^=@5+%NZAq%G2j%UdVKWJ|K?
zko(6sB@BO^E6IYrK5x$Y3EX(D_*;<fZz(soUR@CIR9(aNFrWT0<AGCmFAJ&gZ2LXp
z@x5P6LrSGHX!{RIN$(K>mkTPp4X<M}iTli7eyBKUCRkW!TRkbKbWx(#awj(hWaaY=
zce}r_zD^MI4R|^qK?%v&ej@LI{IEXJXGv6YCH7N8<Bo^@_m)Ka&Q%D`8y_%j@4509
zrMf<ZdC_@r(P%uyfh+WJ-fmX81Mb#rm&CPsc-#;u`N$guKIKN3%wI<FHb*ud08d|#
zx%XYp=^kEFlHLQkXs9VcFfzHi3pN;&dkzrQT;|Xc5hP@VXjNnyF=pN4tQ8a{R$;5f
z@`qZ~FP2*{Zl<4Y@O~%luk+HfYgSYHk7Mmq%FN^89UY$=D#A3${q>6Ykcw<FPD6R(
zxA(#is-d9G)|tqOQU0w1&Fk^wCz29f4R=u5Qq+#hE!(&LG_pzal#S(nIeK1XRe+CZ
zXf%KNl+3i?(1cy?joQAUeF2fO6H@0+26Lx9?+fz1p;$gbNxrzB=^FIfM&YwG`m5|*
z^R%IO#+PC!0Uk-V43>!e!@I0OdzI^@!F{MLSgV85q+BPf2&rV{Sy5mlJRvp#mwy|%
zc`ZvvxBcKr`6Xs&ol@+9`im*K-)Fyu0_BtGEE5ni5QZW#eiTG(jc~$i@j?Y?`F^;9
zh^0y8l5E-DKQ4b!k6+y=dDppeh4=Qe3<I}KR^gahwXdRuFTEaa(%gQ@KT}pwcO~MH
zD+qNrf0oHM#=ba`dg<aoe7J>D%<p26vA}zixcu^uyZC6IQ!?r5@$S*s`Jhj`K8s}E
zADIoWdtbNZ$@&{f^dm*tYd;IF%!?W-E(X>V>`ugK?!TbJ7%RQ98|{(uHb7eN_~Egx
zhQFw;>12z!e#SUlSraR+lI8%V4g#@#4#e(Wl?B6U1{TE3o#v$>8}(buAm($)PsJ#B
zYF@dlZAQsuT28l08C0Ph37gaAv)_B{`nIQCaE!DF_5C+qW_*GGdjU~>0j%dG5T(c%
zgLH&Y)>QG5QG1>qYqML>gRiGv$LMMgET}!W%<6u2WOmB8WLUO5;E(LIr3|mixXrRF
zg#==jS<~!?p?^^(>D{)zSHgTU2Fs->hWg=0p646x7Ox@SV@&U2mkug@vdSJScPdq8
z#u(QTRz0n*`q#v7V~rh;+LjxVvD@*>X&vT+?$jrf6#|Q+#|e|w&Tkfo*`Ess@E*!h
zH)<Dyjg}{3SGm=6EWZ`(`rv3bbYosevR=Y-?e+ep2a2huUFw#9^-Vra#$`OdY4J^L
zRn04R>ifmEdRZ>_=;6eWO1#`nKpy*Vp&-Oqw;;`hS4_ftR0yUlDCY2Lwt#jq)d)ZS
zZ#Cp%U`%qYEg*mN9V)8cGA}N8U<0pvd!Gi9zpv(&e2_K*K1N_afm`BHU{Za#u02uD
z`d&|51#e6HmWhdL5qXx&B{^#~79%A!G1`B-gZ5nd1BP6agXMpx@VRE#RYc;XD<7xU
zPacC*0>6v*D3+&2!ZT_2RPAlOih`Z;lY4c;G~L<?a_aB=%JFG<a^@b^e3wdXQLF3$
zL%DU!`}B^G;i-p5!>jLF8OzT8$u<8Kj7hk)FTeJF$}=Z=gHg#<!GVd&k@lq9ifLEH
zZ@y7!_}u+w-=aabT=M33&bg8^y&fVXLSb#o*qgkcmZa(oZrp~JT|tu|$??@bgwbWJ
zPcY?|s9_w|LHLoEzCuMFSyt!PoV|t7rrt;9Z(6v1FsMf1JR)ORpQr`57$Aw4d?6UU
zU8Y_%L~=Y<PbGXC!PDp%P^TQ4NR9%6g&tF7ZWkE#WW&z5-oZ=n11_CRUH;)MvnE9L
z7;9?!i^^cH?-~m?=$h=?bV#36LEFRBH?jOI`-HV1@w>pSHfQ{gi_!b8ub=;z$ST(^
z2+MH(qddFsf-`WLiM32<sG&dSz69TgRMcXWk63pBf-U|K=pWSX@(B_Q6UIxlAd-Wl
zz~seO1iUOuZ-@*)D&A_ufR-Lcq@+OGsS@r84wdEB8nH+Awroyma=)$Io=^bAsQBgY
zF>kV<*gUystqbT2^{(joB{xn=X^`F$tuL?nJXC8k!v96--7Ptk>&dfcAiob!V(fO+
zoUfeEHe3l&8*6W0(FwHUKQr*mGxE#evPW^RM+PXQ1g<X697g1tkdjvt-`#PzLSxT4
zP}9!9rF+3lHrnQJ0pukqiDwZ)O%Zd*9lJwhAN&hRXKBEbXluS{+%oo4vD4jZ5dH2b
zHpuY$IYaL85J}9O;y85bK@QEROSkvSS39?X_2=7NcMNG?KXQgq_tn|ou)Hq3BcPDb
zy5>+^AYP$3x?Q3vePT&6C%JYx-AeCgQ<_EX;8$;Rqx7_a_M_*e`k!m<lNx?l&PGpO
z7TYuNbKGnDYwWRzx$L;|as9H(f~CYeFNW<iqkkJ3A8fMtUcGv=X_C)7>;{v!ig(Ab
zmB;6h#z2Xlg~JR-uJ4Bofvj~wb?i03=n8IuxMCIy1XBYYr$c<HR~4f`V2Pp<D$$a2
z2JUT5{D8~Mi?(qBI&v!p?-R_RObihi;uj%_^AAk*giwM-uLJZY`$E8-3(8pc9$Fy$
zz1UxX21v<^v@Pl_S+rkD6Xd=*uTE%?d7fA!w3vSM-4!;o*B;ark7p>p8Q<9zdqtfe
z|IsnrDla8<XDnOGl3_OKA7h-pIYT{;kQXW8-&PoIrRA?#*8bRvOW5c}!tg6uCFR<p
z_&aOKyF1gQ)y88(ZXML}bvWf2QlV=6St_z;Rg*K>)~74ij_}6${%wcz)tkSI%GBqc
zGr<V7%8ptVmq+l_!{p}#aF}};S>LD#?Wz-fziGnWcihN6D6qd};s^zy^3Q=J0<7sg
z?h_jf<sgbD2M#J&CN|LOb&>KoTtnf&|6~cbIbH=^ozd@=?=egA`?B6xVZY5ozWY&b
zz0)_iD)NU)xR!)(Pt9EV=AN^je#?kG=p}uf%KsN-hAq8p;GIc3U!7SuH4&f95L#c6
zEmxkd{&Onq+{<^~fhjgS4@*}xp2yrvJ$o}fm$sly?|AsDFk!UNzc#F<K`MssV{Dz^
zZk-r4d-85=--yQ?iy$GWYiVzG?3XzO>xIv2G<ME9dsL;Ir!cM1qVpiNGkH>lr)WJo
z<KkZFuorXk5~kcAk8X~(s+fd+qS@0iA&LsTb-{+bZmB0^)TF|C<Yx-3$6|1`eLoNY
zhgOg_F7~H^-Rv8=qk1QAqL%`Tmr9D1$2>rJ(kJLGboxGn>$sa}v#313s0my^K!c-H
z?%mOgcSG>W{O^2K{aSDjj^U%UGqdOB*C{sH#1j|Kg{P#go?Q&1*~ZShZOgp=WPxZR
z^`c<^RZXj#Y2xe2gqi#t4?XXB!SJ|MF5X#fo55>->aI7@E!8_Ns;5nHmC{ePXRjYu
zw=<VHhwcB&k?bEhvX2q<Zc}fH%OQN9@T+f<b2jYr5i0(>9$7F!NA#nC-upu@L(bq0
zgj%dpl9dZpL7>JONGl6~W(x4(rk9-N55JCCHhjVNf;1=bI!@Uf6iCzH(g}!&V{m@G
zT+{s^2@lJ<Rjff+E{EUEVVwkaw74J?QnN>ngr&6nW`<@!@UJ&&yHJ0H%e3&=>a$rn
zUWcbETX!vJo(SI3U1$6_<h#h}`w{ovmW(Z8XdIF`S!a4Rt?a`q_@13p3w4?c(Dn46
zEsB31V9WmJ#Ps+x)w<I)A<3tR=05t{m)@^$4zNfa?vc`1CK~AWSlFI_+d(hw5xier
zqv@~glcbAoY-64J?3(7q3n0_lw@|6I?N1x5ZYugxEzz0lfXn0f-Zl9@P{3PW>^vd`
z2A8e^@Ofgs^QwE7s}auUInzL-@oFQhQOx=c;)!|z&w1_RKmfOJfRZJ49+csaKze7s
zr_ca|e4*;7$ZcU^0crnSZDR7_I@S|2^h^0E&D-AGW9M}=2UE3Nx@{%&G5XDl#UJyf
zM6+_#Ej3<Wx1Y2Oa5$iD`0DpZ=EcI+Aj^==Nqf}Jvq1N7|8(z*>)JLU+a_-1$4#c>
z`^m0VXAiqt#i|wd->i+s?R%Zs#UJn>rg@-z4XrRdwwHr{J(|84?yM8ubx5IT+)}Y2
zg7~#y$4!p!R|R<>$>neXt?(gK5^7w5ptr!oH@33mL!mUwK*udy_D?Z=k;4A^E~uK`
zDrWr%mX|~`4KTAnMS;PHNmxYi%cFaty9JJ2tw1P&g}F^<2zO>U`0?>M?+u;bcQ?#w
zyFtSJdLy)Sj#jqIQ+e-82WOQq=jvIjye}Z4OfOdJCJnTiaX;>FPLcr#wdZc3*mne`
zWV~Hq{@Y__n{h6pzKimYo1ASiV;Vd5+yjbHxP}?o07aF9Q>}X9xD{j4R6(QX9we4#
z!+4ifX>8DHNrIFof{%jwTnw~6M9rIu+@0v*B6?qB%A>zRJK;M7FlQR>%nH<;uc<F>
z8nU8UJ-+L@e>t6dtTsGl{M1;OY*FxLhm!?X#Ok8<!n~Cu57lj?HR_%nm#WtfxTtaM
z^1G&l`-Li+_IBoBEQzX-@e5J|?;8`wJ+Li~$`5zCOP$_(JE$f<nVX3R5Z=5{UZz3n
zg%DBa4Y^f=Z$@Y86s__-q!5m0sE9g3WesdqW`yswYLNsuMAK~m^EDnQ?QsS>MLyiS
zY4ZeFu;F55%u#u(P&#MVEN>{qU;Y)pz5B;K)S2=l<*o1DA9uP$3`;yzJTm4JyC~C9
zC#N(jAo*o*sFZK<-OJWfc8|Y@_+u&A2^Vp-(8gqY<f^zdwT0~|9)i|C_sTXBBrJp;
zWQFC~=;H{hk9U?!DUMzCe8~Ul>T-?IZV3}Z+U~1c!aBtO=a|f8xh4M4506UF=E&Up
z6I2Xu6(eoJrXm0W1a}XDnQ08jo?7Z5yU3UiXSasB0|~^k;{d1o*vrU`)WgxrC!#K?
zCY{jjv4o5y5F{%j)Z)j3J<-MEa_<-W9<P78*xL}RxDxHD?}k0Ipc7Fce3srp*I-y}
z@_zobls#W`X~s^hqWW-yJac(_*0Vn*Gm}vyrMmYwTNmR^{oW=ys(GY}pFLd>l{cSP
zBwsll{(R55uSG64aRu2|#FLWZ75_AzV+ZM&j^}?=YLMt|EER^a#sFnH7wmYz_=}kL
zA@IXIZbX*d0jqpMxhF>?)og@nBGF=mXF}grpI$4(cy_{;nf<@j2A&2kZ}|#>L<rW!
zEielG2Fg{bz-1Vu43R`FFc?C^@HardGpLuK%o;3~013$4=vOp`RP3Uy^stO^s`o@2
zV>wv$_P~iA#av$JTCM=qg7~)m0&Gvcq`=RMr1j@-#f(yHK5n&X$fPy=u)HbW<Kory
zWM~mYC~CI%osG;7q3wy13ff|v??aW&OD})RGge*ZSmI-)-`+-R6-<7@;J;0XDeI05
zZOC#;2^KZfUfwdk*b5+=-)o2aaF=4%5`onSx+~$P1{24-dqLUMF>;)BQ*z?i?6-e`
zP}Lx7lmq$mT&~x$9Mu|v?vF*7f+awY=0j6Owadx+KY(2gIRjYwqZ>7CQGGxuGI&**
zXNuzv6gpgx6_9KW3f>hP5>Ncnw)8pv#lQpcr=ENyW3!TTKku&i-aB7gb}hDhcrw~e
z?G8P-Z+@To+}HIS?JT2Voit|1uZP7S9vr^sv`a~4;KfA$-N<_mkMCam@_zfyd+ywV
z9HIPkr&G~Ozr~=>6|`BQqyd?;)=Tf}ouG+5D<wwxW2?}O`zJ^@UWN!6)N6>e1&g)k
zU{*6;mkQx|__|FUj)cU^6v2I5Bz5Yc$@hb^z;O&ujip|di@`n40uVaDdn7;t?!R7;
zjgUy>1Mlf%RlN!kpPTLl@>S=i_xT!i+RvnFxpJFyID`{36pKZNd2Po0wtbLg?y5Ht
zuU@~|jX%-*7xl|wQONwC6A^v;mSjK$GgBx=;tRMhetBaY(jl}@9=$A-CkrTEf$ggk
z9+p%fwv>aSiy`{X!3Bdh(Vhy!>w@A*U+gV@-IwZx{Z(!a`zshEXU(BWVC_vYGb!Va
z@#y_env?qTX4CTvoz#V`S3GZM+^W2udotkM`%fd?`Ds3(RU!urlBHb@JNBhdn%8)$
zbY}V<B_t#q>xujpON3^qn560WIu#HG2Ox4Lf{e{To*sHo5WXOaeY~^Nx4aJ|#3w+r
zj|L~;VL{%YPvn`^=JNc>qsmR0ib+B_q6ldRBDW>c1FfceOD_4w$a5LLdw{l&sbq;1
zmlqX=3I0Win*b9fucO5Ygb&^au*4(;7pBp14^pQHWD0n=-0!dLm|%bwK~%w|AC8hU
zkFY?hlE)>FUK2Y}uwU*T1e=hN&kY1vE7CP1_I(Fqy?H6ggGar#QCw(sT{vSrgo;(r
zgquw0o5o9+LV#gc4J7AA?koPW(x*I#o`udJ6$b<|i~a@m87Rlvg$C9`1pzW#<-oTY
zyI(!Z!Kp!zC%FQit^$h*1bAARH8YR^6j)=Wg)bu>tz9=`TV}`J&*?1xTp*%wO1MVb
zH`g+Mn;J&JF0$6<F^lt(V}X?NxOaY*4R^I$^|OQvlAAqe=$J$!lS0j;PNUm?(dLaw
z52dOcwr}ywPpT-qpLEIO`|J2>`OMn>Lnnumu0L)|)Eq1(kJ*Uph}Z?u<0cDrcf@|J
z9b!tw?sM$aPdpfCwPjIqFd^VMa)N;E;kypu?7o{r1r55LqH&k<A{{~SJO=pfmyoRd
z9ITmTsXDnR2V@R)0Ap_wbjGngBf)geFkWUz4rxs#Knd%=yk7Dtq!$UHJU4o@@)sp`
zGv2irnEm40CZ%+}x(Y3XH6M(g_jbxC+UF4_@$|K@s&yGtm8V22gKf!v{5CyPOEMu^
z<exp5T1$SzuIU+zisZ5KDO&yw^Ly6SqQpa4%!UhBm=7p{+SRTk@i>uyoRbNKc$$zk
zor$F@VCsdfv;6Jjm;%kWQlaXZ0Yx^J$k%%h;Y8&Dtx8i&=sduP^i0E{ITjwv#^ckO
zahS4N_2iVMK!c%r<>)s}e1)>gRee|tIPe#N3{XJF)k0zK4Rokv8ue7|5G)XR_poG<
z_tlvHF1tKsnNQBz&chklJ7*G2j~L$Hlbx7D*{mgEm`!?i*lNIMzBoQ!SJA*XIu5Je
zP54rWBTyDenP$bIeA=hfx#%Y(H+AXA-xW$RJt@k#C9N-v2~4mbA`~(oC^>kMjDG-c
zQXqqi;t(771{$1wpbLaTp)>^}LVi$$Ag%2)$IP-)pm~!BZNQALrof&6r5$f?pCBB?
z+~F02Ho3aTjSK2ADnY~xhi5hW07lgLkoISV<+smzcX=FsI3e&11?ymKdg(P#RNWv4
z%RGEgZauy#lL$cV+#YjFep2YAH|}{=ApeC6j8XoT0oP-aSJ%#b9Y|N0BEyLYeCCdv
z?bq7{@+5{^y5@_SlO|RPj!<!X8hAEm55@kjWwojAp@_ezy;JkL50J<vSj6T(iIAYZ
z-SQyj6-okNEVSbBI*faQeoZ}V{`y;?80ZYaaAAZFfX-Z^Xg!cUZ5w|`V-)KhZn=b^
zS<y4);N~f0!qp3a+$K<u=4<=68an<}L3$L3$G^}(djcwpZ=e{Z2#?*(G`<ZNwEyyI
zkmZ?Yp`oBZ`wcouCUYmB{ZvI8jYDfBUM@L4!?sKvd>U1&+JZ6ZCaqyw%Y6B7!dLUu
zmig0Bpf7udUq9opX+R0Enx3LwFkIQ&28nB{#0^4hYz!_DCTHVmhopgxfV1dy1=qG+
zD!)sDmB*X`PzOKFOLIJuFjfju3L90~$qqgBsH+o-c^B&5wW>~1{&Y_>6ltAS$~Dfy
z_=%(K6vO0;ou?jJNo5pHnz5j|)xz7gb54vAnmbfglo55wG!TuYBALUucY*bdDvq17
zN0)P-4cb5%*?21)bwCkiybJ2Kw~8SH7Kpgdd7y<r{3cxELSPGn7zaW;^)^00Yh>Pf
z#E)27dWANbVEK>AB~|HDMT$A1juyRF(Q?04qrRQCeF(+WYfl5ef<9qt@AXfQCao9?
zmQTLjg9ew)`P=7YyxJ1=a|r==ri(1my3M}~DG}XHX%a15sSw9;%XH8w0Ahtov>^qw
zY;J+52gsE;rOjIoOI)iz+3jNzn|8kFMn#~G<xywoMmrU~T}<qPyTzMz#b?1#aZ$ZB
z0V&TiLChv3iu$0XthLO<zMUwxI04;ip}u|7L4sr<JSIs>woze^)w(ZP@CyhLI6;f9
zvzuiYg*kCGAs-59x8>3OdSgj4Sx7twN<2t_CzNGrZ&iU6({AJ&vruJ8orBW}Cic6l
zCnsNlP>ybaP!eeVK{toUM<6?wH|8Nh-5B^@m_K+m)_o@h=b*yQb9FB$sd1Z=qS-D>
z2HYPCJ42{Y0$|w0*9(Y2Ro~&JP?r4zrytN>^iSW95~0o&p+{08&pNJMxkx?(1g>5E
zSswep?9|XQO)aTi0`-wJxFdcmxv2QnwqV`$-3)#`%wwTp)AD@~QAkA)fnq`q)<A}2
zK7Leh&P0M?`&KL%B5U5bUvTR-ss<JZBr~I=Vo97!JYCrrDs$u7=?jzrXu{`x62N&C
zn*W;gd_zQ6b613EE`ax-v=`t%nS$k#^60V?Y2jw9z$?a6-@les*QomxAH7hfellxT
z(I+RewB%E3L0Hyqz<NG5ibpypK$qQ+0v)MXh-PzdB-jB_c^wuSy<FVW5h7o39?(*h
zxwnN2f&hcuA?QkzXc_ChI{?telss-kEw~&y)7=t!ciOq_$wKUo0U<imz?cb0<Jx?+
zn+u{CUAJKQ=EbaJkItqhP9Hv*6(aC5ZlpSI&vnKZDaEgLO73m=0+wpGWuesDn{I**
z6C&u`z{T5d%Rf2IKh>YrRLqnLE4TEst@Rg}QYgRF`ir@RK2wxt6<L+%rF#?Y_I~0)
zh4pKnWMQr@GF8v;&h<38ACGhPwFwQCjg24k#&xB6PfXV6N(YBiI+D~g)^N`{O%*2O
za%^|IsdP34wm_HA9BV%|$9Ak<1Zdj+Xzq(yy1zXQgLDx{$t@Vhbz5Ls02~3uF_1)t
zwzv;^k=i{<XzSi+Zq>IFO4ulCRzgYA!6mVmKO(jQ*vmT@Ht07hZ*{RoRIP@<9tg@8
z;QA+&BlxgxSy!(#PCCJob;SRQiWDv<r@Ah>C5lOWm|OB-bDRCd@-})OefZ%d`oPE`
zv2!{xMY0ZmOyn12vO8bZ1fTp8aiz#{qQx-%hv(KKpASzb+Fn+!_d55bb<btBo12SY
zE(^T~5EPf+Icc&bo3c`A^)4ZPXH84N*jn71Os`VL`AHiu!v)?~f`3f1aq*H}Br2^{
z<@6^z_xrh1Iyn`+Ije@cf3l^NC4<BtKFbP=t%x&@k-|#f13vbi3wWm9qr>JhUO@Qw
z^UTR|hcI9HeR(@}jYrvjNSh57g3w6<`?>U|#m{_8ILDBH%mMJCReG&KOmx9nr~z=j
z2eQ#h>Yia7p9!>J#H?@4c#wPD*WEG*^3#y^&!vBDZn?<4B5k099hkRO!#jWR9Z0w+
zCNwWZt+FbcHDC-!2?gL70`9o!n@$Y{)`x>k8Bd>|TM0{S;rS6H<V(9z=xjT%IAZ*`
zGo|DTR^Mo%XRt7r%Y-=?XOyEOa=vOsr}|Y`!KZ<|{$i_vx+eKbty2-wsh#tyP2F4y
zdAoA53#hnBf?=lo#J5n4x?67Cc7?&9ZA!zng{H3>_XeP9bi+K8@tP|=qx|`&9zQX%
z&bfm5U2x=tn#6Rl!96VVQ`nj7>+h3ghjgk*A<4bc-Nk7O0#_qq*!_+M--4?Kj^=~c
z6<Scw`e)-1`#1jU5k$sF<jcJ%K4cI@p5v#Lsb?$N;m=!^0s7Ij)8Q4cA<QLS2%}kd
zc6F)VIc-+tXCf~wrM5ZqV3I^jU}(KVn260G<8D*FC40VcpRqxC8XqypTM{L5FO=_X
z#=$A8-Ge8O{}3s7(IvS)f5t%Y;whQjociWk>qk<AoSA~e2>))Y7eQvj9W4K4<98*X
zt@h}^?<J+vO#-pJ9=(1c%)h8QflU)>B++qy{qy%}5<`OK#HP)GcOz?AMq_qIqa_>Y
z6c6=uiF4fzq%*GLuEmEQ27gO+4N}z!@UPx&Zc^i;t`>Wuwnjf$!RQcH<Z+tHK&p=A
zju{6b`AppeOV`XoT$Pvgl9cjFzu&h?8HZ?9?i<e{Ce}qx9v;d4?1qI(#F?*(!_ZF-
zlI6V2iC|xZJ>@GDjD`>XM5;nDLJk0lV}*-B%UYqTWmm#8;2@lc-4YL^;}{IbeN#=P
zY)pkbpM`nu-2}6Nf0Ml+1rvqvuED7qSl@Jg={o?mRBTHEV_27rI|OQ4Wj^gP{UWo6
z_VVicb`?{X^44>rr{2l6`RsRI%`3rc)MlDEX}t>a%1~6Y3(4M6cCJ1{dH#v5>*Kt>
z6!%x@N&6qmyVY#<eJS2FqHCBPCfl^*)9Lg5K-f}tXJD})!b@Yb^837jIGdyA3?+#P
z8DG$W&>qe3V{hBCu$Lp@a)xJt=CG?{r!n@2@A|y8pRv-yw9f?j-&Y;`R)~jg<QT7d
z_m{sslUbD|d{k{ufV}7x1&hWVWsHaQ;`*6|!}~@9J9;m_Ykr^iID($C{_)L{&T4iD
z<?<-sk{!+7TG4#@J{KncTY;eb^hGV-CHz>qAjScyaS$|Gw6jdZm0%d<^#CGm8Bh^4
z@$O>hVMRi)<>MIWlHxs?gsI?bfwR}cfQi|MHgqs;^hCddSTgN(CN}CH<uN0Q@&}>$
zt+m^p`eV7$FZjPZFA~=66dHF<-t@dbhT_$4)cG`9&`WZqV}R1N>nVTtj8qCyDg;)g
zWJdWPH4(yxNuS8wV^7pyCT;IC96NlXP`7^ivSjnczK|b9;RzzE^X9}iE>5%B9;YX^
z-Vc$C-q)u2;fD6?sd3-a6X_a?mxV>Ov1SrXi{X}b#fsXMUD-*+3H{A82S-@dvk4Wx
zx=Ee;Q}ar8u76pjIsc^Aw*O1u3&8xGYYQ_k94{k^cSar3Ud@p@qoTShANrJ6{P`a0
zc$`JOI3u}qZ2E<t?FFi=Y%6(Qf^Y(8_@F{~RWjEMG=2C!_BLq)WM%?$3CSUS^mrC~
zDM!}=ws|9;j(38kn*`43dgX5Y!p>*+iZlyP`KLUx@Vcy|{J~V@I06y`W_;T+mEg_!
z9^Bpl2ZOdDp}cD=*?HI6UQ|4eCVEXfipeD%D%CoxuNFKQye-o}!1syVq(CAwJI_bf
zyYqcNWv}N-qsN~cP0>V8R{?@}NcMwn(+i2omU%?u30+>3k@2(NcPNJ+w@BBP7~7i6
z6hEUGw{4-yBC_hn{M|{}BDIs{It!TRZr5Xy?i-EoK7ZncnX%om=$*d`_gwZeZx{7U
z?E8ya#<-P?GI*N4V$>ffh>#k-mX6)J=vM6{abBTz``6L@<I<Tu4a2EpFT@j0>)#IF
z+V-iX%DiKz%dU#rcWuJ{^1Jfo^*5)RR=6mQ^ej1!K_@wvvcmG|I<8sGv3n1rdlBIQ
zMEgDd66AX337Tap5fH9H5a_|J<)Nfe1*z&IlsdzBWx|4%YL1^g$#R;;0B#afNkZp%
zJ1rk0UYcSOm0FU`ml_Ctkf2hhr?SATt~q7@!pnDeIY?`m!`(;>s)SOhNO=pmjKZO%
zM#YNN9OHMR+@`^KPu-gI`g{+wP&ZfGw~D5G)llxXcANmq=f_}0oY8aHW0{ON9Vzp_
zsC7wu8!L+kvzo4-xAW!?FHkIV7o~E#kKO9wv~W|k=qwHo79M}~Dvp1*@UA|^&NQ^e
zSm1P$bFez+wwMRiF8Kpl!YoyW-uu>XR^hQ#g-46OE|)zDRxuyU4FoHpII-jmjV2II
zK{G1?7Z1C&LKISt9FuYBXmtQQxb|yv<iaij`drbS_|)-ZdtONHE#F*`HQ^|9-NiJ}
zXQEx@n~k<u3>CN~%;jEx*6Q)vawFeVOtA&71(4HowD^4)`jnZFESv{5R#<ni@IZY8
zLbx+4kjszu7!TgN$IHx8;&tvn%v;sg+a9)<_ZK(`CziLXj9%8f)e&AK{I<P(S$L=Z
z(ZW5vGZX#d7~O4(TUW|Wv_YW!ay^foUuf2TvuGoMne$`Y_R+ISemvRv=*+&tX3L;D
zBfHQcU7q=yw)c~L{B<`K8C}_%UAfnmcFNM}J#aNN#|}-D#}z4CKy)5|tKI{2ML~0i
zijb4eB0L3I@)F7wfiePHdLf2OjugLt(`}RDrh)nZ)7stz#joaoZj*58%_`waj1xJ^
zTlBq^ghZxj`-X>2jq`qH&Xks`rdMjygpy5{pg<O_1bYs8e>Q?w0Ba&dUo4UQ-@3;8
zeEXO5*B)(27Hhy=^wfLsb~;!=*#9Hjf8g!#`>&pLPp5qziXLjeUjJD6H&@@GkmT*{
z=Su|V!>h#Hb<ExN3di}@iVqlN<Zn%=h>XOBXjw%%Uk=8oOrIgnS(?mPHv2F}!;}Kq
z6-44Y&@f9JVkU{#fP^LR^79DA^l>AQ-AurP(ij-+mLOk%=t}KnY({7#0Iny~=5-;g
zdgRzuLCF)zKwL5j_5LW4l0eo(wY6`GuZcayQ?v*!JO$dj(-ijgy(vwk+w6(&Ap!<t
z<ARa%4>0sN0e1d(Pt>i#U?L;cW3G@@WbSa@X_@icTw)?5?@q+!llKJ|q$KPx^3#Sg
zJXNnV_g<Mu$r8jFAAG4lk(gqqAmSE#Z>l$e5Hg}DucI$rUH(y2t!@#rp`}3w17BMz
zj3@W3U4GYiY@8q_a~-I2Ujz6IaMi&2*R-YUW$Jh69YqK-5c&x!zpPijRe^G(evDg3
z-PK&hUeGU4|MyFf;KgC8wOpR|#JxuI>gfQv>yH9jHn>qwx#T`CKr?qn0ioqh%foaD
zO+{0^ePU^uNOiAO<ixf;9-Hs1ZGlwf@#eKczfj+PQ^|~r+AESPFRYZxLTIYBrG)fs
z+uMdcqp}Spg4beaEk>)NhTfN22)tg6w9meZ*3WJgIsV7T(u|Qsov#jm+n9X+9{D|2
z_Kow$m0KuFOp1MbPpb~lVX+rfc;kM8^M6~gWdqWVvv~tK%OmlSd<V>IrMh+ExiDEv
z(A?@!`dT<<=)lu-0d!ZR@csLs1G&#Yax5`s4yut9rp?P+ATb5`2TLRFI+{#^wHfcM
zuEHAN$r+L~gIfzskg=)oUvW|>GQuGQ1i(Foi_ouID^cw*Yg>_7G4nW?dAD8;%dY>6
zqMYh+T&*|4e7;_{7^E|Gf={B`>PO|1LCimBfl<$$pUcoS4y)}3=#nYF(?!kianEj;
zwc)<`-hUTq+_RREXkDI?@L$FAs-&XgS0_bo7Y&2}ZW%{`9w`6-8@=2-4s5@nzHYp~
ziBalV6f9JWL{cY1j3NVR$fXGis5y|cKV2?T49L6WpU@zb^@`LRyTwgn*-E$h7DHyA
zJJa~!B3wAChORz$_+e=MG<<UtCABHgMX0PLV<6b>q{Ex=Y(>U{SiXt6Ad?=$m2&}p
z#2r_>Pj&}{34ItinlRjwr%^Z`F|P4j5|e)wW*-2{*7awR6$KehaBk2&Sv3h)Rc(*W
zfZ{9<4i+hZCQ=~B$lnjdBPPnxFMJ&*t-1}X%}uy)v=gLDSul^xNk{@jK`D$i(NQ>T
zS3e?AjP3`LCg?^WxToD}+*=Mf!nK<I;q^nmxyE{9N{@z^y=aVKos(0as2DFFzMb8$
z&Fz(+6OY<atYW1|o4RGkQu1%+nMT>tL)Y)#YbWKC!czm+b2H|wZWE)jo^uYIlKIp2
zbMuph9gjHc>n1|tmvsN4WY-(H`$2&8D0lE);C-AZKw1@`v*Ub$P?ob5Hubrod(=<G
zT(f~W%*@?qNSCbR<Wap@bHW+O>v>ClGM54Nv>SC?y;0i1qNR-E@C$k^Pv@mg6*JG;
z>~)3GKkSu0YNKfJw;aBo-*>4%asiNguxNc_`gpd)R|C*5u*V@}*~-?EcQR4yB}-Z@
zpyFEou)Nd3R<k4T-5u^vm6<Wik7CC~+&^PfE#Jk_<o6|}`sNh<ct)2wSYe-$y3_8P
zk*QD!pM;gC@>;~oJVtV(?Z<#i&@}nfAP{TEKq3JWRlD7>My73ZL3R%4F$2sP^J$M)
z-8YIE6y2di{V8wgjS8-->Oe?iBcmpG0WuP=p|s5t>g*R_!7J_GhDjK6coyU4+{owJ
zZv!bisLVkI!ijf-N5!VAt82y6Xfu-=7M-uPx|H2SzdmW}%Bw?1m-*HY{jo(9gejDS
zl5)rY>SM+8%7)4nUlbNqW^)Au#nU9mA4SIl4V{B&leRsGW2O!-D&|A!Okrul{um$w
zc3dnG`Leon0kUub?KA~w3n@jKb)42j>mezmbtF^ILcLDaNo2`J$LX6TWY~-sCvbJ3
zNcufES0S%%QkA-YzZLxkCDQCgp=@0!(#v8|wC6Aa!oT(#mw6&}B=_Qv!e`Y6E=J>_
zx#OvrOQyb{stvo-VirN}XYO4(4jifb<#|n(e&65{0bwg(JS>`D!)%lD8W;?Q%yX9$
z6p?m*`8t<8H~L3;<>vUJ;w`1lr}I~<Wd!tJ{@cBgK8LRy2QBH^{O^f9nqh+mxdiPc
zK{!=mE{uS5$L$~-#Dp`78(=70Ad>O<uY)oQDp_ztpc%+o%$|ZZ1k_MgEG<*u=7A1i
z<YZlXsB+xH(wwJ0V4MYjP=pNzsQ{tx0yT&>08s$R5m&Mw7;opiE4>BH9YkIpRUDpE
zF!23gfQS<oc7*^3Nuaq!Ke`A<v#HxCTtIt-Op!cj2+3L@$w|m)b7C-OfA{Z$4vuAL
zBL~zyqSyC#2*tmN1I3BjghHW&rkbeHXYtD9Y83I08Cp(=D2GwgjyeEYLewU5CEioP
z_%G@&>JNQcXNZK$#_Qm*Y$Baa!j`17Bj{}$MCHqU<RUg1$0pPJ;Qf%q#inAT;Qx=)
zNJsyoR&j&MPz%hoq~S}lEh@;B-B>C^t@6-rDoUVo2rq|1$))<CqWsbsG;2IDo$W^#
zAz$-*7qa%Qh0Gx*ec*8d`c~8dR4O(F#UUr+%|#<$8&wf6`_CpkZV1dlz7_r*n~bHW
zk0L)9M|(>9Q2*bmjh};6DTrz5qrXT;YX*7PphvNqBg!W0IFVbO{wVwLu>C5JdsiN`
zqvKUL=q$G35c!GHlj=-5UI~n%4tiH#t0X7+(QT}`q^z}5eVz$9tRKg+Ic;6Pd21*5
zWzCP%o10;r<_KUD4jBge|J%eyY#PIRX5v@`Z1cuQv6ZNJlrX%@j*3^J|3$5kPSseW
z2G#zeYHN-wB`IR*L;RHtC2A*kB?ATpn?9&iO&=n`rxLSCII0f=FV4VHT}`kN>uhp`
zxHY+w;fv)&DTFl*wVJ~iEU*J=<ZM5n{!CO}sR652$vQAn@O@wv8D2_7^bM;{^6ebN
zvA~2^$o#Qk(V;kzcfl-YMEpfD6zSHSc=F6+9o7NyXmZIQk;;bGPGLM5#**4YruS0G
zC4FQ%3y-D23rr<ruymbPvQD)Y?IdT{Nd>kX?l_DDeArYoJE9Yn@J;{fNxB&n_5Z-&
z4<X<5Kch%s6qpX;GU%<00K_OZk@i2MkfE~HaV4)tOiewcTf_GUEa9Ao2->&hXfO`s
zbUd+=ZLOl?;upnD?^f}%l_Tj;O>is{Jv$tajYIp&IkApwzfg^`#vYfxjQ75Vvc{bY
zuf)UH%v|H@nKb@R5`Mn1reGL`@O1b(!p2D9m5Sj#8&++xw2hcUtnx#w;z~iGQs8X|
zcw7V6VN+A8bZZp)Y(Rz1kv>upjf|xosvd=zIuM0dDiX$*93o(|$kl9;&JejMn_M(B
zM`-?*se&SAtdeWmY>8CDoD&0&isDBJE2SJDz+lB;*`?V*D%vb$q_W`)NtOQ@1rLgs
zFRodIzsK_GxkmgV!FD9V7!D#PA=85A>egWgOfB!!Ie<c|20YDZX9vLa;^xRDR6-FA
zuahGAXq8OQCKR#pICBm<l}L55Y|Ej)XOq%l*c-TnM_XO7=^mG<116|QPgDwjHPw$D
zv5-!~Hjo)ytOb}QWW(=9z6^$mm<|>x)vRue(k2!=VD9V`eGu%Z&g7zQn3RH-V<RRt
ztf5ex){5jO*>727>=0rMy$`RQO@>zqi%{K1)?wl4@Uhv%6p^zVTh>~zSy=mB#2-AC
zQ(hDv4Ta+3<dD0@o6c6lkt<4EIOG!fS6Pc23x|wDzIDSUa0hs$GMKt`hMWk?l?>Yi
zOk$Br;2nrb$ohr75HJOAL2Ob9Xso$FBWqKUwOpO|=fA1araAnflC=M#$R%u;Ogx=N
z#tjkZRDu$CkH0ADfhUu47{ByqCOpm7axfm8sLCO06bg+M@~eDy$Wsf(X=A7u|8Evd
z6<&U-d3BwEx(P!KR@DrG11n)6b0=(zSk+0zM@)fBK*8D~!;*2X`lizd-wZQ0Ix~RB
zu}RI%1T2fNuacnC=T_Xz(1w9tAw+QyaO><D_=n_3xlO)uDJY!~G*>FN(i)A$^Kwzq
zJou?qdRhA1C}NT~vclm7v*!o1hn@mt_zGI?veSPf#Q>X%V4T1vydU`TE-GH*4}7#7
zmTE#$d(th@JcvcSk;hBeD?|7q_%j<%@77-f2M6=XLN>OoS`iA5e9IKeW2#mBxEg2o
zNrh`@u!QTHis&`84t00l)L#_7`VAL#oxRYvVGq2Y3?@<8)_i1SIn8n?&$)oX$n=3x
z`2RNwIbdGX@ZiKNBfzKwc%`SWtiZrObT<6ERrrs+n@x%UXUih2mt_QG<83e430vcF
zSup8%9XN}q2Ub}$G6{>uVzXgEm~ajfu!F=rVm3aC&=6Kbs=-n#-wwi2hHP^5hFQe)
zA(U$X<bC$=8d8T~L$Je|@JitK{x?!s6dwDMif^u|jKHFV51_Gw-$p(nPf<zGrs0dS
zBZyffEWG;gxL3_#;00(z@CDUWcq%*=e992sR9N^pHYEz^=kTb5N|nbscub4){L`tn
zY7J9OBjqW%)VB;+7}rp8$mGFDk?AgA(mF|wV=bMHSQMItS6e0h&mwCq3N2?x#g~Bz
zY{VoEZ0g`BeN=Y^W)1m1*n-+rhcz}{d&t94t)e>X7i<wY#=%>FWw`gRgnfa&4JE{(
zld5*-xQdRoc{={|;A2iMAs>1V?3|f@i-g{Ga>7zYVN*D;*<g_x9R`WKLXr~%x8MqY
z5lF==u}N51-^f10!tW=j_{K(<PS`&jSlX+%e^I04s<Y%NDjQqGMprWY{-Pp=NU#I%
zYT$Jz?_&qRE0E!Fi{Od#rlL^r8Jv80u4^dNHET{dv4qvTNlHx^-D`18tC#;|?*UJ|
zVNMEa!y>T37O_YnTuz*Ai?{(8Jg_JeO!@B@rwt3Q)|d_v57n9j%N}iGy?0$_Ox3K0
zRk5|%Q|oMe2`zv@tfs0GiiAnh#)N7%zPg40!(Yw9a~`-R#ETV@(-wgP0nHVKitQ#=
z+7g@D#?+YwuECXR;pUV7vxfr*6NbEOZ2zoSD-X)?u<>Az4i(weh&^C9@{7}@53g1V
z_FPM`M#)J5cxoLu*P-`e9bW?936n!37qO|*ewD6sB<(B*+biHSNpQ5l=j&vRavs2P
z+R?#-3vob9@HmPCi}%G2V$*H0=>ry(o$1gf+aQN5BMxi|8V4h?VUlauD!z$<5+A}V
z)7cyQWMk4yHVm@nbZ6N{AjfP9GLf!-Q4l)80{DyaqvEv@=Z(w`E;ao)ct%@1w-Sq_
z<VP;*bB$nm!@;+OO1Q==M<qP&_7>u_OYaqUjN(vmB0lCkKsQmUbo8c2DiY+E!6p<K
zi;Tv9HgSSipNqJRN2L$p1Hh;u{N+`W4j8}zYZ-;YbGonsjw3!ZMHohIYcp3pdkV2>
z2E1Dzd;}i$7M%f;L3QQB>-3SH43TSAN%SG`6D*QWIss*kjmDPXJ-MQ-QLv!4zz6ZB
zP_41j#7ZY?9HS4<!GN#>Rv8vv*>I!~>8`?HQ}b#i9@e=nxtZ+)zyC#r>GXo_8%FVR
z4fQ!O)Tj(4It!0NaY<Vt{s@M+5;;I{Yy#MXy<IsNzToPF$7RFbfH@6uUHFS4_J!$W
zpM~X3g$UU%g@eo`%zsVj0M0}$C6%Dly_?>P!g5Fx_z;`+E3MjbGO_7u^g(jEi#XyT
z{|F;Cxx!Z=?h@P-1Q8j><v8SU9CA^3l=UrGPnAgcLPDYVrTeScQwxKL<H;t|hw#`e
zybimPP%~%KZ5_WthAD&5E6HF5`9K|<5oL|qR5qnOw2ne!qj+ODc;&8{RCi<R@k-Tj
zi$Lq(qrfH(^K_P7EG>lIpU&Q$?vlX@O#g!tCg9$)XTTfJtb<wLpppL#T|<+eMq*30
z0FQbs{s1b<j>?9EonIE0e%6nLj1~O=&v9PxiTG+7K?l4(MT-%^KnAW4Uk!%~_=!Y1
zneH$%k0<Yx8zkg%n)#(1ND<~h3893$i7}jPELwN@y)_u+sl&3nW_w)DFn!3SvX?&R
z)%-_zuj@>+2YK!<>Yxl8In^TmqTEuPNZ0{X5sh@$`{jTcA;pR6ikCy7W$TCV^xIhK
z5V=l`j(0`2S#fo4H{58};grhuV{zaHuPkF(WbHme(U5J#FIVK~o?0b5-o!;U{#Kx~
zIt1<tjbq4hxf130{*SKrfp6lv^F>EOP%DJS5m~yFw>N^0L})F@QlotK<|f#L^lA`O
zx^&CdXcLoGa+son?q*Z6ga`?T%?-__6(cHwl>$wL*|eddvN*cIK}z!SwkR$)soWZ*
z6EnHHG%Xtx?E@mk9whGfjO>u!eUA~+Xf$)q_x%3<e&@`1{T3?OV{OVH3kv#w{nkf`
zLQ}h+)_US|j^8b%x_w*32lssznDot8lxP`abv~*!u0rw4J}G)mWeOlAe)~lw6H31}
z@J>>c-Ek!Yi=7*LM2e2r*^#|syi2*3lptz@Md?My(9t@pWG4E?^u}IGT&(Bop?<+0
zyMHLyW>JHOJzwtyAlvNM-aH`2)^~NDN)!%ifo!$&Nl$w2Ln%}5cQ&v^3D(H#S8kHc
zqp~wDUOkKK;_1mX#T_MU+Dq(0+xyyv>%z_fWl;6~JSiF8W5Sh6?`Z0gArRF*<FSgH
z$tk${J2%QHRVJn7><jDqUSH7HCa4{X$hJfmT4I)0u9@bQme{+l?%M!JZNk32;yboY
zCr;6K{*(4m{@oSNcm{plbh!8N&+DSb8ZbU52L9%zQQaS3{lu=e#itYAJ$CQ?)JF~r
zvw!quC>>=HJdk$qAS4^a;MVGc@O5E4lO(B_MU@xzo+MWh)oI$c0VzB*@3qC1Q?>Db
zLG<?UV~uR&!+jRvF)1&m?-LWJn_5xgxzS(!dQo}fx+5>CJrCS|iZ<wJKHSXcj^KJd
zbx0>T!Zz=Yr`L>n9!+}Y>~fEsTe95+NXLbnl0Ri~XK@QjdoL9^d;a`BDV1emI__kW
zctlKh%}&vhYkwt`l<5FjrY@IuUh6m8`vr4B+;`e4-ooS~wakXv7)-FIxT9!=&$YZC
z6W_1}9mmpd|7QLxS<U^rEB&uq$RV&}u4b+ywJ4?7<^;<`<H0A`>gRVYhmM~JK*7>x
zb9vGSZiIsS<@ZMwF|N27QSA0c?V^ZK1YSl6J!;Uc6IKfmX43XV`Vy^CuPDHpn1(ja
zEQXg|+M|mI357iazV*;H!M^;Q#r*l^k>63vs|&wW)9CASIzLdHRNJ~Hav-mhW<wke
z-u|2zQ#aq-wK^9SXYah23eZwe^?qmP<8#TNO76<Z@2l``s3yp+ChIGb6!3ujo=_hw
zL0o|sfF+pQ)OKcx^YwmVq1{Z>8rz<_bkWbqsRP0@;_f{W@A_`waio$KbzateEPdP6
z14DFO=Jfo3`sXc9%C_F8%+FsJLI{LMJ@ZPUjd`4Hrsl6KZQDvtnHEIV_pJ?&Y*b*N
zQ)vWkTdJ28ItIT{FM#8s3TZ&_B)N--oFq78eC_uvh@78Dowgq^S;hy}A@>{<$+n{M
z&O;*9<Uo1HDqPg%a2G{&|IoLd&18P};-WO<rziKe>}cZV-ZiBK98z*MePi={&2oFv
zpE>srm&k5mSxNlIG%jWjvwyWA0g;?#(pd!0q`^@Iyy+Z;{Z<B{drWaB4Ro;RNoi%l
zIbN^Wq3^<?UpDm*2i6^5Vcy0biJ9D`ZXy1d{su?%84<Vg97-Jj+td-^KqoB>o-N(`
z{xWA-|M<a)pX_8Lsi>@L?#VN>yy^XOvz=u3ZmmIga|*Da_IG~`;|EG8sZl9~W3SDD
zeMU+LhH<%$E64?28kLSD*A2pWNzQe2JnGrq-e!}2cPP?)bFRl+{%uSM9JWbx)zWbT
z&zDpC#YB1R&A&bTi_kYxwKKFRpIEabwd?N3XR#IcLUgw;XYZMFAVSx+0Xa)75ZsK#
zz!NC_)8`wArbN($zsuR6D-$Qkjt0wc`(zLq)c(?`xD1==9@w%1RB!CRE^IdKp7D^y
z)@?^X)DG{z#2#OkfL}|CzbJAF&8y^7&kl_JP5kiPEB`i=m=IOPy>WBq^xfBmnXGdu
zBYC+q_nPpyH2BWU20i1rLvXSU8GsNGG5qN8gKQgbhI1#Q-k&M=y%kr)`a}jwHF`#5
zJ~t$2R2J3yQ>ubkf+sKU-DF9{x&@vv31UBrbk=!tK4F%IwtDPE$*y);<}W?`!)D^m
z#A`Ed|LbK*sJZgUz~;`RZ`p6z`|l$(BiWiDkRL5-Gc{`33c<2hvYiG7f5I2|(F;kE
zQDx{jI-D!IUIwF7;>b_fZ<0Eb@?cS>5FaKv2jor$->64U=zkxZY3cV^7VHwAA(>k}
zV!Pl(7H$)d2Je4ry<~yrtCGFF%QA1==~<U}`D8ewD$CP~JI-olUw9JQERiB2wuPN-
zmStDS3xcJ;4K`o(A=L}qb7C&Z3`x!;sU1?zoPhG{h_MPORYtf(<8Ump8&;9Y?sA02
z;SFnuCF<wSS)3asaaT~hNwOp4RRbxtn<qTa&3plomrbw$k{M-#ZnrO>zH%QlmC#zn
zJ;wcOFbzmV?Si#OP=wHJ=3$;zSfYZ8Bzz>Wa)pog+zDMT%b5ZAW(NYL47@Fq+$!lD
z=;agy2SGc8lVpgF<kh#@cyq~rvYUnW&wJ1o59m#b{iofX>ycXsmZ%j+EI`J{ag?yY
z0op8yb+C&eyd_k2zXlqCO?bIFyAbD!d=&h|FI5*1T7(&vK_(|8i9M7mDm%|BeMLFN
z$Y7kLl177Nz+yW{`WytOTj;k6s<enbBt`{GJEYs>ey#fp58d$!@I>1TJdzMYJWoJ7
z7a!7jz|hf9CC5mu)EO0oZUfIoc~Q4ulk5hb1S4Mv*ktD@p#uucAeb$xvY3?JjHJ%$
zS5<N(2`(xZklJ#qL{t-vJtXdZY@~8>wQ!$vy>*e=Z5Mdjy<8X#T{iR#2?EA$mMk%o
zz=9UFYfdK=yUap$cu5_e71=GvgZS7hsvG6wf?%`Sg$^(bJuE)DQ865=5~=GU1da@Z
zk_JJOR?Jy%0O6515GtHpOj&`uidGC&Txpbx6Tf=$v8K%*jdcmDy2NLiAGcdgKUsLo
z3m_7bF%C+M3BsO-R;$t+3>u&!U>aIR&!HOQXGI~)n^d2CSO5nFp(E|XZ%Qb;e);(v
zyA+pJfI($Bl~i~*o3GF&qbMw?lAGZ?LP4%FcW$m+Az!cb*u{aFwBXoxUAWu(QH@pe
z(1Fm4f3rW`LWKOz3PhGCy#^wKOyL@u(L-4sswO@sSYm=&_JB4ZlYkt~K{=Va!rA{c
zgxFO}6~{#|XMphjlQMb+c~WHsSa2NGC1+7u+$=n?UxxwkA#9?Im)>c~fw-Hkha_E{
zGx&~ps~De#WU*QKm8!2xl}yeg&H~$D8g*nMI?1Bof<#JCH%fLd;q(G52Sqwk_IRII
zu19zlEO)?2ayuz!5OtRoH%F*&(FVvk=~U&^*<v-GK8@lz+@3@IHWDSfC1zG0Hwkm=
zll|`qw~9kkfIOhU%{DXK1d@8e!U6e6jvH4#Bp|A!R4__NuaQLS9oXTBOl~QFpzfE!
z4;W5_iI(BKM8q6SP7Ppi5F81X(Os6gR86Fc088-He8je*+5=GxWl1XEc)5PzP{O;c
z<letpyb^r6`{ubF@#UO#7C?y!R&k&QSuFMikd$q~e-KIIc_qmeKt;P?a+U!LpD-JM
zWe1_&^8EMU3%xwXgtz4`jL`DB3LSUH!*|uWO_bc{Sp+T(%#y^!`H`jStcy0OBXJF+
zYE~LKR>?!wQ%13h*xQAUCtXPc74k_T1{qhGvxSNi3*(2%0ak54BRL~e1R4<G6I9)H
zZvayJldUh1YLgo$C*-FnMT*_mRC#b<3(2uyVx*Cl@^$^vWbkx6)W}I3!-sL=E=ng5
zkT<Ri4<k9;Y?g#}aKL02tQ{s-i&kO`sBE0BB9sfPTEJ0dxyAt+tYS_B2gNAPkU=G!
z85IPW{{;kB9)02kLV=<P0pGF}3@j5aN&{LXwg1|is<1{$6=lRqz}uN5h@{ZSvO1)K
zlTTNU7Cq+E7PH3^HK9${7ee1zE^+I`357m+w}51CLCV({scQxzURr`=iZ+EKvbJQ(
zINh_j?_UWVN$UhRRJN#eGG1z-liogvJ0NgjASqm4l(O8oGz%aWSkQz!WFSvYdT|z$
zl1g9GTyX83yLP!3^g@G3qj<ZSC?az^ChcH&jsg?aGAgJHOg;D&K??}PsSJ~5!U0@+
ztMez4dsGpVk~1mO%Mw<*UK|`yoTIXP0LDp%{5v6LXiAkK{}56*-ezGXCsGBrrAAy-
z!{pvq;n#O9{uLzJ>NPF*`ME3^rC(7c6#aQ1!jM*SlsCmvh|97}RP`xTF_73EvreIk
zdcOR`X>bZw7J45M!K1S5RAu8*$qhwhP)Cx1Ur|Yokp|ToCuoUQ?uk1dV@jlCpWzo`
zqP^W?!iB$Z4K4Q?G9oFwyEyV13;GjGt_&kp-e?I{5K`q&LOC!@6$TK&gaBV1g{v_k
zDExBc`GOc%p!l#w^~q8v>!DCjCi8a#Sw`(mp>>QFg{mg$Jg7BGR?%j;XOAUzzo5=M
zj@b&_4=#ykr7yd(>m@O2g%?wRD8&(pQ4VP)>XcQCh?X$EJZ0D7*M&=vZH@?SukP=%
zM?DrpeLvUy=dE*tst;#FL>g3b7G;=Cu*uCS3}CQS2ou0KZc%H$*_}iS9ZSTdC7QpV
z)XtzB)5ss6=cliASujVi0S?Xzr>gkT6A(p3L+~w-B*QAfpcs%UfD|-@0+5_cguIzK
zu|Us1r~N40Cm<l#g%pb5MKV~1)t1C0R1kWG;E6a@l8U3nNneN-z9R|opk=M*OAX&o
z2gpZxEpmAsNk#j+1e?h|<Bl2I0LtV~pulE;qA?E2++|5Hg`U@gXxhL;;Lw5tvKxX5
zOp@5&E!f(?Lx5W7yooL_gqEYapep%Ml1Ul_%IlG$4B~eVNS$aN%8|qlsn#xDJ+DKZ
zE0f&2Ybad-uC%<8vY>!2NcO*cY}*?g7e*wJG#?#q+KWM-I4-I4JPSU<ab;Va2@Xhz
z79prkFAp<gpfPkoc-Y^+oiK8u$@uknoz<UU+r^}uN&=tkWXiVlu(%vwhHZf<&`^V}
zQ1;!+$_jjg4)V7O4PCP&6_DRza(`ug3l(O)R1Sc0Q)EaolW2kk3)W6htHv$59#j=T
zN~wy|7ofmnpfa1j?R?9J;K9LlQ*zeMq@ZQ&foxjJly$`0U@IUGOPxtF2o+Q%cUkHj
ztAfSHKw`LyLY|gQM~-8B+{(;39^J912uTO27&mDEHCQk}c`T||q6fh?X%3>CN>1mf
zz-7wKpFGF{n79-`Q(!ZFJM*@eWEDpZiXhzuU}Y&hEHxnZRY1B{?5!lpUDt&(xK@lS
zVp;CIZ>ue-Aoeh{8;-Gd+O{=jLkzU`)u^yfSRz8OqBqHqZ4vqlR4w862xc$l7xbXy
z5?^z0y=rY_7^f0&1|bmLW#_msL{m@dm#~NM*m;SQOGRyR$A^gc5DMVTl@zry(wWdc
zq<S_UNl`X<sB%h`Gin6Elm-WeOn-<Po%mMu*ad_Imbm}6v4R+#jq4Q{8$g5J#jI#v
z==YKgt_n&PI32Vwy*2XVI60M6+{{7aem=v`!R^TWbLc|S=Lf1XNew6>dM@`iR9~o8
z7*x-`uS$c^;@xS4hX<6EFK5rBZJ-z=9(<@^w6r~h-YQ5P3<kauIKBZSP~p)HamD)-
zFfq2k0Oq#>m@mR%8R+2HdHg_NtlbbJC;%?t$RgrzA@sPju?$d>H~_#%w0xng7H%>!
zIWuFGSE$>EBuk2s6)z>+DsQNXO8doSe2)j+)hp<WkzvK1?7hU4R~CvbaOf^G0NUXd
zG^i%Vz)%+1WD>-IUVb&0#Xq>k<Yy^(=Qf)M@ZhVj^S8dvgRC<ROx%`qmP^j!j-40M
zr`w>OWN;Lb%N2<tZJ?|nrrk{jW>0U%ZbOY!uH4d~R|q{bcu0n~_=6l9<CK%p6&#My
zEjMSaV403aHt~FBx{({76@F%tpAVx1=M**B1tCKB1|fqikn{Cnx&Ssy^<5Z19O`hm
zasb!7><b)<Ov)*AK+BlStkj_MoM?EWZfmQ91RBg+18Gkvi3Z3L*EGT;5OR|67p#ZX
zu1YRGnXpN49m#3aEnIHxoH0&>m_npK0r+=~SH9Xq(gpvat!Kwq49lI|Nz0uB^2)S4
z+>xjjnLD|zk=Lq~oc0pDx9-rD23CdeCM!*?!gi1vlv*d1xP++s%&;{x#;HeAY4oAE
zr?J6enSC;6$R-mxW$@Inh{hWfa0(kY78sKsL7`xNjK1~&tPnyXKL5i+?NXVDqcSae
zw?Y(?gbwufjWtIehN-I|>dL1VtY24LJ){Kx$LyxY<z-^Y%Q>xrOZ$+)B??R0EF4>$
z9O2mD_~k(kZ>`Ynp`VviW%ex-`5B7u*#<ob;x##_8Y6D-hM%{I2Sv8;3acLoU*Ks_
z!LMyqB{xD0Ra$vnDEjgYp5s(b3+J^^P2;xi!-0`EtoVa*5Z}~rOWyKePrk7@UvM(d
zytw3!D?e$mGmgki=CGJdM_~bEM7+*}p<rBZ?pb$t4S=!}$MzrM(VVl4ws0nZC5aoc
zxCfl?U3Bx#f<s-ilw}|#Wt1tEO7_evjzgsE#S9;0s;Sj##_UNL<jJOvFJ-T*J0|)V
zY3h8ORQHc@6E`j%s<@-p#dmaW0JV~2*-(o+KxP<+`-pQa6ih~{ZW5{M;WV&hdHc1(
zGUF3Hfn`93%DUi@MD?$QwGVm&G%_i54n)Y+Jv_>1oMhJpPzIhC=lk3m&+jqH<~o?P
zN;Rif@)skp1^hIS)956_O0YjiYudFkvKj*)la>u}WO$)~kb*;>O*+r+*g8roHJCIf
zzdS7=W#L<`G+kV%aE&vr5>`oA`SZ2=xAotz5)9x3r%{D_+jt0MVZmXTMZJO+LOC*I
z2%A>rjrrpMiXt$&-Z^%rlILIEkE!6ufEbgAKRa>QdM8LIYEp#vq&VS>0dZ9693{?Q
z?g++}@MwdJXXtWI9SV*X_}muXE~erVygx1yY#g5LHya8h&4M39nu&rd-#0+;_aGqv
z>q-=qsW*LK!c~=1kRgAsjbV*hB+_CwKu(;h>62>&fhnRI6=o<vv}nnlWB3Zw+>9Hv
zOOjJuN*VGz<-hztdNAoD#oI$!9km=x+ZovMdL&yA;|l18=`Ao8udh2}=P?)HGf0su
zpQ&rL0cCMr8XVr*(;7+^6NwNvMlSn86lC|IJDv}M(@kJ6xi#W?ITKBwF+uQdN8<~Z
z-1g5*aWzBlBSJYcOtPFX`SBw<fnp5i^{pCuyFA0t`bu4cd2GOzVhO^ys|La;i&+E~
ziQ{vtI}T%EyPtN^YQ2__hqt_6<}#d_+&VJ!K=#TOH^c_!PQ`4JauhZM0#}7Cqo>h2
z+eRySr1wdwQx)3HGK|{7C7qy}D;xPJe<q?6s1?^{r8BKK^4jk}ER6KBaP$*5199Te
zDs>|%tf1n)vPG0hj*b@fbm{v(qQXAiRot<aJ1?Y>8ZC!vqcSY97?<;9C6ko1In6MX
z6}*$ngn%G&merav;O7Pq2B8>0I^)uhyMzt-f@?T(6+OS)M1@$7fKjlocV)>6s)KF_
zN`+ZXRKv6u5*Twx$8{lBLDc|s<C_j2i>BXp9{7I2kuEq{C_f>VcU%`nd5s7Ch{z3+
zJjdbn_z)!sOhv#2F(PRe$zs+BLW35hj(8C3OfXmMhUxBhZ^)O3u2+B}N0ntc{fYe{
zI#%6QQz&F&C`_`i3BAC=wViL*YUAOozLHk?)Zz5j1nCH6Bb|GH?sL$QECUTFgi=L;
zm55lxa8joGjYc9xt`s<(zmRl-n;0PE1sq8tw~G^}D)zhf#Z(j+w`96(V|gywdgG+t
zkyBRU)ZbyW$&syB<hwAS+S`@vIpID%{0W#wRtx3XJ-~-1Z)}hk1vR47xu5Z47Y$1Y
z!~_CF)g3`BAUP><5WFuZD2Zxpl4p;uCTgbq5(5V8A|4!Gu-Fj4c%uDG0)?q(-dPY2
z7b4FbiY#k}^G*Fx$ocROcIwa=R;^VlR9_@SZcq%RbC`q%7}cojktv{sPR`)1;&VT%
zG2%bVFj$gFf{5G=hS>uv=*U)}h4Te*G1GuH>J7u{VIa_4G<MQk&FWny5#?V~2rh~2
zT;J>Hn%6sxubK3%tV23DR&kWL^VOY~<;8VagHRZw6oq9ss0JYn<+YED!Pi-YU((1G
zIM<S%s=fy<kx&iY!zRm_xIz&|{5UiLf*@jtl+giiFp0v<bz!@h3T8KtgV))K)Tn-0
zGNZI>>$W__dNB<P?XF474GJ1`7~>t`DnwV2cm&odGAUz0>m@Zra_rHrTdKxcH6zkN
z@4(0}09L3uhlj<jmsVbR7CuI)P!8AK3K^jd0}{SE%QE7X6dXE};j>XaF{lTl8=;WU
zibX}O!f5`wux&IH(kh|lTON_szpm??`39X0E64Q-)+Jwv3@DB|>l?*tO+DCUH7Hd<
z3y=sF0avP^cmvTI0AiZrAR~G(a{1W|ZwtX^MwRgDoK^`ZBM?b3t}L8Qpr6iiaL?nB
zj)aCNvor1sryn_!V;{0Dm7IoH1wpS6m!1sl??qQ(gAfqkDTuu19Q!@`&N9iB*E_F3
zY}nT4=5Tqd7!>Aa!GIgiX`mr21yX5fz5e+7>YkNovoDT{8kXlk-axyS1>!rjC;{l;
zD2PbJ0cFELV|lU{<JCe<h6d{--Fq#6ExJmw%dvh>n<+5Vi`EEw;2RL)Ez;EC#|E!H
zOGl-9_Oqjx8NQOyC5{|cC9rT!tJZW5vytPc<I+gE1aN5nzSMzZijze?vg#oKNOQI}
z5o_<3I<(Ayyj`SnRu3L5q{=%aCE8iz1_5CQixja~N3R%M%$7LZMvWb1c;3w1b)SOQ
z0@-5SS#+1xX_+(*>dSIK2<0g*S&`@QWFxDiIap~VYxrNuWB(v3`^ND8Z@Xy^6BDfa
z-gD-3$wl$<2GdeULUNrf5Va?e51o`!aC2$9wPU!!6_{UGLl+t(D^%Bmv^043Os%|j
z2J^)i-m3i(LQZF@1=TGm8!EI&!uz#jNA)n^T{>rXHnmVq3uAl$r-d+HPWcw~a{z+A
zn!h}3on3Gyx{|0-CPsoUI!7%}R;-H_NiWRa?%WFjM2M|mN@K7XX~dNbuQTf;+o;fn
zi__<lHgdIm(TSy-RO{39A;*ed1D|e~7Cr#FE`sU)w31~?1matdw8&chrhWH*K3Ha`
z2l&ZZ{0Y1!f<chsy6{W8=S8p;g62*lASWVI2FDY}@^WfGc1Oh#e=HhM3rnodtna#z
zBq69i(;L}~q$br)8l^MmXh~d&VWX#<e5e1<TL7nwU$v>VW-*{7S*0=g05>HSeF|jb
zoTdG&9?I#E8m>jy1xOGVnE7UCzJc=ao@4wInn5z7%HT^p*YmnCq*~<*f!(oPbAI*^
z^*Yl3-gmLHs0_xHP`>0ICCh5CRoXSmS0g&pG688RyTD}pwJVL25geP|npy?ioc_k8
z48-GG2%WUH%Etodbf`8}G=TnT2qg@>s88mM_JI7CKJkl^IK0S8QRMJq8J?IZZSE*s
zIOyw83x3+;!h*B>|NBMhX!DFXKq_!tW#Y=A<^0IjEXt@bol@uA3S<h(r<hHDZeXQ&
zZ*8Cter!oV>j;65Ywck~pfFCDu-kA!1ls+{MG>xuxR5mQ3($hrq2}*}z0YRKo)Rb;
zOna>966Qd3DV)c8vBZQ!{k0=Kxpb<DYVegK&52jheHG$_o(RP+HMT9OxXX^i-J(qr
zurfV<2bWP!X>|DMZ5ks##<3Xov&zL1U&45RA&wXpA3tcHy0(Rx|JG#FKKzZRMvp~3
z%e@QT3xevI8pM`B$xm=1HMK1r^xV4rFJDqwM*@|BMk>!QzIe~c(92Yh0DnX$8P%hj
zEHj9p24ldTbgdXSq~lZl@{9hzx29gLLwjl(nLvEU&#Ci{r(_<?X>B@$0iAIr<vY!P
z9tr+@ge;Ok_!?BMEQv`(Z;;B1_ZSE%E4k7=K|WQ;f=|hT<I9?ju&w|S^Q%2~6=8pa
z?a|NjdjHJyDyu!|->Y%Z=!5tF|GzOXi*Fcu%jXY^YO)vKS?I<^pTM3P1}8Uf8=wDj
zSbEltFR0c6F?ImGC>e=0{~b}KI^x3IM%v5eMjZ*@GFl)*+!X#w722FONODVRFFdD2
z#q`(OELQSjFK-IqYX*3{fr9_%#ugir!Vib2MgTmHhalEk(htbm{|Ot?f-oPw(F@_B
zAEqvQL)?cVP?TZg5c#c8y%F9TsjFUx+C$RqrUY|fy*b?)$LiK#ynacwx?Ih*w%HQk
zSC*Bmf>j-r>{i#V2RLAZ@laNL=D2q0XpWobBUugo0%S#Zp(=6%WF8|!O9uIIt|9Nd
zod$hROdiXpecTRZ5J)4Ga+kdo%KJ4aqX%dS*20m9yN-LsBZK?V(&g$_H7H;$g7xPY
z=7JPS*;YJ5E_X$R&RNOLR@Y&JsyLBTH4EqI>*alNpds{WP?KXvpoH)&DOKSLuM0au
zITkUnarMBtVFLorqi@cOKZ6fnq*g(6;yhGYQS^n*vq-J%4XB6#MqG)KKP`P}mbc@_
zN3&<QCB4_;mjF;O+2=9x0#<Z*EV0?;2UL4hvd`~0-dn@ifUSxDbp}9_<2Ahs5>)D(
zt1M(l0p6-tWM?|WBRmYG41yT|NfV#|5EtrP7PGI-`Ze(JEreA&ob$+>9eGwePbF8b
zWOY^=hqi|jh+k~0%q;lj6ZyTNEW31FcvC0{yI=49VK!K7v#74j=TMvUBjIQhpK-)`
zx$vVcj`A`bU2d|sri06oV<$sLaokXhyjHm)VXZw%4}hD0{DX(NZ0Ug-;~)H>0WtsP
zzS#FkmPPc53~I-yNx<WZMK7Yg=9so6aOlH>>}R~N;W%|+cyW#W2x)%2>oX9hxMCNu
z(hD<8>vpR0petjOEDHkt09WDQIE(WKLwUVGOMx6Z%$*NZSOeBGr;|x<Us*=<JF+sr
zkrNC3@~wvJZft0V<X$tk|BPo!7!$@^sTDG+;#MVGkPzcL_#jd+SzEqM5CY~oK`4?-
zrZ@}Yv8?`K2@B3=o~VV$u;z!7@#ysqum7jL#^7;sYpbz)aLrcA(9x92Ryc(t^)Ti~
zXiefs8=23)-K<wzSMZK*YzEM)v96!-aAMd;i*o0v0wa`BWurO2DjfsC5S3unSX$1M
z@*vIS4QxxeCPN#YPYdm<^lbwc!7Q!NWdxd?v00vu?*S!HO!d@&5b`hy5p&}Tgp*DJ
ziV-B&w}&QSMWoLB2i0o_<@$7#bps#Y!`RT+Rn_|gg27)G1|-KW`^#m9^##si1YW+6
z$qr8c9H~KYC#B9Lq`pF9^L@vfp%0mq;R)@T7rO1{HdhdFOnJ-Ws}m!spZPwGk_XC$
zsjM_{!-p2W(hEjuAZw#8hfh}A<u38inG55&anzUo7~k|a7Uo>P8f8Cd=Jg1D+<QR<
zd5uLZa&3TIO>%=o_*hhFaas}&@e41g3?vQL|Ioi?NhsfNm*qv9#Xo~@!w?3GTYbvB
zlKmDRw1x{M-y*OihRx%`hcO%9v}&De?il{F%4O}*_}WhGVm4oX_V=u%Yy7Nf0YJ>k
zz}(+N<q#@W7s?OR;z~9>fuJbNNG81nU4!C^8$dY>0J)35zh9w|H}v1<*GAXqm0ZC|
zQ)bU+D#K`ZSRPDn7TPRNy3!meKe@FTR~ujr%C=G={!^o-@sI;>`Gzk86QB)<HaY)&
zz1jq5zI_@nTD{tCc~AoMw8YCrNl;ti3a^D4XNj2LepAV)>+H+UP&iT-tx|muH=mwI
zH~G@y$x!2!wb2=CGeueO^){$x-t5!>7ARWV6jm6>7FO(3>slqtLO?dX8NccvI|qyZ
zJlG?jz%`5)1$gin%!uCqxTMU&S>Y^hwWX-Y@syb7hq&m1N{+_E&f*ptTZb|O3#b!$
zmO8|(?S&sS$4@miIv~ws>moI+LiI;kmsk8L?9-=+JB_KvcL0o+W2!ZcM}AX|@sqI`
zcpFn+b%eE;Wt+YnBLH9xP3UO+yn6p5SRPuaVa0ExbqaQATyc->e;GUICR<hxEZXr(
zP*~tkj09pR!G~(4A7w9^o7^~X4|iP{%xcbEc0siu6pP;cvIc?}FsD2H<IgIPS9oPv
ztF(TYLJClPtKe%W&mz2WoH@NU@pDTl7-^nwU%P<?sE<377LjU2>0YMLxP-ebc7&KX
zAYm2Qxh1n0NqB>a!5pjPSW_K;pss`Sdax*~(#=uHF_<j4+1gU@&>^&AkdpT|C5f%B
zn_8s>nCo+h!<do~?0xFwDIBe1H{LmF{;h`QwdJeV_XMaH2#g*o!#%hce503~fLU<e
zx#=cL=GT_5yVTNVtcBe3qElVL2aNJ$lAF=@qT&)>tSTZa3PwkXSPqs*{&on-ClXt&
zf1+d@at{48wJPEQjvBw=Z3^@hsws`X0r$opPLCL%?_9mDfJ#Omu8d1=CT9w14{i^?
zkS=W&U-MYJ@6x~Q7#b|X7^5D`mSy#jO2YKoY?H=;T{x|zg*#Mf`5G=A9<M)nJ6g?8
z_Y;}k;h$XG@Cz(*1oM?Syt6Q3uRIhGG`zu{)-_YY)B2JHI)ek6-LiqXkQa<oq~1wI
z{6}?-$J$v)6xsxvWS^xXLno-Bcg6|Ox{Pb|NS0MF-fNf;e!9&T4+>9PR(Zw2?4`Y~
zAKb!;FpUzNYi)pp3mRFy@=>z-GRJ%7<<Tn7c9AUeaR6{@GmPGi0XL?QHnr&L4TxPz
zeP!k(PDaFM6D1SyAZCE?FJRToAPgP~8|BUYZOjEyMyOWFl?AI{>$2||#b4YKR&|SJ
zkL9>vm&gxhv~U^({RHMGAhP<Uk01^gk}yu!g+E;0@p~S}LJHZZ&IQ!*mM|G((?Aw*
zb*}#rUjVnp3G;DBLz_#4ae!qrvH*Ipgq;i`;CRU?xRJELu9Q|{YG)ms>F>5d{)(5d
z#5Xo4)-K$P)j6tUDUoCxGW#D%`UzHu07j9E;0djfYvpgeRI41V>7zTU4<dRph6e?1
z)EkaZnn0*#C|HdNGar#U)UiE66rA6$r8p@S5I2+D^60PaG`{*BCRn?Lq%{gd-P*h%
z7|~(0OI3u0SOA7{Z@;l{8%q~vi$AkHrriJwg8j!$*S~lWBL#+6bw}mh74IkXfxk1(
zoV~JjwjR!3Y{3O#x#@-UfVAfY%jV@gsgvm_=3YV<uzIUV%xvd+No?!7ExPLdWpzHE
zHbkh7C7=k|AO$e?oNsPb!c;vbyv$u#s@yHwJ}=J}U+xFV2~Su4Bg*L`r-oUT)2mr~
zHh*np?zxZZKr=LP^|96ofLd|j@+$WcMl(=odOo#R9xO^I4v6O*i?e@F?@w6$)=Y60
z7K*o9=VCY?Aj;t3jRemZNEl@@Tuo@fiA9W<ziG2zLLQiHU08|e@%ay_JNeIu3wggy
zgXPl@Tc1mDV?>?r=Er}Y{F}vf&k@e|D;EcigGK;kAm#3?k~fqwR6GBEX>F`<2Xbdr
zjY_U0c{Sm&x9vMU%NL;FD+F0`rQtbvisMKe<qFFON(-3b0$7W{?`d%1YgC9v*Q_a#
zLuv#TiLx+14KPE03{qFX8W6CmUp$gisyS`&p!P2(3y-wUGLRa16#rh+40qn3aA3cA
z<ZGL~exz23f<Nu?N#T!Fb1>n(bbP^P3;LHja8g1gs&oW96bt8&ijllwOD3GJiE6iq
zX!zk&3MMV;#UoUb>N*1^hSlmCs6i5FVXMzts}xj4Gy=5=)i7Fq!q(hKC~siH5Sj(d
zUw-OGcub7;<NtPGbLvOAk&dWYS~4g(7FQA_hdS@{@{PECHkdfKlr@>mBWtHdz03G?
z8RVuFZ!`umc0MQD1(ri%jg`;k&0g`)4YUSM-Ywr?7p8Wmh@G4MB?3}b>spn6uS_3o
zeU2Nbs~8f{;#ULpuimqKXC$Edt_!bIhUF<gc3eoaZmILf!mR=IE$b4>sgP2o4iBE9
z;v>`n)K@WzAYIE>29@nFvt~>Pm<2(VvB0B5PO0m3$N>lpvp9|524EiLafs8u7~=w5
zPC^}Ddif80R4suwem3wnYW$RUe{ugdub3{Sp6(J{YEe15YM%G=vU6Z<!I8MczT@ja
z)jQqe;!AXV!NgZb<xGuap#MpP!YZe)S3YirIRLz^rI!w<((Mql5Eq8*G$O-9LV|%o
zhfg&qZGgUU!l(Bz(Kh`x`^g?I;)BtF_}t~Sn>TvCOq$KIyaa~!$d%>lTqza*mK4m~
zb?KzfpDsCK2R37&DN;#c)>wg3y$^V5(Kak?<B-<`qc<VbmBi8ScrW}&l~zIzd9pGU
z&hp+Hb>0=<0KAbmq?TtsH#TEHTsQgldkBFc9rWz<8OAWH{2Lm2=RM2Djfu$uEqWF=
zxXmqY7<QzKqk^p{eV5Z~QZOrS?QQ>Vpi(1<AygE*O0)o%YcWf%6r|Z3jAHHeu8va2
zkJ2s3pm`Y3U<CD6uLcAoR|B2Vt#Nn?V(W|s)o>h#0JybfjNj2rOO6b|`c3iX2Yt2k
z?BOEA1mOjSx>j*G7~fWu5f>bTEao5%(#jh1CookPgZ<*vjN`K-=CP=Li`v#6#m(p>
zC+xa;WYb^tqnyUA55JQ4+GG*_fkU7?|NIoOr*ZMm3e{ScxMqOF%&YcC`5f2UkH59{
zK_;pt7%tQ_P-hw&+p|=w!}}%))yA<{8&KWoTPrNiBpdXidz1{a?T9Bk%0j%0-#+to
zyU>pN%IT$k>-;F78lZ3iUfg&_Kq~@iHo_|~>(dCQ_jR1hXA&Y!T#`4|?r=ii*lPT`
zkKeQGENya29bdY<O2$ey5chJ=GDe08sKC3J-fD;<u-ap9JQ7sHxyoQ&&$0(VJ=nB%
z?sYA{Si(gopP!uHc2CTT#NI$lrs-tx*GL)-^8S(Os<t`-qxtYCCK&P28l4sfbb|k!
zMWN=kO*6-Q!StFL-Ywr!kO7Fb^BbKcg!;=g&emsom8AMex+Gd&cnv1LA<{}Ji2#nI
ziX`n-I+K!9l~Qf8Fz3<%`|~kSm`9Y5#2Wa|w<^x%JyT%%<E9O;4UU!DIi0ft*f=LJ
zW`kgc9@bd3_V1VF^{51&dywtiLmPD^MK71W*oy|}yQ(^ZntTTkqKEWbt29Vn2Nnw;
zzQM#_bNqx#-n-?x(B~26l9cL6drZQu;!E6DOo;-wXe6&209)_VFe7{?0uYUy0TOy=
zV61VBO3UW*Q}_@8(R5i7^SVFXz=FgJQRmqr+hoHkn$p^*e1UxhSNg3kAr%uAx>UqB
zA*zH<YA~des+Wkc0GK$KgQc)h!4$=_B>K&N&qD;Tnj@+LAa{mb4mgk&3jT8?tB1H;
z7(~2bm2M;opE7&KSrE!^dB#`>%&DYtf%akZjO9qWCy-{C-&iQN$82~8%K}Fl1>}(>
z;ybwWla8lG<+QaOwa%CozRQkK0?SBIi#2~<I^NvbZvf>%0l049BU!D|0&2KTuIqqp
z3<sYq%=bb~oCEg=AQ8iE@PmBQL(SSI>v?{Bc<ZF(NPlx?+Dj6?fP4LVi7r^>4XWNi
zqST?z@7QOv;O5KxwfLr2<H|9tp5A5gfs%qHYPY~0ATPj_>3?>ed~&+&DI^@kC&VP+
zEy5riF+{G$cMXJ~S4(j(a4d!>w%Td3NUiwD%x$_KyhHSv#=20|b92;Wn=euyE)UX@
zYq!n11XY9ROf;Nt<k3<Zf|8a_RA`SwZ0{0m2m*j{=HaUYN`ZVY-fA0WhBWm`omF6S
zcuOpd1^}wXVNk4s@03l^9H&<RyQOPf<uR8Q!P#`|y;<Af-;tPfGDUntnD3$pR()Hg
zFLWG^@~i)_mN!EnatXZe+JT0WJ<g)?(gDG?dCAg&hujJ6Dpp!s1&gY9JHB_wAOP<l
zGi>M*Fi<SNE_|oS5yCI&QxI!97lIt)wK+q>s-I$S3uu8O2F8ECWu6o~+}|w<_k9)O
zx5{I3vinm>VHr&w`(3&RZV?bg6{mXfn1QVjnj999N@gPVdv;ul%bjK1%WtGsqjF)<
zPY+G8N1yoUP&miV0BoeB?YQ&yN*);j@M!=Y%IkdG5X=@nKsR=H27UJHaJTZq2DbM;
z>n@2=^PEIslMuU!43-tTUp{Q5a#LmgbiqLjF1pGr#$BpU5m$8$@#*mjB2$ET>Kc1{
zd|imIA&h&>xJw|uc0~Dp3tWVjmG9=a*R&C;{r1|xi`(x8nlQ9D4~}y!H_#dU)p3|p
zx`8@Ee#o%;ks3mo;=hEJBhx6j5>MV?cgAHb^OVFPpGXU0+8Y#-Th%pwq1|IQxzrKA
zI^)_8B1h6-$>Sb-bQSKhlT%P?Z!iXruC8U`Q!QMicJG#zC$}HVNuXNz{BYm$DImC#
z18W*zIgtP{46y^*H0N1}n;Sxa8Jm_6)`|};UGDR%!1qp3+BS+?K$^@?3wAY@kWXdF
zL=rXddF#9I0FgwGi&~dXo~ivN7SI~U9k@*aE_3s1qf%~E!mv_hkJRy$j;^Zt;YvQY
z{pI9Vs=a;c=Y1^>0#t3vt>_#frxdV(gV@P4C!k)8;4K9XbFrgik9>Ap32A#u?S!*U
z;*+~B%8--^!;-KI;^s<>;2rQqRoce3Vs*EFblTOY<&?U)y==@4<Gr}HB)*^nfe{Q>
z>%*a()=^`<uO>R*%M;)MTk$K`1)gVfr2O0)8^AK>+4CPP(;C&`Y7M6)wtasCvPf1)
zx$*Q3uY_#eT~rWy+4hJ@F<lT7JA$#fAq~Ez(4jW6SQE1SSREtRLwZoy<+4k5AubQ5
z?dAc(TtC;*VI|A>j0uc$v!;b}lf9RL1%%UJUI5KNXAm*MflYZ}K*5y{gYc0<p_&Fc
zdDOe1;9BnUSWv_Gg57Q_+EIHGtQJYcB8miv3sVq)UhGpP7-UQr>AszODh<M^UKF1e
za9`0L+`b1-SA^9I_K-mx>;kwDu;;YuHhuhxUK&HkkTea+OBFl(WnhCL);Y9DlRY+2
zey}-Cg2Q3%7?Yb;JtDc5C*{~&QT2(5lIsO5Xe*JQ<Bq2-SiZV7%?*evgSY^eY-O4p
zPx>nKOtx8hsbCkt<StYX!D_o;M*jE|WQT8AL=8Hrl?iCVYsDmsC=s|A$4~((d@($+
zp^^|RjK9ggJHSSoAoQ3XZlT^b1K22A=w`u0iKLn1P=(4mIk32<H7=;g(K#xdfS=aO
zhiX+WEifQw(CQhd<Vxc{o>h{&3`*P8RUU~P-IC|%n%W13-Q`K(U7WLxOriwJ0H5Jf
zfo<x=$u@;fe{`UPQ+2|0q(V^U%U*X6p1;*a3mATSqsNIGMD-p1BsV|}iW*58D_vku
zz~GUhybR2W3?X8wBR7Ofhuj`ayt`6yzu*yUo>@}xNmx5h98GUF8u^iz4MpUK{84KN
z;PM<CN0v3f@l3wf@*5aHw2ZG%YJNV4m)AFa$tgMDnjliNN2GA}-nu?LFz7V=b++jP
zbP?i4m(>`m!_A_EqPyBH9W4up_{9O?z(T<+%^Gq2xCO4Jebg+4YZ@=S^Zxu%d$m^L
zt=7;;tNAH;gJ$G_3Gi*W_Z)QRbm7qWv42^FB_P}>u?}_BT9~>tj0oYkH`XVaf56+v
ze<Jv!DnUBJB{<LcT4$IH9&_eLQUyV1a}D{IkD!*+W$`+G3^@bsn(X1`_{;Tjm49n(
zepU8Ec7TQ>yz$~(1y#dYzY#SUS73~K2uh8QZ+$5tb*Mgn%-rSXwpcuL4G$~|LFjU6
z$pOxmmWcbW3jtmwD%tH7T!J8vo|oPKa2#t=k=#9%Tm!O8ir~(VGZ7^+z5ziPh6KHV
z1H=3X!~*e8A0UDu;Bt$(1}Q6Jtt0-}s$K&h1_{kDJluOr6Qze1c^!O2g&_A8;Cc86
z?q7zyslE!714q&?_Q`gWxz+bISe2qo3!=?r>4yoAZ(ttN6Ui-&Jn_12McbS>1|5L|
z1A<g>14E??U{sEC7zZJ6dWCn2X?Wvk7E$c&tqzQ9xln?|`WE;F9^3J99}99&SjUU!
zd%_YhOle)<Z4?lLiYY82C#nw*w&@&|Ql;BdiHXeqeX55)z7cfC%SvvAYbRM_p)(|8
zU<3~!9qs+_=_Nv!{T2u=Hx_wLT&;r6sb;8YmW5xjM+Fxo_0;<uAV>%rLGjnY#jAym
z1o8?APlgMsJ)G!VxI#7Na&cr#RM$%=jEerE;vBe`0l2v2E?A!Ivh;UZfGIi?`Ni4;
z<j*0F=s9s+aOE|UIfiZ;p{y2s@z2z@8Pz6M1WfQWte%f(LtCbxfrX7<h%sTwzpt~S
zwDqnT6hR~U>cmKZ7L3jrEN=j-5%f@9#{q>TN&)(6oKp;ty?tYnS;pGqyPC}XKP<c2
z;2ieF1fk!<AJ`{gTE+E?7Os86_teWhzk41jrT#!x-@4+J1NP5VU2*cT4if<<@!*>$
zAaa~MJHX&39?IM`C;b_X=1!jbeoIxi{~0a*1D*nTe71{Irohi;5e^tJOS-2cA?409
zIf24Fn8mtGb2}2HFDRJ<q7DARDlY0qPL5U)<=KT*%KYF7a0e9IAO8zF+57&!b#>e+
zRM&&|>TTOk41M>?!%LxOdHwBbZT#5hAQz!y3jnZ*VwKRleci{-MBY0E?BbswC``mj
z<^Y_$U*%?^qWU6eulkZ@wo5(jS0zWf2T{Ogc@PS>F4yQ(70f^48B@aPJMcUM6?IQV
zv~|s)Og`Yab-Mv9&u{%HZ{MnpXCvf9GN^oUU~}dRxBaUhz%GK7XO4_LK%T6`W`r&+
zXi(PN17x4u;wie%W-r2|7^M8hZnMX(3iC!HlvE%aHAV)cFm4P7Fa7z*m9n>kVo^~E
z53(j=D!wi=N}Rx@Dhv1UyO(jpG|<^3YbGxBh<~CRxU%fqTR)!NR|*{I4INFAH!eLX
z?OuG_3y(VM-{s6bRy;56OT}hzGeWgyfus5hYrN6O?ikY|I77cJ6*Dadp(H%z#$8;O
zg<qut*7yrfZ(jm2lO*yrlJFTLejzhf*>+w|#Zl;#4@dDs>jp$I7F=@Ny67v-Zx|vI
zdX*!qv2{?*Yx(LaMjje_#+q8WdS9MgfdJ||E?oX<&`sYl;R~yPc>D$r;WHPzj?fF%
z4`6MelRF+b9uv$AxQ<cDZ<$q=oTFbUpjzUfK9~*%@7o8uxl%Mqs=0Kb2GgZrHLqC4
zUAls~ONe&OW5Gnxv&gsbPYhh-P&$;yLu^^NHXNz3pgpm1?Y@_TJ_rb;12DG#Zs0yb
z`<`3KqmBxfXmrxG_=ClJmg>m~H_R?v)N{QV%sF+Z-<TbOBaC%BeL)xw3i6gf5W}_N
z-510nBI2T@6>8}HvMd(GwAo_z$o4iX#pDKpyCwF5<A>aDRpPS)gXf-c>>(O%xnO>i
zu*zil$v^SCr61!#N#oMNw=Da_=t6p))d3X5>hGGV_w9BcALiD!+5ES6Q_&P(dq<q#
zQvrv|lb8S^yb2;M0FIa_IMSs6EH7DEUw4Vgol)0elfB)nh`c;T=$Yu#-E-*Z+5?WX
z>X|>t#&2n!B%#a*Rwvi=tpPjMW&CCL35`#{@a;ok4`ewNy`9FTyFiv^o7gYRd|YtR
zv5S}TIv$zyzy4&v$BSEo1&}2`51@vGBnpph3MP<{mRxENLXRNUg+RRIP+mzTfj5X7
zTHSHtj0X!_q*T(QdaS8p?*g$O?Cx*7r{{l`NBp<-<a9WmtwHDK+bs7ZrmW==w=rKJ
zX0t+TdJ&Um!OY8T5YCxjly0SoB0HovavvYKO)WSQJq}J2iN91{^oB|GC)R(30|6XZ
zcpL|nULBCYKv)1#l_~7}u6fCz%L(yPCHH{duDrkz-JFpc%Y7$e-nv#nq856{{yjmw
zm5jZOJJjoEr4Ea&OHA$Ox*(sZ;eJu8&~A0TxXI$)fA-h=0qNQ^f0h{GM8XQ`baWSm
z;DA|P4j|!Kt8$|cTjy6X)8KUC*<4^2%HTXE%CO}KvH6XRNf1`DoAucXpHfYNZ6=kj
zW+ys^d<j;qGP20+TVkJ!8_06WiKC+*-)w;gO{tK6OSajFSuHk6+&shXE)TJT+^$~}
zzCpp49&pMKX)pxwGFLc20WyNsqZzA<e?{<AE28`D&axEF>+W(81$X?AN`gd|m^5d>
z!hTp%+dtkrvT}dq)XMSaS>=6S=-ad6OYFA>E?07gCSJZW(QU`%xTyBln0OavvtwNX
z5S_g5aOU0Q#OFAokK6x>_Mmg#q3UxbKody>(Gor^g$G6`9O{6aVpN&~!5j_ojFdmB
zuaRTU`ZzcUN9hJ%1yMEcuqrPUO?KO_?)?bGlxz97pVz~Ktn$U)8K%y;TPI2W#0P_;
ztGZzP#h6Wslq|TpCDvtcx7wp$6$gL#;?TW@{7>5c3bM)--sew9NHaLPG+_Dy;Grrx
zgFZjdsRdX11CO;j=?kSx%b=>0;ry^-h!xiCh*6Um^|U1&Q6Mrux8U0+Bcy%^31sL8
zECgxsI%3aY7=+f9Wn<?&qS<6%dPU4KISs!m2+7sYzf^qjBGvGa^RHTOOZYmlpjU;O
zVTlm%lw9B(Xr}fAmw`Mz*bjiO38}B*g5~F)5u&N#85bE<PC>3QPq9Y;KZG{>e3E*{
zY8Uq2+V<23q1+F*G{+ln^aP%s{Z^Z$e>KvL>%xnDMRS*kzygGF7F^J774Z)U<+0}u
zs=c_UQT?j-0f<bZ#|3k!`bc^N)h0jf5ff5S9f5%o)q<$@1O^NKT#uMHk~<@QXfgz)
zCxB^;3_|+~f*ESpWl4&ebC`B`{;s#aj~L}vxwG{4+2Km+slPYb-wMnmvsOv>G8KW2
z)e=?hcNIVe4k-q9dn~A;?H`Bnr0N>qe}U;qfN>0*a&aaiUM_X0U}&KO&gI1n!#pyq
z(u*Is4iQ)qzE${;8(0-kF6H8DQQ>JdjsPZU%9JI$@O*a5Q*`jcfd`I{uN&2ZF13V*
z{4D$o`AYFGx7piGu>EjL_PGUn45&O}j6czBfJgQQD9<x{(*;;(m&$E7q{)$Bt_0AH
zU<eP*NtN#uZccNr5or24F!c!^2|Pwr(MUg#N$S|GfCVprKKN&~C#^7pcHtns?^r(a
z(J!zv^zDlF;lp+5za9Vm9g71?FWa`+`|q&#ckz>LV4O!(yDd=M+2U1_)0ct|o$5D#
z_70?&%Qegpw2pC!riCTq9+K|r?qGIx{py5N!B0%@gRW-1T%LfAODzl43LR?T6j};I
z#i)Q{qJ58Ipqe$v+z9tx9eVrKlAKR&U7VJdAFe<20$H>kbvGow!p-0!XI|6;OdQ|r
zJNA>jj`D|=NxbX(Ar$TSE(e!6PLwuF)RTZwA0g?DIOtbPtGRUt;zC+l2@W=<A}~G>
z{xz$Inlca}&WUwflP>r`JlW^V7awDx+(&Q!uq4;@&`;-{US#?9eTfzR%k5AJByXDz
z$E>2wV~KU4ZLCBWUwn8!;qFJH^xc1ZQ#~iCnA4NQG?->vO?$9%i~tsYv0wt2I2B>U
zxLK=#$%wVVHHZNAj++c-UhWn<i!$nS2gqSYFcEP3oUC5SYua9hNP$IulKu`7AL;N6
z|7Dwi708zMwmW%b$c0Ocb-|pQwv%L-FM&jjv;OYnvBbHxs^nLhi%J0%UJt#2jhvZ|
zgwK)w(j4U4^Pma1JvYm1kTk@J1Yo#7m(aH>Io4OQ-w8IXvhVR5n1mUQr|+)T$HQQu
zwo4EF<ifupRt%<D!2D#F_;goB6*8QA9=iZcf+e<rg&jjxtR}DON`CU(cm9{hBqmA@
zP>OO7$ITM$aRC<WXrd4Vu^{$bFDF9j2*_ioV2uTaNoIImWdiqpihCSUsCjHuIzV=h
z_+x@~`Hyz*-?H>X;r4UkJ^A2kmOI)6i}>dMd>v7V-P{d1GKx}?*xtrfPejoq2R!u&
zKtl8HKK^t9qb;~pFNDXDD8Vd131Jx^{T-l|>LqE;Sdwn;SAe-?6u`zvXOTxPZf2pU
zEm2=EIzx7kDoIC*`D5*}?ZCPr^6FmSH$FadCK=phw`~^JTP^bu%|oI*mZ)HlcG+RF
zAWjPs8miIZ#kjeVg@?Jd8Q=a7X~~i1HUY$e=zwQ<?*QErkT6(M){&X(O8bI|f<MU-
zu}BV6ky5It(0X9GNfv7bXHiP2Yw?_~*{;JJ(#^}UnnSvDoAld1$p3JA<g>erLjS1U
zzE`jao({OVo7zO^Uq8&I`A`EXGuBr-M={_~u3<nrmhN%z-eJLkjSChOE&zC79~g(%
z;TViNQ2})&q-Y0mLU4(D6iMqAJU9Xm7Px?bGivE?lav14OiT+m+ehx%U%9Yv<oVOu
zM+?Gd|IjYccIohEtYSMDC#eIRN^k*o;j_TQMU+LBVZ1{;p@-3_jeh{G>qr=CrW(eI
zGfHwHwo7=qLPGlYse;8v623rsuS@mi8kwe6siQ5!Qju&8F3iG(R!|l=o+>%gT{CC(
zJeI8-ix+mE)?U4*I?JQpF&Gt}`mM?Gq+stisq;*S1^lzxt*8v;0E|t9vSIv<Dh#@c
z)fQnF>~(@c;W(uX0s#l11sC>#EDeboam2R-5<M=gTtIoQ1qPw?Xc%qqVTRq!g0UP+
zP?Pe4jzL`2b^qt8&ZhT-TKd)Q*HeG?p)l*YhX!A};|oK7Y7tDpWuMh$;kX20PEM;*
zt&-Js|9@@N3cSq<=1orc6Iw0-;&_9q&!KuRz=DFCX&*`mV!Gs_1%IOFGtB~mDi^d5
zP6qjZ4Z!)c(Hi)ZOTc*%kNHrgI|NnP)A2%qq}N-8BTfD7kI+M>I&xwAhpF8+g`ZM4
z-Dxt3w|0wZlQ^?2WfLq>!Q#qkd}NKR_V?THHycL7e5AlNf#<xATu4g+2p0v@KqIeT
zEu_>D_+}6;nApjBH=9*2B<peF;<HY8el|oNr}9Vh05@hi+(nP772K*kEOK0uV4s~|
z?7OR3xtE3OdgFkly;C)t=lvSN)M?a2Ec*(U{w^^pEUHittCfm20kEb8YOInUKPswL
zKLjgXa6uKR7e*8jFakeNCqUm~=NS|L3R}rmRb}>Li2{fN27;M70aa5fNvX1%DX=+M
zmymQlg?2n9yY=2%ckFCF@UV{BN@bVWe#<%Y8mOHGX29IWm)amehze%Hg}tVMIvOO3
zmu12^pemh!%z#x(frN((kn$dwsQ@Cb&Cj4$PCQz0<VvAu@z7vj-iQ;p&jF$7xQGW)
zPV!Sa@z-LI;4aoDo^cFf5Njm+f|Z}U2ub{4iyi{7cSK>`rS_*qq0NG<Ou#=LN!iWT
zxh-Hu-4DL00ZYnV7=-J)eom>4H`79p`a0%$5ny$QCO8){-Q!Bncco!1uo^myn-#x=
zbak7~?HqSDRXToGbA;dSgfKLYs@qx>!bJv4F4fa*-?MW0jxT;#U53+HB_;~DNOxKx
z<rcNuE&&$Yo=1oZc!U|Z4~?c4YK@gw#j#uZ9YMH|=J8ad+(XjRvdIz235QsK|MBBJ
zp$FB|mP3G1qGxH+&9ANxp|y(PdGHWe5ERgm+c23HCESWs5)<F-5=XN8ytaOR>psV2
zF@FWP>;R;8317oUw_rn5Wi@bto?v4LR;Q+c>jtb~#_r*&SMnnoR!2H$p+oI~Km{mb
zc9u#&^x0Aw(@@)kbeEIuAQWu?3eq$P21R8O55Pnw4|^IXOGp%Yve6?xL$7$3{@c1f
zoCx8v2wwq-L>SQaHlYi`#UoLKdOSem(pNW5!+xGKaD2vG&t6LcHq;y7ss;*4n}r1d
zIs!8af8(G&|L)8HdIx1lDuGtCMuL+G7dJ3Hl!}wVQ6eTe9}!eN<2oz>RGau@v$TS;
zDsmenV*;L+Trvxms36$7TcAV(lo*1DQy69rilal(iZf&WdU;pSCUKjE;uwVoDlJ@I
zfGLNO2&hoZf<tr{Q9L={6HueF>Psa<P(r?iMcm$YX_U}nID@A~sF*<}^IcB^l@F0g
z*BmjSk+F$T{i59@M%%v0g$l1O0so?cU<NO{MPa9oaq>9|P_zaQtAl;x&sbC+4)_y3
z5h&<Lhgx!QD+(^a_rYfL1eb_$<m(*9g+w%`Vc{uhYE?Y>IjH%E$Y4>XhBrv5^4+eN
z5XiCsYLENoo1Sz1<HzexTkR&er2p$P?^tXmyAAt-1pxyf*}6sZP9p~kXAMg6Mp-C7
z{><v9SP+C)VW_-*ZXO_4l1nWF5<XYj=cjLMTt~}zh>Z4Vl4z6K2f}~uc{ovh36h^H
zh@*sSUct&2gB9gr$3OXklzIGrI`gR>5*u0j+}kq`+uQ80`&gkXnC#IWAzD}nA|b@H
znM5J712N&@S?t5-r2tIMPLK;!vU)59m?*@W9SPv^i3zVr6>bF}=jo!AeiP3Dh)+1N
zq(GIEF7?+y;K&+|JwhvJBU_YUK#?}!_hG~^L@g~j{M<~N^sDCz|M<MzEiCewqYLe?
zFF77Fb+*rMQ|I3wZ37dHe><`E)80ReRp;z;Wit90nj7C^f?CciiL?|*K-Eee5GfUW
zC?F>7NX!P=9&(Akhc8^EYm15%-C?D1YZ(pVj(P=B1i=QYU<jzHEfG1So`-Jkzj>a|
z6zln`=g%y(+XT_J#Qr+TykoMcvxS&-Va#gp2cj-K9u63=$3m_X$Ld%CISY-IZt4<L
z$*Duh@Ny)AV4Pna4V(<sK`}bwNO+-qU}c}|Oe!;$(ohX1v&l%VacW56aEj6sxGjN~
zg~CgaIY0=M!}ispwDz{Q{_44}oW3B~XMFJ<=lo!ztIckCM~FVXIcnmfB$xy<5dAxL
zwC#OEmFvQPubzk9JF{jDIo|CCmr4Y-0Ny1%5&)s_N|f&EV~)hmYiVCl-TG_UaViaU
zfcUi(j@xpC$1K#Svu!GF;nA*P{-)Ub7i1P-d6(ItP&du|#;SLht~T8!c<kR(7PW~f
z1oQp3+M|DJZ^w>dt89RE9Pt(0kuz1VJy}_Ygr@TN4BQG%M_{-KD3g}_w1nCLp<;52
zoR@=Tc_lasqve8tQOMsY>fRh3c6BUriExaZS`c=uYwC<|xtqKSv6MgtVMgKaf8m*F
ze+N!j5C+=Jg|@@MXOr6YeNAW=Ed3r7JM89&RsprFcYm<#@TFgQW*>q@yZHDFEIJmv
za5Gp^)saROVZLLdw0a1WrqQ?ziw7_)n3e$8MH@81<r*;%FmWr5sU+NDS;l<YY80!G
z8^G}Tg!WjMBPT5U#Ww^&+Rasej;MN4Q4<W(9P_V*SewNxh=|ksv6#R>wBQ40`hOGk
zHULhQce?nSaKJ>M#=ugQ?e;+51lT5p;uzXmB|&ua5+am#U1F+$O^`S_OxIexE*%mA
zbw#)LhBx#elNBh2aotL#hj>yQnV@z1CuXTM(~BnOLR>qzr4VNqH6w(?`@F&4KY?K;
z$>coe_woEb-mkG=4Tn{vqm6%wLI&t~>Bplgs0a=8?4zBu3B;xw_=zj$Kw1bp5zgoy
z^fsz}utiYjM}>1R_tKmX+xY?#f0!I07MgXz|2kneHCuMs$@Fw6Z?h5`{KzYtO;8yT
zxeJV``-fYlq?q9F3FceQObjP+r|VCF5Q@>~sUC`GMqFKub=9sy8R>x4j+~b8PW8xr
zxq{eLtmNSAE5>H!c|KJOJ4ck^I&AXve8f(~9C(_dF`Bf}B%zbtdBbra?4HZ5)1MCZ
ze7k4=vgJwplx<6v%66YvW0QhXkAu{S*(;@{=9934#(zPFH-2r6_GqVm?TS`pc%q)g
zY)elS;!@C+f|hjTzefC5_6YahIh1$->C~aZ!J+6tINI2^$GtG*E&@=iuh1TA8rL-$
zTIbFh&TpdJgpCPy8Ie@@E35v=^p{Uc<k#Qd{`(|SIVri^ylb6I$aLQ<Y3a8jNvjJ6
zmYkBmI=k<`U~V(R@1_%j`n`2yy*}}=-dZ)Mg&pvC+$=IEz0}dly&WUM!;x2lP?Cxb
z&v;O;-FIour(}XljX8!AV1#a0K`pt9EttE}SE6YlC^8}PQN3}1MRDBHH|<lX@5b$~
z`=pd<mf7^ByDWr|zg%Vg+Hr`>x*V|5Qcl47BchCjguS*O{UmkmGGmkGcr~*;1SAxk
zslxpkh<uI)6*7m`=w7?i&YdwfBgTLylow5c1U`$wt&QDD33Y)=;(yZ~e(Op$QP_IT
zH03Z<@*bmSQkHx&F5R&@`Lqu=JmFsXTg0+3ZJK|~N8j`>P2I7<>?M1^5{cFZSe=Pn
z@|rgOlw0mKny*X*s~9M#@OyAF2Qo}8tmKgF8IEGH*xT&rpa&!_ld0OU_v-p-BrD?4
zL1e}Tn}Pm4y@{rQTdodf*g|&mH22~D_m8pp7n0um@1o=+*}wfY6ru~Cy#C+**^6A*
z!FM|E3DMMK#AC9@v7fAH5dy|1Ra183u_FO{{ry+2DFKaw;7)hCxZx7iU5=>7$U>u_
zu|ukGesG0I<g!TJT9k+AaBTJ}2m=!ufa>Z538N<?Jhdmgd7F|>blzigMw6!fDYs;v
zdN{R}rYAq3dg%@1K*dCCX3rbnBB?tBpIepv<`tikqzRzp$#*<;x&G4eX=@VNHC6Xv
zn+y6PljBGV1Ox;z28WEGS81*SuhV!RyZ2t~LsNuqbz@t)F$@N`>WwD6-)J|3dBEjN
zkd%^!u{D{yvIE1*Xc^r51xUbbl5DUbt8)Rq2I&{p#?t1KkWyAjni&NkVj=GEj&sKs
z9iJKKdu77I7~A}vC*aV-5uqWVF%Vtc2TE=p-l&(l)ek(a1w|!{HF~Y3jI5P-eGFuP
zd7npB0~Oz(0rte5pjg-?)cP)a=z(LFWQwg#+wV{MQBO%od_mvB<p@HaoCG2h2`iGo
z1nMH>r7h1VhRfdy^d(x8J{hsU7SYQQk;AcQI)p?#t{Spzii6Uc-CAc1v%X;2iDK3H
z0xHk(+DY?uNXJ`3pIw`8>_f8Btr$jFzM{;Gz|21-ud|JBRF=zV8=CE_(<TTKa|I<}
zN_xSuB<+>bOa`fj-Nft@q5{7kxz6VPY4Udy0<cd~bs+ym-Kin=y;D<eJ0vow&<4T^
zXX+vY*D*1B4MtSeGW1J#9eS<~%4~_bPom;h3fLDAphWhPYcUMkyzDrv6y9TtBT2tB
z(@)3}|Id6h<)gMcr2SS|u)63`(-f15iHGMQ?IeUqHb337d;$I6>4TM+_+%^?QUlQ3
z*_ZnU9)cnP5xT5WJ&92imt2mkEro}paD!~`S7O#VP*%B%NBlC{OSBS1tpM;IWH0-a
zn;m&YzZj5N(${T7*_UpMirA98mOjjgy$gzaDYu|sn}(2p2>;SVE>GzhJgqOPfMmwk
zC-1}=PI`G5J9CP7)8#9OfD<c~WltFlCdMK4K5*(lqU5Qg7KM6;fz9}l(T^pFw89h!
zF1H)c#J=kO6+ZNGU>5RRDaGhkk9V76t)0ognt&#PY=-PmCoD}1jV?V*fblpIcLMjb
z(r@b92j+<NR^#uP-TJb49z|VkRqOx-JLrgaxa2#KZba`41UK#uoQEW_X))K0kA^NK
zT5Gx{1vw$i<as}|cUyrug$r(e*8R_2P-_u+z~&@5Vd<}_Y_=wo0T?<ldtvA@UkbY-
zB$y{KGvKno0q7IH1+$m(mknic=e?Pf2NuJKi>vNC%R?82t6$04*}^g$Fs_DYv-Jb9
z#jO?X3u*~M>|ncu-3BbbuL#R(K}-yPbC~*sS|r166Egg2<QK*9*VzUT0fp@(y{@PI
z^za>b+E+*i0E7^`SFSn;!S~B^(47F~mAOT_(SP`B-_67fWj@B1XFkGvv4MbFt*#$W
zG%tqZ8W?`2qJ^W{@nylmj&d-QXq<)Y3Zq%JIFAKC00dxTt&nE`pkPsuIA|i?WYHtB
zTcZPW#ye#kU*MO(QE3yZEP;;u1rGe<K53@~DwD|6G@cOO;7P1Qw8$Ioz2%{in4|N-
zjySh@XP;y<8gJKQkVX_nt1&1rf+*-dT2yk`MB^Y2xPll)TXt4&c|IlcTbSAODLJGm
zqeg6O+(;Dbog${1todqOpM!N+$7zbV%#4_m!q<ow)4{2@KBk~xgvf6rCTJ~<7x9vr
z$v^}RCSVRZBj#v!jz#31R;U~yUqzu8Kz8oZzz-~<Rvp_LL-vxQo?nEWzZ7?knYOf)
zOt0`f{QDFembSY6mnJ5d)xU||n0V|<vB6^J%sc-MFZ;$5pvouEGC;tS_&OL4gaS3u
zjL)`2fRcd8Y@#VE6)g@z+v6y&V~)aQD}`_A%a#h!l#nJk{#;rK@&Z1!>L1#|*TfKc
z<|!pHo8WArQ54D=u1xIa3wnEgVb|$JEo?^V<ZC{TYXDVC%g9y$;xI%&a81b`TLASD
z2!-Li`Tf_LBtaNN;wIQ6MLn8kZp?9M0+ik<Q?*L@GD$bwJQUT+E;XbHiG{&Tah=5v
zJhq;nf9PI=W}2DUu`9x^MJ^?J2uicw8^_IL8pAPI)tpZ0s}Y>%S-baF@=c}*K0ij>
zx+;S>l<8*EXNc@ew-W(DfCVd<@qt#Fw7LsM;(f8##AUW<3&P!-%XK(bIl@C5^}whO
z>4e-`$y3!wE{7aEOozsb*y1Y|bCW&FOp&fMFO#KOXi<G^xI}9~1mZkrLoVH2s4VkQ
z3z@Xj|Kkxf2fQ#Cd8e`q3}DBl8MQ6VB_SAKz)N5zCZ11P5e^>9W29BT5<qGWBZA><
zxfnu{;OvO+sY(JdJ^N6|jiClDFhL|I=N$XZk$teCdH_Mow2<8n?`K$yrGvk-BJj<N
zs_tUNKXnGUCJBj5m2F-bR~3LPatham$t;BhQ5&Oa%Qp%5+X|F2zItk_VK4v`a_AUL
z)D6&3L@mYj@*&s{Jbq7pK+*SV0q~?-J&Aq{_K@1)&{~oG+_ybxnVlFmaB9;QVNi`D
zuu0G+quKD!sY=gOm@5snq%gr26hz=mQV18e&ZzihsL1RECJAp(o6rYB^#a2tfan?d
zuKfu&{P>D*jUE!T!^3Vp5al6kGJGv_XX|Ap+n^>~3A{Wu{PtK8BmK(OcrLjfYYRHn
zd;uadF7RNx!t`uxp&8MFor}LhvH#Cp+%@7W*@07yE#zq?GyG_^*O0j|nu6Sf0m<n}
zpAws?YurS}&oe4+ZgLkopRg}>jo3M(>Qa%)O+6@c2ZW}FhhsF;?x_Te!)%ZTkTQWa
zJR@al^ywnmZR^-_FAJw&B6}JUJy%aw8S?!V)h~j*cQCrjN2TvYxL1Bsnw?`(nn^Xk
z0FX9Ke7q#58Rd(B3G(vdL#Wv#&)ACjH}2;;pPZX<dNF3&&ki40@=*0bR8?w;9_hiY
zbB67@J+)vi8|$|H1t_CG$769#!5|83eV`9W&%unG82}4X=(-kr$o(Wm$-dz(5YF#>
zPO@{m-RG_NX%c>Zr@Y!qCnPEHnV{J;Mr9WxCFz=diuGju(xl@(Pcb)besU?12$tqm
zzZ{u{c`D!o3se#tJF3Mn(zOgX->Vcaisv!kN@s?+Ck!z(a(u-Y59TW|3K?aC5GekB
z&3`-XRM^^wueV6A`^`@qsEf%eCN}!=GP24dZ#zcqJKnIvlHmJYF;Q-U<@Y>&C#A=o
zTiLvP^Ga`J#pn*PsLD`&JeX0g)b4>gsu3QDaIj`zBXBfv3J>E3_PCtvtRJeZXln?!
zYu>LjPc7!I#0Jqhz)_s7p!bzcv{_;6mZYuHajSWN%Ql~!dLl`Dj>j!>(qc(Zpyt=g
zC=^nJCOrm_rI?gr)aJF(Cy3#rx#pEv7V~`%`~B<|EL0h(gmP*ig1&$lvO_|L_3W7b
zm{1P!K(hs#32*6(BaG0xF@GF^x0snz3=X2vb4Jvj*wboLb>+-(POseSpJcO2t4x!Y
z{Z5*8llR|_%Vs~B{td4=X-U6DlTW_^(IFX577Wg4X0DXB(w%Ovj5>!%dgsdd(VRUS
zd?ZsU{L5yg*=$Ah66MUL8WG-#>D9vhfeShe7;GIwUfX|Jp+{D<JIQI0Z@j<Tc=W^A
z4>8|R%$W=Kwfbx96VvBmNdfPS*PeVMiLTR5$!4+nj++ixOuyJM(e8JV$LViu-jqoM
z5VFYNKAPD8zoav#f*SUN9R~Gx|1^MxWvpkQx6wEyz|d}!wT~9y>Lq8#7)5cV1(YL7
zLCIy5c(%TUZ0CMy&*Ca<XmI$vYd5;mF#P=9Mt&srbLymOC9ux6%_1e4?Qt(bE=TT=
zBr;L+Iu<e$H+znmNP4Se1Ej^R0TB?yRd;%c%}2~wX3-h_hi7&GQUXbKl$Qicf)w;B
zedZ<17(|Dl0cgsw(SVpBs7rb+NF18iiu`)^vVp*+TF_~%`Ol#kQYTMMoVGah`M(1u
z7Ln~1qFQFWYnk-=EXmGfGm+9JuqKKsEHxrFFY<0FnP{OZG&TrKY2;UaVPnqD8%GlL
zmRmO>Aa5SnFgqL(n1sBXXJAW0T8&PvI|Mtd8T7Cj83V*dRr3JrsVKMh_5CQT?3gF4
z#cZ1V{Jn{;l;TMfF~51&<G-^c1G5+W!P|4=Qa&@9^kWT_M;uNKIE0{GpN&vsE&-yH
zpAwTT=@W=P#JMyS(H=2Z3ji8Q0IC`=*G8e+Fd0L6H!1>h)vYn*a4D*Wh_u6@Pa3(a
zdUoKE?88gQ49o-fKm{YQFHKA2_^DECI+=Ryft51%y+ACcozD{zUYcg>v||&~PkO4T
zpM1bW%v)O1+@$J4mryXJ&AUJ#Iz#@=qlGB$UX<YH9B_NR3U!IFhj9Y;<;{;F^AL2d
zf*DXE?1`u%3`ZaT-Io>T)6^7K{eF-s7MBO}Bg(VWHs^t8mW)o}t_katrd`B3>;3nZ
z&JNRQJ52|kAp28#Ns*EQ5Fv}})1o}QmYCi01YeKc+w(n))7~w8@0&>3)Id}dlA>yf
z7TfUqAC2^e;ByX+sN4oe&Li8Rx)D|PMAO$tv^RNh@mc-&wd*`PgUG#|*;uZk^bEv)
zK67!7H%qe+xo#L?pF3%>5sURZ-}BDC*|VCgm3`aa-R3F9rL>h0lVo1JW}2dT(XCP>
zlb4u-V(b{6_e=dEs@RTTAjsfRAsy7T1vP{&6GaM+r%^-dUcbaAA{SRHzMD5@LOFf{
zQ#JKW`27L2=VBg1@qjLC_kDfS7b!o*W+APOr*5z;@gdQlm;A#-vTTXhH+8@Nc^Qz_
z+Py?h!#x%oKXB;B(5rWAg?Ys8wN3QF#K}QL?=N(OvKFCJ-a#Uej-YA-MGA9AS~)E$
z5fj<^;E_!kr7?hFM9HCiGjQl;=@yQ`u^8fshYPxVe!>^2S}Q%U;Wwm5uaePaYf0b>
z_J56fR<B)FU2a}IH&@A%JMCTwgb^{_YbyX*NXDUwm}doj(KI2u<2+I5yU?i?3Vfwi
zN&`w9_M(Q}oXwjd24S7eB}-dwY|xp(X-C+o8p|paEb>JE1m>TJ@SrCn!ZTENYjCJX
zt%V$9m5U^q*-leQuay3SteaBzep@zQPj{8_W=SZ@eihLKhk6W$A|b#L_@A7%&=etW
z--NMA<+)}h<U&$FsuYsZUiV5M4i?crZq{Mr7ppD?4I&vB>tLj>p;r=u$Z1A_#cl3n
zQBx`i(Yadi=fRmCUJFg!ZwKRe7Y;glh;>2Et)KLANQS^prXIEv(oJj$xf?A}ND0Sn
zVHTE!z_hQ02YTiIz3RTB1(1wdRp8zdN*6$$rffGO)ASe%u4VIp29b?JBsHwWsyhdQ
zm{++OJ&KJL)-!@_Q<R=e4A1sd*erL^1B#sQ{mF6H<Nm|HSo@v*<PmxnhZ*oC$nH$e
z*Qe&M2SxfNyI)GK03=}_95+1e3E571%)m|LzoqI5_uy&?wF^o4C6Bd-vrJ=1QC#TW
z=eYIDG0Fo@qp!&@8M0AmeS9Xkk*{eC*E`@KtWVm+gM*onK2XZ>@Ga3IjuT=JN}1JD
zI6M5OyT5<f`o_~Q6urd0ijb7W#;2yY;-ZAueIj*A!ojWf5FxvENOodN`H784E8<#c
zva~fOVxEu=D57&X05wE}s~PCbAg;`BA($i(QhFn;Mh4x%R=X+uQVSL@_BSpE2O-AV
z3Dj<L#N6aFfpwF1hBx&5L_hOA&;ALsY%gSO7JCS%l6V(zYkDWbKcU=@Y(&EUY!;>}
zk=<U|ZnF8j5Tx`rrfw82AG07NPbU-QJ%X2>u)24%2#3Q}`m`|OuGqA*A-bl9qMXHI
zOykhFflRX1H{L2@ye=EeWgxJ(E>uoCy|<$L3dE?%HqRW?D%+MO%^SY)!oDRIJK`7d
z+K`1Ku-Qxq9TG?0JCz3j%Od@wMN0X5?(hQZ#9xcsz48Qdf7NWYmt4@RYAE|X;o#Au
z_+Edf50eb^01T;(I0w5a<d7p>t!|CJ(e8u{_+5(?GAGubx|BO9{0~PLA3Ab5fCG4M
zFeR7sFTeTM=_MB13-wiAJ7xkC%Yq~UGKp9yVxkSwn#e<OEo_mlu}WsVL}<z@j|_`&
z)~Pylrf?KYU4j&Tj&4)QQK=PGd}FK%Ky-kW>AOrUJ>0jut)a9tnnF~_0YDdZx56Kd
z7GKI<W?zci&1?4u(>>*XokVsF{(Es1SvZ*<PMW4C(Ih}>(gj4|{Spr;dYDSvz?~N8
zj&iT$<9&&<dUc{+^K!(xWmZh@Ut00jU{vA3N*E3q)$m&|t&TX1GL4!s<(+9LlM`H-
z(?KGty!5f||GWBVg&pvIQLkTS@>*hDo0dp5qNVRVjDrqW`R&G9TuMqx$;n=sfR`X9
z{E-kkQAMk19|UAt4}47eH=(*j0~W1^-P+1!3a#L?gN2M95FP~}!&(@4z~zd5ipN6H
zUFG7%C05KA$^ZEB3Yb0fTtVYbh&cvyJ#4WD!N5&c`lZ}4d2r&co(Yk48`YJ*o|Mh*
zeFL6n$yzOTq%l`PO+?`2CD-(mvP`B<7q@GX-i|!6P0>m&b@eJKb*E-5yVnZFjO(5O
z?5I@>pH;(NCyVNDVL!F?omkmVtgegGUQkO8sejx6XE^sZj?RoZj|2w~y?og_a}jE|
zuPi@i59WV8%l>2i!0fg2oa(9@)(%Wtl|DJeY@Kq`_44E2l9R}dyu=LCjc^PQe#`Vt
zVx4XmoiXX|_~lUKpUm02pdB_4n{Ql(B;U_~ZH%#TWYisg%y%R*5Olw)mE-(I_pYRs
z9*b``ZR{;9tT!vM{BRRHHi9P|x#FL`eXH2;v(L2iuYZZwa`Gof4*aX<>ksfHH?Y*V
zD#yM(*KC<<ekY=?l+6}|V*rBaVJnmV3JC*L2^R3vzf+DUN7R~kHn95lURsQuFsn<z
zrorci4P`Uai+XKYknTQX2HQrp-uV9-D+qZSUuh*eC4D3&T(M_>{o*2QyL8k&1U=T+
zmz7&u=RDgsIR9y<^V#*cevf;<?eqAWrCaBwt#i#g*3S4AJYRI<rU`4%7@hP}%k-p0
zTID5nzx4JNWBn*EyEVhDuXgW18Jo-YKZmSiw-KquE=70<r1Z!DvBP#8`cs(VHaO5i
z%lsU$06EJc8O$n!MlhjY>dC0@OuH5DhwmxrAF#nCRDVHTaKQ8RFY{&BN0Sq4ft%)H
zRy27zvP*!HfD5TW06tEoX_lT8OPaf&9$eqT@j79o+^${fNv%vvu%R9<?Dk@n;ucQy
z()L2oMnel0un$fP*F*=U(|74%k-Oywg3gNKtORLMJYQLH(Dh8CSYxi)nq`t@*`)t^
zBW?S`saapT$gcO<Rz4g!8BChyrk-5)`1Jx*%1X9&l^1Zx|7e=TmqUF*Qje}zl*0H&
z#S?UgT17+~h+W9WdK?-k&v2ahXrOgIg3IN^WC6$qP{HHkORF&+6cJ}Y>FM^C29e3}
z-o8pB>Uc&0D(Z<?XI#H5_O<Ql+<VLCv@Vyv;26B=9^<t;J%lD_Z-y;%KbS!3qEZ=^
zcD_baK6G|uyI1=A3iB9>6PHz98bs57WDQD#W&Z7^xE>Jt#`8ec(960WTGT2d)<w{P
zM`DZmBy_aZdx2xAJsbP5_mMNz=OTNQg3;@+J?@*ab;H`fUaFQ|qoznL5IMl&+OJI#
z*g3mF6H!$}8EWlbU|F2-cTu0v6q8}<mwE7cr2n|%+(7W94iC)Jk|rc!S_wgys;4`3
z$febyLy2)MG~XyW@iSsd6H}%{_EvXbzb+hSrn9_?8o1u3H!AQu$L!<&8LzEf`@){)
z9v4QB5IzZuN=r;aIZ3N*N5Tek^r9ywjC{$RWWsTpw2*5}#73mAV>2Ga1k`uP;Yt{8
ze~G3|5)X>l0xM(AN^**`>L`9(G{hEOCk8_fL_9D=e?F_^vVZSh6mr7UM}nS#V1D@2
zcQ076`ky)b@Auxz9*Vs5e8r$@0IuaqF_ZMViFPreJueG6eTE5yxqZT$$n$!VR$?d8
zzZ%uNj|ZZY0I9E2EA{JTzn!`88=S@Y>M0!RbL$9wfCmy6QbKgw?8(HOnZQ_vT^M1s
z!e(b<Z3NvfPi-`LSoy;HmT#2i+iWq^gdBiirYnhL_R=JgILb@t$SS7^vAR)$A%o0i
zo*+p=%y2qFdf`|KPc-;r4j2st+=YVvphK^^F|U$76<14?m!tY`voUr7UGRMhX2R>$
zeC-5~P##91;H)Vg6bj6(G&Mtsy>I(PVAhhtkiRMePoY1@S6W1>OMJjX?7#z}HetU@
z&A^8%`JFomL7#z0DMH?hsDTKc)Jnn|yMH69$Q7a-VQ;5P2N>qoFzD5OQDf(v7a7fV
zb~i)Kk|(3!7+K{A(p_k5MW+_H#r@0)T{V<>@4w?JZE!x?%L7(wauxe?>MD~&e6r1i
z4})N6(o54s&c;=gLNFxLF*#-34?y^PBw9@5)N({z6@aQ3rHBBEh9ASw5}sLG)K3{>
zh953#MAJLmYT2!hA;qL8ck26;{H|BNyRY8P-rVMWAG=@12GZsDeUD_PRA=41`Ag>e
zO>`%Cqj8p>gop@gFCdbVCcvRdlazJi+$6o~)<6F|<vad}zKNYkU-G~#uBaiZm)sge
z162VtbE!gJ6k5zxiMuEJts!MU$hzuM%c75xZRCb8={y$R66>i$Mh$FL9yb!}=iUZW
zEIXC@lR?9DYh<tT`zNrjx!4}%UWS@STo6iw#*)D2!w4@fCJ<GSm|OEGDdB0DY7C{7
z6*Vk!Hyv><;08FnjWdS^`I@KZC*%2fG@T87<W@0z2uoqsD8*Qwt820`s$P0g54H=f
zE0aGZAA3<->4uyuTP9(jqREPLrJrP@Lt?{EFSG^AWe}3gy^xkn{N}K0IE)X+k+NQD
z^NCh81i1*~5iUKvTxZOV8)R|SQ20sM5r(|toGj<dg88g+bg+cU&A|rv8fV><l{8Pp
zdeHUioHzt3fMW2aD83>!jb3}7u*sOEvLtZMH1|1u<0|0l#-LFQ&H@av`)C^2{7Bhk
zLKXs&p{N)D_={t7iI3Xx5|KRgu%PIOTEb00J`kO<OALTVR3ok?f!*%sKC35#yt;RR
zPhV}BQH~q^#r8qHb#)qN0+pu7&HKk&j_7TroN7cjOrO*Hm|vcc{mi^@h)u2bL2kB)
zbdq~<89<^*0?yG8o0<l6qwwh`iJ)%bj8Y*+pNgnq*nz;oL)~A2fis5=N`dR50g%Uu
zTLb8?zFUrLKW6&!7{r3P9+*<vl~STxAV7>lSTIjaNtxsQ!uc04bk2HyJ7;Xxf*l)I
zd6$#$?c%ju8iF9;V%QeIF_^liD#ookO_)2xmVSDv8McmG!9;I!>D<3?nW7<MsJd0W
zPl(M8L))G*BgtliwV*x{d=b-6##`*sR+lxKCtmdyuQm-a&H`bf627IA>Dy~KzxY~g
zR5#}T5B;^hr7!=+^UC+JL4Y`Ir9Z!&5X2WSco!A;)5=R!7;Ediq5+k9(u;T4rx5Ow
zijjQ!jz|*FAtu@tnWBYgk{?40qtyor;-G8>;?^rM1$kVOJX0(oLUtu&^fgPxe0@cs
z-YB;EUv0QN=dgEOX^zs9y1le+@q-U+NFug$>WOBOz#zn222qT2zm%N9AptRmqpzVu
z9Iw5+)NDzTpbXUW&2+Zk4#<WqF&8hndDwx?NYN7GhGTCSshfyroS~nj8?0ib8P>D!
zvwQ^-D%QTebZG1aHH3oi8KI`g0CU_LI+xSgoaOn{1;e$wPm#@V?xrv!NDIiQ9h!u|
z{AdyKkgL!hZ?zj5fSs(yO@E9f-8uN_i>o%^unxRG83!u8<RZpoPspLNh-M5M@7ENy
z$d&Qa7#_=W2Ik{ly8?}1Xdf-`vzJR0h599G#Hfms#>c)IxiOYg*;h7P|ASA;?jcys
z2JceLysLwB0ZIla4mJ5<^^)Pxm=HN8;^c-yF8yAwls3svx`ij6wiE2VjqVS6WMHi_
zdOd5kP((x?k!oDK<-W7UM6-=a9zX?1Y;mjV@KkXNk=B@ALNCl|1anz7IbQwQrqKBp
zRKxLjD2^CA%p$2bBRX?ptTV}lltfBkh7WpY7;!EpXNX5Vq2s7qbu}YqA3Zq*-a!9P
zKr*@bx7`}XPbYrJ=*U9m-ZgVBTtW<ZmLIr3>=aWk6-OoL%qVl-U074LuQ;fMT&)CX
zU-S#%4X3r`*^sRK(siSfQ%fL0%z||dBV@tBf$3AS51b89LO!@MW^SI3XFrp79rH8-
z4s8kGM)guu2r`IS(6dmyVFTZ6eXMv<C@{_giXQGe&Z1Uq2^&(>L;8&NFhl;Gj9{H(
zPGg(U;V~c}h3;j}Qz*=5ltGnxg*leqsEi^k=;0_%eHuo<3Q#Qod+zL(Y+lKZZW4oG
z(o9?x$<I|>V#3?SaFkcP(lp?LnGnf{sv>ibt0g?+;^oa4a~*#ZHO-ciZXEHyvzs?w
zz=<e&bxR$K&(jYkKM(GSphu9wP#%);<tM$?muF8Z`_&SgckAE4;=8a!Q+=}=XO4-`
z6$aII_U4e6*nIRVu4!?OH5Gzd5<sVOBmn{poZsDpZ;s^zLzNwR;47y}`YA{XY!z0p
z*)DEUE8|lULmuSkH^T*$8!@Kd6sIuf@JSg3PDy5(mzgFliDvtYu@|q|gE%ilRow`!
zy`a|*f@iX0!>#6-9IQAE-T4<n!kNfG7@r16?WTZ)NM~b%_^ul3f%jehBWH>hcECN=
zQkSYi@ZreNgE$S_t)Cdh@F6^05Ws1CN<x^X(mAbH!cdrkmZH8@TWQgaNHTKlWj)Id
z*@F-YZc8=rhOrBp0RxO=`-bc9pGqPKnh8Nth1Haq2&y7u46w!^sbRp<(o6AGFGEsZ
z>6;foOnM*=-&3m*H2|5%^-=^m1T2CFJN5n?FKvWD);rlCt^~UacSjhM%gj@uFu)~j
zx+*AE2=2vIV{L$fa0Mt7qa{x*5Xm<5`bTF#3h;Wt+B*pX9S;o~eL&UMG!h@FMns2S
z7;aE~n){_!b37bZ!CzI>rx6`jf&4>M1D}>7NXqM?2GEWEgZlA-b9DDrPsTkNbPpXo
z#>4UyG>>(lGJ;r$P_Uz#fvEKGgW{cqGfb~RRzS6>p0p$3^@<3G1&n;D8s3B`4~aKN
zVpt*PTewV8OYUSv!qhCub%B?{#v~O&P@y~s%JAwX5NKR20UK^v(J|(Frs~H2N!nGP
zfkN{|EIww?Sr51~t5E0?vknJ>`n|zcM=iT_a0%ov4DT{xE(5b^2J0u&KqI(w;xA`S
zOnTL8KFa<Ou-ba{>-I|3WHV2@RGRBDTO=sCywKUdONY=wieVRE3EO9|JIcdR4I5XL
z6u2hEVldW??jmd$j?0ShLTnIQVE~oS1an-$<c#=#`~-HTmK;zG{$lcbKm6=Pq^|k8
zUjIFG!>u&!ocQ6oS+r!&?@}K|VDA6QiE`M3^S;ajiI_YaBIVXe_aiz)&dmW7w+06^
zBCo)9O1S&na22YE0@Gl41BZEB$SRm5gX<Y>A$t7HS3nKOkWY+37;g*l;DvbYKiYZ?
z$P#)BA{2k|Y@_+seVzX@BV3ti+Uj^V9vfWEflKrs&NibTt@4B{#6E3PY!=%ET;4gq
zY3Pm4MXWi8RfWNOijHVgrPtNsI&<k}1#5rAynjsPiN&+%wjlUn$3ijOh|QOs2Be1<
z6GwuHrZ92=z}QlYj_Ea2aFDEcTbD1D4(6D?)o6N~$nKb&yu}f+C_Vp!=YOw$6FWye
zczNANf3>jWms2;{A6xq+6{PE(qwY5&c`qMQLCBC+K;a=PBnAt^z^n#AhPTwkHLHt<
zVoI?wjzx^a^^HeXjA4Y~6p|-g9olQ*(hc1Z*<kz_YBY@g3upq*Rmfr&8m0K#bDG!d
zr*^b6{(jaT|D6j~Y9H)-d;Zav+8@ptzvnJ=(NneW+{B_eLhRnZSo*w#E_%tOGH6cG
zK&D}@Ju4`ThvIty_SASwx%eqL*vR%*nA@EdmxT0IKzGR#-JG;3mfJ#U><}zV7-9Th
z+=)UR9EKn2Xn%-j#RKc_nrHf3Yl%s=d!^KWI=0>r_1s|Zhf0edw?_wJxqDYJu*uu=
z1~4=67_phYaL>QSZiNp557<mePZMi(ks`ne^mFtgI0S?io_${q4eAU;lH0LAyL5t{
zXcSsVD++~{f`rUaZZ}vP7m1%=Ct%cJYxRx@MXj!;1<F2bR5!kKp%*PhLe^2TTH1cx
z94xlnxV1)8w#`DCDA|5=jWO3Qat_lMe)A_e<Ur=9GAvX^5$)oF=%RWR?&#3&InKlD
z^}vYYsg-iI;0I3-w9J%ZD<Q6_mm6@|tO73ucQ@45rsuIPsv}@QR9D|4aJB~VBUW$?
zkD`3$J~3uw!?51c8DcE>t&VZp?F_rM00#T4v+hu2J2F61`<^ki+$D&R8%aDOxCq}s
z3TI%-j~zd=fjgbG$DBMyGs*$~%>H3;6lBv7j-VSSUaDWoVE(5lfXeZ=$FB%z;sHxA
z##;&W1@YRSA2^>v#{dK+cgmtLv><&~#VtT+E%?G9CJp#8ZR2ZS$JDmR(FHKISLSpm
z6TvEgi$SY#y&O>mkwtkp^3&Tc;#8=Tvoa%qF$W$<yin}xloLQEP-6OLt9N}JDr-TK
zpfB+1#@vp+@j7_h=qp23;WhXohEu>V<zbTc)70NgOd~Z$JYRbCV9G|Pj$8!e?Hq#&
z$*l`6xn}(u%;!D5WR#<Dt41{lE9}rhkX^TAt4<Xv4jVC@U&@yjhLFv@Htz$51uU(`
z*$mb)f#4DVWr%-&4h|rX&GgXAedFhU)DDtGh@OzJzIhR_cTFyFBtpI9z;&i1AvTPS
zN_@qiWh(w3C259*T6F7%RhMR2KeSM8IUQ=~B5+33a#R=KTpCQRtY{)!DSU~f1x%EV
zVhV9J&y}`VSHWB1aYJbu|2ba5IaU3@+r(Gn!6JJ}LHn-ohvpeC^2xUV3c}mH`N8p)
zu!QzyWztH=u_d!aV&*3{i9@RRxUNnc7-r((7)`Ez`21RSGgyommqwvP9)jfHsZ=u*
zHAH#H&C78f#+J0eI!iY-wAq<BuLYs5PP^X!k>^CA)hL{X$@g^{1*kV6a`sn(2cvu1
zh5xU4>;bhtmKbmJt56rhS`j1^ef;~8&-ooF|C3CzQhiOaK6X`V8wg>8xK`!q?_Q1v
z&)f%ZF<ReM4HujesD$xX6<`yRD#I__+EY|5Aq<~#I4i6Nmq-+hzE{RX;88BG!KXGv
zEU(3`1}r)dTtsec=zkd9>YToNY^-y7@xFThSUn-8J9kRbjq7h(LZL<*cHp3@;O8+N
zube}wh80r!9nXL5hB&|Pr^qRdv(q6HTv~`I97j0U!c;{(5FQbchT?qMJgffg*dT0v
z(@0}Svk#j`8GS>s!DEhg3)vX+7=4e_VrY9K@i^$a4WI_Mn`b;O^8emP>aESVgIkKC
zZ4jMpGN-(<9jO_iJ0keNtB)>6$3zyAjp!A>CX&K+*pbHg_-{3YrvN%^5e?8#Jb}(c
zgvY#e(>za18O+2ywV*R?T6j-A@FjlIR9s(Cj_M^3XM=AWJrxMo$W?-RR+$s~x}v}5
z`LVnnW9|)M^oX5U9MQF-Lf(e{Ad$dVcB@pPT|h89AZ*P7|BA8U>QeZiZmg5Y=ZydS
zj#`eZfd~i!`Wr66!tf2W=-~ckF%OI72jV17mP`oZWR=&coct@uUOd`*Nu1Nz*TjxB
z7<1yvkFvpBjm?QZ6C1x6jvsg{0dP>3U^L;NS0BntyHLQvFNuYuOp^ivA3i~3+N~I|
zGbe;pjl!^0j1`+dXMFr!t*i*n1=RvrDGF>_SQ#uUd7kY@Rz;R9H!E9VOrE6N)D2uz
zT-LZUT<WQ@97lU4mdk8@WJJmBzC=ecT!20KFum@xn?TvsO}V=c(rUl7D{Ue+3%n3P
zh@iNGW5TcNn`5$_NnljN0gK`#Sx$b=82b*$R)Zj6W<cG4ke4HRSyn43ioJXXO()H9
z{@}}lMsVQ-S%@FG&utiva-!_X#^x957_QQh*r5;~3HF>l8hn+{bf^ZR&zO+N|1+tv
zB7S-)mxN=YEk-16kpfVWK-Mw(*1;Pkuzb}egjSj-le>)<zpICZPIXW&i)%ZgDtwZ;
z^^(k&`3Q>nbQs-nq;D?NGHxL;_#~LZ7(=R~DV_l|>w=yUL^3v52_77MxslVWko?s`
z0I~ui2pSOZ4MdpiVxywtY9ghm3cl+q;k+Vq?+hWfxHiGXOH4Mes29JW3Tg%7C<s<J
zsP+`T0;oHOnc8$D)-!N_$&F_~BuSA);4YO;L&+E~b&y^vJy{Hkz*UfvJT~X-Gw40x
z;{t#q5{)sa6Qm)+Kpke66kRb3NqUKynFF|kE->BbRg$P$jEHK)ahwoG`^0>A{oG&Z
zVK;yYWme&~aW2wrMMouL3*TLuE6-Grfb)=hJ2B@PxN}HJ!Sz6HW3TkWFFh5O2YF<I
zC<>laUS=-5u9myV;3$o&(7O15t&v3bP4EXJ)WpsraDe{@?r??*WQLL^E`FSmB>eo7
zzeQz^XEi8}Awt9Q;y|Y^sRfINxX49Z$l}y;<WLW14dnU&2ua{VdZ3M3T`WBUd0>R`
ztxY&eBD#kOsq5h_;%&$yyXb_Gc3S-xl4fKb(57ew^Q!v@#EJVHM<sSR<Iua%_M>)K
z?ZdR6`nkW&*}Oau;UHb+<}3<kQx{`U>Rf~~JcYjzZL{KjDWlAexzqW&;G%AwG}i3)
zHx-S!Kz?YsWSD*mU3Ax;fHDYh!kZJ#7J_Wwe~MoLy!H$GiC@Ug?2{2IK~$&3^L=`&
z*OA}m#w|e#he?#5w89hWo2Pe=C5ph7^b(LqgKgNMLkmM*G@W+F=s?Vwz28uBSvMR3
z85dfy9TZT)g5VT9_zQ#OFoIQ6oG_kf^k~3AK->b^7$)H06UF86&Q|+^7}m3q(Ueb0
zO|XM0>M%m##=Qa-44ns*JkCt00GV~)C!beC4qlFeZuf`0$}OW>i0V}eC)P{1z6LtE
z<s&Hr$+#g1+V#;sNCH;!qup&i!`9=&59#%TL#4*>=Fv{;hwfLIvf&AbWc1a?#@+%j
zNbECrxFDsjkyFdJBUwDj>;absk+noMa|t&B6M*3;hsi{3r3u!!@4YaDr5f}p+<h3~
z+?eGtjt@em<@NtP$YV%pnq94`VZ~#(2=>Pi_5&Z))in*>c(jEm(z>=*Ph6~zK|E_$
zGiB}L(%R`W>SLW==iW|YJp3tnCH!dLQo16?eEY+GlJwhQy0{}z2G;xVDP2@zA$m|3
z<0k&>!z*9@2qWa9Dqb?;`jxGdKp*TOHos|vyJi-^^PuARuAD_e-$f~I7CCp9-$!wB
z!T6mZtNPgGQc-7~5#vjN?=n~9-JeN4SV%tpR?lu2TE}>QxV6y!AE|4|9+Ot}du)Dt
z7+Y~cP%wrrEVtlQz(PB1i6280H+}!6WEN~e^{XObR4;EVD+4eQVfO~bo3(l;hE{s6
z0x;qsV1e6c1qy7gXn%O$a~Io<4`T=k2vrZRTFcx^U;jGkIkiT-`GYTZZ{i<Xk<M6t
z_qz0T@*c(mjc=s<JP_wAU=9u4fWSb3;2Ohm)!k$>yJ>Wq9#%BqFS%euyBbG3z!Xrl
zK!oGTGF9K|)ZZULUcq$k#r?(H7lV`so2WmxRG8-!Qd{L0Tv=FSw+zI7cF!^9IQDuk
zrF)^J&i{O-31>KR>Gt7swC4NV=j40k!|93NupiR`akcB&Z6H$wERjZUi%`3e4@lVl
z`yo9H>Q}TfwkFg4+aOMN7kX5$I`~`T^~R7gn2pUDKnp64-*MnVjp6DdeBN__Rf{do
zAYyKAJrF)P6kHU3H|%)#*xf<b?(>OH$&=Tq^|2O?{A-RmU%Ru1(P?+)5G*k!#f0+~
z;ir&hyC||%EXw6Rih=DEx_)f--~8NDqC6Di8{8ZeA}WB0kX9+KpIC(7$pjxk*r^~k
zF(mN#tk~diOD`HAT1<578Ca1U+33$a6qv?5Z(myU#ZiMUJvJrK=pJX*dF7~HN9PmA
z6(qbeBGOHFy_+&gW{C%e0SV^SHHm^2f)d?HZYD_!HG5HR%AdFEN?r@ORDi?ZD(oqu
zha5Wal?tuI5OVU~-iSs=Heyvy+Xx~Bgc}+^9O{jrI0mOySYItRJivN(hsVG8hOy3D
zQXXKU-G+P8p~|jx2}Ao6rrde$iSbcXnzkUOGl)+TFQAnLmjN<m>+A_I`6&yLSpwO>
zJylRGQYk8!ts2)u$TE<%xiy9Xiw_?v-CnC=AB2%jm@e`u*(<`wd71@P48(3LDhH7F
z@ZjL7&8V9`$S9GUyy4i=ZPXd+oM{{!ZbMa9Q^s+zvY6PcwgfJM?V;mZ|2Sx+3k>iv
z`7waJFuKibrYXO?a1|j|GVz0Z4B(CGMnr`!4fl>kqi{v89Ljof_`u8F8L*^`ES7G}
zbtCKP4v~Ja+i27fzm?ojsizWtzfw;$>-|G3+7rBMK9rM@?JJt%darg)Urc;FTrYH#
zv>$>%rH8q*qIcJ)pT>tb1ayQE%==*uc;L3RJ_*f@{azrGCq1`G0E$4Eg)#6@C61vo
za7Ms!6f8f8>>sOJrpS)*b9v6jGHd!pJZ{!8XGrHPk9=&rgg#6T>Kut3nYQ%4@re49
z$2-W`3R7^sFc|F0hs+5UzY0s{erac)CJ@#Uq5k}67b^_Hfj+O5%|7{tsRU`X{keO}
zVuQ<(&3&Cxp2waW740+JLVbf9X<$#4*!bNcDjzMh^|wpzp@h&;Vi-EW|5uDyZUnf3
z`T%;O2n(@!VQ+ivkGoDBcz?WJ)Pn8iW^wgar**eS^idWA&M~u(W<fmN7w|EyZngr!
zRRV$$e)jGSqU(|J33R0Ja05Ba)i+;0eL;;_k!U#JE_>^(U?##DB`GM?ppJ{k4%tVQ
z-hLQK1OBBHt27H^Q;?SpiyTb>$VDdaeV3!TsV^q<>ilO<w>z8h*;N<csH4m@Er#DF
zZP;N{oNx5!2jIZ~0FQ^JqI#v6zat76pM-1gJatM`Ieo>}6}_PkEH~c`a9Xv~K}IB;
z+Hsgx6v`=(=>uWhvrvaIHs0F8F$k>mpm^azt+~DJw%)r@Ne4fS<%4676pY13675bH
zk~v)*@E^v^7@q)_BVJSqFKC|$1eXRSU48J?Q}Y=nr~RU(1q(%_+I{NO=MQGZ%cq^Z
zcW@x5M>u@fT)Vp()62>rW>|`Xw-y6Ph^IX)azNZ@J2!B8u+pLfLBx0*fxZZ>(BX5h
zuHAKdoDH5hkJ~WdjF|D1=+qm<4YLWc`xdv4B5|DG&Z%8&g53W6XlU3=mov$1B@2B*
zc4-K1@q6LL^+u-?Mo0>f!xQytTw@5FXx^?P<{CA#g;r`_I0?T<`wMkQzpiwQ4-6$=
zgRwQ<AHiwixRU4;AoWnv_`kVEBx9ad;?;V)h~8Gm*c>BiB#Z%p<cdtJKt7sHh@U<a
zTf5%Apx3Opi1~m?T8Oe9vO|5b=?96AaA96`JntOva3U%LNU9p)7`V~k(zs>j{0PIa
zOr!`6<bnp&6QfAkMwojTwR>>CnWXJ(foXo!onND$@->avSN2^w)Vd8;@NrSAF*Fc}
z`3}H2PKoIo#sz4qD8yfHMFGE|uL1z;#BW$+6j^C2!LYsQ!8bu8;W+ngQRpf47*Pc%
zs+HZ(J71%G3N>fs*`Sr0#7m3&;g5XtN7)!8=E=r9r*}9xE<~}Fu56lj(49_<E5Gfm
zi5HBGpKHx&h7W#9RNOlRMy`QLAD0EFZmri%8ta=K#j&&6k|0exu35|Eg(yf-UHRzN
zhd1JqRj2M8frqJlIS)tm5<Pl==v9+6fk+{Te!Ux;n+EgfzgJ>z5C*0h)Y!r$3{UDc
zO8oN%cLom=>yh}=u=rrS71;RE@NL<!XMD|~Hsuy61xZa5dh-&S$zvp1S8WSgY_;OM
zg|NJ^De7q66oCe3AaV53ZQpnZS1w&C{%ro?t;mLyfvQU{fjZ)7VCk6m4qxV`9!%P?
zb&O(XU!z!ZUM+qzK~Du_f|U?41dsW{#-kLtFcab5JH3y*q3z2CUOn}9N7{1QL;)z(
zwI$eu0Cp7@><fLb#_q8OX7bRT@qgT8qA4#-hM3`%4{ii89g3&|NEE%+PC!vh2&P@u
zHJJIB6&*B*WGTPP*w%!Tq%rw0p6#buNG+KAP$^tQ7!f&5>xW{qH6)8~HhVi0wcB|3
zmAC&c(}vM^3rwWKP^QlVBk8hrj1Kj|zaR_NGtHHs4QKSCdDaz{=eOZy8}Y`f8$b&A
z-Pou>gihv0=S7C0k`Xc+Vof5NVcLUUo0u<MPzDELD5HAR)*v_fnsreQz#Ho0F^H>K
z-q`Hz32M2Emkt%{w+z!X*c>0=C~n@0ddXAnf@;NFnSKl@fd!EHI}Y3Y)DDpfK}p-T
z{D~_xDe*91c^H|Bh>3bvM|I3guGD0joPaukg6M2)zAd_cW&#KbMq{+t54$*8Fd);0
z))&Et7;7FCn7V;xJh*YHcFTCHF<fsSqG^k`CtGCb0WV1g^}wG+<woJE3V8(c0oG&3
zl~Jz4-mw|Q?eCjw$eqm)589vtlUuJc8^Sz79GNWPofurhS7dxhbHmsM>jw}1L-|2N
z5NCT+3<LR)@vaIh4A&o=CBU9)bj~Od4UCmQsOfd@Pg|s<iDaCH*GtG$E#!u}1t%QR
zAH%gE0+tJi3nTX-#6kdX+t6Kjb^)0r8UkHN0SCm23JME+HV_jv6Df&M?uYE%Y0IvZ
z|7r%>#wWK6Z;2|-2qN(HptA%O&WRia6Ls1z@PG@upn(T~eY2t;gJyhWr}Hvyj!fJY
zlHh!WdiAV+$pss`kgpJGVp7sCrI1z<8Njw)e87`-GDYj+Rl~4#2R1J`nz!1B8SfF`
z(^GdP{jgg1HK=r7!Eo~$Vc&iAbC?o2IDjrj!wElZ)}M1f15{8hlBc-Z&FulJ71t5(
z%w9>%Jb>7e7TI$DP`fyEU-56+hgKw-m4%b#Aw{-XWHSa!khC!}@lSW#+#0kEVU`pv
zj027?s8w;I1EWbIaW6XG)&<uJeWxpHnj=9S{cJ2(Xdh{fojG?D_u5}*6^r!;i<|qJ
zt%HtTr@Od<_`G}-0_+k&42?`8b0*Jp<dPt8FP$(0w}l<<i~4yb1I4Sy2)!;R5*@f9
zNRBgd{m<{V`)O6=4{8~IBO213Qr3^E0bpSNxh!q~Ed@lRM4scj&Hdw*)36Q^2{E$S
ztRUX$^$RsdTUZjCzE8Ze!7Z`qyA@;tFJTn#kHF&PK6z#YdXh~M%s3er?}9Kp^_Egp
zih7yReT^irLzl2)?jzH*4YrAbr69F3tc7afm<tp1gylpC<V6t<AvdtYj#)}E(O)cC
z$6HsLEhIl;rANH@7B(0}is(YN9p4p1Q~6G5<<wo(@iXG;eGR$_A<-w-{{>PfB&wf}
z>PE4CsR=|w$|1p|xLZ_hFljpwdC9aF|0D=h(N;%z$)t?#&`S_oWJVbn-4y*ZJwwt)
zJXuqI=EA<=M!yb%Xd0MrLI+r()A~R?(Jox-F#;bMYeu8_V6kJwX`0E_&Y8MMt;P|5
zfykV|O<9TThSsmB)gtPHw_DJ$0(Ni6LKmP=VDduM#5(rS61M+FEHd<B(@g*ZY2`98
z*&gIpuR9Me!a)51Mbq%>am-j1G8ZLv69Uj6rZvjA;=CACI?qsS8_8ptc~1umxJBH$
zJV7`la9vUa#bA{K){-GLAYEK~70MgKOy-TaM^s0p=wX-<Ob(j4itPA-hq#vt=ZmXQ
zAzBorE;2%TvdrzN1$(kV0J&_qG{ntB=k(Fyxev06r*>cdlu;_SkwZ9(+HXL8M&L+n
zu>x2zu$(-()4ifd67VCjS!Y)*@MVINFU7ry&N71kRSLNnio=m8%G@O0V8wl?Z)$hj
zOfz6sVNgVzukw;Gz~PgFx-nOo7guA{I=IgmJ6LE$lJk$8J~S{DF$ZAu?3$gMT0hJx
zwAs)HK&h8*cKk{~F&&fBUU~BR{~uM#LhS7GUJJ1yNV{<~ml&h2kB#ew=ni1sf1H=*
z(ezbN5SL&5;J1HD2Zfba<9wL~tA8FxIAn9-eH?7T99DfKyy=vo90@+A<kw`+(YH5m
z<m~nu!z{!g2Db()r*$sw!D+Aq>@WKZyc(lPlWZWwUUjr{LIQTS$m|Cu*$Pz6x`;i0
zet`ePgNI?uvuGN*FoKW3#^Z1OsU#9V&4_?Xols~k!o!a3Ww&0AXCfaeK)6|fWeqE>
z+4nD2mJOb{Ch<xyZnU@ERMwBQisRMM$_g~ds?X_T&$6#6I<J~&5*g2D=vu%E)AG->
zd!Zz{1K=PpSiE?MTpS<Xit11pUKdFa`~&Egs~aExfPsWeoVhU9$IjwfD5~8HnMI3<
zaj=<QcJUSC*q+0Px&L7>^KgEzksGOxHS6#1Re08LRurr++*k>(2Olh2_qAZ5V|XbT
zD!WCDyy2n(n3N>n6iEw1P|?9x3Gs>;92B7p<b@0wss?8z0So^xug8D$!AfE`2YJYi
z!8w>c2QuLW4t*J$jZwmK1Or2*7mK}BN2R3j!AgSP*>a#T)U)WQpBNl~Wh*((7B|ZU
z(+$DmDKHx3wP(v!ZMN_ek<x4z{42?3l#p)>0}kXgc2#DO9^_@X4?}}jf!Xo@iu%Ui
z{%HwxN<94EH3XuNi<cEG>^e-ml3fwGNXe~**1Gj#u^h#Yn@kV?!?d;N)zlB-E^)Ju
zi2N()90(Tm-Z}GL4Hla<hMS`Zgxw7BGNU#ClX7YB0L%3D(%W>SA6YnF2fBLuk7n|~
zTKF&_67B%+#pZe&|8ZHF>S6qXOU3XTGE?|7da<T>+U~OlT&Om9*}-c~URvkLkDs}l
z!!pK=M=I<^C;?*y`7Cxo!q{8TvpaCJ{A%zyEJXyZ(j8sUXu~i{yvQE5bKe5=Ps|hO
zuSR~}Xk0?y`us;<SDNoi+aFqMPRj7sB!PDT6sB>$^@-?bM$rmGhc0h6*)eu=J&H<`
zY+J6EP}own>QJi*?RRW8R?Tf333PZL-s{j$&+n{hUlzwR$9h^Z%&_~a@@g=L!B+ul
zYN3ueTtDoW?S3geC&%qQNrE%Ue3r_Wl^1?8dg`sgm3^~c)OP!#scp!V0Z4+Vh-J?E
z*ZtRIs+O!;Na6NnvS|)bs4!M?T%M_@lc@$vVh@EIm5rgu)=j!tg^J~AwZ3MB#X(xU
zFP?dfR#rSX#BxUX2g<KX>+?t(36rwO%+S>_Y*zUI`<X_;-`Lc4gMC|Uc4dD1(2hrH
zvF&>{Y=1)fsW-|eBR~rWL;wP^f1JG<WcH>Bj=8W(Q>8g1aP4x!W|t!zs)VJ7qWn7u
zTQ`e7>t9~%^_($~$Ml@4w_!D$^;GaRh@`@k+5AgoS77~iI8fTR^qR|KN-MmOp)YgR
z=eF6HoMOv&Jr&EHoOc_R$VW$%U3=fzwS$s%-n36ndZlz15WdN*f5*(!GM&Mqpi{om
zp$F!YZOCcdS|Fl=CYVE8en>YKD#cN)ai$$<w)^2bed94VNPc9+rVM;M_>sXdm;PCI
zep)0+7<wF1R9o@lk!4LF3La?3W#pu(vTM>?4c%@PV*bg;uPToZ-8Vf=uGtW=nOsSj
zl7*OI=w4~-gP)XLszJB!j0$!wIY0{<d<~XT?nXuo)ibp7VQp?b@xQ_s{u_yRv80cz
z0i~Y4J=-^EAp8UeA#ii=vve1ilISB(KgrHsJb#-WxX4Iy()Z`Prze4F+Z26oDgV^e
z`Ja-X|M2#H&j9m0MJF+D+B-MpC21=l5kT6x{<kaR+;DF%MYL*EMQRtIg5d)Zj63RZ
zHb}{E9lm#|w!)?7K8%gkS46T(a=g_(m2g%9=sM&wu_Ti?x$ap(GC8|nn%u&Mdab_U
z-&W8@^(}b9MUlywQF!e`Gsv<>8ESaihfZ}~_sYb-Anlca$7Yq1K<}qJ=YL{go4m7H
z*=sMOE<;x3LgtQ9Z#4jdil~Mg>_k_YX4e~;=Z?hkub=M8C=`Cwte<{y2n#xDC=pru
zIJO3<|1$C`f>VGX_?s`GzjWKh3rpZ>zHeqCl3*UotlQ}J=_8h8boYHPX|PhWNlN;1
zNeBjigpW$ZPq+5{dl{W(HxGFAHz@}e@RH#`=pGbb6?H{ZE4e2;OxE6`EgRENWw)K#
z7%jBM{^gP|h)bRd`lyU%gKY$_U%K~Vb67wMHIbaO6#f6a&M+Pzcg<^J%0KI1Ja;%X
z*M6spCU@9^1dnpRlp<Y6BY-U>E%Jm-{}b6v3A%FY$4p{E3X7`@J|Eq8PQi_6r42L?
z*Eo@XKT?@L6uiB7IuF*>hw7g!Y$b+EG$YzY`iKZzo1ewWcA`3yO3)dG<A+bJ^3kM2
z-lHT_K03kbm=n!$!;&hiFv=Fp9*zq8gYt-!Bwz@LN;ci{Bn~LKJ}#_z8QVV-HWU|k
z@Rn$?p4GsgWtyrwq_kCfW1!DYt~-=z_3y4+*u4f`e0r>TEU9--c@P3?^}ka8s?Z!k
zZ>*skeL3Xyl~q=Wut_0D607i^@=h7NGiTmvNxg-<2<pBH3e+e}+6;5TFGHq}JdFz*
z6gI4u<5--LNi@~-j@HM;R^oyR<y9568c*+V9QaKp3`L>Mr=Zg}HgxV#0h2Zb7DTIr
z5l{1A-`k&(XX&d~L7mWH_Q(<|$Ro*eITYERPA)j<pQ2BkV&675^F;yb9YRt*IgMYi
z;-B~fP&ThT;gNq1Aw`ifO$mibHyRFK>MLSyKWG4<1hV2~Vd?3oBI;d6xLrVE0a}K~
z;mz3jyHRW?MbQauLx+n6#G75a@bQ>!Aj+^_s`U;3o>;Ip+RUQ+h2@y$PM_#9m*!6R
zu3o}I6UbEnlRpDy>Vku&ANhdPVi0D}zF1_h(TA4op}pale7$=RsVKZda1~Aka%*qw
zJ%<TX0EMw=)TCl!AyNV)PGG5X%?w1ts9GPI`<b*7O@LAn>fW$fae8E`<Q6S#`;Azq
zPwb^B=^Js^yV9_29~G!fa<Yp^m=~5G-tgx3N)V%9y73sr%JQQXV;y1hY!f0nN<WZR
zSkdprx+8~LvoF1>u{Ls9WgHV5jm9gi51HB32Vx5>L6n1k?;6GxxZf@ol)#BT#dc3f
zS6nXzf++H6aw$1Baqfa7<zL{TtH|)1Ff9{ACag>3QDCaOeluf)&AEtCR^{OE#W%sI
zIGYSjA(YS&ow;}zU62SbyLGQ+USHFIWkG3`H5b_}LqenH^16}BD$Wrl|91UDwinq3
z_wqG`%L1l*I}nkn`IA1HvWR5iX~*NKm6W@93C3mt4|EY(fQTlqB0{$>@b^ymFzSFF
zF%$%+VI<QvI@H-n%|j6#mKAcTrPyR;{Nb+cp!%z~Tbmn|pq^3ON0hP4YkClelWvT{
z9Px|8LHSZ1Dqao=FWPTs@>ZV&y5a<z7f3{sS9#@aR1HK93q%YGBm|%ss{F2)Fj?RH
z-{|x&rid~6@pU~ZM$LYjG|_2pv>-Jjs^d7WutU9f!~H+X<}t?vc4G8(5}o7Ld^Yyt
zk-k1X=t1j!M4|QSFYg?LRd7lFzpT9nSW`{cD2fP36+$mklF$-*5u{59q4zEz(n0Cc
z5s?sDC?TPvfCLZ_QHml0BE2gFREVN<jT(6aA|fE~*}UKR@Av)pJm=nXpXcs}nLRUm
zF|%g&+H2ODS+hcZaX9f0LiA>I*})R@W|G`=bli-<ECkT@3C<RhbpSd2PlQ;2QIfdz
z51rovmZW?G+*1xFIXc}H+TCEP1Uh<KWM_6rCs~Rx<p;!_6W)^9p(P~$8aT5w@XZ~b
z=tjV4E9r4rwrggO3GSgCfZ7raAf-`qNRYY*G*`gbJ_Wo{Kt8kpx@{<#mkrRu374c}
zWB>>oP#L&@{`&(Ie!{430p(_N^rjMkTySzROX342-9kzLEBp7Mo$o{L0lr5UjZZLn
zA3|^k1QuX^_5;pqeVlYYM3I2)UjUuX<v)GPd+(|5p;mPO48bCnV5Kx?^}TO-Kz|Lu
z=oTa3NPL+r$zXmF7#-;=W)j@MZ~%=lp1dN$DG541pi;oarzE`@12<3(000O92|9*b
z%o3>vuO|H4E#yl8O11zPlYA%HA0QuY3xyH@E%$&4fT_Ru>-Z1A1SL;CuwgoJ(5e5r
z=uV$hLw72I?$#BM?L`T$g4Vy^0UonXpE_m0_2aC+zuokM`&z}tCaUd?pPZd6|1%@x
z?dYFw^w++Bx?KsM$=`JJ{dAO5`6+^*x1Cfw&K+yhMbnKL<X+$ZCmOEZIy^ph7(FSa
zoSLPtn@Tq;Wi3<`rGr~`IhIE>>Cx%Ysjgyk6a^dG|IcsRcK&O`sS)GpuEu92WsT2D
z`2<wgh@U(h-Qxcv<MHG73BdpL&|?alQyz&;`*+APnBo59;dA$2!S~&tA7TEZaG95V
zmLrDKUH=5+RIa0C*S~x8AE}FHLHfOoYNzj>xp?-LDBVdP8#+KX0Z^cTHBA}L-cLDY
zb6pv>6~L%B-QTE2e|H)v{-4=~XJx#<)aa`|U2#~N=Ct|sTI)aZtV!jm|Nc(^lxw-?
zuhalY)wHo*88%6Uch5(~`nvLehW}RWY2MFj|33u%=$!!As{p3|=Wv7N+qA=-eVz+H
z@BJ5`?$hOnPgi{I{J(|I167Ylv}V%ug0a`jQ~x7mHE5LqS>SeDxVNkgdzDgmwEl+m
zdiB}Fntyj(<-d2+pgT#8_wz|CPTukR{L3cb1Otcv-T_A4e~M0S|0{l<f7--4yh?HW
z<^K=Vjn&~mgN|SO`rB<Y26v{-uUGRY`bhoyb?!Lzl&-*{Z+W&&%EepN_+QD(e)%a^
zeo1Lf4+Dso-FsG6^jVF&3iywz6s>aBYvlzEdUT0&G+?1B;plVqtB}9&nfh8|g6DQm
zeg5$;5FiGU2Fc5D%O9_+@-C=>t*-w~hXUIKNNLR&RF9tPOr!(odECKc+R~vs8+%<v
z#X3M5Ha$FY{@(#0P6xMwk@My3|J?rT0nKPp@LmTPdreiaK@D<e8hfoAZ*_eY=L%GY
zy{>Aoat|ooJw-V^^I7-!`oAMJ-Z3>5%j>GJhXE*9{{JrcD6N#;)5>~*_9^=WIcCB7
zmX}*zR|DYvRi>x!X_Xg^si~HI3UD|#J^8;r0^0jO1^uTMpOTZBfW2PUf9g6s(*raV
z01Ny79`qgoJW&EfM^8@=u)S0NlCe&hA}GT4it@!<ToT}df0!cu2~%7{YpV@OZcL4v
zN}!pcs!oyfq#`A?yZHld6rkXZ3^slR%?zYo6)PWnP6eirs)PpfvSiKulrm^i-o^Vr
zmNp1xZzmsV`<M-msoykzWI&UzJbMjHjxc0^{&vOv)Q=Ni_&ju^ci+IS72byXMgdfS
zeABB~!qs(g_8YhYl&8vbNH~x9g0FFi@4?Z&?*XN8BY(pUW$-8vXpUee;|FiTeOsuj
z?pP$4ZB)Mo?Ws6i5&RK#@=S6Mzz*6XDl3sE>U2+Mv9ZKp7-VFvufy7+as)iCP)y|c
zjaDYT)s7)WzRiz6GGyQ~33eJYvSI&u7tQ>2#>wfPE;TqfwpQSL`J3i#K|VK50i9_s
zlQ%nNGW5Oo0hw*jOa^O4Rl3L)ZNS7Y|4{Op1Nl<quTQhqJ{R6h7z*<iK7db0Q$aYV
zvae>tj{v1r>%5ELK_P15!cP}LirLy727)!C)2Kf4qD+TD*)WR3X;gbH3I0%p%F}hf
zYb>b7=|L=(@6wo&$)ZG_5NnI?$GdrMtPf-T2sW?b)6`)lA*hOh%)NBVsrXf($J*fd
zHob{yfn0kV^}@s4fq1gy7b)Obs^quLb-!TaRdp`ZTc?@@{Dc9zlV=X-qZQrd3nqyw
z#*|Y|pO?}4s6$=g)qs|(2b^Y-D;@8D(%WH{$TPkFN&i9{_Mz`%^Z#40=xUPIK>}c(
zn4y|_7gpfU@<~7_&S@O`{NrOo|ERVwb02%sYtV|3c=dEj!{<Wr95x5vK{LtUeq*&W
z`RuWY57^@c0TO~vN$ck(qxDTH*l(t&yotOjNELdNn>0<K8Q<{z@f$bu$-tnD`qNeL
zBz)Q5vaxfP3A{i2<+tRWs0Y?FPT6m^Z#iYZRyGtIajjGm4H|$CQ_SYx_y7&q1m@K^
z%RIHN7L{_XtV(xPnVmHh%^CwcpPfNG5CsbKP?{U?0W!rYtz)+NOkj;Odu$Zn73%JP
zE8~nspU1!lsI7(!ApoTK>Q}p7urqt&Yk-orm`@7Gra+v?40Ww^V{P^ADEL=Z(SYZp
zs7O<f`?WyPGbhv-%8((5`Cr*)CLOJRf&T9jt#wuE@PYind?XORLOc+p0+`~ywa+#)
z89WLHnkK|VD>{{V@wXnxCly`G0x($cuW<}7#+{d(3p&YVRphamxLb((0Hg}9gI`XP
zKk3|9FX{m>66A5W%w8aN68V(kS(0wo7Z?ciDK^30qIV2Pot-(!oq+#Se3E#MYD}+f
zss9I;uj2#%^?ZBY7G=z4ntx#>&k48Q!S(ojXWLP%4}V6VLc&hl7SAJa&^20TiH;Uq
zdz-vh`s22*pcg<<sPEgrjV%CAu@(kU&p(xY)sD$c5NzOTW>O(HI;9CxU&9A-W3TdE
z88ujP9S0z{89?{JQ6s4TiD1w*TU^e!KYcl5(&LjFlM>AqW_#pzGtoYa;&P&Q(wB|u
z1ivzmfv@-!{KsgswKv5D765QB*lr{X0I9F|ByHwbqcGO$WC2sjJ39jY0qY*gMTJc>
zKGAjX<Xz9I6Jq&S0N#QE<9BLolt6;LWo#6GOjt;AlVrsNn|fc%OY(;N*QI4+@vMgc
z1&pGZ8yc9?UG<TL-r$5}UC<(8{c<!o;VAuM+yOI`zznU?<`&4|27Up9kg$@wCQy~L
zlTM)@9Q`S554_x9NyYoy-0B!br=NdP_#>}!N>;S772kAuW7s>$3~51ZdKLe?$$xxk
zG9*c`+0uOW)#RqxI^yg7a)z)uGc7mCDxv_^-eyZ5RhGV1blp%`Tfjg5yWnF*r*%gz
zLvdflbbOM|cd;}fGQcTGw$EJX#Y3roJx-#6mhYeioZ`gCMi1o75?Ps{EDWG5^m#`m
z-UeqxtKB~K3v0X1c{b@isS@L$^a6Kyca}#p!_i&DrtA5Q9Rc`?^PJ{{WHmsaI~UV8
z9M@Lz{glA97t+x=dfLOF0<;0MM>F2P7%=&IKl{uj8$X9u1r4T8K9cm8F{<KKbovYt
zfH&C)NFLus!;0YJ01b>E5NsH@4Ee6|WdTzT_5*9X!4i;_{59JYxCEe58C|*>AEWhJ
z(FxhNN-6-DSS4c93HBz+nw|kZ{%ilGjok^#{|b7q5?0XaMeq?^3|c){=5BhbS2{}N
zZG?#rgFNdITqXY{V(`8SuJ-&IpU+r8jk7#JfB!vU8S5iZJ?>Q4VcUCzR{>t9Tcy<n
zKlsRzjtXmsjtbk@gv>?6Ha2<#ZuiS;sDrEba-JdE%!U#lFtI?Jgoc$23VM$bb*10u
zP3O6Knn$vxz;Q0<!ZFEBXlwfCos+uDDYM3`K)q%E5wdleoeIUjhWMQ@fA~cHG~z+o
zH`!15vis?7D&#DuLa`Ca!aKP>{0W@beLJ{1#*=P~mZeX&+iYyfC-H(F*EpXbUOK;2
zJt$B$Mf+Dvms7wAF&88iu?I(^gp{Dm@h0C-P&EL)j_F~=NgCQr=7|=IOgdY8H`_qd
zoNbRB@#sE4dg=YWifHC}+ir)oXo7WC`g&2&_(@$y>Fcj4rcO0mwuZvp0=Zl$MP>=y
z>q0W(!6NlB{3R|VpZU#fvxeToD1ckP1`1;HoLD=2U;8Hh7lcu6Q8X}}%>q!`<_6bC
zE@0p<^QVBPzq(#?HEZZ7uqi+7)CS!kL?h(~haRL2_5dYM;xN87>qx{@TX;y)(p6@+
z^zCr@q{jfWa1M#R#1}B;U*jAS9RkdY?R-I$#TjsbK=QSRQ-~Jv1wXJq|1t3VB<>np
zi}916F6?dF^}Upr3LpkBTMd2Jg*t-Sg6&D;K+%U3K#Y>-0yIj%b{V)%G5$-IIt2oA
zsTkUMM@pnB<7?V3+77JJwn<ka{t>AF@9+j(xb_6dW`XW-&8}Q(*CETGcaW*0x^Ufj
zYLHD}XVwkE#W$eE`F3V*wQ-lkP>$RlfyMBI>~B#$n5DJ^6{#t%cw{~&T{nv^gP^i~
z=_f{*1&z0U|3#s%?;O32%Wt+nPAR_`f5)-u4NK%&&>M4LHbJ{~Hk;+#v4~9`^5@Lb
zUAr(Oj}c51M|Yq;%MEV90Je5C$3%g-%UDVu_DcfOhb$$ZJ?KaLo(WkzarP2dR(35D
z$j*-5iy2BZqwMOj0bs!2H*+kna!mqHTh%0V6TdfpC}+%ED|D&Q)XJ<*y5VWR^NVLe
z?YppvkCWfC>Lsf%jad~MS|YOt89Q2DcwQ{cU9MwsPj7T8zydG*)W@@JkBi7aGOjl^
zt+d>=$4+_%-(<xXn%dmLNVir<Oaeoc7gm%Vlp?+pdHg8lDCuuHXna$WrDps-uV)^U
z(`-3SgVG@U@|pNm*kwl-|L~IJj_-IL)2!U+ynPw8K3AcrsvpMhZLs=48l!1rz3hVt
zH#gTuf#j+Qn&3=jJO<Bz^aO!8ARs~ID;0ymiWod7jfq`Q+1C;%7(mxDqS*QP_;~AK
z(HbOUIhl2gMDXl<?QE5LJ*==^{O<T~$z6G6b5-_hpQ<czZD2`Rry5pIdoEK9SbYFZ
zByvhWVZhfrmkV<lp5}Ei$r_k&s|4-2`0uF>XCsl^fC1}BDd5j#DVJxnsE;X`MNQg^
z!H-4dZds@Wbx*vYwB$T5GZMufY+;@&c!N<yP7!OLWnW}j>(jDj;Bzh0q%qH?6FYrD
zBzp2dj`E?6Vb+OXf%loHoSdrPM4?_e6u49aEz9?3Fg}*Kxwv)?v3Fd*^qxQ<K<CmJ
zjZy4^FmwCjqZ($zuVh@wkgwa~`?Rx6=|%@4A_}~H$fIoH!{PKxS@p7?jbmy|+Qj|i
zwiz%%Id#bfmkN!z3_&1)OG;i!i=(PsFKt!~9`yrOA;N#t(QLi4|E5dtGTS;QJMI=#
z>pzfcX^EA6{E)K8ckg)I(^)sKm=9mY<YM`7EdATlT$&i+PUX0g%1;3eP?UrsO8BzJ
zSXZf%*EU|m*qXhMTLj#LaX_D*v%bS7`$CE`MXE}V^XbFd?<=LgU<{c-nNLn0s4Wbt
za$(*U^S<CEVr7Zx{=9y<nQgiSS^r!Er@2h}?TfigU#CLlJLPLsN+EU)r)PKPhP=Cs
z-BZ=fbnu#f3XAx&pNzxK6QJnw71>wWf-s9FH;a13dd0`}u$wMxi;F+WDqfkuJ6qmQ
z!hQZCJJrFnD;uJpHh(zxQ|JOupIvfZHROjhPGvJNz+#Pi<)`w>b31)bR$Wt?5(#$Y
z+S@W}qLN_`Sxh?SUVQm_O6}lONO>v5dVO5?Oyic$r8)Jm3Si88vZzV8Lx&FyYeX9c
zN&+NWAhoNpDH@fzVoU<1iGre&0eUpogWXg!MQ~+hRjj*Oftj<ikO&@A1PByVu~ILm
zsy<p4c++CDJjxn-dK)+Bj(-|p^qKzhMEKo#LvG!?b2WzgOugWz#?d;|(RIjC<#g3B
z(Hiq45qu=Gc1Z~XW>!b=xEkjAE~ulX2LGnpw)hiyEVD1ZZ|c|<W#p$y$WDT`>&QHh
zGfv>s2rTo$yNAD)-kFaVpkC%Ae=@nszXEJ<sb=VWsr>7=q;=ib>?8qT?aV#p^w{0^
z8YP_q8?M`|5jf|<63UK1o0tEN`-HiD5iq|urjIYQM?aR#pYSU%JZ<F;x+%5mf=W~)
zQKokTYZNgQZR~-ZTCKEYIY|h(kjaz^n{*h%4f<qFlBWP^%^!A?_i9EOQa>w%Eb1GV
zXjv^MvNhxVaIC<ggNxc9d#<T6P2?q;?u!lc57{M?d>(9fgB37YlNzJRRXGflN@1DF
z8}?w9x+67L1Q$kFh$+j^!tZ&?e%*@&uLI$wJ~j2y9u9*YzNg%VGHr9!wF6$ruH!R;
zq(9<&(BlGN6><VI?2bufcx6`@L!7#<W4<}D*?#%^5j+#g$`=nO*+cD_W!gOo7Z0;#
z+HH8=R6Xbed5W6WuS_>cdL)=UGMY{>el4#gmU%pOH|vK1dFu~W%aUfdD+-~fiKSDV
zNG^k3x9VwLj5PZrC=`4$2F9iJY&%|4*GZxvq8&rw@+8k`>sN!~R;CpE-67xp%jvt2
zqFwQfv{0G>J;NG#Nk6i}qRly9Lu5-&SQR2IrSD+&04&xqzzlSl#L~o^3oyRc=C-?}
zTs<QvqxZlr31AIPiDTtpdd}AG*u{MbV2co93ztpM7D7iiv!EQ!fa`Q&Pe_lwu29zc
z=uGO(;g9y(v!343f74wuI3OmdyRwy<nSI|&ctJ=iY^(@1PF~)1B&V3T6;M3h#T&Hk
zVk2-Y#zK}=FPRU;-f_o{LzIzM(5z1+QYoIi-OhGgnAd0s2qZWMWn$;!ZA3G}fh~7Q
zv_8`Y$+N^^g~nOy6xs!=|0tbQOx{Z1)Y{O+(1ddb8)SFIB$iuo5)&+Q427XrL-@{S
zMC_X037V+1HRD1{X9rbCrV}ARrGm1JL}utlx03I8(hX053vv{Nxbn`Ep8*F>F&KlN
z>BoLYIg#s$UZmNPEt%MCF{8NzlIQC!vB0+czv(_DVRU2+O&HgY)Z59q>715fYhvjA
z(r&f>)OWh;(|YB4F9OStcMZOoiiv8`Fc|B{jxv*TI|?k^5|ucVi?>?AK?db1qcL);
zMzE5opCJx?UY756j)lMIEK<YR?yb1%Bl?{z88Qn8i=@S)5Mqa<9~3HGfZ0JL$Pg0r
zvJPbm6>VaZ^jMST!OSQGSBBZ_Ae_?NBCos;7hH^+xNNdC8Rg<1E4P8y-pCp69V6vF
zL|G0xpW9h1?r-a_l&He6w|#q9I?N4Gwu9Q0Y!;KL&R+g<b4A@QiFpIsdIEK13!;#@
zW@=l;XqIX2b6^9X`x3BMqx^z?$wjS*q!R@BC`0w~T{4lMiPRPo6{KZFWIZ-a;@T85
z{1t3^Rmh}p*3DmKU$x73c;?gwQp?Mqdbx!idszPcICEoiFYiwkyY*=NnXJhhkRzUH
z(30`@1F5GprL;Fp{&7B3Nu+698rbO5Asbz0*VS_vB`M@yO|!6P`B#l3a5U>vIHif^
zTYFkWOUI2WMfFHMzfpy?5inaxK3(V4-?dx*d~&CtO2?$EZ>)nm*6<v`Oc=~UWR@<6
zN}$!G{JLd{8OJE{zNrMjNWsh#HIOU$K{yoEjH8wpPc|_lG7Ag+s)zk*@hEb<646*c
zS86QetYW?!@`9x{QA1jD<E9Z(Gl}K4ik%WfTcNrCLoG1<HKqI(YX1FbsJ+`Xe8L=w
zj()%4<y(VUM#pG_KV5pcnKNx8pHTf5;v5g@Z#o3zZ@K|4T<z>GECLY6<Z6dAHhEpp
zM*_7=d(dtZ!kMM<?_ZGHL=Om5l(!pUjau+Wa!)ft3f@lPmW*PZxHXW`Cdh|u6b~!d
z1dIOT{@iQ{Y%H{2v33+w&|GJ7Noo44eNjE|KI{FHi|hbCTMx6db576fcT|88R8=#Z
zysvUs__-}Gl2E%LuiH9gTLdPMqm%YUEnhD22-ktd{d~Ni`e*BbfKMfe6y(aGuDmN(
zix~_w)?{(Jm-n4`;dB_+n0RDocxuldErFxx7e}}MrekET?uJ0PY0w(Q#wOBN-SOQU
z=5+*SK#;UCn%$M7JR<gCr_1N1=D+hc*NC?8y!%t-XOu}z9{&NI)`!?QC$<^*+`-bu
zmv?xKm#`a%B}Wc-XBBE%0M)Eb7?j$rrkHWKa`!gxN6E_cL*4Np2%5}!nu%@gN;9|5
zCjk9pAyq>m<JhFDCVnW^EQ$Pa8HD=SrN+$x<2DsHi|uZhIo7_^+`iP~2BOri)6QJO
zzG&IvOuYyX%!0AT+>?d!PVsTLFi-#DKBCQD!_F8~E)*1`ILHZ6!Z%>_(?)Xgna+6B
zbbRPElS{><1@<pha=d%=HyuOvPRjB5qv~~&Zcc7->08e*A9BX4X22W!Br{P#%9t_j
z(n_*Pwiuy(DG9uPzI;;+Fi>UQ#L55QypS{H>j9(6MC!~0H8868vb`-_t>VYi>NHn(
zy%@A#&TEDr;Z*f<wYXWBONBL=n(-3-tCIOWB0Zf*qFEH*`Yys!uAtyz6D<P|if#gt
z8MsVa9C7x`jEU0dsm-c)4rIM5qZbIF*OfMiXvh~1#jI}RW#(S)U>84bLtsI&OWDyL
zP@sVUK`Fp2YW<D+>rL)qHGto=GKIE1A-nhmJ;YKT)m$zY$&jJReXkGf%NEW97$|GD
zP??PhTHMZ2<F6=9=*y1yQlcgMG;_x6Jm%#@t#|7Qw_ue7j0iXWdYQ8g=DWcG60SQE
z7_E<X4qPSNVE94+ezvN^c3TWOjN^VQkzoiz6I&2W819R+1oLvMailjN-&qc6z12;>
zh~dg?_d$`zSX^}MqnaAPb%~Kn2S?RN<=st{4H;IuzKmr*2fRTn{6kJ<!OBn?18!GJ
zOwm>m^Z2l^>vsELV4Eb@O~z;!lQi?kk4n7!_mV(l(Z}b-P$qsXX5Y;7Syt-#0I^ww
z*If_iyM$)Okc1U8V9}TCwt(+ubTm<qRS)~Td|lvT`J`-8Rfr!>{ksfGQs+;e4(x5g
z{sK+0sd$JpPhcJO2Vi~zWFw#U_UAk(CJYA9Q!KkgqMfr20QgCzb{9$OXyuEu-Ar9N
zA}@~4aRWAa;|bN^dLI641NJ^v`%>54c=&e^Sua<}#_M$88%leQY-qjjqFuE9JDFj|
zEqP+ONo()8vg3I{?}?TCI<HiXm0@(c!o)`S%I%s?X@zx!H3*94L4rZ2Y7j{ZycCQ6
zUiCi=IV{v7<sVg6oRNCjR2PmKnwXw;)o2owdZ<6%1EJ=0oM=A|K%|JIDgS^Kmtc-~
zUH+z%HXSD$x3$)d6Cc-J0wkC!dk=gpxYHs+-6MsOCe8!ZA4!@^ZfGqG8~sr2bE66z
zeO3e_z?_8{5c1?BttosnxM04~5A6|iwjP@;wcHk?zhD%G110l&1!(80xJv|=6tht+
zrsj28qBT?YWu<5@j`4HX8`brZG>0Q-c=`bh;F2O4i8C5CzW^O@TCOADFJ!FbW+j3}
zj6>;Ep|@OGe>^8$DMAr1aH7#g8<X;L1>qhilCpiN#RCay3}A+&u;ycmW9ZR^{79~e
ztvp55pczTqhAGj1q=bxoMpaNVuX9H&J~O)$yldh&HXA(<N-#dVpp)70`YxJX6O*f`
zuCn67OjGN3CacEy5w7dc_b1wD>yGSs`BfBC>H=&hburEuf+U0+$>n%)T9JKNq{wk2
zl)T{^x-(pj@84wsftV43aJXVwga(C%q)2lMe=mbW1S5%Ea_v3hk;z5k-lrPCJjYZT
z#EdeN;brW+>H@7Beq~Vj{1G{sA#bK2bybGG=8@612uF|F;%A?_O>z%#jI_=M*gZQo
zuN|MWNfU+stQQejx7e3?Nxac8$OQ=VHbdt2d8V^FoQ=s|19o1~*mH5e=d%iIOMUd2
z$A=}#tzKp?4vBH0!D~4KYO|GYtFuqXd$Xt+ScieFoR!Pzo)B#$=xsGO7EmlFWn0W>
zm=NL!n<dh;$gH<^wjE1$ZCAe2K529bB9Rpla*4P8t)OLN2Ig^He2JS$c1}%pJGroF
z#pmWku>e##ES}l(6|0LHhse`I-HppwnMI5D`&{JAplE@04s42Vplt6eD~~qC(F!GE
zw@7Hgg8dzVC1-9;7ONjt%ipFPU%X!UUG!p8+DqQVbfP9$Na#fB!Tff#FLiI(w=e3p
zFI@$C`^6lX!Q#>iMm}WK4y>lxeU&i3_}X+rAXX|$UDF65WZc&EGmN>}TP+q^$<~?u
zT|aYh#NdOpmu=s7zLB^P;*?(b7T|8Qz_?yE_X`NbE(k{;K1SylAs}EQ`g$ms5UdJj
z0E1GE82KVD7T`0ciS6x}_agpsFgHT9jFtIFtXJS|nU1LjjtLUqg%?3%ZHc;uGDh3R
zyV2m=;WBq>N<^tS?Xk0Iltjp`z&*7=9}ffX=z%!sbS7!u5YSF`$ecHA>=e(u7M=#b
zIL+jgQ7!MasFHHuV6^YrA1%O$QYwaArjebCnY8~+q5}O^g<0HF2No)3rlc)aCJH43
zEJ^E@<!TSaS}|az4DGUkPyVdbl0ly+o=~~*p7hJE;p7*IQV4{Q1fF@<*;#$G4lIex
zg2PD^mll|tC$AiZ2H6Hz;$(IbIKdW#1PIlkgG~8Fq9z5^8!KX_Qh7tVRoR`SYZ1)C
z2-C*K@BxBqj;!fZiAvUcn1YK(nPokBI<rMn?F;M=N47`}(BVY&B)1z+>LD&Nm1U~q
z5P>FLL%G%|xw7TV*~OSf&E?<FsQkBX`5wWVHLp@?hs;*RYPaQxyGjnPAF(GUt@s>;
zdlREJEH})5!zCIUU$EL20V2BAH)flkNyl;`u#{0R-Fb?}O$I<6Mg*%@6TZM?R_aON
zJKdfUp0H@sr;PFPA-kxJodQRmXlF?t5ssj2yn?%FP&Rj;VDoC;>}-GLtLjU7=OHQs
zoJO-h<uLI$(5fm;)4d)0onxfW##_cm2ij>u5G-|lbOwVH7UKZ3MBDY)vz)K>zX0p2
z_D8X^5ZTXz!Zaf$)mDpi5%JF$V*GY=1SmgLbvmW8&kh?g(DXhRc(`C3_bP@I#(HEr
zJ{-;mu9^^)kVdHu%zI+&`rFp5ut~4LM9?@@uClDRy5sn9c0cYE^~`KT+OUZcB_HFW
z*_N~2sm<@g%Dd;ZM9dG9u9fo7VYYg~?)jkZzDSAkjSQ2^tQiuOm(=UDB3E)-hl<Z;
zdz4E}l(TjT_;5evTE@%x%&c_3*u4r&YJgr|Dg`swfd2+Ezdd>D6^p`GMt{CU-<iAc
zp7-7xz3Nv*FV=M5M}@y%v8><!5PVaBHrL+XUR1BNhB^HGk_&0HQpMHQ|6#7a|9#C*
zG5V2M#F5Ysh7V{%RkBLM8@bl*yxGf+7Tqk8H8)t?)y9E?5nG96e0+k+j!r8-#T+l*
zdY#_C54wD2Y50|EFIguW%fmKaWyu!{xXUv*{fWd-)zIXx+97Q^<C8!d<eI(5<aOrO
zHXF+Fm)~^TS!6r3ig7kdV!3`d#tJUd=^N6no3RkmT;gRwws{aF7?N%^+^Pqil#8&=
z`Bl4NQYq%GdCci65MZKSDKo2~=O;Eq0}Wh8%@cY%6eoYx;a(@;zZ|G6TkaHC6Qd!N
zEQeZlE5`jMO2S#S_c3#apGA1{a11n<Yct`3Id@RAJr=nc!}QSCEcQm5n^(bv?Y9Tm
zOk&<`K44ydfNhz_-0sQ0WW!1Zbvh((xpF@<pnsstFL=w?93yvs+~=LjqoRBBL8gYb
z<ePE8Td>d0S(&siMfd{?P8@l7@|mtTpty+&V=lkydpf~|Xp?i47m@?ktMxvzejj&r
zM9PgRqvWPR1RrwM!2iz6*2VcME$I2AGR_Y%SF2p#_JYmKEn@bCi(S_?96~E3T$Cp@
ziug`9eQzF3hQlVaZl{%Am8xSUbnh$gEBm~Uf}asHFV*jVCLDfWeI8jtHEq9Ry#<56
zcdWIl7ldh$q&6(={d)5^3FwNsjiB;QQxvd1NC<PqO4}D%A1D91>tgK7Qj9HE+pe%b
zI$1TdXtKDe{zB$2SCb9gJrmN3Y>SmLv6{ZSSOfp+>5a7MnbTYX?xx)g(Ky6c1C^(~
z)MKWZNF|6CQAbeV&y`1+B$uz&ka)-vvG%1h?-50LYj=GB$Ixl=hai`sR1=RK5u@#h
zcjmH+nbiNK5;Cz0SPOY#4w#Y1D~*lK#?czGj=9fy$f9^{5{EoPo*k5b574hhe?uZI
zX+nE}-kT!3hv2Z>jGvSr8c7+?1<;}G%EKnJdFA4PZOlD?3bITU7%*ocxfyAfZ9nt}
zyMvoCf5dko@c?C?#lVJDKkw_NFC90uZuvg1jH70<Z=)`VU$QT<id7UFGsw=l)^1#{
zH|(U~`qARCcl-0Y@dZuBYrmLe_`hf9qk6C(l&|IBg}U<h`JEe}&hJOSn1q`G3?e)X
zlqcnJq0&;^gNXUKvD1MT6S=BZ2g0_r_u;-eiVbn9PQxnSX@33h%IoGt{n6C~MmIQ3
z!5~iJF~e6P(k8xY&A#{O@?gm?2-h0dc{3T+VaXXkBEbqJ`;_9wg|T6!>gT|x$gbN$
zF^SRV8goi#_c2;;gk|PpTq-dVIjqjI9|~)n>-C~t@3c5RT#qG4J{<IoN#^#;o6LA#
z@`fZ<0CSyN8WxFJwJON#Xd;eRHWsN5F<qKkezWM(2kP2|HBo7Qa-_+vrttx9hyVH%
zfV6Ubi$5>2IXBnKbywK@z|!_Ju<at{ma@+t<AnwrU+uIbM16pVyv6Tw$S-*)9a1cN
zOjRR!2sWu*4=jOYn9xnt!>M!gp^+;2tQ^j_b16vD0W|7~lBw}p9mhH2<eKyEf(eiC
zuB~Masgc~DYwE5s1+yS)eKenmzR%Rp3GQ|y2{z{T;Aedeaa`?Oc$ITJR-0M<Uu}$4
z`q<16ucoDzrtVq!Uxkey<qgk>S<c;$?1F2ry<eH>6KQj~SFSd`9#+>CS|o}Bw;8;8
zI!L+|VuSF9RpZEZcD5D5+<;w!OwHwKXE{Kt9ZQsxQ_6#T6eQqOUeAEV+&KNFR?ijO
zeHS5ct8f0j9z%@5cawrW4gsH*LZ3Chm5}BIo+^awnwcoiDl>vvhL|W|w}#x!lE}(g
zMHl%Th(L#D*mPxTIOh@#0`KmCV~M-w8%YL4os^F8ktN`OffZSdaV~yFSdRg8o^AYj
zbd_ljI8E3QAq*noK$Z~R^;ZcqDxW5(rkkP_k@3naRh&({owDuNL*J>FkRQYxzwm)7
zK%K8uP-`)4+{HZ)GQOM%4Qw-XL`d8hictTJ>xa!-tk^EcLVBvGd#dNunzFthaCQC=
zy!};kHrjgpR*UF2gMIN=U-!88&V_JjbhPE$<d?v{nhUq4WGJpH>y7{DoD^6T$Eq8|
z4AmyT4Czs+n`}02(-I9u+hqNcop^1pIF6)Ui7V;o9}Vz#aN{3!9Hx6{mW%RfOj7a^
zV<){FQ0(4c{C!zmTGjNyEKb;USL4$RwdwiXZ`n8Hr=i#Q1R(C&?s2kqt`$idqOJUo
zt3H75b($Wv!9RI*8yfqOj~NiAA8r>oMCu7!sqgvAFg$I9Gd+*rz0`C6GhTxo4wj=a
zuq%3PtffBS)b)#O>Pq4bjeyzb)L1H1SlQ!Ol<5{AwpC73EDT|`J=1<~C3A>J;ex>q
zPY%T5w10>=TrCd04Dy@u^8Ow*w>*6Z!fL~H?Y@)$WB47Lbm0ikcWVCJAB!FH2o`x_
zxSFdI?w*)A_bf_BYIk@$sdiEJ0nR60xm~^NjX;N_Ze{39jmC!ynG41uwfNZl=Blh9
z=LAMQ;epjlI^Zk!lzgN%gcrM>-5k;yjl`kiLl-+mY-2;u27L$`@r@I{Kw|yPZsD9h
z6XN@V6TYQ|39V1g-xMN>y$An_RL$MsVSnN-a_zQ{;M3K?B7B5`=FXVi);!-OkG8C}
zprs}+PoCQ~*SfWsTI2!$FSd8CZv0RJ)(Jgd3~b3@W_S}zNk)<BTLx-NPSf1qv!V28
zLk2PLyYmCk!11FO#ycvDyDWRE57&~}CaOuhe|R3y@F|S(!AU9j8)aG;%@kmT{E}i(
zd_nhEsvu=o#OO7k;WH((-3W5`*;CuZ<BOz8z?M|BvkTdc@NLhDjD9$pZ?}2LZa@z-
z4;#mZjKup_W|JZMlizt|3fW@7i+|JARC(txuY{94q}}tPaEKS2%{(<L^VT1#5Qn<t
z#h+d*T$v2<l4%itru_NJRsFon3PsTfV3k{(2;vz~AI+<|5BTsKiqg%%v(!m<tgHlO
z3EbK*$<eQxz1?F<wb%4v(>PLj^08p`CQj%N4=^YsDsRLA>J?}!dog%tv^0PnqOF>h
zlGC60`gWGlMA0w-8zLpYVHBtKM}&vRXUbcV(W*1bu{e6DH9)vs*EetZ68Ns<JL1?-
zR0+o2+M+3BUMRxO@O7nV+n`ap`xYv}Xpqy{)dy0OdN1O-{0k{1UhZpn5suEc3kUqB
zVT#k9V<XH@as&U+4Pas8)kd{bWNJpo<421lYb=+R-tj>f<a{8bno0_VZ{!(o_eER!
zvQ#{~+VV!?{rV7s03D^>-}L8TUaGETw6s(G{s$}M9-GpIZhG2n<INv0@w3E!akf4o
zWZv>p!|m%`W3kVSmOj*~u?l;ibrWy1Gq|8YxY``(n`+@><T$}wT{#e=#%U~w==U`#
zol`cpVqE*HuMzWO*ge+`8et9%uest4&#!YmXRQ$RHXJVts#fIK&>Ta=R;OaFbNsb2
zz;7AvAMi17aaUqqR;40SZ6i9*og-G=>TUD$Y$I2Zy4Wm_vWe7MzNsORvge(~qqiIP
zeC|ooL^Zp06Q0fs4N=}bC0d>*_&ElHzbGh|-ZZDXthq5Q_~s5Sikh32+T(wbV3}Ta
zNAPbteLwHqUxCkxOr;RknCILGNi-8)lpx8?Ey*n=2@Amqi~KVEf;b0F2YoVfp5KjW
zpnQ4G*PpT@u(%Jp!Tt<s^@gXhpl^A)d1}uF|J~OG42?RJJN8LAUF7(5VP6-gqcHJq
z7+C8wo0>D7HJ!sC5iCLdSp#{Iw0d9Pu+e2hOvop(0Gp1%;lg7^Q)lfL*|S+w?P|~{
zyl<v74xjNW@b+&#;?B2W24;#4gW7~Dg4omzeT*u$z-8RCLzzxeiKBn=4uu~^3q%wl
zUoq_?Zq&t}F4)J5uzp}b``qwm(oi#Q?=&G(r+3E-%K1mleLI+#tpq1Ma)Ye1vilqC
zv<L0BAxg7UuSngtO&|M3(-mv9h*Zs~@t_R$3U*COsD@8?dkV2V`gm-gP_`HGby{A*
zzN2143~mpe80pzqPr3&jlpX&y&tcMGYxE$QTc(CvB?x`3gIPpq(3cM<;F}cX0<2>h
zb4Z%zb0mFSf4?>gZyT(I2TQzxS8<G}M?{z!+XOhu3wm$si%--u<^I;aZRCeb@iC9#
z4U8HOLs(TC;iO`Pr>)LE%eM(7ZhPzA7Jna9%wy$?Jm2XjU)BN<4vpIfDM#Y`J}+g$
z`|sU*h3$rSX!y+;g?Z?x`H4Mo=*al+r`-8bCp^VSdsnNTzbjk~&+S^#7>leJ-l&cT
zOg~;NFMmd^`Uw9#BV*>z^UTF$boET%dGmNR1EVhy!@3#Oi0A7uN|F~|mL!*}zE?uF
zg`x9W+6(G0!+FcK-7MN_lYFH1E<ejaDU5hdPNJjVY_3Fy)DfhI&pXdE&eq!H&I%Q?
zr`E--+_}jb`pgt8iPb#f-pm_oeQThY!w`~K^~OOsEFGP7s}CYbBx!PF%{dwLU%LAx
zY;Id6M!+zU_mkaU4GyzP9g9`FGHk^90nwoL`HiQG%ibbl{k?;}t89%iXX<vOb3b$+
zV{y{39<}K|RVIh?+nv#{^_*it8u%>!teq~G2~sSx$)AnOt}^087<Qh7r7(~rqm3{&
zC;J5;K@m@Z!N5Q4jp1W^{{4CVebGOrxHbWLqSC~?Edu`V%_;P^49^~6qda~y?&%LF
zKwo96E>6i>N)97}(fN5QX@T-RUF3=c)r3TcRh4~0C(AxwvzI3YZl8-WvH}Uhg>Evk
zd>Z+Xlb+W>wVBAz63){RFL_e7XnA?iTWcay_UcC3f|Xuw`aAu>r7vc25~-LAy3E;w
ze*`I3ii{<rp>>=o!#}oz)*c<3l*8t-lZ%gsmR%mTnGX-*pi(Y5UbFYRoyUsXxo_tI
z2LW?Z-}5E2N9zw8KB?Wezw9Rj_#tVZT>O5w)4EXbdZa_ropH}_p46A|a=WkeWl<2T
zdD(}*AQfSqzkK46nP^E|A-8uRrD_LRZOlu!mms1V&rmTab&kXtM+}Y9JrtW{Xl>lW
z*2)~wCkmUXda^eXRIU~I+I~mXzU}pI#BR0Snw8=7Q*dl~z)7$~jTH?wpWAZ&HU^#6
zzC8Mrft)P`$xgX7e@zDs`Ua2lQ?cNMe$k8*%-|sKXR^~|$BTOb<E0@=C9CS#vsiy}
z3D&G@d;!Lsl_T|*L#j;!uQtJJNw_CMgbs@>$nh~7*A>YK*DVpb-ffgs7QNx$6i>au
zua0orDtqZ?wTwd9a8&#93v*-3rB-sAS}mhMqauPt)TfkzV@ua<?k-R)!lK5wMRxR0
zS(@?0hRL0s@;Fd50dO;D!f_zLQ|UE)fY1wLKPTA9vlc%RC4Bv8!aY}0;l2Ve^(2cK
zzV^+4wO4X^;a0-54|w?Zc@mUXH%!g63o0TTK6Xs7Y%^!?2glU!`#qeZ{(|^ZOcGzy
zq|<4F2SMm1w7M7L{I}KoNo<xxVd~SjXm1T^XV0r16f@CcmHNe_%=+`V8=gMwdIh)Z
zl+t0*-3UH*<)bWWP)4mi!*e#GBGqw`CU$BWh3OF>Q?)iG3dOiKF&QI$46oG5VKd4P
zJwa8N=PXEY`h;GUpBbZ(VPkb3y#+?yd+_LcwKK5YR>iiseRWQCTU89VLeg#0gwIRs
zgL_ysJyuX9vrDb+R$cceyG0jPc#A&ph>s9uL{wGUbr$!6AEw0;Yvj~YA(E3S?%GD&
zEfGzltrjNAPl4U>8j(L`gYEbA0B@<_MYmxr;|O-r5;-^3D^hzWrwDhZsXJpUV0^yk
zu{Ff~^jVMseAX!^pRqXR^GXh^r<9}8;dDFyQ7OA*TK>cE)Tyv?-QOT0XO@Oca+D-N
z3NWhS@#np_7{*?3ThF_R`mW5}@p2}Q2Ct<L|7{-al{K(AYzBpM^pPc@>z=%a)K1D2
z!P^fti{+G}%rP*2EI0Xi<Cl%b;f79?)Q3YH%`|7-lAGjO(v7*Zi$3pQC`oS5J9$-u
zr3`s6W_x5#3~9tETA?`{s}bQvyS~RRWG;i(69lyA!o@c`f72oL-LyKbYsCUvch2IL
zs*ehAc0QD-o{RI9yu&xTG&rpu+7xDq5)D&6RJD6Jh!$+~#T;?{=8IKA@5@mh^x28l
zPA70}b_+vC_(-dAt<8}NQhQ0yxsbUmUtQ{d&lLgIHr>Jmbv-Xi_>chkaq&K>0d{Ss
zbx$H8eUg0E(sb7bts^j0>Gdq-w;1;>*fha-`9L>4)+=S=GAm7SfR~5>543e^i!~{(
zKkImgQ~4wXEG}92sYkeHNYkWQzo^%7ofq<ML$nD#oRN-I15t$8*tsJdOKiUwqfGrK
zXwu3#^A-WrzVS``Qhzf>X5}({c#S3rnWnJ^n-JM7{@D99NsNkb7-X4s3o)5}@!}O}
z+4NSVnypF1!(`@+Hp#vb%w-f9_3}pRAMGnk(XGJ?y38V+kOGh2Q0r87j1OvJreP}k
zjBINqE?-U*V?(qhM5#2(b@TqtFSH?;?kC(FttmHoBWb9FZb(K4Dv(sPY@o*_IJ0-F
z8kEKsCVQcK<_X#AiMsdhKFdSIn|eErYpU^poMKis0yzXmaJ8=EY06fw*fTts)h28t
zHR{TYKOr}+=M6WN{$+n%99wK8@osd}v`Z9oC04R9Rf@m0O~=<{eDSShkc-QE!Au1w
zpEFXgKIDE@u&u&bBNf1RYFI+dxtT=%a10Yek<V3Mm0hq-%@I8tRiqRsF;}a(f_(P6
z3TOPom8<BD#f+`HB}=v8hVl4s)!ga3?-xERj9L#LD&_ARESUxDwJXv5O@hm_4CS-A
z3>Wb$CaE{GNu0Bpk_m|K(A!61&NgSzfp%Xab*?PCP<y$m7`03iS#DR&{x-m`9FK<|
z<)(ZxDwzSztpna#b3WYd9Egr~+f6vXx4Z*wcP`DM)ed9wK?0^@s1aB+M!F6HD8h4J
z(iq_|a|i8>+=n=%b(Sd#71*KIVZcJXYN*FE2PJBgm=QM_SM1yYtGV&Fd3EM<*2-eu
zdz@WC8~9)Ri_ZI^5%CPeb6i=zHt5^JQDX|D1^4ySUWd<qXcuaH<~mh^6uvSdMT|6(
z{sw(KXQityhq29-OUm=WU*M(TPKh6hBS*g%@<jdikv(8Dyc*acin_#+1nkstGvKSo
zjE)|BQezi6530Q8-l5yan|H_Yn1SZSe?GYOR<^>{V(|rYq4i>wUG>T@P3+Ln2Yi-R
zd*F3ICd*?^Ig@%Zr%a8+y6z(E=kE}Z8Cmu91Lyh})fF09m!Y#dSie!Umc7noez^a6
zD>vffJJfrm&Ej^$>9>01<Mht!{ly#hp~4?`;#Jh~VehKj9sPNJ-POAjHIY^JFo}1G
zXxiO-;deVm52TnE77$pO1Nbi5m-Lzt40y-G-TPoKsyx(cVZ}E_Wo<D2h_$)<T7B#}
zm7#own=cj^VX6Kfg;o58!u=dYgUu&KNN2z1aA@m-LL^tb9Y+O)6a~!fE~o{vE-%)|
zg=UaT*a*yp*ntq}`9l%7f0R==-b}TqUkYts|2%75Q$#_yF<knk*HFgsbSQG9UiuxN
zZEe+gHYjWgj*k1Bch(@rE^!rUcbZ7Oy?r272W%DoP3MhU3?U})42Tlcp%Z(mtEHT(
zipzeU`dQpJSkp!<W^>l!AICE#*9zE$Z6##({$SZ5Q+1JuucFKr7{mT$0<oq6@7#v^
zRCa@yB#vOC*t{(U;oCI(kypB_y5&UVe!q5f0#<+Q(nN`SaQFgXn*@27(HJ>U7EW4G
za!@>PW9V6MVgkz6WbE$`nCqHm0Gk6Xg7d=u?NdtaJ*peO_}8P&SH(`na2e)~avEix
z{$}v)Y=`o|CTwd&q@5Ar01i1X71VE>BkPwQVEfh|46ey!pke)e$Gv>)AS8N@eV%9;
z3IbePWP@**&}H@h$x8;T>;`|y#Oeo3elK}8+nbAim3uRc!C8Io4_icY51Z^l4eWjV
zqDz=Rtn2N^r%BynRndl4q3;s}zjiPphWCRq?nybi8M4;apRo~}tho_WedeC~nK#<|
zuzRk4%gk2oc1UT4qPWprErRy7hnTOIZf}^y?Qq4ra|W|CsQf7|Ur;?Zn8JAz!iue#
zX!dpqW<|!N2N0Ho8(B0a=@xl@&&<}Bdb;;AEEMl<OkT<JlQ!^g4ba01ni2+W>$$zq
z0*w+6W`|wh4r}ZU=at`h%N99N>Z(Wh;Sw-*od$*8XHOs=2q^LYO(!MqzLkp{1zg!Y
z2ykh;Jzju=tT@%%1Pmty5^NTN5~v;7-0wNuHa!v-xO~3xMx}DZYLPm^TpQF%G+8sU
z9TYkj<oxsqQ_(O;c9B?%u~%Duri1^%n+G2ro<|q@x@E=PZc)y7Gk#~baPgL0)OxDE
z2LGqA$-L2Bu>AW+M5BUG)4j!E!M&HmtC~wI6J<+=kDH!wz3*tv{pJ7H;`EceKPhQl
zZfoI|f)J3Rx?|?}+tPB*<1C!g!1@6zb*cxmy$aqX#t1Oz2p`M?3x%F@idV*B{^V4Q
z-hJKr7il#R;p;M_&7q+Wa=n~&P`vR-{%KF(GpeDm?OY|Z%A67!u;MWl&Uk3@z5&>!
z4D1kC?3_+{jVV*IA%D?j@SXAp_OXSMtADBL<m!ky%+9FoIt2a^V^5HI=(288ELO8I
zc0CYXSd%`GF(Uh*R^N2GTor{vO_ErC)bGo@Rjg62!cHo{%pDEc2s(hJCv>$jL^Jw8
zc^^W_&nU^uy#OmfImll|vX`OI2xPO*0FR^j^T$_Y`rqUwLTmbGwc{NhN@rJ5&BjZv
zR|sneN0g89ed7O6_*_6`mHPsOg89qb;8drxHo%Cnq&Qv9otkTrENLN}jPs(p(HeR=
zj<`5{QIzArkw|~Dyqo7n`H(W|eE6+ljl)yNH4OUiY?$CYH(sO*FVmAC=8_7~7CctX
z+N|#WryVC}_U={nbC#2rO$Zj=ykk{Y`nm`<Pwoqq_8H4>$=?*IqRqt<9h>Gtgzy`?
zdREU1^4fH3{KAI8{yp5L#@zjTx@w#h<Qc{bKJ}%CLRVHAT7*o3VRjGmtcL+pm|eXH
zrg@6ypnI_2j-)oW<r<Xh3EW|y{r!akFHPPTsr5@)nB)%|5&J#Zzk+LYE^U3vX!5m?
zSxgVF;tKA*8aeU!fZ_gE1<KOXSA!LnV5QTbF86W=AN6lK%f3${Si@@V<;L*s?|rAn
zD{IZ38j8%9eRiN;bC2D*AkUelt!VGGH{#?x@to_eMTB(FQz;#F7c!c+wN!|QPn0J@
z8X4A3cx>Ls-FV1LXj*z@+wU5igL_oDaEY9{_Am}AWH;6~l+qT^C3t9Txy}NaU)*Hg
z?pp9s5AdnEn9%KK-J#WCu=1Ua(E<{SvZ}W&|8bzurN}Sfz}B0v!D+0YHk)g(aXDKE
z2j_B6F}`q&Q`?Pg+dCcjH{EQXN7YzmA(a*c(KVFpG(A8J4ZrO$z!FPMzw<0EwCix$
zpghYc*|e#!$1J*Mi&Y}3iN%0XnjE#-?`B`dO!>y=^5^1gfAFhfQU}ob1L=+lU3~0?
zXK?Bdu_1De_G=eXNv9gA5)S58=LXgP&WpKQHBL>h7yrq<FA^QkHfy<iD9BZz^0X#L
zBy4gsUrVMF@h%&KC81t|6jAxsM27JSVb2`-wx2>c*%HUg*r{4=%B;Oq+LhVa*7R2+
z6WxkKaPB;E2{nJshQ6XG=2fYK84(m5ZCMXPY1Y1xGx?dKvCb#d9MjuIv|&#y4RHMN
zNE+PiAwsKbqh9?wLNmJE29Fr@MzD^mgvdyJPiagvzweARvzF<gF2{>K+xp1T%#ZW_
zg;(ag7naPzK(m#EAbWXNy0tvV?uTt*3O<ec=-6B*A+4}H7WK`OApc9X-eZ`louNvN
z;%SOCdt`clMSR>0y;O{}R~pk(RIcsyA^BhZflOzU?pR-ZH^fiM@ebIlp8W_@3P_h|
zEm@z0c(q+S<#|)|Y*o9iB0;QWy?l;d%qaNvCU?d4vkZFnUFeW^pM6=cZuom>A>N95
zrEhWjLh|VkuRA)-kuR7HL<rrwV{hJMl@~O!K0qJp<2#}MntP3JA_lDJp%dKY+Vabo
zy*aY?^bU5FIAOqWYPGH9Ky3W#7`{Yq>4=peGN{%S#%q^ZFOuaDoq`Cx?J8XV^v_w{
z%51bnVf$7@YxS08y+%Bb%G2-dOrOHej$in(>-XqtAE)`dRt@u>%(}A0YnRg1hD2lx
zM^}s-!d@*y-^EjWADR%RB;YdF8J{_6Y;4vB?kn7SV%z~DF?g~{tyCbmfI}Px!}YKm
z0s~$s%R=D9g1*J~+fT*3$ti(PjaLK=wIr`dH}T;Y-%RTIzDmf3POQB!s(E2j<mjG{
zTXY#9Gk#22%-PKLZZp`0KL?x+>M}<qaBAcVmBs3;7VrK|XF}Rd7E!<j*^sEZ^ykz8
zlRwagxC6l8JT){z#&Tsw*!SVCKX6c<*1{69+<0$8u9+BwnNs!GS0#-pZ8#n$_>M<|
z?T(9HE<OqnC6ag>L(j$<CFQYD!^xgSBXK4kcug5II9NXze_y7b+!|*&(Mp*y8G7@q
zizF0+srHlJQrQ~KDP*qcZIi@e5j-|T9<!B)nX}5ZOTZ^#TEU&1GW$23BA3<k{HM~U
zU-rgCc}CY+<BhR;@ft&JJtnEbU$VcU>n;(4R}aJjh%`wZk+#TcY5!GJQ`kGRh{a3T
z`u7<7Y1yLu$J=>1zN;v7$XJ$xXK-gP$8?*wJU?SWXX@9x#pnCXi$R@d-gIUc^Ij8>
zmb0eOB8QGjnrj!$RP>ah`?Uq1#cG#}EM~|#&cDZ~j$7W5E*Bnu6XN$fDJNcq;aA)6
zbF9ad?8gUTRa2t5;F<X85bkp%$0dtQ{kbnRn08Mlcm}6ac88-fBOE6C5$9uUA*FSu
z)7!k98luSLI0oHtKbIf*9Tbu%6V%vC2={g$%8=quq5mH|eRo)r?fZU13mmz3Vu;(+
z9BC=)D}uPmQI>m;a#dz-({K;OnVFb4dn<F5tK3^t(T0>JrWW>+Imq1oZa?4S_;DPi
z{}enA_j8T&I?wa!68LEzS#^=~`iQ^Z&$qS#kFF}wSgx-kjO+6}&a3F>*sn_I{_EaT
zq=`{D|K=k`py9TlWxvu!WGwTMMUf77m1%z%zr1fag=3m4X6ymoMCwrunqy&9zA_-6
zym<4jx@Gm-4U{lW`hAz3Wq?EQ#i2_bsUi<=+!3fRx{^9e&Q?3`Aj0;QGWpWMP?DU%
z-Rl}!=XlU>OS(_0o&W~bCvS=1q&(MMi3WaV6)M{xi<>=R#b6Xw(1~qQsd>%ByRYt^
zEF^@_yGvPpz&B%)hvFmrSM5*w3y!fyhaoaIX7`d8G0<OV=Y{L8)a-94Bhpx$VU^i`
z@sA8-mhzVNp^s8m<X3n%_P7wbRrq8lp9H8QAOn-_An}lN0RbM0#F=xWxfPP0Xh4$-
z!A#_yvz~Q*C-=Fuz5ItkAWnW!`k_&ix2ST~!)~>gLcP!gqm-+4oIQp%0X&r+@5!8+
zK5ashf<v6V(*d1N#KOIKXsHYZ+N!|3pNr3ff_kgt;sPN!!$9sW?oXPF6CrAOrbF%e
z=OveUZt|Gj%-}2`iwtF+x}|bc?0tFfHt$nE$?6|oMyrWWXl+HmjI>73e$SWT-hxpN
z1l#Z?Xxw#|{E?8eRkmIZibVS>mj|B}dJ%c#8CU&f(21{ohvBCrZQ*0{^j|lRdJW@k
z$LiCm%*LP-VWf$v?i(IAGd^gG7hkmNENH2ksBtW>b^Rkb6FKe7Ta8*CHp7;?+O?g!
zx+jB0C18?iZ6i10hMJ8{tzU;~YbtAN23-Dyj%Rkgzi~WA;d9x9W5!#(S7#rjaAWu!
zPA3y?7GBwOtv~CUH5Dny8FtR7ir10rj<v_?6w-k+sR?pT)YC|+zE$M>x%7<#54TGP
z#<%I$3M1XESKNbdh^xTZmC>Im4XlF;J;F%Y<iS#D|BDo;`=+}OY4IyHzdtW<+U}%$
zA3WELDphEA_2&&gzx>F#IAx8T{*3RmRQt0mU6)5sKToTx8uyj%pO%4p7-{w{h@~5?
ziM$>97A!M%Rk>eMz-CXnxuebYOhGoF0P39w#GX`JK=F|enX|u46)$^oqL@f!AEFJU
z6#EIZVq0kAM^=-8O_PcWBXh1^MH(mj*djP>nN#p5-yG_v50UiEFi}r)pWF3o#7WzI
z0q?PiM}18}n>N9P%^Q5wMgKKAZ*xi{TpY5@n*wzG;H+^Q5peffRXVXA);bgMVN!mS
zX!R#*kJJxxpLYQQn|n}~#`&@s_K8(!E3Z<&>csnCeYk6DLULcLRG}nh6)y>gSjhI%
zzp(WG%U6lw$tXGIK7#BMS+}hFiev$YSH42I7Vn#VqQKDjcInFx1nYtRlO5Q;1HbhM
zEBRh~)g=w>Doue`V$aR=yW(Ur*rY7763Q(M+U~ho-I6xpP?X#JsI9Ble5N#on2|-s
z$GN~xD0&3R-9UM1ph(aRsJDXi(V@2C0NT}KE4)v`TFe6%A~QxZeB!&yg(%PIRfySH
zSWjw535u_Gb|Je6x6Lbe#WO+S(mduFmFFq3ja@cb)hBe+=H7dkVxD%>6h~$CldoP{
ztje^_Pi)EyLP;OlD}H~elrb-|tk=3ozWUQSA$dX0dS0JPreEy2^}U>3vEgV{PPd9C
z;&-P54lKeez$<M)22ZxIzTvNWQ@S7@l|7;**mT_FWt`+C`|z&Tbia#Xyqr>Hs*+Y=
z>iylOMevS-sF&_834wJQ&4$L|vv6rqyv14JSd?+muf^$9i=>iO;(oNe(O{ve^?5`+
zIRheEHb~%(Mhqu05S0;XyBXW{pZ%_dgt2mBZrMG^q|y#B@4|>QQ72(l3e$x(N+9<N
zEpbovH>rE~?B)b%blq?3jt1$(!O8S6p0DUbg&^Q^pTP5M%mFn+{R!8&O~l}N=(Qii
z^fi7OIdynfY{P7^+M|JCMAff{AfjN$AAdvk{i&K;n3bTy3Kg{`xyRAj&1hvVYX`Y7
zCM<1~$X>g_7QTI~cUDmLY}xcCMJgMZx1j-N{5gMxqRQ8f_yQvh4OK`o-?8zhd+)bo
zo}}G6+p<Nz9;vElu6ZFPP99Ab6|1O+q8yYOtsPVeDvM%a(<2b#1!}T`I-yl_?7rLw
zxOs}^j#d?SIu2J_@sPKVyCqsU4pWD|*2kUowL2vD%NT@4eRoHiSKhUY*9uX5@x|>B
zqkc_M3uC!#V?*?{_ihI1Rr)DEN*2D)-{b`B_m9Pw=H?&mks9{+O&m_7<r?Al;2{E3
z{%?0)r7k1q21@{`xi(Y+@r`dOEuX{_0+;oaW!d+OFEF)rVO0PxF$@M{q+!ICnrc0n
zSD^ib3MlgOrcOfE`J4@mC!AM)@-F#4-1KQ)Q`7d2%B<WMqNY)H^~z(cZp3}%Z-Uv&
z9w*;!F2e=UoFN2TX)0em;#PPnqh{P)$8#G<fXICFrb!$EQLK~P54#{(3aK#?dO9~k
z{rv%cKjEr>;;cYBq>nwa(0B7pUbW>zg)`StqB)h#XUk3*Q+`+xZKT#M6y&YU$hjSg
z74moFDgdESGc_?D#gccbq+{<gS@88<v(P|CPv@)cwP*4zcUq^_is)JE9sFaB9rHi8
zO194XQe(M*VW2cLaL(ZO-y^c=t59XU6SUlVOH4}<P+K`d0p4Q07>&P(&E4Qz3|ZXd
zI;&YQhKw)SP+T<(M&5r<F097-XYzT)c7{uXJ=u#mmT+tNb>-2N+Z9e3I@o)-Tr6=n
zhHz6iZGIdq261f8+cYkwXzOLZlaSlc1$t;gEI;AwY=eq#37cZVzOM?ak>kl9SXye)
z3RZdC?S8)Wsx;x)xz?!@9Q;q~A+i|dtP+Jbg~0H}auNZRbngYZBOQT|tJmcPRGx{P
z=K{X*CWa{JX+m#M7q%`2{!Mp~u|Kj*GDM_#ooS=YTwEf4=H1>ecqAPvJYeL&bs-wV
zup4}n_%VCS{S6P1_=n7$7{iU1d7z+xsC#nUPFk}}`IUJfE&7DL2lta}T2BG`!Si2?
zz6<g15wOug@qKK3M`tXp@k^gMy<>v1Z;+>8Z*n@tF|9d@Wn1Dw^!-60SHKfD*pdvj
zKehHfxjkI>GxC`;A`0{O$S7ftB+*NElbMI!wO72Qk}rL$KPiQmeu>HxW{5>Ymk%=E
zIru34JtE0;sc+Me@9fL|;x%;4`Ol_CU{dS8T9qPmXnP0I5R+M@l1Q=hxDScu(Pj&b
ze7nwY#%)Ft*!6$7)1i*~Xec|Tf(i0Zu75j6i%=;O5dhitT-TIjbT7UxT*4M|eXf`>
z@Ag_9Fn+_kD_0CI%^EpS<kN<DWkN%gqnjn^-3#^yD8R=?;{7ykKc2cjJ$wV2F(aV9
z$do+ax-Sr_`8wN>TA07Zt<qUkw<)4F2Btp5Ms41GPBE3m5zmA4t35+b3C(8I>=E6g
z^_)e|oDvCQlHnrrY??-1+sO0{cdKJ)w1@c()5C0OI=Y(DgKE2PiF-JJ)c5fabrzyG
zv0OMfL2#XAAXx(0*_-c-2!?<9dqk`x4=iPw+w*xLsv27bk1cZmF0c<shR%me*Oybn
zQFeRqTd_xjo6F=g^A()9s=X$~wJ<y4iA61pA?n9JWjtT8kCT~Bw<msKyrR4H45K;<
zXQIA3z7YwYqf9a?8iX^#bLS<jA8|zKjjSx(Z9BC&){5xS=Ck1WdH`AG#y_l=h;TyC
z$BiNobfERtqa?I)HLj;Ge84NSQmEg$D~sS!=#sMT$#9<sks4BAgri?kn0iV2K=+>P
z7&m^BJ4RZ)e^}%;mjTB*X`Ei{AZC79YufTeFrog%&Aa67o4+*n2?fU-JRh+d%hD=$
zB@ux{-Ba%j8FPkC9iA55(T0|zRHO?9x0OddQOx<eY?u;=8kBv=<oz(1EvcL?tTN1J
z&PlErP8SF@%ni5tF5JC1Oly~zZsImyOiyUKtprBiQJ!J&`g5xF1t>^$KnWALf?;Bl
z2&<uh=$*EYSivzu-tLnm7FUZ+*dD>dB<a#a->_LGe~<Kh-a&F?7XFa=&Ju!XVn*L*
z)$?Av6vDA3fiE`nUTT}Ud#+97m*^H!mB~_7FW(LK2M!1gXrLUhHK8IxoXM*cJ0p$|
z7uF#omF${v_{u)}E!F4Uigl4(Y>G^RfyfDs@lf9Nfws+17+*g@6e%g)&F3UplN!O<
zo;vI~Bq9PWhwI%oHbmINd4#;#QrnP8MYMu)q;z=FgY^SeEzdvNSIm_M@N5^{+ZX>H
zncnGJ8JI@TrnWU6;M;NZ{O;}`xFzonVm)mo*|DZCvhHLX#f<7%M2$(A=mKz`;+D^X
zl@$w-OgL)6Z6WkvsAJvbY@8qLeqSk%oo9=bugTX3>r_IsgSXdev`BDz**7HGdM5gT
zXleKIer)c6%DM|v=%A)!vPNyWOa=m|Lv?Rb+Jp*q|6?~^g>iBjigSeE_XY9_21y-n
zRG(ptfgC$o>Ktv{(;0e@M#&1(&sqLc`IML>sU!6{7biRY%jdJy5)mYLptTV)Vy~c>
z;&S2K^T0ph;uNcV6d+p}0bvHoxW%U~lKZ>KkGoHP*kN?p0&^L5KCen{g$zmYtPd?F
zi;nwF0S<cj{SP;u$=!M=<Tm}M>)}X>5aNX2G`<}gjBeKJU;%`F?uV!-HwZJ58(6%6
z_u~iwPp5){(x*J;C==TiBx>q@k#9EVq2cd+dPE8lp{!veE<7JEi4b)Z@95F*ixsEr
z3GJ`}b!O1s%iXt}*>6IlF5wTQpefI`T8*$&CnA*vcDyrX0yC+wEF7FFVk6h;IYtmI
z%Zc{s=XRIBGm4NSxTngsQyAE@#mQ;YL2ai~7-7(5o~Q2!!k?>DYnHIVo4P|kv%|QK
zvd4!>F5eJ?i(vAvMckOz0dwcJhM_lHB6{3ps?frr*80PtFbDjEgXz?n4B*00xV@N8
zZV`I${?_2F@cO6>bXYs#T#Wl`y<ukhAIviU+;627L_R=!clXnZ;mj8k#NgysipG6I
zLwj2r{@K`c`;JxljxC|R%YTo2&u^+0H>Cj9V{#WEU!y`n{_u9mheS-K%{L7%yM8hI
zx6&<Ji|k7>Su{~L$7a8f{5kbQI`|(ZCV)-8K0<~2bLVu%(4tEHE^Lv6=nR9j@;XV_
zqj<8?OZ5kUX%GShIcIB>B5O18+;np9HrW?kOv>{~a&E;|0-&)MBH$dtj4rE?3dbyi
z=^^e-^Y+Zr03wO;cq8p+;mW{_^8H?<F{v@gJnNzDKNH0edO_*q-9~d*hHAzA*lppC
zy@h$kT`zfb?=bk>u%@dhSAsE0(?C*8<Ng+|C%%-7jF&s(#6y$s3M->q{zqSJ%%ge?
z<%}gIuR@y9yu4c|zHcnxYyu|_@R{T#_z3qZuUxg!MqQ@_p1&1_-k#W0*Cj^Z00k<T
z(U2i$O71<Y@RHBbXKOnXZ;XZ|Fhe9CwICDLsVlu6hwQ8N@lX?~!=wIt`;(^^;2vNj
zY-q(C$|oU-cupkYar~j0(g~6d_v|>ZH`mo5Rin5cz)wgSJfK8~%65bB>iX~u1$l5a
z6{X3WLCnl5u%XFj^I5mE0V6%*fdQtii?;QyGNDwLtb@(U9NO4XsSIc3z|$kxmrCc3
zg{vGeT(dURc4c>@8E3Wr9-(|lbV{nW*mfyvFl6xeh9XDTxM{{T<7PNd=F2ChqXkzD
z0j_}NHK04pla4W`xE@sM4kC(80rL^oSRo(FI=KgAz@oW82y0!irkd^n&whL`@b}0_
z^ihhM0T~LB0xBA~=%If`7npKciE|r)C)}>mcyO?{4|>4&LmDx0#u)AX0D~fv*^=$g
z{MPI6sYf*8<ibT3J-5}UYHdP<4c&4&{vfk=pQ~hxWPCSP4Unnt?&o~plJ1E`YwEg_
zy0%dI`;ZP+#y{;VHn$z?W~kqM(pPh6C@No<kde0D%Z{@#pyW-K`mJj^FI1YJ&;;yb
zGMQ3g93ieiBnij@IqSHgkcYsNfJ`nhTW$52zt3vuTOGhzTIIPJo}UXpx)$S-0c=UM
z2y3zT+z-xL=Q4pr>l#K;QN0|lf#Ts#PO}4wM_XGgIwV4c>dbnI?I-6hwpAA$0?P}Z
z-WxDfbsimASMXTHuMjQXbLC!p^l_g!x1AFHCVL4sM-flTzH_kv@z}waPZMpoBza*)
z>-*!OCBm8Yx{x<oc(XVmV^F4{H#|d)ULGE`+=>r>j}SsYVPTycgn8%j*?Lz#Mnq9*
zu{;##yAE@!64hm&SMMD0@8dG4oLT4GRto1t5?P&%gQv&(M5{#hf&fkD*Ivh05XolO
zt{6Z-3VUb>D$;ULZq^I~ri{d<>WjO3%ebn?BR<F+*lFDPI$rrMep={1U}m2Fd8}VY
zpU79jc;le`2-v@i%Y62GPqo6a`gAb07wyhY-73kEZE5IbnpN=8q!~&3kjxZXJbL0a
zFc?9*`b909I{BF$AmNZJS3JkNW0TDPGrbI?!JH68eT)n9(0B&F?onSeeo@aT=7iy*
z3x{`~dMnGE*B(7=ggV37q6X+PzTq}+rz%vRCPoeblL+#D<E%`Pl+hCO^?3bmU3607
zLdzQwzw(f?<uX3tYVW;Z_E!2cTv)v^ri^Q|CEhxx8JQHt>_}{~#;wSd<$Q5v37<_n
zTKk~$amuGOdQGg&3C9(Wv3+8=7v1doq!deW{K9@_IVQ{n9fMw}+Ba{X*HRAEWZIc5
zAms=U5|JHm`SLah%24h^w$x<s%>a2*-rlz?0GsevcmXS&lK3s_!!TkF2*hJRC3P4}
z!S`ouz;d_ZUkTK(QYji{!R0O_n=Ogtk}0u{5I;VyVzTK&L1t~AX^7p`sb+Z6B4Sz1
z4y2bsDxhg4zc{8P_nUnsYW;I<4xsT*+?9S@0x5yejd%F}p#&Pg9~KS8MVy_<sbfr%
zof6W19)O_E9N*VE0d1g=0#ocDrqk(#PSCgGa>dll5sdg#dP03n+HGxDCwYNZ#m<`V
zLn`^|i{T<arF$l%3Bqd-diOa%18n`^MFcNW?>g@604P#Y0gs(kaeEvO*JoAfE?Bjr
zM8L=sBo>C97r`3{-Bc-=9W0Z4xcWQ+P9dpHtPc@}axAS$qG`_S$|kp(u7_LpN7f6;
zdgr3bv1n#AE+mp$%qG<1mBwt&bJcGisfsXe%i~N4bQh%fj>T9C7qNuO1y{M~lK_MB
zxk>5ack~`!9#IfgqAp^yHhA@Kc9_37dC#I*o)U<=P=$EgK7KA&IyfvP7-xv-q(IKc
z8R>loAl194>zZT)Gf`#*v&eJT{fCg1G7sRAfZ-AFaRBj(#kXS9jzUTu-P$kwDG`B_
za%upGQJ5sKZ;&^mwX^L^$wW+sf&3dTDJkn7dhRmC%8C`(s{EgElFS{v<~A-*-gLO1
zdhOb^92w)$ERwP7Lyt9=I9~;qoR=iGW7dvQ|CEuH(*=XUzmh;nXd^2^UA#i23wh95
z5m)Qe#b6qipgN3?X6y5gRG<FbFw+Nm84qN6{$i+kl)2AKx&y1Gz-#_5xMM+5Hld2O
z1+T*KHKU(-XoqpU89@${DTG5;gN!CVi5bQoiB$m(*KMb!d)=SI;@z|;bPE*^I^t*E
zXWcK2Jj(hUaD7lC8H$4Idk6aOt~d?-gguapJ=4?4X^WYIK-WDT)Ut#v7DsRI^H6?-
zky{mSK4JL@*_AB@{Uv7AM#}T>1yD6Z39kCC4us_(Bi=h_<xGw;gnqwUsrQ|9c|HS_
zH-#_i#s1wz^lTaut9<r^Lu1PZ_c_1xZCVnjFc$A^l{TG|4{EL!WC!M6Kw|G5UI^a;
zw302=*5v_z#bj=Un3zP5?*M3*4p;!M4RoUi4yKS+&3RwE<c%B|^W{6Oi$fyNN*&cN
z52hsk7vX`=lYGqGg=->g;QQiRy=TBZA77d{Bl&2X^3kX6Wb-OtskRL}K0+YGN&Z?f
zi0V%KR+yC6u-;9~^g_vEG~I8}Ifybner96L&}VR!B(pEFDz(?stJJ)HyS{x;VUDpO
z6rpB8gh5tu$7;^D51gS3yr!*l?xwBll$eC7ZliO$rF<<DIfN$<fVd0r(>mgBKn$L9
za^FB0qR`|;7qD^Q!2a)%l%mGqY76LM5G&uR-K1I!q0vR#_VROi7cb=&*={vpPbE8V
zKbaT$Pfq1<tIvn@bS8FEj(niY;w1b5s3|)&<XPpN^KaSO=&I?ZOsx9dMa{;fG1fiT
zsk(&P5~($XJ7vH2eR0qOf}0aeDvY><-5-2Dc(?YyP-a=h4&pg`8RjdGG6x;8YU8z6
zuq5O_YQcs2G_R|jZelpeX2jszCCE-t9mD0-V<#S}(Sj<B>;<UNr0*>(Mr^5vA??~>
zDF8~+J&m%6bjS2_52#rO13Q4yy~^QoAn$ellc#4_W~z)V5l=NM+n&3>ZCx;An%ZbU
zci|W}FNW;svWBKHY6fca#@Ayf>3yp6{2(Soz*>X4T#kKr-j*vJmu;0T6(@7mIjOb!
zG(TR8m3{on7YAx79}k%@D<N{p(RYmzLKJ^;+S#h176?i?A6Rr0dK(`IPv!|6)^cUy
zQSu>1Z8p;#(#+9zHo-qWbqu50L~R9KHo0fd{9oWYay|c7RipQ8{}-Yw6t`6~IBwVy
zWYwd9j;v}ecD^pQT7SKXy?~M9%yK5BeA4%mBPh8ZZ1Cc6Oa-c{ostlKRfX#Rq=kX?
z1p6p)XE!9pg?%HFmq|&fm=HS}xqW1*#M7}Ph-!zyp!skiSf3&^^vh7g;J#2ZC^ZnB
z1T{=V{*5{z<<O76EVS{hY^@6aUB-?~3tBn-u+GaJDj|hL(4K&~qNK!4so3{rgDTuP
zc%czpe7NphL#JiQy<9O88<?tlZ5qEAV2u9s$@+t&N|9JB6WOc_u5Iy79<~m^r^&kD
z{}NFF_@z)0ywp4&cKhNd-igXoJ3fdm`2o<LiH`Me_bZWowBcQm4PLp}9llm1eKoy^
zqscw;6|iJNs!;9X02r^`agXS)2!8WG%2vP7?}}YjaU(f+t}~MpL2bdQBU`Vdx1=ZH
zK#YXGGn}S4(!U1koZirsUNc@p^0%KP?+bNP><fmXz(J00PX~E&C`Tp)IO?Fd<AqfL
zZkP4LlZi7z!m3JxJX3f1XX|GOIzq06>(Yzi4_;T@1H?ULyzVh6%{JsiJkSOsPlDJX
z8xJlNy7<t#1_UD$pi7@90)}>Sc-d^f`>Ysr6e|UzGLHwPzO^rSCtMF_qUxSj35YG$
zu?MWN&jXzY?BC8&PnUx))&;tsH__S%ofUsBRL2kv3Z=q~+o0O83yND;4nHam#buD=
zu9EW;KDe&lJt$1SRv(kPgMC~xJiQy?r26rD>#UlQ3<a6OhnirjkatwbzhzcT{Qz)x
z09jEs>YM2ULsvlU<fi#gW1#-T)9Oka+9zFK46Lh2X<Z;S)P@%#ueewMST?;`Ih<wY
z!Me1sYqtWT9>Izh2F_fqNKCYz;iya$sdrh*@1M@<_#ajv^_YUa>Hv`i&m_2$vG+mE
z;qUf%Gy^Ki7Rc$au-35#*Y9@!dqnl`kyOA#wk`R(;@tP3h07GKnu_?(KD7-{cSJ)E
zQWG{^SUo1nrU~_`$a-a}&><QV;!*KDR@i5CvS#on3>{9;OS-%-(4Sc3?`C&@(k3@l
zj4Swc5%2SLvv@~IX-fPa{Cveb;tP*;!yn{FdPz$|ZwnbJMeqjIn_6^>?&70cR<??-
zIg4uBYFmOBX0#u?5UjLHF~>*qsAH=pfQ9#O>Jp@_Y2x*-db(wAbk?Cj7tlZUY&-tv
z-tfO?_J4ovY5d=FD<8ueRxqV4JQ{{G-gVtQ@#v8S!+9Vh-IYKC>rq?)q~0;%oc6Ub
z_IschuVY8u=Q=R&CvHl0JAc8yuG@4+5~0#vc3@hc_T1b209Pgz&BHXhR8-mo%y$}w
z(}MRo6mu_aw@iPJgn3JEZw>}#HS?;{4$`_*DDU^U_<^0qWUJgj4n4B8o+_@0dPHJ8
zD&pJhLy|tZR+VA|dSpza3UBql{<s;Y@jAHVWE@Nt2*mi9Qu8Vl7z9h@;O{s>4IT*|
zSI7K_{Xnt}C)l8pf<k4R80+f&<Iwr7xBvf=yLbd^OCX@zvIw@dE}Qzs>gDAPx~#!n
zQBRrLU6__PXVM!w0l`3Wl5Tzty0_EqbF^cC;y`Q*oudW}xVr8p<z}Y;$jpu90uZ*#
zn%(w`y@CVYnQO2f+`MpB*30Ur;Mlyt15F)Kc<F&E`S*nF8JDB_kjgt<kPZ0b8-1~c
zG5wt1_s!Sz-IvW6LrD+IPLn0PJ;T+Is*}>Me5?h4S=+vG<!|VJd7E~j`voKXESUfz
zrKbz>e3EU%-PjB#z;$ybh@^Dv7`MsEC($stIwn*G3>Q2g)fr}0b3<H4FRLr;T?15l
z(Q4c!qPCaG5DM_9a6w0lY=N3U-iQhMN<;BH&YE%MhI#4W2!37HRT!c_jALk~<1B8r
zgE~!RoNcIn3-ug&7xd&bVxpaZK!u1E91mppH+ee#Jp%s&xzFPh9}Ko8)xVGN?gLGd
z<Lk@}ZKLy5(}hlIbrJ{IvQ5LgyD;KKc+?l7__Y3+jpT(*?Ti?<TcISka3WRP=>!YB
z%%G}dVn<*~smq~&zUf=ZcR{ik7IIh6VX<enZ-#eQ=paK92N|ecMTqJ$2RvL5*x0hZ
z5=^K^vhLLMw%syUasQ)_)45B9NuC#c%t%~2+8XJ7JIG{5hE1EhCAtRc2!&Q`Gp@CX
zUf_&6akJb1Rc8^y=7JgZy9;VbXsEL>XfL&Fyh=CrWGXPeK*0Pr_$--|q526(o@gNo
zO%1-@FzVt{_iYH2xh6m;>B72&OTd6BP<?$mG1oitKNA-2uoQY`aw~v!jkHzb>*^eg
zjCU35d+j8{IkdRH6+`a*r?kRqx7+yhk*{V~!>#Q!hIE_H>yl$Hf_{JS8cfwxg7s}%
zr|h}V{K-BiAf5+z=kK6oD5}N}{~nPbFw4VsP{COI@8S&D?Jh`43*Wa;&B%l$B<deC
znjzi;pGGH__{{Y%20YO`uxu5KlL%Q`7Vm;d*rL2i0r4lwC@oauR?@WEss*7o+$+2=
zV7H+gY52Ti(_?(v<^DvA7`<|RG;6(1Y884K#v<(~iR1}Q7v4evSHd8(#ixx}-=rbz
z3fF|C`N4SGF1#*rXh&QBUNg>7Xi0Jd^B<;aX1lgU42g~72}uu0D+TQ!&sCCaMrqCT
z>tx<+l;k-N@HC7i4>JE3Q*k7{SbOXfNSi7(bGS|6CoF&~taw$ygx-^A1EvNacS#zl
zi@(?d=Nev){M>^t2wCw`Jtz@IVc5!)PXZ&N>l$rknK(JH?wKGe0DOy93rlvsx)<uP
z|6AIu?)T<3|5G>4IMK;6+Z#S>rZ<a*RXhZ*0Cz2+fjivgP5dHkr=X|#drpsiw`9|l
zEpw01BC63BPhC)Bt?oM@>I3)x^{k(-5+rWv^}byRXUTu}w5}{Syq|RZm+ojyU%s$8
zn4&XfJ;PkcVxYaM!E&nC)t$5kcB|2e&RptBL!jb0DjQO4a)t?xNEb*a*REyj4J}rF
zPhOP_69`Qzgt;S0&uaR(41ahwa4^Npaz=3?O7*(tElIxZ=*_`(V4{>pW6F_%Q=4dD
ziDBsXnaB>GT2hK`vS%|=Hr<jyas8~)|DUOTws~o7QAd2uChzUqO$rMd6qJo#PeLw}
z_28zXh3e&7*3Wqje%Kk(#V)8V785Sc3^VF40;lIQ@hs(GlD_0kj9s(><4x9=xLi_I
zslolu^wu-YH47{Kt<J|H5EZ4j4_Ze8C$C(!xG0rT-Qqb^xSH}yX2xhR^b4!J=JgEI
zNgEDyCMr{P?XK{;WOld2E$<iR))^1Wy!eD4Z1Row<*Bls$+D>;8EW>t3_Lrlzb}#U
zJek^_qNRB5`J(Mintks#VSMF710;0j+L;PxxsMzT@4EVJgtAh(S^UP=N_Om+@Wy6%
z$OXh3vC(aJ_fUN*cgbc#81#49L~_qy+OM*Y$~dJI$A*C#XH^>7jF51M57hc<FiP~x
zx=JV(#mqYivD<k?T*sLgJ!{6i#y5BY`XX7C9goyUS+c$|w%Cy<JYGBXpzW2iyCYON
zjRwkbvkm+n->QXQ6d3|erhKFfsI%QeWKnuVlzAfe(tP1E?3ns@3^)anp+0R)7jtsr
z?1PObx>H^JR2W=m`k|>xM;08qhWCsn_5RPT7wrV5T6qH8CO;w?dRK+4PG+@_y954i
z_!)p-($mPw$p%Vp5ELi8G><d4Od<C8en7KNi&g{r^WEd`HwE-{T&A~FgN(hYi!@a3
z^U6!11p{EtdeqJ#!4DUlP}I~(QP&hX5^HEguP$Z48D=6H$m{ipw!|My=_je~ws8O4
z&iWO`YqneGr4%p!3-LbIw4-L}C~npHVXfiDrbv#r%h}J&hf=Sr_X`G_kpg#4v0LA%
zf5T(Ikjf&l&Rnqy1HFjYQ2|lgVkkv;nO!M?Wunvy#s&X2asM4tF-im!z>~RUo}no}
zIZr>AlkI%>JGN|gN3n_i8h3E~xlm_uR#EFg;@i(myE7lNy4v<hJSy?0CAB4&V{r9_
zYF9i7YZfq~u)-vLn(99q@JfIl5$h#r(R2$uUy_b5s5(5$gnN#D2g&oP6kTjq&9x{x
zXji-hmQ7_C66p#K+1nnlt2y~h=`LjmktaU4`l&Vju$);2z!4aRsOf?h@X$yp`5~|Z
z#8M6)jV)CU0qzI^N`z1)!DCPkWJ8ZKSUO&4|8{o&?PJpGbLQ3sCb0;a_MYZ#&a~**
zl~^4UryfP-)-x5)(OdzqM{KD@Ms>7KkFyZB3zPAq)=ul1KYh-HFBv2+YF+xR`Id{;
zq0No_f?v4C@SbHqu}OiR)v-wzOGQ7cbZSt`lP(@8%^7kz*7KQ<eCHeUe0<&&1*>5k
zb?}Sl7ZiF6+;7OIQC<={^ziXj?hY8`Q~lO^*O-@Sz=FX{BzSqL*O)9OjqK;ntp3+@
z58Vvhs{K8>A|-W@16D9x2Yr`+2q^l2?d01*1pwQLXKh-fw5QR4{UdP+>d+i6LI0z6
z!Q5h|?hTiuS~xtu+lnO=>yzT;1)Q|pVnJ*tE~{9JX^D^8%#MAd{+v{f<E2$Z7f#8T
zr_E>c86WUhY*_>UYLwskLkKuE8Syp53M0^gmmLLQ!8`;7Y!G=uEE9mQk-X^$_@hc#
z7}yVf<>mgTfVf#^vGR|=RUvaGzZUK*fBUC^7U2r!R<Qal31B=}BqeWNJMhovG>@ow
z%Yoybp_d+N!{Wq7d~X%P+58`N?d{oP)o52M)$<_lFLwo%`ya&fyaL<e>N*poHUksn
zrOW_H6EZD~@{!V*=fHlI^4Bc5R~sg@vm+NDSRu{>g+s_*@}|j*4M`nYj9%})rgJ(|
z))w<cGK^<4qXmPy!dT}b`&ToS+7fi9*FD~r3HZxYtRV*r<f4;*yr93>N?VS(Kb#tj
z-S>6O#Hs8`5A74Rz36Emah+zL&`XRI4;-V&#;+Q;r^-De*SC+9w>7k9QTZNc|A!-Q
zeUx%X!!_*R<rpI*gi$;xu#%$?BP)zwjMf66aa4GzS4?Ri=1Ax&H-5_l3vQj$VN9#|
zuw3#;rWRO=sn+eOdg~QzX$10L*R`AXQrSLrJ`XMC$6n|^C6MRodfNWrl=Q{2!gJL1
zOHJVs*IvZ+e$u<3MvZN;)9UPOPEV6@;G0e(cz5-_SG)Sla^JW1MpTNFR6FFUZawpi
zg=(osh#*`cD$%Qa%D>KJM{vn%E7p+5eRRs$1|OaBC-3`dvEe-Os_MV{7gdQ)S=X05
zx>t>}9XYLy<wn|07PQ3ZJlz7eUoOi!j{@b5bWN92iue~w*f(u*?P9+=Wt5QF&*EYi
zJ%)HARcMZu1K;Ht-%Tn9o&RGR&6&F>5B|(XqLq0PC~k~Mfr5iV9my60P*mO(BJS#l
z#kv=aY$|KU5TUXE{AhTr$70X1lMA;0Mt5(!>%@!=8>f}K9^No`Hg^fQx55(MIPXNB
zDRsag_92T2%d-lo)i=nvWvi)5MxAp72bsHsy)yDcZij*oz+XxrxQs(1JC`*urco|%
zC=(ThWf_~7^u)%G{8M5@j!BIFG&iv!Vz%lc6)5>Eh_jGf`Y%ZJ4B3*-B(;~&&FTe<
z^ER}h0J-MARb>s+M^(6!Drq8`s3G;+Zj*#vyb|<jQK5q6cTxiB?6c3!$$u({KSj(~
zH%}}SUwNQvz;2BY;Ry`Z@b2l+G8tnJyn2x*uc0==t&%iq&^Tji^hH=If5?DGpj7&7
zEIp+^a!OkGg(pfQCCWh_Q^_jQIZ_s(peBZQ>O+C5<nIw6u$WlMqt~QS3ZBr=fz}g$
zik1j&-(u@j2-B@B7kEM%!L0%;pH;o&-`I(D#LI$DD)j|24zEmPQ(#9!B$bKU2nlF8
zFF*qDry1)iRHaXpa0r->CBjVd`yV0RMYv$^t_j`U)e#2rRH-#pik*GGUQp@KHH2WV
zC2Nr!l097Iq0N^JKjt7#6!cl?p}_q91na+OUI2-=9**wyPkRE`=+Gq*G$T3Edp3K<
z;OL9W_ON0MJvcMyWG>u%N7wbT2q@`mH#|Tfap7S=(<{ucVB?}N!AnDGQrDlSw57?#
zvMG97gnHL0Xf$o{Gw*0x)68iV7v#RMKj))lq7*NmL%;Sz4_l`^@-w|zCO9XwXvkYv
z8_DS)dB+yvJ9|!MC8{)N`ao{hcXv;V_M7IuEuXO?$r0#oHTZR1*IqT`0qolRrKIFf
z{R?K!YJIgYWhNs}*!OjQ&1e_rb%{&iY#PcaPHlCQ(egIimkpmUVm_<_PqIqS2>|br
z!F&uX5fC$8hoQneSEY&s3VA)@hZ8A!MTF<0GoOdCiS774MC>%KhzCv8N9)sKCkV$U
zqI|yG{}2u=pWBEdK@HO=n#rB2DfL75b*GGQ*?gP~PkhZdy{FJ6mU1=HgtCi-M(5Jk
z5re6`O)%Wzqq%jDvV{ZaE#6THEu;j~&i$1NfAT);^3G)C;`S@7Ki&kOpZ~d(iN3Qn
zkXmLc0;O$9{{PU(z1w^CPJxXT02)=?LS657?XqG`NLfAT=yvb9?7>2<Ix)5Rpqv5!
zDk%3`bu{V4H?&Q>6)xKg+Hh>^!I=ZWgNmY0<cN#^bylWd&iq={;C{ynXI@9n$3L&Z
zjC29}T%>`{8`{6JagF&Ho)#SefthQlk~qo8C(Tb395_=`|4iHFdiD=8Nd49RWP^Y$
z?r$-P`CdfnP^;04wvoJZXO_#S{*7{Z8|D0(@U3snA^F5OC-F@RRw?`{vb%cq@@FL^
z_Z+p$Z9*NTIr#w%mkrMD<wdC_T;X^erAq%~lI|Q}yD9KANN6GUy9eE4u|odpgRHzE
z7TEa2T&1$0d$o`$CX=!#o1?!YdCR#^AUJ(B1vqwe>-a|Eph?HH4Jm&_+6!H>iaKA*
zhA(1+og?7qN#k`ipE$m5BOU%$$_h>!@dlDEfhw)?HSb`ta95=ORfOr5qM~m4rjgwq
z`w}4(r$qAE)nA!zwbCYsxI8x(rgz;J1@7VGE%{iENliOomwN}tBhGWk9*EfAlaINo
zfWVeU3s78)&|aZ-o{SO&>jU!`g=9v2gedfaN}a`{%e7S}l=VG+8UEm8T78N4w@-vU
z1`efRd=^i;t_A-T2`oD1rFlP%Bf%@v>skHVu!GT2>i7bxd+fc|^^wA1L^#x~jwa|x
z-}s#tdA^I2F+~Hk!iA;V#-@p*`33hixucT=h01(Y>I#OD+x-9KL_!4x`vl|@6Q92n
z?R<qy^HHdKboA1@<!nZVrglIFF`qioq;31hRlR;+{U+VhhBnXc-~RYxFzzCZ6RbNv
zG4NI}VgjF*s%4zd@;p+!*;IPz`6uCogL`J`^Y!VrvpQwxHD_J#H-$xKcm?q8FE=~B
zX$-z4trU~k7fI+QQ%p6VUQ}f@HWgDNp?~ZJ?fqhi4dP@Ri{BR8T0S$OJ}9hh@1Aq1
zqY2EhSbb7zgZYOSt{|mC8ZDq|y8ISAWR2G>PAmoNW)<?gANB$Z7Jwypd?3}SVK(P8
zX!3V>LR?!d;DQD7q|r^B6klj0DG(QZ@?jPh?Zo|X^acbsU=>nb;grW;&J7mQ;caMd
zu1Un*mfsAgOsiZr0K?}g?QTGWpznOhtu+7da4AWiKq%Ie&Wt59y&d^VEs2YmP#6E*
zv`*%hz9!bgq`9T=y97J@$;9iE=Ar}}&9x41u@?S*3D@axHP|l<ACdTkx~H|JM$Y-!
zIZe4myZAI;V60p<qaYr7({1-BJJ05qMB+=|+0~Zd4Uvc4p2lJJvMU(R`8UXas(;+R
zy(LusE>JaO{0~A_&U;K$lXpS0#OOveyoIS|^jX!=u`xDXp#0@Ir?<U@t*djy>u-iu
z0|MCU)4vbhd*mK@{^W6LrsXdvKkcUOhIzDud{@C@)62?OeofT0L&l1djn&*2E_D~i
z!n(@+y`aS=%_~!9@fok&QxPL$`ZM$1<8_y&u`tAEh6BV(tAIf1q_X5eoi69A7Ic@C
z)@OdIoKkEIYyY@o>71`@QZ5@OJwf7rBajD5o~>~yfY<>B5?Hy>Wn~wyb7;;iPNc$O
zF>5;lo4R-vHlbUF%P{zH$J#h{qR;^-!|Il>kq2Bms4yHetYJrKU6Bli-_F&flVFP4
zrjkGI&covexMDi0RKP-!{2vS9^29^=5ReaDvA!e<`0r((fX*GHrA<V`pm|U2YnG4b
zJBO2KQ<r-kY%Uq{rgl?`%cO^(l@B!i362Ow(i#6o5(?Tm;C?J-Pa3FWsIEkE4F<5a
zd5qj_$5s!&*fVzEpOK&RoA_-;Ck!L2{b$nyK2M}ZPME^|V_DZ?iEIsLpPf3j8nDM{
z(473D^-^S4rO6h|+v0>~UF)Y+C7ZCpgGpN-{o{ALsiCEkQ8zk%@)5>TXCgVCTyzkT
zZNAj_%4IC?w2}M_oU<w4P~zi$#^gr>CCk$ng>Lm4Wa!eWGzSOIUs<Sd@DqBFZYsNP
zkox4QL*~bnM4fv2+@?s|SNsU$`mdvpKYB!*800S)eDpH+nc!UZ-y<1XJ#LE08X59%
zW&9vbV@jz7rtX&i;a|roSBKQV8{VHNY(Dw2aSJ=BSjt3AK)q~E^3?zv&CQbKi_!6m
z_%m5q@#rm-%36D3Bsw|=C=B(kt?ZMy7K@GdAOaU8{v^=bFN9H|%x1a&_1|v!I<TKk
zn)^vQxhte_fSE*^F{LEWH3J(9LlRHS$pn%0%y)dZzBijpXATh4T9$#Q4v)|kCBH*{
zM439ZK#A`&A~y_p5u}w%y%t{ecv)@^Mu_LGxM<Jihoq;c<7J0=NP}s06;vTlZr3z&
z81HH{Xb<~yE93x;-GC{uc^^9%j=Q)d<mN;&C(U_=D))?rjSJ^>%c4RgodkRVfUY2B
zXizFetk~2~Jaky^%2L$i9lML}9SjOqJCFavS92cT&o$dB-AbO%ShRnGthTyf-xTJS
z9{wf*J~qT|+QHCJ{%I|KR>R}l48l0_SqfIwy?+0y%)0>H!BB<@qVi7=I)L}e;|!th
z3s=e{Eu`ZNLaXC+7U&Ug9>)K%@Ah=RBP{SD*x>w`n(}d0u{NWw4MW5R?BCmE6poAl
z;!T}<>zjt^w%PIO(8ZjxYly|j8lk<EtxHtcpyjQa0rO|iw~^@jO94vc`<EmMaZ*I~
zTNKA8OS~TzHF*oa(U))XwN^ZQUYIjd<jyBvV~gyPWgWuVxu+C}i&zB)y=efzW+94;
zhpvd4&q9oZms1`HZJQ3C6_-=%E<oQ|zMwAa*dsAatMyxF{s=X#hx07z|2mbo$M=~T
z$u>_wK@=P94%pe>9){YkSY@4awpgF64M|7R?a@fCyYx&x`E`kq8eNba!(ayCB*q^^
zz4B<N+o2*bmpSev@FQc;2>XPG2TY>?ub>QVaI%!{0(ns{|G4;H#Wzr6srVQLmtqAp
zFV2(K5^T>RCD7eIq|mS<z;0c)9<X)J42~ra>-BI#T$8&vUmQDNJy6?Yx65N~q%<2|
zgG&>0?A2bT$(bl+C79i>I_7SoIXNPV<M4k{c2Vs;tTEZNl^0q+xmT}sSDX#qI-U41
zQnf-;yf~rfm<eAIy9+!(`4~niYf~ieE7atgiEm#;4epqQmK)ENz^@dxb?q=pdi+m8
z&dEJHhcbE0dc}u5xx|aYWw-stPX#^!9My(j<eNooYh!7*JY+WNES*hwQM@zSQe+L~
zr%DHDG3xs~DlgND-OhuW8W5?_7+QY;=(xG@XrL{t=Q3cmNN{9#fW?_^A%0lWi$BEg
ziUa~5smd^}WBaf(fW5fP6XNZs#|LWdL=Hc|6_m%hlll-*4Z5FYhh{Nj9XY*BXE^O3
z>8pwhJ#l2`tRDO;Mq4GW0MNpnMMYOaH4XE_jm7BUm13g)OBQl7`tCkZm~E&Afr}~Q
zzl2K=>Sani;T(9h4?6)DcvFNC!{7B)R?P3^&3XP$au-+vaR`N-7u`I6>pTYnKQ}SU
zkmJ)7*piK#R%xkm%)B~^oEzdJ|3<$uUhR1@y+ewNe#{i=BurFE?)o&yw&GAI2r)CY
z(J22-dj03p4)ax^QcPv_-y`?;2DYpl-e0qKY-sP<4f#qE=1urr#wV9lF_uf-sxyC@
zOVs0t7wbYbczH~F=A4_pe2Q%5^+Z|pLjEw)-|FA8%Myy_xjqs?s~+KkpR|M%lg-In
zuQX*VzoyKJ@oKT?!k*sua1v@s8Jn+DkK=d0HpW_;s5zuyBumODS$Z+LwYO)B0!UJ{
z8dJp<OJ|a7+R%-q_a!c%Kqas}x2om?=1aIVG?=B<P@PDNp}o$>JHx=&p)B;@0vJPG
zE~V=qr|u}PC^h)z{2;jNUCX>?ldl|CRZH{_gb^cz%oui5hLLRXm3GIZHuXW}r;nW!
znU;w{qPzy*q<md4cPboExcYF}Yy>MtUKFrfb&<(gA5l!6iMiUW^2vUHmpfOlOMcGm
zSDQ_VO1b^Gb({@}wgk(q_&ZX$Gl!&B`6%L&VG_^PD~)@HH}%2D4{e%cP1Q(y#2<J%
zksi*$8@NF$udbob3d5&410E^O8GaIlPCS2Bs_|Z^rAL{La@(`#ihM90t+FGfFE!~r
z%J$apCGncV-y^k8ip%>KmzqSTyomn+m58dpb<SeM=SR^TrH}GY9?io)$S-=G67fNQ
zun!J><$GXd!y=zI_&fATp1-vqcQMg*Bs%J{#p>lo@!^YX^<Ow7Zv+~_MDHF8dE}cI
zQ`&j6X-L2y`A7`@x?9y-@fG6Wq$O10{-@PUcDD#?OFPN9rxuWcnOaj5oyD1P1HVwy
zj>)^x>S1$+lWC|}c^a7W;-ujJ+d-$GAaC|uxd3!Q#@ojVx~}Eui>!fB8Ca1Q4`7*W
z{bU2sIxR!D4ml}GUh>82RAt62(T?~hL$^N6z4>NiL`rwexQu0P2-@RNY!BFL2+B%i
zwgC+ta>#o}8);N@nLROe8GuATP^DV^FNT3PI9&Z={`O@3?}HRKt2!m_>2u8=Fk1EZ
z(*`Tn9r8lro{!%5j=#F{JU#ofSo-pZ8cd+rz)8IYiyim3t)H|`ZJ&OhM5z1sHu5Za
zC3AI`&)A}*%Lma0HLq=8E$<2Q{rAIIN}#2xoC&{+=fL_IJj^=eTS`|^EaHRiSAN$A
z!<J|dAzArUvJJnPvz1AWp4Xap$ZQ?2gJ`S&UTi7BA+aSyminz%tn>Un;n|fS$uSe3
zo4n}CT(V$>-hxB7zt@+`d#8+kpFB6}hd35aIscPAT0uAP5zCD0ioXdx79|nrar3(7
zFA1`gAEUqQ?Q3?oV;@gROz{b_=Ku7V=JyHB7TyRc(b{&|dN-u{6wm$7YgStC98hEk
z9F|?XDml{?l&B0H^ycBJ=^qjFoWPC>6><qH-8rc;AySVy;H{&84m$(1*y}ut@#k}3
z0C|@J=;;U@aIWYzsq(kUI?}hWzH3~dlgawSHzi9Jp}R&1OzQH`8>wEoc730C&%@hF
zSRm_>5XaPc!nqfc<ll16<d|;2=3vNnOIap&(>m1l)K?Fyh)F*^+v@5#I@UK{150e{
zA{CO7cMyolJmJ*{K_sgPaa%{ksernI^+X$KpR+2MH&P|gGD;b5RG8*!N{=LGPCuZg
zTR#114Z%0$uVk3Ck~4nVXkL8b`EgyxVzsYh&UK)6z`~Chjpy#Z>f4*ty@fRF(^7nx
zJ}Ld{q)xvEs>#rKm*3s*Rr*Bm+%A{HCwo2VQJ2Air?)D2$|s416a%iRebto%h<_F;
z=PQNN@0onri2zBMDRvKXWwu+QGa2(61D^`4v#yYaqV~5gh{+{CT|e89!H{O}xs!hP
zjA@?vG^*fSYjNcMX)!#hw39+KYIN|^vyT<p7m3UQE7$gmvWB?ar_ENx&cDR63jW)J
z5EINDH=Ox%aIfQ3XW}}b8f2hzsM5@_N*&huy}>=H@JZzH7M@Qx3UpF`!a^f)A~fX5
zI1n2p+_zvQ)a5gPHv?HFr3Wrmy-@3534{lnKfZMS-OJvhvKorswfT3DRvzJH;Z;lX
zfV;AkSuRnD)BAj#O@qrp6sHoVok3WK*WKhby)<OOpwBD9Aadc8GLJSjuAOyNAEV1<
z(@Efd_sZ~x@cH%8P4`{4Rpj-8R(s`TW~k68FZ<h95e)|u#?!h-9fgA>;$Y}dwjbw<
zjQ<{)uW{))?%N8XIPqVxdycVJ>D(pthK6&gpRwHXa{HQmz-8iQi<ADN=9u#mFvYa$
z#MKYd+X+W8u>MFmD?P1DoO4SN80xvCsn9&c#3tzv3Z``nUNA33giV>-gtVxchNmxv
zh)w%JS++wksVk|$bokjcqCdUkb<i7*>ltKn0Im3EzM9kHV)yQvg03-?oz%L{tuQMS
zNz8fX<G}QKre{iR?%hga=ganMdsm9IXkJ3OroNIz9_lw~5vA3By=eoUI$FJ3uara$
z9tv*rQ|>G=1C?^nvy0b%9X~sq(87QDV*g1^$8?Qk|4mJ^oQwXC-=#hweo)H7R}jw|
zmz_H+OPGG8B{-po{Ja>Xc`*Uws2Ruaq#A3tqgh;<1l$@tpl_7TshpQXAFM^!?I#CG
zx@RxF83<@{Fg?J1{d;8W%7@r@>cN(S(Z@o`X2hfAxF=zfTbLX1t;$KQ2!ZLTx<${;
zGYMTk^m#fUKaAf{PTShuMNTzhV;2oeK1=%B&WmH8Zl#Uc-5ybznr@{mVt(v%BUUV@
zdkPp&2geVxNW$%>>Fx`91SoqRPE%**=ZgmMkL!c}9llk2=r<j4;(+}7_25_xaaj>;
zC2YK_8clFx_$04#ze1atRVcoHI>flsBAf5EL^>K$)}4H&m?%zf)6e=2R!;4c<Tlx`
z5CZy+l!o33Wy{1+s3ZP$Aw#HSQ=6fe9g;US;asj>=a{@yFRIJdN{QmH1Vbln&-6+)
z@$_)KWvtf^0yjYv8${lv?;}23`Np%&RXFYjDUozeapa*iv23Fv5tJKx8lVakY??oz
zp~fAL-O#5Po7lXIXI2@HWE;ko(gUA7x|%)xRE*uDZ942zPFiP8p3^*E&?Cu{`8@%!
z#LyQ<wKyW#*@z5jo{)B)sENdH4fh3CdS-YddC^x5{8KdRb>}s<wGNzwr*g<9+GS_N
z#Ib1;hShbv2TXra>3r$iwb#|DgZlk@F-hx-9vh^t@TU0E9Fuj5i!lY78}!Fde_~Vg
z6^>XZ^`Uzh{?ZFt7&}$_dl!B?PHo(gaBK$qBlaY)<Z;W;GLX~F2iqQOYJCf5&gKay
zYQ-4E8VW}1Pba9kxf)9aJv?PH|My6Feyy2*ys>|WPqLFS9WG4~8s^4RSa@;h-{go1
zhnqoJd-z*a*xw^Qgl(i;5+o(K1LMU1?Ah}|H7wO&W&G~0=*B(?(Hq9Yvl;yW&>p}>
zDj#`mn|E)}VKD??7sN`7fF-$Wit^IVuOgFjt7PLy>6?OEN%#Ik1qTRziqW)5+ClQk
z_oxi4YXp8|B(&~SvX+O7oVIHFj8z}>oC#M|Ohv!C<R5Y;=h9fBL;kG#7rl<fgxU|1
zp4aOkfr^SBh*Tisc-Hy>lmTiusW(IPHOUBGW-<+LACoSG77!3xiYJ#%{HA1mmcQom
z6)$zpb)Y-y>aeC0d#}eI#N8dmL27;fBLm5J=~soX94I7Y+VN@Yi(|<)+)1lYMMy)8
z(j{uOGFTNg<K-WjA%~;-(6HGH6GWbuYDn)C&9fpNJd~Thx6EQr3O*@S@KJR0Ti#X<
z@nq|FwrsptVBk{$h;f<;ue<miM##6QTcRbWRrjFQ)OyQt#oHmvxXzzl)rOZ-Hpk<h
zYKnF?K9eOt+-Ev3zWA0BRRdKlN7if0KT8fFKsE7CgF@4M=#c@z@Eg>b`eukuLM5+>
zf5(K-{A}@wfV@J!V6q4FaQ~0ldd0Sl^!nvb@Xj@YUhPz+=V}_gUAT7jjZhNqV0E*g
zOxDOkB4=B%(W8|c&qg+Qnrr<3SUU4~DA)i0cZ#xR&z4~phQvfgF~bML*kvinR(6G)
zl4Wd%24fv#myofvN|q$bzH?$|F;NCnjhbpOO_rhG<$QnhUwJ%myRZAYulM`?dOcq`
zO6hGr4SPEj=FGJBg)RJsn0yj1WU;Km2MS(+AqGqCnM3si$FNiqs_n|E{0kf$6(Ijs
zti+duEJ_SKgY-v^NP+{jOk~XdWc)4XUX18Cq1_|oKGe1t8)-f>B_*k4@m<y<tHQo_
z7mXlIFhMXmpqBvfA8j1awoCqnSum8Lv0}w5r31h0;!&{`W;?kYDBIhoLxzJFcczT}
zkomJ`-%O61EMRw7h{(Ok@@+q10#(Is&o3WLKpFfiyL2A1d3_~ufBuaJL)Dga?*r^a
zRjUUCifYsJ{7j3XR!eAsigt!n{5*e>!?~q%7aAj<8VNz|UWiYf?Z0z(*1U~;gnoUs
zsvST2LDEK0*5%{qkF2F_)czQY#D6#vLAR^)fL78C#I?p|3tya64jQ?)sTNU2`b#nI
zgV2vpQIG)l^y{+7Rh5l^!yhxsgd?3Nk&9k-%0Zo$9{G%l%%e|us*96bkCPgZqY4WS
zmwrmoQGYEGIdW05^o*apSAEPM0zYP1#RpjT`zv~*sm}gYpOv_X7Za2h6MDM-&+bBu
z1BvOA_eUY$@eUNNaQ%?fgsJC-1#{kk&k=m7zE$gHcHCILUOS=E4jJ}wGht@c6X3F!
z4@v0{yeu(u_iw<6k4~Emv4d6{3eF?T3>CL4EKFoAguE5}3J5UdO{Q0dfhI0S!6#YD
z+U100{_9Rfj<dAGwote$A*uz4f$7`|yYj3xAX}f(Nb?&BDIqr}kfNq^8Lqyys3wlN
zJjCRg|L<B9f&%h*iJsb7zU$S&pMZE^(tE6uBMa;UJF4sVet92ER5s?{Ay154j;mJd
zbF!K&7m3ha78El{{c!x@|HcRxKgxVa|CIkR{je@(C!?qjw&L8K`2kYz)_J~SsF@o%
z-D^(Y6i1XB-Mt?#eGXZgAb;&c!oXZ6WY^7aylfR7O_n9N(1M%ZOAkU<cg<=?E&unQ
zWI{%ik04dMUPxt{h0;8lVlm0Cd=aR<(Qqaui(_KoWJO2w0duwDZBe`-ifIQAC+!KI
zjQ>kj!vo>ly<V~+-2dXhKk)}l<-*A^O}Y}y<P*St<av2_GHe^#Y|yt~K!?8U-Loib
zA$ze2bUPP%Vi6{HkMGnusC1N}aX@$^E5HVsfhTL(z$6IQq^l4K?2|PreWG!D-DH1P
zM}g`YZ=8ARl<a43{v9~)aKhQ0td)uGjUz|Eu+EGU+d`6Cs~+P?F-WRhja3<exgac)
zR7Bjez_t21(Bfn`8N;)%&8OEm)i1^VrlSM1r6@-{w4`gNSrwM1yLSadgU5rG>L{pk
zCTz2U$>a=)R>F(kSi@yHT42bO6CYf@E3O8^5*hL5IAYotQS>uTv^@RGllp+kbBcJ0
z#dP4{y8m$?YA2I5&~IY>D)Nwf_iQo-2Gd~!$Hl8=i}X4?yiEEvs_E%l!g|(jt6qYd
z4bqMtLIoTJOkpq>DjQ@Rc&EHwj__6PQ6AF|0tPh?-WYBN=zVn(OV9JsNs?wP%g3%9
zobQO?U#Mot&>5TbApx&Elo?DJk^-gxi(c+iWIECm2zmdrCRvZ}#H@pR0lxdL(dS-I
z@?*e8XL4kw%>8AW5%^rfg^Tx+?xWT4g;cj_gtxT+aPYJ3RF9^(^5F{7Ue!207)&L2
ze{LR({lCRI{c)r_{3hSKAdV<S5G^V!5ZZgn*1aefIHFW{J3MM&94n873f5on{Le-g
z1gs6UsNsZKjw2q3aI!!d3w{^ic#z<&AweXnJO26(cnsDmhh$n`4N*=^_?Ul2QZyQ6
z1!n6(y1I6TQ^BK06N^C0QNY0nfj61PF8BcaI04}7D`=?3$RDRE`>h(of_@>B=e?W1
zZNY$?7}$20DC$p&z=Cc&swO%u36ejB7X!gY%GVRw$e<UDD2oejRt0nKuC*&y?+{}4
z$H_z+NTN-ie1Gs(ig&bh>WY!vG{0rC)wHK>n=MzF>-RO1oIm2~590ewF*Y)dOcREz
zkCX)M{~CM=XM%jyp4h+=IAMSG=<=@i7Sp>1zR<g-yf-s=nTyj7;34qgfeFE7tljQC
z*uAVK-UDET=3)fjK(OaKX{=!+#S)bG1Sq;PKtzstAGn9&<=KAOrPE>79Y<Jfz378U
zcre9g;Sb~%tD1w4A{TeG7|xQfiS?#>cA3YgPUpS@^E6wA)enQ)S8?T^52uY@DUw_}
zQ#zdA65ZnR0m7`f*z53|IUWf{su6hUS5F8V^S9NA7el3M{qQQuKtM*RYHDhn{%K{!
z$`gQr(t!EsRbgRaDQiU8pX(af*#Z?@^#{xm6&rKozdLxXPCW*xONSiW%&6g7ney$5
zQbRpW#a}>k;$dP&;vr9W)FQrt?M+|eJp<0&z5>t-v@{h25`A#^5C`{i)f>EvX*NKv
zM&{t?kExy&Upf$Z*cu7#|Gc-tY-vzfpF0=0#C35oVU7WjgaM9*Cc=wGX!*k&Xb9R#
zJq)dnJuu!E(04y2w2Wm(@0<{Sd?vFPIS>nrGd0N4BgPXOni^we8)J`S3N*hL&V-J=
z%^@pBINuyq3u4Io{;P%%A4`r<%e_5wyXC&5&7me9)W{XsT`aMtaT<;XW;mJ)$F=vj
zSDf5Z?ZB>LUCvhhfD_^enq$RG?_NhW&)PBxZvh+!7=?r4GZc;FX2Xr8TGo`Z{z>&p
z9M`oPyOfMf+D=-M74Wf5(=N2VVmo@mu2c%mxYE=3BHN9js`<^f(~9krTiuyk_;e-2
z!Y&Ku5O-TTsL9Vc8jm4tw1n7axy89p-reUfwcb=<wyvk<si>YhM$vwVtKVq)X8wRD
zwhdob0D;(kxVs<oX@Vfe$Fv<AUorF?IFu<s!WIWM;o=4G+`J~JJ-HRwlaDFn`ZiIj
z`uIIy>!@s*!osEvTq7nuu0o?Y$xBV2&o#0$<LBO1V0$>a)1RY34LCIFY93$83cf~5
zM)7GYH+Q4QzuNxV77E==qZ!Ov581qd+#^k4_@i-;g7N5g%*VZ6;i|DJ$<;4(xxVow
z_kY$%L|kCR+_mh2-O`6Kk_Fmz*1_Wdx>W)#)d@n(S6yHyjPR|(So@J0>)n7qj0gg8
z!3tU5AM{epxi==`6M)#!Sd%NcI+ItsX9@a~AqJmd!apq_4C7U<-08)JnJF2fhZ&mT
zay^J};)O`#B{XS4XWxtGI}Av#TMKECtg~S;h>58FvSh4w*muU~-njg4r~&Dvs71z0
z;e-!z#0lZ<f|ehmx16J`F^11DB*?IjQsc#t)Sp#Wr}act6ZfjTxTg7%tfrxQxgPrA
z{g6fa!($)M$uV9mgt8fem&zuYEXjf|Hq|rMdi|Kmn)}&wCr3@MzPD<k>4pkF{7h8E
ziPdMTsl67CLDWnGkGvq4FYrDcyKM1dGf|mI6z5GRsX+LD`4m{(-)yk#N4@{sDuHPc
zC^3o|LXPGE)?<t<jrig|v|RbnKX`{D7Zg5?9Q;gMK*Q8V;`k$7{kfz1%)SB#jV%Rk
zv@7S_EJq?OLy4I2g#&M#w+w$(aKxK;0@VU$_i@D|t|3Dqoz`1{&-%6G!!WC&9eM;b
zU%$#5$(vD5=EP`;r3*%W(fdq10oq9^f0%k8*N!GHu;WszK28SMPC<C}jH}NUWN-Ux
z(7akM)wM&8@tdP1yv;&J@^p_Mv;tSKd;1(oR_sJX;-~Ktr}c1MjYPBX!I81;)T?Q7
zANl%cWx0v+pn+SoY34KIk~6Gke;`|G@T7%(oN=mcR=2HWsk8{okJxT{F4^(>mb`Y~
zp=X5tCuen{PW>&wT-s{|*m0#dgt_0&`<#8&=yvwVf;*jkE6a9h-{+aHE$R+G7t@=6
zyK!U=XWg~Vzp^2TY<x0!=-&@<3W~SH%RLk*W%pqECz-sgf3w1vk8MNE6MgOKk8g^9
z!RG$H(dHfcBoR@1aullrX)UrDf5}Yg+DAi+CacVi(-tmg6fA`-UgbNgpmXF(IZ~=G
zMp;xnQsA2gkriqg-XV~|&=JZ%6`lgSLYCi<p#EN60#7KmZj|`i_7XTLzk2@POVDNa
zv-?Jicm=2kyRStInPm=6`iZJP%rb5W0$4Rm-OZotXzmZ@AHuALIzRgq@Px4fI}FRV
zE9RRHLvdzjYwyiB@rp))#|x8P+r0`|G)S!8sc{9WFLJCy@ReBey^)aK2Kb*n>J?X+
zwhTK=2?0up+uvNvjJB6X_UueV#9Ol`H}uA1sbE%<d+TCyUHla(H@~Qw_&R%Ie=msz
zGp2WSt?>odo{0ZC!$6pb<vIi<8zOD~twjWJ{3FJgof&MEN3WjTqW3uRiJp1aNQosq
z@#l}wWnlzII(RRuTka0dH|8)ZmLyW$J_+edNNF(x%BK923R5GV!A_F>yfELrO?8P+
zzq5e9A=`6AXZA=4^uEu}DiII${^u)h`Do*+>Jz$60}=kq8f~1Tg~1tf@WUMz*X+};
z9~BbBuup5~>c=@f3S5SWKhI}eLnr3W*Ro@+I*Gx{`&WH${Wt~ra%tabLL&P^nciuL
z{`o@lCXZS&%*Z|97v|RR_kho*XB~cPL(*gZPVT4mhsfS5tozp`z$)`O5BKYVal79~
z7)tAwd%5<57zM|q2f3+(4}#llVRc_t!<|h(rr!S+&WS(w-O8qXpp=2vG+ydj0W;}3
zaf0OJQxFXQ#b|e<GgS^SL|ys~A0;I+TgN!6i7##_3v;9S1*x0J=x)FQYQiTrOEKdG
z8CyY=c8v%ey<s!s=Q^eowS)8Bu|4hG6h~2BkGQ+3z%H4kq@{161%638Iy5yIZ|f;~
zR53aU)3i#9=q<v(T_xqcvHs7@mt{wCJeO<2&1X4+kIDQO$knt;-|y8dymIKl>vHqm
zp`+I)Yw~UbXK=MYYN1w<YVBxKgSEUDIKxY4<$NSuize+w<d8uyu?AnR5iyi%N5<>k
z7HeaLw=x~zuSTxz%+T?f{Sk?xzTU|g1Dj9zv#QsyN5s?RYkM35G@AUQb*I)U*IODo
zNKb5vVt0lq@yB46MHW$exu^9Li=H10W93&iF(?rVq>Hz+W;C(|tTY-k<A{&i%|2!#
zSBtp_#mj@Ug7T%;-U!F@=e#+P!*}=2W4q55Bly#0?P;BjWx`ilj3>=vbS)#p+$srO
zjSr+sBn#iZ$e>G;B4$2gS_(Nnag304=#_o*=jn6YKYLn<GVu<{71s!1rm@kGncJbq
zgLQL%i%%C~5oM#qsJ=i*u+7k@56+5Jx~)8#IFow!M94eE_!3vC^A_4a5t!aN?~92(
zZ=LrkT=-~k^4E^?ywWMoPU>$1nf+ARPBS#^<-juvMIU-B;;@)p8?Zq?ed_%Kqi`lx
zr|}gUFaEi}21#F9?cyji_7BhZ0GGL1zFW;S$G3@#%Qrn`SJB&{%muT#B=^E<L!nC$
z%jJA3M$ArXM1Yu1-!uk?K4^FF0;PPOxe`b%JLsITh5753n6P=#4+q`Tq1<R?+F>X&
z{glP&l+}w5>u`cQ8J~YrziaeywF#XA<HWn*c-TRRO`si=_~*}_2iR8IoXl7)A<eZ%
zY9+hp(ub7C85y23g{8;Wb{A~1wVtyoUPmeK6sme$-Qf01Mgqx!RZc})(S-jM@B<sY
zuK2Bsp|Zmx3__$zk82Wanja2-mlpvuH*;K`)s$$t`ci=PQ-bwU9z)<Nh8qm%TVRdM
z=19KkXg=np5WSKdk?o~<-t%yqEhT{{!An3(Zxikm`_$pRlzXqb)ze5htm@UE6lL2N
zMc*A2Y<@0iYE$|AvOX7H&iY-qcc5Ke#N>sTNY0PNqUdPBj|bEjgRb}!h6@?VMeFbr
zH&Wq)%cQ*?Q+k71XU*$r0u8D@Z#1x@?PkMI&H8k&y+3);aoTHnRG5yau*mn@I1^K}
z0a>bP!gRZ*zZ=^+rNQoEX^g#Bxo;T!Pla;NS2fk<Ye%fBPmv+oriw9k>EE_{ghTRs
zDy!K}9%DC-2h;U$cBg9?6im##O?TXoIT_`Y{gPYZKKTqcbKt;K6&#b;HdfcK)+>ol
z+bO^GB72;1%jLO>=TNTIG^)`D_;sR-SsV(uLxGJ&EQh>PkQxq=eS{ITTbX83Q2h3e
zpK+zYk10t1<<He>E~a-5O=?~%u7?b%V+KFWv(n~Sp4f<V4>&Ux(!XlXsF3%}V6q{6
zG32m7s%ILlEAV>4@K0m^MY1vvtz4y)K{sXvrf*?jzrCwb>xIJ(T3yYwEx^wNy4oC2
zcI|eFck4C5x|3V$ySmu_vw{oOiO8wXESu+A&t0#HM(BUzL<q6Hh!y-%J}jD2;}Ny2
zlautvQs|Hsoo87=q$boT+W8r0@-7fd(^x2Iu&I<acm<Ly;__*HHa4ft5-iFKuW`_~
zAfSTB19V>D+Z5lBh@hKTl~UQ?B;MHSF#7{!?a8W-G~>&;{MmpzHI<5T6_+Vz-tjYA
z(%&H~sBYI+tmSm8KB;4lJ__8cs2`;WyS=4Dvz&u2&K77%MLXVTD;Dd?-x9d+#`K06
z%0H{+(ebCTclnl#-621dWv}d15sQ|z?EOj}kBlAgtDP~8&r5!@)R~%bYKeVz<ka(<
z9sCC^f&(;fUI-f!e}-rzRHWpXJ!nu4RHIayg;Y<>6dCsP50FF7ImK+t$2d`ih0Y#F
z#8Q&jLklHwvKypD<Eqfso9o<*(O0xuPr7T3i%E<v=Ta9QXbpIs=t|OAczLWrThYuX
z%4$xbImCGS?bhCtBMi-v&km+uxhcJpP}ET%oUjmjA^9AgR#N;FbN*BF!S`+J=IiEc
zRMHRPp$&;K{A-Q_j)I1=91yEm6Ho#OjIu2p;HxlDkmfVJrFL?Ir!j6{y)AzQ#R$(d
zwCZi)0h`Md@NV*6US-sx#sF$6scQGg1_WfOfV~7gS?!m5QUQ`y1x%aikzGCsXUYJ<
z1)9{@NUd{5=1Djw@BgbY?r25`3F{hF<1lO4Q9{0A!-l+95Vc6FqN92i{eAQUFLr`F
z7Kyl0_qMGxUxVePz?AKBegt6kWde+tpz=R7Rv$AUPU>++Ou+e1jz4!|XO<RTHK?CI
zC0VJob68N!R?H&oL|w1=>o1mh@!6K4Ds8I8&xh}hp9y<$I4Gs<@oL11=`EL&-DUQ2
zdE~+;J7OA-`z6EEYcE=6hKh4lpNBDt8U1ZP5u%Fvow$QVzn^1I>AekbICI89ppIxu
zE=Fz2y|M4VJUfx%HySM~cwbt9l?PSSX?`F>wTq)ZL<?MQdy6P=3!y<r^*qk>PxS>K
zv`m`19Q2e=Hf`N?1;zfer$RwA|2>fZFubt0)x4F5R}Rq8M6N<yWAV|-qZLFZN0NQQ
z05Sb)TNE<^8Y5V>EI~$Qvr_SJC$6rz%2V2v<_X-hv{##0;0%a(^ayy)ZFzOizTCK%
z{GXl=+!O8m3t4fOzsp3DU@9a;j4mUv@vsZfaRTb1YNY1VO=?xPyaY{@Hd7icJHvCR
zy1N6n*D=T*4dDuv9tX(t+~;tsd~CFuJ1y(ao}XGzCW5x4uZszxT)dL;Znge+)1a9F
zdHf;m_g;G~1{DQ7EO-aoNLI1An#J%&7J%mwwjq4@6G3PKFyI8G#z~wq<0xW9tq{q8
z`NiSu8uX!xEp|#O5~J635<&+^d8fqRI<G?;5ITyBha`^1$%H@j8*Vy!;V2^IS~~Mt
z*U=SXfjyb*?04+$DSf-hG`-@yoR)iJgFI?T#!)<q^XuXF6wlOhNLn31Qc@BIgNn$=
zNIv&#gz^d2yB~c#`sqLG)b3Js;)U&I1?<)u?Nk5#N5P|p^PsLocAEk{mQgSF2z<m_
z#9}7;9m`9Z_63*aW;M19iMqRh#zsWKpve>!X0jl%`gO(RyT;Yy<Nw=gxu<rG?p(cJ
z$mzrRpL`<fHX{%XX_p_DnnlXZpH;=#uMK&TG^IWA7?&AEcNcvfk*a(R$s)n|YB#dj
zfDZK>O@dR|O*ag%8dK0XlOY4QCR&6_@({xqKI^I8co8$dX~W|s2v+N#hA8Q;>x8ls
z7wU&7)8(gRd%ku$tTx;P2lqdFz-=lyan&hfT}nU=L?LXU$Xjnt<PZ3x%uzrF_<x3*
zZ@_lr@DcpW0;~}KL}2EF3M6DjX3m=8{USA4{*9Gu5I$!@jN}_bR_wvg6JhtYlMG<H
zB3}~&m6YynK;%gv1yZ&&nIrNxQJa&*SN-3s4m?>+^S$d^k^YN&dR&SI^X3U#JJc0M
zI`}et9YY(Kg*{(mGa!%7A^l}9q{?>+g`6Dk(t<>0#x=zRmVQkTqb;f~BhnEkqQ!CD
z=kzdHr&X`Tv>-<oOTQ+*A796E&?)*523HH-kEIU*E2v}RONsk?@2!&(e+8~bpMor%
zmRr?2lqjXV4m^-~13SQ4oSt2z&vt<MY@iBxZpT+D(N~O)5!6y&P*8lHfPcCR?GYhv
ziOKz9-V5wnH7pBEVoFFx7!?)N8Jb;b;`&j<oHloCmJfORcLKQZW`Ng7uAz|n<6(%2
zVXM7BX7xU1LyhYdj_Z5PR<BuQT2f0pdU#c07Cr);Qn3~9h3};aWL7<IQare&je{8i
z-->&qbDkk`N@=B%tm(Hc9DVO>J2d?78a3k2Gndj3rzexe&Uu?FrWv-lj$ywGJ)cCA
zn7^~50p_mSUSJ|+Q)9$#W~2yJL_;^Kvmr1K2;c5=Z|xkAU6Wt6E<-JH_-manuCbZH
zg+T^b&NR`#yMO=L<7!5Yqa}Fk@Nal8eqlEzKJNM~=Lxy*=g`ZUBdls{Kn3r!Oq7DM
z#ZB=@0*+qc$OXd~B$<J3r!cq4RA!d4r5D)|BIuEa%QKh+om94?Kcu%yEsi;H^Sg3l
zPY#WwW)Og`$O^TqAend*Vt{xuv4=If6*#NIxDP?##dOhwp3R_&Y@M%}D5lvfH{Xeb
zZ5qB~#%2?LfZFBoe26ty(ylAX`Rbxu$+C0zRIar`wgAGaCtljvOkxBs`X)wo6snu}
zR%#l3zV1~{&6Z|3nML-<VAy*P_eG&1hp1bUGwTJsBj`ZF1SKlvneP+tYlojzEy+58
zeo!yHR=f7sLkyXwRG}K-mGx50Nnd?cW(+D&Y#@2SxhEm@#lh<MGRHeVd~W_25&xg(
z-pBoHdg3=5`lyZ$4>F&gc*nTE(}DMH9l5Y`GZuQ%q9;Ml*CQ_czbf3DKLRwk;M9bt
z7*arBGa|5&&l3PFffo~?<}zuW1?>qt(hpkzHUl_Q9*zLF1Shb@VINH*V^>=37m2+T
ze&SmCHIh*}`<xk#@V(2#-rL7+02TC15P~%%#{&R^24kowDq-<z<j!!JV1tih;@(8M
z!M~@rt6CRCo|!Wre=ci0+yPj_%|pBj=-XtgkUWiwai01YKhOMTq{T(2O1^DUkNh=~
zgy;t5=6;Xwfy{?b4U_<y_woGjas;wR=2N^fHe_x#j--#o56nbQ{6#l)<+CsVdn3j0
z7`%DhXleL{UnfbWez@*59kw9}9Xs}k*@^J)^{(#8Nj2b45NICPxHCJ4`2JE%0wS>}
z2;7+#gM*VZllY8LX8i}T?U=*7(>&p|hKWJaS{0ad)K}J>Zq@I|M%aOE+9SV*6EDx4
zz{UeWF&SV$8(Bc@i3LV+QZ@4yW5O?<%(W9ISfFKOWP)M^FhbIr3y@R?UpQtOlRt)O
z0n?8L^;MWh^+?Ua$l`$Bwa$U#$y|5R%qqe)>EqQM^@=aC3H{}3T*r*c*Dixz8Epqs
zwR%L$;ZrS9bFNSW$M1<gXDWuHWto{ZzV#Zm;Flm5IO2C0xh4{Pq1kQH^b4vn=BsZD
zxXSR5pQ=4{33{}n)Xl?C;{(+4aL^8|9yjW#E;tcs1+j@c{z|h6nT#fKgknoluIoj+
ztC3v=IGU7J;#_%_|8S1kq3;zssGHq{9OWuRtBtkeOH26D>p<+Jci@wEmo`M#j)Xhl
zK<8WFajN__()jCmm^3BIf?lpCXP;$eae=U7Rbb|?-V$=b@$j+@+rg%G!~0k5`s%}$
z{=1EhU3>9rYG6+X98jtN_F@m1G2p7=q&epP{{Po7cg+Kyt!jR=E8Wyl!&_(~Ft>qZ
zh~dHJ0C(EN(fUSG_p=Lui9gUjb0C$(*ZC>!<9w*pa@u5<za1crnW{GJWJTG847qjv
z)wMU$^6{6aqn}1%TPG4(>_#C)s0goAz@g-t2_KjwAK5TyakOFFZH$*40qUZ2K-vJ(
zcXpYI4M|NI>Jxqd2X`kYnC|1K1WCoKmlL=A0SDX?ursrD>NWWCaK2ugcX>>Fe!Jl<
zP4Z!;RJ(q`*>1K^%8|erO6PmTis*sK6CM&$wg>M&|F-Q(l2Ir;?z(R_#6MG-`?rXx
zf4^puPWipK*rH9Sx&DELr~}9I>WY>3XJxBaJN*6mQPssMiqzTkC%2voI~k-xt9_1^
z<Mx@K7!eHWXT}Ivy5rH1EK_a10$baws^aYaezbI$AhivNd(!pxcgg2Z*$;5$Y)3Rp
zZ4-49N%ijbR(%apH<+F2-$87F3dxv)CQoBhxEgDkYsg^wf14nct8sWavU`oHS;^=l
z;K1Maf6OjQbRyNmnAvq437Evu&`rTeoKMjGpI^~ke}9}|chS7~P;%BZS*onJNCIf{
znK5HQQMcRHHC91$HMkg|1+gSC3RR6^tb-x4ZaEl$nb(MXo}Q^VfV53qpWz~stvtlB
z>R>G-7qo*>h<cET!qqxgb2>?83(lNJnHu6DYrR=oWV&X(kP(69>iZ!#aVig&tlDzq
zm{YyAE7dJer|!fc5qt;fxR+wvo@2i)pB<gPBy4x&wbY`$Zgp_=cM+?mA@_D%tlVc=
zZTS_lh}nr6_T<B}Hz*!W8gE9KEm<#ezH*-*%6K<~{cF)=RDaC2!Kn8)DK&L+qvxN(
zAu46xC%w|{=Fr^F@98D|0cvS2L&_9*liZe4t!|3!<@SLmDHa)|AuolIwl`5>F7FMC
zsX8TI2e0+XXB+9Fvra{aLEpQJ)~C7GrDo(^z2<N!*8FsZc#o`5hT$8QQ+|M&uAY^7
zW2@;@*~y(}ExKNVCpz+Jm^>9roUNyn7Atca%e1f2d<wZfai@x*`UBn;-~T&tMC{J}
z?iGO*o;Wb!F21*~oAxH2c~OyKzDi@zl$t-(E&`rZG#N;WFZfL)qN=~y<^t^`i2(@5
z#GW%$Akh4_Rl(pIF#wi2eKn9!##|Go7FXqUZHSq~3hmlOnVqSbg&Y$!{Pjfgy6s&N
zj!%`>K`u#r<s^}nx$;57b$tKu`qMZyuijtC^!J>?8Lr>8=#g+Ny_W`CH)?73%c=)2
znj@3Rf*1$qJMH2zLM-lXN=C%RTJ?|FSwRG~jQcM)|Li$`A44U0g_lfKkGWirxnLPx
zLcj7+i?DcJCqza^%d_N-LlQB{-k4SJX3XEkJq|Kt*!JR8Yi7@w@ogP1;J)ckoGh&S
ziO`s}8$d*x*}e%2w>jFkd3YeLWB8ZoE#G4==l$L^mBeVNyIy6S$60C^2_ksMtB@!T
z#^qX=;5_B;N47$xS8ldGeY(_5-s*lFR;yv`qflF-b*FpUL)v{tYmQ;&ak0y%!_rLq
zb-#4w@yL^E{bZ6|vl!RV=)BiZldJc?pyw1KU+i1lqg=grUGT1eXOO3H4_fT+*C<M~
zNE`0XY9YiVKK)Y}yjA-<Up<kT9D((e)3~t(7;hjl#KK#op~@K}LC#HU6d<--tnMbN
zkchDmFz1Q|!>+?%D%E!|`Q!LbPV3gO(5*l!H%6rbZqvmK24`x+W4wuG<DC=_B+Kqs
zeENF4496RnPiix#rOwWrx+Q<Z1P~+w3kCE?Du!}n`zyp0&;p!eF8|R19Du#Y<t%l#
zI(}OyBi(89hz&{%qj?|*o94$xS^;l|t2xtCD6R;=*mAMH|5?1xITe4gcFNAq=9SGi
z{K`7ePd?sqQ4eK|Q)kySobdZ_rKWX{hGxM)PyJ0K?N!%#bYJ>wmDN|rbXvcAXP^z|
zIa&iLxVGe2piAeRUketVk(<_0lcY?~QDzH^XSlJ7u$h#QmaeL%b<CYnpqrlyjeRir
zr222p$r9h=F9P9y^bycg{wZJcRyz5up_6<kH!KmkFUU&OGD^_AUDa;n*6NR%`zF?o
z11vKwB%!OSjOQ>_IWwjh?Dz>wPI;b{((@wVco=xMH2&FBA0oUpv)l;8u*7pTB|b#E
zw-^{oB<!OW_Q|$*nO7O)KiJN7*f&ht(+QHD0~?iceXHz&&kQsyHLP)dkzsSeG(O3b
z0P}#U@Ev!s{T1jI^J(!51^s>t!KnTuIPA4xL(P9uxmx0p7<lJt3tnDhf@!gkG1Ymj
z-RxdW-qEW?x}dv^7FkJ{*ozj(AqzSGT(EZ9`EP`H0#vxtk28B@+Ej9!^R`f}vAVG#
z6##?;+<4{y=LrWbEFsTpexy7@6(Uf($^bZ$hRw?Gj{g7kJ3OHR2$P;=>nxdrMqXWS
zA>|@}HPsk0ibFCI{q3)Ea}vzp52ta~Uyf|&<hzsy(cVg%Equ~}zPg}s<ov13rIGi>
z%nVpxAndW5dGM#VCH>XecMqE6zM@21R90=(Py4-mS~H}3;{^fXa?7?rwRI`pv@z>V
z-5cM{GNH)P%QD{`c&y_N4jPV0(}zwPG}8{*jS7^z40aZrfHYq_{6%wnNCBbL$g?rc
z@37f4{vIQ7qLJj8wP{`XbKhF}i6z*fEt2!Y4c^u5BXQ55i9v5~B%8|VfF>PK^8vU|
zFBJ&X3rkUX#ZHC*$rY3{9NZWdU4|JrU5<L_N`Y5#tBWKSvg>RdjSw+po5N{I&Hu{1
z&tk`0{@R3m<0MigHhdEOqkSSdwq#G(3y2+%IAx3d{W%bS6`&)taLJoG^n2i@Zvt95
zHvhg^RF65B07k$kDH)iFYlg7b_eU5Qf+W7F9V2fbZa-X}PWNq+MMPTMRL2N2JBLbK
zm-r$5EZoPm4wdw3ejuT&^$)Q9c9VZy$Z1CbJaMOZV0Z$q#+&~*6*QcUhEFoC3{L?m
zOZPB%lO_I&tou-9q|LAQF+^>jhP)MB&h(m^S^?y3!PwT(ComhqLTqc{+$hhL)qnBZ
zUW)7AhU~z#wEWwUFM%;ziih8wEi%7Za3$*2;dvcfZ9c8S8ms0+Ut71vD&bKE!umwz
ztu&qR(0MZkCSyx+_?2Sdq1W8-?1pFUhnY8EWdjaX|IYZCT?r`WxPLxr@zwI(OocgS
zF7(}_%!{Kz_J+qE%q069j||(Ea|rtKz{Y0gwDE9_wo`_6iF=N1S%>()51%IJ?^Hb(
zi?5E+us&9D>xg!GO{7tOOpC(biMS_Uuh()STakV-2dRJd#A-ZWor>)ue}azi*-PLh
z@X}QtP{WH2g!D!4^cmN{$*ohVKUBj8?T7?hk!VyY0@m!%UD&rJZ7cAI62zIziKq<@
z7zOyU<JE{&2Q{HxFj%S47z4Nxq>o+Pl+tii%44J<wB_j;3oUQBSl=gWDB9-{R?bWw
zL;$QLEon!$a$P$~lK%I6#%lH4UC28sJtSv<J&^d>Dc<UX1Q(nQD?=(Q&3F4U1wh+*
zNNVX)+;NON=r1VyM}nsGE_M9~_+qINgtXB1kz@QT;uOqgO7Rd)_Xy+efuw5|(%J`F
zb;z2-_db^ua#T8HM28V3V|pQ}DF^pSX*cp@$J9*D$;D=$0SQW%62t5yo|&P72ieh8
zf>s@7@;{2%?oJOc-Ave5l6CwlmCslG<B?Z;q%S*OOUQDLY)ppD1w=iS)zu5>J^S2O
z?Ho}jzR1}By|sbXU|v7^`+KuqIeRNEQt01pF(=p4UiNpa1){$j&lGOhnzc(aM+!ro
zxK`y$-XlWmfkfl?X*<L}dj{vyD%LbSbt}g4f`6NO!2+1)S<^2FkU?jr^Uf0w(-*Rj
zFqX=f7|;=q*4US)4{hLXNw#Ij-Yb5-CLdqQ#YwR2B8k6hrz4VU^(^!Se-v7$w0ctz
z{_5`yQ!up24uN<n|EqjuM0dZ(d6=%`M7r~b=5fzkQg@2Ylwh06WQ-ULot9fsrqx{7
zdSU@KClN|t^?gp#U&VgekYZzy|Kvz%CgCA@*Md4f&ozDuS!qJ<H!x>sXtElX%GDb4
zEgg#UC1wg64WuYA0LL(2p%jFc9$($PfxEcZU9lDsyy>O^1~yBO%G~8nRsadWhEC)1
z$+@d|<;~8%l^Uy=6ZwdnGt3czyq90obqT4`%A{g`NjG|ICbpvYo>um>XQru#l%m@y
z$w3<N{L#@jB8WV?X-tiV)uk}KgAZ3#t!k6Hi`<Vo`4<ioV2G$sa&os~bZ&EzM<)<L
zdH<f48h$;E(z>mL$e#Qh!aZhM@CNZDJ>8>6_S2I)aWdG@Rh_vq$rwa+>UlT6`5QZ}
z&rH#n0?75@GcXhsmEWyh-A|HOgj0Za(Ok2t(e^hy++p9OQSa^LNAW}c@OhRMy=Vkn
zvT76z<{=|%8B0`c8xA?Ycr+wWnVdbVP)`ocQ)65ESx8_iOnyrNGGA}&e51XDeabV_
z4S^rM_CGgtn3clI$E%E~QU}LaOHur+vFMDeMmc?lTl+gF1fd2ZT6rjfD?PMyO6rp%
zARd8^ZY5w+7Mv?AeviR5O(o_Pbe&%L|1eLf3r}kbz|{CvvHi(TAy2?D#R3Su$v&}e
zpk&tdo748{de=tHT-@<E%-nGZ)u-ML;w+ubKHur_;SBBC5!;!pd{5#7pI__6n>h={
z!5!8ESZJ^%E0_kY&VPk#{x12lWRqm}0rLI~AZS$QSiA&jxBjhvYt`gdc9@2lvO@ZN
zHr`O=Z@vx>W*ad2@@K3fKq?QZo79iR^m-^&i{`RXR5UbiepNXSuJCCm{Zksq&lHTd
zF_&3c2f{!9fQSitD!+)_+9t9Iv{3HVhdX)+{WP1&p+OvQ_gt?MAkW`Qfll};0Sm^C
z;L;Q{tR0)ap@TTJDEbHhHYM{kcbk%)h2VH^!iP0e9*g<!1njN)fakD=@m@?!3f47V
zDP96oGo3K>?=0KL84I%amt}pXBjrE88dUA*@?&oylm6_n#0hU|hVn#SaRb|u4oW1~
zSk#c9v9h<di+gww<R&r|m5R1yW2}j?`pJfhq^xCp>-gEH(vjOmGs$aW=M}Ez8(E@F
zze2;G0^ww>(bFwsf6JU<>Q_83AtneMX7U+1Qm8twgZQq%*H@b0pDKO#1^t5f;kI!p
z+lSWiwp>ADK2}B;dZuC))ESvI@i6=}YYWEOO3YqPh-e28qS{@|<vWBQ7`FnN5}2t1
z3}1y+9q4$+9rTq3s%m+80YIi|o7+yT)k(}v+xcf;r7K04hug_{rYOR=+>64-i`^+c
zq&kii_;NOy|20WcGU92icD;C7qv{FFU7{zF=wRtsJ>%=o#=+}vJ~nRBm+o|zJk*<3
z?3KpHO!qi6Xe#HC?iBRi%zpCQGiF0I!XMJ_*$lC%N_`laHfOBT5CGt`JLBsKqa0+|
zLnhc^2p#Zti@kAoIX4n5+)sV=z32J?7G=3&KZxmHon-;-1ELK_X=>U4G0`}`3KGZF
zww{&MhIFwXip-p=_;_1J{486#!8HOrb$Y-QU$0RPhn|E2)B)&g{6{XfsQe26l`jAp
z@JmaTWG$l>&<qK>^7$`gFt+B|P<Ij%CI9R(o8MNFZ0Yltl9G}J(<|Y66r<fK*P`^B
z!+pMJ#oaL`>Sq7%0FN$OZ%}Ibqhrrdf`S!4r0`;Q8o++6-RmSi)Q4azUb|P^0kPOp
zk<{CPUH2DqrGc4P3=~}sx}N4!+i{nSrNM6|i+e|Z^`ue=Ji5EBrCgASL;!+Xooaqh
ziMOC>Q2%xv3`?`JKx#^66HGv*{D6oo=<4)>vy0KegQ8hZ1?R%_HWMFC{MjSj#Sa*=
zrY&2*{ZO8&mC5vLda-EVAIT9;PXTi4^<E&4UDsrm3q{~3+?eQ_KXHW+oyGDbl(tRL
z@G9a3Swe0~fIt=TdDJ^sCWpLT;x1MthC*1$egaD<Ss;BK`BJ<xCjp;<*X<;kMpw8*
zS5meh<J?4mJ#$~}5^~}N<lcIHc4mwG8GSdM+<Ujfx?TOzgx{*^74Qv80BsgayhsGJ
zyjoOG7T#$&aeJ>ZyLgkr3))F)tJsZu5u~3FK?HsRlIp;Em-ln+nDcWtR5l$K3?wnQ
zJ1@zxuvT`b!SPpx5l+R7K#bJhQ~HNIOs!(oX+yn1WaHl2b;d|uGr~_T7YvtcpaNJs
zqo;sm(NsL*s0@M!zy}Dig{f$X+?&8Io}+CkDStzpYZi7WAHW~c_gVo+yxGvWc{h4S
zZ|0e`=hfc0%STj_#mpujy=q>tS*tbk6+6Qd$>N-8Pyz8zi92PaUoIBIv{^7N-PesO
zkPChp_v;Q9*T_O?1p|&&A#lQ2n$!#u?$CTx(JvK=?H{hMRc1&n8OcxSO6;JT`EmKy
z_`vFINET=THX|FALn<s@kR#1$vVm6tP9+{O_k4~~0mLPjs1Wj-{i4FLIt;NporE7b
zJ)wM&6U{U_+R{o}B%(_MIY_?mDZ>H1W_b)*8Ne6ct5{KGu~-~P>~o?B?2cQ{|HM_D
zfcR7*pa0PHm!uOxZwhnBb^SQchWE1TI2fiTW&o3c_rE)CD-4eh|7Crs-MPg|j+i%(
zz>C-qKgPBF_FJ9lVoeAq;)9kUdB^EjOc+>u$9z?d1h=WeD&w35y4E=ae4og}Pl>7;
z5nT#S%+1RVems7e+Rr;e-wU}Mr)*g`4v}9y*?>&1*!!WdLp`d(bR+Qj4`XQ(0&O-v
zhKG)!jhp*Pdtn|Zf*(|Ht;dc-z+0R!-bI#}@w4M~Hf8H|<MS<ns2)gtT`X#){Ya)M
z+!$THng|m#fYbbQiGs=y65DnY1~`EsZ-z1F<PQFW7*BjdBjSt~V2f%}p23Z9+xO2>
zzoGBl>CnM6#pSkQ*`~qPM3p>CR?Y}G`K$#aVjZKK=2OkPmt4C0iOFV~Se2a{!PZ{C
zI;}g9%_CKZrD_3EzEMPiTTE;Hv#0Ih{aK#CT0&~qy4YQgF-PpccJZ;^guxEzFL?07
zcPy4sh{IL0Zonach3in0m%%8ubjpa@0hJJu(Hfgc7RV7|og0jky3ar-C%7q*BeNoo
zfB~0*6vkZxsEU1uTgtOyn_}@2Q_vnI7+M?zPMmqKVeL0!UVL{rX9QX$YawCniFzev
zX{Mk9S_578jQ{L8{LA~>7|(pexHkA=cMB{yAAtvZfG6apV2KV`9G-9`6qN(>V_GoT
zln4khhbloI2AcSuOU3({7JZS>6D~M`*;<s(spC3u^9Un^SBeZ58G)BTyV9IKxPBy!
zPsR{_qeWYH4s5H}k8wcLvTxk=+VcZtWZG9Rsb2IYcv4F55oYawV^pvucfea>$czh8
z)7!Gs8qG3XNP2w!17t+Bcv%m_lC|vs|Dq!TvZHBAHtWh32HCCVVImnnv{v2SeD}Ni
zwaJ>$RhuB092`aJ3#ivJFhqDG!cK^A`5gH8W!=iX<*<vbsh!FS0iSHiK(MA^-aEhm
zxxuIARm{=8UnTVgcotmRJyF@PI0&@k<^Zpv51irXSg@yO$$`Vdi)DOhqk`^p#oNYy
zH&iD$6USC}(o&kok<UFe77FQ|q$(4yE5L8}vb?T><p3Y%xVq^q$$eY$_J6w!md=DO
zdf&?m#|^>n02PN`dZj)Dk=a5Ivs;>;onmhngME2}1T~tcYE^O~P77b1sq5`fOze&q
zl-Mu_SsV4T{l6tcK|BHj(+`0eU?ae;iv{#xz-hAiVr}wyYEAs)g{zd?5M))wdn4y@
zZB_ynge7&!tb6xad!BO((dDImx?hAcHR|fMIlF7@kWX;iku_#?^><$7$?ixVd5L0v
zocd}C4hQdy74zz47p1F1xF8}j*$M-?(^)B*chQ;5p#r^yj4;_qbng$JpMlw5o4!H*
z$|2`uE;;2o6`MoDsPu%PDQbi_C>?zi|K21Ws~cQ#`UNvqUm*$HZCkax${Swe72=V-
z+oB%cRnt7jNsW;Crmf%`!nHE3LBZ?7CGL@$#mHt%8p&o*+l9y?)3KI4H<e^BctTGO
z7ToLF^E#OJS^3guKh6+%8oJid{`7R3RKMHP@!}5QaVsT31CZz-?vM>$B)j#*f`G$-
zx7pRIbsfUjXfbHojhaqe+g1zuf<y%NVg&NXg|5*quvbg`*;9ReLwSAJ76V;%O3y+O
zL3Y6ia8Q0PjC651TRr(WLw$s04|;whtREl|3bHo>MA49H;z7hK_%Q|dY9fKLzfqhK
z-?s|soyX|j?wsJm*(PL!YqhOm9*?Z?$Ut+10Q)9epA!*pjw#PfnUI$r&uI7dKLziy
zwww5hEK*wP27k2^(GVmP_^avBlADrP*v1hR(mx>S%5S86R~Hpe@u=d8+`YT0eC`mH
z=H(Omq}<VyD)Yf&E~k$l@MA5Mx^6yNFKbhRt)!9RFPy7BgW2BrxEsaFx&F!s8~C%1
zcSnnCf{YpXPg9ec-QGS$0F1LK1?{~8Pw!nSQeO!=^E0sZ><o40b%H7<TB!f1PU2wd
zh4WXd{i*5H^kwsM_0)lK*MWlC_WEF8-+l#1Kj6BAmnKF?SYtLd6kc+$gEW$dFohy6
zg5WDb50G`5Z}r9CzLGJu7+#f;z~?pGP?Ukn<J&IosUVJnTu~|Pv8?DL-^J5efs+kY
zt!?H^BClTNv_H4MVaM!C;)v2T_tK~`IEB5CT;CDAWXycl{;E}@pG}-2%&*IQ=uC@k
zR3h+IsCRKQAFGCMOSWMF{v`pt(uqhm!>O?F+iogV=F0BH8XQ6YWcB$z+?3Rl;g2b%
z2|WwHI=>++TEU9anPZzDiQdkih0{$OL}J1MOJXe9Vzn~Qv2t3vV}dauB$yXJ9l?fE
zT={H(JH{j1f*8a~dkXBb!wjKksr}U8mX8sNcC=3h>CC%Ly6Vu>7M?_b8>lE)u@Ymi
z(U$Q7!O&g$Vb=JcJx6^Gw=iK(F7*!Uaz4Jc7Y+G0+3C4rH7j8MxM@B)#ipL?t4ET4
zU@9tI2|pT^K~!{D?DwOJv^-wIPpDU4*>C0dSX4A)ZhRHhlw}v0Vh}2x(o`3setk#x
zzx;*bZm_cH&|40skAht9e={09ZT{Q2?PMut5S>nK`a6*(&QTIbUyL3KU-gO01%XD<
z5;++}kZD0>##_ys<C-@r&T?^;j5J)D)sjJWxWT^z>$u6P2Hh?BRY{OShlY9orAfFn
z!iSZRk&##xt1=IIG|`9fTLE}h&;#^0H}^hTdi?!y8aKWmS<LzMuXUl(aWb&jweN^+
zQKxz4iU*r}3(opTY*?<0{)i=9Q-EN%F!HWfWSyfAKf^-&&Ln*{T%#nHg6$gPXh^v>
zd3qLo6ZzV~%-;0MaQBAl50q0PqjLdz-0aqPC+LbaEby5E+ODLD0t%!Qd*ax#?7=}=
zo~}{<i;Tv6<&h9bng8Ljf@9~EhPSEB>Pg?thgVDGvOW%-`p_NrwEG?F-_KSK0?^yl
z!GVts6c7%_9PZ#)aLyfA8Mx>vQnWv77OE8*g&Un~nP=UX3>nPeYAdzI+?EqvEr!Q_
z`71?R+;1-J$=y9A+C!^8JZ2^TPAspU*T!?8vq8+&2^V@?#s2%}#+$2s>a@Bv+lY<?
zJWO9hz^d%Yi<!cgc-7aY3(7X2xx4{y?nj#q$B=x{BL9re3WuP8_~Uky(f2|?E-kIA
zm}lC|VtT)!{h|duK&?RntBHuc!Pav$t984d+@J_-HKp*~x(;N>{V#t!MiB8!v;H7q
z>`iI@bGpo(oS_cW2#kcQWe&;i|I8j_r%z&y%G|ULPIR5O?=Wg-=yf?&C^8>d%G9WN
zv+Gj5+p+*^(Wn%k{lY}!&z|HhszaVpukumLpH>*DDWjA~;+B*Vt^-jHT#BHKutD>+
z!E1ob4`>I|_N^pRIz+d0@Pft~;va-0w9SPwGU*dopUkOTFpDez8=KxZXyJ+BV~vIZ
zBtV9vx2Z^P>0_V=KblwO`pTaQOAmRmA5!GSZ-N+UduMjP#1co=N#c81+EiZ5aQFHw
zuU>99t=b-~Jdc&GY10yEc_H*AJNr=elIgrwUP-Jw*-0v}aOIIv-=lcJ4>E>|qRj<m
zw$|qs{Hb9j|Gw(iInw7QfAU2QHHsBhrT%GbJA1woWM7N_3hs*(A=VxW<(`~z*2^W7
z1w2%G^6|oYYrfuy<M%JU;g-jSL~I`_(SlN@LgYS*4zxaGJheC~*#40c1~LcQ{W%~m
z=LWA_0cdwDxyS(h)tIjQVJ@&R0HOmM4$iRfnt1pxHYkjWKM7J|pt~k&?6I>n#)bY|
zeA7^g(?l{TqPz`IaQI+G1SGk(7bAgB35ebB&IYXM%AhTyiB|Jj$<^Vf2MNH&1m{-?
znk!ac73)~XR8|iVc*WE>9c_LurFi+jk*1aWU?^3{S5BM43Dk`gPH4n`>%Vv}85Fq<
zh2~h1{{d-^F!M3#FM0qT?SlNLLrlF{DKl)QkiizjQ;knRB@86wbXJ3W*RSfw?>Hgi
z(hYok2aC2MpkE*KiGLpyy;&r6_`~-hI|Pm4hj*DvNPTdAoE*}0R76OpBS5#=^4$2W
zT)OAm^M`&xzKyC(f!*(VOfUMFBm1b|p%3HUnddyJrH}WUKAhjre5Zc(G0ykL`%S5l
zGBJ3^3u<-n(+j6b1qP>0-;kf2%;bb6%I*NY^4{I}hvYdQYKS&V!L@qtRI~on;d}=4
z)p+!9t#VhUVzgGwi9_Pk=N>4u{@PibU9HUxeP@k8+H?bL2(cYEr5m`~d*wu>{JJ)o
zGu+Q~P2MtJ&AbpJ5A<1SRXG`L7YcS8kAQNNT+pQs(B0^gPYp9$vMjc--v-X6ldQ~E
z=ECpj#u-fjJB8T*#fTA^GXyU(beB9u#rYw5d+V(u5HTl6m_xro>mqyB=cmM?*Uibi
z{f{jUa#15@+QKpvOV}}|<pjB7)*wl0zrgFe4pp9lohi1<mMVEx>*f}gBY>+?m@Y)+
zy8wcAe5Z7Eb?pA^NzA`Q`skx-XQxJlH&LkClfMmHn^N0P2fw#@F2IwQtNEyow8Gm*
z<q@rG@+GB2L0rY^4aL;fQz7rIE|ooLx%aD>#8oVTbB6FY)G+xk8g3^2kN3mEawd;<
zaH`$lSVutLq0vmfC)}Ld36lc_M|YAt_kU#07qUXq{)!iiT1ou&+jz0+UZfST*xr$(
zdjE4?tPeB_e+jJkfOkdfFUdzw7F*gIv9kN;O45s<|Av2QzPS0rLIBM32A+NTn7vsQ
z)YZ3(3WA;hg$lohbt+X!g1Sr@Q>Qco!uQbnN$Gp{E<eXcl629-YZ6~L-Xc3Bp6SYt
zUzqYsS)YS%wv`$<G0#}6KyGZ8&(DOBkU&@FZ5c3Hbxb&+3DC4G*O^YYxTd0S0MHCt
zeXhepA(?d-y>(aYPJZM1%phZ#hKiC=cL}K@5)m#=S^t|Pn4p13HG?<unxmBU*z8O!
z#Q9Ezls+q3q+KIj5Njb!$v}&=OEcE|NUF!wUkbRxHebg|0Q~9yO&0vx{Rj$j?-j9^
zkI}Yx*%!~x+5Cd`yiBVKv^g~wCr~Ima@l2){;Ix$ub<cA%&C&HY~DIdSg)?X)T~(V
z=AU%SVEUtSyO3^OPA+aW4CgEL97}$R^SkUh#7=G`-QYK+pDS{&y8YbEyt~?g_Rm6=
z`}{tc@iVU6>ik(qulqZV?Zj&3^-MPAJg7Bg<Y(qIBu)EX1x*<|kEz^*mKv9YYq1B7
zUX?7lTf11yZ631Id^!O)udA1?^z}j7n1pO*UEa|FkI3Rq(8uieRRVTb2Pu@DJ<4`^
z=Q92aAQTt|37Ugjvdu^0bk_s$0eFlc9XCF81*A&cK!OC6HHn%0F)bErVka<Z3-td<
z#&le$=Bo<d*{KPhn9U}NPl9{Fxe*wqC!TfV<ACN$B1VVNlU#ocNViB7BO*v@i$>_;
zs>};;ShgK#&v+|%O4XqYBlOx~EZ_Uy0ugDGWG^C*B3sgV{$@XD!K^dDlwef7bHewa
zn^-FmhD?UVDN@~ME!<VBK#jz;Aeb_h(|vnmvm=!+Df_|H>gN+P3hYUxmtI0+#2+h%
zU`;C?_{CP>&B#_3Fg>4~+#1c-1_#K8&GQ(6e1T=x;mt@*$datiEy>72sSUgB!VBjw
zeH+;)KNIjB>N^*Js~?0On>9OyAy)^y6OWyH^j*aD#+e7HKZd%$><>T6ZS=D!eC61&
z=BfPY3!`gNdHw6zQQr}V6a6k$F7?Hr`M^28ZaL>^ymdG@HK=nivrk9nihD`?{0z<Z
zQQgF*;Q#*YDczP`O3%DHmZ9~&PR(iAI>S@}<VZlk*&uzvsRLpaOZ<Q-&t0o-G`27#
z7Q<78+BNbK65LpU?_@^WeuE>+aeP|ytGJ*mpkCrfJ8w(RJiHG7zEy!_b(sw)t{G#a
z&_kS<q3llDmxSREinBjPB2KJIk<0?g!h7wB)LoLo?unUFuWUwxe3|DsFYIJ&2H@e#
z66=~x9hepq&|9wEt+GI=6;W%hU8~3w2d<GGK!F7b6TC|TCKO$uiC(N!*sZ|LjFaQe
z4;p7Ec7mKm)25)b66KlCYWE|2$IH%t?9($-)R8}#)JC(g@Pb}+pD`M;zcmWIpXZbA
zL%d3EwW0e4r*OjyhD@>j$@a2j3mKPw)FlD`m&9uC$vT*2DzacW8G0$Z&C6_JU`KVQ
z@<7)}Vy1pi5+d}hCBpsj$TfIB=ET|jOJa%JsnCy#dLn7jJ1Nl1iY3Cr)#B9pP5F@!
z#ht{)+s3vjo8GeP;eo^eO=d%texwf@=@<p^>G0<spL2+^f`}qKTZ+MWJJNM#BFciy
z@#ZJUM8jKffin#h&eYidW9iJpp<e(0-zkZZJzI^8osfO!U}o%OiDWDLzJ)B|WZxP4
zQe<q&mL*cQv9B>mCS+*Ts7aINjD61U_WAy<%U>=omznq6_iK4RAJ63CQm@63Z+YQe
z!W9^Zh>9gSwU4zJ7L%oAN1S_97bV)v0w(Uuv1&79vomZ0{!l1SO@l52&2`XOD>D_T
zHS>sc7Qqb{4+D)oNCZY(-XKd7nkHJ+&N|vS&Xe0<e0lHynHeYY>q}zE<o$v`UYDF0
z?pGD^2~WQBP-ZlpIFjrMp{O^?^HZU2U~zk`x6Wzf!l#!corsdBrqZ)}vg&;qgVR@g
z2K&m{&5;BO@8qk>?OH%Z_&bh}gHPjai*sE`oxygc`!oim*+_=z-g!OEU0ULyY)@MQ
z{<v`aJrU=;z~EnpSMm6m)IM@`m3tdb5|u6~{wx6Kyq%fBi9P3XDGBoWdH?vE?D6&L
zEeuG!xa>^Vf05|f9w~4&H*p@!CeEHM)b}&v7ck-@Y~c}-9cwv4fV6n_5{XV_Q5ot?
zb}G?yrV<jGJdQaxdd=s}UR2QhzJ<?#-xEh*?~VfbEG|9Bxj--Yt2$3=ox=c(l*7b!
zzy*$)Z~z%>c?~ja_P{|Qr0X?GvT=)-^UCM|8v9}}7$CLfY>s7Z#AUq_5uZTkCNFwu
zl3msnCG@Ox*P87wG8MX0pssDkZyLmTO&m%eA}_12kNeNADMp3U84<Z^%IS<%ud|An
zh#=BlGvnG`{&(*8wa20dbEKd^s_xa(WdEX6snIjOz;bOMGTbfv{nrQ#Bs1KA)GUUJ
z<@IY%pPB)4XaI2b2B9l^2!znYl;+&S&)#`-f+!8vvo!WfuvyqT`w^Bzf%p~-UcY3Y
zCumM@(ZE!T+*qbJfZ_kmIaU-o&@ppbcob97Krx%VzEN$1Aw+>mFW{kv)w&b)vBdXZ
z2F((k8Ik!W+!Dli1<>9ha8)2jJz6B-!Q_SK3G<-rLqyA&dWXKxO}qhImq0A5eF2h#
z%QT?=caG0v-*mgp`np$gB&Uv8t(V)$Lk9?g@1b#9h=QD=$f5EzgoKT)u;L-VHTJbc
z#La%2;YT!`1`62Ew11jZFtLD1@{kJpcIcs0#9|I`8SovuHlpzwTjqYoD+En1twft0
zFP*ww84(>`0fj`7+%i0<3{X>~XrR|OcMF9t$S#~kPn?1!0sFR*Gg|9*Wf@CEDoC0~
zs5y(*0CdGDT_*5Bgo0`UyjcAl)v-qDSrefBFfas^_`(9U^@1bFO!nsj73dDj*sHQv
zD%OCC3swk%z@qZMqsz!qY-MF-W3-1jklAdLESB%y$fgy*QqI`uw}JeAXk~w2mR}@y
zY;*ZiU3nJAC4N}d#Q1^$({QMUuBW77sbe|R;O{pjNGm&d%#&Z#Ye4Gi42ccg(;gZa
z@v0;+#LP6h4hj);9OV$Y?B*yr5Dpm+j!9ss0J2&~AS^G8lu35aNAvJ#<WFk>IAeD4
z(AfkyH~t>@pnHQy^`7%n0pLCW!AWz@MRIcTORd#65jYbI1#lj4+#s=r9YaAV0VYhW
zB}X^Z%OR=o%oj9)w6LI?gDWoxUUU>_r%kYo=FQbnlY|#oz&U`-lZQ4r`9Avg#OM8k
zPcLG4N7tfH<m&43lZ;xfgt+0$`)==>@eVM?s{4FX&oFw0Z12j_F<o9o3FY0qDPS+x
zL-4^ry|z|vs>Pu5I3?XNVSZseNR!Wo{BSh?RV3IMB>i?K7@yQ16d$ybzUJI?pHF!O
z$QJC5sRBrRKP1N{s(4WxO%S8Nh>`Nxe0-%464ZcUHHSm&2C(~M;3OP70dG!W!F!;K
zth@n(e>$_{u!g&WPcN`G0U}EwaNM+Ydxk({x78%&R>}N_mkOLuXIHg7=gAP(FL`SS
zSFRy(xL?!UoQMIq$j&}Rw3U~h|AF?fey&pE1S!I?!lazb@ZB6CQDn=A<2r{94T|;|
zD*9TlT*YL;W=XhvtO5exKNpA3#$JjSBxF{vP=L6LCjZ5XpE6%o@;%6d-t<Ko?L5iY
zn+6SO_(gFH90=k#M9^<7*kI->qn(ndf9cVaLM)U!vahjhtTl}VDvO3zGSX<F8p~L%
zxcU0SU#Cp}oy(>=@*nAOR7bLfcV<?2MAPl;*vR}<K}dciwvVA<Wd$%|ko2o&`Qv|2
ztLbEuo7=&czSs;LaK(s5LsJ@PcEjVKiIfTGLaC&5$t_T;fuXIrZo#CCPzdi=Po>J0
z>O|`FsJ4E#S|!;E`>|t0t2zAnQU9Ed>D;SqEX~>OZ|E`M<T9a8L5B}d0g+Hz0Qe=i
z{qb7G$bP*P(##oE5Wo<Uy=+Ik-rN=n5l?&WNb5i!QLtV;TY+c!pOa)JOAde>h2m+e
zS@hAT8(;gvF{#e9!PL6{&W#M@#ojsicv$gIH3$(9a(WO3xd0mATW=+ZjqU1!lY}{h
zCm`y8L@I=?^h74u-2+n1|AqUA+qyd<QD855Ch#u=?*Im3AP75C^!zNj>33)0MnA*j
z_V$4+aT7wtg`fJ52zM*)KEv+IJc!cdpPBwA3mwBiRZ~w_)s~E@n8Rvm=egVfmX41?
zV8sjPB(d);S>M6(I_Xpw7<(Q8=*}8MlXtC@hu9Ca%LKEx5ZAUBu(gWK*2A=*QmK@`
z>pLJ~w%DQG;`QQbwu-;P*l>o;(yz*LLFW#Gxtg4X0<+v2o=3HJJUIoSM_FGG6N!xo
zkue2@t9jnH?n}(4OsqxS9!pL}oS(Ys$-8@?T;;iiHF)bl0J$9SXX2*sFFTK|RDdb>
z&JX+mX5pCbtwuMy<Io?s@VQ{XR)Runhn7GNo1I-{7Ep}GtKUe#6!!BI$n}Zj3b!~c
zNFF!CZo2)bBAk)jjHXE-;CeFu|F-;<86a0h+=L%=k|1K{F-i?$%2lj~Z7{s<v!NTq
ztV<S~%sUh}5D!&!sA~1h_2N<FUm?ILCe!h2InPA{G&;4QT<5La+;TegaSA<GXdZbM
zRs8ikmN4h9J|YRxKbD*f49YcHhsX*hg8em1rJ7{{m~DWP1-KLdPzl%~a=p9~mqotq
zQ}4vva<lR~xK-P$W7F(A?(yd7mq=v!dm8)S%Q<gRJjz*^AN=gvvW|Pm4vz`aMqwt4
zj_LaI{5EC8;`w@>JK8VW^Oc`6txY0g$X^=*x^N%g)q&rWT6I(EjiHUWM;`0vyQm}|
zTzi}1|Cchs`p_Z+Z*faFJtvwQ%f_~9@+?C#v@ne=huLLSgkqc*@FqfNJerZ!k4)RL
z$MXO(_vraCvS#%_Aq|<@@*+Xf)lPn_b!=>AqiSa9Y0XQG+}_AmTE#?HA;E?dj=D+F
z%(RAHhR3NA-Wh_NtoakaYqAcMt*zLmVGwts)QiMP-lx<{2V7He84nxv1k`^ANQvFd
zlC9OAT45TMPTE6XuW*Z`6RZa-p;N9``|J&yAQPQnx91WaJEy&;_usittF48Ibi!N6
z#3Y!nfuO7(^etMq$A^9x16+%Rk+rp>I8Bd!sTZ#*m9fi{Y#%sp<)P`WV&4Ar%D>vb
z;Ctf_W8W_aVt=0I*NW;aXwZa$tRz-;lqf6;TH6juDgu}fxrt-VoF1H~G1I`s!<5qN
zGkY%jLV$}eO4nxrhc#oR4d!E59V}A2{E#{=qqzez?VpJhA%n9}If7v*oKy>YJP0}f
z%R+Ejouui-&<P%J3vPz-F|{oUMWx2i4Q#9G+Np5Kp?7V<4OUlHe6DQVT9|vI<e;P>
zi_LO$oojL*B)4!s{tPqI^`^*g8<j+`eDDr;ApERd;_0f|>DZ=in*X5J_4a8VDKU+0
zjOM3xenzOKuF|IiRYH^R`K!%k@z*9B+PpZ0ot^*9;a82*v*{n0AC27D<^D2L4i##l
z|8CNH1@Vd^F65_~SavP5S$ZBdrLq7?w)h4~s)BR>Wcq+xSx<jN2x#prMS?*_TR!Z|
znk3=g1N8#TrYIZ6!z!AZT(8q*9;^(v#hc$Lrz7@oTj&v2>7fT2d#z-RD`l822PNKT
zL&=RQ*gyvB+MFkg%B#A}h{9(DqV}i5gL$77L~MHyI`-E#k8kz+{)0Qn^LPApm_d>$
z6EfQ(K6t1VulLi!&isKuhfz#&>=l8rU{AE0p_B6Ab^i*FnTbY+#O81#`*#Kudnkio
z@Q<+kRba{U(<*&L*IGSq{#Q73G+cAMlsYG~2Ue{n5$0G3%^T-xE%sB0I5pjGyvPBp
zasn(qOXlwib|tEa>RktwCspOF$9n-Dq`-^fv0M%n6I1U%lerf?pr3FRJ4HGz4MGyG
zSB_*mQ^G-zI2E|!Px;B!;9eho=J)=I*#+s_&*G|pSo{28cyx}uiy?%@>>|mzgrioR
z2B3gtzw$<Ylo9_40(2{cZdKuzoS*50C`Gob3*7@V;zXGFy)WQ60)0-yMLk^*d7eEX
zG?#4UA&=XulAiFU7*0!w=Xr_k3*{-m%+VG$h1{SKpyhM*d;iYH{1@*@G^434(hX;c
zQ88hGJ84~d@l{rZX_(8%-m5V*kgZ)Q^>GXLxb&WI(MsK6>}E<)6x(nVAI-WQmgzZc
zXtkciors<H_B%~AdZt9H-5oah5XB7~unJ9P6Gbfa{4Q5h-+thUwx!&3e!wuBg8c8?
z4k2E{fTex=!7=B9?*krR<CJvsd(01bad<iVQPih2g;V}zjQ^{~ekC?TU$;`hXc^B9
zG>3>N3GQW1(JZQ@oX~4B-I^Uf=3B*t-UHfs(ukKF#t5b0SuZt8QR2FDtjs^RL!7^i
z?gkF)s*u$qPT}sjpQ;c;l>OtmlMj2w4Mf`~>m!Axe^glcE&_k%{nyAx@9sY)CGTQc
ztYzc<;VH!(no6ZI7m?q8joo)?X+F|ZyB(~<LM&}Ppbv&dYC{bdlHc+~E0{^Q`<-N6
zm4vT^yo;09I<LOT(31V%xvfZL0_@MFn595QGAJ-Wa3RqCfzs*TeRTFzzY3_tYsD+a
z!;a)J6**N3wD7gh`9m6R1Wlm`U9DM}YAV@kothTip*heNwm7aKgOTP(g(CB`rij(6
zIMIAszV-yC?+Skq&E=*VSGMVQiYQmHz{^y8Bt9Zz(Z4ZSd6KqaJ|Bs601uQpNU!F~
zb%gy}WkJKL3TI?vBU^&|m7|sfaJS%Au~v#^$6oB`34jOCe>vul(s)rn|HrkzJz54&
zVwK7#$8fyQO-+OyE)>JLYrUG@7aQn$162g~s+^8v+6J_43mi*vG~>pX=#1q&dF`rG
zy}K`8Ftg*~Zr-kw_kLH!_|Q&I;M%pnDL0g6g(O1d(o_DVhS<DkGC<2UScmq~_bkdS
z@rhjE2(vp8C|&mPgXohwU-^~s*(({Xd~g@kD0j<;d~km(`O_K8q}h^cX->sq`AmeR
zFVV6QWLrgAg^SM*bUfM<-XEy0oJZsyKiD~`C)+_PN^V8XF%w{Tf8{?f*N9`IYWLUs
z`sqJ4=1^|<Y4RsfF_DjuTv5?V$D%T37q|rDGkq+W4Q7`)=D5p~@;+tqNLH?_#i^e%
zr%*}vK};GUpgp!^XAA8YtHO^Qi9??7<MmFm{$A)|RoY*lIPqB0iu=jWF4>6^`WmL>
zJYN%JxJ-_tu~GMX$6f068`wb5ixS)xYf1CSa+U6gtLQvFZRj>ZULF<6K}1{PpUC3C
zsQL!kvjLc#-D4n$tjPW9iFhb978Ko}V<DaS#*I<MoAfoZzt8%8I#I{uMkER!vw`lQ
z?-Llf-!e^^IzedxU#CMGapG&516|@%;UjH*Y{k^E$0>DA$1BP!UV0*>ifn!<LHC_x
zdj_3S;86wQp<Ttd=fkdQoI4D!OtJWhv(g;-xwx9WizjU>M?yU*mG}gbu@yEv7$IVh
z1AGJennNJ^tCa%?#40*??V^xdOzxgr$T%izF99(uQuT&(W$Z-7-{$5q)iiO4ni!}}
zdv3?1!=LYc?%D!Q#81JnRDPOc#Ckk%q|5W6S6y{eDT98K#j|9>)Dc9zMp|XljFG#K
zuRko{#>XWKsVj1uoffwq*CbKeO`W3tJGV|>>UQ<@!Oe~1!6X}!MrzKRQ)HTl3~^2V
z?dOqLwVYQ2)el*R1^f)_lE|YdXLb)oU~iUe|Mt6@Wl%WZ@{_VV%{t2|nl5}5RF<tM
z0oUOF&Y^g`dxheg`3_`0rrFWroiqJKCgc3+t?<s-aaE}3xE0UeahM3@PB-k-WL{a?
zS9D28SjOPNzo-M?ugZ6wzS?2nF(q2qHT<`lVQf-?dUwYirX9uRvh}R$k94b82^X({
zo%R@sAF1A|+d8dsT*TY{FF!(Eu_9TO`B%x*&U&C(I=`!#iU(esNQG)NAqom0pPGu2
zk^sVaHc1f8=f%BvLB@nycG`HtVhY3WU@a^tn4Rhmv2!pB_wWdw$1`)xgO`ap?UAxr
zTcL1$FJu)>MP`3GpWmDt2wusb)ViH0){=(HF|x%v14~I^=|W*)g)SkIwKULx4Y>f7
z5Y&tB%@~ZT-ciTxv2?qBg;U@Jz&NHRH$TfdkRI)Ra5pmr<@`|N6>^p>0!-Sf`5E>q
zQ-oYhD*vLd5h;<FM~c2Ux|#$w^m}tI1`bTp$-T@b^mHp(195p43v#*SwZ4*18<v=X
zzX=Z3+QNPMKN<IZ#@|%?#V>$vh={mZqN8e+Wic00Bh)=j*W~etW{vfP{>h@eMKas_
z$CpoPIdYnjD_XC=bZU6FCyPpJe^NV6Z7-P~hDm0lYrG}mTwI1t?iEWFc)9j197x@p
zeo*DM%gi3=>UMGQL14;hf^Ec87SF$HABdH++26HourUZ$<VnH8rBr^U3fr+;qxw=a
zW4fJ*UXvQaSRd3MA5o$!(|jetz4X6x%SIxqh=Ob1IsD~5(g-bH@V9)uZ3gQDDZ1xJ
z$n_WImqm|Dg-*C|-Ew7X&>LA#UMj%~ET_%zez!Tnx!y2`@cuc<MPr2P^mEeQ^>f@H
z-4wN-40Y<h8n2QQG2RYo-1_g_kX}^lvt`25tT>mdHT<}~VB4E?JDUqOLkvaB5H`+>
zLK!SQ7#pwNjK}oMupgVJnR4>i(QR^BWn=zQFR3upNe&;i=Nc+g%Hync1zsFDj0^o3
zR1jzF?w;l08An94JsL1*;t>GO5ij<eW3xb$$Bdc&3noXb<(|;cc13u^L{P>Dh+yLp
z7!=B#6k^~-w(c>dKXc;^xn;o+HBl>bXaArzuK;AodV-Q;63kbSg_Ul5aI8AhKDQyR
zbV<WqgXf_FL+OC~!((<JZxB)e^9bJM_JL<`n6#lNkWK<^IDqBW=v-&VeD?(D<zeD>
zaCa>-2q*-CQ_$3<nx`BO2qz*+b^R3?VLa>A0h&pu3aklkQ>B^F`^CLtHw8SKJ5hRY
zaA8e=N8zUoJY{8^ZrD__)M7yuL7j<5@dPCwbjPm?+RI+iw^vIgRB4!5%vk;98(s2S
zXvn3(@vTz7ZzMljb~EoC<(3wcN5J@P>y_<*!EwG9%2%|$??^2HyJ<D=cw)tPAb<2_
z!yy%6nGbkGz}t7sZplPzyO<y$No&siRnLj`u{+StuiPKbGu7M_i{d17?IL-&5XKK7
zU!F)h#9X<ZC8Z0M7&vBW;J<ivySI^hCB)DD-hp7q`U_G&Ted{)a8+x7uOr{VS}a4>
zm{rW>Nmu=C2CH)=b4z7YK?YexvqNj%yn$C`9XZ!j5rJ2~5#FpDGgP(pnS8}UHj=2}
z+9ah!ozCZ_pV1i!`5Qom=h~F!E<N=`GKnW()6LAJ(LUrxj0)zFJHCEprdvc8+n>)T
z((sO!W>PG_=q5gVLdcHrfH#uE!f}ni-E(6by-`>XQXuEq0~B<C>&F~wxEIX=gXvx8
zM;!`wDg<dr$^X<$B+pq*NKTtZ$%?A;iuddzSN_Pq9|^3fyx_^-teg~+_CJ}UCMsI+
zPJgXt=~s8HsZFOwhf;7=0MG@Ud7-#5De=HONxF@!k8x&8bSxppYg7y!a+0TxG*gc?
zu$qJjI>HbR%(-dF@EgE(8ku{H<YOIWr?4yei*Kc%Y$x^tO+e%>vn+AqmmZjba5!*!
zk%3KGE)vjXK`aQEd6mio1l9qj)A9%MUapF2?e99%`h6-wFq&8S4X?tmWA2LDdy#-v
z53vNFcY0$@$@gdKFzdY~KN9SkP0T!|O}zJ30;(e)8amS`mLAYnJ*LguTd`bB^{Q@Y
zY_yryC~%!(vXM}t&-p^X`*-h^TVr3h>K-C)JaNCGNmgE2ld{LvW_Cb-N=b)_h&bd5
zGl_=SqiPImJ`vxrRb*{*1iZ%ghLN&O?UB{q{CeR_June@zuckPu4<3VuSr2rS9e94
zPtm8DJcdODb1dVKDS25D#6VJ}vr-^c-AP9`$umGs_*NIrU%pdT@o9Ft7FTr9#NxVB
zZoNA@LD%VO=(f%AWv``O#3VPak1y*(=-*$J^EQT3UVfyQg1qiA;^>OPF6Nsa<rH#W
z{VK)fWSR(kw&%4kRd8&_zMmEKFd?_fyAiOsfrP;Y<SP=C0(YTs1E8(uXyE+=E8<bE
zG@U~V|4m@X1T+mxRC6#UJ2NzWYA3H_V<=aLhI1&2VNt)0-}7@1ut&JL`<RQU-*TNJ
zR37k~<e=WdBQ%qrrhPii75j1^u~Q&ZJ?}J)h?ZX!Ez-}EPQxXhC`zsIhwzdjAesH6
zk<>Ze-r+We=r+w9dA9=*Hq&H2Xgj#nEP;<W=uG_rKr*4g5l2-6N;Tq(eg<4D4MCJ(
zlOCyYUZ3{r2DeU)f`YD>pC3ccz*L?HQJ^ggv5W}pcAXJ(32OiFOhmsHL21rXc!3WP
z)bnQY8wVn=k{{+p7QVnCArK^B(O2wPkom#V#8?#%Oe>dY@*Bvv)(2>{lHi#|&dw8G
z`Cg8;I8EU`%*8O%g)~1KT5^A#9S`+zPV43LZBVpzomh-KVG7Ig-?Q^qc#!oV+`VNi
zSRW?v=x*$IV{k33>LXAFCA=VHCP=S&joCaR3p;M|5eGaOqF}Wrf7Pwm$LTRd$OXc8
z#tKZHejg<`jd_%m>O>{~aS+HW!{(dZ{~^Ntt9St_@>JJ)ihbpt$9h`GN+v2Ki}R_~
z=lqNLBJPIW^TzQ(VtIjW?vBm}4ktzVCwI>F19#d}gMIVs_IRC*CqbT*n_k)H-5`yZ
zJ;8ngWwzG|53L?4`q>BKL|wUX)gSm$bQ2zD1c0voA&|yJ4jGT0mH^<L5q1D!636!L
z8T6fieG2he?lb-3--6rJ<YM9TVN`=&uICj=K!!=u#%#u+x*7jFSF6seG1J3N#X_=^
zPT0puYV@5}iB3ZgIj21fp$HJb$qDKk0)R1)X2^58p^^)dNQQ7SkT}3eQ{$)JgwuSK
zu0xqTsAGUv31Ki{HuX*;;V$xLl{|on`}JefAauoWxTH+pp%ykr$jJbB9-xnbF~Z|j
znW-$PAMx{>tC!vr9MDyDTBc8(^<qYr-z~EW3+Bd7s^v+E2u4=0RE*L{HD4)ihrK`w
zWpaP`5ejDg+>H;{T!7e95$Ai)qDv`;Y7_J7qtD&)Cf4UmK?93|P20l+`*&8s<qWN5
zV~f11m*n<vg)al$AYXW+refmvp?iuwa?X>JKTjFU-KPXP5{{fdKlyH~z+^2cm!9yL
z&P)DgIj(i+19Eo4$!D<oWq@j*q3HMQrGSA?Zr$QJhD|kWX&qFg)vYa7ONZqALO*+y
zgwoN4g@2tze`j0~dhzD%y;jj_E}gN&s)A2PN_hF7%i0Mss+cPyE52QXik-{;Z@#`1
z?hTR6yI<kL1+CfgAGO<L%tn=j%p?e~g)d8Kn0<Wm9iQ-iI&R`M=bqfLa~>k{#@m@P
zaeQQ#h0S5L)x_xmM-bMvnVZT$+Sm)k8H9ESRQ<C4tY4{n{Znv8HP6d8EOLv10Y&y!
ztD2X69cDM;XeLsJr%u?GVwm%gpWk<};v|uiNPEL@tx3T`Hydvy!|u~M<2?!AnCiUo
zwh9r+HLGDVEREHZ{@>rc5e$PGD>0elIrss}oLTu$&hbzLHOKZyGVy(#44R<1oG{yN
zp(c2RmH%I^VGVW}U8PA#mT-t*an+Wn9<TF7MsntX{3pT36@L1DzvTc++c#fGv(sD%
zy!95{Sc>tqV3+W`S-doI2YRFlGSrF)NIKI8;D$OF@}>jnhvq`dWDC2y|Nj{&7|_Vy
zR4h|UPfV#Im?aV-6g#T61i<bUWRvXJ@$k?u!fxpiDzxMse8$F{Q~nCxKeT#ErU-E4
z2Y>L|FirD%N&gMWswVEV)kV9?V#(=IXnD+28o`$IK+NN>syP%B_ee)Vo*db;No|zp
zPH;dbR`_b=EZtS_(MxHzhyq1^sPl1%q&u_uQ7q#cS!$}y46RC9)VjRjeJc$mivCvf
zxWm$&_wj|!5C5IBNbeiV8uo|3<3GP$>E*VjFe2W!I^V^O94pl0h1~O>omq58_YvNI
ze>RwZ$2Nla;*Z<Lp5I4vLV<8boU1Doe#~<G=&$byZvbv)?6(xr!2MdcAncI3z2@J%
z{Tz8PFp;bl{&7hfp(^Z<-C*BHv_Hkny(*`@T`C?s+eha`%@2xd!x`>HwU!_Lrd~L4
z+UM#hyclUMqh;((B9tXncDuZF=wkXOFa8_AvC%SDwtY|1v?v<6R=4!gRTlk~N@MiK
zPvqs*MydK`fr^~pcWk{Prr5*9Z&phyXzzIDMLKs-i=y<KW3CU3OvkM#gv=c|%CHhd
z)8L$;Q<ALYbdqg+$w@`;9!xC$(pFT(0KM^nH_Wol*C-k?DhL7+D=*){+D5_g*`-?o
zgBgjZ{Lb+#WnlTPiR2QCE}rRkCrS{V&mX`E6DRGIsdqOj(m<|DA^=4=RbJLc(;t9h
z7^C%fg#rIR8Otel5Odv9r_{BAXbzhs(8W0t#^F0WJFH~S4$a&|fh_4Fj<?Q<Zf?B1
zyaIGON+<8`BM<b_3{=XdOf$DWb8zysb(M4{8OeFx8`DQ0a4Z@&b?{}j+>#Q@_m^VZ
ze7+T&ne(Dx;B}l!(^sOEox#kPhVRv!=S>upteE(i7N)+}AYc2KIUm8%zp6@nCGv#>
z`U)Bfn_R=RQ(T|q2VX?8Q|JjM^;h6cT36Vw_SBWDPmn8E?=P5k{UW8cb=mvP9+qw<
zZ|L1#E#UAl{we<u<I`<^soc(J07cDDi75qbH)css5|zQD4PolM{jWZM#Ky#jKIzZ2
zzx|aKInx!}{8HlUR%vuuGki|b6<)ee?Gu@GdW76F>?90_x`}2cR>_V~+@${exA!kH
z8X7ljsq&bUhFD{K(W(%SDSlIP`cl|2eG_Ag=|rHl&vejh(W3PnT>E5B5nHIcmhGkT
z+#P)5#E$GPQ&+nQ-xd9XO4q0=q=CWhl4pjChayk-L!->q;30xVqib`)UtD$s+F^vm
z+$!66I-yv#n|`3>^*dgqRUEiR2Gak74caI-|4O0}jHJ#^mfnhHeq}}^%h(bcb8xo)
z7X!G6{j?4>`V)J8mB50H<TCz3H(I3#-L4qr->){;OP(M(qXYN}^&U=u9u^N&9Bxgm
z?fo+pkUY$l7b*&i5CQ<8Igc>lFkIxtXQI+&%zK1#;g|$kMh-8-i$SH%#e^@ZA`sXk
z#_sDg6;HogV_7jEUKZ6M60J4YLCeVf<M$94`p~(OWOIxxGyRWG%IV)lpb=`m0XBRA
za}Vy6U?W7Te~Wy`Ux{-1nJlsUqJcMB6Qjqj$(S?Gx8~ysy2%}zeLh+pH;J$*FXOM`
z#z>}DuQ>+TQ9LLyoldV0dX=mqAEdX+TdWXH<JGJuf%|OPyF3qVU$59&|Dh9?%8k-A
zb<Q3USMoYd9_T;@tVJ_CcvO6ztnl6t@@4UTOpd{UR5AZ9*UXBU!Lf@$C^b_<_Ho*2
z7Up}(s!6HSEHO}2HyNw&O1P@JnmrZOezk~cWkjQBc7WhcOSvxaVsVO1^|EY(vSc1i
zlmVay@Jy(_D|@c~bdrK@ujBB4>a|4=?K3}5huq7*UH!E#5Deh*FK9I4Os?cTAbX0G
zcl>v*{&Tm1T{OdrctQ`m)|9rnl&|-$>K%Ll>|RYdl;U^d_gZOK!z7I$ajkGs8svz0
z>bm^EH1AI1`$0e4+CK1f0ldTtrUpn`uXy@BbCW7({i=&S@I<7oy96c_%>|4>xQy0J
zZopgD61ZL>AroI%NXeumR|<mMzZon}l#23g&+F2CD`Vb91Sk{J2MVB|`nOs@BUcwF
zDU3P|PG_d(Xh0nyGZ|AcQejF{Jt+3()|)eCYb3Cw0+&F#W8fCp8UR04shp=BSVI{%
z#v)WS<uPiPN!Fi%4c8>q!Atj2fdKzs22QHix_U6b(z?p7trJ%RnEmf%hlW@OXy{IK
ze7DmLb4n=VC0+3kO|*XF-<OqPM3EQFK>GH|N&iGEmBW)rH*|q4WwZ2Ijb22AxCJLl
zRcy$=<szfPvYqt@ftjymZQuQuj-<)he6=;_jp3UJq`b<70cZ(7)V$Q0xY44$qRAd=
zX;~`KwI{nIXnw%@KC|m(l#qmvK%R)zbz_J=S>H;!&8Fr`m9f%FNbF_B*!O{=6m5!J
zobM@LRhjSXX=W7a#;DLnnUIn7_=vZh&pYU|>1fHWII>3Fs;newRH)d`CFr4Coh>2o
z)>r#Rwd=z54ov}3*Za0s6}v`TS&QBcVZ0~)J2#qelCyWGXkQy~rP-{2eY~dkX1b|;
zO=Ynm56)w&HU)Fp*zNfX)TfZmhikfPc6OZILpSaHr}2^Wm=Q%+sGXl{DaZb1h~c4j
za#Gm$c!145!_U+%vv1N(Q2zn}tzhjUyyx0+y=3nUzyd9d0PMXqVnrnZkn(LYcdOK@
z7b1~gI439=I^Ec7r9jtdV{r)O5~9Qmz)91arI19DAW8qzX~leFL-s1+31QCa13w_=
ziPUm6^PcHk0f+$(p|y2^Pd*`gR*Br)gN(ObN-&~>9#{@M6E|*hJDd2WCTnqR=Bo{X
zM5X_6n?X3FXCeYex<1SKxgbYg^9$|UE8a#nO2c>d5BU<j-!5BN`J|f&uqhq06mkl2
z>V@Eu(R>L$1+jE*SNvE;kx&hWD+PW5VYS5HhF_ouq`yZWXMH!?6Q32nd;ClHweKv%
z-J%aOgq9{b@;+So7$Q1BF~%j~cq~e9&yH)2TJVKoRtz4t{Vec>iYU#vB-pLY{JHVu
z>K@*0iQ6@NY{J!aDu?m?g*Z(6@*(#m+>$b0r1Sa(n^Cx~GAKH!W3t|DrcOTYs%|gx
zA%beKz4SWbfUCwZhs(6V*UO=H{BBkb-_RxsBdwGXq1Q~kbio~Z6vHm`Gp>tiG)1P}
zR(hqPuL|^bPQByMm3jZ2OD#0jEX(xvHC-+k%zfEFt_Bgy`P3QVdv_@Um%E>a9*M8&
zul|L#R46oSo?#mEuNhj;b2V$Gj=uRf6Mn?%GPAdmO?ViRaCLd)x=0cttF!T0V#Csw
zb^UfESzw1~<B59Sf<&gV7B;Zm63GNC9w4yh1(;UXSc6~i7+}BZ;Qv>CtQc!tpfKzN
z+dfJA1U{`_ob_8b>(y#0R^b`=X!`An8E}T5X))yFyea8V3Ff9?=Le-gYJcXxWeSZE
zq!vW-pI%ry*07q-c!Epqr=!5;*|9*2?V0*p8%D_FKJ?tQbz>)yZE7V1iw0cb?=%F5
zG=3Pb9|=^6_W;Lr`&s7#m;+~b#hEmR{WmbHxSoysU|zK`JUjVyh5J>+$!!8iaHeal
ztxBJoqMEp3PWZr<K>SZAe5mi$ovo0Te}yGFw@Pk(aVxLl?e)BHJ!r*&>91TVJFFS&
z>tcMzq=1H{V_KBtdC(Z|rT9rLpgxHn`ANxn_a$lZ`c{mpQgmIg-;4aEMyjyQVl{qS
z_Qge4ubmR2PawnA3;y)HLke=$)@o~M*VtB8I$ZCY3gr!91`p+f(la~GyH9MZsj#!m
zjynB!j%hhNHd8PBkwZp^h4F;n9j=oMdcCW&`)tDg6zR~3H6JDA5n%!>O)4*8>4WmC
z04igwK2`683HeAlK*qh=i2im8-@0Gs$Zmm!(nnQHo#{*loxkDWGd;uIbfv&U?RZv=
z@*w^BPdFm|c$;sZ{DL+j8pHlMJ=7^*?m!+F5E>1Qik<ugC(Y?7C-&{IPsdzuq0U5m
z>W+RqC|-%s&X=}H?+H=&Orws@vpE27;Tc_h55QGY7h^a!iyHyX8HK0L!H-4JYJv=2
zmX#jr9(1Xt*MYkz(fZEkH&6jZfMEv?YYU_Fy9z&J$6EM1W)$|cWv_{x0@;-@gc%zJ
zGOZ)&oOJN+@d3AsD?5v6WloqZ>3<ggwmbnoN-#H#J=2{CdFh*`bq{OOFa7&0$G9hZ
zJoF1%jWbjq2*2GuV%fg|Ipn>wLeu=$NfUarQzf%KNvdTm?wmizSO%kF2Qe_pRRuAy
z=B<Rlm!~9-7nHD$<t99Mv~$hF%B#mrdM0Sed6u=vkjWGl@6LD)eRUZB{--0J0YX-M
zGNr^j=cc5d*qD5N;hlE81If@mXANJz>hd!*LWVFHyzNNntL>q)ZFk6d@bg2*FLdBz
zFGE3Z`DeL3`A_)?6AfJ*ms&3Wh)sOZQ6lNGpRL)$S!+B=n|`xpbh<|p7cB6T+xP+|
zJrJ({)x9?gqh{{?M~SbQm#;=us&7&FoQa~Kp<~@0dWTEWaZBo9Zd5wo2f}xsjz#V;
zSG~!5oty9ZYEon>z3WrJjg+Qu*XY+7vpNjcWU6+RV?=U4UL%{RDeZ00StN6(%g$S;
zbRWqbbiG((>6YQFesGP@W+CFAuF#QJ+k|`27*-k2Hu$V8hS}0B@-Cl#xfi^x;Hqup
z>%oThWt{K&9jX;|M8f=L)!n2?&0HhGG#c8X9bX<>EsYBQLy4FmQZJa@^#TTe*X%&-
z6-EN}QU<kCxq;vE5hA)912O+=D+U=<Oh|a24;Flt5;`<qsJ6b~1qtDh<i-O1IFv@w
zfhW}>8WcwjVBRsfSN$4w#IEBW9&$G_QN=AZ=CL?26>u&=#6;nuOzp8uO(YkgKMtjR
zbJr#E;rqGB`4RdBgDHYzGVEu&jCQ+pPE8cFy}ciscgi}Sc1z;V30KD{gySeq#r&TH
z5Zd>@C4qDeI~FYA<i42C*m?qrUt|uC8YHz}0c012fk&MRxtw|<;Vi$2;9n{z*)j8g
z0t(Yi(iE~@D%=S1F1J{o<zHqf#6cZei0V%^Y-RNE?l>Kh&97FxpO?;*<3V<ZeZS=c
zK(D-({&`n*4g4(LCz#*&>&2l~sPydWAAAnX_jx<U@=)mJd&-l6E3=#CRoH3B3wy~M
zh0@pVNtFdNMVrjIuG}BEbiQU*Iy!<w`KpO7t9Y-{d#}f;O2i%JFi|N&y3nhlp#nu~
zz4>Wz^cEjo?_!w|R+u+s3Q``n1Zh?dDa$}+&^?Ic86g^?mWrRu{s3iK9!wS9#2cR_
z*l6{Q25I7SQOHP2mAjeEhA)OLU8HU!C<5K(VLhHPxV^89)<%b@i6sO;bptvluM`Cr
z6V{*c6`qzbto}2b)wz>qBNJqN*Z2JCG)drOIp>6D&nRnG*Bt-t6Bi%1ahF)c2uZqH
z^CwHfwhrsSU@<zt#M5Yzn56JPOz>5K;eY3P!m=9WS<3^q1uikjhu_pSl~iT_+OIvt
z*z&~}sV$%tmew*!(M$Be#wPY((v3~ulsl1S3=Rwz-W?v<SScPMNEA(SmVz6>8+>U+
z3ee+0C=j#mDC2=f14Q+4#n;nkOK|?wis_?Us|TF(d?bxB(U3#WcS5AtNy1?2xCotQ
z=mxLtM0|$kgq|aSDo^aNMx64y19ghKxGh}V7~j}wt)F)gSFM?bW_V79z-)CQD~8SN
z1!CF4Y8<%<WU*`vUF!W8F9!ZHCV`NuRPL5i2eY?MdTYY(I(7jJI;|bx3NY}K+kn#H
z0K)8O@_t_?-i~59yuAHC$B7wZ=zZW;t(C8r>7G3);QD83PCXOh;bC<vMM{8lt>E)v
zQp<;)(9G)xG8dz>9xMoDbuTJG$MVftbQb*AWFL+;rI$IUYM`emf(5J7a0KN+TF&$l
z%W1M)ofhI{Pm#ch+@C0)X2p+3E=p4hop-N|D(RaBtkhqS$0PH^fnP(R=<cZ$&)t#t
zcVw-Ron1jD=Ay_*d7dM2qO{fweusaCr#?nVnW0k#hwfQv+Ek>!EYSG|9YZJwhP*xZ
zZ7a@0=4W?YS{f*leo0`dJ<y5q>%+=3cOSEme>cY^(2(POX746f1>UCkthz+P4F!k4
z)0y6z?m1<q+;}bXs@d(c2u*toeE-s$Mj4;j@yELlLhY65DB^}US2jb7w6)cgsk+}5
zIj+94eY%j&2j822{wwjOOqKuDVGi#uqi`#Tah%7>0SuGZ=n=GW{`0}jDBfU)cJ|PV
z+)ExyGxAT}oM~k$_vU*%whWAoWTnrbFdT2!rSl#B2ch|&p+LdrK$CteG6P}~O_Sl~
zKO`4Z>51iXBa$^l|9M8M=7TlT#RHq9cLWb%=D@w52Pa9G##g}d?A*Dixd1)#FMKDG
zk>c6TA;m4^(<ACafiVL%07!Ov;y9rH)TzFvZ+e9S1FOaSX<4xBTmY+XY{g`=)^sh5
z9YYW~k|ZK<t?(W(eLo6B2>6uFNZ8JM;ubV)Q5ChJa~EXvw3r0~^VgABA!H~3MgvLv
z6EqrBMA!Ks)IhvpZ-kfMzuWrZm5gt0IRqn7|DMv_=L^PsU~i~WmghPytg>sfnEJ`i
zGIPv)-i7+%G<&_`J&#Coh#`ThE?~ok?^KLQd%{MBdLYU>B2*U^a>U+6Dr&F5*$HL}
zQ&}Y!>kL<Cc~23=){<W9Fw?-^D7y0c1^NJ4x?Upk1HSZ*x}N;~fEJ3BAtj%!Dx9@e
zx!oD7z^K2>^hwA)TQ^QkB=88G6m{k9FA{UiEzgCT96>Xbnriux%7-V%_ap<%VB;Qf
zo)iJ;D^9SI0?f@#>5(nw{3q|V7?<8TV^2Qrqeewj2-fra<=LLj1bc1$1?EE!yH#HT
zF$k#z6RaD_xRZXGX=+--?qMX*y1Nj!BN@Eg!*v~%;J0zT<NlZNyS%j+L6V@C4}$A8
z@#AEs=pA#AufF=3=A|sRgnnWVZ-#FQLfro0X$B6gg=`&j@2%v~y~qQ{icHc1OC{ZF
zAVMjqGnfLRg#_S{67U6pK%kN#U0MU(5fGDW?_5dOH_0x=fvU)TR!NH^^U$Dc3wW+N
zi*?kv?F9*S!|sl`)*vO5^^NAm@U(6A$!zSEu?mylA(elm+6|f~p63713k=B^AV$PJ
zQ$Bf;)DD*ej9(4ld>0gJ$oVH=4IqHN-edn<y9IUNUO1D-P-#GgWkHjV20T}P_BKZ#
zsRmCD^v54-%L5s(B@zQ1v>xU+F$qqWFtT|k=_7dD%xxrB<zycF#&%V;iz$bi-b{L8
zU*<EDf&xc6L(Tbmf{T=VCkuX=H;KXUc67xodg{M(EB5iOW_J+&?+4mkg0O#~gS_Je
z$E3%9(`Y8{4j1dE>bG=$k`BddmL<#$953AWop9$s)q2;ZG{+VXd)X@8OqQN&hGZHH
zv@h7#c=;rP?G{ZKpi4X{YZ;|(g`Ga9$i|Pxxd>&O`ViJsQ=3@3Zzx|3kJ%|cEx$I4
z51HH**d^Q8c+d>J&fdH((hTdm9<cAM4_#XjIL#sIK~?P^HeGH+7s(-eUr#F^gs3}X
z&1M6>5$&@i_%tF1$@H7=R5*p2^X4$J#KB;dkBNWprqN2<cg9W3{5-hm)9Ut0%1{w+
zoA;tjAK{wQ%r#v|RXZ`vI~hNI{=xC~U=S&gETWZv1pV)vqSBF=p0`|Y);IgQv5rJ}
zImNEZT>*6gS>fMooeiRI9ZJdl?uT*Qk~MqceG^F}Fi3CpRL`Hxp1MKOb(f#dm(0s?
zUCExrJ0&@S;NEDaX#V@t=+ZzBwYxE}C}S+=KIhsovyBh{*i;1lVJNK_9-HIY0*z5<
zy0b8lLw93zBUldvi4+#dFYFdjq}0Snd5687y=C#Z3b<TM!hwS|`m(<<-m@C`vgy|$
zuy7tMMegtnAWL#IvK;s=abu4ysu+XAFz5Eq#mMP!Ip=}rD6|%+<pn~tJQPDZMJG<|
z<~7=%>aD^9gR#&5CYhGt-l4q@m<J#AT`Ff+zE!E6?%mS~&&-|-s$sLjOFl8;lL{hd
z=Jjy&BxowD^7cK8Pf@uYRKz|kf?>wUXy8MpBMt9Ze2Z{nFexGQ8RA{uRVVP+*aV~%
z^0;5A7O7XrvujfFx9f(X+|XEKx%=Ut_(Zs_H>Z5^-&@%znuEYMZ1AZVm%B$C4{8=I
zWG(19tZ>T)pVBgo<^=oFoknCobZx-x&0w+PUvorhw;yIl3EC#bWST8ooyM2PhiH8)
z@I6$fcl4dlb<i3S6?V_k_4$PJ{w+EE1<F`M5pnxr+BTzXR7nWI@W&xOwH)#~c>CSR
z2&v!+%X&9a$#BG)=)F+cwwYi4`m{8nl_H&-@jltm+VJ+XkhcB~!|6}diMP#<4_V4M
zBY17PwO$!;K01N#ryGiqoN|jV7BX}lqQ5M0j=!xoM0_%B>_XgJYP#+AHFk7ss2v}E
zT?B6a1=yRpj?~@y?|t_5#08DNuIYVnIsU$F*cn6H|EqQ|LxgunR{LY!#qy#9xwOq{
zNVZy@j3r;TS2F6|BHwf6KNCl(pFRi4MZ0-NSSxlf5ZyfT#n%r09PnwI6A%>f0DVEC
z6`KA;R(rgUW|Sg?k4aPkq-Y~zzj4F&X)x&~$hc~@i&o)e7udP8ZZ?3DW``Z4b-PRt
zOlA-8?{mvQ_6fL+EDNR&{O*83&n4m?_EA96i_O>UgeXL1{N%#oHB*a)9Z@Z0QOm;6
zI}4hu<XemGJ@d_zl(WeaaXc@=Bg!p?eYq59VL(&I{>8$Ah77P`Uh(fVCyCGK2=Ox0
zv2oyl#7E^XmCGbLgnGFKkUvajn|r1BJ+*@VIugXyjtVYBTC8X$hLd~6hEdnq2dAad
zWzdcVc6q=&P2Up&->wnqQm-BxS^0c-e$#@9(Qm$aaE-0QR+F0^0E3Amzu#0CGSb=9
zx`sNYmcE+EySC0_xk4IAViR7tu^?178V|o{ZCt^@en&F%u^;c)RnW{tG5Klvb3!UR
z3C)QtO&x_e`>c{C*J+W}SW17Q-xEl=2y$Y#Hb}(Wd_{LQkFF|PEB)LH1E(n};%BPC
zzH{5_)M+WJbmhhCu6K7E-ODout5(SxnJ!bAmXFI%2?HVJ)*}V)lqP~JSJLmH=7vKH
z;w4Naw<zka*)?g6ua}P@m{F&hYt|Lv*Wgcju1<3ADFz}L+<b|T=cGk83_e-Olag2x
z^XUbe>DAIQUrP#4bUKS6af!}2nRx!G@_bxB%NJaVf`Ji*-K8U{T~!ZxOGx^nLybQy
zYIzARvNEPTWtLeHEFzGd-l95pUMcua-Rwy7rh87<TRR>>1>?yq6kDc_v3T^!yU&%2
z8YW47E69mAm!1TkNbcFhEh8#HVp`;x?T5;L4@`8+#&p@+kPElLe8$w^9dsE?eOXzo
zHDKhZoRK~w=WGMWg~5TSn}_C@|F6_N)$0LU@c}m#-7a>5Ir9;wp)1yI>of<-)k=<z
zLqrdtB3UC(k)d#mh9Fr5UolbMzTQYq9o8HGp8k&EWAEaVzaIa0j?GqkIZd59LQtHy
z#_8*2JvOhZgmZF2bV0JP<PR7`_e^60+%~S&{F?ya`IT4o9$OUWq7;cNIaw^+4aA(K
zKQ4|W;q-?zFM*TsQSBE&v|9CfR<%~6PRs8Y#yy~>V2ST=-Pl%V3m14@@#<C4%on-w
zw)dC3@56p7=2Uf3R;cpo?l)7rR%3QUn0l15)kPoJP6{Ury5|!bJH0JgVlt2(S$lG>
zJ|TKtHalgILe#oW*R8@ZI%y$FkW1eJSWoR^Uc$-)-1I{p)<_n+V-eh46O}Y3i-8F%
zCcoSrxc$wR(-6ZekqGo;;qLFM$ZobZr%%V1?l+Jh);-8`Y-{8PF~MRHgQ+bR%ufa;
zyx?io_gAkpvni`$(%=ehB9ji~%_^#kPc%5yvJs`x42Vu&sf-KW3}U@pw6s94&^nPQ
z`URhvA#5)tO;Z>kvG4l1P1^8;HIhM@*RO1^<QW3%=OgDCH{G{G-KOa`XDLb}CBmS~
zzVxJHEimIr4XZu>DB;YJi_phjijp&%$h7*LCb((jX;@F7Ol10!_5Q8ZKdsUbJ(5h(
z-t~MU(;*G}b|lzVxuE@recE+obsyy@gHh8)tU@cs+l)7n|DNR6??uoZ!$MWMA5se%
zCO7L(#heeO`0w^#TG~sFWCgOdstP@hXc7Iz8~V6`i;)I<QtFHq{6xb@Su%=WH<k{H
ziEqikoiVzRfaI2ZIUO`#`ht4v%&)w3OiSK8&zX{Wn*{>t20CV|*|2h#2~HQ@r$?s}
zsX0oCg!CjBThvZ_I<Y_QGE6{0m)+qCV^j>+Smt-7)0{oo;%Sr6QG`)9a@#r4!%2FS
z+e|D>y`JX9Px<@rwbufn8nk7uw&C%uGzt}85=vouZ~X4xmB_y3Q7J=s`Q5-UoR2J&
zUDMDJ_`D@$BA|`V{j<Pv!++<ZG*cQxREmC4(MI+%iLb0Pqh-*WZnmZCQTE)Id|bF^
zYlG4MkX#ra<a%X5eG3;eo*ps!nsRC2?wCM6D;PO0*Be-Er&spzaR}8%OvJN8t-ZDM
z{eRq3Hh}lsH4k~SW5fJCl$3w9yNXFN+RyjP#h2Ck6#$~ud(8Wk>D_ooy6pJdiUhv&
zyxVtuy(F8;{ogvj`m?ebiRV#nvAnwU;vl<Y68up{@>CP{>Z?~glBfT?se5U=H~Quq
zuW6G?UgJUi7Xx2h@d5pK))r*UVTfdBu&Acf@4YdP!KKM0ZyqD9^7WF3n$TenkyhhS
zJLpF~QJI+OH}!=<@R$7l8NaKkIWPrT#}0~iq3Y^4vS+d=^;??O^lsz)afjeLH*NC2
zT<NSINl_fN;^w+-pC^D`7{Oev$*&u(pSu?Yg)Z;J%74od-UFn*YBZFFaLNfnEFy)7
zo>dz3*w~yS%$k#F|C}fBcbw=GhAN2gk!X$H$}#ICi&A+#k(dcG@@RT)<)@rpRf{VP
zq$YgP?ai?(>=~Cj)+jyAPtdLs9@O9?9#JRO_@ku1YQmsUKz9eCtS431v(#Y}tt|mf
zxz989u1>Tu0(%1MNBMJq)K{{9R`3t@YNu7u+CwQ(%nIV~Ab#&b5$s~fo`7M+@H6W%
zPqa)~{-AGKIJ0fua^2*IxrbX2!D=1JeM>1BlF{n9F`Kd3>00UGjEcOTZ0+8SfcjrF
zU(2pp@pWHPaoNrNC$u_TQsgV?vxI$bj(ab|*j7O1v1zYM@8xY7yp`cw9OB+%L>EW)
z0pk&^#8#aetXl&0e%ullzOH?-<zv-}{^?FXtuZ(W<_pu?5iVKi#)-uP@#Pu6)OE&D
z+1ze5IX8P6sCdF8THG<OnihzGADr-$Wz*}bul>oYbbT3`m7ByTXYkC>^N|{~bi0*k
zfUGN9Gg|f*C4SZc#&*jF^QxMSEY&Dpu&=FC{tYXNFp4{G%#CHt?Hl~avu4#a{E%R1
z%U(YAb4A^Mdvu>x8Br7GJ}%XjI(u31#usTer?kaE{$b$!V8-nG9`gDZYY(=zi9Qqi
z@HRS{Wu?qtd<wNG%q}uz5EIv$ev<Rs<b-R>2sg}v+QfTvej~g5HCgrtHbhoc6<Y76
z8B}x-?QZjaWO({msV}uqrNp@H0b}~umm%`~(EMnkLh4h$0)^t6_OarqjU&>?WPY#!
zl%;Gbb*KFuPf+T3Qc>Ej;N(&%_+P4P789cd{yTS1rpWUZi4{W@-Wx04)?(P?p+^D7
zHafnw;D4j=%hr0$kRdX|(jQS5tx3gNjNMtVxeV-kjyaC(z9VZ(LlO*<EEMnsR`7?d
z9s6YI_*(3dWY3q_>8;4lzjIy{7e#AIMf3LXR8M<adIF~vcIn$azKGL;HKpy#0W-xn
zhZCO9@xLl<n|{#KK5w;Ve)Cax?Ul>DP<t_*ow{kUrrKpz00S1FOHaMge;FV-MD+$O
zqGD2S#0vsaQ1Lw~OX7|Bc?_KnYxl2hEqhp#xfj19MuI*lP0Xl=<tM~cYH`Ic{$?**
z#M?+N8H`Ik@2#>xtz@L7w1RvIufL6m<8?m2$)AHe-NVi+Q=jDwFf&>yYqs(B^j6!a
zSA2*ax0JATC%aHd6{oSxJs!o(UC5Do@*De;Ed2rEJ&+EQfb|$V)a4Z|AjF(#7jt%b
zPkj6Y!P|w%@-^Q6<|gbXqx+EETO8t3@%(9SzA^bmr}y*|FU2Po(lv(JUcpR}BFt>V
z<cG%AMV58=mvsa2Y5W)ZGajwI>7duHYU#?G@~?cpALY`y`gSF%-D_O4G)`{~Hipf(
zEKJj*^kGUgTG{oE4_ca4<do^^m9Chqr;+Ho;6Q<EYh~Pa2F2>k&C>O4`9ABX<>UXI
zyX{DW$-9hSo#?r)<2SG-Qtz9%ou!>a*7_xR8kb5n`1U7|_3bAgk=0G5lweQVjk`DR
znWP8xGNX5tMv~L^BI6ZJQ14qsFteC}JmjU#E00|6aJQEc8<1}wzUs!c=4M&2)<~tG
zPSl<M?NH`~w{?qtF1clWzY)%B;z-LUN^mMB_yb-;MJ5;!B>Ct@9>%bYJA<u!0xz`S
zEyzR#-xdI+HcO?3w_U!PEvacvze?lsUd~BM$KS=M2>q&92C2O%Vj#X!yWRY}hL^!6
ztC~iF86k44!fwm&cgWM68@6aJTsY$YadaMzRPX;E?}{=rPO^&3V}$H7zs7NF+1cS(
zWhG>vo7J(Mb8I0ymA!ZNv9d}|WJMiy=+H^WyuY{KU!a`h^Lf8tujljec;2kD#kwv7
zY9kki{@exONJZp!aQCr_7IE{g=d(mD>5H=uEkqAygPsDDWpOs=0(c;TTcrWu7U+UU
zGXN$(NC13r_Ny*Q=5!33Q&DHeWy4bQTmgoi<>exaDOLR?bC`9rt9PwjueIai9?hS9
zQ>?Xgf9-{F?VlabV}|wqIZfQTJ<H%L=H@o9=ugm_5Ge5*dgeC-Ywb%#6xo~(E&e`5
zA=Z8CKVJXHCXdy!D4>?%sna01zgFNK8Jc@spjr@YH)EzGO8P5O{7*~w^VrtxZP}k0
z_YHi^B!BmN&9(mM3NZfMR1?LxRJ1#w>7kG*<V$<>)?PfeuJwsXMjq^zJklTym3Mi7
zZS3>mzK?6AcXLf}@hMk_g8ioDQya3MSKCc|S<#T?;>&`ejTfAoL6zU&dy}b)evVyz
zWsomQ<l6&YVJzEL+<Ka~T1C!=jqS5;3DF2oyK`6Go-=jT+xBzK+|#K4n_;=@BeLtD
z*qztntFgs@vhspq?K%4gGiLoyv^G}hRE{MAS~L!W7hcrH?}bxh#NO3|xmYKsfl!rJ
z4azvViv5@~p_(M<S_)O<y5(CI6)2%0GJWc!ZhUmg#h*6b1<#0)ha);>G)?3^8`Y=r
zMt@3WliWF~meX)u=h*-X$P0;}&dZ(WFol6kzy$YJI6ofjfKQMP;LiK?X1*5kel=#I
zR2;azU_ap{5S{T3Rm5G{FU7Mr-VnA=1zyk?XKd3CPHDhTMb9C|`=0pS(;~seApV50
z=yYncO^L?d#V?0bBK7BAj=Ol!?gHZ~K(nKOF$NzXIC&db08*VN{%-}$N%w@EtTJ<Y
zsmn{$6K76b{98upOSWQ{5hBb#q7oK<HO^9M)?$l&#GX=ix5?<aB*v@Xzl11mDQ7|G
z78C7Ee8hn;u3Vj2l_n$pw9-5WDoiTPca0{xokpvVoo8ye?`F*2RYj%pY@~M53XSPE
z)QqWq+ZXj^o(=d!?*X&wYw%@FQrEAwwLcpcvHZyT(8es>wQ*<o=UR(x@BvxD-^x$v
zvan2bx~tuQ^kQ~=`^=KZ#@##Rgm1-3Gr#B0QqS_PEq5v-9$lXKglY<r5h@!;>Rx2B
znthn0VEmm%QHp5u#8@C!Z{7L8?t`ydPx{3O<ULT(bZyGO-|##Lg?yyVeN}eEYn%8>
zlRo#r`$|P?g{M(vHBaNL)@%`AvM*z-5*5B-GO-Q)w~}BwOPUzdE#P2E6c#Y#m}!b)
za)~2FrRBVGIq35gSx2F)SGibn^W@xT@Q@Z3K!#+Bq<lPaO5aGysIlEay{L%$MY;Cb
zI(q`2pzVWr@U;JpSP-{Jji1<&NVe{Lh-AGm9k0`{r9nfB65Q{yevt)|=$IGSSCY^L
zTyicxM#(^3<zqTBCi&lmR}&rV0n-H2McMn7PhalJ+Iobcx{=Tl4m%y8&hw#`&1<hz
z-ii~u##U<tFb4w1a4vBqW7OR-i*;T)rZH(`<-D#rxs72hc6Th&q9YbVlYDA<w1Oe8
zRS7?5eis(wdZ!<YQG*Q(KpTV#emYbW9zGaiBg6InsQ&pn2EKEEDG;hZljvh<c{_%7
zHk!)&9QEQ^wg`UKc_Z|k{DxJ_Q(wtIB|Bgp&Sfv2MtR^m%O69#Q2s}vlzdLk-PaNX
zL2UD^AG55f-nyKwd9u2K{r1PaZ*2B-W$r5AGE+t9WvvLq86gF!64lsYoX%I3JU2FM
zn?UwrMuhmsM21`UzSKxTXjfof#=B<84~?szRWx(QntWBojtdN4=Spy`vWs`vl)&sq
zdu_yH+=>QGJrql`emz$GlgC>1KF33ZXM-pyR5lT$e+w^0GJ0uWMTbO?(<5nGpHI7e
z-=X6WG@cC_9ci#dC3>_@vXSQU(v%7amX#1?yL`M2(QBEt!$0uodgk9}d7+mYZSB@I
zmXG(cnevp5Bpp3dhg5Q;NRmK|*P=YEmO5F#N);tRQv0K$`KTsLo9h{m1*XVWlEQ*+
z{rj|4e!<QpdX$&@whWYqWh(K>y94U1-a+qYO1a#QGJ{k*iY`A}THb1N83d9?Z)I(r
zFxGj?)DL}By>g&`7wvQXZS+v$-pytdHqKt~>1KFCB;VN2_iy0k6hG-p-?dn@zHCL&
z2ZbqbrK3D{m96iyJ@$Y86W%4iB;Ipn>{Whd=)Sh@>Pb>kzkfa2-iD{RDc)B2h$X$n
zPbd}Ot{a6L?Uin3Qu+!Z_TVod#N0&zayGG}j?X6aPISNe@4{vfTbB>%l*&*4_MhnY
zBj%e@x)IsXrbXC7s*36muUKOjZ%WbSmnns`(b$e`Xd_e}B-JSM8mjD4@MSo5XjSXS
zSKMh{@G&%&_*<}rW7Z$J@l;M_@o!&KNgra}L-{N|RI8gEZwsO-J<{W)wX;`D?;e|m
zts%&8crAv)r>&NC;C$r1Kgrif5p#}dQUgUY(|EBUyE)<QqJelq6z#FDwkaBN>;E;R
zrFj%EtDsS%KW|N=aN|V4SCPWqZoU~qP9pR$l)ys5T1`rnS%X{xKty7$w3Jy(9*orF
zW>Wj9{HC8Q7itB`%zW4L-xzfNwO5&zPa2)i<ZNwn^)ygv#M3==@yii(Sbn$+K@D`U
zmAX67y7eo$2WJd+<#g8tke4GO^}+@c>p#YN*o}>6<lBos%sh90tXh-T?)%Zc77M@b
zU5Pb)79g3XqrYw=$WAwvFf?%+>(|ZHX7ib?JKgtnM2V%jY3cP9qT53Wgq4gUw}WpV
zmeJYOD%}|!&>pw;dpQ!3vSWn$9DOErU<Dz#yj9>}e(Y|asIw%H?$WU7pyEJn|BnFF
zJQ&w1Z9Zq9uDj0l7>B9XfpSa1Y@YEnmUv9H4a&}qPPple5MJ?JSn&>*Q6Q`KjrHUH
zo;*6CzVrEYlRJ#rpITo>|A~V2qO5?q6nX7oko$xqE>E)0YeET$pi{ylAo&XU-$iDj
z#RBvFZu`yuUFdQx7VyO!+}UwLtM-wf&UAAea)^&w)UCDA>?HnnlN^dd<(#FOtrd`y
zi252dJDHkzUbNcmwB%{Vn!{6aDD^G_@lQ?z=n7)TRMB|q5Y4o4P}JdOQCv?Z{&&GA
zc);1s;R6i!w}==jM;MC@rJ1$C|4@Pasj77-vMu`BazjsRomDES1%SW$5FzO@n|!%N
z)ltblBf16Uz0+L0%(IcbDIqFUaKS~X1hUgVEN&Im22tMX7J6h5`&GpbN|e1zmz925
zKM|cS2-SO^bzpvY&df>pMB}IkR#04mf>76EwQ0edu{YS+0e~axL{RV?83btcAWa4E
zAkLS<1yyPGfgKN;9Yzban0Zq+%RURJrfS8=>)r;6GR>BjR*NZe$H?opk1G`n(&&nQ
z^Ha;;wH(x;K!w^Nf#}ZPZyzPNRw%MdUu1jQ&@Ho*{J<I4dv?gSQ#pZen!JXRb5xHX
zPtVTyxt}b;dQDz8;jPa?eqAK<<CBgPj`Wz1zh1kBm!W^v37jmQMmkZS+2{mn6rAMz
zmLAQwqVPwu)ckkhO2aSk4hNv1TA(Po^7*vsY{b<4l>Lq&!K;ItRB&tWn9rO#aOpNt
z>Awp<Ms_e!sa<ctQDEcFUnD-PRr(UgGqU-j^RsZ~YCcSeL^f=ttv@;dw6$&%j+Hi(
zw)#AijuCelhOFk;g3?QDJuiAhu<3njR$njO%04ivd>?^Z7J7Pf*K`()AR!Il(zX@I
zAg_NABE`x<cO=^`I3rRZH<ofRMvM!)iHd`6rW^<mN3-23!9>H-;!Al1b^>7(4jotK
zBZbR$iVaAB6<%XkcoJYoxrQN%eEGBi-ZUuu9L%gfMgq^G#=i8=(wNL%M)KH9OozB#
z1e?l5$8op*4?@ncAg4IUIVe+$7C0=tgT-jXg2swk=%pJD8xFA*5S`*LlqDW%Pi`&C
z{M6n$7uV~dwoZXm(-EAn^3Az<1OFG#z7YQ$#Ef5=-C@&_oJ{l^5%;y%ufcHZ!x|Tg
zgv=dspR`&^rZ!yJEhvB9K1|8!Ck#_OUZf!Nqd65~W7Bgq3JHB|!1oa?Vz!+=)!m4!
zPtz7kYOj!0Ps*WvnEgFt+JpV_h7+DI)512!zK@u2cr4A9|CPe!d%1*TLdygc!f3#^
zFBdZNn;93HNA(%Z!r<A?tbd?EPpdb*>R_AEiGld5-K;m*XMJ;h+YO27bH^$i1aQK9
zm+&12U4rs$N|NGy^PSSBKlI**%ZKJ>7*hqF(SVtv4ZKWZ8$3=mLE1-dw)m1!M(N>#
zKBsidk;Ghtz!v8=N52DqpF2P^E)iGJ!(C^v1GXUWc@iY|7=Xy2dYsQyHEz$A0Q^Qy
zZG2p$&aaKN<0v&QXCpH7Uox*&eBx#pm+Hia=TxD*qI;6`P!Wr*65_PNOd)OO?WE)i
z(zzE2Bo_gyt=MhJNncQB?t~_2oZy>cf=kaf6=JvGYuj*RU~_^KCyL$P*LfiKSan>#
zh-7?Rae8Bo%f3%xkg1v>`K(5%@EAipUPA=@ON`Yf9!!zz-Q!gU8q2A*%{3w3Z~a*N
z1r*M7)$W~1e|=x~^)ks$a<q;QbFw;^UaYKk_hhLcI+*g8vUm7-9yutZ6Q6^wx^(uu
zHO?j^qa#;uRmc&2O4kAYocv$wrS{YPcY(<A7rEF`creYH2yUDtn1qCMF6hLr_GeEX
zi8Y;OQB0f0(>}s6Au0j^i=AEDIkC}+#ALF1X7Bib;zk(IkOEAWP2q!OJo`T8dXR`m
zn5zIWhW5B@K#90EmB&)RYPeYey9q7;rKKl=gwjOt#<iK%jzX;pSYQt4x3dGn`P<@v
z>A>@i#OOYdx}Wm;AF1S>7L&^yA#AOu^u(a_3;vBd-=2I)ci9k(L$mSAcg}saU(X|e
z+b%&B)v${rqW`-fqS~m+%Oq7EYH$Ktc$YJ3rvt<0V&pBA*8TuOn6BQqMAP}1kkD@f
zBb`4UWle^n2&m+*wzkf|C?Hnks*VGoE=#MDDzAsKCkCsG(?*3?{>6e=DdfH)cW#h{
zgUAm1rKKMqbuzKM_A>eoo|t<ZKQ`wjF}Gjr`C_V5?3)-r{FNc^JnEDcZn}H#(Z$RU
z6=5tN=KN^gBmy$Pra`F{ChKqMJIUPeByMDYy<iVsOmdz=A?XUkPF|cz_VM6aPk!q3
zJ_P{ScYurrrM&FBe42hyM9XC6xez%V#d|3<x2~zP$Ov-{SESC0PS0zaTv^1v%kn`x
za+I#@!Zq5XYBJW+1{vH!;drvdLUys>^1dRRN{}Hon$;+l#{b>?6F&3RXA8o)!mLi0
zR%z~@nTnqh8WPfUKLts4NV+q)@FPl_BWfUHfH=NWnmqHSG@Pq(Tr6BYuPBdCr;x9V
zJBsB%Ef26=ODUYav&7`hupM+5ApWM8s&S=_M}Y@HfOWZEKX@Y&EqLX7OwHp7;gEhv
zi*&=cp%9!Zb%~I;4$x^5rnS5zn#MAqN~BU{cCf7Ga?DxD1HXTfHn2ic=%>Zv)CZWU
zZK#&$xdPDNRDw7+51*s6LOHnu9;HG)gxy@uT?AzUK7kUerw5`Bn^s<F5lf3k8rbL`
zCO&?B4Vxg-b-g^^eqT)w!()Rk?r;fIrxCBrjCr-V!}40c+_fMF4{_mF+BDF(_Zu(#
zUKfz~|3ol=Y}=<{KT{H4|5dx+BqYagWa>%r+UmYq*dhNz@Fuc_o-jywK%a1&nL!8C
z<KD;RdE^t*Y&UgSplk25gyi>*)#Lp$ZNU)b7X{ikt126+-iNKPZqPxk{qg*_t~q%t
zr(eepc8P{i#?;Qk<r;$1HZ041q-;u65^_8BxYcct|4l~PpG}J)NQH!~rvn;XL$uC`
z$P|h5C=t+{gIrKJsB`5J;hWwE+B8~j@3hE*-S%HlTTKmpt!@LSsXSVDOv(1{$C8Z+
zk!39)Rlm|4g4@{^IM1#DHX?X5f&nC`rqeB2N?W|sTaB}U$5r83BpP<1V^Hw1N+kj~
zoFfaMDyZBn%8N?BT?;<lu#9Jsd0WkfXzcRf!x_rYq(AUZD8EC)jmF3#Dyef|`VHPD
zr_sK#X!*9Sk@SnxgIAbUk{{<0jaQ4gf9<c@yb<f`M<zYr?^~}+K9L6Lo2lZfkKde#
z=Q|&zz&3BMt>KUP;%de=-Ft0a-KL~lO=EzM^$fMuNUI)CKm&jf-^V--)3R;vY&c$L
z-(=WC%zn|7TE4?w&mVW1E4+wd`K*(4lvZc=Ao{`Tzf1Mj<KAA8>5C(cN#77@)M&sm
z(E<Pe7-cnjpiBn8RWtW(Jss1_Wg-#4LGi!8{KH3r38z>2sN%z5k5x2aYD+>zpnq=|
zlg0Y_lW~~DcJ?8Epvj18X2TC?oHNznz%0<cH0<iDv1oGG0Z9A3lw&~oMPTG21;`Yp
zzh(7dyua0T#UGnIQ@v8MKC8mcj!PT*TS0{C!sY}Bz4RySStqsW)-nS>EzkbD02&~?
zKi9A~>k|^1os%A(*%*)3X$U!!{N9k%{Brhvq_LDIlwu0qBY9`>$2BKQe1DtfgbEnm
zrJ;VV`IiF_Jw|;I3szTQw<8!Im_;UaB7(x{dRdjZguh^0QR_^3*;@1{0e1fCIi(23
zFs5>DyzZuxX({9~q2d!6LNTwMpft_O>*yrK1te`}u1$xV#B+)}WXKr!g$slO9jmXG
zDP7heuRu3*1IK%S+X3n+P$_}I;8?-7k_}b87-f;>3fMyCvG`34ogh#;w8N8i|3U-R
zbs!DTeps%hpXBtv>@g#&N=hbvH;DaEa-CTdfPD|#cgHqiO9C>a3X1u5Fi@Q9DO|dv
zk>1xgRv)6Pbd@4Zr?zs=TmSvft5KOdmPU9DH;c*Mk*lvZtaBM}RyK_7+7dHt)$&<3
zEepa}g})_`>P6m)bZKjTDEg$RO7qpV-ri*!t>huXz+LkQwh4>-<kk6YRUo}k>g^~#
z!C0z`)04`XIt`wlPtjt@11iu4!Wm;RJB7FV1xPOS4Mc-WYe@++Y0Ww3>{o&?UEB&i
zg%*`qYm~RcdSD7|poWwX!hs$w4IPtd?Vcw{p(Q3Vv92YoToPSc^UjJnMcF^%(QY-S
zdlg$<uJl7`=LQpRNmWdT){GWJM}TAsx^%PQpd}ZTT(ywRo*BVKeAEiVKNsfxC-~|Y
z;e)6A;K+HgBOSW$GP`z0>oZLlkRFl<QsFGZU!vV%#cx7mXA2p}eKw={eLus~kHF0>
zejM7U$1TN0cwe$EK{grm(+M0KOY%qCFxtE<G`n2LK6U_E+;4;yd+eYBf`BYny1|6`
zwHk_d@-O0WCK&~ryZr!U&Hxz_Ol3UR)$dXoQa~oIWTtp}ztk0CB}Vpf{&6D_>VG}-
z+NZTe(T{COGO<B0qcg}^!fg#{w{eQd2&3+hi)L!T%#g)w@7J@#+(8hIb~Bv3>hEO&
z7klX3&@CrpQ8j~g;585_vd0#;Z&wGMwRrt^LE>0+)n{$71byEzUHGD3?eH3Wr(8@~
z&Y%U*pc2uo%Vb4?&s?wkb4Em01<c(nbQ(~aZSB>N^`2E>5u*Hokg(DnRdWQr&_nyr
z%#ZgKROgHkp)3paO}5`%O0~tG5X138KkqBDE5Wvy+R+wfCH;^3Z#Oz9y7`XIEX>2H
zE>_lPzc$EP@$pRZovfz*VeBsRL{(DP{};sk116FQ5#U=;d(mysru6N!X^^^|Er(DG
z*G!M%l4(C9c2J}X2|&ieCy(r3U+rK^3j8uZ+dQIU)v8n)ph~CHwKB=^r#hO>%+B5K
z2K^#tUqv?ZQMfy`^PPnO^I^v!*{Bgo=QZ8Ivw9hzT@e=39R~)xaL}?b1EmK+ux%Il
zsoBxRmqCIN&_%XsfmQ>e6{UXtfimJZqqQ|lSTas$Da@T!$we-6W*mBXyHPw>E2O{1
z@ly*nRC!!VL#=gc30~r$35fr5lB3Dz%6F?MT~iA6B1@ob#>c%;9dn2XRB=?*7<=wt
z%7dgT&<|zpfVExwE4N6>N?i0tC8(&Ca}rJYMU>f8QNPklRMcJe`|!`<t~Xni*@QeB
zY6p%r!_ghEr|e762DKjMe{Qf!v4UtKW$qjy2w<Pv(abhwoRgg5(1yO8ui<P;ah`;i
zdvyZZ_IZ)1Izc6&qvmpIOFkjdX(`^%6;A1oSZ1`p=>BllYPK6UNL2Xu^|sgw$AKDC
zIWf%3fN31)cs$ycj&7%O9L8D^+TL&G7hfDnjCym-`Z^@sh*r{~1bK1djZ5`kU&dWi
zDvE#hUZzorA+ba9ib?bI#nwqKX_A1cQYSQ~V@t~N6uLCq#+yJ1yETGZZKpommt&xd
zRdnMi<sjevX6BD?98@MNynZBR5cj#@*v*#d%2nr{5sbcYT<P4lb%*RvDp$4y?Anuc
zA$tb52!8l|uQNU2>jj*&#@+pXk6dUTVv<nru=1_TcI+ZaSIq0PWXBeV#8-CcR!iAA
z*im!VAN9TVWx3pxjqr^#-+q)&S6;3{ZZm9kFb%s(<yZCHD&aL&KSAfJ4Em$B68qnm
z$YvRfwDMts_v2m|2tMt8@$iK)x#0Ts&p}`8DWX|XEG}Xc`Y}3}LpBAdSDp()P`Qz!
z%yz1CIn?GWd>2{oA|R{AJ_~HBA&D=|_qR3GyziH+No$eqpl)i(gt~rbueVB@{b}e2
z!v&41=vEJCQ18&2(`VBBOo{snT0M|3N$zW@(Dn}~elsC2+}rmJ=bJ_V70@czEsy|v
z`mGEX*6M5vY3yQ*6v#AME8>P3)X2wNff2mZMnOW`Y=8%>mc02UJN-gUUeRR2hP1}m
zJTsLtgdAR|xA)%#npT+q3Dliz5#sFg+vi;>Gn=Xo*cCViBJu~K!KMrRG|)pe&zlRQ
zILehT)XswJbfzi|Sv|JQ2#%*kb>5b}+kIw!c%Zk-rfbLkB|!Qm_6g^{`;2leBQ|=>
z>azyn%iFhuijCf_<BiLCC9t|%F+O@~QzL$F$D;e1ZVVj{vj%LQG)n#(_FDQEQd`pI
z!m^0411$SuVN1dP$o53%#hv-)KULIPRU*Fc-it0QX5Z+4_(o!VWIHA2`QJ*j2VY57
zwXAJI@99|;#@i%{2U9#VCXQ~z*eefZO-{<b#TtjQaTgo(Axbc2M@pCKT4#mHwx$;f
z@LHTj+cV?B#;8ht34KP4B0ef9^9R=H4)3c-jX#XSAAC{{HwX7k=TB3LpIyV$W_<JB
zf)dKueF{%<<)&k<55<yQQw!cD@ZZtZb{znxRgv}V_q&nIAwAJzXG(loODb1B5kmUc
z1LJzFZbc$$Fh;M*BiTNl70>SVHERV`bNnWmFxu-Btv9ElZ?)Pv<?FnwsgNq(N2-BE
zMHt_eCFHyO;x+g_-z=50C=fY7wP)UxTg^$wR~w)}{nrV65oUFIG)IcH42Zu<nu_s!
zacN|pK;y_n<5y*|A119*!X9V}6cBnZX=fYaQz=^4OO#@nhENasQ^8K^RfRLl0XTB%
z9Kbzb;X=l@tx6i4-N8YW)9Y=Woz`-uJX5F|1JMdUQV^%GXyOH%Xj1<cYJao7xUq{Z
z@DE*p399H4JsNndw~;hr=XwbsuQRiGXR}exq6w`A^d}haP!#$gLRKXl3{%-z<cURq
zgMH*q7>!JLuYp;XUOcpeJvq_eJh=4X*S=bcLZL><uDJHAN%j7HNfFP6dS8hpD(OUe
zTw;TTw;B%#FHI!LKVG?9+ng<2deye>h-Wl5Sf^$0G+6@+|0O#}#DA6vlHfGK)ZXW3
zb7J}lczrioLYd*YuSXADb{}<yNV0RMSeWV+=z4@GL^C5Rp{HMdhO$pS>C(q=1&l^X
zAM09OW2BoNJd1n#g|WY2AX!S;2w~2L$4Y3#dR?u1R{PX2W`g}yAz59p-B?lggd%k)
zxMEg#=}USrQ~EP0=&k&ZnfnRVY|E{4c@ZVIkRCd2k<WcB5G(v`Ay+~IBp0N+3wq<$
z<`R_iZ$|^#!ON(WORa{wcLNSA1=QIhIgLN^Z1J2FZy_(Yt#|IdKGFM<?^S^E2mUar
zAI&a4Rz8Ty#xb8b8Wtq=t8)D*sT2m4Z5dii40Dq||K3Y+5#o|RS+aowz^2_y$uT*t
zIgK`=ktx3m_0Q6oXKXx9xdP8arhlTRW#P|`)P|&NZ!`v#21PN3a)(&0lB_mIF|~OF
za8E}HlXM->J*$MuojxCf^#hYc9*(i70+R4go$K6V)OnhT$rlAil1RzT#zRMtDlC-+
zCUxy$nGUuBfU2W#znP!WEt(9S_5~_wHU;bQG~B?ashk<9C3!X&e_(l+ub44AuKJnH
zaqt>`s6U5g$2EG$oD)=}={Wrk5+VUGq40S_?o;vg==X7G=gf2w_bwmo5L_?0T$@-S
zAtBXd`&OjsH=tQ7C!t7iyC{S0th}4)z<&bb76PTE^fH8TqIvp&8azUe=v;c&tp{Yk
zm^yG6WsSt^JkRHxS=+W4?Kai&c=ZuRkk!r4?mT-(($91)!!?iJLBhLVeTDK;zlgJ+
z*~824AceYaE7+>e(~zc0N%8jC>(7{6O7amuk2$m6@0Bn6@v=1y%9_rM@psq#rPT{v
z>3ivJT;P?CeO7jg^bFb5dSG6;J;zIq9ZNN>8hkn%4{dvv?rE@oId|u=IY)bf8GY^>
z-IlI_YgH3R()}#>nb<V(x)Rsc{7+QX)m^rF@7(Usf8nb5?}E<VoQe8TL2;u-zDt|g
zYWt_TvJ}nKiTJ)saXs7o`$?f`q;KLU4y;0wC3j}BND7tk^5T8fh9_%?R)3Ocy{B6M
zg8P@Cex#^}t5xe=+{6T{s8Etn-%xnXods|XynOjG-?KFPE^hHo^LK=VrjzWneq;Rz
zMQx5-_XWN=_DJ<In?Q1?ygnnoyXp%FVenqVoS$x+-+e<jm0A*V7T>mECuxF5>a;ri
zPW**LP@EeRRHSEq6qi*ymxA1r#M$1A!~l?{qI6d&H{y%Vch^dI;%Sk_rcQEPT#jB*
z3J2iMH`E~xJZzYb)Z(IHohli<I;?X7B)7)$|B-7meME?*CEaw;0k*&r!NnNe{o5aI
z7>q-w;UmbIys@i)h`rm#O5Z6NoAAhEL^}f_XHIUhuHW~wbOtsXfR|HW&f-3X*niGy
zbt_@`#jUsNK7m*xI%xz3XsJ@$xJrRx+Ep7oMt$0s<^Oj?b@7zx%!TtepBBsiTTz&`
za@K>w*It1F+<Eeq9sjEm(7(mSKhzNOlzEvXh`slUoMN6zqxr9UD{+*{ZAr2#sr!W$
z6`zVJXNBtAxp|zvTnpiOaH3mWHlBWApH6Dn5bL(O`GxZsQPxYn@Uy(UO<HoKL7DSQ
zg`Zl=%E_F#zv00hy+`6AIt;OH_HPUn?SeeS-%8^$-+!7(`1Mb_!e8p9i3yzDwb?42
z>V6xuuZ>*Wxs}}7$wL*KIg*{urSgd_!StI>uFLjV)&8;ZJq_2NMdtrv)JI=1Y4Hdx
zNAWKx&iS^FUiB4ii>A>skZ$YuWJWJ?T8FWs)-70-ZhTx1^lcC`yB($_<V_q()0e&i
zx&22@pj9tj{fJGfpjoLSP)H(<t$i$BL-<>&TI>^^zb(U@+2W^75tMPiGAC-!f!|Td
zV)Ua3lNIL@BT3w8dV*lJ5kF7M^((Qg&MTMnKP&hw&Q!RDVD??RXRUc*PDL=8?~G!$
zM$snG7Noj;*4>2d=oCMR{#&goZ+lci`L2k=*Jb>RHcZD1c`&{sF~+>2i?LDA@=K&b
z5<jo7@Bmx8%!5F87ttD(3D)~&)<&2|^4`^zT-o!tF^W~!CK2dp_R>4{pU8@v{9A3u
zYTr{pK_1Mp$$hYWfJ~j(LT|o_n<S6>jQ;!QY}#~QqNR-QP_S+L+E&KO8JYK=J4mY9
z;!@>XJ;*3&TcUz}6Hq@1lb~xAoCaGgo1oaJbB9G|L+GssC-wZMQ6S#prSYzm0X#ok
zQ8H&A63uEt)Aj_P{L+!uaPDS*xVCFAZY8de{thQGNo;@WFOW3O&Gn8SQ-Ep#9JHWY
zplB2g!kC7Kel1?}Yw0}@JT)uCaIJImeLD1IVd6KZ@=GfG;mrII5J2h;VU;(#NRp^w
znAh=@ApYk4ikp7CF0)o00EZPZT{}h%sE;hSMr@m%7V}V~{9brlk9^gJ80V^aKjd+@
zbEY!Nn-`r|VLMh_tT(Nh$UG8ge*9|g(O5Cf?iDub?4jOd>io7Ft~mXrHer8c+Wklr
z>TAhO*Y!1T`FZ>coCRyd%e#&WGzD@&>sIBRGJ8Dq{XD5}!(&+DBA$~<`a6dkBQNd-
zci!bui>cE!ZrKZC%3Xe4`Uq{MHJ<;;J}Pa=PopVNYgOpSXv6A<&+n{jeO7S;AA4+~
zn65kxikJ=U{I0?-zlX(N>3mKWc|9A&U@Z{o)$)CP^a!fV&tLllJr&HP;W04t?n6OX
z57&kpQ=^IO09#S1kc>8)ajQ&=;{&g$6a6InJb?;X;$tm&iwzd2-n$&7;LXRx=o_el
zJIs4!Pd1e>c_B8gw&)GdMD*M#c9dv(k=&bnCyHY%H?m<`cp(S&;RrJ0J#Ur|?@r@7
z<rt**SuK<7d#n<qO^8k#1rv0%Vp%`OtnLsg!PXzeYfN$d(L3A>9*o*g8;sH;@k~RB
z?V0;CANM`Rb1{(&)pQR04AhY{?nrUxG^0GZy*RKe^eVk`GxaLC%Ym&;6Y%Ijr!T1o
z`Q{xC0QN^BeNvZTBdwYN@5MQ%p(t$6IS7Qs6t>M#LeTV!Pj`(7hm>&W0Tzj2rO|<w
z*Yn=;kCpNxC{6xABy!}>qedQ_0KYOmAOTl<l_E3k9m=n*vlK<W$QVLWEptd5{;*ez
z_E8{?JDLjKQ~7*HoBq))U(V`(m-kg!Jrht-iQyb+!@G%Mh1DTXT%O1u$k^@<uX{&b
zvDXp~kt@*hTWkMopZ0e{(UqpxD=Tw(^T{ogiqT(wwWwHjp`CU;Y^<UB@T_JX^~B*3
z`{E^HfbQz`0D=3%J!~$s*aAB7DO3;LxTY(XuBP5ctzd(*foEXTu2rXvWx+U91imLp
z4uf1{O|4x>lf?HLygl?wS5_m};e*zCw%4XULdR}vCVw3ZjypkpC-v|qf9sXF*m!N}
zm~jpHF8Ju$<<f!KRzGqjWMYgeXZ>BTg!8cz!&-&tl3ayj$$<bjH0$j7H#xkbvxhPb
zMQ|wTL>P$MPFTS~T;=}Zl3G(spQrHo!UO5P?hMzCMt!BQ4@RdofwqMDJrde@AVL;k
zvj$*f`5mTkCO%Oo9in_>i8X{MwU)i=CXv4AsDTGdKqEyz+*TpkQHa9dA454r<a?HC
zP)E~~B3Px&BKUiUSvSE)XB%Ac;q7@w$NZ#$6qlffa|ul9(qS>!oG4LY$8U}a+5;dd
zzrrX&+Lul+mAxO{2zA-Sb0?wf{>)3Pa|LKFPCthF5a&n&!ocU=OK;xTq`|2L*#{FQ
z9+KktXvMxX_E=p~XuRHCxEWFENGIGj_+TnyVQVH^z<?g{U4e$9<BqTub0cS^xP*^v
zYFUS;+hB668+Xlph@MsOKwsJI_s0?1J<pnvv!iy-W};~UrC#tSUZn*|t@Ha&n&}zC
zAA%)UJ!!E?+<TNN%Q&=~o?EN7{rTI+6>xXjhgfx0&RQ$C0iHlsdYs{%o3FOgO?{Q3
z#l;?P%>@cod$jFqx6=6jLT@XD6wmx&tWsR|dG;|FQ*)Vk%+=ggMN^sf!eCc{v$8Oj
z>M_rlTKDx-`*;luONg=nb8s)7uWqVp(-T!AvR1E{r;>;3*I5cNY3jFWN2U99XDfFn
zS?^eC=e?7j`OEu<%@prcX6w5ACGO1)qJ*0Kks@Z^mZTIdcQ)Mol?PEw)84@;7Vq1G
za!vY9<6)ZmB1wCk!msp5^1<X&X`0Y{{ZF_NN!yX+Z^M+16N$AynG?8ZYv~Gi3;K@l
zA}>%wZN)G6Gx~-79g%{mtyrVq@V$rT^0##L_Cx#Gda^e-Vb!DW-Uwy0FO$6b*cNo{
zR%bqtr0c6LcY+*4-(&y8W!?zEvD4hgr#VT{d1gUM+k8j(9KMEc9P4F<s}v5Li$+E?
zvNg01bDf#~lyyBCNVwz=xI@$*e8kPi;T`TezDRBMl5W?ZV8~9cJT2nCqA|7*XNbsH
zEGucyJ>lYcK^v)CC|_iABoHXHE_?9M>Eef{3Khf%FYH5w5stO=n;%Sl^iqPHFw&!H
zTok~z1o8n{H9gqp+a2P>(%QVpq$756^00tn{-GQ$ftZfvrSF86iRg`s#yMt5nXU1<
zl_-%zMS_{jr%OY`+N0f;4*5GEy2{eAYLF6D+Of^##xO`p8f&B^|9L(_dNa-rY&J{f
z&700quU_EF3oH|5{w29X*EVYhEhb$5yP%Q07*w;E!goznSV<5%u-2WbI9pl5B2-87
zY3XcmF|O^u3tKOGek0w8Pj#8&$#O22YAXj{<0m3HdjfJ(&(e>$6bq-_bS1wY1DCO`
zfmX1!tC@m8nMvIx<4n`sk2jD_aQRM9FdtLl`O)(0Y%NiYA)49kb|G~!UbNb<H6Z+>
z#9gvu!5VsuVQP%0H|0XZt;a@x+$nS{U5Fm8<4R#BONg(E&|~hb<Ez_OX)zY@_4zuV
z{WWVnd~b+E^z}<4wBPfM^lqJ^3J3I^;90Cqy5HL1&f;=tLnSLsfpnpEmR5(lgX^C<
zif*3xnR6eGF%X6ZjDu$pzT)-IeFWB`x+^iqkj5JCwuU~RZl>&u$!qc6StmEqbUKM_
zw%6Y{GClNv+)H+j-|m_v^qSh1Q5-oFqvdn6Al@OPup1zHH1nlTa_ab(e0@qX<F^Y{
z^P)8(ch?^vX-=3q7niggduhlCLSOhH^q<f|`65i84wt`foH6)XALl`r|K7}b*MU*h
z9=g5F`_F$Dv{e&)rv+4TX>lQNH<Rf$f=R1U>!g!t_naU(jCK908eXlbzm%k*V~9ZD
zCeai@em>m@NQo0GfiIwnV|MjKclzlO6yx=!XeezJuMDQSPK-skxLtW3b=U!~Y{xDf
zS7h2Cuju@51t0FJYffoq#ZFcT+%x0vxrJ6xErkqUn~o(Os}JP}>2Q9c(Ruhlff)5X
zqX>e8w^#KOlOv+HC>%razKaLZV_Qu#re(aWp<DqIi{aRJg(cdH#X?~M+pJ)&2OS4z
zeRXyg_`7iSS*Ik4#$ymqh!Kzk6<=<!2b7QnOg>uhb%d31_FWisZX1{N)xEuhKCu+w
zHp$cKz78?`A{7`Y;Wc{QzH1Rvx0c`?^Ird|J%2Kdo0!HOC1w}<vr3xHC4ZuY-x0zS
zra`<bM^wub>mEK$8E1#wie*%J0E<YGR4|g1Z9@|D{!Gk1ytapQ^>L_d(0w@TDdh(h
zmkW(ix%>6#`p@8sM}gPg=*zFH3>~<B({HV#$Oejx9P<ot`^sC<T%VuzX4xNqxXzc7
zri)EG%bj_I>T5Z@k`-%9a<6H#;t{YeKsI@~i=P%G3~_8t6O!lz^bg#E_V$7t;3ueM
z?j(gh`2cP$J@2|QH0~RkgJ=@z!FzAP*{$@h0_$Y+^l1e5-CrEXJkwM^j+~c7%j&N3
zKaD$<`B1YSacv_41XQ&@`pK97^dw-vQ8Xt-J~~Hm8u*cy_WRxe+2FEWk5%R4R{c}@
z)f&kl*~SK?d3w7t4r-D8^aFl21I?4GR_59@Jo<`O`Yx+R*VemqssfjgLNYH5rLtT;
ze80>Wi}p$~c2FXVfcm_|G#O5?2|+JE+pi;LjS_bKXN^IB0NeFAD0&oqKwe=5x)ZP|
z)gei$aq-KKdeC=$xLG^kmvkf>!Bn}!%kfL|8?sDliX0l>>e?kQ_Cd=$wNtR`58TDQ
zD_sscIIo?U1_3@%cXtmxZIPe~F<l%ndxsa{y8IV1;9)Of?;(ArEKQ}x*LeK07|S&3
z`CC9EP^~wu<&@!NZ?i;TIs%SwZR-y*b#ofqZlp-x>fvjY!wV|$l>+*c4{5#{0G*EJ
zAlhoY`JjBxPYs;mj^MbDyb{g4Zbl8MGkV)Yt2%rTfvo^p+_|j(&mxx6{)9=$B93AE
zR`|>3;Tu8zODNEyWHi2WdS)!PbaZvKKld(PM^Sg4<5^oI^-#4;5&P8vmkxOSTvXzJ
z7g{f=#rx6ELH-$ikmSOe5>tz;S1V{DB>IXWecTSmeZK#AYw|_6VIC%|n+*m7ZKre-
zExg&K?aqr>DYF9)`6{vyMv^@%cJVBks8F_G`N^$O6|q0tURf+`CW|o|N@N*c%c}$x
zroP>qn1^q4hOccF#7A(!oE$&1=&$$HGH+>=usI(jh35N`WLBcBqDK~Y^P(rBN+V+L
zW>xJA`?^TqS#F9Ib^FvA#XyF!ki%8Xp17Z2Li%>OcjRoYeqIo@$YC>1MAd$;lnD~~
zvUjQRi$PRy=91#z$FAbfGGAWyOF@rcgx8M*_YbUytm3K#<CXY7@J#p<B7QtiS$HK<
zfVao>wtk+*Zv>pEM*43c*?a3_ImT#x{F~*z;?(b4TS<Zt5qIme?YS_5{s$7%{kKvK
z@AIoV>FDbVD?^JOJsFHs;2zc95OcNn^ABp%4&3G-;3*vaFRXX^N<Jairk_fjY2%Y$
zkpz@wW`jmRxx^B@+r*BgW4FDUkNCNYaK}3P^2F))=|lX3WVbk26u;8i^njBse=PTD
zu1p1?;wDJhp@~rtiAmoBNU3w_VWyqMxBG$pRRP74r&$`CxIMv9GW=2(J3i^L*qq!J
zQ&)kXdC0V!>a9KO%Tc|Gxf~-h0{7-TGxD5vl<w-Wv=2B|I_qxA;A3eV3FrTu6PCES
zNO7SUQ}a$~t-G75!yyqY`oP6;gH!wiIC;u&XM!DfWEQZ9s6r%&)={)YRE|kI#esB!
z{qf`C4kb{=ig|VBrX7y<<K}xZQM^>s2&dh%Th=jFGsEt3`@Y7Ci>2nK4fcLOGmYzh
zEy=$>xatrHIZ#$sgr)^ZCMlkzyCx5`oLpnUKe>}2bIs$w3zz;GlwJ!MyF(0+WS}+-
zS~e~b3Ul#`fTHJQ^r7!Q@Rc`g2VlJK$IcO?wLd0T@_s4uJCuHpY{z4)CP1TH@$m-x
zYhO#wKbUtz`h$JPN7sHM%8E<$)h7=W?~tqz>w?~PBK)NR+}qpom`ANr-}soLFX@Aj
z<NVADEB7Fl{jpz=MaGg>znVysrJrNf3A#9j$??0O%H8fdl)QOZK#2ldI}Ufvn#0X?
zi+aUpJnHK6;JUS^t@ca-$rGc;&#_4h>Ui18nyX0-$b{s!Q$(O?_+qm9g?OQ#*Mgrl
z8jhb$#q5Atq{efNXD2peKT^{e-f&D@T}l+Z097I0{C5tCGm2Zd|F_hteJ!N}JjwJH
zzgwrubdvKsrsV1~@qU;!Ke?ea&i7kWCINRo`?|A`5T$&7uV_$Fkiw@--?z*RE5xNT
z+jQrEnq!~0^hFWEq`0-h`AAi8s^o^;OfNmT!Y<R=)9||iz?|PZh)#GRGO>HlDegze
z9oTryt#;!+ow_qv&kF~?-qE?!JRgh-3v{=ytT{|(ak^>bzYCtRhbkG4$l0~%#I^Uy
zhLmli99{CUdh@GM_Z^ZzxSHWQ<F5&c5DRR(@n_XR_^sbPP;h4jUw@MVLT%E}Rk8gD
z?E5&B1HBaEc{DcjK(L1`oTJ|v=I$O<dU!L8ZtahU!}w^a+)vHT(JpA&kWdaa0pxq3
zus`KJW;~f8jUBEkpa>`<fEQFlA7~Oa*PxmebDZQjh>L5)p|C%r=6rB(^u4i>!irB1
zXZFnJ+!<^d?~QL4z~7WxNSWz%C^pK;CGIdeTtTzW3?pvc!Uku*Il8_%!BolsX=t%1
zY=|wy`jm}7eAWJ4_s6)1(zA<`CM8@J&Qi}i{3w<meBa(|@ZT_z-KJd*gCq(M`ZSH*
zS+l%~<kOXyx{6Qqyu|%m_f3DyWJLG%P_E!x{9}@<f7r4~H|=9}HQC%T8#0gX2%XR`
zGOew&w(QDezP&9Yh?kL^Op3}(sxx$X<oy#@nct#)Ra#)V0iIR0!gyU`f{jxQokyFH
z*x|BZxHogC^zLyw-=e%z@mN!^OsZeQ=E#6g_iXj%S7rnBLtm1s=$s8rZc6PD${${`
zFGtETM|S8G?GyFQW4F2TcX!iC`;OMD)F9b1m|lSOo7Y>p=hj9YbU=5=)5<Uw=P;jT
zL$KzTKW5x?6I&nTYS}7%oTQv5(5-PFSF_F$FEZUQqPOJQ(lD8+`uET_f^W9HLa;gP
z`Jw{%p<rbYP3vQ)@tz#&ommDn5}x}s#s&(8ydWkx7m0+v?B*at8vq~|+ka9783;9t
zZ~T80QwqgV1RaFx)Jl#sPdb(^WD~@`1Ht<w@SXU7(;1-aYS_tZaq__Ce;SkDWGi-Y
zcj%Prb_;a6hp-=>#hszQW=@7BT2HZ0==7S3WKs~Rar1Eo>(R}}sT>kOSqx65O0#CB
zTDic=7mIWDY*52!UIdT1o1Hgf<d3!+37SAY(S&Ca1XAWuR$coRK)U46<j`>hp=~oW
zBl09UoFy!23%i)M(gHNKtkjXh5)wQfMya5+euW7HEHiIM<RGuk8nLP~zRs8$5|%EG
z$6?3Uz8e?aoy)n*=pZf7ru~`u;yyH0N_S|jolvf1>iO(;1jRUufxbX8es5L5;(?Kp
zFjY;xKv<KpmvXXH8{=m)YW8QMs0r$-MnPjsicnZ^ilC3wu^XG-M0YY-?OV`77-yNj
zHO`e}u(y~!)f}uTvZCRhl-v_`9c0F|eL5H0uFt&-{dp1ndeW`RPDjdbN;8G`e$n$m
z@sbG6n}51GG~C5LbxGAn_Far{Q=JMN<rf#74>p%W(4<W?--nIgP--KDYqD(lnl+?0
z2yM!>Ta0WN3v@-pN%!y0BIA15Nl*p-)}8$)@o@{sm?WzOhun*)MWac>Z{N6E3LC5{
zmt${!3h&v!#q{(cQE}7HC~3oNjA9Lz&7js>mZ1riEE@PG1kXB#;`Y?hYS>fWD|e-d
zx~&6+0)aE#i$q-KwHPR%rt^S7*6>`8Aw38*$lV7p9+u=LZukN^!^|-YA60a8Gm&%6
z7gXqY<H@F8zXiwBZy)E)<Q_FFY_%e1c~CY%r#W9V+@D%ZQ#b=%x*#$Uf^R6Obq!q=
zt@jB@ezP#ghUTu*M29lqS|?p8UJI6!;&Ba`J`&tJxwd+E?aRGxHkHiUL4uD(Ljulz
z5e|5$5)$g<a@J}X$odnD^ia*wDP+(u(QySp4FLkk@}d98_le?gxML_rMX*so5&(vr
zIu^X?_P}d(9Mqm}&Je;`Vz9?2oJBv{-gVDk`;Mz6yHcbuHorCkKakRI1@qtPPI?%r
zJP&!bFlGMZ`3u&(Cku1;yI&AKs8krrryb4`|L3k``S5!;(}x#p9@?v$-vycy*{Aq~
z+BrAe3z^~dY3fQ)Zj0b55uBy#oWYAnCF_K@p$FbgqVE}p6z;9|qt@)bpQzJ)P!x{G
z%m+cQ!2PIwP1Hla#_XMODD$6PK{8Q%62=m|oi4Y&K$!MVb@VCsI9=h1q^)jUKA!dO
zy__^QN%!ge(`4ue2kxgY4E(m#2AJQkdY-2IPam(LLjod?JYwp;G{COKmET0AjV%^7
zJJG%(OP9UtHEpW|#&=MQzHh8Pn>HCXA3g_mMDR0aEs9vQKo80x)j`8;N1Nc*leaD1
z6rJ?vNcz7ElI)Alp0Otd;MgzN)u`HsP(ix?gbrizryT3#us*l1dR^17Z|OMU9qji*
zZWD`R;c;!}vm4&nc`@QJcucI;&9r057D1<p1t{u64WpvNV{wp(OivLbDi7=UxvbJW
zb(OL`WhFXN0~O~b#z?PK5T{^R>!qKh$kohk8%&}zZHE4T9b{$(R^B=5U@||K**O6D
zc2P_El8Uyg{2L!N{SC!STVhjOky%W}tzDGn{PzR%h6>DOZebLe*M9W0`KkQmxH(ZZ
zPjIw;){o{_ac8~v=-o0a%f6YPP$dHCV;WG?21R2Ohv)6QP**G08*eBc@X7~Q+9T=G
z5dreJn%Nw0d*Yy#r+(?}kA^Ya;@BmZDbFl62L_q3QV5j}+g%Z?k-TlVCdx-4U#a!)
zrLVZJL0R?-G?`Ja1Jy=LpRS!TyF{)Ocz57PXV<5k(jA*0txu1gUET|CD#q<U$Q-G)
z=|YGY$bWG}35H{BNiwZH3qy5x$l<%gKC$MSomr3Y@4TOw@tDT_Z^Q}Z$b2bKFnuIT
z6<P01cTP7pTqE1b0#^YjnZ~HV_ML@7G*#ObNT)>?KH-j$u|v}kIeaql7r`;6Pc}52
zb;C6mADEK2Z&JFghHM(|3x_`FH3}TfYDE07TuggjLNrCBhDs#xL~{_U#BQL;|7$`1
z5$rk54f*c^!djY;u-Z>|aZFsI`IP<0xwN2;xO?X5R+xtKTViX!{)=@eu7i)}^y-1@
zPevOLXI-tOij!+!Dh@>KUmq*KHVr{I;hEkhOdO!E)B9}2`|<NVCz>F4=?cmwr$fQ$
zKZiCk0L7OlbP!q@-UGCofEM?5tO%ziNPdufUyBuFa6ldt{7H|W53V+f6V8`~s1$b2
zPXN+nN~KP|8Ahw7x6S*C3}esZcH>6ybLH9|P#Vgt$w>Z4)S$j=-2#ZA*393HVC&i#
z>ShvN63|AQ*q}o+w>J8}*sJFep<}xDcHx^3C8mXc2`FLy_w}KfzP{*-=48XmJISJ3
zJ_$ID%*s`_;V1E@Om76Xu=g9WK?(y*{dJcN59B3TDxBqmV?CZ<{Hnw@NNmhO|Le6x
z7_m0^gew;ps@Ix-N0OQGXqZ~M2OsM+eu5T{<Zqs7+FejMrL_{a5Hv%{7t4sp3E{Sq
zU8a3Y4cccthORp{T((^sA6I&yUy`6J%(qM9&E{<>GoG4mwTGnH*KT3<<)0o;B}WiI
zZu|4DLLslsVeN*^ZBhKJXd=_StUYua)s%+o8!wQ+oJM%eGF+GQGBO+;X7rTws$#&1
zb<aV6%zbVy$34Gz-=%U%ff{CSnE&l9e>6L&A-#%`*Zd|EXsT^B?(C*Du0A}+cRy(5
z2+*=(j%iw5+sx#=`rzp3S@+^iXF#t3G0uMin;?fiY5!fotM|E^u-?ef4~5TsY(9c7
zn+~{x@si~AHZLM%j19ydK>a!`DMG}?Q~GQ;9FDSh;=648ls3HSLvId|;LW~~y!&%z
zb{FJHx(P)b$84c2p!5i&zRzs*;!NFOxZ^w#dpQYkpMxr*eb;u+v~47FeTG7wlA&KI
zh4=2kN%m*3wlkPg!xNdD5NRm!U&D3QJ1`R8V^!zb8*(3uD%9x-Z+lS3o~?0l0<<>=
zOWtILy-z;kCWH3Xm*Kysc@!aS?ysb2;5)04-ikvY|9>$p0LBFzBS~OKq&HT|dD{|>
zC`vI)<u8}HH(zs2In`oyxTr{%*ZTP$|7VlttR)v)C-Ebh+Y;3@&8BYANfQm5iV1@~
zBfdSqPjfr`ux|hJ6Su%tx#d_N-ak$8q^?>9zd^~<4N=uPqAQa{Eq%uc>_L)s>&Sil
z0-vV_QN`yQvV^ND{uo*?=6Rr)U<E<xFuw2*rosqFS^tjHq+@9-9gC#BGSwXl`y61?
zuMgp6NPKM|nV`1dzxW=5sWtSxTsm&vB)&A6lEqB@{;I*$WU)QE<N?&DVf<P~k}2Bz
zv2@IA!Br$f+3Nab6Z!T9QZM7uiAl5#xEc*rmVfmwpGvK?Du(%TgJF>Nl5gb%dOWg<
za-*ePe3KZIJtlSI#dW$DiT;f~h~H;goru1Y@2B<!=^EQSjd$5+dKorawn>UVBz#Zk
z_cUOK&dnCV9Af0BcW=xe1o9E(e;tVk6Z>LJk%WY)BW<S}$L=8(Bd6E>*BLs_lK#6u
zN#6fNvVVG7SUg#)1uev^7hOIWUI^9FZ-l`};q2gPAyF#Z1@G&Jjah@sf6&-L)K8=H
z5Iiiz#<OEFl$iybfAv;x71>^Xb^|5kgE;~MaIim?;wP*174msF8`a{D3>U|o-S;&-
z=LOf(Aw<))eb0=<F`LJss2~IUHHsTSmG%$&0erEr)eX;~Loya~rU}>RHv~gcsXRdK
zHU=bjK|WCsU;skI%(oTp6vVU+!ks0~v1phGzuaCD&S#$dq{~>XZzbp7Q5cL1jnatG
zbUmgtR5dHTlY4-v_<G9xB&&X%eD)kql;^kcS{~b#JJa<Q&YY&O=;*zP+6)&jk=Qi<
zzC#X_5p;0}v5w*ruC?#S^gEH43ddHSD$X>oV<~scThIO7s>l)_7_$@yF>pm=>4t**
zG%M3<za<70euhEn_<IX>9^Z*}mv4uZfnu5Cz<!`F#X|o0Y^1V*-C+M#OQ$V^aDrNj
zZg8niA6n3LvBpe>0kVDu6`UpdwB+?|HavP?w{JRlHdydpgm(}jsS{7W)$Z@<x2gKq
zkVY{ReWVzVD!6h*-e9eI8{u0vOK<xgr)gbeI5Id(l~gRy*KjI`&X@5$<h@IpSsyy)
z`(N$q?0*+n;@_Vp`5()5=zwtgKN2AxQb<=_@X#jFw4UmGJ&OA_9>qJV#rA(3oqIgf
z{r|@MDsqfDpT(TTB2f&t!<<<p3R%pt916Fbca!rl#}wf<$C8{+i8-Ic=&(8D&}3?a
z)*NQLf1iH)+v9=9KHK~CeqGn~yu?qG+UW{idu>uYxum>UfwjC?nZpFuT^Du_Hb7P{
zoy2$*3gWm=|4a|5#KRiSjEszwhDFnDX)h?Q0>K&{$%XgV8Htu@iHrtOquJ?}r7ap}
zlvzcb&vrZ-AD>44{W(~9_0m}GOvYQ-UeZZQjF=dAnz{D)tfx|yVwy@gj$^OPs^Qaf
zE$<{n;9%m|t3DOD4b%-dX_KjH*j_P8W(<gU&YW9F1m=UvN+$uEQ1Q?aq`Y?N>7DB_
z>>ndu#olBsP3|}zlzOw3m*4e}CZz-HRE)iAN<*a59~hAACk5()i9RJ_Jt&i9vvmlu
z334UW{NDELiE@_W&By#6b@(9#g=Zu07fy7OA7v{Vyj^<Kl=<8X`S)zRt%UNkXQSay
zDK>OuLMh?vgl|*HqZq-C(Mm?T$IYzq)29Od?sO$}l97qKqR6AKp{_S}{-;D9OY~PF
zj(5W^&(o!^2IeF2HvfD}Pm}(H0g@~gGJ`1JP4Y9*An~yoHd_V3Jf0t8?|Hx4?8Qi&
zenRgbgdLD#$Zc<%PlaD^ZO%>|<H#*Pf^29$;z~)rS!sZ%Yk6zyl(!D^D70lH<yj-`
z3E`21cTr%164`&QZb~)esRD<~y^YJx*D9lYU~3Eag*Qt2^BBSEeFA$v-+43~`Q57e
zQi&Pek~|X4|9X<un>t*`{K1e0gQIdkB%(PCOo=61m`DIbGMg{B)=37sn>HC_6*_yC
zfo{@R1+vNwxjK$DtN)he;CWn>v*$ieUf$G}HYqJrgq+pj+?GzQsjbNXiW!bgoxcOU
z!+S9Zvxh({0A15VCXgyn0<Tn@#p{O<MNX3_#5KQkvY|XiOFzdWY}}ygM}-TMQ>gz<
z9^D`p{@<BA;OwlRX+d-GQgVHloxNwhy}X#>QQ~(irE(*jZfybX&HPK5Zzd_30GcGt
zB*5B{$-xu^%No6}hV9+Gm_4f{ehhsbFbenlKzSobm(Vi}QQ<A{83)|1JEe`svh6j)
zKRv>|c}8;{-sw0xn!W&6A5Y5QeJ7Sbo|q16x$$?TRZ~gL)-~az#RnpK+PSr@>IF4=
z4PFcP<TI^UUZiQwuk*$!q`ezU)-$UcNY)=vT;j+Yd8HPdvvPC~hHGZg5lqic&j@_h
zR8^F?c>R`EnC6|Pmj+fV$qPj*goCONuTNy|%Xip%^fb{-GWzo*T}l@o=}_dfJNfe9
zAbD9idHGAtT>PZOPf_Ru9Z_LdB8LDa$1RhlSN^5?mzxfQCvLj$JhucPy5dxH^3fh?
z3y3Oe7@e0<l#~3PZMO(P%ZzT%evLNEg4R?Dw`*2=nt#=8+01D0j~Wi<G;zO&WUYJf
z&&QW9?#+z6qad*z4<0Z$PYd8%=F1{1w;uqLLP9XY{BuPocNo#jI!n&N)u_MiYeXu7
zp12}M*E2JXI59i)jpC)jHp;PXC_xhfq)kd0Co>Q~T=LXm6M!BAvxbW@T-nTw_a8>?
zCXc=vH3~b$7%aJNR~jenN1jc=!GILDI_?0<OI)E#7p}$0hP6`d4(KAh6vc<*+oeB`
zWso$}Tws3j`31#BN?BCHedGARo3jz{a^eQCgO0Y$`tMBD9wVEou#&b@mRW{zp(GEg
zx?`PZkHsURc)<+5E8j6@2$OpyEj<ulpM!C_as!`sk0vojC<>B$n+h?D;~-#~B|HAg
zBkzF(ZoRQ~xWdvoQ8qa_eY!RDW6P#q-bc^@@jnniU~>Xqjzu=Vs#Wq%UrD6x`u!7?
z0vhz`gGC@^z32lL-Q~n@s1^l#P&@+oThCT|5Eepnj;8=~V#Lg#Ul#idSycH3!M{+*
zJ#gcO?e``vLq*ElU=5c6j}bK;>vT7`C-!*LOh;d#QE^i?k@X+ac##~^=3#}7b(dY#
zCkwMt#b0LId<su|=F85R5f#|Rq;+Z9rFM!*n*6`9NUO>G2d#0s4hQn61Hl3L-gwr^
zQ{`x*;r7S{l>)>C*uIJM38F~PDDjjJzLAt4<rG&@w?zrK_wSd>mlO7&EJA^ZULmK>
zi;KkuYt-=I?A=TJ8FmL@_=(}u?g%7Zd2hd_&jN|_)@4qX%n}VVBW3uR7vyK>)x5{b
zbSP@Wec+L-P$xt9Mno&wro{6qNwygvNF=4YmLi6Mhr-@bxo4#h9qh58=XudGXL4g|
zkoRObDt_YyuEpj2zh$f*XMY-y_MV4$PtDCDCaB;Yfo74=W8fW=r(V_-R9$IRuUi^l
z^7b*HI#)%&qe1$bgI?4)0X(!pBMTI63>%1sMXOjlJ7#U?X<d;I#13!ur5gQ@O|r~R
zW&3%kfp%S8-7lWVMlv<9Mol_t-{M~1SgU*<d;42p=*WD2d>4W#SYS1ii&2<K?J5&e
zX_d0KSfgUm39;FD!@&H_XA%4nQa82}d!@&NmT#`4QDWHlqUE3!jRjHKGOR$vx0Q#^
zMsry!^~mG8WL2{HYsne$Tgok~tFZh0-z8gLUOR%ubaN9<ruc(LHmjDNeIgk?aVr^r
zb*t%=l^8-jCp)(v30l~G*PAFvDZEc5WDHNJE=q=*hp#VnsaUk5e8W@TJ4{2Zjv7rh
z3?)u=2&S$<2U~0HxZGGm+bZd|tI;?!gcmsDhhR^ZzuFBWp3M8%253c{rkT%oK46SV
zB-*B)ClXJ`gMpD|OuaOCRyLm?K(qPp%s^oW221ZgfxKfBtkR`T+>#0Ww|$EiP+qr2
zD77uXPb6V*bLD%ot;+?piCDfaGg+Y6*l-owWm^%;TOs402_s|^q3Lr&3Ac#SY?912
z7ksIhgklk%=7Y6NleaWx6PPkZsQs`ap(-27EuKJp&}5f+i02n#+3zG_&h0yZt|X3U
zce;oD0w-g-gU_wVMq^|6`8+3;CZUxZc3DS^Sq{N%&>?aXI<&G)6_9bb4lAP^dgtOn
zfA%qmG~m1e7ITCK%2*X%44U)Dd?EGSF51Ap!tPlX-XnY9TTCYKWn?oEsj`l%9B{$U
z>v<NLHbpfnrR{;PnOaypzn_c$t{d3)E4Iws(07veQJ8MUSEc?U_Bi%O4^yVxWp#6B
z)eF9qmNGTYg*fNjurV(@xTcU8z^hlUZa29#u#lZbELt?+Z@>@8?GQw<&viFNF>$D~
z#n;mAL2)#rC%}yvK}*$azi%G&2Dfy5MrSY)-=F-T_)VSBP=B8aJpahxIzCnyvw;)B
z$3~sn#JS9nqH*Z!^Y~NI|IX-&<DUL^<~mlCib#-fUGCv{QKd1U7OgbINRJzCs2U`4
zu7lhPJm5!al&F<vg(XJ*0)JPPO;xywz6{cvCbaBs9q9-O_XY2?!~m<lD$j=i0X6P4
zB(X0uzrsDM+uv$3e)@V1kOc;?snJ{qImev87y>KZo6in#-ch`%My9}>vHy((6%j%#
zN;GISbfO^Uj^nF*dg6h;hh34=^>8CI^*p>TaM9@#V;_er1JI^<Iym;qjED)s%YzbG
zPfiedx7$b=^M;44MHk$OD>Lbv?z;mV?qvA6Jrne&1_8Iq&mP!HRtZ_qzN+$sbW?GT
zF~6#r0P(B`=Fx|dxO{<gJ48D?U#HrOzlJ>LVsCsm!70-K|JavKsLh6UXcSvj_^hx>
z)0+uq#>!LDLFw2W{9*o-pQ)w>4^45Mmjy&i8x`Iu8wAWhSH6o~nO~sLB`FVsGyq*%
zM!9iy-yo&4LAh~`Y+VT-&xkIR4JSQ9El*zB{>>a@p%^IFG+fG}CCPuBIhPvfz0crG
z4|Q_=iK2MM<37DW$2U@JCr86%mP7(B?~`xelDD=~$k#OpnJN{=u;4=)7E(uJi-yXR
z`_qpUjRuMvM}IZovmCiLsE7?*(RuLE9%SH@yrXj`#ApKDyv?0FOS!sPIdNjPkANA%
zd&Th5PU#bv2jC9o?EeIk6yKWh6kVe8YQUX=%EXqw3{9?xbil!8#{kh7=S`(`DZk+Y
zR2LO$n)1&>&H9K{;zr+e_5}M50=DOIXPs$7^{BeKbfUhCplXJx?c|)p?a7A6XdOp8
z76EWAYXo=W__ztR;Vkn*l;-jh9U8>QjTF+cidL#}h$F{`z?6H%z-k*>#%ut?PXvaC
z0V}4!=WBldv`lX183P66TJ3y!mPd|Qb|?s8?yePXpb!zI72|avWh~$IeBFed4&KZS
z9ZyT0s|#2Z@5ToV#0pLYW_Am$;h04$Ahj^Cq2)UrrJ*B{Y6b_jaz*SBcyqi5w7~Ba
zScj=<XQ%T{<OXD;u^GVnN&<Kth0^1dnoD;Qg(JR~?_Ga+6y2F916Edg9}u->Y#aF7
ze|AK8zOL!AWXZ)hw2ps&-*@O#NzluU5;E6Q^wIY^soLSW@dnj3bz3!5^FvbD^=kXP
zOC8`#OZj3w|DBopa@;Fi7p+hd&lB6YW^#l`Tv|i#ZGEGiqSwdXlTqjQ$_q~sYhRu0
zt|@IX1E=@WC92GktxDr>)8~R(U=w$<`X(puYK@;LRW8O4H_UCgY=b&6AEN4s8}LB)
z9Rs1s`1IhRoe&D9yhUc0f9bo)t^dvps&{^u=oTSDh;9XRrHkcz;3+Cky*ge0GqIFS
zam055ViHfJL7qg6?77UXn#xBb3+}CY11`;fBH`kUG8}OONRSWihEVnpEt1@NN(0cg
zsJ7Zl4PG-h<w`?K;@x~0C#Eu<V=c4zgJZ!PK&?Ux+&6F51LHyS$}O}1i{<s|QOX6N
z4u7&%^%g)NYQaEE!WvDnw!R3|z&TjAd;0~SgqXZD!pkJ+=Jm96JI0)g5vcUJo0=#s
z4)IQ1_elTZf;1G(&@;&fm9qh{d1xyeC;BB!(Anh|5LOk0%iqnC1>IP@NWBthDE;4=
z<D<1ojIw2f?~dWQnG37Y*h|*VnDGo7gJAVn9jmvCF`iFPZU(8RsNvZ7t&}@dYwxcJ
zep!sU9h_^Ig_qQR56x1nlsUBNGVXcao4=gfZOijk`Sxe_XT;;)CiOmj%>JXw`GAhP
z+edpL^-C7DMi(p%?u@+$^aB3n7$)1JW0^_+KHl{fJZ`uTH3{rwvUw{sTPdv(Q5-R<
z&NRQd=Y{m5b;FqtqQSVz=DytJxc7Y%|DB;u_Mig#rE}etsOVQEls6NQz&G1Q<lMh3
zP|g3HxgVKyR|yaBFI|=x&kWIP<M~2-X@9`L4$|i;JtiV+TNxq~h2fKLEo#Xzjrh1!
z(rhP@>^2Bt^Vlj;@-^QqdJIj9czlIm7O1;c(W%12Nfp%E$*&NxsLF75sRADraQ(Y~
zpU=t3V%!n#f1ebWcEA^@s2C~`3|y>VXH}DCz4Yac*@SC=4=4?|MH}R38;HeSI8riw
z7k{>C#&*_xc;Vjjz*=kYyvJY}1s+(=5J^-rJAe^EphWB`fD5WQF^N@d1@w;541g$6
zbzmf0&6YDy8O7}$AqL_Fok&i5bD!$xaOZ4;iW8sqsJGc!dflrQm|sxK8~i#DFR*F&
z_~23X<d0jazqZynnEJv+>!bDSP0Yu)_7UmK@VnA4;<?@&&Ynm_f+z^DO=C&iJaj;6
zfgXKM2aaDWufQe5Pp-seMOXqeKA2(wp~<*V()%4yd$71d+Lgyh^F{BXH3Ml(EoE@i
zKL&FaDj)#B&dSc-%{_Owxi;tzB;Liwn(jx;Qc8^U`!~szuZ)wtR8F8cr~^BKOg(`(
z;bjpba~4E%CR=RuWz$a7%9jD;5AKkE&BDf!&luoeMf@Q_Sc>^zSFrlB2Ug!_DjeLB
zeDItv(6KmQ8};;PkO)6$9^79zDJKdd|Ed)JvCfk*i2&PA$R`{L4MH3md5JQGbZA=N
ztZy8~L8X7s<dKC{&=RAJ$R9c$YFR8K#AjZu2DJB-0S<XHzUY?3#zZYcDxlV>{E1M6
zI~Gisr=$fdmdx%RX@`=TcJbbFYp5rfq*ETQJLP~ijTa%bmv{vAX9{eM0&GygKN+vP
z7yVvoFt&X3R0cSAHq5PG;3}cGoSvMZLqwJNVQ5VJ*hG2YK3D`$HTl+Kh2DVNz_Pp*
z$vv;fY)^bS-|^_nQsMzH(%jOXM<PvZGkv1W|D8jujB%tXp+Tg>iLObo8OJo0H|;CX
zq8!z7rP6*fa-+d>N9o%IG_{>m^e5V^nf|RfsnPDx|3!`OX(X8t<$%@i)g&`>kW}S#
zO13k4I)*Ko2I)vgBu0v?X@{Jp#LPZ#e%NF+#L*?q=e`4Onw2G?T4Ife>oB_twY*5N
z+Lr2TdIDF}dzkiLNzmy-DK0y*!|lXOfK?Z+IViTy)qksx?Pp$6=rBkGhPW7`Rrupi
zvjkt#0-K>s^u5t?Q<nrKv+%Soo7-?j?gzHp-5fRSaeU>FO0ccD_Gq~;@Rg=(@B_u8
zrEn{jpQoxtOEm!Va2|?e8$w1#E&Gm`V0_CPRD%^8&3<mAEQBkn@~d0-iiU&+j6PHu
zvMrW3`jT&!p6LHfu_0FY#1>Q%z%D1?)F;3qoo|H|8RY^mgzPIeTI?-<-N&N$uD<p<
z8jP@$J-?5Vmwi>PIV=~+@7D~XgWdUFhNe-<FTHtwH{%2_qw+(`seeUE(1uGv57SOy
zJ2}er&21S)3h>kuCAK&z&7Cq{%C4pT;lpe<*&ar#G|CH_oq8(qu!L+sReJ~xh|9vU
zgfZCZpe3lQ6&@6h!L|U6+<+VeB7!(`fj6Vg&P{+6$G0*N&q*mKyVqbK2Y!={E!XUU
z(=VF_quA*@rUKO?{6{xO)b`XSO+#n*FO}#9r9(NOa!~MvH11+)<yyHOCG5M(?jl0|
z(_zPT;I{xpd$jSi%-EL`yD=^7nLyYB3^-?=Yr#icyc>v8xeHf_TAc%VT2XS8(VWp7
z0Ezwo%-8=n{0DGs*@_xwxKyi39C4~q-i)IEnHu7qzPer5U)S~V)|F!&ft=KajAHHY
zUcb3#kvfY@TZy0hdH6P7H%EBgZzDEq6!6*QTQRoRn0!mAm-)XD)r5MJsEbI{dE-1A
z_2z*^ExT)}!saJf^*o&?Jr!C>J=ty|ZkV7vx2#~<+t^T=x?i(EfaYuQ5c7NHf<@qK
z9E!2|OYIF&^T&>gpG_!e*-7vA6AiWDiO6|Z+0N$Mg4&rsy9al@GdtQI@y-8Wq0PCZ
zUHv@3rdyp-J>ILHKKgIB)Vn99RP1HhWa|NuTbERtDsmgO*B{?UKazbY;b(DG(l;}-
zaZR}o&7b+YlePR5k+>Z4p_M8o-#`DPxqn&U>Y=W9HG8L8vsSOr-~3az#}^><HAlbp
zxs>m9{e`OG(sj`f5qZZU2j%P63ftarLn7i6({IA(3X!jb1_|Q|F1`df>+_@|^)XFv
zZTl_QtmXx;$|vn=60SO>{M)iN%P$PuqgzfL27fXH%szWZ%nDA9?3tSw&YIoTRKTao
z40v#FAt42E1+N^s{5r{Y#NnADVmy22TmI#|!yraFR7*_`AcE~Pm2P(;T4-YX41xSO
zo+u+E9qQc$QqMUt4=?}oVZ^tu{gSPUvace_9qBA`XY6hfR^Jv*WTVm-Nyo?r_R`+;
znu!yYz!1P4D7cc@8C1qb{qnyJGaLVViIxfhEoe3Wa_Y>`a@n|?D+$1-alE{G$#nS_
zAF00=&IPDU5~o)+97v|7b+g`+JJs8~s*qN|WD_C3XP+pctZ8%9bZ>0%v_jiQRTUtY
zh$|}aUQQ^Rp4=@Y;>_KXezJC&A;KhZ8w$Zg2&3^6?)z(zE@aeii>$N|?yEB8<1|?m
z*O94gi*3`E_6JJl1AUyS@X>+<B$cY05EY5?MKlQ42Wnm{{)hopPb%{rAC=bq%_QHv
zu7j%}EqxMa4&&(06&Qf9H!J{7i=<vDLhT5?@{tgBTG6U#$GHiF$_0Czds~`ya5Z-^
z@~G5`UkVu1L7ykJp!-GyJ^w)b3*9wWetdhFmhSD>*UT0{0`5f~H};XE@gt9uKdq3g
zbW6Xm<rQ!}dFUPFI}v&<?S(s(H}uuRwva_Y1C*EjVq`?o2rkxN$$UUWE7#f8Wf&&O
z$i`*EQ55m_8!M@~oPP%ulYEZDj-=yROFQh^cO`y*7@5LR^t)N!$Dz6owbXpXZxqW{
z$|8Ar6GmT7-|%cxi=h}15QLZ4f{enWA06@I<J}I*F^Yz-9vL<Fm^b=STs$#CySmrX
zk;Lf~L)P4?LDL!~OxB4&P}R6<49_~WnWKde4eBfic`vb434Q7wBl4bWo$-^r+Enu6
zNCJtq&q<8GkskvLSr2DH@i?E*x#Jg4cj&S`Ov{HjCmZ5w72?cFTZ|8@uF0`9rN9hn
z3A8>w%-S=L!_GL6#CiXxR1t(rxRt^Hy^}_eemX`OB1LTaX1VlOvT`t00B<i>3gfxP
z_~YfWs7f0jhY=InJTfNLeqp(GgbgrPKFN{C%=S!Tf#XXSnk%Y-Eo&qC<pdrt{QR1)
zM5G~Gu#$O^HIv+3vIhl+{33rt<uf(30hdo<0ZtfbKGyo9xEt}zXR}kv@Fdys&nZ3+
zA2ju?_D@yVJF0ZMk9TwGchZ$68P7NZi^vbwaTT@a|6M=^C5QHkG~*)a6216eU8k8v
zY?ADOvXhV@)0+2>k9EYH9hvEZNk5UN>3z5VJCi8i1t<F&7BSG$IW!U7R9f)K0%`$V
z!s6a)Y8845|KP)#FQAL=`cLrJ>;7b(%bXD%Rb3Qq_w5`=4_ozGbJ~a#n7?|*(f4P#
z(!}KPlg`izMpjL}o8+v(c`?_xJSn@93&lci2omY;scPCNHaBpja1YlsZTayicsi<s
zi6;lDkCLU$17dzKs`UR?^QxYF|MVRE19Rf{DrMQsouke5_nhfy`0SA`Y9TLFF@%?H
zf|^@<FQDN5LWLC}Z`CAGRu8ViOd~}e6@;BI2l)b)Yms$fm_rH2%`GO|RyYGgLU{(O
z_-;wBHORn^b>habr@U*am8ZO?k{3KDvy$(;d;_mwoNs&Oj_P{EhWY1HxyB6pgDfkp
z*o)PZMc#Iy=^BqUI9zTLwhtQ24oS-<n>;uD-g7yr_dd~(*P!;zjLQbE;(0z(oYu7@
z)VP>(Jn#C5cXMgZE7F;`4IPq{))FO+Y$6%xSE+}z<txs;pyA-+Pw@j<+4|wIaNV!g
zzfsuBkusEB#Q5_bC?U5;#G1hYF-0ivvt`30b16f0@8BI8xkbwG)9i1<ek~5dX&0+i
zY@f+(KO>iEH&(gSSi;kc&jCbGU;WEYjs*@dA73CuIeGscP@2r`KkL^loC+}4W`r{k
zr7F%igsNjEV0HfE;~)<eL_V>6u!%{yqW-*aMXlL(Lxilfi!8kkiTEph5ZuBOgJN-!
z-SaP03(fvXiV*$QRugHfCp#(L#oKx$9x>hFwWRsi%78>}X9P6!7ivNpRmvK<e1~#c
z!a92AucjrmWV%m>x%&4?Qov#Q!s3-D(LWetEx)ne?*x9@5e}Wgeksq;VEf;av%BYV
z-&xRZhFK|)JRYpPO1@~PWj@kURc<DNq6ko&uZfJDQy8^L()YQxu*uzjcZD$I9-#C9
zj4!iXH4T_2UQ2DI1SbA~I!@UH+&;WsMwD)CXSUhncu9U|KLdtLI9)N0Tvpg=S2*F7
zl0N3;8rVn`gLXQ`i2j~NAxmd?Z!l3bqo>T@ZEt~%)%%r7jZow+f6hCjpXLlWBeNF*
zG)5lRQU**Vwo#uR?VH|FOndZ{56q)89lUMCqV-){b!Y}QZ(9=ovap?wW_&8Me{Ru8
z$uyXy>+)3cFTJ{YXhc}uYb=(Tcp(9N>Qcr8Ofti<lS^<EojLeg;X(bA;YtCTqDj>H
zh3bn9UvzAKY#NlPy~}UWX{4B>-|S0i@ID-Ho3K05!O0O%eA3U&==PL9ZU(24tu$mR
zGHBD!=Otv!qqWT8dLHl_*=cyw_(>4Wwie8!{(=y{rADSuOdp@-tsa9G<5R~N(i(un
z4E%{MaSh(w?j!R@<LxghB3si1R?_tty>d?cDV*<$s$5LxqBudlgzP1`0~kIVQ%O!z
z8s$s!o-e%b0GND$-cc&XH3MDc?H$iS1SL{6_a`I(0A|vu*gOC$mS*(<0GQrMvY<qO
z_UE_$gpEBk{lM@SA@YizWZ$R}>vhYHZ`}r&8s;{$gd%>rLJI<`4_SLK#6y2d0Q7hI
z!fe&QJxSPDFagNtbn(O><`y!h0x2CcZi&F%x}?A2%|?{LjQj!d8;9HVHSo3^fRvI5
zomGB%7!CjT#lPc+rpv&0SJQuIWCP;bt{M*?@d@Q)Dv#3=b$_4;xC(}(DEtDdYsKnF
z%QXb%_mh!kuS1g>*m2l_Kl-}CyCoNGvOKwDL37x-_#VHhW0+WtF#X&`NiEgQE^#w4
zDcuXScKQg7<KgUI=%>D0+`iyV{+YWqvy-a$_D|6$0|U_oU6Z@hvMJX7kz`lPqAT6Y
zby)dnx7hN{K92j{6}kW<>~XG8>#>cO@Wpl)-cU(47q9nshU3u);)T0CbUaUN_@X#V
z;X~v49ln@t|Mbt`l6NR`pD9K`Z1Bf0?{2ZTwzJYHzg|VD!>`_qm5Y3}>+Se=<Lc>!
zJ39G6jdC4delv=g6?cv&8nNGT+vj~YO-R;g<5HXq-H><fHzHxzz<E`}j>)_CQ&ZeF
zlJ9_wODN7Sc_wjBlepYYwu~s8FsjIx=JrqO8(D2JO)46&EvyQPKJ~xT_Z-c!%?>}O
zSQwE;$^CcsCG0vlLW!~(c=rKK!yMy47(UkpB4<w>T_&hA5|<!^r5;3Uvti1rIu#SR
ze=Q%g?7B`H8EuI`V0)N#wcQFoY5a7$w~htA{5*s2b0CcPL`nvx5+4a@XSeqFEQavr
zw%dR1WW^?YWtKg@+~K!?r?K%Vx9GtwJV07?!50na@kQ~<8DPAwd)s?pE)<fZLC1*7
zt|(1NDSHu{9!vwhKj7lsiG=^TyoOsi{oT~4wS!_we(ZCY`&cC=ZFe{k%J=n1PYe(S
zzuiQ}bDj>C6pgR8T$c`$M-FJDh2~1%I6A*Ey7Qt@6X&mXq(9vU*jR58hO;I3J04xS
zS3Q@$kv)oo(V$E7`vAtnT~4A6^@IB+CC)a^2>GvWpvu^uOPg-+Dy3HGUk8dX97j9F
zcz)Q(8aiGjTW#}pSbB%4yNtKHndQ}x=}zH3?I<}5j(0rv>D|*L_wi@5ic^fy(=_ax
zo=D*=%ht8tXHCdN-BJJ2hlZ?Y=E|{ak-w)*stuQ<hkohUEC|5-SDQJU$6%>eEqguz
zt<7u0bf8uND?fIkd*h!tlvl>sS)(Rp+WZN>(q3}(ikWY+-Qu;8Ab#{yt0TprbDw&j
zpFcT26>z(o|A|jItHGmiUFR!y>dBjtlMhYpE+f1*ll&-ViDV0X$s1>7LYd=m3Fc@P
zB|Y<FUtq;32Qk+2>uZj>$D$>H#tzD@aehq8S8MN{H75`W-PZF3p&<pU+s|m9#^ySv
z4-8fZLW*mTgi{>!$FV*gH0QVtX6o1i5WbLm+fdKv*(Um{<I4S6;yYFZ)h4owrl@13
zJ8;^N>w~DN-yD61_Aul|2punKwzSxF9LBM03q$H$U9NYcU_qWr-$Z0<1;EBzY}^jP
zV=;1pDL>EN$hQAgQ<8(>uY}A?&za~2wXAy?|I_|YYy+|LtNA{^r?z>6AjRs)=bG9i
z>>%-0*MsE`B)aSp<>^PN#YN&<oxL)`F^FcdV)$%+vn>0^iQer<$dZUD#U?4*{4Nk-
z*;V<5YyO)#{35;r!bSDNjj@}60x-!ywjB@{fn5fO-6UEzU(c~D6^jR`z@yvF<ra9L
zqDy4$4W42mU_zE{Q3%W+?+bolq%F4+zZ;+R(RdD?$G(}BWiby7*4ixk7NgdyvI{Qd
z>Q09`)}H$-62lP;B27#H?Q=jv4r7WibzzY#i_l)zT9qrgZd$S~xFV!K++qKd6osK=
zPftfM<=$P)za)oO+Ov32^jPz{FLTs!T_`r~AoYt`ob7%P*S;Z|lD5ptU|GVE+;LrH
z#$oYk2T9oyM!%m0rv78Zc_I;S5e14EBSl|$!*?h-yE*zT7}^{Hvoldki!3GvSxD5>
z4C56yg_Q%0S<ihMyy*ko<YTd^XqD~sZ%~l?+d)3btixs<_HoJFK|YjkZ4;YbEOGxh
z3$HA_DRf<}>nXV~R6BfT^m^JQZbw^D&`rT<n}n)int#fCNG@OdRWq}d5kd2*cCR!m
zvx1~*8Ltt`L8se)cgONV)}GDDWw&)pk(Uq^pP-;<RpH7yD_I?Db<UE%-4BL4Mp;z6
z#AM$s4hskER;TcoiEv$QAYfM{SY@X-OA>#oOf0hLWpDMr@V|HD3tjNyY^pAcknYj!
z2^HfF($Pv<>lWzx>PcBK<F-+p_kGW?20NABlWC?-<;Ogi2#Fv1)HM%IRP&>Q>V8)F
zM%bY%cEeuFY(!&GSsZzkF_0q7UEt+j(sZK(%P791U`j}*lNPFy)ynsDPYr)<_}}rq
zcI<ZmR{y|LxymZ?^nqOkh9KvCsQI`Ug|we!@W`CPCNVzek#OXuOO4K#&DxdJ%%ZTJ
z0bLH-$Z%t2yGSLlK=!h~vknFh^kn=KY3sK>dyd=eCQ*N?<?8C*Y5*Hql{@*008BQu
z8Ja&K5;&HRFZ4h846NC!7XS$4<aa>_*nMLVt3^Tp+~VKjEJm3+=P<>^Cn6@lf|a)`
zG3KJ*hwrcs+u2lFL>Y?o2C()tlcS#=F{v-S%@=meg2#)3ME5+KsFexImwVH`P|B<*
z*%MP{C*s{Qr3}%gXQyHtDy@Q>Cu$Qxs|Scj+23MnDdot4^11xhIP7f=*5tLitVzhb
z!YTCgW;F~6!@hskLfP~o{q~92F65t;rjF6rmQo<(0I;h|Og!#<QP>Z^s$MA6#}Y8N
z^h;YQgboHcH^bo;>}BjD*PyK7I*HeTmQxP8zU$?!Bg4T)r30Lr-H_|%HeYvy*kA&z
zI_qYZxH#JFtn0-V<^QD_2M>!@Rec%W3Fx_E*7g9aq>|_nE$k_9cGP;U_9>TYd4on5
zsXU#hCq6lM5xRCYJ(6<k4xGcGNYb{$_<#58p24n9sqrN~RBDd+t5|-sQ)pyY!KqAn
z2DD}xBz^Cr4u*)<+7&iyj=?Jv`8U)4E#6`GF3j_=7e>pG>%PL=czG_!F9j$}i67A3
zoN}6dn)UKI4Zc{vd>fDS;JjIS?b4FH{oh~L+7MioAA^3Cm)<LP6y54S#J39xE#^DQ
zcuhSBWZp2DSQ`<qYr~2kojiGhMPvoEQ`*gMzx3|;L9uazw7ALlq1~=;^ANAbFc*eb
z33*DLWCIXj@^5Kr)r)(!2~^tjaZXOsfnblSIwS?bJ!=*Zph0LcTnk*KJhjTd0wSMF
zX|cxfR+LTC;!YLwzj9y!E4bWwXxSpvVGbj@>L^Ayklrf!AotLz%Zv~HWV=Gwb}E-V
zee4f~y;bWv;A(CbNwQg&IZCnzIbO_txm`1E77k~jIv|iCPvf3Z>W%9wCsX<?K8-D_
zr6yin`b<9I+Qs^kt=DK4;R?MnepDM@h7|YT1F$5aO|GBP8Jp>w&)$`{j+~ZQru{yV
zozz=`p7qFB=S{*fg=m4_9D#2#yUFzHS60#^6dub3CjX#{^!MgcVvtg;=kZj(By*_3
zaq7e5KC`XU*-)|)mUG=-m-6?Ns*dC~uX);-l3jTrgZDtT6Na8>zUgY9P7FS}bXM-Q
zr9Se=z=Q87<wC^vu-f^=m=FQ@=%%<tN|Mij#gar%A@!cM^--elwgHg?4AliszSDZs
z@yw5{&I1&oe{p!j#@^w6J;koL+k}kFNU1zh&l0LSP2<V_NvOdKUt#U;@{Q!ntm<p+
zDir@rPU!7DB)J$l@tA1U<830w@dF<P^5o?rmZG${?NPo2$&V8^ujO_YE7<O+K~|vr
zKNNWgL}O7xZ>;lxxMwrjQAvViIi_Q+Gi8})@D0s9S#0x2t2pdtePub&f}QKB?Y{-<
zqRKU(OanIc%7(O;#ck-K3iD=5tNnT4aDjnzwv5IBSl4gYD;NxJPgppMsc_`nh8(c4
zu&~@61{lh=$a`Mp$r`KzK6%8~+3pSM7L}j1DV2rY%f~SpSa;6gFy+<U=g4C&3GaN0
z|J7E+xjx=CLd@k`q`=RL@K(y0kGb}VVf(E2L0q9!wzLm@{^?%Hi44;OcO=(0W4=v`
zW*tv$%DHdxM24Bv#p7N;EC-swYTnakMQARycti#@dz2w~_?T$G9D8DH!2Z^3;paNl
z;!|xiKk2}S(uUjSAje;-9a1(YL?_SbJN+`4J>icEO!%G_9>q8C&97_Uwr^UI@*s_l
z9@e<62uSIr@5{iWkrQ(-1IiPzlga?;V|?NOD(zWWNSE==x90>b^vH77D5m`?DJuKS
z)fJvv=7mV+S)xhFxs|MVehtVwvE2@Z;ITiOPSCngM3T1no?%)XY?86vh+j^mG+JG_
z-NT-{w%2B|DW?xD25kZEb{kTC3k^)%@&{K%gMUL>$Dt*ckrnpnD2bclPTj7`rIe|Z
zNX!{Ru(o@aoSf*zwFt!M^w?W%ShqK%JoJet<PuurZG%c%(b;F~P9kjJcbAM`!RPFs
zYbkaG4jfqB84rQV2j0Pp4upgkYUOt7z7?&QLJ&K56}?}QPjpXJunqU}`&5iPxwN~Y
z#z`JS$2=d3$v^f+?TPYOx)Gf5q=xj=#jBwF(tl^dct7A8q@8F1MW3+D(RWbG^hLwC
zzAI%$1X<F~_zO<Zdz}9_PBWXT+lC^|=S{XwZLzyX_)ldU81PNJ+0s|f(3%hV6$`wY
zK%Tj@6m=>p;RDqwe~0<W80%u*e56uoDohm9L*72>WrlbL?VGw9coc24qtexc-fX^!
zuyk7vLxDpyn@2hAHWAI+g>ONUg^yyLZ6-S%*}t*h>%$RTvoLhw!IEQJ1FBKuL_1|e
z!+b>p^QADtG+`~^WqX{+P<*mmT%sSMwZm{`uCoXIc4m<|;}}=;@55pSpfKata3=Kl
z%GyI^p|K>L1BTG-F1Uy;ZA40C-)o%93v>bsFsozbTJ_8w0N<U0y#opVK4Gn`$W8hz
zgaQ6}ijh6LY^*25>!6YD&W4S}j1(L}T0+QhBF@AXm@pq3+FH`$XL%27{>h^xspl<x
zzk}ebMg~sgY4!b;1o)SMaxsH6z;7n5z`tyE0GO&0q;M(uaaBPl3=m*Vx*7KL_AhqX
z@Vs@QYi464rHUh-X9O3)G9UU!7eZ09QZ(FC#Ih+`pdl7K4smw>CGFog%d?I8!V3(J
z;TEP;6`Z1ew1f#<^L#7Sr!_ul1{N7VS_XuVCM}*H@PcWkC`$dnR}(Tb8?)g`L}MJG
zXWKjmX54^e435>;d^K-7rmOzIAn6Mw#E*qo7;?jRRv=54rNK<~%4XVIn6?X1R}8OU
zGh`G+iQxyrrs#%h#k_CA99mXN33}=74h1RJ`gF+w0UUq)*o{2m@8}AAkM>C%`~3ZH
zEp<sJQjS3(ymdH=59#gr+XZ!vC*gt!vq%I*{C>W;ikfgkQczW0#H=~gE$Fr`$k!V+
zJn_{kFY7Rmx1+9S=<&#v`ui2a!}ZC7ks|V0VT5$K-ZlL51D*IO4<ci9VUqZ)|HX@w
zE5F;4yKReUhB{ZAl8PPgnUhf4k-{7NzI@Ba*w|}FznCl^zt6DlBeMKc*~$)o<nBCX
zrQJmr+hIg+o4E6+=U;uF(<E1SG)|LFbIX+M&9V+GQ+B0WFrL*m(sX5z$yU}G^<`-#
zBzTMUrV>>Xy&liLNV~*Y$^BZ@<4*8R%B};P23lTd@9YQ_p}@Y<6oA>SY!xpVpSXm<
zzGD)IfQhMVhXK=NILI87X7#oh9YFtRCS$NYboKnUm>4Vq<E3waUOr@Ez@~MYego?&
z6LBM<7bAK_v8x3Z1@M39=a<L6X)R_@HUW=m>A77$LN`dNygwU=ruL?V%F~qk(^TCB
zuv5y|jLGR-VDHRsr`cfG4!~`JG_f~N47ercV%~{Qx3rWt&?%&2k;m(-!IOSBhrbKv
z0g|nCHQ-XOIK<l#uQavy7+`c>jC*IS@WfUeX1RHnw5qfgeUG-8sA!)Ut1!H?>z6ZJ
zd4V-@zE-xrC-Bgy+3j8zzbjakWXzp8>+NFF4O(K9&K}GCWNWA;XHTA(1Uu}DXPye=
zFCZ={>@X(|n8R$na*-F?jPqOMWSenkk<q%OkphHWyt&5Ed#<MsWzzze&=>ZFU#AuT
zqv=%@Bt`4^oxZ4L3cSnpG7;Wi^?;-4UH0?P&ypF9ww$s$HZ@jz!`!e))cAM7&8mf5
z@(6<a>cssmGP+<{R*`c7PROW~t<-lcpz1f}8<}V-KOJN6a@#C7{La0!r^&yPCl_#_
zc;4pj)2N_DpGI?ozO&M|>+)}+^KLDN)z7n9y-x~uu2Bf0^O>GkVDFrKG`I{iF%Eln
zM>Ahn(b=d%?!->`tKOKD?t$VOjIVZho1eHA{s!&trb|^Hb?Qn!<#_szOfA2_RrN7e
zX~Xrd&gb$c!vCFl?%yfws$lN>Z3W`3*b#RzsYJniNaD32TgGeod2$SQ47+vJ%1XLC
zW=ZgVq0+hM9Zy}9ha>~C-D*|5Z9Xgo^o=}!BlCGC7Sv_Maf|znj9=Z&zdbSl$FybH
zUD{(8noSYKhG_V{3w(b_%Vl*?z|!U=5EV6@+Y<yD1hNidTB7*L7G^fU(&Viy{<{p{
zT+o1BZak_&Y0;nvAWU={d2PYJWNae8%u3PuBUFIL)lSF81~_oz5GHSxD-CjU5BGtQ
z5paS5Hm<jEKzp%|S>9hC0*K_Pz>$%75K!(^%1TpuN?h~u!0^ihcz9LSi&sN2S_Y}l
zs2{Y<U0`IZDcK2@TIC{Tj2hJ<gdA&;h$Lv*Tckr;SK`M{LGERvQZMs}t3)AB&Zm=*
zDhEVhz=2_f+`5=n8vL9=F3I9q_HzW%0Cc33cP!I%dgpEi9VNTfVvz^I0(E>%Jw>^=
zKqo>f`wIbT5H^Jv?i)UEk0_v{=B1R!qjk-=q9V;sH?29$2uyI{nmCo+9laED<r0WU
zzUf|*;_){#EnH)+1VPANQK$&XW719LC?ENM7p(L@6`fLbbD7IBOw9Ns>V5rs;R%TJ
z#nMWt2e}+!C%H=5b+J(JDYU*Opr^@8(jjJ?7!MOA#DZ<R=w8Ox{yX#NQtBWx>28Dk
zX>-hFaG*P?8Lhl!R*<*MAi=Zm>7h);g>&^gy}aDIRD(7Q82r5Uu$RlzrnmLJP|RAu
zajpoWPza=1D><fiQpF+RhPy-vB@BpnwHbWdzA^XIERy?IiXK-};6R?<hZo!(m5X$9
zBWj@R#ExP-ShWuVtPL;3LsvW+H>g`{ziWUPGYN0{5!xMIyJ$FE<%ToW9TwfIYUP&|
z(z$WU9i-Wtvr%FXYL#sE`qm-Hvd0r3+e)%_RpQNnb(0e={$fqfQ~F(b8ba!`TAMW^
z17!pzV5uu-$u5=vB;F^kFRjP>z(5R09a3FgBjCtgR3*W}wht7CGgAs>aMIvZL=0V}
zr`#`U3bi4>nLIxB<e2{`E?ojlu5U4Q1b5h1`(rbvfo$3l;2`15cneAF%QzKN`)M12
zH0}TEBM&?IjcBJF?)*zKbW_)h?zPWPHCdCt8&KvT;!lGFDfU@~rnEBW0tS6_JOg<+
z_uLXyaKBW#t)@6^fMk`0H_im@RsQ=4d9?z^nD3Dr{NzUo=tN;(V06GwXCJLWy{#x2
zj?Xj?l8&9CwT{6%SuuW^Yf$zvb_Xobwk)LuS0ct13|PAYa-R<H&-+<J+dU1r?F1XK
zZMPN~St3B<-e9OVAJX(gqy(_+2<eK}jR@T}o`Tva9$&7HnR5hvc$4iFDj2_+wWo4h
z!iDAI=xAVlv(b6E-fBjX?sI30Dmz(^vh(#_U)~i)Sn+ds;6aY$mSgF9)hBY|{C1k_
zutCugIaduMw%i??q{>DXNSzEN<KLB!@Ah<r6jF|)Pc(U*|4DvT>>QM7jpul=)#}xe
z_a;U>?XBs8st;%XdjwEQ*wpM_RF;y&X;t;*Tafa!$BmuvQ6PtNz$ubre9M_HcFGrR
zMZC(~w465GQDPjpV%p;+?o36iCHqNLkj0H)1lrq9Y3AtlOL`-zgIVbxI(*Mm!HGP*
zo+o@y^Jbqsv%a#t*m5mIUxt91SPNjXy??EatBC9Ec-#Le;0?kUY%PmJAMaEjLGnL`
z`mG%<lwQ|AYJjY7k($(OY}Z58Ur$;esu^pQki2ZQ%94;;S?RsQ4m0DIJ1okmSMQ@)
z3Ew6o@1K%C-odN#R22t3bnJwa8;~~l-P3lod{E%v(GHL6{AiiPSY3$MYUIEkUzu&h
z6fwuiSW>!JJR+sHasr^u3>d&X;1}~lZ!Rm4<jbVa|AXxO4o)hE9iB>W`p1h0Rt`@t
zUc~l%0@{NcfMy$U$HMynfY$`vt3t^}Z!<Fj(G?xVL+|qmPi>3XpRgqHU-T#WZ^n-;
z8xGXk#tq~xUAPjDrmTAT?m(J#F%$~({NsIo-N15gtBq+E4Yamn(N`#6q6NF<T9QGV
zXZ^guZNN40(b}O|FJ_Ku1>y@948(Srdnb8&2PSgJ%1Y)vKj(Uv1D|Ce%;ru14BUaW
zk1T6dqyx&<3BU?TlDEnb5S1%B&n`rjck>>2n#HBx&al^hk(npx_*~k<v?-rGCn>E6
ztD?qc{9Ez)xtYWdIcrN}{_@Frkh9(!AN{{ny|K&8j~TcZZ}m8qqu+P*iR#>Hf{9Wt
z2Nr%-xKH|~+O7L|H*1TzBxHl&(N*HNFC8j1ALRw~g$I*CR5?d}%jHVKy-Oc?a?R|&
zHebv7HxSkQtlRqKppCcsb>R)Jo&ii<o*5)i{*vU5!P`gcYlP&Ntr^MgLGo~ux9v!i
z`&d1z7$~jADx%(8;t8|)zVdFRDo3THsIt-C&w*H(sVwP}yVB@eeVAYFqzboN0_T>@
zH@+yBDbB%!HQL6&oOV4E5AYQ4U>6oEem8}4H_7?+Dn?o>A~e{`Kpi)Lmxn%Zv2Das
z@n!hjGQg;6C`OK!G5!9hi&3AE^I@Hz41nQL=X&es)qp_Syw<uLONfdLP1Y!!zFn41
zw!#BG=<soJpY~npMOmI0Q+ps22Tl8>`4aA{K8>S;j%+Bk%JAi=+LAyI<G(p1LGueh
zfDPFyl}o^P`+ucF2aeA#%aqd|B@$39x^g|;Gz4m^qVKld>K;VyoO>dI<-44_U}H<)
z)bTFRVh2NNWEGKexlgyt6K#vkKFR}+hdGFn^52;agGTZX+Xwv(+j-w3gaF}<k~UyH
zO1wu5Pm;C1dHz#V(!~mnDDeZwg8GUh$#W^I2^|OpOW2duI#h1WdPT|0+RGDyJ>w2l
z3w5ST-u3)8Y9%~@z7@NAjDuBJ(2wztcXaS*q>Sq`{*IEHEYe@sqE)JdLOb)Wn6D6O
z1{xNmw?z5Yb1Ww;a2`A2_<^<8gp4NA5a;rVxtA@F7O{FGF`mK{RfJ&&nwXN?7cN-5
zS}x+Ymb-1^Ry7RNME;naAb+ag3swK`%%}}mtRu_I=u!;NILutvcm9pHwf({)RZUfZ
z1R-;WUuzZHKyrLupR1RzrSLe=+d~Xwn{iEs<d#<2BblX;P(KLX1|9u>2b%rx<J4N~
zNCO-4ULjGea8qt6!w@+G4K%mzl1{@UvIAAlcLAd(*}pSD+O!{v8~BKV?Tv)_L@*}s
z(odQl{_5!_1s(!sH%ZGqgt?yT#kxev3To=S(!`E<#3lEe*h?tbW1r;Cs~Z|SCn8t8
zzQJko>3jSM<`aU|KznqC<Bz-tM$&)of4Kiyi%jY$<;j81`mcUXJXK1uSb%buv2>?4
zgY=Y44Ds>jOF&DssH>Tg5;u`^3{JX)4`fKuHt)N6(sMJUnVcM@de3{Dam7H3ub(b{
zHw)XSgog?qK`Ih8Zp6(U2y7jxj+~e($MWtkJj*(VjV=w~Hm>>dQ27(E50x(ogREr*
z!`1>@FrO{nS|R3AuFR^W8+Q?E$zN7Zq(_n;1jjuELz<s{D~Qq6wy{XQof{c`B2|<9
zGuKgXZduiMK$AzvGW3Cx!Lvi*>3#{NUzK>?`%>Xo8={Kl(4IzBpS`BTOF=dH9L$Zf
zpdnL)V4E99yzVP$0^jei_N+%m8^}AtXiKAEY77@D)o4A`oAATLd1L;^gxq^M)#Har
zyO0wJ&aVM+k^uWZDDKk0t^r8))8PPLf}%6Y<I?jN6|XyPCTe&aEGdj<nAayhrAeXA
z-7-s`9q%-K$G(?8Of54x_G>N&k55EF&!_XPp|JrSU7AFPb2*s*(IRD<fAd223^}J_
zSZS!%G&IF!ydwoZC6zgQ&w8`c-R6Y<fw$7lDPAo|FZ*WpbyF80LtL9mxxaN}uv$UY
z{O#H48JFA7#9YKdXo^2ilg^bNiGaDr6&}hz6tUcw)5sJ8;lCpiGboK6-bqwD5Fij`
z1=#GDr^+ZBvjcUOSAUqK5c7FP)oPqc^7PBa9T2|<(x_p}RjNYQ4budJLh--ZmgVSj
zo58j;WGkEa$E+3|^Ry8{B3~jOy6s+_oro<EKy+vUu-ZWpQGgM`3C<mUJz92s0AXCK
zLSKlIAaA4a8}QUaJ-*(vIp|MMpY9r(Ly+J6cCw`7fk(N{?$L?t56R?7nTgojqZeuu
z@+;?p7kMgjow<?&krS7kuC$%ij|{>suUWhH1XQm*5cr~M9AVW+3Kbj+W22$XR&G|G
z2tIc}DEf~HMB3d$h^BdbYUwZ5u$><>7q1+OmgzTV{v25RWPRHFVEKLeD9ORY)ORA<
z<O2~~5Udlja_y+%)MbFlW+#0jiw+sZ30@Z&`TpM-5u>l_bLnnEE^8)jg&0b%z~u6H
zc)eMn{F0*Xt&zGLRr~Sm_kYAHAqUZ(^YO#q<_SyDLZlAdKVCnaZy6@5PFg|6+h+W>
z{M2YygE1{Kj1;zPHyLs7-i)R5InEos*>53lxt`j8W@JVAwZs&D^V-$CG&BC@oDFd}
z4?LNCuxb`Y=QxNbN?Z48HUepj$rEuM87ZE#YdlNxhzec5Q=x93$AKk}32z7EU4SLu
z+uOJ;%fKYNoH3Il2mmYq^uWU0T;QsG^oOQ0Dt5d&#eT1@wx(Kz-IOJTJs2;w2%a?B
z(LNRJn?VQuy9BI?bPhxL>`40{-inTl6M;Ap=kyvq@VdeIbZIPIGW6<zW4Pr(!LkQc
zezUFj92wc5<^9CWH~I=*Oa^(k9M)&<9+xICD6%fik+S&)PFK3`1jyZ}e+oPI^h&5V
zeHoyWG9fxF64KhswTUv_GD>KC%*ayGr34Vq+d=??=2@g$Qp)RL54cP&#W7Y{*Pr(A
z{bx8`BD1%gjD&O4IgY-*@x(0*N6}KAsV5QQSw}mA24e%{_XwoNTm~<D%t9X;jFwDV
zPDvFTjCHd#?P(NajxPk;^K(>x(t9V#aVHT)oCwtwx+oed<{=R7_p<6q)eI6)t()9x
zsSknRl!61);p+x~zz@e2m5&)daGqb?oE<=a9V6j55#ukAB%2FGET6;pZ8LL<`0*WX
zU9&>ms^W&-EP3BXj9<Qznt6Rk;(xk1CXDg&<}kIPIQDACsIc5GF1O;(r_@_rZ9WYd
zIE4&wIwDT_)k|m^vi+z9^#yXR4Hc1ocWsyLpE3%teIQi$piI_$H9P^S-=t5LGFzy>
zur2zSiQf9IPu$7*dMx)a=VL41CLc8O0V!m`i0ngSvL2zZMn3v~9O%O$5yJ=LM0hhK
zA&dhG!E1<+Q&}|H>4v`t{-h_c4RDhm&ec*t_ph``-{kZU4iezu;>>8ThPi@akRzb(
z%%PL-^(?Z_)P4!77gz!TLhg_(<*|gjR|pn@>yXxYgdBI&MV&N{x5+dYpU2oX2lKy^
z=K0F?E9c&z2c@_Dx4X05DNOIjk16S&DP<G!X+fvlgb;ijn)wOw&uoE&u1UhHA-19j
zl&|Ex1Z3?86YL>F)p`ZT|Fsm3(KhKF4}7^(KAHReE8j|0|HAF-QYreL5&B-=XaxCQ
z&@TV&rG~M~>8;HZS}*lHh8vzJYT5SvCatyg132~1fiy@Zi>18~`GAVB7wo>VJn%3m
z?t+TY*TI@;QpB^T4^BXVI(O34><4X9z|O0{9pd`+w@fo^v@yu9W0k<<cJqIJ1vTu)
zvg8bz{<wPTY;f(Aul(ws*Bw<G^)mbEA=>o0|0C(l!=Y^7|Nm5E&%TFdu_Y$^&SM?B
zLX>T?Q+ATwgQ2mGG1hDuON%5+)-v|BFqDiXM4?6vO=B>8ez)(>e-6jtV6N-Fuj@R|
z_xtsRIL**MZSLPuh0eZr&3_&zPNZG^H&ha4qQN7`oIzWNPXyX)3--CixAF36ZIb5T
z)Hk!IP+#7#Bty&e5Q1B|4bddHHsMpOd=CW<JJxL?nQc(v$?MGFcw?%LRe~t#6=Rew
z0VqNJ1`_yp4p@J!0%6z2wiEfwwc&teF66iMC22z#kRqyj<*{M>nQsT+p~xw}l*WqK
zGHnOR&jg5nCg@x*i<2o%mhi7MyW4=<uA_&Z_1vUK3eRjEnU07opTk;1C4VG>uRlgf
zyB_YX8boJPsH0niCfG*WSlM2iY>9=`%v3UuO~BT@&#>XCBC{+y_C8TEF@`bXnlL;e
z9!X0AMFV~HAF0;;K%(~2q*(5W_=$=JCh7$K-%o;Aq(1xp$OlRn=@@bvn5!R?KAxEM
z;~iui@XkMMUDDw=+2M{ex4U^aDTgE<#4{FGFpbpO5?v(S-RZINzr`WP?AY+(@`4-*
zDL8rWl7Gkl{_*d9tnX#iFTyRbc-Um4GRD(*bS;_HDtm)B^xk`L$e<8%vqCW|K%6B|
z-F_!abY=)G6+wFZP+Fq;p*%)6^TjGQ3S3jQ?Po51Qra_+BlGN~M5tp@TcVzvP2`&M
zZ`%AKK4q?If4<FmKdgRBLTbYAkN0|rcWp-4rE~mxL#*e+#OM{9x1J(J?@aVf6<KJK
z;=E5ET~K~|##bg|9-79Z@{L}vb4J&Dzr7TC%{5oW1tm@^#+XGJPJi8SaQrSvQSg!;
zPtYgM%C?-c((xwRh<>-_W8MP<(Rz+zxUxFaRq}ZC5H!;Xr#$F<b@|mq@3iHt7zG$R
z$(@F~Oojv&pWp61@BbK}&x~8xOIqe@>!dt1o0xec6pHXNE55elM_60+GpznByZ!3B
zW+-NuCz>}!*7DT@ATL@r(G1usZ}@rO-hy*7FY5qgJa;7!XWVQ5joDCHm$z^FTIaIi
zW`WvzE_$UhUu`!*Y%1na4;5&NP7U^1Y?}+2n6k6H$^=xOKx`_(!aKVjF=)UIe4QNz
z9{I|ifTmcf_dN6E@bsKTk!Uf*eDafC4*fMm>atUd`h`{MWYW%e`B0I@890i*JvS)*
z8qf?aeE3?%xi2H-Sh#I>9CC@KJ{6wNt?X;dxvxGk?AthQhAj<@$bT-t=xGTpu73Bo
z{G*prH?yBUKkPs)0rl_K82jYOI|iyqBsAJg_o0#EIKk^eW})r}3UVmJ-s2t{GszE6
z*iNi4IsG1KS!mRdGRV{)oMq$X75c>Io+9QvNinYeZ{F#Mn_ug~d|@;km0jUBKLXnW
zMl)#veW$Mer}p1ZIloRZJ{7_DTOKZ`wwk9~x{FrsM7WiCT<0<G!_7D3U^cm)r+m`6
zsL!F|!|rLl|0v~YSe>Oz74uz(FJ+<s4zk|$Tatp|R<p(}C;#+==R$I5jYBQ<<kPI5
zO|5+yqWBjQaYv?BI>FTA^eKtJ#sQn;p#S}2_hqx9$Lo!f4g=OZi`Btllo@y;0s2fg
zV0x=7uNh0Pv<=fymTeV8N&lMpYZ=!pzm(+Zbegwatd)2%T60!*7V_XoLc5ZbFL+*|
zlhl7HvL=PTu7do`XyS(54A{Ic5}FyO(LP7A-r@gpY2<_@#@EaO2ondOkb8*1L||e0
z`34^y$swm%)^)Bm_N#nac#Bp8Ry=Lo$(zdn`opfcFLPpI;ys}M0ILAy-oDAg;x*i{
zi2gBqkc-;sqrNiPUHVjN5{*AA93X0ZkJN8{9b-jpaYX`eCbBTO&8yi>9JaY(olx9$
zKP-+Z&db94{^(%L$+=ULo7Q?CUMjw>J!g!&$!79}VLyDe8f@wQb)bS3q#kPdVrnpJ
z<T%6FjUz)yI4Thdy=nYIlm_Auy;a@Q?|J)bjS_R{i+k*{kSH+9-CQxHS>*m_)7E|t
z>u`mNXBg|EVU&4!ba*h1$1u`O$Q=4wZYt}~-X8?hp0nM36+jjbQ9JzoLU7>J%(hIR
zqy{!Lyl~*UqQDs={^~(h%T|Ka@O#aqlz6VJS8wc2)76_u=E@sK*W1k{0K$v%-bm+l
zHI!?`;I5JT`&?0q6lox&?U*O{5%Q)_?(gLomt)D%{YY;60;_-lzf}L{1EvYA!9!7g
zsI%PaMgTQjC0$?bp26MCR^oKcR)x$W*)n^|S`2aAR@oojlqU`AuK?L!9sbtK7RwQM
zj9@(!5TRzoX;e^ToQPHZ0B=UNPFqAsBu1iyEEuI@_G;`~-IG5#!UR?Y?p#cCjmg+y
z>6S6o|Cmz%_+^J2G?jYsh=Ldf^Fv+2FL+?S|BV@2@4RR*MbN-B^7tz0(S-pCwm`D4
zlZQKxNYGmS9u+LvqdZ*d<eNA=d!w72s3NsW=ROo9a$;igP+<6cLATU&>m|!hLdN4~
z_lY7aynWki%z-lmqGo-p{>-P&i3uOheR%f3BO#RpBVUo{A7zDGn^I*Z!?Dk;H?5vG
zdC$*(cgP6-zF{+nB{<zH4>X&+Tx8X$cJtX#&ir#<)fm1UD1pVb!Y(6f!@<API&oB5
z+~g@_PSojou%?N_3#FId15zy|<mrM^R<A434FZ#S#nlZgZlmen0@Q-uK2;5<720bU
zB;0$_VlvaAgn?hX<e)p0ua;jq1a3m+FAvr_`HiT9e@%#g411w8FT+w7sGr4M`@{WF
zPnnSNH9e%xEa)YBoant44T*?C4ks02+uB4yl&Glf#o{1AZ=!*MfQvPApVV?e9iy}L
zpnwG$-g5mBI+eS(X??6Wgl??s&hJ^NXxhaGb$rpLsG4`B==ZHn-wlqYOe{}G3xB|4
zIP%*09cko{)NKp%<%aD|$c^{m)3T)@n}+M(BExDAW4H)?2dAj@Dibw7f{TZ{%u|`)
z%FDlkazvXpU7?mugn3&NvX1}0JZtc157k!wl=7B&K)h3yp)xHbMOQjs-Vn$&MM$vP
z6?8jfgsbjHKRe0{qMLt`>rVitD>LTdUv2)LR2@BHfz5mh1XwDa?}~`rRgBxcq?qf4
z{R-SX>q1vWx{Q9$=IAjN&;SiK&pfXg2@Nqnzr;q8P9@dvJ4wj_IZ$Jz{fX}z^I;OC
zgh@DLpF_hVtSczxRaS6X$BEgusgWelEpQ{20>9hIx@QvFP?SCwmEOyssWm6zcD_})
zZi3GL(>#c{DLBcqKkg66BpQ(FYp>dL3Bl=c%|Q#$u!Q_|D%?9`Gxgn)w%j$|4r7|J
z+gB;IHFpE`YYTOvssFx0HPqTd!iwQ9o=%8md_H#xwkXnWo)h<c0n#GB2x&b9P&^ze
z&rlLZbd0u6?}t?B&Q14dAw!{`uDLt@^8VJsA>ha9i^RNcFt<vn=x)Eq+kH)!5UHUj
z986Mxx831bV~Ir}Yd=TnEy|)_{%y{^LY#S)x1cZFk^=7DWnb5p5+C08@)$hbT#>j}
z{$n`&vF2kP4T&P$N(k({!O}fR#D#*lPw!K-qc?0mjis8j?`Gv^8*9#LzT+aQiMHLA
z(VLn2)Z<<*z-B#g2~<#)bJaV0Vee9Sw5ZjH#hm2S#b^UraC+Xy=jyyvU;ovX%YmP0
zD9<IN|9w+XR;F>%LgS8se2mU%n%Cv#(k^hwp}IL0d#GpO>4=bPGXlFkY0QwERlDw4
zs7(4I-*}Q0tz>E5L}PicD4*sT^yV6Qv-QROWWJl%5zdOfllMy1RuS^OA>TWMr#EnP
zc4z&=)Nf@>ij_o5(Ug$!)&PFo<(43n?&HPK)l@+X?5W<hC2*COWMKAKY2aZZL$%oG
zx%TU|*d;y#@b#iSQ~j=Qhg?YY78dz=4UZyfl>9o?DLH{4xJr@zHLw}3jo);6P7UKZ
za^@Zo4SnA`PMxIa<UQ3VgkUbWpALUA&PtWgFbZa-G&DB>R_2d8)}<u3q_M^pLw&{_
zf#oA#dWTgEh#QU6MfJcu>f2qUoYEck{qFvP5ieEM0s=3N$|*zuRgGSY2(yh6RP~NF
zbR6^GaCtJ<7f{wrmI@KP9r51~5KO*7BG;-Th+;Eu#k9l<$<ir6M~(Mz9@Q@*)b0dZ
z=qlrh5j1mw;*AY)tUVGQ48Y8hzm)%1v(5RC?Ds|7;;C;qh(h?&X~oL@xZ7pPL1+uG
zx-8#v4CBP+fiJvWmPia?4jjnTFd_LB-CC5K!Wx^+v~=qXl=Ic!b<H-BHO<0cpR37Z
zS<wCp?;mNw`_u!o4}8MMOv(hmev*FM%Q%uDv-A44xceJEy?<B6mieOvW`-(FmSfUC
zvm38oa2n8Iel(zP+t4%y@tfu}5c(oJp+%+xV87sY71@JgGGZi~<q6o-(9P9Pn%DQL
z7LdYiHTv-Vq}St80bL6x5lSbhfC#VC;Z1`tx`1-pux(1j@V0|v>cfti4}YWx*S51=
zX)uz*Evb>~1=LHHH&Ji-)9%X>?V~N-0+b|IIH=EqcJw`icbo3isDt`2%FMb-jS_|(
zuCnx8dXjNHN`;!_QT1u2%dLc%d>O(-wm>$Na7|XSrAP<3CDw|Q1~KLO)%uo>Vf8=P
z?!8u7iQfu)xziQLH7`^cIf`R<c&h)mEFoPZM6NjfweI_{5lrTR<NL$0UvmM%N#1?a
zFdye*b^n7|4o_Z*oS*IGr>eA>{E1J$xeS{NHP`mri$>zA$Kf{$vDsa^vxxa7i~GX_
zpX;xWl{cQ1jWAkc3z4s%LdHHheopRCX+Cr_{q#&Z8RSLdYWMuQ3{_jM378I-kfz45
zd>l2^G66)_7KVb&RxN06k8Z9N)7SL(mBJL4kcFq`BvL5$4|P5t)E5u)XbAnZB4<-f
z<**BtRzlZc={{rwx5`*P)#zCVbt2oEeT@9CQ6K4k1$OXdK<D6!{xAAD07}H+_Y^u!
zI_$c*mf^Uko^f(Va;a_@;YiVg2>B(?A3Pd7uUL#tOl0o}3zOin8s8yrQK1t4G$p;~
z#S1PdzX|Pu;5#~bYv!TZkqH47W7~TnGhbehxb}A>KrU~Z#g(#eSMuwai}Pqd30~H|
z{r#prrs3X^R(UUDv{8uJzLB0)lV`B;yUGX0l0ArK*-%@g`YvN%tZ=A<dz&{XA@di?
zbwhx_`CsuoW-HxmQN}(@2cy7kbtd@8Oba&N#R{_X64RaC)>IF9ZuGF!6*UgsDOwaA
zhP;l0DdLu0{`b$5mDfOu{Y>qFr!BzegLtuv{xvT4S(vj7VPTIOG=9gjt4}TYG7zG2
z)3hCrQTy~(?fLxp!SR)QIay+m@1aV+JZOYA>F;@_b;}?4JvLgt)L43^wu+CM-(6U%
zOTpcM1L_xVl@3QpSlD=$Z#9?G<433rOc3tJEj4_z=f~*#$yy)YRsHgc&1aF?QKt*(
z_J(F_T_o9q>dU=~Y`i6rKVx1!>g<3>x43M^rFgB6lo9-Qxv-OB`LoMi!b|E@8GG+1
zo4MEUh6!^nA)6uANk4LPEecLYCiX|B$XAGtpX<j(t6D=+XUPKNoj*O+U$v;GDA)VJ
zA5jpg`U^f`bJ&0wd6}h<YPX1(xJdaIL4Mr+D~+QH`~uKKhDT2^dhb6h;`>%)E83g3
z?}vxNHiOt32B9>FdHqCLFP;hL?3Qlr>p$iV=ATaw@^vaMUu*9GHH_x(z91-ZLnq$9
zg8hUyL3wmj3~elnB+M$iG2o3`Lc9*2lk^zlifp?1Yv0W_jpaP<ETfvt7PiE5OPEB$
z&Pm9Nr?^x>{4ACGr`p^n$my=!2}>0FWkCo-rR-N0g$Z{ylSoHzv@tgr=L9e1L{WTn
zu`oM*$$cJqIVDp#Q3v~Bwi30Cdd}H-UxewSTHGbx%P$kk>-0&6?j*}3QkmHpKxNb_
zquP6$x^jOnI0&qCk?t#Of!CK_$*cRAd;og-A7PAUFG3#l9Vs)g@sJ2VT>b;N{WBOM
zkrkXb&&)eOr&AWVTx_^TUp3YgA(iYAEB~c224j}fbf+Ucsg=-Z_<s~ISdl=n%i$0w
z?N#zcSK%(hazsNj2i~G3n``d55COV}Kd-vaPF8P1#$_<C9j_H%ot&6xuW3a|@*84j
z_(ifvsN_y@l;^`ip|#)~4xyE9j^n$9fO~dYK!r32hY@)z7&83B5k?wxTaO0bqbx}c
zTN0Wh10(w>7~&$7WHf20ycuj6hK%8~oB+QgP}FV~GUrPoJ4I?ku3&^)?}zJXi$%6`
z2i!$8D@hKNv12VBFM^gbU*Y%{Ez3-0ESyD?Et4R#s9Ub8Ck5dqMXybbH-AaK9uRRK
zcN~dy$Y{&BG%_a|%~e~XW>~8g?DBTRVDW(l=q9;ZPaETW+t+>MO}$O~k%T~*GsW|c
zNX4@?7j5PAM;W*q<z*%(Syf%K+5wUXvoEJC%y&P?(<G#w^DNs(g!8-Q^f`Zi1$>5q
z7tkF9tnE{iu4bR68ehKg@>Y&$-o~Y$R-PvKW9vce*WjVt31#<(vJG{3BCASE0Y%f(
zZgdCjvlB19z5zc4-8;36G^meDU||T;wEARPf1H`LdH^H_fUr9HjoyTyMVIp1Yjl>3
z@^^9~nA#Z6+B;sB_KlJNGj3%~2^FKFv2wBE<FQ^gMtSz?dQ*Xm_pvC=@Im29X7KOU
zu8!SOH9!g&n$6eC4nW4`thhBzY-|<9ZO?s_3pL#`D=wnEh5wf^aMc;G<pZHl=NTS6
zeV)V^6*AM1o)w`t5h#``Nj)8Umzh)0%Ty=4AH&oN8P^5ya`R_h_0i>fD;FzmL+Ab@
zzFvo|Q3X~CR<e4ypky@936Lf$&HzNV|6kImAK8v3_kT=_esuAdlGy;el{}Hwe!=Qu
z2Hs|sI@v_7Y%8_&xE4(ZvrZ6q4$!YJHe2m8^hb(_7=L3DtrpwbP9IB<7|HH6i!b}x
z{&u%UotrB)8sDcs+BfP|wQ&0{6OQ|h1mzlB^K|sWZ9Z&8V)~zUI5q{@%DEX(@ARI~
z?O!#v;VtqgN=0%a8bRc_rN%e<GLQ0zQ6UrG?0V?+pmSumXqV*y5l`{qTM97PIv?=V
z{KJs7HGp~|2=BI1ECRo&3woappR@yRsf*Xddr<DCBG%I0<toVHOuaJE!dH6Pi~N>1
zp4Z{J5>U>a!E?JTeD}aLl*;Qlg98HDAGrJ<DKX(^MbdPGGt(!mBUM7sRL=14#@O2r
z3Ed5A?TKl%^+ZQHI`joqs?y%dg?|!yx==00+WwaFL%_Je2C?PGO;^Pv1|&G(*R$_G
z<^@^UPx%kl`)_E~yPGEqIevXuh-LZt{Vzh?a&I|FEo`qF@mT&%;r%9~w+@oS;x%>V
ziuc_)MvIaZi)Ih1e%eSR4_En`W8`gPqYb5_MV72pFCT9gje!CiJzJElCyu2--7vE_
z3rYN@;oIv%T`&_}U_9^zVBeo)#6fM+FRAHe9_brKK}d~U9zAfZ4O6NBapJ^mSrYfO
zN9+Su&_deyZnvyeDMUIy2CTU7iEzmgYb*H<V~vH7hrZ;Fpp({!<6x6!3IFLNT#UD|
z+ZntQ*d|KemFAm(=tHd>BVn+H<~m{W&xREZ|AOvdW_c*tu^^Q?*d*x~mj7xZtBz*!
zrLAthpz@4bk-$y4>CQ$vfVB9#@lm`j`-+i(e)VI0B+w|0NLTChl?wc(Rs2P4m%Eyx
zvrwFmqkG92gSb_uClu?|_zrZZVs8Hq0RUYQg13Ewbj5-_npYTphI8DzG3n8izHvAP
zv}Wa=o_HfhPTx3w{i9Df(ePysD3?Wn`SRKY4O%v1!*9Qv^VR(@WQM7YoFqP17<=lv
z;y%*+Y2jh#o5eDa=m`i4mJ3iQHt7tzRn`1t*}Fp)_T;<lyH8^oE1o$guLpZ>;{!T>
zEcWDFpC|IkNlwpXkC#PT+0D=KWx1-QTX<_(YC|DnY0djV;Z))NA$5?d$BmMgq^9`+
zq3A85!5x*^-;5JYlSr~;fx~o-H`r;)wAB1{g`n=tua6+pmXNi$0xQG7nGDTv51WkK
z08rgEA5S&kO+A6PJTbo37O&hlD&(BL-Z!>;(D><PwFubQ01S;<7uRT!J+8B*NQTtA
zIe5C?tjJgDsNaXSPM4#&&0qAmqDMu6$9--EMp}pu?8cXi_J+Xs@d~E+QLA>dZH!I=
z>S?QGSo@Irq+GJyFtQ$HusCTsw(#;r-bQ9i9uUly?rc59yH=WYmG5b^P4zo|^|5yG
za$pw67Irl@sW<%t#UQ0ggo~8RE5G~1X-9mxLp(61jJSSk%S*}_q}-Y%0Q6VD?g!uu
z3pJtnDhQM8c3X>qL*6Cm9*YYUDpkmr5@!U5kc>l}tU4I?bWz8;ZAOgn>TRbXT};06
zSbnTAm$9*N_Z*}1+Os;qwG}T+5Ed3z1-#S+b;`AYHWSyX^<EZ&RiKq<yH5Cju}ml2
zmE+&?P@~R(iAO)pM{uFBuNcA>WfZC75YJ~=#SYwX>&Y);IYis^Sf%5Xd4tnME7N?0
z=YRHFR^vnzL`6dv-H}-b6upikEqkr02fFYtGyfI$@C8#8?^p~`Mz2x1^ivPFlk(T=
zti6*)3XH5R<%lUXNf9G+M>%dezOF>;SB`TL6(XNae&d#J#i;c%y-}_aDgoapFiI%;
zwEd~Ze{sQEDqHqBmK&O%V*DX^$h^wSJcBgz9M|}1ZNM!f5c=eis48PuNSuNrL(K^D
zJ=GhJGYopzY#0<Z^tq)DMe1)f=f^~BCbEYme^c47I&YQ_%_m;IM_DT=d#|P;z?BBI
z@!{x6QE2ndGFAzZ71C7KL1AU}y#Qm8A<ony%8u=P398KqNrhc_KzI6DrKghg)yM->
zvPf3Hd;4!qfU-uB3-4pgIF?sOAcGEOt6LCvhB;yuM_p)SNVO(M0$a$~qX6)l6B2z@
z8{Jts1_ZfcoPdBaN7J$KZ@lBV6K`atO@MaGdclUpF}$dnW-B7ikeQ_d_fLQ=0MJ^k
z(i5?NSDW!>=<X_-t;(;nC)+uiE8hjScJfnY)a!maLQ>!F|BkV<tJ0Zclq(&OfN(#*
za|hf1@{XrxjCYqYDQTWP5|}tKbS4h{$XIi_mdxiq312X_Ne2=Z+UjVA@%w&T@rB#7
zeyBar)%8SlT}@D^%MgEP3`bdW2$r-aNVqncvXkgg6FtvE(=kiyn~fJg&zub+jY@M$
z#Y(89xjysjEZmp$>7IAg(t?Q2qOrdn6ccq=iZ7&;svb#Yx7Rr9K3_Ic%Yn=W(quj}
zI;=z)|MIyW_*xRPoxxwSidipGW#agQul_}r<nVmv3)k@4O>wX-Qv5~n`ot~S&R?P3
z>XtKMM&p%L1~_~vkF*+dBfAxqlTr=Uc*MhS!m3>ae+pdsI-c#towwqq$~nH%CRM42
zut|N9|NV1*9M9G+X;a&>Z_n$X_F+ACVwNYcwrumhj|<ZXQw_oV*225YGCDV}ePBB|
zE!4{M_OmiU`1`GVHP)AGAz6pjTQAjAk<wVeZDbtJ7T#u)y|;4&_{U;A;eP)C87Kix
zK1PnD1doRQj-S!6vaLl6bIA0r7oz(JWhLoAS5O#FJU_)<7OW2$JIPzYPHanKr9F}n
zhBRT)SmhmU?nqg}-?PO#)e%yBWp?!OS?hv15gN_Ijf<8}(aTu7S#|x7rwe|Q3OmOZ
zvUVeAt;sa*(b#vF_LLtG92hnZ2m=c0fR>%NPdtNE70|GjS>mSp#IOP44S1O%B*i-W
zOf_(~-$HelBc4+)^C6T)LaX`Jy8PA=k2}u+r{)IVrrPl%!Z9F!U$02foe<ep{b(Jh
zbXiF+T_qi($>Rkh9BP1EgZH(Ncg#PHEfPPHx9c1~^kNND;0orqA0#eX)I7Edd%c&|
zT-kPN_#k(@hQRX0@-ODchf!F>U7a*pC?;h*OKh#qGLbcNGV->$e7oqoNBNOMQnwmQ
z(mzL`Z6%oM)IXu{X<~GDm8T4ZFUY^8gq<;p=Yq51H=~>(2qs>9ORH%*k(2JzxmZWy
zD7c$8kz(X`D|2LJS-mskTt|_q^tD|YZ|WY|qR}`=%5bjdhAavo`~Xd>z0`jyv^;r&
z3RCIvSz#S1ZYuU`?wl0(dZn)1w@Om(v~Ma~%Bwe$z?5r_s`u;@Fz)^Q#+!GOHBffI
zsFA=2P?WSQDx<UzB=gYs<=k=bYH^-p2_J;lx7Oy49p1KR&0PLCn6Yis5A~JAG$J~v
zwa+Gt;#G@aly&TaK%jKq4Xh+!l4U-|5}^S=ZW<ZzZ<|B|=mE*`SQW39B%FrV6WVC)
zfwWYN+H@CGfw`BLeicx<bJ@{+$O$ZZs4+^s(Lko3XAD4Vp}U4oUlbrr70KfPbOI{e
zG<(Oi0`*HGLBd{n0TJ@m`uQKmi7G#OpWBSZ$|_FuGc^)2q<3*BEqNw8)d&hN&)MGR
zf5S_S1NN^I*d&dIoBSY0!=Zi(Kzd*j%@L)CH)*%7Va=gjcS!inh*;YLadq*`e2*o*
z<$TL8N~+y!R+kA**`k$0wlh*PPea`t2ZA@uk42^)J~lM}jtiCay2F*C^+j@o2|t}L
zFJ8`OjGHazbx(}<r_SD2aSl`il`{iRC)sx~&^iWS>Ij}P3J$XesukEb-w4K=stJ@^
zmFWy{N#^X4eDym&&-coOwX3s09S@ZTUVX9ngHP!HWjUlH>wN1iE^{SoUw+u<`Xh;P
zt<0)(_tuX^4r~`b4JAxd^!P?YCj;f$FdZgTL_~a-HL-H7(=Ee^N6@yFE$i^=PYRmu
zd8n5ibqrb8o3<(!N#H$Iv0r{#wCb0=SIJ;PNTJOZgt!8MP0ya?z5&vWfmkwcAHK<J
zR)5LR6lN3}r-y>qEM&Db$x&mc(h|`ZVI995*C7gw>cZhC*d?9waOSL9+uWo}7I1Iz
z>_+mm;RCq0&Is$RO#q=NlrxehJ`^Qy)|ZR6<}c*B$q;GRQcAPe<<5BcB$_)*<}pna
zzs%bS?)^<u>Ue0x(L44{LoyEMxO*(QwIndkP!0vP7`e~%^=~Q@t9)J7bAlWmszL7r
zqR%~v;nWnrJ(pGu;Zal>Cm30cTQ<xLu9?M~3x}#_8{eDy#*WAt=<%tH6VJYw<Gtiu
zo<>v`W&PIu@q}<Wjt%hlJdjdZ|0+K!+_&UEP?a1MV0J4~SDkCX>eY?K5^9serR=|@
zvxF1jq2BB#?c`O)j4`r~(n}r<FHMODcM#R5OfSV-H^zIj+-(@huJqFO3ieJB+&*lq
z>O;K5H^jG3%Ku>w)uP0xcB};{#*+?;jP16h%|QDi0I`B#?1FxKMDQ>$O@Dsr#jWRx
z<(bV+uS?OzNnoMPey3%*whGUE`)CPGxn=H^7TZ(vq&YzoI$QrKYE7;K_{j0hHa~ZG
zTPdb1SzA=`4<{9eIx>%#2i7RGG(wW3Y>o#Yevq-bQDPXUi(z1anh>O!2>FUNmu`M@
zPMjtHxnI865e%Y6&oX$m6`*yVTjS{jE7r{yZ&7b8ex&%U4Fz+R5jGM6V9m=EPqO=?
z`b;42N=u%fdAl-eyZ6i}Z{enZ_tiFWt2BBa(^i|T#DzK|c@ek4F5Ebw{ky?|xrKCp
z(Rjn9fpWIsukapk>hLDP&7X`}XO+)4GTjybr>ZNU>*|3|%&EmbP@}y&!LBC`RbD9S
zKhhfpdTAC!jC=)<_ei$BQo&^$14~|6uw4WaHw;4GHQ@VR@;4`)7gkl1#PBl5v3oVw
z%YXPl=bd!PpmYhh9kAq{Pkb_fmU)@g8@(C3pVLIv)!>?C>uLhK$h=<2x2bhD9JdP)
zS`IVlT0Pe{qyE2tQ2TpVnt6=qfg5B*Dv#z|@v#5As*x5ApT$EU#T+WeBi6R|@MZM3
zxpyl>_86b-yz%*mxyEaLXo}^HqU^VdmjC_Jbm4^!$CF3HHoB0Kh_Rx@WT1TbnUnG2
z+SAuXmxI=&yoMSA5hLmM%oSVu8i97`WYm$7uQJd$SY9F-TBZp7NDv$WGF!&;TMl6(
zr4XnjbnYT+TlbdPP$|g(2+=b18@zVp50p!05FEh#yQuvJP}WIYzyk?Y#M8;ZETRjJ
zGeX?@bZoma!xgM$?9Q@ZM{|>P^3SD69V=e`$*DEtbnT&{_{|@d-WC8wwI4td>k)PE
zUozH!wDW)KOlAvm$4P%Uw40BaLxITR?LB^?(gFWE%I!nq`%-JpD^=HiUa1`GA@m(7
zc38cd(qmBeJLFr7DkhmXE+9uv`1zyYKDHGgmPVN+IVDl+@$~6(bVkj40DcqErQ$Yr
zobirf`NgqxPYB8Nz~D-rAu%f+Xd;`M#@)Lf+I_`D{CMWncEie{;uYJG*L}}mv)}r@
zW}h!44!v(EY!O*PcHR->rh46QsEz}D(;a@{Yr4EW_0VO+jq3+v3BL7Fqrk9sx|g)u
z4QrO#Y?o+}Th@a?B?@D`qsiV8X}Yg5^eBO*KsKP>7Sc~SZ29qqNdNV0g}zF&^!(pH
z8B^K3+~8Rc;fP25dNySU!_@BJW~AwM1;5#Ax1bPx9JeSWM)&@0ug5=GSziS4tXw1T
zv(K@T4W*`N?o1EU90Y9{zJQ31-)X-8lr-&Cvz1LA3)r^ca9V!<<)RH~hTYd1OGz)s
zg;|=ubSVLGkP{@Tag@Q2Y12E2Y$U35YoBB&s@qf*e{S2+F_yQF3gM(drm13?$ASh7
zG&Mjm(_|H3VHamiwZ#%jNs=GkmHS}W%V<r(6bb~_tz#CrjZ>c<pFv9H0SXCO3*hmr
zTpj!A`8J!D)Cl(dB{ud@n4}P;$(rHH9GY84_t#|%ZG(mPPHuJ=kH_a6!aWnj3Lrn7
zw$D7Yp^7HiZ(K5r7OdQw*1{F33f?D`ipw}j(Pot{kwc>$&bJ4rZ@R;3^YUXo`C_=5
zJC1n(vn`ZV<97~fxMao6q9vUZ5TAn~id8qHCo2+lHfDPI6KCl!-}n=?MLAo_-|z<Q
z>pHu@QXHR|)5bO|g;K6>fa<Zy|7e7GfFa|50wROJ&%0%0U*w}5_Vqi>^@%;kcc9d9
z0pZ8Ni^efp0eH{OCh$I9QQCyprq1;Cn_N?z{8z3oZLROnXyjo;qd-AYPL{iBci28$
zUrqAG23Z&Ob!&~XMHk$|X<N50;m{z(#$k^xL2vH@f(HwW&Mtu;a;>4+BA{+|rR<#6
z@V0EmOC@Oel}xOtSfX*@U(dF2=C%flJIyAMQ-CmaJMS@+X1FceSMKgtrn4wsExG+x
z3R#&J=}6N<lHf$Q%Ede+P@&Sc_UyEQHM6C}v4+o-do0dl$8$&KDsFDyB5XUQjyWwU
zFG;uX_A?vuZk<tZCjKG;6p7dRCGPo*b-^X`b>OyK@e-I?M^y8siV5~p<bbcwlDZ{J
zUDFQ*HTPM>&B!PhsBKmpi|Cb3c(zlu-p%b(i;>B1&|FUiQ_r6vza4a>|C=1^mLQoE
z8M2TDLtZUZIM!&v)*48#9j3_DMsxq2P5b84<yX4xD_x|M%=XRGrMV^woFGS)BBO@a
zk}y|gdAq<EXt5BV^0$m%ccosmv-^K;Il8S9P@f^o)gkqo#t)WDl)cv??%wu=%R*+B
z$8XKZY7!Qu?tM=mqA9k97dPLFji?3YiV_oOvxpjrjVivdQ%>nvSe2Hp8shPWyZK{K
zdRY)y?Xlq{NH8NLyV?7tjnvOOK+ze;$^0vIa=20Tb;?9Z{R1P=*Y>hRhGk5@rP{qy
z6P$S|gv-GId7${>tyvFugLJFq41y;?3|=GO>faE?ybLg|m`VA}s9uLUI;Jf6)}lyM
zwDgf2DOS9J8p9+1%dd#C{=8p|2X@x$#1hYRKp<vgpbL7_0#ObGbOD?#3lza33s+^d
zaBu}RI?(d&%4e)tP8|?CjA*7-CNK?Ux5H6ZM84lUybZcf-?|P!H?JC3<Cgt5Gmxu<
zgpyZ0+j(7gyB6iz47s_5uG=v~Y@fUFZ&P(ki*`@{o^aOJa%JJ-pIbO0F0615^2T{d
zsFQ*gR$GU}*(ZAv@9>A>98@~UMv-A^vR5~wH8dsF`M&HjZ$0RIGDkAe8yBAQ?G(Q9
zKy@!z{#u#RA|45~v+A*7QZOA3A3V;HE69bm^G50}1wN?*(wf)vT`vyjEf!;*-OH>l
zQx(1A99DC<+8?oxPy1O`L%*E$CxRAzC@GLd_uDgKD&^*F)+YRmwdvCEF>#Q~>LFZ#
zB%L3By==I&Humq_N#?yDUq|QRdKaVXIE@|H{b%dCOV+P8Gus9B-v64`87B0(KapX&
z038PNRsl!G-24V2u8lg8VmYNyj+=ls2-|0gGy$d=Ejx*~=wv$_-9zPgRv}|M1!`CD
z^h*IC+eAvWT1DM$umhs6PDSgkUcLIUiXo~kVfrS%sdWV~3msM&Q%)D?PNXJA%~SR+
z?}M5-#Lx-4Rwe5CzG^5@EGtekAuWpg?d9+PKQ}3g;cY?Adn@gPk|6Br5qITPT^V=p
zgcfC}XO_|?DUeK%_EsF2EvWm5Yf+2g4A+LHKtzCVWi7JySb^IQR(f?{dmad~;t1f*
zS_pK<%jwwakGX`T-EY6HC%+Ud$Wr5RWAgbm^uk_LeS+Y5Aq&x=K&=efZ3`?%psui;
zEZL|kgp(*-YkU4WPMU=Un#OZ^uoINw5#_D4LI}+3H1v*op98AASRn#v_(zr>gR}nV
zI)0@^(C+k)4s5K4y(G|!wG0xO4&9oi#-2xss&vR;?eCm07v3Xzx4b6&V0WAt@KXE!
zvutMy<@wlEk=eoU%cb5y)w`M5Hg~Z&XV15L#-Bq>5BGz6XaiYa(@V5@pC-q?Z#Ne|
zPs#kX3dh*W<6D;%utSC$F8g%~ZFO`z?8F&4lM_dqld_^3G3v6~71E^#&&*kNrOLG+
z=Ir|xnVG$+Q4R*soxid`I?b(x+29fU-<(*4-$}Lg?;}~N>G4Ss=oj3ZCz@p~=Hj#5
z%(dv7?sl@Z-^{WZbTXdOnRn^tn{sXJ4rC{c+j8%()5&jm&&`LHL;so+1PAWrkr8Q>
zIPT3};oFnQSXL5h6c`Xk;?ueRzeidKkcNh)URO)s9^|VrRYf2p4nf7i{j-8{SMM%H
zM<mO|-O@J_lMGy>PLwj>wjh1OgdEbBA?4R;@C%<L04HX6P$X&y^rpZMbM>t#xTPn3
znPuS=M*eVTbi*0(7$FTLTqkM&i6!fg;LiglQ$AOMVp7Ju6vd+MmdnB-+ueHlx!kC#
zJtYp8`qRsz%KX2k#m)MqlP)@dq)sPasYRBNp_t5lgz6rAdEC69Y29MNQG33_rEWG3
z#AztBE7+og#1LFrm%$W*urt`0Zf^&MTx3sVsWYK*nmRfrupiXw|N5?tXdOeX&-JTQ
zoJ?psi$)HTJ?w9dHI#`g<$P;@XLfz4u$DOOaf*?B@?Ty8Wrhy;@8%#Rw1_TKp@1$j
zYV68n0?ZWI1`hQCW;Is)0V%kGdIp|Y&NN|i#@9BdN!A!iZV00@fXqL8?{OievY)-f
zDBL(2<UC>C*C+he8PM>8#ES;eEq3XgoRsJpx(wCtu2bG|>WX=DGDUf0S_99%B7QOa
zCTyY@ir*EL4txrG1|lYzb&eO+^nGe8EJ(}zxzI!^Ks0}r`FW$%+#Z3ai9te;a+vjm
z%!Q1iHc#%(^C6^v14Wi=B{UTyWl;vKAFoCE<Qot6TdFT<svo?ms>WtsqRJJ;hAYw~
z){`)IlalX;?JXcg8)y;aB&NOY4zk~c!)dY<%I3#IZ7jpko%eGIAm4khbcE$D0p_DQ
zoqtmk)A3a~Wsykj(m(}db?+sHgntv{tnKltdoCHplf2n=3ROOC1-;Yt@Ci`vL>n|0
zeQpNUUj>!hj6>}<%cV$P)B=*~w}Bqy9HYtXyWoOuxmpNB8Z$NBf@9WYJ(43xc<*}^
zxesaageAe3tfLWL05vgme%A=DU#qW70|E1?%Rqn+?pXRrcfj{Ibuu_*lWhzdJQkiy
z^J+O|Ap<6Et_56=T)PXY2{|^{m-Vf3lRCBd00(1q0~a`Xg|xe!J$ml}VFLSo3wwPb
zYiHZANQf98@TLN&lE4q=NDa+30Nl4;C8;lfs7}#Kt(5|TO~X$FHxcoItA`a6PSVY3
z%S1f{37}jM6}Yg2Kr-E$?W677@gp7m{_PWhV5>wL1(LXpe)EV>XM6QafMgC%0D$lt
zgE$s7_{z@tuUV4ZzWvja{<Zu@;cS78I2U8dV{uK2q$*4bq*7iHE3f^HbrHr4p4&R_
z(0uy{gfD8`9s;{e9xKIn;%vs6oVB-jlGmQ=MG({H@-~fc2cWCB%p*mX4~*#3-~Zw<
zFth~YQ~h;(HUYhM%C~yw(7nka(=j77?L#3)Bt&hM0+A5E^1=T_;OfaSZ>nYLgR7cr
zE1C$>9X~d=2b9pDw%5?E(T7pP(hI2D2hk<zPm@tLgO}@>oW5D^CK32ogmzh8Z7adk
z$Kn(}I$87Y2xp7|`cY;~N{vVYK$qS(AJ3ikVkA#^k5{FUy~iu~aomFKfcqp(k5v~9
za(%~KUfj$~)?ow|c<*9)4+YIP+PaHvTA-kXV@>pr`3_FLyH*jiuwzsEx(Us-(YD)v
zBnauNPqy=WAVl{IZG?!TB=29-3=41B|H{qI#&CkZg5*)%bDqVw4VAb~IO--OZtROs
z6I}Ij|3VaQB52}CKLM^R1OieW9X-pM2DTub?^#2OX^zw@7T25jOY${^)DS`vqxz|5
z2!fb58O3|Xfip;G*~=hR;aroLa?l#36$*Fm&Us^OJcL|My?jYd3h-;dZQ;vwaB=qD
zJ9T9AYR#||34+rwLuJ*fbc+~XY}<njAYGwST1+qLIxF}?@!<hbb$bT;Pq`pu{@SZo
z!)nWFY-8KVNh8Qh(^*fMu&Xr#;zDW7?{c&BwvOpMvi<g_(>+Y!W#^3U(}DU9AhZ^n
zZ1MKAR$EQ2xe(W{^a6IA7136;p2SbaR|M@j>&BQN#8r~kfJa&+>{MSwe(GvbML8;N
z{kIxmvbXqo#k`@5we6YJX8r=TaWpw6-|1fa_J@5b(_iOm>x#WKZtmMASIEpVHmhLk
zt$(ygmZxY4lm&<XMP%?==u9tkPxF8SF=$2w;3&4LS>d`8CoV8m4@+KqEF5k#95r4u
zgT|#5-2qN4*6M@OuST~PdWCj`qa^|or91C!mJ@HF88sE>7h-f+gt{bfRj^vy6HeZf
zN=YD&1Td7;d^j@sbBx+QX@(#zh)_cz$*#-NN1~xyh<&;KU39&KOh`wpS5QyLk{0WB
zT2!hYk91^K2ptf>i#5x*0iCq8k%GEIQy}YPhA{pM(lpb1`PC$K+FKfwyV?@vz3+M6
zGWq;Fus%Zq<^UH52Hw^2fGQp+`)GzjpfSSO@x^D8-i}1PN%W%<lCOvK8LAr8molOY
zBD1Ye1m15GTuZbkwJOADYm3IZb4w>Ha3g_@<O5K#Q6p`@8^ld-vPML3TqiHfh$geO
zuWhy6Vi~OH^`QAS*T`797WM_8*G66jwR$c=9gc%?Jfm6+f`)EdrxVvWo8LAx-gwkL
z&vTz-R++bD706$uVx4y%%~!z&am_*V=NC&ll#$fWb#TyxGe)%g3T%y{Zd)pa?<?fw
z+c))5w!XH{Rq#yrlN95fzHPbdT|T{}XHS-oR-8}NNy?`73-bnN1l;Q|#yx_VPe@K>
z<c7BeXvDEDnqhhR?>c5@8u75$j-{QLxRh>wTyJ^$-%tJ{-SPLPhlKMvW8G(`+T#RK
z|9iyJMCM-UW}{ZHrl&SWpHf$YUd!c$;mSF?I69<~enR4ZOkaDXRpo4pivgRMJ^KS&
zm}`lXPPe^Jjb?gWw0WAfJ`r@xN8~1D2FrHE>HDE9*4&uWTBK9B>FTt*X+1hmjdbk2
zz9zK{&{)d)UGXH6AJr(<-w)}InojA_S%Ri;A6+FijBihD_hQye_H@T|vVp+Fvj-Y~
zk)GpHpP7&oZPmNQY>ndWvLuZo|A;jnLnf2_{*&B2Vh*A+5dd+qB4do%rueANLgfOs
zf1Rt{><H)_Vf9&+a2MazbmfRD6OVM*sY=I!#H`25P*Z`KU()qT2+z^C2R!0`7PY^t
z{?flfU}t@HoB$g``WL<6N^jVQpz?470zk0p$F{m}H)X@%*X}xCUOc{xm1{D=d41n}
z7CY@<E(!90<W5N1yrc4%o9VR~=N|0gIZn;|M5^f$<3W1}u{S2u8(U=EXo<BDiSGto
zue;w}Nq-X?t{dVNoTLEs_5RD})?-c+D22%E>q@kgSgSW0!BSdi%4Z~iXLYVEuf|0<
zT-x3>Q=1l!M;FH9TYue@TBG^Dn-z`E+=K_BCG`yVd7n`#-v9Bto5?m4(vL9lYJ6-I
z+5|rIh*8q1{QadEX|^Wm3RZXz@N*LR`6zNbg%n5cF*?hduC1ACN=(v1K1`?h5n1qr
z;MIqTkCZEdw2$p8&h)k+BxNmuGmv!5PZ=k$4P*tQ_9Nc=a^DmKjL7CH$WLz;wM*v4
zCIlvmzFe*!QG0J+kogNoVu@?js_Lj;m!zPN2Zxo86%Abat%ucFVnR${DZt;61>B{9
zz$2CZ?TI+Gls&VMK6KwJ*TjhTL%pIr+Pohr{IfjX2^gxAUpv~@6cvZl^5)*m1$?TI
zC#V!qEm*{u2VC=&jF<C!$92k6Rwhb<bj@?FjVPx=G(XWbr9W){U2_g<sCY<wo5)Bm
zJ=E`SFGp`ufz!H_rl-1P`gb!{DG~7<#5C5q^Kqavr2WB*W7%2do$ZdE49yT~G(KN9
z@qNTiFarp;765qU*NZnPu2n#nP8RUH8NEGCbzM2hh+!FC=x!qJaBrg>oF3A}Ju7#y
z4`XjEkhvoR-DDa_<GC|)ZP9~Yb?=a5+%x!Tya^W@qTpxZ(eJ=aIP%}gD6qR#WP&(q
z+W%OOl4gd!08n3!19W71*|wkF{G<7OktLpMT`1B4Zbi0gL>t>)7;_3dAnf2`rLoQ1
zQ|XsMFg9yGwK%XV%kF8${wdnSEtBOn1(NO+HD%B8zkf`$%Ih&3#vKY%^8$bL{hqBQ
zCO1S-^s*fS223>Nx`J00ii?JFDV9%woW^w^U2m3m9GD*a|0Mw7db$o-$sm@OK<V%0
z+zVoevh~)K?3-(L70)~5{(`>VJMa+5C$(L^HIlWC?$4pcR+i;#p${a!J#LQY8j{8l
zawxD=!l}5i<bJoV2Ek{LFPMIVdf2oF>p$fHCTy$wfJ5I8tS0m~R`K2k;5MinOnOo?
zR?`hH<X+Ot2F`4^a`JSm?hG0$h|yMB&V6XR(0A+ti#SpyE<EKtg0U#Ay&B63=6xtE
zhv7t$GcE(e^~_91_mQjuRq{#p7pEg7VwFi$2(gNB7u(I-M1ikTL|j{sMZUvIIwdGl
z>sbI~Bia}SHC@)*&1y#B_(3%l3boT=wn18C)E{8)Vf)an%eFyz#@#2_Evk+ekU(h+
zq#d;bTuBn8rZY-;=@5e7Okt0dDrTy-u*0c7x7-8}ESu!Sw2saiQDo+R@k|$T-%dSg
zTrd+Dz@K+X;s&<0jz0y^DKY~Q97O}2QTCo+B<}Dn&pqUc8gf^yX70}ku*l5XW0UFT
zCsJuhFh~Um!UM_^yPnreE`@Z)ghJwO_4m&@(mA|OmeT5cH$lwlXKl4Bgv-tEF6^^(
z>gQ&hGHLxGce^E2A;gokr$5|#-G&fX-PlNUp#vtkOioLiBx!x%6GxAI5@@o0rztV)
z{}+J~yaRn-O}r>}BG8$bHY{Osl!G^WwF^Feo<Iz!`}Os*Iu)8U_UP=tfT9h!X0O{7
zyzajFLieJpm?g<^Q*UlNARM5ly|OuJjzJKeLcT(S=odo=HjI5{rKK{1br<dDk*0c9
z$SY)q3Vma@04&j>=c+7%*&1zZGgj|%!daR~yp-7kTXPMLvj7IKCM2}dQq?j%glZsA
z0I+4xm`o4?)3=Kiz?hPMOW$OjZssC!G<%f#3!XQfk1~*lfP~q9Dx8Ni5v+hRT<XnP
zlDn^M3ji{fPT@jI4-~Hf9Al_=(m9~BxDg5W>(a##b(#+#V`$&{Pl3+{M-K~@EbHG(
zRUO^#`gp9%gb^f?)JcI_V`Ii9M2<L5P5RE|vkx%2HOFNxAIM@%(fpi>In`5LLp;jy
zo7ynDyS3iY57c0I!6tq22T7%6^3wUI!HIG7`KG=C&#Qob><IGHl^+PwbFhmu`jI#-
zLtn+k23k?WmoyD9-()*UUdvEYlv=dOtZlM<9>A5*i~eolo;jMb!_Kc@tE7bjN+}qF
z&w!HOodK>AxppJLo0nL=vX;&V<N_TB)RkwwY-_-g`5rsYAYTB|10^#%bQzIma?z$v
zw&q{Ye1^bPMm!C;FUJ=WzZ)<}p@V?_o8s98$AbjyA+HS!;J2hiDw^a>b0e%aCrFim
z2MF3b5jq|kjaf6?hG!W@nmB)1yb@dr+BEJGwdRExEG*M3|AZYzb5ZWZn(QSdC(3_l
zTiuAP|7lKZ_tZGlN9V*RUb^Jl6w;prrs<EoNm6Y``H8^DpCkW0lftM0D?0}Tc6ISJ
zs$arM-WnNi_yd6h-LwPnZJq_)el=27^n!Zn3b{kBBOM+QQl_U)2Io1pl;}e(xf&32
zk6_C))$K&~A7cKE<J$a_epubHHl#FA48R63jYeJz6`9{qyBG!L>`1q8;5GmSF()Ki
zINYM0{oBCh;#>khAyTfT=L)Jnw-jDo#o+A?5-Yjx03VVeM!uZ(b*&t7!vIkg7hRMg
z)&0yg6HT+}+r|Z*RcrFXRaj@+0=f?R*8T=LR9@iq#<%J9jjyL{7gjD8?}bO4#?hIp
zlH;9dX8u86u?!%NuIo0_@c(ij5x{`sQz=%Mb(1Rc@X2RSvd^}IUUtFp)ho%F6BOqV
za#acqmJ{whox<VW3akx~1q22O8Y9@qRrv+a@EJI+$9D(5aW6vJ^frD`Y^?yv4^Zb?
zmC`o0^!U2H$1eif0pM_s<tN+$UQ5OO7IFZl(e6<8>_xI8o|kTg$vc*=u=M6xj+l_B
z$ce6Q-=V-KuXjn@ZA8W82I_z4!;g340aVWURXPjJdG>xHugG+jp_TuVyBTn`W`L?d
zO=YP5-h5?^S83CIix$njs$s@oHV@y3h6F0@Le)MIUbKsEsGiF&d7O}FEG;cE;cQoi
z5xOk8iF{~VO-(xA%d5Q5U4o~h&&$0VAB)S_w>7lM>`dd09pHKlVyW@@)KUz13}~Cv
zX9*>yhRBu5(JqFW2AmemqE-L@)VOxVu0K^D5dqv}j7n`~X1tiJ@V--#eUVe3yEheJ
z-FBld<{?d0fONxIUMnG)+%`KpQ+1?Wb2~6!I13477XmmM_$xrJO+KH^81O~_?9V_J
ziE%pjT5jG_#KqnBx=T7z8kcf5Rb3RxnV}f>o8Y-SOZs1@`wk9u6(U`J;4f_in&*f3
zEc9j0#(1Sy%5O*vskbvVeDcR@(~L@NOdG0sT;4e8vXgKw_!q{{xm6Lt9HxZrWOr9N
z80@13Rp8iOD9MLYgWp4{>epC(K_#4g(`P+gv$RGx`@x+Gd&?T}sVXm;rdE@p!fmBD
z4zo@iD3)6jhng?6!(s+QE^r06b?ocoelti~H;4(bzT*6zA>S8LE8oq=39c(qU&#%C
zX74B~k^17!owuI#$L9n<m+mJbEk${FQ>|5I_B)vSoNekh>kESv&7<!9KxS{lJhWf(
zU%rgK3eA2`aJ4c1VA0NfJMKtgfIdH<e(BloCn;axK|tiQb?GeLH-WuW;2@&89gf~~
z5*OS)G7Ee;9lHkj(#iG`teNIanet>cfJ<8e)M1LXx*3`}rI#G53}<^Y$xcKnBGwMi
z$mi(P^rb}Bvfc_);9j!JIfeuZ5+T<@q4}2HwzH*lW;Ta%12ev23-56WV}fe;ZM*F{
zA787|`C?Z9b9BeTO&RFPVtyEMJ?8^7kK0W(VE=3H<ScFi%~+0t=YoXbMJFpOV#!<O
zKD5DiBgge7$p$oOv}*HKX6T6EMpbduz$vf3^h;XxlD4WS$IWYl7r(yrcne*`gevDq
z3@0i?K99eh(|>)HWp+fAWsCIhC#C(hAV^x?<Fu-AaLtQl+I3aCKO&-y#THtPD|Rr?
zw-5a12L$Fb9gn^28LDGZ?ZzEu^fGXDr3uJ=nh#6Us11Txzf$|(KRZs5Zwq(_Dp@*C
zfLtt!b67p(6|@105dxa8qEjZp-0F7J;Mi)Ywo<RDz*N-KjH`W~==~{`3X2KYg`Ym$
z3Hh^g1@)v&6N`_C@v>b6F{}3W57x?0)^T(;M;+5SD>Ae^eCmWhD+Oo)HwH?17fu;t
zeiz9s_CNNLzm0H+1qH4qqkjWepsj*;MI$X$V23x`GMG?8WLwb^hvFgEn~>(auWio7
zDh<I2Vhi<FWKAW4Q}Wg+r{Z^Awm8;h5P~5QNKXtvSehAfrF(Q~>H9sYQs4f8Z<G|B
zNnodD0Jk;nzMw_OOdPI~1kcyvQBe7RES>dVQ;qw^eH4|H?vjm0K)ON32BW0~gei=Y
zR6?4EZpP>wGQvb8q&r8K!2|?!=+J=!CXRf*$It8g2k^@{=RS8_*ZZ=}p|@`C1+W!>
z6Fmgl+Na#u>WYs><G1vF(QPV@0(O(i(!ipqbUFwCHZZZ<qqj#+Ah`2rC7-=);6hFS
za#o=<o%AQm&z{vuVY1K#RwOtCB>!&$2taEwVQYXyf{i|AtfCIR43v&mG*6o~TmtEy
zIPXu1<fP0QVShmO=vJ%4(mPpFfJi09Xkq+@qI2A(8Y+v9;T$q!9}Qb}cMQaH?c8U9
zlQe!NrBXkB{cc44_|56R3(hB8?p0(1_nWwiCb<{(4_n#3BquU^pQ2(MnJ<OIz-4=C
z&9cf%Ij$q_g`({Wt5d@6E9B8Y@mQJowBqELg%H0ny<ds^lWr7i7jK9_=^%>G>gL}5
z=tu6ynMvwRgra4hg7{GRk02G<zGCB?ML!KCL#emOPZ0^D6FItlJeuEqzEq1m8~zPY
z3@UOP!hW53Xp*^}LC5fXqU>OcPZ2z(HWPBWMcp}=D~-Q4X69GSp?9%Eds2f8aOTIE
zm;I16#x+|!Q&F<ZqgR_Mj)Z&pG0S6#v2<8A^{otJb9Tml_5iik&lY<te2c4f^LboT
z-$u!g&?y-sUSgfEc}`m%se{Lk_<r3qCbB($y##r$l|Yp(Bj1?*^qVl0tY%5a7%Z8V
zlC6w;FG5_7QjEC%7J5KNh+(;2;#Socf<~h=RQ&1=IeR$PFLnwUMmc_ERR?VFFi(?p
zK8-1aAK+152<Iox1%8mb(0|MuROPk{Y>+5nYD-$ehY^;1+RXEL&Z5@YM;?dw#-Wg8
z1aq~tD5R7ymbZo>M1cq2*EDUw$n4Rj;17k576=jvnjX4{mVwp?hMhEhQJ~2IJnX<N
zc^WuW7n1kz1S);FIRMr+AVvYi%<)KRFc!W55WJ#Tz|EiEHISz%bN?)7+&GkI7Q?MB
zlZ9<2(1(_y`}dy37j6BGvEnEExt`!1$zhAQ6ia_>RLukE4UDoGFhbBMREcg-C(P<4
zGF+I55v;s;5_f3~ZT@Lndpa;I&Z`VaIWLVn=C;MKQ^Zu^zRd>a`xba^FH$52|6Q)=
z>npW-{}i7xG$);zo=Tmw9rX$iMr*U51c{{wOkojDQg+?`avj{-w)$N=jyg7hrKfQ>
zoC-4<L+Py-#M{`6RZYBar4B!`>B_GE??Qe;v<CHnOn|KJ4z0B!YhM!Woy0$e)(U&S
z!}x$;DsHQ{{fYR8gvdFK!i@)9-zF4Ou*ztUO=jgXr(pU?x%?oM%KG<*<$;C3akH?^
z_l((KG%mjA52e|Kg#JT@!!7x9U6PdI+~hC;dpy4l{&BySKJAhB{oy{SXz7?gQBe5<
zUQFg$Y3!(id0~u@Hl&<mXuD*FGq<vsB0E)}YwFQH#5pXMF5YhN>FuJB@K+mtkty3V
zrK@aZkEv|qEl%qR<n}jz7L(^{o}fsIKE=eK_tWcc#(wA~t&G&xuwV~dy29r#hRtZ&
z_nM~M{V2_b+c{S#y3bn80P=25>2J%6>rN+Z^Z}WRGSmBUc%6l_d|-G$A>gL?expPo
z%N1>?1NKny2zecS#Rjza+DSe{_ln(jirw<uMcoWAm|ymp#TU+2J*B41o%8WsV<E;c
zl(aVKkbTG_D06|m(=rTDYhq~hbCNy+J~UV{-UV11dYYlp`Ov==?%p#BK-Y;BjRMi^
zBvrIOAy|$>LV$il)xN!1Y~K>e$|CNO1`JJ}1_LNt*ECGJ$ODxR&2M$af@Pu=b*HD8
zIkbu6Z~AO8X0PE;r@smcQl9Zo2&9)4UCQo3L|oFf+FFdF+v=77N0YV#LWS*TZm}oI
zRDFZZu;YH8Gd?_}<sOU%^<O4=wdhp&0mvT{t2X6hp`?mX1&5kSjnl`D<nOMrrU(54
z&lj#(!1G`^4{PLn(&BXGnjF4h*hNINK#i9iRN>=?uO!>xp5eyFsw^*QR;)8b*e@co
zql1HS)NR0h$!Gu-XNy~}rt{}-mrL_^hUa&u8TwOW%lI2zmXjZ>=D8dI#$*^M9R%xu
z*6f*PZuZxPcKWaI6#R=#Fm5ECvmbu|*0WoU#DLDS3RUB5dUdkNzH6!ITlGaEU@1Zy
zP7D3HX}w~Kc`QqjC#Gl_fwr`_M5Qfj&r=qIeMD#bA@53~P7Iub{6*Lti|-Yxr=#u6
z$_dUW(|{$YZx@r8=Ze&@kYJhqWvF?-x*7wFsy1v&mDCR7e3emR@E(h(X3J^am|wc@
z7t7(7U?Zd)e;K;FV;Fd|$L?%_%E=rgxw-@zyz*nD_I?5V)QRk=spS^a&=H%$eBnm^
z!>1u<7}jgfl$@e7&f%D1yhHOD!$cf31}{v#v$(sC3^CG|XI{|u;cas8ChK`+8RbQu
zF}B6wBV+MiNTr{@t7BTGj)d$^bo8*H(tc^goZ<mvL0YBF!X-`C-x6+}ei)oKZ6)3O
zILw865GG|x=reHvn&PNa0n@NdypTB^>2@zG1-u2IJ^^u61DCVpXiN)uU~0P(M%76t
z<%}fGk6-fmjcpP2%gV|E@>eDzuBX6plOEo^c2D3t-j!(fh?DDYa$fa-)Xy(@zVRK@
zB05?&qdmAAMRTDr05hfx3>Yl#168-kSSkjGYF4;t+N9G=_Z4tbO1b`3=fMu;6={Ju
zm+B{<Q_h$&#zW#ZPG2Qab>yiI&QFEos~%<McfURJQA#w`K|Xk^m~~*8ATkIoel+yS
z*<`J$zOcEVN(NOLza{ZWq<H%UHox`13k=lY*PHmCLWL-ZXly^hiKFgVpmJlYJw8H_
z$*>8qiMi+Hb8VOLRwjq)&i~<j6DmFtZg?hUk}76<SSl^!Khq}Z#}2hmEPtVyQIyNo
z;_-lhFfR@WQ;L_8YK9L*imq+OZFIa^m=nHS)_uF7(*bon)7G)-?)8AOI20KFtQbav
z>md+!Q4hxhNSsR!hCL$2;!7JbFSAEds_%^)EN3`-E=840oG<K^P4fHSZC{m0+g`73
z;kW<#knpLBc9`+)pzhV*C0biSlCPClddFlnsJ&s#_wte%I|R0Qp7Y4sJf~<L1kblF
z2H3tAq;qmwA1l#N@?yWs>t`iAkAd~GPXWE4`GYO<`BZp7kcQYz{rvz**JL}xswXHV
zJaJ{OMl`~w$#qjWu=?d%l~>%MWjq^R&*<lGa*6a7cbjGW5@Mlf?pV`#CtZzmhv+jJ
zY4rGg;!MBW_qjIH#%6hOfpfGFKM{34N;%PnI~4R~4D0mccL0yJ;SRWtaW$jEBQWs?
zN}e2%tg{k_{PUz3EBCezfV%!&F&u$KxREZo`m8x~B45s!xnekn&x)TMnlA9{1A%%%
zX)kL8xh7>~!1FG!LM*M54t-I%<f~UeNJWGKo|W5BlWMO<POzA>5*jz{H~JSd-qEfj
zpVhr==3KmrkTdenSjAi#OX|%t3QK~*2SfU4M1Z^?kbqRIGw{e+Vs(U)VmPdJp0g^m
z(sxM{`+y6x3>heL`BMrjH*M|aKN;w|#u+_X!FKVrDGm+^9nVevy)Y+`rq(|TpD7L2
zm}~MnfGyFfV!!{HmA(@sy~zvRZ@v*PmG*>s<=PGLH}^}!S&p(rsI6h^to<uX3B}}t
z7rdgiH3421Ut}Udv*D~?i%+!g%cTao4V?(;919l1%wL!n(R3^VTgB*#JbqJ^?*oJY
zx5p6xePyWPQbS+X15x8^!*XnAXtto#G%L1Gx_8E3<#D~_)9kN)`=wmm|7P)^cBUJn
zu|AzZfOhraRkU{77;z}TF62CM!vDzI-@i>uCW_zsTjAHHP}a-%$S{$AbG~--^w6=A
z^~2z|NAhb57Pk1+O`+4cf^R7i0>4v!#&GOumAaqG$Efp3=oid8&<?f;vDm**?nrn}
zWXVlSiYYk}QwN3F<q;(R>CC*c+&6D}2TFV%xD_{h{?-bUW@t`V0XuZw0#ZPk44Hvh
zyB`|%m}1FjJ5HJ+p_(hISB}A-qRz&{&WY>D$bN$jVCDSpfJMo<heQWP7~ZyPc|&BF
z`~2&OXfdCQiLG^E3(3$3I3SVI(t~0%dbk&&3yzgO?#JjMNojMr5v7_yBuO7Pv5zZx
zKEB<fF9aC=>a>KC)WJ3HxyKUkf1W<PV~4l0ie{n$QqJi^8@_w>W2d~tIj9$&)-rNQ
zILw3yyEu=f6A`6?(BcZti>Htl@z5%I7Y=)ShgV1uU6v{V7Q==@0B#D$LF%dj-|BxC
ztbR;x19|b>W2_YAG4C`C-16Z&ZhF)~V}kHZbj16wfDmj(bj1er?qNRKO<&VJ{-<@d
zLc6rtEh+IPDaQKXM++V4`7R1<XlTa*UPXoGzA=dV+jn#P%~zto&@wuFueu=y*s{g@
z6H-{af_g=FHH8nOxc!Iw6PEe2neLA&PD>NsGe*2m|87SuT45aX`SwV~{<4f+ML<#^
zUqzRD=*vh2cd%_oL=5kZiiIc<|1RsXHlc(bthv|3gXUh|9zTEvS~CJnK3Sb-_x1zv
zp8*Ks_2-j-Sb>{eH)HdBic}ZCQn}#2{CddMWzl5Acu}I}GVtaC{J79LC^8xYaQIon
zr9+GoEIIHH3Zhpm!q>7zuy`N`EyzxI{dcGan?=yyaG}%!iEwo?k}E*Lr>m1s*v=;!
zg1a&1Xxa`#dM^`x^c0^<K5l4ZJt+n(QxHi@LP^vBJ4cw%cvGH2dy3k?$pTU5VgPC1
z4chJp&l36RnCC--PAGU)CuO=<cJEn&gCv=5U-{sabqU-Q=%r~G9lH$3?ZIzeyVKs=
z?{C(o4L%j5J5T@mo8vIJfg7(+fZf-k|8=PQi<BND|Bts-Wc~!(TJe(@3ehZwx6&dU
zM&=WLBwXOLg4}0I$|9%kC0#$)j$Y*3-JYM2KGK*<tZd*y8CoYhzQ((is;X4jO1qQ=
zhO_DAFI#4FT{?g~stG;1q8LU`c)%PY3k-}(T-K-`AL4fd7U<oORYl{kUbkw0N9bK^
zEqz+asuXG>+Bk8}G)fIuUt~Y!TH^O8lJ?Jj6(r(V-*zfuAOn`Vv2o0$C!H^uXV<18
zJ-zD6)t%<XQs!THyXxOIMZxD9Y0nBN(e#Bd&3qCfJ`lJmdnyZA(pIG)U1{gA6}l<E
zN(o<=!%-Kt87LNMKoNGtw&<E2i3p{9;wr#CN})kr)^@C9IMgjB_09tgmv%*Zhd!W%
zmfjRfmdl-LxrW<f9dgo9hRA&+4Tog_%Vfzz&**f8D({v&T6yDAA~0ZU=cWxNR6%1{
zsr>t`;@#5w2u4&=78NYhEay&2)KVhZE$OflQ=L{ZIATs3a6iW3jwK)&^Jr}aHSbuX
z&To_@Fx&ecu6&c=2I=bw#fT>#n=Ngs(?>-euoacn67Fx&g!2C`BrZ`h2ida8iQB1Y
z5Ws=F98ow&5_}Yc*$l<dSVBXfQ2qE&wRjAp6`3K(b;<b6QIlZ*@}Ni#@R-xDUq&L`
zA#%8GJ_x)l?w@GaZ-w3_FV>bRuf{FW6iK>b22@MP>;EorC~X?`Y&L=xNMbbp-8{X8
zV>;*Z=!z)6p*0apaOTYpX9g)9p{yt?bz<m{HP0Uhid0@%1o<I$3c8v&MrXJ~h}Pl%
zsp8~b;7D>4NL$fRrDyc@V*snilujpw3cVd#tM;&~6rCN0oaY#^y((@~<9fc_{3290
zB*kd>H7$0gUO~HMYs25!S9q6^Gr~Ky=95p`OUFHfWrLNC-(ILP?osbQ8tPo@U#vY+
zLGRhkB9;f+Z*(~PcY)@&kqOnTojY+E%^<yzF5{JW!QCeuM{)8me)Eq1D#4w?9_Kbj
zu@jL0!!-|$d`^grGJ2|~@?wS(cO98jRxo8^rHRDlgC+aLwoz9Eu-es6(R&}llZ+N`
zkTa~(NH79Lnkb=~=1r)k+W^+9AaCPqJRjXOg_1ojJ>~DWBlokoo1!y1VE3tDA`0+B
ze8cO)0W^HwA@Wrp5JmnHcbe?N%lf-U2ZQGZ+)XdBR9Lu7bKf$>wH1e^6{o6$Zt8<A
z^K)gHXBC~pBpuzCT~nSnt3iV5_+8TJGg1<*Hn?exJK#?xNl3A+^Z?To+j;HVeMQCS
zs+ON-VOhdR5$-YI>T@iYn%lH%DVrBg3nKam8v;iBMRqt0C_YV1Pi2|JzZy?zC)TGN
z1FwpJD3~61z&%P57iFvPsUQDIn|`0aJW<VyCF$A7L{b!@HZZ)C-R#G9;l&ZwJ}{q7
zlK-82=B6K%X%TI;H}BJ)OzV`M!Y3PBlt2G-(I?1Hk5eD$8S<tZM!1#j>d=B0{XFlP
z%A}+e#l2w#`Sd9}=jg{Rl#hVyl;)~mWy15{7}`&-oeNycf8i(h2!t&{Td`W*>MbWQ
zSF3PZ8MPN(mn>{?u8=oaG;fOHD@qxyYT`rwwgs+9MlI%H&Jv8=aa4`kI-58xRHoR<
z>`jr~FD46_;pUJ`2Y?dP49jS0?qRrw8tagcHNM{(iqRu2`N52_rXSUWo5a+L8@@xV
zMAsf!ovKsylI1#UsJ+57nWv|%i=Xe6wBntet9`5-K6Of?YdsV{o}Dl~9DD<^Rx2h&
zN`!PW$Z#|rpUu4FlsQqQCf41!F(#Ya*(D6PjSw>XpA>Xd#o0I?#cuKZ;%>7rK9MG$
zG51q71BrWXt(hB{wO)Ws07j@}hKd){u|Y_430*SX9G#1<$_udGiMv>_i3BVyKd^&W
z8%q=IhRuhHhtqQp>?8l4offRbbkLE!-#`{u&mgZ?*bfmj(I7=#3K&<JK2K4YZ%Gm-
zJFLj;oC36&Jz5YLYEb!urGv*GaLU_TZ<_#SIf~H&yqWTr$1Q1<f4<c_DLX-}S<%*|
z)i*1nT!dP52fuv?_-qa^SHfFlnjGf)je`0fRL<lp5cjdP3#NN@fPOhok$scaGHFS$
z@R6}#?vc|K@vAGR*LWJMDfe4^_&d(8Mh^k?3Op+i@d)8Ck+q)pYv2#p;W^D3-_FMc
zMN*BKZ4uqjv_OR4>`I_p2OE`W=+XI#_HDG1!1pcO?CI1iq^jYG&~eK}5Q}#IrT9zZ
ze)`$sHUq|@MQp|*52U;PT&7#@zS%X!gcsy6Pk7aC*Gzu}pJY+<@b%lu80Jr$q9tY0
zYS`6BzWi%D+3+6vh-Ne6x(SSNN?~-i+#z{LEt)H20dk_oxbzVab0)|e`-G(A8{Vj(
zvQ4DDG3MHM&aUW_4j((12+^+X;PG17)){LM)I2wMYj44HqRRc&?hv=5A~rlb!JuIr
z+D_09|7U2jT_9Xmaop3^$(`%xU|68(Ug8z1PQ?3D?$O5S5pDiPpv4ur2hV%J9+R(3
z)XB8d&2_Au`7Kn<HxD*0Y1L65s6^d`)Y!?%iGR6X#>1VX%qzzq7pdjKsDnZb^zpln
z`zRGf(jA%_>fZ4s6vs6SFyG2g0<z;h!aQcvq$P2aH*Cs`yj?IA^m6oFx|tgdg#{L7
zSqV@PlGSZMx57y^2yBv>;tPiJx$xWt5noMhrZfACoP3}pBlv{ey3gU|jif3vC_+h)
zb-oV`1EQi#x}@?&{fzU+(rII8aR!&6(LmCCmDdX3CCg}y<m{sz;oj(*dlcoy!;TeU
zi~Nz#znvfgSFG!n+VGg1Ut9J?nyi}3ww^FKfc2}3rV(zHBkad?oxyJ`!%#O<)`x`o
z75e{aK^k}&b9S({Kl;2I)p|Wigsz@GBO}9M0XRBGw_LluG_Xe-Bz*7m`<oop6{(rZ
z0B3)&;>(c8r|x-wf~WDDZVtk@!hzt|SJ2CGHUF0L>r{9((mveH2#XjWBK~20p?%5y
zv-6kOnZw=Y@lMzvS$H0%)}}W|r=_|;guf6d^^ndn;Z%OFW>vBU)WcRL(w<KzvDNoT
z>D^QgS=_<F3nrTL(uL{n4r*f}Okbo!&La1OA5#Ccl~Wqt*it0&(q*K^xyYz5%~t9T
z>PS6O!Kq=Bi)6cKB3B%5d&G{+`0~6pU@4xp?czaaM-Tm04q5r9&I!v4F9kQYZw&I`
zRxe;v^m*oO)A-J7t8tz~dM6!%!w=FeHYOT^geb|-WA+uWec5Cc(UnuR%<8Ll&dftF
z9jMrAf>e6b+l-6@dfjxS{R7*_Ku_NJqxE3FOm<=~oO@yid;0TKNSF43b%RoZ%dPfj
z%2}@_GUfOpCb+6Mjs-g9t31GB1#eQ?iWGsrOiRvW<YV`$iMi6YFV+UmB$_K@^GKF?
zpZg~4->qggDIeH!d{9P_eEk>Yfrqhg%x4znu9|fSd<K<76=++NTv)QQ<~;T_RzTb7
z;fk9K^Oz7jyy?cVTq)^fap-Doa-fMvm0OHrqr?k9xYJmI_y>8bBI)4fGPsfS<n<z8
zq(SI41#*+dX@S$Vi3ll*C586w)3y`|b7EwQRETwp4#(6EH-0yY6_+{GI{&d=nR`*P
zH)Y*pUyV?=uTOL^bFx)&Vxq#r+<q&`F1P0}5yZa3-@lpnEEi$b3KI>Og@Ri|T9~Lt
zkVG#~d6Pi1^PnXI#H>RY()S^KE5GGtBzHA1XM?sFW|Z^AhU1KEDlO}evySAu<C<5C
zR^_G-gqD^Kk__TEd4kd@Dvwki>4CKcDG09<+2$AwD}MpE0_pxioI*_;!PWOPTPAQL
z<vY=R?L=l!IH3Ydnnw$94-%)FDU5X$cJ{C!x&)4(HPrwQN&o*YOdu>zh}$;1tP)9<
z35JD^dc<&Eq$bTVb8x&*7spfv-$A?#+x7_q|J(^|^3N-sQA@cQOAS%i-%WDeW@~PA
zF(}vy$z94+2L)B7Ma%b;r_e4{G(zO8n*IA8c_9m|Daz=(5-UQ_Jo&^UaUUFeKEjKu
zz3v<9Zu~%de_ml(jJfhCU7loJ+b-TK2r;cAV3${$WxxvJwhgo%H}R2G#~$4R7yXq^
zJ~M@Qfdtyd@cXJID@{7dhI<G){arEdn6@8Tk+ObYYy;;<Me7eKZsKi`{%`(JZ&VZa
zY8Jf5kE$l({{8O)Uj@p{zcoryHD~;?X#uNB;qu5`W0hm6^m^#xCewpe@gtsWPJe60
zB1`G+W&DgB{M)3);kFIeCVzkx^>&f<-2%%pHqn02Dq^L^-CdhyZSb}CcMrnULgB+>
zpt@c^HJA9xo*yGjY1X3vcA%_1W~3KdJMYmnXNY^vk^#+14bfr@K0!F~ra&p~Pd+h=
zV8)CBTL$ECIUKTt6E|1dZJ`7x5-FZ^OolCJ`yK+D!N!tzz-N<J*eJaS;nl^BNdsf$
zX<)FA+8nc0)hC8p(mQFJiXGg~`D@<(cR`F<JWN$zkTnDn{xJpTd&9~2neY6^nvsAz
zvxa9JW4bi5qnF7Ur$e7fI3K$$cD_}xEZS@eERlc`ap|Q4v=(*?CIIE+h)-v?fT|55
zN*FLCU{YW_4IC3isyjSe^xeQWq#S3Nc-|Ts2UtjcWRm5@!sTmpbNl}kMg-z5!$4cf
zer-5_7>Mj(bH1|lr1*+*UmP@OrV`OH^(BDS{Tl|@tmvbR`iRDrFUn)bmj;p&?EEI0
zoFQ9rSb>i0BbH)!$-GXkte|i{$)74;<6m&zbyD)b<!aQIX*PHx-1v-z=9f5D=39H^
zY*Lsa?Lm+#%Qm}ruN;5<8x5|S2zXdM+1cJVVtvq--REke=<Bc9SI#mWo4h}yJE{Aw
zsR{Uc&~1G)YNih|;~)g<^9A)V$J{=5mbB?dPuTL+Mtj|e8h2v!8a89ut*^P<b6q|J
z>nj^3x(Yt=Tx2uP{$QkOyZH7G42^C0j@EkT7Jx1-hr35xFMSH<sj~B)v)>oUw2+>n
zWObl#&gQr+MB*rlmWPaP+_f*?0|V1l#^(jqd9S#8AHSDR)g(WBN!1&T4LtSOgP%3s
zyZ*eE;5M+R@ZkNHzMJu{8kP^{8Z-mARz!R>OD`Z_aX(~pX_h}@v7|*bxT@9DcgfYs
zEQunKfg6V`_a_M2<LQ_$R3Vy>VFk;_5>3C5E_4-fSLdH?3&TR|VhTSep=nYV^<YvC
z(<atHZ+5%yCIz%D?A-_7L9YOJG)gHNlQuz!T{w(Tb<c?5B~q^U#pSQ&(Hans4NW`>
zjA;z%!7vIk@#X#GFg@_Hu~LI^)s-)IlLZ*O`E^AiIZ=<go4X^eLvH2W<@xDkea<!M
z!WWiC!c2&I8xz32vQ)FS;a4H@8*_Pcpz?lF-<=#>=Bw0;+9KB<EQ<^i=jMn+a|+@b
z_xU1<;Y7kJ_EJxu<*IVPIW7=^R5jQ@ED+99ez`AZl3J!&7bwNklRvM_HDxk|X_2?z
zLd%|VoTxENEq}Wkh5hrOrOiP+kOBwA&NRv5%HWIVlY#0y)B9=<-)m51lR2!G$F`i4
zx)e6ESEtyZ&;rQJ`h8Uub7)qCRl&4q-o}G^5rvocl4(8!qIX4R3f}Sl^zTgQ6kDFE
zIeekqZ-DemDtT?XD*J$k=0;E-?X!;O-5c-{crb^V0jJGCIy_xAw>Z#iyY2?3qU-Cc
zSsCW-tV;IuAoIMbb@)$pkBlJEBI^Zu+0yb;rN?7EryWaG#d>N~jipLJa)0sMWx4vL
zyb+aeka4K~vBv4oV5)Z$@%l>3KUy2F8g`LWt4_=iITL_7uRA2nxaWU0ox53PhZk7j
zeay|D;D)*Ylt)oR`&!l&(7pxwSkDLuNUV-xG3L9`3?SmCUU{%j{FNmwZ41{3mh6=5
z8)mQ*J%+Wf{`-0njMh?hO}+~nRp+eCY7%IJS2D_?f7iotfRfHO?e8*g!8w~9_u7Pn
zi9>^Ev*a>*m{eML95Y`?H9h3aB)LLt#h^m6v|Tsv^-G%HRJXO&9f`Q|+UQf%#1~8*
zC3OTHs|+Y~SSkm<Yo`Xe`eD{dcZYt02R97)MbpU20GdjoK2$g7l(kqg?78`Ja<_a-
zU*ALka1dpB2LE)P<{ZbwqzdT)g`;i&YXYlppJKryDRs?!6g_Vs->7(A=iKD!FWUkg
zZcQ<MPuW(a1Ss9zp^IwHbuD!4*pbi$g7<Mm);5$brHCSgLZVP|DEDk`iHpmLQP^Xl
zw8&qHD9wj8RLjh1?qT$~;Q8HH*ts?P`{#LU8gh*rtq=cF=xIC=H+`VCR+Hd4a>Vj+
zT6e}5=8KZ9Bln-KEO4VGAg);)!A+K6hins_4^@vCL+qZDCPXRk+HS8H+?b$+csGi1
zZyX@QpfEz3*hbQRS~s82h<2&}Db0!}IF1sIc4byg`vo&cZ^&T#`Mks(Znz>@?j6hD
z?rpWn>(;?N%AysTp(pk#Ta4^SvGQz}5PLyA%s8}8UbuKDc-~_|i~DoS$gMm>MjkMX
zjJWJ+_7j-npiFQbUbj_FB?<I*k9H3szryOT^Lq@>;2T35{@jBm4%og7B|K)_-Z`1j
z-_mk~5!uk>#ppH7zg%8fTKF{es>$-Pc1g~1;)q<U7)5;x^i*=6j$yNqV!aWF)lCWb
z7Nv`GZq*%3R&?Q=8vhn0;C+5~8_?8!iQ`A*FMc<C>$moku2P+gp;G<h+4!4`V6xlM
z9ywCvbhi=u-vvQ+`R`5#RNnqCZoHvJdT5r4UEEVo|K!VeC5Z6+Fue<9hdNc&D#N^_
zNIhf#n>McTR7hXf@*i%v%^(mZ%TA6cELDsFd~PMAm-Vh_idVJxrclC(QPZLB`0L^q
zGJ~mL)1rHQG{=0A63H>3s?((ZE^K*3##HtX90Yf^<K<`&=NDnk=M0;i6r{!U(+Qni
zAnrDiI9s~NlG)4o*9NzsV-heY4rIXHs{?=)q6PPKle;BxuPvV?(E&u62A{kL05f>g
zb%e$E2Li@n@71idVhv<^fv?b87bNLlZHg`o)2Crk5<E{oIM#8D)>zbrH^NaJufGDY
z5-{5zM=)BBO@a7wlI=*ioMHe(-5V4P*^%a+oZRlBZQC{SG<|%ELE+Okk!@z%ur^%<
z3-s<y8}=!8tD5TKs^ysb<o=ESF4Xdb!_@qf^vD|2xyf0dw!oQF`|@b<wAC_fy4QE*
z?_b9<<AjAm5JHa^a{apI%8vSrFLO_!%HGFv*DP$u0x3p!9v<D!-soVOL%MXGaY?j=
z^9r#Fu{vOl`SSOCC9$nYaqGS;G||edQ<P$4RP91H7rdH-t3YqDZN&o2*o`hFV%z$&
znvoM^05_@rM=z+n3afE8?i)zjOV_%d=S(l7G$m04<i7F{2c!Ro2T1}pR$<mur3Fp)
zf%f@>7*WoUNDeC(%&^6T1gtk>wMEz1IZljwxvEHClww%>SJRHb>Sj=DsQ+(l_xcJ5
z35`WDsnvJE0|<~t4Cm*^C?(>7Uk+X}C*Ap($~W$`zKI|{4Q;<4_w#{NAsN5ir<aKb
zrU51x=R=iWCyLH*PeA+WJ7<DY>6?@yyiqMhm^$+7v6~UNCnaum@EdOj-y64~Kw_T&
zQTdBaQdTGYS8UF_5B|j6a3^-GxTZSpY8qyaIPJQ3pZVa%>H6k()2hTf+=Q|{GQ>zC
zKhdA1C^U}a065rFjD%D39#Fp8%jSg6DjY=0qxc%f+Nk}&x(akUZz-ug#xP|d4*{ru
zb)HcYBegl{>HtOz*yAIR6&c+562ph4@wyJ%`9pz_mDs$MSW@<UJM+S)^%`6^{tUw<
zO2TG(A1i%G2P7{>e*22z85!2an>6*nYtc+8MZk#Z8;08DV2CJyxs=(-Ia;h+_-0_G
zDb?)pP@@=6vM5L!iV&f-%sRUFJP5&e1KT%wB0o;*Gu03_IE+9O&Y6Rc`5qoKL@6wy
zos$fgBIrqfMB%`{#R%#OsGLplOxn3yXM&F@Dw~;p$B}?dhbX>_b3x&CevwKwvB?_*
zU5|&SnWG|o{!DKw>*8<*TNrWEr10C{3oQS;a1D4V4V@KY#)0P;aI9$KuwN-Q4+Z+0
zb<XGb58da;Mk!|?t;@Y7d&Qs8+lDbZ9FAe8Ywaxqx^EUq#S7d}Hr2~n<bJMo&ZOM-
z$hqvJU;A-P6I7l)KLaXYSO82THmHnK=I`UfVZnQm@)<`Bt#Q}%2ABJ{n#x0oVqstq
zDZS(6m)t1OpLaF&fGzn91Bl`USpn4+!;3g<d7Pum)8(hqhy=h}ZZoAj`(b@KS*Mr+
zl|aVQp0XYSI0R1CMzdG1P+$I%gXjntN3Opa#DJaF-ro9%Pxg~Y7ju06kcUo1J<eg9
zbFF|7p&CU?7Gvqq0sjPqc&Q7t5IT0_+C^+J(Ho#K)Wj3O_2J@i^K%y!uP1bAU-oA?
zqL0qUOZ=Nk<3bPIQvb5!>{WHrmu=S4=CvjFu-c%fG-Un|8zf5l42S?((OJ&TtP%7z
z-P*=vm6DBs&iq)SP0kPx3gG|`m=BWCA6Y(u1HOg<TiRx0wveP=Z{vEvdzUE|q-@3m
zZm#}is2jbFJ7)%P18j@2x^eXHdxm4UhoTX4Eis6xcwlGM+8=)^PgVMM&T#7S;;8z;
zAuSukyE$p8&V?6Y%+&Xh^xuWf{7L2xitS-wbGM$~o?hhDN?p_`ZvQOR9jFO((Q%MU
zRDuj3aEL_nEEdB&=7CFw6QE;9Lc)RHaNLs~1u1AW!%TJ}HTdB9OWJ<*Q^zuc9BYNE
zN9w6J_SiP*<kDJ?!NjbrviwbzI{CcpLJH!<Ff***D=Xw<t;$!DrtnUB1rRZ=M-~ab
z5j}gBx7LVHmHdNo*G7<t0Ge$`bqDB~rA(<(X3^gY>g00EaxF!nhJSV0)HNToQa}T{
zfJPUXgC&7r4Nf4ZV&DKZJF%I$Ih*f~W}5v;(cNqb;|m*0i??>$2^XfKT;rHE{rWGR
zPVN07ufXcZZ99*qQJZt)sATFiyKr*mmsy%tgt%V9LbP98POOPYtep85TgaFUprh-y
zKGuoDb{?0S<9E^_cWEjMEc}u=zE2S3P{BBhBn!q6TYyZAg8-NK9AKcFgU=bW0(*X_
z!pedJZ11^aKIudOPY{j-&lUt_0QN}Wl(Yg;FrRPRl<<S*4;1>L={AzCCmAw^xYq&_
zYJM_OPPk=EWae0eR4&^RlU^=;0)A8lL7iBPpXIOhE&s_1uY&(C$I7Ek7ypI+R1f-X
zIKbsF%s$*00Q^w3$W{L~l)lSK0*Gt<jA6DY5RXM!w2&+#<}(UzIY@Wp6IwUFQX}SW
zJ|$l?r%E{$?SPWgAP;s^YfT?r-|;Ixz9=yjA1#!A%GGE@-A#*JFCS80g**MJsWbia
zW;TE_QsYsxK+h@U0H0Zh2oL6eotWE*)XP`r0ygWZ<J8Q9BwPa<MU0r)B1i;8HbTi9
z_J9^VUjlEEf%F0*(=LZav42i-0Sn3o<{ioBmGHw4z|$#~IOR=6%(R2o8Sg{-fU1_B
z*VA%iBN$pgv>;28(n6((@>u&<TF8TiE~e<KtD*JhS#1nEC%mLQt%5dE81st6<XK68
zNOM2LRiJIXi6}$A3si~^q!Y2_!U1DB<%uFA{0<fblGT0NPp{8PxBoI9r@XHxBKTna
zui-iP${Nben21R$rl|soEfNq;v%?{P28Wa5<zH_5U$Lxmog6k?j$YNs3%J8JE2;rG
zZI(?y(|)}ruo@oBJ>||&NR{~J{coL})MWofUX3<KR%POh6W5ooC=IfXXV2+}S1aFk
zWxrjqYfdJ6=9xYwCvF6PH=GeB8m~&CwB!v8RhM}Ev^YN7#2Q-MHFi=Gy_Naw({S(u
z0r3cU+_lS%nhkf}b~X9ez8Cn>KM?mtk8HOz%r1&cgYfWEZ2vN6qSdO2>3ELo>2xbF
zB4AelIGHZ;d0G-pHRwNC64f839q#+5!1M@+V3Jn3*IAZ}NT>*C18{X_0mbnkNraU(
zTQ`&FyLR=z3nhS*nR8e$Zn|Meb})ux4W)tx-cgRC^?{uA5uzHN)xuh;yI^gkr;Sn<
zgn}9x0bPKl@P_*)pN`>~psC^EU!~Pl*a-G6YY1$gB`H#br>E~CD{K2C7E7OD-9M-T
z9Hde#n_uH3T8+hEi`H5<Z~N)fkXu*qbj=ydBz_X|-ndb^fIg2Ff=hD3<V(7p6i-0|
zHsDB0HQ|R8{=>QNzAn0}{))&SGXkMR2z9p{#Pht%;KSgE@i#)iHm)_%n2@qk0ThR^
zEO@xDJ>>ZiuQ*?%Sn9%Kvk$6V4lgDFaZ|{<Fl4cop1~n5Do1JPN&f4+h9+>Vvj3Q)
z+5Itbm+#ntmu@%L&jw$%E-2d{F$>R^8F{R_r8x$ereTJt2M~T=Ige(PHn(4mL(z9z
z{HcP#r<Is=KN7@jjbY$C>ITD}?C&@f8UU*;a5a+mWlxervW%YO;7Jk&f}A0(4tRtY
z{t)&#E$G(26ED6S<vOEXp#cm5n=z9+Zr}V&H-$tklaJZd6&7;(^qe5<<a^_J*qZqU
zCI`u4(cXNI48N*oJ@^}7>Ux7g92&$C!dRJ#$swK53h(M*{$Cf&*NIbM%UyItK&0^3
zv$Zi+ydU$^%8KSN;!4A|Fqvh!ZsBtlpg7Bt`0fPQI*wfadU#&Vz9}1u4&L*Z#j_b0
zam27tm!z7JToGQmDZA+an&N)~q3ibRmzJM2hCzhftwryMTMxF4YZ-(gnh*3TDU*BG
z1RJyXxY1jh3-&RCwlA&I3DK<W=7ZL2DBcasHRkt9*E53IF_b%R6`}58MYJx`?-RJ1
z3fxHPMT`E#GKk+O*=BX%^>F0T>Xd^N@k>RRcN*n}8QOC+NoSmF3O?MVi|Nvl;`v19
z=o2pg4ZcM3@fxbj4rkwH(rJ58<@Rkyz|G9_XC=DZ_*V|Ct7vBAH2ue$<83d-&LI9o
zPvrNUBq_yI>6#nxhhff7T`w25rZ7mJ7^7gHYr7zf`PrOz+i@knJyp$@+{%M=<E0$j
zLjqqcN_^mmaQ~d`A3*nk?QWy^9^U6^^I6#YEvDyr6BxJ?XELUm<ql-63wfz|>CSA(
zib{_HAv6xMI>aYi+l|Hk%G+);3?Dk4)E>~Ww(77X%uONo0}Q?k^bh<wy@@=nU?B7l
zyPe+k)wS8^=hF;XrvwJ<#$-VV-j?16;4$H#)3R$%Wm+;{aI9fh%7B{$bbtEppTqD3
zSDYL=Cu_S>_jeLdwpa%L#+YJHrFU32Pk>nL`Ep|W)L}Tnuc4Xt<Siw0UweH*a6Yq(
z9bZX@xrXqOA5kLO0DGxjwhnMD1YlFq?fq?IiuV?2CGn1qm}%5s%OBC{wnf_Y^uIWE
zTObqbf1<U?-xF0?;%MKTX&z~Ct8zdpT1bD--r*mm`^8Arl}y?3Cc(Vxz`E5CH<Dxs
zcnQ}0AU^RHtnrDiV-26QkA2=GDHZsV0s+PAGvA+w$^pZWA?}|<qa={cim>>7apyzD
z6%F$|_5m;XH-!APl6Vv88-KoZuo3e7@6Ug|R20*pX}C0~5yf+9c60Nifbq_o$q>7Y
z@J}mm-nm|>^A}~}gXdAJC-!f78vD;eOS{(eHkS2nh<(MUDEaPdzF>M3=;@%x5$x)G
zMx;bcgo{c!m6FDidt>k9E#Y64q9g?98Y1#sn&1(_XoPHj_F(Epd0>b9Jr#6e8{Te*
zhe5QEC0(%meSpixv)3+e+&)(K^s@%cNeVhNrVsvz@cJ#Vd0r+)m3*a-O~ws%NAZZh
ztz=a!e7N{N00aldr<NeY%C(&I7el4hzfMe~IlKD()A<i&Vc$llpXQHUgK11#GZ9CA
zv<@`9usPvtDK1I&cd}%NL-W;dHYYz@k~3I4HwmPMFL<8|pvXFiEwlb_cg;s#S#bmF
ztig<(Xf+0$Wr!Fl!Egd3dT*?hI;5L+EcqG-g~Ij#xN%K@^$8!4*mO+--Ct8ZZV=P*
zOx_wyk=_?FJ8O1+3$*q`qyH`>h1AQW$TnkpYQ2DCM9k`e7k(GW=w_AU9`Xe~kWSXc
z&IkPrggii-6yM8=*ObQ+fglq!Ylu6xk1Gu~xsMnP&1(F&3}85g+2Q44$Cp)tCU)vK
z-vB`a9O7A>_0FnA?Q~pgwYQCw<s!vMsDHjdL&UIJXE|!##o!_Q{Af@lD<`oR_xXz!
z(ooYN$P92LP1Tr>gll`nvj(N95PWuEe}u~cKrCMvPKUnu(u+6)!l?ha0q&3^QS|BA
zNDAzR=SZ#dKX0TnX0jlw_C+=ruU#hBS;c8!kk}Q;Cu({)!K|VLpl#&wi-~0YX_Z!N
z9<TKZVO!^$VPi8*Hj22cWi7=&-YbGuRCD<_O8+V9<#`BH!$-92th}UDH1*%*h`JUb
zBa^v}^JMH}%A0!vPtU6JBZ(%%Z5!>1t;?R^REmEhLj_cINZXwh06(rdy=w*{Y3Iab
zmr_a&=~c@=7<Fvl=)X9r{DUv8dNn$uNhMr9kt?L}CzzmP^3|7MEyiwNZ1f~E%)H9|
zT~ovPG(L%XAJqmu^PYuG-XE*9Ubrcz3W~U98u}D#h!=iR*IGf<3eoMdeMuGi;zX9m
z9P8yn7uk&t5GqGiTU8qkuX8_(-z-V?gafKmDOwmlkq^OF0F-_sZ_NBjnoAznVX!#M
zjHg86oJ@EA<m0ELU$Nt7@SK@riG&}Vc&qzGKYF=WmDbvqyHFP;Y9s&Fj9RLB01&b(
zqv!>vB#AfS1@XEeEqKMD9z}(_z$E_-z?>-ZgE}~=9wFQ+3j|9D#Q1vR>N6u4*#XFd
zG9&h}lnq~&exxsm3c%<;y{uDCaRA0yhw9d=fVD{Bkf!11vBLen$qk!S5^56m=TWEo
z^hnUAFhpI48G=Hp+{KKAH@WqQ+JEC*n^TRcI;;(c5TO>_#Ai7SudV~IC__eoMs>A<
zKSbQ~&M(ifI^VnntkunB--`;W1a*%TwKpH83KXJMl`e-23yrO8GtE~n>4Mvj#qBvk
zime|ICv0WNyzXn&vvIBNnQB+Ufa=@CD?~X%PI7iLntRgmU%kY$;=b=aCvzfVmsOP1
zxJxd^Uw&Cx(Dq?}>HJ)BeCeze5(rFQp99=hL%y$-bhwFrVog!RjI#K*mj+0vop6e`
z8A}cGepEW#{d$%z?Guc4r%QcQfsRd;_@JOWeH+xj;jdcd_Wp%Qtf9F?5U+fm+lj||
z`2*>g-Y~5I!}9u4d*@XI<oedzm@f7TOI2o=PbN!oPx<o9dxXJ4GXPmqRq5fjSh9=q
zgHhvF@renxM~oHr6KDMN&{-8rm6Edib-&|R`1392rvhan!5kCH9|#7n8_L6;LKZad
zJDCnc#@)^f9!oBIzx_%Ui$DFmPAgAoKffrk;XR%WhcA~|ioOS5xpIwghWhWqKXSn5
zNvDz$az3ig+F1ot%j@(sI#1uNb!8?MOJtw7rrL1hugn%>b-#Qj>2GLXBtp=Q7ok84
zSttkSp`(2(Wt8@6*^IB~Fj4!fTLRxX1mg3XS^q$RqQn)}eWIY5yKV5~!w$z5#;fK~
z*FK$I9Z_vKMU^y9T8H=14p*uzSTM{doam$hVMg{D4zLXr3|mLG?#nm^QXJ^87M99<
zMorm!o}$z2c#|{GsspyW>6b(|l9q(FK?3knUcc$5pn~Au28kxFCUiOr^gbu+$APwW
z@eo_D<xjt?i8X%N>AH??GEuDi4l)Kd!OZueul)6om=^WBCUZ*%4Oa-Q+WT|O4ifQ&
z`-(w3KC!$*uU)SQJgTVUh%gX}S@`VG<L@DXj&&P%QWeePA5wWU9rj-5$bF++?a=S@
zj>NtPLmxayI`Y|i@a<b-x1UAG0=K(mkYsvs`(*VB0^h!4IP88?L6s#ThmQQ;g<ks?
z=}$&GdVrIC?&s0Yb~_!Z3jcKiqZ5!)&37kY{n^c(#A2lArvNlN^L&$AV-LEXU=ozp
zD8qTDC+*N@cazRUQT2-=W$G(lhy>p0gMYQRoD;pYeCHUXnm_yj_xOdwg6UFuv3UW2
zZ%1a`5Ah&GHM-+x)9n1Es=HP+A-uWb7v5KqJ#yNx7>Vl>rFott*O5@qFJsR|Nw<{l
z`P{@m8||wF$0n`Y?UeYWR=<cfQ&34wQGZdIM8*P7CvZ%YAA8zDb@IW+c&8>qe~mG*
zwLa&Kr?YvMY&^DjsM4{B8s9RdFc|h^$j%r2JmFoBIuoSpYZa<LEBbmk{^nNx?xNOR
z!bJe=ADtbG+ZSm;K-(<0Stn<!mS0bR<&yv04`WuE-%A(JgNF1wp(zlSvEo6~Mtt<P
z>A|rggx#c~8h{tbZ8o;N-^O-pmk?sze*j;ss5wdFf%@SQj_(w;0H59a9jT(~9Dcr0
z%se^o7U#fk?h{gQ7~aBxUMp8qk{rndBwoZ~nP7aK63KBW#Bn`~KSD@<9&n+g=vZ>4
z8W%%?!<^S}&TTjzati8HBH{B4*8C9ONW1G1jE&gd=AanF3o0Jp*ynE2&MgBCHrLP?
zUfCBxdMDH(C6~vBIIh<zF0Uf3)L@=wu{MXgKScM}VU}K~8bQ5oZWsU$N;61;ZP)~L
zL!#+<lxM>fTZ^<Nm$DtFS@B*DH>zJrS2Upv*Rl>E?snFSCBM6GJ$VsT{rvg!<(C(u
zR!wGE?h*#xIfi>4MZ{)FC66bl4Ht<ic}0!|Yu@*9DMD>53xpzF?>d*ZuHM=SljSQ|
z{Sw2loPJGLIz0WwF+WS*H%I;L`<)zUu|^zaBp*z<nY``6G=7I&Q>>UQq9~dt3Lh~v
zQ5CNgOJsgnONA&v&$A1JOp0(_e&S`usy{2zJ@-NAF25)4lXA)ikpbBSOI1nBzG|)(
z{WwG_ifmm|l-*8*rDA+J9@Jyf;s%5Zq>PhBG&&$t?9E8-QC|Az%g5|qmn83RYv;Lr
z_MTDh9n9I{@h)p8DTb~dM#tLuGz?3W$rK)nY4@*i?VXi8MKe<k3^~u3)7N3W2`3UO
z@<XK$BVib4VrVzBQWUrT7dn0sZd2f40Twhoz~sYVGC8$he>?<y^cdy-1{|28ei{FY
zGs+Z%l#6V{fMr)Jo7kj-xNp60_R?<vurP_<`_P${Lrb{`$JujS4*D2Eh%<6NpBjV&
zN^z|J)~4_V+$R*xhUe7UFSm>Bk5s1h(17Gn4N``9nqRd~yS-XiFaWTW7R2TvjG#Ay
z8fux6dYyRvN?n8!EEhiA5F;6dL0>LNBm?=_3em!sV4sOo4>c5uNQuCN>B3=MSjxE&
zDHzRh8_=&-i)>%9w{GBfd*jpQ7@gaRXc2NC`C|yO`R<Lm-NYCcOXoqoFt)H*{N?5o
zx39cAx@j@_OV^Ku#ybfNo!MM7*>bp&?`~I9Mq{o~{Dc>ThN$MVbTTZt?hdHre-QET
zPwfjm_mL*qqusm{Ge_+&7UaGxFRyQuyYjvDN?@#6h~{dLv$&>=Kh;}xlC8QD5G31*
zNUn(P)S&7Weyl2ITR3Hz@8>v`ztt8Lp-K*JB9);H)(A#E!#c!ICaX|R9}QQBGuY%x
zd@a7yVDw|5l9$A(5!qzO^~hflT!_pXUlAJd6crby{jp7lYdcN17yhPtFTco3MOChP
zla31qWI};J@`aJelbiPdm92bs?4*kbYvN@CZ!+`E<8>ltI98an_Yv=!xGfJ4?PmJ?
z>qKlnCjBlQ3$Or}ueGlWnfz(_Zrn`RY-(c;3Y5a7rkaVXv;SRasrZE#%sXRW4(0Px
zr)LW0T19}g0`<X3rU`OsZ<D|>)?hyABfxDBut9(>$<V8tq4ONKyKv2c=!>_BSGXJd
z9l}&0YQI*xi`P=D0$CVC;bWOA3uwoDkk{pON)tz=<jqvTg@1{5lRIp_x{u~juwuK1
z%auMogPp)diBv%ZZZ6?OrWFBekE^wc>ych-zDuWcQ9=^)i(>jf*;(UV+Cgo^u>bDL
z^~QuFAbcgVAc9TJ-D^rX_YYIIzET)$Gu_W~gk;A}H6pwF%z<6mJ+<2jSuS62wGvJ*
zmd}~p{Pn<<{5|L6MP}I>PBYvS-lH{>Y=d7ymS;!MDy2u-0V<sDeNt!JmC(os80Ao`
z?*O?k2tlKBuBK_W<~^|=r>n9~)^T0GOhpiWI%UEQg8S}d=T@8n)JG^XAf2(v963K}
zRb8Du7yaJ_cQ$#@O{o;!RB~8l6TGYErSw*{Yzhd0OVP30xmMuco^TXnc5OzQpnh5^
z6(;ydRa^nU)HHrB<fuC><P5Xc!DPMWdDWH_li!pBIHaJaNYO4svu!C_gUj&W9x<E+
z&z@y#J?7Kk)-u@P3-_`Sl0%e!$Y4FH^7s>Cvo_M(@=2~qK9!i<j1b`D<dK%LV#vFj
z7hW^spp$kaG+oIUl!3^Sn9YU1pAfLFVy4)vrEg)}o2?l+&1wI)Ep)nWS{Ek93Zy+M
z>)A%IhTJ+0ZN}l$B{MqAdIZR}4h@QtQ^&HqL%>Jf_%rJ^H(KFUF^$eEqIEf#7)M5%
zuc9I)kJ%n&g~>O&(y8mTrUeKWSTbrG;pcPvFBARex_>{p#kM2;Xx{u#=9(JY{3mH?
zR-?XmdMRE$^eLS@@2&LTL0e?bvLGD+bZIDA$Bp0EdhJm8lN`t7j!8ywJ8g45w@rfK
z%Z6sI8F708uAwafmhoUKqdwpGb53J=z({tpL)jLC`3;Hq#)MC#><r@1M>ME%PqGxB
z53p`Bf|<N}wo~H|E!KrUS7|53(LC>139`Ojuu<8`p26`+BJ3{nohFp0NB~$T*_O+d
zr7&L8!$bn}E|o7-Q6|1e*4a^}H`=vvPu6iPAB5oA9G{83dX`VF?k5*-nD9kuUwc~?
z?6a1HuzI$%RM*d6Dlc`<M%^y0BI5;FC+ALm#@;UPY943oxs1BT!$}R6Gb?_kNcC}p
zlta95U2-iQuZePQ-h;=&Tj#eB@v?j(iB}i?o;JZ?vc;#f?7A*ip@)ZUx|fZKv0}aX
z@?x|&+fGFD-RiwoXWwHfvMjEb6(JhdD)c2%+iJdb98WR=#!qkqX93b;RM+vFN5-z$
zR?8SzjLfDEZUE}@6}#p46C@z;Z2omJSth!;ZBArIVOj1;IY1TeRkVa6q^MFM3en;b
zmt&bhMnMP22m_ymYZhN`TYj!mO@kUFabsDjLh3-cL0+Wrzvbn;o>B67MWy_nh9xCt
zj*&;+5O}}>O&oc2`Q4lG^ymkX3tyQI*z<+v4ne*HIp6=y;VOpc))I7*?zXv4<fN?A
zF&)YTjJI6RtgDkNXiq^DgC8=F=Cr!4-*^z!6zBcB?rmTpswKrN_XdX9^HIllcvpG0
zd|E^9uh~zVBGc_HLp-M?kNe*#sTuwe^0Gj$O=FBnmhp4#rd3`_NlnB%CS(PSeWxjU
zX01E*jgRPrSPyi&tDxkKGgN6UXEy-b+-N|vL<hTK@gm#+|HW^SxUqy!8<`vnx@X|#
z$TH-(s&N1y!k<{lzDu*tCRfLYz|JS1)7us-?p;v;dt6O6tKZ$R%9~|qiULV5cg<W`
z3P@G4t?TmKqGUet5hguO)6z4dHPv;^4Ojp8xKS_JzefF3-u3@TI`eoa+xPuH6*3|F
zE(~Me_p*MBv2RgXvSnYgC)qt13}YX=kX<1LS+i!DK{1qOLYAmelP0FId%kzSUw?Y_
z$GFXXo!5Du$MHV&M1_Pl!^m#jM^&$;knBM{$&C!iv)oEH`$ictl}n!D%F(NiSI2U)
zw@m>$TDM`z!#cj;|Nc2y_Je0S_+S)&;w0^Y$Ub(vdA-sd61kWzd%v+)kxN&ixIU>P
zoKc@s_LTK&Cxl>5?svMMh@6545@rSmxy$7>GCt2!BU(uJ-jA?u(6xk7jbl&wT&hpA
zLqI1=D-^>>i`oDFG25Hj*@gTavRUO9C@CfKB&+I!$Z%v>U4)x2K%Y*8>6D&x5zId%
zAkG>niJy_N-QbY}rFtriDw7gb4=lL%1m}{YU=h%`c7RF+h@eGwvYW2Fvu2!$)gu7%
ztF{?HwSfpe^h9@!%hyAkSsypO=8Gc$ryt7+AV>3!#Il)8<vI{vL*?#_r4GVw!HkRZ
zmwfw%uUauH68h}BNQI_6(!vx^&YXV3N5(y6{HzgK5l9?LdpE*%x7)uTq5;ftM2Yvv
zR>{URMBm!Z&cfzE&%id0n6$yNc+Vp-a+Mv=SDj8)^=XRkIHnr^d(`SVjN*L6$9{h{
zPHaVP9r$C!bC-P_(3w*A*MP$YL|_p00tdzwL6S&D>#NaSowicH%aK&*&+H7O>Cgui
zBTS4~#2o^%;5kk4`|nUbX{TZOM+|X#9t~nuF2tUIff023U2CC*^!gxpbHBjugSKyk
z*O7_;I5>W)EjY&&Pf24UR;)y}&EP?HX9ic^>A`xICpNJqhHc&TM`mnN51qPv0dEJl
zUOK9};7mKg_`>7ESrsa>LwPm}AHEuV$1_uZCyB|sze~1^H;}7BC2?`-c%5k1+-gOA
z6|JUSLWo~u+Vxu9Qmm=wXqq9|UtSAGlRN=W^HVkKr^<Z)R~r0!;+sD0`EQ<_2ytri
z7US1zswJ~ubIuBmyX10;Z>PB?359<wIq`isjCrl@s@FeZ-}PM2kgC2V=%JxU2wLq#
z6%*!#NY2oh``TrIMHq!a)`D{Yw5*|_jLJ&}fa&5}8&$ZaJEs<MhNu{KuEs5x9_!AK
z^W6&X<oh~h_$rWGpfvLI8RU_IS~Xxk^QAxo^P||<_*r)iW1PecFxqX9&n|JS_lmKl
z8xm}K&eQ>DjWxlc#xNY4APU6?12j78x3RpcchEI71!BaiG|9lZGG!=>xyj5DVsQ@x
zkPLMZciiJ__JGyX11Pw8maf|`*)IN5I@zBD6-Ni}pRz(7I<Yz#O^(4w2EXp8{y4p%
z&#N2H&zyTGn0kLdFOQs_683P#XVq>m*q%f`_5Jp6hj_L}VolAry7TFF05H6-7nHwp
zOCXY}&QqD({G2bgR?rAhTAx`jZQ6On3UfrBAwmmB&z@ETtz^gHm}FrHtpX_|pxCe&
z2#rbQ&R>HSw2RJu3`W02JVRu_l`jI$<*^fW^S~+jv=mkU9yvTQv`cb+dKSgia^{+-
zDKx%2f7HI!9n+S&UtXcK{en*c{guO6>_s_QeH+X8u-?Sc$>n3jYbDa|z0}$}olc;N
zS$def3m4r~M7eQ%W=~saP`0#VfKq~!0ttx|geM8-dWDub>k%yoGyKcE;WNQO{lkeF
zH_#G+0BVumnTN@_rbTG<M|<5Hir|Ba4$A>puu-ZjO=Lo@y8z7!-m3X!3Sbh<;%LB;
zY>%ZS1`@E3;%`U|r!Qz4+g^2UAims-yd}^>DfwSM$OOpuaPc>DQJ5+`<y_LH{SCYJ
zvuduArAl?>OOE+w&G*?K&@3ucA={(~%U9SyY_A<=-C@k9#tKfb^1%E%GkNN~<+CTg
zPA(SKgH;jZe4%&MoaE$o4ZQ+WmH%8cb&C8Hk;mmi1|{k#qD%zlwN8qcTzUrcD^BW-
z;1vlY!*L;_I3~406TiOvXD($vH|rNS0z-^RCthUTFxt~$I{Ae1i`jiCL(A^z{7+(c
zZ}yqY=(J4Z{9XlZdu41TNP(n(%V!13%lnK)6f2%KyqEE>kuQ=tW!LKD`O!Fgo+{{0
zG|dI))wZ>xGdCWzN%sM|0-VkAHhx#UB@QeZIkk6f#jHe#Gn6r?U*pyWx^9guX7sel
zV`<)-7jZp&fEUcNsy|qZGpVoONBS!h9ky&P#<`ym<AR(5L+@K%?tbwhSa_z@@@Wmj
z!8NU6=FLPPLFsCB#Z6pd{RY^FNp+YBv6%6++)tZ9^m6xVj1Egr4_$poc(dL9n=;{F
zU6`_Nygc89OHIOxiXev!DRnXR{bQPfag(qRx;7vZuj5pRn8QnPxYu)!YSnE2?=Zmc
z+ym|XB2{0dx_={@jfdcFOls)Pkm|{XhxvN43W@{tyMCYc1oaO!C=uemU4kNMai}gC
zLk_pVS%Ae!%r&(23e6O}j4sf{=qE^87k;FH8M@L7=ZgIuuey0t-53FDG7ykQKLCgV
z8hjjG%J%ecfWZ*r5M6ps`m^-3!nwN9p)9YvSY@@8St!ecdt*j3K<p^M_NWi+!sp;v
zb44sk6*A~89b-?EeYUkalHk>I40%CEqBN+VFQiyfqR_M3vr911=n9S^ddaYAE@{i)
zJ4{n)C!V<@!C(UaF~v&s5uB&;-B&jsi_%#8G1QF}H+-3|<NehKqBgKo&&A4MZ%|Fr
zsE6R&>{V<?q*RE-FvUA&CcEeKMqoTjx=(p2u>7uYR><(e8zA|+#Kws(lp{F2#afhQ
zao&Sd8_7@h805Ozy$MG8Pp#Tq$IDM{@=Lx#naN{mrGi&p9vb`2d(+mEGQIrwm4*Wc
zG1L49FIe!afzRzyJS5Jq<yC-pjFi<-5qCbYM>zUJ)Q-NMdOW+IVnXI>-<&=Iq`$X5
zR#SS(X4(4Jxz`C9cf#lC#<>q-N`|#jUtr;&FPP=v#)Ur~9DaM+CO=gZru?;?aq_;&
zwdWv>jTVhpR2yVh`)of;dl1urM^LA9kHL3^+S8GRW0trdr%L-_3dFoS_@$7oJI_$z
zC4|o-U4wtLryqJ)bC|DCcy<$z0P3qt^gb<Bv}+-Tf{Rg+H{>N^shNXH-zf1aItiGw
z>;3S^`OpFSVr+syKFY04%Dly&eeL26N!yYV)z`S*rSQL}<!;6gxn~nQVM8oH_6e{x
zjEIj@0Ft@rVGo*H8r^8)IRRG=4q^=fa{9~$TR`2>ERcNAL)*;RT6801X7Q!TPu9o;
z>}E#ah3~)i$%8uSp1yE$9zn3q{g89mQ78}&{!{PG;acsX<rS?>bJK_wh@0P_@JGb(
z3PMkPd&)#>O^L~(3<i#1ZB_(vr0a`2i4BO1GgwJ#D`iMYR4{6O7Sdnn=g(0opf7b_
z?Fmbv15C`{b?&zoDd!uk==Ia5NYAk@50HMvkD==fUZ;ja!UW+9`WVBY-!Pet`*J)J
zD(&)9X_dK#O`?E&jW|gjaQoeNq*rjHM|Lb;e9oZ%dOg3&qT@B_tfFv0<7R5(L;MQ0
zYHworU59lxTYq0a^b;|ohZ;t<KweUL3pMQ61M^!l$a5%$e5(<~X4|&h(3|a#6)3>t
zf24%A`@zq@{481{6z^TY+X%?Ei;0&sHxu9m1i9dTWhjqRl?3VGMua^jN<Fs*5`XQN
zkny9caeGa{6j~!ah$1jl`ujSXM@BCkwVW)tUHCXbsP+*jhPP0&30Un{R1D=FA(8eq
z=O4x+6@nhE0zlz}38vS$nY3c7FsFdYGE)DB3a;=-Hy!Pg8>Z9w7%hxQHEF7S5VSUP
z!^=xq5y$P95$=h#-9)}&o;}r8bZOA=Bosh)*#>0N^uXHa@!Sp0VMdg`+a;O^!$mGt
zEJ_^UM1r~cq@~R6K)k$gsxZTaMIuGU$Vd`}AHFqTs8#oPIpQIvGtWzS<XYcqOsH&)
zobYnS!wz7ZV2*a?e-W(wzqAew5W8W^!9^<Uh}q5a)wbPm3|l2E$GAT1{jC@IwjYxM
z-qF;IjCIAev3)=~fCbK6TfPoIw6foD+jw*&adcaP&m6}Ewym<N=4EBFXq!l7+c77o
z151o&Xx%Hy!|=?fdL$=5%Br5)V+pE?SQ5>7=e<oSr0rxw^tFrk-V(3wswY^Ei;)Cy
z^e`|$V7?`W1whqNH;32fe%$q*F&sGOSd}QWjyv_DIXMSLJ8zzAG<0scA%OAAc1cN%
z1PrVShaHWJD5Oe7Gyr$!gBrqFmWDNf8CYH9h6Ub%BMH2qojlX&H+uAz0YBL`PlenC
z9Zt|TCG`7ka^Vwsh(Y<3g;sBzmY*pk0$8|y6vLUMs>0b81=Bm+Y7WNYQwfM|^Nng6
zXblY$h_^O17@%BHp>ejk9$F$m`G!51PPvh{R;$gdbhYMtj9+VOb`gPO4Jg$00TNa&
z{K&RutXs6^s90o1XK#c~TRi_GF~)LgHn)b6pgB@ETI~1xw#%=%R<aI8P4Sfs-_*m<
z08WWw#WJxJ*!Yv*zKBGM7$)yV-g^81@qeB{<BBihyF>KDg+h8M!S_w)p;cSbE5B;C
zSDrJqx#Ycg(SbRNuXnoVVn+WSsUR*{Ch45{l%IQYh^y6uZ<!@g>;*8pJJw4F5`y*K
zBv_xR9o->0iunA9aHtP#$At=5gK-BBelN*PZYlF}2B(&WNv`p9#iScj<%apwHHNop
zmY)Th=Og$-@-q*ci^Zuc3zfS{9|PHrfe{`uOwVsdT)}(awvNVtWFDFR-#-ygnClvl
zzu9k^PyoLkej*(IIz1zP7mT}oz`CTv>(E|r|4Evaa6>tjnClhGO9DinB|zYTOMcx#
zgS7A;-xkfhCbv3yQ$pD<w;pS#YhFq6-xcP|YZYVdfW8w>u=bVf<PFU{V0fgH-tlk#
z6tXeKxK_f;-gi(qSw5eudtjX9RD=}70T}9a=!I+HJ6!T}RhWD>n{G{d7bYzOyPr()
zIXxCH=K!tWh9zJuNHF`EK~GE_qsPpMBM{vU(xRPo)-ST73=&vfMv^QGPn4wwFJNnv
zkr1{5h;XuqNWh=+4+NTi1AEBLKx)aG;@w}+Oae;OB2Zns<7PjsSHMm)boZQjbC7tq
z%e{_j+9}=;TD0)>BO3y0Itnn^3b?GbufKolotIi)A<1d<OW}&?5-j5yfA${2FHAgE
z=bmdvG#)ygdy(Gp!^ikEYwx#H*r{+!g<#29mRjQvh8oGny?o@S$q$rUJ^2kc*`Jw*
zwpp%V6-4k8HUTV*2B(4&_4-c>Vyx*f8Cm``YRkFC`yNN_fMl8<_i%bF5ji~;_w{vx
zy+1NyVJdK=pM7i(KVngMy}RhVD@;~7lVlMv<F{TQ<ja09N=_!i$iiW=`(k{>NFu@?
z=JWB2)lZIZ=}F?pj^_xeJ6!+$lSmFa%P1WBVv%L}ELCtCUoG<8ID=KL=DIy;U+@=B
zN4XlT;+6gtaSteMCd8sn(M|R);g^55a}a<~MH)v0+VA|RH>))|o%Hb!L0@cH;7dQg
znAr}q>sSm(jR{5~KVM+UVxfjOt}<oc&gdAh?k$yiQOPO`K(!E;ImnrLa$niF?&GHL
z^c$t`+8BTQX<DUnjJHlY|4~&<iP&w6wOn!h06d4Lp{LNwCbsTfEd1kNpGQLXa~4Ff
za=>;1ZD>x3v^e$pXxSitIKI#s^FI>;do2rmF-o#VX&cA3!_));V0qr1-prQ_0LC&Y
zfhGcugj<Vi8Cnu_9e#Ex6W(<k-HTP;yt;H^5KPhtk^F!;$ltkEp^qy5xIwvX(O8Xs
zF4nIAEkH5@<NCe;6eGSi0;PxMT!l=D+QgND#fhcIdN?+I{hQj%?)>_#?Dns2X-Rr$
zADI9^1j5^kcN0}f{)*$Z_g_EZCTscjT)>}~`itBLFLU^k-4>|IL1ESCO|eb>8`(vG
zA6oZNFOj<i_msgZ!Tk55gQz8-f&TUNl^ei8M-&)UOB#h}bnLZY_Eo~ETfu?61(m5U
zKPnJQ$s?--e)v5mNI*Bv2}AKS>b9CF9q>vW<d8Eu#B#KbdfGPL-(u2l$9u4p+4Y=#
z|7--06bA4!Os6uTu9!82y51vhpds`M2R^hwQ~f}lfRbbt=%s3d!k545C;t>aOQ|>@
zl$Zoy-ih`&<Ar&z_^Js&gV+FM4wwI}xZ|vO>7Cl;HsjuWp+MoZj-;HN0s4RhhTkKX
zNy(8Umr}DeTp|Gp&sZ@2_v6Zm$n~D^L1sgy;f3~7nsdDW>;?AMSr&JqSNiBg==41%
z{B7S2fDVaqPFv~g=jV0=Qf}FCtl6C078F4CN>wj`fLjVx1J%A)UK+^{*7GfVGOLAe
z!sOtKLZet&lW1KyBW8{S;auB|IvD4I9hNPFAlkIGyx-v7QqjFH!On;5D-xKBLdfyu
zS*IYCa*GO%>_9<FjISo~JlEaJaKVljg6h^`0GIXFOg8?AM;aHCFQ%PSJDz<rS!<_g
z2APTHM?7l5-^ELtAJ$Dc-2*6Sana}WLto;HSR+ORKtUiQH<rn0Z#}5ue)*DG?>jwX
ziKvV8!aHv4=V;Q997}*9+%T~HkW66SqHK&vv>faG7F-upb<?_#InsNJPLqvN`W0He
zaIDriAWkFRSq%-{Um<q*2HNk-bwg36E#YEM2?#qoO~uTbT0RH71826%#_b}IyiL$C
z3sU9B316=^lKL32iSJ8?#ux7Bc0bdp5XEvsr|DbUe{cUqN?>#504()u8yD^x3P77i
zIRWv86zoE!s?K~k5WL8HWNBgv+*m1p<@Z41=}a7l8$13bVcQt?rwy=2X1dYJ0L}0S
zV}7As-J|&Ip!`JI!@F)0-PSRu0^k;RofKEd%NbqBh@pCu(unsV*I{54{`ERqAt|D6
z(Qt=tL+#+~h6csh-P$*vY15u2=63&Fxc2~O6q9qMFQ4(hjdO0^GUFT@$X_QJ$DRXS
z)6dK{LH0BeF)UTgmTci^C6cj4TAGq%sZ=5`3m{Xybsb6bKJjp!ga>MT5vo6IubCw#
zay?I)yOpMEIj(7R5-9L6?c1=?q1cgS$-e%$>838mj_uid*t%@^{W_ayrLdJ3vI8$X
zI}uhQ3A&87UTaOHcV<>B&%0|ACdCNCKZaEM16=r7UKvL3@5jPt)$Uu@yborQe3<Sa
zV&V+RNGfeYNo{4rN=!uV^uD(Cl>N|_C<0{D)gQG1m!Yx5tNR;H0vEJ(M~@B?Q!lO_
z0a{?1V_A^0&RX)P0IJV7A@(yBk+bFp%Oq?19Qe`1g9a%_s+*Xh;o^}(@^Dt!@z{oh
z0j+$$=wG)!EbH+x8PE(iNYPY#0Qw6kMZyDEYO+OGAw&Q*A+Vp+#SAKX))5ERlVRd@
z=4h)q!21U9#mG*P1{d}B?tff3=IK?SO6<+V_CO?RB+cK^4{@Uft?Jk7bV`TUhfAX7
zR{F9l^Vg4AcJ$Eg-e&@q;2_j7KlOsP7|Y4+$hnqE$Z(~&T)sEXCeI3?Q}CRU0HpXx
zccnPx;e0ofVH%S0is;X(`P-F%F%TEW@fUoDrMJ#rJi+ntd#Z?UN2XQgZ`)t-dUJG?
zq@ks8X^bm7Yk0d=aN2a#_o`q6W{j>#eO-0`Z2L`@-{Y?O;DUY$mo8Lx(vOaJ>>UTA
zl*XQJ$^NFuMKeQ<%i>4cU|kTKOb(=!=XRSn!q8L)!H+Hxf1E9Hp1q8)mA1-K5<i0S
zVs8o#=57J@1N)=)jp^g{O}F$XzwaNsIAcsK)B88JRQe~w^HVlzuF}=am}|VKj*1?8
zKV%o38UX*Q{p*bb(mZBKhhBL=n>RuSSVe(gq7Q@{c`<+E3IG613r!x(fJVMs=NhrS
z4Y(pjOm#UJ5|Uzc5k|(d;1pWmP1=oU0J;FTe?|PF5Kc;#-I=5US@vFKM_U;+s{5d8
zP6^q-rz(9S#47|a>HVf#Ro`cS1U=xk*#dDh*ox9WXR$U;);5MWK9L*LA?#rRD=V^&
zAvtSc8!%k512RM-O`!Jyaxd!^T8FNkaRcT&zI}-vQ%HuPTsEF-ohvsCl<iQz=b8hr
zpQ7#NgXUqQHwbY&J>srgA0#)@J922Vxvu+Ud5k@Ju^e~n?BSW<;}o3VyT=<&52xRL
zzpzf+U*FtLVW~=l00})bYF$?DO#P4gM?74A>Y>>S!M}l#&YdcWiSw$?5eVtL%JPC@
zNCvB#{<Sc=$>srKY4~EpjM$8JifdW%TvWc5b7!^K83YD1QUyxO)jAm>kY*4oQ2=UK
z`b)oKDF+f=v;PWb0R=UnU@A=hJV!BJlt}tZnWY~TajtELobpB9B^y#=XMWyZ+51-e
zORi`Zq$R5V*j9ky%|?wHaMHcn1`<i#B%z+0x*Wi>v2TpP(ob{!1oHf5Syy$r)^S^2
zkutff<ZEXlPfT9I_UFNAwMWfoDO=19tUV?!dE1Kl>}RyCZ;yg%&jil+d4`!Cfzq@&
z<a@+?CebSKw$@*@^w=7!dfz$)_v5-dO~pfbCf|52vZbzB)w{gsx9$meEh<tYf~BW6
zUbF_uO?+?htZA_C=rOTIodxzC891$)v#ue2(tsrB7$3Ll=w1dc&b;&n0QM!<V4<~j
zeq5jqV+<;JbHiXUYk+IpJ}^lFCL+>K<2QV5c841qUlcLB&|V);KaKJNN-~ZaxB*4}
z9X+(Ev>w?4I-EL~u~$`VJDQlQAJ-Y{#R4+~o`TE6PLucbrYx!KFjgL575n!AzaP2e
z9|7#PMp#~Eo+t;3XXuG*XNwcn!ZX4j1DS^>Vh6OE3TLpOdC#iA`I9@N2Ji>dhP%AP
zO0LwjzsLK#WyI5(rkW=26PSbhpGVDZUise)!c`AeWffA6EV;1>;-LVd4gy#JANK1q
zTNXu`1VJhSbkY{L*F8P#hI)C$gGh22_vGIc#^ycj@-!1i_?eQPPh|hBy(boY2IB8S
zNAvf12b<mJ<a^>c;rx^)3%Wd2J(f?S9$RqgHGFdS_S_!3vSL?+RSQc%ek>cU*AA}z
z@m)*iNS_s8_LNW7RwZu1yvNE!Lt}W2-tCzLG$fA+*hA=xm@2p=Gf`%m&(aZrwm~09
z3_%4`C%V!v%1p{TyiwxVA+pGP+Cp%r?hSt1JF*V5;&qOejOKUDnEw3*k~@@s275dU
zCD@2lA7Fe?dY&fO4I+x!<k2)=>5pQ%7WsvNqw9rJn7W69#oS&+0|pLH$)Yx7WY(oD
zS6$;9UcRA+LSq+@z!QV}2;sbB`P}y6YlFf$Wk5DY5imLO(JocACB8&1P({P%#j)D3
z2-VDr?&yq7mC4Ll<{A>>RH6VFLBgzQmNbowfXl7)LN+JGra*TdXnPd_&G|jT)jIZ2
z9cC~z`o=aDMs%YeW_64M$Yr)8@WnWH;5`yK`bM@CFL6+-SM2q4yBt>;bdh@{(U80q
zJoVaU2GJzhU+^*ix=AQ_o~#-ffsRuoOi*fH)320=YwToz$8_mM;m{~iVij$;gm&50
z;;d-x0UZzu?`uWhWb%0Q>1^blu`H!&OcOj!@4fIR%Ubhl$1wYssgCV!52KT}QR0j&
zCgqm8*hZDgw2z*PO~`!R6;M-Goixd0$^ZS+S|fj*G}YimmgmEc1I^}<tSjO?uf|`}
z6$DP;I<R8J8}F2N=#m3JDEeDY3Y+%Ki4J<g#hM9zo`Km(sGJ-c8K;f^$?LMbRhQti
z0Jh+{^=KHH*6-HsH}r!b#w<2JlZ0sjM&xFoC41yrX+A5LO<nRM%huIw&bG~5Sp(6V
z={f4{`39Oj_a(E&QGNUOuYDBXJr;a#k6vQLh|wzPl<qyra~GB27-#l=S;S&kImu_J
z0X4MLcrRvBaXrAyMjQk(oycNY^oUie^Y=o+YLV{sWo(QFe98~GGD2}R^%J#=UKzVI
z2-$~hE9SM(56q^&4OU|mFugb9z^!@kNCil=Y7=%(rnc7@E#%t;#97b$RZ6k!2zqE&
zF+Cfm`R|`s&kno}8@P>e^cX0+DY-eB*>*q@liMXnpn(|y&g=$-ao_;0%2?PJ#a7N`
zX!1CLl?)Z8*qwP6fs!PZz*7D9KS?EUn`I92^$CWx_Za&;zGcPA?-w)uu*yO#SaeAm
z@HU;^P_+*qOnCFfl(V1z3D7Ql1@F&a+Jb%u%x*tmKp*g+c^D{uwa5+E>wdAyBBI0?
zHA&(wZ`?Jq8;w_==#Fj%K4VsTw`6o>mvbWmFl5T+z?hs~=%x5eOVzl8oTw9O<*>Xv
z6-h11{Ea6^LQjY<mBNBkiifs64f&m}$I=I4x!epm@7*k93wu%~+ug!d3|iK;P`t@g
zZd!a7))M@Me#umJ_-*F3Hy}{2Q=}&OP08qI#;u@_z4ZPJPrU0%llh1;fhNyc{wX#R
zR(FQcr|c<%s8fh1|C37PyW0}n4DDDWy9~pPJ_)#Ih*WRE=!#!lu$R6P2UToyio{`3
zkU@|7MDi)femOHdQ2^^GeskAxqM}zsVqhvGZrjVB<~gCzrf$O@Ry)-dF6Yy}P1*jo
z`Pcvl*!U8dH!frjly)(_AL71kxM}=$%GrwG<yi{158OvfqH5)bl~7n!C@YYoqy(X0
zc1y;k#svCa`W>j?qiDJ5ftAwSVf~WewM^tDAhO(HCGg4AQgkv&RN-U=vO6z5kPkF%
zNVc={bp#InlKZm<aJGGz=VlHjJQw^G&^GQ0jsOPo4G-cZbL*)T&T<;1GN&aaKk1SM
zHh14Lw!5~E+4}sm8ZChUrX{s?#Pah}X+_|cxo*J;{No604J!Jsdd$PrB=JaRy-uC?
zoO7(5o(dxqQK!-hDPfJ)SC@nuRfr*1B+`_ki-bgi<Vq&?A(opA@*1nA5Hl99fMwXr
zG09qd+V$tfcH_0mfi=Az_J&^XzT9AYmapxvVp3YT7@jr+g4B*u#a5GT)@-1bY&mB_
z<8l7eW8dGYn%hzCkCfvA{y62@4?kr%YuZm_xM0=;kv)!dhGq7_%8~4lMOsE4MIbN!
z(3<coo_#!|wokt3)W;cOA8D`Gk+Z&{jmdP)YH0%Iz3p8(dB`!FZN5-9#`f9~+WpK_
z5++V%)O%dQ%Xn5NLpTj+fZvtu&?69k?d9t;|NkYCW>-i_pvP_lym4-)?bOi&KNJ5&
zVa1T`cv8Yxb~ac5t9l}Z(g9X~>{tlT;u(6m@D$ex(lIBJ-zYO&BSC)VM?L^77cQcr
zGF_x;Zx=wD9~o`+LN2GiTjm>#s!FPw?%1o3t-4CgVW+qOM3Cl>`SvAe8K>~qJ@PET
z34>D&%ArcQSLSwUJMO90F0DEOBKQ#UOT_j!<90o|;U_YHk(qNj&JGzQ^!3}9h}bip
z>e>8@VB=$<Bki;P|Nc??g3cSBIgzPR84HFToTjy{or(<@X^EY%^?1xfxknG#FZ8$9
zNtFfC;RqdG=rD2vv0AjE^SPUBWbfNgL(%7U+kmqF1~OHEgrY^@+=$3(w4q0g#eiOR
ztIuxx6GRObUS>+RcO@7-9b(2ZclyO;Q3bEOkIE|=XSK{odR>(oVdvYWCvb7o8X}@-
z!;rGj4)?7Sxv9?1v!Bjiv&rO%J!6Vf2at%9g0;Pnz4W=+U&m7W(V(3k1;zZa8?62;
z3zf`aU+RA6t;w-|ITG#ER4vk$uT=TBVgM{&>;h3^5<rbN3}55r^Rx)G70<3re{pb8
zoKF6>F>LP|vF#}I%`LrpblCY!u;pHp-WFM5Y-I96?SjM1`8o<u@s0!QU5Xa81W?^O
z3G>pL^s(ZzsvD;|i!$>RH@%Kry-pkM6guX1awn8WT3WsgtS6qx=;h_@M+LCafHUxq
zY!IMs3QQsryon?9q;yfot!XEs%q}-i@WsT(bmPhE;yV029KqO7(NWa8e*Xm{Cu_|R
zN8Md3WURP2{RPdzJLFESQ_W>9bW8CCywB71O7LWtH$z$HN-wF01lj4-W@VKt3{uyp
zKwL3ajtY5tH}iuhe9Y}-1A3Hzy*YU3Jtxo-=%-7DptqG*6(;~xY6huHE`?Z1(!^Qa
z!MXCj^N8HkT7$R!V7saFSTK!`!`_p9^&<StnrsWm-A5+ODiZ%quV0KS*-G4cbj*G7
zsNws6|2&dsU$8Zf1#6WN4?IU8eMZ^}IFdx8jho3(*09N`moTFI!B~PK9Bl84GK<kM
zBenv;yz?IP8NcCRVAV3qfWkHqX9a1HJ2l5ule*~(hdNkNCC99&sQ42RV7e5?dBh(-
zPIYiaB+2U~JKb_l1c!m+3)~Fi*#^5<tA%Pq8FK<bWwf$Gsv}2IUGXULVohCPG_(8G
zWaO>T{NVgJ_|q-(#BIx|5wDYTCN0rAib85quW{eO2#j{Zn{2S^LT|EU3OMf6l^Sm!
zW$vETC8{XZ>_sa@Dn)=u$@j)Z7NZmmL}K|^2|BqOmo_EF1g|1;@>v;yXm=)cX`Im%
zKcBqK*M+*VzJrT(?)`6D!`#0Qj$B_~@E(2KO59K&^PS1;kId@fzUC?^ar;Dxj^<VW
zTOw{Omk09%w6V!YhRT4HvAWW@hG@TxSYwVf$x9YVkp5eIEzSZ+IWw8BfcY6sgt18u
zTu4mENg+*})+16B2)P>+QAm<gpiw8?NqJu-xzq7WZMJF&JEZ?uP(%~z%7Wvyc62m}
zc+#3v3WOS2C9i6of`eByCEtiV3M8iIGhtbZRu2ZOlFiuU8RPiV4zI}LTvp`vFgwq_
zmAj9#$TvDA3)qn9htfy0-cmuiz9(#d*tX|8?MQxz-6H$}V_#=-&u@kS&i#giT8pf5
zsIZ!o_zZx*knI@h{9}+HVm-i6s2$9AOB6m6D{c>M%{4=Wn{gX?Q`L^>8L6wozI+;q
zgqr{%`zkO?PihBsexADx#DsoV<m*-bKJpU)E3}`}2I^6`o30!UjV!pmi*q|3S?3p%
ztb$#HR;LI_?KddA-8IY1{Ohac_68srOa^48zpYVV4!d{gWCUw3$D!vs)L8^@MYh*c
zaClru!7a}Y7htF@&uaTO9Cn|5DpT1DesqsO!xFPQoaK;`{mH<J??zzRIB1QgB9jUH
zLM^G|9aBM(PO>rME~amedY<1<D6)w}kD*h1iqML$DaS)J!=O+RGW+eaaqTNyx2t92
z;Ws5Z0`N0<d{2;sF-7jcV-I3PM_WGhW}k(C2vl5*F^s&RC3CHc?`>H*c^K2Yq|dkA
z_-z>BJBH7AB*y3DPbF~gh$^|PQ`KWTO3;aW#<e=^8QfR%V0_M<5Mnwp1kgfH+}Hcf
zJ*I#emT~_vee4_;1zafddym?;M`X_a`)6I_doL@{!p$3*U}F7%U1RCrSp)C`fP@UN
zk6&i4(*P4&3uG4)2*w1Hzm2=PYHiWm#esK2_+pSj`28ScvsoY1HJ=`?42{+)1GK#^
zateVh#nIfPebyzRyqCxb*X;yDDg{epq}FIMLwY$M7R}yp#p)`>-Y0L&I+A^4JJ*B^
z3Nm`C#g9bcR~2tkJ+wMweDuy)zNQUvtTER^Hd#Idouw_aGPd|IRQlAjUgMGG17yHn
zow58NlKW0f0!Nh1mjpy8l#P3ujvhfI3yQ=*t-7HiEMveglR^g|rK&Hfrg!nNHfS>H
z*YbA@JIv2do(ZmRjbNvb+UXN_LKP}lE@U?6k5(D(T>!Q3Vt3?TUq&M%H}E9>_PkD_
zuw6SKmZ69s&_&?*1OCIrK5m@P#Nv<tmG-VtuEuv!s=sNFVNNU*5(?M@L|Do<%HOLR
zmmaXF7+%}gH;T=m2eWdIu>zC~SHz%b7to?-B1G?)RF-Xfxz2~#8yegIzsp+@(}9gA
z2|<`DvtK9Dx+$rSELa^@d)+fD3Nz=J%ck3sHkXZ`AUmYMR3$r4gv=E<j0m8)ebKJ}
zy##21E&19b`L){tP8A)`|5rh5#a0H8(lxmV$|sBNqan?%2vSlwQ$Lt>T#hX5>i{^D
zfh<<rl8p-{PXns@prXLfvSOh6kL<hVo-=pgfyv|jM;G#+^xmqu$-BM=zBAKD4ByVt
zk6v(fHvlgJLxC0mH0Xf^=Yl=)`+ol}s0HbHx-NV~x99({=$X51PrmCG$sI|ML0|pX
z1qO2y@He{$X!7;1xIC2grs_NYcvr{@^dj`Y=P+|a*ju!ts}K@ki5B1f8~KDAK(h}s
z32hD{`}WpJmOXg&?o_CBnMxldv;dg)fKNwerz*i5>`XaHHfhR-%3g&ADt!Q2^0NP%
z1e{!oe!TfrXTd9r_65Hh<FDW0I&_x^G*1%QK#X02vo5h&kVN*8pYm3WoKzIj6wr;B
zC>GQ&n$;mG`awhULe)_1R3L}u*qzaDN_bJURvw<068Pl8`?!7oGd7@C=*R-dp=7D%
zuEWD>E1aMHP=)3ONq1RhWgD7xW#8m#yC)WksIs&PmTm}N<s+z$^%u&bLI&>IVduev
z^_zoyjh`o!L6Zh+eO%cjT;18#qsEqRXXt$wc9JPUr%LJ~C#jm28f`W9b-k-QU9zFQ
zZZs|D_O^Cbg2(4MCwHeYm$!$H-j-%O&Hz|W2qIz%joJYS0&0AIAxm;&n<_fKo-*iY
zWqNQZWS8AURjeK&qSU}ITo8w>_1;L!tfD!eY?}zL?ri)XE%{;cL_3>-6x7i=N*I(h
z-j1fVB)FikR7t}&9!&E$;62X0e5$JZ9s^Ou{|MLe+)tqN`0E*1W!*Gz@5EcIWpCK4
zw0Kw&1B_t>F%RcQpXDLSZMtY+Fc)B)R7)>x3Oqg=R@;p)^c0Dl%)yQ`!s^7$_|CW*
zRYI<^GJaNSpYi}8)3WAOOJ}+Wy$!sfQoZl?4G3AK(a!)G8jnoR;w{IN5dm-Svn)45
zmod=Q;;Ant1=SKXbKRz!cCubcMkuS=4on3yqA>Vts9emOHMB9`E#6*Y6x&BHv!9JP
zrfri6@A{TN#9dtf5bW`MgcWEqZ!u-O+<IZ|$l>VgV2}an?G56crS=YVEXPu52Ff39
z$RQ5YnGGW)k$Z8Bt4NRK&rV|<Iq}mASpF!3uyD|EeE_#tJrb5dQF@1uGB5aFuIMek
z;)CZ*c`tHO$6|Qh3nHU?uRQB1Ydb^x7w17tzJfb?pS211cCXY0yoGz`i!r}O1XlQL
zRm1`Xt$ky1S?yv|jI`uo@Khx&{Yc_CafNf(_^Wjy1~mvZmh>a~;~(GxZ?6$XEqWPq
zvF?`RLo3=@&RY)E^vL$YqxY)L*#gXxA2A>QU;Xq^m?{sJ8x$PZ3N**v5Z$c>%b7q~
zL0)NUsWaM_4s8VzQhy(c3?_Vg!LEULX<(IDCbvY@*us)VlAV-D3f+R}-p|)Ro>zoV
zux(VDF5G?x@F>zOF*C}NwSzj{H#eTZ)2f?lS9;#r3c-sPa?D4`EcUhbmwhjPecvi?
zBkE<~l{Sf1O~qmuH&RDUTt0CZ-y_iewb`irE+1z%?f?v*LLPU_>}prSV@p<-)-ryg
z*|uPPiU`rh*fb%UzaFjBsEf&WTZa7^^ZR#qSjQYoutG5?BjBj;yYM{{;8yw4bMhv~
zE=SBMgNNNoDZ8lc?PS{IgL2sNoTY6)%)TfkGOT37(X_3~sCl^#tQ$0C>VydJF-fyn
zjCKaSgAB#l)}2n2sNSWw`^a;PJ$B4dH{w2I<pV`a$J)a`L%T;jFQ=(vCDb1-Y*`S3
z9sH@Nm!ih>JpqBtag61W?*>%vW(GYUAD;4p+l7HS53efePA)86q+Yo0y5r*icAn&p
z3)|-O$PK*RHm9hXi;zNEC4;rXWzKTDM)h@b$M8ij$>;lGp}pQpRL320QAP<+xH|h+
zW`m8z^9-JkBBol(7jkyWu>o7GfAmPsxysEe6ijpXizvH|)Th~kn7kUffX^)Oy)WY1
zj`(vUnEumlxwe%=!;ZzV*NMEW>fwjydGBNCsgD3;u~!>&1jvEvft#?$VzpYHt3sZ=
znvED)4gWl#{84gZY_y2)M>qE_%HwJ17RT;TBEHxvtDTrGM`gbIRQ*?)oK2VE+vlwb
zKr!fFF}z^lC(Um0M*$cnVi~5Koyr~dHP94@Tu~D4a3$C?0`6|o$wzl8xGBFmYU-(y
zGu_YBC?GYo<ny(DR(#=Id=-$g9QD4_QsNDt@k44<inT=2NZmwB?}4ZsqDjPO&wViA
z9p%*P3A71z^MdYm6O3z<>7x|y-j1&0)qhPSXdbk2mThDgM&}dg5g9;sJ2V=%TQwbv
zGp;x7hsEUu9xIgaJu%tzPvUsCOH{W$-wN}5%<v5eC7?H{GTo|_BybbgFnhC@=L#TX
zC}up2_vJO)lC#%0b*@20Y02}@bg02b>Juj2uT><mpSw}?V6|a7Jl=ghRRDl=eL1@(
z@{)yMLGY_U_tVJwRO|-W%Rv2`drj3H(qGv#6K+Wj`_%ptS7>F7*#CHn%D+swwI%rA
zdi^qgOl?5%Cb4Jbv#b>tNSr3>98~e#G;%jXy^e3WBTGTlq7Cn2^l3$p6aBud8vS$y
zq(W7-u^>C}&A=rx-P1T%GDF2uW~61u&6uQT0vKf;UClVWpzD<CG!`N`eoCZ!p#_b;
z1itsYm*`O<qBQK#D<yOabCh|C0sn$<PFp2fl>Y2t#us0Qr*z+XUBO<s=uHKOId!Yq
zyPai)XEQ5ye!tNa5jQ@uo~F}U^vlpLO{#pxLGMWNdgId*AFY+InGJW-x#y{GcMcx;
zJ$<21vJ0*AmT5oL{=EsZSU)ZO?ptbotWHzppL>xRsDY|NJSjU*75<RX)Th^Ei#-+h
zDG#(U<?-2x&%sW5lFV?%{rJjx#`s<n-^(^VQEcd-I!?JC$2@P}1SV1~y6fF2p7zvr
zYPiF`Ui1BcHEpFk{+TE2?gCA?e~Ll>{iCokq6!44`D$^3yA2`{WbQ}jpqsw9kO#x#
z1LNkTN|T?Dx{j~Mo@cluAOaDA+5&Hq*E(A50UD1`xR%HFM2{bW;2XeA7w9TcbE?23
z()eDY!IR$anr<ZfH$O{ubc84(p8?%^&VT>R75~)hY<H{W7mq-1$&7Ik)yc|hxjy2d
zus#@cs^qdKr$kI1JViI-hj}|L0?W(us#+52R}N)t8cSfo1*a?<E>RzdRNO%`F}SyM
zw$rC+ZD+jUcb!ZZ;h^wi(s^&O@^t?o>!c=cQ%cy0@j`VAyAf_)MxR6~GT~<(ujd#U
zZ#SD>m>x77<IaE_ht#97nwpy00O8fciz_lguVU_6WTG@JnxySi??xyYxsjzt&}`>^
z8SK%wlTS@zz1bsF3rqta;2Q2vCpY0}97u2l#MiyM)u)W6O-{TV8phmc0M4hzhSo!e
z^J%@gNV{wQ$klR~{;^hbLj~j_ZQ~nz;MNaw)2(L6^CHtIaR>60uEKx+u!|oWTMz%B
z%7jl~73$)bm>)$iUVm=UPY!Td>eeWvbyP3G|9h|CzkkGK8qV_y&f$jGP)0N7-fAPl
z*mwGMLANvhO$*=w&&oFHQnc&XmV@eN64DE2$`H5~Ouezw<I~VI1^1H84f#*LGLgvQ
z+jIDdJAJ(ih<?@$r=0XmJ&!wTQLI&jiD4_$ozgV)!FdHzbL<t=q!4-8;nNf|ounb}
z2j4pmtXHKBxACr~Cihp8baGKYN$Liv38ni#y5(PtL7Q}5!EP2!FVv?06jxBxf6nHu
zk#gzRrF=PHTjEAHzvzbV!KqLok!T^pwiClC>)htZFXMn4di6@#*z~yKU$`60ivZZ0
zXKm%be|-7cDYX4Q&a8=h`C5Y03kAQScH@Im@p(-wvn8i;{k{D;(1Wf~NGM_#xbm`z
zh$z*I=hcTV;ILmMGVK{>R-`oRFyebcVV_321Wgh*bzyDDXZBtSGX`;4L8}LWF1gSc
z80`LAFl|k72Cy@%^H#zCDL0^%e&1T#<Av2_|5BAj@q)>rFMYt|?{!CAG2jLDHCwjf
z34+N+s0hSwnmvoDL6In?)oiWoB&IxSfM~%)YW@otS1g<3s;P&FIJxY7*_KrAZ<z^E
zGW$SAzmL>50%o1MSml1`XKX>Ov~u&P7`QK^C`?iKw~}msaJu*sOV)0xK9cM70w!`^
zRWyLAClZ4oBHL4ro=<!RzMnU=Wc6aMDSq&8IMKTU<%w5`+D)B4uV)$^FRI|y0bRHI
zvB_Y!rG86D6n?Ba=uz*zLcJy1TES$II!EGhD6I91_8I#OO~f>63;PX8=n`HBTYeEl
zjSEd)5BVmg&{-k)(<n9N!Lw=RNZhp_cDly6fOf`Uqlr_V4<6Y4TXT>X){lmK;xpU;
zj{0)0k~8DR_rY_yn@bm{8qUHx5ZeV&iO?(K*?OUe{t+7__RSBIUA*F&F@O&`&y{~s
z6C^O8b(I~)V*5f`PSo1q+E)hC;VvBU@<m#kqr6UqDQ8z$T|ob&nD~)tm)z^p;<tI*
zt7J7RtdeqIslP*^n;5`8K6U$?-eRi@7rqPs8ky)p*2=lNbaBe=IrS}HjF$f)Zsti?
zs1xuWOu5`u9=;_wG4#~LQ*NJahSz6D?Su>Kq3?CQK;cFv%X29mu=C8)#k6bp=RQG<
z-#e5_w0s<WW1Jp35Eq<!Qhl<-9Z^jvkM(=*%bR%hk;aX0L`R49RKjiBzSElk%)#Ld
z%|X4kMUxDJvdM{xqfF~{r;vpPh0*iuz^ZOCsNS`CL?WVS7M+M;-QhIKuAo>x7`|ee
z&$L?an)v+IE$tpkO=y<<H5oasFMt*oswKh&2Lguyi&9~qR2)}fok!+O6P*c^XPNa+
zEetGW__!Wsz{$jVgAL5|3#=5(E(q{hWWtyh5`1wIaP?U6bB+z0;QwWL>bLvwv!k^@
z;hydaB4+<<>gc`9^~D>`7xcLdLgxALmD1ciu!9vqmpq#;G`_gbrfxB_Iq?o$pu=X$
zsyY^@qQhdf)6cbS9k>^0rV^O@k4YbJCrFTh@d2P=VH-{@fP7Ni{%_~9$m~tJHSeMP
z9Gx7?hT<us1SeUq`>pfc5U24xySehrk7RiKN3~x~ubQax{ASIO`zW`>Ke^`DcX)yn
zqWJF$grGsQ^x1ImO3FW2F(<0i7q8}i(5odu@$KS-3#%nV;0*9QN+cF0n1qDBrbQm}
zG0jyBp;Xrr%mqQQDfjp*K3=94PmCBmp|u|MhnRcRAKnrCGIe}0J{j$~{OBh@Gj0QL
z48BNxk=JN&6u5U0+VWK|$+uaiCrmctyQMWQFMPBtMO;u(OhE2eaOUxDN*J%Y6@JFa
z><_~=nI}Z-n-yXTr2Ngo&lackZI``A6!nVk$mgZNVTQd)M%9He`jxTAF~!++)%#Kt
zdt^(IxffUa87y<cP?~#O-rJwULnNQUM{7OR4_{|Pr~EI{3LvHX8u3NFR3FTt>%Q`5
z+kEoAGuNJ^2jY328`y?3ml%+(TgIz0z?~zgP$+xn@kDcB2?I+S>UITAR;|<e_$JGi
zbiv8>q`)5XlxFJ_3V3-g%w$t)Avu6}p8dZSsbj%*B^Lp2?Z$iottZs53#K!3a-t4^
zx3%yy%b=H&cCPE!^E%e4uGQjxQz?EbTq0r|_R{&6FBP(%B}h!e6u+Bi#)lACT{BP1
zgTD<f=B3k{MhWQfauTyF9APZRFfueoclMhF_ZY?NFYJ;r^1(1eZReVJ27PsMAB+_R
z2qC0}liIEI^)9c|3WR_hnyYTFzhaLM9HS3ec+(CVQ*U+tKAmLk?E>b1TD<>>ZGQ~Z
z;-wKnZ?|tq42)TURR93o>?2lnes!bmjpAdmK!kdJ$c#U|Ww?=>H_@*u;&s-4|1>O;
zGaegSu_|VBv~+^_SoDcbE3n^!Z2f<9>l3WOF34Fx{VJYUZd+K#^GUh3=A-v8@`7$W
zQ+iCHlTB*0rlX!~B3#-6^C2?P(@Jq=H#BwR@<sR1T=#$f+$9|OB^Im`^k0>1j1k;X
zx$NBd@+`~ll2%4YDEE(uTX(<QwvgtF#a4J4$4$uuh6tTAZ)5{H8wZZ{hv!U<8gMu?
zJ2A<|Xj>xkYZ#0Tma%-!YU^=cXzpaaFXpbRojdCu>^$jm-Ht0{+xpIx?2Kzc3g1wj
zT+EKMuQGxhKCQz-<iA{)lI~bTCbzX2hjtB5#CnS^a2^9T#J%L3bYE&3PqCngCVYX<
zC*M&@#jBp^Pjug1byk_&|NG~*Ia>v6P>^J@MmHTik)F^fqHJ@o!`RFD5xRGwwD6(y
z1*Z9R9&Ym<Riiq`qO9(4YlL7D_cRY*T;m4WpJompoxb8ZcA3D&zQ%7(^borT$FZ-A
z*zM<>=e3l6AIeCws7b&WE<&<PG2698Zqo7sW7~5iANqmC7ZLnyES2RNm}eY@XCNRi
zf+?WL%We|EdXvyBn?~DLDnCtjBQo;>P$>iUH(VSmRMuX9&{5)O<Bm)Tsp@CKyMLP!
zaqKH6i&0AN)Ao>`SPwa8>#10x3k`9^Nb?w~Btnnq9!qW$OpWk0GTnx=%JFOr$kvt^
z{_R-&r9ia-u?mEog>}!-c{&-m!y4y8g7K<vm;WHMct$Xb`GrwQCm%WOZMvsaU(Q6O
zpK`1S#MJ$2aiaeI&1}Uu+>6mq*Y@GP7Z6veV2s6~=n$VV!WJK3&bcc(3fp977XA>J
za}7Kzm;Yj%Ef5M7DQpl-S$Ig0JYXKq=9)qJu!W%9uhrf`)L3zoT$%BMHeVIGR;i+$
zvD%;Jy>uG{??xBrqM)F~iY^rn&8rB@BOGrPVy@bkkW7U|>R(}0dYsJkeh@fXNy;IC
z1tUEnIFibPBSB7<iVB8c=!f<vD84a&;A@8KG_TL!7GL5#&%U1cD5=1k9WOQ5ADgCS
zF8eI9VIQPiO)o}aLVcF%`*hAwMj7c!CF6UY1kCX(Y0HpQ{kmB)tItoEzJq_FXW_w>
zM1%3A3)w<rVXAf)3+b+U+Mn&(IDM{^|Ne2wP<`Ex|B6&Vg(UO6C{Plw;*DpH=W8D2
z(#$CGZuUh>2KYDT(n_k`RX%TS5{JXzaop#%gGY^qi13|iGbBvb!FRIAUm*qZEFbbx
z8^7^ER4OMNv)DI3eY4mYgYFukjIN0U<6g+p`yJnRkOj-)keB7gr!w@v-K)b2cdr3i
zbh7Kn<B@%(7yXF_E$HvJr!`rr^O3Q+EWlzuHMw&s1tKbXL~kirszfkaTlP<(yEZIp
zW?P2ybbzwv7fKv$8wu1p=O(GVN4tph`*R|G0(X{6DumS0k*+qWFU)6fjmTR&o%~8n
zx%R&m7TV_b)nzNT|97A+uR8HGkyQiwgl;KP?dn7Yhhl-NVNn3o(*SIV;@m@sZrcRI
zuqn5$eqQy^GB6zE=t%v*(hEjFyNxo{r|Bn%?;bRHd<IQusDlT&Sh};h$EtGm#iueN
zJ@zQA(4bMgTo=bq9p9nogy%*IndY=do|at$MUjU^v2N(levK{K-dB2}LO9i(KKF$k
zoo8qnmkGk6KN9y*Xzh7}ORADD%bD4h_N%@PYR?T5evw)9X$;K2Cd!=AiPofRDlfSY
zwiTSDeRYX^C^qxC$v^=hm33<@^j@wuk5Kv{e}kJ=gaL2l>z>Yuk|v<#3t3l}BLwc*
z$>tXO!1c!V4NPa2_#WH6t-&EGyhod~8m{&g>6^h!{NC?_Me&BTXa)4Os@)UGneU+%
zMKZ1{+2V_;y=m{1ZC@Jrdrj1`)dkfXmqnP}a}ol>Yr{39Xp3CM0tcSjZB;F`mgaB?
zrVSNZR3~4hd#)isOeK(;JD9nv;P0I`)KHo*6EzN}(LI;Xl*whOzQ!H_D~AJLN4mSV
z*c@}^jmE>*cg?yK>zjzNx!W1}NUH<}V6J{ub4X|c$hM*V@1L}4G*IX^^9(eN)HxU6
zFeCb-RL`UMB2JF6K$$OwG4fr&bRI5&(cO-8AI+8cWDF{L14LKFnPrk<J2kzI&hyt;
z@BI5=*0*EdQO4{1hUtkyXZnKrY4`9<@JLsm(5(zCa_wR^>d8X&ou8cQrre#b>9J|~
z-Xq>LO<1{n_N(<8oRd#HZn5>&pg>m_0<jkGeRw4+UH-p6E<LvIXs7L%U`GomLo#(!
zJF8xnDf47!6yPoXw4ez=3x(q%A|eLf{DN14L9_hLX;_P(#Vb8X4L&=gk5bZ3?ggX9
z&Vp>*W(u$-C`HrOC=S)Hl}U07JVmWQ3N(dt5JZ4$oGoQzB1{3qt}incELl4Tbr3Ke
zrntm+Lf^dJ3J&_yXe0SlgK;1pYO(>#z8Jr!VqQXogPR2Tu@A;bO*?h1AQ~!p&)vVz
z@SHiU?j(fR)uZ~{$ftQ(uFZM!s_mH3x6X})x*%nW7*_R{sIqcLCh|4rg~cc2!o)rq
zme(Eo$*{WU9FcG4XK;V-tx`ZD1^UCNU}Qk{Lr2=HZbZ1{FT*~VBhf}H=A1{oU=vig
zVsNFu#w-isD$b&_{D&$Qlrn_k&**HmabbSGqZKXiAvT`LCrX+43Y4gCrsfe8dt+Hj
zck*V~93!b>V&w8trNmog$awnEXTYD42%w@9)etX@>F8WQwG|8z{})#fsPD*=XW8~u
z%SG&pi0FgRtry`5q!Nbu!VsP{tfC_g+B~Fjlhu|f!SzHc<&SaiA#8}cLTSn%A=xdH
zXMv`|xYs}AM?c7d;qKp~p@`sD2cb&cBha*d1N$Oxe`B57V^0l{*mO}`D5lE3j|m*9
z`yS){Si6Getd|9ZI7IZcLkPhU9Sso@m_!&40XBl_3G`itA+^Ar@HErI!#tRfconGv
zNWl>oUpi9hJeV4}88vKV8X08sYZ|@>SRm-zq4+u`@MAfG+!twNZo<IQ-H7b3(Qjb5
z5WGa0MHE-UMcEf*UJQ0##qFp^N2ZE;9))yPWXY2x(Fj6mgo=iPPb2Ud(A((rC;rKx
zV$09?^YBf+OHM}3(SH*&iS9?~qoNMmWgL!Gcv&ylq;4=~@aaA6R(cfGBt}lyuK0*)
zb`nbA2C|Wmce`6nzXIfDNg0e-RgH<|-4?y{-(+g#w*J{++N3tlU4J9~@ZFK3f7sVX
z;ksD=0I@>6B9CELM#jNW=*$`*Qc&Pi#f0dHq)70Hk9qp10_&W$F;k?{Oh2Kw2A{z)
zpyY*@{V1uN8iS6a$$o`O)nawZRg8IGH;<Ee{EPFi@E_-21SKNJxsJ>Nd$X@3_AjBg
zgnCQO2>WF6^hZRT5Jb`wSVA(i-lRzKB3zvbo-B9{lzWxrr@JGQz$CpFvE)RF=<#9k
z4ERLuQ+XP48s^(CMm>*Zar{KzWFAy3urEySU)a*_&+MMn?3T;=A#8f|CBN{`lm5nc
ze<Haa;de0-Ujphx^oT)YS)|f3G805yj&@bvgblZ_=#wL6L_~zD3m|w2oy3JJrvi||
z+-kNVG{KE2b)_DIZV^RxFU%?K#mGy9KEWU(a}h8|#D<1i#TyK`LvCe_?qLnGCvbzF
zTuL--bs{=F3hUrS>}Rq***uVhAqnH)Aqa?!7}x7a$)nizA|fz73(#s9FnAAEg>FXd
za=ZTk7k|q2Ulc#Xbn@wYrwKlVL7W(*6{Xw%0AZ76KHv0Nns&AtlXCkIq%K3m%{6O6
zP)lSXU5lXIGEe;=zSubxq?tKJQr_|4W%6o8{g_0`Z=_+d#qfHHzo88hrg($Nv~lc8
z_GE12m8<NnF7%xX^Y(j$rm%bJg|OCF`-2*&Sa$juxMnf?#*ICUgU01cQXBT*NkmSb
z2W)3A(99C-Y-5}>=#ogb*N>C<zsUap1OEU=UlmU${UavqgTRE3gd~>*()?p1yA)Y^
zcab9q@+}{6;@S@kJ;tAsoU_UQ03>No0w;*WWKdQ!?1@aYBINl@_K`MWLmvoF6Sbnb
zEpssB&jVyOH^P?LM$QWSwf0|s<aPz!h2&ya0_Y)dh`JIGi1HpvL5B>thoF~;^c4A8
z5z8<rmn%XzE2F^6<W`7hsZLKJ<fIwlq&5>p6D!bEkxiJ@320j`lTQJsxrs=GAql1r
zvl{&$UkWz+(T3hd+!8q(SB56Lg#G9=<%}AwkeRT1B+fuxn;cymUJ#p%K_5*>L_}c0
zg9R^w_$%nX1r$LU89b1PnpR}VlO{~}rT9Uj40N9eANGH$Y)pUjY?APZEFo4c-ewK#
z!QNldC(2BZQ7m+ZEExE3b9oc1MvrQ(hUO`^vVZ8Ba|a|+>6B_TghHDi*mmV2*NDyc
zmISD>{{T_WsT)!vx@11xC3ny4J>r4C(UaMhzQmzTScq(VIQOBFR>rl+XX!8OdzODG
zk!D4S9IBg5e?{w`(CH;mY>FyQ%e!Cobb9hX!v6pY{{TsUnK1YM!eo!5z%)!RUVA0G
zMjWN~KGbzqc(K$x13r%_ZP@dq5a{}7LgAh)S7bKieoKq&_B@1A^(x?2F9bE+eGNWf
zNgZRb$F@sk$>4PxY@pl8IkCT%UoDAF!F3Yh5om_e=zXYZ<nS>+4km1fUTGBH_>}pc
z;R~1hBJS7hZF(ip`oDr3H|TY<={#u-y&h97*pIeHLQaI8A*Iqb=ue`)QMpP|vWodC
z?P+>5(33Ax6NMeYlBBOBq$Y+f9efG5B*6;IneSW})|APTGK<j>g91xKM9A-xE<~M@
z+=m1Yu^P>p1faF1NXSG)L}0;#1u09WesrZNU&?%_#y73d{zv$W{64kk@bD_xs%Pzw
z2`-WmSv{MMZJ1Ow7AF&A+u%~k+<&7AMX+fxXMj6F;pmKtztG%2lo-=yf*_XdHT;XQ
z5?}ZcF>b5+LRnQ26hmV+Z~p+mY(9*FEdJ{LP>`+=?*^WT-5j!Wsu?-JZY7Rv3evS`
zjmv`ej6J8a@2#taOi2&iI}1S!T8}egaGJ>X_$|k+FX38ks{a62*a)%oKhfwjf#OAZ
zNYDNQX70~sLE=PRB2BcXLM!nPa)I_|V<!z23l+F4V4RPC$i@Ev??nCNHbn4_fo+9(
zgH6b<A-64vK_v7fCxPh>Q9e-n&V=95Z)YtS{SNgjmM4(VultocBggP%!!yV{DIRo+
z+{*9!k8}GWcl$l$9>VA(z)A3X5iU`M$}a{~M?!akK7_=62qx^JwAq&?e2~&gKkG^K
zMe9Zf;i87cLS_tX(K?JpkcCW5Ey-IUu=z3!k+MA=#9fgX?5L=MdOc4;w*v|ehu9|U
zMZm*u=qD^~vm_-bcouV{A{M1H_^EX%N;mjt#_RNFNI{_nh(a5Y{K5YKo`>@1@^CA(
z!e8x*a;c)vh@wj4tc`Mvl=>EGGPx9olxywrYxF*_?ibk?EOKDlmGUJ{$$w%TYkVfm
z8!kK`?M@CQ#1Z#Yz}q<OG@q(|zx##F`H5w*i#MUOc@Kh#v0Od!dhs!eai7^v+sE`v
z%Iave>on>)InfKMA|-sfM9+hqir8dt{)mz;3|+Ppnplrt0{N*mkd9lb2?S}rezfOE
zgW6<cEi{C^WP6F~hob!NVSLxY{#X7>`D}&A(FX7@M1Bvnj4n#9<9(itc%<|u6xW1`
z-6P3T_IX#PcDo#MNS_Fou*DKQB|hahWs%5`^%2mQqYgA-65XNni1Yjyf7l)DT})ew
zJOp>iT)*t>{zowe)nYPxe*rH6yV$-Fevun;`vtw*5^_s}sSseOXx*ef#Cv~2;&;&A
zZY%7h-4YXeDZv%z^e5$>nLM}I<|Vu;Z3y$Fmk;z;V&4aVz6X(hk0LzhkgO<s4EE2q
zL}Q~p85(5C1c*i@V>wJtuy47FGEC8@WT6}j6iXYI!aW5Q_D?TJL_!ds0SGa-(BA|!
zaLPiy2qEx&2>K?5HaB0*N>Y@iGpr_%!B=RhZqG61B%Y6PLWib=o1@NOWgg@ac6%cB
zLti5{g!Foq!b9*4*%z^2W_5of{3re@-(G%C;G=k+ODXY}*ojdgi+WtTFukjUJPc`N
z4W#xY{6S<<LIvPawmpn&l+sPd(Bn+^4jY*EF@-lb*qyn}9j7O0!&VJ_x^9bc+`C2k
z7l^5$GHyFewpn-{aNtdwwM0tk2~_#P8-mBhC_)KSI3_pQ3<V2gDPp9QUj`FVFP!{u
z;{N~()&5ub7x+`;oTgsFF?bQ}3Y(eoJ<RC8vKuGad@A%?Y<flPBY*TeAM$(YN=HIo
zguM~ak|INCa)zzZ9&_G4mS?jKy9;cHuPGJZ^)O}s0A%<6WWVNmpV2+Pi4C7iO^~+`
zppCm7jPy$2Nj5}y5~6n^ZbOuLiSj=&(B(K3<s-b?-H7i00Fg15_9NHmp5M{bC2U^+
zuEx70irf)x*q1ZSiF~p|I6|EmK8xc>I<$n<GU+rz5t0yuGFCa@L?HGbN0K`cSdK`K
z11|H883{?`gd!sAL?H-GDRiYpLujaRi8@3$A|mL)f}#lQsqDr~>-IrWb*PTW()B${
z>q3ayPayMbp7gvYjSzYf(dc@Y(dyJG$)3#i$W2H>6IO&F88jgndq2_F;N#<EB@?tt
z=}8n4ZpiXQ8NqZ)K$NXG8nkbLk6}}6f>D_vVQDi*u#`Cv)fUC9T66J#25@2VN?(fn
zZ|3ypSN;R}FTlUWt-2}L@_!;{yJAvB&@Q9Ij{+hfq#H|j3Ge#`ul@+~&nfge*~Hz2
zo&u$^7kNeKw&TnBos0W9$VdMG32k3TLGk^Lr7P0<Jmh`FIbyBbF%on~NJuzjf=2Cr
z$nYO&(3=zR;(p(`AB2l=J6uW2limAdP9L+e{2dAYj~5%Z4g3d@4EGlI2bqhi`ys+F
zK~9EOrc9K`LK9LnMon?>Q?aCDM3*8i8Bw_JMn^{Bo)2Q6#q%O-cd@!Q<TIU#x;%`O
z#GsNsrVMFHT@mtSW_8D9MO~5EE0HH;b~SP|oSh6;;7#40OiGv=NN6>NOHnq;mf~-+
zDZ49{FCs`Mf)n9NUyLW`nqa`%8$$>lh1Eao{{TjMug3-~yTThbCW~X8In{5f=kTp~
z)1QkLi&weF&3qT)L;RCYf5czr?u^P++7SN$=uCV607Tp~8*piag*IQo{{U!Lb~#&q
z%oVNnC2TS$52Qo4^iOAJQyoFSV^0JB0FWDZVw;8LCG1@UiEu<gQ+9@)1dkJDT*_=+
za7DRDT%)o70Ar1B?1z7{D3-~7gs=W5h;G_Sa*<fycN22wu%8I%MmqRprY2?`2?!vI
zp`sW(i1Q!zno!4p@DE0hV?B_BAv_S8-vwP5d<d;@Cf@^1laYrbYr95EbVlGu8<I&c
zfi2*AiIz~X(vXiLF997DnIj`3D^I?pA|fK}$7J$HWWGozv&c@=N{HY{BoHv<XcV?H
zV9r?RqHIaBo9IY|@}+uYjD#UR6s0M2r7yrzl%*+3QL(YHv_S-urq|lP(fRyHMH(+B
z*~X7m`FZ&BqG#dA{tM;D&3qU6i~Mz~{{RtN$H-@)T%-FtKVXjPB<|m$JT`fcXNsLm
z`Y+??uP@l$yMD*G+3ZVpE+HvwB<Nlu7(BF1$UH(jU(r3^MO+28f3Y?R%#mHrY^Q(n
zC%^JB<rCpQq8l+h8tgB`DZq`m?{O>LwtI!pc#x9mX;~RGCPjQhZi>4{koFPNEO-#w
z^j99tkG6TSAtFN^M244WsmeTuxSklXBjF)FsPahkH*9$t#B$%EK80K+O6Fe!z>w4O
zLU!$w$sK7)+3pgdArXT7sY*vjsa+M(d^AqQ71>?r_8^GCf})gdm5s3lX(n#QF|i)R
zIU9Si%z{;<M1;~qDd_}=V_H({N>Y@%Qk0}>R#sA!rT7m*x+|iFCt<f^*Fca&e<^>8
z^<U(qHZAD#{%c3A(fl9HAsp++$!N|rdOwoUh+ma{XYpU?FYu+qv9FN(wk}F8fg^!+
zAt~@hB137z_C;*^A<ytT;=YEVTW{npsxQz@-hPiLM(zICle<37*X>H2uh7x|00J8a
z)yiyJctV^C`?-DHi?d?#61Y4^fe|fX8!JTAnj%}YLv$Z)JWP(oa2xnXl+Te(+t84N
zAhZ(ldlfdxdSbEYPYqdIvF!9fmz1=;20CP2X!=JxX)i>h@+X#Ov&2LnVq*E-Dj^L>
UC!$H{&tzX(PZ4?|^k1X@*;Pv=jsO4v

literal 0
HcmV?d00001

diff --git a/docs_new/examples/index.md b/docs_new/examples/index.md
new file mode 100644
index 000000000..c36b98969
--- /dev/null
+++ b/docs_new/examples/index.md
@@ -0,0 +1,18 @@
+# Examples
+
+This part of the documentation provides a few cookbooks that you can browse to get acquainted with the library and get some inspiration about what you could do with structured generation. Remember that you can easily change the model that is being used!
+
+- [Classification](classification.md): Classify customer requests.
+- [Named Entity Extraction](extraction.md): Extract information from pizza orders.
+- [Dating Profile](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
+- [Chain Of Density](chain_of_density.md): Summarize documents using chain of density prompting and JSON-structured generation.
+- [Playing Chess](models_playing_chess.md): Make Phi-3 Mini play chess against itself using regex-structured generation.
+- [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
+- [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
+- [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
+- [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
+- [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
+- [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
+- [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
+- [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
+- [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
diff --git a/docs_new/examples/knowledge_graph_extraction.md b/docs_new/examples/knowledge_graph_extraction.md
new file mode 100644
index 000000000..6a6877756
--- /dev/null
+++ b/docs_new/examples/knowledge_graph_extraction.md
@@ -0,0 +1,155 @@
+# Knowledge Graph Extraction
+
+In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to extract a knowledge graph from unstructured text.
+
+We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```bash
+pip install llama-cpp-python
+```
+
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+```python
+import llama_cpp
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
+```
+
+??? note "(Optional) Store the model weights in a custom folder"
+
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
+
+## Knowledge Graph Extraction
+
+We first need to define our Pydantic class for each node and each edge of the knowledge graph:
+
+```python
+from pydantic import BaseModel, Field
+
+class Node(BaseModel):
+    """Node of the Knowledge Graph"""
+
+    id: int = Field(..., description="Unique identifier of the node")
+    label: str = Field(..., description="Label of the node")
+    property: str = Field(..., description="Property of the node")
+
+
+class Edge(BaseModel):
+    """Edge of the Knowledge Graph"""
+
+    source: int = Field(..., description="Unique source of the edge")
+    target: int = Field(..., description="Unique target of the edge")
+    label: str = Field(..., description="Label of the edge")
+    property: str = Field(..., description="Property of the edge")
+```
+
+We then define the Pydantic class for the knowledge graph and get its JSON schema:
+
+```python
+from typing import List
+
+class KnowledgeGraph(BaseModel):
+    """Generated Knowledge Graph"""
+
+    nodes: List[Node] = Field(..., description="List of nodes of the knowledge graph")
+    edges: List[Edge] = Field(..., description="List of edges of the knowledge graph")
+
+schema = KnowledgeGraph.model_json_schema()
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
+
+```python
+def generate_hermes_prompt(user_prompt):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + user_prompt
+        + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
+```
+
+For a given user prompt, for example:
+
+```python
+user_prompt = "Alice loves Bob and she hates Charlie."
+```
+
+We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+
+```python
+from outlines import generate, models
+
+model = models.LlamaCpp(llm)
+generator = generate.json(model, KnowledgeGraph)
+prompt = generate_hermes_prompt(user_prompt)
+response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+```
+
+We obtain the nodes and edges of the knowledge graph:
+
+```python
+print(response.nodes)
+print(response.edges)
+# [Node(id=1, label='Alice', property='Person'),
+# Node(id=2, label='Bob', property='Person'),
+# Node(id=3, label='Charlie', property='Person')]
+# [Edge(source=1, target=2, label='love', property='Relationship'),
+# Edge(source=1, target=3, label='hate', property='Relationship')]
+```
+
+## (Optional) Visualizing the Knowledge Graph
+
+We can use the [Graphviz library](https://graphviz.readthedocs.io/en/stable/) to visualize the generated knowledge graph. For detailed installation instructions, see [here](https://graphviz.readthedocs.io/en/stable/#installation).
+
+```python
+from graphviz import Digraph
+
+dot = Digraph()
+for node in response.nodes:
+    dot.node(str(node.id), node.label, shape='circle', width='1', height='1')
+for edge in response.edges:
+    dot.edge(str(edge.source), str(edge.target), label=edge.label)
+
+dot.render('knowledge-graph.gv', view=True)
+```
+
+![Image of the Extracted Knowledge Graph](./images/knowledge-graph-extraction.png)
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_new/examples/models_playing_chess.md b/docs_new/examples/models_playing_chess.md
new file mode 100644
index 000000000..a0df2eb2d
--- /dev/null
+++ b/docs_new/examples/models_playing_chess.md
@@ -0,0 +1,88 @@
+# Large language models playing chess
+
+In this example we will make a Phi-2 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
+
+## The chessboard
+
+The game will be played on a standard checkboard. We will use the `chess` [library](https://github.com/niklasf/python-chess) to track the opponents' moves, and check that the moves are valid.
+
+```python
+%pip install outlines -q
+%pip install chess -q
+%pip install transformers accelerate einops -q
+
+import chess
+
+board = chess.Board("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
+```
+
+## The opponents
+
+Phi-2 will be playing against itself:
+
+```python
+from outlines import models
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+
+```
+
+## A little help for the language model
+
+To make sure Phi-2 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
+
+```python
+import re
+
+def legal_moves_regex(board):
+    """Build a regex that only matches valid moves."""
+    legal_moves = list(board.legal_moves)
+    legal_modes_str = [board.san(move) for move in legal_moves]
+    legal_modes_str = [re.sub(r"[+#]", "", move) for move in legal_modes_str]
+    regex_pattern = "|".join(re.escape(move) for move in legal_modes_str)
+    regex_pattern = f"{regex_pattern}"
+    return regex_pattern
+```
+
+## Prompting the language model
+
+The prompt corresponds to the current state of the board, so we start with:
+
+```python
+prompt = "Let's play Chess. Moves: "
+
+```
+
+We update the prompt at each step so it reflects the state of the board after the previous move.
+
+## Let's play
+
+```python
+from outlines import generate
+
+board_state = " "
+turn_number = 0
+while not board.is_game_over():
+    regex_pattern = legal_moves_regex(board)
+    structured = generate.regex(model, regex_pattern)(prompt + board_state)
+    move = board.parse_san(structured)
+
+    if turn_number % 2 == 0 :  # It's White's turn
+        board_state += board.san(move) + " "
+    else:
+        board_state += board.san(move) + " " + str(turn_number) + "."
+
+    turn_number += 1
+
+    board.push(move)
+
+    print(board_state)
+```
+
+Interestingly enough, Phi-2 hates capturing.
+
+```pgn
+ e4 e5 1.Nf3 Ne7 3.b4 Nf5 5.Nc3 Ne7 7.Bb5 a6 9.Na4 b6 11.c3 Nec6 13.c4 a5 15.d4 Qg5 17.Nd2 Bb7 19.dxe5
+```
+
+*This example was originally authored by [@903124S](https://x.com/903124S) in [this gist](https://gist.github.com/903124/cfbefa24da95e2316e0d5e8ef8ed360d).*
diff --git a/docs_new/examples/qa-with-citations.md b/docs_new/examples/qa-with-citations.md
new file mode 100644
index 000000000..79a2214c3
--- /dev/null
+++ b/docs_new/examples/qa-with-citations.md
@@ -0,0 +1,255 @@
+# Generate Synthetic Data and Q&A with Citations
+
+This tutorial is adapted from the [instructor-ollama notebook](https://github.com/alonsosilvaallende/Hermes-Function-Calling/blob/main/examples/instructor_ollama.ipynb). We start with a simple example to generate synthetic data and then we approach the problem of question answering by providing citations.
+
+We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```bash
+pip install llama-cpp-python
+```
+
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+```python
+import llama_cpp
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
+```
+
+??? note "(Optional) Store the model weights in a custom folder"
+
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
+
+## Generate Synthetic Data
+
+We first need to define our Pydantic class for a user:
+
+```python
+from pydantic import BaseModel, Field
+
+class UserDetail(BaseModel):
+    id: int = Field(..., description="Unique identifier") # so the model keeps track of the number of users
+    first_name: str
+    last_name: str
+    age: int
+```
+
+We then define a Pydantic class for a list of users:
+
+```python
+from typing import List
+
+class Users(BaseModel):
+    users: List[UserDetail]
+```
+
+We can use a `generate.json` by passing this Pydantic class we just defined, and call the generator:
+
+```python
+model = models.LlamaCpp(llm)
+generator = generate.json(model, Users)
+response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
+print(response.users)
+# [UserDetail(id=1, first_name='John', last_name='Doe', age=25),
+# UserDetail(id=2, first_name='Jane', last_name='Doe', age=30),
+# UserDetail(id=3, first_name='Bob', last_name='Smith', age=40),
+# UserDetail(id=4, first_name='Alice', last_name='Smith', age=35),
+# UserDetail(id=5, first_name='John', last_name='Smith', age=20)]
+```
+
+```python
+for user in response.users:
+    print(user.first_name)
+    print(user.last_name)
+    print(user.age)
+    print(#####)
+# John
+# Doe
+# 25
+# #####
+# Jane
+# Doe
+# 30
+# #####
+# Bob
+# Smith
+# 40
+# #####
+# Alice
+# Smith
+# 35
+# #####
+# John
+# Smith
+# 20
+# #####
+```
+
+## QA with Citations
+
+We first need to define our Pydantic class for QA with citations:
+
+```python
+from typing import List
+from pydantic import BaseModel
+
+class QuestionAnswer(BaseModel):
+    question: str
+    answer: str
+    citations: List[str]
+
+schema = QuestionAnswer.model_json_schema()
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
+
+```python
+def generate_hermes_prompt(question, context, schema=schema):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON with correct and exact citations "
+        "extracted from the `Context`. "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
+        "<|im_start|>user\n"
+        + "`Context`: "
+        + context
+        + "\n`Question`: "
+        + question + "<|im_end|>"
+        + "\n<|im_start|>assistant\n"
+        "<schema>"
+    )
+```
+
+We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
+
+```python
+question = "What did the author do during college?"
+context = """
+My name is Jason Liu, and I grew up in Toronto Canada but I was born in China.
+I went to an arts high school but in university I studied Computational Mathematics and physics.
+As part of coop I worked at many companies including Stitchfix, Facebook.
+I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.
+"""
+generator = generate.json(model, QuestionAnswer)
+prompt = generate_hermes_prompt(question, context)
+response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+print(response)
+# QuestionAnswer(question='What did the author do during college?', answer='The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
+```
+
+We can do the same for a list of question-context pairs:
+
+```python
+question1 = "Where was John born?"
+context1 = """
+John Doe is a software engineer who was born in New York, USA.
+He studied Computer Science at the Massachusetts Institute of Technology.
+During his studies, he interned at Google and Microsoft.
+He also founded the Artificial Intelligence club at his university and served as its president for three years.
+"""
+
+question2 = "What did Emily study in university?"
+context2 = """
+Emily Smith is a data scientist from London, England.
+She attended the University of Cambridge where she studied Statistics and Machine Learning.
+She interned at IBM and Amazon during her summer breaks.
+Emily was also the head of the Women in Tech society at her university.
+"""
+
+question3 = "Which companies did Robert intern at?"
+context3 = """
+Robert Johnson, originally from Sydney, Australia, is a renowned cybersecurity expert.
+He studied Information Systems at the University of Melbourne.
+Robert interned at several cybersecurity firms including NortonLifeLock and McAfee.
+He was also the leader of the Cybersecurity club at his university.
+"""
+
+question4 = "What club did Alice start at her university?"
+context4 = """
+Alice Williams, a native of Dublin, Ireland, is a successful web developer.
+She studied Software Engineering at Trinity College Dublin.
+Alice interned at several tech companies including Shopify and Squarespace.
+She started the Web Development club at her university and was its president for two years.
+"""
+
+question5 = "What did Michael study in high school?"
+context5 = """
+Michael Brown is a game developer from Tokyo, Japan.
+He attended a specialized high school where he studied Game Design.
+He later attended the University of Tokyo where he studied Computer Science.
+Michael interned at Sony and Nintendo during his university years.
+He also started the Game Developers club at his university.
+"""
+
+for question, context in [
+    (question1, context1),
+    (question2, context2),
+    (question3, context3),
+    (question4, context4),
+    (question5, context5),
+]:
+    final_prompt = my_final_prompt(question, context)
+    generator = generate.json(model, QuestionAnswer)
+    response = generator(final_prompt, max_tokens=1024, temperature=0, seed=42)
+    display(question)
+    display(response.answer)
+    display(response.citations)
+    print("\n\n")
+
+# 'Where was John born?'
+# 'John Doe was born in New York, USA.'
+# ['John Doe is a software engineer who was born in New York, USA.']
+#
+#
+# 'What did Emily study in university?'
+# 'Emily studied Statistics and Machine Learning in university.'
+# ['She attended the University of Cambridge where she studied Statistics and Machine Learning.']
+#
+#
+# 'Which companies did Robert intern at?'
+# 'Robert interned at NortonLifeLock and McAfee.'
+# ['Robert Johnson, originally from Sydney, Australia, is a renowned cybersecurity expert. He interned at several cybersecurity firms including NortonLifeLock and McAfee.']
+#
+#
+# 'What club did Alice start at her university?'
+# 'Alice started the Web Development club at her university.'
+# ['Alice Williams, a native of Dublin, Ireland, is a successful web developer. She started the Web Development club at her university and was its president for two years.']
+#
+#
+# 'What did Michael study in high school?'
+# 'Michael studied Game Design in high school.'
+# ['Michael Brown is a game developer from Tokyo, Japan. He attended a specialized high school where he studied Game Design.']
+```
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_new/examples/react_agent.md b/docs_new/examples/react_agent.md
new file mode 100644
index 000000000..15b6ed369
--- /dev/null
+++ b/docs_new/examples/react_agent.md
@@ -0,0 +1,273 @@
+# ReAct Agent
+
+This example shows how to use [outlines](https://dottxt-ai.github.io/outlines/) to build your own agent with open weights local models and structured outputs. It is inspired by the blog post [A simple Python implementation of the ReAct pattern for LLMs](https://til.simonwillison.net/llms/python-react-pattern) by [Simon Willison](https://simonwillison.net/).
+
+The ReAct pattern (for Reason+Act) is described in the paper [ReAct: Synergizing Reasoning and Acting in Language Models](https://arxiv.org/abs/2210.03629). It's a pattern where you implement additional actions that an LLM can take - searching Wikipedia or running calculations for example - and then teach it how to request the execution of those actions, and then feed their results back into the LLM.
+
+Additionally, we give the LLM the possibility of using a scratchpad described in the paper [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://arxiv.org/abs/2112.00114) which improves the ability of LLMs to perform multi-step computations.
+
+We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
+
+```bash
+pip install llama-cpp-python
+```
+
+We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+```python
+import llama_cpp
+from outlines import generate, models
+
+model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+            ),
+            n_gpu_layers=-1,
+            flash_attn=True,
+            n_ctx=8192,
+            verbose=False)
+```
+
+??? note "(Optional) Store the model weights in a custom folder"
+
+    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
+
+    ```bash
+    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
+    ```
+
+    We initialize the model:
+
+    ```python
+    import llama_cpp
+    from llama_cpp import Llama
+    from outlines import generate, models
+
+    llm = Llama(
+        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
+        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+            "NousResearch/Hermes-2-Pro-Llama-3-8B"
+        ),
+        n_gpu_layers=-1,
+        flash_attn=True,
+        n_ctx=8192,
+        verbose=False
+    )
+    ```
+
+## Build a ReAct agent
+
+In this example, we use two tools:
+
+- wikipedia: \<search term\> - search Wikipedia and returns the snippet of the first result
+- calculate: \<expression\> - evaluate an expression using Python's eval() function
+
+```python
+import httpx
+
+def wikipedia(q):
+    return httpx.get("https://en.wikipedia.org/w/api.php", params={
+        "action": "query",
+        "list": "search",
+        "srsearch": q,
+        "format": "json"
+    }).json()["query"]["search"][0]["snippet"]
+
+
+def calculate(numexp):
+    return eval(numexp)
+```
+
+We define the logic of the agent through a Pydantic class. First, we want the LLM to decide only between the two previously defined tools:
+
+```python
+from enum import Enum
+
+class Action(str, Enum):
+    wikipedia = "wikipedia"
+    calculate = "calculate"
+```
+
+Our agent will loop through Thought and Action. We explicitly give the Action Input field so it doesn't forget to add the arguments of the Action. We also add a scratchpad (optional).
+
+```python
+from pydantic import BaseModel, Field
+
+class Reason_and_Act(BaseModel):
+    Scratchpad: str = Field(..., description="Information from the Observation useful to answer the question")
+    Thought: str = Field(..., description="It describes your thoughts about the question you have been asked")
+    Action: Action
+    Action_Input: str = Field(..., description="The arguments of the Action.")
+```
+
+Our agent will reach a Final Answer. We also add a scratchpad (optional).
+
+```python
+class Final_Answer(BaseModel):
+    Scratchpad: str = Field(..., description="Information from the Observation useful to answer the question")
+    Final_Answer: str = Field(..., description="Answer to the question grounded on the Observation")
+```
+
+Our agent will decide when it has reached a Final Answer and therefore to stop the loop of Thought and Action.
+
+```python
+from typing import Union
+
+class Decision(BaseModel):
+    Decision: Union[Reason_and_Act, Final_Answer]
+```
+
+We could generate a response using the json schema but we will use the regex and check that everything is working as expected:
+
+```python
+from outlines.fsm.json_schema import convert_json_schema_to_str
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+json_schema = Decision.model_json_schema()
+schema_str = convert_json_schema_to_str(json_schema=json_schema)
+regex_str = build_regex_from_schema(schema_str)
+print(regex_str)
+# '\\{[ ]?"Decision"[ ]?:[ ]?(\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Thought"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Action"[ ]?:[ ]?("wikipedia"|"calculate")[ ]?,[ ]?"Action_Input"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\}|\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Final_Answer"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\})[ ]?\\}'
+```
+
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic:
+
+```python
+import datetime
+
+def generate_hermes_prompt(question, schema=""):
+    return (
+        "<|im_start|>system\n"
+        "You are a world class AI model who answers questions in JSON with correct Pydantic schema. "
+        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema>\n"
+        "Today is " + datetime.datetime.today().strftime('%Y-%m-%d') + ".\n" +
+        "You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation. "
+        "At the end of the loop you output a Final Answer. "
+        "Use Scratchpad to store the information from the Observation useful to answer the question "
+        "Use Thought to describe your thoughts about the question you have been asked "
+        "and reflect carefully about the Observation if it exists. "
+        "Use Action to run one of the actions available to you. "
+        "Use Action Input to input the arguments of the selected action - then return PAUSE. "
+        "Observation will be the result of running those actions. "
+        "Your available actions are:\n"
+        "calculate:\n"
+        "e.g. calulate: 4**2 / 3\n"
+        "Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary\n"
+        "wikipedia:\n"
+        "e.g. wikipedia: Django\n"
+        "Returns a summary from searching Wikipedia\n"
+        "DO NOT TRY TO GUESS THE ANSWER. Begin! <|im_end|>"
+        "\n<|im_start|>user\n" + question + "<|im_end|>"
+        "\n<|im_start|>assistant\n"
+    )
+```
+
+We define a ChatBot class
+
+```python
+class ChatBot:
+    def __init__(self, prompt=""):
+        self.prompt = prompt
+
+    def __call__(self, user_prompt):
+        self.prompt += user_prompt
+        result = self.execute()
+        return result
+
+    def execute(self):
+        generator = generate.regex(model, regex_str)
+        result = generator(self.prompt, max_tokens=1024, temperature=0, seed=42)
+        return result
+```
+
+We define a query function:
+
+```python
+import json
+
+def query(question, max_turns=5):
+    i = 0
+    next_prompt = (
+        "\n<|im_start|>user\n" + question + "<|im_end|>"
+        "\n<|im_start|>assistant\n"
+    )
+    previous_actions = []
+    while i < max_turns:
+        i += 1
+        prompt = generate_hermes_prompt(question=question, schema=Decision.model_json_schema())
+        bot = ChatBot(prompt=prompt)
+        result = bot(next_prompt)
+        json_result = json.loads(result)['Decision']
+        if "Final_Answer" not in list(json_result.keys()):
+            scratchpad = json_result['Scratchpad'] if i == 0 else ""
+            thought = json_result['Thought']
+            action = json_result['Action']
+            action_input = json_result['Action_Input']
+            print(f"\x1b[34m Scratchpad: {scratchpad} \x1b[0m")
+            print(f"\x1b[34m Thought: {thought} \x1b[0m")
+            print(f"\x1b[36m  -- running {action}: {str(action_input)}\x1b[0m")
+            if action + ": " + str(action_input) in previous_actions:
+                observation = "You already run that action. **TRY A DIFFERENT ACTION INPUT.**"
+            else:
+                if action=="calculate":
+                    try:
+                        observation = eval(str(action_input))
+                    except Exception as e:
+                        observation = f"{e}"
+                elif action=="wikipedia":
+                    try:
+                        observation = wikipedia(str(action_input))
+                    except Exception as e:
+                        observation = f"{e}"
+            print()
+            print(f"\x1b[33m Observation: {observation} \x1b[0m")
+            print()
+            previous_actions.append(action + ": " + str(action_input))
+            next_prompt += (
+                "\nScratchpad: " + scratchpad +
+                "\nThought: " + thought +
+                "\nAction: " + action  +
+                "\nAction Input: " + action_input +
+                "\nObservation: " + str(observation)
+            )
+        else:
+            scratchpad = json_result["Scratchpad"]
+            final_answer = json_result["Final_Answer"]
+            print(f"\x1b[34m Scratchpad: {scratchpad} \x1b[0m")
+            print(f"\x1b[34m Final Answer: {final_answer} \x1b[0m")
+            return final_answer
+    print(f"\nFinal Answer: I am sorry, but I am unable to answer your question. Please provide more information or a different question.")
+    return "No answer found"
+```
+
+We can now test our ReAct agent:
+
+```python
+print(query("What's 2 to the power of 10?"))
+# Scratchpad:
+# Thought: I need to perform a mathematical calculation to find the result of 2 to the power of 10.
+#  -- running calculate: 2**10
+#
+# Observation: 1024
+#
+# Scratchpad: 2 to the power of 10 is 1024.
+# Final Answer: 2 to the power of 10 is 1024.
+# 2 to the power of 10 is 1024.
+```
+
+```python
+print(query("What does England share borders with?"))
+# Scratchpad:
+# Thought: To answer this question, I will use the 'wikipedia' action to gather information about England's geographical location and its borders.
+#  -- running wikipedia: England borders
+#
+# Observation: Anglo-Scottish <span class="searchmatch">border</span> (Scottish Gaelic: Crìochan Anglo-Albannach) is an internal <span class="searchmatch">border</span> of the United Kingdom separating Scotland and <span class="searchmatch">England</span> which runs for
+#
+# Scratchpad: Anglo-Scottish border (Scottish Gaelic: Crìochan Anglo-Albannach) is an internal border of the United Kingdom separating Scotland and England which runs for
+# Final Answer: England shares a border with Scotland.
+# England shares a border with Scotland.
+```
+
+As mentioned in Simon's blog post, this is not a very robust implementation at all and there's a ton of room for improvement. But it is lovely how simple it is with a few lines of Python to make these extra capabilities available to the LLM. And now you can run it locally with an open weights LLM.
+
+This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_new/examples/read-pdfs.md b/docs_new/examples/read-pdfs.md
new file mode 100644
index 000000000..dbe4ccb02
--- /dev/null
+++ b/docs_new/examples/read-pdfs.md
@@ -0,0 +1,376 @@
+# PDF to structured output with vision language models
+
+A common task with language models is to ask language models questions about a PDF file.
+
+Typically, the output is unstructured text, i.e. "talking" to your PDF.
+
+In some cases, you may wish to extract structured information from the PDF, like tables, lists, citations, etc.
+
+PDFs are difficult to machine read. However, you can simply convert the PDF to images, and then use a vision language model to extract structured information from the images.
+
+This cookbook demonstrates how to
+
+1. Convert a PDF to a list of images
+2. Use a vision language model to extract structured information from the images
+
+## Dependencies
+
+You'll need to install these dependencies:
+
+```bash
+pip install outlines pillow transformers torch==2.4.0 pdf2image
+
+# Optional, but makes the output look nicer
+pip install rich
+```
+
+## Import the necessary libraries
+
+```python
+from PIL import Image
+import outlines
+import torch
+from transformers import AutoProcessor
+from pydantic import BaseModel
+from typing import List, Optional
+from pdf2image import convert_from_path
+import os
+from rich import print
+import requests
+```
+
+## Choose a model
+
+We've tested this example with [Pixtral 12b](https://huggingface.co/mistral-community/pixtral-12b) and [Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct).
+
+To use Pixtral:
+
+```python
+from transformers import LlavaForConditionalGeneration
+model_name="mistral-community/pixtral-12b"
+model_class=LlavaForConditionalGeneration
+```
+
+To use Qwen-2-VL:
+
+```python
+from transformers import Qwen2VLForConditionalGeneration
+model_name = "Qwen/Qwen2-VL-7B-Instruct"
+model_class = Qwen2VLForConditionalGeneration
+```
+
+You can load your model into memory with:
+
+```python
+# This loads the model into memory. On your first run,
+# it will have to download the model, which might take a while.
+model = outlines.models.transformers_vision(
+    model_name,
+    model_class=model_class,
+    model_kwargs={
+        "device_map": "auto",
+        "torch_dtype": torch.bfloat16,
+    },
+    processor_kwargs={
+        "device": "auto",
+    },
+)
+```
+
+## Convert the PDF to images
+
+We'll use the `pdf2image` library to convert each page of the PDF to an image.
+
+`convert_pdf_to_images` is a convenience function that converts each page of the PDF to an image, and optionally saves the images to disk when `output_dir` is provided.
+
+Note: the `dpi` argument is important. It controls the resolution of the images. High DPI images are higher quality and may yield better results,
+but they are also larger, slower to process, and require more memory.
+
+```python
+from pdf2image import convert_from_path
+from PIL import Image
+import os
+from typing import List, Optional
+
+def convert_pdf_to_images(
+    pdf_path: str,
+    output_dir: Optional[str] = None,
+    dpi: int = 120,
+    fmt: str = 'PNG'
+) -> List[Image.Image]:
+    """
+    Convert a PDF file to a list of PIL Image objects.
+
+    Args:
+        pdf_path: Path to the PDF file
+        output_dir: Optional directory to save the images
+        dpi: Resolution for the conversion. High DPI is high quality, but also slow and memory intensive.
+        fmt: Output format (PNG recommended for quality)
+
+    Returns:
+        List of PIL Image objects
+    """
+    # Convert PDF to list of images
+    images = convert_from_path(
+        pdf_path,
+        dpi=dpi,
+        fmt=fmt
+    )
+
+    # Optionally save images
+    if output_dir:
+        os.makedirs(output_dir, exist_ok=True)
+        for i, image in enumerate(images):
+            image.save(os.path.join(output_dir, f'page_{i+1}.{fmt.lower()}'))
+
+    return images
+```
+
+We're going to use the [Louf & Willard paper](https://arxiv.org/pdf/2307.09702) that described the method that Outlines uses for structured generation.
+
+To download the PDF, run:
+
+```python
+# Download the PDF file
+pdf_url = "https://arxiv.org/pdf/2307.09702"
+response = requests.get(pdf_url)
+
+# Save the PDF locally
+with open("louf-willard.pdf", "wb") as f:
+    f.write(response.content)
+```
+
+Now, we can convert the PDF to a list of images:
+
+```python
+# Load the pdf
+images = convert_pdf_to_images(
+    "louf-willard.pdf",
+    dpi=120,
+    output_dir="output_images"
+)
+```
+
+## Extract structured information from the images
+
+The structured output you can extract is exactly the same as everywhere else in Outlines -- you can use regular expressions, JSON schemas, selecting from a list of options, etc.
+
+### Extracting data into JSON
+
+Suppose you wished to go through each page of the PDF, and extract the page description, key takeaways, and page number.
+
+You can do this by defining a JSON schema, and then using `outlines.generate.json` to extract the data.
+
+First, define the structure you want to extract:
+
+```python
+class PageSummary(BaseModel):
+    description: str
+    key_takeaways: List[str]
+    page_number: int
+```
+
+Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers `AutoProcessor` to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
+
+Images are denoted with `type: "image"`, and text is denoted with `type: "text"`.
+
+```python
+messages = [
+    {
+        "role": "user",
+        "content": [
+            # The text you're passing to the model --
+            # this is where you do your standard prompting.
+            {"type": "text", "text": f"""
+                Describe the page in a way that is easy for a PhD student to understand.
+
+                Return the information in the following JSON schema:
+                {PageSummary.model_json_schema()}
+
+                Here is the page:
+                """
+            },
+
+            # Don't need to pass in an image, since we do this
+            # when we call the generator function down below.
+            {"type": "image", "image": ""},
+        ],
+    }
+]
+
+# Convert the messages to the final prompt
+processor = AutoProcessor.from_pretrained(model_name)
+instruction = processor.apply_chat_template(
+    messages, tokenize=False, add_generation_prompt=True
+)
+```
+
+Now we iterate through each image, and extract the structured information:
+
+```python
+# Page summarizer function
+page_summary_generator = outlines.generate.json(model, PageSummary)
+
+for image in images:
+    result = page_summary_generator(instruction, [image])
+    print(result)
+```
+
+### Regular expressions to extract the arxiv paper identifier
+
+The [arXiv paper identifier](https://info.arxiv.org/help/arxiv_identifier.html) is a unique identifier for each paper. These identifiers have the format `arXiv:YYMM.NNNNN` (five end digits) or `arXiv:YYMM.NNNN` (four end digits). arXiv identifiers are typically watermarked on papers uploaded to arXiv.
+
+arXiv identifiers are optionally followed by a version number, i.e. `arXiv:YYMM.NNNNNvX`.
+
+We can use a regular expression to define this patter:
+
+```python
+paper_regex = r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?'
+```
+
+We can build an extractor function from the regex:
+
+```python
+id_extractor = outlines.generate.regex(model, paper_regex)
+```
+
+Now, we can extract the arxiv paper identifier from the first image:
+
+```python
+arxiv_instruction = processor.apply_chat_template(
+    [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": f"""
+                Extract the arxiv paper identifier from the page.
+
+                Here is the page:
+                """},
+                {"type": "image", "image": ""},
+            ],
+        }
+    ],
+    tokenize=False,
+    add_generation_prompt=True
+)
+
+# Extract the arxiv paper identifier
+paper_id = id_extractor(arxiv_instruction, [images[0]])
+```
+
+As of the time of this writing, the arxiv paper identifier is
+
+```
+arXiv:2307.09702v4
+```
+
+Your version number may be different, but the part before `vX` should match.
+
+### Categorize the paper into one of several categories
+
+`outlines.generate.choice` allows the model to select one of several options. Suppose we wanted to categorize the paper into being about "language models", "economics", "structured generation", or "other".
+
+Let's define a few categories we might be interested in:
+
+```python
+categories = [
+    "llms",
+    "cell biology",
+    "other"
+]
+```
+
+Now we can construct the prompt:
+
+```python
+categorization_instruction = processor.apply_chat_template(
+    [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": f"""
+                Please choose one of the following categories
+                that best describes the paper.
+
+                {categories}
+
+                Here is the paper:
+                """},
+
+                {"type": "image", "image": ""},
+            ],
+        }
+    ],
+    tokenize=False,
+    add_generation_prompt=True
+)
+```
+
+Now we can show the model the first page and extract the category:
+
+```python
+# Build the choice extractor
+categorizer = outlines.generate.choice(
+    model,
+    categories
+)
+
+# Categorize the paper
+category = categorizer(categorization_instruction, [images[0]])
+print(category)
+```
+
+Which should return:
+
+```
+llms
+```
+
+## Additional notes
+
+You can provide multiple images to the model by
+
+1. Adding additional image messages
+2. Providing a list of images to the `generate` function
+
+For example, to have two images, you can do:
+
+```python
+two_image_prompt = processor.apply_chat_template(
+    [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "are both of these images of hot dogs?"},
+
+                # Tell the model there are two images
+                {"type": "image", "image": ""},
+                {"type": "image", "image": ""},
+            ],
+        }
+    ],
+    tokenize=False,
+    add_generation_prompt=True
+)
+
+# Pass two images to the model
+generator = outlines.generate.choice(
+    model,
+    ["hot dog", "not hot dog"]
+)
+
+result = generator(
+    two_image_prompt,
+
+    # Pass two images to the model
+    [images[0], images[1]]
+)
+print(result)
+```
+
+Using the first to pages of the paper (they are not images of hot dogs), we should get
+
+```
+not hot dog
+```
diff --git a/docs_new/examples/receipt-digitization.md b/docs_new/examples/receipt-digitization.md
new file mode 100644
index 000000000..0fa256db7
--- /dev/null
+++ b/docs_new/examples/receipt-digitization.md
@@ -0,0 +1,296 @@
+# Receipt Data Extraction with VLMs
+
+## Setup
+
+You'll need to install the dependencies:
+
+```bash
+pip install outlines torch==2.4.0 transformers accelerate pillow rich
+```
+
+## Import libraries
+
+Load all the necessary libraries:
+
+```python
+# LLM stuff
+import outlines
+import torch
+from transformers import AutoProcessor
+from pydantic import BaseModel, Field
+from typing import Literal, Optional, List
+
+# Image stuff
+from PIL import Image
+import requests
+
+# Rich for pretty printing
+from rich import print
+```
+
+## Choose a model
+
+This example has been tested with `mistral-community/pixtral-12b` ([HF link](https://huggingface.co/mistral-community/pixtral-12b)) and `Qwen/Qwen2-VL-7B-Instruct` ([HF link](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct)).
+
+We recommend Qwen-2-VL as we have found it to be more accurate than Pixtral.
+
+If you want to use Qwen-2-VL, you can do the following:
+
+```python
+# To use Qwen-2-VL:
+from transformers import Qwen2VLForConditionalGeneration
+model_name = "Qwen/Qwen2-VL-7B-Instruct"
+model_class = Qwen2VLForConditionalGeneration
+```
+
+If you want to use Pixtral, you can do the following:
+
+```python
+# To use Pixtral:
+from transformers import LlavaForConditionalGeneration
+model_name="mistral-community/pixtral-12b"
+model_class=LlavaForConditionalGeneration
+```
+
+## Load the model
+
+Load the model into memory:
+
+```python
+model = outlines.models.transformers_vision(
+    model_name,
+    model_class=model_class,
+    model_kwargs={
+        "device_map": "auto",
+        "torch_dtype": torch.bfloat16,
+    },
+    processor_kwargs={
+        "device": "cuda", # set to "cpu" if you don't have a GPU
+    },
+)
+```
+
+## Image processing
+
+Images can be quite large. In GPU-poor environments, you may need to resize the image to a smaller size.
+
+Here's a helper function to do that:
+
+```python
+def load_and_resize_image(image_path, max_size=1024):
+    """
+    Load and resize an image while maintaining aspect ratio
+
+    Args:
+        image_path: Path to the image file
+        max_size: Maximum dimension (width or height) of the output image
+
+    Returns:
+        PIL Image: Resized image
+    """
+    image = Image.open(image_path)
+
+    # Get current dimensions
+    width, height = image.size
+
+    # Calculate scaling factor
+    scale = min(max_size / width, max_size / height)
+
+    # Only resize if image is larger than max_size
+    if scale < 1:
+        new_width = int(width * scale)
+        new_height = int(height * scale)
+        image = image.resize((new_width, new_height), Image.Resampling.LANCZOS)
+
+    return image
+```
+
+You can change the resolution of the image by changing the `max_size` argument. Small max sizes will make the image more blurry, but processing will be faster and require less memory.
+
+## Load an image
+
+Load an image and resize it. We've provided a sample image of a Trader Joe's receipt, but you can use any image you'd like.
+
+Here's what the image looks like:
+
+![Trader Joe's receipt](./images/trader-joes-receipt.jpg)
+
+```python
+# Path to the image
+image_path = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/cookbook/images/trader-joes-receipt.jpg"
+
+# Download the image
+response = requests.get(image_path)
+with open("receipt.png", "wb") as f:
+    f.write(response.content)
+
+# Load + resize the image
+image = load_and_resize_image("receipt.png")
+```
+
+## Define the output structure
+
+We'll define a Pydantic model to describe the data we want to extract from the image.
+
+In our case, we want to extract the following information:
+
+- The store name
+- The store address
+- The store number
+- A list of items, including the name, quantity, price per unit, and total price
+- The tax
+- The total
+- The date
+- The payment method
+
+Most fields are optional, as not all receipts contain all information.
+
+```python
+class Item(BaseModel):
+    name: str
+    quantity: Optional[int]
+    price_per_unit: Optional[float]
+    total_price: Optional[float]
+
+class ReceiptSummary(BaseModel):
+    store_name: str
+    store_address: str
+    store_number: Optional[int]
+    items: List[Item]
+    tax: Optional[float]
+    total: Optional[float]
+    # Date is in the format YYYY-MM-DD. We can apply a regex pattern to ensure it's formatted correctly.
+    date: Optional[str] = Field(pattern=r'\d{4}-\d{2}-\d{2}', description="Date in the format YYYY-MM-DD")
+    payment_method: Literal["cash", "credit", "debit", "check", "other"]
+```
+
+## Prepare the prompt
+
+We'll use the `AutoProcessor` to convert the image and the text prompt into a format that the model can understand. Practically,
+this is the code that adds user, system, assistant, and image tokens to the prompt.
+
+```python
+# Set up the content you want to send to the model
+messages = [
+    {
+        "role": "user",
+        "content": [
+            {
+                # The image is provided as a PIL Image object
+                "type": "image",
+                "image": image,
+            },
+            {
+                "type": "text",
+                "text": f"""You are an expert at extracting information from receipts.
+                Please extract the information from the receipt. Be as detailed as possible --
+                missing or misreporting information is a crime.
+
+                Return the information in the following JSON schema:
+                {ReceiptSummary.model_json_schema()}
+            """},
+        ],
+    }
+]
+
+# Convert the messages to the final prompt
+processor = AutoProcessor.from_pretrained(model_name)
+prompt = processor.apply_chat_template(
+    messages, tokenize=False, add_generation_prompt=True
+)
+```
+
+If you are curious, the final prompt that is sent to the model looks (roughly) like this:
+
+```
+<|im_start|>system
+You are a helpful assistant.<|im_end|>
+<|im_start|>user
+<|vision_start|><|image_pad|><|vision_end|>
+You are an expert at extracting information from receipts.
+Please extract the information from the receipt. Be as detailed as
+possible -- missing or misreporting information is a crime.
+
+Return the information in the following JSON schema:
+
+<JSON SCHEMA OMITTED>
+<|im_end|>
+<|im_start|>assistant
+```
+
+## Run the model
+
+```python
+# Prepare a function to process receipts
+receipt_summary_generator = outlines.generate.json(
+    model,
+    ReceiptSummary,
+
+    # Greedy sampling is a good idea for numeric
+    # data extraction -- no randomness.
+    sampler=outlines.samplers.greedy()
+)
+
+# Generate the receipt summary
+result = receipt_summary_generator(prompt, [image])
+print(result)
+```
+
+## Output
+
+The output should look like this:
+
+```
+ReceiptSummary(
+    store_name="Trader Joe's",
+    store_address='401 Bay Street, San Francisco, CA 94133',
+    store_number=0,
+    items=[
+        Item(name='BANANA EACH', quantity=7, price_per_unit=0.23, total_price=1.61),
+        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='BAREBELLS CARAMEL CASHEW', quantity=2, price_per_unit=2.29, total_price=4.58),
+        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
+        Item(name='SPINDRIFT ORANGE MANGO 8', quantity=1, price_per_unit=7.49, total_price=7.49),
+        Item(name='Bottle Deposit', quantity=8, price_per_unit=0.05, total_price=0.4),
+        Item(name='MILK ORGANIC GALLON WHOL', quantity=1, price_per_unit=6.79, total_price=6.79),
+        Item(name='CLASSIC GREEK SALAD', quantity=1, price_per_unit=3.49, total_price=3.49),
+        Item(name='COBB SALAD', quantity=1, price_per_unit=5.99, total_price=5.99),
+        Item(name='PEPPER BELL RED XL EACH', quantity=1, price_per_unit=1.29, total_price=1.29),
+        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25),
+        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25)
+    ],
+    tax=0.68,
+    total=41.98,
+    date='2023-11-04',
+    payment_method='debit',
+
+)
+```
+
+Voila! You've successfully extracted information from a receipt using an LLM.
+
+## Bonus: roasting the user for their receipt
+
+You can roast the user for their receipt by adding a `roast` field to the end of the  `ReceiptSummary` model.
+
+```python
+class ReceiptSummary(BaseModel):
+    ...
+    roast: str
+```
+
+which gives you a result like
+
+```
+ReceiptSummary(
+    ...
+    roast="You must be a fan of Trader Joe's because you bought enough
+    items to fill a small grocery bag and still had to pay for a bag fee.
+    Maybe you should start using reusable bags to save some money and the
+    environment."
+)
+```
+
+Qwen is not particularly funny, but worth a shot.
diff --git a/docs_new/examples/simtom.md b/docs_new/examples/simtom.md
new file mode 100644
index 000000000..a80ebc24f
--- /dev/null
+++ b/docs_new/examples/simtom.md
@@ -0,0 +1,127 @@
+# Build perspective-taking agents with SimToM
+
+Prompting strategies like Chain-of-Thought (CoT) can improve LLMs' reasoning capabilities. However, they underwhelm in tasks that require keeping track of inconsistent world states. [SimToM](https://arxiv.org/abs/2311.10227) proposes a simple, two-stage prompting framework for LLMs inspired by Simulation Theory. The authors showed that this approach outperforms zero-shot prompting and CoT on ToMI and BigToM, two benchmarks with Theory of Mind questions.
+
+In this example, we will implement SimToM with a few lines of code using Outlines' prompt templating and structured generation capabilities.
+
+## How SimToM works
+
+SimToM calls an LLM with two consecutive prompts:
+
+1. **Perspective-taking**: The first prompt receives a `story` and a `character`. The goal is to understand the situation based on the character's point of view and filter out the rest of the story.
+2. **Question-Answering**: The second prompt receives the character's point of view from the previous step and tasks the LLM to answer a question using that context.
+
+![Figure 2 in the paper](./images/simtom.png)
+
+## Outlines implementation
+
+To implement SimToM with Outlines, we will need to:
+
+1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
+2. Define the JSON object each prompt will return using Pydantic.
+3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/latest/reference/models/transformers/).
+
+Let's dive into it!
+
+### Using Prompt Templates
+
+The authors have shared their code, prompts and data in [this GitHub repository](https://github.com/shawnsihyunlee/simulatedtom). Below, we define in Outlines the prompts they used for the ToMI dataset:
+
+```python
+from outlines import Template
+
+
+perspective_taking = Template.from_string(
+    """<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
+    Your job is to output only the events that the specified character, {{character}}, knows about.
+
+    Here are a few rules:
+    1. A character knows about all events that they do.
+    2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
+    3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
+
+    Story: {{story}}
+    What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]""" # noqa
+)
+
+simulation = Template.from_string(
+    """<s>[INST] {% for event in events %}
+    {{event}}
+    {% endfor %}
+    You are {{name}}.
+    Based on the above information, answer the following question:
+    {{question}}
+    You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]""" # noqa
+)
+```
+
+### JSON Structured Generation
+
+Outlines guarantees that the LLM will return a valid JSON object, which we can specify as a Pydantic model.
+
+We will need two Pydantic models for SimToM, one for each prompt:
+
+```python
+from pydantic import BaseModel, Field
+from typing import List
+
+
+class PerspectiveTaking(BaseModel):
+    """This is for the first prompt."""
+    character: str = Field(description="The character we extract the events for.")
+    events: List[str] = Field(description="All events that the character knows about.")
+
+
+class Simulation(BaseModel):
+    """This is for the second prompt."""
+    answer: str
+```
+
+### Calling an LLM
+
+Let's try SimToM with an example from the ToMI dataset:
+
+```python
+story = """
+1 Aria entered the front_yard.
+2 Aiden entered the front_yard.
+3 The grapefruit is in the green_bucket.
+4 Aria moved the grapefruit to the blue_container.
+5 Aiden exited the front_yard.
+6 Noah entered the playroom.
+"""
+question = "7 Where was the grapefruit at the beginning?"
+character = "Aria"
+```
+
+We load `Mistral-7B-Instruct-v0.3`, create the prompt using the template we defined earlier, and generate a structured response. As a reminder, the goal of the first call is to get all the events a character, `Aria`, knows about.
+
+```python
+# Load an LLM from Hugging Face
+MODEL_NAME = "mistral-community/Mistral-7B-Instruct-v0.3"
+model = outlines.models.transformers(MODEL_NAME, device="cuda")
+
+perspective_prompt = perspective_taking(story=story, character=character)
+
+# Call Mistral 7B with the first prompt
+generator = outlines.generate.json(model, PerspectiveTaking)
+perspective = generator(perspective_prompt)
+
+print(perspective.model_dump())
+# {'character': 'Aria', 'events': ['1 Aria entered the front_yard.', '3 The grapefruit is in the green_bucket.', '4 Aria moved the grapefruit to the blue_container.']}
+```
+
+Not bad! We will now generate the second prompt with those events.
+
+```python
+sim_prompt = simulation(events=perspective.events, name=character, question=question)
+
+# Call Mistral 7B with the second prompt
+generator = outlines.generate.json(model, Simulation)
+result = generator(sim_prompt)
+
+print(result.model_dump())
+# {'answer': 'green_bucket'}
+```
+
+And this is it! SimToM could be useful in agentic workflows, where agents must act based on what they know, not all available information. One caveat of SimToM is that the perspective-taking step may remove important information, leading to wrong results. As the authors note in their paper, it can feature as a simple and effective baseline for evaluating LLMs on Theory of Mind reasoning tasks.
diff --git a/docs_new/examples/structured_generation_workflow.md b/docs_new/examples/structured_generation_workflow.md
new file mode 100644
index 000000000..b9d7c28a4
--- /dev/null
+++ b/docs_new/examples/structured_generation_workflow.md
@@ -0,0 +1,215 @@
+# Structured Generation Workflow: Generating Synthetic Phone Numbers
+
+This is a condensed version of [Coding for Structured Generation with LLMs](https://blog.dottxt.co/coding-for-structured-generation.html).
+
+For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this
+with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
+
+## Unstructured approach
+
+Before diving into how to use structure generation for this task let's start with an unstructured example. We begin by loading our model:
+
+```python
+import outlines
+
+model_name = 'microsoft/Phi-3-mini-4k-instruct'
+model = outlines.models.transformers(model_name)
+```
+
+Next we need a prompt for this model. Since we're focusing on structured generation, we won't be engaging in any form of "prompt hacking" and will be leaving this prompt untouched for the rest of this example.
+
+```python
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+messages_phone = [
+            {"role": "user", "content": """
+            Please generate a realistic phone number for Washington State in the following format
+
+            (555) 555-5555
+
+            """}
+]
+
+# This allows us to properly format our prompt for
+# Phi-3 Mini's 'Instruct' interface.
+prompt_phone = tokenizer.apply_chat_template(messages_phone, tokenize=False)
+```
+
+With our prompt ready we can now generate 10 example phone numbers
+
+```python
+phone_generator_unstruct = outlines.generate.text(model)
+for _ in range(10):
+    print(phone_generator_unstruct(prompt_phone,max_tokens=12))
+```
+
+> I'd be happy to help you generate a realistic phone\
+I cannot generate a real phone number as I'm just\
+I'm an AI and don't have the ability\
+Sure! Here is a randomly generated phone number in the format\
+Here's a phone number that fits the format for a\
+In Washington State, phone numbers typically have a three-dig\
+Here are a few examples of phone numbers that could be considered\
+I'd be happy to help generate a realistic phone number\
+I'd be happy to help you generate a random phone\
+Based on the format you provided, a realistic phone number for\
+
+As we can see, none of these outputs are even phone numbers!
+
+Let's see  if we can improve this using structured generation.
+
+## The Structured Generation Workflow
+
+In order to solve this problem we're going to introduce a *Structured Generation Workflow* outlined in this image:
+
+!["Visual of Structured Generation Workflow"](./images/coding_structure_diagram.png)
+
+Let's step through this:
+
+### Real example
+
+We start with a real example phone number, in this case for the Seattle Public Library, that we can use to verify the structure we are creating.
+
+```python
+phone_number = "(206) 386-4636"
+```
+
+For a simple example like this, we'll just be using a single phone number, for more complex examples it can be helpful to have more examples.
+
+### Draft Structure
+
+The next step in the process is for use to define a simple regex that we feel correctly models our real data.
+
+```python
+phone_regex_1 = r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}'
+```
+
+Next we need to validate this regex against our real data.
+
+### Validate by matching examples
+
+Whenever writing non-trivial code with structured generation it is *essential* that you first validate the code against your real data example(s).
+
+We'll start with a simple method of validation: just checking that our regex matches the data.
+
+```
+import re
+re.match(phone_regex_1, phone_number)
+
+# <re.Match object; span=(0, 14), match='(206) 386-4636'>
+
+```
+
+Now that we have a match, we can move on to generating structured output!
+
+### Generate Structure
+
+We're ready to see if structured generation can make an improvement over our initial unstructured approach:
+
+```python
+phone_generator_v1 = outlines.generate.regex(model, phone_regex_1)
+for _ in range(10):
+    print(phone_generator_v1(prompt_phone))
+```
+> (206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234\
+(206) 123-4567\
+(206) 555-1234\
+(206) 555-1234\
+(206) 555-1234
+
+At least we have phone numbers! But I think we can do better!
+
+### Inspect output
+
+In this case the model *did* create phone numbers and, impressively, got the area code correct. So using structured generation did improve things. However these numbers are pretty boring. Let's improve that structure!
+
+## Iteration
+
+We've walked through the loop once, so we can go quickly now through each iteration.
+
+We start by improving our structure:
+
+```python
+phone_regex_2 = r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}'
+```
+
+Before rushing to another round of generation, let's validate this new regex. We'll add just a bit more sophistication over our last check:
+
+```python
+re.match(phone_regex_2, phone_number)[0] == phone_number
+# True
+```
+Now that we've validated, let's generate with this new regex!
+
+```python
+phone_generator_v2 = outlines.generate.regex(model,
+                                             phone_regex_2)
+for _ in range(10):
+    print(phone_generator_v2(prompt_phone))
+```
+
+> (206) 867-5309\
+(206) 666-7777\
+(206) 444-3333\
+(206) 444-3333\
+(206) 943-2222\
+(206) 323-6789\
+(206) 444-3333\
+(206) 867-5309\
+(206) 466-2255\
+(206) 222-3333
+
+Better, but I don't like those repeated sequences. Like good software developers, let's iterate again!
+
+## Reiteration - with debugging
+
+Here's a fancier regex that should give us more interesting results:
+
+```python
+phone_regex_3_error = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4]'
+```
+
+This looks good to me, but there's a subtle bug, that's why we *always* need to validate our structure against real data. This time we'll make our validator do a bit more work to verify the correct string is matched:
+
+```python
+if not re.match(phone_regex_3_error, phone_number):
+    print("Regex fails match")
+else:
+    matched_string = re.match(phone_regex_3_error, phone_number)[0]
+    if matched_string == phone_number:
+    print("Successful match")
+    else:
+    print(f"Error {matched_string} != {phone_number}")
+```
+This prints out:
+>  Error (206) 386-463 != (206) 386-4636
+
+Ah! We were missing the last digit, let's fix that and regenerate:
+
+```python
+phone_regex_3_fixed = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]'
+phone_generator_v3 = outlines.generate.regex(model,
+                                             phone_regex_3_fixed)
+for _ in range(10):
+    print(phone_generator_v3(prompt_phone))
+```
+
+>(206) 494-3216\
+(206) 374-6218\
+(206) 494-3337\
+(206) 476-3216\
+(206) 484-3548\
+(206) 495-3218\
+(206) 494-5517\
+(206) 375-4636\
+(206) 384-6216\
+(206) 385-6218
+
+Much better!
+
+Now you've seen a quick example of the structured generation workflow that can be used at the basis for building and iteration on much larger structured generation tasks!
diff --git a/docs_new/getting_started/index.md b/docs_new/getting_started/index.md
new file mode 100644
index 000000000..ca30f0708
--- /dev/null
+++ b/docs_new/getting_started/index.md
@@ -0,0 +1,350 @@
+---
+title: Getting Started
+---
+
+# Getting Started
+
+## Installation
+
+We recommend using `uv` to install Outlines. You can find `uv` installation instructions [here](https://github.com/astral-sh/uv).
+
+To install Outlines with the `transformers` backend, run:
+
+```bash
+uv pip install 'outlines[transformers]'
+```
+
+or the classic `pip`:
+
+```bash
+pip install 'outlines[transformers]'
+```
+
+For other backends, see the [installation guide](../installation).
+
+## Quick start
+
+Outlines wraps around a variety of LLM inference backends, described in the [installation guide](../installation). The following example shows how to use Outlines with HuggingFace's [`transformers` library](https://huggingface.co/docs/transformers/en/index).
+
+=== "Transformers"
+
+    ```python
+    import outlines
+    from transformers
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # Create an Outlines model using transformers
+    model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+    model = outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(model_name),
+        transformers.AutoTokenizer.from_pretrained(model_name),
+    )
+
+    # Generate a response
+    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
+    ```
+
+=== "OpenAI"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # Create an OpenAI client instance
+    openai_client = OpenAI()
+
+    # Create an Outlines model
+    model = outlines.from_openai(openai_client, "gpt-4o")
+
+    # Generate a response
+    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
+    ```
+
+=== "vLLM (offline)"
+
+    ```python
+    import outlines
+    from pydantic import BaseModel
+    from vllm import LLM
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        role: str
+        age: int
+
+    # Model to use, it will be downloaded from the HuggingFace hub
+    model_id = "microsoft/Phi-3-mini-4k-instruct"
+
+    # Create a vLLM model
+    vllm_model = LLM(model_id)
+
+    # Create an Outlines model
+    model = outlines.from_vllm_offline(vllm_model)
+
+    # Generate a response
+    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
+    ```
+
+=== "vLLM (online)"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # You must have a separete vLLM server running
+    # Create an OpenAI client with the base URL of the VLLM server
+    openai_client = OpenAI(base_url="http://localhost:11434/v1")
+
+    # Specify the model available on the VLLM server to use
+    model_id = "microsoft/Phi-3-mini-4k-instruct"
+
+    # Create an Outlines model
+    model = outlines.from_vllm(openai_client, model_id)
+
+    # Generate a response
+    response = model("Create a character.", Person)
+    # { "name": "John", "age": 30 }
+    ```
+
+=== "llama.cpp"
+
+    ```python
+    import outlines
+    from llama_cpp import Llama
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # Model to use, it will be downloaded from the HuggingFace hub
+    repo_id = "TheBloke/Llama-2-13B-chat-GGUF"
+    file_name = "llama-2-13b-chat.Q4_K_M.gguf"
+
+    # Create a Llama.cpp model
+    llama_cpp_model = Llama.from_pretrained(repo_id, file_name)
+
+    # Create an Outlines model
+    model = outlines.from_llamacpp(llama_cpp_model)
+
+    # Generate a response
+    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
+    ```
+
+=== "Dottxt"
+
+    ```python
+    import outlines
+    from dottxt.client import Dottxt
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # Create an Dottxt client
+    client = Dottxt()
+
+    # Create an Outlines model
+    model = outlines.from_dottxt(client)
+
+    # Generate a response
+    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
+    ```
+
+=== "Anthropic"
+
+    ```python
+    import outlines
+    from anthropic import Anthropic
+    from pydantic import BaseModel
+
+    # Create an Anthropic client
+    client = Anthropic()
+
+    # Create an Outlines model
+    model = outlines.from_anthropic(client, "claude-3-haiku-20240307")
+
+    # Generate a response
+    response = model("Create a character.", max_tokens=20) # Here is a character I have created:\n\nName: Ayla Samara
+    ```
+
+=== "Ollama"
+
+    ```python
+    import outlines
+    from ollama import Client
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # Create an Ollama client
+    client = Client()
+
+    # Create an Outlines model, the model must be available on your system
+    model = outlines.from_ollama(client, "tinyllama")
+
+    # Generate a response
+    response = model("Create a character.", Person)
+    # { "name": "John", "age": 30 }
+    ```
+
+=== "mlx-lm"
+
+    ```python
+    import outlines
+    import mlx_lm
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        role: str
+        age: int
+
+    # Create an MLXLM model with the output of mlx_lm.load
+    # The model will be downloaded from the HuggingFace hub
+    model = outlines.from_mlxlm(mlx_lm.load(
+        "mlx-community/SmolLM-135M-Instruct-4bit"
+    ))
+
+    # Generate a response
+    response = model("Create a character.", Person)
+    # { "name": "John", "age": 30 }
+    ```
+
+=== "Gemini"
+
+    ```python
+    import outlines
+    from google.generativeai import GenerativeModel
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # Create a Gemini client
+    client = GenerativeModel()
+
+    # Create an Outlines model
+    model = outlines.from_gemini(client)
+
+    # Generate a response
+    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
+    ```
+
+=== "SgLang"
+
+    ```python
+    # SgLang
+
+    import outlines
+    from openai import OpenAI
+    from pydantic import BaseModel
+
+    # Define a Pydantic model describing the output format
+    class Person(BaseModel):
+        # Set name and age fields to string and int respectively
+        name: str
+        age: int
+
+    # You must have a separete SgLang server running
+    # Create an OpenAI client with the base URL of the SgLang server
+    openai_client = OpenAI(base_url="http://localhost:11434/v1")
+
+    # Create an Outlines model
+    model = outlines.from_sglang(openai_client)
+
+    # Generate a response
+    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
+    ```
+
+### JSON
+
+```python
+
+# Define the input text
+person_text = """
+John Doe
+30
+john.doe@example.com
+"""
+
+# Apply chat templating to the input text
+prompt = tokenizer.apply_chat_template(
+    [
+        {"role": "system", "content": """
+        You are a master of extracting information from text.
+        """},
+        {"role": "user", "content": person_text}
+    ],
+    tokenize=False
+)
+
+# Generate the output
+result = model(
+    prompt,
+    Person,
+
+    # Note: transformers has an extremely small default
+    # max_new_tokens, which is often not enough for the
+    # full JSON output. You will experience errors if your
+    # model is unable to generate the full JSON output due
+    # to the max_new_tokens limit.
+    max_new_tokens=100
+)
+
+print(result)
+```
+
+Result:
+
+```json
+{
+    "name": "John Doe",
+    "age": 30,
+    "email": "john.doe@example.com"
+}
+```
+
+### Regex
+
+> [!NOTE]
+> Insert regex example here
+
+### Multiple choice
+
+> [!NOTE]
+> Insert multiple choice example here
diff --git a/docs_new/index.md b/docs_new/index.md
new file mode 100644
index 000000000..7ce2641f8
--- /dev/null
+++ b/docs_new/index.md
@@ -0,0 +1,147 @@
+---
+title: Welcome to Outlines!
+---
+
+# Welcome
+
+![](/assets/images/logo-light-mode.svg#only-light)
+![](/assets/images/logo-dark-mode.svg#only-dark)
+
+Outlines is a Python library from [.TXT](https://dottxt.co) that guarantees structured output from
+large language models. It ensures LLMs speak the language of your application by making them follow specific formats such as JSON, regular expressions, or context-free grammars.
+
+[Get started](/docs/getting_started){ .md-button .md-button--primary }
+[API Reference](/api/reference){ .md-button }
+[Examples](/examples){ .md-button }
+[GitHub](https://github.com/yourusername/outlines){ .md-button }
+
+## Features
+
+<div class="grid cards" markdown>
+
+- :material-shield-check: **Reliable** - Guaranteed schema compliance -- always valid JSON.
+- :material-puzzle: **Feature-rich** - Supports a large proportion of the JSON Schema spec, along with regex and context-free grammars.
+- :material-lightning-bolt: **Fast** - Outlines has negligible runtime overhead, and fast compilation times.
+- :material-cog: **Universal** - Outlines is a powered by Rust, and can be easily bound to other languages.
+- :material-lightbulb: **Simple** - Outlines is a low-abstraction library. Write code the way you normally do with LLMs. No agent frameworks needed.
+- :material-magnify: **Powerful** - Manage prompt complexity with prompt templating.
+
+</div>
+
+## Supported models
+
+> [!NOTE]
+> Provide full model list with links to docs about each model
+
+- vLLM
+- Transformers
+- OpenAI
+
+## About .txt
+
+Outlines is built with ❤️ by [.txt](https://dottxt.co).
+
+.txt solves the critical problem of reliable structured output generation for large language models. Our commercially-licensed libraries ensure 100% compliance with JSON Schema, regular expressions and context-free grammars while adding only microseconds of latency. Unlike open-source alternatives, we offer superior reliability, performance, and enterprise support.
+
+## Who is using Outlines?
+
+Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) use Outlines.
+
+Prominent companies and organizations that use Outlines include:
+
+<div class="grid cards" markdown>
+  <div class="row"><img src="../logos/amazon.png" width="200"></div>
+  <div class="row"><img src="../logos/apple.png" width="200"></div>
+  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
+  <div class="row"><img src="../logos/canoe.png" width="200"></div>
+  <div class="row"><img src="../logos/cisco.png" width="200"></div>
+  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
+  <div class="row"><img src="../logos/databricks.png" width="200"></div>
+  <div class="row"><img src="../logos/datadog.png" width="200"></div>
+  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
+  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
+  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
+  <div class="row"><img src="../logos/harvard.png" width="200"></div>
+  <div class="row"><img src="../logos/hf.png" width="200"></div>
+  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
+  <div class="row"><img src="../logos/meta.png" width="200"></div>
+  <div class="row"><img src="../logos/mit.png" width="200"></div>
+  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
+  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
+  <div class="row"><img src="../logos/nyu.png" width="200"></div>
+  <div class="row"><img src="../logos/safran.png" width="200"></div>
+  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
+  <div class="row"><img src="../logos/shopify.png" width="200"></div>
+  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
+  <div class="row"><img src="../logos/tinder.png" width="200"></div>
+  <div class="row"><img src="../logos/upenn.png" width="200"></div>
+</div>
+
+</body>
+
+Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
+
+Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
+
+## Philosophy
+
+**Outlines** is a library for neural text generation. You can think of it as a
+more flexible replacement for the `generate` method in the
+[transformers](https://github.com/huggingface/transformers) library.
+
+**Outlines** helps developers _structure text generation_ to build robust
+interfaces with external systems. It provides generation methods that
+guarantee that the output will match a regular expressions, or follow
+a JSON schema.
+
+**Outlines** provides _robust prompting primitives_ that separate the prompting
+from the execution logic and lead to simple implementations of few-shot
+generations, ReAct, meta-prompting, agents, etc.
+
+**Outlines** is designed as a _library_ that is meant to be compatible the
+broader ecosystem, not to replace it. We use as few abstractions as possible,
+and generation can be interleaved with control flow, conditionals, custom Python
+functions and calls to other libraries.
+
+**Outlines** is _compatible with every auto-regressive model_. It only interfaces with models
+via the next-token logits distribution.
+
+## Outlines people
+
+Outlines would not be what it is today without a community of dedicated developers:
+
+<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
+</a>
+
+## Acknowledgements
+
+<div class="grid" markdown>
+
+<figure markdown>
+  <a href="http://www.dottxt.co">
+  ![Normal Computing logo](assets/images/dottxt.png){ width="150" }
+  </a>
+</figure>
+
+<figure markdown>
+  <a href="https://www.normalcomputing.ai">
+  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
+  </a>
+</figure>
+
+</div>
+
+Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
+
+[discord]: https://discord.gg/R9DSu34mGd
+[aesara]: https://github.com/aesara-devs
+[blackjax]: https://github.com/blackjax-devs/blackjax
+[pythological]: https://github.com/pythological
+[hy]: https://hylang.org/
+[.txt]: https://dottxt.co
+[vllm]: https://github.com/vllm-project/vllm
+[tgi]: https://github.com/huggingface/text-generation-inference
+[lorax]: https://github.com/predibase/lorax
+[xinference]: https://github.com/xorbitsai/inference
+[sglang]: https://github.com/sgl-project/sglang/
diff --git a/docs_new/installation.md b/docs_new/installation.md
new file mode 100644
index 000000000..ad5443512
--- /dev/null
+++ b/docs_new/installation.md
@@ -0,0 +1,86 @@
+---
+title: Installation
+---
+
+# Installation
+
+## Dependency Management
+
+We recommend using modern Python packaging tools such as `uv` for managing python dependencies.
+
+### uv (Recommended)
+
+```bash
+# Install uv
+curl -LsSf https://astral.sh/uv/install.sh | sh
+
+# Create a virtual environment and install Outlines
+uv venv
+source .venv/bin/activate
+uv pip install outlines
+```
+
+or with basic pip:
+
+```bash
+pip install outlines
+```
+
+Outlines supports the following model types:
+
+- Transformers
+- OpenAI
+- llama.cpp
+- vLLM
+- dottxt
+- Anthropic
+- Ollama
+- mlx-lm
+- Google
+
+To use these model providers, you must install them manually, i.e.
+
+```sh
+pip install openai
+pip install transformers datasets accelerate torch
+pip install llama-cpp-python
+pip install transformers torch
+pip install vllm
+pip install dottxt
+pip install anthropic
+pip install ollama
+pip install mlx-lm
+pip install google-generativeai
+```
+
+If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
+
+!!! warning "Hardware Requirements"
+
+    If you are using an offline inference tool (not a remote server), your model may require specific hardware. Please check the documentation for these libraries.
+
+    Some libraries like `vllm` and `llama-cpp-python` require specific hardware, such as a compatible GPU. `mlx-lm` is designed for Apple Silicon, so may not be appropriate for your use case if you are on a different platform.
+
+## Optional Dependencies
+
+Outlines provides multiple optional dependency sets to support different backends and use cases. You can install them as needed using:
+
+- `pip install "outlines[vllm]"` for [vLLM](https://github.com/vllm-project/vllm), optimized for high-throughput inference.
+- `pip install "outlines[transformers]"` for [Hugging Face Transformers](https://huggingface.co/docs/transformers/index).
+- `pip install "outlines[mlx]"` for [MLX-LM](https://github.com/ml-explore/mlx-lm), optimized for Apple silicon.
+- `pip install "outlines[openai]"` to use OpenAI’s API.
+- `pip install "outlines[llamacpp]"` for [llama.cpp](https://github.com/ggerganov/llama.cpp), a lightweight LLM inference engine.
+
+## Bleeding Edge
+
+You can install the latest version of Outlines from the repository's `main` branch:
+
+```sh
+pip install git+https://github.com/dottxt-ai/outlines.git@main
+```
+
+This can be useful, for instance, when a fix has been merged but not yet released.
+
+## Installing for Development
+
+See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs_new/licence.md b/docs_new/licence.md
new file mode 100644
index 000000000..a74661fa3
--- /dev/null
+++ b/docs_new/licence.md
@@ -0,0 +1,34 @@
+---
+title: Licence
+---
+
+# Licence and citations
+
+Outlines is licenced under the Apache 2.0 licence. To comply with the licence you need to add the following notice at the top every file that uses part of Outlines' code:
+
+```
+Copyright 2023- The Outlines developers
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+```
+
+If you use Outlines in your work you can use the following citation:
+
+```
+@article{willard2023efficient,
+  title={Efficient Guided Generation for LLMs},
+  author={Willard, Brandon T and Louf, R{\'e}mi},
+  journal={arXiv preprint arXiv:2307.09702},
+  year={2023}
+}
+```
diff --git a/docs_new/logos/amazon.png b/docs_new/logos/amazon.png
new file mode 100644
index 0000000000000000000000000000000000000000..85e1dc81c405d777b6c5ce1a9d571e87b892d91a
GIT binary patch
literal 19489
zcmYg%by!s2_w_JzclSt13lfq7gCHd_bW14>(%l^*EiE80ba!`)Fw$Kgx}@uO{XXv>
z?=w8a%y8%2bM{$#@3q#6R8^6~#iGIjfk3$L<)P{z5I6w%y&D4!_-?F))C>GTv651j
z0)eXIu^&uPfxj8d<kgiyATJgWC?Es`x&xjH_zwhf<pqKEO+X-#G!Tf~F|$QY40r+U
zqoN!X^!)FU+g_9i0x1f=hf2Xb7LKw!eGD|0pI28!6JHw!RKR|lACEEBy?y(YjUmH=
zP^u3VMF?en0kQ^`=-S*=Z`@57lzR;1v_sN0canKK4-+Ir35~)s<xl7OoQu8u-!~Ti
z>-&4Npqje6_m&e`={~EK+QuYl-ji42s~2x}voFM@Eg<9L<D!UYCKyxPP}~m@S<{*b
zB_yaOE66pt)A$7>5MCj>Wf7FCEq20;8^nAW+EY8prF+g7W(s2-Hg`tXjFp3Jq9Rz-
zvVwM%-?>Ib#9#*H8bJrh#lvk${*$TaA<$(${r1)Y$ta2pD=TYx`7=Y6o&b(n-3wEz
z-1hc%`^zZO6Y0w=_8QhI@?U5%FHQM)kUr$!y@vS5;Ngh<m3BD6vhj1TYiNkW6tr8A
zfg~&=Bba!nf+N}a2ocq9C99}(xT^GITk2K+qnMq&HXl6SK47mmIP(zWmO52AYg-PC
z#7Y!PD~4|ps0AY6Eh-GO#S$Nw&NFjVBrh1Y#@PkUSjWe?P!rs}A>N;-$PY1<xJ-$t
z1FT^6GSH`u5{bF=QE@W?2N++i<@iEHXQ#h5X3#R>ncA|7i73?V7sDJ9A^v~JMrsM#
zC|V`^9L1I|Acl|d$!+ImO|3-6`$kIs?{yXS?wZTE2ur9N)OFwFo)L~J=5SPMh6dq@
z*q0nsN;`FtlaupT#^_4f*;z)Oyw8mod*|wJ(zeOT_A3u^{lZyMB+GA33svC*B;gb@
zC?1Oj_YggXcEIKwD<0Y1E3{{_#mh^imveedhw}}?IpLaPpgW#MW*KP$I%be}Noi|&
zX=$l<-GaP^#`M;!ssCOT)BSYB_(dWum=ZZej}avLnyOtqS)sgo#U@0odPx>0$#6<%
z;@(BS=TZJu`qJsFO|p3CanI$>vvlwL_sWV<qji7OHFyvlj&x?)VJ&|8wZ}Jsp{cyK
zwY3zEVq(MDkP+phzMkJzhGkR49P#i?dVK)qe!Z!=#dLKUv?hPqHt>x?h5!XTiPMGq
zB8^J{LUteEvGS&v&IpaV75_$xJ)?b8Q@pXh9&z1%J~F6jj_3F`(wLs@0S`g{AzSdM
zp+A}&f2Ku4i6v3=lZ1azmmJ}n5Pyg5VwAf8Z|Ig4QS8dX{Csms6pI16B0*;S{vu&I
z@5tba#2V$=V2SP3Pxhh8x|^p%kOY{9>#NIGOICR2jQrM)i^yX<JL$-M<o-Qjy3x0H
zs^if^l7d&)&jpQ*?!Aty)4zY$ytsYXCD~D5;GqcC%f5%Xu5W#wwx=IXQ?o~+c;EN&
zggjvq7j9ev&d}q8r@CxA3xPwUT(3Hwl>;l2N>q&h!d4g4M_qA-|MPuT4UJ~2h)JF|
z9HD2%=$K0$<>KBwwA+p%{Q|uNL%)w27IahI%}zHjDbK7$)t%h8dO<Frm#|zalo38?
z#eb8#JbQb4y1G*vmCd)o8=j4tk@2pa<y+JxN|t`imgV*Jt}$2F{z~E}ip9r-S532-
zqsxCyl5Ib{AJ|-#y?o%C_tE<?b&Yd*pk!vY*%WqnhpqZZ#5*A+(6$ztxtcz_Wb_JL
z^}k&@%98_!8#88p5P5M;dCCb{JsdIj)BEg6O`><5GC%(;kUZYtu*qrrgB(Q?x@A6y
ztZKCOBT}QtFMo!^imSsvqNlLSp7HhBba8h2$TnR?=$Q~*{K;;&keaVUGTAz5C@N8;
zPm#1FzT)p|{<mW@D`XGQ*Wb$u%E^KyF2GU7q%7L1h86+zm*iO5fB*iqEi=hEm-uj*
z&G%vG<h@m`9Ioluobfsu!+MJZ30;!3sS3PkH^&s~EwCUt(p%JVDOHA!#thDGEQy{K
zA1x(1{NFfQ&J#yLA7Jds-irD0{R^5~<faslo#n>#-&$HWTU%OEV*hF?D=V8F{Fpk4
z+JD1X`IAKTab@nP;oj?w;IoMCazy_EIN=0LUrcCx3OOxQxGzYPXZoo~&}J^;FzNxJ
zr#ohUUsC&<C_1YcvWo0hw#I%yh<M%~*{-PU8+${A?0Oi2iMzPrPu$_3-Q^zmDFRVL
zCJapEO3nCyP7T_77AqFtX9>dF_qdUg0li{~PFA1ijJE~S=XJq}7NI_V3;KAcXE))n
zt##FY`hcNsDT(J#Vs!EG@>OGLQ!w&KF$g^8#lwfsgi!9KCi~@l0O?~9Go~A79x~A<
z{-Ok)HDP`9T`FC%?^in~uOFd><_gdO6I;8`0>(ozV0S#dU-dVTttfRoL;qC(nPuKQ
zTkYRGGi2xWPc(2TSI!W$v!gjBI=hh$e8$#&K(fVu_|bp;?4UnbsgCg$2r+VXwuh{Z
zmmPWqMr)<RI)@cHoPwCa1t<P|s*lX1YdSCWynkZ{g@<u~i6i?KgFpS{0iP6C!B<)*
zF6iE~B=VPbbL;u|`}1Ht<((SC{;bETJAXnUWBpidHbzXb$d|4-uz5lHgmGwSIHJ3H
zQ1aByOm?3su$d8_1r_rPEuJVT5PO-K`uCBeQ5Gdm9@Nm#V4w5%QD5~jM}ZJ*k!?*+
zNP4RXVAbNMhP&>-lNPMBv$ksA@8Sq=OF==ry&28Ylg?m<rko9|56nbGsvg1#@`u1)
z;~~q)KLtTV<AdyZ-$U&kBKs#Aoql;lzjD@7<aA?&8yAHbFB$Lz6c6<S6DvR$|DjHc
z)&M24-%r(a<D@<_<@7PSkp&!X_myjWycHc8*}u@k))#iU_jYStYjxEdM&+w9O$TxQ
z(jVFB&u^1kaZFy@_{zWIgcaN*XQB|Q{rH*(<JdauRsIH^Ca@r31t+Vy5bVIasXA~j
zguMRdN!w2N^~&pacTZ0uyr)8>p2o2xd%fBFX8%4f{;name}~~bG|ElIQN@{qc*Yd7
z%lylQDB|J?R2TF6Du<-&gu_?%Cr<7B(OH|mBYl&1ahPQ5;&o@Wrl<;wdqkHZ?8%P?
zL_!rTzm@Xy=g;coZ$$$G15R`gR5vr9XmLn7(qjgBUJsi+jA{>T){^>_{?aj%%_z*i
zc8oqbk+kkN{&I25?(F2Mq_(~2fY6d53mD$nabNfBEUbyE2=6w>3~DX43x0|~YY7ZI
z(H`9u55;b1P~ecN!k8<BQ%RZU+{tg%C=8AK-J2d1nn67NjOuMo`nC4AGx6+(Rq2ib
z2wVSLFChfsfwS;o?rEzzr`udH@l|=b+02FCyGvWoX!fP1!VlLnL;;b|jU$g0UzVq*
zeihg3>)QF``!{vwOdqSS*)z3db>mQ~BDgY_RPLTi;9k<nf`c#rT|15PlY>Jg_3ST4
zbjRZ6z#p&WF5ytfnOkYplp>JG*c_?rJ6Zk2Ae}0GAM>rx$+s?ciBxRo5(ut?*1}&9
z1+<Q(y$WS$BAuIPxjhi(J-8QWC-f17%5I^<R=JRmhCLoK#}}N>2EJME{04&d*VSUV
zUdIDb;QXIQ<3VA}po3mp@)yvv*(IN^5liw%*?#`D?a>b{i0f5a>4klf?sR2ke$_{3
zSCtRj1^-qyoEfyZu&{rBuy0>pedEWRd3R9{g}f@&8cmHJoXquL^t{ZwD;M(nc@qDQ
z=f~e`ei3>H!3tygkto}q>q%+z?8}Go8mrY7S2sS7B*OT?%?-~^a~Ikm@F$G5KPf3G
zmV<jD#xX;35OTHloJ;XYXom<V(p;nc@~o=(P|GbD<qN6PWj#(EZ*Ol}I62ITkYz>|
z@=WzLQCV3OnROPrS)3id`HN`y=TC7IKNG_zu2dH8Q(u49(#WaqGTqt+YB7xZO}6r&
zz3-FHUUz@@HkaM_-Ek_M{c2jmvBh^{u{HH+#fT#Nvatm7mqqARn27FY^zyVSbexFR
zw3}7GCjzL9Ob|I%<aTA-P}jF9C#=LH{n|?)-yzv-C=u^Np9*R`(VdX}118)%=G=PD
z#z&^twMjqyOAENTUpP*E+m7kS-LM^5KrJJ?l%qht)ra08Yaxsdo#1|vQnV4?6b4}j
zN6np^@77sDF8|fQl~*On5$?JlR?&=ey}UFu5@}OPoxXMW!y|l#tib`BMAdV9^;*T<
zRXlPtP*a#>{W!Qvlui=&lTa(2G`fdtVF?-rbj2`GJ0I7*#baj;uB5DNNg@#uEDG1x
z({qE}w78zxSuRmbicZ1vKKjhV)jPjfYE0uqxpz@j-svVhE3zb3e?PCzcIQTP-d#Ir
z(?!=iXo4CV8)rYeo;+d&iv;CbYNet0!DQ_CrC>H^5~H+wzMY48azQ>%iZBn*DNtW>
zahrt3th;6iKGx5aixNZ{pw2CyR$6&uh0==>D5+e5$gn~i)waE#1^IyzCF+vwTDshz
z<AERY4dYGaWAPJi0xN;1La<3_*M8Rpn%E~M3^&-M;t#K{T}H{dzft2*xdkLmgsGQd
z_$b+&v-ehlf-2*^>3ONd>ctYrk(^pOkSaJs4I>tD@P_z=Xvw|=7+o_+M{~{{xYHPN
zFRzp&L>{_Sr`7UWdDDI9$_i<rk0~GBGY(u%OO<j~`xnYl6_nG>OC+_mW`EaF>eSUl
zNqJmxJ)j)(Z9IX{D{Qf_p?cICU6u^PcCN|3&9kQBY<5Stq1qbXe|tq#MC5M#gi6KI
zr613*0kI%$xJ-`0^U$prVQ_-leVNs4FxVjWYi>@sBCq23MOP}D3XaSdi_lMh8P<xn
zm6#HTITi1oh)lvR!^Sb3COJNjP5uMQVn2>P$To@n^_u;UsUqyqo^vIZX`!^n;i*q9
zKZ_f@;e*9P>%w!8bCzE6+mk<k{^Yp2xmA=qU%V+qYwhc(WxQk965;^q`}jA65nr63
z!X^(!wvQ>oh{r(2(Sw_kr?5`^Oe*#5IQ>q)$CWO#9WPwDiK{FNpZ22lkn#D5q3b^m
zOE;a0soaK=N$9|K6aoo>h3fk@Sxfiv10&aq@x^nOzyuhuOKHO9?1lMK7`wf4(KLSY
zdFbWTf4gd~+zqH+gtt3=FE#wyJkB-(#~*(d6<KYs&0SD9>K?>UDoZh)b3s+yY^gP{
zk&%tT9hNLQPBABO{Z~w>`fc+(=vbz<MjbxCXG&U&!Wf@c7*Mo}YUZa&RM4^VaE4@y
z94IOxl5J9QV*2rrzSE=@^<oE!tEZ36hO<Ad{5R6Vmu~Zt<O=C`-hYz9zhp)B435z*
z+vd9r-{M!n!B4gr&BkT1WLUdN=b=E|me?B+4L{ADK^!5HhdGzN-7L~%DW?1_aBkK>
z!K1e&P7S(@u6JRp`FYyr-eW0&BqSgZiyHSOE&1*G#h&ezdd>XrH+{BO`zbn{KOopL
zX(jE8oMVpihKh=cX4=}??vyJ>E_}s~9||gt%?ck}lJ@s*tsn`)(4aNUrBzU^++y8l
zVv!UcV{JD=9fLiuVj>)8L1=BqvubgN>Q<2YNoQF!l)LZI`kc#8@cs;)QF4jg#iI7Q
zao{658U%lh+vF032crX=8LXyTeB7psB+NqeM%~mq9Pim?&$fXxoTbINS@!*E4sMe~
z>?|J^E}S|coYV_y&J2G9YryIW6CMK$;dSP|U8{t;EWe82n<zg&Rf`dGFege8Fa><W
z>Y67QkEA)MqVT6>BZ*VCs-+%fh^u+ODoNQi?KBO&qoUgAn$91l-TGL~6aMlEdf8|=
zZ~LD0`Cl7spIhlrvH3%cC?Zps=<J$3t9gl^tK-=zbakDLqN68sXJ_Ym@A&wbmNbm#
zJgN9eP}Mh6ev7$Ml??^1!9?_#rcL>Xm*KCb8fkj7E)qq=7@4cv=m-}u?VY3P0jZNY
z^w4~I@W7ig<D98(v^A>BHu}Qpz1>xH!p6GSiIe_DrgX{lDh2d6q;PG|%S(KBH#hr*
zrTA~dTS~*bl4bqgPBP-Jc%)R%C+p^+A005k+KJuUPze3r%itUDWSF40_V4Qxv^eK|
zTk%PHWpEYZ`WkPIgVEAdw);XtCQ>GJ3&;jIkVR3>K|P+FXAkS)Zyn#bqReaYfwEW@
z>v_8v=h8J{$iP}M?7w!f@yUq!(_gYq-CzGY<NfW_570}Sw_<&$8R=E&!T~?-=;U;D
z=p!<f(>HHjxOIOcFC8{{v5tQiuxX_dx{;f?a8tCXwQ^8?Kyq<)b#=F>HHHw)<9hrE
z8g*q2?KuY8RpJtZ;|?yM3xKg-N^5gpcbc&(&B;cys-R1c)(_z^`&$vv9P&OJD;BCZ
z*RUe5V&~a?j3J_;rXECfBxXc2xQ41sn)C4_w~*Eg8#Ql9q7PPnBeMI{grT#qYCHDX
z#-<c;@M5-fzer>L(7Ux}UefMpFq)i)PTqoV76o!bkYF61c7noCNgQzI;R9LQQjrOa
z9{u&c@ilViit7=XpnaJ|k^dLYjI1n`qVTrQ{K*R2;Dq1+boMeO<YGU@<Z(X+#Pr2v
zp4i;f)Kt!dWhc+6MATW+l1&T)+ii#p)Bv-q;=RUCgCnLq+bb!CgC@^+qvS|b`}-Z}
zV_0@{DB5n8e2uAh?ObO!_W|j6(ZAE#&sYz&J)drOjgg*aO?6iqoR(YSQ@nI4L0|mU
z#7_cTxZQhI>Br3ZN?ZBT%``8szMdO(j>b?Xt~P<2c=Vg;Dxg~j=eP7t?0Z|oiS(`#
zyxuEydM>9}*L^kfiF7pG*+~F4r)rC<FlxTs_z=974=lk3VN^7Fz#CH%27;H5QMK?=
zaRW(;k>LR%9GdIqqt|+g9c#*pSdo2%Ar!m@jHGs-gey58IWTtKaF4pbG$V=rudC~l
zq_L%meR5vF^0*Xn5vNo%?N8q=ij{GG`vJ@oq81Q4^<P{iu{WVy#sVpwpZ)N+R6|mi
zvyFaatawq{ufZnLv6Nz1Jf%Y~@l#A>oVio=nN<n3Ya@p;UEJN@r)&c2r`*LgF`BDq
z?M=KygQfvQZ@ro5Q{wbrn9UPbdwOiF1~WzlbDMC;Ek>qrQ;Vj$p1yt~Ibmv}Us==)
zVp(+p1<m2zZ7lyY%83flgW2@2$ZjS5qeT)riU1A-8J>e%f>TKzfO^A^_i;B;rq_EM
zblW!lSFCa;vm?T%Fe&E<7(^Bpewaa71~B(Filv|09M*4$L!+PyPYND0ffV=pZ?4{G
zCh-uIG!L(3r10$@@DzE_8tY8#iDu2)>K@krT78x>Fhc)%chR`uZP4Sl=C+~9=2J&z
zYp7XyFk)`cF)y4py3BftzH7*WVZ;MMcbLARVb)pq<Y;4Xd|K9*b>^PNR_KHf3yK@P
z_EM(wn%-ti`E2N7lDk#<H8=;A&UGPuw93`&^l7!`IF%T&nAzy{`umq9K>;q1{>Is4
z@%|Zr1!n~>H99ojA^B%&coW{H6Hgs`c7A!IM%s85I@&+b0z@vRb*SRfWV(5)5H1z4
zOP%qr2%C~N-aYQVw1Zxwi?p`38j%0d@a=c<s;>tZduhJ0EZl5ZJa$`4=8a;i%AHB=
z>-;IRy4N9G$eb}Qv1rV8$mJ;Eh<@sBqKlb0M-obH(C96Z<8JN1&KF13fBhH&H=_AS
z_J=sqr^dir9zFj})5F94NMDHty<5Vswb@L6Va~-BezR(GfyLG?L?IqG-Wme2j9K<Y
zj|_pCfr~buJ}xm#{lk55^DJ_TAuTU^wycU+>Q`B|kW7?2L=Y*q>yMylCzOegCnijT
zuqgcP%!QaWMl5v1Ip^~~f)W5tWC<GS^a@V>U96zhP0v*)Z3?dFcH9-NJ8>goiNb$`
z)9k9G*<05zC`1{)zy=uy$HGOB0p@dMKtCL;7J6Ycw4F(h@*UuETQoVeRFCX5kUZ3y
z2DbZ=GAR2djuZ+<8Y2C2D^D<TyVv1k$awfrBr%-BJ@#s8CxJ_Eg!So<mRy>Xqr=0c
zq`!}=NE6@_bO>?78!8W$L|xo6G?9kC86ILQs`Mza$Z-AwHWF&tg7+q@Fj3~5&L8|4
zN0|BRFxRYz<wOZ3qq!6v{}-P!(FvV6K(g5i$xdrJtE<mrH(_Y~h@Tz_U785o3X~#`
z1C6EZll$oxQoJ&Ggp<*pRT>Pxnc&F9E1479$Jo7T<UB@VU;jQTH!aR;CdRQ+#^+QX
zZwz$I+jWTpm@&d9v}Bu_;faipY;g@fZ)$11LE@_TsZ|Q{Dd0~%U^6oli-Gea{nvf-
z%?Lel*!-@+)C9u{zVG?SjobcHmCI(VqwC^nr`etw;XQhLSE!g;@!{@CQFwu(5)aP-
zw_dk){xG^Ln$#MB+_W@jY*u&oBnn12grtRTvpZv1tn~hboOI_oL`4YWe&dlkZE7*3
z@O0PL*Z2Lmpa{zLeV&meGJmHfvdj9r&mX_U?s}0K@z?sDwGEEkK7|D$UC)|FHeq1b
zE$KSJuqpCu{6)z9fleBna1ijz-dz<jJ>r|t#m#T2Bvn7YS2lc+?UHU}#eGfFe?MXP
zs=;Bo#a%0ZnS8|O%OFT>vkpH{$QmX0doO*UHdZ$K?g6@)w8%2(0JVFSmTA*AKRqH1
z^m?f&*yjlXyT;%`hGS1R8_df@;?5vwLo%WXp9D;m3e{~4gjj#=)!i;F^THI2hAeCM
zoLRn(ZfhZNEGq{x^Lc53mPJdx6J{O=+IdY#@!4O8{h4i*pPV1lQwA+;O(OwRlq0Mc
zXKh7O$nDTy_<S8*gezU1r9i}Y!Y=%mHAz6lmF1NPzmO(zz<`!+t;?P|Icu@ugwL!S
zP>s8OonWzI#sU)uywi;w+$1J!{!ZX<j{%j?j9ZittkhVlaK;LOfchUb&F?jrnN%((
z<5-YNX;hSyK~(zN%yN6e_b8lIeQN3xD7obzs_g-dc5p<fFqS_@;_$<~t`|h*`uSyn
zOo>*4{os2QYkPb9GM2^o2;uo6?y3<}|MMR+1htF{i!-m=C8_M<ijPCAr%l+tQ(}1w
zACC!~&C4xec4^ZzC|71Y6=oQO($$CuI9R4q^8du~SW#)CYK7HPRKm>BxXoG-0X*_|
zPelRo;c20)EXo;Il3Pm#8$|Jgtm-jN2j7&~1gKHOh4B)|xS|dcZcK|zUz|fGTRumb
zB=^30p{@z$1&A#Y!W^HKYv>012C@X?6)eM<m-bW)@xv8Is{qxXG9j7sOVYPd83x6y
zf(_4%%+RkGL24E|c6N4`zBYgQW-ff7z$3J&Xy4`)?_<o9P?T6SvIwW8puqT0Z@lza
zj~682Lo?1-8q6R;S34f~XixD{*M*ls_&U)X?u~D<OQ378)7MdBp6td2L|UGz=l{kO
zQ+!NT+z>g$9dQ_At@dBZaR3=U&%uFcoadhN8+19v-(YIP`X5PG*1cRs#3<H3%x>y{
z#<*O!!^aZIjd%_>Aw7WS>m0zBsz4#<dh`r}^BkkES?D7T@RhJQkg<&AQ(t|$Cwo}2
zukfybSK#LR0eu|bEV=gZkYr`6ZFV7EY}a0DrYfq57+ns=ArRUf5r-UAsc=E-skcf?
znqaufRdvatBesglZdAnXFR#*;{IRf4@h6?)0go_qV>RDhI4OtQy8;#Kl3dMqJ=dKW
zG*sF5Q;UEKpgEygDZkX5B@40XDc*c~dJ=HO#kF6oZvyLZah4;Vj$j}`LBTki;!;r*
zVTCnS&1F=AdzgOrw7)w#rL_u3gsdJB1^|Rn;84PkyAGN*YzlM+!M#!pJ+tZtPzUWt
zvJ}F|G2Hhaf-Oev<Oztp;*5=QS<gYQE+M2QN@~ZpEzK)52W#u^2IP@G(lAN+S0MA?
z$o73kN9GimxORpyvisSs5~w3vKKg_{&r!_LTvaU1v?G_O-`M6L&BSEw9SvQI=k&w-
z%Q_bk+L8U}E|rQSK8&DxM!{P>lbYVYKk@GdP3ErFC&-N4tgi(aK<xl`R~(U9;7G$e
zaV4$EO4rm}J9W{ASQ^w#hxH%QeH%89@p&SS7}JQ*uY~)!n5->!Wiwi?N5Qz=<vwkb
zduANG9*m)^m8d)Ku?ArNKu=Har<<!Q?)6%<;^2aI>0?^~xXNy!4wHH?UXm8U1L_z*
zEX^^w;jhcPwA4&3wNO#g5TB$T&*3Mop7WO7^1%x2CD7>6oVFhyM}O(_s*lqpZn<hb
z=xb@WD+a}%O5+^-(kgWWBp1;Jl<vg<i{ur0lFTd#{H{IHkIaMt<p6YIAtUeF24Ks&
zMm4LNQOqpust^8X`1tAUQ3kdJfvs9t8X5uw0??HCfuYhP(h>x3s?BcCT^Y4xP#Q~3
zL9buGMlmJza76uO_Z=>Z^jy64-e)Rd^lkaoH6eD_ACCu_?yjz9;`=(YG!PB0#R3G7
zLXW$d;a)LWE0yY4p^BJJ!5uw5MUB3;VGW<B2kde3P5@vFYX&z>HILv0?P>BQxU+y{
zp@w%8rr2EoQD_*nR`d2!_~QFJOq|u}1!COn-aG8tb0*4oe>#K%^5*#W(!_!jdhAS!
zk7Q3tKy^>;CXO?M2z5PW9$WOuEGo|ys-U_HjcOv<N%qGm=)cCFVrU|q8(#XU*7E+7
zU11@7!}=ft9#%700Td8=yu6OIaJh6$ZJvLE1y;25QgL&(^g2IdbZ+<O3?Y^fSnyXx
z+JXH|odZwCjJvN^k=1jkSl8cg_+zltCmG8?4SgSv{t@3@ea6WdAF_50nR%I^$P%po
zS`dZe7-``irb8KZQ;aUO1F!X(aSnjj`+o5#+x1e*F6>py=g2|HGPn>QMcaJwMng>h
z#FV4#gI$Tz&UdWu%At%(@A5%3X6=<&C?)`bVrbtVl>;8sOQsz7kA*Ngi<j-`hZZ#=
z7Le-LLz++Q)1moaI7(Eo669yFtt6e(W_1@?*VKj@e86#Kdes}2#QtHX0NYFq-C`m3
zM@-sAyn|L?J27aj%@Cp>bAR|rv+Uo2ibSE6cfnYoYt_zc(6HY}-0>_E)#wR>a0Ko`
z%_GseiNtt{q^kOebpx%>J{~dX5TCV~yxk%QEG-+xHB0-a4KlpM3X?~*tfcqA8_H{e
zLQ@#NA*Ya;Xfk1(gQ>;;<CsC<ev){m$$z+T<tDqgOsl3s<vtj(JSppfV{Y$+&##G7
z{5Jin8uTZewJz3<FaV&am0*)4xepTMkZZZMr>=`^ATKxD`SR}Fd+UCo9^maahO8?R
zEWO*dT{5bGUwb?H6GT3Eo^K6LavHUHm|#4az|h%k&mJF1C-BuC#9N(A)`AquG8ixf
zWyP);#d5MFd~p<0!-%j|VOf>`MGv7}l##N7^nIUc1NZTzC-e3tWw*SrLUY_M|GMZU
zQika-EG`~sRqC8+;I6xc2vyPDyo3uCIX9zwDi3{i5wXLqk5$T%@NI(|MazF`^!-GT
zxwN>bp8$TVDWi*&NfIl_$+h4?VG-BVQT_^p@usdC_g^v*LV*i~i!73R8o-{5I8s})
zPR*Do-xX>!kd`NNX!$6cbP`}+@tyVtO0P#{XJ;q>nR#IbkjL=>Su%f*j1^toC5=ih
ziBY8J39lD~F?Ro+nIwkrQSu_`M1oae&)8|C>e)s_3D`iJbbxsRO;VEq4^T+_Q)xIt
zE=wvvoHT4C9o9XSuj>Xj>9K&4W&2W|J@Ix`p(1}}Wd)G!PSE$|qCrd{r6+wTCfk6H
z!}lgpnvpZyKL?th!#Q2TTfHL~!w7etYz1R%)~z_hBmwdDxI7t&$K}=P?;5kOlRP-D
zun!8<?eFIn%weuCA>;O>Y93%!Qv(d-E$fRaQWlUv&hyWRCifdSwtDl@{I)Y5e7|{R
z(}V({1R)be6Xp#mFg`uM?G;Jo+<SQf?kkplozqE`MyH1cov8TXyTk)lK{d&xS}1V|
zJTfSvjNqh9w-iti!3J1*@2DY6MvB#Bph~W_$fo>$#~;HOJ9}PE@cSr+3@bcE_o03G
z8X+?5;X`*CNKor&qs{Wg5JAzo+fE|=T|?&ZXt7nLMwQJaBcq%FDPq5w5u-31ATRhw
z%l{LfyeMnNHCvSL`xuVwEuzx!LHwlYA8RZuZ!>cf{AXu}3?{Sa&iYF~XEH}k+4ZN?
z=|>ER?cga^S%o&}VtIL4aeP`U0%ZI7^Sl~Na!2?P0fprBbY~OPlfLbpyQ&_#=!tmt
zG!n^r3P<?urQuF>UD%_rgfMdjOy}s=Y;Cibid=(L1zsZ+S4bk2cF1p^pwfziRk1@q
zY}`3pCOp$B@a_`+HW|00&@Tc&j|wjLMy26J@;J96a}))$_Wbt7StaN`OepcIVoS|@
zYpDYo)3ZyKIcQp%qud6G?PHEVhXM%RvU1;Odw6##=`6JV%jwTjvLXRWLrJ6h!bg)e
zk6G~hwNl}z>vTny!T2yY+sOU}wMXYqR9M+Zle^Eg?Dqjf{1e2upHcr+a)VAo)TEvZ
zVs@bzT>yLcZZ^;e)$ez;&#j0%G2Z;d_r+q}DXj(Zx<QdE)ON%9!=sAd(O1_Wie0_a
zBz*5YNSh-arEfqu|Mu1Nv<6@#$?iR4RUpK)LH&g<b-17vuRs`}meA|iCZ`g-W|eR1
zW<h~~iqScM@EqHyPO)iLu85(KX`6Q1!Fb9GUlU86z36>y_S=T|@&uqKQdLH@5g;>R
zWEKH3?|2(9$hUzqv_-A25zCJu2k)x|8cl8c8Bw~CgjaY<r5uZ<|5`7R1El^nBOhxY
zr`S^RsBh;!hk8M;QxsQ-lZYYHj<9U<-CfbxBCy6#bNWHNoA*oT&VS2+n6b&$&7`L&
z9^K6;J42eJ<fo=C@`BG28O;O>1P&FIgc4^TYnM;~g@*<Pv+FW_i-@S`?DxiU5R&lC
z@Zeq%7}M(3U3+jZ_SFR+-0_!p>n4e}(SJvV>wMjB>&Gv<j@k2k43|l}7JK;V#YKUJ
z&pMAIVn}9U8YgHuvk0?1ZrXc>#?FEu7Jf|CgcrHy`6hLE+}8SgLili8*5H+;(K}K+
zuU1hNzTLK%wU!tayj*7AXV_Xk<M&Apj}@s%kB3~Li0`0D{TJ{962XWe1z4(e-mgu8
z*KIR^A*PM!zUQa(di9;WdE)4>PwoQV(q>cfiKCce6L~%#N^aCm@n&T;+{S<<HU1qw
zUS16GSX&Hx-=!92QAaLNy!dUlxZVqMo6V!H0al~*@MrRxBU@JZ7eW3~5W@fHc1qjV
ziAkE5_wDa%AtcH*N?49eL2s~?dz#2)$v?|Zh-MmqlK^u0gTuprpez`sI5|4??L}3S
zU|j>-m;|YK;Xg4=&7#$3*~?!9L1RbeI|3nsMAgCpjMEDXExx?GYhD{0Fp&LRm8m4s
ztP`lw2CyMWQ}!Cko<n9gMbaGhD3Hg%mW=)7(MVN*NeLIArG-zcdh!E@UZg8o)^Aqt
zGeQlpx_qOb9c$$z=pzqzo62*X948yAV?C&P{0IkUYdo>!!c-euGdw8~Py&<*-@piA
z@k=(iT==Wbd0tcL`71YOpJ!>B@MEVw;m~Wg2Xe>oq87;BJwB^hHTJjRf<)f~EPP=)
zYHEG6SE=_PeAlIQe~0XbAQTgfT2ryKE%tIK#7UT7U>N>C`P}kjSZlRe$YYJF<1zMY
z3>Fbrad-|+z7rG12p^;E<zqW)hlGNy7?>tt$o(ZrHqiYi0yjkutcC@4u9)XS9eSXq
z3hT%0YoqXC+8vJ+Pr3JZiT7-f`j;Tg-8cg)72}{ae$zLlUXzQfN@K;4rF-z?7ePuO
z`Cf;<rbIymkEnSE?vsYNFUrjw7M^x?cfI|fDE4##l-|P3Pc3zT(1QeXaJS`9(JImZ
zp5b0}v$s61gVPScHd4$8VE|Fkh3ceYi~QjABg%Sq6hP3Xsd15rO}2kkRbKm~C=c2V
z*J_5yX>2fbM(P~u%)T<<Nb=GYr%PC6^?gnU>>%(ySKsK(bD9Mavv>2X?Ccv}0uL<K
zj1uNtm|dVdX~#Pi!R&YIqj?i?vJy%Z3Cp-<9)Cl>?#!5LXv}YZWhROYMzsvrX`J40
zz=&4|KRc(SA7;#<Hs$qS=hy?qoDtUZ0b=%<&e79xYFfUaY@-ZtG~o$P?V)IFz2l$6
zA`XDmFow8XQ{lWd6x=ww;l^SOH|RWreyJ@7Qz6!I&Tn~rkoG;dvkAP2P@ZYM9^u)B
zaQAw^O~T`tt9wLdi{++?$VlvXzW(RlU^>|cxTykoqsVyYP@7bO7MT{e@sktuJ1tqs
z3!udx=dA7Vk&q$?o2DlY#80g?pyT`jz8dhzcCScJSHp$&q~ky!^c=hQfSG}YeKYA$
z4y~~;CQaJ^+1rt}t5alNJU|j~;(Um4{N^0G>BFj3u5-3`?&|8)LMv?=nEnwJ>n*7+
z<phcHZu0p0{S)~}-JX|l*mHKlN$0APoj!a7BhhbCpHls|91Bro4N9(xpukbgzHIW3
z>KCftlI4*^O$0+w(P5SUo`1DOZqAo3`Qp53)5pJ=CV}Z&_1wHLj>gZ^^RSK%n=u>?
zRZx^QOTpCB8rdpOGqDZO!vsdNmwZEgsszGwWo*3I94+{;0#AnGJ;8%qiLr>ypR%I}
zs68tCp58!wec#6yfHfsDy&o^Knkd%K6{<;xk}5LsMbVe({7AinH$na;!e`X!4YzTN
z%bpj1Hx@|%T|w`(tbyi8`~N&6FCV^P-M{`(br9xy(D8A{5*Cby^KzL#v07*Ns!jlC
zg{l8LXeIqrT6ISXt)MAcB~CN_DFT%&yUY7$Sn2icSo4%bH(1qw2qjyUEFj^r2EZf;
zrTeI<{gq`td90E3AMHHwX|i$#ABn>XrYr@fff8oVTP7gl&e~Qtd58WI&|ZadRqM2-
zXR49ivoC9Ul~nl<LT2weBsVxn%&O-K!an`AH$ZQu4O)#Z_Ol5&scIbvSMil$;yctA
z<V#n|Y3X#B*Ju)$UwWuWXfyw3ZV-y2zW!qJVet0ZNn3lCkG~y@G~l0N%X8}U2|0B?
z6DrmbuuealT5yuXA?3(ZC4)Nan$<J@PnMe)n<$2nkcFrDC_*psUql9`*xLJ9!OUMl
z(6v3pb28+a;wZW~4naeXXxNj~Q{{}*cX!-qM$VZ>w5M&Wc=irde38%|`qf{jypBnm
zK9vcd{$^OZl5&F3v!=}Q+!PRR0O~C8kB+T;j*Xu0^D+SABsUTW0$gMB-N25H(ec9=
zEH{k!uw}WRPk$p*bPk!qPAyC<%{FHWPK?8L#j%*+d2m2}04yZE?f$xBddmR0_WKfR
z6_0W6J=QggSv3bB;R$lN)B)x)M#^Yk<)F2fCY1y%Z&n)+og*wFY+ZC32^!LF59sKp
z=Fun5FsS`-uLFPjSO$2SC2=^73`-7Fgm!w&UUoqY5c%H!3hS|l7=?V(-mpPnVzqNU
zPONp5v=a`1NY-UPPi&E1|MRabw6zad<TdSiG9t)K<5{`Uuyc3#aJ}ZWCNI4ZW`>hB
ztWm_R*Zi@+8T@|mJvQWmjYJ|nrr%x*(F&-)4ka)+I)+>ND>DTNqI5csmF!}=_Ufh$
zp*IEA`!qtPJ<OI|nhYQTs`hYcDGn1Y-J;XanvZqpzL={gr2nBBC2Zgnr%Ri@tN-Au
zd8xj7?VmbsY5bbl=O|<fBO`8mbQZ`q@r<kA&-M!(=>vVw6NZ?9iO2LH)QQ^nfc~@(
zNhd;4E4Q~gT`?Dv5K104F#V%V`>)GlAdEJQ7@CwB>A8v+BTZ-~!D8~Ie|v9C+bQrs
zmnY?D1W>@8x{4N?NCoO8H1K$BofRrHD;;#O+{;6`qfDO+UJvCgQy8QXrsIb)hdD<r
zZ0+d3YCkg~A&j{F$7GBh!oyv-ucNUJSd;6x&qrO{uPaB>->SCAIIw`C6YXvbHN$0c
z@a1&4D3awRF#JM&zqt&@g-V-#C=Fw$7CYFV==0S{Qz&As6d8o)ji8vUdCnqdlqqDs
znu2!~Lt9!~y$PGAt$Q8<t1W;sE437EczNpp4oq}zUScneVkZvRbMboVn5Y>^jbh)Z
z$IJdlLeu|K{aVcS<>h6(o&bz~53mSCrQ%b-$z3gK@pWWjc_5Up3^z#myVH|zBWSnt
zKFRcYTKc^!<oj>vSwtZ!gSAaYzEn;8>UKBBzrU<PgtFDN_hpylJ5PZn%F*o(%Rj39
z;HuKr`_F3&dOUSqwV@xOe+@8(pKByB;$DY4Bzftnh54Nqs;N3Wf>}Y>((<ix;tp=?
zsoryAjb3{?9X?rJE9ik!R2TMVcO4z^;yRh~l$w1RDpT)Y-{pLtq?fZ;g}?fyc0LSf
zg@lIf?HoX$mP9<z$*~s0w!XQrT*%r6x2vH7P62!--)FM5CMvFtu(w-tN-P(P0H&=9
zLi594)YN9^7K>Ft2661$OW3<{+6qo<idvLpjo!I(WjF2!bnd;t*w<u=8p1>ArXZZB
zOT&8&c0i`G9{&1!&qa_jnl^mM_w6H_Gco05TRs3iEh}P{xENg(6%`hM0*eZ7$^C=H
zajL`MJcHfdUWx?tg}#8(5qx8to=6AUDdbP5&lqk`78GKyn1tpV&-gqM^kB2;bN}&#
z3>NUEqa%3%&J!lHYK=fi-UpkPjme95clVNr#l_aVpaUy`DxLMpZ0jOv?KMxsbfurv
zY%zbg|M@u%Wc6U?<bj5S>8NeApUSm@6}Fx60XOJ}=pkHK?>Ov*RixKSf~Iifs|F?z
z5?~ey1qN#h_wvC_zI+1KIlOh{bkEh-c=dLZnhFoyGUes60Pw0v>V129hML_!?Y(he
zvj`2|s4hAiU;cLuIV*iE-+`tJ$i&yG?YTZtdfX}&e7nig(B!n;z<cOc{P#&76@Y+9
z`J(=3db*ZK2eXa)>2Pddpyalt#h3JobdFv)fL)JEY1dg2DBHZEUV&}iG?*=52VJ{a
z)*9|G+bDs+#k*Njc~$6DiuL5`vSiBHeA|aWrQEr<Hu$?xakla9X)t($wz%-P>J4U4
zzD9q0a+q4O3m}IBX5eR(r~HLc$SG%N54vbHcvo5JJKuI`X$p@dN`kLKri&D;b1r%O
zsaP8RF$=!(GW%LT5_){JzrSA*qSXFtu@&6(<rF_QHg<F1529U8b|rj2p{q=xu^KQ)
zSOE@U=(_gIT+y+V2wKou1mNZ{x`;!JlS#S)Xdy5qz3#}M(Z9`+>t!mEnBMy4?79+(
zJHr=XGZZtgB-?zR0=S9&-*G7co=WTh;P3%UaVJN_;c7kV8G64>$B$D?6J<;dqe>}9
zTWxOghMPr<B}O=F{Bo6P5r8OXFqSXMg%O?OQ~!d3Bcok3dDsWA5gc!n9LEg_1(37P
z&-`nPQRC3GX3{vCBnjU9>NAeTejD%)%!_wvG5^9$=e5uvasUMs-SGKO%CG?^q?2u&
zFZjU{h=8YmPN1|%N`at(nt4y=gBNiZcn`eEqwSK3!(}TA@y0{j>RlyCJ5_(8f81rO
z8^;#P+=sO3tjApX`cA9QAHFhdUb>_mO|2|__u_DDTLkbZJ-`G-ISA%Itg)M9;4>7%
zYfp`VHu=#=;u(+2&k*j072yN4_jJ`<zC*=GNElyLigI|mB)T@NUjqJ^Kcl<XfLEcD
zHtK^Nh2#@-ZH2H9S9<_(E*!y<)NI5sZYz{_%z`<15j~3VyF6fhxMc{G0m)fAB2%EJ
zPITsCWjT2?dC471tTs94FVbxB?oq~w<!btbWhGo7Cu!LHX(@R;TZ)NK1}4dKdZ<M=
zY{PV(q|x1C{ysiuoGxZaR?`5XsCiioW8u#kS!}gV(g1<YYv;)a?dN!@oAP=_Yqi$@
zRu~))t(SGUy14vBhYY^pqhTdvaJJVL6y24I4mY3S{yt*96Fv0ut?Xb8Ea3n-!4S|a
z9KtSq*b{XbPFH)Co{S&~$N<DND8=31+e)N8qQRb5-xAan)XZy4@BKBlUS`lJnfVTU
z%(n{L24$!OSgaa#s*NFrxD;Pn0?2F;`>(J{toIM#URm1Ojp4K(Y5KtgQQh2t<Ddxr
z1Df^%>Kv+OyULHL<>~puh^v<teCY_}(4K@v-VC3tY0{fd?1@gUzJK0nlX5^*v~Xct
z7FI@f7&$TTblf?z>NN<24VU*u%qzIMvWq2PPWfbFFClUOoF&TLHN!bF9<dzf3Fz<7
z3Q9Cqa})~yB$dWw%^xOa51l=D$Io+MttNhYoiMW9A*^uy=`+Ap>9c3}rgq&IQv8++
zlpt*`c$4w?u~=Flwbg7X1he(W9cu1x$!<wTM-u>ol&q|*3S+M$;H_uUu35cxnRaTb
z34x4*tqQx&oiwecGR~*`zXqyA#T>)ER#t$t#$UKFwgm80qPbq6DlWo1?TCFsAmGaJ
zMvwM8`rpbPyU*kqVn#AJYO*9qHhl+SE}gVoSk%tD0J$h0?ni*|*Xn9I%`DE&oUmM-
zPaa;<V5d(Uv7CH&DcwLvCX0UkKI=*=d_sW*%%8qEsFKF}g1}SGKaW>DYdm`ufe_Gh
z$oI&1>}=)8$P#z!4LfRYX^GrZ+PT2j>Pk(Ii;H^#6d0#Gii=aUsK9y-dFca5*~=}M
zVCx5M!TTZwJLAbL2i#LJtJ0k8)@oX{$D;_K>k|?Z`s2N#i&cayDq{Q03>J#suLrD_
za?M1aZEb00RgIrDVDa;|s6Dlw!Xg>tKMfcESrR~T?4^NS?r7R?Gv_wf%8BV;)&}3_
zz__n1v@NV72OTsJ_$j)d{mdNwRX+Vs>NDziAjYyvW(*n8uf}ZZCBA)c#qtH+VktlA
z^l`et!@|G?`}Tk5FwD<K=PWZ6n0`j@rx$0VD&fB3h%esn2WDRzHZ0Kz<-F1$ywxec
z{R22n%Of%kcFgVHThtDwGF92w0D>3Lv%Lzw=ZOQeBzw(~dEj~IX!>UZhhH2m3jU`!
zpmu$(%-HMwZ~?dacUp{b-k)jx9ys3<9HI2`tTojWf-MTny&?7@P)WOx-a1mUkc>jj
z)qu##_)<+5KdFYm$5i3~-uc5zmOZsK{~OI0Mb1AJ+3sz{z)^!)KBj7E=o6G+=?$pU
zaULDeuU<jUHB6Rlo-X*QP%eHiFZ;iKZ-x{8&zX9eFvN<++7*suM$2@c+2a7)8cbK(
zX=X1Ie1G#)0v=Rvv#d&LLovt(L;<XP10b=!2MqC)BbPsV>;Tsu6$6{6`7>ZB?Ats_
zV3?DO=0Iu%KM79QqXjTWxa5D=9E#Y_IlvB;`yV&T%f$tp{UKKqFJ^oG`{bN?$RDHE
z3!FbeS&7|mg_-jNqVNVcVn{#hP2ss9FSCCmzR8Gsc$uXlN$;HQ78T*6qsfXbChFD`
z0=RTV?#k<9)&Dth?Ui18rLj84z$e}ng#I@8(1hkelkm^>v01)hN|lf7^mzuz@YCXx
zrPo-Yn8A_Po>fly);STjW82O)Y|4jRy}ay6^8`nB*5BiHV=<vDaCn6LT1my;bsGal
zkbo4y-%H@!QQZQc5&_UxQpX-g?H&bMXONvJW-FXU_fVLGcjr`~{ejewAq3!eHRzKJ
z=SlA%v?VZhtNPCrVW$7z*DF1OQgAOXj+xDWlaW<Xv}$fBmTQj(C$1Qac`>L|g8U<8
z(z>xVR63zmI78Kubhk7aWtO<TD7nBz2pfL2hM@_b;e`k$>0;Cd7_!l=MW*n*Tw_kM
z3M?YX9@ymJT>qTz;NAWI{y|Nm+nv%OKW-l7FuaQX6fakzkFy;7gt0UXY8;wxvn@BL
zU;*Lw^2Q`!bliHp{vGH1+VS--qi(Zq&<HKYYD)71LT{2+2}A|lf()Tw^e34&xHl-L
z4c4O)J%xXrE5`7j6@+?=QGaNm0`CxDzvDLb9yoPcgC+TvHn9ac<B3+2+{D$FnI`?v
zpakAg&-_sFk<U{i{`8H>M_f`nR35%%yhw_Hp=#2%sVbgGtc;oHz)>RHZT0j)UoHMH
z)1<yZgAtN_;c#${oVCJD`v{C`p7g(}KY}p{rW^sseXI9`$4g_+E#%vZeDR(7T$Ew2
zIG@#HrrhyPl}%cyc#ezAxrKB~sRxEm%dj#Lf2;LxB*R}kXp%eE8!6{&CIPoa<R3j3
zjd;MkS!W~gXHcru62c)4K=getuj?(@@{}UyGp@91(@Zs`aCrH^Z5qS970&h{_mJNU
zYl6haw-?ZJlvNJ&a{W0vsqlTWV%u_~g*A8e?<vBnDczkZi79#?yT}8E*f{+9IJ<zE
z)WYRUEmD<GJiE&;pNeo65Ixg2?D8(Z!s+#(RqMb%lvH4yCPV+ltFZA7#~tEGr+ona
z!(f0Pvm7RO1-;4Q-XO&Ck)7C*DwXZ!4M<dY=s<KSm>^ftd7$cu-keZ3GzBmyLDNx`
zkSqg&A3Z;Ufh$cX<6INZam>dDEC$ks-~88(VVv?t%)b5;MgELGTMtAea7)W?0o~;V
zr_|oNJtQL(ErvFfGLOsnF^gO#PyHd`q^!Vp!WG#wc(YDq>6&f&sD&kFm?^kQB<p6I
zDbkUg*M=f~72CkX-Hd90?UEk*b6(T4gGkT2BzqG~X(6QLnQ4xqnl?z^zUHA~S8AY{
z!rujAbM8)F4&Hw=eb86=kq^}mwarR69=WriOMAOk%mLUn?_}>=Wog2lk$>^_sM$1(
zt>bIuhrQxEB#=+rzjx89iM(c95>q<!6C)IHVVvF9tka973RA0ovd3%8^>OOs8JV^N
zuEP9Zq82JCFF*J3-mS>#C1D%O+h*)|hXpz*V9WoW9cLX;F=wNrpv22a>+^UM7=-v~
z8iS*cRE9#zT|8zX(eu()Tx%_5QT+xqV(dbQe*nR2{4Ri#|HEh5=ls8jACv`350qh;
zT?=30zBG|-w|D(cP_BIg+P-f2D!C6{skG*32Tn`le863Oe;%ZnW10pkQzjIS509u3
zI{k!c<j&BO#kq0P5|HT9(eL{j=Y`9k7!&7vp;C!|fz=8*eeIRrXS)_f)b5NQaxfRU
z>%y2bT`JcddInr|zj#^*jMOy8{tJk!KMfdSfqv<uR%#OCDKF0OlgZXTo^wuxa?Wtf
zBdiyAaf(o0qy=Do&_SgFhC9WGvg?H0gY>o~q;eTAO{QzQ<?HGJIIh|^D9lv&W4NhA
zBXsBd2Qc6yd1D<?7`oR>jpE&%ZW+k-+UF2Ook3h@ZtC&@T(hKm8d;sL)>9Ai5frhM
z{3VVr!?FLhW9x)C4)DG!&9K1BsY`hUtaAP&<8vlP>5ZVH6y@TvMM$Y?MN+eiZaMV;
z+XcU*B^7cecr35UZmS-pOS5bjyElt%U-}pwk8<FXfZUE_Vf|NN=4jgSya?s;43L$0
ztoXVfF|?-!%g^5k>W)*t!@W;NrHL67OEL?5rGI-XKDPb{`HvUH1iu#>W!XAW9hl8e
z6L$(fi5+nrlY+Oj7!2YalQX+(weG{zUifWMvD8F<crk|TM!oT$Q`y-1AI3s=RM8Dw
zj)~7II~)4Zl<)ScWEc?erAd>QUxcY{#9|oAdKzAlZX1BJDRH=czdZ$1|G^4V3;pqd
z(6Fn!CBWuCE6g5d88Uy_IJ$FU!iDhN)230qiGN61dDpi=>YTb(&GT!WZ62Ah!c45a
zC1BJA>7*frH-??S4?A*Cx?=?L6nTmsG6~vzcRG?MS$fDnR%c&J_Bw{h2BS2H%IR3r
zgmd$j?sT4Dp&ontotS;n8#m2FQ&^HQ#h?=gD{^>>$8Z0Rv2AY|l01J99^FR~o3y?-
z=boyQ_d+<mC~ItvH`u~x>((ZE^13(tj>Jfk$^5#OTpuaoe7k_O^lje52fUq+BYC`{
zUrjZqyHoQKW5a=_>RUNIuV-)Y65ECFl-{3zx1;Y~CyJ!amnkk%_#*yPG>JNJ(cFI6
zp8pljW~{FfVZ}#dumn;jsI5F^7`4lZ&;3)5g)8BSOtTdLK40-OfW!!Pbiv2M7A9N|
z@0$5qTSM1`y{-*y(ooqXwV~~08#Z^qU)$_r4G0NlnnlW?gWph(9msaw%J>{w2)^p=
zQ3ZD2TeD)7MhdreqBS$pZ-o*`-hsvmyR*)MJo+4z0|a+}lk1323D<NSy>hPep~CM~
zioT?Myu_q9f4Qb;&YjcvKfoV2)y9fLs{noHVADqp84NI1T}yRlQ7kLO*~ImAm!K1^
zmh3@C3x0iaD2(kHa+2?(xx>e+^vqA{ff<2wHszK+YCMhw%xvzfZ-GEO{hquIm}aj!
zLJpBr^TS=@`r<jFH%3wKTkqj@O%Q^QME|3$@u`h;a^vjto8bNz?2`}`*=DbZ_(z$Y
znj+m`AlhFm2s9ZCOQ9H`j+L$1a-;g8KcL-YjakIjXR7^Q9oHJqblb;wFq>Mkd&#jO
zVNsbOqvbGJH|elKNHw;RL`Y2OZjLcT8E%r(QgUdd$a9M|Ioxyn<PZ-}wpd5inkSK^
z`~UBLo)^!H=hgqsb$zbi_4|FV@AvzA@&Ao{SC}WPX<JXyC;c+9qkkT9lvr?1wq-nL
ziIw>XEBg*42VK<X9Sle=Y#3h^*HQZU%-k`dz4HT6(^QUvfQ!RM7WU8Tgt_qFtw<6o
zZK8$c1P9$ywYhcY`6Z?cOPAcm@wQXfLUe`+DLn;ExRZk{z8@EiN^FJNUb#*dBxx`g
z$8MBH@K)58lDqV5YU(;RQKRxd?G`R-@*czF#+3>6JhLz7jmlIa^j|V&a$EcDGC?Mw
zwjyr_LEdIYVLd8GZ*eZta4xMw>Ur9lBZ7NO2Pks=ns$D5ovW`v<_>%HXOXfBUHMfj
z8G5%RfB#%YpI;ku<4@0`mF)f5p>-i~EvIqJA1w)$KvR%!UmLBpWst1f``lkf8H!wl
zF2D?eA%#CrKq%^ysjOisw)Sz<T;Dt?>?02n!-zmMY42U;%@bg&XVaS&Jxm}Dhj4Hl
zVzqxC*=#|rt|}YzB`=6wu#+1%g(KmYg1u;%2qVdYEWqlyTM0+eWJG>;rDrf&b$4(e
zm}yiP=a1{IYL8OHlO-q2#|@RNLmO6$ptvS)3ZwSBbWQ9Qk)yC3zXDl-U~3!jL-5ys
z27KIh@#glqSV^)gd+s7JuD2x>d=kU>R=8!US4pzMTYRzO)PxetMzdkkPotmX>{G{r
zq5RR0mJAqe5%vNxziY|tgsjUSngI7DO>+iaYdULx_e|7E{1%?T+yvboTipk~(|jAl
z9ZY?(JxU9bM|E#JxyLVSFA_Q=Y5e^}>BFHO+u*4}>kL#l>I44627$#CLVo&II$WqZ
zx^TWQPxrWxoMMqC?|shPF+<F>0jnreFL(_vgLh|(%-IO3fr@eyn~So9SWc8ynK%4v
zUy(>V)|h^j8~5g;`aHt3erz+`3HV7(hJH4PJ^8c~k32IwwCirjdizwzCNGeky%fFo
zb8m74usiVcKE`#)mD(z$d*gA^ctx9GNaJ8lPtoyl%oDS61S^DY$E}SI7hdJ}HaR89
zP3)6<D4|y@+%<+D4luIhCyF{B2R{LGSX~NjLCEw_W#zGsMft@h&Str;y_GTiIgb8p
zZ^6r&dK&8-AU!uuZKmj-T7GU&g-E9@+fDHrquJ=3W#W~Yw0W&p@RKv$WiwmOE%W@E
zLHyT5O}AceIYM8R$iA4WT<P{O1dJ2&&;|U9)HYb4j`;9+Q`@+pOE*i$Zry}hpst7U
zu*%az#h7Be>S0`%X_Jl=Ig=YdjlJN*l|~)rbnI^~yEBmFs0XkZ|68i$-UgC^^vJ{=
z>ER}I-$lzPR9_+sUyB*9tV51V8w+4zm~o>zzR+SDBRjLx;(`uV<6SjlGot&Y6^8YT
zXb&%guyioTUFgLN-QYCPo?W4`cIOG%5NamZe)_nTBOPGg7FxQwL1DZr7kf$@JhU=Z
z!m`0i$tfd=pM8uqpQzGDBr``Jzv|H9JQOT{{f#{k658cDAzk;03!8#sxL+r)2FBpl
z5T#$c+dFd0;5}zM4+?Ty6UK|5fjE1JSJ5JBk}k@JHq>G|wRb`zox~=@oEb0X<qqD~
z(MDz@MeY*=S?LYf>DvTY5odlus^m-7otzQcApcQLA0aE8f$fUi@xwSoekS$cQaegX
zw{SXkrJ=09qO<s?<nY~4t0fnF6kAi8@T$|mJedWcITb_J*2roo$9qTGzKmT%@Mf8~
zmSYRx0gMOSC#mz9az7rHqYA_XkqGdG&xZCZS}%W?tW&PGrGKa4CdKLJ3i_qdVL>1m
zwcpr!GC6rPN92}zY8LHR1G!#`+nBC~pH{SO<hx_axn0CUf$tHpXr1iu5TZd2Ro$%!
zuD|bd%H6<G;!tJ4UY><4#>Xdo=5g_d-jTin$}(y#DS3!nK0XIRlh8~q4>)igwZO^a
zy!A1ru4{xY>EJR&|EQ^-`+nZY4E{VqdiippeZu5Z&@<X(g&X+3!pg$+6MHi72Hg>f
zgql2v%!M4Wr`7G%{oA*{y`}sHD_QFPk)<dj=493_uj4Dy{WY~Uce@rH1QhjAn7pQ8
zeR)Yr<gG~o4MB!0VL9*T(9diG1{3X+E2e;{@9_?XrO_&8YrPX`*>5!a4Mlrt2VjfM
zx>n}tILT3!0E-aCT=korLhDSFQw2M2!9b%uJZcq`yg*)@-0$fLyxCzg>3V~%6~z!S
z?IEo1pFP+-GXQ%0G^r1|86$_1uewamoSDWO4;d;|W$Lfw=^2301yDmZNq>_1lw(Z0
z>Oz85xA<F8ogrkkX_v$sU^Czy1R}MwnrX$h&r6RaMWW?K^ozBc3_V{}CG=4tKCU<P
z-~RbOiIY%6Ll2sX(Tlf)Qq&bV63VCxMsw|=$yw2sn2)lU)zf9MK*I~v_#qwza41H-
zdzQd+O>BXej`dd?uHUALqb}`!&;AuqJnzgvusvoxP!pwkBGjSkddk<HwfV(?a4`dQ
zdC#)(EZg%y(BuCs>tsJRR9cN_(+zl+G!~(YP}5@rUWk1frR9%AUkT7m*2f!38}-t_
za-tb}d!%DFMCV;NBm`MGpMCs@*3Me|Z-bXqSXLq|u~4htAN_Y?;#H)ScXIYF=m+6F
zVMOweGtK|pY7cidqaZzyqLUf&l$*o0z*68o`9^3<p!>dW4RtRMKcw9Z{ezC&-%Q2z
zR*;iHs@JGUKFaP^KMkY_MASQM+FBjtd{VA+C^f^_rn6(UEu81ob(m*^3+35RR8@bd
z+fNBX{!S`2$GI)VY7}csv4#$y9K~o}DBF`#<oVf9yoH{C*chlO-z*}+3JkaDDZdsT
zD%iZu#k~)B;bH&1FDWProrf{yds0-pDu`-TIHM%=i-gM24l3a)LRa;NBeYsixuVN`
zUYgi|sX8m4B;y3-b$fmH3H|BoaA6o98htxv{Qk8u(D3&GL3!=?VK8|<?&GJj=mub_
zmPw48ZwxaqhULJBW&s}vkwCQGP9SY3+IbO)4mP$9)<C;Sa3Byo2O_EevmyK}Gc-8<
W|8Llx+Y1i#g1EaK*mI9UPx=dL>UA3c

literal 0
HcmV?d00001

diff --git a/docs_new/logos/apple.png b/docs_new/logos/apple.png
new file mode 100644
index 0000000000000000000000000000000000000000..345f6277e05770142af72b27d334f6d631782c88
GIT binary patch
literal 21938
zcmZ^LXFwBM^Y=+YlOm`HQdEjaub~)fKn10%2vS9)NmY6$DuRLvNbf<fh|-j5=w2=g
zgd`{m(k(zhs`Q@sgy;YD^$XeEQ>Of8=IqR@kIYRCI9T~v0RS8qFPys!00aEV@P~!|
z$B8&JOTTi{HPHp&T_W4g4Mun^bo0Vx69B?a1AvPGU<=;DO#u+70Kgwd0Ms)9;Juw&
ze@PQQKsp&2oCCDD#O@XVEJzp6=~{#g&W!{%+JDQ_nO|>GyLUwBLXXhhA9C$t?#aKZ
zuSure+Mek#jVw|}uLenteYyMwdz6)1(x<)feFG!QVIg6k^}Rh@?S=8KWLb*;b#z5-
zr<U8ox(#N=zt1Lo;VCeeJzMFd?@TC7I$G(KPP}X_lHm4NCb@um<=0P@`{#gp7}2%q
zahC^QRx)4K(OyhqOWnKTjktq{K<byJfl~sm0s3wuG9$#XP(|W_gNy*XS(-+wx~SUu
zk?(WG*mcF<F&CT;06aSn@$`V^kjpDX^!>4G2pkgNr()ec&8Fs6Mb-1K#0yM_yn7CS
zMA3}_Sq&GeSI^>u!J3!^ILuwW>efeR<Djk|r<>Z|=;<SXxuB6t!EBBq-{|=IC9tS&
zg#gw49ka>1l?JiZ<0D0`my%If4ez9i$?}!z4Y&6{5O}S=4mW>aV%%PTl<E<kM_K@s
zo!Cn3{%PIv#6Uq$6c2!Vo|r%ODcQ$+EW9SQ2#WwctoN&epOT~aY#zS21n48`Yy2#p
zN`Gw+W(C{;m?$f6Q|~ioE?MnA99ZAuq&780U8q0;D22{I*M$h>@S>`>24{>#3*qd4
zUM@FBoblw|k%;q<fb*YJ>YH}_efgd=&-LU1<nxd!<v?igEp0PqPX+iG)thyHxzo%3
z%}rGll$IlQ?4Pkbid(?+Q`>#%6~s<{j}R_Nt-g(b+ZD*>Cr-r3Q2kHB3wxRDb?UNp
z^H-{Z0U8Q*YUl@zP-?!loLMG$34j&yznizUZGYX0Mgs|m$6_=y)z-R6iF%U>Qr?Pu
z(n)h^Gt5io0}d5lh|Go&KYzKh?~4W9U*5vulP-C=`NT>WMv7PbN3nMdU@^#uO5D8Q
zqI2?<ha)81SLMvfPAjF-f8B}!6yqWf^JMtZ!G#L?#arx)rTep2?k`s}<4&m}kf%C-
zZ3fN1ZeR3e0%!m(K6~0IK1PnO!HiicV^{&MW+`RXp#&Xe{UYTW68!PHt)0{0sk)Pc
zXk`cH{u-12l-jl}TLMQWP+aD+lw)_5_){mVcvBGtzV-@l-bt^R8yG@ipZ#-|0b3?h
zaFUwDUl47^LuMa=yJ4FIvkRO<%8rDzFD4>EamLZ*gjxG-Zik*jAx}(jSo`MfF_#Uq
zrCb(sXsn7dI>c#feJ)n8*M1n72Zm2%sMLRkx`t$1pAnx#w$TJMg^}F=T&;81OM3F(
z;_1W$`YtucQ(6S}j)Lw40R-ZzvSv$Hlv8Gcn9@a-Q|SOFHHc}Uuqbx7hqUR?91nov
z^ryK=M(ng@M+AMc7zgFG&-x3x=)MxuF<PSScz{=vL4d1&PZG$B6@V@WaW?8Jj3qnU
zN+zAsZ|^M0U9nCIDD3t3oWMM2q1zX2%-*D+k>FZ%EaZt6z!F;ii%_KU?}bkb!%d-a
zN>n^pN)VKcb!fQ4AOUbowu;f6=}uGk8_q-ENi>-Fn_3>akHqa&G6RVhZGL4xTPX8~
z5v%~;-JGi9dPs5Q^M8+zBKKDsdexV>EUdUPvmKqt1*w^+zGk}qwjin!1)*hrS$cBE
zONxmRz`3TX-4CoUva<sIwldsZzwheNBqgV*x_owGiPMkkc8nnPLo?yuMIE6;zC8$9
z7k_86-@q8Kmq&6#79wO`xxPKd4e^Up-(1+~5Wk8V7DA48pT^=l3e3kk#oK#8h3j<p
zA(Zd95*GiyjmEjO;N<>LfP)67_)T?nh87&-{|`y0<Wzyls|glG<;SwPM!I|swd@vT
zuRkojQC}5sI{vO&RUMDS=ps#^E&A+;_l5xiaMP|#TVV{@Ib0BQu5KUSy!iCc+4}>?
z4!iFREZny?bTA9NXWQ&$>2Nr|=G_*3L^DbJ0bQ8o@)K9|-0vBk+rLPhEYULuJ?ir)
z!2RJ8rk(d$QD&Blk~MoK0>r!1b_QQvO0o0Z-?BG-vq}0{_N3Dx=!xRpNx22*+k2FQ
zfCA?l0CUQk{cfqpvRxz6&FYRRz}1K%nEu7F@da9JG!)l|fQiJ=xMSIWF7MBt)EF>h
zG5L0rna;iYQ;K*@k8?DoN)MupyVG>@5m)xS%pnEeawM?v4JfBvc5>GxbFv~J9xutH
z_;b)?Z)g03e#TtzzJII{&F0^IDPp6N?PGclyT#K*!d=j*sOSDD&VL^S?Qe~mPcw|s
zpc+>z4fn=hN6W&!*?vI9s4!8xyu@R9Le)!(4CrhRE*Qkpc($tcH(>ZQJ{HsS1qpas
z4}+cvfr(fdk=XQ5P0wt)1L*NFnviLWHroBTn8*k4y9G+ri99Fw!u-;#PwhB(r%C(9
zhVMI1cDl-ghI-1PZRWWT^$0+-o3vK0bFMyS<6)!k{G2YZ>RwLBS8y$T+a3pIYKU<4
z@M{r#Ff=vNJHPqvFHY|7+^Re#AJV0;t!-<H-l_ZGyAP(IEvFzJJ%MWg-w#JWsk7fp
zms6}BRvEa&Lg&XNd1r8Oj7NuuZZMI}>6i4}@87}Lve0P?HPAjCQ@dg9df@J86*CMG
z=l$XqXx@q%qRWr>srp3Y-y0e?ZqsqyA6@=#x56Z-TZOKv>NY)s>`Z*2>^v!U`VcDe
z#sT1-S|PAEG)d#zgATAdZko21m6@h=o-S4!-wRPt7&%w=g|W8b+;>DTr@J#T20Lbr
z1WZ2PhBF|ZPPY&*>Cq*7P)BvA?(;ewrS5UTE;_>1r0f0-TvMDzptV)zSvdMKyZ3J<
zES<k!(y=tv`p$h79Pgvd@yW#2VuYxyFe>$InI%bARvjChyjxn}qu8SZCBGM5@w26@
z*=6Gs+$Jo6T39L+TN|qG37r#N4TDs6n)j;KKEr*s7*6P2s!8ul(_TB#$?_Oe+Nt|S
zJGRf?s3qYu)j&Wit45X{0Nk4v`1XeW5M!gE(Uk?eoAT-jXr->AZTH#tywEHcQ@izi
zU%z4E&a%SQpGP*UxQ?AM0@yK`ytu0BV}tIrTyX~a?6vE$wA!t3GA|O!-BKXgz>YR%
z*+5kirtecjQ-(xL{BNH+!xLn4NMf#6dc%Ko%zbNi-0&^A{(mHTgi4(KuSYedGARED
zioVu%Jmh}w)4sG;^(UIy51$><OkY(xyH6!;cWiOYv_N$SMq>A-RlhoUcJBcDpbLrr
zlTAjXP4%VVy%Bo2a&J=cd&@OA)lV>?L#b~&pgsm}v~*q=`Ue|df}nJa>XzQe19UY&
zes4ex%3?0DDBmvuxJ_GSGgm&v>dss*m)(;P&>9gPi(~pa%KQ7qeF!~O5?^TY(Y48#
zhfZ`~c-_yIC)%$2D$6%}I8^!Q*x>lA;3)>^S~MEbRJKBs(=JV#vP{{Z4g&>Wn()2a
zO(<!&aCm7>c%*)Hr|iIK|9$Zs*Y?`*P2meI*jL1fgo$Iq)t$V4b@W&<cgH!MThG?6
zV1e!KB>Dq$_ej}*F^ji?O3wHVMY`k@=O8$iB-?c8mZU9nulBvB_<_59Zy7-2$l_y?
zo|8M_(SU^6Bf5<?G(u7Gi2qLy%LyMut?k)mgMAKi(N(%O(7QOsM4{f_H>BQoBJPBF
z!8~d^9qt8Ul?7Zj^D_jCeG^qbmB3QiUfyX%ci-+6FV6KiRg6z{^PnINd6r#I_1s_Q
z3ucJ#Ppx>-D;AT!<v>1!oco~-S_MY(IAVJAqA6=xl1xxm;ojLp@cKxl^$A|o95mPs
zUIAeHTKtm$J;KfPr>(OTK4{LbilJ}SqG+wlBI~c)yf5ZMf~W7Lhe{X}*|JxsnBZD<
znh$E;WWt|6J#n01MaNFOqefeFLICx`zH}rs6Pm9!=Pt)!x1o7EESu{eC}iAwcK?eg
zoeo8>%!ZgA<hol<qZCs-bgfCJwiED-4&<o)Xpt-^ajz&Rt!=6J7}QGpyUVIa^m=#0
z<#>^h=WAoMIcAUAlar_DK#Ok|!;yk6YKt+%pZ~@wlf{d^#%>Z|R!ccYSC~Zl2s_PW
z)ob+Deh8z?i?~0xS|)3X(TVswD}1jgEA6ab-2N_AVSQYBDJ$bBf(1Lov&f(kgURwh
zt_U4ticb#drbNh&9YVHGC}E4Y-CF&vP1#KrQz9(rE;imHvMu_`j@-#3bOPT89o9XP
zoAdjNEZhZ$#GA^Ie`bj>#gk8f;yO399t%g{#e*FI`}9@TaL_n{ZjHL^(>GEp;$H0O
z=UDUz8jX2?Z36zT9MK_5guV_RPMsjM?e^xF1ZNH?Y~~@wK+8Rwo746`%k6UXY>U3X
zBh0+7NtbtNEd0ORFj7@?SV$z0&h>z?uDFraL?tBz%WBPQ7N%nH0`mHQSj+$eOJL2p
z(;E!OlCna5_k@_B`<bYG_cIz>H2%X*hAyaq?+67hic8Bdoe=D%*Qt9^4dUCA7ZuHY
z%8|Hwo9hg_0ii((J1+p}dM*0mL63Y(=V+D?eK4jhSL{KL+sJ0bQ=*KnIpZ3@YL$uK
z)6U6#vCOLo^~VhYl~kF>gA~n=rvrTcESSn^aY64xXs3@`_hfG72-<<&B0&t80|lMl
z<<C2Y>=uc(clC~QkgEe874qhHeVE@cP_ahh)cJ31ETpkjKH87m$2}i9-gTw6RQfs`
zVz}zU3BG>*BXSM*YB}Fdl+NR~%6OK^nM$vHWLQwTL^59uY<jG9``^R=kwIXOmz}y7
z8lr44p-<P70?J(cmCKbjoxq>kz}z6EGLvi6$w~X!Si;>cX#A^HF+XYEa+G(W`s`mH
ziGkvRiF1pmdzT2aH|Q?QJdpVC2Zv8b_;M26jarP|KGf(Frk!z|od&_v4Ze1+VzfEw
zfgNgjD>IwoVz3{6)UE5+4&-Jej^8>?cPLFV4l`BJ+Misb+jKG!xUJU<_Mm>bAwooc
z8VM!Ih>aXhPaHo@;;ex0M65XV%%ns&Y5Gw5qX4SaCZ)0Xr!E4l*1DIG*ZxFk)_Dem
zNE-ojowJ_)MP}6RvtAp_aGz%Z-+swlC)F)QXO$D+_(s>;6O`MQQF-zcp(Y#<;M!<F
zrRprbb&cJGX9n)QtA1B&JE7%5TV0g|<&@0XUpCQ;CLpuH`l#lshNTwF+1`6dkXiTo
zs7Ari(zo1<6C7-;z%lAgz^&oHO-r)c#(ntuLt$^FU-afeX8v3T0o2)uPeD3x(MZLn
zaf4|6=6q&lQD~X<-QY1X(f_hTUkn_%<)bH=xb<<TWe0;+ISC){MJ38*@L|++I#&4)
zLI~Qk;^LMQcBFo*@WZEJ+yL$8V@jSV|H}+*ECczYc1OL1D4op;hib9n-27alQifCx
z1LeUV-51S%Mmcm=R91<nBC*Sw)^7`GUuQFfvl(qV4?{eiwXWg%^w)_@Ebu3kY(q&^
z|0~MZJ1H6>L!~y_r48_-;F>e8@C0^wFWty5>O?;aB-lm=)#KssXCOy(Aq)oSK$?OG
zuUr<CRn(||>~aK@tY^(n9T&mb>=92Eyv*6y(h8NRn@k_hv{f=1rn|J*{NcVBZ(Of`
z^gPL)AXew{JPB^iO_AiwT}4N<KKKtEs+>6SdAJ~a$$gTAr|Vxx>Odu58cd!fTT2pM
zH`ys7%R3DzN{UC&!oQciK28j1X1rPoab-CK$=#IAeo*PGt~9$!+v}R4MLX3$K!N}7
zKic>$9B8k(olc6kRyA_|E0cA@TGOa;uOP+w^{LJuGHXR+o1anyu9I?&yJn5arN*Sk
z#>5!oj{l4|s1#|WC4fihJig?8(u74TBk1cib}nFqk0@#XUYZryTL?{zqK`6WbIB`B
zxn!+X;^J&cO41YjmaAFdhL-!i6mWv_J5=VkavD6NJKv6VdCg9IE0ZN`t(@TOp5Be`
z7nezSP}MU-yJb@(K284b6dLk<y&yX`^HC_K>ios?vMC&urhT3D*5abQ*6}y<^FJ+z
zoFJh>B{*G6^(}z8FVW?kWw)QTwzM<hafhF^R9mI{S%(I33iW|z9`BNIJ_Q{rlIU`;
zL$Zuqnq^XXQI-7Bny})A7UP^Ogi7$YIzbd9l>q#ola3{>Wu@s?m6W92(n_}|zDs3G
zeka0rsXvEP{$|G<j%Y4VDXYAyO49coD6%?!#sB(zmli=Ic=EhSIlxO<c5Lyb)hJ25
zsML?;`SabOTPsv(eSB}Y>=@oLooG2AW$RYLUh*weC(YK+$fvU`)LPsem}|FH-RL7S
zS<4GJPd^h}{yv;->u>8rTINqu<xe|ad8L-ZUt{*`-SD#<|Fb@w#v^(N?7OmDk^{~9
z^;ha&V;_<ULAB2LjsDB?>Qzxe-3JF0MPvT*^!gi>QvwO2nt4c^llAE#=NYdwQl(cr
z*~`+0!a>mPpWq`I_m-xOD4lL_=DQ@*bDtn*H5sod&B_Sodec7B^Kjdxw)fXgW_bEr
z%XMAh&yVHHw}+$M0xzu$+$6|kOvE>yf}_jO61L8-W>e>KqI}<p)azJHcG*qej6)oy
zEOclONat*f3*zU8v!wm*l#|(SNC5ZAxIUf%F%6gPE{BI-!~Ok?<jCg<12x2zPQL-^
zZRhnbUXNuHBng_l<9g@-`0nprGn=ecIW)t`{`0y6*@Ym%OZXtkR(<1Ow+4jA{C6wM
z_3?XlnKMd8+^m4f_2=v}tttck1gGXOKO=HyD8Y!kd!p*jj9}*d%5(`ES3NStnjq18
zt`OiCV%=blu*rkaH5P3)`75^|!Tz1`vwq7hPLl4?_b$E|6XFC$l<NE(=Hgb!TLbdj
zZk|WdcGvvKWIn?1&4(Udg^xRR2gY0?85Kk`sHHpYwf2l)`6Qfw!`h0A<F6r^!bj+^
zaVE(;8}V~L*m3x|L@I`$^lGBinCAf!r#7S<H<|SK@VU08%ql<qa=$VxhfFrg?_7}Y
zW?R`*)5T4rs0`vxxCK0Y&5Rc}YIw3ff9=qbg%P#d{{Fi7!+_{QCxRm^MuoIyx0gCL
zv2f2EeC=Pth#0sU=~ge^q;~HJ_rEgWKDsqLq`J^~gHJC0OQ=vw$LkZCrJy+HfxqL^
zu6e#w@v0%GUQJYM%02utB(%^OUYRXq=;cc8r1*!{pDzT(_gDP`Mtyz#hDNe`mu$D6
z8F@?y=ANy}6O#IswWF#-N`c#2^k2p(eR4aZ%l)x>N9@(aVGWd_^R<JN<bhGD{X};s
zBL>5p$p*(!4sMzstkobSeVkawhy}Af|Dj%GJDPKUNBT{*!P4&?^$J!!ggHmTWJiRu
zz=7vK<~w|44siF`U{pW39@YK&?>{^0I^;IEW9shBncYp?tEXunyK_@|{qBn0MOvTP
zp1eOmqS}%IsPQ`z)CMO5ggLqIyYxxQyv@s)6U2(898J-OwKRc2#ZMDl{cHW-BR6lD
zcCzbAfUcVto_4&dx+`S3xBGj1r`<bd$oZvU+Io1{$%_shX3O`Dxnhv`jdarX(t`CR
zuCzrv(@y!Nhe9?s+6!4P{LSmneb8yj9vc@3HRWQ({o+qEqq>KuN*9S+O;9mJWoNe1
zlyz#Rj5y1rlb<P>W}STnaDUd{j5xSj_OkLLH8nRkj^XnuHfmC%j>ZP}&1To}uMyIT
zbYTeQOfF;2?Wx@MG$r54I{h%#M(xvN$okT&jCy}p?FyLD<NhGGsTXwg?32dKv2kko
zC7*}w{^v#u${2JD+>oZ^{<#C@Fnv+GxiEX0$se)OHo56In%B;#t7AfvU#e*oVqwJ2
zc&E_1nw^Skto&@EoTVx$>D@m!>%62`n7sT<yVaB~L8kf!l&+ITorEq1%TYQppLUU9
zn|w7lM+?r{NuEVLp{V&g6>Wbyzx;%(=7%c{O`$H$>}V#kIccX8-(j?8Sniz;5h>{8
zV+Gl_1|oz{b_(e#^b^4P!Q2M&Y&Pd4j{Oj$Nr1n0z5q8+Zu_-qto^fmT&DH?j#Nv>
zk$WK{T^P9q;sxEFXCgb*J8t<`LC=Nz%Y_l2cRQz6KlQX@Fwte3cCAUX1?w)HYM-9J
zEE>1n^f_KZxdfCyW$p;gB;3^w!$gxG5lYgj+Ug6WJBF@<<5@dyrX@r7V~zsShS0Xo
z_AVc8W{H@a+zIo`Vt(gFGs;vny@O2lWY;CZRK<-=JMH|!dycrGiJjbBqBQM^^Kyj3
zLZ=gdu3lzZ!K4oQJOX<?a00b?GFbf8-Fc$Czl}rbIH4?C)Zb43mNV<*=Llt9X)u)_
zyVUC?%*5~Gx|iMSr~BZ9Y@_o*^5+4?JpPP(TQ{{Y#l2#Z0Q@(uHD@Od5{!?SbQ3V5
zyw0YtDbJ-F#xn)Szg#n1l%FSV-sz+@NdlXsAIY0O>AVcGixyD@V-xw8Pi$Laqzs&`
zpIo>#oXL50J6n*kwe>OCtg7mm`)iHmp&emL>STx&B^lGj5h+=iwOJ5>t~v@Dzx{|%
zz5e!hLG2XzDs+83I|p4MwhwkqW+xop4u4A8Wd)<|DLel%3wQ@#CwGJrYHok3i8@GX
z-g{tvpm#YfqAq?L!3}ob-EPh`8(1;8p_#tyuU~E!Je2qkI>)sI`Y=y!sfpM|HBL$J
zT}9l@MA7`rZ$F>CXC;Fxsml+luB;IoDxb&VpB0UxH*eMWkY0kR?O>W&LsS+rCw<1x
zT8=m;o5Oic^rq73@P>kuX~EV{DOXr=VPOWQcc@B+zaH2n&-i)I`wR|Fs1GP=xKc}H
zygY6?X(yrKn+*EYb<(S{?JMdjerL;xWrQYOD|VTDld90#&ZpM8@4tSzxOpyzXOJ6s
zzq@2g)OW&<`y+$L*M@WAmM?r<pH^dban^SWP$f@Ve>!}uCfMpA1qILXQ{CdrEqo#@
zQsZfwx#CN94&7Cuit^4ur`;@zoP*#QM<!pGFaHo<@*!(QPgmL9lcUat+SymQVI3eM
zDzB*>wC3cdCpuRm@%3xnzNG<31zygH#o<e%ZFSjdzHHS0?A$k)9n1U39GegMaDN6A
zqC2*nfMc^Mk(n5J6^C@3f7)qQ_3#+DhSyjs*R*bKt7TXWVw-Tov@ddEE4-fBRvK*{
z9D3qC>M>{cEb>2qt&n{tibnTKg05M6j84fT1WtRpJF_$^L8(DA&Wp3Pg`G^BpdCSh
z-GIo0oo5)a1MHM(7rQnD9-B}4KzJt@=W*nRXKmM9Y&2kUZ@0@5?YdLE22OQe!k~!{
z5jccE>iyp25^mt){-%N&&R0r7_HZ2^z(`a74dC5XnDGuFcRvl21{^`AqbaXhanJeF
zmaA$VSg}Wjb9a~Y?r>I+bHJx@*EStsQhC*<Lmhh5O{`pp!tN1Vk+o>QTR0;&Ibb3O
zGw@LdWP459rMJbabN5l4Fz8RyX`s3lFPQ4)*MP+Drc+)HTufxZ!_(3A(>x1%xIk&p
z*F7<S^<}_6P_MKKTT)}hZhNIkSH|bFV2N|j9{6V=@mgz?(6wDu&*>qz0lQ+DNsA^*
z^1Y!UdQ@#)2zHq#7_riksyd6~XThKs@o^WSJ;V@uoE7&pLgRYh43NNapw+3|g$%`e
zRiNJQ@@T~;g!wV*WKXvRQ0`C4wfq{cfH1dpqrUC-SzrGxWqZnP!hVYnFjo2WbztU3
z4qlr;I{vlG99dBUtnM#LlUk%e_Kin&xpONDfYl>GCM|&*JgIe-BWtY40amY>fC)oJ
ztSARjY`}(s#QzBD>b5I^sos}<7i+(i0ExRJN*Fb|I|(W8aE0$zV(-E~P6k1T(6W0y
z-(=-9Fnl4p4bQ^8TXkj6#Qa`&Mgv1P%e7mpZTCtnfXp+tsuvCn*aJ=3$Ne&pcpV<{
z=D<}>B>tcLPSw!*KwqmggYko%(+~=VUN3V~+POg_S|YludFvo}S!=zQs#_LTC1uCh
z0}I9@TO8n}nw0c}9V6B^Qtjk7H&EOm(^_ix0!%%jb@Q(90;!w*n)w<zG*y=t)*s(@
z>Ydh=0Dj)a&kE<qna@l$I~s*o1c55TL*u%Pt&OwUY)TgL2>iVh9etrSS-$3@M9@#n
zg0x6^1ekixAG7Z}uyTkXMVvVaEc(=^JIA<zZ*7Pi$vqN<GaQnH=X2Ml5WaN~Fi|>N
zaUGb4rc|2Ddy<VZ?r^pyveUrVe}7@5jig|LY%}6qBEX;S>y#w%l@M#0f8cY+A^N^{
zOig+*$yg!YUL^ZZp9CtssU)1smTHrp4_Rsp0cw^#YCTr+5X35G$HH8f4Q70LdPhOk
zg`3waHI0Dyz&0n=TQ+0D-{CkLjxve1If>&Rl1ZE>|JY=QSp%~v%Z58&VXiNxb3lpn
z>uYc<%eL}y5Ho&lzOyfDFf&7H?{KRKLAk`!A9VQf<uA`%20NPJ9(TZnXI{zM{L2Bj
z=kkc1@TlS%#E$|8SaG*qMgBNWkUCqf0vbPzReIJlZ;70DKpsmbqK!3BtvQdo{~0zP
z;lq{NW!qOPfksldS9&<}+Tr%5`aL5k*UW_J%`dGB%;sPDaC!dL(<iR715)snr=I^+
zArdgCw{}s!8_c-6$aR`st!MX(yUa%ySJan>9`|&Cp2rJ&?Rkg|r=Wg#k-M;0Avt;i
zm*+>Ajs_w3ZT$m8nQ0<qz)qBe`)s_zY+BZOuPk0~)B-x49)tYAVLjw2Nre>h(}@HG
z^MXA+LCDChfnrQs^t&}s`b?uLwT}sRN`4%LsSB0Ao-2kc-x*!xgeBu;f4h%r%N+4l
z42zpy6-@}Cs^xEjJ;Og+U~1oXe2cwr5_}y!f)P<up?SwEq7;-{RCiEVa_3~(vAu-L
zAVNjJ-5+@PSZZgWmi;_OOIhGFs=68tfuGFpK16sI=-a=?hb!EC`ST<$TSFtK+HXT%
z53$%{eZ2u?ezR(sAeeffEV4I99?>Y;zp=-OebSUE>UVx-thR}<RgZx7^b7|cGh2e$
zZbN@R&e<b3?!krlH82g;ZX3P}?5#`$m9y2_phj8aB<0au`1YC>8y>BL-a(mPk)KGi
z`$k5&Uq>t^dzI`WieENVMd!OO2v>U^aBLsWxe*3{%1VEuct!2D&|uAhMR6!wkLj6#
ziru#8hIDYS3js&eprvW)*N^@-AAHB~_QE{TBMbpl(2{Y~f1e8)Yq)}DXNcRN{LW^=
zi!*V(TYR`n1dX$+DBv+cp3Qlc5P&TkW7s$Xy|s$JZmo3R;@B4Q$KvD9oS-;uY|=jK
za`k<Wxx=`*Xz-Pkxb!H1cj2v#25QlTKH*CWWg<r8AgwG9aZMw*VOs~(C`vC-UZ2-`
zZ|T6aaU71RQB;YWy!Y#b15DKfbsh~v0To3t=J6#F_`=SLBkySCOvCKqzcI5Z0`ZX#
z%WRNmpxUt}jY(Pw0hE>?GLXq<M(aW2Wy&Yc6JBk^4P^7-Xi89gY-TQXnwMiy-tFwI
z9m53=Rt^BEg?F>hUcqoy_Wn>0tRo?c+v{kGE_{2oim10Y=H1&w;{d+qqbpU@GQZ7x
zuV=@E&c(64U`bdo*x|z+_p5P0VB59w@|N8R2MGg=t$_oUEdc)|pCU(S@nRksCIU_c
zxc423%PUQb<~E;6cC6KW_f{AaII61HFutaAA_T2Ab)4=hz7#nZ_#LE?auHKqk=C|j
z0AJ}kX-{}t$~@M{hr7u^0~X(N1?@+FwuGX={bg)lj|M0%&X17LkzYQb?8)@w7$olK
z+JGM?jg*ZzmdjjKKZ?Mfj}$ME+?G^!WBSoV_mne_Bg%KgKR@OJV>Co@xo4$8Q3g#p
zTM>1;6yh*GOkv^8zl&p!p-&htxD>_$jw)*}bclXCllGAU1U(>`&x%}VZ`f%h37}mT
z2h4RxFiL6`)Bt{a=)XuShk(?Sj$crk>8sD-puBT!iv^@D=LlMr9r5r&ygggqX8?+G
zXR171EoM^r8yFTtBN=dAp;FA2-W$HN?5!UOO^*_JnQ%dw&*uCMR!vKi(SUXX?&=ig
z7jZO>rcVZsJC5)@e+A4XuXkFNggQORL2NvvQ;tE??7v;{{L0??Z>Ut;DmRcC|LK)%
zRTAd+BpY$&BLvAyW8#f($M*!Tf`D-bEk>N-=4}-<no2GBt}*jR5?FL}VqOl4f6vqc
z8P36xjJU4%T0^e&NJm|Ls&+5{c%{?M37R&mDPI2_ZjGa>L*MLa#opo=Jg1#s4v3wW
z2NwUWFTGaJoosajPnDtC`RXQgX6`&k6(HWL<j|l0o9dO%Jagv_Di86!50c~(@m#Kr
z*2;$CoRNjIF`J>0ra_KIpbf3cHmwLOT9*gTQ2MC>%&Mr{xsXf4sZPZ}{yk;~0I)}Q
zgfhO)#4uMfytOa?b`uoi!mC`0zRjdEE1?#(wU}_MnOVX8-Ww+`Al_a(NRPg_ii!EH
zyZ{;-uC405HVQjElIGj@w|e_J5UilU&aNu|2v;c(=9}mBq90z4kz`jj|L*aCg>@E3
zV7TyMC)nPsDY`AhDxb_s5AJ<=dHyE$X3em1=q3!GsiTipkMC{U0>PJXx}Vu(%b$p?
zU2EhuWr~ZU7BlYhkOuF!&6RC5>NXzo@=T*-v@M#(iW6NG0EXT++ScnVTMj@bLoo~<
zCe~U)^$<N)3!Or6=Hc*DV@LGH*Z>;<sYA33KE0`#tH5*oQX#_J$!(HLcqNkQhe9&E
zxUwuMyz&a1v2X^dnRC$<Ged`Qu<5RA_NSNoZi@~Y@W7_BDhmJJwi@cTr|J<tAQpIK
zH2H+$N)x!ImAJP2_XFsO(ipc9-Z^_6dQ7_CwQ&>J+lDS|-Ah;j7~Tp}?M69G!Th-4
zfSnyk{Yv4W?aG2t*!+QemYqxE(Tiro?LJ_{UW}Bd9Qa8J+b)ZG_n2aS2Vt%!?_6el
z_$<v8tihOLewpUIu@H^8jVlHcFYj!=<l3?Y>|NT=k@(c+<Y~Q-hS531ca@9;R@~t*
zIZx;8V_O=iw*`1O*jHeWu4HrlN36Ind<Eao*6h2`rH(*EU2LJjhvr~b79RDcUgn;P
zxyud0;}75tZ{^R>sNNsKL9BBG4=+3!qJ>4#m)Lg;5Z}pg;(^GqMV+@uM|oj{x%w6V
z=Um(Xdhpkw7g?68@LE86e853Rkqg8ZiqBA^6xp4b9L=D=dFlV?XEdI%wOj;=4``OZ
z$7tQNzAavvodhJ(n5%rk{!>VT0R%c8!wp!>)@emL@*)uCN)_XEda#yYt|smh&&5-E
zvpIUr%aZF55L43dbS^!{q>6f*PlpJmcTr~%*(!zusU*rdg+-*J+h1TVl{g_K%w7$q
zI@W$6@D@=M$kx&9_0z3^OSljqkv#vn8s+!{4TcUt%n`oBwkUv)ijurHnzMem6=iLp
z2~s~ASxch-Q|RHt{j|0{15!IAV`=)_U`TEea~GJGB5|->76j=#tu`Sito$6@=7bpf
zkEvBVd;1(M$+%nZUj+VCk=p}f?lx%as&s9N@E+R}xB|Q+j{(E+M3H^&uK3;kpBM+S
zL}K9j2`<%+K;ntcBfKJG>*Vxdb|9vxn88x0$hZ&E-#gvhur!0K$JjFLD%J@gw9>yZ
zqOg+2#L8Nr>dNH!`Y|^f5WDl}XUiqV){prp7VNEDv3vSQK@JNe?pbC^3v1yW*nnn!
zP;F977&hk>3yjgW=gMYw?<;kt0EEP&C;k~+{AiTIARQ@>Nh5mh76?ZO+{z8lFnaw4
z?<2=^mrg-eO3Q)5JB+PNA_zS9UcOaKPchQ*nkDxE@bdO+7Je&GD$NB9e<>ZJ?~)@)
zxX1!~VE;^JcXR3eX2q>CAhD8mvPeja1ekiErp@M<Y@AJOy9@h^iriis^PYX-3i=eG
z)L7N{_XMoLcLc0XUIXs#2n8EKkos?ph`Tf5ZTcDbKI5|v++pY_qW|7re!~T;h30B%
z3HW5j*2~t0a4?P$7s~^hX(&9bN&>8~2}w(jK08(c&d^Y@WpSmGICQ?}JF@jW4iW@`
zZGIv6A^y<bU**=qthFL>6cT%gME-Sx;cEvU?ke4twMIff@WO!+aNs+}R*&ZZ&oGR_
ze_(8VL;pE{ME52eFrxDZ3Bn>wNGW=-nur}4F7WxE<k<j+97rcS!O@E+xagx{z?Haf
za1PSp7$XE}{aOL8>2krnZvGD~ztOt;Qv8U7WnvtEm<7WVxB@Q&c;Lp2ZDRwluhl%_
z=J}o}B&^ZHn&RM5z~*vM7q*)g+tRfND~lX#IG1*S{b|xwqDPn7<6~fPw<U_P^@|fE
zBma=@3pU)({hvqf@<D}B84oGfgSX6ZF5DBaPKDAH3X8ixkgX?RJ)d&;f23TZlLFhB
z%p;DR>$z}zx_dT&1&Pzk$rGMFbXL?2@X-Ba$Po@2onkH|c%Y_BpTWaR?*)VY_kThM
z74cZ8=w0J|jBRC|M#4v{=&SUhc1$cdBy8#d+(}}%SYD98La&tWvf>`ZKz^)|?jZ;T
z3l_jV%6<seQ1MZd58+afIJm-o`>h0B`FMI~<o|ApeU{x{y%*XCFE|@Rul+|c;w~Vd
z+~0yJZaR$@7$7thP_R!2VuHSEybJ>VfmG1J+<Jb7Lmqs!0Ni6D`wG%Yh{u0s#EAiV
zl^Cu8`E&(;tiTE+6%AIUf#wZ9_yf*`dlZ69Sq8W)yj}kQa=>u}%AO7w?y)cQ)%Hn+
zC4kfi51`$(3eu-}I7-sRuYhWWfo>g23Sr)cY>kBCyYF%yRvU3!e8TrW@_?%U<EB+v
z7_I{s!yzHxa0NyHsz0$#|Idl=&6<KIvY?0+fxCR|Ji`+iP{c<6X+r-=*X6;I1iHBR
z=mzn%^}_!&Pj>=LvrtOF{H&3)q_A>1$P{4!hEB4F;Lnh&EFkqibrE+vgu*43|5K~Q
zi}P?67J+REbiUvUGeVa$v(76Sus_K{WKQ3~!U<ezC=WXXR|1UQvFCA2JV1_t0WYJ<
zwU0r!oBK|rz~mf#@2Ebqh#q}+HG69_mM2oc3(0`joJG1{Ln!F7!yTc(8&5F-!<$DA
zDZc?Bm)M~<pX58m0C^9ry3og~cas(O7ef48?cx6|?|Desg9&fIMW6eR=tU((qlZY`
zvnlfB`5V0t`2d|L&)GpcmM4<H3&nspu08s{dCdO<S}oC0n2rq}9%9kHiSN!z&>h>?
zM0>y?3G$(Rib(_Y2QWDOu-1ja$;jpAz6U56IZwaW-pvggTygbBfmWy~H%RR|Xl=oQ
zk_9qIV6J-B{s5inr!q*oA1&1SKM_#6^o~JSNFQ!&uy*7g-Bb+_^pUYaq+fc+t|1m5
zO8|+yLO#)$yGKEVb2jvK#}fbwDi$uQzp!G_dY1>7v7&HatmTGSVM!XhD=4yW5dN9x
zad3_q3kr9b|HdJR0POMOQumr?HmK43uu~F&J)o75_C+>!j|?>P27Up2jAciV-c&R~
zVQ2)#kk_K;dT7YjKq$64a7-#@j~AFdMBy&msD~f;0PT0xznX;(C`y#IJ_|e;<#RCu
zF_r&h;hwNcuP8E=@KkMws+*Pl0fv@;`Y4fl^s|&#14Q||-{*T;kV2v7gg|QFyLqOb
zG31XMcY(PLaU8p^gD7pGebH0o_{8=V-h1O9e*IEYG8+ORQD;?YSoE;40u8E0*jaFw
zZKW0t(a(%Skm!aZqe=}X6|8{1P2Fl$%6=HpIeA2Pj13oe4vFoYEBF>eH?exY)I_!$
zx})`%o=aaDzQ~7rmXrH!sq)yC5=!Bd2mmkZu4#+JG#>?jWn6&NkDJ!l_I<W1K~R)_
zcb%0H$GYJRo9|#VLsTI!N4)SW=9-+Ft1}YJ?c@`Ihl7P?uS_4sl^`a^^l$_)KQJ?a
z6<%ouZ#lSdZ#JGYqa+~`3Z7K8&`%DTD7t<)yz)^o3Tpl(39-(GOplmwhVRt`+@2l_
zgOMDa$jXF^zor9uU0?;l5Wt4bzxGs{WWhW<UZlO#f65CV;56Vl;z5LP0t+1qBc+*U
ztUpQWlYd*9b8q~MfNwT#bB~X--E#$Des%7kIH`T<s<6y^@RkK&zctDW>;q^dU}9+B
zGIhdT<b`+I8c!s@y?y$nktVR{5*-kN$Eg-R+^x`SXMjbw&1SGL{A7hb|7fJ5^nv|d
zW%7OZf9y7jYp~oH%>gM#mmi0KVtja|wJY=$8Ezv=hj502LNwtR^5)Kf>@NiVS#z!m
zH&_lTIf2BA$>tX9hg)-<@X5Df%sq+)I=0%BX<=aT^?hl9RX6A+E%TK`1M^FwOS|d?
zQPc?9%EA^iNDK5jXb!sbym7!{z^pPA>F7hZ^e)TNg(r74CbwT1x$PkLm_Vwv*;@BY
zR$N{9*RsQ)xZ17ilEa&^TfuTaDBf`Vpk=rGpO8U&wvFX)RSnGu?8_%)G=aoUj~{;~
znfb2PU2*6+fjf^DzQ4i_q~`m)a<T%|e=^D<y1PDD>%zRiFH2*$^5WhD!E%efb`@5@
z992M(W5ub}>lOe!kLa)0f%Wou7>6CI{7lO!ok1(Vd^HPPR#}0^&+oY&Og~h#CLuVU
zq*T5}{y0fDvi5&5T5a6>;mWlv*CvknXW;6ug~lYx%AW-U_GoDeAEEY-_&@_cHQlRR
z2bc?t42bhuZF6gcuh!8$g?Qd(q`uUC03dzYyb>z~h7TUY)elIGRV+=f1{M77wZ3>A
z;9aF6Z1H4=y{3{?B=DgA<7K~&{K4M@aD&!osBR|oBd1!MzmLiS`}fZ?E$epay)YY8
zm{8^G+8@Y-f3>xK4v8ZSqIJqfFRt$Kv`D+kM*^J9pv)2fn_6yQ776g=<_y%EK)t5u
z(Mwhz`+p-I0R`Tas<ZROzu8-J?Ov@Sip$#l94t$3B7VSk->*Ymgfm9{@4#C&*hDlW
z#OwUoBiGR2<lg*`#>;8FOt`3enQd47CRgAZDx&~u)K0g|_fA=x+edB<mAG2Ta$-r|
znLi152gcUYMta^Z-1a5fDs7uveQTwT*(D<V>pVy@o230gqk8hZtDdSd<7SiJ5t<s~
z#`?Y4Gdy9&^-FS!A6->{TyO7Lj2_0nphVLKO7SFw&_yNbIO?o=H&_`rEaLz!1&mYM
zW$aMe*l8o&Q*GKuk5D~NRI5PhZDITZ8kkKg%5s2-e8wrt$YWVJs?!L!#|<)Wxq(ch
z1GmqG9;|zDvIVV4vHFR*0D2s?HWPR7f^mc2634Q#QA&E!sMeuo%Z0Q7LeU#Wtdf$3
zy9><Q+E*FNnVPAv;jA&HY^YW{?>rTMhx&xbq~p6=vWbjLxJ`_FoSRKD=iBI<P1go)
zE--ns#V$$gjTJb>d&?GNCc=Dgy8Qglz?J27OY${uHn6;UZZvEAjS6Zp;F1{&V7=*F
zPL%s5MUObG_h3C@)A0~zsV8U&uMUE*Gg040<&CJno*%a~Xo>cB|LCST#|oNmhdm{&
zIDHe_RS}Ku%?CZwaTAs775a#zUnXt<d&4_JpwdBw6~|U{8mtW35VW@~3qi;!&RZs6
zsz%J;rx}BPS9hbHX=O;Xs;rLz=P=W;HXsvCdaOhNr>M6s!gq~bKkq*GXfY)&`%rD7
zJN-_Ee~6qX6~XRx052}!cD9TKL{{G(O_?SXmO^=*igP~?rmD!Z`6T~4MK#np4W^B4
zqseymzIf0hf7Ab(j|_z1<Jo85YEFWc<%E@RSW8Q*02k89`GcHb>oxjj<wZH4)fA*U
z(+=y4;e#xIX{t~#UBlKknEh*8D7FF7iYv{_SsJh$6@7cicOnQ26wJ>qlk9Sj!iZ?D
zcGJ&b&6i{wsQg1^n3D_ezZ4grNFlXoU`1WQWF`vWVk1>(w|$63q^O+o*ha(P(Xw`c
z*Nvq%O_MpMeO&WCvz%Ra;^}t)J=oyMCc=yFXy9LYv9JF!ydgcl5ipUx!d(Y0FojBR
z0p*C)QB3$`{bS!9qhRH!KpqO4^#vmYw#|gdyL`X<)@*|6pZv2aepx(7-05p)U<PQQ
z@}2zPx7IcdtN;Rgr8Mccl4ifF>kEgsFK)!u8s`&*;3?N+IYc`*L16jvlknMABZ3VP
zV2_mLC-~W<d;ez?sIvGry+eDD31oY`h!E=aCt~Ff&bt2)E=zsK5vc_n=jS^N{CBYq
z_++yaQK!{z9RlumP5it~Nq)Q0cgtf>kS`CM%>h`p0O#M0Iw}6)wESZ|-zB%NF8y;J
zenwZwy!<+MZ2kI-*OcyK(i0o_k&1g>ZiZ3SjqA~6B_HD6g{*zy0X>&J$5Gb8YHl72
z(JwzASrc6fxbJop;6^iZk1Ve?4(9XWI{8w*Rl}~q;UupD&VSH#E0Z5|?Kz^i&yPTr
z4mu}4?)(`lye>bbulr=Iszoo~N6#7oRz{mPgQC=0vTz?czok_9TV`m2>d$@{g*a!w
z`K*3pEV{WwJ~dK|6L{Bp9qca5QeGFq^VQAq<=;8M&JGy=UTaRf?h>@@28vv_OC^1Q
zM9j;|>!W!C1o^6LTX5=ZRZAlRuaMrUu}pJ0H^2EQ5}EB!-Fu0aJNpj@U_^^XcX0Z-
zWv@-;`**{@t~-*Mk`ve-6?`yF9&mr@8hB@o25b`q8Oa+!;-B7})gSrO7^8Y%ZqBQ=
z&>7j1A_us?zC3OcK26rlN)uxT5B!5wJx1BF!Fv*L27Wn(e7d<2ido`0Iw$xfu75a<
z<0Txmbgu7$ncJ1IU+b5daQ86MS|oeqg{T4_3<~>!6{p+)LZtdT_yFrTVYp`p*{Y8P
z5`*isc}s5a%I!@lEG#o(Z7u=hZcHVd9wHg-X@W@Fu^9yBTvX>EnC2%gr$V;5_BM-~
zCwajL=ISVX%SrJIongg|t4$|kK=sIVrKU8-))xfJNzfA%z0n^%>(wbR$if2V)|})#
z0};2wPIn#$gMU`KJpvI5Vs2gZ3r^o1`8?UMEAY(UV`?_B6~qnt$1e?_utPh0r2g@m
zH(;t$vXU8ByV;gQ^E4oihNNlc%Wa~t0>ddBV>>zlwX<cPCPm;4+s3)_#yr`G<pSbK
zFo@qSV*_QfU@zNkA3|cG^?8h~vpa?Y081oIyo)$4s|Dcqe)w+=q=&}dRCcVAYKA-f
z!sj)5RC}imzJVX&0ycAB3A>5MWwjsbhAK&WumSgE@yd@m3yAL!odfys{ZMvyZ|q78
zAI^DI^UD(agznF4@fVYO%K3eq?TJ*SIRucsv)Op!WB=2IU#dtX)^o?tA-d;t?0vpm
zaW4i2Tv52Bw2S^77!)U*y^+9VeJGTJHkXCCpg-Qp-X{loEON%lp}UvD*M27w2W%KY
z_Mf6qPTH*@#!zt&Uf}-bI(dOpV=(Aiu&byy3YaK~Nl$+JMFOXEzNvy9<>+kqi3OPt
z*S=(&-lGkAxQ8aTHh1rAyMohH?lL(3&e%>It$MY5M<;9h_W^)IZ2i5rr)OZxxllP<
z^aS8PoOXM;xd1NY6cDLSSm53`Cg;B4yUyN<*G#`!eFP|fJ-yt(XnqJXKoz$+0QmoG
zF>=!8<iN86m-uv;v-|TcFQ|SnBB{5^u(rn%^a#mM>}J`+I?ed@be|2pZs}SX-KkmQ
z1)LL3WsHF1b4tPbynGPP569@@CV&@~>uj5l1ah5`Htc{O-=8=r4Rg>p4+6md<5BMQ
zBBT&T_IC|GVDCv(dQ}%)Q^LipsnJ$>KL~|GWahp=zpz7t&PQ3VCx9b}N_o^oBib`!
zrY`d3J*hqpR<4+h4rMT|lK&Ol;G|y<JqX3iM>m~eY?UQko<U%BR{zG=laa@_f~!w}
z72<tDbt7Zz1HxH0BvvOvM=NssLAl=Xy=uW?4g@~Yq4Q4G)ltY9XQ-hx5_@u2^UBti
zbbB^p%G7P`>r>Ff?C<AwdSx|3oi*ri0`)a}Z#wAlj3yUG(q5L|%lofeX1b^qj`~EM
zZl+x}_-T4`Q#1KNH4M95W0n5VQDs%P#g`(YN>mW|jIB`Kyr;<gE_Ff$?!f`r+>El;
z8TpopxNxF7)P@^`t?sroM7Mmu)X_U2rFQEIxQ3&mm3p_=OWJciS}+-&SK*z7_23Qp
zKkV2_qw=0|pl~v!(qqgQSj<8Wy6j3^5RTfW^R&h5!w}pBJ6DI0t(x6kHmJ~hrg!A)
z8ieXR>_KLn&cRtkm_4I*O@@Ew8Y7Mn=9@3jQ4!3&@m1a5s;~cFu&B!GR&lopo9qU{
zlq8uwZi0{;&WXS9jzOVWg*2zLY#BT7^}9K$xF}rP5vC-)?coMOi0iaC=B>lW8FktV
zLzSg(-Gb%lp~~}NZ?sTJ-JP?uWo59M4imr3dm_gl=#U;P-L~u-gk72hd4aX{12~s#
z#06K_LGgJM-opq7tNn6OH@yAk?D9hl02h|r_1%I*9g0gB#5KgxB<Q1`nN8MyY&18J
zG!4FRqRYePI4(O}>V;C3qWm}O_#^pQzPAE_$AZ8U{ym}+Dv6zZmhbfhqWIk%YtdeT
zRi`{L%(ONAC4k_o=CdJBfu7_Hr)yvtEP+0u{K*=ApD`mtdPM0`qnD6ZNAwTUAboE+
zVd^MM(Nnn=Ov!xiq;y%W!erBQ&XePNL={w0z(jfJD_&rZ{#|KVq`D@iVRW=LH{zl^
zzy~$=P5o@~LW7G2J{^5wNT9N0(MNu)-{S7psW;WhyUGBo!qNOXI~l%ALKo#I<@9px
z-*bJphB<IV4OG&p0<*+DfHjFA@U=x_+`u#c%9axdEQcoi5U+txe%QL}%%1ef76VXO
zd^yQZyJ#i(J4D#iO|%J#i*FD=>DgBqY;aMOFN+y{jsmNy2fMFluYR!?d*ZC8?`))n
z!m&1}6nKp$1gEN|yDVQwrx!#^BmHhHJ2aM5zcVHYt)pNOwVXoD*<3m<_SE&KX6Dru
zE+BQQ-)^8gV?pMQa<ArzW4L(rSA@1`0wk||IG<znFpzRJRV6o(A`{5Zm{%}OutK@%
zQWZz5gon(|mwYc^@lm)Je!}OomJ^V!U1Ua(RRC#Ad6S$pmuB|Gllc^hIz}eEl0$a`
z`o*Siul3zx?SfldY~bkMw8p{sy6HDqiwf6HuUDU7c7T(*(I?{_Y#~zKzHDyU;|V01
z&ic7}juw|3zC-FR88E9r;KzNsUfHZ#9zXi16s_jwzh6lc?yvRvQ5rj+n{<L+AJb?E
zY>U2{4xTxCsdk5jal(v?c-&-rUz5GtgXh};MKe0|LY4n@>9O(>z0W2c&j)m2Q%}rk
z>JMX|34tI>spbl*btDTc5{+s0530?+v4}t~UmktQ3u`rr6I0!{=4^v3Wd40RoLRPV
z3K&k{a)-7MRUWmZnxmT1;LXgJ+0R>Qb8qY(->lMC?aKiYj|MPj<g53*HKdPCxGCvI
zBk`w9yGQ0ihRRv>F=D+{8L+MDKx6LNp=I-Jvf;<g-@}Dw2y-MYD`Rl^<JA$lI{Wf>
zN4aqK8q{)}gThDaOP(hWva(?R9?$<YxZ3w^H@z|9M2A8o8}46qt+=`cQiGJhOsX+)
zG>jK8s>m|eF2AWEvl@gyFT(-uzPjae8e`3a0ux@04EW4^@|5u{Dk9*j`sh1rt(++r
zz)g<x3lHbS{~1e+qwcZyQ-DOOXXTZ!!EhG)qlqCN4l6iE!0Hlou-ku5`ysjOQ-q;h
ze?O2&S{Ns<ojiBEw&O{cz4e7?6rL-;duwiFZ#}!5|J<%$ePzN4W~{JaUU<UoW2^5k
z_+0Rzgl!1|?{%u6&12MS;?*UMao>5EB*`)9st+wX7Wd_9N=HGL)POMsg)h@iA;$Yj
ztZrI12Fz3ZwoU+Vf6rQSF6`2<@^&gnzk|83fWn*9```Y^L%pr;$-TSaSGdQF&DlAn
zc69popBhCUdQZrf5RhHB>+`0zMMLFV{x$20nJwW(_V_Ijz1!9C_4fSz$q7{rQ7l*^
z``o5@HK9II$8cWz%6VYpYNOiy;=D$2`JhT$jIpf4aVG48NHHqU0{%}(a3k3+Z*Ii^
z*!*ht%FbfNIl+z&=NplSaYZv^Q$p=S=Dw@M$q>mZw0;ExemW{OB_Y%?FcNEO<<pVA
zB?e4ZomD%=ole-endCT|j%M(|{}09FYYHcK;6}yFVeGTUBtE)ej_gY5NX&l-{QCZe
zy8rp0FFFYK_&e4*72Kg*r4{pn85hGOfXUy_BAm2Ucn<iqd{K)VcY5OrS#Avx^tJgB
znWC|J1s)XIy^?rgyGUM*??3BDn;yZ`X(CUw*Co)H#5w&MDEFN7*Z<2|X{0f6_i>j#
zOdbn6bnj(d-f6357P^?;ZEp?xo|IcZU`p8IZXFQ1czSXpAOSW$ZhK>5qm2M<@fr`M
z7B*>OKf2J2r9TzgAUEPZ5L9hpt0m%$LbN1t|Bf}Lh&@LYNR(brbK9t6Sk#xFXrC2l
z0$US*1+NuRyQRSBRBdSv2W;6a2&CdQEs9XEwbPZB{t8w=3L-_LrTY;>3>%gjldQ`k
zV5M_z;53>T*b!S_7pBDK7On{6dvBH5$Fw;-34Q>Jw@dMcE6GR7=*W%fN`pyP(r|uz
zJG^6;t<`&l-ZrYBRxwTx6k-BlicYh6k6biwGqw(oBnMU&!tPSs`edx%iT7f-P`-4H
z%Isq>BXDJDwDJV(hZwD;Ty?zwGGD!d`7z0v4+1|X=U%gupF#<g*DsfAnvb@rp%irP
z`vr~)Gl9upLz`Y%oYxTw|F4eg3~MUs!Z#q%RaR^?Q7kLHhENO;5@Askq==LNiUkso
zrU}JJ3#-5i)*vkiL{=#hU^k&G5EK=W6as>Pk*1#{5dl#`F%<KSKffQ}ulr=4bLO0T
zX6~KbnfHBDt0j%f0B0I==}4pLaGQLL&w|9W-;%H!Aok1oUEW_ljB8VrTFt6MkO=52
zd{2)o6r<z{ZEAFh=CquH-p8pEW-Z26JYc;ES`%VpV>m=*7u=i+l6kz?hK*t0@nbb<
zKrS9*#lw#;?%zlmlO<Yn!Vm!MbzOGI%Dn^FnB>|qA8SAtW}Zae4IenDnh#qXG(y>u
zIv~&UNMeZ3&(Tc&_ApRU@y$B|DL&4<a~n2Tnd>8TSB4s5!xxV=DbQ<uIlFP-|6zml
zn6==oVeuU_fItq`q`@vsjz6|q9aij3SR%ZPG=)<FWaKBf25;uN-h@q}^2h^l%Dc;!
zMfei~?%ZwI@#A%Up?e|YcK$A}wh!amuyXbt3F|3H?gJ9V%BlcR=+a>(-dar&D>`-W
zwvV)98uqEqS(>2o3%v4Q3G1k8$*PUbA~ZjAZx>z<!aj<VR6gD;&x7FqNx#wa<Q!ND
z>m=#}Xp~Yrc7)atv}J>*myP7D;pp;G%`{A-;N{%Xjh}tHb(NFLxO!nK?d@)j+aWsD
zA!r7q)85{8CtvCgASm3UGT#9YFPQnorjVIPeK>%{*ks~CQ=c~Fu)lFr$ZlZtd_J<c
z^>Hd~5wiTF7|=Pt&Bkf`Fo9WerG*6mgY0=FRUr9%sca+Va5G9P1dV|1z56J%=?6#{
z;31$kTSa8S==J!J){%hxoQC!EOS)pZo2nu<yx6n!Lolq@KS6ZW5z?>1hzW7FrxKx^
zQ(ta$shtDQaV9euKR9`91{PNTpjkH%baS#iWNIX^&bx74Zy1tzu`?SqIh<(v*^Tmf
z((gdCfo_?S=RK9s2V%eYcr<=1IAR~kbwUQhu;+t{Zz93qrrYdm6{XvjfcY=}#95gu
z(113|!;_P=Ya~@h5KW}oq@<wFmh*6N=r~uXTbfkDJp{60Xpv%*ZK8n1WG#c*4Q|}U
zjIhKG8t+f|jV=z*A2NUekgxlq;8EP>c5RB7R-!nnwt-G@2%P<yUXsTEIsF$pJ*9MZ
z0IiS+(Pr<ub!&ruZvdz_AMEEO4G{)_U<$?YvcdVeGL1-u$eZbNO_4fqN@Gzmn)pa#
z8Wxs2N%kuNw=}$#`C(icTOE~CZE;c=ii!=mwe@)fFet8&p4_|x7>wjy{rr6d10d0(
zg^!pwHGp`7Y&t+<CpUoJP4z_<dYyy9nsGw2gzOK&AXE(sWq2DDPvnqSo&NQ(HLP*$
z(@25NOQ_fJDd&>oWuz-qW8<JQVg5)R%L~eP8ohT4=DAMgp<XL+(kbcsLiSMx%f!FB
zf3eauAX@Z^_@0mFF57K?J;MyE72rXz$3mf*`t`HX06%Q%&5Y>;gINVMJ8Tp^e=nM#
z8SBC+x-w^nz3Y^)h?!#P^b05~E1OlPB^MMut7y~aR0~~|p<fdc76q0FeJE?H)3~Un
z1M2nc!8ft@X2||o*PC*cSb~En@GDk!Vs`|;v>JLB6Tg`^17hF7q4)qPqMvz`Oeg`>
zaav6@Rr^DMZrm9f<52|<qtpS4`+No(5a7?7ay$``a(f*q=}JehR;_~mEj{U(&AwFF
z{ICpVIC>5cmU@y4sb~<~%my##Z@H>KzaQQ8cjRwyibLzLQ>rQ?ia+YxU49Pwl=F|!
zq$m4<8RMXO&-+!K6;P|8PQr@4K3oaVKdG*)8V%@*XkeNa__EjU3#f!Q%j3%yMXZ?{
z??*<e5;o3_ZzsK`0jZtAZ|015(4k5yfmNZ0gYk!^ca?1V*+9#*=$uJ|iL2~hiOoQA
zCtr^W?Q%N64m(=rl41q|`HbFAMuIgiMtXhKZNo;2VXN)>kQNPSY(gq~0M4O&0P&*B
z7y+@*ewUYTfpcppy!X%SsR;&y5cNQ5S3)|7lASn*-F#93njpzafJ+=PMjQOfkSDo<
z4wrMG+ad_i#skr)eR4eza`$M|KA_TslAUqi2Fmk<1%WL`TUzmw$OfeNrgh!rH;&+w
zq@Uy>*y40P?oax)v0MzkQX?!YypZaLM0i`s7PP!P64!iBA}y>qe-B}-)C7)4sXELs
zwQefTI2-Tv?ez|CBGy#>ssiPJ@HYZsE21|Bt(pH18Szo-Skm*IU7&8^P$wSXYxYi=
z<~m#0^Fjgj;j3|x2x)Ibc%-})wPw9<mcIb<`Kn?!v1+^zeg*yhcJ~Xip@;aoeQt$?
zH8OA+h1_3_&t*`3SOJGaO_2Wc?~p>fgzQ7GcCDzp*~}Xn1r;3f*M-u|xib&64F__4
zSn-3LcR0H#g7fj0_M`c27QR6BXzaph5G}W(y@*X+Ax1=Z*EC~BA07P9S>fIyLK~yB
zJ9yj=s%u89gtCFk5x<+(tRHbbjeTE><Q>hOAwTFKAP#5|iXw}nnB}@U1d&URqYseY
z;EC#lj>=X8#jT{|X#R1sK3a8ZD3CBUeWzwdt+Hl2={eN&unekXznqgy8)O7baz>jC
z{luTyLEn(44i4ib)6^OZf7a0B`Zb1R4X~naK9hysQgbq~r)86zfB3e>y>@1uTI7%6
z{M0McI4BEA)*%=+X80RV#?&y345$UHYU8Fs`%G2lk;QIp+=={327!p3S!q_FI4t`=
znH?0U&WHS<J{9jQdN6h-EMAa?+H3S@88KP_BfEpLEMK^N|4Wfq`bfT%m{ME)cJWE)
zB)?n;jxOr;4L|mv8a)#jfLFU;OU!2IX7sfhq;iF1$1QEl@LT-WQTNLE7O5!dcyl#G
zmuE;A(|a@eS}+xoC9=zw<dR?CS|5>YXnJK@lTQjU1(T4Hrb!DHyCCif`cX!NiGV}O
z?T$k95eal#+pyS<cu{*husK}+SzRiM-B@<uG+$G8wc$qwPB&B36*<*kW9iPxL>>oO
z?+hKkpLh10{=XcQQEFFcp-EOoAo`W)vYq0_7`#JC>CYmazsbc3vC8Rj-YJC`JasQE
z%yY3jlbI>}`}q~)SwT0LbiJ3eMfoT<<#gn}3Q0w?(-srviCf$w%*T5-N_u2i-9KS8
z(bd92ZH$tvFegI#%(yJDG03lTVxB*^ev=ewrS~Zh4}~=r&$TiRWHct~J{8o>#70+J
z&84x;c)yg*eCcYHQZ(qY3`x@X!FS7T^3~}n>ux24<*Ny`Hs(#tsO&+ey^|5)ihGyx
z9(2@qvJQs7)Ar1V$P#apTMZ(~x9P_#ho&%d+Kb=%`$_{DOX}}7$|6e!Jpz|U<pf;N
ztE=(;`=2iCdsFtS`%|u(?R_N}CgwYNf9vGMUibA9r9IBa*T@Hfj+|{kRb&wcpZ8>{
z6?1Zwh~lk3kz>8GW|i1F>2|c{eoP4a^vJ!^L{`pbxbvmT+bwXp8^4EHd&?(HDXn#0
z3^&V6ep!;LpO;HaQ>lQjjfAs;(_=&21iW2qE&DJM{6kYE=h(|73tO$|r8pGY?z4+(
zy>Ci^-@;eenth>S>;e5+9jQ~(kr_#kR93#i^8*XvnJ<I=D+@@}IR9MBmK~$z&BFD*
znA)$w{zFyiGo%)yb;}3M7Fi7XM=VO2t2dXWQhPdj(eAZVUIZaG$)sm$tu1JwNmK0@
z{&LE6qCmlBWvwWpFd`!C7Wda5s{AEM=XBX$Skz4c*`Y(cj6^@}j<?+)w8u1|F4kgf
zLd-@LxlwLkzMfsovs+2F=6%*!?Jq-Fr@A(!YROldO%p$>c`6m7EL6UGx_anJY=ns2
zE$c}C_&yPPIZkpKgr^9U3#zcpLH34!b@tSc^(e@TZ*Zy#1BL^GOY;%7u#>Opy8?OZ
zH(-ilLhHp9Lm855BfhLu9Wzl8&bi;IA)E3SDKX074PMNcE;>XjPVv_Z<|a3MPap<6
zObiZrs<%#zWQ}VjyVr6vnK#Pu6l=kd)H8kdk8hj&USQW4g?sQs=RH2lxs2MhaJm)I
zv9{(-oC1j$<?u=T#534cbY3-JOF_<*w2JWn7f$t=d%bGqeN1j9DoGL|*TC!AQf_9J
zzj|7fFZm1S<=v;SDRe2xX6fG7@uLaRTmq|6mti_|u>Yw_gCfZ-p7)-*dx_L+u0GM{
z_U2msTdojghp#xHT*)FH7YKsq>ha!C`-n;I8u8*{mt#7nPqWz|U;KO|PrLhY;LWaE
zLd0;$3x&L%(dOU$Cgy^240axafzJ;AMb0y!k?W4HZ}F16ph6?ksqzc@8$%yNQcYN)
zjeOfE`L2?j6;fPnS1uX%LLZ{n!9#h|c?j~6j2-X8P77`}8`y|5-1;ncjFkPP{I5k8
zEHH_#nQQBTBmj8BJEbuoN?PCnyAZcVPC2e#G>b+ghy){4E$N-Alw;xkgWS(JbVSGk
z<F1d^X?G9K3-SRK$=#9CR=0U0h_2bfjBf186Ykd{gl!)u0OhF{&2|Z*T6@!Y!$lc8
zD36!*Qzzl~Jf6i*5zp(!ch9zcpP2+~#R(a^;G1T?x->*vz;XmM)+uE{*A4DF&0H_d
zT%Udv-LHsNAr0yjSJ_jK%w6pKr63?+e>OYCX-uat_o315!eY%rT8W#Bg#j!5ngV6N
zCM}fZn8#z{9C!qw@x8N?U1gfoeRXC#-CR7tTCImf?gQ!;+oi{JCm!A4Ax+;ahTV|w
z*;z3|Z$o1~KoF!8>+ByJayB;9_B=5ZTp$Z`3#5rT%EZF_B;ug0xuvbuK_hc>TXXX)
i(hlzbI7G#SgyYHo-(kg+0f4+fOVZi>_gcp@fBY9z(lzt|

literal 0
HcmV?d00001

diff --git a/docs_new/logos/best_buy.png b/docs_new/logos/best_buy.png
new file mode 100644
index 0000000000000000000000000000000000000000..a273b3647904ee2342e4ce2d65f8be40f098a316
GIT binary patch
literal 22328
zcmb?i_aoKc|Gz|JWEPPT*UH|#Z4s_bWbc*iy)sKigli`&7oqIEm6WY(@5t!dTlV+7
zeEx-R{ZQ%kI_JEe=lOh|=i~8woG49oMPdRP0vHTNj8Kx-hQV$y!eCg#_&32%<}oji
zz#lkvvTCw0SVbJ+xdkrxH=~u3wi*oP&kBQuyoSL}!H+`z!eCx}FxZ+o3?`NagHgGD
zXw;Gb|8dh&MNuAh1^v!xDNF=E!Sh6@J;7VNiI2xb6DDZg0E0b~K*-DL_)cxk`n@#v
zYPs4yUR6!GJ5`yS-;G*VZR|r)Miai(#xW*)!n>yHcE;gb#$KNjV%&LHS<e`T&7-Gj
zJVE*WPVz%@+!z^wVWC8OL?UA9F~<;Eq_^+n{PMXhy9xL!SJ~>Hg+B|-sPy#XldOHo
z?f)`v!9vK%W9cR5UWg75ymB>U6qhGhH}Q^ax}AWLNmO$`fhV$dSI5D-bvd0buC&>Q
zhmC?w<Eu@z+3hp?#=0+}xUbCC;YPG?7J3*hIZIAU8pXsJf}&Yquz(+tFqmeHZJ!eT
zl$n#--8TqAE==hpYd=!RhG>4wzu{TWXOp`;hkQy;C<{pDT#OYxJoXt?h+*uEU`?8O
zb*|aw*`cXQ{tZjKsTdn8V<GSH5R^#b<%3D-y}dI%FH{P`J%tCRDFcI9IsN)r+^8GZ
z#>I}q)^py;{asGj=7=L3c}7mSG3eaLIiB?*KOP3d#ewlDvkqoSR@7?l8wv&|s+Og$
zhY)1A>8UiT<X+@HWH=!nPlSD%z!va&HvHI<Jtd#a`5EcKoyWA1ME|<f{^0F7OizoF
z_U(v!UY-X!oDnF-<8&FOk&G`rtM*D-Uwx|(N5B0i>%Z?y+esKN<O_o}WbKDauev^$
z-U#6~<IKRnyOsg3yu1jHi{20Uz~2(&n|svI2T5cGy@4xRr;5^OxIe1LInuKBS1lby
z`oq7t`NbLRq#^#(o0zsc5$Wj}8vQ{*sCq<-Fz*xnT1wkgRdk#(D|R>I%H6=X0O67>
z6vfi=6t|nIX_~mT=%{vXZbj|;P63T=zlJ|)ZKba!Q<F09pchw5_I+i!J7h_(WlF!n
z=~(SU;pwXG^Zo)A_8;^(X*5a&I5yGR>G?OpRYVOg$)azC{Bm_=;?K2hpL0SRXxE3x
zzu{t<jItH`b+p+gM&oem_xJMK+;UMH1u={OKYEsO-hDhRajDc*g`Y$IPq%mXwin5z
zj^^m1rNjtHjwhX%!{vj~gq=3woGKPNd8PmxWS=x;npCyT^q1||v<_`XT#8R%A=qKL
zkxK8`CM5)=tCU~S$FfbUo*MmaY?2oEC;l!THbLQ4g(Oap$@U80o<7ZyCMugmmJ%Ip
zUO5Iz_yy|fF1$~6aw>_$1-qW#1@HH;&q|k);Q74e5XrXfg&FAJ`-geI%pFHY7LxfN
zmv?7}Zbv)Yve=J&H@Lv!Z4pdW_{*i{=T>{Gp&Q6j)pY_(EX`tUrcb_qjKvV><WbqQ
zk~p?2eFhgXhtmVxGYaC6)shmUdJUm5Pq$=ZsfTrgk$e18FP6lZJzc<zWtjq2TS81&
zGFY-jB_bH>M6-;tfimaks&mGooPI<Vu;-C%mx2>@!RdX{y!K|~_okT^(FO^{N&$gf
zyEV_5+!zyJIUzasM;g};7>5FXbuCHf#j4qjjdIUK=!f(OI31ne`{lbw!gCgdx=GGa
zx>=nb`1n;yY}b1!!N60`cS=~`ZPog_Jr|AU^OwoQyGwQ?>Tp(40?o^PIXSDn#Be@U
zguk#G<)O$aEZyvVSyp`XfO2z+42UmHb=IbX{3)#U7%&<INETU2A$w0n1^I<(G%hDi
z%#u432GRT~p%*;h-;K_;mJi*awSKq{-&YY1l4Faw2;D;urTHq7Rw%-`nPQP%%A^%z
zZ@M}atCt5NV8A4kkCrCCT|WKOofvh4-Qw+C-6ymZsg2VbESnmPEH20<lUY)gxnWY3
z0jAq9O46hZgWQ83Z@QNHy<%gbH@Jsa1~Sp{Moqr!U%a#wW;{P<T#k~e1ceAes}G{c
zTt8@cn1bwuys1e81;gFFgy4FNKAPPqapvISxrCqipf5I9$bVv*TjNjubc>T@3SW^f
z!lK7VeD|zLz6P+kb-3F+;x@|+&XLN}m!Sk%ie1s=)9tTgZ}gGVKG;v#X<Yp6*e{-;
zgHT#pi>yXpISim@^djmpb-(l}R+Ddh*O9X9_E8DSLp?-)Qu<T-0hwIyyzn)Av=a-s
zx;sU}N&eyPl7^o%KHt&WpKg*~&pZ^qy^ff;NB`-inTH8J2(2EE_la{0e?-p{qHa@v
z+apuSvOo(jPLlqs8(%8fKOXH@79qujo~w9$Kedc!hSo`<IZG(y*XYp1fvw<0$fi`S
zl_}}p)}o3|iI0%w@9n*AcJ=wTG3jNE74z=7)nfkEC52l*o;O-ct}5VTRpNdWHVB|)
z_O0;Zro#SA@o@GKvbmBEl1=K)bt;QATpJ(GE~Xw;ykdk*oz+J}c=B`Gs$v5E6{esd
z<bU8xQWDG@HB4W@S+}WIBD(6XB0_P!z{&Gg^BG=>D0yNn*tfwSF7*m%Y-Br+LXU>~
zVd8v(dLiCVXyQA3+nCpfH@xr2!oHO}m^%ncoYNqmxR+Sk?}Zv1TKw*ZhQ0#Vi!^hT
z)WuTq7B%az?O^tTd$gqYve7JE(Nzj1IW*tH^4I@_fBmPX2FKdC@NOY}Bb_7^%3dzF
zlDaIW&5*eC8rlP8VN={Ivhi22fI_L1e`zQ^B#w7`|3+ei>gh|!T3D!W*)-*Eqp&6f
zA&;}P&1So`@#&?KSfTjvsn_)rwE`Hy57D?A>E)yi+T?Y$T}PCvN1?`wYew>52JB|d
zWP$}@qelFY+~_Xk+nGZVWV`o4ucyC{;yTH-r6+g-o(N@0#NH2?HE;q4PrJ9H_FtC)
z8|^km0^s}Rs{v2V%wJ;54U;<fC!+EY*wYh#?|yi1lyxbo1)>htv=sHc8AG=OBeHPf
zZ+lHy9_u9#tgBJ5Jb1kTBEPbpL8|dEm^cWK=4+epORTJv_1K_Eu05w*TvhFt{+@3b
z%o{zdwwi#QWvBalp~PPRqZkQo$V48wCBp+8^Rt4k6&AR70=#^5uUKi6-()i1{@MaZ
zY1smSr(qY}SQ7(N76~Ls>{>(CxA*58HWt+X_rO5?OSMotTG+G_oLrbEm<l0N8L?-2
z(PO-lKwk-&`3>G64p!~%K~9S-x&y9-nUxDJqlsyL;dPYz<JJwZ2x^`!Hq|W2047-&
zuRibPE7FH^W@nB&HJO<t(odk+Y~LxSGy5bKN*%&&S)u>Mx-ee;$=MeXjhC}iM@Pih
zmWYAX{npop)l!$cl_d#9!5=23J4<<LZ)SKjuw)>vXDPyFl72%O!u@y>fzqsSUMtNw
zFI`z#k^KnMEG~|O<+r24jR-xIXw1WRD3&`lL#I0&n#8pkwD6{7VfnGljO=&*6!vFH
z$5LxzEI~*#ykm@t=FCj>_CZ~$yu5WIhsYI2ev0v_t-M)>E;oh|;f0L(d8{f$w2JUg
zf@^k*1Z+MQY?UZ3-*YqINo);)LafL3TWRo$K#^W-m&zze{)&IjsPyH3WE?zexGR9;
zW2Kee&xotj$XFi!A+TmVE9D?QT^dee`1oWh^5m%nsl=uQwLF2WuGGgHVM8SQGQ@YM
zh|+oElahpjDCY}RPJMY2FInFb0TTzH>gxOwQEe_4^H&P1Lc#`f(#r)ti0Jqs9$U-f
zZ%P*W7mrI7BQ5$$4!<?kDXuBHGJ3^HJgA%gCL?-myE4N)t39#VR{mT|S8n@{qA&Fa
z_Wf}s|9pilos}?SyXDHF?KoZYjFMBmcf2|+MH?&RgsseeCb(sU>wOm&%SX9PC-1#Q
zJSa(tWStEhN>ImRqO7gX&f_Zn97P+}qW$t%_r?$t!IQ%YtqE{t=MG`Cw6}|$igRj-
z%U9j2Rwt+Rtc(2bZ}(@>ucr3eRA_AnOZUr9Jv}nF3Kp7meLMQ4`Nq~)$LwOxi)U`i
z3uT?Z2p=p6oShsgn-DVRa-(-6``e9T=pNOOdbX09(4{VtE^!QtSLJ_9ZK#_#Ev)-W
zx5|BTtXE+%{od^6W=q4n5m8W!{h|Ep+R(~oZ_az=*w)aN&-N|#2Jl8|`?pl99_a@s
ztO0E(>`+HLm#>P4GY&a_M+Vb9;!zg}%hc1E3#SIh^-0Ksg=u)x95iZ{DvL-O@T%Y@
zEcV0n=jpMb2VGl^?_CP9G-dew8^sNSpRcUOcfL&*dPk|*&2$nnOzPvr7_YwBUEo6Z
zmTJ4ZcFCQt$m8ZW^EWfCCAsq=$C&o>8WXyrCjHmYnl8wZ2(xb(eclyYKW!Nn&U(7X
z1mbop+3asJ{|WaR(f8UI4OYwWCRSThwOz`SPDPodAGC1Ef*cf+ZfrYe$Fik)@b%^*
z?JcZ3EgkPPWxOjA>I=maFTMF&^0tmfq&?`0tXMLcG2h2?)1DbobFCT9P4(^FkiqJp
zZptUUc}eEfKq67AI+B5pb%%ST^>umTDcN5n)2yw3DYKXEb8nP2iRAXGNe`9jWJ@#4
zcQklYR_m{m#=KpD5yH~p$`vGaQ*N@2=Tv5#0aL?1REHvx!;PvLZ1*`atOz=b4zmQK
zUt=9#q4(VSi*ilm2nM&zCWcWV(WZE+!l?2<+gg1V*Um?&MwS1KC-@&f+7>VNdD8fn
z1rHnriy5CQza0Opeh-IxY{PMNUXuK{9k%Pq(K7ES(v+2d6Fg#h$Ny6s`KzT+Eg4M=
z_XlA9G-5?1=p0$d(6doN$Ubz%!Krp7`1$T*f(9!tNY3W<tk3p;D9MKZT#2G2)e-V=
z<6NUtv(2?l#0H_wXDm)<`+$QzxtDuVmFSb>fLM8+wQAva%L)V>3z(fq5zCEEnIK5|
zUJm*WlSGUM8|%^ZKaWT(XgA-PxSWJV^I8yGb|}f@mpuIU6jeU#wEUeocHc~v93<5a
zQ~i2ZiyD-kuK#TdczxWj5aqgtv1iDy1LE%k{&6bfgB6rhjeXu0*F|66q5oEyWfNyj
zP5j9-u5frz$xww{yrwPSU&N{Y9??e_I8cwCpG*{TO5yH{Qxmj%(6fY@(6|)P=2a~5
z7C=?PBdC0T@Z!1Zc`7RP*-O%`Hgh4^ps>68HyU>xe~$+Y-3p+@8~D)RDvgZ=hVy2c
z^7U<L3JiyjB&_S7!uX+w--T>s1uh+>hbmp|o%#+_6>4bQSAatwBT_rl+v*yDif$`x
zX$m$%3O)?sR;^8<Jo+ii+Qx3?lu2B%MEsPN;s@C+@Lh}?2}&c5wQhAiHSt2#90H^>
zGDPidcT{JP{@Wvs7|Gme$zQ&e+%yhi!BZP2f&enGGv2X3J?OO1AemrL6-yU_H(Z$M
z>8&T?V4WMe6x5iY#gNo@J*+R=!ugEJV%dg4J<bBA<gx_cqV(i>q-cwP8J&=dp*#%|
z)z8zEL>)q<=aKGaGP0AwHoiJDLgu78M-|I9Sd@|M$aoh<9i%JIS6cwi3TB48Jl-ew
zkWBE(<P%*<I`03%s5>I_r(RJ+-GW!x(Q6cnzUV8}*gIosNy+DXT=Hg}8u0@sv^1tG
zQN^Z_g?*x|DWy(?f&_fE^0PpR3dZ=rdeS!xd<Mc%&FvJ$W6Pa<=(BR~z|D|Swn#}q
zmaGqT#mu83cgyH{wutLRzdcU}>-*X+4Hj3047zvovUl$2fE@8)7l3DL67R1<K`*R^
z`x0i2@P9PmCeD6U@DT6Gr(pR$5je3R@SkX^y2m^5^1MC4ij^{lv14f#`(QM+E<Ylh
z;PIJGphQF0l26+Dk=zmOaVm6>E&_A?)NGOPG?V&RX3pBpBHCItQ!~F(S;Rnbqui>$
zzyB4mY}vBdE1|A8hzMoxf{X#<s@AI_Q2d+Zo)3h+n7#BByCAxS)&9<nOf0&aVynZT
z6Gmi%H{r|{{;{I1NnNy#_644!now4=FsRU1bN5%bRfK>G$?g)^cq?e65Dtds?x;~*
zGTJjEC^G}$ExLOd#i9{$qC%`LrEgj0>RlHemmb~#E*c%x&C_jT7_3O%KC}et*rs)L
zB>Rg$vtIt@M<t`{XH9v#OUcBsQK_Va;Ay^<bGBK!FJ`QG=Axp%WnGCmlHPES$NatT
z6TnEL!SlB3J<rO2%)rswCsccaX;v>7Nb9O_yp4PcmgyTV5#%g(wSCeIIBSjotro`#
zXO=#BThBoR4NP!pIH7-5;3zV!LD^29;w$(rENSQd1{z@-<WX(68GYKdg!@Z=QN|L0
zO)P~jmukPob^m}2Q8L`})fa*L8YYHv<Sb<!#%8}l4i#t<Mb)F&R>^NcBO9wqLluyH
z!wj<jHP%KZn8IbREK&Z%Q@LRvQmF{wB!@t&MOO&MVjX?xJvcFDJ*3Z~AtK5y?ugav
zu0g_koP&Z&8YC%W*@9>J#Qi8C)Cy}_y%m0WYRu{t)p=tfbYWz2a*_?hgk)D?#ZZzQ
z%0PP~#!1FZ4%N4*=Yv}vVKaLN+uhjI_Zl46Shpf|kDe@jA#k)8%BtTgpCH4dfL6O>
zIaq)8!-Ec!==b$MzCAT*{PRDHDPzl}hLwUPj;;oBT)Xpqe;@<5%^LiqLP|~^Brhg$
zn(KJlppP5ZY2jLAw>_&IKE`$jK&5OXRPiNmka+|<l`<c8>BiN%X>sxO6z>lj!|gc`
zI{4%Kxwu!waiW3zci)bZLtyM7qA8<IbUZ<<Qat1Gh8xT(ud1qw_h4rhSwC#N{E7eM
zNuWq;gT>f6{2C88r@ybGjBAL>Qax30yZn!D#lK300eN76{3MjM!)S~qb@68~fPnMN
zYoxPJot1m~h?Y_G4{lsbtY}4WCWfm@Get!Ub}V0k;0(SG;qXTw%r({j*oLOJS-r4Y
zeB{(12@hd<D!kK|j=Yp5zk@NMX<OKaXcDO84RhyzivIDqF5u(t(jI_%+FPC2l&J#)
z1FcHB3*5*?L>?12KK860g9g_%)=0{@=YFR8T(<Z@mSbrg%YdZk*tuLJE_DFs^whI(
zS=UOGE=ntbM%7yb*qR2a?PU$zF-n4G3SOk%*YV@vC>Es?QZYI`ho-;TaZBI}YizVo
zM}H8&9v_siF|?iVmYAe#8{nLzMgU@G1c{oIkk6Y%KD5OUZHWFf<WxDyO<Yj`yce=~
z2exS7RT0*DG+-j00*a7<2}{8Y6o?ML7nl$xD6JxvPM6X+HjM`mR?@*qyq$cK-_d#z
zr|_qHzd$2f>YWb2!pSD@OKA>LyGK-Mk_{fKHvTu2=F0(rryS&cCd}T;!`A=^52vFs
zWMFg+J3Ifw-}S$J5k5$TCp5fuX6|oIp}{|XnkjRLtdkrXdUSQC6DuZ5n46o{<1kS}
zaQsTyoxJGL(=eGfu}2AM*JQz`M?7>G`b*)dv}XSIO81qR+Sv`7AI)c~nXp%hM!p^9
z{1x^d>x_pVT7`DHvEB1_!+I4$zDD=-vH9@acb*ti^rvd#vN^I?mU%UD;MJe351lW2
z0Y9a%m4X|S1<1Fptal^30^%xYLAbUS_tF~1<E|<jKD1Pk@q54sB<ivKDx^30UVJq!
zI6Q6Na82K0eHUF}_8|L;Ula!lX_Q|st_pAtqeyIONt5)kI}NI9@YFzkV_I#p8BfQ+
zmy>u12m{8mDmMb|=jD6$5Uo2$p4Fy$!uoeoPi&AkBT>!i=Whk=cxK|@z&Byt@Fp%A
zji5w+!kE&5frxAXGgtBG={YcUaP6_%^pOX_Db(P(kl-_YIj)o77$froFPqc?P&<m&
z<0-0K+L3b@t;*XC#8N;RxI8&yjv|(=jk8<n26dq1WJ>Uo7@95Imd#^&u2Ltj_~#WV
z^w`d0hRCkOdB&UO8wa+AS0lp-{!d7!v;;1cv_o}zUa!*~R-9;s0e8hR{rdr&By3Sw
z)J7(~Mj0OIBLUdQCjx<HXj^R#ztPaISFAUoq(r;b==egyPXGaXqA6dA$p?0MB^*A*
zfgDIaU@uDAIT9X<Ajy;tHbY*I%1*8<y`qDa<NI0v*SRN1C;i5X4_DB%EVHad+1AQm
zLjQ&o_b=STxn=|=9@sAQ8&-dcW)JyDe&mVT>hnTf=DjFi<4e!9DLsqKM>W62hsEG{
zPpN9WFY)_qI+q{F5}Ud~+PzMV3|RX6W{yp2j5{!sJ!Els@?_^JvNYB6BFWd?x%0dQ
z*`inaO{p-gP?3%TFhNqOey9kD6MHlM4R9kL>)$cbNcuN#R!QA^BQ_THTi9J%OI!Hu
zbNnX*7lco3`&yhv^PD&lH`q-}#I&7|1LGf`P>mjc4vbBo;~CF@t1)GMq)=M0N0zdJ
zaElj!3Rf;n$Cs-n0pQ}#Uv-?#TsR^Gm7W(K#p*mMP4FbpB1*t)BcG9W(mP8y?bvHi
z5CUKnfSOl2RpYu_GK-qh!wbz+zsDi4Rrpi}Ms&ubB$RbPFn_B_JQdMW{AvEr$0xeP
z4K)nz!z{?#786_6rAg`-jM9azUK^CP_$KA<WG0^k1PEYg?+LTB{`&jtk8se%djE#g
zme>=t6iT4rYX@IirLAADm5pH6Vs{LRAZV)Ta7D~;{oWL{it%exg1fu}K6X#%wPFX8
z&)IM)tE7qs6R<TK-2o{8&YI~)!J;~c{m$;(vmzq~&-zf$E?enbG5k1Y;3x6QnQ6lq
zuZDLAP1fEiC@k{H(o}gn7E=`>8NZ~2Imgp+;5U-GuLke$H49W`c`A(lk_Tn7HYU*>
z1f+Ry&Fu^&tAM;I9#)*3W>0VLBgwLu`!AebyvGN>Cypy-VW=Mi_X=$k)Anh=EsSP<
zFzw+v`x#Ebs@Y|O5})>0)DStm?Udyk-_)Xp&P48^1hI_?3!$tBF$XzWs=e9$hi33y
z%LEyT4ysQu=$MzB95VtiQM89?xc`6y!3`Ju`~r`O;vF3FdnIIJRS|=$#p&Ay%w~oD
z=M;Dd!9(VBMSoxgP?Tv$;N{Y>oE$Fdj4n(XME;Y|2FvSc8~vMy%wmjO6(yw`vimb9
zLn>nfVZRUSO4)a&+J(}kc<t$mHvS`wwbKHm^zV6h%z>ECl>ov~(Bl1^I)RMMdcXA;
zpg2^z{N{-ocoM1;biB=;zef;&tmlh0FSI4LOFT~~=$6UYt%NNjl3|kFTOh^zwl8I)
z5MCLedLIj<vJMfb;z2Y<!A*bveq*pt(Fyx)(9p*$+3;@6r|}H4sFE@Jo01fKqobXF
z`S07>6Re4myiB;T=`{eKiC_JuM$Dr4;SnPczYb-n9qd2f)Q**D;8Eba2=GHT*PUGK
zd6x*w;y<Sfd4KV5Q#(t*2D!(SF{A%TCFMaB)uS+c*v8Ld#p!1~hw2Di^hARz-&Znd
z%I37*6V~nU0v7Tp8cSF?Gr;|3Hj7^mO!|4rHdU@alPqy8@hhQxR-a<MFM0cbPpRCQ
z2*(CfUYHg{$d|X=`}&`Js}l#|H);-8fHQ<t0Qv|nig9t6O?0}5K=pHNW^VFR??aVn
zZC#xc9_iP&$_5`uLnN<xdo!nCkKww!-GYpdzw&;hZl*k5vyR%fj(ETYAS+1l)1yi_
z(<zm|86U&E=c_08?QqDNjL>7`+gC-qw3kwW17R4OhNAdzu0uM)*q6|XVmSm?)Dt}!
z5spny7bp-|6OvD@_Q}Kfnl;dUPM5=BFCPR4Uu<-i^lU#pd%$&)xxrU|Qwa-9#Y<XL
z>QBt!=#m{KUI%N+;}N;~ITZFCS+Mb2?{>w|<pimQu@2Av&*}z`LM5@Z{I)8!qC<d}
z!FQ$8oqu^<-hy}vw)~Bulr7skVy^SIsP3kJqmjL>)J~44Y9>~Hmb9;CP9E>PNV(+E
z&y6F6_VR;$)#_lpCpp5=Zb0Z=2ux0E{?Sz@W3{vr%_&9MtQoCuznoXB8cWLsf1Qs2
zws*xR=1~nKG-uXR{BLjdv&0=4bzj3aKECWsEKG~=Q4Y|V$O}DV$gAkATU)!EGV{eS
z&*S}jcxE5_M8nU59k=ViNTG}9>Rdq10VqxGJ||FK<)p-QB9iEwD&*`^+}byA`v?>8
zNlN5uIsf3~%WpHid(wZb(}=viYNVbJ0oT`LO8CTI$DkY?-*x2o{6ihlhKyV@=EC+g
z3bnQ{nNg=uwKMnpGiI&mX-4#AgFNI9{}q<q{%(c*06B2Z;f*6T1fh~8+p8H+6CcUb
zW(;_r91EYDnaf*WCX|l}{_5(Ci1sSA2jCU0W@ZQ9c&`$7{m~;}3zpCk2unSR!ZJ2o
z^Nc+*htrsuw)_Oi!{?PVrEy)zW6{nGQv=oPHs@!1)T#5iE#jM>k+moT0gw@j9G@|`
zoPO+WRic!`SGfcpdv*PTK3=--$vf<+S2a>u&1_Q>2aOe%=Kw$X;Dl}=tEKZSeu@+J
zBCy(B7q(<rY*+fi&xmm@q?6A3QT#&ASb&K(#HCqOGxIrVu-dl?d6#(R0vtMG=3Ci;
z_3Fjr#(cgP-yH@%ywZt$IXXM{Lvunh*Wn%qsIPQ#-slf!-28k+gE&N0(27guj>Ti=
zFw!_Qjx94h3@gKZ$5&<aakrL!e~H*acj*m8gea_y-UjWm7~1Np<{UyeU8m+mU;^Vm
zs_XHzcOjE4o4KCsqAKv3#K=|`yE6|2nW)bmh%J^u6lnG8E-jq5Hez$M3G<kfk|(AM
z`GvqUK!1oKfR<G{2k?{u;<beXFT@0h&LwGvnoPvvb8CMXMX@yrd~s9+gkO(I_3n=U
z)64#?qX(%B>P!J#BVAE+o#mN<1a-8aVk3t0kM+umdfPo^KDwW+&-H9N)#ywxy`|x&
zk9)?_3j*vwo~f#%dvDpU7SM9;0i8^9^q6B~P}=t2Cdg3B5I-&l=i)tI*Vo~f<f57P
zdkXo!zR6=cYwP6pg=X)Kxz{bXKPn|m<n*rF>>=U1WU9OMF)oG2-g8Mlhf67rthaPN
zHa3Qz9T;3~B2AxKx&(e5v-9`>_wZw1E>ZjC>n>9e>n?fNOt%<?JmOr^rI2VEt6ILl
z;{N=Y9K_Ynovqwf2^x)em$Q<pXG+D%>mjlT_h~~>)tA47V)&b;l(3Tj{+73VJR}WB
zAHC#XKZ+Ad6wp{)3%@>;hDk%|;O7nvul))&B}2@>h#y3esb)@j$~^krmVDBSxDjCg
z)TthWWeJsMZ;TH<V24EuSSjocbv(SvL!F>PEy^E(yiIW<WTEpFzSi7fUgHiUe~z|l
zhPY3UdGba2Ug*Mf^qMzx{$jo9qu(;9-MHm0!;&Ut*KzPz$~|uB7P_5}V@dad^M6zP
zTyA5*)d}O+U7C5B>za)+08@Alu5HN0T^F)^BiQ8sww17!OL;`LBO?#=y4RI%eCsxr
z1}6|GH-8EiK0B@*Jj0XC`5-LBYD%u}nu>abxbbZxy+5JD2rz!21;Rvf!G`9>MaoR3
z#Q6?$&TbAOZ!-}ST`7zNHP*HS-jD^Z4wK6iUzTRYO)3F^X5Hj6z{!u`5#Pq+$r=<W
zJg|vonOi3FNyu*WXvuN870<Ygr$l_h{x*P;KD0Hud;-CYcG89uQLs>`y;+z6>emz@
zK6hFz6Ip!dKZhirNtW-fc80iv=Imwak~C812@UmsN18q&UR3{GFJSrT5FkvcFq!|N
zx)N6#Ff_H3)Vrq(3%JM+5JCvaH4v_kGf^XAFS*Vts9u(P_U5lIzCe(gA4Y9|zejPB
z@0j<Cz5S#TIeY2@69yI22m2c2xC*lD`}_YhMPq*Ai$h6EnTDje=MpDs2nH5{7=pU(
zRWWT6QXG)A`8tb+<)p<k)j^9(c^;E+`VYO00ASlZ=`-Q0OTwsYd%9ngZy$1L%&L*H
zdqv5AT)~9aW-A}vMYliapDgqjOs|1snOD1#i7C%UyG@;QSjtRp5WU|+h5X`VIQ0VU
zyQ_rjsf7G|0(?MGqCv@yhw=2!5f?npWHnKwp~iJ!eig(D0X&$TH5wNbUM%p?J$IGd
zCgwwn)K=BI)UwW7S7lX#E$`W{4O0`x;N^pARVHQElcXzpYkj1Jcwle-uP!&p&qw5f
zGwj+IpGjRx@ob!t%I3^03w+V*-IPS$1dj6Dt#s@6+y7bTeG=Zu)2c~FX{{V4Wu@9f
z)>QSxPB{_&`_G8pbWGRy@{5XJ55-7Z^su?nZnBas5<3}H$b3kPE2dDIj0$Zj=OGFk
zFEg0iqlQp2tm(IQMu6+rp4wVnRH$+iXjRA;6Dq2BOWT{19HisAY=P2<?Ydd1r`nHO
zCe_L|jbtX$sf6Sh#WzK{keDB+&{@9U=wMGf!>wzmZay9~Z1;-o^MKUBBQ2P=ot>Q-
zPomhj{&`M&!6<Vof-SK;Xh9#Lg?eum8~>OQ{o>>@FGJZ*cxa6leSflROcj<%?RxBU
zVEt6|y2D)Vp^<^11C5#gK$I-PYRkkEgEVV`P>42GVn?gIhPjDEwfSojqJ2Q}d1?4;
z`p94u_C$@bSW79`5>|kFXNsweBo>_603Zm4Uw_b#Q?*8RJ?WFQEyoxCxxt#g2Z$M#
zy~UADY}jIc7o7qM$B8V*b@KnMv)%ak4vM)^T-$7ALBBoAXmW1vxsT@_Q$wISIm!Cx
z>bnnTb%pg0$(?HmK`?IA`ycddeBN?gONP+a5*E=ba4vw_kc!e@+d23_yfTKqajhIt
z<MMao<v)%^nC1rPL0Xx&sLoU0x6}J<Z@=)t%N%DZrLYRAUhJa!!NS=r)swU+DZfS_
zK!a!aRGdBuCr1YqczO1DsFLFS7kxI`shUzDWUG?t2eG=ZvXTyNARZ(nCC|$j6YP5c
z;(-h|L>rW>G1%+LCq8n-#<LmSpIJ!EJ-xgO^i)oRy(;|l@b12pL+_%r(PO<E&45>E
znsPtA1Kby8-}`+zcOygc*Q=n8@5}8I>-{`|O6ZRuKJOw@5zz_C-6v0|*FMZGf0>22
zDQw5x*BL?dw=Uk7I{FvEIaJtX_`AJO&Z!5zr~1Q2{pY>NMU?izv-*PyKy$c6V#^f$
zp&=!T@DjLS_mn;yLexC<7buF8(365umXxzi`}gK1zMycvG)2(KYbRwTVD_gZ_z@>C
zO`b>CVUjgHDDlUH&YRtRrd+ey@hDSwEM;XNKg&VUQt=r78~!my(%ieesx0=wy34o?
z<se7<k9>Rppp_hLNghm4!z^w6<w#BQp_%{I;*Yy%te(8cb4@f!p$O|ckI5f1UNrli
z`jjMX9g=(Bvm6P1#&7{O_Wh?ykJ3@3fp2zrYdvR^XHoq$fhJj_049KCR=lRd-u3~J
zw59EKf>64(LOcf1T~<ajlx(zZ2m!IAFEs)+J)@STcRd2&?2bx4034DLYhs?Dyuwi$
ztmILIW5;-99w%qk{w*jP^gW>Z^BQgc_gf(=76;<}%C*p9h!Pef%oZnixw_9b(bpUg
z58bPtn$1nx$Xo3Ls9_8i!bhnwue8F={kQOT(g6>GH^CZSJuQ;H{w(!^G1ciJ<543~
z&LTnNiA|^G<FDq4wZ<RVpVoK0Vst6R@~c;uT|yA<uYaBz*Cn}K+G|^Nfm8b`)a*_v
z$5IIo+pU<Bhukha`NvI%VSh{DCU*cCG57Urku#7mP;@3FWOAf=)-?q^qZpfOw0yEZ
zEH5CU+Bj!}r?<B1<IzPQZ6S#RGvDq6ov;kyt(v=kDH}{#in4S@*f3=)jaKcirpLHk
z{9=@#GYv+K^~)eCt?!*)&Yxv5S?y{jsrrDfzDsS_D<_NWpJw%?P)8OP!aoG$MB2Z3
z-ALZWzYIk&!;??#`TP(ZiCaFC^IX*WGqZC8f8wu4WLNmr^=*BYkkDF8!qtIjxk^>!
z>0Tq#%O;I<&dnA3ZtbljjrEKjSDrd+K^5rG1o}mhz|#Xai_um=LR&$#_Z%E+XzT9V
z0{h+WbSQ2`jobGNQtu}nq#+!P4Oxa?cz}(GdqWmnErzf6+0nu1I}&P%!E!!k(x>z1
zIl1Rf9VMf}A`+P_NN>G*touA63Q2jZ#^t%GYgH?EVwNB3x5jeGMtbC_sAGzVtGGLm
znErqKihh$kjaQtV{HTLT_QMmS`w|EJ$U|I$GeX&N4tz<5dZ+mCj8`wcT<liAX(XUc
zV~%}?4GMOCM$d*cNd5xMrv6%X-h-_3cfGHoXVraozJbh$T<!AW+I17`FWRUP|L0Je
zcgW_r{KhY%dji=s<zI>6yhlu}qH9wJv=7-$Z4pn5;Np?22%c09*Hl%4ewOsFYt5kA
zwMCTl)%2*9upBnpCpXwVCPcS*;n~#qx@EzA4v53Dxh9P`y_u4qDL0lAN&Sd`zt!0&
zt+kIrv0PosqMpzPwN#Ef2Ok4moBG+&#V?6_b|-Br?GcaX+54WI-tk$(`K2-&ArLpp
zqNZq7(f*gtg;RAKhIm`iN1y35xhA*J=yr)^a0V`2W;v?7`x~-1l4yV80Yn7hsNsi?
z-UJeae-%cW4n|G{To)2pO{M7G2-I{_6-Jtm2|B?_6$)4r-y0u!BLoQV#Iot7V@y*o
zIPeLN0r|<3ywr)C6rn`=vXb;oS)x#>r5s;?Nj1NpAy48dF0<vfgz)oh0ejQ%A462P
z!Zi44G>Nd8WTM_LmJTlmnLL+$NTYEy@&hUTwhOtPB^B6l*?A+WHhGgfu5qK}4#Vs9
z3_Lb+3Tm<deuKU=@&4umjnSF2+$B9%Wd=~t@eT6Z`ZuXc_vo#Ti*zE70dMY60Cio3
zWb7;ZDO#DvdkErka|`BqEQB^1lysM95NFSPi5kxT{j`a>3CcOL4=zA{hvM8HO?fNA
zdPVN!D3<6#7D_&C)x$mM7@BRi`=rlP-NbQ@EG80k8dO@<(6rQYL38g@QetE9!2nth
zncXuPkJn$|eP*N4Fs8~J9+BsgvHv@O3TCx;3;&^cN8gps<7g}DyJceE!J+bChw-tO
z9a`E>9t3|(ECQKX6jmosn|SrvMYY-PT3&}uc_P2OFo}G4;XT7!6vsAbEktCJ{i2Pl
zsP(HAe*flux5T?(W~SeU&z(g-8_Z|Niil(ufqonFeN9wjrOGf*U>(|n4}-S%SXH->
z-!S~Pv7{w%##<m}GG5bMlTL<3fUjv>0Hl@A2d7l4D=?~<1N;Qollpy=KU4@qqXqsA
zJ$88>JE4sp3|V-iy0(qWpqfS_v2mX-DCF5q1($+MgyxRQAXplCgPo_r(?2uaOnX<h
ze$Hfk!jqiY^2eH$iu4*f0d7HcY%Ac<jYUbC^RM4$-wf=}_d}1{F7VqqzMgg~CY+dr
z=<NF0Hc2S`#6(hMi1=(#3OXBc+1r<H<v2t^FIL}o3IF=-m;SKbBW^<WyLFV<`&Hj@
zYj$6bz8>VGH~kzRXUuKi1{s!N90=?baM&4TF!y-(Rq}RggGKt(>7U#O$9WHGTX>io
zI20zV5fDUzOqbjM!iynX9^7iG5zjbf{txU&QY|k1+BWfS!IecOR+lOD&Fmx{42f<D
z@z)(gL|0@yAWOEGj~$dAXOu7{fMOo3PC}>YoDej9v+Idj`D*<ux&Lc$6J@v8l9SlI
zZ5baHh$qEnukanuqOsXy@>}M$%eXI!<;Itt+Ew%KYJY`wsp|%PTSn4<Ik^9LQnWzb
zNh}8`_&iNmn&ae$0q2P1BcA#&Y7m4yRpBGuPHsT^x(TF@u@Fc>WEUUYR~4(~8)(z{
zb;f_!l(VRD(T=rFqZFWzpJUMSUdH;2Yvk?1SEsP8;NuC6CTNC-t*`5)j2Rrn8p)*t
z!0x|t7OP$Cjqr5Qs6BjEO8^RHAYxQ$Y<`e(PM`&|5{haFevCHH|6#r|{sPF>l~0Et
z8SC<;%^3Eb&dnlOEUbrbGkAc8An+AD<kxtz8gKG}6LRSMbmotp-q>x`eo?rZEqSE_
zV7V3bL1lsU(CLg<y6!>BQpO{5qL}^Gw>E|*nE%oysn)Fa^OAQA@+u1cEc@$yCkYx(
zv=|D$WdtC7h%)zs0>X8xC=UcSwaKKTcZ+p4^DQ+Y!ZNO7l5{2b3+n+vO}N2dszcy!
z_fEk_Bum^&3*@HJ*n^vzCjqQ6pLoEC`1{4HqLa4w<c8F&3zgFQZy#$teGY!FLk<cH
zMKL_FV_T)b<tu*dBd%y1S<=3-Cbv2`Uw2>H$a7_;UUgM)hNU6<3n+7OKnTGQQ;NCH
z8BofS2kkVdj!t)Gx~2AIk}|0FzI56Xruh`1QZ!Y0y7Pd|)3^85z@mIx^;;WvOe|pv
zdZQuPZ1#cqtG^5O*JvpsQitt;!gAXi16ODeQ4A{{B)tI+=Gr>K9R?uB^|8lN&;s;F
zO$HCkE#2<T7s=h>Q&8JVqoCK|GjfJ!)A-1hO%}MYIi5pWu*1|$!&7|F6g$&98(+#Q
zg1n^kLCVYE|6ZnqD$Oq=CsLGW7Yg2&kwcSs_j$=Y0Fnz>9y@@XuhE4{kKKpwM0|hu
zs6I(r4)W4&sh{X1_?54V%RZ|+$N}v<6XNT3Eb5GS<_8D1!)+-CM#!O1T@n4pLKL^5
z4oltogjvGUQL3@xsmL24FvXkNkGMy=yF*;&p5a07P(fisXDWv(@gV=isTj61>By2@
zJ>rK}|I=7ZfJP%rfxP{1H@AizOW<3$_1y^{Ac}P2Xu&k5M-?G?Y<CgJ)=U87u1(}!
z7jDe-E)14e%u9V9hSG)SPG-Dv%VsYcqrBD1Vvqe+eO&^1C2>7EiS1Ki;Z7tQ1rDG8
zSZ$pP-()!aLHc@dfrv;ySNJiFlWLaj-cy4Y$X=ceh%!M!-<qWc4|TiE%hv-mY%cK!
zB*>wEnvDe$^GlF6V|~la&JBU2-@>HrOIkvYn+lNz_d^^sDGp!HtaNo1OE`R?EVhfn
zg?pt=ziP70x)MWVgm7L}uA{`q27VXt?JM6)Hhx%MiG!a*m5ONmYxrcH?SOH;8Gmcl
z-;Jf0soDJkNcfpO&-AI@we?LsFtaFFOzTVP54;$A<WN$_&vuq44I1QYh6r6QCMTYy
zeX48Qyy^V_B|C~6e2q}W*{5m-ztz3W@a1j4bY<GyQ#q@}HaJ~BIzja^83+izz;6K{
zM3fO3Q5cWSciOnl8vIl{8cunO(v&Tep5u+EV4(jZpU)Zj+b76_=hm)&^q~9<`}WYY
z1*lH>%YEIXgtipe5CU|gq(=VF_?weE9yV#CFS#fY1sdKE1OTZS_Qo&hSMS^948<HY
z)a_iVG)s8BWP3Hu7vMds-x~2oe{a;*UtC2FIc)|M1-#cLVz#gh{DX$`!>pJsPwU~W
zHYu!r`hvEFXYzT9aX!O!21H0ybFnAS^(HSplOlP?h->cXX=A9D_BLV`fjoyUACxg9
z>e06?R3U9Eh~S9vrIec5Yxh@hUaWDw%>lD2EH66hMK&ruMxKd6Z763Uv&x`>g;>_W
z=p7b>hA(`OZc3BCe%#w&YhrDhb)L_iH|K`6i0RfaO9mT;QF}h|L@G^%pnU?f=`=?f
zD{8`cqK3-VZ?&B#ga<k&nGSL>1Su96eoVCvk{ztH(I5rpV4tO%4&&K0$l9X+GM+)M
zG4Ms8Wo~wk<-fu^#9pll8Z6*Dmc<N%X@K|&d7#WnA|9+PaZ;sAS5%u2RP@Go_{{k7
zp+XLPQm&zIhC#{xqs;f;lw2D%&;whlYYw1)#H{AsJahT@q1+3`2r_KGqo&#1h*jl>
zq~R}^&_3`<rB)@OghxZmsbt4vQ*4=>vtR$+0A%%3AO;Esno14N@o5%GH74;4Z%@8;
zM!q9R1p+`I`~gBtibkc?L0&ZQ#*r=pkD>;SffPi+%urorF_fq_`*|o1Ir5lBobo?#
z`)IclM%i(UB^;CsALvMh4~wp?_T5WxIHRFiZTpHkCaZ+oKh@s><Q5<lsh0sOu+TM?
z$8wP%<UJ&FdJ4{Te#bI2xTDl6Lykms%jT~2e!EZvdi>Y`-!BUPD9ex>Ed{)XrAi)D
zB4GZ7vFYjNB9IYC5-*?7Omi^W3Zr0@h+CMnfY?Vbnrv3}Pc>&hFY$jn5gjL)@I$U7
zl)~+w9JULBHavKptP)U?a@Irhj2wzIygIH7jJP@aGuk`t((1R3j!3YFYY3@$KGVOG
z7P0A?vF9<7`D~PS_Cs7;{7eaiv^fsMI5Y`Xv^~e2@f=dnhD9rC1f23BXO+p3s%mjg
zf3GH7&x`(7j*=H!qq7^UFlGtB9onC(ZcHA05jc}XNU-%Sr|b|@YdUh;{>_fBj)KE|
z0xeEI;0_<JxCjZc&huc1Q5=XKPNIJTw<$@P+W)F=;BNifb;*GL#}E6S@8ef&RKN+G
zxmrUid;X8Z9Z1sc6BtXK&FvYad%TxIi2w;VW~DRB<TM_Ihj@(i=+(&aZKF@~P45-I
z^%#liC$8u$3fR*)6T0L?uHDi?9_V|up1$n_bO|4;NBR~cE}}s378biY)R^7-Jq)i1
z&j?%xwD;Go^Ncr>-FeoR7U%#Gd6NmK&yNF6alX6L=Hr1V#@)?_nlt{DU+N-PsU;m=
zY|89N%UvA~GZT1aqVRtqST8tJ)h-wifJul+C~oFu#_a|@Aejmcq|w$~D%&l@p2mnM
z>urV}m$4ryK_mIh-Hs}*mA{P<545gd2|OMZ(~j?xkRy<BTkk1I?YS^S#=~{f@y0h_
z%v(TbW8Vh(qQAFQ8G{eR=-~tG#9z&tUB-gj%~hXG8A6`Mju-Zp<qHd#Diz%AxpJ{<
z)B34cno!MPfIw#P(@JDh1H{{<A3N@<NQv3KzHbGKzLs3_T#z6nre!zKa(4aF)&W->
z*JWlQ8QQ}ju&Ak}C6y6+(|;)3P>EpiFP`ROe=+Ux-a`$rLOOklf6LRC<)zLGQtoYU
ze6y-veh@K3K=Sq{?by;ai-LUIws05=3z5lcWJH8A()cp#bTUe>#XQ1t{!#_r&zNZ^
zh;Ep4tj&f)lT27tFyHpp`%XnVNEP(`IZRn)dv9c(<?J3p3E4+*stz;^$tcimPDBo{
zq^$P*Faf-+I075cM)*2-H>i|o^cMEL#p*tOdG!mHw75q5femvnEZ%x-!lh;vK_EeH
z%2xDV6OMRAkHf$MDU6whfP^loDeyDe!}Wys9<hvlj}1MBAES-*^$xJ50t#BQlyz$K
z#QUiA;<=*~ZzSLzsI7(zH5gWNTRF3$X$}8MTg$`p9kgT39z^A(8O5a1)};rQ#@nko
z>7PBhe%zx_ac`6&7;cYc@b|c38N<a7%jX1TOWL#_Jd|o%XYz#enUFM)9O2?v^?r<X
z8b@3*$UkT}aMFe;oB7lp-cG;Bg70iks`rCjB}$G{?w~6lk|5yqVi+sj@IrII$)|2W
z9h3ELY(C+JCOyGccBB$p^xLur4HD;(sCf1elRx<Q3i3iy+cx1-J7%msKRgb<d6Fek
zXi7cvLQTzF7{Rep=i}G)dv4f*8nNrKiPi82NcF@~KWIISSo)6yaLC(?H*=Z~6!%h<
z44;tSagx*3^?kq(ll#G;KjCnVz&n1+Hw?s|O?PIN*50|fl1-_hd9!B#x(Wi?q)(hr
zrR;S+^Oia~<h{nGjPJ1xozrA+@VD$VWyC+`$E8qPP;C&Fl+4LvnNMA+8&n(j1`0wX
zj%F65jsOU(3f__=^ibXV!?6kI)Z)IzW(rgk2{Ws0Lbx}?l>2-snZMbfD5W{~nDBs%
z#H;O5O~Y4KV{x|WgiSVl^W*Yd90m0bx#0Q#-AqZUgT(V{N6g4i>}`<TR0VQZcBUYU
z!M%goC78~e;^trla+2H<q8k>ZFOg8(Xv)vZwe@+S${6u;+$Bz*zldD|`>kxQcT`(f
zod^W*SkX3?jcsjXgQPw!*bfJSgd4TV#S3iu@`Z)NX`)%72MS)IbSYXdcRf$<Rl*)>
zh-K2_qvDYA|GBJO3<D4FD5is+`l?NM(==11aGVTRvjGyKAQ}_lJ@M<=fI^PD1w0*Q
zH8}F&c<l*vDdII&yYZ^>;?d`27M<S=3E(0Y7@}*6;nZ>G_2^^-J?If2M>5s8Q%{oN
zTF-*gdQl9F!0oYP_}ioa$4;4bh-GecF?x!$zJG_CgqBG+9CS3GDePO)uk1>h15<!l
zhXOCeNis`a`p?#CqC7@$cvY{q<_ps;XqZtQ&$h>A6Pbvs7%td`Dp1E*b#Fh2LZ*Bn
zJ}YIto+&&QDV4dN=Hck<C#G^A_NHC#*rPMNH@|noYSJ-<_L}prW@Bt)=K9wRE^!fH
zQm{i|)Wvff=!m6LEJTudK+?s-RQn4$1h}3SjUSSEheWvPq!8Rk2!22W3wSZ~YVeuS
zKL+FxTP%`4N6D3*ANDPdnX%nFAmVblv~E?P2r%%ljW5uJ3y<T$MEP6&H%iM6xdPyF
zP{dJVZmqTsU5sqK8_|-S9O*}X(18&}Wc}^X>m5}l&*!`M-0RCJ4-3e**fA)UyEoOq
z1s1oK$%8Tudf1y;0N&&NHQ;DrKh8|;w;iXz0p?dg|Du2077gTa9{}a0jwHJkvQT@N
zG+-7luVsYnEpTSKw#%D)NU6Jxvp{SoPHWj+rV2@f2}2er@17eYGL$|(qKfMqyrb9w
zB9iZBK_ezCvRa5nMT$C>O<D}P`xS-TKDTC)u;qR-uJjSVAt)Z&+Ur$v4{ISy0lTF4
z1SpTkz+Gm1+;&r^XCfM>W)07f@>&LF$;Vd@pfyk?y18dsR7TKPb_dW}s=1JlN5dv=
zKu6=<zk+ZriJ1$J%uLKE90Ie8A3mx6XguM~+qG5!0%b|V>c|Lx><MnDr^CX{8pSkG
zpyCv<DY``i<vbY<rT5dszp`qah-16!-ZC@6yzR)3b{r9YNNS_Ir)=*tGJwtzI9fh@
zr8FDd20eoibkvP@O;4arob{AM#5@@1_2?aLV}rgasW|oLOE(~5<_1?7B;8;W$x7KK
zaC>e<OcYjlxCR9je?1*WOX2z8g6{n>L74dP=;$e`A^06!AzMWcx;Y*XZB5U2I6`mi
zM`Cd7zz15k+c0r-W~NUjN>RyD=a6XRUp`bPgj8Cg-J<2+1Eqrx@@B{;xJi?i&7IXQ
z`QB(%LmGsE+-)dZ=uxn=aR8}l1$6Rtf+&LS?rb%t!xi!Z{%(UBHk<r<#_aVN^ZC#K
z=ynt{5o|g=1qk+;|BLxFw6q$Xr`DdfuwdW*w=SMp;<I0QEC23kmg;+tRxF7^(9m;;
z6g64d<Bji%bb8e`o{5mwugIRG5cc=aX8!|<hMc{E5V6Erx>(`Nl-}CQWC@rAr$GHe
zG!WMwtGt(``PO$}E%mwK)5qQ)6r*+Bdxd<|d?}A`38|~nW$=EU`y5u0vyt8mvGW2j
z2XRV?i23P#jD(U#*~Xql5(%&bf#@7|lLw@F{s`Oo41&`e5zH86QF8!#LG!sU!5Uy_
zsod#IuZj~za@-KRJE%AnAm3WB*vfwH>|9zeF>6%=4Fq!>3<?eRQ3tJlK+nXZK&3B&
zm}iZ$a07i@6F0@A+N8K!b=p#3hsv3s5}f>W0!PXlc9;_S(RP!5!6>{!@03Ya(`|f2
z6MBYEN3o1yBwy~z>yOlt%stQvS!ZcWgCAKc8V7yvIZ9FFU;sPRew*#~nJ*x9r-p*W
z?*pkp^RfVDTg5*Fe#F%Zk*P@}IrTgD<K1WerL{Wk5nCOeuGD5*|B*Z4O0-o0GN{UF
z!rirWL8?oe@6juTKBo?qLCoAaGvQm_rRno*QMy4sE6l@%W7X9<eTmB)4B+n3hKLNV
z$yAlZtO?h4;e(V9EEr$W7F{#~S>I(bu2NCjR(!>Yl&Y+nCqQ8hbUci^Kt<DU4rFF|
z=d(>W!htg(N_`T$NcA3$2LkPA1_y)8?gAxOOg-p?f`0VjE3gB^y04_vR?OlLaJo0#
z_s(4d9WQ%*#1@n;?zB#u?wNgr&}G?yQw8USUF>{KF_yFj+8;@c9GV%JtV~BiTc%Sz
zFmSMD%zDvK+)<o{;)E|FY<{d48vHvnsZJ_<D3N!e98_5O;vy(nYKXH{X}jy_sq>Wz
zxg5>YaW8Hcrvhk<-+mOBZyG?Jn0@<kK!b`Cyi!haTv#Nt`cL|ghZjod*8HI9#g|{`
zv@x$FnA`-zb49kMcGleo<2%*UvOH-&MT$o>{CH90OipqS2LLxZNEv+gsw1`g{5E=`
zLwRlCJ7qlG_6rH&PYo;=Sk2Y*SLN2E&zT)ghoCl^=ouWb;VEx21A3)j()Jz%-{`#$
z5X^PEI=;6rprqTGF?LhEpez#H|Mf1vS;ovjXWHW;<*A`_ihUoYw4DO_o0-TU74qpC
zAw8cV;G-bN5Or{eZ@n8MNyFT&oG``vH94L6n)2w`xkBG7z2u{hqaULg=0tWLah^Df
zj(c?02=daal*T#iE>^bI#_vH{dgkd`LDaS|#!P;fsSuLJwAV_1rvGGWfa$0|r^Fm6
zakL(HSJ`PKZ7e6Z&K=v{4o2#}Jicxwmoi0Bjl>y&>y6UdQlQ!>4@i9B+`SSu*#viD
zjivg;J+kHmitMWWg+jQb$-?A)N+;u^HSL#w{tyTkDi@dTEZ3<w=C?m=Jv$BS;8h+|
z0eG`tH{qw>^_OiV%&D#mr1P*vIX!!JEa<(H{qT}Kaz<JkxjlXUic8$9(tA4{>Ql~q
zB?OoXvo~NvlDet?@g;QTg>|@f+~N81AZk@jroalicxEGh#i1hRZ{q3g9>meIP*mY}
z3-LqmJjg0X@m_0y^UH}VL2XT*mj$(sZPou90}1K&S=zGTruAia?}kxrm%lHJR0y5n
z;mjAkZ@A-!c((i9ur+L<Y4wGJLndX?79b<!PrxzKZdyXXbfrizJ)ltg_lY#Y4T7I3
z>8CXiA*!<QMd8oBDJzDXOQ>7&H$J}136z#^b50x4CwC4U?z5pbXDWr4@PD?6KDV8h
z1{W(9gDU?4b+h)8_~PFPIESUhdrDXj9t6{!-dw5Q<mC&1RGx!Z9mS{Jbc>}HgJSip
z_-%T9)3v8rKX_Bl@$ZhN+-ZI&U7Ae&zxT-aTW9CH%+;2{H3m$Z|F4lNkB0JX|5Ng{
zC$f~1eJhM*G8$PT%h(etO9<JCEFm+<p2FB-ERkesl&G<%<(uqf7a<u;V<|LKW0~Lm
zc>A8;dCz;^bKbw^JZGNkdhX}Gujjt6<@34J;lH)fS1EG}t+iT3Te`|^#Dt5pC#ApE
z=rcwdp^_^&U7+nN<q_uIdfeP{kL3jVo5;TP7DO4x5ut<nAwE-dfL`>cs-k&vKGbn^
z*|YAs53K=Tt|gxJi+-K;O^wJ&6V|JEM|)zxa7U(yVgmB~=oq47a!1w+E6}?R7&u1k
zFcK3u8yeKQ1WZNa{b-+-^qI@N+jEU(LklLPrJ}&TwDs)W_t;g>n^WCAQMF}y7bW3#
zXKCqsWJPv2eensU#HD(_=5FcUPgD=+{KqxmKZKe*29UZ;q8Vh@i@9m#VaBgMOAB}5
zZe;AdpGjaz0u<G#Z(0#8BW5j`(C1)j$~5G=U#`Wo>l1d<y_aF6rV(s+k&-rq?9&cJ
zT=rdC8b5es)(n6RrO{uLu^L}(BAx<nPnx9VN0MuCn-dz9jDRM~TjuY}#_Vtv7nT#M
z%x()pJ#Vy!LFuhJtvBuN2cL_Gsg1rsmk@mSw8@+XTh|u63@pv=3{?1rPsPJUlS}y1
zUzZ2ck``h(T{<oVAUWzlXO1w42<Y1I0`eCWKq=&weOkm&=0***eTx_eG(b-VsI;B)
zrxhY1#i4MIxXc|1@Xnp;Ts}$Oj`V9{oRA*JpS5#`e_o$ITHlOy^OjeBqE+-u7g4m7
z9*lS3NDehR)*T{G+JEu>O&EVw3ajN~(pXQQWWw^1ipo4uvi?GUq$e?tY$l1=2%HDx
zG88(#N=e9M$E5savu|b3vt!>ntJ;f??>l4Trc|&fFw#8ko0mU_63Z$@%xR9XWio87
zjCg@=u-=}a9=VEp%e>fJiE{c$xfz1dvEuSt!W_;Z%c2eQ9D;K5_?q0}V_)k;tVAD0
zo}M~!m3qPo*h<A5o1m-;l$Y-}b&^gNKB7dKp6b^C9fphtij7$Hmny4bJTQiwe;%`8
zNPj6-86z3$`zel1qa~^`kC7B-`PP4d+WhAZkmu35GO&rh@`xQG7i!ScGGgw$%p#+&
z=qj`KVvN}<lGc7KmIgvwQ|uME$d9+!ripCHS`!Tg!NS7AD8<^$#4A~p-kwJMAxj=A
zHb{}g1j74!-G+c|rns5Wp#1v4Sia-7L7B6*s0^PladkH<4eb{9@=ixm-+~My_va-c
zG!O4#W5)N6(4Qxu<3&nAOr9^JC3a@>Ees&6oO=*&^QRDskIkF6TFaD!SL{@?rd0=#
zXb%0k3lNGO+%*#{m>NMQq{%o140t&11kH{e%n|Mi5LT(BQcqGguj5MH5^4kbLixpD
zP*&ji0HB;6B^_(|IN6_$xMM^u#WWDPNi@jU1>SA7J4UY3B~_u*-JW603HZH6<BB%L
zsOPnn5%o@LJqr>sHfCLh!8(!~l_g%~XhU>FvI9&0RwE1AHQ-x;M*_Wi`}voX&zY!l
z=uEc#u2R7aCo356SG&R)i26vq--|D=NtRmHv`ZnI-m%JUT<{@VVt<2JGY`y%jUr?3
zN0-#Hh&r+cfZ~0hY&*^aax&Y1*x9qPb?cjMug&Ru2VCC;*CGU_K~E`=^QIO}Y&UOO
z*gp2ee41=KbK?*b_nb<l6E72l*U0i`YQ(1P49~MUWe=HQ4h?@4`o#<~L%aXgC~Y)T
z=cY-3zlo>t^1iK5-i(4$iTn0I$hL>o4*>(*$AS?U7P`WPvrFQ6;4F&J$UOvu%s4BL
z;lH}muE1P3;~QwqX>haRTXpNQ!ZKiIL$WRBJwho{J9k!0yEPSG#vqhqn^>r9ZMx8O
zvRZyZHBFpn9KEYOr*@;LOeTp>LM)i^9x8kPOI)g_%tgcdBk&qdjy^*3ztJ*xL{dR1
zNSY1S-uGQH`2H`hgkzi(uk#?p;|3TAFPyp9WzI|+wf6XhcB4x#CPOw?9cfyJyoX(D
zln}zGOA+KRF;iPYgXh-oq(gqGsj2oV*|U?2R(!{rzfXQnf&lQMZc9>OWIabm0svGh
zwG@rdHT|kC@!yQ9P`dfb`%#Dm+HXA8zrhSeZqUhfI<Po%OIBPKQxMC{7sdEqZ7{nE
zawhif<!R|4jh&kB1iCTMxj-cBvz$5LC31W*XWHLh>++HXEdfPhVpdocM*??ysZcxE
zV}SeakjQFZ8Ag%J@;0^g`%Ia4p#^9A_pXJ9s{L@!@wJlaQ>|flDBHz7KHGn-fdqPc
zQDC0a>DedG!e_^Wp7SE9?dLN@Z$yGTg($o8jqOFjl4Ca|Czr7_Pa>sDgpc(A@c*0p
zO^Nd=WEKN@2*lsv7nnNkrCI5Uj8iPEeb!1kR|<+JB=&~^(+vw!vk4Qrem4@1g7WLr
zNVPoQA#<Njf+yt~jUiKnyIUeSXF&tpEJ-%!U(e;Q6EHXX8QbmA!8BaDC?sL8Nbahy
zIn*n-1n48#l=ZV|jzsS53XM#<&>5J7t5?SqR&Z$K6%7px<xo>1pb5TI7JwIYm$%K-
z6xXpn&0GtlH9TF_+qWmRAjD#Id-~!g->Fzozw@N?)zUnF%x)cOZL^<#OEaOf+}=BJ
zqT}o9Hl#=0IHnY4(1+Pv*`6N5^6m=A_Vjfs_y`YbSV1;^fy=hld_Y3}!0G;!sxw$e
zH9YAmb2}?u*>6=uTKz!`3orDLQU5M?yC?7j<N%adUU?-KL~yIsu?lkz4FWaNljmR(
ztL}+-Y4fq$h0Qvs9~aqKp(^LQp`-qfHS6w8Sy{Z0TCmBu&|IIuY&}kT;~<ktHRFjk
zwsll!T_r@e-3|q$8a<iwI=~E{O4MX$7=5|Z+S-dxStdZ4j<08~wYas)S9B?e0NV}%
zZ#0U1Df3~7{u$1*X@S#P|H8Tl8fD@!4yMRu2zB~5{jbOg`%iD$FYB{astad`su!(+
z5<c5z{?2Mj(*8;8qLiBhVWaSgwHzw`0SJCzV{IWRW9Zhr5%<4$3?NyX_wd>L^pqdY
zQ`KoCxrO#I?*cFSPnv{xejP<uS<|~|yXm-a7$0<4!zNtqmgk=cQ8X?u-<?c(an~s1
zizk}94%Qv+`OPBg)l71A9k-^rQ2Q9(E}&_h#iohQ!6VFA(-|IS>;z;aZ45A%KPeFW
z=!CI5SysB2@78f~m_cs0hJr|2lZU>&VPz`~LdF7q!R2!r&P33K%cO3}VBu$?sZdzx
z_3Pg0^3k(A`v&lrg%{19<gmvljcAeC^#)(D7~3p$mhg2Tcwi4Y9B9LH%cMqLd6E7I
z&`s`Y0j==(VA`cpcUPTxk9pL9$4MmeyjHZhK(bMyxtiIjILHl(c5IfzVdu*XMa;6d
z{QLk$7z?3pO^b2X&L49Wb4Ttm%Rq4{?yudwGx1JPloIHa5eC6?Penb2>R~||=Xggs
zvcv~J+t|cQG^r}k4*a<}QLpYn(D)Fo`zO=<GXj>21C!Ob;u@Hn8f#kI?>=AT&_j&=
z3qp!=e!kxWKe7-7hsRGWtq`X8$u)Y^YRxnAxoGxFTlxeE&YxWncRZNmQU4y8fp9K-
z&zs(juoL?{Lf`>tn}Lq0h#zaeLGv)}!~t$e{llF4Qdt{>Yp%=jkdXmix00<oh@gs(
zA+Z7A^=byXI@Uu%%2ul6hNMaKB1w;neDR@~h3OQdLX+Rnp%pzq>@yQgIO1~zA`*_1
zbD(num1`Cyn|}H0nt8(ed*iZ14r=v}-@E0x{V*M6P$P7PrzB3coifKi=?QfkJ=k%3
z_Zijp_dc{mE#Q|<qrV!t*N+ig#J1d4Q>-uC1+5~%nNs%Lgx)vvYbMgi&Q6WDqt8#1
zv@lB_+oCA2z7!ulT&7aG<MOQk+FK?F`D|=2OD2|#BO_G_Sb;~EmR|t)nfX9uY3&?0
z-(q)=u-@@E$N8I_#_(ngfFXy}pKTnvT8@7c+cw>Nde}@n5}s?OkQ3|+lIUn2TDJoV
zXHD&T<G*vKsHX+rKv3LW=JJOsj1bY!<JstGlcN$3kb`s?G~6rcWa~#J2Ey=cDDlXq
zQ_u1m04k^1W3r_Ap|*;$JmF~W{H3t)-vkJ}d{)K#^@=Y=7`Fle>uP3Rg-dJePnHB2
z7AccE=D!SsW>JX!|B_?QJC;My3Tq~!L>jB17&6J-*H80P@ao1Xj&g2+H4y}v30Iz;
zDmkj{-c(y|8TmDZzt`d*D>PKT^%{J=?tnzlbJ?dXo4zweyLi@e^$^9#hVujZcJ+%m
z&(%*z|5l$7igh^g`?AWYS@>X@t^29tEYOktxAf{-L8op0kbrU8jf$L@s#msM2FZ3U
z>cyEDF|r`!cRqPpwc0E2ZCOPFZ55J-`Sx(0{!pB*9~N14Y0{5N)-j{XKmLm`dXN=z
z1M4?T>}1AKat<<ss76gOV)n7P!Wfpq-#0O&Dz8K;*v3x>W<Mf468<boPJp8Aqk^`@
z@}m^Dh{7^DZrizBhGgeEkEP!H)dBXI{W)ru$9%2U?VO0;v)8WGmK~Pddm2kW0TH!K
zG&z60DB=6dY38@(i!(U!nmruPNhKKD#S<0Zyr!*N{V94&(Zcyc_*~hGbLy(vwl!Z<
zB1*#u!DNv$&fG>{G(YEsGiFB`$0p3kLzm3>qfSshtW$0`wrn=VpO+KU$rC&P%O<!`
zmr%5PX7*vp!7n%@+B?fyH6_6HfrFp4Ldduk(&7R)U0XwAu=<j&4yhy#*c&WcrUIY+
zIBlPrME$jd&Z`Z<Pj!Xep51cCq?aY*q&xXCs_h2i2#iq6a*r2F`eRo+kw$hhUKF_)
zJCCc5E{8(%!o$kCnr%S}Utlx<c7InVEV@=fEWQ~u<q`sG1_R*Qy<pb>exMjIJn)-!
z0GI*{rUNRcqTuq65X_hrd@=ujX{igmFz~n3zW<_RSLZ)U{^<NqCA&KRQDUhpEB3!=
zyC1hZ!vEA}S-V#pj7b(A>U=#{J}A}!P?f+e!eFUZY?Y;P_kX>U<^fk(aO|uU*V@)+
zm0oeKf5!Re-JK=3un!&<y&pE?=!ae}SpQ<Mn@g}e$~DLx{J>O{l+_iK)D)D|t(A}{
tRRl`q<OwAul#<d19JBYo3HSxLd0h?v?*b+Lts_7HhSoPfTchh7`+u)CYcl`<

literal 0
HcmV?d00001

diff --git a/docs_new/logos/canoe.png b/docs_new/logos/canoe.png
new file mode 100644
index 0000000000000000000000000000000000000000..3876a3fedc342cbf29b3b28210f4201eb2127b82
GIT binary patch
literal 3719
zcmbVP<zEwy!<^(`fG`*#B??FjOr$}OF&LsS*hnWSB}ypW-6)dMA`%-lLO_OejP8<>
z5+?nXj-UU*^Z7jY;_hDDy||Ybt@jj8PkW0N007WyYiS?>fE%cPJ{d^<k6D-y4gaJ@
z?};Jo`ubW}(hyKpbx*>8o&$bza&}wj*{!=zxCM22h4k0gw-S>+2Zp8!N}vGc<#$E(
z-?&Hb?(W+<hCp}`U~X*ySc8^Llbi*%xU>eKSD&3*%E&6+KlrcVd#9{2=F8V=sJvNw
z$3RScPGA4n=lOYkw2QfgXGK+$@?)zW;t1l|D`(e;ZW8Ig^K-$5hK0jJ06@u>NQ@jA
zG3)B8U0p@AwI$EY_zVyK`2Bn4@X*NI!^<y`Zc&!*->w904K+irtZnm6pI`OhPVqNl
z(NfYLl_qpy0*ngt%7p&+$AJ}2)o6!mUBAf8^UVt)T&&@(VX1BsC;k74_(0A#6+?^H
ze|E$>uWx!(45<eAd-v~85o}KbTXwcL5`Q$WzYUOIUI6>M+0dR{7U#VvA8z!4x>2s3
zC{>*u6T;8KrU`_rbqI^PFBUOPxVktbk=W#OsZo~!Z=_V|@s!u+-w~=euocIwM+O$h
z7x0E94nM-Trd+Z&FNg=;h7-KUtS1Ibo{*<d^mW6xmRzzxBWo0AklTrOPbT71aQB99
z?74U4`Nr;|H_|i&#f?L^@BAG~yMuXG=R@u>^kSk(<*MmhR4((7qyGfe(!)Q^eGU^d
zTiYkXg(y2LZ%gKQqe!O7zJ1-cI9sgkE<yQ|neB_IPlEFX)xAHjgMRMU%x`bYX0G`n
zuePa`*2L4!2LfhDf@4LT3(lR`!=-Q4`eHsd$2wYA5wluNWkn#@+>!Kj9W%vFz^F)S
zS1xX|FeCG_Aw4VFfA$#Af=%5!?i<toRQsAo(T(i8tT-HAAeH+e7Dzo(B9)ou?3Zjm
z>prq1?9E2NXknGivOOIvn917<Le~&w(8GlwpI6vo_pEp?ix<lwQZ0)EW7TuPkMjT>
z36r*7e?QhFn!R+6o_qFj6f3wx@zuFfIV>WNnyKsu|NK+7d?(;?#!9q#7CJFy9T-`6
zb2LJ)-wE1B!7nF|o%}A$*aGr}^&2}U;)yOkd@{3zUh&6Ct;}1sCfT$8YbaD!*%@J&
z*~53|9Cl|&qu+r+AwjfKA<(5`zCPj4JUotvKZ^&mNkJPg*EdR27~0c4D3K=n)P#3Y
z<FcG9ItO!>umF46@**Ne6H_BH;5eFwd`v}ab&=Dw%(VWs#g)LGfCr84N{UYdNc#2w
zAIo5y)1K}N)0p^>-rvtm1sR_aOBWMZz>MyKwqYz<?_0m#g_Jl?h$x3hu%}o(!3vrE
zol2^WT$Y*9vtGEL6ge#JrO_oK)fn8EtEWzK)Wb0};04@}zLXUzEzhk}e5~K5?AE0a
z<7q=SEKFY@?CWoMr)`M{C*Ni%HTyj35*78lolPpW#*o@WL1X(WuN~r7x0NO!Qv~TM
zuWHkd;<K&QFo0Wd&8?jS`D>1dA3iyd$ZF^K&<(HN$w}xs)+GhW{+F%p-ckD_8`X}f
zOp+B0rQN=p&orGXAX8oDg$V$d{%~6>E$o;Dbv+?txB3ICUT#QOam(l^3<cw68443b
z(<@y%wXW{asPezz1&&73EW%O^GV&deX(T~)O`wKRV&HTD^GDKD!E)ZBAMgh(ctvd`
zvN795Ge_>yWyaxN_3{2At`1hbBfM2Q<4Mwt6=P)wnEc!_w<u9=-Qp4NZwrfqD)4^&
zxbI*>YkY{av`_TL80sFkBx}Nh4}dAjpg!~luAT2`YI}G0`m?ba!TMt9=p~zP#s!EF
zS2r2|F=qota@I;m#d)3e*YOW;90HgJ?@vcNQIU>H%2$TwNI{pXc8|r#hqAVt+KIuc
z+ExB83znn}5ccPSi0CKMK&4TX;<!px1ZzvG6xUNC3&@mPh0d@V!6w&XEaOVOaI4vp
zWqA@)(<wZi3~Dps9PRoNt~|+S>YQasq1O^i<8?9cfK$ptVD{icfK#9(c)GM>TK~OE
zpB=_sqFxVw%MJ61{`5|sHaaF#(YpDOjGZ*auiJe754@x;wmlNq_zUO4uN95R>tChp
ziv}03{0sv}jCX+(r&!=KoSOM49F=o3p;m2ni)?}U6*yF4QGt!QNS1*pu#o?@gEfub
z+P)`>ruMV~?pmkNR!&S7@VL)c?Ul1P%LEsf7ftu0#O|QSTbZQ?I~wY`N)d~&GET9w
z4nV#isF$@zM6N(3i&bv;qvaZYf)W|7g=tU6v(nIzdb-B#Emg(sQXWLMF}Qq@s-G@`
zj8c0FGY4fD{oVr?a|A0DhVz7dS>oLQgP3t$J$=N6W!vZNn|qT7_lkOodQ1xO`h3|B
z?&|}k>G@NPCj1TijvoHXS>6>h)g4yjJ;4U@gwKeK$|a>PzKUTo3NAHGq^rzA8Q@AY
z@aplOHw4Di2k{kPi}2b@EA|Z++hdfk6n7quMd(}#xKvJdoBUC1>~}VP7X>Ai6pE<)
zJtW#WA2l!%LfW*NWSm;S2>&rvLKGvnc&sGvKb-bfOd(4zij>T;hO!A9>FF5|swh%W
z#8NB7vZ?2g3KvMP5(UuRLs8zI3C3T|vWI?JW52i(BBEa`k<3j>o=~v-mfx~lrcK7*
zA{s5R_7hEd`0&wAwJOFv);Y&_BlYHZZH_j8_GA{b=l@#wT(~nK9^Sc2r5>Oxb5Ij(
zu+-0!i(#{0#H~8~4B#vIyLUbh{Ad<MW{P-^;T8j=m3id3gFqUg>FNS$HVj5V0%G3)
zZs`0=MP|8>WUadc(7iE(*2IXiu#TUrQvc$2R=&`Rk6UTBG4lWjBrtl{JeJQ2;;o}E
z&$&z!<I8Y237Lsd`i+Xl@WOntPW5p_YHXawz~>;kVI&(CYLB9w$K5QebVtM~c4kId
zp(_I3-D|SW4vFO66#NRX*DIMFEf6Y}Kr7<k!hufWm2L9}inoE1cK(-tLshF|cv#0D
zeo1dtgLo~80gH1{#%yX;3G700isD3PdK|4Q6q}8}k%?ZYwv=9~p+O%D1AEDETXugF
zSo3agST><j_yInmNJh1SqTQ7;y+Idrt{DNP$8wO@gUINI8n-s_izqCdVFGQ!iGa=z
z5XfqmtR^}{9Cqqb{SBf-Cj|+gK#C!3Za_?6H{8`Yf_nRhe*HqGPKnVJ4aXG3)k>yQ
z2X$Z2UUL1kD33)M!$dr4ikCQm(rD?|r?&bwyL)kLnsfkG+-p|vScej5@Psk<EdkL%
z>O5qgFTTxwV;+}>gzf2X8t4yi1kwKIU#lfVTN;fr=RmVGx%FE{Uv8<=1%t4Z_F&NV
zU4UC&otN(+1+PNttTC)v6LX*f*w!>p<Da#BN6Y5n{Z>!=hq^x{n_n@)WJJ-?8nP|P
zPM!krlKtISskdB)1RDmiX(U)>eh5!&1DiFr_cHu=h#omZiFT;tT^;f_6J9rD3$(Hm
zUxpnB^GX-&<o9<Zry1`{MV}eN<_4)!u(Nckzz5VBfL8gZL5@c^`ZQn7Oyg>UuVp&#
z!JB*4yCf27N!AnH-AZTpKYk%oG1SCVuf-gZrpKR!N<oquyAPTSab;!kPSEpN<31Y5
z*sNuO>twe`-BgAHWK5wI{*x;?ZH2u4YfYOHlu!dM>o1WTVk%qv@PV+<C2PDl!xSnP
zqHkh$<Yga^tE*h)XEZm*H}<;@1LG|d-o?bc>7ze{?%<Fmmhl;5@7E(H8c)}~3W7J&
zo!KQ!lzI1!>K$!Tl$&45-EbctGu77wO!7jDV40ybve5S4{eaBp+GaBasS#6=Q^6}X
zOxc^&-NW|o-QC+RpmMF7aNkzSGzAmqFa1qP7D};0U}INB+h-7Wv!5!yE*W2)*+Isq
z@hg45vG$}??7UY^!_K?e?WdkUzV1y{5F%<kW+J|D#QA<4d8_d0Ei08LfZ&1N_?Z##
zYk0pIs-r;AMVuBa+(-*8o0h=yJj|ZxQ%YFbw8^B(W5yX$R!$46GCFOf34Y>nkDZAy
zYKYLti#?q`zMWUzeO=*!=48G?kn?qucyO!o8gQP)Q&F)T9bIVLosZIyzM>8!gX#-<
z?(uz{6kqubwW;3HC}#N%xVbE(j^fWdN>iQrJ;e~eQ0k<l;(E%lW(=D1{GiYd?)PY*
z!fozP=9wI8SIvX6{%8`%4+_v~X$HzkKNTHGSn$=vDwVwqH~1KTF>YRB77$kbB-_gQ
z?^2HSJnkXRHUmrG;5$IVda)xh7rGHA`5g=CQya`eYnO*yJ3b6?FSc<nNrfGk7T^*D
zauVO4boiMzWz>DBlNXrBsju+eEg2Lgfh!t)0c0u&n&=uA9R;SHvr>kJec$T}8UCv~
zM~S6Qc`R@w+xu2kd==_{LZxxKTOOs~;aD(X5K!1c0`0t)ief_)-Z@&DJi|_W{Sz4=
zL1+~8bl!=ffh^kJ`;YH9<N6)CTIi)%zdVB<#R^i?bFG&uH(q5rAn$DJo>MFQc@&%1
z^TM-pZDJx3G`hOvu@Ysz<UNH(dtGJ=4OL)x-EHjD(3x6qohPNzY(x2+@H8-^!_1dw
zSFJI1GTRDFwL=+9y>0@~#(+JYh1a#b=DT)wr@E5UPoC0XxRe1w5BW|^3b8Cof~uC#
z+<59%nIgrW_k$uF?ezqexL?wAig&FQOKo*TF@!gCL#~H4ZwT2<$xUHWTm!5o)fd$c
znyHAu)sG(EehD-44G4HXzyG{nGsW>U{G{MW#ozvA`<{|DLATVusjTuCja@sRXdd-^
vjb8Csqi3&k^V6cjzPq-x|Nr4a=j7%GMqqtK*t*rfX9>`TJ=G{zw+Q(kKhq?@

literal 0
HcmV?d00001

diff --git a/docs_new/logos/cisco.png b/docs_new/logos/cisco.png
new file mode 100644
index 0000000000000000000000000000000000000000..31422734e45f9c2977e8730bcd88ffdb17148dc4
GIT binary patch
literal 5784
zcma)AbyQT{yFNq5&^gjd4>1fd(le9@(jbit9nzsv!cfwDN_QhlBhu*5A}taU5=w)B
zAPni9@2<OkYu!KYT6gWU&%1WKXaDiM@B2La#Odp5P*L2X004kW^S+uP-p=8>FBu8`
zE%BPm5pRf`m35Q>pdp3w%9a>E=C;3Yr~?2&`~VOh1ppWLsqifT2!H{=4jKUDvH*a^
zJFne90lz?E_dr7p`1@}w>Uy1upCR?v)KMp0Bc)|#mND1Y6a@gLJWVxa<7e~x`N1J+
z{cVF`<CkL{JmwL%)p@K4qV`B!M4!1rOH`mHmfLRzZwYY^#Y{96HaDG470ggd54zQK
z<=G!`s}Ffbl!)#O78K~!%$raPUnqB*Jnnha%10LQ@R!<?r{B+H!=k1ra`yJFdqVz>
z=zh<3-Z&pskUSXaj52ii^Y}-!^(pNq85-&ilz$o;Nq12D_#RZ#<+lcu^09ks68UWS
zYK*c({X6sd+H;qJ32V0v!y)OF>taeBN;a~3rJu7)U-q-ZgW_HF#$U9~P|@30HDmMx
zcJ#n~9&mA$AAO_c4vhwG9)eaU{>U|gaDz3Mp9cHefo{1zrJR>ApZu+D(eBf!awk7c
z@@n^ohgvWi4nZmgn{sPKWc`m2gz_s}N6g3383L%j`*7A9S_VtAj$M)cU$Rd7weMo@
z^SZ0HD%B91)GP9u3i2ti*L~$Iv}fGS1GO+`-7B3Wm~v^tO5GBLCEY6}u9$SexJp@|
zZzbKM)c(&}5`LqRact+$dFJXyZBEG3>Xu0<A5()_G?cBj!E%?CmC;!~LOnnue!id?
zU9RpVO8#8DxeF0x;u}BD`FBPlgAOV0Ib~^P-153i1D1(|n-fL+g@Wft-QDRO&d9<J
z)vNBnYgiY?z)_yG)2o~>@^tx~-@&#xBx&2ZhUH0FO6x~XCkT6ro@V8ZCXZU%S+y>l
zkg_n^FPPnY3cO7lr1{TMFaBF3`^=a_x!Ik=vR8~*P};YCu{OX~W9oD6_E?V1Kqzg2
zg)tQ?+4F?(Y2aShvNi#8g6}azBq5C{Ndm(W8Kpc<TM7x^<qC67<S!wuLE&=y`NjA*
zWr0GubC!bbxjFrrhJGVo0zcFlaLpOxFDZY<q(iARC;dtpxqbGuB%zvT(NKzdsG%-&
z0<yfBvC=`e#cm*S3X>h4JZNCybXyouEa$7oCQE`iab?1j>RV(<!W1u>`q_a<5_gA<
zi#4D7AwtTpd=vVo`{A3nD&K7#OIu;m7sdo3n6?qz{q#<eK~^J1>TU0y<Z!HgygHhQ
zGk?OVNN}$q_Mx2X)I(3x^f5B#)Cvndw_Nqrmn3X)9>pbr#3jZi#JQBc%<5c*CqpIK
zk*-p&@B9n$zX9+s=Km7_6}&Qm?+)J9y4l8Bz|Xxy1`A9l<sO8^mFb(cc~q~*+YqwG
z4vq*+7*3T!%TX@Y0@V$rW6ni}a`1?ap(O!I_lKtMVWSA35X8C%l(f5vU-m9H<SL|W
z-5R-rM=^oIPmz1vo)(Dl9tx%<^UUZv)eZ^$Y+$W<-V2V}q{Fd*IYB^=6b7}G@`Z8K
z%xubj1+%|tR&gFu*s5zsM2B?0P^D_Wbxb5l-oiw*)k$-6f0MFht^_I+W_DAbf_)6+
zwd?scI}R4UK*ASc>LJ^x1$68dGT}E)U8wKC<E15*(EZvrk@;pqr@i7ggE{Od>||L=
zYKGVOjWU*n!M{{hrzIe}cCvHipG`h(RkI?fFio*-n=}J9n-D>@dW#VINsUG_tI9up
z1N7pd(z`Nbe;O%xEvg+<`Z@fwp<?Qzv2jwhY^|~(+?AOoLkx5Bz0}c{x~7=_q}G4u
z+5e>0zj>ykM9i#{_6fwLC%^z_*aQJ-PD--967=)EsEcOW;lwgcV-ph>E<Adm#)qM!
z=_Z$gZfmWed!{N(@TV0<SzB~fq)}>v!gfA9xBDkzwc~aZS@XA@qxR!0bDAsm(^hKx
z5XWr&69^wyb)b))SDu}NUT0S!zj%cZ2)ISW+uj&t-#OGFD_WKL^<MW^M7{W=s+oRp
zE9<Svm_NLrC34}Xf%A~;wT-ABWt!P@I$<^uR!`<M5|NDR{*#7&v6vSOWfsvFLJdt(
zi(B7_R?>;`_k8dCI&PwynADp+t<W3@q4jBhGCa4E3sP9tGe9gbFPTYU-g{m7uvU_8
zp^&Fn_b(K}v^&O_XF9Ro8D|H$kUICuP@(;hkN`ap&IgG+xEsFPP=>G&8MPnmYXLof
zUsza}`YYaYDn&g`XCaRi{S<lUIrpzBkGrV8qkAbSaV$V5Csqlt=Dul5w6kbxcw*@k
zl=1e_c19+{4|#6)o#7WY#5mF~*3UH?5+D8D>b}(}{$MKa^NNjxEi}9Z2+F~XKPcRM
z0r6vAF7smTqa<sh45z~K>!);&DJ<M|)vWmaQt3cc+g}<x4m?XBx6F2lcJOms+zc(R
zG-0-Iqt<K*CSkoBd6dbZbWMTrO7oOu`(yuoSPe0`5ogKAHig9zvs&unPJ&2ATi7}i
zP<?#w85Y1%8Fc>~8p(fY9w-h}xhvIXKYi+EJMdIkgnH@mW#<E}e)lfzy8o2tiZtST
zea(hFJU*ez_O%Ape(Q*@1jCV7-Z!0<PF9<N4}wFI+s2NW$;fTa{^?Z~tta5~<Z>mQ
z3|HiqY{q<=<SaZs?BN+m#!O+8&-V5w2%yFmVq3(=<0Ly=9=&;0U5a`f+*1kwp}(g;
zi-j}CECCc<HXh6{*UAxPWcOT75d6beN_Vl^T2W6b5J6#S1?uTGDxn}RdWZ<h$XLW$
zHzs}i+#@aKoBsoy=<u8_p;O$2Mx1+IMUrUJpKy_5_edex<O`)9o@h!ZU8Md$W)#cV
z7YM@_ci!f}7(%Q2iE-{|u!JzkmkuXLr~$A$aBHt{&&`RDW_sQK6Js*T`GJwzim+Tl
zr$b;KCX}ZsPg+BrobsTZ5$k_DgfX$08%$gY?i<9)P^@rE1kGU)zJe+aUh!?a?fQD(
zK7uRKsUGjmyn#K7=~YM~W5_m8NAaN|x?fb*iBiL8Z*!BaCBcJE9Gk3Y$KAe}c*^%S
z?LCNv5w6Gq4CGcrFkw)#q<0k73vSTo?ZQA;mvY|OJ!Oi<jL{L45Mwhz&y1d_FcxK&
zoqwM;!yVacl8eyJQ?BF@J@iXC3G^I^n#!4eDNc{u1(fpFjFRboEFAG4ymgmSHxDiw
zc#>Dwjs1g8)a_14COAoRKe5W3jF3o$#D)~avVdACQ0A}n$Rf6QwkNJ#-d+bQEn59l
z7@W2oM|hkc5dALKw-q=i#!AvLI0DJV0OTi)-am|<q~4l2DZcNNHyuht5IC{id~pn#
zds;69MnZVCt(`FnR@=}I#);8ZlpSTs!IpVGwLvfwbq?B#$r(ASZ6}z#??bl4JxM*y
zy+jvoIcvht)PlDpU*YN?w5BF1%BwU>ym1{_`Y9^yx&2%av8&a;{4yU2ztGQKzL=u`
zmg&SGQrF#KLS1z3=4B(d)~f9`h+p^i%}}T*^?N_3M0<BLd5-R?s*79+hL1D4bSY97
z2|qAT<Sr4wAH&%;b!^?_^!MV!-kH_EJfrnI9`ZW6wcb~cc%^!uGM6H;8J!-R-lZSA
zMvqxZ2VahxGtMHN6BF5Lc0ScLG*3NTB(P)_VqjWp0ml1WBkGew3NaGIruE69y-}_T
zRA}GsX>*OAw}F&GvUOjA$hvpQanwzX22BK#(0v<bJFo1nwoy4y&s_@NvGuIS*sIA1
z=mMTbWv(u*;&%&g_gWcnw3iERqMd?q7IvQ9u2HP=zYPIrA^sH1w$N>b?x_73%ttnQ
zgN6DR@G0-xZ+%ckvF9Pc)ED8|)`_RxC*L9%KR}+dt`wg`iI({1RdyXRevOVKW9K8%
z`>uu=|1OZ!zr=|4b+V{+0fm}srI3?`h`s2)5SkG~^Bw-81y~+5&Bmc_(Fin);Soxq
zO<XW+5~M&AM)vldAY%El#=@1SM--C6bDEf?->TZ!PMk1m6y_6AGq`XOssf2JlXQIY
zb#U7+6iv0cF*D|kf<*k<_T*U_eCJ-lu|B4qBJCu{<BAhrZ&90M40Z^K5N^h=i^<<A
z4RPUQ>>HLphW6-P-|P@i<QnqwBraezU)b<CeduML6d(zTL%IZbcZ&M(r8bbQ9BHoC
zUr67TKPly1?otO?*v>VI&M2HG(L7OG7=zW|*l)Kp4Y@x;-&9IJ`C1}V-*`ye{KQl?
z`OkA`+xOV3=8h{iEz80Xqtkh8B*5t(Kr~J)97Ks%NhU!M?U?f?x)diU2dM^0WLq2Z
zIWK9jr|GwxfDqdyY5)pU*I%vb@k<BEt|w%j-_bPw#c%6<eH8uu=HtW{L%&ZXoGQER
zHVHnZA80lf9?nA8*TQ-&9;I)Kx<Xj8z6HBeqk5-v(=z6D3giukpfmbsLAL#|Ay$rO
zknT1RR?@tXA!m$AGfI1;i1<nB=Bn}U_`87!{=4%6lE_VG2mepfjb6N_Sqr1D=v$}2
zM7=~oMB#5}oAqc1Ro?L(8ap{tW&i5u$f^1+w5pt4(n>)`vTt<qlh}#7l>n;BHROgN
z^Gm(I;C5-qVEV<cu6bgx_@<p(#NdgdasQv2q0YKr20+VOQ_$$}Rg*$*l?wNWp?0XC
zr_9}wC|DMPZ!*h<No-Gwu=p|>86s;j>xjrX)QBO(W!bsjh6nwu%?nT?v3}X7e<qC^
z;L?^B|F(ZF6h6zyqU#iyO<{trowVd9lrd^B$Sk(~(oxP^-zr+r7tWP+33w7PN8(cO
zQm~MT?0qgraaDcBsQ<8CPXTrR6uowc#!O0Xp281<#loRoN28^8+W{9pVefW?cVZi-
zyXPaAvOHJ=c8YnC$%ghvQK3ZDnnxeU?TRq$3PM4>^3h0tqvhr*#!}S?3rpIv(c0y8
zsWkqaGGnO$FZfz!<USIRsg_#%=H@t@fF&D{=`TAN$W^E-C5l@(H3YUf;?RjCtAizb
z_dlJysMFo2V||@<IPI@eZ(KoFW`>B&Lpt(~UkFKr&sGI!8ynRV5Gwz1-14OL5uyN(
zrgnV8W=@{<C@eipom9J$rz2Y~Q0aZ!+@C93Qz*f6*Ua@#PN_z90XR8lDPA|kX_W#z
z#@!!9YS1%B^&JUIe0-sM;Z(FhAg*KL0KQe0CD!6$8|T}f)@5eMHfpwffc|LL^Sg(M
zHFcf?qUK!O*AV*tuf5EG0LcB5{<@eBzZGxgdt!{=0<+JXpp<wv{N)9{nZsIvxTo)h
zKgwUNA($h5-<}{ERsE*<0U}7uwjvPtQsDQh8(J@7Sg|4CmKqf98{FiSdBzHqGY!K9
zEb`mFCCqoGz}c-GzYamhbo<j(Yb1(~z2#7S_;{uJwv|n>V(7F;v$Dhpebyw~*Y-%r
z@(JT6y8;u3Wz2fD(uFSFf&LCJ_(y!=&o&W5k{(C5z~9Yy2sPOaxf+vpz2%hqzHEkm
z_x#=b>6hXiRBN8{^AqdHiW}B@U2BV4??ttTC2ph#-#4g_k`(09ePnkZH&C<{E87w_
z!}Ko(SbwHu?XwvqTo~`(zP~FX#MTwo99p1GU)um)q!c`Wel&BXcF7UvKN_(#RviZ)
z<=M4w-=^(^*W}GKYO2-5zd2LO74Q^Ylr>w{{rM;U&5}xAhI7zkt;toku!+;aOjPLj
zVzvvV3BUGx@3jT)9b+~jeNmF*{WkcwdXM#_F>t$hKCKBmrFit(zOnX||A5En2_d8j
z5f{E`F-iB~xQPV<Y@aSt^~h_*Strj&%jwd~^Td91eoxA>-ZR7C+?K{1O2#VAek@V0
zyZ&_oM%xa;(o9!jpOvi;J9P{f0$V8h_|73W8f2apu?1!#I@{{=DD;nmB=p1G1c=sl
z8O$`2JN-JrHJywoEa$VwgWI%PJ2x6yJKlH38A&|6YpAO{>@DydhMCAG@<&~t?$rCw
zDp?*9s9tBCp+fY0OF1_vQkxS{xL`8>WLClbqUr7<6YeH6T%<Vs5We?4-f$w*h2~Lp
zbDB13eWk94t%I!&jZb6no$cTm3A*O1VyEiBP;3X>>gJQfFju<)k%||9^MJSYEir+B
z%U8I}6j|zzo)@an8q`(~O;XU+%T<$L9Mq|tfympBevJTgFg9@g_Z#^rUMi`!{$vL3
z7{$Xevt(a~u*)Ddh?O$dj6~`ATLvvlYSrVkYfV|dj9(Qk1gl=~IjXfy$LMRVWzJSL
zN>2Bz>Eg$6TS_qFu@{Qa&`0aJ?H9EN8+U=SM%8XS4TgJ+JY!RP)txRdkDeFW@36V(
z?0r>Lhh~VJH#}+V;&vxKR)Rgm7D=EBh@yMbxq58M7yOhFL>c)Hyztt+0ofWA%c)~<
zVgyn6^Yu~Dihx^RAz%;6;!cheBq*kph|q2#{JAH<O3JZu1|HDJ@<!0T&F;+v8K?Ix
zF@DsmxIdVyu%I4X&!qkIHToqvT8?n)Qn*lZp5*AB7ph6a^ZwEI5ju1tP&qCkvtCo<
zV|U@!{NYjR`^O7=743r~C~rQDZdKL7cMV9mU$(ax7GS_dMc#yYr$*?pXMWMTkV#zJ
zzHY0ewjo*Zi=O788%~}eSX~hZVm4ocLmn?~E@jmWItO&T+HcMTaFSBlzA=A$*2phk
z-F!6hErYfLYF><aI#l16`lLt$aVpf2SB>45@7y<DE?tNJF|vc3oD&1%ZUJTcRF-9Q
z9y;fLHw%MN1O4lO9OY=!vsC9hSs%lBpS~}<`)rM;Blziq<`f;ruwIV`59|DgtFr2~
zs(2rIVRI{NM>J1WBe~G{stxBCBRO;^&KAtwDd0mm(NE;BSF^*GAzp_fBLnKCZ7g$f
zFGoC2#GKZ4CtKi^2Bx!4Li>=(f%y@46^}Qvyljf53@+ppd%e*|74xP)YPY0RoN6g>
zXk@fCdeg7*&950N4&3PWOS+B*&l-RAE?_B%;9A%bbR<T$oUXBDC#5&$UZ>mVQ@Hua
z6&rlJF_tB98ZTC8&jL3iToI^T$&;_W+#l|Eau)LD@+;;)@KaL~)+a!s(1Us+$>(!e
z1=yIz@2bN=;5(3+$Q<t_gFQOQY;ctfU%sDUVyC>nTFu<r-LXKzTq^o37q$gOe~tX-
zbGi6s+&>0`={f~SP%8@8o0;0JGlC-kp1xw3H3eDRu`dywEldtwt~3YyqwZxRthJ)_
z!k^tn{m5#I2=5NHNqHR!95|QMkjV2mJy$>73zeN_wP<=F9xZv8)KcMkrk2`<yy0qg
zhw!u8!$ASyTdFG!U`}$@UO>zlO$$m7eR8A>9E5Lsfv-mCuNLEWE06D)7gfz*TIYB|
zB3K!)3K`_<L|{3eqEi_55{Rdj=lSQzu(RXzPj#56?|K1+EyoQbI7c~A6P#$j;!dy#
z?$l7z=$;#0fsy69yVi~4J{1!DB|Q`7zR9QgI*k8y%jd6d=I>zZ?<i;Q>xef193~+n
z3cD*RfiQ-{<?xR*0t$o4!C+ODe$M~P!NbeJ)hY1*c9{P&ynuHAG}U$0YE{q?{{=!i
BuZ;iz

literal 0
HcmV?d00001

diff --git a/docs_new/logos/cohere.png b/docs_new/logos/cohere.png
new file mode 100644
index 0000000000000000000000000000000000000000..85f6e6d9904b0d49302a2d933181279e974a7c86
GIT binary patch
literal 113536
zcmeEvcRbbmAAg-FBXP_^=qR#M*UCO+Mksqz%FLEMI@hfZWfdvAtn7q{aJo?<dy|p9
zPxj{b{;2M~<#YP|9>2f8e_Z!Gbl#uW`~4cv*L;5jTvAn_AY&xkvu6**1w~o)J$v>i
zLI08L2fumH;CUAOZ?Chu!udVfb<E>?_OR`_AbU>J-FT{p{Be(t9{-lx^b<qRM217O
zWOCukwvF0Bw&onfiEZbjI-604o;-O)LgwyoMkkt(KqQGpO<cZAOH?VGmm2)pTrGNE
zuey@C+O5Dx<-~P+_imqLneEtfo1^wx&fvnI$w{?SPd2;rmbZN7Bs@s2Wx18&EwV<w
zd@S_hxXnsLLdAx@wP!B^<^vh6B5|8`6#PH#tV5F|Ng&{setRnBJTZ~_?Ky5$0tCRf
z*h1&_|85I(REl)f8E0Ylx3lgRx|)gL2}r0;YNFY+PUZ~29PbBN1eZPG{h<>s?Zs$5
zeG^DSm<UPI8^ZhB^FJm4{rVph{>OxG_Vzzb_@5^HPZR#}7ymPd|Cz)8%;A6L0IcAD
zO!)t10_HG_-`bHhr@zM)4g|+MLaC~qaqyIt@>y+1j{j9vOXH-(glc+kIYJ?N)hxZ@
zeO2x;ryHNRgp;b6*dMi5y!Z0!U7s9Yv@#1U+Q!G;kNuA*v`B_ll#<TUI+n-RY)g*H
zAjdBSpAC0n4y=~kHVAfkPCXuHlIpEc=jfHJF8bJ+Ia6@{t@(*V!t6c13Bt>M+KoYc
zJE@0hB?6H~r=h2~BWF*A{=Wli>U$5beo9uF58EeN_X%kspCrIu$<98iAfI&iL(dfr
z<3lm3)3G02l>Su>GzL*~uKlf-?}%SWco~)Gp^+F$ZB!2n*H&8ReYIpI@AS(VewjF_
za1IDnrD&-X7r!Jk20-Uo(=nYK>UiQWT4yIeKNZttrct4XoTWfIY6owOxu|_TZaUG&
zY5K>%$z0Fzudl0SDY=H#L^OsqxaLwdg4Osa-ls}A*lud{v^+P4V9Rt9WO$Q)2gJT*
zM<#Sd?$)E@%2xevEQRY;!s4j7SQtcOkBUkQt9!F^|NC$vl7&!Qs&j?YrNy+ECQkOz
z5KZ;@w%g-O(#<jgmJ!f~yN<JkTGJ%`vBSOuVvL}mW2K0!l&4+N$_VXjZO!%(<fwcF
zyZ&=Cu74~CeDJ*7Zz{W{m{cT$>eClxEM!?z`s}G(x)ou}EME^j>%|{?F)h_GnW!CY
z%b@NM6Ls~iu65Q>8cm`d9ge9BsH*wWJfcOHzr(k0;Jg>H$@7(xYi^q%ZSISCMOGSh
zp>THey6kqa4aBl=WsqfNuM6`*JE3KJvH4mj#&xA3Z2p<LsK^XyOkK%xvud9v&QXc_
zKWSvbMTDDtMPG_y5cbXK7x@ci(H6_TWf%9DpFrP2my?hrtqpV3o`a-cNj}wUbYyMp
zKz-W5gNC8bWo+t(RHAhu*blURMKnLQxWlP_I}CGt6%rp3zMQIvw{MwXWtV{-(ZkOc
zI<Y0*bPU>Jra!KY$4w%$iFsowe4f}t)IHGUoUBA;WS7C+8i(Wm<tK}N|8`2%dZbJ`
z?d+DnGUsQf?x`uG28d=t2@#Qc&rR0iW01Dpr0x*b^HHkvIG&o%hq{9rh`DK9nY**6
z9}4`!9OL3#=kb(!d_-KBxyx;`0KYjGrOdq}2^(rE#jJpOb4W?X<S<9YU2m)nq($6|
zs`;X!+KsAokd=7DV9_Y>-_4OQ>-1-1H_EUs?;1xMnWP;ygSID&#CahmV_uPyBz38#
zvq1}7qH%Kd9wYA#%!tk9V+E<Xq=itn)2~<-cX;#9CXZ4p&W1F8a;{aH%3>D%lEajj
z=jJ+g?E`e!7GS{YCgop9sn{Cg5e(DCQy=$PeH~D&wG|y0pdJ6XDVmn{lO#-ofOrFa
z8b66J_k`MQ;0hMivU205A!1+I5C|?ce@M+e^-XKY3vYC6+|GM7q}26@-Gmug)3<K|
zDJdB~0%+~<W@m$@r9@IJTy~!t-^^Bh+;(c_vo9X5@)Q~{1m|0bzJ*Tjvulw6-#m%I
z$?52nJN7G<@uDQz)xF(hV>JKF4f+N>KeiZ4w{lh1Q%AN%*k|URU#@*$S#li<UJ!a#
zBzRW&OtT<()|fJeuXQ58KXqt&pQt3ef&@D%t|ii+F%a7Nk4F&|tFtL=7$}X`tIOJ6
z@thgKRvC3ZPXYhj0HNNgGUhBKVBjZ&Oo?k8Q`+2prNE=216Ih~v`;MNoc|sW$jlC@
z7+)KfQ=g_Me%j|9Ur$qn6!e&B@EE)DW1G-pjw41*e8w+uvi7)KsLN$npTaVS@O&J5
zfBdHh8PiqG24D?qMBC|e%D*I*irp!L1V;*rIh^6KJ4W?^c1L07%l2>&3}8F3Kc9(Y
z;W;j==8L9MZP(umLGj~_NAgv`2LA>sf4h<UBBaukv>j4XzH?tN;^q8+(WxC_sJax{
zAIko1J8PFrO#&AlX;88}=_zr;f02r*R|Jf+)I+ni{_%tY;$cG|&WYq^JMO8)2-ZR&
zP}zqpoler`C($Q-AU-6o<b_I>ww&@<R`B$up9b|PQx(7#q1n!z&3vHNJ@oXDZPEOk
zQn1CSqFM-6CFvgl?R-FgB)0n~y8Q*G!|k<HU+a@LME{->o$wcU2(@u>7$A7*FWjn*
ziD%OV<>>cUS07$N)?=}?axsY<{}nb+3M&k=NdRqLZIz#Ug7EfMT%6oIbf$CQOo&Ix
z(3!p#APmx5X+&Zk#6NM8;6X_~r2P3A;MC}?Yr)mg!Lq)Sd<ARe1uvWRA?1q#%GYjp
zu^c>Q41o#zIFaWRh!0`bFlq><|K&W=B%elN{OldgeNChfn(YUxB`%@ahE*V=!8sDY
zdgD-Yhr-1BQ?Ta85|n}vN>y}*gZ$@b?7_5>c%M|*=o%0;*y44?+hwQRl7vXh14R+(
z4U7X{Zy-ht4i4_y_8V?UPNObC&CwypTTWRjlS2pnSLqUX5%G1%a>t=y_Zq&%HJ$1|
z@<mCh#!SG2wtKTag73cI#(c?mA9q-JRRL8=34E8fRt3j@%m{j1f)S!{6x-Jg+<}2e
z{miC!Nnkp_D2yrqGcmpy3Gj87r?}vw;=)HId`xwQs^{6~G)i8`@d4BLkLF=sk~6C!
z+ea`989|dHQkPv1&eO0pcmYxOmIxdLx|3jwc&B5$mG!!Fqmg!1$>c$fZ6?_-&p;77
z$$rvtnr~mK|Lou#b|4Kh09P`As%dHT5OkqU*H>0P-rbLnQ4}pbt!~({kIyFTCuAU@
zI{OND*GVbIC??M902{RX5+LAidqqFMJ<Er<S4OQ*6C>QO1iHJ*#<Zloto|=cp#x3w
zJJ@Ir&;>~@t}fck9RRD&z5+Gh8gN`REBTEh<wGqL{+tq=a=zze*V?35S}lJT%Pn-p
z-Ka3X3AF9j=4nWpwh)$-AU7dqQy>gD75KAdhl!CM5-3$mnF}X=k|{VZ3sMp37w7D$
zXJ$V0PO|BxFew07a0H^r8?p$Z>SN!0{a8ceXfiE77VBJdrhw}w+Xc@!z)*$HJgDjG
zx_#4`H5DKPW|1B^yJd>A@!*6uw1>;Arvh@rO&Gc2XpqtEx^=F<YDGCw6mq<{T77<W
zxxQ}O%rD5C5%B*4FgWGHWF7GB&uXUEtf#EPqWM3Dr6a%8JLl|pm;ao`{UnB6bf7lk
z?OPpw!}U5pIXSQoRq_?`PpbFCL%!`JDpN{)c;$(r;}=Hd&R?tc;cEStf>(NKqxWyE
zmr47!znu%RzltDg1F=tm@mw7gw2*(Z`nU%sRH8;S$PoK089Q+BlWOMBR3Wz?p^Q7v
zTGT|FEzfYE!2W6Z*h0DWb$!6c%ya!|gW@cXE~a0jVg}@pvoD0gU+#p2|2Ykt123lA
zsZKigtUx&>ge`#}*riL$5C$K$xlVhxcF>q{l}o4_(jlX@_1+8r!q8i|Kq9A#Zl3Wm
zW%?BRjZcBXx{z4tK{U(vrye*t21WczHP~2!YPN|vjS;zZ{}i=>?Jp1M#~>ipmYz%A
z=rWjC%;^Vl5L%pg<GO{oP%8*8aT7#M4J{o4iBC+T)DLE5|7G-xYJwPrbVEaR_4>W$
zd9N<*%f}EzN+3+*-lW+BR(q1hh)9*qw!D?8zlV*Kt$5g$K{@*0_;JVHgZV-!1vRz;
z_BD%qK97aO2wEmtS`EHnv-u}aOGx>iXO!3&6rHP=tw_}WY$pmB^j|)r6CNT&N8Jss
zzw57|&-NKDcbzk}K(aNM05jPtR%-?pQ~8g_U3eCQ-;!-%FD}x8T>5_i2RyY*o9)Bg
za=53KS<03t6g=n#-9lRnSua7BMvSpFGfNdz3${Ux+MovR)4$UHPhJDI4Zc722WFx0
z@`ZXU$j23Do|y);hru>Zm9tHQ9hD%u<Gm`a?o#q?x+E<OV9#c`aW?KcPQMM&Bh<iR
z>QTe4Qa60>yBpxJFQ`GtuN0mM$%h>AZct5{oziaOb@<>1RAm6)je_uR^#46jgA8Kz
zVT7i~+pY1}<ugxH3Y2c)#7Ie!+DLmrM7t1=^B5dtVik5@iK1P#@Ij?-&$P?_%+*`C
zTw<pRR*IAn<z8u@chmR5HolsIsLvj<OB)77cAC}<rxROVL++ouYSMmTCE$f9#;;a%
zyS?(s>ZDh;GX+3zB=95(kB^=QxdGWhHSKa9R4)~>uPH|$>fW#We{mH<i?>ZlW|ewj
zfw$0k^Wi+utYbj4DutLH01PKTINjx8m#*mx!nwwou8Cc;^fc@b4xi_tHy`oiL^*}j
zKLdvOZ$cI>vy)45&Dm&F%vYbu{T1-`A~NWz4mh#I=*D=or|KBIl>+&v0st#%H!Erc
z5gF~-?FnTyvDvYa$W-&KaYe_g7zHliTz*!~F-{EsaJb(MM^9fhX^QLg2qJw3D#Zdj
z^8<Sk*Qq;7Oa9m>xAX8Fa=$4nk$y2TdKW}A$YZ2?^7B{kF4xxeHolKi0!kzU^ogaI
zH;bCBK@aogW3GklP`g5fhM3pO=AWzr+Kdm^*Ku>}EgWW?$!@%6{Gi+|oDA@O5>!B0
zCD5jp?RCx?$#d_qE2qx>B7CsemQJ^t%eY`=h_9Tw#7(4IU+)Zps8}B`p|w3x6hS1$
z_2Q<})F-97oB&#2-%6bIu8>dtX%iqp<QmxAw~r~#XZ|LB#qi09Fc0Pr5D8R!uY<t3
zKLf!|lbG{~s}zM2HiMF|pC2%W#Z4n;WrFlKjgE@e9s|gl0SwY;^{6TU7xIH@_kHb?
z7E-V)DT_0et3Oi)0Qi7x<86wHNwIk!D_&f)X*EbCffg`N-7xPMDV2KyuBzpFU+6pz
zFJGZ5_UmHcwoiSXRjSBaYM<~cpNX{u&pQuA&M&x?ZlMdwZureC)!sc6KUYmVD**BU
zgY_1MpZ5+R<1fTrjbH}ePc1ooU#8?a8kPa|vhu6@sM#FiabH_H-0N1`H7Ytae`47V
z4~jKFKGHi<?=#N1Jl$Q`5_@za5Q_9jlC%@efwPc2aXQ~-^mUs0YALp(S0l*g(J!21
zyJU=c-;cXv(c|4CE~R6Pct|tG56sW6C)J9Cswf`!$a=AYg7X7^3U-?j$<+UA&;l?b
z95>2Sd$`-r$0s+nxQ%x&;sbC{9*4odB;y(?#<Ihkauh1e)($M6xc*}nzzHq0%Mh!N
zIUJWW%)6Aq@3+V#4RGNK*aoFZTpKCXiYLZaCn&aAA}Vs5vU>R|i~>+%!Uxtg>`m6*
zmU=C-%!CA7JeS}N*f`hoI)D^xq{k!<pR{s2d>|Ed)y(9O)yZEJ@D@4%pje+)Q)3MS
zaFZ7Uj@(yGumDa{((5K8P(Ue6SU@AIWy%5**T4<#-CE$o=Wtq1!se;=c1BrTlLOHj
zh|gQ^Mp_?7Tt{{YpK)z#4DX9Uo#LtaH1bpWwxbKW^i|T`&uh2I$U&$Bp@+CzvIc@^
zhV!(b=-LMD>>AdE`|z(D-h)vm`y_0=Xc6h;BttWH4&s(^)FaYxJ=0Z<%e5`u2C~IT
zpAP+;cJ7cXGDtuwy9EiG_s!!a0Hf`|Qn$X4nH_mo4UNq^RoJg-hFk39jz1msYdr84
zX`(9&w<r4+F90^U<bVLt3oma0;xe9ojmvR}UzHo=Cq!G4ax6yAYef_REX4w$%)DVR
zLb5sJ;2LBXUvu&KadWHspa0dS+0gyRhz_OFme8k6;!|XAp_hSMD#ok2fRC~_VwuGS
z<a})ciw=_PqPPbWK{lkwZ;)AAYL`5j#g0x807UD~ir)rt578aYbd9h9w;j){x63mF
zs&<%|c_xPPRdEwq0B2#qH6MlcTy(&Qij4}n<5MZ@RuY-5U`Y#+CUtnVM-CjYOKUHG
znIa|({24u9u)$%vV`$SXP#H)US3l89j||bZ;amSDeuj?nnLTvN)ym51!vPEX4F(jz
zGQeYNpUqg+gat(R`}e9=akTc>s(LQT{d;!n1cIo^f@h!O$1RWx8KWIfkW%SB1<P+=
zdIWgyi+LK*IL(zT32LG185uv9Np=Jcfc8Lin#V`T5nMTc3%3T4r0qjb8i>|TO9(C9
zsaoY!qTIV%R0$dgx1?~_Qy1$?b2JW!-Q9!PhCl#h{VtlOmIHww6vWoDX&uigP`Vz5
zkr(z|*nHjlPa7tG12oKogykt7gia%t=Xmcr5^jMS7*dxx9QP}9p-m-;gOg|n$(tA5
zax7@-I4%hTMt9nMenZrD?pv+N{ezirbL5!80dIv}@gxbea>5ZV6~UBI^I0I|BVesX
zOzioDd_<Y7iB(Q-|4*t9d=AErp(?C-o=L#jetJoMQ6Yg90R5)JJ#)5(WN1yK@%>l3
zA-aq^RAX(na{RRj>KHmf7f{J9OQ$)Rg{!)@Sm<1FPnq1*VYeLPkX2m9xo3^GhvTet
z6Nx+_IPs56xsK@JePQdp;8^>1Asf50WH+N>|LZmZe_7bRN;4L$sa*yX4+Dy4veT(g
z0yNH^02P7)tcx|*-h;cM5VnS95Fd-2s{6DpD?1d&B0Eo143WJZ`RoLu^yKM3MAtGz
zoBSp#dlz@=FhA1;nj_lV@;Ggq1PKBt9#GNFNwp?Y;&k}3ay^8RIrEJqDk$8}ec$fs
zqJ=c{R+hZ&R+QAyI$feRIgn(XnmTl%ffe!YVJkim+lh6T^}BeDY!{REB9=*#!i`H$
zOBP0hNNE}gVx|w5=@8dx6lFA8<5fg`&!F~K3%~K$ZDmTR&^e)?@zp-=kNNM9UwRHu
z?2>UMDTsH!SBqnUv9X&Xu>rf~N#Y%g3jw$*-4E=e3uu7Nd|os<jY!_rX!b8dJx?`R
z&+}>fMSvgtnL)SEPe7Hyu=9DV&2{<H;+O&uV6|~uN@E_798{b3_8h8?m#=uI^vBCH
zySGd7nbSc~Zc6`oYhUnRz`p||)E+fwz^Ienzj(jwkJN<;?DKGIhj+VmoN7f4liYGY
zmb|^aj~w7%8*ph^(VgcK1VMH%$XS7VDjh_dn0%xw;g{V)#>QgUf2q#n!@#EvDNxGG
z=Kw;jt+7TVBM(0*o?>!hDLRIVQ;&K|0f(aQ2+^_!PB%=7%C_DP@-40tk!+-bYK9E}
z*h!*SDpTIp0$L-e0YRva`)@k(J+cEVeydbbxo)#qZXt6ER}Hw#<NPuZZAEk^Ro?Fp
zr?o}w;ghuu`#~oDD=0J+<H`kjlo?n$@|N?m><j$St6=qTi8Bq&*%aK;boT*D$+d8i
zWbEj%Z+(%U>6^+SSZ%c*-8nSXxn4<^^$H8(0wVPjW!{N2MQZ)eMkVII>{AH-B$tGQ
z2?79*jyp7)MGE`jr>3SFAGzvXOt=Qn?152{BdNqC(`)T@_Ey^5A}QFs@!f|87*vc7
zUP-Lf-@L2eXLSHG2f(-5xt-(0*P94-jn*1xffV-20CyOEQR$YVRc8Y$$Jh9C%KnTc
zS~CE4+qY~ZhM?qzaC<&%K7;-5_<g;kyR>o-MwlcC%c=`htM`t&64jXaUO!}yn~pUb
z;se9sC+T%V%q79)O>ovnq)%N{Ya8^0iC(|U><oq)Y*)$Qqx^9Yv_6_uR_!_K{&z3-
za0-lPzlA=AC=}7?^)&U|@?%gpx{Q$I22GpBd(FZaZQ@hR<x`ja{r9bd&WiT3zS=Gs
zgXC5j5znGni+A5%peQWNueplw;RO#EtPx_wtbaE}tC@RMA<3}R6^Ix?+PpTi6Dw9(
z9Xj@J_N|+Ndlit2zmc*LhdpbSRWS8a+5S8GOotJo1Jm~k*4$->BJeLCp#6X`ugx2)
z$bsk~OCvE-E6Pe(JqG*Nui6QDkvj74pj9%@DISJ)gTk6gDHj<~sFcx2)Jcf}!NERT
zo^-0+vf7KtA{%PCZ^3J5;%A@1l*QSg2Gq;9SS63DNXPVA`r6BwL7iK$szrx5pdq$q
zOWrKvo5EAvNQAjb1*#|eirGnNPln#%l=l-?_q>d;IJw&`0B%nr3gcWI#-Bm_vh&F}
zO5|w*6l~qi5&<B1r(w5pcPUAcMiEXo$D}aKjy+eid~Yt_OkoEdK9ZzB5myA}Z(`S}
zj@<G7G=D#OE_rh4PkVnh#cyKI)Mp#rFuNx(UtT_e5T+ix4BT0Wam*`H0Et_zk#R0f
zVJi3l+V1DDr(Q%TM5T#(o37n4C@#~Wptei|NFVfA2|1Sky-gO(yXhh8?Uh>Z$^ysl
z4EmMSM5g`xE4cVK^V`|Qe+ChGUIb@1V@NSVER!0ZbzJTAZ`{aGoRx(IgoPMf*?Uog
zG7(U5UosqJ<aj{_Dm2^;H;%pN1$F5l$uq$t%W@To?IjGOyImtF=!ju}NEs|9_)U83
zN=OzSf&CZGC>Ii4y*BaEzQ&`UpMBsUGaTa<UZLy8=PIAPe%%{)ewip3l-tD0!sv*K
z<@<Y2Po!Yob=w7#fv?^zvNn{1!oehadqa`sfCAdwNOU0(S@Gtb1%-ZNOqfwCx$}vw
z^M{8oz;Jv_9AhBjGox$(fVQy3?cotDt`uqjV$hVp%S|x{XnLs0#(ds4ZT)NY9P$C6
zDycry&et+x(l#JmL%MJu9M&RRDe*d8k;r}ht!pi$*M&}t>=qM%o}sGRdrSf%yy2$W
z?2kayqV>13h!dXVpvrwaHLjkM_a|b-n}6z_e>*lti#%!NgSq}%+di))U&mIV8^=K`
z2MTeehi#FdU%uoMQMHKeko~p9ytCT}BuS6$F1F|Jf*67~4D3}l(zc}+dARJQ7-OyO
zp~Q_{ehqL8I0Vtr0I1ldEd0uG-4N)~D8r36y)=iTx;!paM=*i>6n`_v3{EgzY+ozi
zVkr8&cWNX|)?DWjcz|;=a-@i2a|i)-ijHQdI(z`O_Tp}#f^KO6u1r2_OA;BToMpOF
z2BM|KS}n$;cH&pl>4$_gHm)a~&26`6l-|u>5L;1IZD(DM-Y#&}RQCrZ6%e|t?~hSq
zL%mX+{;d8n!UyU22UINDtrDMk9CQSlZXd9AE~GxzFbQ~TTz+nI`3<h^W9~dpqqm8y
zSi$JsGcZ62sM4==Fgy}Aaex}})2A|7FscA&#nPsE5tLMiy9(w%Yo@)CM%CT=xwrj&
z8yGY4p+4)Dk({S836a2wmwJ17;;kTfR!bjRD8ssIs!=}kfZMd#u@|M6Gx}Nr{A11~
zkO06zrdv~=4F=T=x)8N-OPRyNX}b>!Z1p_G&$4AgP4|m_4T(!EP>9KFS((jKUt4C6
zhUhXLmZ0J)%k8reVuj@}jbfNd;iYIly>;$%*#u|6s5(9I0Q64ZC&-_Z*YA%BLs&zn
zvb@}HIplH$YTNM5<i+5MA@OXI30|V#X=yX51rOto6eRNsL_dUKGngLO*aR_)$!T$M
zul(r<y5pA;z}C@*$pB(O&!=02Rm!{C3pHQ`Q-1;ie?bpQFT)@QGci)8@FI1A3V&9P
zqY<j6>r2!AcUgVrda*7S#_x!nhC8+#$>)f9oaj9P{F)20vi10-zW{9#Eg^`um)rG~
z!9tKYWZ#1G;vjM%dy@mkfOME9x#WJ_UWha@bx&X83>%zyDgGF0I{m7Ym?iR|Dt_I}
z+U+arJQ<1!n}E|x(dml7bAS0bVZ%N!d=^HHLCZK4_ZAr6=(j@IC7<i0`6}S3!b-P6
zWQ+T%xp<fTuAji$L|CKO#W3S7KefbbO&v?8(FKh`>*+h(R4a@iA<xYXpA%<)_U9d#
zz3xTaCf9t{)u9p3zrlQfO4lDOq;P*l4HYGa#vFndbdb4E4(&c8_>I`{uYTMns%l-*
zBFo8#Vn+`n_<&mIbL3$_z5bPvkgMvvrSW}i^Wj~EIOz1CRl?O|&)YI$-1Jc(RxW`K
z7y*xd8?`$XChMV@g*IwiMK*@`r?H`b_&mwxaD0E02h@$tJ!B#N05qtH)IXef32MsZ
zF@B55oU!@&R;v#$_x=6L9Kv^-nml<44D*;xx|b!AN`TC)dXVkKehly@lq|jrTmati
zdNM~0V3H&dK%c@)gCcN>E1kw&G&Hj_{XjD{93k_t71Zywl9_uO{@LY!KDZ>}`b4a6
zn$)H-*{2Lik|rr_?YUp>3Dlu6QX#258;YpmRKl_xtvddJA$7F3W2>5*TDKHPXVKO5
zk*?`V^@OFM*(ub9faZXksmm@M+>0>ccJRu3xv&Dcom}PhW^Gc?P4gt>@2Rq(*_qQi
zF?Urq=EW-|SGcGEoxp~nE$@7sz3p<YHR43WOJFqD|6r=-{hn<zuYeh%blBn_knJRK
za0OZ{$Fb)#ClnMtIuG1sppEb~aiVVlnUy~)7CYa4S&af_<Dh}b9cor~Mk9yb^UeT~
z_DQx-{bS+RiSBS@_&q|!$^8kd#*)XsA*#%#b%7dC2OrPe_trVMuh$k^k+bB}S?P?h
zN=(L3G@Yv)T^}DCE2+P7;Sk$bDiBT7`-7UIJV!=-HPUZCybhI~{84mSx{CLlVFl{D
z{W<vp85NYdb~~}?-p8sChwD6m-E1t?iXpG9?|C!Vr8-sSk%Nd2U|`FGfrv$Rr(K&X
zcy`4NIco_`x9m3dvFjjUO9cU2d1sBlSZN{Tczpbd`4ztvq*f~PmJ|ZX$;Hm=!7|t7
zA~ay6-?ecz{=vlI*ZwTfy|LpY%rd5yFFkDoi^$-ueNCS2gEEuAV*7Hf#Qt$Ea9lN`
zNOSRHQ1bVzYSm55@sR-BZrK{R_5woT`Od*70HIS*cYUln*Ea2meIR!4Y4{V0V3^0R
z-z#|CzSwOzGVTkAE%${T5<l+nrDE8EGG(Msu44SCSkoE!)z5KrsY$~8wp;Igz|5J;
zg6Z7|W1xU*oO?|5`ZqH8CC8JJRaoz1odaOF3$(YKc&=X9dcM4=ParD08`yY{EBl%r
z1LjM>L^pfoa!$>cLoj6lki=LJAF|W$@_wau;U<0QR4Qpw^%mk2>pKlI{=gW&1H`Zj
zSd9v0H-4>~QJ%s(ALaaG1XNc6^VBG@FO1vj3`#iDj42ozLfy3<ol5Mjl-@jWLRyVY
zYwp0Mk=ergCQcLY-_qFK!{${_nf^Nau=5Nh?$9QFP?PjcOKN5+c?X=+A%F2~8=|JK
zLt|d-O(%?+!{p&5sRv0i(=APnMloV-twwwUSQi6opmmt;^se6V4k?s3KUWT@+#=Fc
zFNmkw5Xq;!k9*|YyUdCSU+xnsd<LH!T<A9`f04gB)opHp*H4{4YCyf$@0iJhd%`1<
zh;EMeb=D~xVc4_&FqH;~L!UQuX0{d0inloEe+)9Xwv&P;F9Ygo(IjItY|;xRFW5eC
zG72nmX0^}O@!bVz$Gm)_8hbd3aXlF{Z?C{=$}L<wsYKr8jPiTrm}fvw<UXc%YkePs
zDds|P8?`<*hvy^(LR6~bahgdBMb_9;X<)&d9yuw+@wlZ?W5=#9?-ZmNks%1!m=iR=
z=&K@JW(VS4Z+er^LXGJ2TD95KhodrkLs6$D(^CXt*J12vs%B^2U)&6|vr(4b1nVY3
zyaX%#LDHn{HzX}}WO4l#>|W9RMTpfgQ;9G8+S{aWevY6A(IALQP>+X8*<akWUbqku
zV;6+2Ep0APh4q*(d=a8^(F`{8{D$R+Vm{{pW$-QSd{BS0cV*?h#sSJr6<LYoAzCm#
z2dCjY2)E?7Z}S+SX^K}bKk6es&`>EJRja};C*i(UjNxUK_$NM&2*K49z!)uU6*5OJ
zv(@BvFyGl?O2{j?r(I{Yx~8+(WEij<-t%??7D05Aanrf^t7n2bJH>Tg_X#6RI01B>
zTvPf{o_n+<g{qs%czY0@)v9D*rsDm=XC^U^q`dEyGK1prJDeHL$jf{^{rlmq*^5}J
z?Rjgxt^I}8YEWz9o}XUsoO(xNCQZv#o?QR_4|rg1lS^p2|7EkiouPf$;qDTY6yUS-
zWA+*C6b{l>8?k58Vh}0tB%_I)-Xx#V@bit%;MP$GID3J=Ea!bMm<L$MHCRfemGKdZ
zyaKBhFUk-mTXX$i1nf(fN==OWgkyRsA<-69KBzza$C<V2m?r*+n71pi42I$w>Cca!
zc-gFzXXt+Gsl2HlIoc((b~Nq5A=wJk_#ks;c=v@t8(X%F6y7|?u0Q4ob{t}2`lOqc
zg;n|(W$$VsSJT590=P#)%XR69V&9wX;@2Sz9(TGnp(*Zd@tJPYyQ%5rNKxN(`vZb7
zy(0iB#pl>QOakW@=_<qHIc5Z?;(LwK#92GFN>A%tT~%K6c$e%;dIwI06ZRqE%O}JA
zd_O@m#)=2(oMhw4R%$k+N;`a6Cg04?2BjYA^)LV>ed&6uU87YynC}vVuIuZ2{KU_q
zeIYR{4j91V=Ljoz_m^v3Sa(rUST<3OJ;qT4>HFlHrHVes18nkfOWu@hkSmerKm0tY
zJOHcYr)Dh%yVNBDo%f)*bKPQO>vcwP)5UsFb@d!c7cBN!UwZp4L~6JVdp2_aE@$*@
ztJ-X-?v(>Bn~~yP>CJ1P)6&W>qW7mL0Djle7!$2OFaI(+{armUGXR~^06Ik%=bSmZ
z_$N?e2D=L(XTFUV=xw>QupNTkER`-R#_$3I|ICx_a;a4Dg70bgi6#AhwI8zp6Fz`d
zPq7{9&%%OW8VR5(luc8`<?H2Gy!Fti{qdb*5G}MW>Lk5|!gKkg_ORVC81x&=O@@_7
z`7B>;*`b?w*?p2AyqQRw<TlBZq0y0$N=_`4s~{bY?yZ2e08p}wnRnx_`-Gn?t8;0o
z1*T(HcC@q-cvMh0V}vb+gGM?vCfv9DjO4PJQB#nxr!3efJA>AHVFiv4+0|h=e1lWw
z@OI8XzOfE6aA_x3*Ck7a<>v*bfEL~Cylpwkg5evCj45{f`j;AL4n2YQbD)GQ$Zq5b
zZsCmfv_{w!Llf7vg&G6B0Mh)Jk<z+X5sC0wgx3#%Sr&etX9p4O4jSWx;(#UAK3e-#
zFNM|JLjXITpogo!X6fc=sf(=+A}zZ#%1PMuyT|a6K(Uyp?$v;&C_)s)g_k%j>1Sr|
zl2P+xICL8X9~|j9AGv%FkrPneQd~FaXgZgaDO0%K1`5cNeU7H)M+LPTb#)(Xv;u_U
ziQ)ITVm0sg+SYX`I~%X*KY0qIEXzC%I3K9N_!|41diOlZjt2EoDRO1&KF%&L5h-pR
z_JsO%8hxE6bl>ZOcsbar1T=JW$);sOL$Wp3y#0GO(<cGsIyV6n#VW(+`9z*d{lP$8
zmDFx#cHRWFxh(P$89e`KK|LSGz!h8A+A_#@y$})MWft2DpG}sHN5y860An6@)vbwP
z?xniM`3mYt9YLlJ5<1>L6({`7?7{Dqwagq)h|Juw3tVx81{+~)t050j2-=_Ip0B)u
z+#nrCn%`wsHl2)!jMuRewhzMYJx^d^3Bq4lIpmFpD}L0=4H3)mZ4h?z0NvrZ%wAfg
zn#?aZ3Vx86JQdnZDPlg-+cl^Gd=tQTOp%gw5w1#eaV}=i2pTwqiB1yuiMH}DXW%F9
zT{w6d3m8MCEyt?#V=O=~9$cni2ZO<I>bA)iTHf%yHl79fnl<H&%sL27rbV=7&m!0#
znt`cf6H54`lW0wmO{7rB+)j5aG2ng^G+57cCPJ(XSE(VcCLP5SAAr>eF7G0Q9!D@S
zeraa}w#g-%bL7RLe%Q&vcXv|X_Z}AAK1s-r2w<35Z+vRupU6wQ-^>9hMzckTmVqud
zzjRL1jb29Nz7Hn0@VO1i!eexq&3Ik2{gcgm{0q8$UDLM!oAy-&7b%#!zx33Ql?VcD
zbvR)zq@q<<(<{eUB{xN5)7ntLpzg($^3Z=PeLv4}_rDFd3@EHKXq**;4(CLijp4i7
zpdv-a>55xfS;1KxIOM~tE)hn+buAR&MwtrM*w;PJzsphrBLdvQ(fUZBs%2;8He7kU
zGGTK*Fw{=>T(i!8=PF6Qmn{RhX~#{X-RVVCSH9NtP*uNcAq45Tp9c1jRAYX?3xx22
zc?jQaHS52%fuGK@)a!l;O|@mHE0q}ZTJ9sjpCokP8;hsHep+h6Ipg1N(nv^bEeKq7
z7@SK@RDDJOR+Et{7ZbJpysc7%#sUF7o~th{g&>&fyN|We&TD5Ou+uA;<c*H^QybEY
zUy1bX^MDv@6RLaIUo2K14apKbuzd_Ae3}3ZIFSy;aMD|h<nXt*r1+fN3zEOfSZFdU
zL_KhnUfu9A0g_-$s;>crJ&tZ7X(iv#!{|ZY_sU`!ABy|Lk<Q2PS^)0Mu-dgu^zFS3
zFc=I9Kb?k-2yO_{+45(y5nyQ`I!{keY!eq(YBCjZXHhQ`98$NhNvluz%u;M^xU_EZ
zTMC%D6v)C&*h8ac!OYquzVf^&@Q%Yr^R-OfofhgrK5fgF4j)yLEIh;U1EAAcygKM<
zR}QFcrX*(-+3MVUt`{J0wfEq!w+6nW&@#hH257UCDw6O5^?u|33S_szoAua@g;#W%
z(AYW*PdxSd>p+)KHMplFx>Vw?-fng<Y7t^Ko&iQzeY}!3M3V3Bfq}T4v(Y^ta{03L
z<Td3nkTuTj??M5eQF=_yu<g;+Kv-bBh4v#!@>VhP+$vl%JFJljxhQ%`FFgI%@spGo
z_zi;D`>mwQ(d*~LfKvA2qKLs1-J;@|;*s-Y1Zb3#dKZD(yh1s}^|<dtCyvs4<Adjz
z;o%w#`^{L#<PHA?#+@&ueFhPiQ{Q$}D(e$wAGct^SA+?fkPVs7zp0$GU$HbfWbvF1
zFkSoN^)LH~Pjv}-zD1wd<U0b#a<Gf92?#F9k~5Mb%Fln$OJR-6+_I9rpOc2Ig^?3j
z)X)wPObE38dDcA@-yPLKv)wdAkq4hR{m0Etks*n%l*r4+2>3A~mWyn1aR$$8O;mpn
zOa}+dEE83b@M|T$v3oFdCk%N>;#~|lX{L8aXfzkNB4ha)gj(suI%}!RLRBjJRkNCC
zTQ~;@V&EoHZ;!8EKNcFV_wq;DIqdV(0?L0TpBD-%d;;As)G!fRsZN$u#ZF$ZD$jLS
zWT6KbDdp1>nCbe`f&6!OL$;h{V;Y*|@4;3SpzDv)UC&#$6F%*^9&f~^wZzacumQ?!
z(yXG}SMPQ6!sTILuYZsyEtCX}zh;Pdk}#D=4pry$AN#p`Z~S#wB7bB2d9!|pIZOw4
z$DZ6x61BHa9UKN1LL!!D4C1`qtrs}iQ@0!E<eA{rEnx~_0w9Pp^61k~^MIK5I$xcX
zpx)*OT1Qucd)$j?%;b&rF|F^de|8c~Go{BY_s+G*&?J_qD+%z~=Myj<00cwc7tGAJ
z_^*U(fKq3jm%YaX=rx^}Jy`gr5xUI>XG1S|LkCua#@7r?e=M?^2urIe;~T|L7Uq8h
z_mN<iIY?ueXQP0Jjkr|J1-#v8X`$dEh&{zn90pz2&cMPCH2C}7VbO&<fjGC?1J4uZ
z3EiTPsH*9^X6H~D%?q!c(B06g-10LT<%chd0_7ZjvrMNZ`Nw4+71dd1o?g?POV}{S
z1DEBbg{&7p{LS@^k7du&+;*lxto*l`krtVj{&Mi=L-)TWo=d6su^KG%S%1V9p#~UQ
zCT=oYA@eY?UM;Nm%>V^G%oyB`CEca6OxYOP*F3+?r2cIxY@SX{LPD>U4^?M0yPq6h
zpF!V>SUv2N<fZ6!XT@QVBt_usbOdL?_@<Xrg-Nz$E}@;85`)?rIlaxiU(uh*j~1Q)
zg{eQ4Z!~yNO|y@MtE}8;v1PYUUh$ji@p2#5G2-8<?|R~@kI|sDztvBP)T9b-*jFnO
zA%i3`c)2gx!qSxt%f@DWh#RGDd1TmH?%btwvS?6~J<Jt5b@8t9TclX$lfQ9axQ98;
zTb(1z&rgn3HP7i2kV}ZP;`}&uysxIlZ{5hIv@fTCg(coMu4hL^80<!%uiK0ZbuG>5
zQ0v}U)|a!6u>tXxa^QZo{@slZ=u#pa)R>MVdVC8-3Kr^p^ceQ;ODq07MCEl~Mde1F
zXt_Jqpo>7fOu|exwp>zrhW0zTKHtIGXoJcVUWYcsCO8{lNdcrdMsC=PUz!mflC;&?
z3#gG9HtEYVPv)zAL~g%)WFfZelF&ET$;cimV(hyZChj(4c;lj(oIUO7&^anQj!PE{
z`0q1^R1nDKh_q!)L+5<l)ard)g_d4fcn}{<s`e7GiMuh?@}W<>aLLcEmeQ3#)FP<B
zMj4E+zY%Y*dPRkj<glX0fj)+lVuLvWiuLCR5>#dzT`0Q!q_eZNtFLSrY!g^<rbjxB
znuOzEgj?@d6XdLP+lIpdcs)m(5vdQQ*1C=MHa-43u=k`X+wif^Cr`H>*$!eJuZ3@J
zgG+eUWJ!vCf_<@@HLViwdWV4maQDS#DnGmW(&uv0Szn_=ZYSZvL-0a>#HJmdScN+?
z!$i?z-vQsFLN}Y#MkVItz$Fk^A(|Ett5-yil$2UzE{QJjN8U0~jF^;aJE78ORgOY!
zUHc8Oi)`hIeX(BW-zkZdWn(m(<8dRkSpyFwBjqcMqE(;5POpF&`T*SedQb1W_Mzi<
z!{u@4bU~e5sdMsbw(*bk9pA3S%xJ>HsbvB#b*-m&lc}SmF@hqA(PtAWm%8~1+tK2%
ze-R!oOs9#;+VD%hPSe58>PLuxvN(`WhJWc$2@xBqqEPUYL<Q@nk^~dTS4Kz=0(y$`
z)*>aj7)^hC`QxaSo~qTG6dafS^Rkxz3Zm^?|DDRonKu@bw{^1KH3l1U0+YD;C;z)#
zu+uMC_e)1!RuPCDYmtX9_<NZMmTr>QT4knTG-dj59cdw(m7lGB-aK7wg{@XWg?r24
zg-|1bJXZY=B*DgJfS%BP{u;;1YwP@|O4f)IA@Fl5Yz@aTE?3LN*XuZh>^yb$rQIpJ
z^{lx}5H-`=ux1TgF9J6Z$Sy9Hl+LWlJFMxG9ahXrcD+z1=4(BG^pEM&d2seBtbW}>
zzaw%Qvfy<Y?QMC{6_jnm){2|t=jgT%q*b#wy%_;7xrBoVu5DjG&zYPWD!Dra$kU<v
zOruRPPE!IQk^@y4_@!EGk(t$bQUZ1xL=8svi%$1=c_b={yvf99zGK0SwN%ik*V)_a
z+TJ8UQx=4i*U5W&?<vt<2-l(qG_Uu+v^BZ-3b=^4ORo&)2plkx$haB5w3To<h{?DO
zoqmq8|1Xv3nIgA$iwCxxeG~+rze<NUzDs{#m{(cpi=FFv^>`h%FXPbGbdEKzZk}oN
z^&E27=yduB+cw#YsHIj6tB=LkNv7##MfX57m!-R`%K7Cao}G)(BB0MsWHEjCCDl_F
zf@#_^fRvD?a*;*u_!((xrr<Xy)MpP;yuxu<+Y`LNCiljotlP`wp_oW%#ur%`V|1gH
zG`I`e5L#?a0Q$g){nSMLg7523$m>kmw=gXZlFB9Bx~ZEp%(EHE6ekI!dV-2<o8`W4
zOXn!KKd+?kSEq1Cd{Ddi*~;PWMi%z4721f7fb=#H@zZWz%Y1$pN$;L8WfS84NR1ch
zda|-eyP9<(_$2&P1Bsw=KMz36)F`djowfi~9Jb|BC7F=Z`!%)T#v*~Wryp~;Uf#X_
z#acGP=`doy`nyZKvn-V<DheO&Oy=CjvMV!dKrb|b(G|-_xH8|u(LMu_u{w$*Yv0bq
zhl8mp;pW!H8j|oUo<Bm*xOrt+%RO3h0PuB4AqQb|Pj%fE++IXnmV*xiF(V`^w>6z-
z*50S=R~IV#x^CO$GIXYIvwrcY(O^_M?2OSZ^deFAq??ztW@!g?jwVo!WG6aZ5i!v-
zVd}lwnCB=0GR`9^>%lJtTWJrQ7Yc=@&u%h2oEDJ_Ol>cEW(~^%wo57)djKM=Y343z
zIw(`N0_uXmOfIo$9@M}ox&WSEUi$GPz!6mFp9E6+5Uru3u`eyE43n$TP5Cjas3>Kz
z1@7tqh)k@>y$kxLdkVVfl1zeu_BrKqFW5F(ZClC4=sm4^M&LkZL`}(?GgFd`8aj=q
zL(|9EPl`=bDpdGe>#Ok*5Ya0f18`@h=|<<)hCx)pr;vf_lbX7qRx~-V<rArOJexq}
zr9lw9GikDZVWv61OJ&A}P3v7Zzu2^eTm_T0uF44l#O~r!0WtoT#6Wsw2owX=E6cbz
zx=R($80U5j5Ioe}hxMM<2ZU3ypKaaI?CHgA3fqd!J*H?kIf_(h%ZipFAS(vUo07g*
zrYvP9$wPeo4C;4<(_0u4V`p6h1r5>~{w6@KnW(oNzr@m+-?VeAyl&R%{B2=S8mzTW
zBY<a%)MG#?0z9>vMwWn{TQ5lks;&zl9pj@~)$hWWL9CC5veyHzA|{lcRDBFuv>+QS
zK3x%SO(E?4T0TN^cRat7c!-e=t$f(Jh!>gERY~<HKdjg~cKyEVdp`lWwY!X3>bD6{
z14B%2T(1-_$1rMZj0^9O6AV=KQkIp-B}_}7Y)EIJ#5uM7Wt92>I-pPW?-b=MHVlyC
zkJmJt;BF)Ue3^hqIRH{~Vy?~O{QzNvJdPx$21}nUXo`z!xePb|3@zggox?Auj&2p#
z(cLV0eG8+hVqDRoxxOm;fyYK&0e*gr>Ul0Yc;o)<na|PIhL;Nrfc~})U7*#NCTIV^
z{+m^WIL-3z$+nVbt|HK}-g=##t)LMx_86tmk*j#m*m9Xb`fYvcwBl-RN$TjS=N>AH
zcEJ9%{6`%OEe3Q9!upK-xw&Cb6G#ockEn@_MXWq|`7OZZ*>Vu}TYseutXrvJZz8w8
zBu5NyEta=%ij1GvpN&tOSfaJ)D3T`T_XmDlk_#JHSra74eqM)n12^4z^A`W!=NI^X
zTSykipTzCk$SJY-SK`u4X>%(w_vwIE1%c9uImwW3VY{`w7GF6hINfZa!$wO5`+m{#
zoB<(1Yeq=Cv-+ktuQ^Xxd!j)=Jb*X7VSUxPZncsw=a|%g1I&ND3P8k%01+9!AtK`&
zOrz?$57z)v==rnV_kMSDT2Q9H55%*iwa?Rsmx(scr~hsL<u4l5ju*?>B3ql!6S)m#
z;HD!1G*~`B2hXnwlxjq2>gjBMEK@4SG&XYttND9hPF5y?yAuQm)>2o^z6W-uSIeE1
zWumCHLUCWUKut~5So;gvGxz`#M2hOck@fWp#vLG1%mI-2h4YdNNAHm^R*l~-s;hqz
z|HRAoo%K#0ayuRbC#);$m)$ZQh#Lt+3s(@L$r|)27WFo|)9>4X?;4{y94$5;>BO?C
zC^vWyc1&h3f`YE<wNUZtJX<%JNs(8tFq(2XzAjE1h3ZN<IhNkJ1ZX`)__qy6y2h6b
zHr;8nzhS#CEzsDf@ySXmb|XCSJwr4+<V10W0;pG%9WrmTx3d9jdN(aCHXhK9m(d0B
zLvNh<R~i1X8bOWzIp0FJLUV0<SFnVvcYBtG#*OjwF~5>O8|4V(5FF$~E2e%a%MBd8
z%8ju<W&vmf3raAW#P(Z7X|oUYMQw9TU7)vh!K5z<orqCTI(QhZ((q|%zdGHU<;Jvt
z$fOt(&9`GH0+N16tiIO0wJ?Pjev**ZLy^SoJ`aFV5;=aE?*6Gi2pmqIExs3DSNg1x
zh=R|aO{?Hu-<#K3hA&6EuxHb)8;1#mz2@Xe%xX{K$76eL+%|tD5}Mvo%8)8Kv-iwl
z@64;YmGG<fBEr-?S2h-=)-MaH0h7GyW_+<Ao<+1*lc1a3gV8no?AK0dSK6O(G72*O
zTyg#sjx7$St=Pay`bc;wlT3l2b&FX=rVoSBtUOgVb{`+8XhwIB?-V=}s8)1A@ELzN
ziVGs(1tJE#Y(*pC?+Y5Seel?P(#xe{VQK9LRHDZl1VP@i8(z0j_h4r{09`v;kX}TA
z^nz7E0AMY*3tld@CjdK6CP%;PeJ9@)!rP54VqRbrjKb{wXN|=`CU)b^Zx($%*`PG8
z<Em1>#S?4}GD@wn-g|W`UZ|!-ZdyG8WdJsZMBG)oo_T+P?y~!dPk?)*D_(vk;Co$G
zEaiNd93gQYAX-`~DNVd2r~{;lax=NXSnhR0F;bggc!uszLBx-_c~PR93_(^M_<3_|
zs6tlB!$58z@B)0Vn#hj(1&BE%kr$jHt~o7Nwp1Jmn~x_P1ETv6Ah=>X<KuitAB;c@
z7@D~U1r?83bcV_*2*mfSBtzk<dXM8ILw?W!)rA0Tk>782R>uFr@5T1$sa~hI7NI*l
zAag7CaLy>YR$qp-)H;@Z9}Z4{x%YD*)`8aula1m73Wr%|m0Zj%57p07(_%}<5?V8Y
z;F9_rKXde6awjC8qV3#|v)d0fy7UQV5_>QPBuRcVZWew=mGdCxIc$UEh+mb<HXE&p
zgvXs*xLRT-i5%%>FS<wOAdu7Q-}=AhMtc#3REnHSC3*OYrW;VX^0Qx~mERwz8J*wk
zESmJ7{=MZ6pP69gTtNe%K8R%#m#)`{GXxSaK!kS1wY*z=KRF=rpd2$OKD|zflDa(i
zcmO{80;eZU!Y+Su^9qp%nR)t<y{U`!04djKVWdQgJ0UJ>0~ZzvUh@!#xdWoTkA3LI
zBij$L3xN+l&_ZukgIo4li7nZEW5dOsi4h3qgx@{@x~%L?%}Oj9&`bW{;N5`-!-Vp^
zp;qWd{9{e(raf}DT=_R^$T2Qf<>FqHyhMjUNExwlls0xr2Q^@KB7_U}Y9}CF^!j^{
z@q!AJJ6D<hi74BD#{)LRD)HvbXP(CIi?Rh~aoR&6kABNGu7R-mp5FC_Z)*;Jt5&Q3
zPCDeg;H9mbSq3>W-^FtyeRCt`UHuVdUo`?U#bpsmf!x>^Lyp0d9;Y3q19Y~JDQuW~
z&?w0hz>h<y&_UL90m*IO?!lEZAk+X>nMN!8eeB)Hv(J8g#o!MS_iKauK@Ueu@<tPm
zeU$_nM_qpVYWik;*kBytvIkXB1ezzc1RwRJ^dJRTGgpq$bm>}W|GVMEhT91Oog!}&
zHovmI*mXTNfOMuEmz$QqhJO6eV<yAsot89P0iTHPMI2!Gx}5mtzCWZ13wrnCgdM#Z
zhwdpcs__z1xl4Qfq^sYZeGE8op|7%|rQ%xMYl9~DtAwn_pN#j^O!?%2LgTMm642h*
zw3?PrgWAs^?P|U9ZwOIAg!0Z7L4-`?&xmdnD~`MUZh;N7AfpIk*#oJd+`5%l%vR5S
zbM``NaCV8nZfc;@<J~7yy1Y;y{C`J)2O^LNa;gp{P<q&@iV%zi1a<n^H|^>v`J%kG
z3%ft_Cn>F*4pfN_4yI`LOZ}@+mY_iBs<=vBu6QmTlps390hsl-+iZh|S}|VHX6OO6
zJPrI^G9nUY%>*Ir)%Hid2o?y|J@%Pt&}@s3ZrJ~(aAVA7(u%L!bdVi-ud(OsbEv!U
z^$M*9O*Cq>FyeF+{KXyMb^y2y#N4*Sf3zg|{0NALv`_ww&}>QA{G`%h75FM86&t{+
zaCHk`PuKmBfXGTQKLSz01s`YPIf{Gjn_uIy7>rvD5w(;jpgIS&{#`$*t|NYCfeZI<
zp6v>Qheoui9(YySvsF9%l^Xq_!S|#<!M|1m2QG~2I&q(a8XkkGif*HQ&y|ZWEAbN9
zL9yju_{6dM{IW@dzdM-#=3K_j?VeG)by|9e`Cj{g)33y?j1vwmqd32If@Gj?q_xfc
zHNy8;fnl!DO4(SxJr>K})_+m~4(&lShoK`M3O6o3E$Q_>P~^pnGmxVwnHW?RpY_Y%
z&y>v#D%G%#0`#r2h?Xel_>JCNpGmYNVOGMSCoY}0*{HwAsj_insy^Om(CnxJ0qA?g
z89L|;;#8kUg;nh;ps2vzSsS7%HB%J6pZvtV6YvxLz-uJqctDm!{(?$35ES{;!`r?_
zKNFa>L3(^(WEtP}X0h*>`QirV9OFh#PjmM&uh&!JAgHa^S-=qM%85!*rdNiarw2nS
z`!KX76Kc!8jHH-k5UufPSqGIdP3HTFD{M?N9tHP0;q~N&H-<R}l)~c*H*RYSY5|d7
z6Ok5cm69BG*ZX)ifIv-*sEj|<*~aCHdAHqhqDKJJ3N+J%*0RSg#?XbSM6_3R!(ZxR
z1Bc@Wp{^I7R|IhoXku`8E`G1dM)E#73%^5jzdR<qDZWmv-tPT5Q&AxDnnrzpdj;`n
zzYECJ%(s_=;SxVclr{7n<;Q$S2J*bNbEj_83+h9sYX|WWJ_u{V^&4H;i&+Lq<K0BI
z7f^UT9GEq?{H;mWz|CsrMTnC(Ftn>y^F6N%cjU$ZosI_J<UVQVjcp~ul5^NBU|QN5
z`%ve*TyT<v7vuoeudilo_%vFXAdzl*!)7-KR56HvUJgucx!}2K42EnFeLIS&e3GNP
zT|SVNuqh;rBSfXsr3IFIQv}A!{{RP!c%9Xi{;at!<JIdRLg8#R(J4@acQff!&wxt!
z<Gld)#<OLi7Z~rhRY7C6)P<(o<DCMGlzYJ+dI0u)wYXK!lC3~2+Z8?w!bFhsj<iWi
zpUs=D;pux@odSa9h$&*|$N37_aw+;9cnd`5>dfas#=3;9D?sOua`eB5WKyIsXz_*H
z6~N%L2!F+p-kw;W`Y%&g_#j7F3jHv`!t0)8K6=X=pkZDR$0gsBK=Wj-F!yqRwn|<;
z4LpD{$DylNMdt;ab0pswWpYqmVbgl41H9NvawL4MSOvV6J@#$YfRP(0!E~#fXz+3e
zbl&ALSF1wg@(J9JV@^TWtuB%aIG6i8?$%Z9mxqoaN%Xt3JjW%p7H`iu7y6`wo-NBu
z07;?eDBZ;OrM;ue-~H0G9W1=V_971?jaNz5JO0E4L1^nm0zobnJj^zrF)>YC<jr}b
zLNAJ|GmZmXP<p`wdVrv1(*_?P;X;YZI>3~MP*Uj)ng5q#JO&5v1P8ab1kF+%Jzy;n
zn8=b+&_t>9CSWKsv@6=6FcEfa#)1bS0u&`;>e%lT>miZ4Se~@x`U4KFwwtz~p5FG}
z<6|no)F;(<;T7w&A!bs(segt+#eTcZp|F5yZ+R$p4-<fZo(K}yF+Yqvc%4#vts97r
zkk=y&oxuWq=ecg_1^`hED0r8`hzgyrDy28C+rBC@k^cHUkcZ7d-{*s%`F3I`9=@7|
zdqM0p^yK!8p+(1fR|swyoJMJh9{g}UHn7`p!EX}wBCJlk<2D{m$3unb_6vZ+vHkXw
z9nYFs{wkl=dTBsFlZEINa}%X|#M9nWXz^g6u;|p66eQ?Fn8@BZNf2xkQV{D0RwGaK
z{;}K`)uxeg98$8^SbQZw&Qhe~HNrj!+zZUj0tvSF(Ts7;Q$YWndK5q^2j18?_Mnm)
z-r9&JxW1Bu(sc7;W{TDmz>8`ZCgg$r>N*M)M+rOE=iqg3zJ<#bYpqSs??d}5iTGaB
zxt0r$O?xn&qz6$0TfRrNa+LX;nSfpPsDf;s9p#%ucnu?IQq^dk+i09-qEdReA>^$X
z1iP&KB;__N3G-nQdH8<PTc1-=y<4|I=O$@tXNOOpHE&tVZpac|Cxom*+Ut!i#c4kj
zv(}W|p{C{11qNh#e#44@e+pcN|1J{(o8*%FoJja+`Xt}GP0s?UBcOz%^CsQHX@1#2
zuLy#U|B??Tr^i8M<5Qe5Uc5Eu*|}Cod0owbD^A0%+$JpK^i!b&BJ0hiewm&Ws~J!+
z@}(Eg_w<IKAAbI7XM~4eUN+m6Vmt%N(_TjNaC9@s1rGj(A1}~K<6q{R^YA&Kp3ie?
z2kLpn2Jr!kA-{3K#owyZ8B8W~Vo>TiMgKiACm2m?YM7eXoY|hr&T=+MH}_7Pg!)r2
zqj|-q`H-^}gn7YzG8DDh>g~4l8WgXcYIivJ6mYN&yRt}lRXXQa6A@lxw?0V&{x|8K
zjVH3BfkA(?%YPi@?lZJuKsf5vyvR1}UcGS<KLvvS!jv6d>;NghPS_wN;UMUFI64mR
zHX59xdF_e~RLpxhc{KV)IdAIOQo=elK{90%lyqQ9%wp-mgzTgInH^$ZKs}FPINVtM
zr7k?c-9pQO!MD7T66g9HMZ+^pJB++{N^H6>1gk?h3<ZDnmU(tK-@T=NvCytG8{sH-
zBNXDvDX&a{Ehf>X=-(*GpXfTZvHDM?iDZ|r-+(g4k258YnWaV5#c~KA?<ILsM`Qd<
ztme|=IUx<mJeQ@~thSYc-R%h*^)P)$$zy}%)&5Vf_#?$rTz+ILabjEjElOU5C2%qq
ztu)Iy-&`uq?<m{IxTDAiPP<MF;$scgrF951Q;*om`^EPBhv4@kwf-;V!|y@I>&4~9
zE`_ihnf;B9RdYGWFy8-c?_0AP&u<}-M&b=Z-}D>&SLR2w9Y4lxUcYzZ*FTZ+y`CR_
znQhw<gk=oZFLKaeq9KR=Rq{LZ6D|={qa6CBTSc11PobRjP#@3rzk-ZmMHqDRUt)`F
zND~8K9ra$sFjbKhg}iU0UwRI6<gv=9#K7O$=8u>FiscFw^*O6=4-u$_D}nco)f<h~
zZEBtPo9qIv4-9C6%E@2}s!3P0`2vAkWs_Xs4FFMKNa0o7kr)-zf3joWOo1{!g(zMB
zL}miFf?BD+#!qCl`CM4<gG|Aj3{2Xogbj=n&i6ADe?(KdU__7O1q!%Py8px1b;ncr
z{qee8GAf(CN^VxRmSp7?GDG&JGO}fF-G)TjBV;5aWQ6Q-E6P^(CNq0wuitsDk?QvQ
z=lkmWa-Zj%^V#G5InTS)so2;Hn{%E&A2bV=PhpRE<j$ccH_R=HZikgKTD}H6$EW@H
zLvM{pp1}QC+9P{<>;;AviM|wiBg<0ngj{!%$U1v=)F<E!_}ELD)dX6?9OQjm7RM&c
zq0e_<iA~;w@7Y>)b4Pp^Ont)Q)<o0dYXBiv`goV_A0Z|+A%LuM_eQ!-QPYlJh6#g2
ziJEo9@)8@<nN#E}?hcm=+y4QrU}c|qnnLNzx2T|z+ave=SS;BLPh8b9?k_OMvvxh~
zr7RRTS)SXL?$%b3Lp4scS{50Nh;A@wZ5s#c5*c=G1D{T$^*Ud6R+YkDcW9=p;1zPd
zWiWNL+5EI1Y8sOBpN!;@vuyMY=NO-#`RjK4?bp1GMS2PF1MF)T>#18NFcT;{Pw0{o
zdix`rXdOO(MM~$p)ISLP80oW>cWcX*r~@DjQjEWe;e*sosfS0-rh}ip|NGDl=Krv#
zJ#QbNd+xqJ)gWZw`9<ZHg$AiA4;0Yytab`HyuIKi<71;PG4S*s0-ti)Zy<Y9+QNxD
zC+7@m0TevcWj8u`oozBVs)G7g{(aITYT(Yt)yHA^y{G1Ka_Hp<fb`01eo7U7X4g>~
zN~itrvMl;6j=amV2fy~Ef|2LWi$|?BoJp7=%cW-CPs!_5IU7hHKmYG4@bM@#j|Q)x
zZfnhi{-@3c8F&={^?}dbExU#9g1`P*cSU{2J0dUX`vUaUwk&h92vJrLU+1;^Z^=4+
z)S_CNzHpRlbi)%Xtx&h{2F2@je$+*AKP#TN3MOcPnwLa^)Sc1i%s)VcC!^4vph=eW
zB=>PMzz2eT_B&HgW|s!hP2KGGjQxkoRKc&<jIdF;JAUz@aaPIcVudwBQQNF5MdYH3
z2>HK@_v{)7Ehpi@!;_<nX{w9zgm{Vov6oy%rDD^eLD&EE|2!z$*XelnxK>-{MsH|t
zO)ESZiUa!B*!ai{MF&TM_cs}WS8?igu>ilBiVPdn2FYv%2-Yp6bo!-W3s4(nR?tW0
zh5-K%Xde6%YRP^RA?nuQVn4+pwcce2H-`AxKhB<TYVAcus(r}*#eZKA4x)lkfXk)j
z_^uF5{3agA(9`q<fJ<1*wpx{bO7HcLe##TligT^eqc=oEt~4cX@}deLDOfk~mg(F=
z2Cw^+(PIHK+(%F1O*}1seX6CmV1Z8(9?lMSIoHXR(6RL&@08j<n?z_1fO4)s;&b`-
zM8CV+7oI-iy9JP!)&W=)Oatei9VtsPDCqHjdELIYMjO}T016LeGNlgyo~M?jpXcA5
zxn&fZ2js)ApnMoc4!(rL2JJNT>?PjK*q04b|L4IpbTh~YZ&6u<tucd}BHa+sgROGX
zIf-=T`TuUwiD=R~&QxsHEJ7cY=z>zGL9W|xFK7>nZ+=M`Tw=i5*?Uga7L9waVLha$
z@C4Vv`{=7sK(NVvYBeZu`h2Fv?7zE}Hz+Ge?x_Qq*`@!k+QV`oKm}^)rcK8d1&Po<
zA*nI<z?GhcTs1YtukMvoFl8J8=1N~TwKY-Huz!PO<<_`ui8c1}Ti+%4PBWPn03dY0
z>0n%OFHE;w;~WfdV!g0j&<{%GxcJ*C7tfoFWVkXA6;OkFuc2zwQA$gau3bnL`HyfX
z_lzI?)~9Rsbwv0AWH2en9Md&9xhdN)xpB}<tJWW>`v+EMi&FfaoqNI3IEg>Ffg<A~
zI9@z&>bS}sg)<vAb@V!#t;EOvflNCD@*(%BHig^s!w10?At50Ad~cItyOH@LJKfjw
z`8f5*@4u<tu5y3^U5VwuQ0vP)>yf(h>lc`1!7vrBJR4HOm$6*YS$5E2#_8pxG<J9A
zxn$AhuyDt&#C8NlOSCKm)&;rZd+CgYRhMn2)!`I5t%>#Fi>iIXC#Za#%IZMl)hGis
z20tbRS_&D<E#|XrgUjAll3Z&4CNatAp;q&}%`)G}jVTi&Kr1UitKiPm(W=-+Hy4qj
zJ2!}Iu|v?q+5K*Ri7b8ZFbBSPw|FTGNZ%)5<ETCQsV8Qu6?`|sHu4u%xGz_cV5Jv=
z_?_**kc#YNSH3j~7@>c{rHim#_Ato-tz6pW>A;>_wa09*->pT>;@8^+x@86;gV$SB
z!qKcCZ%u`b8Jnrue6|;jdk4)R{bSAaQF~{1H0l8AJ1}xbTdkGsPeh*FDhz7KQduC5
zz*@J8c?EQ(7L@tiJD`~lRV6?mr6HQBVSv0S4VHJ>u(St7Hms}}403b6<<wH@zCe!m
zHULyPSwEzG&bbpgOD7ht^SpQxD?kR700voQBLIRcK51itDFf<bnY<E+yKriGP^|a+
zG4B)U|25d<U@t^B;-u9-bolr&q{k;g_=hZ*S~?P^tl?0xMin_*FB<1e<fze!)z8M<
zf~_Pgn+qC)K;j9tvZfXcO6$RA4cLUTe>{t(@|V$GIi`#i$_%RbqYW+M0>-VpS)muT
zuDQcZfX{lb+#t+n2vC3JEb+C%9d(zma^!9}`V1A-lF-;g4X$P5Mz3)x@P!Oe<y3IT
zG4!?j+{w}~K@g&jb$kVtK=@OmHQ~!dcw}+GDDf(pstMrCdrG02`Oi@<bG{?x<>@rT
zmqM^JRSm|Q(+w_!P=IDQbJfb<Gs+zUk6U$p%9IDd-mPaRmo!`GXt6&GNkl8Ko*ObE
zgC3-BZs>MHd)^=jp^f>GohwpC7W$9t=kYE|BfdZwQ5EjD?+d`UCV;grVB@uk0ZC4g
zr%lVN(x2FGzMy8atVIkdfRR3}uc`;j@q@&H<rN8i*Tf78C^MdI`AyTVSbK&jCj3E9
zu0&5U&_$BvNy<%wS3$0NxHJE`>FzYfQ%_}_0dK4~5ulXFE9TdHFW>IRxNGHEVhu7x
zhBTm*{)6D5c~IGN5Lv&iHdlc)4+Ni51PD37xjB=l0>~;Z)b+anuK{KB->7cdeBL@o
zCzu*LTaML=gTQ!kq{6-EYvfwWEpuZo82Anjm{$>~=I&1VksVDzVxih7OoKJ1LCxX2
z@F#u%<^N7DcAbmhlNifUAj%N!{n*wdQ5rN~A;wCrV~SN}PnLkh!;}0Bz}d^vZW`Eu
zFS&9<I;3FFz0|XZqcrCBeTU^%Z(s$gSp0gp?@l`RPg&c$6uw-Eqk|A~pcCyzbSHAL
z(7G&LIq$&_th%6@hOA>LHU)r#8qbPqtqIWGYY_t_Z`ZCTR>qJ~r0o)-sWP~PUGrCP
zN@!9qr@EV4q^;j%WZ1uI`k?31zgJC0Ga8+pYg^$Mad%51bP>CcbOgail{HnhjMYa4
z*~`Y$J^WfUBqu_O+hABprYMT|()C_!B67X0eYzxv)qe=AJ`A1N9gH3p^V6MDpQY}{
zs!4siu!kq_4PVP$IxhgnQ-n}s9ZOVqBA=Gl_-|$W3kr_rz)6!PNd!2sX<=b7lLPp}
z>h|Q0!sM5slOrlc9?MwCo0kMWb_&KV&RTC$y#tpc69FOi>Kzb{wwlwrzR00{^Pii%
zkCgG(b{!n*b=$0q<fj7PSCXRz4^wyPgE~4Nom(2Nk|MUoFXTD;y;=}55xou0_QLp0
zOmdKkqSl-S2<7K24Hy1}$q#4AuqHOLvbSF}sePNyjar4~FZz7ED;E{xF6aJhH8t3p
zbsd-5^2*jiKwFXH*SBLGq<5hb-_fbH(-AVOl}Arp=S;s!L4mEJan}bulzLU*7J#i>
zvt~7sjL8QK<j!9M8m=eOmCj__g!P0yuuidtzDg*#RxUO%7Q2sW>ln{be&(BO9ro-^
zk)Ywmw%nT}WZmo&LSP76v!m2OoD5;}ZbSiWk(*VVjrNVPH`ojcauPoGWgZb-kfFa;
z+L{)A4S;RrL9F8^(!9JkOsFC%l0yA79$OtO<}{2~)M=rI-*rfKvFKiXiHD7HAcU!e
z##<yn5$y?Gf7?H%TcAlP?s%Iq2<)7qCF}%)9|J|gF-sh&>=F#vAX+6>JQho>2=P@B
zb#lZ^z<6+!zt?Fq7&KNR4xnf;YjTrEKGpHA;5Sy3punbE$!9EWeSSI<cDK7HYuh{p
zpT>eZ#mxa)Ph4dk#_+`^yPXqL=x=}GJ2@?m?fGy{zk!EI@7){Mf8uPl+YoHv0{|hQ
zbgj0?cD~j6HU!HD0^xW;nusBn&+9c^Qrq~&mhIk~%6Qwmq{Vq_AC<J0%|)#3bAUE&
z*s6ZGp+KmSv&9cxWP$-E)%IM!*D-L4_~C~y#>!GO*fv{r#?n>to#ST7mrYUA9sy<@
z0bsb91HpE~%<qV+dDWGL($C2rHRA8YMrCAQ2#l6m;-LZ;=sw|KK5@DYbzxnEX0*X~
zf0X`#Ntbg&nJCmH)`O}zsK>4hNQ&P_szD~uicGlX-1czar7{jPY;_Ri08YbUd3BdL
z41pa84#4r)ffX}u*BjEAL7x#?J!Y@kwfx+(O?5My6kGHpXO41F0-w-<W{avK6~GOO
zJN}vXnJGxdB1u-xJ$d)|uMhQN;?1_6s?o4Q0|PE$pe(3QV9<^4WHD(vN5bc5GT!2Z
zG7?d2se4Qj?C@rmODPy~t>dx~sl+3G142{9U~`lI)V|2Lt`}RIY>kAxD@(*~MNiXh
z4Sjyzy53%wJ5_WFx^2s2XN+{}LdSol459@V|EOOhE5T=dOON5MxsUgAP`M=NeFGiK
z7+lEC<dt<Mp9Pb-tfB(BpDC2EZM$y<hf|70&>q0nhG%j!p=+#;L4m|T=sBm3vRNZP
zuH<{7&Dd;>YCih5$a?MP^4E23`L^TSY`vc)f&85W7@V=M*OfNuKq0A>zMoTx^E5w(
zu76m%Pm~q&4_hiOfXS3eg-<>05TXMEx~%Oj-!<g_7)=dfnt5I^>brI!%R;+jqy&^_
z(a?aZ$AD^wmQMt=P*QN?VgoMuvt5wtzUnLLJ#kwoT`{NDi3t1JI_+Z0!+g+48)lGf
z#W}AROwyjp`o79W-~S_AMF|^{8TAl9&S4rcmjzt}K?$W&Eg&&8E*@;1>56UaAtjO3
z{6tTKO{&577zXP{x>i3yH#*q*(t&lK0}zoh8Uhi?h)DCqBzs|#4iU`=g(8<;8m={O
zL5r>|BIrYhRZ!W)Lqm1w$d^dF{;j)U?C=8Jb>$W2PHYiDMih)CfhQX%1kp?Do}uyo
zKDC2|@}OFlaF58hFb6{sH)()4z>+z|+yvhCwZBHa`iu6Q!kgv6jL4YtK)8bQlRpvi
zfpX8YmklR~ET~ej6*ic|&^&bIqZ^}9?n^3}-T+sCiU3ZRdr)+JuIY}ugS)wFw|h?d
zX{xlpM{kkcIOd)w)6HDZY_zAfx2KG2D}w|+8RWA{8~mD955LJcJ9}L=sw8y5RviH!
z`?4}UVRUqJ1{%O)*0~IQz|GPUn0egdXYcCD0<8wKPAIA=VX>IDq|16D>oNdOlFJv|
z06vvw0GtZ<K$L~4<9b<HB(8mhj<@IQ6N(zx9u5X|BNY)zL(Z<H5AL)PJVdNo*BxMR
zYET|)&Pd_nBh#Rt+cLs_3xMKHT^*72Nq`77g9ZRol2wrgV1Jl4$SsRoJD_}?w>bq^
zq$HUs!_{w&b|NZj&9wPJHRw11K?VDs&^ozRbI14c+*qM2J>67O#@HAZgZbtGh~nz%
zHC}{usy)s)4}PJW1eh!poczrm*vn|TIfZ{XQzfcp5s{bQfYnQPB3S`KV0jiIa*6jg
z5K`<=N(z<~D8>|~BhGsXOxx8x_0-I<gf2JGRRt@43V7RgMKsY>RHQ2&`+Srf6)OFD
z9oJ-i&4!s4!6;;-JE2QzeykmjG~TkUnsq;+u49IUuj7m};NFwBv?W%(I@4|M?o2y9
zwuI)KT7bnePA!+R>iLm0KqQlzoyc&n><?oF5nRq5l|nz#YTD)gQ;i>Vh!dvt7l26x
zVe^qV>UAd2Weg;_SBgva2FrbLpkk2bOP=oQ%2SkGx}epoIr(V<V3|;vP*Hfnpar4w
z?LT*I5k8ZnK34F*ttBb}pFP1d*XJ3C>JQiPPR}I-w?ZRbxuAAuJQH!;d9G3@ii(QM
z+tVNS0@n3kz<9-a&bRr=lrB&7xF4~;;|%H*F;#%#N+rdo^}vvHZlY(PM@BFQ7F&N&
z9b8vdo9a)sP~W{cdtI0r+B$@&=U!I>m-|Q$zbrgzD2L@6D`+nqE1$>y;-$GJ28+UM
zJ!Rmd+;~k;@phQX<&D$cR4eckK2KI#gs>G4`wU>Tl@A@hgW}6EPKXbdYhc&f;$UpL
z_XVBcA*`9b6zQ^Q-O4l2Im#6k(QY81g--h&Ja^Rnz~M2fKwlc~rjEzhK2d?{q*#{p
za`_B(3SL6A&~*flRnI3+wu0GnDfzgymYkT|SQJp1bLKhuD8(1e{Amm5^BML-ovOf|
zqM*qt?xiV<mZBHb?lOZvsj=RP;-b?x=IO8iyy@ock(Sg}ODG)$kV#O2>q=8*^X>lm
z=8jL7<<C~|Rj5=i7A&Y;-4)njH=Ko4)uQi=8>!Q)!*ih;m}nmok~yXluCsjYZc>u7
ziVRpbPtBSkM7O2D{yA&|z)y_cpgO>hR4ITaE8R*RPIw*Y%g-7A;Tm_N6qec4?T_F+
z{IJ=`1x<J<mv{$b_@n^{*Ymj+4o0QF&@~*&z!G2Hy^r*%7!flWEZ+dI#iE0*N+7{N
z7XZVRdcu{L=u~95gI(3ibXAxPD;A9wCarkKab4Z;xX97UMyL{Y`O$q;&>5=>sKZxQ
zzS-w}>i?WBT1JEI{-`&vkX%+y_uLH@>A5EopkB5l)WPUTQ7hF2P4#K|@(Zn8s6Z@E
z?TD7H^Pr8#)uyjnX2A2H7{O*ao+hc$QUsC_c?Ah|y>03E+rl=<Bg%bR5Rv(iBD|kS
zBnK<9LckP%Z}ssPqJ}OiBGek=){$7MO}dw1qgtsaxz2l1`y^;V3Xlf~`-Dxkabfog
z4{Y~Cpu!2jR9D>9m-!}A7OxU0nn9R^WQ3A>N3#^#)t%vM`+|#fZnqa?t+C8)qdt*~
z&-S!zI{060y2Yms%o<7w^a#}%+xSfRL0EZKv?%+Smd)R$@=YT6d}Zs&m74DEeDAS#
zIW|b^q1mNVOL?puUN?MK{M#%C^-12{Q#zJ`<tjh$+2&=@Cdr|<ka;Ztb|#h3+4dMs
zq-iN*yo0eVsnp-2qjH9*6I<66WfH+q(S}^4^BAaLp&&VGp1l5~EAhok2f4WQ@24$H
z__4@QHM#h-@ep*(^X)FsF$2Jw!Gm;FC-1y%+yrMCI&T>C{I6&dARAfi$`HjFLKoKj
z!!Id9*Nr^~wR+<+Zd&ci+s3yQ2G?^wb(R}rTavWx6mR4$;`i(qplT@i)-bMoJh(j@
zhmCJCU8JnLDp->Gug6dw_=7LJ#SN)Nnx_mX==Lm{RH_ibR+#&&C&#-@V?7{oh5wyP
zKvRA~23-H`Jcj5PNFW{6`<`>&J99f6bW_<qYA^pr*~*<j9eVi2{s?%*#1;F_MB_$D
z%r)D2!9d}}-sz#<RW4Rrc{PR>p|JsIk8%3ZH*_9O{dxN#vn+J;L@h+7&w@npL!T~1
zF5mG!acR9(?o!s@SZJgIPQgRv@p#;uH;Ubg@rmyygP2vp!tHUfchb{VvAWXtABRHz
zF1=q=&~=Y{;iJ5owaJ?mZNh|V0oLJe4tO?e+P!^}c6D&E_&VhPdaLoP`r>WEzkOaf
zQwEpSy0wY>W|o%5NJ&%n+$yBhXd;kj6nW?qf~rP{r$c`xmHKX{>X#aPB}Wy5?<bG=
zZ*cWObDI(J>aq|o^UMw&VoAT#4*O!ubgIp3{hK~l_P*3)$L+^V!CFNJdCZUV&eaQ9
z);Wf@7Qe-y-7{=}IgZy@nPY<OOsmWNy>C2`viJT23nv!2q83o8{3SBm_k47LdQMe@
zsy`A;nN^67ex%^Y`TGXi7*1Gg+{yIXHH+oy%@%vtEYA3!(5X=0fg3v7n|+R-Qejfi
zf8fJ&mbZUj0S8J{CBfF4<F0Gz?zPZB)8lQ(&CVpw=874?v#5?M=ic|jL?HL@JKbuD
zp}5&g#aA;>JJ-Z+am#R82A%bG+GWeV{X_YoT`WM!5?MeDqT@d^tplm*f_!w88Ub%E
zjK<P1d}z58AYx|b+>z8KYQ%3QJ0g7{angq5UXtg1LQV!<n@8}ncqQZ02Pr7nTaHqF
zHBPY=W~`p?>UIB=JyPRt{FtHMipwG`CA4_Vb;~Wfe%xLK#Z|vyDkWy3PIvzR3~x{v
z$j%jVg(BSO0Lj_!3N#1d#K;5t|I@%Q!+$~R$3}3dO>VA*tirv5?FYrcgKj4k3MG{W
zvW*N%*xw%@7yIo&%zFaZ`<$+!>ZVUDbf&b{>-0tsCff)@KVyD-;84=X)r8v%V+pG|
zQhetr^jV&Q4I!Xi{rXSbRkTfMp<8B7YL-?&;;ZvCB+%{%*nrW^C`nQ+PDt7e<m`Uo
zl^G<0WpR!SNy^Q<Lfdk)$}|an=wztdgaBBP<8Xc{3#|KOIY{w5QK<611-F;>b3mbF
zksN`YTcg@tv&jBVd18nrF$^%!X#W^hbi*V(k-NZIV=nv~-hN_j(@y*?8f|~=J?kBz
zA0;INPs?)RjzeccGQf|Ei}oKUC60|}eB)-bK_77J2J!^!uhZ`RtCGU9m~kElby!F%
zD{mmTl0)(WX7|kaXZ^eyjLHP0Z11>cNqgVcez^SC=I<lrakN9%%Jb()sCum@Yu&|-
zC7ez}bQMtmOuJam7?rd+uONS@m(f;4fY6T`%L!ud3yu?TO}KyJ)ah+sQ6@dnfDio&
zu?h$f>wf*6q-vw$mPDxeGo`y9G4imtb<B32lF!rzMD}*K@N)StFh7P^O3Zc;SN<Je
zn(CwXfbt6gK2Jr}uVH(XEaR}&=Ii%QZWP?);JDR4b`wKkL<m^<wxj8{V_L3Gu94Sl
zuk$mG{voOmf8+2SRPL^RFK8LcWk~(c74wUJE;N#L)9`pDuF~^`Zi8emz021y^!R<#
z5~OFlIa^EZb8R=zOT7Q_0NS-5JJ@w4UwuV(Sl7kLGNhZc9*hKzg<v7~rAib}xs*EJ
z`*Nr<`T7YAyBG-0`bqXDU5`_+RoKu9Fs3JcavTf0Na301-{YV*ENd5Ne3@Pm(_nVp
za2o8IW<)a}<jalpbb>>+p?cPrM6pOY<5#w<l!u!$VkRA-781(b9hhUqT?M?RmdSra
zV-h}QVLw_%dR^B9+STuyN7TVA@-EivkEEN7FKfN@_QO~JL^q&zgX<=ag--Y7$>U+^
zp=DXhmS&g!eEI+ij$d@!AT%iSFjcC0-p^+ELM#DB)+KR{lg9n*^HwV?c(+Lqc15Q~
zu$}PtrJ<d8CZP$ZR!24HoklE*rWSebbfDU{DYX9hz#*1XUJnRgem$!XK36?q6lTY^
zY+(AkeZU;GOrh^)XuXpl|6IqD&d#-dHB{5iJ8S{>1ZYG_-WaEC=?t&lx_Rji85SB`
zfprJCb#%AnO9y;8b4dbYV~CHx1QzOp<Glr9oCN7G`Ps23eP2mxp)y7=sQlME#J$ky
zJ|jowt<i#<p_D**uT$GQC<+9pi1LlaSLAMvHfRWlHLJLeMFO(ntD5S1|E(=ejq^u&
zDqn2amP|QSu>Z#k`U_i3a?Pr}W(lmie=+a?iX6X)v%Sdl68~DyqK3BkO^f7hwVH7U
zQ)shi$(3RX?sc3yUt0PmBJ)3pIloz0gd8<$q}JwMU8Z#_hx&?>5nTq`_K&_1>mn$g
zTZ=1sjU^bhif@i81}Cup^?aN}A4!<O!t>S1_v0h!(|ZmAG1Cgn-K;6OR^OuwGTl58
z<Gt)gdf#{qh5sTRGR6`%&bSd%m}+hGJx5o6`v?6jK>njfDA$^rZ`k(wPxRi=dGy!n
zi2L`mKOS29)*(t`g5Hr%R4$b+qCD^>DT+lQL9wVnrTvc8b2f<s*p$bEzR#(8s8v<%
zY~k9f&l!&_@}+I!`QZZ68(kjqw0mnDWtW`aOa9HF?Ajm71OB&1rq7W_OFhAIYOIAK
z9r<X7wuMZf_yW*9$w*l9jfK*npG!FEQTp40wO@%)0{CqYGiN@Q+^o+{jV>(_lJ(9!
zvVF=V5?w@a_-h}7Z<UGBRwBe6r6FZm$7|{7DE_2^5v8R>?u(0sDXH^y7lprW`*46+
zISW6YhFBufh)I4ob^Zdsa~t8i-&v251V38TWI&qIXM@=dxs`+Z6x-AXtUD5<*|$8L
z<UAzM#%`1z@Z|L0567wX;aEqL!LK}-ej+)42WbyIjJk#geC#7ulVj>J$w`}ROj_rE
zoX9`;x6&NXxi2(mo~2W-lyjx4r-x@^NOhOMo7e86$caSt;K>nVhfAc!Dk|V~cz>Q7
zk3kb{P8d(v^4_+~e<rpG6IdW)1|cWy28S<aA#DemsMRuXNwA1q3@mNsYFMqdwd<GH
zINeXmcX-pc1#EWzEzcmN?R<8=)s}qRHGntz^Z5Y8j931!fmOS=TXV@=IRkm@q3zAx
zBLzB>*eH3mrNhJHX0g7YuauK4!I(VBU#}qGn#r!=8tXJ5Y_Yd1QCgmC+oe#C3^;qe
z=pfuUd|dWkYeNMNJ@H>hFB1cSk!IGh3gw=!Jv~w0vX%imf-4PqEP>Xx++_2&r4k7W
zSk*?(6s0kACB=DmPKVt2&f+%Z4xmao(MYdE`T!X1-ODS2`ubrgm8IuiddR=sB{B%U
z%4kA(Xsxn9wEC2f{MpPMyMlIy2q@wC772Ww(8g?G9H*}-^1)yJ7MTQs6ce{gp@n^r
z*<!XmzPJ-V?(U)iQXQ>;WO|wa%fGFb-a%*ODJqpgRxX|9p3Uq2JcgJ6577De17NsM
z$J~>eh5PW|>9}>Pj+UjzXJS`1f4z!?a`}C}TgUsn`2svc2I?Ie7d?j@k9P&3B2Tlc
zMkcN373PC~DQ*lBwrP!HcW)`SZV0F2{;p=npn>cW9I2W6M0NM>-3kGX7U$?evL~g~
zY=2ta`$%cH6Qd3}ZWv(;hx($W^m#I}9U0m5f;RA66;JxgMiBjjvt5WJc1jff-_fx!
z#d2|2ZSBvc`W2<MDyk<1cuPBEMrK2y=D2|P{4to)jfGzB(Zh(pIY5Ww#*s^?Q>~=A
zW$Em!Lw7=+0J8-YIS7(LHt;2tFMZ0p^Qk}8AJU78B*>G_^)RPOGa1jjlc-{SaEEqf
z${^@4_2rEG;ApGES0CB(i`bZdJtb@7Y_;80yQd~%r*|}KnGo{S<nPGdm?v+kNNiCS
z_Q9|38iQS*2mI7VCSry+MMDL2C3lvLgAx&^I=*u^OiWCwotC->v@6drC6J5mKDE1d
z@$CD8%!FJMCl*{fS69Z@$h3m3cD6T@1K8$p_BPARhf!NGLzwf~kni}<|1ft*C%E#C
zk^QNdz_KeXk&uJ;ZX*B=-n})Bwk+qshd-r7Jof%L2a(O=YA&tTg7sMj<v6mPr;lg@
zm|tIQ9z83JgX8ML>+xy-i!7}E6{tVg_QE;Evm`Z{+Be2cL(?fgfwbKroWlu-aFp*T
z--K<>J`}9jP?y(#>rH%W_td*S(SaaOFZT(hc4nIGiloG;dg{0x@}jOm8BzAxaVpK2
zgd{f)DQkifk`sT4D>@pd$FH6H)kkkf4q76do&8S$QBBgi*&mM4mR`oBOLJH8J^$SN
zTO5}xy--|>j^LsYu9m|R4%>V80>&o!$DpYy9*V5)vTd(v+7^j~Gx=@TmOpbjL@kV|
z`#XwrRJ$qvLF-)nibM?m_6~%BtLL`ipL+HJm>bu?xE9ZHx??+=+dIQO3Wvt-^aUi<
zoO(&#snhw}{PFYwd;jqgM`8{k*R5Krn4mD1K`yoT(2i*M$sBA|Fn3EjoX-aIbj68}
z<nFFutWW|6ZE>@OEeScPM%$Z}%d>PsVw%N>_|l(=Fv0-F+0B^i-NHHb!&A+se<$rh
z9sycf-sk6e`7)~VhV7TbnT3qAE<`>)DSt>%15=-qRKoa!XIgo~t3G0l%F!h|+mjPG
z==Q0^p=Cl*1eH8&)XTHv&e9)ZipjY4#d*k=@F;&58Bx*p2TFvYCDe(1dOMi}$^n>b
zuaSIzC&L{PEM#=GGz^_v`eP5o;{IsXe4Pf@<#*Lh4WtQgc$9X25mN>^T$cl;I^Y<I
zvb+eq>Qe#PhTrXO?L>0FG?9)gqdQJCACYI-9ji|yc7<ctvLgJ!J4hYh2CZhH2Zsv+
z59LyL{>2KugN>AdoNQe6H3RQmlK5Q-=rKnm<^?L?&rb>)0+CMs=51?N?%wgtQ-4YU
z2}NFKmwS%8`C#qj<Yd050R!dEPvWG338^|J@Z8q-?p<bDg_c5@xu-@GvWCCAeUu{J
z&%CDLt7pA|2=*84lq--E5EG{=@C3+F$)9~`?^c-w80Eg|su$d0(H<K_6wpz&^Z6~k
z_Kx;zW?W(y=#%y+^Atrcd({}8Og)|u(ca%gp1}P3_K^WN1zH7n%``@iN^nZ>>}*UN
z65+v@j5qAA#uPYq#9Ij&QgIHE{=V&l@K~m=XBtV>EZvp>(M35<I_@ruvIpTx`Mf_X
z%EQB>rbW8m_<~Y=^5tD`wrhX;*=HUU>hxG#GIZ4#k-Eou&2D%7bg=%<C|Bc%)|V?!
zT?Ha&mIL+H;wVx5f1Ldrm1VbyLFi8(H}gu!_P(Zl@o4uaXdNhAZ}gW$4J5RxJVq%?
zRmG9sZ`l3wjxo7O@xy-Q2FgZ4rE_lTjRynOE`82IInqOlfz>iRV1jEew&Ofy5&0wF
zLtQ2;`q8kidC7jXcrv$$c!W-GPepg4tz?|!6XO;EqfLtRa(@`lm`AY17?vWvnw0#O
zU^bP-kh6Fi+mZ)Zi$ejL9dg=~z6I#G)z$7QF{9$RPp_kbUH;hrQ<94O?2Bg$<5J!a
zOqFHF?gg_m#~|M$AkTF+KY!{g+Dw<3n{VOxfj@+wT)vN@GQHMm_<=JOSEqW#bJY{O
z7Tz-)FyT!ko}(n4T;pZ4<OqAo{w*>_-h+SSJEZKo1v;;4+vneJiUK?QVMLmHhm8(m
z2i@i)55rMH_=-HTYrR%mLihhzAK`ZxbtN!YD!j2dr-+|u{2hFk{=_DL{_n0_`8c8x
z15$maFqzt=g11k8_W^#C2c264UbdRpt*qwMg*G&VNA1!FT?h1MQrfdLniVCnuuvH6
zJNG_-COEl2=#N^*aZqBh$7y)Ekrgb}+1Y9RN&Q*Gu8l&$p-M~ND|sVuu|0RwbHeJ0
zgA&u!J(69!#(HNtKT8x(*G7S~n&oWxIRB{|4k}}NanUEB4BYHj%w>|1H|DcT?U(o&
z^#eNmwRd;Ab0lF(hqCzU^A;T+Ay@2CM8LCS&JVy?z#D>NKP8OyfxKnBgelh>*Pcpc
zPgb^H#Gr^c(>nLnw5}k>ho7{=Uw!xe$Pa)L6wg8)2LgwrSt_13%wY3sC$HH34Z06U
z)2T9!E%TJ7-E_ac#^F+!{V!G>0f|>XFH-jmh@5Rr7$c65?{aC|2!RkrvqRTB)P`#+
zk5w?$$ZMBN<RIlx{BEkodm8qIN4A@d3U{f(M*JG<%|p9VwjD?GP2#Ld_n5!ZDlm{3
zQTN-mBM0GgXF#&<9`ZIw7ZpAn_@&UWMZ(Gcfr<R@r4sQJwDtLj){!grkRdS=jp5nd
zpPV?5qQ_18lOp(v(QZ{Y2|k%i6rM>wuK#NZ@Gn#oL0h_}o6*pG=jNT}WTp0|^KvJ5
zcgPHd{d9;dQ7d?Ic%oaAK%SC36L0EWOdW5}2*-(|7ed0+t*=^shY}Z=o7Owk67UN~
zNSEc^F8RJjDm7-;zi_#AGAr=E-S=ZCApQ;=XjPl~rbhFf@%IK5g<pZ#ZgIFS;U3i+
z<K(S02_a8sqPgFhk56S8Hf6aa)tbYh!&ynz6Y8|fSfm~()`%|N#iO)EjeTBqJnwpL
zjJNW4HHwq#vkgqcXD_;IJMMKxp)I;<@99AxI-*nJc;shL;Ryx$oDr0XHM=f;=K!Sa
zGbeb~+BL$WOn9{I*COzzBz6q|5@837&~1A?xK9AmR``>6{m(9=gk<I+I>KEBVF>6)
zhjI(;4Z2rqTH8-iD!Er_0)Ko5CB7*Gbf)AuJ{D00Dx62A{73zmhv|1i-EArL6FC@~
zBN<K=6wL9XU~0Jc%scwXT^6GlKy_QP!Kfi#55`@4#@D|7ZQ#$TfXBLbesgGu6&PnD
z4U<;OLWxAAu@QctPMF6Y9nJ$fFb<qFA~}c}Yn0`cyw%2Gw1ii78ut4+7*L@%a?WOS
zy=+cpi;|W<SFoo;Mo`ryBrw77=B^ZpHh9LGRxs@@fl(P`CeGA_75yO!i)>K#Og_`D
zWVFXB1`VJ~nf0M|jKpt@-$s}lcuQq)YwXzw%zGq=+osYOulg0{r7VOT4wUme_UrYe
zAfJH234jeyT%xJmEIuHW#qj&?r4P1}8Q3{GidqgOKbF$DVIZNtXT(N<3<vjQB|4&T
zk|PqeSeL8`ijG$QE`|4~egsWqPUZM$3}`=E|FyEa7AQQ(9)>+8%NQG8H5uf~ved{p
z|J&NcLI*>&1IDXfKHy#A)Xw2{J1YKc*N7l?U9l4!m;P%5X_G)sNwZetVe1pEzn>nT
zB2R4CdeMb3S8u2)xEOy@oMVr}s9eB+@;fLb3Wu7S`h0!JLKpMs08WqB(5|h=h%<_r
z;Afu6J>o&mDdXay56rj>zvK-09jFpT&kGu2IWsdewds=Ao&QPBp50?HI55%}t|?pT
zl9sulVwc;yXl3lAK)$EveP&SIR2Lz~fI_`etS@;T8fIhVpQDGV>F@n2rVWT;DrLvZ
zv~rqU@wwZFb%lN%;w}_a59j-_zJiHeycp!m(phpX)0Zim?)|Q?L+?RE!qQ(%jG?Wp
zm2=Y^&PT$tf2F(*XMjNzpv5Csi&=czQPtD)EmvZ7jn1Oh?iatHgOuQ$@-vbVUC9wm
zp-daAT<YKFdi_p<cU1C9g_nuz%DZlu551mHu9Wq>X&U=GKio+(Oa#BKzxU>ALGH7b
z=rrI?Iw%;^uErdI9THCRx>-;gKR}ovAocs>j+8UCR)w1RB`F1+7S6o&-dQL2WCr;W
z(2{ZeeiD`0K@a(FS7;mdBo*xbzEuDpM^jfCW2C*fmGnN+9EbACS8whS!5qnTnS;fo
zL-v<mQ7-rbTt-*>E?Ly}`}7W*+-$kJ8>$kEZ-=CUwe$>VIo4}Wu9EL@<siI)t?zud
zQCj{`y@f`OadWb%6U$wnJ@NIk0OVSs)E9(%FrPge#7Y}4BSCwrpR~(k?<bHIJR;gh
z5k~1Tmp`r^Ds&|XC-3l{OdLQ(0u$B|M(j?Dq9bUEQqpMe<afdmPNIv=+F!LDrO5lD
zWzEKx%c1?=Qc2e7zkR!qI|6o5E@wal$32afGbUEm-1E|q1V1FREY?p@Rx_7T0X2nV
zQwC|jWCp4G1DJ6S^DmKj<kB$(uWrR%|H&Ra9{0fP<H!9{13G@8LX;JpIXsgEm~w)`
z%h+Fk_Xi3{5%wB^2itN-b#!!O_PC1WN*beouYw+bfVQjb7|M6-_Exbu`rXa^7YmRG
z3$Ts3>y_8=$VJegfScwwVU42}DY{>S5l>;I{wqZ#n^BL)JKZ<cNeQ<*j2K2D+`tR^
z?F|@YUO`pF#{qRWZrQT#H)7~C{GZ|L*R7*-4Q9P`CObGZ_ockZStyXI504~19)eU-
z*s#B``XPhzyTLw5AQle-Jz8Oja22(icFh<os>oaQ%S13p0wW&l6L?+%IP)mGocwzK
zvMu{Sx3-@<rBaCJed&Zh!$u7PYc(1>?orA8A|YPc)%(3oKA%|pNl;sKa?P6d1^HTR
zEC+rgVFN>7gw_X_>P78u-RNt$1iM%Ox6W!KV;Pmb&7zh;fyQg=zXrr(;1(AW^Gu8g
zQWW-uB`9)&w&$-dx@&2Cp<!W{{grg19svcGthg#q?+W9-VcX1^8e}TH+R=D?pTNcu
zwq*ryRTx;$Pn(qvJ&wzbr~Re8bCB|GJu`ZtiyEkVVW;|HzLVj(tmTd5eM$^M`_Kuy
zN-iOUMteJ`L`zBCbNXc;7<2#~=jva0;8R_Gb*rkUJ(J{6>D|Z|Gs!#E`fYyu!#Ioe
zGE**u6uLm64VhYi{4Z)If&4IdRqVih20ZEH^xjmLAg;iP*AhR35AV?ksP^eTtpwev
z7b<I=DruVS5k7U<zp6;$DZrd$M+p>@o%#CptDj?<=^8gDzu>(g=6ywBq!m0Te{CVE
zQnSGPTjrBM1@HYnzbFf;^AW>Ip-$YA-~|&2RgSS=@`50?M?IgqvXn~V+G++hG6%Fh
z_IXykBcL$aq+6bM3UwT$sjZRb=B?sYWe84^eL5kYj7*oB9RA)lRnArrpdPS4@z^c2
zQNwRDTWZc{6Fr`y3AkQ<_KN?$7kH-Lucjp=ui;9^OVfOBIhyj0<C^6@8qG0-Al>%`
z`(tDioC!qPBv+63a2jpB(WfWc<M{y?KG}~8jkQb>xe4zF(jM8q-d|4p1WI1qrq?Ku
zhz?j_L_#sYU%KB#mJSJtz32zHq?4fB7sfHOS%T9&nOl7F3$ETi6BTnDsJ7Runp1c|
zs@t9|>f7Y`=HSZ&o%{6L4~}}PNi!VY`aW*{m_*bgd7FJrvlu_{3WB8S4m=Slv~hSs
zt<_4+-Ez94`_Ax=#Nl&1&Q-Pc2WjKxo35tvWWU@yxC;WPg%|oX@j=vWf~Jctt{-YS
zL&?dlbr$zI8AJk%H!!^-Y&g|Kt0X{Hxcn#i(_guyoE0$Aj6yINLe2#>Hk?Eoc$i0$
zT4}wzeP5#obs4WM%b*~$RLFS+Q+9Y#ruR!Gk%%gAj93>wJ>i)9ux8NW3lh-eY|Vq2
zf7)OBD3e6TYrq=U(^~5E-x0{D?qeq%sGqX-9~KdS6R^lRw|rTem^{kQAoR*_E#uNY
zTHK_xG&0e(3toCcSkM_NL^bqQZ=VfAq(P?E$!hRALKb%P=+UrLdMAU3@f1h(fc?3b
z=R>s5v`4_<`NCuFTaxdyAKk;daed!j`dPtV($0vP!{^w6g79?Lqy;NRjl`=R+h=Q-
z@dfC!*s?a-aa?#5gM1#d|I%>@NXLJYCD6hi62IsYu$Aw-ni%JL_}+=oU$z86Y>Rq3
z7<sDW!Fwq_i{QbVMABFGZ6N@(Tv?6INYcVRSS+u73M$Uh*JoqAK|L(r^~=ZKM;hU1
z$~wldJ&<mv&}!{oUu*EVh3Q)E?6ZLbyrp|Y)*0TT+~?`xF?WvMDa~(O+H+rY;u(f4
zm7=-&W<^ViZ@RBh?i|11Zb=u!2MkaT<hF0fZM-8oI~8@0QquhmAUR!7$jq-lN}*Yt
zmO5aM)ARhd;8qusZ(RJ(LwIpFFs*w6mn6-BhL*_PqAv&o{HZD=3d)6RhYDZOxhune
ziIsRhFm*vf2b#omRp3nAE;iCDp8jT5H`;^8^Y^(Q6pT=SYl$thiapPMHZv8c$NRS}
zAc(*g=I&RJ;>C%^B;;>uOs&3(lUC<z-0e|;;PpI+e<O#(H8$6C^iz~mtFANscXc0;
zkh+QB9d?psICzK|9Z+3MY^it<!BBkozkNLOLd+Zvr7Tz}vQ4d~<@)Gm`di}z2LKS1
z4k(6*AVy(lE=sraYbdvHj?xD71iSr`oOse1PU}LK8k6=D%E0hzB$H2g?~m73KsXaj
zE2exE;R+SFo9`@bMdZ7ykZ!R{^pb&3$?j}9h8Wh_wc%8)H*CkHz9@J4lJ1S&KS3^v
z`T2p;r<&CEcA|)Lu;%BswEpc2MIX7&iAnC28e_;|r6D$>5&Ffg7$}Pr$$i0n#u*ut
z0q8+6UOk2vcWtMyo+s1Lq)49dE4@R{;AmP@y1bK~FVG0>y|9j>w%ZiOY%`V}2xnEI
z@Fn<d;D!VyO|46j4HYj$&v5SPJMtTR{j}OyAqU~sx1&GWl(Ws+e?zEA0!2-ZTQ~JV
z&h2(bEdp%x<~sHP*>pW&f&tx#`r2{rqF5786U!5OuLqPZQw8$<jL7!~o`{leRS*kB
z`jpz@9SNPlG9klQC%M5rBMLg)o^g(@?UO3$7F#POk9&Q#flWIUM<m16uisOY_0Ys)
zK*kt2ZAhH@p)N5kgQVGqh?@vUM#rIKYwE|%$Jd9_x<4ub&LfN9Q){!s=O}`e+oP5A
zb%!Xzo~aC9Q^I|v5?(}iMA0a%ipE_@?t7Z#h2GiHeCKwVPW71t=jDR1;+z4YTT}15
z&^p!U)0Irq9VTZiCTrPb?Y%01m;$}JNd|)7_hhGufE#?#q_~w!A7>)X$9t>!&BHyR
z0@5t4(PieN@0|X_ymV6E5*!HihCsmPTj2CDrFV%>z8|k#I@0w;yrc2ZiTN+nqNix8
zdh|1C+-1|wou6bms(E3i^yvCh#kvEyAc*0_GJMRZwBQLpt}5rzb?arN{4E)SQX;`+
zt>b#`?h}rOuh545gkz#>anNf_YNFwNOv3yesWhLR#qm?^{G@U?97G4Z9FjYLa+Bbx
zc6D3oAF`L6gH3!000%>J5_Z{tZ(Q@itNvX0an|5cUH9QlKJB?#&gVQvzFOMs`Ba-9
z;(GHZ94n)P*BUc*O^7`-K4xZeh(#A}y{=4^*@hp=gS3`4{)#ib-+5GJt?d(Bb14Rz
zz|66xy;3bH6BN|0yR9gVck<~-s`7#u3Hb=8rM|~WJZ$K5*x>=;cGWOiJ@Y$5ZZXn_
z#V$K_#TGM<vIIt1w;Ej7xOCB(X`qC$X2FbiHR8$o74ydx7-g1o0{%8$=nyIR;5#NK
ziP~6Pj>>lzlV?&IV%gL8G7Vi|j-|Rku9#@3tUmx=d+n@*FR}^e(#x%%S3~fb)@S)g
zeF(GyYiu6HEr|-+d(AC~chHSW1PUH^%3}${zjVrWs_i+;`50D+o@fM!e~*$rt0qAr
zgkao_<3fF?4#wy8^<Hm(rO9%~5B8cG5ISk8%o)A;h?m6*Gooy>+$1b~qF+q*@GDbA
z_MHO@U%oGiNE9ZvGxl_ET(vhNc%Xad$7YOP1XJC~HO(S^e@w>#K?@n(R~6KFh_ehR
z2tOjIFBtmbXZvXm^bL%3V*c}(6rAyKfW8)7p0$tTlxf#|?h|!R{;Qd<*M*+Q9@^rS
zBP|sQULWaatmLUOeVu=a_2+Z?>NPjlA1XwXH{5KgF&=eY?Y{f>d3L~^N}S~Iv9TL=
zsd0i*r#ueqE`&_RS-z;4>>E9YH+WNbz&EO$O=XmO_-b&aZ{S<Wh6(}m5AsimXT<bB
zTzAa-Tr(YJ*@NnS`bHEVGMlJeP=+dh2&}n3@f+TGvNh&Tp-%loF*@SYJK7Jy*7985
zvE><K(w5{KN^s(e!28u^11lQ7+2|DEEaSjd@k=wVvg;)ZQO^{zrAtlw$P^32hR0#^
z&YM$3v%W8C6!ehbF(l$0Y`qNePE6_?yq`txLh&X8(Z%TL5U-IH|6OeBZH7LlkF}IP
ztJC`nU){Db7ZF9kU?yCSF1}T!qQPM*Y}v+PxK}nVGQqxi`bncy7&SQNbqYGWeQ7Mw
ztBo>Qn&(p*AaI0u4p2e8Xs$_O&v;Cwlx=(iRb9%&LRL-sXfMo-W%+8D5HS?q+uY)4
zc#%E;JQ>TB@qIm!C{tQ)d~Qsr*lGIB8+``ZWOz=sZElz{TT?e#$+@X1#rKI2!jAY3
z5chLe1Y+s^Y@mAjnSxL)j+L@F`(MeKoGwc2Mh-fwSj8SHV4;?7I|4LiKH62NN5G)W
zbIw=2?~KzM-zpo+8a5K4c33ClhWu164yxC#$?@<eQ@bsuU|FUDY`IIl&mdl$a0~z?
zJO139-&~fNa}UgR=96qB-3{I<ypidZ7B=k<rgG|ZhV?}$uO2{&5OncOmyiruloWpM
zmJJ#ekqC6q5T*0w<A~3ZEa>jC7$2|iahu2`a^v7AABE5>BA~vg*d*%-dSe;FF$pV!
zvuQOh`7#fF;c5G5a889}$HdZQSvG%+PrnT%^+l0rkNMp7NZ8h!8eF&MN^C|9@^}o2
zx|2S#vXvG>W8;VI7_;f2!Wv2y3hc|zk$`<~vdeV~q+Xp06?Y<H-IeJBuq|2iptn`S
z(0E5k`ez()yya=|;i<Xxo~!P?U)VFq7NXO?cM>te<<ImviJ~?%(>kj%+bP60=dW_b
z)!qLHUWCL4Mj@CuI+X;edl&Zp=+W0s7dqV;vi=Kr@GAAA*6vbBQuz8Sl*-(KVXjVL
z$;L=g05iUr<YD+6^UPOKjWZ6%-5T_1Loc58mC7QANsu_)?OSkr)joHp?nEIgOZ*Qg
z4iZrU<ihwzM1}$?dDM0~7o9dtrOY4S58H`N#6!>#RvdOXge{;L%c<6U7kH^p6b^kK
z4bR}-T4o8OW+=AE8^0$6I#bIJBlr%rHt^`>w$_|16cp`Ff}94qC|Kj>i*~>Q0wz$i
zRTQ%8RT*4#m3U2<Lqxcjk;PM~mJSrY$_(~aexF-<bM-O<<nr)L^JYeuA`Gdott3kG
zF|g)WppegSa6NA3Mwle6y0kx<e&){jSqlIyz+tX~h4g$LAb@c@i6XlAay?2~ZCF0=
z3hztXyghu8=L_`sd6VFF{ci7&qeaS|w1AxGW2nQN9die(R}yO`6asn(K~I8Q8e!PL
zHE29OK3%%fW<guSW<_|t7O=<92563i(5=<ez(qlcnKcpRi;I`^o0E2(1U{){t9l@Z
zTv}{uVR57iF5?f=qz7Qm{*va5X)ybG&|q|c>AGc0zcv7p1t8N4m8*&-IdCG2S}epp
zx<I*r1wwuFe@f(oNl?5IOv%YixA3(nTwB`LDfAOZnN7L5em1QO&#Ald)Kr`Wyek_m
zOeLU2rapYJQkKGf#{8>W*L~zcoSuND?^E7PZm#QduqjODhrA8-Fou~~fREYI(w!Ss
zIEMEzlaL|OV=ogYzJ*#iU#sM<&lc_BHN!%36jU1pE<2BJ4u4P7w+o3|qPS<0bL{}?
z75=)+oC!scmgCR-H6gxi#ykjuEnWg_*kJUzqa@<R$mY>TpRBJV5`5=>HJy=)a3=<i
zdql%&T1BQ>ir4%>7f|v{Mv}gQ<JOOF(Pu}q&POt{Fl$lUT|WSB_I%5|lv0&d$Z=&|
z!`b4*XTB~-uo$3@V!uz&H4y#hU;=OYCD(5?ZirC}xxENcl3^yVayVE8u960U$4T{d
zHGmQGLqE*Qb(3otxp76S;RT*3JwTKnk&XG%ZIOvi=UXj0G<~v19@E%mfTmZ@*ZETd
zgW?Sfq`tzJ5NkTPq{XKp-_F_dS$|g<0ekmVn&-Z=e=Y__)*W?pMChIGd5Y+BD{7oH
zBwN=n0;#qqw=@f}u~p-=AM)RF`AT>fs&yGbU6q!=L>?Fm)eDzZ7BTp(<qHzDGQ2no
zJ0U)RFxMBh2e;5__jH%6=YKT&1e$Zihf!<8-8{F;PE1EJHL3+{(Yqr#k0oBe4xi)n
z3tYeCXxUud8Kt`TDuhUXzT^-g<e5o;d{KP=bQ-`B$+fq`;$`+qAJ%pA3Gi3UiHwLX
zN7K+7nQ7(K7Sh5xe02BG{Dh`27i!|pW@Oda7DzjPhYaVsM4xDA8qvqDb^PATKS&fI
z3waT$2!3K|1Z4_4Cf(**Tf>4*B5CTvEbo=IJ!<I?ESs3y-C)Y7Zqe+WWd<ztGe;$=
zM<sV%;%oeoV9h`iVScoO{yw@6XM)PWZS-dP^T`ZKLVGts6)1+DB?l*{x2ER?CqQCq
zJW)M@CB}2p;vHEBG%|Lx9R?Znsf$Eh^Ia9=g`J(bXVOkRCPwPY!ItWstXw`<m9RIC
zFch1}@mQY&r$(6FTj?Sp7^da5o~Xuv6C^;S9-K`U4@y@@G%)$p<ERy?P<J)?l?*Dq
zUB@LJj!c*4f%9tRfQEHPT<~k?=Q8M_obL}f6lzU9;!x8b{Y&b&VinCmP(xwG0hCWD
zHEn7f4-4Qypj$ukgKhyqL4k1qZaApbcj$TIGY-O!wAIaJT&(kT)ep95R(7*bBD6Nk
zbeaNGyk-+a1{@yfT}SxBA`C2CRZ5@eoXz5r;5&;k^<b4PxOB?UW+VQEFC08!!cbmO
zrUvBrrL6A+9zu?M+u<9!{Y(^^R1Fe@Ofx2CqGRY&XUeKB?Bp|lF2@=RJke8Nw6bk8
zKY}yNH8Lm#G2?d9-A9hWCrBi+s<hoV3g|U_CZx0=oD`A{GAjMYS%70HCw^o8cFnWK
zjm}8^+~OUg9bdwng!T?dt8ubyFZK>K<1iuxFg8>Gp+NJT;*iOm^R>f=9x@k~4x;b?
zi<2lUM;$A<Mp?!S-;Sj`DTz|Y^8$;o1J_tS*M<iQW%1;1jNYy(-d%lF*`rc*a$>i3
z=17I7UvT3U*JT7x=Ii2O!O$fdRWMt1n=xbdqhZg90@wDqdI@T>l$l4OV3>!nh-F2;
zYN4JFlMRgOkAga5C8!DnKeLVl7(~z<%gM=kYhU^;I&j=MV3BQZ(R8;G=}3`b$XF_=
z7TpW!V-aWVZJ>|*2Q6lG2Y#g58t{FICtJDF-bQm8MIs)7whJ3xU|S!3H$t=Q(oxk2
z98y0c$hEacMAd;s0^mZ>6c0Ar3fJoe$hNY6lHH!qW&yHpkZth8dfKPnQh5%J)2bL{
zSu@yJv=CF>qS9Mrlo)78z3L3!@(SszH^0iUM24$yPC$^3-j5XcDi+?m#(-(Jw!7vn
z52wTJiWEvRGYa4F0r7)0{)lj&k#*^L1!cFP^t3IaO>)r6%%?!b;R)D$XyoQ?KXSpu
zPcbS;`zKKECW4>hS4w-=zZ^-BUppdq5>){;A}78P9{~qqKAV)ZX(5(Zaw{Q=agyZA
zF}>}86%ht=%Zkv+$+0c34{Y~U1|Ir2Dy>~Z%x7-q+1H7afj*~diZvnsStg2(m&&5t
z<bRly_wqO|$uCcWjDf?x+=#fu*Ci+*mO0Nbsqg^LbQiJQ;LTZEuWudN)9F@Ch*4f5
z&7y^`2%Z_`Lt97EyM`UX)`ReVR@mfcx(aRcnlUm7JWc}aF2%QprvXxrL3qcMeA7AG
z$TL{<<jQ}El7x`#f|B}M`}TV6dgW|EFh)T2ut-GY018IXW%<zXn7g@2@tRMoiMt4C
zDV03ZA_Tas3IbD}M$P)gjYl68_3Ax!@ffu4M?YfR^kf0h)tkg`Qd2p`JMEjoIM+hr
zKE?VyJ+jRR2I&}`hUarL4N!fawB<P;AQXJthdh|$s`x1z@cGy+d`u%OfO#fl(0cU~
ziICvpiM8UOvIjQ69shusG1cw|m_f=a*QmP0lG|4pAS-=W)Ofq0A}5VPsm#{a8(3RA
za_F3EZzlxL2wWZy8vv$CIyy-4RCgQ$RJ@gBo$;tKWYIV`{==eVq&9%*mNoF}0Jxsi
zbdnqN*$~DMc76*nIvz0QU?$!q^F#o1hHq3ew&ts9TW>}zRE69((21{?fk9x|PFS74
zI)C=L8g%h=Ik_SPkHck~;SW{Uc66Pmpfz5Rv&-PI76dYR_BasRhu`{vVjyD<z?u!E
z`OB=`qVnqi>=9;)DK`L7VGss5n;NRar)r=H#uB}tSfH;*#B@K2*3%f|Pw>8KH<a$U
zbJ^q`7kH(g73z!T;^fl{UwU3z9UcRg<RGvl6vUzNz`LS8;GOIhuuXs28OpHc1&YGf
zTswFI0Osm_u{M-1Z4R(JY&x<SFc(?|Zl3qy!V@#s#eqyq4?S93KyHvDro+a`+gg3D
z*iRG`rBwlAaQ{HMXa9dtUm^C)*-`Jn>xxAsI<F&Y2b~jQ?XNIkf)r*@eyZO3q!$Ve
z1!mqY%p?Jktg`N>71@b_FfXpKb?<;hUsE~EbOS!IWov`9Oz-8f=r3m5RV<_keEE^0
z$gu~|)E~MNzv*JLyXstezA5#!ZIOz*hRzn9vR>0ud22E5!a@<r0)&2=IpbbSNGZ{w
z$U6;Z7@pt9jLpT|g{j%sro6!`SrRI7LDPEWu$?%4A9)+liE;5R3usYGkYXU<-I*g*
z6IlUvvhqATsFt$`-8s>;zEueDK>`($KsDs!6v65h4kHg~_a_)6B61Y2>C_z;Ju&NK
zzIc`6@InwLK0n|?7i7^l;FZ3>=97ZRCrFT8WzjLPgs$;w6058lM*ZUO^WqBi3({gs
zttyij7kL7d{jLBpaX=lV610CYVx@C#URLUo(+ry>2D1-%8x_2+!&Gz9KD83pjq?lS
zfPmizHShBBazGbAHFfm|V`5YMq6Q<9t81QdbhtoBfGsSfo5{=9=}fQThQKGbD&~FE
zP)d&0;;;|isQt0zcFo{|X7UX?jSfhko)YKP>qSSLwVn`udRG`96mcP>WbDl>z(j=H
zaR9S2`ch$&HuyZCiZF{X1ydG5oMlIacDqXG7Nn(^IWA==_k*sn_%n1u^p%AuQ>NvB
z2-k0Mu4D?&=%9`A!86V6K4_%5r-bp^IrFhX39|>7O*YPZ&;=-lCnBP2)tGI6m_Udv
zmurr%)bk|4IgBq?ezEouiryZ<fNU;jp$}1+rykWD5#PKgp<jWq`ekb9=zH}pcKxkH
zd2DqY6(ENy2bFB4lZJK>3%yt^H3yo=tO%|cCg4gGQrTKx!ZdZn^Y07k5u}X<>tvOr
z&c_Gc0e4TrHyLIsWVKIrUx*#KfLb>V_1SJjBh+A1CCuaWb&!pkoiEj+;V>hvfLf(R
zR+-SmGBL+a6{_>Z(4yl7B~dmSBh2;#==vI@P)#9OGvT(WY%OQvsRRKFm!Z&Y*b{B)
z5q5>mDEE7M1p$U@B$N)#U0aSHTfcC5T3(L@Qu`}~1_RjyiZ`NHg~;L2JfqSQYvk4z
z+if^7L``5HQEGQjOzidvd7#6%9G1OBYuAH{-_<7>97-Q*#DA!L{{&_~{hTvw7lcec
zIu1lXb0z_pJER+s+o9ILX2?CL;z4@Xpk-L8Nac-wgz1%dKD#fG!uTGCGAGqsYgSV4
z0W{Pr$!WPIn=_%JoWPYoN*K_iFo%R=`i{);XuAt0j(S^%I4@6LtPVbp=M2`c0U?#>
z2)p?I*!t?QD7Wry89@h7Iuwr5oq~YU(%p#CA|a)Klz_lkAW9A0iZn<!0wV?uN=U;<
zibzN|-+pFL;r)I8oXhJl^URL5@3q#w?zJC?3#2f4HQlE9>y5*LS`6_!nW?7K)8?2)
zM?@uj*N|q@)W0niQtr~ke3=B!X&j`^=(x}(5P<43TLqPH)nGSW4*mK|lG(ETA1%Nm
zO}hJ|IK-cs`7V)4gEAo~AOyBT<v}3m53x^lcj(wHVHHvy<kA>u#&Ez|a3Hj$D4(X_
zHthRWR74_H$O_!Hf%!a3=;Ec~`d#juPUVvUjTQLKn7k#3J}hFcL8N`Z%zC<iUcNt+
z2OwoC_!&ko{*Mp>s&HVcdEA?MlXP5GMC+J{fxQ|<XKaYPRO75}-)JH~mUI?)oMjlT
z%{>Xu$U2ARZkG9rTPR%YLWlWFF%85icK)}K3xCQ8QaK835bTl`4M22SU;>D&M)oBe
zByZ<y9{k>sA$notyNo%BzmO;{<d{RTyoqCy$3oWyV~~4illAu$XN!wjud+>vDt0;l
zg&-fo%2q19bpziCz~|TUjJ+rU@_#u<h*#Pf;0mo5=p>}gyukFJq>Y9}ieBJ79K+!9
zQy`>bUGHnSf2ef$lmyxWe7>$~QB<8qJRuxDh~U|=YPl>sZ5HN`@}w^v(@-Ey3iP+8
zT{p>XLTYLQhl+u!ClacMJSX;_hkVr)0?;*f@VG{^;=%`xx_g`<HcAkHq@zL@B3tkN
zygOGZ6a|X}F^CD7=bdkYY}0MlvDv0N)-iX-86hwQ_`!&_o7*TZxVRI%XG+sDKMWFt
z-;lVnV1<zfI+Pq30(^caSd?SJ&V7IbsUXiue+%>7<Z#!LxRK$tx{ZSCs;Y2&@bM-0
zLz;6x=zybpfc3&7GVp%^I?7Xc>fb$B`B4yBn;qzlnTRx)2>zD=i1kxS#e}?kbJ@g`
znKcfEWMUY3PT+GCYfLof>cNc~%dNdj&1z}@Rl`xA2aG~pi0e5^td(n`Fj)vV67=+f
z?X-J_3&W2MiGCsgg@A+_dgEqPxWGhTrNP9W>Bfzc^L(MGO5uFI_#gZ1^TMkjZZJia
zeTKPl##9yAH+Qu^UsSyY$dhPPMzs3W-W$av>x}bAePa7dcMoIN0@!}mqhej)*AEm|
zs<~ePbUD0c0S<9!%tq!eD?Q4H(<f8nX*MrDx{Vx(h7@$t+PJ1}TW<w;_#(!NtKB|v
z$qz)S1Ka-wBQ{wYzER=ytW}E7cJW3GgE+KYp#7q}^;fQ9ipY8pb4(g7%ogOhQJ-})
z1;JmnWU&29`wn9q0=XiyACR1?=K=sf2+RRroUv~i!O(*<(9#Qaom_d!#sy490m?vA
z-?H7zy3ed0MG?u0UvubwuHbm^T+qJxU^;x}*UD$~!lWHIB{=|-vy#tLL1;#S6~S1P
zYOHFpwXQmw&c{5G&13|@)phfz8qt<VP6MD5S(9kpIv?b3s|4as-*vrDnXcnCcSgHv
zd9h%nW+eA;{8-QAeUH80I6}L4F%5ha2ZVLvr!lq>PI}I>EamBSo8oN0U0hVrt~aNE
zqAUsILH+BeyHq0Fo?LVP1jSr@gWnv$!F*1F;T)|SmbmWPAukJ%rn5t<Pao!v@U1#n
z`D#Yzr#;B<BlKuI0XSUrlV=>>HiAq)t!^b*vZ>HSeaY#8P$xO^-F<VoKBz3$7`!!%
zdJ3h$=U$>FV#G~<U-#Q`jcX6P{D6PM1v(GFlrNFp7#QmtbH#AY3USW?5*`3vlOhx&
zo|opvwae&BnVeoLzVIRYg9T}P82A=y2(*MOo5c>8R2kpxl~BF=;0!wyhp=P$2lKe|
z*6UZ5{lp@L4)uMASbPWPn;$;L-CleSUbX|at3U=DK9U5!K{=mwn`j_i4_!ulkMxu4
zGIjoS4LbM1Af>)uHDEIBYv5rt!<22Sp$*C$HU6#huL+Zr$PBN1u+(M7docC{4+F;}
zf#uTIGZz1QI9hr0YlyNahMxA339^C01Znzpn0|1kS~*G+X>Ts=Ci9C>8?4nhFaJ`l
zW_{!qpA7K=1p1d2o;HEtBb~bcviDp~an}sXL2;UMx~AMAX{^FqSghv$Ib5)@&Ul#_
zI5>bs-~xVCH3(q=zEC6ZZIVlO7H)kC7<Nmo(m=X;8W%@^!oLMqp~<oka>2E8_^Pmh
zubH>EH*dKk97HDJWS;0$9y+*85|eFpT8q+Cro$fgbZXBItITDB`Yt+Le3E`jU}80?
zAB8ZxRz!)Z3_YnD{!Qgorq-yO6XJOO5%8-J`}mf2lS0qD+zxUu+V~1YU@f1aba^!O
z@6smOx(;0m6cs-I_729z{6XKuHp{{ESHVUz<H8F!fVyp&5g-lUOKuFtg9$Ow(6nq2
zc#WnIzEdcsiLhgyr;XW#{4f%}`+m~i*zm%W^<2{WA@{Ky{oG;}<%SGkx_1yeHWF7Q
zZ(Er~&)x@`gCE6_m=H^vehiio5>yv%^WXm4+KMS}+i$Wb5W9d_YYAn_#rp9Z=*KqS
z=qM)NeU{PZD5R+xrJE3y183|z;jTv4$bzrU=U)c)gQ6m5{gkgmpg(GQgp%@${8-Ta
z;06ZKNzPTHbXmo$8rq78&pyOo0ArTHUm_ooX)eap{G}a$vVuk-CvWC4(jY-^z8KNB
zJAYwF1{eI82bGfq!%B_G`;ldIj6F5qLmB}HdA}UeQVo3ZzZ{e&2x{r(usnYlE6)9g
z?jTI#QPZ`lW+XWccFXw+ivR|hB8^{dzqAdw@hw+YH>%l=IpxN(7GygN@J)N9W0-Ai
z^MW?(eVaZvkpo{ip)=s<p(5s4YR9BJCv?s0-OI|9vW8C*u#<wcA=CRMil0No7E8Oi
zE4p7)fkg-cyOzV==78C?u*;ZkApMZ?=GOrcEVQ1V#&DGbSl+TdgsNO9wxo-ZtNv-h
z3vs8nR(0DnBRAn6_0ZYTnVe>{5w^l~=j49dp<^#H6=BZRILDIe+?uA^X2+(%`1zC`
zII}8nb;TVK!Wc_rp@<c*Qu)N?Z5a|dfJ^9pCX2PnV>haAFKWrSUsMP-WLe#4HIiTG
z{p6ju+Z=)K*SPqJOm^5;ihj@Uf{RhV{v%$y8Mhu}t)hcX<H7AgnNhlXU&sF2nENP_
zP7fmm>L$wHCmFFJ#4eu3m5H8*J1ztDgcKUC1Go_(k*;dH315Ld5;#Quy0%d$=IS<9
zV!8DUnE@sC(r8WC==l53UybJ4TtigZRL1*8#4lim4v#cN^F+k6wf$F-x2mn=FYSf5
z`P|v7GkN-OW>6ct1IMBm_k`rN*K|`SQ!)fDu-N+G-!2jYVuya-S<(op|KcSa@&g{f
z+_SD94PqM{8hSwN)1&c19G+&3-d9Z?6N{KXc$DUcyv#fGUk|o%jvfh{-rj#w5W3T+
z^o)rD#sQIXk7P4|9|GRe2iFc5<yz#%#ka<uk!>LaY52T*Nof)j-H+1XXAEB{7AvZK
zXN?iAI-LL`7Ix#gxt+#Lw`w;GupESkXn>9R&CS-aHy@`hGu~|ngcS4#dVUI=lJyY*
z37d=Cy|Uf`E@B%1#=n3-|LjPJ7#2U-<580twu>F?J_W->_*08zD5B_{_s9#}vm^3^
z4GQoxswk0Ghmx9PxTU5SI&`|Ys&V0_OY7|qhS5+j8G@0}wS837B{IOK()O7p`HOcF
zL)b0#2d^W3S5wa?HU+9pyMu)>766g^tsp#8`}7n$KsjAVb8UI!n|D?f(HXUXXBRIz
zP`xu)J~>6h)qVG?nnn&YLFMkH2npUTau<n<hD$eC&m{jcEOxz5wgW|1T`WtePe#ON
zT|^a^JC|msT$3amG1Ee_gP}GIH;x_zZ%9Ubba$1Z4+1fIswbx<hNX}ORCDiHI`5d^
zrk8~G;~uu})(Xt?4Io&9-E6tXjG`2Om2yvcsaS_9@6*O4yACihaaTM5I$MOnuisFE
z7tlhV74e*2nkoNQLwqK6l~j3GR8*Ty4WraTL=wI`r^H4lbbt6EhsP^Pm^>+(txNuv
zH90U$aB5R|v_`Zl{7CxyjbAw<9P+sS<ElX1>P4~To!Z|{z9T|!Vv6T}Cb6`4(nxlo
zTfkf6s^4JiBOA=62c{bqmmtTM3Zw1tv!KJ9G7Fncao^nugX(^m_(2bGF@grNZtiTn
z*{8rYfRt;Bx#<x2KjKMxTJK%6x=KsS`Ja7$(E^Rl;H6TUO`_cTsM&kW^Q6CNY(q9q
zD<`n{X%HlbHBhU2aYULDq{X;o#3(eG8IDwsM_p$63SF4e&+pxe-vuP$*`#N&6C+Uy
z^-p-{HvZtYUv%>4%kt=h+l!{CiM-a;(6LTt2?P&k183NC3X)P`auO0PExAwRsTQyC
zA8}<hX$#m!P-&+!wx_k15>|6@S8ppKrTb0>H5BJ%u_tp{8s@MUvP3+h4@w-=iEmpO
zEBwhw5zl&mvvzT3-p%<f@8XX4!p;h9!Hea~%4>Kxl=R0EpJNx~qanJCXL0{@x`R!W
zf>|O`qAbg$B1wIpvDqVPs6(JutLw#nF^a~>qV~!+b}K%cIjjBC%EV1;r<)h=mo%!^
zy*Hs%D}S#>JLT1ub~;1N`_}@U&(vwnNQqOYA(X#F%FbQur>N<;?;jV;lNxWK^8$!b
zFgsDW2dj;9@Or;q|CtsX#fvypIuqLC7z+WbmsN>+y)xa--g_^3Dm^?Y<HlZ|c9f~P
zHDBZwH-}K`hzD}OHOfb&#X+U{-Tu9U0sRz#l<{H?m)_RP@-#cGb8D*yDcPnAy3DyT
z_CcTbcXT@?*8C1Oqz)E$#MDxkWRVy5TFr=6mornQbefH6Exp>NLNg=1@+Q+28>5vp
z_dHZ3cvl)fM=P^vIx1~7JwvKL@Ocp|Dd}dp{VXNlxXPAnz>^Y;QyOgxdrG;nVE=~6
zCb~`Xj!zU{%Af5%N6v3lWfgE*Lwu?9WW@>Kg{COY8T3sxHLz|U*gDpG!Ec~{=GNXF
z`@Xgq9@P={OC3j%<<I3IJsRkM?(()P*7Y4G7jsQ;S(!k`6qU~J8eN)0Ac4#pT2XP?
zSB0zjCbf458`bu;m1|C7qA%z#4zlKNf9l?AO)P2jN&^mC9LAm2GSS3wxD%M|0RO=u
z$TPhz-k|FnV%dmyE>!98&m@KZs`W9uJnaLY%WxXOttOzKgMl(Gm+^l(%2A%U1geW;
z+SzQH(XP+b3NChviT~BOi3gxQTwfLu;kmO*DFe3qDwspW%7o+L3SdYP0GHC>CccG{
zK1Ib&;gtxMg!4+=7FcWXmtMsaR`T*@O{lxNt$)WKyo%$7Qg26c^Z(2tf}%2eE@L~%
zro*?Ot;1<HI6^$64-4hs@iS!o4%Ul~cJ`G-AUTeWJwcfLyz2L#uKhp1bQbU1dY<>J
zlF858ch|KeOs{`ne|f0HK}^D5#;)#XIBdsHLRWT7k%JADRMo(u=HCs?DnvQ928_kJ
zByntcGxr*J{@3_d`Orzh9yide?s(Qf2CU8y8%oA*=bh&tS<zuQ11_BQn7o*{^)Zr7
z^3HMzOKob)#Q8skD7d&0%1t3dVvp|&@K>AQJO_^jvHh9kp;6_qJIxipHt0=vaKK-~
zQRd4UzoYQiV1}s7Im#TFk|FEmPTRb4IP*W2gv(U;vrAw1C*9cEL+jgP(V79L$zKAU
z1=dhd*Kvnk?d}_Gx1zz%mxG^o913JR9QdCl;N*6)&#yFkSSvZ55vt^6`?S#+llMYD
zp~Z{i(H(G8CF(`Q{c&!T3};7$>Yob;u!cbIgpgVMqAegM+f{`82@6#f|6?cS_Au`B
zO=Eg~A_y;}$N!3Du|EvnytMmkT+F7uUto>SD!ZFytiXru{vkmDepwJlq$DDKkS0Ix
zChQawNhL4<6knj{pQ&LMK-w7h`b@w~wa>5I*ZbOCtCg#%7@8US<B|c6hg5b|?Qi=)
zF!vHL_utuvw)^KWXT_oh^UHJ<-Qyo58r!INnS46s%zt=9=}DeRH=Dd|pP$Wb2$Utj
zZGcWvhe_C<iY#^!1A|d-tH$rTt%=G`(}g?q^;5k0+uUF1h|s?8?V;=aVWWJHU6USk
z<WcD~SJ}s7V(P#9NJ)L8#Nh|e8*7``*>lQnlypre{;VLq8iK8TH%Z+wSmfFHP(8Sw
zng3J0wf4B<m#lnLLEBJ|UA6xlot@eA%T>7`Q2$6b6FIeKgNkUi?X_)1a;Zlmq=BJe
zAvcFv0O9e+0KmTxIRtqf#GP(@NlYczmBsYju6+3Ocf=jILu6)1OX=e8k~<$+SYT1m
zLAq+Bad^%S7t?}6tBLrbR&l7ZD04I($KQ^uNCR_O6|ae52QU2xp;MP}Q(1r&8E!zO
z)guFV#R{|cT6A>cO8UO6%QCMML2~A=tTdHtDYGZgRIGyOp@%D1*nfy-{KByKM7n!_
z&nd<xiYT3`D6gNX?^O}i?Gj^2w=e=K{69D`g%_0F)Aas+XVRYZl;R~E#1g=yrygwB
zO3&Yc*l{`yExYlR>Wpv*B;D;A6*~LjZx0Q<1}uLlGq~-F0|f6cba|@(k0-=FhaTKv
z&to=bY`~c`pFyQ8NTwAR^mia@cxqDlev?ISrd`_~6mNfs3;h4XqjWpo!n@55TiOaW
ziEGw2N?wlbCrSTqn9?Kv*&+4N%kTQC^CF*AyULlX`9GydIeBbN=@HO4HkKnZfuE8^
znj)oP|Jvw;DEYh7)g5O;doSaDAk9hzM!{(zh`j>p@9{_nl1%l*?TVN@*JsV6yd3(i
z;%)zS{R^AGz^eQedT=DDbd)QtfE48qDuFa074O@0fSEXB66aQb3Mwh-&Y70K`-Q--
zwD-O5y`6)yAU%yJCM*o&6yv}IA%E8C1nd@kxIEQUD<@5mC5hhF>s+=HHtxe5?K23T
z>s<*6RB3)s&cK|hz{#8$N<Su4XVySx<R|<zd)IBcGPzxu*2hYOQSf150~Zjvmsz|Y
z>XzS9$DJTQk8lPiMMF(a{1<jH1Gz=`@}W_k_SX3YQIm|31`Sm}${gAX;w@YpUEFus
zhIT>zG+=%U!-6K>9_eBPTSdl%f%SBqa#R{s9gmqsep2e;{78&&YHsW9krzlzCO*6v
z>?j|^avdCPi-P3we=7j!7=4_@WxZS9wr{ITt~+jxlJs7j{I~x7@&0>#_cr<`eoTZ`
zX#0=^Mu1JcGQw)|XN=H30#Y-VxAtBD2d!eqW@~4_Ak#N1j=$cC*$~CoVM*J#zeHM3
z5*Q4=-*3z!f9(7Du==gt?IaPYo1%TjHZ7~EmFsDL?_cppb>r>Q2@ZTNuy_Qp_=eJ_
zDc0G4`vmNvh@9Q{oiFj2j-h3SE^~XJ$u)w1Yb>6EqT<_1h^M~1e~uM*oE{(pQH9dc
z&OVaBDCt|DYInAOG_ke+!c|@?;#B+-cLLO!C-r2Xa%;ao<Uu!@9OFraAOSG{)RXHl
zH9B-Oas(91pFXgdF%O85;Y-{8+d~_|%DZFc?n}1@o~bX<lUM;e)~=H%{%faTF93mf
zosa9Hw#oom1Rd22naICbLBd`D4q;sRYSENLWH(7xK3Kh<UV_Ijz1U1Z{oGwLj@eN1
zqCL{_1Q$>7Z{@wyA?$OC1vixw%sWQ}v`-ybcyNW@tyu~09A3BZw4xN!Pm8UL-VuP<
zIXs&=oM;xV3nhL-t@TE|*=J;5slj$XXR3w$_cw~eAfJOOCeD28`_pFFEhlZ;QKEC~
z?=wI^d>$5RWV`w*1J8RDaR%ytV4y6cWrC{me~+@92D*G|&h$=tkA<L^QP;~RO8UPp
z(qH-tEWc(hQ}bT({ptMB0P-MaxPy`Z)1LBRPYi4MXSgJ>g(!31JK@SR%EI>>0+ujL
z!68^^k~?jAB$Na%!tfC6;1~vVIfi4x^*>)`WYYR^K~%*$MmhdM+!x%xet`tk@v0<;
zdXo3`EMerAe}nLd?11#dpRZvSvxP&e(sS=(UApV@xKZ7X`&W-#LyNE>>>m;wcs3V)
zjh+B}In_zk@?S3%urBl=1S@9pB8}Lj*f=-m80r744PX`nOxAx+JA9~h=8y~Bgb-kH
zqzj`r_%nnYX00DQbw(MK$o0G~rO*Esm3}GGzsh$jvvEQ4S5rQdBK<XbOJIBqV?+Z-
zY(Sd+4x+a;z`7=rH)1#I)nl@M{dC}1FN}5%k6u}i+TN9+zX87d-t_GIV_&vJ$lUua
z6DLXPqEP(r%rPsX+D5<jd~-5!q3OPC2hH34_T;}s1@n$4pokt@VSVsZ6=`bPO@HR7
z#Vq1&KG5-Dd<0?c>Ke(v5rH|&ke+KJUc9@^rz9#IsW$oVq(Fuy)oj^aposVXKcVKh
z`E`SHrbWJ+!v|aKx-n_gxJRN`U?>jisCQi5-!^TXNZ{%KDes8LTs(SQ;0osaQa(;9
z`}cTjIb*f3Ei>2wpttZ;t}j1Mpt`(Iz>X_J1_s{c7zW9|{|@jGxW;~~n)+MRPp^?x
z2Ibv4jzA*@+&??*kE+Bo_Q<^~6mfI2>+FPYz-iK~!V~pHIsE+pQUwSqN&!^R`PS6<
z>38vdRna&a;p}UotYU<0Xf;0IBDA2;!-f3l`ao+w9PaHjJ)NB($=BB`bhhDd+yNH0
zg&ECJ`*v%1)h}}PO2{>H9k6SxpBh^MQLTp~wKLy+*(%dSNS{x;*)&z}=O-7xB0aBg
zOEKj^<@}q-&d7lfq<f_O={Rd8Nt5Di<mY`FZ+G$fi{YzjlbR~)PmB{R2ia{O4~r<R
z$Et;@c*!j*ruC#)r3dI02ShGQ<>+m^5|&tfW#Te2{rk+sA^IX&L8SCuBtykFU9E=Y
zt=PkVIgJI+68g?w<ki!OE5)a8k04#bNx+&vw`T<6Pd!}}b?1=$8@FJU>}xDSO637M
zgX;S|^{>3*()_|!Q|+udC-S6JtIfI=r?zX=HdGAxR@QWKvSuYMP4@h_C&WVzNv}Hy
zH<P&xDPh6h32U3v7b1^d5~_9sRUgwVTz9_#lucn)V8``Yl|n$*!X5n0)GfZgta_Q~
zL8EDJ)$9ReXLM<18ns-m`apNc!?ND$gDrRY1D@bLE01Q^%<>-^bMv&EJC}*DxbR8=
zmK$@nHY2haCt^Kc8~5{(h)0VOM<i)*IC|ej|Jfy6#0;48Cq+Q8S@56<jW)DRW_yvf
zDt8NzyHUt_t}UV>UehWUTILakSrO!cg34Qc+f{cXuilX;+UCPNInw;a=lMTc00Z;h
z^xO=!fkOM3G!139DT$=+f96X9*D?R_gQ|Aizhm{`wG1XbMqptq<MABNrFZe(*^lZu
z<`0>@88pk*Q()<A>RIjSygP0ngnKb!$1XjOe&bs3=BuD|Bgd(huXbl_4%IdjlsCF{
zhjaDraj}f~y<1%RcdG}eNRuor+7t&XetJX5lgbX9S5SolD0qeOIXGU=aSgM)?N_U?
ze9|0W4K-qA$fCKYlHJ88G+tk5WiWE_hRQew2AP%2-Rd?OVtBt3=Q022%Eky7FivMl
z4IwBvuqyT8ku(pON3Tp$98h?hz(%j$W8-j(Z6k^rdW{STyX;m&^!>XQu5k03Y@9D=
ziP7qp+ZFv*V*}?@uDD=+;-fBDw2gVK`dgD@YNC?KXKnu`B_JWJ#7jQM6a32q^=M!>
zamhL2w|N!#wJ25OeZaan!0I+@#ac5jFlb)j-TICs;vV|`A4Rf%C{Lzq$)ptW-fsAs
z&rZ>j3=(@keGaT*J-@Rjd({n;0_;w^9El*btPv@ft!Rlv;$pUpLuAnpphOf07-Kx4
z_}WTJa+Y6&idRJ2DErnmA!2#|LHx=Cx#((Tx9vKU9uM95NBo$}9_W9<a#ubH$1H0v
zUh_SITrIrl(P-Oy#@g%muaY2UK#uyU5g45vBYLLnvTe^8=Rm@gSBaXN5PriK6rRtU
zw!3=G`?_v#MEa4rzSN%P;~2+)OJC^K{N=^v_EW}Q_9vVJ|2lsFYJso>72)liy`lAY
zY#*I0gEB#w<2lOpmISVWnrYAN{<ZDVv?^RR%4+NG#DGtMhcHvEHnr}%=`j|p4Odfu
z#)|k!aE6!Gt2WMrrTcDgQrqPt6JTUTZ)*d7<XK=k>)j_{ZR9tX14E>L!0J3ld1>DI
z#k7sF42UtS6H5qgF|sV#D33*LmC?kxyydB}kWK3%r-E?&)i2oNt?wmjb;ddm2Z(2+
zk34^z9WBb#oxC4&<@y_4Pk@^ZHRNOrB(^w<SDcBP$Cy?;RJ~NwT1g~DVGFz^_LJM?
zI_o@h3$#Ype`7)gdIp$?+lb!5_)TxWo68YoI&twwf}==Pv~$dSU!r4Wh4E<@B3LI#
zZi70(5mGZmiA^uErjN3%Y$%Jmbe(oN0gIsC3~8%>-9=<pk<2G0yftwJ)89+4hZytm
z-Rsblyji{OSC>A=@Y?#{4ToSW7zit8r4oj^c6WQj_&8VxK$Jbnn2nGw!+SRo8Q1lB
zg2nqlSBg5w4E}@zjode}=6YLm&nlKw8DsAJ+6W`>KAWGvvaNb9l1n_^ID#9}=4z;e
zOA(H{oK$c<9pF}3+^*6h&JlRp%oQO!V_e^*>7C)y;i!WVt5&N_pZq4V^my}I+Iip@
z!_ENKXUSQhUZ(4%<>d^PzrOMWjFQsxN)*k&y@je0x~v}*M4F|*V&u%Erby4F`16T>
zKXEfhKjXSGAMKct_*3aG;3k-Nnz|Xrw$!FYyg$;OVEEQeF`Ss8_}%Rn#&vL-BfCk$
zo-1v}njl$}$L4YlrU~NJPmN#^%Q?{|bR*XiE{Z8Hur>d3<7WT~R*C(T@6)O6f)Gc=
zAjrxukybKopxfGYg-uvy=PzZ;{o#J0FyIjx{{U@5@mZ-eoD{b#m;j~;8z__FwlJYz
z7!m0!(?uq4+j~9Gn4-XCWN=DRGmDC<vrx_7Hen_s$0%mPQ1PSQv?|xDuPyJ9=_TYG
zu@SwSsD=LGiIeK5liktLoW$0~FbJN}$m~E=Ij*sQTbq=;HC_1XASO5mb^eTdt@EwX
zb2Y5%CXgl>E+>sHr~LF-hsA`=b_V{D<!JC6PUW;UKJ9SZBTbb=QVB$X*GvgtlVpW}
zg}zmg8LeeaH*L10*O<r=PF^XS(JCadR`AVj>P@(+ra4$v`ctr6?||v5fj(yS;A=-)
zujxtf5WyS;L-+3PEPT+0tA73g2k~u(NjL?f7=N=NTBOM}UX8~}Eb=i)U<Il_CfiTD
zD?NJ7jCL`n6xzWC(i=&1+oi3Nlp2i4BXKW~KYKjPt&-RqwJc4yDgz6luNr?9j!Taw
zoJpZC>9+Ech44pO53i25gwGHH5gutMF`{?HI(*UGyV6}nuWeZV-Bfb|=1j?7;UQcm
z;=`A}_e1|7ScYB^8%CHu^-@T*pe3M#a&sgT5%7=tD7TW~dUMs&`9b#bn75*ok$|Fy
zy+N;fjp_29{iPJ(YW{)Wk_civ`ocls1|svqmGz|Cn@>?`0HM_5!%mXGK_HV@`bLLo
zrpYxrZKOd>nG@H7p-^F2m0M(~pa>&5m45XM;L_`>tTm=x7lmHFzjM@*H(&U-Qj6Ho
z*5F)w^&h~cme)<^;keA`7n-9t_a_^t)}570U_lBLK;y9lF6CsKJ70NaK_2RXfa}RX
z!ERx;>l;f!E@w%bBT<My>Its!-J-71)8)0h8{}RIF(jVxCkQztiQpgQpb_W(R2XCq
z-V@?sEz3ekQ{a75HOx0YaETBP!fS#qbJH_2FZ9?8w@zPFiPxSVXggZ|K+I4(ZEz(F
z(MtJ~!y-U^MCj(VQN>rEVWYjvWR9zN0<9|XIulo30{OuKd8MXK!PNH3E6jR|SD?%n
zztpt-qTRiJCgtcIwJ<8B0fxByw7wUeWK!j&0S4jnJ`aTA{ThwLboDwdPM*k@K&yTG
zk#r%<G%_>VP;+eNaj+6USW6I&J@s*}OXFfmx1&H59o;&9UK~1ymJDM4^J8bNjZMW%
zg@94`@Pcj*|IP%1%mH0dmc`AKdxYh>AIHh55zK_U*YguWA<76z`UC1%+|`GcIJ&vf
zcj00^+xOc?I1OnE+%`$1t$n39fp+t`j1%=+ge#^>Pf^$L-o07lj^jg?ipgQ9o1OH$
z(vQ`r>(2L{!mu{1T>P<ozTeVT*u04rk3<Om(n5H<#5765VkJxIv^iI!ZvKY2=9HQ=
z<(GIq+Nv+oRwljNUV{5H5tjI|`F&Ytma3=kYFu|7WQ4S7#&<zvpvtnNguH+m`k?`X
zBh<Dwkg>j+nHrxmTn1IMjQlc~mfpd9P3OXC){3Mgc3HE)!ODQ;7qqCRl%*StT<Cs=
zkznn!YBf<e*3~v<haEW%Ku3uuoUX&<-67$zelpxqdMO^H>pABo9+P_B#WCxz4%Lxd
zXFDqij04UrkK*llxg+UlB6Ut#<+c|XsSEDsIsrTN#igmcqAXXB+&>C}3<tcr*yaq)
zP)D&KeZXYueY$h?9Ach&DR-uK)<Cy9EwhdjxHtN+D>z&&wXI&;2RzQb9Kc+QD9z7P
z4&|m}Ua<tW4Lr&PPeI6Vkga5CY7VI?z49F3R(v4MBE1$4&f=Y|S81B9JvR^&R{?4>
zvX~#FgBL86ul`Jt;}pIHG8lvnyzu2X)6v~F#@n5XbC}qf7z{Cu0{fzW-atd%d>A(R
z)vK^gifQ(TbN$E1kVC}T=xkN!tQc-lL~6%I0}EioiPm<mR{wXsB&bO^AR=AA04|(T
z82-`#5G9`Jo<s-%!p#92&F+^{z8V?p0(9-p;LCgnsisx@xrKUh)3Q5&)N!T;xwV<y
za_Hw;w!I@RT&WP%9_O$fh`W)w?G8AIK@I+kiU3f+@KW<!!!2iB5*yEh;hn(gTg$2V
z<gPtt4jY8tNWsXBGEl$-TbOmj{0HSu5W8grYA*MNJNi7uUjtYo^ZMPBAMDTPFrt{P
zQ(?K7s}9Qmn|TIT@^;lep~aWS1Yoebh~C;yKsFs2`|ZuK3f*oksuPgJuU{RZ%3J8s
zNh{P!-*~d~&onxNEE~u)wepGPW)C_o1;2pH5M;>NCj$(W_J4nI3IEe-g_$uY8_Gag
zzeD&gTS-{ZP;eL96|kieH#io<=gz)=3cV*o&IFyZIul1YU))+QR}0?)(&TL@a9%Ol
zc34gL0vyx7r7PF<FF+1z6sAdb$Z%$hXfEc!JsWokEx+kU-8uu}^E85CrrG~m*ZkPB
zPpksb?jYfr=@)c@vu1WFPjHW7jTRwJok|YX&#wv1GMru>3gdA2=6JNM{_;D^06EMp
zU-U$<>6AX%l474Sz5jS4h``-dufB-+B(@o-jBikVtrqP%0^wR3(Bl&0FG;)@yZm|8
zB3oY!XVGD+=U|sGPP2~}Cjw0BJq&0ZjDdAXzCQ)pbr75cU~Kn<W^0f?bFV~e9~~XZ
zs}p7{l<0I^s^?Gj%&BD(e-Lawc-v01=7)5Zb2S1*Dd4xzhISHr7m9w!;Nz0OubZnc
zN(4l4uSJ-jBs<P$aiWWj;!DmJ<o?upCa(exYFhq_!q)_zimg=wJj6nDK%@#TbFbHG
z=`3P7DWz%YtpX%{@_46KWbdnOX9I5~8u;NKQ2;)y17r2Rty}&*Ae=zGKL>&n8llG^
zW_Mh1@9WuCWv&$o+4}*C<IW^YPorsXPzA8Mf>X*55iRla`#l17ef(tM$EZdbQvTUz
zpy>NUYdMUG8=(HJk<+%Iee#9jO!FQg(*3!XLn<&}0ld2G>T1FH)Pg<9xHVhQp*Tf$
zqwH5*jGD6JO_rOtt2!qCJ7O#(fico5qId2J*K(S^y`u|NG^eyCUw*{*w6Ql`F8@fG
z2}n`7n`qy-lv^Vw$6mVN0~Q%ncD}j4O%fHbpVnl#pjAVm_K&;=v-pCiwr#B|yjS<-
zKC)mNoVA>L9UtL@n~$PdduPp<FHn7_kpp|nOc)z|!{gwAGAs1n1BZI3CgZy~*HOVS
z>MK6_m%{>!`2e<%JXlcmFm#NiKr}c-6DkQ2<9_BqU*Mrb&gv-Ze%>Mog+k%i1iP6E
ziW((WTzyYz!i>QQjt2f&tD{Y{K}G&pybU)%<u_T<Y7mLgHQO<uWPq8sy-f7yi}!mz
z+!18#1cUroC(jaF)x#dI?K&K&e7ocWd%FAOG|LWw%27k~N4dizRfMbi_8q-POD-%f
z-Dv~C)TiA?*&<hQ%!Z3oRJIM23sQ!a!8=%Ba9pmX^Q~)UBo3mWze-vi@00;WXgM;!
zz!v`OnCoQ58(w=ly?)uRNeWqE5ch>UjdP+U`x42L{M!}gdgJ_?m$GE>z(%6CaO<+q
zcWV^o7l@yMu8UnV0Nm=g`I+o$eWbhGF*=YW)*ls7K43OjTKT(^+%v|&0wOt{#i*n=
z6K<w{<E>huJQ<NWi3`NpA*e;>gRrSncb{umB7s?@<9^HAT=Zj)`sv4FyP?vKHa=!V
zVDm7nKkL4_?Uh3rP25Mg=g_<J@rF|Cf@C3@YPLZT7-&-SOZ9@_;S6Y+U^YNU?=HQ(
zpkFOtJUz+z<?pT6(B^{_#SiSz!E13FXO$i(f!;gSY$lAeD9*r?UdPFyTP}MQA}In>
z{cm<u?@-BK5{(15S1iMxPJObU^xBxz90wCQVn)&eE8V~a@O^+mA=4?6u*nyOArU(V
z5e}!NF`r8&+pM-0_bqmNXv1_7xR(DA4a=_yC%;7{0-@9qh(&P1^XyxMe_06>G;k~6
zLh1o@D}$jMu8(%BOqExTU@)X=k$!?}{8->2%4N@<DKb6_=uNIcy!D8HiQxOb2gj=^
zAk641id$c;U{EOsNu}kO-}sRC%&uSbuhoyTC!oMxJjQ?SR`0;`^0Y;7@X{M;AH2Ip
zb#+&#ZcB-%2#yaSt&duymMkIxFioUFiJd=AfK&v}312R&&fg>HifV$&L*(eqy;(Jg
z_+w;8`&O@=@wxP<`8fzWhDa*c#>W<sGy^*PvsUK+sC6h-iXwB`$_3R7AL?io97VJ-
zl@4hoGh?&?N=(Buox7KhLw7XdDm1IW8h?{JRdt%5YgzV4l#r+#RRi_7?ah2C=k9CD
zKh^(T0SZ2w^&Axg+-QKkK-}wVAfO=_r-5tp;4H@aMYes8)~?aBX9cX)34|y9W(>Co
zN=p3moQ`dTIih##HJ9-^0+(D1`&zI=_m{&~_&|4W+JyM3O^RZ!2vV5=058F;aFdQU
z&SE0qjG`znG<=8lh77kYzju{7(cC1)QBB4~l##>$QhQVh7+vi1T|KI#0R?rn7GbOI
zT}kG5x!Ed=R66>%td-v4y(?LpKo(@V5lutK_7d6VQ-xoifK(0AX3L=uJBJv$PC={^
z<@zFR`(xzs_g;Wks;Aco3_LT4ZL<eg6X{cf*vrl3)O912n2{9M)>dzFD<m}m)pn0o
zpJX@U;b{MSJvn+|j2frvK}*J%(x8NJIy**D4A};>J>jHIvp=RCwx2cOvfdeh2)|(D
zV;PF2j&Gf+zSe;QO^^`lqa3<ZpmwZ!r=BiCvImmhKlS--gbWFOs7-LhQTt2Bf?VBZ
zuxmG&xos%>4kwik<<!)YA)rrMDUq8DBW3S!t$`QN!Bug&U?Z`2`v92$f>;u$6t>A7
zVMgX82m0X-2L`S7E}qI-_O`@KVvW(IeSO<aSLgs<v|vN8O!wLrY~O2ITj-B^1Hv47
zl4Adz=5K=h>j5~MfaSp<7#^&V)U`=BQ;8=t7iT>7)JM3+)rD6R_AMWA>gtn$RPa<0
z*;0tEbxZPQ#-!pW>^1v{A8?;Gm%SJTHeO{?ptjFZT^zdA@+`azX`1}<^O4(Bft7{f
z8E-$F{`yQJuT_O%hS>;WG^6?=y6xD}gc^hT+2Qw^UGJQ|&5zMZp?a+)_PG4P*PL)q
zt$fM_L~NoQGv-FKL(Bn<{*^nI+UJA&SVtI?_iie&%$DCC?k!oNz5FRZywnd$e4z6Z
z;=jW<cfLuUjbE8GumJUQ^mS<EfYonDCBdWc6_|w*0`af}X}iWRd?@xmK~wOA1wE5e
znpytRN3koh5E%|#g{V~R@7LvWlA6^7u@CRjfs$UXHWfVOZu_{$14lu^AN39Qr=o(H
zaOUmKov+HzAlw<g?_ji@O8ko_*jF)s&%y|40c~F--f&Hx4Z2%9U)miMb<rlthPB;G
ze3M@4dSYRJ-`D}o+JYDDYnoB(tlVa#v&aWL|0cWKbrf~^!kb7xgD%N%o@-F`2T+gE
zb2BMd$z1w+K=U-e)GkLv2?^1RiqbKY?j&1!o#Ei68FBT>Z)&s|(Wiru+XGA41uvJs
zE*8BEnp%Ti>^08egqYfvBJZ>-m7y0$2ys12P+>WUO50xQ3O{y5AfOBUqk!0_c6TTC
zBP2K4UA*FcHf1|bn~7;`Vrcc#f*l7Gk5Tg{-5c3UdEu?T&7ftEC^NCo3~7tk4@<mf
zYdqI#E{Q#E*oAy4iwz<!Ha-?7a|T9DT8~QmwDD{AlzPz|bGx#k$2N4VE|LByHTKg~
ztJXS+pvSw;$G2Y1Mf-o=*4jzcUwW;41uEI2KZ07b-n&JS>+?SEm1}bDPI@WtG#(l>
zXiOf=%k-1)vg@Ws*unE=YiA0P>Au|0{W=OxT@gI$@m?9jc!0e1m2rTxMl86OP^itD
zp;r(W!>(o=gw1L}m;6Ct^3o*`lF+<hjg+P*K{Ryjc<7Bp)JI1WF_U3Cd;dHX;Y`q9
z7U%Hc(u7ZJBO_GaPo<7Y>zBM|nq=KKMn9{52=D;A*ly+@*yI0DmdDLu1W7kvjH?|~
zKRQHe>;u^gP+5-aCh=4v`&I86T@00g#mH2jEq2gHQI{Dr_0@dGoTtFo{&Th8x)0u-
z6MjPm)2BSwMm=WN<)`Py{#VO|c`U+(pt7T>H1i-mQPJi44e+prYT_<&S%9ngVXHXU
zw~ShFX(5La-exXl^b3PG7*kSJvPg;FR@_QLx)&_5ha4#hgO%ZA{gW5;pCHPw0!Jj?
z3<oD!6+&vhtvEOak3O2B8RF^(s!!j2@lcn5Q40sSdFO03LQbd4ibqlP1Q@!~0~0TZ
zdDOqqQurq;C|5iL{-G73PRtkN4AFOk`?bFfiBQZ^Jy8TgOh|W?xJ=eM%vmFNtTfHZ
zgkJsxRIv*!i5=Nxpm?o6M{B-cf{S%e{+P>A>78NqUtY#38unSylJ_Mxz}2L&THm!f
zKaQ<^a?d1#KO2l)#}jxy{1J|m8&_~f)vhuYg>eQj+C-|O#AYf~+^q(PCg~wy0!7DI
z6Z?^ejdEO<ZAr6&rC)peFplFXXm<s9^gr+ikppR#X0J%cRgVe_D~*{RLy*kpq)V0@
zXn;xr9ZL!AEzBlExYg`(l1`%<wnyqIiFm)ew7^=g+jS=G_}F6ceHbIu%_vGVdcr@-
z1chEBfsHyha9L&3sSL6NN<YDXZ&ndD2KjaccY+k;eNGR4pSoYm((jLZl=j2HW#vXy
z%qQ@hp!tf!2zDPKNjfp)Tsoa{3-fC}>}$C}nt=(DoD(%y&cU8TQh*EhigU?yV^fd9
zDWGub9L>ZF5})dcFAC|reOEFC?)`kJclai0J!RCGRTX0Ky6w5!Bydx7$nax95zT>B
zF=f89lH7k@7zRuZTG{6Wl_@cx_$BbQ?*X40z?EC0sunsA4g<o&*JuGwqyIw=t;8^I
z_0RQrpYTS~z;tN_sdqBVp^e4=r5|6ldi*8Fo$<|EfPQN<3RYq>>nYaCPY(RZ0vf3S
z!q4NL5;8pSGIIa)Y*5}@09HoL&dpP2=w_(9!_0i>&JIs5C!hJWw@{hTzQ+{XrPqc~
zcE{45SWmtr&(qkvwpfVe)%YcGcAiabhtQ6dfLe$+NR&TFUD~?!+d&cPlf_OGdLVic
z&)D!wqWZ=-86BAC@2b|O%fCCw1&||*14QBY>ROof7$mF&^r)ZW{NpfCb0Y+T&Wh^Y
zmY<r|M=J)9vi30SP3S=X(cVd>_*b+iyv|;eii*63RRJRAsrVm$tlr@pTiEc#LBI)z
zk#xyLZ+w0Y>Akz$AQTAEq}W+NX6#Vle~+Vg(F^@(WWTbOZ`mgn!j*S1{Ta(0(YeeQ
z=!PCqV2mp38R~oV+<~*usY@pjB)9~%pV9?*;0%Q4kIwf5!j2$zPF8(>wm|PbYY8BJ
zWO7`}6L-(xnDq>@Y!Lcm0|;{$m~V{r%ersX*Cj7k-S?zLEa5?-bc6J-WyqfmnyNC#
zRuPm5`z(uRHzN1+i1MC8C{@UIFN@DEiESe{qA&3?W;#)%OZHF<9d`XxjP6M?V8#a@
zCj;;(p0gjBEwBY4JmGIr+x;&vI?j06IQ{crW1^g=dNT0=`?EK>U1eNyILuc*F!$F}
z+mbiyR<7V93ULWq14j8OqHd)hyS5x`j3bVp2kX#glAWfN=@t}KjYX2DFrbroM5s$5
z_5Nr9RJ|nlv15reK|M@fJG1LO&^aZI3nO8ld?AbM!jb!*H{NK8;cT=D&Ii+Fo19>c
zQ3XD;iutNdF0Ag(K8uSO;h~U*lX<ByZ-IVpf6#<dH03sF)^l)7R-K!#_gG$$Jxbb~
zK)k};B?g?R6fWg+esUnM5MmHEr3v--DD_dw;0v9(nS8iCWU#Rc;{)vo&lCev_#c&W
zQ>h!$^gliRhJVHb{i}@ZT<>sO!|Hq)P3Q13h}<|snKFS>qAtCht&z*Dj)SEF5kZGW
z{%M9E@Lq6=Xlg#By)&i4rDH4*N#GOzn!$vj?k%M!ER**EmX3eWCWUyk=V>1}U9JRR
zvA0kN$2XZG<;6MhoYAt}pBW*0zp`czDT9B<?-e(A1?87Wbvl#i4_DHxfo0n%(2czx
z9f*&4$5VIK7W0k}k}MRR{^IHWnN!bBwkvYumU9$VS8TF~Bo()xe(>*E(7(XVjnfx?
ziY%6PUfprfxIXV$X2n2Ru#@+I)~rWVbHSib9FBuI4oCp1Z`aQJ7tyv{8d427nr(yF
zC|@QH$~E4}<1V;`2i^6xePH$Mlw{0G4s>3;z7?fRVtKS7=6SN6VhW6i1%A`KXllDg
zmPkn$w;b^N#D0p)9<Ptf1_4ivBRG{X*frlT2Uu!7*G1mS$<cx3v&xHSdCwuRLt}TU
zoZ9wF{5b9!k^xmeiMVds{jzadKBBOLpbK;X<PQnX&sL%TJSuB)^s7t0TI*v1`EwZD
zVkR86uG&9|JR`kv2cKeDwnEAMPN_}=rnLWe;j;3j_nLmRPP_hk;OCYp@YWdYLJLbA
z`u^^_Z`Q%v{S`ka4rQqf_<4oKenv`&rbi%l2T4ao{3_j#<JJau(0#=cp-$|u!f_D}
zRldXe$lhv!DatA<%5e$VV(gua0qvUy@_sQ=6OQvY5i)g~y2=4^|ND+t1Sfv%Y*$A4
zYGcmjkV{rjY+?N+?ZQkFPYQB%Hf=2fBc}xS+6?z)U$)AJbPIB1<bt<gPe)(Bv2?%P
zV5j@{3QP>-$G-AE5RPyhzW0c4#V{DaK<_3jReY4X;=df^-RCZ+vHDbIY1FR3+{G(A
zIO^HwuHn>)az2Vtusn))S=lF~6mv3mBWOBg>~wsZeFwy54)g@-YSoIRPL)FDlgeQ3
zRqwXpCqZ*dIgSGHesoPSKB_JyvTpUL4ieIzp*q7k$t}x5Z!zdQ{*G;13E84<%Hv{n
z%fggONIY=*!c^AoZ+?5|T>s<F8rS3tsj$^&chXr8oAeK%Q_Bd)H$=~Ik4-S?X^-V9
z*>=81=|%396t+^yGR%sDtZ7|-ZdObp=Xdi3K0q~W%UY(Ve9onG{2%WCU%H=`^V<0N
zdBFF*rdbhusNG6DxobNbYbLk`J_=rJ8mZFQ-<6vmY;)C6UinoRU7hywo7Bj-yyGa&
z4g+lgsx!%AI|#rR-}V7U^?Bx_t{>@A#}*;85y4?IM>qxlPMO<A`*TZonMMWwf{>tg
zeVt=Lx*apxx>2Y?$mfm1#Cfr6z!#7|Q60ChD|Dkv-xWdrME0k2V1BxY^hdez!FYe^
zYo$qGoU(oKzmmcUHK20*Ut5=^r@-keR`d3{#qQj6NfJD4;{G^?^61rrIb<#I8SF<H
z`NEcE9hy%_5no!SK3|JzZOiTY(NDMZJ>|n2kB7|1$Yw2EHK~r$bYfx_qN|_TQ1_lR
z&O~lzf{LYFoFC6`k`tKY@lhVl+PYTH!E5z{^a2!Grn4U0aV3j5^8O5Pq=DK@xBJam
zzCvCuytoi8BT!oWbozDz%y)gDyY$23`UvjY_JBjgLo~rRZ${H@%2<w*U0w$2xihs*
znFD*s428Ngct>Y^0zpG~qnX&j@XKZ{O{4j)JMaL5M0shFKQHjKTH0ya`FUY@-;dYP
zq`kIgHCYD;keJ-if`mVsal75uo-!U@J8-zcB-%JJNh67ln?7<AYy!eSf?~R}sp_pb
z_g-Me#^GhkOKcZbDl@uz*lSygEWODxI0M@bWL;~1quAulfPc?%W4Qra@U^<{J#iYa
z<nc3*_+)0{0bsbMyp$C`U_1CgC3ASpLE_{A<&ZB0p$xD)N&V&4$8cHr@%<$Nx@iE4
zwG&z_92bBZWGaL0OShPzBzw03>k?-H7OY5E5Jud62i5F&toEz81Y6?;EVTk&&C35?
zx&}pFPq?)Bh8}2?e}C;nj{G$XEO=8KWQGOF2ATS69<4oa=*BDX!6Is>BwLBSga3At
zDe@rSzAM~ZklXC^?06b-iRzFcJ7@4xUx-SlsJdf0>NnuWRD6B6G$xs`^LRw%sYR>n
zJ;Y)lN#m;0Ag}W$cR>E{yR&h*4>K=+CbD^K3!Atqmg_hmI;=I2v4&L``*}Br>Q*#>
zKA<4)sGn{IPaB>7wxL1uqz_ZWDT#_Je>M?uQ_}10O(pro4$kC$Owx)7BDOdTN8jV8
z#dCX1#iK1q*^8k<gp{(oh9x1%G18m_eiP;9ks(p*(L~z3?gy>e<=rIzls>OCg}t4_
z;5<wTr(0!iVMg;h%kpxo*>=Z^#l-SkmV<O^@%sDSSC72x1>@(`%v(d&#4dUc-lhEt
zhnSO1553+ez@?ZWNA-GNpTVY7^vpQjLvL#iw!;|1S>nO?ma>s%7u#3Uj-FNwG8c;o
zThaQS3odQe0CmNIjRTMQhd>Nq{P3_Q4=%u?Gs-_$dP1vJe+DXYmhsl+oKb`-sz-et
z>OMR+(y=8Y<|IWO2zi$lv+VEzTGCXRnv@~V9I9z=;OxAb+TJ%S{0*mi2ho$}5eMR%
zuduf`{lz6lUi0MwAuC$VAkSdAk2VZZf%c>B%Af?RIvwM%qnwQWAU!F}<3|2L)bl$k
z%|UQbmoyWem8HCzcszy#_q^~HvPP(jTq8Ri);Rvd(RZTr`clk&3zT%9PaL8gllhEc
z{$dIvdCaT#aBRcixw!Sy%H5Cf<w=eosyf)4`jI!c27fmW-SOf2{N_ARDh^Mi83H-m
z1Ek+woq(mo_Zb$EFYT<&+_bIvRm@YAtMEDqG`~>~8Mhd?!UK+0mw-v_s+*Ti-Y2^8
zWz$}%0TR8h@9BvrVM_$^%3}IH=T2f_EHDvEC@Sy@BkVbW?the}+f?Z6dx*(90_@jd
z)4o2s-V-gj`XQG3R661B%9o$6|L^g#U^e&ClSGySF@Z%-5GN(YjoLiVCz-1s2tv@u
zR7Akl*}!9W1=R;W6Z+#rbXfm}?pIexJjvO6R`qkINLJQm?40C`P`staV{C6m5a70D
zkSfHeik2C!#Qsm{$Ips3>$IdD3zhE&$D*MwB`5=U`w@Y;ydNY)*q9U4)HUzOo&7d+
z(D=jP^)(i}Z*FfB_J8D67j-s=L?wPW&Yyz%AjQ&{b#d^{cQUAIQ@EIEF^lEvENjy^
zC}|tsV-5DBJPPP#b~<0EUhvXWFfXj|d=z(z^S)^2ae#u0hoXKEttEI2&m=w+enSEE
z-8d~~q&B}Rv3i~+e^|NzTlZ1XyhNe&G^l>35Wba*kn#OdgRFf+W^*5t%pbx9_-WJ%
z;Y+WaE&p{1f-6tC09iL&lYK+L%xAfS)3M5cOiC=rqUApf=7K9^yJzxh4tD$3Qnha4
z-mTr-sObrIs4~;Z;Xdv1xA7B$R-oL8U%C5a+F_!1XpM~6Kp*Ule$|I_(=Vp|6z=Xi
zt@GzUroc@6D``J{@UOd7QK52d0^njgsN2QBx;0lQ^y;yb3>K1XQx0@h!(*I4RQS|e
zgi?7LFz(uJOdHDa`r$bLVMWx^@9=mFjSp<bO~T#DNr9H)UX;(l^LMRc|MRp6YNy$;
z_>x&&RzET}J*b+WZh2CoPt1_2^<(svFFVl1t02)`g86m53?Ip}pNe?T#$w(JYVSrB
zDVaZgb^qeWU;FNl$EZ=@cN#!YK)EoY#~kNT{u1O=BmZ(P&@R{wn_E~^m(dpu(FDXh
zyN&gQ3GevKqvr%Xk&^-4wU|(1yFcQ=;?(NON|-HhP2u~w#XDDJGx+2fxXxp9QqYip
zv2|_nypv?A{>@imeEBh@-cgML$Jjqir<0L;-q>$cxBC!W?U0)YN(0;ksBq|qf4GVb
zall(>EB*StPSr)GwYZLegbRa&|Ir;g!;lmtO(#0e`|gLIgKP}%zd>@&$@|}dg7H=4
zS>`leA?(Pq3jIgys_OdmJmHOs(&kQEP8C%i3iO<$*gqel0-~R=a=^LW@$6XO!>@r<
zmJF0%Ukr`*kA7V~88=#3yZHN$6BG}SZV4@8IXU*3Z2;1}Dd~s~5MaXie+$<@!e7#n
zQT{BMzbkA!4xC@IrMRW1JomIzq?DxfIk5*xO9E(t08hOB^Q@)L9pE`W+&Af{cm$uf
zXgY=k%2b;cKx-?}e1gHlhJkW<K9VDW&u%s%H(jmK`A@L5tB=w-ZL|?pkINhfO3Jd-
zO>ep4<{>@B5o>$Gjvi|Tm~!4iKtP#0&0?Qp=_%W<biJSXv3+TndvQ*{gvgfoZr(+I
z5Tfbmry&Fldqo8y@kSHpfSM3;uYdound?8duI3#-mWj9h92nV;Pmx^F9&L-!=cVq{
z7{5xyP5~u|_j6STM2R@&NR~qPPY-9qyPEntL&yKME@DOq8rVVsB}Dy<Jb+mfR2V5I
zIv<BV#51;OeT;MuasKnNLy)-iZGk1L8hmE%RR(Gkbl^qK#$_`C#K*=4R0eqEzrfoP
z5SjDkq?w03##XA67gjW*2BxgU!X6*a#GDIuz0}x@>2foK#8MHfgO+6($CD-ShC1!`
zc5bz?i-@%6FPdaHKN~D0@qkNb&yDdxS6kxX!pPVMewUZtvyy%usB;ZV1)G_@C;jjE
z>Vt|pj-a=quF(fKa-XC=P9SE;c~n^@K8VdIv#m~nDkZcUKnqmg;c{|ONn^}X#H7~e
z$D0Lr+hWQ)&~2vwLJA-<@0tZn1N8^Bm|uEeH@yx<2L#roD!VNVLk+PXe8R?$jK5uL
zurK<~oS74dKkxlcu=did>YRc~wtx2$c&}o(ab$Cbly-E)m9#uyuq`d;Ki?W%`+A{t
zwn$1j{-Oe-4)&+R;0SQeJy=t1AE}x?EbE)E@!ix`7GsV@{5&!^fLBd0vv$kwgL5vZ
zH0#?8lphi{I--jTUXH7h-S(`|!vs8VL$oi*bGUyzd6G~neQQhN9x`*^zU#E}v5A6b
zrQ_zeT6X&w?WxRajDuQ!1d4D?e=+vpc3drQkI&>`^i&RGNx27!iXEkbvcDJ^b9Ten
zkIfQn6(#>3XLqzD{}<QiA)jP8t@{=UHLpl#_Y_!Me<6yUfSfF9&v5Si6;_aqUnK=}
zgJ(_JRq_Wh9{=;K9L7RP5IE^sk;#?=GOx$o`RPHSM=5tiw#054Gd6Pvs?KR|!wr<g
zCC8W_h85F6HNEOzYo9BC`$Uee7D*i{0#v>XH!nrRggw=3TNoBIT8Q}ASr9l-tAioj
z=v83r_^oRu6zLx=FQlM;PRY`_Rj4*@T-H2VWCrWjqw>n~4c}nBV`qT9P)d~%beP^~
zKJzHQ$hB2rq4%8gaFUuR*BZBt&R|m~vU{q)$#1F)*4h|#{@yr-ilBvs3!OYz0g0J<
zh4H#mIDlVPNWoupn2otCr4W^cEr9}&D-mEAL*LLO4|h^kZrc?MdF(B{LAs-k2?s$f
z)f2%46cp|6Gu3WLQ|Vl)a$3A%mEq9cA48vEq4S*PaBv2w-IZ?=PVTORVY>j%e11DA
zOUWF0a~^cs>3wiS#PX}&ey9D;rhZ)pde}Xn#K>vm4OPBihc#C4vcl6mef9P|s$mBQ
z;EIYDGc~I>kG|s=Is*+rO&uy!A+jv!f3gOM2M{JrauYyG=O$3=n7X)fh>P$@Bh={a
zsyh3YhnzFL3WrRFonTV-(TxP>cZdaBRBm>!{qTVjQcF0Gw-7IdY}GXX8g0?EfF16<
zLjbCeyU(x7UA*%Rf?&7@ykqx;&HP$y+AKnW`IveWght)C!?l@JwMFESg}}=_b)zF!
znYi%Qrg;~MU)!d7I07vW1RGFYmiz*$OFg?2GuIW(*OtAB+%!jYl5EwmT=3sBAN$V#
zz#fp?Uz<}t;7?J2VQRF~BuyppjK#QG27ea(!}=xbq71qFL{4ax;+^si5N*#B8{|<s
zjzr8#C<m*Ix8FB$g+k<@Sh_M}hG_Nz+X4gES;cp!eq${vFc;MuJtxJGKZWed4de6G
z>>7&bib+1UdO**5x6QKq-t%jrH}rww{+^p0T9e;}J;(eYi3ftM>^DovFQlq)W9=Q|
zod+?Q*E?x^<+=#R2g6PO=Sg_it%_$KTl8~vMFBC^_BQjPL)A61gwen;ChUkTxV4tI
z%3aYp=R{t*mG!nhGo9ctqMSoU9xu9Cu%f$60M1HY%B9>A<xXStd5R6|4;UK$A6r))
z)`a)<bq*;NCLkz1RJueRB}}DTLXb8PK?Ugsrvejc=~hY_MY^U^qJ*T<lG4)6-a9sR
z{J!shK9A$>oO4fp&N+8<gzgIx0?lr*_24|4bdQfWcdv%|Z&PcX0t^nQR-dj~dHuuR
zTl@U{Y(@u2FW952n0``#3Bx_5J_uQj8r!^k26xbj#ae4c@rw~-Tto|Vz{G3x*j}m@
zV991YOM@{2cie`ZC${sHSF2`Se9Z~gn{*hG+9q7IoO%DgP;|&kS`gf*l!nziRog=T
zKJBv?GEtV4lFPsPwE}0vZBu&eqSyL>^M`Lv{&rcKI$sW+!fxl#M}5bO7OX7KzUXRq
za<!$XPoB*AvRnREK*vj6vZr-F<_}=?w9O_7X4__|ptD&H-yhyk@b-{yiEfrK2n<=C
zMb~j&#M=NlP5$e#%~Lej#TOkY@y(bJ!7KYiIe0zeFEj|fPE~Y{GQ%`)eB<fH>j=RH
z2RWG$8vIBHYLq->o7&*q@1;??AghrN1Fzzu9b}3R;@?>Q<GN8$O8<zv<|nG$jh>J^
z$pRC{CA=q1*nnynw9PDBYSQ{+N_Gd=1g1}}+c()!gCTQJ)yEpTj+B@TUH+fAX$qB7
zmYR4>5>S2YEcNA@D?CqFA<}uNipe}OLgFF#+fc}VjQwhR$oDJ*n{6lM#;}~C2%J6+
zvrk_SK$H|oo0`&eFT(<Hh1A?FFo0>d3;7%oxJAN3fPulltrCY|K`QS&6~__o1a;G<
zMK|Lci?m*$fkC);hVcatE_hNK;r;Bl0M_bJaG}oW13zplQ<6Ct;u{E;X~1U2*&ABL
z8yY|zunI8nmu9P-_fo*SGQ{UR1__*GYVn1{*j=hSgr|4>d7nEq#bySqFq&mDc2i2?
z5w`=7pJY)cVCa_9mj->h<qX)dFuv7x13z={k1`o8jM@Bs@Zf?-%;ZRPjJI~pTqk{D
zg=+3J{6uUC;j4dFL)LEOdN~Pwq{n4tk@k<)6&3eCw!;QmM)*D7bK*a7qj(bnb>TG<
z`mT5MMRR>!{>;*Rq~QuTzO9^Bb^#<nu4PDe6MLCIx<Jsg*Ste*PsSgVA<B*pM}l^E
z&0L3sffxJ6W!MYVKU1S1*|k79tSwR19Q1-yD8I<N9^VCJ3c0sWRKWEg7vlf?KDDK{
z>-P4}6mE(-VI_rgV|Z;YyKPxoZIkU7A?HHrQ@D5A?#ySJ9k?r94XBC}%wXmGAJzIr
zQ8M!mylY_b?IHEvCEBNhPNCb_Y5<w+SsubQNg9|w9osZHpi%<*7{lJ@Jm7uLPVaYM
z)ZdF9@3@`q6|(MaS&+f+dX*iqt0=f2=wfg(aQF`-`X&|0v*K!CJEm@V67{2nAB@3N
zpPAJ2|6h6W4GjWdBFNA7X^9!AM_w_mG)I^yi3?o`!9DuBClUjiNAY0NP8mb4mDFlb
zp`2FvQAM8TfB0J<UpGD5TY_Cr7|XB(SD5Kzd-AUHL`8hcrtg<WwDJU{LFSOC3>M7o
zI(3O3UP|Q8FS<Y55pqKWaeH5p1XSJy1<E8VfA<#*py*w}{wQB^E^o;+-kugyq~I?=
z)<zVlQ=a%Xa5ufskUIR@7rv(v2{eaE!>%b6yiACHq99{Y(0(^PLXinc^6qNZ<D`b9
z95#w876t0E7|5(xI&ErXhK5q?9qDSu$~7lcTGPqZ-#eab64*t_`yO3eKPrq>yG|-C
zMwPnM(ZJ6R=!TZuI?jfCd;p9^@;$n3=`eRN!x^GmNUt_4%e0+q??pz9|6eHXA4z?X
z!CZu;6Ym;Xr;;_hPUzZKgLD&SvC4O+ZE8>NodP|^8Qhj(u{Ed8JQs!aD1r2^ZVSq%
zNHbgM_U}8(^Z=xE;En2$q})(m+WcUB_P+eQvGK3I@ZkxceqCwE-WQshLd?^yPZS=?
z%KhyJEt8;c4x%-<ZvW#9{=ac#gU@k=Vyaw|9c4lw0d6NKGeX@Vwszw~^XGnhC4^Q)
zM(v$6ln)UTw~#j<x!EEnq>Zw2`)SQFym0kl=E=Py{xEsu%24@=+-$>T+Us^^kQ17`
zc8pUw?yGf2g+ua~EVY{OETAq^8pi6!kC67u?2M7y_fk^Agi9+7)M)-3gA*F~?TYPV
z8BZ?xd24@Q8P%|{wcA9%+gKymGSPb!J;vBVyK>9DT$>WTEW~u<nBy?D^b;4P9s7rO
zKR9Sa<9Q>T=)?8939#70zS=$|1n4@mjvgKV`oXNjI*j-{O;5R=stKbnrNc8r&@y_@
z$@<q$Cy@Ei|I7jm6+E3zI=7$WYeQLpW2vh%>EYhm8`6hLOBb|Y8)1vM6c8`EEvRi=
zq@%sD8d#jND?gDKZ0Bhr+@^4uAbvcFS+wgdfWm5>u1A$YkgBVhK~Yb+!i(cQW%?>s
zBQAGbe~D*^p%mk*nPu10phfoE1Zc;{7aYSYqWe@M2?7zz&9lyBHb5!zh&%!E15ZGT
z?6m$$4WgG!b0$cy5n2_mA77p`lWDMX;#U#902<x2lbgk|k2`-E(^o*t0#XXne1Vqn
ztdnxIs8tJTy59b9+fN?PK(;npju`txuign=*Z#^~@~?&T1JN7_1J!;!Fyafvyn))*
zOd!r%a{6ahR~6IYj!TCl%y-$~#u~q)h6x|$rBj%U1_q#feeblHhk`Szu$5gM?dSV$
zRE|qX*Scey+4>pP@25HZDHO{LzJ8(R=DT8=+uB6L>9Hlo>bn`g9iA3gRZZ$OUbm9z
z%djS)gX7JOppe%n+$#p&Y!t6ttHMw%7X|aQ-e$68?~L1Ul<7+y60YUIfBL$>Z$w6?
z%VJ|*`41}87r~NUsp+$;gRMdOmc(}c7zu+ermoa=+<p-(<AXf`IQ`AZo>;l?EUE!(
zJcNEtU^VN60%b;Fuiy=%!jUNoLzkZ)uby>Xc;2aF6T6RBKX@<)w#lb1Oz;4I<5N8a
zx?g9G!OmF6d%BJ;H{VJR3&ej~uMh(O-IS%>?X0ZaWQ-rlxMOo-f>xPxEr}t)yAasD
zLOzOq0dB&amZu}icY#@xWcR@6__DQZ${_zEnQw$od6KErB)7T5Q>`V3o#LhWRWxC8
zB1FLO$HU_N>h>7ob=|tdGxc$4-NjIEZB(KZ0#S7dZp}x0)Tpy!&o`H=(n=X`&!I|J
zhh4-LyTbb&U+iO)f-`s|V5O;$o?f7vf%F{^+sMTDS2~!DEsDcW4zF&@vTT)7g^t*}
zCQ97z;Hb*-69b{P>y4#!lch357KOS(F_Vq3G@yunpmOzw0^R!0rLF*kY_mL_=<9oU
z6F_=F1ypmIZ*OTSg@=;?H#GkO&6z!#D}n!S0%aOVR9JB*LUl*&;LCuArzta@NO=s1
zdl-(|t6m;?E=wRvcWAR?im7)nG?QN*XjuguzSX=umL#T3OcI=Obo<@rc*YAqCY!!@
zfT-o#r61;TCunF`x$a56{UdhDKHGG+R7&hJ<BmYqKpS+mh(1>xCjIVxyV^d3LVhE$
zZEWe!x0uS2Z4p!3;7g&{>~V1%uMHVqAqn6-fw$yVYg8OdjN{C%O!b#cCubCs7?3ZO
z%ropoDalA0Vph4lk<A<2j2G2+9JoIQDk^wQ3jDe{F>Wsbf5<~j-stko``c^1TSnI_
zKT+DZV7I@$9Y<p&w-q_}>jfXAGR$jyR8<SB9Z@k7+<ql9DK9RTagaGAQNPvs7J-+Y
z6Adl~J#Xs;4Q7j{k53sn{2bmi7Icby)B)QcrJiF@#!?Gm{94%Sp(VFrxZ_&lVnys=
zHkw2iVZfGJoHkpY*gOAjirIK5lXBdv3K1gZy#7*czlq*}9-Db_Tiv27S5SGtron<;
zr&YJPDcV}L#)$HiIv@VKlLjH&SS=6}%s2PAo#_|D&V+@U*t<FN?BiDB6v{REF7{e}
z_aYmKTkunxMX~fjgzNokP0T$wfzzinB~pdr^~Z868jCqw`^CKOrYvtg+b=V55S(h=
zlqoZYD-}n7OlonYs-F;wQNUjl_>t_uiuMqT-b7EQVE&G(9L#_bWuA(dag)zJGveXM
z*R$@h43AEO<GjUh_NA`Lba}OTKGX9d1SPvnF?xYYO=4iR^m38LRuL|ckigzXAKmBu
zwdIBuU6WUX8&$MIE(7ga4|Z|v2@2pIDo5i@?XW=b$ou!nORrFC--fn8f-!;R(%f|5
zBi1T54La0v_e$NUR^~ZXopIm`hm^;EL^FpB&p_m82u__L1=(Dk;5f7Kl0c5p3wrtH
z^aA#MUb+$d&SvGcC2wC6H#J~~ml{5Zx=erVh1VBQFsPo!-<WAIM9f-QVs*Qj&ESoP
z0dO@RNo{K^9_$zMImkp8W1+70-59?kx7c+szF(W7l{tpLk@%lIju(ND9!gp%r+JCA
z_sR+@Er^d&EHJ7wSBcE|Af%JJfjhqRX8Y*<EMZlLG#qON<Kth=;lm&H9aOED)BF;5
zP_IDB32B;;o#M!#!^D#20S-R#?JHpi8`-*yuM8(NN!%Oz;x_bU`m!E>lOcqv!#u&c
z3OW7KVy!3q2IWkx-F|M;E^yr2&)6vlcUy87;;gh`fJLT=H;DZM=c<><o&1vLHu|5`
z9{0s73PT`fC8JZ!Q8dSXqZ)Vl#T>%|GXHy&nELj2pQbhJnx%?XJU38))reBzODCt>
zu6(%loFIxL@=1~U@fSW>q<&fBOF4IRpUu7r#8j1MSANM56djl7jDB<KB~Zs1W}uEQ
zInsG4PAD@GE=><YvlXN^VPyL$zh{!>+e1M;g$*`hdrcCDV6z6^^cRbdXqz7gmtSd?
z5WlgZu<(J1sR&F>&GX5MVr(xs<u?16$Ug1yBR9MTG6v6ah*Tynr#xfr(RDJ!&5ekR
zdyvVHjKVFBtM}7G=7Sec5~nZ7Lg;Shz?(}s`W!m9PvI!IZDr+tO~&&okua3m#>`V@
z-3BgXJ8rHY<4z>LWHN0<wz2J=mDu#=>^2ph=-Mlprh?J2?SerHdE9z@HU!JONuIJE
z@V;`X{*z^Tkat+ie#<@+JUp|R)<Y5=lL7yP6DGTzT^b#%faJxQV+6(JL(vvkfIYpN
zh3)lM8qN1<tj_TUu{6Yk+{$Dn5gjv->_4fzW7WKUS*9Y%^lc5n+I-FiH>>9K=P~F@
z1Z@9@2?e&)fluAm;&Zp@PGy`+#?ipNH1l;%L0BtaZ<ugrpN=;y_yETciX*~FVdKcY
zyHtW~Y0gJ!A(+?3c$&1sZ{Rd5)WvbLGNkP#<v*oxhFB}-G~4-ftEF4Hv1<`u2AEhT
z>tMX#7W70a)tCGGCy>mi!WMoJ=|T*0GpnO#=l!d~_J*LDrTtgw0L{%rSWah&s1=?C
ziv-`@fYnyZ)L{AkUBgdu!x;Ve`eS+VYE@VNM~cAi#Re*G+RaXOMy0uG;Y6ScM4%Hg
z>_y8LAxd!8zoENl3dN}yjV0pCX1#Y{Z+)Qq)FI5+)V%UqujVE<k2vTCe_*+YFqdM5
zQ4tYwlYDf&`+c%q8_p_7z3vP;!G{%2?Dwu#D5iziM+16T!`%F)(ux5$z2oD{4zS7Z
zMAAsW&}(KbC4ah~u2o$5I{;lmO1{rfxr#9|f{Bi85gAnBM8|y9g%4K|sg@*aS}SYh
zg_h5R$8U+0?X_;|%0l8q8iRozTV~Oa-YU6&_`1nY9dbQ|9KSqlq@7)z?yuEE9ofAv
z)qtaxb2@rF){c`1cyxeLvJ9OM<=as{MWm`IVR}&w>yEC#(#x;vaZD|Zj;jdho&IAb
zKPmd+^y8C#dTlGp9eOEE&}dTs;wO{jM0%#-ifZX#EKKCup`~L1O*qq_`|LXTq%gb-
zJYv9BUj;;4liwR?S8e6CAOhR?F!I3EkytKaCTu&hmBKEr8xwElpujD=8Q|g$8h^#$
zSNJMBrm|0Z_AbnRe<*0racgItA}Ws9VCkq8Tns$gJLa`bLmLAD`ktR&2$*?_#(PgW
zwqSi`Y`{gCH6VqTFq<iA14}|iNnm8>gu=%1W?Dw|FQ*$To18<e5}W2fB9%^Pdfe20
zw)ah8`rw`Pl9uA)%5`<_#ZSJyItXGGsDGtnWTtj%6k`LoRHW}`cA6Zd%-D6!EsOE)
z$Rwx38&xVjc#n6E#<6!PQZ7ZFsyNGPpN_T4qXsq^IlEIIW^~ubD6QN8(rih};x}h6
z)W5v8Cv)Tm315oOmQQ~?<$Mrr5aa*C6-hx5-A={})r=)UNl&@V*e{TE>U6vh)+b5G
z+WV>`I7oaxc{BhyT!Jlx{dDXMAeJL^lPf=c;7!x=+XT%+)PUFnQ<h5l3ORBuMtGGv
z3b^`Nb-TFISy5&;_5b<v-+rLFwql*;zHk71^k7nMT8ejv<#OMdy%^DnJ9AFL(Mez?
zkhUt+s$`~&IMq(FhH$5I<_t=1R8`Sw-(v%`9=d7da*Z-0yth<wPi-^dU@*TgGOoAu
zH>X%_-qHWaP!LtPMYY}Vyr3Q{Up<u}!T`wlP?B4c>D}QsNvO}lVK_3h9D6D0WopL%
z%LPYpwWhVo&Lc0&jB^U$?=Mf~CKY*iP_!E2T{j&K?(d&EIyEQKaBtig`DE6;PmDXq
z*Ly)iSK~>iW&Q{-@wC9`ni*vV-CIKG#B&D6#6(>01?+ZQd#ulYbm=9g6o4Ic9+|0M
z4}c3lj4lst11QfmJG@b8E8}!={zrf5`iqc7lOz8T0zE3uGo#97QE*FBIRr(ar*JgP
zX`_}uCBk|fpAfx?0WH=?Cd-XRFM?tkHs1YL2JFF_o<p0eS|&5{{1*=M0)q7h&2)={
z^)E(K;#J9ZPnFt0k*O%Q*!QS>icmQcOv;101R!V3oxb8qM=k?;$r`>`++;6g;y1@@
z@`XK3&eeT?BJf6mK-Kxa+<m(67&ArEHrUPhQo49koae|@Z~{Z)mtWw7BYg7s6J+;J
z$nSnoKHiKN@MvoKG@Hq|&-Z8Rv(?|9ioIW5I3-STL>8RE^;OyW7JH#RFDL#C9o!k%
zcevCHKAih$KAU6z0fubYE0w}41L1>Cc2PKnJB+UgQf8cC6T7mfk30fEGTWqcMH6`e
z3R?gDdu>w_>sb%=e%r5D{Yj5CADkc!=RY|X1dH>99Ks875qGZFFtAB2m`&qviD?BQ
z4}B2AOIwl#5+YM?CnH7_xHhuZ`8TP!KqiAv>x3XUqfDOfFZ}%>Y-GZB@P_<8&!iWE
z3{>;5dpP1ZS9bn5_-si2pr;nGgN5P<ke@H27l#R7+HMvnK&^P^glU8>-T)xWeBMm1
zK{C#Oy*HVc-|us;fGPLoN(p-7NX?2n>+ufzVGufUl@#C3mYD5Gx_$?X)8NZEZdUK4
z;ru)`KG0yupx_#U2Kp~xGaD>=|0@F68-(R5BQJ{aCo|Q>E8ABK0d_>>-uJ#A1Q!?B
zM$muxM`5Hv6(XNKl)%dY(6v+SRX}>bYw@;Bg7XEtSM8mer7vo3mj86>3fbooBlv?y
zE&!}#P=ECjy&MntCe7X6i)#SO`5)pvI|IE*SG(%|p9{!ECZOxBMYC2>V*_@c?|}f4
zSE^N={(A-04|>h+d;OZDs<;5(UR(IRu}_i!iFLg{Z$IG{w;q~YT^Xm`9B7O&DUrt>
z9*o2VUMe^~H)X(U3CHVRV?8wb^EdnM@ck@tJ#wkM=GJh$Mf^G&&<Dt+x({6cdmD0(
zcLymmf@Er10@KQK@v1;LC7%|>^!EPM<SxC&8|Hq%m+54$fuiiY+n;dD4x~h}@{}^8
zRywGm<)o^U(VnsWA;Od!!d=_mZJk-^#%&(Y9Raekw*#L51M}<z2E2iqb9HcxI|@{Y
zr#p|`E1Oo@XA;UVJ;uoa&WmTOi^2+;X@NV;)SCxe9QkY1cn|4+1=u?66wGec+oTC*
z&GH`-$RTzsayeDc>;h>u_L7!iCy-uVCgkQEI5j`$Dl^mX(x0p~Wh5RmSGM0MZj#s<
zY781mY~@cdai!q$$^=}Y@t-Sj+byiGk9jWmA0a*xMLjo`noZ%?VioTb!@3CrM?|qI
z@ZW1-fki$ne=tkczk@24dxn)`X`=Z>EDe;z@shDOg~fd%&qUNqNVLRAw#~{-=kF3F
zVg5)U4d^dyx>@lhuZJ=*r;H%+`)*`N1HwK0begfS{u_55j+uex|1yKcdUt%3-k!IP
za2@@1{FlnU8}X%8uyORznuUmu;SlHVy#RLmA8~GCElQobKN@HTnw<Q92%w?K@NB#B
z($2mS_(+}n;)4IRaL6*Y{avrIqy&7P;Xi**4MBh;H2E`_E5Rv}IT80qOon1JI!h;h
z_f9!$n9h4s1}RrMOYg6sxxz4sAa9axdaa@UZd>St<`)FxucMS1s6`&NJw6-R9$mF`
zaPRC!8&COleH6amRU0YSVQjF^(~5>68`HZ1hDj1<ONlF2c0h-R5oQxaR3gM6cQq|%
zOR=pz7WRy|22U1j;p+dO3)B!&D?j^UU&L8!0hv?m*ZEA@3vZQuFa+pHo=lAgsg{ur
z{(F<45EF%*=AZ~Rb2E<SLVAHOu%OJD$A<|x`d+Dv%z)zKsBUJ%M;1+-cv$`W;%@kp
zna~;KJN?PiRoUvR)0pF5`gEx$_mX{`jjQHLU)DUo&+s)4pyCjn2z4X9$EV@h;3&)R
z$2(OCvF*C@r%mYkt@T#$-{=CB3ek`Ejo`Fe33YVh-*p|`mhX)HDnoo2S0Yl7S#)<)
zwcf;_YV<fZ38!{aG^Ju9^sp+PGSm4#&g?bD4D<rNQg2zT9D?*y(zBkb+_n)SK8j7q
z6183GdVA_avf3*zFAz#LN4^*Hl#1E?W|el=u^*Hnv?0#WG;HLx2VPR=zA>d#JDY%I
zb$WiG*t1;o>dAe`-Wcuo8`d2~VIpN6(p)%YY>@%gyNAqr=C3D}F9=o1Jwm6hb<@IL
z0aZEcou&1RHF;)<xZi|=QKT44+U`X3$jdQYo|Xcc*b1zGHPzN`q?}_>YwMaaK@F3_
zAl3i|qg0NrxbkYEhjRWaOg*nvL5>)hLzY5q-92{3Kf9$Xlp20Wt-J($_Y;qcK*w@m
z3o}GImVgQDm^$wZ3i_+$Ah21#2s4%?AX0?tUhXuZ0}%A4yy$PHKb|9!7sRxyUUW*v
z37;NikSy|Q-ft;QWqAepSuR2EC9Jvog2Mk6lvwY>G}$Je*FX5~j{Tx^()ia^z$M+x
z(~#okI7YE%BFN<xc?x}c*Bs(lPQ08iZOP&qy?T^rFqWBWU&;lXRnRo!!ObB9eTAa~
z&NrBC?e1QL>whU56~cc`gBf*RIKHTQhq?3Qn~Im9p)eSTE)R2pd?xm38kz^CbMsQD
zoa*0#uGMOGsn`x9Y?6uy_t1lgsVPf{v!1#x;bCQx)SH0aLN7DO?bob3s1COW9_==f
z$4avt4}libDWTuCUnX)ysu{$#tDcN~6jR^x51H`bMQ)3&1m3K3YS9x>pgft18xA*1
zxV*p6RRg^7_P^x6kni20nd)ON0J=-sqoeeNM?V`d?#Wa`6y;GG(E~cfk@t@rnyq~F
z(<Xy{p7qnW0Ah!W`bNHP<B?pz`p6tqS%(CR-L0DA7+~14KDTB<dzTKj-i(-wGu|pL
z*1i*?8MLtQBDZQQWNUtT1s%n;!uFB44GbO&2?8k=Rxh`qO#qiCb1Pkyd|50x@k_4&
zEA9JTlb7IdL?I^XIn7jboPno7&TJ?XC-dY_Sl^R_5B!l)<+WHQB~DczTxx8;_z3~r
zma&X;>?bWTylrHj{r#h3YTLQfhi6F=oWjU4?EN1}$+F4zR8<MPsNGm>ER{T{95m)~
zbj;qW*<B1ihX}hHL#Q%jkC*TjiU)CPCfT&SbDG#L&{DEyb|PIOX3mrWLh8CvqgAA`
z(<|gUDeR8Rmt$FrZYRW-p3{6FI&+zn018@?r#QRy47fNJ`tYosw<{lK4Yu37Qkou|
zK+I$+byOC@az*nrovkd$<#=x~&ZJg2j9bI&X-Ioz9SD+o1QBaimk=-T{&Ha)1&}F5
z_^{ONW>a@!%Q>RWmfB%hZDCuFjOCKj#O3CTtYt0$a!poPm?i?Z6;N>|5sovZ_dq|w
zxC&W(2}=)~q>uD!-yg>GULykwX$m)^*Q%|aPcZYo0#5ZHD?jKZ^Jw;5Cu)(<Dr}qt
z+Uj)lop;X{`l$PY4;o>!V}tfjiTQ*^oDCMtLZof0jiufUUkw9iB?`1yxf`c!*YBr|
zH5Ab1j;C$yW38{8pqKjwqJKY>m`Xlsl7UFuePCf8&y$6t%tXNY(#ZP;(3T_an5fLm
zJ&j!cGYcRs9`E$|!=Ub{ItnV_)jl@e)o;Iy;23Uz)E@UHrd9`5GKourws2`sD$k+l
zkydayB-?jJf0$$SQVu2Z|ND{;<n9uM;!cc)F5$Wyt^K9He=01XJ3Agl5+4Y);tjT~
zaM>B}S^BPR0^;Ul>$u{b#`uwO11kaIvkb%a<5!;Y<}F&{S~JD6N_qOKNAV8y&v`>d
zkkUvm$4?O%LGLoE0PmLvMG@vzw6J>G^9g!*mxCN6X){UfnoD1)%zokl`|Xyp`SA@I
zv2XN5vbA(GYSrI7NriL$RdygM-Im!_A?nK+vCS*xTnea))X5iwh}p*67xcUYpZga1
z#E%HFLq5h_M(Kil$?lrJ?;bH?U_O8ko?x)AU$_topk*J=NS!<{(7U+82~XS7y|!<r
z#lH|a5b0W<P&Qd27*zO?;V784X{~QAU^F?&U0s&6P(wg1H~CwqFA2nB1F2Tavt4&+
ze$B;sS*6(je&MrC9N`Y2Iw>ovL9Kgp#Y~&)1&%Tkp2SNcG&FC^i4O~2TmZ^@AJ-Y(
zB{#JB7CB#j7h9FLocCIZ*n2aWpqM64GM;)lf64^45Zebjnbrx#R@<K&HWr`M(GYs#
zztWLPgNnUQgQ<yUI#2ln+wwA#=@E*W$|TPCO&>saZ|?-WtWSHGfw)M+cphZyo9qn*
z)sKo5oJ!vkQOEnp%+f(^(ps_eFuPE<v815<fH=Cxa^GQwU_@j7;k^!HF@dF+rD|O2
zzpdENG`WG+=Ka^ruOdt72W^rY9^9!7NRn{@J!hE&G!@nR=flKYz9~W)mt!%)oXsp)
z)=rjo`@NK92$!-{NW)ol|II;oz#8&YaxPVOM{4>V(H<Ipu@u8vZnY2#`zg<NOk3ZE
z$TMZ~1`q05t6a(`TZ*1p1sN5^_K$`Fk#czL4U#~vQ+TCpr9y=zh7`>&yfEClDfa9X
zD7Ku&(dksL!uMh4ZxWD!>0OBK0j<gpD)yp4b3c&>{Pt=wF!#J<yZw!f<DM$MlsIxV
zDUv0udC+4;NPqIJE2`9Jsij4%(qbiCo7j_JzLRg)1DrHSwO7-6<}4zAkX>iA1)1Q{
z^$ETrb$!(T9RX9!I#I$6tlY?fj>P9|KENy=___DDp0eCGzu-qUN?TodQ)||V_gtWG
z1ox0cN(A}EuOl0^>4O*MidGFi&1BHy`4zf@Vqygu0~M>S?(B8$)`x=I1~S|0JYP!i
zH3A`s*c>mp&>uQqyc?I17u4Q*cgG9ohp^{C^JD@(e3>MOn=t&7Z;L{tUZ3#ht)3g_
zg09ZRS*M2GmH%d-PIM7uVb|E?E6lDYuiiKNDqx3CZbkEP$*qk{oK*f{3PPkyrfb-c
zwWv>!W_U=}AIbY@%$t%b`d2Mygj<UX@p=9^n3@>U?iTUYA%y9=NE(hdul2q@@B<aw
zF;D7ggYYg(F9RdvsV8pTQk}hA=9UbmQs~DyUSBMOiW5gx@Px?5JCJw_n<?^6m;yC9
zJknE~xiJ%<?*Dun;lcZcnvk;bp*YqCxgVB~cHQz0y~_>pd@!gP7DsIRz$0x_0+`9|
zhJP_I{11cDi5{Cfw{FcwQnF}C;yEytg-rT^ZqrNNvMKPy^|(M3_OoP&0ILi2LA>k<
z-}V4?AM|Efqu^t=>cgyXVc_YBxvM*UaiwSBfzwyH&k%0v7WwuyONRuLY=t&Hc}&o(
zL107yk!krSBLkP)js|@04j7rayXy%w_qaP8q}!qB_0;1sbOfMCF9PZ5X*s)*Y-VIr
zq5-j+tzEJm&GDu<a|BNgzB}s2@w-s;kxY)~>W&qU+G#~O90ts@VYb}V6#WBzkAK83
znGgm}LR}|E?RV!Y&Com}uSm+ub1$7)tVv(r@4z3D*rw<{yr)tXa}@OyaCl~&Y4k@x
zP+*uK|7C3$J`+ZzQucxFngst&$xB(f-?z8?HqY4-0q#P#P}Xa0sLwAEpgRJaXsyB_
z?j$Z~3BQoyOJi?JHgc;{-$Jd=!1ox@^T7-Pb)AT`8Tj^uV*RZLi3h`5BDphUAe&w4
zeVVkILVQ{)G{8~bu7SN&FJna#{n1*=$8}Zkw>b0&40~9CVUJ=er>|IJcT}ZB-X*cr
z=e&N2+E<8G^@!?Fmq%qG2rpw%*bDCiPth5JDEwP}dGMf}-S}&?vxVZYAHFR>#&kVt
zt)|1ZevoA~>QE{*;8kfs<qDD06*`TI8(Cqz*zIYjF#DeSInXgh79IpbO@363_!v{{
zC9UaDYy2_4=JZ7owsBg1`NT&dP4gjJYpKSw9jf&nF8L3&1!Hli8S^33^JKH*1r-g2
z3G$aXQ6dR?lPGOw6t9y}@lE(zZg#WG*?oH72x5W3#q*ZcMelm8bZ|W%nY`Fo5iARg
zGmPuq8@y6r*h$)QJ9_#bYw-ijV6TbR8PmA0I@e41TTqvSA<v+vfTrihxWIlI{vbkm
z`R?mIX`rsoCohV$tzMdc4bB2(YZLeI6H8UIQ^b<zGh$0$Z$BDqSzWfws22Eg?<k3u
z=55=Q_4ebkCkS|=Xc(r(Y5rxiTet2{ZRH`q7gg?q7q-?sxZzzo_s_KxE3By^g!JrU
z%!qoA`^<2tn#6fWoX)x~Ryb^yG)%_)JF_GJxqAm#Q=(}m@USgdS((jROEq&IV)FW`
z9uG2)wcFrh3EhlWOH*`<%(FpNjlXyX-wn3<4KPUS_<4evFC5dCTgy2bH>jK0`jJQ{
zkftk;DhbOKpYS|T<-g->Zkj~Mq|k@+6_pUgj-^vrcm}2fJlQ&F8$#Oqnro-wwZffo
zt=dF>g1RIJ*|;#i5_`Yp_SK<4;0BdeHyR4st7*B7G~az6CVV3)i7gX4d6hN1$jFvA
zc#7idtf8}m+Ct-$+P!nc+_@({o)26^>4gFiG&IW3j?22=Ij>>key!qZa=b^$KZ2r4
zxjwP*MRfiiIZ1BMM(Sr{@qCSszu?5q+4KiQtc}$QxCql9^hOL4IQJ#xLP#0OlV9NL
z$vQ)}#CUx-0S3EgZyBn=;HHMmcAwsn`wB!P)!i^1tf|@k{Ib~<;}t0K<VWOLfJmKB
zzY^cr6b*{+L2hlXn*^WwiINHJT9zc@wEk2r%m~crMFpbOeoO=>{z-*#vEJ!)u;zOJ
z06$EV0buY6P33Dm-=ihn>OR3$-LwBqNX+RCy#Uw`Tgn1;$+l`_o`F;QH<RaRx4`<E
zG@=%R=@AkueZ~yOTesWtqE3FQH~HLvYnJh1;=yo_rVNwTV=g#B7nNZOU|^zhDKwSI
zA?*~%YPaX*Q$(_a!K2MY+WPwtL=3CuvD9%<02%=x|3dAIC-GmJqNSgsO?dy!F7(o`
zdv{D#(6_FWeKmU2^mM5vrp3Rd<}AL~xu+ne{?Muuq=9cbTdD?CZ=Wx?rppbEQ^H&`
zYQgKrxy;>7f_{8CzHa4igEeYD6Un64#qX;uC{%3slSZues**H%Ux52v3E^(ewazp;
z;0<~Ld?^vhI!OgCy;^~O^L;D5As-n}{4$ckI~m^-4fD91Pr-Z}WEe4CYK@aE_13mD
zM2e!J14bLh-1?k{4%S;s?3U}qe=~dv%m2#YKdq5D{$p}YPD!0bNko-!!Qqsn>xG~&
zhzO^DRFDn%3r=^-)=)({94;jWVF00GwM!QtWL8E~pvC|$#IKu;oB+<71hzm*;&++?
zug{OTP|tVd)0YHF4>mhffZ@ME74C<uEt9H*#1MBi$Z*_P^K`&x(o}^~CA(-dQJ^=U
zY*-W`w-?Q^3QVGpc94F&1YmqwdFU#byUH=C?ch$tm>VGaW+NOBSa}&HovP*IM7$%1
zu;eLm3GY3ZMz^A<^*~hUQ0kzt=R}wb1fb>Vxk9-Jj$n;qbh!<TV@@ae7b8I`c~i#o
zhcj;!5iWoVrNj+4?`VHEr}laVh~uc4m!AfS)jPENO=-&v0xoz2$)>K7#%wLUF7a&p
z<ZaZPY<Iu(>e@KRty}&l{fQDu<Z}|+x%hWCmuz$+oT1<30SJ6ym{jlKR)x!mc6L*D
zZc*SCfjA^o08H@QKCF4;TN?oU_-CFonTH9KV2FJb`n!>JM*_(cj3rQ*>&d#o<sgN;
zAtRZ|W>Z7Ru6OeuvvyAu{JsRyoh!5YBr~^Rc_op8$PYB0&7a3qiOz-Ho!xv#5?QPV
zxU6*zqD%eU6dfs7Bg<Ph^JlD26crbwDzUmwwfccPGS?PN!%)E%8rn$cqyG5;)LS}*
zkfJ<ml?0KZ1Bd5vL3EHyUYDyyjg^Mm!@{22{wzLsvf>CH*|CtHAiqQAo`(ZJ!qCes
zJjC2_J;yj8WL;Z7F}_4Zp3aLR*&7BQ;p5Z?!(%2gt3k$JF55-yef1Gdwy}v+FgzyN
zaKbWDolRa1WNa<+HHR7_jc*Y#Tc^q<EBVkPc~y7hVzCP;2M|9)_bc6|0+H%d+}p|B
ztz#GPKFF8$m#oxcf<=97b*(Rk3i$OHFVE&l)jPy4MdKc*9o|^yUaD$6>CGDsSj|Y7
zUi~!A-%_fa{R6v#*3%6M@yF{(BVRsep}2?{3la#=(BoUD(`NIo1oL+PE$Q<m0zAa(
zYpMe&uIYKsnv>{)oxZM9YX%tvgzCHsx(cel3-AP^S0VoRMzs=8a6*KB`HsTV(5eBU
zTTZ={g4jV5h2@x=%WeBmpD!mGHpW((v?lns5-6c_pmmgag*B)z&QnunbDARH4(*<d
zG~`f%>DMQs&K-_)afT#%?f{j+M$8My)m-xJ*5M?QBN?A8o7dcZgy({nq-$^Tn&8-b
zkw0EFH13Omd2Rrc)|uzLXZL&ou2_r*aa(!iu*kJ)5?oP~7FM)Yp6g0{-Fg8?zvVY8
z0iUVz(tOIH$8bH!w=J>>@nQ(gIIXvD{qb`@jU27hJ9osf%(`Km`6<cA*C$qoL!6#h
zP{Y<v>Kyj?A5X=-*xQ9jOStNdVZ&82uf#vbY{Z%G$Sk`ivfPB`?0p)uPBFTz4$7>=
zlO_jV3hE}6-@mCmD(Y3b7RO061&KKTt+J_s(MWHORJ&mFjwKa^LO+7{30R->#qE}(
zm%md4-o+2nYS^Rb!RnBzEETtj4b(kS+h*X(@qg00Ef_pf3%bvUz{*39QDDyNsK2AV
zx1HUH9mir~fB>p_Pb#K-Bvp*^1G57by{w+*eCdc~KN5gMwfKcO=9D~su}^Y2NI&(Q
z!SfrzwA^_`VOvlub#TG269f9<#MU2%OKX2y@W+ov`EmVD1JNxX{oxKz5hlIZV-xM+
zRNY(1JuL4RuPy$MMko(Y@CY(HFVSeL+ppO?`w1Dq|3}$3nK=;Vj&d@|gASd9ew3BG
zDV!ytdD`8d3KdQ^(_gsW7PchITRKgcsl-e0I%lw8CL+zasO*##jc5gM$v*>>T;PaA
z2Td94+_0bCc&{)<vmV(oC7JF1_KZL%aLJF1WS9(6RNfKAaSa!gU`}OwJgfV#a^SAx
z80Uqd27tB5?LwbY2FHrb2chea=00u2%Ln5{KDs@PtDs+9HoX7=6IOzLwAOn6gXHD&
z{b>jYOb32~UI*==!Yd``48Is45?I7c41KcC;nVEnl<pVw<G0ets@Dr^g7lvKBK3;w
z*exHNokG8%V-Gq)bSc_~Y_*Bqh)&PaVXaenNZ?dST>UT00H-P_Q*-F>H3L<=u!pLv
zBr?MtXMQw3K}tix)(KWlZIbF=6A6!^^i?5u8=LN1U;25|JmC??8OW#OplcDajNZhp
zVtKJiSj)^762`IUf30|)2e8~zlLNz4s4CoY2CnW{PK|;S(MK)%4B0rTIE_%RqB@4U
zi5n-<a)VYbinUZaEwA=d!v<WTyZSh`k+me6_HC=pb$;p=73`F_G5*o8HaFkC>k>rz
zQYXtMpI%iuKL2iHaeySIJ@hp)NcasG=+O9}mtSA*)WyTF{y9iNmhk3zcjpHah;<Bf
zyoZvt0hWF=8}*ThIR7k6cbA;_&zic??m9Y5UovhsD}JX{d}93M3i^5F)%$N?d!B`X
z6nJ1Y3Gtv)zSo*W-BG2Kxh{_q^p>Lck9?qbC{t+FO{z0i=XhG~6%FH>S!)kBWB3r}
z19UL;>HF7e_xw6WlyZ5|UfioiqAnUU^OP>CSIq00+GOg}+1KLutY%M}N5IJWz`Ix1
zwhC>cyW?sBaXbUDM^0ap+VXF|b!+(0A?{o*JbBV6FA&6^k06Sk`*I684Qy8~!X_z0
zhK)bXWDx+Uy%WP}OU@o+veDd`+3MH+Jb6Ti0wngDl3HF@z;vy&TusYzdJH}`n);L2
zR-B&wRdLgw#9!((DE4XxRYFW8QXG4TjYJh?++3-Jcpx>`Xc=gp`3{(G_*m*@un4|5
zW>w<Gho!n3d&@Tx%?IwXChRQ%QdL>m2*_{O{4Sd!!qjk?8XlS`Qxb_zjRX2T=3Y#f
z4z@l@(V`ZAx1>hER>K&G>J>KTLfGHndS3*64Z+@YW_BQ&bxpbg!M@wA`|O+kFhloM
zw7({<NfIj6r*wM)EK<p&NE?EwD*S~+QT8#AcCow^aL!6By)@4H>}Js5L%|A8c*41v
zs!bx;L&Fjk3u;kTX^W+|FH#e(tTzIdb;h|Pr4&GaNea$V0*YQjQjR?nyf~jW>*g7I
z7<H2p7uu1{la!z4vRH{Od5o@sTiWSUT1IgM=t5?J;z*~E-sq!bsn}8*y5t_YJTsx<
z-%Z5u3%{NFZa<LdMMr_UhBG+a2;#NnF2dd+E)_%Bl8-ic6}%E&8MRG~D4GHS)baY+
z8!d=HG|P|nL<L$u*t4em<EUAX<(i+N?`H>CY)g?<NiYAjmDIYC=Gv_Mr3*y)$D1@{
z;JU3Espl|gxO*kX(8OeBgJaD+N_EDY3q;Ii{fx5sM(~QuK#%gUF`eCMC$ZjXEIIO^
z`4$BrvmN%0`>uEf(V1M2k!qFB9h>}xL9}cy$e>jql34mnQ@c3cW!F>*6b3y?oSI?U
zpte@2%x*;jv-w)k!@K}CtwO-JzA-*EyB_~H8ARS7aRftJ{ANc5oeF`E&fKYT5;aPQ
zmlDz3+z~!2geOh!39{<U*<&hcpj5a??#s+B^azzDusuw*=6YY673Re3EOL~X!XRXz
zm(f;^_0H(?UbTus$yaBXIf4GzxXxbv8WSr*>P7Y8P+i=I^*w(wbPU=y5Be{<aJU9A
z7KsUyV9m{^aI;XNHWMm@qAzBtMv?L2Aa+nb=87G}X<3lZzRrLRVQx$f+ZDcDA$>Rv
zmkOXjb$FLa?3!}#fG5Xkh%D%N7_L7`ahCaJ`<ZCH1@oK4wpp+*-(^Y_H#`UsyY*uk
zsRlZ$cR+KJM{_%<@z^~f{B7StEd%h=W6m$HT*34)E=M0Pd0sIN7PD#&^l}lGN(&^8
z-He6ibkb?hMsQ8oxx~ll1_Oi?udOE|A)g0pL{+pVebV6qrK1PvA+o@K6^h2%S<zic
zG$it_M+}EDle-2MgK3&2e@393+8$$g!0lWo*9)Hpn*_-rZR)yaEOvYD1f@essFAK4
z$sdFqqAgh;t{AV+EuVV{+Za7qbS>uZZ6840qHwpHt?60eOmO&mMR-I}c4!Nn>ki*l
zW<#VW(uVAwBE=gRHH>@r!o&HvnkUGUU`202T_vOp_BQQ^iy09DkOX>NsqQ^sT-B>P
z($bHb;R(nZsDr<>!4c_T4=zDJqQVv*mQmy>-=m70#1CcV38ADYfnw)9M@O!eX4?{R
zO{>i&@7n28H=4D#qJ~?^UHq_)C0XneVL9$HOr!9};q2`$1eT%}l(~Rfy9>ZswD*HB
zAYIt50sy(<m$G3Y6a<$U4Z-|q*+CcecFOk0zshw)$5ho-Mcir4?CNaBNt6^Az$sfz
zw@p}|kP6cz>ae4RDGWW9dJNP&zDINR>Gh#QVWRIJdxxplt&xL0w%hPpdXBdER{VVk
z(>fafcw5BBj<FPuOK2wK{rKrl)-w>X>Ryd0zJ04HW2GSQ(t0~=n_WX37)aZtA47-0
zki&#bpdIO;BI%w2Y6zi>T{XnL);dX)&m2RYp>%icO%M9+wN~!LvG_F7{qyC$kL?91
zXd_#&mXf-DwZ)?K7wYSB5MPjlES8F<Ac2QH3*Wx#)3SHJ{vpWCN`UA$3ZMG^hK{ae
z>SOInrSYz|Om@yMbltx=Amj)J)WX4l+8|^3hogqvfJUWD=2#z)Bbm$9NNoP;OKO;+
zn3im&c1hFBjct=K+_D_`cW|a0%bhx_Ts6c-<g~r%{%n_cALU{|ox!0KFiWPOu^-XZ
zezes>fd#R9%k6b4Z1s65%<<BU1sMSbH^E}3LvOdod09_q@A~tEP8iPKrvNU}9jm#S
zNM)@uV7PNrRnGxXnOXgtXZ%Z($P|pmP50VI{M?qtVLX0kM#jT|8Y45b`2gl(RI+!9
z(V~`OYjH&QQIZg35PhgpYs#F}N%iJ&@++ovz-?z?7=5$AN?vKmNSsYnr$6pKh0-8m
zm(I_A-+3+^Kk%K79yDwc=|iw3!?#ZnS~|G=jE-bAwZm<669CuQexIT(B!SV@E%irA
zIt&y&gjX$<dIix^87`h7-E-Z@8&KOB6$x>J#^btb5MCH?Va{GD&V5RAWyCJ*9#k{M
zJjYsPZg0%;h#Uw5Lb>6Yz`?foqn)%Z!7sJ!g|!CCh4?KguFBQIVho#x8H!h6bJ3%F
zrgyDW!6uKV#p2>W{po}$RGR=`KqtP{!1n8(SpXN6D?_+usr&JW(;$B*rPm~-*V@G&
zZb?Sim(3FP=O*iWi!pTjnISgZA4ePyDlop?gQ6Qz@A+3K619<#FSsWi&VDapX#w7m
z&30EshDkX0@Hno(01u##I;C}i=1`l=U3M_737~RBI@;tC7}k^^QX>wUlFjkL;<(Lx
zFm_;bhy46mAtbo>gTvQz><7dvxos_V9||ntXas#rCbaS90%+}I)mzyf3M#~;n_$Ds
z=c>ob69dsbdRH0#!AVGA5&#^!rgybS&BffRd+7(p;ep?^7QIdfUTAoSUViiTR8Kzc
z;x&X(?v7xhFzMFSlfIZxTpq<a)&N>A68w`-t`eF0k*h;pOq1q_^Jk@n{qMR`GXNB#
z<ISjY528ga@f0`nw)_$NI~vFhsHVH<!};}(ev6fHH6FBJKE&#4$j<J2sTiJrdow^4
zD;pM;H^+P2u3!7414#u#vplsTbi5#RdFrIR&(X_B!RI~7)D4rbEb6UHY>J>f@|pa&
zi3=Zvh1Ztq7DRQV8MQ*Sv9-#s;>plP1=B*Dp3^b4Z-Iv%Zo@Wi-WIQVi&M%Zf5PIa
z+R5ZEDqwLqvU+mSVcJT3lLPj%dSWwx=xGID)T3egj2Bg{jD~IM$-;2$qh)o`n1eus
zKVg%Ja_vUC_}FTQ+Ft1BS<?DaIVK4z0Q|P~-F_X|<gfjT=}oIJ8M>MF+5E)0SPkgm
z(;&LCeYfF?Ew<`AAy#e4boavE0zsG|q0-GB)k3Q$%>(b-eTh$VJ{g3r`C2-n&^6Qw
z;g;GgwD$b@Eofl%*AAM@AGBLKL@QSh+-)joXAKDRbUlz7j{l3D<IXyk4#xd?x2CAy
z#Nd#ycWC>Yo#Wue#y_?L7V=ZpJ9;?gw%^d0*e(}?pI7Aqu^1FZY;&Sv%?0i0a!QFz
zzBiiomJCv9*A*-sSZ*|qUZG)l4XWthk+21S_M#5Izp)x4O5uKYFlWnisw0PLdtCIG
z^<8i^QirlmG&BVmz!x{N1XA!H{#A^Lv9|?5YyixR1R*<d^ExZ?Ipr4EdXC|g%%;06
ze~3H?A%T@_VtO2;JsP8>ZLeNlL|-=j1&+PSnRoY-q{PmE!sO@9xaPesi-r=4DYh*z
zkoD^I#(El8T1aBm@6Ddwf!$JcNj=x*G@%2C(~Xai7hvV_Rl1e<+!wlpZ}*_bz$~$Y
z9-K4+9$c)(N-yTY4nG@VJ8*K35Ft~PQ|k5lAXY}9cZ;1gQD7a%PAr{D?pZab_PmSg
z-;zcMf&<9ylh)c-3WCb-LT3w4Yj7R(`o`B`7bc-%)eI(*2lVH-Q})t63bH=x;yG!t
z*|Je(x~N(*sE<Hve_{R^+jX0-S0lT@PZr~akn{v%GM%GHC9z@lQdjrUMGlx=5xDnv
zpAAbmP+*#DC1j*F44-E-Ks8<O_Ec?!hOBf)AxUr#%nVR-8*8Jk`Qw8v?qjCf5K!EC
zI(pn*%)Ca@8j1V{KrFviI9<cbSVn2O5cLcUm)xEkZX-`I$|x+qR`wl2!xe=bh~ONb
z^XxZ~y;IHc-qf1oFBXI3hE_3WtTk+=jpTC4>a&v{Q3Re{p>?F=taY^Ky=+BR*7WG6
zJ;dFcdVK;+^x3nAl3fPRqG0=sB^L!kq>sbWubVP);G7?xZH5Sp(=GU=G|#Tl>LXdN
zq@aypol7SolIKP<02mczT#Y{8da?JG!x-IndeY$z`^)CJ6&wm`NzRZ+lSexqU8*zH
z=iqZD752yf6?9y27$;jnX)D}x)d_)UB$R>;ocdV)jpZ|Ns<O;xcxJP?08Z>&!N@=p
zFu`HXtpUGA2KMeVjk<T&QIZP_v^HyvN6e`IDd-OUy4BoRVzK{}bhck7SQ5!>HaKAw
zy1hM-_rk4z{WDry&`=bGlRVg&=B^faIJnd{b>z-*NG%O;ms02CcQs_Pd5y)N@7>`Z
z`~HSQT?&<)9SCNp$%dC?$RrHVcoNSz9+@u&EzSy~&(*bWo)E!F3&~C*1seB>{CCd1
zBGXne&8;jEmiKh4tvmPkcEGF@Mc;8TX|U!sHLq*X-~rJA+l85%l=P-mlI067V33fW
z&5vv^LYzU-BINvh)0V^&bK3OovTNMI4W5r-x5`tDwf@b^fX_0##;C#;o`L1HpsO<(
zbf`9v6@{-(HquD~EKj!P#+VX6KTd&Z_1RL*!TS$_-1c%m$+R_aYx1CxtjjAJ6Y`H?
z(1exm6aDrj=ti~Kl$LVUC>z)}1)gY$3L{hxdbU&mt)MDMH{K)i8Y;|_^Wk7+&g=G*
zFOWY;z{q5E0J@%D>dj-^|KnBYFqon+j&Yv(96ML-Bs8XTMne?@Tg$5Z?CFhpqC4M3
z^H*<&+<q5)&>NRK;@)8e>#HJ-$FHkFbZ9{jv75DUHSG|qQV6SY?Z(%?7LEk9UzKZ8
zB&$thM21(g%wx<VuB|pPw#);P{^)v{W7R{};eojTdIz9Nna%WWqEfj)sf+(nf3+lP
zNTfCP<E1X<w{sk@w^h`rXMgpbO$fmL-Sg^`A<{phwzrcEGSrj&z@_WBcRpzeGfy{R
z;I4d2c=rbNk$iMsX4E&mOQOvC#aU{UFx~ps>D=<QoZ{78F*wWlG-w?j`NZ85E9ySy
zVvKy;`;G6l3nzc1KsSM>>oigCjNl56Wo!v*Fo<yO6W&v2>X6|`fXSJ>2CAwDA?({p
zvHYw*M?Y;wh|u83Ux~=g2TUbnAu9!5%e*^T8OKozI4rFiA^5chU~`eM$&=pXr}qXD
zX?59ZAM^}}RhJw*!yz;Po?F6A%DVHxnJ*0--z$Ijyc_`Y%icw<N`uEq0O;Ua5RJw>
z0$GgDbWZOY*{Etd>&fXZ_w$<KB=||05)agN9f-oWt^fry04lXxLpzxavuU}-8BTb6
z2S^qZO0vs+jA@}{^T%4b#b*T5Y2=V2E-4h1;|Xqwc6qcsJS6z3YF^>#y}fRlEZur6
zz==!W{2YZNL6w<X@#rTzP35{TTGa2~I~?HW>Qm5EETyXvCZa#T(pYgXF}<VV0xfv$
z5aZh_1*18PUIMnet$(l=K&lm!*aG_f>@4C690?wvb#pRa`xWLI8NvBB<)6M*nXpZ4
zuO5v}PiXQ!fZ7Gtw9>6rNJlaHBiba8)~oE}bS}=1Ay=SQ=sveQJqv*?+Scc3U~j50
z1My4hYTT6)@GfmHeHsEB1qI$<s1KN@pI_4*CWtpK<*E@Ov(W(fLy=sK%lSP^L5^<Q
z0U*1Y^YUY)3m70eaTi|S%kgngtZ~af^iTV!PBAG`YrD=zwGe@uQadB0a?neRDXqxZ
zmktE>S#-r@J}EO!lTj3uqvoky3H@{|t@ogHa=;t?&Jz1*@~eKl_UNAb2mEs7V48q}
zfcHnXlqFC@P$irTej<OW7!5$MF7$5^{rcITO~AFIu*Vg^X>)_VfH(&zu<}W?b1dti
zZam@wVY~sEbPIx5ZmY52v@XLA_%)dUMsdJt-%B!kz$l_h5H5|{+hg^ANhlv}V!mfX
z>tun>EKG7Q*23O88s5Fg`6nqtl7r2OpU461cda5x(<-Qi>}d^G;5XW&Z(bm6Fi5bA
z)GL^uhqryo6nnLYZ~--Go7TF&X06GNOe(`cEvt$&F2hE!x4C6nG5&vJ_<4}K#Z<+5
z-P{q=G7(h0dzn1U5agJnRC;6U03L}_n4E_X*Jnwl|G?b~+dzbV{KO+*hMN#wBO&Zk
z<@4nMbcxnV`UUpCq@qYxXUf0mCb&&es`cv*`M-v}i33_VmhfKx(3ss_7N$xtq0v{Z
z|1x@SWd-0RwQj(=wJbmhC<Q3hckhsCCXLi|VouCZ{c}5P8nh_djm~LCtvntj7O_7m
zrlX=cNHIY|$m)-F3%7^vCD<S#jjW?zON_uFLu?Reo$^p4B|S^1>w**a@Z89SQj5LX
z5GMmUu1CMiUnLk=cdYsgJ?(<l;dbP+6RRaZO^nyc%Z#qc*My8I@^j;u>&rz32FWsQ
zl)N8l8hi~rtTTmS>2%pqn#zIH5chB2oIN2be=sbDObo(KP}&6-(2o^n-G!2YnHcXd
zDj5qXfMC>KNG_n-peMp`Bv`2rH3Fx<F!`j|lLP?$Bm9p#y<bebRk~Zq^tn}>QNgcg
z0PzbUQG#_fF`Qjs^wI9n%&YSUA<Dpf^dX@fwXg<r>FLkH-z3&0NcSd^!LNzBi?5Dx
zrmQB^+SWzJ96)(e+6!*C1<B51d^v`Fw~P!f?KxMReF)PxETCe$`xn3Z#=CqBxGiuK
z{8l`{GT$<xBqf&KD=g%W_1Ww-pefjbhYQ|1&eF8cSFXfBQedb9C<r+LSq88DT_*86
zLzs&Ij^R0pX3Fd|<I9OKFatoYtp>EqR$ZLs(AbZF7IYZaRx<6v86XB8_D2pW7t~dH
z2Uj<gRBn?&pqm2t9zuM(8@#?{xw7QRU#~QB`jMU5<PT66RoTV)T;xTa!$JN(!|zC@
zJWdUx!daFQ$0LX3wwG-i<#__Cw=R=YGd#zPS(mx@m&FF|E}^4X6c4EWRRbk~;0e2$
zmlw~cNf|9(uHn>B1h%ClHNAc$yjV^8lI>#Wv5}=wFcy>a07B}eVv_CRH}B8W`{}c?
z>3c9P={?v8tCr}YE&Jn{$ih^HK21h!FRx6fcvRjgtvw@j5d_WKOC6oe=Mm-!$6@ms
zbI;C#rXtdUM2dO3OR5%}H@iiyW|{*0FuxSXSl@FrhukR{yT5nznsOOFF=&L%>B(v@
z>fm+N+aU8Vz|~pRp5Av`%v&U-&VsP<admG^%8&h#*~}Mp)!9ogYSA1d2O(y(zl=-P
zTYrsJ2coyagsZ>)mE|G&5_VZ}50G_#Iy9eh@E{}$OdjZ8cm<J#H5jC`(un~FK6fol
z#TM8<Xw%xd^SRy!rsX`XTW9Nem4>ZV7o~3E?D5dc8tV&&gFX1O8F1Eu%Xg4JNTezp
zv8YKJ^B`={J1XGxY;eoUk&!Gf#g@d{sRE2$b!E02qWmA=CM#)GhuRvjCT?picg=4k
zP;Ps69|?X;ayj!l`9VI*qo)t9w9R&D_GZ>P&O2r)9KZh}N0XOM@$u=_>+)I1X0-nO
zOBz(CKfEs*d)=jv(7XITC;XZ7cRJZCqFiqtK7ojRbuFfOm~;Ki=daltUB#ZIxrV3S
zRjpOd4~lr5Ub@qzf-QWvZM1fgKS~lxww#Aa7uaaKNy=o8>CA9(#TEtseXYav3Di<`
zK*)T{zuWmm75Q<8BPbTo)?Oj&xll>-W+LA6ijO#}g01_!l>-e<5GkXX?iOoHjXp<{
zaj{M4Bm|LDHXC0!T6}iof%)I4xa^JTR?1j17mV8osZ!uQeGoFpY05r!=yf<`rH@2J
z#1Jd_`KA614Q_DcdG#R=vma&j2*C)jk@#NKw0){qKxNUVXr=A%lH5=|W!PwPYJuCp
z^9iOko$G+9bl`=(npLMNsoyB8N${OyF;LnWHm>MX#$A?aHA$mE<GaG@;=v#{T#dKI
zmA}x|3tk1hkvRHlYyC~+U*%p-#N1QbN&e9rxXWSvt=sF<%P5Q&*qL2iXmv%E-PsLS
zH2%K%ngXz9sswGE{_?KSOZ1FxdDsCI>cskTqru%{4nEkkpR-Xu8h_xT)?28UzHlp#
zpM6VrjR%KgGGPq3XD?7ES4`=1UQGO7d)FD(bhh+0h7~Lzc(H&2t^_tB1VNNetQfjM
zK@!?tL=X^^D&6R+YehpSQbJT(L_m;834!dYumR~sP$0N~fly5h5L({z53tX3@BRLM
zk}vQie@@PsGxIyYnKLt~Mj#4z_(_ZdgVk<<4W$+PjVr3po#6oH;T|C2e~u9{w}(1u
z$*W{U*Nlo)4?@<tV-!{8#)pOh(-OL(d^0&q%r)i>uvI3>kJteZ&@MWn4vg#Pl%r5#
ziMxda_g-DGz&GdM^?L9zQ!a!Dd8PX`k=xh(lk4hz8*SZ~N*^9RJMR4)*pvuVi4S@F
zN`Z4*CA_jbOE;gn=n2iR2sV@1DfAT1iHwSB4WYgo16#XFzIJBVjxXGt)X+0Ew_TNG
z_PKsOV(5GKy$gQizjqa=quu2IZC6{{@wj1I?)J)I$JTt`__?hOHv4vi1u3Y%Leh+?
z_gC}y4=mN7%yi;x71ngq>WZT^nir=zT!+m8qOZyy6rRvt2jC@bt})}F%~`AS2qySb
zBB(9jG4rd`Iw=Y`KH$?X{Q)|R*OdFtMbT_!0n~RWLcM?9LeTihdIcc}cVu(%hBPhx
zZ#^FV$;te*qu0>(fUTw&?krbxp*&0rJw9R9AB`~yX+;GA_<+yhC{?umc+uUe0u^I$
zi=zJON=G6hi*%0|{8K^YpR2>h$mR-mzirD#rG<39t?ouW9#TcSi@?Cs?8bOa*<4Go
z>w2awpNQcvj$C8HH<I>iCB4KS_#?*<V$vqEU6q0LuPf36PY^T4SU0u)*`+Q^LPQuf
zJ?VpKSk3ww-dfaNG(`e{ujY<t8wR%L-s5DX5)%f@SW;jj-uKW9RMx`D^7FlJC0UP^
z;F6oNnH+Z1{M*opu__b`wK@A*=jw!Oq+U)Ywy;euekl_5m7Ryls%TTexbbzPQ!#R2
z+{H-}&kG|xnbT7YslFWi5Tcj8zrLX7pp=&p5zN{CXGjukWY$ZYz$p^^vY1ssILSPW
z<r57!O40J*ZtEv|DEiDP5X5cCy%)Q9g*7~c6e>T0po0RNY-;TFBIzkkC^s+p0POuY
zL6Kc1m~Hefu;OAg3#hBne99eItdU_9Gk_d5ZQI-S?y`=oO0y!cs6KsqRuh~{l$bol
z4VZEAy8<b!DFN$>*K#t_sXAs17!2XUMrWhI@hREcCSOCSPQ=W%v69x+V8xtkKiik;
z)#RejXGF7)gZ-}21R+HKqw7n~Qnb=bwL_XdSImO@c0%YzlAmpzWj(_8a0xv(6o_$_
z%&bYklU*&!br5lXFK2Z@>q5eO!xARKZE;sWxWorm9Fy8I4CVS(DL!E`{sl?)s0C*n
zWAY|xrUEK?GWl1TX?o>L=gJ9@V9Lf3%#0!n%cvatdMsz$a4OCciSjwFvS-yO#z7Ys
zJlE0&r_Uc4&xP!ED^$2}n%H<TA8iu0^mu1uhMLeUUy};Xts6)?ZfZSZ)XgAyhz5LB
z<_!E$LHAr`dqg-!dgz}Szp91_`bfNr0KfUxtfpPzUndq#-Uk6Fb_DE(@w&tBx&*gX
zSe=pb(RQlnAvorlXC%ig)T`~aT*HKCoF{A`5I^nCe_?bXxPDk3LvDE&1$GEfF(6L`
zO}oIMse+cE3!0|W8V?$gyMXFIZ9B3Sv?dP&fc+zgSlNM}6?`5XM4<btC3_yVEciQ>
zPpFBLCZ^aDhw?uKhFUl!o>x!?9|URG8iF#)j@Z;OI25#Pcl2(k7S=c!kcS&m!f2S8
zik*4mRqqM*w@y`lKX{x{4X;Z982fCgZa)+(i9lT*NI@y<p1G}6$Z8TSg&tl#w&srZ
zD`pLLsQagW>(s$kL_a<UvaYj%2jNp^u7a-K(2n5OR!!D<E~dEo1J$<ZCzApl%FmDA
zBKYL*hja(|>WbhJ3dOpsDzlSwT4^-}t>gp+Z_#IZrvB#ZZ4I+0@fIV%*9rj87%?!G
zn7E;xZ);|d-k3@>&d-wX0N$%eVoAp+th)GR_hoaDV+$ll#`YcaeOTK4s#-@&L0{j4
zWkH|^;7di>OVd3kbj?1CjktL`FyiO~YI{<(?2<lI>$6(3Ii$EXgtc39oz5i|(u0|}
zJjm<A$d+AfNYtg%UjD**8XrgiTDrVpo8@y^E(oWCUM(%7iVCMeL(sY7Kn-kvLR{cW
z084AaaJw8Wce58t<m6PE0|{}Sxs@-ROF}*BBVT~d$OQ6+sEF=MaSDbilQz72CX6o6
z)jgxiUR7anU2box{(`CI=gw!VUB^pp@rH^SQnRV`q>b&mO~xuRjG0za<-hCRN&-Ej
zY^coMkDsYHqFHzt+POWK0XJ~P#_M6EN8d+Jtf-iz6@j__B%QuJJ*w{tKD*!F5zGwA
zOG%V=$fwE!lM~@AoYLDpat_`345;%i!W=8O?pffs3foOKUJ!5A4uP>3eW+oo1|eK*
z*=i4M?LT0886%-}8n!t7f#+MJiU)x9ZP|!ahQrk`6fNUR5z-2-Pq=wmYInMd;F1fs
z(fND9Ir$L=p*Nv#V^2v0=({Dz$HZjMl+stKlT@^S1tr@&Zue)e1<X!luar^85eQ6B
zVBbu4Nm><ylYBBMzW7gdq}t9;-{3!;rVJ{#YTm2o(L@J-7lZ&aM%%-=r^fwE$KE7$
z;%fSI1hUpF#LD$N+8qU-D9m!6W;U<eZXy^PD4l#i%79v?%>!Vgv4fRVUj@KISPPz*
zZIhqvT_J!}-lv_?p5@s+mk<S+7sUXU>JT;bXw!J5HtU&cTS~muY6c`TcI%G$$RWC6
zdW)Iyx3qz(I*6e$fN#aUS!^U3<w((bGE+hd#Q2ccLRlK~>UL(2{JkzZ6-|F#jO??A
zR6)BN$a)Fu;kL)j?AxJ{`0??T5^VPj2~#^?&h>JRlHg`^Oe@rlX$L5WZP%YNws4P?
z;QlS-WDbG}&CG0S0>_m%Oz@_?@usKnlg3Z^<yGM6VMl`E#~#yCdwMK!5V`p22(VJ|
ziq686CsjFVWy8a^v=<3FfIU6{TF1U`(fmfL{RIEQdHsM>NR&VRaJ1x>WOqW!*>cfe
zY*au*+e0bS?kmt3&Fcu(zHcg3X!jY>{3|A9jq>7=A?f|AtbjCE(Y?_t1`frqvY|d4
z`vkq#Dj4xj^2{+dFo3lL<t?-Hb{@AlYTWFxSVNY!ca=oc=V9x8v4K}brD8jWR<VvS
zimliiv`s0<TKPPvY9p)#0#>-TC$CEx?A&#J)uPEyuBq)e=Jytn;F4A;=!V3A$vijv
z#KOp#%-%2V(4v#&2)3@F{HJYU$|b6-JmM}L#2)9B@I#3+#WmG1_)A()+}!#4d%$eb
z84&01X1=>rie^lGjNL~Mi_0YGe(pU3@&sc_(BaYKGT+R~%Wr=Xmm7N*GPC8~qORVO
zj=DrlDlN?d5djE3&;b%$P-^LzQF8ffZDeHOXYlH)Fre6xoWZm3h<yN&B5_#P^g)AV
zT(HVw&bMk7YG`!!87J)N5r-QX^0Q-EVM>g=iI_{&o%2=CjxiX$msZB#2oKG<a11&0
zH$*Z0k0{Fn(o+O>^4{${oM>Zv7T?CUbm54F&9LZ@0${y+;RP(a@aS^yaoTUt1l=!7
z9CC@M9cb-cyp?Gjds+VA3PMjG=}*Okd)=z9Gw@OlnKw1&WI;020T_aO9(I7h08DFp
zS~?bG#rvHRTc-hFs|3yRB08VZaw;uoF}MjO)B@=9a}F8Bq_>u5sx?iFyw_Suvz<H9
zuR#?QyU;^L-{KQZFW{?xON7$PJRp;omi&GPz1zS8H^IJggqX!%Qk^D5%i)0s8DmEo
zd>CAh01AHi?J^-Wv*@qU!^alHSLo>3X0C?IPw{Tziv1@3e9k@rWsWa`)3u{8K0eZ2
zf-B51(MkV;z%=eW2Y)a1J%{-wbAOX2JuzXk;>r2f*V`evHv1B+^WK=6rAH9npD7&b
zUG8t6^0QqbRnPY;*@(~0RwY6M&x4RQtvZmrSNlCIWaRn)(JsyB#+o$>WX$m+XVNlo
zZl>=~JKI78J=ciXc-IQ0QTLbp=U>H4fUqdrRTfy^R&QrurYEX5av6)Gcr7c9o7C7o
zs-s%^(Ch6DT|JJhn;_qU83#OT-j^ekCJC+tUpb>CEtQxw9WMV(3iyYPJ}&7`Th0#s
zy?Lf2Kg|?cgSH+lRf{oj#WOhgil#81is^S)Hhp(Fi8on;yq0vQD?H+JkhT=L3xYf(
z7(PWVG)XS`DxqTbdj@7?zZI|ej#=*=xa8Ok=0jCq9mC4(mr;|w^HXU_Vtzni0C&$b
zZ1KF;Ou=hQ-d$DmF~JSjmDX?vttqL4Z}kMcb#G$%g%=)t_|#o;jR&dRGspw7Xf=Q<
z{A}0cb~(y0{4BP~=alLJ$xU~e(<ELE>C557du+bo2r^$BAhpV8Zq5U78hzn7(QsNO
zzNW9WTV@5vWZ*u$_gKCcjbxTe_paF;y~XJ`c<VlBvgFhAo3u)^6<Ey)^Q>UECnUBd
z_m2iEq&&evvgb+4;T)qVCnsTF0PWdhIq|M1!4w?>!oioonX%&Mimx?8FQO@nuo?BC
zCd<+)kLhsp*q4hTP*DYQ26Rsn{tg}YhP-FTg4||al1u_PfoJMRUw}Tyunfxvd>t<&
zwx`~*cR3$H$*MoQ18+-3fY5TA(ikmfrQB?ofMyBw$``J!%VETJI?0GG$oK~X5(`%V
zbyyvL6-l0O4+IXV5M|v^A8&jm(Esv2fPT5c+wF6m#s6MRSl8?zZl*fyL1-%Ii+JbF
z4NwceXiaXeJ$roFmKT4yjJNJ<Lkf5Rk^5TXB+f&`_c&O9bxAH?@;d>|aQ`EyZJhjs
z<Tiez>}NUeTWD6r{jZli-&e$XpHxALm6F-F(GM3gZ|^%ob|QvjvL_egU03e4$DT9i
zuv{Z?Xd^nR5pu7aq2P6A+_@t#GaNVp&3wn)KE%v$j-8WLnPS7M$T;?69NX2wS%?rT
zE1JTVFr81ND;G!BK5UCOoc3boOh+mU#S{#^7j98tTSvbwW9-c+9rS&799&;fo&u=Y
zy!-w+cmxu#+vLRg&;<tz*TnPwsL9angj*XH2D;?-)qkwcQ>8uK`F=T1Ye23mcw~2+
zM#%RMIKCq0;jWLrDlK1Fty8xvCE@yB(<fp$pdPc-v!AO0st*Z^;GFp+b1-g;_iG@f
zCFDq_E>UOt&K;Xn&`h#_JzVJ&Zz7mC0THL>Wk9;2`s!jwCr4f2i41R45p*sWqK>7a
z8KtVRE|!d_#y>p@x=Xzn;#@;;DFiwqPaTus8h9BWTt9yG3f3)+r0(%^Gmd~WBV+qg
zN6fzU>F{!9Xm4`<EmlZCXH03l#=W<V=x%mCUvmaWmI$q`lyLFYM2WNen8wq<vxeo|
z(a-+EVO?`_Ieq9RRNlbM1SgucLjDb@5_5{H_9T*;E<p_;lm=aFa$aA_a-lk<esaNs
zEg4x`YPYPM3w8us6n*^kvA2SoNKaQ{j{CA4*X_0Ap=gPD6|@Bbi=#xl=pF&pwGl8N
zvY2w-y>_d2D%!<(QUW>jT@~dGI0O#WY|UHBe2J>>TnZWeyl)vJ)a(FuyP0O&->zA@
z(%O}=&davH<>B&2GyoN+<LWv*%N9D)UdL51c0SF7&Qm%7k578Gmn6Si0lqo@`!4+`
zoR5iS*22*ytx#D)w0y464l6P?m|gS~(kFU9)-qN?P_r>0;Nu_b#~Qv)`e0c%uw1N8
zFqNOhS(qmKm+$LcXL^vcU>=ia8~Hl~h*ba(xv8h6EGXrGV_Y=Pkv2oI_*mZkuWAW?
z*Wd-#n#!jtlz*xJ$TmPT&Ms%fnaFpQaXaHJMk-aPP1bciWY>x1!r51GFoe*#v@_2V
z>Dc5H{5@oTwoDvwW9oy5NR8AmXl+sD2K?(VD<6)d{b*tK;FBoH?{@u@_Vs7kOXs`X
z5{4YFM&4BVlfInya$m}?>U*Mjd&?JlIwB`A#Vc#@GGK8KzTTCXz%-a!@V-24X_0Bb
zMU6RT>&J~koF3{ku+AiANXn`JZ6Yc$zu-OVh#f6d9V*92C-*3(o4UhfR0TJvL)=Z{
zRTa*aU&w1m3$*|i0aEc2y_=feaQjlYMMb*3FN=yHC%<ExDH<L41AJ(+0%1MT+08;Y
z-$IgUa+w{rMr6IoPX8JF<8aX0*geDiwRdePZgB2j7uSES`ohT`Od=Yqvkoj0T^dhH
z^$>|No@4VzdyD69c)q>Wax!b{O0%=sr?3oMIOj2FUyDk>)+E|D*#xRC7g}q_5F!G6
z%J;BBvc6V~te=>z9cVI*shw>Kfk6|tx&>(R`CP;JHH7u@Al8&l!jTisk8x15+h5CA
z97;yz?zCo^6XzapN{~sq*V%h8c?(V)v?GK{`Bemu+WX`;Y0*q|k+#-4442I`F+Ih|
z?TGZcKZ+eCOSF&iFD%uo_70qbq>~H>aEC#MfT-@gni!6D8^xMF7jW!(LF{PDq-dje
zd3aU3B?B5LW^s<cNH0Y0RSbQTo3>IS$dr*3M`Yeg8%?u%aW+V#89rX*L_Bt6nJTnL
zArT+dhbn>%KQ<X~T0|D&gzq^h8WHQIZb0bV(EeCx{*{nMO!3O>^vWc~<M9qf2MD&~
z!uk^<u1UoM9z%auFBoovCM#xrHC3W^>)V%};C(nG|AwYSqL*rTSN~n;LYhF39ai(S
zr}mE05g(IUT$Iy;mTcK3+T-#N4Os(jVm31$omFj~gbC7gl~P8GG7?+nN=QYe!6RfM
z=}XL{9z|GN1^VtI1zV|tt(cix#eQ$!tfkqle3wnvMnVeCA6FFOa7$;p!#;;J=oN1q
zh*A=WeJY@645JZ08)~X=ae1O&RW;s}UvffsBEauFq-kjI?R?v?18Tv5ecA7M#fh}b
z;tvLo#@qAWsj6F_kPb6VMhk<nA_P^G=FiEfze`h89+RqGy<@wo4TXJ84bvlzp48Pq
zs0aJl<RWZO1^#uzWK`V0nvw&XK?r-gTB-44PUr-k3gNq;hx0zR_GRKbRRs_d?)d$H
zB#2mEmEdaFn*7ohr;qfU>3Zg4gXO3W9nT=`C>(P&6Xmz}kZih1{$cjF0zIS^7%elg
zpM1?wD`|B#NbTu_Gr%)5d>w40KIVi)y!Ht^ij7fY9T_n08-O-xYCs}1X>eojpf20V
zcu3dN@leLAt2LU4tUs4I^a@qp;!Q&{`oGs^CpuA|B;d%8h_&jGF*X4q-!K}E6PDa+
zwIM!I5QJHDK53MJgxrdYj)tohuG%()REHShH`=DA7wL1qVe7wPOte@Yl>)Z@uO=X_
zyk<kYCZ^m#?{E?^l}_t(zI7{r9Mt$+6td>P8R*<3M1!ooPv+Vvbu?u+Ai$L##JR3a
zej9Zt$9~iU8|YWe;T-$a5?GCo(q$#Z2U@l=5VWWRwE6d&5>O<|PId)5Ct@|<tXZ3&
z=m46{{pB5PHQw$)<aHI9$s~%L9TIBw_7~_wP;!%*WdE8*iooZ4S`v*ca(Zw&5u<lD
z#tEZPurhIp=qD)0q_5lO<TYl@hW3p%^Q`;Q4OZ&9tqu1xcO}>;C8U@mgjChIQ-QNF
zN@g8#i+Ezu47W-H5tf~56dw_3*q!44=7M*yW%<>&Cic5q7sd7hR;niH=NeLbA7>tC
ztsY0EHNMO;-cM6<ts|RlR3ZmU+j6@^%JclE%HVnL7feTD&mtpZh}=Hm*f!D=@6Vtc
zk?7Jr1{EQle|yxra+X6A$p<0Scpjp?7ItY{7vCgLT7T?|fwp-z{-IS)&~YD=pGTC)
zq_@Ofi7CgxB>~cNEY(+;m~z9urf=UX8_PNZrY7bgBQzk!&$3vTY81~y`%Kufx(?r7
zbnsbpz)o1Rrd2-^D>O$I@D{?86}<tO_SK2$LR{!2I=ylvAZ)C`)AgS$X%GX+i1?ho
zUMAM}`;-n&rpZp)w&4p>U?fU4E>UZ<WR;{emgr1Rw;<ZEqa`?Ay4%MhH%?@=P4<wH
zS?$vP5wNCpNV`evx-K?<dcmbo_46p%?c=48rdSP8ABGVIe)nFYC5~r>g5m~x>4VOv
z<KX(O`Hhp8Nz~YG?S<vg)dN7deBSK<3~wbN?|=SSjluqt<M2NZHcJAzp1gkq+%ZFl
z@jvea8W9Oj-A_jNm9L=_pK$;6;E##?x9EOI<cCClSj~Uvzz-bxZ;1UNkslKIar*wJ
z4*cLFz*+kbD)vJnKP0l8N%4clulgPTZ+^|Jh@8IWRq`#5xi#PibHe6$!BO|?{|Dl<
Bsw4ma

literal 0
HcmV?d00001

diff --git a/docs_new/logos/dassault_systems.png b/docs_new/logos/dassault_systems.png
new file mode 100644
index 0000000000000000000000000000000000000000..1ed0e876d39c0d88f482ba032833f1561217e94b
GIT binary patch
literal 21865
zcmYg%1y~$Qu=OtP8r(g>B{(b^G`J?Xlc2$!CAeFH1c#u(gS$IyaJOK=T^1I2n|tqj
z|NkxD0y90+-KT4Ms!pAXR9BV9#vsQ4007vE3bL9200J2PcOM!G{2MnT!5RJw$wo>=
z3IM2y$Ap<7!~dtXP|#EX0DPDM0B{HZa1TEO{sREGa{~YerT~CQ8UR4zoY|rw20wvf
zt}HJLc>epzZ7)s+0JOo1vQpY!OUGHBUX=aNL5b6Bw<I>x405z_$(I5R=po<ml71l!
z%<ccdlU50qd^xAcOj+a3GWaD#-TeI@6%jHlqyX(=fFhcyy+{$`3oZL7iYZ0G%ZDvl
zaG(B9&(`9vj{({8YMGW!9v;>nH=?I)mmRq;qw@dX&q!IcEg_P^#b;pk(GNTdxo(_Z
z3n96%ASx6r`I37`y>1yk3dysBgq$UhE^+W9X-$8i@2<ne1;)>BYqJA7FWzXpc!O+e
z!b8{2taGlR-+;LI;)q#aV8FtH{V>dJzOc)$x{CG!V*sS~_Nlu8N}^C4>+x^@tbky1
zb8Dd<N3>QzLC`U{O0Fa<8lSp<0X3rkq@dm`+vD3m<75RqI?MXJq?&y|SVvhzTqw$^
zpR_}(&~FYkJa~1i{&xTx+JMRkOCM6^YydN09r#f<k;Rv?hUBOHP{oR?QZbJEe~p-H
zGlCw4ohi9(i2|ksrHZOcE<aiqx4I9p{T)*6J$lzRG9ousCrSW<*N@3$iQummAJoJv
zdyW70q3dM+W0lb9=iYHw;*Y#LBh<}s`*uD4aKB5nv1|wTyTA7sIT$`*`mCC%F`oZM
zIq2@=+UfE3rTixwpXLsPf4kVM&{gw=DjMkz+cs*+iv^H87-<)%<o#U`R)C)WlGdL2
zF|aGR>%dRsW*N)y9<|zs!0?~$`E*a?1L5oRtA=pEG44z!raat_H9{85@B7q_^)>)`
zT#&Tn1j=P+`i{^{h#Nt_ATS2-fVlLna#~O7UyN9-qDPm_-Jb1)Ud*u!Fzf&18DcgS
zL=h+^wQOmD9DX>KHa9Yu{PrX%s`B3jW-h;DKCV4GQ}o5ikz#wWBsl#d2`YI*ZmX;^
zt@^jk3nH*mp`tN2_+-Ke?<dP|f7~NM)*1OK9#)AYw<TMae+&&nvXD#I_CQWsk!onN
ztI65Px{+o6wH}4hf4yj!nR{@F6@gdzzvnsgHqZ1g{rU&2gRKvQSA^C@)by!K6)Yj+
zorqlP-zw@8iT*LVT!J@C87Bjy@%<O^lA}b>GJ(p!_EG}S&R4R#M{GIv+9=pSqD8DX
z```9f*8J_!zbN&+!_oxG{<m#hS_Y8f?K{5`?6Q9s;cLmX2$W+Z3#{a8LZSWN5Fdkc
zHJjRVRZ9PB^CacK-5-Gxr28KRS2YOsq>@srZr*|OH`@OloJqestOD`_zg&_2WagMe
z`R8eY12&V`1M2?L8@rVDcbfSK>iZ-J35W<U3kioCKU%Q<J%}O6bnrH<R`~1~pwyBv
zcZIr?`(N8gk(#n(gnfxk+<z~x^`TordQ6B+-}Gw3z~cD%on^SN1IN5x<L|7kuBAA@
zk3Z+h5%2<OGYfvJ{hf}k#0<j1y3pSR@#aE69AV%8<WlMc#|?h&%Pej^;6|{lEgF7>
z>2@mX)8DzgLCz&`CrEz`Lf;{=bRB!U`s4q=dYNqh4)M+(z~ui{Z}!#yH_8D+4Eq<7
zim$9yd<^~%Mh)gkKXzv?DDrSY%n0PyKZtypr4UD^bZ1swCa*dZ;w~9W0-=z}b6%JI
zyGG?}CCJ<(F*yQARlkVAcZl0UM1lW6Q38nJ-LI2(-()t;xTvS;WJn$PQ1B0U`3T%2
zW3g&L9JJi;Ecya>Zyo;KB|wEr8ML2Y;QD_@c)3qiuILPTm9<pz&&~Xzs1YJF??#Vs
zSMb3?&q#JR`TiH{b^n0)1}fHqO!?|u@cOW;0vI6??EF1S;x8)jZAUJB`%6VvD+Q^c
z$GaZJCGzEeaewo!MqB+p8(m9)h4_K+>?@$<s&QMh_~ZYLuWUwT8SfR0Y@rU;b2Ph6
z9*<hHRQ>Nr$XbwBQe&KqNnWkcMBM#!p(<!mRX;EOYj{ddCCJrPC3eO%iKuAwzxV_O
z^DAZ2vrgv1{xPjWLHNRwFc#ZK4KZ{_NH}W!dsP%hmu+-s1$T|9-&48aU_K20zg0aC
zwniu|xw9h+<-bKdOfgCsf^L&fXy^PpIaoe%my(8{yG@KYDFyhhSG9n{J?y`UD@ox7
zY0JW_Sm<w<a~sBzFuI1f$+@QG{`Y;LT?&EWjqZHuT<v(V$-P8=I`!|VLQ3-`Br~(n
zdoOY}63r*whtBSRRKsbLI-`G{(N`q(fG{aus$z9lWb=5O+&bZfg`he$A1w;qzi@w{
z{Ut*y<+|X_z<rL2KoBUH3^^_O*QUSFYG9fsF#crVSwZ&r!K-{AXUXeO`r*ITg_{cN
z_4s<X5?i%yI#YYFGq*$R!RbGc!oeBRocH6UCsBV=T$le7MKxBg=yaQ5Bf%}6K*!H-
za<`u1=HUY0!EvydUMDs7mjqb71PB&_W>7R(NA5nzklL5sIsbAw5iZoCkktS-wbZP?
z;m0y`c{NmHm0}sRzPSVW&Yu-OhC0{QMAq<8RICB%fvEhEbx03v_JSiz#s|OM+Z6u{
zbVi-xiT_R?c4`ewnU7dZwatRfv5Kn_j6cw>rGZ*IYlsu$@qUt8n6T`&OWSQdOcHr&
zHRP-Fqwo|+HUK)Ld3-0hqXI_lDmV0ux~8%_q58(sUg$Lth8p<~4tGuY5a*dIDfYuP
zlvm~9UKHT+NdD)?j&h=<M@-|CWvwby&AGn)Ls}yP+FI&4d+-o#b<QDTDT=62Cf66Y
zsW~sD{(A>$AC<C?9z9FPFLyt#SMNEI_6t=8gYPf`pr0Uqx6Ee*a~()+l4D|0hl^Zc
z@Aur6EF56~vBj2oJ3A4st%0lGzm{u2t%BSQsotf@Y4fUfy`7Kaxcg=c-v}u?%4qn$
zUy@fUdWKy19|i3AdRepe{a`wHF_t`L-}2n~Y@>p#cL7MdCmNfg91?_YTG+C2MI8LC
zYB@G=FM`^a)(>;^Bn`SORZd*<#{2z}4;@bV`sMtk>%+Tu2&6%>*00V8&Unx0Q@qo9
zv{pdSSsjMu$jq?%hpS^=$Y&xvAR;4gm4!D_&CfHG6~#j?(EiVRs^m&5ML!c-^rKL-
zcJsyua;I7G&mTEV8A6;$BoGkh+_dIlg0htbY#x~K4IW+bPdth+&O-2xQ2}0A*CE+O
z3PxokN3aNlU0GdBs~vYvl=6u4S3NXvu9V({mqbBd?Lror*tQ@jSgHdzW?)GZN4|3M
zGRF?l;(1z?86|S8Q_6)_FVW6R;OrjqLm0QF*vvPJRbrE4qG}ASt{BnxR)mXa%naP=
zWSi$h($^ql)Y{~OhXflNzTzH(p`Mc>ewJyK-ymcx2C@PW^EiJ}gbA3oUx@Tie5YmM
z=BqDp%C=YvQ!kM(LG=KN-CMKcq6}Qah*-k(u?QCX7j~N?#()nmh6=XHP5y)<I1@&!
zFxH^=Fw@GLilnILv{PD46t{nn+99Tzo~N60=T*Mr;TG!?)uF2utrndPB_<f>@G|Pe
z9$U&DibM7VR@=x9EW?s2#3FZt(^<5@91J|x75=PWT&6$7+k7Mtm)@bEljzs$&PKV>
zCKx|9)!d#45T@lJ7+fHvstO2O+&{tK5NY%z2leC<V9*=}RFJ@W<^}?hV)pfNa!J)H
zyv(&>sy8g3AF0^ZP@JuJ=!>W4ysT)(N~DdfpMOW?R+CJ6Ad=qa;NBL`zO`HfJ|V%-
z2{XpW6e03~zZUrEjl1deeMRUc@+*b=|5Pd~6sy>JF7p}nSKmPrDqp84HI8#9m{|M{
zPZ*<Cvilg|9bIxnXIjPP3PST~6Mr|=mSEeAcLXWmf8D-;Y-;2ozq=HT->OykRagPR
z{2|E}6lyp<yF>D>9a|{6MEENY`|GOH%*!=i$a6yVR&`}gMNe{fr?FpSPAM_w8peb1
z7T_Lg%k<5lYvQGAZ6_{-;*CLB?QgEEn-V9`H`yCu*%O2gndd8@5v~wY(~GJz%E+;w
zjJ(*4)4lYDdwUOx%Q({8^x2V+?XIgX^%L%ZB^0cx0K84<T&~b_9MT68`1`hHn;?m!
zmEq>zHd3PNqLF~OAvvGr2}~M@>vo<o^qTfYBP8qoN?fKgpECnvBXQuqseu>a*QFJe
zFKeF-=dW3DL1#oIr;gY}nccv2+&T<N5@zc#KCu9<q3^s(@3345nrAvXvW!QW5}_!r
zTk&Q)?>Fpz=%3mnTJoIhl|oeMv*LcIf#T6lfK{@RF~u$S&kDxev<tVkOLKjg@xgw4
zywT`Y&Vs|l&(c?YWV58cOmEhEUrtJm)!h@;E+%$0<5NaLroLeepgKd`h$4@$B~wa{
z{36E_Y(Y(9nk)SgdL=P{y4v=KdQjtCH0EP0+>YqZY+ppCsJuI;t3x}K@V!m7b1xJ;
zzv368F1F=i`zm2>v7DrAWIQQ@J?xe#dn4Fo;otVV0?E$Q0Ta?^Q)?sgz7o56Kmx(N
zi|^59Dq)p}DiF<)U3C$oiko5-1>bi{g(no#@A>0oPb3h@5BHm3H|5cTb~{foAwL_l
zSWq5kLekWS_|0<)a4*AM{<!JyD|e-*=9)@`U$jyNy&Ye8CiHZnbxKPIHA1I`if1Q?
zb8&|@>;27W>0Ay0I$msfoUIHCt*^3=ST|R99jaf!{rew&tHllqdCnB(p6@E|$9S{x
z`CpkT^~r7=cOAq1h*W)h@iKd%(`=3zQ%jBnGc-H83!U;+^aY{_DEl!U!J_C;p(Cdn
ztiC*`#uLo3BXi5sL#UsP=#9^V$x}<2xMR8{PNx`VX60JF1*Qth1DVXwaP#oRrpIzY
z@rLB@8us6};;A_fj3`I=Q?Yf26jv94hY*N&_8vSg)Ygc5$0WXov~~al!oSW}Ul3Dw
zQ*z=hD>k;_>!VSCn`b|SXl73WTx-Ol&#<Zii97nBqkRZJK8o2p5;Mv(cT?A#TYbW+
z2ds_qq>y7VJ-duHX*-_}p|Z!MgV^zckKqLuu7EzI)8}w+O7nHLOMh{s#4W0-4y^C(
z-yghg97v(`6{sZIfcwG@sHnbZvU#~^ryDReF<u3SW4lj@<$6qDp+{ySuIWc6S{;k5
zDj;B_yvQ@ilPzGqcY;C+6zB&D9}s`aT(}fs7XMkU|9)T$0Mj4EpcM>mC*yhsS53DJ
zW1^`45u&Uq4sN=??T;9a>>@j%F<xg0NX6$9o%Fe}n<P)*ItFB~dxfag_?>Y+(gR%R
zAB?J#{S8zS*#2BB`J?Wrsji}LWnoZ8{VW{G?uz95)&%>`X>#i$0wRE1KUnT*e{YZf
z28!4*VxVG4C4fCUH+Ctap++pRJY(MJd0X(h);+9zu;KlA9#q|;jJy$5GCE_?9H|NO
z4GDkQ7cgg2ti@v8(02wXFZx|2q}Wy#eAn3ojR}>+94?)=Xj}FzC*pKy1}cAyL*e9d
zp!+HeI2^sd;9k%g=1*xb(uuM0A{zYeWT^=IxWPU6rdQSmcxQRNr2SCGi-9Yr1_SB}
zux-lE-(sGpwwVHowr)?PV{U_(+Y<Gv#y-|woC`u}x;AI59HpARnxFky(@(|)-3d77
zbn-A+WEn*lO0$4d-Q%Ri2D8GFq+w0(0UY#dZ`>?f40!D9r2SE@fiN22EKEFv0v0Xd
zjh`6&!49UH6WMt_Ezq?k1;TD?6H$_A!agz+N$9*nye8(aZG`9?;VQMvXYaGJlic6+
zGsTPqrW$HNJCx?Xi|T)oAipc#;u0vRTs&?s<uv<b_0V;2Xla&kt($HG&GGTb40}j9
z@=M4X<~P=;y?YQlj{ONXX_}us*<D{S^~(fQCIG%HkkaTqhy!>@7pBe+^w_c=2i-M(
z2jE?<i(`5cfxqr&jy0Zb<KzwuR;>YrBj02#Po{i4ZXGemYdJ%W`Vg&40z%3LGy~Zr
z3TZPE?Zm$4cgH+1z`aT&;nl%~rT%x6{@I(e@`#ShQuKStcCdH3=cv(7&U;OPZEX%&
zm4{`lR+?)>w;NVpdy_Ra5l)DqJlXI#Ec%c%5L>IAqFK9I6m@NAG-!TxfO`3k#Nc-a
z)e_N<7(m~h5YQfm-X-nlOY?1KSRqb%numNZOOK)jS*sB&VFllfxQ!Ztm3dqsq=4Ym
z&Rmy?dN#U5@$@y#Q|%Pa5Zox7=?mZjE?BMl1>SOUVb=p`4Del?0?V975c_dDp2U&2
zMT&Vlr&uc%qnh5Wy#8=9&AoA_a!-J%4=q3c`ryrOunkA^%-6hZQkq%eWbboN=TU10
zuUgEFUb00DOQT-q$7Hi!%{;MUy$9~1g6R($HnGdUKg@k^EQPckP7it*5G?*2KL4Q*
zt5~B31&z_WAY6|h*YS&!E_#lCBYe4gdaR?n;VqvmdfDD}_{+yRsAU&5lk%zJ&ZNm;
zZ@S?WdSk`@7)r6##L-?dm@@%EjCb|?5p#4?6|dDt&=R2oyn9tssa)NZsc8tfk!~pq
zSZ6^M)9t(aKp~k|*yCx0ae+10fpDX<p<{kQZgjy_CQ2eF{X%|({*wU~-!p!>dwQJ-
zVl(~B{rBlG*<G#|m!~b<bE$y0)}4``mMVERsJ51YYfuJQw3>qDuBI5(%BbHV_RaZu
z77#XN<1oRL_%3577!on|yRiPeIr~#}c)W>@#g6I$FU$0rmk!cwj;TSMCn3EHPX%@$
zwi>Q?;Nbgk!kjr1rMloE#}Hq)C}E<^ExB+#78(|Czw9)^AGU}(cxcqm&X*F|Zg`C~
z#DulqR#Z-yXr^u?MCCbZRqBRtYq3Ku`xMf+h#7P@QZCR|C-ceuNVZV7%IWT|=>Dhr
zi*0>D#G)0I%FV?y5;jhV?TV4a$8EvwV@0OSx@R)P)PU%g-JI#QD`&>AqCI}|F?(1j
zIyu;x={o(~jU=G^*rt;pU9W<(mCdnuqV;_Hv-+XA{%`BLr1vH}17bAhLm-VrkJA8e
zJF!pWwA+Dkgfzd0`yQTwcyKO4H-zfX2HDzU4sHD$r|dQ{_k2<Y45Bdi7;&X8BJ`?H
zv@N{8b6|vj%--;OaN!KDlzrx9q__bc(Drya;)lVvc+voti4yC{d_<}@^PxmunvJ1R
zEGN2$8JaXnV{iK|@ElphB_J<lsp%&(3pTRwZO2+QZ5{5|BFEl!L7xWkwj%p)=n|Xp
zxYjArM>~a36DC4)z7C6UCRIRI8Q~aL+{%V7JsyH~sRLbC0!U`NrX1j2Mshlx{eG(Q
zCu7Q;g_P5KoIl1ROKP2CFB*F>)}GKhfuh*kmSRq?3tH!x=Dl1j`)MR?jvGIzp`4`J
zm5w_}`A$PwpGI`SAl)M(7t@<QNFLH{|L+HyqafvSDLWG0mP0inuPJ`2PNgSR>;?gl
z9M9O>NKu3&TI-ww5MO9P5$?hCX1ZNluF&mQS?pin*iHri%Sn8pBi^+ND3e~0r@hLc
z=MzJq%;2D(X09}Xx`#6fSmz08W5a5BlQNwl|0(b^unGGF^)1^4)<+=hOFFh1VzK|H
z&!?!@4ZE#$U;U_?2l(7gkpx8V(H2eFUI#6jozPABaj@OG!-OVqu4YbCw@42se2bAb
z-9+8fPw7{HSX_CIh}Cbcz2|vm_;bepR35U<Ftd536rmi{KFb}wxr2~v#gIiBDjVGt
zd!_T)+QF3jbLi5kB5OEVAlS3BL<^q<GlbhMQaq)<sS)dp_1G_Ve-e2oH(SI7wUeWH
zwUB@COv1O7iEEfXLJEWvA<v7PA-p*Q6pOhk6nJnOK#>zRZqGG0voGiB{yivNld<(k
zgYihl@_oMV`1MPL&7Gu^$+bJQxg)@ZEgAnRUZMmy*ApXQM(4bZMZLKDEBbY0rS$*;
zkDdz~g7kF8Vr>DF+D~!Cf?P+WcKtjjq_)dZt+e0L$Z-5-Yq(e$!fxK`(Zv-fakwg<
z(cFIWqMv(yd}g%-Y6BNeYr1V<)i70DFC>M@D-()BlFqwN35lRe4_BZNvk~T;1KZG2
zXZTahY<(r>_rf}Xuhgss{uiD4Gf<A|adlnemE-em&;j3Vf<4(R%T_%He}<PI)`IBB
z*Y&goF@p4HMeDgr^<yezQ{(|-{6Wy`lW1HB#8Eo5AuwFeDfm*VFUdp5uTShKt_b;;
z9=WInX72CTnsXQdxw`f)hAJ&38x}h@Y%Bnd<%gyPj@ZTDWjsyL=546u(9b-H0-nm`
zlS`{Xqs*)m=K0rTY92)^zbI1td3IkBq_ahvWwVZr+psgvzXsNX19yaG^bjTbxpP|V
z%UB4d)BG5CX%_NLiZhU354eI*+^>*wX`Uq@HgRj>(>h=SFETq4m%(kp2=e_p)QQVu
z`E7F6Bmqri(`S}DEzi%?cW74}$wL*el8gqr0KYlL-G(<z6PvHHf2gyI4JJndF;3@Z
z2-!zYBNPzN6v>$S!CNJqZL;*PoTx>px_(Cu{Hvc+*U7iZb18$wUeu7i9gLo_io@8+
z9pc%4Z)(^0i98^NUj)PA(`Dh2Wd%NnSO&xnuIV8pB7YTt(+stHs4wdb`{7=rd0A|-
z+uw*}xD^8;`bYpE0{?i?y(b!@W8g3c;F%lT)-Q=cR>v^8y~SE}rPSMqGoKHW&>->W
zTE$UBe%wsUI)M`x*lKbNRu3XL2my*>PwejF>s5Dm4l03?oYyi5?LjVEj}9TuRFPx$
z@HkE6mk(zpS?IFjq$8;i7wop9x@zky<PIz3L?p*g@^E{eL156m#_bp1(=Kb|E>2Wu
zctqxN4n~&_kj_i%Q{y{53IW=w54eduae9IJHR#Jaq(p-Y%M%|mGFj}nV2X~DJ%qqU
z$^mA^Q7OBRAcxY^&*C_DxnolTQrnydd=>t+9GS}Ek{}6kOOYL@Fv5jOTMy*{pBBP0
z@lE-IM<ARlL9eek_%=T%GHByY<2!t1>O4|Ct>uW!uGlZ^=)2c&YX`k$0_;DNNwV7Z
z8$R}QQNoFpMFFE9V|1u&u3dAe@2$t|al|SPH~`tP4naTmO;~O2QK6N-wC^6$wgz9t
zeG>obI9~+Hz}lINAqyEg>B2Qi6~j8lnxtn;q0~zrTTu-9fHtVE3ZqgMpABtSe4Qg%
zx!({Kw<kJ18uU})^*VSbDcghdt5C}DLR_%oYnEaDagM5WvcoiY3q@NCTtdV}tkHNd
zew>7h3D#oSWD|wsuD0b{F)EtKLA*iF?#`wfkIaW<bNY|KE5d;y7u~TZ4nVvDE|W<+
z*}yX8(SZ1m$9mtVxkAzykMZnb4{r5w0x&dT2*AGoZkQ7>1@RQ|6an+42#8B^+a#h5
zFQD%(iKU(8bK}-X<=yZ@S^4+CvK2nCHY<2P<cQhSKV<KW+PSrINiBEU#9U^o<+p)>
zxL)E_Tt)0D;-F?~GsdQ-A=R?pn0kVNqUAu6-CG?$!@Y02pC6gs(`jfg>{c$PDkC_g
zs0%oK8IMcTHx1q$RqA_Se>nS^i|rRdyybB0hnYBn)|E!J*IRhWlM~CJ;1RRcA5Es?
z6-uemH;d4OCD&1caCRA!(*T@ZEr=p4ieXhHwU-+#PF5YJNmA3}u=5wde}o(q+@%h&
z1C0UMy%&@`ry40Zer=HC7EjNobT2<)yv*RH`t>)ZSJAjd(!jSCH|i1vz{Ay@CQ`C_
zwTF^g-Qb%6`31gq#NOW$GD3|nca@5Q?%#4a#1`vxEcQ)J@Ka{D2dgFng%hV%qfqE?
z1PB&oh6`bVJ<*^3$ldnV%&X`%w(e+IvhAH_@j&#3&}zbezT_V;DIc>O-6frK&Me!#
zdWUztMzZ2qqo5p}*o1-JWohkzarAl6v1>j$j}@a=W>|m2g#85VC;~9h)m2Ev$QvL0
z$T#r7ro$hIvD&!Z+056Tc|V4zKX@hYXs?!UW#EDh$^Y`BX!T9v`jtZA{Jt&C>2xxj
zq@F%MeBF@ap}F%I2G1!5t-~o72W_GWd7-+ilWlE+^t~Z|<4KkZPpt!)jlmhKD!C)8
zE*z#5ZR8sff^n{gH5}ts54)Js<8X`9et14{eqwC}I`zJaK2H{Jaru1YcA?`?*!rG)
zIq@p%GF1L*3j9?3(SQ7I*S*vZ6efFv%UH{0>V><B!){}?7(F$`rW5tv6gPoc%on$4
zOPnA*L&1{yDul5FE35$zLtMAtfVL^)>9I(j)gh=JGa_d?bw2d096uwsz2m`o?Aw*V
zacnppC0g4%IYpAI;qC;;bhc0C8)#N{_M81pCzQ{Ii-dyXgKk#$Kdu+o7hEB0mm2Zh
zxZa@QI3z>Fk>?yI%&hAgaNGg9DI^|0x3(4kD+w5$t%37m2N~0FkhM54@t}wc7BDE&
zo22d+$-NMGcxgnicvx!p<nhNdcD6y|3BC7u7b%IuF89j>{HbUvqLaz4jM!gKxcTIc
zYW#$<cS|l2RLBsVr*;3-rDd>XG_%`?{NSr}`GM?hy>I~}r+^g8S0p<I#X41gByKF0
zX5*CQarXX5s}8L)==t(%DM;et)437TN_V>nriM6>J+a<gR~4X|*@1Jts5lng8L|*)
zx<eTy^*&T5InOj+cAJhM-J9`S{Bfi*u~@&u9t8rYJK9`!K5J!pBisA)9|!L{(2tI{
zNCc<8c^K&5VY-nQ4nwW1Ro)XsmaU{2zpdOKj5xw%zj=OnbEb~Kp8A@la5q``c$N@r
zxign2?y6Mwz8Lw}<zu&}`*Imh`w8Kco%u0(oE{hRI#Ce-7J}Pq98omjNuA-{HWAfg
z6ncu7HfApq^fR$CQh$%|^AGanAjAIrH!3rNIm!lG*WKX{FAcG3a()wgueonzEAn1G
zWKTy&oer<YxjtWoQj$VT_Jj4apWEZFbx9Nk<{9@q?r)f6nf1DO&oQe-MQ22PmgO8u
zA4*1#;?HbMqTNPY7~cyp4gSE#pHC4e(n!6=%Q%N_Wo1>pEf(M98K_B|kZzlo@wc~-
zqjPD{ir*YRf7f!D*GmUa-Z9SftCTZi&z%F39WM6{71SE?dqm>&AjjuFVu^=Ms>1h;
z*Ut}ZY*qHy-nE}0nO-!I-t4?0osiY-^$KRQ<|{W;&fgjITFjaJhQX<#c5;-eXz7TI
zgInDp^O74|J+NyE1h!a>rJ+NWGtTZDN<Gcl`M9KYBHhEU|6>Q=rF8CB1}G6)SRfa<
z+o0jQ#lXmWi}-M7=vI!Yfe#k<`z4&UdCnX~regRBSMHQLyi&5`kYHW9?{;Cd^6+l0
zQ`#<g{jF-{ydM7P8dVd^7n^dYHOL@NVt*g!g@nIHth>px?94O7?V;wa9*nGK8KL2l
zt4|3-8&;9~$zx!lQzKQpm}Syky?qf2$rsC5C=5QYo-Z@w6GhJxXK$Z%6<(_$3(v{t
zbShVT_lqN(f8o5T=F*R-Z{k{hn~YB_#k9Sh;<ufhZ<`Js+N)a0=@r%r@qguZuGa=h
z`vtn2oAB(M6BYV0s0`<k*ygu7vYcu7C}0nZOKG_0jF0Qj5Zxy-=@($i0*u^LqHhbY
zdw=6PdwEnyD%m^u;jl?PAYd%834Oe$J?522-maSO<;Nxyqlns<@a_g2@TLa8hWfZ}
zvOKBcz5j$WzoA1bbevRnoJo>Bb*&vDKRNGjY?3QtJ|~g8=T^H;IKtH$*l$SkUbD2r
z<MGeK^9Cm9;?Gh6f<#E#&{6z3b)UG1JP;!A<z|Nm<X2Mg4cIrx*LMW>jAft7SiAN^
zpNO+=s@?>K10vvlmo-0nl8XM?*SU#{%eC8@{Z39%{ONN_`oRa-PG}wf{U}0mW(&Gd
zmTKM1cCW-Iwi>vP({bWAR5&#wNE78dF%KEgV3H2JGd>^=h+OXtpvOh#YO=qJj{hyy
z=|FY3_1rF~a|?x%)4HI#r(am~QNMO)AV@a=x1PH$j4qMEUGIgH-y-22HGY5E6gmb^
zAnLgcCg-+Dn+_0<bQ(>jMtVQxS;;E<;1?&m5m(gkr`%9iu_AdrpYxI;WB`2VMSi{5
z`ayYo?!?9CeD(49Gxy`Erv^46NxXDha5d#Fo9U5_dq?t{tYp9Eu>nK9=BgC&j#ZF|
zBZ7MVuApV8>}Zj`E0*rW#k1>4^jJZrP{B7`2~w5rYz^}SyIUI`kMe0CqgK#LQ#7$7
z_HU1c$-c+<&h+GLmc(?_7#CPSUWll|>u0hiB&V*diD8VS{qD4H{$14QN4S0db2rE+
z{SektDqK+~Bq&TSPETTfkE~g@Z*#H!PhEzP@uRub3Lz^Kz=ppa^GMO(Lyi3b(HFcL
zq(rH9niEsR5Wu&9I0B+u`Rc?>ErGT0y6e8bYpv&Q_t^Jd_Z@V8LG9v{5dPLlp%V}c
zE6fNYVH0Pg5442lGXB(hQk!}wIx*bEK;-J;K?oku@KS?syT-TZ`mOxGx12x6;~>oi
zA6a*S30%k!-7LP}zNH;b(I9nwT}`%AnESws|30>wE}GZ@OZzMFwz9Ex>jY|M&oYZa
zM-&gkN{0uLvjpCNP)D#6@8#I7QrjrBiyy>xqq32a7UxKiP7>6Nc4~e^kWdKl!s_OI
zVTQ<V!EKMyFk{pc&n+-~ZNtfSOfk1(gK8C=ji_aI4x*p`m2%P8bstjWRp}(Te#ada
zC|O|3b3gX+b%(c6**lUpcRaB>!)%G@&1o&wxfc=M(Z0G~qfdj)i(dTN>DjPGUVw#N
z5Px+M$~j6f6;>^EwLs4jX{$hwPnXCGJ}=qC$I?-1Wl_DaIwM*8J2Q=$Vf>A#V$D^`
z^|z?i>$sqpan|sht~!g6<22spn9bfm)syd)1f4!{EMT@XAZHlTr&07n7r=MNg8_y4
zb?`Mkb<LHgqgs2%h0TX2(7+1yK^u)MMWloC1nFir>(O{r$j{m}(#G2wtUy_(ZR|8X
zRiswC$aR=QkVDN`Xnz>?M-el!Y){(eFq@H=y%1Ql0N#nz8@t=RJ~BJk=6GSvG5uST
zd?ybE)<Ee-Tk!)?Oihtvp5omf+BHlA_Mt<%Xa+ZJc>Wnmr9wg3gJyfH-SC)@UCm2y
zMJgYPlMsv$KzLIC@)dE&EG9oyKC`($RR(prupY<G>&YIyC+q-Zk}OkZN`G$Z2fR4N
zGHs8z-J0IMY|mLD9wxXh!J0ExtQ=lNofRntXK3Z2Ivb2dzFPe~4QKl-Z%|38_jzn(
zBR#t)73{VoMk!G>FptDTu!-+RXBdB5kl5<f1LR_Vk6c`5cXEAm!~NERuTzVe;zU=A
zTkTCV<E(N4O%8G$$%($>={Z5lQd>;Q<Gdt2j*Tx_G5~hGefF0YTsZubUvqQ}YuUa2
zBz{c!15DVV9osFE&Q1Jlt*&dc<&gojL|r^ez3}5%ou}la#8x7z{Tpsez|qg%RX>M9
z0)$@}ma9m#H5crlic%<J#%=dV`-1ukffA4LN_1qFr0-S+@P&W?nwBNFZC5m&XKoh3
zcK;-Engn+z&u@S#d7XCNjpA5Ef@xlepjY<|9W$3+6SD*+<B^i@G(veMp7dkc`A}l!
zJrUJwwihBAD$3V9&SWnd=;qxFX*)6}cow|k=Z2j^pF)_sNbTWH20=NELxLOa>zbFH
zKeQy2^xRx`Fq|0HkX%#-p^)7TSY>Euzd)Xqh4a|#@~lxue{uZ>9O#13#wO^bTMu72
zsbS`+PL0(BMwLkYeu=I@eXykS+!d}ipmi+1;|SckxsqoE6by_Fqz>^vMHzlnj`2hF
z%9Zw{9=NDHMb^~x{mATWcNT3lSGZB1^mfoa7pAr>yKnITN};H6c?pG~Y?B8{pB&MW
zyrvK})=MX<6e>nKT+M4Ov&Nbn(QyO|>#O#T28K2a#I%}T<TC?A@lH^E(UR7`G=z!z
zr6#m-pfqtvvuy^FyBhf|>@{0lPz0^G!WrdZ{u5+PTy3G42r&%G;MDbR4KuMp(9l3l
zQD&E=zPpwD$M3GbD^s=a+d5o}r`V|`%(<`e&hW0&#j6d&T+54-gaJ+B2}u}G8vN)}
z6noO62MFiFop;wO5y=k8LUwR8ev0`!40a%>=K?1?3%5<O(K~Y)62N4sctK0e1u&@4
zW>+lqtM(yUSC(?LQ3#Mb)%r=pa#DL20l8@f<&zG&zN!Y?3C=xHoM#s`T+&#DOqPZK
z*a7|b?+hd@KS7!dqjs=m*Sao5YI-Ez*&P@A3C~nT%mrgIeyc(e&iqPq^pIxQYnOm=
zyWbzzUFp{5h6u4Zjtp8CO&tLWC#4&4C$R@e91#Q8t<cXrH>h8(kex|ixE|}1D$j`7
z!tU*tj6LOu5xLBF{9H<JK)&pcx~cHoRN1>LDZT|hnYK|sm)P$1E;w}|zg_8E_A#D*
z-Ypmh3$D|N;Qn+k(sv;WwvVpK`0Y($F^-oYMJv;dvk?tVPesT9`+=*`?AE(S*t&k`
zJ8>=tUgQT#2r-S60NBH2cab)3Lo6y!tnbc(tg4ba_18y(xcv}PgID1hbVI5ae^pWh
z{xhi6lrFJ)raO9cUyb|oZw1in`skoqU~btZ$pq1~G_uNI>3=0!mG?0xnV)xRoPCN#
z7QnOc6i}CF=7uXWqW$MV{_oQwMgJ<?Hds|f9L|ousE?eaoOF$j@ry+dL1aG0B6<}S
zDWZn{73#@mJnbPi{VVTL=h@3=?;AByg=Ztzbw@BFj|`en`rK3YLdR{zAL`Z1GKY-U
zlea}}vCMIEEB2}V+y8dM^zDiZbEEIb7_RK3AF9w-1ny~-G&3TIED`u|Pv9&ZtZ!cX
z?ZpR?NRw&lIx;4%zBS?cM`;Gsr;dbs;YZKs>p?dCCodxJ6*m%hO>(a{q;0PqRJe{K
zb7t|S_w9n(V=boY@1tJ*MG2#8s$?`1<Azt4wCO!4N{-1&s(NSfELmnxQ%%bD2`pO8
z7|`5d{<mp9LN(HPsyNan?5iP<1i^J(Xn#-&%UU~8sVAqqu(N|gA<Y~A8!!Lyf5FNX
zjoh7#c<fa={6OX6ql=Iz1x%R>RoKb4Grd*|vgySixbd9fkDu22Pe^*BT5E-!K*?>&
zq~|_Og%6A<$2Zh~h0cY=v*}(RLJY(5;?yp^P?S078;KsY{;nHbSjtvl8h{&f^ybxS
zFFYs9pkcyhfUKLcIT1A<XuYMZFU<3ON^mfZ?aE#2-}WY|`B2jk8DA2kn~^`-OOAx<
zQmcJsj{C{-wrZ<U^{HXnK1s5>gZCr<f%9zTciBSANx#i+&Ttwho28bifqg`v{^54Y
zXHL}rhMhq!6XU{Gq0pq{3H&ccXMv?D>j=HRo$8!u{HobBAj3K~?$-&m_O>i8tAA6r
zEtuiv$31vTRZ`o!*QoypDmw~UHS0?ul;E9&wKqvdL%9bMvPEkt=)kXa5o##wi+d1P
zIOC~ciHqpVg@@U6l_>rMMy)cg`bLyAJ>TzE5xl2^_-mZpOFWvp8}T#A!0M`+%fn)o
zkWmIa*SZSjD6M77A!*U?J1(7<u`6Skl7W8rVdQbzDP<LHM}`OT9=W&F>usF;HSS#l
zz?W%x5qf^8G5F?d-@0+JMdJ+%40N*}Im^t>1~VjQH~z#!PjGd=jqrLK#8m1D<S$uu
zn@Ii6;&QcEdX!v&v!p4rj^{Z%Kc{%?){gi(ug|B$)GZp0u0?|L8ZL=U9IEp>INIvy
zY8E1FV-?eBhJd4%qvUR*DB9^k;-_bl<uiUB6rwK}7o3WR5Z+TfwS2_rf+s7?z31P=
zaqwip4S_h)=QO<%0S8k-VPYd&ON8Ovuz3b?hTjHHSTtlR`U%`TFsl2mT@Nlx6m#M#
zKf?0BvYa1OM@ASRt~&4dbxSwysq+AA$LxxnZWwSA3mNoDGp>w1XWPhhUgG#kK?TD%
z`xCgdKU%_>iS{L+>1FzA(C0^-9?mX(Q^Z_y3y{Qqz?^V47(1_^oKmUk16hT7YE~4x
zI5Hsx<r)~><;eS8?x)aa6M9i%zGq@{`^7`AAFrv-D?Y_WaMwV6*%J-A%Xi%T>)T{`
z+hzJM=jZ>bJXeCvL<a@*Re2v&-fOAdb<Fpr9P?Z{Nwrm7-{a^7s5_$G5rW;d1m)Tu
zYijw-J&+cWEKoMONd_w)s{&DAM{~vq<@pHC-g=nTrM+KIc}nh+hGu+a{lrq-5nq-P
z9`z^s=*f8&rjx9rFR+A6aec3Zp4JJ950^>9eKM|q7tXBQ%wH6{p4m?3I;I}^60RvF
zjyA8;7p5gwR$hs{%A?jTu@fde;^OvTVMN#(SX!G_VCGd!7znrqrEwk7uq8Mk3&fjN
zq;AIgEqb4y4%c_R2aT)~5O%&0YB5)PUE?kK-5Iw!FWv;ZubS&D<n(rQbPdIxth>V~
z4xUn+(3@9TB|D!$!a*>P2<^z^w%^#dX84Vmmf1qM;UzuOAUklcM|oO|dWGMVeN%*-
z=wmeXA$U-uD}(Nfaj;h-oJLE~k%p-#R1Nc;KLpu<dGJeT4aUSS#pQRd-8~bDVY1rN
z;w}0jf5=SU(D3`?MIN#Dt$nW6lD)U(S>B_nB^wS2xK&N-i5gU_^&X6fl90^yxZyBq
z^{_F0Bye<jaXtPOA3XdwU(W^VUoXB}XH{v4i7vTAw5&8(f1g(g)r+`AQ<pz;RA<sP
zK``;UZ3l>blH(s}ovKMB*}%T@w;a$i8Wrq5A&+Qi)FF|*$ok;mM^l1&L07WJQF8L6
z(n%5s7x~Mu2XLjd=ku=%SFYW{Q$sN>h9&IZ>o?ao^;vm)gZ^rh=2Bbx;F^UL&s88L
zUd`ks<9^*J^B<hYql&wTWtPqsbF)9kl&c>NZu?_-3c2XZ4SV|DHc6L)GWyN}yS_0^
z&_M2<VtFQy4eVBsw2gLGsOO7cQY9UC9b$<Y=&eg^^x#ea_Ttr2=1lwu@|aGl5bo9Y
z(uDV+LmfiR1D(O*Y>Z!B*sm8YJBh);DVd0V3@+Ea9!1mcb~=HvUu{;PD=~DN@1DiT
z4lRbatEd@1s-kufc27?G(hyZ_Gd#?S(%qOv`C7BxOAN|2taS|Nyo2zN#v3f0=6_Sq
zW|)MzmbW99ylQ6p{>4+g$3Hq4xgy}8GU9FAMps+LoZE(6S_ct0@|6MY+;u{C=bh!y
zPU3!i-wYyt2a%-W>I?JE7+6TQ-95)I@;G&#)Fjxo&y_q5JaW}ljb1Zk@#Hh_i&dQ;
zaE>3hb?|x$oV$;}-|yA@>xSxQL~Z;@A5ZW*b$#E)9aggE@J#O=^LVXv&sj#lN!W61
z1R5}H=202swcWo0o6zThe5r?=Mkx8JQoP;^lzlQ5UAdpjKuA2pdJ=QmOy}!;*@TN5
zq%T$5IVkd;y1FxJ41Lj3B<}n@!pCqn28YLop4LPOVl6%txN8;t6+i97uKVO+M>n`6
z;NT)0v7QNbp|T-d?ey6XsevuG!4U1{)siM|b7LXRXa-6CGYu#t2sbco$LRVCvAGTK
zj6FpV4~q7do6VVx#?*?FkKzeZ8Xx0EYyCbQ(Sp6Yu>kLxt%reW%@&yLDOi!LcQo(y
zwk3theoH?-6irXZx_g^LX4}J>zbVvKm|Gn0rNm?Lj?GM?n1VqsRkvSi?Yt>sq|&+N
zhg%zi!5<`2g}mOwF4c{<<N+*+t5|lOMm^Xm=Esz>#xi$kcB!a*YF;_CZ6SKDzWSm$
z{w#pQpM!U2={}KfjOO3$7C3#Jz<PPT%*)=H-}&pCu@b%Oi!+j3k&UTa=0x#q%0~{H
z>Sw|!-6^<3_WRF+4!sI!fX^apzHN%?H@d0KVGvzP;L@PHhsnB~PvT|bI7=vn{o3Tq
z#cUr+ueNdnCf+?!Ov$Dg!)cW~v5UV9)E>?zzKaTWqTiw2*0KeHFeR<ND?h3VRWCe7
zdTI1m(Z+v_mq1g?VfN6d-cBXGa$ZBjG*4R7QU9>9et*kd4@I}?e|w~yp_;S8pZ<;_
z8h5%;JOi#YRjAYF6(z9Ok}S-kX-@8JPqc%H;?1R8zh+i7Db8t^DeyCLfLF_8sA22D
zpc(u4Ht@O{IU-~9Lh>iritz=Jyo%oqKD&+c-u93sIR=X7;YYM$Yzc;4i)i(H3i>j<
zlB#9yvNj4|^O2N`6MHaEAkG!uw={yKHMNX|V)}(21%A&xmjR0={U+<^T?izfU9WDv
zV-7w>%i0xAQzQE_RpMN-*?^24lshT;3obh4nj}J!=6eoBs-9nm5Ow98=6hW-FtHP8
z$^ZT~@asa`wx@n1?wW@N`Y7bEad{u;s?uUHI~kzsjH^=jMFX1J^CJqkQ<$y=LEZjW
zhodjI*onlSr@jvcT1H-2q*Ud+FJBE5;hLCdM{M}k20gE*kvv?Ese1#{Y}ZgX(P#+e
zSb67?Y<^<{sbGNZ!AX(HBV{y-m`C856j~w8Fx=aHETEbo>MFKca(pX7MHe0biMO&<
z^`g<{8<3~CuLxgX4aSxh&01uk@3hid2_}-Dy6`)+XiG1qUnvV<2UjlNCYRcI+)61o
z9L4F~v7PCJ|9IXn=y19TdA@icXl+y+%OLCqahRn}+Arl4hm8b8;Jn2}YMvPc`+T`6
zYGs(z94-|#LsSa1+7WRnMZLe4FNEusdF*7)Gy(C-5B;c3cX?M786W(22h<`HUNNvP
z3v9j%3-JaW@_qr1lIy#lG3fRjT1cEThkAa<g6T3E@TJEETPX2wI@p)v-2K}3JsQSl
z<g>f=vGOm%u{c2u6aL~$GAAltcZ(U-9N0gJL1xW9d9mBm{B)foSh!vBFsllf>w^FV
zE_v`W=qHu7GUofegq_`;kFCz;wR0xQBn_tB(SCmBT5#;|Vj8@S7O0*w{@f=XFMA1j
z$PzkhV`0$1-S-*U6V;QZS#&LdyuPu|xp95_TJ~wF-eBtlL}+xFQ@S|!Xh5qr3EEfb
zd+MuzE&_V_*{D1tk08}`zlr@fe8O<6WRB*0s1Pq35m1`P83@KZuG`W#MES1QVj|X8
zO!(gGz_TG^_L2P%`BTu6=Ff+=P`Ic;gQ-#TJ!=*bKNRv;L--@cfsW>Z=S&0bM-6kF
z;=J4ww6#6DX3i;aKr%pvo-Y}$OPXnnHnthG8O@DLl_`~}$uUx!?Nqf}$~o-5pwmNe
zx^pB`=8HTMhZj6~KL;p^dmfgN<gaIOG+}BGnmtQM1jH2WalbK`<^UfKAHbP1nE9I@
zDsXkVY2e<~F>!zl=QSdyMCs@}eb<+bXB#GR-CY-3hcwh}n1txHcE(^S<fIhgZ1mLa
z)rrP1N4Bsi0mo~xfB@%!Ktn0t16^Gvc!?4Po0<N{d7Dq;HIEz6R8!&1B$#}f<kwsJ
z$X#MMX-ZV}5VRkpzh_u<8s@=muIT5xBBKd?7Fu;xVll@-R*Gb3!L3=iWaRDKiMaap
z1Wgh_H9mn$2q0m4*fi4G^nM5n5VLTIMEKq#(p~D-Y3+>4Dikv;LiPqDK{3ufQ_f)P
zLr#DC3z||S3nny4no<mvkcYW5i`)<eR1dat|1{ZM)g-O^SQAq#BvYdsJXoO_yee$?
z`OoECDTEq`v{S`T!}W}*`^G#20-*5YwyVK@>>+Wso_?g2=zqQeb^#I;)9(0))FMf$
zR$tsU=|xNb2p7sx=Xj)5THcnI3eeA2uR>AH<9iZwPcOKiZ8(Dq-Zla>AJg;eU)ELG
zJvR~%q2|SR?dOONO)Z`A%su`3lwBg%Mj=Em{skTcaMM4@ZwQ626X6A1h2BNJp{YNn
zw1a_;lh@d#0L`7fb8?MNHUgYI3+P=BOexL)<{S3Nv1ak^9z9oJ+FM+?^3Gq>+38bz
z99pT1*ml7fN-O9t)l~~tkCAXXV<Xh8W;H@;H12@JAwk(FZYWQ(RJS&^=0@Ov_ot$@
zXNf{q#re<rW^1qOmuAWpb)Xh9A+(pr;&t|5xPl@@UeVfPpJc1HY&l#X{k%6*)t9!C
zepAK;yXsA?_#>qU_)VSKQ+|I{`@WehryJ~X*R8Lru;q+}ecknbB7zdWG%;MTs=No|
z_uZ@Ep6JVad-8r#&sf<<7+>B)fL&aXgC;M@6Z2b5u}lC<cZ+uyImy2FEBC3k)UnPa
z-F**;LecO-Rb49=>|@$M%r#q6wLgLmzfwN-TW}dRdg}_opu_`D!Lm0#NgO6{X}R4)
zi$EPw1#YyGr;hijM&)_Dmj`ZcjJ~Ib4OV=@&<e5tvt*<k;E#9B_@`Jwbl3b0f~i5X
za4tfvQ5625%i$~ZxZfXN4FT&4p9D=>Y1T(I4cVj2!sgsgmWPnKP0}?O_;bAcNJ4x^
zc}?|w5hITz6s$Xl`UkOY?_A7Bh(Wjk;QDcYwkLS0Y`dUj^x!qTu*>|bJbRKj>gY4F
z0lR3T?P7{y@6X44viB4;7ScXgI_GTFGpQZiKjqEsKg;emdKoK6MlSju@*Xl=YGiu(
zIc?lX9y-GdNr^dJ#3L?neloWk1<>%1)1|Kg2|G#eq=p#?llsh2HFBC?46B{{3l-8U
z`-x600?c}2TKAy%bOnl*Ph{pl+PX{p@)_yOt=OABPICz#`@qZh;MrItDC$YTIopFZ
zC!=-eb72_aVXSMV>#Jyh-5^vUx51&rpRFx4rW})T;R#McBN>P#FeK$&y`(i6<j%Ms
zV`onu;_s;TL6eL$3c2R*dX9%n67I|S;V*jBmJ^`O;GMBC7WGzg_v7Ga9Zq;QRKYU4
z_ffXkPDZsWu&ur0FS{w3)Uw;_+u*CD;v@*L?MmBHc%3tPhM$UhP+-N=!E5q<f<7yo
z=nqD~WKV?@rJ64)DJcI>+$SveMbWdPetm(LHQ$9&v}(?2)aO^mSY$YN{|3H%#H(sH
z#D5O5Da=)1mJ_6LzC|UPFZ}~9z7AV5onSN<HP7%1(rG&G76?}BJp2;*zyw<8sIB^P
z5eLP;?RAbe(m%*-;K0nmbRN}VKh9LkhXy3D1yPNbEPI&-9(>~(9DrEurk`Fz@44%e
z2**(^7UxYI+DiTB2!>Pm&st898?Z|Hko(^9YSILc_<gEG&-?ZGV5L<jxEf+cQEsSb
zt!Ni)IF^sQn>h3q&C&4^7OP@Lva9pVx`DnN&duHJq|s1pO}{3fZ@uBT^bX&S&V?E=
z6{O~WnU{F>pjZ?Rgp;|C2%bf$;kMYM6nTQKj%3kk8@%Z;CoOnhO)-03K*(}dIvajq
zf4R)EXv!^I@cN|I<r5RFz7Fa=a`W6M-2JxuJ9Y*6hzo3G^+h{$NbRh}xQH`@-n$#z
zVPEcyWXdZ}y6u<K>611W)_J4IMwx9C{ka*kr(|vA_xWouJ;g9+2H&a&4Xa(NGXlMd
z-dcb&U;G*|W|hLLe#~P5BTbS15&#cKkknfV8tA`Awj9;N%PAVKpk^aQGdA%oiOtxF
zn0{E5&J8aOC_qK+=koo|YMPFY5S1DiKKFD9z8ddV5S4F(5Br2KEDv=rHVMG~iEi_<
zy0;<p>)`m?V(isj)Gx&H*Pni$i~Ee1oV%^4Rh|kn-5bl5I5~I)l*20*hWL#Xq#A;2
zjk3b%1%mgz;_k<QKT-f!{%44IANDhJT&x{Lj(gEWk+sc%qmy;T)Hd$lua8KAazmYn
zMn5(j&mZ&t+z+^bYD8qJeaKS)K|E;%Qd^FMtwM+2f&K`0J*U(z6qKew;d+!pPObGV
zLAK}18cGQV#Dk)fw?;+T@TN01r}T;5B!s^8dm;V=x0LmG9<~V$1D)L8U9oy*l9qO&
zFBea|E*K+sQ#vqFHwO=_dmfyLgR`or@)~REK5g$+2iH3ae?3iuq0Bwa*r-r{XS<;)
z4nZKepv&E!sWN>h$SZCTd6(E&Y|Jv>#FbK@tblN~ZvS~ll^~tceOjR0%@vsokgV0v
z<5C7O5mzzjeI%lR3+4_j(qda9+iv;4>@Y4-EbF(UgP3!7>>fLGPTq1AbGV}nsRA?8
zD&N?D3sT-o8uBanP5>{`BvdZNSq#SaeJ#zlpr>Hjs$fa1k>edLyNgJ$h*O5?M7L&v
zRuQ3MaeVt~m8O<APYPa<_-oEfz{{crRysCSCdG^&FEP#X4gqFg6-|=eR9Q(*QhNJq
ziA4!!BotJ>oZ;Bx5^d022e0(FP3Qj*xqQvM!Yw&RK~dUKVVSUhGjLd1x=4OB`(+P=
zpdS7rn;{YUi7A8aPKLJVqlkV^v6bcf{`kkgiOd>Y@G5;%<p0!hR$)<n;Tj*hyQND&
z8kFu<X;hF-0f|8dke;DiK&eq+1O)$z#L$vMg9D<1Fd!iz3JlUcFlU43=3JbseX;kq
zpKq;a=X!tdn?O3lS))Ut;Z=J}@Bw^sxV%`PXg(usXBj5(OSv6PfrINEx(B{P-}S<v
zwPTaL*@zMu=IzwpmO}KZ7a!IxQ~(GV&p5gv&^NX5$ZBd<3xJDRxV)=*J?}pKUAX<;
zf_-DT;Y3v{xieW!Ct~~sfAd?Y?0s>~l5zjRd@^4t(CE;R)050hrfPxZSvtvxa(B=U
z1hwMe`|twNu|r&@ig?`-3%2=68Nqd6RnrrmZ&@tsO&%Z?44-3vsj1O0a<_y6b&4&v
ziFiHc)bY9W^JHo?jLE7ARd5C~Ar!Y-0mKET%EPmXT?UyY8i#V*|60_8NRo;aUqp72
zD6xtj{M6#YQm@PTe4GCRrl#F&O%UGk&hdv=Qb#?%W-PI3KNi1V)VE{><2@pu`?hME
zz4srO$JEZs?U-tI-#L9RrpY?Ov^pscAtN*y-9;V4D#fGGmt&^c3lQKyy1^5P)1P)j
z&^CP|p0x|l+Zw+Hp6M4><o*hMmCINRfY${(4GCP!ic^MeWq!M5SonoXZzYs+SA5pZ
zvDypE)&2Z<Mo79q>(VdEODI!xSm1fi>)m`9Yi_z3`Oat*8{=_3CXu7odOS~s$@0c|
zs$eK{%&S<eNTs_Ki*?rEixI~}WWI%I0fEc4gVC)olPl5d$1j7(Qby=v+CI_0UUV~F
zNQo^~Qh~eg-Dh60jQ=KC@xXGzW8sZ@uy;#tH8;_~_kNL;rr>#%(Y@g;(WHn#7FBL)
z-o*u)3ihS77}R4s#b7tR*$`6+vwU{dv0FYCiqwUJ?~N@6ad|5>;Qk=|HKn?A<zO%+
zw>42xLR$qzYS{OrGj?jkE<xOqwY4eEBNG(}MJ;U|rLU{-w`OmMj{tv$aoilS|KYrH
z{hR+0x0HtxuY-_2zZjoebuEFv&mJ4buVwoqY*LhAM4M$5%ITlCXanVH-|Uz8N;|tg
zLYr?q#;rV}*&*HrT*sj5<(=uF#L?ZS7#GJD<y?nBI4`~(usH#CLLyFO$MZ<c<cB{a
zzdSp1t9QK8=g2}k(n-!s`eBO}Y%sun1<T3m)~#MTmCwKb#)@%Nb8u#ZLSXmQ)83$Z
zN90_f(WHEv(pv{9G|gsx()8*ITjr+xk^cnbA~T>mcvLfUO@zNN#jf-;@T9=ifWbPg
zSB{+{<HGDM4Uk1)BmX}Arp2v3a>eCr1QC@&-}rY&4SNPlu@6(ch~AFRj$%nLT)OPT
zMyrzQWVp$Zd~LXD)<T{}&7DN`@e*MyxdDh(#f${EDtU{!)ib^;@b2I{v?JB-E9Zoi
zqRR#!pPV%mfzvDya8fR=f&IHC7i-@uBq5$%#u4j3Y)-_oq)Xy^l>s<sq3f}a8Xa0O
zri&P-H^-MLaA@C$9Vo?-t8M*mIURqvO8R_8%n0y=r8Nzl!`wvsU44|3-LJQQ5nWnd
zb|RN2DMXmxbc`qATPy_8vK^`9FS#y?^cm;$M-|vZSLFiVS?yqoz6JJy`?lYomx_Z#
zfKMPK4X|A0=&yN}YtGr_Ka<G3RL>9*gsShp39P$_sxL|JNmB0NV}E_f`nxTvSix4~
zm~kWTaH2f9$h>F$2|aqZyuZ;G92sxFthj?=@{#U`8V}O`M4(uJfw_kCqHFY}8ae$;
zyl1X!e!>PR4je11AG^D^iA{w{_Gk!`A3T<bh_`&}>T5I=<O!CoT$?jWrWaG=SY(rH
zq2~Kq33=C<AKe#~bEXgpe2kWfH;D`TQ4Dr+KWO}<C>eeN(Z(S{#iH$d$~4+UnEfe~
zj64Wd?geJB)f|@A0HZN_O1MH!t(q7SspbFB@5e_#r(=Y3#+2bY^6PHHnI;i1O#;pG
z>aZxaCIYb4hlf;XdcF4yC**Fe+Vezb*B{I;_2Md@CACJzPw>r5#sM$ItRZ_^+{gkK
zyEAj_n0)AT2>b881B+phrL&evP8F_NN8wr$cPc!*=b(f6w*r#-PeJ_*RNm7h{FUCG
zP?lX(YPm_77>Eg8Lok@XjfOJ(Seg&u8Kyjt^mOr)FHiEdVL+p}4E%SCD(6rp70W?W
z{u$O7Zt*rgENw!Dc#u~^oYZ)$tr=I3JwEBZa0_x_z5!(O#gO;j>EL1Z$X?X0L(iwa
z74civp?PpXnqja1=vMCZzCg3ZDK&}P-l4Gc35I_)BbYZ+xZP>_Elo`9vWq6X>vZ#H
z<XR7oW8qBMWt7Q-N-KM>wkbz0nm5&KtZyWs9UE+ONY}D*iCp;aYFZC#Kit>9alzuQ
zF2gC1PRv1(-dD>|oD#FI|2hk;J71crk(hn)V@^+<uJP>f-iHl%^q0HBg^|tRM0v+_
zAFLudz&i}k7uomy5?z0yalO;aH+TT6#Jj<O4}Qir;Ovkv{CFMTEI=td@+<Gl>h3p^
zOGQ=vT@eoYSq&1ky(HZxEgQj-M?)xeqD;9P<&VIHyOVyk7$-wERI-{)qF!h-`%3m)
zOQ!DG$8pM&XA{_{GP~1bVv?!*>#o-tj4hZu6R$#+wImoQ$(p%*>>g}TEN^3NgJ!}J
zUF!y}nFfb&oWEIP#a{N`GWT$qfzymBk74U`<($*xOcdHNx%1&bQb3(TtA^|OUTqv5
zqTkXI#o+&J@jlBUF}@24rL9yErqmJ-own2*q049f8tLKc6?UOkE8;?AEp=ou#LMWF
zgS*nj21!Ga3^s{ucr1Cl8D27u-j1q{G*6qn5p|6`u2voIIR70IIT(vXj3N&qc!Tgx
zf?X$kho;Ne`gF9t`fbzcRN6DLzNfRFL;z2dQr*}upNLcu!Fj`1uCCSu(M%sT$AGX4
zGAG0j*MgHW?ixouAd*2BH**4j-`f<<Ye^!oIWx)p^@uf*u6kQg+FyFB7^27{cm50W
z24d&)h8+x}JgX6m_I!9mN@A;%)jbX6#Oq56PJ;~Pe!O2~Jczty=(efn%#3bRh{dpt
z1%LY82ahyu7-sWnc)4R3n31nz)$sV1Y`w5+*R#;LI8MJ1zT*A9w>N#9J%nmW5;wD-
zv)-CAX!VTpGs|@@GICec9Wll5jg!43NVc~7@SI~;`6wBg!)&50MOE%MTIobYYOvCa
zl>)h~woQ0GrG{=#6Jn0Cs@uFwp!C}i_FbfM)I~HPB~i5SaD^iE7Vh#c&13bWLVS-M
ze5iZcm{<HCDI6AOYQT9CQ=m*jsG9{9r<x0-K0aCs{V%iG>xpDu#}jV30RK(;hM<aZ
zp1qNr=Lxy8PxBo%E~QWhBe!E^j19_fS2nNA8~pNq$nA)VN|~LA?r6?_I>>f=uOOQn
zQUk(}hlsL3lDWEw+Jf<h2l;&)p^Z~|?a<yI#qZHK<NQ#mY}og=W3gq5HX46Y170@C
zOwMge>20h7i%_Vk*)Hnm?`LmVqwe8ZxO%paRj{Z?05bme8=P;gJ{gvk7ua)A5F~lG
zu$P)ZZms^Q$Vzl0OeB-Attzs3$g~f$0qb5IWm3=M6btK=30<9)u^Z_E6<R~cF&4t}
zTf=qD`|EW{d<lC`j`(*Eg3OJ+L?5DKauiBidD2ofq0(Q{g5Xhb7Nq2e-Xo}lh++E<
zWwN@)u2{aWQHeH?C*Wr6)^|1}Fh9Ilu9lYEWvix<$_E)Gn>T(|f{x@4NgjIemKV4s
z0Byv_fWPTRPYveN+2JLLJRj_5fzy9U1tgzj{vk&Jjv(i8kL2ZV^3*`2O+tzT;h`$O
z30hI^zf+-VFtrqOOm>aIZ#Dg{pW5x8<KVyAZ&?kGg+%q8)xiP?{Y>Z#TJ!oh-;Hu{
z&N+F}FQ3&Km&@8j-1wNNs?(m-27bVQ58C*IVT-DHB6f02o3(Tq|69d+nvs>DxQ1yU
ziVwE=iAD(#v`Ivra(vtl*FU+3QQQ#$8TMuNScWq~#4o=4&Na|v>j)+;qm*0T)f`X<
zP;O>)pP5W!ii})tm{-kIv+cyHXX-n=;Ep3d>VDHRB{&;BMlkqo^7v@zr~$Dl7P+e1
z^e9QL$*yOWMeSKx<C@vxTo0xDpwhG7&C?p12zbf;N4{?cOCbKXYtXk239{8yu(e%j
z5RF?|2m9Ymt?YA^uTijmKF{YqEv1Z7=6?xg!tAK`?u12OvkVwZPy`s9&=V)EMdmw@
z;-JN?&_E%8HY>@|fPCAV@SrtpVSPG?{hsmBM!EQKU{KJm_bU6$&Tva`z%A6iyRKU?
zp}CxWfAID!_j*;ZQx2d!?715gBaSKJ#s7>rjRBxy$-H;Z#3)}=#SaV;%2%YoKY4wI
zRqM&;sEt!D-nz&yPNlNam|$);otj+cMV(Urxy{k{Vixv=?J>EPD0|b0WkRF-D;0kU
zYh~~O@w=u!Y?C74!&q8&h+5D&NJwIFDr2Z7YgE2O1P%uK3Qq6Rr7LqE^M6%Tw@y8A
z9`m?2Wx@|l{xlerUI9LbtHdb66zPizPE^00i&lAdP2yV}>~&C)#BwQz!%&FS;WUPI
zzsk&vCzJ`z>5S-K-m=+a{|LjS6jAfa=F`0*@OM8>-Y!G~7zmKw_Cx=y=-H9~$qk;6
zfQQFX@xKcGNc0Gi6|V5D;Ory-0hgexRK0)yW!gt+lGoayWa2e(Q6$)fo%XKoRfx`%
zfG`yb$iFEz$ptIqg3XPY@%oo5O2fRdLr>V-qJ~zdPf*zn<u{rg9;7!OKuimtkWy*R
zl29)6zu-ou3l#QBWIMd%S4+>m`*0$RtbTlM<y5<>4ZCqKVu7kF*f^%Z+8FMRl~V&N
ze-EJmeB<`PiCT-z^k;$S;_M?1u2);CH}#+T@)yT17?SF;^b_5gn-6^Bquhw>pPksY
zSU}h+X6j<mKeQ;(3|k(yoC|c;l+wxCqXb`z5{$dW8V2f02#d}REHR9&u4DR3`&*s{
zNxa#Gf&3MZ%W|zd&YU{YJTelbXxjAJ<9hK7dHY#0r8*s_w#*b4t=;~U_C3K#TPQ_7
zVuA&}8rXW{Yh*2+ctAY5T+0^OPOOkPwCem6)Fg;AIUyYC4=sA(Z6cj90yC*dviNVF
z$K~dbXqF%Hk&uAk1CFm`cs-${EGoaYtFX8nK9bigjMK`fKhe0tJs8u_0g=dw89dAJ
zh0S*(Jp)&Fp&{0OTzd#rYFI?s9mQ=fOc%eJadDFMnSJLa!t-s5*f|QVZ3BtWjOsM2
zl}ljg{haoEy3rR9jNT|TF_3_XvS_pw`sz2_Y&c!io%O~RX0#;YuSt;2lZ3I2*pW-e
zw5Jz%a!jb_`+Va*eSAq0-GQ>gQbsgk+-D1A)0%OGdeu7#-&>PjNcxy(@f*3w6m-+u
zF(GxeU2Oz^R-JsTT;4{+sobS+l?@V<=&nb<-llDA+{!IbF}lQ_a-#V`OrvM8*&k!Y
zYuvA6?AKd|${DW!C;b#q$D5Yyis*d_2b4bGbrF~ft@__Qt{j%VUoOt;^%1U{Jm)LQ
ze@o4kaskv(zwIi&j=B6EftMe&_F}Yu$a}<C{f)y_j9u!|Kx!0-X&UK_NDt^>gW)K_
zd*Y`8;Z2u37%;3y{j6F(W|>~qD|GZ{s=N1p8{XJ${kKnZ3F_f^`TXmt)C!JY|HzCL
zW^T2;bH}mjvNW|6mxONwg|FW8b?;jU{CUSXxZdio!4=5HTyEXzMW6=A2G5SWVs4!}
z9uKpGB3W(6!VTMQUx8o3O27fwbOdH_2`n2xGuiZIzD29e@ZY{rjryzi?otit)xNg+
zJh9Y22Jc-++P}_1BOeh&ww|hnvamsGVCUSrt&ajsT6NoWuj<R66PmAJKFv*DaBQ3>
zel2>MWy-lpdevCYCyQA^)20tGvRj}={&^pcJkI2>{4X;yX0NW|ADPR+qPC!_IaR%#
zBOpF+xxc!kana2$@L#bgxmGU$f#=0BdjUxx_s2vs?GFl#|2oPcevRoM6cwja0|gT4
zs(40+{3lpm_2Fuw82R4};<w5ep$i0u_}b^aD_avc{jO$E^y^zn<JsTfRkzKkP~=(G
zGqPL$h<_)<m;O1)ADX4Y*2m(Z%6V@Yc*xF_io@b3>z&3Q{kv4wwg~C@WId>S4^ERa
zmUsB)?t^6j622kj3g5e$3ZU<D-aWi`a}CFS+Rhq~PAVLvZxiJ166B%c7U%(NKyos2
z@{%$NlCnzoWRz5H%Bjf9i_6HU$jH$2pAh_?0za_3_oJ}?zaX+ey8|cy-PSkNtJZOj
F{U5Pf%~AjW

literal 0
HcmV?d00001

diff --git a/docs_new/logos/databricks.png b/docs_new/logos/databricks.png
new file mode 100644
index 0000000000000000000000000000000000000000..15f64018e6b41458d250e2b9ff925b3e02370893
GIT binary patch
literal 59004
zcmd?Rg;$he+b;?RDk0rS3?&WHEs_HQGLljPlF}d@DhfkMcPS~|F(4o?(jnd5-Cg@e
z-tT<hK6{_P;IP&#7WK>%_kCTzy6}IcB#VVkhK_=Qf(4b6f}x<?rA9$PO@D|6zSGt%
z5dyw=VlFB9%G}5Z1?6d&(c8DrY#r_Fjq%>TZS7)VLbrE@!Qnx$x8gm&n}2trp!v7`
zx&I9nRgBt6&r6T_F7w-WZ%>@Cq@Ho=4Be0W^PFr(p8Y8^YM=MVk*D<K0;Fc?sG&)^
zdN33M)5N>)0_lu>M#-OVq=r(PQlCEz?TUG9ppRM{LA}99?S%Um|M?Md=$RvDg8yC1
z$6wg)8pfonf7~#(?dJH&+WxZVsS0OVS(&Qs%Qj_swTA_)?9eVY^_@<IZVu*F4tZ9n
zB5KapDs0SqH0TmEnWzJ1KcK1gkM6t@d9Imqw_&#9sS_HFFG^d6wDBtj+6XQKvBn@l
zJWR|h1yT~tD{ACRH&PO!6;93_WodXe=`+rUE8LtL33pK%OO>xQDWO{6eo-6A!4y$Y
z+~`qIKKh`boL+;^D<~+ihbX9onkXnjF(@d+Hp6eILc#B%87jz1fzROMRI<h!e0pRn
zr{#cxf`NbYLPd#BBmv)i;0RTeez1)C=rNDbj`YknI3fyEN<!6TY-7^ZC1N;^|LS~{
zjudIqX|eEi@NE&-TU4jc$0cnq`qGsmD(p=AhUnWmg}Gk}dyX$8O~2f$*zSz1h+v>s
z>f=*-rj^L>h2brR(QhKX(BsMGnuRBvdhfK|hNh`1pnaB_%_iqw1@8+ghClvwpk!Ol
zb=YzM=k@=CkKd-8BzI6yPIxkMeF#G25*yy)T9JCsU-jdvjZ`G?izJ5jbjR)g;uG1J
zC1P!S4(`dFhbSm7gC<m!gML47QKAS`d-FuOW09WWa8KIMBNMmYB*_m0(@UaKm|ePn
zwe;MB1@$i2U=OKLP&mcnGjVZ?Elg_)=gj1B`mgXfG&JI)hKL}Fw1y;*J~$6#$eG=a
z=Evj9ttwTo!|Eo>=b;hs2~3|tuuz_(prU+yKcgfr5!CQjC7xW5pg-iCrJLxktB4fq
z`F(R~dP70OWe(M^vAUujnn>L4mSwF$TKTgl?p(1qO9@=1ekiSsx+o@0-4|MrFG4r!
zFc-PhjAG?t=J;p+J=I6}JX{-ZW@)&faxJxz{*iX&czUBqjP<$`CMQ_%A5l<#Kdjv3
zvHLM=GJ2-+O9vK89YuI*arbhAaXi0SLd`#pFR3b{L$^9smv4w730bIPrAFV&wsJsX
zd3*L%uiKt%B3~w$z{uD|*pe><)R|i`UI~x4<<9I5^909|a{EtLjpo<9&r*X22|?7O
zwGP9y;YoZ?e6VkCp!3Tevm8e2;(7@Y_jU;3GL@<YGn@Tu(J(qsyQG$TFCOCJ3VYK$
z!VOch9&KMfuzS$Q1GTrsAexf_D;Yeu-&Ry2cVft*#&SrwdZM2$CY3(@z>C&3pL`O~
zGJGG`mxkR%aQFT2_}3Bl*Dl_#p9wqbVdG{CaAAPTcn<}|nmf646=|CY-P~5jEA2B{
z*uzU3ayO=P3F1cyDzJ1>qcYq9hltazJl^87K5b_#Z;;6KI;Wff&;A|`3QE;r@ZtU|
zG7HbelDT_a`>MbC`pnchoRq1blfdIP6~_Kjn_qfPS&jDb7zS~}F?81cS_MNFhkU7s
z_^hO_!2$jqL=BqiN22#IwDj3r<lKd%p;9JvbBiaG6_;xs6U#{ZXY@V1N{bKQX6yXo
z)AvXFGaC1c<&Bag@TIraRY-hewMx|td=RoDZ<UCSCL>Z+HAX%pRLr2=xqZO-&sW!|
zn_FZHj}AoWla&Hv(Dyp+jhVt&-HGvdf=5m<{ILUV?G^4M_iu}bX1Yo}SQj3qP8Kr{
zV-CP(CbnP=D0N+!ep`#dpI7hyC7(;lda6<CauRz4+36TF`&`8O>FxfR$eY-%m%cbt
zH0hDuZed{yR&I+jcI)Eh{9y(2I6<q#sqJUB{3ZrDJktK)V1iru78e&6`pv`;6D^`o
z^SI2BnOB{;n@X^GIKfIaN1kp4zR!>N0+&lDGFJ#MGAdNt+_IlXvi|3WL4!VUYCP#;
zKvZ9Zx-7j*)KR1SexyToDRrFMEm_FOZWgBipR%5>K+^60nAIPZ`|GeoxBGr5UV6P{
z7pA51Sx0Xp7Ynlp`E36Bm^YZj=hI1OxwY56uKHEM6X=T_`-HLHwK<{$+a^>LoaZPg
zGcvo0ij3-o_>r@-rVAx=KH-~Xg=3R#;W;iE%MEtO`)stj2ag#|(O(icwTz{2yo~XR
zS&a7yjrE$)-(2%hVXn|#nDf}H7L)3S$BFuZ%Zjm%`P1CJt2jXLCp$2xG5Eo4Ey$)@
zG6^rT3sh<E3c@Wd#aj%(3+!9t3{7>>#q2)odj0k16VZ$2uD0rxCrt7VwYm?rgtwM6
zEy%c}P+?Gs#K;tPx%QX**_-)b2ke%JeSzy1#hxs(Y*r_o!>Wu~A`pu-D^z!GD@TMP
zo+uJs$j9TySn60QzV!GiNk;M1Pb@4q{bjEAk9Tv`g&k{HvoGR{nl}0fl_2(|_^}fq
zzBuZ3Jf}Hb!H<%I23#^K+Vjc;3(=ByMZe3Z_v2T-{h>`<+R-K}^JJ+@C7UeFNAe-+
z?WL_NJRze~_M%*U(ZmY#>Xz7@APv-Unfi`(;I6yV^)f+eB8QjK8u}aK?D9CPmHxg{
z$`|B1RXnqT|Ex=deK8AEB#DTYc!6}VBj%U&$JP}`WPUI;4Qpe2_n|xN`gi46*(?pe
zu$7P=SdafIvx>Y1BiCn~8K@{xE{~4O8|H5Q?8&6Glo374N<*zV$rn&K=lIv!^VgbG
zDa1S><4qjeyspExr;y@0FPV&5UYT~Wgr|#%<*kIqXjU$*=P10M)$M+Ho3q2djO$>Z
zK{TSru#QFG#F)1mAZS?gfdl9m`iQPL>I_c;4vomzs7$x*3eMm`*Up6Pq0O>`J(?fa
z1%ackIuXVXgpGzGbz_A7$_CkGN_6)W!(A(+Ry~x0`j&XYS9aYnNQG_0cGn>kjmekU
z9&#%N)>khaRsXY2c6Z(-Lrl&b17Y5?(p;BS^?d_x|4uYb5!&J73OiFi(%uTMBBbeA
z+dh?fV6-EcgR5&|^OtVY;o|9|8l&L3-83!>XVY+(tstnmIYXc7_L`(XvgS2?j`esl
zrNL!0?Z$>+PmT}!BSsU{+aeNU@-@uJGUKyO1Wd(EuCx3P6I(E9{QKYMR`T2?c%VTB
zPS*`*22O+{+4!`|9}&gsE5o#K?K&holRFPw5Ut77DLqGO?(D^Sg?l?9cApd`Ur42^
z!KcFLq%2F?2s7AeAb1?0CEb)m5?Z1K+)B56EumApU#jnvLjyMI-sl$&<J-w((OR#f
zR~Ffos*T`9AoHi+y|&2JX<RGkpDfQ!Yf#8$=4OgA_Cw|c2_Om>o;w_d)W@`G?mQ#&
z*DfB*2plz^ue4N{yX%wB97g_Mmp0Km(E^jJ^epUkxG)Q{FXo1}m&XnGQ>b<9BjY~9
zwMh*Lhh!*vH^)rX#<VA2n4Pe3QA%=L1)Llvbi$mLjO1n4BkZ+s&6^~pSoynUZ#ury
z5yc#BrTj!mR1f-(^^hIhgumjT*MNRA)1Bt{0Z-WblY~%$br|{{UdG%-n4u-~ppL|S
z)({T@dD56%#bdo1u*_3A&UEkIq`gsE^`sOld4`j-*do-G_7ywFxA67%vUj5S$4lVw
z!$Sv*rF<nXU8F(n;*_dlbY|HJCCa~=WepmuGhU6EzludpNLWvjVq{NBX)7-wTG!?g
zD3Y`=UPF`4SEnCvMPH2gzDrt6CP3sg7EB|yqp4Uvd=CNbXh%~#AfhASN}|07>s8^w
z%l&GxFUJ1*bpa<kSWaHslT3}_zu%cGvF5>Vp~OTo@`A!ppgUb}6^f^<ZInLeannm*
zV4a@8<?tY#>P`)fj33;$t<%OgPDXnhJzyNV6Lc{|A~cI*lu>(@Zso`AE0u#TmDx<l
zlb4Lv(s1oc#)4q|@+D|~{d7*?)HmYWUh~_E`;!TwX=x!$Et)!6kd$LspEFl?Q;N)0
zZMk9P5Xr7LM^v%>0X<9`8)Kb6!$Ruu5#J@DcbS|h9F8gPFpr*1ef&K#3O7yX=QE4D
zclB(vj6ieN$DR9YHJ;S|)%VHL&=<i~1|b@-vu{8`vve<yA^grecc(<1NN>5$m;6jg
zn`n5_+h?wvn08kTbL#oU>&`71MG^`Th>tUr!E&{)8M+MZ!~m|6DKF!c=>(M|hCp*6
z6YXwW_;DAh=cTHbgCll*Lt@@|9lvyTVext+-^_i?e}&oh{VP8-|J^u~9SYAD{aTMz
z>Fm6bR8@iqhJY!>aNgh%#6BDw%fUYL_p+n=QI>hhLY10QMLZ_Vn8_}a+Q(&1GEi0B
z@s+uC(bH}si1hBWiDKk^v@p<r+t@T?f7YC%CPaoU^-W)*=>5GdmHZsXE!+51WY$E=
zUKF5gF@R-KH#S4Pe#rO3yx$*NcbO>FN%QpQrtCmmo%)s``EG~0L6N_lW!3W_B~uvL
zdP?*&m=QogD2IkI$HC->Zid7OI*m(D6RZ_p;!%+r59(Z6C&u2ty}W*s^|u9%&hZQJ
z?UH@j_;&WLb|2k}$JhNiR}YX6*giZ8)3$u^j)0+2UowZ8mYw>9I636%*YkE^=RzGd
zKIaz=J*H&4@0)Ng{<F^64ZR*0_XmF-b6+IQPT%<+`=MRK>b)^EjIoSR&Usjgv8GjA
z#rv4wGIg^|p}ymD!&OcvAKo2PPN{%p&34G4A%8{PZIzo4150*p*q1zG?lTF~xn-(q
zqK(QMGAz1|;S7>D;%_k7zLU63i0NN=x2Km0VonYF1rc3y=o$=bZ$S-wH0T^qC7e}v
z#cy#bR^mj76dEF<4YF)ZA5g)~C9hkq;(+)i{@7{AaxSLew_VY2@8EpM6UDlm(#7wp
zu$>?mKHXQsV>0!t&E6MWtB%Oigo(y)wMqv4o8ar)qbhR#hzi8;$Y>Auk#G8Hx^SUm
z(1^9Go^Q9GJ6xA>z2say&|ICen)(-K+YKhGE{SR3T`0|7#u#9l#|49aQmOHHghhP|
zYOMR98_%h@e&Ng%dl4|k5L1P3AQ;M*!wnYj`vW)CSrsn!)R2BQWf*o3U;_27vYPth
z-{5GR{z(+_H4!7yS2EcdOfucVF;AD@y;;by_u&&+_#=Tz7M6p{ZRFct7`TGl$Va!%
zdYs;>M(A|e?Ri5PKzaYTORl~9_w3vwV<`Hb;HAqioj32IK}D;WchO&-C4y86fSTlY
z814R=SId6pT`m(QL^lc9t^YRE{ux`}se?bP;glq!%bX5+;D+&}#+D0kgU6Ge(<0r4
z+SlAxwnZ`AWLzV+3wv0pvzyLj>1Cjf_F<|E^PqlqE|*!84=B`Aq|9M5$nIpx#vv<!
zt-DGuN}^;>;H{JpSn-^lQB_DrsP(yC+VV*+$)ce~xMdF)Q@6=>V@3|%uE^=dCVrBy
zY#s3r9-S3}p)UrZs3^^!T5wQ)C&@ioUsh2wfiZVH_O~%bwte6QP4ewKn9$VDj~rxY
zti(h+FT26`R+ty%rv`Px!#n(m07B!2DX_B059U8~m3vdntjAu$D7RedIXK0#F%mGy
zYcLx2cw=ie>^hR}hP2&q9u54C06B6>zQDNEUHdRm39*#TTmXNkr5uAYad$;_J2Vid
zhY5KAB|<GvWD;OnYlfer4U23b)kRq8=JC4q*PotsPe&r}bBNQKm^L#7e@MBf{2xN6
zC(;rUUL-v9lv13GHc2>v&NzjOiLFa2Ug+v|xM~z2l)8A|*D5wwarvyp8aZ^Y^7<y3
z?1Y}6oip%OlPxl&3WaPQgm}CddK0<iODw%EdnGTKB;kPVt>_8oJ99JepB3P!SOuip
zYYX+`c=G>=f5x_WEGR>c|FrHC;ZW5z&};-KE=s<@cGxEtdNiJI&7^@%MY6IV6;F<z
zK2VQj88kLQ{MZUymlXTJTIh9sH@kYWcp8?(cB%6m2Mus+P{2RRS^uOrP8Kh|6eSX5
z-2WcwFa*(vbaka*2_A5LnNqO#V^;K{`mpV+>iqRW38Y5sNDT9cO=)$FP6p==GH5{a
z^u=L6%&u<Lf>u(Iwr|~E8&9Aw*4BqK_|ffF5>v=wrS~PigM7_sNNC#8Z*Lq2t@W^b
zfwfvyj!XANT65HOKV^UlE-P6*9wfvgm%In-%hxmx3*3D0hkKPMo3rP4cdWEdD_M=#
z_7#DXtK73nTe}dMbcn~l&iXq^>=VW6moDaSCi>;s><vbj0($0o+$>f~dpW;#)co&#
zs=6OI2{3-*g>P%qFF_$ST6^1AkEsd*McU|A>fzyL`LD}VKR#W()a~pJ;gz;~_+ysj
zmcV|@ouz%w%mj}>x5tK>_AAOK^gmA}R$P{`xC_@^P`()3L2u%xWHK`8y3*Hq&UxlK
zVNl2fiI<R)luCd5=Pt_3*>5II6k7#yHab4|YRlzdK{s)RVU09<H=`nf?73^vylYX|
zrn<wWoMwGn#<+RgoO+%RlKs&DelPvK5aAB98Iss9uZ&*hv@0%<UKP8Y_Hog;$!116
zcV|o+H+;YQZ@G7M0{WIc`NX963;;s)Na`5kbI`Y?@?t(viuJ?wOS`*dtt$7lx8(8e
z_*hxUCScTiG#ye8s^Ykcm~fq2{?dt!p-$>KGB)jYlQK6iv(q*VCr~WS311;{bi*V<
znHbv1M@Olgk}AlL8bNq5Akt^dR|otxvjxPTZ(1Ioy<@?<DaJaDnO%Ga`?}B<15UYP
zlZ#74tjM=bNS{4(mPftg!7!OOL35T+ywaX{u=2x{)})XmrZ+2?O=`1fZ|UbXwR9Wd
z%TFW%|M*aq7A8aeOHkgSP4y?I*o$lE5UrzvZXaWR?)$hl-|X%8iRjk0y?!&8WY%Cl
znB!{Y-2Kk#1}AhxLzE1pE{x)3Y;=-jAZGV3O=T*tioN#z<3yKLAa3qQ>#|%QQEbN>
ziC9sp7Jt@?rn^pX4-|BV>4&wDi;?(_j5??q+=35Qu(edyEH(hzC(()J1acXY#S?j;
zQvqKt#g-p6WwkdClW@TDu+!~Z>HWL57$_CPVTgkI56EV6R}8i%uxW@Wd+h8HO>T|g
zUyL2rFE#NC`L{NGwfbGv_@zzpP%519c}Gd}<4U3A=#BtI3+LtX!xV{p+<&BE!j}iq
zx9Xo%{}J4D671L$1&f7uEb8Ent5GOEG91&NTU?Vb?HjZIkV|#D5AsFl&o{ik`LTxS
z=u!$q4kSHAZk#bt5!rp&-*p{w;M48f9tkY|YDX%n&i}lCJtP8s@3DWDQY?XK;HTy_
zk)j;be-NZnY{-mmt_ro;nY}l|8lp&VSg@;M;!6Wlg9(po^*pdSZ#&wc0v+}1qlB~w
z9<2)K=_vc;&?d>dlXy&#G4G|aLH$j{L8?xuX&Aaa)S;$j+8gMBpwi4NWrmC@cIwnG
zdr*1>iDW4unR%m7EqWGv9WD!YiI;n+Q+fEw2DS?B2Beu+_AQm|s*LxBpw@+tJH;=s
zi2k<-M|5U!={c;v4C<tSFENU>Zqqg$KxigJ+_3qm=*+Bz09Y-bz;mSiyf>#Jtx5CC
zIeu%W^{p{;LQs`ct3=h63TdqyAeeli6TM1#y=LL>nUjywQ{Yj78?=|EJezK4A+r7}
zn}}s*&Y4?S#to|$g75D3=4!Yl<Mk^e6^<-Rb+mc=$dQ58D^gM$7hjQufwp@rFOvJG
zFQ3OVC_J_ni^rQPZb-{cuKII@w!6U;sG8{gs-U3gsp8=jc+6m>JVyaF-iOW&KKm8r
zbiiLABlRFOHo+KHBaB6-Fsw(lw@Tf}c_W{Q@k|s<*NA_`dK2B*eYj2oiE7|W5T55%
z)0LRmQ<KT4c=nHu6OOXUd*;ttB(*GnL;r&l+J8pue#%Sc7wV`dQkx_PXuKioTKlur
ztNhE(Yatt*3yo^-b4<-w-C`@9=YJ#O$E|obCPNzPx?H<6@tf=zGU_>??Q+3kx$VJv
z!>&O|LrzxS*x*Ut$UfG&kJ9<IO?ZMyQp~|geppF$PPc#GgpH9<0akhNAkMv!_YaGj
zI`ahS2**9;!f*Zt^`0aON@Z@3)eojt^3fNx0aKR&U7X0y_<VHoPSHj*hi8@j;i*IF
zqlXD;TqBid?~R{T?);hB|F*oaTLXHHjhUn*kS}aonz(*}U5uw*NDr|GJ1JW%MBL7I
zpI<8-$WQWCQPPw)xVvU=d)*3)=Heo}_P(x3zG(Vtv*z?CGrOs&$NaKZDKYxPI>HSu
zAv;b?WyqPlvzsl638W{+O3Ouuv#Pu1$syCup{%loW&fJkjMC|M3_T^Og=pcikVoUk
zM2agUgT3NeLfWnUqm<uOBd<dBE5%-o6rI8|&D8<W2Ml_W%y$~fP+qei7-Dy1_;rK3
zMF1#2xGWuc?>Iwa<0K~f7D;UXp}wklu^l2d28eyNRTe&$_*siGS=)`$eCfH7b+>ug
zPzxeWgDCJNkS&$At}4<6MH>BP>#3CkouQz2<}QC&@#CeOv{5>Z7%DQa%f+f5o)QWh
zY*ICPlQBo|fDvDQZt{s)xZ_pYm#aldRrVZBN=XeIpEQxU+`fMVsi9L?rn%FV_9Nr)
zdM1C2hhlV#%Sa}xZzxt~W*+rk+?eK+{B87K(J~ObA60ehv@4BLg0+jDs&w|)twGW_
zXZRoWj5>zzm)tXUIWe`@+LdT?8q?U^dIweemAgM9Fn8zk#KqbNbx>Qwtu^u#+k2es
zI>|8kZV{XQGMX>ERZ;4LVc4Z?wR^ptRA;&8>x&bgRr&XUpG1;(2mLrP%{^=Bl7!!@
zva7b(NpxK2P2(p(h6m{!Iy8QGw_@E$gE*vC1j<=|Tr9Kj7uW3nBMqu7|G+DzD8?Yl
ze$iP#DaJbYaCv?|Zw;5;w?%{b_U(CbAP|4}Rpr#o{`Iaz!!&NcSU^lJOJ_joV#G(g
ze2FjcfdVC%J9|RZP&B7*XMvO$WUt6}xLZ}apMv5=Yh;)dcJb%#?>av#Fza{aQzIit
z(|CAE%6~M3*$Ni8|G+hu$}6P$KU@yFFABGi|2?0O>jpg}JYfy{9DwI-S0&1N?aj55
zukU)|@|~Q&6{1d>dBVr0ENttqhtTUP*61_ph6=_B6H?8`FvfO5{eWR`R~#By@E9fn
zDX!IEh0UJ&%ptgm5Iesl$XBQke)F66)0mMjQzS0NY_W++E<Rg3>jda%nJnM&`7bhn
zfAI=aPbufA5T^GzavZ?&C(c4v#gU#vkQgFTp5fnB>ymORasaqw6pXQIM1FB-=pbG-
z+E%cuzmA^VnMotrS(h6GIsj(;ioY6cDt0?N*YgcK2Okg>Qzk-o|MEbR8NIrM@H%^)
zCZ{f$HYlcv%VrHida55{3fw-0!-h3SDXH4<71tNe)BIQ)B~e6Tz_TzmZGPH?C1WLW
zNFyc1dgB>DU6&8!b-QTFyG|9ooGo(()&x4w6aK1rer^_u&5hwWW6uBP1GBs0Hk8Tv
z&cpw08@`x}{S$BZPVqi*^#|=rK}7ki->K;N4TUNa5=Eo<*PPzkKctONtXwWTjGe6!
zZ=E4ZFeETl2#@s~O6TGKI|x~N--yE{wYI20mn@7=Jv?~vywd8}XDpLV7GQ>C!hHw2
z#Ms3rJ=v;;n7*|EX{XXmw1fKAKk?=Eg6dORG-(@LF@VEiF>Ey*(dZ`)maPmn?(>sd
zS`mwWrb`5Y#HhXwW8MhxgzMG;p3|ui+MTouRKe1D>vuoR)1AtSw~nXxBK91-I}7#5
zGXuw>)BPzERAZO>tR}xnT#w~(_Vq|IxAHO(X84JWn$F)^T?Zt_HfOG#@roDmoH0cX
zDZvT6!YZ!`80$su`C#2PH=~C%@1cV{HP$g1kw8yU)IInXGwFSma^K$S!^a`#Ke=Z(
z8MB`yHqC^=?c7jLWaIN%ub@-x&-*77dUExgT85vyb+wbe|8GH(V|~(zC72}XTh(Ij
z#ES*>Ty{5Eyc+vp#q7cltLlY+-rl`MIhw3Zp*iN1L(`6*4*BWNNP<5vBa2-!26rQ2
z+8On3$aP+EYK3_B*oF3IX>(@Z@YgJkZCTVsS@JYJrTWV#tz6{(_-oOEwQp$B&DS=#
zsO+wHshby>LjJu%P&|8|MET^^VA!4gnMGv&8zaI-$hQ^4?Q&)%zXZa2|5bfr9H_+Q
zgQ%srN<Do}Tl1HBQlr~+MBmC%y2IfCi@tC6PPdK~24`m<GhIskQgxhS9aay1Y1=Zc
z;`u!V8KnA&t@}Ezgl4bCqv>lC=p5Pd8d*NtMLUllQ{FQ3?o93U)cpsGdw7AsM5Tky
zZ!dF^sgL|ofKR`5S~=cQo|haF%c|B<@*SS-0(=Y@mpjJ$t-WL)f3ZgB=-JCLw+0$z
z_B@<Xz{ubDb*gL;KP2<owwj>ck`Omc_Q{HH=e43?K#{|AE=;}INqLP+v39_U#M$#9
zZnkI)=A8LDdPeaTea4^x@!Xbz%DH<)9BJv+&u?qMM+8!Hu)LLVknmxX6qEQcL_>pw
zC&c~YQbY!$GcK-FmBvyMyWZ>dGaCfZ;aJN{xlQJ<ZiT9yMUmz1aG}38L?KkcF)Z?U
zKpfJ~uMeRIohsFZgNiIUORp|AFQi8NPs9;1AdPiUZ>5bOt#pg{2<%6S3!kI6j69av
zU#(drhwXJg0`NKI5y>!t6i;xZFGhNWV3)k7Vyp$?n3}k30}2cVH^vDgE^)sy7EuOe
zU90fOFf!UGV)F7SmicK5UNW&*^L}9EuHf)lK^KRmDToZ61pQmtkQ?zAfd*p7`u`Eb
z=)om(p-NoOK*g`%Fzv&(^rU&96qsaO=!k>H7CL8km085Xr?v$uN|(OC>Ls5{r3r6t
z{87p54;%7GCsWcI23D2smmxQRXzcm*$-1NzcAMHF+MnKELE6=G#<~Ybs@ZhcoK$tI
z73);bhI%Bk3lG)URq3mgGybrAV<;UP<SO~hwe@(3IZRJEk?C*#cx}Bjd@8w<O*>9r
z>ceX;YolSwcp=lBYq3qq_yB&6f3<NUU?4d#9pMMvL|i~mPI!E2%4;H!{X-~-inRyT
zR_>Gp|D`slYB4|hF&lAqA>#wr&M6QusoMU5Nl*P$roEIN&|J2E#^|5jVZSOgId+c6
zuadgNark7#f0hYG01(Jzs1>G5TrnG6pB0c9CPP4-B#h5UyiA2>l?B*ImI34gC7`*x
zxssdJBj+%I*RP(HwoXgt<h&f@^<VDBndi1D?3>t0)zMxo#KNBM!noxVbiPlLVz|>i
z&C%b=o>>WKYaZ9Md6kHsw*WiT;KjfZog$u<y>zw$9G*Ck#S}y_zZTA{@ER+`V;i&P
zZLxIDQ?Z`|>RJS*9dfiy?=RK7^TZb9vDXD%CzJ~cMCgVGn;iff%&FC|G&HI1u%*5l
z!ibRc0U9}X)wIW;{s*jIPc<UnECk>I3RwcAeWxv;V&*&Be`@GiJ*JZDryFfx(|blg
z%&ndP<o4KwfIdk>CzHfJP2%qewC4H`w=(yPIwlM4CEo(ZAX>(x>+^!#(R?1gqFwFG
zg6Za!fIw|~9;*4YO5&foN#3C)N?J&6M(E}miT5>*B5elZQzT!O|2}`lPu<NMicCA)
zKe<?^rqpR)e@z7JNE2ni=Azn5?w!6UFXR*9xgred-Dw~(4J~4C?3ylbVvgMwnnSqs
z%70Eip_h1VFKVcTo4sXF>L|qaKjg5fL&jS(vFcATtGT2PT?_%q;F7Pvj{>I<ieDF6
zUR~9+@tKNktFRZ(akArwn%OlYTOZuFF)`_!pwo!F2>yR;U&pSm?W_vvt?6YvWP@AX
z0vQOqtPRVGy`>dU^eru9a|5+4t5>a?1_?j&WkCzF1$Zv}i02pD`%n(|GIk4HaDp8<
zhu8fv-YnszR*rvB)DAz{QB?05+)l0=@wWIpP)yKgHU0j80!V1d6%1`h6cKZlR|`)~
zzF?z#?D?GqL7A{eOTNbL`!jVs^tWE*5I#1!I#{@f<V1GItrK->CtXy-o7_5Tv*TF>
zM@UOE9U<(QRL2@MLi&*bn?wkKlxAJ3XHKH=GMyCqctbY!t4>;$0B}etsw7so{!Cb^
z^FI|tD)FORY)6@<fBxlTsP0=hHrVj3#<YEe*MS0{OF81&rna~{_+KZB`Iv<Jf!Y!m
zzK$46p=0caB#Fm7B}@!a8B3H2NA5nzRxVh-Q5LM7=%-e0lm+yMC=;$?gEmbLdna=B
z!Aoeu^~YNT<%Rrlae}9yqEL<JTfDg|c_3+R<&s$Dwn?*USq(LJAd@3#(V%MNOh444
z;o_crI;`3Zl*%iuN(cRb4Kw+rmlFJC-tgE2Xk#;*3=<=FbCMTC<%UjvV`v%_!nPrM
z)P4B+YmL*<4tXWrI^beYyAGYyZDQkRdbjADdhyZ*RH^1&`Po#u+e%tN$+|IZ)hOmM
zqx4I+(CfYWIkVAKjv`!A?)=_7=MADk`KaCXMdP^`KdLHxH%@6H0H@MgS%<vdqATK!
zcfA^Mt*3l}><!@4{L{OU%Qr~xl1J6T68ym=qYJ7xi2ai}Oi6a4@sIW7QLU1f<buHz
zMN$r~U=<T4O+Y5D!n;s%cFl!}JgHtMAsO>LehW#TseBwWGOp4koO+A$rk%f)H6y<>
z#NS;i4oWQ;WUW;Hzs$4nJ84cCE$3HsLQi&Q$)O`vqwSr$o-kl!WTt;&xCNBKnA9oB
zIAo?c^$_Hb@e1K63@qW!!bb^xUgW?H0@6P9;<5-tBkGz1@#|ff+v(gFhN_9Nhj@*U
zyqh+=hK*D{yXSlWL>HhSvka<HsjgY1gl&v|=KH%nSqoD$%v5Bw9wfuAq?@^ad3Bdb
zXGF#yZsBicR?^fmctzK3qMwn&Dk0+VOs}B+t$NLvKhK^R^;yD1b&b<x(Z`1<wrOuM
zh){mV_JZ?)N_+&(=MxU%P$C0iAb)NJM96eEk;hH#gM{+V!{YLBt08iNrfl}oqm0@f
z%?_gBQuYRvOuUP)`)Krl6l-5|cS3eLID$j<NqFF%qD2qA!NzG2|8v|9JSsAo7hgu%
zHzrdW&rW`Qcb#}$F~>EOB2{79(k6ZC2ZAaQeWO=xM1m%X!;(@PbM(l<hsEn`$LU24
zEa8ZcVQluSjSp_Y8k!8F=dj4>lU!_sN7$jJ_))}r=Ep$(-ktdT)j!95guyr+ao{Y@
znXMc*xb6@CmV4Lp-N};kw_OPZHp65#6`>Sv*aE9t^R$X5o-i^5Jn_4pbub5`&yi;_
zE{IfQ9uI%$@ckuB$&HKq){{yK98`E`+tsm2ExcOEC6afV&mvsD)y)^`)g7d&s#~fs
z2Z#&Oz5R6w1yL%57p>pDjj|DeYkL29R^ij%Z-2?&YH?1Dor%)vIXqi3wwRJqWK4=!
z=@pgnBY!dpGP(@<t^A>VJY6}6v&jfiQc6#cN?>XjiHTq6lO}pU4;q_{190C6_9ot+
z=flJ>59yOR@yT87GOvX~o&i#x<09uLo&NA7;gBzE&d`Z~Q2mwv&J-l)5Z(dKVZ_*=
z!T;_Yem}m1g4`Sx+0s&4h4yJzFZ9LRBE?NJhh6?zuC`lm2LCct8*%om=56n}e8N5#
zNRn`F-7$-%R}7akx8}^zePSnXG^z9h(c5k_8J*4)ptxy~a^OXt#=_2?o%d9Y*!f!W
z<=>M35iJNV?4A6uhI3v1q`p@0kp+;%hVvYD$B)Mx!?eSz{xIb7NgwhX4X6FbdTjE(
z@J!Uvqxi13adhS;_ZPPkxq!zfYr0y$E62Y~wZpA=;K+#?$j@-QFbaSM6pmASNqqJP
zwH$V~>jzYBt%7Wd>mh1i8Y>q#oqK_V`A^~BF~r-H<MOm>(L%Mlq9oWfmUpw`O@>>h
zG@2jWqi14y@z`IB)Xv@$2&5Rl82{2-RXbf!vP9}vH?mnySQ3K3lHEH0Xyh0(@x1s4
zMs23S#T#GK6b7A@;98F?92;s6hlGf`1P$feN_4>aWd#wZ7<gFNo-LV=?xf*jg(-#b
z(}NR!`w(TuDGQ-PNk(~5-ZM>i6{lUdXN}0*yM~)8b%TH!X)~`mrKrFBCgSCsc`m53
z@Oh0d7g{z5Mrmtv`K3Fb-O8Zs43wUh{J4WywI{h$q?DKBrHq>#yK8?6vLZ2f-K<<f
z;fu|?A5%7QdWKauXEh2vqxEt35RZ3<%jYX^d@|rh&#C)rel5q9b*wqkiaJu~?@;PO
z|D>9z>@k9t21JDUwobXo_5fCSg?YwHXtVZm!#|~Ltpa(lL*CWaSTI76QPAAI;^=~Z
z2gT!To!~}Wi{GVa%B^{CnuSx!@?z6+n+GTgNSv-gUq31w<tO@}(TmY8j@|=}TKP`1
zi^rOPn!&kVR*_e~0ge6p9GQjK9Jl+~=^y2?`ror2M@AljeOdOGb$HlK|3Xn}Lr1(4
z=QR8AZfZ+IXQJn&kk@;%VhrL49Nbl$BV>l1f`&*IVp55Q+DyKA!@@f}njtXJRDr;X
zl&k$KSsLHUUY)qV9UAyG0nF(OC1RjNNP;wn1!V7g$^(?&%Le!Fp=^<){g^4qF)v)(
z_H6xaGojO*uj~D)yHqP3A`HTU^99&6S_~P8-tQHc=3ilIN8vDSb7PZpv263PNS;Xi
zAsYx}@8?}lzpu#a!{nAT6$%YK54n=%_yLxfmik6XK;V@>z-=sAakpHHph}{Ay4`I7
zhjsInGU*n;f&8(z4srWZgw-b0A<AU0_BAP`-d29=dc<7U-tJRjR>{KmKHFJRL2->&
z?u9`Y6w8W}0X=28SgeTe8OnR)m5#W*_pQ(qXm7KH6J!#|{3^6>tH;y&>EjfaWy?*@
z%L?~hi*H2QHI>}q00X%Q`tS`wI*q99#o@e;*EPZn<<rf}1?J7HpeX$i$GJu)#<>t9
z{x$c#PgT66k(;j})9xOwA}yjK9EcN<faYc#9fuD&xEemGr-gvXR#qtnky!S)yE8d1
z;t+#KxlL4~S_oQWFZdkvA*BLl@<(R)o6056zMK>VJ5AXu2Ncn90$j4rtv!VDPX?Lc
z7MRC`K(Z!OQ&#p*@X@-b-{E5PP^q#HXcCxmhp~oSBlCVV^c;-;-K-T>tQ$E@<P#BD
z)ee$jtjkF?kHN$ZyQZ(?iLzr+zqR!hGwd!7X(-IM3}A1_3iYO|Jlq1u{@M0JO^s)Y
zAOvTGu3TYQekG=#T~04i0U{PLgz<~M)&5p0`-cwu<ef?ex+eyc1y!nw;S?Krqb`!S
zaRi<D2L?O#5J=^Mnb?xiH+B%TNRC#@XqNwTifUm>j?1NYJpO9!y?1NA&XcEk!2+pz
z2+Ny{QiP}B(8{Wsb+*8D>}-Q9cU1<9QDk#9-EPnz-LPukAY|Q6r>BZ>4O23aDHQxr
ztbbQKA@$qznw;jtR}_kN;iAW(v0)alH4B-Fkl>mWKbdsg;?iXtDumixbpFNHBEv&m
z5MK%56G^wPbJ%6L^Uu&CsqZC3&sy$mlS<hZH(q+yH@Ao}s0=-#s2y@P_81S5Ik$S+
z6$~2Fn}O)e&4AKT5a<a@ho>rLvwm1D_%Yr5rNDGyS6ANDow?p_UlOz$m(>fzJua3e
z4J(I|^rusVy54F#e|Zd-yMbHK39xHQ|K?~)W?Q&Xj*yrBrZP-#qp+f~?&;FIH|f6|
z(y^+4Z7V;Ul$&VBiD5UX44wvou?J|{gG_76w*g4o2Y<IGWaBeOu|lpCv)zhwCEwI@
z>6In%b!2=^M@-HaVu#TI+XPEKVgBQUtwCjxRiz>^DQ}-jQmg1F%N+pyZmjf?x9mnS
z_vk><$1go-Kmd`q5+UP0Oyj|MV@#4S{m<Y`^@vPk4)1IGX|eGXI&q24(o>6CQJyWr
zykgEj`?QzPMn282H?4BODM3el0XX(^ivKxN&)}=-O%>kFqN^*vX(OMQ7U-Y*;O+1k
zG6v15d%3U9LM4dXq@B{nG>Nl@)Hi3<^IfC$EA|{N`51QHCUP<dSC=?Epq0H7ki}&N
z7qOSSQMv6)7Ct_C8o4w%H@VGa8+%}NSyr=gK|S6@syQ6id?jsWLTDt<%R(!Ut<%nX
zY`i4A@i^Ah1-^}Z4OE+CR|o+W71!w~(R5HE25Cz%MRMELBLa;{V7y&4Uu-<SCe^Qa
zda8DM*ND3azhq~7lUpWa<}q1M3h|ZOtbLJOR{t<hblLQ?;95;U>1vVJbZz$9^Aw*X
zuJ`sFa%%o?_RyXKw8?Ltg=U_$r6=X<`k+;?$xCzT?mv1PFM})s@desRUEw5PP5D>X
zP!&N*%?q-};?t`qkt>HP_F4ure--MlzB!7yC-j|VU(xC1xfSgb--H!?rj4P5L(&vW
zOB{}PVMwDI!dpD8>Wc$`sVUY-1rl9n^&F?{xlDPxCrnK2l)=;tU>wf<RTZZ?o0WFG
zGlHkP=ozz(3i7Yhbm79Gsb#|Ye|`N#E=trQ9N1{<=dgpXdQheN*}0^|7dA)j>{YIA
zC{Ma$1-aZsID=2xr@$m(<wbtk;~nzOwzEz5*O6yr!tZzD!q0z{3H$3V*hI!B7!LT3
z(Y2iwLzycgKneOJ6tc=pRZ`)nPoA7N?yL;tZHr}M*2X|!j-mqV(f2_Z4DlnnO4j=o
z?ZI(b>wUz7MF&$>81Qh($tbQ@VL3v5Hch>j(d!nC2zt~&N8#fP%-XZho@2);@dxkc
zNPKw|9Ud;NXn<_bwnKvgK4AFsi)R=n<+(|6=X9LGA1;<Oz6dyuJ7wAbFz`e08h}hN
zh5>oz0;OD3Mmp9(0x~`a6d7I_<}x5#b<k?(|H|8+Y2ptr@_H>KJiHCPygO~t@9Q_=
z8y3;~pIXdtGDD%8^y<M`-o>Yre)%t_q(54(B#+N+b}bT&3YJK1JRi!&w<y{}Kk1q0
zhy&HEJ{#~HwH(%3p$1dNdNC>bR0Tt?ZiA+#kc^)iZs9;OBFRqw{qs<V!DQ?p_euS!
z+WX{zv-A82&%_-@L4))Zzd5rbH8Ku5IAcgl#|K`M@-$#X1PA8eJrbHDVz<ZW6;AX3
zglx7T>Ge<wu**Pr9`cNn+qCN)rdZ5FEX;q98>NnqKgf|Prt0H&{i3T(aoY|JPucdi
ziyyI`e9z5usJu&seQELwW75Mvqx|W;xa;L`;Pb-$eTnt>7D0AqBm0-Uq=WZ;(lTLv
zTaJ6UOD+s9u9z{->?+yJfyT?-EQE89({}=i7`eGkk6#Xwu*A!dseBZvkY-(vgsz&M
z@39%`C%ps3nQf1eopYpM8b78nUo)wbO@C1G^#C|&GXX*Wfo{H1v&$2M=HYEIr?Bh7
zgPJ2nMJ$}#g`OWw$^>)9Z_!;C@=cjh|B_zXq0!tCWvmP`Acw%EKS3W9&&D?RGhpN;
zulR?i{OW~BjjdfO1C+1VY3xQT7<doj@o>0~yoM=rWxu^|@dUoCkS;E6`OCq3u6Zeb
z%O?-6#-fsm!KeZeqHHGua=QB^r1y%h6JUY7K-<-$tx6EH_K$}MpAymly(cXg7t6vZ
z30lA`%yaPumZ+mm2$JFWfE|U8^}QoL+EP||qpkN>WDA;$dfRVe<0EZ7-E=+td-c2h
z;6F77dHOa3*a$m&bzjJ>FetATcL8mGW>Cg=-nCmQooAr8sNZ1jb9Cm@d^WzxrIoNN
zngXxmcYdhW4KE-Ne%vQX?c<BkbBydKRwl}8e!KRr(9b$*Te%3dUt*&HTqJO8t@S1T
zY2(GT3i?M99hRVh(Z5tl@>6{+v9KHaXQOo2e)jYp({f&++bTqu=;dPjGgv%XC(L-&
z&1*FPP4TrjGJ)vURhEvI0R}aKN95q{<dNv>J{zCQbnjO0NABC-W`?CLDX*FW8{}oj
zuVEIs2V;L{s?PmFS;Jr&MylF?6TQAv`jOEFIFS>wf4jt<czng_q};>uk0fHi3jD!i
zEc#16C3ZISR~$~Ro+X4gi<;H(Q9{MoJqOkBBF~VUISTHJ0gpphC2c5}k%)ZOXPDj@
zM8|cW2Q{~*hDG-H8b;H1iM|9|PKljqr-7XK_N>(~CGB;~zt?##npQ_=Zr}Le`yhTQ
zpV~sRau9KD6LK{EmWaLJ&~@XyWDU0_;{bmbRy@f%tm>vg^-M$xHYMl-({TUDAyWT|
z`@UEUXx$8eBxUAMdpx@-AG&Kq3(aRK(67tb0a{regM5q%0dt;7eoUeNW=^(#NnKu@
zTN$4DXRj+iyUonAvlUfMyU?q&cL#ylsUQt(#*>oePY}o3d=1u5X7DW0j*BA+Jrw^a
z0Pnb9yhdDlbK6sbobt#vVqo2%Y`S?Siy_1Bv>xgaf}uymGORAjwwOe|IwYFxzr0eo
z1VjeBo+K{2t4z9#+@^x*Cozyp9tH>02GKDBqSwlinu*FfW>qR|Y(b<~DdG(6^!JGn
z*;CI;HguC7G~sZ&e`B`fJbnl&d}EzEEOHH9SlD?gQYU2MmBL!Mak}1CnH=d**nC_^
zAQp(y*Qf!T?j|6qP0)nHgx*F-pU%S@yV%}m%g*T%2K>zS9pN^o6dU@NXRk>_xjKH>
zUUc}@d!|Z{*bhr^(RKmZ&h!<BKAATT4dl=03!r9enOEU3SWYKAm9iFP&6|$Qv522H
zdHf%FN;{r4*@(#>rY;j?|D*jU*30kRO#(huzcU_B4#lckTb&pf^s6)CA`pt&nG7C0
zgZQ)$QTI~3h;s$vQ=)U*Z?Zh;!&J9O_2t*FO~v%!tL5@6ly7CT5&-eLWz-#o_AOmx
zdO7Rl^z}K$Vm-9S3rmQ*@FE8WiQ0&V2E%*wvZ5ab;*mXxLE*XWrtb?-A=VV4Y?{$L
zP{=?2a}D^<>)mjeiiI&Ts|T#VtfH+hd6#T`nTB}zJOGRMF>|?Ej~qe^P~@+0nr)$2
zat}C=WQR6uYJvgZg<9GE%?S*O?^{<99`_`aef>H1hU``M<&{+g<+c8cuLag)bKOI+
z%_a3XbVk~Y?;9P@ntIb{MD5QPWRXBQq*!GEy2!yP&Go|$6lgp>5y^Fvg4P>tN9y?X
z4J&jy&fh7|y?jt2y8IydbvCNJ+9;!J*SGD@aMhd9H&riPsSunB8ZA>ms|#|9)}6O%
zcH`{+N4P!^8+z*cFLbLTOW^%%pV5#6gc<Pb`9K=xNlYUen}W&mnc^AEU5!H7emD(!
zXnVliLU^O3Ve=!PD0BUFNqFTol_qs?G~qz#lJ=jpT%!YDl4h;aD__5qu#Nz1`HKEJ
z3ui`!oEJzJm9kDdB-<lfkZsbs!J5^&K>ZaT&ls=vv}Y`)WfACy8>r)&qpg<l%9SNS
zfp-PQzvASgYZE#x{=azecK-<aClLN+5!S{_A28arH~*nkP)NW1B2DIBY=}-uN(f=h
zlG5Mu1XvMmTyF1nRCObOYS^d0x#gl#vixSD?@r&u)_VCMI>0gif;Tco_!5J?mf^V_
z*sjlD^O5oGDCNt0D~#g;;!~<ozzXxn&;!zY#rVn3P3T1Xpt>5~WQ3RD(9^cwmtd6g
ze`iWH!^4syz>0!bMbW4mSRrvI3p~{`Y6nHt7)|^~<B*++<|=FN^$!w-W(x;YNx}Ms
zMDH*8;&U4@?2_}>d~n_@Ll0AtMYcc76|=`E?$1<iJFK$G^2fm9?-&Z1TMYkv4?qqL
z@)grUholD}pG*A_1!ZF6CK>~X3ad$aGYgxdG8l4kv$B0_h(CD<1k%l#e4uj65>&Rd
zuYTasVl=trMbt%)2-Xn+kw*Hs*Ey?WQF_1Z6KZ#^e?ca0T0d^3<5z7U0yA6sL+6HU
zN|?&k*bNTH)~_a7guGbPj2SXYHbP^ga2baO@9Bs>vY5aYAj6DdcV0Dg8r5`g#1sG1
z@M_D9amb9Pjh9f_y-yk0E`1dCcBSix>t^N#ysAW>>OMi9oU9D8h${Q9GS1PvS7;UF
zr{>C5e{<+likI2=Ed4J6rVt~ZHIAQ<2`%wS`Mfo1q#Xk{E9hGrVM-A{j*N{Uq>Uj6
zU9xi*4@=>SIN@E_@#thvum-14Ir450J~6zGWX;ZlS1r<*D4V$b(gE{5J*MY5qA{9<
z>mtn+Kn$@Nl(td8tFW$wesztg<K42jnGo$Jc#G37Rw8Ho=RwpMGq9PrFNXyF>&!3Z
zZs#YJli;tJ<+1${S|F6v|GZ>D*YjZ$4l&6~53?%W`t4*UHYZA5Ma>=@R{+dNvP(A1
zPdA_uI^Lh;4qS%QL*OHV4p8_Q!!C=4ct^gT`xkwDz8~wm6yoxYTbgMuU_9u*)lymX
zOPhu!)bus<x#+@Y%;iyBu%l6_Z0erWfnl;=LKJUVJs#Z&qoV;!0W62#n`V~xfj)Q`
z)e(S??0mf|uaKryJ0g14wk+)lZ+Nb=V_#XsQW1>t&<*BLlkZ!5p0(hn3XHX1Uq%RX
zS>w}~3{JeSNPg=4^sjcC$c>3Gbei(0#35ZI7_t${8%J=cMbO_2l$DDY7Jr*Q6(qWu
zg-z=5zp-!%8$mpW40nB9_9Fhn<H>H%Y!Fj=Z7;HC+Bh)kcr&kaa7+&+&E~RKL#V88
z@1Jp^G@FvyNhl=5nfF5rSh4y<D8Y?>j?w@U?wgk343UpPe4>ZvS3*vjzP}j*Vlo5~
zp^vw?<eL?hgC60Aa9^08NT6=CNBZT_&@@`!fHwnA4*=e*vFj}dW3G4sONn&7LLMD!
zE+lsPz`C^pbyQ3Mefl*`X_6=bQXKsCd>3q&lG9tTgf(vlDJdoMw}7#FM8&rQJ;3Pm
zCpFSl<!3~iJ~2-syAv;xV_nx&tjuu!6%S5O&1({Z*myhhMsskAv;Fj`;ySk{Q&NXi
zZz6Y${2!m9<KoInac9sR6IN`ONVB(!K1~sRAJ&eJ72mO_TSUfkpjY19#P-NVqW=1d
z0FnNCD<bsdv^ekolJE}u#TLq#%qYf_2rR%2P)wPCtS{SYN+v6H_jPaA<yv$t&;&Bl
zZQpfFEe7I`mk^f)o&4nlDep-p%fBV>x^1TDhp#VBwGB<qXvhBI{Hq-|+O3kEZeT!5
z1=neU9DHK!?`DAFyhfqB()#uXr9pEojmqay1e)oM`wwoS*4fOAUE?M$zzfV11h?F2
zWos-73^7qHsTWXc&C(bW%9a+6sm6dQ4~jybp|~i%tQhUzst`SDr>hs0>=rc<fpC#d
zoAIQa8zn`l;wA<f$ERDUSA7!Ma${PwiWN&@n3ah`Pnt~(ViwbJNTX!p-6E6tlFlf-
zW5IMd?(=`)gPCtwkl}qzH8py?dviXZiCNdG-J}I?hIea67MGTk6&*w}mY;EPo7@DX
zvoiXJG(kA~W8cJcpzaL&btYw@_X@k6U@eKSI&E~t1a3HhUFseE>_4gW6;IrYU#H0X
z?B(kf+tAr2NI1ecH#5&i5VzOX9&dM@KmJcd%Y4+)(7J==K9G~Wdb$rg24w2=Cs~_r
zD)P-M=tRc|Q33%m4x4xCn~L>VwV$d&d99NK$G~i|(wzF&vB+qdVO5Gr1t~fZA15xG
zU^PhzjzPDtXERMn0(lRltGg)j-@vbdPh8{@TDSr_G(xH{gxAERhsJBtEyj9Wk6hTn
zBsKGCcec=|MV?(opQb5*Hz0?cZ+DwEGB&(~7W}ZI2T%K7ynUG=6U-As**OINJCV2&
zooE(9l(P=v9}gk}!n2U2j7G>r*1ogzU&&_pO+iF??udK^8|s@Oxe@yqwMu#hf9-M;
zd7rptOQB-7r0<wDDPe)#RFO9UD-PX67w_=F^}I&lZ%-CzCPb!OW?VvVI;~jXzqu7l
z$MKK^_!<&&9^!cHH4l${f1Y`=`=B5EBZIraS;>ZFINIy$^s5tEX)tZVmiuhyr8k2L
zV8${|TkKRjSksqAO-&T+IO3BUCx((w4eRiEJ|?St@4^A~Rge<gz3*QvF8nFqTu1Jw
z8XPW?I$qq8Q&x$HYkNO;OloM&@F=WTX4mmeP>Yk0N&$Q2)BMLxgZbw3h|J<EIt(Ju
z0`W&p^SK27bdKK`X)SeEHnk1Q*YiG~>uED@CT%t1TPAIc4tlbQYvgMCjMT>x4Zy!F
z@YgwMZu02tR)rqCv?%5ke%8Q20|_U$$n~T=pXK`CQl%g-j>|!qM8#C0uC}c%Rp*o{
z_C~VA<q_aZ0*>cf(D~?m4<&Lk`)6(`Oo)*16&m%wqzmn^4qJOzYNTnPN<3MRM$>Rr
z!Fj|L9k?{fdf~19OK5s_jy5VvRnklZKFSP6tKuQ@p~pv%`O*ZLuQDBVOOSpp?6E7Y
znTG-S0AvJdu!XJ804)HcFql0F3rm?A8F@jQ$pc35ZrS5AgM8_Q4=Fo4wQ_K9N1F<D
zQ#!S`6jhiNHhp7tU)u-EYdm103={B~f)Fk}A?QihUg&$_vqZLIe;)Uu^MDjth;?lH
zsexm~sU@Q|%J_yp1g#qs-N!(gDJ{AKVp!5t-x@lgt82EL<E$L#oynt+q#m)S#wo?E
z8PfA9f1hE7rH1Ojo&!N&3$%je&oLZ*EmJXbUe|0jKwD<KbOZ4O5Q}7_!i#Uc&o<d|
z(yjY_U{Hj1F$sU5gIL@(S<g457%lj-piq=Pv6tpZNf%Vg+906O?<OebdMC2qyg0e}
zENgMk5ZH)Z#lNdeo;+GItFgYBgZZ`%!k<Mgrn>j005!T9>W(l1+5(+hUiCR;`@5L@
zq8%_Z%_Ph%102I|HPdJv;d~1artL(%>|DWHt*DMs&D@-9yl*~v;|vF4q0YN$+$1^8
z_jI%SnUkVsYGRyNAC1>{Z;#=$93UDZ<L^vX*E&~%d_?lVg!Ot1p^?6g(UVRvbM66x
zCijlpLgp9SVWoY23%b1*S?3X!lS@EBp;|iLP#%2O)}W-~jrR{rp?vRk!}`A^EqSiI
z$T4vx)Aj297kh6RRb{_*4`UY!2nv!Cf;0$7vsFN(OS(isK)Rc)qJ$twHz+ABE#2MS
zUDDmL|7)wybIy6rIbYt-FXJA=F}!8(eO=f3tu@zNbImOT`qeWa+a~*!zvA*`{O9cD
z`T4!H&=|@Eia6>WhySE*sy*oMBX>!gww;ihQgR=%h6tTJ8$z9P4S!E!)_EgR>|WiA
z0qD=P0;Q4;=p}bx)t3|*g`U)3EBlGK0yFO4`R4Q2ab0o>3sH_+$uMDaADB&@nyuQp
zBG)#!LAJ!>kefa<*j9wF8}_AdIASe#Fb-vUqvlWTU?is<ZJ4ce^FyzpClEhK(-d+i
zK~EX2qicJ^jb4F8gkGR=adL;@vfJSQF_jSYcU7A~+}oR2%xbh(p)M@h{-V?FBOgw5
zE_br?RX}~tBCCzjX4+s)<`v5;ynBN`RdR4)O=+`PaQ*$CjrX72|4?vxW2uJMWZu{*
z%R;M$yUZ%T+T@bRgcI3-a|%(pwVFTZ(}NmXB{xOITf9|-ghr2>IFR#OHHcQ(e@sv)
zNv7&gBh4>@w&E#|i|L?^ka=-BxK3!@WWP7P${izVfs9d+LXa}!&(l1sJ)rzdS-L!5
zNq||ys5N{0VQ4x?9H_JC`v(FKRYOduLjRb^RA{x6i@d?e9V68`c-g;a@HJ*&uq@L;
zTuDN-{5s(Z*@MW)ZOWLnlZIvP4egSyMZ2b97ovZXe!fYyxr**KBJxS^w6*+Lq%7#r
zi3;U#V7!V`=+Q9CZTX;c{zy8Udw8&Z#{c+ya&CuvM8ON?e^_0h`=BD)p>QhOTwFHe
za8ZMw>$&p@<24|RJJ++@C++i;wB1=z0hG#UTa?s(mqf%B`pb6mXD}qr5LaKrC}5DD
zqXpGR0SJ<rKhYpGW3LxI!_OJ6GzE+h5hu?<afwsdnQ+EU_2u@(QXanJo@eFPVGf?Y
z4I(jR$*SXz|LFEUn^>#<F#D^aiG_-fQJ8JT%i2<3#DBz=uq@I4OEI^RtUJ+R>h=*)
zO->R62$~-!Bz947_Te`A<6oGII}yFjMD5}__zgBgK4YmIMJ`%!9|(KV?C_Xj{*?_*
z2_132JojGGe`h+WcDY44A%XRg<#lhT@321m@BR>bfqTRHZ=6&l5TxYM`E9_KM9!@|
zB!3y|r?mtYIoGoG64y@?b!=8kM71Uh{bt~^{U>jkT?fptJ8TMnq@&$|do&A2r`0u#
z8XLiO4^MFZ8HuKvlN?;X;yC5+hANLahMjDJ6ZWH?`mmInZ2!?rg6m$psg>aX=8jl9
zVPOfEERXG+?7}g3DeT3~$1HIq5C##rVwqPqKlLfr_)Bn%m(AxF7yKs|o7yE8(wJ=;
zPqA60kEFNrSE9G4eGrl$rK!_rzm7WXUvud4G4A~h?te~yyLcYMggg8fc~|b8p)X5b
zG~1v=UuU7eHF$!jw?z)6OH#Z2^m$?GGz%wkGG6-89jN);7el$rG>VLiQ2}cCoA-$b
zT{@Orh#17-WeO9!zUJ}-gZx&j4=is$J$?h$W~3HAINUr(l*-<7*rZ!~CHj9mA?KaL
zdZK}6nKY}JM*$MDFOK37jendpR1lv+#ILsYV+;J1xPq<6a7+IwcV4j<M9tZmqHuHy
zwU<#Y-?+@{*cQJxOkj8@B1EO2)I+f9_^YMYh3K~|bhLkOc>n!Ur}{aVMqjvZ>_c|>
zvo%&qx{r?bI6v*n$I%)E2F;AN^!GXiKP6sh0GWoEIZSP6y5ntZ37-Xf5Uefo{d_y&
zcS@DDmtc0LQhqE(h|ksybRmvGZr9MKWmhSj5HsA-?)WhnvOza8al0m4C|oQh=c9Fb
z#4^q7pbLYLklU3`UVnDc+&`ul7IHCF%H`LSv)Oys8}{0MT)V9n|7ps#=_B((f%)}G
z@ukwLPkK^^jrb{8jf|f+MoK<<zFE5!5gNAofY_sa!PKj(_p!IHucoo7qZk;ZFyER9
z1IExV>qY3sM_=X@X$om)r2ju$J2cWI2~GdjsJ9U<zi4HrAM}~VmGynK$tU@;WN1tO
z8Q<tQIC-gn{DT`VPT@U~EmncodY$u!#@+=;kXyIi-n6m_<wDATf^7}lhsJjbqgR>s
zo<ow-1nE9SE5{3)@)K2XLD~;gS4r$wKR3PhLf}x9U`d)nXbrv7<NK`iMDgY}%vOXL
z8Yr#xI~bylxislu<yqK9D?PKn_>?Ms^5pvOYR^gZ)b#EY7sX?O!L@et{?ibZ;M{MI
z0djl4u@ofMk}q_Js8b=gI>C|bPMoo#CBN(8(XjA4OJ0(Ok^lek`mt)&`|yLok0Jm(
znw}RA!S;Y#RCZ2nlZdf+m(E}!SA+(vGDh5G^GU=0?(;f8UBg9xe|CO)CS?=Ir>=$s
z2if255*HXPPppwQV?KH%&4mB6cZl7pm4BciTe5N6l^q&wr^LzYH!uoNdt7d?NyY)8
zUgK_T*L8B(?AustlQSa2ndL%M=3-J=(%YEn#g$et5&<2UE6KtQ0bN{6w5Q8Z??LQy
zqk@Lo1NPO!FIh`qLIF*Nw#EnX0^9P>J3Hb%c!S(wJ4{-R<*>_^+JB{KeVy$;EGE4V
z_G5Bz4`91MeYN<NvHW0+5aa-x)3dID0S10;G2($&+v@KdOHqo(+N!04BNVK0V2w}c
zIG>PGE3`(P`XZ`{NdC}r)PF2>?nX;}4rmg`f{}B;kYyvwX5(CKP1Jv|21%n`DqtPj
zXa2omylIIg8n!^4SK9m4o^OnERPBey&x#ktF4F2(3tn13C~)cx+{MDNWCN*V?fu!;
zbsmE%o)*HM!Uom^&tywh=PUoS{}D%Kp$$!zS1>_B;vvDo$e<86q8_dC2VBjZuiDu_
z=VK0T1S<X2nQAd;vA!@9efH9;wQ3^bzf|L5$6Z!|Zsh}~{LJS{&N5Iky(44jzn3aK
zMU2)Bn+lTG>G7D;KN%_FH#a^`;gEiFa-s+-_~gQ9KkO<xA8M_<Le|OvKSJnr!h)nt
zm+s7;4>?a_eht6SnzcA`<WSQ6<1dncF>=lSX17%Oq@pEU@HnOS28`9|%Bh!Gimg6^
zN8IxBkIBc8kZ=BHuE~Xj&hb16w7TFCl@k;;eH4GXYMR^0=o0zhH?$jJ{j>31z^*dG
zrxvT<>><n4Ly-44WDi$!8guGjr?>JvlR;>ArH>dJu<KkqxS0$4QgTodJ3fS5yH2Tn
zk|?@_oq6NBKMlgy>YMg$TEzTUa10V=;xp~Icvz}M9tz%Y3Wj_m{%`TOkJhC9#IG(3
zT=A5*z{=RAF@JbOr=WWURueZYPA<SwfPftRs^s0C1+=Z##}&sYP{zA|Y)B6v1%D`X
zBT8-9rTM?QFv_TQsXw}~xXH!NjOoZ++-mHlrKHV>I7i?IbU<=TE9U$eg#6loFaBw{
zpL6MJV1zEP-Zc5Mdbxsp1#cv?VKD;_6kaBF>z$tN6{}J&?Z_(v92gYQ*a!?hpc@k@
zHz*moiGNe<Usa@}E`uqw*YYrB-x{_11bA&?c|!Om<cvOw{WTiQMe!2<02g$(83iK&
zRTnd#p*3^t6hXqU@C0nxTeXPL7{O($xUajK_D|_(x$Z}EFQnR7|CnAZ*a*8@brYXJ
z=*hMeO~DK8*wxibMsiQX#gTjGU=q6ch>W!^>>@gfqv0-GZLbH}%+s6vWBB~-JLo;x
z$~7NJ5HHyElAi|njQ?p&qz^sqU@sihrFI3nMhx4nY7>1cBmK?pA5M#zUeB9L1}81c
zyzMl0zS!PrdkpQ2<{vG^M$L^2p-UnlQ(_vG^7_1`<^R1g+j9Ha>pvGUTsyEJC#K{M
z3V1#$-T1o3+1fLUwI*vySlj=>;$<~CoY)bjieLcti~IDk_fTi>|4K`HmlQ7AR8lAu
zG<Q*^jaz+);-6Qs?yznpnc~-zy18)PYtgFYel*`|CVRiH&GB3vO5MXF3O-mcfu|PH
zX_`s<D2cakyX@y*o6pm+e<ECB_B+gS>^7%CQ_KE1BYC~|;w6>Lf{g<SN*CkMi1DJ)
zCQ~1&zit4a<*y>b`*Yg$0ywhp{?*JzvUm6He<7rmNbI6vB4$Ev<f<@097QE$T+pa1
ztTjM>YDrR_(_Tm0HiywVyD2cx8Y&9pDwE5RW(1Rk_;02=@%z}^8<~|=;wvdhQW9L>
zefsQwerUBFv{G9Q-~4(N<1H*fnc?n_^o9GM^^#M08a{K<M`*LkH`SMBX>ko7ku;7(
z!j`p9@{nqrk7Q(OlZ2S=AUP{NrIbn9wT?pv8X2;eR3^F};sDR<OkpT+&K1#3t_ele
z1jYg&>FW0AT<}o)H(_bW4~I8Cy(i0xrB$vk`)IZD(u3~lzE|7HH?hnvv&2PfNvRXb
zVzSd6{<Pej-?FA&f*DSvoSa!hfPGWE=5n~Q`38unnYFB|KY7(|4c16-ss`3wV|GX*
zJD}%K0O7savD&w1lwBUMIOyR!LTBErD1hE)`!G)ItYTRl(g0g*UV2h>O}oo)uaI1C
z`A<$;^0&tR&5Z}KQGNCGQ3}vlq%MLw*|5kYoE%Y94QLCULQE`_e3jBOHxw2PacgqG
z8AT&q1vXSpbRq{fQmT@Cl%lGPh0^|QQWa(AzW+HQYG!&X2D2S;KP05432j_3<GgWH
zRf}$CKJc3xjC9><H!vxXzkE47d;Or%yj8_Bc16NlPRY1O9T$<I;xQQ1<PJoDV0E2F
z3oLn{AMT~Xp%-nl4&oAsGFdvlHOo=LmeyTR@VECo43)oh_|&I5@2~OB=GIf58b>v<
zD)XAdR!Y!~fcux%ipGEU(vc_Rnw>hUm7G1p!=CmW%%W_n2^IfpGN+ywL=cpo@>)d5
z2T3J4%L~Mi(4)8B+F5X>Yni80#JoGdZyHuP?*H)6ab7N^wNS~@AafkYlm5BoJnME1
z#w#g$@godtG$TsVgPUMU;p;&KH2&`qEVy725j89qQoi$UfYpR%MQpPC&Ni?=!U;Z4
z4QiDgY`;$mSbvg-xoi;+m^)P5jgl;QR=ewJp<5`NkxIRG<PKXdYL6i2+|af)A_Vgt
z%c1r4>w^}+eo~34pp~?AaSvH0MuX!v&a;$8!y4h!{CMHtBW(5^HK0KM$vVhNBR_s2
ze`Rr2@xw7>x;@BrBp}tdoBqT1N|;T#WO)%xV<HpsX0hb@47}<OabO<_m!aqemC5WP
zk(?Fd?Ri{$LX+p>*Iv~u=XiqbLD(c^FXPuC;S>CMQwmsrD25UCRia$>uxpp2hhmSH
zC14RVw;@k~`BReINl;Pvl6+vfRc+3pDPLEm6enpf`^p2}F!CFJT7b3(4lTxCsU04Q
z*ZZ3If7(LL6_N51-3h93GtH$9N?;nr-yRPA?cKG7f7RmE2^*|7cfQh#VxVoRQ4|7C
z8&*)!&q%y$U)WtP->AO{<30V1ccQ$(ptnN@eMWIR$5``kKR(=ufwoA{=wpZccoidA
zFq=U%Y`12m8mX>|2O{^)Rfjr4HaFp)AJd1^j{A(|$3kU^ja%-Iy{+^9<0Z>tz!#R}
z5hgGi+oNSCD}jJy%s~Jf0DC8bb1nDC-$a1@>j^`F!wQaz<UrtUHpi~hCm`g`q1man
zD=eXUoWW4Kdt>9bo)AB`=N{<UyN!LKH+{y@D#8I0Lp<1{gqR43UuL6DNH~xe-kQGV
zR<l*Zq#VC!v$})>qpYg7=&dxKU;b8pvUVGP_GcwZv!-7EUt2}+R}p2lAFMI?w<%ct
zbIGZCMIpX=x#M21QD!mS+S>`Sq3j#tJT=Ia>4MdaC@C91WGvvO0ij_7@afbQi_ZD4
zCPXKqbFDNk8Xp7)NJ5(2Kahgu%M--k_>OSz>J|0Ad)}_gr?xR@);<1ZvFOrLRK-RH
z@dml!vwt<0WK$Rpt>dX_-d*M@ypICU(s<E@H=OFmJY2Vfo&MnlH~z;B?wWkhh8irO
zfaO+P*BDP%CEt9SaK?Q${c1tn?^~f6o8*R$`+1X%Mb6-^6_j98&5t#8;l`h=vnBi6
zJ+2Nt_O~#eeLUECjK;J2=T63?J(B+J^VDp=H%Yn&FPSHObSe+1Ly5PBH{2<&Y;3$}
z9LXz@!Jv_3lZ(S<_;3TH?EjGbQIeD=E$YxGX9p5QbJDIDi@uyoCMaZS|E^g!<6Qj=
zcA@|J9TmDrFOuymMnJZGog`U*jp)y5n4|V2eR(13ahZ&<?8+ddP%G~)Wz!zFi=|^X
z!6QTR_!G9f$x-xkCiY{N0*jqM$$b&&z@f*2g0E>g4U5%Pf2vqwl_w69@Fu)_Mxqj=
z{ZF$k=KkK$UZ4C<Qx@h$J}V%V7_63HNA{rMJ2)cIfg{p{4*g$f=JP*0W}IwKTiV0|
z>6;nAg~1j!*NRPY%3M;gOXJ+fe{!qfmr@EF04+prNMk#dd0$9~7^n5FXwP$zZuY64
z^g=VyDU8~2la!B|+CK+~UA=+sZoruAuIzR;Efr4j3s%@}8?pwKrCGBgwPmwlN9CVA
z{E5DK?g%N&`pe-d{4gsg#mjl%)VnGb{QRTG#>-6*X#8J0*%n&*!)P%juPMZ7h1SxF
zJya7@K-(~QT6+?ih{NW113b`}Mwz@mM@>h3r~QOW1M>&;xN9UB*Wj;=!V}7Eqk`^l
zFB<8w4O6n0+YTUyEBuf?Z1mpV@FrQAftYV;)b_;g!mD@mVz{u4uGPv>;ZXXy^~GHQ
zo%D!tr;wm>LlJ-Ml95E4+#*=0DS7k<x|?W$)hDR=ZWpF33ri1)oI=+Iwn27UWDlHq
zu4ogD9&7NX9Ve#CkL|A&`TE9%5s;Auz;1GGs0*LsZ>gHK<KBz^;`wU{R?oGU|Ah7t
z5wFH452srqmhVIT2!C@Ou=6UsVPc45P}3GAJrbTLC7^#`dFV)rR+>~^VlKCSF9NoR
zULgcE-877H!W)lgpX3Rfv-^5`Dp`|aW}Q;s%&+4rw7-VCV~co!%hs*oc5cHC=Witq
zqv{mIY`=7WFwB<Tnxh<BI;j<`0fhd_3JR#c?Du}N(O7n-HF{+B*Rd;`)4)9MAF?pN
zP0O2T31mW6+j>GPR;D)Qe=%#maOw9YV#Lu?<Zy@{C#b7(b)VGHKmHWQEmq1^w9*hU
zt-IqeCGpNgTQ5!CW57!J@^~n$8=53yKXAwndTd~;T2o@_*)<pdK7*fOjKAl?8xD&o
z@5D)p(pi&7hSsaA64QUZ_>&|sv|i8DFto+*?$LOvEU;SGd&(}APC2{sgAww&BvojE
z>fnJ_bPse8HgPmFuhcqbQXXg}6it0~`gVASmPTAAQ+=1m?2olhA&emm(1S8LBGT-|
zS2mLu{;bbqXo;Al&@h=L%O<CP!r3zN1O=cMs2D*8u;v>b#VUWR&vbFJ-&AalaWC;c
zI*DBm{R-=JHS15_+Re>8L--yO2Y@dmOTP7+k^VYx31X&kS#c=Fb~L@ZW-9i2zD@9N
zCdOE&^I+KJ^w5)5iJv-l6_ZR5Hc2B(#@fig52mV9oS`H|qvi%+p=${$yT&8wQ`TGy
zK}J>wS4|R97qQ<Xz~NR4$y=YdpKK)liuQ;q|MJf^IpI{?X4!uP;usQEn`50)Ex88N
zApY)tdAOmRyzcaNXiQ8qo$xyyPP=a!VcNf#FgRvQwGcY&TmV<eh+4Kjuzff{+dlL|
z#>?CaWeot`z(|z_Z#q1je-ao?*Z%cr?)RE^dYDE2?e%Y?#K@Stx8{$)GbgQJIOMPK
zL{%cAA~>eqETHR2?CCFp;S9~OdT(c)SJTHg0dgwf)ysDatqB4R{WLNouJWH9-O(Nw
zE`wsJVl#6;djC+C7}WL3BG$Cb%~T;Fr2fl;=3<tw-a-;Tsd1TOKrhr?DV{bOX;JYE
zkLB9o-E}x2P6)pWc_iY06|hHXnM0PI_Y`(aif`59;A1ZXCQgGG{Jq!k*UAI`{BODf
zL*kb0;T1?kUPGOymtZ|bD><DCjH@%ixO%{T%k#IH$Z%BE#}Cw7hE;V6Ei2r{_sAWJ
zWt*OP$wo4YzT8y_=o*>OwG<BWn1#nCbL*we4#+J<>`?dG>G(tLY3SgY0BwvOv`cVE
zc0Wr?B`3c-5X%#O(+xG>`zYb`Ng#TZ=j@S)_D->!FZ}JDvO|Qm1;bhJPPxo;0&*3!
z8}77(tGen7zvV(tITJ<iu8|kpbeMkVy_hnWnoO1Xj+K#+rc6y+P&xh&KEU{?@^|!W
zN5`mW0%U`_CX9ZIOIFFA@JR!aRAb6Ayc2vRr(AB&tS{(FYEl_1ecI}#DmQn4J3Ri%
z@0tZx(|nt1eD)L9hg`o-Y*>K7=}+G?t8CWlfDtUX)(l<868J>)$2As%lL-9V^bav9
zR1UwXtL}f8pT<&b&=u}x?|&Qe?h*OxL-=hs$Db7w_*QUnklu^i3vNC==ZigYu&<vs
zqK^8z<KXM`*-mA?GE+kuLd#}!^(P+F-Cq}Am)bQT{SV}vpnm9Z`1~%;!LwWU=mWRr
zZ-@<sv3|;_<?@R&RT(d=Nxy9V%T`5hur^W4z<*wVbtTUIkCEGhEJiJtt<gj0y=@D_
zz7t;j(vzRm&02-CV8)NevJ=!+dH0WVVS%7QJTSGHQ7pB3LPK_6nX+1#4mQO6VL&Wz
z{*XwYLU;}Q?y#<;cObU)Y0H_W7~(jI@5MHM-}+-V^Hwc%WK67Odts|Lzt5_DBhD*f
znXPOw!Ao-wi`>nFeRFjUEYo452_!1w1;*01@v*Hu)QUBwg%TKG-uOFSMh>4ky8jSj
zS29Qgm%AX6o9)WeGvO&1>3g4ShqLQYR34o~9%?vXG(4i)3oEv04`8%cRpVj>Ab<3%
z+6<RKN{T|j$yFsc%3i`cpO5y|peT_APS(f{s=mjM2g_x_V(Q70^d~|yIHNxNph#@}
zJOOh=^zYhUR>1RAekfLIwLuGhbeEd$9mrZ3-vZBm+n0S>M5%XFN<Mt}96<uZ>Lhe?
zopk*=dNO8y4&x8blA;2fWk>#~?b&B}v8%rYaXVeY=V&!A`gKmgVq?Np96x?1A)N=m
z7xxppGRBENpMfh!GgrK$%AY;Z()Z^<_v~Pet9Ms12Uf>_ah|hG#}e{xXNTj?bKcU5
zuAJ%k#}8d1o=|vK^bvDn-SdpA`4K4^6i>APYakhPoRF)uG`?q~)WG2M0<-P4;+Mmq
zd7$Aa7=ANQOm+@(4*F`>d;hbHj$QV6`Ax$w8+s;|P~dxtU0_awmQC4YXu)?lDDS|<
zq7{HYU!#3q5;0sucm0IvvdIEr^FA=~B83VCh1<XdFApvJ$(ax)q*MXTL&*T0OH%n~
z|NLGkb}|f4#{Q~BU5M$mzTbF(&uFt`vzl|2sDM~@gp@w1AN@@Q7a>65Cj+ZJEC8H;
zW8F^ok4zn#KCLRTx8C(6#RLP5|NrOz%N-brC%`--A!KRE)*8jzFQcwbtfsEMT+K4=
zC~6S?^mUH?fYHMwjS{=gQnSe?m~<7^{t*%4Y_D5b*f}_AA98RUc<+@a_b(Y7ZZB|$
zC+6|(4!k1h7_OL@xSPd5G|<&`Q$a&R60^F)?K)9|&VyNO*=9RL^<>U-VRAQJ*CAE4
zH~|3{^&LCD7Q4FV3JR+EMt6+I%lJiP1B1CO=U&K6=GVNs{DPvWxY(dCO(n<c(<h&j
z)5E1{DXj2BzseWCMzVFqg-lINX`bRu9MUo}Ce(W2w+mR#wfMEPw3wG8oSmE+`uq1}
zCA(Ew!V5B%myL#gn#Zvs5XfO9Y7dF5f0K7nDn@`mrMx_b!0lUMSy|1sCv7K}ZsfA}
z_0wF6km&1Le4|?K!hf;7nOIIiVaIl@>}WN3Y)q3-8pBUF%w}y&@^#LEtV-}k*}3!B
zZ<!ykusr${E2E}HbnDhF`yB!j5)vUHc0NAuGZ%jP3=I#9CUwgS3c4z-wHtINc8(NU
zigvSO&M~HBXZvqExGuQv$@!nI4;ixFyO&T`Rwm%Dd%%X2O6nE{_1;tCq1*+g<}&AY
z5~nFocFCJ0-e0Tl`kZSv->x`^#kIRSI{aKyvuw62L44Z9dt%~+ziyb2Y|Y!VXD?Ua
zYZJI-^!xJ%pPtFs((%ys)K$qGSHrBvt~bAYx8Qr6h;LU*7~2VIzaCMxdAO9RMo36l
zw%z_I_L;G<anIP8Ns-fnoPSJA%ud9{hHrmwZ(h1`)>&82HMZXg0At?*CeAml67gO3
zij`37?ucbeF5>>(J@(ISx_SY_(^>ljn5cPm8ykwJxEL9nw_%MO!;I)7mSmhp@m|N#
zT0%`}#O5l8yRS4o9bN2r`L>+;$mxk%x+49>TfZw4#)Oxyu5PE6RAi`<%=6=Rd;9rP
z2TTQp<gZ^Vtd2n`x4SZ251&i>#hHnsN^oAqcsty%)sB!J&Nq&5hJP@K)G!ht#b?%x
zUtL{Yd;R+L>0#B!Yg`G~e_uZ5>5<I&NCfh<R_k~Qe~Wrk8M9hOQSo+OZtj=n*{X@N
z=nSilfTxN_Ow4z5vNw*Jy+r235}_TsLF!m@vvW%*&~p2iEav@+>vL8-dJ;N2uOvi7
zE+K8zM<TIMoA_yOe<x{mL<qB1B7`Z?8$-i2g#G5s6K3Bpd{2eA2_)YVxzBPH?TtB-
z@Y=2#QN%Lh+A~OYH`Leb)m3k73=i}2q^PN>u~z)<RQlax@5dVudd5Bz6ci-DeycVR
zIhaGs!t&NGx|}rTx)SNhsx3lDo)6<^wAt0$81OwkI)}LiI;y$f?ch(`G1olp>WdD?
z8}$pPM{D&JzpoEUh~|2>?u`DL-K*(D$q?#IF5>#mwkY09r#o6G1^X#a8kQdzW}|3R
z5qQCw2`uEZI+YwZw~Lp&a&mH}A;3!=B0m#x%E`!dWcFofIA+Dho;`OBqyM_%BY5N{
zlx(lt6XEAnqu8eOINj7)d^_>B&Rk1auH}4NfG^+Xvp26_Q`_6KT=-MbDx&UkafKd-
zFo-&DcZ$RmqfU?RJ;tuPB|wTi-tfx0-7JsVnv0M^AP|`MG1Qd?-ox*iML@7KX_YMx
z{xqLP9W{Ovd~ZIK>sl{x*iev}=?jP0i{?C^^qDhftok)<RmdEdvnZGPGi`6;oV1Rg
zGG6>MsKA&=g^U=&E*U^kV02LH8p2eX)i93gIM|w>9vT`t8!1I|{rYvBY8H0OIf<)R
zug2;&2Xo&b=K6B`&Ygpiii)?m!x_J!0DOCaF1yfA&+n_Fw&N$Au2=#DJAG=V7iMSY
z2+0oms!=Uwd?W~$W2Iu76$}2&2Fg}=U2em^w91MzzpFZiL`yXmLP97o&Or6X=bpDN
z&xE#T0;Dx5C~nb{{%Kk<`bYIUWIVqU#ht<6JS$ZR^%Z{5r5S!6VF0H2r$7Jx_5b|u
zPzNrzy@y?WeNP#hFZT8Ki_Ff<$cTzw;b3R)2%=Zg*Vp&(@OTZ*>6Mk0xmF8u4<cG~
zU)A5Zdiin#d2BH60p@;X<>T;&#^K-M;wbwd84j%2q$DSQOH8D;C{(BVL~*dc|0Oe%
z`QYeCUSI!ySnF$>lfxa!5GIW#PjBy%gv=Q2&{WiXx+(H#7{|2iV9rs9YMd_fo@B5G
z*O=MtW`eSsjZZt=_T;oQZ&fw5<fNqeMdug}!`qCSC81T-)dtgbpYd+p685^ySTsG(
z+0R}V{!|EWU|@j2damUT9RoxB;^Ldd<>mNy@6J$Az$YbSRQ~byy{jwjg9rXL%L9+;
z71J7I74meUhS@F;Xp}jHIXO9Dj*gCI1qB5Oq^GC%3=gAg-q$Z*upcroq%A;(<t8U5
zw{5X&3Fn|sfh*<W<WypWcdDy<y12TU2=&3AYGHB7pe>Rc4;%Z@yLa#KZr^@N#A%`<
z_~c3W;o+ghTw9bE=NzM13_U%)E8Bwy&x=Q;Zn(QTI6A@wx?Ay(v_^6VD5k3lrz)n$
zhK3Rh6<IH>LkaBiX!zW&a~?bN_{!Gu$cXw0J6TNq)vMPiac|zNfz&o(zcnYUQDo%_
zuaqejt*b&!5epZ-xU>{^c<4ZP|NhrFGN(t?Cbn&hy{Qq>NlC>QFI__3Ds^^TgZ#nV
zZ9Ed1^7ZRm#Y|29qmTHS9hqtct68W$)aiI)PtWtw_ZP0c%v3LY`D<a}+v%wbAu%zP
zsQ)vC-qct9gM*^h*49?5Bivo=>AH4yT=V0nE90R*fBtk44o2pgcPDB-#?CM}784Wm
znC4Q%6Gr&fv#s6|jq7~2HP^b-%idaQw=vbOL5XumM<qHvWl;%<y<QR=9Q?36Ik|C3
zp>twl;=|(NBB!IHW3h#WMQG+#UXy}|!=s~2DH%N=-XQ=M;Z^v<G(zE~8o*=9(Q;hj
zIB5!^Cp(%&p7v9x_NGMe8@@Jb57MljZ4OB;C<t+Lb9;tYo4SYoS3RSnlC7<+fk8pS
zrYG&FFCuuBPXq-c+6Rr83WiToTh;Qn2|w2v7Oj2${5igRVE_trd0APRx$nXDLU&_(
zd!|x1`=0+jaccfZA2y>OOd*CDE6am}7hNY~P#a{ZPGuF9J31XpaYaSN2Fs<hHpQ7z
z1V=|l=#Rm}!}A?5fK(*c<I>Vn?&e@dLzk1Cehv6I<qp5TxjQIjYC5YZC<IOaq&<ji
zXlx`W9HgBPnI2*NYCw2!cqnIPW|o;~U_wR~lX=3su&|K(x|MLcc6HU5Br3}Hq#JP(
z5FQXPo#S+{IRNLidVQS*DhFSC_f)OdLh<;?>Tp6r0=L$#U2Scx4hzBSy`7yqWvgrx
z;x2(^hT(n1)=PavII1ck&9i$+h!fEWHodmQZ{NOsSaV);q!YxuM=v%3H4S|w5KadA
z`o9+Ew+EKcp^Uz=*x2__L_{FR-;0v!e<H+<+^;56evIuuYGPZtSGLFh6#D~j2F~;^
z0wQ?8ueAVgE|H-D85|_lLMQGVtAkAw1%EuLB%=OTk=&MU4~+&`j?m$?Jvlk)don|V
z7_^#g3Tmycu6`F46@@;q=gysj<O3JU&@88=MYglEBWrEVp{u7C05C?)KR(xS4SZ~w
znVDr2cBHknwfW@_?9Tug%FEAxs9Iu|MNUasEvKavqO7i7vY>9W>n$F_bVr>D|2h^{
zw~})wCKi@f)c$9j^W=2!8jB|gl!``{#q7Czv$Lf+uUmPHvXtdgQ&NZnxXB~m2n!42
z-Mo2o2W~XeI=T#)7#J)h&c`5K>Lw<>Y8!POSXY`Z3wXQiUtF{*s(>uw01v-+eSQ6s
z&B{=3F`SF)V%s&TQm2FL=~NfvqX}26XPNy+JKNjwT7?$6XK)Pq9T-zmx;AH<2cvjx
zsp&p#PHLf0fS*_~afb&7!`R`AH-oe?0Mh_8HOv`2E+H8X(Q+Bg%+7u=+TX84a$<2L
zm(8F>VyK{C&>1VdJd|gMtDtaV_F&Z!Zrg=Rmqbbr<|5vq;jgW@nApRU&5-g)Ru+~Q
z1cZbXzV%yqzMXK>S5{VH?8kNV^h8Rx+IZhU9Gr8sSzFb})$g1fb=-~5&&%UJJ~=7h
z&T5P17l$C)PEAb}&rD4v+H=ObeE9_>B_)k<iNGz1Mt%0K{?YcVJt13L+edZD^5%e~
z-Q~KwUibI)w5*U3yr`%4(9(VJ;w_VAiQNv;k{Dk>Ru;#4b4Jt}k}g(me*W$aOw8lO
znA3%rA`VIMrG*7^>0Rc*%(6qe&u+d}6omcGa(EvpC@2Qp$y~T(gJ-caXDt^B%1?|Z
zVo=BPYdt+ZMyZ*ZoVz<ae#0w?MMXvJCsB~<KYjWX3qQ=xBY}Z<;Q-W_mzI{wX+OoX
zM;q#?D6$ZKUh|B#GfSbVMOP5h(w#mh;b0Uu8@=Jv$B*AV_KzO~yf$M~yr5d$133tF
z7=9MIl(PVzJm>A3=UO911zlaw2?z+da|%CA99pq)A2ceXoWl{KisUk%(nBY@5(tC-
zNpa|xewvXN=ucKXr*L@hak2iW@E8s!?fSGY;?zV_b1ac_%<h%j@XABX8#i25oL85Y
zj<#(>D_^jT#Ds*1PzNlOL1`z`58q8i9)utl`q;12Q(nG1kq!vdcXw^vl+R{axuwu@
z{wAbTMu6XL^hy~L`&suy36D75S?A}sH8(Yd;}a4d%BiY`JCBz+)6TcYNI#5d<u6!d
z7J8-i2I^-b9B4u2<5gR8;o90W3JMBZL|hrOS^T6}_I#v87PD7}J33gHIXJ8}?d<l&
zX`~{5jN6Vo(?TU_Yz|>YRc)YST^~xbk)h$zaejXOh3S*M0;9nvP*Vk7(W@g*4C$&_
z1mK;vii>H=C(UhbIUToubsmfzk3tUwb#Lp=T4rv)#dy>W_vFN)!;UpAC)T#;7&1Wj
zPa$9Q&pLNQ<U-a>tDX!E@$+Lf9WAnEb~fAoeiH4tDoRUBD|!P93srUGvb(IJ`s2l!
zh^Q!j1PzVe$euI&ORCH7ePizjI7pY9sErLU_ujjAuk6m9J68-03^wxe^5!--H?Ncz
z;0cq5Gz&9@c%f@t$u)eY7PTKQ9?O}UW;z0VQOp^PWM^mB;kBIeURo_)iw;~uO_QOX
zv1f0xBH7=1@9{r&+*{o_$X@Rq>wCI2*Uky(QQ)K-9@{`mTH4!`l$6ZthW!r6F~>Fj
ztC*OPK>-15wk4VNKGXV@?=R%2Gxf(KZQxkS3kiMLI6l~7{rK^tr+r~za8)dYpoYdM
zhi`43RPd!sm+D=fg`Z5ilXZFG(%g3Y@S#M9ozQY>dRh%GV+`}!wPrddCL3W9k#QSK
z&(6K&zU=I5r2Xl^JY$kn)B}kb%bNCfUs9*_%I&$}V7(N>Hy6q5XZ)ZdH8wVGuUBIc
zX*nz;gnEsRiY_Ecg^7uaXD+yZ%`D28oYW>HB~4siH4&7R?fe9#gr-oPX)hXxc}JU7
zF>tEt>+GC7TpB(57UQhe<~5(p%*@G|#(-eGXnX1nsJ^TKnxBR=TT7OUIS2;p(g76!
zsL#^g-hN=L)Dai53>M(aaVZjFVyh7`$82B{8dhu~5XbA)``t;>G?9GvT?M73QKivZ
zZ;-}zcA;>+-*a<&A_D?&cRK{JNZ{HXEnd8+IMl`UXuN6@a4-<Cp`IB62|*68)`^Fb
zQh<?>akBbVy%V%(gPu~K%0J!z^jul_8FY+uRaI3#pxkh>GBK^185tR+BpUb(uk;sm
zw#Q(k_7dvlsbV?rPr4`i_}p+EUWv{{=A@>2b#--Na&vQEdh+Cn4L&~p@akwe0@woj
zM*j#LtAd&(2L|GhkPx-gql0;@Xdde(3;rj=J5DI+TD^MZ6%9?z=vj0!n@W~XdI>lB
zGe19n!0ySe@Y*n>{3&~ZMQDFzm-^C+ZR>qV@{Nb{?-_)@j`|dopZtKZq-DM>>S#9;
zd9rklMmjpOgX`$vpi>ZwUkBZWk!61WuI~uFV8(*TH|YJEGhKF4YYN6)coWOY#<=_@
zHWW~bP-YV1;|U(Hv0+6vwhWz&t~u4`1LDp~Pj5X?pznE$VF6Rz&zh+27!MDI#>`Ti
z^(fo%JfNGZ2X22&K73)r6Xn*g>1Zr$ZM}o8D}AjIoQPxtRie?E<}f(!zX8r*j^3*q
zWCD@fZM<utboBH|%i_ii9>65sIbAA0HB?bn9%zPSXs4>BHEt;>DGA?f(LXILa0M{v
zcX-+v>V?^BaplLey_{oC-Oryt7YZnqTQ?x}bq@~@$1F#ktSR+8S1Rx68+tREZM`%N
z=h_(Ra2aFWK~J*$b2+&wEoJ3>3o)@zvk-&Z!l`BfJA(tzLqQ(o?Ca^NQBY7^9n98k
z&MCH8696xL$#OZ~!#nZq0=BF5rX?G7Y-=S)8S4w_RJKif!>OX8?0oDxqN1YQ`<0i-
zl;Lb)a#|4Us3ZdP=nP}i^M+Gv4o_y-pkO$CZ^W{l5!HM!Aa)O@m6nO_aJ>9<Cvyz}
z9nH>VGMcr*`DijLYikX|{SN%ww@>#!Lot|Z<s8R5b?_R|(Qwq$qbWVv9j=?2n)-?E
z#^F;2tGy(3#W>4ZLx$QH2}dByW)Y{;h|_h*NQb<{KD1GM$ncbuKwtoSdwU(pAQnoX
zI*?w$!a~}ZnaS`VPmd)zZGhjPF4Z&@gWG+0vW6rJ@bl9yAyqS>zA3c1wRH~q5~MN!
zGCZgoCrX<3wFj<b#~94pO9#S~o<b$qUm~JNj*iArrweIm4BcA&yplT~=5!o*6rL-p
zGTd$^J3U=Poua=1gTjJ_ifW_@_l_Xc)!FsQ>Pylw0>%ec8ZFPEBF6>pjp15sZMyVX
z&WgPsUUBJKuu?Vc+sGMS5fZA3fJSo*ZpbtZJkqM?ZqpIJIPfNTO*NOLq1SOagw#Tv
z)NNl~HeU7&i9|%f@sd$hb@&Q_Z(H@scd)9e8kdNOC<-1->1>^onc0^_>6ld|6<NR2
zoEVf-Q16?lt{~s0CNT<%$<n>Cql2Z)@>AT@UOI_6rVq@HqY6q&0o@BD!~OjsdJn>U
zjK<4O1_5_oNhQC0_YtL|4Vl9)^67EZ$$0&9MMY{L{|?6+8n=>mc6YPU(D@iZwTr|0
z#AAcCu~O8+{QP`jSlt>>3t#8vUZcFdy_LqCwj{+>R0btCN}eku3=OFsR+F9nFoodB
zG#V}1*+3p9A>W8gOV@jy@-#Fz$3>r>LA?_3EjzA@Fja-yW)gly($SIk<e(jy-xAI5
z*zdj29ix%|vK9%R5_TptHO^;oMVuc$uJqM%IhfTGI9?ajrdvFhF>B9Fe52abWh+7#
zYJH~BSP8dmH~TQ0Qw@MhB@1<H5Q4{|a&leLfOtMe7<BR9vaqyFPEBn_Ffy*pa%v(+
zEgrH3nDk*J8W39B+RDS>%F4}UH$k1AAYCBo$gxX0zVqNKXpx<cNjcbWly=E}z|XIN
zq*N3T*q?BXDcPU$LbNtD6@ANe3~8<{103oCxeu|ju;52(H}MYD!`|u|gnj-C@6fNX
zwM48jN9Kjd*4y%$vf7a;#@G9srZ+mWUe$kq?k1*KooVvjCDQ#p=tEXs+ieM;j<=vm
zG=|uA-T{;~*5BD_E<ZNCy=~Rp(a}K$@U@(blyucdU%$c&dDe#vA?51onmB$u$C+8S
zQAc#7s;X*IG|vSjF*><?oP_s4RU?jF*RfFW-zn7m{opWX0I&fb&l*}<jt|y=P>mtx
zwUPKaC=dbN$H^=+k`-=Cl8}(l!0z&3z;}c2t=IcTg!~tuKa-Y59mJe&C&i3-PEH3Y
zX?_Eo5ekr?RB0TEI*HhHhT`=rJe&}iN4A3ee7Cjna-^b_)pkz3FIfon(z@%?tXsd6
z2Pp+bWpv2i7G&QQ8crT3Zy>tm<mEr2?}#HG>7?5DVP6hpU^$?3%4BdJt@1%$VmBUQ
zM-_}YECoSZGX{Xrp<6&nL16(-k^{xzg7+OZA)jf1<;BI9rZbif@ErBc%sNnzMKcGC
z@}2vqt$s9ip%W)$H%=hBm4He!GXu_l98EfQ+-bg6ezIaJ3O!F!e0<3!Ae7am#l<3n
znAG~|Z!`T`{f0uGowBmDSp54FCwfvbWyx#B49#e6F0S?UbV^f`%ON_aByqmlC!iF)
zIY?fThmV2r>Vv4z<L5>Xr^}&S)Y)IWq*ZuS@veSYEqwU)rHi9q=Gkt$K#h2^R=TGt
zBPQ153*~IbR&Jl*=)y>6*Je+5J0C9cc;DPoTDqOlmIPYPZMfYFz+t-_pVX6K9Szr$
zAvVW&*ObW|_9r7YzxsBX_bUX3pw`eK!I?G}mzJy3OP^U5uAirBn$lmIXb>0{maIoG
zmju0X5K#JFzazE%m*`zIn4j*|LLK%WJ>3%@+t#83$aNFg52VS1u+vqjKQ5e$g?vZL
zIr05Vk}4`HnXh$qSBm|PPBRtIDU8qFVy&Y?W(=CjmW&LhGix6<?BTkAU9R7rN1zUR
z_r{UP@dWV@lf!vFK0b-e(%o<^fZ=uU=w2%-Dvb8*jm}#5b?P4vqK;;nIn{rfO*)GB
zl9n1&*CzpI*inueM;T6DnDh-kIVwlSDL#3wa8DL5C^R(m0_wOOb^PSbo5d2|<hfT?
zRkS_*y(fIMv<ozVJ-WxQ#<Xi$^FzLHn@8>ypA)REjzFK(dc^7ebVSCSe(p~-I1EmQ
zoaG0SBO+wZLb%*hP9^kYLPGHkLd0ZBa&m`@ylZ8=dJm-UMmC(ZG&S7}W>DE=uQ#9d
zr*if%RfNO-)K=t@WW+<`xeZ?dvBMQpq@(x;-_#i9PfrR93*CT7tN!e8gSlk=tnp}3
zFX{k-cowEJM<xl(1;BId)_CG>Z7>m(V<C>e<eQ8xqL$!E1YtqO3BFABijNd4A}GkZ
zd1OyMNeY<V*%N)}OidCRbNCT4r_)O1L;W2i#7b&vN1KpZepS3XCx1wslA5{z@sHFi
z7)G_d$s6McN+KA;)a0@FFxMJGW(6$|**zMXCDV{4QxlU!+cEok1RvjVC1m#qy0_lx
ze36h)Y1$E}V`*P91e3b^x_K)-*{k-+VqYh(vMuGR<9>gw2uHS_PtLu3b5I*P8nch%
z&{MB=gvHFW;Y!y_SFj(MI3i9Lqzw)>XFH*u%Wf-K=YDKAG&W9vhaP%*RE|1QNbN1k
zEGWQtwYXN?H){`F4B%x>=yFOA0T--Be4_TUeV`J&py+HG#qStQrE-cuqHK}<{^0pK
zi(4BUeA{8QaDMeAf!k_sdlI^}gi>kZu#(rW$>nmqUouSVbm8K~E*PkoMhQ3{mJ2_9
zI>z3<Bu`HU!<d74q{Z?8t9Gp9#ro3C$CU2ZOE#1G=g{QC@qA?D$Q3R8=4^JyU88V*
z$A=3$4lwcpa`hLGTK%&>{DA0cYHQ1nPe?e4L32-2@?>Xb@?(rxF$EzP?CqV*BTx0@
zw6!N&8jtt5d?=%&MBd-UZuSW@{GKBADfSDb7By|P+N2%zl4q93$FE6`)~n+I=Q?WX
z>CMNVI=;XAMKiN}DAG$;1*ilnQ3_eS1qYk;GEx|C22cMoMeUmE_4o8(H)kuY9fio^
z=~j`=VOD$J8qs7EmIV%Uifh$yPcDYn)(6NTeK-o{s4WC4C$q57-U4ygK6=}M-<;Ui
z`WTu=m0%QVw~aSrVcK~@n)*!Djo}|Um$@)-&Od*2Ug!ST_l{ogQ&rRRdU;fhkF0x&
zY87s&e~k@pN_o3{@9tTO22AZW9RDvfTStQH{ifv_)aPSM_m*nO7%d7%MlEgI`LA&i
z`_!}^k*%=#PuTM3?qPchRZN<#?M8n)5m9%pVZqGK$<D=XX=Pr4Zc5|Tt5+j{#zlJc
zP~-b7{(*t#A3l6IB;rF{7#J8h^YCWVCBJ9qDg5@Oq*yT>3q-Ky54*a%U9S^8tZ5Bn
zqxn+%k`>s0I#DdHH3l+>UD6w~)1?e#f;IK^uJ7Mhr2yOQtnGz4WyU3=pm2eWEk_DC
zpc`JbMY8hp0de2HJzic~s^;P4nQ3ipu}O?ipe?i8c#=?9c!#-(->vdR1$;Dqfum)%
z0)Xmu*GD_CpPf1Q<?;FRM<{*0Kw(6RiiyQ~Rp-5vpmqfc7xJ5%hK2?S8z5&FXJ=;_
z85xYX1?HE5<XxX@B{JZN5-85d@PX;?1%D$ZlBr_U@ujpR33D#wX-RdeQfAHO9pHps
z=<8or4G%YZs-e*`koD>)Ej|5ue0==dMjb^Uy(WPvc_ged7~MObsU|CX28e;1(~0l3
zL)D$sBVB;1A;GNvQ7-q01@j!RLq}6nFEzBi1Q!+;t6+xpMiWL7rvgL;kM3e{aB!4l
zXIr$6Yy67gcf81JvuuVio2U>l@BQi*65<ZorHX)&(Z*U*vPD_2+U5H7&)+0uI{Nx9
zLUw(WAh*Iz>|<hY&&@)BNw83|i~0c9|7&sathpi$iVYrKMNQ55+}zv&U`ts6W1hgI
z>n<GiubuiAfq%m$B2r6^jeYRBdh%D9{nkYwSKKv{r!5N2P^ZpCweg~QCT<_O1M%<+
zzT!SBf%JI^$e-Ph?(W52%}+${Vn33X$KBr9c?>rjUbqXW?^I~jHVE9#c-6j#MzW)~
zGpl}bqey}p_syF(j{;x1Q&Us_fI<0y0!%?*zM>-)@lcYYPE8J?WoB;fePm>$uCVY&
z8C!{)p67RWcbD&Z;nxoh4}SFKc+eCS9K8Cjvg)1S$6GKlprnxwiyi3c$(@G8N1B|R
z%u`h~{gU7!vsPKWl9EDHB@5;XE(8kK9fxEt;JrHX4Ewho930$sZass0_rk<vY;JP0
zvvJbQ2MPv9M0j|em5og{Qp-$_1tS`QgX5^~Gtmqn?WNPx)84SKuqL<+X3ux8(y#A4
z`X;lyi?qB?#n11o%qRkl-PhRIE^f$MK9Iz&J3F6@!1&5s;ra8I7;mp*IJly63Eit7
z1C`YvmprbhqyR*E!u7zQkaiDG&u{S0zC&j>DI$KSvOMp}HbzATa&v8ZYO2CvbLLCC
z?V2WXq3hl^0ZZU<TdLi$Oz$%=c;{tjJ7ku_H|@RM-rg=Rm5`5*U-p5yBZwv>9l5kP
z8g(APOy>$QG4T-e?M&@aY?sz1UbF`($s7~zmz*K?+eTa}1K!hF)1_>zy5U*M`;z0f
zU!erGv9T0EySp|e@%O(G-?%hjwCkbFWzI$1(A4xkJX|#?DT&5iEMbi0;lnlvs#!W(
z+S=pQ&99d*e%P*!y*c~tMNDYucl<kdL=|LY2JR3Mr7CM^q<4*uekNnmg~WZKEeesD
znei=av0n?xfA}r(mK`4{C`qtAYit#(z7+-r2VXq<4nrt*hIm>qvyO?0$plbCoCxsi
zKk}t1{2sx{X3?(t@Q_BNzD@PwbZBa-;a;vmw;3HB-2#Tp)^NQq^j&b|e=svK^&$7y
z$qR~$9j}u5Y5}VxpA3()HX|bgd29v&aAUZ@)Hecwba^2y?OSu7%wOjK14G})=scjL
zY&g-mEUaPY&z?CxKIVc+AK&7yU;C2qXy1B!x1g3eFz#Z!y^WzSDkg!+V=?35xE(;F
zCmqd`prN23Y(Ktr#=ZKTzyD1uYujoseCDs&**7&5{myP~`1$*vrMlTj36W!JY`pSA
zx3aUd^9&JDgeqW8eFcTy>9)4ED2OOyjJx|e_yz#{2uhR@#CQYs!pvHlDVdq$8=!-5
zq@to;EOFQod-<sd;^*xrH|h<*G%kLAe%q6BU!GCj3=I#@1`6vH@IAyY4Gh9~0fpN#
zGnIckEI>UD3c677V+|7%^Zdn&7dr+9?!#0xy{EG?YZ@NQ&i(uMJ8=49S^OIs8s6f4
zw2%-MuEoN5{?ULE0Lv1P>4>Z;W%|<6(z}$Df+dc78CuXeIR8}VxqSYtH{}13k&!p0
zr6sYw9PB<3p`krzF-+Fyk%6%00puCRTPw`!u5=pdI$LXN;l+g|`iJc7%lkY1TKq5s
zw5nOTeP8rTqfRJ`_F2cKK-w%A7?LY0Di-$j^&zPcf+6nXO|m?0+6yKoCU(>TcVOIz
zH#|IS3%S{@gn_PWfsh(Mv8zk&<+EoV<M>)m=JnOpu9GV(jDD{?t0&PJOF}~8tyn@I
zOcW9w?Cc2c)6)w=JxYN00Wv95D<nAB0>;m5aIB~j*JNd7WmHs>32)xCphh5K1k|rm
zy?cbg#>N&ZV63{rpwzEV7vwO~)|L?<mT)?I@Q^CW)esY7V!kRw7^PC+pit2~HKm&*
z6@$b6@SzkvGc&50t!-$C5cs*<MA+CrcHu2suHaHXtFgmcb&ZGj<VS~lLogQ5^P#-9
zCPt4QRSZ>CJ(+{w_~6I({rmUzz<_`)ygE7~ZEe;cCZk2wK0ZDo7!za)w6<0jXE8jI
zS|T{}^3oZ(9z57gg#(UsDv0gWpO~7eie}T3T4ulGzE$Mlz!;~^@$~BDCuw+icop%Y
z0l8SDd<21CzC0rJc?fS^IW`tOJuzYSdS|hB5<+kgBv?3$6GYG9OTn_U-}?~rzrn@9
z!C7A(%xQ&?lk=BRqhx2tV_;y|hPkz(f}+B8@d#kC-+w%FW@T(_Z2r2pcbUN(eS%Pc
zgowl8ab?1>s4Ey9;p^K5UkXA(9930^2E{{|4e{~tWT?1Zu|t|KJd7Bv#{U)o$(p}h
zMWnhZU9I2+1jM0qG@lBGde{^^qzf=sSpg(iy1KGLa(cM4v}`_IcZHH|O>+w7lla>J
zLWiNRAxF<LBP=Z};-S4lEb>bNrTfFG(ccD++*WV^4N_h~h5LoJHhCA<>E>4ZV*$(g
zB0b&L(WOp1DK_TjtinCb3?a=Qfe0I7ePbvX%%qn8Jt-$gO-fewWu2^))C~x!0Vr4A
zaHX%HM_;zKvB{X*I#|HFD!j&tbypEX<|+n*=%y^0gJ-$J&YO<WV%r^MhQ1%U;&Gkl
z(boWj{{H!vFx*oZ;xF&z{W#rqs%G)Nj?r0{__{pQo9!0R19s4&uRH^&!^z2sYUjxb
zdvIab<v@3YSZn;zZLfQ*P4yji^K)~nJkE!?e*$QIch|?##^y%a_wU$ZklUcxoH_gM
z?&!$K@z@>_l-O%??U^>#)}=Csi?`}~T9!+9ePM#s=X|v5Oat5p@Qk~IT2Z&Sh;Piy
z&Q^AHW%swYGuO7ZGU|-{G|xhA>%V{Ggcm8`Oem&Rdez+Ayb|Nl3>iEyqw2#QX2aA}
z)0YI)9O@CL9T(lTL-7I{M>c>h(t)agB_4xBSx!{EyNFrM4r-gz(}e0y9GH#Xij9w_
z+MEZh*fQ33quA`!sby=vz5XS^(>odcN0nLC1$lX$fUe7~su%ds)6j4(zv8rBe9pkm
zZnYteIH8h5Li~MAPnVp|X%{fW{)X_Xy~hc!;0{spQUA@G8_ho@Ls`!7TK$rdFfv;H
z(Z-8W5y5GS{tcKh7`ta?n)aEc^&%Egoa0V8`_S1T2SXUu@++X5&YYZ@>S7sju9)5k
z&B<Z2nLgYJ1@1koK^n5kinh*-k~lu|TvI~>rg%9J#^VYqyF-*S#|OwElIs}y*!5$V
zFbLi~s#soG$vQdO)0+iCHZC_e&`q1bg{!Wsq2coD)^KZaRn^1NBVnqWtWX`)1hb@(
zBF2)}*&nj+U%Yg|+>C4F$4Wc!AwHtw;zTyw7cgFADZkYwfEbDfxSN%E`{H>s=vT*`
z%V5xPn}~$ORXfz?T)=_v@XpqjE8K{2NCr=wZOO>Uyc>0xG7bYo@&Pkeuy_Meaj+9B
z>@^@nAse3s8MjU`O&JIG&Yk2FI7VLViRf0Cm$$gK%E=ZQhUJM$*Ck}!i&9SK($Wed
zf*-}Z-O_mg^N{=8FzE=5iqaTg=uW!!$;-<XfT1q@UB?CM507tu!HF*}CYqU<X@L8#
z*xc53yO1BSsB*rM$}&g+@;j}q6ENOxIckZ@5;$F$0KSw|tJuc)XIa_lESYqqS+n!0
z)APDDKz)E@b!~rjkYmZFvkFn*BuYl~IlcggNfDZ)>+p&Cp6CiPKOk_+<Kg8c3knHg
zHd_02SxQXI0)?%fP2~wm^{dQ%n;7TaF!%2!k`Uv|E6XqDeWa&H`1#JkM|LlL6!q;J
zIR?f&boKRm?a?F|$;tKylhy7fFq|T5jf5)oBP&arK_z>pWAhgMON_6I44Z5f#ycB3
zCoUz0cSOUUUMsVh{q(@Aw#uu^#z8gQX)0%CXUX+_bOzSdDW5sRBmi4>_V%p6&)i^C
z|H<OkJll7psMbYhJ4gAEUvRMRWh^aJa{!Hw?5&UAcJl7uo#;qxjd=BzQ-<Eho{*3E
z!T?5rHIQl|TQT#u-@GxnP+wbH2$S*3`C^;eRx>Ni^B9S92(t@>S6mDa_BZOuY=I%v
zil6!_7>uQ))L+rpr#QDdS}eS%B`kbl!i)=>IzQ-X+&fQCQE7es?;XKbDs;D2$BRQI
zxrh1JRV+4xpg<a8spJ7*1DU`s@S1G3m6emn5H?L`zLH@m-?(w(*AEiyb^XgveLj3R
zZ?`$)si&>|E$REadBEO3AUhAhWGa1SX^Fme(u}#r`U@ce0S2bo9U$H{a!{$IS@Ma>
zzWW2E(2e}R_TD=h>p%V<z7QIURH$S}Mpjm3%LpH$VQ<M^*?UXNCY2Qt$w*da2}L3!
zBO{@ZGBd;dxW4!OyU)4L@7(|1f8C#RKIhZvbGcmC`~7;op3moFz0r5!z={gMa(2@i
zWTV+_di{D6>W5EIR{UjIAD?q%%4~O@j)vO!4gmCzhQg`VW2DBrq_wqGg<+c&7%>{f
zUFL3XuT13RNQ`>!+=HNv@toVKFP<a+MA5z6m1A}(DR%#L4a~8X{opH`_WY~Q%^?t#
zzOVDVqV>E5-t8B3WdhOx|0=H;`gUX(9>5?XJv%oy_wskVrheqhwLNMI7VI1+xqQV$
z#e9Tq?UWrD(K(K_3DlYTEwzjP+PiO`0CWp}F|Bx7(Il@9i<sE8AD_otlFHcbb4OhW
z0g!$s2{X$dCEfn7e^2IYtylb_ifk<57JeBGxbV2w_jgfho2zs6B8{J2jBY4XDrg(J
zMwm;ZL0n;!3GnNv^<7nYVmW&JinRIGzkgo@Q`D7AO*2%s4$ZH34h^X<E-$~L%`x$l
zW=82#{51G;^7sC<n?IVdCjuQDwpQwdUA|xFuk$}|0TCYwM*clm1~TK$mD-F4bV3aa
zm8SXR6a<Cc%F2Uk_Wrj$n-~Afq|T4K%P~JYdlv}(XoDIr7gII0eK)ZT&FAWq3YkpF
z409@?olH$+HW%x*HivOPl<faK$;x_z<yK!<=$QEIYSD=sAA<zy*4WN*adG)%i=`2d
zSYVY8(98sbXiCQPGju8D>*yhRddeR^e!Rl>oy4rnR(9RSj3j+-=;+UhS;pNkA&D-~
zvqNbT1o1{reSM%QketclV!?{Bk3QnZ`q3CxS5-N9T{dQF8DDs3Y~&|+lkNtfxa;^-
zzwX?=BNKh+PU#`V0KEVEp8|%L&U?*&t3D^fz9&^9QlS1{T$SIz%Z9GLqVfBTXGw8+
z)=m4~mEP!(m6J2`*4C!|FTrEpOA^2446%Ca;%|~6CV0~Ot9_)IQIKF-n|-e&+dw*?
zNM-k-(%%05-9Xk|A&MYq%{X5=m5>=y_8H6ecAB*R>wBTKTV(@GOk)sk4Kz-3htSB4
ze5Sce#i~>IcFT9wGVN)O-fis%s`~oJAjJLp61szSOs-TnZ!R6&JRm`P6s%73{K8+O
zQs=R^y2tK`nWr1+Ym-W)rk*bqK66#krlqr!CPdu+R_f%;#^2Yy&z0Qvqx*l0n|Krg
zr@>hiw-95a?KHc-LOCCw*VnIHNl@X603lvrSjOW~p4|5^E-vn%|7X<ua}$;~&Oq8y
zah8Q8Bm4rJfANw~z(21wzrMm(KrkhLgS}&~Sf$dZO+R^3YGb7{^SxfFRaAmbWlasE
z1mh<Srs!OrLLcG42poX+124l%>_5bb)+;GCX{1T6hbC#=?TEZ}YqyV*QiKVJXRRxG
zdR694eNRuGEHO4VwkBh|&~=)5jz3`+DaO(Rl3w#Hnh%ycF}4LYj(LcQi$8B}YKrQi
z=SkdK|8H?$<C`~TABZx(HGE<nP&UtYQS4p_;$vYC3hJO_WO7=Ef}*3ltLw1hPzF(|
zS8Zx*tM6cId+z&a(}xdx_>Eq_V-$BZ%(!>2eK=P$eQ=M#^s<ymz<Yp2*X$e*)``(1
zs)j?Sd0nEaau+J6FrdHl-^W)6Zlnj_zO9N;eK8!6k`eB#wu=1r#y5WmBo_hNXa1^H
zQB!M#RR0WGXVSp+N%|kw*(OFXC=d?xJypmR2Zr<hhNY#S^AbN77bTjQdpB=-jsgJs
zA{9qXqwsNMo<K$xdI<GDAG1~Z9)qyMje@tLqjh>90#m<u@n}hi1tx4LvKyCISE>D$
zmpWYV4aszC)<Tm$FW;qNiC5`HqkiF~Dmc5w+edi6Un|l(B$e&PwmrHCY01r~3O@p2
z>enx_1CeADp$j95%<F#Al74HKoX~H`5eMTCGX~`Zcfra&&jZ^m98{JhM@gKWcrPUn
zReMHvL5g#*9QFN#ZjH6pGA4}!Bn=JL4xNzu#KgtD-AzSx7fPAKzTu@ax5Yd0oS9jB
zr`5wxw?omtIQg5JPGIfh`EGuMn*A?%PoF-a``T@q)zfozD)6D%k5Lt1Uty7v-GN(w
zb)>ZN=`|lq#A1Xve3Y44#<q-EkkGIz<eJ5y_>?=Oq)$L{%T%sAdw4je2w!jg;Z(ns
zBy@~;E(71oEUHMZ@cBFQK71d&@TtR@6_=kqe}4JuwQD11hZ6gCZrdC<@a^ZXrN5sX
z(%*f0BtF+z&$#4eI`>W1FX^#8WC^F-&U4W2p%c*W`7V6($B)R^SgLHlfAw$o60)C!
zgh=ITrpD>EwK%0ndJF3s7=#v=+((<74emw?Bj+d?Asgq{E1DS65Spa46|rxts|K|^
z+$D!|gr{f#&)@Lz@-eQw`SXD1EAgrSj*A7&G0t_0KEGB5A;_J-rJuqL!!G~)q*l)_
zK!46d38yiq#$zw@g3Pf!&$T?f1w}=}(bBu1KJa6IRUN<ho;e-}Ks}r{B5bkNT=w%-
zj~_Gg8NGgGiz?U$m9l6){T|6+mP*%k?XN>;N9b`}uVwr&0?q$8HMI!()N3!GF}SiR
zpBNuXJjNSGMgl6ZygR*HfX$69jJMc|JB@HeU&y?QO?bx0s(HQr5wpJl8>uvf%<}gN
zx>LZz^!Qbi#4%=kQm0SehncjuE+D{qiHyK_AaZ(Y>YMQqx<AqZ-{1ZD>iI|I;>$^k
z-PF|77-=ZX;vG#HD4;;n2G6-O+aJq*&_YB+QjjVOa`!C7;KMZUHffI>+D9F*tyff5
zc2&7w41;x$QkMUV%*=y!H*Sc}jM~(MlM{m9FU~Kv(&uVk$Af$r&$@;IPqfrKDL*EM
z1~(!--Ne$?mZzh+*{Y|-35?3Osi_C&AmqPdNxmMcaH}O01&g-LxW4@G$${Y7TB&~Z
z6j%El9g{MuoR$+;>B#wZ)s{rqpjv2bUMgTggNZ%Pu#b-KmfzBk{Z1br2xX1rg2D=Z
z{+ze6s_MvYJM!sYzd}j)`$~%{`bukSVT4bUa52A=pFjAcp+T17{y!HlW8>GZ<Tza@
zE_S5tF{+S5HK5wRKMU38<`jlpviBPcVhq*Kp1l|2;{7!rJKnN7ROA2X-K4AHRq1!O
z>kMxP296pU89hI8_%MU~+ztZ;q^t=REzz5^1NAchR$ds4&DKdXQ#GFIC(r|Dm;0w6
zr~lH8``^ZLo?+Q;JwFUDRVK3|!-fWz$<Wc!x^PcJ%LC)su&KTMW7d-=)5yvyD;XZB
zbN;8L<x*9=-2`clphffD9^x_?ProKXFRtVy&<Nc*oydbnk5<_)t~>FJ@H`TEnb6~E
zR+{z!IAd9vd&KXayw8;N+xl#Nt$BO*#yooTRcc^x<2&eQnOpIw2S<TyJ`uO?k1?~f
z<T&Wx)!R!>@WL@=uvvGtFK}O0;eEyvhuBXXKh8ViovUpI@pvlEs@b8{ULEFpj}JIY
ze4CxBta<W8-a1|KT<U@I(p#^*H&=ceYW-NWc#@Gp0AF3~uzCN4$xxlIjxEDb;XB85
z2B^o$YHDlGVBXhOyqI#&j6p<Hyb-{}vuf#`A&Zt9>B9pH)43Hl7x&4#zAU}B=2ti4
zd}Q_l3;9G=c2NbrAk`xW8=ERk-9u$4d<@5r%V_m=ddc^Fj(sr4iSu1fPVO!taL3l=
zsYtEo91|G91Hi<>978p7mwv2o{LoCwn57SW#M##0zt*+3Fm4k6!IFcoehdxe{M2Xm
z_Z=Pdliw!KN=RG)%N-}g5P9+-PMbymgXsW+rDEgb<JYP>ySi*if-IiAGoY&ew|iGi
zQd0BF;^Im(b9=)(^sWI%sVH!8Q?s$NHw}(LH5ihXHc}E46hwD%uu<{p^XlU~`T2a<
z<WX$z74}`#`12qtYR}*r#h9~1Oj_Cjs3cib<C0@v`Te6)Q&XG&`?tOC%NM33hR?5G
zpYvLqIykwQ$XnNjk$N?qUAYUY;YS|Z|JL21i?3sd(28LT!S0<q<^6YW|JGI*;9769
z&=cH;RV#9Gekkhb=oDtB<f6frQ!_Z<)7k0V?>|WHCv8~es;CdyV^%?d`c&D>w{M-Q
zZ1D<Q1_o;Pza8AIps&&h?&6V5*qterxR@9Mpe}-0DI+3!9|&fa;d5Ovkmq5b4sIM{
zkvMhhICPw)&lLYEk#ce6eu87h<9^TNWO{3d>E!n?o9^77yxiP-dNqhZVoA9%6B84^
zOv90Ax!V`mg7fojBY-w9LqsK=KdV9PSB>Jh5AXf9ux5w0;A)(C>b5R}4eG$c;v(tg
z%iVVP@vCpM$F~}lDXBuCmh>{Z7E$jxmo@|V>fWbx8D^r2>#ISzeN}401Fs}4gKF!b
zw$x!m6KWM!xxMe}(<rH#8Z`R~r2-eehhLh0<SfCcjuO}lg}6+dsLVDuz+_WX(-<c^
zyEfWx<-0c@i2W9hFq2Qw1zsu}t<fQVM*N0%Rh8-&#lz8*=$!|F-L^w3d0It(n_2A9
z{<O5TBz4ZqO1f1puC5^v;4=Kqn3j1G-4r68nAn2K&EYcwvA{PdpIgBF7}z^FoXmIX
z|4Mn{OzzP${a1Ldp4w4I=fKF{M@z8e;o)uS(j+G*e<0=SQC43sqX73-R@`OI_&rc|
z%JZ_|NXf{{OWB#uO<!z|W{wuCGm|$qCIZUl^nAL4Pm-cp@n_Sqf9t0`%Eel8G>&xC
zy_s|!!>^|~cu;Q7{{4TAYOUHMUZnfEPNS=%pHoySR8BhhfwGt6;F<OH^-Gs8lVSbH
z+iiJxcof_9@ng?K#&c<^_N+b<v7x<`*YgEm8IbQl$AydWn=*q!!GwT-fOU`t*BK)A
zB)+!AxqV5?=)`M4uYW*3stj#7dwNoW7hr;_y6%+R3iEMb>fnQ3K33=4cO+9Y>odC2
zFjCoH)=v$$c`?<R!oDbli|`LBg{V|Y^hOVaER>?5Eb6ooZ6Z&T4Fa_~VP|XW*xlWI
zvmh_;GtCOEk+wD|m^<@nuBZz$Iof%dTU*RJ&va0c0n*1zyS;u@Owxs4Tnv1X;hXCi
z_Pq7@aZWS>yMaj7d>V6B>RA-pg<EmAjf5pO>v_K2cU4v6swm#iWD#baI6lFwz>Z@(
zgRw_mR7hyghxZ>XzxMaPd-yOV9P|`er0nJU*N%C-OP?LH-u+Guo-&{~LB-q?fCDaF
zzrI8gB<k{ZSCpTP-fK5*cmQ}!7>j&S9dI@Gyrmd}izEnt35%5)TVEvFyS0rCA{`o2
z{4WAR*=L1?6lUETu$qx1x3!O!-k{c`H8RdoPa#!OQsO&$G{x^?8YO@i=bJa<b=Ltl
z2BU-oftTxaz|ac^O)ywi7pbmdm=>30%IUY!Ut2plt-dNQo<IDh_Q3@XVfFw1dw%ls
z>mAn1Iy#gnd`T$AEcdUS*w3i9myKD{)XUq~$+$kizfGm;D7BsvRKBMu4=~OTpk|DX
z25dU4!3Id3-jRFd_J^)0ufQ1c3m;*dCc!n8P8aAI#r;?KS5zO-+|=R;6^`13=d03E
zGO$@_j4l}&NfkJ(uB^zTJKOEskoW9a(2EzRovY$foULoB&Z=CzxB!Qq&%9SnvkIMJ
zyyffaYUs1%tjWUw$e_8n6(^<ixO3V~P*8BHdCWQU3QuoVMTHLAW?uHQ_o>3(uJxya
zi@(2vLmU0TI^?cW1eOHy?5r#@Fk=}dC0cy`BI4qape5x1bVGqQn|1E13P+qsdIvpE
zC`Gt@@R#!PD_F*N$E3x#e_zqjQGk+6Y~Q{+Oz>mJ^4{KeXY{N2qu%3Ris~E8N<EEg
z+%DQm)N*X^rD_Cvx?fmW_?dul%{)YFp8Vh^|1K`>eIv*>3G%)+F0PlFn`L*2&G-`*
zUfzAMUdZzbvIhfq1W}cto!2<x0Y%fRii&+(TYCQfROeES+v6{hhw*SxN-!1@5hjLk
z`8{SOx+>O$peA!U1A|AMx4AHLfgV>a_`YZU%hs;w76t4Dl_#VjE#s``xqj|Rk@uEQ
z%AnG%*Aa1%FuW5S{6>iJ0R3X!_6RyLh)Oc*>X^V<N7M#_<FmB1T<q=X(PcKcJphFZ
zORS1~OCZheU7bBWVQFbhug0Q+KYmo#hWwZbxDv7S*1r+7e`i+!w4_1MURKj&XCC<D
z>-_xZWVDA4Ip8ZQ6;Lxo$l)_(=j3Rk3GxHXZTO|YIMw`yQ++V#S&q{eKh*D;2svrh
z)7#5>>Xb^<DNCN+!YNd^VYEL3w>H%X_ms$7vg0OQs~XHp!dk6^@zr?GpHGy}w2aB-
zD4p*!jUz$J^GhzzYloRBn$lj(U`n0SQurtwZzuH~aURQ|u{}lWQ^iHITb3<$!$#zo
zj`kUGSRjJYf4JcW;|65%bNNZ{40kxQ3kXny@5%%dHq2~5I0q6Rw<z-k#%vO6kwImx
zgT}N6ckh~CTd>H(nX>C7C+E!8a<^ud2NtBnr{B=Md>tI*?|)@;IU+Xp;5TIn(Tx~I
z4!AO3|61yh0)pa-WpB;5_X@KVaqp<#bPhrbeGU>7ox-dvg=cCau__5uJ@+uMVjt*k
zNK|_J_t=H0b7J;9zOY}fi&~Q@n(Gy5NaI)#v=hOyvZ3p-XWW91lwX>TdC7agE~TbM
z61*pAdPihPNJzBftm5#aJ-bK;SKkp;zu+tsju1W(kv=0253xox_E$C*dv#3Y;D}PV
ztT0|aUFmZ7%#Dsgx+6!X?<c5zaq|8<a|DxgwX4Kqb=WGFpkK%`XWgg><uLg;6gW?S
z^l7_Tf#J8s{3SGY!AW!<+P|Fbk#SjzIuvyQ@Ur{~H|&3BiN_+3+Bwx+DIS7Z$Dx#L
z?V0)cQtPuT`JKs!aD0HLWwAf<H#P&hER;YFF%3l)8p55XxGDFjJ!_XtOsLTYy1xA*
zqwpsSKzied45lhL9U)R#bCIH%Uw7X8bm9{3-QjB=m|gG5ZKip_G8l~Us2cu1qlc2-
zFIRy5E58uHyR#@}KIS~Qdtd!ft#6}_Z(YIvP^}^brYHbwu*42RnO^s-;9395rH@T2
z2USRpu9Gy6EBZ{X)9`CU1A*TD)_9!2j3QCemWGd>_I-Yi=>QPLO!xvNz{aQ@-dWhW
zbn0+Sax%To2%vPFWF$x+;8;0-4QulH^^<p-_kEKI?)VuBNXZzwWFte8!6#jVJyZ-4
zd4&e0R=LFs@_^(ra&jo%(w+Ioi`v665`Inh&rsdz=UG|frwV?Zif##o19s}qABr$9
zGeHLDzcW`?`?)j=_4V{VD}SA|9*a9E*w-f>F!ufAWzE!cWuCtW%LZWkWpMG7L}w$j
zwS35><x^^a6A#bSsNd3leLnWxu5W11K;t6I8J}5NdMIp9jjfYYBp|RVD3%)p)!Bw)
z(Hmjy1xYOWUCz6LZsO2jVh&uA|A3U(m~ACAK_MVT<j+xSlC!9EcckX<e)1i@hz<aU
z7M%Ub7l0Nx=eEGVy*h2w%+h=%OBqBF1K3L&2u>?bmIQi-hYKDI>g?>{aJ#RGI@Q_T
zedrD^4MbQxi9990F6qwiA{9*eC@q}wPen<|DpSzGxytZaK|un)fB+X6fxfE#pAbH0
zMs9B3Xiv{B^wwQ(d8h$BI?v4NRg7#0MA!W65cq+QlbG21%rEy%6&*p=4wecRU*ApN
zXeOzb{`230-JP7~W)QG&s-mJ|JpA5{A2Uu(79pX}8ag_wUE5omeHsr%qRx8$F?=;2
zL%KL-0mT%~xcQC0cA1B1=<blCXF7lR@@00D`he3e0lEy)l(H-|tA6WGpGpk9vlcnV
zaP(-fAW06v>9Qgx2ZxG1T?y{8sc;JM9Wd3%VaOw@i{20IpZj5?o%k9@R&jB0%uxo0
z;)Dspdr0HGR%TP)8E0Uufjyl9kZzcx0xmBcIkJrIdx6ZNtokRo9I(yLm*t4fAK2QW
zB4pIhjV`VG35tj;7v$$JPPkT4e=c0D2eTfu6GA=!Tf8U6yz9UvRssLSg*iY-f^l)E
z?)TS<zJH5t0wr<tCqR*t5NuICY=?%2ndv8_>O~qR|I~Q@EYw#v_&VG5Qdku_CS9Ye
z)vsT(@biBf{`AQaElr15Mq=dQm}p|}@0mx@*47s8UlmjBzbPh1%X>XJAt@xK>g<2l
zu6cN)6)&zR%Fo~1+}w=57|>>hLaxAQ&Gj~c(ZE$BalN-M&0yP72n?(@badpuowU8b
z$fQ03NB1SVJCo#uIivlxd3k$2wznq-t`2+F!~N(6V(4Pe>YZD+AQ^Itnc$fsI2nm+
zYim#Z8LA?K)+d`oJ;2=D+!P97*N$`<di{WAYr953c0s3d!pNEJB2LhzP+>J6MS;q*
z6!tC=EMY`LHUXq|I=Wq6gc1DeTQuF{j3-X?^?vCQJTN<2l%M;&`ui`ZIw9JX5=}_P
z%*~&amy|3#)s$YhG{U@#Lui%bZK*s6D56wqvY+4?dgsykr-kKTYG>OY?JrVQEeP~p
z+~sSeuWILD*SNa8OjT`I8?bfk&q(dBh2IsUgAV=0tm77Hk4{HrFwh7=wm?7&&qmHl
zclhvooI+|BKR$6~7dcFfy|9^{Hn&L+Y6%4_D;;%#b|!4*_iv((unnsz<;I5(wBi9+
zMyBTGFH2F8j$CXDG#SL@i7&z-aK)4=y@Qf7-UjyY%PG*+s0^(Fk|f7sfXc%>>XRI%
zCn}4U`@>OY4{WuxV&DDM)YPO#3toBC)KvD=I&1s~oGj>*uZwu_(blY9KrhI5`gFL)
z8)JwO)j7Mdy6QPpU7sYzLUwq9$iE(EWUQfR6l{K6b#tuYP3UQO1xH~Qb&WWcS3O`a
z78V}<c~0>Hc+;5#TUYJ4cm9ElQB)9rN+~J-{0v+<xj?_xhbV@~#r8J>Jr_`k%!l&u
z@TOwU`)F`t;^P&dcrO{nZSM48B>A7Z{77s+q(N&73k&i(Gted^!JTI!EhtEH+Gez<
zu<*d_>}*oK-+CP9{S^a!gE6RcGewU5?#8s9*(@N&e?KFqkRQfNR;}1qp#l;Uz0Y-?
zXPO)Cv_r$v;L=Prs_8j<$!F=uy-N<TFm{*kq7)GDzv_C0_U9InR35M|ui@xG$qxnR
z${i6N&W(T()t;!Kfq^|ZUnUSF^lZ=Z_qZbz182`>60-*GU(j<NnS0y(lk7lrg%`%F
zi7D%``$YTUw*{{@lUkp@3K$sJkM&mQ>g&Hn+(S|6fV~x@HPt&9&%Og4FD-oaYFS%d
z{qFr@{v2fFI4|~a*G4oqXD|n>jj7-yYU%ISWm{z?OfmlJP@-xS>mT@f8T;P~NgulR
zhDYnq@ScB^78fUvO2k%e1E?8gZgLS_2s1F+&pwCGyeDO|LdQjjY%H_9e*MHDu{Kp$
zwunIb9PX3bMW1lUaxrpDVG<6&OlYoU1rY%?wuq^N!)dqUP}}8Wvz|YBde{a#9{CeK
zfZe=RZ8@`7Sy@;bXd4Be#2n4t*>}jo;?mX2zE;4+URlAkcEN9LsUrz)U0P9%yMy}H
zw)V?>yu3??BfoxkZUg&V#S?X5klSY9HasD$IIG}QY`fe~<p<D?1TujKXKoxgR}eN{
z78lp$KpVxe{Nvc6L+m(mEr6Ffh&&9%8B20c>A0(!PYYMNAE}0hMn~OXwRm%H@73%M
z9Ov1fO#W24PPQ`POx4D*_%6E5)JU-W_U+b2EDe(bdk>WFjIAyB<96>SQvL1D+12->
zqm%RU^Iu+|riF;}8*8kJGj!w5E-uptzfh;PjLShfJRC|%ElC6%nsX)2o;`{QBnMpa
z08CZ4C(i&TXz1#S!|hpPG8>o0P{?}vbo)4L@E%kY3T(}V70>rU(>s~|^G9JvTifO-
z)eP9p{Ld0PBP{%(O2hi&otvHgchT5bf?W=GD*I`!CS)VUMmWJP+|@#vnVOn;OU<S6
z?7Edz0L{|K;Nadpn(01aM)i(;ycgK^mz;8;qQLe202xi803Y9BAI35f(W;v@H5*b=
zlAUIM>yjx+Nf1r45x6m(yALCq16zT>Ix22Kec+Dt(cWF_PLn&UzIh_yq(Nqtc^)2`
zAftiRG7EIw!0lIs5O?lVYwMtcsVU3f%y+!|dDuuLnMU*Fvre3h|F92r{8%yA_Hjg7
zFq#>Wk`Q3ko7Cg`;(CSU5H|}mu|9tr8`JvlsdjkXR#HNOEM$J4Af4+)E!5h9xJe7Z
zE!B50B77+0O)F<bj_r4rFjP}eXcWQ%%@kimyS)J^%N8nzLPFR>9?!RP^6pK?Bcb?4
z*OU`&o=RO6yT4@(F(nP3vy{CdL0-2|oNq_MLN20`-gSKV5Or#g&=p8W!%_RMa>iF(
zrj1wqcMyDa*BRld=)izkE?xo6ca@K|=DIOTW+EPH46>2~M)Ew0S9p#%d;L44czu#<
z=Gd_yuRt-!q1}<x%n@i{_YNyTZEI;|<r5JZ>5G4}<hGvQ5oe;7kW5QWO#J&E$jWb|
zD_8)eQ?y;eTV_AkDFomiKOW1SqpRTp0-E#+kEm~V&VRmjJ1busd36dHXs!{bP{^iv
zwXLLfY0Gj%T#V-oj*ixhc>LHfzc}1u4b+&=vWfuGWZr-M>U+h62MTZjfuf6|SFehL
zW-GtdeQy_(z~B|u*4IT*k@sv#UW0%fYtJb0@#BI(xW~1OFgd;%bPt|rkx_mTnDm%$
zr=(yBW9hS^qDWLHGaILYxRb@M>x%*G3JRG&Yca`!UMz>~*@3xzZ-c@T6C)#EVruG%
zWfZDfYVPyez-rz#Hx=;+_`V2S9jX0%^_G|9+P|fa-+LFA6kvi1Phtjg3$)h37;?FM
zIaUtfl|bSK;GUBYFS|lYnE3fK9@451KY#zSlVidEb)}qh=XuxEB;;UkPYSW8Foqcw
zYO0F4Ay#Io0Lk5&9Jb8RCtO&Or)u2I5b<GupR~lOM#e*zyXX4IDFOgU^7g3hfvOJc
zUB&I7L3R+!$UcMP$8~FMkD!0vJ*<RNkNfOdZzmU*5@or7?PvbUw;|NZ%nZIc^YQ2s
zu(R8hGEQ&9_6S~YP2j>*q#4%k6KuwS&y<JOHgy1^1|vg$i=Av{f-Gz!Fmk;iqrO-Y
z0KANls7kJ--iPJ|#KJznYK>i7|2a5Z%K4o3EI&``;8P{U`j9TnXv(Z3=BOLaG$n_-
z=~7!qy5i%dTp_)HKO*iMnLG<kL+E>z-h`f4XZFzx94%9AWk@FTA`O0!6gTTRIdk>(
zsc2@*J*OeAZMn2pcMOJyP+*4h1qB6*P&tG<=GNCU<J=lbZ*H0MMNXTw=?)1*gq=e@
z@77#9(lE6KrWK3E6GrrlaPfyM%*_i64JtGWHc>)OLiC@C$PIf)7T#PZbiRqmGw#ly
z6*jf7xcPoy;H>_GO*+g-Fs)kK*(uzi!j`AZv9c0$c^h^L`#GuD0dn0+&<yJk9m+*0
zDGk20Vme4{h8XOupMg2q9LL_~+I#%px-{-c1IlI<F~^2ZG5FW)Y}ov~a|M38SAk&_
z?Yrr7EA*|Apk-2QfRRKXmoJJtp9Oe=rLbQ1s|Y9;X_%R*AtY_|cwj&LGW__rxzU1r
z9>1jTsL_>yJ9Wcoj^$sKl}(->{HpBd=QL8Yhx5mHnSe<(-7*K~kI=j8D0&eF;!dxJ
z@7$X2FL&pK_}TW|rt?DURTWj0P9Vm1&#{mdZ9&lKrNUk?gK?kzT*ifl;o%cMI?~hg
zpFMlP+Iq6`*|Xgmk0qzarY0uD+zwv&Cu%%0f=S8K9N7<CRvl@lwGA-SI{*4|x&r5*
z8fyGt<22hj>xG8ogoI9nlayGj{uw@q+AMP%Ld2K{4<vlMo9_uePKtrE!3C9H;QPYb
z>Zix7G`orO{3Xe_H8~qE<y#O41OpZUAycoSp;3Ow<L2*~nUM6r?E(LniM_oi)shPv
zDJrQLZ6^Mlsc;-7H8A+<FYY+xT$Go0w4|fAtBaJIJI?}Qm;g9g<VTetLYBj*Fbdqs
z*4)BE`4z1_hD-oD(U_Lo`U;IMLq$njDlOX6JIdqBR;dL){Cg-q^wCo7e`F<T_Wi=8
zOIIYVFy`s#>2WXD;QKWIQDbavYdZn&z4Dct*rjAViK_L`noOdKzxIEBw1YN2E{KP#
zVy;_L=EQ<(E3hPBdTK<({!`d!eGyM9d2d4l;l1S|G4qfPW8m(w%<k!8>oNJ}Cxo3D
zdUA4)d-D43<N)2Qth%zEJSha)@}d?pEWTsI^%1I1Nh8c8F^9fK?A)>B#KbE6yXSWz
zaVV64L+AWGrLV~;u|j-D-fV7c1htG$L;lgge*U~~3lX?Rdzw=`U=J`kqTu7l)EgbC
zcRqHs9WNId-P+o6#*Onsdv#4&IZ{%eJ8sX*pu02}u$-qlGYtUI<+gwPc)|8LG%5j-
z>sbapS&_*>DqV#|d33%zMMl*W1OmW6N8gp%1r?^TCjkZ(_9PgZHIUJyVRLlMuELCg
z0+(YDzomKtJu=07xo&3VMNLgg{Oxb-`(0ZwW@-VMi9I!N+3&g6D``0FAMP%Cu$Y>f
z+NQXmye*|hCqS$z?i(9xZh*q0dafag#Sa7REr7_!V^tUr?(2sL53{k}c{aB1T`0~&
zu>Z<T`Kx2N^7E?!g<HYj{!p|H=ZnLImy3W80JPSvjJHG!z>4?!@2v;4{f;wN{*=Ek
zcw~pUSV@V3=82i36-E5%(+ce$+Z7)&03AUL(;y<MYDK&kEh|9hM_Yes!PJNeFQlfX
zW&jjQGGJ7H5x$%`Vy7PW{rg=#e*R#SfCX}kyXEC7zpq<bHb5#VwY>aym@p#kv-jpF
zSOzgj<f2FW-Ct&>1BQZ&pP18uT#kgdR;48092z|P+aLMJC@Yfd>Fez`u;)!tr`Q`#
z`u6d4{4!v)7hIK<{+)>-x7;;Q=gtf1j1Cer&Ol;1?I&$ARSt}7>|18zAHD;tGhqDs
z@F|4?($Dx+G7KGdgY0M)_h7aiBx&B<TiL$P!QUDhl7bZP+#eY1595fsRGZj1)&PIa
zG{N~V#7%Y`kI%AyNiK&ngC+jMG7czX7}K77btbo=p1Zs&!1}QZ#u54TdMK#n?ClFi
z;4qp3h6Y8}&DiRbYEMZIMA|{7`xWkSFDSKA`0Wi~ECoC~O&a{o73U-?C1oT7Rikaw
zksw2GB&cD_L>;8G_hPm|NmwuYLE`qQa2n0AVDJDr;PxA-c}<)$p}y2lGc(_8L45xP
z$l9ML8@WiPdJD&Jnzgl!lw5Ate6iz}4-D+@q18Lfi!3|igoi?iFLGXAyzUbixYLVo
zdB@#<%h>pX<;zcKes18Oxu@EB%V7BAP;4nKG6$Zyd9%`U!LLnnC;SoKpPzYWj52ea
z5$YclpU>`YZiPR7{w%7!`BS5SM)>&>ANL1#-ZM9q*J^#2(L}bU>jhx&QuIM~kMW5U
zsmWx9e;UUasKI;bmf0pwVip|PbPon=ipB=?5=njp{|`ILV<EAj`5A;a$HlGIynL1S
zJ<?vOfHUkZ<i5kO-v=Nx(4H^CEMY9*bSo$ykhg$koJ?x)DO4~O`HSq`02VC~_=iuT
zx{{;A)rA{a_>nxcNYYTYcj5+qZ`w275^2hyRK)z_7;9*etTqOn0T;+0VVzyedQ^P1
zR@wFe2H@Rq={`;@cr{oG{{|o=g$->gK{^|m@IQi=PjB9^>Z3!>Ua2(X=iuPzgw0O@
zZVP_Uini1h=7JJ)3D`gPq!0-DlwJA0pGa$qlRy-1*C2p{8fz;nI>@GKtE%)EQ8$|&
ziKio!SLi!x?4L$6xrgW6ZCBNmTkhm)rd=2)ir90n!shN*aAB@Ca6mjB(>1~I-i>uK
zp=fj>m4+!<?Z)jlJ`&5?qn(|y7J>}tk@m#wf-H%cwKcB|q`{iv&rHKOya)u01#v_7
zjz^O1>_5&r`)OE$8U6t@<T=IerGycwE}NZyQ=gxPFa4}Cl5v@JI<19TOuWu{z++GB
ztZ#4LQiwcYBr*TDPYES3_g~`IJm8_gfEIB~gnIeatstT1LRvRw09T+`K1|W8BY$A&
z))uts;-C?&75Y9$6oEH)x9#J5QE<OK^-Rm$$AQxPC;GHFg1<#F2ROg55ETCU@KJ|m
zaoXa4u<18oX)D?$f3OU}S~)2!+_i;?`AtWf)J+`sOuZ$IyDv15sGK$_FOc9A`ywGK
zdNrHM5a9~gP;M)KR|-2mws!qpO7UfCEY|(DzgY3!^NaueM^1O=`XM?xx|oE7SL+`H
z>Q&q?>Ykk5@rh!gVd3Mmm(|xlrwR5X>}moAcvf+7Pa=M2`)9OX#$F&9xm8djyW8V#
z1OO(<fAXs7=hE1Rc&<yv#((~N|L)p^lw!#Uvl7ss_F6A)!7GDCLB(7+#RSqsaw2F@
zr6zxSb-y^u5?Z68?&0Cuk?!uxp3odSO;QG(o@v^)@nc-}a32KN#m>UYYIf7!p3mmJ
zgTi4LI!)jFTenflv9g(Ys9)^<Rm|<w{%&$FF#QeicEmh?{wpjbgw#Z<bobxYRRveq
zmm12-EgSun5=NLs%;LK<m7gI)GIfeeQ}VVk-5;;{`Gd9D*<r#7y5Su^6+pNfsjAM*
z;C}WgDJ!e_skZ{vgUeHD1xk23pnv>7CnrCX%UeAlqGwC2tZwf6@11EQ4Q0eQvD3py
zgyP{#$cDiG53ckuMv>Ehr8kW+k)b=U0#8Z|A~2#qc<{~q>MDKyXpx#4%)7ohxw*c3
zODVvOsi#YyAhPJi#WZ`wzmJ>EH-)<)Fk}nUkp&4sj8O)Z)Nza#J7`^*VW%A+N<XeL
z+)?rKlf>n@1TXCm7qrEZ&=YTA8T32<U5XBXrl&K=>_M{(_hI~Q5{8I_eFb_qFDNQr
z@88HkNM?Kb&%VHgx8aeB<m*iSTU$LCGR%G?Y3pg3CCvXSqL&`7tF=N~X?~)bn8Jk3
zd<UaDS8SgZSt!Zx_W$YvFllRTTzINog3!J%Ik2SRoH;EZKwsqq9tJ**1qfr6fdWN1
z{KFwC1Rdix5nwX2vdU2gRQT8bc@A~EFD9NDjLT9;)?+?)^yuI51+S9!&9FU(iV<^T
zwGd23x&6`!qLARZ`W@rOrlzNIkAHz=KWb3vY!*0m?>Kdvs+ts%m;U;{ke=gHK6z5f
zMM#KVjDoTcs{I27%nM$U<u=`%8Ye=>Hq#tqjuAQq2N$`iz_SaYdp3AGI|e&N_m?kq
zf2IFQ5n)qINKD|5=8k>(dy<#;4x&s;193a3nWcOtwDk3Pde*NQ8d8CNTtH2l!P)e+
z1o8;J^XJdsRp;E9ltAK4^$P5?j)eK>y*-d&yg?;%`fM$dSyvPtP2FrMOuz3@l<m*^
zC+5H5(84~tBe{Z)Gla$?Fx>1xPx{FK2sJ3F8n3L*jhd*k(1dOC5q(CWbo88B+dNGf
zRz`$aF+_)ht^F<KyYg=L`4uc6T}KU#_qksyvi3cHfKx}yi+kqR?-LK8O@zf#cob#O
zfJ&JL9C#$i1y#gxYNFk;@YXx3jBJNyOS@l&Vvka>we7Y)I@nx#?+z-%A#h{a<_%Em
zYlrFSJ5KTQ=itZ>;W(ncw7MGE_w;EE{@0Jorw*f8{a3sU8N@!O3)0!yIZro%J-$|d
zD8j9$p}{{3<V%!_f_=EX{kP^$(g#b@VA;-lw{O2}fB(MN(az2wH6bC}@aom@ofz$P
zo4ZHJ2=m#|O+NCOm;A#0U0`>^0o~C(TH}4lSljTiWP}DaY}=yf`KH~@ZE5zIczRa&
z+t_fQ;^A32e)MRQo2O@Kq+ucGfXvsgA6CF<^&Wk%B4CQ;F@~?i;&|sy7Wk=kNb;&-
z;gn91&m6+|0~-xPHkoXDdhHJwuSoD2{h<VOxOOec9lG<|z@pRupm1-073M9gs8I0o
zk~n$x>_7|NyXoy)Hoa@t#0pRp5w>`4xT$GUbEh0W*xkEA91&aR0CBfFOeiw|ekQ?}
zn<b{h9<pEJ+&QK$rZb$!Xw1!!01Ag`e7WQA!B_CoMvtlX4)yg>!W5E#@oN)m2xTWt
z&HIlT|3<`qs2swr=TB6<7c5g!v;Tt)HqjQ+Jy>AwFxJ!7Hcw4X<|1a=u2)}KL>mU;
zaS8{64wmpMF-b|o@26Rx_Z5&JouKH~@#F4~9z7a9%2r8_{xTj?xK*T{#Gvu5vvF`Z
z0r&jQUnf4!|40c_9<FYFEAZn!!o}P3R}YgT4HJ_iPIIXbK)Tx#+ma$=?nSqGxVR%b
zW*;G=CSYrD3x~iPv~?aZ03Y#4x3+PxJFj#kbLB88Hw$f>>6gT9xF+clr&RC1HlO+3
z*=P}yP6iIRh4-Jc1QFfbfvrU8CoN;7ppZ3#HhN9SqB#f`2(#Z5CUZFNb?XCaA#gBe
zj&2-Ct_#?YsZ{Wo3)9YFi$jM9K|N5lHG{RN+OHFC2l)n~7Qp*F!*xvq!JSr8QX*=|
zFTNG8DIH9#=`d77UPClAgT3~Gc6Ape<vzd~%NR+1BRf>K|DGmof`y5RNnyuO)fNUU
zmfLM=P(A?MjsdJWgVFtu_eM|N!;OdtH+uSH&oU(<TU=B`-F*K9?<MUkJ$>SW#Xwub
zAO{dThE#%b@U7eB|6$zfFE&$XhU!xsyJraE&?M$UZzoXgST(lTt3RRLmclk5<6sjq
z`p>AAhl696jK~~_)K1u{aYWD@KFk46cMlhB)C%UA5G>8Wzjg53LA%y~&gdD4f&cl|
za0;mHk8Hd+T6L4_K}t$+L`3uX{{bB!&~C&yKn@yp18vL-4nvmv%G8kC_ZJ8Fmqk~s
zhX1RPyx%+VVTy&83|-PCtm@x_4ucfp;uURF6pap*FR74BfYAUltEVuDkb*5a127!G
z4M}xg9(QwRXYmda#fV~D7Mk6iSmPM+1p~MJpdU%E1BtL@da!0ydK726r1#<l10|&&
z+NP%8&Jk`k3Q}>1M!X*yVtiRz3f%J(6%v6bCsBECi%?_I)W)aZc`K_~#C4i|e2Te!
z|AELx=tCt-UcP*I3A=)8cKvgAHyQd9?R9VO^JzbSY;-ug-mZg(WC4*e%S0Es9^bW}
z-%jkMCcGjx-{+;;U(f#66oGDHmFx581P|6T$*~_Som0-15}TOro_D+Iwm0lJdVt$k
zHjFJa%=~)ZD=t`B1qu3DlzpbNXU{eU95oeWXfPEFQsEfHS-A;vg@rZ>(>!sNLD~T^
z3;MT>&CQeg(`#FbXJNfIv#=<&f{jB5DHX%Vj~$C}7$S^Nj2)q)!<^L(!29)6r!Ce0
z>xZctK{Yo3hX9WS9Ysj&Co0YZ>UZzGc{3tyqNdhX4_WxcR%Th*ku;=kNHA(7UtpU~
zN=~jn|B9USO|?q_@PnVBMOHw6$%14wqWzrbm@Y9ZXvSx%JQWqM@S?g=!bb=~4|NLk
z-JA7tn8wSwVpZlQepD=dnsg0rx5^5!36zEC1ZD&RK3^i(6MvYgciuxl7KtAu3Wvs=
z-*L*I)JjT89W9lT6zgzvcQ`stn!O_iK%I<B^Ra1V`@^!TtlHDRS5m9<8~w|H*jWPy
zd-IO+FffHtoPzZRCJH}~gO}t_-e<_KGx1k6K?)^#!xnpP5C#?1S0~>!H&gH%RyNEp
z{T=4}w$xoXTA9gHpr{4F;L4T5u`T24-abAfXWeHGTOdDjE+Co<LG%3>s{)aAZi>Xr
zGn0^(v!yaaDUp#=e(h)S)RgSG)IU0ghIg~H^QwqALE*D!x8jrI<EiIYNA--gSao93
zHS_XbT(x&}q?=wa@5;WIk@wX7<Cibx*DLqVu*HzFxx9bhDi3MdDw=!o0^O4DuKxbD
z$KyX7Zo7g0WD17vSRZ=!@TX5tqCpdSQBXicng5K`-E7fd9Hnk7_N31aBulz>O*ZFJ
z;lSV^1zP!I`+rK*&5f-p7u`ALb^@jOPukc-&6^gkoke(qCpZgP<}&Z>lYixXqjW%!
zk!0en^fmQIVlR9l2s(~7haPv4;?z>&+Q4fYc0!cli;N8C%#u8B9l$CH5=P$2w$X^(
z4aJ4AL;Q=eJ_7z<L+aZc9ua{p8gk)NBc6&vAmk?|BnpU#iT(yAPwT^kMa;_0{Q-jE
zr7q+Vlw1P?R^>L`wX-4q>q{Z|nVU;?@WKQ`dw#~2z9ga*ixxt?aTK85G?~ByfGtAh
zGreHmQGiNmqUOfqA0~2{ml{xB8%IZF277yt{654^P;ESa`%>POO7A633-Hx;faxuO
zx_-{6p(NtTUj|$IOK)gSzy;;SY(AEe+2?){9)Bi0RYf0Y;e902s7^dcdGrXwf-rl-
zXy6hWYXj+b@~&_LKl!m`(e%lYXLh)h%K>w;4du$5GMTXUZDzf_1k(nyjpatHoAdTC
zR?L_9RiK-w4A?p+b@uG~EDl&=c;iV=rV-;Olz_1g2<X3>ng`i@;URkS1GjG9b`1nV
z0Q57dGcet_@8b|7M5X$$8Hjg06ul!f<jV+k>gXsn(2K@GXFvdH71p1hyhn>C2z^^D
zP4v67w<1j4^zft^L!{)}+_*sx7l+$fY3Y&Q3Kzd#%)G?XAQL3@#o2xpfTA@{ZT7;>
zPbS59I93b5jw)OoBhXKY8<z|-(H^~NY-E33P(TG_SCZq6{L`$grE(VfsGg5jrBynG
zhOBDZv>~h%CL{R}eij|v9o7L4?9v<EJQz$GdV58EI*W!Bdfc9Vk5I~507RmT1JAY5
z=$FD{U0vOVPbbcg8^UTws0nsFx!iE(d-5p2rTg{uJFB3aIi%C;q3$zKrCp6*4OV>`
zQjG_x>*=j~08jeqT(xuiCKGT7qCai3Fe5{@u(WjPjF$Ec<n?b+Hiwq~F3U^S*3gxS
ziF<bU^aMPgZve0}&Iz;u;+P*A=-FzrJMz#RpnU6~g<pDDo<s*t-mX3KyJ3c-l{GbG
z0n2j+Av|9Y^8aqH!EQ$m5U<7X<^_{&)}5<y8CRmm$f6Zh!{jqoQd}%i1A4I$58fa{
zU8I!>uxPm-j*HyPB<16F0yr%|WL}|@I$*Bld_O(_3-L<<3?S{g>sg68Nh$xW_o?Xt
z8`iz~S7)i;&_01i3{P&jC4~<2HO4DP^ehgsZf92)7jN&cb+Kp%1Y{?Z@ZAv16olJ_
zc*@(-mm2S-2v%Do<}hdC#E7R!P+}t<fD&eV9x+f9yPd|#rB%>m>;oatD!tq;?n@_N
z?57K8)D)yY?dTkY;|d09`<!-0Y7sB?hDvMSYhXmANJT#%iQ(h??AFZlo%OB2x-B|1
z&50|$X;L=$0JKtKhxd}x@f*H(c6T3UWMJqw7j|T%HOvnBSB(@j37pe8Z~|n&Lo{T1
zT`#IOfe+89ICPkCxwbBIY;_lDX1MCpy-LqzfNSdV(cJ)!FgiC_wY2FSQ1>6&Qd&C-
zl;u#QgS4QXB2PQ-Sy+dCiYj$cJrHR;I1*h90depdlo#l0$afz+z{7g<n7;mY`tKoc
zH!vA5p#LJhnCgj%Ko<^O9O9RW8vS2TmHTiXzJ1}*FU+4{Q(5eRR<9=U>%H9Wc1Rb*
zH9uKikCKEg^8n@Q?>h#syj)oBG?`-zB{5${epfj9%uf$R?dlK@zPSo%{A22N2k2Y(
z!WiaT52ADmYvS~`15}JHW6qDMxix~I)1VIAKHG@s2xo*>t7|dO2$#N`00pH$x3*bL
z(hx`<_}s>p^iI>v%EL1^8i5RchxT6})<T{${!MRh;`EOnbY7VBDq&U4>ZisvK0ol<
zofn*x;1{6XRsaP`z&6u*v+7~8Y0xp5m6javX&pme$D$xr<Gpe{|KjEkLE(wLV*>R<
zrS4;I!>8k(y~AnOn42DOj3}|mK_HmR*jw@_!#g+-Of4;Y5TnaRXEI`TO`f9p*s%am
zfalzsZXa2IfReUP6a6L8xoA{o+xrl%W+li;{Z190g5dW={GN(WkMA0ly7oTw_gxi)
zg6-sDoCFs2pp?m<s;S@OrG;fr!xwIpMbog?&yh3)MMrmxZTZ#>jZaP{z7U%#XP?HD
z{`SUsPhkWT9xG;n#TG<>(3-!r1l?)<U1(rSz}(vbk7RhFn+2^bFqO|l9_34KZxvDR
z5cp{V(`++L`Cn#7YR|(fdhH4SYG^AZ(a>&(DaIV;_t*?J@`=%Qyhv6@C%FpLRm0o?
z=KGjs|0EtxX2ZIf$Hp@%_p7R(4vSJI31Yt}Du9mm)7qXxt>y6S24KuD0S*RTu~MxQ
z%#ToF_|}iq%lN=+6^zdPW2RdvtzQYm01qSp1nviX_9rRK<zHzXcqy?|jiUY<@0%F+
zeX}q%`JQCuCV{{a$t?Y~zMexURHQ;w78JX7`8jXWg2v2{+GII>{=8#TelJo>(gTN(
z-(YSW)PhC{5}A)gIhp_UtGyJA2U^xH@mKE9QA{RW-pKjB1_uqDAD+ID2bag6<emvu
z?2HhU?Q+Mj<-gZ{R(lFSrWy?eb1l?S%(IMCQXF_JgRtZ3&va$Y6k)~e6>L!GtZZb!
zQ^x%&(DOJ^-{vEVXaNQW22<kG3bw#Ch^wbbrm*f=J&NvjHy(U|4aHH6*8ae^7hgHe
z!}{2dk$xab6S21$qyqv-$o#h5eto#cO3lXHg?LdVWwoYpvyE3aH2BBvtoP#J{S}WN
zolWelta7vhpUC&CbK37r#LM1^I<#tB<LB)G5ehnr?q|zUum_+}G<J2lHcAHWBmiAr
z_VM=ir?FdjUh5mo!<B4iXXkr}Py>&7{*q^6sPaDi11#LMz^Ga*1BcWGx<vUjNpE?2
z9iLzI2wx@vqj+ue16TUH%Ga-%Y;El=t^K}jHxUz5Q08XQR?{O3<hKuSVx4Fv8P0px
zzEhzaEiNm&PPw+(!mnSm<~LF6TY~#S!^n8|0PUWp(NSg$f((cUVa0$Hg0^E06Q`Tf
zFTc$!(aEW)-N5M<+|K!TLz0m~Xh7)By7doJR0ct$nqVM5=7xZ_A;rAh9;j@%G*h35
zi;8caK$Zg+8(YZFxw)M`t9V~$2$J<NblpLf=>_AGj6hsX;=?Es=Nv69ciHt7m{oS9
z)&2o>c@HV;>1H_lR0Ly5Lm15G6Q8cy_!%zON%_(tK#!b*^W+P-{I2{#WMA}!988b9
zpuDh0P*g;2WXnSM9Xdrar=WG#U2e9vc4P!W1R{zvO8fOoiHoncLa0&}9~VcwxED5%
z()XsdE6&2++|>$g(iq_40p^E+d5>!(O&Pw}fUIA@40|^wCMFT2+iMid!x!CHk4dii
z0C+!o0wGbdYHGYJ$P2r!p>f|87^*P~P2jCtckfaWhcg{MlTU6B#2hZ)ynK1KWkSc^
z$qDKam%I1wMI&0{@XT?BU5M4LlnLBsMB>n`WnxHTN*=fR%i>#q=jSErG3fO}M(tsG
ziDnlCc_!I~w@2@?@4XWh6^wbUR!Cf&8d0r=DKRk_L-0Z{-+NLC*$pyF`;iD`2HDjK
zPY;j#6hVX0LYxf<7yuWSg3z{LOtSREXLfP7R_PMuEoqVi5Tx~#LI%dp5V4i)?uI^w
zK!E)4(9fU0juW3flLjE3gPlE8nD(a-=08ZCiMFU}fZ2aV?Y@K&DJm{bPSQkGMbapk
z1ewRa&rmQ#Tz49*Sr5PW!l55BL;>S!kN;|)JyXfe$x-49n84u(H_V&<{z_}qM4|^j
z9r#i?S<p9yHcpEhHz;dUQ%kn@(G{fr_*N`D`{Tc1T-8lr57ssZHA&eZi77_P665Ue
zQu+;n-MQ~Sq)*={k$`t*jirxeWu?p5%G_N0APtSPma_7Vag?WeRdsa^gn@@<=S(*V
z@*#YfEF4_^UL2Go$jKVhGBz$`=fATXye8a{#V=pB;{0%_%*-4^!ejeNw2R5H8pC4r
z{>=Vmo=$1$MROxNjAx%*Gd7m4uB?2(#?PCAPX~)1u|RA6I^*eh7h+oClba~^GHx9j
zIHD!ibowh!TI4FdgrRI7GKG(QpPVeiLGmUGUE@zQ0qjpY5!)9G`w_GVH>#_vBkk<$
zO#1rz0)0pP-~DO@m|$*U&QLAA$(JC@B7qcFeWiquOu)9;h^U=Cal`ULd0EBI!<L!j
zEUo`MsgladL<tR^NYXhPm{#O;@!i%Y2?80#kjlcs&&vZ73p;z(U?T4|k(37y3Orps
z0>}xEqdz<eKTH#*OYKY5C~-MQo%<0y`4rGJXlr<^v_dNNG?tpZB&V6P=}#JyY3TsV
z>^^~Tc@lI(oqgRFnq{TWA0mP6Hu{>Ed+)7~67+ZM*^^*uXNx@rRlc2=kkA{FhRrlx
z3CJ+52sF3Yo4<)~-N!xgK;3!`7P1_>?F%XoFP?_+6w0_GzCJ!-g+BjbDh4G^3>S;%
z{LgmjfmC&OfB)1PpH7AKxBWfaiCsDHRN)vW6;2s~j&jp@yyfRoutx4U81D^l$+R`&
z=n)E@ml6CNI3%Uel5RqMbrBjp(;FaSAk@g(`>4Gc`cssTpg{A?#_<K<9>FRcq*(Mg
zM&UgYOmJ80`77vlks@djkj}0Xl3tyhxHK6EPtgO@d<F7|YH#`<Pr@ODgae$It9Xc0
z-Fj(peo6%7Uu9c?-e*asB8-~ATw%F5f*6TipgFyrs$k&BC@PB9(bAgUsdj;mDT4cQ
z&U>9JS6<<1i?tPgU{%M!HE#$}g`$7j?wF$-=VKTc7&uORlvNbQHRRhL$I~_#FN4ve
zyQ2vE_Yofxr1RjMlLMsdU}pB&;`;SOJZgveHQEF`U?AU3ROrTpu#ga*ZcmUsLr}@~
z+de{cU_oDY-QIqv*{XE3tw5H8_9DdPM(9mqz!!O3rz5n2Kt3`tK5oHGN0;*zS)VnK
z|DBVO(JXxNBDD)c;S<A{NQOa+XIE@<agZoqyf{sC?ZZoCK}YDJitG}C+XE=q3?x>z
zm$kIKi(Gy#c)_i959j&T_iKdVp<xn2KJpIS>uPKL^*|s;wCtll#s}pNG<E3@)BZfq
z$?<soI_LGe=8xAm=K0RnL8R)7`}ses2jt%KrKP1r?AHZ+Y?gwLKwx9A2=}|Y2gAVr
z8#o_IQg<UB2t+~1pdv_2y#dC1YjreRj}Q}W#()DnZbP>q{l>XZG$LXZ1$(O9)nN6S
zg=NMa1up~m)a&Z%o#k&(Q<EIKM}ZoeQJCM#^lixk+)&kJ2Zx(cmW!H75fNL221>$;
z$(~;sq+R)ig@4XD4oUpm+$$Dg1}7@px25$L-w(Jl%^K{FFcSq*5`E~<p=h-6Z4VIw
z*75OU&*XyF3gi;-hLd7XG+34H%$tK|k;OBZARCfJ=YlQLjpQ2LTKk13#*9pX9|H!L
zHE5%HD0Ujmv%zv;1bhKg;#EwnwH^ToNkkC_df$CkyNN!Me;-4H6dpgsu#MUni)Q^c
z`sniz5l<4A5obKNgN>QcBLSLA;A}?jvBDC?7Q35+?I2QtZC|>&cEPv?O!mV$JUkN4
zFiGr%$1h$ua1n~Z(?`4*1>8wI@NVEt#LL#1VHE!bdsHGO_Jg=3zS8MMUlG0o86L3J
z)yO4j>t;&u7WlRyXgsbs2?#t6lno-fU$CQ}`#`u9xa^9ByBEFU3%7H_iEWd0t2Iz-
z660KcjvCD-pElYHwaGXbuB`{^R`e<A&9%|?h<*z~m@i%s2B-rV8WsgnjAbk$E?SkF
z^YjBy3wu>OF4h5oqmD2N0q9ZC(V_pBa%X?k1yUSJfR348zw1Tz$c?k-&iz2^J_1k8
zciaLu?NbQ{_2xoTb{mj192q%d(`NntCo?w-*jN*QL(G0A6`Zh|p~IR508_%l5-2z_
z4XMpEZo}vQ0XSUy3Nu<CKD^FKFHo0|r7iW)K>rlld?HQ``O&=2=svXBct!`zNul}q
zJYZMg512-*p(vJJCc2+0Wba4?dVZ=EfF>>jZs73P*w1e;i>$V8zZ+jLg)<tP_`LFz
z%bB*1A7{e>PzVLf3MTN$o+8i2Qf}qt*%hm$^&;@b=e6J7vlA1`5_9WIYJSpzKes5U
zpP^y%K{6`DN=(Fcj=HazsJS<M(XoTLDw(2_rQK(GX3)8>;IlAc30P3A{$q=<*~(;H
zlOVE#kaKS0nTA^k9+SgB+C1qxS>g1xJYJo%&4QE!DDSs<FGF+rxAz|O_df<!PAerT
z*>JPmxBg*-71g}AJu-2Nj#s~FOLTN_c(DaF87U+NL=(auF=VG|8>qD%!;`5Z%$&s4
z84o1K#ksQMIi)R-FiYI{#3(2(u1UXpm*}9=xW;xb)%}y<jbk4jVf8}?>65#Q+UOhD
zk_FX8*Q@sY;zmkIdkHLLgmD6kQ#N%JKp;%kWgV@pUD46e^e~L>Ma0wQ&JsUOL(C8d
z7uP5RgP&3miH>Ko9fz~*GG+H?eKJ{*a}(RiXx>0bt>B2>K$2_-WP0X&bZQXWPW`SE
zsWe3*!2PyKk90J85ANPouXyz6_$RyxC=|Wkk&!h14`s?U=A)XHNXj7?1FdR@em`j&
zX{@X=GR#>1yVB<d%Up45cTN|D{+f<F#C0F>BT2<Ry*#=1KYfZsuh9(jFsD&a?;K5|
zCRiU>E`~v_Za#Z<N=02g?B74XH(Ul4g^Xvc?v}~2e4|d|3Ip9Y6rZbMRPpN7J8YrQ
ziVAUrVdU=H6>=;g<Ty^q&%>1q1liCX|EgaXlAi{O1l%UfD{H>6{?}Q<9ua6l+hAZ9
zO1LLv&M@I_Ej31w_<-qEq4aBSD>7NGhdR5!rGiHJ|Nry<{U!Kr@?nQ$>K>IS?WXTG
zo>6<!O;Oj)(#-ApISbe8_zyu?Q25MgL7~%vXRio}oD&s3Cn~`!D0og#(B5kC+5h7Q
z9Gom|ti1o<f8fM<)x-D!apG^l6m`zR$=Tc0=7zN!LE(S>iA4z0IE$R6llgU)%icHL
zt~=ahx#(!&<m&9?iVtHcWWVEpw_y2SZ((`;riH7Gvzv{RBjNx2DQ#tWmQ#2a3GU+8
zGb?y4>;HPKgPEhdm6?T`yX$q=|Lw&gG2vZ!@v;B=#ZH#j?f<t|xgEFf#H$Dw6*Vpt
I%A1D#AKQeeCIA2c

literal 0
HcmV?d00001

diff --git a/docs_new/logos/datadog.png b/docs_new/logos/datadog.png
new file mode 100644
index 0000000000000000000000000000000000000000..dc758f42c58eecf3dfe904fff6045a16bfa4d194
GIT binary patch
literal 32779
zcmeFYWmH>1*ESq1c+uio+@ZKzOVMJ*Em*Js#fp2O#VPJm+=CYn#ob*C#oZlV+TQoG
zzCYiOcdh5=d$N*4W)3qaGkf-2*R}TvQF;Fk6O9B7003ah$-Yqo0O0)q0Jta=;B(2K
zld;Y7iON(~O$h+-pa%f_J_7)^&qaRQ0Duc80I&-J0EChN0K$)HjjAHg6+jb(cW<80
z&&Q2)lkf8h*;!8ME%G)z86MXQE4j9~=OSz8x7yBX#;#N!9qqwZ5OXSL_mAdO#vea&
z0nq+@qEY=lxOvcc1bGBs8*_4U3Ov9699*27++5F}ub=<s|69Vz^{4*-bMWx<fuCRe
ze5QQ7e~Qrxj6#zE0G}ecH<BOR=J%Ie)ATpoubxI+GLz?)C+R#i@`%E1G5CP4c!Zow
zn6&;5a;LcrZM4cU3{xTM`#;7i;pBKs7fBDh`S=XT-^8lSSZgg<Y%Z>>P+EMdEeWQr
z8JE>Rnz5;EnrE^YPUb#KZ_=-4)2mYi#C#kU+)8u$!p@CArG@I-##@~UH|84Q^jx|~
z%_QrQlbrr7_H**X;Nau*V1KK9+Mk*iqA`vzad{yct(!JWa30=!fp&zffB&*wRFDyq
zi-H~e&~{Vq@TYBIGfeZA)w`d6x;c#R`gdU(vc&V(=W4joKjp*j0Ovo~I#B@9|CIjk
zRAc%5ZFBfPU7i89|8^+?5c=Eq|7!LB`#t~n)AzqR0slX<YbQJG>hCm6-Y6!vE{usV
zJx5#Al?%>NlA-QCo-I@0(A)6r75sZ6#iFB1O~M$oR<BGrX?~+M<cffX3V8^Kut*rH
z9fG^j=`o-U8l&N3%<AZzKvqmiUktZrqy1xB4G0ItMOXMSUJWvxy4fU9k`Uc6YZL-{
z43!kBBtZB(MMkVR29#RRr2iCxOO}VEs0vNZk~a~)<d9a|rqibm>CCtzm#x2MxzO`|
zjped22hT)3<FavnRDZujNq(A9UI{!6`i^6X&m--ih`;WC0@EIK3p_zgc~uODCPBf7
zf);%0XWVh9qL-5~cey~r^kq!{Fe4f{)aP^B-+{5?-x1rF_VxTSh1to8jg~OdB|##s
zVk?!cpV&QXU8wP$H%QZDetLRh!n*LKB|`VB2%H-K2)GoMU+i!J<|}wqAmIhHg-zw|
zVGg~9&&kxff#bT*qxN^%)*~cRJa&z$5G8&F$yMgZp{vT7$Q`D%DUn)t&T7kcZ^-qS
zhBH)#6al-bbq6QIOXtEjOI?(Q%-FmIxo7KKrvZxbnZULv8ON8PG6^<lR}}QGA2@uu
z6IJja;yhhiRLN6@GpTZ;79K$Wfv@C|>&|VqiovCq+ll*7qf78T*muhK_qS*ry3tB+
z7IKjAZYFtSZp2y+FJ29Ec6U_szviMZ8pL<27<{fD=7|*8ii+lP33BN>bUAHyM*B~4
z6{C}2tF-7zpLJ5-n8?6T%nYjO1KXo5P>iXJrA484G`$Du{hiTVZ%WQAg8wy`b)u|(
z2Z{IQj{7Mgg&Xq~H$IM*dh%V)5VTT;><c6uw4bx`)U7SaQaEK+Ccz?8qJlme&Xk$=
zBYTqXx>iYH3F>B{$02rl6)44`6PD-n*O)Y-Y%IKKj6NtUMZZ*2$CTtHFol!qsBZKP
zH|IU-rg?q^f!2d`F-RmZC1~<JsvKM|^o-8x%Dld(JNHnG&zp@so8%uiYWL*9;i@+f
zZlWZWEv8)>SOZr|Q3RJ$Nz~L$6dubp41)iL3R;&aOWF<Y-j(AGo&rq*j2BofT1#AW
z=ggj4_-i;EiTf*&O{oQ1_(`2yHBor)@P+noLiTKs$7iLAr>SZ|))7^jrITQ=ZT`4G
zJ5iv+oIJq?DGDU0JQF@$d?qH(_0gxdDvNfU9dF3BCu-k4b+(=_rZyAR{JY1Q$JKDp
z(zZ4t`OndeK;w4CN#<e7r?@#p@Oh-s(9e_L2;X_0Z!r<5qpxy^E#YkVa$Z`>&?eO-
z3#j%nNw<}po`J05$PI1*?^fDv6M}vCrRRmpq1KJb)uKgYGJ5mSxiCrLznjdeX@v(B
zQ~bQm_jqe1&9FkGsZQBI9io{fy)+C(KLJ%vRm&J6B6VVWtuS*s)z8+LXtL5{O5M=o
zy7lC$g`TF}bX9bSlDiBg*aoLWV&v7{E0${!OaV=`5}kOF7c_7YINtvmGvPSz>q*aW
zo@~X6DS|!gm=qJ!@M+q~a1-uARP-T)E*IM@hp(vp{n_E%L^hJP(bEANl7fM0RWsl2
zr}b4)J>xeyB&RpDr*IB8a>W52Fbc{04`A{5pX&K)T7tTRI<l;sIs_}P+%nAy0!gsU
zY(RA3ThX<sXIF54eaH(L%9K+Cg_MO@Y*bn++`u3&mG9A$2|2n_;kE&MEH8VAYBsVW
za@%;7roiAb`dagWZ)2e*=pX3SC9u&XuxV-ZDn3zwg4rFuj+71+Kj{=0U%i_n%!M+l
z8axdttf)VJG4)Nd>D^e%WcO5WZoKF@NU7Q+=Dx%HSh@NT6A4AA31p_jz*9+;a~BW%
z>wQX05-HE(RSJiuHc@!Fva#PR$EcM2C{h}C3+RShhYycZU4J=I2Z)e}&6de3I-N>V
zg2G07<C%fJ+@3{ASN)oEwg|8*%sa*Lh5PM@k#r8f5{pGs7TXCfA}_OP?~z65@nYNy
zj}@2u^fs|l&5?rK5$bZ8lw?dh5h_nXxNgqUi|M~eLn~<4I@TlH)#QX%GUfB!n=a=x
zvkM8#d|fCB_mupkU%*^1LDM&uK|4wjk>i6;h@Go)9pzEuc>G#FDJ6_-Bc&=AA))pt
zl~1HFG(UTB?;^h`CyOO_n;Pf5+@~;smoMI6S}FMP4Xz-kdD5Y^qNfSFX+u2}kMM61
zyOI6t;Va;kcqhide0|}3D(T5V<QJSCum%2KSP-^Z7x$98uA#6BMy#j3gQ#ZZ*JYji
z*VsSqQzi)UgP$I{gA(nm_77SqtHt5&=n7d6jt&@zYKZag-Gu&Fx7HF46tNCfv7=|r
zTyo?EXmDcG)ylF=r_RV>U8UDYstRs3vL<fvsVnGcZ)L0tH^{CNJ!^&Ks)xJo;m#tn
zfqt{&mpE#-5y?Fxy8oSxFQ&^ie~RVtS^uA{v4J@u>80eJWS%wJn_l^tX@$!gVg<#z
z598=DE_>MhemXzsQcmz#8<qvU*55~IsPu(*N#Xfj4T_x%AL!1Pd#*)PM%!H1>RjW`
zDz{(iq12Q{f1)<AC1D<3PxH(>uB03$|3Eos9ZGOFol`n+ETL3IPM{{_{}QO|CmOoX
zq?isAgm%QaD7R^t5JZz+$W=(cO;VCTqfIJR3)?h5sMzPyL-f@DJmPFM9@><cSvyfv
z*IHe-I6qoHkKzA5?8X`Gc(E?=?x*Py+lb)Fu+g`u%#KOFeV4lvm1z{+3c-^*=%O;m
zmZ7RW!`~@$Cg-k$%*q(wwS26I3!;)^ngy<U!hAu$qSW`5-~M>7dDW6iV0Z#@dVPGB
z<`A`-iR~MF4L2%YT&Ab9r_ZcyI#<Uwa%-~#tiv#n>Fl4$2yTmmjQ98ky%KNr9249r
zFLzowJ5u+1A5O7~h}UztU&YVXR>P6oK>11xS+|l>PYFG#JNM?jFUQBjm)&}Rbm6aq
zZBG#?!Ztny#7lRkZt1}VZKZE*5oRke4!<1h#pyXrd=EpeYgzUcb%^)FRk@x1!9e-L
zbq#OHY{PPR{)$qBx0E*Qfg}kTgXlgFTHxQ)tNSVyNfZ_<!N!SD3iYhFHD9f#zhnJu
zzx3mN{Bdcnis%DRDS8qZ+~eal@9J|Kly|gf{IS_#r)$Z{g7q3)<3_g4fqik89a0ad
z(LHAe{q;P>DY+S}Umbp0u4vVjmtCNE609@OA12P>@v}91Jt!Dl<y1Um1eqvb(}$TF
zUBRR;f*O@2@88K495%uC>M(&pc-t>JBdCA#$!~}guyR*(&<x`WA-Dm48i1J<ZYNmp
zEmQTX+j~T`u~dOJNGtvyTTD)ShZH!L4^9$;UujcD^YfH&IM;#~8cq4xTT9P!xDOk%
z@#QDz*sD~1BI-(4uZZZx1Q-9>^zI8ke~MRoEDq)?D`L_UZVQ+c)GW^TftEY!{2ce7
zqr>I*4W3SA$&s(!z>n(HVNM0hH>PyAtldWFk@dFBxQ9~-V5e*?{|5&7&83>_@e3y~
zs{q+L>UZzRX#6*~B;0pe#4B}^jS^1&Q^B@_njM9#<`$Dglf8;rhIpi~gp?ES{$GUr
zWsR2HOAVsdt`S9D(WACN+zK>FJlB^1xN<$OzLXO1(#cA2dQy=2Ps^cQO*%lzcUQQ-
z+nLr9YP6vu=`VAJ!t|iF7tj?rariq_I)V#RQCHR{P~$s(f`Jez0eDq-&BY)4<I5#r
z*6W01-EU&l``0bj$csV}<z}-aP-XSvmNM|$an$Hbs2I<VO;;q>60d2Ko>^l#{pJjt
zLdHSnq$i}c4!pV-?8As-MoWK6<Q{9yWzCG9cafxBz|+QDpmtO}a5n*&TNlM+Wy*1O
z+6Yxw6?#a)+Y<MFI2~{m4Cby}Tvtm&7ScEQxqX4C*Z!vD#_i)jgGptY35)0d5Hs2u
z0n|}Lg%O3XznmfusGYIv)2y1sbNP*Y8)EP}{bJ12-t(~0LM(2%>beS3Yxu|6i|Sdc
z?JK%;Nc8@&c0KD=k>w~mg!lNFsS12Pz91zq80B(n(93KqMZt#yFH{KI>pV1Ow>jj>
zmqH6%KT$eGT^E(<>kM4EABYhf6OXHeH~b(_=dr?=ZQ!ExK`mXHy0G&)9rrJJwsslo
zPrz+>72`NvxWFvI=n}Zw`+dj5_zv+4L02deOf6fhN-5@G!iq|7APtIC^J2g9>jRN?
zY#644vi%W9g1%bhcXEOJ573Ta`1v&j{y5V|yYMyyAI8}a$p(YHT<*pA1_Uvou-DKE
zv`53LDAYSroyT(0j2|_o77HKGc4s0M-*N1FWG?+UB|9iNMpjWY!Eu9QGD+xh&i~4w
z&wFFTk-TYQGD5P5E+Xzx9e^a$!pe7-ys^j<Ru2$zjkdDQ6oOi<nI3@gzwH`6DymWw
zgG7N?f!<-AKH90u1m2Io6o8kPno%&G#HJaA4SaO_dst1yP?PD--7A=_0{d?3?Qh|>
z`1~D{z%&2)!1u8b+ytfe!p|AFQ2gu4zShVM3OSA4MwgR~jlf^p@lBq<)1lfCs!&`(
zD4ySnf^KM3b-r2(n8D8DbbNe~*!5l?EDnr>+qm>*$%31do=0n5wMcZTAX|ptUX<LN
zAT1bBLH=$^x#dq~iIZmbjRx&xqh7c+e@M?v<Q>%M^jwMLvHp2@c$S;DLtnnzLeK2j
zHwE+ctR5+Pq0d!=umT<osFG9b7mc3R<tehAuJg&JUcHX%mBdXH|1#dA$L5Ud4YkKI
zdNs{#F{?gmApNMYshZdscfsRIqKJoWI{6+(s}<DbJmE@Rbac%k!Ks4v&#ko=A>@#A
zxl?Wiko78FLi<Ikibf1C9i1`_FE0gLH=F4Kus(wd3<S?EFY4KESHA+T>Z1(stNl9B
zs|%JMBv3XS5NqECE$i#H-3I3!?NoFl91j8q63dQ`1-TCgamFv6y`;%m<XQYQnkPz9
zFsj*v9}7W30UNyL!E3YDYgB`Ys@0jIE=ttMYVSof%ek7&+pqKYagKjjLwUa(glqQR
zv08$!;wPZ7&w6PISN<gNV5Ib*k^f=coBNb?GlPc7p65KFbDtViz>8>a=y3u#HrXgC
zE~8<qB_Y=q!Q0<&;NIfkMvq#89*-aXN-Qap6TSsHw)wk#dn3w-by>$qk;?7VPH3;|
zJjaeuuI9pQAr3I`OR7;Og~yDZqVdAWa;o55M%rJL?1jDuor%fDz?b#>fZIQ^%{qo*
zDtm(l#L#{HAo@-1YKjFpAJ-TPmb_(~54~La=qgpLt(7N^qeH~xD8soPIyEM?l$%U;
zXRLu7S2yp(;?N#KAxOAYB(u+q8t>&u^YPcC{nJ1<YhJoenqY#%J;T+0$;<z2Lod+!
zZMkNfuX4^mS6S7Mr0jD`-cAmMu^V|!OdPFj*<h^HT#Mv=W1uY|Y_5!QT*`JNYOT39
zxIK+gm|>vyLRg0juh$+GpW3`zaXA=l>b}?N=x418TEtylkobZKCh*&7zwW5(j>M}V
zf67a5C0J=bmx_<T44`ZmGX$sr3x@~)!@#C8pI#YAd^vdgDp~^DpAKmBb3e6cOfY4^
zhFSz}7p}$sO)OUJy2C?k;-@~eE7Hp&qF9ytnV9$|$j(?>(#e*^J@ew}Z@P!}eM3+o
z(75g=@)6bcRT$nv<m}WzBKm{3|1tJxfp;=owVN(Lcay@qqK#7EFB}{izn9R-zW{kI
znpgmVq+A3DJ%QzDpDT07bY|cNBzw@8-Oh}j4x{xvFE@h-U)t+C-(aa+<E>iBQ7VsZ
z@m+{|K7J`!PT&t3ufr4gZ0(2j{n{5i(ri;e{c$-*i%Gr&n+1}&C9YVgf!XCDUi}Z{
zZ{oR9h4uooGenita2ZLjpEQk+fC;qJ^>#e<Gb*TV7=;MKk5dmU-4w+=-5xeR76gwk
zs_oI!*>}`$^nWli4+Cmna4cAYQFsHyKF2?ThMCSH=ih4Y!Rrai^JeSe|Ie^0dMmw#
z<VYSF0t33`iSDf8ysvH<*eRywJsQ-Ec;;Spc=&UVCuMjQQlzG4v`}d%d$IODE<x}7
zXb^odQ_vxVUM#Uv-SCu}$>OQp-j5-PiY>lPGn?A))!G^3J1Yrj|D$YZRGV1G%oai}
zB1Jt*_<Iv{)&*CWO_<0pVpwL_Qfa~oB}@EsI6V*I3s?&UKSYfTjTnBLj@u>t{+k=L
z0v{QB0{--RVR}%456&4$(swFu&-<cQ{fH$T<v|jOsMe8fh~NaUy~y%)2G04AYI34l
zlGhfKG+-=N31M!X+U0G#J*|&Jk2<B|RS(}lDS{Q#HpZwRHNERc6b|vrn~ftj1BbO4
z%8H?@NAyma(FNCw>;(+kc|=N1=MPLPzISQKu8+;p1cqd?cTU<~c$_wIeDz4`ftfmg
zRU3`UA-&e&*>eQa%2^f&M*>3Kr$A)InxhI^KB;;|1xcng4FbVONNd|EVYAYgU5AwV
zd(oX#os-{V+`d=J-xq<+K8bgX!qM)l`f}XV-c>}m8P%WRw`*Ugn&ni{rs=|-ajPfY
z$@7wYAtk;=s+O_`$9L#JP2}GW3}oQvU`n{0;rpWBs$8}4l^qn0zv6iQP4q_kL0c5&
zP`p>+`_<AH`BE?BJTEGla{@7t-*gbpTNca^xjboQgLFn9@ee*2e89z=hJHa+nY*nJ
z{TX`Yu;F9l-Sn#x{UxZyn8R;_bGY0}K6SVv+2YG><?UjqYmZ9ljo_?>zd45iY@vz+
z?{##<`=*uYyYItUeIv~O8r7FKiql5eS7I?#r<#14aeuhu^+K&R716r6&@Sb1>5tRu
z-W!czSr(@XI-;^eSjbw^?d}5RtZsL;LDW>R`8Ew=DNe5t|23u~$Y@ducbWfLfs`Q1
zp!FB&oeSB=@H!n6^90MuVj|@|8v2Sj)9FvY=?>cWDezXRAg|uP9GeYcV>$H!|KqG(
zjiH!hgO>sT&E4Tzv=*`aYzJRECnGHbe&}?SZF!GCQ5`5qtDozx;wZ)On?>dCfu9)?
z8hCJ+M5$3B&$9F%U9otMgD;cjdl}q94uH|0_yV@4&Z4xCx!`qFoKTgmBx^AD+g#XC
zF3C~u8p|_}te%6dK_sI10TYbQxff)O>Q6!#TeJFSC&VxWb~gl0Ca~GYu+5^J%ZAS_
zsd3J2U}P6X8-Cm<auxB}Go{>KqFm75RD#{JXOT|Q3y@7zUi6UN;Iq7lSU(-JGOD)B
z1Ja#r%Zj`x8qU4Qhw?<!{Yo%TjGrzwm=kwD{UMB*&atnQBh+K+x=tA$@!I}*dA$7j
zhX*C;>?2QCyAH`lz=UIX(jJg2T1~}~uw&n-a)OA0>8T>OnRqE*v-$BwD}0YPBGh=q
zAK&tOig7T-IQbOgcA!zcN$I<v+aaYNTgt~nIcY#A+P*Hu?t%VKFhlmowPa&FbXS90
zee>2}c_xvvt&L9`TO#|^iJLR8{MRTd#pQm|hkQ&TLB%hrTyukwaKl^^Mk(h{n4BN2
z)olVXC=e6W$sb^*CL}FLqyK1zxW6<G1QXgb(4gT#ZX>bvPGQA+s-|N4l8dvL&4C52
z3dBP)P5W7rGf6GyFhP1FDctu#moxIsr#P#<L!N5TEdFR0_ET1y-GGDqzTlt_8Z-GU
ziX~tufz~BcM0xXhPQHjyB0QrLa)83C@(IH_Dvjw(a*p#ceKoH&toL5WI0@X=HV7O7
zx2bSM$ThR8K-do+xm16b9GAAlk>(Z^?r7S{->eZ`Nt3RG_TJYy4m^i89RwR8WKlaP
z5Jr95A)W84`73xQ4)NzY=$Ay$x0Z|kBxp%W81*7x+l@mbu-?u;_y4{&f#agErh9f_
z6dXu>d~UfzsWa5r#;_Md_L5`roEJA4zTW~*6+mR#EoKx9xh!>Gmr$o7V|T8ZvJBTL
zil#M2e|YwwyiUvO`Nuzp!Z=tC2ks7zY#8+pB>1(9#`_zTW?$3J`=Q5x&c2`UtTo7U
zibdmo?@r(^SMu9_Mv#9*PmSGbWkYO>k!Dji6`A76>xD2POElagyq(Pw<>aR#Qi`_w
zpt0HI4;wUo>c%z;Nh;Ofr;frF+D_w|oa9)^1D!KE-C@aDa$s5v-U?L7iezt@wKo5V
zyHl>^sx@+l%R%=r|BbKQpT-L<V12KJ=RxtSNTVfKlV^<R#R*AP%~D*(N1BNae_iYW
z2QWQ=%?ywE@=;TTeC@lo=2jZ<S*zQ`8jI%Pz(KS1rl??%fHbF#C8Ei+lqR-^PQ}}!
ziid<+vISAejqw{PLU;>*-8zDnxnZH)6#~{BvUt<)W?Tu6va^aR4|RIcwT;YC0`r|F
z#HlZ~V`crm0Ze1OWbaEI4@b9RCDMeq5wcu}Im9KBh=Y}|Z#lbbXEGXBkG)g4q};1k
z*|HouNSUj3+AphsZMEdp8y~CB@q!;o+zO1#z+mvgieuh_AQw#{fsDn^jxRImlm2t{
z{yn+rkp<`=(}`9u97|*t3g}ieaRBYlS;y%A`kD0bLsp6XK>s@(vXocO8bxW3?Tb3-
zktt(~PgT<6`&RW+z^-}(dpK*E&Y9At%I5ld(XHRYq$b~45S?I=yw%W>2K@Mh9;O&}
z4W^9ikl41?5$RRIRIqGS58_pT?c!LBz0R_xZ7VfeDLRn=RGF6UO@z|K<bMGdC7?{<
z?jxDrv|YDZd#&n8g^}yEw|Ygo+koMh04&O@+BCX8J?u;XIj>im2_1ngm->^T=qJv&
zjBw|doaR`X(tQZ_ap%(}^Wx0BD49m*b?i*8_3Z`$-w|Viy^V2^#>ukmM*T7Ubj%Mc
z^E?03s|F<i^{yQ0qYX+Yls*tEb!SDv)MmKF<2ImCv?aY^Vry!{=5CqhnqjmLOc4A<
zJQll-;OFM^tVWGA3ab82hjDpMhd=<P5*A+_hb+kwk%@Bre<o<iDL!MCJwL*B!#9^?
zP=ak-mB{w!TX*8JtUDDO_9()@PlqpI4E_SH(IqT|jknV72s%5F-nZHBb+WftIs~Y7
zo4tnzZ>}cuJ#NnwDrU{+l5Lb*88X$xsPj*7zdHGMe0n`tVFqOk32G^Z)Bc_Z!e};}
zHLO|Fc|pXWVBJSQw(oRCnaZ*n6!pI8&ffI<$h*96mPK^Ty#g4sxF)S+PSeQvSijF0
zg@vZE&Drde&wJ-7%SDH5;QlF+Wv$TP0pt70XY(uuU+>P-%JEV7vWujAF`COce%Vw%
z^J-LYTb8qkRD~H)_c;}nAJ(@9Awj3V@@m}5S4~Fu{HMvm;{MRCq=%|e(GvTSjv`2%
zcMgbzNLV1krNVK<XZyI>Zl^09_LaD2X88V4J>8r9*Ga^ef(l_1RenwYvK>xjKSe20
zE#G@>)6xr*(SNXqzmhlyeG)n%{n@`yQ73Un*k3EP%Sg3t=Ua^SYtQNX+AJTuT9pjC
z2|-*7e{vwLaIq32Dzp|jlUKFkZMQMm!Yq;&&HVjGVw3<lo0TFu;d!t0cqgCdceR()
z23!O0Yz%+#nX~p!ka4+n{P?d=VeW;rDfr3Uv4+5G?POWSpU!Z{nw=@EZ&Y#dF_hrw
z&D!nFO7!W@;X+|TotUV7>Fa!C6AiAZW{s~K3X<MT9~#_cHhyC(aQ3rU0joMAwfi8a
z&=rbvD*V(T0OU<`MqfQgwx3tvq9iulf~3J64Q&!?-*{>5H@vE>#0RB7q>TD7ENPrF
z6cMyfHYk0fGaSt|4;v7rk^J1rMMxWc35WJ<I+1|D4WhPH_lUK8EwTt;^K2u+O0o6T
z+zGCI$7Q2}=TTJY=Rs?vvV^#2K;ioJOb0ehj>b@B6)U;*h=vJfg=?bNmxIC?&%xWk
zGwJC<I0?Wy19e3AlTcsuSqACXm%*fjhPPD;Ga|L8o%q#9@GL}J=(?N`Phh?D_nFR#
z07_CXoa-UN1f!3znU<Er5Uj4;UK)BaTj;SLM(e=zMhUN_g4FpHaO;J`A3B~5b37=l
z33}lCTIxv1GHCrM$xIF>MC^<^TM5(7qJRL3&>`edM2MAya%U^o$ny2z!quEmbrhF+
z!9sHotSsh%$<nKCIK@~8ZJ$pYi&aVljg_xPNM<_rw-$Nj_L|s_@60cLxeQb=4vF1f
zX6gk#)Q9*2m4pBY@HXwbv7JqRe*<JC9^XdkCgFNznus}Vm_m8q+TQ$BPZT{5iS;?}
zm2g*5z>vcuuu-CBFd=ypV(Bf}5J{R~u*j{O>NF9nbv`6!)h?n=hJ0z9OK)6)7)5+Z
zqwe*JxwO)xYA3&dW2bm?-R;bYa{p<&RJc_)VF<sGplJEh9%az9=CjxEe?nVY2OkbO
zOd&tFk#}XThB2r#%H!a@9{bMv&1`V&g+Rht_3pekY%_Rk^n=}~KHyo?IO)62p@){I
z=Evf|{M<elbn#Ixi_K;|7#MEr=OgoA8lC4G(}p&Etn=&SjX#E^|B_oe08`><8wi6q
z2Ke#Bm294@G_M()T%lu?GLlbS%xM#G4p^;r_i!j+%@cz26#A(f49AUexd)*ST&)ur
zv?+e{Ra>!tBb?>dbPbkPjX0D1jc4yL6^Rm9p1p{KZ|#1-6I-i}$lI0dj{+5X_u1zu
z>RaN`2}yM4?{L}f3K(VqyfR@CK5BxlmS@R^!{qSRp`lq90n1~<sPV-d4ceSW+@J6b
zo@O`5!F-hacfon{l;}=o++Hd0Ja;%B|EVP8T%4+(v)Nb2Mo?8uJ3Us;a!T^Df9-%4
zU$Cxl^zhoO+J1<87R~;0=!3`I1}Ag015gXeLGX77k1m3?g1W+&M0Db}PdBMoZ<s_-
z7VS+Yqsq)L$Wl3ZZ<byE^l=EH^6<ZGH<CHB1B_XQmTgTw%wF9whULu<^|mrYi1!ow
z1|vX!448WOkMk}>MEdZG#X}!Rdf?PFHP{_Rif3GQbl2iA|I6UAF`sfW&a|kGEG`x)
z)}%;WdnDj(;)|C9t>!m(<AUl(`ts}#z*(l4pWD;G8jsxksPGmRFZT($v8cGeXlRNI
zkY~cF_eB-BA!Eb-F;90<TRJwitm3r9lSwO6s6vo!{u{3PpmWk3KRt#7slSX$gboJ%
z?cT2~r~IL(TO|b!N|F(s4D80(%WKimDcJ9^CAE#gQbBP#g4qwE)y?UDdlAMSl+SXZ
zmrt6CToxq`CV@pR<Dy|oBpy9{6LYvlFb09713Tls8k%=<kH_SjD!w+wundRN0|xqc
zVWYQwm+$FxG4CPAm#z0_VZ5I60Tyy=9k8Q_-R{)nEDbf$|72N|7xzgqVb8?4AVrn4
zRw$xd3C$s_YC0`*j!YWAyS%9I72URlm0&8t>-wi9%k0C-#Ic^rO*w__+yIEZJ&vAC
zl??u8-yk*9sC=lBEUGR(Z_<T7lVW-i3b<MpvhXFS%rcgW>~i~8Pf6sH2<`XRdv#dA
zV;#mvklCvGVxFu>Ci;Vu=LEP1%-1g=sW1!S;nR7sPNVw!Km-{#f^FiVjuJ?QYgqE1
zl!#|LD4+L!EINBR2F|h(r>p8PA2hLznn57LP?kKx)rTjuQ)$|Y<wb<DTQjKJ<X(Wt
zbDCMx{B=41khMSYyaNopLaf;zzK2mlt1i^9m#-e9P5r@xvkUCHfeZmZL(@Au*d%6r
zl3XYDpD+Itqzj}+9JlWll4(Iz0kp3=bCbJX$hqFxpsEy2)hEI!Y4c^AzPrha7Zzy+
zi|%Jm8R~wEg8O@?hgsM(-ne_03X?q8HT}*QV%eoJzxIWLA7Ew(o{gXE`K|&OFz#*>
z2@F)1-A|_<8Ji|Bc&_1G{#rX#H2sYj#1|398X7%hJ;db3B;nT;7G+5HQyiY3X4;a!
zTK5x|9>;!$f6e7ide31Jpl>@8j>j*+m@U)$fx|l>>K|;U1So8X;q6@wq??aX*{h*%
zIIek8`V1u(qk6nfdSI#Y1^Unxxy~GXg8%U5{;TCmv+YihFxz@gX{+gwWh7ipsRKVw
zxK5ROdj0-?2C(jPFs=#BJXS5BdS8vj0rA`YT|>wF5~J%-d#>BH*u<WW;RZ;2pPg1s
z8(b6dIZq)gCcUJ_r%Ygz4<+x9rMPD3-clJs)#03~SSf8loxc~;kwv`I7y2jX=ft?*
zsR}DOg4~-fel7vsT({Md)?AUO2_^QaDSVSXc?MA;Z?9M$1nwWUl%5%1zRy+<STu?r
z-`A{r0kMKWh+s_O)5rd=JzRH`u|A*H#}z`G-h=rFESIpV{qQ7H^W+{)-#yZvPuFeX
zhHA2Yqd#C7=Wu<jy}J1q`cZmS*QL_b{$s3_W55H4zAO4YL(fT2%{|^KlQ>#32Zv{Y
z-w@2(>3!>70l`ELP7hoUXfg!a4mIlG(Q#ZK54-#1qL-N_#4&^X1%2x13n-IFQC`YJ
zvC>`1<`tO4n*H}dPvv3fm+kY5%zs-PGkDs{lIG)@&kt@|`fs}?;W?<R&DOU|HmAJ(
zg!7K%nRh2a6!^?`9&0h5vdG<juNN$*$HfOPs=iBXJHD7Xm=@cm4oB_O0UDp5Wxn)t
zeV4Zd&l{f-m8$5BcrHwJi}Tl(%0m&ym~ssEzkGTH;L+?}@Dt)rwbRV_z;3OkpSp|$
zAsO;~oK@v#*vP-_IIaA&PDs#JaRm~TF>>2q!<TjU!g{4+xK0bMq>L;frXQmn<EaTq
zpQM^3praH%{JV+O5`S%AyV0-o79Z{|go^~mDsEBF`_Bw~X!O`@7Pi-Remi5ibh2sU
zWT~=a`}THsYdGbpEJbbTEYs+dal?GnSS>ffWC2*RC3~yeg2She02ZZu2QJTK<Cc7P
zRH5A#7^G;;HIIKh&iRatB9Q*Xl4~G`OwKg(bPF7Gfz=mwY;YTjX-GoE6BvqG=$=FG
zBJVZ}k$@L*>T)*b9=@N&k4CW41fTg(aiiafhWb${2e3s{!z&R@bGA%{(t`wZSP(W)
zJq0ME{*rj_@XyP5Ic<2VRh7^5`jC`lnoA%EEUsz*T>_G&j@WrV%|uQ=EH`XaeGSSq
z`4lBrk+{~zSc^53A?WMKOOc8-l9tGj+4_FGHX!HaJ{{KRqLw!e=!8hlo@{<52PZ$T
zf2KF{%cC7)mS9|2_sHviO40-{%ffd#8!jL%t06{hH)y%AkOrJt1L~hD@G3LsIlS6@
zo9Gfm)P~`O19^QslI5GuZL_RHDR}uPRZA9K^1fxlu}UlXS~`#k*75B(-5zgns|oq>
z3#h4)CamOMuSitj{BC*#MHAi=&JBCt4r&h`q1K43FxJk&(Y-%+JY3x}eO7{K=R8#S
zM{A6ZrhqwANy&0T)G`KsRtCC<wk_7T%{M)So-<Dup$&I6R;R&Tl(UW-lSuVJ`uo5K
zrr9rahOv_ui(;;u_nuA-i`+&Y_iN++LL^S_Gby*DD!SHIP&^Ti<FVa-=La2m_B~)r
zI}mJpsRSNu9t&dzo%VmAvO1o{SHeKK#W^4vfg3^5GoGz}RgPe*CM(3cYsT7&#=56A
z!mP<3^D^htKLTJ3SOn~r2aMfDWTC?d5GtxZFI!m8+aX`xgfHnO8cg#(qpykp7qmxQ
zH@0I!oOC_;0arus@c?9%C3JpPFZ)JuS6Dp5a)Oq)Mc^Ze0J65}@kd)<<AJ)whr4wj
z0!{u>TiHBtS503hQL-37$Yo1)r0$7Q4kY?hNP-;fOr?h@f-FJ~eTi%Hb|0;3*(@u6
z>$ZSnpVCM8Ihvn@TNy*Nd;p0!{l_Bo<Gb5>Wsp2BjhK?sA@42^%tBDJDtp;g8O4Q?
z>!0i}R~EJq@b|EMa|?XCti)9?5pW(c1W3>6LyC&d>D5QPoeJ2l-+DLrtXpLIj1(fb
zmlQ0|#BJUV`(O6}Eld+)CmHw49(WN>ignHsbteIwK%%B`%?OQtvnuvACMk8loIxV%
zhly!_cX+%x?GdrS2+$tpbYK~RW{R8SEIQY*Nq`HZHC1nDgTL2F*oc13z}&c03k)_L
zXhlWxKSG2WM%`FeDuO{J4LYWanYjsuTP>65(><Pe&2{(2^Di*Pf4cI_BpAKxoAABw
zSRQX=%RRwrUj_=n|M<|oCp&(0MZdI?%>6L@1^?`TpQGtxrYolJv4#;AFG5Rf&P5v`
z521hyBNnNxEFtPJ_c;$Z9#D7^FfEU>3Bvv$=N<Dl3z6CXWgnV+1yB>d1f~gZ6H>}G
z*H|?doHj<8Tj)=_d^||!ssBST^gFR$UEOQ64idBfA{}O(B6J;#jYNcIHW1^Tmi4!N
zvW?g~ka9oOQSykWm)#9%rsuWRmA91+!>o1ZMpg*;6PaasXT#_79)j}&L2kl(J_vEs
zNpQUV5(8U+R``={IXKIrk<|yKS^-no(M_w&T$mP15t05i<Q4QovOEHvd92&5Kl`XP
z2x(RwR!v|VQAM}OM_PB2PnURbrW*!kh9eIg$>`VQOR>Df8~lqmnDI?ygQiX5{JdV$
zj%WLtr$M@jt1XtAjyHpjC~NIzU42iT;lj#OYFCXgMQvt#Bqk8uM|nCT2~WI%SY4YD
zSI=(fND{s7G@pUvXl3E989bf7u6B`*<?!OzT4m|DhePI(-ak1WhddrSMLj1pipH(S
z?xuU3mMlbjOMSjlwO!@^;Mh6XpEcAGSdwKZb5?7yY-)5n^|o~|^B?+slh0V$UB#t;
zQ#{AX9i2kgBBOcFtfRGA*l7OCd{5l*r}qp)P^&qJlK<p|B%Jy%sLJdp==%LC&Rd^9
zd}xK}vaoBh7?d~e5>_#5b+y9)UbGR-7Ei3^^zThB8oxqY#;i%Q<p*<+#BuV;mGbSs
z<<^udMf7YRJ;OGX^(2|za9Y|R)zbHoK+seyGYjF1Ditn7980(?3NCdS|EKQrGJIk$
z)_-B+Z}QU45CBG;tP_W)f+c1)W<4#QTs<9cqqd9?>zaJ5G<ydxIN)J`9xww}L7+>c
z4bHA%{TVaIeWXnvQyY2?410^7uU#i4%afh?Xc_>~jC&84jzRW<fC%TYl=51zn}b%*
z%7okmVO*xG5p$0}{OBN6C8r3_w}WuH7It<dM%8XVT+)_YQ(F<1=>v`gNU<i$mEuuD
z7=a|C#cGh_|2l_;x?j9rt!KY#r4TfJxVNt+f%*hMU5`IH#6|8&8#00&%U-3jeoPF<
zKgotTe)Tu6W~1Hkqmg)rvJ7wrq(;=tz|&<hGSzlk`vWDpEitt5!u9h^wXg4}KR>+_
ztoO>QuH1^`J(~5ov=tVCp2NtavB)ll0x-|X=Du+>qnyKa>bH`y9)Rw2x!GP%#g_4{
z|5Kj{3Pc4ooWR3IiAV`!dSVd!5BY*MZ@ShWa~~7FJXT*7dylVOWbe4nOv<wLkL41j
zQrUKvqLx;smTN+}vMQxa>4f1KKlB$u${hI}-%W7}s3I^Ae8;paW{%Z^q`u=LdIXeW
zU4J8fTIL<SJvLBoq0wA-+IVkQd&KLUNwp&)<<kxVuwtpOPmWg(k$o>kiq<}|Nd5<I
zWhF^Nb{F{4$gUABsWwzad>-N&$&{;=`=P(fyc{ta2Ui8S=l$wROG9;;W4SQQBnrJK
z(`I4QZ|yO;D|5X0yy9aMXV3gD{JhokpYnrOmTI9w9h5NF$blJJf|W?;1gK{)g}uGB
zPOY90@uH5Nt9;n!{|d5dxrwEzhY~Y^%c^J?uk&eL<UuQl5{j`^bQ4L9O3lwf=TXRS
zTotJ}MP}4SW*qx8vj{ze6}w<tSl#<eZOp-r7}C(2gn}ct>jMqa$8+%9BJWuwLOMq&
zZ!qV1#_?dMzK02ubJBMi5l=H)Af`{ie-bUF>NfXu&GSHL8g=fdCi3i<cgMWF(^v^?
zecfl0b@(LdKm&=yEIjRqN|H)6D;FyFZ$x}D>gh3Vd}~TsRx$Jlc!R8&wb=7eB2h@K
z0cr>Ea~a|TFvva6|94`2*^k`n`AnLb($=O7&=y+D3sVDJULMnX!l)K$hT*gsnr+D%
zj8x#bH8pzQ@IP;5+8!D+jbL?Qn~nj2kSG{u(#fNzexDSD44TS)wBYnH8=LNhNWV8$
zPwDpmARic)&tzpY<FI4Fmxg21ftKpC)-f{md*7x={@q*CV5EEnvaq>t&uP3e9lI<_
zCQHCQSOy_$_x{WqAFcVc73YN%1CDA<%+2OazA#ZC5;sC^(eHhV7e{3`%1yqj%FLqc
z)Asr0rKC%R;j5+m=COwVo_v`igbVA}?HVs$6k%$srL#3!q#)NpUs16!;BnKZzWHP@
z<{{p2Z=yuny5G%5u_ho|lLiM`jh(Q!U5S|#mn=#fpfgU(?)cmd;`xwJhcHGnD~nPR
z!_%f22ei|7boYRY2=}K4&Fiyg&<K=OAvk@VcEIaFHtZGSA=*W@CH9aZmZgp9_A%Ks
zxV(9r=*DIF5Aw+|^%0hdK$o-7^l~s1-!6QS#>*1`K`X~M()pPHrW1}S&wJ9#6DQA)
zdYO8TdA*l<#M&A+Lb;!_m|nFaK+=+<S#iRPtoo6d%esF8A=&`4=3)?JjJ2ngq-`f{
z;i&@d4_4+0sWivi^RFr{PnQ{kPRYeASp4F_RVUsXD&=6AFhK?54aGEzlgfCU&6>AM
zdEx2o8L1^dq=85lCNk-zxeU$&ZRf;VxW#I>@239Fx%Jsb&kF4T8nLUE&TAw_Q<3tj
zOC5AIlZ2GY_oGHk`ncG!0pkJBEee6xa=yTAA;f;6ckRW6a>=n{Ai-!^%v#Lvt}r98
zivGKnVvEpcp~^k;2eJ1(^k!|lhRZc0x_}8<Iix)t@4%^iBs@A1zzgoXxEpHR5u|j%
zdr^sKly01u>?f9jU1XQTRd3TYsdMT#|3D#4h+Ux4lscjcfN3@`AL<O(RzDJB#g(8Q
zTgJVOM}m=328`H>Dm1@G^Bp6^<7`1D5;q~oqv0Ylm6T@ldt0)kO85~#8a4PF82yu}
zUm_EHAq!E9=YiAgepMd7zkt9Sba8iDXK!>3N4<c!JG3i-&b^Cbyl>p&GlD>M+|4A8
z%C+lU<q{cNHdP)}<#N*98J7(ZfabT#YDzo&o<QqHp?r?*{gMA%fH2wmWZNo(IPtlt
z6D;WWY|1To@#cdu^vj2Rvn~qI1eDIxMkNi~MkqG$<ocnVYXwtl<VVHjuKM8$^3rTo
z4#`Xn6+3+QLY+do*<C5Gq_jcP6~x^W(FGIzDDbR~G`gAKUl4YRQBDf$!#i{kodhfI
z&-8cilSG8fH3CA-x{@`Bf!bkQ@ACs+11=|BBcslJITJjVMLl<$Hxr%oC<QmmReBxJ
zJ|148FJZEOD*z6tKKVx89vpA@HnjtTGo-+Nj$nty+KPv``qF<Qf$e0<9$_6SBSkMp
z(o%H`iauo}<<mjbGXA?QpPzo=Q?KLW%aHWLlKft@0^BwRo6rjH`B)!o6*C!np8ljl
zokyiRj~FVo_D)UEPtd`B)dd2LrF%=+yXNcSjEF9urjgF=?&jc=w7nite6jw+*z}(U
z!ah%a8~QDR9l_f_-5=jLbRF8YD9X>#SZXn8IEo)iKb<U#NnNVRxxByXS*GwoOtt%E
z0F8f0R#eD!mkFx@pG9}GgQugOQ@%G{*0uF2nMzg1mejPFuBj?p0!EYHr1@ym5U*nv
zD(^+@BgNXHWPibluxoGi3u$s#?gPbBRbTxB>>j(J(q8=_Sk+sSrQ4*ad-^LY0Ny>k
z67^SB3Y62eJHSJy8O1F^5*98mn97ev9&)lPSiQMI%z8D8scg7-6*J;TD=ss>fDyTH
zvjsr-z>>pT#DPqu`PifvPXII48j=g&1H+zGa$(t<0{*QR7uOcfelCYcFF<mBx7bDu
z^qERF7mEjbD|qkD#Pk-N=&CNF6atlWfowQ0ic!(ggbY4XkMarVk6;X%WtsZTNj>3f
zG&&#3oiRjYdlyLVr*KoE1jKSJ(|)KvO91)nuxUGea8Kp(gOI+)<0j2ScmMBzGj||5
zqqxjJHnquL=OCZ3L4B?EEDgQRXehttc&|K7`#8i^c7;c>j`(HCb#fsOoTLb!4Qpm&
z!A}g<WtM(>NKA=drrkMud7(u(e80>@DRff)lcsd>kTX%3zXm0pB?(+tvJOuAU|JZj
z4>iuDYMk@I0AYs@Y=r(zXzc$p1e3`^(JuRKolJf{9fu6F1$Me_H+Xlv$1)e7QvUUA
z*4JZ<>scs%GtFR!uPssO7UNA_!{}^!36Q{1)vPG=_r-mA{D!{w6|DmkHBsT=J3Q^e
zevJkZ2xa$qn%ZG&(}j#ysRR<3h90GKtt4NakXioZOkiu}=od(SsSB<i3pWBy+s@iz
zKN=YkUHyI>Y1-}57ACSmREp`o?<#3EoGR0x6x)5z;v(+j^;wW8*67wBEU5w(TwS!h
ztx(Ip=g=BSdrFBKPSFA+i8}eTtS-Smc=pjSz^wf>=^Weu`|uYp+Dr#RA^b9QL(h_T
zbv?Y0EB2|f{V@5OJXZq#ED8pIH;3S%DmeQB#zx>YQ=xnMXwjxnwPG`Zcq*#qZmW6c
z0d0z_H1KYjBA?VN6HK@e>r6EZJ^+`#mY|PTireYOf3Mu7wV774YM9%HdxS`0bPL2(
z;Vc!bRG~_3mW~Y-T}n8bqu(_BPVqB|3D2k{W$Rw#U%;Wb6Z9HFqm~(-GG<v`6n>+J
z<i$l~WVchQ<GBup;tSYNImCogm{a#dWr%I*r1JQjFOog)1Kw<$YXn>c?%HB0ToM6+
zF;jl;5rh@Fh7r+{$Or5u0sR5ANtPI6Bz|lFD^>H7T08UVfG<}hQ~f43-s{1|OBH{i
zM~fpu)>mq1dsfezy;fG_qz$n=*z48CfS<NT&qlC`aO1YO8}`t({PR{mkzT46WGgR~
zQTnoGHy25XX9qW2hhyv7EW(R`Rtb0~OP4e?r((0M@I@kvE58y|8$kt9__9dI|CLHA
zxf_Byxn;Q2pIrPGQewc-2I&V{%HFVE`jBOBU34LqvltVR7nes|wDmq#WB!=F0HgbZ
z_@Xz$qgqzU7!oW5q$MZU`Y!vn4Svg?cgj2`ZtK^nER%|+6<y+0elTd+LnS&^IW~Bx
zws}5~P~aoEsMZwCoMk#8)BG_t74bgwuB7jn683c>^J2X|@esg0(a$GPA40vDd*9dT
zT9(MBHWhV7zLY#AN$XcC9N$2=`7vsHm~z{LiSh?V`j^>{-uKW2{});1xSHK>!iHe^
z#pv6ZL8dP!2HGQQy;Jfa_EEMu(Co$#Z(gNIEN5r5x@oVzZnEiQOKErIe@`N6r0yl^
z=RZFiR7<V4+qYbAWQY>?*|-HZqxfqg+kPK!M*OVK`Py7e<YBp6e>;;{@!gxExA}cA
zhxj8?C49~cS%-BAb{$^N&E^f*tBwp=DZ7pfWyAX_YamIT&ghF2MZWT~h{6PGVLm5|
zp^7N37NT<?6<N~8tqjmJP)1R@EDxQ+JrO*T!as}DeBmfV&Y4g4KS_(SqX>}smT#aY
zMPDLkswTm2-VUb|B^ftg#y1v%Y9E(XtS!HCm>AQj7C~u=B3|=Skp?=X5nK2fBByDE
z*xCdo<W`=qH*g=_^5t*Y>AUS1)?)ORlCNz-VllIZRGfIpV|PT1O~$N%S}3}u9*!S!
zR3`Y8vtDEgT>s|6kZaH4HDbi~i^(+q2R@Rk@-_f_M4DZ$_CGVByuf;-n31%m-)p>y
z7PXpnPBUkO7|}y$%flVLV&WHvjnN!S3K%{8=k%j`7+};Eg^%=cE9?$e&(ZzX6$<o{
z*1F%{MXhmI86oVti_I>M#q)?7cmVOoZr_^6i!i-tF;?fF{?D?7*<5Zia8?KiJxzQc
z909j@yD|-_Z1dko2>!QF8*__{ZTzc1QOLIDu}qMcvqP$%I+VgVbi6wZ?fQ0-v}HI(
z?|kXKU0r69B75My7*dvBcvKh{ptKMtp{0!z$k!VPJ{<Pg8#tcWjjRsos9S^#r6PfT
zzTvL&#szTL-dK;jxjiX{ueWF9r1^^F?j+Qd^S9iDlzieSv`fmUR`I~w_ToO(xHK!`
zFFol;b^W3@V_|RthJ3601gG_g6|!lWjiY_EmE?p>nE9ON%p%d67FW%#$#Ud=jR0c%
zSlymzr{gtO>|JZ{RDDT)lHZkR+hBZR-lk;w=&Sf^MAC?cMam1GQ#k7>kv5NiB;Ccr
z*wy~ZDQsB>kq=P$8Z}pwa}!VfNZD}sayaAhYT8Zv>CGw!?IYn%n{4P)+H{<Fv9`b&
z4L3n})9&Xc>qo~ySJ#tyW>S>A94>Ed&IVoGr{K3uWTX1r-NBcxU#H>lzi`mzK*~12
zDmj_P!{lDgNTUr}evU=#wYm0e`OaADl))~6L&$bNtj0EMWNb*EhfcNr#|49v#gy2W
zHQ*{ju&#u?Bp*)yjg9i{X?^SY?|1!8?y-CAw~`kUFd}gkqC&V!54mVd^}PTG|Kw$>
zLM4;Xc0q|RA-o}ZBihtrKODw4k+&#5%{X$qEFX8~(KNmjcbvc1#r!Dtv0P^=udN-^
zI;5ed9JjMCA)xw^Je<ms@4s3Dul14az&d@Ut|yPS5mgvWE0*?V|3X>6(YmI7;)fdR
z?>RvRwrgajPjY+<;U}^MXpb%UPPLTDd{3lCh8{PEpMUUX*-46GW6;OsyLgfU#Dz}w
z<U-1;{cT|LSycoapzj)^Xelq6*-|`0Hnun^Qbf-pjc5oM2r?>E5As`0R&30@P>O1A
zb)TxH()_1+x`jYa1YR1iSW05y`&88M%f|WeWYV0Fk&fV>b8%K5FR$J~KV@mnM`$$B
z(qgkp|C(l7a$&jJx?Dd0ZNAsuinvHOmX3rctU8;yB_(o{vVK&&6hT!@v0d_}l!}N$
z{a_g~Qr6pV>Jf?91fMjl0$+S@GnXVVFr6`gz?K#JtsGt~(7{yx-Z}H)DWa1h8vMcG
z1Yep*p_sIegJy30<3b%3|G@JM|6{y9T2bB!FF!9Ev}7SA{Pt;DO}@&#^Ym;AZP-$E
z#5?wbI@YBnzmGrZbEFns&tdR2O<ccWea<r~dQ$^PiDpBu-|`VUE2pRP;OJgxH*_4A
zp~FZ?-so+VwSZ4VeGojhgU(w+9!fezJ0>A@ya1poUy-RBpftP;g2-!dmtyfhh@Kc=
zD>0|InivU~tMBo@dw#Cw`yoPQiz`Q)w&cxQ5I@{NRj>j&$ilRdS*a;%b|Xwrkq;m$
zJ<uNF0IutS`#2naqz!cw+QuF26t!&_NHFIfWc9O_;6;|)`MrsA$Be29*6Y>uox2T}
zLJCaK_IZ6|k@O$oMLv|s#`SD%3Ai#6%PJPV;I$ru9+rDc%*m1Z86bhMZ6HNBoKc%)
zXE76@s@L0HRy6)@K_1$&UB38Qb^HrX!SU<V17OFhWY*g`dCIv@VUt<0wvB$~1c+sM
z%a6L^9sjf^DY*ttP*q}Dmmcw}>wr%F3O#2XkF5H;%{(pV?`CkvRA1$NQc%1&Wh<85
z=R=uH*i=gO3na|eG8V5W(=R&w@(2N2fl~`C;3;k!dOJ)%)dYFxfIPi08}93MqO9Q&
z&VMv~^f#LlLKg47*+r!B(JsPI3Ko{D4^y9KTIaU{=Z0Y2@>-ly)vr!EoGgZI7XD9r
zZxt0s7j=u`E+G)yEd<ww;32^sg1a>i2~Oh>EV#P{2`&wdTd?4+jRtpjyPbUh8F$==
z`*6<ldFig&WA7SOySi7cnsctX2Zr>%FW6Me!JBQx77dGDm9-@OS>fMygHF~wV|C5s
zo>gVOam3EW`LTDS?91deEz4x)!kLfhvHQzj`|^LVKEC@0S4Sm9$f(NwYn}ivtOxse
z`TJ4}sH4Yq+q-X60U~g6EGDeDnH-0a5jZVgs8trp2)4#n$Pt<2<}6nyvj>I;J)9pn
zQC-eitV2FR9FySe+tz!La@)Ts@Pd_8JUuuXX?Te7ubtfL|HXfYj@Fhk_3?VJm{zj)
z0cLf6djsUNV#bOd2CO0nxjb(Y15KgFMOB_m(bE?|TRc<D7iaghkq3tcz{BLYYj)on
zM!uraOFv!)CL32Z%2vM5RopVgS@ZL|u}PaZRRKQW;#PPN{bw#PBi~18JSDGRu2+3_
z!Q!htQ+ITcY(UQDDAwhn{WCLZi4M+@HW{eTZM_vX2YxZQ7L_fBf5Vc;fT|$1XmOW1
zr1WfvMyUL&ang8_=Fx8c*j=`?{0@pp&{OThaD@(%G8}?ig!`KtQ9mZ`WO{AwB0!M@
zef#|YwOszpnj3vY=S}rby^}GVqlu};Kxzt{*802!O3S0)8lnhxnuJzhqh6#4>)**W
z7NPJBQW<?T>|BR@q55m;7F(|o2jC=YlT<-ipkZy6GxB-0>YOmnOW`mnymH%H;axiu
z0Y0<ohmx(Ou3i{A;KhJv5CS;jrZ;}yU3y+`viw~e0M0=eVhKmFz$YA4x^UYZCQ*h0
zJRvD|lCh~3^Ld&q+C#-kt)`(3VwQad!Sqg|yHCW}^IhIQL&s)Eeo;xKkc`3Mrk$~M
zSMncHbyi}3yFuD(Bc5aY=Ew2MR|DMaW;Gcqi<ryS-Qv&En`>)6dp(iRk$#K5z0}qH
z_l%Ia>>|ed3=9ya(E!)!-Ny|FM&6DZu86Bp-x8p5ddKSx%g#OXbzJkQ4{uk+wi+#7
zM!d<;<mfSfArxeZ5BnAYiWKx)w_I&aR>GjqkJCo;-Q~i_w`f3V4ZlvZ%xG0~_oVK(
zae?_JA`)d?hH!e;rwD%})7tzmTT809i+YK!4!gj`Ipu`L<Tk>b0jwVY5Pi+)qm=Le
zInhv9d9t|Ra#o|uqCqQzS!^e*UuS-{p}584P;)itLT1QfkoRLTZRs&dYG0D=eE(7l
zd`97kl60#p_`yYinTbHJqn(#b%t+X5sYzM|0ST0o<KA=bUYe9WYFb>XYc2}DxeU0{
z0RSC;ta+~KR9j+eVnD4hGED2-zyYHf$=)7SjMu~2ZtfN3`Y9s_l!PMS_DgckzVb{|
zlt<S8Ql&xfuj>!;Fv6t<oe9InE%C)4W$jR0kV;#_K4x8Bo?UilSo@T<Z>rNIX*wOT
z1?dv_0ed`bLdN3Ah^OpQzf`jby%*F~jePuk_Z#&W59HaNdj*k~a`&{MKE-u6REF?e
zuv5_A%|fy=r`JPs9)o8V^WgQqYSXySR4wMi5reI0^kvQ5@9C?k^!zJ+{7FJ}J3jDN
zwLKZ*ka5}Wgs%{zB<iS|Wo^2Je}J&Nf2v(VEvG6tL7o7~63+2qb8q!!aEH=YuW6qW
zu^b<Q<T7gf3cfEO3TjfYOj=}v#rW+$J5S6z-y%;0T74>z_r7!eLFm!id=XzK)S<%|
zapAW2D$Esy#X=Z${m#GQ`#4D$Vw}oSq5^T_OSrT2w>FNG&F5#+B?MbXN;JefQhBF4
zebYihNyS)ei5kmi`ang1%eNl9U(xZr1GzZmj1MqAc8jwQ?d~D`KlVt@6km#H;kBWV
zf|d+E^4uj6KbzjQ7U;`<hpt8x^5-UC!q!7F^MO<OLQokDJ2n-x`>Fea7h$W?isixf
z>e_WcOyjwJ@Dgegz=2?0XnX}6$6W9b$;KCU-0cmkz@ASp&|eZ~pQ|d6;uropK}iwO
zq~7srH#;?v)JZm=u2O{mi-+8EYUZ8{_)naKIwxa>b#0G%5TgQ3*_pnl#QbUWDn(n&
zytYMf(|<B7je&VIm!o~Q<w$Cmkb;++5=YkCm&$C{lM@H++vie}HNUBIOgXKQ)SJ4r
zHZ(yu>6nj60Zu`P6fA|x2%PQhfRtixlf2JpvhBZl6ZWqayDaUO!KC^+-D)I0C_du+
zwS2_N@bW=5`n5CMP;z5T?fxiC4rrN%_VaC`SNy1UX;ePDWoe9Qeh+#b7I}{L**5|T
zYfeI|-f1k}0A`&r6fJm?>x<uG$U-oodf@-URNueUaOq+9xOu({@^@kHc~N-x{BVeW
z;=8uJMRfCdXant(^|W<O8w$AnFfc5Lml1q!9Eu&S$}^p;>&`QdtKy)H;)LzONn3s7
zxf|3qE5f3u>GuV{PYR#xci+B9)WTGkN4DC0;vh=m_IG}<r{}hGtBW%dqXAs)$~U}(
zcEXd5qvxG*zoPjB0N-K0I)S8-!c4aB_^>JNx?Gc_MS;)@O9C6Iysk0!{~p4LwG!jY
zEqB`vxpRTf$=)UQqvf7A%XO#U*=Ekf?Sd4@AfoJr#xa2+NUc`YhV#1PpT}SD$PM3-
zOtDIkE!+y)$i3FF?{Tu=9$-Iq{@H9;SAGCttwXkBJEaAdDU?e<NVP5InUn0Z7`6-x
zlS5PzuI=sQPPAWUQx5x{hPutkNO#_9R@1Y^y(%Y(2u-ubsD1Zjh4F=$6WN8n?^AAj
zZL}k;OsCz0xGi1u@zAq)TvB$4YM8yR+kd=Fp@agYN9CBD`A^M|yF1(vO&0L9*f_XM
zDc${sP1$hbYc5r2lDt8wEa3Ab6*0p&BB4_bwnYWMTCZBr=UwZ5mv>)0ww&0AD}ovl
z#v-^Oc)Zoiw)x^JNLL9_{D;U;IN=x%LJxSm`e8sIAgnsCsIY|RXjg8kPN#v@tJuww
z@i;X4&I#=*Si(?~#gt9<r%}$z``$G)NeT;eX@3~&^J>Gt#&{&@s6LNQgQUtGZkoF9
zX~DJ%{Bk=Yz4gBcHAY=sT0Pvx-wq17Kj|Z=xe)hauvq*&`Wkk8EY4bRtVZ}5?yRgZ
zU$G<*B3<^UUN6;ce$PuAg1xmLB+d)4Ce@`{Gr{#eqv(A6nvnYmLh694O*D>dhoTu+
zAr6|2+(9D45)3NmdDUF0=d`DE(_JBS_p$&h(%tz%IuxGHz-JQ4Hx(x#jo8oJ_k*hs
zK`{`hl3;Sz6;u;MS9fw#g_3YJ{++7(SUGlv`mpNT1|Ro?Av5gZ_YX_g!1&Yysup$g
zLaDHr`!t33VzeU2Z^(7gvYOSfOHqh;#W_gB9Crm@LedpWBTu~m{FGr|lK8pAMXMCq
zio$Ox0b<fCD+jfe?VHC)Yfq}DT?8lBAB&LnU{37Pe#^&;;1~7mH56Hi8NJn&e@FP%
zcn^t(QCzq@bC+pNK%pPF+@SgPhjBg}{lYM`>kRZ`p}aWHu{Q2S<1TjmT{-?Vx*a@(
z*iNcEp~kQl8>!7IoUPLObVl+&j$tyvum<bE|Ji`>W*&f`N$UG88ud9ZdZaB!_&7Po
z*P&MfwLr-@GL}ARkXdIOUj^ekT{HE^wCGf;r)p}bofNdqPc)d{mmr%NVk>1BlkjcT
zk?-LCvODGYY$MI<zKWGilt_a5x07OWUpH%!w4_U78ii`n4H^$h|DhrQZ<>ir5@`tS
zgXfoll`i|`gtcOOUmiQ{Ml#Lz_bog+1U-YV9gm0)SnxFg{`RqT1CeeYgvN35L}W4&
zj59Q>_M4vL+@-}b{$=e#GTN1fiTAS~4xvx}d3Mi13vl^HoD<{7K794c@Eqi^{c+fy
zwg~4zc!R_DkxOB4DOAL}Y%@FjvGu?=ofxmpPVTFyX5gMW)+C<(e7*lISJzs=qTI%k
z!%}>Ox1V5)t*Y?-X^g_YXM<h{Cn@^14SpSb#p@^)ba$+Xy!mRHRK5P=Bj3%b-#CkT
z+d_lHh>YKVq4?*2pi&d+0~uGC+hWhAEeL)RQeEag;F?x^zgNNrxbi-3a=)u}_=n+8
zFawl)8g#Q;y%2$jrNd71jJk#59M!!l#kCF{O4U)nXZ<zJ$^*@j82}VN`&DIZF5;fK
z^rjN>3mjPb@s{g)w~MT{0-_e|4Ei)^r{<qC_e&B37p&)nfW9;(G4>1PCiQ7tN?Ye9
zc+ryN8AXDxd2xQC?!R))eLU4i`N)!VsN<Dg=d)>HWo#&>9b<vH=o;jzMc8nF1fYh(
z>1yrm8{y7Io5n=aCX=bunZ_6NCPM~#m0un+r^hh<#ZsKFH!EpY*O7N*1ASjz;}y{s
zWB^ThKBt!9uEqzQv*C?hxq7wSU4wZA@jCLJ1LyaMfDN8f0r=D&mC@RGlA}U>-%oJm
zdnbAPDNyryVBXhczTPN?zI*qq)6VVgi;M&-jIFxNkFfH^a6(?$+VG^b25E6&gjCuE
zdKEKGEyvx6xOz>yP_biyE$pNyLMBa0xpMETgy0jLZb4*<PGC@Uo`X&eI!wcgO&IT%
z(5&98_F3j*q_u(u<7{C=rAJPv@IRwM!@a;g$uu7n)tsoSTE$d*jDTzJYPMc(29`9Z
z8dW3E{bMQXfk~fg;E#~23O9tT2G;g3J=XwsJ%zDhbpYu;@%3s#e(we&;cSfZI@8+k
z*d8<g@@WhOUNi4PGX4e9tvbHLh3NC8SuS_I9d)dy#P|wPHGoWT(g%iWoSwcx3^p<_
zH_u`$vXCHN7Na1St6UzP%2X1!%J)xOY|6@NWuD5d{qj`;|0dn7f`(9V2(iN5?>Vi3
zJQ4AZBkjCCbDIARm@YJlStn?|^~mBS$#05$1WNydPH&soUrTS?kB-5Zjd&ysNX-TA
zautF3g&i5^H65;+BoUx$&HS9kX9XS*K;d+rI@xQ(`EC7Of6e;OqJUk$yP$)ReWWJI
z3flEQw%^Eiv^5@%ef~FTOQIKP3`F4+k<UU_2zIt(c*HxJ-)YPiKCkUimqRDE=$fJe
zoV?t0;QHXqYJ)%2=JL0a%}FWD&EGp00t{j1G-eSb24QST{td|U#`B~kFo~Fic8a#X
zW#6`%97!-Q%H*qZ?+g4C&xZ1U3G?VtbWAq4m=^1SMrT_0entTbOJuc4KRo9Vm^D+Q
zGPmMjO~RrtSS^wXuWL}}=af{nU1Y%oj-AA8bPf6fqp@>j4FyKVkhO`kwS3L|v?kCe
z#YNk*RY?!pt-rFf4E~4pzrS%xd=m|r#0(hFBRi@i)E0Igx}`{V8epK2+{4VRhdE-<
z>de{U>h&*IAVQ6+Kg>%NzzPJ7nmJuR?gmJ>H`845N5?U`&V7A}F!N01G4`<C*A<ic
z@7N7v!`UEdHtTz(HS&*l<$9X_BEdBtt_RFDHV+5b;!SY42<Y1U^PmVwK(114JpWU2
zlIETk*Ph=DrhV@!mB+zozqG@t+4{fe=ltV1M)}sHmZ9NUArF^WfkYPp>3fLpN1d_f
z{z=@U>Xv%>C+ji>T!vcdk+6c`b4B@;lLXfPR0<A8P5aZ1slnX*VoqJ6WG|4y>OJ!=
zBXp&1G`+d;>Y3l|QClE4x8(;S-%sKt=UabA`49>0T-T?nWv~z_r+Q63uINu+80ot7
z8lqN}S*|RQ%B~W~zL0xFr7>`R9%Xp5RAv>^YP~0U#<+R~^YDa8+}({6@2Aw*b7mzk
zyW-E54s|YkY4&sMGcQ+;U!L;X%*)o^t_X--+RqB8WhBM7V>Y6>OdES{4RTvgRIjGR
z>pAL!tZ&%#w}n>T{}+4`J(b(4ge|v^<f&PaMZ;b4CLX?=q~FDiW$`)c(Ce^RN8qU=
zvZ6$&b~(vrFj39SRhEQlnF->%kgMP(&zzPKZ18AfU8k9nCO=TR|BuMz%SJw~+ejAR
z#a50O_a1S4!T5l_4Q@|rXtv%Sm0q2qNt=UrcAOx*yQ|NlN)<9<Lq>q@_4?5HSNm0x
z1rd>tqtL(qhW;DQSMnT=9iToD{vu6ge;_6W^5<kn9(Y8A;Y9N)X>PlR!Z~Jxb~uBO
zCx!VD8b+~3>j4IX_F|an-x%^L@@Uh9OZT4$zzk9R3t)p|<hJsIJLEE5p-7i!#&l2*
z2;JqMKb*sTTr|&ly&f~Uxb(yix`o+Twr#V)3JFwinxr}Qo_nUfPL7rD{oZaIq3p=I
zVQ?bT#c`VbTvrZ6b1}WnBvB?6E+_U~gjAZq=-oEx+_>w1lMvLklhP)Jz`~y8vWl%(
z;bWv#JphdSXoILLes>r@i#h9H^}|?Q=gyUbwb^PahRaBO?tkuxlzU}cZjY`ee3@^+
z;p~t0WM#}k{rE(z$kd_2JM0a)xhkPCy?dC2>)$dP6=wirb}`zmg5gdkG{zfN(D>Ku
z74}x~&*ByKgZaDhzvn8u#B`P<o(YwAR&Hl0{=7jy1LswzWM2&zP|LvF$4ijlP5hI>
z$msF#X61e9!Lsk;bQK>~;(fo|e6>9QQKyv6#2%V_Ahe3-?n<^Z3gRTU-~fd30)p(6
zsU94-yYDZoeJ@gy4Q}JhzUg~TQq%{SW5o$CHj4m>KZtBZ+pj#9JdFA2$PfAa>2ah-
zY+6YUBfYszfRJ=0GWX?6R3-wr$Awf8-B<jq$&YsSj!s(oeV^<~ir&@(BWhMy31Tsj
zoN@_N<C*fusW4>wZ<A|$Uhcsc*F()iFPHZ%2}FXEMYR)FMWnKxCZ0i=$oct&so!Zz
zmEVIfgCT5Wa}+x5URw%AM2mmXV+!K$TZCQGTt&y$Xz{Y~-xEpv>%=H>%e=NaBfyrU
zetqW3g}sgJ3)-uG>irMb3uMcDRn2WGtT*lIr=nP7H6n269SZ%9?O95Y!+p1C3hlU_
z^d%_g6J>1V<S}2kWJxavVwsz3S4<`-CRlCH^%VyyX{Z=?08O$TsQCe~)|3MN@D#j)
zlGsk<F+?e*C2mm?x!_;2<hVRLHD2KgwjA-&<+^cUub+|C-MyQfI$3@jaN6BJHWr6z
zFB>-HSJ&#1Pa>Ue{aOkfRu7z}bqXZlg}XGXd7Hek=$bLx&IzmZEdD5y+(UZ+Gr}1V
z)hjyr$a6fnrJxwzYxc(V7dUH+&O#8&>f>~P4VI0+Fm0e*E<A~Woc)ma>h_hQVun#F
z?4A{~Ze|Z60}O6@3N8Hj{u%K+C`VqUw!Jwl7a8&kiUJNUmJYv?;}Z0r;TFNic${av
zd2|^yxyTW{59R^AKn`wGD3GR}mpmi{Q9Zm0Cmb_IYB$O|>E@K$PWX*<-MW-cs`_N^
z-zdYgBM!wVAyTI<#IUW~a=d{%n4H&_I3c>{RA~t^c*;x^eejEJwB33sZ#A#dgHKAf
zMbCuSuOWi}vYYWLENmi5n}XFxbo*9uxNtRx=ctl_EE(Y9opo_zQ*7w<*w*y?*y(`@
zs*lUY1@V3PR_mWM6#ak$84hu>K|1)PTf%E#g}&MX6V~VacCvLa@E$Nkt+0)GR8q4!
zB;|-G4jN5x5$u~tWh9=K<7Up0OYC*~7GQmwm#9KwhJPBEDEc{i=`qyt>E0~ouFI=c
z@*hEQ#1FWv3_iL`0^<yJ=?%hfyxRDt)OUzIq5aMJc`8)QiArAk#tCp4^wx(@nOp!_
zz)TvdzsHxwzHzvs0>au|weY1FwDP){HxlG5iHCDLGWB(^s1Vfc%fRCULVldOHb(Nh
zG$eM6$<3Zb{Af>!nGY$PM6&9#`k+80I8^P<GSR@8UAG0yUb>haWKnKp(_+2M$?nwP
zUO{jxFu6<dz;OWkB*U4}N8t@4jA94mWE2|R?oJpzUAc}KH1<~nP~|)y>2vJ~yxB7g
z8HE>8_&HnqDb&PH!~uG&NA^6z9Ag>ZLqG1Z_EvVTum7>|cazB9s{FE2nY2P?o<0P}
zPmgA=Ctrjtw;%D1Mn|?IeApz2ewRm4d@GQH%7e!<oO?Fhd4>;N&!Ji)e0OwOp~M?4
zvC#{j?CLm|D3V#48mP_Z4EQLDxl`hPuYWw-9!K6;dYdi=E<gK=TFxF_<uU5}pXG3G
zrydG?ygRySb~|87B}sifPOzqiC#LSO+u<@Q>kc^{_|a<9*II*Ua$!O1i|6v@OQu4-
z&RQ9g7<2-H?U3(wJ4^z?kVh}Mf%{v1)Wn>7*v2Z!e7q~ako)jDR*u@b*zp&T6t!7u
z_hfrG)x8JCVNod<y^<JJLK&675W=-U`7={=yBIUp2(AVXNh%my7{jSbEY74rmib?6
zL;T!8B16~8vMn<2{CKZ-G*|OwiBC{$+|#AObzdUaMqt8|)z*`khF3j<kdu24I%}qb
zpS5!plgV~)R}w^MPtp?tcZ*!B9s^v-&Pt6us&$yJUa&GPv>nVbt7`2NX;|No#!iQp
z6HSEj#$v0I-|5$xz$^awyhc@U`8{Sh)u=%)?{X8lf@VG`&@aK`YVB=?y)y_RhGk+A
zZYNB8iO`mM1&^5IL(Ip!BSQzSb9RXFgfR#Y030|?gm${axU5zY5|s!m1VscPDN|a`
z(a6VZA`P5F5pGR=k&DWgCX)$<Bd=P?Eqd$3KN$f(>@(K5FEt&5y0@lLS}i*N!~|Hn
zxp;|?rVvh_@r1pLjU@ib$E^&3KOqtv4hhCHhC6KeO}X3~s?zPjh;6blN~z^FwW$Ak
z!?UxFf{%3J6i2HZWuwA+8ewhS$9p`YlkOKKoE((gS%Xp@J0C9<P7ouNbpeUT29<-m
zZUiB1R4%-O>aE+$5NiMG!{yP?N=G-f_g%p%BeSq;NLLP8bOQw-R|mrmPLrB&z+-S2
zKzD$MLCn?}*E!J1n-?zbYp%lcEm~@LgoLFk+xNlUQTx1Wa@6hj43DF2n02elsp=Xh
zw5zya5OJT|E@@3CsMud4kc|<<nUFhMz-UbSh4@RLc|rhJZkWXn!hf)(P$!m(YAtHM
zQ-l4BmI;?`PvF*M=7$PNvd}#O6367rzQd?d2nsO|Jzt;?Cxg#!`qc6L%2kZM+Tt?X
z{#z0FOHi)EsT9jeM+Wv_AbK}@Hku*bUgG<=t5-{k%KEj-(1UL)$#umg2VJj00#e{C
z-sX3*0<^EkN?X6ipeY1WF=4wO(!#}mQyF|S*e0|*b@jEJ#1UQ>oSc_;I)}U;D_}bB
z%R9~mY`rsk%kFU)F~lK%v;KjDh7y1M;=4&pO#=EmIkXTbwVwg=fm|bz7vheZvj$`#
zx<|ELU*9X7B-T_%-!kM{O1tYr&tW=*N))9<w?JuzHAnb|rZgYKy$|70NFcQ>x!`AS
zQd_j-%N>Fed>g*QdYKdCQ9M#Hfj;p0GW`gF_ork%Il_5}b2r}DV%6PD**73!KQ>iL
zNnWz@y?-zY?yDm&ZcP+$CG|b3soG%Df!a`t>LS&kHhV59LVcH2(pTY0+hIG2&P92|
z`sdn>znM~L5lQvVEJKRj%E*wI`^d1T@~4f5KzDg>1YX5jPHrS8f+F{nh%IUVnHJ}6
zXQj5{5VbRfF4>QcJqs;g_)ja!O{A%QVm()x<1(h5MXQWH5%11l174hG(oCx(9Ed8j
zAi3`t6B}@UqfPq^)eZ?>yP^8sh8?hK-)X$ITntO2Vh(yD^+38s%ypRGutD(BO5cTa
z)Ztb<@h?M^Lug4ERXTK<|CXR!os7Cxrg`4P>#qd_Xy2osqIcQW%Cgc)ynG63ke2Vj
z6>d7Rf1HV=Vmz?x)_Lv7!^&^UA>(h4;=nIoNw@etgV3n9%>Docp#R=^#TorNwmQaN
zG|*)paPbMFlkCSH_S|smPTB#HL7hRYZ}pZIHVP57QdbD-V_x!TR42dUI-$Bp;z+3L
zYVGx5iXh=~(g}y}X^&S5p;IAM@T>HM>HBwcw=B_rX#*hvgO1(rwAD_>=nc!CX!Zym
z@ppIQ)E&?z96te`%!)^iZhHIK{RT8yljK9qKtbVL-K!x1u2N1!KW7yjR~8j*{r5ju
zN|L3bJc6p!PubrNEe?@3B~{tl#0ij1adjh5>KM;7&_C3tt>9z|J}OxZ%SC~;6g*$r
zT0NkmtQ4kRyzjQ}C5ci3u5F7}zdDlIC9ItU%;9!K4Vnqgux4L9y!EvocwzTj4dZMB
zEC{R=@LST@Lp1>`H0@GYFK$zl@>(ayZp+QP-(Och?vq?}5JzQJ$Iy{Bps2@A15dMw
zhl~&*WYNa8d%5n!v(-n%w7Fyp-fMo2y54u{Uf^O!qvvOW-BRx}RD8e9A8=VvOi}cn
zCT6D%bt+P{&0|CdB*2OB>rIo~IiiT^j%oW5(XGm!aqBQ1KH4v7oxX3(W26~6;%||=
z2~leDWbx0I=Z0^cW_)BGf2g-xZrYcUIM$SVkDlJ1)dg=_S`Se29ZH)UEOkf4?TL)A
z^>S(8J*If)W^Tsk5PNs6kzmKjEpU;%-HTi)@OwNxXuF#207QHh{If(c>`Qox27K|k
zm>kkP8FAY>KkJR1AL$&kZ}AjJ#s`VeWZi|A7_D@9g{>fy^LRr-UoqnEX2kojyUhr<
zRb8)J&SmZJ{Zn2qb)AGrrd-N-QyjY6iD5!01<)g~Ib4M_SsP?QM4To+Z-_X5bXT9u
zX)iZdw$IiZufi`jG!=u^MoPAn%OOu-1~-7QpdP2s7h3r(FRIgz5!Z)W+0Ct1r>^79
zL5e|1Ix`^xWGo%mL9PuNl0y?$F-E$h{g1d^zQBds-|Ny7oJx^26kO5gyUJrs#N@b#
z#=u5FEFZ--h=*gtUPO$aIM;}n84&Z^Ra7cP=21P+#4-_-E;X#y?$kGT5HKY!l>V-j
z`PH$E5X?PhHtFPM{2x%g1aTS^vpuC=z?xbQpoYic$HX^0mr~%ttD@7>ji|tjl~x2D
z!jwzucJ@BZA-OrJ+O>+OaCRuCeGzXtWJ)bMGG#+$x9GmDVDVz%yp(zS!7l`DwL{nz
zx^ZKIM0lO3NOQZ9cL<^f>otZpm|6Ca9?(sva4=TMt>+8y@Hliatx7F;dK#Yn2&WC}
zUipGnGTl4M?`rtjntu}X{0K=tUK>Ns)7iRj$78Fexu!%U#gl7DZd*?&Iz82zZJpgY
z`X}*jpqt}p`iZJ71u*v2tBA*Wm@nHtszW|V*SWy>wG#fdCv3$?mEf^XjByNM>Gt>A
zMCM9_cEn3-j}tl@gtoR44RzR+DCArP!L{Oiojjf6kj1;@!-#P3*|x{aC2-R7cSJbg
zqt(#C5s3L#qDAqtT<U@C-Dqoy_si5`OP82rzT%DDxkm9)!_^t-NFe^aFS+|dbuqoy
z`bWR1({sHA({)J`sG|75HZ%0AtmxdiWlB|p3Hqr(Z9E&486l?5KC^vv%;7)eE}t9h
z5|xMNn3}|@MK`_FKT4)XEBJaN${1UZND;0(TY7!ReK$Ujq_u(d(b%S|9kZM~U7}N^
zY%z^yJuxQx`l^%|Gu7O7v7Lmt!&ME;S-!m1B~uGLt!-d$Wg@-StL}fb_uJ<uzM>${
zKt2EG?hVjoFBrLI**M9*VKy7#CqrBa1!5V$-ML$W_3(G@Y~Wi9vukZd=eOoVlh{--
zrFMhz@;0P5zj#<6D?Zg+;v3gA$%@9xPY?GAZC7JW%<EwPtTIDAW+NywY$2Zs$mkOB
zwsP2P*W-ma4<s#>dCi2!OM3{ywnCmvb`~1=gLuMX*hBV20Ttxi?N*`wVD$OUWxTAW
z%YIlc0*w$Q*l4+VGav5?k3lbZ&&#+fg_wtxozE21B5})J-AIZ935O@iAT}fc9iU!G
znZEB=afh4-eUJ|-3%6sFc&62UX`e45qx7;BgsJK4xmh%`-SzNuRXIy)wu7JZc1#Kl
zEk>@D<qssv7qT8&L49peTv(Y=0@sD3F~;k2p>uCEAD10DF1dn7k7!^rk4jx-B?cek
zr|uRGmH}$VT9m11EnBGYQ>AQR13a5v)mwiY{8jI%U-He6Hl^j%`{P0LSPeQLCE15B
zhM#yug<(Zif`l-slMpOlK_M40{Zlr9`^Xyy(QDe=?oGKu1sjoqOg)&mI8EM~fVyYE
zgqvmS9{qN`(GG5$M}*T%N%;QQespwt+?6P1^X7Vp?`WIVj#`kN-b`ceuewFVVtnG8
zrqB9VqEg$3_u=b7#iiN)V#NdaPk5ML(%!}4ANgIXAA`lNi&B0y|GZK{abzL(>f_^b
zKSNCC+b1>#yHDg<cd-((40J%Qr=SUAS&8GfREu4c_Ad`p?()gP2-o29)bp@T&w?%?
zXQ?UpWvb*+xX}^aRFoS$YEATq(PaH)9KE*4C8sV#EDrrFzHFsfkGJ&rWgB7r6+%7_
zs2o#avrC{L3SmTru=c(>2P(IE820%!11_q%72yhz%?EJ!f{7JHNt{QHh{!>5uH00n
z*~9a#$gx7c3Cz(tGxiU-2!!YS2{(NR(-<wN_(dNbh==wC00_qLNyS)YXmvPFpS(SH
zmK*UCAoZQf0rejb_JAX*SDgzXSzmImd;5!5bNB@?d3{<s5j$!>4{eLxwNnuGwVSN>
ziZDHr+YN*{F~%AT$}(r^yy56mMOQPh__iEYj8N2WwE$1`qRcq_U1uKJu0kJsz#8JC
z`A3TO1GHVs>Qtvh&B#}~a@HT)2nZ_6lx*9Kr${c^2#!L#gJE^+u!8ooMI@`MRGS9B
zpoaKJLml!nz(d@RZ3aS@z~RvmMx4mTm5bWVw<1|>g)t+ynoLt%hV+Wnd`xysu4v$>
z+@#!Y=|rLrQ?GWs|7f_2J>N-lupgfosOG!puNjE5(P`HGAiD*oNX)&NI)*IjH}Nsm
z(a1|34+>CUvYwXa;kE;*afn*)N_E_yPm^IQ4T$FU3u#Nt4mJn<8l6pGS`dswc0~u-
zOE;KK&&*sibhhiBG!m8rAF{3GJ{?4D*6~G-o!~Loz#J*x%mXcA?R_>ktF}!AIk8>(
z#+!ft!v)*^q%Qi~+su%N<4vf873pMu4)Ggr@gwx(w0qi|Mv+qTPEBUa8I-)oKe5lV
zqw1fWGydB1&`jbP{zH1n3_`y^yoUW$IQi76{@~27@A~V}n^y^Ol$ZpQzB)MM>H63r
zrO?@O#gOU^dvw7G_YXI7IC|>n1qY%Zfz{x^$prr#it*N*3O!*}sC16ic!dhjG#`~J
z6_7$m+nxihUTUfuosQ$e8cmV&JV?fZECP5*5E2E595=Hvet?q+O6xm5_m%6_O)JWX
zqt)ZF+q6aC&CsdnVESvf+w^g9H5(MsrafsM6ix~l?x;PMq?#=HI@B<W(r*51j+MT*
z``e0|C)na3>;u#oD2CgV8kMXZQEb|3#5pw%u<mvf@Go;RE0(`rEGt@V`;*K-bkgBO
zNmKWG@`9@#=Qv#MU=?oHi?B`*lWM@^6$ji(jXgVzvbB3?!1s3+6}*D1>NgyIL`41L
z@$!6$JZt+k(D=_OTlr-nVO;&r{#(c89WJa_b*e!ZUmFI4C}>Ke*UczzCf`UeYx%bi
zh}SMZf6K2vUNKDxLEax7d~)D1{A>n!GOk(v%0zVkvE8F%oP(sP@eEJ<@#7k~pT&Ri
zp>Ad?oiMUQMzOow2;kEVBm4JHP)8|28(AK9=?8mqV~D^76-MIks2%v%2V|5_;-Ia8
z3){qf;kEfJq&A9<_Zc+Sw7Xe^lc>fwa-ddys%$su*Ak%DyNY3)-mW^;#YkdJX9FxK
zqF0rnb_*BEImnrHZ(}fd=xt)zS$E>^I;UWl>mI6NMyL$1TJD+-;X~5&+iR#IA*{2*
zoSe$#26kI=&G{Fh%BTA)a=5qH*go{r7MaqS0X5a*uZ>NfByE*9R`QlfvXdsSusBfc
zjcZY=(_q+d$9mt}m6mSO;09q=6B3X1b3XIyav}ZzfTIyb|NEO-QN0KDFO?YK8w#K-
z$V;r|Ct<QTQi;>M2h2*`5j&~YkH#CaTSv9sJwYbaF9>k&q!AsL26QW0&H_{|NzM{s
zQi~3wm2g|kCX~RHkgwmQ+V4^VR<(bN!Jl4Uj71`5AuFtu32_&hwVo{|5E9_@tOJP6
z{@8^>j^0{BF~hFtZX#S$YhHl^>>2fu)>6@U437a%CR6-u^y^lU>h7HDfI)4|I<>j2
zVe)&k-`2rh`*AUBEX{ONwv%=4`-XOMe+K@Gl;<!*=Q5=)mBu}hS_!ZA`?M3m^Zd6j
zb4pG8GN6hcYpj#fp_Swmnun%}QKD5+N{D|^Kr}oQVR3*+kKR?Dlq)=UvKF9o>X5j~
zu~B0hsm?GO{}UWn;m1Mxc)2@6<O28U4|?b4TrbNQd8>#5c3Flf@|&;^*na|fvGGL(
zx62;*?7W>8ecr3zT?J0f1MQ8fxvOcJBGLSF{0=8g3C5@mi6_?5p15Uu+zmtyD*IBv
z1|p%Z{DXg*I@g7M8zFXXU~e_!U7=W3_*7$R|GKu-4&ogeZ<-?)>zEjG+QAWq$+9a4
ziMO=S4X|cypak%|+V-{5?>0kiRthV&2*(#X*Y)}_F{<;|X!5rBuUe&vAWJzsI>nJt
zuy9jZrWtl;)9bhPQ$KX!vdBJ*ZXbcSIqp_|4M|rJLZzcy8l*@)vP^h*r<ob)BRXFX
z`(F6*V~>hsi@fT<eN<bt@M%O6z!#9G#bSG#a<roUX0U4F+d02gaa~8WyWgarom=aB
z(5D;MWHo-7Yi;&-u6FK{XRVR_u|O=ts>w4WTo9Nk=js04LC4M7KDw88WM$U<mi{{;
zW%w0KXshyPLd>OahZ7;~U+F~YAJ-ROW-?)-HgcEnf_HQ+8x@o;s!dWcqgie@V=&_D
z>%f@X-xTuZyBlD&W+VDtsJ_pb3*HN{z}Ac7$x2Xp`H4`pUwg++Cb&G}lS`A;^q*bY
zT?9a%TCqAo>!++#?XCxg4_1?abfuTwY1#f21;MmB$dpwit(Ag`c!E__0aLzdlK-ou
zi0WqA?}1L?L#~cH!X1t;9;1e%$IJ~g;nz>;3pydo&e=P6B9=Ga*Y0=E(`J9#3YZoy
z;z=#rxeJ{?h}XH)S>kV&mH$!H)nm}JVm?ePfY%V3+Pz~aI?X>R3vrMs9g#mtd(7El
zb{QGqmnEg?jZl(H;LqMIPAc+YtS|j@cAQQ<6(>UIcNlx{J_X<^6~Ws82>Bzb^){6L
zw^=5uXVzx75C~FhB8Tn1_juitJ+^qfvElR(v54{}wB1SH<%K0Ji^bLcXp##$VqT_r
z{U=7t2wGMkubZ%%S(ivS;F|3<xAI%(NQ!Fjm_M4d)ymo6d~GqOV=DQ0X8H*uFop6)
zg=tIfhrzwGtXYND?M41lcFN^_M~6t>oHHYrjvB_EBLmf+O78~K&GT}su?AB$o0ghm
zzvSf!j}N7y7zv2fa0oe?sFy{6{mmj$WxLkR)#OKn0gPE&G9w@3$$b5v`%4ctx=)#{
zEiWYn&Bv=oD-0eME_$A;ivyCfpf3kkl@&H~&%uyw>*7&Qx$`muQnTgkx#cnoH$z*c
zj@Y{O4Xf5FY(IJ#whi-V>HXb@LtUh5u2eR5`cU`lz)`zYwZ!YHIfun}xa@9IPB4v{
z4X5n!dwwAv88ZjU^ZBXO)$1h0#?a&Vqnp89vIDAt;#twfd_QR6yTQTC%UgD+{Uqk`
zohbD7nW#9iUGzB5Qi8ROelA5Y)EZob!JcKgUAVM<9b~Ii;c4Ahwz%`$x%&94RllT7
zVAjCT-ocOYf-?BGL45k~VU2!GHFzvpX7Z@(N;~mc`<!NZY{*>WxLM{ry)C#1+My|S
zvhXlxbt0<7^4&lT4uSgg-79lZ51XgMA6|*OOgFhm!N%{KFc`6&Znzu-i(4nJw81Md
z`$yGDM4B0}dqqnfldK=VYpuN=WvF~QYl*IiTtVtVoc<*nsu!NDs7_L{cl(mG#^t6t
z{6@s_*FSt$&a0j<X<Fg4L+7LFQ9T#Sqj0Nj>a2NDe1T*<-%8W|5UR%aBI71^s0L^W
z+#J}=LYu-XckT?qwl&PNUu^P|T@5lvWDo<wPI315M)-h}2RZmtQELls;$e@bV;0k%
zzAPOBg9tdMsbwfT9@<K*d(XaOx;@ZnWS3aXV?a9P^Sl1BiF(hp!9$Mu<*JU$d3$IB
zC1x>6yK-c!Bu!$!hFU8MS*UqF*;i+hAzmuovpuHC59V|^$jaLOa5dlqc@eR#WLKmT
zu?z^zVc!E;2o1QrY?%AE<T%#WypI`M{<5FhM5Aq{uz}MsZu+6HhhS>jsi<yIMC8LH
zHOYba-${O`kei&@b}PC!Y_p<l)1Aa)$4+=;YQo{s8LjY|>8M0ocaT>AIL|ltjZRWR
zQw%AD>#~!A2{DsmYP1w}-ITjEPz~lB`dCsLl*@8>jLxfodes_Oh*a<H-rxN!rnf&l
zL5cicoVdeB*c@WbAgi|tDoSKTZL{r<F=`W5s_;9tcJAOM6J$FfjZbQRmwLz=amFPp
zkGtlZ!#4LSp-=?D0^$z#_FqDkC1oy|EevpaQy%kRz+Q8n-n2ZUiRC!z(0csQs{>e(
zndxh`nQ2svOts^%>U7g30(u=(Aeiz}m~TY0$CkDdCYPbVV&i~3fI8rk*%A|9uDLKu
zCGxRyQ;xx_%YtWhIriZ<x46kuwc_4ywi~3F9QPZ{^4iZ)!^2BYyG|!NcU<?tv<{Q+
zdT;85MnKNC`hvRW3G`V#62|^;Xph<8Pm|X1eb@JLT0alwCq&JAQ-s=AHlpTc822m^
zo-eAme>Lk};j+sTMSJwHzQX}DL@-cUB_+J%D~$Eflb@g3l@Mq#hCk00;0?N!sx+y~
zZwvCo#bkhfjR960)|-6~kTV5Ah98&>nD$Z)CF4eXH;U@TtQ!x<qb)&NwZTp=AKkGl
zQvd9VE*A0>$7F+Twwue+Z!Uaum=?m?ac3q>DQ;T4(l!Qt6bCt?rBa@xNv|@_)L)8t
zAA}yz_N$CwFqrSC4|`P-ujKjS@mL131V^K7V?iEcmf^~$`L^@c_{x)$E{*s){o0oA
z*6UlyUT^(2V-F**WqlKTo0@g(#khN0TXNVI{0FTu7-hp-P~?5T6(lg+*lnbZo;|BL
z7)?@*k1-)4Ub;W>3@CIdG_{MdS^~q<ka-UHYPX!4T9$lL3W9;x%k%}94y&9^VE4T`
z6*RRo7oL}VP)t-Ci}hJsS>=+~&$-DZL64Xw5q4{J$-Trwg_tC7Q?(?aQX{DH9-1sZ
z!;FhTx_qrje0G%VlFQG2kDc>8dqxqug~id`!f0dy(oNt_)80H{<LdSLjUJU}2RLv!
zev>U;Gxb-8?!eCS>X(`0jCQbR$9DUcZ;F?E_7afr*sbA82c4?QYUXgB&^(zR9GIq&
zB(^iOQ#>g7o1*F?pZxVV^HuC)<za`^R-?AThN`AIZHt)*#q}Ke43?6T$HaYK2c3ts
zBtQ1^QsIV7pWSWV<!4fOaX2JV`=u7Y#!Z_9yTajP0NC#x!LMH=Wh@O$wu46%UmIkd
zD70!m25rVyxPZ5wzB=G`xW)|~H5}_QdwRt*-Fl{F6&kx5iQVP-dJ<*3Y&!$P8RbgL
z4xf15m034#J}1plo)2ymJG`G8IaHLzoUt9^m8Yj-xj5sG@)Y)7x`8gMrOaj=eo$|e
zB&5u~TI6fN*?fNjI;r(IS!7v>{YAm}(4Lm<Hg_T@p}zw!c0VkLU^GYQGI<bxoB$P*
z8~BxGTeadtz2mx*F0ad1ZR7c6*=Eu{K3Q(%^BHLM<cea2EwP#2Hm<y|G_8Sa&6pDj
zi+nJ{b7y=yfJ5ytE>%|nGEsOHZ-_y-O;qY!cA_s@IA+*-?G5^##yF6~F*d$m-My)t
za$d;x<QgRjt_Zhox42k}J}Mtb!A6{G>b*$##3AofxsA2eaGd=zXK5^KvQxB<;hUEh
zy?gj>P-^&xIfu3`7}8P6QX=Qpx9t)a$SD6D9L+NfX`A|z;#Xhaw)It3F{xn$_p5BF
z?0J4I1rI4hMck!%b@lW$WxZKn$OX>*o@GVTpC__DvCD+&x$MM81g*#Tf{E-m&Z4ak
zwljjyI`7G|n4#N_Z3CK(ysOS%7kpT%M0S0GE!kUm<v`A$rzNW2Q%21P`e}+P5V&m*
zcM6kP9n+#eNfS@xL(X(Jjh7Jfs)0J~A`L__nOjMuS8xH!oR2#5CP1<H!-=5p(Q2ya
zZanoYHYw&O^cZBN4M(0Yq0piS>C2#b{=+<qiAP^qTrK){7e)GL%A=sYbYGT?v#ENm
z<f15_HJunYFoEdmhKFj=)lAd)gGX1o`FbcLg*sl&5|sV~9LmjHL;c_=E8jde<Xkec
zl$|wmngn@nu58iLKc0)ayITw_Zjdotb^eI$zN#tL+(%S)SS=!H>(zr>dKuNCUy?>-
zFw1UT)zV=}N&3-l!F$Y>{}M_*=#^q%s0QJ*?5nH66)xq?+?}l9#^x!JTG(G}m4DCE
zd+7Qz-@F2N9pJS@AKZ|*4OoGG(dnqa8RpX>WT+5&N!20^!M4dvkpE5>cpxwkom|ov
zh~YFOJDBn!6UJUh_&HH8qA6XmU2(E6{+tXl9l6lH)~lKr3XkUB`L-<YtY?`nv~EPm
z)Yt}9P+Fu^k<9k|Y$AoR5cMkP!PnJHO%E)(BWuAcY?u2oH)MI~rIQTSs+h0ctUK5t
z)OANb03`0pX^_Uw-J05iq+|NA7F-sNw-+s(&D2y$Bnt~Pj_%le3~zv9=TCUdBbDr$
z$Gj<WhTau`qmzUCh5dIej+Bedxvr;^!KGM``%c}N#pxx_VW9@2$@4Zb?RKk&%56)W
zTXE~pmG{3r7Iyl2&&K?mf@3Tznt^45k`>C+>=~RkA1VbE6Rw#t2`@g#PW;$1eaJQ~
ze}8y8xorDP|I|3PH!xD%1h8!>IEvSlndFy#VixOM?9PTSjIDduxHBxV-`is*NKIth
zK%Ta8eckiA!`1)U3>zEU0r)~Far@BREC|_#hl9Jtvaz!p?kDcpIrbYm)PXIv{$h#d
z4+jT_Bo6y{1qbT_3skU8_*eh?jR@4RG5?MR%Y|WwhC_qx|DQDe-;;q&-2Xd={!iHc
zO3tt*Z=fr%O<($z%Z<00*^dbTtSP7Kawl88>4L)&=yZb~@8EhPbMdz)BeHvrF;E9L
zH!W-Z*m9ixZ&w50g7L`<JT{(JGRcW+Caj6UMN-?v+{DE~(9FpK_65hy$<52g$<4+s
z@R{qqAn$uYE?yQ+PC-u2qc^tW|DOT&j^@@t&;S1cnKDy$umNzgl1iT{C5!|AACglA
AR{#J2

literal 0
HcmV?d00001

diff --git a/docs_new/logos/dbt_labs.png b/docs_new/logos/dbt_labs.png
new file mode 100644
index 0000000000000000000000000000000000000000..3e52f4707dfdd972597f22af6360e5cba2d843d6
GIT binary patch
literal 13418
zcmX9_WmuHY*ItlDLXc39kPtySMM6SCK$@jHBp1b{OIjqPyQE?1j%5iYq-)8AAJQ$c
zEY17ye?M^T#WQne&YU@O=EObUG(ISj5IrRVfj}h6O7dDD5H=D7!XhQW1HM!wX*>ab
z2wjy7+&~~^vHL%)cgg(bz(*`MEk#*S<p}*Y@DGlijG7DxR1-^lZGj5{3EnBo%jkGw
z?W0gGI^guXqs=+oUbT%XcQY9<>9K-B<m*(zw1mfY&n7BTxjN~iLh|GIz=3|awDvek
zet~YdV(Br0b~+z!^NPt_g&dCEJ4dMMjV?uL<@Prb$VvfP1afdm*Kz;mM#<bt4*RUK
zY}p@Kmt81;b#@VI!0-5tow3=h1yi^!C9&#X@H{<o-XmV$$TV|)?2Th7NZ9JTo7&ux
zOG{r*AG|Ljz*h(rtA?qGzUF^kyArTffhNHMfhdaw@OW5<nt$-6gyRTeM92vJCK`z!
znJI}(vs>r?@s3R>=aC;(5tYXbqM};+lz*CS1>BZ$gFxob-;*HE7!eTV<l&ZFS7kh+
z$#4ymq`fr(EIM6Furbsyhqnd8GqBaloY6-J0$G9ISC2e^_EolF47KpEML5mDq28L2
z%}0yh|JSVhv6@Fx12LkzVXa$~y*QP0Rs@RUkkmMQ*NF!*e?>%y^wRJ=^DWFR+>^No
zgu|MUU|fRP*#~G-PL+dCwXZ;++{Y~Vjz;IdAn)6`5(tQ@Sj<`U;e7)nPn@(ZpJ>&T
zp6FCzfjE|wy5^~+;48bX6v@Ig^p%h3$%FWt6iGTOxA&{SI3Unt+Y)nBmSr`3W9jCt
z>^rlU>?NYxt1g~jy=e^PKp>8(&aPIPvcti(l=`bUh%2jvxiR~aqGYr1YxJ4`T_6^S
z7WKAEyjBBFcQeD0<>%jpzr*&8MFa7(mRVbUgt8Loptjs4{#OIO5H>sYZeC8Kz=aXI
zah25L94$_|@)Q&pho*vYQIwhG-95#P_!;X>9^wcBDa6xb=s0Q0B<s8|5bv$zgDPS^
zre(z}^<U%reMyAV*#j^q#3z8kPK}rf{e|vOl5bw$t~^i8@*(4Anq>0e(=+f4u)K$B
z{pMNVFKx^}@#?wVj|KLb#gwXI=NUt$5)I8qgh8NSaUuPMB;u2UI*&q^EkrL*(r*62
za?$LR0Ruh=bb8onZZgoZ(^YU#?wSy3;<ul!2xo;B*5)-KM}fn&HQ1FwgT9EL64mih
z!@O&@Af{nh*sS~<YdCOKlH)902`xEOwRK)ecn2t--yFEcJ<k#e&y{-IF6*Pj{pG4;
z-<xDXDn?d?0BD|PgqbAi?U+<Z1yx35wJZ0RB%AIgFbaykm>osw?^IOr5Q9Ks6OLW2
zjdaxbJ)Rwrv(t+E3(q(uq^dy4nKN$!9PT?z?MfocQgHh0TbbOOW&$L+BI3Dy1zwCc
zGU^4eLuIDKR0%<WR_!o-%*N>w&BL(U-I6gyN$d>ogk5@2U=O@Z-^0n2ule9;EA};m
zq!Vu}^C*OYa9U2eaTBls^qf`toA4t4c3PFc`9JT(-=f9Bd|!#6&n_W=xamviU~~2Q
z^}<a~{?Rrk13L|(7jphm<d1{ZqJY-21)`tx6p!sI7Z{GWQND?Ey?+9RomaZyOW9bm
z(71R~7D6nLn0!^=ycFf*s)Yc*oJqPqmc#rr_(AJFnb>nsa#t!n4ye;&N-8IVYGI?`
z_wmK;?ssoTKRJzN(}i$|odeMyo4E|{O@uZe@G9kB*$p_W|81H{7f2T9qZ(t?4>9vn
z=|RV(wNC-n!7{E!7Mc6fj)EghoEc-E<cQjqomM;mxdw*N3=r}g>|`y8@zN?u{qBto
z--{XRX-;$f%K-x6i>;WldBtUBX047(TS+V*i3#)5ssZy|l@LSO2}1t2d7b_y7*PJE
zlm;(RP;S$VE-;<rbu(OtoPdV@$tWiMdg|A;j)p9tfdm<cHMbwg<Fn3w^g**3{w_!z
z+cUD?uF1zdCnIKI0R+-XIPi*!8Y(xnm$lT)IXAH0S|8r-bhTxzQzB<G$BPXDogGy$
z>Ey`j<xkStUq#5r!vGUBtJp@>y^kQYtY#L0);LT%USWvqKFdP*_E{)0+X+COUys>;
zw5KF#P?Al*uMY=4E6gJo%(7@eplbCAzXZRR5sJ9;a^B(X60ACk3j4%Z4zM|=;d4eU
ztHTbT%9kM0sS16M1sMk7NgB(<NVTlkJZ}_yBzw4R@Sx1KUnxop4!<n|39km9o}MCM
zY@Y|FHik~;M`Iaa@1xK8vRVQVW%WY^7nAw;5;XTh4VXx3Py^qzzhs}mt@&o_t>m9&
z(!#guvY8%qo~luy;JjvWd5X;Z=IeC%g-39oIvXZsRd1Z?(!zIB<KrQ<OC<;MDKO>)
zk$(BR?xU^ERq3$L^-ZV3YRZ|+ey;n4DEe%Mi-H4Q*I`YZ<I;-wM<S!Th<Rm1mD_C^
zW)DrW_Yzl0Eh4~vxYJ-q3~n&}>0m-wDil9>nG@iklG~c_cCEveg$BP4&@RWz70uUG
z;r1CxS$uvt;Ci@mq23Rah2bUf<jV@C&aMybXIy1S-SNZ^?$T>#<TtCAVI~(SgYjSC
zyG2aarQa#KDuY15#}=1Ubn2TZ!ErlVql%#bKFFsMuuW^ec(Fr{W*S~(-J!uR?n&NS
zRkN2F2W{;?n~*7NZ+VPG31_|v8k})UV9W2J`M9~C$iS#}Q!K;>c+lEu(%4jv|B>S_
zWiNE&N;4O{ETIVaa#Kq@TTVgS4IN9x3HnI&q{!p@OMM#JA`ce*2jOC${!`>`d6kZ$
z(NUg+KV57XmA(<HKO+x_%(=M={eEhU-;atSB!47LJ@QnfX>!9tc;2(w$|H3|)plx`
zXzJfCNVPjJR_KEjOf0EzY+lZ3h3w#WwVOIQQQ34vclMJ(?w1A;RJ-w<li_)mB=moO
z<>m03P_yZSlLz9tjZ?`=CFLP^GwE%4Z3?jgOxJ&?$V*Tb;=7^SOHu3dxdJ31oBfKp
z#HJnIVlPTsg=<P3^XkKAlZd$v{<L$Z(kJP9^t`K9&<LX?21hFE!aP0%(p%YF^69|7
z%4Lcj*Cm@!su{{IJ5CROa~W-MIEtJd$e<*fozzx2cDfR@5L{ieegdB+kVTQb^12a2
z9}xb~Df;Tlrsy{T;dDzX&j?Dden(S%xUfmHVH0#czIjk|(7slDW3hMKW4>s`m|ESm
z{HAF`wqbdyomCh<p6$K5ZFMKiXI7YF9QyQaPP|n>xU^7Mp!N&YHuL?}kknG7sw6u5
zQg`x5<@H0nyfJxyZm41^-p|iY)g$HMIGN%OtvhI|tLw?O1fp5*D=wOU_1v_Kh_oaJ
zz*5M)?y%&3gVhJF)p%ojLg`HAWif-Z8Xx8vNDj*NzP|nHv2^Epl!Y19-C0cR2!h6p
z8$r*?B-c)e>jV@o`_Lq$FP~6$jDP|u_DLtTi<@q3i$r_*ef%>oNs->0>dwjQj-&W{
z;Z_i`yitTN>d3ns8}#<;_0UCG_%`aYO5X>W=MatBXP&Wncq|oAG@1DQ$Qu+Gbk28k
zmgY9&mpVe>WsK&7F8~g~@{~0W{aQMFl8$kq`^7rrbNT63i4Mczuy>$Ng6knK#v&A%
z^z-AIt1&H;?o*dwU6^EmCq5Q_m5_-*8`#~4KzDbdK&KoF)br2cvH~MzxE;D}IGVZ4
z?^@0N-+ymWfUPWS4A*GjVOT|jKo2~h6fNKVgGi~C&+D!ftfk;p>AwMP1!Y=9c<^3*
zcp_hRY({H0zbP5Wl~sCRwUcjn<sH5_bI<WAMyPyhPv&vao9x3Ab?M~2g7S#ULaoPU
z4gQUi$!ZDHeNnEUPJby?b#=(F;qw&LoDmxy`Enl%4+}%r`aDh~!z#uVX0p%RILo;<
z(&$0)(TMLU!s6&`#t|D7c*;3xH!{3wb+qJ}j!|vnkW{zv=99zUr7!-Q$H=a7ym#93
zcV3QSUPj?JAea*P45`zRh2f$@UCx?JD#N@|m$)=l?2*sb=*V_7nPN7>=dy8X5q{Fq
z*=hn16uPw?9<nVPrB`zXEVFCKeL*gaadZatt7MrVEUxu4x(ozz*vt}STq^I`4o?vJ
zTz!CUuZ2Z*wTr_)nTEwZx#{XLZ)X5aC4U<sdN1t#dwzVu?Bb%k6cjvXPv#}ziuU%7
zG?x_X-NgcF{FDuXj>IlM2%Ds@mn@het!3WI`E&2o6ux|=FX&IkvS3eUaBf{wwVd<B
z4|#LEigu}RV*%9XTV(nF>y4(c2~m4oQ}`=i0gs_WHjcA<<pCvDj+tjcBDGd<>EC^9
z<=R;q-)A;3a+SEBl#lwi9-tO$LaP{!L>He3X=B@E{tb9_e}j9y&LNB!xZ^lXR=U!U
zd^ozIq~ZJ!vbi_e>cUj?F;uJ|P;b$BWW!$6vK0#p(af_^jJHoeZ;;%_jQd%m)L(Xq
z+v!wj#=H3RueM9F*6FzNiBLWO@#z)f?jD>=u8@-@_E{_IR1%$?tfKG5-Kn1ht~q<v
z#_q)PpqJYQkQ-<f>`;9Wz1;Q^)T{WU3Ln(TnrKvW*1C$(!r~)+Nk@$ZN{yNbhD!O5
z(q7<3g_}@<KzxtZ!LZ*=A?HL-S22NF7bM@C5Vk#hKwu(}Y@TIx&7i>}@(LTor)fou
zyf~Kxc-J7Z>Uj<VC5UuHHi$|#utqIEh)mweLbCy~rxZp4!n3vp$X?mF0i!U}LV#T~
z^0MKk0^yE{+#d~sp4njNNc<qS`4{9$4G!on6FUX$BsvNhM`5*Zx5nZTD9~KB$Z{Xb
z<?LjQ6P6$V9Fo9l)mQqPNg^Tw*gcx}TS1;qDWD8osesjt_o;dbx=gAC2u(2uZJX&n
z(q}znWLj9gAAfeu$1c}#(IlIo_bWkLu{N8yVzPB}>974R8HZJ$)}>t)MqVT3?WUZ1
z1C~y8`Q`Sf6WO&<IyGvP7w@J;8uICsC+2)lw|MEWRevCgii?|UD=g{`WN$jnMh_yh
zpfvxYuAu%VcKFvVO{1B+9jldgHLTEBEa@{*^i1=W8{Lc3cYn~)&(#_&xjWdb{idm<
zpt_EZgn{|@g0{!%+BGh$Tz?R5aQulu#Jc=K_9?G41fFA(yJY9lKN2uH;&VWmd|o~A
zqiOlhRa+3%FOP7$`*wCVsl6}cT&IXvI7c|~Z2GlF0wxVnbn@Tz@$&<dQhbvs{2CYX
zfc>Vnwm&p&l@R4Ul>ou*vPEaCFrg#!EJ1*UAYs>sBCnor%Jr3sW{<R4|4aMcU6(eX
zDr%mZ*d0oN$${86fj&cbPLtbDmRj6-BVaX)G0*?s-DyMeq)9+$8jNgS?xrtQc=)w7
zasO}pQpDFUqA6y#r#DTG*bkW^%=h~a=aPs;O@t51Kw=>%>mEx#xFJ!>uB{QbkAe4Y
zoh7*1L}smb!A@%473<EHB|(IlF8G*>$IC|C%X=Vv+a)=B2#T8g5)kSI{+F>#IF~*^
zqf<M$+A$Jq6ZAo8(^8XL#zF@M-!3g&Qs{kxRShm{LYO|9iXoe-l=am;-TqrX_ULH_
zbY<0Lqs;Fx^L5jXY(Eem_iZ=w4M3azPW<7T6dC99OO>#0qwBxOqU+CLl+>qsT=9}J
z(I&dPc5`h*|8PZApf$%}1CR9QXH5+*|L!iML%(_fbg#DAfWiD+E|oZ=tZ}FR9`+e3
zxJU>jrGYrPZMixSZ-MaWs0g)+{^)4EM5ei&VAC_Dn(qTKuAY1ns-@G#zvpiSlz~Qg
zg=^{{^vWAq9bV=xoctC}gGJ|avCA~_7OtKU7~O+I1#?R?r3Ag1t=4BbygmsBTa{}t
z76UE-;$$iOd(ixRi$Rox#vxk3Jm>WZZymN$VAnjv`U_+>lz0uG!GDt^G5DpB^Q(7k
zQR&h@tdJcj{QR)#s^f~P>CfUThR)a~-~OG4?f*ED*)|;x!%;d@&*-QfJHe$~`U7uN
z1y%N#2RG5xZq98U(m$Tg8e|<wt_95#3FvQ9i6>=XPd%Zd<{HLmA>1~4e5fk57`sQ=
zM^Z+sh&gvFFw(7>2*)=wrz`2ZZ5`b(eZU3o&`KfFP;}CV2_cLnpx(QA-$s=ExZ!>q
z`98OJavD!ibiHmK)35_cwDGbq72tyO9U6Q#2)wlDu?)Si`MY%NsqHYj@gdvaJ?Eb$
z2|R6g|Js!*0?~0pBIGxNrf~Uf=q@7QU=%2>X<*x9DGa~4ZBg-0YYrP+4cDqM=!Obs
zdRy&gPZi{NgN99)q1qb6GWffhP~YkF=%|wKI*^JgBF;X@8byL`msH5^`MTsZdN%M)
zl!UQhd=*0w8PdsGzGCcFN7<i6yCaS`{7(;>&PWtz{Afm0a=Oo<<41_A$%7Rz41S=X
zP=+iKbH;2m=gazDg`Yp^5KB%U5>#X6fgmt(%YkjITfrmeu4+a~4euV^n?JzzLyG{x
zq)L1lyX7xo+W_@3lSMed1Rl!UZM9T3iUBvfuA|P=>7B9xGqHrdN^^4q%cz&%-xhfl
zw!Wm*S#lc|HdPzPaRlZSYT(65T9Ff;lvc<S)Gp5Xq)k!nakHJqnS#^f@HrZ+EkWg2
zWz?Y1B7&Mg*Xn<kx?|pmO;N|Sv7}J?tuJ>?Y^%+Oogv6C;*HBmnMdPqErJ>GVXWP*
zv;3eW?>i0TZ=Bn{*RXMXZCsy{=Gl4g<1whVK)tDFbJbGbar7Kal+>6ja7@S)^^6OT
z>P|x#H<WLh&Lz}wHJp*59yf^`{CnOi82oyuV+@GymMP2Jyy!-z22Otdd9+~jU*hjS
zTh6UU6re5b<bIROigK0?b4kDM@HZm-)C1u=Z_Gv+(MAL_g>yhphb^@3q#&WW#;`U-
zqOhUp`A=e`S109_mPVA6ESj_%N;uW=NC%se0`zcDmF9_{0m%{I*G)E>e|ZqH^DJaH
z$R=7?=Ja@SD)k4*90%s9E0@^?_!xWwWz7F0w-8A)F(KhC_@oXp7g8$<`NjJO$W7U+
zmC(%-<ay(-v=aF4moepRst^+dVyFIriU{;_k3=@4jAw`^V&9K|j@YV-G3~I3^Xho}
zM9a)qBk633(Mwad;-}oh>lGy<oeCD}Y5HAB0=HmJKAppUyYIXsS#~36^4bFK2*czx
zziG8nY#N=ZFT?Pg5djrSm1`%|Y4vh8<J`gRKC;?i@Cy=U9>7Gu6=bp3diFHAfl=~O
zWx49mKcI(rvPe09T^7?f-Hf+iR&H2TV4p$w@lm9hzTJK*@qraswq!}BnPKAeZgqpt
z67GO*|22gBbTFRxsjZ{*^}o$Crhs~y)|=R~k-p>VETwYrZwQfVNINzvZoQ(y%=sbR
zb06l#yMBzR?&?!tg9cLyc||tMM%ON!J}!P8#2LaMGlSU}<E#y-x>F$MUN+h{ZC_pd
zeV8(?262&2<Eh}>Pz<-(agpE@<Ngi*XMB4x9<2K$7`R5u7k4uwE9{Fz2c?|H%z}|W
zl+<M#hWB3}yz935ircaoR02imD^<M>*k2}`X>xrDj%cC!r4m}!Ki>PeOp<dZIG0z_
zUGq%u5N9Xgy6n_u!}Oy=n`>Pnn|G|pBe7++v{$69;sUUg8FLJ~SS?ZGNTvi_cyq}h
z>FK$Js%Y6Nxd(6Pjz>^WG{4TFxtvuF?8HN+-ZB&CZwe|QCi`Z4Jh<@Fnad^Tn0jo7
zWCpl^vu)*JAHRp(W+i;CwtNczD9_o)VHalPa4u_UIn*1gx!5%?kX=9~_*DZQ?1~DP
z*OsY&ReOdf+;+BV#+b?LQ0hi2wRmpMi%L?M3T;dKth;j}9dNfV|H7nzjxIym@6Oy*
ztxB17aoF|yB<(7t&u?)mdTVE`dj-4{`=e%2aJ^CT(ObkPZ=>(<uKhXIzb>gNdI#$y
zxrbb6QSF2hEk>XLVR97vO<zx4yv!}AgikNuTkAF2x;6vkZ5rnINS2i|KjGsmLEA~n
z<Seq6Q=VSy|EkC~P;G-joGN~SS!<jF7S~qY(jt}wGi7sZq*32g7Q4i)ls`$gm2#~N
zb?jxwzZAvwDGcH$owcKAH7)f1;nlgi;R7!<2J2Pld4~r$H9L6YI5Ue2O2U8lE=Oq8
zR6cbK!N{h&0%3XH%;9WeXG^Cm|C&Z!54cjzY`9B7IE10hqZzSUCSU#8k}!kTO12s=
zP-@*Rt*43J@TyNUJ7=+#1NNUz*fIy5T*=Z3_=7B>MX=`eyv$#j>dJIhN9B*=&ys62
zNG{?AWK>S5zek*BJAzjpTAC3k<UcRZb4q+BpHF$op)@aJN#RlbL5$v(F`>MXb|lN$
z|7re~ZkMT=jLwqx?(#LUkv4X+`kRO!Rn-|BKK##%4KnGp{6bCYSho;wCV}aGw|d*I
z&Cy|4L0%>PjVNkBnoMas^{h^V{M=oKXy2V}3jBU~?w#Gy*cIXndE;pU5j7T?D;;w0
z7*cEQ`_NiIrcD-&9~6A8NB)H5B6L{YLO&_uBAGm3v6V?-+$_K}-k@JwHhtJ*R2aAG
zdpnhUMo@6tusWwPW!cpa8^j}dwtsRK9T6>t%~n1Rj-h2xUt?it1<{O|$0Vd#cURk?
zK3JaVm}Pi<Ca6b^>7J^9&@3k+#2@BLRTj#iyDMg2EWRu_{c3S;r>xxt=*!?}m9qD#
zm$s5{`hh*s&p5Ufz49B@Fiss}k2PIK`Fw|&<+_AXJvkF$enRi*hmzxXy9f1z-moiy
zFjut_hD?(S6K!%_63?M*+i7w57`WoK5IY01zax=o0IViJf(F9zZ0Pyl3Vnp9OoKN$
zTbQFP`bT^LTSQQ3n|N~X7g|*WMa}!rhMn5soLfu)vcTbkb=Pf1(d}qLGSkN+JV(-J
zPmBpy(lsQB4AU&fGPANCO){0syFBUOrb^`PWAWrPglp_Oi5}0?Gi?()69rL2s*we3
zw8mfWX`WAxsUHOk!iAshk!VcMzjYrYr3hDDqbtqLt&$eu!{c2tcUItdqK5@^Hi?+~
z@nPUs7wpl&pKc`bOLcyig6(A%)5mC7x2W`2(*P&)hh~Xg^HMn{ykBtQHKWJ@nc>b$
z4I58%+9V6Hy>*!QFzIe%C*_zwNgeX4VnXN#yc>QSwMW@3#)UkCvC*Q)@23&dGsR9R
z!S$-J)7p=8>G!K7HxpmK%pk~G!Rf93{RAf5!phMLy6=gd0BCu)ZJvv<geK5c>MTl|
zwx>`FCBjO0b@?Xii{~(b{+F)#${}GC*R#pTkyDDz>^dp73XQklc3*$AkXp{%(&yAo
zv1b$Q)0Jf@|580-5dS_fTP1iVD)2YB<^zyc|9!IJ4P?Qudu<Mj38vSz+IR@Gq#SrP
z)U(^;-lH{ZNOibf_J)F-Jd~rPc7J63m`xqS)5?B5zHb1AyR`A1ivDG6Pxt)~$U~X;
zJ!la^9`7@?YN4V()`(xkPO805Q}%SK`$I?L@zIjb_K%x?pu;DtVt@zL9w&7<unxex
zajFL-JxJg}FT)GtE+y^Xt;&bDP)vgwRSC)!dOD{XQDdl9j(Es>fpAN(%O#&)?H%FX
zE|!tHedo_bpl4jc^nM+i%I0&DTGUnwaYjIjcz-vH_9w$~3M{TBGUDqyO4z=%|MIcQ
z2DZ+mYL!Fu4_1rHAEkem3mCLyT(92;O}#yqAUs*Nhb@M`%8rD-fT)AYzLyDKu5ls}
z1c9}M*bg}(%~9oglXyX8`3@v|wZib<TD}q3s44sqrWk;LZfDDxNxZ|sud`n+H|AJQ
z9K`gz3*V9bYp$)We2yE-H<r>*&zV6X99uv+=s-tI;(@RgciZ2EoWNik!`NXf5P3#5
z7xL0qxkN8NMyPF2K>-B%OW{=AILbQ&gu22=YZeyL{KxblCZ!cUnA!oMh8_FNTt5Nm
ztwhj6WTl~>Zb?6=TR~yUz|B3`okLZ%IKlPC!n_N2w_Vn*n?meIUGjYqrjv*B*r@dD
zc7=VATHw%{sMdpMg<f>{*HCbcZO`-S@FjDWRcq_$sBJn*`QLo12fYU~K?5s+*}1q>
zu(@gi`iST#%6JL&?uSUd*uM{`dic_i1?Nc`Yd!yUm+9A8JQAaQA@CujJQeQ=Y?s2R
zHUVds&w#sw*C4lill}NRlYSSD;|G15m|x$bD8~qzs0v;}$CPbxu_n@N>+tsNzg9l{
zbnGx2M1ah!CO8eIeA2fMxR3=DB^yAPuk~pIJCjMNSp6>lB~`I<YDUO@NMVq9sOCNA
zV~iiHGedN+g`fiEzQUrKS8vG+HYZB1(RP+Hw{*d%tFMc{f}u^2M3V1sb7uR++@G#}
zPwf~h(lIXS!kd=95zNVRv8YUSa#n*71PNb6Hn=R)fNrc<Rz1`FUNCD_d=bW&EbmZ!
ztp0kwnfHGAR28Cze|cCpEaa?RLpm!TEKEoJHE*byZh|hTX6#oGj5JM2h4WRpOu5RG
z4yT`$T{It9Wh&7D`X4*>y&R*gW1^#0EtIS2Leq7ZIzMJ9sF2q>m<6OYvc5eCyF3ia
zRjahja=Pl+DFV@ch}Udj2wwA|a0=Z@G_;9r1INYU%DhR_@tkI@VXBg*awHBCi38NC
zTrC-=^WS~Bk5WV+`4wYJvBgOAu)K5v(VwgGxn|n-f}f(&2bYea^ULa7O6Tm_H3u;>
z5BLP&Bj@nrw2%wSB8R#piIV(3Nn=F(r5lmTFUHO9Omwpdv<e%4*5L0RguMfDF}P`Q
zc93dto>P&<2fRmFuhHejCtCbP-H!5)zpz&w=tV{1wo+Kr<$u7@Byr=p+b+;f!|pog
zLOSW=5ZQh|yTDBzRw;=7V1%vP;7-1bXQ}X8oH}hsxAezSw5&FE!Pm-6s=W93q|w6q
zZ8QOlWf|DH`bAFGug2Z2+PJ@drYg}ZEDNju`s-btUebJ=K#~`s4*PL*RDUgBaiq?6
zXI>5N?o!x05ksuPYCz+)<_I4pB)d2@ug|!azf0CBhIWVI>j~%o1mdJ7&ZGHkNIFR~
zNokz2kpj({f>YCb-L&ZH)pC6g(O0ztL>$q;d9d<=1v+RNEG|3WqKKSsy)Z*9UKYB7
zvhEE;x{lxq(|Fgyi;$m>(j0yOk@&CVNLrQZ0|vxU*t=Y7wK=KbG7aUYX%)(y#?#`L
z?|G+p^naT*kircsw3HeQDTcKOwV)#1F~yb@3mDwD7pj%+5Npl)itUPMjZFg|co|gX
z`MU;jcX#0p-Gt8a>7Qx7%(xoFbwIXEXpuiWm_t(Eqi^@fnlNf0sx8ij#ftD)YF8DJ
z7zm9Q6Ro*?s{qi7PinSsJ@USS2V@bq81BbUZIDSzd{lW<%h$K>RW?`1S^qfcDeUCW
zlT~!!r;Iu-1cZ<%<s<)Kjg4l_9FWlGYjs7HQ;Xngo=`e0c^&qY7|OPCGbxY7KB3~S
zCE-po-yQ;zN(wno1}pSmEZmY-1L;VWXu5d{CpG7a&_&X}Y1@{n+|2(=X7p52Tap)f
zuMVXJe6wXqXWNj2YGd0*!>W^NeIHo&b2}ega24%tG66}@B~Lh8mp|v=0hf*v;<0PB
z?==J66Jlz)HxVgp?!00C;5%%Gv`PZb^-!Uo&9_TZZU?Q(|NP+F>UUStzg;pLS(5@U
zc3*5H|2XDE0+j@@h`FMrTW`~XPntod$uf(l;;Wd*lvLRJe_z;D*8KZddhbxz#3i|r
z>3L=WQaWv-DiqB>i@vt^?!6&2I$5h~{ORYy!3Gri88>LFJaDGTjzyd99MvFIi0!2|
zb{wb2kgy}4--Du$aF<9YD4m)IM7dfK8DpIKHeIFxH6ae_f{1RDe*uCu|HUd`ZZ08N
z7h4ul3AXQnVt49$*T)Gnm*D_nfXi~u3qUA{CG18mbA8=b`xZ<rdiHN0BSmD}vVaM&
zBk|z~*%2q;F-D+o6{T`sDdS8PnEcA)^RZcoZj|qY^d6W<bAMC1&q;gx*&o~lE?pc7
zLC9)`bgN5E^WR=d{CT9$ypTN~S;DJtx5#x0H%)T+d9CofUCa6XcP+nj*Bj4?nH)0A
z!C`ft?KteO4QJuw1^?TNJ3PsRfY|Gey)41agI+DT-b)RS_9c-)pBs(W<xNO3lb8XB
zMr%G%26oh+_>=omj*P{U_uZl&Skpn-c!F}G<$P=0the>?p6E;W`cFN7ue_i2%k$h_
zZ*`SLeI*WNPqLs>E}|JM>1JRfL+@ZijYYp~tH(05zhwJ%dNCM|)D6)8@W>5*gEvrW
zCh6FMuU~G_s5J%kl8u}~RnI=AABC;^OP&5!JA_ruCoMO;gAUw8`a`1>1kD<F_LW;>
zo<1{hPfASw^&3L#yHzHA``f-{FE9b4RsIT2-oi>18BUvgS$NZD)pk3QnQ{0`V(Lqt
zq2*r&o=}Y5VRD3@{ZH<1QPLJjiZl;VmD1(e?GD7UkbrViW0B}YJ-mttp}jt$V@KJL
z7PAd2yGGum*uHB#9<|3v@4cuh&<gCXtT%1AtjU#5_AB+cZL8`~q^Zt`7dmL+V?aA8
zs;2&_HDWWMc+8G08zo)t*-pv$UfFS(t>oB`F4HN&{IoQAXyzw0$s2HayU^V`u&&!b
zX#{40X&ffAr)Bb<^sZJk-8$>3+mhk7EX}lW<8tum^BVul*W~;(a+T_|vyWbWc#Qv%
ztSdrfeJVi|TKG%_V)E}e+7q=+{psW;Ik|pVx8}opJF79hM`l#DVZ$v<CtvMv8z&^E
zaCU>xmD1CDbo*Pq7SnyfU%>CaAbB@meDl+Y%V=58VUX01k8g1tq>;-A`WFP{{Z}L7
zpdoh(pZ~iU<>bu!Sw|u)pUeBI+%c&8muZ944`T83?XXgpve_bHZbbWa;<=3uBaffw
z2zWeQOAu$5e=BTXXa7Gn_o9PVH<c@@_HSGbP2Qp&&fMRAu(dxAhVn}yL)uqZpP8*d
z$c`1i*bGY#yqZf|u9ygj?wb4Q=fCnjUu49R??r{%Fos(tsk!=ZkjkK9KfH8=)RT-O
z6AD?3af*(CL0TDYFnWlyJMV1R_CVCNUUgjGehZDgA!V3NPxxQ(>cv_uUp84VpUsHx
zSKoz@@}*vZnWt}?Cq7k#c>3>Uw~5*l6A5cokBGiXA0iGDwgUSv$3HsXX1>^t7U^g+
z{xxElZ|dQRIw>tyc{Ch~OU|?$!(@f5{%mVAGxw3ov3EDRa)DGpfTlX7HGI(3o8>Zh
z&cw*+laiv$M0B*;z+Nq=(>1wIa(Q7PecmviLj#r<48^2c=Oa9Hic%tp#CR9v07C|B
z>3DjrpLjZDzvaJ4%w63E!&qoE$pa0#ZLy+2`g7^CcUbvJ6-qr|*lYF{4#DF+@QD@^
z{Km|&ZC*|bi`4RSQW><Su{gVC6RYzr`#5&(!r2s(NO$+|Ln_$Gk#~F!eS~bH;9UL;
z!{_WXa{6x66uc|uANIvgtWR748_rv^sl9nt!?*4yqekg$dy|~~uJ>!Ba&CkTTLUcW
zj1VOW?OA;XZf`vZH7BpuZzWtTfy>Xt=0&0b*SDiw6oHJ0JLH|DR#!hBDKct?>ziSz
zD3`y>rz**iLWlOsHOH!$V}E6vrkj$I+i>M~rLHkNez&Jn@i-xUM*Mh0_au{^3b_|Y
zzV+e%K>?fDb2d=%4imA;l+AqMV5<2dFVC~1d*I<(USUU7>$25C-*4%q)B)+%F3x*e
zaD4Roe=*?j!Sf?uohoLa;3bpSdV}l@$zH($5BK;-$CSHHPzaRNd-%F@V*Rl@KFJi0
z?+6xNMfY}EBdMjHED+SYbQVeCv=sUITPSOA-2I(fILaZqTn#den3mW}376Yw!i<_M
zyeM$vpVHXB2nxJ=LllM#_-)wq8qe$1>1DHGF5HQX--57YgcboN>PDLjWemwYq5a_D
z|0V+(yDrz@ljR2#q`I2IFA(|Zrj!p0tISc)s$PU<l1EsnS>qU*!aM3ZQk%j{d|#HZ
z^d2UKxT2cYHHuhs1DueO3i5g0ic50p2(-p!xxSz&WQbi9o11L53Ay=^tduKY|Duui
zakQ%+Y+X4^Dt!FrbO!$<WcNX#vf?O^T=TfT-dcFqQ5^h6L?2T&t=8pf@9q&lc-$dr
zJ+mEMLis_Ld$`<nxc-ME!8Fy|=B3THAEKOR!H3)b#XLlMtw%bb3xrU$*OF<~OWV;9
zf$XXez#7@_o?lJoLq=<dE-CU_Wqrw4=cy5P>)-cZOjdC2(J&d4ZEQt0NFg|N%3t*3
zW*&T%phP=JFzz!qbmj3E4yG75c5DwaZmO14X1AIxTtIJbf-gAL8V7oWGjy&#e*_Am
zGjt1EqV(O?1JqK4maAKU<|<g|rni;St#0+T9_LMPQ_VoKeIW%5bX4VB2#Z>ni`(KP
z<-}{F>D~DN^)%|w?LAybRF6414t$q$rj8=Q)TiHIUarjLI-ha3pO*@v?mS&2EU!E7
zFGy3&sh7j}JLT|Y^-j~H^90$Z996{@<xk}PhDzMuu^x!=>3hi0hj-yv_e%1ki0ut~
z2G7W;XsW#eb*G}7a*7-z@oqKn5Wc3{rOcJ@_fz@<u6Kh=MBOa-F0z9T8Hd`cH3#&%
zuNB56T>@@z7k;<WJLoi7cIJ{+>bH@<AW+l(a^?Rqf9w=_iB~MDM;qQ7l~nb&bzRSl
zq)dl5KeFP2Q@DZ=jqb_EFNf3hl^HLTwq%4ZMOzv<*n;^3chC%1L4Yy`QCl~knPx93
z`6x}hejLJS--E~a68aUVV^;H}@%^F%d-LBsdU5QyP|@C;lj66~H+o<z<4}o_^s9^H
zqJY2y4sL;(p7&P3IcC`pV|=nNxXB=zfwD}vf^yw8T%8|CHz6#CJ>g@x^pze(;ns^?
zdPL#9s@;QZryOz-8}J5b#m34<e1JQgk!99PBD8wI%?adsSjJ>t>T(^-YKpt(FtmL)
zCmFzIDT;FZFXi^;I@safgFmJoP_<dQ^FGY+;nvr#h*1zV#;?!&)O3F+3$(<9Y%Sfx
z$#qZO9BUJfm%q;@W>_|5)mRmKGjsNV>10J$_HU&<;^n7J_BbN+Q7hKfLY=?(`&XiB
zf7Sa`#r;qm_8X9=dvyoAr!$tVhof1ZH|iP{F-nsr^%wwAX#-{>t%Zy?0cfD*2GmW4
zG&Q=6$?|qAwWVmuKn0g_+_CUqESd^|Ggpz^-zP9&89~NiaQv>v35aQL7pW7Olm6RU
zE;N_^BDY&+=F$XQyLWwdwiEH6R?~UpA>x19ft)brcG^!mq2-&j+5QClTODshzGHA>
zX&UQtJ9<fxYvZoHzXQ9t<2w45i6Vhb##bl#6zM%i2NWn>-cK~Yu$9r98^>?R@ZNPS
z3c=|GAcQ-{@w51a_>ZLLSj;vWc<GWG%B2h+Can7y6#JtKY$iS?dHNOe5Hz3~Zcu#8
z#KEuKHq2}V?T1=5F4Krh)G)?c{_I<l5SujrO-O1Z;Qd@Wuipnm7egkI9XBr!+uAMV
z?1Ki+F*lq)1KRfWgeFX8)EHpSpIj8OouwbozXfo5hgjn1FS_9F!;3!)(XCu8YBEm$
z^GmFnM|S?;$8B#{7=Vv5pel)s&3%TGtN{Vl13VoSJ=~wlW)*A)cu5*a1`5=+Itwn$
z&${bNzlqenvS)WdoXuRs1@!x7#JkuQ1*pfr()DnDF)`<PycV5kx;?OPksZliw~<|b
zt(JJ#+!p6>=C5+nfA<#PAs|+_b_#(OeV=TQu3yXl=LFeyx2K2yM*L(Fvs@``XJ=Q`
zSWM&B@4pErtGYeE$$@mZZR97}segbY-Ihs^sA@Sf8)Geuq>+2lj_x;dtT$RHM;B%C
zXJ*dI@gu6q!lcA=rFR3mZX$$qAihkOYSZ)6j>Ejizq*V=y)U=5_L6$qCFSG2ysz?W
zO98kn3xM@z9PazLeP$(tQ_s!2mwx*RuH_efYNpux|E~4_fTi+pLel=V{HZ6NYC?#c
zHE8UwuU@$LL6eVs<CMhR?h%>Fzr?v^ntd0-m-(o-l;F?d67cc0A|~TxU#9yhz%_cN
ztic8Q2k}=k-=XQ<KfqN0q-CR6w9NM<!4F{oX}jkZL^JT;d1I^%9Bz6+gx2)!bcbxD
zGEfv0fk1(N*Kr$aGPDj;XXCuf3=?k{P>g{I03!Kqu5Y!i>GrP+d&bX=F3ya>u2?=V
z4-wQ?vV8WLjQ~gm%QqIvnVL9-`}@0Y)|qB~``zW)%n9n&aeY_#W%3PxH8Icm1y}Na
zLEALk-mo&2P{+ejNr>0&*+Hm6;o<-Qi~}LKEAKwTuIpp>U=ZireJdfK%c={*nD0+U
zfoBEuow~5-$ZgMpi*t$Pv+*p?uR*7mRihIDlI+FVjrZ%yzkOlU{7)2PtDu3meD+C@
zEP%trU!qmypcP{nB;eGn8&vl{0Uqb5m7FG7*!eQ#c{{bz(Idpw@1V*+i4g>H%`L!$
zx&M%zKAHOUV{Zq+Zcg(ae2-36@>t4yRzO~Q70I((f!3AbYu#sVUE{&11Zj@pVS_p!
z#ityWE<S#^JAE!83w1Tk2pGoeTREb~0(Hij>cURSOpo;ep!wN&%TPhACtzBd<m=iF
zP(`_BrP3Gi{{u*GADGQ{7h}?j#+nX)#lM4}6gM)xcVF*r_rWoy!9F|E(L*^5*#KG+
z?hT{|xV$j9igRDgN^Rgb=*V0O<4)7hu_5K%48c84?qtG04M{+qn(-*amY&(RjO5hF
zJzIsYokRS`ulGU#X~*RINRRC{_;ctSy645V@}x4=s;9?t16b?6iD0RZ#5p{zIS_(t
z{<W0t{7Ocn^PWtlK54vkEj!MXv@Z`pZv#%N)n_+zq#oB9l)i{zR;)zQ(GR;^%EbU9
zoU*+>6*SoVxkiU}2(#+18LXtFPboBw4c95k;Q`Q9TNw=K0q=&8b14UtqWuTqrJ8rb
z+HXN1+8_oD-J;g{d&px0-Eqa}uR2oZMxV3$9sqE9z*q5cU!T;~EH{kkzbdi09Ehx}
z-<>%Qiqk#DQA`-@xOoRap*t^$k>A|}Yj3XoZA1hwrFOQ9Rd5_0=Lyi!6HWv8tb!`@
zeO(sgNyX^n(+{qvTyNCAk>vgFt-v!FJfcU~Z;JLi!Bv%?%9QXuggT#&L*FMZ)Nw=8
z>iKiR*^Fb^kzK|aQDBKtPXC$1S<da5Pc>OUs6bTn?vkUj{vziG_v|<T2=>@+%X}a0
z{k}>?!<P}dDex7Gnb0AaMKlXQrdbx4?}rM0j*9+9upIKBk$A%001^Oa2VP#d0xvHh
z-8TiLSKwE>+#+Vo3|2WzvPl4c6I2v1kEoWM#0RG24cAA{2P|8Bs6IwxgUpSINRV$@
z2T>i~Cht4hHS)FI>mSWMZs28ggpEJN1$CY$0+h1|v(zNY#;se?oEEc@{*MGxCG`^K
zcH%Ec&{TUx4~KM%*{-xcNSdtk1FC!uq>VggEgcmaKa93;8an>_p{D=#W){eh$eT7Z
zUe?gl*P7@WYGY>4dqa;%JbGGsS08`^2cD`_FO`D_Tw3mYv|n9x`u$=QS?xKqJ?sm+
zniaJyUer1><a`?qg{a<@`Y>s~SV-)hVvhyh6?C86VNqFMS6MDu0#JSsNcr6d`AS*y
Gp#K9FV3WH5

literal 0
HcmV?d00001

diff --git a/docs_new/logos/gladia.jpg b/docs_new/logos/gladia.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..4c5e505eb43da98db8c9042bd42e10c1c0efd795
GIT binary patch
literal 11198
zcmd^lXEdB|wC{)>JwilJ5D6I~Ix|G{P7qxp6H!L*j4lX?(TNeG8@;#DTa@TV@4bxP
zYq<Vv-LwAZoV)Ia^Yz~SuC>?mJn!1i-tUKZ@82%B6SoThN+mf(IRFk00DyD%0B)xN
z(g56h|Ij}i_m1!g@ctnp0s?#jVxs%^iHV4b?~_uH-X|d^AtolHA|t1uq@<#}Px^rR
z0VVa_n)07ZaQ>0R#Ur{Cq$D9GxjX$I(rq(<iU`nwlZK1)6mX9U2bT)xwjIE9XXHIx
zoI3*i2N4nz5Z%YeyN7%CTALE^kK#Mq35h812?+oBs(ZM2cPk>Q2TzEp`JPHWd@Z7B
z>RLg=Cj8F9#jTu{?lHg6TP;2RsJQsDO3Wk2n6zPOo%aEO=>nf1(HW)tceg(LPul*s
z)_+c^{<YyhPNx8*xOW<<aH#;2fJy<GH!?DK{|_<Eu?DD3j2*&FKX+2%MU2Vz*d!hI
z&BCy(rXai8&U#ljN5tNneZ~cKbq#g^G0xNU!4CKS9q<6uINVAZhxHMIEGO<JNawJs
zO{Vfn@r)|;*MOqq6l@gF(L?YhnX}8o=z{^7ArN7Q!mcPI)yxA*1*ra^?*nlBBAR2G
zvRg1|!41mfd^&aMJFzpR{w(`36Jr6^>;<$2W+ZCKdeFrV{Ao0}Yq!dgcMCA#ZL&)w
zl=pPY(eue)bS;!_w!H<o@LkP3OAjbtIBZ#$V>D1t;5Iltvt$%aJ=Lcl|3JYKWgF*X
z?mK0sRiqWW{8LWczNc(!lq0>*(tlK^a@6tLQou9#u2aM<p!l5vE?NqkK4cWrp=?RZ
zoq2j?Q?Buv$kYP@wW)G{jF)kM;Ieue6Y;0nC+O9V9c&6#_dR?mK~>;NA~mg4W_--p
zPWxe`rjTyRCxS+<M8-`8tQ$@}L3Ok|a7BoNijs;7=YPxyo)n0>^r#<42oG5AWBNXX
zFDGy}ymTSs2|d!nD00*94-;ED0az*B*K0I(huxAgr@LCsGW=s*ud})2U*7^k$G~Oi
z(iGFZA0jxKYP&_E%hejtW_Bm8tsVGg%*Oi+$a^LEQ6t-lxTRn?0fm+RYDV0TkC)27
z2#<@b?Y*{MV;HOElNLP4Y{G5<7D@e?g;~BbQR0JvyG{xmMNogJ|C8o`w`4U3^b_G<
zre3$@T)Ies<q)3*$q(BMjI3SMK-aYLzU+HfoGCv`XQvp?`K_jr{Z*(uf!d&un$aN@
zd?0p+J*Z@6;}XqRbGGSz-WwBOu$P?G=Tfl43A@y4G$Jp5Q~75=O7Z7R;Xr1M;g`i&
z-3?FgTuXIOnP#W{_hkWi)dPA@ZfO`yv_~i^*KwMONvDmt{?CO&PtWu~(xY6|Pyc45
zgu8@)TZkVH(Bgwo_!a?4F};)d$Y1f=uxaoT<!Olhv+P)(3%owOW2H{Vap7ojM`Aqc
z{;IW`cLv2LOIN2Cu}P(c>nn9~Bhp$LQ%Tum-ms@lQ;#|AaTau%DNMx)S19a&>@Xf{
z>Eg%u8863~Wd4+1c8pH#i5kx>N^1<hK_IQVKQJ_dC`pVceEtra+yX*Yje!JR$E~%s
z*<4muP{wjqfdSIb);2P~`<)^_0`Q_?02)(w_+D$iW1{qNn;MioP}Kq3NvqUJX;-Pj
zk)FF;O<CO(+jewPhlc8hTKSw6>l$ZU-i%Znk4zWF_vn4mvB$oQZqyQp7K-w7qZAO$
zTDf*}-5x{M5Go2JfW}EL$FpngAn}{9I^`h-hAuc=XD34U^7S+FU58^9QC7;HDs<J4
zE$7QzDU_ZwFDeU9+r)P7TO)}voy17-@oMK%>92q@N>SVdlIdV$z=X+s?sKTE!1*pm
zcryf&`DD~htc>KBv^;@lq7m@|#m$)T<hK{L49@!*(}m#lMI&1@{!mBSe!Jduyu8CE
z&Hz+>PkVbC2UH~#b&%;>tpQ7zWzMrr5sTp)F)V!;y9)!-TDeilL;X9KWaeQJWZ0-V
zOcx^(_YFzcq{%we%^dZZ8KnyU4N1%Xlea@#`|9w*P@V?NKe3s-Vdj+O0cOSNvRH>_
zlqk#9Grbfx6dL+ch<+zQWT$V|+8Ek@3#jyyR;SRr1a+~qT)aO_efsp?wW@&jfrQ8E
zX9v0AvxZ%qLJP)d#bdzpCP#*FOkas*)|X5x8pwBg9}3&*xALvFI%b(=ME++;VyB2N
z6%TwMvf{|Oxtd+X0blQHVjK777zF6cH2-pHQ0gIubsZ5(co^+8kBn^D_N%>~_F;H~
z!FJnRJZoGg$R>2|VXD>VQG<l!tG7Jn#DjI_m@s{BkNKQP^Cqi$Yqp3JJ<B>i4c$l_
zxEiJ8OZ5irqfe@*ZG}B+c24u|Bc^iIsan*jGox++QH6MMXLI}8b@UMSt^;U8ZZtC;
zcIxp*^6Q`-#?;N_hrI`+yF)a$0HQyIkMYY)dtKsdC-PE#SG;V2l4A{Vy*xj$yV({U
z%CyE==t2xbZMzh|rfQoRE|TG;dAw5p`XtO~?j~&G6nBZmVwUU`##<g9)|cz7Q7b18
zc$wYsrsHt)BOWr;0EhQ`${Ujb^V|ZmaXmTjA~C@%osp^rl~ENfnu5Y_1FHB-d-&`z
zG8Bq1LpFcl3RTxmmwqkEjv=B=4(Y9%u;a*Dp(5;nz|Be6a4;e<u>Tc&q|w>P;9Yj1
zs-?-MH@qI-(V0L=PYfb&evqm2c8>ZM04$I`@}A#*oxk~C0yove>Bgx-P;r2F$7a1_
z$5LnM1(4OrhysuD^R|@!E>#YlP?kP{GAc1BiK8O%17TaCPU)!BS~Wb|r9%PB2$u()
zk^Jl}KiBcNUevto<gVEFykOzqlI^*Em73xvU8^cQ1y3Q}(qPZcMU+95ttvJBtZZo*
z9S!=^aSeucM0Y^~4%nr!I$Z(Wjz`KT<{B2S7;gcp`TP8ydGN4l$E$*#@E3{eS>ZVn
z+G<}#+XX}jD;|h_H|z<iTPm%zQJFD8e$syyXE^U(_i<Ek0MSE@n6%o=OQJAb!m^4U
zAYAtcek`0hAHIR++u2^>&tw;1z0=!VhI>&I<yvF%z_BBriUBDyc2axZjUxnxz1|i2
zxO)ZeX+YxVO!w^@(Ko;U#_}JvKWZaJgA&4Brh{^C0j_INqtyj!$XA92%X3O$tY8k#
z1Thx*LuC=SlD{Q?2Y2Fc0V1TnH#F*(idSjVz#DtITfj9}%5kLAIAek`%}}{Chfd?N
z-7nE1OJ^<207D%mOf&)-GAMO_*_9t{E$w5V<u?6gt)SvDxuSXI=Uo>qcj2&Slw+d;
z0VG`MP6)<RN_9gc@P~H(rX;mCVO&**QE%39BZHw_8@psJ)get9;8muVzGy_Y7@V-2
z_v7;9P1+6yoQ$<`jj;7qbMiEZ;NpUvfXN*2?6F)4Ud*XpoqO-<;ob|(Gme37m31xZ
ztYOPqF4MbNgW4JjXA<HPN-Pft$$i+k(xG1)Z^Sap8Oaq-G-Qwu7zQCd457<q0GMd?
zr4PkPq|WZs`(_vl{64oX?oUbDN4*fLs-xZ?>COpa^J?y-_e4S@eqEp<OemTh%NX5k
zZ)lQS>%Qoji2b>S*T0@u)p7VKw&-0WLe1!?0N4m>;DII5L8bN7@z-N&@Bh5Es1$0v
zgDZOd37z6MYoFfJj?tmFq1@~oyO%yEG$Jg*Jj-zPCz+n*`T952>ux~#k%9nN!|vM6
z2kxuabe&+z_LeD;DGD-+ZOIqxJGN~gkuvXY9gLM+PRoag&CCac?F*zjy2BK%TG67`
zAUKmS(^ciKy4>FL(<jx&6`w~3t>QNYt8pqOH#wj?HOVH*iZrsYsl5ojHcoB(01+x?
z#+$#!zy*n!Tw$jQgZp|G&xNV7jz>m@s_0BZ*wZ0w5P34F0Za28!e;U#;$ZynsI%mk
z<Mg*BS5K%8Vn~&ViHwF{>}dckIsI5f_uh#Fs`lie<duU^+Vx7s_|<^?O?osRV<+{%
z0am4&up5|g#o?Q{Zx~*c8}|@%3?6DqmwPbo+yJ*+I0drJ1Ww3k!q=Vh!uhAGGR>W}
z#OQk@`0WvJxy!1Y=CxMeCDCs$xZ5F4G=||bvh%w--AAV9&Sm<1=Yfr<r=>IM+JtHR
z3d7P3URCWBzduN{okHAZ=Q8?KopSQ3K2Ge(j9<kv5*wK?t-fnEB04fL@L;Fx2w3jk
zgjR++6ECJ^jnJEx_p-Jrb!{Eiu*4W~K;8Kza%VMSE?!Kl&0ps{Yj{3H_H1z5=Ot9B
zL|jc=u7jXqq=Z7kX?hA_?!;cS4Mx=<+tx^U(bK@{JmMYQ@7{vmG+Lxi8144SPjMlm
zMG${-7AX>}42|TH@#1)psfdf~;_-uy9znq}x$bDlNa0dk?B0{~qPW1bq7bTax>GUZ
zk-_Fp!uN3M#Lu2?36tdoa;qHp*fv8NZ!%>H_qFhrz7<%@asXx`Kx$i8<Iyxz3TK(_
zvIl-GNl?+2Dv-v8eVZaIotR$le2yaDXjqDcPor~(cJl1VwY8}n5hyVPRIA>Si~2;L
zZ~ax8T-^dyBL+}B2Ni!jGF`bmR$%V;{>zo`Wh+TW?{Q-Cp|Y#6nz-eFNx0*=eLXcx
zMwwn=+ER2tcrs-x3tOf=)O)63*KL(UJ0ZMtF+*J<zvmQ9DSh)_T)Q@KQSxxAT~c`V
zrj4sH7C+BmatT^V_MlcZVsy0<ZL^g4JMN)ggE%)R0k;L34VxrnJ?fDOY76Wl(FYj_
z=X!PK!^e-S{sQp^P}Y+{HZOz936usmatsb;(|VsA+Is+D3<LNJqeU-5S0=~g?b=^$
z(b%k=xct-_(im$0jPRg97qYxzPkDpG=|?Q(q}TLgZJbwE&Twy{(L17r-9Ml<dsZDp
zRe}^-q3quzWUd#C07oz)JFT(RI9Y6RyNDCWAImaQA&ebj9c+`<m-Fm-(kPp>2c809
zK2w|j_8YEWz(&feuHv%4qdDcDsxAXs4{UQAOGb{)>pR9}2X0U-N^l!dpL**Wtcjo1
z7jCCY!=r82vT^&0QTtEM#i85T#o3Y6jQ+)|SqC+hy<1E(W5+?sH`_b|O>up`r3^QD
zGoVL&<=AIb5mp2sbD;vnsG@t?(c}<k3-*b1=}C3W@NNQJo-+XLWc*@Be5EwQH{%+1
zU5oc1&%g4@i@|Q%MzI3R<HndUh$AGf?Mqs0!4~-$qOJmh6CD}#doVg?usqnAAKyt(
z%q{JuEoni1=`euWNGo8ghF?)g6x^fWWm1I@+F7p7Trtu=D=4g)v!&LZ;le`m^S?OX
zA!B&F{;I<Yx*wDv64T9z=D_dAxR2Vz8%DIOHw!+y|JI%6SwK*`+sR8Bqk~SuQqB!S
zm$a+zLwy~P;T2EWB+a?BY@m(Vt2M4#VR2-X0L9|=dSDA%)+}d$J6A%*ax2PREHnbM
z64Aax*8n!etJwc6%n~@#@bw-^T6nRS`7I#+&A99Fv58JO^(;$pd^H3S8o%W79_u6;
zbf`5-Tv+|*seFo(*^l@`!ySTp!pfG2%~L6<LJ#x(yrYZOj1#!7<-*CKg!$O_gmY8v
zAkQNBx<@furjswu<z4?+-3TElk^(bpycam$b&+Dg^I<WFpw(8==@L@;%VaVE8&yZ*
zJ=L^pBU=6Le#pg~;mRq(XpT^sY4+_oZ&=Z2CADy8#Vvq2hDX>4v88C9cC@`1b^SE`
z%WoFL`nKUY6kMIqFMXaty_uHeq}e+x?L9fBQTtN(@=a$`??<U&#EsJc{>um2BfVW|
zp+ScVX9ShuxY3GWKis*dbW{pW!}f>Q1;fi#s)7xwia(YQiC%|=S&}e06|*o$qERt;
zpkOo#o-2*UlN*+{uQdM+wU{z?MD$2S86SC0x&2(F`^~_`3JK}c<X#SwI__8x`fe)m
znnAm}D9X^0`@K!rYu)2(p+%c8&oB4fSLX@h6$vrzW?FK|n^kW4qQ=Z*+2@r$Qr^X!
zUFqo&xvLp<$R&3u9iihH<sewW(46-x^TBc%Lq1S1w(_k+sg2o56EWw3*moJwEdc9q
zoLb|pFe8*;i3}UFC0O-(<i!<Dum9USA-zPMNmj=Bpq3|rpxz_88cAU~(APRvJR3B5
zc8D%MW%C`?vDncaBO@MG)t;Qony`%(@OYdWS5>>Occt2TA8QpO@bQ$d=OGUvDsp)h
z#EgUW({PcN)+cKXLpJ`5AyZL7yp|`-YNqPC4)S$bS@LZ$6kfaZ9j)~2`WIh`ru{8Q
zioUT!#Qnt--vSgRq0d&DlGd_(owsfQ!^XV4@mnrcdOLA$`J=6ffcZN|Z<Uv4X6+zF
zO>cCUk+2G;-c#iCyGL#XpR8sbG}>vzY5*J+e{)~?p2rKKMk>iheprR?gUi3KGRg@H
zhY((DGqJNs57^S4pZlFB83}643K-g*y|A#ac2ad|>KtW~UGbE7?oI04If(3!bOpVN
z+@Nktk=_-Xtu1()Tw&{-^dK+dS-pyZ@?$dYsq>#cPV949BI`r_;5H@lz$hvlO_uLo
zj{KgF#wCOW6wN9{zrt}tTAEk9mLuhlX<%X5*U*Z0joha+H6+H=E9p6zBERS4geQab
z<}PxLb|sn0)aKL=Rhjn#DmGrtG;S&V*)rjFFUXEThURXVMZp*L-JJ}iKFD8z25tfS
z5>kH?CI51{PQou-9$8GX<!<U-ol-a_7T4)<#QITuw0iLiCi}|H?zcNY#1)zZLr4}(
zITX`@Y11rylvUj;-dBgviihTAEj*G2`O?2I>{>0lXkiQ>CWrb?uKC^q{0}L%yKe!T
zx_ygLwlxM-;Ovn{O7vl3IgQ&Mk=imFE9SzUt)fsrwn(Y#ufTK}{K6Zyr%q+b+7GW!
zyn9fbft}_1U1CkAhgDK?hQC_4T!d{eBm4(tE}(WIeXaL}1RG>WC6tqimIDV~3nG(m
zY&S$RjbH9qW9qDH<t&v|Bufm5B2XCAt`~^MGVT^|ZiX#v(iZX!mZJIWI8#cL5^iJ)
z4Rz*0xl!T{45)1|9l5c3+T&X37~1W%b2`+y&jd}YIM1z`26xKarRIjlxn>WrdX<DU
zCB)1)>SPugjL(y+HrYzP;}%rc$akFB6Rg<RQkZu6^ZX-ekpGXxqM%MBq0a`l^qGlU
z%q_tCQ_2M2Enp<2_NYGPYU|-mOT<l9oj%1^wZ4Pz_yZhuCj1*YHUdt%+Ff=UfUa~?
zSogD(nXxz}_5eiyxq9ju>xN<5k#YIo-#cRw?<%>{!=NK)R#@Ri_)jU81nOq<65_p<
z69cMchjR<eK;GfsBH1b=<vbhIZj1!AMPJF=vAjQoO3_)pGWz!97C?OGmB*f3H13@~
z);Z3j-A}o&`NWd6dkZj)Guc^sj0Fl-92Wr*29}Q>ldXr)Mp7`}Y>56Gn|)h2>ix0z
z<Y{BpQFqkb77wXYH<(XCm+q(NF%+_4!Mrd?lY9=eFpEzh`BP2oni>ZE?cFkQlCDTc
zlAHSp!Xmy5a(ylFz9hwjb$JkJ1qv88SdbzNpg^1KeT`FPyy7X5Rbg1uGT|xJcGy~@
z4-s)XJ{C<mltM$5>vbz626Cs3YS*pPeyoUju)BHEN6t6W>w-&1-x{9>3w!+X5da?V
zR?A9@_vfOrrc8=^v`;JHZ9GE(dur7fZ~;p!x|wzD#2Hnw7u^wl6LGM2xbaBvdlnew
z6)#|S>>J^m@FpPfT8IC#e%>Qd2Yo^j{hc~9tyHt3E^@4jpzk^E?AhkpE#Ru8r~b*Y
zSYMv_x8nXXsFz^Xj(Bz1vN~xY))K996J}0}>`W_6US5OIdH<oBFVK0l$M9ZXn?k>9
z-G2Rm#6bTRP*G_+?w_SY_oZC(5ML`iz=R?{gN1^r=?VB2Fh>Mbm8%OMGV-<e<cqa@
zFRe(Lz9-As{4A2(k{dyiyKZ^&!rDLc06UkB>}Ty(2#iZik0$<r*ZVSBm=>JrW=R-u
zzuuRyIO<5E!l<e3ewMdvZf4;KBeE>;77&a)GJFZHxtd0K*j<0q)%-WIN_{<_@&S~5
zD8k*=xnJU%e25xq8YTeWK<<_A{okVBP`^#IRPbWLc4Ka88>C9dC=<`A$24@9<wDT$
zuUGv{nfPeilloDK?8Z$UppEtbs8&DI12*Fmy-xj?JM-^=w3cR7p*|{OeeuD<_?m2+
zXM}$0PhtPPy-u?}dXmz8ufZM@vg%wHc$t5S+sVB$2$Qtm7JOQjI}QG1&%(I?A${K5
zd7frhdsnT#^DTw(o7xpmkcQTK;j^4kYp=84cmD;O&in-xhDjMWzOH?2*6sXhwCGq{
zv(vjRKdpL9TM#*NI+but0E#dPoqcR4f&Bc-%S@F!mEz(3JSCfOtd`YW%X-!5Lsd7Q
zT(aR<Kf-<y&P&lV9^XM1j?h~GzV&?4<}Kj;=oT+0_09dsCSlA?VgA^^<r1lDE9_>@
zi!Vh5$&naL$EI1-dT?ntBf9;%`JN7~-FQ1M&&=<&I<<YJ#O+#B*?0O(;>6Zx)pC1Z
z(lc)XoL%IfB?8Nnd;B(Q{`NK;>@?|QhW{?IrSb;agD7S3W?NHd3wF4*t4))tAGhR$
zGo?rMxh%8hg?7Ma9~C=Fdc{VBQi5tO<10k_NcrWQwcfd0iDh8=fF0{~vw02G$5*e0
zTZzb#9u(z_HY<+HSOvf%N;0&Du7&Vd6aP*-v^Ht+yVpfr&rm4)u=JgG&2N^Gs$fQ$
z_z+4(j&?%Pc4FqD2wRg!;tqMW1{+VK8x`7ByoxN)D`&4TgX59`(m3pqgfc2v)(75s
z@?$#|{na?Ry5U+Q;(obx#+i8XbijakS~yz(ZL}u@0>c?#KU`7{Wq(EnJ)ocUsnpfJ
zpR82$`{girPA7vdYV172^oBdj^k|jQVIYw$q!{4T>c2K(EEBEUE#Vebt1FU#K2z-Z
zj^0)0Ia#$!304oe`&D&n-r3?B6yQ%7lLn~xjK7;>@XUck%x7AjWJ8iBn1{uyT?JdF
zsB>6_!QBRffJb4?fHUul4fbT=Ioge5$8}ZRP$-Wso;|G+aGZXV`+24*0B*U?6A>Z+
zU%+j_qngO)n|8evNb;%4>ao6xyz|c1A#51}z3|5*&u#meTC#XK2DM9DiXqCsJS>0?
zY*`=a+e)mwo41=3U3?X{ORL`%(vih1Y7f~D)Q$W|`{D7~lRg;dtlG1NJaC{CqqE=V
zQevzRdN4X&^d9SX3WhuLW7HYYcvNjaPiwsKC<9I1BFcbDQ)=30Myc@!fiqUdObJgz
z1Jg~U>;uv=&T`wNw!IUaqg&rRsfnLB)5=TI?w2&1a&h<^>yk`0yFF@D3tuHse^P9*
zI}TS@31946xvW4JRVGL5j^ocii=7eR{ZLr-9cd~r_XMr+UT%oyCOp+apF!OD&oGX=
zIeyV5OiXGK!GBj(A-68JK8X_xMF9)CoQJbNsrUyg<v-@EZ~=K)H1wsEtWUZ_zbm&c
zpWXtB=;D`-5x?`@_2Rimdex3zn6viDYchQ22v8Gu9It@;{tz<~r_7X1Ly|fw{6(42
zUob8*+t@>KK*13Kn#_c~qU?8g&v5Oo(k6O#Krs#Mw7L3PFE4HD<sNJd8ZYp)pKtg!
zx0{>me<t>r(Xg{vDztN%H|5iGU!x77rWJQy&X+n<54teYfN83GCUnNDBP|1VLK4+5
zcE8h_ibzkR>AD&C^zphgy`3(~Gw7{5`+yDl?a+RbnJ%7?Md%Gm;MB>3a7LEa76O?%
zsocuaPW0gekJEnkub2dvaVkr3tN%VC+Ecr^c2;8OCd{m{UwVYe^gPR$+WQpoWY}A(
z;<G@%opzK~A4W^fm)3%%y_9*md<y)w*G{;HN7M#!6gj(5==`vc7m-($7AZ|F-!R}o
zW^tY6h*8zik`?F5!Eyvx47`lyj7K*`!WX@T#aFS%YZ3?Ug+<PgV0uKz9|H-~5p^OU
zL;LskK+^7#Y2<v>Rix12bPpMieAGK}gn9sSAT!&S#rufRylWnk8edVDb3FPf(G^oM
z`*D`XkS0)_?;^=6_<hx;Z*F*BYPFoV^2#yne0L(LXZ8p{5-inNSP!W)9oNd5C^R-U
zu4SO9Dy<Y$BTx$c@JMAiQd~exZ^hnXd(NHt8&JHJVa3Mgvc>r?<bAgLOzE<|^X^;B
zutTVy8&fEa`igI_ufen=kzG_}^%g$nBwR1#d(}ialPRXxN$2=<NN{Y`|9+Lx#)}MI
zB-DA%@JE5g?h$qRN`>2I)b0%z8TEc{VMhAg+(RxbBxLbng5J|DHzr4V;%^q0>V*D#
zx(n!!RYtq-4Z{WJ`s{O4i!<WNF*;gG8`;5fNP|;wrf3n~Jt9z$l@A3&@CfkD2$O7c
zCBKFvV_sF>ki2^;;=%=4tEaEdpPpFY-0yH#ihb}#^7nNP_m+#NmSK9*w}Fbayzr5I
zLpz-ZhAoyj$NVBaDhJ1OA6N!jR$%eibI@)ALkrJYU+uc?H(<qutP{`ARaq7PD6s{-
ziL5LhaNsKR$xB8*V+Mq%(-e94`A_Tjt7vXA<~&OqcJ+aSJb5ie83oh&!2pvd&`^LX
z!>_}BcPxdI;(7f$m&4z%4%-t~QuvS|6g{Y5nwL@NKRAFFI9d$xb5b9COP(F)=3c)W
zLf>8ycjee=TQa&T>J_Qz7EDaPY+Lo4&5YTOFuqZ@?A6%8Y`|N0Ca_K*bhc>YWjw*s
zX>yAdK1Z3zG^vJcq;>??D-qK0S#K2!rW#A!AnuK4ef(gZVKR-GwVdIZg$};$cSkN2
z>hI60V5_6C!=lk4)FbB*->OZ~#XYG04#g|L$jLAg)2K`@d@0%4)RSs7#uIy*e5g0n
zG;{;LF}`;{^uT=>ThHc7BbsmdYuXiNm5G*qZb^DBLsf<Ng4@X?-5tK-dm?pKAl_!@
zL~B#IOM8F2zeaOO-;ALk!8756Rx^6xoJ)auO<{1H#fjbHjN4;w4zpcXIdpM=EAUWM
zHf^`KWB1(=247d5JnSuwgi7J5aVLnfb6*)CK$b_ZnS<a<82gAw4%5}+)mB?PK6$yr
z<~=NT(DT{!4E(j?DvUfimunNcX<C$)8<vg87V0A%tOrN)wZ)^0n7*+F5N%5toHT|H
zLFzsaZq3f`{^oZg8D+1a>uc&|_NshG9`V(g{Kn|X93*)H_N4wqw0`J4AsXCl>78*Z
ziK^}UsRx8aPtkMN4`E7>9xuP=XJsBL0s%)YXZXA}k~Rhnjmc)g2MFW6ackRo&4*|;
zWtRek2Pw|1T)b!|$3Wt8<RI}TUd=wC{b-zr1jProHE#>FZ!nEgZJ<?J$vPScA14lj
z&$<_6qJGrWr9WST8Hj>nnsL#f{Jj6vgK71$kp$@Rj^mti1?h>aU1)90Z{jlF0{Hy(
z&%#j)B>0yL62CR_2`%TjyczzoLgv?Av2H&2`z5_=TbI#urN34@i+x0g?Al_*6cTbG
zQuL{k6Sr~GC-t*|OEgA;1S#s6Q#DfA1~;be@eJu72QX5k-Nn(OG#WeiRBMW-qkqP2
zAIZMlt6ACC5@HPNp<uS@AkuGlc_lo0R2ojocc?ry<_yo?-#-K~xk<xEOkI*b_1N*f
zGqH(_F)EE8QJ|Zu)sHYb9y*%n>Q-LeJaPKsfEz55FJe>zPT=O~9_pGxkObka&_-@#
zqY~lA5^%`W;xXmh*yOr~;l?h$aa#tBaDaI{+uz7h7iUlBlamvNPFvCCuM8*447D#!
zHq9$-l1Lk5jUBSlLR+!5$&0;Nhd&qb*}Aq(jGsjHo!BL?NYeT1C_@jzV^^K8WJi)m
zv3*s`*5^RrqwRO_{Gu1wd9XJ>dvO#B{jHO2s(BA?w0_WR>YBrfRsK&X(nO8(nx@>x
z4)xbix8UENQef)<ABj)u!UZk98`9mv(NM=YJ<q@quOKY_k<8hv$T#-TZpHx7<x)M8
zj<43Jw;TvAPZf27_RY(gi59Q&KTCj<Q&djeHMt^>)3;r~4zK_wh-r`<$^ZP`Shl5f
zJ8vGZHJeL$w(R}}<flM|!kEhVT?b9X1!sVqZ7vWAp4?3mn+?v<HWyWaXI2a)z9^t|
z75`aSVKx(d-5)n2(9N0u4Y%B2tm@PL!Gz1Slm1uzv9u}Z3xURNULBRpS~iayyDWB|
zla2NIXyHNfNa^GrTdowV>5rL-aRPPr!y8|I9Cov{?d&+I_Y;LgX4Wbx-8DjTaxBxc
zQT_oUX`%i<!u{;&$KcI4yF`-7dE+*vQybgyRz%NjPS4r0GP6)WUO#^lo>mYZ&V(4N
z006cDBtMy?Z&+65k=(u}CnIlH=QZv|2rh)z!{)~u2xS?>FP>YRjlg&(mJAe^JQk05
z)VQme()1>y_r*|uww?K+v13<Tso@#*8R)pe-Gd$f5ti{f*uPR*_sES)tmk=#ulq-n
zO?AVPB&1C9Y;>5kjZ+>LZc7oCBjkZjMnq1^2r0ZEUl{*vHg&xXr_1MkP3f5F+-cdk
z36sG!zCBC3pB>`JT8}qw>Y#(2SBHk(F^=GRn_Ivubhl@ZGrxMV8*#*Iof!uNS0d8K
z@q&#<B5<U*m}KxTEe2MQFe^Sx%1Mh>Cdth#C|f>`<!^i-{o{)Q1o~YOs%Qst96Dk>
z7f_aR-m-hoJ;$4MI$2l+9o>Z;q&hc#)-G3msM#puu{+UtNMy+wu!OVh$FBX)u-1pr
zw6Ju}7LZEx3wpg%aB5{ATL6(nYS5JoWxQ#bcec3;MSj;YE}6GX^1L8ZL&a;*^{$Jg
zl*6Uss)rCS#cbkoWq48b&JkU_98wyZ1hn?WFz&%9j)O`p0vF3Bwrgg~NO-^&>6O!)
zhC}!qG~%qm&#wC*0I2*532mQ%xUAsQ?Vi(giyP|LnI-><g2D#KaU7<K7Ts_CNioAI
z5C`_n4s>tHd-cmkV0-(kAH3=XeVwlu-*$%^=(A<p;3Ma0n8ZzXckSWNDpONDmZKXg
zWT@gztH8Bb{Gqd6jC2X?_aEurxXSlj=Guxk*U3z3H_`4U9k}TRqFL{HG_w)9p-Y$n
zmbjOv-@)LQ!{w2lAJ}ck?N@MHI0s*jQa<#~z493XT0$kKvg4N%Uk(Y#)9BccFlkZa
zaAw8$u|qsKOYHAU$xzbrl_V!GJ$k-7X4AB^n;X=!EjZL|G*;tD`6r~IKB6moD>?YL
zXPNf0a1s#T8(Ee(;v#g3_@b;TD4sODkwWbc{b`^8WD5|tZ#<Tw@fq1E@D)ntx>=iU
z8hWRbjj<{*$SHWwZN^bMZHt0O!{O1iZTJFI6+$8z;i-Z@pl>WJ?s|qKTQc}R)@#jB
z1}F<!CdHNleQ-l2I`+(f2&>N!!Hf}W*>E$Rt!M6eCPTlaxVTz|<vN{7o4x^UA{O>1
zKj*~rc&yi}N0z|nl0mT8qlo328;&nooeoeYeB|ic78*e^e16sdS|4t<k7O2?0FyA*
z@7UAwDzVGl{H@c~tP?@O*J760NtD$Mhv3cx(SANhMfPZQN2R`3+mDDab&}gK_5Ceo
zeyazyzJ8NC`$Jk6@a5baBUPwp1a3?xr(Z+66#UDj9m+G&E&!8rPAeKWINZ1Y?M1O`
z4R?B&N%TBlL1Gb^oAS36a(8yv7_a5@vuW!`2HZQ}?GR8~v|%1wenvrN`^FO_!hME`
zv9xO<@GgxaB)mt&N|{-4m$p;6X79l4#JaVwg0nVG`MioF?JqAZ44EJyi~uXP%~(Ic
z^3+aiXGwCz4zD`T{(Pw_Q3q>{mWU$WiAH1P5Xe(Bg~;jB2;A;=-%*CyL_KWzaV0#T
zuD7(u__EK4o~v+NOaz;YBD04{FGbRdhX@ui2DVs8q{+`$ibGlYP)becs=Z_Fyl;|}
pZ8RX20HQ=j%?iFBkT+0+yXj3ZfaLW5m@fEl!v_C9ZUb*8{}(r?m_+~p

literal 0
HcmV?d00001

diff --git a/docs_new/logos/harvard.png b/docs_new/logos/harvard.png
new file mode 100644
index 0000000000000000000000000000000000000000..f6ec08d7210d429e8f11f8c3173ca3557cbe0aa3
GIT binary patch
literal 31910
zcmb??Wmr^EyY{9*8YyY%l9p}<5G4epn~{?4h7pt&5ET$491xW5ZX}fMZjtWpvv|*W
z|9?NeOD<=CVb83!*Lt42R>TW6CH#9-_aF$uSAGi9gdkJ|cwUT+3I0!KpU(jg=+<(o
zau8G!i+g2`0lsH|Kh;!)AYWDpLcE2b3-BYv8U%UpLC}U71c@g>5QX#CS`A6?3rvgW
zN-*g5?q5bj&PNFHYfy&Cz4V^iPWAF89lhvmvbORXre;QupPj^~)gVUGdC9G}w^0E{
z!<{HLYe330oaT<ZYLm8mZZ~iVlS+^$2V>vU{5bmsKdVZ$^!H9*eUnUO@uYpca_J<x
zt8{-`o7HJgkC=7e*YmW~uZN9LSlgnrC=J2?=ZAH%d|caDbZ53A0RluA<lmn6=I(vI
zO5J}#G4u~r)G^%VwXlHPvbM<Hp4|OE{%J7CX>V%b?vq%XVar0zk3$?Hm{fk(|HEWZ
za}soXG~L4${J&SAd83Q+f1j2D@!t>51Bw0LQzOXr|Atcy{r&IAP`ds*NYuFhju(UB
zzhlOg{qOMqUoUv{0m^?V<!tvlom`Ja9@o~TAv+O!G?yYx)KnxsTe{Hie=m5A_N1QY
z$j?oRE|(h>iWXv7F8<5Cm7WyyEfl?2bjAGnb0(8QkJfdE4%_t_QNaKkqryu?W1L3E
zA8)ZkfB$}v^eO300Lf6?!;YMb6d|(onm_ycPsB>mnRciI?tMXis0#eg`&2{GsS{@z
zjid@`tl`mecwG{8<FwDTKD_!xQu3fEA)&bo9gkS`DVahRPUs6?$Y$iF_qP_Mw=^L*
zIaw>MRx0uZ8|_p7e1&W|g<f>!KHW1sbH&GbdLQp85xS6#bl05K)u0h!`*gHi2&P<~
zC)_1~V+b9MmCZ8ap>rBu%K(#kd1DRT{9~PO2FHwFBHl<63s(q15bPsyovXl}iZYJB
zG`%JUaR}_bCPvU?$PWe~JE~;m5XkJ6ynvc=w?eI1GlBa3N<XT_hrwY&-e@`sgv6yV
zCIJOB%_MAkY%*y=Pw(X7!W!vU0btU-t8*#G2}*;Z>c?ae!Qu9U=uGgdh=pIId857z
z#1*gglC#MqQu{U<Dth1@(aLUhvzUy}ETUN=kv1>2M>h}8H&j`*LS`ME{yl>CTmS9)
zuud0i{qpZ=DfcabnbTKCVrJz(148d{g5z+Y+*~S3`1^>6h{EmZs`1_LRmsZnpMKgJ
z2q0d`5e;F;J=M}4{YBZ0{XFr(SyKXkHn9dR@n|G+t$cdh5+!1&>E778a%(j2)18Nz
z{H>p69FbKUo=bwdVh{rp6B@iTrhQ{H&&xE9e`I9j?k6cCP92eCTzl82^G`HLDc{$%
z>-zeN3kV8+`0(LPRu=QsUK^oaiNDxEbi@yh2AiRmcn<{7nUrE+t_M3;Bcr2o!=CCD
z+26lE1lMW{Bkb(?BFOKu1f%hsFxRVbwPaV7mq+C$$Bm4NaxofW==syLo3Pj5I-dR_
zA}%>OGT?NUB2Tvnb8yG+W-*L@>1bFztaG^~hE}53s+Y=swwBDKNaPbMxb}kw55ARB
zkdc|+T%GULr0|c#K{-_qsIW0ZP;6~&A;iCbR@JjR!ajd`dVVdmgz$T8yzEJSjO)~R
z&U$m&bX)OE(D%rus}!%?dv9;RrE5bLl4$6yuCNz`j8GR=ZmxbAa3?egc9@_pU<?qQ
zP2p!efWo`lDGxEEa|AkqG1RdbLZt@qnuYWe#!qxiSd?qYnhPLFhziYA1nU47QiTjq
z)9<4&gO@{m?^X?o-U@DcUY6cC6EZ7P5tu2ZsD+jI(GqmhUgw7t-SaOb^z#`<8DNx-
zdQ(ddy+WT-vV!@o<2fz(3m_a|<#I%CKbsVJ9rvY5r1&mMOX4?^*eEeD^1|~UI`h6{
z%#C%89SZLvS~-7;vKPkPeYm`bcS{f7Vt^hko--qTJCdjpbng>?hyDGG(nYGU-@3H4
zRBG~i{t)7}&x}W3d{H;sn45kSlE~LV&e!1&`+ILhcg@Y)5c`NL!aN|F!?_6`8ge%&
zC6{F(cDRqe$g4UWI4I=cqpTcJMtJ!x2uhkap+!$Sy(sofr(yTPMEotjsbuqp*xy(O
z&=MD*6&oe2Bscs))yn~GsFTi*d(L<B(ciy+Gcq!u-rinHNpHSN$Jxkeye1nu2~X~h
z7}}`kyba}tQ-|I!^z_1IZm;gYd-u+J<7W<^&%vu=N4<jhk01Hn{;5RMN|4{42Hl2H
znEDg1rUyyu6q&rgpF<O&zg7z!#bJKN*2-3CsoQOgGH(ktpQ(0X=i%x2`}^bmVZRJ~
zZ=nT(KK_2|vfOd^tJ=QViHY!O$2$DTq6c0#hiW}(Qq=Lx6}RYPV~m1=Wxc0)It8tb
zmpf5k1g)0}>vmdeR@0&=OuSJlDk_phT^Iw4@;V1OLQZ48<uZ6juruu;=}1;ED30Ao
zZ_o#(32mQK{Cc8fucKX2N<SOAj0{kQbsn{`-Y50ccd1Swmi1hie-mBu0EQmYvXH>M
zE&(~=2JZ#pJ$ko{LPK285JvcdX-7k;Ur9^S5gXovxiHdqFx=wgnBLFUQ5TSx>x+j7
zd<bF4?Gb5U65~Lw?laSA@i+`0`Gw=d)hwdrNEP%7#_(a~dZg`%VDr(s6o!&R>o8`;
zypWSA&1ilU$`(p$Iam%>N+h##UiccW2B}Q(^mfYfoU}!NyvRA0m9CKshO4XV-n3KG
zx7^%&V4HSeP#8fPTXTmt0$-fR^sP}K$HsFj8utOYS;u;wMBLLkKVHYCYtOIlL(1&D
zr-s~w(r0E_tN03Tgu-@TgKo|#h3zTF3Jk+~Q$&`J$4xui2xYc{+T6>whO<@Cpy&sJ
zm|n9Ux?xMaHs?`_PoAJT)~>T&Ez{pNU!5O5c`qd>c>iuS{RyDkdL&8{)BgZ&1sy_L
z8yy^tEXeCu`g7_DE6BN?Jnn8>DLQo?V&i)m5~<7m(IxWw37s5~2i7vR_`}oZHyKw~
zS2sdK8Rg~WG*IXL!IZVSpUKP+8=AN3a5)L6>8o8In1<m4jWL*f_|0rmlA~BK_c-ou
z>&{GpSNe@A!{s=zBXeCWd5GuW+<!6>xwZAa{!8yaW=N@u$w+-m>3V_sP2ip@WO%WK
z!ZbzS-1JiWQK=p`eRA)=>tzoI{RH~w&z}8ZMBb0Eagm{Rr|YA<jLo(UQjRa5d0o->
zySw}OoEwIj*~Mc~shU^6NqUm_(c+mOva#h%Gh#!S_*Cf7Qx%mh4LU9^FKT6;x-9|!
zqd}!_P1MxX@KUXFu>RZj7Ta9@eX(X_W5a^X_}u#!lH=otHg0z`wYAAE{8v8fE|v95
z<EqMt2n!=4P6{o#2*Tir+}*K$l4&@w@<!Y+o==wN)!vamhlYmkDQK7s32n<lK_1`W
z0dff|OrVfCUH{y1b92*zXy#2i*&>&euQ{5iMI4X1O0O!A>=Z<U5BAiDJ;SJ=zUFRE
zT1et*xWRa5Cw$E@wzEtjKC8@EPf|N!#p#v)jRfM$L_<#o*9>R(fqo?NimLx|$z5E8
zJ<^pGQRhK-@WQ}g3SB?Jbo>4K5#I+9{2SX=8=9$fOi?h!Vjq7Cp)_C4N~$s;Tp@)w
zo((K}Ub^Y`-|&0ZGGK2qEBi{av9Ud-^C-6LCQn4Nn{r0FV#4j!L@<4Np-Qwf;${(O
z4)@BBuDR?@&@LwWKLco1f7zdsSjw7@GBRIhQ}vB>xo*Tcsk56bEGzqBcQacxg)g9X
zo+vIYZO(z1lqbWNkr5>$-~Kc{ZaGGRKxEck>kOBSL%!b{ztUkklqp6OHM(Trlb6>a
zEw`8Y22bgXn3x#A#&MPpm=vS;5>p~3s5u@zD&R}$6?_2S9Q-ZB1<w!W_cSzJ8C`eR
zF)SEM314%g{cVhcgCl~Z=iyOPQTZwdqb*1kBW&KhE@@zoJowOVan1`@=@)OON8u(>
za4~%Tyt|SB7A6;;g#!_wIWr@)Am%mE+ZWKTe!*ByX<1p3X`N?56`SQp%X5#sQG+Ki
zdXI0nyz!RjEsgbT0F@ZVXC0zMZ-JGQiDZxM;CN2qHDZYy>@1A1zR;?hYYgy<nzT=A
zr<4fiXZ47$f5$JZJ6!I78Degyw8?&WaKMdZr|Y1Uhz?g9cG=0;U!Tg#%7XV=9onkp
zxEa!5?Bm8Wr=l;Y(8}LZ$`J{BSkwNTw5MZ^Txv<qru{=HLD2kgpsI&0F`Q2A>C@`T
z*A;UBRnKh|)P~n!@SU;VGl0kxa8&h4G-7`l?oFiKYfeZ=7_iP9#7^(0xR7{*!&g$M
z8%4xV>rgdq_*sG+wK!+Rv&N|4M)-@$i_uw~CZRz@16j`Sc^F+c74h}Jc%4vLSy>ZK
zw5SB!GHBLE38w2F|JOUJJ$~GDI%?3yU(I@G-cmoXsk+)bfQ_6jox9u4Qs0ER`(RQr
zQ^O*9_UE{!EX=LDyL&#SJ!5v!nHy!`v#O}a^*7~*e?pPP6vtZoBtceXV(I*z&)8ZQ
z;{&G%Yf<WRS0ArCi<-P7UCHVD)em|-nS#dL2TPwcBs&u&+?u~bXR~KWI+)?gzQUlh
zy@qpc@MG8J@y+0@Doib>Y--Wj{~P-s8B2X>6>aT^aL3^J$lS1Ru%pn?*Vh(jdyAS*
zFXdG*%sy`@q;B9h{(>M?Rn>=!NLzzgm`Xmk`J&GY?TXLX0s`L*|9!HEZQ7+5_|-I@
zQCP_HsehsH@6JC7omP<p@AUrhXIE4uUUhFCBGtyn`8zl-#MAA@3y(=OYGSfhu1nHb
zlsVCzUs%VMF<`r0+8J_ia>frf{M0nwnkdPfS#-V%3qL{I`<?~|Ljo<S)57H%)LkW%
zwTjy4_p7y81jUuQskymxG({@d!OmX@9tOtsT1jNdR658J`MQ<8=?CKk9E|`TeT`Pv
zK;iz2i19*zmzP(<#3REtnP%6QmxIHFGx8vuyF2p%+*)o`lb%t0aP_dD!k~bEUUHlX
z<t(p`z$?M2UFMJQa>r1Ub6cV656uSAzj06E(s$>TH$CefV#`0pJ`UqpFbhORWc1X3
zt8TK8kaqhFw|#;U_&S$h+n!+QA%4+V!ot6yCcY+n!NSQ3VnQ8qb5NN@_kZ0FVG0{y
z{hR@9ZE`GhZRX`tabAAn(AtxOi_PUvNAOZ}7MFiR4XO}2YN~0cYo<KhSfd<~Vkw%~
zMrJjQ9*&AjRC1w4tzU9AIO6D#TIO(fv+;M(4eDFFJtFV^Db=vk@t#@=x>2`2U=C^F
zYf%53J}Qpaoa8w*-GrJ@V3Jc@Jb=9*kjl7x@n)(J#fBhk`C-Szpg5Lh0*~ITUo?B<
zoOFKBD455%j~@l!gggFn6tSfJpb9g67VqxRM1;`JrBK(I*7$2HAIw4+RJSI7byyg*
z4B$@NYWl88<KKO61Dxg?E0kzbgv`|*8lB+*u~pC>=Sj^m6*qN!v|+b!rsC|Z$y^a{
zdm=s=rN=}|^?u9c{#0jF2|)4rrcKi28c(!Gbyi~DLDAp$-g93*9ybapy%4WA-D!T`
zozU#L{oOmi29JIuF}Q6IR2DiIY;{8=?GbjXiisVd%r1h6arb*$N=h*_Uj~Mw{*+!l
zyuzJ69>^IpP#so#^66L$RzA1rTu)Zy|MKO_ub8sCL+=Fr>^2<LTNa)3M})Sy*Cx2-
zVL8h2LJk=ty;Bu-4UtGmPVClKsVSuR5FO#!<7k{&9nH;@%|(`0q(T&**UpUa66012
z{`J`KV?+swJgq7UmN60vG!3nqam9@h`$w67nMv7+?*F&ICbxywYvzsLPEAf*eaABh
zGj<an*UWn|q%Mz5B?*CZpS9VdWT7&zutneH-j~kFLt=&fcs1c5m;hn3TeFF!{Z0Zl
zDf*(T>TnF~QY%$1)URQ#45}a0dX7%~$Shi_fN6VBIAt5Eo4xTfeMtbf_ml{ksv}l<
zE=t_H9el|j(!W}V*(`(U+U6;q`kY1C0FsS8+%rV+mx|lybgXi!oU{)k8Z?BtC&xUw
zv8yo94r?pP$Qj?M#3*K55Omxa`MDk0^b}=cm}6nV$CndoOO?9tZzMS6jb!YIrH@T`
z(e)Vrac-3s{aeM%lFxhF5r`=58m7ao!*g*-Ye4S|o??V$R{bOy{A?^~MPD}Qk6I+&
z>{ult*x|0)O`@S9{D5eXYoI2fyeM19L3kPkx|ex%r;^Zja?!^#zqgK7-Ci<xTUy!3
zl7aJ4(hbx!UL*wvqi)Ljp8%jS37HI@&q&);7^v^;2dwYB@lm|R`}K~(B;2h}3=7c0
zVm)qAz*3(O^CEF%HApo~O=Ev*<+M6hg$;er&bDX^#V`0u*>p7Y{A?+TH!eLrwr;1E
z>s7(Mdq2zi#Bn|AXvxUPC<R_&1I{}*Fn}<w^E3ygORv`5whk#IgswBH(`tYCuh&^T
zwcsZ3uvf$-d*QV6wHclm=KY<f+rX={<=C$7?kA3pJOq+^vR{2Sem+d(rPYVHxVTE@
z11~G4-qZX7_kSa=sN<`cn|vo$jx7_$U%<JYXFbNhRaTN;ZdcJO#?rk93kZw#<IiDN
ztZurVVL;PM>?cbHoW6hm?tQ+J_@NoyS=1N6gw9Mf5n5V2=R&OFPvZ=rG|qiuUPpoc
zsm0TM`1tXofriH5z{gFoRJE$n*8#!dM`x1~<5p$8Rnyxp#uUihjV({i8Fi)@ey@nN
zTt{^Ju4ig$jo8f}(Q1qVy&l!)PPkonF8K9QTZ*{(7SrtU`78ajN6JGCXR|O5;fcUD
zdtvpR2;5Xh+-8p7Aq$$?-1FsRvDceC;b!coKFxTHjx{UfGhXun4+n`GAGDhBF+6@u
zb|(zfdTbg*eCXoi&dptFIGuX}REC|}&B8dMdyxqNCzDw8{(C6UtwCEDoiDbfrKR`v
z@puWqzMnpQdMlh+&i2+{+VrX3_EZHf7M-_QN_Fid5(jJbBGDDwtlDzVxn8#kHzl#C
z)#@yOp66u%rMNr0u*1~b{CsZ8_Z^-hZ(dniE>6H>+o_$-tertQ4HI+w#{l?5)qJ2t
zVfE5IZj*XS`hX*15)zWBPZC`~K`1eA!}M-20ZRc5$-Z>4S=3o<-Zr)Q)Uc|uxHKKC
zy_=+F;+`!<qgQ>+Sy*+xv5CyK_1xXmK97PS(Eq<s(`vc89^61WM`s1ghf*w&O9eLb
zB`$I3+S;qAXV&$cOd=v8@A%0bQf<omSN$HEoKp%~sU%Nt?~M6sGbU3I<xR;tP9WKN
z28C2(FTb^-7^%JnxXy?5qpcX+UY^ZX5M9ojxLIS?!Ejc1mk$~qc6D~zp_CMNr<as;
zr@2Y2;tj>c@!qLfZKV#P#?4^B_P)J7bpg{FHY9bn5YqMMHErnOCoQ!%4A@Che!@e;
z!_#`uZsVn6G|jq*D9~}Rv~twi0V&&C&q-g(PH~ji)YQyLD0b%3v$SNHaTLk#VBShh
z3RPBqASw|+xzkJgeuQsz;JgE7BvTE=;?agFg~qgLwVkwAggCH2dW0YG!%VELh@vrc
zA6$^&v)9SiWc7XrwafbV#HHi$>qeJE9^*)G``MXCkT}i!@Q#IpZeGIcfKFPka;_W3
zZdwi~K$;#e`Wqhy+9gZ&V$!FU>POb-YAU^pMX%Iy(6b$o+oZ;sWfEUH?f6Hmc}BlD
zzT_^{1)myYK_Qj#_K)|@%pIPDWEE2|+a?Spymq1uXEw95vvC(=$!I4lkMoUbS(Jx<
zqcs>$Z=aZEwC%FM?T=&&b4p7~6+0#{2C5AGBHJx$kagpc>sKc649MnBGp}<6Tt)29
zoDxlZZPyL~JwU#{;R99S1>S<$MCqOzdQNn~*jP6vo$uPW7i-lqZ{x-{9m?v?Hso}A
z`sv<KV^S=X9~oucOQn9M^ySL2G+Gh<a*_JWZh5)T7T{<OZ)aUPX~P8jk0jV{!4E(u
zwz0#GmG<=#{R0+ZPMf4BGla6RkfkysS`8hRDVcdDU|h0h?N{+D%@=EH7rm9Kw5*oZ
z|86@qDz5HgE_sbOw#LJJQqq^DUsB5o-_n4jX_10Tr0uv%?4i?ToHE4yeyx-#)GKwM
zgXA2NitHue<6l^$TH_x@-s=gydTu>kUscE725)SRQov4jen?t_>)mvH(l${7M}bU(
zu06^%D4sog#%DW}6_*rEsv>tX>t%XfP@MW0k(<k?K=i?B%Qt`VU$a=7$S2sn<{06W
zw{9Ch14T?dJUj}}&8$;oTgTNmh!Gzw=^dUe8k6R9N7IN2d`mS>?iWw(GYz--oVK4i
zXj30Q&MFJ5WhCV5upqDPTy!QC?3f$>(zN+z?v#pGd1!H4C8QwSddn`V-IW;;z_byV
z*S2^@hcM}!QKBFvRiRvSNa5GX&J2jCr+`#g0o|u~%$x^`t>Fy{oCi|qn5uw6)vGYY
zF1N~1prGqx3SlS5?TjlRhqT{Z?$S?<5cQ%`W_^>pHCXF<i~#iPTfc?K<>SYE_le3O
z#rufGCJz{3+lW~W1@ThI)Ao$49hK6gO0jV<WEfwPu746nTBDPpqdjU?{k`+pzz_6V
zr={cdpX2(bmL>H(#8UZ!>wtr}`qRAKoAGRntvcpSCc6}a0PP@gCqnXAGzUo0&%~t_
z2U{+fU7l<|iY^J-u+J26XERXW!eG$j`Y`^5Iw=D}N5o2B{1|mg54sVH5nT$`3C0L_
z#PQp%7=NLy{SIt+0-#y9u%-JWpJV^br6Q4&PAAMyS)N*ukK3#F=p;1r@OVuA$N(Ms
z_H%cR1l+?)@}_7R_pgo~8-pfsMHuj9f(HVB10Xfp9IAh=HCE1CIMi8!>wS7Ax(O>4
zP~+(F(%$aO&f`!5+T=D$3Bf<=b)eSBgpW61!x+S*G&8Hp%h6{&YzAa#1JZh}sMplA
zkIT1>>vfNpL2nzJ^db<ZdY^M#1Ci?4eUpcMn=Kxskqc8t3&WLVY>j7m4t7LLHo`RE
zSo@W=_3~Yj2NF|>+p5|uBRe~!t&Op{*rVt@*V;|XgUy6f8?x^$DSCT#s9|Ijt)(tS
zrfck>ozq&CgUVOGg<6gJ*Zmwr>;u2&Q&NpH`#m>?%6&m`b+Lu<XZ#&Uu;hVg#bx>i
zhJW-!Uv|)y<4@KL#QCV+C!&OtZ~v{4rL6PDcjQ4ci>{WT|CEpr>Ob{GC7S}J+Wjk!
z+;DjwruSjr@Ze3*s+1kJDaAz}&#q={0|Wz{VQ<fAFdOk^R;SrC@En7Mg#|+3QwwL*
z*VBa2dg`$oJKqzXdbu%E<MyyyU@q{|fd<qPz+cu}s+o+xc{j|(k@ajIdm5r*#NsAM
z`ESGblDAb+=6zOElcWR0Spf!hYPu#@m)u7u=QWvm9p-pe180HSx_7D|SCp>x16+YA
zp;@5v+(g^h^~=u}Rz_Hro&P}Lfe6Px*pC7+{trCBK|!O@BHCeKa^-le&w+Swb^&Of
zX*Y{_5Krs@f$it8JGTPOTYpuosC?2s0<FY2rT3p4)Zvlq?cb>G-A|SM6Ma!NHDKT6
zjrh)UlyG6KW`L2}Fc^pzmt(JEW)r#fgE_Ra-&$;j;{LQ*t==l_HY?HNgLmNN>y@Bl
z(KsV`h)HcRHeLG3x~Sp#qyH#6^y&tm)SK`&o-N_p&((|l#TK72m@q_*3pa2;C9c7U
z8^*8~He^K{W?qf_V=*=`Fpz`QiVMgQVb+or4}1TB9R>0lryLoep!96LPn{-@T5*al
zvEg@@c}Z$%UGEqSXtV_4snU;>RJGMujf*jbfz>DnY+_?RBI*hT>`sqtF-kZ!D0Gzv
za~>iDVVQbP`DPsx9|AZbg|Qo|X6F`SwKJi;%iSh=5(HbCx5eKs2yxR(Iv7ss<B%ig
zC;BSd0tK%sSaBH{#dGf*CiF!`L&WZvJcQYQ|3o$v_A@tk2DNK_1F>WQeZY;i6H@gn
zO0`~gNzIXNw<w9G4ENwv9+HkO>62gn3^Vv<p{%CXCjn#Ld^*J?QI%O^ghMpmOFx)D
z{pLHqga0iClHdQ^Vf{E7np3$dOk{0M>jp=y+<FN&b-h-|Z-`5)c@g!I5u_MujLT+&
zVN_}VfSkuDTsu#vf*8?K;VK2nF(bP_Me+FNy8%^VQX?4(6X^#2xLC=9r*~kF^6_iZ
zO8d!R>+}HOe=9MhO^ZH284YatlP;*1c)?8WAT^w)mNQJdjR01}F|1srfB6?>og-pw
zvq@qzF=8!d0W)RJ44^v`2=_QLOAoNGl-^9DS=5U<?rxdRO;(^BxIi&6XSMpn<+>6-
za2wFR2CyiLdonzJoi{2jHD$1f9SDRJxHGb%KwEs6cIflrNhGw@g*q{(7`|t_K<!wK
z0>u6O?bZFW@tc!Zg0NNc<{g~o1&x+^kIHkFuK+1Bv#=}zWP%CDJ6!E|s)%4=hx}@B
znCrI&vm3#3%A`<^k6}XE&|Ri)7Y7lcH}L*`9!f}`#^1bg<O*4P#cz24p>%z_o|%nJ
zhA=}%X~>)>adSq`SeLXtkEmU|gE9(~a0BT%LYQhqB+w$!CbAAm(5${Z2@Guf5^l1=
ztp5`*xwzB(?UR(IHJZ;}9Ffm>RZuc}S&5{`oA+$ZAkp3W!>KHsd*3pganfYiOBv74
zSJR&?j!fN0e4u_<Yxq#E+uScj7WrAi<~?5MTZxki_?>M6Y`Vo{yFy#79UF4l#P07B
zv5RB<PjADZGb-7b8x-{;+uHP3`a63VYtEw5C>d-+3dy;-ZkW#kfG0LQP@}a-7R+FT
z8L-&t`zzs*q0wGv(P-OoNwt-UiHQKQG0jTOYi;fM=<$ZG-@lbH__!k3{VUyOpGLWx
z{9{h(750S6Hzk3)fXSB?^2!nL_30Y7oL0Xv&&yt8=;#cKcI5MWhEP50m(iuTdb=;z
zCau%|Lxq5|muT#%{}P3=u?AUELTp%Rj{<R<HqzW@Q^WgSP*l|UJ<%kxwYAk#;l+!u
z?=Dfcc1WFHNMaPYe^`krpBo(?UoL98L52JQ>`fB);DpRtf-%mn4*N^I_bu-5eolIz
z?^8%Iua39rO5mvG=g(#RZ_fMFs*tM)9|YgxQ;w%w!_t2+W4dVr<h1zyRSA}7v+Mea
z*#aGDBg^hd;f46&jPoWle__h;JLr4r?!`RO;p70-_viljVH2}lRvaGg2y^%1UQ_`k
zQEMCcfoVqlHbQAiu)$VD!@yi=@EijN2ZzDcqAJk6M<z??Vpl8PzI}T+;3F!SZ963i
z`4t#aD#sIG?a?93X_=>g+D@&j<s4Eti;up6PBzP*5G!!G|1#R4MWL+3c^h6^{$QyU
zm-iPCFr5E>P(o&8Fyum@#7F^&%EsOv9h!0Ness2yXv*il_Tr_s_F`U9WAN}go|sk)
zYjE%qn?I+}x0FFX^iUnt;xgC%uVVL)j|1gyfddTRnW^dO?N#vf6ej1^Lqze^VvifS
z;R6{Y$rEG2Z?_@rbZ)v~?xus<=~pLx(9DkGGMGsYf8dUVprPH0ZWA)|D8Xh-A?;{>
zV*=j?a9*OpCX_7C_z_UV)=RmYs;WlK&CS1+AdyS%kQXCkj~==SbjlIwDsr?f@c!p0
zkg~@i_tz1lLfe>-6bWLxnpE=vT)newCc30X!|GGosGGgMzJ4aNaz9$Ng3^IL=wP>B
z)vS-_Mk;~sd=tztgSUX{F4=wdG23=kEVYkhE^kAAkO_LM`h3{0H3_%bSqAZe-&w*C
zR|FWv_S2OGvqF!6E)N?34CsZ4NzBrc1?cG(z^R%jx4r+`&hA589O6!bxRd<DQ27L?
z6sv$|vkiaY?@@UvCL==+h$=&5WaNTporacH$jr=4oam9a<k@~F{i18l6d3|Y4?1~|
zk4{gvr#1BTBfft9dSWu8@osmA!;ByYeg_9#YmD5G^+Kj4)7!QATAp<cW3MY3r%Zkn
zVHXnW5nr&tggA@oN=MB2UV@r44s(0St*l87%tA9pVp?!X9|6QyHMLtQgq?Vb&aNuD
zoL+|Q)b}<2nt}NqtHY-`!!->O<0U{HvPmX)Sg7~~Fmz{LeLS=3L*zA6qbHiTb`>#<
zH2tOau_CSU>rw{nL^SUVdSM4J%Q-g5I)%1VeL%{cYA!|S{0E}f)J(QgPAd_;4T-Q3
zN1%xh|ArhtTb^tGG8bzzfra#03v2J<z<D8kJ1)J2m5k>Qs;sByrQ>%=I(07MS7npM
zV22hO@J?OPmU9Bf5pDB^HLeSvXVHSMbVgn3_~jZfoCHbk-#4TcvF@V*43+KWjt%Nw
zf}vx}qpF#qf>Nbv^ww(7_;HSpkF5ndT$w!L7_j?2*M0WYc@>TK=v&OevM>YVu9X?X
zf^PxKkG~@$m(z%cf>{~$>chie)>Bd_+c{6O{Bd^+iX`GY>?jEo>SK>{VCGvX$4g`E
zovPIZcKkzG(|O^mI(+eNGto0YX!m!iQRXR4dwV+pcnzv^cFf;#QgU)0?5+A8b!L<^
z7!f*lFjO^7Yd=m0y_fs2w(qUkhEJPXjMjXg_2YUw;?Q1ge4yemJ>0z$J!X=Xs45%z
zMabU9^ILW{E3-eemWH|$-Nmu6L)43}@&X&+Lvy$1Ok*OOPAPOl3u7<y?Fz%jH}O@K
zl?)&WPOs`SyP_Fc``PBVhSGup0t}IijrKnfVL~@aF5>=jitEmSz!sky(l&3QywuE5
zSh<iY&K+7X#Y(`CjZOLF0qqP4IphNk`V?!QqT`A9bVobN>Vh%jf{OUm&T4NdchLPJ
zd2&+Hza)cSH2#cQrbih&<9sv)%k6}2q)^p-+aUJ)&#&PoucA}AyAN#i6DVBw#{<rh
zhhih!jJe(9&Cc&lGky0xXL<5n#*tzwqF<%mZR${q+4`cHau9x;K>Fije+!rjHgitS
zV#^=KENn&r45rYOc$a5WM1QHhqJ(&fr9D3{hIMeh%&Wt4#G-X40s(<1NZy223R#tI
zIv#&5O-<+DEQt%CJ$C-drpm&>#N%haBgrbydt<DSlPSF?^qtci0|Z5~N;5s)nlE1P
z$!U*rXaxFJrsVk3G)IV=WLy7RS<$~(=aG}||9bHn=KdbZu6{a>C7-ae63<)!IqR`0
zYoV36(Qo`Y$H#25haxmgUa5DK744t<yHUbFSX`?EofVn4s!6({q#QqnlJp}pz?5*I
zw;|sct@#wxmYh$YKR<TeCJ&*O|CDc?pf+5NIooeSQ8*8sJqkd~A9$#@dwx1Mh)Yd<
zpUI}K7#JvhaCqo)JgQqof}mA>cTxf2rQf3k%H^G<e#hg$ss}ntI|xU-338fwd+FB6
zV0e_{_f4@yU3FA+<42}*4}YiG+Q7FvA08+b<R18D>hGNVw3=tv7LfO3p*yw68OV{D
zVE3)8tmrlPKBn{eOZ;-o$2l<6H0(#+rNz9-RUAho(tmjTJdXMI?EHqJ_GlimK8p)b
z=&)I_^tMGkP}0dExVZ9bAxgY)sJen)_=^It2H%Wt(z0Q<-t0Lsh$n=*y3_IUb)>~J
z|CCKwIS)TnjOCG}Amr~L-+b!z2%%IrpoASxcRKz#De3GPO%%I8M`8F-<L!8exZhX)
zllYjdk1Wc`B<f1zzH_InAAf`5Dyc@=el4Ub!9bLGf#CVNZ1eGVxulHDtrOTTYCMa+
zm~d`|RYY!~2)Y>yR4sJtcCGsy?T`N>(bE~X{h{Z)3aK5@6&M4{`)|4JOGprl<jsiB
zXl2&@1lIxc0jK7_!imosv+r5w5Qc;aaXaO<^oq>Sbaa~iYNL0<H3Q_L=TBko>l8no
z7cYpW0;B>i+=U)LUM#_gUD!G@=M%v6IHe*a45Gmz2U=+V-nl;o%yyO6X}+=Hd);nV
zvw^G>`FOXho7?VA?ATi*lJNlok3sbNBVf?rAyS~P8HYkuPvbvvbsw-?YoG(SD?+Wr
zu|$vC8a$0jJ2WU5`xyR%;aQz8%T#vAT&qRNSJY;beUJO_t>c|E!G{kQAH{Yan!0bH
z@pc_U7at**S+1wYg6Uo$8PVWps4z6h!Oj|K8O$EZRiLx^VSlElERl-M^;i#j0QYQI
z-TFh-9O;Zg`3fj@+lA(v?;j(6{`@JG`^zQ}`WPQSMNb~dcyMs=VrRO*hZ3>B-09_=
z+IIzcDnSJJDg~YL5U>-}#_7PK;<f#Oui|#ujIkiwzItf<w$d)}>2Ud~uj?F8;dC>2
zmtYCMU)9<W{^SASfFfXReh9xIkln=$){kd?w3TBy1g%ZEuwe)9uu~*w8`EwZgXF<S
zm&(GP_Xo2Qpg&gDt<C^*8kcyzf+5YS!T|Teugz1GroEJVKG$}NvSWf!oKW!ope_am
zMom52wgz>@##0@gmoG=S$eZs$KM1j#HL^*WaT)*hrA7fonh0PSNlG{fTi8Fwr{aJ@
zsA08SZ^frrX2|%-c-{&KTuIApZf@>_mJS~;G<L3=1z&n)R`@!FMiFJh*Nxn&D$Pl6
zNqbTyD1xr`vDC!ZnZuFc8kveM_QGL#$2!L^bals{k)8lw><EpL$3H#bME}lEVCUjO
zgSLR!4I&k=cm~Gjcw6yv4k$+loi1j_QsFzro_P$|I?D0+QetAn1$bN8+bViJXnS}U
z>q53dqzdQ1?1kyMv0FQ#j&;$$2Fmfmz%pyP(IJ;3G8q=yc7&ct)TCp9Y(bm2p#+=q
zkTa|*E)ZTdEz+51bbfxGi9$p1oyL$^S-dbTXFgjn75kQ8gZlKX=eE1}^Lx0sIodip
zJR|FUwpqEkyA<T)bi;&RlBa&CgSJQ52p)%y%}B`p6=M2(t}+zI9bRn}N7hd^>2k52
z6GrkLB)CAS;yD2_nlNf{Ru*~`1+SvH`9q$XzwtbVhrX{-xItzsinnpC==&o>AORTY
zw@Tlf!F!TNSy72FqgZJ^)=KcTt>}}GYftZYF~u%?P@+RwnE!SOl%K`iyuy1p&3mp~
zk-^=(O*9})L;-4M-6yB2=_VTciDL6vo=advyZ|0ifninXFSBY4zUmv^c)=mpCxtFl
z6>%6PF*5OJVX_MsHs*O9(3v<UlVO9rOF6Uhu%-Tuy?U+;Ko7mIE9mlUr#L&3vX<i%
z;+d~Gx(^O+UJ}&J&e@mfrI|%vD|I&1(ltk_4H>oTz3j}3TGHpyAYEm=2o0;gSY<+r
zfg;e9$3^~iu1C!p2UUItkk~({JgMUFX9^1KA(XN%ac~8Y5Movx7T;!n0GIJtsSo9;
z48NE7uMC|&0ViVJ+kQeC_$>n&=P%X|7e3Mw+=U%C#|rGoa;j=;slYmV0IZBVhsk`d
z-lwal2d1cqNkvUPt<tL)EqO-@MN`%A=p-pO$BQ}vfe)*&8*g27&L}M<0ddN%t}a;}
z9deNU0WlYX^nR?^&XA3b*H<93HBs#XU;9cZbHx`Uu_G!ZEF8*ho+5IIm1+fNJiw1I
zgSb1GX%fXLf41)A{^3t@K8+jvq0{g~s{vT<XN4vA#*Mvb`=l@JfmZiqxIE4@OV>1j
z45VGXfwQ&`$P0s%6z%%@`hd_Q=AEhKADJ<&K)b^H_!~W>1@WF*#G*I(F%X)uo;`nV
z&PN&d`SZJ@BllNUR*b-3P|PF-9dOTE^6mxFw^wd=-VGwCtW)M@pHN6h=<Oe4fWvch
zIhh4emPs_tEl@SliLqELEt-DKZW5$;;x-%HJSCt(<=wc6Pu)cEugn?!*0wldDV(Bi
z9{osZbw@B7c63L27eQ8o>rxBoKk6t+MVWXxKu^|-0wKG>K%tag<>-><$aNL{1esV}
zA5hY@prgd^7`D>syCe|qWg7_e@JUGcnSdM34l6+_`UL`$pR>UZCKd>~*HdfL>wqF0
zS%|D6SA_RUlo+oK4+kRQ`|}C4gL3ZWr?@>dM0(%_P89j0R<5G@V8MRL`=Cbrx|>B3
zqf|gFifslNf8<62A)XAVW)KwgmngkI?Bc?gl#J}1XLsdc#`0IUJ~=r#6u@Ejfumpn
z615YR4%7s|+h2G}tMnbObwxkE&>x@VJtZ@h26X%BY{odKb={p&LV_Ch_OacV8Sqpg
zL}{r2Gc)r7h+%dB7ma#!iu}E0cY<7kC3M5FP+KdmSD>SkB8(4mf#9xyt%n9+HWd{W
zGQscQ@9*m4F+<>1hJe2??}&Z?WN9XORA*2_i5|2Cg~i}+yT>TrcfnndWs;GZ@1c~a
zptoGtE1Rru8eX3%&Kdm<9vGB|M)&H@)>Ukj6*KoAL&JlyGduu^-^cdw46xoMPwxjn
zB4zr)NO}nSejysF)_YwaKpy&;kUfDTDuwUnQFC`Vwta!<G)K&&XtZ$b8>H0f5)^WJ
zb`J7^9so8N`Vi)&EPG_x4ha=J=ZUK2gr0M~?7yL9;d}G$t*NFa2_-y9*n!Hb?`s6`
zXD%mbWCHzjQrH<mDSPfbq|EkCSU~dnXt?^|50~M0u&v|*0;Iqs07J?^%sXagkI!*N
z)F`N-iLh>8@wPN7IhpEdN0#Gk>RW85wco#gpMl&mJ_L+t4xpD`I~4(CyQGsY@3I~Q
zdG03cKgrC(LIRM+{sr7(ew%>|#}&7WqxI_Ze<?1jsa|)fRa9sM<XOK&&_>E+jFLJ(
z3ZlcpeDgv(^yX6z0>1YA^w$Ll`e`f2XQNcV=(M?kf%Icp+oJamna=<aV<6d0@S7)o
zN8OZuU{ac!jai<zNs*ZWOS#=UL1A^oQu2W#@Z~*3X`(Jg>5Zw}|C9N#DoZW0of27-
zL^sK0kKDy#ysB<lCrar}0P@8L@%X@r(TlCuohC*(7kHf5782x3&S3)pPf&_D(g4pE
zME&kkBO_xQ!vw4!aljgueJZM&EQAAQ#JT1w4Fv$#F`c(8%nTq8IjdK_XauUf_g)JY
zko@}wSmUOq4E}lwk~^1;`8ErK_;9{a4GHAjFLk;ws%!dgJkM!+bKu((mkIMQ^FN<B
zY;0^0oy&4oR;&jHS2t(LDJftJF?0)nCv~6tM@C-K1fDJG6`P@ebik7p;dS<^=ZA8T
zWO859+pkh7t*!EKu#dlb1~%f~45*d-TenI07M^%=4)8@483je+cEoY=r%yuT+n)5O
zW?2MpT#I_GE`K{#6>adI-{qE`#UJq?aK)9AclNz`Q1<lB63L@1Nf5s^{#;loD=TZ)
z|1qyHAY?(>?1$N*7|r1(azlHjw5L~?P{1RouKr(}V-@d5=5_P6!|@sr0GD$YbYGB_
zA1-$jr?$uNuE<;6$%V@87<9G|89Rw)4gRuLXb<}=MX)*FB|u|kZkhpdi3QM0Si(hu
zavctUZu?uD`+GnlS^}-Kb-QXl=5#)Y9*Ed7itnl3frc>9r`-U=1||*;9BplF&?uU_
zKHLOW7Z$1|5=ENqj8WfT?N5*O*c>wlO3Ak%M9`tai(j5VlBye*Q~xSTOb~_!c0x7S
zA+_}pS*e3!AT-e&2dRu`>~N0aqTT=FQ(gT=&d2Bc0TAQ0#0fxF7RI~P@tN7yLe|I=
zCg03)+;sZOced0rsvE#u3O&LJ{Up)MY+I*E`^pF@7iAJV;EU@Gm*cuKHxS%(iNBsG
z8C=?yCvC4*T(EUI<%DmkhPCaRIE8a6$Mc{Uo3gFP#4-PT0jw%NgXzBCziY}$OI<mM
zY4ew9-QNQbppafq;!<LK?P=MQgq2-iLx>zr_WOpdE;N=W?fskDHN+KUZ{ghul@cFk
z9Yw&pd2)*-11zn#Fbg=GN!We@ak*1Rc$a`cBb0<XD6RU_-rvpXUAjEQsWAfh-0pQ&
zAh_`%DG3|&S)en~*lnbxe8~v<wIE&(g6xX`g>D^HKb8BD%ek?+SrTyR`PHT40Z{aQ
zDNq=bmcEq+`ra8pR4(0*suq6TSKq3HJ9OxEtsjD*(-W-;T8-^PgzUY$oj>TGiQ$dY
z@HbLnxSS0`u39;aulA2snUq=8bjs<kT5VFxyCKUk4hc5c0ifsuSsy1}0WW->Mfu`h
zYLV*e$KP#*Q?Q5!Ycz%)fka7DC)e$hFoIsWqq8<>Pu|H^tz$x-amek4Is^Vi)@fsT
zho;<f+n|V9c@^ypUpAD0_(on%uElBFft+ehk2~SMJ07Jh>{d=>?7io6DB)6)GLcA1
zPSDo5#>ji&4VE()P&wC&`1trZ5Ch52&nMvZKKRWBA4DqLg(bMWh^@c6VBUdkAQ~j8
zux4eQrN%2s&=Xfk92<dN4?L-5u)@Q-SLEeS;v8pxn1$C<JJl28Qt)hj=q86Kg=}7Q
z`M6i_Hs9@Z=Vp|yzP`}zmEc)NLRIt4hepnY`em5Yl@4tm)ukgA^WAXWK5x8B{1Gmb
zfQDS2q>wH5U`u9?<UJkpNL8~|HuEhhOPnWS6sK>F@_S7yK@)?4BID+ZeD~G1!k`Gy
z5cQKxM#-!v5T<b}HzuS(%KH$R#csQ_anb%tpcjQ``_V}qu&4_tu(X{>2Cvs7#k(ef
zD@70T<uk)m72Mql;5z_1)6P2db8{QD_4K?+PYEY5QN$DQpov2EXH|@2?%EJJN+7P8
z-&RUoXur`)_4Ud0e){cIPG?^f2qfKwqkyUN4M4;YOng*g^ZljvA9;Cg)h^2xr_QL9
zub-}sl`x^kl#J)=hX6b7S2<bzUMp@jNKbu;j}La3t~8%4wLIG@oA7O=98nIJ;<dHe
zGZd@AaAr7?+*TPzIb$WrJbou$K!14h3I#jpS5(wXZDYcWVRn!zZ+=}-2x4EG7oeX|
z|3?v-P#$7J(@Nf)nshI5>xz3w&-rYomYjHSnJ|R^MBd3{0;I^E#~%qHf;_S{z;dN(
zzTJ+%H`Mr41)|wPjI}VOgC|^^oI4nMgmZ^Git$<9kjeo6k0h+?k}5xDXp{1V;fCFB
zd)}772;1t{ab%k_9#1V8CBHx3!!TiPG0u`EGC<wzN`6eniT!pq@LxSF+{jwT*!Vrj
zk68f4`)*I|13BeaeQVBlt15|#*#|#^!wt)n7E(f=@nPEhrBFr82izmhcq4CzA(gd#
zV>~R&@B>yLJczE}6ex8U2fJ$@XW&%RqMd!DZH*QNF<X{ry-T*{G3F||_lw<zub{7H
zgV6m6`UGx6Kma&h1X=<U(vYDIWAG-PnHDhqD!<pp6rh)YWR`w(3GaZ9O&>a=uQiCC
zzT1!fIQ?77`)SViRm`%<D{%y|MxdW?+88~E4G17?2JMoywFAnkig+WJhC#&k9z)5;
z+6G2m-rc!W2`}O82#30wn)4~(IJ6D;T-iW(N$VRBC`;I*qXa>-Oh_`HX`LEao6~`U
zb@hm7L_z7eaX3Mxuex$y>!-i@NiW|TYVE_d9?#|Rw$le)5l5EuKNVV<o8f;EvA}ZO
zO{NmvLRqpXhyOIi{hRe1k79B>ziS-nM_0e6ZT}J!>oqrFKZ`I7s~qnf`-h0dn6DqY
zL|8MFH>U*hUwzx2+qTmD@R0J!p!5_xQKWq2wP29ssbjs#t_ozU5tu9RMTNeTiSSK@
zrmqqG474yURnthOaQ?@lM6#8#PG9Lj1DrMLM-?qf|9dS392zvUP3Cp6U<SO{u~D30
zqVZy{?E(kO=tm$lIqE`k?Ih44?q(@8VaC^kzSksGVw?!nIUuYYhTE_p)vKy<0~vjR
zsvl&bto$1xEb#Z=zijN_J#ulc%EB}M#!Wxhf*tlW8P9VAR*%2;SQD5r^0mTPtj!g)
z@WDyKI8I7u&DUzN7=}R|ztF*H%up=XP74P&+A)#w5NuISOQH_uvvG;+ds0nm&vJND
zu0#VKqEtsb&uXA%tSwDVH<<^j#&gL7aC4*+m~F9;hS9BG0QBk<qlsxc`72H@@TzP3
z^XJb>Ym-vSuvM^zfcDyd9Fl(tNk(Gp@@nv*iVFQX-JRS%8S=Iq1O73@SCHx(@G$21
zgdOpGAfk?}W8yuO)<#EqE50OVVqK!=2#=ORGWk?8VDFLviZ6y3M~ehJt?0gnIoo=2
z#9B|9;=PTSw`WgSNS-x3aw+&M)PlIL>Bi?3M^+|qSSO`jmt+14FO1o8DUusUmmr>x
z=V68VLCNqmDseteGY`*)6BB#k>6g%@;w3nB<6T%aoWaY=9U?7)bDa#xsIRtmD$cdE
zFuh<(@8xk=J?N<QP=;Ybne5-#>WA2^GoEvPK-LMOFY*T`AlTgX3L|XFM_)^st2<6d
zBDtB7yzZ4@PdK@_Zlz=|=YT^u$A}yj-=vOM(^Sw1+}=MNphdwhFU5AjgjgD_u%p#)
zceZVk$I><vO_yl{A8~N3E39ElJcUt!b3Jl?zO&$Xr-Lz3nQ|~geF|vSQ{(46TwHd{
z$Q|sOc^ZPwNbi+ldto->!S?N|D}g0@OL+97k&isZQNHS-KpAa^o_QpAHMU!b(V>Oc
z*Vo&t4mZLs3nN<F=V?IbKk_%?2X&`C1mz^+HorkdnOJvV<sICu7d|oJCVx(cI8jV@
z`XwO)98j0_8M8P>>{d&^HmNwvbF#J5H5p;>VSnDlIfb<Bp(A)2NW<9+W6~8gBnf5x
z(B6$N8(V`wI4<PtIimiN+4d@JrqF9^LWAJ8FHov+1HEIy?#+!2!%<x)HC`zG+N)t|
zh@QAthjit;o^hQ7YH>V@WocsA9yR-?OM?LwDzrVeqW29Hazu`bM{!TGayPI7KIHJI
zNSIoeO_H33UM?J-7s{JuD_{w-C~vr(K6q2?U*x_b#TdC{=+c9H6F!xf<59|e`YuQd
zKd>8a%?DL&VT7bOMXZ^Me<U=I4v0@lNhz&5sq*phAy<w!0oxN};M=lUhu%8|>|s&`
zAF%|VwJ~8-RAKu=q1**L>`EVb6KtK`MTK?u&L{@#rL=h64!s`o);sNfYJU)uQS;)p
zH8@aVhcdFsflAyPvA^=sJxe*p;|oZ5PiEOpeG%`*hfaYMCU{xflAC*heWae*NYu~H
zqm$=BKINj+N@+N$>yd4nJm(Qcfi7Hplw=cAJ#BlLPm#8IiL&N#teF%Sx2Ndk@r=!3
zbVB!Z#$Gr!o_XCRq@@_6S%vnv9w1L;&$<Bp9<1wG$(=;;Q8t~QGcKpHW_3BlJx;Zz
zRbdq0hxfQ)pN7?cndUsS6;ds0&!y+1#8c0tZF&`5vK^pDfo>MnM=`w(CS|gu`p2Yw
z-#F6_CA^`OxNY4X>{H`>try@#nt^Fi3MU80<TDuky3cJS^30vawE7(8Zi>~fkzTQj
zz#jNa+8n}PyTwjx8!epTpu$V4KP==x4mEh(cqzx<P}mW(V+TJ@I|Xfk0I&*I<@iPt
zbaO$3od+HtHDtureL$?Bpy2BCH?Qt<tpGcoD4HB3+|oRe4xLQ++}77i;<|flV>yHK
z%}n<%1d2fW%GB~#>Fe*OpB>%=?W+#@IwKd~Bi$Gfa(vG}q=o2y6bPSEtdXW-fn65j
zn>heZq45(N=L!0dz_;!LDyo<sV+7lONV|P9?A;znE9UwNXw>u>Ay2uYOQh}Qle5<y
zrp=;LA1V*kr_C{TOw{<2<mOIs)or3RlB**EV@ccqP@o2fNlVJ7)=QJX61M{aHwLgc
zaq6FlA4%!z>DiPU6kvea$YoYeG~`ZLWl^4zg{hrgi>VC{%AAL<DWvyT#d3E?4Me4U
zthHO}MSUSA<q+;H*MZQ>5i{G53l(Dtio55F+U((H@jwwAPyj-L3^>~7(Kte{L;78P
zZ07IkDj~VTJ{#l<rzsSqde4SQuJ3hjX;xvd%ZQwoSjP<y&jva^yneF{Fq&@4%?*Q+
z@yDulGwbf2cFh*TDTE~GixQCUI6awQM+_j_?F~W<NmvWSsjXv5uyOFGD)a<qf#}XK
zEajLerK5yzPs08ssbqUWG_%T@EF2uR{-!Y956tTx5QY095ii`n5HS2kiCsUCaCj&4
z7TH%A$MgQUwu}L+c^cvXoMz-<TtA8td$+49;pUb6nTWyhtH?T5hMo=7CCT^U(NZ+r
zRgkyRaQQ8|B<j%(q9CR7aocW0;K0bI7QC+w3EOF`+;+JZdzY^+;e_3dchP-K?PDe=
z%-uXIK+o6r;tt9?fVij2zMl=K8^goHyhP~c{Nk~>M16d!!#sbwx`@Cr$G`an1qH)e
zK$8%cVq#iWg#_K*-$wI$Dq2}(YwGG&5ktqG0pMJ5mX)=&X+Ndm$os(8K^+s(Jv}|I
zfmZY#iIgArU0G}kdk!e42ymO8jE|jIq9lF(@@1VGKz|Jf2M3iL#f^;(ot&H;C-dl3
z?2IyxbX543)0Zz_uz;+11KviLdN+)Ri7At0TZwa$x!bjfpzu!B!z85h{|@90)PXWT
zVQ{+T<M7as&dB(9VtZx`sso@-tDY6=XwI({6+7g>)lPsSCWZ1z0U$G&mDRsPsBbsY
z3|OQIX=&2pf`Y>u;Ekx3J&nb@_ZA3&5J-Zcdp^X@D9+2o)MjF8I<KAhD!AR^%cn12
z&Y>%pB)?q4c+>!x&VuPZah&aO6mUooMRnK;1RF$vA(6LyoPUx<k|x6E`5V{g5;yf7
z@$y!U{}N}LL?FHps_bA|ywiVTB@f;6zi;-`X4+2SG>9>R_P~l@BbK@{&|MqR)cs#m
zls5w|^~&!UWuXvJ7eAL9i;E(Oc1>O#^Mef;#1(m9!<O*jF+>pEaxBDnp{MaFNxdC-
z!_gYr$3BpjJKYk>9uA@(rg7D9cXqCRjw_zo0Y=ItOl(}h{;a5`kRNn++~8u8+3BxO
zpujPq?c}*59}eZA=aAJy;ms+U<m}XCck!Tzlg!cAsRRovdif5lj9>VLjT2Vhm*mF+
z^!z`qy>~p-{~!PTKDO+XnLR7(7}<LzA<4+*P-av%<rs17(IACLBr=MSb&QTaNyrSD
z$sT1(x}K-L*YCPre_a1v|9sp&KA&>D$7?*s{Uscvl2bp$u=G4IYbOx#K@$$o8X5f#
zm?0$PY#)X78x5i|Vlo&=sv&^Aud}Jhbc;pO(?K@msG;|>6u0d&CYW5^qRCA)&MZ4A
zbYT!Ke!Cz{hVVJ=IZA^RQyPJbRDR$=lKoJAy)gpKa?GCENiz8xegZ<D|201TdId2g
z<%$wXRcFO~WIZ7!P(dG!(;--eJV_MhKS#}(CVbS?TV%`GNB>xCp$tiGlw{FIT!b^~
zV2Rve;azA3KvxXUGar}<q~%8W-^YDTC~jNJJt@k|jtNmZ{D?>NbscxaF(37H%B0W(
zZ1#U5ls@1N#Hh5&6`o0egHJMf)*$^8ZG#xY(l<tC>}%xlvlokjNODr&jDBKe?9*tA
z5}#|=Zdc1oip9g!o3C0f9<j|l-bBX~B|$MsrO8OIfuHa$zeI1UH>v2g<)t51M%Rzw
zU}7ifcm~V^l9NKd;n0lJC++yC!5L7!aS**Dm!~FqaM047h=$0>%^I<Y_4FFJ-3{{_
zd)MIgN%<sy_5?DC@N1cw-KFE&H51HliTGL1KTh7VFN{<>t&uQkgtTvPEmKA{FXSD=
zYU#xDQU&mjAdX35osaUmd7RgpMM8a-MG+el!|r!q7TWM%ez^(LSuP)ta`g6bNM#JY
zACG|fkCiMBI`21baBQ}JCB{v&3WdnYDk~p((kr6dJ*rt4du0j3Wc;sCXNCBa(1T=8
z($E&v&E%ornsJ7BCOK1DWSa>zjSF@D>~6eeS3abS!)2i~)<P;ljVY-6dM&EgO?Gf-
zNZ$O?CE0bBL3!e8A4b#}HVh@DpWYRX$7jeX){PR;B23-@Mo8#eIG{Vs?7uG5L%565
z=&SmMW%KzwTC}=A!bmI1?R4(VxSWe$^Fz|@)R>gT&Qwt<g_HhNiNF0q-db!qU&hD*
zb|i-VDEjj~+REJ7E%`t$mP^M;@e@<I=cfPCWL~rO+&$^YSm762v=Fanq<c<%J$Ff<
zb=1FpT^n=!{CN%7M*R{gz~r`HDxitUw2eagpWBPt%iCr##GTiPpZt>{hC6%Z@K2j*
z+=aZ53*S98>$;<wtvM=yp(0)o5uxd#QO1WVDq`?{_CA*VC@=fn{Zcjgs#-j9AM1Kx
z`j=yKbuF(*xTp^iYjW~M2@AQl%gwkpv6KdYZR9UtxhZKgyx>U)Fk$w%BpE};8AGHj
z%#|<khBQ>L-E~9@x3-f3CPpCgNTf|)&+}q8{^LejN92pwnet4{(gXo!LCbdS9HBP)
z5*yyvVKx9he^4B+GzI$H@uS%Dy^;#2!?*>d`Ko&G`W#dm&d&VFN{gbUcJ(Z<=`rmT
zjhu?EI<a}b1WFY0c_VXnQ>0v<c#i)C)ogQzc(D9RMm5oONocE(JIc5RmFa0#7E!;@
zY(Kjs1r;mutT&?$(|{Pb4tCiTWCmM_fwuJYqo8wYUyxI9We35X@osvm&!<?|bw?K$
z7hZ6iU`l~S+VdALv<wWWVW08iOXwQL+=%X6nD}$^Mq&Yp*n5Eg87j5(V1huN;`#IE
z8rs_0XLs++u_9!e_~FVK-(Fo-C$QB)cp?yU8w-6!v7h0Ma-3?t^S-;ggXr<-Ca^eN
zbfzE(w#y4T_gGW^b_)GR8=_@IS()HJIDD;+bx3z*nXZm}**_H7Y<-}$^kB)SHHtwE
z2ks0vctdiZl-E=mz=se$c@l*ra>~>HIha!q^=quVIp;jx9(Td~IA~(+)kF3)nz*K@
ziKVR=ohQ9+Ivy{=V>ip}>cIv~zISg>OM;IN1&KvsFof8eIsk!$O*KT8Nbmp!P0YHA
z%KQao<quq5?tP=`mK2_ij_AJtrN#uSUB3MW`weR4eNnM7D91i==YUc{^xnqY3@9u*
zP$L983KNh7dKJT6VWZaFA2Vp5@rHnR4i`3|4#o%KmEa-b2tJZ`I}cL(^%~K<oaq3X
z0B{Vz;Ig#;5jvbINUb5M^KG!jBT)Q&qBm(pn!4~mL=%ebzj|c7q`u7L2+?&31GkZA
zNfCbusz@v-AgRYCETE*$i$<fr?*-Ne>X(c_m{}@x+skWWHRbNtu`yai3e8vM|F&<4
z_RCUC?7ZtC?(fUlM(!vZ&sH(T8u)UMm7EAS87y~5;nR-cmF#Z5tkowACh01ZJg+Up
z-HeUtj~+dEWDPM|3Pg%<87Ui9{ryWKT<6~nC6PKs)?l-`L1pZ7bBreS!2=5T%`~F|
zlqVP_Xg^QDBj>&m=o_zSm8>M*q$Xr}W&r<wcnx{8q5t<XpO8RCXH@ac*TlGGJq|@r
zfy5WtJD-v$5I@a*5)s(QdiaoiU&X1FOo~u3Xcs%MY7MWiv$>i0xe!w~*3hgAi=Gh~
z^Kyql{|22vXIyXa8)?Uxl%#=QkJHp-3x9~+hRD|YzUXrU34SeDGGRw=$Hpo4tINqX
z*QAyujxbdpjsDHf@3FH66Z7<cBDw9%G(2r4YCMY`9=-D(8ZnOe%__dkF`BYKi`w6L
zH(H9q{~@$UD$Mvu8%&EsR!2X7##T}nqZQ}o%8rB$S;g5kv9KlSozGR{BIYUS&#)20
ztG2!^<at$TEchp`QWlpiEiqs=z^m8_y}eK)M9-S6{O#f?TmJ#`yk<*4#1IKC_xY7S
z5_D)~1lz?Vxj)0WDaX$CYf%5&w-n01OfFj_fKWoa+udy>rzJ+_81=m^bgO=nj}A5u
z(3WF>)%vL9*-Y{aeX=kSoN)(TkGMc|ZjGkyB~R^$;<>$xf)C-jVFSjw3O9p&=!%>z
zw1&_nRhyfe^yH3$kSCFX4op0VsCl2~;))U5-b+z>My?3=ObW&SKEh&NG*%LB>tm_Y
zG7j^_OX<Q!Q6h|*Gmu(dyKsQYwn5^kplEh-IY7D4BsUqNHZ}WS=QgH@g-Uo!`nlRQ
zowng!_dA7vF1ilFBW7!HfK4q-0V-_ka6=r4O)byzYJ2O9?Ch_?xl`Z{PA`==>$$t1
zgh4D%^vLb~1radFfrEzjrkyxr(So-sPMN1Ubo<}K{>0wiO?ah1ZdK{TgU16;nP=I~
zzn2qy@^L6K{BR?ecoQIy!G82S3=?e10n4kAWMWtyGJcO4zu>)}v!bU?wWjiE`+icK
z!!P(v!}vjooVc~w>*nqrA=U{2^0-rnL=zCNPHqDQoyf?@=X(;8k{tl?;p@EdXo5fR
zfW&Q9Ov%<Qz)vC#ZJYWq#f2|m=|o^q)`f5GBnN})_tiQ6zVgeB?ISH?MI;w~n`^tK
zn6x9da%Z)YTcQaBJz`^lSldUte-=Cs&LXT8+Zuv50uu7E<Wg>$hK7co{!LHH@x)A>
z%+*irMEcCC#kJZgiiZ8g5?Efo16I|zuq5XhFAdlb&mTZUA)tQm_Q8)1DCp1w^oB^T
z0;fxUJV*ffqttp<$(odyyViQ(ic6e&QgHrpYF3md<>-2Q#SaBxSKsqKV?%WJ84*b=
z>mCLw>k216jT@DUW4Y;q{2vwl2&o^UdxS357vXe0*Ah8)eR#^d_00pl<#ky;UB(~!
z2{eE!vkzGiQrXT7ue0tUVLNsqZEEC`(vnyrkGh)d*3ykSEqBPyqMh5Rh(orid{7vq
z{GiSZ|GYlxU8CEz0n^?Q0vqvfN#u|_f<P3P1a?-Fu{^ik?yBABa;TESQi`}vFKQ0j
zdy+|X#hR~jL{4OzD29vu9lE~MFC;NCn~*&5adsppJLt=PYIIUT(VMD_)2nS+jK<iQ
ziv$6~iuQ6(B)qiP_^Q&`@86xFdY;!@0#j&iqkZ$!S;$3I-w@@JK<2t0#li?evTj2b
zH8sLtTvkTG<TUpw`(;zEdbk=Cei9_8@s2vFzmI+nsb&hBup*aa>cfic3jDox^sLyg
zPRgWeHshDHHDkLfg@r1%yIiCr$Q$VOARs$x#M#6xDT+Lu`Vo|xnTfkY!%7tU^VXn#
z$|3g~o-`B!xj#KqQ_~_>NQg7v8K)s1I?tgNXr78%W<^f#{Xah@5_^aH^YBiRd1Jae
zT5A*KJw(j9(RVXR>?<9`p0+G&Z;xX}A;hM^{4cjb6(m#Mha2oH&~bg=^Tqnr+DoZ4
z_mVmKio8Y!J?SQwV$Rjd@@zz`PNtClZ%EjVUL&heSN9PWk#uiNU3fzxwR_ooz*yJr
z+ZtEgyYi;}Q?mYEufJ$1r(}rR#C-q$hVKgmDx%;KX9ZIomK=EKfyDx8=>!m*l0yz1
zZ$G8R_iUb7;zA8;!7p{ct$kPtaoFhi3rmD4Bs7{Lq{?`!GWU^(^X+R|Oa9?g5Xj9^
zV#fAHTgW#3-nM~kbIhR{BoO{GMRuVxlh3_T)BKyR`QRswvBDG?f+yD7`%U$At0@-0
zkbSn1Unq$=*YBwogu6DILh`@gj1W1bQ;}|B>sk;IGD`wK<OVaYW2RRe8GQ;dq~^o@
z(@+qXAWqf0P``#yKvvIM=L2!ionhkQa^dqc;<t&<G=$I{2TL#SLfY3R;S0SpDz@EQ
z(h4CRRF{^1!!ouH<xqG?S+VHWl1OeyO$|$~zO}E@(Dy`?p{A*I`aoqW?02;T>t2`|
zabi(8+?_!oAhsb`Ff}UsRJa4ju3Oby<DysHya1JO+kh{wFAHx8Y_NYNVGBjERUD!v
zd%X|Pa8qIAp}|}IY(OjyO>dptilLJl;ef||@xw9Sg1&iP!v|t5#Ci)k4a;P|naMl&
zz$wHuk7UC-3DVrHUNic2F%t}62rE_XOTZ$NnB#)*M`iHFcMFU5_hAGK9V}q2rca5G
z4wp|3s{h|8Msyd){Oxz6A1w&J;!jvE+yz>NnMk_K*iROAd4=6y(2|sb$QPmknH6D^
zoWy(gjy(>Vjw80VAi25@QMoFmYmgJG{hsAH^4O>6iRLgYeh>tLJEj{<ClPJqml!Y9
zLBBc6*k0$$@Dub!mc)%asHTQHucU=Q+EXz|kAZ)6f~)r5t+W(kFv8wGM~Hi<26ksG
z2||*9VY*+l*hs2<qU=X0RY1qDt!GgT#vc;n<WAQXRtV5I%9^qoyUt$IALuQ;W0)bK
z3);+k^7D3fTv=IJTFq<Ce-4QurSb4Tw@&!?o93-S-s%TD)sFD7^Drn!7#*y@bAx~n
zjOqnFJ83^6U}5I_!YXm9>-j}htLYcKq!@=iT+?g{(g8|6b>tJ|Iq4)t59Dl}J4sT>
zc$u&^4`y%G-&p;(2V9*+E+YL>1P<JLO2Dc;(9<bStW>0r6?oww1jzxCb+@Vs+`JxC
zzE%ys0CeMsqb!yJkwA=E!kA1sFkY2nokM|s^q}{=fDQt48Vv{#EU*XO%0PJ8HI$gL
zi$y=AXgUc}weCbzjcagck#vYqEUL<^?`dCtP`;$McZO7o<T=CRW_FDbUiJTwOP3$$
zQ}#j9>>H@mmb8y7Pay(K1Dg7o2GBX6kV2<+?`wqqpm^J|sqyOk?$bMk6d%4*kJaBm
z!bx_MaB5)fUlmIdg8oC0CgUecUp6e0(+VSWtkMMsn0g(MGe;sT9^y_QV?^Fi&8Vlc
z=!P%SOfjN)bCSK3^a=UMc@m5moyHNSug^q3^tE6Bb|eHEx2kWBFk&9@M#srpYV#a*
zRg_zSe(-<``Jp_DtpK2v7+Ua*=1@TZl*&L(bh6!KPD{YqCZZiC_YA1s$nb}pHNB_*
zw#S|>qG1TwiKJ(*CY}Tl_|#hEju>jVAo9@syJzrpY~f;Ndxg0_6@Ghgi~Mp^Sn&J?
z(UR9@P?g;bG&N-+a_A5|G*weGGKg6>VlMGBL<~uhoZI{ECvLpwtDTIbL6Af=D0DLw
zZS-wWrX?m)wHe9AU>1bfE-`JD1iECvyQ@$+YdbkvvlP&x&%xv?;B$Ba0s6-^xRp|v
zWn+Z3`tCAhU;dPr%#A|~g$57qQW?d%C=-Z7pVn}0O@N0QQMP|DJ0xf@LOqE_W?I96
zUexV%h80{Sp0~d(Sfg*o%}evYRpzbI&~Z%k7RvWD67GW)nYQUY6Nm#|M!12B;-+7E
z+f9mX6|^3E9*Z?jXtU-Oy33Q)3iEpK7xU%UK~9n<pePI)P&$4B*t-H9y*iszOyabE
zUe(ste!TK}1#~@s#92{$axhr^3S+N<4p`}?1>rv~NMO@Y*C`0YcOg-5I;zFjCMY|w
z@25Ym&|MN4F1lL);Tvt_#L*ldVA3G#Hz~av&ZZX#pj4<FL+*v*uoaEAruSUAY}G{<
zXCis2<lV9Yl$pLRcMH={PfJOFd9)=_HDnk!FNfS=QomK|K4R2RGrmFQ*mERkCU)x?
zxmo(_%(yMD?{!asm(=6iOAKu01b=_nGvU+lgW}4^%m_1MMREXzWMYzGTG9ZeN0b9n
z93DalG`G7`&(=dIfmq;HUacjCtDrFV_$Jh>e#}2jnNHIWU<mvE@nMwms@sZ>;^z#<
zs*{drJv02MN|M!=OQaMk26*A|14SqPE0ov-Ap;KDNgpQlKhUawhbqykA$YwKQ<pIK
ztcgfaE%InG8owD`>B?4J1*=ijZ@%X#n0Z%_sY(u`NZv3<kmmiBvPa+Et;D*1=QBue
zaC$({GdEAUu9PjLWTFkFJ_YA?+)ShW+8ZfD^VGFtx{9Z*e}7pCP{snAZiIZYNp$Y^
zd5AOAhdKc2CJ?!uD(J3Vi*-uuf>d5tz+%BM{7Smu3XiVBZR*J>`qNRO578EQHsx$w
z@?tzfTNV2b8f0;}aolcQCOUu;+bnBz8k1N@N1>fHv{a2I&_f-wG?5>wPVlKR1Q&>L
zs(Itzz(F$Xt72^{#oR}7dSq8<qFQ)E8|}+`-9GlpiYq(Tsc_+Y?HZE-XM#GgOD~u}
zC8q3tnc3JYTpO!1Td3kYD$Y#i#Km;KdZYdhI>5husF~EQJjqp>Kseev!M~+7`DHIw
z(BKHR<LQZ|G}paqtdN;csGU=#l?9p*6FVCmkuO@a`Q*>kW+k%u;)#vNr8P<b4PV#W
z@5!OY(~Ds>OD^l1;S<qB3dw)}9kyHq7$hKaxQ-@6$?C!}a=7P^X(pS@eL=^HbqM=G
zPVCuj?7?!5P3aI;TSx8`jlOcFfx3GlRH#ne{D^+(%tT3@Vlb9NegPLjet~@I8HWg2
zh4t}GpYct_6%|eg3Z-AZO>w$&hB}_?7*?{f*LtzLG^Q{;!l!;9P56|u>s}!mpG&Ue
z=+Dr}b#U_*YrDfP{*`TdfdlFU=k?WY<q~zcy1?R5L4N5TxFolDCnf&3_X?~QGyUvR
z{`|WFO{2hjYw8XD+Cd}>>TFinqu2-^MEBC3eygXz(oX6lf9?w|NbkOUXQ0VPHF?`0
zJyWc6C(!hOi}FvE|Ahl@mqeqR3McCM>wn<?sFPX=0_6g+r?(QUkCg187a(^`qyI5<
zUZ?0}{O7!PBTxZ2p^IFD2dSBhD<d~o-TK(OHPkm<_xc1ugREnWlT(@SRkZWxNSSJU
zsZ?oqkeTVdN5S0<S7|hm-3w$kTX7`^lLBTGm0rF5q^RLK@(Ore!A&7Xs@5*>o=fll
z{%gI96GVwjDJQLdOcW8|3DxPf3b83qrnZfVa|q$d&t?Eiw=Gd%mr$2B3cKCcbdS!~
zKVXY9lq=1y@tX;{s@!@*7PA}$TiM)->oa;Lv>(2XeU(0H%9X%Y!DQOVokm~_<&7U$
zy7Z_kz}vF@h9G1sEzHfM`sSdd9CGJE${K{GtRM`032A%V-{EOZvqZoHd14Nna{%R_
ztxbw2Tq~_$0-QBbojWNeCTn3~p=O}dl0h4nAEbHx`VR@NV5@7dza;m_dXr-#CEoB@
z-Ye(yJF%bgsolW@1T(Nht-FQuXIBL4xban@JA2j$>r!JfgbmVg1hS4x5Hqn{m#Eue
zoDN1G_lwGqZLq5;(Vx5S>A{)CT1TUx;LmxQxaPl6=WH(`CnuN2<-m!W|J<Krr@b}Z
z_}0z^v7w$cA754s6}6_E_Vv)#Zj-r{-Qt@zY7VHz2?2h^=_S6b!mdSwoG`VC$5QCc
z%aB1C3>u_nQa!7)G~=(TYpO=USa;<!+q<6w5+F2oF!s(@{CfNHv?n!b-+acq;`am4
zH<R4n)`h6T5u|gJsZkF3uDiF;3L9+dkS!{vubhL|zC#g47W+DoN&_h&o4jvsBscu|
z^YAlL8xr&#bN=_Y4J~5xCY{9*h(FjEp*rH<7cW97Xw2<rTnR1DV8%2o@eg{el0a6H
z4df&I?5sAfN-f;e$cQL^I=TFeu~#TvuxqWHzglLn%x>v88Q(~Md(zjH5*iwsq{K>J
zR<VR>J3bgP3q(IGD>!W57oEudOwJd6QD=1VCx_YP2hHRaT8v~4l7lNSicu2NIfSPd
zT`%W+9?ycc(FAFQ;IyZy`G82oC)QV#w&dh?R~w2%KUg+cj@Fa$%1&O1sC=8@as0Vb
zK+x~+Wo<G8(kz!)W+Fat)z;ileiL%wtvW`1Q$-BhPqSJI^+D+8ie&p8g+d&+;Q12_
zzn+I3DDt!N;O{`FF6iw@M99mH+pZAq8LVB>z&9diehPPgG2DHU-SV-~qGao%+Oa%U
z=^@S}!d%awc?Wd33j^oE05n#F*?IfY2xa<e2qVj3slTV>bFlkEIC}?ugYLo=dGfv~
zhzbm71dHmLHgMdwJUewl!r)@8#WK8gSEDcg$)cqob=awqhpL|Wsot0IHSC5QAC%A6
zz9?oXAG|+VROsR5b=~Q14Rw;@<+JEHC>hN_z@PqEk{>BX`k`}{JM<_taku>a1%Vul
zCqnCHJ2Nm)sUUWipmOU=@V{C4$u7k~nYwcBb6C*JH_;%YMmPZ8K9pQ%lN&qwbjw|5
z^92}@{j7V=iSDY3w}CvfrL$>!FZb=zn}x*+rBD)57YfIYWG-{l4<Dd$vh~)$AaVB>
zFPONA>cL!QwzkVqkU9Oq<WKb%ua$VW@}W=>EMLUc`3R`JJskQzeiCp-iloCM3`>T^
z{X>UJGw;eSUfNY5$}WEnchrGfu3>twN8&4>o5jPv^`6ssH$pPZN>Y){N?UmTA<6M*
z>*^WbK*>Rc-Y0vq486hY%7(MP+Z-fisyQh?&7?8QcOj`@$m3$TvOo<umzj0t4Z1lu
z7+o|hKVi;1+M)Dynd~0<{{8z9#ko-+wYs}0uMGF3X@9qn)YGicA7|kYp`E{jUk+u5
zTn!po;&1u{t@Ks5g^i63eHlU--NEH{wou&A^td=8$1{JFCv870>GNu;?EIf=#Blz>
z&(6`LFEpEn92x|+i)I)el*oL%uq`4iEUcHDe8J3&wEk-{+?ScW*n{p4$I4rl^6GNt
zGpvE)lQjn~XN<C6%R}2$hP8L*G6S>jk3Qmd<(-VWYBlJ~rlx5pDtz^}>E!*2Z?Z<$
zIqij<9H&|IK4Vh|!_`~KkvoO%cb|?u@_FkU@0R?i8^F`-jeYM)*?5@jm|Ue~<ws%|
zO^@+JsE<(L<_S5=L37LOB#g+RBmOJkD<?-|K0K#B%*mlatQvyQMEYH+eIp0ZK;IGR
zZg*U1WcmNBD_7in^XAQ&!iWBKwr5RDQi6rZ1d^g@@YiPoC{;4i7z$}=X(_iJ-ecf$
zKmy;YDNOOm`TrEe6i+WIc7}-u)e#JYg%8S5JL^|IeX+R=pxPlvsmFBT1<~(W56Nmq
zf3rAq8dYkCyrqy5TkF-jn(bL>^>Z^`5@B6kwWt#Lr4jv=5m)h<)t2d*F;8EpuZPFt
zFC@RNq;_of#>+Ip;#BopD^Eo~lq3A3wvXueU`M309B}5))sa&EDng~<hX1WcxE=oV
zn#fAL^zks?i^89bc_?y+5}>ih6?@_am9SA(G(@=>VXI4o1d4k1kkmpX8^Oaju^`SC
zCq25qdJ!!9D{ldc-e|7^1=Apikvi3kYO7JiiMlij@)PkB*J?OvH9(Mzi(4R8<aTJ;
zliZlhFEQqnJ(wa7Rw>9P#dkbXETSQ~lJaQ!)TvW1;AZ<E=bj#fy%IE_nI3()^%%P`
zwkFVY2KKU@_u~rPSzCC**%37cosc-zd_e;l)Yu5?3Xl_SB9EDVY~6&R<>B~Eud&+R
zHnvAm^Pa*xq!s7V3yNYHcqD$<3n)S1AP~WAeq|qFya>6NV`O}6w>ku<TDKYorxWs0
z%GWlIb~!(p$SDqBJ5+SJ^r+%0<W8sG-_sQRaptM@R=)j)Q&K(s=+fdkTnb_ieRD?v
zMKcNNLZZ#JI&va^N}_S|qHV~+Gvdn(TD{ZjFK`OL$d!?50irTqiRo~LIu6)sP|Lt8
zuQfzHGt3~?6f~eXLxOsR+U3%reCGyb4K@s;Zsq;0imIw(#9k!~Gic92F(8d{|JfgE
zUWa`^j94ae!PFF0K9mrZE`2@t0l9(JTudUv;<?JMX2X4HWHg9*j+TswMu$DnF=&Rm
zaWeSlymS7N5Bu9tSX{4D@wdh^^&W@&yaW2+>*A;xD_#oU^~}t=+kE*?g$zP!vJ>F=
zg>xFbu7r9zkXpz4dwQf&aI4x&)gfyI&z{-!4QxO_vl~teK>PAo-3y$%57n8AkZW^E
zR0Y;b{NCN#N8&bbYQ<POt5HQ`QwIQQyk%=@dL50ej0o5Qmuupqr?(^2!FHpte@0i`
z_t!eMTus~BiQTMYZ^pgvon+s+<Q3aj7YduGVH`MxQ?O-fA>vhuG)bSvEtUrWR)Vf$
zeqJ6bJ|91~QGI>Bbm&|ypcTK(&bG|nnEH0tgkWxIcU;L~-ND;i99FzqX8lP?Nm38b
zGwL#dj1P3nHek?EzW$OJMXyBl<;$1n0AJ}ZHv0iiL)8bvo~&TK{R?)}K6W=%)vn)f
z5u{oRl3kGlBoisMuIV7e{5RM!Jp9f95bCad=X`Rp%Z)mM_2CCAuf*rC#3Uh@BneyO
zCT_9#X<cOG_-m-ZSZ6ZNbV19UtBrqz<WK&6u14D$ZCEt>@t?vZ%E9wq{35H1jq{w2
zeywep3=-{gP>e|g@UlXVj(nhc=c|Su=gNC%LXPGrTQ@d*W#qn&tbTeypsdlaa`5V-
z)RXeA{uC$~*k@wOX4&>HD|An&Q2xGK79CC2bjck{Y{C!YhYb>zIt!Cc&bDPnCl^^?
zc9gQPZyHVd2sQ@~+-&aC7NyAK1lbLEwP%1Vx_aXZbi3f#5gZltFxvq238jj>zi<iS
z@duCLexOf>0rgVt_omr0$XU<J$_TC*b}Y+b2ohJi@o@j#K-AS2TrSR9rU`{z-;Ey0
zXgDbF7^eF%WAz1oDG`R1Ay6j)`J=tI#CO7}zFo|s9j}>&AP5%j)-l4y{UZ|1;yEO#
zjPhE1a^S90B^h9%-jRM?r%F)tOw?;3)95)pq})Ag+ty@SJGQ!*7@>cG<;-5-YA}TK
z)l?W<jvD}&d<h7=;|FnMu+#*QJ2wq)nD*cur(B<}kU&-~FOE13lUgq=DJERi)SJi8
zVU-OIc%fIi{O1p>Fk6f%!RgkHp>26#UNY&YMC6`;`(@Fqg*1=eKg$yVV$j)h=R5#y
z_b?Uu1aQ6xR2845sBsXnkKCclXd<j<7_{p&=zBnu3*G=(dHLAGH5}MBK`P#=uYd<i
zG}Pg-%#?PG56`^!sHEg)LxGdieXeVttuN)9^$aqJ#L75QS$XT}QUAV=ntToIz7q{_
z^2ty8w+$Sfcc~-SvkKk+XyWy)v1s6s1EiuTZNdA@BZwhS=I`&fx3_1;+;&O41)DXA
z6;UH1mZqWMfR<oY+IBZMZhit0MiY_~7>{Y*234nP%6B<__6})Zy$GZSnb|NqNamRR
zai(4&AbaQ}ypsk=CzoP>!|LM2Y1kXmBfV#)N+G%HKkiBAg`z@c)$(axn6J&5SavQy
zSe{x@vOPy-)eZ(C05$;u!+`9<`$=@}+Ep;b*{5mmJ102S02OuwnLlXz%T?l~g3p|~
zOL}n4F7lB=qMpD-Gc!Ia5}0R^WqVdOH6uXFB!VG53BaV*hgWmht)D&*?EBBK+W1&#
z7BWeKg#WnG#Xc_@FKv=8$QiIEMe8nDKJ+UMINXKuV@6VhDl6kVBP-iLK*=HVwQM($
zN8E_yocg2nfvU|7fY!HcSi1d}-n9mxu3jC#Fj|wZFpl0rq;7nBPnObgQwUq%8{)C8
zVbo^<QP1Wr!X+<9>K02_Dno3H0efr2Kq(QuT@3;0`?0d5E|+g#OfRTI2s$c~HP~2@
zSrRGyMO2XEt((8Sd4@8ksy63YDg=dNQY3sooGhBmaJa1*5)vZ%=Gr$xf}2H?8=M5w
z6Ve3mT{<90u^(>NG~aow!2{4d1PdVxrq@@xJTOj)LK9&#fi|x5jz+z6WjA2!LPr`M
z8`h)t%_yr4g*?Aj^J@$e0X&t;O>U{$aq~jRT*v+_^J#ghDG2-*(NeU9AyZt#{|C){
z_REM@+{o%I4?JBao3nHPc`%`ltWb+7qE+l5BJ*jm9Y>BFdGO-J+ha~kN7*ruaXk$g
zSI1KEh?dc1Uj-J@T>u4t1fKkzK~ILj(Nk={E~FRycD*dYG2bn~wBnqY6z3+OGEK>f
zbrP_eq!++YHaavfS(TimW@a8Ot2d$>W&H8sbopdX=xPhtSfODnAbx;5PQ-Z(3=9zG
zRoJXo4^~w6s@K28bkKug$&>1;D0`giKs790zBZ>qwe@AJCOmyacrI9i9PIh1hbszU
z<2}o*HvaXg@7pCG8Dw`qlDTRUfgW6XdI-7;G*7}H##Mr33Pg|8LB0==KMs{^t<Igq
znlwz4<oNq9B?>^e%sjrmLyLTg_<3u=Pw7`bjRveZj$^Hi|Es#J4EutHcfb~g<7QZq
zsbtgGqrJ*VvgTS{>&4@`!6)VAcfJzIScq-Zo3eq6h?tg*p=>ND{8jY*hJJ6-Dz{87
zed!9MWAFM%7)(Q_lQHpQF2t=v^NAA3LC0Wq!SQN6ScMyeVSWJt0hc<OvISCXObik6
z3!sj?XV1uCJqDZem9mLn0K6jBXNi4hUERZoF>u!tLf2M)ufGlNKEdLDDy*oSJ>kWQ
zi?f%FO!3t%LPD$z@14t!PM#-No#*FaOyjH7et=t?Hsvnry9{WVb$N61aIH}}OyIfn
zD)9%NscaYzpXC^H^DR^8fTb|9$-w*{rfV8NaY+9%_!yZ}ul=w}<I30!NPl+pywV{b
zm|%&5(lx!d+I1fCPI9el9`mGxVM<LVZbBS13p0YkY{1^yfBxfW_6w2{R*U10R=K4)
z{&TWWr<jcH_Y(6r*CHi$z6J)ke72?0D1xY2F+bZ;a?}FynXZ1~CK<1dtka_+CgUPz
z;t5gUSrXkgi@EbGI+^*8<sZi3{3xBs5*8a2FEhff>Pe)phO-hJmvxbxlp;azEa6Sx
zpxg^q0OGe8UYZl~=D<>n-%~{Mk3`)&E^z%G=XRa6J^3k6v3;touQK+EFx}!NU2hOH
z2tb6;?2R5-EW~#T26it{zs!(uhP6SG<h(zim-9s93D8im0FaK~6Obif+`I*25T%qz
ze%Dus6<aEuNM;C8z6q*<)rAYbk6*kv-5c`O{(||I0wVcy^-df46Sdnz8guvet;Zd#
zT(ACl&pqxTK1R;Iry!!r0P!_8cAB~v18U}YgGeMVwz7W3UOuNR0Dsy14jddsVGxO<
zk=dvW*W!p_XATKFEc;wg98~d&O76wbj6fCzHszEeItq6xAD2~$R=Tyj&l9TBISUFO
zmBO>6&D;#c5zX|jz1%GED_&>(5qV-|vwm2huD$(u;}nw??)m8LwSC2$<lA2E?saN>
zd`?G5KHZ6`iN@Ij(7d@rfa4rUQy~fOUSZJz`t7$~t&iX7n_%1zK2&qTA0&xO9F*%_
z>wC)19`z*?XRA#T!*-s{Sqkz1+JBt+K6w!Y$8wb^YObW<>7r=XcYAI@{fPc&Qrvac
za5a7vd+*r3(b7GfF?mIdypwU_7aUH`EU~ifFJJ%eT_{fok!G?S*q9Hv{*V9yRyU(*
zRfX>tWE|J)+He2yj|p-+Za1W&#j%zhuNcv62UFc_W0iM^@_r>aC_qnD!@jT02NR*D
zBZa3t^kfu_>(!%@=9;@w@j<M?5E&$?NN~eP@kg^-$_`9M?SEE#AsF+Uxliik$)^LC
z{KA$Z)-P3)hM%d&MO;Dcrpr4Rm!`SaQQ$w|%KL1rttD3e3kHP*Rj*y;eG>%uU-@D1
z2t?uXR~~dHz1s`zBVpss!v1E(wqtvOo;52CR9lJ3z<Ot)Q3yfd;Ya|OxE=203fI||
zQP%tNIKUPE72x?~b;Ddo(Sdr+P65I|o@Za$xw|Ycz1w=W`!??e{;MF#dRS5<s9U&5
z-vkb2Px4siq`9XJ_|-R1Ao!~I5H{Gd2hYyy?Ci{DrsE`ZYz(cvat7>uNtyY2)OAJ|
zHXPNBDWaz0ZjThwCRS3CR7TdHE>jKp3_ApUM299kFdL3>ssH3@PRU>9Cz2~L(%?ou
zMLGV`w-#Cc{=VCAGmsh644t|5nj}(vvZGi_kL=yNN{+b5b)MWp@yV_SeVzO>{)neQ
z(<kA4Nh0*lPBBB!;3`w^#bx_rf9mr^Z*dfjEcm^JYS3|w>poN;?LRd!;9CQ3|9S!8
zY{PIbGeDW}AV60gU0TPzFOe@6jh8_wjC#myU%Y+$aNgJ_ztEht)+)UKQ-5&rc=TD%
zy>i%7-fX@wMM0XtZ?$ub3(x)p<bpBo{297L?~r0b+!|*IMLS(c{WTFdU=T<)Dql+7
z1Ahv4#*uO`bz#+4Jr$k;mJayxe_#xBsd^pf1*k=K818JLd=fPwr{D8%&KO7zr`|4P
zB;hW$*;^BGj#YrYMSJMDS&c5M*EIc^&D==??9N6u@rlIB%(t*PYf)8=g_LlOl~@9D
zpv<cgFy9dHCg<(S{Nd4Fc7*>{s4F{dj0rSBDvXUyncXIMfU{)G8ohc4xe}vp2(Zhc
zJEuBvKkYt_qyX+%)L&KF`1d(ZACi30E9<Lo)*M3}b7CK|*-QxxmvOti<NNPn=t@^`
z$u$u;GDN_DF~`?)q`IN*#&n#Z!3?S)Ww8EKn>$&>agF0xJ%W`m(<~_)(=(qvqhs$j
z>OJ#5z-dos*l5EMHtyX;cCdaI-PyqcbRi5|BCX2lYXIhaHA{4?<MMZvnRKl)4}nI}
z1Gf<q<a0<f$>ZeUo#7fx>+BLi8*Tkd^+RI~>8|-)=QxR@U_fkPNrtPU_S~&HksBi3
zj*d5~@<r7H@^X^N*O{>WxZHT;F&w%?2uKd)I_$~PRHIn05bHdhKT!G3nH}J8ln2+`
zw9I}YWpFT1xYBD+FsD>UVlo7fJ~+1NjmVBH)FCT)sV1Ze!aD3KAFnH%kleIzaDM%z
z&?9SGoIlYxu~HudXE6U}Nh(g$%1Lj()?>T=)sa(JXvHpyie%l|n;Ot`8u`3xqou``
zksXHLHc@9DiGzwe%8s!Nd`H4eFh)!uhoo`>$22;-<e3#<tg)QPAP;8d?|e};_nOhr
zD&>!W<2kFt;}ob+E<vpUoCyORc<7T5LiJrQ8O-t&@C^b%KM7f`fr{Q7P2>eK-rCn&
z6ac0pOg`X*=k8%*rA_adXnT{(P;rXLrh21A7#`5^yRN(@A`gJ{-0q!oAh=s(^H-#z
zxA~cLB^=fnduMbubNUu+nI)FtZ-DBNoT;%T^<?1gZ@o8U(W|@oTMXSkir5n3u|cm7
zl0Y2zcB9Ui)R!3>HDd6cV%>o-{MwGrhgh47n;*xI4DoRZ{-(e;!?{A-C7fzt8}iwg
z@0>zv@~eH?eRE)0RJ>jh1MwVX3@)xRSJttAD~iUdD)oBki(Mk&4Khe)nP^kwJern*
zWo9>-`GVj8tUlz}(Oxy28u>D+A_sSyiEOHVd3o8SUf`fL`coWK87ZD{Yj7PHcQYde
zM@>&*#7rN|a<5Hxfw>QlWJ)U!PtavcvL`TlUu29P#l0tQHm@SDFLHZLvS&n*`iZR`
zw%i9S38BuFR;yPMxsGVau+h!|XJ+8tbMEoT@FV@CPtw5cuUw5W>m*BUWiu4Yr^~)G
z{@VH_&1B8bGt8!{n9?+yiq1XwPgIX3)6m;@!7e-NBMgwKBv8v`kjHe66L`l$CXJI=
zb+E{fN9?vDT7$2khIv9qN1s&TSVZ<<wxdY?oSadpFyoPOGr&D{<%?1jZ9O7&E~m#5
z*bIkc#Us!UD8jAkGB$A^e8Ee;2+*=?NBZ5luq*^C(n*JNnmWZT`$G8>K+Za4@f!|M
zQhkW~PBEIsOVt%IC%Ajwg>2QcvTdq1ZKn!mhC5=+81?7)Cd(@Z%N5^BRo_Z+P<30J
z-YqWXFo>+^dFd{{F8pDk2nPo~X){KjH|o2Xq=(_cl%8SXFOY=8-Cj!((l@O$6iB0p
zF&T67fgp2p1|(IIs`Uvz>yrFI@6Gs2=D(j66u5BideK3kXg5wn0Mi+!^UERgVG6U^
z&F%MEL1){T2k1-vhhWCP=G|XK8NBiD;Mu;@Pj?%i&~|aGKS-;b8Il`f*kE!Ba~|P-
z=EX@e_4Q13?R(yMOIgnm{&CrV*;`-KFj|~r{R#<&(#OM{&AvE93Vl*_HE-mt_Or#7
zTgR0rH|0W6Ki1SPPJHi5|96rtEH>Al<2m`I+E@KeUWJZ{p8tT2*m7VJ4gFeM<4n0I
z>+m$g?MOszP0-(j5^+9d;&+Tg`e7wzY&B%ty=`({KfdQr{NBHVYGYEWUS0W2Pq8lI
zll(8R4U=Bk5i$#u)mM$h%j##$IlTxj-#Cu$8y!A4b7}IS2Rn6SZBtyoHcGVN<o^!e
zC8?P65^+3<x_U5n>F#4ZS)%dTg+AI~gEkpC=?AqR(j+(>0^FO-CU-mue*<B$%<{vC
z=!_IIh9?L|uPUZUli_;VaWQIm2o6L_F+bk6dfoG6q?3K)_gj(j|Dsk3w;~jf55rGL
zBxy7vm%Vdd71q&Xj$TaV<Cu96p`K+I^yCJmhq(`FMDoqM+iE@`|L1qH!yR0w4-dE-
zvKUBb9>;;bPF>vSK_Y|@kJt4|iWC&S=E3X#io$EUEBWsgHg;&GaC|5vfZ+*_6JCYB
z_~}=IF58iE6`C7U#UAXOSxadrIdWf^HV&aK1m6WvKK!J<AslJNM}Lm>NW{%mt(!?*
zZlK~sRNTX)ICMR}@BxL&g*wvuFmZ8%d|%`2MH_^h=il8lR*#@t|7jvCz5OP`&Clr~
zg~Wfd0wt7#8uzKz2G&}ymfa|zA$0j??dTx%;~)RVJ*PHB+EMn07wLlNv6cNpH->kN
z24MIph4V06T^l-wB4b~~L)t!>g8e+h!1Hk#RR6m0zl&bf+!Ay=I@85vIasncI<3&<
z<40Lc!O6P5C@Oqll6lb6H)Pc3pa1mmddoTP5aj>w)-IIe8jff!_hYZ3?hUv%J#F$S
zmA((K5T|o`7nSqAoib<OE`K#s5_7~Qj=<m{yOyTr7GQt>-^JgT86#oK=Hp*?pv!c1
zAyJm_mr@oIXI+RyY4j{w#>smIw!XLG!YDM`o2$<a{O`$lJ@+~{qOu%yfrPoL`59ea
zXiQ?^4{G;z<bzz+I4bo+N!-8p;OZLY1=Tk50YUv7E3G*@MH{-o`rife-M*cS`0`){
z?T`Qa9RK_F|Ic4OX?*{K^2@f&XuU71iy(-QpRSdki<6(L66S_0d_kn8q-DjW(Be`u
zm!xHs(9%lM(x;`Ql%%AlZ}_JFzy5%?kBj@2;Q#MGa98rZ1pEQSK-X9YI^L207lnhH
A+yDRo

literal 0
HcmV?d00001

diff --git a/docs_new/logos/hf.png b/docs_new/logos/hf.png
new file mode 100644
index 0000000000000000000000000000000000000000..9ba7a6f25bf89bd94fa1ba973197e0ede96d3f70
GIT binary patch
literal 91529
zcmeEthdbMQAGS`4qCHk?cO0u$)u`EWR7-7%)DEX+W7OV7BP~UHdW<3zMP*3T9+3*F
z!ybtl4X0Mj8bOTr=RA)-@Bi?=uB%to)l0tf{@kDYzQ51SO%2boUu0)tVL5y6?jM#c
zEUXDEEXP^cP6I#jxp8bC`0I?{-3I|IEc{mw{~Wt#d1VFo;jsWqLp_$VF5x-gA1B>!
zo7`q$DNp6tg`Q$L_E7ZRAGfVTk1gz-eVMZnHWT3#(k=T}!i5KH_$xi%KR>(X?TA^>
z5MmYlu<7>NfwSyvv}nnspeZN^n`rg@Ywkitxz+g%^<nqd>Z)m-R^E@hlpJSTXEJL!
z-PF#gKKK;s&`^8rM0NSGZ*CtaJFmaJ&311sIPYM2f-$&b=6zCGJ0xU6du`8JC4|_;
z**|_DI!<KP%PRxp{QqD7Pp`nO<!_uU>C)Y6*Ltd>Kh3)b+0|xx|0b^x{-R7_sc9f;
zFQGrWd$K91%yxdit$T22-o4__Wb{F8@Q&}vqlLpK405tK8lJ1<f3Wg}o+z%WAnz{P
zvE#72a|r{}qMmRu)^i-6u3>K5`P5yNUCQQj()Dh_slW+eDBt>UtVh3wg@tZ;Zpp2)
zWP0@48Mh3L#nBpE`q*-yQ$R1J*LC-Ut`0%2Zaj;4=V#Eh@tut?QCSCKkDeb0*iFv-
zUkCYp<&`m*^|X9Y-ZT%mr1+&;1mR`P&4H@6K|;yN_D+Ux)A(P0l$f&h8Snj(U@u7I
zSBk^DQVH|GsAY9}!o5G}@RtAg%c0!*^Dyt*k?VTBR-zf;t=~tl>i?Lmw|yLe-B!9M
z9`*GhGCg3TI4|EntW-R-G%u!<WtiXQYzf~$C+O>X$*<p}9q)W%7P|dqo3MJ{(M(xr
znLb%<z+LVF#+OkzCkPRY9Tr+`FQsQg;#x9y4|%zezQ-}AAvTF-YW@6tdALo$X>7}F
zZpx~DLL_ADm73E;q*xSo-Ftq&B;rL(=ej#1PxFoU0`WJ8{AJ?ofkQ@%wAr#<BzDVt
zW8%$0Q_@60uA|fL{>6ZdV@ESX6}Yj48`;jy$GV+kr%#6>^K$YS;*@wZcy4rBlWDlv
zDYOC0#S@m?aP6GVzkJ_1s}efnhu1$W^+!f16T4cR;e<{n+)Y0+m_boJTfa9We|%y=
zdpXRk^FXYOxWAJ%;TW0xiLCF4?8E&awy!fD85sDlG8r@CW->`a-EV#kgyp06O^1lt
z$|`^VYyvoGwh|i<5HYRwMh#=dX2$FzcWKONAW7Rz)4{#bcHv^k#I=3GMap*6Ze)0b
ze`1E@@4xT#<3=Ni>+UALW-+G3PNLh+0nB8vtY39-+jE+?q#`xL%Z@Ox_Gh5S(d1bi
z{xCmgeb+L1zNmyM5#AdyC=1V)we26ZG8lYR_?2v@AW*aYa1H!*E=uq1$d(J0I;<e3
z5hyiUALKoyMB6VXJPv(vmC}YS>)(@<l$=_e+n9^$|5i_|-VtrvtMH4+)ghyk_GW&L
z_Tk2Ng9vFs0Y4rGaL|iTe<!Ua*_x`=Oq<HQ3Lxru-W&JOM^?W5bx=Ie(I?EM4L)aH
zt?(D<<n(F6$?AQ&)8KZh-ner&?Z(Pk36H6r8}@Q>1J}atH)0lBTb~}Mh7!8G=@~(J
zJhQ4zZnfG`y9J{`*<JK<4X39~N&AvPgtGo>wrASHDY{9B_AlN2p_ry>(Zam)Eghx5
z$p;Pl#30Af8!MX+TIQC&bqm~X6t(MLRy^61oIdI)L=UX?->JRw4~~&1KN2c*w1T&?
zEl{TeEiC=#*F4SMyv{z^n*7bP%1`6BTDB_{mDknxq7C>Sk?qSz?YCu32J3aQ)fmj;
zV@tW(5g%T?di80FuWu)a)~GjbRc)9eygDkNtwV6yLH}fKP4(reoGB9065mv;-99j(
z$SQ7n{Zm;#6r-@u{AuKcyW+F&=m{?$jCaiYk)yD&%U6*dRU9xCjN|<UBT_=miak)O
z=*V9Tr*juk^Ofl*n^sl1e}7%wGJGDJER=B-cCe6N!kW8dOVt=@>8YGo#nzfg-;z5%
zXuz;AM`IDy(K=^?_1_ECw5N)7Cm;|3u$Jz(cuL~h>VfWFbZ6=?G3KCv?ywUTBnJOl
z!i@3cdKuI(VP<z~Vs817v?qG5_*QLmo&a}3QsU&uc`Q+Yz}33E++Ag1!FjZRMz<6>
z1E18yZ$7A;GZ7V3?Jh|xn*Pq8Pb+xS+!^pB8y!cuPCnsWf_ktA@q98*E0qc4tCV@R
zu`;MoLWy8mknr8o;Se2Ag*UNVVA;a@duq?|3h2g6RTp#y>|Abd_qK~n-VMC%zf-|a
zj5j#;LAR`Kr*@n1J>P%zpoeWZBkO_vZig|L&-ka-bSK~BocHCM%M>OZ@lx<;gq2U7
z1wAh3F-Lh%tGzq+md_!dU=nzd$9RNaqtvpq@5y!8+5#+(a7h)r#phsi@_n6CdSxFt
z-}hpfs?F+rBWtDTtjzbpFnB$CR8RxSfjg`pPni)ap)EiEP4dGDR<6*y!pjxjl(JA{
zwZ6VSutAyHkE$v8c{jIn!va1MtOv`tmZi=-o0`%V-|QLB%b+lCYV|OxC-yfkI;l5`
zW{o~qfN!@Rjn?_9*Xg$okzE)kpN?#k>94A`rz4~=Jw5ug=?l#Tik1|_tq;ZKedxHN
z_be&RWh5x~^{NeXNZQApXwj#|89_fkLL^T$8mtI)%i3=|0a56Y=jMy^gTRmE>Imod
zWyUscL1y|7)Et(ZaZaXUdDL{U2(!C#pu6OC7u|cCQB@r8(nzY^ifa8vV}RDYW-EUL
zp38DK*a&KLnAy;J${a|XQOsS}+6FU>(T)8Ndg@aTu7xi}6_<DAb)=X@?H$d7<(k0U
zar5K88?qb6X}f>VJNXKXt=HB@5$3<If3P`KGDzJ2!OP2pPY)BRn_<s~2}!nEYBfIm
zpjEGk?;m5g)@DEG*}@L`V|L73M{9h=w3!PPAzWNsDP*sza=bQi>p+O2YOi)t)wOLK
zS*^#x!C|JF$L+-YvQqW4uC7k9#WOrQdVi(dwY`|Qzk$;YTCAdolB4ui#<ni*SI|Q)
zM6+^;iZeT0aT=TSwoyCgvW-Oe2ucHcZ4(8JiSAq5kJ4^68;P9hTqI?WCYGeNI6`xU
zAuso{8s~FlT#lyc#~ufLds6hI8X|Zuh<XWf?5z!Z1l&*KGGxsQ>v3+e;!!_d2`RLY
zdT&(!o8{V*e(?b`tUtZWxXK&6!kmfOpDE0+fBdCOxm#<kZ-0LuNt>aTzE#VSKRY87
z>)R&Gb?m1Vje=Qx&HhT?V^<E|o~pAm+u|F;9Ocd{!&V)L_R*ub9{1^J?l~rZgb4JJ
z`z<Rs9oMzyKcM3JhkF6!Z11GZ7>hW+?7Ezy+j>6jkkS9=t89V=1F?`8%jB@PubI{H
z49!R&P0sYcR6m1I=03O4^X|I(8s4Juk~&{mlHIA-?6R-<zpd_E+I>OcgXOpm;$*y+
zo0v;T&ESRFKxgOgk?^LdsHltttV|4AaJMN}2_#Ryzjg4;hLely(_;7DSod5=wEEN5
z33h=gaK68~iGzO|Uwzl8L%ZLQoeMqu)I9rhZ4ZUGNrS;5gRA}*$$8feA>BB=?}0jj
z=PcWo-++1SRwKjcb2ru<F0ph)6|1DR98E>s;K2p_@q6E94f6f`?Z%4JK6*aHPng9z
zTY1y8u!8c(X$h4V)z3Uzo35+Xz->2q!o64bCZl-O!;NPnmg;79VWZU!e0u%Hi9B29
z>76XKkCdcxv$L&&R$FjQJM;RqkUhCQYGamdh^0Z$PQ+$<|Gw6hnSi|Eoa%M*B9%*P
ztm!-M6o|D1msB)z-+R>ld9Gtd4&!C5Qv1kXeix61iBUM&@{xXbt!jES-HQ+TG-P$Z
zla|}f0anL+RL5RS<6)Oaae@Ug^G--JeZzYsCwRJw?=c5^ff|XEq7kE`)Zi}(XNAhn
zN~~Oc`can5mejJ$+{Z=!kYyZDm-nc~^%7^t`@gZj&}Jh)#7TkQE_b_+gxk&#+A81S
z&Jav@8*8%2u_vPmq<^OUYU{9d<_5oDoF&H*e3DeI?aM~<ibi%x)_9^vz1pHC$;(p|
z_${-jWTxJ&p=CP<8+)G|JwNrgR{ncWoXFBA<K~}h@tDSI<mHlr^h3sjC^r$@Vof?v
zGh>QT6yc)&<sdh=X<a&FH(qS_d#A+TVm~K{d6a-3-LI<W0w7v3tJ@nhb)m4<nF$5v
zJzUy)(BQgl5{|I@5B=KO>oQr>bbHmK)x+<bD{)?v&^1Z(8{rRMrV==4ACA}Ar;+OA
zv0PwK2}(0a$CJD9S$E!sOkd>&exe;6XXOQ37;HvaM$zB`Wfw182(z&`7g>~F$j{F|
z5Wcg&(ipW0^NkqYqk7q<wKffEx%g^g(<j;@Cr{^bEaHxVc<AAvm!p&XJxMYNZKJQz
zwuDS4q~tNbM-gSpCTXp!T>G|azs6TQ$ocV-=xKq$jB@S3^ShsVu(ghS+2E7@$^5+|
z;iI$S9QnQ``AEdaHnN%0$xU$Y$j-udzpshn{=MBL=3A~}%ppkaUX57v7EP1X&D<5D
zt|Ta&<>ZWfX<n<z`<c$b&Cr+=2m7lljgi~4y@ka6Hic+^`+(;gSu0PR^6}Nt!!`RC
z@jKqXcV%sc_3w{ZBzfi;IyCYN!M;qLFxyNq<^J8{DXB~L8P?PE*Rnzy?w(}v?r`et
z`{q-6lN6n8<5hecALq*b+<@k&qzhhqx>j)98k$$M=GL<3bq`c3ZGKJ0Y-*3lAS^FN
z_GeSqb>PI49|`({tpNc6PLZo^0>rJ9kj5$KDQ5lt$Ji4i|Cp>j2dN075qqnS2AH)u
zy5Y>8PG<n3DAeSW7Zhou;=g(^g@P>mkU0-MT1*zqJx7U!HAs{yo{uk0N3yooDLg_B
zDKe(@n=111qqZ@K?gJ$?^)~{-W;(J%GXn{VTO`-^l)V^z{8pN<w!+g^D`jQnn9YTX
z?)}0)<pVV?*ZYS(f73cgm+ZRi6ODavJ!!}kX*?FXSX}X8Z88WZIdfhi-0A4!pH#(y
z`yiB`S9r-_(BCM`z?Wq!Yg@Or9@-~YQ2qn!wNLFkoV0?y<%0Ak)J{%L&I?#UkfkQA
zvI=i1!C6+CwNAG49p6PRN6y518AkRR2xzNnY3-hqc}huGIX_aG-anHYu|y&6E&A$~
z1%H>=6{D;zRXjsFig`)Pqb^my+o-iQMo-!Y>NHkcD+w!<;O^O2bE{1|(^VF64;B74
zv$bl+XY%(SjTa-gu-#t&mw;S8$Ah&EPYC$rLs696NHMlRn+L2|qDmAUlK$wuzt6rC
z`3Vy$_+D}At1NCeqsn%Xawe9Nx4pg1bc)&-BsKz&-n11#WJD;aAX>eYsJntyVwCF4
zdq^hTG?Dycwb3!dGs$yRwLh%y;y8E03d3+U_jCVO=*ayO{~b$D!+XI`;r$*NcGiZu
z=h+1+Y$~R24Ry=%f@h0gqvDh1J3I8Dpwr&Lc$<h~Ene_SiA=A6V<-*fM@-#2Ecap}
zf`VHUY;cOC_Oiq*VbNhZoD#sTv-SO;m;&#<lwJ_7?zl-wbF-vZk$0<j*#$uKaKa~n
zDzVSsaX$(wHl*|M@@k;-YBGpY@uPm^ZB2QJn78e@#}5rw&>XqJ+psYIh;+%Q0=}2h
z_pOSGSeqvo!?i=LIGnT{u?1!fWvL!xw61T%Vs&}UZtB4*4(=Pan8g<7&XuP{xDJ}g
z%`Edxy^n01khLn&ag&aB-LUfN4~^2C97Relm%R4zjlQ4Cq0lawDx$gBTap_-zn`~1
zmlxfK+Z=82mbY;hcFaTZcY$tVWL&rt`ZxO~K`s0Brre=YO+M4G;(nn<LOszv@k!>_
zUl$_$KJgg+);W)do`M+Vdegz?FeU3jJ3~4z>HDWWOHp=Xwiz$>DLp9x*_$dHs-?U>
zV`hVSjIn_3ZJ<DJZ8SGGb0?s~CEM0NGTIRHvCACXgSd#*d<s`+!|h|`h*mp!H0$SR
zB;uR4-r6&@M?)(D0uF(%Iy*Wdr&)Eknho`DX%)qD@-CU0nYpua34h+mvf99pn3ODj
z1GR~k`C9X!jmEkCdMI~gC-tA$TS0QbT+D5k?CQUN=l!(_tmimgO@8voA~Wh+g=vCp
z2%e4;lXt#3z&#8iX73CZIe{BIa9sl%UNdMvQ|smjCqq{4W_1-BS4e4vg@t0JkwJI_
zK|#x&J0^%K`y>J$?TEV{J>O-Z1}}-y7=1kK!TF5uqHkTLC_j;Q9bF$A8_R${*@^;S
zi9!7NRrX-jJ$l-`Nl*_uTIFrS?--zi)<__R%gtBK;f$+P1PIYd3K>j#BU@qs?)Ob{
zpL`hcUD{?RLqh6*GvMKL?Yv9+-uO9Epp(b+#xyouT$tnpTXXr{yU`&n-7%<CaYzjP
zw$qLw!v;16NUy`mKqWl$lUmbinMt;46#lPRJc&B<duNK4=jOscUwA^#awx?1ZXJ|b
zUA%~{htTk*qh+W-2!_%(J3E_}s=FWhNSWY2^{IQd%54KrX?$2G^DK!{v_0l6D-)IU
zWN3zG#sD5U6?_l%Tp;KHqpsD{+mqNNc<27zquPGUK*?3mTd()bNVmcT$LI303tm~c
zi@uf*@iJ0;Iok{0*SQ>}F^UjaL_ZesIA*6LMk9C-WJB=v^INGIWFJE5<uAVu8I+CX
zW5ccTgbr$5i=O&J9gg1`JA*C=<XTms*#9=egC<l~x5cuD^-;&-X<HDHtsX7ShEV&p
zU8t-rRzc(QrP4;T?%z0%P{Q28nzL{QS$2I$-&yv!5{}EXXtB_W8ccDomVp>gv$f+g
zs*{=+)Ce^vfF_DzI(D78<3MLI;<Xtgw{2o=tm*5{{zwun@l=}*B9yT-ZWhfPo!IXf
z7#MJm{$UgS&`24}$4-~0sbmXayDAFW0yz=-QaPS<<!3>xNChhQs>;%`K+N*{#d<`m
zNvz)(KdT21*RRpt21e&zge}a-k{+|ay&z$uWK+93q9F4OboI^E07&PF8&y9*5AkKP
zG73R?k0@mho8_rdj~4O{)?%o8w}#?7+?!m2zX!*C4;B)xi2T-l-@PFkcH?xLm#)Jx
zlz1)@x!1uRv(?i5J`NJ_EGpX1_D9(_oy&HKH(Qgz{RS$l1y5=`ZblhxtemfC5Mun&
zm<F43Bi66^iXl&1d2)E3B4{j2;AJGVuNb4B5~3=*BH(!5!L;!Eg`hx-tB(%Z*mM~y
z$&qjOALNF);CNwmKkAYwvHJXAt0)im*|SmBtcp)zmxblIeDDQ{5LV4=>`!oj<Q4j0
znL|YJRg<q}xUc8vm`nZEesaahUmdFT3GBPxfGl);68$1Tw?U0cSYCNloE%Kjp###%
z%O=aVsR*%2P{b5h_p<kC@sc;`P6s=<JpWG@#OA(oLgCF52kT$>d*3Grh@tYDn9svE
z@FW_?#6;Cg2(A+Vs?=-Ot~Ht|KU4D=+q&Q&rZ)+hbL@~v;cIP+<)-@BTFVJfoCsIA
zfB*iqYkXWxrWvdAyT!t6^oFOx+5gs{TPo&X;f}9=<layLNgF9ysT0Q7;s{0%=*kAg
zX(F>6WnP;pG8Sro?zCD{x?6Shl+Z-&<Njh3u{k=up-<x;EMKfHO|e!0Bz}b-#3L)-
zwBBc|Te#U*Z^U}@?nQRV!IlT1sn@oqUNi2CRwut4nsAD~b;n=)j$@=5h)((Is|)ED
zZr49^7St~EFdKb;ax=a63oKpbYMX&y(Mz;H(&mgTue=-DP4vY!Cz<D^^loD8iVbmR
z(74)Xd`D8`>{Gte=<AuPz1|4D$u*oZ6QLFa`@Q`5WV>j@LY_*u7+9szroDvqE-&Bu
zYGX|%*)>b{8Q<wVw`spAaV`nrU!WD?aW&b<t^iiaYT}p{`qh>9bcA0TnCDOkTze^d
z243j0<i#1!$BzE71gjZW-Knl2nDCK^Fk}JG)+!I1ZjigS@TZY{)$d*^*w-k;L}dJr
zhUh4b#vDVe{(#F)L6Us<Lc_$~l6&-ep;**<N%VY4Ze>Aut%oeOp-nB#hMO7DHlaJ>
zGhRP9VHalcDsTMpH#x_vIorViiBF1~EIy&m_WbCK&%Vt%##g-izLZVf7!YU8ChfT4
zLD>iriXWUgxLd_o7%nuTVLUD%w}qi_(!_RGh|X4kE+Zf{=n5~GW%s+BTZp62rkYK*
zNG;Ml6x)A{#fi|oa4_Z)e2~1+QyYK!^l2)iSdB0j`<tX`WUK7d-n$c*dC$`c=a#az
zJ7B9XYwOY~YrHAUo*LAw0yQIf9*TA+C7;kdqFT3RdW<fDsyxzHq}9kRlN~3dxl8(P
zlE~{j@_KSrFS}H1AQWkqe(+*pNlA$}x7NzX!{UW}a~b33H!mOOJirHP-JE9FNtP1U
zdkYN@IzNB@JTAH=BhIxPf65nImfjZ~bE<_3w$0L?w7%{%M6}Vu#j7TQSN?Yc1KFAt
zW!mBkCnL#P6*q>%{p_s_{Mh1LKk|0DQ-6YQ<KNBjv7uoA2nfqQ1u1PKVzv<3F>@Kw
zbC*w?I3ZRUY$d+iX_qPTqR4Z9<p;Ba8VWLWKiA#W6+Im>!K8}qqGL92#R&-s{HLnf
z5mlgD3#_2oMq4Ag_=}bO#Kofix)y4e_Q{VN)ms$_b-%jsz+=F$rh?+WLmp;=*-HA1
zpm1v)-C7TXkUJY<EtwfngqOKYTut_Eq<1KH@49w6RRvoZtT1#%qI=8&8)1s!R-waY
z?!HaX0cw4X>#=ea;$QxhETk6g24HOZ=~hv$B&r#zkrH-_)YdGRrIsVG+^>ZdcSJoN
z8Z{AEUTrLWsCBeaQWumHEY3J;XC^eWr}WX02ov`BAWI%lWlDOce5oEFlR<!1*tyee
z@Ln3Jh8j+vHUv&@0I*qk-gZ(8Yw@}vvBM|Seq(nXCXG7ZsJ=k)F3}HowxKhGhqgF%
zbad>tjfYGStXNIAVBr&6VIqIh(pqQ)0W=&`x6JBMyZ(45I{gZ7N=DVgb;YAWE$Tn#
zuQzRr8-a8b@poS`8R=AYM0*K1RHk$#s`k~Ft#0sId(sC<rpix2$4c>ZP2hyHwrr<9
zMYm%_c-eFxdlU3RwY#}R=W)JFuZ`E7n{vzL%4?F{-3kL{0d2vAk)Ze>eB9Qd@U{Q_
zFJt-b3P7T~;dUm1xq(m4AUuZfwS<cSh)YxhY1=>jSUwq3IuEt6H`Mq=7le2?3%wI>
zspnShcq{p2+je|1(g%skaLU_u$?*6z2J`)~9formBgPP?k)N(u-X@nhG+c+1Q0KXB
zlVN}awEQ6$q7aKeKGD1%{u#Zdu|b*7&_223>#863TYv#1htkdbVdGA>BQ}DA6vcD%
z@`wkAU9GQuK&Oh?{NoYFkrt@MEzGT<nX}9d_P_hd%bx1?jg+X<<+u~)DiB9NA!S26
zHva1LEvgIIowJ?8@gAf@2Grv%%3cD+OFgqJ^1E)RzV`KwMVl*IGH$%mvmZuIk)c%y
zvdNBtzDIm|GW@Ce^$bD=|LC1=%jcdxhDFe7`EdB%-T96$J&>6Qr0sqBK<txR9B%pR
zctc3cT=vB7v{+8q^qYfW9D|Wg4e%OXVtx3mCfH^W-z{uPbX(2se!Uj03X?<f*_`M3
zMJ0~AaTdxpf+}TBXVg>Yf^n+Xju#7RNVDYvO(RA2B|BwsK*RP}n%fyfk4*GAcniRJ
zmZt#)Q<p1$Wd<XXk3SCWjE(YhT23Ib$F_GVYPz>BblP>w5Az>y7m?W0q%4@&%~R-`
zGEYh0iV`2fKe@whXvS0w8@N*rV&u6GoCJ_oMAHK^q0?CRAdl1KZrrIaUW^qFw*WVt
z@d*@*OJ9K6tugGBx;15~84vtFOb@rT*{>KZ(j3;9Vn^rpZ-ld;$V}Q3_SgYS)wA)u
zCnK01-0&oEX;~EalsA~Kg*}dsT_D!_AK*Kzb?+9?cRTKI5LY<3*Ecrys?!$hzJyCL
zK6VlW*kr$hh9a8NUrPWp<2H_vW^t6}exZHRtYs2bNUNs3UV9k<KR!vU8(C<!!ypgw
zx@4ZlDjNFzE3hgZqecRQ+p8v8N=neCjj)MMzWqtyMEj}{4;(CR-?#jAiq8N=FU4r2
zMHO{Io9(A+pJy7rjQqm#)O`)Ud@GF6HMcCcrj0}cWbooh(PlMKBo3`x{-N}`6<-#6
zLIga{@l=z0LQg=ydBn`Y<|Xc={^|ik(@a8SW0UsdF&Q(QF$f!Tu*nmMSj?XAU3@!F
zQ){H1bqZf7Z`Zj<%@!LB3MV4f{V!E#=sf=CF6VO<-`dc-jybsP1={2uk`FeSq#)E_
z9AP$hl~^@6^XsrY;eW}M|N3HD>=)tQT8qKAK4x~t5$X#t4?cEay>a|t7Ep(3@(kbo
zs6}rlp{Vs<LtUG2DGcg6tBy79d>Qa<q7A4CbsWOitP!T7n7{H<>IzlI*7tT)n1uzl
zdLW)pH}}r5RU2RYZdcwLANPp4L%Q{pF`Th<DCO2;=Aj@ibX8@w=^%+l5aOO*VRz{~
z!$|L&zxbWzx3kYsDX8_axT<t?Dj+v)0}B@Wz<BnAqzBE(f0wYq<?6<28;)N^S`_6U
z{~dIi{@CKGy%o<j16jMy-<sBQ0=nw-LCN6XK~Ndqvd7;%;LYsvYQBn^u1{EMk&ehx
zoAU?%c<e={?Zszz?$b=%vOi%(NKEFJ49D;}HR@({&3Fvp4nVXQ<cbjp#8mK~e_l}-
zX_{#}CYsD3+P0!)pE$`=50X{;*;Dqa-~E;O_mK_sgzwT#z4p-2*#bShk~^etr9CmY
zM`P;PUovL_>O9Sccv*W6Xrc>wHTE{^W^$PumVioO?-fk9@x{1F0qs%6V-+!*fH)AJ
zmL7VxdQ=f|K=9Tnw{fM{%j_&k&f{pyF(V_*cE-AZ(Jf{nbuDWZNo{wJmE|Sh7LERh
z(4+N7J=kmK7VjOQ(G5-KepLjyd>26&$eKg;vu6i?dV!aEu8rL=&+c?IRg4aE=kQPW
z>s8hGnO^lTF392<fBvV2e~>yE{SSGkgN(8e#H9DwsmbL&Pg>Gca^KuNSCNqdr}X52
zy2^w)fJzD*Y^kWI$lD$z?u{O7ghbDTOj_`wgu#P&DtZLv0-=<^Xe|+r%B@q|#iqwc
zZpbU7v3kpw9|7%?hZ2H!Mx{RYnnQs$4ILMi{vOfNx%SimTe#O~dlzj90U@%kE>I3O
zoDMd&;DDB6hy^#ael6H;W=wc2=5&vo@8qNkAE(I)m9j0JDj!^z-3aanT~~Jt-qxu|
z^8FBO?mhP}r*$~20L9p%{_4JHr)9S9X)T_{o~tVQ^k*_~g4Ylcq*R&X_iIBR&Z@mV
ztHmo{T@n~$0(6WrP|!Zg%bYJhOT@l1+2<sw0HqHjm@NKFsg0157#SL?@0q!ENz8R?
zk;tWI!$YSl&YnmRE%5R|c8xv?-pU;A?;m`BiA>*(&4$q$4pvJ?6r#3wKcEpC(PKci
zA){&chr$7oddNB7`9WY4z?epjGzs0@A{xB<F23BjjoT%YAJ01keZ75E;>Kpsr=xrF
zk?xwRVFWObl)jI@pH0vn@^`erpl-ukbYgClp6}`lBeB17ir)U=BzED#_LQH0$5-<z
z87L~AQj7wftEmeQ1w`~hz+4oAm7Uz2hpuk+85qyqw;P{mf$ClzY+3a;Va@*MfBdjf
z^}OWYMuW{djQ$f&r_|Zwvqud}82;>Vki~oRA>5(*G1X{mA_gc!41K9x0Weq!SR7(q
z{Y9YK(2}~&lvH$fU69~)^tQeo+00*ITs<h;%-S$kng3q2>tGYvzZkm5&BYb{PwVsF
z#BPSKdvwa~1vw*Z4g}I#RvT^Gfith_AkbGg<++94xKejt<_MKP(w*Hkz7DX(Wutd^
zC*Q^5m(c8h7AUg-608iOlLgts&<{G#gP&bN34-zKvxTYq22M^BObjPG$@(hBi{@k5
zFk7KYj{h!E-+<FZiI(YQdsh7o-dF)Z_+fT-b_3IQf7&;CYQ!6N_Ff=Cu9DhM(<Z)U
zG{~!Uif*#IICNxl{r)8?`5q;spy2-!*G%KJXCW$8sq)io?w*jTzi#HAWsgtlF~l`T
zcIRrVz?*f<YU`rbu{G5C?RhkA-}2uBzOrZ5ZuAABG%Tm1yY)yW9y)mD*+Fw>I&smg
zq-8XCz=)R&mlTUwD31y5RH0~<egkx~75F6u+fVHb;&DLu3o5M1Hyt&y)4rAh#p7;%
zQT^p59NLN)TMZjXmdxhG#X`P+-Wv^-YC!DZw<LJJY6<>StpNHVp*g_2a-TURV82qd
z5)?AxCCvt=&l>7L9rVRtP%Pn3AcHN$NT(>ff@!}{50ASdCtNoChNh#PpCMX)`|UTB
zj3X~k!}M2LrSUP^FE}c>BzWf=K;4O^v=`m1AQdMK8+9a~j3{SP&Fc;RJ})6`&uo&@
z?qsNEKHk<aBp_WeRuNn<dTjMOgiy8KruN_sFPHA#djE3WWTI?S6`)+f*Q=-fhV-F9
z0zJ2{j>zzAa-9t8&3#)Jq&@hnb;wB!WJjlkZOL{5zx9rQM2fAxx0^?f<Zg@DzGBlR
zFB|hd^dNjERgV_2Okni&=|QVfn$~_U!AV(SkzZ6RV1-Xf5YbL#(~U+q`JwZABk~=&
z;2^>X_Mn^l!JGAHYEu+5W54}iOU$?Tn7D=~awU5&+E$>_arPgIv@rSNRI^`trQ`|<
z>h?>a!r$0@TX?k-P^H+PM)7+n8ACu3$*O&#>+el4Br#tRIrir_BG!l9^9bvJ%DQJ+
z>(v3K$8rbZg4RF#v29X${#ueh83D1&eWMz&^ckA%1lfylf&Vn|&FYHYUN-#ZBe!>W
zls*BOsNIacy}dfF)6&!fj`@p<T3R{uC$I3PH%5#QuqFE40Pnd#D*j-|!0-OxuYJ0s
z_=5b<G2g~|MTI1QyC~nKW_wps89oK3rP{XdQ2Mo?$gnLMVxk7xlAv2Xm_R~>HZsN~
z?q^q~Scx}A-FSrHl<It57k&$)+K((Y-^z<_e)R1jxcIVRO62#$_UwM>L3`D$W98Lf
zM&54rJpcv~s-LScdyAJF?hs%+H>q{(hz2-26qd+=-@WHeIdlKIZmrV1GHk7Wz@Fxk
z?w4~Jm3Rx@0`^rF@m0{H5%Iu4ye?qLy+|&&is9C7r<(xMMC!%Q(x3L&qQUgW{>7q=
z+<N-sZ({r0{a`?Upl$+-6z%8ZBfn-|%R}LUlbDQUCtkKBvWQcBqWq~<XFl$@AO}nB
zUl3Do*ru4?+2Vw*af3AmH_CLovqvP^a3}Ch1By{AQ=l2orwn%$x35gCDl|+Dmul27
zqk3)*3eovmkne7iI(Y6#mOkE^d)eRB@xaibNg-#`_KlmHo6bdYe5#k+ic0I~J^HQl
z+=#0f*(@8;7O1byGV8nd;gHcl2N^EMFK={@Mgr-+i(je#D~#m-r#t!YviFwVGwiZG
ztHUHmxdr9h{b3{&FMzvDX}0j~9~s}yz{MKd7M3T*ht0QM+^$IKDz#z%jlJ-7*9Zh>
zN~wM=I%ebB?>NbP1gNr?q1s!|y6Dd5U{#L14yRmx;7zksFxBfCLsxF{v^5B{ir3y`
z*^IY!^j6h4DnZ^N(){VNYwTJ$(wb^nSG+S^88y9Djf2`*NnyU~f>?I;^xwNS1TQA0
zkuG1pjOK=mZJBH1=XtA=$-=alP<dEkuQt_jXHrdc$s=N}WC>H_ZId+jmz1zh@B-wr
zF!>(+k6x&_PgVMTQ`1l1ps?q1?}ApI>$y0TY_tQUUAafd+qBm)>sS58%a_Ls<4UPf
zDr;*=KC*flcLc2}lT!vGP}4v>4Lw3obvvJAO9~&v>*f7f>_@k*3I!CLw<xPS3HpMJ
zgm5WV>o3av8=HN|51wNS8Ovy2<MX5m?#2zQ?_RgYGPa112eDCDR~5J<a;8<2|L%zG
z2)EY*mv1~r*^Q6&w-sOX|NN@KZL!-(at6^nv%47UFNbE%piyQprz9_ERct4q;Ru1|
zC$>JRQB^cFc5d381<9g{7L&Li7>1_dr_Y=K1-W!B^ooV}Y_5FI99Dq)g#x#TAKv1g
z8#e7u8n#EC`qRThic!}B&I=L+I$GC0XLTy=pI35W^$F*G3C`rpanNgD+#iQGc${h$
zEuP)E#8&-IL!mlAZp2W{Esp<~DJL8xy-=YK$b&e?$3SEuS~^Fh0xZqDpN%$#J!EFx
zFy88Px48O(x?0$RSggyii%G=}@YNs~pPATejyEwVsKp9>aTcV6Ibrs@q@-&`-1HNO
zol~K;3Z|+9ji_>*Yh2P-{B?E}?$+Y(uI-v|TfWQe9<I$f-BMDRj@$Ww666}apB8gy
zEOq5_z(Ep``)((~zlOQGs%k_I<%arWE)TW<irY_{yzDOdy_!OB9AJCwtwu4%87pBk
zNCuwFRrR`hCstC46D-x}-km%0`SaaJb~rx>Ysph*b)BZnh$RXCrPRS~e03X>*P!j`
zQIU9}f^+-W=K^;axnm18vb-&fla0$zx7zB-*Y_8S%L=s=w9i@B@P0}Xu=i`82P#DJ
zlC_l9kZ*AB&8MVU2Y0SipeN2oR`qH~j2;|ZsABakzy5T?J&?J%=oAh>WqZ&~jD$bo
zOf!z0dB^5>71~TJqq>grWAZKknhb3^>{oII!+Np>VmbY=x!R))C>tp$6lqSdGv=!{
zw>$5E6c2dNk<I6ERgj53ZDk|a7OioCJ(V-tA<}1v>q1Ze!2WKW+{X40a^<^@tg>+Q
zH#_R%?it;KZGQ){?WEjX6?vgF<mxkta#TXLSlJVG0?9pJ(m2QF{E?u5FD_RWK5HN}
zcjxY3V2&iz5-%GXxgJ4X%MCS;9jQhcsh@07%3NjcXwmZX^X(U%fPzWw1Uhlz$}Q>q
zl@6%4CWs9kbpub%$yAP5ZzsPZ_hl?|ln}rP_yw(PO3WTbSBRXO!OP{ngy+&YLpNTc
z;itI-7T*cUy;EV@M=tq|h~_=`wJRGyysq;)cl_(b`p{l9Jc%E>+NeicH<b`ifjRDd
z0QMo~-$s=}p)lmVK_!Q!JG!4A^|ZWNS~z<CYDG&sX3>ZfA;8byv|**YVda~hmDN}O
z7c*8;#|i>S;q@<bQbjltWuP#j#dkq(lMt=8Yk$f8ulkr2q}^~S@_T$ePwpC3uU=gY
zXC-BCO=~lPS^y~j7Z~7`Y-krq{z?duIj%eR{19K0Dj;Xz<>b<;_3WD9;;O*sup;lO
zk*k2+wgqE{2Mv$@!OhJ*6Wy0(ox?v~l8F$d)D$ppH#*Sm`b1)@`~C~6PgIgbzCQzW
zJLBq9ay1)V+4mc2nB>Xy=-e4tM2)S<K&!&FgBOOD5%}pAGRhQA0_f9b@5k`9J~U$S
zxj_4)=;jer>sQ*FH*Zpxs01c=adGkF?qd)a6Xx0G_jtTO=DX%nt$v0OB}{6;u+{wL
zult$Amhh?wW-Tz@GhjR-yYxphKF_8I@5V06!w)Bo92y#-6u>~7!GLi0-;9TaG4-L-
zXIu&Nfic?o78-tO_mkbk4!L`_k+sD(esf{?w{CB$U--fE=g%Pq7$d02tR*mU$<mAI
zFa8JE={-i_B35jD!lwbF$rreWBfxfBal!{lpc*Eo`;d&fv%!52QOPHy$j>Qs+bNJM
zU-2!u&#SNr+A}iW9F?xSD^`p>nXL-jdCgpEShhskjD`~(>90)Fg}2_BLqIL_p}_wa
z<+Ss6evub=?LPPy;UNnKN8?<*#E@O*sSP2xvXwegdV(ZC3HeU7)Gx@rOJe!`r{A6Y
zmH*J;s$g4Lz<R27IGrSZKUR>%##eUlwG9SaRy)4XxeVBv-b1e^_BU%{R{M+<qAUXt
zK!#ylm@F-H;lh_F{%}ak&m@H=bsKdIHbw-*r6zXevh^=!NIoA&3qoZ|3%{ln+1{1q
z0f!cdtDk+kkqNBc{l&t)eSo5LH%6_;o?!g|NLCR$J3D2SU>^1p<BX2V;r=4t&0MO&
zxcla(Gj>=w9PSIK$WaFiP6zauaSsJW4UKy9${I>|nm0k__~aNx+4nN`gv+kKXTD-U
z!ckA|eHQzUl7EFw2PSn*Ld$dPy`q5@@1_xy8?G;oIb23DpPRhk(b<TJaq55gME6zR
z(~jS&Eh-iZ$>10<DqTVAPc#TUC|?IS<GNaEYBleXfffLZU_7u@xz?21rg_EFuU?ct
zfO0eR@Bz``7v2ruAF=nfF)Fapocs3VBdD(gHW5m!m<Em=*KPbrg%>)E^gv^DjfN15
zo&ch1`y|^U^Q}@%EUat9Be<FW1pdBmSP|8>-a$1vd+%q|#-fY}(1gaErFX|z8+$r{
z?L70M<ZUZ3)K&2=A^Cdof1<@-REH6W(Txkr2txY}*I)84AtjL(^v4^-^%R~b1L>J1
zZ=`v_f@8w$nxOFq{)(|~A*f4cAkt&Tc*rtmeutPHTC}l<kiT;C@>9}BBV?6l2eok!
zgEJB1qiRG?<FR12SlyqkznR&if;a0iDjk|a9_aKwpkuy?IpL6qkh7iHWqIQg)v@+E
ze&`<jXdARsokNj#M-9<N5D9$sSFJ!tH7TB^t;wXdY4a_F?k>~>n%K<;{yZ6*v`Z7)
zr%fL!$<jG6><9Dwm3SrwHo(u=?L7VVl&kxI`{NoV;jV*e5wxg>F!@6-QZr<KB9A#X
z^JxGa4!9pm!nwJ*Wu=eRpEldJ^YSQ!=V(aIMpK+(qNW{wBMnCLaK;|p73WUp62{I4
zk7B>`Y%`)Mp^BKq`hsz<X>6gvVqq1C!xO|I8wscwO>2^eqCzEDE6hEL?9mYD0+zvx
zHfp$T&Zc%w3t&Wkj*De>QTM-qh!5P|-E{+i6kYUNB2ky2VgxH{sV`RDDle);A%dK+
z{v)b2X{L4yYSB#viJ9rY5Yn+tbSa3`{%0|T@0ekF*U+`4*XVH5e81js5Y4YK)rMVG
z3;g{9s*dD!Ib!BVQROR#^ZAPAWvYml;O!>A5pQ%+Sj%BRHg7KSA{YGw3ZWwFFq5<a
zIQCP|E;w<wxpPUbB~*1D8NW|N|B#*A$nQN50+Ae7!5u<AIlQNtd)mCgs1IuZ1?qX+
zt<9*^v8inxE-=2td-`h)6i<K^iH*5$H!I-kv6a(jWA@kb#B7Y`R##Vnpi-xSK+x?T
zu-wo1YmdA=IreS{2Yad?q3|t*OM;G)jj(k~N|@p`3E>n;pdXdSctIU-{CyhITD@~X
z{B0Njq@EpnTEv6{H)1r_T+kyAnBDL{5qE4FR6VFGcGREg5&N^L`yTh@iu!xhkg1Ns
zyMLj*hH%FEv`SlbYI$_7>Ct<IC%LU$ODbJ<cNKcMK*#Q07L>O&l$BaFs=vdW-D5HU
z&#%!D_Ta}*xhqi@$eTgE+v00lzkaM1=RK`ah=diXC1*P@-VQr^w<BwZhlh2AgoSs<
z@zFm}qSRXdsW?<Z%Lu=~r|GbkHVkx)+^6F`RK(X(`S+2Pynaz8fufk)&Y68bt(iTx
z%0!0xq4K(EETGRNkx|L`Ry2JeQROLVwH`=&vTE9K!HF>uOg&n_9X5EZsl|_#>>69U
z_byv8(VxNIKxVimFJ^yRc93=}twqu_$3ys^N$hIX^i{fJagceXPPh&HolEVx(*kta
z|KJY`i+14z{&(rhyBfLEZ9yfKw?WRBd3Crrf=k3}tXQ>gZZB~-XBgl+n&8TpjyK`j
z+!&pO2YnBOI=dzwv1oYw`FSBu&dsdRxk%HmVL1I;hkVl)uq&O6t=Xd->^Q}27H9i}
zf5G)1Q!oLw`0uSf0j{!w&ciOe-Xx&^*{;<eA>l3DTbIzYKg}Vx@p+A~%22uoGH`gt
zSBCct^j?lE2LA```Ee-x3b<0_FnFYUzG>6cuJlcdTTDPulwGy*X*!MfNtgG`?F-yv
zkG9R4lwjz$8}o;0!R6AlN?TflW+c|v;MN%AfRvSqc2P&(v$sAg%iBeu$6aYVVqPrk
zC*x=;cPkX4QvCdH`XrfOF20~++&@}8?}N(Xi1k~lfCHZXB5DLT|C2+Z(eKB{j~`jF
zd|9%XMB0b5o7oP5)yHNFRIB~F_eVDh(b{kSkSh`k{T?gZ-rhcFG@>c~hX&gnk@LJf
zXfKE)HWf%&z1B7Su8fEZZm6OEQmVpPG%G1m-13t#dWtz<PGj410(vh6+0mAVOV@Sy
z<|I#UAs?@`zcrOwWi5xPBRLUzjb`#!ctm<*jWWaM@FE)5-5zQlXjFtmHiIu?o}2C0
z9BfWdH-QYPtH#q7|JZ@CKtSYEUtn#0y2jDTy`jvWI2cac3ZF(^2vkft!eVoOYfjv4
z1KBx)e7sHlbJAH*>)gBf+Dp7C&3D;e2r;Zc<2^m*jk}9Lc%EU$4j)yqsm0xq;<WDE
znpP6JP~E2;<PLf1Q#IJLv$3XouqHNK?uxKAYTvLBIj6@X%>B$lUF-Vwf(;)FKD^CY
zZ*<Ur-QAL@k);1c^kkBqLl>VPgOid1MctGnE1J>NKVa5CL}d!-BB$u2(Yya*i<qeB
zKEA5JApovom1!>cK36Rdgqv0vjc0<#Q=&aQX5+bacUQVoQ&a6Ud?A6(FUX|zeI4Ig
zKwqJIXb>G($7eq{9X%4dm3EF3CxNo{=8|kgb)u=fkDuc8ZzHP90agAna4k@!E=#~h
z_-IwLnSQ8g{2|AH$d0JkgpR%+o>u0(bt+6H#_5Zgg2u;xs3D_XFlytr!v1deLFE0l
zgguQQL-QYa^2b}=t{7o1I!)RWhiGowHc&?Vf?M{>f{yx)k2|}KEe83`*x`JE@8Okt
zK&e|nsN@}Pc0I8sRR>UrO`5*|j<inSW1#qKr7h;119-^kw<PP%gJ~?dUO?<RGY{=;
zg4<XSSq^RYo!bRG<!hLL9btRk`2>|d%BQUhbIK!oEJAq-V}Uwqzv4Bw;K1}n-qfh2
zRl+M-TP&?P+lNW0lr?hc;ftpMBM@IpeQFt#30TZ)`p&y{WcI3;@}n2x-DZ4MS+t%1
ze6PJsS&k+x4BxX$2I7JfP4u#cVAN&Oz<_nx6G-Lh9WZOh`$9do=Tli1u6ao>Ld$Cw
zy8H5fT|sM}=9P;yWq6P(R@f9Ui~u4;{vrk-kwzhqerN#}<gaxGV@5M%Y4>ieX9pN1
z0`yQ7))mIb*114ic~xJG7=uDKDh*ng)ItFuI}HPLoFw*VRUpkdzbMzJ`))WjUhEbO
z$jOBDzRK_Q(YUGpzn4wsHy0WmPu#Fmmjhc}%|`IGh2cV#x#6HQH9%YaDL&C0;yx7w
z#POs5?F1eb@v_A%9hV(%o5_L+-eA#oi*8=f?1qbnbQx`M`=~O9Q%AiTfm=(gefTY%
z8L~K%=3QJrIC|R~_QFor?{`c_9obyO{xt?`8!;8DPR(!bM!<ZIfZ3W8SctceNZo31
z-UlOkwsxmlyu`4oeO?*!b~ye`g$g`=#l!dzkun_Q8exCek!~qyp3|cl=Asd;4|j@y
z5Q;h!%kKT-bqup6hFNhi>9m`aal^_1dH#kgLH%y=IUeGiiLdQ?yf?U^`;pBLe3>zK
z*Z)FJ*5|PAA|U73+HYL)6cIrqqN?#E4rCvHA1W?8GxPInKXpz2$v>t8R1a+y29O%A
zDTJ?2D+6W8QEgx%H>`y_@-tYUKBh@s<?6IZtw=~hK{>qh-OAckEG3;ZK}riye09C4
zHx@^v8r7@cLY%-IO+Ag2LKmHudq>cNBtE&+b&(8SQ`FCVB|vEt8r0J`3+IcI;;~SS
zJY=H(x=3JQPT<2Ur`eqXgSD-X_b{W^{~8W3#N6M`J(vg#o+HAxXJ6r?P5<@N43@*X
zBW_fxQ#9Uhm^Ux4toY@$I0im|tfcSOcQz;u5L>0a{YL6yf1%K=t2~?x{H0I`FBdx8
z#LhP$s}kW;`HBoAW@y1yMR?m1FHJSIk|$f$O-)UVVWaV|x?w=gDwWr0fAr^P7R<TD
zOf@aQnj4Ql^j<-S#$nwGi9`488sHeJV}dL)__(~$PaqY;`sz8}a>&d~h`IX0u~kdC
zB5dY^U(ATF$EaTIA+>lwAL=Vc7(fp0SH9`);=Jt#=vWerkQhdYCV;^JJJ`<0PXj{d
z32$Skt<Pa=4>UrCzFz;WX<$!n#EX-YvvE0=RkYS}M#URqr~CfSINu+vgHILJMV%`E
zzi3P!I<9aQo)p}-`ho0eYjh-$pyHfs)x!8}r}I}VZ}<*%BmC55NAf>?12BOxItk``
zOk0QANmsojKcZbPh2IE($5wf^yq!Lq`ru7&_Kk?(KCtVzbUCrytSrkmd})MUt&xa;
z%eScBk<vt(-`Au!Em>2dn#zBExwNpj8wJxD-}yD|l5?Dl+IM<bAezb!I^!)`Tn!j4
z$$#oeAP^*W?LP)dfN*B!JxH2pEE46)xEBCFhH;pAk(tbCSK1WrN(iqtfbKQ?+@8rJ
zUaym>9<>>bLTZVV>rsi&Jhg>^a)~#xkzX;;qmx&kY*{Q?SpDwy(&9ZrRNx5t1|h~`
zn7`Boa9e!EutreW%=C2AJLgugUhK+@Jl+zR>>%mmNC+TVzM7d2Kzw`rFUiJc0U;b+
z%yZ)VH2{CGGI9MkTcVvnTqU#>;F&gUv2Y6P9RPW~R`578@W_7c39{y72>M#hXQ+(!
z=|hpcE;n@V#Nw;j;XuwZ(Of-Ze|rXJ>o0ZfU`-z^V|(`mz1>`hx||wG`uUrJsa;Yf
zmlqj7a?g(7qRvpUqL+{@oy-WA<4v2Eyw0X@bJ7EW*WBTHM^n9Z)m&-Jwb4=?Y4@h<
zylIy^u&n;SPXo_zCxg5YDDrz0zv#t_mfv3ik`j=LuyK^z-5aH2?Y;$msSbynblwYH
zZ+A0HvGWWbcMkqhf&rl}E-fz~w0+7Lt&*jLm!fDccOp2$tC~<nm2pt`M_V(3SM7Jc
z*45*nBdmYob6Idn$x{)-1Bv~^-)h8C!+3_hMD87Ek3TNu6U>P@LMhV#9w{!Dr&)5$
zU3cf`*SiSG8|k&&$yid3n0i_@->$pAs?f9u^KEdNNt&mf60x?j=AMB@Iq?3J5hsJw
zS_}sLrh>8U^9yg+{-=8S6r=^Fnu{+~$PjEH(D<^o#2c#@TG(X(E@kB(Re044h`088
z9-i0gG|MW#H~ETk?_xKM9gcU2lv`?tCSeoE2G~@<U@GUUM35pj7kanv7|*#RCv)Is
z5J=MZKo~j8MC#}S&#K1JL@iE|*rO`0JbLC+_al8_hmK1vxs>ynOGM*uKDl?wRl1<e
zd?VT<uNdw@mRu*l<HOm<iAi0mEc8r`)0`V()gQFffNcA_DkHiJL=9)Ij``+A#Q6<r
zDtMTP0P;#p7Cj#y^5^GBY6JP9?z47%m*uUa6Mah!|GHjnHq0`}kMm;USLlRluKuZ(
zXvf=dH|Io|(Y|d`y=iPaAgn9_O+Lnat2qjsRiT>YfNx{L1ONdY?VZN7>9I2vPEFmm
zppJ|CGqguKIyycmQQ%Q{y+gBKcubJzUE6!cON;xj%ecZc#6<u5$xloRTxxed{7Dax
zaYb9M&fg&B9f#uhfsj^5WD{Nns9O;nb3DAbMq9{Y;ozP6YhDtBWd1&uxL5f*v6mi7
z=X-6vanG%@1kj8VeTpa(SCu>et_sAEwKbZt4XKi+vxsmbw0XfBczO4Cq@xd&pI5U6
zw&0F>4mfvvqs<;#^{NZFK3LO!?1-W4kROmmB)Z9N?3ra~+J!PbqMD>wK5u7tI%>G7
z5}hAHww8&j){8sKdA@QZfMw!_k12XDm_pgcY-&erdNldD!ocdnfEG(jjt{%RO$9!+
z=Yi{#tE#4^XqQ8=ar#5o<k6qI_`#pj-@FkmV&UAww9=oSvJOXg3vNLeIAN-m(TDD~
zMi%=GjT$1JkP^2^{=XOdtknT=-2Db3vjN!<vK&2d2rm=%4vva|MQ>blf#urIL!GN1
zoDL7!r<8cj>^=TQTS<-CopOpie8foXiqQ0gw|I3bc*zlB4)#CYz6ou3S@}9$kX@2>
z15zIN8$W>Lf5ZAyp0Bu*t7xe>I?2glN;dQ#{nfu8=2hEpng08v5@S*M*U~xTAaT$)
z%RE_OJ*YPq969aig*JyY-(h>M`$!fPT(y#&Da+}6X7mns`w7FZKXrXS3=VyG_zh7N
zj6zn<-pVcl^yt6IInd?rZ~cZ49O$;dKGSfE!^aa(&OaK9)>qdS-0!LHe%ufS&&R(<
zZ6$1NC6DGkbK*W+C9*hhTV9<$!!Gy-6#rDQmOt)H)pdt;K@f*`89t7k8xH9D-KmU?
z4MI1#U}2b)br}^_z~YBeYWRM(sOl$L<fh-gm;2cXz!=dSG_|F_uUB~Q0x6lB^NdRP
z6t)ETaCP;nTT@z`<PnFCINztO$G!j>&#epA5*TsHf)$uM<eQ=>QS@Ie{E5ml=h_io
znb7uGh-b0Yx$CniLHYMaCTns@;9^gXxh?fm7p$zT>hzNHcK#1fUm4J3`^F2Rf+O`M
zrS(Ng5s@4vC@LzwF-8ugTe?*k_@bDgg0$p<9v#92q|1#4CrS+@M(4Tlf6m$0eR-by
zx#PNibr+<Ee88KIkebJIS?TntU6j8KkQ&%5TE#wCu34a1;~9tN9wAKZNyVdjZe;HT
zK(ZwDo+qnXN&hGbLO4ASCML;sfY!^_U(xd@2)BY~tX{h6x$$^=+Q<4YFN~uX+oayC
z4a(>Xn~bx)aU;6&8tFHHDB5;DKxFq{PESv-?gNA{q7#l;bft(|2jtPZm=7wB`ca{a
z1@xi65$Wo*cMYu`V?D{0N}fq&%)RMnW@p{Eg5_iWgyLu29}>00dIHil`O}rs%D9F#
zTsP-@q*{F6abshqKqWc1yc{{@q(Mv4+&`E)I`~kx_q1>;$_tPh2dva$wgID0N4jWl
zQTUn0RwNtv*1wJKK%skRZ_j4c>6w*-OU+y1m4a7hp8s77uP$U_i)U04Y8HJZ-^Eds
zhm0veyLgIoG!8TJU2hkFJV~P&5mbw*5F4fo0i$~2oVmy<e8cK+@q-8&%bu*w*u&e_
z>fvXP4s$ypv0IjY<A=Y;UXAuHUXjoumP&7_xV8FLoN!et!cZ=X9i*8mjN4rmAb0oM
z>?GRk;5k?#uh&~M{$OEeS0ayRx#^_r6pqFDtG;Jmo~Ruf0@nEZP#5qDi5Q@gKX(P|
zvHr-4yK#AVoAkd!B!us`9*HZRpFMU^Ql35b7EKbCnmh7dl(Xl%Hgf2)M>hYGfsR(#
zs~Lw;y|6z|*Pfza%r?a>hI!tr6}7tdgn3s9@BJ@?!GP{D8@Y1!Uh`N%71ZV&4W4XM
zQ}o+A2!*0ppbqv&0AOaxMMSB`55>|AV`E7@nrUWkN53+w8Na$-OvU|H;W|D_rOHul
zJkp&-Wq*Rdz0?S=K5pN&UfH#JzcQptiD?#%?qi|c!e_$>AbNrc%_GKkY-8!+ifwj)
zGqZ1+$XEu=Tt3)*qN&;zLDNhz(fcxI<2!ZK&@x%O9KWj8LF$=ZOrDzP@eNSLoEL)H
zcj&#<TxT34XCd(c?X+IaNOTXDIjSOgs<4oSNF-KQzT1Qk^f)!|h2Z_`lz+6zzB}e*
zrQ2dWLTd#eoen-Te?irC@u!ryH4%g<3~rk2FuI@$D=4rk5Th?P!~S=IxzWSbwIc!;
zr*?@I##2Y-i7_M1{Wjq{_xDH~lkd5%D2DDAH1S@KSi7RVU0S}UVj|w<yLldH&K%m)
z>I1(!vRPm&+(*_Z#c8eN{oO~dL%{3-C<I+y-ABvA8BzXzwTxe_<!}m?j9bnRhAOPD
zr}P_x9q`8nORVOXQulNivAe>hE&A_JvS%3=WW{~S=+~=psVL6ZQ-<dip#7LGJ@OK>
z>f);4u>RVXpSW&fpuSPh)y)&JDq04;x9~d070AA<x0;GEI?IQnm#7I_4K#;6rRDJU
zz8cM)MaMmNrWmNhPkhm6!_n_k{3FG-chn;0iXRB`mtC{`tC7GSuE`!f8e2X(Ty9f(
z&sH-6L}s9cPCegzUVvv2_$?l+p84Jcg~a-cyfR}`BD_<43W~eO4#QOkCX_}cg#Z1~
z{=A~UuH-p6Qx-Fmi=3MsrR%L6*S~gJRC>lgc`8YY^F}P?bYq9FGnzED9J6^g7C>k@
zRZY_Ko$0G#L%*s+KKFPA<%2vw@qqB2yw%nr@oo8`xb(ZJaSj_PHKzxYL5?Z_i&{Hl
z@qn~awBJ#*|2`eGlD*zMf*(Ie`CX@3!H7bTQ2zN}U()gsvnpE5V3Y|gZ2>Y$2*Q0s
zh2ttIgXfKgslklA_+T7LLL067EW`DB_i>@{VhBxlVe*e?w0PCAL!f9x&$)2r$r%1R
zHLPN#;mH2p*eFM%5aIn{>QNP`_*!b=AN_fiivpfm=ERF)^|9LGDMu^5GQH=8czLTX
zutxc>#It?t-?cDhZ5#XEeWM%%07T*`5g7(GJz$3ZP+0HQM<_h_TaK^uzKxA*fBx_G
zb<_>}h@8<o)a*{tXU~)*6TDkqplXjXjm?b>D;KHN#q31K)L!nxUsm=dkn&HQdRVb*
z?{D{-S8Aj3ygw!pA5?tU+WQu<J@UTsBYChW@_=dv1Y7WdyW%O>s3^AKvv^WV9ZGPx
zEZ@Perq$$lKvt3`y<iEmLw}Ftb>62Gh(z;~Tl7zw8N4(x71<D;I&}G1V`QY7M1tX%
zapSQgd<2L@w>2yZ00TdOn|4K^pI5^E9XGl6%aieYy;kFANp+`45(Q39ggcT}eBWVJ
z>1J?cJ-Z*KOFpB*V`^t=qy*s5BaV|pF9lTIa!m-FPb6yCBa&5b0FI50t}b`+8aB7n
z-My~v&E@0$yN~`Pk?WlH4_!j>QwDaHwSR9ZTRr2@{_j3bras<}2|c|5SP&_<i})`$
z@!6flFR|$Kox70^Cv@}Plglq_s2_N0Y(~PcrN|=D`uB$7C9cD1HJwP()YOs2sb=6v
zseEUxT^JQ|qFdmlTlwf+H}j!mV^c%JKgYW6>to77+LXE)3UzU@>VGR#R172TC65t7
z3;;%gRE!w_Wz}jA1}TNS(?(4@M#PCafunkAhb~gx<`W)mV-H{~>zrW&Y1`|jsTM7>
zixskKDGK?nRYd{A9V-W8({&Vhw7-A3^X1fnR6jZ2s<VrjM+mC#E-!v!Va~fvFmRh*
zYc(KUMc@|z@*lU9R=b-0BD=!yj>owEV=<#l)3e?3m)`BX(@4J-^EFv&5r5bJ(+o2<
z-p>$}dXM(ESCLH88Y4qPmHY4k4CmqaDR^8HN3m5+H~Av^`OyBK04%2VXwGFUfSf|f
z#%MZ45|+)!!Vh1F204x{&bvtIL~P9FmN6H4Nd%`SCGB(q8tf>>26SM(qykY~JM;X(
zb>HE!BPq_!_X8~wtWjr<iOysXP=0MM=ixA}1pcLOuRQ-^&U0ti#ql5Qa*Ib_%vj&e
zrMt0o8|DKy{iHBswoAkbNj}6_nw<J{WyqBv<sk-_^QCc)4S6NnlaP0mYWLjumz0WR
zE+Z_Y8rEzHv$9kbz@Vha+ii}YE!#O|fnN+gIx?#L5#rVYuyO?u&JwAyY0<c05&K+0
z34Rir)7TTM0A;6^Gd3d|r_bZBDw|JDBW+%=e+x^YQ?;EASh5wWT-Yub4BrjFmOg-9
zYG(E?g0=r|U#pSmZDXYinP^>Fh*<H_sf8U)h0{Iue_~su`p8RWLH_c)&3+dJ9~N}|
z0~aG10Vj(z;2(hF&9NTYjW)*ofU|RJSWzV$m7uVty%EP}OQ2B68&)hknH<-+yY>t=
z61g*)%R56$NkZNo8u~E%Y8c~^TQho>y|7LvbM)>}SMPq87-jJwSwVnoP2*dJn;v0E
zjm;?hh^T<4DO2OD`f@?|AK!F~zVqpRu_6M0^208uJ)eeinQwAoYj$5Qp8?bP^6y|3
zdwsf$Po?hCIH!MNHW$#5z@JJ!<;Y@o(@2inK<{*%>$M`q=UXa%bB6ahOeJue*`h2j
z@!3QP_9r*%ULlhtvyTgKus!C-mutVhmZI{IN>b7mw<M?#y02GmuI8O)^{-z`c>u4&
zRDV|duvPo5ajX4(=WJVR>!ev1-^+>f0|AE(8u$7qZjhRjV%y~Cd#KYfga?kS;Q79g
zTuE%!*hAXirS=cljSLeJMiMjl?0RI7(MD2qMy8RRnwpxx*!vKw@u`1AIl-fujqC2V
z*W#{^k-hlm+r2h*g94|boOl;Peyd<Z#6oVzv!G2>8ec8WH@XEk8AIoL0gi8vn()B&
zLt=>%g}pU60@AI5Yw6HmS0<0pD#fyi4~zp#OUo|Ps{<q5oA<Z74q<BYJ{2U@pA~}Z
z*F(x=A7po0ODtMMh8f8$mZ9+k#eqr-2~IK#ev0k)R2WVPlY`YTseC7NWaMRPupH@k
z0RJiL;(M1-Dc|I&>ZNk6rR<}2&7I31wnwoqSyfqCS$~6~!ON6RN`U-g+pY&_FVh9+
zK(yiGLB3)M>^0<FFg$c~a$ql{keYwO9%3X1+b!-&nXW9zcNcyS18z!;{@{|UDE_e>
zZKiu@7?iQWtD;gZs%bLPuQu*HcO>Y;L+zq%w+ra0r@&!aPF_!3-$S&7Z5(@FkY3@e
zFD<-ymL>#lex`d%sO#s`_Ko$|W8=umxeguzs>bzns0036sN%E1hjn&Pu!4tSF^;(c
zc9;ka{J&EAB?HnYAtT4O=w#4uvr!K<8n5Me0|Qo3s7JO^(UjhI>jE|Z+b>r)&H6^~
zq1;GLqJat8lm|K{@EE7dcvw_aRAU9m02%JEVVvQ_XY(^I&d~`O<;B{`yka~E3c@K&
zYBIcKh-Nz05Lm*cR)uk59kinVdxC-|F<}BG7awbqy&UHX_W<OasZ7{hU*<ONDOOR$
zH#Ki?dZ9U7QM*8v{s%{L{V=STK+;u(ui&TDrhf}zXTlVtUyN0Bjf89>UZHn-8~VIz
zd(-7OCxd3H9tfAgXaNJCT0L^U>+c@hp%Z~+Rdb6=@Z-CaUm=nFj4SzrQEV&yNFnW?
zqh!1VEjzKCyXPYtPx`4Yey~d`&-ZxOv|}!H{sCk8f4|}nnVVdHSFKTtH_NHGk)}7J
zuXEY^Jwyd6fVK4@g|o<1yu^35N6KAH&*Nf!dmq#^yvC_AGZsLz?cWSR^j}01T<WOl
zNIlhR*XuU^)F&mch6^W~w}jP#4BY}P8`%Yvi^a!bEi3clB5)C$+zAMpr2kjInX%H*
zba$*OTr!2fa^Yj?XM?G~L_nWCQ%tRN>g56vkpdAD|D>l+f@)(S7*!Jyfv1#K4}F*`
zu%9lMC5mI?*KH%!vv2hBjuh`+Wt~&?mkc_5hKFyiPsNYs9FdXzQ&UBo2a(0#b4sTS
zAq9q<2niJz#4GITH&uvP8`d&8BqZ_!u>F6EeFwhg56;Ty!91L*=ik<I?nXko{~OQ@
z@xpUHO2%-p%&DepN{F)b#Nk~URse)v2p^aqGUlHr&oF=Zez4F5iEVCk?6PKR>X(;I
zR$tw3A8bnhUKF|mD;o*6@M4M~c2=}SFnUDxtXr#GuU`-Sz4HOHV=duUSUq~TSD*qO
zeRVs|!mHk2+MVXt<E`fwfFMrq76uF?E8%Rbl<Anj7c0uKPKCOVv<S!T#N)JESI2K*
zIP_;AZQdD7&-epZ<%J+_g)6JmZDr-%<&MZ00HFSO!U*SvNS!p|HI9+2cmsd7@%@!f
zzW=wymJh_9X!~7<R5d}UQKSBoq3^yMD^vuPW)bRPy__r06cf4>*!SKxNVG^;c{R$3
zx|~QXl0LBR-)o<e8ik9AYz7QCAZ{_DfdC}+s1)(sKSMv4rfPc`H==+aQO#&uroxCw
zYWVv)a)if5l3~~t+|-jL@Z9(I47c412p4xcHaw3vlgC2G?=iYsBuVjB^i&ssBQlbj
zG2(-Fzw040R{%KMHOuUV{c(fKwKSME<K`}wb`Y1C<?4}b(3Pxi6m6#3XbEZLd_mw1
z*+8E9Q&AQo_05y<fGuP`xOSzE%x;Mo=%3!$iepPbAexqMhyv?74}7Y*n$z{N)zkfg
zBUKL>NYk>|yhD+|reZ&`@!cCbx%G`6j)<#<XPNS5&m6Dw@%}tvNMuR)w;Bl$Z*(L8
zo|>-j6PTu57Po=>Ui)W|6%)rJG%-RG#`3%w&7mlb;SDH%!)6_lV=bmu8Rx2twl)w(
z`Y7WbmzXt^0Sb|s=$C1hCa-Rjqv2FizsoGe9Q{ecki3T<BN4?2?|RY>*76DYjg3di
zK-(n^GPMAEzPe(@xDh>=m9#ovhcXgh1U8x3_G^DLbF#lTn^WAC=JpyMd8hw9wAe55
z7Cy_eQ-;5EkmM>03Ci$B04cS#yd>8Z7FJdX>2I7Id(&R^8|41uo;I$*#oQyiO)q)1
z8VRY17<kk)We2H2GQ}3v?0?*NIx$<6@irtrl=|l%GV^A_j_*Q!mSH6ero)R;Z;UA&
zy^F0sEPF?eD|TPr|H;!Bbof#bYDe%^{@sEMhX~I7qfHs+x{xY|BTPYW2mL;tyvg5*
zBl>1+k<Sc|cg`=!KCijt3rF$SowncH7AoKpy}`nwEiT}7T=L&%YF?^trsOn`VP;l@
zQdbI(FytsBt<>0EgIJ?4<I@SLNr@%SrrX0UtZxrtWg|%%d%w?OYt(CB@gE<)pMt1Z
zc-%D|RNhucnQZh{#|T*g85{BFJAA3S=wrQ4GtvhrlB|7Jel%T%$>2Ir{sD&2`7B}-
z_nYy`$x|0#<p}`y{g()yGtYN@-e7p8IuL0y6;2tbBJ`SW6Bk=KAO`<~66sA@0{?To
zEy)7UDK?97;dU@m5GcAcW8^w`=aNB+a;7UKNaG}jYPuI_?HroC&1vX1Ne+lq!tN@{
zjXB(dFlLo&?b5I26MR>6`h4+h*0lO+O5(X0>V{>2_@b%Gcy|)<A`u?PEfwS7?3`b6
zET^5W!6&WgTM0LhrtRz&DjWfv-Rs5Km3LR<?ZL?YP!Ndplsx&BC)!sj`NX)fL##b@
z?#UD26F8nEII)vNtBlkG4EJV5W=pV?HVT&HkgQ&yh49GsHwLf3EUfG;!hYFfdI;E}
zI^pBGAzYNU3!kL=bdfiPEcwTVg`h^7GEpa5BX>5>VYqFksRnf$2(_i^d;+<7Xn)^m
zhFK|i8EPivybu^f*(qDyDwxemEsNVw*W1};WH+~%uOG(09}x4u@-00z<)uuH#nr1<
zy;~|g#(t&%?+~?|ZKgjM-F%L`$-D2G%1-ao?ZmnlM#=R&_ZO5Zy5$?IF4E&Z0KeFI
z>4C>*#$DNG_|NPHgn6ci;PvR{*pPPv|HDLcoyfn2V?vi@#peg#5dnB;d_Wdp@w`ma
z(HWxFX3<1*;1BDNjOk34#f_v*WOFS1W;1QXoO{%^c4!m1kt$o6qp<2_4HZ1m^}*nb
z|CXkrd*IR51I3*md;z~mLz?7ihGn}cU3qn<$_a(tnytZQYtiCAk(A&(cx>cmr_*S9
zSU_3gQvI@cX2;g>{ZRG{?^qkTvEALq&H<O!Vob~{Tw$rIbld06W*r&AJJ9}U?v-is
z*U#sEVsq;O|JT<>k(U7MDeHPsuwn$9?9=DdyFPC9M%6JrbCCc#iedrCvu@AIl2tii
zUz1KrXUe-rOgFKo$YNGa{-is^5;ePoA*i)-2f|${_0zw9quYzgXITQD1Z>W39QG%H
z1kh-LjH8h*Xo0ziqK40!wB^dK<(>lWl9iP&4D6ODsmz^bd9nS_z^ip(F*W<=jG0Jh
ztQEpqsj)=NV%^Nr?WI3O9_81>!}f>0gjGGHfPC$pHl>|WY1jtsy$0&>_jO(_iFs}>
zBH9{vdoEG}EB!Kxt|FEyi-Py0ip)LmKdgIO7!T)C9MJ2XnE<2Rk*{!-_0R7vWBWe*
zXr^@b(T9buOc*2at?+*B{9^n+GC5;rmOf~<ft@UyD{PbhQI7L;2ju9hE0QlSuk#rS
zTsJ*ZS+GBSZRWV7$7!jLr%LM|Pm9MVo4f<Rhu)3FOA`Iiqqq0sZ}GdN-$*Jr$k!KQ
z3nh3*Mm8e#oWMMha=&Ob5Jr)&wUm)=L%QXSzcq7u&&AToSg2mN2$>wBAv3-8p5#2e
zmnwABpAo*==oM$d{vTJEWYyP!h}nACRy^abmzvvTAsedLk)!BgL($Crsm<En58P}Y
z;FR$4Xug)t5{{=dvzO?NOlFUn=@g0!=pXwSWIm{+gOC6-0kxo&=Rn(PcyAt{qLA3|
zo?`N9aavncI3zxfvR(hY_cXe><;KnbC#}=s1BcSrz)hY++Y%4wvn*d9BQm|^?3E{b
zbxx+kBp_%4h($w~^Us&gFu(E-ZYVyH|4$iNPA=No`nfagH)ez8ItDp%Wcp}!PEVDf
za;S8x@4w$B%KK2U{Yg%?4;%X*j49!`=(y>`xnFzN6t4LV@4DJ6Hg@ohE@1kyfEFMq
zL!T+uUdm<FlYRb%#-#o(fczs~zZRRg`29gM^xo_LFMrlE(l0W>Ju`dypBW3!ti45l
zIl=kDq-=B(=G@E7V#xtcwIGfXcPx`Sw~Fs@&BD!CiwnK@#k~SmGg2)+IfyE^_k{pF
zT{i<w7k1K@Rze*r7|LFJP;}%!?LqqP+SuZ6O!Yy0dT_a$58fGRwJaa=T!pB*ecMX(
zXRsje!<O?!z4YB_TRXHv^W&ZH`1i#XtcTk(TtsUL%PyXt@485?hqnmc_fN)=KUY^?
zU>&o3veFdANYmW?ta+Zfsj0^ya@H(&A&+FdJuvk9i(Mv>V#16XTCUs5L8#XlW6m9<
z9w~7(*6a61N^v#jXdb1qxjLy@W~PjW#Uy66yx&8q3)Rzlpn<O8WL3N^B2_5PlijD^
zl;mAQ8nuZgzH#gg<#+jC!*%?89okpNPq~}|wdf!ar2HinD?KNYr8ur=Q-W;Rx$9EP
zwPUe`HLeb%W?41&oZ+jsR!TMpUvBzFYHl{^NU<@(B#?T{D8IzCqwu3nW~q=DX<G(|
ze`sn)c*)Op2zb9ar5Q;4F1TL<TAPX^2|N`nfnd3N8WG&vxL_Fb<y0|ozW+DV5`a2_
z`7L1bs503(AC(F^C|$EgLYfh36~Zw7`I#cFr|3WBZ~knCq2>cx)~-qX*u#eRUa5Vw
zZDuy|l$TTr4mXO5dMR#iL1Ey%VHKIOr5DN5HZ;VJRB^`=;>;I8BlZE4SU$tD1*A{M
zjxx8v;pD&wj&w2ET^7SOrf{L_gjcx_|2AG=bPOR+DQuD_DAFVI%cN#j3^yD7jlyCr
z(nAa{H~Nf_xN~c<bHuTDF|pztmI=zuw^&xlFI{lWyE+DW6oC;qaA%5*r1?9xO7fcS
zp>Q&dv`HSoXZ%!H{>*T~)6EPx5n?~=B6&2C>qj*u(S#c^TTaN%4|}`2piGtj;R|D>
zX8VT%lNJ!>dE%%bP6C4dCu#I<oBWOK#SGW~_n5`5Cy9clCxO5>MQnE3BbguPVWhzA
zwa!O;rM7m*{o;~`uFmjlphB0LN8&5?cbbiLTfEJiI$c@U{S_ZC*o5;QK5=7DSKJ4F
zyNW@?CymV}VOev9!l7o5qs^Tlt;cL4&d=pzaGzBV4w602pD#Q2OIG+}1vhB8Wxzi2
zji{RHLM|q|0USuijFHLzDInGv9{WkE-a@E4RaN1lV%0b={r}J>oF30!c%jYc&NA&H
z*Jg|e0;M&RS^V@w(FGdajYLQINrqc)#-~_Bc?xA36U#AEj_@fIp`0(xf%p7XXh5^7
z-C%Lz4Lo}J@N?er=@6H9v5365&}l;e#K7;(Q&DrOMd!C>tv&>%D-9;iS}_{s*V8Ox
zXSluQ9$mE@IG?C~L@<I<o!oPpM;mBUM|)r*FCR`cLU92Ds?ToMQKQUQ^6=R$?3|GF
zQ%I`)9b7Q+*2dN6$6dwo&lU#whAxYzlAKp0o7S?>&eo=oHacb2SguR?1NS+Zc<|X}
zm(zl>pfvRD8iCzxMXyEMt6*{7^^Hyvt2^{t3fA=1K`>ZXRPZwiOX*l0I!n03cd$nW
zs!Df!CO=I-+DSK4iY0<^b&jt4uNApG&EWUtH5;e#7L(2Lv>?$5{qkQ|A&~imU~Fw|
z^YHW>FGb5)d|Q~fTIwFwAHEhQ>FKtE0<1Z{F|vcr-~k`T!QHZV3K6D33mY$D?VbI3
z<?>;eqj`8I?Z1tO1zYXLZfQk<!NWq~HQA{G&bIX@Rze2+e-aM$xyW&opChBXddNcu
zza!Y@<b!RsdOgl?xlkV;L6`j+Z4NYu8ZEaXX)AlT+QFb0kZcwr1p?RIVk4R8xi@s+
zhQLH8geC?vv0f{>2IiJ|qIGx$Xt!t{V%9;!dPX^Mlz1n<+;hg6$aww2Ih5+NpWaN_
zJfoIDyD?{}s;35BgEbpFj)rvh@Gy%C_ux*jQxfK8kY)Z_1_+|Oe`H^>hML;A^JRS0
z(|tP8dq~|-UP~a$9WMs3Ut0W_Eb~#lbl3??S7y|#7I%+DuHdzeu|Xr+j*A9_JI?;)
z`p5V&{|L#j_<|LJP^t8hUR8J;T@FL}XFtzBr7LRlD)rxStCm;PXMXzi=GjaIX%K0=
z3?(1~h;dPetfbujQgpq1z5ea2S&P3{K)`;&p3H`j`JI=w_osx-1I*NVk(Fwxh@!4b
z#KYuRL{Z;c>Cii|Zez%wH;}S-hMsi=r0w$)6f2xl5j|4{^B(ia%pcDiVYPWebOzFg
zN_W=w6*bd4cLj#ePL}QHtOj>KNN1q#`R*k2j)vLwH`0uYwv-HZ_h5!84mST$e@QW3
zW_6}-@{Vm$-wU1`-YM#I6_tUU_Ok}T2@D<w5fIKeJ<7Nf%W>G96qYx_p+iruYS+9r
zXY}0D{8^jQ3ZqlM_<OEr<hclK4d-scS*i6|QQ{ces-t4TRC5xJvY04bZYm>vyoLT5
zE}51k>4jh-uDUA*F3?M_7D3U(V=in<LYH}tCK6!^5Y*ATKs@*5yfORJ%gytn#p|)O
z;1XKP`YQ&h+loSm@<I(WH-}MyzN_K>*N=4M)cANmW+OkJY$$76bDLfc3D_((SvKvs
z9W*TQ$t7>V--mHgd%OUX>iLdR-SAvcd+tm~V^@zgPfz`n<E(yRIM!xsVVS**`cn+=
zpETiez))G*BUV#lowOvQEkc&USw<e!2D1nDwKp12DX@^@o~%Zrdt(Yd=BvNCkMI}<
zy{?AlDU639`%{8q?5uL_pNDkrVj7LBW#Jbzih9zAkP$;Q%2g$YTRp6~k<^^X<(t@p
zPkGx1<n$t{-A7B!KD%ov>t1!x;3?Z`uo$xQH-YQVeIt(2>go-!I^0KvB#>?-jX7^(
zVMO2|`3pgKDs3v>rZ-uOQ|JU?6c3zX3rL3w*&%HLN%^T?S}o1q8C;n2<j@=|uFbLD
z(hftXsL%2hlJc=KnyWwk6xU<87{7v0D_60JkFwG&qaKr2kn+C_l$;Rh>@7YwgW<?K
zbQxk7XRVx=X~LC~e7tov7mt37RV|&tzD*AQ_EOC^ZzBsZ0SnOS>6dG(@09%3tI}7q
zTBiIr1UJTVaTv_LrJGal{{Bs6iScAaU0L2|?Uk%Z_8FxiEF(9j_TAE)JkPMJ$LQgI
zD~Hu&$%~)L0ovxg^c{aRTdHXvcJL{+w&}a1oGW!SFUro7jcOU><(|$$os^*}zIa%V
ztiDE;%HRpQ!&|HP{Za>O5mHe|U+JsKLB`v5@54a~Z8Jc}ZJn>KIHN~EN%~J$KP571
zPQ~h*r#PEuoqCvJWXOYXMOjFc(`Eh1zLF{0O||Zx2sNP>4gpe(W<NGKSOV`DEx`<;
zpOfiYQYb-*NL??G@;A%U_Qm$-@W5Qv25|U5;m6)Y?C3yrew4G!Q5-1n@VXkhX1McY
zXyyT_;PFZLJs1(H9kloTRi~FkeU2TYVTzn#kN&IqHukX*%`v^J``jL+L(OVcS+=$Q
z^XZpUMbAe{NOD}Mxg)(Tu)D1V>a2;f;;V2|n?jwDDwrPL0$OhIS@5L1yb@Y-w@7L<
z)|@A>Tt<|?vIFzOw_bd_qix~#g4yrS=1||V$&g5#sV52f{6q3*U^YF_?Co`~Pc9`5
zgkI|1iW$6l5LZ}dbLr>SRPmBNzrcQ9&B%?8kW;o2U`QM%g~`$X=?1|HHYl9`td{%i
zS)1JY_2rAqNvqDjeXrLGT>Rr45rzD5dOs~fRJU!`epD6?PWhxuvK3aPZn5g;InE_W
z8-%oA)$_F}9J`LH1+Mo2P4P(j<mu|CQ2@dXQeKqFu_uiHNTw7(J@ciFdtH=W7lI-S
zLM-?P$>wJX8zj2dwfdd3;RA?QXGD=WXCbn7;}{Ts>b_f&N=`0`Vq|wzo!1_2H76nm
z{mp@bN$2aEi?>T3_IQ7eFP|%G?wwVsz6zWqOCrO?)65j%X<4+O+@5=mRfM4of2CU+
zDV!3u-GD$3*T+FK<kvWOajo9lp97~i4O+Hdh~fXOm3S^ApUy<TYHiVpD5AS=awnf)
z;1)%0+MUY%hiuV`x4yN1b3csl9<q$`?*V`Iq1skQ<Biq_^fRqX8@;pEA=U?hRtLQn
zlS;}mPbocm*yf+@*tR&WDjy~TIVWX)LBV*3@LM7&JiakY21n}#x67n0B3#V@mCFLJ
z`Zmr`Bp>5W;`B=`8+okY;+&BJY~&A9<l)_X<HUpC`$puSqXio~hkMJyQziDd3sr!N
z-Sey}$DMR~4CiJNO~ayUrxRoc($SF;{)l{Aqq5ZFg^Kf2ck^&t0U0Oo7T?<Nu+fq+
zrL7E-#S{3efCRTxYd7d^bMU56g@03u2{&#GMY$bnr6=AwVjcL@OA88%@9CN3(d@<G
zB33M94-35zJdEE^x17Z$L~rew*~dm)yM4`*3zF@73HBgYkCM<#;a~ws!NKRq^;d9s
zOMu;|o)KZExm!(uEI{IUiK`=&YdfHu9ilt666fn5HK~YhhgBb1n@iN+_FK6LoAufv
zwf?-nzC$8(_1FZ5s7jB^5ZIKL>u5cu;sWGF-j&U*Ev2WE#SKnC?oyIkUUXnBx2Ptt
zDT4=hp=6auWhic~+C^Lqv8q47Hz)cqkUvY;cEDZx_j1=FOZC5|_S#RIZH2krB1_Q~
z=rU7AyHOs9PC<s&IEHY~;wMigZ8|OdJHExFDS2){?9d;UkLexL-|(j(9(~2Pd(yqW
zYf8yHU^+Q@^~HOj;Z;dy=bME*N9ss%HXQC!`B5dKcc}@dzaHB)!dd}t)d@slaBkBC
zn4Bf`jhzC>VV}$Sv9O_S4O3q@gd}ugluw7CG2v~!_(roB>pWQ9q?G+9{RT74c}Q(0
zNq8?oa3`Yk;FpruWLP^`INz5<2XQDoWnR^~??tc))Fr&O_|R$6NIR~CLBk1rcrM@8
zm!ItX1-P2)Wh;>5tLUr~w&RaC+}uaF2q9DM<A<w4VN)-7C@WPP7r+=u2xjN&eHp3F
zg00uMT-0p1f)xp^boH-pCC)c(Z=p4oYdju~`LEVHkjFAt?oO`~bT(E%B8hjbI0soj
zoIPf_YML#v#3lD(A)VF|;BR;+T7|Bq987IX!0}?2&rz{sX+#gbcJ0xB_^(nX@HmVa
zi;~F??4UN2seaeT?w%n7NEPfszR2Q6^Hp2C$rq@1@>X1Iqf_+Oq1LPLht8LFvEcN=
zBA|PFd+Me=#GT=d8~e8D+7-YOqU?r9r$4SA=8M?Q8;dMXTr!ohEPr!J@bqKJKDMNW
z5d*YCaC6_@0yIp6>qNwi&xvcApR1!yih|RR%RK0>M%#M+`yt9<)q-iN?)6VMAdj(T
zFB&u#>#8#+vmd4Yxsq>oLx1?omw30DkSe3I8nwOT*iXdSb$=`B%Gt;%3N7FUBVAF`
z7B_5*Dg=rswBW(v-3jq}*oJqAF#0Pzniw%*8Wj2R)vLBn$bX<gJ6;eEJK!Gqu{KvA
zPhn@mV(5cp+~(^Crt$>Uuw$rezEu(DEq6YgL9OT$-cEugKEt@-46kcFnIG@hf7Z?j
z>LMW-Gks9H+kZ2<iOx{W=kj<;Ge<y2(J72B!tfNGaHW<`+Joq$$>7m3l1$T@0b-e7
zws{GQ=P9<P=V<*X?XdLX3?lCEKn}WBQ2fq$octjKw$10Y|5`A#!Z$MTdc>4lgH&6P
z+*madU)z5c2vP;QV_!h`fPCDUD_Iak?3yHg9Q`j2)8=OkOVXbV@xyFe4yJUt=Lnya
z7Zad^e)Bm=7xImd;rC*(m(KwUbv++>S4w~_4-}D9TM=|K5KMviiujGj;GW{jgxvV+
zbf(eGuZJt*_3&wCUQe*xVQx7x0)}bmu&{RO^q%8b9<Jh&HcIJifU;SrAZRp1Hn^M9
zOB*;6<XRs})T>|m;dlP%5YhIT!Iki;ZGo(hh-_T;!B+l>Xhd0w^8R_9b;{X(wxRcR
z9ax^z4rn6a%{G^(gt2Y^TD}Z~@Aa>)Bq>^l6-tN{l3ZuU^b-2@LW+OVZqkd8YzA(R
zGvfuX^N?MNcwc@1^kK#uI-ducK0l+$d)w6+XHOuH>f_7B1jTNq^Cv5~(>?w(AdpMo
ziow~8mS6;9-F&a`7l)V!kR*l_CY455E6Cpim2DhhUqeob^l-69#%^}2^In;-1ykU%
zz@~Zg{+>>*yKTh4=z~-h^78UtKr>|nZCk~_3E6vrk^l`n-`i1reL%9Eyly`b7yK?O
z?i%5o;Q8}$>?wMxsOX}~eaTZ^=o)1GxZe3MyS}=2X+)=7JQt|;X1&HLWD04nm+|kW
zW~eo~r{`=%G)3<0KNj_pz|%sv{^S7kqUJ7i`g-h{`XUWsr06{b9g&Rlt=vndX3jO6
zQ%b^>7*zd<jD?nw>dOm}u>&QJNg<j78avfcR51s{#v?ZUQ)xh^eC$O5>)UsBRlgBY
zQ%y%{TXlJ7&jsxng3Idu9R>orx>qyCuC2vDinDuPM$e<VBcJ=d#><OzeL8v=-X5E9
zPgUu!QlF^SJV*Vx7)o(On}1{e`AtJ*U_z=XVu)%64H?iLZL56)UNA=Pmc2nBoUVyB
z3|i{bPY?M$QAHo*L6e7E%v<m!EiqF?UguXP{Q1Q~9BKw74)LXI-?bVCAvPP&*gU=b
z_X`<sJcdTJHzn9kia#8Y@nA<gsW3_Tgl>;5-wpl!BOCSL;yjgOsc%T)-dxBjIt<&O
zi=AC_-J|=khc)WaOx3---04PqLJhtnQ(?K;y_%9S^y}vYyNsLZWkyzKk!R3=1fR1_
zD<~)kvh58Tq{o8MSFl;i-r)Pu?4)B!<zfr{aEAREJ)Pi9@wy6*Qo7E{*H-*b2R0PU
z>fvMjfgEP^RIbO#>h1_uG|{HmIt#Ljc2^g8K~OVS<U7$3PQC)J$&wj8gEk31L@Ov^
z`SV10|72v@4F_&2VWJ7JoR2-&N5@xFyaWCjCu0C$HKPq8*4{}mOX!!APj7G@)LGK4
zf4oM<!vrj8JH3$_LSLtSV{S42z1py0rmgoRjAicyB`dL9GKR-LX~lel?lGHDqCn=#
zXl4>>ah)FiGlnCfuu>(|UQuc2K=Cckj4m74Zko)!V6||Adj6ZTee?xg|L8w3KN^oG
zU-6p>lCi(lqRuFeLnS<}Fdy9=l!`ehm2Bc<3({Krjx?pA0W;MPKKT2AN0Y&`_{9e*
zw6}?N#l3VHdFw-LV`nT#!q;}B7%e=+1O)-7f01sjxVzpC!MNv)H)|<oJCU4{;vNP)
zxJ@Z8-0FQ%r-7S{qw#@Iwm!#X>c`%Qj$+09-R89o4SAL?kACeuK;53ZTtf~?0Q>%+
zDkyhl;4OS}_1TlGKYLd8Qc5vd_xg=T$Fys#AN1Tn<SBW|%kOTj)Uvvr6U$iQ+4?p0
zs8c3MA47NVcvqw;4xH0|IzG$H<B@GIj;l@(5>N3bn|FF$lqphk$(g&%d%Dm5^sZUB
zpGCmW%QCUez(T2JGX`WW0*o2Q0T-%tcPPsjU#Y!~?R{L193EhZS+CCQWf#5+$fL&%
zypUJ?A?ezO$xHHd2q9Sexb4K3kW(2W{duq(styQNJe8#W!*_af>mP_xza`~Zz_4Vz
z+$Dz!@-Ppu>NOp=Z^`HNl!y7d+-|5mxv@}mK?E)?u(gIo-lrTB+t)yz7;nBSNYxWR
zeJ0tUARxQOeWLF+j|rW6Zez5+8y7xXheHjpVTxFis}JYnSRyLitc__RXO&N+rAfP*
zk+_Q!*U+r9Q@$v0`my3z)xl}|+fAcJ>^m+|r)|Y1h4oXtX4{V&4f6o|aA0RgU3mFl
zFj!@<{~~hL8Yf;SyHYJtFsDyv1Fo3W|2*6*^f$gS2I?!m=WowU^%r$qUD`L}$k0Mm
z9#3YL=a`N<U~HoNTu407p(ILJ5JX>!()C5{*6e{xbd-6q@z3KZx>;C%E5@Uf$fIe+
z@Dww)=NhtphDO8ZAPicytZm%tp{#BUqMDD%mfV5lz;j@&4Hm>Ey^A5W(_CS!ZT$+H
zr*tkr-5F?R79gZCbzw2lxJ7+waR^u>*j7__j&CSIe;6gE7vP<Z?&e}pgm*7z71E6<
z&0oG~b$UFYx3!W0V!Y`e$<buL)PXq41%a1|L{&t7@MhTR$PEx7xkOgN8U~FQ!#UO%
zoQ)z<{ke~y!Nko9M-KAg(BF!We}#?=9Ouz+n>$x>XR@KXS?*cUa1(v~bj(ZPtFaZI
zS<1WPHv3Ch^SQWCLM%3s2OB=Kk^Q<MM#@*!)$w~X$GL1Vz%l3E^78%&hIP%DzAwr(
z37P7*#J&Wgl$7leq-Nlt@aZYmWpMsiizx;sQzyPiy7L5YQ8@$57OMB=d@mTx4!YfZ
zg7RpmVVi|V_HZ5Ci^kJAv4ksd=0KOie>=>kR+T*I`r_0o-u=0cy$edpqT4Vcf2G0s
z)Kl<#ZhK2)j}1tdkIurTU2Z_=_nGa~BHc79UBz8u@{Q3U$6Mk8V~=-nsQ#YA^@x_R
z_8_ZKJ>;=|v<?%)c=~j}d7uhsT~GW6;?}cbOSJ)^&;$;M>|9;ziF`X{5X?|I(?QJ;
ziRWZWKUMvT@AXX2xfUlqUFEtN=fb{dnE9FMnkOC9PfLFOyra(-pmbO{a#L^u*r>pL
zH8I%3<<<r2A&#>9VvL`qK!X3Hmkn&659(PeJv$YGHR)EFaOT5jLoaT1P_pCwrgdS@
zb`Ch%@&Ufw{?-*mU0&Yecx+Wh*XII>u~-KggTIdXu@-M|ba@X*@2H(@9svTF=@-B_
z4d{z%4fgMPs%q{=iN$li*}V6(fJbe~=N$|gpu8XTGt;|9e*d}cwKQqB0|r+aY^5rH
zrRyh*W^6#{!46BW%;UKEFZvI`=yv94U1E=9wDafYuGvy*ROOVL-ZV{?8t>sAP#AMU
z)Q6C+l9?o9pgJV0Jj;1XvE)~6-V8lC_A{KkUx9_Yxq*vtk>ELZ@zv?RsG4!T8aQ@o
zyu!mG*OR>=0`ME3*!Dp#0MHW-g{pw0=Ui_DKHFa<^=@O}^5p(|3(xj)s|DIgx?;Ui
zx%Am-xC$g{iM&GbsTE91My_4que5P2nQ#e^@$G;FA4Xuaq?w_9v|4un0nZL(9)1fR
zYDw43!_d#wj2D&$*w>#ZKF91UT-u;9>5F~i@Q+XH@;tdBAkG(E@=TI4#s#@Ig=@Zl
z7;nmki5cRMj2V+olEVq^n=7{n9=dX2-kqR;Gz9$3!Pg?uP1V|{RXp<bs9yaw2Q0bm
z-YG41%(qbj5L~qe<EIxO{b9oRgBHXSU2)ty$Fn>`zYl{?(21LK1$Ies2+cK$>=Q$$
zTy#(>yX!g>zo_nBE1@glqay#nC04Zv?C9CG1!d?GA>gLaPJb#TVz~mjG?6Q*PB-Ca
z&cZ(G)+~wMfaI4?i)wY2TCif}9Lvg-5P2kvJFocl2X5Tit<<jt-m9qoirR+)EEVwA
zn*?*@NRFN?s4l`#>HPDFs2VI|Gvt|U!?qo|DIN~QdaZ58Glzjr+CYZ7VDtps2RQ=n
zRQyb^kjb9*KhOFDRm4dHHvNjXZ0YRgJ|adeF);Jvq?SCWksJ%1ha3u)b#c}vcz1>i
zcFp;A<_SCX0ClXj1-W45tPMK?INUn-_er|LC;W!?yP^rqVC#XeZ!O!We8X6|rE*%_
z8S+%j&c#Lk(RGtW3V)EP+zR}Wq2fn}&mNQQi8GaT0jXToMXCEhEX7}+3e2rv<4U|k
zFLJ<hyqTgEJ{Wu4@ed~t4Jo<OJ$~8FGm}R=QTLNj3CFfX1)Iv@xCd@23uyBAU9zGL
zdIy2Ll2y7)0O45srS&Ckzp<lx{K#FPvCBPR9CIP2nQ_Y;yFUB?+S<&h+v0SovO5~4
zJM03zI<(t$9eqHYwFV*mu25x9I#t=;OU;pE3WL7Ql2IEFE_tYLy2rR<z1ZN55U`2Q
ze`$IE^<)j@VyZ_5n%^F~JJ=H>vE(5N!fk{qI|iBrs}CChM5UGjNl#~&sR<4UklwE~
zJj$(=b}z#84WBNmvKhJYFds<&6s3?6ie|wr?*}l4#>dt_QGstd;>Gza_vuU1WwlXW
zw>ucyReiUH6w!map_`XCDi`^Clgs|=e1#swR>-8>-Vz8VaHppqykZ@hHehX~DtaJ_
zZNg{I*3K*y@sgi~Ssx*s=QC1@J!dfjWV2)IiRBPUZx;xXT%wyG&vZmmu!(-#mfr}6
zSWMM6?wL)V^TkxB1QrIZc1PJVj5s!RREp7y4XSgUzwG&u*GM<zqUR{GUkU1h@1Sk=
zp0uh6+kn<_R=AHBD&pCZTlbR!Ch4jD+WfP<W!3~@LWwp0>_`=5eRoe2RpzUN-tzpk
zbvAZIvMjbaB!`=9(7GfM*xpKvlvO}`LrOx3AG^Efol?O)9E4*3On?7*yVrcy>@5`$
z6Nr1g!QlR&F?>BH_($|%{Rf$G*Kme+ooZK61DwqZS}`N$Xd=w{ZUi0}&DFrjVtF7#
zL`s$&V%FE|OvqfHD4JMZp<!X{8*_a7$SrS<<vhg!%=LS-rgq`m!1WO@X7!ww@1(6&
z2b^|$!%j+L=-J+{9LA!ncbNX|kAOrq0@qsjX6MD#U_?j5QTTK9&F_y7K5aETL^N_T
zQL{S$mZJLYvWwV+^7d_dpKX?BL-RB(3p8;ahYI)POu9;HrZX&DSo|nkmWq5IF20yw
z-CJ|1cX4j8(xT1b1MJu5N3JbP2!q?c*-D;6`*WV%%|ew=o<N~+7=!8E_g1bR#1PVH
zoEbuR6s|CyZ?~@(Br8#sjBNkRPl0ihAJ@6cfB)11>ALGy8A9-8AGkrklP7}(_Xv`Q
zLS(?_@eaBW=GYL)cVpGmViqzKnZ}O`0jD;+Q}f~NG*3M9-2>mH0DcX#-Zl;D<Sg!f
z^w#cn+z_KFL_P*Oov?5P_QVCtcYON_Cz}AE#G&3lv~ZVbdXx)>S8=isk)zHb#fF)1
z(;N)Pn%P!;|K@?Pg`M+@%|6;{JI8<xt#)i_YJY@07}NZx%tm7{E%?c-y*dp(kSyoE
zAO+j+&h9hAXN6bPd69IHVXfXileGq4@o_P%bcVS;*b{>4){amDu(u*F2Lz;A^Ai&W
zpLsNmS*et{JDB18xAOM`ddyqS=LmKr1w>2=Nkws|`yC`~#H?0y4cOPa3IhU$2Xp?4
zgG@ZEsWOJJ;|dM&E(2zS)*x>ismxc+QtF$Dc|{dFJIuHaxA4g)jWByc@q}$X$SLA3
zLYt*EsPw#vtRH&H1{t8&EfT0x)7UKUt!?G*s%)GHAz{$8mB8LFUw#MtN+P|=H4J<S
z%4Uv;-b$oWclU(toD1N`GcYJvaL$D0^VCBhgT|+ivy5@5p+c+GQpwJ88Oa#?oRR(g
z{dY0`s3Ut~CtHYrW9`viE-&&N#|J6aXs=o~X>04kb}!GX>kT+B?>xdB@Km>xmTp|e
z|4tTOo)Af%Wybp6?ANa`l8{a|I6h-7Eg0w$q_Mi4wQ9h3w|s?~c7|H+UZ{F8(6j`8
zX~~=F*w^?vo`0prjyJ5cU9f|3N|50+?6i!={4R4`YTFWGuR~!hMr9pmwnPj;Z)92D
z7~Xv%*iCPHkkP#ModL)Eb5>j7LHX}5>87$U<~=GUG0LoHfsoO*5UjLkSbsu2u`;l*
z_(e*d-ii<QCHn}LvYb{5HFx$iRMK>lQeb1ii|~4uWtB?NTxHp#2!oo7`(FEp3Kqsq
zk1?&<wy&{v!$eeO{QB-XLar<zhv`{oNwU_k`r&7Cnhs8I3{)EB?_&w-X~qkjs9S+&
z;(7YWI+O$r9bRx~iN!&qB#lcHu4=mE@P91-!&j+~?Xavx98AP0=9+I7`y0$ked)I-
zpN`BpdU|0y)sk$#0Da|EEEXtab_WkX90lRJ`e3PPF4#jaeK>mjlP>MalWr^<-G_VW
z%oBy4YN-JN+tVQh$SMt5<@ivP<g?k?E0ym^#L5qf6+Z~Sf|V?!Wpk(8phxMmvz`1p
z<&Ty$i%)m}O)x!>D_yi9f}0I9Ku~FrHJ2R(4O54wn8I=&SDh6J+PF%7<z&XUaE_x=
zw^52Po3{NUScb9raw9xF^EvmL-lDB)(2A|Sbr@o3IvB?4_E5`5V~Ae+0sG*RIZ|bj
zk&m8E@0vx@=XUnW-*Y>kP1yYv+zWd7sVUOfZ$1CjsXuvkBIQrgRa+MmieK>f>Xhr3
z{M1|m?h{nWW~^Cz$Bmf^gFfIbPLs~NP<fbES28zK+^eb{TRTRI&l`Nn5NSEJ7!9-D
z4%C*33><R=w+dLwB#<y+whae~(~xPJGPGuIJ<V-P+TkFg*8Gj8rcL~m0uqRP9D&g_
z<rV62z)Gly`z`dx-b7Pn5a%r*{22sz$ko}mv32Km&CJZettBg5#^D|jO?|aY&pxy<
zpbiTE0N*9?eBW<^6mG=8g`d0<hC($BdD%>N&$3|cqje$MPEQ;;uw~H|HMm|@;k#NO
zhE5ElC)Fz*`fc=5da0+&Y)?hbubnE`4Y*q`O$9pEI@C@*pC?I|_Y4|BWSa`1{vh?U
z5HS^rJ)_<nXTcDZSfj!V%!2i=1Y3*>Rask&tHazSQf13&UZhXB0lyXJ2!9D*FPrCV
z+J@M*fXdhJCDE$~zNsodWO~jo4?Ov@6n>P^4QSY#(6tYvNz0;-C1b3Z&YY(dVWMFQ
zVL+9u7X3AN2op1V<^uWYb&-e{+-<BUo85-BRhDkhp2+9nPr8aNG44N;jCsBTjC5@b
zv0<l6%vjtSiX&sP83pL+;02Og7Qy_)G3igGo&|a<{Vq!(8nLmp6KBvFr_falYdRZ(
z^<wbU?%w4x$qtY2WJ$7=t%oz4RcA7fMeY!!xYi-_K_r$ucE=2Q9V;`A=y+m9GEM|e
zo4|jWcn$>L`_F%=8UH9hP6TgpAG#GD+Zb>d%OL_<=f^;FJ-7#G#{5Uhry;7{Jv|yO
zUnIpv4oUT1<?Q?YDx2}ZBwRmmR2#tv-U)FRC{i+&%FE^R;n?5cGVl9pn##&mKisOx
zTcCI3T><xO06*_}`MG**v>6K&gQm}Txa7<}8rP+#Lg#Wfc(S4BV0S7s&mW%GXyfEx
zvt(bk!-qK`Vp7taFDOrY*_N9B4z3W~Fw)WXzg6p{=7+ZvsY4J{bsn0BI=<cNhS#<f
zvi<7}qF3`s#O*WGhJu&n-PSmZ-ro*>b8}Yf2}>_5J+i}N8v@p`;rTPHQgq6|qg7`O
zGEC$GfXOe&+(Y*{A93{w=9*hU;`+C$5#I_`dEc0xvGUA%5AQ8aRrf|dePGBhW4DBQ
z4AmWQS+H|5O$7`&?;o|DSAtHFG!18FXY(pO#&QeUeg4(zKM;)}s@}Pi&WKtXs!8Xf
z_W?QK9vU{v9-@L3d>~n+nEY^J;>qNdr86cN+1{ZUnoIcN_LG1B>xmi3kF_{gc{?`(
z!BVa(<P?qzI7j~M-ry^8vi+g|DfJ)$)>cJmBfzapJkrC*SV=Ay)kjKPQF&2Xihm5<
zR(c{gL5-z_yQ|ZiX;&q0_c63nrt+&CakRZiOAC590Z33a&S&^rVg<b9ln$~sYQ6jo
zYiiP~(Q&<CnI2J%HP`OLg-)kd2xV%Yeb!H!b25Njm)XWzaTo|gROSH`EiQPqnr*+o
zeaAi6|5erWOK9msw(zK4-qUm$PPpG!e-l_(aQzvlsN<$7m&gC4rh24SPi7Iuxzjfw
z2yk@;<la8tn6h>bmwm}Vc~?<h>Yab!Fx4+^9vEGVX=$o4EbPN1Semgfg>JQYe<~=&
z4!6d0iIy~@R)HPhCb$OgA0i6&&+|$SbpWqFjouDIL)u;nhF%+=dhbyj9d&C)^&dk!
zIoMjjTgI;-pRLhu8Pv;T;gc^mu5}xy?a%OSjsX3-z@Ks@&&<&VtmEtk`)88OQ!GNq
z$<Z;FSTm9TXG3LmQunxs{x9{yzPuK=d>0+`sYWmQ0q%S*`Z^uX-%7bARHmbp2uJsJ
zjhLqy;M9}NCjaZpY&^<b5h_)9wE_7OlF>(K=n+>t&m7tkh$IDgCDgO^^co=iCni^2
z$vsKMe?BePF$W4FhT>mWs4oro2b?5&z?Ve?$0BxNqwVQvlN1G9$JNFg6fM0c#b95t
zn!(>Dz4UA8U;^7pOl_tK7x7-dfL&@`_j=ky8&t@&Qjfk^GX1i!+GpxgR_%_!g^#in
z8v*cML{$wz^3r`uG1&|M8>pk5m;_J$qBhj%I%AV)Sy3_TDw(thk+XCetgZxNh2fLl
zh4E1{^G^<+6PbinGe6N1Ow4urPL%6WXssS5l=$nOZhW8%BxrM$x?aZqftWM>$Nf|B
zq$#1qM0-f`@GkPXm1qYQ{$>_@*Hkb?`9fHGdR|&oI`|0aerO2Xd}wyqU1@!r5f9wS
zf?2Mz>1TsGWi}KLcx@Qp48_}5$_2(E#Y=Vv%6=~Sm?e-26}fPHtJa~Cu=M=$K5#Ay
z{1h9!%LC6qD&a4r0jq?f&rXSV2;tF_JdkG3h`knc^jXKMfX%w&4=uFzXU%M)oPAZc
z!gU<Q=fO}&Az%cK=0d>X3YC+4gH0@tCH6$2j#y9)SLqU*Grc7D+iD6^k1~3Oj1(Th
zBv(q^$461y<A5MWE2svlw*WMi$78%O?rWzVU5eCx_Z17y!&d7g^GBIH32N`uda=kG
z)d5&E);K5i_q&S8hkRT_^|*v0kV;o&B)aI4&2{owEIoupP2g|vsN{1l$lkQ0Bb@g-
z>wDA3836-2F>Dwl`p+*VrLXTsxW|0^eL%B!yB2vDw-UJ4#2Nsa#Zk};!Y{Gtd#FbJ
zAC9hqA*ybR(mBA;15!hWv~(lgp@4J=AKfJlLpMlCN{DoK2}7qK-Jx_!_q)D-F!#(k
zJJw!%t?11veLX#Yw*9Yj)X#v7ujrrMl5gL5@FHfF^qDHJ0D@1jRDZ?Yio`2X@Z8*6
zGnO_D!0lC%Y$t)bsZ0E@t?4;b_^9S^F21z`USIV$daS~3UUd5T`TPV;@nVct*801{
zm44m_AoN;Q=>8}x#c~U<>iv6ECWmpftxoQA%2#7oaOqJK6X@HWRyE#r98~r=UZc@I
z%<tGe*Y?%7JJjW^IxXZR$UUNqZAoEJ=c_II@n#L6=u!UVHP9XaBw4LQQ)8R!DzDDN
z(vrM@r8Ec26(w$7>SWuu5iIWSXB}t1%hosB%uRmJ_ntf<^xVATb;weF1SC_=eZ_$m
z?Kse0FudnIFlkPw<lM{;+pIUhO^56I1K@9jLx<X7%bXV~TwHr?7j#&^0WiB4fB!BT
z0RI-w7_tU&UVT79&@0}aR9zgPj9}y1_7XkuWAhEzJ-coBK)Ne$HMg%*I*Wny*RQ1U
zZL`;>KYSr1?sD&!UN@8ezlcC&Yi7ons4j2b8BZ3ekpRrpDY`4l1AX(11+3RSrDe->
z*jD5^a^5IEp3QvzS^|)Uf=+*OPV5@<Rg`g|_6b)jaW~?#vkq(lzBLq~rZ2+icvNF=
zYZ7N|`TR5YRW5qd3M+J`$H-2w!Ybc`!nN=3a#;Xxh$R&^8XdMH2kY^{i#EsQwAOhC
zsmii-x<Om$BV~i`43#kK$0SLxGcj9;$!Hj?cU7Rf<{s4<-jvoo@XzKF4pi4MhmQbC
zL^$u}5u_o25e5ewdh~uSQHN~2hQWc&im+bc|Ar*kzOVuN=pSR=8>~u<7}hV2F;=P|
zU!;39YkNj4_W;pMrvPc?sLi=;Y9y_nvT=^jN9||G#CN5vKc42Dwwypo;zP7IHepNI
z!f(%{hSrl=NzFh-qn|}x-TV2Aw!7Q415-o!18SOeQ!@J74G-XaBFjR6-NxmGsX$<P
z|G&X*hS^?jqB*kn@stt)hP!Un<bX6?G;cV3u6*KB-qAXAlW^A7uEV&``P{DaeOV=f
zKqa%|H)rqCpq2x~TP25O9;dy6JAx$z^gA-A?1{AYHAWH=l3Xl+v}jhjtPkJL#jo5g
z1E{v)^)=UdfA#$N7lz?37S@>KCAy}-n6^&L9^m7$&#Sd6vi<6n_R`N3Y79LtI(Qe9
zu=tIVuvpjXH^14c`{jNFXfL*r`k>ahk@S$*`BNDvH?x)HF^69hFLb_jMd?e=fA`D2
zE37Q-zvh}6#R3Wr>S9e>09Wwl57W=QH)Uq$|0X+tYU$s*-rBhDtb=|N&I=q^h(xae
zx6K*pjWbJN+vl(TvpspU7H~?k?Mpmf+7(zQwFm1u-#hu+_GVq<^+qQI+lvF~|H}cK
zCDoDF(3}{K(>){;_v(LGgVtRJDbH&IOp~hEc5n%S0MN<^-+!8O@VO-$7`REDQ)>ef
z9@^ozs<EGA#Aqsd^Qjb<PJjwQI&u*UvIfswIe#V(LQFsZ)CHlLe7`C;Xkrd<UPb@7
z=*=Kve0fv78v5~aDL04DJ<0fHPsmOpbDziQW&dl9=RriybEdw6-EQ{|hz&!7rh>nq
zj9drx13nh#y;gu6pm!5x#il7IsKT~il~3t%x>TQK+EPBEUmvc&UUL3;F)8pYn}340
zRQ3=GII{9-cR{GZ#TPqbG-;L;!D^4#xCkR}xHMAf_6B<+`p0$BmQ-%wCnra+GP&}O
zPd_Cb|18?A&g>sJnfjY{+M>^~ZP15Vy=ppc1g6u{%#Wz$eLc)}dbzrmbH}uJsh%4)
z#Iu9v2zpR}ku?;Pb9Ww676b&+-NtQ+>@A=1G!Q`0&;P*|?vHj3uvT0fsjp0Xbe8l>
zV?Cz_1@FJTnsN1+lw|^S%1$mjN<W>jmX1nw-6h-D^#IL(bHjTPo08UZ{ZW7)V7l3`
z$E8*IvUCsxJ&(tiO?BQW1h!=v%1@ay8(4=fvXmBus%<x4FI%t9=HzHxT+CwL13Kg%
zKmNous&_f8s)hu)0+$UyWAnRF1sq@AfneL~Fe4{}1jpeECLDw>z%7x~`TVIEIAFdb
zk(Yl)ZERt{?nN1x27B{ILC%Y85V(Adh!xn@`3ac(=;&~F6vINTcfLbT(y-@EMfMvV
zCmnkqtJqi|6X9z=wb=H-AoV=A38Bn3(OVDU7J0i?>2;%aPrjWIKqc9TE_yN;f<;P)
zzP0_gV=9JO0N^G{-qlsGiX=6<J!99l6~14>Ul8!k^Dyz!pR;kaejgI>{?E+Y<T;0x
z=VvU~BaaQjNdKo<juZd_yMN`neVq`|jhcs%?nnbZY4BSq`=(>C%sYR9^*k~EDmOG&
z257aJZgvfJMDBnA-T<&WF(;~siZbj6DA5Y20kf*}2B7B*!`5yh`1lS~&d%l*-o6E}
z`1vl=qMH9^d#si-xC?o|Jw%@!o0zaA>A>9?c9%@>?yp%=C^x^oJ>UK?r6lYb`*Sx)
zSkb!~V`?`X_chjXhv3ghKyQ@UAn=8$LYM=anij9=osb_75mocaq1A$M+;v4kV$p2P
zpIAzPu$MK*_cdwOzYtgZKfUJ%5Gvg@o~4S8DSExH_A6zKlk<+R_Jwf+5cDOvR-pZA
z{yF0*l+P|ALPBK7=I>){UO76wOZF+B0uaux&K=J{=8~vO^?s{Jhx?uU1PfRZvNONL
z-NDwC1k8SQT{Omlx<2>QPNp@mEA*4jjD6BQKVi<D^09CVvCX(X?5W6+0N4t*#$4S&
zCsK6H2Z}wFtF*zqrUMAx=9zz!%LC46Tk~YXVV3XDGJLEooH2F%W)NtF>bdlNPL3&-
z0g%s~7R|bmo^eFFOdC7M(O+9G{{E7yitet>H($<V4@lO1Tm>rpy{><ZsS<P>vN#4n
zUe92(E_Pn@XGp-?Wfme1&z(V?sU_RN?y_zof(RO#lSjZpwh1fpz0kLxKLFvZPcwjp
zu*-F&lmDS&81^WC%c9a(A^0nFoSb3}@2$TU|7sudGIgovruy$yr)|2*|Mhf5^FOMK
zp3B(Zy}Z=ia7sEF-ed|qI+;5DcJ6~RME`b{3Bh6*{pP`Ytgd7~T-c#j=Te~%tsL!S
zkaq2ROyBOqUYfnsH!+ty)@3)(=%k}ne5BfuMcnvtu=ag?X`hSZ!-vA5JKAd1n1Zyr
z*Hg6KuUy}Bm<p0k5I^PaUfzX@Z6s14tj3|8J(hiHn`!}}2?|;}=U#{gR&x7Y?rBQw
z?u0f5a8bamb_=tAG|n3v$k-qXhL7KxPFjExQ5B~>RlW+VU_f0qeA-nK9$XahgzCj?
z81FaM$ls(0Pby4}zQW&pgxwq+1>JeN)Q?SwK~$oPEF)_OFZgL2#U+XKre0t^Lbfro
zacPJb0hDh}ck@wVz}IP=>7Cd7v^N>p$jOX;5~?wb7&aW!iza`@cAr1Yn}}K5Uv^s2
zvYND|JrP6qd)ltF>$pOT2qobCJoLt2)>GF$Ye3WxRQ66+T7NBf`M2kkV$R3Cwsv*~
zH!{r%*-uJIW@Kd4>>7efZ_eO&qyc|)15l~kiYudAd1wb~i1nG(=8$CwwIPn6lYkpo
zqA;Os>>ItL6x&u{dF%qJspkq!0T&dBh=|OxAK5vW9o0HxY;nnKp6P8K%)F()X2umD
z>uNR-l@Q|&iQzSEdV^^NlbO7olH$dT$NbaKc$J+mXE^2i)A$Sft8Jx2b!{1P#vOb-
zagiHmtQ=ls$M&;M6}Bt)K*DLPqR#GK+V|JGup;cD{;`RGUze|trKhFl+p@va(UvW@
z`S@SCvccdmJ~i@J*}WN$5e9mNMJ}=AaDtXTQ4IJ_orJLP7b#VxUGnQ0KEu}}c-B0b
z#r`fqEx`&indA~B72JjbU{%oW*#bS`#;*wkJL&nn%>V9LyKfu{<>HN4h|TYJ{q5$f
zt^ikYKjMhg)>a`r{}$tLgN0aRemU}9t^uT(uY(Z$Cai7nHiqmbBRw7G@NKP8vQ?YF
z22+^X^yAp%X`=nISnOswX3G~Z<<4O2YG1gT+hs<g{=$3R8v$q~!bACY(?D5$WSN%2
z-wOI85nwHKKM5vSV)WPE&j=HLK1%t$kui_M4`T*d>tz5EnR?A!E9N^3Z`qal<m^7b
z36d0*kmy!t{Yn>dSLudRr__KM%>0Nbtf@WO^y(XgE(eW~HZqCddvxyIYb<0sPy#L<
zUOI~5<?0?B{vRt!5OeG~f_Y6_L~SIqXn-U*+&~Wl3hNsrV*LD+BW(*6yP{-@hw8g%
zC|os6UR+2j5Y%*+QC{m75lf0~zLqtUR){{!>WKG9kv0=VoH-}1!Y|d`lphkR@Ce6e
z0#31ZUccAM$tgiWD-t9J5vNbB36Qi!i{HFs9k+E00H1!Ej?5aA3q6k0rj5YvIX}x7
zT5xyAuTdQ`axeHsHK3p?>67Pd*<kvBV!y<cEu?Fd$Hh%J4!b-!umv6VQtpyB?Y>bW
z#{@65-}hiMn=sQ$8GXVAeinuzhT6ouw73W1#l;0)=H};ft68ZVN~mJ*a~a8kNzx9w
zn%Oxs79qjF#b0TRQiz-)lVl4=kzBBF?ZSX8u#VwPP904Sk4y;H)-&oK0mNI+B)jm#
zj*%xOCa4Y$4w4zP-&T&H=1(}Q?@l{nL&*y8P}XZ%KJYk@UHFZ!H`!+18Ck7o5)-s&
zuGhK91^BhRRk#l-&*Gj=NAySRGrmMmA#M4@No5B<A^en{HJU!9P(rI`s#VMk0-)BT
z-W+>=yTYQ?kX%&!=`SEe!_N)J+uV$hU~j3H(*c|T1wv^BEmK$VtUrAQ#olY$rCAvf
z*d4y`!ks-uXzITY9R@!1(}D>fb#C*rAEskP7(s5|jUN;X`0g|jMK&pF4p;+)gM$wg
z2R&>PrZ=_^Own<D)ld;IkVnkbcs#PmdU~S&sSR@8TwV?(f2dWYH8aQT<?5prd}hTD
zg^v--Vx{z;(cqfQQrfdm_&SNnn+X<!I<~WkBNWNNK~S`AAril)a_9<ELk_n|SghpA
zeEV%d(j6^H@yzqbQYNNZSHQw49g0zI0Aa!Vwk}ucorM%kfS6xiUQT~$3wuLbtm!Ff
zHyyE6#Dewjb5RIIZZZo=!(OL$#8R(9Z7jDII6h1T#@uOL*CZBi`iSgz(G}qW*OTH}
zcU;k6xloI_d~l{Ln>5T#TA|q6SPdhJ@Ol;pv#{t8!O#StW;IN3^@Y<ak>hfA>N}we
zknz*e+{%HP<}L6v7cg-Dxo;*XB%`IIe_zFvx}6mm=Qr$R?%3P@GAQh=WVx<EBd9*9
zT>dvHPel2gyV&n<dggs2!ZF@;9+-^0(wK+?Jt34n#D6d(;A@9OBy;>)_rbMhY4UEa
zuVY?xFF2Bt%Dkx2QIrHhRaJGdi;^##STMq4Q-f@cmI3qM6YTG+?cCXYE1f3y@CZbd
zfdp~$0C5IOmLlEXOhv?|f$uSosueC$a#lHr!W0{cV#beooEj{Hcj!2ap#Q=mbTB=-
zcMcOx6S7s)76!nZJEa;3hH;x(kKgA=CoFlM(IswsLhv!uIB@Bd+UP=RP5thB-M$ll
zx>dOhaTbdF7XIUMdsC9eQHlmD@OF0qRut2|&I|R|9mmASs1MU4iF}QQVt$G4OwYFc
zE9p^tuM+U2c^AB&E-&%4uTDyX5n%*xVND-he5S+>(2y7RlZZvf8bFQ3Uv744PDOtQ
z#!X;=fu#u$CfEg&hW>arKv(pSkHH4lSgk~NGck{_2Bf3vG^7W%>BrZH@}GW$N|U2V
zaN5A_?qyMl!VfX+L<L?Gg%e6MT|u!d3aAGG9jRvCM=<gR2^`>!Q<)UXLb2f7D~pRC
z>LObOwR}PIXDcf!7h0H_Gi2d{BGj&xJ(cKYR)SKsTBzoNkx)=0#2u$3p?zJ9gFS%*
zycpOa;3*{tx(`LB+a-;Oos_93uNxtU1L4FO+t<oTAo~|NTrgZP7F7ZqF*TSK(Cuy#
ziyrxTfC@+pAmY&f-`SiX!l<pY{vJDCmm?P&6s)@td;=O<4f|HX8}8)PM4}(hwk$Q>
zwBtCmMZC{+(>30x7lB*GbSmOdJsD=#^8I<8%%}Xs-^}+yHdZ`jSn#VNJ$uz%V?s$Y
zLFgE)>%A5f>+p9Me!#JiAe;t8aK6S`21(tb2#O1TAC6(u60TWDu*JpI8!ExeTd&Rv
zEfd_YSG&~Iywy6Nc0rQD(5#E;ALi!<K0C%iKxMe8nVuQ;4ZlyW+pi*V4=KsDMh~U;
za^c<a3(;I+jp_ZLU0^C2BYd{tw@V>Mh9HqQiGA<-`c~?t3!n_H+g}GnCe95?#+e;?
zK=GX?%*x;}uofOOu1RbOIY;kB#{v&?QS8`T4(S|lGTMTP5fYF^Q2%Seuwg^Ni;n(5
zD6gCOLg<uwZjrDc3%eaFFtM}M+RMG{$STfcvAU2xkPnaFlq*(MpluoUQ#rFpxKEnH
zLDSR=w(zzei)IzyDuQBUjSGiEK^CFgW0Baoa-Y8(;(2&^ao+h19YhWwwIHT@p%k3C
zV{N%@l819OevI=mJ|+;k&C(on?}FIti!|1=FVtl&<e?kFO`amC$g)%>@8hZJaWak_
ze@#~lywbm@&MXn2ZwzIv%)1U<KL^A+E(&~Bm>15g)fBgW<cRQ+z8=A7?%}@(@oGSQ
z__)Knlok|^!Oahb9hx-Lm+iWI^l&ue;^wZ!E8yxg!@T#S#`d{{@P)fn#Ds1@=E&jt
zUTgNb(eUemT<b71BTE?!Xc4X?!f~rrHPI83s@K{|V_VvQwpGZfl#NV?;&8jA?8~-Q
zkiT9e3hii;5s>sqkE%(6pNMS`H%QVJ(HX8JLDzXa53AF2b4b0XA3uB=9{fF|=v38g
zvz~QV8e&;5eE^#zhsR-sZCcihj7<QaP-R?n8pb09MUW&4vVcr*N1`|<OM$}$k7!Gt
z*bSN##PcEqk4R%=UuoV0hSi;Z568`%cTqAskZ=qGg(&8Vjn*Rx4%czkBo8qH7VxH8
zTue4!;LYuf5B`bPcdlmJ%%sw^S3QDDQc6Z}lTY8uc)lkgO<ltSfOuizL-C?m>*>-m
zDW;a!4hsC^l)2tG=*ROsR(|?-U&ao5+bu%BEo`_r@VX`lIR@ApGh0hkiz4#l0B?%U
z!5sHzXR+ZWYB|CkR^#N$PURMQ!<na^4i38!0L*a!8Gs<na}6LDDQr*_6xBvz2x{t)
zlHmi%AKgOUQ+d9v)9Nqc;f(8a*$=VR|I{}7ArkkI?oLBl694%gMn7fzOntC?9Lfhv
zPh&r@nE@4B^bhuI;|<Vtw7No*f1<fF#^j}T09J|p&JNcwAM^OoOjHLMb=VAMUw=wC
ztk)UoKYz29?lB!#Y_<a;Z5O>@=2$TlGv|#Th`@`L;?KGQ!Z99|@TbeY@A#m!Z@9Li
z2P$zEo}C4S>Agw3!J)2iQ9@W?f-~%ua<?Jgu@7dUg>@yPk#4I!B=}dJIBzHJ`F?dY
z8*K$rDVU#$1=J~Rmqm~I$2<kxv_pjhhee-FW}%ihDcuwoFWq31h{@;<p{52@%s(6T
z(KuC~S{AF1aAmgiNm>7k_4`5+Av1h`qr{i-9B?@ya_T3PGQ3yPi`^cE&w#T*B1fGd
zD=A08LGIE*e_O^3>a_^-^?T5B#|5%9WoF$KNFWO{Qs+2^nh#0VVur{VE0X_cEhum=
z_wyVb=dqhttsD~+2MFxI`{z?=Ht?>_O#1N>5-QEnccaake0QUyoz8G}VO92;iQXrX
zNRpfYh~tF?T!tnYKLsA9J;)i34b3Byu4lMGU;!*mNC+1PogBGuOBhtxG+`9*{+RS~
zclVP_!^(DMLMtvMbM=@kCyubghjHF#sL>C&g_e`4ZIK7t-TtB$rxSU;R8k~}Y|Qh-
z-7H2pzosK-@nu8pbzA@Fs7;VJer+H}_fks_uU^%ToMbmFr|E!zU8bJF%^gi8_mlGV
zF}?Xpga7i%C_w%MdpSs~&d*0cY%bTYf4KrJq^}6Qlq7{5yAP)A=!02F&6?bHQLU~f
zP-0s;axnD{eQTTU+~l?q`)k!+y>BawmQq~_nY{np&@4e4X5oGZo@TD;HV3X+=+RMU
zvKmz*_Pe3}VM{ElEn}?+PHF)O>LjcF(IY-u6}5xO{JYo{2G|>feHOK)qW{d_FkTta
ze?KkwTFDiwpL_t(p~3a?@VEoxWD6eA1rSNT3t9*ghJW&<1HBZ!lYZwiZG(=qP&+kp
zg3OJ3vZfoc9Wxk~5q`TGT7nc@DF<|dE%xnZ`pH{#I}G>bThp>;e^Qje-lILIx6Y{D
z=m{^@{gU)mmtOC&P&U=qw9Ov$S@kl`M$YqbCi~Cu7$~m1kNn^EoR-0K@WVvPl&!Ne
zGkw+~+mT!#G=IBAmB_IJYDz5@?3=DdNzlJG1hrNluaDQB_WcWjni*h%4dM^$O~$5L
zUB1ZXMOj|vAGyJlt7`#vzoL5C?$PN*ZVH6Ty253)lhR`Y^D6Pi_vEF?(XVxZ3W@F9
zMUgnt<>sV-xdX+@`qzE8Yc-|(m-L1i&#wMCH%4Hg&_d8^#j-1nxJh$*YfGAY6K}8v
z3u6Qz%)9?xKbhr@edyt<P%~@!Iag(Lm;-cRe@;%YwxZ_WlLFd>@3ttyH&A={OEc_c
z9;81n*D-!aG(Z-4o>jaXiodFrQz~ckGD!UzC8qVE=#AscFl*ra>WR+@7v~^VR0QSA
zx;6o}UIv1^6M^JPKW$`Js`5+hn|5Z9t)8gc-`~Th1>F%BUf$k%*%k}J{<D0s;y^9s
zd_)1)!}xu5c}ZlXh)8XL93%)b145aZN>6P(3t;<{_pa)G;_s-HSP*k|jzi^A?|15X
z{u>Q#IWJGA+ZHknFU3()J2!EtgF}M0DYU2uXg?JW!Su*Q{~<=+iAD#e#|kWN{b;GV
za(uuptw23}&~xTu9#{B8+u{2$fGYmQNLpd!tA(H(mkXy9yTYXEZ<A|r0Jg)n<gMfc
zrDw`1A`AYc!P>rU-OXKQN_GCo#?F;mEEK_HjT~V$-IxNcIhHVPt#1Xj=mxTL+?1g-
z!Zdfi4B~h`S7LVbEpsqP-)n7jnXsZI5b}LL?l@-07a7@o3^|fW%4ViO%S)P9j~+AG
z>hHa<L)C=L)Ohaz9+lMBQZ;~0$GIwXWxN?17Tvy?qoW$><A1FPlS1JR9}w-qVM&Z%
zulC-?o5dKIS`I=4Ccx5LeK)@ZpqcR*6>|l%GG-XP&f&4t4ilf^M<1G$M^zT|(kIny
zoR;m^CMk(NVAyainBY(ATUw*jzrPj9;^1cVoYuG2J8>`;u=D+7d|}T0F+Nb5B%;?@
zdavmD`gmEn{vqR5mj;b_xheIxKzjsp_T;W?PA`Hv{Oqiz^BEXXlkv^78&Sx~9JC?Y
zu+aGRd~Jsa;<<QM#XIKPnpcnBa`lZsXp8YFaf12E{@bDK|AFJjLPHfONGmlzM|mCk
zX-pHW411H^QcZEH>SrdM(QCOuW&Dofng;8p3C0X6GZeMBznk&}yp)muIlb_I9pJjR
zZwZeG4?p52xojba=kCut1g<SE+8lx`c>TM@N;IcA<d(Dtfrf!Q{(I<xluJgU4*W(y
zamBq=n6W5K#%%W|LF+75@3uU8G1-T?mm1(UDbypoP&Xp5)PSd3fMT}j>a4?;UhK%+
z&z;zNO*vWZ(??Iw_BV8dcb&oQV#FW_qBs+*hysY|8Yf#?UZl1lWR+{XoXteRqvOT;
z!8sDS&u<JZ7dALTnJGN6>7<3T#`Uju<R$K>24jDo31q75FqXn}+;EU*F2i^Q$1>e|
z?%sfd1!EO#Fw!*<n~zKAJXfxVi;X(?>x9kYsAw<J8F|u)i_&}Z9+x+VWUQ3GY-K`K
zI2mzw)fJ6o59y`+H;IKx(SCJScN~<JKaT>MJrbD#cinz~9q|H2FvMlLh04Eqf7E5K
z&_tdV7LU3`g2b}JEeb#nn$taSlkYF)hUdu4xP;UCAcYkaX=a;GnhR?AM|CHrBGAC2
zq;H;KmzS$-WZjuef){fRC5}!iJ6gcA;|6L71-s;LmigCrInu05dPm7RR7oc|s;pQ=
zL#3+~NTN2ZdKQA_4c}^D6ov;Uy}zHIZq^ePG$h`cEUY;$H_@2t78c=imn9@&lVs{a
z5tE9XbDlQPC0gJZxZTThbDk0J-~+xg*UcLRoRP)`^D*vL$Wk161VG6wm^~g@QLfet
zWA4~C$w^j!e%3D~j%?10)g#iOnGru^H-5SK72tmIn`{0pMOyxVOQYF%_thqE64oit
zZb`KjCyMEe&{~IZODw+-&zf-K>+1U#Q=Wov-K0n^^74a&YfmR$>o)htoL*jDfodV-
z=huW`VdCFTuZ@8J7Jlsr82X%RT0rvUxckg1r7G}_^EeHiWX;m0r8F}V1v9fomHF9N
z=&UV%Y<-bDUOUTZ(4t&bPRjAu`f|`%Q)R01MU?3~_gqEQMt{(dNw#mDJ-be*#rmYw
zMM96;^jg>QC8iG)_D1}u8pGz}G;~R7#eTq*RxxoLSE)9>^m97;`qfGD&Ni8%IpWo{
z_fJ2Q_2Jt1A~NCd!a?lCHa{=vH@NR)&5od0bkd*f?Z?)-N=gRT)(oxtk?I#m<@A_m
ziysRCS+3Nbzk&MQ0Z;2X>{!Tx;UWhWS3t3DkA1DkZ32P`3P#lCDQ`;p_eHu|%*snU
zwVRD#Wi*0Im@6K?`s`P8iZsI?CxRh?)x;70r6WR*b~i-!hF(RFzb&Du=VMHR$T#;x
z-d}Y{ef#NGnLfLXbIQ$eTk&MZjiz(H*w^Z<yiC44PG%X#Hb@fb4082t0tPI2ud05X
z(||~bT7}`o>JhJWj|%88`lK;z#CCR1Qw1?>$;4@8OkxnD@=8$7$1}(L`A5hIcc)~&
z{t4CZ;itOSd=LrAm}s|uXU!JX`$M?<38f3+BUK%4%LZQ{>(-51v#$g-Pw$9TreNDh
zN5YLziIFh|bR6-4O<F(Y#1Pobo5U=QH$6YeFG@$e5i4mu>u@io8ec*MzylKDaJmA~
zfnTai)?6;9*+#;xr*IaK3#6dNEkcTZ;s<-hR*>*;Ll+l>oV?9#wap{YoXR(421~|F
z=36f0bKm|x!D7uc;Qf~uDk<XF_mek;v>KIy7`&ySEjZpG?z*;UhzjP4P-_~_dwTna
z;6H)_t{=P)Wj@O<BMrCIS=%PpoJ7&v2wP{=N*ff$tjMQ=r2a-`f-aM}%GHFVlhy2m
zFBb-D&^p^ph10qXy`EhnY<e1%NqXkoYU;%AQhEMBqtoqrveNp}V!v27+4@VI<ncdR
z*#5=r-C_He3#x&Le+-CRIMr6ovS0JUP=FS^@emWr&E-MMx%!IUgD8fYR0cs_@dizF
z2NxP_Fq*Xr1(a1%e&e-Ss8}6w7$NK49#!~I8CS;<Uv{h%By2on_+c^r731UtP&(ht
zr0iYA5!3{|nRB}KmEs=M`OT;7toHt{yngdjhPBC#RcjtpxAqSFcUVN47&g#<xvI_E
zt+Nyp>&MIIqEDiyLcip374)T^?oV)Tss4VRLGJq?Z`6erW8D6LJ^Z>tp8&be{)x`@
zMDWl(<sE?LB1toHh}2Sd*)tv~J96;W0-%_i+NPG}o;faeK8M*%k>7FS{Bs;W(X(WM
z<*&|RY`8xgvSv$mlL`&U4DO(&s<5Tj{|3>T*Oq$Sx$)g-iG-j20HfGld7ysn?+2Qn
zk13chgrA!4{JoUx?0LH{n#Z1BK7<JRJq?f<v<qo5T5tBL28z06$KUspM8bmIAg`(d
zuf{F=Ha(DQ8Cl#tJf2RDj*fCTJ1?i1)&XB=s$>m*ie%c6v$2Js&MwR@4i2pJ*FUvT
zuVD#q6KJ`iwHw9z*xz_)fO?VE%^$7cj01`|Jge8`ecU9upgKQtBZorUJ;t~h62TmJ
z9+h|5E{_4dVjO<P8pbCL1tu+5P(2W{Y8;23eE`~cny2o`_KwJgauRk!I?}eGM%gaO
zFA~Q%Kkg(Z^~}Y(2nKdbp5<{n(TqJil_3JN&+cy)KKO7-PB19~ZaYP_5q6L&u6iGG
zC?#Y1sE;!io!ot#z}oH8kgQqFK9wR-3nD3D*k~{W#pU*!cv5*=@kg>Oo?idPS!ZPN
zr%hg+1w*XwhwXH;3|s}M9`O=&N7I2^2beFLPVowpe=$rTGsNOt$pM;z9ALuEyAjhV
zc)!dh*B)L~rxt;kwrFXEnFi9vKhx56au@Hy!|h{Qnj%>mMWkVTfL;=6upmEf7E6qK
zx?hslgsx`~Q>BD|?N`vD;83rjfyqd5=S(QC2TuEaolvZNeK?rDYb1@-CLqc@JVQLd
z`eY=%#gL^JyG&!W1WWv)P}fZhE5Z}J{3T5>$p8+kz2#vVk2mT#EPi^C?iid@Qa5=_
zF+(_?U2G*~V4nTO#><<aczKlO9z1`WmE>ogKq%OvI4Wh<A~(Xc!7Wvta=E#b@D)SD
zY5WUcR}6ZiB*n?Wf%LJNFBxrcOoE)lri6`?(~D|UC~>Q)a&%ldYg;QZjD}`2V3EY3
zEb<~JH{CWk|E@Stq=ViCLZF@3K$4d|WO(Tm&iu%<(+C&Wick1fU*$~jgTZh-1P^1V
z#c8D?u$msRpxCM7e4tyz`)p$zu%&D2p8t!8u*H>3pU4^vi%JwUEE9-jzx4K8!DLq|
zu**pNe$HcKw3$_(!ZsT))fatyf-|W!VBIk*o%~_-^nnvIinQe*rIgU2>%+q&fcCxv
zFa>Faw#YeEcVi+)N<`N0^|_v(iue3^=f1!p>umxQ4d=6)L>!at^Kt%DNorDEO{YgL
zA$LK4p4Q*UbILs%CjzFGg5*c^qB168h*E+_TfOw+E2dTj>vD>4{2hK@F&F|J*22SE
zddXEqI;<BT{;a2T0;a6w4@8z>tHmwIq!tbdLvtRql<XOPRY}u)`CD8Ezl}jN#A|aQ
zS_B6ar+U9n<5h{mhXS|k?b^c5g~F9vDo4!FJtDVJVk~u-Zk|xKYJfcgzzHPFnXJIs
z+PdU48(yq~84R(IyL(DvisNPyooo4Y-BRfYOZ<NJ^k%;K2kAKJar#|xz#^zCTY_bA
zF*JFG-n}p}+hD<u(Bimh;ruB<S?iXpEesPn3xAFVCv!N~nmZCT6fje8F~>=3xZMsb
z^p9?c?ABZPwYMm*+!T7@lxQd7b=4Q{IFn;cqI(C>;NAr+6;3MlX+Wwd4QB9D&h-!R
z)A_b|LFSU1Byuo!L2><lGr<7_88f#>2H<4sWX5CfUIIm_XGp|rlnIczryCJ&xMYfT
zh$8S=#Q5185nOKZ0xwkWh9-?5syKs35e5q;7*daPU%LaBaen=^7KxV!#HYTJRA$Bf
zcdJB*JQB6c4KlkuJIOcc3l}F?R`GLVwEC`x>EgZK<u61pYkHoz|L_E?<LYaY7H|z<
zRlZGdLH&aPsNjlfkazRx>pK<VnoBDb6?HOdR6p@Z*mW&zj^0G$BxF9(>HG#Y$*;6p
z7q`E}KDa*TC-y5in&<Y#RL%~2s$tk50BJya8oh62Nmk_23EZczPULwq8y<ZU_|reH
zLY2mt83WI7XRcX>BW+^*IN5!gEpyDGuA@3&Y8rbz3w^YT(mAF`G3I9otARU>oPMDQ
zH<=U6%)|b}O-*LY4Oi4Dtwd!g_SP9tm;$bT1Ftgw%_lV>j%IxdcCmW^Ja~<sBoqd<
zl{|uc;s6WLV9kgAIo9H3#qGZTcGs@|I{Ieezq97~As{y_MfU~%9lO(?%)3P1wPy>R
z->4pAx;)7fu{-1$?42<*R*T;)nhj#^_`I~0PWP%FQ73Gpe{T<GrB6Xma>X4sVIBzw
zOtP0>=FYJH@5B36AvPm7m*g|)k1KGPrX9qFkenI^k<0@7+Dc`LTP}swkKF<so&HTp
zMWND?$A?8u&q>@EZLB0<Bc%Qr6}>3?$2b>_;^yeM^cG##hA*ZlCli2Y;{Y`!lqQ9x
z=tyKRRIYDi^!HD-QnZ;W_KB~apX?ZSQAHWz(tn}^Lr$L@d5uNsj*mmNFl>!Eu)5xk
zPiCCvulbH&=i|$>`gZe@(VHCebkmVr4**CZz92e7Wwi)xgSGW(W3?0G1<XJaObrf=
z>A#6i_p!u0#aWA4P3k12pVffG2lZ!v5EgP`Jaw6pQ-HI_Drn4*>ZI3Xo;BBBx2*Q{
z!EOqvpZ2`yN?Aa`bT29!ep!k=)<R|EJ>ZDyc2;o2?&0Gk0YxmhBdrnryv1$jE<1~u
z?m5#znpV~bK0DwNar;{g^+Vun_T#3m?(kI#S!>n>TnF+p9WgO<B@w7&zO4WX;T6|C
z9INlH6S=hy8YWB9{`M57GV{V4qTj0-*_&S<4=29=S>p`yI<FK_F(Ha81=UW^1MBPS
zV&{E?&Hnel@;GR;$gv}wQ(wK+h@r}RXW#`f=bCs+F|dXS4``X-B6xjyzDmm&FpXIn
zwz$}tJdS1Ky0eQ;{~;t&Jm5SYeX?CXS>-1)$$3H#2Nag51hi8ah+TPIFPNc`o9N?H
zYMFvfhjq<^FHLT4%?o-<w6Io^yy*X&*4gXb64<|2E;CF1APSCTnO=2T`eu?_Fl+Zx
zwYa!)XKVGobKM)%2a#aa1cr*LuP74VgwD_3obu&Wcl`Kp;-fB0OK@nd@#uTB{Scsb
zWVo5U_G*SdEYvq5@f*<nr<xWRO{n*!+lrcN8Lt}LtOs^`AZ2wzq#!AGP3bZYW**qv
z++!|ixV4?WH*2f2vsb4@Bx+gW_hIs!Cw_KlU}gW~RNCqUg;l*V)Ww|x=H>Cofrv>N
zwlUe?<Y->qzJ^VmAC(4Pe(BTEG<gdI4ehDN4PCvmrsvl%XmRFX(hT|1-pU3ji^oi&
zH<kMy99EfF<-&K^idm%o_FmvcceK#)1VgsY9@bx;U+w`0h{VPSmY!~nF`g9UBgW+p
z#2`@;+9UWI*mniVV!urNl2w8wlGm2UAE$5X7$2`a-UdwLdv)^gB9r#}OBbN0&<yjK
z7M@K|1{wn63W9$T=f{encyWxTo0adh9LN9>`=e)vCX*#;2oX|X@l=i@H|-w#;{*{B
zwxsT*gqd*`TJ^f^`mbQ@NqPSD%SfM`+^8|Jm_fyQJ8!5^u=4VaCK6V9=wt?G#F({R
z^;33ZGbx9&gB``y;~2ZvW$Z0mmc8r3k^P>C1Mz9Xsf0b%gA0eTV}PDTi_(frdt8L)
z`Bc*Y9c>Z*DJECxDMfGbgA|e6xI$c_&fwYl)9SiI+g3E&v!v~&uN-*I-%eO|b3d}b
zbnRo~{OiiJVwvq;mR1iVcQqS?`Ln(X8C2Un;}`TnnG=Vpwq^v}&2<sSC9T*3uE@3~
zuBG`jKGNu3vyB@6hqn8fKPS!;&&;LG4?f)QXrAwP6wL3JmzGWf{-o*#5Pn3hQB_y>
zjr=?Z!;GZ{M?eXXUkmf>Em$Z+76_#(eO=6JIV@>mGqHKs!tf%`<OI(FrSlUU&bi8o
zcWx<u?CtrE3!YthutcnX7ax<P@Ei5#N}oRA5Lu1TTSebkHyv0ZvmYXW@nr?xv(y>I
z=c~#BgRQvXpa)U09b3fiJzI$fZ2oMx&Se+b%G?@L^b6Cc9J`D+LPX)^^hYDTzkP3o
zF?uK1?Y9f@=_=JYBYSrsS7Bl*_gTY%Y#h6;ttFd~6x$#gL@DpU81`$!Wx~Ea#wU?<
z@;Ez->Uo7CB1Q(+6TX+MM~OkCUHkH&?klfwS#ei6OP`|<3#kbexf?v1l|5yTEpsy@
z!<|Kx_D@ean(`8Gj~tz!=x-?kYOh~}(%B8J(I>?pj@aGY+&m4*C@CAaWxvnWc;M0n
zBgMpPtWyCAlVZi#4%^~wc0+V;K=PD3><wG)O?KkWK=rcB9Q(=Wb3K1qweQ=%NK*Av
zjT}r@NJ%lkh(KQ#$M#Uu9SSqV0{BIVVb0vP^@zL0OzIe|6v!M+qU1<7=lO{vn1*m6
zs47ljQ8~>~S1-LN<GPqNbZ*!uNYQt9kF`I@+%*uWn8jCm<pi7f=;9<F+y9SGp~=X<
zOg_-Ml=qLc?qyeD2l?I{xohb1Ic|g;8DIuG?WcrCriLf_y`~-9JJWOLGk{AnhY3Ee
z64SPiZ|B)hS0U1#O(vkKZT`U&(ok2sl-u!?eYMijR?w~&OoGiE&+r=*ks=ZQ=|}xV
zs{i!?P@VQvHnO{}01kfh&&z=iPR*UC{-<pJQJU!n{7B=%7*w)AB<M49>6hlB+HWcH
z&9=^d15M)kuL&&yfi4|QGuPvo6J2I|A_T&%0y2Lb%_@0cljN^ih#wTa`{%f;lHF;N
zSGq#qHAJDMz`9Aph`(bkH{v+4FZgs<d64X){ZZxYdGqOYLLMC~F(?}V0qIpc=&uy;
zY0h~9h#J3WM!aVFttD!OGQ67PkDi$qs8F{UD$Oc;C@w)?Wb>8F5c;QzO=T!6%F7p|
z2}!xF8aNC2z`%qNLkQbsm3#L+0%JcuK7RePZ4mS0al4EaxoMY`FrxtLN|Zqr&%sdC
zZJ`x(sOcjs={FkVJR#CJ@3?Lkdtq?)_J-Itt$+$D(z#cAmOM>W1WiSDZmil*nO7BZ
z?y|2$xk+BEQM%E&=Uo!}wi1|w1Vz(GTQ3w#Ns8oG0Ma#t^x27C+1=A~DHweloxe{x
z>-^iLRU3v)*1L0{I`ak`k>+DKq}$$^2nt^Q_lmKIWNGPo<cQ)BltM2g!LpYg&=(Uh
zpXm`J3`g>4nIjg*BZ_1DG5hgXn7$&t57BWqdJp%S`hzj;z|Tl-uqBUi?kN-Ww=@tz
zwXDwq=Lf^yQ0Lwx#-K5h%3H*GDG<@*`pdpd3|W*Jgrbkz-Mq8($WuJX2Tsq^O6)SB
zc8)Y00_xFqLMs6{i{H7AXeE`sxC?5Y@E!p7!aAE@-vXItBjubi;4CW`K)llaIWx4Q
z8^OT8F;gLX781i#w(DiZBW?!yAr|Xx6WJ$W$np0^ERGF4uMI}~@1}6K03U4Y5bDw%
ze*U=ST+ddh9Y@zkvG7s6u=Pte{P_6zWnLL*#{XnFb&;`M9=8N@qeT%$ewOQS|My^t
zW3p7IP_TLVKe#N+k^_2Ngj#eNFkmbg^#u{Cie(pQ1e46aHd-F^`f9pGMR@ING<Kh_
zLSb=fc2FBK$-X1jUfeLT@%RvRgSYIqL!Bfgd=0ecVZf+cb7$8}ng1apaR(V`Ir^E&
zUWKz&5bWaX)=qo4thw#undW$Gi-B?_1y#kg1L?{s5)E|twls0U%vu`2ho$FJZRpO}
z7Um{QH=9Tz9tSfyi}u$hO2v}E&W7Oe`P82uwGdLU(sqS_pKP4;>s)GExLnUL=0$=z
z(h@F;IYwDz#<PTa)Cw0n4stHr0s409+uv0Y%%BaTglM~TJ>6ULin#L&L`{fC*f!Et
z*x<LKuR(6!fmaNcW03m4|LC%(J0ZzbO^ls{feuAL3`4+9G%RW`D*CO(CSy_Lp^gy+
zvxdFlBrjA+Q<~_oy1}Ad9~-h}JitpkcsG|-#S#6O8j6a5_vKho_u^jmL+vK_Tn8h)
z)q=@;WPm%<@ne(u>vxZ7!-mFV@aHIKx|L_sJ*3OkKF&HJX=s-F9~=UE!$1_y2Is8)
z_1)3zpZkX81_6IoMAWCFD^F)HGiq1j(UDZPmz;NS>;(Q?W)N0N@DD=L`3fE+(}=#_
z_iRO01+Ln?VOaCd$v?y%MVOGj5{DcPWZ25hhj<7leAyME5S{<j3WI}WfBy_S(=S)1
zs6EtVS7R<&iaadlAu9YOB=<;4MV6{Pn^m&3y!=v3CUOF}?Zv2O059?%gERYbV`C$v
z2je5iojW;Ru`LMGmsCI%qcrvU+DQ5f>AKSxB9qu}`tRCz<H-suuPU6CC|>a@p?k;u
zD2`y3Abz5*Sbk7c;DtZ4V*R83^^S@%6bXG*Xnj{eOuspI-2uSwgEyI`KZ`UV#<Y<e
z?pZBMaSWs+e}y!c@dq<9<SUrsHxHy^8^qCM%m9qz8VIg<j+-LITE^?deV_SRBPJCp
z^4FaC!>uS~-Fd0#%68G_oy9ar&rfl3ya>9t4gFE{9@RY1Ue?YpPBva?ZM|k!`9oJ;
zNjX~(!e$;x6G13c)z*s;qo@|IOlKY*o%;8soc(V%fI5Fu0>;OcW|ac5L|Bn<tNP|p
z3kY_~40kNMs9KmIJa-ubIt!j@cttc_!G$Fx?242m=<F(Hke9mSzAU4l1`u|0i#DVc
zW@@^l6aU5KxcJjklIaeOB17`71aZ-}I@fJ^FO2R{T{!zc^iTQ=E*6})Al}Tw+>D3H
zx25rtA9x#eR30=}6tt3>ul@p&#1nWb4{S&ro20M&R)XxjUk@m3O+XH^<@W1hSx?Ao
z+blfdJ;N11Oy?zQM0Z<BlG$Mzv0u8t_Ju?-=X%a8-PT(j;Mn!jiH$hkWMQ@Jam9BP
z`W#8%#zHE!oi4~BR%l-Ao9hFg%Vj<pB{<(BuozbnVj-9<h)9F`s)NsN?pEPD1Ohpm
zA*-+;Y;wj#K`$)EEJY27axxZ8=yetaWklszS4w6dp*$%|Z%LxQ3^KQX3CNDBxBd}h
zR2bto1E-~+<Jhtz#fd4ct^c78-}$sOUP!ajr7TjFDyVBo#V+QJhh=Kq%yPf(DH3K}
zVR(%JuMCUw@o@f%MjyZWyORcq=%M>@w0i|M98mh1`ow+=W8W2*{M`BQQ<A;PH$Fkg
zB>83oSkr%=Id3<0m3EL;iKTD5@56kf7R^K&c0<ay;XHeTmb|g-ZbzO5{y(g{Rbdzv
zuc*`$UPTQrhTxwTl5zQ_%g&>vH!2t9KQ;EZbp7W&<_G;^^B5?44bW>=2W54B7m=^s
zblp6?s(<Ka_|9i2ygoboagh48T(38BFr<`tLd<@Ug|e{sBbY1AOa-83c$kI()0iH=
zFSq^Q7yq+cW~+DebK17}q5AVQ`T9CXYiO6j7;XI81?qAh=_7H<ce<;_8{Dak0d3@B
z22KwKoStY6WcCJ#7o5k(H<?}gvY5%ZL*}>7aJj@4;>8=ZnX{8M=e)ZhvC&gM;;BQK
zdrIjqXu{#?ky#4OgT^1en#p*Y(^3tBL%5}=(XMDTUl9yOA{zk{6i(%0NwDK@jD4=i
zc`=fD%(Er7!Wf76>5rK5ZtoiPer=F<QLvuSu92OXxA@}U#2(X?$UaY|><Wk_EN?WO
zW`O#(<OsuhYNg$AL|L1;BI|ISO~yRJ@S*;<DsD4N(w{H!$^{@z&CM?+RuMcX*F0KG
zV;S2P_V~Nc{R_Zp4T3|tIfPsmLS$6~f;5O2>5)WB#j%2lNM8}3{NkgKaCYvI@{8$H
z^cAsyY%?)M+J7`;gEsa&vtqjTK1aVH(2G^Fwr~qQrMkHA!d%}~o&sRZskxRgItjS&
zBr##+CDE@ejDIidDhz+JG=%E_8_o_YNsI1XDD2`|>2>xnqovl&2Ws40p{fb?6_p>D
z-sURi$N)=h^9{jAvAORngY?4RcK&J!rl&AVzH|HB>-^5W54_lgP)#F2H6V5I2oDYu
zdp}3czdbRsxw$FA9k9VeLQZy0S`6~5pM;Up{~HJ?*mwo5E+fBoCy-WfXlN93d?=fP
zavon~smf1363%CWWM3?YGLL0HtN}El5Wb!idhi2F6IhG#s5<%IzfU0=t*eShQK3WO
zSbCb4y6JN>68(%Am0U}AYdkus`|<M2Id9jgBfSeVh?7#{cT?MjQe^M<^GFMo>8$b4
zEM)-?ki^9pZx&DHn5q`$_-_(i-Z#-hMo~C2XL9d<2`GU*C{wdq@eSdX14+CDWV-Hh
z)UCPWGtZM&EBHGR>NiW4!B&Og`tWj{$~KX_N#$0-*f4pZ{H|vJy;0H^cx~MKjW68e
zvA`lD5#tAn<6%oU6l<LUmDdL9s_-SpG?|8zo=nt9zG92ZE`&*OVgNdDjKWm(!7MP9
z9w3TXfLp@R!`=vhBl=4O{aE~76r;Z6H~c7+X5GZ+sIa91J_?bP&LLQPA-m4eEdT12
zk-ym?!GQ50*6Dip`iTcm;Ak4VmqIZG_2+MvUH_q+CeiTlbAY{;lj^jVz+wC3j%o$&
z*r8nz!`2HvIk!PdsQ;FI4<ptaI^gj#I5;?X0m^L4E-3qie@s`Vm~CR^Cxq;){~0Rs
zfMMJLtqL~6;h0fKhvbf)h*N56E4wmgkJ>A&+@smq>dAurW@_IVai)#;`sj+t7fcv%
zfZWfl(PE1VcCzm`<Iw2D5c<&cj34zS%Hmc{eb<>`Y(kE6I;Aa<r~h^5_id$+Pw|mw
zmkj}h0}5G+KKuLoEUOx_W(>e^f|8Olb)Gcfi6!cDZk1%l(Y|P)<3MD63NNj74zh>%
z*S0EWRB4K}HZ^l8x?&T`v@(|a45B#xX_D%|cU$Rg;qX(YQoOrgLQz-^N!}<62g|N2
z_|t^EK0;bUiplxmg5n}=x<8q6DFl{bGH3uX@2B-s(7R}SPl2c1m&ZI^@}%a5keN8{
zrU{^8BV}>mv`|@U^C(}N0gCT#$D5N{7?=Rj8OO^hC}^R`B$fs%YZy~RaWf;2`m?>x
z7pe-X)3q}fw>LEOWHh)pm!0$$Z>I)RCnqK77n?{`eXUe0EcZOPE1TQA4W);)fUEEZ
zxC-|@FI#N)c{NvJSp*)#b?fz~9`kMOV>gx749xA)y})EgHL#QI?=$)O)2AwyjJp@6
zMcnkhptx&Xl7=KSfTlXCK}C5?0iR=`DC0JCwQYsiEY@2mtWB(;;pgX9R#v(Wo<&JT
zfm)ux$mInoj$j$Ei{C1FlGkO7eHTxpxf&A2bOc@OIy-ux0ej%+k>6!#2=Ui8dY&!6
zr=jVfAdDn&)-7jj^~w3`N$qU>X$4&G<8jMpM0f~3#)`ldYK#_tm41i1j6~$JOs25&
zmQZQhKl7O8ZFCYcvu=mwrfF$<)$$*=q1>3JM^l~gMv%!BX(oWLD<cJ#;A2)Dq%Gj8
z;kITcaG6^R$aC%ws3YoRvT$+yx@x6Y$Z6)v!~0eM5t<5Y7*GK<#lH0;b9VB0#!9{J
z(yk-ly2xI%juKx)*59Wpd=(4TeqN_YSLuuYVjiEpe8&8OPIqB%M2W&V6G|M-a`a3<
z#4v`v;*Lb^QNUlB<OSR07D>2zM7V}oc9V|{OWE^5iauBYq)9)rupNVVW1pQg<cy<h
zFn?pH;*nVBZY*t>Fukb2YR%kx9V3hE(h_F9Q}%++I)Pym*(*UslF0@b8*h88Zjp4q
zzB=u3pox@!PudYajQ|YMyLtV5CUeEec$*gV;om<dEL{sGgekS@*#MFhxg3^QQN1K`
zN`wUX&NpvT^*-Tiw!rezbr9I5j*CEc{tOD$nc0CVTL>v)73-O=bG@qnkvw3DBc_3=
zqQDiR<C0jnk0HqeGHbj2FnJL{(uE{NuApc6epW8H;P){t61xi#`mXGq6f|brC-mAM
zgJdWLEu#1Aqee5=wUznm{Fk`4`C+g8#xQQhgwc$eL8=pu8A!mc$huF;yOGj#dV&q_
z!_!Jb_h9CNs@4`%a|}Bs^D_1_+#XxW6%UG(5NCS1WAfaauo=Y2zfCzZGKMbW+rMTT
zw*~2nb{{2);{RT;q36}VnnOc6En}TV7N&cAP}l9=`xX1vN^lsHfVsg@>;OkyUA;UM
z!1LEJ`kue=+4+4I(ul=~Js-t38rZId)6ipm8C^~BGP&fK{UCceX@hJ^IW6A5N#FBZ
zv|xD4^l=Y2njKU7atr?+)5a5sN_Y=!$(~0q#jYkPk0$ru@4C)SSqKDg^F%l;p~2c6
z!z57MU}C!EZQ`(0$<N&}7H6ox$L(CxZBo1#{)#8c<a7DqbJ^euyw7YOIeAYcC(6XQ
zvF^Pc2Lea}K3jgey6FixgBdkWc6RlHVq#*c*zj$iD3nNY2fmI~Ma(|=iN0Mw*>|3N
z(ytHL6Z(nX;}<W4@A_<2B*HQ;u#?)uw#!K4^k-&+E`#*AURJvJ4I_bpSAJ!Pb7eKl
zeF8JEkYerqt?gyMe(@*{40}(3=8Wv+>6tb_(X2NdFI%<d9Cf_h{Gj){2s?KV!bmUu
zrDxTm+j)}nPmHl${}(Cx=3a^Z&AzjzV**klgJedg)mKzR(h7fSx|PVUb?KiL`m=B-
z+bJTI2?wdV&seMP+YH7**_l#(=lKm3*AMfB!H`U6O7Ga7i;Jw{KZV8PzCkatG-LJO
zoPSA~-SWT~f6nq55`+`e-QI)40I?MdPN2Hr7i<k$v~w67>F=-YkdcJ)a0|Q!555nD
zOw12wY-I^lS8)(r+rt_y;4;VLa`Xb`K8mgs)DEC#uKVCAH!bcxk+lDD_1*DQ|Nq-1
zdmj6cy~{jV2a$2?d8Eibh|G-aQO4oOu@f>XqsR`~;vB1l$ljyujAT>y%jf&K@89ox
z|KrhL^nSlz<9c4t>$;wGx*}{hCxq!G2gMt|Mm@<o&10MxLkuVOo1Fi3+aAo<{CT+p
zRe%I+6GD3f8im#|ymek<Rqr4+SwQhj)zco@X{)-Jmn`V17-;vY<gsGK{<@XQQC&Z3
zG+PG~K5mJBVy8P40@&jbfT)~7h4z~sQ7>a{p4_M;E+$4revi$!ig-KN+Wc}&`m5%_
zC1df3u}~ig>bglD6l2ts%+z1?6sBcu9wTTgyztXrC0?f;p<pM!{qb*DFw0_PCbll2
z3><pLMKDT1W-)sLPKW9VedBAAv>{tTjd{%0ToAlEX)>Ig*$N7`v0+cpWGk1UJ21iS
z2j}}yao*(T<L1J>gkPOm5jLpXS<^l%-E3H_o?%Wk0_ald`J+pu6K3^v-Fw!v+{s4w
z>7rO-?jG##rxE9eC_U_k@={Of_d+jW@Sc?E;*Jrn@a!lR(%6jXtTt(zqZ@xhZ-i|o
z^UoEQmF-8NF2!gV`CJ$0BO~TB*tv(k5^k`=hcR6hESUIz7zO;E<YoSSk-_~>^7!=b
z#+aXn$!`ms$!gPe(`9w2H#k0&{}c$DVI;?u5uz93Uez-D60o0Ee$x@0<ff*frk;d4
z$EQU5P^A9GBx`I6<`qaERIfcDjm#MRN3Rd3!FDW#sq?uN640={9_iY#PAIa%+uz<*
zG<~0x@07>Rwd6x#$~>5`@{MH|NAqO$=R#FBBDPDzzDmbJPtD@PAGKwhKR;|dhIbtR
z9P+yj5iGM$9%UKzqUg6%bOfv;?DmT`eYw`Usj!8sS0}9o&o|TYOl6m&?<}=DP{v-J
zOPs5{&rV9k-aW2G3Hk}Z!p8sU31o;=3YVw8K_?S2X>9qjRI*u$fO#WQY`MN6N{5ke
zV~wsA+IQAyAw*io1FjYVfnGdp2(!>pMUEQ>{V}S35qPw8*EhL32)8@C6(q(>DfQEE
z_Zy;QX#*}ctLp2XYwBwp(a_VSg@BHQ_>uy<1a&jb57|XTD)rGTYvfo)(z3UL`92i(
z$?BFKsE}xlfjbA5yCI83g49U(`T!kMu$vY0+(x<*&|9%vbnsi+BIynsMSk^eQ1G@B
zEIm(f#XbX^%+7t87U|fCN$XV2?UECUm+8*>V-l8#iOp2i6fVIQF~4hKZlc=TFz(%@
z(``mot$lDZPy>~G@)I+302t4xI;6f6YF}LFqEp$wlj6+!+87kCT@b!3eXg!WDvQI@
z)$8E913zEftAqv5s2UA#7k$lobXt3RI}!a}eY||bJah{L8b9r{`$4~}C|8{MDboR=
zO4?Dh#!}eBk`AZ5Tx^^!nO%R|Conb=eE5KwXpM=?ZlcL0qmD&K%jCXSFIHH)_hQW%
z?xssH1=a9NWaqrVE=!Ie&x0a>(`o=KcmVB3i71mdMOP`4#ih0N^z;x>yr}#ddv7R;
zuOs3G!Wn|u*FEHS4HivP8`Al`9Mo5EX8-0&=-pL5orL4#%duBT&t|UfShFV2CANO2
zzK6YhD#sILLVQ{h)CNi|CWtq1!tns<n(^T_n;0>4F8tkTD@>^11`~Oj*7F`+`y`uH
zqcVrfD&%HIjkoyXO+0!=AIi8FwuGixH6u@_FF708^^m%=-po;mfwSeB$<CA0B#1D^
z0E3+4;X3PR?1`lqiD^1xOd+d0&&lK89TNfeE_h|5bt{2iC)&xGa>Xo+F>`i=ulh?7
z#JMXyi2DiaXKUH<{RxqMCF%!M+Cb0w6V&{@Q)mPwVgDNG<Oc#L4=24D7-BPZiE*Yg
zOeQZ(2^oj;BwGsnz+d0Gw%PDN{yUJMGVu~JSN0U;=jUUHG#%aA<>s(*knm9UIHN`7
zh$`|=Ob7B<8(^MeW{z2CvgeNKpln_BE1SCihb81|=h8ACl8dyLe);eg#{{|U+(Y(m
z34tF@E3M86aP4fVuwABDI=$Zm{T{ME{nw;xwI6-yU;Z*Ms&|#Ez<Atdgc+<dS{8+f
zUlL7`QI{vgv_X!}mzB=+x7eP#k~iB|izKt-J!k9EDv{H_>SuP&I}bUfy0w&hv|ZW`
zjsIS&o|}=fbJ%Izes6IYZSlpaJ!B<4n9sMOiSmWaZZ>VC{~)6tLqDkvn=qyQkY#?s
z2#j)`G+GGA|LAudZ&>r*RFqW_#~erHQ=}ehCUT@@W@d`l*3~`MyS}h*#RI^)bmU})
z>_drD!Z~l!a!8~BR+~?`=&3myhs?mwoy@0OJYzOiNoiKQdej_p^fMboM!GDtO?>j?
zw*v#u`?3*acH>KQ(T7xwSWL_BgUN;Lzeu`5A?%9(&$Lh78ZY4%NxbOETZd+wdsXD8
z*^sL;ZmS%utnVWpiltWvi`-n_-29E~YI_28=3Gjljnbt^`4VtHaYVG#K1<<EWJVHS
zyF<Jnr>qT>giof@d~VKNY2MI_$TMxVpqW(g#|C=)7H_KBp$@~NAKocB=r+!>alPjT
z0((iBHM9ItNxJwy@BRQB-KVV0_IoQ<(CZbMdaEPhPvwxaTcJM9bGMlJ#ltCpHbrYo
zCc?qm99{0~C^A5uvAuLiSP`63%z2lzYf%7EP4jC{Ta1_;&#WT&#fn`6UQ((%KEibO
zEwK9EW2#3ZEueKTb91AGHu12{4pipLm$w|EP{kNw-V?D)$p>by8z&n_2w<$WG3ma9
zc}8Vh3-D~bCws0Oa|c2^P3>*){f(0V#DR<DmvjQyV!+Q-?>JcnpD-2bm;Kt$Iu`ox
z5IFhV{~Lmsxyx0=aW#IK^VOYRENjsH_hP6^WBbV%rs1;kRzumPlsjW!NKUCev^((d
zv2+)Qo0XN-hZ=w&Bn>8ZduL`O?M>mA@icMxs3bzr!q=_LgW`Deyw4Y(4F6GQM_LB|
zyl%UB{G&of;@YiV?c>I!Kv`Xe@mOSI>qY{oEg;?iK@?sA>zReLm4<S1iD>bazpgJ2
zW9OS_zG-UZO51s+y~hW^?g~X<!bg=m@3wKFb&8(AL&RYFbH6vyIUScg7WD*7PqxIZ
zD?&M41T#n=f0Hr|$6X1J+Mm<SYwLf7g?LHnO!`-4<&HATq;4OOv;Cue$<5_>4AjDr
z5?H7qu$W3<r`n{kKUW9yXF>6I%f;iAzxbK>RWDzGyjBKP_rhA={7ZRT;i=(9_k@uO
zF}v9ZFN;>iK5%$iCJ&9&XNY%??i1ac?H5n4_1<~(qYAwBEmL^o5=8#(3dQ?vv-{g}
zJY}*r*OGUa&8(?&K53WPCX*3v)cUSARYK>}mRge4W9AO(0|z*ty9hDGot8H&zj(>P
z`aXFN8dG^hx>&g_W>5LcR|Gd6<>NY9y6Zb_JZ&c!7nk}*RZoj`Rho0Xp&>gfX;Yf>
zYv-xg+(4H2bE0ps?|tbXZ!W#!;q!EPoojEP)k3!ZXbkuMPYnOkEwO?>Tti`JQ1+uJ
zOUeV&wF|n{@A<DB{Ne67!IX8_+yn23f`V*!lJvhEtTl0abLtcfJg&;O;sRRY)SI^&
z;%4vK1P}D;T3F0-?D=TBJV{uHvuDR$Dn|j;br63)x5lJi+eRR17X)2qi=BGaUzu4s
zc6wA@T|IKY7>Dw+Gvc|7?1a4Ly2P*jmcA_y^DKT+4qJa8^CJxb$D`QN>HW)s#UyD*
zD-|W^UiLQJfy>dB2~o<FVs#n1oC<!SFNM;6SSdoFFe_tLj3EZ8j{P*&-9j7M6^|=S
z;%KAYS0?7>wmB7pYWTj}mCAvT<uyMs9aj5ijOb*~a4KhJ2nL^58JQ=+?aX~-oWpF)
zNym(K|1QhCWtE*hjQiF4bMaV#4|Wu0O=kl|A$mGNsVD}}u4BxcoK}Xi<+Lzg;L$xk
z-zeNpXBBNWas1-%VNt9ikDR84#wOLm7o5A1&3I?vq>!^DGBF6RUOm92Gl}UCudr@0
zzwY?0WsnR%KVNK{_UQ%Vr^mEUb_l~)zGhBF8Ez=cRQ#oK&8Jl;iS^|0Ch%ullyE6(
zl^0149}`{^+I1vQ-7rE(dozYmFPgog#o&yqGFa{xNF#m=AyCaqT5?+Q@~nh;o>JFH
zK0me8M-BlC^l7DLU(j&5raQ_qS_nINP$Y@%Acsd&$$TOpDUzl{)lW;ad6UQRz9r)>
zkk+2PmYsgk(DX%ItY>;!RBVW<<hN|5wcN^LXVA~w)WfuIEelJ_E7aQL15?*PKVm${
z?&r@st*VUbZQc$v(Y9dDYE;sV3t<URN+w0U4&51gHjNusgZPyT2N%*tE9L}{WF*Pf
zRnh;(Bx?NZ&ekW)sII%l6o+}naU*p%?OH4*8OSeI7Yi6kRrcR0QVgz-$h!J>s#?bB
z0H}|XD<9kS<qcu(*Dm;ecg#seIlua0k&#!~Y_HdCo8Pr)plbfLLT<X%7fMHVp<pez
zTZwtiF*x{Ya%|Uox)JX-RcQVtWAAlIMeB*h@>l<I6c*MRmvVQdG7;`<sgXxyi3g)m
z58msxR<=nu`>Yty1WO7a83dw|fSyCgTN)iNbn|gKgWJ#I&&tzgwOnr~xp>4d*P13q
zK77_T&xGGSzIA-}`^#V$Ol9WKulM`lX#wlI&|6MUk(iHMeqs<`mv>h)NSnT#9w$Lm
z`+-gpsFi`UUW@&e{Sn=(E(w7zYPHc-b+@=6_BchUU5^1g%&h~(Ne`^u40o~cGh>;o
zmfP1~<IgHWW+Hh`In%Ymuh3CX1l4OP<UsDY)<{i<c#djtQ>#VE`-ov;%cJVS#X$8@
z$|N)H*qiOz@6U|Ac_jnW<n!n8=3EGiI8BD~tp<$|zkCRKgN2<+z==X8NOAp;&GOKz
z7=8P#0lRgRpD4AW<nwVYXHiqzuB!@0C`#2J^i#Z}+xxgo3CGL4)b=YIv!r<N!9{6!
zZl(fmzyBQqWn@8BahB)QBs>4`afe$im5DRAXP6+r%=be3EoA1ibxghh=m1~dQ$J8~
zetwvm<QRQE^5poqa;<uc@nY>jQlP#)D{dlO<oG+lJ4V0eX<6(F6%&y>_OQY`-k9~t
zs|2042ZjjkCD{D>jVxIyDZ<xB_lu*;Xmmx|_?sRILDJSQPk(^Csg{;fXN+0lm1wym
zoTPH|>{GES-+)THD_!k4W%6-1ZOh%VHelXZelh+-`DiolD3m>mDey<Y>v!h$Exq}%
zHA;&2itaINt<g0ste?c<R%s5wTX!WFHk*zVTDuOjA6#u~dlH}0?^s<(1G@Q`w61_i
zA!o(8Z{O^hMbVtNc+^3JE`Db^LJ*ictma35HM6HM<jM-scl&9-$bf$sbEElucDDR#
z%fY4Stbvf{u|Z0HQ1UIYu0-m|!`t7H+Vr`IGec=^D(cgcS07pHs2za{1_Is@8IEN^
zcRm{zu77m4+2ABF)pF5s>kq$=wjE{x&MX&-jq~260$Tp_U%ldwZTHtlY*Jm0X7dM|
zqSFin?{*^h9$iDwin8A+mFf_MbCz7LSh5(^*y!opFfInG<+oUGudl1Osu(qsZnGT%
z*FefXWxJO)7}oFNJEKYKtSEDy(k$Ps?Q-uYHM4Jvi*0_c=N}WG7V4@q{~I1DO@?=*
zJiZ$od>S(k<>25r07!x__0zZtd2RGwJ3*M`L6uf<Nl9-EywskNNzyqU#fSmorNjZa
zv#M6_P|IbHvdh{}fkhNc_49cR@ZPHI?P+V1KuRB4{VI!;A0*+I`?Y-R{#>d$?{8MI
z*P^r4ZB^3Kl1?*^6XC>vcNHDD|DD13xW>}qYgl{I4o?RlCwnA_K*7=B8d_=$8pGh<
zmey5voioOwSzF2DPb4T*7^bpcHs8L4W5|y1q`XU!jW6o85Uz!1?8@W30t9fAZCR#c
z%SF~$Tthb7wSrHyDPmI3<y?Tb+0!Afk{SNUl@`5g>B5MnZLy3K!hWEtYPQbvod59L
zOdZ%Eq+@YhS$(s+I0|vM8+nz=3>pMPLqZG)?feiGx$ER0B*<7)>Kr!{B^^!1U8V~`
z$#H!`YSju+PEPs1?ULTSb9D24^`}wIo%3|Q^PAkx?ZRNYXn|lNicbKRUMvf3a&kO#
z`(dc~c+N|4+0>2Psj1$g84T2!mt2kQAYN$G)oLLrve!;vH?$tDATO^ZFyc807`|KE
zspH|d4*(juE27N`OxUr8(bWg@&c$S;cQ%3wTp(j1wgo|BUUC6ec?r*m1^W*^$HKVp
zw<(@1EvSC?!XY)ZmgH2wlorOT?}i9C?V76?l^K})1U<$eKE`?Jn;`<2*NMFH@?T<A
zk{^d?xwK>Ig!|xo-EmeiRm4jPVtNr4&|oamtWE014ZChJ49=5kLsd{D#O}EicR}Iq
zxr1wEkoAe+3yMJAu9HA*B1Te{7P&L=U$6Mc?ApKePg`0KMZ8?dGf@REqw_*%_s<yu
z7={QRqd<?vX$=oMRrYdWw~Pls1Rje!P{M2zLM{l~#zaqwtd`K8J0*9|TTQMY79P&=
z-$?!X_l?Zx1cv|k8@h3rW3&-J;_cn+=LN0$z%DI+ZDE-$=jpOOey_MwKNOabP3D46
z+52v|C<$o^W>^d>+p_$zFS~aGc6-XFbpKdyfZ4_Sj`&8AjUvjIj7G3tVzI6KMJGz~
z!?f4;o@9l2rf64aoAhSNHJy?SqtQ$DsiRPhY$-xJU#io~c~`lkYWVguG9I!SFG3GU
z(nF>uCcZCeN-|JiC#Xe`ac@?zUjEOC;Oxy~eC-7dJer$yd%Q}`mD1w1*$mFy#>|!C
zqVF~83rsk_7e3?CAW=C&BJ}k1oYIL^+Sl*jd|4{Q=IGFD+HWxO_MnncRU!V8-}}Z!
zJR}C`T5lfCQof*H;*$90&!<n;6OBqQUf?m$W*_r-@@alD{Ng%Y+sf5DwG=nx&cL;-
z$#SLqm7?Z8iz?yK(IW$w%W)hofl4aFeh)|TB*G4*m-j}V?@aR5N5znTBFCYC(LEz^
z;%B0EZK*}wR<#DhA5B5(V9>@=7eRP0R)(s5ftr|C5{^|qSN?kRfYL*W5u{j^NZz`|
zrM+|-|Lfyj`CO+D5kiI+0%MNJr=7n)zMK%-V{<Cpl8UfMPAaf{TnoQYfrE`f7YQ)x
zD{1BhhAs|*DuZ`wAdsfl?chB<x(tey$D6Ov8i>^e0wV0xs+a-OzE6iv^ZdC`FoW;S
z$*(kD5$kwqzFu)4Rx(paLCaiW#n8ccE&kI-)A%0Q%m{JR{`Z~C5Qp74R30`dH{;V3
z>5h4EA-nVW50|C%UGrCOJcsf;QP;)AIfaMy8SYvKPE)duqz^>69$h5+<dpYI=r9Fr
z-PB>Rtx8FxN<a~jF+*Z}2aWGNs{!CZck%Q}%1yZy69aLphV0|OxZ#)03<lrRLtN)7
zx_w&j84tLz&gvEmF=Zz?#+;9+1C#hatDhm8)|#ziT5tbCCxqp)^rp(P*J}&*_{DBl
zNm6&As;V>E8<eYWu67y-D}DQRMc4Qmh8>L;5mnA;ZhZUUq-n&qbM2eT`Mk>$$el7s
zbkDsGm$A=aOdH7~hPF;iO#29Vx>WAUFF$9wXbo2jgy%Jnb)d34`LPt3`Jzu-^tTvu
z14!q5YK}nFh{KOk4LI9;x)!l6r-|?<u1B-X|0pBAK=;m;`?4OtHWM-LmgskWCZ)px
z2PR3>T>R~N1CrXEdB=FL5@&M7fU{!bYs<pDJs;(i74D0Ib4cefT9BWecB<*iMokSX
zC1lFUX+Y~2LqSWZk5RfDny@17I7ZFe_~FfMF^y_Ep3Mp|V&L8ZZzl=y+CG5xv-(1g
zg8#cG)^tO@@7cX_GM<RjqeC&c^8QK@o7OXX<7-92*#bzYGNr2iWT@Jw#u)B*?Q)i7
zTvUS<=7K^Umhc!?ex1SHNnzKD$IVbrSvA6!nGefO)V!YJLTN`CbF)Q<MaU!LM_AKt
z4l{_E@RR%34|wL~RR*?p4=}ljE=a}VE?t924_}RTJ$igz@z>W$Ahl|L(~DZ0aeSC@
z)65gAuJ4GSy*(yj<dp$YHOB_Y`PaQ2m7WGLJmq;~Z|>o6%0mSUS?<fKU4Gia?LNKQ
z=bvsnaCfx-aq;>WhVzlTQrbi%LEY?SXQCS8lQl#H`*{sm_pj3OQpG&SH$L@x0$t-g
zn<vS7Ke3r(eC_X^lC~%$`hcARk=r`^V0}f5XqXSZQOQb6lc0jQZMqbV?xpQtFn2jM
z=+|<XmwvKDzaO1X{CFvhx1hy6)O*`iG4?itb*;$c>ULG}q+s%HX4&1MFnds!7+hCZ
zU+5={xf){aes*pdv?Omx`4qI1Pc9cnkCd3sS&1;c)Xc7FD3Xq5E_G3DxMFHwZ$jKY
zUn0TA6a|Ecs^&Dqr)9by(<SNW2Uxw1%1z|zp6}msvrg^iXE%|7ZD;PbhGjA*_WWv%
z(4+qOtj0FSADGm5pPofYDS*5kpa8i*P6c9~Oh=2hi_@)2p0?Z?tf_1DrMR8(@r;>m
zU~AXVPQP_2+Y&fpCOK<>4&{vrZlej2TQRHuTqvhJ%q7J@d6-nuQvHOEaHQ-Gc`lL^
zvpp%4Z2WVhl3}N{^}f4L93FgsFU;$Jay4l%RPxW*ox|*Rp=Hk-<QKF@=yUfdymuT_
z$XC3FfRYFj__x}aN~4Fp&*4TWsK-_$TqyIS!dc=rP3Vn9zKq}JElBnp{Pu0B;$C%E
zw;6G>GSPurgoJ$ozz6F7b1J1J$tTI_2YwiQV6kA)wd@@Akl#-VVm>c^F#S!FL0i^%
zS^9d&W6kRw&*y8~PBT+$zAAle$O{^bl9Jr4xTrw>c_AJWIze7cm}V4koKxsdH2S)(
z0gK0nD<st8Et#bsEw2CACpIs5u<<unN5(qH-Oz|a)&jMMzRI+ZTq*f_Ye&H~_F;e5
z4ifLv=z{Jbm^JMn9*`}*8}(qf9)Yqbkt2?AD_guisVhC;?9gm9;)?F%iH*gH$(;PK
zZ=vEPzv4?}8cyqD3IY#5lH^IaP&IvPWeFzCDv@+c6*W3&!`^H4p$yjQ*trs+kf~T>
zcX7A#c~W!MNQVZ;!?UC^lDfD1U#ujFA>l9AS-&vj_xv)(=(krQePhYD8<wptWECLU
z-YFb@NC$4d`;{LXq5ruRhHe-^#nc45eQ|7FS>w}DjYRaDj!x`mj<}gzJffLkjLA;%
z?@7hqp#03{IleTBT$+vu4Yj4hy|)yxfQmn-7%Vn!XB?<DuY)aSl`saFsvErxfEkH1
z(N2I(H9MwRu$GbCX%~HzcU86&D=5>zWU91fp<gPk)T>F6X2>*u=-zsR1qrPh6sKX_
z+{-TE-fFRCD+=TO3RIkrSuR9HA-3?nno$WR5$--5zs0y0)ep!eV~J)X4=G--1PaP;
zt`aUf^&Ma)Jd*35Txw~482mu>kBT;DTGRO1hZ=<U%9n$JFa1Ojj}7Nqs$#_gEHS6T
zGB*Wg&YcR(&fXO0&eJ@2X=vym%I*B*oblhYk)ds39wKJ!6E1=6g+zzbMoD^-<HLW5
z1$#Q6FBoC$2@Z_=KF^>{{0LGa|B@wdEAmk3r|b?D&758k?+u@~VuCsWYP5TI9O^Y@
zh8{m(&r179tV?8Yd-$FQG1nuUv=aL_SF|7JeX9lUglt+l*xf-Z9Ugb-alR+yHYMg^
zHvC*N@{g~JI|(dhDH}#mOaA@U)^2OV?U*g^fLN&82l^W!=WMkc6fgOdeSLIa!2J<#
zkp~eHSakzOy*=MS`UE`9wJ6BYb!9gFrH-0TVT=viGreX0-KW8w0}KUaq94DHui1Kk
zG+lP`NX|=_?3T^yZk8=7$#$7%niWYK0``N6O%-kNWyRln%W*$^9PwT%lT&-T<uEFh
z$9_!PGos5=r+tn!GJj74>I(jx|8CB?5%wtNpQ8SH&VR7DRvhT{sy6R%(64}Az#(VA
zd^n}Ra35dOTi8(16h;wTz}0WwzAY2N`iyhl4A^mH)YN`_TE5DThw$NHJs6YdHWt|&
zccQ@F$=UIZ;k&$8+;Q+8adpKa@2m19Ttn}i3<OQFT#Gr3gUlK>6Ys&~1bj*;-aD#K
z3et-kN~^9IYwJL)Tv}{uyfDMAC1#h9pSH>9>m<s#WKKFJj@WSlRPKM*P9zW|OVv#K
zE1S&mGF4&9?-s=WEQ+6SgdL$2kW1^7CG`~HW}~I|>|d6ahyxIzcC+i~?&nm$fLu**
zCk?SadN_VREX<OXnwM2~myZ6owE=6Ixr+i0{XA?Tm==4<N4q&&N?;`RKQVzr|6af~
zY_NOKt4srZ$)^(3S!9pCzd9-E#-svB-jFZ_$~3GL8-9TmkJ;PY-mXhnaFYp6HNoj|
zn}b5@Yv=k?RU<iBVQ)vEz=>BA(qMos7XI9;SGA+ZDCF?g(Ro2}{R`hl$O4Z-$I8g8
z>r$7NrEmJbHX0tuO&3mI`M%E4vDWpskj87sb3bp^hYd&<YQ68ZlRCr;Q>%SR)Mc<9
zO~wjQa<y7du#?us)hgByA<{|Q4c`&;uj{CzYCgKCuNFoOSpIor{l?LUjsy}B+~?}A
zZVEAd3`_qwT9oj)@bOXRJ2XmCYzEbp_3_410%J7ykzMi-NGIJiLkl@SQ<TW7bn23_
zVaH=qpC_qhcnw7*F*#I76-ZFR+Fj2&gMPG?-MQ_qDQqTX`xKL&>R1r^2bkw92@gz*
zBmPt0nnTF(_ieq|e50*OYYg(HHn_+y(ookKK)VJq7AxdD%pw!~*u#T*9~~@QE*=t8
zf#~{?9Sp&8h1gz!e2MNRcztbM%kHMwbgfsHu3B3|L&f_dD~!TbQ_~ez;*%4U`!(y2
znSpQ$3JuG3>57L2DZ~|JjLjuQuR{WZevF`!%Qi|Lx2%)RZY+@VW~7r_E)?C$;d`(R
z$z%S|qCAexE8jLE%9Hr>$F<_-Tzbc4yL_6L*L#XTWfSgK<`upbUIm1?_|JX36V>B~
zCbU~`-)*0!i`yg+X+1{UOA=48;>(v?#zsbd*k7hsCL_e!fMzfY!HupuqXxoTCI;c<
zpcdGpM>vpc69u~_>>xIaw`KhTV;x^RCcjJ)h&jI#^dBIc#KccP^jgjc2DP=e)TMG|
z*T_L=Vx^`{>*?sXKo}B;2i_9V>v4VdTLFX<e4fV559<Q<Q1p7d5Be)%H)qG4dKC9F
z1X+i0{XOJMW{-|~<)5}^>=XPlKa;&;37f;h@s+Z4sB-i9ajfuX=Iw=K+>i@JAz4iw
zZx!ap?pUHV`N~u-V)UTv&2vA6-}FbPOAKi4Rgl~`7K<Bvjhw1Gc=`A0^}uua94-!`
zrNH9ndII*$T-=2O6scU~uJI_MJWr_Pn=~E%OV4ykt@P8tvBrw!WT&*t)Xw>&dwLdX
ztk8rW<}}Q;%`s3Fqc|;*0`1p}Yh9EQjUt(D2gWd2Sug7sH$fkSLtvbnT0vzz(gIqQ
zvjBIl1PK}7k!=O?8j4HqYUy+7Fsgpi4;@qpp-8n`Din?{E+fR`Ta1a8mW05T?aPto
zG>fHs@;62=OeNB>Pl?F9hV%46{tt6{+;nu&QzEF>pL5WY-snMJQTlvXfo~`q{J2nU
z9GBs7`wEX4;&!COG|0QTZDaE?2C(B|3yHfD0;rE51(eT{bg`hX{(iS)nPh2R<tX%E
z?FDaC4Nop`03<+B9JIN}`dXpiZ^K)F4ucl#gclz^v}{(aQ~&4Es*z%dR*PcC7qu7B
zFGJ8dS9G2F-4TM+T%<tv<D%bf+~pA*9Q*^&3X96IZUXX$)=pZ=vF2PFCO=zWzjA!Q
z_Mk>8_irbmUGrTpb&Ihqc;|ww0jo-j3WBd*KYcx{EXt7y-QK7>Oy1>0Od)Ou!q^40
z=k*>33Nfz*JI(aDMuuH}tY)(1GVGHOasBl1GZB!E@;(8T)ca$T!=9;25QE=(bD&0n
zgpN70Ml&0|gJFMDf;wb0yc}ROWXDJ0AyUKok6JA~S>kP_-?V#-g}zgO$0Psu@qiAG
zJ%f@)A3GJg7DYIib>+&HgZC5ZCNN(haYHG->~j=mj?T9Yj6oP!62L5Di!3NKoh)~U
z0$Q&_)SyEcXExp&P501>kaYM>(@oGM+)MHM`X<`W^NlPBF@t=f_A<LAeq=HD+{622
zVf`OJ5rzr>Qd5KMpJ_#AqqL+rFLmp&2KEe;%YE?a+F+Ewj+@J*v0h|Lsc-JNO?{v$
zM{Pif*P&v;B291MdZf-QrM)J^Xg;9#+D3v>%`~r|?57v^J{Kqa<K81>y7wlal}+(4
zlNl{=SLQki^+he_sM?LHtCR&C(8CxrgJI^bs3zJ-MR;@-%`GN@jlv0|aA*WCVO=Uc
z9a#VyscOJ<JzI0Yw4*PH%~Pc@%KU^aCF$f;Y2n2(Q48S*`AAKzx(K0CWwx&8gg$TL
z*sl{(Zx^k4+WAetJas!CGY!26j(YU>{BQ-hJww1>>9*s4@`wU!ih1@NbOftD(@@~c
zyDHi7H3bBKhM~-@`}Qd0iYgsW4lS6VAxwTn?{~#&_i2Nbmhon7e-jS&R3yrT;6M6@
ziw)SJ(!Ha46QR|l{fo|%;wo~yv5bp+BaoNX{@0BEKHGf{kN!xDUN`HRsD(_q*qqA9
zdf7bao%)yswuEnr)9=jMTe=#*#d&?#3;Jlp*dLqC2KfF<0W){XxXA^{n9$AMH|3Gg
zZF)jfhm~Lzl#k!{GcA97RkvI;vn2^~9D`S~E9fIe)=$$~(xI6f5+JHCuGU8@CMM^8
zN&;OT3toTn7&6O%KyK|F6R!7P12who02vG2zh8wLemR6=V1(hoZ)@7wh8h0u3t!Nh
zw6Rx!t*!DG4g<U39CAE3K3U549v~J!_BmGj>+pwcBlBf6lZ1W)84WyT-h!fAoUIrc
zI|P}XH^IL16C~>ZRUSiTcCyx|GO?Z)jDwwjG>Fzs4K<s!2;#WS2Yor1z3QX5No{86
zt<L8QJ$D{S#X*$(Gz#(?6Nf0on|D6YgzZGe+{nbz)K${PB-?RoRk!V%bvXDbb(I79
zd)((j&?ffez1kd4Ap<b_Mb%KZG+1ZmgbuR(i0X)3q0*U}R!bgbXoUhd)t4s+rAbFm
zzJ&qxBX`s=5Uw2k%N%-Nu4$E<?x=rPe3vU)f|3M;Utw@T7WL5!Z8uaa%V1ctfO<5H
zdv0#-;h`~=us0o<V1L(bivzm{dOjuuFv6BAyt^Yk?&|Y-)82Qus5RHF-*Sm+kygYI
z>LtRle$ZN>>AI9W&u4o5OkbRuW#1o?N#lXOFEK0FO^4@o{)a>e0=2mSm3g%ldQFCP
z>5icx4U^QwRGS>z)07W&WN~;pY&f_E<l9t+Yq!KcIl|;#($k4*8-B(U&9*EcFH-A^
zK4l&5ahtx(f*1?`#}jfvgY~G}K6}KH!Pd=`D`g7e9~LAa=*xBrGY-x1dr?{-pZ@XP
z>-^VIQ0nGm`7ax4aJBQ1B0{NQOl`lRhOVaS+mghq_w|Px>?)cAWvsmP)M&?UTTkJM
z$dM6$od51V=m@sh3v@qUSIRF_-}NtTRd*>g6A?3e7Dc3MQRh*cj<g>~8i&!cgjS8O
zCMko73}twbsPJWAjvu>^rPKR7cZd1yXh&6d8JB(@K2@-w<@_$#QV7b^lGXqFWweDT
z6(CtGxqSW};K&(W)$j0(hNW)#U1_lF2W_eCrtT6;k)~gcv4kDOAT`vNd;J6tvEt%s
zC(>~t*)|;JS*aRa_N7n@*vuDVBf}(wh4O8myqTQDj?%VH{FYSAF=no`{b7cPgOr!*
zLoXcQ8X$!)L_`N?1}cvf<x+bFAx^wRD=J&<n2<5D>M-=>X9>_ceGO-$T@HpI5`qjr
zlSKT^S9R$mJL~2^g&L^ZMV#L;`In!+N05D{m>4QL&(y^i*aLH`Q2Xr&H;}iq5BU4%
zJhw~R9^W848?baZH8u74_VpjVB?w2JlfTkb$S=rIPRE)@kJw;~)3KzkP1gRt)<I2x
z+d6oO`H*mf3I6DY*GEULygvbO)<KR~!XlN~yWPI}$Ulnn34(o2t13XXwCr~X<c^C4
zi5!aNk4z_efg@utI4EedO6(DVC&R<J)){}h%BQ%K3~VacIbv2D-YI=s5}KUM#@Ct(
zLc^SRYNOJ1cEWB5G~03Q^wJNF9>9s1RO!IurTzPO#f0+F?5x~WsvaoI-^9ekOdZ+)
z!63AOIyt?=4uiv_XcVk~X-lZ$Xx5H!3gs_rRTz6;VIrRXsWEll)y3`Nbo$1LCq1*Y
zujd07-3-?I#vj$>Nm%nf#nnnG+9a7KFuO1zf$XJfP>b&Ypyo<BlU!g+eM*dI*q=-B
z$uhy>n^0nK2>gyvEXBQe{(4a&AJoaBG}(9{)qU31lE&;Ec=O@c#u|YH#b?)%&}{#R
zTl@}WMN3WY)xI!ezz@g)KPKtFm9WeFW(8g!v$0VHt~||3ay$gRe>Ob)Ofeb`?9cn`
zeGQZXD~37pzK820B_9RSyhDt>Kio>uQi&f1+Qcnot?)_<xS+Jx2stlx8Iv(D#2DBv
zdd3?sDDw?w;n1C$o=*TnVwN%21ZqJ)zvYJMvv-5z4}RP4J9^*!-lhefeVt<NG|wcC
zB{+Ak#`)VvWk8U!JZK5-(B_1DUW!p-!-I&vdq#AZ`%%2qr)}|}VdK(z-4fhU@Kugh
zwx%zyc1`fZUYmdgxyHmv-?+;Zo}i#=OmteJ`!7l?U|NMLCB7|3&E1O`kuL4O-l55x
z@J+9Bs@Ie9itMqqH{}&cxxepCu4LP@)4ICi?Rie43Eo;LhM)dwln!62V5+;`aZ5hN
zm8ACG!fL|P+rH&jLLH5HmZD40!&>vDW%>*nr-*ESRs`yIevv;17@P1xNu*W+jKUbe
z;D*M=Ot0U!eFCIT(#>Y^D9Pu=s~d9Np4VWJfca)w=g56I=O*x)1g%vRy>op7;=zD~
zV%aC$bN3v5ekXmU;<!{NYfU4+^-rF*7_dHF9CBKGMo{YTNt~F*H-mEf_h)}jx9gBY
zo~GPN@|G4*VAFJa+gNjQdis;_T=8%^RhIr|Oabc$C7$FfF#EYG)tUT46^7j*h;v?P
z^3=OF?cDgHtC-Ja79-m-dA~!b2ppg_h8kL92BKcQidJiwcw$JsO{ipRr;v1qR%C2!
z><Wlp?v?<#3v!z`$@n8|VYkf~WR_lt1^mLVK@SLt>{Lnbv5|Hxb#J{V@Y2-?F^2{C
zLzqJ!-N7677_v>>p|dnM-tk{lnjP59_n#PxB%t7415ftvlEl20V8iJpO|k4&+C!09
zj;bw1<tU#`kQkii*uT&fVGn?P5cQTQ=mB@UHMzp(z+T1o>tIs8#git{VC0a~8IxSJ
zKrU-rTYl$EbdLe%La>c*#7t-SbXs3Ew4%v7Y4Lo2sBT2lG!kxP6EB24z;u#(y5R?{
z4VTYeX`Hs&zX{dX7J2pN>?u&nfZU)DG{GWTAPc+bCj`cUYNz|UOCI-P+RDwD5Q9e{
z$0=;%!HACrHWZ21H^ZUeqt;5@{-)|JS0Vo$<0oyEMmqz#dnt8;B`B|t6SIb@FIWyt
zBQFSgh$>Hb^~aUnY%a^Eql6w0t?+Q8Zi#Fh2ne!)uEfJP+l$i%RuqA8j53St^zpZu
zne$U)D)B9;De728()XbnPqx13%!DTq(mK9Tg^?xTgqD&lXyNWQtWUeVnzo)sSbob2
z$ywm!h+jT_Tgc&`U;Q_#=ewS|e^rGd;y6eFfDoFOjDbFSE>GKA*&tDKf8F$Pl`KvN
z_p2vf!eUxk{7Wxk%@4<GIxd5<CEz~2IK%K&86045r-+d9vkgJ`LPY@Y`RRtaIV9$I
zwrFjL^%7s4{Z&QMc05vY`q2J7Cu)1vGcJ&QOL}`LIP$O>StSi8>1+FwT~oLHJR+{1
zCrXOu-{zGTqv!G%?U%^lLTFzl+)ZGl46pOhHGx?Ow2AJ3H2OrK#N2N^8L2o3TUFzf
z7Z{Pa4LLqpw&j#{ro$`{aAcY_DpPVhUt!OcjvP~bn&EZza<*rvbv+uKMp{oIb3ky`
z-Z@_Pqv#~165iR9RFIogxXwd%in^Fc@*w1ONy*0CB1l)w)a!bRtH*{qIf1@8$SY7J
zQGq}5x4C`eNr_}>`ZY@6Totv=vnEun$}Bqc%1(Wh!d#WGlK#bK#+5j+1-l}WFFWSi
zI-hy?RTdXF?WWOVF?8+B0?Geya>_cqa&A^h3x!T%x=8Q?pP%#|Pxi5|r2*6rIh31e
zd-&zMyd^#;M=?<!4Y1fr{qnClU=C0yWVH@3`JUFz4;$k7$`!$+%5T|`9*OgmYICO;
zbN|Wkoqg=Lj!b(w(f*tKqQ6gx=saJgPTqSOd9`Y0%z7dyqViXjq;xuE!2XEoV@F$C
zTN_X%uEheulJs_H-Gu!!mA3Zz`B!bCSo>!>Z5;OOh{dl*g0D~MKo~t^PHBwR=Dp-`
zz456~gGIrI@&^$)PYxZ1&#d)2wx#KJ1E03tAy~MT`blo^uVee^gSi0&VMa6KZYY>F
zue%)hoj+?v7m0?v2cF%3R(uceMJO4r)wX><>d#RqlW-IGS|(JA9J-W%5+t?w;5HEv
zgyRrcIz2sA?9f`e=|Yv5m}tr@sF=%qd|=hC<ScKMGZLgJC9clZigG_k7dkGC$mSOp
zHu{FDJg*b1yK%gqs*#ZU>;=l=X@YJo|1fQrT|zDG$cJ(!c;7%ay_|GzDOYKmDCL-E
zT87uMCm4n}<$3ewO*eoUVV8=eL7NoJuiUA^koiV~>8C~n?X&$@r}rSi-*P2pozem%
zTWe)1qAw=;L|8YymvWMywf@#{_##xS9;Y{X>b*9S2nmXmZ(Hr$YnKy$D7Y{+MUfq1
z2o<@MkwZ1xJAY1@5WfAear0bpJE=)@_FIHPdisYC%`JM|FClb{|Gp!0R&qQwXmJw*
zv|`08wO_l7g^OILPK=MA*h7KXpAO(m^n~jK`T5DD_Bo#4I>X=EJRh++A2I!6f7j9V
zygLR#y_wQT5Drzn*JA%hTma%<yYi!JUygealD2p7;ttOS9mDwn6_I-Z)4X*{gP|5e
zglRi|EI_}mXGC$ogqPBPr=GUd5UXBIn#VNH6gF==QjU^jIzM1EwcI80S>@FTdhnp=
zJGI~bmTm$-K3Oo`VQETY5)xhHP(Mrr`qwiv0lcDrmiRwEoDGC1W1hgViUOT|e1U7V
zgRB5$08|zt`I`F96%S!?2L7^OnsU(&?Fy@=(O@uqPG-h$%KfyEZk|pfbU`XS!x@g*
zSmTL4B55Zhfbr~lmjvCTkx|__QFSOwlNSCGTrJPi?P)?gj!jO;<#j>`II2Idiz_v^
zwvj#ft~XC@ATM^}Jv`3}ewluj%(+ZL6)9OTsBwOR2mu03V0y0a2dZ-}YS(P3a{sv3
zH`969nSeayVvqj&LyE=sy?Y@ADegM?6%3BGBDSSOTm*s!!%af@4Je~s13+KhlVXY$
zZ@|C=uWy^0J@`QLkNtM|vj0r`?Ql;FAkg`!5w@fOvX-L1o=Y>J^NA~*7%JAjyLne3
zQ>*F7%#M`vi2A)k?#h<GOB4bhA!~`~s%y(%M3Q_GfPtHjEgjx8ChA^98b`;((DJ(X
zGAT23jpy51lG}HQV(%&O6x?CxGK(po#NZ&2eO_I$9VY+v5Cvqf-wTYg|A?8vDd$W3
z;vE`kF+bBBBhDkbJ*h=Em}K%QTi7ivFCu+`I{4gB%O$*o;i_fR3_DRn<6BcO8H)Ld
zyz1@B()<In<%Z*IzRnPY>fC!jZhahXvi;LYV>vM|`6=%l=f2FD<^4k9j)XhHkALTL
z&YX?bwfc&S#i*m6NwsA#=+smi*bXykoDe0<_!ic$wn>P1Ld4;Q|My*Ay!C>>SN7wV
zRiIY|x~&S{6uFL%K%;Mu-$t5kPzak_KvQ?e@$J&LZf&!89T^l8c4)nv6{c$M(5vvJ
z`53_GNvoE`9pCqhlyGRR;X50?_UX8ee@qfF@s{myPqYHGHa=$H%>a7{ZL{p3QahyC
z1c)jMPBM5+#%Pv7Xc>yKuo&mJe8waNM82YrhR}v_oti#{H2<|6xF1obF9>wwtA_j%
zSNlLclmCdr_bNCunt@%Jh<+7<p<N>z+qtOa548#r?7Z=qX)56>nhseOB$wL1Pj4PK
zW&!mgBOFu3{j~n!{heTiYE@S3gT;p6KFuBsKJ3h+PpIRTMvAPM8TQ7F=jnsRmJX^C
zlWN;Hl7FzP5c<`Jz8XM;99JQeGcip3Y?Gqi-KoY@Ao)g?ii#?Yt-TF{1GKMnG{4!|
zeOHo;uCIJFqeuV&CUuk!54I?aZw&f>mN!6@-&E8)UZ*Z@jcd-RV{l-A(V=BLO5u2G
zT?)h=Gj}?}=yyMCRhEcvw!=a?8BM48(!5t$Vkt>$t}?;$Rk)((!`<Btxyl*)iEtie
zJTSIwPns*kSFD58rNm59DBWq%q+j|p56!$%?|Z7MUG|r-%Nr1P0ZQ>#RIfgo*PadS
zCvh{C*oMqIa50RLyHN68vr{*?UYMW1M7@~<f({I;K7O3b%Bs$*t++akgGBeX9|8CM
zLBAMG4a%$kzn6|a1tV>hY9;dv24+U|WdzFi?q$gakB1Hd!=WA88wK*&xCTp)_N5QC
zweBf8;mQyqkZ#MI-xtdRHm3J&8&5VO$csZxnnh6j4;h;9tc84bZHj&^F%1O+>24qX
zu;&5!cd3t_I7IfD5}(g@-AQLc<J;7mf4p0r>@QybOyU4yu{9VDsvu<Aj1Y8f_GL>v
zbyEh}qJ&)m9;k6``q?+vjszM0duuwswl^!i{TVn7>apJp*bWxn;Q;DFx*pjjrVgxd
zQ=fXewoAJeB}z;Wet3keXCvi+=x~XAloRkivHY1_4Dqa_iYHFInzjRo$*?yok*!PL
z>k-i!o8os!tto$}Z3GciHQw!_(XKr^e)%_WOO5lyh{Kt&2!>VHS%y-iZ!Yo|f`k|H
zRv^l$qVJNC-Ae1QttcG_Mb`f;lBDJEyK-b~c#nrOxN$Jh=I~OUL|y?tIlBl}T8<JV
z5AL+1MGl1l0k7(HqN==FZ*?lW7=pp27k9$rpaeRags!AgjzMz}_73O!N_p~S6MBSH
za@=|E7<KcAB6ixpz4-`d$`PVeIf|}T`8Z0iaw}Hok5JL;Ih&7Rq0Wqu4ir2(**d(6
zY^AZWaq~z@M#fP*h8^#y_YLYi@bf1e%JlTVk8B!-Xns%1CMwm+iq!97cHao<JRhje
zjJkDn1cN}>TQC?hh#dw?)H1wTI$^p6O^CS9PR9blEc`+z(WQL@&Q!HC?lqJc52{}D
z8yq>rcRzq=0~h)uAczP>gdED*r<yfu@cyA8yL_o>4DKu;E&Zci6nlg8eqQB_{cb9G
z`)gMGBP#m;cZGmGMz4yh$A0ao_ocb263|ZAPy*XTy*~$+P@(owh*F}EbMRFRJ~T9R
z$mBuCqi50>ZU~+Vs?H!B8@;AAUHz#6t-$O3e&8EdCV8b1VJJafe!f}3lY&T5*F5GZ
zyx!QD*X-b6)eu$=Omkac&~PD=v8;RiH_YwN{5+Rfu;ccHz(U3^K9$&v;>W1k1R7bM
z4M^Yr^??N<rK8!+?ag|H-^G2#;Nb^=-_Rj|f%~y{w9G<LuX#f)LpxBfXNsa|S7<-a
zQ>-YIg=B%X<HME{Ve;32>`(WSUHg_ColQ!F12|pywMp(<AY3ipe3_eJfBh}jq<^Bg
z`RscEQYNmpsc+%CI_J3Q{j&W+y3Q4Tf=UP(KjriypQyX7!tINpXjrceFw?2FjuTAM
z;DWdo6N+HGmpBz{|JVB@b+!|t>^)}1Z;i{Ny)2cq6IkOlxVj3Od~Gvh{WdseL2eTp
z+_P<#*aHQxtGv_sG`lCj+QrW8y63E~BfpFZSE~!U=jvORif`r_41q0B_^H-8CU4e5
z)504z#7^srS`nfgO9Lr5XGeY6!%xX^riX~S2lK|1iQ0Ux<|@KpzvkWy!74x_pU}pJ
zV)Y?tFBWHJM|{QykS6-TePJi}f8H(l`g`p{&SWl>_B2kO0jg{s*WFkfV&&sO&yVtu
zQD9=B%Xog2-0+KO=eh+S9rep(g!0XezR++j%MvXTl~V*lm7<tUL?9zO`Nf)xu7r+!
zUMG#y+o<8>E4;Ph0?HldqrTy~a55zHWqOYbm#Ixcn4@Et<1VPtIqhi{3#wEYKty@X
zC#L#Y?7dua90)^;M_~i&H+6)}{=WmXj{?1L<53>l=G@1RkFewVl9?V>rXCM@<=?)2
zi*y|~zyX?n)_Ml#_`c=WVuPYwEFs)8mEj)uX9PEdrfSx61{$Zu_S40%A4VDTfMIZC
z>6_|0;o_=m@wcgO{NDJ<SKHdxyF&UOyuHQoyr&4IlEmUfj!EfX*<&6LfhnPAyrA~a
zN12I1Z~S#PF2V1=p`<Z@lH;wR!OlZ$1`8J{_tw$XI$jCF3SP6Iky;{jQ&x7Rwzl?o
zd>H1|&;DHJo-9|g*LN)KDIHTY7lfBO2>%)UXO8oZ7x-7;{3~)k20YwrTFduo&y+}0
z?HOV=RBQ7p$#JU>lPr_?Zpqy%(*8mXJQw*VwG2*FL8Wmi64WTSh<~_DRza$C2}?W&
z7(@a+fNiCzYkq?P!uNkR6o`_eOb?Is_Fq+D>ndoFkLh!0ymjM7#3b&37?T2e*b=nh
zVarAJ`j_5kx>Z=tTmGF@=F5oHx*8hF=Q&hnE)aWpjAPNVLW*a%m@GLtws5Q*eZ}PL
zJKI~nOA$9Mmyut;-O>83*qkT%VkH5-RzRd-9Jnb-h!VM3+#O8xcBefxP%apxuma74
z%R=Ku`4mZaB_YN0%_DSG4_06P`Gg{qH1|&e*w8vt*Z2ou27#lmlt{Qz{MVs%8-LH1
zWVy4p&GQJl=m|Gw#Z`{75CdoE-H|p#bFNQqbc?A?=)P=IQ`2Lc>p!OaW3!>pv>M*j
zacz_rCC<O|_+p=*&Wu@rM<?m05P7wd7;5)D@@M$&zLDbY^uVG*1F{jPPPV<uoFZZN
zYzNOje81u?7p2kdqc$xm-+KJ?soe;22wW|BI|B~Y_k=gW@Ik)2N8F&*6kVl>Q#UkB
z1kM(${rJO~#*uEQQ`gO_BTrBR@sI!ivll1i3}C_2HouDUYkC=15FU0sR=kBkjByU=
z&K7PS1#V({p8!m%5?~vZilZ&+biZHtNfsnEqKfe$x7a$&BcX#8@P4tO-w-m&w=iuE
zQarLUPx82vq<4z>p<x<9c#7B1r^rK(rFWebw_!)4e@+k@!Vj2>uCdh$HS?I>4n^@&
za&mA`GpSQjQHj=5FQrU`kmAO{<k8#L<1dX>COCOyNBWG*CtS>H{GWxSwWr!$nav!n
z0@v9CMBhNYJf4-`oybnI*hBo{p(C@*mI`3BFMEI7tdHTQd<<HCtLJGp$7$Tf(c%?P
z#SG*r5tDULuMd*Kz9c-oUz&R?VFA|I-7*@aoX6#;zd_B36|K&H60Q}qSkxW3aQj~R
zM6lu*`RCH^kg&XPVrBv>LX##lEFeJ7ZXrQXy*$|H-F<KGUp`~j!9Q29@Xd>;ppyf~
zn-QmD@@Q~IC~k67E@g|kkT;=cqNPCgN$wQI|JoH%Nz0K5sAQt5%<|#983p7xInF#7
zD$PnlOpHsoZj&$HN!C*&4oY#FTW6k^kuQy@EI5<Ro*V(6A!Ngn=Fl_a2RlzoF98Wm
zec^;Q&7xlPcG1)+^bVCfruPxm7>Nk=;GavUL2<YTBo%Clh78db^~8V$R|-7|?9eI^
z%I^Jf22_8~Osv?mF5Y}24QNDF6@9|G=h8<j3X)g%{}<?yg1M`TT}oZTj70gYhm&wr
zt`iFMb3?9R@E2iEveH>3rc7e5%6X@8>(H%@jlJcSmA|mmPXHdj-$x~n`Nv{c_?e=y
z5+N=;7>{U(nmPTjzM`C1^4F$PCF#=3lM;DD`Iq7ZR#`o)=INBog?f6-ntwKQrCfFl
zU~O4`;reF~D|v8(iT5&A?RBRPKWM6qGlr2K!dpj~1peow0iCjGzzCD_KX^WEeD}Bd
zw18Mfi?7*2GcCHzY}3-Lu*msRU8T|++NkqXLBlZD^OOCL*V2XtuXbSn^jXne^`zmv
z$q3hwwaH8fa8401<AN}jG$n^xb9zC2U1+bct7q3VMz7gD6y^F4XGrNay8B^?bQ%Zj
zcLQn+7ye{>>S^3@$gHOGmH4EtKuM2d4{rJ)u}z$Sh&L{SxaLFU|Go=Jor9qK?DI{e
zeMHGam~oSAA(3@fP=DsQY33|kXfjWWEynfwEe=OPqDPB=>t4<Peu94l%xiX7YN1nr
z;2`2u$xv$JAy(xx>}S#GL4o)0OTrRRkUQB@aHiL`215fq0V>u?cPlCQm!HxpY(Kv1
za^KBu(_)-AlOPzYxX5%3tq$hZOKHes6MP^Tmj5mx`l!&T7z2!J_M4_n`=OB2gP%&-
z_=~<P7O0Of--l08i|qZvApU|v0STsjFD$3{tn`Pfy!+8Fc*Xx7U7MU<c)f4iDI23{
zL7P<^e9aAE#mc0cEhkv;sVa0fN$by}hnr$3FilzdV6D?`cgJGxWK`Tf1S#?7ySH30
z7Om%4O_5&tgp4B7lVd48icaj=d#Llp2-E*L6^04`1R%^SQ0r!`y5^G43WW+;C2I9o
zRl$O%M@G>1eD4%6A#@eIjw}GJc3f(5g{Sau;=`vmeM<O<JLudXlP5teuo*06A`x;C
z8E!tZPn;()xw(^p0P`I+cD8OL>gJRdD?9w|F;g+fPV4Bfz5lI613JrHdqY<CFmbDr
z;Y!ZHN{gk~3h2=&*ANnP2|T;$Njr4FW-MOWLv}UqhTrBma<+{{_`hp2a>hhRzx17_
z-zW_;T%Xls>K*<cuD&~-?Y9fNwW>ixOQ}6Nlv-7L?@q0@R>cUdQbl5~P&2k_e=V&|
zjhfNgTY_k<YOPukN)Qo>7%|?j=Y5{{dHVV*AAk6K-1o`3&$-TZt}9&}+R%TeDbR37
zs`EzpE0?k9)inY(1c0_E|2+3crgU$x-?X-<FBK+LEI#cUzI4)UO^Ds-GHV_Z8km2>
zs=(OfhmB6bv%sg0n(#ft^YqO%Mo;D>MDDg>crmY);ivI5d-PbL;jtG)neVheQAEDw
zAN@%{!Z&w>N17Xz>=DZ29Nk{tg1(Dg8(<H}pPR`&X8ry5q;)?4EHsvNjJ>SrM*^>z
z2~XW=?}WK3xeS0BFul2X{W%-UrTOX+*<6iEqcTEHPLA40h3!%80~VQ8*$LxMd>O8x
z@|@*&e$?@+3s}1nJU+h<^Lp>|T22%ZSYxMovAJgIWjqIb;tSEVvgU=&XB)Tn1L8pw
zXI%4x%>ley3lFpZhGEHvG*nil@Rtok=1jE^>-0&qS)F)psE(9tM_3kbMUpV9FetAa
z&hYePuiwU7sPwiAV3z&ahx}MBu(h?-_V)RrY?s^oMdCncMZZxzX5GGCw7SHa5`5|y
z%=~Y1K1S+fI#@<|y5ve9!p%Yfv5xqLa)>1P(gfq|+Tx;9!Iv*TQQG_-;6O)=jpMUt
zuDw+p(MMf(rtu=x_uUaZ+Kb`$yHc0B=s5@`Bj?;0zfaeeKl|vFhwv#D)_WFM7T1)i
zFAfR^1fc`1D1axQp0b(ASiBS*G|_&03RB(@S|{#on+J>=9H%gb7pQQxt1o}xe4Nj7
z0<IhX4gZffzcQcp96XJ<lc~qqC6}wr6^%KOyLNa$BimS8K_ZVGOyxpDW;<dA0DwVr
zzHB1DTsVw}Zmz{aw4&ua|C`Nl=A|5Fhj&(tRof1XB5WsSb2%;^&YC0(wf^$T1pvWX
zd@xN!;Nc_A{&^2Gh|fRzuxy7%D$)lS6i@sU)<C1I4}ZTB@>OkbVVAmYD?|Zc1!_d0
zH2DW$!s!FJH+Ff=em1-{H&!2|!G((J6rKIp2|cO?;*b4CS@y!(wjkI0n=y6<+=zxy
zJL<Rf#br>fXj;XS_yAs=2|*>gwHUcijSzmvsh8cfmh(67!WgR~wKj9u=ZQpOGd$vY
z`A-cv%iAF3$gOng7j%z9>UyQ_nEoBUfW4ha)nx7gp@RVLgC*b}!FbM+0SB2VZ);#j
z$YE>>A0Ezvku~1j+OjkmTbbv>?{pPRxHXcq`sOPBd3=5#E{Ilcmvt$if$k(*U~I?I
zOxun7&)yC(iRP!&vfMoBTqF=A2TVLYCg#pu1eUI{jy`4WeL9ME0O&>oPp-jP^J<VL
zE|Y1^fZCg)7N<x;+TI7RN?;TkIs5k^Dse_}i>&Z6IqGVJfAwO<LSpsKu4KM{S*vvY
z&05i@lXYimx=e+9{8w7F%jjc{-K|mEBPBpv>*!`KaQiGtvHa6gpBU&66pv)no`|8-
zL!6G2zSp1h83Bj_P}xUzjb(kfQAiV5otY3kTfeFk^V+`BZ!EgqUEYCt8RU>WNC#6+
z>9HiDN4TYZ0cTuZX;0RPlLHr@+~@h<xAxVgD9oCuHcuB|uf<y|ivYuTwD%Q&sId^7
zw1A)p{kl>{M1JWQY@G&`_oCFQLRLn`0iS$w2ui;*POZ1*;e$BCySCM$V{e{hi}P!6
zqTVo?48j?)0_g=siyjaGB=Jr1@L4brqH2_;4G18`gaarGOp>eiy@Slo@dm1wawbDT
zn1;4bS9MjPilLalPf_l?2=~j=e?EWpFfmc)ml8M=iR6kz*i2cp)kk$kV>EL=5;g(<
zKEP2vm`z$XCf3R~S{*KX&P4Rg(9-rYy1OMu8?DcIHRe>bb4ANd6J_v1oG2~nl!EI`
z_3Q(15~J%*rx<{NeKZGjch+cG3He6?e0(%Ajnaw-tmGpU495XRl}QcY*$lWLeRfF3
zqQslV9F}>wTcGbG{(22Qo_0R#ak6~X{)%w`u|I<<yz=>V+b#*9`2J(u_UWlUfdx9|
zMu+#DMWbSn^8O(^$Ht2&w>m=9QL8?q{^@8eO^AIGE6;fIA|%FCyx@&;z@=y5A281G
z<F1#0okjgww(1m?%VDbRj4BF)X#vmRsXW%0naO%j+osh9Qp-yRZV}f3K$EGt<f*$0
zi%UYFxh;>$k9?dCCQ_$I<f_G{zRq7?rN^fXZ_4P_b)k*3OCIvb4F;$w{c;n_$8~kx
zA`<vBGc%<-Z#V#o;XCs3^6&4bS8>GwntpuN%>>yZ4OYa?mh}(Ws;Bw^DU1p+fepP+
z6$fHf6eX@*m$;yVt|Lgf^l3_d)P2PZ_mLP=isfOoFTTj%6)zyeC$jTwa#jo=QrBf;
z)_{U2j#nh1hrr`J#52O|oGtewR@%${pX>jV9|2BS{A(MA4DoIH$T2}ugsOe6Tzev*
z<8*vQ%29efWJW|LFzG_40PC_PCATkcu;p7UNtV~mvb=foChJX6XgC?D^jDE`&Feqr
zd%5Vj<tEFclZM5gDBFF_4@K7;F&Gd6PCTQwIE0F8X~_IOBP`16x&_zyjWS|Ud@Sf%
zV{&X^TI<IC`n)}7q={o&{6!&5C5Nwp0qp8k-9HQu#L<x?ki%p<nL~&T@ZFO=C*$l)
z@6D1{c7d7xsOf1cAYAd^bRThD<zN5KWt1xqF8Z}uGEZT~jrgyA>~APQlC853b*S5y
zGv$vGW)0l|Bu+BluWV0xXKDGZWsd?w^YbwuhHDu_nE;&t2=?Zrxc1ct?abx9PX$Cv
z-o;#1qRo?aFF<}JD!f9E^lHzjD~SB4eek>3P2>!7JJGSDOYdI)Keaz?oG%D7=y4|c
z=dkG$LiP$n_Bw$E%T~qX&+nURJixLBx13|D9-iMNsH)lc=at^N@QkGKZ8!_?Jk4H=
ze+WV8VbEMCJ&qqfMgQB{F^k+~{}LdRU~Us)LSl39n7Y#tXv|roj56yo@PAy%lUNQE
z!`l1Rwen~81eaM|nqlekI7rO=Wv$(>lrB*)%2_|uPwtU;ORhX5)~ZZXj*;LMP6ceP
zk{8Q3<=D2bk6(fwI;BCv`7-WqGgS_rz93U19z@LW6~sLKc&qLYJp@&m&Fl~(@DG$6
z5HAB~PGOQzjsmy`dYonm3u`Q|Py&J#_w?dfz*6*o>Vz9_u;NQ`C&Fc^5)34{OFt`h
zc)C8Sb=Z~7IZ3;a?b`L5jk_F@ZU&jVjG-KO%4M?O90K*opNCpo50B*DSuDxIo7;8b
zQ<n^5Qf*k7??>KD#y;!l8Vo*yLo^biACrt<V0N7AWJH+eC>{7aCd$p(NI`Wj{W}iq
zl~upKTPzaflC#n!x;1~Mz_56=pqh09*ydZH#{Ye^<7l6|w%vMkU<<7KY5DQ6M7Hhk
z2VJqGTdd1*a)o_-lpPO@1e(EfXPM(LJlIF9BiBwdW&Is(RZL#2wqYI!<&~NrAF#7F
zdPBTX+WTF6?*DZ7+!+yDIGYGlMiCblR$RR)xjF`y4*@j_h-}jOulb&o5VP0Wmi!%Y
zgYQm5JK>zz>KF~_T7~;xYUy_BwO0xtlVfcLtj_`|f7)(NI*}o9QjWdBK?TBRe>PL0
zUqX3-{8_7+zKA(5f2nEz*ius{OkDliwON!4$YuAWHS{CL4~b)mUm-Q2^}*=mmZL+O
zcg_z0*!E8#-2ciDi}{}~4dlnM2qClEkIA>^!HL*WEkI5GeAz_w4QCDZQS|%lTAX7T
zKDHyZPO0`S!0BPVL=_O6wC%Jvnfpjq5tE+ik!jukbE>>$Z&^#RH6;H+J7G%nh;};p
zZ1|5&ZQ*{$bD%C@b!Vx2q;}!ch(qL056p`^L)J`5Pz+CW2;{(+^0t8vVd3yygm<*j
zO>tZ(!nUgQ*KSyP8d<>#UjA!BWSpU6rYwm8clw&V{Ki|YqYbSiGJdZN&)fMlx^0Y|
zogKw!g?BwO?EsaVi=%ww*5-c)kFSbti<Nc*)khwaG5%Y*-e4KL^uz=@Ss=t8kZOmz
z&7+G>cw@#5s9G=gV><SM%|{Jj(IWrs>WB&}j{7;A%g5{Gvbow8aB4#83wAfaPR|{=
zk%kZ2zIU{FFJyw5k?*3L;xyk2I2fzLdKvL0m!p<Npe~e`A+(&{N>MqAREH&tTzb|#
z6QrQ+mzJfJCZsGYQg?UY6b=&c^LJYWe6Pj*1JMGa@|0m{U2kFQ+bJ?VUw`5G=)f^E
zynV!I`fi{iV<6P*@?xA?SdER$_d_&WW6v_&ec;^3lCsen`4^V1;ilztSKNI!I_=z(
zGI)Y=cC7FQXKq%yX+Qa?5qi5$Y*)+Q-hOr?P09U`|HqXXNSDghbAYDf3g^fK3Bc@g
z);kSvkitH`x;>`I;|)TY0N&;QdtxpCb^B>s1aizeKcZunF%f2WX-h*tAPi8a%%SEr
z2LluB%&r^Z_di~r)zuh!6LFh0>_nmn<X=|Aet-dKNI`S+fy3(ykfTI)MfYAW&LWGT
z%rK6%fP`+QVXvHTLpLfLA;u<FTMtYcAW^2mUBN5&;p*1Pc$;tA)w%W+ZhQN}JJc7d
zA<{up^F_7FP!&e8b6M_InJFoyqa^*QE?3W+^g25N6X8e!*}1Qa1T0Dm0B>3veaPdv
zvF!oP=jm4}8&KF;xUX+5@(aKcYQXS{mBszjJ8f<<#?JKl4^g}JuV<u+hchbW#KMq?
zIgc^nGjl&n2kg!_`XT-<{R3dUAs@Zh(PzC<*zD4+a@EnD!h+zT0RVrQ#tR{q+|c^I
zXJ<;(py1hj+mPaV>y8U{S90Oi+REi!&iuY*qg`iFHx-8Dq*Z$X2?AfX*26Dav@a<5
zug#5%2s3V`P_%eyb)-p!pqpJZ2EZ5FF?nd&7h)fE>cvZwk`SzcO7D&}c>a4o07>Xr
zeiw&Ud$%Sh&W(~2an~ort44J;cR60euW|sZ`N(E4M)p|z2IWT4;Zzh2t!r+k9782{
zkGrckG4E_x8=}+Flnn?~dWbreLFIMThH$RB2%sdP?M_cm%goLKD!TgCm<Hw>b01Ge
z0IOVMJ#HsCo}cDU)`n5J>moac1xLh}&<`zc3rA!b*N6WdKO5Nhg#5Fd0^rja0<f6X
z_|>l&!!pVj)yOgnsKmg*dcs})VrM3hAt514Clh#1z{YIM5Mu@jppJ;}Q|!Stjg2(<
z2ls{yf}DK!NPVw+=+{{lBCSj~&BVb<?B3-9P~+~4L1UYQ7V5kNtC%XlIq6oKd$}ko
zGL(a|;$eva*g_AAC84XFU2`IzF!_({SRc<YbUlr?`VbK6>2|EV{@>#S7Ry-y$@>!<
zZyDY!QN3AWuq-w}jb3whD*#L<cj-?8<dq3xY=yiP{nn=(MjPY3!gk$k{LUf3`0*!1
zax1owtSNiXFTJV1xy&Y7AaX|U>ZJsnw-9$ddHoTyllQ5_9imXKSPREs(c%`px9=&*
z<ui(yRqlJ4>dw{};h_}T<9j?2fXtCJtV{}^)drSHvVIf$h+Q9C^-Z}Y6meUGnep#F
z+)|v?{AJfI)GfuwlTR8CI1=BCkgR(Pn+1(4K7G1!&xH+pn#m}2gnwj0hyUg6F=j@p
z+<jk<;NWSS!c`c2Z*wWB99sY9i+5ZKIM3uKXAB=|LXhQo`Rk2rfe@N>EH>dphh1G5
z0c;$@Q<@rlD0#HidN_*TZN}4@O`w2WoCe^9EyP+?F1r`-lVX*ER*huSJjZdR$UI4o
zL7N90gUM%hhMr16zf)f8VY>P^7Wr!AqS%wKcDfqsU%}jF-3luu4}p-lw8F6-(+BNB
z3ZJ8;_M%8Vcd0aZm=3?;LHGBH&o~IXFi(+q>DFb_M|FVpc@wZKA2OeHxRe!)IL4YE
zxI^?<lbO7q3n}wNCq|23TB4^6H_JTWv|(*hbZV`2<9@9x%;fG5hiSbLJzh6Hc1V?x
zvZ`58(%kxV&(eg)0XJUla)1_(Kb+iJYN2jLCwI~1zy$&{+Ujx~x-&Hbg+W$w7K06#
z04?Q+t<S2F{~rkNq6e$fm2FS_>b)YjE~x0xp3hhd-=V@i65`{p!X5JrFO1c=rU5pi
z=k!sSE2SK{(Pty=PF#K;i-D))tfEim>T=j)kHs*sQ&Uq}02)QC?eFakGWc<#Lm@g5
zB6Q_UAxj5&7D=EXoZY54s;11!8`IBy>zq-`V(pmW?9Aeh>%>AL070AS5Fq;=<w>(*
zyl?oyZ?dy=ozgdQNNSm$tcY&F%${RX*{;xtdT4=TOVNgL^TUD=VetAHE8aZaE8Kt0
zr+|G-)Y(qc*=4rd%8z+*(UgwLk${e;-rnzT2D#hWwF#zuMGo9q6nntC3JVk(;`bT+
zA=m+XIgTCUBWU9#2cp|&J79s9B(qr+k@A1-78dxn(x$THbg@PkU1i-Gg0*i1;;2Ig
ztBMQ&7$)@nvc4lo_lzk2*#raDD`lZ&-Vp#7vDI(b)cjYZ9wcl~h9n5+5pXB(zwFcq
zFlTOuGv~G9rZ<nMs`5-@k4_)&&0w$apNY8LYL%C#qpYY3eFzvh8BQ`Div|2Qdw%Q$
zY`hpHK$7#`sOklDEH?Vx2x&&Ji$RNF_81`5eB?PceXo;nco@vy4~xb|^N>lTFAAt?
z!F-PB=2zORCo!6W(5{7`g-C9jx`h!d_k$M_jP}{0>p9Qpl2dc7)Z5lUIBG#)<j{$6
zQk<HB4gTVVOW(}~jUQUb1KbP@&_vnpFPpdSbq)kA$mdBkJU{TAWI$enVhRSN*&vT^
z^GL%s*T-CLF^GP|+RMl~=ple<LXA87C>K+J6<4UWHB76?;$~Ld|7I{|XMZ3mgr1F(
z8x$;;d(qyZerbs?_5kq6Vv8TUC&p~M_5rIAkMr<GA*_^fw`#+0#FIGW*+O|D4D9U^
zH7@H~pguI95c_@Z!Ja=Gr?<+zRHL!l{84XqhHR350{;uMzT(i=NYZQlQ;f5VFIjlp
z?3mjj5q!cme940j4i2{04|dIkndkWT9uEcK1lJFh3*&sg8Nk9Rdme$)^^vY+{zK`b
zoTpa+0UB38EKms8bbj{_&Zx1nz)Si9_~Bpg8^j*CqKhe>Y%c#?rq-2m{7fA(A4N}l
z1ILh0BLi%uX*UiSkMtMD*0TzdmA>mAsp$<-dgTf%`f;s0ajkU=H4|>PE5emoJCpzz
z{_I2Xa#m$2{2{vAP&LBAfN@r@ammM40w%Ka&5BoIB*@2y+6YW6ZRL=qWpjyRPGStC
z;Tl0%`@@sZF=`xR($hMz^m{Jm59@wI40#!-kO+j;Rgf;(A9<e$0c){^Bs<_c^>6P6
zC)57E5tuluk#^lQy78(6vt+;A3d$iL3${;qYp8Mabh8X3+Eo2}0QQz<)_|Rm51Y#Y
zj7wbDFv7Y+xi)ZnH31F*HuAB|=lJqDUam12t*;a5a#FWqxX+a1zVTvMw13M`7dGl2
z%${&P3y35_JWpbEXHFOfHvMr--l7i`pS074XnOqaCb~D0YlsLHsGdd7_AHRNGQ>+i
zoyiTXi>MZ&EH6tuZ%rgs707B^10A_?l=zFT_}Az-IsJ#?+Ws%9)M}j1?KSp1l=^Qe
z4ScGSPM1qOp>#eZG<<A0uu91UhaTaNa#0Qn1VFh+f}oEA3aJ6W9n|}Q*HU_<>ewR~
znOos`0vZ8BETe)W6N_5``~m&co$65=S~no{pfw>&hv13ELSg`p7q;&0{q2jB?GQX%
z<>aMHmo(kSXNj$cTNUn2)NvOjUJ;l1Kl43GygD!PowrrQe16qDp^m0nyYq)#oei3T
zwKMw~^*)VmW!!kL{G?_OQ1%fFo=Z>yAMN0@r1xndM--!0YITKc03{GlBacm}jEgYX
z5B0kZQ<*QX?QLHgzm&%mclr*V#`QDm@CCq!7V2*oSb<^`b8U6l2<V4AfeE|5$S(kO
zEMf-W<S=SotQ_q$N=e7KAzo=1N(F2*o>LC)Sq25p@?&GqMqgq6-!_D~{zN~&QG{pW
zTgS!f9kgLUl)gQp8g+Z{$1=Z9a!$UJvo||dfpz@^ot!a1C#yRFkr6nDLV&{<X7jh|
zfu1<pd9sT?F#4@@h1%``CFCe^i!>TiC1PJ7!<3!|nR~IhdE3x+Kq>1B5BScIeWW~(
zC=`R4SX^ABq*8Y}(Jp5e5j*_%jV@9luebKSN3DJP9t7iec6Jum=WqkXfFX1U^%?uy
z7!Q_AdS<PpS7~^oGV<{&d+1{yCoFshdoy0ytE$H_HqB5?pGNx12-K@>Z1;tBJlX`*
zRaAD%j@GpfapwPeQVZ)14N;8apKBZk^4<ZI3F&JMzt4;us$OLde=BKz&RWZzn=vv2
z0|>5k+QkdNxRfu@1;P+x);MP8|7k;*MOd<B`9(w!%2%04zk|NYu|lh=riUMLw54xg
zYEc$1k43sTl!}RcPptxZFzB;_@T>5`^DvC3H;V9N<N7$~tgJpOSIn>uf4R~AvFcnC
zA&PHRmd?PTA)1>$mtVi(p$ubgC8*n(ds%o-_k|vVdq$^{vN*8Qw58@AI5E)@@3Z>6
zWV{9`hp^a^K*poLJaDl`<qcqon~JMVg>>m7-+cz!NAckf2O9GOM(#-(_CpEE`_Bqo
z@fU8M^{I17yQ4NwC!2^U*U4y$3$4)ry4fiQTq2SKF*x5p^T(&`G;8e0Su-7rsrJRV
zEpt0hWDS=;0Wv~@3b`lAfyH4Ws{A5lvvB{rFo>$r-(6ZaBVe(d+PTa|U#3!89fgQL
z9SrTNjO~C!TzUZ`I=lKno?JTn0*%YkPdw_OOTJi>PxcHQF>o)hF&$RSsP&!iE+Fpp
zd}fvoNCJ7<I?5^KA9yJp8U)u!dobUo|6=3V7n=a!%$`}9dRCzuP3mS!P)~2UQ<V*p
zT+FS6nnfr%4%Edo^qq=bUxXwej5{5WZlFvtssumDGc!GX|K72p*yw|M81VMG9t*eB
zg_GkMH5xB6nnQO=lE2@bInlzo3|2%^%r!UO55zb4xa7U>v0h62fnmxHll`tr;T3S*
zdumHMCHdIM!T_dklol+2k^ROzHj$eNcWVF6I4j-7$}J)_4~Av027&&Xf$9QH5}dWj
z(?S3{=5S4DR7bc=jw<~iK&SM2#A)s;Y=kxzgE&Xw3w3+gd_3djN<pBP&*7=eAJY@>
z^4Gsp;>nYekydx%n{!7M$iMBh(@1p`^K@<N0hzM2x_Vf|p|<+UOc$7YDJC2^!8K-!
zWlnY2e6PF)!DGVG1q(T&pM~t>LiPzqYeI*5Fpgw%jB5du864M?4PEpeY_!=`am~6_
z7wob_4bZ%L6_sT^cC#tfNnKO((DG-2mYDN}CS{$BP6t%KtrGIR0ZT@XPh=!hc2l21
zGRgOo6M1G1mvp26<jMPY1A7*A!_gc3PHBqd&(Sr7<cG8e+4i!6HZ<FpV@}Eup@us2
zWkN|U0?6j$1wOsFu-Je=ZxG-UdzkpcBCN;jFFpygC5SU>?y1g{QJESmrZT(k7UL5j
zx_k$nOzKdTsO%Gnx{{KT3uhWro%v+S?BR7xm71v_r+GS5tSiB3sdF-O#>JL4UNGU_
zdSDA^4+LjB=N5d;&CXl{63%uc{DnNT?Z^#}1b-h)iS0;ckZ-?@^w~HC!6}=hp2uQp
zLGeDtfnsLQbQ1}T11iwlYAez{0|=ceFY9ISn~qSP7kiy4ceS?bk6T%Mr%FN`xN!n&
zY?bb5y!m6Uj`4KO-KY+`t6i0%(tFS{k<u`<CZlrX$vsyh*bv{M-JyVNqU?p(a<jD`
zme9eKRmTsk^KrhTgIE44eSg1x647DkcLr}wo$5C)Ov%Jn#?lc`-{}E{{{?NR_>pya
z@WtnBG=T1(AkQ}3_^8nFt467)07Db};109cYS0x?nt8~UIlB~~qE_4I?XKmu_?`C5
z(Xl102qtoue)`@|n`!s)A#TKZ`!~{@+O)Q(Rczi2E@|z8#K9JLOLKE`k$0p>!Ka=x
zqjSo?8?Fvxe7qmuuI+S~MZil3o97Wjpx`7MbLr(Je{a?j`ArRq|3Homa_P2KxX;y(
zpI%M~#PJ<&0%#Svz`9u|7hRH2rVv4{ls>23Xc*|}S~FzkZKcoIh2gX1o^!etKEmO1
zPj4;{xj6wp?NFISorklEz11O|CLKKDiTc6EE2+P2p&wCcpoBd9D>+9%=|EVy+_PJ%
z{uSO)Tp2qk<#-1!Mqvmap$*WA<7vC;L#`e@CnBnC@f|Q~Wpg9~0Wp5p+F7}<Ux69R
z>j~1K4}*8RTAm?J+Q`sEMEYXGV3!4?^&a_;0yXG<6dPYKc9KY5_=6HDZ_l{=5%xYa
z^AH2P;ztKl6hE9=ra6I?QV%MIiPW)5b`<-SBzkF7I0gP3FNKVt-%oH_)2fJYRt~T$
zx0Sz?IoG?XT&C6){~~4nbFpO>rC~z3KdPmUy4?4>igq(vgBN<=+Y&`lT}<|X7}}Wo
z(V)=5paP)A=!3vs)?AMXquMUsL&USqqR%Aqup6b2R7MwEk3K0xAkH61Dqigd(t>nD
zr1C`*p7kS0M75dvYoXo^$+s*oZ^TD1r<^f46L}X;aW0MsTi3&#iS8sA82q;>0X}s<
z0G>LnU}a<8q8lrh4w096QuI&OMJ3qXkFYpV$s_l(NF#>#9|$KJD1eO~y?D8n4b<|1
zM8TTM0ytWQgPx90UpY{%ww6ff_Oj1x2IppuAiUSxF3MR(DtOnaD1ABt7A>|aM9Qs^
zTf?KNxIJcr*)F=U@uNfh5p8M|>e)vLqJGI5rmJZ&T%HX(Yj%74w2>}$os_;v_2}Mm
znR-XSN7gW`n_H7&rR|svGABsYeajdoPYYHJl&Iis>+ZPIIKNDZbCNLbs=9&p<N50|
z+rvL8Aq&SdHO?Qh*<|B8cY+QnM~4*m2%aDyG!D6kI=Z|GAKD3>Qe)8gDyl2@)l?m;
zpKQcIa5X;SMCn!pB0hL@vprM#JHlSpK95Z4vKQ39fxQxMpy|hYbq8_aw4~DTnyyAj
zhC3LMk$7>{9d?EJuOQ`WkD33y_Kk6t#SQW#Y&qg6tN{z^u&;{&)^l8$PJIA%7)Pu@
zy*|}7Ba^BRWg}P<)FZP)H6oOwg$y5UJQDU%u({o8&Zu4XE0(uh1V#?~(6QR&yE0bE
z4)q>3ZGF-~WKY#{&wuskBx(gHwTQy8(&r3HFz6OafaY)3b5kfWpHNW4ct2dAe@Sq$
za=NQtMx!}H(|C6-N=TBmzn%(JEtDJXZ|4s_%#;rLzSljTXzNFaOgN5jO^0zZF5)9N
zx*cd5@q5QEnxn02EUj>6;&s7s@<H6@%{{@C^xwizEN{x7&~rY<@qVvlw(%?aGfPX(
z69sVoWlyr}cNv!k`j?P+q@{qrBtu@-$T)^qpsYG_s}9QJc=s=8N#w0AOj*S141Rby
z^7dFmwcI+phB7pX#~wX#TiN*004g#PeIq{lff(Q-_cWRlBAp&+QFYEyCGNBbBnBbC
zpvdf^Y$kWtI#r^0u~JndG*&ixWw@1gjBErR6JJa=8dqgSXAMQX>D}DKN|Ea&wg=1H
z!y=`k;|968w+9=Z(tSt>wz;$74c>Z?_B~+K_(IL-l1hFMfo}Xle@YxKz{)qHu+`kX
z#8r8PdY!mJ{+>wyc?XhWM+3ex$y^n0+5B;vB1ap@?E`({bE;Tq4mod+UZS=9iMbbq
zDc~sLVI`JH8@8)(-PuA8AD#|FtBy_0ro8Tfo>uuqwwZ7+H#Z;L%&!SC2m#6Qj5by=
z@o|^k1jA3wUD$g%lbeUM{+%ZZw}3E792P_3HA6TG|H6<V54bp^ND*ga$%x(GTHhsQ
zF0rC?daqNBgV4BnY5UVE$=Gd$TvRjbNw8ZTTDjEzaXmb@zXv@-chWEt=FJHV)`*#b
znPxAr_`{Vi1JiYY7xp$!<@_p@O5J9@8G#nwe?Ji9I5f;*SXG9kVBBsQaMRroA5px@
zB=fmet}zx=7huddy+(o8Ug?}$c175T4Pt9Mw*U>C(uc{G6Ye}!Fm?QW{c4-}$h^uP
zgl;bKZtxCPXd325v#c5Z{W1+15U^*3OA<OFQG$qSwC^_F$uT3XW-trr#83-Uwt&nz
zWVxF!WOc$zC&I;unqH&fic&Ve#GF@AoiaEoD)HEXnlM9(jDSgkTz+5guuC89)EXZs
z(Nn(TM1UCl9cv0Yu)1JB<I~coJC->ga#v1+wT7&jU?)=)M0tyVa-Lp)7T#M`uaWL{
z2#|0KO$vEKc$sD&`*(Ql1bEv;G1?hecibvcMBHQ^d7QN@JTfucFc5OMA&Eaa+{RZv
zA1>bJA&|ivG<VK3R~8O}f&q%3q`QBQRm+*gQPazfu}-kde#+91@!RM$9#`Ew@1Eck
zbLsLN<frZ^(u_SWm%-I#p=KoVlhd7whAHe!126}R*G$a2Af9q`h;~s~i8{3$ZLZXm
z$!=7L5{pjI_WyaD^D2&2r6n#rKs&W>J`4fs{V(hsz)P}d#7=0cX++D{sb6|*St>y9
zT^Ek-5r`1|dK(OTsTiXe+pKA@nRNmwK_xt<L??{2C|v-{?40d<%M3?03D9Gb2PNoI
z^!{!DZg4({7>fI$ua(~R)D0#HRz}HI(Mw*?<^-!nBP=8B^=~;)pxFGy2V(cz*q%qS
zDo>zM=%~Kv=NDmYxdI*6#a@j~V7Hc}Q}j8?L|9-3N{qVkK$aZxj%|+REix^ks|@SO
z+Ud9qQo;vq)GE4;Z%!Ci(cWwA<!Lo^srMTJYr5M8Lua@ia6!h`cm-9x77s|H6-W8U
z_^g3}@k<AQTXLR~(X}G6Tr|Yv_|hlzZbm)CE(1bpoOm}Le>RdVM_o`fXa?h)!t1hX
z944948sro{n_GW4F}g&-Y#A$4u>&&uN;A6=nGcmE#y}Aew!xb9TGZU2=_~!l&Zv#!
zKHlcq!2D<eynd_qy?iU!yyR7Jo6ECtVpc*z!T}Y3bb!YnuQ%I%LCkg8Qpn_kT!4uZ
zo*}*htQ<1Ba*sNAZlTHkO(}l199(6TifN+>uQAI+Mld8|(56}=&w6QktGY(WfQ@kf
zQ5CxmAf63pl#!UxW$m*0DJaYY{E-#CxX@dom2Q7M=QlX-^OTFZ{lJ?Ko=Cy=%kjIA
z39m}VY>u`nLZ(13BskRuJ6KrG@;ZPx==&9cQDR9QjDek`4UThbixFutq(-@P>fQ-W
zjk=rh5duTM_v}QYof65O9j?L)%7bGd<uPdUQ>0Xo#`dRs$mg46_Yhib{EbaDpkARU
zjc%n9*A52<2l2-RlS?du`;@(GhgbBwe~}ZYz$p5t#-BY4M=A;XFv8IlNTB97&H--8
zTt9_#KWv|%h$Y6sE|-Hk0<@T&`ga?(S4`K<WF(hc9vR8%+xzlKV#=WPKR~!TuszV(
z8fUrqHx75Q?M&^d!_}%_2_K?PL@UO(-sV<~FwkR;+1pN~T3ii5awI2?QdS!&nylN^
z{uwobTEY9=eJ+PjYkK1H$Y)r%;=5^lkgz_IU!HSTYcQ}yxX%qhjG<U(1~;9USPaW^
zI{=JhZ=66_ar?UsMoRisY#~S3Tu&;54+^5$WnRd-4%DbwtLKqf<6c~s4gkQyBBGlB
zrh&((A6H3(gIuh9(#R#=h<nOhx}AjYv%&*X;K%Tgzlw#AW=;mTPTBYeIo033!)RY-
zbTud1`7)T~&nM58Y|&}Y%niAAeR(MQ?0KMk4Fy$>O;v2tu6sNdD=@vl#4BHx_zm9c
zrCiKi9X+fA6_hFa2Zyn*sE$E28{R{e*GeyQ+xWre?i~!>n;M#$N?2S<gJ`+B)joI(
zNtjWX*Z`<?N-bNz27+rKiCSj5)!`akV;GKa5Gt19tZZ*Sp{)$&HP1#ypDovwBi0;b
z3lz_pXVIU#7%}G?TEF6)VdiVePWErsa3c!VUXW;1TqaR|rZB__o(n8no2))~F7;&S
zV%O6utLSJqp~O^afks8wBgd*|+YRq*hya{K^F5X!Pl3d^+2GPX8pu_w_ZTt<pRfFW
zid}Gjbz?v)q<iz;vwK;Z<UDEgnKQ`h*2inR?8BEwqn&8uE@gH|eq@qvHTnRa^)Z=9
zsA2gg)>|!gk}Rf2uGrjdpFai@!XoK`(pX5^sMC(z<6RLUP2*jc{`-3qQFZzzJ<D^D
zp0Ie*yBZ}>MEGmkos4I^DhV499Mt-Dg7g*uT-Wk-aMwix+^T)F?5UFs)zOHRh`RW6
z-_T+Q@Mub$ghZW%jue02GDFRS7m~7~2~%o=XJ!)*2W`gsBtjP*-Jf-2*<0RquoXyK
z2vEM>I-i+#ltBr4_Vai*o6o$|;$+UwB~N@ppGC!3y-b#t@9Ly)62Hoy{RE}yoogL6
zoe>|=GlH=<Z6J*-Iu?1kTgVO?sQnS6Lc$QEVZ2P#@}r53<(0^_64$Tyue%mjEuIMD
zJ11gSA)4lbV(GA-LPga5$XY!o^Uty)R;M(#@)x(M`pAT?FnIKenR$`4$Nh9>w+c^b
zKbgrWEJiH7c#P!2SDwe_TjL*16mbhCS>YQN_DB`2SSg#j%JY%-RR+;*juWRi#;R~B
z$JHbDIBOeL-(7YIA7$7MmA9BAj%h>{D){$k9l?}pLG2PCRNhKwedJVNK5{UO8Ihf$
zFgsao9cWd>2lUiZ#%M6?=d<y{dtt$qPWlAKitv#$V3~e7Vo>$TMh50V2X3(p4sN>2
z8}Zlwzehb;4+jsp(yRmt#&Z0W;6-&lN%+U7yy^#+Ax{m?xDJg0mZBO)X?ndZjVdB&
zLM(dYAwu`ol5bHOr_=B2aJQeY!|U#JC%=+~dI*th23))<R8Na=l2lGqPN*9rf`c{B
zmfNmCQPh9Y{SwTfh|Wu5qSIs3)8J<a_dMlzEM}skqoXH^2w<5ZR9XF61L)g)TK#D0
zg^vUUy(*>2c35qG%YsN@X@XT0gw;1+wG*Lo^<$(%j!0Ya7w!n_cei#t-qW9TJ!9-s
zvCr?uq9c7l)M~U*&9|L0nCg|T7YJ5KIl@y{1ub8fjL+_Ty<KZ_BI*SD+)a)@=TBTZ
zYp3bQpx5@4$(4ik#KM}Nbkr03o0u6r4kg=i&u!Jm^^&)Lym__Qa&Tnu$lT0v#C+kC
z%jUP{fyq?~zmS3*%`~gKlB@$P{2vnsmPAjPf?W>eG6kE{aa#TD4M)dGKJU1t(o*Z6
zD)rFHq~YClElTAqxW)E&cQS<pal0X!2b({W4%bjx`<AYFm#h8AkUwv-O7Ru+Go^d?
zEc>K)Y1=#Mi8MXN^$p!K0`ArwGoCpcxiwTl>|yt=(o*iXDU!KDqs;ZSXP*#WXZx3R
z)8^bi9ZqWsI#3Bbg@wh0bo$^&-Dx2elH$|F$W4cXGHRq!MEi;V`}5MBt9^|Rv$5{D
zg;q)!Abw0|IdAOu>ty94wFLtmtLpdDxvFl}JKo>l4ScF;wu+K=8*~Bxk`T^pM+r94
zjlt5w6i(^5X=QLeQT2Ox<55pZh7bG91x%4?!s2}T#Lm5ty`KY#l*3gH{?_e&0GJGk
zct8=t&uh3o>|9^pX<nD|So8^^Bqh#mCx_LB9FS~<kUIoah#74PzrUcA)f;Y(&u^J1
z7VmvGsHv$L+&twbFfql21`Rby4cMARszJ;krT42(IhQMrI+U5Gnr&3{w~gqyx%Mpg
z)b~-&a$iaJlQthnkrJ%fRo&V*?JOk@2bL=i4_-rhqOHq{xPk{<1$%zkj8phT_IPXE
z-e%?^_X6D9v)q8mCnb1uBkO4YU@rZzR$@wFA#sA%;H1-or*^U+_XSEcM|RX&h&UY2
zK0Zn7a0i$X&T<Pi9C*ezDrBuT*WeD6ENK-Ib^=02nG>DD8+i$;S(&x_q6{`64=1G?
z`X{Dk9KY9X`41q{=W4B+(6rh2pm1FD8r7m!I&lp@rs)fB&AyyipEUmox%uzgIx%Bq
zD7zokEh}%Hq>9Q68RNnZKi|8~T{&?_atKnZWm>WnWHNAb1#}<f503uQ(9e|$9`%s6
zUw2@11AFm<pMAenPS*=p)>Df7u&3!d)wCT{5pegEF?eBe*}O-taEaB#+Ia;zxmQCK
z0Cxl*(eSO?tAyibFe&kbf5`5i`9|D9Be*&#dr2efa%&W(dM+^J0WBV2=utkf?3|rC
zE|v~<WI}d(F1G^mK8|Z?IK2(DD!fO+DTKSH9#C(?A%0m&Qg-)n#Nb}VLH|tkBzOhL
z81yrsEzNWhjMx%*Wc|smPsiS^mNn9vYjTvNk}2P4V_u@vY8H~fT67W9sxaayXm6gJ
zXm&b_TjTfsFH4=$4HEIm{mIAKS)27W$iu31&z0qk{q>}fAWEUu!BXRKe?j!40lz4{
z<<NZiy^~eN@$Q`Q+(1$iv9Hp%X|)hHM*Gdv+9)w#WV*TFsj+I!$nSvId^ayhL6)!W
zXM#gW>#K8TJrxIV;FriCm#JuP&p<P;QG?4#V5%gm*@=A%?ut^C@QcOWL>^7ef$)?A
zYtPc>E6*)kBU7=DL)5m-fPo8hdF;#<a&yeJ!1LCBF7!mpZ3e7+X5pXD&u^-lL(TL0
z?(4%UiDFb}N3o#K_nen^+tGTENKr6lva6P^wXBw@+^4w&Bt!J*Q|xG<NEKMm6RDk#
zZF{~*eD|C-d|EIe>aLKo7+hihF!blMEF|?6%&J1dXFfL!=9<*z*Q3}s=X^N#(DX?Z
z%)DZ1)cVt!b0#+KdGAjbi=!2j_W=Q9LA3w0X|sXAM6&0p#s^~?lm^TL;v>%;!+F1&
zMh(+-Ls@k^&h%Xavj}FgPiJCSb=~t19aF{jF0in%wLg9SoSM_f&TzvEH8OX6eYC0b
z-hCW_6Jll7x&6(TH1{JX#0)|U8hJJV+D<O}N#ssMAY_87OaBq#+FKYC92{4{TjjVv
z8p&F?+p?XBL`JogFK_R>;CAzyT-^$!QbIZ?AuWk|GXYJ+PL@Nyym|F<!A8@I>Z2;G
z{nWX#4@h@c>ZnTq|DNvYqdrerXbAu3JzpszbUkn3b*njy?ce9#R!fG>+|AR5U19-o
zUkmEC%dLHr2%5DdFDxrX3W6f8KPwQFgdot=_ESRYs#g)&v(JrBy}%vhNyu?wl$A}|
z%*fG5E&6T;tv>b!qx|JesVCLxIj_P+8Ji!Elwm$r+f*czUmI*JcI`I=GL!}~S%Q{1
zg9K`E9HE7WCKd%&rVz*bK_4b+&5_jIVC%1+0;h6^tFu~1pD8o;1t<8i3rUiRlqPeR
zMgKbz|D@4&3+9$2$jUX1L|ZP#991n)gg4mxu&PCuP#;nqn(3fF{I-0iOzw+BfOm4J
z9xOp=DzKIHWyOdkdt}*2AEH1m=xW8TwSrsqR0R=ap8yMSoRT~-<U5$^O><fIJtVM)
z>=ey?#~LO%B-kbmr2h2T@gu5_gS5TlWW5sjvhde%*|?&^!IlsA)MT{3t0ssfJ;Uz@
zG;EJ;b~4%6o?HtmMXC!{-=z%d!3JsDaf17STbg@!D&0yb$USS?j#IFm?GOR~;)qPW
zKi0FCW|FhkG-@Dz$TwH>y+C%1_>lX$GN+(I;z8|ghmT#@{ew|Ci?UpAKu}>_Q8FxP
z-e`*Cz8q{EydAfT=dKXrWcv3t?iWH>TkEMlN1NGUrx|%4o4rI#Iw}`DHObC__vyHq
zw$29?1)BB7&~dh=U||D2S3owL-O3-oX8S$Th7P}L4Da3_0|i$Z*A}0dl=Pbr^Rj{!
z5pmx61tSKg>rmqNuc5^w=4}dC(^KWU{3dX-92)4gz_QB69*MiBFqx-*!OHcN%6=lv
zY$d$n=9)2h5N+R|NE{+H{`ryhkWgDJC6GyKRMX}bFhV^KhT<kgb}~0kN#5v@mC}32
z=nOVPW;r~4RP;_{&9B%jYg^C-@LT+V|0R;PD)1-3a?VVtA_>v3>S=6Fd03(3&UJ<Y
zF5)^d!pJWFeWaP*{#29#`+5U9r>he4$2;B2b{N|z1HW&z8Z4(5wbf98SgOxLYWkvz
zH4CfVGDA?`&VMUz(9r8FdG1xQVe#`2?5wa5loNrgm4PnK{o*!D#eJ-$O%E?0-t|A?
zjdQtL=ep^)XB}8uZH>EMVUF|LmJ?Jkc9|yDx2_`lr*hinY7bX;EPB_?9Q-qB4Q)@8
z{7cLs{8Aq`D6{EVQ!(z|_G1BSX=&9fX-V=h*?(WyjJLMr(z7;#apW1AQ@cTly_Hl!
z^HiBhp26~nq~esv<DT-a70+*ujZ01n8Xk%FT5W66eVkTk=|_qZR`(nf76LM8rrre5
zE$n{F_Tz&E^+NNq5SQ}VdYcKEaXL!#+v+E2M9s?{S?rcQ;L-z4w=Y%$mQub48hg;b
zg9GdHrRS#47cbQ{&evLzm-nA^%V;LGFbmV+Z<j-Syc>Rht^*|atfuw3rH7!Uq}_#=
z#sgcy^>?e9HG2dpt@R4tWO=LmNrl;P(uQ;6wL|n2U#(yX4iWTSWz7JbiSipt{n^}a
z-k)E#u_(t4TN<(o51fQ1<j71^bPZ3}(!R{|^zN1kP@e}Tmh3nG-f#V7*eFH$#jBa*
zP;-sa8dBCn{Vun?agzz-Jxz7;U!C#^!KZJJlmB=g&hVptNvNtxEmDez;Fdkw2g=g-
z3~io&AM2WGj{VO0=-BJ0M0yNd9-<r@Cvqficyd+NcBYrRAZw2L&b5JK8;iR(|1|k-
zW$J!)I@P#P%N1z|j?idzsebLPuzo*(!9ciST#BSqot4=G{xTW1XJa9`-VnG-@_#th
z^@}T01y>=d;j+8A+hil<y`f=Pjq-WxTs>i3kuzV)Bn=XI=2w{A$5m4NIllO|0L4gP
zg89e7Zts1IV$;pNJD><tQ_JBR`8(bQZqv|TX*qDFhOfh|yuJY|ARigrLGg?20I9~=
zu^f2+y2N?rI66(+nOEGZLt|4=aw+Bg#xs|CiRc4W(gvx{RJ_z-kLT@}SjiNU|4W09
z{)A_Sd6I#7qtX1@-QhrDwjvJEsNLRrvNdaTAQoYsYi{~y_1$Q2hRs1*mbH$1R5MvO
zo`W>%J~;Q2oK;<l_p4c?`b1)YvvAY!fhv?M5(jnnuDg-ei@~EzM+qLLfrkghai>No
zTB-tm@%{4+PB7QF;+d_C>J{14`~U9_8h|@^U;Mn93;HAyJ#ojVtwti+s*}j7`#HC6
z;Ar`lz4`3?v+n1;#U1Hf$YXv>4S;=SrG*PHx*MISGPUpzND!Nu{4Dx&+muo5`Sf6Y
zMX>`7(Y<@iy4vz>VoB~7bL{}jq0j#6k!@B?&E*TG7G{K%ASv^(kqyW00^u==zye9a
zjt0Qcg5`AD&E>T!RX=G%+m<>Xwg+T2!hH~t=p<h=ivXd>=jA5`Vs9Y~&_cGROpUKc
zVQz^P{O-uspE{r9CTfY$@ul1_&jg30iT!v%Ta`o&w<Bt8tE)HbLjB-a-}aC7AILsS
zHT><lFNw6(=k{4~JZYNGJk`4_(oL%^P0|TRrZdZO?}q2<Lqb|`Rk#wrOj~nFTJ?TI
z`STGTA2fd!!a7~#P}974%11CxYJuC<!CVBLn780tzVU^+j%U;wiM+T+BY=Esqec`n
zWX$X0Ceq@p3<9N>%}h1N|MWrI53LqQS4{y^mQRrXJOxc7%u#qcH-#x{M%=IdTKpqZ
z;<9}V+TSxZwW&X}&$Pt(lt!_%4{2$AsLOZYDdX2cGoMH!uHFpY`arTXm$84=$F|$s
z=f5Wue)pMVS>f+>v{*1}I)C`o^b5YQc2a=rzE#cEutTd&6#O92|JU%l@+wZ#TC>#p
z(RHtR>Zb>F8CvG33NE(_Nz1+o`Wb%n1I;b8YYTvdZ38=M!CU-O2MyIZZj{C@w_q(s
zlz*sR;wWeP?@v>k&F7Y;1<MMI*v+SSr-(K^2eXm<p7j;Dblbd4QYV+@HgQ>$ejsTQ
zuO*Y+U76yV&9%9Ax5-t%h{VDfguuyxxmei;eyzlDE!Q?1&1mLc+-ykg*w9JZIoOWl
z30nQU<zExo@3WA%?$?i)mmtI;>O&4+$^O5vbrj|aruBMe>R3nkW(vvcC6~lDys?=l
zEzhR+JS>amnk|y}BC)SHK-@FfXy^*;C7<^*Ec#vVS$)9S)ydyk=rDu68s1QHVx-CT
zIU$pH%_het;;E93ovrVdjivRkr`7qJbFW=<1?VoH!l)Q|RiI1fxr&{G{O$}f6no-J
z%U60-%BwUbg9672(2BL4nt^x61I|hlX|_lvcVEAZ4V)uw#ztpS_m7I1HzK^tJ@TKt
zNDQg6>xq^U&QtJnn4*7%Bw&g1%`~!{A|K>4-0*vh>M|La9$JBQX?<q9>hKwbw<)B3
zkO|txhil59!)I6{C<&GWeEj5o_XQRo;M=Y7het4hSkez71n1nTu#kyp`*&V_sdTP=
z%}RXYj_>)V>NZ!>$(&k)`86BU&J7*+;S$Aiu`VdkAq(8GG3CUmcgkP)VOcJ{##x(}
zSUkJS&)E4GenTO1{HR<&g4uHmQEt_4wOY8=fAYQLr7y8c6VTDXsj7f-;tS*vbzDsA
zp|65FylUUZ^LK-F75&b%bb$U-v=!N=I=+85Nb;jbb$AP}5;#&|S+Ru~RuM~0J7|)!
zxlBJ7<ko_pl6!If3~AbQZA+kbL;|_7bH_Ur+r6$ji-}@ZuLbgzhEmCd6wW*`ju|=1
zLzWuaYT;U41*ZhsSvkBABo<4AEDjT2J6y+QPG~nuOlAFg+g3ec!?o`zy%?9h?{y8@
znTeCd6&-2rsHt)$YWr>1d!X@QGp5Nlu50GB#ZsSbO{6q|N6n=a{O^-HF>S|A@16lt
z_Q4Zt;%b^C0Z+82mps5`O{=n4VU*1I@+;kl=!YT`tl?TaGxQqIqO`df%7{+8e$#oY
zqy^9~>){9u&JhpBdx6z4*7&<RpXo|1w66U25PwHbtu_`J<EYZut8@c(e2)Br*9(QR
zk8TDGdid^W>}v)m(rc~L5|stG7MDELNB)%8TOhA`QDn!+h&(n*H-)roF5tDmkEuQc
z#9r@8E(n52Edv}uT!$k(9~{c2)WImB?SGvu?suQ3n7Sq-@^K+rhTZ|mEl@H2Kp9t-
z^tP)4hn9@_@?)w;#nw{*mp-w3ovL1-?gqWP0yqMwP&buz32BJe1Lm(tqcQjWebv0B
zTxjc70lfpzInQ9Tq=|rgAtS$-##^t1)Xphuj<l5Uj#VXvR8-r*faq!u6u@#pB)!bV
zOIz7h4;uuC^OgPWlSE^2bz7OPRo+6?0R?GlQ)+vUidj{;v~1n^lK-4~(^nu1dVn>q
z#W#Xb0;SGr;Yy~%rW3=aCM#PFzH190`zgsuljw)T*f&@q#*_wwx7paa;aekwv{h5r
zBKW6w_wXgUMdxqSH;fIZfhvk;sh8HaJRbe-7rv8}x9rL24f@e^6*Rs%Mj^<SF@^^A
zGB%l~)=ZD)e31+@JzW-f3XgbEy;@l;O{|OtxtwZmq)3iIl_-T9gxEPmV9rZyjYe0G
zQ;+1061dM%V)%8GuVAAUvSsT$C-u{3jj&b?34_sK5VuS3YD;&Hd0Skuxi{3CB4J)O
z3YemQ!YQmSc=E?CXMrAFMl}e^EA&&F#C1sOBhOh~%W}7`cO@PyPzt&h1Vbw+qpAI_
zTj}0V4MkF8I>sxc)(3{kJn5hCb@%alU`8vk?mi5txTXU8pVcalI@tDRdlEh8=Q$>S
zX-_~;rgP29nHh&ru0aI<bF%94Rj<49odyHtTS?6-){QT;tfdOJ(<_vTe%_7F2!-tN
zoq!ApW(}N6&03%LbB(f@E4okm$PNB252{T^kTo9@WxYIS-nzc5F7I<_YRM8ZDry*+
zZ#nR{xtz(ajMuWF1RR=1x*kezY4E>F+H<3s_ROWv`Rh+OWv~c^m_wHjxIQ!tdt|Zf
z*stQ{Y$%G26T?AR7l*WmW`vxHT-s!y3Q>5ZGTxzK++%x=GO}pWtl^wKLh4VF%am|K
z;Vj8tY~u%;Zp=8w|2p!u{&isa2f2R;02^QW0+S3h4qyA1BKbeN?(f~*CRXE*c)uKG
z%=>M5?)!E*_yp%reV6PHGN1kiN1po6fedhRVe2?eJ*}!1yq_+9C{&PKTlU}3#2VC+
z``&f@Dz<=j7B6d8czS2kUx>U}9&FnE``avNdUFyt9>VnCF+OZ|I9OFev3a#%*@9H;
zx&_LC%2^I~%R70BPdcYroqopWaw=dDw8uH-eB`91zAvz4$q7y>Ebir!HRGG^ld4CS
zIMlZkhW8y>SCN7FWs>*dg3E$BmSm?N`Sj6;)_&OdQ?4Kn@zQx_l#xvb#Pk!%+gk)%
z=~kmV8XI@tZ?$lpm;9Y5B{^liD<yatxA(9Fu0_>sNP7O~q4u)ed56+k3)`}c187p{
z{?+l-ZI2sd{5`Y$Mh|>zrC``-CBl1)H?16!IOn~+O@WMs7;Re83@(X1)^9A)IZAs(
z+Q?J!%T&^Z{pY-1%12(Yio9ww-E_$8Ea)MmHPTE~)6Jwf*N;?KX^v>N2qd+R3I%3R
zlHkkpmHWCqMl+kbKzDgi8SGXvziaV-+PU(tq|-fI&25>MvF?nirDiqR9lL6_xCFwq
zSm#c+@pK$iM4@6#LxFGsMHX9}rm>)zN$rf=otknKfm9X+0h^MqWk8mKf{2QNNkT-3
zir96|xpx1Dd+tx~kDujzp6By<S6v?H&^^*bTNIN&hzDj1Elg(dq$~#-t0qgzD<m?*
z&YJgcm4b>HxF>+GH0+SRD5bT}5*&7g8*DldCE`1T>hw~R{VZ`)a;-$FNtx<@-Mw9I
z&IFV|FCbLL1-{Q`%-(wED{fUdX#0b6ul3O?&=DGE1PF7DpC9gUj7`gg3WE{jFOR1F
z2*kN){^<OGmU*|z{_16aMz<_Buw}mTrf>s#;G>bDxMhohgGmd_(fq`|2Xsb@FbA_6
zf;r_}1UE>To-}%P4g7H%vZa$-|I1#}Gkr6a4P_w)XK7&A2#zCF5c{6*+N!^&Wu_@C
zk-fN<N(b9yY3;mku9G<YzIN=^koxc3ZPFbw_*Z1(*^`{hFn1<wCB0;0&rE+PsrLA7
zf%b~qEx%R8<PX7YaQ#rjH=Yy=csyE@JZZ%7tCWKA6ob+=o96Uzjv(xczV?RCNp6CW
zyiw(>RH+EJj1f@8q_7h|+>C8wQ7tZE2+a{SL4bRAAOfSdXogO7Oe(T!L(P|^m@K44
zV}r_Lu6|hSvZu~0K^^9e>zmW?%7ItUf)}`9iT$Hyr1)6(47i)v#Lj=MzSOGCz$FE}
z44PO~ueLbv!)hOg(s%I?r>m^|tu9qZv@W5jIt6aY-_`}<ivr`fI#bO-t)|los*6%J
za=Sd%%i094<*pAqB<h<LE)(=MXqzX`^G+%qB#hsrNACEzB;A&q$*fC!v>kC*yuXQ&
z9=JwWGvz>E<COYgJ{k4<ej@dYea`83F{EATwKlMgh7BL1NNSEFEYy(;azzu)YDruu
zIJt$_5t4GVDUzv3F@%2=tvJ|dGZsGZ=+7`r-m5@Q^Vc_~d&(DZLz&tg%@C^|1k#V?
zG@kWs*Xqm>?b9zl;Gr1*<N`uKmSnynJEC=+M?uEQa)7-fNAPQGXa*=<$d#nn%<lw3
zPvq@%YaDB$a!O=$2t;*i6;fF)rWw>fO?UQXDDsmgTe(`R+*7Z>@rKjPK?tj<&|j6j
z<>x!sRoUTsAgh|0;6`sL!fz&Ne`TuA#K?hk&_pW3?i#(FPG)pg_*ihrop_1&Olzir
z77@&))spD4?J2jIZrOV=9jX-(n$|9Y_?Ih7=Qg*IlU?jiOP3Sw*~wA%N^meZn6)BH
zccK#m+W<-ApSIpvE$7pDWssPQ>|?Mz(Fy)kZhoC^^oNR;SWZX?AGxq^3ImPn%Onez
z7I?Ka6`>`3c1WehJ-&BZ(mqtCO5xfny=4+MVUTMaHV!t>uOJ+vS(vV+thsN7q7E(a
z1hhQwMe{-~$KI1*T?Mhqfjfbg_JZo)(}mWwHuhvha7LUMw`niLwx2d9YWZub@n3>!
zNbYrETPhmolAbJsfai1c1r0Y{(X`%X0~DZA840C1p%(G@J%j4^rV*P?CLSXYjyn$Y
zbJS9oTXl3<6neMvN7hYOMzVgfz3Y&zWB+Dn*RL-Sug+>%_u8ax=i`-j5g;hE4eu$(
zH^AxCnmc(a09RkdF6+cLI7KxNh0B_<SAq|N`KtVaMD~S@T?ippofV*qpS%>%IO80p
zYRFf3-gC1wK3O@e7)kNE7?7QqWtH%I(+*}#wt(`ur{2XVw$Aa#+Iia#bwg}9SQH?W
zkFEY&7H!m7<;(m8rit;7XP0SaGfN@B$*xnRx4xUN1Lt`OP=2#A(n6(K$2^{aQE|4t
z5^m^px59YSQ5J(E4m9rBH8<}eeT!v!hGq$C{xj<xNpamL=74Bl>rzvnNE&L*z<lF<
zzpgo6^(m_;mUyf2hV#$DgH%VBpg8`@`%iQ4AeU!u7*bZ&*fc%lD;aG=r<;dd6Oy@9
zsG+vc5YEuQDtc+o+5e^@K(p%g$C|KKHj&Ym0=4A^jusjBgn#=c=BdMXqyZm=4mCfw
z=3<$|o?_rPV33PeqI#s%S_<3Vt1i5r1qE!CvN~;s!d{tlp~}l>UJ!8=(CKqz%opdg
zzY6kRPFXzRTSCFkg@ce$bs-AYjfF?dM#uRVQ+4_p0jTTAmmDZ@ci$my<CEA}Vh7Rt
zU18B0I}XW}#najIez#}e^5R#NW8DYB;QXujNA)ku(T+3!9Qw<X`isfoDtCLw2&VDf
zsp+q*pN_m~B`cHQwIhUV!`|0cpWxIs7sBb#`I@#ZU!+j6u*+1eD64Fu>E#Jsc_~S`
z%%Ru0m*HNXn3Ou5z2B8(u<u*x>VtSK{Hn$Xnl7SIZ77>_7)TP%rP>Dq=VsnQT-_es
z+i~|JB4sZLNMuNg+)hVynDX2f5>~&d2)njgP#6QCAgb;ezgcss;SX)OF@JF0Bw*R;
z8SRW7c@+AyMR%4&ah3&#Yr+sRYT1o~BjfwGt?fQ<MJ89cO;dpo)bf&<w6|*L@%MwF
z-?tZn-&D7)esv5~kz~a32$5l7luTy)(l<Vr+q_W3gNLQZHf#Lm8H5eXP!uV_DH-iw
zEC2k1A!*abuE)giMsbp&5Vmxnf$ktSH2b@D1%*E}OYvN-|Ar~O&WtSEv2>7r6l9I^
z`C?E!J&f!ppL{@rj}$orQTtZ7lSO-_!XjIb!8{uML6w#agMKdifrDxHSq02T9UPEs
z6dAXj7yGP0mC!wH_8!~v?>r8pQ}b|?6{2O`cdpT#`@4aKv#DO4ev@frp(y9=o6v)a
zOWgv#ZPlW`$b`8okoBMC1>|v?k9i;zD8{4Be`MP!q&}e4X)rc1lsz;x?SB&~E%BF5
zMZk1Nua|m1V?=HYuMumqAnUAzP59hGH|0z6Q|x~QbP2e~GkP_lZ>)^vMEdB+cX#-l
y>mcV_lQ5b=xmrU-Et>!T_&<|@VmkBvM+Yu7i-A4YzWXz(f5d%#gmdWB#eV~YoBA>U

literal 0
HcmV?d00001

diff --git a/docs_new/logos/johns_hopkins.png b/docs_new/logos/johns_hopkins.png
new file mode 100644
index 0000000000000000000000000000000000000000..d1ba7012a7083fdb14a255bce47f5981b5d22c6f
GIT binary patch
literal 76335
zcmWh!1yCJL6FmIkZo%E%U4y#>cL?qddGK&caQPru2ol_aySuwXaCf(#zphT|>h`L3
zYHNDBdp1f<MHUt56A}ObsPb~s8UO$d`FPF*!GB!OWj-AN0FsEUl$4sSg#`fkEN5w2
zYiS<hiQKQ>NGe7`k`<jba6#l6k}(jR1X@N4L=1&UieF2(>Rp(hKf@9Y=0?KA#zMa1
zsxzZRBP}BAP~^pi{Emzse%|u`Z9m_7zc=<)w;;M-cAZr_fzSy;NtNSL=YgOUOOj%4
zh79!fZ|$&(2O&|r02rWJOA7Z_Dk$I~Kv<ZGvJ;^lfby9^1_7PQKRelpBi_-EB-0I{
zAfZs59tr%4AP5d1?iVXr2#Cu-L9+g%)dX^30n-r+^KC$l1u&%x+?xR)Sr6&{P=G-)
z1uj%>0zi&!6)6qa2mxgiU!p$)y6gbXSH)fdV37r2m)H3!2UIlxo#PnDH2??&U{{X}
zV*p_N0n<S$Djy&;9l(*f(h<I7s=(N1{D@R~jc^mSz~>+X5Q{5<jt(m+{kS|n6ArH#
zv{{BYYqxJYA!iUb`sVd00OTg%e$4jz)n^>7Vtkx8wjRxd<){PxozmQV`)zlk)I}Ts
zwmbr+-k8~HKS6}xA&zf(G-t3j#>lyz_c32<(8cS3+{0y^YnT7W=JT)krlqCbot;I6
zUP%M9VIA<Bb+<{U&a1;?fbh%x?PmKw>L5;|AX(Ul&F;|~r2^uKM8r_@)x9{`mj>kb
z7m^vuUU~B-T}GU3bu8Cd>5T8!oCTCol8Fp`G}G^v|2CMPVMI50fGRstDE|bOKSmZX
zcZzfvx*nbX?g8Mc#kp&i9s~oi3E7(Pem@s`mCdCAAYbJZT>-#YnwnK>xK?Zc1OU>x
zL5x)rL^oXo%$@MWU2v;i$S<b6p_0_S-I5rRNR|*{H&cePU`hJ$u__WqQ?^-g67EiQ
zoA4wTG}dnIMl=x@^d~b=b{BnX5F)I^APC!xVmT7ZJX)737KEQ3_CV1k2a6R$O+FNf
zqfV)qz$3?@5v@h8Ek|-C;sz}cVJOp_AT$VY2ET;2%5o$J)u@ZiqSQ$>{o(r*DwBsi
zY3ad{5r>!ibJFeaCz0>GzxF3fZP1tFM5PAUI}ZpDxOyGfS^Ik=tNED`lPpH6Da%kw
z`#)8EDT8T=y0w_7_J+j@`rL`g3+pXOZbm90r>mr^vMBe5oS2#$a~Y8j5hhf<n~pN+
zkK$k2X3WQ40&`A|2wgd5s(Gw9Y&H!4NU2^vYOE+JPP(ReX!+`Vt!b=jnQ12tszc`D
zpWJevsFH@3E$wR%1d^G_nFhmlvUYlQxORwkD6R~UvdqO@^B*+-(Q6FbDG=^r>~inI
zTm8o=q5V5UW9`o`Eli#`eaZ4)^%|Ax5r43`Ju?KxLh$4j3(E6SW<J^A+uYk$9(Xik
zuBDKwq)qBfj~vP#q#eNAc%z_(<Aw}GTGJE$BtRgrCiEt#{7<Qna3+RuK#QG^<M_Y*
zbo6v*ZC%EC#&5U@!$>I#DYPj>jBHxPC50vHCCJ*>+9O(V#m1W83K?xK&Fx~u+0;_)
zk}S1jO(IR}Vxd0~CDwnOwZ!rR^j@Rj^?H<gRHgFObE{3ZTl>k8EeLCUm?`X0OT#^b
zczS88$Ckc%F}D2oO41S|Qd0a&K#EI&OQ%-jI{B*@!uEfz51Ae+i2L+KY1aYK>p15O
z{}_l0xs{qYtPJf-Qqxj(OXW({OJ$GScq>kevI^Tix9B|wp-BuV&>7OnTE~sUW$kM&
zX|5F^72zq12;Nq(me>~%{c4f+?r<tP`!BoQ7?Q2zT(s`nw*N#7W_#6ob$lwil||8o
zuSKDTUqKls5bon~brvRJ8(6timon-RmntwzF-zMDO=y}D{yiu2Lk9i3ZNK3*^|oWb
zWB+ejcv=ctMdc&qvkVR8rb327AtfQDfw9rCys`Ooo}VHd3!M8u_kS+@+-NY>_t9sr
z|5M*+C110nU#NFh-|7^mFQ8YTtF3QTD^s3TPBY(9-cp2F)TEN6Qkp$)-BWd0KU81b
zKv=cA2)|NM?^q8uvo^=KZs@D%IZJ9w0yl$yyow;8Ml7L~lIW8Jxc_y-5&R;!o-v+z
zCI;@iNj+q<)}6qZSmfHS#~xwmwawD}IU{S0Z{6e`avd2Hr5HTInkLHnlc;}fux&@d
z<JQ$^%WwKN>u2rH?$bXjm1n$Xwr3HmIiuV;4B2{I?vp)(Qk&|#Q)0+DDeU$eWdW*E
zszKIu`&4WeYzaE$3snsT-X-^D=lG&`qHa0cqT9lJ>zwO)CZl|v4TB9Msy)uxCSe;=
zfmH97?|$!7KoNun${HprNFvAp`uV}{8%0N~nBa=lUCq*|#P{4L30#S|;PT+zu5t>S
z!ISuMiu?Wl#uLVE(ZbXM4(2Bj%=XM;dIC`B!Xm<S!o^};B6UAaaisBhPpwM1Z%98$
zr%JE?wB}G1sAN&#VH7Q4wG&?EYvyhi>u3F{H#(=Ar7Q2CNG!n0Si$S$(7y3<dL8jE
z>mTyAKmLDggGMgCzs9?7!iw;US+SS}Q^l1cRrr%HBrq8N;C-jL!uF<#Me^+?>9(e1
zo8p~{T$D0oBvrYRV^sD_JV|Kf@ulwLDq!Ai3Z*NKZ&gB7TKg%OjhEp??#L-3*v1i(
zm=NzOSwf+lRBwG`z4z5?q|7R;=B>{urh$z#1!2g(Hogo_j|Hh-vhYAlpMHT6p~l)u
z=WEAU)UZHmX*Sc?>1gfnN<3fYdCvPORyWpGVtqllq7Dm*?p|A&oj{GStM3ZZHZB}f
z_!syx;l}8CM#nkT(o>zZihwp+KV|zM<F4W0t~x0DV`Ocl-7xU+>cXj!sxF(}MxEJH
zcovZ;Zd7yy&I*1SZr~#Qs?RP_g!6Dm`fBpu;?mwG%ZabkUk~@zS+Wh#QcKeI($Jhx
zmzMvcZhZNBdVBYIg)ME2ex)U(UaIMLhj3%7RngG+n2?xOzSfP_lNRsU^je+S!h_S)
z8IgbM9rwxHib3oBtLRYsiZ+zC^4!-st7rM=+Iz8zsFK5n)MrPdaur9{(*c-im{$ZU
zwEBgn**TJF5iO0ld8T7ov;ngAx2$t-7F1qT?H`%_|EOG~eRmG;5jXBhQ3aR1zWbML
zA<U3@kyw+k*-2(xWyt*0%y>31H)vKJ6`h(tTK9XAT_Hoy<QCTRRk+p~vRyt5o=LL4
zn)sfTk_8T6eOxtN$*cF@w;12BLw7!J#_-sh{`?-ZFVreDZg|z`aaFUji>=eBlTj&X
z;&|k7B>a{%qdL%%*^*JwWN6=X=GRu||MmWmD4TD5?WWE567PL}E?Yokt?k;+`M&S|
zb^~uMz>V&8e57T`o9?W<RkhXRM*LJ_Nop-(PR!tKb@6;P(5}-v_j+no_8vYgWa2I1
zHV$11E-pQ8BpfyTIGbNoK;WO4+w0*=k<^6c1Z_6{Ti)A91wI~Iz{A|}!uZsGQ@_%#
z(vmtHzZJY5bsiH|uJm{JleQ1}aB&#le0e_Y@?P~OIsd(ZuDbJ$Jdpm~=CS%#bF%0;
zi7+YY9Tw)l>lZ=R=h}~YlH5X0Lm2@4XaE2b0sxQiANNB5@MH&oBU1nn{0{)|ofA#^
zWIn!yH&>LE2HyXF<hB(heY`<*k<)Vr0A%d{pHM(XCc(#WKwerx%V+rnY~@3;P<N;A
zKS4~(q0d=M60r?Q;-1WR%w@=DXkgedUtX_oXxptE&D0o<FO*N>9?gU&X0Nr>E9(*F
z@~=D}7$qJP5QH8{Jh`*WmaWEBwElW$d{N=eCjQ?hXQ((}DhG7<y5uxIz3hrhQ@}ue
z-*TZqFvwkR(aE6K;;XLJi1D~#x1va-F18dG04ezJWS2E?i<z?t(z?sIxy^T<U%7}6
zm;Rz4>7P>Y00N)}czo+q<@~pI&)%OdE&bmx4GPpp1Q67yKf+}FJ>Yp$)MrFM@q#VR
ztMZeICEk4Kx18^VZq}&1obe0<Ryn_1LZ9OCmx;Q2F+1YNanBkb{#reExX-v#iTOMa
z86q)^Ex^fx@?DI*oB?raa!H`#;&L7v5SM@+S`lTj-7F1)4l?*G4vm2%{<ahM(l$vn
z(KCl%2B}&=)lfM2G~Ria&0l7sg&Y{*!^Uin+Ouuak%<h6gax3*@gY#a(0)s;$4htT
zgTEcfU43d*r$1d>nmi8zwJ8aF6;U+nS_#PMc=_9halxoYbfJ4rPN9Q?ao4oD<@j3F
z$N)ft0v;;32!T*kAb#(92b04Y)D;go;d%OZJ25<A;_0S_;DQM-0^NW({N4opUQSeM
z!z-zZt|7Xo1bVBBDXfa_Jp4l2y9aF;tfV_x*+jwRygzhP{e27+5Vr1BJNj~#QUZms
zQz&e24Zccc<_y+M<%O~U-AbSX(BuG)C)C1v9f@!Ww0{LSYXP{>Kc>ZnKm$9G)QPar
z05z<*JDkFV9Vkc+2!#L;^fRTXXTI~Wdn3PZr{$_AdZQR=z9JQDeHxE6Q<iAv>Cm~a
z04xB?We=6>MCGAi`#TR0g&ENNY<lBOS7}Ei5a=kipnAS)OFZhkHIl|=ntJkS-u;8q
zM6@_X`0E{yLW$F7!eCPTgQh$}lmqyn<Y0g`!0&nGJsw`#XDvM@hztJtd4@fcv2)2j
zLt+ahdx;OW;bn_3mn26Ws?%9lk#TGzE^s$|>?Vqd_k&M#3urI0KPyr~QNa!oQYoJG
z2m*9~+E^PK^S6PVt3kjxa9+bSGJg2jWo1UhWyv;6jqD0Cj9CK*1E1vkFyue}@{{@@
zTYAxH$QjD@bG8kFg4r;>PGe@L7sv6E!ZQU*7bqyCcm@RgX}zipc507)eI3d!c%7QP
zlAF92qZ(T6@6v_m>{B<8k<Mitt+B!^c6sP_p>!X>wXRrm-?mu6E=R@mxs_)@qF4-%
zVHSWrR)q(440~K(pk4b<-f`IhVSv0mE&)m0^)%2Z$6!tm#*ruBAZ|CeZXBKOy_ASm
z)D?>O(@>|p;xUEhBt;RTIPAmL1R0#)6Sf%)WGQuH-2x6WxaBPRR7z*6?hb2b7hKgy
z16}0tq%huH(_^!uFNu-5s%rUD0Hf>G0KONFXt`Aoq6&ur#0wr~5bf4(yrQ+fTbo48
z3ySR^35I@SZ46W-Ypr5~uN1MzgcW8)DZ0)j!U}lpa{mti8BB~2il4o&wesB`j-*QR
z<lLonIEHVmx8QocohVDdh+`Itz~YL(ch<!fU_!l+!jY~E0!6IOp)Q?p0cl7X80;7m
zP~@HH>%QSS2p|jyIj{#mX!JG~;U4&YkiMGg55#|`B228hJ`z(u_jJBF2AsGT@VZ)y
zFx#)viuu|sX)plpu(r)F0w;%xT{)!d2uM;s?;|==^){Wg&oZ0Um4Tp0G*}d1E`SRW
z!YE#Gc=Qtq+|l@*B8VE;Bz`~$c>MmbC8TnB0En*aAp_Fx6Y5Zt7#xpp_GDqYg&^ZJ
zXlYm`p`*>Q{W3^Aa_8wknut85p715Fp)0aMek#S;Zg@Zk;B<aFgXyc<fD@r2H0?B%
zcLxcxqM<vSM!3)O_Ly>_S8GANy%i&`L3dh@lt7!l{0o9PY{*%QdSNB^-D<oeiysYy
z0s7NvD_SZM*1gs_7fi^`ji>lB?HtNbTg9?wSmS8j*^@^@;-{0zM&76ZSTN%(e5)G%
z<v!$qsl`9<o~)yc;XdO>*0b1cOVM;u*%?Gp)kYd1r&SK@Nl{<xj}ph#BZmTZ$Vf%2
zD6?4%Ln9RlEzT0XpoeT2JD5^gdjY6c#-i}5G=}d%6rs|VfN4256oE#!O9|jR29fI|
zafp7<A-5of6{oPGO&Z&H5vvrrl$ep)w)|<6M|LdYftD|4nwd`Yv_H^1esrXo*-PMm
zuSVIT#YHVO!ww|aGDB56P+$NU67z6vOSuzZu*Hw_1g-_o)#sb>w{@ND+2F`Qkm6>~
zW9e!#I6esw-~4`gzR|qvAby7wcFmq(`#q+aV-|nxWFd^*?!^i8yecl;NyrIJE}w|n
z>pmUd`>ChBFn*(U(d~qq`Y^LAX`ig)VLpW=`=Qa_PJ|5_V;kjp8l(9#fyDY)GP~Rj
zPg0x`nDnslA<sU=(!l#kX<SI2WXZxUWtk|41w~!^IdVT*4gYFxF06F~LtO$02lXKJ
zh_!FGTCD&O3Ye(DO(H(gz}SJ)_GAteasVqx8de5&+Re1@m-JU}&LSGz@y2)!#>l^4
z)h({(irAW1cHpt3r=eYZ8pZn)?S-l+0LV|id)d6=7b2w379v2x*T}7ku34urD4sCz
z=&6R90@ht>5;cL!ZV+Jj1M}THPSFv7g{Dl_KvSRyxc;TMa7Kq{Dd1*CXet59bwLZq
zSmgw+jKl&A1D@Bc=U*?gPuZqu+PISzj7gMtVI;9&?4|zX*Ki}jj=*9BK`a;du{_N-
zYCo?vR<j<gy**v0iec$qhTxnAh$UeSLk5YV#H)CA)tD#9mPI`RIDd+fLG=Kj$WBxW
zuC<V%{V-}hY!`{eeVN$E;L(QEL>vCB0mH8NvAVbU_=z<{$m+CJDwCaoXV1eJUBzO2
zRZw<a|K_CfvpfOxpiliDL+_0<Md6ZTn$Ah9XI5oC^N8^2mV}@(bidByFKP;u<4S+2
zqF^y43UhltRv>=X<mIm|%S-WQ8)-}^o;3oy(rXj%T|(rR0%}0B!4~=_B$W&6H<Ohh
zoTT{pf3QH~_G1vL5E_~@Cv2+~d$wLBQR32kOeX^wP12Vdt1XBS$(`@piN=!1K5556
zyQUb}ne%!i?RTgKIUE`t+&JK!Y5kQKXgzCRE6Ts_w<olq=$#s%x9d0Ekb45i)fSKQ
zAUY|gOK24G8NAZdYK*E)d5D03RC^VJo3QmkT~5Cr{)i|Bko!@!wAA>rG&c(5Rq8G-
zg3n5p$0{HD+rnlR*uv0_7lTrbxtJa(k#V}E>-7|%r9lpdv290Z`~LVXMs`dXgvAb*
zo#K>22igLqh(7s>X1y5cRv`c)#C@lk!irn7)1fQ3KeCbn>xQ*984Kgrd#Ke|$4Vo2
z4H@PDa9p`owe%TMdD~<}#~B;IeMyGHO|{GuzZwow!*`RK+=h-FgoD6BPoqGAtC<04
zW*boXP1BsrGgnOWdhN$du5pW47a8-Zg>=8Jeh+JZyWdycMC#~1)*xcJlz;`+FrwOR
z{B$a8Y|J{4?O~k6JEy#90eRTADP|1dX?1l*i%9nuJ05d6C-hM`|Kbusdu4IO-Q^6P
zKZGMZiqhxpapz4lJgEuU5Xl+3in-CT8x$G&*G8~|FJpKBCss+0GA?F&7|yfJdHL^b
zrJIi-(x(~+{!L5_pk8EgEeG-Kx)`SNveAf+f^uS~F66a(R3cVRna6l<FOz_hD<21l
z45t2`kM!pd%itI$%Q$V>u9S42;M^wm<u1<?RcP|YHaZO7VXelIy*60g?|vd1ujAWB
z@evu>a^RB-r*|yvAq&-|gRBI;X3cv+8b!QQisnbk%I#kIp)ovWP1f)*w6Rv}`QE}R
zK>CHR7zMGDFrE>+EU8>BlgTjP@M#R6wzltlmt=}o*9M1VE90vifjlmRgQ?<=kA^&!
zhZ*IPWj!eB``&|2@aM5DMQ>D`_1e58_X+CfLE`#23W$e2N++DhihD@;TFB)+3Z0v?
zoF#IMhg523ep@`FHO9psT6cO_OJsOC160tW8YdaC{BGqq-fgLWM8o>u0N0s@#_o5T
z{J%j6dNR#7c5>KaRowz0DRE|W&-CC2mq2FUs7Lv|fu#ddLh*VV2n8sUR0|8j3JRq@
zh_U*`mr<Ra1_9+#oU|PEy@vP#4=auSNO;GVpF`0_15GWsQ{Ma3I2R&85K{x!nl8p(
zuFs}63G{OFV(AlQhtLp=ykg<+)b0gDaBZwM6_+%CDHDZUIB1Qdtu1$ijpS2Q7Dd;i
zfZE>kq@Ttb`|hkbBzw!_>Rgrh;pQofM{I49upWBs(1_QCn%nBzKv#lQdgfiZR>c7$
z9I0ja`=tY>j|7|h7%Xf?yeR=zGrXqNM1>2yDE<Z%;Lk$!p}B&B;vp~qqLdstd4D<p
zObFQjWP3eiVQUrCoMe^X<yIZ~lvI2Y&;ZRkGPqQuV5P75#HMJ;lRAi}y8S&Xm5Jc;
z{N=PM(D=PYXFYLpk)=;k!XDJP2m7*gb!M8igp80!ptvi#!1Ld#4M>mTzd~Y5BpKTc
z#eg7CbjF8^&~AhGlt&ox%1anj7<rF4zSt3(taTSh`@jT_gT=x;RO{Bw9<|#z82^fg
z{X-i|28hbs<7dd_^so}!KArG@q0CeZ-U3Y$T;4-(Rwl^(wu3efNFX8jUc-$NO0k>v
zR*%cd?<%q4?hn6CtcY)%G>jX^rt7fuM})?Oq1AGP+d!0AB?6?!tc@}-({~^Gde-o=
zosbpSu_Z-k67gLmRC1XYg$2wQL_;JT2Lnq(**$}hI2M)pcKP~xse>s1W=zd<lZJRm
zMC8T)7RLT@R$4fcXv7;DD$>SrjO{=ti_2uaF13sDw5=Cu<?pqAJ-R-iYCP|9kpFpW
zMZiLJ<D$~I`CmR>RcFqt6F(-oICWg%48FPEcE(v(vb>%EhM`1S{2*=+RcUipB7oEk
zAOay#%qX;Ow5JjPX<A9$Ji;}QP^r@rNLl)bBC&BgS9>eH+ZwF=V2K=F_eHn<zP|X3
zR4KI_-5SILwQ)q!cIaJchfEt5Kl()9_}tg^d2rFY8A0Qxj3a^|5Z?6Wt1QlK#Efk^
zBNiVeJ~X@FVPZbJU_kjnTJpYTx-Hjaj5y@pjh=1wOV>YS!N3e*+o13gfSnE69p8%%
zvCvScpX)saNJY)$PivLT=))|3hf+<NEe8UC3JBi%`*en|6TjYrLdAbh^i<gx<OvMW
zOs7qc@A=g9N6?|v{QSvok@P_m<%tue&A=y7hp48%DyXN}bU!i3k1a#bRybXz`JBF8
zo;KyZ)kxPb+Jb%_a#3+L41Yi&040$as?-Hsi(`%Gpa<c4z@VWO;8FaOg&|Ej^Yc4@
zwa1fox@b;57O+5+Qn1a@cMDSm0gmxBVHBGd9lka6h$!R#X_!me;P|DY{Ep8hbko8s
zak!zcji5cZ4^T}Mmwq|B%hg%RMRuK_2Cc{P@?#7VWeZC0mpgllR2MkU?1_ZrjNts3
zx&EBB1ReH4^S*M=J@|)+E>!wO%uIY#-hYr8xISk|2B?A;&zqhmU5@^G7++IjV36Mo
zaOS$*4J%c9kBL+IuM{gDw^x;jB!;peNK`mAEPzf>v!a0#6c0ebWjq?_3=*$SC<}x8
zkP7H+WWv~bIJe{On)8}sGk=c@{5)9AJfQ+hira!1kGex5jmQ7vel>1_N|cTtUqNsJ
zS<x?l!8;D*3cuq*%#mH1Su(6mm!a~(ykcjxql}B}EgPVtBNT~Mju&T791RPLKVPo;
zb4G?zjgM(o!T@cBY${4F3#~>VM*-M{klGS(rZd=h0)tBFG#FSP6bp*Qu4h*M(;C80
zmHPEhgu&hFFrkz~Q&!b_py$RPuF^pAhdN^7&sYksb*2V88FOVgMCWjjS1K~p&moX~
zhIG(zJ;DSueAsnv8EDEpL4DE}cW=619VNJgwW2PdB(NlyjsB?fw;%SwNa54E_iY=y
zny>6b#G$)0jlwbRpa713HEo|@3Os)6Xqsnwk}IuQI2>+u!8qnxLhh#k%OHz>jL6)0
zCIT<&q4L3>#Av}O<4hQl9Iz=|QgXwb;aDI7%oJZt`NA2N>rZ#rx)Sl=z?dayXpCD5
zXL!I12EvL6T0CeQaFszJo&*7#;dK6HGf%t$FaHMnBeKum6etaZc`6B6pI?;?!u{MG
zb2rY_D^Fv-W|t!S$|2>I6-q6h+oC+Ko*%?njmzzGvnq8HW2NWm>^}6<)!$}USMquF
zJRYyKx7UbdQZT?_$#Fz4ImGc*h*(6H%~yVkjF*w_B5<R;K1s2eHH?S*ewCpv^GPd%
z2KKKd84G6%4k9!rg@A>bG&{8fw-Es7M@2GuQf!SE-5Wp7s!YV<o_dP;JE)#))qa`B
zFG!}QhLX)>H>JM*?<zKXOKHBKD(*Pg5a@Iur;lPbHA4HM@HTPB^_v*=KfBmO4tgXC
zqwOIHMWc95Ulu`vi7;RaPMn;EA0Nu?GZaKd@8a|ZA<_o!kQ_9rCV)iwsKn^J6IB+U
zqf~s`Xe2t`L`H}aqOOqm=S5A8kJ-eSkmARs)bIMqyN^LQo;^azqnhrEBb|cKv_Meg
zj&KxF-MbcUA+kJ%BCftOxa7HU%wKlayVV0u8UH1S>aD873c}D&J|s{|Ctq+&|0qhJ
zXqnPP`uy^VfIXZLe_I`aF!YN)9RH$C{Zmh18d;NN`SJe!$y+ZB+f6I0!K{vbGyNf?
zJAIzu^0e~O|D7{YPVCsVSWpVq2z23Z^s8#(dihOv@1Ld0H5O(mN<I`Y?MuQ1_uA^w
z?sIhdzWh4fU-jNrDfHVy4M|UsUyW8A$*YVO3NWc@f6ckftftQxHpv*nH!98y_chAx
z{x65aVHPDwLw`Qxhtr8Jo#Q9xh7t(%kvkLff<BoN95_3mLl#`Yanj6mV6{S+N|AHf
zi6RT#&!6O^jk^B2`pettob<&-u#%0vmws~RyE=GrVP%QnsVV;EXG3-ib+nIdtdSqX
zpusxg`RifNMbdpH9Qt|mx+cB{i|Ay|iw3vms_rC;m=vs(LSzy0i1!gyjQNnE#h0`k
zp~tzv4Ot;E*x+H}F|Ge_QJSs#x6R3f+)slI2>&(PFnag92(KiJ;=VS=kA;)}B_Ywj
znfmjR0GElWg+D?NPJCIi`UlnbI%_Opqf#HE$sVcO8TyMf4c5gv36jr7=8(P4U0gnI
z{`F^?BtEV2Ds6lX8qNjGtbL(zVOUr-fh0+d%7wldPl-yKKd9?x+MCPD$1BaF9^~<f
zk#i3|@5QCd?IMYzbnq6%w?nP{NEt3%y}U+p@D`u@Y;!40W_3iWf4^sJ<A;m!lJt2?
z2f_mMp-rtLD|YoC`WoAh0_1bj?*)8!1GS9StE$qs&GD<n7bvpj2-cuXvsDZndyp>!
z@lC#6NVZ8YR(eY&gAy8@vkWfeB%H2U<?K`BJAcZI>ZFIQfER4saEvF##G0bMeFw=+
zQ-g_zq2@_#ppU+f)=U60!&XmP*%qJZlDnn>_+lpuAwmp$!mjARoVh}Z`-e59kqx6w
zKO7y4<3r#NiUu^;X$<pji*~}cGTncfJjy=L?pGu%b#I=*o5@-#LLwQ1%*0ANFTi98
zVQd|1mi)gz<mBmYl^2<;J5|GzHpQso7C+witQ-^0S}PJ_m{Mem?=JCmnVID%BYjfZ
ziYgo)#W_n&KG>`ZiPDZwyG<j`cm}srV?V)T^g{$ia2~sq!Y3`<FE%L8-@_x{&*B$u
zkEQFYls_e61kLQIDCpZlO9Q#H-=OMGAM)RW@cq^w{<R0GS|ayTN@>_S6Xunn`Zli|
zg>2lnJdEjN)is{?+B?l^^>1nu;dfcTKcGyqoQFTxqpf-Fm87zX)>&@`8%6v>{cHq_
z356n$F@b|;b+sIkeb;2HsPuTQUH*3Ljq<6gbsc4dS<>Q#SR5YE_?IYS;&2=M<UAMg
zx)hW!yEJ4tl{9>;w8cyvT88a*^9?quvr<w`zycB(A`zG26#2m2u>c03V9D9#JHssw
z`woK(_hPyy2zU>}rOil}Qbn0faMUJyqYbBYIWh;cnyQ-i)_>_fDOL)Cr&#-euw?90
zn54w-nJV{AODoR@;Z~>j*xf|o+e<SM+j`#WgMJkB%*8L+Ru^+-Wc)q(-Ai)*V0I(y
zn=_D4N}$i$QQLO8wY?F07MmSDn^UoXP|D1@F!2O)q!p5=ltxUv<&~;#lrZyhDu~}A
z*X4WNT*Q;U9W=WC>dRD+ja}(P-(7a|UzxPM7I#1`Kb{in;!J;9<GLw4@?3dJ+RLaS
zw+UPChX=(Pl674(#jBTExJvFZ_j_y-w{#>D19lx0-@_^^A({{R=iqGri0yPS2VE^y
zNriZgn!U&dx*43u|0W6oI8#Si$y9cS;&<R6GE#E5w4eh*DLKm+Cj15&<CX%UyK<gW
zO!@6lY|T~&q#we>^%VxX@BP6lWMq+raWRG0pvVANhiMOv<|bTT@hwnh?SILfm{-k~
zn#!hCO}rP!%enh9fT>`x7Gl2VztB_`Q^T!)LFHga)w548=FC}s;C35?OAw>=<~_a8
z7jNEga~S`E__Eq8cWT(y)cGwunc}S`-uE+(w}uczeVrH7zNv4>?S-^a9Wup)-)Q~Q
zdaka8qjJd=yR6Zwp9D>E=&F2*HDOkR{^++Mu>NaLK0vlIKjLX58#nAL)q9Ne@Otb;
z5^YrXfvO4;_(Y+ZK4OueZ%((S))ogtb0nS{-@f^u-zBB;{aQK2`$!nl_L1@c25lcA
zIU#_L+tmPma}E}Jn&~mByL8x6jF6dj9@*h(K-C8TL!^+xPv^JRmB$Yn{$NvYgI+5D
zkfTW7#Q*^yLKkUT4AlaI>?Vf)#_i9S{7etPhpp{ZtGd4&DsJDedke-)zCEVBef*T|
z+#?%t(vn<_0Gb+;9?2T$jBpc=b3Wb`2|;ZCAf~PpXQ%J__oh!uw!fbk%LpuT`4%2>
zbmx{XbUGq7ZZ41!upMtDe1^hREPQJx{-j!Kc&(bjV8J@=Dh}iOPoJF{Nqu8aD8Npu
zn-l5q-jouV8f|g^OvRpHg8sO6Ij{e|G0)=gaMxpiPFjNOXS<yN*YAo9-QOmp>2Y=i
z=%ajRPckB0zy8O*u`V)L3x_&k*pj>5^ik)Yk*aj4;A^BSc^yfv%d)OC-kj1gqnwkl
zse0sXGsV2^&95s~ks+yz{-rwdbKL-(qkUG(fAAdVtjq-NcKw>whBKPQ5cokB)UEfc
zj`!js8t%43!MfHPtsnWRBvxP8R@>lL7czMmHNy44Nr&tB#ks17NtN%$X2XmxxGvq_
z68<x0M^Q(uz}2x}$5jdwFY9Q|#ZZgQ^#ZG<Wa8hAFvjj1gY2WT#buIrJxYUC`Xa*L
zW}bb`539px4aQY>o~8P)Z2ANFZmdA=Zx9Sdkm%FTv75DqtbA&u&U#PTM^ompH7A{w
z?DgJ-76j}x!o-qb)|I6Xdbq}fH~ZS!HvZo5k>a-{atv4<if2@euxKd?)Q3mh4sx?5
z9D*dMnX9>~>gaZt=Si$23BE~K<olkBAHlSs_pN^b;ugnN;kKSIBa!|r@#FE)c(}=#
zAphTueXpT{)7z(hD@>S9Xkys<b5g$h(L`Ap8~BpIZx7-hI{mksPVs+P>8)S3TdQNZ
zN|2l0ea^zeOwb<I{1@t;3f{BsHLt9#(+;8dY~^8rd-pZO^Kq#tzV=4vgWpTR>s!>*
zTCtN~$ePa^X~uGzAFEGQ-#4?Y8NKSG;FL;_;_He&gu2yl?7xNUUgJ44hr2Uk{Zy$B
zHXcsS+V5{-Em^tqJyd&Vxk4p1djpS`X~x8aU2Jxio+5LOx&wwVeu?h<V~sr-A_3R1
z?tHQ9{&oBO{LguPgUo~)F?Is)R*-@wxe=O4t^P(c99-czf<GJhScZ!HzJwXt8ehYF
z;JWi?nOEWMqQ3m*!yWBOde%Iee}zFx+$qUxwKEC4QjBtN4JtG3aWJts*Z8q<+xtYr
zvLKU`-oa*zCdI!wr)rEf<@m1V$P#crDe7~Le+J!r-Zm_jeYT76WPZ}St6W4%tg!L6
z+HO|)HPV#<L)%IpJox<0TI#}p2kM^W1J~<tSwt<iRs!i-0$#uVy?Q4wD?aEl&6LoG
z@iJvU{S;Z4tW**C^Zo|`n~3jaaF^v4?!@U?i*U-q1UX}0i#A(Y_dn#HJ@*wyf8Mv{
zgkQ-q5T#{ED+l<kjhMpLat%KF-wh$ah&MZ4Lzt3=7@py=Jo=x;X;xZfL>n8}?lSUN
zyXUS}GT#cT`xn4Z!>QONH5CG1x%3RTxaEhteFPwXTv&6C6*|Mm*+pB#&X(Lv|J&su
zSzxx)ZzvafIs>;|x+67D=`}e!rARIAH{a*umW!oG`r9o0l5Q^-rK}YGEyd@4MFSng
zUCXW@uzz8)Ai}in@^wE^w*D6E$e!KOO&~H1-%w&H^t1b(C$w$B&~vXFCXy^(T1=XP
zhCimpfeb<yn4bN1?f>mwc;h<CQsvpF-qdHn>Kfxl`53=-d#=dyMP81ftd25M(z85k
zpVqO*P6uHns<Q*^l7r&2Xlu}H-a#mj`(+b0Gi%~^=k)w?T>EvfmDr*ey<?ff?eA-!
z(R#1cYs=C6y(xWZN$6x0YJ|}9)pqO*=<Ey|!S}U~qMT(Z_)EW5lQw*v#Due|$(+ZS
zfW7|qd_SZX=%ihAfccR4h2Xh$m-0(KZ0QXegXZ5Rs{6ME+IjVKABOPNh7)|8lfL%8
zQj#gQ_N(?M_0oUi>sojBbpGStJnSbN!}G;)_Xsh74EtAqHvV7N|F$e$&%<eMC-J_}
z!HH_$)oBHlxUj}hLlm+9xWK!x>-PqVuDvd-wN>jN$vY5-;T+@f1-!MQB6rj`8smO3
z4)=gIV=^=RGJ?((D*n;)RcD?5_NyHHen2SiO#MlWo&p;5G}|5gw2AP3Y*iX4Ax&Ox
z$XRhQd>?F=@;x?(Vb@YHhSDLufw5;X&S~Q*p$%O8j+-bMOAag_hr}5LJWVg0kjdSf
zG`3@6V4_eAizc~ngTtZREJ1<(FP*+@CuEOaj78-I8qwGi%9AVL)#*aE)_=)&8@K3I
zpRKbN;19K$ipIh1AJ*lES;~h_Q@cR&N#2AzgHONR398a>bGTNg<M3g;(?W7s!|E)P
z7%Fa}7aFLtZccPFSQzvbp-S&SrK#bW%cRB>AKYKAbCWo2dB$3nI&U=<n0;*0Bb5F9
zB%}wG`g^l$99+vJ?W*V;OCWx+zTYC|Z_0kjk#{tJ<`gzE-oIHX`Vh@?4$q|z2P<CT
zV!IZhmyYZOD31@faO5x`0zt25Z?ARL4sXve)hEZbpW99fo$y1NbNADVDjl9|6s4%^
z@|MIZ+J)9rIL4`g#<z!y_K|w8&mE~O2Qf@!`Q@TFi}FnRqF!$qPl0+*p-E-CjVax}
zS*G@?NLpxF-ooI``gUC}RLmbq@!x0lj{|r9B?%g1j;Q<1M9VGh>yyA@0I=J%sc%=e
z1_v2BmDj5`x(m3^&)Yf$*UNoP!Cj?LxRj8p4HElO^f%EHckr8|*f-8!8y8OsO_rM%
za4Fga-53uG11$_X+@gPr*xve-B+$3`>x)?r4oz0E+W+$I1n%t)@Nwi8;~Qz?96)V5
z49V3zyqJ&*ZEm=_<R{y0qR6zeW}8f6G~b<|#<T|-4&XPVoH8Mi%Z2F-ZrKdLApVB*
zR;2QRg&|r5JqFpQzs=Q!Y`vZTRln)>Ya8Jsu9i>={ge!R*c{Ob+`MeNoGp&&QNof(
zvsnQPk%TmMTNPw^;$htu{5ivKaDD?b8#T(QBr_lb(cPQ_Y_=~Od^xo{&(u+ogHiIG
zTMKlC=L&1Z^k~VWN(JT}t<Dqv_ji3A2}8%EMM=(;$g3+F$AK2s?qi!vOm^`~4$|GD
z-mF%`OnCgCp)E7JG@s35&iv0d`QWjebP#LhP(Xk%YqoKF?pbq^PThNl6wKwXG-n%`
z;|E<8%l9Y{=~6-`KKc!KYt!XPvnlVF790fShx@#@p?hEPsaW-*iyD9B*<z_$!L*a}
zzrl_q0HPu1Gou%?SgK-h(`vBASB#yXi-|D{4}SjOE^*M1>Y6zd#vL{O2EMYB2Z@XU
zs@35ED=#Zf${J5(%3r#??`{9jHXm@Cn!Tq%I0rAXz+#UYc#_Feewm;(ObA{g@6(ge
z(wNsCV+DZM5@kgvy_TvsY>Bw$jw_=li^CriXJ^lhoB@>ED4+q$T9u4w(~buAP*lsE
z)Yp&}XyZA^<%1sV(iA~c4#)cI6}G;@<^*13F~4U;a(NvCN;l3oadHW9;|?7!-iUWt
z;1SWd^MFL<&BC>fanL*kGzRI?Bi6ih7)^I&&hO=kIl?|g{k5U`Wibd8KW>|y`Ssgo
z{r=N}5IcuE0f7@CY39M&sb@_`!)~Xa7j3>Y1C5c}@U#i<!h?T$?KC1q%O8QOq=Fyo
zZ)v0pTIMF1Ifp)h@PO_l>SF|gSLM<&?yCyoiyo%(;C^9j_+(urf5EG&QBSz;Dh~aS
zO`VA6TSr=%JsgE?Zf)27RQ$v(dyi|J>B!5N;V!T|q5o)AiAhnVnL67^c#KDU)0nb9
zU=e@|7v%19$Fd*r_)FEe9?MR6Fh730E<w{5ZD3u|pTxn4zk&+*ER|pT_%GzcfzWnE
zhN4HpR`&b_h1{N`$DrmFXiNBByvsmQwUSMCVa5Gg`udR=xZv7QbpNX#1)na5)#XH5
z|DBEg`X2gp_VA7cbkcuXCgNpz99RzX%(?PUK2uPf<4Y8LO*aB>v|6tW$}nAYF+7Bp
zbfpYalWq{24k`VzHgg~Qfg+K!3EQRnu{Ov1<;d?7=RD^`8~(~hf;8lwM(r0QCS0o7
z|D%679QDjP*PF(AlpZX-(}96R0z(eJ?l#q~q^7jqo;w2GZQqD;NdJ_f6(zD%pb;Dl
z&j@Xu+>&@>iSvCKfvT-|ihQ5CRyp9msoDMsVFrwBP5NP|R2ci=ZCLTspeprQ;Lu-4
zX*LYEpLYz#O>PTr9oJcZfx}>88KjeAv%J3ij=KcgWJh-u(A9T0e^y?eu3ttBOXWH-
zT2p^N{tI8sl{%LZ;Tf57&X8HGj(aQ)6lRUt_sh8(xx}m#5O7bqCh0*i7scYgy>{7D
zqY*-z&5<W5&#?-;LQmTuQ^e47;&F#NyOh2J_+x{BU{c&*<Aw8$r$nm|>+(NPOyEWN
zinNO`b|npnE4%@EzZAN#_k|+Og>_q+qS>TA(Zhmm>2gQjq4~jd(y~GhSNf~0^x_nk
zTv4UngP&&aGxm5%=9`)43CE$NKvuypztW~YeW_H}xq3vKQ<zQbBq2t0^W|ieB9WIq
zveF!H9&~a7ZkHU<cW8iUID|8q`Az48FP117v-uN;lT!uUD;F`_Kit@>m=LFETxEzu
zv<0D`MO6m05|nIP-^%(LDvVZd+~X97bY#ATT%H`No*g{S)=Le|-@iPlyAj~<D?$Aw
z)Sh1=@MxlA#{<#7l6pQzAxrUVjyE|(l#On~vsIB8^{67tFH7enjN0Yl&{T26R%Il8
z(#AXP2!OjGL6FTReKpEpM<Q0)nI}1jA$cG1zvCR(=lG*B#s7m?!ih*RCU~6Hp`r;B
z@GuC+?UkHRdwDYv@o`Zg8zSrxG9fNGRm2I%9<;O1L@#t>l`_6%gW`KX3<{ET)&8xo
zj85x8>MUrD+4^)tv$Ej9ZHRx-Y>pjblj)YrPInkIZ_fQ$ST8ZTB-`>wCnm+gm#70l
z3~4JrUR*kZv3h}pASPlAOci6SpfDC5{m%U%EnEubYA>6;eohzIN(%@1x_zv00;?#N
zJ+Z903Q_+5zJ>TkZ~3Ui&SKoH2V5eOn6&MONmY-i9153tOG}#~g-}A?!atIren<Xe
z3u#&+LtBjY#5CDh=H=>HyM6sFtwH(GjqB11!yvk&{u=%7;u8}Zbyr~#T5=77nEFud
zlDno9ujPQ8lrhqW$&s`k#V-rD?a$q~Vup)2KJJ72f?Bw+(2}&3QYOsTg=JsA?KXzh
zksgmVf%9PbeZ`vUBkfJvZF;lrI<Oq>2BzccBaC`9MWTrjwioY8_xmaqi{Q-(t5#T+
z1Ln@v-IOuU2iE<%qI13-&l_>r?lGh(a5_$^6S`;*+u-`r0G)8Rq?nhb#jXM%X?mM}
zwaGBm&{fd(X_h8ifAt;~U?bl)jK@27Vm`>3q)sb|wG;ic+>(77&|pDE2#_cJCTA8A
zx!!Xi4AqcpcV=HZ*M9V>9O>qf#<*#TFgOYZ-vww5)9dilBe5B@5CefO_V3oR(_stz
zGdmbi!N#5@))t!<Lk_MNe?NEi)O>Wno(CK%_;@^<dX423+8Z1}v#IPoW%~CO+BP`u
zVtW0>a-sOk6Dqq}l2}oAQg1n+@t9pmt}3DD<$W9B0bp%uVT==q+(=F^HF%@H>(u9#
z^6Sfg;B5k===g?*WTVNnCpdtDrdH68n3udtRI?J*Ctw-pPNdsK3o{BtjY*`i$J%l`
z^h#zjk98BcH_!U~$6PSjS_Y-tOy8zeiyw<?>>oyINy^Ec7-wq5L8=idM2-=PJ%4&i
zFh^-fK^6y3b6-6j@22Fp_O{k;!{lhg5Uz*Z-rvg57rW<;Y<~VD)iQA?#K!Fh&*qjO
zIdNpWHb2I+XCf%uJG$^!CvkDtTLF?JFcu83T~V%{K90K~`rn{qJoShs<*LH{2M_Fc
zdAVhy{?g!6^y7DEyn~*>Y0k0Kx)AR8_siI|*pEfZCSe3fg?|%9u(uVXpd=6C<{w=~
zh6iW2cI$bv3&NzO+<vqg>jYc}JRDmfj0Xg4bRuxUjTzKf%!^P3c)t&i_vesyG`?-#
zuQw;CdYfp+OzAG-v{%$JeB%=&%E1as4h2t*TD|)?Y{<4>HjJ(N*MBaeiHOD1YwQq9
zsGzT49Vz(hwqID+EC=bvIB7adzN<*9Xb)7^yf~_wnA7>fazd{)g6DBUs?_>+xb~{!
z)_+nX=<zo?T}`EpHCdW^WFbc;ECq0b6R}s?U|m7f<8)?!=trFxrt?PQzUkvEAT|pJ
z1;iyfp!mK$mua(f2tRW*e98&jb!$<H@3Db9_MpKlNZwR;-N#;e7B$nZ*AY9H$qf{p
z@uU`w6ric1-4qp<P9MVHq!e$r@p(2+k0D)L?%*E(uj-rXP#`8WTx{z?&_@l@ex24i
zUH|0Npg6QVGE8`^KdaBh-N%aa(v_v&80fxHxMle^9n3mg6DU2KRzLUp@*xR3t^LjW
zs*i35f+F)7(zkJrJh^FpxPK+(5z*%2VEL<I`rDTakIOpylmo|VNd!0V4<kx<z=GM@
zPbFs41~=319FGGC_hl=7rWe1{TjzOGGrorHMV_<mS5yQ}AJwqM?iht7pX8w`FWtev
z%i2L_uV0nuLMx{!2$zD{iEr9Yhm$hqsMJIOAa#LQ&yb&#^|M##J)!DmRxWYvXDJn@
zOA-T~j2?zU46rU?WZ{p?CM_)iw)4SL?!U2jWDrnDlp&b5(rAt=vpS_cnQPy{PAH2-
z7C6}c-PY3#ui@Gtm~q|<P;9Cf)gWL2<eTbv=YsrKGRpW)`<3)aA@LJ+uisNp(H-93
ztZYCWtjC<`tK<N22m2pK-Q^df%9sIyakI*_!`i@UC&>3vy8@By9@jxwER3|;g^{Z1
z^D*jFyzr`LwVDVonXO<{$I=F`xNiamllVd;(aEq#){p&F_uh1@#T{$rzJ+toiA=<h
z4>h~d)s}jOz%B3>?dBn^wG}orfd=DYYk)HQP9Q$ipBdoiP8HRrM(Is{H^)9OgEj*S
znr8!<nlLkMc&$fAq}9*AP@Z_F_>q6Hv!<<_J3k=x+c-a8Y0KvLdT;(yNxcdsQ0wmf
z!!r+xr=GJt$O7Nk1e$ijRsE^k`tDQs;L1=8n+kTKnIpIixO3dEY`x33LhVQ1Xzfm?
zqlKj3D$7CxK>!LgrZRqF5&?L%F)vD02SFisVOcReDQg*npXyb~SZMnrY2)4ktg%WM
z8Etnbt#^lIK`E(}By<Xp$Vo%&MP#6QIstb|&Ob|x1NQ_P=q|r%{s7(e|3-1rzTDN7
zA^K@7ecP%uty}q7_$qst=%5_)lT`@3>nKtA+!fW*MFItA@gr0clxHNO{Fz<LFjeZ@
z&n*|%CV*{Y_3!<eS)-Pr^s(lfkNbYj2U%VQ&Ir8j*l~Ze_z^gAXusSZK|ZB&(}3g9
z{3C=080>RIMfeay?X_dJn5ttz@l2l<@kflg2o!Sw(>omOaDHJJ$6Z?x<$saBxma7g
zwZLS;yV|fUqTif+-(auB?+@&|9B-V8rcl7$OF4&Vt5qhgc~;ekOIL*S9C~Mg4Gn(h
zYh;)FU6;ZjP4MsfqNk7%-d}|;X}mzt?8TVl&Q80zloheA#l3SC{R-1hUR&07c2||Q
z1pDZL`GcGrIaBS(K=4x!`X)8_!wH7R82x%P2!h4LjGgF>drR~?AvRMT8OmVkfrYOB
z8vW0`DcDt?X3-qw7wZlQKiKm*Zn;~5LB}yOv#6)WnrOk}%n|e)u=mBcFNDG}2A(L=
zN^D%bf{o{@9y)30%pm<agH6z`x%i(PjP<Bfbzx+hq7>c=5!J#1d>eXcrglFHjX}!r
zVNvgO|J<u>?Vph@T@`q{A_vt%rk`?vw<Ae-xkz#0M~jVWK61uDD!4;-h(}uZ-F>X!
zIPlY5@$@jEH?))6H!;U0a@Z^O((Jr4+)>?@4X(rl+1&d*ThvaR=cZWRbQxgK9{%gq
z)6e&(!S;grZ^4`@{({Gwf6%^qnJQ>VDo>K6B)XwR#TQH!B%OFVhi^6+f$g~jUQ@#3
z3luz`?aD4UyjBxsh?G>2TUWH#B^7p1BJj9F&pz5ndbR#F9Fjt%CCZV^eFnG_@=TXT
zDXC4BRLqi;C?wE;KtCJ&+aJ_?Kd9Z)GEf<rVg>kT4D7)h(N^Cv_1>6gV);&2BU``j
ztQ<-4@k+lKMD}aoUMdA_#XQ4{c`sbUI2DO6J&GnlNg~5?iV}KS@R^Hoe$H_*Id%4+
zK)0Scd5IdjB6r*!C^_H!&$ox*llUAV0R!T+K>6(>xM$%Hk`#BGgW|w~Y&V<Smy%Qq
zl~2t(ZDWQ1_{03=B>s}8vS#YdqN+yS{cQ#l1?QpcYz#VB16<Wb`(=((ucWsTq$&9Q
z`<$`qF`zDbd2|&F%hBC!G65l;{%QirPWF@%_g(jAbBB?)w<HAfbj9dq$nbZ#@IMU;
zaP60qF$m9t<}J~(k-iC#Kjfwop}VPnWt_@U{c?)?ull$iQ?2iB0`FAgynLD7!ARnh
z>x^5T3}uFQ{oy{cowGuYk(Ma(a!p`)#hp^phGmP1!BXMsxzNyG(s#Ff!hgYG+7u$`
z<4cyZ+6gYa=Gi60-?!1Jp~R(ni&op3o&&=Zo2~~@Bqes0?8|QhAAhHj+#>^%h!AQR
z+Y3&-Ol5?!FV_+uTh}S1n9u(+Y@(;Av~Om#oyZ~8IUWn>-p1!9r<B%woD;}m9mn;=
z=VUGgM1wcK3+!30ScVXXWf#o#5B(qtir+!#H^!X6p^Vh!U4VizWlw!v>9_J^C=KcN
zV2Lwg4l9T1{9(WB?84CaQyFTlnu<HgN+kw^KqZ+Y#6mX)(bnAKuR72_d3xzdOsoK=
z(VR5W-NK9nKJ~nblKUL}I_<HFc4<WXXOaC`f#2SiK*eaKUD519Cr*)Y#vugc!k;4c
z!R6K?f3HbivKv(<$I57G@gR^`&tNBVC620Yff^NNnMkd$`P$3bwL-Psl%YwjeAc?3
z$R6A5zP8HAaD^6G^MkjfRw5SIW2Hf>XC>T51Lis*DY6N>w&~Kahpn`<oKJUt!cD*L
zJ;a5}9lDcv+}KVucjsYx^axm-@~GGi3zm?T%{IpG!Y8S4r3wJo=^-(9G~kpiYP&S*
zLunYV`DBgYX#NA2mTNqh?#qeN?)P&`<mLqd<pJa5`DW8V>dp4+p0tWO{}hEZ=4#lP
zi^lmJMoP@CJ|cZD1{5dH(p7>O>Bv5=zBD}UON#g7H=B(l<tdZo8`>i61~RSPk1eo5
z(}&z9&FzS+A_cGQ&5-{ANI|#0dyB#tjoC+QnNvBf-s^F{^@ZNF&&t_mhjX9m@#<kq
z*eDRR4g@%7aG*c&+E}J|m1aoQ?A6E$i<(2<yZp38KSNxlNg}WUU4_fC19eeBcuYI~
zwPZj}viP*KH$g|RccSX_EQ?&J3wl%Q7d77QbE~=WK13H14h^S1eCKlmt%zv1%#uf-
zy_L^{0DvtK*WQ9=j5zaVX$mKb3g__Pb6s_roc3l3<!FFRRu?!z)^XqI5&&e8X%YZn
zS%WLOq_y@@g)zOtkn&3VF)aW%37{G;j<?5U9nR*A=2`vq-7=Ogyd|D4wpROWl{0Xw
z?CYnkXgv7&SFboCD!yv)i~wzZ5XUWQy4UMcc9k^Wp>p}65!}%|^p=OV4=*)`ZjV;?
z0sv0s=w>p5GFwj2S@B(JIy}wKI!jD#?U5YCKIIS>B0f*%gr&{jYK-`=cPR`inFu=v
zGgr{pJ-xgCEdc03D$7HhwriP6m8_V8<=240?n5IrnY<Qo&i1sQ(4#2W*%0K1bKgmH
zP7)BFwWcX<2yuhLDmsWuFyI)vY<WXOIQBCZkNC1}8T;3^Lp+|&Up)}dc2rYtg(3Jo
z%6HCP+uRF4r3Tiv=pErRFp`7-Bu-tm;7*@g-8pK@sHB1UqT%l9k6-$`51qA&u~IsL
zgz`2{6VP)AQqF!Qk}X5FKVksJiq3{V0YYjb8`?LLdhIjY`@;Cq-YFdy;>_gKvI-z=
zrvo4l|Bt-;xt^8V4h+2~ku5fu_5ii^hR~m1e%kWycGUY5hS&yHmV~tOXU7~v9x2&z
zIKVcVE*=TfSY=;EKQ8EV-*M*Z)(^GT`tM*^8G(=fPApSsil*}K1<;&!|A=zr9qY*i
z5!CsWlp$qt8jctTqAAyKG8?FpISMN05c0ZtX=0f&J^SxWy|7QMnrghiWDJvK_pDMp
z%8pb}4##OJoEPgt{^ViYn^{Eyc=wwBcS6@-^6i4iRk;a70Cgcxe`CbIeMMVH+N~~T
zm=oYc5?^|J^V6!rwyGr|9_4c-0Tgw4(hCQMUisTc4%(9@7+8H2Clg9Mh7$Lu^DI+o
z$Dvpy@)Q0XB7}mTXuhZqx7PUo=<}$1t0_MsLP0lJu3%j9=STLhx7%013=u{)I5`52
z!mBDuM^P7}(}c)*Yy=?WIEio5ic$b1p!BLBYeJrVwLwpmX!e7L`D;p!UjhI!&Imlg
zMF$2FZ<NxSd}HT@T2m})gPswuTiG1;sk$^Hjj|S2)3j}5CCZzP^(EmstLrxVz3N`o
zED|Z3@kK<C*Tu%}q108@sg87{SkO%PgP90e$G#*1>DpZZkD8uHt)iK{Clc~+-1q5L
zoalLH3Q&~@Pa0b!#~9+T{_*yP=MKcaoX8f>LLlG_p|vjf+(|3izOlHD4GJNdm0p96
z9-1@0?uYXkM`9W)`>~b`lp~*fMT;+07jpk^ZOA)Z&dH66pCL#+FqnAtJzJ8;Ie`4M
zSis6PwCWC+EmqUf`-QEwD+;<`W3c0BUdxP`WoKH|U?SJkQtLl3*&eaEw1TR_SvcTP
zZKOJp<Ff=@5CH9rO8MfR{;P|+XdBCWm{qxw;Pws6+cwWMh;p!YL}p5yT-DL=h)dzg
zN-Z$k4QQ@ZPJ30@tFC&!Ck>?nnRy>Pbty(5w@VpjoGDkHx$vRdpyzS7OQCAK9wI^_
zTU-%M7hhh~g+B#sjBz|_ufo6=^0^lfjSX0U0OHvqC3Bix{Ba~tJUVht5(fZw%8J&l
zs>1IEoY^YY`D>c{<0v+GdROoJvPHd_9EQWWM~vY=)-7(@;;@sY;*aSTfK)nm%P1Mk
z+Up)xI7^rOTd55VLawMowy0nE`-k>#IPudDJ6lB^!z)cJ;P>VhG={dp1S2sK&^s8<
zyz|NJ-Fz-hn_a0&2*3;=#u2yl=3K<s2cqf1D}^;0vOeS;XsZu?LO0}2Ara8@BghT2
zTGVQ}^&dr?kDdgW+g6Z(&GnJ4h~NE7yQ-BcBU%C|=%W4kz5Q=L5I1)m$tmK~2dEZc
zKa$TupWD@m2&~feBO+uAIt(WBrCy*%7FjrTlj!0T7wWC`!5aZkOawgURGB4_P5@c!
z+X*chlK?1zIAhc|6hAYS*Ur}rp;Y2KB7}fPl@Y&dZziv$XAP6kXueuq*!Krrh(pz)
zPXK^;x^VKrf%qBWppUcBWY>8&JOu<(N|2C1s)^?1x`p*WQW)#Taeh-55@M<Rd#-=1
zYe|Hc9lIaZOE{0BD^Qpuf}$>zH4E!QaOA}wWnJirfL6EohBZy!>Q>AYS_DKw;PaAp
zRQNwL&yk3p+?QB8oXnr68PZ)Q^;V*z-=j*m!gmxjBSB^jtusHA&w!jDShHEBkj(2+
ze~Sox)dFXR2(ffw`CuY@{69W=no?0|d8`@|anjN{E$sK)fk0#O%S!quiS>`9TT|)G
zT)vA0Gxv=F02eAS0HpnWv9tCZioG*$wOtK*)mTfd?>%>IjXreFs))(_nU5|cFWI&u
zDs=mI@CA~9^~G}emb$>5bs=w$-GXhj62$z{4XLIJ`h~lDhEL-s{aVgckm6{LJ*_KY
z5Ew)e1Z34zMl%EiylNOqEitF*XUu{kveE*W5Cn}uMH`H!@AbJHV`*_L(J;qOk|Yro
zrox7i>Bcz#@E7+EoKw(@rInGg-C96V_^u60S{^xNQD9cgx7~uqJ_^{nuRq<{7#b=c
zyE57zu%HX33wp=SuAvs>SjwCEH*Q^V6v9DZiM%cVfELyTH+M9I{)G(1SkW=%Ihrms
z<}~r{y@N@&<;#wG8x*Ci(z!g*9}z)Oc#x29VvJ3ju^iLMp8)ONVeJj!oZsUzGqI3}
z%?BYzEfN3#QaOizZSQ?$K{M8vEHGA}0|KvG?OL~}@%c?F>VSx_g6TOTnYPEnIgW9q
zRw$-B8^eQDM@RvHZU~SPKKS6)zIw(<Om+l(41fm=dARYyzJdFQDo6%rBYyzM77f?w
zYdRX|)T~+T6cWG<|LT)}d;Gxa&3gvFma`aPkBi3_HibTU)uzRF{O-Lc+U0Zxlo#mH
zv0&e3W&f;U@QA<!fqe5>9k8Gw{Dezke|D*y#-;(8e`a#p^5^&TUvby(E)PsZE*&n5
zZcm??7TVW{NkLwkz{xe%1kOGbEBKHxJVLn|B|c1`M>Rozwl3_;QE6J`*rRba$I4mi
zq&Zp60wDlnjPBbqu)HYbMLEqFvqI+2AQ-~fb>_<2e@P(^tN-4ViOULfe$i<wWoJ{(
zp8=pS#=2O4>yM_7kEV;qli9)MKQ>yMO<A#<h_%<bL`Os5i$Sj|J01ZciEw{^{59Kq
za_bo~s#NwPIX;`V;Mm9i<CUZkg!c5tF6v3h8IcVuK#^!M)$NVpU5sI$eGF_4u$dRb
zxsPSc+}AswA6S$w=x1dMI>#z9!4RRS=`XzWxaKFFY5*|f%v<U1o7S~aW5oYkA}Ch<
zK}6_}rB{%obpRay4S<K)OxdGIpIfQIDvTD8sS2+fN)=C^<7YE{E@#5$Ryw+eGT%-Y
z^b>?6aJ#tZY>NE;wVPJl;B%>(5E5|$zcE{l(2wr!S)xfGIhuBCB$+o>E~vlF?^gFg
zmF^!Vn<h~c^6kCD@ssb~GF)QDY&_uj`<!)@s`W8LRapN-IulRlFTUr6LqV__K_3Br
zv2#lW02qh}%i1CduS<Q#Of)P>l63k(Hn*CIla`m4Elx(0&+Hjon$8tZ$10p50)R{5
z&{`LaF|(i8j8Dk0=O8<kZAIfhD4)+isB%^u@Aotz)x2i>*BzVtTy`_+SpcX1`OXs&
zydD)2Ic<Myt^fDqF%<v+MP1hJ>5YECw6CmE#?4CX@<!CNdZ`G8lvEfm2~QV-9U4w8
z40$j#WAf@qT3@}e{z*+YUSMWLX3^_r37BFv`usJs*B#IG!;XU^>kCEW+)->-9`U$U
z4h>;n1`%=&2VWWrKNH}~ajYe|Ij_{{ize>(xVSbR(IF+c3Yvb+9h-aIU^dA;mM;ka
z;B%?v0I;&-ocZ@)B3oQJ*U!f|0|W^44yWEen8?0F(*?K`4owl?GYc9*KVpb+Atj*o
zE)~Ufr5?Qh=>&i+W-s5+1ecxM_DFr$cb`k)Vm#a!OXX|RdHqcQ5Z>C8E`dD<@Z$i$
z=Tv?PCTylV5X-iy44W$hvuVn`(^kOg%WFprVcgBFG8RLKA?s(DjYW>xgkFVn$Z8l4
zcwFbEbDCq+b@^&?hKVJub&m@vE9GYP_LN=?@vN4t4SI}m*<k`aIG9||5$fmN-;rHH
zO4!$*F0|AIf8JQ*?<(Jm(d`yO@@Oi*>5k_H&vPggS%Al^rkul$BMLxfD4E?g-VuV7
z0Fq$C-yhreR%PioW?XKInuxFNZ1_ieec(`gLujYZjeCWZbJ|mR{sLf}vDKaRH7-@T
zAeAd}tU~`004SX0I+`Log^*<|kL~Pds?yW>nKga}ez&JsAMzH)V~2?d4-6#NDxB2-
z0y~!F$Ikz9ySSAnH(Fs`Nk#wwGlW!U-F}TSJM_qoVFCd7&3(IHx_==4-h9ze8AAyB
z)jbR910Vd*?_b>G=MuSus{U+-tCIhH0HKh$x})xAE`{}sCw!6!24dMOhGOZ(KDWC}
z1?}P+q`5tf>mEuxL#15Qu|c11NKZ1aefHrUgDSxEX4pFegbLZvJDS4xE^3ZE+1V8C
zMSv|(aSVLS0>ET};GX9WG$gadx0bih#tIfiMq6vV4_eAUCmV~1uzpEXp`phA1dTJ@
zhyb9Vi*Qz_*Et!ed3QLyCI9Se*5TsT@D8`ae&$jbIrqXD;pa7@{h2+zue<xjkpPT#
z$G1H$0nMq*i-1ST`aH@Q#Wxe00rHv=8jNRN^2lv(ZxAzlJiD}Bh4W8X_@~CO@3h6u
z;dggL+(oEriRQ1_W_=#na<Dm**IqwSYdqbM0}Go&cg(eJow2SxUmNn@2jfgdBLEOY
zk#tdBg^2PP7ajrtY^@LHLVj<L*<%T-fm-wBX?Bma4I%8jKjaDrkKYeJyL0Es`v;QW
z%NLDCK!kut9a*=q;ghf5wB)IM-@T9n0GB3^W&an!UNspd1a|!5nr%zlYH#4I`j|_r
zvNn`07@rDy(Cwg$$SUNNR#p96`C(Zo6ysLolX3~Q00R-RL&ND42NU^a|Jac=X3{nz
zB3yRLf)`q90+%eP51zTMvu3Z55}k9Wk68d1asx`jj*(Q}SDL?7;kpn3JPIppJ22R5
z1FX4Hn2k$oa#GUmP6}%5bP)3G?~9%H!oe)Sd>-Ht$ua=&<JyDzlGeHh!hZL*sy_h$
zx*?cm(5rif;xE19#UX*C6~TlAgh$p$_plXkNo#G)=TehY5h>WfRR{^0f_}-)-sEc-
z%zTyZ-u^^xYfrrQ6(_cih#6>)Fn>)?6jIq@cqEbakM{!%0T@DJZ%=gF9NT4F(d^9>
zirRfd)xuayXe0%8bPvDe;XSE;ha5mk=ZmT?M9p}^G3;qgG|Owv^8k{WzG);Up{VJL
zQaSC{shr+UMBsNTa#?%*wQCo(-ucYFVSAjJAaIOW{Xc$LlQ^j^;4(<cAJharUE?Ay
zBJ3JSY`W*UL+9j(+2GMUJH%2)k2#aY#jW+l#z<gbbUp!<HM}MODC%Otp^@y@E?g4~
z&h!Ip?aPZR&+Htz@|zF#u51hG)<cxA$T155AKM*sDGGnPkfe@pG!YtW0=aO&bHwvs
z4_mPr8vFa=FQ_FpYV56bAz?U?tK|$^|F2T_GZjghmDq%rpWOOv*yq00<5G<AKvyzb
z2*)yoD?BQ1vId2h!9-;a_u+)a%?}$wK2&xLOrtD>`MgEeGn{(c4Ug|%`js1=olyXK
z`r<JA#+y&#*6{fGYV@OUJ-#Qd`hD(K6@@HoW91AA7(-~S4Nx+t&DNans#igm5&?{!
zk<=cy%Eo$L3lQo;LLysi^t$-MV?2<R004<pUd<J>5MuTE8JeR3)6xUhtT~inFGgP2
zJMgjoSazw91ggTIHsrm@=TUyVqKOwaEUdL{o!yd~&FAYn_Ei$Q_~a!cjS>IPR6b65
zj5+VQsVAEGct8O!a8n)xH>=;RX6SQv9lxY0=2qG7IgYxZ%9J3@5ZE8focf2mc3%G8
zpFJ|8y0xLr2U18hZhRBB(&VHQiyU<T=p2sitb%!NZ$IxHNu5OID5P?wbqGaW*tcut
zg4(jNVzYhO&)5+1px>=Jj3taQGOEg#?(L6XIIq7m00OO<sB0HB{K2K*j&UPK0!U;E
zuk7lNt-pIq(%DCCP9thz2oRD6>cie%uS+TCC`MTbKtzaV3MXwn82heEQEINeYuD)S
zSE*#Pza|(g50x|yA+UM=c@FlwVt<GTwjCID4aKsTnD!PcSt#3)Q8?i3X{ig*Tm~qr
z<(b;*0ub`KsH|yu1sUj8`RcCz_{GYxAIPFEq?}aV$sX-ZGcE^?t^>pwlaK8gVBfpx
zg{uw>B(BnhbTb4BdR>n-23?<i-RYhEcG{<1=;3fDY+kdhKAJ$#7U6}OfakXQuzy?C
zT|)pvNJ!<18-Dx1zN-{83oyAka=CJ5@~l7il15h!ksgl(Jo=b%ruLtNkS`jI-9yRG
zEous#Atb@wFZ6+K$q*ICVx+^c|KRX_WBlK-B3uk$l$s?XSkYcvKa$M*sbch3Mf9vi
z{GNx~YrO_IDqH62YH0`;f__gzjs>Qr=hbu(+0!$!)|hX<am4Z=i9ia;&Rg5KV^MR>
zxD{al0O_3O&1>=_wSK7E-Jd2KquUb|=E@N*Z*k+rCoRm>hrJVu01-esuV1Wi<(Gbs
z8u{t%+iV%<j-bPtM>^Fwme3p&E?_<lVYY)pdw<FhlBzOaUn~<G7igHLk%h(@-(9CJ
zt3N8oehV6F(teM-$2r=v@)?V|;ooy;q+=jM$GWYcfp`w$*`gC*S6lQ{L)f=-HuT<l
z4#pVbozEU(k8U5jd^lbBvMyu|Ab{Va_MNb-<(qFif8{Qc5*0RPVw41&*H?hD;wLGI
zVAF=qJu1gPcvMz3Z5k0FUo`3t4yP{I(jBiQ3*uYpKnl!pF8+o^4NtpO{>REze^h}4
zb46nb0Q=2#59~SV%fJ5@Fe?vKn%2Z_;>@(ejK3V!<6}s&V*voNib2L406>@QV#Y8$
zu7;VdEsR{D_`F9g&48WLRx6D){<NyFu5mFp5acxjhm!da2XM{1$s>vY0n92XAR=7c
z6#Q-^;M!G+2`cL42mpG9lA8_;Bwsd|$RfbpPB9RMut{OwTI2gt#P9yIOJSnA1u*}j
z=@NQJ(yts!WPj;VmDV5MwF}5>n`A4$bEdQcQ<!+d6>f9r1OPOD{5$(hMTM+kYhCCi
z01$AB-dHe50tgN^g^!Avz-4VUaYf;~$I}i3fPy9<nbm@R54ex<J>V-l>K2MgCa~*5
zB<e%%7iPgGFfiMVARtm&ljrCL{UTQ|Y5@SeDu)x6Huj&huI&yfB$JX*h%wH&dh+7`
z(aSETAtKh|2mPc~ZFffk-YpEP>5|zwLes^o{<VAfWflNJn&Sy7hzL?h*tp`Yv6i~P
zy<xvQ4%LRUTmMfIz+gPPbR?Po>l-duc?Lsf4(-8Gk3S*=Qf0~hbC_ar>3AIkjs*Z{
z56E^_=U~rJ;tV42Po(CYp?GF@f3L$4&S~=WrZo*w&e)3xRiK3#)dLYh$nRd(9W8du
ztML&<)TO_ZBv{(s(6^v5@<WB8a8^}0uc9vf`}&g~UEW?FvKo-jX(hHj_&wmp%$m-+
z&)0>#JIh6w#&kBy6?Hr`l73ZRJo~dH&7t*N;l#|&YwWY!`dBRK4LCsIq{^?%=b+9u
z^Tez`*4`5W02qj8&O?U2O6O4KumoujZ#lSmt{uKL5{{O&c+*nKgFxde2oXU|&~w7R
zzU<;-*;ef6&$u|}uR}y-+|6PDUYC-+>CxT0XJO4H01)Gho<Epb*fp5^+)yG{V;}W5
z*ZOyETGRaZh=*)2Qf8N&(|zUxw$h1Bsfo6bQaN{J-R9;R?_b?2Pgj#zE3HP2rt)vN
z;nDrgoHJqw#%x!8S0@8*!MB~e^ro7C=bx&=$!Y&>$7O?9Aqo0M(u@0|+237v|L#i|
z<4Sje#*DtA(cE@AlI|zR#1XJHYE4*4ngjqi5KT7=DcMAjAmnpFQ`qNJy17Or*=#{O
z1UPA|w)(mK(WUb~a{BU7G5A~x0)qkF(o_?8p}cs}QMpJGV(I+CdtT_dnwbd(bDIZf
zjx%9GN@%JH9Biu(emme%4X2T!3NV0(P}C(1r}CFPyLa%~o1f`>_1<9(8E5hXSqApJ
zNBceI=ldQrGD?JIg0(}&(6VJqr6gsd-w*m+P#^M;=`iLz@t?U6Qt@>Dz_>$6E$tCP
zuHDu(yzH1o;fV13-a)l5nmQdE`Quf7PA+bW>^WgcGfj8Y*95Y_IAh|#NTKom=MH@4
z&~SRAB$M7i0NGp{{Gk!{d`+4?fw&bg%rkux69A==tVke)lyKJig}(`S)qR{{)egYc
z6p@Px63-(tn?TPc0brd1At5kfh<s;L<VRt@s|!n`j7Mi0J7rK3KrB;Oytgm$<KI5G
z@BKnjm<Qeuo2_mlc1s&u){f+zjtK^knd!Qm5I7RBo|wiGISs@l9j8=<Nw=9?FD>?P
z&YZh6BmqEiFrMwHY%5gxU7=_+92|%@&+G9tg;bUSY{?Z49YJ6how{U5<?JV=6aXyw
z+|t_XLWU0vBwyAuoUJQJteOWgYl1ChDe+b}gb`0>{v7eUzUo$);XJk=qA~4_3_&oG
zDxCJr?*1R#`&`%8AKg9}Zf^N6VCIMuax#JQ5^`){K&Z&GK>#S~1Otg2SQU8VUVgU%
zwE;IenU1+qFY8Y|nJva}5;t7XjNm{l9XwVy0THMn;$JwD%!ejUyT*LLtNv(BN9~OJ
z49=MN=hFw3Tb|tW{{C3@^^!<+6eFW<$j;%OoNDFx37GE>JGpL+c2AM4X#Q<QlY4M^
zd)==&!~A#^-Fe-h-l5b-Z+N^bgos2q=choXy=KTi`L|t9HHE$J_q&y7*~?Yfe*gj^
zfK*Oz+uAkqzqdWH>xO?k^vRVcEebNuN`rXN3L1`R_{C8NfC}e8Xm-H1dmtrpVy)Mu
zwoX<zQ4{nXSlQViO6!Po(+AMbt8yU%M-+U#<W9?(=QSM%6Iq;zYP0`~mclCDKpagM
z06?HdFBqq-Zojua?0W#K_pk*aBovF{%)6iOdy8K|u5ey;6x>YD3g!y8dsh+w(26E6
zeB|=u3a>hS(T|#I{J&K>({05^#Hu}bt!n0$y}j`dKCx@?y6f-y#3lexPW$Ym&LhVE
zbG^dsy*id!WK=n>L&g}4WQ`WiFj!6XOZyK1E*CE_&c#vKQwi3H_J)WDJZ?VvY!x0D
z5h0b=pnEtao!R$CIK%XabaCv@BJAx?eMCxXa4J+K$=6!zA5az8wWQfQqb8Mz;HIZ~
z)$ZZ!yN45nj}~;{snBV*w>Ns#?w)8XLk4qAT>M+oe+2d<TT(a&;VI?7p~CRb^&!vx
z2{sXjQn}^8*efj|VB^`6*<ju+))J6F?)lbv_@0Yb-@T%}_A_1=kC{o04*Orl%NL49
zxO*^p{&RbWZv6A3-B;cDTz}nF-+N3ljx%EakCh`}yF6O}01YIvix^|IQ;4*5hG;J^
z9gP{4t+jsixs<Zzj+K{YNI=HIme3w?C^Q?KPo+~OA~?nhm+dN6Q{Z(XDI}b}EOM|e
z<o&zfUD0g0WcU$MdCfhT%)RQL&mLM#RUDTi$7E*bNV^nJRHXzAq>y}ROL+Lw)0Td&
zrOy8cRbfVz`J`j6q9N3S!|BVmcaPlu?SH)Zq3^wQ{gRJg_q5;^HR=Q&#~Lu2O~1hP
zqj1y1WCRZkWmc;SuOF{k(e`X1ujfE(U08SS;~d&w0RZ4rRpiKkXl#z6{XG#t#P4kc
z02dK~W6tUk=}r~rJMPTtDtc?#D}wvC4X^1LN}gm$$(*@W&h>XG9O}cq?_P9#dsK><
z8*Czp=x_h(15L<_qv^sIk~z&&O)9cQqdAeufAHa*37TjBrmS|A9m)+Nk{4}giFGuF
zzwC9Z<7%NH$hWH}`d`l-$~Pi`K&txLv(gn(n9a8&5=fGg06@NK<D%cLXs`LK*QE?k
zdPf8>B*93!u<C{VgTHz4!0->QIIZJU02G`tI#T<8th9lV_IMJma7bnftpE_7R4N!2
z+$x{(9FD1gr80+rP|*jlyf-uvKx4$WLIC(?Kobyg*7l7#j4_7&>M<c7HoH<b7u9Df
zypR$uIC061ez&^0l8JRj|M7Hw<Drr41pwfd(sW#wA+V1G=Hgui0Fck4_Fj7G($BQj
z27c#K*cjY84pWRHsl2~8n*I92FLqzMdO_q30E+kV-SJAf{LkwfM&=$?#K|ZXvr+DB
z<;>9p-%W|NK!o9BZUq9?O+oX}>){;CT&Fq9RpJV_i<Kq}VjT<?G}ZW4F#tbQ6y8m^
z4+JnlM2R-wt4R2tXbnc!Xmq@zYg_m5TQoziqUw`FfrudDb3LFHwOf2{MW4xaZVdYx
zSGNaO?eB|ydN`S@9oL}?sdf)0Urt2p?SOW^*ke*vluE@+Y)J$`pG*4O%ENUb&z)86
zCK3oDNM?)6AA8}z8=vmVRILI=nt7e$3I{iu&&pI@D)tR}mHgG`ulUW<)|xK{Jj!r&
zT^SoJW(&H~8%<xet0(%~@7=QXbG!R<?f=|7Q1TbkAu^O>VRs#l7ytkUW2q*?5R<kV
za4GD?rOkl?kpwB`oUh^lAPBL_Xu&E~Nr0ZBko-v1fDo{tiT0=VrQCPEFgRlz$bH+Q
z7;tlx1myQ{g(HGXQQQFfF>f6TnlZGbCH(VbzPP@iiMlb$WX{}$csl?7KR>kZpI4o|
z<d77SNoMBXl=@H3Vj{LPHNg-9?mCd^J8whBr|*5PdoY$Remqmu6^E>hZNzI`njx`2
zmN_qxDIEWU+jd;Iye;z0{X^RJKi<DpRyyPVn8`C}0{vW3=YM^)+x6tW1mp?=Y62?r
zC=8Q%L*Dtq;EWujF~%UC&ika4-ts+|TEe9v1aJu;t*|hg*J}vKGX)yXaF0eXN0ird
z*ukuY{F>`RiA*t`N$0h|L_S0kEPA0Q?LTfoFku@%3q;cRri-TPl{k#D{S&VknPa{X
zfgax-#TWMtzH~5_eSM**bEq-_ga{xK@FW*Bg|B_vOP2P$QwRjBFX)j1FewQZG=|@O
z&vOS~+aJvwU)8BQb~mpXi=W!r^Y)H<&&N9I+{R2i!++{EnH{_0w2uCt-kJMT*zdkD
znJqFb>kVf9y`YOgG+lUgK^M1g8z^pfU=N(DV|e>Lgdva-<=%DSitB!I`?i7JSoTMm
zg3%-;RfiA`*^vY=lFF^f6tvI%?a|$vRxGId$^9=5+yWPVTjz|SRX8i{U~ICrQ;wm1
zE_FZ86?C5@;*(hn&M?*-b{S5)GZ&5zM+f^sV{IUtDH;XH6+L5&rlT^|kn)5_wv99Z
zK>SD?ArbIxT|;lpYsLzOI4jf#GH3t-Bj9z_MbncdQy=x<fmXG6V0&NYmVspceO&{|
zGo>X-_NX<moX=1qzr3?Cd?|pu*5PcVm47Pc&;YZS9@hYzvBAM;=F45t>~Mc9^ND!6
z&?N0=rgA25bZUrc1TqD^wreo)rr~(zjCi{EgEbv>e^F|`I7ErhgB1UHDMh*=p`go`
zMbr6va=H;E0_csFp|Tl~Qgbw4zIh~>3p2#B!R?dX!9E!fR^=Z6h~ap;-Y}#xi7g|a
zTlv?rW`7odz#3XVcZS!Rl^XE5^PI6sBR(N$L4PdoJ#ImeDn0`ZA*%M89PgCIQNKxG
z_Rh2W!cJz}tlUMwvp#vN{P?R+IysRyzLm;rt(CN&IYA1+<Kjld=lWeoL-?OW1j6*V
zHrptVRBm?mj4aj->1gj}fnG<Bmn~A7A=FGkf6c#k58b{e=(+p6wGHz~?=UZ0&WMOk
zT+(oNGF$w6wxGXe^i&^9K7fcIlh;o8*Y<<wUH$Uob_0N6Xhunn2p#$5b8tyS&H!@n
zd)b=5|LBe#gZZNI<KB_<Dnm#G?LmNmC05=^G9e)3iw0_X?j@OA@wf#Iq4RG3(rbSV
z0NXfY;&AN$Q3rsA5LQX!sCgalD6dK+YQHI~FbuP8FiQ?6?04rl!z=)JCix|wexwYb
zbtQ9J!;AX|FPUA=#gxH+g8?RJXa;49dVWED;4=fU>^-@n?kaaWrXmTFd|!Y3Blm0x
z+<xBbx-J{Y6M&rKL>vhKOb7`Voz$LxXjlA)3!7_R{MU}2kEe3lON+WNJsV)vD9|=)
zx*;)XXiJ{mJMhi=koOfg-hK6VauS~EeAnM%gFk-_pMn$p_aOlhAyd#pyADM+Sw(U%
zTbUnZGCETs_N(a)!%;>I3`0(-_=_=2)dgMB9>A+`K9`z@06^B{0UE<tRWW8lnlfkz
z!7X{7xSG}aypnIT?kkBxse;6S3e29>3T8{NaRYg)2O1I)eD|iU;-jxR>6EU)<fr?i
zndQ}UEde+~stdUP{*v{rpF3emeO?M_HxDT#>Q9rl06?=sEmzP>t({cPq}Q!ruBf9h
z$f3LDVpuc4TMrH<FKVj`J`Mnxc}RD7S7}71lz8g$`oSBXJa|Jor)^5+icQWjoA#f1
zO{kLatxs(2yX*Gn2Y170Ck5`n4RfJ)1OP$`35+v*|HZ2weRy}`if4Bn`fz_V^QOEe
zLbh>3Cv8l2&af#zk_a+Iqi$Dk{9Vy>;j*D*_G4$QU-<kN|McXJ-+%Bd1*S0e?pVkd
zLV!~BE&>1~vKkpePHJz%Ue_t{tPB7k9ECng!gA#_rQ{i$0g&Oii;#q~_ys}|$YUS<
zp_TuG;eZER3ftCP<NIfM{6HlGK$0MyDYozJjlS>N2YaO@ZxR4d&b`bN0+1AvZCV}?
z7aiAl|E4u9Z*H#fUt1IK#H;-V;#f9?0HCM~S8p_P!QQ^a9e;Un@0b4h$@6L$XaA9v
zAeL;RoJIMU6Yy_S&|&GAISRn0$SRCZf1Ue&rgl#;>A3d<Pi;sF=8Br)u!D919*0R5
zu3`{^vac_GK|#|!c4?c{=W^0KrVsso3vq>`-|M<$SI@|G2l^A2R<l<`1Pl=x!rt4?
zKcVyUCoHK?SS@*MPWQ;x8^_LeRv6P4G=y(m(^3Bwze^batb&BWZ1t4Wj5l@-#Mj;P
z;>bLro8Dyx4n**(Qy2Ur<aa;jY_?n}f<uYy3H$mJui}h(5UqgCsy{IJ-0Pkc5~Pr5
z7-H{7E?N8G)*AmOn<D;QrQU`m7lvcfQAQ08@u6g{wQDH-r^mMS{OV;Vv~S=_o1u}I
z)PWx}3FG$QYyg0Gwg~#vXiSOay>RAa1D!1&$Rui}V&vHj!znrQUsU=3j&%fRHV$XV
z0Dvw#r6XD&_I|TA<Q*aBphX%r*1C|czG&v+6)pY^iM$YmfWSb)+|CHDRA2xA>W$`x
z-uANf*DP<Zd2dtL`;^LAS!gjC9`71V!mgg^`|f<U`$v!LiXGP#GvHVE?x{LR{NMB%
zoso*OA^%4v_ehTJ{B*cgBUL(?&yg*wkVBj-bjcUA#-Ui+LpGf?3f{zoofyUzJ7Mog
z>WtsqyW_)|g5hSI(QhBvM~pLK9O3zc8F=PE3a)>m2Y!C{F8H6{Jq1Jrw>{g-t~jN2
z?T>EX@ts2>>1+FAnbl_M#MoPUT$~~S*Dp_A-u$7Hm(=c&RzF6&@|i0$J8gq8G#0ni
z+;rNiwrjq9%Z|@1Yz%#0;Vf$lg>!ik0b=QVLn^Po6#xR?y>)xZ7aohPj*elQV1sXM
z#e&+OdtE#_p6u#ELOfIWKrESRWur~o1ri8F%ncr@V73VW<aulA0RW7Tyz!SmUAw5^
z9c^{~-}pTW!C};iCo%=t+BLj!$D!D-e*e(pmkYtv@7=ns#FkdG|H`qlfz!ECdF|PV
zk7eN~%s91k4don<mSKp{(G-5JKbpCws2jyuV+9alX=~)veSL|qNFwjC$^fEC<4YkW
z2r1#wUGXh_(adiZjvp=Mf>bC_UNe?Gy`$%fHA{}$f+WqftI?^Ua*k!*>~3!!r@=LU
z+L8Ogbr1gax1YY?nU;EZ<L*Pz_Z2kJC`nFg37A<a1_e#L{(%>|m#<n-|JAV1eG5PP
zCx&?FefE(sn`@9UQycAd@dpsu=krCwAeg;2MYRFX8M=^PPvtZ=wu*F$dpmk^Ght?t
z-M`WTj8w%z9CMc;Bo;Np$-p?L?oLt&f)$@d{-I?4Q`g<UXZ3=H@V5!zz{9&@+NKq?
z+HniaNuaSz!Cfrs3TKG_c&hK*-jVFPdUy7BYKB}<)CAf`V$K~v1`zbQQ%w>7&%!?T
zwQNC;OCb?(_SP8zpuIk{Yh_2>XU|xE<$g{3Cp_Do{>?xl_x9mrepxkyN(69lDD|4=
z8sG0f@~yibE&I38abNS-6l|#!63$-T^w4l3f78~k;j1MPvl&-s>Omq?Xv`Glhj#X*
zKf0{hZ;+J%ibuyLlGEoB5q|Zi?T>C+)w+FGZ|bo_BgqfP(uLK+c`7juC`3fibOE{{
zPuS8m{9_?$eRGZfyT7=1cN)r%xl4ZTr~^QS^UBhH;mnhm0)si{ZD1;te)1z=nn4Q#
zU@UA3|M{bru6ghSKYwh--a|7a{QSRu4R8F;o8b>%z7v+d=6HBpjju}*eHH+YRRBoV
zm^gFf4iVe5vNqcnP2ZI(8n2C}@(Zw>T>$`qf+p0WAz%5_*1lWz4;P+!{{;(ea71$@
z^L7wX8N7V&4J+a6S1tvw%e8MXOJBSFNqVq9p8MfQGPhhaq}xsnE-NW?R1Mp?A;XE>
zNm2~|rm4pN>AQaMnx8Vxq!gvmivLz_g^0kVaHtD-`ybsg@JK;7;GQpj8{YNfFMvy#
z`JnxTWi7a)XJjaqD>jc}8mt;jRAnQkuiG<HkZ^cl&M`{c8kaaJ=|F_sWZmrm#+ko2
zVNh9Z!HT-{cMT@5>W`)`YK#OjyAQ>l?~i7lAR<WwN~WNno6Tz-nh@a!U+nTBV9=1#
zPQSB2r5O+~L~y&5jL*%VKWSy#H=DxV`%YQ%p{y{j0|X#8TSNf?1bu3zKIr=7h3jf}
z3voLG0=Bf?HN3wi@}*=>yRN8<sy8oaMosrn>Z|*2c+dHZ8$MDj`L)r|VZKrHI82&l
zL%AV@00LIySBol#AB24BrHQQ8RI1)DPoLuZ`(iI&*cklj%GTgcVV*W}Ik!2}Rf7;I
zB>eYfE0HsnIAdkQ?{9hP;N5!%Qr{U&<j%<#^t$ns(=0f`l8)sHdh7P?k^gC}3#_S&
zc;Cl3iwP;wo=<-)F@Rt@a2rT*26I~mcVxRbdEXtZHtywcWd)r~j|KqhkKTRejP+<S
z3;{oR_Zf%)@Y@fbjf^w+pWi)o>@!1Ds<;Fo5JDHMF6_TI<V*cIlh;06)TIlnurAS5
zZq1%U(bpZfq-7`PEF-0yb5vBR{0cV>n#wsdgdonqc<W1+J^WNx`r>DI_P%Q%mVH|y
zTU;Qd6Vr|_deL+~2nb&%AbECE(@$JAh-6ZU|30xjIsiNZ-SU~2f{>EK=Ps5=K(oP?
zUGPjFqBAX&a7w7o;gCv97}}iExLB%GB+3Ae+EfSzW;zX<6u=UIsw(M_-(%SNXDyLk
zNqN}d9s$7+5)C11yM~grIFwuf2(NHdV6#%3L(cfc<P=E~D4apSs}>c8k1c7fy>(Gb
z&2P?HQIiG$DU9pPB9~cYRtW%VL;f2#ENgoLfTNjB14ezsk1iL#uerwmMAu;Q%xWN5
z5<!0~v+UX3Ll+a#KdkCxdyLrp8Svzn0P`dWiHKc%%95^sJht~&*+TSxG(%JkJTyag
z?m85I-v#U6|3M-D$}9$O3bp>TWBV&YK@?H~M--Qy(s5|dVD>G~>>0UyTi4JvIn6jh
z)5W-r+MQRnsJr`Onb#1J@%sn&f8<T)EFKn8N~ew<D>;C+@k0QBx_}2<oc&wS6dWfo
z@fGu3r(a}V#{Lq)tT}O%SZe72|2`OebczfB$P3ps&?zh1|KRtiU2Gf)7m_fU)85iM
zoLFvs%A5`gpO7dqbIF(#lBAI6@wj^5d(q0ztzOvh-u8yjT`ET!+oqVR1{u**-a8b}
ze)GC#cfB7Gi7WiyWM8SA<a8K1t7w5{>!{Qgf0!I&k73hFmE~Bfq@T381eM?EsM%yW
z#o!>|bq_Svg>n@#%dBR)hh^33L%wKj(DS{{rtmFo^}%S+s{#Qa5ikN^l7N|68c<f0
zMnu3ELd55S&Zh7SO*OtBbT)=Q`07n--^dt#@2nM(q>z%VzG!Ig&#Wme0Q8Nd9%YP0
ziOgJTu5b`i;;T>V90+*Sp9a0I31o+w_VmTC2J^JVw!ibt-$bZyG?$GbnWJPSk!%XE
zOl?s8TW!#@ZH)3-B2?6c(jQH~>^Bd7?R<u1&7O|-WmnXblq7_ZxTL|WNuoclSy=z}
zRUP$@)doFd0EayZ2P#?#X7f4@#ImoCX7b;?{)w)-r}ibw-cmV6HnFVPjo+)XOwj;1
zM*EpLk4?r+oE}5f&VX!@O_$Dm4wq%QrBT)G9=imfy|5M&A-PK-0D#B$<+m(uiTvV)
z{li~0g!GQt6GVuo3-!6W;Cr6hm-xuZi^BzrcBW%Nt}IrZApx#wQ~>~Fe=K|3#?@`x
zp4=|qn#^jSN@R<!vIvkIc~*{hjb#d6K>Yk)9y>TpM1K`hf-nUA8#juc)tq0lz^Bd5
zP1TMO0Fw;J0;q?p@Qeru=f0Z(tFXT~rS%!rKRW=M{BXP?YXKP=%iC*z`TDaKe!e#;
z8y?$!Xvu;8SWs1z(1M2Gahf65NTLoPFiOdT2%s{)UrPC6Uo^F)#_!RNU)I`v?wYz?
zKe=OPaU_+8*Pqb|R{k#nfVB2xmfIo`6b)%?Sl|Og2nM4wU{Ml@fY)`e->Y7qD;lpg
zeTJ3HVKJb(At0UCPW$Zx`(AnDlih#1>A%nUce2Arm4kuNoMwguQcB5RbNZrf-@SS3
zO&*o4Dj0%cl@yL_(OA+S&0O9SEj-#5@f44dwPuc%cx1JGrK?WwK*re%yN9%QJ@R7L
zNve|lf6-LF&CU%R9o1NF{}BN)MGg96^!k9uJ<OZfrvRWR3?aF~j|m2lS&Aa<b_8sS
zRu_-%h9LkzG@Y$AgpAPmq@A^^FcS8;3e!h^)`psM+h-i8SmlqLnUNF%05fz+$GG_+
zi+w;VS}r}Uq7On#jsH*eK~MjfJ)u$0)jg27Y#^3if{2o^e9ovL1kl{K!m%B)o!baV
z=8?b$E?&XEe#_3T^EY&SZN-ARx7LTfFO~rY90h`v3{xzfuie`h`{+GehBo}<u6<=c
zVHl$Hb!=h#5>_w!e%68OXy>5AWp16!IRXIqye`2Qn^2Ds0CYhMcMlhp+3RjN;oI81
z|5PGklZiY6>I!3}9<28J;zVy*vRGxzt_=EI`v5?1kMN#v-MZt!ef_Chzxeu>|M4B?
zFZ=unOPb%hxTWUeC2cjYSkhYi@|6qfE?M4I`-+XL+ur%!i&lK?FOTf~dorip8qXA7
z?2Z+S@4I*<Y+4y+0HD~_*hs*c1m>vFo`n@0Ei|YC+9VO7kaB2IbIsp_K3CMq8QAln
ztx;d~y(8%hK5@mmkd%_FWA*uKa@SP~L<oTqfJ|wiAHTTv=U%t^*mzAkG($i^6K}kK
zOaIwl`TJI?491T}6}Dr~?Et-$3o3<V%Nt!g-*Lg`ez&+Ka_OSx$Q_(v5piss;NZ#u
z0U=i~U~gaS&D{gZE15Y%7)#*msG4~b_z@uhlXi1uZ$?#^n6&f0q=vl%S?(B0KIa}P
znJ&0AU2x_YJYMZb8N*#CEv-o&sWUVs0@*Ao699k#0HR<BI>twp+YZ;{Mje{|kyjiC
zr!A`;>THU9%cCmg`r&A(T+odA?&0hwz+}Va?(zl4qh-)6Lm<GM#b;LUBFLLRaw^-f
zFj9QurbX9Zczowo4I$6Xs=~$u3YA<A5x{UFd*bFj{a3fv`hthH9|9s1>>?hucbzom
z`FFd&17t`5fcl8PPhqS$8KNrc#?nJWY4vElsr}3&scfAtL<lEI0#p>X<+#OlX&7y!
zTPj?c5M*H$TW03DkuU^2@udskr|&%zi2B%;L%G~YDnFdbYlkv9tv8)3_V&keJtL{Y
zp}}}Iy5&%|@cj2)17CRk32@oTEwHf8V<UD+INDW_5R3}jHalhr%VfYgA%y?{-*WC!
zICEvwEnXLYKpm%WNl1c%F5bB9VEk-FRqTAw|Ge4WGw%bO>m&ex0$jqm>)U!dn!*ow
zTuQz=)rJsD<-@Ub;bj{Z)%Y1_rGOZ^kD5&s&Y+C26i_lS`^LgIZv0<(!&!@-zu?5q
ztCzGyzUx+ba%|)4vDhJ7&}&}UKls7BUl>@)xY?ENs0){5fRdScPj-!BFyJ0RL~SxF
zT-3!$dj_-KqvWim)7Mhfs^!r$`9jecR5<2mz%LOH%@*37vtEib_`CwD_iw|n-z~f@
zep^Gt`|y}lEPI-UB-S&OJom4Uc3sLDgJq-ns0;x3QF^Cp?_+;oYI(D9dEL&7PhRxF
zhOp=NRjXlXjDx{UUVmFEn?J2{{r7G0Wi<NE$07lEbncDQ-a8XeS!*9~yM_TEhmN45
zDuG1i1pqXctb_9?J(Vt?VMv^egd{1WO#xN_C^c=CmGj(>1ZN9La&@Z*E;*?Mt~|X1
zUVC~6TzPs2y#CAuaOK8!xNv<VtX~j-fL8@XR;`bpsIX<T+tU;$URMB+hzLy)S7BvG
z!=F{os&;)g`vMk}M5ge@zdW+PhMP>yd>-KBewe_@xpXe4Fa&N|)B1zDkoWoVP8A6N
z;_1S>b;2{s0{_mVYZEKpr7Xw&n#)##kP=rldj~)8^3`8n&=B~n&#ffO4N8@FC4!<M
zH*7l?z1sF|$5M2`08Hw&F6iA$1j7?f7P890ye>ZH@1Sr#3hZ=)1j*`;N?1CFABbo6
z0YLs()grF&vah4rTIX<|*>22T=?J8d>{S~(4%Gy`*M<Y0qNECIYk@*t$d<kRiPsi%
z5dgCrww%=?qY31y+L!YkoyToLN_O&+x}Ftnk!$LL?%%kTiN=SSM*;C{!I#JsKlZ1G
zKTr$67??RH43EOu<A2!Ybk-aYIFW?5y1+gFiaW@@ssW;q6d;ny5y8)CTvfp)jsHC&
z!f+zD8vyb`NM@y8Fv24Xmbk(}nLwAF`f<2A<g{yZ?Yy1O>~$${^5V#CEwzEC8QRvR
zylo^RKr~%=HB*%Hg_LZr``i9IU6nqonvDGK4LhQpjiK);9CKw+n&WfgnS!fpAaTtf
zAKdG41ZX8zdoI&}DxZHG88+EK0xpEifAHmNf7BfDej?yevJ=NC2)Uw;iA>>&KRmSm
zOnAdTkDXC);7Ai*))L+$q#T*Jfht0ws4wV`rj-eT$=S8|L=a6EF<%tb!GiVt9+gAT
zry8z#7dzK-MF1e1YW=r+RemGqmB;kve&~jPf^J-R{o`G)<eU*=={T8=ieOEVX#xOY
z2*Iy7ZBcAhd*l;!A<rLFK2GX{jT{M~KbG0#ar4zemMRU!Jkv0v5xdQZVPnL#yP%7a
z>TJ>I(P1)^Z|NRP91U;J?9_>9IuFIslmLg|Q;Aq`DGYSOh=G}!Kowu}25yq;wilAc
z);j;UJSs0Z1;VMa*@A{${qc|8zPaB&zhXz`1KxU82Sog?d)n(mo3R3LIFKOr$1<nZ
zg#71N7@Q*=)9rWclLldn&wpLA_E*idzK;dGYJsXTfsX8;R8Ctol+2wqeC<osV*vn4
z%Rg~2X-Vc*@qlKC*ksluThRQ`WR6uct#c({M4-qCTBi~bc-^XKjs%Q(eVjuLv2t&u
zkQmPt6HD7_Zw+|VSb1Wc`DG1ej||4LuXu1vUp;4xth`i$V~81Zur<znM<JzzS8iN9
zeBp@;KUEWO-@{l1n4{6kRzQHFA=G%L__l352@cjcJ;zY=`RfS4lm)1)Vef>kVb`$2
zAkT2TKo;yEM9md6Yjfw+mPsVwIZ2Qz=u$W2_>n9C_}r?|ToVw^N;D*3gywtC7cESD
zy?jFrG=_cK7B)q8S2|~>R178aD>Oq~zNIHSzi88S0R-}*b@iaSd<m7~=Yw7~L)EA!
z0MHB>e(}K2hjp_Jk72cTD`Apj#r`{nX4pPJ3MolNCa>DG><{gA!SA_Mo+lzeh83>P
zN_#Q{NP<^Bx?^zRG4WzZCYmJ#$TN_QK}9r0d{UX@bXU^_7>1~l@;+zC)zebzUqsHz
zwknrfReBb+)E=DI$+;FQvo*k*&gjHTPHMj?>~r6)a5O4?8xet&0MS(bHG2j!=Kug>
zoY~b^$c~-!aOY%zQeKml7>Fk>s_8p*)q=0p1ib^5zvX-dBI0}dVyDK^SryIV7iJlh
z`KGf+7<5=hxvYA#xh7EHRT@&3##0eMI;TOlpwD^h4@U*PyT<Qnq-xUU7*p)?xcZkZ
zXt3prTY3h^IAe0mTy3=;JZVYouuFw&!ai4mD%<0foV^U~I2e6LP0(92ziQLcMU>36
zYv-*B<Hog}FVuzoe^<td#0()Zo++&R(*t{7{;#eKklEL=1h()P=@AQ~MK7h46C~6A
zH-<dfvp019thpxen8JA(0cgI&XrahZBC~$$fq}-O4gj6!-AEvP)<EW#x?mS)I3a76
z%xN&3Dy7xQxvB&fh5$w35Y17*?NU1?tyUId$I3QG#5aG<3HpQ4Dq-e#^TN7CO+Rp}
zd}P#}jrRLM1jrT*Ol6Dj|KkIDBG&Io0FZK;Ng&QP0yETB0BxPIzZ|)WD20T+So)E+
z`r!A+09~hXB!HqO!;0!%59T0rva0dtD-LHQE-b4~r<VHAHbm5`?Ss9m1w*3W>vF-|
zXB;4yy~cv|Vc+t}h8;_y-OE}-2W|ZzM;}0>`80y5thdk2<f@LQ=R!XBpA?Q`0XF~`
zk|0yi*WUH?-dFdhspNNrF^{tk!~Sys;F1Uw&PHm3?(2eH*ZyiEj)+jujFyAL>9-+b
zyD96X;tN4DpqKXk9qB$+=D6A)FQkOEt-kKoTK{)FDnD42I<T&MQScQD8m~EK7(hqP
z6#@XZ)de3QfXrlTamvcp<wxtepWh#?F@#(-@n$hF4AIqD?;4ovo<j3M;An-g&;=yY
z+06@^!gnxcj`^xa?Dfa9rz>vvr8)p`F*DA12KRkD2V05;Lq4}`t@Zu2r8cl<q7UMn
z)ej_c7cj<%%+`S%=9|tQvCKhvjyVAUHH193NJ*&)_k*EpS$qA7k}A2N*$1o&=kWNB
z;d&uy^+Z3&5Fwi@ZeQByIb{DX0|3S9S5_I&&tK!~ur5S=VN2x3u;0^N6_5ZxzG&1A
zByyKza$2)Nqg^}Zf#4MFKY>9eXGUvV-R!OBF8x=?=X%hkGO8Zj0OSh#*^h7OI%~(U
zL5SroeU23PIe_Vr+Ltko&iZUdB=HSrFTQ<2W9R{evvGw45inmgt~lxd&=A6I|3cPG
zS0X7fmdf2XnbAZ9g)z1~p4IKP9dk<4_6{W@MNMBp6YQ2pRb+E*AP@6N_D91R+lqn6
zkX^NDvGjY?@6-m|Bc*??nD(IQLfO_e^ntry7>Za&2-AN6B2@U9@ZO(33d-^?f^`fy
z?a!Q`3bMo{U4CK<)P;kGJTCryx5~$@V3SgM2BPUT_U~HlqvsjGBe)!UQvySX2t7mb
zZK{Grhg?*Hcxq9X$0c$GpUVj?_PvH<xv*}C&hcd#z~@nAb8R36W&ud0JkAFXhan?p
zEw6!-SG7OtR`4#D!m9cPAqg`1;stj-v-f;HhjVKG!+d}Zf@sH(C>+0}`L}+zdT=~}
z%@lN>LGZqV1JRCqwvPaqDKgW-6+{HLKGO>y`qks`fnPojj5Dw>0183%z7C(R?2H#+
z+4+E00Ld5B`@iK;Sz;o69*Sp<sVZP*O+FwZ03?}ftMkX$giDXE;bi`z?%}LD*Im<z
z2o4P=St?r`zmkRk7{kFOZ8eWbQED4Ff6bv|vz5cn&3e^o9ebN<0zXzAisT$HPXL%J
z8Y@QPnRju;cSN9F;T(T<=dV@;z3R)p{FO^ydiyh7K8^)B3ruj?1d7v^gyf>8$mY6`
zckg&Zs-+~zWDEBEV>@`7rxj+v!H;R>z|=x(U2qVs)*o0UqX-ezXexKy!`lbxXq*%b
zM3df3t{ACmqAdO3j18>ptbe*1b3T`CR{m2!;GziAj$hXH7mthgj#+RkFiq!mPeBu}
zyZ^=BM#dRAngZiI3pla;=L)wAsu%#!szwgyu5Wy7K||<vg)<Gu@!12>^u|L&sf#2s
zj))S#Y?C#e=WwaK&F4}ss`0yC{=HjYY_}Ra&&I|vLIBB3&>Qrs2iqG$+p3M0v`K+n
zK|6LLek+NX>=!w0b*s_d5Wa=L>f}z<ac?4%TePb;W*>pkTnQHd05S}v^ZF=-RB?C4
z0ECbSPFmXV0GK&4^Vb|*4&zuV5|ToK-=h{h3f>&{ySJH<Mj3_z002!F{sRN?*Wd9%
z_wh&a4EJvxT97S@AL&BgmM$1K3FbPth`!lis!QqmgusaiFFC1gOHng!uHFC@K}ZSF
zbaAwufIV$w{+jV?(}sn`rkcR-0jf7}w+6Z6U@YC~;gZjFph!fJ&TDG6sN*=|0t^v|
z<o>lCwU5s4=#1DVltSXg>uVlt4ErAB48`aHUpd?uPUbG|i51QS0A`&cNSG%Rrg}*$
z#~Bb2H~=ke4Szl4bH%F$ItXA0;?Z>BBR(eE&@2U!({bHfdy=l9MDCNikhkYF@lVc_
zw{ymnN`hjh#0mhA3qUC(J9lk!)~)idG8~%=ZYy~zcWeMa(pY)U3;<wzjax4kwL9Dj
zH)tI0OcKFBEX#Hrj5{6uoLs|2a<w;vH<sPAeAEOeoIz`SaIoIbQvl<`*!gSLE@sWr
zwdYKekP_Z{-tukrA@B8mkBg{ERs{h-wxF-s-ZgaD?!FY8ZVmA6L$UU3UO!7pf`TSp
z>70Jaj-J$P=$y18KO`VB`w|trZe_^lR;nHYO9_zIg>4Mx2hY#q)e&JLmdZZjbt&UU
z{z4K&(*@R6>-Wqx_>)BLmb%c|>MdZ*OH~-uhkc2xCZh8@Ix~Uke{qn2DqGZ2`vbSi
z<2a57%N7kBPUNon!-IQ+)=$h+t%u<QD%M^iQGa`V@LGkVs1`;O01PK`OLq4pK6dAG
z1JlM7TMk5(!C2-ch9uN<0TS?nhj)%N&vc%w<MX6_(ggrwD3<LBc+@iU@APFwL&7mp
z1?)B}e^KY74Q|04?b6>=;~yyV%w@(4Ku8F_?3Bf;VVnfXnNr$Hy8Zdx!@j&`oQva4
znl}G;VtHH5Z)Wq^{?o4`temx5xm1o;Gfr4s|0kEio*2*Q=8FbT=k)hJzH@j}z@tv9
zGh42pF+&K|1X;v;h7u?2JTT0F04v{Vx>dqfB#BIdf&_{!YlsMoTWd$WZZ$p5(v%Ia
zC!7a>vx%~nw$==`)CQib+AkgvhyWO3O~zpB=Xf93qdTJkA>k4jcU%?$BBqzL)ZVtF
z-Zj6UM-E630N}?06hNYYM}2WobIq-Gia?d!$ANhE_<-Mg5tNULoWG`^ebyHT0FXk$
zx$By>&Zf{GT55wk#w&W)g;ctSk}vnV`P$oG9GRB??vG`tpy@8Vn`|<jKk@lJgSAdX
zHUpg@OaUP)*PTY&d9CVfEYyX3-J^}%X$%=~tQ`4B03bz{vv@B%p?%2fQhv!C{irHm
zK;hWfHI#nEY#B|%qU(`NAZQ2yL<IlZc1Rf*Nv|32rU(EbuPZtjPd`K?aUKC4ra%Lb
zi!+=7<rhw{WcN(kxUykyTV3!$kID;GvLpZzA(}2U#xwcX_`IrbjKQ74W~K`o6m>x+
z<3<oo<^79WBR*!%_$*U?ET0H9o%eJ)1q1+Stqbi(#6es6%hs16*i#f8Lqg`SS#tUt
zS2PU(z~4%J$SU^(0T^S=d;8)S%|<2d1dc{--ZS8erg9s`M=S_#l@BK}`9ITqtLGW<
zb=IE-0Lb&!HRwW!9|nDDV!U6{4e8n5HT<DH!+OxRKK}=DE{TYa2p~yQ)9kTs%mRTJ
z@Ok%nT<nKlHy2dRngT#xHx@p<yZ`g;wca!R9<}~({dXc;fJ|O5+4n@Y;K~*CKsBal
zhAO+6!{6bm&bqY6t!^8=Q90Y8a(2w+{18}e38awlotwAi0qCi)&z-UzSY?umE~IyV
zUu=D}fV0(t00Ds6O(pH`7q`{6#j{0qg5Q!IP2qp8=%_Cc9Nodv|E|g(aC1bf+3ex|
za;0&s_?`(N@s#EDUyt}b&yVfVi=~KZAf9>0fq~RJ3z|@!2+ghXNuOyXRfKHOaQJ(y
z%^P%v#Tu97$<7IaBRkrWP{DLIMSR0jQUU-g8$pA@88k+`w$jbZ=WKd>A{p{BJ>+$5
zsR?*u<S0FN0&~p}u6@1Hh5LuJ*|K9*_c-i#*F{q~W!zsxl#5#;54Sf23iEO@^Vgj_
z3v~Lrh5PG6{y(yDc1aLnGFM#w^o~QXIhc_Dx!}){myI)Em#kkFP(G_0PhT3AOIm6l
z33=U*jbr{n(*;ZB^vn15#vk%~)D4H*4mOd~A*&fBu0R)rZ4Ke`_eF~?2e8G-^z$$-
z;UE)eMYAUlK)of~UqPVN2HeL60JJN1Dy!$dbk%We{nD1_T`IqAe1=X!BCCZEp>{T+
zW~W#X0z(2EP8Ah}^VczkJ^-ppt@>T6Ar0}zjZ14aqH<<t{+c)jAR;VkMpH6ZWI)p|
z?x{8e-m1a^0Dzi+FT1EE@-xm@-pRx`@0KkXu4GpGXgaTXOTi%mc-@C*fvt8@^?Q5c
z7YRwucd~&W!ZeR{Du<$%ij?gChp9UN0Du5VL_t&vu*&KqL|};EajBwGZs#itXD6x!
z061}F`({;TcQa?&umhOuhJ=DH8xXN+ww}eAY9OV6=MQFBz~{L{NN|t$J3hCfc~t(h
z^VT*MXQD6K|K+PvPZ1ehyvhrVVKU%VZ>R})d&d9`3lkMIQMZ2}@oL7fAtU}nw*k?v
zfJT~uTt`LWpp5kcQVPisaYu8F|F?dRniwxNBbh1SlRJ7@JYB%UO$H|_Z}yG=ES}C^
z{@m_CAB=WiAf$0I#}UXCLbcM=(lihNaJ!UazUq&3Lkx#~uIB<Cm#CV|P6Te1uY7Uu
z@Tt=r*f3cLfq<n+2}lTR>%kc8>yK`dM5@iMR5FuIwf>vec0{^pw9H`s6nGLRNj5?Q
zfK;~7kVxkk0MINjyAlv^QKOe4e)XSi^}%gUrFFt|L@bNqY=D?6>SD6v33QZJAR=gn
zP`U<_t=$R90D!E*4ZHs(zzpa0lng+{sT~3UV@$6P`wx!VrXy}rI)6<RnoEi4ydJF$
zc%JrqRHM47B5=Eulb_kwf9@>UGF!!%!}-9Lz5V>)K;jYraM|Xm(#^Hh1s|^q`S-(o
zmH4b(8`{0F>G(y@s2p!rctxXPEXfe$ipCkYJ-Pd0Kl=|Q6D)UWwaOGQ!~IF^2_b~Q
zUb?;!PFcO+W|zXAVmMBp5dg>)4ba8RnCAXix~+RKp~7hEJNr-DAIva(83%MAKqTlr
z_5t8ID{A4a)h&N^sr+#^PD)s=s4sYSPu~eJJMA^>ZQ&Lm5Fx*BaV%YElu}mQ&=g#s
zThR<5{&L#NrV&ZR70yeM$ow@SSh2EM0SqarKb{jaebnJdmIWy!G(<vij_{d)N6nYg
z49e^gReW#81a3u{RMzaSoc0G8f`B1g)fY{ls&Ecm09xbM3_A~RCc$nhuenXfUhUar
zw4*6>uZ6g!J@<It$<La2^NmXaxM@wx^$KTCk1=IdkdVz6Yj$=IFPv#x<~Ut(2BU06
zXMJm5Jkuy0k&kqMPoG;Uk`zBVbxl*0=CIY)|HUhw%4gRtZu&>S<2qOt0MKYKn$ByX
zbV0v-b62csRtS%yO#_tX)dG@q1ne^l0HqBCFZlY?(GAO*f9-WC$w>qG>0sG%bN~fi
zdR8oGc&TPcL;_(|?Sf<_Uoc~OKa?tRBJkljp<qjGpzoM9_FUn>EN6K9;@aMYjiCo!
zoM*>w4nn>tm@Z($=B{M(OiN^GoQ#!2zz}Im*T`!Tu#KoPY%piaM*N-!Hm>b_R+tDt
z*u^9B*W_Ljt(g=E0AhBIIc&B|0)X7zo0J>Zb#AT;`~G25{8S-dW55#-qzjsy3^0`k
zET+~OLM+>MVDQAfBU<S-olbcs<z#Pfd>tYzuIjUiQk;yZB~htu03nPO>G^9$WX>+~
zS=}D)?Q9C)uPQuWCK{Hg6m<zMRXOvS{qeS0d=^9kj>zQRzT{PmvF0(YA}ryo33%>4
zWo7%nY7|LQN}O-@JbTw^i$k!gv+2H&PkmTbxNri8GN70)T>A8`;nV-qf;H#i3lRua
zG3_g#!E+H^9Zds506FlzGZ_-y+!XOWWi@@Cv8y)dvfKwb-!5=qFn-m6{usgN{A?Te
z&1iu2)V_GFF68M{#RGDVw6{0<?xP_<XtE(<>OW6}XRdAeVaV?~gyXdD)(r8|CwBC$
z?#r0@c+StCZmG4xOL+s*c8~%fcwiuYVo?`ftnlwd0Dg}e5BXhxcB#A@Mk(IIDr7hs
z-2RacsXo%*$I5T0N(eP`cBa%X>5ZEffU59ezf1XR$mfnt7FFA5$x)yzrxpR4KcC5K
zN-SMi?^1Y8$>@y|k8CFQFPhBO7()8X3l0bvBDFUJ4;+eSGZw*FHeq1i6;J(|qHU3c
zAQ^xpDRJY<<{#7qJbMwx3O^A*I;Wlg(3ak1+xv@Uwrsksf~q_V5`qK(@7~_%8M-0-
zSfPM~00FNn8TPwxbScUqLDL_JHGfV0T2hO&W>M46T?!kqb+$4H%@lN=E$Hw0>%)7)
zwtUP~;CMLpTgaKU&sdfl3!#Cy<^li`IP*uqNI-_!c1~V?%7WsemYQ$--HJYw@8eb#
zByxHU07FRVAIWqW%)6ju{&944<dL2L*|I=HaA+vuPUnlwSj8{ZhkgA4pX<<E1Hf3V
z-(!!x!HEEaFa#7xfZ#w<5{29>v8Xxx6P1r9>^oVSM5YiVg7XYR`YlJs90ENp=US;)
zkMEAtUmrd2GD)yeNI3=>s2m~Wb3J{*$;<wE%EF+NOq#N$OqM8!vulEx{oYeOUht@#
zIevH25k-?ZvC6y386JFIZDiGn9ey}-{lZ5=KG!`ih0|pF<NV7+3OlCt3ZiW^FMoPx
z?}ir-rX-G5nICENR|+(U2ns}SU?Abk<~4?ujs&6D(G>pOMJIGnHE=xFfyXgN*O@CZ
z)peMv#U^1IUpg?D9%*j~d`neWp_+atvqiT{RbD293|ODf9Qra7Hk*`nKa)2+*>j;L
z=tQG3zli`8hTwJcCr(}6aodJYAAr@Hb*4`<{vGm8g&{2cnM4p6&WJy42`Nu?G=}eE
zh!V%S=|DVl4tKdX8HPZo-JR~<7^1??(xf_qXRK<Pk@t1lN&D}TVCmleR4vSK?z|;$
zf)FH^qd1|#&!29o4ct1D{&gUplhsH8L9lr9p20WW{`{bm1*GW`C>1`nH2Er<*Fw=$
z-e1kecwOpEr>$%|z~<7d5>TS8J~-eooD&*PMFKKu{!DX?PcCk*`Ez?i_-Vwk!XpSE
zo-SN{-{$@8P;s1QovqWf9@=4W4Pwf^zQl>?oYsI=wUSd|T&mK)rlam#$xI<yf#)W5
z?l$P+i~&v*Xj3U^e@6GtDnvN!HBzkEL>8@HYGpEsAoEIqO%+UP(GXA&0su#qbQYMv
z(rq(&Ew*-1!_Pe~rROlB-wD@itnoHM*ia(Z&^3@;$vIc;D3FWkt-ozF9J8<)0N{?E
zY-df#d!{aA1&e@C8}MX{1?`EDcg*1m%~C6+iGnLLP?OTDF!W9Vde8brm2`wtFs(0S
zrBnjMDviCXfgCg^)8H#kY^4?LHMcKl2tSNfva|^RqN(g#UfAE;W;uw`y^drd3?Tpr
zfHOwV-1bKI?w;sbxq{w;V>6k-<5C9KEo!`WD48F&V=+r_kB|wZO1pE+gFidCm6I@D
zG&2>_v~Eeg9`vf;^}Chyct2Fog|fdl_K6p|(*ZlL*=c{L>!h(1*gC*$6*7PTsCy_w
zN4mDQpFl(;M{{~2@)k5hvG59;1r@hz4p9op&Ro-ytPgpAT^ICbs^x-D=jv$ivr3Ui
z1Sn{RZ&z>ZxQ39|3#0uBodsG)=*ofhCpA}7ZSZA=kgQCzi6G!n1(JMvU3(xohX9bk
zFk}u9(!vm6wG)`I!p8z`0Os68nGt+MW%R!Rj~XQ^cd|$TBk6pdl<<b_2L?+$)+_@f
zrfE2B@W$%iemz8!nNtAp%2LGa*eX?Fu&6n5_p<i7hfi4)EU`n*7B^D>o9K0MK8YQ%
zb#$hg8XKhI;RvFY=LpukMQN?|WWOt%?_6$X%>Tx-MabrLvw`+3IEhLo@bqP2c-iqS
z4>p86PjQBmhhbPfKd^!kz$)Al=!OJ9{L;bU)Iw$wsHG;%(*$5hXwJL@1a|A@0eW_4
z-|Bq9xUlj$83Hud2L7^YVe_zM@a;XvxdMOAy|-ir(m7*s$md>9<HY-RZs;7EtBQnB
zG&9r9(J^E@dqOicWHOnzY?%ZAl0vfOZM885^2<T5Iz&|t3gYQvO;MLuCbMbh@jN?H
z2M+-J=BImM*P-aig6JyE5M!R)rEpl(9JzJv!iGC8TwTK)_CSEyyqKoDsQ*r<$fe*E
z#|!AnFE#^}xm@9(N(rkw8+Wxegsva26+RK*a4NrYcVGOH2X`f;gB=#rWYrO@>ev!e
zasCsH1%th@EX|M|03tH+tF_otlH1VD-m9}<|4RjLNI+&f1i|r(Y9H{rm0zfqUpUg2
z>x<=O>240bKu8G)c;3!K(O3NUpSF=RxI2P2OMKMEqj>Sau&*zgc_Yy1|0k=^%<pkM
zyJ2bTGek6pNu&sn$`|_pV4vl%Xfhzf3=o!sKtg~ER|fGd7p>mh*%Z0f>sEAf7Fi+y
zN#fmuiFfp6`AW`MDY=qx`hXs-i21kgKO@u^O}{;w%5_vT1QEaMk@h<G=U2D-^ywNJ
z0KkG~a{B+Dz3+~Xt2ozwX69_))$XcyS?=A&fax7VNgzN-BRwSHl7#ef6H*B2A%WDJ
zMj(Y$2<6g4?*@Z$@3Lj}uI=rdnfLo+&siO<C3gt9-+6vNVIi$fnR(}(XWlx!3MDYG
zSHmi1&jK~ui5<wZE1H7Lv4glDBO+*=6>FeQ)G5-Ao8MT;&1T%M)waKQR{f8IUU%w9
z7@u4kV|(R+n7XxRI9@T5$ep>aE2)_q>*1XK1TQ9}D@Kr!fIE=L6;2&ZW;JFr8W9m9
z0dF@+dCTI~KuRj5O!D$?D^aA(O*0&DZyifzSF>r$q0(zd9qh(kHnshVO&qsre==_l
zt~5W%i2+r9hL4`VaCLKa#ZSC$EmN!|Y8w|KZC8Kn!_RHcEMm$e_b8R4I&=#UhZ2H^
z2FKHH9Z6&x%XJxnP!;k%-&*DUMx{^WQ5MERN1`k*@E?;)Y}S6AV6XcKjnNH~hntxz
z4GBeH$Fzr%9Q4O<z&kwMCG&=iZtWhuD&%uFbHSB)2tVY4!scPd3<I*oI8D;*6J+pk
z8a3?|nngB?Fk{H(_4?kCxF4@hSc*lqB(b7}T`EE>ozFE@1s@K0UENc;r4JSbV=B8-
z(HoP<<e~%PsSD3Ps-c=q7#z4d^h1OM08vz+ERmI`^pB>h)TC!;@VG=g;1Rbr)KqO^
zjJ<XQP~#kx0S}v3@0~aiv+3K}2u(m1Gj^<iWhj8_jHMbYL%*pCdY^P@lMm?xAd%Ip
zH|-w$Bmg%fC>(}a-dy-a^FR1$gX5WZPZjtY4f%Q(&Zzz7n~rJgIz(Z$S;6-Ez(hpy
zRZyYK%mOVHtu`%K`*m}g3&c(&rX7RwKvydP!cYoQsgiybf_tMC{srby^flT0Nf93z
zi><DU`fp-@oQ?z>D8fFE@_Ss9ih-7hvm#gTNb+NeR3>cJ>^!6xK-2ZEtnZoN)ISn$
zkEin=oM0bCrR{ZTs0w-i^3JmtY;#S)?qB0f^+~C5#khw5fQT@fyd~P6cz^4j+iiD-
zQc4p%w`4}`pX#H5=Uis-R{MQYS-o!4o}o{0jcdHP7WtuWhK7`21aQWgt{W$eCUPHC
zWm)z`^5-9!*H-hL^N()ri}*OVbhW)q&|zDO|FaB%6LY&C>8uQSqlbDhx4A<(0Fp}K
ztfOab@_Dp>Xwx(uQVPiF@~Ac2dft9SMwCp87||5*jI`IGNF9C>2qRYPTV6CWn&A}y
zGPweVV;RJ<>eZXiqHLEE0$0wh=I=Un=EIdC--8~PhUvPG9Vq5Eg?Yf^nrx*q2Tv*>
zmDMlZ)i-(+Gpjb!VKf3Y!vFw)Wew@u)jxg}fx`A@*91dl$h-Nn(-!@0PPJk<5>})G
z4P4X&M|Ie*hzJRB;LYc)_y%+2RD=|}=-qeksf~C)<<Z32^0_e#<r<8o-}dLHJKx6z
z$5{5j&zw^vConEJdt_Z--G+|AFD5d1_hcppg#2#VQXTwFdt>xz0O}BhrSxnbP3*HA
zBm?8Rq?X;I22=8gCP*ll*-_T>RQTPTv3)C+=IUU`ep)65L1n~#nxqe_2r>hJu~eQX
zvU<t$z!?j54JA*u7VB$4PKZdo^VB(o%AofTHDP~>=zzp#S_Hf<xHUe_y@1g~X5QM}
zqu=ODLRfM>-Fn(jrk^5KbU8n_eJnIEp8ei<DpzBtDvSdmzw7blYX42ij7pVhbt;jR
z*Xlr6BoU>!N4}p8OcEiLD`0JRdde{=PcAx?fNJY+Rk1ifoATg4(aP5vtNq3@mtvUI
z1c57h<{1jQ>~5^|e?Ag$Z!Gt^8B)1M;+e~S_2AYkW$~Aih#)jkdRO~<THh;O?lm^1
zF2loXdz-d&jeISZ$^}psOC9pMjk>7+7h%8mQR_frtxsYQ2NO$8L{QUNPA1Q^Fiavb
zm5kfp??`65?rVKblA~gB7R%{MW8AFsHO+dis_@NLKwzrp-<K?wS;MTbD3Q$@<5d;D
zH6fonKcy9l5i)r_Fq+D}|E|@&%fI{2bqCaOs(c&`DIxM^nX_UMYF}Gjc-G#NW+M}v
zxvqNSlF)QKp!~hd*QUL+8UV!7g+|7h-}?JZK_U|JHif)ggBj0%Aaide!c6yBO$duq
z2uu{?w?dbQNXhxl(Vx3DzMUOrMu<n#sKV!-Qe@0zufS+B_np5yy}N;PUh)7PMl}r%
z#@H>7Z2$OZB6pHhG-<WEHQwJ)>A$8WQp_+vY=v)o$o-BKXDnY632p&^3R#RNIl%M2
zvYoVxN@?>)@Wq=~ZCW|6{)<7cYiGHfCbN2Y_dx9G=eCc(xhBZH*5Wc`Q8|_i$y5u1
zi16691U<50D7d93{*zdyaANs;dfnWJ1Y9?@*M)wZ%;a-Qs>2~<^w63fa=TrgOkUsD
zc=A{(yL?0EkeDF762<|RVX;0l)grNK5?LpSMXlA<*hiQEV@$NvMUEvvGuw_ytH|;c
z&%vfM#VbI}X{wwf)x;S{D)Mypj~&g}S0POMgJJAl+UE#w2&AN9i`$|*gFe@fG{L7o
zmJzVOP9UaRZ`$BBU6Ok+mbq;8_JKEZ!NDgw0I-93FTpuD=L9g`JDPq+?`Z0Dr6xak
zB;f5hW?s|vm!8nRvoTnlAIlnRv*^4k$L)P=;sk)!^=9$Pu91i$YMXx3BBb)tQ(x)#
zPIDk&Q*5aF-}ue7W1q90!B{w#*K4P<e-n{5p2{@ZHf6P+O<yrKsZVpsf`8W1TD<GD
zrE8bZs`+Bj>)Jh;12UxZMtC%p|M24*M^5<iKeo^`9usCri5{~9U~Y=8zPGI&rQM^3
ze`{~z$D_&IiOjaU5fQvDQHX}zziNp1en?PR(*tPV93`fRrJhRXSaZ`@74+<L9vHM(
zI(OXTn}#_})XSK1e=%OgY;lVq6Pqkoi~vo7an7)3G(Xqxaa9pbacrL`uM{0k>pn}a
z_M`izvgVUbKGd=nKTcZIvPRSRrfG};=0P&MsB>Wal3Dcux2Z#veXZFK2=@20|G#pB
zo5_WwE4cg~SMziaP5|6)t)XWuf6jPra;+;vH(<v2Pxa^a{}<-rf_DrTuxBW?f{3E_
zj|6<4H^1DO3<z7Z_I(`^%CcbhtClS5cDYgFf)$OMW;a#-OB4LCgf5lS%>xvh*6K49
z%dF59eHeht>NW;|NpVnys-p52Ts^*d<k*o!?p@h}5wb>{irvyOi`r^`{(&>+K1)(r
zIT(js8im;hoU`d&hEmZ;TV2KdrtMENzxV-I2`mLWNy+ZG>7BUr__jxv&#L=e!0X;C
zPGTsQUc9|$^sgJclP756TH=!fwb>Mc!tGywetHj9&W-r@3?)9_JDPlBd1#);C8)Z>
zd-HiKTL1g(r47lG=12Fr(5E6^h9RM(gy0+xt{ud=E1DSr@5}1aH@&HcSod&zVRgt`
z%S;c=KHh1%^Eb1fHl|QO?CM`VuL=gehjZ@V7a1K-WKUVWW7LIW#}xYscCUL{%E-1*
zx!{~O>`CxMrtl612upf&sW6_-H$J}U6Z48CJ10uil$FF_^2nBoVasMjb2VZA6O|$F
zp6ML&W`4&$<Mx6+kKxjIdS8N(%;f9V?(F}}zt;60hskZaSQ(Mv)<$U96t@+s!vOL3
zXLet(c4z<RGkHB=-`p$wp2C9G>gyNJtba%qv%>x@^*`;+&n=Snn<x6Pw9$*LT_dN(
zlG!EGDoY{+ysnk2Hus)-^CLU#7870~d{j!3IG}g5>oG%0Yb;s8kk!ddK}==*W&jF?
zWP@WVzj?3e!Z9Wh3ffzz+17N5zv-CPhnLT+`Bc#7>X^hp01S?$mb|pB_e<|Pt?e=Z
zu6#jRmPdOqJ8vq+fikx$StdCEocf)YaPs_`&@-EQuI?F4o{`rlbQuBgx&#`l1Ao2n
z*co5G<oLGKS&OO}Fv*HvvJz`TOjmYd(!1{&&9HebRp)WW-1|5n@t&c?N!1nJ$gkI6
zy7k0u?|M-6R1Zcr0dTG+tv?QcDG>hr)Gm5?$Kdgdp>c9b=j3lv*}|z$ZRoFiYSTcu
zUOo_?IR{WGw9N?gug!6I-J0B&6Q<Z=QnBH9_VYPIoIlO@G25ZY9_u|3hiz67h<@lU
zdT!gOFPk@xrYX{zO>ZrvR5e?>M&2Cpz!l3A%E!%&B^2fRlPNQnr-^gMXhyZm-#3z&
zQPAb2w7J3Y^jUqw@v4_PMo{*=(;1=Ifv){B0~6T~!o9J)tO~hrsPMZ74o9dODueL4
zw8XvyR+o}*?i~Ku?;hRu7IE$`6c@$Z1v935y2JCN-wDC!ly5vmPiz{T_uQuLYeo~<
zC6f<#8fG>`Z=2Z|{WX9B+1A;iZUo#k<64ug03e>uR!F71Q<FKC%w{v26p@v&D7k_C
z?%MVS$5vJ?69UnI2LR-fIgO7jnOXad3cq`InMt0#(lO}y-&H$5{`}5QFXortKDogI
z?Nv&pv={{Ehy*>+zdpX>+MRvlA5LX;pQQ(>Nd8b&;rY!eiyFVMv^5$B5t(H?#DOew
z>%K4TN#n8~KZkez@DaFOV#N>dT7T{Kp0ST9BKN+0#i^`b_41B^PZfaZ-=6QHN4Lhn
zimum#QIM0yKyZ$yI?@o@8T97wJwi)oR$biDKmML#zvJl@_@2S|={bp6CK&}yVMQ{7
zlkJ<9bx-xNU$3I)xAwhn&tUuwMaw*uCQvF`w5e<Cv;ApA=YIZi_2TZdb)4EufgK?t
zXqL=V;@zjsTvZeC|H1Ed=~H@N4q)ZXYN|>Ge4ei92n`5O1dJtfCp@=#;D-;b?Oj>_
z;kyZZ(iqzN1XGzD!Ch=g=laL9otLcJJ^X`2MxSkWOGSg;Y-45c##~PSeRD`n+v(#-
z6d~P@+;bc0XWzPWE3WwIQ|g4-(YZglXVW)^VwsPp^M<W*{@xp1DqCpVyl3RIE={z3
z_s&fuILCNKSypR%C@`l3jj3VaLU8uge{97?-+ma6Z5%p2mDfK~kZRiFl%WU(Nk{$p
zsgBd`da(xv<0WD=#fHR)O&;I!u~v$}LGo=UwLh?|z54SZpKE)$E*l(Azww1_12;dp
zW%ShZzWb;>=4VnrX)4A&3zHDziB0|Z<p&mr9$wS&=}n!(pGoKRsC7;+ZVnukA<wVQ
zTsGqy3z{P-CDo+Ep`~+eMe)n^7zcpSQ~^7Ob9iji2tISmD^!rIX?I`jJ%p)s84*D#
zU?h>PTfM#Ssw-E(`=fie%U8Nmv~wr}z5HOxz|Go>uX>!Q;GE9<_H*jGe{LX62(Af!
z&X%s>k7V;kv}8D)99s{@7>48N#;&2n$Ezy5VNE-lUf3D44g%@`WRV1CYHMExZ}|Rm
z`1u1{v7|K|eP(0#+cJ55#{SfLB2)16kEOox=*E$gZhCYlUfdNgMeLdTg0LTs%<Nar
z2%0ATLYfd#JLRY9mG7NfvubnCXZpvI$J%|DtmO45exK7N>lsO$^rxpf-W_(s^N&^C
z#06Ju?~?x9PFixjr1nT~PCxwL&Gd&ycbuEa8OIxPvdeO4JRRuhAG_-4S=FBVUKvz7
zh6}{a)~sdLOiLaGu}S;Q0dWfZTI5`yA>@M3@9S%<3jTo~h8uW(TQ${0gFCR#gi3}?
z-l*B!HTI34edf%|0Jtx?VV$&*Wxfw1kC`_aq#wL%$<k*x_x*4vmbuXK{ty8^w}!^5
z&_g~=eEZZ(9vYQOO^PEv=(n>b(=zc|B@i?r)Qxwq|KeaQ`@K|7w*~qVIOAxmuUH=q
zdRD&ov{?mH_+Sd6I@tC?cJrqw`p#Ydcy7-HYdZ$No670)>>D=c4Am9>=VvuUZ#1O5
z^S!6fG*C*+CPtEIeP0&KS^@yTxAdjiz*uV5i`xdiH=fR4l*;PDK6yc}OIC-yzn$M4
zxdy=4`OBLvtEW|m37Xiud5vYoAqFNmS2y0Zjvu?Q_2@l=v5U8LkIXS7=IT-|%@q!e
zz_<iURaN+(Z><Zz(pDROJeEv9(_HB@R?Mo-Gd8vOTbq+r^N$gT3(lV2HpVvZ9@LJW
z*Vx`Ql)89(&&bj7bpE7F!DyU5(3ean1oZmK;PVUHs&8tFxbGfF<nqn65#y+qkQ~YZ
zRc>X)a0)KC%;?ab-q_1i**tHot(>*KWAJ@FBgre`nS6Aw-ibu;xV2bwb!gShhRT1{
zhJ7z=?i_r@<6_iW9pX(j6~^r2Zz#x}*VxItv^LyBkZoaYsDg8PeDm<!$6ncUYb>2V
z*8aJ=ux~WzaV_}N#Y;w%R1i}XkCukS_MvRe1FJiKx@#czj(rSasRd?b$h&4iYt1i|
zlz)2XsqNBSsO*^?2|8tN7>1Nw@RFOD(hB^+7rH~^>HL>=_Km$Wlha$tcFoEPU&qYG
z=#A|Sk>8y<H>#T!E-RxHuSHugt41NCB41h4854|)g}eF_7jN$wS&}UnExMsrX7c(0
z1zUUF8hkGCY+FO*l{rn75BW8$9*kwi&tBRfU7W!@Tv~?>No?xchD3zjqdM;!8uj^o
zzQ*p+%p10L4V@5A7nY`Sg(ak>U>Y>xQgLY<5x-~4{MPF0TWWlFtnV1hT(Gj8TB|$+
zKq;xXZPQnpn7+WcA&V9{#uzwbuD?CA>w;}P<JXTR^RtzCP@tlVQPoBRTiWX*H!f<f
z`1A7i%EFn~J*po2+F1aIDM^eA%SHpar7xxJ>K||1+8euJIF@}wDq9HI$Jp=DWOar2
ze^$({{@U3~n`6?D)-zigWXfLL(vX(>j&#Y`ZO`szUN>{iX|A3>6wA)*9Z8<nJCc|;
z63?6<37-9(xZExgs|@*`n%NN9*j!Wbyx+|`x9l0}$mI+b2>R7KPHIyCDQV`|a@$6J
zBwfG?MBn}UEA0H^W;Kq)^Ka-HPMp>|91k0cmS^+E3|-PhEPweLo)}Z|d9;Vy>!aJ+
z>%w;!M*gMotjb+>T#GSPR}V`T`^K|kd+(Sx;0w;|9!Xx%JrrL)6w6c@idJUwdU#sh
zF<BQ=H00gfSQ&g|UTgIqUf$ZX>8zt#v3y3<P*RoR?idwy@e><|G%0mYWh6ACdpLF0
z&c3mgBk@eVZm2VIy4?3-H1F?oYXd>Qwt7}W)w<b@kp~Ke+{VtqRPOZU?M6#Pz+4B!
zizSp;x&S=2p-1S36g8Eh<!g5hU$(1n?EHeE=1;ZNxHX=y40#_|&{q4z%!Y~=Vu|#&
zZQWzps*s;n&a0!=YM-v8g58p7wfQ_0`LOnBE;!x&V)v{~U1NXi8%-T;t51mLnu_~A
zbl%ecstK^3eO+s7e_E{HJ@T<<H}(8bsxm{$6p2uNkJdl8sq)r_%FqulSW!Q0|MZsr
zEaujE%*0HlzH|FJeEG6vY->;4w{B1T?OVG?{%b6mU8!uxV2@iQ=QLIQu`2Al{<7nm
zyDai&*Z+r108R2HO+vvmA?OFUul>SED);?N-k9`)oHPBz1x?>PX<^-W8^Ug7Q!v(p
zs6!!u6U0JFz$I>ZblcM2k@QXD>HLy{p)~6m7-R6c#Yj_C=+Vjw-;aV`_nvc(X-Wn-
z@-wgetJ?MJce+>Yj0bk~42LE7J30r(-<L=i<`<;0OhVwCp~CMOXs?g_thU1Qi*=pD
z@oV04l$9yOO>#%mDSK0}1QQH<hBMLIp4oZLa6J1KFjimCrOS|H)4XhNA(=v81ZQw-
z!XTm%MHrh=6I!!kUfowuo>$Yk58-C!DTxD+Yq}ENj{d}lcJz$CZ#<Q&6oS_qQu&J~
z@Tpy3G=)3VrSUP3ix1XCgPH2E|FLMm{fG09YTRnp2_(Te?poDZV?ey6cR2CBf$>x@
zZ%}|UG#3o1O(uU6FH1ylYXUA!qz$RY3%ZO4d>SGFk7}+7KhhBO{q&55wVgJ%gR%}J
zhbY^YTye4RXY!_BzVX{z_KbW!Z`P!;2w>RnUU2n=i$|1HG|eUnE;#?`y&JFWAItnC
zl`XiYj0U!^69)J_u2@4f@UO<2&<&ZK(Rt3%?b(J9%Uu1NXN{j<`Fjxk*7wl5L0`{s
zEEEp<&)%|U_(SnbVMR`tZdOv7!U+3atB#t}aLt8BH$D$QS5j$~rQWK&_i7VBn>9=$
z26}k?5W0qwKiJtfb_oL-3x@O<QkCjc5z*fDm)X?rcehIrfFY$+thuIQS9@LbbEhn-
z-x~FTm&!|-Aq`Uzz^As4v%%5i72CQ-K02Drg*Cxz45dQ4uJ&^*F0nue&Y@{MCK&o_
z!+}&&O~r#10oPwE{GN^zXGe7_#)%Q+bcX~}e}WMN0g4fUYeM_$(;dsZhEu;AjAa(+
zWxCAo5wWJK(EXJm-;DvEyX&kKP05fzp1t80ivIPn;0wFr(VgAHnoj&(yZXj1Pv!Il
z1zp-pETJOc8EL8t{j4VB{_&lwcPIXI_34_Wzii&;>13M~la4s<TQhL%?*6f_cMc_H
z5D*lM1prj#bvbns$P{{v2yh7ww@c^*8Z)HIlA?6j??zo^sINZi|JJVlv8O+M;S!>W
z!oc!K5<oM~jQ#kYjc1M}bHB_LWDNn4f?f;_U~C#;v=YV4@+bs?R(uzw6b-lq8=u)!
zwQ6x&<=2m&UD-Po0eoTGpx8B(e)opX;m_p?GAKB&RYd-RE}^JwPr%ge;mX336cM^T
z8c)t@s(fr=TlF=^wO3@U!x0yZKeBOT$?m@Kuk0R-Ey(NAFF3D}ih>10!W!f$Ppaib
z!>PCrd0e6(6%BGmsk&$|KchbKyG5;4f0$SA%@LrwM>1?}=XkRMyRdgS@z(C4M3p3}
z;EXlo4Jr1Dt+d@v;~XxHrxek+Zpf4-xN58nqAnVIaBfTG^($wF2bEHsCU}LKPIeDK
zXK`W>j4>l#VB+Qn)?e8_p80W3ml``r0!T#cFHd#M-_aL)ax{^vp?&P3){A>wTA?cF
z-8iSY>U;fTiDyn((w1A^8q8k!&HHor{&p3p>pIt_-gI!yj)8#R>z>mylK6Q4XzI*N
z!HAS9XmE~zNA%5YuKe2Z3tRp<|H|8A@}KXt);HPp|KSioVpM9X%u<rm*Z=tZ7o}9+
zqq2$gxI|`Id-VzLIeqpfixAjV{K0P4hLol91g4TTHV^8wq`|}Pc%i##IG+7-=RoYz
zY{95Aq$({I&KP;!Jk?ko-V+IU{uN7QU)nto?_JVfd!iybrEes8;#e~0Cy?8+e6ixD
ztOq^Xvnv-ge`juE`04f<Z(bRaz?j`UOT>y+wW$Qp8aKf?J+yx4te1E6|Gsx5+02<$
zT8gGR)J<Awj7_mVN){vrkH#tB)joCIyH@^m8sWCw0mTIZ&=cFok9%Qr-z_~Oi3O90
zmvz&9Z-bLJX3k)~j}Z-epMLk*3;(OF%DvsiIW~4Dnjc!T=Z2n<<V8xsqlhNgSF-li
zDXwdmC&o&a5_sIgXsxce?dmrzyxf*qHf^lcfylPacFGxmXLpR&J@QJ|y`#z8QP!i?
zNBpBrm4U_YKYQ+wl1lB10PY@6*WLMS$CpM@`FCY<dZk?+ryESp7`Zi`uCELZR)_tM
zXY=~HO?yVx)r5V)*4pr~*}QS~U@WyPp3b|af>)VwfF73ym&SL^ZmPUzR$buwGZr@t
z08pmoPuLbeJJb{MArL_8WX}KkR0n@~=kN=LqUA*8r<((ZwdZoP4w+z#j-J<e(@}Hk
zKe4FZQ&5ToSg9bnoJtlx{k!J`5k03At+X_fy%0H-hj`KSvd<WU;0%gLZ>kCn)K>=2
zf9FXpn=~%OzBVJ<fy~mSW(W!cV_e<&Vo%G!c<P$2!T80QysR_i1c?)zDNV3+OKqex
z67)PW5=*b@98B~dJ-6XJ&e)ZE2F6`ushn3za+8fX_`R+|$m@D$<@}~^2fVIlPM%j;
zkV>&B&Unl`%<dgDWk@i=F__S!e|u`jcX#!VUuhaDQJy)!&x$Xsm;qMaRH<mgWac^t
zT!OvqbBhzMeaA5dO<YcCM~(n;Q4}WcO^@vO`np|1pSP1NvKn^N^zT%%!C5Y0+fV%-
zSMJ#Pjo-iQxaMo8B5psucVjq})i?H!r<<*2Gnh^m_o;`=zGFbp>mEA)xY<{oJUe`c
zC94Kl^v8E^_+j@*;sXGfGAC}Tf|k?2>%_h^Nnx70*`@Ks(Q_JpaOrW)-}Sfze)F$w
z6(fo4s^NHM1`#|`77LzDPc5g=4CX$^m>Gl3pdU4>?z+>LG+p0T={9Jh8;ZdQQ+j3&
zL>4UL$j1I0jmER{pWWPb^Wb>q1ROL0RD39wbKk$F=kkvJ*fp7goS82a9gyXuzWjj8
zIm=dtyorXY(DNGSYd3cdKjRjhwbq8`aL&&hiKkB+j%WOaBEKw#Z9oVPk4wbrqk$*p
zG)AvK@93r%T@2FH|J-)?XQoaft=#%yU+^}INa>_%8J_?cRyhL3z&tK^D?{GKp;%_G
z(|dDusIwyA9k!IAjkq0ZJv(JPIWuBli|SnjfG$0*rO#0GslPw7`?2of)O%C8!nyHu
z-c3YM0L$cMXlL)(VgQS+i>>V#+&exCfZyYS-=n=)R~fwLnAx?zJ9bud5JXH#Ww9}4
z8xykm=hZ$`9)<nhoi5I|y994mir{u}xHSQfTgz&KXKkHmdI2)q$}ulvXd<Ejz$AgZ
zx*>zPyn(z?ELrDuYq?OsJ3g&IvWc;?o+s$_40_$7$M4n_WD5p}08QiYx&*v#t-!gN
z#J~^-&__g)F;+uFb*5y9f^LL$Ln3FMmO1B|$0aI;#&djjRnagto-wKkY=L1Y5916z
zk7k|vbDZ&vO=1R%UjzVt##l2I-AF`4?t*TFGI;|9LjeR$6Ks6H2de{wCIqA*xs7Ac
z{FbWWJ;RxwCU^ARR4}B*7z2S2%jD$&kSH~?HrPL!%75{pb^U90bdR1V6`hmF<SUEa
zK+H{}Ob;j`meXahb1)w49E_LBF$~AE7>;Kr9STg>0sLO~)_~WwYGG^D&F3y_cwQ4s
zN~IW6An;<Sg)Kf#y@nO{N^7Dv>}R!+z@FiFX1Q)CxHS%sOTg{YAOs(`#+@@ZtcZqf
z4OrVzEU!TTkS`R8bUJTXZea$Z{!})v81pJqVOjcRO0?93sG!ST<H`KWbY3^-pTXyL
z!R-=C)5N%0vo%G8E)Yq^Sv3*WGR6c^QSJ??kkbtqlE4{U4B$kh+%6YXv1sVN<kCur
zleQR7G8BPG0bu=7(I;<T)%8f{K>YG_PCqY}&iRxgNChqxg?9FiEn&bC>jp1x@1Ite
zc|F>OrmE0Cj+<L|)AAYNq2E8TTb(c`VqZH}aK4%@R@71;0$?L4LwH=m=Mn-rqZkD3
zbBm${%&lbvFZDWRjIlV8>I0bN9;#vtb@pZihLjQAkX%Ya&a@!9Tmn@gU-$9zYKsX!
zfD}hQiAhPNxZqrBf_H~}ZltmWC=`1l``sEuQCFqFSfN;xoCW|4amH(js1^(?UogUk
zRH9(WqJ9E{OECD{LXo0f)7`u>=+!kL)+V!s=2T9%T%Xq3FofWVQvFGyQGoRpWt@2d
z%wi0j2rO67qlQtGN2Q9_s0g?R6``}l2|1&j4y6gk1n0~!6a;7RdtGq5M9!s&Oi5@7
zjB^Ysg<(tEFfcC|T8YSIZGuGP%@+*6E+w*s;=Fk@5lZLsH9x$2le*@zr9u;&=XE6v
zNuHvoqZoVVatS3kAGZo50~jZwK2vNsfY|`t_WR@w19@G-t7$IASnWVO&u3J+rDCOe
zm500^nDQ?fftIxdKtx-&bPxU|kuEIA<c-L|%2i#|pDXC{mVa#=+_7o*@Y{0*<LqR%
zFfW^*XsA=Z2Ph&aNxq?2#y1qpTnyl10LTK^&^cVD{|Ro5Bjoptx&(iswJ!Y7yH08U
zi{LCxL|hq)z!|r9UMqEt0-%-L5M~V`L&{PBC1FzCr-|bCrt4Q4R&jf%4V{T*zt?s2
zROV^7CT==ze$$xcj**5WvjF2ES!%3+Yde_{Tuk&81dMl|cE#<3smGtX>)CB@i22;d
z8A`ovB$2Jv4K=MMwqCR<<TJuS@1uZva9&IGBRhIy8yRCTPs(IFu*$M=*<XoM#Tds&
zZ`rnIK{fljM-#VY3I=>`fk?oMnrI*v4*If0#E8IZBL3dl4UrM0h$XVRXU|}~Ih)r5
zj4_!vq<1uyj*iCCY&2Dj@2Cm;V<#<{@x*?NCsZQ!4D8V@vCgKD_u5Q0e@CuhKmqV*
z9F-vtYNMf4MZl8+fd&2Ucza!>&+QUM=U}{cG?A$VF&IjU!Lej@CR<=5sT|Td1&`qH
zyLrcnb0gbeTBnxJs@(p{j{eVa&e{Qnu-}D%$Bn8;AQAF=($+gyR`>?zHbw`DfU#88
z*EN)ADHzhjz*Js0JVRrts{Zj53WkEm#S3#<YM!^=?Fc3@l*Xw#z@^f&chyA#Z|NRR
zUPx0|ai$`H_MZvPvJCkBSKhwj_CG(h>#VN9_&mSYePS+eoHLrt`1AU~S6^{y43!~Y
z+UwTt(skpR6>}TcpS!ef)lVPTuBg;r5+qy#N|(xt`?TD^uSNWyjV~|&w0L$CDen5l
z<7;={UoaH+x&@*^52E3K9u4}&irkt}b=W`BQWF~H#ZYx@=p1e@=rT}r=`!>U$D+wh
z!ILQzH@a}pLlM8Y3qXM}2181<KTGgOE?lhE?udQ!+4XxKOXqb#AZVPUD(FR|!k4R#
z1jdPo`8}@G+@{JsZkJG<1F`DSM7CBc2t`T^ji<utOu;>p%)*ccT!N!68rXB(g4R9e
zfr}2n8ZD>t8WeRINg=rLhbMO3{=Z*7<FP+K-toraMCRCnq24?a&$L>)jHYo7G6sbG
z9wQp^{R^ZXncH0T>}PLT{o-w3s0RRi;Eb7g&tyGX(q{*-mI%Oz-y^Q~dtJ}xk_H%O
z2>M(IcwDH724fXLZ<fq}^pM|^Xs-+RdECNajA74UymC06saHhYWQa5}o{lEcId8#`
z7>Z}0BoOqu<Oz#fR-H7j(jZa_wAUf_uwmOzaz3y7w(77ioz4~9qzIZ2s0w-E^SY=y
z5*%~6gbqclGUOeY(^NUMdoWQu981@N0Rtn6NHUZ6#nX9{00KUjhUyAm;<QE8|D4Wk
z`>xYwWuMtP@|jeoaBn7WFi8ZzTSH~Yi%7_yj)r`RV(KJV(B~dq)K=X~M2MyHo*lj8
zvx$fkz<LH_Rq0$o8;VOPM(}$yv^PYz*M@y-?E6y4g@jk*zx24oi@G7<(gdnP9#mEY
z5|v?pnru~it0RHIw%X7LXRO$=F{^t!ddFK0CEb=aHdipbLu1KsB2!@5LKX}_*zeAt
zw6O7ke?Gel0MxwJ>Lla*)8naJGv`1!;DOJrp|ZlC3kQ8AqeVr)JwB(gvJVW<b;&pF
z8JTS;<pwe2b;C0-nh1}lvOHIi2>M)z__cqXFgubp4-jfP<Ke+N7`RK6tovQ#@~^*t
zXze=-hCI=*6;C&$+~+!BW~QJ^<HW$8{^F&}p1$vu{u9@A^dIGl1eOwT!FV#;oXYA4
z{jiJy(U3=0_`R$01^w>!`snt{Pnq%fckkSokV-PaH7Fvbxio8B*i`yovoT-m+-xU(
zR?MxMq-ToOJB<+(S5Qpj*xobRnlBj3Cyykdy24)&g00CHjJ&NP03D!`+2^9&hoOt4
z@hHFbg`uH8J+<rBC%<+6-G6+d;|~F^)>IMjotP^ar>8Rc*|~xtESFnF&^zpNYwMEP
z!X1pzee#mlJ;%Q3C*8OI?tS>wn-;*_SoR8@w|-_4+i3;Q{=C0<|6Kg+-i^-?=oyR2
zD<T+@pc|@`A)3wW7)fM7M93B-@`l{3n{kQ7vK98UID%A!Y(YPuF|(fai|04v507^|
zrwGs4C#Ps-P_Q<cd_f|X&cP)(vUz={^>e8dlroFH(uC`yd=RpEWC(a@(kX`shE&ka
zxDx`9%j+0TWI+Iu*#ZiN+^8GHch+U`UNn&?T*?+af`n$$0wE%L&zbWRe|Ws(cd49y
zZX%O+SNOey4OKyf1MJKY@fYQOUS~XINOku$Z#{}_AIvvCyQ#Z1mCb97RTXX3VgE7N
zg0WOLWOXu~Yn4(>a(%m9f`Wceuixw5Xc+RP-2-E*8>@o)nakU@|KJ}ljx3yEAW<NE
z;{3U0%o5+r@@rb`Uo+dE&?eS|CIq$*>Xg@wR|xb<iDj#z3>h}voh|4XPvuIAsG!Ro
zmc<mNY^t(WO;Q0T%!(Ya`+qb6Wb?*mMc7<seJ#m75h2ed##1?Hf+L$ZIts>wyi4<6
zdnkjo1{rf7$XM^M1s_SJaL<3A#sn~a%VRtL^u#yM{olVlyZiTyR83uF#p(HiaaJ;u
zZ_gJDml=kkDuTWtuUlJR(Di$yA-hgqGGo`#?V;{pKC~TMe|k|d5ZojqWgedcY6)m}
zz5f2O9UBR>(LNVSS?hX2m*sQ=v2+evj)O6VOkRg>sM0(#hAGFBh>$DjC>R8f06eeW
z!9_8~@YiSd3@Al6mn<HpZnoaZvJ_<V21XNEWb=B5Zb*PJNL74xDzm^UA`Fb@_9+KB
zUG7jsJM4R>GS`MKWyvLx*NgWhBBXOVbVF_?Gxku?-s{4U3Ofg4m|t5<WubekLA1K`
z^47hqIV-)Vj<Mv#TFezB3WnTdsEPGxNCl-Pw@gz+NM&_=^=-@W6(Z6Nr5lp&HFGaZ
z<5Jx7CJgntf`PGQwshTsA-Bs?q41)-P?@Y*2cIKh0GMhYiKla;GwZ|OPUrNSvqW_u
zN`*YMy>oXZQP=LhV;h~M<Bn}x9ox3OW1~AA?AT6+9osfKPRF)wzx#fk^WNtZod5P1
zyT+=jRhYHrysqCIA>z@YNB!qa9$V6cEckdx?J3awQQ{P}aW6Ji-E<<Q>`a`tuI}yH
zs~m+~nxU4iuh(VVa&8{N9+o$Sh4!@e)qr5Lq2B3i<^<8nV+Pc=QU7Jp<L%=b)t+R0
zOD>fmOg)my6F^I_UQ9kDj|*-@o`|ojLym08MHv~Ky!H|}x39bMG+rU$@t%FU!ILq*
zo;apBd0<yw`E;tMl>L5;ukq2V<=an-RCcfcM4UW6=N6I-{Jbk=fb$|T;>N=xmh-85
zhMdAlGDZHU|0Vu5GykQpSOipyR=z3lg5&oz&PC|R=?wSzOiC_LQ@py!o${*PWZY3*
zS+z5ycCh$Cd4>gxQ@=FdYoO#Bv^&>ap&7^L_%<EEVAAt=^Pu9ymboZpMjKlthmBkW
zFWGmxz+qcf%5u=Gc?`4_v;q8Fq5lZ`z1bOxFB>lhZ<YUn);ErBsre^rYt7iH1EJ-!
ze*_aIN1OlX6hE>(VOaplhl32EtWqeSpj>>}g0XiGF7TzW;~A?2g47aMH<wFsIT#9V
z&KZgP(P7Hn-<Bh%6&$%*t{tcw)@YOS{Vg6Z?G%tEpJ%58X_)m5e_p=5nM{{4`x>of
z-+>0$tF+0_6$*g>Jj5^x`jrYUOm(0x-Tk1$?~1P@Y#zK2kP?B$pa^lkC3d}&?&NnG
zhZHDH@;r4}*DaCKKw{`ajtsI(2PLP^@nXSmn{_-?5a8f&{1X0&mD^^0JyDGlUq@LA
z-!2!5fsHpFL!XFdll#6-_jVMy>vDJ^r0;4ErIyel6OKlk`LFj=5E4Gf;Vd#Dm0#)^
zUyi_Bt`V(JIs|p&#hSio00R9G*;xMlB0^E$<pngm((w`Fe)lYws>8%4DY5pJN?S@r
z&bfr)9?0+Yj?p8-w>oQt1}0%&YZHvR;TB?72?Rnk>)}QuDO<|t&YOt1&{P@Dp#$Jh
zKW-lZMP5^0qTqyP)G~@m0n=Va)4;#plfa=Kyu*_&E!d;z@AG;%7)xj$tub9!DHemW
zkY}ngZg5aR;P3hpAZ}P|zUoChpZfb$AsHtN`k{X3YuGR?rgJ|fp5?!v*P}>ZSW9Ky
z*U~xzbLYi=A}Kz0tt~dx3ws29-}EYl<TMkpn0_7c9-e&OLIC!ySCc*0!<TF1`Se8_
ztg-Uqk-E{8{@%N3f{E3=9PtvywsXNQ3Y9@nS<8_J{y~}=H>c)`Z?%Lua2=GBAxI%i
z)M&x6Y!_VRMarm8|IAJ1HXh6Kuw>phmL6qjZ3t<fAn(>ZD>2E&CEPje>Fw>8M{TGc
z?WY5?gWq|7f68Eqr9Y~8T$UGonov0l^oV|ucCHeaS=>Gjj7F+zle(RctvYS1D;?`5
z&!)Frpv_-$DEA3C38{UUpx2XG4rx_@5g(iiMj2FX^e-4~^@xg?+LdjX54+?jJPC+C
zlud);5mlYDbSST>k@b(8?cr3Q(Odg#i0@lBgW(5vAjS$c$XFNlZsm9B1k`GBH&1O(
zXoSTwAg&l-IQ3j}j7|-6)A#x0Hi=~|q7C@o)K6~)9jZ;6xQMu)ih2#)+Nv?pFs)yh
zC)3Jixfg~I5e?F}t@~)sFxNiKEZ>=`fQ~_mM0kw5*wIZ56WB{ITmkZ%Rx&%Dz8XF%
z?~ser8L_4Vj)-OQjsGO-+4iLd-`-bf&*&jQNEf&<S=c=e@CN&{XtBs5P=)}@ki8Zd
ztjjy^mmaq41XU<cBRK!iDZ9c!B~sVu*Cyb!IA%f*!<-pcD=hL^6HdBwkd1(PEgmv7
z_IsaF{!XT-=h?&IqKfdT-39{ZYEenl9Eh9xc>px5(0?)#{IH|s>itvuUAB69A!zOC
z+zEx>JboZsdZkxR`5xz4>g^VV>U%@xPB{0V{sP@F`i2QN`jS4RgfRJ!jTe&v+I~+9
zr(_T*9d=z^wTUu+8SEod&A4_~Kigm7&&X*kcY0H>{FDKR^c)fx3bkK+r5;D1zB=K!
z+{m4!Sf!G&Ky;p+sxS=FbP`#$NhiEH%zJYV9;IYIVyQCQ94JYucWrIJfOy^H%Y6NG
zfz)QgDkhvZbj7C~rL<=|8TkbT_X`1tfpPrZ-k`jbS@@Y<?fA2<QOUx0v=^)AqxK(l
zGPwg?1{IFUW6zULR;gtWg0_X3{aHPb^kLLqluK7*Yn>@3<s>q*FW@ptQ;^K}azhJi
zKRUaFEsC%A2G#JACe@7pOktepZ8J6aeYDrm1Lpe_ue0;sVgXg8(T0EZUoE{)J-bh7
z!B8&3DfxpUi%u&PnfP9>i+71TPWr1fB*j3TdAWjbyO&4!G6ofL`;dU%;IX_zk7g<-
z0TV}#U#4NEGe++OL$nHXeZJuo(<*tWhK%V;n`HaSduzepQe*shoUd{l4beQKi7z&9
zWr}2%H#X@da}4}<OFWPdIhc6-@0yvF*}e;8%Xago%mil5TZG<I$LBr9C*pGW?MK#s
ziJ>csWSm3YG`i>kzrIlkd7QVl!V^@^_@^h75lgdbkCrFibC(u&yY2R~Y%g<(H+^n;
zyk1so-GpQ<g}}vhr#aT!3`YTfmvyevq<LS(Jng3Zf7t8zuKzn$PjF};J48jrl_>?Q
zzZTXI-Yl<RTcy?AX|43dx8p2Ber!3oLUK`cw^ymqt}mM7j!z3AO&M_G0j-EM?#SCs
zFTwwFNdI$e9}`;|I3KHz-zV{)62K+T@0UwwT{G%G)hPfyyE}g(o<1IvC?`m|ADSE<
zC(AI4_R#fE=%_On{(D~T(3MqeSMS<<foEG@Z4QAXO)IUe0MTyW4k3c4b<j$W(v;y%
zv}3+e&^w?<NLCj-7yoBcyok_y$|ASgQV4blB=54X&6@_}2F5o^63dVfCx0VU0gxO2
z{qW$>gsIlyDR7)fpfb|sMovQ@L7f5qcZg1`9{aAmMJcGj%z3=!4j;Qs>epe_dJcsK
z=ONn&5s=AvoGqb5XniIfKS4QUcO>QQ>ZXUO-b2;Azxl&&=&O|!AMkVde@@2P+Bhs}
z1c#z+2<d6}Tjq6gGs#!ObylA>Wr2ESmSS7FxAuIK%ZnlfdTxl!tMMj`ERK={T1n72
zU!WzkEM(ppW95G?_5TjnfodW>icH)3JZN|_xee8A5Zf)d__HZ?9WFd~`+>87rSM-N
zS!MoYT-Xm2X(|+2@TMGvEUy`S|Nj6yUPPhzs*vhu3lOmNWh51R0i>{O1|5V!4eeei
z;vot@5ZM4RNK60sLZCW+*g3g6ai%o~NTn4YcN8l)%Adsv|M#9*6vk6-*wgtj88>)Y
zIe_{bLHhO)wfDbs`QQEizx-iLz+weYNJulr-=|Gv9yDrT4iLfmwZUoGcADf_3IQ<1
z%3VEuez+ER*i_I(Z$jmHZ6E6GkbB;g6%WzAey;S@&E=h*<X7{*5cz0&TjykLg^GJm
zN8QJdN%K=+^-y!C<5Qh7HCWvWl2aU6*dG5>pW?`tPdfNXwV?j&9`jH8@|1T>*etry
zzmli|S^oZQ;hhqWgZI7t+^M73m-zMmOc=~>i(+%ja}6jQTJ}M;izYgHsXJ{KpRa^t
z+eH2N3^_m!4i%@eF232r{aqUmmM$nRSc-p-tjI8jJ-8)20%8w9i`K)>V0^8wyEb-n
zMqJ%!nh{#?b%+UL@GovS*3Qg_E==<y8HA`RCLRtY`!VNWJZE5riC+}aT^m?TN~yhU
z)97b-_O*yp>WmbA&J7OowSE~*V4;zcilEF6PNc2UTh;HzSJCscbg0_T57m~KMX<GD
zB|wq#QUG@32m(Z=074<^X*r8>MpOt2uQ6kZx&}!Y3B(T7Smo|oNF#LND*r7eq36G*
zPrbc?nh8#rJYHzwpg<&aE{(Y6U4v5RrCs}C6Tk>x1vLzLwDpOGlm{TSsSEpptPB`=
zYYlq-EcTmGUugA5N3k^gg%#@NV)7{nMT*y(PdE0450!_9bIHgdIvZDX8kE-~;$dMe
zJvw1UahpC`%rd&>HtI)=DngtOGra@1z|O7!pAd$9aH~TY5;V)s1^{VaJ5{`{z{8sj
zP62^yqfNhY0yR6}lYvyXI}7}@c4y;(m0#3BJ>ynpTsNW*3d3s^_y8)7-w(Aq=eGnY
zSuAv}v{K5EsT(jMdB|MP*$%sf_U~h@&pETjYiZNaMDU`VpFh45&QkVLM`QGMo$5}h
zVj~BMhK$9|9k9Euu8=0`YhQ|pVPW0sCRw1;jfH3?nVmkT@d_c*e5{jUib(N=w*zk9
z!&PK97)>Ya*#;+Gbx(6_Oiim4@YB4$(;0$W77!U%MbxtG%caB{lJOQ&MFAZ)QW>6G
zcK5NFbTH5X0)RRYYaIro$ygiR=7_w3@+4;KM1s^o`wWG_ES#9Q4U_DtEUU;ap=Tw2
z-%t!h3v@s0?N{fA{ES(Mai+f(zuFT}Sr?J0VI5CcTS+>t7u7O&g=S}yQy7lC3tDPm
zG9`>#Gk9opZYvQ)Cmj!ZY5{yaG;xih3ictGiQE<zS{ztIqB41%$WUs*tEiL%8Y;ay
znW?S3_{Cz3{)*Q-R*Zd??IhgXjwo1DwhpA`{Ll*IV#Tga@C;_q?lqbst5ikf(UHp+
z->F8)zXoS>a8W&D0g^Z<bTXSAvXUAjbR;HT)J%;JUdwGzeL)!5+#<6uj+t|_?&0Bh
zcu@3${1Ah++}u=k9tK~@a#8!FJ}PRdY^2$Y(xP3J-DkqbahSV|Mt3;LJk_~k5NRfw
z-K_!E)l(Nv{bo(f`*l2v18+3^GR=X8I+q0s8DJta+G*TxWLn02W_z|#7P6_a2mj~#
z_-tdsK9gJs05CLI_B7e0_c``rz07r)@Z(w?Zs8M*nK=6T#h51^#!7j3vPxs!<!5~=
z=k_d4BWV$7L`7)CJBtaUS}!29RSx)jywzmDUbAaQkH(>I-I0hm>NGCtBq~(6!SU*D
zTl=Nq1aXrn){#(G??H&;=LcybsmtmcmO=%jok4+o6*Pft;S9KC(=(5yM-i6*fe~06
z-kO}R6;4ByY-;gh&fIA^me9CBQIKDu`%BYigjS}rXwIImzv`Pe{wQOmLI#i5652=#
z12`yX1BlK@a=Z6=!Ko*?zNt%a*}@m6Au1rsGn@BolrEji--1<^0<ItIsT^DTF)ADy
zug21SDLHs`-cVn1S-$4BT9c`%>S&T=92D9)+?*7xyYOA?7=3GLLA~S$GL?l83)I>G
znC*$A3<jdAGP{xUuomC0C80YAa@Ll9{i^cC>Zuxn3%Cw=dl=gXVf4UQtxyIkyNXjQ
z`mZ9g0IV2nB&e*mw9xD!@Zz3^;gqi*Hn)MP#6KUi>#-+r<mq*~#ucKs(e-I67mW{E
zkz9Ib9K77bNt*1x_HbA6U<AY>N&NU_c-*Tj06rnUukCP&Pxnr_{ri`^u!gZ=nbkt;
zbxpIWj+e^%N;{({>qqzs1-=ew$|QaKO1W^Q@bxEj%J?^VK%NMFQPb~%kV^iKV#bq=
zJO}wjUdP`(x6_zAzm|8o<8mW#7`z^){;^^RohRx~GtdjhcPBd|kTnS~!wd8>h1+ns
zk$=1U*4?M|;l}B4H+cPMBp4bc>_?un^e^^WZxfCZXBSr5W>6tpW#*&}?QbHWr>e^)
z`M1kf3CS}j`k#Kdu6HR3abkf+7eMrUfW({Q+PxSxW&|=z3Y4jizRP`j%W=Ull7;J5
z+2eYdsUED$Ae^QtMSTByto?AwraMC!y};AlE`LoUbxcOh7qI`5E8s2u^m=`@3%0jJ
z3&{V9^)wN!wDUru|3hOnthPfi(o1s~&B>eiFjo*wZngX>`)}-aBG=^3bqG7;=XEH5
z^m!-3KMi>O$JZ}cqjIVFRA+;mxfdetg)O)#jauY6v|AA<@7CogJL#*25I3-I)RO~n
zQyEcU0hkr~JK9Qh2ayxM16fz#vl#f6g;XxH4#r(LP(mIG9*9w}gJ75?nuG@-RpGV&
zc)@I*(rzsJ`~fzNfP!l|5}qNyg^&Yqgiw-2nHv`*Usq51GGj7v0`|Xkd+BC3*~4zQ
zxFY4LkBW`(6acIJ(b0@NA%eIe4?8g~1NsGMVDJ*ImR^MLBzOY7**m~>r|IZnu7{lg
z@B3-!{!lxgtA)gat<_7z-emQyknQwy*3E^BnQq5Zg4UYS;+>{MTu%Rs6u3oC7y+iY
zj~OBWgjS9)35eD2wAxNt?YRG{@)Bn91LlOF5ASX6b4y{-A7yx|d{9g1P0Lt^eIxLg
z?d^Fqd;0PP{c6aFOm%zk*JS#y`)sqcO~T%4;<b?2>~3_F<Ny%m?bCj!p7Jz`xnnn^
zXZOFy6n7#h;fFQ@e=FkohYvGLb$gf~r;lA60+2<X3+yC*+b?$Kb{+z9Ar6UmWzse0
zMOfrF8_ueiL5r5leaG>NYLO{z+|_^wTt}U{k{ePM$itAY73Z%a*g<jt4Cb!uUaT#k
z-JrH9`P(r#tG2Qhi}Rgxr|{YR53-lb`TlocX_BPJd{@ObDk0=iPp>(o*DVl;EV(&N
z$x2S%Oy%W2pj3l;eh`rnnMwZl6jln8R}~r-66`QTowA2F^4y+@Hj|~3KcwR@j|Z@Y
z=+;wjTpbQK1&k_}Wx%WB+>jmtYx5C6tafOyqQ|Nx0?B6HrD<CX&nbxq>t3@RW0iob
z#S*^v{)n@CqW4`Bm^`k%pHl@8<~~J{k%=gjb$_mWo4ZcOCA)Jpc8S$hml++QHQm;U
zkh`8<4-3KyN%r0Mp}^ofD-ALd1j7za9ECrUic&q&bidR!zxN!PGS2^|7V_A5+hh~z
z{Fh$gV>8?{pV=;T0HqwlPJG?I{n&y%CLftnN|!uw>!rypsD(_;Wc#tIr)2GoE)fZM
zzSq+V<kW~m0@Q?WRzezn8L11e+0kc%#Zy&b$o}3&+6y?CeLC`;)W2FUu>A3wMDbUZ
z?os?qE~V9It$V-hVta3Nxg$7kGrBDA3nDb?5WK{Jzm60)OUej)DfU5=x&x2dGiZ~c
z>=p#63@$+T@AqGyW!Aw!WwENm3NmcDhcXnzdZXBhV4w5<NMvOh!jr~e9ZDl7AL`jj
zhulf;sQpuU;!DD7^}#hqVY^%zj-k!Q+=*6A;#K1K$J=aFUYt*DTK<%F=U5=Y+idJr
z#iceWhkgr#f#9w#ZZqm3j3$E)%CmB1<#q~5YqBsbd^Jc_OhMb}NI5Y+4YFkjB1pK_
zCLBMKG0?*5^A#(-4|8%Eb3HD(HROj0@dA7h70(*Muf=m&x_pD5AzhtRXWY1riyymp
zgN={Acy|edhRz=a?c7<5ym4M=yC4jVpB)5}kcOx0=eWG+LGq#{GZI}-cCZ;5Bt&-Q
zDDRJw!(0|TR!`^stKz$*Zz~tS_mFeH(>qUU(iZuYhZs6Lo`fK9JI+`-)Nkk*I<Vou
znt^mS41z>4Xm8=L7Qt5J`%h3XzN`sF{H|f_>5@6sCh=XH)7wdV%zItzx4^mn)zrJ#
zk}eLO9I{FLGWrg%t7|q>ajo*nKEhsfa1MqBGNTOA^ZijTWnCunt`t7cn`P*wtKoB>
zNSs8k8MsBmG6YF1iqZWG&x{hL2Txp)d&yY<3Gxrq*%XK7i|}Vx{lP#*|NZZ=*wD{R
zXZ;@3hlmbQ48S4MWJeYd4c?w5hnAmT{)r|WJ_P8Tmly6<`cJvbcuhp2;I%b*pO}J0
zbBdQ=FQ}On)|1L9TwSx7P<t?syzY$~ITA3box;k+fba=j^OIF7J){a>HoE{}0}|Gg
zyyROBV;XUDgn>9WwySsN)m9Y$>y(hkf&#Eif86q#mUc4qEB8Lcl6S_C9bTxVv@uQm
z`eDT1bIdK!zWX}fZU{Z3wt&_Z{WP!Gl^G8dbx6v{iS)tU=%fOdC`?4;wJh`zNHjIl
zCrs^RrVTK4=r)Sn7m-!ZF5*mmzL+7eE&?qr>5PO5iTvfz7ke8Q@Xpn2)cVzyxjb>7
z_#a9u2<TD!mvRk8{5sfc$j^|AOdQ&|my{<F<N_t)z^C^_zl9GAZalzXiJaQTu?-RA
z)l{>1t5-J?b>^6RTPU3lVKV5#{>)oQ?mRhTXtd0i@SA(_DW1!_onyWDPHcFDcGY*N
z3sdQE1qVX56cdYk3O$jO2d<g7X7nvn-cT%62SjmuL?YL(d@>$4b^lDwg|?PFxfgKz
zyZ$hy>ltopq+Qr|M<_82b*qIL@aH>Mbi>(2l)oPsv%%<X&?F}b5bwLh4Xtd%xq4k&
zPq+W=BanViSqGzJm>Tb<h=x%k@7x4#^%HzVpd*wcE$sbVTEC-8d&-jUy}qJ4xtz&s
z6%yQ9-zFiH5-d~t;F^R9NF9O*fq+e4ysy&ly5gMlMh%j+L|N?7UO|F2zX|)=Q<?=K
ze$CJm2^?Sk7ob31m$tbfq%%1taE|3#jh(z_vZEVBhI8YC%GBE{9c_g);;oOEZckQ5
zVQ@4n5yT3HFp<Xe4ItHx@*x2@qC??fV)af<K<jd({(PMyFozLty+uiU23T+MfSius
zEJ-7#bzeGYhBLMRd5B>%V7rC>H=bV%iZf)#Y;dR$ZdaeQ$THOxS$fS4MNAzO+3YDE
z55r90yePxpgXqWvOlo=%0{pXnc6|n2ZkEcI0BLS>D~9CbKZtGLOYtPZ5uTNvs)X%m
zgQs4|?<K$KWWTh7wYtFREGxM7&ZabrID?`i<X8h|Bh`vaQ*c)KIM4&QIKtPvOhaob
z&thAOKylFW?F&<`hUY(HhI&`6&7+wJ&_>3hMY$>^5-FaUv?bDomtyySj5&QBD{bzU
z`mxE{uG3z-V-FZZ1C(t83EF(kh4q2<$I|4Bn$s5i@ix8C^$ZOD;9VM2y2m$i;&lBo
zkna+abwQ}fV)8xY)CzD80M5G_s7O94zl{q*?b0Xc_*bd=u3oKYh)kDP=4@)-`nfN4
zFHo}V%{<A46@s8yCk9<ESlNXrBbdL&SSYT|a|fI#ld~Jq-lWXA3@L|)b+qf+^;dIf
z{35<@e~!sOoU){i2u{WfERGz^5U4!(@cTD@QjO`kp2#@j-TpGHct#MJtIVP<sDq)4
zeAUwM0nUF;)Mu7oT(x<TU{s>D=zRH|rtq?N&}$?;su^vCNbm6&ojt>ylA*?;9!mKO
zcN;59CReKh#b^DejS`^*c#;^BxziSlG+ZYwi-3G3JME$DQsNQzeiZ46jS#si9B8=A
zXd;t#h)zpOtWQfVs!_-=&2m@E-Ugc5@zG7Czk|M?Rz*MYm-qDw_3CJ1v|^R&pBgjl
zj~$mqvU5=<F2+xs0?as<I}F(grx;LQ1yeuHmj#92&VC|YY=@L2fhEywLa+57>4h5a
zAGO(khi0EQJ<tYIzb0^0>qT*D!5OB{5q3KM&Dln&|Jp}Rctem4g)nx22nGtgs(=Oc
zRm9;In#I7MQsku7a>y|y<;e+}H%U0{CCu3>@2k&@8q8SQ?wi&p4HvyjGG?W3I~`2a
z>GRq%#Kq(gp5N^p%+-JjgWUp5LKTIzGt<#&s)!&Lq3*_)keqFa>AFHz)TcEj8Hx5=
z7_;%qdGnuB4%9E>4R`TmqhJ5-S?%A{h14RDAG$<IEL|>_x}-FlhXu2;S+MA_fbHY`
zl@a0S9lw=X`r6cse7)8nHE;qn_s!x|Z!FG}nWc-fIWxwMcnIZkr`Z#%$B$hDL{k@h
z6}8oQ3#@zUJv7zk%i3f=Z^@b;ID*dHnEdH`OO5}4rA$v{%P2Eu-xHPx`F+1DkwjX=
za2MX>70Tg35-%<kXGBhd&4m3kL8msVw%Aw^LW;<Fp`ECIQ3>UvVn0vt-_FJoc>pOR
z81BdxVoH^p!1Jn)C9sg5;?@8x{a{X4xy`{knBSPu+)V{srDqxqWDUVr34q)1%y+Pf
zKOXft2}wWto%^NCMZDiW3o?{^T?G4o?ktK%%yRx|_!<m~moV72=(BmgwWQ-92l=?o
z!B3_c>Clmx<oFzyI_#p;O_}MqubepeZ{o@AD>da_i3Xt~M{5KdP1(Tvaiz@7UJ5q4
zk7d}yA(DrP8iOr0$YbbwKA49U%e-!Xoyqm={HczB?#$Met+rkQ?w_o;na(S3vZlc5
z(6t*J)S!<>o3u$IO*C>bVq;H%RTxz7lct<&VYI<Ig?-qnl#FStnV>@=pq^crVz$vJ
zo(PigbNAhmyOV@PxF}T1Ru;gRCg%_ziu2RYa+T9Syp$e0eo82ptlIZ2$)YlselJ>&
zNq-p{V!BmW_a~fCV_nj`^+}ve`fm$d5&HJT^hb?OnVL<(8dTX6VMX(MrK2`T46-Aj
zCv&!r|3tbR!N&#C(_IbHfVN8_Wymbm0e1DB`EpsbFJ8D%mle-*wz?+T(r|tk8;iEa
zHZN5&-CZk((Ei&GD947ypV}Np+up8>>n*-dmA+x7)L<c`8e$uy6*4fB-pIBBF~@s!
zthhnl+@X*#XeVSh@M3+SZCLeejJq;iptI{VI}>c_;mYP-Id>i8{hQm`aXxqZRh%24
zcm5*2B4ZPMD;r8?&y7eZ(0IA2DjsZb!J?+D|GDcfu8{CMGKhLn9p}YDW90EZge(!f
zlpLbCeGNtz>E$KX6&CRIAMV+FkxKmjrBH0-kXT=!;NaKrq|?WIhpyI1RJYO3%kEDv
z-#R3?o+gyS<%OK5%mw?)c1wY5;V^jgtms@{`PbG%W+yKlIv7iWiIeIltgw$OAy@!f
zB}tzb8@C-;+FT85c3eM}Th6vOK_u9I5q+E4uQd4wDL!)*^G8x&|LL!3UcArYhmG35
zh;60;v9P&%0`t0GyltPySK7Y`*}wFgt==R<Ff$BaxZK(<X|pNuzhJd&J+(S`3bmhW
z`NHlx)r@?_FmVLV3tCmCxy*kn_k7SZ|4jOc;-}{%FDXuZ;EB{xWA=lI$8n*fa(%Z1
zyf_qMOrKYp+KYxQ#m*P41}kB1qArEa7NWuCLVzttQPiu^MX_{`pp)Iy6Ep14tQGc{
zY!g__S{?hMs6V}Mb>scl)&SH?DjA6+^!|pA2@1w>g31|_5$J5~e!0FEWVS%6YxTO?
z{e$Z3#BZi^+DP;s!T5z_aR|z>N_z$ijZ%a4PddLlfQScI1P)vY_*@))8vPZS9))A5
z$fPj9JNzv^=&+z*G%HhKk1wAJ)2peoYv*SZL3fg@f+`1zV32P*C2NxU58v(ph<}_3
zYlM5QHpb>}Oh+#MZ1Id<<FSW&w%@vWEH&M;Sfw<J&z7U?sNki+q8Gqc!*enz`SaYX
zr$CdrW#vkZ?_Ns0m!;IGt8v?T*zPjg_mD0@4V)D;ST)rJm>>S)0yNSZ@@9_JH*Aw+
zKv{t~*0iTXa3{NDo1&2KyP4k;o?nIhxARS@+-*d<^8s6L7Wc(=k-c#V8y;fpB+OT-
z&G!QzkYQ6wu6C?pAxwb}@xVhgS#hz_d=l5cN}Hge+?0_ZHg0=$ID-Kgl8&P%-p$tA
z*6i<(APS|LHA`dUu!QYw!Ef{%{x4aXKc9s@U)B(MN(%Q82S0{}8hEThU;rEabawd2
zZL*#g9Ex}_z-yg=fDY<WKVWP;5RyiF*BB4e{4bj&NR$BcFM)u)PQQL*6Sc3_qgp02
zlyl^J*G!(1v8KPeu$2uXYg6;?Kh4Kzh&-iFhsVF&ori;|B<&Yjtx6S_9}hj+2W;eG
z<k@zJuN;-*Bc&BqwHu#>=Nu0kNKPMv_sw0q(Ec3jhG{hCkdXKVqvKJx85ziwICb{{
zw(5C9f3y%Luh%A@GS<aQIZK#r^lLm8uNJ@Wwl-7Ox22aMQp|cY!{=&7c5DY7)&pE?
zEGS7tf{kE%8oj5czrzmA$+?$qdD>Z5AiiHd)2{|5PpXX(TwZo)rC7zxtr3t+IKgk+
zPigpq+}pzf`7=fq*oXXKLMP=9_X%<>6X;PF0{{9ZF|_qPFGrQ_pEKW`L65fkGr4GL
zrg9m|r){cM>YkGO9&kYysUzCvV<Q7<ngd^%+~<FO=fta%{s7w|d~qAp1AGpR>%`$>
zUjBU9^jjP&MI~z4(hh#yqxt^h44W%9z7?fwe21I4F|}Uhucc;+5NAmvtoU_bO1svh
z!}IR2)`s7y8vo|Q*yQ#l0U{%)UVr2ezqt}<kzyEsDd>-T;ooW{N3r!$txgqqt1X*j
z-tF5&>dk}4Q%!baghp-`alu`m2`^6|8!Qgf3HiR;8jF`eCmKYBM-K5=8Kp-r?72>)
z2S@U>Gs;q!RQ$2(g!d!#Gq5+C%-LCFh#^R(bZJM5S=aU(e)TX;W_VRCVwlOWOFjMW
zFEgD54wbql1O*XmNGJZ^_z=0wtI9YlX2)#N?5(XpXEj}gf5`OBEVg1vT?~>>5l9DH
zN3E6RJZ6uIsvxT9Q5=d(e8qs-(}fLH3}^2NZeI1~2#qN;Xk!&K(i!dqzb^H>G~m4I
z*W>r`%mcl~MXLC1T83gmfRdF3UHYiIjb>hsg~FY!Gg1z{rJC_<dunWiq<Nw0(V5&m
zw+3w2;ez%g0W~bCQJkN(z}$&Tgfy>8bL_d>KtH)Ji`|IMawH7-n_d4jX6$OCDn>AC
z&pD>u0oTt!!5r}+A?6f;CB~YAMm)g7J->%XQQglC%vk34zjefmUF^h_vmBrxlJ3rI
ztIszNkHx6<)zm55maekCZW#<wS;=F46V<&Hb>M9=FSKmH*JzF3#HmB~GD8+H_csr{
zqH@HzGW6P5%BtH>59nR)&nnV~AV#?rJ0W}oFUR|qmaGwJV`+0l@G`BlnQuoJXb>1Y
zZaVaEn4y{4y_;%x{HSy$;46^qYoy4n57_dEn2@tn-xSE8DO@4<<9m+Nyp|>x13KK*
zO84AS=YCfMtd-iEc(9eQ`?`W0IB5ZAMAVHYmS*!Un}>=xmCabO_R`N$4=Y<NN#H)4
z)H^{CZHIbJ!aB4IJ$MD+HKMe}i0}Mn?&;anQ7dpD!BQC9m@0|QY64}Z!6#fLJ%%pz
zaH!^MRHk3ce!*}}A;kN7$>fKfnu2wvoHI*J{7FrBiC;%W;FR7*P_*w*^OQr?i^lXR
z!S7L38lSM;OMh>xz;GAK<kb9|ok<JXAECj)P$9q-t~lEhLXeTY^i3!PSRAme_JvE|
zcN1(3d+m0fRlhyz2^I$s5enExmKF|iOyZcaP_?MIBofG(H`9|we<Lv;*E^v@?p|#C
zJGKu0D)<Na866vLn6pR~53JfFC=rU4lSh^1Mi|p((`#=JJX<94AN_opB{5Pj{<#$w
z4vc_Gn=&A}ArP%bN{MIOdqmFkOnv>z4eU>4r{nw=1~q<)i^a{7=(LtmQxWcGlqqXd
zz!5&lTLw!U+f-ItX3D;tqmLm9AMh{K?nSWyD-TLihjBo4ARlSVuIau~x)6v2fUVZs
z1E;*!y+0xkP)t%+F5Puz9*V*t9{qgq$sxGwWjmL+^m69+KOi??sBC4~Bzqn=-LYfR
zjk;m;+SfCBq}hH2Uo)_%q&knW8iT+x%d)sYq#z3V)!qWpSW8s;blRFM1$6G>!oHvM
z8B48lU&FK}OF#UkPkpM^=+h>TKTwE5b8|3e{HVNX=6{%Ry!(D^5g~P&fm_^PyLF}1
zo351()t-VRIhiNRa#j$`<KavPTUbl6RhPvq2iL+@{T&ynz$#9;z%J5JiU>-6`b>BP
zSOB!&1?1s)O%ys*O~B%Gq6sr5WmK7trwUHxaziWd2y@6fY^oB42yEYbHTZVCky*`0
zoJMtgG|%@2o3;b6HdLEG2h}oQP@d#}{PvGHLLt!~O`UM;5Zv<=WaLqMSe7wIWQ3?b
zM3iT;?yBZ29B+$;1Q7cd9gbx=GH4c}qk0V;HfwD9tRxH>eBS;}{E;xW+(5_!J>83u
zf(7e{15;IXHH6QiT-7d>*m{l8!-7Q<^9RD!+fMnhWdK$#wv7Wq$l|u{eRt`B)Q<^;
zz)F)h9n*#80G4fG=1bJt?b?GHy`S!VrP8?&$1RD1E&m|C8T^YNZz`2*F$B##I*LGG
z+^I$y?ljw?;`M8SQUukP8pcia3T9_`%4BWUdLN11rwF|()hr@2<WaXSZ}1wg!1y%9
zwWgocO`Fve8DO}^xw3cN!Zd+fh)xLav%Y%39Trd&=~{2JOlIMs#*$0D<vyq0_`$=B
z`<`|~U>~y9lafV|EVgL8qP?V9PrC&ib`!T@Y{U^l55A-hO$b-k**;g5NWa;%K$41w
zW`{QwF$uY%Z!glOrJU4W;Si`UddU^Lx`Xb8E>yJ_$j+@V<qybzHhqx4w;6U=ubSZ^
z*SDSjdjRUabSeGaqW+5S<OuFFI2C2Od{#D3V5P?k6*iOr<b&HCG#$iZ4PB_GvAkfe
z6VOhREeVk^v+)Qt(N%fydQ@hxm8Q%n7h?vucAf&m5Gb=yhlvAgHo_6;by?HWX3WCK
zZX?4l|CWzMnn&Yu`?wBII<)m3&J3LwfL{R~cQJ-5L~7fr&H4TYlN59M&fb?NIm_a;
z<E80Y+Y8Xl5|9^xS+##INMy@Yov!_f-oQ!{_`J~2l^%s6ONRuGW?nYvJWpmyX%Qkb
zI0srjQc(+eh@?gxoUrl&h+XCC5?RhJa5zMxdvny*2{I1)4ZnNbM@iiL`W-<2Ye%ux
zQ?Yiax|K^lAa|7phhQ&prKa{>zZ%<@TV*3gfs4B?RQGkfOpdL7q1v|ZqY}(?bZoz_
zBgNrRXy5xt^*toYmy4jS<Uw7dK=|qr%LEp^PeDNOSp7^#iO&=da6)$L(c_1L;=z%*
zS3eiR6A}7y_)2yZN$+fuwDu1=h}r4dWH9Rm3}}DMD9(t}pC&+o+NxX9blgPaqGb<&
ztRffR2Stw0v+F$4X{TpE7fF|9ATCxy#M=Ex?j*#>;JU>34wbCI3#LVl@Hb!bL>uZm
zwl}HLnXPjpD2=jyEmWh5Pye@0kq-gP$VY4Zr@nUlXdPHxb+sZxvwSzWcO$YdsICkW
zHU<SNq|i$OsL65fS}qPj{y;^!{YA}j&`A??*5XR148Z#QB7P%f<*MF&W(&93*z9#C
zU7vQ42glH{3Q?$UyeWo7jvj0><BhFczK;o}z)q{|Jm9d?pj8DlRK@?JT<!><GZO(X
zrvzM7PNN%fatI@Fwae$Tu<j1GdQf>M(Rp~#_qF)u*~4a#BW9f*xWpiOFz&^rCFPwh
z-Owl(GCe2`W(TY#r+r7D<0`OCz#f{5=z*&Srv2nru~JWF#IQav;c~-zMEGf`IKKKP
z<i=Vpdq|?<(kmIXM8#6bXxqLZ^P?QP3~-e)V@)c$*w#=uf<_)an08p(nW=b7{FOX-
z(6A3=vKZdg3S{}vN*(GpKKKVP4le?JuU-{gOHps&9f@?YD{93U3@A`@7>R`9%j=j;
zBAR^q+ipf?up=1Zmct6TPwtJ$Q)*Yh)_W7pPkXAZuCS};Bw=<j<Whl|VM^pRDD~Iz
z9QURuJ*T{kO9(3q>0*BMtCHr}x=@vvTj0-fZA>Cd<W8E{v7(lfw-#GQw4_JR`zkP}
zZ%|7;!?|XugcYY6=|%U2GtmT_)w!zpg4~2GXHuogIH_+V96lXPPYyOssezzbr98XN
zw1mSiA1v@Wq&*Le#Bnwnh&D;XZ!gY5!ua)@u9$>?YZOy3oiZ~58w5a=Jo)P+yz7zm
zxW1bC)z*13Kf~0?(&zkdnzmWHo%+E}`kK|wF=3|4{zay$7Jd;|VkE$0q3+;KIscX#
zXZJV?+}wAn3^9>@Wb7h-FS8zI*Ku<FU6mHEEel>UeT`qQRUb0Y%k8xzUgALfxC2$;
z%1^xLb0|^hD)HRwL07u}`kl|2AqdqFX5VTRIq4bY{50vA4w@7Tu*{i7*tniNV}dit
z>CX%*FO%NdPoewNDhtUGZii0wI91>)DJTeh>+ioFhl{mA3in%9K<imIogc$0-lwdB
zcbO|nojAD3OrpbuT{D{Kg~C^pEF3#<@L}V<mITC&;i@t=ZhxVB%Tb{>&aK@6e;;=>
zJDo|ZFWteyk&XvUsQrzB?!XMG2<QWi-nV}j>+HYg1{i~#ibvpzLKQs9L?s%FsqtG6
z!|xhl?q%9Vi3v_kMV4anKU`l3xf$KEFj$%=p-TTDwI!jQ<ByuF*P^iz1?6EbZKS7i
zh@$)*&MwZ2Z7|ie06UR9g?m5<@r)~%cL=am461F+yyJVw*vJ}niUg5yjDN;Z)@!MN
z100FatFL0Ki-OCH$(^j7)9hLJc;R2nYg>_}rRxJc1`E40XZiRBgI%f{4F!;+`&%x%
zm*x)}l-{zoH>KR9F!6De_M5>tpr&_1*GgvuDT#)TN>iY}#g*pktDL0t;hh?ZkISKV
zjQ{|v`wOUNfUCZi?eHIamb1)GDwsqE#5c39Ks~pvjQV&7YGui(9odvD<U;)di)rd4
zrlmv-?O@c@UEi)w@y{gl@wR4v_zjoV4^IhN4rzT8ee@5hml+dGW|RP@OuAntzqBcO
z7D&u8<oBU&8B#_~)unS90GiUwCzh6YIV>LtdGewi0`K7?9+aX=GKY&8Av#RF=gg43
za)4YB3Mes@%*$<7c|#4V$pqt<r|%BSZIO62RlZrif>rdYyx;P%fAbM8Ehyvb`yagh
zfhNVvtA$$XseqD-*dos7$iT%l`DQ*ISpXSCKb86GF;@u5in)rhuJFArkWqY(!<!tW
zAQ@Bq)+MxKk};I{kgFpZ#W=;K;MM&olVL|5fP2@*tDqe#nzx&~F5rbTjwuK#ZrV>J
zq6lXVJNWZeE(6Ls+^fgig=zdrFyr77)uK~vIc++)Dp(7NepQ^uUz8G@*=wGfr5nf_
zZ;g)(XSA-A_u>f9qu9cycV{sbp(}r7hj~sZX-4L7o~&#fs7y2AY;Kqw$Pg^mS?|A&
z{TjLTXx)lf?ctzV@^LEc{N@Jd0ai4Eb!oyR8R*qwjO8*F!S1M5ODgR728R{apRNWR
zHBD~?#~nhS$uYl4)*tMLS-FLqCS*x@^6vX99IXJ}>te|arO<2>p=)cd`*ys=YRN98
zNJik2WmLs=tAP%GDyRrah^j#@XM@f3XFN`({><ILynK+#ft9*N&zG|!qIPXM*H-}o
z0fNuhLHbPYcyg7!{%VI3$kkg20QgzA+ItOr0|&nBEZ2&dJ(IByi_i3PLp50eD`ev4
zs8uxF*9<7YpGxEdKf}dv%G-)1O$9I#%kb{e%9Hu^mYrypoM~?6_&r+m`IWFre`VM#
zvi4aih|^#8Bw@qnq$3p^Mv0hAtj2!KNF0PXbdDzC^+?v{2SeuZB)dUO70&(>B>a5x
z;xj6B$)noocDk;3(YXEkw)gJ|xP>kPhl;Apuf(;1)2>d-kge8gVQ4qu4u)IjGIHXP
zf*F>~t1VV%-vI|vI}!f5I=`E|NOkHUmoLWfv=zB!dp)XS=U$yysvBQ$duuY;HC-RD
zwoV8@-Jbr3aev2z$Y2#;Hv_R{`k=lJUB9LDHV@4`YZhVqIXb+9R;C5BShGiVdL<V(
z?d-S7u-gla_p?rvh}hd-d_(Hn0-OCFsIia3%4!D&$#*lhN{3D&9W;Hp<+VSHllx&R
zH4K;()(k#wy{Nj)=W^mN*^P0(5N5Gk3G+bi5o6(wCs?Ggg1EJQT_Ad`8(<yMg}F@{
z*Q#G&1qQZ#gJhvk&Bm9=!s~%iQ@Wfa++@${7>OIhx$)-vN)OddtZ>|klu4fj;rYK4
z<bTQh@whLDKn{Z&l>Z{V^-)~5_4rNHDHy)!%ve%HpdgncW!9<del*5XLBGa`J`YF4
zDM3=Yd=@zpw`XGpR2XQcuf~f^>#}p^Y$a#l#ds*c{3<xi{#<;JF_Os>&oP=Yj~JuW
zjE@PmOMq1}UVi6zNhGM1N_XOJM~8XW<$R3bz?`{lFX_p)8K_tPjqepJ0;QpezSEf^
z{k8hH*F%}QDC3*wRR0N4=!zGeCmTS8q0Y)fD&7`MDjjiHz=y(OEuJ%#)RP1y`7v#|
z%hb)Qs6bQHzohf~?gsi<JABHB%N*~7Bw~qrc+ueGG>aiai%Y9{t<s*y&Pzf~`Ra=;
z)1xATN@Cqx*Q~^Fd;#DC1tCoKc-~nFoO(V4(v<v_YNk@x_<30hHgZ1S{gf-`{YJgI
z_@5;$4-fPSw|DBTb%(aN9K^MoVf9sO0MV80=8d<+!ipb0fEEXKn4KC#0SxSTEz-07
zXmsgNr&&Q9NR>MF<>#$y6Q-$K7ucFlOQw8lBMXj{uv0in&DgSeRg;mN0VmNnArzdM
zo(dTT{V|M(e)_r2^~k`4#&8ve=wYqXTP}VgBqkyOeBAHi9fV{l7U@(S+LTJII9gS;
z%7n5P)ktMKR?O@L6W?{t{#>~RZc6)Uj&jKc$vN9rp*rX;>|72?bPD<g?%22t-QX)p
zglr5)ZHHpr*h}|m(*^FRW%_*jZJr2fZGO5co!?Ate$Vv5^o?F5!xjJh#B5?>pN%}m
z5^+?~o98wlxNK^7L<)0tI>&!sd)d&3w?rm|0$py@>U1>#pZd%-yQR64uG$(E{#q^g
zSuOsF?+@re0}Jz^r-MU}HSWwIwUNtJ0Y_^VoL;9g>^GMJ?hCPCIYYO*ojx@k4yr_N
z+!2vX_8ND>o2H!tOPiGh9MzZIMlC0`humbMWgU*K?VWp^)d9}Qq#<iLFh;TdJ=k5m
z)~lH1Gh(8j-u@+2=f@CvFYYoM0u;pF#tNsK6rGm!BDB`M%BN57GO$Tun!NY>IOZ5$
zxWdC#Dm8BspLItKeS{!XEf=lEQe4stm|Dx-P4^2m|Hj_HG7#5B8Xd!M4S(2}!PkYn
zTx+G{Cp?^>-65BTUvOHD6&Cwkh`{-hOi_)g)Zhge=EG?6DS@UZCx#$Gp6#^0!g<pP
zM%Hl~0GDUE8SD{TH9ZGPj95?~0*A81AZKCx_4ND)cd@cn0OY2%>H(9*lkv@K-IN9J
z%KFZ>;V-rxXffTCIg=W#&~t~)>hBhYauW~u7*?wu|MNh{&=55wvXo*wl_OY830p2a
zo|1nETZbsEv;w{R8_NUV9Ll*o1OJ`Q_E}^WYHUsr&z0b<(u*d&46#)=zZl5UN*)r=
zW1~H<akoKdHpJtrGWxmLeR?yWDe6?6t2&r`C1W_oVdA;)FLDyXXfMo@JHY%8=IP@Z
z6h`;zMYlDA`#A58A6?>V@1TXK@9f-roLq&y;9i9OcBI?(@KAX-^~Lvk*oqJ1&iiBt
zGrgeBn+Oz)nci1!qVt<J96J+lan_lodC2PQEXrS<7Ez5=4IT_Gn=G*84%n?I$auxe
zJA2OxeqSgsSEc=^QnOss{CXX*S$RD^Ag+KtGm48m!<p6MZ1;G?$i3G2aT*X?eqa2n
zc{w+kU$X})l$XSVKHC(un_rL_{o=gkYtn6{+ML!wRsa@or83(+ET#n(V9Y#^g04Me
zL<~5tX2GToMid+@1G6l9qyK0Mwzfv95r*L9=>|XE<UKdNhsWO>!3EmHv_I&OEToBx
zV6?=r(EZhlLF%5{?32Xpg3Lv_zXfyGM#2*wj$AARiyyIrf&)iQWfD?9SVd$7Q>w^?
zyzF=b*C%0<rh5D(iUw5Zw!w*K1n$tnU~c?Roydna@vm?}X1B2FazQa=5yy#mty|Re
zQ#<&JyhDTN`d0HCl$*(jcW7}l^fl+{tpcby4j(8>_Y$mnx67Zq{?_bO?3}{3yBi)O
z@Cmf@EAwy|fR}fkgg(qdn)Dg2*HCged--Qg-rf^ewq(@O*9}~*Xsgckl<)VqA)2kr
zQcND~M7xmz9BXCO9|h{#iGlBnzh4d;-jw@T5=J;4Vv0XsW2~A#H0XS$I+-8s!0I5q
z*kjk`w$M@;<5oW`jJ^OZPyc_Pt1|8olw{e9v7Juy4^cn1{-gXnfg4K5m9@Nuf7SOf
z(>Myl0iH_9U^G7ld2DStZ@EFbY#CoFEZepZRr&uV9OAts1TFfD%D%{4TJKOK!DEQ?
zm1|XR3ceAg4%IpF*S?czRyKW_vp`DIHd$NxTP-T@D23EJ(Mw_xaUl-jSKD6-o+MS<
zuQ|Nkvq>eI;f*r*79dPUmvB95;2^&kb2;^#E}&o8cJr9X*g1OK$7=~{>hVCL%N=%>
zbKUF9iBt`sIb^TknTZZ<kwGq^j22}Xw+c1$%uZ1%8m_JR%rq09>24TClY9V)0-P6p
z?9Yixw!LC5UQeD^1npkvehsr%Udw-+W`|z0Cy{sSVX^IbR`XjL_BU_T<Zo{fb@H0Y
z)L+K!>(ZG9a4Y4sR@}d$b#&OVq|o{CC$o!D?u<u5Nesj-CW8q;N}!TkB;=nsk5$K?
z&zYJWXQq)1shl_xGp^4h>KXf+sHNDG`l`9d#4?+PSPy~LEGej8+GC-PD~O4e_L<ja
zQ~dz?zQfYw&GODM8PZ!M^ef+D-D6LO)?|ODtL+VZ@`Cd*P;lQ7!;-cTC6N2dDm4Wh
z5(TVeCjR=unhH;VHv}^@iY3A$Ip+zu+++MMi`}X&`$@Xak3X+Wi~-UZJX&EJs(Q7&
zP9sas3|J<XkU)WKPc02ECI6GJR_{^lDe)4KDx-+n72eWuu%ZKM&)$OK*98$lV1>D2
z?rNhA$Hz%)Cv_QBsLLe%4`V}ZUSV8*hB`upR9QUL-Z=B-fx56ixg^^V?vF9W%(*&7
z+mLxbwv3JK9nG{QpbYPY(HsqEaQthWC|J3(Qz3fsZ+!-0@bBY?trmZEc~91XJJZ^#
zWWPGYfB8OxQ<2i3S%|#+b~+(AdUp~0L@kAtZqde(`LR;b841s8Al3W`@)t;+=yoyY
z2pIgec~w~gEZywgcSlI2YvgrvcBY8S=IMS&j`uo9Wg`1nJ=u{0NF}Cof4q-6XkoiO
zZ-kI}gQ+I^I5+R^V0K|k!kFxf%HGoF|AxPBk{^+@#F~eQx4w`fvY2?^{Eo7Cn?E16
z&vQB)-+1C_)mJ`+QkcfK1a33+90Lw<WDP#na1O_xm530<YE^=D|KV+M&P=O;@f)8z
zpw4Lf;s$f6K&S7YPZQKr)hl^6gtKv)q88DS`!Y}K%x4(K`rC;xr5!@Cu+4IBZ(vaQ
znO7rYA+pHy<$|6q9XR##>dA%*t6?A1iyulJ;wb%j<yA(NGHT;(K$J<3H;h-eL~gcd
zyc3S8E?MYsO57=nVUWy-2gV-c0kDw~#Odsh2^vv^ehU%yg`|wxG4odKR|CMJ6;k+}
zUs*I}Dfra&*ztXJO|tIlzV{Q%7qS&TuICp@1NASyX2%_|3JTf{8}WNt<_rFF%p`9&
z8XI|!tffiN@$-^{WWjJULx&tc5K(5fx_8q1^oA3bbrv#4PG!x;TdMNpz{q_NtYCB3
z;az&{%z!qSOO}B?na0oKC+Zf~(_3yL^Qwe`+l<4V(=dZ6N>h5Z4LsP;CAxX@oGve-
za=H;=&`8@9PI;ybDu6J4jT5pFFpOTR(~5#f;Kcn=+h)Dx-94O~Q-g{DzCBnDch-6f
z@0Ny82lBL4s{gD==XLt>xP+(<^ao5GEJx310Fe9RSPUzlSm`kTf#J|qVP#vXfH!B-
zobC^&srxsU_?f^4JDT@z0Acd2%+7*irH&5ZQ^D|L>HD9l%1&dIBCXI|9}YzgGUjIi
z0t_g8S=A=vvvuY#$jAB+yQPHmN0_p-mMO!$=J^&L0wy3FBtdq_&}ADjx)s0u%#;8}
zMYTIDcZ^TEXr=cXS%`=Rp=M8xcVXe~!2aX(j@}j<N>AAVLJkUmet_<)no0MeOSNA8
zF{~JulL%c;(RECL<@_crsa#~zvcEKM|6vbV5b8%Iw8_dQ4$g8pCib37@@rQN;cvQ!
zu54=wVM#OhA}^KafR_om&n0!7js;4u$p91sS>!=>xjFCm@>nMV8}Ga1b5e46_RVc!
zYoAKU>3`@H9yfhM{G7c_u9lR#EHsdp1d;w*5$?VnqxQWV3PF1N`jamUY7LFPXNhfv
zx(d#aGvB@t&D}*Zw6z;{*0QYMI4TVI5pxH=MpQm6>9AkNDv=BeIS(QGJtzRL9<cOW
zp22@?_{I#b8}X%f@Bet12TN;Ulw2lt90md<Xi2RqSiq79y=2k<udc5SYU_*s4DJPr
z7cJJ}9$bqQr?^ujNO8Afr4)CkXj|MJf;$u~?!jGyTiMs|?9TjlXZDZGo6I}+-aJWi
z-Z>u`FaFT+kQ4^<QwSONJIL(cbN`deKfnqrWwHrZLTIBK;(*;-$g%7i>a{^|{09*o
zM~hj1VG=xMbCqS1b?4+PCi^^-)jH}=Q24>h`uEEHD_KShNOfesdO<lk(vv%yANp2N
zl$<vD#_CnnyMwC4-lYsS?T;n@YXtU^Esc?~l(Y4d@^<0Dp+c%gYW0hg;%RSLwsrse
zh3M))!~CcL5xQN@m`f_w>7(FumP0fzTN8{L`BT#?Hyye%5>>v);+smEMn*C?3OgjB
z<lp44ebvnm<;`wXImO0?=2h9!cys}bfy$gVGnFJR)IR0X$eExMckDb2rw`1mN+9bG
zibbV>IBh9oQ=L_G45N#(SnPux3CF9k!Hvl;)zq!Ndhg@{?7SuU4S|Hp_b&6>Ltvdl
zOyKAG&>)bkC7LC9I!E6)hO#bPTKX#)PV$nd0~?*dO`EBV?*qOOhxA>%*L6zJ`3Mi)
z&|~en()t2N#{=$oX250Csdh+f4u+Q@Ms1Va{`x?|-CggU9H=QRJ#8!DV`#jSBxtSp
zOi4@--)WodnC2b4F&ns#B9}_V8(bWomP(jgE1USZm>BUGRVgXPN90K`b=oHRc2M6_
zoKMX^Q$MgHbdsmcRlY!5Y0In)kFrDL?TT)I?(OKNU@ICQ`z28R*LeuQm0yF@bB}n~
zD9_}fU<z#DFh{P-e$~|GFZ1jF89sWEL3>9B;r$fh)tq&yd@o9bR7oZ&pPx(q?NjMp
z<rb5o)uW57(H3>ans*K4!z=aXhO4O7&GY(}O2!|UFKmbo{Z>Lp$knCP!y`F+jwr#E
zf<A4G^LrL%!4c&v=F}HWMj7%Bf?N|9luY#iU}A~PTn?ja{l?lh=DW^AQGysk;3iD1
z-wf)))K?^K#&bYdiq1=e9Eqp98Q&ftl9dSag*hKVhhr^Gd<QTcEF3yhPO2Y^u?_<O
z?N$GB=Xk&@4+%T}wxQw=i-XCGdj{SYr5kF>h?e?R3VFQ+N?DKwhzuTkbcw<j-bewv
z?>s6fQM8<vbGY$|$a`T?8$zybs&jfP?0H#`N=T@)bMK_kLox4f5V4)8Q0=zbdDLIB
zi((a0GiB4jd<3I7uYSHQLHw_7^$aQ#0INf?h(@SLGB0wGzpg#b#A&qW=9etM&$(i{
z3c@_nkSDlp11C|MJLWUgT|LJH`M0QZG5+|CEcZL|>FBhUVQ7Q33FlWuuZvAYo3`oj
zDgxIsM}*r~>jlj<6$0b2`zDF9-i6x(S5AlY`&(@Lw;}P;4cqI=Jp%{?WO?jehoSvE
zZwhNsD%V)sEi8{I>k-Dnk{46BT3_^caXT>4df53|2PeSwyB6x%gV^Nmjg1@M>idlV
zrHxJw!_~0-;}7mT@N1a1mY1;if~*Or=E_M>FG(XVNrs)d$1f6++VPfz2$tf(<Cg%H
z;*bAK>rM&G$56j!9>Z$mGGR`OK7A;_$Ja4a@L>qyH~CV_wt45zJoXKfcVFe#>7^%v
zjmxQHY*+qO6Iv_K5m{0qSYYzu*8MCkn?X8~Hu3E8rI-)Ix^rVBQU0L&JODUmZx`uK
z)Cs;Uur?tk<mu}A)TsjEtQ-q*4XCp8eW!I+3v?_h;5Z_M!`8bWHx$J?JU8<>+_8D4
zK3OK0`aY>tmA|R9;oEc+^!CVOJbF&Hv^DKRSCi&3UYPZ)e%betv#VLZ@#l@#O5E<t
zz(uEzMc}PwRs8sw)@&jIyy-(gUrn7ENCDugrJmR@CpX_;r|)6XgdsgHPehO)l7VI`
zY~cgizm-kAV#PNZd4#9~>I34pkJmT$`@?~%NrEB?YMh6^mE9;&1BR=WNtJix0Kdx1
z6+z`c9Zg0*J~m3rtDDJ<&u3n(a6Rw{I;=gFQbKl^B#F$0$cSoBj(rFnqS00|JpL=c
z0IlLfg>V!j(&6nZ|HC37Q=&L5&l^)Eh!F@xNhmKZsX0KG!d%NAFHZQw(v|5?3FXfq
zO>OZd`rJJ#6}L4`Gn{mraHd)B*U?OiyF3|POTX)qbW;eZp08}YZUUZdwE%@$Px>{g
zw>PtZTNec8b^mR*@r5yFpfjp>QOw%hV4p0AUMCSfqr@o)xn~Dz{V{=DKqrp2aCpji
z8QhSBUgP>!E1sRSuz#{ScTK5bCd9D>%XfoUlADk#Y)+YyN4o@j@bCRXZJk_2i^ThU
z#^{@PQqRN-9s>%SA-Km-HPK@<i`9yF>1&^swM3jl!4>o<D@^KzS#seJE@P!t&Xsbg
zP*XkMjMYMKkMno;lrVs$%2LE;e;qw7`#2|ycoHLD_5q);=##PQ_MZY_J-kjzQ~^=O
zzEiHiQZo<!1w;JV%sq7ojyi|n&z*n&^2YrAuZsyq5ce%AIB5j<8^<=pt2G;Z<Zi;2
z>byXSzME8>zroD?qyv!#4}I@N+DJh_7ZzyfcAHL845JTj?%RDg7<_7Gu!-P60mVMT
zrRs`~rmpZ|p-5EaEswZ~dOu47L@o|<V2$e=3gTfBp}!dcwxPQPKU1#mWIn59jEc!*
z)xhZ=|8BurQJmMwGqJ|?`Mb!ehYBKX5GN1TtPUyW+>&nLq~F_r1#88Aja1iQ&QpOS
zMb*?BV3Wc!jdgr+-^dP!?B>{qs@zpuKA*_ojqX6VJx7U3pA-n$iWnC5fyAk_KP-ho
zonJaQJL_89GRl-b+h%D+{5lV?`O#TzgTj0Y+GbO(C=@uvhIy6-&Lbbdo7t%kT-lR~
zIc&m7dZL;6Hrowm#)2S*<d=^63nC)5)dyCYTz(r1fR$0+xtGH?AdgWPGO65<+=Y3c
zspSX%(o-o+vgsj3kNcJ<i%@tJ`LDrp%b%ASr3agqc|3#m(X(X?66(rK@FNjGUYF%s
zb)i4hYbLEkF7ZQWp=!t0w59m?%#W+9>Csx}Js2@Nt-1ut_^%V0tSgXQhfCTBxe=^b
zOEyW`&fj{>phTY77GZxL+9eL&4oLgGm+WW0@Z7xHm#uo=c9I_2^w7#!?YiCO%|=Ul
z7S=kzmK}>K+M_@e3?dCwGTw#$;(xxltE?x-B#G(eksJ%e?6~oK!)8NW<=Ak<*<e)a
zpDQ^DeZh|wQueHpVcu}s6Iy_A=y-JTDc)C>Ab`WpTs$0a@v2B3e5aY1^qW|ZOumon
zk8rv^eUoNHm)h;#9aaNaJEWcDvc=V7RRI~hk2}@zK;EC%fRk67@Nv==kEMdrR_WK<
zp<CAec{IaUCN`)$iydXQZ^@lmqJy;@WAJ}rsZom3;ogaMzx|oD{zuaz<?ikoPlq8W
zEa{dXs$Cf0(nc&~mB`xnUzM?|W4#R1DKU*{nE8((<%zQ`t)55r4prmr8XY+*(lt(>
zosk<2!qN4393<Y$=@>Lly!Q^7nD<Eww-R8FEOBm0=4_TL8!)MZ75csy+vjKp!Lbz#
zw3*OA^2o)6553=sVS<(U2R@CAW)^oIBbLp@nZ_OMuD2J}79Q(epWK&j69v&I9(MTU
z2o;!bml5Bv*_*HkbZ9tf!>Ngf0{*rF-Hp2J_1<lTAUUa3Y$U2TKisM-(&rBk<1jdr
zH+JXOGhA33GbfOFFHCBWa2uO#i*re-mIOKj9?pnox#H58Uk88hCzq{@4P_qtPUJZ-
zh%KXrZTLzZF#S2fy)(SAef6Et;yc^>Rng2xUHjg~1ZT@wIjz>TBnCo`7&M(ep~#-i
zef}86+K%N`t34K5Ph1|BuU+l4=<~M5O{JRP!lZeg#(aGsk-Gmq+v3(Doo4}Aft_!;
zOF+;ic(;U`#T|GF{&bWFUbPQ5$JL4p{{wW)4aF~T7ZGATb#v0kWEeeU(k^1Oz|pF#
zXgo${PtGIGrN8J_2fr#q{Eb$n!?RRi>@unqvOEK@2F4TAn?F-YMr;+N6xmXQeo*rg
z$udVtlz55obLerMZ~;=LDI~MaA?7?biltxryyfA%VqsgbjX+AFYEB<7nl$-hc{U%J
zAe_E3>rEuSG;LPwoD7n!l)q|)=5rhueXS4xftdqtZ5}dos??^#v}X6iojGVs87sGw
zHod@ggLT=SM^=%Fv~)K^nt=4;2bXbtr01Lgp&8dDxm58E`@K`qpZlTIdUG1)!qe!9
zBYIyomHFy3l3(|J{OCd&RF@qO2<ffvnTY&6K=W_3wLK>*+ImHyxbrD{W&SGp4fs;W
zMvC;>xf&DOpd33Kec=msp$l}DKO+OML9}%0D)(f;-{0>apxtBqQ*d9B3->m3{!WP_
z)>u%l6?)+G@v_o_D|@Z`IeW^Ez0&h`S;S$CGC<UgtGW`%-~|L_awz|Ss18Axkdp`$
zoWG-Dse9AC7WlunqMaso+nM$80d-Ykr;1KVyt-JNA0yp;BgzV`5B-JKJ%EFn1(r(%
ztrTE~SpQLW<rA%X)ub1$KduCAX(D^%Cvg0bMhMLPJp;I`!MUK@QUOs*tak^j<L_ws
z?$X3TS!vM<x5wVTIk--qcw}y(Oe#>QU373<wB*iV?CeY1+d9Dldv!^y`{lc^@wWS4
zed6SPL{#tePvnPv`gP++I1T*Z=h6j!huQ1zPTU>zbI8n<*c_s7XFahEiB%!6A2u7-
zv5Kx2W^tu58)X~2nl4dGpI5Ng<i?9_tMF}Fgwu~g!xUr6cYy~*_+;izNr#tchB3H9
zBz*=}<+FQs<=Hw<z%|UqwbZ0^K=^?SevJAWwfiWw0oNa1b1G^IF(a9S2hGsaf(Ex4
z2;*%xc!AE@j;A)VVMpIrmh{uuwIqV905lYU4D)tvTfU|J&XYygk{s`Q%$f=g&~5}c
zXFf<sVLNr2ur~s#otDT+5zD~s4D-7=$R>cfUU<-=%426Dz(MNBWD)jq7@(GiZG_E7
zqER4giS1xA;cRtCrRk9Kg<po$kC*&-+E%Nh>%$~)zK;mh9Q*j8LjVx9<I;{w4>n``
zohj5&!?EY#*Iv+llh7ag{whZPTwhAp0I98@wj#{Spgy1~kGpATGs7Z;S<ihrL)Z7H
zJLLg#awwOS_o-o{FL5y3A&%LcKzahmW8a!9gRsX)`_1;v69k`vz2D`rzg*<3Y-S9_
zAaM$RL<n4qil~akn4*3CY__!9Y*StPNl_m;h6|{*V){6}a}U{-9Ag{+@3;e+mUo^p
zZ(Mb73)Wa_Tb-Bkkn2rPJ!A;S57n{zCkKpuq?R6en`puNarNSbY)GB=vIMk=Y|~10
zzNSUkJ`bKZF&;%`7|P%L=!2B{%p*=*e4>3VNK)ItdY-Lfx+#8f5ooV^7C!w46YO|!
zVg_*<xPX{wM{Lpmwh0|AeT0PQQ<b`{S|9Gk1{|9g@;8V#G1(-mz6SlyE*6l{!z{o+
zkX$SE>`NNxBbU>-JGFqr1rXO<BUeN0`2sM!Ll97ne`hb{3=O8_+lU2J)pXC<J*^(U
zB+}c-5?p~Dp)s(0k(VFxh=6U}gyG5sXV+(*Bl+~hCj}dJ(=5sXAzX#jPaVX4#oXGs
zC{hIZF>gWE9<DQ|q*Xd?BC(1^C77qfQfJ4OEDgbnE5pO6$yN&)6+;Qhkzt%Ezu%P%
zHOlW`@+J&^c*JW^i_9!vAs9HNUeK<t^*qv`J6ECvF#{^x`8=g&DI(t?_6qbr9C`V^
za}n2!W7R&bS_aWwJRduV$zuNk`5cDxOaZXG)&}qfOaemjI(w)^_`YBKaisr*A9GC6
zb}EzF0F=JWn5=*oXlG0%o-*b~PqS9Ytyt(6b$s~8g$4Fx#_uf4*K@8k|3iU-m8Z1$
zz;R%-w03!H_CeWxB?FO2tPu^8J|u^jQpTyx=u7ZSa!khIWjCwiPJbFe121Q3zHwC%
z;_akV+%<UfRGyB!{T~dtHAT)cqx<DMKa6CD%F^5Txls)Q1JYb_%$1&S4dcaC&#Y8(
zp60a(EC(V9c9a~0Sv<va?*3_UlRZ?DOrHVnE1-Z*eHlYPK7NSG+cj1D>W6s1RK&M8
z$?DoVC>iXf)9NV&3H5JkI^zg2@0Uj{7_FN4@Rhft?zB^}YNJq*I*0lD+RD@khOiTt
zw<hDfPWO**H_eyk8Z^-=E}?e?#bBupIrCo%+X7|he>ep%l9o>2Nrx3;iBt~Dx0E@&
zw}1-?pF-tw8NsA6;Za%-H=&sMzoeee_MrU-YI4mX5kXgMKw04MPdNgr#~@e6=aOfR
zH&s!&^j)K*7iwfvgG9fqRBW&GOmU9-R)*T%tL)oG+(D%2>&D~ej^R7>5B*LezQvjn
zyn7#-zqB(M1m^v!>GR7-*`|maB(&uGCXueoUHmpPwd$n=D+v!mE;ku1h(38^*lMoA
zwj9gEgz`F{5eL;gtk9JW@Yt*HT}xg~Y+k#(j1l1cU>V9%e6}e(M^ERk8pQL>u1n&^
zX(KDZ+hy0l9pgr^g#a_KZsspR!7|r_ms1zh-o@>dpttg>8dc<1uvX|SzqB8qw`;`k
z5|UuxK)pEtj{e9SUonY8^fqT}XYtHT=4@dnjnaCh4}Qn`rb_y&vTuJWs7{THAUR^k
z;W%7N_v=i`NZ-+%FOmFf4v3v%-#}vX-NzB5{-m~ZCI8j??7FP!l*_a!kMaK=hqdQf
zNcEh8uu`;YlWTwAj+NmC%pGso)aWs}z4HNX=GDL5DJN(^3N-6_Vj`?I+DBfrP1pFx
z-P7SiPKPT$@~(nYMZ4#m$1TS*kIL<1R#7xEZKLR%H@xC+bqioA(jltDxy-J-1$0pX
zRl0-**rIjucK2e=Rn2seq*MQ?cQNj`-H{{bFMWp-LN)1nVINr?|I-*8-5GqeDFq{P
z1?M2N_lU(&oV6qYUe(uKwY!yKJ}OVCUyOA+R#x&cPMXhd_3S~l=*~=Ou*cO14xW|B
zOwl?Abl?6OE#M)Y;MhSGl}3Y84gZYYZ>`h{qhDb23l@)aD1uzsEjLmptj>k8en$H-
zfy?5J2VXAp*1ltm&+Oa$vh2c8x4kM>4PH}2cAT1z09xS=mVMZrXdoMd>9cZCC{XVO
zI4OL-j?v1(ip|2GuDBciB)xj|X4wm-{dl+UFw%Z6Y53f+l-IVKbGxbG>ojam*|c79
zb$bn)3k($t^)<QwJOefQNxjTN7~rVF;p0Uo3DbDP`iZOvD`iUo4+&uTob}^N`(gX_
zoNF7U>gxr5FLmVgZdaLzd0I^DMQQYBDN~wYsu?bxR0ND&pNk@q+bJ=8-qd57N4?Hw
z9|4%sFx8WQ{L_&{E1f1P7#KV>d)3a*YRYyMf1JtFM;bRoenl=M*AXyq(E@AFW!q@E
ztInct|6Ptu%+eM1or3hHby*HnC#@Y|HU1vXO1nWrv*j8+Bi9Bi!M>9%o?q`XR>ViP
z{=#!UJmBlp-^~o>&8`4gxfL$9^{;2r$)SnuOD20EYnfb#oA+YT-=<y_u{B#@11(Y5
z^Tlp;spn^Hc1PCgfjI;T+Qjcu;t556_fEBgvBuy~(@H-9NOT_^<jGv-SmF!M>TUQy
zGM*thsbba(TGfSl>ntR5*HqrS`hQ=K!#LCN`wd1tPe9T$xfu|Yw8m8b7<K+fQ|z*_
zPzc~uzCG_#0&F3lwq0!(0A8%?-Iu@Q9Rs4P`qbeZ*&lc`31(|HZCvbEMGdEfi7Uw{
zlzoEzVN0R?8YBQTmj=IUbCSFE6^o;wrNbtFtUc}vK*G^MMivhgRs5SNt>WdF#N|1x
zykO<Y+-vQSq$y{S8@#!o?8{EBy4#UFL>3hk<VthoDK6B)bbrMYlQFaKYQ0V14%N`>
zFYWSqUu~I$ojYf~q8%~6Q`cLp9l%^c^ts++cWrL6S&GOU+aZ;-W$TwWwTQRFC6!hg
zu1*9)?Li5~Sg#)J^N6n6Vb^%2Nb^Se@g|OH<>_a0?$T$Pc&0G8ay|RYFr5$)YH|C8
zs9j?4J+$)g70>;6KrKdpB(-G&e3YKoc_K(>M;K#@n?As0yuSdzK|JYl>*&$fv3f2)
z?u=837M|$T0vZu@@Bef~84SI$(|=#2%bAMb$$L!jY2l6P093p}%JMRF2Df#7s?PtW
zb3jAyDWbI^&8V1YX_BiX`c~*O)%bY7#=gB40`(_a1Ud><{b|B@vrMio(*8ujQ)|}k
z;VpMkX(EKmL!>8w+{rE2gM(_FnOD+%bBEh@UPL{ECZW!x?BtMn+1GM@bOS3ZKb3E^
zTNZUY`nJ+DM_(I~zkRWBW8HJ{ou$&=vFtPa5=Bc`3?5mC<tPTBTCFQ{$T1&9DfBft
zAQJsq%yAX-tDR7Bw{;s+<Q3clkYv3ZTT-!m8aByZP4vBCvR9Jt!*ROqWbsQCt!}um
zd>21nbA{5nXzlBFsbiXRc<q@v*}e|#s7&to48?sC?+AQ484|;6X2@Qh2x-g~scbf<
z`Fm}Iw3!o?4`;&;gq23u&wd4DfC;RY^+~Sx6%yJSYlW_{o)v8!ZbidRCc&4G?<T9O
zo<28faK+%w;%49Ia083OpMO=$4d>yP@~%IObe?w>+$N|yJz;H`NNQ6d{Ato9($h_A
z2j6_bc|ScVo-Ymyb35SMze8s)+2*T!Z|VGg7T++_8Wc&NKQt^xm;kQGk4t?6y!J%_
zu$&E^Pr(AGY0bBb1|&@*f9)m}&V+gsWPI28OR;XRD;?iyKtSXE&!ZLfM*FkTS;21{
z4=CE%v-tHT6IB=fYfF?ZI<jYcco?in(eUJWyd`mdZ1oXoU<-<&&M^@Z0OTqGNMNO)
z9epTfl$^iobN@{*cL+xm(Ctt-9_8R2!^7}dzk&LG{YWnVA%Sa0L;9iy#m|y8odCrs
zNCGdD?&akr;LgRL={Is@kk5`rGVR;653V(|u@geEe~-~6s1dm9iR&nCqp^Dij|glO
z+Vac+XAe^43?3w2lXQ`mbhrC=%(X!QzqnRS1;<hx`@50vsbT<&n&L?I<6%uMY2a6j
zJu)WboRtZ8d%5R%D;_~4sz%b5+umks0f_OosCqK`$7(_QKg{n$oyk6*<&W|3AHB+4
z(Iv{*o$XCn_ws=-TAj(4&h31OMHUGee6~H+H!mxxIBUEXyy3gyay=b^7-Os9ks1e`
z70}2^DsHDY6yQ#Qd*c>+b@`N+rYC+WThH{?HGE1WmM-P9zel2<jtH(pssBBU26yYW
z{YbhqHsY7<o9$cqx9R4Er<XNy2CwsMrK`h9Y+~E&>cbMu_|PDJ?ZohpSR98?8`hiH
z<)@iDI+zB-Z#ARS{ttdQ47}=<Z2??9C2v~J2N&9Xe@oD>5aKF7{H+;UQDE*u&bOVz
z;SE>R8Ds?^;)6XXa#j|y^emtWiwV{H>LuyLcFqZ#+QryLsISskLh*1UJ7VHsRtM0O
z?l`HJ+Um~Tb<1gQ7+ea3EhN&(h88ncG3W>m+ncRNzzjrb5~}tOd$XEjnS~hm<&ocZ
zcCim<qfF77yMs-w)dK{!G_I;Ad}W4Lk{g#@Jx9qJ4;3{Msr|Z1v8j-u56Z_g!MoJW
zX<Y3yIV>tVS`q#uQPB6CL@)L@RLGZ_{Q5?A8p8%nEq(4!<!ozZT7L4IkE9$-p55{W
zN%ML|p$FK8&0IFpn->;Dt7PvD8`K*W8!p99iT(sR%dNVl>?)&)Dd*;arx$cdeC$@T
zjsubvh<X_!bO2c~#_Ij(O&+n6v1R<T{c~M<0bvt2htu&49~3VmxZ-{rW3KNI^sMX0
z_TNGdE00mqrFp(QzIOKyzTM@&EDUWO@Pp;?p6|-1^8S|Ht&CmlJ55xgB-F2mXKP0K
z^7NDHiK-Df2CvVdB%BX=BN{70qEsZSNR-L0u2*XHUtpTC?~<?>F5vG%*|g%^$2<J@
zhDj#-;^TA0@v+c$aJNRAn|~O&)|aQDhs_{IdC}jhs4m{7?l)Q1fLR9`8HL)bbB{S;
zy^iD!<Apd=gFPN};4~>bLhgsPh+@@_y4~sZ_Y+M^Rm=5$3uaLWU51l37ld@lY={dR
z??H5->>iL0e9@Y;uWCLawil9{<;B!1_Ma}pC10=<ehbZv8P@r(?hn70D}2aIE4)QO
zDH|eW#okg5d=F5}-&-1vkVb#qV5ZeaL8YHc`4WnRZ*-}B-{UV^%IhnqX`a1E@%BVm
zL6xDq>s*~!N1Ns8W$<vI+$X3uV%5vf1^gz5@>fN8yoFXDIcL;^N|@Q3F>g8Jxa>sD
z=M}@NGzk!?E;w*xsK2FAwr5&4slZ7<DWlfEpDhgzWIEVSoJB%rzwO1On;&9<G-OJ)
zQOy;DUcaY9QJXlQ{<Neaju2Se)+{-PQ;yN@LVyCAN_o5ZGH|;?<XSCdsc27MC!2)~
z^-Ea3`g10KzP)F?BhR<C<@4!Sit%@2ZY3?knoi%>FlA{ws|>kWJ&6PLBs(gUhV^jQ
zN<dYJ6GjYrZ`BOA%T7$Lw+H|QVAzbeSQ}WU?sSy!=q)-TjRlUdjqVtoZ>(pf|HZ^Y
zaL1DW?O^E2EOm#+(ywgMP$(7peZ3=z(#@({SBcihmO=;snmC#2^%+iXiUa$+01$Ti
zcrrw;iKc>ADVsrI6L?sHIQ4uS{IelB5BCRFH)V?UG@Ldiyw}KqC>W@ct_&$l75B|H
zOYcdJ)a?jRPvNDz!l1O6!kp%`EX7X4*XSryVuEq}LV?md8HvzL+d*VvY33}pepHv)
z%`a~)TVC#}GngV;rs!0KX(nK~F#K%cfMKq+C#VT_xwTnAQIBp0=pT(CDa%TMT2+w5
z{Ss<uf>i+)F)F6^_O#2)<4a4m*{UAB9%HG_my*ciiG#!=XnaP(kI%U#0(Lzhr+KlC
z{^Sr{W?E$lV*AtRhg8*-htC7mspOmMWYB`ABPkOH&4AUfEM)IaUYA<$NwbD=DLW%S
z-50dGnQq>ssEW1T4phF1AVaPI{33`1;*=)7LK`lTd0fYm($CAoMK?1xZ6(f7U3LJx
z+U!$=rUFj#0cg?SP%hS1*$(eZzfZQ1sqTra-0N}&rGg*5h}^uzNsD)qs&l{flyS)$
z0G&)YagIdZBQ<kyXP0se%B515in3j_W7fpWoF0a(^zOc(?r(l?*dKk*mQY@A&AAU~
zsB=w`ys57UACSWRVB<ghexecf;q1#R<Y~2qgz>E-7kpeTEfciQ?LpA@YL-|jRdzS|
z3!4Y7sp?%7JVn-OMQI3c8PdPXdT98$9qtv}9{cBjV?kaCs@Vtf&l`IS?F2MZ`4<r;
zBvQ<lMg&`4E)#wXq9+N)u_h$l$r0&Y!<Ny;_&8_*pIQ!HnpG+^ne`R&glte_4-e%U
z*5=y2`U~F~I<9saWj6Zhfr<ymu)y5;Hiw)2G%&eeUMtH$=V`y_{ip-NdvW{uV$q<T
zp!gi7Q&9DQB=+aLxJ%;z{yM4fmwLxsw%7Lj-x(~%&#g~c1m8O*D4ruWheGcsWe3mc
zYh=*A?MEp}f7z5mfQ{jQfbF=g#ups7w^*7!R8u|(ndtK7UoB+2IbWwo>Rs+C&sRtb
zIU-09kScE*ccP=XEl(zN{`!D%%jZ+apx5X<5H$6a>4U9atWe)#d|s;=!TY+kU)sWV
zDQr>r3EB{be)dbK?q>cj^r0q29`qTC7XeVJiLd9D!R5!Q*-{s4LU=e|H~FB~Oy@n@
zNT>*FA&CakU2wytmZY%35z(w&RR0O@*RC%<dL=YxBxQ+#1x=^Jps<&(wf`<$ADYPl
zp`GKm)O##~R5@Am+Rr1);Xwl-x!)4ZzC3g<g|=k#Q0{B#PAasrXU{@&1@ZlA_67ls
z!d$P8Z2{kJjoGFxWY;Fp(Qjwl5Y96Y2xR0Yqw8kwXm1a3a0AKO*g<Gq9o=2bAwo3P
zZf;IO9321lIXbwqd)nCA*h4IAOxYb>tT_I?<M<aYq~K`n;A-;T(aF{Z*lIc0ctBiS
zZQQ&~>>RBe*_|A$UdWoO0}C<yTgbxH4I*Uj0s%H4ZcZ*<HZCqUPCiX8ULh_%AwC`!
zPEH|C&ST2g`v2Dedq)c!OYi^xfJ}+>9bka$zZF~^E!{j#T_7NNds8cjvZ<Agxh)Nt
z9n3Ds#!vI#lAJtjoM4&{T;Tt6gr1(B>^A?Rffo?{@A!hIYI}egN&n5LY3*)rCL|>7
zXzu>+Bnb(r0imuSuEOUQj{hg-zcqz~tRXg5)@~p!L9P@GI&mO|<bTZbzhiW4EZnR?
z`~u)1ouUmOgzSHWC<7<j&D#kA`tMV50OwG=MqB0onMYX@hyro(a`ABqf(7|_pquEw
zfDnTJF^`6g4+I3}WCtrsNx9^e9s;|0{>N@Db$O8Gze~u$!Qp1_#9`-XZfa-l=;Up}
k$HNJRKsY(gc+D)>xc*(lG!SVKFdQT=qas})_3_L90>k0aG5`Po

literal 0
HcmV?d00001

diff --git a/docs_new/logos/meta.png b/docs_new/logos/meta.png
new file mode 100644
index 0000000000000000000000000000000000000000..33e9e5888434f24ec087347b62860814245030e4
GIT binary patch
literal 20763
zcmXtf19+TW*L7^Ov8^_C(pZgc+l@J~?Z&ok+qRuFw*Tq#e*ZPOCK=@1_c<GDW9=|`
zS#d--TsROA5JX7{5k(LXaAM%!xiC<`X91`6P~a~Jv!B0yf`HV<z`q+n0$&pwNhtmT
z0r8*&0RaSpfV=|V0-S(=IDZELInxIL;Z6ks!M4w6Q{V-D0A(mGE&}rP?~&74oB#qM
z&nzkOQ`v3pBGcVXW$~eV+xfCn@*5<E1Q-l$?0^K6eUnOsaARfWgBgM5@hbN;7<R@n
zH{D)ErOLWx<<g^8Qj_LzEJ`FfARI;##U!4l^(pJ&jf@kVa`*Cg@dH2d1sS>7RsG{y
z|MB0>Wi-bq^sO^V>+x)qy*J`~`S6=>Aoqf>xiJ0!azK(GUJaDG-wE{dEoh&Y^3i^X
zWsC1f$hXl$v5VfZm2bQ*Jt@qAjyW&|Oy&YP$S7c{AP)L8Lhxlo{`7(eHt!;Q2a_bU
z4lE@G-+~RkiM<kQd<8>u+?z(|Qu*V9X(HWA)mdozaiDsG!+sDA3U8CscD1Tvr_JfY
z))CbZjk)>}O1hnC?y&7p1Y+OxSEm#IK<NJt`~i*klFeREX^fNh>sJWpc3}l$`2|0=
zEtpN2%XHntem*hD-iN6v!s`Mv@<UJ*gn|+L1J6w^C)XDx$;Ardza<dRi8qf2kbPxI
ze?YGI+-)<^6um$kfC@QsxFV-pT{dNn)^_^+27sS%N+rQ*0xZnkXcaWfqYVFTB-j9~
z;4Y(PhaN2~0tF|vs7PT4GIl+c{ckn#pY%R5AL0{=9zqM{a)>^?>j$_zcV~9;|NUDk
z$SUXXlmK=pX=dkg99B^&WM`rUS=KlXQ##r@_#vf!F>6&JDK9v8P?$R#J8MDo-dtHf
zhOKlEa^S+cr@?%rhgHzMq50@ac;GWrq5hOdZei`Ez7=Ty_s9dZsS!=0kI!nD(aN4=
z9Q8NLSTnB6fG<B^*!iZkti~<pLR(2sFVZm!K+2<`_REivpEA*WVZqZBwD8uU5<v(v
zULvo>dFFcel=?(gf%l{&pUyo13{E4v`Yxd8Zf}^&n#7xs_F7)5(c-twN}ivY+XJ4^
zhP|&5h8ghxn7~UIIKG_@>|K$*v-iKt00QFNz?BTK9h=3hQ%6WH0(`1l8XSG=MPB8V
znIA!V)&_ubgph#a{QM{%)mldQ`^UuQnoMG1Zv|NvV)C7bOYKCTG2#pH082jt!6#vh
zT5T5BH-hG3-@0Uk!0UjUlhZg^0vZV^?4;2Y8Yob7;$Fu-9#Q-!A9Px5V&n6?9HKrQ
z{PrCI{5xIAp)OI@PwBlIS9zaUQ5z+{v2vNPUk472q83QR@(S9_#~h|%3rgDTPv)Y-
zsT4r{s<W;7Wfc9g<0j<6BSjt}Q;&OSd`a+We*}X(DU8+($Vxr@IHc&??Qa#)v_7$K
zIiWA*)fByyc3{Jjte!bJ`tyfviL}(fUR0FH;^_zmqqqv|)k}JR0=$wfav|)x?87VD
zrz&fEjprVNjqOK$11jsOC+tTHG9eV7e<DQ1uTsYIplg(2-x?w`uCUU<(@v1^Gh*^j
z^6O3f6DCjem}Rsm0x{OxJVYXmNfh5GUnKjqG*UK+c2`@=!4QK2LX|4i+_l9!=n5FS
zS_~P$H@^fQ@tS`Yvwy_lM;O<~m<^SWB)&JiwfBrXHoxRq)SyOMyZl?k%WFtK(3iRJ
zgB6dz&I`1{<5SgSKYRj>=EQhi1tC7#xbUC$!=eo1KlspC-C=>|S6YBj3%e)n9~W<<
z0?gl`OWu)<MUa}LVUo1MhHUFRNi=?4g}x?CAkd_yl5QOe64eL(w?Z@i+g4VfNztIr
z7nHZWd+<J1Mg}|EAN()OJULLl-)$y=EEbUhqYQuSZ2wfAI0UV$C8rbk=bJCAFz=ZE
z|1~nce|_h@q9C6BV;m4Dc(v2=qE^7Ge3kf^AV>sCF8~F&RZ1StKE`dVZq!(N;?=8S
zXMy0Nds$8>m2^Y|+6)3JG1Y_sLbmt}f^>K)+JRxT)-|Vir(SIQ9@Q|T(4lNi)9W7?
zoawgb?N-g^??%Q(cenX_b!Xm_XqPnfmAEm+D&OpY%13qqF>(VRGjNPL$A@q4aU)H-
z^*iHMp+<a>a+ndiiuKMek<UCrVBC3aTqZ7vbYS^h#;rX&2BM0C1?U%2K^%r*iQb{2
z1kCsHdCEy1i)=2oxr%<I8VM+^O+zsL=yw$gtOyhMr3^oTp%E;vqO9niiql6BRHz~j
zwm*Re6bYr^hW%gYx1HbFVKI}MDBjT3!*}@n&DHacl+D@UW4}zVW$Tf3-(y4;r)tb}
zvJ=X~hojCdf+hHcLoSzu^}{v#{#j!v*-dQ;?ykE$&s5FJU)dO5S5jlC_hsVdf?m%m
z@IL_D9f}>q93=Jl<jTDF7BZHOIGb~0)wgao7V>P7t41?Wt@MA!;~1ddJ(tC7_sypE
z%K%k2e$@6{bOe-_+tOq%3peTuf6-HNl5v%%M?mm1D%dh@ASBe9lgC)uQ^*>c*JZ$r
zN#qgwwa^caD{ZvaH4US+SDV#S{(kQ#&tLm5al7u$m5MlgIB`{EjC(IOrT>mklAFpr
z@DX`SsOIjbN@RqF=wg~6K-H$}IrJj>Bp(!h5R>}J|BHC`{k3|@OvyQk&FODw2k6F6
z<$eh-qM2>LJA&ew9nhq@rCHw-Tg{hdh9g24L9MCCCEZA(Xw@OOfU2U6`vy)YiQNH)
zz?{H?L2_Z0eY6uiIbMjS)6CGP1j@LP0(Pv3LMX=SIK?p_o^sCQKq>m!ThP%zb8Wu^
zr|&-YnsOqny`*Y-nZ^g>c>8MXK4hw>t*GE{-|@f0m$(IbYaWG7nuO$UIAc#I%buFe
z37EQtA8|AFJv|_2xX`wwOuX@uad<6){J%V46<$Im<D_S=U8p(0E@;dX_@B43LoMwh
ztx04?y89+-UP_L2@{<=9{&TapgTwyW0AI!6UlBOG`M=4BnNB>8j@sFTJl|n+SOJ<{
zQD=lQ;Te;nbrbFu3K~+Zj!1u~i}%qyXA*Y@iNqH3x=fG-=+?US+D(Hmm6TOGsmA2~
zPx0c%<0=DCzG6;0^NIOuz8x?qX9*tKF7V*Zk^#^aC_(1|Uts0Tc`8YFT%Q3fZTCEt
zVeNk28_T~g^L*XJ^W5zqY*Wm8d09J{`a}guci3NOlnkj1jFHcDVuH}F9(&5m)Y5VO
zX%%{NO61-8XGnZ@@YQDYnG@QU4e$5&v?(V2MWfy~BbXp0!NcO5_M%#t*0_Ca`lGB$
zqPL~&%ncvfli6si2CM_Oyj^i6%aoGmo<*7HP9$EjoTvB$k31VD)JeOBTq-;W7#{2t
z6~Z}Q&m*GAZ#wxc4QFa_?P>q1G`cGf@o=Xv%ZZD{V;3DtE0&#`-R0)dn+_ULHiPcf
zx(nqKLU*u*))ka4MVTuI-CUiG3eiCeWI8`3Ger~LNL2LsE$rB*j8rWUHxo#iP~l=A
zKe6>o`B1ME^8SRx5#Wj(mc@~YAEi@p%elL|6B<fF13i=sfIMBVTMaZLOYEu@>mkN$
zE!1a&<sGa@ik?O;lf5F#Q#?aT>k}mq9#yqS<L$_gf3)@ral0mGPAHLz$7c%jmQmM>
zDlaZ&e(H@WNz~l!>OrB~-;jQQYIP=o&AOg;<Cv927`sbsfXu9<F7y91F+QbpYx}37
zFS{n_{dR>qU#Mqu=21mP2)x?w>1!)<cpLZ)A@RUsCj%7SXNzd-i-)A{PAxKt_ZY$2
z{yIF^phrWYn5X$BXpzFqxoai`Z}J-Epu4c9n8sZy9+XAtki^T5LXExkRga0(Pf~63
zWT1)yBRcKq4`3(}u{Ko)CBT|xu^xFGcB{`ABfxr(UW6zBF6Rn5cbw5}Ndal|<V(S8
zC;jfBFBekCp1a%GqKsAO&BQ5!W=+HecfSjeLLhVGh_M<8&asv!zP}68u?~KAF}#@}
z*9ZLH4w(msLCE7?!_*&yJ34kZ1paQ*(D<H(;@sa1&DG+?Q<4S-0JT*bBy`9(nsf4C
zBXS6tn3l*yMh<up^8;#SBDj(Pf;0kwdEg~}%He79yA-S%>?08JnQCD~K$ro&61fII
zVoy^s^FrI*&r>{7-|}XdrgHdK{}Qv!GY)kKNcp@`6nO$y;ftF{F&DiF*B+5z!|;3L
zIikW1i6d{{JwPf$#_0kEFSyR1cf0u@Fg?S%V}=c&v%&#9taxfc+2<Xeo*y*2`T{+j
zWRE)8>tuW_uFvSt%!KBW5s%SSm*iMiS0^v$C_WPd?;1GkWO;MZM4yn~sQ_~xB)jJ}
z-^P%XxmsEZERpL6IB>MtUDS54O=16GnL8MHnSoblf->ddZELn?2vKITYM0%4^Yvak
zSgBrWdGNC_j`>`DCzbpQCOX;I<g+enQla@CvAzXJ;fFskYPO*sgM(0zmTk!m6Z5W*
z0K$EoF=(Cr=S>e%sVDpk*;5oJWhDcjb35d~2j5R!X?LAH2leFp1q$@g;OwzKS*Q%D
z0gm5M25i@^{t9`x<cw5_7A6U>NPGP>kX-BqTp-{85l<3>IAr<)9_`mEUg(o+5t0Ba
zS;a>W`?$vXqxViy;<Cx4rg|s4hTI!GdcT^P&Q++Gu1G68Af7grfTuS&-}+}29TQWt
zae@;$0Y>lT(7*6p5{npeH$rlA@b0O_?3~Cragd0hF}{Nk8l+U9ticO%fSt!=6#uqg
zoGfp3RkRNszyYS-AbW&+VI-`mT60k+vK3|urG6G72PyWS(}+BPjYZ$u#EoN7RQw&M
zB4Ruo+3*SkrrBg)Vkyh|fbu>n>E5d*7xkcLfny<!xJEv5S_f7K&K>TQ3Bjd7hJ2Ug
z9CIggcX!$2_HPO*i^xgR6L>)BM;^eT2(W@uOZ@CaZ=iVGnc5v^Ub;obuA+<~0(YjV
z6z2t{YbzweIH0CH;xP7R3d5QZzV85$;*KeKtT&)R(cZqmL5Vre%pQlb)f=V&CJP(E
z|H3-5zRl0cge{_<8G6MfTiDk#O%FJBA3PyU`zOgCu#KGv%-kjq>h)N+ce!RUjPLOq
zLj~8_jzz^Xp&OHl&$&IiNYbasH+LX!mVa?~pQyz-8HC<h!pF+4tkZPfepaP5*ECoL
z_MN6}1rG(|a62^<-do}w?%VCdf5@Mny8GyfzUJIRc+kv6*R7!^=U4O^HPVhl+Q61h
zeASdTLCV(`#nB<qGhGqL9;5)GQF4j+JCz$!%B*+Y*VK|?X~%DjlS;aAjgx4{$q$-6
zE;_o+b-h_M?sp5{z{sbJiJv2q%c_50v2z$liErsO1fmPPqlOL85rZqK#vIa-FoEsL
zwhmao1~QIRLY7m{HYuT0BZYEmD`(0o+Y!@B|Jc~Q<Lx8X_YIL%0-RySdPW%g(K)0$
z*&~h==1>aaa~$r}g**1Hj|Vuah=Jy#R}c)n){GH+<(G^tdkOwrqC<EQvZO}AE-Yzq
zByhWe(X5urr2VpDxKtazKXqVfk2^)c<+5$JC$Hs@N)m#lr66}&F?;=)G!32*adMo7
zD$?~DLkhUs4bj*(D*!FTnr)T{{G4ZY9r}xqMb*~(<0@&#9Z_&5S9rO`@IgO|S})0Y
z%j0hd580<vPRKe|N|YNwkQ@QL`Tc0xE9qOF=l361lNtCbd9_w3+C-IUBLyQSt=#t5
zDawSEq))9MvAw7y9U8DVW}Jk>Ue_+R78wc&bQfxzZI)CwZUiCj0wND}M|q!0;hHJK
z8(mTS+jE?EJ=$l!L5GX$`{!GEk;O*C93#j^<v(??+}L-AvvqjQmONX#Vi{;F|9~pD
zd-!Pa`bHGq>FQa;8{B<UY`aT_wOJPzd4zm~Y=SL0iGld6%9E3xz83gnWw~VB48mpm
z$-1*5XjO(Qd{8WdccZ=JaeV`iB>M2gNOk!O4h6v9nPZpz%Y8yrCQ>`IK7Yb_ErRWa
z3y4UH{6%JN5XYmI=){*064FYW`Jxm_^C%8{$-nF@D#p`{6GRJDzMvkRR@^FY<h>T+
z{lJ&gKqM8@lT?t0m&s6~A*AL?EbT~M-&#_hLMRaJnu;z`RWvbGIOi7Eu$tS$?j&Q3
z_~e>dmohUI2+ZD}?H$I$iJtR@;obq0Tq=q9{3>pWykDQmu4f!PbCw1g57na1@#SMK
zo9e(+0|q|}td#Xf`5<%;pO(Oo!`(1IJA;|v_0*DjB%}i|tI9KN?%J8_BW<5oV>0nk
zvSeewCb3c-kB9J?AXur-4SoM4SCA=XUfIE7@aJUI!?ba?Tg7V_HzpAXvxB5?KkZNq
zayqnYw%_4lC)v$)FYT?B@{>q>Vk&3r-_Wt>U6H?MeRE?1p8U-m(`he`Z7Or(O&Ct=
z2+l#M7K$-H(+q2J4~@hr(LR67wY^hghLd}jEs5`#dtII6&6Q~?+gYCye3G%TZNAFy
z(`Zq1-knz=bD~a>K98u`?R&@bJ@K=juJ%)b$Q3srjjLV1_zf_r|D^VU2$3K?HoKe~
z`2eh+=#8z4EMlogzW~i#W5;9pkb28G5)XUS(kPmUNOf0gxUgehaplVAZe#MfO+j(4
zxTN4L(w>yr?n8FpO+xzS#^e@)OPTM*PQhZ4Zm<v&diS;fU|<c%t2^5Tiq@41&4<iB
zPDlY)tr<5iBtn(y2rEf769hVBuF_uz&j~4e9Xlxp-TEbGn8kTmUDnM_+3USg9PF2*
zuZvm8xf__O*M~$&R)1mq?^!SQ5t4Y2))5;LW(3LiiO7pAku2X#q56G8sc_YoVbgj7
zcx)yC)RnEu_`9iE>9I|t^B+d9*eR;A(KAM%eOG{9oLAa*x-Gp3sq#-Ax3_Se3KPU-
z1Ogqdf|80`S%pma48iNCY>trKo=wrh-&Wz$<u23OpmSti2sNTPY#4+lYwXpBiOt(;
zuQYBUAXZAMA#7F#n{Csd=M<D;m;KNFlDlRg!`L?u$}A0@Sj)@RkIGanMB}V&Dw2|_
zrtL)E9aFmUWSq1o1hoE`^4bxuiw;FaU?i#PU_=BR(J`au>Lquq5@ua>YAUNf*zb;<
z`7O<e)!}eC8v6iG7ZWzdy(FW6kxaj>SaJ6-FTTtZDsrRH4p<Suj7<Qm+SVj@u#olm
zY%q8zDPoUX4VNQ{Q;$gDBt(r^{+z@-#Egbti7?j+T3zLv{FP6nY@exRZa;9ImNW``
z*KdQp_lV$1TCnAQRI20DQdirKRsCpn0u!;p0k5YZA01mW#YkJuKc~4NffbZr?xIO>
zsROQ!pcu>cA<D@pu)$tm)8J99tX6af1`NF4(43qQX3taGHvKa$K%b&p5Z#zf4ynAJ
zjaNNosNI%>^AW%QxL3<y(iCojM}v_gSN^Eaz|QxOy+PtZ7^fUZ1j+=0v%hk)#rI#?
z?QUKcE+#x&H;7_U&GW3g;v^N{?T|siyQHMycC>Oh^3;w{-znW;pLEb93EPdZ8st#p
zHJJ>;JGZ5QRtoOQ*SSmoK@PsfHaph5o&T68p!@OL3>OWZKo<3JATn~jf`u@c21Z;t
ze#Et&L*^s1QM;Lg(nmy2Rm}7K_5SYYs&R85l^Pfy0ah({qzp6iMZq&QUVHg6)*psE
z>{t!6#e^UFpQKpuix{?g8n#*bPOVpwg!MKzmfd11Ra4jds+lAO-%Cpx&SL&0{A8Gz
zCmA4I9$Vi<e7>1aDpy^!z{l-x0Cj4_Ku`+~d#%}iW3Trc9!h8~)lUH@bs?gqAtZKT
z=*B-tnTVX72lUtJdv)QptNK+W`Q>yD8K*kCZOG}C>62XDnLw|A&in8cby9=R-p{@_
zUJeu$GHc9tZfOUL#mg*g4>oZZgzgEZk1Z=bsX@N7DcD&1XAX!ribJDY&kX||L%#Xo
zWsp){o-T^GmB}R(JYrf0k&V}~F$L2Tee*OPBlaEciSptz6H(^rz-Nvfdw9KEjXSS(
z*e@)x$ZvY`%8FcwEqEg=5i`Sg_ou*GjnO2=P;+=p{8&DwE@^;c9JzbOrfA>m-p&Or
znD>{g3_dOOPrb{wWFDx>I@=peR7d!cY>q3EIxfQ3FV0h8c=&b$KObaUvgC_Vlk+AP
zVQ67qj5L@QLf=O^S{zS7%JtHkiZf`^x)`HEb^{0R56io;C>HY&0;pz8`Y&YCrj^gL
zJlednd0-&K#A5XgPYVMxSy`u^pGdX}{)wO?^hqd|hh&>kFj}XE=Q}s2t3HLUVcA0j
zVr3-Sus?(^&DZZu(q@OFw~q>vuA#wOSXy8EH87tM(hi7$_+Sn#%+{%olk@G=*3}MV
zq%g<IeMn>P-}5KXlVltW{6aU^@IA7oQBhp+x|-^$e7y(;UYULIXXPTC+w?a<FmW~f
zYP`1E)5LJ+6G$U={si<b>80zH-vw4ckge-8VhPb}&YqbE8xB_;LJ=RyoinKp<yjo<
z1Jy9KL&id6;T6Whv8$hdQLYu2I88ZlMx*TK9`~Zd-VGtL)OZ#5$@9LGuHuzax!0$i
zCFa$Q)lZ-e9UG}@EFPYNvwhPu5nm!F)US`8AQuMUZ^VXBL6Zpy`Wm8i4X*j3V_@4z
zXJbH#Dn{=n9(_~-_dYN}*JB+*k@&sPCK8AH8y2INbL9?COvgdFERME9oPxp>RC3A_
z*0MrrU^fTH)O)e}o>-9|fy*~6se#Xk^NG9Fz%j%@iX`!LlE}Wfv)}TJKcuY-%Bi8T
zlT}+~>E9T@3iPf%NO~$tNvFTtoCsNcM%f2p`DyoKd6Q(+orL>WD{&p>Z!HmT_gpxj
zKG5T;0$4dFy-*6LY|Qf1*MyIa<8?^GuHE58ks>6XpIzt;$q?2ILV9tL<*>h5ES(s-
zqPMoUR9FhI0KX7ZSEdBfn1H{{&+$v~{<svK7^3YBl=Mln-p*1pdK>ay90Kazwf^T)
zs~J+r3g!~A9yZKw2<Wx}Y)EPx_@LUo^%Hn!R<p`#OP<X4k(&53Ls4z`B2Vn;D&O^u
ztMtT_l;fbiT>-H5KWuG3M1)0>#6yylvD4WcV}KonP0`j-`q|UPo#zFZ*Z{m+6ePH7
z6YM&t6M>3Jjy_;0Pc88I-hFqMBL~Mpk~q)lzFW2QU5}J@66VW6T?lrToyQTCxbhc^
zo{;;L!;em~KcS}ne1?Q6po}q*mEn@qmRZdP$wOq#;GSs|>=Awrv)lc>B*0y>dA!I5
z@$>qLwoBo$GYY=i)e@0@fUkdCcf7d^{q5U@!ftBAEZgqDn{P%I>7%(PXiYB(uJr4o
zpSX8uR(<j@7DK|Co#En?m4(6QX{j&_acb({nfqZRLI6#n(0R1K%EoZyyj{>zapn~W
ztcyw-RPW8TV&TxKedAz6iykJzuF0&9L?O--Kyq9SzS3$Lj6!S2d^WW<MO~@$9sVi~
zhU2&7T3kCG&%$@_J2g~!$;IJP|8u;+>VLPNOEUdbC3!X4*N)W;Yk@6!bscr&uw7v9
z^x#*fon*S`L3A)(<`@-<U)+a``3Pkve(iEUxc*Dz5&Wf)^qABmrK}6Y>kpPNo3^I9
zVr#Rz8q$6C;td|kN0kO?=gRZ4)h+-;-2cnyZ0Sn#w>4n&CP0vb(4P&>udCBQlw;`L
z_ylWlob$Aws?}<R)QTQhjbNd8Buc#EqBU^M;|7$_vl&5)lj5gigohx5zjD%vJom<4
zls{?Q%U-~rK7I}sCcV+^ImybmG!zz@M!B!6;b`je_sE@eo_*i<qOt&YR}CuWn@-=_
zke~0&P~IO7flp6S^CWpxP5Of7W*5CeM~iD0kH+UR-Rf|+IkH%loPEUH9!!~^Yw)mw
z&=pUs%)1KrxKlP(pdNs;iKvEum{|llSH3II@jwknxK$1z{#*)g4k%FtGa8vOGQq<l
zredssXhN#M^4H<tMo~M*Y8bqFU&=mThhC}*cvPpgT!n_+4AhDJyj48IiM`@FuaAG`
zzlN^=ZSQsexE>K9RaW)0Wy6H4t9lufXOBJgr`ioh>yk}MZdGIHCCOa_V(7NW0oH~X
z!5jNg*>!sjwx_j$aQ~6-b1FaI7M$JO&GA~_?b!uI?l36T&ZEz4@V7?pjW_(AvB@*y
zEZ^*G6so5$krt?xA&q797sRfrUoRCnw+UIs?gt}rFXi58QXpImZGut6a{iyit!gK~
zEG^sW>^-l}eHDNXRZfTCtE$H*{2lCD+$z|>J~Zf(0j8CaZ=;1Nc!jif%=ZE%0iGNI
zdR?$G9aUe+9er8L4^&so&q-Xw>yPd#Y?CT1tG7QOu!N@sZIWtKvS7j7^hIqo6|bR7
ztBn@*-<TLn-=l;Im4eXYf6vP}6^31u`{dcXg@@Y-wN?~bEEqAO3pKGXbb)8+@Oo?{
zol#7IO@VnpJZE8JU)A*YPug{FUJT?<UQJX$TIwiQ0pkMF6sB#$aZk(Bj%77dqc08g
zOXnXF7P!iXWPzTco-s`mnQ;qV29pHsLtpA&#~pjKNuE2SwQ${FxAU#u>~)27&@S{c
z;LrJ6cf2)|-07BGLkBBcp@i$F*)f4UE<%b;+;?_I7090I{h+h{Pxfhv@-@H5X>wYD
zx)`p#*G1>>GLze0D5^(G$Wc`su6@*ia@gv`>ym64{I6ITzJ=*E+9n$(2`*F{o?O%>
z9xx`h&0Bo5pFGqiEksfc_N)eA+K3Q&hPrF{n^)7};VH@}1MJ1+X3%sBtv5_mwX?47
z83|`LyH8D|q~o1OM~9G-DIekz+r%{bC;>9o*WhDx{O5?=cujN;t+b0f1&d}EFT%Py
zc`gI3-DdwJCS6f}T4=2aC2QPAh%VM65p^NXHj+(@^v-j)KceD$pT6fYo!0#8ffNPd
zRflIIW^M#s5IP`^yB}$64-|h$#rElkFr0}tmNnM8NC^uId(v)s8uUlHGekMR!QN{5
z2~J_U8v9f=T@S5_Y;*@LY(|+Yg+7K?uT>S6ULGA`Y}jbcU7K22DP07S=!wJg&dhes
z>i{a2_!LIr&7sMopQp9tUU04U)Br9_vkL&9Ql1OyjNveb<l;Q=W<96Q88;RO%cY>M
z*JCL8sh}XTa_FL4QRs`Beti*qGx8MAWuo36T_O&WqOkTdJ%juQ4{#St8_L^>NbKge
z6`d-4zC`*C-{(_wt+P@HFp^cTC`jK5%n9z9VXNC3Yi%95Avn}I&5rVphT8H)jeU1M
z?y-pXe3@!Ck!Qc`-)7`0+P%We->L+k$k-1=b6l$z7Ia3*3S9l9rkahauYLVoLYmk?
z<4LhN7Q3l;Ub~d^m?{z!U$S!DjF}npeR|8>sPWli)&!rhTK#{Cmdd!e(=Io#E`1bn
zP3FZmb|`)6A#b9?kdEzriWV9Tr#+eQ`^(BI%qnR?J+1Qq@>O-P@9V9;Fy|YSNO`ju
zo!0;$H(l5(AIL&-T52z7&F1{1^D7(bZ{BZ0oF<aG-Q&O-+L4!57FD_5GCopmyH<$g
zF4&LV%$T=>K&px_G(t|%{GsHZ*q1rZaw2-21-p}~cSY5T|B55A+0aMh_PYP%bZQwR
z8pU$#cJc&~K~;ir!}J|_!a97s><QAJU$z<RFche~=xO|XSyfFhEv2XPN2|^u;U_@_
zy>S)QaIxsgwtn)M8%QwI>aH(n()%f3hw0m8dymvh&HQTBhZ9e48UeR?{bn%-Qk!!!
z8MvB9Q>hGne9_p!sDtg?m{(LJU)&o$etP7c1Y+Ui#v;|EIROpfGL|+K?fjd+>Pj4r
z@CUw7qnQ@UfaN?BmHY`<7VL+#Zw=2VaYU{V)*}e$Y|~lph+}a-H*3!_K*9!GC`*jR
zh4`8=yX-bi_>|V@P~)E-C<KT+?qwF&5sLVH9jBU0Z`<Yx7qV6cxn<8J8r~FoY@Vfd
zi$zTqJ)vHJsZ)|C-TREiGD>U9YCMupz`sZYoTt6Y!_La+B2lLH*WKS&9BL{~M#@Ba
zu3_FE!f(?jV^<LAH`>dvufXGJ{2F7`YGhsHE6e*re&Tt_#2_YYClMm{-TA6$ub?U;
zq;&<0E8uU|Bf3^h;#HG9Obk+K-K^Q=uy38DX)AZmTBv>Ae2oEl;=w3eM%Olnl@qD@
z&C5QaPV{Tyd58mE7S`A)cv;IML~d$86uk_&y)lZA+i>6+>~09*ufU#mj!$DdTnHO5
zv@|PPD>r$Z=+5zArQwc#cv!TshsTZk(sX^9z^Y{dY{q|X$ip<9+zlq8)02>UY02tI
z#EdD%x)Mw!Hi5?*-Y!41!#?*o{NTY%Q6<71c*96iAwuHp45cIB=}dOIw!(-ZY{ooc
z0z)C9Lp%E_i@-^p4|cA-KJ5_-A+?Wiei=t*j+vkrQ&xEpKEVE&M0H}qP;d50F@Lc%
z$9jhvl$fD@)0IaMlnK5Ds-XO3rUTWY9TL#84t2E&-<KQ1Vb-)3nd^nz9(&n+`gc~U
z(K%;<+4u9_xkUG@4PUexpUbgezM_dtERsp^!Lj*@WZHPoAxIwx3!Es&dRyLc+l;nl
z(<}Sa<6Gq#B%Dad42FKn%#n}Ij^rp4t-lXjkA~7Teou4&9RnOnOX{l-AmekkNl-pT
z)9km^EZR9UfT$bs<f109QRFrTtSxquJuDqds7g=5cYe1X+A#>JNj6ba^N%S%T*qNc
zDr$LOr<|*XpjPi~J)n{R3w8L8i8*HVTD#&T>=Ik{b{hzZ$YyP|fAFk`z%4fzBJTmw
z1T<ty@{5AEM=%vIWWxzM?Ja-wdVciyFTkM0l=WQt!nnM{=}W!CBoc8k9@5$pJk#4Z
z*I%|+K+R_nO|G(Ix%Z#m8Qn;oOIUn=^1HL%0>eh8*<2kJ&)b{bZ$h)j^;b?Tt-PAB
z{_grhDN14V1TDt-nEkCQLju=S<We>pymMf;pUmw>c#?%B;9sT;qXNpc<xa%D$ZnR7
zc)YyG@WjhSIdaqcHg|%tAtc6&PONs8T(Y&(k{D#(k{_wQhXyuA-R>v)T$<6v$$>rA
z+0SB+b2gRVSCqXlzoyQivs)b<ae8Idp0Kamg`Hhe6V%^UEi>R95!pQZW9~}>g!ZM^
zNhWqlM-$iVQ#e+mD(i|)fNgjJgX`>rGu3Bo90)t}b&HV^TczHhBsKPZdTrpOy@y{j
zv9Ph~e<}52_k+G5MGDCl4#C&y#UI}7PPMCG_RlZ4+(|yHHQ-y`&d$%jrF+kT%YEEs
z4y$0x;;=eyxQu!o#LkV)5<14OTaK%o=Lh59*%N>+r(~_Gw2;09;&HU6^RF76%xrq3
zdYS@0dtM~tr};6oWnBIe+a|VPb0IJ99&!*1(t?fgw*b<i=|@ht(4xGN4`6@=z^T9)
zqOMzx`6i!lZd{_m1B4lCujlv#gqhxj!p0511WtecR`N#wzUc77eOF?ytiV$J!19+1
z%yo5LkCuvkz5B=zg#H5~oT#igah$j&-uaPPrsG?~&QlMqcz~7Pw%LX2gz$X20mb5A
zkNT5lN{Z%I>GW&Nu{7?MHRio*FKp~@El(0_g2>OAX2-T>kwz*~B?mlgHH9hB11l*V
zjO+wSl|7s|E3TJ-&(`G==H=afX=^wsqW2aVEcIQdAEob(5*E&ieJ<EBRqseUkIyk@
z6!6$xZA%?S{p5++E!ijN&1U%*OK*!TQ?K>YT}Qd6zqIc|AY#pW#2t@dPS+l<qk8J}
zAtZQ}Z&xquk9|cP-_Y^~iUGFAO(@=x=EnUz1W&df?6j^0Cl}DDv@N_(Ya68?r_)Jo
z6k?h)+|@3yd1|eX)Y%2(R2aOiTMjzgmKJ|rJntfZAtHAvfs&qBP1c{A8CQR}ZEV&L
zLP-2UpSdY-*21&;%A{l0dDNYRn`zfcnI{nxN`!Fv0%vlC97%kX+CdmDtRTZ!bt)Mr
zJf|dXXvxH6=hn^`D6MQw5}bZI=JfXAwH>aR3+-t9up9Dr?AwHwF5{p(RdvufBne+P
zP7cqApdi?w78X}S0;O@@fO;~iM>Nv?t9N;+Ymrnw@>{I#o(dE{VgBv+WR0%2-SCY?
z8WL#Umbt$o@z^@o>aRRpu|pr?zbysnEz0bj&#lf2QZvef`%fsXnE`BxEH>L@%-F@)
z%f^Y5!l(W3FZ0W=+JAkH7OmRSGio|Ic##J#3XKm-wO=h)2+#<t1cnCK80CPskiyDZ
zal3iAd*ij{!YcPm2~>~Yg-0K7&X0k7c(}oMndklhFS_9)q2&v!BMx04U}*hA-&Xb!
zBP<(yibc&gFu3}1K~8q_l_VB}0m%tJKuc~St*Kj(A5wEOO(|qmIZ>CDW9b95S&|yG
zNr&KUv+_hGoBUQ*ZE#gV-Y<1W&phkeMrV0_)d6*0>v+-Bdrr?UWdqi`R$JO4{EBJ_
zaz^^A@n6(Y-$VyWy2TE90hpkmz2GuCJXu5Sfq2TV(G^lwb$JZAxo73>98CVpqw9!i
z#z-?p!emd~0hBIo6$9S^f_8{C1|=c`)72by&%Y8TQ^`C5E>WkIxAJ$Rgb1z8#adt9
zbi80E><zI+M~1>H>T%gio$<GJoV8LA1Gn$5Wa3OY(r&af;|??|G*M0$8GAZf-BFHa
zXq2s&GbtkPe>QNa2Y2~U2rPv)UTaHgA5{{I39#q4pU8C9EJ6o`(Cyu=e`;L5?9v>8
zZV&a6^qgXAGm$VI`cnw&nkV-UuOmsUogq6g2@d#f(O*abMA(q25lT8X(214SV!$5Z
zbEvY$N#aj#uP`gxG2*>RIupW04kcu*K6GtE<ui}3Pk~rl1zvFiS(SiZc6JJ+OObKk
z)dz^jtYUF4mX>>JF9K5tBdZTPi-@_Q3%@O-h-bIij5Uo`{qZ0~mMw5qo*Ogj(@Pz?
z{)>w6H=~Oy2FnbtVHx0^RA~LcK&>W&tKJtdx3w7+cSfP*fRLYK5h43JTe92Ywxk#l
zrNI}nJ%ord?qmvWiR5M<2(G&~R@k55;`c}~I;KV8PC4YpfYU)d;HsE-2bza%ZSvIw
z2W<7k&DmdPM=G#1<Ox^OYmz$3>gC6y+S9?F+e|2MZ+z^#GL5xD5J?C(KNa;vt)r+-
z84y(5naPg7T+SoHJ-&*ph<0+H@%8n-9t$z}{N5l6h(-kz?x?<KH46?ux?h_P0t1?>
zS45Na+6eXz_g?ObtLs&-1`f_gVN#~mav6uoH-*glUBMz)Q*A%Q*Mr*`h2u2q8>B(~
zF`%&gISvk^LDKLAP~wEasZU+<UaqW<dMX8v84GwuIm*zv>P<vkcmjd$M5IRj!D7r`
zqdvLRv2ovUK)ibEKJY>(W<0XD4{tK6F-9Z8)kP&@9y0NDVuHrA>`Dd@YSunsLB)Z-
zrwcIeg9sJwBM#tR$_AZzw2+onXDjO@aY-3(_2H8a3uKZWODIB0AA}{^3bNOPqhCsB
zaV5LH`|BJ&A};&;bzlt|EW$=YS`KdQb#qO%<{hJ_aAvE%rQnQe@|W{=%d@Vs=F9H|
zhd#!e=|Sg(YHLlV_6JB2w~msAF@#J6tul;s%(+ShW2<G+KvqS&V3&}P@85F>VhcA|
zt!->m*?b^duJA);j|^YPEvCoyTw!Bz<EY;wm9%^l2W4;K)(6v?lMu~L$N#yTu$FGC
z!alZ0yr)GTmHi=6LHw(pk!-E4(y!`9r<9=%RMWsn)Ci!S`c&?DL_L&3&f7Bx>gyA~
zg4;~NM#zR;V9hWJkp+;sQa*$=M{9<i43^%BwMGHK$fQuu>7ZNV(w@x-jC|4}OXqF)
z0XslPQL%;6;XUOm&n?N9xrL{n5d}U>bFR0FWkV+RM=FRU$H%1yw1~R7stbc4c}`Ho
zI<??wFT<Z0N91g?R9BkMLwwo9gOH~VtCaxauNP~R-If5~-$J%2@O3M~8=K%a?_KN}
zg?zLl8l3z^<Wdb^P5VtC|F{WqqEwu~q!VkZD&1;w-5p9z`M%HFmYU3B92=eZl;h9b
z_RSrTqS1~`>>U8xOSf80nKDQPatNAu4>o)Kdn%df%#}pgW3Cl2&(FvPI%ncxRKOR3
zYo!PI60vUxg>^tN1+Ctk90B4=l$Lmyb);j;us*2Z@~fm+gCt<74OhL2HUIA3eoZ`g
zRpzfIAg7mJJ8-F@$+)2tj#B^-zZdEoJdXy{nIIQRSHA>?c;1cyo1FL{-MS3}Xco|T
zl4qZH-keE?A7+>~dN(?#AC>b#X(NXbxZBSInvD^Gi!K7p5m^u@DnloZ>74lM>IRg+
z^YdQ=MO8JeexCVGIuG&$`=ESjXxQ}@dHcer<Vr&DxZ5o4lNpq?$0*>xh9pr`<2F_C
zgn~~+RRG=fgd7VNV$!kA^Lc8sl($Jt4%7gR5nu%bGZn{C9FWtt^xrPL!Dwkc6O`O<
ze)~CqCN|&;C4jpZYc*?!jZlo@b+y0VNZ1_|Ed5+Pq&|}eNDxIvlAy`T0AW_BSR6z7
zq0jy9s8PepB_Q<zV7$S)?VAJ}i;qSeZU#2c3W8c~Cg>_ySO0@aOU<8|CI|8_Mjo2B
zdJ9hk(et~IM-B`z_(%d?y|k~J4bm^H-jQ!u6Mq_I;^<f!Kf$V-*(bgFH0#<~s{gHg
zy-9d!pgKYTYRE(U69#1j-gIcJZz!@Xgy;_x#I=dYR-x}fG#~{hJbMiO)REJSel%-?
zc<+A0cO|%f!tmp!@QK-hA=nWq*~y;eKIAA3lR2U(&<u=x(^R=xuK8vXc(bt3zoE(D
zo9jDPEOt2Cudy=*?I@o2kqEHou4H!=kJ)8ksHHDCIC=ep?7+zTQ#tH()5~-dEgrCT
zM~{ly)G8o3jL1=ZGc`#1i{3g~2zsTMGOB1q7uVKXL16!mmzpjEJX?{(4{rfZpAj-}
z0twtqH&io8D0Pw_jJf?`=B&~kLncH(ike>oaOXlZ%giNX7g@(T)(p+95qPTUn9SC-
z1%K>RCFqt~Z0vu4?r8&Hb_JPp&E`?*ReEFOApKjO(?Px_61%Q4`%{3OxcS<QQM;7L
zaj<EJ<^YqF85S2-Nr@46$3F7M!F4S5iBz-6rnD$R<n@=^$n?j4tYD5#%>frdW?cX=
z;ZJH5M3ZTxk;D}R>Nn{5;R8a-qiLxZo^6eHkZa3uDP$54WQEDU)$gPM{8wn%Uz^LH
zHDSK%3q*2sHk{)!SkqZIxYxx9C1-W=B_D8Bwd4qp?m)=}*Se3Z74r~>doTQE5`=&?
zh9)1luBcf#<i<ZJ0!V57)HOH!Ba;S~0Y~0dD}fYjne5|bYRVI1`-SMxODQ6rLU=J~
z%{i{^^)NjQ4D==mz!<ewW@;8R>0k%&iHLqgRt(=D!C<6E>A<6ifUknym<SGO(UlQ~
zJ%99IfhM`(V2{T)xPP?=FXTaZl6;i6Lnd%}-?*gl=(a^BQ(mxVm@r19pKU{5N-V#I
zU0;GQ`Ls7GJC0ag)x$-Qntn@X$*!`0a|=$51FnyX2)ui=vG<l^-FF%ND?rh^>nV|J
zH+?@^H-`Eg*I=xZ$DaEW!1PQ*+Z9Zth9DM!3QbZQ$}2wF`OdqwCO5ZN#5O)v!1tAV
z#|TI@cXFLdKqg4h=OLbgl#u5kJ~MlCZ=<m}r2d)Z%TGmN;MyGpBuNP;IFod0aU0rA
zy&r@kJdnC7C@*k;#_!UnvRUEeow^P0`BwgF^sU2^dKC|NnuH0S-Q<$ffsgeY+#k~D
zlMe~JxmFOm%nZtlY@FM~&1L&U@gv8F2s}>v=EH4s;h=W&k+I>$TVoY2mbLA06z_Y}
z?#F=Bv}2#<HPIz&1`6Ksd@=q$Qd(9pt|0sn`@42u2q_r+i0<#-zb|FOfp_o46LX`l
zg5NDECcX<7z-Q45W+F*veOct;yUj-eLmELI339+J`)^1ID=)O~>^XC5*XU*iFAJq^
zlf`>?P~-|7QpH~mAbrX^IS6+JSx9fo%y?7G*}=iVkHCf^yy$&cgFF#!43lM<_o61{
z%QOGo;I4rprWjKy+xSpW!H;p7Sh142o(BzUCUQ(5Uuq58_}6!LL1TF_P3pBSI{Tjh
ziY<u-ec843!^b_Go`r=d{`XK80dC5AF&T$%9fF$Q>-?9B)~HZ*Wdq6@Li!y`8$;{)
zd4RAooE8=J7n3*)@-hX7dx-u*8E{ez>MnhB0r;1uA){$0p9Yyfk4^lL|BJ@%PO|v|
z;e&2Qzsdc33C~9^<ijcs(QGv!#+8tU|B?KpGTCWn_u#zLYW)%hI%Hx_jJ>YK)tDq&
zIHkOeht$GJq*pT<=JLx8=UXs`1Ne9D7971P654X+AzoJ`Jtz+a4mFO^9HzbPY*dBX
zytQ_+(_*^qS;%jzEUi;nH-E}_v-)W1?}TpXDk(PNQl`_RMNPl@W`A2_4~)oKS{nAO
z%cdCDD1F%s+ZYVFPzWmf!$Ov5#ww&$<ky0u@crPt>5Y&<5Xf>MWG#92+3}^R=zZcm
zPe_>WE*{)ugb;#_C`m9W8DgST8HG(61#X4m?))qn{P$)w&laeug;9deW{mXFlJ1tH
zwqT3+!1j2@KoZE$BZxZiJy?l1WENqs6IGZeH0`XbZ(Y6~Vjl9!US3xs5(b2DhniE;
zPY^-6h3m)Iz?@aXrS>13L3A=aUk-GD?yh7mf%%n8EN{wQfBKmk8_m;MSBpcr=HM<<
zJnvmu>F!y2p$W9KIMNDCf!yqp+=#=l!%i}Y+_AWVx~uK3!z>rqKy`+O5bo*!+P|Cn
zbR}@1&z3f>1xiL-*H2rvXA~B)vnL)2%GhcbvDr}31&7yTq6xU6{D_{Fwyd;u-}gTt
zQZGT9g^m=09TN7ls=TWVAAH>WsLwtCil&@hnzU|3Ded6f%OTTMkd@3_qU^M$M}<<y
zm0DE;V~T%=l&_7I^Xk@G=x~%hdE2}5cH4H2^)SFQL7zFxylC+~JXFNPtjy%qGpyOc
zOGYKfZRGK8==jevVpYjW;h11En6dwaC|E=jGi@aUKi8wTzEBU^r&nGb^M2<;k%83>
zE1T;N{$&nFNHd1jAAH9s!L6QdhOlBwb)TFI8zA9|dhfp{_j+g7sFbN{`fl)lllKx(
zvB&c}Kg+t3sWhk+trg|w<`Ot$7YlpyksiNSG2C0_RBM6G?fO}Pc@%%2b$%-@Y-CzC
z=poBqC;z5vDs(biakH`>@=i2j+BI!DneamFySI>}Rg|PtWX@*Z)o=Um_iKJtdf0)w
zbLNFS;?v=V=b<`%5WJv>vgkTvUr`J@S>>ZAGBb^OkA0zHhOQs+KI4t_`Ecg;7L;q>
z3D_VMy_Z=YtkAPQ5Ao|v#pb5Eh$L9#MY|dHs|2HOM0}7F5vd<@@ScKF#Njiuy4`y!
zQZd&Ra0p+Yovp{{)Q|;ELkaN6%B~cZVW$@=)%YlSk2&MytxTy7k)Gpw>R-t;MIVp<
z2E~}?$r|BdP~)v?F<GsKC58-Y_HinMtzjuNkf<NX?5XOFrqRxt#h&IE(g`9WarBWF
z+ZX}7qEG31sJ=$Vp0Pjs4}F>@I6WqGH*NdGNf-hm1Z-~Je*(dTA@Mhyxf}*>%$xoA
zen_I~mFd!ypkH{~1L9(Cnsv75Z_9JPpi4D8kJlX_^Jn$a4(O>Ll;b3SCN9RM2zU7A
zc(@5}(qxZbP%+b7T{hG_Ca#Dm$H2BMq#+>bB{ULxt?#^XRz651cH&2ZWgc1rfkDNm
z`LLX(anOVtPJn-871sJ^TH+U}LV8Jk=D$=7d?}5h?mt$f29SP58a^M}%I`sNts^*j
zmR2BWTsiJ}65jl4$f_aQC=ysFH!GKykp-_BHFdpHdqEdn;SR0ib`jjm?S^;_CQT^{
zDx@0r4T9=&iXii#r;?)g)KSN{FTh|*CUtv{`>kr5w6=xcy?Xi}XM#i8H8~d?H!EF2
ze6+(CE>-SpJ2cgBnn+=*!!#KH?rb?oG7UBNONEhuyUkIesckLxwv1Np{%_9MISR3Y
zMi81dwVVm{?6aJb<*B)~-V8J>Klcct%tSfKdTV71>U|!@vf(#e&t&vJO6bS$+!v5-
z-TPgv!=~o_Bi4p(k3Wb3=Aj;UOGZQY91&)7dzxCCXLtnZr|f4$w!VZgq`1acfmF*T
z`Pu)h?eOhgmjtZ><TQs-P!u_<=n_V?!P_iQ05vCfc|I^WpNqv2iM|ce&7uC?_--{#
z_XQbEh3Wc9Z~YS`bR9x$%qcw&k*xk-$3+R%KQxq+oDhC9HC=UT!r0NjspT=Oa%>3G
z#BsbzF5^`~aPXt7rQmgVb%{slV(Wt1<Y)5BV4h3|6_~e{35wEQH<GT`z@x~)mp%Tr
z8~8KKFnZDkS3dzJ(L)Saa<Lo47X@9)7KVA6QUm^h+0^(cdRKq4U2xnB`8D|IIGi|U
zlIqcdLdD)(ctE(+@shD2waRva>avN*_0!}%yQ<t3Vg>!|Tf|sdh}o|%>hcc3U5=kA
z=iZ~^SMc;CyJm$YTT)J&O@5sh2hV;m)5U^bAdXc382POl578t<{ux=X;UjijZwdM$
zBm200G2H-efvx8F-v`17{UdHzK&lAdv#a3#Za{$aaO5atJ65uCc7>ZSC@p%}h*2aF
z(Nw`}i1)O6AeD;1{@p-B7fb!mnPo6qTR_Q|I&efqw`jc*$i%c)X=TPV@g))DzV!1t
z(021u*}E4Nnsz8ea{4S%+6(dLhrE~Y%S=DPX(Ki>TUv(_0K07@Xu>EsMd3=Q7ie95
zaj0-L`~o5P`_8uL4W-`y-CsCq9h|nTS^VU*FK!XMekDUuUC}uzbPLqW<=wJ7Z*aVN
zni_O&$rTC3)sp%k2b9e9H2vlSx|OM?&PC@)oN#m@-<x2s;}#_l8kQaO2?|1D*q5Im
z_WGHbVqnjgsw^NQ(L3S@J!(7rj0m7yvdFHC3}n}4uCrMEHO?_XC)1ZFJ!tl7`zx2R
zeny$)+HgTL;|<1q*572z68Sf`(PAOnFBlNM9$fQT?snf&SvMK|ub~PgmGV>dz2K?G
zWGHAWx;qthjK5y5UvV=Qsp%G6ghr}mbPC0QMS#ES+uJLQX0%I=VlPhS1AFgosP6aD
zd%S~S<lE^P_rmI#o?D#ckKD6O;RlRR%TdM}@0cP=DZPp<speae8QB_kM6a+XwN7+U
zo#gohT@!L7UnWy8#%~8Ne;&!#9-+qMz|fQ(D8+zdpJJ*vB0|bw;}T5rxw@p~KqR+#
z&_!F)j0I<JqepG6Z)vc6LeSLY!7j|p5CC?%{AFz4{>`q!^<LVC;cEfs43U9F3+4eC
zT_E{0sYjWA95NLP>;TL{EIv^|vAt~}5jFG)D#0kBMf8rN{4breH05xWxKKU7fd+$U
znuWv*T@1Pz^MhcvJwvWEGan9i(SGX|p39r%!^ZI~&iu_#*%fXgc}y$$3pjEcb5yj}
z4OR`IZcLFoVD*nkZdSKBHa>m_CtXen`wDh#(~5hG`7s|Ghlvvhsl9+~sAlcv*&EKn
z()nec<x9ocC0hfEs@OvN`Gl&X@-er{9?)F=O<7s<0eL6SsU)V^FKCj;ACpjz00Tfh
zzIVk;;likJv8I-0t_3-}q5_;;20I2RRlbXW4T=#OsGIUB3_9eUVI9@Uh3rweSN{=n
zr<yUPLB~x*LQw`OVua1&uu1wpAYAfz`R8+WBkNqxY|8lCYS^=fK@M4NT_*Yc;y=S?
zp4APIrtNBGMa`l?DV!IVDwsj`^!;!A;q=powiiZGG=I<p6DJef_#b%?)fT2>I{fc^
zQ}i;lonEap_gZ01!I?f2DT`l2nA_vz;oRn0Ld+iBtZfFo^rs>ZXa_<^qy*mysh2P0
z&aNc?nGNKp`SqCKFgV32vu@5i!2G~HU5caUG=k1=xr-FV1e~#TwEJz&*X!fTI3T}2
z_)Clb%^j2kYD+rwi`K0k#b)qHHI4`;Dw}#Bp_iJ6EOA#UvI6`1w$QQreFHHmJ;Sy=
zI&7D4^^EWzm9Ct#OQr;Wc24=J#CZt86v%y7IKNHZ&wYPM9!i5QGVYbvd}t7yd4a+>
z+-ef}<dQz_>|);3(N~vrcF?}}*ilQrNfiKg@(C`DArt8RR1{{%t)7z6Cis?+QC7=M
zTrQx~xLN-<^dYhZu83+(l1sac-C%>N<LgT`cxNFByQfoqo!{^_Q`g&_iS|ZSqZd9X
zew4Q%!Rq+`>NwLtsJ{P?Gj@?ZjBF!eguzgjjD63pu``xImWi@`Wspf2LW`7r%f1U)
zA`)faQVAjZGWKD@{H}-p2mkxzK05cFd(OR|^S<ZvdVikrx>`aC4?|N*y|1f(B>m&#
z;NT;XG5@j(7szkoPdZ?fuAtA<`bUf@3wr_fJ`h>_OI*?lmGzL1Us<>uzbEmNDAzPj
zm7r>U8|+?hs!Shf{=jW(a&2>4$HSVeb7G?2(_SpBRPsLWD9e~%09ZAj6)$FMm4!N^
z0(}-s^!~(a@0?rQzPglu7=js>J2;5d!n}WKx;2*Wwy@aWHWanfgO0__fA0#9dOBu?
z4OtenwDlB45FS1d>hAM@%XjZeTuK>>ZMP1vGg-0z9za<el<LeK?#ecn6z&O|wkefm
zW-{%L*5XDb8T-qy=|ks2WiqrjzZ;*AOijrwv>R_CfHWTwL^eWkHL``~0*L1H#hkHR
zcxZ4LK+c(9>u~q*R@uI~sWEc5Hh?nl!vKx<=F1%zfYjSDJc;~-DH=2$Wnd44LhI#8
z{Exp&d}ESMK|c`a?xSRrYu$Dlnc!ct4!%1{mXxlnl7UvL%I4}!Aa(Q%bPNK#ts6=j
z%FHLrHU8?;;i0B4=eU!=_Gyhne|D`pMUQCT0r9fk+TrdA`L|w};eTq(b7BF{)dA%w
zRs;=nzQyJ(Azj(h0q~L?g|jMPgqnkc-TUy1Fa#j|gaQ)A`m%}V{O(qNaeq(<`z})P
znB%&#qBfI<bABp0U81Fv*T`_^Y>Jf>(JiX-fZ9m$&)8frwNH#*RZAFWc|9T{vxkCy
z@eR*Oh+->_|C5}u$}W1Jtzwu*uFkbYa1R|JtNZSyf;ib~-aF}JSPI*8F2YSHPaO$)
zbQnV1`63B+|8n#Jkdv=z>d{h0dar?}zei!gRk<k=y`p|Dajbv`dk#^xaK;B*Bs@pT
z{j4<EMRa(IJ|=4O6u}Jt6an-{;R-}Ty1eD?tNpPGiRmaLSC~PraG~7m$&&}%0NuLP
z^wTuFn)e-Ky1?qB&qjj?QRs#916G%>wKv&<;k%=}M(baV2U`1?Mww0QPa{;bX*Acw
z)D<DpEl|7X`x+~b*O!wAf<u58J)Po&1Sun#S)bkZ4py|ng;<C2H|S1eh<&5;S(rWb
zAWT;LSZz(@n;N%osuEg$nxOSf(aVcZr-@Qm$cn1l^3;y(>wWbw#wv4~fTQ@p(+s(N
zuV6#{N^YHv&g-x)qzSg9j)uS3moyr{`^ZQ+*jn*Rfi!jBg>D|(o@6Ne_q(*Sx3jpD
z(Pfa-Ap?{{p^B<71h=e7fn^@igTsmY)-U6elTHR>uB$1=Ow*EM<n6@roz$ty-AnS?
z+y-r#M2KF%7Pk#B|2#sK+uUV!{5AZ2C};vB=U-8AXeG!)NmHeU@yFw5^d=DxgC0|}
zwGF3nu5p)Q$@ya+rryH?gTYm1kuvvry#r*l7^6nxo~2!O);GFvNC;Rs1=Zw?AYTM|
z_RV*>3zf$Ie)X1X@H#o@aU08QR`8VDi#A<yUpzW?@bB^`BxQ1Odz}88;9WHO->76a
zOwq*1ZKep~$SY#`F>{@c3Dj7$xDk@WBuBLE@v!bKvM;(wH&aD6&W<QtQJ9#FkE6kS
z6{iH5fk{$3WpD1q%!#OJZ;gdLM@x!M1ar19IQOR~;R}&g3}Jfd$vnFLrnlnbBAE=c
z-ashaB%7F}`C5H`4N7M_#YeDj%K=&2cMPgtX{+oei2tyy6iR?Ub%9>F(L71!n8&ok
zbOORInv@Y+*@HpQdl{-|2uWBW))8Ms_ZpSrP=6Ki8yiTYTL&C2dA8lVHJ9!P{CXaI
zX)Of(J^ysZu7vM^F)QB;6T3Q5w}#d@e`!YvVwDPIIk50a9kz{Mk2dbh@q4E)*Tiy0
zJ%?cRV$Of+(J5Y&86Vd852zT@3u?$iM-#-3Kx#0bW_H0nV+iN8h+};;e#Ml6{kw(Z
zriy}s&jP!$Ixb`oobmTnTDI^?YPWJ^s)0xoNXE9c=JJx5_WrD#<*kLrE}{ljTkl$&
zRV0o6(hYhrf|_kEvkR6uY;)l&i`(?LBtfV3@xCKK5}5+o+rq8V<mI1-@@D?;zppC`
z4p(^_9G_!5A@yHqU#m)?UTKw-24o|O_CeR6YV$kQ>WG2)#`u`4nsB+9-xIP;Ocfdj
z>MkC-AxneoLjMWAU_2kr`~*v^9+zu9ib>}_xxDktRz#u!zb*@m|4=dhM9LyMhv3XO
z(1hNIjlz^g&S`bf-B9#-WSw)8y-h?}`UxQ~wOskO#crFpBkvM)S<x7jTJVMw`{fZ_
z2z&F`TAM{bo}NuFKYx57hF5$4#%2|YhtI>HI|Fc}xeMZM4uCThW`}7p#N~Hyh@JzG
zmXBD4Rf8Yq75=psB-ajBUvHav15~5=nT*{~k>`61t>36E6kDF8Ku`us>{+Tnp32Ie
z#(nUHYR8eOG6i}KTU&*4Mi6bS3P5p}EH>eG*YYFk_MVDS56<*;z0)iJ9GP&pTe834
zM_5~Prv3g6mh7H=md;I5@=s7$KcQ0?uF7F^&zmtq%S-HmARV*|A4Xu}2$bt_z($n>
zhj4Xn9)aT0<71|jF;51c)E$#6zkb3JhvYD`ZH<67@Qhop<>xI+GO5uYp|S&Sm~8&I
z(`O7YJk{o}5OxoOlY1VZ+gDP_{8r}~1>m+xZcGl~pQDN-Z+nhs@?fvN%Ax$J`s2Nj
zayB4vHr`1rr%_@!MjfN7oPnu26)qQc_?b!p)IKP{<41uZ(^yrBN)pbn3CJVe7w0VZ
z^3OaO_6-%RZVF*?A=0##YxB(=7a|wN&!Xm5Y)ngD+;?5+srPJ4OIT<hByD*eW5f~E
zX=v2=z%|ZUpfnH0j%dtL#W!JiOIG06XK(7<;Pi<<gMh7xhSxO?_bP=36wgdiQGjmJ
z0pnIKbSj1Nb!m+DpP*<$1AOrGuq!sU!nO%MHBrIKns7$HB;m8PJz9&m329C^r9Fnz
zk-F;r;!oS(&S6zNqKn!Fv6}Ebh?6T)D12z7R8@U=N5BCpXxC&cUnS!(m|uy#?7;Z>
zdCi{b0wE-FP@OOmoGDZJy%VsYyJ`ITWfh*+rKYY`HC-U$P<vYUq1Jy2fHcV-iz<q6
zctg{{Wqy$Az1z9}8iuMO`iLCX^fI1nl8u)xw9bqMAlI_T#0#7(vgPtzq`|Wcf6tjP
z>DEGeY|Mg#VRgS+T;J~Dt{u<ZMG8HlOMk8SwxE32g5tafw_`SxR`k6F=tGd~q|MFt
zq8eweZ*My5uEIa*C1g&|;r-^uZ2lEjL{7hbYXihjwz9y48Jcx9{K~+8)gDJByAX$z
zJHb{ZM9s~y-!|cgE0QUR+(H!QwmpSAvO*oyZ{=(JoI3GHn8ayrPzTfeyUN+z!D`xe
z;{MP}3F2*S*SPF&pZl<BUL!CjPvaZ`jCWaUm}9sXpGI*CRj*Oh2{iMH-K3i~5H6U3
zV@pX?HRB%PK(o$XFBDY}u9^z0HX)U`E>RFm%FkY21kLT@_ce3PI?O>qYZekRYo?i;
z=Nw>Ztq)HyS5pm!KPIf!8Z!}kaRR{sGULr|BT#>E=v7E7RaS?-=*)R9=G+&Fb)a05
zY8!!FU!FlRd-3S9-UzQo>pWN-Fvv6Wr7q*>`*_Wl@id1X;p2T(p)M$RX6phFQbbON
zn`=2b8V#F?cG>=2-bPw@`Ip#^{I&^ckiiyp&dF7fwi13{VbRfTf)BN&u8ogF11nel
zmYUldzyoRDiqk}-9~i%^XKa4_c*hI4_tQ8#WWS65?>z;7aE%JI0}@f+t+*(9+1<ij
zOq#8~^B#}xpBuhjHopLpaL;Mix|{3Sw}bj^#(nB6c6LMHOFhD5I}$lnQD0AT7xlky
zw#>x~f|K@C0^NxH40z9}jv1h9DaRy7HUYY^Fr?qB{3eJTzz8250M6NWOJ`#><HdTH
zW~s1u9&Tx0dj!6JV??Hh(8jr>xYyNmzxI*q3^apv>AJuo^%+Nd+8JncJi7Y&fl*Hu
z%Onl1?ZF55DeiqLif65}Eo}c+6kRkO-Uw>Qs${J4%nXx5IT^_psv8ke7Bpns4<bv!
zhf7a*ZuxZfIbP?&{YthNwTvEJQ-~;dgX(`HLnXs<Vb7Tl{p+QZOxJjah2z@|L4h0M
zBjGDkgAUn8l8x#{_<NMsu((yW&U5wTD8nPRcLjlhU=oLUZNKe@E+wNc8mrWQdeu=+
zdn1E=*Qd_~S$`grF?%8I{9TdsPlo5&!=}Oc`%DruVMaN}^shZ|ntlXNOP(QzvjsCN
z<=_fS1Y6`%*Uum&E^zAwwa!kWiav`X(B3k|+{+uw-(iQ%Aj!_aR^?Xn!SX!w>_=Cz
zVIuZMpGLwn0}cMW(QP1-p!xX>;%d0O7GXUU_|g0VTtj%37`+$Ok1>1I@uZ7nQ%4A4
znyqq8;bvgzzf8I3;Xffzkw|!QpQ1+6jOMRcEiO3<sh*=FGR*oa*>~V8Du2Cp;RQcK
z7=rdlrw~-A*M#ef#p-HnsOcU^Sz$!%b~1TqV2Q@)&vCmnTQvQ>(3bMJ%`!up8vbn)
z)8kH%%jH!j>uTE3rVHI;?{kYY3jl9ae>kG_T&}{dtY&?NT%lwJ9jdi$#UskYcA`iR
zTVJFw#U@tpl%<XkHHb(LF_(;MYEIEVp?h!X$VSKU-14@G#ij$fCCl?YUvg)?wDsmU
z#I<QbPdYmfol%etl_#M7wPyReLM`7_P3`3<_(JbS+?9b^RL>7w#}nJu3Z_@J`d;vS
zEwgI+QDLD7R#H?bz5T-j_Uc7avseJ`eW%=xo%(Foz7uX_zp)<8quXfd<doe6$#z~Z
z5Mhmt@vnsGyF_V5zI39GxVcmA6V>>v_QjGoP@TXWmCAzeB<J?FNUaRzAZEUd!fp8c
ztre?anQ<%}$$2GAS`B7G`TBy<k8s)i3Fc~Clg+s~r1hSjeQx|WS1wy|tZM)y=LsOh
z9w=6RiYqI_`fZ)_81(5Mw)vUQ*tr<FNOF<#ogalFcWCZX_=yWL!2dz;x6|<}H^)SN
zPdHfCU$^(&CsQ^LF-Cv<ZubMICn)+*$L66s>Y<0STc8JUqL72h!KGmGQZP9y*_+Dp
tx0GdXO2S~uFc^1o>Hhy|;D>SdMu+|X4bWq3ra%J<1056XDovNi{{!6=kVpUk

literal 0
HcmV?d00001

diff --git a/docs_new/logos/mit.png b/docs_new/logos/mit.png
new file mode 100644
index 0000000000000000000000000000000000000000..60c5528e8d7fa4a812511189a2ede58ad9d8b2ae
GIT binary patch
literal 1483
zcmeAS@N?(olHy`uVBq!ia0y~yU}|7sVD#o-28tA3Ir|$(v7|ftIx;Y9?C1WI$O_~u
zBzpw;GB8xBF)%c=FfjZA3N^f7U???UV0e|lz+g3lfkC`r&aOZkpoB+&Plzj!UMz32
zSl)DroH3BGO2%@joGy?tSKS&&E|ND{qF@SS%$zy1v!|z{w|A+W=^}a4<;te*?d@IN
z-9V|X?w<eu|6g=U`3f{kv?RzcnBhcp?lvL8?Yw6vg`Z;$vQi3?72GZ!!!HF?z?kIi
z?jp{8ZAL$k!(QU)>&pI+gOgd+nw{%m7|=iQo-U3d6}R5rwG=w!z~g%GU|Rg%qW_D9
zCPz;3x$3Z=`;Wui)n_DgPkov-<*3<)9lnuwbmi|X=aKVTQnqe0lR^Un6UWdecCa0I
zthu<D<H6IYbcS^d0gVhS0uBs}xQPj(U|B|x-?>041Hj?H1kxn}3X=t36Iek(qk*Or
z)8uTnirwj7IPPwIZ7jv`hqd9c@nX%A<8%Byd51^v4w)3P&-9CADFf3F$pi1%hkmLa
z@L>OEc|e@6D5=&*q-YMX^ieHwjVMV;EJ?LWE=mPb3`Pcq=DG$Jx(1dZ2F6y#=2oVb
p+6D$z1_l?tX<S9okei>9nO2Eg!@|`QGJzTxJYD@<);T3K0RTnuA)EjJ

literal 0
HcmV?d00001

diff --git a/docs_new/logos/mount_sinai.png b/docs_new/logos/mount_sinai.png
new file mode 100644
index 0000000000000000000000000000000000000000..f18af0686149081a01f463017e2cd78223925260
GIT binary patch
literal 23749
zcmcG0Wn5HI_wJxbBMd1e4Bbcz5&}ax(jhG=-7TpC(g+d?0x~qxDUFf?0wUc=OG-D~
zJ-+|<ez+g+@7@m{$C-2XS$pjj&wAG0MyRVQ65>(gK_C!9WhJ;K1cHHtKyKONV1aLt
zm?^Q~FI*QTeRl{1p9K8}<JWB9B?Q6%QHINC`%G;&2YoQ=3|PIOAi-e@vn_w$En=c(
zkr0mgLuiieRlwG!ZK=N~(ZoRY<F+@?j;pt}oGM=xkk8I;t?(Ayx%gY`j6`si&&346
zCGXmIcf}gmN0&L9{y~K``p~r#d<*-AaboI#{U%G1iHezd<9cWmx%T7FPLcb@$l1{D
zn(sjN(9;t!&w)ZM=YaX510Gl|CM<-Qgfjmx39&3r7BTdHE-*4iS)Bh|LS*4Mu>V}t
z!@;M27h*gpc<$dt7RDGt{GSUu9DMqB!3hlq<Ndq*Uk$q0{nF%?Y&dV-oR{C-$1+#m
z*U4$lb_Yf8%5gYY%5K4Mms3vyDv#PTn<VzWM=Y@gc@L#^KkFI@n(p4c-gge_%sNfP
zNQ97KMEhMg5q!q|?Aq5aI_f(PM0!V1{;*c`q>N2B@O-xYt=iGWg#yfFkEQj{dH1?6
z=u8>=miC<oS!YQs5FI=cAbo;c-p!ul+xs_@`xBoB9msC283&@eiLNIVj};jni$Odg
z!x_>%5F4MsXSW{wIh&P%SWS5Wg(=~k^(1)*9vRLXHauCYT66Xvj<`8#tkb7k(_tJr
z4DTzB^e_l$dH*baI9jnfl8gy?+qxf`KwN9wdNUg*y|<;)kQsE|!?js@)kdeL$E(n+
z7?O-(<OXJvbP+)enQ=Q%DKT_vDI{1RIif2F81Z?A%fu+F*@4aWrA(^c-{%d=+aZ>6
z_T}_83k3Tmya0zG|N1DK@Ut%%q5otn`!i$FPEHt4-HgB18}F0{jEu~`{g-<mwc2sY
zLR{caF8}W3PV<mwi@xUIc*SvRV%ECxc3U=Bk&C9v!_O^R8&zuP#b$<TAqM)~yJOSc
z`Y)s=D~Sm`t-Jf@fN(0AaXgR@TGN_?yKF;LOs`3@S;d#0KpwbmI@VG8rr?6Lz&zML
zNp1;X&XK?Fap!yLaMg1y3G>P#?bLRvKg&Z-?L&?a&sx0ct>&zNi7{tk)+}H8{B~{)
zI;h<|X??UX-l);1x%**|_<YLAj()bFu<elwTBdj9$)8fzV(CcBBo5wV-wJHD?-TZv
zD$a8l_X}JRg=ICCDbG0y;H;)HF|#`8xtXkH2K}Kdojw0J7;wgQb4iNfJw*Mt4%@kT
z+w1*$L1*LBdr)YCx`Tn6Nk+iopSZW>>rX$*uNgB%-bEV-g@d7+iOEA(TD~CXKhqek
zQZ^FS`BTXfna+16DA8WD$cP7Y^uF({sFCPZp&wxjQn11lZPG=BgWLu4C8-Uwis2MK
z<7kzH5a+V2tweMO{{6PNOYc**+MTvBpS62caOs+q@qAmIid+?WY$k;98tsTs(i(7c
zc%siiP|#(WkhIC}S-tpHdq#@xtD5)ZMFGbuXp8@P`hB@9u=j|c?`(3@*?@N1aEL5f
z>}Z~&hGOW!%g)th?~zumjT3jQT^1<tX=cujS6=c?4GRIzH%BT1vA+DO?ZSMSxRBA#
zie_Iqv%(|h2ea9=$%;(|klK^W=Cs3VQSqX`JMXgwv8$X(N$PP{Fn-8c<+Lo;N`G5K
zyk(SJRlpgEZMw78@^WA2TfGWbV`O`aH?V?n$d>wzv4S-nV@XFY+Sl)-!wZa$(Uz?<
z20^cQAk1!&@C%+z-QJGj4HNQ{Sl!i8g_F=oUj`T@46nVUZq}rGLw;u!GIGwd_L(G4
zS2x{)&DHLscP7wJ#Pg6Qja!g~MeO@+zx{3$nY)|r<^Ptx-0Hu&_DAI<8OwD$I5lq{
zbBZG*4=alq8+T6d?PyN$?lZyAFQ;NmREz`qeG_z;`L`o3=*7_o8rWa!w}DWLvLm`{
z<B94@#HRdk#$E<TfBRt8`>eG0q-cGI;Aps;GR-+HiC$j?bw#no8a`U3zb{Rr)t66n
zG0_4fuV#a?%W#4x{Ce4B?QRta_7kBEi5t3Kv)t6QVo0=9-1>FCm!Gd4Yi7U7-9tL?
z3`nao$B}^fFd&Wif-&)J7a|MtcQH5R_v@(rthjpUtw%&FPg^vf(@v)?%`yU`t5l{7
z=nB33x=T9!H<s{s<@Jn<gs6>CPaLjs1v5H6H6UTTLN9pkpRbcou){xFn1m6EFUbaO
z=>Y>-G-g5%!6bY3S)(fgpHKE(F+K)&z3>nXk2qR2T$fiTL~U&e2mAe-b&JLpG%+ot
z;9mlFSy@=XgTDsHzDNh$G)f<>+?v`}zo@?<s<k)tNZZT%oIJz%Vo?8M%TA=U?+5RA
zuu94to%7Ov=ko8LU9U1}E=E3*rgyTgr|rPzqdBjri}XCDd^><=e4H=R62=$N?z4!Z
zH)+_O8gyEjWp*vkND1!fueFzMJgT9bs{8a1HW<IzaemaMmKk(%);fLpo`)zm>(Y8a
z@_0V5V6NNH{|_;5-~xt3t*`sSm1M22pwwxP*6k|Z85a+UVis)3*Mmb<Om7hV4jU0c
z-$Y(@U_7^BsSr3&v!6dp%g^wOJ2{~VZ3}Jn*m)kbSH`KO%zN{1s@e0tc2#v-LEe44
z81=;N5G+gM({hR1trx$1)++}ys$K)pzXYDd1^t;}YxTbwcy86t%$p^wXuf2zdy~y}
z2``wIyUCmn+PNCPajvP)Np7`Aoh8kN$)EfX?i6kf{G#{Pk;@{hZw?rU#My~-i|MpZ
z(D~cqbg+^FgNlKhTHca}TL+W`7c)1P*Y3OX=kZs52|joK&HS8a<!8<M_k|SCZ?!CS
z3UJt7dv}VHH6d841lxc4X3+m+a(?*g5<lqTDy{RC7jn*X^5s$fZjia}O0`U8{PUoD
zK3*Int}m;2{q-_lofJFZ7#~l%>;4rwXEO-cnW)gJZMliduuwq{sc+)D%0|>=`p)&d
z-lDr>)Z2e`uLI{#P3G)|HL4_nH>U$z>LMj8?&POpS@;aFq4lc4%(M}uL|_cKo&V<2
zFMaci??WJ;GU1XM{qu9=&PLH-G3__!z&)<4SntK}QGd&BQlf6M2@yMks^ia*m-Ws(
zbCuiYSGzAt&>~Ph<q_jkE?+Ngyo{swJ6|iir{a3OS4Y`y;BoOJ^Tt~AbJbrX(;cyk
z2lw0guQme~9|q2yzh$-O%_5HfVQPme`IFV{Fyz>HTxFPPBk4>F3#neGSWY(bhBgX_
zM-5ROWDmzJ_2tuwNS~?&dJWVVN*{=gnE=lyWGi4<V^NJ7=twvq`HtlB4gyi)%jiWA
z>CVj8XeJ!pW46|Vk#5@|5NVZ#&484ydLA;|Vtmrq$Be20tGg;Km5NTN{*!C-;oefL
z;5V#~_(9TRY)EOXJ9HegM8ATim}RZAMxJsc;lTdq4Jw~#J<wmq38jM#S2lQ~D#KGR
ztTD9zo6xZA`tYi_9~g+{TOPzf6e5Z@8ol~)_dAm^8x7a8R|5YDN2r4SVQr;Qv8A>C
z^a6XX&j2J*X5ooi5R3lS&EI%yWVYAPe_e_wPo7C#U)-Y@3-C65T-TLw0L^9%4tOI)
zI+kz{wXWKD6VGZd0yH>cI-OuhYE1jct5<mU!jh}#N-gLTe>&6uyqgkUv3qiTCVGBl
z`;KLeJR2|gPHXFAB{1r%gJ)yUz0c^SyQa{u--BDBe0`VeDOr(Eex3E!Wt{$q53d|1
zDqrsnYuvwTQUMv2v}A2w@@kLI!J3B+EV>>NNDPn9IhZ1v+3OfqG|{KDst@39NN?vB
z%Pg?D-!?EeKM8K3jVIY|^}Vbr{sVMF()UH$Pzog7I@X4VUR#OtAJ3N*i*f&;y=(sL
zg5G!mxwU(B;QW|}Pl~N>*-qxT!q(~;A?K#7cJilr;kaKE=zw>N>25XRd2@@$wF2xD
z-GP5`_kdytOO%BlzH&_YxyZU=++us?O)oB{zedelf#4N(u<&sbC*Ltivq^v<q~feV
zM&kQq{QnAAcQ@~izTd4Klgn&34)X5qD{EdotUry(k1O`ux5(^!JJ}6VLCMYLRGs8K
z02R9V_%40A7boUu?WKY0bJVdP+=awB&PA)NeIh@@D0^?c_$>C*)Ya}XxW^6$#<MUJ
zr0vRqq+Aj*I@tdRQ70Py9Cds9GBaSkx6jz<U{0y}a_i3$q^S5DXeWV~FbVSX<a&oj
zG>|~Uzymz>qpA8uOvWsXyP*LAoyB=Y6Bq4$^9WigQ+E4a+<EFr1l*fDcAICPgMo~Z
zBp+vG!8%IbX2_S?^|=hc#9Pi_<!6j|knxs4)`0-e(Z<SG)E=&%jj<k#Hm^$8JGn2P
z%#cQe#jd0CX!%GMa10H3N~sthzV|?xHP)XaZg1ZTo;88SRLbmDERPu>-eVbi-qKCq
zj7l*Hd;aLgF+S<azxX5y&epWT04880FJ})1sm+eUi1GQk{$2ESWRVO#Nx#ZH1&@~W
zV81%!7;lFOVw5Bpt?36^@V!DH_@ZWrn?cMA-nZ8I^J_?`Xe@72@~7u-u9H1=wffC2
zBC4?D-60Q3hCdT+VJ^G*tY<&dYh9B{HYx~@1!WO#l(X|p+)5z_*H9-hVt}uq;I2*2
zxPhA#;kn>AS&4CZUpdCJ+Cdy1q<DHTDWypWK5wrfmmfVa>4)4;vw<zkdLXm1l0R*)
z%MOFA+JhXnd+@yJP=Jf@TmffGB8QHf!2V~AIL1kUInA7K?NF~Iy(#KU-h*6<1<V;q
z29rb#n0uJnpa^4l!}Q$(313d|r_jW%J$P|kT3jY?J%#)ArmhT6QX$_A<y*)byWM|a
z3xFyKxk2+@;E+6Vl&Xc174a8taS%7&KCs|TF}t0YtVYCxd?k*t`>9P#>;0>~Qd`tV
z1rJaUDlORbOe7qb+YM1ge<47QBk@NbXMFw3c)9QVVNF)Jdij$2P+>x~ZRYUpEwyaX
z^5v$dp(qa-fYVPS)E|Jr^h%zx68fCfIJ`0Qg6IMG-T&acq|1UAgIlZ>tC^&hl-QCw
z00NZbWQejsW8}7GS7{F(ar~H5<&3G*sSX|QAmKyCm#h}NFz$LCZfBn>-OEGzYY7MJ
zwq=&t;N(w1*q?_Z&{eFgyj&J=y>@PK!x*2GM{3R%*C_f$p81y#XJY5XUe(>4v*r)Y
zv<OaH(0|>GvJRV#L)|mRK{%0Dtrb5t3Fx-IzorET(J2~dvSQi)<@)^Obyg3J>*&2j
zTMvp&zj4X_9dnOYXRAksZ)58(>AMGNN&5R^S|i`OlKnS->54<zUqB4Xmm4(LTbS=I
z4OwSOpD?u^6*is<fuw-@gf1Bv?3GDFwFBxmVlhSs@m#`{mcHTQOV-IFn;$Ff`vgta
zEdG({Nm|dljW_3s`{{%HpddSYV4MCVcpQ~XyCZQGQz?BgYaQe~eRMJ3-yNYI9&RgS
zIh$~>7@)|8u0>8Vu9bbFO4Ypf0Kl54HP45&E)^dakw1%5iZ}_{=!^G+K)xi;pxDVE
z=)y)ZBjCFApb)@Hb4J-AJvS*2zjbHwO~$I_rit?P%gU&E7mt}DkVERT6<t@>c53G^
zAX|oR_NC)-CbT!`GB`<%5+GdTjqsj?&2hQ9(QqO=g2svJ^@jJnNObf+^TFS2p$WXa
zDZ3f1VcYI306O?r1XrU&!8%+n-q1}-@-Sdmf3n=vKIjc~J8vy_gLNr~f*tB?Z%+C~
z;q>;=4@k8YYS#VucZ&x5&kD-vdB0=lvqYd&DrVUM?)v-5Kb70HYAdR4JO=jLy2>eO
zC}z=N!<e#Zvvn$b^*+_34SSSth1F01Fu>E<gSj&~n&g={Rq0s1OyJ~e8G0sSj+>*I
zEtzUGAl3$s{06hh;;vhsO4Ie~zdb=(3%f}@t#1jRs8~NXJ{5gyo%k8+#mdOI>v)*e
z_Ml<qI)&6_U70a~sn}L5zylIO37h*PO!M<i|M2XR0z15ehES|-8M5`>(5-(bNTg(|
z=oXIU&_|JFV_5?a1f&SOs0Jg%AnS0ag8cH0gxO&_I-7o1@UYT3Rv%iIw5IQ}c$jx}
z27q*a9A7}Z8XXYS1R6P<U^x5`>rWQ8ZBJuJ`T7$#L}>|J$OUdqjq>!}Ss!`;oxNYr
zB3jlv2Px<4QS|^$qIPqcbTl5g5X*LYs1c6yM^o`>@~4xOBAK!Vo1R9@>UC9-Wi!Tx
z&+$e*;ZCYfDBm(ya*Qo`Qea;1)oxB-J3hPmL%+Y5O$=3#v-(=Me0&HZpt<(#2Wv<e
z(ZpxIQ=#Q%X!?64UQ>A8*0U~#<HdmbOd5SRlUSt)dZ`wVN13u<<s6~Cw+qCiwmvJ&
zOn)ZQxkp3j8;sQ~@ff~_vE>0p-HBc=0&@}myk&5_*jrd;anm4|)ptEpVUh0;a6>QI
zogXg?S4a^zFJJbMss@=GUk>;_VZGU#maYVRlA+1SWdf!Kv3izgG@8epLR<3bC+Qx(
zhm&iAvuFf;`cN~xXI);79HJFV^r#v-?)cMDbUBr<!AVF|2w6pivQ_Zt`&mbbc+)$_
zcSob3XWcNW?!o4*(D3MFoE5W&u(n&QF{eMVBqPI&vSr2{d$1KeYMJWi!goNKGrv0+
zz$5m)k_aI~mPn~e+s!q2xr`x|PR1x8J`%5vP|eqfE`sek{tVz2&$cOLLq1w(OxUbE
z{RK0+Y)d`%FJQ^n@@gw(S+lx6O9}ivRg!r|ORt6pMFk2a`XEZxvL6CX?=DuXzgBR!
zXKY|+s^{I`Bsx*>U`?KBtK7_Ez-zCHle%&>@BrWm=$T0puqGtIR+NOkDc-Zgm?nX?
zcVLyQhc9@;#S=E&;JqeMs?!NH+FsA-g3;pk%^RMJWa@$!@_K%)Sx$xu9H}^GpFWqm
zTpdK4NR!Xv!|Q&^mScyXVNLm4F`1W)EgL{0^bE*$n|JjAb5ynkd134Y$uN>?`rYtH
zZMLsFNfus7N!V$I(rFi2lw28{^-{6@;?3W{UJq*brFrqd-vG#%PXriWw?6k>KjrwY
zKPf7^Fh6@TlSbKn9A>igdvWS*>%R0Dy}N;H;L)zhr3#G|8XnR@d)u2?L@8@UcgTey
zr1M|*EZ8VLq8ebWdp)Ryiu-vQWY|;kO-GK?ne}<=TZ_*x&OiEe|MsN85c=)~iA_I<
z@~-E;=zK^x-dgS&Fi}YgfE$G;FzjB6&}tl0YzRZjo6`j=RqB2B=f?9_yKp5~+3ob2
zsKD*ZIx*XxGT*ZwJ3;vr-N(9MI1`Y~ja)S5lvmD9zgSOlJ_2qU0+%bSTV6mBTu-d_
zhyOwz2rV0XVOFdIIz|VV3bHWg_-u)eMg*?wrN->N9|*uoKQ{XX4IKmmaDZWs6?X9E
z^2fj2Ba{VOXz0cy>uU+BkZS=>5a{Is1uFUf4TaIRqtp6h<u53{T*y(LREp(RQe!jS
zZ>3(!TaK-t>BW6a>ALJyjfw?z!&5qa0E^(3Lq{NNd@%56iMat}KxD`21}Vyuzp`GN
zS@h4dV~{oTlsE@t%8o}*QGg@Tks1K%Zn0r^PyR$XO)>i}tSKs4e!L8}Yd*b7O@c~q
zk`@<9ab_NG7cHBQ^p0_?HgNR^@xBqPAH}JqR=CnjI6Kiv@{9~6Fi7W_Q~oBa*_T{|
zXE(m<4%<rB@8m^`!C1rSu{?)r>Nnb@TyE$c>U56*0{iTSuB`5QT8<_hB=mjj8d*f+
zp5xSG3;rKy809!=lp1f8@=9d9T&eV=VY*TQThA@fs%&pBk2ro=QO;G*t)(vzoBKu`
zF@Tt`9r_RS5xq(+lh2=h$Mt>6C>6}=c^X0##BQ)or=xYPVk1{$<;&Hj=tW!(SHHif
z=O$t;tdj-6RlWE0=qv7q7~%<E_P>oq08$NyHT7M!2I!ckA%+u%W2U0BW6Ew0Hg6kd
zV*$6^R0Zfle}tbC%siMPJquKN_Cuq$w^!0hA6nb}I<*f2Kj=^xFDQwJ2-^YBXYBpY
zecNnomVpZMD2yM2_!|pWK_1P^0?ZZt0)kb&+_Yr^XdN6H<yN0RRu6-Q>lNm4O;Nrd
z_0eR{do>s%3Yw)(sQegEnVy&cf~^~Pz<BC)B(Cw<c4QixA8iFbHTgwohAL^^QGDL~
z$btFlS!<QCpZK|9X9M7LQY(j%*<kYOiVZyVB*C^^kA|xwWewcEoDPpae;G9xS^BEC
zRFIS>hovskZ$L`oENk#G<z#9i#i=WtCf49fk&K`M0L~S(Xva@VutfP%y%=)rN(1C9
zB={fC5bO8MgYZZBM;m~SAZ`DXlP3%&)r@?gfoGz;_f9_&^$GsXApdf<kp~PZd6EP`
zQ-8pe`5~M#XA<mLB~u$u!qiR;<N<Q|%XfDY)%ACzYy2;Z>*?~%yr)`~cf?K`L2AT{
zp@jA|k^A4FN?)w#h3KwI%;{=NqUAN%?4L$Ifidnum6@c1S<=2+0|Wau$l36F{i`ry
zXJGi5=nP-R!@GtQ>6y(o!wG3LLF7bB1y_~XHy$2o<8!xp#POgJ9Msfj`flxiSt)C5
z@jXG3t##lnL*H&FcH}uz(?X{|xi=6YY+(!qtn^neXlVFDimY2UDBB?wUb|t6^PxZ8
z&p?KBm^5lWMyeooVymXuc<Kv!=hEJZ$ReEj3;FpuL&LweD>Kc>6#;u73OK~OoPIV?
zYmIU=x=-7$S6z?h`{E6N$+s>J3>u2*iozU2l8sue-vD)W45^dDOg<4Pp`T^rZL*m!
zqZRXviwOXr%1k26LQdM+m33;r71ipV8B=)z;2(F32M(q5-^Zu7SKWO4&LVH9D_GV-
z?I|_DV2ot!Df&v9wkY3hH0fIeq4>U4@Mu<_5+U;JC{rjXpQLsR^7l{G-6KG1^^{3F
zWN#s2qFAjVTkb~I-h$|vSay5qy)$$h!wNPA_;1Ka><1b*70<rAt;^joXElykZ&IQS
zSL}*i#ArgIQO^|M2LK*uwzWzjhY1Jhe(-!H3*S<d)9Elp>2tTbBL-;t9StUhi5T9=
z=L4{ZYc^EUYl1H6@iDsp=&A<Lc{@HBFMmgoE}FpmsO5yzMLO%Pdbj|OEC>vK%gHlM
zD>eh5?564Gh;kaRRLuM5r?aV6<DvkoeN3i<?A$l}Cf!5&s`vg+q~UR7@CN3Vhtb=5
z^jhjUbr@GN)GZeuu`X3Ehh3^Xf<FLL$1Go8#o}&s(mqB{#C~z@RbL?diT}dQoU{M-
z6>A+P`LB0EHbX$ArwUDY25!39Vx(zYfM@D#L(+Uu*G)3syat_-CW$IH%+|BcGC&9E
z>3T82Y9Lj-U{QM%J|!E$hd!z90)}%<ei&(JfjUcU##mS=rwaw){ql4=s;vr<1+P|A
zGPZvFVRcF#l!B#DZ{;x_vFtftJy4>B$&2<R2y$L%(LBcoyo?wli`3uAR$2qr^pk!x
zNZrIK+_sL0(BOb~G(pQ=CmZDq%`ij_Bm-;`EN@-iH4He5ZJEpyR?C`i#8CqtSskxr
zsWGb80pKl%#<eF*i{T}IQUjO_S>=fG{bMmiShMWkI~P3IhrA-GocbImZ{4JhB+(rA
zq9V=K3=cPB+94rDC@CtvcR~?vrB4G^=Ty(btTc!`eG3X1D~-2I0|@vk^L#&blal$)
z2%0_rI@q9YWv&e{<y`o)-(v|?&0F5Ipk%ix^@d1tf@mKnPq}gK$J4+LjDL_$GLlQ$
z+xw&ytZ?SLh93=&(HSvJXc5>!SMoarp|xKLRU~Lz(z%(F{yq-}X6`#Ksj~1lOD(MV
zdbWr02^t8;)s<iG-oNmjQ2kL}Q5tUg@BP2R8=1G0KaJY#6iI)OBL8mudp>ZHOyaC;
zXVTK*Ii>$KX^rCGLzuB{hT3)KjrW9t?(u`d6xs4L3cAqnV#kgzpIImkh1l*^_xT(c
z@*b6PQ<F<_l@92=?9RP@cC#g#j#{^x2qwDzJMz4mTpv@p*>WkB9}P;IhH!o*MDeF;
zF&f;~E#szU$sm6wCmcR&wbYErR^&i}9s)9)yVvDWt{=(2X!nKMHRWdl4FeL5mL}Md
z_`LqQMuZZ<ymzY)nq&S1WKM)V7h8(%ksqmAEK2HO%qzzK_Z+{z8%xxRS|dxHF5F!V
zZZ&iS;@RuGEPc|5Yw4T)jh>L|!OE2}#;AK$ygxzah$+SafnVS7*l_4^1CQ8qXiP8i
zK+i3Ux}%h(Zj}gd4%0ya?~f{ph&86;<*A%_8gN~nh=$sPxEi0ew;W`R6D!d4C-FSL
zY2o?e2;h#;P*kmaKD^qQ?7s~P<H&-V|1BRi%BCRFV=$qw{pztZizYB|`6wqPA!v?F
zF;NtJ_vX#^SctY9yfHpBNgS#*b_b+bWIRb0x1O6E&37DReM9bMnPlk^I{nui%&#)0
z{&XU)VSB4bqIny7XUo7t=P-xQ?VtB4;EgAxCmZ2Fn*k8;$>)t1`tM$VzJ{Y$M0r#y
zHp1a~@8uk#J*oals1Oy4cB7ARVMYH|nbM}n7tm0pxML{MXx1h|bCd!z^^0Pz;P6`%
z*f$ov4eBevUy<BV#60!)vg;pc^#h`)@-c;Runp4L5_%d?VssY}YgGs+9dnLMM*$wI
zLM}1XKE||t&+KF<p^6O&FhiTo_c|GZixypC7$FDJK17I_<!kyc$<B@fFtVmjawu1_
z>5wdqGGc(O5_A$|6-Wc)<%J!X?HF??;3IaH{5Ha~IA*}MHS&;IjKOrajvbD0QVgXK
z7D}PYa$=QF%YOAs&mT9^wOrrBDP)0elWWqLYs(SnQ%=0r2(M(2b<Js;1ng=m#h_UZ
zkJxa}cTFJ-h#ts^mwV?aa~_@PnRS(|lzhz&Hs&LnIg-$Alb|RHW2#uvY(QR>yiyO>
zR3)!oe#*hpn{W`p-BNWW*Ai4|gTnZ4G}il{z7N+$(|`J5H2ybUQezgYx%|^7!+G+Q
z-LfyyhWo@*NE?%<egav~o;?(Z;GL*m2CP72Swv>$W$Wd6`{U_>k^sQ=onBveJ!3*}
zn{Hoy0;!GKbnFhu?sZU)q3|#6E{-VunAM;97VHAhBLxZ(9x=#Q`*u6*jqt>BtD!U8
zKc4$fFMH#)m^`t$TQapmr;G7P7to7Hrg;bbZZYzh0l4vklcm{oLYid01)wAxhdfEM
z6#y&(pXvu1%HN=E#5OW4(#08-WP_@YQvkLK@IYZq8Q_du=3A$$*LQBVzm#i)qp4;s
z3Z90CmH#{~S2kp^eEGPCr`|Ve?VT55z*gwxJX<YY)9Q*~N%L9I+MhS_U3v0vy;T>A
z`ghMZ!$CXYq&r!@yBHsU`d|g(w%0QWRb%1S9SQDKF<$wpDUMqX$Dhb0VwM0p7(yBr
z1RkGrK-@W;ei&ODe?X&7tSn*{A!@f-oLk5Fm18o90X?|m1RRj^e97DmZck(CYskd;
zt&9uM*o8fU5sn!6q;2T-*D?slo|F(^488Jo)v?SEbAuJhvWiy5O77Z<pbO%FS!Ie!
z5<v5w_qyUc0rN2ih$O&mjE&GA1*)cMt#?32UAGzWG|Z8(@fe18frtSEA<BG~wPC?b
zQSw9j^g!C@QY$VCuK|_OdpmG<|LT)6Fe+M&S-O9w@ET_*AxbP&?gu(E*;ta!*1J+E
zGTf$Q4VJGfdwzuqmLG>aNLwsG7fM4o16F3qMv>VXvmwP+HWToZPXYJhe(SBvQt5ff
zb+3RNX~!5wCQfmpqil~BU*8T>g$82ahQ(MUSeh&Np6frUc{rW(c}y06UBZg^wgw3H
z%+1i^UY`2oPWR6&)?q``%I(|@U~dB@VU((0k$*IYM`UtG0q%)DB{fJ;Ft?aZ1sA}?
z=o?aoCV#53sV1P0+G&{?ECIn6OV6WAYh^e?mM!c*G1`rqFgF)UJK=YmM~q1ZS73Bz
z5Q#>iThG1OD_<W2)*xAWJEq=Qk%91^Bqh*lN0bi6Q*rJaC4D!km^02OH2l;B<|Rd8
zJh4S?)qOOZCvPp0KBvGI2Qn4An+AyG^?<IA&8|=A+|L)L=E2a*_=k&amP2<c)@7~B
z)ML(6y9Yn2M8)b;vco?{hW(ntDSg>COxq7*0eZKX2;BQ67_?h@<Y;{-1#}_vn-7O+
zxRd5REu>xzd=IM)XX57^=S33hQ)YebVlki?%)+Y!yt&PQrw}!W1^X@YjXSB{p9EKn
z|K4nFI)ArFD!Z!fKVSdKpII^bp<5_74UJW|PPT?;aJno}N&46CSNJeKTzAFLw=2LH
z`7RAR#T^W(s-^_vIQnW-q6R$2LR~#r??Cb5lN54r4BWcDt00_ogk~cM+{MC)OelPR
zJ3kbUtr%C0K=PM;b2QhUUg9aO;(Y*YyT^bjVM;+gU5@-u-Lg^gyq(ZSOdkDo6II%P
z7+2-@=V63CgS}zwGzS8j;&UYL<)5J=J0p#xtAQK8ygL<ptFsVYl6>92R)6Khhw12z
zRF^cwKUHfpY2JE}KBiEvsFZ$u4eAB>FjU_|ly>mQz)eGBwRD`fl6n-4RbEX~dVai@
zBx_j@RhD5Y7{~&6vO+%lp1`-xT6u3O5mUP}59=7QnKEM`N8hGSPwj4W4tOl-n!@2y
z3H(jd<hrtRpE8U8+X}iR<g8yOb-0HGD(D!oZo_xil%5>$$+$2hLpOv{D~q*oYTx6t
zBgS^4yMsnr1Oer+WFyNMv!_cuV;c$td>g8-(P4}lcs%d5jZ;qV4SGMr_&|{1!ji3?
zTfE4sORLwAHYNof8uO9J(>d|fm*a9Y$1qr<a?3AIoEpX7$y<Xe79A8-tu2O;_OH-Z
zgc}(KLY%|fKZUWY3bmIQJmPW)-pZ~AVKq?>h90HX7*hkeim={r;NOC8lRtItrx|#F
z%?$-K2b?pi5gC;dEDZ#86iYt}*g&d0JF&Dy{Q(2K-roc@F$h|<Rq&MGUQL}-i8kn5
z+DS+FFHW=<{79e)%agCW>d-giuJ564|3Cp4!cpP)Mg3qh<Y6Q+kC^)lTT}^W(4Dr#
zk}JZhb>1%9uayA`<}~dO^WY$2oA)$4kSs#$<cpscFHgNBN8NJrPn4;{mrLKA>4ZK-
zD8|gi-sWB`A<fujAQPsO1c|a|u6$kXB`*~UtP^90If|!;M{GJqw<x<lI&r>5;FBvB
zy5O@94A|bZUYvP0uITCLVBo=%Mn^I3$pJzElkHGU6LF^%N?+<N=Zt|tEt{TRO|_6x
zHDL2QLMg%alDT!}=h4G1{e9MXdHge=n|-%V3&{pb>@n`-`I^p$!gdiFRM1WfRKKAE
z=2H1`5k+6-+i96?Z4h;KE;>Y-EZW1+NyxTt9W-iPjIZOLM2r8D1&cm$`y(EL>&;3K
z$f%~EsfMgx9{ZFgAH}<EPg(QMJStu`(FlrUvA)YwUqeS_y&~=UTizWs&XoMdUhBVI
z$t<TT>9XW+@W5w>^Di{8sUn;_&kmI8Sk@Ytst7*Q%8lKcrZB$aR7GL=4avp06rDqy
zJDbq={gE07rcZVw>#Vv@GAL#@_EUZt1mbD9!P#Roink*T@9>C`BFmSJyexG~e9~TZ
zAXUEh8<+}L*gReF{&QbF+~|<kd<!*YgBkTxZ*SO5$oisFCxu9d6?B=q%JJ&7A{`Aq
zbiUTyvj)yWr>`0<t|?3AttJv0!$900Sz%kdY=^tilL$=cHHEQ4uNo{;Erev~ea%n&
zG#8({K>s<j30${6xYEXqVF`utbPwe^0*z2u6!m2m9P2+3<CGn7p(Wg|rX?iDkXZi;
z<<{d4y5H8+ZevdMcurR+M_k|__R+1Bbq`FyHHy&TUs4E@Pve31>Dq6vAH!Z}S^-y7
z(0Wpsf!pdt?CJ*4SIV+Yb#$VE2Y1lI{u&4T4+;2fNaVHn2?Kuer%hZ&omnF54kW_;
z%+Mvx6Xs4a%MOPVc1!L(!Fqwx;YpSz8c2weppJp@=a1I73#ea<&%7}H(EDpj{v!qQ
z!^=g_sFi#c%FmZfmZ-DYM=2EX+coZbfd~86jQn;lLfN+^_dZa6OHbyh7x?9fBL77e
z^P&xD|7|<bW_rat-4iBAVVul_)IRC3C|^c?Fx7C^yJi=^RT!m^)R|-urPPlF1WQ8;
zMjD|Z$N{IiM?3f(GpgDfvk25N6RVi<0yeIl`q!K@tX?WtKqgW~1GjIn*1oSpH~;Kw
z=(+I<rWvL)khB}#Xy_T)^0jVyNW=(P8~2#Q6OS`|pdeOP68G14yv$C+i28z2N0BnR
zQg6}I@NT1{;3EkbE5fz|L<ozcbRPOw*`2v@I#FC<HQ8P{p4-SDI8dH?)9!4Ry1ct$
z#tsi&=PvotfJV-qVmnLDBYVtH_TQ3wBs}#`72J6vn&MZug)CTk3_W<d>6USGT%LHd
zhOlEH299N2tCw#HEhay@Rah!V{W`@U$_Z=^!J9EKs^t+M*V@P*uJzK`2nF{#alBsA
z0k!Ryp9y_Nbkrlnc~eMta!CbRy2c%XoL*#~?0+T&B58F#%v^}_Y4qZU&=<o<cS(Fl
zH&=(U{*}(9b%LB1h2h%xP#!Un#cY$ex8QTU)>)Ia>qSa?+|||ISnTj?LZ$oF%ZpAO
z!%nD`cV-A)ShL$ST}kzzEW*<|ST?a4C~k*GtmPmvOPowSI#sDoe*E=?zKu@QN*oiv
z-6Br|sKe32b40H$Rxb;Ta>bc#GXig=T{6~T{xE{8L6AHHfaVwYPhwDirrYr9f`&#}
z)7Ia28vRH_i9BVy&od-xb{;>Y+Z<^fcaHqe1VP6Rs{TpJY}Q}SL-l0CxD(IEI?jA3
zLwi)2)~6DtEqt@Sf&xw$+z3+!&3wy@7wBvrPL;+N+4W$~%Ja#&5}pQVacwV+0u4qn
z5)M5zpxOH%erq&GN~?PPwPbrT5IfF0#vHDVra6H{3JR2Jc7_9HiqMSI33~we{ilTc
zRUcxbiH8ypPnj}SaVg|;d>nYq`HQ}9xA~G$r1$NW+V)YT+w;emH*rnNCc+T0yAm;~
zLX@2oMn$y2_mGjCGqK)4*VK>~bjd-i$`KmF#L#+&juKIZ>g9X=xy+nyS5Ejm4K>=o
zxbr@eBkyt7bMn+X2m>ufQy*~tavTz-i~1wfDYPU@_J+mUGwbU;WIeYK_<50Sln~Y_
z^zK8KiS>J0^q96NE5Xz+pg(Ra_rWeI)|@)rU?xI?lmou_xDoFp*(fUJbi@9&hoRdY
zNF!acbWO4mR(51@c;Oe0*$P9C2vX!d>UTh;ECy~ZCvD{V<zO9&Ir%-$kTzuI2A+D?
zdgHuLfER6GF=L^pd8Rn>-Zo~HgoNBG3-+|}L&HFWi=q*DAq$*eh{3jQ8R@fD2=EB2
zWmf{?v7F7LBAZTzyxKTX5^xU_9QeS=sl(hGMjKQowFVop0{II)V<%V`5T@8TF(<so
z^S1d>i8+8|aL_z54=)h3La{Ixl{*gQiWh~$HorTKgK=PR+ml0E6wBiM<@oE9fdozk
zke4cnbu7g<Ww&G#p=zynfdpXlQSNbb6`8t11Mo0=%G_CGn#*JbLKeO(kg&jAf4VE9
zoor+`{|KS1J#x!>sPo*Yv%ILZswQfzo0f0|Luh36^$X05z`J*!X8=a4&~HIAoEjSQ
z<!b}>wg%WaK>d9Rbg`V$)hy1Yi_AQeJzu&yIGaR#EpaaZZ5>ZB-jCkMJDN|xCVH5w
zD;^)le|Ove|J!m8HkSQA+Q*OHnEXG71OC5>n}Y>|%iw4%CxvGo;T90El4tScexe_J
zi}hIb<Uq5R^@}KsEi;og`1Mqz9SwFlHY77igOU+KM6Fx!xqFf$x4852K_Z0K#q*)A
zvw*CumNs3D>|hzbWGl^@4w9JVl|Zn;4=eVHUMNY>+hpb_4Qi%T?PJ|fV{HUqz+OOR
zK1T8I8Vh=j&|g-Nq-c3L1jHa?zDBPG7i^+Y$w|zG#ly+loTAK$Q;3szb~fEq6&OM+
zLp9P`B(qov9?IjyV?XHJPyi<^kWSi(t2W@U1UuZ8l2O<-7=7A;45@d|n~J>fgoagb
z6bHs^C@VxM!Wk(>ZD2ae6!OjkUqf>FB!3k17Nw8;E<>=$m;11g97AAVI3b=P#8$-t
z^hE=9ia0P1N=6PVnl^`?*+h%*u;koCLfuEQILbkAOD4`pp_RDG>pZ6LW%TKe#CHV+
z7^Y(1G~zRlo>?b>mu$h|B8<`O-4RWUbsg0bA;gu*)qJ-#<i19ak;;ZAY`fA}ov9d5
z^uv+dJx%AkP1LPSOyTp4DO>*WKk(xd-Ar&QLR|QwlNpI4DhGErIGjw~){AW>jb{bs
zcT-QejVAi>q{?dZ5-XEFyYCzBiLAK^L@0JNF6SnCVs{@$DyVug8(@u>jcD<8VUN6J
z@z9lVbg8_HGKY^iF0@;&@#4bVlKu8gU<OLlBx_InpKKVtW_CzdpVl6fN=aNFZ#N+t
zeRm$A+&ce<n@hm@7l%fS>o>Kh=a)|3KdWL(dVM%(rIkx2)+l4v_Gi2{^EAV1izWv6
zY1E^}vel-K#J4@5E@33Amq4=@?hGk{SnjT#`ORINKny3%eGyU@Wj<2F1p0I4A>)tA
zsK3{SZi?I3ht=<BB%#U)`q_J-OBGRa8=ifO6{{Fz<}LFwwRXRze+W6{TRv7Y#K{!M
zcc@SbBxN9R2CL*|x1CR4F!yEWdkBIv;lrCr?k*0<MJVwB^w=vR{pAqE1>)t-s(Q<&
ztKLpP`L4e5O9T9~D-IftpmT0)sgy%?6Wni2uR7}qOo@UIauR(xtTY_OASrEIR!hnd
zB0QKKzok5zJ&#1$)p*C$$!`aKRenZ&5@b%!rSfc}94Fq(pP%a*KXWG)Wwt%xSj$bc
zVo^e1$xoV1uy1P^ic%X!{@x$8;Fdd?!1jx!PyQ5#qoUw{e`(>ulSz$&mFw}7NeWq7
zf=Tx;df~;K&<IOyM(wwghbMPDx}a0kCzd9{FL>6c{f!&*9y$oTNvxb=tF*sI%jeMi
zrCTRVs(YYNAwb-+xrYl%O|MRr>SJLp=WUKqA4(;Y)xO%XB}yTh?0$n0Cuz?f65%EK
zYbHGZW##vG%D3_iVap7y-a#?pSgBDIrS`LbKWsCMW2gUV{?%DAve)Y(r2oz3d*mnz
z?{;ju+}{XRouez?!%S06TDR>t_Zwpd4t>YpKYMr<{GPlR&u)ezh|{)WC4^YU1he$c
zN4F9a8zB|-X}U)}cOIHZ!aQ9$@+|7)6n@u8#rLZU-<<%1aDn}g0zFCzHRJ~EQ>JG3
zL$ocYba!8EJ-D25Q<|=%E;qw_d9_O7QfSB_;Z>-HZPmErC;cS)waE=0PAL2BkZr;^
zd{{Ia&!PTZi3r9-B6jwe2Yi=B%7oar{XL)H#E*3qQY}@yY2lWbR1DdO&1lDci?2(9
zr(szOC;iP6&rr2?$yjJ`i%)~_8`h9xxZTeHZLyN+dxrY5FK!#86Lt^M1JW(FJZ3do
zxqbXSUVch$PVS+GhT*$O?ON`x&lmU93JR>#Sum@6wGU|zS=?K+;ILjy5$ckJ-`3qs
z;S3I1gs07Uxi-R2mxS62dIAD)KL%Bh|Ml}Vd5Nd75dm9K<6z{krXnOK=62;E_HX)j
z+|uiBG%GJh@q!+!d0eaMk_=f)d<^^b<ZRel>~AP$-XX5w!hE&fUA!S132CrxkXuqI
zH#k#yQI7TnWMXyyo>_=Zf?$#6(Iw_i+k+F@P<B#w=L^%GS?ybatQt#lcH-wg;r6z}
zsrWP;ZVEbw+$;>pH6MmhcEJ$xkO&z`!6VV#MmcO!n?7<mfA7f`GTcTq$BEhbD8yrS
z_%U^}I8x>8eKKag8TIZNK{~!GX$^@itdg7HO$a*={sN05h2{sUnD_pcR8T11Pyued
zb8n{95^oi3zOh_Y7W-tHN`eiXXzQUz=)Xgx#%q#fXO+iz3BFl_gJW~KyqpF{x~=Dv
z7dz^Vf{t&h;?O}Hu3lFjr-_#h3x0C}@}b6M!HMbjyvsrYC?o3%Gy0nxjiDT3#xktl
zX2Cj=6x%d(8bGy&dFm=eCf{a5-|-r!Lbl*p_^$-RSoDun(UGj>BDrcLf{vs>8uo9p
z)egRy%5h<-u-43uKl8M4Ody!6YLDHX9xmRbYIO(=nd=r(w>$th7d+RtjC#XiL-zij
z<T6lEY#cWbvRYAIWV>DgZ=jCw!pl;RRe-hlY9V19tk1b75PwAoCqA^xk5g`Kk)ZU!
z6QF=lVqr}o-Jz%Iw~c?iBr9Tn4G#ZpyzPBWEW+f?B%e=PR7}y)PJw^7V_uh8>&39F
zQf&x3{Wq!h<uHU@-gZIA`05d#c!PCGaQD4B;v@35JC``97UE8ay%8;+%=W~|RUgHl
z0Uu*gHRerY<RV;!$bQ6Q*spKrh$K!L*`I}NWQ->lMfKG0@dbw-2Y83|KPg(k|3${g
zbLkL^486j#{iOBL))05sp{-2qh?n3VGJb$*Vr1LugJGdc8q}xj5r@7RN7#jmf=oD;
zo{PxKKQdeOU@I3nW>9{C<?mCO6AeOQ!W!Acs(a<7V{H{#4{=`(9;BX#@`bQZuP4SI
z%EMQrS~Kq&!nBuOP_yLdhL4=-T96?bEZFchdkuo!9Yfb)G?`zBkp!l1)tD7$pf#R`
zcgpG8`eeh!=gDcHFilyl%9--eh)B&$TSMUunXaz9q#6p=O5vK7WH~%j!G#mn5MuH-
z?wca&frON<it2q5W%@yW-PbSLygL<G8H+6bdQ3f;Fo`hh%`E1|HjypQ9d7>=^iH9k
zO!KrIUN&kbD43)(j9EUtHs~(HC`J-ROxQSBFXdS{fXHx#k;sl<JS94tcnG0^RXT;*
zS^f^=iYg^=-?TBnfp*juwMppw^y07*T=o@#&ScPF|NUHP`ox$<FG5Xubddf~tnAK2
zg@Ce0`V(hNqx%f9Dw#HcBJv+~G~I8LRuy7@#eGceFk2Dl-)2D}6gyPzpOLq=J{w8V
ztFGRxG;DAD;Gu$8+jU>^=6jE?B@I#=H?Ma1f{nrtc)RrL7CGN+Xk#@yTV*y3ZyZb-
zB~0~mJIERzK160z?pwPTX>c=KrfnmN_J1tJeCs=Z5T%GS^nHOpdRw!%Vbzzskq~!m
zV5jff+>atF`bWswtMtUmB8{<6k%xH%afWm+BnvCv&xW0pTKc_gB1UFGji*gzgkKy;
zb~}9!T%733O;kC$B$M#`(IvPXko<CUw8&aHh9+6^Soi*KfnuVfr31+7g=AEb!~v}A
z+xK_Y+KoI69O2>I@H=BKpK&5LrVRxo%8HguzV5!zTYVBd=Z6R=E1=GNeO^D7&n9N`
zMSiqTdr_Kx@5t#iNg63`;&FL-7?3QhASqLqr(hhrF(ItLH{_779{DzwR9NA&pr)3r
zxj=frtu{KyeeETdxS~w&dxzm`+C;4+No%FYhVXJ+YH`Wqbj@CpZUPdW6l<o5qvFJb
zKWU7wWkcTYrgKLZ5qp-sN~|PUkbqneZ~HyKPJ)oPt3LTj&ctDb(RLrPR#SYOw6^sX
zKDHtdyiKut<W-|a0#&}pWY-?FVTG_hcgnP_^=;Y<d~ED4-8}!NWJIO?fOC;>u*ti)
z7cuQ2k@K9&kDZ<CiG8%vk`IdqmSYso+65mMSK+G<<8GQu=2kT-3ch!xEZ}qAUoSTT
z@$IWHfB40XA6$AkiqjnR?dosPZElGwjMSB4`vj_vnGtnncx5@clyMbB%gkAP`DT_O
z2kM8HM4nmsGVFIxeZE+yso4d#WVmmPpL)_v=u*2ABnaPnrhU@Dwa}a%5}FY!BdCIC
z{#Smyd*>4UzmFjQ-=9PNKZ~MYgD*{9FaK)vsQQ!87h*G3BJwPgrD9yM^k-^87YzGz
zajkM@95<vunPaobAjC5Y?7kqA91#!3jfvyKV6R|9=+%MU%oBF(y^v#?Ku4;_@q}1r
z=+k;6wOKv4KVUz&zp&W{+1Kn9DPsoT+lE%HF6L90X9b^|gr>YJcY-RwnV*&kZxQMx
zYp(jRMC6;SK$>XXL=l?35@oCrsYg`)kON30q~S*PYUYu8V|1()285?Z7~=;~730Y5
zxoItjJL$F~k0c*qF3E=9y7FJlGBu+IfW9fQva?T_9o|lJcuRa(6Y(>RO6acXldG94
z@4BgRl3=W00M&mV%xa!gMT&X6VD1D*jY)LUd%na{gCrt#OIvpcmD4$^3Yow#eM@Yv
zAE7BH|MU(-SeFJ=S8$Mg;vnQ%Hzx3VT^K(KV%`avaYg<v79$ZWOciDe>A{6<K<?8r
z%JxBCU^rm7WF%^Oh(gv1=Xd4SR+=AH2$Z%TN;5OU3|1$YvWQE-hA8pli66umQ*Q<?
zh^#oQ+8_Po%ylT7e>M{RnP)vm)HOEs9zUg<XjO4nN4`@H_ecp=8MEuJmVC2wuQ(jS
zmu%)3=_AY_0~Ih#_)%i#Cgu{a>1snw)OpRylHl5P1543j-@}!jhk*U!$|R>*@tR{l
zKSrBU(}j{?Sbf>XKRwzegOQlcB5^<$=%sajDKt0HA%{_e$OIZ&BF<rThYm4RomEhU
zM>@tM3yVzV3{&gRsG%Zfmx+>#E`so4xVTK5A=_{ezabFX=ji_zz~W<h76Ys2rrQ`#
z!o+8$LJK(7wHLDmGbgpZ9nSCjN1vyqyfB<C!yrO}(!}2z<N>0&|ABR`=jG2hrGsMI
z7kst6;OxD?6Xsv3TBKxoLQ>R!9ye19a#p1h%l3rhY9vqxB{}afe8Gc-L?;s`k?3X!
zi1YQ4$l{<Hh&c>ZMc*2Ilm9rv;KVZ^fIhe1Lnr@cFqZ4Ln@owCb_aKN2yp^Gtd}OR
z<x_A0@G7OpB%%R^?o(cTee29(j*r+4@3#?2v}U~5q|i@_F3YZKA5zt7U?h(?i3&4(
z8YdSnrLULeNt@37Ej|iFB$kkeIHuxZVS~Gu`x>ox<_~UJE-c}}o@`tbd`jjEfrwM_
zq(U~7iCH~oqn<oGXh>)F<eg>tvo<Ow{QmTpjge>6J8pTn@e5F@V6Sl9o^U3C9{Hv<
zN6qhg3_j<g)*_5~PBeEhODpuOQOEuf;iI+4sjk~5J7N%>@lBpWKcmQkc31TXhjvYU
zixR_-2aE>&(UtyBpeN#Qgd*>RAiC)(^1CTywdL^CowLtA6Pr^!X6e1KQ-0|A0qX%z
zh~6#rT82v3aUKB2qzIoOa5N%%b=BuuO-J_}f8eiMA~LI+HmM+YU~qmsemwS12)$-J
zyMez=Y>7QPT2z&|RQjr@1+kdh6MP{bo>AbtGRd;6vqTW~TDIAEn36Gve84U@)Cx(4
zI4H8od73=Y-<rgO3@BKm?z~UQ^e$-V9f$O+*V0o_Gf|KDj6j4IQ&BnG?UftC-q_dp
zIL;V<Vxc;takGP~SvS92qz85s*rb+TXqdCSRzJzHb)l<%KW)dmy3X?RNh^8^#SFPm
zNAlou3f005M2@lpH6;7~sjygMxV6Z$pB9v(fxXn@P4)pgqgRe^C3ekwHZuezPTYbm
z5^sNc#)9t|MI*X&p86P81dhZ+kg|+E?Ea<4&X329RkSv%roFNL6k~@C^XT5-bWYVU
zmO75lr{v$hH5bNJ)uTL4Gf!)oW1+%ybk+uw%6#}RYT0nNr&0luT{d3}i~AUs6F>Oz
ze71t5AUOK2iq$!+3{`4~Qb+wcZS2@Q<Gk<1iizAFBOLa7RNBFG<QYXwO2P);U{M(V
zSUMh_(0<_9>5qDIzjZ^frLx#TaCQJL*3o?TW!!5^FZyTk4N<B(9CE;nJn-U?pj6Kv
z`sg`MAv-c+>N3KeTfX!37z*2Vz_F{8`{_ql{f)Ol+@#Grq-&G_4%VDsOXV?D*HDE^
zAVWxjBDa$MdVTWkyk%TYu-|1@`RB#MKYtCT$dF<p?jf>dGnyY%)Z<^g#EjVWD_$`T
zPZT!WN0Do}I8xm5e`{|91ppt3%qfW!f1|pJWx)0Y_tHeI$2FQ^NJVg=;`LZ3!$$^I
z{agU_#WXZXv@)Y_KWtlHXRe~a>U=Y$D8u;nGgNPQMcz|LL>VVkNiKRrDOr9kCN`BU
zTaOa|$DIN#Ld-PU#d8S`IH(F3_&M41qFEJ<qLaBxsFq*GWRUzQuGJq$XaEHNB=6of
zqWqq)LiP28`0i(O5%O}W{%pY^88(5bCG&JHF!4Xc;IJl-(i|q%=YzZ;Ps~j<yKGF3
zm8L|2I8~ADH$E1h?3AyXu+zDxbwo%*;E19zrUXq95L5`$`BRxI3~`2(zVDCSwavZ4
z9C>J&u7UxJ4u47=>y%M>@UE#Mdx~8r14?QPqK}OV)5JcGCJA>V2MiuR8c8gx@HpHb
zlOry6p5FV(`~C2dbSo*T5Mr9*!QO}h%v4XvY?BKM3ts8GCVdiUXj=3Yo+m1<-I=1*
zEeo4siTnFy4wQPm@7Qa8Qk4~UGCTpPk8%Eya>@F#b^u3#W8||?P-#)i7lxCmIo>zh
zD-8CX!*Lzpz_lmNv97XnFWk#XO{|10kibPzSv^HbcQF8n{`oPiC6f`3!-NEf>zMWN
zM@E)*_7})&VhpDw-)FHt58+(5^dWjVn4&#0y%7<(HmbK$G(cV^0{cUOEHx&fD*9be
zl;KLA-ud_Y_YBY1*$f`6HBYf;I=O?X=d&0)?ki`uNWwx~DwRIG)Tq6+;O1iImMmXD
zr5-}@Q~4~rnZ&UtmzLcFm>6l`E+omA;&Hzhmu9qpGLLC>1LX5P)7f1A@Rt*-+l*w$
zWp%HBwv4SGWZ7GAITp-5wXenS@b8a%aPWt|5;OAhb5=!Ka#G{Tri1B(;;cVqbcr?S
zcaI%qCLL1$S&Z#zA_G?cf3<R5K}~J_J}6a069nlc^w2@N)DVg^DWNKz&_OX231T4$
zgx*4=6GOlV1Q4W4S5QzoNJm5gsnQknoSSpLx%c%x-21v`&zk?t-m}*FReirPs8YE%
zQvlBYyLfU}L_+-v4fr0he`05)UDmu-a%AI%Uf=#)M#G;IgFY7CMj^R#NDh(@NX7m3
z^h<EZw1sIOhI##KvUzW>C0N8$M2WU2i!cA4#cf4Sf<)z27BGz8Ta1H*sTYk;WQ|4-
z%hTWA%M{D5HP@%7K2M>rV5JyT_-}<v^jExeK_n0ZP!so8HEM*xGn}Nb=7!F2IRXNP
zmn+1GeAsFt+E`+r7G(spf=2R;9HT}=<#`W-gC4NzWnx~on$c}oOfwwcJAYhc7Zjn2
z1x|7=)zIc9=Y|OzBR&qx-GEe9*(Qov9BPM*O;-l*d2V)-;}xywrrCTS3UiQxlQRP!
zhhISuoQl}Zo!a1=4RrSI`L&8_Uz@aEN*hk@w}5EOf|T)rnxrBn_Pg%axx5P!EmCda
zYCag{B?V=6=-0`fq|2wc5!)_rhu94yTM%+pNkrs@Uu^skf?c0>q-5I0qcn2b$40hh
zhT?Su!=Uy?RD;qSmVfxXtMG|&0ajVk-zvKdM<-7Waqdl_Ik7PRGy&l(1`J+V6ASVZ
zj3mqypSQM8p3gHj;B5G=;uI6!hn?W~tgKbP+uA<Qza<ap7#Y`S1iWfRT!F$zqq(G)
z4t_zn`gd9wjl}7(xPe-7+Co|45c?%+u`-@+E6S-%A=7rK(m%k2{!FjNgiNZs)efnC
zu&V&}3`u?y3?P4_mTjecdF;IkKkw|A`eNwe9Keim%|7Mxe0(U#&2iiN1pkcJP%}yQ
z8|AuI<cwXvi_MnaPV3v$f>b%SsoPD@Z!vpRT2PADFM%=cHJYaB53E$r$whkG&g@CE
zf?d`>Xe1|Xw?|U(byNu<70)&CPdcyAHwdPuK3Fv1g6#kaugv7Zs)ZghK!#aL?d4q%
zh9Uv-j)-T>Ww%l)D&*yy7mwoVRyK6|x8(JNR=>?6#P7X&Yg$eVo4~)v8o;T!vj8ir
zJKLc)Wwzga%_s$F>e2pm)8hyAB^e(JUSF*2E85q6w14)e*1O5#jeTkvr%L)l+05FS
ze8~8d6W#i0)Zi4aurV7LcBN>qOF6FQ^K9y?AlrdiAYCDowY7I7XD{k{@=&v%j3@2r
z84(MqgN`12R#WYCv~K@M3dHoNNF&!@CoaKzOpKQe*29BaBTW5>k8ZF8iu&t(*Yu4d
zb^8o=w2{Vi7_UXrHmnPL1nAA(`y*#YA>!I0R#<V}qke_Dfc*=!zmF#6WPz-f#fAS{
z()m~1p(SqrPyI}MRco{rGq%3eAI3y=kGQq68cfmndKyR*+liN9A|_1^mMakJgQIf>
z36eVxbLj9oBxS+7QI^IqOhz+b1N-CFn}%vH8tZ^0*|Ojtf#UJnk6);i0QzZykt;|V
zprJ@s*SYlh@hOd%5`p6EEl7o03S2K4sG3kyb>4i%p95qdNw}<VZ8Et?OUwKE0t=IV
zf3*znqTN3uIGE9O8iJE}AP81{R^2OXG!_LACH9Bk|04*R#sA_P7HH&_*d=0rzBU^)
z%EviyX9uNlVDKD!`XT#dF{Qk_gVFvp(ppAj;6PF==nFcH3O}dF=5g68-=UCGi5)7Y
zC4Dd8BPT#xgM7}|p@~hG&yh!At=HI#Oy?2xcozdIj0@yyEqX^fjgD*T)k2#7;JA|p
z0zql>azSNza9vs0#0#qA-+#EgnErX*iWfbVO)T-vVqH@f38d4&sh4v_zF$sN5mD>8
z*}0Oh$olu5#4VjB2EgJ=V&vm{>|WQU122FZD^1Im%&CKb0g07rjF>TQf}{7D9SjS!
zL>oU!$JYh66e3cU9iHtzAHD%et_l<r_kuePdl>1|xK@U;6e&H+M-f<VAYd@&8wuPE
zc%c1#RUE5^@&9yu{i8qqr{n8?lkw_U?S7{)>#+XSkSmh{3>uy!GAOi6i(~zLgi^9l
zib)$wVI{-ik8bzy1uYDp_m^1iwjs?doFe?(*C8>zGSwAN)<5P<W!X-N0*|wEeaTbW
z!<xg<jo!^d3{^w<B~8(Cc48WDi~VFSr#2K!)K5_>*^W|PikgB4s(#`}ogHQ{EDe@N
zTbo3bN1+-jFnT>nD?Kc=#-bvGw@UzuV+bcGtP|Q8J^IFppU(K3=4!+Dj_*Yfq(@zb
zw7R0LUW71qJp{jzeT>RLaN~baL7XMXPtJt0Q0WQXuPVvKpXN&@aj?Xj&LN(Jitp4)
zAoK$r<sKw;vb`A=vf)X{zolBP?7%iKol?t>m>kan1&2fwx8Q0JR;q!@U#J1gZj2qn
z6)&5jizCKkybn;V+737>cdgZ`>*RY%Ne$pgo1m7H%{j(H$B6+R0FDjs1gs%ALE-8G
zeWT9E7vdNSEL_k?oP3R^)14C$JF7j+12`B^6FuIs=j^(Cac8n{xO@<g>M;IG&=xoL
zi>O`U$teq%zVuleg0T7rHqbvd#}~b-0-HuLgYyR}Y9j;@fXkO^pQ8p9Q0n%_+wqrU
z`A?)JG5&yN!}0iQh;#bcY<Ueas_DzDZt`@~OrP29g_3m(tdEce^}Xu*CmG%_2+Ded
zr;Vo(6%qwm>jRocx*bu1)>E7>#sJ#dn@FdjPM}#m;%|EhBE)HiZo(dJZ+Q1|=ga|e
z6xrdq>xeXMS)GV}6_e(!$UzM?EJwN@Qyv}M_r5Jv!(;FFXNEVjhCl}omkgi`Phx=)
zsq2CK0y1}f-r}J6BBnAvBfN$0H%j+>g^OUT5*eftaCll|Zop#FnXur$6xh{7{MB`G
z_7M5lW6Gj54}22DBGB2f>jPB%MfMVoUpUJWQqfQ-tGd)I8G+%;IJwu-=NEn*SkJAj
zNxIy6L{TF1_YK<*t148UDh7+tczM$7t{_6t%E9#5hasHtVC|WDi!j%|lhOK3A1T<~
zM>60+_j*^ykvyPRcnQ;dO~vv{8AZO|p_nSa5U2bqk6+F;#@fr08*64Kv+=rDC1o-w
zBik<sJyuN)quqjN!RFE2%(~Ijqy49hGhzAPY5*GkEu{|qKf+!uA`jc#2iH>|aJF<;
zL%%A6rvaRn44T7{T@=x6xatN<)kh$(l48ErBAxnKhX@p>*<|#07l^Tqi#M!Ozh$i&
zjw{Z`B5*oh=(@za)35I1nLqrT*1{4U%GrB#7~NS!S3iPcI`&8Sgyuj$bR_O1*fP-~
zlM{xv`hJa*iiF<d>5B}dPa&#lB&2TdgL^Ux!^f{bY?N{D$rbDe1h-x>ak;9+#F{Tu
zGSV(-vw-HvRNwlHiQo&D@}j4SPFxGc)T4{cfqe^Ep_;04tzPG9ib7sz^O|}3aWk&-
z=RPD%b5S4qTdp}|@?vps1lc&B>3K_67&}JMc0D%t?r!uVAE!XdFSVVQc|Atk=1oXN
z&7@T43nKapE2;DC=4Xf7uNijLsuxFvEAmTj)lILsT?sH4=B3waG*bPlDnnC-ua>q+
zd_eBbctJ=3&8mYZK4jX}g%wO7s#{Ie!6hTKa$9WLxB|FHJU+5IDZG4RsqwrZU9wz`
zDcrAb5d$Qtd=;PdgifSZ1*;ZYRt>#Y-2I_uEDn$;LbGE{leOYIKP>zyl(}G|JI5hH
zLML`Er8LW0u6KTQs{8R{>nhaAx!S-eDGWAfZX=gT)M9GY`FUdy<m*sMs;E<9itzW#
zgIssfFve?Jl;1I47%WMr%%i7k_P>|qP-5yN*hG*c-)WZ0E~_d5^sRHXu78TGamMFW
z7t0E)d}q4P-I6#H5>j^a6A5g?0;#!ufA`ql9MOU7zZkn(nvYk${5d#$>`IYb8Gj1{
z00QftUitW8%L_fj8jqta=SKawt0qTSdiDAWjaQC?H|G7i4kd^dYO{-p4AduHt3C*U
zAhHP(Sbw|{h(st=&hvptIR)ys$G!Sj!$6I2sHNRi0{4)gkb$(Ity15c@hEYaL~2l|
z{>^&qkKN-B01aV`za$)=ZzI&D_=WV(Tcak<H&-Bfsh;CQA?<s5JECSrQzA?LS+vAw
zq8P7oDxCdEQc~kBTD}=SrdG@mkI<DDGykDOI{Xr-*pQe}nS**msdk%P0NK#j3Kjj1
zYPhJNtaj(17#IteeExW{avoK>$KNBTJWU2>Hox=xY9M#1AVjjPiW$5L@;I_XO9eCy
zKwVb{abgy+4p2MHFXbKB6&D&jakAd^<EYhQyU76>Lql=Am0U`|k}Te+;Pdx^MI8Ee
z9AH|wy_~y&j5i8*6Hc+sh@|A$vG9oLBfv1$F1qkO0f=@QQPm{aTRXC>s-9G5Fo2WD
z+)e?Wv-kvkKu>vWWqffwtcWItZ@O$Og)Wqw#DE=`N9p^IX~DM}_-jI6+AIN<4KS2*
zNcPoaby1A}#rN#R;u>vMq8Ffkamn&bIsQs#aYp_h#S<K$gWn!VzcQ=0rSWWl!UQYm
z0E2m3jh_dqDfHLcFDK_K&(MIOY+C{u&z1zl;-C}V_P&Vh6{p=Nm0{#+Dc5rzLWhph
zN8Y!MTmpQv(#`!r(mc44NO65RH%;1y(6mi#W0Ef`d}`|eCw!G-*6n_#ED}L}-~YxD
zb8logt0SAI83GeEG+vAd(0uKd3cR)>G2!JO5Dc>%unekxWCHUup3lOMtxi0uzH9_G
zV~`-&joRc{`2Nzk^AkSsmx$Ji>7eL{lr?UPIrEbJV86gxY$>qp3Cz9=fIj`Gd-j3_
zoUc=G#ARVW+~6;49Kb2Ea8Du3qPI!@DjWSX9uppuFcEsJlyffp0oSf5w<VlgF`|t)
zFt-p1I|K~#CE3tiqKrb%?^>P17D8d*vCRdeg-Z3Azlk)a@mq)K6uYe#9ARh%z5w|e
z)<z4nuwSkKsq#oBx!{r*OQ@HB7Pyn)Nk6ovfT2lIA~vacLlfKL5-3Xo@W~ZUDHx<w
zFw+_aff?|2<l$lxol}^@-xl0(8|Qm(i!<H8NYzoGGY?zQt&@)v68B?^J)!tMy5+h)
zKLXn(kca<C>kf=22K=VbjX8fzPd8DWI`?Pmyg9GJ?}k$ro0P@qit=|ACkf#alnGmv
z%?h4grCW?8a!##L42pb=SSf5SjC-o|g7_swGGhDWN4jW7BEtbs)fICdx11^RUsq`u
zR#)Ngd}bR+{8}EZ(~6UGd-V5DJW9PsqV-Pka6q}<`^!J)(wGPIs?=i?Eq4Pco2B(C
z#tz#;<L}5xLtc@3*1J^B^O5>yS9e@lx?UkNL;hvtlF)he93&7|sr2DBD@Qqp!$Mzt
zLoiVFTf*;lQ1^LN(|J<<JCAl7#dpb#*FLcGv!t6L89!BGCa<pI*Ul`_`<9f83kJKc
zf^>bG@zaxt$s2`?cbSy4)M?J{V6%}JO{5uh*&E&Js{(KQ*|Ve}D?dzxQQZ~7Aw6P*
z@xa7@TWgtiY-?{a(uZ7z5?3!QoL(zKT2A@}>YlWXj&}!@)|1il4Z<gOiPq=49irtf
zX>?|lT5)1=g%A8)Mus`=n&d7PRIEo>@_EM4Cl6Yl^T(#9sEFVwF&Wp^Wq15mwDY-j
z-{}-g+56ekKQCY~?3W1oiEAvm<smY=vhgq<hA`sXj@~V*1~{l(Pc(OS{z@A>Ey3O~
z<(P?6R2-_c0-O^~k?Nz<mc~%xYoUy)i%bGK&Dk5;8wPa`Dr2Sx>X;eUM^92MFhET(
z?0k?Pr~J=J`=g<`!sS1~Y%z@(AcK7xQ>@4{w5{2dgcTnj>d<gawv&l5(AG?yX+t~X
z8{IQMx8At*{&Q0uTw@OH?71JrHUj8boi;5uJ7#wJZG~h=GgLUbN`A68)ZeWX4UpEF
z%G+q4R2d555x8Cr`n>2k%A>8;?xq^yI?Fq1$bgn~mIX*L=8<XJw8-tyPV|QE@JY;`
z<3d8#tc1#r(c4#s#2v#+mHl-+6W`^Lx)H5qcDj@Dzq^Q5a5@z-ep|4zVs<Uxb?nZ(
zg+P?x>eS4=6E{@#GV!dDo=6-BK9!X{Y|(2!ldVvjob$y7*!`KUBFt_~9sIZfDGJ_z
zzSmIey*H5CNHU(eXO<FvGGm>Jyd%EtFq>Zi*E$P0_b#FNBsJrcuUk%2h2T$;R!xO;
z!O#g3Jg}tFyx6)nq1OD@VvG&*eL?w6<T&wm19O`Dlt!q*LV{F{VcW^W7k;$}&(8L<
zDhVo8xw?QBykNGu$zx8Ct{*alzr4F!aFx$%oO;#fSM2Vy!1;Vop^5qPFw0Wli@HhI
zqzsxj`Bwz5EPwshjQa)_=MDSLuO3MGR@Y?sD5)vHw8O;6paFMW;<yI%!P>jFt>1$n
z-tM*M?K*276S1y>^tDE2dBgizlYv!1=~z@m%lio1DOYfu)_v>5kq0H-8);n8A>DG$
zRZN=#f|ZkYqpPEOpB&9UJ%X=D{PC^-3X&b)5lk1FtvRR8*4KBMQ~Igq4#WNXo9h-a
z@|3+MuXjpS<@-w^m)B#2`mzK3Cn97kHsPcEqC8W8i7!LC)U+k|+N8kW5dXrtxK_7?
z7|(;Zb5A1XzUIH!<LS(Mf)zY_f0%DSNqq1rPC%W~7C~=uXQHU6qR*uQ+<*QlK6!-O
zL2m+i3%e-x%(=n6^C0N>*G#jez({p`U}e1E_@ocRm-u2v60YI=czd`c@)bUH%xpxH
z0~K8T`ZVhG&C-QFK6L*_DP;+63y0?0&y{u$h)s4sp%uv}>-F#ORULu3(1j2e-B7_h
z7W|_vy9-vgbxfU6S~3j72fP98Ruk>zK+qQl+J3t8$lVcrGIwb+nuQrm)*F!X-!p?T
zNy_~Hw^W;+F8AEW9FEjapq^@<Rt<eObf&J(?Emh_A(u_F+?Yr9m-Wy}DdBWmkq?~Y
zWvp;<Q#1$YzungvqgZbk=Fo3wEh+?Xpxdc%JYz!FrJ*1W{<F+YE6C1pq7{7pW3T)l
xbp0=;{@;N0zcdX^CjLJ)X#a-;dhpjH!%ru#zbM@Y(|XAPH;k-c)liSP{{p%qW$OR{

literal 0
HcmV?d00001

diff --git a/docs_new/logos/nvidia.png b/docs_new/logos/nvidia.png
new file mode 100644
index 0000000000000000000000000000000000000000..22f9e9edd013bd071fd3091bba4221a48ad1ea10
GIT binary patch
literal 15146
zcmch8g;$hcv^Fus&?z9@NOz~SfJh^al+xV;3|-Ql3P{TUQbTu&h;$6y-5uZXyZ2A{
z)?F-Mt#>$wz0bSrdG<b0AKoisW4y*dKtRA&R+7_1KtN1@e@mmI!k;98S=ZseP(P_C
z${{?ze0^&xNk%}RK~R>H(e}<cg!%XzSzX?ojzN>%3pvC%_$D-~2RT_)K8Af#Nv;cG
z=Lil)GDta2|1nw=)BzD?VdYPzTZ=9zjWi(Y{m1^Hl$-{=6unz7jq9J~Y<j~R<0R<z
z^?({<ieT<<(9xo@>k?$9m1Xp|PhL@`$hG8%PbaBGEKmRc{SPKh0xz`RD_YCq0jA&g
zLYwZ*!_ikwq4QS@Ye)Sbw25Cnq&Jzm*Jxus3do>c*X<-w!fI_2GmGW1Hme1EisdP0
zNRx((p3+MY$<v>aq>ltwLyJI#2CmGzPrzPM3sq%yFkG=WQH*$S9T{u>e=XAPgCX*H
zTh8}86f!U=$u#P$>ZYW6R}xvK-nCBMLxW9E<S`0jyC-;PEHo6N1{=ynag>nPgUoCE
z`T8?{!ezg+l5t~&aU3x2m>~Hf`qA>@TOmmnqI|g6GIDs*CONd4??0r2=_sq6nCL9+
zfA?cMK&d6+?w2R;&-e#d(=g==d1DI^K}7igT5mdT`H2j*Csx12<{`7KJ7{p!d)I2k
zYQ?GTSu(rN5afN<w&sUM?~BKZ2BreUMpp^-XDqRUGlJ+c#h4G64TKeR;*jF{nSXR~
zps*TZelVG1m)yt)<N+L!2053UX0vJ_>ZhR0=)oGK2xR(Chow9r8H@?E2WMNSVorEY
zSu}i~H@#21bXeK7E@VCbtdj>xU5U4+7)yTX&GLW^yz!;u##HK?m$5)4Q=yUl-8y}-
z!TO-}K?I74lPc3nz$JDG^^oTK8%u}nFoetEiZQGRzV2O=us{`u|9NMt7}HUrl6yYm
zU|G3ur*!)WOXjz)8`OwYh)>MF_$B1!#y*Ts$n+1u;0x$fXhJ4i`*ss`EZoz&jad9i
z<>+$ushnN;7N`-Z2CXxfee}<SwYQL%EslW}(bi6BG7Rt&5?~aV3Q-TrE2vgm1eLo>
zB%h81S)DH=5sEL1;Xa%nR`q85`~~1e|LvsmUg?InQA=bYB+BL5=9>XOeFxH@$*lOT
z9lhlt8OC|gbsvn&f;E*f0nuK@?(T!B2bE;W8*S{qvac*gde;}eD+(+9w;W}-ekUR;
zP2PT1s+%<c-cHM3-nkJ>Hgv9u5K)e!CoInxTd^b5qrBxb@wXVRP;|a7&3Ya)r%=T8
z?k&oGoaDr9L=vECbc>W>HKI4!_$#xTa-mo5vCN_dASd@@W{lRb9b3~~Rqs~a6HG1#
zGue9&4+tc>WWTS>DbqQ!<VjCEgUZYFci>`H#i6sN=)grhV*-W{M4&BX+paD0L1<^T
z=^nYY>ck5-PxIo-Ngirm*S>E}$pL!i^hIXrQ)TZ>ZYLkB_?vQd^s_NQScj3${hC`<
zJ0eN&caAoUjg_YRPwe2{zVK?+Nh>ea#r6fZ-?C6rbNeEck_IJRa9}>`HRi*V`blG6
z)E3KER2-sBX*{AGCYfT!%UV`k-@K=Bk~Y8Jk6aG|`ZC1v^7WKng%tCDlX42_M1uFa
z`mx@mNwfa`;dNN!@ZAr_2-9o=Shesn$4zw}%WZNJ2T0w4uLjm05|(;7>3(abek^#q
zoib^6`IG(JzfG-Kke^F@cWTP&Sfg6xG(EGy#p_$tSRpSio0L;H9(i{;9;V~+msnGY
z2fEPPP8=)yJF<j@$#lkwfkW5lOEjrPVnMS(+Dgqzv^4ONw*Db3U^M5d-pc-7YHuOQ
zFNSLdv7zaEt35g!lP4O%uLIX|U5Lt`%S5)I+~G-kvF><Kla+?r!1D6beyL&!9Unyo
zke81~K~J!b!RZalvWa&3vz>_8I6wU}19#hZMgha`z>E)nuD-qzYC(?DkhN=8NM-s9
z8u`2EyW)v|JMLC7f81<^6IS{<dKJe~C7`5yCeDZZdBGL84<14sLQEg!|K3U$<T&8U
zz{Mei?;!&^zX8d&GU@z`YUNpVDt#zzHVC|`|3V3oUox>0(Uprk%t`VsZr7<&sWiRr
zaUD(Fux_nG)lc}M3=2QA2$dKauj1LX^rEqH_c_1GZE5W#X31wd+xY3QVw_sx@N8Un
z*lLIZ1Ggrh6u&K0x~`X?KCwzPPNJx99yVmzL7DaA#CQ?q-ikBC{*}{Z%7%^H23yx~
z`5=+@P3oPm2Xx@suVNN()<7b6dp2y`|J>PE0RA-9$XF7?C1mDCU-c0wYUrDUYN_UL
zCpF3@>vwx5GVR3XXR`?Y>XQo<WyR}-E==)=pY({^C9YCQo-@BF2c+;)Cj8^457fJE
z-n*uKD55wdBNTuih7v@1(nn6j=*L6z%8=E}WStb{NAgF0gd^!CnU%5Vocs!hz&axb
zYlDB(Xpfw_uFEL!MNu%Y9L+>lR$0K5#W!rZH1P}C%Qg#F56Ti9%`?(SA{=Cp@I+GU
zRq&=*{L&p;_sCSka#0>t?3upEJRcKAnlwdotIxxq%(O`^jibowmDLWJvB9YgT!$>^
z8ZLf450MVoh~E}=qDXG=&tvm*^%sRFWr*D)eka%PcpeX2eV3wkM$o%-1}fw&xk))h
zeg0;{>(QyLz;P|6Vd0V%w&uF;j6M_p`N!l>ju&Hx&d2fLwI!Hwj$|PAY)QY(n#o$C
zUA>>7b+y)ZK!1>xldP{Ue(WlpcQ#*IyBO>xy^?#tUG?*$eX1xptK~XNitHHH(>Q$)
z5|WhqzNAmB5XC!8;@N8x47g`xx)$&6y&6Tp7z5``X=guciOi3%tEqQQECd;kj4#oA
za--p?0T8^EHMC8;Gg5bVy17I*{a8P7Q!xx(bhoP=DKq#bc`Iouq8fT#F>DEznA1<x
zz1%V~ezGuM7^?faU|he)qAv0-u~7Q@`$NYsloUuD)!gZi1MTKQM|R<K<1_>`1=uvr
zCm;h;n)Ey8uCrXlcmuHma`bJvndk>X_4i|=`uO(@cAm<v+BWs7@|&c=!0nJA4MTr$
zA)vM_CjF4Ns=VdSQsX^J+>h8)BEZvu&hD<B`j`~8han03vd_g}eNO;qcV+3qYX*2+
z>m^W<;?NgeH8Xw4E@W4u-N9|x-C8>U-ZmE4TG+kLVs2Gzd&@DbnqUFOm+ufR0Q@t2
zOxk#F^47ZIglq+*(J(c_mgjueofWsKA;j;;69wxpc4bMIgX>`SpAOMKl!^s15ur&t
zfBf~|ly3`7Zw8WSJV=fjCcW2TeBjgGVDI*4Nv8#0=q@LTze^dT<0UVRYl1q4p{UhU
z&d4p|AYA?|8;d+stE90o{U3?(U#Go)9q<0g@XxI}tb@3ju*8IK|JH`y&Vp}?<9+-^
z{|*K2AI*&Dm{cFD_cY|j)4Tzih{w-w&VY398G@fG{pzoGU2`Ooz9{Eo-MIzVUT(o&
zHcmTrl$Uio^S*3;qWW?cWJ(KDOy%>L<bruL-fdk>Me#JhT~P?b<?9Tjt||weLi^_*
zN=}=vHJk7Y*jB3s2g>_*Bra8w67yG+DjYhnf5ffyDp|0G!@%}3f-y59jJG}sTzD|F
z3Oautp~CfH*Ten6x*{B0&%u!>IRSQp3Ch9ZJwql6x;4`XUdCF5lo9}>p~kU~FR^~y
zqU#?J?LKe$?JW)!Im-bK0~7LHG#_g^m&)O+2qTn;LB(CFuu=Om#Yk-+eI!x*byl(>
z+|eC6K{U*Kfd0vBTfGNH_&;@)8<Y2T+lR1#t4pE-mxHG-z9;q`Tdn>k#w{yR<O_<A
z;}3~pOi;5-D=u+&Siw(V!S(YU)stL=hjO)02ZXNciN8q(#dLv45|VpYta0C<Sr73T
z#<fy^DKywxZpyDG8yO2JwgACr%mqu@>`@sQX27EK@U0D~_nlv8RH8tE_%ZQAA<2}-
zgGj9)I#UWCM_K;*?xVyxx$L|EbDh_Qt(Y~dt0Z1u@3r9Ch!fxF5FuUUBhFV3w!9jK
zPpn`@yx`oaOO7F@(m`yw=6~`}yIqNxHf<+z+8i$xZ1NX1lPR?|;7Z}F_>36;RD&t5
zE9prGeke6~5JGOC`HLRDS(=<GBk37%<IJjzQf%hOJOIl5FFa-7q`8>0a{qIDc~p4A
zB>O}8fCe<>$^L=-lWYB9yf67QH65J>7OgJZgZM;Nbb0Ak;Ke&SE;(2lPc~l><o!N8
zVfwYR>&(&*djg&`euUu${ygf_D~Jp{tu^k~aVmX!^%kmI&{}-LGAM;VT6k9+dCerb
zfd16bK8XI;h9F?AtiE>3bY`lpy<GKiFWu~i@!q=?-o*}*X1v_oilOBr;iHz9B(d|R
z1C0(sW%S*iSXVXN9_h>H_tb?2jfe}1bMGgI=;5(E>P*%won`HsCDJwi?JuUfdU((s
zYQMjTBLt?JHSHkE1upAM!Tb;F(oLhQA&+Talyh<Y$*;XO_KA;+isy+P;?~<$#*W?C
zWY7Oh5GbjJTOyS?y$@8Cqt81=0Nb<_?Uh1i9u{u#WUE7R6lAz)3IY|ww&gZkOi~}*
zS5Mi_`q~}Y6z5q_Y_tRF@)`p^zVTv~0JY;^i_Ok>w|gCQXR=YbQd}GRdR8uX<LMW)
z0aa*t$XwlMGPfq53CHsZd_JOs@Gs-y>3m)k&LHiP4SpigV8C4kCj9D;-I?Xnn+fm;
zOj6VsUv-v)P;=7fkMWnZyU$S)(zju<D^wLfT2o74>u(ER2(ms=!mFYdvABEA7cfQy
zrP=e0o%=k<kO0M{)HYtuYDZjsSI}{8u+HY8aqzy%NzL@&4Q|lUzGwpr7h*n{MC&>T
zzA$^(`Khk~J^%PU(UZ<{(jDOSG39B&{DYJHB?Co1R%y5JL;04^&s|EReppWN{M9~w
zJD!JFe<z+Y&H1Ym#W>Jr-*V(0)avtl`3=IDy|X|MGX56t4)e>OyX110`%N8wnsOb-
zI=IVhdJCs*t`s?SQMmv|`_F2RTS*H{4tiS+%in7_1DU%&HbDyOz24rcdEOQcZnL-w
zP@1ATMmJMhpn{N~X19L|2`wIVzg4kgW0=}ketE7RNk|upUxvE<9=6h8kk&BKRJYl%
zJ#&NPJsrH8IlY%%?JdVQ#vQd&Q;+*a4b6({i&m(vG>ktoTDWw$c3F0~h}=^T5jTo!
z`2AO4hj5THcS@3VQ<Un01WHD!*Nm0%CA&H>U0VXdw{E0KB4B1y2lNeYZB^)>()T2Q
zm(Lkrl+*VUN=IMweVWAu2a4db=kRrK_=KJ}W$Q!x)q~$hHdFG|Vc!9gBxs*sb&n<r
zF^Kat8;-esBy|ZeJVT#+rZ3$G>JvAeEmw#Y%kvCT=f$FRa8!vU{+H?8jAtY|{c_Rt
z0@^++JV|c5_+`En1Boo`jPpNW{a8-`q!r>;Y6Vz0i>bR0z~LpjkEM%g7>xn#KU`RR
zt*6ZIu);wZsG&?c)hTy)38(shbhkOws6W7T9ub_q0jv2&WAjSetF<agg#_gtC|7Z{
zlon=@3<+U;dg4VVW+@J(+*3N0cU$g1qaR+)_Urr(syzE%RVMiC*633By;@O^B<9yh
z8LOfnHpgyrXvsBFiwG`f>+XFH(Pn%4iw(j)GBzS8Dolm*v|&d9kb;JV3cikgaibO;
z9Ae_f&yBxN)n$_~m253-Sa%$7DfBejp4Z+O(cCwjAYr>RG*bnvV(yJ(##nnc<!kLG
zj;?(L-G-6BnfZioLHZVGsk+_Ji03d4&25C6V0;sL5uI04<iyBpd7D%VA$Nw*k!_(8
zoESQHPZ1sL(X!(U_V_qUw8qGQs<Ns!#^_n{pLt17sB|_8bKyMMuX%wCV%w+KuP4Q#
zV}p~8J5dN?x;9O)ZjH1Yp45-lQjJs5q~Jx4n`n{OkVA6AzEQfZ#ws7@`W4g8*lW8p
z%))5gzkjpZ3rbkQJXQjh1nANpw5J>iEOuO!aN<Pr0_QX@Rl`ek2N!`g#g5F)Jyhbf
zBFiW&dnPN-NT6Upz$~jGBU@9y{k>u?iH`pcnXguT?l#jaSyr%-!2Yd}_N`;6TI7ez
zk3xwzHOaz3EG1Uy42sZblEQP6y5hH7x*2c`O84ED3%^R6;6~Bh`0~xnlti0kxgamc
z(7Bw4hW6>#xh&dLX@uqo2Yoqy#Vk}gDHro|9E+ZCvNqe6Wdw-G4*oX_u?pdFwMy?>
z)@7_;`rq3beZsd#U~(;rpTbx?Or>t2LWQ>9zKi&94qjq)U00V;$R4F1&wt_SC{jwf
zf7_3ho@f(_P4IiJv<Ew3N#y8LpYk>3{V-<w#kI~HyGgNtPb!TN;bA5j#CPX{MLq9#
z9mAMb;xQj!xuduiYsj=SbWf*=YXKHxlrkFrJm(8M7R@2PoY+loc)vKNR@pQ}ven2K
z!#R-%XI0d;Q#K&Jad%(;z4p(nuNI{4^VvtRG)Uf_&5bA{l08eW-rnQxCjQv$w4HLH
zY+lGuwfsa>Jt-4`Q1tCdT(M?NaQ^&9Y0Wc=*L$Tk<^VKT*Iv+`X6hU?^AJ~VO1hZQ
zLy}d>EVA86ZIi6FBXa(yX^f|Q+Ovp|OZ5FW_PBt9HLI3bHxe%05`ns6yF1^vpoJ>5
zTa;%m8Z^blbR@d`SXYQ5$gZ5~21oWr!j5$;;nB)ol=hkR2EWQIF=XG6={Gt#qu#=k
zSb~)34qj*E9`;m-bw^OSN#l=mf!78TCDF%^h)N2J*rs7_t^$XN#0T)>>R@>nN9zly
zFEf%A1S5Ep3$i1bM6tH7JjcffOZtYz2qU1Aq^?`G@@BUB9)_c2KUu*QftD{J$+fx9
zkqATJsH-GVPLE+C$pk&=D?>XEE%q*;_M}KpFGN_P!-7yzpsq-EhH0<Nbkmmkwju8h
zJML58dpXX0uYq0(1Wx-rad?Gr$8(}n&|z>bvn@F7xkxI1G8d!FUsv3YDf;>8=kFE1
zB0}`e^4tLs+D5)cpWa0Ur><qxp4G}$CB6)&(cheREs;8i3Je5c`+MmOUavXr(!b{z
zhYPmING4W{F5}6c_{2=mGk;Zvm2(S4-}E=}ZV(Q>0Q${B68YI)a0RYf#1hFr{qzo!
zo79je1p7j|#AzGWU!rA61K`5pztg@t@S?&Ye&hG)Ncs;?)!1>XcXL7kqGITySd_Ao
zVCgZu#I*~G^Cj_dE-R8_A$I$mx3n%~D@FX_1=J3oVFJ+>uIrj{6W7zFCCvO&n}LM#
zCwD(uj;5#j@`DEBTVcJ0RL!_bI5zijzdyiBK*dxdUODYGIrI1_izTeKyWOmhV^S5Y
zGtE1MG||^>nAyxNl3te8zQ*HOrGMh{m$O7ah#yN12PUl$IQ85WzM9^87SdZ-`qnN)
zyjjM3-G%ibP1^TArUnmI9oI>Wd6a|atGt^L9Efrf8^Q^$9pj!hV4V4h=dCLt!CRhm
zb2++epnfYc&Ly%G8n6yeChu`xB?Gdot8RMAjr)&F=+Y4??!9Yx;M<bJ4!Hog9g*Gf
zr<hMV<{_5qgB?He0$%A6?sW&>r2!+FA_bISdOyN?vaC1qW9htQ?Oef;ONjR5@(&!d
zH09fyw&x_vyb;~Srmasa5iMD^^N>pk7_NToXnY1oHQ=AhRp^6Z-j0vlb@;8ftlM2i
zwG4cxW6hjYKM28tA?o-No>z(~iS9@pWU6h)AM(NTG@xhCP*X}Xwc^e3{fUQq>6gk8
zr}}R@1!T9xa5y|6k=R4wXe|&D2`}B*D|hwdc5vx1xLEJL!YP|>)HC@#8B_Q1H)f5Y
zd)E|wud)0S*j(!=e?c-Qj{Hs7Cj=eS|L{01(Vr;K)Z8;Y6J6pM>rd3QN<j?tC!WmU
z)f8#y1bwTl5OM-z!l?c+WCVeY4&%9fYG0Ei9NSe*?S3EKX4P@(_ZNxFwD9H`9V^|c
z28o!a4#M1*_xM<j7DR76aFpG^SRPI}uHRTtcgrW^CO{c8vmXhPt4H+iad`eMZ3(pG
ztE%t=5f^L|jDJY$)H9|4WmgPwC7aFE6Be-lt}806GB6x^GGJlAi7IFLB<r?kuMSl=
z(<C}=lv}ye5j*zno|?!vCKH_Y@KA?1RgP?9&?V!yy;d9<9_@thj_E3JYt^YvhavTr
z;WPy9jiK0eqmglJ6mjZ)--NJ7l|7*5@M?#?D)aK;R~5YtJatSp76lY_@$_wq8#Eb)
zjlmr`C-AWg*9(f}|DrV<`&cC<Aj>f%fq06~J|gYP>d^Awdeau1L>vo><^stX#8MsA
zFI5+b!zdVUM$Zkk&_tv4c1?7wJgXWWwdTskM5HY3WmH@t0y<~gXJcL`yn`3TS#ie-
zb4u%{hw1{;b@{Pp<}2eY3%scV1WMFZS(gvwbgrBlhkuCM=q3cspR3JZO?%YE><u-E
zS-CK)e-s2H4JK9XRXt|!i5jIw`b5B?*RJFlNf>fLWCtS#uJ5o+mj=RSV#R*bHrxp$
z(fz~G*^IJ>KLIV-UfazZ;yc8$JUIJ555Vh>Qc?SF{ml<;1$WAeqYDQcZ@h5-tg?R$
z+kwJOju4W2b*55P_;O5E2Aa#7iKYk^MgzymJJ&)xgShfOO!sMvyb*7!8rJ=y%2)la
zQsn1897$%J)dDm_$9vA7krzocZgAs3m-5VFE9CoU9Ynq3%8>{MEH@o?pOr^FN%<Tb
znaUrK;ctyv9^W5MK>z*2>H}z^iM|2p#}6+WsRbOtU-W=z4e0X1MsLe@ft-Q%K4g)~
z##FVA9ZW<kmU<+Ynupj*rHnkxm^nR3LGt~z-2#N5X-U44s*ot)<8sael4y{6FJYdv
zyDwbM$%>Ngd@ss?BUv;Zp}JzoiOw;~l`CZ++tXo`Ka1YX(<5)=LP`JnLjLenL!g|t
z&FoWR*ARhHVpU#Vph?27NZa&g*Pb6g39sGo%WI0)`OueL@GBxB2Z^yc${G;8<CK^O
zi3{}{uAlN!J^pYsS<Ha?+t0c|<##CP9m;i5+R#Fpv#v0KL)^4qu&2+JBg-H4M*}2A
z0+w||`>>4Cr0>4hG_;X5F%_H?ViGxibiRMTI`A8X@^L%v+0&mn7-IsoI<9JWr<(3U
zU#qZ^1vi5sICQaxcQK9z)``nrpbc|tgba=+Rfm&t!TewC1(!7efy41dGBNiT-uK0x
z+g*vq9sUVeSp$pm6(CDTu3?|6u2Mh`<~_M78bj8n<aVNxg@cVoz57;v+JSKB2wUBY
z8>*R+-%y#FI+)g9;E{?}b`A1TnLx8N&<ezxf*{iuelEmi^H+~|fcs>@EC=-uc@N6K
zqw*usURyhC9JpR@UaHEy-7f`lXI+5bL;NLN9t6sNi?gmU<-fQR9OqiUsyn$>b=!-0
z>0{7EE}QyjIGhMp=1;=oQy)Za-<u;^;niaML`L|QHs|UteabeVSB;~rY705WdZW-_
z5?>^+XI2OWL<euZouD2Sef|wEk>XGMp=gn7(o85K3!61%nQ0q7055ut3olCNn9=|U
z9Bnxj3xJr)rDDG2s(FC<3!ktH6}Jp)wRYobYr@ke?XknTwKN)jp*WR&SmrNW_d`Cc
zDr{6;Xo8&*#!%wk!r=V~;VE-*Ed~`D+LxN^TkTZsEaRSLD`A$cUL6!;_HT~3W|%2B
z){FSuIAs8Eq$^Bj{t3X*io&jVF@w^&Fv9b+>}=z!&T*plcfHWj?tH@>_t^r|?);ft
z{oc_sE18+*%)-oX;5)~EJf2c!nmhgHog@aszq_|flP>^egLNf%OM9NAjpLuP=~T9x
z$8G$mVU}d4%Fbzls)le;FI}+DiW5yBzSXU$!UNRko`TbU$^3v@!G!g##P}lAK@CyA
z$#(nH)ZKWbJO|Eh?iuB`5~QNYoo8D7$L_rywmuatDGI<5o*Sqk<06rZYb`E*{Gx}r
z4d>x1U;m8J7S3l0K=UOSEDi*{V1f#&X#r)z=Iom#xiAfHQPesN=}A8=otY?RHh?a}
z!_jlh8h6Rhq5oQS<7{Jk<73ET=`A4ixN-Rhb;0RY?r+T5+x5lF%-Vr+b?5SF(x2eH
zQSiZ;GQH@G(`VUDS$u)h*qbRufj|C#rVup|`0I-HiJnonWep6KXi^=oJ`oTqmPoM{
z=ph_Om61#C2%Tc&kBoYMQBwt>JG^{4P|S(HsUl|zo`o|X_QYw(@G$+eVz25BeNc+t
zc1<E{r+GVR8s_OiRtNV+X!(bFu7YJ7^n5=FIQa>u2?spMlYNckoaov&PPydcho-i;
zs0N4D|4m2T9|{V3Kv+bE4cnJeG8yp{{|o0PDU9SPpz=Du+g8ReC2T?eJ%oFr(egxL
zkfKq4mHXUgvchCzvD6fsOG*o1NC73m$h^L_0}e4uGVbu^7b9@W|I`&rn*QykXVF?a
zijT@;#EW=I3nUmz{NLycn6BPw?;i6ytb2M<dXL9GfC{O1gynQ3t4?zEZ&Ej2I-fH_
z%pKL%4l`~Acc3*D3WRuXI>ilmL@f{wY%|(3Oj7&`M~@wNSRb)tTBM4_?Qs~8Fpp+E
zsHIjPab_6{m`qN>=H&$_1LNs#!bX!LaBX07ktXlnjG(0WA|Sk;c$o$05g`^u(|n<L
z6QG^;W^%t_KC*IB&x_>`UQU^O&ik_ev9v-_6~KUE7f{|Tg_0^eSk5xQZckbr8ndH&
zT6qzgOfv1x9?v5AizIp0LDf}oFc4Ys)nhltI~r^i=C9X;u#y;F68;O33!KvK_`dyi
zexdar>FY<TWiP>jG;sE01nkqBANxP<63Xj9c$FU9p&I6bFHnPc5e#ouhexhGD;7I+
zKM`H{%WYTJi&bP04*u8m6B`Fb)7Y?3V6Kkkd2JjQ+?js$xb<5<QYs6!+8=TYVxxy8
z%uD@ejtrkL4U+tB2VrssPj5STeC66D_cW>gy6N!>iV$an470Dm*m^!G(OizMUlX5N
z2T$T3h_Kjwy!`oD@dESNB=vbnnyLyT4YZ%yiMDn$!t_6!H`#@Ku>tDAD;;tquZkIv
zH8etg3(iv{oVYi$YDoyog}G=jC%}79o3y)+t)WzJh7&u-<!*j)<W?fJ*}6BHWWxy-
z-Nm*yI-rqIGO7UG)#|4ET9p3;E>eB9C_=kS@?gEviJt9)xVa8(j`O3lFI2x&@q(#|
zLljb@=<X_+&+NaiNZmhVk<8!wUFN>D#D@g}cvp#;R(IP=4{#6AC=l$)hlE!8QRO=n
zg-mVXK+bkC_l-`mfCzRfTM#{58C`v(s3^t+-sX!ci3SdLNC&YnXScO~3MBTW3*w6P
z#o^DZ@>53y6NKsDN<?F_hs>{fTR=1Rf}RsAv`}L(<;aX3vmnjg`dhClotfXhPjgLT
zTIDD|%gWk^iN7fhz|!g99W`afuY|b1dg_6mn{ad74<DH)2gfT+#zvgn{_)st_E8bp
z&dNM|jx5c6dfguwo_RNWQN@eLWj9^m8j^oP!kLvwoltUU^}mg5lMsmS1s<oW*<BZb
zQl{}s=R1Xuk7Kuk_D+Xx;ddDtqd3){Iwq2@CiRabbRz$zDiPbdR2ZsEydR?d;rjQz
zxJ10!;chNnI_gLchfBs9T*i=ZBJ+?cr}q-{-CsKe@$0p}4*$rt(9KJLGuS!sx;LgY
z_xsi|-Q8h1D|FRv;w&-8pR|4?d*FDHaS5gR3MtH(-6ca4ew-Rw#siS8#^OAJO(8}&
zf5HGGNo|1>8Ve^Fmryb;c`b;Yv(}2E&g1lX?kyv-hlnG{(ADX5Gl^3;`z(Lc@nw*K
z$54CX^!ko*NVYUB|J$U(LRL<!sZRD6WfQKaY6z4QSBeGbRIqqJ{NTiybJ-mvJ|OnO
zUglaXKcEla-l$s9{Nc@vwgeU-@#iK45?finw0v$cK2Vyw(Mxk`LiV*@$feH(eyh@)
zW%!BMG#ur=6B+nZUX~@2jhsg~sRbqPrxpe7wxFLar|^E`!K)8V^ep=2CbQkp&Ar<W
z?OFP@SBhmH;!Q7?;jPBCV6Lykb)VRn&Z;q_D7U<pDF&1yIsY3xV2qcwaumwJ(k_EJ
zWkl*Dm2qR*RC({fNmNR0IE5s-vTpU+RoGd#dm>(lza*`!WZ9s8qG}ZR7V9WLA9Im@
zSQBE9ZX(gPe%mCbU5>&r)aIY~OEmSD22(GLnvWm8q4a+g5BS$!_JmN(&ZTrS)@U@w
zghr+$yg$4A`Li-?tx(FLd}uiaE00mMptYzw7fHpk=WaTTxvg3BNHTS=R*cDUIrYQ_
z)3*Uv@ta_y`S3`<D(fPAlmrJjW%p(bHn=6uQC)<fGElX26yJBiJlFmHLii?Hk``Vo
zx<c+2J3d<vbgUg|ladIkQ%R~$j}i%!*S!*J1==(F+OvbP@Ls^K3`BmGuia#j^{*96
zpxow@&U(K@E%bX;20<j(NlrH*QXmSn!_IgLB;n<TJb%!~;7Oh<s$kZD;TA->cBB;s
z?{X5V&>#=~7thb9r(@1U3KsllsqJK_2&NRZz29o*U0^<~YaUWkYapG<n4<asy%kcK
z`;rGh%Ykpj+L+PK605WC)aU$DC(xOC3PoYA+|~cH=ii5h4Zv7BQ?P{`+c*b}UU|}6
zcaqSQ?O({<q&nsJ3NcFc<+{jDz-C0tuWyA4DyGtdX(}ey1=KI(sh`6_wcEJMefvgS
zh$Gn};P`gBQjMyFs^q&XZ?)DU=rzqtxMUhf`<53Ba^qE*(CVFlu=oofNw$U9>Qz;(
zYpjenwP@~g>bIZoB<_b(bCv%RopGV-jtj9F3MM;%AD`I^yM-|ElJ<a0dT1YR4{15z
z7}@KxqNLC_xY9y&$C(_H8#$3NLFOAg2{Wfy@Hz|;YCwsE#g8O28?Dl{rl6c4B5p$W
zSCW31w#b9lo$!GheSp}U6PhE$Y2lLQv@HhA2gI@EwGupzi}LdZ24;Pm5S`({cE4}B
zw~Vz*lSchsOvWXSl_L@OW$)dXg`^A}8KGMvYQ2xjzc-L~CCc2?0@xTW{V-;W0&Acp
z-u@i_btY!oan78z!zBf=2#GN7Q2;h-`ci6+K8v^FcV7l)ik&5w>!A!+xF`UBzdg3x
zPFyfG>b8vcc5t4I&$}WlZ+KSC;1<1Y2P|o@`U0G_zh~D`c3Xyv!UvaBCt@FxJCRaU
zNa}vzo}Q|ArE9IhR!u&>;xGBvUV^qdB7+`*tQEv#3+&18ss`|C1Dg}B(7<Tmk(VS}
zvXWoa9l@yICqTDFmDa~Jv__+Wa~9@OI?MXTqM@505|Kv8qR*n!qEoMSg!!wt9cX4X
znxdk705O8MGv*53oH+g<Z!Tp@c<)_)k93c8yukl)c8;u{XpA7zCq0kwChc0hJ;S(l
zxmv(td6qi=`o8-f<@m7^{LNzha98Yw1pfK6YQ9a{Jn37qwv}lZNZiU)hqA;Q-B!7q
zRPo@a<-0<qtv@cwEZT>hI}Gwnw5mv+XjGIE1<gxS8r5)8c6)W*^Iz~msP1uSiI~sY
z@!tKC9DZ&Qx3U?5uCsG#t${v#D&SMjzRsUELT~I1oxegKbm?x;?c{IOT{SdhtXlG(
z$uW}<8W87RxibeB1e;(jN0X7hSZa8+YKj6HE)3iI-c?@cBf@<g&yS_R1|9W&ZQho0
zzW<{U!2caK6ifUotQVv^c8(j@LT&x|1VhKEABOgSIZ$c+EvZSYI!X&nX<^%fxdUhh
zPqjU@yz)l$E>wLKMk3v0nT7?4M(f1Fz}Lc&=cFfVui6_ZovT9xsSr)L)_kTqd2r_0
zrVY#}yE+!gCE|rfW&?O6Rd2@Po{{rh{-$f9J5!8<o+PlvNi^&yB030KU06?RUfr0k
z2(H)xh8$<m15l;BX18$_{&uuK%xiQkFDQ~LirQKjBrZ?j#LZ)#vmii_o)H3&H-g3p
z7#b4g1E@J2X5R(I_!xb>GU+(Xki6Sm8`o`j>buY#0$QrXmjo%SI4>HF=WJCw<-7gO
z|AdXEE3$Jg)$s4n2UQx7#57_cpk4JuOdkhX0{l=H9{}|xLO<$P%f#*l{JTFTa^|U4
zi_Gt0G5cEh^uQx0b!Qh|C>iL&cBqfEdw<H{{D!-I@EAgbECRNVXt~xijps_=Cy!1h
z@j^q*9O|3_2NE(Q$pLwmkdE9Z^Lv1j2qWgjg~A=4L^h@Km`%|uUMB`-_N}loVViZw
zxEii}x(q4uNqFULp%^UZ%JoBywdD>^4NuiOEhGkT#uW^7CtTuQfy2fYx6)P=K9^hV
z%z3i-mYDKrD<Fe`R!~F`;r(s17<xF+sMLd~gzFy3PdL)Ti~8s*T7c5YFBs=md+#Cl
zN)T^@P-Cn^UpIl$MFYLQ$oXzS>z3LImEz$H!Q+%oAmK*Z!>|U$xg#_3H6kIr%#OV~
zr*bz(tv1YDU}saI*`)3C3gWvz%`&8{m!J~<t5L~O=7zT}5-d-|Bzlh!h{V*l7%v=)
z@}A2ro66boe-|awkiKK_bm~6@j363Ex{}(V1&_v?J1ok6DiA=_M?E0$QXD?O>i!wW
zJ_Su_$gq~K256e}fK4?QhcnV$8nJ|i0h$rnLXp^}sbhZVN6ry_utZ+g>ys&kFN=%c
z_258%Q;4joU~s$nb-!rN4$}wYO8#90DtEjAqGP2MOc$>LRY?R2B8&SFPhJ>-k~rtL
zCLNW=HEaEn=+9~Ay7;O@;T~)&Vv!bi+nQ^SW*q6&UerPn{veJJ_tRj;1aGXuriu=n
z6+}e`1s8G<Lv>q-aMe+r>a#Xt%-WG$ufljoA6xbvNmua-_sAY$C$ns8`zM_~7?}2;
zP+H<-UN_H+t?TZx+7?|m1e#Ntw@DK|0?m8pN$VU)ZMqz8=4fviyxNZ(Bny6w@>D#J
zKp$>qkE#ADJW6|$nYO18{5FlFQ<-)x2hAVF3)6zN+f0=Pf+Uo3(rgaUn-IGYU&4*`
zoQ0@iGb_<8&kvtFqP!CQ%mr-4y=e*2Cx<H0beJc!l{Kt<Sy?485ei;;f2<n3oX#@r
z>>t;aE!~Y}N{o)d_3wO{8)!+_<?Q4ZNovhzM5N`egHebf9Qv(kc?9RK9;#SFI1uPI
zxn%YD%ws=psR$Bn&p)JEErq#~XjSfRUqeM&5N7aljMvYu|CSsR+-1LJ=_9(voEw-n
zsR<{{p(bQna?k)3y3~jyaO|T#y{-Oa>lKW<%sJpdfpwtWdu}v}-*khwFxUqh*za^j
za|w@52m1gRUGqFh_0YrhQM0Q(AdUx@+m`^9lb#HDZUXMl`bcm7yk4;lHvIc@^y5VY
zf+ESOxE^o=z56TL^%foY;nlUW4~uu`%Fzve1dqB0X$3(n!XJ#&J0bLNJ-_?it-M)W
z9Y7cd8nz0}<FA|*)_~lei~*w4oB{Om@^BrsQ&}MpnTku4lU=3vr(F}YV#41)n@*^L
zY4f|*EW0V=A2F96B|Gv@q>j~jA|<`hMujvLa1qnRV!f0S$0~}<E7yS?F+Rl3=^^b{
zRuIBUURMtiy}04gO#;v3fv?%OVknLnWL^X^Yt%&S;FPUw0CNZQgFTtM9<W#5!-S+q
z&kHY>E!+7C!)TjjIx%jo*Fk+{77y~wY-qAldn&xhy~bLxmcy9v&mDQ$naGZ!kJ!gA
zV<iPztr;pt*wr&yEVm%O^G6T(z80&($fjKJP`uli0X91@lgB8`!Wl~#bv3wJTLaK+
z&UWE*ry=qpDbVico+dKePB`q4ujUzPI(@wwp-a=0p)3+fSYq3Kcz0yIFEvt~ylZ~1
z&jT0Pk+EwIgOkY6U}2M6x6BC3R6y&F_!s49vri0V-`s*mPE(Fj0=mi+Mr(EI_5*=F
zc>y+hcn#{q_HocN>ZdC|MVDyUX#n*}LC95HQB%@c_IkmcV}=~5=Yvo69dh-&Ob&>m
zH&Sc1jarIxO@s~ngLYvKS(wu@cJ?C@a9|4P+m}{y;_@rIKI>ok`Fn<k-SB;MroUee
zjZ<`4t+bNiYeyU}I%;K{Ac!i_{zvL>>wb*~<*POV2>wV>_U(=-e{sRFfHvJXVG8LD
zZ^4X={)Uo0MJ+%>?^W+b`XE^LC?>%1SDlKaE6lP6NI)AD$8c?ht958c&#2>#CeuIT
z-+{ART{}(sKZHIIDg~cx$1l&uZ+qg~3u0N2o<t?^e1ga3oXgM-4Dgd0mY3FpgQqDI
zUJ3@+G>e(-pkNF7aC*jPcj3sIkPuRrQbu9P2JQHN&$OM@6*V*&zsRaFW!b@@jjZ8<
z#>RcuHl%i<k1bmBIW0h2tS}z4WqYgYewo-+oKL;9(Yf1(=Oinq35V_JYr|wZ@PN1q
zo$Z`59dYK1|C<jWxZbPyYEL|!2E8$+Sfx>bG6EZ+^_VvwqaYlY9&OmOd=nhtP!dI?
zkBT=|p?K`ulv=&#p(ib4n4O2T78&h~6um547d`J(*0%AEoLjAfHE)01kag`1a=-uH
zWI@Rq$Cz-^?a0EK8dGs=PSp77F(h>P`vcm$z~rp83mbvuG$O4p%}L?-VJD3LoUl&k
zw}Bnh_QKAdgQPWZXp$G>vaQL#&g4VvW~1?@N3YmvK`c_^H}CYjX5>tT57F(~5WFYH
zr9R8MlCsKPBcd1O#L0`KrV$)))`))P2pyamQ^QjLcb80T(^07kvFF)?O(^8cHCs2X
z)$!MCKF{G{0wr%3ldsU54|z(<Xuz*%I2d#N`C$J+QtN#tvB=Zr5Yo+BK4;BAQX$cu
z0ox^zulXC)#Xs(FAhtkY77`*1f`$F2dpI6~ylr11+WvsX6Rmua$~$=m?54E8LO=0R
z1i5Hn-}0l*N>aS6$p#D@gip=5E>%}Rh=eX%RF^%w)EHr(qCIrGDI!1Qbw~9zC;!wt
zg=JN5H>Kw?CV<EKin|Nv{N-0xL!|=F%~zlaJin)*O<JMf#K)wa`K$FJXPjt+Ck9fH
zWH^3SqfT<C!51A12A?sR+kX)`){f>9=wKA|*P+@LiPlmD|9wZ=jd3s20`Q)aSM+tM
zd2J`Pf{x}g+)XgpdCj^SKN~uVYcY2Q1Ft$Sf_8|KrPrvs=c$AREMv8R?YQZ<dIMru
z8M(V~K!irpqo1upW1fC2i6KtR8@`*35JS;7yk7jzZBp|6O|yNfy<oNRVlGQKV?x)e
zZnr7%QNNgap;*vK1m?%xs~rJ^2$5)=gIxPXS$*@72EAqtfO0LIh#{j84ML;r-35_+
z>d{MeVmcv?IDOLIIA5da&O&OH>c1MXl2$D?L?RP6^4D)srFq5_f5jU<S33v5B<d!@
zU(}w>dcVm{BdJg6O?`C4PHzFUJaJBx7UbN9C7OK_#3FBx?k8O#@8ohZ^nLue=ujRr
zOuBN>B&MN2oX^ey#-dUud_p!8?)nEa=uStCX9f4RA+c~4f>nZ;!I&;mxIB1ihWRo#
zR-4E}yPdO5V&f#~jU7w@jYw$Wu^K8jt{BgXADFsRs-jm0S7m`ocTs@k3Jn(*Zp}&7
zr|88RhoE@GWguBe^Y-e)!ONABk)NH!v=)<y3*oWmJ$lEU_|C4>rVu9mW++D&-kRT=
zWzDH&LUw4C<uMvgTni!34gaG$4KP@?T=Jds=f$SHj$;57-K^X#G6lte1JKghC)Vst
zR5dCBMsd+YFxYareY4_v@4HCJ(*2?3es%{L5Ow~YE<cb)CMpf&_tZgtBP5ggq}*8}
z5&w+Hi4cHa>;`|i`G(*T3;PaAZ<_uqwdaSZVJTa0qOK3`NqrWz?RVlCx+taf-QIA5
zDb$=5{wl3729HOYA~;?6gM_(}Z+N>{A^C(kv=cku;)z!wp<K)#j(!xzP+r{9MLdek
z$^>?cS9#pD+5cIMyA%02c;)FlX~HPk{N}JS{Ss%sy_WhA)GgzeNzgu4-B0)s+j9jH
z=)$OLkEItLkq1g5<I?EwwVi6HI144y=(l;PjJbh=i+EY|kzCe4qJda1@h{746@~Xo
z7!%A${N4x(e{PWiZY(7@5C^$OE!*jj#2l<oQyG$X-@hL_UqZSv8zC`Z3&O=`NF4Bc
zIs(2knz9-3Mk{FwXD~rJifO|BD*W`pg)o8bw0`qJXg2*Mj4n`}yY6*FU)<V-j+qN-
zSEvXed425aZr&7KfiVuXT|PmH1fQZW_t59h5<u+Xew1Kvg=mK1$^oF~rxD$PGhudu
zxkB+y9RT0YnaAhoY%%`nj(9cRFo*`|?~G}BPvb;lx_EKYpaZZudx&Zb#98-yYDqB@
zCJ+qF>CuIINX}7VQ}U(}o8lo`3o>wf0YLcATyk>w(uq6gg!$qp^O6o2f)MCVWT!I3
zKes=Yt=l0Pw8FXLx%4Dr&;IfJmD6%&Du1sH+*r=Q%PG*&8E3cFqv=^nr7z9k81OUQ
z%_DlD(KBiZ-a!<wFsxCJwr}&5qKjnz7W0H(K4!;mSeH`p&^aM+OX3y6_Vt~w-dJFT
z$}W5up{rt1&$y#+^yl5nZqE`*-2IQ(il=bN@UDKlMS9)X78<(n)rp~I{L1V&2EDou
zRcpVZ>B5`A1oRl;656wbnvoCmi^coDp{xb)dOzlweKdu_Q|a(Id)L&~Dpu=1-elmr
zrqCLUrXCs7ZeO5RYMhWz?4{mP-bm@(?-@NpdkVk}lhbkyMc+bw-x1djKDaSMID=b)
zUG;(NY+8oJPIh4H75dJ+&dX5h+UTAVbSZ65@~*K$f~ME~S#PU>)r~p;Sr3Hg(Jf5w
zB_(J48eH!96`T8K-}WmEXZ1I^@EcW4sr%9;V&3-{gDErTfR=c2V3nj7gYd|vC#v;Y
zkeT=sA<mk1Hyngih1Zy!tN1=gn4$ZS1T)S<5cJMH%LqEQJ@g0{!Tx!T*Ka|2W=`rI
zItFi&ThIf;uP{g`cO#_4BJro7t6umy<oK6i-PiPS%L{f*&@8{Q!EFb)uLM?Z9}bt<
z`QH6SX?=Inb%JriD>JqJ7v=QbiCuS~=SBUh7s2XrUQ1AmY`-)7vR+K599qq^{njxR
zvp&tK#0tQSIar=BaQ8AN0=V_I9W@7b?j@wPbPa)DbHuoG366>WDmIqSFdO4a-(`~U
z^@omsFm@Q@&vxE^Qa<4y=;c`MCr6r06>w|MN$GDCxP(9UZi_H0G#BXlyG@v418w_1
gzVK<yegCWgI7it?QjdXOenn80e=k?{)-3q{0p_HAE&u=k

literal 0
HcmV?d00001

diff --git a/docs_new/logos/nyu.png b/docs_new/logos/nyu.png
new file mode 100644
index 0000000000000000000000000000000000000000..d93d0fcb3d48a8beef4b6a5c6a12cdfe08e81331
GIT binary patch
literal 30682
zcmd?Rc|6qL`#(G=Dp^V;yONYb5<-?zBq4itWy><gHe}aEo26vQzVG`s_7RmW`xb*C
z`!IF}W6XVCQ=j|3ANS+?&*%H+?|v_gm-9O3x~_AbYk6MR@$QkjGQ%<UV=x$uLFN8k
zEf|bC00yIpJ3<Sdj2FN>z_&v-x7BXLV8x;I`%eyopLs3rYpKCtUKe4ofY&hC9(WWm
z4}&4E!C*LZ7)&}E24i)KtJb^)enI<GRrxNA4E>u_ml+P8(7CCo-J|<Mdx-u5vxk!d
z2?lc?SGjvz=lKYJ%){rj?s)UsN|j`AxA%%rGW{jqGet>RhaP<!r~ln9$XsM8bZDu%
z0%<Z<enpUZ30|zv{30wMDWI(ToC2<HZdcKRXH0iu`)gGqVR5TAL02qX3N7ZkQD<Y_
zDau#6fr8j<{buL!K5b8j8Ze@AHhnJ(d{A&X2Ls=jVN~EV><}15#SFe2QUG86%ZEn-
zz&pIqYyq$%;3G9J`0~Gb_+PI3Ka%~wS@(Y=`~Ns%DB}9Rx2d(@cDNxfZ0yS&XzLV8
zZu<ZE6mzgYY&TSX^an*F3RT<hYm(2!e~FnLI*sBHNW?xm1?k6XNqvypBc)&+6|H8+
zjvE&;>F!ZMbBl01*NMXJZ`ek0rS#vBfAja@dge|)c~u>~>k(|!Z*>Ytl^Ir>GZ#>i
zqboV(e?WR$y4SvX36Q`Gd)_5#LEK+PBl@a}y%-#nUqFi3pvrzDJN;Fl`#}>~?oZyq
zsiTw`rlX7LS`ycrdBRDID`|RM>{^#7v{9q6nW^fy@}WdgC3ZhUv-gLg%{9HYnk8)a
zyKGy|f4QYcHo`zA{v{PyGXN7@ZQkHVGFGyVVU&`H{rfPM>t<QFG~ro2R$kuif-5y3
zjXEGVxX#k)zI@^sD#EbLUSX0#TksywxM*4r+FxdA$MF;sMA_*47geUC_BDJ;Tb5SL
z5KT_8fO@^TgpQ_>3YGS1+#s6a9(a5RhFZ<saa-O}<Qr-3j$1x;lJfeW%Dz3N{qxP=
znk_HO@<lwQEZ#Se6MkYn7L67t)vIQdE)hF?7@8oH!@Jddpg^BbkSP89vHr1R5U=Wk
zwQ9ax9$uS+8+h+$bdEGM5c%OIVcuV~v`Cs5GcTo_J?@bgjg*;LxV(%jATDr3E+-w<
z_w<x0!VRvN%&FbHevv5m?0#leh#)mYTgt?eeNYxqo8qHU<$VM8=0@+2Q-=3cDZ4Uo
z^a*63EqGBcz(}eFHn3Lp(%|I!u0@U0N&sKY_kCbj!LPQPqMhk-O@=Otl4X$`mVz}1
z`(upIEbsaY>{;S{){-hoa({_dpiWQQLvEF^Ol!QuHjm#{l-7I>3H1J#;;AXuMZ!=_
zMM35)9H;FAJ$;K}99Bd2-g|X&y@q()QG<@6*SFHsGv{;mcgFWfYOq_CT=5hewdU4|
zR2-;KoqA0=Fr~rpU-f|)ROpqO2-}uk-t;>v8^|u+*PpIWFHHjx{WM?w)k71htwU^d
zzkZC)nJ3Ud8yFCMU8(xd;FJzpQ%#wZN#rGE#g7c~`Gt&}xHzdDphE?Q$m{cGDfGR8
z=v&@WzPb2n5YcaME~7tG4ulERQbOP?w4FBDnqq20V`2|%7cPg$NJwhXO>J<Dt$(*|
z^PhcaHTYoeE)B456P!`AT)Ai9@m1?e2v)iV#UKyuTt<e0G%J&M*Re<HHy~klUUPWJ
z!{y~8+01R$r?=IgWxz!seOl!RX7gY5KiHiQ_sUh9`x)m1F?TW-9`&>O8sSnikBvK8
zuOJscVNp7equ7K!)L2?56&KD36iDGuf7aKG#TB;nvX7^ldF=HPBMWl!*VuxqcKnAP
zEz75cOK&~<*-8iYo>!D=M#E!v-}-f1Yc&CZ2ioT1Fvv5$EW6hDA}#=0^3Tc_RoYx~
zn0kTKctt!%t~K1_7u_=-B>^*^Rh|RFZDpkHVo31Kqf|ggziW`2WN=`74K6ykE6X0?
zkw@>dT2x?%NnbO?xHKA&q)t%ADQ`$@Av?JEEPZ|=o5!<lT$_1Pu`7MqQCuZmi84NS
zJ%&C$gLK}h_#=CE8>#(W0Rwlu_f2^iBL~I??YLsye}CYU!1i?7At2=x+xZr5cH}th
z*xHpStb1JUT(axl5%Z1_HfgxL68QkWW9w@ob&?+Nds7k?g3ER-<jI-L+)6+DSt-FI
zjFBw?kCW&}==All+zDlqzmH_MTqbl$BDBq9nH-v0kAnB^s=oSJenG{?Oy({_O#cI%
z;U;Id9{A}y3)~sU>7`~g)VF5y`rv8v34!fDPY**xal!2SA8tI{IjTH0i*y?s<C`IP
zPx0kic-YW3xE6Bc^xyDr!EpJw>01q|&%LC)a>C1xq`b9yLiP(%p*<~E5!h`tzxfLT
z43iaUeR#N6#N7oya}EaSYWv5S46C&HM^V@qI<OXJvWZ?ecE42gLA9pBlvuUb-jVVy
zJ6yJVIllk5Gg8j~&xsTGskc3&w6tLK`N66=vijN4d^7tk?tWogJExx!LADc`CbxzP
zF6QGd^NDqcHj6DykK=v|m0)^JyviQ|wXAdk8y9eYlf8S}gG^e`=Zrap=kalfRIh#(
zCV;X2*k7G0HYien&C2L}-`G+j+MC?$%yAZM;SnP*B^*!v<&km?)h*eySD6@Wwz8Os
zI5?n88X`j#sw~Q!%|9fMYv)V%eAOW`gYycAY#eemH%d=CSi?kTzP_eeh`Hfts;te;
zoeET1gWwDf(MiAKG$HeOx0|Y!;Qsh(iQ&T2cOgT+9|BDH$fEzsNX?sSZ>ROKkH2oH
zdG1&afDPX89nC|MULPF%GF+(M34Ef7T><B(G5})j5nE!<3q@)C#+-U>gOTcYHa-()
zj5BbtDq5SarGr~1_lRA!drsmGdNPpxwX_j_3QjMtj)|9qKItdXbSrfwnaiZ^;<CTY
zV$3-eaj!2eu;1k#^bGP_0ntk}CPNu^qbb6y*Maq5e&3|L`MWStOo`pMUQ&rFRIUo8
zUh8D8wdD$&8&3QqIe)qT23okrs`8_V1mgGE(iP>i6y6>@$dOOWzQHI{AImPDJB5=V
zu#K8dA*3%}Cf3+eD^Yb-;)<d}IeFX;0ox9kBDqh5spps!{`4c)%Kylm>Y861*{)a^
zeay1V#qa;X#qX5(Gp#EH=)$3Z3U<%V+rZGvGCHKjl#5j+QQjAI(+LejNqIfwNw-O>
zF38{F80rlUjmrZ9RKZ}^Y7ZNix(60Sd-?owBpnb~!-0i%xeZxdqE_bZj6JRq(P_Nh
zKx=n$vxlp$QQvi~FT^O8@B^zC&uJUhBzaPQc1#f}6;@thoHZ$Lx$LZsHP)Ywadr0}
zD(l<%c;IFeQoU*SY17wKM16ycv321D!D^(o-jD&3W4wGXpZDuJcx-3z08nTBvW=H2
z%61K(Oy-?bw%DzI?t9#DFf?LSmlIru+xM0f#GtY@1fX^N(-hsckxPSDQ+zr7t>v3h
z8?VKPi{SshN%Y%6_FDh%D})6NkD2fLg#Ax)uP}YBz!Wx&ztl{l<sSagqnZVU9i55K
zFT|6s*oyjj&gI!NY0S4ct=z`XTLQJ1AoSaYWh<tv`a}Y;pX2kEE(zwU5${V}{`IK~
zF~4d?*Rc5|8t>h3B3^ub^t<}0aX>y>H77VPJ7&!v7g)-9oi<hKI0<wp%~I*=$xq=C
zpS~*DkUe0(-m1YI29lEt66DTL&fSUM6MxRg484XpIF%S8%kO1hQt@o4K~v;anpN7f
zUK)Yrp^={|Y_Pr*9+=qmz8jPEJvo|vY%gST5m%(yp%=?<D|_fRS(m3)ccp49<KI>Q
z&5!VZ34DZ-GWkXSB`mcBaGe!X{TbSP&1IrQ5dHz#uW}SW=i8omVi58;(0+;kmH1Bd
zRcTL={mVUd%CWAUx7mI2O6?Llj!?#u!DJbEG=@sPXj`_+?6+<epEK|7F+t+2f_htt
z%vr~j;AXx|_@q+ujqlEIhvE7sjTy;QU=>b*$c;%ZcjG^if%ap?p9(5TH^&b~ncC?k
z?^?;jr6mWE9>HI*E486+7HQkVY6}iP_DMiu(w5JcB4=^q=yG<6d@h3f0N4!<9<EfN
zwr=ZkhjL!UtPORYAJ=HgC`WEsSVS0Ls>BhA1Fsn&bE*{Y9|l4qQ4D(RYc%aI>=c1`
zW39f&uD3*(tCnd%mrC~94Dxh`8B33!UH`+r;Ls42`@fllVArXSFvDJ4(9iT^;6iFG
zVf$robfR8#9rQ6HSFkaR<3T1eZCi66%zA@dkMR*#w>eC_l|x35$b)%yKFHPH_mZ09
zOal%NoWgjo8`>W#*1o>>=q@T~lbYj|_TU-ZFS<9BxUM0z=c4<a$nHNSD(&=4K)<X8
z*{t+HZNrrsgf6^N19|buQ|c=mD6HhO$3xNNkVrho{Cmvv1q-u!qEGt8_=B>+!tpa}
z^__KoS%Ja$shA3)wvXLF@(BS*;-7&d+y1aff0D$wc(#vZ103ylqQ-Q3sPA*a%!*rk
zWmK8Hjpuiw^i)pM4O}f%K`xbQBUR_;J4XX{TCzUnuyoF|4q?9zWJcHxbaWSR&4><@
zD`P{P`*T>-S&l!JbtMORZLNu<l%TamWbRu0<XUPoykV~$%0lhLtOR>;lTD%`6nCTL
z<I3S>Nt~+_F?xwu*+;Yk9^P$<=Mu&wJ39)^S-$GgdyfWEpUMjE4nr0guvKA}T7COR
zYv8OOFb!NKvAEi!%Du(@<{N?WZL%9|z8jP8vz8G3o-}TQ;)=#5;nt(nw485@8WcD=
zMRPJi-Vzap9kX$AQ-+DzrQtL@f~WI4<%?3BUOComTEQ`8SN{02@x_xeli7U_U_(Rc
zQhWgcsn@8Fe1B{gV<sT5T9v;SY^Q`wnfTy%(@+Q_nPlPL9JM0S_oXy3_`!~<%GP4c
zf>gp0|IhtJw(-&Ge%vbYM#yI^t@OZ3&r4e`QQ3{ws^1Aswm?~H4tFItTP?Qq)Xn!}
zMTxD2H}E-M1x0e-KR5eY-iPe7ut5GkI(AE&ClQ<F7?=2Bj7Pxa#v_zGl2+gvP~Q}R
zRHYpC>gSIdLsJz|Em42$MsgB_t9BntZ=ruD4VWQ(l{Ln9UpRCt)uIE{?Co*dHd(;@
zC*<nly~H_@x4xG5im%7L(m~zMup8{0uvF5TWK-Y`43RHQ#7X9SU7dRCc3If}y~fb^
z*VJDlzX$|--3Y2NnI|D~-kCQAiS5gn2Q14R__wYX&%{YkKmYYxjx3eQ&~y&inzwE9
zckph(a{fttA^dz|LMeESD$AnkZO7Z02W<%^r`ynf2C}X$UL@U8>;9n~;y&%_(xa7G
z@2UjT)*?wy<{glC=WuN>0dXjdk%WlMEf90OGJItDp*ybU*9lZ{yd^Pmta|28)TW}Q
ziq6ZHH{qeZ#hIKFyVV2PD{&_pn$+jFW1~`Ya*hN*^z1O1FwDfcCRv2o^i^{`fk%P-
zIYL__4BJI2_Q$ODxzCvFyp-54ll|j6VE<T~WrZhysXqVnbhVQb?4jDmV!tZcUklDr
zc1(v(s_HiINMBh*II}=L!s&c<@WD!#@!7$OOf%<8s&%9`rhEJlVP(GU%Wnw-AO=$N
znz$=&rzc>Gi;A55C@<BV*kh(<I7sS}7u%x)YVmdeVXWGuvA7sLwQD|j=d^!~#55iV
zEBVtLW0InOF{GUJt8cRMa((94#@u%En<5fE#psA{vW2f&Wa$bh3Xq5R7;cd%a|Ltx
z7g-z>A+RMz8slrXTQKP$fNaYaXql6A%S!u$7{W!aiP&n`-n&Z_l}iP=gmJ3K+2}7w
zx5iz!HmjihuF|THu$Lg<IS&Q9Jpu&xUu(DIw(5$M@;WQmujvtYeth<nUl?2y)2XPr
zo<aiNXC7xOFb_P0ijk_>%S3+oct(9-;vQNA1C9;EUYu_SbMwQgwT{u5z3GmWFx2Aj
z{I5!JNZ$C%wrl#iNDRN&dDF2})25i1{caUp@0F^|-nHG6GStsOM(9u2yNc0nZDj;A
zt-9gI?$Gf}ee8*Edv8Y)BG+}36^Ow<=eU~k%~Y_y^;+5@-uF(=e0qZ(<l<NeR94pm
z5$)&tjEgU*oFEqr5C=37yS<dd3`G@tJQ95I-lP-P8C%uF$Oq{U8ly8Tcd~s4s+2d<
zR=N=KyPCWD^(k%ZxVPJV_6kO7KIS*tKaz+aR9<f#*SF4hGx2FElHXtPb{C~EdhNKh
zbi1f)72FV|SSWXJuLUCxq7oZp<l|n#%*!r4JKT&&u3+O_BQ4jME_A8@BE#Au@&FaZ
zKTSwLc1+s+&f5m5)^#Nqq+;v$zTk!ohPaobr#%0JFf~@gJ)bOVw&g{P+LMHhD%JZ6
ze8m5(di1N*A#KNtM<0+^V3m9aR7Hd&cj@i^I*s<SSQsS#M^N$$BBGVI7MEktN{Q&w
z8`Jnbu0hRLfk-u+-@fe+>oIA+z&1XgbLN|eX^`9Hikura1H$iGF#dSdpA(*Z1tIU3
z==<#mFF1{Nv_KsAGerKkX!AzDtgGVqPY^(jE(Tph$~z7}$}#%c@pk@9x@fP#(D9yk
ztc=r*J2Y$chi@%sSZsSO60q)3Cn>zWES_T0gUpcg)kJiS&b^JH*R^A-eS0~`=4q2<
z_}rju^}@@!CQ}iEy<Pl*>PE<broh&0%I^=)Ltr2_OjZnSA1_P{b_-Dkfxv}N8TpF~
zm128Re;!l6NJ^+DG=-sZbtM%uP2KPJZa*1belT|ZfIpw)Z2y~hO)m9{ql~Gd^7cf^
z@1W1+n5lK|5jVRu5Tur3<-2ZNyRRho0QPz=FPF*V7e@lmnNxOSZQY8$E=L`#lss-t
z$+@1i-)`2k{)6NEasTEM2ir}Za4Tmvsyu)b0CSlCk)4R<oGOT)abpJJ+RnFMIm7Pb
zPyy^9^AWqBX7co&HISnQh|)zX@^I3CEX%R7)pp$zZ_w=st^$BiO~3rjWbZ$EG79TD
zEYt+${9zTt^`4SkhrnL?k|)tl>pdZ!?Hf%m@8p}+2~FcW5BeVMYJc9~n@IHNVpxPP
zm?hFgk#-+*NmaA5e)ro;&bly*wwU&$TL_y>_7}xe*s@aQs*`ci(LgRQd>pvm$VFNl
zO|4>NACbCDT&dHq@5>WFAD7=}TlgFsI{7A)Iseiqb03KZ3WlpBYEi=|_|rqoAinng
zWFTW=V<Ph%sKYQuwTYc1gMY6W?vhY_&#a)A&)-ibWmWQE)<PUWAHspmLL^7{ukA4a
zDuIwl{V@w-<#PVp=XEw<6Q7Sl7-coXbuQOilOHXuxVp7Kq!{Mri*g*0P(yoVP!D`t
zC6d<_k-r`{nq`4(iLEDg&|4eeJd~_2>+6}pZOronCWxxOY70I9x5+nib~TFV1tN{`
zW$jFA)%91f+S#`HwI_RwvYLKw_WV|*bv^fbvZhP_rzIM4hclC+kbC}3oY<8IDdy4u
za4RnTsw|}mJA%{jYM$Rpb#&#b)%D)1KPAh(c4v!FD*$ErId<dY2i}Kf^#tF3!q>=8
zDXmNiY3i~u_sV#vp)+&Po%F<n>vY`UL8UI%Z%?rW6YtXcO=ZZVl)M%p=UIAb2`@u%
zC|F$PzKl)`ZFgF_oPT7|c#H%1UY2FHbZ)xxchAj07y~ksl?Uzjdpw7DCbKNOi0KF`
z_%lpdmt78om+4}K4(!Sks=Xb5V<1dJQttZ2G?h1NMv1isX*q2Q{o6J^O8RQDw@E>6
z@7oT}QXPWklU{rI0tCDT79p7>wT3*uj8Cn*^L5Ey%ie=GgF{H%qdrVMS+}VPk~~K*
zhdkWP`E(vscFyr<KP?iK#rI9N=O{r+N<~-xA%zQ<2dQbdQ=C6kL(_&xXM|#nPp<ds
zGQ<ut+P^eLmFQK=6UvK}J$ie!-I|&Nay?`Bj1Qf7j$8J(1#Ve$II0WbHq*~?!}2a6
zi+gI8A>Hj1ke}b;{!!&iLQMuytu-S4tm_iZ@#7kovk%f1P%r)c(sy}+$fLtMv7wq#
zrb3ml^Q)a7565h5N`mP?PG-OIt7pQZB*ia|Uml)W3S~58FNSBi08fy{MS}=ABc1+E
z>?27&F}3EnUIE)Z_`OM5S>KhQCkS}}ZRf3UBo8IVXEtZnb42*2XY3-!WK(ZUyqzyJ
z*4Aa~jtr(a%L!IFbkA;hh})>M@6F6>(^2vOY{~~P<L4BtU|X&m0a>bRa(Y~qu8y{%
zuZksPg8GQI)xC1=#Jz;^uf@EA#9YXFtZ?Lm6tsEzxnw@(97mZOC&at^^t>qmC*Gxo
zv>Gm$9sZy=zC0fuZqqShvrKxls0XLF$AI8&vAV`3dYS(rR>ZI{$k9vtN|fWrwgZq8
z<#TXL8Q@D}elA6;4lX54LHw73xW#QRB1mevGE(?4$NI(txU{&|(0J?GRijBs`Gs<u
zxi{tcSn`ILb?q8QwTVpo)~`Aq9G;CnMIvUfLe*kyXZ|b{&}1~5mRQShK~5%-%U?+z
zf?k}Pw|2c%6~||@!arMV7N*obU>aXl`b{-mc7;)waF8#xtGKy<`Qf!R=uu{|n2?1p
z1^z1LIrbTV$3l143Q5lN@nAyOWf0uDbh^*_c$(FtuBIhwsBPHaR}W2_M`bQ2<Zk6Q
z)ofvGoV?@RCKc2Vn6K_Qq$kvo6@Agqe02yi4h}rQHBTOoMj9I{!hoI!T;TGyGgBU?
zH7<7umw5f0h$2tM>wx^Ni!j`K*naEh#s(+^sFT>M?Qm<?<(gX2Jnv@<drFyn(teob
zhv5ABfWoc?H@Wq{tNSyOATIBat_@uDXA_VU@O8<wd}`jl_cqo4vs+P9M$x-B3cOj$
z2v30}BXrcLPaaOP;72Q(N&ZyC{4i0zw>DJot}hnEe<0sd%C#U^JHxd-4I(58=lGdH
zTAAk2vg)8LM-))mkY~g*A;1ctnlisD=S}+RxmR3yArBsUcA(FB<Cb(sTRMmMR8QCD
zu*{Eh8uDfGUX0IjffdQFK6(~uMx2U_E;|Z>Ut!nL1rn#HagAzwxzAc)SjB>DR{VAQ
zbnj|(h2v;CEB79llS8sFmEJY<Pa2s9D`Hy6%5$KzA!p1T9yYzf5zHn3IMew0%_k`A
z(0;>)KbTXcP_^(3ZnU(DFBCq>ntSTVA;l5viJozu;XPs*q0C1A-eXJ%dN;bV)(R(~
ziZ;rO;@-V+5#>eBH(2zl_8K_??ECd(IH~v<9~X%1!j|G{bw)fA^N%mg`TC^-U~k*a
zU+60#T+TQ5W*(eLR@exnsVyqxD?hLW10O<tQb2GfYK?9>yD%^6>vr&(@5`sJk*g9x
z#CaId@LT7F`0l;py!EtVK3De~$p@6gZjrVN%X{UQ_w^rR20=_iU%5q$<t6=TO2k3t
zf0F=kVQmOhPJJm!1w0#XE5pr{Qj5eWOnIvDqF$hc?!CenaM>}l#5$8aEBc`~<@l;0
zCf&L~@$iE<k$tan)8{+dP{RJpe+WR0H<o`|aBihrwg+nANA-RN@f-<IxGv-K+EqZ+
zw8F`xg=vy-tEGmjs9G>7b0wZ*umYg3-2muT$GO(6#wKo#((q)1?@teEV#*6LKOoL#
z)Cajc8FQT_fpWur7jb)=`G9YDE^i+oTaJ;#@i=`@B}l**qD=QCrjHh<Cnn0*IIU}5
zSCPhU;8r9Tvc2W|C0(PM_dS3K6W@FJLV7pl_Jek(*~I9XHyno`IAd^SmC@PAU63<*
z9Jkb)`=(b+OLkcQ^o(xI+PuvZxXgvjAPAsyn<Hdw$!zVv;#o8+Wet)waKGfmWS8kI
z$*BFooq^LN$_ks$0`}l@bRRa-q{>bEzC0En!d=?~G9p6VA^F~S#)W~cwJ{O9klncm
zAtmb;iHpH=v)Lsn(zED7=M8CTqJ|=J-g!8rFu1(81k~jMU~i{oceP14@B4J1c(Sm4
z9ZJ;G;L?`jYnEoZy0kYo-ib7gxNq*#^e3sfTqWq4obHaxtRHL-^-Ug2pAjYdmM;!;
zx3V(zf*4a!XpGBW9XR8AHSS;NA!Qa^42TR#IR1QVz2h9BuQQcum88#VxA@cK974aw
zlp*+LjlWi07d`rMcC(|mr0^AE%l`Yiol<mm0A?_jOqM|T16<;>aVQPu=H8B;5Iuey
zVW{-pu)==6$`53WTR1HGh&_fy<m|*Q|5^onB;9{pR-dQSfODf9PP#0|<@r=3-hKF7
z4)%9^YEE+f<HI2K2wec6R%$4Mgo6VO#;6C+Y<RDT4~3eVr3Bl1yskANDTyOT*+{p%
z4```x*R3txk|j?FJP~<^LIZq6+UG74sN3<1+naO^Ag}C2!T@CTsSNxEH-Z@!dFWt@
z7E4P+3=}5|5OKkfy&qFyWm;e2=z$JcDylX=Le-)s0S=NJF4s0Hc4s|*ja$l97K_}Z
zGM{h&QGZGUP~Y-zskQ@n-(I&;9Xl#(vC7Y|!YRJqtX6IskLxc?PSw~RPwVo2aAk8k
zo@X9Jl^+)H<9osOm*<uaAS9%~`zHc%#=l&87S##_IAIRp9O;9zlKJ%TLq(dUBMFO{
zk_oXqgYLlJce$qe)60=GnZ|cApxCbz+f(986D*Ag2%O%0WEX>VozuP*Sr1M}z$<|F
z{_1Jg2kyBOp4{O%R9S^mf_XTMrD|XWk9pRB$^w^%p;=ddvV%u@LhS}kos7lpUrL42
z?ud+V5%#nf%E(V^o*q?S-u%_P!Q|LdWta#468&zdZ|dwgJCzSW$aHksj(CEK^0rPq
z?bn2?ZzXRQ-3VKU1&><n){uIpk|dxLhMSt0VNSJDRmmRAA8iZbQGneo0SI0$0P;E+
z17s8vvnu@f5NhuG-+sM0hp_ng#kKQA0iuS%u*2t%sw}ZV8E8*xHNql|Cdh3NY!Kw6
zOK**PBo<CLG4aCAe-(%8cJ(Zh9~`q!fBEs@%B_uUQp%Pg%Fd_-gOo1{o*GIXi-N~`
zEFP%1mQ9q-(?y!njNoeDDQEfW^`+G;&-s?UWReHR%W^PJ&9|AyB8#efg>adFK!yqL
z4+5Dy{OG|fqGw;cdJ6Y-q{7ofhk!|e`^LMHPOv-OolC2or5U*GOI!%`*`WgAU+JuH
zbW24_tsrpMgzkAWUgmxQMBu@?GAPlpyXoi;RS@H&&de8Mu)0WHUOVEimWO>=8;rdw
ztPJbZMs>tor*VxDZNqZ_-@G6y`4Ki^{tOXd*}tj|+h5*@7^&yqJW+H`YK%)BHMNp`
zwQ#V>6JhK4?N1iIZ%&e3A|q5Cz5l4ImYL<KBboY;Lc{5enZwk{vO6mT;;lY;9(4VM
zJbcKl;;6-uYWRlZnBT&2A+y?Zc0>xyUP;<X^ZI6yNVTd6;of(9bG5O%eCNL@JmU+9
zI6aUsA8XzaC~nvX3i<jEmr3@yQHa6(DVt}MxHYj0apTVt4f5ydpemOOH}t@6=YSv~
z8JuJ0)`9|h$9D$&Hn4raH80=|&aAt?3hi*wm0|TO*6J&WVJn~O1?>w-Ak8*4peviE
zCW|E2>OMRH@ONJQTP&w>*R3OF#*YL{%H<+gMVF5TkYnl|gL9yyGfNPYON}dK>Sxox
zjw2{H8mQ3%bH)vDY#orLPlby0M@fT(82q=GcsP#iJWo{w+;2e@iMUs~vANUlky?^w
z1S$(|_;|j7xe0gU-~JPsr(?_QhA$oi72!`yT7BA<#zp<MUS|<ok&YtTz}=Z(Bry^z
zgYO%=%>vKPf6g{D^;>e4b;Sm04(o${3UN?(04WqB>z3(jv3Lk(seRs<S#9z9Du4l}
z;S~;-kVvqbr94gSY_3hx6<wo*WH7@<*3#l{ZN~1yRm%KZwSL4!?F2G(mM^4dR84&;
z)GkzN!8nW;X>-*`y27!iAcPDKl8F9%#1&=K;=6^F+=)ap<=Vu}k2=BV-Lni_!>{N)
z<@X=yG-A+T4|Zl=d$hXPZpHS07CmNr*I^@2OP`>&_OhGC+#RbI;<r0lMJM<E!H)fC
zN<qdgx6dO;c;mg=#BqP*xqGnfM&+ZhITck6&HU=*U}{HGbYD4M2q4r0?TY@7(I9*c
zf%oNx`X@(mldk?Dl`W3k?MsyTu*&+f?LDq&C>+#c1L$<Cp4HdVs@c8S+>p=zy)@#t
z>$kQSCV-AiGo^`VgPHleCnQwcIDRKCh!@3Y$oVR@%-ty=uGd+Ol$XZc0S97&tV}za
zqFLHKzPg{B^}U=_;&%Ia7rDI4WjyP5P~mCZTK0-kZ6d%uGhz><l|VLI<)pZ|d}Z_i
zl$;w1b-w6V%{~T5^x57n#8{=OOvSRZ75b`CrSLQyaC_eWeW_gIG4!qq6s>>X*az5c
zK*869sTdo*__XOvyy51|14t)ckXf@Bm8=fa3eYO3>^*rWqf9lFcZ3A*P?wDZVe|+O
zN|5|s9~j5C&*bM!<#pzaCHqfr=t}q06#B46ysLohyr?4^KfM5~;Bm<}BuP`3?NRFp
z{>h<+*$~^N&U|mT#T`(3P`2!fkN3qh0}BshwT%{R&)ouP-6@b5P?PO3872icI|0P^
zrnUpvLN^nChaFp^$07l4E-i6E0Hs!|`*Rv}19;yhN|D-{-2S|wsZ!v(hB7a9cf67c
z9ZDwzEDr!gYDNR}wy`O@_B!>My}G0motAZeXbphsyM13H%aZGgE*2k@Q|cGZJimUh
zQ2{HPS)Tz-u;XiFauUXWIai}S%&#7Vag7K+7hV56fHAdTZrB<Ooe&A|+21=^xAg~g
z0&fA;BS00y>50-jM#Im1h6x}Z`=XL+(WBzuVV6{ZXY0`=6+7xeCp6aUsQmH(7SXFH
zS;y8R=h7+l2uSkFhwH~E(O-ipX*xyoWgg8PrT{a))w<mAm!pO!_fEQZ7!GO7i>?go
zCfn+F=y)3J^c`bhAJm;Q@jt|Bc`DDiS`+KL&G!a$0XCRYPq+~BqI*CtXEn$OKR*FN
zimV-sc!BNIwoo!UVk{3?_nLd7NfMooyn+g>^Bo!&a_cBNab0fK>s=r#&+eCOGhm*I
zJQnp4t{uh!KxlNFqrmOU`E<xGCIyNKX`H8-fV47P@<Zczz3rkGfOcB}crp3(<I4zD
zN!$mIC@qlQ9$7L;l0G}=YV)plr6}YaVq?MRN5PS*!3Q#^H{7sAiL9?-Q)4B`I-z6i
z2iFnhILrB;%0IIlUp*B5e#vt(S>vxn4vyL^6yKGa=w>{J6}&oTnt_lroZP9vj~j67
zJ<2yJRJzlMqtP}UTJBxF=&i1@We>`wf#TmJ{jdP5-&-eHQS^$K`Ru@Oh|?YfRs-2a
z^Wog+gH6d01ZG>PrfH-=owP&OFqBBTnDVZo3sls+IQs)Xh`cwo+ysu8I{C1IM@NLm
zYO%pNL@9vt46EEyQ!(Y<HuI^rKW%h!)s>7xUB`0ZUbuuR04zLvtRL@W?Jrz=Y0t06
z%Xh1A<uZV+4&B*#Rqv8(KX@oItx02EI8lC6Qi=XxVX3Nz`o%Rz@?ghXBBs>8%U>g6
zX5a}NlZTWwab9#bcGI-X-~7HwV1tqag?WDk5L1naH)E1igtl%L!CNrC`#(>@F}0Xt
z{4pDb3HI{dDX))-^}2bUPvkVdMkb^jB?L2i3i2$CDHpQlJUd{g9Q1DAtrs<pJPd3-
zveWP!L7JGY?@=Le?VEf7e*XJ%6>aYNIAC{!>PWz5XM-xcaK)vZONG5+1i$;Bi<Or-
z0U>8R8M6xmxd>nTaf{8xj=4Ge4O$yvf7Z?;Ng^RA|3vQzKKN0K^ph%}j9U@Hh6$n<
zYBSx%u#QZp%y2@M;u@5*w%kL#04+9sCsMrvYfS)bV(PP_tz2<A?u3HH9~PMZRpB0v
zt_~)93@FUmjRrSWwxl=!uN<>A9YxvKPtF+5r_G7OB6+O5a?w@y?fu=xr-#0o&r@BC
zi+a+yHqAvob|vXSFF=q~T8!VA5>S{Xe>OVkkcVoFU)9pJY5>BlRU;MY(H=eY@`Rej
z$twVT3PZhXJ~Dwy+|PPrmUrA916f`wW%tQH?@L0>%o@u?zyxnYdhu2TM#^sA4JJDT
z2s6a?O;7|0u~Y51mSB6F)<T95#4u9S<O0R=J=3XGMJWydb=$-Adr<TtWiCDn?K-fw
z&E2m)wpY$*k3o;`g|xNjaP~zyWhh_9n6UL(?5%F`|HTDF>YdM`{46Gljj%lcl%_D>
zQ@WF@3j)H}+>qTbs&ql+V<3AIpVFMGW&Ug+wh61uEIV)A3_)2N{<|R6rsD}u2YmuF
zFW;Alj2mTWV1+5>?yazWx@2PPT#m&u01v7OOo&rX+w%2;Ah(p_g9VCFS7x5KzM@b|
z>O~0}l*&%yLH3ATe$7`Ios#F~{q?}B9YZ0fc&`bqfRVtv42ULrZcRRc>w`?qWaxBO
z$@w*ewzXx(n|(dRHFI6fGbBvI3tW=5{H`uQ66(zZ*d+^_-kh2dv4%2$RQ(bv;8ojo
z^>lJ4vJu9}U&Fpe+2lzskOXcKVjJrR*wlW$_TYH#aomxHoNMm<nB}-6Xe<Jyu7-Q3
ztPYcR+K_!WK*-wgX58yZ{R33Z;_VH3mryOc82LQ74Qc8Ml*%_^x|3}2VW{Eq)LS{t
z=C34{_%}MM9J2s#RRJl_7CaU=`|DgOJN2{><Pz58sVn+iuAt=&^m@E%WRxY$)fWKt
zjAbM+BOe!7+mV5kN%1;vR*QaU@Av!Ux;xm@K>!w1Jx_Ty6yh?5#9!4H>H~=yRn(Q8
zxuGX@hr;d)gvvZ-XgYf@JgtMn9asO()zab?YJB{Ly{^bKT%zE^vSCr-0scc6N>oW2
zbPHb(T)b$Q7qOhU&u&iwF)(kv;FX-M?_~rN<S#aHE#&RxPfrl!j-V@>jZKLcmUd!a
zv%rRw5Qm&5S+8u7{%oB?cua(K;`&qBSk+E-Jm1vzB^Az4c5aO@Lx;6LXqK<UMKf_n
zrhjX;ep5$M4fV<C!m9jwf9c5>wJA>>Y-dj_S%b>*<Bi$Ruc{uW%=e+*Ls)?}Wa}?<
zVl(UJa>QTq+l~e|XMuX(6S#f;W)uguyE?*X%5Sve<W1^IE5+>kT_;dH<GN8+3c58r
z-<;$J^`ZHC71yT{Xg<j8m_bN-*C;tTH>3&z?JlfD7oo7)hRGOVU#Vrp9Z6dtow@l3
zDW_mq-x(Q@1k^d0e}u}LPM_|!?hRnR`2V<4Ky0Ms*@8gShC4qV=yPp=me^+iV=O|M
zp<Hqu(Mwjkk<EzD^Pp41j{VQ22@sr|PpFh(K7t#&Huyqr{;e?7t4)95m^=Q`5Xc3J
zIz@mHkDHS3OIy0+G2k3p;~Z@WBChRiD5<V<_7khLAat8$>=Zw@xQpIc^cNb6{M;Lc
z;%G1R*VHuCOUH<KOPvelf?8I+%hS_zn6UeifJLQ_8)}t^vg>xQ<)zl<R)hzzQ#%rF
znb#6BiJ-QVg5BL##Rk@@0dRGom;-U3lmkFll*x5PAhkzOGw{tmo0U_G5!YtlsisN9
zv)lTN7lSsza+~@d)lk1PiLkSSN6dGqiYz<ag?1dQfZ#!=l<?b2uhAG0PlFF=Lp4z#
z^a+QqQAvH^HI{NLxUZ+w0CO#n;(LqGpzM4U4Z`y9wKv-rTS8BKg!%0&X||v$IRN-f
zrR+BO^sR!GXL*L{_WLu^=11=Uddw3K>h9dW$wDj83-tpQQo%vDbswQ~02G}*E;w5c
zKG6SAu>}A{fPIXV=AQortGqA`4@?}?KR6|Ky-61sE3Tnm{UGv7I}62fp>6ASYG{`X
zfM(9bSZrtRK+R8fJq5S#5_jiJ$h{a5p|=ag?l|rd*MPwXSdnA0VdS!>rFue`Q@s+P
zub1831tI+M8XiPWM$KJE>^tv44K|YiX2_21^fv|Rj(dmbP)=+z)LA6GW&gU4$vzzb
zyhez+9r&*hu^O}7FqFZmpZVWIv6qf{#wv!i>j6-}8202h2wlRvyatilg233o(8aSs
zLe-OTHhc*|!U>=xTL@;%-Tb(UJzjA;7d^gSVJ&we7W2)QX@NNkw$mD$OxM7tyfyuu
zsR<Zb+dm5$_A|Y~r)YqFETguY^Q(7XvH}pDdx+B&q+ZCaO_pgwkpHj|aRWaw$Gvk2
z*F2PFH5P`@6#;o}+8j*(>M-+qt!r@XkS5A6Uur?IY@&}xeBq-_nP-#wdl21Itvs&t
zDRsQHvRibKMg0qK1OvLeQ0u@6{9L-SeBYEd_QT#tKHYIgp587x>f1uQvzerWU5)YM
zB(-J5(3Lam9Br-3q_6bIMSihHKNZM{I=LU&_Y1!mx2JR0wf6_+iGW*Qx<5Df`d$qg
z59uJOsxZ4+(96D(oB1bv;!(z_UF+rbFoL}0T@Uq5l8%99J|t@Yuj_%-1GkS+Lqv{f
zY2}QHB{EzwzR;#Au-$u4JMbPS_1u}OgF>w6w6XLa1L4Bv+i-A60(5Kn75h^;O^tBv
z+5VUfqtqVD7e=GQedP0u%GFpjybTd`t3p3ORIK1~36ux3^=ngx5_W%zH~Wk(_yU8e
zc_OIcjah-#GFumv?4RJ#2JO051O~ik4G3g&KyF0buiE7yGq104cFtq{c6Wl$lNP20
zOvFzz0%GspyJo(%(@$_)`7Tx9xxE1OAHK(ghR*+BUYMIZClI$?rgzHm?N0O_*O@)|
zMAx`Ubywo<Y-;yh5{H|czWP#KR$kU1pC@8rdC@x8!Gr%W@O%pEdrs=f1+ojRMHert
zzXny~{?|mMH$ID7Xn>5WYIl07CxTW|3{b@m#@+M}U08$auB4Ki-!EGjRv?nB^B0}(
zkYB8@<_~RU?u*T`yoOY&pm;J9ALF*y=LG`mJEpD@Culi`?@7wLzX!zv<+8+e*$4;*
zqJ*%hnkS_rxN-S1Oamx6WvVTbpI3isEcDm%G|;_IT=ICVtUiqeHZ47K`D^aGO!?F6
z?*)N~1Lh6`ta7zm095x`%RphbnFkU2YGl4kBQ6JOt19EMPuJ%m-`Ul#W#5K>YYG7x
zwX*-G&dPrWJLk{2hsV|jorIAw0LormpCV^KW4{HY^8z(%;4#>o5i}8BcLq3Ub535C
z`)0BGa*we`?Q_t32S2%1U%xj*h;hPcKY4?F720}iH*#~_kwyeqy!*tqmWTA7r+V8f
z;N$r+h4JaYk9ZdH<pYXge(9j%50yjoXKxkkChkwnTfj$aLo#H&GoW($A72tMB?UZ0
z4d2~wgmZ9x5bC$^Io-n+?2f5asiLB6JD8Hy$i4Oon74`VHrqU9h&*Ns?^zB&XeBs%
zhxXR@#DSfxTWzrK&7fxh+EqPYv<kjPLKomNoSt!)v;DZP-!eLFH6UnOB+<yMnXiGp
zss%b4w%EJ%>PbG|n&u?;mhjAXf<PZ${+c|W2fe&mIzWj}_KNa&D9Ixi2F}}dY1^lR
z<$B9<O$P)JXQce5*%ev@j|M1E=NS~=L)6_g<M10Pa5ss~(i=P8%#|vt<V%av506Ba
zPxM8cC|oLX`)24%*nPltt&-$d=g^y{D_eK4bW2xvnf!o{-@XEBsBPc3b7>;j2<rf}
z@Eq47n?832fIIkSs&+xdTWxdag{i-Fg8RzVxxZcWts{v7qJA51c1M7T#FxZmS>xTB
zxon7e8Pd<5*?{OP0<^{4D=QS@cpwh)_7Qj6lv()V%9ahl%RzsQcDrJ<k&}n*o07dU
ztqln>dIglR?#3!UcNCK4(`B9JF30jkm`btE`Pbeds7I$7X!xbWW8Ix!fS3p-N}A#*
zGc2}zHSXhAFOiSUz9@vtF@25rqhtgCF68&D1Q5fT0Dv##nc3-?fxY>yI}t0{S_B%*
z67nPt>NUw{oekQ~*3--f=j))0?B|24IagO4Gmx1cBAxA_73U3_M?2!Rvxu%{+TY~S
zgM-0`0LprVx8UXNT%O1;^ZBgkffNya(1Nc9zy_yZ_jN#-4+_1dx5Ua<KSZ3+pwX%|
z&#(Vla|deE22cG7HaLkEoi&tw*%SFlFCuCyw_+i>*ca5jGC(<MA)XEU#?3g0dq@p*
zC~Rz2RRFh8x4kJ@?eT}RL+EiHm(ez8><7%htFRqY_uY0c2H;jQ2o48QOlSsL9=d)n
zMlKH<lNj+?{2P9~o$ya%C&%kOQRCjoWV@mK<dCsg@^bZesPHr;xlQJ0<S5J0me&Sh
z_d@DW<TDUHDoST)&T&E4zP#3J1wo^c#+aBs)7WOl8J}&wb`<3Q)IrOb)`xoIbSVwr
zeEyBWOOLo7h3jp7iA?z1J!RLO78xR4i~{M6pXs{r%Wn+|$VXV8$~#vg9Udj##(y#8
z+ma;L-NWobr^Nlu>YYXqH4XsSz-ysLL(wk*zyD4-2gFKUaf*;mZrRK(OvaA)gP@`p
zc+bmcEl=9;4$~yDeU0xDpqe-joh+yO-dxhPvEcSr>Ggvv*M4BYrYyX;jG!3$m0W-`
zn0)rgqval$ktF+dYMU*t^o@sl8i97n?ou;T{GKaR;PX8g*Ye%dD}IkQ;cL&l6@VQ9
z@q_VIpog-IcB!DEzX(coEbfag8n$n^LGsS?%kWeLiOc<tq{%I--~JiR!ys~{4|#rR
z7In2|(=J)Y^d=@^u2JKPg#EbkmCfCn_Ohj#^4D@uP+|bCN6gJ#dw$=0i~YBHIqtiM
z-#1U5{BEe(lVeCpj{Py(S;!>DBae}DMCJ45Z}`eQD6KVb?gWJ_CFibdI5lSyI$cit
z)fsNcAKcymMN~n)BFM}ccg{)LfzE4bf<v;fY;3M;;jCR>)tYIh$-BV!>w<Nw69Q16
zd<^8jD=Tj`Ws%EB#jS&1Z}#TKTrIqoK$Bba&yO;{Kar}CQ^KIlC;u4wK;;A4G+!5s
z_&j3(5f3ANPWY6=I2M!R+BJ&<?M8;~&$2)-)FoNIy{;1l0H1>5-AEy0*0uVNn#PW{
z>XnEujNFkCt9~P8ewluT^#(u{y4UY5me)EptrWN(hbDazl&jQ%%d#TU+6~tygK#_R
z44xhOF$Y_zcb6|v;@Q?LJr>8~;hWUvGFK?zjs3MTP+%WkxnBSr+6ZX<1YJ~WKw<0w
zJYz^AUCcMFHvl!Oo1#?TZ@@Ah(ePFo5I5fTfwZ}NqBH`{5G0rC#bBRvd&)2en-<A_
zqaYpFxop3wM6ySnl6VBTrN{O)I|m%gUu46!&Yh1TGIfRS4rBSihxD(NKPHi8i)hSu
z6CUX5)0NvRS3U6fVoL~diCDfgRKR65Qj83(`0~XLAe<w$$vQ@o_}<R^hxNyKUzi=R
zQ5|AdBQw5c`)Zp5+C{|e)9rtJK8Y6=^7yf4W0`T8e`Zdz3lz7~55C2Hka$*}S#}3N
zp}Jr6d)8p1uTLnzddh9|-)t|rtO)&Dmqo|eSqqSkHpKdTfx-TFmxS;9Fl>NbPyxN&
zd8z74_s;#X1O?P;Qc2Y1Lm*)Y_?}HcScg)XolA1d;?BeFpLL|7`LAy>gOmyo2!c}>
zwB7$+2O9goyWSi?vR1O5Gin2`huwhsHo<pG%54A7<p1M}|KEpq6_4@5nBkzcNe^th
z>`+0@dpn>yUm0n^$N;j$h0L1HYU{UiBf2{s8Zj>0rEb??LsMt21!Q4rll#N*BbDxS
z@syjLXdBg|fIUFO=PS_N0*FnOGceT$R{3-^6Vedm%?sNv03ELpjv^bTFiTLb2jdTZ
zM}Ce^u+soVwpj+5g~!lqJ58|p;yD<~;gW8W^!;Y&)drs7rDDk!RV2z<#G(jr6A_GT
z{o%?NP)h}+hcNHCavK=XE{Jg7!wR9nu*ke&^-Rj!;|``P`Iogpw;1KOmqc=VmwR0_
z%J?5~^+yJ$y9q+XEq%~Wpc$?8%X~?t8bo<b?p3&Q0?k!}n)pu2i=@ej6AGZD15H9=
z2;gR;8m$Zw4}ek*-fU`K?FK04Z8)G+fHVZ$`gB=qQ{{!;RDk`~Fg;6^S8(jvVup#^
zMapuapr$W~)n33!I2fa}C0aNE7Xh#$tVE5^sW4FR0Y5OGN1>0_eo{%IEb$%Ia)cMU
zSarbwTn1i;Rfa%gj=%tH8-=A7HULdR5M^n{0zjJ`^yv{z(uYK<JP>(rcl-hYHyC@)
zdBcu3b`bx?>KKKtBWep3vQV5kQ|wA)g?6OTJiiQ?DuYOkyhcaawAVK6f$;^mmen3o
z(Lhs7Ixo5g{QLaGWihtl765Fa=d=wcf!HY9%2fdR@IZAYV*~n`HxD-7Xw<-Oj}rSg
zWqYv53#Y&9oiFljchuE~t`q`_!mc`W+hHzxsL`tZ<>m<7eb_YsVhmiw4a1BdvSzzU
z`F!L5rDc-U0<g#^UE8^e>(UGm6X%Zw2(yFT0exMx5N)4a`iuiW5ebsYC8xe4=8&#J
z<4{QL?3Yv?dHEI2T>l=OIsO=qSqf=Qk}QBAU>LQ6PNkKwLc(9=c?N`ii>~EDH)VhL
z9iOvQ0?-U#Cz+)bR@p16d13z%-Q7?QCD>PLT4<@17KSd^Iye|$%DZB)NU%PjjCX-L
z3Atn3{g1&ETuuNmvw(LnEWq;5<{a!O7zH%JO5Km;`B0||Uf7E9uP_7!LSfhU=pgn)
zi;#R^2{7(+n>Af1tmGJFnePGNbJU|{jsjeVpi!Dk0UU80B@YH2Ghb0?9f{jNbFJ$y
zkH>W)^>f|4+5S!QjwO)>szH>DMYu6SBQB`It{ODW^FW`!3&W(H1fzo&EBZt!#FfJ4
z8(Nb*4=DH4pR&Ou|1p<+(^>E=cmaBc8QdCzHrP@^#z{OZ_$mz@L{qUg2&TN<b#Kl-
z2Yjcl%l^|-(V4U=*uUexb|iq~pLXcPZC-S0jlsM9<u4{x2{r`GA7v({07gRwOTI-I
zlm;qrQuF?2L8|U^T`<Ed@TdO-C!6_@whmG{;8f00=(h(wG`Y1`*S^{FA^yI;?|^K8
zo6r8$!J96P#sU2Qwuncd_UJE;2FiO~uofY6=+N`Rz#V%)MMFE)wQs0^e9#DI^KUgX
z*u&K$6NAC&;FM+XQRHrVQkLxoXt#lmMyI<YCKWWc{-Ngmx3rHz?C;hj)j^7aQt>XI
zlw#Yws067H74N~x-HX=%Ux4GVx3I^bD_L(#6pX3=y^+38Gg0t*-;|?lWGNth9`c5~
zM`_HT-uVqoTu#BUZYm1Rh<7pcVRK{(A-7M8Q~y)tFkXNV=>dEDPxYR>(le6$LB7ZQ
zcl$0pIzLNo`S<AEw}d@xa1Hnfu}!T&{e!4Bp}-5U%TP!Hu+J5aA6LWzj3mBXoc_T=
zv4|pnWZ9eXW{@!N7d@A!0c$`9@h)A%P4wY^vJ5jfl3dohW<JIGA0z%9glQx7Rvx2%
zW?;upaX}I|32*<oGe&Fu*LlP-!=^Sl%|Y#LsOTs(=T0i@Y^LN3$wDmK&>RDWoTH2Z
z{J=v%zhyic;9PyU_7kfDB%1~{iDuhFPE*o@Fj|Nc<=-bWXC|V-VFkjCI|+HNsZ~c>
zVAR3F0g5`hv^v#rrhmsnB`=ZYpEW<l4x<V?33Gb?7?^73G4)|yR~U5XKfsmR{G&3Y
z!_e9)tT2?eB_xge1^+SQH``{TDZ-#EKdf~U^KaWB>I|ryXixzZ6OsJ|n6{zLb#O1Y
zB*hRT2^mqK3^&#Dl&C<HEn8IjEW{8jnYQ6<-O*n_&|n#-tPYN+8|UvI{YQ-$eC%xO
z+@^%+;p3E-we!G!9y|MY=;MyVEn)%C@to*~RlbO#2#7b^A>^A;1E|u0cuw-iVTeir
z(Abg;Y=Y*P4JdbmvkLL1eg;oDs}vct;w4`IlLbxX)Z@nz4d{Y~FhS}9q@wWoFl0%g
zxv0%AJf=GfNt;tRM&t#=B(UG_kNusp<3vNSeTDWT3W3$200qi&@6}<`yYiGLK|=gB
zDJ6;j98Tp0nB&_{;{cAK6Mv5(7uB@~T~v-x-z>D0{@X!{n}VhmE=F3oQU7O?RPS0y
zRRV@a%2VT>Qnro5A*4|(fd3zF%O`?QC}adsL9?(97b8)ihIj)1>YobT%ZsRhzjwI)
z4Ez@CNae#cSE_%8!0LQFeb`^-g-g>fl*~*3Mj@$G!_FRv1=P4l+Iw97w>LG^=7&%L
zpzsN;B1zkD#E}YI5`?ti&3VtGuXX?G>l=XUq$}`(J6gbhr~REw;#<JLwi>AHR#@WQ
zfzb#2IhINN0=Afz;tq@jM_F(GTX+z5_U8xQbWA(QUqAn$f>iVhjrrN|+fjK>HkqJ(
zFy@3Qq7MTlc0Wc(nd&r<DY%hE5nBb0s0OUa>#`1T2}tWEWfyW~(@%p6r9P3g0VTvM
za>P%Kl!;?y2~nY&KV3Jh@Z<i=pc@C-^;(L!F9o~rZf)r<#}ffFP6pN1c(*m)Evd?Q
zkc#efp|SN=AXB$WHa%Fh0T&2U9>Iz{Y5-ZIgpL85e$cVSW}&00mkvru*Ku^PbB7vE
z*})p%AYQ<S8i8Q_^*b9F6%6771`!9NoCz2t(*OD)5F-rQP#Cmh;3GKoFh~gh^~3+;
zh-nqR*X#WS2nPJQrwC0*Hvj+l2mD1A@HemWKZFdi=ZcoTcQs31jI*$GKe(&XXY5IJ
zpU@@PRwu1T<oP7)o22nggMYX7H$iU}_cBF=;ClMz#*~ZYc9{l0!}=(X@G!m2X|)4y
z9r@<!8P-<8_6S+{mdR~rN1o2zN;2MCxAz%}8;v=*#Ad%~sXnv-{y5A8t?T=Xe&_7T
zv%`q65>#BYueYn$h^La(T;j@)Rhx7so6|FLx0z^KPs8}LE7lI1vG?H;*xh(WlS-;x
zUZ$0E!%gD@*3{CY;;kMMyx9*~t;_x{JH}Bo#c#1-#k)(LbYUwe&!iPDP}vyUzTw_x
zQ@HIwli$fU496vJmW69+-K+n^?!96x=II)#CY%xUVwxe~^U^&j1MRM4`bpYSdI9v+
zuLj)kgiqpx&>mGkFX=UxYV{RFp75r5-x-N~1G1yn+53C2OKcY0Y-7HLi0TssJ8dU#
zM5&BS_RMlU>Vf~bx9Re9=5m*f&x@exx~LFpiTRp|V$Z3Zd8+`>StS1QHnDeA-4i+g
z;HJs9-gk?60}Q6{#Nrs~ly;Ng`GxPh5;DHEmW=D8Q39#NW0C6nf!v^B4L!&&Bu)~o
zKIzzvs{HdrCDF6YbIZ5B4KqLU{cOB=d2*W3t>0z33mNA_DZ}T!Xb$%YC(BH_dA8B-
zo#GV8?hR+SF`J?}mdSHhqd8H0)=`B)*5iSJ$*t6D2v*tS%rJj)6X*|)E9fKkr8`xB
zy{COfwYo1i@v_9(=a)1*<-DZ*fF!sa(o2EFGp$e0)hja59BQajk*uG%f*z%%;><6}
zna#hxKJRuWgS_v!FxT6@CTe-#^!-QA`=8ey_eo^x4lm9>34m3~j~x2A8)V5~LNDO$
zx*Ku2K8Z`DHJoLHJy&Mu`WYUiYH#5+rSVvXZ!x*nR}*XBK^zeYLCVtS1iaa??32&M
zJ_z_fOB0YCsSUinVC^MU5whr~Eaq9<sfXwgPT1sx*imG7p>6OL<`{-kMVo596X3A<
zmHw-b&FJ#C2HDRzuRH^9t(Vmw$3e38#8<d2ChBn-x4LBN>PY@i9DArl7f;#tIqufC
z!EsBrL2L^;h2`yS5B;nRmYFuMDZq~Fmj<lKx3B9sRU7=M>S})$0;U(gnTgpHo1d9y
z?~@hjXps{Pt=oB$SFbLDTjhla;@h9H2-0&m@?#l0o=MN{T6ICA6KS1SXa6wB7P&Ap
zpM5;TjPEkWq9ccK?}lUGPg+P;|5sga9u8&u#t#o7^;D8XL<>`v82dh!Fd1uhLzcqW
zLN%5wgGY;mFpMqRlbu1fWEo3kA7mM_OSZ9Q-+8a;{l3S0{Qmg;<v4O*?(;gg>pVZ7
z^S)w#ebLdam695G*Yf<G4~8;M6}-g(C?|6;HINyfVQUvTVy-oAVeq3>SI`W1AVhsn
z#sGq)ygM}kYhh8%sq3Yz-*kkl`}24toNUZfmc0d*J?Kzw$KsrQ2&j(F@y6r2tF)&Y
z90Tp77+X)K5>|YiJY<sv^n6q}&%KW^nrWQPVjnt~pJ@$mZ@-SV*V{BJd-5YkYqp3a
zm=i)nYjgKpZt2xMu!LYkiG6J{YX$q<$-x!fClq*&crEo;3b#fjg6YNT_fL_X{G_8G
zsx3tilS>A{pHm8T3yH~!&z?bgy=EIh2Uk02@6?)aKA-P3*w&8iC);)oKS*(YPO$`f
z*r-<;m~R8#S<}1qH2<ngkx-e6Uv%vk?PUFec4dyhKdK&&w_RQBm(-X7`0JOp)T>7R
z>W`@+0=C<mV6xMne(=kPUMuHNt5<ydkON)xTg%Kbe<tvOL3Ym9v`urBwQ1A8Z_*<Q
zEEtX6-}%<VmQyj<>YQ=<L)`2J{qa4f)ima@><Ee8Gu$o%T{bsFQmNka%z+@>)t2OY
zAqSG|j9!}-YW}%c`%_8~v-#Go57FD<C!I}(OZuCP`|-CXoGBKndGyyxESp#DxH)=U
z&AoN$WXOb%ZF*Y1Eiu;>2NXp<T3g*F9<~U$m}*Bq@r#sP%ox(Tt%#snNiCmJwoU$(
zvcB+CprORth|r82UT(~;Sc}inOAiUO*4;VO#T`$VppADA>(+23RA<#L7k0{{_;bFt
z?G)&;bh2|_dptrQzhbs@ch!TPyiezjiv3k$GA4&v`+iG~?FK}yiC=bX-RH*D_Z38^
z88`x6clv|Lw^gmGc<$rEWVb!raEVddPV?B=f-f;t1-ySuSmDIuCn7Z)LT9@=3obBp
z${=ZqD>^lK?xXeK@2!r9-Uj$2Hn+Q|@bkGeck_`7GJ)fy_hnjSm^ORV=g?#Yv9R7v
z*^QWP%xVZYKI;Xp!eh$p>%h5tH>W`>wv1La0~0M3DB`|U>5NC-*)%CZUHIse`4Uji
zUP|sazSLQaO)5PYwp*Qu9?S8HDKQ1EUGP$6BH|$tLi2+%VWh{7ms~a%pt+MI4s(wT
z7v7kzaPasRI#gMqvvA>a-DzkRxa(_nJvfU+tI8M1_K#w*gsWX;UWmaN8GV&UD;78`
zcuNN%ZuPz0@?f%CUsr5}DiOc1j42Dd)A=hOsKG7zmW!2K(E(Q{RWf#sntQ!ykTqJv
zxe_Ip6YLhdsa9im$?+qgEd2TN2A@^{_NUS*?h2!VJOZ}IPoIqLDqWPa$Ouh+l3cQ5
zHM&}y*Vdor#80WzbpBQHzSe8@>!S2?>Z-vK#C;>X!Y}^<jh~Iv_B9Y&_50*B#Odu$
z6hrMlvl}Q+6|3&I80tQB@#^`WQ>6)Z6sFYTVJJ3^3l{9s<&0XrgtX^AT=&F)Q(35X
zB4nWT=Y43_i(Gxb@NRGB4k7@E1_xU?4mQtNy15|vCnbtmb#|?Km${#qyY2<|5yt~%
z$FfuIP;A_>LOnWVSM<yp$>6%Kv6FvqOKu+=!6GTC@d(LD3Hc-w?RXL`la;pvHGNEx
z1LW%*0<b8Z4TBE^viHab<I!$ou#jkHV}IkfS34vpQC_-ucvqoHJGlhGbm1y4eyS-z
zE3b49;k@+UZDR`-g}{~3TIV9oAjo`$(*3}85P7#=2lbHzl6Uc<zTbf>Sh?FaL#<r;
za|wq5Vzc)mv5Mo8IgaiAMF_B_jnOfDb@h&C+p03yRb!BQX=YmOfUwK5>mBsO_^9A=
zn_+4<2b{M})lfU}dWV65`)`3$j6vM!MdQ8Ag}-U!%2OQp3a#`vI0ofZ_7qrzuh=<+
z$4E42U4Mro?xwsRT%DyZNkxbInhq3w3DInDLbK@vs3MAAthmH`jK5u?(2b41??D-I
znh#kJ$<ni-L+Vwl&$ehzHQ0#;fa<D@pDsDOeHRvA-sv(?b6Cn$g|$yj?;eHb{TH%t
zuX>}&s@oV`A{P^mePOqgH<nA6JRR}%45yuT(Oi+#P#u0O;OTN0i>zhNg*dWo>H<rK
zptW^uO+*2>qwe;Hq_at;m}=c(9sWt)vUqKO?dHP{zlo|~!}}gp1<@lH{KBo)In!2E
zkj+{Xl4s+lyC$6hDB6}^utv|Faqj6ATi1q%xAI19mz^g!OsiI|R=-=iE*$^lI+N#j
z-95VBRb3ETo6?gKxTlfHaZ^8yuE8gdDFfT-rkQ?WZZdXX#ZC5SGIyo4q(N%S{tfT#
zDb9c?0q_DO(7K3#VEuJ`P)ughyF6UnixrIEMSl@JpQ7DEF%?Xt-J(zKu{HMtqcuKL
z(6oN<?k(&Qw5^`#3o>ZnM_-;bW|*l_Huzy{{pG^LDE!K9<)8arAAEBOvYo0~#pgIM
zDyi$i4=q6hbW-e^Rk#c4kE|d=5`WU}Yr@fo!D0I>O2+;)8MCl<9(dg?7j9gL)!`=-
z!@0YOMy<wRGHtCs5RaI)o3U2CImwGB<5_qMw~R8vlX7ptyYoU!i+PvLk#M-tT9$l=
zU^Ye)gwSEn1k^S=az%0Zq5w_K3yjaWuB|72r_P<S-ZPf7<u&{)<)@RcpdcnAM3ZZC
zkMtU2I*b2*SC+)uG$<-%_w_f&(&~y9$n3DTk=D)14*yNZ{&0%4Jk=Y#gUp}#FZZaG
z^c9)*loh;fDNMg0|MEugNPJOn<<ImgZLtxeXC5Ug8r8?O@ljoOdo#n{6BnXc+m&fL
z-aw|i2q~!#d=l|Km$d)rEI!Ea^4*j#UFLona~$ZUF<Uw9f)DfG*r}xq-!2;rFXc$e
zD!Vf6I=2AOcgAI=zRrx&^$>kV8TE;&cOmwhc`Zp~g4o#Y!wwT;5$wmzqSU=hf(O0H
z&=@p^3>wZ?`nG~_?7jFAbAz)oR8uH&UW4WJ+UnGxJ+_zOm$o|Yd!!9CYqG$Ny1CBo
zC-#8G_tq1H&SAhw2iwkpi$GHl3+ZlBsabCdw~LA9gX4osJ38Hpe0kZX@0=ovgv{ig
zahd)Dz7w)zh+7)+mA4NvG%_LA8XH7cn%hTS8^SKMt$~l7$2KDIWJV^|=WdNSq~3Ep
zkW8(`6ZAv(4H;w_KXeom{-FFqc^C8E(09VW`Ho$!jQ%28xRQ6XAVk5aQEahK9ar)Y
z3Y@6t5%0!#q~*H&gR?S?q^Y*+9BnQ<@In_O<GJR=cw@3~+7m*pht5e=x`?-|>1iOC
z(U7mPB2kW74dl`%->CTdP_WIV+t*S_&c!+2>;#v*vX??2d2;8Ee9cZgzw#>|^#5;Z
zZ2N0IdhLxF(dvr)4nfKjNNEZKgz|Jz(X`0!Q=avi=OX*>+7P+|pNxMIC^`$i95Xxm
z6;)+twB28pdn!3S-Ldx{v83N6=<}S3l0s-0`f$Cp?OAGGoa9cIKvzl&8DIm<i`^E9
z_Yq!c!x5SdXM&p_3OmK;6p5x0-8E`zb0dF5%*V&)rB`*RW|=3>!cN3ZX~1s$`;C6o
zCiQvxx@!Kb_O~$2{7~C}+m`&F8>0DDZs5}uWOq#efT|~lG_}F_yXMdX58E{b*9gg^
z=S~G6)BN)Fn{_#Qwg$F)hR0wT+opcm3cX4Wptblm#02CzVVdlFUYc#C290tAZOpLE
z(^_5_k@3N|IjzuPEz{dJ$G6%93-+Jr7M2cTX2JDxyaJ~<HX7b8jT{)9;WcmU<^QLU
zL<Gg!y#84B`ACmDpswlugt-4=7SpoOezoI{`xF7(JEoC*E{&u`^>{7*C`U?`lMl}F
zSC$mY<vRN~<=@Ru->e5I6%k~v)Vz?*S;Exloa_jgg}-!yt+lm(gEcRGV}2tVO{n@z
zVpvvR9}s(%0?OBiEZZSjuonfHSGe<<+&Bq!;o@93AYWyzJMM>*QoST!D}Mwf&pHgL
z=eH;yM_l{9G;%M%&zl+w1dqIs1reXFUe3VjyFk3Baq<Z7G%rSb<*!ON2t*STPI!MN
zDd@0I_0R-OyAJ2;uHAi9uW8Xsx!L6gkhVGG8HS||TR9z9ZT=_?&?>o$He1b=pi{&?
z`W*+<w~Cfp99JRaMtNL|r4;3<jfPAOC=d$j+kHPfgqs^@Qz!~E%)QYM%HG{p5^6Kt
z_`Vms-_dk*T;0t~{Ovb8X!=8YqpRgzid4bVcG@myh#M2zj!(5WK^CpzwSQUHkXJ#*
zEFz|SL{(?u^P`<O!R;+oesYpP#%*q=jnZX%d?iMT?kN@`L1C?I{aFc5p~<S%N9}!+
zF^<*%aD*wxkc9%Pb3|JQMec}cVu1^kmU=kT+m4kgX~Ky$V#Ud2SA!PmY?_4W`vg4W
zGApcBc>b4Jo>$+B{w!Ts+tNzfc@X2wY5m9BeRUSJ;t|}q=^YJ=6S*9b9lEZ^kW|JX
zoP_P(w2sQ2Zk!Qi$V$w4YUjXVsasr2nLtYUVCw2GXKhdslH@r%Pslsj2yOxWESna`
zWVBR~G=;S)xe=lO017AFp6{@5vR&j3IsLrf=dot~FtsF<3`HIpxYj#ZkS_YSfLUJ+
z6a^u!YU^1LGW=qQK<_gm>qqlK7^2xrm_qKbCz=vRQHbsR=%Kv~S=VTdM$$V@LJt$S
zvq5Av^y=^9KE=J2*#1_Y>m0=-;r5g^)ISYnG!Mi~C^Kv?T#tHB_GIFHrZR>1xMVz|
z`7pCH%hAy-am;0v?<o=$Yj0QfPPhNWDD>v0{JYR#oLmA?;D%(O{#pM#k`C85Nc65D
z4HT?X$kZIyNC{Kp=ObY|G`szBkz0a&B(A!2ACh2NSo6aEVBFZJxH8tDIl6RS;JKv_
znjik2(G^D!1zk)>n<!zg0$9&vTJbxTd2?i9pd&3jg?o-BYnk=W;%GIJ0PX=wjs1em
z9Pq7d{QLG)-zzyHm+b#n?;3r5Cg2}1@uJO4lS)3P^Ww9z9||~y#>;oND=3s6^t&ZD
zYs3aaV^U)`x$#e9EgXbX&D5?!c<mwn4!^c)umkVo%KN?@nQEzQ^nSAP9+);lJZ9&N
zIx+Csud*;j^-GjBa>ddp_ku6vl28ghiJ4*W@sKg`{CuB*(`Po!dtR;DpTij4w))|)
z{H0|!q5zDk<K%#OmOhTMLE#HCQXG_WU>&ySay<H868!b5V*li-O_?R0Mqkl8PSyk`
z)$vDweUNdpKDCK~SEI3h*1Ar-U%NZLCqBc`8u8pT$vJ3u&pZbbdj0zBx}f0U*<@a)
zu#gXlzhYLKM7grbGv$a||0`i%Y4Qe1#)B?tMExd6&L<rO>e7UDlwT1qOfTX`o>-?M
z6!k4HLXDBZKL&cF2A#Tkyt%YAg~adyeqi!?0!m>|5{0SE;dF`>7m!bzNsGEjuXQI*
z(@O#W_JdGed8`XR0OuzammEz}1u3@fOpFQgRA7fhiBL<hl-N6u+7QFPZP|hQb&_5W
zx@TSjQXhzTyIFbSyOm6OeVL@i$?O|n-wJz6PF=o#%V-a4sxgd!K4w2AC?*)k9-PAi
zq<K`&{j%;)o0pB^(=*anBFy&=$*^z$r|DvNr#NCntBh`;xA~d^MM0-=1M(&pAP0eM
zNfxzZntpcb+f0Ydh<hX}u{BZKEgJmWM!f5C1hvIb0rMS@Z%>t<?YWlvc67K(lm>RS
zr3Ogk$X$B6jpFeO&3x|6Os*FGIYd~eSpb+hd$z(1ecvI)LBo7zMo~&?mqcx6+9>Z|
ze4x5PMG^w~F<UOTfgUqAYitPEng&J^U#(X`I9%Nz%lq`XesCFS_m?jd)`qe(%<h!q
zcm7DNT(!%>_8IbXmvs{LsPbgkY~$Lj&0UDcgwb>y+`>JplKBLN^1-X@2}fPd<cRi<
z?Aytc@2h$PkoSzdbeSEitHh}_v@Say?s=IQ@#2G_#M(c|Nd3@HU{XoWwzI-M7LYe0
z)f2^q=Xbke%KQ5wVtpQ~4O)!!O}wpa&wm$+8!!m{bcprRU!lU+G$ITDCA?eFIBABo
zI*zlPa5u2J2F@)H-)n1|v+K1f&F}2bs#g9dr_RC$3~!^YIvM+Vz+CSnx)gCbJ8Px-
z!#JWAAOl!+v9eq>jlo5HF|6I*m=$OvA*5ldEl|8yi8!W-LFraHIrNUam9D3a*F?+X
zf5M8{?zW|<Tm?$_&08UHbjf8aC5I=yA~JkM6b}BV0?}N;o>;Ms5Gc`;KQgXc`z!4G
zzb7S3-lSKTNxIBL<f6ZmT79iH5VoHRY+l7`LbnBW$TKX14;X+5GJh>a4OkL)U1@oJ
z&&b;xU{tJ>jFnYt5M0dq8nqv=2QH@_1-dAh`OtlZF%1{p>v;GUtuN=ZqhMnR`;tug
zyzeKuvE5ctISRMC7{J4s`T6F98rz(D4$6tA@7qm`$B!c(#^@S%T0mL#L#-{tObi{^
zbkP(AAcNXpkWwXWl^1EDFiJX?<_~*M5h)agzkgw6q38mOw#$3Nn=zHYkkm(w!E|sp
zS#tSsd-Cr!C8HF9z9mDP<fhD&J-~*T3P~(R_f$APvDH-_wx89c0UhZLdi5BqO0n?(
zbQ|BDJAV84O{pn<;=Mpzxp1P3O*pr(nOvfD25_DfM~A+3ew^aE7el%CtVfDm0x%qD
z_fhW!!HpbV;l}Fz=NLK%TB-n@)FbSuu>?2Z_a!l}*&+3M&rA^l2$4mt2TKJsnm?b8
zsPbrLit&1S1*9*zJtp!LKU9icLDit)Tj!JI4byrN07_7veHUWf>tw}wK~ZJa$z^uE
zPkgTfMCJU=ryydlQo6_8vf!m+eoef-oNHn^PkWs+?96X>xm6rb8gD89Ny0Yy^S_K~
zy8G59R!-aLDJ6{~4vPxfPxE3^FGomL7ul@EUh?H)PC+0@qd`&i*h>(esGM4xYS@uB
zQjmBEZTvYY4*J;2^cLpg#de05zoEuHwRso^SliKQ_<KH$oKuYYtTZC`#LpIwIs4Zc
zPR~f{L?pGc`Mb47)m{cp_2837`#t2&i!Y(}136*5-%kD1ePns0?ajcI5LF-HcKksD
z68;ADnx$T>l!b^*UHntc+2;mA80M5k35+Wr{V*8QDac2%)P>3OouG_aAk0(oZ#aaH
zFW-V7gic-2snZ*%U0RMA)4uk<a84+h9Tv0ojIOQR@A`T>@%(C52-o)94^YjnOUn*2
zC<|&q!dIj<Fm>VYi2l_UiFXhl#0oeu;){O?sK*5<PE3;g|6e`tJNOLrT5C)npSeOL
z3?ykp-ccjn_(wRtyyaUb0nE_CF05`sn7DX^{RgZB3nI9ky5OLMn(HaJa7eql692w;
zBerE?<4c<OKxbo3Fw5<BG@G6|Z(HAnFDiqsug_~j-JORUoMPjT!--vVIw(|J(z=6h
ze`0veeaM!ji8<C+bS3bgv1<_TOAuRT+OX%><9n18uu~S6seP`)iNIAKne<HOK#9Ls
zUOigHQcq)wPDns@l|Ow_y1zD=^&}?yb67pFGQ`%dCH0*aZ&^d%L_#6f&LfgC1eR|)
z`PJ8)m?UWK|7c8eXkPK~VF8e;M7Ez*QZz@^E}he98W98)Yi4Sd{{K@3vKig``|Z)`
zIG8f?#hQ(dF9&7AY`l<Wed8O%*VD7CW_K%tY$+8V&rDb(KI3*rGodh&#Pbd0Jw0Kj
zATW*co6pfz%R5c@4TRl=8h%krs^BRuUyh&;SF4V$d~(%=HLv+2qiYsjJH2Rau1R!0
zc@crLf93~yI+wP&EUfn_W&WGYY)*P{R*fE8&A$-#(~+H=rmUNZhEUi4I3SbCbvB_j
zETI(jJI$iXsF6l;P8i9T)+YIXnSp0j$b}xV(jB<&`qYOdDRRZrfyGzM%Ey8@=6@K#
z33g(r(vvx8zvLl61L-ng-Nb((o+tmiz`bebKb02!5I)86DZq`U#bOwzE3M%hsR;?f
z>;?}k^8~G*7OMRB_ata1Ml_ypwF}m9*h8sV^uH3CMW$4KCpZ}bao{tPD>1?qm^zm<
zZ)V@IEgqi3_@#K0%%fT?8qVycqZG2fo23YchPu);$ePsb&lkSweO=-}Tc;@O{*?)*
z22w3xJuuKQZ78Gx707N$aMj)M&wd%|*(J5j7%y!@4<?5I<dx){IRySYo#a0uJ=JhX
zJDbG)zUUHh!eEMIgv1YWgh(7@{*_g(W{(ff#NZ!@#-BT-&yE<=dAZc}nsp=E8<=su
zmH{76G$0;9gKRTy;))aeIDtF{2=DufI@TihmrFlxs;(Q%cbc844tPaopzUA`x*%Hc
zbjlQTMR-WisbeQ5hWZ_4@rA*Wghtm%GtpW0oxmQXwHm}Dib`g7ETDjPL!(OweLJ}J
zhu7^2;bG(9b<ATMOtJk<Go5Dg4nv+F&IDbEnJs3AWjqECuSu)g(jl^2mzPC)=k>Ac
z3q2Y7MJ_*5=eDKHoPO1xNYng4URQkWQ*&9u7P<@^Au32Is#$nbSzOF0+UpYWDqMAV
z#-2UV5dLQdq<$j+9t)KsOdp(qXf#}JO~2yzAD2oZhSE8oXUJn4`>^hC1UB)MF-X_s
z6YtI$)CMoumdpk{+lyPRnjzEF6W~IG34B=wR?Y);tOsd)^!@s}78I)mh1>`T{x`OG
z8gDO$MW2OpR9OdaUSyKU0{>72E>h%jTWPOd4?4`z7)DA$ZYWaI3f^e`^=EdcU6^~(
zt&i7S-PW4-Y!r`lYUa)7d<z}F?ldQa<(`6kCp%a<y;b5v(C|d{c`n$vc)Vg8eysBc
z19CmdT#N!m9k5Xq^;mI7Q;Ps9P{BY(EXQo(q<E!Iwy4U4!ztx>C-bsjX|<aXmm$}V
z?oX^9^lNpNAR<8~Q<j5sCPfeRb;U`i9>rv)=(r4=BthoIsIdhNoHRmy0OtQbKxkqy
zG0V8F(4SdEajL;khK{7E+$mwnl6#-~PowQdvoSNo(vb=+;+AA_!A5kD$A@!bjuTr3
zegxx{dG5Y4ux2SRa7GoN%8A<=tA{(_JJxD`kk13YvFqeqf?WLFuqhuXVks=d&ZsC)
zg>2@>h=Ak>#6@3S%tP#90d&7umN~_MT-84VdCO&um^r=P?jLlr8uV_MHL|BB@X6m+
zK04>hW0BsSztt<mqY6|mg-S27Gd-yy*$xEPF5?up%xl6uW^<SFO#8movmBRZOPyn%
zc!I1T_6rLV_iL=5&p+ScHv=K*uuQ?_QxzZ}G-O))^R6Jqfr{x-CQqcPEGyop&CK(U
z`AD!L@g8I|DzPa>soX@GDqV!K=O0$&z)Kx1(X(B*PaQu2EL(Xsr&#Jv`>XB~`<djg
z$cc*(cy{`WUJ0JgY_NMQm8L4pgaLw;1JNj(QI?$=MZzl7_R&%Bjp1zeOKnz9(j<M=
z#XtzlHk)M4E3_SI2tjUKNh|do`qrt%#362Lx8466SK=PBRb7TQUq~H<%C{G|mLp)t
zJ%K>iql;nBE9tS@myxVM7at3HR|8GP1IFc4#fl?HU1@PSwV7?F03<)~8DUI5jf3}o
z>wG8yx_1}a^bjhKILwZLgJ_ud&3O8i0qO2nn~3QB?q1gS&->Lx>xo6I+{cmfXqv3M
zQ6!b|72|{5@ndnCA3SKG#7#C7oEjh1gbF=gq+z69-e>E`C8@0kdJOe|r2-78;$)du
z8~~eZVS@gRL<d^%#>_pzW-Wn_Kr?X4ar+>loL|=5MXo)G={|byL_DWuS%=u04=|>P
ztz{%EPSljQDgFwD3`|f7Q2OuTUu!KD>0301RY9ydrxoIHK4PIsPk2*YX^*Fp35Ojm
z9^G0P7uDz0<Kr%15Tl+@0n>C&xy$7^<7Z>7qzc(gptA8DNz?#T(-V1c6qK1i?Oajl
z5PocMRofb#ZxN9CA(iRc7<Tw(M(KvNnBs&#QWcj!60Vp+?Jj6-E+ql}OqK@@TBsXx
zRu*30%mwxI&Xh^(6p}mn$@`*r{DwEF_+#>mC|mO}Lt8*0^?|(wvYZ=^ER;atl!oLw
zPI;~#A3i)AUm@)ZEBHcii`b;&<x&P)dahVdQ`qB#d>CAf@Cpu25)qRR3>8gc&ixD{
zp`@y0^iN9wN;4a<dC_8C+F1gKHrvJBH_Rsi+CnQLVUUFH)bqWYlBzUtYAYhM+UOMI
zmUw}Tx0SBfaR@>#**W(^MfK2}v~(n?3l_UsVyCz>G|2{VEm#;C#sy{{9{q3(3TI$1
zQ1ObC)o#x2<5DV<N>Bos`cwn2>4StCf3-@dxm`j^UfT5qdWc$s`*Np>?W{rQ#QgEb
z$<UgLtppIbEOT$bF)P@d5@W(Kgf~HBjq|<i+_Ye6pq?u|8E<1JPEYh<)VTt)ir>Jp
zs*w%*BOC=@h|h)5PA>L-r?P~V?r^ya_g;H+XwgGJe+yh%)nkFc-MVndvis<n6Lo~C
z!FkAA5)~<aPfFdcYcY*_Ys7>67JlS2->6wyAFh&Vq=l~#1hQD;yfn@pm+FjZ7XHaH
z(psmtf2p_^z`Az#nY+@in-4&Gj{@{V<CSsBeM#qwVtR7GA^Wjm;r6?Wte#Nv$;?SB
zPc-UthSMJAW|O+}EcS2uw4Rjsn?BXhUTc}%yhL5-<Yk1{)M%B@%|et8L_3a&&?J3m
zr~StWNs~_li7rvJ6v_+MY@k_NJwmC+IJ~X~b#^F7@iSOtAJq@D0kB;q=^LeaqcI<Q
z_90i-8=!6sm^RQsY+HgOfcfn1@m(i406N)ixf4(l_M$Ww4`Op_(~&xG!(KV?j$eYH
z4(V^z&Ag-$Wz3473B2HKOLf(%Kl>4iE_(4sI>5R7sVJl?wUBs*w(tBR9Y43s3%ON(
z(M(}a6@}peIiU5_%Q`KnYsRIw)`w-0)cFrD-WtGkl?*kp<Y=L<fJgoS#Og~|tCb$~
z;++D40>L|K)hx6&2I8OPno?nj*YoHhuiQZ$ThEzjns@3kVVGh0YV_zn6mk!@x&Lw0
zRV59sPI91!L?dDKL9qU}P?zQJ!C5gGrKM~%7n2MV4l3-9)j(wVQwfrjnSX#4l+V>#
zJxKO*FZV(~Uhz74^<;(+HUTse#904;GxfAuzfgg04+>WT=m*Q{;;BnWbzsC>yZ4B8
z>vSgD%mV+;f3g6$)(fn@o6&^C9mtdtV7Qjo`DILcxv8pii5vPOKs|iaas}qaZ9xfd
zb!C~01+vB|9+rJKPLd5Ie#QyvQD~gwNC<zI$MDq-iuslNb(HCG35NE@aSMQqx{<yB
zy;9z_0_<R?HAOy2`X%!f+dCxTq)j7&mmcK%>af1av)9p7q~nHvv2fa#=3VUlu&qI6
zgI#e$G%V5I^ll(M&Bb~bYB^z@E&fyGuf+gX^ULrLJJsK|+lKQ1^q0i3XsrtQeq8{x
zQf}DoC&)Nw@Lu#49n{js!G)OeIA<GJcDHb<0+hS{E#xisRYM!S$!q}Cjovggv4n+_
zZ{p$%&mXe6*cYW*#>(?s)klHHoa0b|!QqNoATcL`l2p+3hxeDvCL_A~T#@Ev^9K_A
zMK6T_G1f=b!=4~E=iQts<i=wFhZU8nOf-rmsY>F;(SqVn25?AN(*YX`q{Q0FQt?@Y
zXrm#NK9>Qac8yl6FV5qdkdqu)!~CD4vzsu!q)q)>0kXiCT?cUds}@cIR2S_0@v{M-
zg-FEqJ*2%6Zz%i0LzHuSWAT--PXhWOfG*M4zhXR%^bG;(j;(vhrgL1=y}EeEiJAPT
zR!k7HwFHgnenIV?SbS~(u8bE<-&i}hqaF*;LEa`MhTXZ6h0kV+BGVatD?5ul%YAt@
z9uCN%@zZ2560lF;;>mjl^yxt|!?+GDO^p^1D()bMmDLq8+BGXpOvp>pWsdJ`$B%DA
zZq-r)2gBn>2c(`C;7dKsM;5n!Mnr~&?pk)5V;u+#te0oT0d6EHnC$Tch~jFZzmkGO
z8-O}ueQps{Ec9B>TTp55W6HzdeTf(nrVOWa3hZ}--gWLf4~R@0%V^T(yN$vgrhbr>
ziql1Q3IUttYnE6Lc!u@q=k&~+LZv9l8or(JD$GWHif=lIfI}g&J1O+>S+LnqD*h_K
zm*f#>Zu}_>k5w~(8b8Y%eA&F2WEtz{)d)}uT_rQx8@Hs&p7wg<iZX!d<;COl#*H3G
z?DeL1dmPWW{yLjK#khJk&7EozmRZ2x&d0q3dlx{n%_!cbI03IJkpy7H!cbF{Y1XNJ
zo!;NL_B_o|`Ur`G_(|_g!q-siszMt6&6^*y?n=(!IE=|hyq*~xkc$?@cRP+$des5B
zlJYRoxhtMLBj&7b|2V&&1MXh1Or*{fdSO)AY1H~_%l_<-NzKda8}O4MLAwbh{_EKQ
zpz7;8WaTRCNfev4e!QRH_~A}shwG2QQXAxoOl62tD6_H}3Je8m-M62KEJJkbgCoQC
z;$2cf>?R@8L2pTdXnaHR#T&`$pJtxva*u}45oitHa<I74%H<l(MSW>N6z_oo`i<uH
zEs!PFb{5t{U3hikC9vQK96KttC6JwP)elmX&{;YNU#$U>qj}k}1vnq(hFFV%LxpnL
z&zKGrME7*JZ)bt-MW((s(VUMTmSb^DbK4Qo>JZZA43Sl)K~8A+>NR)qgOM}59d2sd
zryZu%`NaSwiMc6D9>x32WcZGX4lvNMTs!t$#rSqW4;#}6l0`?gI7zK}HQr%ZrC2@Q
zqk}pIY0xY7X5T)tbl}>pA#UVFMmk)a*^r+9K&>@CpNX27mQPMLrZZ*P&c)L~Ab^i9
zHUUk3{Z``3h>FvwL@S)@+!VnET4pdHjO#INK2^`iU{EJuCKt`d3TF2qlaZiLvKn!x
zg$FBI!)#qYOD?3RSqo5Z<h@b7fQih!Znv3HfqvaRe2b?hx3;?UPv?w}JVl*4Y5f+*
zQs(Y%l@<G0>~mhUYf*Yq`_Js=gxjV0E`$a14D-@_p^T*ZQ_bN=j2PB^AGs3C`oj|g
z9@nx`PJ1k8#L}De3AaCO^*0^!O{*RaNrK^G4LLCUazw@Zvo3|M1y*7B-ZQRWGy=bY
zdZ?Ru*gf*Fm$Q9t4}L(zL_}qTM5Kg7B#cBw<;11rM5P2pMC3$7)>@&O|HlGXcRMGv
X&;NZv5I$uaEPx=@b=8XQ*#!ST?R0w1

literal 0
HcmV?d00001

diff --git a/docs_new/logos/safran.png b/docs_new/logos/safran.png
new file mode 100644
index 0000000000000000000000000000000000000000..ece23c8e5ec4bff96ac5aa7ac17545896d1c30f3
GIT binary patch
literal 24435
zcmYJa16<v4`#%23_Oe$C%eJ+=YHiuJjgxJ=Wn<Z{wOq@#ZU4`n=lTBruTGshz0R9A
zuKT*~6ZTa>3I&k>5dZ)b=`Z5S002b>{=Ebq7W@;>Gh7(_1Z^rJF9HCyF-R}^FyOz*
z48JJL1Aqq=0DuAk;0gQ^=l}qmnE~KP4*+;m0RY!7vsH;7{0~?ISt)Vg{ohw^M{xoG
z=!8p)i+poiI?Zx*`!+i_xZW@^S^)Voz(DF-&CdnGI~hzetg?9$R)~}NDy}t45s6D>
z%uwwQlw#74AKKKuCQ^O&P^$QV&_PL=mR2`^F9j(VFP$+gZ4!U$nSH>#d*U&^c&_eQ
zDPh)U<WtuIhwRNQ3>POI!u)_v#fz8j87CTPQV-V$#CGdJ3BxgaAumDM@mQX~G=B(W
zWwwU8gq?ec=ppQ`_X~Y<aQjP|hU-EXOa|}&pGPhyIlG{KM)<N5GF?6s+r43t3Is{F
zFq++f>qB>hcQs(rHQ&&#dpYr{Yf$wj2-UrgOi#VHJ6f-?I7*tsyFW=P($5M5BZ3YM
zeqfMcM6;zt8KSLvuwv-;j;5e*YQair(lD2Gp5F(r|M)CE{x)BaAE?;#Bvzdwl<M)g
z*)`(y;on<D(8WRG@M!e3M^zt!WB2IrngQXEL4-&^^dqI8(&_8dz=Q_Xn;Ofm>yrAF
zzKMlUjGW?sCl!JIf*1F=+p0jaP(|*K6*+=m6H?o~@E+rFEW9urDQeK%^dY5lABW}R
zf43Y8nBz|;*LGMg4w9!IPyx99`qv4F3{1@a7(%Ll%9NU{k(;qc57}Ue>tu>5{&!B0
z1^5s9)ES<5oafF2c(UE(&P-=5=?iB1eo7KAf0f|g@9$GI3vEVveoyh8R(E;yVHmS|
z|G!g_A6<=@%PB-&87ByUSg=%sFN24!qQWU#z_Y-my$#1j)hjI7eQiBLw^&7eF~bUV
zWIvfPK|SnHFRJ~&`6<QEUW{JT#9zojmPb1Ma5XxCQQc<3#77O$f7P2X(L0PxUw6t&
zUpa@JaQ^Q+*}uDawZBnP`Jj)|>V5qJ8Bk7ts;Q#~P2qD?2?f|8uy=?T<fIB*1%6DM
z`>!?l!^e6wZzni+$g2oRb?5R1iK7TbNFqiOEH}CF2@~Qkh>vFAE#!$1A}pXB2#oHi
z^cNOM;(tlgvj#wOW@u&s!nMfZ`gUAqt5H8aFzsDoTm9^5x|QS^eJpT!Jk_peE=}6P
zwkZDJO&|zsYz{7zm`%Luy<lyW6yf4M<6ioTUL(riu;B^O6}=}}o~O@;5H*dJH5!}z
zdTKaN@59?;A#g%wCxzNjgH?c;>^!WX<9OQqVc@;|a{S!IprZ93QxrjmP##_g@<oQd
zqy(W^Li%lXg)&hMDx=Vo_s$FU-QAY|AjRMm#7s-8@Zr5`>s<P3ihhY-aVQ780wu9c
zv5y@S8=@%7r?%^ETehsT|GmmUPCDUfq<Lbw;C&e22P`df8DB0NNENvqPWk+lmPwb$
zeE2z&srtZT$tuPF5z#~oHrEp#a`gR`)!@xluGdzvO+`gM?drd0Et-KzW{5bC3T2NI
zz{?@32Yuyx0r?Os(!y!1`Cu>|S2^^Uo`$Kb;mk@a@P!Qp&Q9z18@Zfr{FX005!=R(
zw7dTeQ;th!PWrM0QMfllkT@sor{oNwy$Py<@ESmFnm_ZlbWvUoePE~AU-qHOP(YXT
zw|9bnBlm>hzYrNTPXIgb>;LxlomSX}@nKy;qkJB?DO>Gny@UUuX9a!XUuwnF$#ZjX
zYwc2>diV=pb^<B^DUuu@>ft`H{7TM07!`3wQlqK>@n3iRsfrMLjPsHyIjREDI8U%y
zCc`TD@XSL`9W`9rb~iqb&(=-X=M0L-(Vfn+b`pR%)FUDZd>{4rNOE;}x=)|9_P=oO
zQ3u5BI6ojx_=ySqx`Y>6GMD;B5U7*Z@69NE^+_ZztIlzZ8YA>XAuoZ$^ZJ$3T|y;#
zhtTzT82uh|W=wa*`_Af6`oBveH-PzF;$H3%cbGb{F=G+6@$C}`8wgHGTl4=(lEL=U
zoMUW8Fwv6G*%~PKYD<;5JHJ}O<VV5cz>mPPog9%Q{a=d;Wrtv8b-g{Txb)R?s-$AJ
zh_Kr=97GC6QS9W{3UrQQm`EFu{MJ_4A)V$8+`!sFvpMS*x+a~~zIwh*R{1tHhzJ~)
zPB5A%$6!2+O75V!H?8QTU#$u;F~!Y@{%;C+i1?>HG+az_(nuG-Ah;jPsDWrny;)~d
z)IRHztZ$c_Pvb?6If~u*Ko-Ov1hBP9ZQ~{Av$$a1wLLbK<=V8{BVgUTAJF~652Ap&
zd3{4&eowDSZ1<YI8buo5(f;o&@U{@~yw`0;D5J3G+NDrsHg#oB_1-WT>jLks2}0(-
zYpcC&XD&yNEI`#F19u-OChyxx&J|)(Y&Dp7Vjo%Sv2RASPD42xTT{8Qgdv|%<jB^E
zVBev?r+BBhZ+aXG{O{E1VSdNXcM?NQvICM?%U&Suid*mEisPJwH0jnVL!EntDW#>2
z3=T7Hh>-aoA**dN^ES&moCfw!wIfDqmecQH6Rj52wwNHL+fXs)i;%fJV<7>TMW$)T
z{~5j-LhR0u@2jvElxwc^@0Ef+`g+nckTfJe>}F^QE%}n!TpK9LyyuOmI)gwFgvvrD
zgURxILhDMu_?!<HX1Io9=j6Mj1UoZ9R>6XKM|iV%Z6D<aVJu17%%X16{~LGFaUdx#
z2W)MPws0Ye(iQfSm_Skec=x{M#f&AV-1+OP(^xX=u!AT-D@VOo^-zejul1wZJkGOf
zIoq?l5jWvrm}?jAeng(H<)k$b+5JY23>>%raa}u)(S0rczey(ZAa~L;cLUO5k-EQg
z^jjv;?<2JfUMI&4X4NuGvPgDzTrwN8ok7@9eD;<ikWnv%>w3StACkJ`J!N(MoW`0M
zwSVxu&Ii?<p+}@D@AmK<@^VzS*)EjRyk&dH@;|SXfc~vIlcMuSJZLl&k%k|nJm>We
zXQ=pYo4d}pu02nS7!4<aheX6JDMri4*q75~^JKQWHp3()&`Bw}dPDNA;7J^11(G~V
zA)S2@!+Yr2w?kgJ*!i?vF!bMS6KH}>mLmVWXNB$!Pga!dg}LrO8b=x}a9n)UMf`xU
z?BUL}jyNa=@WQHWuD0E6&F64cMV}GO>SFG55QnTjjz<AhP((i*>4<?n+tfQ$Eg{;5
zZyvJ$GgzG2R3=?k!^NNJEV`D&h88F@MrH~6N8Y}BmYvP<b&z2(ap%@N4JpM3bj;Z(
z-@Q&4X!|CFs+K>jHjD^kkjjtbOlaS-)}14(Tx(dfj{dQqEebZbNKIFq?@avxP#rS-
z&t%mI+Ee+DTbK+DtdH1Gkb$qWQ<?e}8$A;?LdT^?U2S(A^7hCcu-&PHh$O!JBut%}
zkDAhQqA!PZwgS6S_`H?}&O>ckax>RBcq!j+PfLj`Cj%sAFmqzFG5>o^{y+8oL1{$c
zI(OzIb{mJyRbIT~%ZpM!U_X$a+v0c2(E-h(7^F*oAAEL6LnESnzV?SSROC1f+;zP(
zwkcb_CeCT;zcO>w&tZROR1un+O*&8f|A!XAiM`H#zQO=vzDL->0G17TbX~32Ck?YO
zBL`^VXK*T1fM-dP*7LRG$)akz*{#^VHjUVvmlHvPE0mujUX7|WBf^PaSe<J-`wraU
zr~g@wG|X@J32%2l^P?Cg^PY<tvUR#a?1*!LCK-bs!t18`Yw_cG2udyq7*{K%LB;jg
z+Wp+S=)TV8_jY1dqcPbK%{^_=NEqb)AE6_!n&)$p5{GY^zmy{iQ~qZ-!OHuMC5|~S
z?fLVTomCv=;ip{4l^k9gOw*p$``>X%!Ctn~#;@T!`)y0iGU~ljC&UGTWg_Ci4Bw@L
z`9^GPZ(yl^4m|Q_Nc0=VRavaUkpB%#g+->*8&R|;CSNzbE@o+lav#00Mb|mb@7VP2
zD%JUXA4H}jBivO~w({hx32{*A9nF2LgkP3ycrRPJUbE+?ag0*bq8Q5RRYZct=b`ZH
zCiZ_p;TOQO8z#~8Am{ic=%NIE_z%@Y)O9Mw{F;Yq3wY$Nwp|N}BVqtSUX>+hzF)Ts
z_ow&sf$66i_YsvBjXxGM{lY58DA?qwbbCQ5k3kEYb!hP<|7RaKWYq8bGx-fM)M;-#
zf^q`YS;G`K#ySVr?`J}5tBKIl*MK;e@c`#*2h`|ANk{ZBtK6o@BVnl}!jZwuj(ky7
zjely<nKGh}>Q%u1Aif9^x^=p`hT}>n_3F}&t88O(R2ie7WJeQ^1lAZ7`s`oq<$8O*
zttAQH4t~dO9sHsiOma)P9NS#;^o-BF*Q{#Sg?8fv3>_5z9gkfDbmO^`k5M$2@@?l)
zFExFgZdE6_-9_fs`bWkD#f2Oosq&-D$(ASl{zSs*+bAKcxPIvK5(o3&U7+#YH-;I5
zV*P|B>;Fbm1Ko^!6?e>eZ21`lwEH7d6t+Gz-vvYsVc3(TyDiTOx|~)pd-SK?_=3mz
za}u{Ma%coR?@4l{x@qxB{x?39M0|d}d*;7bGh&Xa9=)j%$=FM?%;~h~cz6AL?NtFx
zE|<x?UUi`?)az&5PZTpb|3(kuDXmk6$kwJc;Rv1R$()icFsvYFxg2yc@k+E_@9L1u
z+<#5Ok}WX(k1>UELrgz+qCEwS6#~d6C;ml_zFBVWQF8P7YfK(lc=H<SrRU#Yg290;
z`Hf|BOASBo3p%;&3fd%A(z}LVXIEzMs2l^a3Kvv$mtd=*`CpdwD6yaR@ZE4?KXLI^
zl`xbTA#$NXe7X8_gIP(w63ABMy*SKs7;X%8PpWAir~WWAT(T2L|MK2s{E2V8MS3SS
zfT#EH#qhv(LR(49OY7NYs==!E0RYMDyVm-MBevw{h#zH0K{n%e)2gj{a)Z>TS>Wt=
zZI5;gpBK~ZsDxhuq8Ln~pAb+RP%)6rPfRT=1K+}Nh!lK;jM0UWttlvZwD|xXh@vF2
z3ZqFbn6ruoR?+|DU3Oe(AY`rKLd)QdNY8Ae%6UTjRqK{1`oYK$H)BYxPnf!#uEAoU
zdZE@D4fUL8S(-Ih3f`RUWC8yj0pti&NQm0&?nGWT&Yt19z1~ip${H|I<%#GCl6KdY
zH14uQF2MJi_P^jK`Uu%7j90V%q^pVv8;{fs<2KXfGWo0$I97D1sm{%0DsPBwOc{GH
zdLZ8~X50$R(!>6IwWr|Qc!|cI8648OLVS5@x`uKxay?Ke*N#$TO}N;uIbyA(iPOnr
z+%y_b#Vw^WGeJJtLQX7hnDOz&AUUO)hQ`fZ@%W@TQWJiK_M$}wc`J*|YOYA<Jj&%|
zC3)X8k8K<Aqabaf;Q|1XzZOb*Holq{^1Uq()@A1(+#Vwr1js)BK?P_Um!*K;@gq{u
zv2248n{`sc$Gd=s;g7aspnai;T}>ms!z`pVSUA50It3~eN~+)Env6M^&5XO=kJ5DT
zz5B(tg1ef`^(m#&erg3hUPbE%^TXp;4Mja>C>i`O4y|*ThQG2JO<GpZz9(oU!4LVd
zh%ow#z<<lLg#i2J2i*G_o-J>T^?j|~G-TC+1Z9E>6mEV&@P+CCnh;h{H8_zCK((>)
z%ZGvX-qZ2V$?D&@($dl_c8~Ii0B<qADF_r$Ufv%}6)ndkeUrw<5Y>J!E|8$##O$jk
zqOeTQ=4G0+!n_5uZ~tf=;@+6rToFlj4GH8V!%reHV)o7pR+VdE>no;Dp7~W!C4;AT
zq@-47G2y2iW1`Qf_5fg;W*e&|_%XCevI1(x4xqwIv)dc24>Cc+XYKqE&$d0BM?9zt
zjG5}egPP%WaEOq4SNm@y@@5)1c{nK8rqZU`D&<Y7I;+Fb7BdhM1P-o(W_UWwwV0t{
zJL7apu=qgPKr~z|a_=fxx<ZA}Ey(o~k9K*2vUc^SqSh)xR?5m^ryug2ueCHRMwd!S
zW*;L>tSe*`#_*<s+vWhfkD+H|g~?Pr+H~=mhXEL!e8jz|h3f0(>FvWtjt9}1vaT>s
z@pCVLg|vnljCQLF;P<t?*lME`E7>_#Kb&lZLJ0N(zPlFWYIKZ4VlDulH$lFy31{OO
zxG{59<N;BnJ?cWg@eChoWTqa;C@ORc{A^I%?*Dt^2=K1w_2nG;oo_cz#)t}V>BYjw
zer-iNKpGSZy(}}wBaF|iQUxD;{&3%Pbo^QAhnC1^dR`|Ywh1{oX;Q`xWZ(d8@`bY4
zQhICh=53nz)^M4f<1oCn({7bdLMp9$W3M8S;!m)R0H49k`f~n=moXE>m$;wdWDZgS
z5F250OYMG8Y}UlBHm%Fh0+5n&g!rnEcAPBZCZ3OM!{aw8iRN<YLcU8mmDg4B_v%``
zmsU)^#KM?M2RK|1WQASHI%d<Bf5QL;BWwQ511AR9%>fdGy^74tOwPlY)YNo)^3;i%
z8u)-v9ejSPQAMP^QMFGh=ztwB?Gx2-8jJmWvJD!OM&e-tW9ch`XjUp`Cozr@U$s>2
z_V0YpUYjQU`KtJy*BtfFh??+7@mcuPt_K?RA9AVvI4K0nuqJYs<;1Gx<)m|Fm$L7W
zNs{e}AbxO&DFuKJ4a4DkgL=5N$Fkxy%3LT0<oKt-atXdjUCb=}CY1KKrvuJ;gwcBr
zF_tNf@{~-4V)7XqAx_Zw@nnP9CE0#6MFF2bIvJFZK4EGwhi?O(OB>429p8t^f=27r
zTZi|sh>4U<OpW;d#G6gzH_YzgX+msV^^XK1;ord?2NMt+eL7D-V4ObIy6a?MuoZP`
z^tnMgxuHl8T7Og@+NV=Ydh~vn_7G}WCLa!k&=V0fl3v=Z(sAIBAlE>US*6D)@6=kA
zTxC`O@btIsvI1jA^oOMEC5FXrH;2neMf13I*4JyU&qBAVdB|y#nBX~M(5L9@wacYC
z?)eID%EXJ02P`uW;iRl6GPk42L>)D>Jj=|6PFcN|NaX0>NXKMtsh+r}4%6*-OD<0}
zJq=Oj?m3RVU`dGCY2@Rj<_p*$f~0Ay5TM{pmJ0X?XoT6DVe-D(7s8i*{kZBuxzUV;
z8ekH=*lv9PvQ7Gag@DiblSf*?eid~?>5Ef@$GJQ7iV?_1LB@HNFpSDwqO)x?CAUGn
z%1zEyM__tZsYo%S?J*oM_X!2DhYa2C#*!PNSyo21n;fL(r?C{IMV#u3FAffWK;l0a
zZf+8r&x^s5T3OL<&4+*tOl34%BV>8*xJ!u9=G!Pxi57RNUtQ?5zJHYx`emm29wz(r
z=K|q-LgH4dAHsyRB73+RGnLz<v3=f*vs}S9kBf~P8@<+_FyRW!{7xZq8I3S>pCAi%
z4@?O>z&^wRrD3>@EVde8a<>WtP<2W*Ce97J@6<8Kx)kLwYAb~RE#PWk6%cG(q^`-i
z-3F2A_YVwh>_AB6@AqVVRaSP{+Us(y<h`3>+RGNaJsxL`^v9;;W%CGN#y5C`PsU1O
zrJ#a@Up{oHpTyz&*g1CCUYa16#@V<2Xlq=VU}L7_5vxo}%`=oWnNHvw#rCSlTxI=p
zi%%udutKhqxB9LF??&Cg;wkE*utgib<=Jv35|eN3rtRy6Z{>&0(0sJsVJMG$Z_x_t
z`Eda;uD4ptRr?rkPIJ9I>M7RRprYPGCjvx%B;ocO$w@S<H!RvThiW2@P60M82DPBO
zT_WeU(R9v$z;zU~B?yzAjh@3*jj_Dwzr;YEhMk{xlbXoUr|?uA8K+j0g2KW@3DCFq
z?y7X_2(0HYlshcS6Y4K_uSlbzVNs&%Sg3vW)I_jSi3}z~jq%I#$!<-g&*zHQF*!H0
z_N+}oUupNp`|!CL<jU&syJY}UNE)LeEm()USBO+;XfBKuWP0ce4u~#bVb<UI?WYvA
zqd`GrQ{VO&5IY&~j869@19Sh^>t`=+tz|PCOFMKIFv<}hHD}O5sr6^SNk=QNUefpS
zN@B~G(zNoeq|OxiE5-H0hhC`G&%z+!^_$G?JOc5y{dB36{{DE!LcNdWXf*La*n*P7
zSR9(yOQis(M{jz_xXZ`~T!N8qzckLgJSmOb17%PO<O5MbCu~L+FZy%avOp>O0u&Y1
z+CuaxOUUPd7^ATvC?{5o9OPWC3M7v5G;RUqcwcSz-3OX(S4+Ha#yQwK7(7UtT!>rm
z7$%u1lg8IsKz8(wq}v7_92s&2nv*LySj-?I>a0<yl<QzPR1vq+n;$Ive7r7dpdB%*
zgcQf%8cE>p%KG6Pic+Og;`Aj{<QX+2x4wWV<87^|@x!TLVL<`?{iLVR0s+>xn~+Re
zLYL>xM;3P^lsvJfE+(jQ!Le_d!`-41Zo}i6>-KG{*+u+ryC{0jySkg5_gUkSZlt4F
z>{I4?t<c>;W@V=);!GcQOqik*FQdcn*M8?_>)W%H)%f$DyL8oSkBftZNkML=mX;~R
zYsUIo9k!tt#SQn~zaUmng*`!r055@XgiXH5mhUH2(tF7sO%*Q5txN)T8ir&)d;P7N
z<_Sbrktzd}Ej^l7^IvR+R@r>VpD@Vel+19DLUk9k@2+7=XFp>Km)H&HI#oc0kS=1u
zkuISuH=PaMReH9btf}&1_EW^TUF2KWvfQB&*D7KCSaTm_O*<E@j3B90Zo#kG8g&b>
zx)ByZ;?8efRUZI4%U&`vlao!DmTWSIb7jLGJ_E(D?g#Qa_A9uwhNn)vC};=c%w@`+
zOgv}6MNMWH0P?(YsOw6yF&hlxX66#&dQRIDL>rm9g*Q1erSXQvqeUiK9hHv>G8cp<
zqxg8PIn@XTk{pp1l;2|qF$Okpv9hXW@@t&ERD>cU=^+eap#&{qzF3d%IcrKZ>y(F%
zeLYy}fNFI*25eh!`5JKz_wTmG+gy@S^Ro*3qCF1sQIw|c(0=>j{#bW>Gf2A^orR%|
zXJ8yO>XnB%JDKk-TuoXo6s&&nG`gs%^_H6<5)r3>xG{g=8FSs)7wBSB9$XG!?E6|v
z%;1Fsiqy;+&!7s!7B{2&W)i5Y`Iv&X;n&kQnAyM7oYupYpH@KuttI4ku@HhwpW4U}
z|HQ<=Fs4%<(8rJn_I>}10CN(jL)RoWid|yYxuUt1IgNYJ@|SbfWFeS6jlQ$*wc&q1
zbC!i;-T7)i{v6X6OF2V0@LuUTm^nvx;wzy*&N$tOUoD)+wWWLCItUS3n?0|cSQk1h
z9pK7rwMx@2@?9P*ZoPw;>{+5iUi}eED7Zy|j<hHD+@g1OBXzyHIE4Oy4AO=9`h8+z
zjUkoqwGy82s;#d^;kkm$f6_#q*>tds-*wInC7%hTm|nVo$9dBbH}@mMbwLsWzOO$V
z6NB!{cx^xm3c((4ojfA?^o?hI5DPg0rb|=nr$Ed0y+8ACcLufVXyqxJp=Ib4A@^r~
zYp<vhF*;co*r#T?o%4=|+1iCn=Q=f>6ZY?qUWd|HI6_WaY!Pk4Jd<~m>Cc*UEwzQ7
z{^wnBf2rNi4%c_uiq~_k_sD^Dw2gB>cez|&{O?>4I6YecJ<&~zfCx_SEk=`!IK2pf
zxT@mos&`z7Y{VtyE{BIB{hE-LR*L7P2p>XITjK#<yBq3vf{1u!Ev_aWh%Ko155{$K
z(8{V-(5P)~@7`<g=6bcY??SUW^XV$$KfOiiJ&2}ZLhcU3&fr`{wL%VPPOnjD7-*jr
z7tgMuyh*j|_W}`g<pdAIs-J(iOmga($leoU)$l9liOs6PkB%><pJxm9_#NcqIq`v_
zd{CZUe|jFBW6TxgdaH&qlOoe)!3-dNgimT-wW;gQ62J-EtEtJ}bc#MFp!KeExGkat
zdWuMy!_477la#1A&tuVjxe)ie5Y3z6=sR@cU5W&zf}gmTP#ZsD>ud7}9wD|7e${(;
zE&l#fd7qOB6pD_88=$a?X*apKI<=cSZnn;{MppRrAk(oeklpdlw|P0ib(?cBu#!OS
zlSaQUejG%<<}<*MXx?Uy3zZ;)w4(mb5EXsgC-)9<;d~E$-SWkvFIT^c4S}0$VLc;o
zUoXEHP@|$svwzOTFxTFx_z)Pb(%WOg3qtw64%@yXAmpEH|BkHBketEAb?p6dw5i5F
zYcrI9dWNgOSvnm~TM!>yQ<z#>{HTPa+<&R^wH(|_rCJNdc34<_>X)<@Eu6E~;doUb
zQgZ2c4wXm=;Qgr*6CJ7;O5K-xMiOINqJE@qx3!&fdM30P5u=dSOlhsRcOLK8p_+A2
z7BGMtUxyhI!&ky_GyBX&o%~^R%%i}!>&#RtT~*4Q|AT+%OTC@KUHnV<f$?|2xF{~U
zZB8CK82$CWm`rVGpB0C~J_})S{x9}Ly3iNuB26>QRFFWN4-rnVz^7hq0~my~I{7{|
z&IO*min-}rUaYuB_Rh!{WqB)T^O68}0ex4c`QzZnz`yn3-zEl8Y1;e_n$7xMZDQ_L
zG0|5UQ90Wxvp8=hOj=9_s}7v`6x}^8xn6VLwmMa)>5s|wL%*V-8mF_tp~ys7Gobt#
z9sL{kZN2|z6&8dRb}#u|2#soZHng7lbDF&;r8OJ|botb7HWzTx@C^iti3#jv&J~4s
zA1B2gPiurMv3r2K=nPWDru#Bu7aROUxp*jJi4s++2}*TatHM)~H-a&DE7AT->e3V^
zNbYkeQ3fAbR(;&qP9sq>#Erfxh{Hm>+=D?1)n^i8FjgG5M=hOXO~`yXlB3}5GK}?F
z>}4ODNr&O)sIs8%jE}4(J~fyEhP~GXLm8yX&A9M)Er_q@i9V~iXV3HgJnBLi&6B-H
zgPt4MbfIM5l)qkFk-!Qz+z>zz;h(+6o{Vy*^`rz&{x&URfX%6|R~$f`{wB`COfh`u
zv3l4}k=NujPb*xcJ$cHOBi+>$0a+ql;n<)?V$tC9m)ONsm%HLKq~ZJVA9(eKpgR=u
zP4fI++kU$Ry?#lGGHv9@1+(rNd_AIaxoaQImfE+QFx@Jl)5mIu@znOb>P$9_U)(UZ
zw%HDLnD9ANJTt5Rs*3<!bfhT#SjQC%Kv~Q!{K^T4n83oBNM7y-JMP;)xq6ly_sRVd
z+f6{vI@iPV%gUiDJqgmvVh&^tD^t3fNP4oR0L`;kEZgZMSSAFzhf*z6gfl7XZ<RXq
zU0ZC*zvR}mt!yxShxj9QfPi*iiirYA3lsheY6{gZ5sLW(hFy*R7LIZ9tKD2d^<2*5
zN_?6r|MNha!&+bi^kb*`Eh<e=n4q*ntB<_G*l!|ss}tIfk`*U**EuB=5S9W~PLp$7
z<y~p}*>5Niq?&5hwf^v;vq;6W2uql(Y;3B{jY95fOG*F=>d*IV5g-R`z{-4D7>xdS
ziZLUNlh9Z+bAZS57Hz7ubk6!ZsGpL%l_QUQUl&%pc?9o{10}gLxU75oK37A{`oU|^
zx}CbXA;T~c_U8tm1Wj>foR&LgzqcNYzWclmmQMIO2WJXx1_eZYs+q*EMvd0n^QEm!
zq2|#hOi*Ffkx`IWB<APVzM5jk2&1q$Q|P`GS?5oCJ9cRkfT*@MI=zGWxAr9L8<TQb
zwN5X?1=!T7rDjqP#*U0t2l4VA^pjD~%AqZknz{Zg*lKPJ+Uc;oV-t^Bn`RNvKW<^p
zxY7m0q05U1WK;?aNuftP2aETXQJ>gUg5%7SUBIhSneDp)x|cYOajy<lVg?Adl?Oh5
z3&n8QU8Qr;V|>SUjrSHGgJKi3EMSb(EmUvE;lX?vHOelvY^D-Fb1MdhUNQzUgRwmU
zLoUfmh*zSgjw{g{ePri;6LBibA&)JKq>yR``nUfAKAg?O!1ojkViZsA##aK<rygi6
zLa+6yf+doIZ;Y8#8Xm+rbfMS_-Kps2KIAB@EaS}^xsQTH6V~BVopHrRD%OsO^qiBF
z`mW{ex-knSX;B?zXfhkW=+>Ay`uc=x6%A9Q467#sO`Kft@aD+J8+ZImWt(!ij1N~F
zb9AW;3O?bM$uWh@L!sUnUT}ZRF1Gft%$vCde{)c<4D7ah6qM&js=aFQSa^AYW&%3R
zv37TM1-k0aEQtjSFTbIKJs0>oIc7(4LW44qM74@Qwou^ImajLLCOqp`31%uAY7h)y
zYNIVXXkHC|$JT~45I{fzo-+KyR5Rmvd=y1WAN%vH6(C@w?etw=b>}}bzkHtjkc^w)
z*P$~w&Pv*NcB`-05^QpYXPp0Yisj)b@5FXam)Tl<+<j)iucM_lVc%JeXpco)!hsUP
zT>N)~d+n#i!zU^hj@MFg(?M(nQ_1#7JrE1SeH2VA`p!hhXy@>)cd4Gu93)WAU?tfe
z;@vc2hG<QU@=F8sa8!d?kEgINpM}@_=>UrP*XmiOwzK!ZsY`~O+uXEgs+tIWfJpHJ
zatAvL$9C^`3Sphrt=Q9Dl2<en*5yS)ob>rJYXua9HqDGrlwTLLZGHxZJg;$a<TtNG
zFnl7hTG9?)_4GDkB_=Zd(bSLk(l@}C{BHkNU#K$frZ3IJ(@`;W>SN9q*D&rTAKKGx
z-tUCwhwIV3oRC??gsH;%CSaC&JZk4sV8sZl<Lk|iF&JVO+Jv~u<IAzLq(~;)#00V{
zR}%?rL}gKb4Rn>|Zfwgu-G+c}v%4o);%pGnT`}~8>6S#wm`=;f9~_ZuJmU1L>VNTR
zehy-{u#b1Pj}groG}Oz@%GxZMhxxov(`J_VZy&3l-|+O)FPLMwuPp|vl$qYvl)`V9
z?+r|AL%O-?1qV5+>cXF{x9~@N_<jFy-D~qarqJ+=RT!FDEwTYVQQc+B5Se2AXmnB4
z->Vou%gEV}{Uh7fzBgGq(?oY*kO$6EPCBBz+B8LXazVV9>pT97r~UhyTD`4ymRGqV
z=Cx!5guP`qQIFTf6}G>B;4)?^k9S`ubB5X%gKuO2b}+F2!~Ce-Et2`o%+xTh(*(WY
zC^_@eGDn^z204nl)jT6-*|YXT&}t>q5S#|*rv7VktgwH<9w!Iv#%<by&Q=ep=H149
zp1{I)<tnRS-YW055^7bi%V=30waYX2#Sx~I>xx75o1=2$ip4PIweVTtR&;%KjIl<r
zwSig4<jnNAv=gVSa8C5Ic`zeD)=3d4)}FCH{?zzWcYJvG5>vCyTcm<aRMX^NQ2}IH
zUGm&%Ha9GXvQERp-?f`smq}`;{K?z?1z6S9NE^VIqgE>5or)}O+1(%^Sa?d#!3YF<
zrU?NMmz-dx=-;Mvt>&}0^J^oX!TUJGB$=yF6c8kDP?La^fZXQ&Lw9dswWt)X%Xu=G
zBjH(7{1+F?kc;u0$kg}#=tt-4bKc>|FSyQMts$8<@=`tRhjeMounbtq^YNZ|AQdpV
zicBx-(E9epr8S$FoqoH1jyHIB#hn|p*7K$lKzOQVlbpc>ksOQw&fAkFQ$_AgyAhmY
zBfX_#=H})uRDllTT`JtR8u>3XYdJi%f$)pToE5j~BE8O)lk4mc$Nhh6_d`?){ntnZ
zZ~X%puPr`&3q_COCOr4eBsel?>sik9M`#<Fd(8hkI;eB%j3o|fIJcm3)RU9pYIsY!
zd^w*iaFO0BCF=FfgE~aVcTtuu1vRtt1+hh*BMyFe%qIZk1hisvo_~cs`<C7X&WCSb
zWQHGXjX(C!U0;M8RH2_ymLsmt2$`RpBO*E8L9#C{CI2hoJSkC#Jz+y_YP5D}49&S4
z+RWNm2!tmUYg8q+dIoh(+K=4T2(OwO)I~a>H#;x1nCx5)UX{2Xg93MQd9C#(D66Hf
z0Wz7iP_|JhK|BmFh-$PtX+VnHQ*b+&A<%S9-YOPq`nQ7`_h%gn#y_07mZE=qEp?vn
zVj_x}+ECI~_OfEdK22n0iESK1xM|q4p+Q|%Vuzv6bjiv%3otvC9RL1eGr5s?Qx0*1
z_U18$<?o%h3$pzT(#GLrT9fVM+d28LggDPVgBCpP2rSSuTDX3&)xngMr$#?Rd>c}0
z?aq2NXYSIj(Ocq#>@qrSZN60#!P|IQIVldcM*$J*s||#Ek(c7zA`>LyEMFxd#F^vX
z$_+m}ock_^lGM=6nAsYB$Uxll8Ij};T&<MP{7cRZX@ya)<bp)SU$OQU&&m`o+wo9t
z#??yMzm;2c8RC41_w42Hf`*3M)qFA_bm#IgzoZ$%n6j)^^U>&OwEw4}&&NUs2%@Q0
zO<8J?y}#n-C765XNu_c<-^s{INZ5`l_->Z>Ec)<B+SYWJN_Z0UFCvGP9<qfVl8o56
z{Hwn&6F$eAHfm%mC^(;5{cU!1TYbQeP1qmTVf-<*A?*--IFc(Gs0U;A;4X3Ax*x0v
zzt*1^*#q1o`6$ULhOqm}Ka0;?w4i&P#XT*U!no(nR^aZLFv&TYWGZaH0MTedh5h;8
z>ZiEe-JW9Gn~L5sY9XP4UyMhOp~oFvzN#+iRiXLI_*61!U&qU$t@c+U$Jezi7qn4l
z0z;y3O|joE%{R`i2Nvo(@Q9Zs@_#hBWVN}3c5sn5@^6P`8mlj-Y;NxIK((V2$oI74
zkHDglcoHl-bFyVx7c3V0KQw@a92IGWAJI^arExVI@-p^Wa2uH$9C(Tv^)xfM%Ab(e
z9WNG#C>~CZZ!vzD!?@=*C<wypE~ycIYnJ#<CNVKG%f`*vc6ZjC`m3V}>V-_*l|+@0
z>AImNT`2IYmyVCC-iZI#KRH6af1V1y&1dMX=xv7in1X3Dm4CF^jKv%hShZ9z9aJYd
z_yNcu^nJcFO-QCHjwH_Ja$S;QxUnlT|J~WXZ>e*2GhgurfXu`wzCQ{lDd2vmZ!C`p
z64rO+k;5XH(+GL3w#5DH970?rY^k?grR?>5Z9@NUj^%RBU_0O7_|hSzMbz;)o$^D@
zZf}nOonQ<s*Ea6%niPND6RF``v1p!x>ZTxh0Q<7NygW&^kDJ;&C;Rb`3tV}MSSQ4o
zPi?5*k4o>@Pn~4?<5daK$BmM$KR1m`Ov?VrLh)IL6GMqfFR&MeW>l9D8}o18o}Gk@
z-f@|%cLtyd#5YJ0tW}$)Md<moV!cXTvGT~!T^CdA{kG!b?@vLgV8>vTds<DtLNZr=
z2W29~Jgm>13dCHdXHGyj(n|7WgqanJIEP1V;4VjA$hoQAI&XW4f+MjZw5|K1x>j_O
zWq8d^NYsM1R`w(X`JIm4y7Q0r8n4imGye}C9gE*}&7HJ_O!z?;E63*8wPcaT33HT%
z%|$x9qwsCNEyKjLsUf|LFP31`ZCt-6aO<!}ngS2>RyLUr{()@m^lWS@vO?c>kK(GL
z#eyTxIouW&1f|7>vg0z-Sw7MZ+`7fH8NU6Ql{gMNM1oy~&DFd{#U(*2zBHI77x+RO
z6=scTso=-wiPC+|aP~ZxjvILGSeIk)RqZ*5vZ!Ff!lo;h%5~47I37;;D(bvSv4M7!
z>>!Y4;4AP)4#t*)eDZI^>F0B<gxx}YL2_p)NJF)p3%j9g){}?Q4_Qk?x#vpS9xmqu
zLf{tEu2+-i@%T91HIMOXbb^6@wK(8*1zz|eC)yb-Ih3#{p?N1%V6Dr=Em)(Q$u?M4
zLBdr)KqIBtdIk_(s1;g5%r#V03nv9`?+gA$^5~HmFRoTvKlqCuOb2vE*MG*ka0BD-
z`obVLh?xsjew80d_q11ShYiT(#PzqgcK9Jirh~)Co!k#_wCsgfB(vhzq5|U^KWl7K
z49mgX=;y6pYw{$?E(DehKrUAASvxg>$kd>HK<0Mvht}e^H6mjT9P-0mI*;A6g+rc|
zksA^yD=740x&B^grE3#hw81Oo>V=ZZH$P~}q<IGL-8Wh?3F=<rc515xy3|I`c#<v|
z-*Qw}zNA#(_-~)8FY;aOn+W~h-3?*2!|i2`s0!+baQlmKu9*au`t)Ugm>hhaj}!k;
z?~4jm-LXBs^tS+4y?49D<ykZnZZFG$DwveuC<DHtv4a5>8!N^<{8UE1NV^9usAhqw
zp^l}J_E;q?_7*CF(0e{9f3V56GJ*bIHvnpXCPeSCzQ#ZT_GqDWX_f3p4G;k0yDiaX
z(zQRpUfL%#V&!B8G7S0?a9NBX8JfpR$-x|ku1?GnYh8#^Z{N~KKNcuo>!l#Hl$U5B
zuCvIaX%<E{^Sv@vQ$=kryugiYyASQ{#o%nC_3pEZ2DhpFv3aUh?Ec9SnZ-9TG5%7w
zc;?WoROSKxUHl?ho0HxSPluWMm>RIvY}djUa<lW`;z8;|`Zc9|7!<lu>%Vds(O1YC
zrZCR}z=m9q?W)qz=k>o749K7FT%=a=^0ofc4ZuV)FXUkN(rHW2G=8MXC6akE5eAdC
z<KJaZ8B~E$B>o{`e25x4$@WdL#OY@`{aFLg_m5qu<?!>#X#&VtB(S(?<tDHApn21+
zWey3ob_ZeugA;0!sr>6BSyOsTI*e?3a<HfE@0<9hretItfhuuB&!8Wk8*A1eFDpr$
z9=(3YUUsCv771+aosntb_L5)G4T#R<&u-&z2`MZmf_)YD7;!DVRjHNyg!W*)`Z<5|
z8@Vc29ACIiD{aU&lIQ?;Ug$z^$VQMBtt_Tjl0839XJzGsWC1JGz9W3>M^UZ1${Qr}
zMoQ<iZ>+jqPZw>YW}Zx><y@(YQUK9ps47+7aNtKv+~wK6Glx`gM!srgS^isw(6H2p
zI!|A1rpY4saKvS`s*X8dr}^B(=xx7v?VxKBUQDdgxU+8}|HoBPNQsV@xL+N>Zn!eZ
zu_5SWO2{;P;w*o<Tsjw~nQ}<Yt}U|xYbV|)dT|3>Hnrd_t$)D{<a>#On(FHtzI^#t
zzsm9R268eybh)$n)K(0Eis{e%zuUqr4nex!6*HRw!Xv%cH{EvOoSj5FM5fYbbtV(>
zuTP&k8auMx9vTFrROAK?VA=ijL&CTp&?TsZtqu*^3b9DOd#^U1T-^RQre3;vskK`W
zcWSmt-=0jLHU7g;`EtVrYepZ&hc(i3_`I#ahKCBd?M1~OGe`L69$na52P5u}N7X-V
z4&G3dyMODYMLk(?k%Fqci76DDGLno6gKSein{g)0=DuLst}L;q8Ht?obtXK!;g=6s
z7!W;Xc(U$8garQSip$td_xuL!IMKHh8<a4P=6Q~zg&0i2cb6ttO6y&5PJdgBQN8_R
z{<a~r;v@6nU0#sG3NGZ{Gc;s))mj_(_HDc>K}ro6u-j7B=)SE4e=46N|LJ`rF%t7+
zV}3r|B|h9ycK91i^gTM(2{HX1>!euE@@}3xa{K7<p*yCsqBQG^{w7%CWw@tvaP!_q
z7lmdWJHjtNPW{KN9_KiPVN+V!tn};3G6jAjh@^hHKNhU7VK%kpRviV)qgbIf2%9hO
z6=22_GopMr$z6oL!?hlp4j!QQxSjpg-m*CV<0&Z6Fz;X-x}j#8sbDxSx8Y&N`d(G)
zoty9QH{G!54V!uglmRJ#2Q40lR(sJSLLg40ukQ+ll%xS8Dc1j$O%hVG%biP5`-yrh
z&${?Vp5{?QJk<Hact#^H6BixZde$o1<ZI}m4y>Sr8*d3mazMTQxT36V9sRHI?&Inh
zSVJxNkcY>=+if_UD!yvo+c<PzQt2JWuCB1BQS^t-3B1hTR!-phScGJGmWQB1(L6+k
z+=_UP@FkYlAZz9fvSv&w$ew>Yn(i}~%Jw>CxCH5j&Tu^cvOJHr#D+8?wJjiNF;S2?
zFBix)^y+?Rsl!cD9vLU7U4nzNEWtr&Teg#he-`*yvgx{|>w-tgUok;yd_0cnbu~sB
z8r7Kdw^;kq)8YJAw%2_m>Ght8fcC~Xi5dZM3Xmrb>j1nEms%T4(q_0Wk<=eO5Z?{1
zK<AoH-cqD5Xhp=my|#{p>?SjUeW#~FSNk95c^Fco&&BpCm6A4u+gj{P6pc;vPtgk%
zb&vhESFr8bK11Z}+_p;Pms8iA%4rO<C?#@MKk<9g6jBYpNR<dj`};3Eaq@F3__o*A
z>)^Y`u~UANSZea0Dbf}MSdBmEF#$qz@tzQx2fh41r|LN1DtL^VQqSa$S>A1d+x~mQ
z+w-3_Rjpvfl-IDpJ0)DL<T`q(GsCf=f9V+<7qD@f>xOR5=R0ugc@yAcAk<JTuVvU@
z9U>ng=!h1<tsW4PG>R`=Qucod$7a#MJy7XkIStLJ+}5!>Fy`u03~!{HJ)9ps{z6P(
zz&O^CAT%-jc*9eYBmgn?2z}Ws&X_4v(XsNy2Ms8#J)G_AEMqM#6^waWK9RO<$xOg*
zCfQYxLQq3A)BoLG=ye8gwAKvSK=yhhIWKdHT*?C<^O*qF)SnB1Ssd*Z8&Ewv;Wh}y
zwZsd<Z5q*Nfr`C-yvHgNoDCX%sBkp4&u1NKzbpjC4}A~D_7z_phE2D}ebmn^$7FY-
zc8~vbg}|YaEHAR(NRVlXb{7KC@Yhre*w1Hhv2z7nB1f$Lcv3-jv{@@F3MB%VH=|%=
zP61S0N0CPo@^inoz#tXbMjwPtu;YhQFeWmjS;T2IXdt6(s8FR752T)sk9|hHM=S=b
z{1GmJS7brqE<?%V7UR53THJ;FJfF6^<g9iZr=c4?2O-_sE9JK&syj7cHwKN0EfJtp
zrVz0Ll_^#4!qwMv%uKhH3Lj;drOd0GzRrV8x`&WSBVS{JTQnHa?nWqnt=G0So}Vo9
z?%M8}{eFJNx%1JiEbG)AaXqRqB|e{5ZYnc|E0fwD5!^f=;XaC0@buAcdi=fXs@tr3
z=D69%AvQ#0t>T0rrj2+J1m8oxEXps)=g}@e=P9M4-39LFDhGJU0}W3^Mjn>>$pgP|
zp=h>U&%m_IFep+bC8o2$kHXBY<d*mv!<zKO&{Y$W?uOncNh^NKxp51(9^Cmd(iIPG
zC4mV1N^;^3Q6D(@3&PCQ9)DQ?br}O`J@eP{mAzf-9B_d3a?^(kj=R_NsigaUzftV&
z=bT!XK9>GEkLKUx0|MX_G>O-%prU33JClWnAS>XJ3Gu?cnemyWd!U0F$A<j_L!2eI
zEdqtYb42owETKQQnPZZ?)ilBjwAEYAHyE~Q0l34fj9w*?&u>T|tKu2+#8Sp${LvKF
znL=P3F~M<F|C$WnjG)dz*<z4GPmsL7uimoK(LZ$%l_c}cs$WwC`e#o<_IM0y1W8!E
zMu44b=ui#R!GerbLqf-jD;;QP>8O4@C*$=omV6s7HQ1H}G@~io_K|g{FR%7B97*av
z<Kpg?`i3GTKgVPn8)d1{w!w<IugW|0%3Xh62d@bBaBrAVXZzKo_w4CEWt3&7O-o=1
zxT$?F<n~~i#b;wtr4t7Nv*@2-_)ooRUpagqb~VkJ*xk*-`(i*wn0&riY3mKLOh3^Z
zz5io2ANJr6Z;dQCB9=uhp&W&!!U~&j;Mwx^*?YUH@c?OS;^k3g&NJM4A4{ct5t8&$
zdJj>G<hm<}>#pr`l~D0yi;X1^TX0HzHxhBkc>ozoaG?R$Wd^R0<I8;4f=yXXTLU7p
zP%U~i0-9xAyVZINuCLXC0`d0g#BXC3*8enx`th~pc4uWxb?M%V)%U9F$w5z^C=TLt
z7fRbKl^RaK*+oa%c)Qv}CWNYqQc#aUXACNRe^Wow8cB=;n+&-F&ispFn&WMJuJ!Kr
zk_R>YKhLqXnUmMgWGaYjD*pauDH8uq_whs@bPvlsu75x?hdn|%1~P$V>-h$ag{Vq2
zhA+S4I;EzI#GWsND0+opPKp~;&cq!&Z?oXO7Jq5VkZfsa<glBxolj)H$4u2($66*)
zXNDfrmyCYovj$d~i6N^Jm8^byjB4njUqYDwbe#W4iVk~5YiFwY1Tx>aotJW@346p8
z_KgUqp}2qnE7tnNQS>~<Cj)X%Vrf$-2Gs8P(Z5GLeTt3ee=*-eAoS~*4z>`Sz)n&q
z!+mQQR8MC!iM-Fh!CHHuc}<KMJJ3d$WY@uBn9mJg`6x%mj64GuG6A)FE&3?eb^67x
z$s7i;Z3Ils*39vKD>Z!;B=`I1clw!eXpq6P#5?g-u!TzUZYB{7b*giwtDkB6o3Vx6
z9r<rUuaaC*4tx2mMx~*($SD@OQL8UtFz7<jIsZWbR@g$wnm@JN=_2l8!3DR-`#yTp
z4rtcg1~8X~etgGBSo{EX9E$`fa5v~I7->Y1#PjH{BAWv7SCgk!(BBR#Jsvjx=SEQs
zD5G>{ejU7QhgfsSr5jB*E))B8egrhMcAIv^e^OV&V=gpfM(th`Luia^=JL|M;xD%C
z9dfb3dy6saQ84^*cdRRBrokV>z>pzCAb-uA^)@S<(~JmZJ8hP26@jNlTv4|Sd%!`8
zI4~y&UIwe+y|u4lVl&TB&1xp16eLBVOkc3yCIq|#iWgiU;F-b!1jo+K+tJmWXn|WW
zACS$NYkTv@?fa2N<=-JWno`zbdD6roIMp*4cL6dLF-vX7pSJqFeS?pI-^P@BXLWyI
z;i-<Gj$d8fCkrrt_>_6OVv1Wo_ZZL4zl{#d#E!DAI82LwA9x&9fq)Br?%#|bFaHbU
zgee5onHWJbe8q#n9gJf_(Svt(6((Q|(}N^3w<Y$A`>ry--Q@CrF=sb`DIM$$QuzZe
zzimrzkS;@kFu-9j^3sRr?S`wf_4XzrWyM9()7lXk;wLDy@tJ8_n(g;~kkT1Uk|#tC
zf*J8Qu>2-kC|IkM@fCrR8KlwZuby)8mCbN_Z`H1!JGg%Ek;Uw*27MYY8ul~YcG?3i
zTB}#ejvVPqx-9{+TK4Flzq|+5xPk#^B7s3@XjottKJe2jUz#g$ljJ<l-pa6&ISp>U
zF$9p4r_e;3!6>sP)fEh}E+v3Ba3E4=)9`g%E=B6>Jgz9HKId(#n+Tp!n=%5{4cC)D
z7zkT-TSAgBm5f3$-h<@p!WH4uyQwK-*I_g*+(lLmN^nL-p2j@Uja9rgcOyf8b@b*Q
zVPr%@nE8MU`dz)^a#al<D?!7KS>px#Hvd-hnC2F~)OT{+3Nq>ju;jr^NliUezDbCx
zm*Jd1svagbgbbe%ggYbf>kW~h*8i`l$l1++6&@5oj{$-Cv!eZa#dzf1ecYPvOyFKz
zR1TNZ$}si7@cXxucP*GASbo8;<~(1Ig6`6N5MW+$--(z{Y^*UFduccIv1Pw-l>bgX
z{baXb9n#czsVHYJ@TI@h*=tu0g<i&`RZvF8S$jHtX*~FxEFyx2Ba&Do=>1d9D<i<-
z5QLb{ZJpm%NEm3BKbdu%V$ezA)AWmqDm^iMAKd7%G;Iz3tdAEH6$wK=iGTJ7x-X|6
zcPdK>1qK@Z^nAb{&<?D7>4a$O?ASlaRK3sfy>vy(TpxEzO#udvkt?8x)$;4A&?)xT
z91fJxoLX>~O#RhEwErox0XI?*ppHTk;9xp^kdJ3JNiI-|uwD(05pp}gsQ5So%!iVL
zaT|uOVA|L3a5xWBV7)8qHEt$t4njCxD^8S33@lx6{{r_6y+duQBSmokMOxPcx3V@K
z;3%lLpF?lNVxWaVbN?BX<$^By*t>gEZ+DM`0WrNWJJW&debd*suif+%hlKWt6D=f)
zBsgzCj%Jk!i5*H1OY1CX?buI!55`kY*s%@_v>($0uB1vC2-}3(?h+f)YvmoH`TX$N
z2t9R}?FcX?G+SB^iotPs{GYNK-GBEFP9b=eCmEy9l<E2G`&ko#P#s_p@xk)2{zz5j
z>N(s+i|UZ+Mjonmc(zsP9ZwdQGvpKl>$E&`vYcx;c7e^+$|4VRCdvjA23cj~Y+B0O
z?Ti?%;9Zxmw$s@JT=i-0hr(WA!O*4$-OB}b{p4r8<ix%oM}8CAFwZxBFdYM6-L*4@
z7u>3a(xd5FU+JBg_}I)5x%;mR24Eu<_JR*N6zi^ADw?M*G-+2#tK>6SXzxn}VhhPb
zI3899-%S1`4Jdj~pA6K0iyqFCNE7U<k22DWa8?un3k+q)t;uFNDS0+1mn0NK!WCpF
zSmYzbH7Gz0&B_tglu>#(i($JZ!1eiU@_7?&!XPRmZ7d?y{=lbpe<o>h!5e_EqKTCk
zDTTscBqVg1e|N5Q7r5&}9lKJ_x^mF%mkN{tmtD3?`lB#*;#kD21f%)z{GW#ppTg4i
z(HHl?1gH;b^y$Rt=r=lFnkO_UR}HOiJp8*+IAo`?dA5>fi#q%07jiqJU&ay8fuFQt
zGqA5%-9o#)41t8#`-<v9ztaRWh0mdwD`*1n76XE#UOVl^>#@gXt469i_@YN-z%`EY
zc)|Zy$5)0$)qQWDp&V(H5EK|XrAxX)0U0`^OIkXGp%GBz0i>mo4k<xm2<ZV4gaM_c
zr3GO?_#Ym5e%Jfu{WRyg=Gy0+efHXGueI)b-8?!Kj~FfQ2|t5$e!1^$Q7n-)Qa|A8
zJHqq8)3I;+?R(n=<nSX2FOC4hh(IMoy~pIm+CSAm5H2t)IKM@yQ{^V4k%CkC7|-)p
zaOXm4%_{)$o*H2qN{Hn}i-oNRRzvE*YRKT6$K#%y6D0{39jLROZuxQy`)xi!5L#_e
zm&{SO4!&r0m`93758);KxHs+#_HmTLo}dXo{V|v8915g448J1wN91dD^}V0ml!-hn
zu6&hznwGYdAsg}J5dp&oJQB?iyUwtUOa`c>a$X>&?*#|-W77R@N(!=q`nNJq6=x7#
z(+^BC3pDsw0>=TI4yn}BBdp)`ql~tM*_vxjbYAX|br(y1+O|Sh+&%gSPMdaP`ReqD
zGq84DPA!CON%~o-(L%TnOZ4yn6(Z@s65;*WJa`i0p$UuL#(Q6y0A;uCz-`~y?=;XR
z9^LUck!;6a^y_9`w1ho8#$3ezs8#c^`;N29kJ5eOX4P`zAuouGO|m=~H&;{lHWjt$
zM|WqO_F6C(+dWCk+8o54;LJ7Zxk2t#kQ2q?H7kULT$G!y3vFUj3j&ZFeZDz27F6Qm
z$jnI@hDXC)({V7!E=FP<Y3s)E4WSCxN5;k(w8Kr&2QVdSM{VYpIp4NLZk}g-T>+2e
z<$lk{Ii)G$RI+i^vprZ}zN<Hv1!3=TA`Ai(Lf9N1llQ(RX{uRmU%;0u5mrdWAZ^Xq
zKSDCQqc{a}({ItyiFG=;X{wd}PF1BNCfRFj>)P0Fwd289Agb7UkBajWf2f;3MtYFG
zN$P!gr+3s;!Y^bLL|{cep{k!oFB`JXJK(e+yS*Op`?{GO5`;7&r3z{XdSm2%N9b~w
z+B3Rn+S%D{f|!B*iw%~zxIxu>?uT`_b>`gv5Y&;rTDzif7fdU2K@2JA&To;lNaHO8
zbA5a7@UG<zLLt08K9qoRc;JHLOK(jp`r<v)+;2X_TT05hQ7T1n=N4YszUJD=xa1y6
zPEZO#4!LDlk?o=P05Jft`cj|y(7#_)J3hsO)L;Wb-Q2&sOkBHmw6ur|)Na#>u+kW~
zB)>_ie))snuP?j70o)RCw^}Iv#8H#veyRwYhzm8dVCIK>rcsh0GzZU2HsoEv8Bd3d
zV`gyS3r)u)`wf7#{zd^U#5DaGpKM#K&mJ7%z{Uz3C6wa&r8ow8RVu!7B1J5RcQ{x(
z1iSulHrjiq686@pzWanq-mqQ>A$Bmc8wI8imV;M~bR<44%8m;PTa#QtZXS-9K`{Gx
z;~WS^ZMs|r(e*FMGxpS(U(H{IGhp}5C9ZW?%8IVBT=5lmL}w%;3cf+*Ve|+S*-u%)
zXSr2h9x`_@UmmUeseCR<&Y$p}-$A9YWRY|-RciQi?-wt{8y!wrBr9e)JuLhOA>0fQ
z-uQ}+*kA4DAOx10n)!$Cwkg)a6kk~e5fTn$lW5^Xtw9jmLcQdoFG5W{4QByPx?=|x
z4|pIa=s#hNMHZ4x79;V>(eo@3K5(1Q<B8&Io*o{pJvG5YxC~70J*e3geC~)!sJng0
zV7y9&BthCnQD^s;gr=p1`HK$@e28|2drO&Db!jH=^AF(vNaDD-#27=ls^?%qf4*<H
z0Fn&0V=OxdIes=bdB;rL$nkBzaAK3wBrhobU`Zg&Y!*@T(c5hrKWn+*qUPgQcD&S(
zSz}B)Omyq^jr_;3d=;|Zfk4K1UMZHw5aDB$@a!rAEwT<t!++sA)g7_A@58l1_Z)C*
zQ}jvYpz}_L-+$hOwc1*e?HlXlQ9Q^>Q#fi^x=AMGgUh=M5gYQzV`_&nWSNi8+o0Bf
z69t?YMCEwZg};jam;ERWsZTib7fI^ab5<nk)kgMM^8dCT=*km)G{Nn{Y7I~@a@zE5
z3e<KmbJb^c%EDW4N`H~J;UGe|cD25D&lHHhQ_alAu;L}TOJ%Wh4^NlJyfZX=4TfcV
z(X{uTGa=l#xtE{0jMrs+;gI6^>U|R-7%n5c|L8JU>VWC_Uh-%fPA+=M>K588xIBwf
zF7~zmxPfD)KMjOF4j=;niDH&{dxKRa#U~!f$<96Uz+8<(*@y~p&mbXF^sHlfPpZl8
zK-wQuu;e?+x+epi^9a_CIcvr8aQHcG_V!*p+pFOR3rLa?@0(J4d)g;y#K)WpN`Yg;
zC-LAb)_@0$u49^=0CMX-_v(9(9ZWT=1kK^|ygykvs4KFiv=0HL`Zv7dRl}=7F%G`{
zn)7UP(b<ZHBny^e2`R($$moDw$LknkNjX^9JG2P{c?SQ_Dthz9uVit?o*nj}JCXeT
zD}7cXqmPsi4>>Pz_ioQ_;qaBCz~^*~o7@Zr<k&ti=X30iW*L5=5kI)Xg^+2M{&Equ
zFS`3^nRN`H_N$|#r<t+?0QZoLQV$8syOw_xq(*Wrsoz1o+@aCvpSfvMCPQfI89%&9
zH<ull{_D@^H7cy)EsI+7n67?N39E^_+eP2HGiQxnJaQTI5so2t?z2LgtbV&pScQJv
zYEloXpG?SV##3&eCYx%fo1(!;g8?<jU>V_f`AVWiKzmX%J+a~7?ZJM!zo$o-$11(}
zgf}xHk0H5ZYNWS*aw5+x-yUMv_LU#cc%Eb20O|_{RznTiR1maX3Z7-}u$)5QF@Z}b
zZN^b}%)6G^q=zO$A5b|mwo*_x5d}ENc)<d9PFoduHb}|P<~ZTIYc>;V>Gp)xly2^4
zR^qSp_hVcayXh;eLI$3aO<*|A!nHTp|Jx}?Z;tz<hIXZ~v$FJZ9%P#oTDTr3(M50b
z;qAx^N;pS8kVfq8f?$hYp3T%T*?hECm&>h=O`pfpUoymxLDD@dsZo1Td>Y8aQ^bk%
zEUwggXzsiyEM~?bwL1u34i6B^T=$~wmi2(H><SZQ&(8F|<Z3)NY%5Z)fNAtxaajHD
zHP%(&19?b4QXtfUHL7*&QHx|NM>~3rG5`Kq54K=8w<vb>tWtOPPJXDRFPA>&rWpb&
z#1oxafK<h{7RFpimJ7e?>=J$VljoVe%o=iXZdes}@1`-VS^0xjVx?hDGDsX}Nd7Zu
z@IWvCu|7juW+kAa-?$KEC3w1lpJ>l9W`#?}Xzwm$o$^Vuu1`#|JsoN#!LIj14;maN
zu~>8!b(nlhqxeua1w4JV9sz#qt|QH0*W;MM2F)a6T8i%lZ?KbZ<lc4u5hXaqU`p3p
zSe&7rwoQ+TeQsB9ixdz4qse}7_~hs6#T83>ymOnLirbz2rIMKLr??}Y-itmTQDM!~
zJ3+q3N`#vV5o}{o=Zbd>i;P&-n(KImd`F?v>Q46U1hQUGVPma@5#w((X907qKj+h?
zU7FL#!Gu_jd<102uZ^@9dURlX2x3)1c9@$k@jq9U04!@=sBMPgYKX~J8reF(;U!WN
z{P6eb*6co}Jf_7i9sn|!yKExmMfdnPhpp`8{d3+_5xWs)24tNgsCBlm%L<gyrZQV9
zE)S$XET)E`_@qJwTUa~(0g~3RV{mKk?ZEwAHh*~8SE7bc!MPwOyp4Rlbug*EAvW<7
z>>>9v``^4-5|wK5isIQ0w6Y&qlC7UAjW)V|Qn{V6*n3}}lwoNk{;IrEXPE``3IQ54
zh_Dzs>a2f)H558_>rQvDNC(5eH+Nz_D`ENXRJcKcJ`*avpeTG{1afd1!GTS+y&SNB
z+ZQ+M!YN^ODph!VA3cf!?q<1}@tV@MlRu_^pH99VGSh8rFtkr*VWL+#vPz^W4n?^2
zZB2a@>Lfbv!HZtL52)xpsXYxfZhG3_0aAI7uzV+}K<voX(_)LECx`)I4+KR+=Zck(
zyo$d{O1*QywQ4?XHj31ke(NF9i@i?TMf0Ps>%sIR<A(n1)IYQcUv7FQFda$s3vLzm
zwg#~lcWDV?$e0ZE3Ax5?Z#9Q}Q9P2n-`-u^;1Km@>_su86q~Gxd~uvgW6R9DQyXUJ
z$2&NRqCrKui}AZwe#e`Bx)_3-)+vq7J+c65VLxC;sH1+`qdYHfzb?fsU`JcQ`J^<S
zmH49sRdsw8?Ou}W-8y%eTX$fZAm%xBUhI8ifKu;b))Q<$i_UhZCO0SSr3Nsth-Y<_
z%eW_8vTdU?LA)Vz@wWPu#mJ&za<8B6E?Q=pV!Oh0ur3n)RO>X1DnK;2m=ASoXk>M5
zZ7fB3u#~DfIM|Pq8Q5r41Q#)iL$<!vz4ryupcYG4jKdgxdrti}XbV?(3tr%pq$49B
zxN1wnmUENc)?*KTLhgdf5v}~nhD2pbJcnDwZb9%j@^r$JHJ5SN#|S=CY?Z;>Y(naV
z$4D&j5geND#`j9Ao~xr&ZhJn)^JamykL*-&&f7~tm-;v_8}8y|M^7_#Bn>@3A$y*5
z=DEG>na(9*`e5b(v2$lxz27~*?Izl|^3&jF;T80k;eC?WZm;G{my;$$>481Jm5#wq
z4D*L-Qzj-|nSp^OHstC9+J`xi;v{M@Vha7jH+$U}S|mQHKHqpO`vO6eM7g2zzgkQY
ztW(i?xT;myKTaugmpNtTfE?-Dhbq;nRrsQif6hve7lTlQR>-+IvnF(jO2kAbPnwsw
z8NJ!zK*HfmOIXIXXo5hZ#3r0Ndx2VvZ=7=ebeD;+TZ~fH4Ud?0)fH#6bKxz}0$C4>
zQZXDObY}W2f@p%EiF@%_&y{oM8}jG#8-5!!N!X|UYX4Uf1jQ>0fR;f3397#o?DgF{
z<>?CBYE<)$B!P?mxagtjb3|#0AStb@6`UuSR2`!`lfw+6N2@uc?Mxv{k^nqKaD%GH
zPqU|7G<jfMh~&i!(MG}L0%y2wqZnfjHlQxyRGn`MkZ!S58}k3FM*$6iXfw#q0c8=1
zXOzwzcXlSWzt}E@E=PT@tQbod?tVt9n_MY3nb7OfmcQ?3dolqWF5d!h0ju-5Hk#&w
zxx_@WzalA{jC&%))&6V<#tvUiJb$WxPun5mx9hcs#7r+W&w}~}>?Wwerl_I<5GFkh
zMJ!pL42J>-<*_6I8a>h1f+($k(Ov=y)(+9@X38ceq3lY(%R4*Sm7Fdrpl4XE3b83L
zu1mi`?ep8RRp7l{TI+=9NT_~fN&iu8f;n7Y+ze0YTIa_M0U)!+2q0A$oUC<c<k`)v
zX49fA<6A90IcP0ny+yltM-Z%yeHqZbBapV!#l=KdzW>5=QVF=P__fASn$I*YEc3WO
z?OOOG56tY*;~O~^x<uWN?f@%)2uG$!Qg)nzm)v|vY)~oyKqxz{hmeG&RBx%v8isoy
zE<+8XWek>+EZvA9`x3y6-q{cM2bEx!f@{S(p_hP>pWXvx0YvwubB`FM4R_Sx%l0pk
zsJNS$UFj6dIPKuBf;WZXAf2_WmS|yUf3RQFF*d<yx$~K|d3`x{5TsFt<3IU3p=<yG
z{9fOcyK{RX$*gc<tufJaB-PcKZfX3xrpVC?K-LFaOE(TJJFA~qR)Mh0iZW0RHeQ(>
zRxgSg4Zs>savHfk0~AGdW;+g+Tu=FB=hM$k`lIHN!3xWeEK^(RhvWs&sToc6{j~W}
z6Yg0D28YO6k@52wVXFaH33~x?Eg2FM3cOm2NU!N5`9z@5sOx{(d+yVSqP52kdW6VG
zmMbman~b>}$o=mLpC%XFj{mHfRY+qDNvj^fWH}S$gcvY0O|Q7Qbe0!2k5rpPq-uX=
zcV|d+1G>rZ#PNa*G?d&YmF)9oEm04(pTCXk5lO!C+Hq8q<m2!&eyDJh^S)uL@RAcX
z*|iSSof7!G&#DjNfCsW{2x!HYol`%2PMq0obYP{O=;bm_mLq+sKVFb?BaO5O$j=zS
zD`%mq5Z-_-nx4dG?gwl}(J-zl|4!lae>0h7;TTnQepr_!8TlRldm!d|8Ym`Nz+yd9
zr5i~FwCGX5@S_;nBN+wOTV{C+TiIs6;Vd!Tzm(TGo6{R6e#OeCvrRqfM8N<Ex?wHz
zw2E8l1C|@>pF5A!95Q?zP>wl!eC1G5V??{mJD8x((0{9eVgnxn1|60MY!BgVGF#Cd
z!VyGDI^s%7^L0LhdMjU0on*eA7cXL4%^hK{Xv@mbB?m1tjpb#NB=k;iA<0q^25xAh
zw)lu&xY2b9_WXxN6kQ6tO+i0zGgO>1_4rqHUU`iFJq`3i$Ol{8whP!r>{S^jdzU6^
zh04w-O=Bk0Suye1bl;dJ^tt|RU&d}kF<@2TO4emD{1uO`i=aD%SXbG=;KJK@tJK)@
zJoLVS9xl|0TZDFQi2snDMS$~Kacv0PP>p1DJXmAjqO`|Caa4==ICStId+%sp^lYlT
zYq>$Q$IT_sL)&lo;Uq5QkDPHRnKSl9xB{TOg~4~%Pom?8Wcmrz`8DXYpMTU7hP)qC
z!N?EOG$<%vJAkL|V2cI3z@80@(|{wkObPMx>`asTgYBO}xp~#5Db1kX^W|ic9LKxf
zwvqVoMXLQW;;kEkdk}|<rrjdTPz6Q_&zEL5KD`uf2!7&3yX&?g29rM~cug-!=S}v#
z)Wk;jS{kZ`3`W7o{9C(-&zvK+S^*);AwV&v8=XLBB(YL9`{j?`hEV&0dd_H2rc2rF
zx~gaF)x#_BXy|lhXcM7+k`!-YMb$Ps`too)sewdL9rf<UjH8|sf{&MjtEhZYbC%~n
ze-Qz0aRl0FjTrWcAVtoU82UPP-4N1f@1j$9;raB*M7>fv6~^Wc;+XpIQch9F2rfY@
z>o*baS7gAa(?~T|_ge6&n@iuer-3C^wPj$M2Az|mh}Zpgy0b=$^x}&T6O<|OsY%EE
z|5<<~7kCa`fICsg=UZUm`fKqD*o9@_6ni+G+2M)pPNO6=yKhBi1$W27Do~1eRpu7j
z<rY;yolfeM_ppv2fpy%_&t;8XY-7)}vS7LfDCugeRhxZ;;>T-oDn@w(Als#JNF1K;
z8}qrzF_^SNrtGamgtRWQpq#$rS6t)7>FoHLFPC=-6@-SIFN4CG3fufCDl$sBb1o|{
zWXgAVZh8L2d*7P*{tWuh!)=b@=~Iq>(pt`5@46*P(qCJdB4}mn8HsmykEp9@9uifO
z-h(KrzQl>b>CXKGlRTR3;NfxJrp<X4xF|b`p}-@q=JVNO#hZM(N}qcwq3x)qmg;68
zs!N)6G92iR-uhsaiE*uLo{~<q6fihEDp@>(?^sK@8jO6VTJidO?2s~QBXp4y9QSt_
zx^vZwRJRXt`LUp{%cPlbw0|{NT6870HIcXI=%Dncwnm;5SXZO#8a8Gh6<cqORI!9G
z^^<;jr+HpTkXp_^GtuOPGCfy{Ds7G`@GnX_Z66hJOjBsGCoyGw_jh_>q>EUEVO7=#
zRoWr}Ivfw7%p#6Zva*9yWdXh<QsJYmz|kkXpg^YJ+w*hXN-vtsJC(wHbZE^Y?3h>B
zOWlu?gG!&bCsz`oCt0pt85!uxa<?iYRlcRoYsG3-^9Ys2CK?Nt1v=v&TB7ZOYPW_*
zU#q!%dA#_~%^~sRZ$mm7D4xpZko%_+xy_o3c@3*;;#4V8A7<{qVE9i9lwB48(sR%O
zE?muVgy$w4D<w)#b%|ylK|1O}RZ(@{7BBdg<3GWcCM9LD%}<$)w_mg;YxrR9DAcch
zD8pi^`Bb26cNjZ={)uMFEiDMeCUA6p=gIrBoEzRMiyIG1rzpnw_2hqPTiiIJSt{K8
z9jv3+tk^5DlIa`r)z~BUi6khR5%rQMl;j1gXTIGllJz0DaD?mgkySw#;=Yslw=G_&
z>VelHbtot)AjVz^YGzfx)s3PlC#Qx)DihU`Sp@3q@xOn>f$8Vld#f=Xc<3g*GO9|A
zWA9o8*Y}uk_}NX7;T{Rj!bxPa-s{<RFr8k+%@eorA10Rp$7Rg|v)8XJ!XbO?-qW`G
z)&{%ypL|cQ9>q#&m^Z;K{FAm0BBdcU%VvT2mA4+dYyP6Y=BbIZr2|$Iq@<$ga@$r*
zS(kNDL9ZsikI6htnIx>@;rZ~YUtd`1#bYNMc<Jfh42Qw?n^K;?i&^gv(T3>kp3UEN
zVGl4)+=S~Xmf%4X#g(AzNT(b|Oe7&D$EpJUu#%tfB8L^Ta>UT*;<Ty%z;imL?sAv(
zI4@z+t6~6S$oHE+Jz+e9#HwmGqjUJkP{J$w4XMsyA5HtdQ%>uqO8U~ZH9}tl_xyKn
z(B2b~v|%6AinVg0x_jf^L@6Ca{_7l@{L9f8s3N;gY$M`nxHcNASR=M38)~aqMqbed
z<D3I4(pJSdz~>4Hz6A>%Uu!L2yPbh}*WvGaf?SgnMyTH=!Y7Dl3$l!1*e23m3UlM<
zaRZo6;RN&5H>LOgP8d{U2HJK*-Z>>pZSp9LRX&8PhjZWiD)TO5_x5Q>6!ww=8F`*|
zjBX4$W7&Gz;M7-%*gs}`cZPv9_`Xb(9pd*3qAI_Hh7PElUM)NoC>^JHKY!~Wdh*=H
zpxdvxL2T(?)ifBCQ8pvxHi)?>+kDjF@Rp5vjqkitVz~ihr$Els?gpkXMjW>hGs|)Y
z+T{Q@`-8{edCGn{XbWBFO}iTDy?~CB5+2p_NG0#fgq9}jKMr9z>lJDd*YPh5tB_+o
z=k!8ukaC4F(u5rD8v>EneK#7Nvo1?SCjSHEklG)i*<{bT>2(7|-6L-@=K`fxe@c{s
zaOz_xSk=qgFH;iuQR_6RAA^R0YIW<sMh)Vox5s%s8<=<;AKglPs1wU`lUbHMm!<IF
z@`ADOi@f&uCcO5jtQfdCp+_;9@Q_H`lBkqbw#-o8k&OPLE4*0~qFvRA%Xd`hMtA2V
zMcwp&w-t0h9*=u$x0FfFpi~;$Z*(C|ch;lyoUte5qn^ekiHyu^&e>1Fx0n&Y&8Ig@
z(XrkON02R5**>r1gqX;p;?-#zo$BY_Yt!Fs2l4;f5u>beth)YMFR!!-N4FY&X>5s#
z-WWt?A0FN9AC9Z4M+_yM<5d&`U%o{!Kx5xsHUe-yNEn_W8uvS_3miFDDlYw-OTe(0
zhp*ZwOhoQ^<O!_fGq$9WXP3n25|L>;;O21#!zfJQso3JUx9EehBDm&DG!PT$R}C@`
zDZ_=SoK{Ek_G?M{FMo=V(f-fAW2_e9ZA}U3NrYl}8S@D!;Pl!QVKzCqil+px{#?)>
z5s+5pE1ol1p57#xAcqA9(+3J4YTI3;8HKcV+npI_4g7ThptyOZ+G}+aT6zWBzPh}}
z+KOyk+H7QtIE=CLyt2?UCQCi&Ye!SXuU}QZZX4V<2iHM%QgB(qt^(`CWP7}P#{2Yn
zrG)U;<PEm}S-$m?i=|D?G3AX;XPRjFu}++O>Wa~j7~3U7EG4ixgB6|onT(L+4hwKB
zpryG0>TP4YlVt{k+c_h;%nSXyU9Ep(ltiBxvW}}?A7q>-A;sr46i=Bj#h^|-i<gq2
zmyM;Dt+=&^E%*Zn@CgXq=jXpK@IaT3Ph3bqoL`6wyx`-b=pzENq5o%qi<^yuo&W!T
UfK5ij88`q?Q_@tdlD7!^KP>T!p#T5?

literal 0
HcmV?d00001

diff --git a/docs_new/logos/salesforce.png b/docs_new/logos/salesforce.png
new file mode 100644
index 0000000000000000000000000000000000000000..031f942b0f9533d811b31aa3c5d15e3fc9741f23
GIT binary patch
literal 20712
zcmXV1by$?$(_T^<2?3>)Mp~pxq@@Hzx?F*m?rxQmSm_3(yIHzp=~cRxhNZh}zlY!T
zUHq}^v*(=W)Xdy-&rJA7H3dSv7kD5Lh)_vURucq52mVE~z`+JSs3jO)0w0)`(yG!R
zP!$CK&I}9q{t~RHsR{ylvw%Q>KR}=x;8EZf2;{~K0_~WBK%&VY5T#>!lZH6(1NIkX
z1z8a4@$YwQK^zFAj#849{^U8gkMQs{oJ>I-!yC2U=GZGA$i?E%P^A&|)&BJ%Rzshu
zR(fU2cB|~(o&U7uYaDa#J4I&91sqI_KP+cN`9jnxPBH8Ddevo;sm}02;^&E9ly%1b
zd&_^*X2R31xcBeBO_9sR(*+l}7g1PPSbrdx{2M29z4c{Gi$>4B?zd4YLwWS?9kh`G
z--9}-u+dC$(f+{iTJoNSvQ71M)Pn3x>%YvPt4SnsC0~7xoDCc?>fKx6!fOf(BLyiX
zTx-9&v*0Ps2(h{7t0YdA=!`~BC-;wyqLM_vB}s+^);oTwa9Qb*7l?gqzg;F*0$UXg
zH5>k;OVrIvcvP@m;1)#O+48fh!pt*#(%gX`-e9drLc$ChyQcN$i*})z%KX8{8b}+r
zvQ$X77iUTbJCq#7TC<+$XJGgM!d#w8b-F^N=(@b-m54e#id$jiB>u9%{3+T;JBo)P
z8`Ct<?WUw}89yE4Lfo_uWS3P(<wGy*!Dq7<K?Qa*4GUvK>z6z&c`rPg)aj2QtZ8>*
z<<2~H&7o$CR>L~+n(px>Z<fBLvI`&7S6Wc)8UAE#Cc|>!>3ay)Ggr6y^iEb5_gx6$
zuYPw|@fu&-VUUFJ@nH<4;~^yE3Y|M^rrh!I6^bT8t#mI9ozaerN>hphnR+cs8|ebh
z;KlqefjPcx6}P)DEo%dT{lA%~qmQw=g94Z~Zv~d9YE)OAl4zj+4cOERJxQ%FyTYvu
zMUN$&uiDgi&RF7eQAz8HVVZhjZzS|BkoJi?XE{4oJtp`=XS^w*lL{ON`h_aWvIQ9u
z5b7>ugzIA@pHm60R>!>iA`t{t9Zo0J{w)^|`r0O^tzb5C-r<iY->yq+sM=urieL5p
z%9i|?KGZ^764j>L%)?8V42nO!mSBjj!1g5Gb9P+pbz5Sta0@2I0FEkU<?$5u-E>E_
zdB?h%seIULI5M@_+!G4zN4uf)cn2${(IBFS1zO*L2#lesO})rl0zTBKcMs#^;ujj;
zgl2a@0#PcR8L_SqteP;{GpWY^R;Pk!acy)z)$oL9Ri@+7$XaM8xq7g;!GA5Ao#<zc
z;RJ<|20qky?vpeem;@@}$?MOD@S%E3s{&|03Sa^o<|JT(ew&cNoTp&^uYNvgC96Z~
zKYYoqUPc4A<OnvjEuZ=_PA0lLtUlvXDvNkZ#pDm%uMOfRmz-BpzgM+YnC+>Uu7m!Y
z{=7`Wdx{J3({+)OjR&1YU2Bg!4cm2iVliro;h)O99`aAqUfM|{#ybjb52-?7@9_uD
zdyvh@QMr)S43)SxJie+m&6xP<>x?IcxL4%A2rL_kg)Yh>v#1>nMvQwJ3BWucyot&z
z#~=mXyC6Mt;u7ClU}>3pE%xYw{$f)!xu#zeMocX<^e$vKY2zmO9rqJV)i7uT%nKI#
zQc-!{dx&BF@~&6em>p?y0tMiGhTpoK{U2~kTYnCo*zm7%J2{OUGcA<xZsG_S)%#)I
zeRCYL)0ch;2*;(U#YdD-B0akrMpx_A7nd=2lb~Q5B5WkGqsqG2xTNz)#v%dnM*$-n
z^PQxPzc!zw%ag=7@<_A|8Q)DBZcc+3N!aR6xNw_IP>ov~3xY%URAMu}weTY#>Opdo
z&l<#knDJnN-&yz7A^GZ8WBl*Qh3tlzCm#HGb+Xa5{Ptg!o*oWAZ`S|wlo^CE*(LA<
zobiO0K#Lc<mMytLOV8b){q6!wfS>OUZ;whqZcsKb>kV{IG+NLpSSB7sP&nFtW4kQ)
zQm6B2$CrIbvet1s`)yQ2iNdc+A+zCaynuEu{43Ff4F`5|V6iWCKk`$99prXzHzF+i
zvvjrGt+IIN4L<SX+XUbfX#Tx5q5=CWb+v{xJ|lS?{;M6*#0|Iaj;<^(5fP1*jP(g}
z@14bjhCiM3Zk3|a^7UUUq@QE^hm8hvyx^}nL5OXx(g*d(>+hlg+v2KF7q*IPYW4Z`
zgG`}zYtbGP9b^`i<@+ffzD@Z@Hd#i{>{pG+cJRo#C}u8o!B9pdt&dp+Yq*|`ZtaD0
zWjiMNPaAnd_QOA1$8OwoOTSHl8xeVMSJM7{!()pi%ynu1Q^Ls3@fv6{x&JGDkWVHZ
zt|FAmQ)UUj<aO0T;i!rCmT%wWU7KYF3OFB&B2iRi-4kJDJ>5J7K|ttE5AJM6F<kJy
zH+fpk8h->&O6?i1>iNBqT(ar-ip86=5!9$~iq1q$klwr;t-qt5<Mt3+`nKap)gozW
z%xzUq;E9br_8p$V=f)I8gh#_wNAmkuA3$T;3KHje?}}0lW7(oNI5*=_HF097l0X~s
z;?i?o7FW8O(0VrDC|)8K+N$DpZ#8+zCKtw@yx>!@bicmMx}J|lFa<z1NQ2pi^lrt&
zFSM?2a#^rs-YB_C;Rnp!gex3sB6<7@pV}N+`fDFsilTnj)fb=Av(;;C;@tfyxZZ8*
z#0&Wk^Q94I@dIHZs=y+VJpBWNgw6*KWM|`mW>Az&AnZVbli6|2xJ!-)=+%mhfJvRI
zO(KD++~|9(WQ~sCpS^+SL_y0=AJ754-tP8bbJ^F#gW+ZUg+yszd$LWQ?U2fX$nG;Q
zQ-=8xO(0<WN^gbS9vH@WN=3q;Q@4#y=F#7PozT3A8s-VIveqZrR)BpZPlk(^WP^c)
zGgD}1;P@(xyk@Xkwuzr!DCAXa8q5>{)tA$JBh!sGOHS369qTv$l!!7mdm05gn-#)+
zJWScKzvrCAq96YEELd)`!1THxRZDg()*Rm-6=W#v-TL^_*bQhxFw?igZ?rj^<S}p6
z&=;AeJ%ssTwd_=}L~l?e@5GN>fVsw^)f?Q7C;rWS_YdctCZ{1kVY(1nR;s4#8!%Ic
zZ?K5D^4LiYmluzsmNff=Cwm_|mKeFsht72~gdqWK9bu`~k2O)Tc6G6tsc5?1w=1bs
z(Pgk1@1sdDLNsi;#=@8OG_nNa9qzkW(!ia9z?{iQhFEE6ki&}vu;AH{L;J>e^lyKP
z8r&HYov5_XA~9YE#`s;O<#g!(eo=^C%;!#cwU}3SA;T@nn=d1S8z>Y}4DWP?q<)lr
z<3Qdasb+p1P6fU_yIg2;#%WuAMb%84=ojP{bZq6h4?*YPb$?#27Y#Pa;X(><)6Eut
zm&5I_L;wb6wF8fHZQ=(*?4P8ip(4&(BWH)m&ye05NKb>pa%wl9%IGq6E7cKpCAnvi
z>>d_(YJ)ax<efT*V#*NvbDWhdQ(WE3+<l6{u{31K%wHRaUSTD5oG<-7jMpFP%y&U(
zNY{75>$OqAH_qfrMVO1ZoFCQrBfBJM#afbN;E3ahfB3anz5xWCqMWC*2i~>umH@`@
zC?r|ZeZ)4kC?|7A@`x2MUeA*SVhJq4W=smVWto(@yK4$Lo6B~^{E(id8CdDKn9u6r
zXI}|*={k~|aRj6WhuJl-nCD3Y4J6q*==Cw;iywR0vcx9$^sZ4$p4LyjpeUEX26F_}
z1fmBI#(9_OMS@XXdncFAqHUEx6rWM7jv1+{vem<fMV#%nNCW`5v~Vv-P1S?*DNFt6
zV2vq1d?;Nw81NMHSwN2v=2H_kRhb1@3};-&oF0|STSrNZ%SL~bS?s&#(+|;&&(fYn
zt0;pgCWpJ`<gDM2kZk{9b7Y;m`zB(3$d1-ej$g8rof8%u7N+Q+Tk3}cF3+#N<6;JF
zBd+NQ#q+?QWo2ViaXKWfh9?bmp<jOcd9li*ZW)NG$;svt+kb0U>EgCDJO5%|_FU+1
zn{VpUeFaa@74}RcegZJ!QkdCZmgB09696^=$?^F~nw0}cj;!X(K!Kz<9N&&_sSbQ_
z(b+rm@!KIin@V39SZ-KY?q8h9*6^QoJbdleOSbp1+xIx8fc@|El7nT3g`LF-_dSqw
zNqxOIqKE?J9<2Gv!7_ozj8apE(NQhk**oC3=)drQZ7wTA6O)H!0RzgwG4C=f4?kE$
zTWO*H)uEW638~`Q@+2dPd3!2ZE|Jyv%)#;`NpVR5)F+`1O9p!U84%uNrkF5{u4B<^
zm%Z^n3ZCQ$!WA8rm1T<Z`4>QXKC(#-hO(n6h8WoNt<;pbeGLy&#CFZw<=v#QfcQzo
zNxFGUq8~XG6S0wy3=p^zO!;NXX}bVmMKYOnH;|MhhJdsy^_sJ=F_U>J371~@o(^r{
z%scSoHqlERSX@}x*WaPMe=72QR(_vRqv0g&RXpz7n;EN_eocD=DSg~`9=a~c(X_ze
zj^m0wCKku3e(@kr^W;Bk^0Knk9c+e3t4$qM9rR@@%y+nfz<5#8!kf6j%1bbR^VgTZ
zz=-^N%)r2)$3zUKH`SO7{{V7_A(sqb;Xu<rhNX>JST^vF=#?B(2PI%xV8l;EeXVI&
zlU5pg+^wV_aQ@EKQ<4}?iVjE0h~j8X^dr3wAYgCigR3^tHY+3)aY2<X%5wV;tPBho
z(yyX#dGW3D+0Yu?5I8S$!S7{dvBO9s{c77^W9O@*FV{R1rI3~V8u(mJ++Qj}5J|Xz
z2{I^O18_VxP4~moM*hiWoX2**F&`hX3>DQ*8Vd1}fpngM<Yd3TyX6K9>=R>j5VN%p
z3CZ)B;VfO<)}@XQAc8z=PvF6FR|n~)&NJe4%#Js@``io+U6?c!2c;1Hm-3)99AL=Y
z94+zhy^}fwk(i*9$z@<$#PnRa@TXO%ad6*7>l?_*u8-emyNI2Z+~j_+7i9)X*VO?|
zo25!8pVZJI=5KJe%)5O~28I(X6y#dlWN;L7x%?SgM`4tltZZAy@=?-PDLhAV9TJk5
zS~wcZmtja;nszRR4Pk^e=?ko<BqXzC!sQZSkCzaR6UavqR%8csbRU9%ABM%#;@iYB
z=+M~EfOgvpuRcu-4?r<r7P~(KzJA+4E<me!BS=BxrJo-k6gPjR3hzkJiwDh!))fPe
zCF>HnbMAOH2Bgaj0+vO$5qK-5_Xgc`XEGD`YTT^a*0m~d`_&)a^e+x@NfbT2)piXN
z2yr_s>hgfELb{`7X6ODX8K8zwCCtF2n!Ii!8%1l#K=j(y3cz~oi80bU?~48Sh0y+Y
z%M>B<Xf&sY1JuVkCf=#a!u)4AMO=^lnB^HHGb30^hgwKLV9vhsPt-rd!dwkNS%+eX
zYc$p}$u@z*LPDUWw=1iklE(#(<JjIfj}`{d>P+#K6ACLO1w>Wyc>cMt`@(DQ5EhX9
z<F!_V({`-x-1GCGC2P0g&iuY01GeE6rFsrX?Z(gf^qamq(4zpsM(H@-w?Lo*^3*5;
zuH=|ZB7^fBe7FO%!}@<?dUS*MfiZBBxVr!T?t2`=3I!insBJ2&NxB+8kb%F&rT8Qf
z%2!Z=C5N3E(<m7d;v2s%9KU(RhWFw{PGVw%va0g8NmV}m)z3ks+lEB1P}?1LDRCfI
zPpXgTynzIxPRRn?kY5g$^I9!65`B!+bL6F*S~RX#&U)Mh5F&o?e?)mA8N@1mTrI6!
zgQo@zociN~^Yhtw+5Oy&NVh~#a_!x4fPt3$Leb|nLK>-L{T}$`u50bjiywK*(dRve
zh<o$zD`xakDW4TLu6&*EC7Q`4pjA<1Su@3^Kle7Pm4xf#Q)JRS&LL|0{_FUClcsuq
zpwz-$ZRI9$Yy4M{UlIu0`1rCQqc1v{eJIWvPM=&ML&CnXj}4f|DbD|EzExnglY*|8
zSc}E%t@0lw$vkm6{qvn2=j`Bw%!MnU!dm;IoG5r-DJ2Qia6wm>u6V)QQO5t_YP07q
zw`B=T%{U=XidU+LR2yx<G_~&?>aecMP;_qq&{z!&$-IA0QSlGa9`@wfu!#g;4QXJs
zPe}8GJ41>?-!B^fYITjV#fn!uCmn$7F=Prq#J%A`b0iJW84CX@h+!D;(W#!rv57+R
zzp&$;d>n}Ff8f~*Hn;r)R+PVvDu*3z88*prA6->R`+32+Wv0LZEybsbj>@DLADQ_5
z8ab1`E4zrdy8GFDLm4gh?`w1Ig*BO6T{>a;=O=9*t+R|gP;Q4jPZ2`@>vZcFF{j4F
ze@$>_E9D@AQm%*Apk&gn*FsgB_wLZ9-u5=O76)^s-m#9+<|LbFXL{iV+BjgHRjM;#
z(26k(_jJ8WsxUy1BZmv1)!MYpiYDf3{J>I@4XQBy#=eUoaFv!zGKh_<bX=@)%kt{u
zVNkLV;XEY;fo$wGcf-FX8n&<ORjqKFc%Yg-FKS8z#7J5Y4gbaLAOtnN5BvIyOd&)<
z`&LUpDkk1FS1Z|4%H@6j<!b;D_g-B&ls3Kv$37!LTR!^J$!!9h8N^DyUGeogpZ~qE
zsrC&7#91st4D$^WNGds{91tH5b!|O6+-PDW)R4HPRKgYF6KGWc*_-f3CcAJ<WoA#2
z4ycGoc0YV?rbF{Ii#T|Y)9gzw>S*0g0deZOpleu?gqT^g#|+oOf4_W1oQE>5fUxT_
z6n#OkZwsLx)t=WtZ(ye73=tu>tdv+YIlS+bqE83tBg}o8%?&2Z4VsmWnun_Q{~;(O
zw{6g19}|%|oJ;WK-!f($uXs#xZVAeqlx2qVUd<gZB}DNF)>@!M_kV*6`!D(1I@B|l
zKGL}IV(YIa_7x5(L^(~pg@rfXoBV`|FTo|hf6Ef-1L?1!JR<EPno4I#z;?>w=2<pU
z#2zTn5VXME-oYxmyQd43CqiV_{-@*cFO`c`xPf3jv+g8$0bjlXY4a)-Rot%S`#)LL
z;!?qAp}*iy9bfL~&yglhHJB?Kd^<0r4baq{eznIR$#cYC-l5(<?KF!1wwF^Zo10#s
zuh2f4f4M|~`t=wG8gNNQn|7E@i3UTW*IG^tM`_45`^`iM@UL5#%{z+ws46BEyu57U
zeWsz+ZN#E;fnec=xBVL3?`inn_UC4Uz$bTz;(u*rh>hSmce;71m4no&(!%~tA;;B1
zUe;)W<2<uh`Q%R=o5GYP>RL1@H^QSx5+`a;q-8kekUZ&cUk^{eb{28gB{KI~R9~4>
z{5HLl-ST;9^nRD21rd?w;g7MjkQ3O|QI6quQuwQ2?V;sF+;IL{d$z99(M=fweV|R*
zU0bMXq~T`>Wef?CRQe@a(<ac#pT;QOSd`&2UB}`V-F1a7MXB1~Z8>be`bAF@0UgSV
z>3%Oc$~Cl+ydt8)P4qDn$vVAsEE&a@RJ<vyJ>n#38&>0LBc|?PUr<DeJM!xnq#_>a
zd>8gqfndn&W2MifCAB`;<1|a$_%dfE`gYwkV*fn}^$p?g`ypa48<rEYQ5hQgMT^U6
zD58!8Xfe7}VxX7Q^+VWC9;DKXtYE#jbUG&>y<;2?y>qz_;b5V3-p;&^96mq4xi`e;
znUKHoaia!wnwbv|_dBz!m5G0}w0%m9-C2g5v9HWl7VFCHe7(RG(@rKe4kS9NyG>th
zuN-#6nAoG}$z87wjIn(EzDP25wc6C{?94T~dy!pM*yQopzoEeZHUin~^NhEV?N|Ke
z-S1a@z|0!#R5?o0nmqW#-CUDHqo2oCyVy?QpD#f3drx<XsZ+Zz3&@=Zw|HAc)(zGY
znFY8En_I$(spI0-?{7N3v^DFP6;w19<R|b*8C(vavObY7!?f^4D#Vk1uifJ#jK&AO
zvb^alh&XXiz83|jONy#=C@l(r!8YH-iVdzFsKvGhKy!tL^h;Tq{iVlV2sL@hfG9s}
zmmes-ss0)aJGu`p%4ka|V+-w!Ai(j1ZBP*WQHkD=&XcB~H}1*|4$5sU05A8Oy^5xx
zSe?6%Xjr*tDL45yQ3k`zarV1I#=qxI{e`sRa#z1nfywhABS$tWIMS{mLhfff_Ke!T
z*yYKF{g8rZ7ds3mR-y5mj)(t5#H-l5<do}~&kwRW;qb2RCoU6}>zS|PexeW07bKVD
zKXaK+wI~PX_W}~$5yTu651qZGQ_y{Va=+kt^iqwQR&DTb<eYjl5{~S0x*j%}98Pf9
z+tPmjaI9#k^Gz&t<Go~ei{H<y#rQ5d<~Kw2uqMA!grg&$W%R&)Tyr;I;2-SZ=>B(~
zW$H<<f5DI9tb*@uje~#qwW7X82d&Bf5C*Q9Sh0S?+2p=0HCqB3K6Vt^V(5eCVq4FR
zuC5zbV!O13$HUTMP9UZFi^Pdh)F~bY5n!BqQeEVo2QslT{Wrfkd9?&=;sC4v3ot@R
z^?Zl@7<fqm0r<VNcTysi&N8wwF64L@BO-3<I$+L;@qLTeoMW_#rlsFr*we#wN%3u<
z7g_e;@wKnxCXT<y(MxsOk%?8)!8uN|R%XyrdNRD4U!JM#6!o@!+Y{vhr|Jx6?Dl)e
zAmnA(Tbm^db>3K%*uCRAs(*c$!$Nte(Q(2b;_{i^?SZ~Iz{sXv6m|dPiPvfE%AKjT
zhNdVAeycA>SCJ)}&fbfL9db3JUXEbDZxt#Z{!_Ne&8uVKjl%&$p2D`j$BK$irASMM
z?h>(GoXrifY!u|KEltvH@Us$u4O3arLWxIK60}12!HQ~G51C^$p)d%2mmb*J=Ug(B
zXkg_bC-ft^imCyC{qy4R#O)UO?LO{sCvs4OYy0d9GEV7<p7)p#(W<nyJHTO&C*`&G
z^l=lB$lEm-@p((sNqI|c9b!BugC=XGf*<=7x&2+!8PQ-DbvXz@CT^VEDRswQ%#_r`
z<N#;0S>!I)r_3auD~Z-%1?HCPybY@1m+)te!gCpl>}1b+cf2g>zzQhDo5PeTKuRuY
zw0Ep<bzcMayoU@%ocEShAk_CS--6dCCCH&XEBV#+bd^m1X@OugFkvIL|Ax=0`zUj{
z(R8N<c%k5BDC38#dI@~Z3ol=6d*T0n6^L+1xXt`cH(rW)7?Dw+b|nB_FgVS+UclsD
z7lR40J9GC-m5_umD<rbGEoxf+{6++!5&MCm{Ozj%j6lr}-En+};r%~_&dlWPDh@o+
z1?Mf(o16fHhB*WDN>4Zb`^NWnOP$gWHiuocRy7zOzAbp1J$2Sx_)i+<cuT^JX!GIg
zC<{&St&V3;peWn+CS6<Xl};xQalVd5u!7pfJfE7d!*`R`c2fN!t}f`Xa&ro88f%*J
zhSv5j7(WQRxeA+S?H+2EWhTNi1x1Is_&;cRr~CQ(wO?+%%G7Z*XBuL~eYXUeTm;C)
z|IRc_QMO?SRGQ(kSAM`moj8=QYphH|u#QDhS6L|<(kT*R_KMzz7_aj1_wND%4;h<p
ztb%o3H?6w{Wa5n?UB`?q5<>Qojkf=(RyJ};+c8PYOlZ=$_7hVLp5E_iP;FiC_oLD^
z`{^ZZ4=(Q!5#NbCXtn3c{+La7QJ-skek-dt9c!xG!#r(C1(>al*q7s>>z-;Ld~~7+
zA|A7wc0nKG@4h$5yb1&($DZ8X016kc$;;T;#N=|SU!%eojbEo?d4?ri$<+UpnDXs(
z*XpE^6#rrkZym&Qv2f`eFHUdm#M>3w&bXsFPHG7#*v$rW$G>K}GYWET?emP>?Gg>i
zLeRQXeIxIE?Th{poO2@;>)Yd6hEcBFS#UA*hdHofE<0Q(l+}@U{f5xXzdKo6s{8#Y
zH$7x=(@gEXTw7LeXp-x0Dp`GHtzg$+zA#`2C!me%PeiTLY2Gk}-eW(JfXqaXo*x5p
zD14bBS{)l#jJXVOkUiND)9Rd4V$l%gi!;Gb?h4xYOMZS&(&j}ELHF1fuZf)3D``gQ
zr~&x#$B6E#zRYU-Gf4ziiN1RZh{+aK(`hNF;OS%&zcHU<L~)6GO#q9+(GZb@xD2(Z
zJ{>%#ZWG|k%Hnd^nijlWd-WXvV(!T>!=Wrm+s$c?R8g)}en98=XyG;ZY6caEynK>~
z+32c?*6!CnFpfOB@%7rCa&J2y6I`7CcaGTzz^X?@P+Ij$?Rnj<(M=|T^iN;+PH?DD
z?Q!e0M`bgtSLbm2{A(Gjkeug2H_N;EIU~CHtKJ?DmCP(DHTs1zqA$lg7WOxV!P7hQ
z9wP54TnS8G(5;I3Y`XL%S0Zh?mVon~`<@mD6`zbKi@~gzkn?JP)dzJ~o~R=r7K<F*
zSNsZCmuoK}up-gv#r7g&<c)jMbxEq8=J=q&xrS&;qI3tJ%IlMw!{09zh~X<gV$7!M
zdYV<@CH-^P+|Eh_?5{U`qV68Pv+Rf}d9nan;a~0h*D$0{lYa)kU=n~+yiAR#^<Z<I
zS#(%C)Q<s}f_suk5e9=*_me_ilqht!4ldNT)bp&u?(p9s+klg>r7gFP3is78e+T=e
z!xuTBpCiBxNgJh>TNy1m6PwRjxYJOsh`6K?d>bkj&1M!EivE`4#TS*vfVhIo{RmOe
z;|6S#g?M}KhfHBwO0*_S@3t-t(CU(|0}9vk1xaxf6V)Xos<uZAwb$J(DVQl#JVbVy
zGD4!}?`#7R&G-d(1Q6I#N<17F077@=jXd_?&Yex;3`S-VA~%!gh>cx~_K)yY`_f%h
zix`B{+@i6*d0Jc5Y19W=ZV(KKrYBhiv>(-u6WbNcLBLbZ-9g;*X_D&BY|i%rA%&~+
zpDYh`ZDp`?mT%E3BjYV_Zx618BY#HJy&5=6M=4$GRhEIJr}Ez7JjvQ{LA`rPNH2R%
z*WG<Tsd^l3k{zUaNhZyyrp)_y9fZGN^+$APpyc+Q3Gvft?t%PK`sL&mBjTZQ2i2o{
zA&;{wkBR!`g_i1+B4XRsvjXVJ*=F=lAC4wHt8xMRMt?anb%$uj!8n82!_JYrmF|3F
zodcgLXa<<B<h<6xy&^jh*?ib`!~rQ`c>(5TmDHFo1Kx0c!&V^Pn6T+NtrPa=75!5B
z%()y&x%<75`$KuKLrZNL%p*PK$42M5UHX|U%&E##72m-=!r4PrxdPk<f9XJ0n0UD5
zrI7*X_DnGpF*TUx;V98wN5a=v>jxu!(YbHi-SkA4q~7P7?SDla#`B|mkjWFcvFpQM
zjgNyi$fd-}^!wq@ZI?IEl81LLz9CurNV`v80JnCHzGB}Utt7ziKk%l@d8WKp0U_SS
zK<t=5QaFo%kQEhX^u_PT2m5YkYiO^X12OB!{jR>1de_I_SakYJeZX5T3S%idT0C9|
zZ&vBkE&ffM*l{gMT{XP5&ncL`n66HJ#6un3w#N8|N4~8EW@koTmOk-=lsgiw?o(Ig
zVr8vQ+%kAu>$v9j@l7nV0xs~+X^#F{sE0=;MOf&pCK@NGJ+#4a^MTD-u-x%ZKl7iV
zS}TE1`ZQq-U#3YKgg<rTDv;8KMbXi}a&x;xeJ@&+EU;O*L+)q^5=i<OW)}Q>+}KB_
zx~R-@|E51gu-q3MiiF3zm8Ok*;J&jthj$J$;d?#bL5vCV7AqmDi)Xbos}6rZu$Mb+
zGdZ{jBz~$t`g^lzwER#7O$Zfoy{Ee+CUVz%Yn>a4Zz_7&&O0fCeyRa3h2>B_x-QRm
z_YY9&^n&NYRkt&Mv%`5T{Kom1&dcwO%{_3pkGA^OUAru!arvj#9!LTn3d3)w>hcjH
zKHaxbEwe*oYeFqcTiY{$^=wdAaK8z~JyS)mtlWGbdF>B%_x5@y?-po#8Ic6>9fV#)
z({46tCT87ul0TtUJNZ~Pf-dFN%*Wn-N_ptrEG7nxNO*4i`A4oxog5D%Xvn7uW7%t?
zG4r$O+g`(mOGJLIZ$VZIV%chY4zMJl%w9b<s#@twrT?kS-Hv1|??=098GNH}OK8Gv
z@p*<NPAdkuOm#~CP5%fsX`Md5{oxzJ7-FvH^ir8V62w`)NWZ5#yRhxmXd_a=2_MI7
zw5_Zz+Ql^eYQF<^n%k-@5I_NfbEFX~=;_ra*d<RaCQK*!c4ilB3+W0(>g^P*^V3qE
zl#>N!O0@|OtzAC&xY>*){w}wTJUzm8=(UMOa8I`{#n*D#0;U`0!wL2s5sOk<!m8po
zPGVftn2fE3yH&tobpU{@#_VA7!CX$PV|4O*+Rq%})%Y7whY&iUukV=9j0IT-FALhf
zj8!qVWK>>GFdfe3+pILsm5YVnV|!$RCgfX;769iZw)P%7*8IJ6$2d;k39SUcvtJYN
ziyKECs-L**`b>NKwDS<E8>hYP6-`7i4;?eU&zzbPg4xwyJXPEfxrNjX*i45M4$afa
z=QN300`y_!!#Xoa-9XCU1DE&ByjY8egrs`9A-U=Gc6Xz%v-F^HW4<Q1c6YB)>I1qD
zLYP$#t}S(8;WrI~U7-gQ)M2#HfdM+9K#gk9f%rvF;li2c-nv^rp!vLy;EuDY(%*3l
z-VvvvxU^UUqT9~%vdn{?OKKVw{juALO~s@k`Q|k$NXA~qBo{kkl7F3F<+2hf&1B|f
z;o>}hPDsoVplNhf4A;9Iz@52u_^jpktdFJ-bkvjY&VAnCU(dF6H2)pj>AaQ!Fq`30
zzB3Evof)6-ghN^*F!rQ``_Y)uF4zco*<gQ!mwErQZ1tT!8E~ARy1?Y1UMsZ=uPN^}
zw^L3Z`#60KE>`xUctSfU!6~sYb--fBat~{1G3nro96i^VY}uWbY-In|YY;?~6*R<3
z={S&i%~@o>adVj_ZuIwN@DnlaJFVfo{6<TstWBrz6@UWDrQ9fT@X(ti-*qM}aIiG1
z(Ak~yxfe^CVyWl4y!2$<J$F*5R^i6;^!|Fe7)@oTWIrvSYGe|pfOKQpuPSB7S(c59
zdt|q0Yyx^f$6I}Z){MCDUTF|EQW$V;NS2nG!;nx}T}?zELVzpdbJ<Rfg^**<-={Tw
z$diq##!UAVj@cA~NRA^HMXMVr{4`CPqY1C6I#%Hd&PMgpi!<HR_L>HV|5T6G^KoEN
zN0RY0js%pEL%TEBO|SZZ<5<u`OqJxh`w5^Q9YpIwl5*A9FWejTt8yILE`QSUsT%~S
zvlB5fqj^e&?<1*_cld(5&24->K+;;9HR4{*B1@B$2C>-)sCMRSn9(ZtTh&28Gy}YE
z*U^!Dp$2<8;dUr^T!InNd|xApp+HRRbawQ#UF_>qxNWqfx+*nFcm9JNJ<smR;lkm)
z!Z_Y-9k2T9$kVHP;5aMS3&;GgBjt=fy}I45mIuN~I~pmY19E7a`&z?*f2;?U7?cpO
zlX%TEspGLV<XwgSdw8uS^%>!@<h4J5E;6BSXw#C?i?j!5_-J07czOF)R|$09kV8Is
zx3bVmzAac{S$~F=!F~C;|2O7xTwEnmp}1T1-E)lLDEhbO-fP6|+iags3%0+{crLPK
zikq?3^lf;FN;lhNNlP1xxCsYj@HLo-qCUYvR4Hi>*Z-OQJ}nflXYLxIKleYGu9cR4
z@L6J56m!8m%n<p;amoGFv%z6i>*>+OR-O<Cs09i@oi-(qV6FEqNIoRaB~yUuSmZ8Q
z+@Z=~oJRfnn)l80K%k3Ynl)OoF$)NPDl?|6*T0F>g$pTZ04rvakh*v-Y~dLu7n)Rg
z+@f)J61bXcl>6x9p0Ia|OWh7(%e>zX$i|jn^44jAN)*MBU%tk3+<)jTmGEFvj(&Y^
zSUv4iy|VvL+2I3-c=UJ*I*=}WX*EJ8wjP_lPV+sgYTXagDeW*FzyEJyX@h7au*}0H
zy)<<gF?RaDU^UfcJ%Q=2{we?jC>G}u7nDSMr9)B4n*9KYZ~_nmJ_2c1Cj(f+Qnrrm
z!(E5!D+E_<BXrz*fH90L*a0{!=Oi!|O}CeMV~g93<AGSyTdSP+#;jYd^*Kmn<TSWZ
zWJ(D4T^#IEbxI^@c<$C19zlEm3-q-|gc_cWSJ5|dE{=wyfO^7a88}a><k-V#a5eH%
zKDHJAzZpw_<qB@`*~C$!n!WF00NiTlZt3s57?noO)mvXk-MA>lP2kKt4(i?-6d36u
zEJ+)dk;^{^uv#TQ=X)pru5T0gSAr~*hIp8jg|{{8)_aTpi*q7i;Z&N{%5Ban%A*%;
zQW3UH`=VFJybOh)6fVb5!8ZA3jRdn^)w%%2A+oQq-UzthTJ6zT$#ris_wuKcbjL|*
zZOV}H-zvM?m=TvcwIn$BhrFuf>jNlYz{mVAvXP0C9E<SUKUBQk%rUxq%TNk@;+x6S
zS4061%o8<WT;!=P-!B|^=(za|2td!Clb2Gf>E<<|5Y5{#fD5_48T0lQ^<JE24n%_w
zRlfIgwe@b*ifS5|CE!eHVIeaf%wWAwS3T958mqEqi?BTe!fIt^8Y3o(^810Dr}BRB
zaajw4ZQsA&-LCrr^rx~u-&0rzKxfFBb*#*JG?$O~bxZeVUNk+-kfkv!XiN^~P|u6Q
zMFl&;4C&iNDw8YX3#RS$V%|!r>j?A~9dyc;M3vz_x=ov{;~nO4Cn*x#QblBRmJoOj
z+ke8ylqUPb0Q-*di)w)2@}7k*O6vO4tIoW}NzU~jjnAZ{qpoY97i?34dHT&o)a}>y
z)c|=q|Kc!BSLRN6X|fOeFBcan%ts^UuSPlMMF9T;86PZSCOf0$HBo-&i(*(rknFo6
zo2<TnOW$xRsGU@GdUOEB=t{5a#Kdyv6mKLbPxU)(ABruw^_#fXqN{+}`X5<fNcoop
zffarL7-D!E+(n?=C0oo7TR&-S#RHrs#Jq1l=h9)(n(pGz`8B|pgC*9}{IvZ=m(~zY
z0V=ttq+&cj(OK-^vo?tWj_31)iaghQthY%%V%ZMf5`k4xTI($UT~ASAwX#>R%hu{5
zQ6N>9&|YqPFuBQueKmg8nL#9D+@g?K@CW4qRY*BLN~Cjpkhor8-+w7#4-;odpWWaf
zrJ$Ri=@T-U>+5e8blmDTXt+Rx7(3XXjr&E2+ijiB_S85ig91dakFzT*5{QX?nhou(
zRol%9&$$-mGvv>^H;s(v1;*jCwJS}#%}JmiA->efzx&-x47yJK-bCY{hXKP5gX<n0
zM8SGD6a$~=w8F*};xNVUW~-c)z+b$oR4}%G(9_2VLsC3~o25KBsUCXzU5I+-v`tmT
z&y-B39RPLhr|_ZU=j98{m1(%|rs1eHC^%Zlq4ab{LxuYwP0f}E@@#i1&kVh?h-<})
zsJ8)KN=CDV6K>-dyaHq<b*I)Xk_9r?U1Jn2IG8NeNwkV|nOE3b?nzLKAX7w*w=D+#
zp29W`1VxUjF;@I+=DjEQzT#DEXt9!G_fPqy0>{6WYEHO!-@%fPvcdP5taaaQd=N2j
zMCcC+PRhWp%w$TZYvy$JjuIE#JTSXm(3g+>baGd&6Dk)+2{BZ7RP7K?E3{v;KC`~j
zdqaekeNNrN3W}NH0Fn(CVmUVT4~*sq&jRcKpuao2>EtzWHnz-d8Ki#U>2!k(ftE(^
z?&LQVyyw04??z)5(F;g#)tnw3w%kpAII@5bXeijWc%U28p>fVH-lje`t|!S9WK*ox
zNqmM$b8+++O|Ru}E7{t?;q;<0nksSbv}?DDuZs5<NrL3~0ydE7w8;eoXP$LA7k!zk
zKPa?b*@9V1s7~Mtha5MvEo$yK&6GEZ#7tJXa%C!LN2{$a?r$H|HL-<CRrk6e*UiVN
zw{%|!{p+?#I^6Qx7}kM8eO~l&4>%?BF*Rn)0DfzR{n}mCrF+j!UUD=g&xE%aPn&*b
z+L)2JP(8DIv)JZ4Ny}Ts$pVx*5;stWnb(j{S(W1MbYs!N<gl2@Iumxh*O#1jUgi5G
zt{xvD#Iniv3y;~}#-nj>azE2ptbfh9vh^oM=9dO>Tt*Op8Exq5#UyQPg^Q}hd`j!1
zS!Zu2S;rZvTm7JXUjCyT%KImatU_M5wv>l46{FcgoFy&Hpwm(eUv(1Y3<aT!yVVsp
zuKh}K^&>u(bBR5fT?2_$d*7fem$;VJ|C)U4m3ziBcZ}0UY;czsr$ART-*;Ow7?C>b
z%;wizt}2pZezo;$>p^bYw_$ch%IQMlEFYy1^}Kn^-t_v1WkwLH)VQ;M8EC!98;hiY
zKkJZzDTQHv$KP}&6nfvZl!qd;^*&VOeSvM7qxfq#=D{udPFW!)v>(&m(_-Q`uAfgT
zz(tMl?dUcYJ!#obG|?OQHmjk`B59TgJ;Z9!hsC2xi1H<;;^5)k%EWv1fu2n$rh48n
z4ydE7q-CEAIQ0&<C*%ctE)p$&qC_F5zaQTEeH@Ez4oU*)LU2JBcyhZn@*6jv`x9oR
z;x?F+(aH}`Q>M-31)3X@%P$(lRsbc2kzMvsJvJX~4-imd(VR+YV|~#<mTEMEJ2eU?
zOScrvt!g(%mGJR$Zvi^wPfrds8fAT}c_7-Q?n%S_@UR%%s&)WSi&s;!4&9^mrPeGX
z5pzyH+7Zb0gE1wnJ4b7c50W|)b)34Bd)LX$*MB>O2hqG=KG;v=c63epfI``Iv(D>&
zihqkXQ|yPWb}M@x<eJgM_b~Ijz&tWOl{%>IImiAh2E@fVp&*LLWiM<mOVMSw*ka1Z
z5aKs{^c0>hTnj&FYF-2Qig@Jh!0~hcH)UbZ0GmCe<g4QJ57QxIwwcayL7jHp9NpB}
zW2EK4$ic<@gEP&6aCI(BX=0fsv>EdH{OlCK9v+861@~T6j62TVx2l;YsG9qqx}Lg9
z9|ZTVCw){il@BsaVqgGLJr>uI1I-NY;nvZ~gVG~F8#-AoE6Ay3dcL=zI%>~HjT92z
zv?jsxxT%tm+g@km&CkC?HYFl$n%nr_bVw*9I2$IiqV3}jsW)i|p8b6#@)#d^@N@oy
z#7$^k<9A%Id@4jQnoMXy*Q=D%B_Y|@_w1^1w@}kq)QbwhD;L&dt3k}t{%`Ac8Ki))
ztoS}YAH&|MYiO&lk9d*YZBKr`J^Z^7vj;IP|Bu-812^(DygPBXu8-<IG@ykme=>tN
zxeZh298Ef5wij38iCL2Avui$*R_=yDSpW|#N^1pRrlYiHi$fo?L`8?w3J8(exYw<8
zo%o_5aq_{9_hF=hzFy?Fgha8yi}!aXzQ<^b<5TdtRK1;pcF%$uICqE`;n7C5VVhy%
zkd2$hgE3+dAyAf?<+XW5@EEI`y!E8EZZ7e@BwKiZ<|mLz*wP!4yLw(V{CSLb5ptx8
zQ?6d=0yHHz9xu04E4~YGdxnG8oJIQ9#{K{V)inL5VMzWnXMhP-x3fkz&M77dyuiky
zz2GLQ)N%UN9fB?5^(lEw-Ay=2eGo}Pd`{UVSu@Y%c=L<qK=O6y&Ze-(r#eR+RcZ%^
z2=5pA9t6WLlHd-+uP@H_T$SqTmEsp!LGp2n&__^KuSR`fWwi(du}XzqFsEC_`?CR8
zYqak9ixa0~x@+Hn!}o9>Jb>3fk{Y7s-_(Z!m*=`<UeEbc5i%=9m&$a-j2Vdhg*}<B
zJ6_+!sj!XGq*n4>p&CAZ4v1zl;A8jwdI^Kz1iaBqrFXp2Viik!fS<@eZXFZEO%(_a
zEW82po?fu8M;Cw2oE;klxc#>te6hSt>tJiAKtNa02yljy$%nArO|WPJS`b!m5*=%9
zJg1F1*Nd=LgUvme*a(=ayQTCfk=-fI7gn~Jzb@n(-N_f8<Z{wq<t}p?0z}{h;8+Mp
zR&1U%W@c%X+yCU>Zi{M0NCM7yskjeQPP0gSYd-tvSlNDyF}=B;&1xKOFpxv>i-w-j
z#g71A1V|SMIALApTQi;{w_I(dZuUh8+)_AA)|SWpK3bY`v$?DU)U~mgn=V90s(}Is
z6BI@?Ow_Z6HT~KZTE8a5@(-03QA_In<%FMeF!$H|dLw0{altG1GU8-P@f)S}{5248
z-pR)V$h)^NBY?V;Lkrbt^e1{vc$9D=&O+{~JT*s_C~+{q&K`6RXAjO~`AK+ldn2hw
zjk;$KzROM{n*@>^DF3%~c{`>~fG@Y|H~ZR<(=GZ{QLX;9Y!ii=Z}!w_3uYiSF<p;&
z9A1;S1$xBv#kPRm;{nnwH8-kI6h<)p_fzk^q|A(is06+7;@s#TVu$TB34qI3zm5wz
z_hjO>7aNU&z-Jrw+3XYQT?|ggex0U_`s+aJnj;%*WHh1fYSew1J$pYD#OTpFJ{LEm
zR%s`$pCNpR#DIK_>+Ic0Z3iLOCJ%Qev+K?@F<oCzrTS1LApW;szkmaz6!fP_k?94y
zo~u9pNa#<kCr9VS(c;R&KMdcfDj*+jiA4i0<`7rJxYxoyw$4d>569gs@EjfEOjLzU
z_@$m%x=Gt~znP4Sa1J$@LAPW#aI2MkFfYc<2@yMY;{&}s4q(lgEFi1M+sz*sgld&9
z9SAg4DwzEqS1nnbqsTky^*#`<u2q20+~uzwe}ZjFONd!MK?<jfza9p-*8x&)T;uDm
z^Zgodz;2hW+UF#cUK}1q=FZ+_-tBZ9Ah6T?^9}~%T|n-@#D7`UK&nK6MB}7Z_;Q<b
zP$*_~dr5kkZKRX7(55P2X%;-pow$xBkvpF5$|%>4fSby2ufwkg7qx(<j%TeunSxFL
z^u2KTuFFbh>Sv4AkB3Dk;H-?Awd!K|{^sGbnSWmBe<`HUsB{xF#IJty^`SUzw$3eB
zB<g&9B4>F#mzp24DeLpx<f<QzCN^;tq@d<W%4}LIzg~$LMdSGG=GkRl{!oP4?Q>g-
z8zDuMPST=#8iv3g3QO+XcDO@U>h=I7Uc}Ymf@qF1+^H)5B3k0$A#zbQ7eB?*;Fw--
zH*^TVud{+3@#!au&~b&(+tN+t!RN*clO_QN^C|oj+IwI2dfW-x)qNcKA$BVJ!JAPx
zY%odV`7+WrJ2C!SS1Me^nKQ4D3w@*bd-}QIjQgjQ80k}T0*ycbKiL<>VsD%>cfL+-
zKt^+~%SG=BezTRE9axZOHeifa$pL_;S#^gE%OSA<*+c>$MP-&}!et{(tUnRfJ)gtP
zDsDWao+OaF9^K5a1YeP9m2mfT1ykQjG=#%$SNN3sErkUQF6b&RW=F{ay{0;yi<@o{
zw}pIt;p(;<J*?0ZUpt?y2kz;vL#h8(j})Mlbx(#O03oZ8(18c@Ow<%{oe#yWLcv@2
zspi`PV%`SLLpoU{NC1s`wMq)eWKnp2m7%`$^32qwRyrz6g7TdQsRD^3d2q4XfN}>g
zDpb(Jr(P1y0dmh}>BGf;biJMvJ6sqZK%41wr~4L(ZGp_;J)fS$%A>e?>KSC`Nap9#
zv#B6NvPUJ5$b$=+b{^$-^PQnm))NOv@{9SkysrVYIO0{!)Nw@k)<oSiloKO99`BTX
zMO^kH)2kX0#+D0uFfY{HXa8ek#)?U1<a>&X=WUXFSq^H$r@ku(2bK?j7BZO6ZqZvh
zUJ}}wZ#MjSYW4pKjlR#`P1LO~wMiCvus;T@s+Fz*p-4ZQLjq|*kKBB3iGf!^sPc+D
z^5&|q#}|MGNfSzWaQpD1-OE&Oan}QLZ53NfpFD4tI<&_OBdL6HM@6f{9Tz5Bg_DcZ
zjVMk;YI-he#}6;1UfrpnVwO~i*LNVqzZtqOCfxrlPORL{Xe9Y?RB;3VxsA>FQGmrc
z-)e9qNEdxQ?7_0(GJNz4IjLyeTF=<mYGa_+&=d@WxT<@UKPAGq7|;rD93rp7o6ZkT
zp96_np<436!cJe4t-hlpiqoIDaXCyVczx0Ok){K!EGN#=QU~awuKuS_2#DS^vUTJ#
zNMX^bA1QWeB4`e;fB^FBg|Nev0>HuTSba|!1=39c1FNpaT5_If2RU}6M`Wzu60$<@
zT!6qW5jSD=;j(G{&xp4$u0wlBwXIL22Nb5;REIeA$23;uJaZWY62c&OP9K$`3u4={
z4*C-vASZi??)G|<3%es;bW;BRCGl@B0S<qAsNm&pW225nAWzbU^VtziKJmb-^_3Re
z$VTz{ybM;Mx|-bl;Ha_zT50@i0w7Sk?HY1hoF9E!+Kxrr&ByqjhF?7N`-g9PLNoKm
zZ2P#%+Aisr)L%FeFU+m%#2joRSls9)fB)_-%IT)zt6fV+_<XE1P!{%(HA*}TtFrZp
z_ki{Sy2_`ZgzQZspALWXUMe6v!ISD}3MFEP5-?ZWt!hh+z;G;oXOHqI!fOJw0x*@g
z&qQ?@!rWXdQfMa!p+BE#3*;UFA(N9kqF^*2Y4DGLnU%Wk3ovaRzPl{K-z)xamYuNp
zUXC~%epE@E6(lVtwxHNGMp0ke!<9BqY8%cpGm)rx&A$Sqrzpvxv5;d@kvaouhmHO+
z{vDOA2p~zZy0sO|m^r<ZlPQ#T;vCw#Uy{Rqkj;2?s-S9k;(VhLe@ac2c4+xUI<QlO
z2n$W-FP|(TdaW6#_AKVxZ2WQdPMgM+DsH;!t>BjHL{Au#=xAt2^3YV+(9M+RApA*t
zP)wtL-{!S@W=%-8sRHHni^GOYfnKgOAmV=@ZutcJ^66_>Ytt>J>41;)+zB-28Hiyl
zSOO2j*K;b2+eGNH*N#&cQ5qP2Mma=LJS6`HzW3(#P;q7cgy<uZhryU;&k3UxNKbpl
z8S~(Fc_4~HDT8<wUaK*z_J%Vy8^5Sy+{w4?gNE&uht_+z-N0#4PL^z%3}er-I>!$}
zc<k(0(*_G8zWyHhF=%r13zTg2Ga<}%V|gqczUSH?MqH-D{2|8QU5H;b`xMjk-=_F)
zl`7Jh^5Nr3pL8s`{!G<RvTx$nIJZxjndVr=P-)O`#Z(R)Vc(JNeEEv0Ive&3Mbu3g
zKFn2b2HrU%LC!(4JqQ2_@n=y@0U04Extj!mdQ+988b=z(cu<&P0kBio|0cS)O_C}-
z=b-wsH!3t)Mtd8<M17yxj2WuMfWV4R+ouoZp`s-y!=*$DW?MPz$|ot>Z371ArX#$)
z%3qhCle^A1I9871zIzgn>2=fdhemg`><Z4aWb?^X?q+(#*klXFYj{lh-19-)kC%VQ
zMQ@%I**oVh2xXpm;WThc4M0(0S4E<@2=vUQ!x%!lmX9Sn0SxzBGR1>*=b4~);a^zZ
z934EsX~s4FTH0o2RXuB+w{GAvd-U3Y+~|Neo6{Ky5BK6vBGom>p6gm9BHe6;cFONy
zyJFQ=DAyR3Ps&|J3mNo|<oq!~mQxMu`}c0AgBA~0W^vzMkG1>57k=88b!wx(eADrm
zc?$kdO>@!aIF*9i%PZ}%fj$l<Y63y82{c$fcsTVox*11r_pFZ*;hBp2n=i$1KYSN`
zsxaktS#XD!-rwQ8v*lRf|B$h-!R=?>Y;hNn8)`xNZ|Q9PNdhR0fLF@X<;i0@fD>ca
z`fY0z*4sWqIs}zjXE6m3BMz~@ksIqfH+<_<LJA_;9s>wXAgAwrG#t>zFnyE+zTf&a
z?FmLmNTIS{^ANN!k2`ibW1sKyUKpuP1y8Io0imp)(AgR6@l78AAI}BI#i^G|yA1ov
zH00(BR)JWqynZZ+7-*SCxyhmnoo&+A;sMnGOs5AAsP7R9(Et7VTsTov<f44BEgYBx
zNzCxdmvj!^{ONILU|68eC0dH#XK#1FlktDW5h-S5<@XxZiz@%CE3mjCa^Z0QU)jZT
zVY=+<yTz|<fBrYh8I;g>N$ny{^j+&04hDwU^1XHq5epf5CXjpUx5o$3e(iei%zHEZ
zj;Dxb2!B=rRSI)OUG%Nl6^jd<$!<TZfCu{FYmTod)ok~gF+s)TmcZoRe!1eYNZ{EY
zGFdV8GKrxeAyFbryiL}=NNal#Ac|QI4J-+Z0IDIHiH~8g<1}{LcY(qaF=+|x$HEX3
zRjSd0A>>1rDoC^?ui>#>8fNloicw}Y?1Y{b#KV38)O+0b-buY4qWiGnu%`$zZ2r_q
z33MgCsQ;9B&pp{*$B5Q%<3|=&NCMRRXwF?ry=%Dhxj#-5BMnT^%I@H>#sOye>HPk}
zi$TN$oUe3df(|nEG%R}sESdE1spIHEBWZ=noeBDKs=@?Nq|e}Yz=(bjZuk+}Dkb&f
zdOZNBqt&g!?dGkE0yZG)tjQ)!MB(VXiu(38$9IP$u!<@Q9XOS<7-FY8g*+J~Y8LZ1
zBY&g9J|pk~*F!Ro=#{R2yW<-eeu8K~p&g#Bju=p^#1y`vir&dMzA+jAkRd?vU8LW~
zGG?F$!KTO@8O-sazsdFX>nOM6UWPR<umI!c?;%^bhT?6nXr&e?$pSyl=9o?uA%G=&
zd#^b-<QDI}2g?33%ju6Ja644uI%a~cC4g@B%Ah4P_ugm2{Pv;&A1#Ff0yvh7X@Pnb
z@e&vxtCrpaOy#PwSq)^7q!XlJWlBOq;;`7kbYb~fukS2l)E?FA=J3(yvE*f2kfD<?
zMZ({X#qYnG`G6{*NX9~hSI!Q&I44+TR6yx9ce4nY)ZKvsD6)#GT;~?(<lEWY(6*g2
zcMZg@K}>r~frQq;jUfSX9#Oye^5l)j@QKO;L;jEsXX%5`GO@MGW6hJb<FF9FRkjIS
z0R-ye3cxD<zdFu59?G_h<M$Zr*cps1Viei8vSi7UsgxzlR8tKig^-Yl8QG1sES0rK
zvSvvck8SL-_9VhX$kX(CjI2#)jQ1M<y??xa%^%l&?sLv{o$K7!eShZn9si-Apo3rC
zP1)mJ@Biex#Ar={AhoJLQA35nd_c_vZ5~=Xe{Asjz&lOq{Rhq^OqY4MEEljb*Yj8r
z<Y}Z8rsTK(5{3vI<FDNBWBOvi2GZu8Njbx?DXm^!1$B&f8~y`t_hB6r#HC54wxs22
z9WzNeT%C<SLv=NGHR2ix1OX)2vLdT)P0t4E_9-Z1`?iL$YS&Sfa6ymAmmXj`>&^8i
zaJFP$IqH)+-E^a<|NKyd2ujPH4Lx%(U5CC@;I!yh9X;nsmL}C(RIcolmdS4U0W4w6
zgc$}F>QlI|N8~wgnpy5G5x5D;WR)V}ye~YcWK_VC?pb%tBnQ^q4Czrt*yxB5o*$=*
zB(=e89Z}X+=-~)<s*A0!DT^^7^WGMM#H&q&mNl#^*?@#-KfHmns!8xUIcp&4JJ)nc
z+a~}Q0kU!S@M;Scu(OId1|MzH^1H)gSW0!YGY{cyoVCv?M|f!`7*?s3^f}bAa6Nrf
z(@W6b0kf+JgUGjng%8gyQ`lpjfW%S_(SkTh>QcmE>Dh6=#*|<Xmh2^EaE^E~))|zh
z8&{k}omx*|!O^~!ks)nA#JN9N{wE5Ef%~v+<YnQEdw5s(yySo>UXNd#bk0(?mrH94
zT22e~d^8=>8<`VK0sTc67RRI6_AJQEdCzzPSMwn$p`^?C%fXv;`~B)``p5_|>_CS~
zpJS@WU>yn4Mfvii|2;U@ChIGkYlPRe=O~pb>vXgZFG>O(wj!gR6{h`OINzMZeQT@T
z@mZ(r!q1C^*XlKFKRv<!p@fcztdgb#ZWuHF8t%$PO;~A9^yu?L2Jgh$oB5-x^D9>P
zyV>yt!DVGl{M*kg)PQ9B@_KW4U<X+$)*_^5>2YZOO`Jk-7WQ}1!Qg8kH7~b=An<uU
z&hyCHlTI~-Op6IqJLf5WM9jU?)zHmm@E6|V$yU6bJsAXqQ+z!e-y}AUh}|0HrW{p-
zDPG|@QJDJ8H1>*BNzE#A*z;e8l8Fe!>TeC7huNF04oe6xDaM88UVMik3i-#MJF9yU
z;4U8O=T?e65Y8yMhIjpXSBv#r5YLR{q4bH5M~M3Db8nBcD=4&1kOBof>64*ii1mS`
zoH?Urnpahj#)q;aLc&Y4)n?g-z?m*^X6PYDIWNQA^2!Sg9PS#;y%?Fsb{J^Ipds0I
zk?X#3F5I9C{)ZaH<1?W5>g<DD<I-df{WsclyRFDg9^nk){4FW>qd_~}A+Z)Z#-Na@
zy);GWA$U3cl!X2}>RoSbZuPcr#XWW@q;}Se#XO8>Ylk%?KS}79n9R!EUI$S$pN35R
zBFEbds4jRj{l^L1kX%NBT%Wb*gbFhAsv=Z)yz#4Fr}~&S^>)$u^GXWvnQ$@?R53I6
zz}UNhS!E1EGr+Ak!b>>gpRi;rniWf$U7SwLf+Bn!#vQPms`KXh`3ZbN=K@#C0nPw!
z4BNh8RYWzY{am~QQ^=hCp+Kv=K}DHxiYY;bxi@8d$0NjOQEtOd`Y#i12=mgVymIzW
z^kP}W2inMsXI*vlv-5iT<q0>)A*7unZwAFx0kMEm*)#roal>!E(wz9r)k<gRm`qz|
z)y4@i7DYg`F~Xjp$<QaYavN9;t0~+X-3VpWB{v1U2ss0so~iTxiLKiAt^E#*FMn^1
z9=4(|B(gl)os|J%;QFN&FFV|Krz#U3k5u-%_qY=<W|xx|q=z)3uCt&RD$<gZdGXRa
zBE5%T)BvN~yvm1#GMv0QH!UlAnZ&SOX9IAPs6PWe1s<_+rPuz(=RKrcfYXoYKjsmo
z+@6=^TP0kI`?ic9b?QpEG3)uufsvTpgvX`0vv3VljEV(=QZ`Etd*UE>5qsB%{C#zb
zBeWgLRDoWY4~~R-mw0GCdY~1IU)J{Gov7}ZK2}w&py}8(r1IA&<4ccE8YQ^!A{^*_
z(*kH&LltB@R>&Wltepp){=uLZ{op~e{C`Q4Bix%i4=;7r2$g92F!>m#h(hl%DU{Vj
z^y|g1>->m#Z+7#RX21nt6!VvrWt8}34m?DS3D9w~_7N|{yHe0JxdvjIGAzB|iqoEr
zT8^?%W}FE2rX7`Zb(m3cmlVX&8~G4AS~ZT3v_uE+N~HOYFM#w)a(F1yt#v<i-e#2)
z8<(k>?dwMAwpXC~aNM|V%4D@}+yYF5Ui0aRS*g-P0659?ar<dTcq7JrMp3+PHto{l
zm#Gz1ao1J5=wz)3G3EF5{_Si-7SV~Zd1iN#1-Km8<R2O$)&%C2QJF_vQ=@d6Xa4^2
z{9IL9|FmU?iJ;F2$7xUBSN^oC+j&Yr&2GU(1Q{TeVB?`4hOWf)hIs=U9Ut1}EVC(W
z!VUNH6h5`<E}s^h`{hz&RnOYV?3-)Nf)xR*w3*sbvrc*4e@k(s>#gZ!ed|Mi%g?Uj
z9s75LQa4s71Ia+7mH1U;--d8pqc%NE=DVQ=>bGexR=<&D+*teX1ub1M?j;)y0j{4*
zkv{=z!a|;`$rH-fYfG6<#P@oj_>uIl`5qK;yn5VuY%?M+G;lWbz0cM4*xPx4e}wi^
zD|;GUw+fkZSne)m)};^B8}8AU{c%q%9m<eLU6Pii*^U#mcOdMQn7Y7#%LMN1cd2G~
zcx#^;e!FtdMiX!^g32`rTZP9()=iqcJB?y3(3?wPcT3lw%E8_uVoC~HFXXqg=unyx
zPyDzdi_9@ee%&!q<yrupl+M&^_BE|_BBmq1>_z2FbzMPnvFK2Rq&jk&KRl4b((wa0
zK7u1~(jyj-iYus3jjOUc`en_yXziRM3NU*xMv4D0gtpCz>21wBjXC8Lt(cIHIuzT+
z+6QLyp7LdQOs4TZ*VBuEacqZQ2t#LV05Jw;<|2=N|L@w<WHC<QE_0+}Ad>%Nxj812
zO!4)~@^@4|xbgFwxHV_CpNP1vf<2Y2Z-S}<toagkN}VDEYfOjzI`XlrZO-(9^Nv0u
z21NHB9n@<I;a-|!1grZ}u7A*efmWEhPzlwq3XIf}y|lA=k$g*W%4<sCYy>>$vB^t5
z{U~{^GxAaDOQJ%-$I~FA?_fU7b!KOhxL=z{lS1x@YQ8S1cQ5V)FJeC=!~ofg{2&Tf
z07D@8rc&k{E=4wV%7didA9*t{*Zu4<(Tl^ft>4s~OxssEh*f%6C0w{K4{n9z+j(*`
za0da{PsKu@xGx!t*wYPq3pg57QPFQlMbq*|W)MoT3h-Pi#E0gs9OcZ0A~`&uxt;hR
zI}fV8iMYsL+yp*U*0*2oE57ojcs`7vC_TRlumtqjwCvT(48=Y6VaQ`BcyD{aN|O{_
zLO2EDyAd={&~4D?KJ^06$K5D;$M6cM1kewR)GA(6HneJAMQ3EGO)lqNWx`2yh<H{(
z>~DS!e#C=5#O|exy7W?Z=mcP7oG%svt5p9ygD@O6*<VQH6(d#EN_vdCvrb>Jz)DCk
z(R{qIj>;)-)6tLhcN4j(9nQEMYhh*d#~H##l$R?*GF}ojp^TF3bEl=~bqhFy?-o*^
z<un}}og(IaITRH?xeZwlX@uJ!IQuyG`-~43w|P~~jH45QnJ(xU#XppAH1v8DeVA1Q
zVbq#}gvEW;NkuVIg9t40ESV-T{lSJ+HghSQYPjF$Nqb`UMTD*9n}I8hfGk3^VA++d
z&en&Zncanyykh)TvLAL6eys0HRi`Ncg_%40yA9`bFg3MKbzpdS8Di~CnY;t<RApFW
zyQAl8xcVh^Bs`)Je9q>(L!#Kt>q9ExmIu*hnr45Zj!kUH;oH(&c+q33sK|cW#w#QO
z;Kk&Zmi>JOFu8K>!@lw*gsm-7!(hiw#>7@Kt7O$0Ya~z4G$99-%oEwZPQ9V{mmg~e
zf?@FFvEm`zD3x}l+M^hVUZ#-Wi56pr`AFPvX{t+06#$<B>qDR+VuqiRLj-#B*x8ZD
z<I$~>IcB<z?r%B|W~q6v+Dk}#q-%1D;s&SL`cO8Y-kF}<I>DfZ#AO0&QFYd4T;53V
zgdPtN$W6SotFB0C*J7l^?6J5c0J#9d->74url|KxCIVeN*d(28(q%ZEC!j!n!E|IM
zN%l6Lb<xKxrB#kwgT{Qc#CaJ+h}p;Red1dRFZMS(ld{Mn=twz$rAa_{sr%T7)N<ix
z;S|tW&+e;f&#<?^==`KjPG)x{tMyA^SQnNIw1f%Be5?IzAD@kj_inX3NwY`Z^A6-Y
z^=fgFx!s<h_@}J&GCShxbmL{V6^-CWR~f=Nd7L??xX44@9DWo<g{j4T_`>(xYbOS7
zV*wmc>7F(A-VL{Vo<<&bJ;4v8qot#xuC1f4b=*n&gpr=Uk)DpKmX?v0R?p{VssAzX
c^S^P+EA0OpG$g@GfdO>J`hwMK3w-o{0WP_VQUCw|

literal 0
HcmV?d00001

diff --git a/docs_new/logos/shopify.png b/docs_new/logos/shopify.png
new file mode 100644
index 0000000000000000000000000000000000000000..a4f9cb575fccaccb38534bd5abd64e7cced6d36c
GIT binary patch
literal 8003
zcmZ{JWmFVS)b|2XBEpi=vUG@mNP`NC>;lq_uyi-b(j~EU!;;b^Aq~>q-5{N^bhi)x
z=bZP$^Wi;n&di;-F>&tvW^Rb0yfprE%I5$803QkgD+2&%peI@f2kR-H-9+F$70=8*
zefb0cR7Bz38(==wUmHP`zW@O4%m9FI0040N)a1Jh0DR*H0QU3&0AVBmKyIJ*TS@e(
z1ItiW8VvaNU&?GQhyws<wV~ios;&zM>27JN)5%w@JL+oXV|;uf?5pTF>^SU{QlBup
z#6ngX<@LVEea4hB%Qiht2}DBW_{Pbl!Z-Lor@wMY?g~LCqErG|k|6LZzrU{Jll?v+
ztG0CZ=wK_Uq^ow-P-lCAesy(ozB8%zEy-zA=n9^(BQ%OmFA;zP1c4ER|4U>iWNF_9
zUA6LNX(r5(b4&>hOh}HaqD5#c*;t`kij4G@NM6c5jRlfEBWU%6YGq5_yNn5-t7*Ml
zjIe<TZ8gnq2T2lxz!0$Cy5Eh_Xm$L>i!p%<k5Y5hl=P1pulK6Wf7D($2iEj6H(2*}
zyjw}(w@IOSsfHj7<|>g`S#8JMXf<Q|DI6HkGNmn+StH*3Ss-44;#~oGA!__p<#;?>
zo;PBGKzqQwYEj6pG($`5@YKHWX$njrA4XNN$Iq*e@pny$UAOXaH0Ik><lkmOx`^9M
z{0AQl2arM^o(FS+5sIMLxUqY@mLStxZ=wm&%Jd+-RLE}o9-hZSe^)AYTy&XiLzi&i
zBY}S1*-a!idHCI|ec%780GbXp+WUqwu3!iQ%)6TfA~Vh&@>h<M-<36$K;B@<n;X6^
zoeoEaE+uC=kSZ3DMpQIQrN`xc6Z^)<yP|YJr2j+MWXMXkGYbz0ibqtgH|#KfR_*f6
z4+&7Tw7s$+tt+5aQ8yM#laO#It-tMYGE-=_9U6S!h7QPM3r*@BMS4Cq)g%rx_6Ct)
zq8{z%o{S1%WwHEya6`MRk`_f3hPn`7{*sblee1RD??C(!Rs*B>-v6JOs-f<eR~V-$
z;=t9Qv+bswHR#Tw(!1>_DvFo7|81v9S58$geEgUWTTp80bO~>JUoIiBT7B~BUy0j#
z504Y=m&TzzjEKp*hX`lzza|T_FIl#>w~twS4lMVl-Dj<r>AfTSuU?w{Crb&IoA1!3
z<!~AAWY&tjy{j$rXn9G!<N3#OR;Ml();Z%oiAI|vJNZu_{T-Yp^yahjdg#+CdmOj#
zouF){up?5)voFTK)crHCKC*{M1_*|S>}CmV{N~7Oo$aD0yRAJpiZ}dl!g_r(g$G-T
z^Ooz1+!xENEt2s~J{$RP`y^!rJJQ}eG;7+JdWBJ>sK{=Nv=RQ^+wqQpI1(rLJs8m$
zkJtxU@hYx+-k;$cZF@f%iBTSfB1k**{G$Z>dg9uH)gAi{yH6%vyvqGvnI@zJ%^Tyz
zfU|r@#mDJcRAeL7)tlCvwX>+_!Te9s!VwvTw7UKi0@SYwQ5wU!^ftQf-f*;wu2wG-
z_2N-Nf`49M>mGjxBVs5j_^x&8)*N3_7phoL7>xhBv&Gf(W{RbUmi(n1g75%i{54l)
zRQBMQGhCc2S#*24o8o1o{)fAHcd|AkJH3Sp`f*pkh!)X`Bs@~p+Y{@`IPBo;_k4T2
zd6Y@Zr*pd^V=))dbt+4g(T_*(0T?`cZ0%Pt&Uh^$LUJP<bK_qwUy?ALT?()1NchxB
zfm&3gOO7k~7+N=N^=}#zSg86ATKCPe&q0Ah?GXe)I};imE>-&b)DeZc(tY<K{FDH=
zt{9D=O&4pCeL%Bd(fiZpQ$VL8)vuKq@jWLW{rqUGUFaK2EHAw@Mn7YW@x-B`vf<J*
z>H%=BuMG4G&7)IeEm~S(l`}lO7JCen75x%W)7IMt(yurxu6_%$<WwhTFAYnJ!b`mm
zI{fj$+nj`Jk}N6f;#fjFApKlY?~*6@x=V!i8e;!vvaPP;s(P9{qRQ9FSH8`xZLlo`
zuzXaGE`XL3u=5rFzGVp*qwa1^?8831ojkp9G|3!Y)0%Hm|K7r*6$&czpjnT*GhTC$
zm7q0OJI8pW5@a~I^YDeCf4Fd4T5D5j(~sJDV+k*w_nOY(7!McwC{tQhc^iYEB>oLz
z%#KEAQ#AVu_=#&R5B~sA`ALiEUEQ_mejM6xOGS_F@C*ycu>aSAy9C1Z@K=_rCf$h7
zFRYPv`YV0G$LF3Dnq6W1qt{Nm`i3xl^UU|`&ok|Cmiw}3Vr|S(z>2ew>34?r&=c9!
zV7Ka>AB!bmKBwgyX%u9WC@`fk?ovaM)kd{~-!FF*Txr6@l;it>HapG}xr)4?$E~3h
zMkB`Tz32v$Ktg5iWoa3o{rXiS%)kK@yFX<z=|An!sp(eJe%HqN!)6k*_Yw+@F+V+*
zj7LO*JpB9>-&j$-TL1KI@3o*QA-C8lfAjJ-f_1!NjjK9kVP@$cpZ?+_r3K3rZtyN+
z(n4@zkmr~sYmB_gM?-Ps-?fzFUznQX+X$S}5nF<szrBjhusqoXX}WFjxpyCSX^>jS
z2APqkm*bkKet2+=2Y-)PCKhY=e#!e<Q?t!VAk8A}=XVx&<n!zO4+d<(ltH*rcxqKM
zou7x)Z<4uk9ST;pXK{}j$dQ|bN~sb(nX|#$;V7;t*~J<#NfheTV)8jmxhO&gtP=?G
zL3FHPg=;wRC+xfEM-IL(*7tEuNDaNZsD(&qg|JfO_+;4@D#nv*j2Bf4=!#HFN?E)O
z&7n~g?_55OL0D5=Xw9BQighu=NqaPDWF&3q=l}(1Rf2@3Rd}vw8x{{#?eELLWaT%E
z$J#mvi16a%%95P%cFiKA80A~F^d4Y(@MMl5VM3nQuafOL8@YX>JMy1%zcy&pt&^p7
zo76vp`|lSt)CXAgO$rkO&6!G^r<k}K%^jk4WNjoT7aduQ89_FvNA^+VR3y`q&`n>+
zi+!O)&8{URjO0y@!j|H#>Px9C&B~J-e_8NIXU20;7i5tMA!YF*XkC0nNX34Ld6n)U
z62ZCE6>_809N&$|DKmCmu_LUx7u2<yvG$q#ZkY(fp4)8?o-<CA0n2t?wJdFD${`U{
z+Y*Z|UHzQEwb9S2)CX+=#NIMT1H@he=8MZ}`$a1tmRQjcR^aTzLYrVb28N$4^8AGf
z@_F4CACw;rnhe;Zk5+}RjlN2p8bVw_RMn{h%iOVBkf43x>>|D)dL}IGZuN!W*iU{j
z(s?<#bOhNH0nM41Ay4nNNmt6z>PmUkvjIzu^K#L-5UA0dRymc=KL!<S#P#1J#AC*E
zb&8I#F;=ifdcZP`fr}+rk3k{WDZ%rY{pS({B!4`#7M{;zBK@IgF>k~M#8>H8$<bmY
z@|;>_dDz$ug)lW-v5oA#hzieK%jwWpz6os`>bni>@aps0<M-eR6W5~q0K8xBpO0k}
z@Cf<+nW_#`>1iMzo^*mZRjqv9e)IUOhc%<^)er~f>@EJGh})NX^lNy#!_QPL{|paX
zLt~FRv2U)Sf7HV19I<<yML3S;Z-y3E@@o&BPELN<K!5!`t3SG9!@RSElZ>WZD#;2t
zrw5+kQ$+lGkv#9%6HZd*J?W>Db_Y9(mNi>z)-WVmD?OXPWS3LDsflFVB~A59_1h6k
zI}<fU36r}67SsN<Vu=DEgDPH!;UUHv#Tcf8%fmLtS}9#lY{vz@JL|z8<8CVfaRg9_
zy;dSUANopB$gBV=oM!2gHl+6)jp0_00teS-rE!257n2@nfxqPO%_2lGN_kaWPIMG0
z3h86^FohNswscvH;vvU;M`#Z}0Iuoo^RAm`V)pZI8R@AVky4iQ3h;WSb}K+F1~51m
zcKlJ|%f!Io$ohIvTr_E^X3Ap<slGfHltKJ&mCUpjE}J!GhV=}6?L>z2&sa^&&ZytK
z(Ry0VOCHB|Ej^aL5T)@#^f*PKo|t{I)AF=R9Afg#S@D1z|E7Tc<sEDmA!Cpf#W5#y
zxxC3oFL&(IHx>e*+aZdi0x@w7_D~vs7lvO3;t+IcF9J$gFT$Ak%yLq$+DX6L?YDvI
z8SCXibl;dNAVnyLixV;H<p(sB<?{O3LR4YtVz_*jWa-5(9kO_^akYa77!6v8a~)=c
z_Z<E0=Jgi|@Inl8&(3@#TSv%DcZG$nmTyaXsE7Cu+K<v`z$LtUm5hVG0sYuhdI)o<
zAA%L@p03@&gU6*;4okTKY^<debl$>!mA+S3X+Ql!eCdW(n~t-Eq*vGxyv2=mE|_$>
z&OT~MvDs>c6%8g}e+^t>o@xfh_JwJAhvo0yHDs=SR38={n+3LhaEN&ZYo9B79*FI0
z*(<*xO^F^S!N6K{t<yHzXDdm?Oub`bv8On;H8pv30_-tAsI>qo4RiG@y&^dNfz(~w
z7MKowoCq*$W)Iqo4_faReyl#n)~ef}TKCYZdcc2j4Bru?XVa-J03s#c_83fe=q<*F
zQ`ZrfUeob6-)XjO<bE0Rr#HnXc3TJ`AO?;+QJnTvzxBgs=UvPg=|z{&t)nnK8N!Si
zbvb7o>|bcp=VmDgxfYZ!Qw$P61LyV~BsrI^K1RGsWupjzSTqofE9uJj+nmkUD((}<
zS3ewhVj3`n_8xFODQ=12lpYe6*1csoHc8RuArd4f#yXY&fP92&g_LC@oZh~T{ZyuQ
zq;A}f)b~utH`-Shp|(I-u8K%p9BJ!QCO@Rz3zyhBAX;g!pSx<8aE(fUzXzPr>%Kv)
z7S<IBlEXbVlG=2%yggVt;e*e>s_W4{{H?X(xg;vXerD7ONY3KqG=y*x&EMigQcAaA
zS3n4()XHzOuQw`n!w<Zx+g_58X9Z6BC_z#kzYiilEAag<H*>)gQfH0S=yDf>Hx#BG
zy8hOfIFZsVKWhe(>#Zi$Q3Bj$I8iud<D{>4&Q*kiT=$c_<Y6C1{HVo%tFt3lL}{Dh
zYDOO~vpzW5GmjsR{ITx@HO-?Zp=m(lv#7z&5tA?>QVO|**za9^5A|1e#Ik|EX#jar
z2|_+cdOW`{5v;FF8nTVN#;{QAS=z_eWJH#>0|N(r0z(@Nu&{YInbz&{{*09DdF{BZ
zI|{Dc`2d<WDGSE<Q0W-*_8~H-Eh$j(H^z>Yych_P5h^Zmfz2hCuGOe|Y75>Pka*;k
zIwle<>>aUspCmNrcCHU%U%Zm?5xuU9?{d`G{HyUkmcIEdJ1Y>=pakm|D5eVCb^$mc
zjkD{m$C&Nm%i;)mHW)ZNpT5?3MpN_2-^zYel@%W%F4hU(>f_Kim*c&=a=L$P6rG;Q
z#4*-v>xyLuZb^d8o|CUMrZ)RvHv<z7`hy{cvqE?EY-wXVu^muwf2QHhc|V%K7fNAA
ziW`<Jk%93=rfbbi_X9FLjyTsxqqT-J-+%b_YDWp(%nTQc?l|z#UAR~{G*1D2KFFiv
zx5RYG;^)gHX<VT08^NnDa)rt+posyUU#O&rLkFacR+`rzk{;TB7%{Ce8WyaX)zCTq
zUQ9Xg#9Yo}r27m{l!qJrxCnQnW5eqUcfIs>e#VQW+t!W^e)=%#qu5}tU0GvNwdU)j
zTssW9kdn$!a3}?{XTF;FcI@V3GNhG6%PkjXHU&AQ&$jW{yy=QW@dR;vmMd_&QSxj4
z`{`A2YBXd|9F2IqH!_8xT?(li?&P&BD+8vZ57USZ@Z5bt{6J|SvV8Sgmk%duuzCP0
z38!t}r~KVMICGeU8EZK3soMOpvtZ$^(kL$!4C@wj{yp5<T2B-+;d)wpNy)a-u+Lid
zN<7_Lt-U^Ru76dNfp#_ljaUhzhn1!K^=diNW|%$dDt6%oEjfgojF+V?f&y{LYar6Z
zn2ib;veinGZWU!q@1R}K)p^}G0<AxixltNCWden;TrACRue5_Vc?gt-$*tz92N%zJ
zKzVY4YXxPr^Ss~KAt8v}c!~lwAm?F+M@x#GVRrUg0!>gQ4e1!ZSB%hSZ{+38D`PeY
z;X+W#>zbN_3n-W)n(AA#pt$Wh6kKXjcqsw?TdYMl#2l#Ob5B3Jm&9&$b+)j#NBS=+
z=vq%<ht<Sk#J4DIZfBxt`81b?(@K-6y8v4bJQ!Ud2a!?;cUlpvr_#ymKm7n!uQ4hd
zCI7qt=iu>{P=0ZIvwj66lCBA^g{U<@8<qvj;%wsi=@yK2zteaw_q=%`cglwhVsW*l
zY;&b>N^v&wlW5m<I4SG4bO`2PR|QbV9TB}0c2}ofF-1UtBHvd}R91z4>@0_d^eUm1
zL{Cg%d29Sp3D|LquuhbZz7YRm6!WACx=Q@A)<zAav&rN_)9=K@O)BBptWJ@;<V=zW
zEw+`97IAQpTT%cE8{Wv5)|YL+(bIY4uwG0hGc3X}!g`=e482A1h|KXd)so3#WN@Kg
zOfiU$wvt_c0=y08(=CL7ViQen=Qhq{nMzH>e{{+M(JZb9Jt%VJwsI6Efuu4fKcZpU
zu(vMR^!2@?!_K*4a8PI=L{CUh!_FhU)$^E`k_hH1T<$a<^I~_+u8KAJ1(L_QQSto<
zQCq8it$>whnh(o-UH11%G1msY{Ipr+nq^t9Z+LDEX>c!5X^c~4WiMZqJhCwKi1F9P
zmyMpXQf?K*4SDdQPGg8YjX48wUic_GlnUNFig&akXR~CMF+%kRg$}#I(+0^fIx9R6
z+ex3^kc!&r?mG-%jX_-E;%2_%^-ysiv5FQ^b*Eof37maEC(hiv*n&lRzFTw5Qq8eA
zd_@tjYo_^e7I*!gIi&AR@FX#g1hRw#1NwSe7k+dXo|O1U=IMNP?VJ}1W@MWu!*QAQ
zLkfAyIK~P)9DzFAH*s%!bozH>Ee`#3gRBeoFr=@NAd=o{I7M}qrNL&8iO2PB7W~Q2
zjk$1?MM68`gA?_Yf*cYEAz$@Vg&G18yL9!9Zs$F@<-=Q(@z1<DNWId_E)$b^c1{=T
z!~Q+=M$OH#I&rDXD5)K$V}IvYU*w<J^cWTQPi%+zzUxcZ<RoR{8<lke@%|wC+eMwy
z3r^y~Q!YO+G@i%Hys15gN=Q5?x<24Vdsuu(xNN{lZr3%OwO|V|e>;y6qoVdsz+%^}
z{o4crg8GPBI<orHErQA*qJt5kHq!-T+h@bWTtd^KgKH{g(@DyfKhWoC{p&MSw_;5x
zoAe43r^o*2e7&V63SkXpBWG;qlH;vceeo$VL?8NbqYBL~FT$~C#n#Coc8~`yYiEik
zm&!jZp~rwLD+TT?EsZ$}XQz8#W+UQiIwq{O8}9%2x)HmmaJ-<9mJ#?q)@|p>i^U!X
z$oIX|SynjXj(#1oOSVbM=PiA<l$Du_Y>_b3qa15;ZXD4hA92Me5Q!!GSjXdo`u<B)
zE;#nk^I~&lqG(F&n~kVeaP@?Zu2Jbuh72Q~l9WdJ$3e-IXqzemQ0#3y3B%x0Roc)A
zOu6zdWfhs+Ct3^_0&fN@!pHyo<)$&yC+dd&B3AL_hOkvdPuR5(#iS}-*}fI10ys<9
z79KubXtbo?k<J5h^7~E}@^qOQ=|Qi**s0Q8$%6x<5k$BmFdZ#BIl`ZO4`7Wf8)iRT
zN;x$O#+xXO*FVFBUulUoT65%0N#1a5av7Z;)oW~<cSM=n-bd1()i_iUqz6=NK;cS<
zL)k=7j-Q-(2t;BnA#kfhhfl!<B@a$yV#$6~>uKdAV>6j@h`ZVDlNM^3b>&3qOYABB
ziBRg*dpjvWboT^Z_irrXFrUnU{@!s5`N>_O`jflCEoeu7gI(^qVhFwgu*atWcf$B)
zA|&}Jv=QLOw#&j3w5wWqxDSzlyCL_Ji4yRMWUdN~PQPlH{=(IbIsqK*XT5pH8#4Pc
ztHD6*(4+}3;X5mpc(;#*Lpdgc+=ykwG%Q<!RBCWn#qD@qS7{=S2jgGDyQqf9;TpPM
z1&2<dMl}Dyf%QVpmD%kDf?qXP&904r6fGl(Bn8i>m(onGmUlWviVIPsi+aBZ&rRS;
z&*qUuF{?066%&lU>2L#+Z2C;s0dC;=nHjHa9B4Gv+#4e|z%4Gx)WN^+h!`5#5Hds_
zr!MlaY0enAoR45BgekpYet_oRIPz8gD#}BP7?p48K7r;|kJFB(9^JdKfZ`9|@@!Gh
zyMf>Jvn1r+=U-{Us^(_ho;I>NV()juwq}zXpg@K#S*G>;quw}(c)2|nn{D&C&`{^*
zP|ync3aN~%zKdSdwx4C&{)}~BoMNhCXpzWH!R6Np`mO}bT9kA9tDibDZ)DX*n~27v
z9{%(><|p-WO|)7a^JLvo!Ydq6um_+Lnl%_8wUxMXKmAYDcbiXteK;1gnu0Ye-IAtZ
zvad`k!)-jNZLhH93_W`PH1Ht$@aG%-SzPVgH}UDaq`bL)3B)m(zkf<J%<!Cjzv35_
z?XW!;AbwtHg++N+%`XAIXd^Y|FOgvc6)8af0n`v4Cy91_Vb7iE8>|{@NPzG4iToZk
z@PLr$ut;*wi}Qu+_#i;Z9V}TfT)p|vQMQ|1Ctb8{uk7*P5%|q&)rUKE{WuqcY@S%{
zJGKEhZ?bIPwG_TRG%-oo{739-mftRYXHjUE-G<Uz--y$jNr1;;5s|-7<vu=We{ZDv
zBuiDS!t|9bqg#jh`}Yu5?d-cOW1Zna{^qXzDDO3)4soL+pNb&2!K0He)~2Wy?x!~g
z{Y3BNxL^E0f*Y<ebHZLGH9KT1+kK3~7Q+<pvA54F_X&RVT#LzEE0ZWw$^ZxqzhW?c
z9=7e1k-^+j(hdI5XvziA)w(ovG`k7POQ;)*tXW*yoQIbiP0v`u`o{qL`>w(Nd3`U>
zD4$B!Xl<+~gjFUk@6oB`P=;>Gppdo_FpMlk`9&TWNaK}72mbmrhqNrm@C}YhVx4$p
zcZN!T^u%!sXgU|#W*}%;+Nw5qZO?$+s9621Aqj`+fZ!w}aSp$gkoM|b6Q;d@z{8Lx
z_aRo^wA=?4jc>c7(aYoxU$4fqOK+F%_$yqOh)b4QSdVCn?Z!Co>Ra@UjSnhh{#h;f
zd)g}cvkAh5nJwmk$b&V+y9T((_qr~(6ZO|oLa4BzGuMV_!YJwO`YS!!2O#E1$0k25
zVjlDsu0E+fG%^_+dAp@i=^689B({jd3!fSJE&i;5s;5++oY|&ZS^Bgn-_1PY8BI)H
z@zR<^Xm=kqAAC|&T2=fyU-997j(*8i@9DqsvA(fJxo*oT20@ym)4Q>&#e?M9@?s15
zmdmL9NtxxkGAn4|_i)Kd1@gc{1A9gmHZ$*e4V7$16y~xDuYa&h`zKDnKs>XSk!e$B
zBt2mOH|Y*}#qtH3A8>ZOCNO24Fd?%&gjD%bxnW)OfP7O!SEs2k?_y96Q#FhN5p3un
z$fTKgVW}qcxLaL-A^xU-fQ*;^0RYZ;@!w*X>hw0?$h`G<d-+<9X9?uD_C5A;g^^u-
zEiaAU;q{RC=lgi<S7e^4nMK!Q1)Wvu8X%ABqY=CtB1qRBJ%6K(S5(S2iUM_GLt~*I
ztVlJ(Wd;djbTva*lQ;Y9ZEHEIRVk+>@q2}f=ZyM{9=!dV*u!zKLn;YcUAzmxeD0uK
zpmtGvqAB8Y)bUswS!bdRF25()+qsJzASGR^n}LI%!Xkd+I&J<0Vt&~2MPnGTUn)0R
zBM-sOFh&o<%G!lIIGMZD-MRWF7j2<_fclGCeMC@<D@QApAirHm)yWz7(@{(~zq7mP
z-l9i`(BigIfCGj-UahXgcIIX$t{ArgueUO&yoWLDC+=^hvIP8s2(7XBXpZ-6<OEIP
zWj<Y|g*}v5?u9uyS5`vPCHAvV2t#`)Lo>ID$`QMkur#>65A-NmGZftipxZfUW4!;#
z2F}U$R%rU0)_~$pJo{&Hd-$lqtO}O7fXXa>ujwYtD+N+|2o_FcQqdZ-WofE1g@sM;
zbn{q2IkQcZrpd&_Y)*gDqt(zs^3s?LG-96Vb&%Z0XOkg)n_ih57i+bcD<5#m;?Jbe
zKNpkWkeA#^!gUx9`fTaqZpOgj&lRavQvs^UX3Zxf$)RmU?X{M@*fDGDan>(ht#|Sb
zG=TGvAvr`{sPeH^8UwAPElV?8^$+Kk)nh2Lswu-w5jH#$^DkntR?D;(`$iUWQJtjZ
zI`~pN;t@SlA{B_+$s(pRSQ=l5u^>4CTT^6z2@;Po?zV=1pDP=A3=$AIB9iMH3X%|j
z)Z{InWG{GC=KC#(xuk>!>VCDhiTJFku4+8|S5f)u&LVDkfnuallJYq*-FBja#hc2#
z@?bpLZpJ@@pBJJGT&DMQED0C+XdmwonvU?gje|lS1jn#qsi0eTaZuz*?t%`rksDur
zagIx8Rcnjsxb)h+@V@;EQ-NOmxB91yf!y`JL-R~YLeI`W%a$|KOH`Z1zy3+9BYQZ-
z9!zD}M)s01RtM`(a0s21!=w}hOfNy*QjfQ>Duicb_7qM0fAQP@OT2{L`xMYe%)n{!
z-aW;gnVh9GosA8gO@xgcO`Zw>FE=mW8}4^+czIR1xrKQIg!zOxo(OL4SUA+~|1j9t
b8C#gT{r?#@*iR;(7ywWyd2pG8zW@IK(!g>;

literal 0
HcmV?d00001

diff --git a/docs_new/logos/smithsonian.png b/docs_new/logos/smithsonian.png
new file mode 100644
index 0000000000000000000000000000000000000000..0cd759abc48b8ea69b73b7d80ce4595c0b4ae1e1
GIT binary patch
literal 35574
zcmdqJg<DnK7cG3~ZlqgMknTo7kS+=7ln&`Wbf-u+h;(;%NP~2DclWoE_jkYhH(Y&q
zd^oZ9+H1`<=NMy-Ww5-gI0_;GA_xRR`79x-2m(Qofj|(3@UXxuGt@eKz(3I6L}Wxj
zpz=th2R#_zcQOMBMHvvtl^O&B`+-1rz)RpA5XgZU1lrRDfp`-@AY7Z21_gfL4Oo3C
zaZ%9o%ioOV-#>s)f<B9iC_B#`E;>0AP1eJnTqkQesDJ$&=R`FyE*TeYSr6iDzFu-G
zt)Yv@i~(hK{~i6XqbrSo{sHOz1txqmo|46arRgR`v1ZQC;*!PNBT`mWSoOKo!?Sz0
zG`Do!!@(4u?ZLrYV_6JzjIp@G({BG0LaV$wa$4Uum`TVyY~dVG0SriytZ%+d>>aoW
z<+<^VFfK@V3_635!aTD9C(GRJiK|Ke?}xuojl1o$Z&lbPf`e)_I~d3)=05OOxzlzD
zp~6ywcv0KbKtEgIsr7%oX~U%50<psRkm^+qwe$RPV4$R+fF%<ZjSV}}N(qoogywJz
z(8mSwLEy!#<XeF*q=W_G2AM`PD{W4LWO1l5Frd&WC?weneDKo{<j;kJh~Ra_L8hNQ
zRfs_?kQ>Aacc{aKH_<RMqFo@rprEzQrEXJ+T4GQi2ABjo!vvgeOAM~nTjQ{YSz4dQ
zc)c~YB~u=X%p_lX)OrRte%s1>b{cn(c3?1W^zsv9MCqRXCK-i3=!2+e^w~~Kxx4`@
z_HW3(Zvhr(lp07%=>Z9m`hSAz2Z3R|JcnWTMT0cnAq|KXKO{AQTtr~11;h-MbOo7<
zQ_CV06odyx$58MxR_4>&yJg;Z&Y;x|P#Z^X|8eqrA>ur+*w7Ijg9SuRk)`8i->(L3
z5ifFKNcoJc5}bt<^$ql_b&w|Z`edM$)j1YsNL5H~7;rmsFe1zewX|mEu6yvmJ6I+f
zJO9y3QvG0q3L<2qq?U%)_Sxe|V0Fi+^bHDvfX5hXyS`cba1%}l!2<_Q#mB_;3>pN_
zIuPetRR_$9{JVw8O5&3xV$KSra4B5W0P<&eL18Pgz-KE)3yJ^UPbk08&09dx?Golr
zjZtR8Zu`SH-9}5={UJ!xhi>z@q5|{Zvq);AO>MQ~)V=YcgW9c%_}+x%zw%}I3F<Ka
zhQh)d$@VH}vZzT^bgX?=B`7G%$o0~;3=RjiY6S{ZGmsR1SYeFT#HvID*J?>_{8w5J
zhMFP-71fiHBBTm`QkIqk%BD|Pp!<;Z+ef1|Ptg(IJ$va=LF`-?#^F-UZb=zYY@P?m
z52DzgBmWYCy04*lZqR9|ji3BJ<J*yq&hQs?h4luLz$UZ6p{bJgguI3gZ8^Wfj|lw9
zYtS7hILq6;pmQRQd}Jw%-;M3vgt@`@!H~b3&V^?dMCsSYgFa~}0c~yO3EBe0axRr>
zW3LO)y__xeZmf7&*p{UR+%MqCb7jytI1>h?KCT{cOWb+t-Yu5Ea|a(bpoTY~gae<O
zk$q?<t;3)4ChgzyC_G5(xWY@X6N1+DZR@a44MZE@+|ceY@3-6|5E9R$O{;o2LH!Tq
z)4G&%M^@Z~5Hrjy=QQpddnPYqDw1-8?7AY(ohIrJkrOo}hRejlP|tjquv$+vs)4>4
z&%+6A<Nl<<&GXE0z_x>Ah)10Ku&4~RT>WKr$iZ#Be6rWT#mU|cY<UJlx!;>$bbEG0
z(X`w|k2?5>m-vY<jjk1&8u_7cH$yGX9aU7P=dF~rxK{U<1f0lrI#JR7@O>c0{r7^C
z?Ctk|fFdHhS0nn};I>gbVH?rzVDrZpx2+$|IxE8i*9#hi^u@4o5UR+*Tb9&h113<Q
zszm(0M?KC5v(V2|I`TJY{PM1gJ<nEbY-Ni+sw|}fB>Q}O^rK-)EY)@4xG)$BCo?di
zLA2Otf9pnz)4Sf?fWYxds9m0+>7Qm(Mr=MlcB`$A%pX`EVz!yWl<MkD3MTO1)Cc|k
z_0Hh52Z%7%a7Se$h7k2DulWX3xmxLLV!3)R(scC-GI0@J!LBef8Ybk|jU9b$bBLF-
z6A!8h?OknN^uMuN{57-p`*|g?=k48tW$svn_@9yvtfK+Fc;T*bW;*j>>!<fBP-76f
z>Q1NDPnP;ya@<tU)y|V4#2Gi3Z2Z|S-mhCdBn6*?uLIs4wvG5WhaHxPyL!vI&b|BR
z<I~lkkJqZs8(dKa_}%w239&ph#@i$_(yFCG+~2w?CH`b?eRDS4b!?J<O)LJ#t9y%?
zcO>8@QOPM{6eSz2#pAojMU04R+vPITXar$be0!3C#4@5QhvujIn=lv_8F8jFh+BBJ
zxPu|2fY-poYZo8#W<Eg$iv%Y$B{$=)H9It3-Z!2jrne)wSo=8?+3y}WJciriBSkT$
zvhXY`m`ZWo*ZSHChB(<8iiA&}N(@0X+59)4qv*X(y&ZE9D<(&9E<v5)tpj|<vw9v0
z?IRmejYC`{2f3YRX{F7tAX1HQ!=e4RN6F{6#?LhxSA(LoG19_1ot=|`r;rr?vM=_B
zlX80<&<K=-%ta344qilj@}1YNwyrgzd6B4&R*L@EBc0d-_uT2Q@v|VWfoT@@X9DC0
zq)2DfHGk2635zyp#h`x}+=NIbpdRz!K9MQ%c+}M4MKQEfVIOwL#?`u`%V6BWTfFin
zHpi9g{yHo6i)~Uwt!h5_mWf}pnU9I~H4>7qO)js%r>Ni}L&+<?p~lX?Pu9CISJVeW
zJ4#Z$2?@6oTToH7UJ(7&#l7jFN4w`-kYaIFG>YYh9c}w5#G$VfeVofgf{JBLCIXjo
z{cCd2?{_0*-tv*)PIA<>P@hS6-o=oU?cBR&-l9{g$NJWb5n7qz{c9&JiO*G8{oZ}^
z&d$5=+Pb2m$ihKEo?{(a3ZU|4m<(ohIXwk87x~uYC?25=54pjViEr1`-um-{^$I+t
z%Y3BW`M0H4_l=xvP^rd{sIv;#ZOltO>!C<4nHc|aHDp@eXK}OZR){d^tqGHfm4G7b
zIf9Q%PRv?Gt{QJCHMfyHA)I0#sOmsP+5Fy8;(?vkZuoM4bIyI<{wsePtY1WG1rc_J
z`#s;aw>*wia*5M`f>(s?z1JGD+*q;?<j_sz=iQeDsmDT1RPEPo^AMT%Ddu8k6JbQF
z#F7;}sK|QJ{FM4R=X~x`f0h%pL*qsh1f?06w}({-t==?oeVjWg-fE5^UlTxxoN`B7
zLDhB_t@^cwi%v23U7*RHRM?7$_P0{X9Qxm5So~sg8C4AvYd=0qvDQ|4nPT6G7V089
zZnABxL3}&izIO2+Lb%pk$I(PEziyCnRgXuIqF&4A=Uk@u{^TPNA`Ijtek}N*zDMt{
zt3i}TUu?KMSro80ti4AYRMa04W310t%PZ(lu7WCkk`DK#N!0ByzYY!NZ5@}z-boz(
zfrWbRLYCuU68#s#RaDM;R`0r;0gNGWSg|PG+e%N9P2r%fDVsmKgr_La!dgtdvn=~3
zL>ojQpQ3%jiW9bo`wh)=Oc8y0g)~f~d0Xtg^+Xus0{_KuUx`B&EjVdJ@2-}vi7fbE
zO&6YywZ+Q7dRsq`G^}@{yU3WJ<-p8HaAr8Dy0IkKg`f8+$0Z9>+3&2HvU8_R=#N(*
z&7zlZX7kv%3qADq<Px+F84d4k((T(8LI=E_9H#bx7AT4qH;MMk|8s{co693~O`LG8
z<=Q*sK`TOkw?M7P=Lmj;pQhCagkb5xJGyre5@kA4nskMyr+rwq-=A7Ld5`z^8G8y_
z`p|w$#LF&lUU0gz+^J#z=3MD88VkX?Kpa%7;9`>gH%IK~0t)M3@81}iwxq4b1>TkX
zwb>sh$~3N|2TAPEOKb#?2_Y4Ps#7bC{V(pdDq2wXoQ(46KN6^s41yg3Es2*(_~{QA
zWtiBBeF*oMBAm645-tT=2<`9~on!)0AT(<p2>v?oWnUAitbgWj;7`Y;{TKHHCJhd}
z;BgZYeZ6lmRt3*OAOWo~D`HnZK@tawPr1{ghY56Ry%FlmfmK5ZVN`e))_d1ZKPtBT
zQqq9T@ytOzVA9F=cDQRQ@w|g~TTY{<b2qK^14ca+2T)-isyB??Irp5yWOX1x&R&lt
zH}`V}Sz2sTx_8IhFp~fFFl>x?90?*B^}ff#QECY3{3q*&W339~)xfd96|`Y|g<FCO
zhb?}?`MD2spE9_)ud_Y9z8TrJxA??01}+8>S9rqjyQ@0juH5Y1`)WJ0;58bJbfIOV
zwHpoha>gl0kzpyBSqGa$Rl2xh8vi9~NBa3M&{3eT8HXV@GS!GnG!D75uVn4_pQl`h
zn0al)`*k>78cFCn-$!up^qk7JwjwxKAe7xQk%{vzb=Lc)qm?G2w%1qJcFb$xMK|5w
z(oc$Oq=sSarpY`xepw`4-DL>Y68PeQJ`ull=0o)_IrK9QIl_Ts)JV>HYsTsdR~MEa
z?nl;<UKX`cbjw^q;g6Zr{9*b%-FM4<u6w7Kw3Ys0>9|H+=+PLb%tC{Z<Xto8ft`op
z<!9$+m-L{G`LtP`)!zkBZp6RO#Rh7ebuUfJOY*zX>w|F_UZJP)MEntA`-U^o-Q8Vi
zdER=h`Oq&ZnwN!bnc2ewN<Zp7xBpdjT#Zz`Y>N6x>Clsh;a8b-h(BA%tY2B)dF)a@
zUp10Pwi~g?-_4qN2|g)b_m-X0&rb!wNtSj7YNU?uH@qiK3a~S?_RY6-5!D6*#r&+}
zzvnecB4FOI7MA8ceg4##7-45_y=d{#v9V}2fn$$ZbJLZ%sFWO{S}=ba#{YTdV9jfF
zQ1No0{B3>B<&Cf1r<vQ^-#_EC`$@IYetFml_NI)%UAZiA{nZMCA?{qi)>Ju%kX8~*
z|F`!Q#_0-p2tz)fN_YHzKfk)lL1TS&EL2+YIGLBt%xIAL%fQ|wY4ajny0ksBp7fI-
z=(f9tW~14gNJC2Pwl2efJ59~4X!Fp{;(da)BgYb<t!w%OwAZq>KOaGESN+<b8rrjz
z9;AeSK|}SU7YEsAB|uc-&rkZ2wu+K9K`*u!_{n!?qp51oq}W9|{^ap=q5{#Hd1KvN
zvS4KdTeJij#fGGf$cJWqs;yfu^%sICU3WMEipA)pGWl^20_0h}nv$fks;cooAHHKL
z@Pp<R97&>j#}-0hQ+-gJN>)#3IN0*^ihs9Q*K%X!&bHSVHPx>+im&aIS<JS*%H6Ax
z{GbtX)A_Z~;KQAJjwDxz5<!{Lm6P7D(p)>dG)sQ|jY9x-I2RotHVLdT#O6+2-rvs?
z=C0je6^Q1y%NouwZu?9B+l?{p_Hr0xxKxTUOQuUua>W9?J7C?dytg*{qw~?m_Q3aU
z4_6~q>zX&FWTr6jlUva7KLu_+{=(P7ORont8ha5s&KVc+uxylK!fknm&3=JfJ!ogT
zx}9AAhQvp2W3VU|{3Y^Dy#A-^AGqwVk!r864MWB@7UM7*QaH9kO7!;yv%V<_JRQn=
zd!LF}TQVO!SCXmstC+nSl?!LN`n0vU2>fI|Z#3M)I9u5<7pP<D%-Yw60y;Ywavpz#
zYtnMT3HbFFt~xv)cKB+6WJ)cGw%_N^o2HD{fT2ikslpB#scL*w-yi7|8_4yq8Uirk
zjq6{9y8N3`4^I@AY?tG*38%3rnsO->xi<7+&_y_e#8f0+^cI3S>2c{iio;^N2Q+ub
zso+)w59l{;6VW)!k&Xv%i^(jf2z^yrBz~xrvmVw9WzQ0&`i=sw5|vpBm4RD-vkXCQ
zX%G)8VRQv8M+mxCdbjVloZxO(l7qM3F?Q&>DXBq?KI4K1KPmhvD7(Ut*8QSu-NCjQ
zzc#t5^ifb-W}8{F`V1{tGkYSlM(DJIIt$`5&lm@N4179zmRaVY3V_57L!4hl9SbaQ
zWFt#NXEi{44F~kswH(FBG$7BSxA~OcXR#J@9$a{oS>Yb=>=c%%nxVA?TzWHbTkJC&
zXIl(^^B5*`tXkLNS95%ya>YS@tU*5Gb*l7ISA(@H`7%H8<TmWPL}QH!w=t24vV)II
z7UL;hPar^-iyu$!->7QLO))}f?z-YFhw~uw89tVMHh50^@=c7=17L9^jwh#f^4PK`
z<#nCSvn$!7OF`T8j2%*)BemG>ptg1M2&n*oo{WRepPjw`o5AtQquvd}&qvGCB)-?v
z$7U|~W4lr(w|VRTlJ&64Wml3xdDZ}jZpcM)4FUl1PhG6_ttOwi1&?J0?o0L^*U;!o
zB+~}bQzGHEjL$#N{7cHg9qHd0LE;sZs^kd=w@AP?i@VnQ2=cNSZcAZ-(~lI<$(|hS
zI#$u-vw@77l7WgMr!zm*@Ah|YS);(7BkY{Jetw?}t3#GQbvxiIaBiW!^9r^xnld<p
zG*c}4OAZM_%vGKQ4<3=_Gy>4XQ+ygFtvWP@+&&q8DiUQbHR$Sj_5ff4i}m0WH|OzT
z+4Vhu&B^7Uqsa93sS<MCtzMtM&b&kr1m?s4L6T3Wi^6isx4nL&cR}uKKZKw*r=E1t
z{dJxLTa2#oWbTB{Irl*kdnUPKX`8@{gi-rxyDmj8qL?aA41-Hb-a9MF!=)Wdp`wxv
zDTsF0)V{~FPEp0+chNC3uLm8X5&=N3vpJ%?C!Zh(tw{Y#HEMeZytLJNnJ3$Tdbf5z
z76@Ct=i>D#$L-{M_w9|%qb#kI7JhX5#wnd*yY{Tk+Eiw(M$&r=J4P)}VigD6uFVwG
z;YaxQ-bE)Z@|R(Du93I`7*uSsol>Ixhu#FZiuU;6ScwD1-YFk8MC}fB-)H|~wyn$C
z9$pZkpoS2_#B@p+ZZc|sj%8e=+s}Ow0RG9weZy{^ZNf#>I#XMScvhQz1JHv`fc2t&
z<67K_Q`A&DG7850xAB}46g=Kwa|JdM5<+r&^|2+W*hl^<5e(&90Me<kdzB}RRIeO{
zPz$~TWzBF(FKf7ugSUn9kQ<X#rQhb<l)>#yYBXai-r0KXDhe1^Got^EQ<-uwt-38T
zJO&jP*6+Z^(3gl<>YI3BnQ^`NVG`FXLT}}|FT0&I_O_Rc9aEIt%++)hOwJi~T#Ph-
z(+Yk&LBcbvq7ZvXE?U+P{)5k;@XGMeBL7;Tw5zNh=$jParib&R0+I*o!3I3`^qa-d
z7pPibAK3KhWNBdgt<x~h+|L6TJb%SqP91S;+mc_QCYsg5@v7-S$3nKsYIOG(2KCp=
zn)_gN%<@l|^fvzdrYqV&lhyyi@tV|KIKNxGC%1O<Sfo;$QE#sZj{<1X5A$cpaHU8C
z?==9|bKO>O_I+r+IgGHRDYJjgn??B~gu`1-KNrFZ!nzec&$aNRuU8gFbmm?BBDb8*
zH07SQqnBp9M#q?*(!!~w3x)}43|I_%OLS~~c8o!%0Rtq%7cmiAi7ieVTHe&*ynHv5
z+2zEJ@Oi$Jli{GKO?=;}18(23J<>9J%rrOPXBWV=S?vB|a2K!qjH<aWGQxae83WwX
zI<;2eZUn<&1Lbh|#m)^6teU(Vr^^qPE=~%Rs4SRhs&cw?SPHCbykw%WDbhR)DcT&W
zI0Dch=SCMf{+cW2rs*a^V7n(jP>VgxxQ*fhH+Ykh@C<7K?;#Topslp;&m`h}OWoZA
zPek_X=W_JVL2rqV+kaLh8Ci<N=dPs8&7-&F`rb(?HqTg(e#`r#=2wm%Dj!>cI9tOB
z5_Tg)p>GH+Ybu0)8EB-2W7`_4lSD5^R~OxF>1se$YkO)<*7WkAIVncs4|W?oFsb^S
zXsiV{3KIs^Z}Eaw*s|uIGHhX}{JT$3C589IIQGbBX35dJ9Cg++><8)R`J0rr(5$wA
z#8c;Wxog5TasT)jd|SFcdnTneh4WB<q_`mO#R3A-e;A!_dN+^^^^ync$%hnGmIa_+
zhLl*m&Qj^XALoRtXB2O4P-gZKv}sBYpe2Yaq6?0QAveNqI+Wza<$@zf?4es;MXdwD
z-q;W=V3aS~GVNS06A|)fQ*gIc4&9n~ZNL+RTE<Q4x$T1X_rhPL<gFHepSs-yjf(^^
zh?Azat(QyqkpL_($col}aHTL~^<>`1`PJ6)^pU%Er9NDTheC`zE>|P1<<2mr_8?b+
z>MiK45&SUmRG;E&o*ujC8&?9|iZ`KpX)bBJB)FU)p`yp4%k9<vYH@?Rq#xyZ*-g{S
z$~lakDFwb!rd5>-zALJPj@mbvpMEiQ>quFB$`mM{+tL{E!V;<pkJw<8Yl{uRRXY=n
zmF9*|4T^~(LLbR)ov~+@&1+#guwn-!4b3fgWTe8Y_S$T;UJkKq1NR<CJbkMN83$E5
ztGw1AF_N4>)pQ>rUeb1@qh%V+lhXq0X9+k`@E>mm2MWoC-`thppqSyelN_LhQ`w#`
zC2+RqqkL~$bacqzbgC_U6D`+)K@$x$gq~L2vjTvu{NR2Rvrp7=dNfa_m?*-LjTmdI
z_03tv6&XW&YzPK}$LrpCANijCQ(G%5i&aJlE<Ixx3jkC!^W<<D8Zpp{1*v%Xj^ij^
zQXnIc<aKcwC?pv@pBaN=u>E!SDw4|!8o`?%k~s{9g2#{qL1dE)iS@6jFH6imUQ}}U
zs~3r9>zY{{(&ZkY?b&!9hATw)8R2fd0<{or=n8|ZUpu7X;%kczjg=-QZ<SCMG`+}W
z3+wl}LdmTVrlBzu(w*KLLts#ao0?_R854txM5VqOeSN>BOdMptYsh1{isKNw{$Rwl
z+Ih*>y%$xSM?cbDr6ZmGNi??Y2F5QaM%>oHhys=rRF+eTT>Dq2jdjk)V}cW6`|naf
zfy;3>E1+{S#4_4dLI=^R&j8v<EW2orhMGICu~mNS3;qm!v=cN^M-eV;Kq{-ML~apA
zO@ibxLEO*}7a|2x|4owv1%E2lsg!0d4~Z~`)T?MM7PD6;dc;EjMxmKn2(|u19Q>#1
zYn&>k#`;L-H-J_A09r&w1DtB3EJQ8BvXF4wcxWqM;)wp?*ANT(!gpgzGaDn_Eul}|
z*WxJw7`r4m>X_<>BfYIjs2#vhPw#$hWc$i#5bU;rn$PPlS_@}TrtJKSU#;~KKxEse
zM?q}Nf(8%+@7sqgh$HV+(<8Q%QN?~ICgqg#{85h)jb->3tNGEJvT-@CjY#F|KJm84
z1#Ie82z$RHqPbh{JS|#RzVKIH?QEb3x-x^vvnX9turR26-Eg<b7N=6eJ`4RhWjuOU
zkAGD4hM^Z40hLL^7A4J`ADcId#CP$~*GBIPl2k$vCUGkmUE#Y#HPs>ElBzqL?%yV~
zlg7>Iwt20-iZCqym2!?=gTbcj6N_~{D0*N$s@S=t68{k-F<DbqnwAOa?#064o-!Vt
zl@ltqlZs<d5t3o0vVox%8DTG@2m1EZQV6xzQ5%0hIJP_|m=@m#ofm@&?jsW1hf;;s
zxf9oOL4f6w&Y-u|4f@Zp?7PM0IF|!D$U*EQv0~TASww~k8&<FCa-`Grq2MS}Ogn?+
z>+Ya4Lk`#h^P~d$C7>C@VO4r|bPe*%2%d{0e^wv(9^u1~jdC2ujWy&ga+7aBX<Quv
ztnQB!(*Z}QBB>Nsp!@XF$ijGQEK=bK(O+Jz3+#I7!4%)%L?Jyu#m4(lo?t{^0Fcxf
zDsvpi{v`j<zz1TGJdjl@=X&&2=D7#Wh(kv-sIypBB*x7}5!L>@F7@}~!W<LLUy@us
zMhsmgvus?jgxTZq9q2#A62KvfDyyRlXtpLtfrQV+4Tae0rRg`VZ8??<b3+$hLik^R
zA^pp`-^1zqrK%X!91(O2FZP~2X8V$Sf#%QKFAeJ2pNl9Q@ZTZEtycM{f-b`5)&DA#
zlat|xj(n}@M6=Asvi@Z+A}AI4x<*B|M#f!@NGf7`tOvY1^Wp71HTZVXdiG@&uy?+S
zkib*o8L;A5BTIgTOuQ*KU<amY{^pmVwwqDY?<<*arvEyTSwS#lF(G)KAUavo8FZ5i
zWXxq?P5=hR00#*I#2U^Bn%mhC-vuF!R$@Wx#31XhG#Txr6Sk|S4EX;oR>)tI!z4CE
z1<Hr-QTHSIynR|PEy5SdBoQL99s<0>0K?YzI6q~(GYYS4Z3JM47tNPKJX-&kTQds8
z0ic)24tc$q|MnOJY}YNS?#WWPdM8svPZLsKpv0XTpKCjpx^S68Eo9&PjQS@&Te^r!
ztKmUof>8kVGZ;{-dZ+NuZ#I1~KNThhlg_>6j#?B3*S`u=6u;S0!Zm<70Dy;}d1BS+
zvZuh6L2uD9%gj<&N^a2orE&|f?yJdD2s`7keZ-=jjh|lSf6KXsy^sr%{l>tR%Po(W
zcqV*4+@39U>`(L8f*aaUZZxd7x-7zHMM$re?r^gFV#3Cj<*uyvzbDM2$AQb=cg!l8
zt!JUVz0~(@H1rC!WZkkgvYYF}?Q5?sUO#c!r+kUF(yz7cW?uYL(CqmCO<s_KeUCvn
zSL+;etq1g0O=O#rc_i#;B`kFB4~s-s4X};XeGv5LRQqzEOAPcw6}skB)Z!(=M|C1;
zY-gtQ6sL#|k{;hv8xOamstaX|F7<tv(r+cHJyjoYyvekzs>UzS_oX*_h1l@0Wab9-
zn67#5-bC&Iodka3!=JDAPbudO2S4VH4Od9Z1K(vS{1?rJ4H9*()UQbs?`O5d27C_g
zV>nG>!er&1Ysigyi!CCjr*`JUz?5ZqPkQ{ubkP5{WVAtv2tTa1=;&6@kCzPz#Jxda
z5vh&*QEwmXoZ14$@HLZ0Z8z?FtW%kNXm_Aq&^Ks8ok$y@NG1Ol^p-|N{(E4);y7Vy
zLG4zk7L&xFc_}L+++q{{9ZIlPWovie1;Ii5IX=MQanh@$I|BEeA|@=pDCnOc!M~0<
z1w7S>m+)I9x&e7a`2m&B*kj~-gvtY=|3nL{>^*@+@-5brb)V9X3g42&5@BR152|&^
zJW?*T3k+6e8kLM_f1552kXYGE{<4kA_5&1|bpzpndrKFV)HaI3lW5l&?YSf+Y&ILM
z!`8)`>XswsYo{VrR!*H33&7Us09JWpHV4W<^_$?~-fpck+96Je;z>737S=vL3<MN5
znI4E;VxXROu$M%Ae%&hD(MeQbRJkR>68%=7DQG$7F~f1M5%<lf0N}*2d#$Uj@A;_Z
zvfd!ERO@2{`5^|PQ5Qo2vwG!6CE=8=+rP(25ONf1;#)(8B}FLV{snOy!4K8D)to9V
zXXn^Tk(JXfvLhPHkBtOz@Z-c56FZwNPIu&>)V6A!_#p9|!venSAx5IWe!V^<)HZJ0
zk-zyW&rsrwB97p7O5G<sWwRJQC0Lq&Y#9m!+RTU+g33AXJAt(blhg=Ft!V88h7n5N
z4+s|qMyeS#=Cp$9tsYr4PMlI<uPdvm>&WjKL7=a{ERw!H(^Ok7HN}7|S16V=*1a|Z
znG;0H6sHq(h|?5cq}7c;lx%=175e(cq#+?VZ)LGC>QQuB**{*yT&T@RZ}JBco;C7b
zc%98I2e)(odT^2`R(_(Ly5j!|a4%2x)JFrhnW>)@1lB5PzTf^Z8J&o5(_Xi7Qy-@S
zI*($?k(m+QYnI$us5Tb6M+fxsmEJM9^^ApQ)>hMOKtTnzkOO3{G<WfT6G8BEB)k6p
zkj<-0{H<(9RBZL2`J5jRSJ$YGf)MLyk14-C{#(+lZuab{neV@g+X&wdi_{c{-%2<0
zG9JGtT6OjcMf~#*&WyzkH6Fl$-bXJLqd@Q<wjldo_4x4_`DY9b>dsypR11-&7cG;}
zH)<*f3}+jucBIx&D9E}I6Il&|#a?e>Jg4m`+!B0YHA4UJ?L&$DlCkm1Q9vco_++^E
z+?Y)|4g9H^X*i7A_w+<_P`{(D*?flfLayvhzIH{h%`Ew;(&thPNCYt%>8v}izKegY
zU<JxTVH-z4bLz(KQ~<##$!=a;cez_`9h@Sk7u5g#I8e(K2UYr~RaH{rjq@XFUAG$q
zs`eQ-qF0Lldf{;|(34qF)tKo`isB(<LTLrze-ckAg}DV&$-NHRYu}YG#*Ggki++2q
zlUBLwa}5J!hYwsfHp5;l)CMe}{1WS|IBRkxa9-!f@4cm)0Kgxieuw^N%RRu<Azrsy
zNT|qwIO-e$=%gU=)WmV8mEUqo0rVi8`9&?sLp1(d1TBvpyyyYQz0!jZ`Nzg-1OR{b
z<~h`8pHPSN&rni|wm#3KoD7YLb(QW*iNrqDkQFy*^A1!=^|@C`^@e@ceR(-CI|AN3
zT_cN8roGFhbR7V19#4#bu>MT%jtHnv!rsmOXQy%YajUU=ji)m6eo-~^?}`ZuA8$>N
zxr;~|NPY|ieh^jPd!5udW94V!5VnrRa3ESPu^hn}*PIJ$s9Z-{9@N@qyy$=?cekM0
zV6UHCR6j$04tKz5LhU$^HhUcZ^QRVLQ==7E(^3=Kl2$F>+>kv;U@^Fbz(Kjy8{_f_
zkkV#co?6j`pBIY(_4sCBl#dw@t>YRS914jeOL!o>0K`4ecpni;KVf$2^)apEMcN)*
zxXw@^?e-C8sn=)yFZ*^=g7SGF3P$`JlRJq?SE^T)JkO60jd=lCgo;GlBV>~W3R28v
z?>n-LrSRthT;DJpd$V_QgUCe(j(VQGHte@{Zi`ol)17q=JBJM0%n!ER={Da!o*bSN
zSGM#5#si>Tk-ur+SV&A-yo_MYdF)?L{C4$xJLBU7Geb2h7VFk}+k#w~_|-HzEu)St
z)saa8W)J>ec>?PLRos8v#vc#{-%=U80hew4ethAwU%pRJKfe%WIL0SF<J!oOf&7;q
z0?=6seg<t*PFv*~bUf6S)pZhIaH@YIeCg~D(*{{JdWyv(z1{Y#I}MG8v6L7g@|n;o
zW-z6Y<RtQw746^cD8K<(1M@&+@JRi<2yh0$&?k)Al}>c5$T+S3mA8<rVW@+&j}yj?
z)0eq`{Fs==$6SBKw4>stJpKlZ^3C~)K=uSe_<3l>_g@z4UP6jqT1~8W()AQKT9Dmo
zYJcvsnq^b3z^LfWD7DpI75cbzG{!-*QbrZ2-<<g7lwGUl!OXqR^EdOgzMJ<isegM%
zUuI%iIH54B^rjoU`eGw#^|<djw3k%cw=wUxWSE9;EZ$H7SgL>1V&lGJF9}jwX9#?8
z(UB_QllAOnozrCE>XDhiSBv8^!~e7*nfF${eXX6d-At{H58ZAensVP<cPH@-%T#4@
zP=E$g^99iH@&wu}#5hep0FEb7A;57|(Tx-vS@)j&Ew&b`-EP!5r(W?I<RsEA>?uR7
zrm5f_FjRW2Y*$fDxh-G#1d?`-cRgt(fA&V(LZq#0`#@YxF;Lmsy1jaBC(-;g#er)A
ztyV7Gg$1V*pidUfgWsMFh;EX`DS&w<Mq?z)4+$cytRWOms<b;R;`>7L-a%c853rNY
zdyXRAlbQ7IH38pJ#owQw8-w&}=f*NkF&-)Y;@M=F9W|af;s#Ww1F&q(rWbC8=bkU@
zvrgCtx3}YAjP|C8fc&W|(re?s*ub&$wHCL<Vz%dwm$Z4xiJ>d@Y)sSrwcp-7@s>js
z<XSj6G%ykOR%-5Ei-+&4c2ONVP-2OrJrd1U0O5S?Dj+|PwDqGR?mmF!K4mszz->Np
z>EFBO?+O!XJnFK38z5!*wI-{5AsRdQ;tGmM-&uyR*H~=IwNqR0iOeQG{={(elBpJP
zn!y7qnoz2L_Hvk}CJ?pTUnF&)(`@b7-Xl-^<65@sj1RQ|ojH9C<~6IXS$-FT%3`Ln
zuG*YUw)NUg?_81taW0H8!^{u58LV^rMw@c<ld66=m7lG4u5_^uOT9+bzCOf?eq7SI
z+YWMc3eDWM@!Wi;ee$}JEME(2p(F}CPhWJztk{#eALn!8y4$zU9h>we3x;})FRr&B
z{h``h8@K-AJYX?TezDoOUp@kECM5H80A?)+9P}6#ZFsHw%l?a9%<helS#<XVhW<sN
zZKV{2lbFpzz-6V=v~4gCQ@RdhvDVxBt3)Bxge-yV6)KBV9yHL2{zN+-*D-O^8XsAa
zN1DosWW^Bg6=QDfa~o3XzkNl8$?=qg+&@@Ttp~&{dMK&IR+Y6sF>aGjp~@3Rs_~xn
zQyODxLM%t`_dCy9Dk{xlW8;~7bXrfYjnnYw!$ZcdF)7lH&m7l&mmC|b<K|J8=28(2
z?{DatZ2r-oSQF~_W+<}Ndy+*(cen*p`VUJ$Bf75qi9$N*`B-o|dxmBk$&g@bIzM??
z0sl(5S_lsD_w)k?#l&b@!t<SVL!rR5GhV?~b08p)*Hn!FAzD{{Q$h(}bSg0X{uzwR
z{SRPCvQr@9O<?i_^BH*sA|a)tx-<gSO7Axe(|SY0a;B8h-NWH(#K~NN_OtGw1u^MW
z{V*Ut#CyQ4w0!yI*IcT++Y=pe4b)3OPI*}cZj1hGU{L~^mP3)ABLH|Nb>dn)ZM%r}
z`cQemLm}!KQ&7Lx;yrbqeZ$;^?3xUex%7R;?ylQXxN1Zv01NglTz5!rhm=qZS|AC5
zZfh)wKr3fIRrP@kHgnI%a4nSlI8Qj_X3)`8`=T*q3P5*vb^e;BJn{+mv1X}XlhK}e
z4JL(PTIEu?t{g2@cP2pTL%F+t;o;*CUh4sOp<~5YD+Id{&$<$S=E}xz9rmX%5QIIK
z#D6D-<X3YAF1|4@O<}r6#-6%4xA{IKz-_kMN$x;Sm|hO@vKNd=xSZ{pe7F;N4Y9?)
ztJWV*j3}r~lf&voxW4lZDZfCT8QXUHgo&0f61|Clx2|mB4$z2hI4QKsp1Am?U%Jt!
z9KHV4t@gWJCslKvUe^WIUdlrrxN@Ze?{++NG~7u)<<cz7(WUE~b-VqqJE<5o5Jk4{
zGeZUL{Zs0v1qqnpUJ9{e`~C|adN4!U<;G&&Hh4~v>y$LNoR((*3}GBk{K`&0zb?ti
zY?mz$%5xMyDFR4LZI>{$1+Z3e0M5F_^XZJ<fE+0r61wI;XM=_cx*ct#&cv_%B}10_
z2YS%6v|rA{*e!fbf5QB>H7`4gYEcs;-^erWR&`{VeTuo(wf`dcsdw%zm*da78i$n4
z$BQX{bD-LxmGE*DU7$!6qwQW>vmb{S)E}=Mn@1jA*SaC!5p1J+3bO2#{`tXOyGE$p
z_quS2(26CXQGyZ5FiD+|{gC6sb1t1Ki%`q0X7BlCB@FP;xGgY@(CqUil+q!jcBOFf
zXQm+dY{IFnn&w5m>Sgt&yBI6`fTnP!sF&=uzCslV7&jHtTbqHtbeZb!NHC6Y$rj9i
zE$yM(PL&U~8Vgos`fCc;KPBRq2!||<VHpAS%TeP35|Gzx-HE){S}IKSydG~<Cq&f#
zDm=B4QG|K#B;UJ^uaH`}ETV!WelQNqw=6c%^bC2x0q{q!5^9aO#sSdet8S(=CX-nv
zwMPEu6UZKJI|Oa@u8-p42!;b@mdm@^7iY}W`Eu7X!EiwS>SBl-le0@@FM<k)Q&aW2
zQReBeajHD7r7yuuS7m;y(ORJFj(uPk*~#c~tBp*$<Pix~?Sj&*G2rSh+2CxcJvxFr
z{85ELw7`~&N%M8@6)}>GAz*Ej_dty=n$p`yof0>5W-zIH5?^UvFyJ`VbqWrTDl;81
z)KeWi@|W#ZExY@xTN~?%xQX@Me+7ii{h^|vr^kUf%krv;InZ^(5jh53I?!zmv9()n
z?<|zty)^H8?93&ft`VH7o`k;8O!>0ZjuiizEi2F82|}<NBMHD`Pe_%1swuYo;q`1i
z5P>Ij{eGcwZ?LS@74Q``r7_$?!Okal%#AEuVbp!!+-!cbGzI|fHLmn@F_3k>O74f0
zWCi`4aAiZ<W@^N^XSnONHhv&T+UqX<XdUSw<AJUe=+%Q1_wTMK(8ODtAeF8)`ENK)
zRsHsL`$G(xWe1-QnoknTa%&!|3$@U6(^Hr_=EMOT5Y6<0uOnOYbgM=p0)0aJe(B8>
znoBFt9S1Bq7)+&+Q^$U%cFP@)g|f?(%qgrl>VOPlS!8BHtX#V4|DRPUBvib}m+l9~
zW{Fx!?BtDk{}AEx{Y2bh*<lmc$*6I|otlI65+BfLlvBIk8OH{SGZ_FK=MJn}%1>Y@
z6D*BuWOpi)RjfTNNQuul+s!!GY{$+_XDr8%)MjQU4vShpLsLj(aw91Pa-Dy8wbDTK
zyKEi{+vu>Po$Bs7^i^^h04geAX=*1yO7Dyrt_UJAxE~~a=_26ifBl!;unLd60hD%p
z?o>H1?$2&%6`y;7Kl(6O%IQAYAR}S=RAkhcry9-lwWhn~T)TI%(3Ki)1|S%!Zv^lT
zJD<Yo(_YhpcY%Nn4|sNL=Ca(ydY!hn(fPLvT3G5m<BPu|OGf~d0Wcj^`rj0tWrDJ^
z^IAd8IL}z6H>mLM<P#S;febfdx{R1=Y!(&%7)<`_et)XiKv4`p^$%9J95WuP=P!VS
zUK1_*oq?@AYVHjJo#y!~rbW@jjog8Tq=7Pt+^{;g6kK4xDCM5F<3s&=CfbU$q)sl+
zujQ$(hP3HX=SgEB^!&wrw!(6B-mikBZ*Xd*(PGFN&GPUEpg49_6}T_)VgN`5>bis%
z3oTxocnbIT&<^OraM<;g>cTeAz1{j?qfO~lFE@-k^KDG%q-)L~JMIAMe7OxUS*g!C
zh6|y6b3W<<#C?qKHrSH#7?NdY17c|Rl~d_~d&7DxrSAYL^HjGcce5TMq*p4UBj$YU
zN}h|F-iZ_wRKxBFv_mYTsa5&?@h2NaY^^9AafkPo+R<sBHYzxAR_l8@ZBEaG09L*{
zezjZ!BZQ^G%%=NjVXAs!U;u3RPfdIpv>%8|$GyD&d`!?h<pwO3bC{=qCS_Sa2lWiM
zl58lk7cskRpxM(ARii$s1%4YZ+Y0~>HB>>6E+E><(?``NB_>R3<6n4A!&HnFw>QHT
z9y1b2a2%Ug-Fx-^&-rQkl0PQbK}E6`IV5rS=~14~2J<E_Erq77d8PvcINeS_seF7o
zYOHw&t&3#3Ec3LG;aNMU!z^#61x0CNF{1Y0i588t{LXj?S9oxZsO%P_dZo|P;rx)B
z{<IiVR*ktyR|Ki_bb)+>&SMeF3jCR#5vt$Z6zT*w;t%nn6@T%61-HgfDPY8J1)0oN
zquziG7v{3@x#l3qpnz0Vii|C*7tPq_w$>5{GK#DfoP)2`uRS0glooQnpmX-085df*
z<OUa7m0EvXu?v9lrOV6jf&Mav6p2V<BFQN`G#ggjPbRVfp9e`LL!F-smr@(o3H%R6
z+9v<mit+sxAXc74rBrH)tCyGU3yFVEJ)rDE&MJKV=&)TT*eC_mRKR*caOT;H{M|7>
z?>#ggQh}4f_uIv$wpR)G64NhM2e~~vI5r2Bt$2BvzwFN5eV&)m%v4??{r(xKpTA7M
z>sT+*<Q*a&$N<}q_jd2c%ICkT4{5kKuiDWNJDymyh9^nJ2$f_E&<qf4{8F8~TQ|hV
zvA4TZ82a<)kt#@jf(MzjWS%Z>nFIju-zTYl#+O%IX-Z@L_&-<S(}o<)oq%YHrv1EU
zK6%2GbRED2yKi{Gf&K^pi@0mP;-~QpPp)4l>N%25G0*@iP;;ZmJ>1-ec|CBG=cPpB
z&^hLbtc<C@Vxv<S-8$CfBt1Sltr}s|;N>yJiU3$n$B&QyD5~n4tqT2kdn564oy9?s
zi|M~`Xy<DwKyo=+CwFFD`s`x8Xa&GKpb^aS`J+}X7N9~kKe=huvK4vglpVkp?HDb$
zUd3lR=_=oiLzl3hF^oDR55GHJfq8YZ_kBR8fGA=R;#-26ShiR~0y;TB2@z9O7#ttd
z;9@9T7xFYM7D5YT^e6&ALPZ#76VDnEU~0!IEuke5NxfJ0ynQ{=gq>!pzqW8jC%=GE
zN(l%~&fgDNnxBT0gVlLm_YpZOE!{=09)b8Ls@gbyEQ!ztG~aY<2L@`Rq0c!h>p0SM
zLXl9z|1}Q-f&c)wiJpQ)#%Nt<4TvH@DP4YW%WP&bHr+Wi7pH6iKOBStg>~Nt_e80;
zPtZ#>>1!ZU4KHkM`~tzm`LRD~)HzG_npqGn6ZU^+E;#DfSm041${42ttxnK-(p5B{
zlD$Y%%PEabZ!TR)AB)lSyB{JXU$WoAmH-++NQ}MWDLh{QTwM4v3?L>1Lwy$%3BZ`h
z#C?x00IBqcbj$zscoS`sNZ&p903w5PD$o-!j<zPUm;qU<`{1Tk`_$d^n`x_7ZV})r
zu-*JKz+$!pSLZf4djj*F^L`!xvOK;zZ8F_n-7UU?kuslX3o;wnsL{Maae>aSiJ!W#
zsNvWnJz~_{B{CRF-(?1%4}h_&7cgEzf&iQ?&$>Mz*oq@ME;PW@{`s~P3Iuo)k%7XD
zb!m_rMA~RP(gisBm??&Q0U8Y*0=WJXmRX3bOXn*`sud}SJepA6UMi>X6-DDhyl#;h
z#W#Z|ny5SPgH~0I8li1Io@&*NUou07D48Ag2oa)1_LqBYyAuIi6B=~$7*c4f+R39K
zZ3^YexA(4OP1o*qvQIB1E?3d{QN=_U&>Ahb<N#?sHBb6H!)(co<rc95=sdbtOd^25
zi@1W30F0Z*lyv4$9_au$K}7}DE6qY?D@c|1fQ<6h_t*ib7zhlET|r1CZY%&e-U9-C
zZ+*rnHr^6unnlyCO!u~v+*j*=4IZkvG)8q?C|5#omA7kBWiJu5zivkpgaB+KR+Lco
zum@bDCI*IBL@F?~*dhYvA`8CwlcW&N$RVGhPT9ZJL@XQfE<x-N=B0(^zp+!x4-TRT
zojAe75bFd;v6^}dlxQxb3BC=4#QaTAF*r?k4pfgBahA&B!0n4hkwyDr5y&!0UZo36
zLW7#|%19aS8cfdswU`s~fe|H_1IV}67ec<Oe0`m#e&Co*r*7Xeh&b)wcg(v*5SUR0
zAD*fbI;4=fdQD74@%3X+0~gjxWH7iOR{^l5v4m#zhf7)eB(L`Fx4enOuL%~m1kOye
zY@Y*N8aC<=a&W#ZC{oyljarqRneV?28V`vMD>5K_%S(5I(rUOw?4g`d9p~9ov{Z%;
zUUzU63>-!En-jPaNSk+1(Lwv56L!e&NW)Cm|GUQ62+3Dg=m1U7$0xK9NMi}LBywlW
zr`UPhr?Q`esIOzK;sK7Xjm|n!D-Xj}h2Ue3XInDr3c?TF@crMX7HN$c65%HjAlv&v
z6^Z1{d*Bs^JVBrSH_x+W6aufuM<@n>Q=Go+^QX)<6xeQ%WtOWvn4lAK=#gxJ6ADI>
z@`#_W?KTXtk0WZ)24vqM!CJK$x|E<5CE>Kd#b?Z6lJcaV-v3&17~SL91Lqh!OK@#1
z=mOADnNSDFsvl2ugaw6hpTdy_ni=e0M;{BX3={d+3JUrP6sD!OGb~>hbHK0%;SLc!
zLUzV7On2{X!23F%3Ka=0&fLiJjPkPtqqd+PD7+Z5d~*;pT$=?*NEk{-cQ5D{1N-Y$
z*`o&Y96vY^{l!CtvP$@h>P_gwN}Gb=E=UWFBL5fiD-;Vq$PAr1{(}SR%SpqI|IUa4
z2VmOYh9c_9(FwOPfK9Q5qfjNNk#pMpgjZ(JwQqEV^|G)=Ica%y6U_D4raG3Jag;ul
z7+up*{J;K3@)PF^1jKkpKk<G1`_&Lptm?wpnnn*YIR!-s8XsU3f-Ml~ehW80c;pM?
znpbm!28x1WDTO$pvkMG*?a`{G$bM64j4m&fDX;x-<vm-C0{WQ9A^N(RI=-l>{amy(
zPleW%1Gzat^5KR2O#}Pt!C|~6NoHgc{-U~ht7S1!Cx1Gu#$OJ6p!`96VA@~Y4tfa*
zF~?s8#{DcQL}?&-Vc%gr2xkhr30lFo+MvJ2Hlor|lwN1ITH~mnUgt<VtRMbIQBihb
zyg^6`NlIhM5R?>(l%OD!wvbI+bbMt^!SRwU$}jj3Z76LhA}rYXpzoj@s1v9Y%bQke
zkf4IIOf;4Lj(#UQE>|w0)at2y*$wcNFJWr2VN<_|PDB?1^`#t2+F_@(QSo9a+zlP9
zN9aWX7KXrL)@vJsz!E$9CI2W1F8xcbbED$A1%fY=2#$Hk&DjB!$!C6q=r@i4#*M;&
zajbS|(@4D2mB?%h3$4+6q+BINmJa1CV{tW2sLH6`zD6Z!)!p5lxKGvGwZ$cl#QaIj
z5yD=`kLCPlkoJVKjvJd@d{7W=KU9?FH&cIzdPOl=*r+$#=ZeBDP$vZAmxAfN>jnx~
zD9AiiRJQ@ko-AN<UA-dsmt)#lH;9c$QyG~bkq*3Lh2GiZLz)I-sqKn_u!dyap;okT
zIJzfR+B-v_9Ph`#M<wsN-!;gRop-}GEhyI>*lo$5Ockq*!Rvl4skEpM;{QWLp7lP6
zTk~i!lN*!bEp`%<ut0`Rke*w4eu*o7Wl&5OMPOxl=cXW3NF%w8U)iiHs1o!Sab+<9
z)xVQH%91vKo*7Ztic(e{W}9Q$Ew?g{ju8mPv#$?a32oiuZ2cds1#5!ST5DPy1=$$7
z+@dBPJ@T{!ZeQ$J3RVicq6+IGpn2;pnniT}Sj%&rdj0MY$1OZ6Rw*6fiqy`)neJyi
z@JC-HJ$n>ge$#@Mx{`J_Eb#0y&TYv!OKzXHJWpn1=zBaP2IG){&a5hSxhXhp)po{5
zO=>r2>DvwkWlz7-mMyF;O?W0eWEUyrT4P!CG2t!RcYKKNMnHW#>)+_)H5b-ukGV~D
zMt!GCHS6Pe#Dm{$I0&DmyjefMydjAC=zDYfHBU3e5-s+wiNi~6dgYck0kDx1G!Jau
z?wKE-xd`wjgb;VTa=qgUO<Zd;w#Og&U>~31`QAqO@{PCx57ZWVP;hA?k|=xBCify=
z-XEcR1VK+zH#f3Pe4An}0$K<O+30)FoAYyBf=ft#>$%6b9w;2w{`Yh?TE5%+)TP_P
zv=Pu{yg9Iltzdbg4c>ua<rOeG*FfM}>0t9pK89MqCRvvEf-D=d^l>(l_3`zm6>w8l
z#g7D!*b4{=E8E+rEax`@=Jpp<WTVgKXRfjf%xdiV?kH)~7LL!HpcaJHclMPDbW@@m
zayZF<#Wzv$Ia7G>D3u5j5k%SLc|sNgKeDusPl#@Q-y)-O0Ik{7U!yDOpin@f(Qa%|
zTOUG=NuJETp;wuEG`)ws;qH^s%qx;MP*5t`;0cU`(d7tYM((T;vSaDH={(ZueHp9U
zk4_9d3jskd6C(t+=onT|KYa-;%n60hj$0lDF1wq#c93J%bdh@9(vch=vJa^*Pj8BG
zWvrM?d1Ju?U_91@(O|*yHIVv@z>FJY0wN?G?ty6K2~n5m#|76d?wY}BL(1cIwWOJp
zl)6x=Am~&YueyG;2!z;iii2a}M)lM0L#}Q4%*n>WeHF(Jq$>6(o(|Wf8EHW|+FQt|
z1;={WsLzkbpHy?8D;Ogj)(H3G7*)seMRG=lvU1fFSS(g8m<r4jLfb#fXpI22RagcD
z+qKkXE9(+tMny#!GMC{HTMB}|AZm<txkA+u%ivxF_UY*s=x+K)O)zkBo+V7!-cc<X
z3JG)9o~YYso?CO%X$$A4Tt6LMzg!++f*vhB5eCXkZ}3hEGbh?jbW2pd#OF8Qyo75=
z2?Tvo4w*}ik)GQ(?M8_97pF*Ep-K8w6H}Lh;7c4*E>EEYYwADs8$9$q?Okr*CvD~5
z^Nql%dcYlBxD~#($;|oe?y*SWsqlLPL70f~ngn!(Tu=|qk4#pwNan5`*(=Xqe@m|h
z7$)^|OG-*m+Lnt1EH=t<j${083<dM*-077DuR-{Oic}OK&A{UGL7F3#5csq*!_nnK
zq6?RxxS|y>QFjDGOV6izK!30T@{wXv%`(XbgJo+~k;cI^^@lq<MEXxN(vR{WT&W*H
zv{uR}0;Iwk5`+*V98+#fdRGeL&}6;=Dq+y}#~C#^3|3Zo#h?AK)N5ULF(IC+6NPZ0
zu$0smu<21yP>LyfR>{M*^8)DtX*#CS#1&i;_1Kn=>#cz7n&$3My&U={SOktdi_{Qb
zYnyqS78qjZUpA0L6b5h{nipSTp*c}#_Iqk!V0j%~T+&`aJo)4Tej65OP|)+PL9f#K
zSmVpNLDNKvs-^<sWF$ydOGY%7dE`ex(zISD<D-4A8Pvrm9gjm1!UjkzDu#P&!c$kp
z6BJLaBng^{nU&i=tKXR*na{XP5Aop?hbVrlnQ7M)A+*f`Pq$!(ltA;YEF1KZyg^dA
zg{|AUyuRgbUR1*TY4i^!+sZS%FppG35N$b7h&(}MKRUnE(C89D?Qny9$<$!6NN1t$
zI+8f5F9+!)8$gs!ePyee?1IoI)GIs!P%C%Ziu=a4N=-#N>do+ND8WfBMdX{q!fnqj
zpf?QfDYE>230fbDg9P<e<XPvdkm80~oM;J?^>46_{;)%hfOae`L5_6fdtx<qH&NIh
zg|Kz8?STK{B1oSA3_p1IaZXLfEh|<2ykJub;<hASs9t?Rv=vcM^QXV#=`&m#-WDtX
z54!n5-jWibVRTW^3BP}pUY|t*fq#L7T05^NzG5ld!^#C*kX8^vg2}h`o|*>%-!tg;
z6Vvn8jy38#f7uk69Z_TGm&=SZHOeD^JJ4&LXx2V+Of%h=eB)$LN8R~daA(TP+hS~s
zWl2x~$Ft4}d{|)vDypNSpvbC<fSf}Ij1fFn0c7W7dm3|Nbu8e0jP6Ay$cQNmq<L%K
zKY!Sb-@d$t;g1cr&%JH@u&4&M7niyo9{ltVHX_o2)NA#-dFlD&`%eG=*C1BlMsfB~
zp}F-=R*GR+=!0JX25l?|8oj8hEV9YUZCcrw>h-!f;DiKoBO1u&ck^G=`wjrj0w~`r
zQ0O~yIM+YEO&>KTq-N67(&6(0wZ~_N#7JL~)#1m$`}ae3ys7FM`X0JYJvqiB!0+?p
z17&q}+-%e8Usv(=mUwIP^c8{P+2yWk`sM#4>8j(R=-T$~(%m5)B8?!OF0jNy2uOE#
zr!))FA)s`Jba!`32uOEGNlQw9!~6Zc``ej4bIu)C+*4UsPuEn><0wC#+}<9!67=<J
z?qcOg@7J$iquC!#m^nCpvH_PES(cs|hPILjt~d-OX~4haoxf94FTh?v*?rSM)L2DS
zI8e$_B&4@<df(2g^tbginSDarIWAIAaASy;x$p#eV^sD?pvcLyU;*KtG_r$Gh$1ex
zTc?PclY-+Dd5$}}Hs)}cNy_uz+1V#lE%wW^Lqh|V+YhC?{_q-Zn0zji6v;0k1_bQn
zWyETVyeOjc3ao$4Tv*Q`L(^Q72a=U_CU4`r|M>q|!RwYaphFuUCjTf#jO>BprU}wz
z<s^lITR~kN1lY;s=$_!2@L`dZU6Rt>q4Bb$x0JH4;*U$&6sArdXX1{HM_z}2vEt@}
zAwog8DIm-g5Der7#!mL3gfl?cG~%-n|1~p@USlP({h>hhdiN5|=;5oFDa&QU;aKyo
zHNCKI$1qC&`hUkx%_9%!+VtIYD*DC9#H3)~#pY6vH{OG0*kjN!ZLSXtoCV_v(=cZS
zco==HX_K&Mle`B%L<n!vg;$w_rB7`VEX9~Js%D0!@DPVTsWew!Qb8SloBqDO_BHNV
zHeHY_WWSx&<d)ntBt(62^*ZTsAOS(miAG5Iz^R1@N)K;fNihdyho@!pcqfme<RBF2
zpD=k)m%b^4Uctr1*5j_amxi}XmUo&)I}0OJIO{1QT>^<A(%d-}prlBXnn?dddmP^e
zpqjcamLBeUh<(`)#9e+^a0OOj)to-g=tm&hy*DK{=bkD*NCo`_bK05O4$fqfYtHIm
zLu*45gjF#Wz?XKxtwJ3}Tna4{RI-ovX}lQoL)*xISSa&t^vZ+2fwqISe)vM#jYz$3
zmhP5rf9O%WM9DFtM^EPppmm3GM7Q?tRZtTD5Gu0hZcJ_7v%~8U`hLW}4O&3p?o-rs
zwXxBCO_o7^qD`xqMVmLzGdsUV;o^n$icb%7`48v=&aKe>`jn;kCH3`5{g8=!ht$K}
zc-7d{%!w+I09sroXrXtt2to<*Wr#~|o%PHuF#c0&<{h}<lz7a+!dSY23i`=;(?m5H
zE48i2J>c8vwOCtd7H}<?`Lq1P4*`w963EeImo45N|Ev!ie;fuzS#8@<MG@4w8RQeX
z;BiY8RD|&L?((ix%J!?!{&3Y8m+GSof&fDI$--nC!^%U9%;YA&4vuP14G0^8JCiH{
zCWg3ljc4DP)vjr}W}5FnGO*(7z=Y2ar(oTijVo47U}as<!!#p}vzYTxzze(c$4_Ez
z5`T&9JV!M5ygK$0KN2F~f`F~J4!Vrvyk4|o!nU?M^BssoaB5z02%%t{GzHtu>t^X<
z{qJ|k_@3SOQ_%(6k!Xp;@Hz^A=-%VzT}t(HefJFtU#kpylxRlf8wwt&V=jC$zOsqs
z*Cu1}8Xywx;lWYgV7k03pmza>=MVca69&i+XDGKH*q%`&VuEGAOm}=nePZhl#4B!k
zZyPkM-^sR2`Qv=|y{&8u_dubh_>~>BHeXBd0o{w*Vj~asWHNKHCG!@<`3kC%fIW@2
ze;7J!8grHPDIVtN^}l6}o|A-a3qWyBD!8m00`-(XzMG-0e$HkpgrL^xW;dd4gWbpi
ztZhmiJ|sPidtg(WkX%5!MX=)@y=@U*%Y(kY|8B=PZIOX6`x_iQZEu%Zx-y@U%g-RG
z6rpFSE!Z!hkuzj(Mg0^y3wmAEY@&&klAOC&5RPY&Q;l27&%-+-M*hQR(vhmr>%&b1
ziznaMV<Xz-KcIL?*EJfW{X)C*dv+25QNY7Bk)<5;y_99Prtug(#DD?nZMGcp?-$Q@
z`IS2_27SXsR+C}sJi;yme1Se=JqA5>>8RpXQ({dkOe3jyu_Fvk5T-zZ7?_`pA1?b+
zCQm@lv=GV1@EpzLitL@5aUQIYX@IK{%uMPc3J65Lsu||IDj!sDFnoQ&Zf_!IFqJl{
z?riRc5NnwDZA(cJD@&K)Zvj615R^f29hE)li!+t(d(}`y=f7%kZ%<g%-BlIoe7PI+
zrJ8J5UA^Q>Lq!1OtSEf1!=#{FpsFPIE+)7gv#j*vMcSNWQDF@ZO&CI}eZb94Fg6tE
z(I^hvlmdqfK!Z|uoh)>Fe6#buX;#O5B9;SMs$-wMgRoO)){%388wPTPx}U6fCykWA
zl_z!FkIuWarD8;y?ta2BTYLL|j&3tJ!)Dp@7dL@a&X;bkTsI-Zosc6eBUz^9Fnstn
zk>-(2+AbzoTIz2?8i?Zf!uyH*<jc99IST8xILLg;ZXD9Qws)mRecH<Dc~-sU(B?-X
z>264Lgl?n|UWD%f`l_qu5nU`2i|Q5=@djacLHe||W%G!h3@mGGtnSfNVXu<_*<Evs
z*!T*wW7=+$D^uXg3N(>GO-}_0;hEh;bW(YAdQUu|xn_TAqx_Yt$-KQqibB=Jr;GW!
z7q)+(r>B>(NG7|!i0f(?a)hzZhq`HR%Y?Q4Kx;0X8af72CK|Vl05$P@iE0kKA5KsC
zs$1_ZGP9tW(*~&(mN=@-2nWj=7)VtX9!e=cz7Jm4ugUfZT-&~wGa#9`U}tk{YqhJ@
zRZ55AAb)?Ia4d-mece>-TU)#qan?z9Wyzj1blcuQ9nzL>Kvh}S8h9QC>YyE47{uN{
zlTlgZu(lW~cqwxc=oSFLVaee`FzGi$A+)&vl-;9XH&U+7&dwIPG*@qU$K_v-O<4D=
z(1Y634_V$<Eb3kaB~qGUF@IXZ>d*}&F#W3m?yg^<`=sOAu~$l}j^qOr4m0oyIVk{{
z!HbLV&ymfU6!7~r%+9fHsPJf+he8diUnZ7qu5@jY9Jx8<)1gM-WDBK^{iyy|*+4}*
zFtLVNrEQ+5EtW@71nKz3c&;)?Gv_>VoU_xR+|@Hsd$qV!ZrrfVKUm^gZJ2BSPdb%v
zhuTrl$VfeGKQ9TmMPkak9EqSzhfbo%8>D~I_9-|7iY6)kh(}&5X*fnAMkk<Q58|NI
zzcy*v06xJ8X*?P2`$J82Ao|u(_Lh{$^=l;86#Ro7cTG%qYAN|Tf|R(qxsHMSjpCc_
zcd+M0Ms&USJ+@Ito8rt!Qg^S>(XwGF<U#B145n_<5Vex>OeUM}&!>3F5B7|18~7n!
zXoNK>SW2LlIhF07e`kICNup|3IEgKFp^T-&FcrIXEX-%abQ+<ZltZ=(+<3;^R$I*T
z22k85U?a<fEsy@^9lKCk+h$i0h2$i~)WaKEQuM}y|1%8-jpm{WG*qUQ`054;2u^M!
z>c7#Lr2GFRw9(XN1K7kj<%0wOUm@tu1f83%mnTNO^!`#^;#-xm8&`iv3928Ez85Ju
z<&Z(Ql#JQ#0;@Vbx|}`{^t`3HUZK0e%4?=Wj>+gGTTQ+r6S?W{=s9K+mVr_I;eV_4
zY`k!!$velczjmrx_T9{?+p!C&+nRPo^5)V@lzNDj+C@(8oWZC|iI5Sc<%eGQN)mJC
z+u7OK2Bf-x^mUJ~5`~wD47nO4o=Dy<uh*onUAufV;PkjBnicDMQGQ*ypD^7WW8h*~
z!wUi{-%rlHZ2hUPJVNS@?7~He6S)C}HZUKB;%CF2q?5kVZ(a_EGcVDAFFG4s^|dgB
zC!BgqJxkbf*rOl#hkH^sYu+c?Rzp$mwi`E8CnJHIJp*cvm(DrRNC$CR$ARqJFl_YE
z5@_vZbw0lOt%>ImHtN@rHaO&r3YZOvO5g4Qn1x7jnEU3L9PmXjeCRj6?<J4j>6o*+
z9m!Q6h&e(-*vA6<SK905I5Gs22gAJAh1HYCE&`*^tdZ{rc>gUZ2T*d{Wc~G<+7gq5
zSzKn49B3x~=nP$&4nhOS?&C-zRzHJ@S8t*zL2T&nv{rZ>ZaIN8yV(>#t#+1sR<a?I
zJ<q+SX?Wqr*@FTq*R9E`0n#yi;&2f;jvJ(px8*mCG<@$RWv(excM@sj_Wg70)6yI4
zqI-PAGj-n}9MI$_h15=w)Z<rvv+l}^xV>_LnN6L00>m3Tv44}i=pwHtKaa!hVcoLq
zn4~2kd<gl*8^u@zC2Sbn%=6nVBYAvDRpb_h5Z!@Vzm#JNv3`|pe|4B;ztF!0r_=vX
z&LjkEZ|xcbT!GU+rI&mkZTKOTq>vCaO+o2h+C0MHa6Z?G)w`R&RO`aEO|5)HvG<)2
zKmjR4RVf@!2nyp5S-KQ(n`sAeN(?>&F?c;NLifgvaiuogSh!y;KzS7gi@^Y8P?{jz
zT7`sOamxOg^IlKeaYs9cQ_QpP9>S9tWYLzu57V0-v46cpI#c#!oChjx(qb<!Xw*ZT
zU0fx{qcpzM{L~e7ot(@SUWPEg%`ei`Fhp76;^+yR_CxG{kcs<z(#3DrNFRjN&)0Y9
zWt-*Fy!tGS0?aCOLtS1{+j$dK=1Xqac${EnsKoHn51Z79CW0o!pjUwpTbs{Vb#_ex
zx00;#3SG|zM0LQR9YxfWdBV)O_VUK2lEQhA7HC5q6fowVNX<Uk1tN8z{Vx=vw40Rl
zg#vY2M>uliOXyvZq`6bZw|D(VjqkBH0BoHG1mQ#|>f!VcKN@5FSkkQvJWq90|Gf)u
zk^j_AM9XZXI#kp=Y9mEZr5xVBw?{h5L{Ex3(COX}O%W`8rv{&@ZrA8>JUVeo%${4O
z`xX-7pxH#JB+i?1!cz#rkg9<qM&+J~$CZ^z(hgcdo21PKIij5jD_Vv%W83m>jCx&A
z&*P0UUDE;F0_nbyX<pmlExu0E0^B&g&ENcr4ioeex>dQJL!MAPeg+Qn=Lumb$1=B~
zze&SvFz=%xMQlivbT60q6fxOKDzzV%4(h*Tjqf}5unM~sf%~rwkLZsZ*!B!f%mKJ4
zvAX)b;9O%n;&86tgWC|+yw+G@_PA@JesOmPob+)W`Rqf|l)FN;p~zhw?Y_vHi9ekm
zkpj^E7P>{}&t`wyb)!lwH|3xN_PF>}Ak83U1<F~)b@_8OVUWA>HW_DFu^a=UFwHwv
zdYpHP7moX>Q}2Xg&=aKt7x715?dIJgGfeD=O<?|W{Cue$3Wol?o<-?!a?b(MoQbpZ
z-*kJ6gGbQM+PM0jU(EB6W;Si^CZJ)iQ&c?(9BC1M@D(InkKF&*U5unafF8#D&Q9{)
z$p^L(QiFay;ohob<!~lK?Gkb6BMCeS?EP;lBNHE>RHzHB4iTHs)@t-F_35E*Oa=Jx
zYlH>cKFg-+2NC4&mLS!eNBHI6MNLiA?NreAj{92L>Euvk%k4m4SSp!-G~u2UO(xc&
zN5Jh{&A}DUzFC5OpSda_r~zK$r!k)AJ--gWVIV8b@?6M*Na(`<MBUd2P4BzNlp0Uk
zKJ9AmcCU$QV43QMw4l)V82t_kN_x8&&H-Cc0=cLbW8fj%4QfOX!nG(7AOkpWwhjFh
zOhri|WsHVxa}@*^xPPPljLQmJiR1jhZq!Y&zOT>s)*|M8^8Mc@sv~sI^l25U62a^L
z)#fpCO0)|?cn5k}Wk-;Z73!5Jq+6xNj{9n>3gRQDBtyh09|`{-wEyS^vfaB6W?gmZ
z5|es9H<1y_9%to%SEaF5|0DFFhj5cVib($t6S=jg#F5yZ5%U+7pYheU_gvQ06efRU
zi=M07YRv-gn+3OS99LyeMP}x;H>Pe2$&U_LM3bm~(io-^ojwZX$1r3Yg1_)e0!B@u
zmE!GKnmu0QS6P4Ns$y!%X<cGM6x%TSG+xEL)j?ulkQ_gmu~!?j2QTX9w!utNgR_U=
zb}7V9i&#C9j!U$jIFXu|>5uPYSd%YTGwW*Kjn{}d!6UOs0-1emeGMjtJD2j4AaPmg
zwf7oGmEi`n$J(gFyGm_>iyAd!?nxS`(NQ!hm{Fetuv9-?N2P>rys~?`*iv9}xaRdV
zklCL`SaOLPnTY7Vw6nZx&tmjpseINz<@nQqT|&)ifN<##RK*b?mvWWUe(k7KP}0wH
zh<5>sYfC65T9lYTSR=*MbLhASA;ejs(166mC(^HG46>rLS<Q4@C9eW-2X$|<8XSK;
zLx!Jw?lfzQfE#|rg(eo-+9VCCo17pq2zjVk)OEFlA>3|l_9<|m9q&T;@I)I*I<HfU
zQK+R(F4X=S#qBhxsg57#t!DOZvUJUjL?@uP&mLV-;MshpNm#b+obHh9?jFb!pd#+N
zi`aO8lht7&KD8J}?^IPFT>gL?ix|1v|38OVfALnllXop+9!P|nHSJAA2?>~3Zp%ME
zZ+t8>lrbTK18E3bWmpx};>a?XIM<yp@jd#iE1QU^WAwmGI|Bp5tBTdee6{TxP-DZ>
z8<vJi;(f9Iq#RG#b5x81!Cuf-J0)H|4sFYl2k+K3u}}jN8SK{D%#*SzFQF|kDu{k!
z$Nxj%?>rro$V8is!QH!{p>Gw>{{K%RDY80-AbJ|)4%zM657DXozKPeOU@odQ2?o$$
zBxQ#R@z|=z&<tblD{av7C7EaV-`8+*`Wof&(9j>ks9x+bF&6jjS!mA1*o{DP6a)XX
z1bvMI%bwm)#mlF6@V=!DTJ>|$-x)%<`#!*Lyc#-gFm;+<F3aPVd7?=pP`mB}H%_qM
z-zC{>-4_!_oUq@O>ZJsp^Ju=;)2-AAg5m(VIN$d+qx~<PtxKn-$^!$NECG0vrV?w)
zTpcAnt_F5w*`7aj=(XP?n4vsae8jO2+;T7*Hl*hzwvn9QWt?;*>02WxE$vGuw=os_
z5oRiTi%5D$4BtY;gkpx3*^$xABQ9pyp^}!E&d;;!5JdPqdv`qwW7y`+-nM6byu3hS
z|9T9TiNEJy;%IDslmT5Iu%CgpC&*j-LLbMLs-I?E;Z!EhB;3H0y{1*qq6x6%^Acse
zS~1J>DjQH%5jk?I(9h6S6GMvVO6rjwzZ$i$*UA}@SVm!d5c;Q_BS{!=PGlAnH8WCv
zsG<3m4|ACc-IKl06(zl7{f5jJbq#01P910)UFjR-EKFB_?f5qwBjqe@aSZMWh7Fk_
zCqnDyUGd;vIsHYR3`VC}lR#TdLMJl!65fMy2c1O?ktCDwQU57@x&E_9q?sV$q=oxl
zpDYGo;k@r~<WF5k%3xOv5+95QP*w4Imju0aqk)P_Ie*lH4<6SRU2J3n#I_(S#%iD>
zd!ts_*w}D%ceo)5-&2LW>z!Bjv472IVr5hO$cUK%vc<rR40pB>aZu`1=Y1c!bt5an
z2b7eEH1o)>e4#n~(saOB0)XtQm%Ng|NGIMCUR@sG%!U9RCIhIva^&f7BcCW}o!3Xy
zl)rqMINoyZC1m;{rOHHYiAJ?8ROT#De&i$_fgeEYnl{|P2GQbyi4g~Gz$Nsv>K-^>
zM_^&4$hOm_T7MR-0k#I~?$6^$GqM*3iUhBW-*}_xb5a^OW2fwf0yl>LH#?TRW>2Uo
zNNXg_-(oZTKv9MUr^Y>dBR6cgp%ZyxYKr3X+yzA3q4xR%vWnwn@xdQ)-cE)QRS^u1
zqhldW@fiqi(R8Z$!kA`Y4vBQxwRus$Tg~v`Aj2V|cjo}Qzh!4xJ<F2yB`=`}qLF(_
zn>I!weOh}Kfo9b;%sl?1VdkDXh?qp1FE(~QrnJNo?kF)v@sAt$GpO)s$%XH&B{aZ9
zDnq>)f7Fe>IQfreU0Qp|WOG^Iu5)I14X|db;e#X@Ev{`4=BIBGnb^Lj`9m<-dD(Do
zq45I;fT;vI`9+V|wMRZJZ@L8RGjS}xK_laNqxK%BE^Te^g7^r3yr&4*SV}P7hmcsr
z_i5ne*Fenu1anRzTbG-_VbQRx|JAWI*UCFKAs*pV0onU^VCxV$&;x|haoh(h<7wxX
zpT8oR=4WV7Xynu^HO^<}XCkkKHY2!Sur9KpBc68-(pLH+j?qIrP2?IB{)@eZuL?;g
za%$hPw=!qUVhuBDG!bvuaz)z%ecSVwrv@!2jGXt{vgWfJJ?2v3I7r*0!DwO`C-k^m
zgeRrv$+BI_$6jxPWZ2F2vTA<TiA&!URsW-Nm%M$R-Mc!XJWe&82l)BoAbIvT+~2!@
zjwI9`TzD5oZ1{p|TJcViR6E_9S7LF1>`02HFz|0`tI(d#yL3X=$C+)gD}3DFl#~P^
z$CGz?cinpRpg`UvzUph}GN`=HY4#+zw6vM+a(i?%Yt+NP*PuB0C1aU*!n%qL%xkV?
z*uy;l)<JnyOeJz4cx_4dQv;#yzdEcZ9Uw!{6Ox2f|M&>3Y^>)_4ERDv%j>WtWy9*P
zA>Nfdy3q7N-FIe6AL%k5!>XQ-6QwJ(hT&lH>&W$4V+O0pIRx!WEa#}X1n9xVl&qXa
z<g@zaWk2dbvzFyz=;C>Kf-zLR@E^deLoeSXF2G#<jz>YAb;?{GPC}XDN`q!zicy8$
z6Mb*YJ_UD4m@`?6jyg86yAUQCA}A;{Cl-UDIT~V;Z;4brA_o-SzPBycZKZ&eMTv`@
z0B9z8#UDl=`!OF8F=#e=he%vjQ=wm3vs!3YljwiEWy{1{>z$!#_a=Um02ZhV+6^@b
ziz(*y7K$#P5?8t^(ZM|yv2oDSsG#fmvs-3CT32^j2!MlxBE%!y%CtB`rQpd!QGi^f
zQp;{=zTMd@*P__&n6ov`p9i@0-*mhcu9Fkr?v^yc!6F(e0}FfxJ2+CNTPqU0GZDrH
z(Q}wZ<V&=<{*-KE0OE;%Pgno$N#iF#c6#y5{ii;Q!fu!R>q~Ub&}wKy@tTn1IR7u}
zXp#YXuCWxh;<bIJid&8$^-fXz7`fr;cai%9ie)wyL$rF{KSng(m<gHif9`+VVQ0_n
zYDZxUqi34xcVy5*>u7^Veu&m^!Q{r3=rb6kLB^=*<)M036dH)$mk(}lVBe#wuhh&e
zj6c6=p4Us?8Bni-_}{ZnRV=lR^zw24CY&iL*vsDk*Sso(`Cp1<MDV{YZ8d{0HpV*p
zcUh58DT?z7+8KKY-z^K3x!uxZl}@UNRP+WT7TUe>uNk$~m6?rH8&vFCA*TTol`p(T
z{wGW>>jWc=fTeI2@}mx7pVo<uv09DD6eHu}x3;7Ntech=68%=nXeQ0lzuF#8!OKHM
z)=iHh8Afq&oK9r;gP!GL*3{4Z-OvM^fJe3|R5tIX_~p#X?re#%VjMbNA0@4uZXn~*
z{_EY^Ey8Buh+UgE>PrIjH(1k{q97GJqN>ZZG=T|xqILqgU*q-NmMR@pOXeEGx37Nq
z$ejt!e`TAOndoQ3cQIlldze26?-;i!ziH1e5u)n;Rx?UhBkAm16#v%`I3lJ&Je&_d
zKKA%h#VWC1UJy(fJImElvz%%*(`@lM$o{@42*;lZ^*^EgNzCXW^J&<173CHe+x{3C
zPmqDwp1s=>vHMWDW#Ns`;9iBO=ttp+!#GzkGuj^^jrKIo#9{APgg(xzU&W&Lkiz$=
zl>;xJG3{klu2lGdX176ANB@s?pi~PtGBiNnh57V+SRV;$K@XmZQNoIl1ev^^z4OTP
z<iA}0hP*n<ok#NVqSjndtZGj|wI&~AICK^&(o6WrGr!XbnbV~Fq7e8>`XjHvK(7u_
zp3}hbBC8PNrM)s?*8?yw=JL>%<1&Kffhp=k{+sQl(jowF9fAqR`mUvszm$b9$7ybE
z{%7n)r9C?LHh7U4d@&we0{!EOMch@6?sNwTm-62wF&d?;{^!A*ri^xVoBWmcdpoCC
ziDE|MKq&Lg=fht`=h^#3RvJwpc}d%$l+Rm}RI;Ts`n1G@Y{<{O{4DVgt;iGeiUt!u
zS>7Ramy!}y=)-j{^iKBV<mBFM%{8B)d@b)Op+%u#k&V@TD%G-%#*Eaxj^qU88m`Ys
zV&@Hnf5fezouK;GXpET>AMKUB$0AHKQ$>x-W>8nyD5mn$xG$LERCh&W+G1%Ue58WI
z=z>w#4>yJO{c|+oj~FGL0fq1F7ha<L(E2Ve`$IJZoc`TGmd?o~eAy`XtW>)H++z@r
zYP%(*u<q`{cP0DnDyvdyGY4e@*YFPZxDF>&DZ17ipOqTRYH<k_>d9O_8;mU&Hj}d^
zzVZg;g&TImE|=RXL#jN<GC?qPTgi1f5BKJgQ^lVfIF{NA`*p#>I9vuz2D>g4Z+o%4
zjDw#<YCf@MGm_U%Xc`J@tG~%_l42qf<K0=<4K<(TvKjj)(H8pp`k0j4`>;T&?=Eq6
z7S%nYnr<Hlq!JJ{`*JL|D>_yz>S7l-bv0=2u5g7HH=o&-rm?XR3l}7wBnt4mKxX<$
zEh`d!8wj8A8J{@ErE(n$yryKM46&kYXZJ&}HE%P+U6bdPFWp4H_DOUF>^Gs!pI?wu
z>#S<PDbJ(AaDFGEhDXG8*&)HWQ(T{?SClV`-%=K!3xb1ukUS04s`ykN`qDLL+o{{4
z==wphfx%yg&v;;(nIHT?4vqsy+}^)O7+BNDL{z2hbrJBRp6clpyB^<1bUGQgFIJ2%
z5p_w+O!)p?r7fH@`cJi&$NR{_b$HpzZEdchveX)GR5QT%bV^xS8?cv@w96pbN$!6i
z58Z#OG}EiOWwl%lDxc+T?#E1#IhR8Oq21nEYVKToPu_E>YuZIJ?AZGXD6skqLzi>`
z<s9C$^=F0%ixjy^c|=+R`W{1SuCBN=vZN;f@29XyX>}!?Mnk^eSQv3^>PSx3j33ph
zZz=uy)fZ!jSx4I?p?+RZ*L!$gPi;s)74>&W90iO1VEM|{`6I!5NW5RUoyCo*xWpWN
z&t8(eq8q<K1U%9ucLCg+uzj$W06VTIM)Giu+6E|E|376nN2PJU!LYhmqr_ATF${S2
zvUi)6>cfwlIcRgM07Df}!?8oHa7K$g=<CBo-Mw2<?kfA#uk-qLJQ$DkCGpN9&(Ymp
z{{meaHe_`*trbpYdNM#*kk``)uIx9EXx<w>kFTYX_#tWM=Qb0D|4g?BV~?=1w0*tW
zZ>oKKGSg$K%4CrFL2-j%h4IKoOxc;3MGf-B4-`JG0TByEj#vOPwGjg5z6<>VQ^Z^)
zY6tV3Ez-F>>RzTY{4WUL*Ey>>z!JM<zwZ7!7Ev6&#QT-h@73PV<8xlm6BXVe8C&0Q
zmcpH8mPS*3x;(_mc-aUpLQN3fpWZk~qN|iy1n^h90olMhOrQ%++NNWZ16hMrKugH(
z>}<NW=etP_LK9X3ht6tRIBwEe{xQORcV%7vEACMn#~Lm5TT7EP-e7_~jDK6eN{J29
zuL%$5n*!uI=#v4k>qWY-$dprIJ#+4hjFWHh2<E>+@3|@}UOPf!B6KXuG=0c$rK`Y!
z2{pT_wa8T1$@l*ycH=-O#wd(I_*|Vr6J`DjL*i#T!JkCM=CwWEL}iBWYhwNSuq*`<
zQCYZnO$30TI&D1j>Q4Y|a2$TsQewm^&ph9M_wnj*8ojJtA)$6o^&=A108yKeo_SRn
zq6LBgrEl&S=rx&6yS&EzfvAjwmZGSMsU)I{=!!!nhB)jJLs4;doO4H(#^##{Cc*yk
zKD455Co3u2FqNQg1D({`S{Th4uX16ChBxT~mT+jMvQ<{D;MzQ*^r?Adx7gYreK=P(
z?<$N54IwU@cYb~hm)8SQEkY+ui0SUO0V5X0n=14zv+-Bs{UYtg=sgTRv*d2@DFGhA
zJs>OC=Qz?(PekvwH&-R7<*Sib(9mA}*aBr`s0OG+fw1XEen3=%4$Ve1v!hO2M9KNq
zG6;ITl}1*&x9}K9nY%f5R}*oeeF=kg&8{dvf^)(^R$ory%{Xa3HU2xs0O%E!!y(b;
z+yW4Bn++)^f%iBqnqT6Up&cu8`Ei1>xQv06gM-Q~VOZS*H<2qJV)FG0BcClV{4hSy
z1**x7@(K<cNaGznj0^0*jsHVe;|Fv!kpoJwOWhX<ltuRfc6SCTMWq_8<8I5!TeCJ6
z4El}>jv8~_RSgCq)=8O2G;LEqTVO&M+7$M<d(z1_-EIFt1@4YKA<DeA%ewJ*FVfJP
zWT4Ns7K<GkvNHqWuZHCf{jR&8#zr#JXue19_P~wG4ln-n(i(q2GdpJM#;6g<f($Kg
zW|RLZ@nK-#9Ck?~EQxpo&2ti?5mW|>JAXT6sc>mtnFjQECV0OQR5B>~gUV)=^$z5R
zG+t7yM!8lGZx1QF|L2P&qX_X*8Dw1XfmQj#sx)jSY_@E#kq7ACAp!)Gyj%iixw0F_
z?Q}O@cs|1=@kE2Jbgc2CGU|Q%H<cCTn}qU$Aw=}Rh1`Lo;!7~71bV{Ep$lhqPLlX?
z<0VQDl(Ie>f899wU%~tT<nI5C97fyMR#{l8AXz}m=#2?4USBSlIP`jF^gRao>=VqQ
zr?-=g+(TVt`=z#vw@gTzYcmo@CulM1uNFJgpk*&#BD~7U|NeS?)6Ec|$h)}gcIiO#
z7(sf;GH0EoHQ5uo+$Arlu7JIm32>kQsbjW;Q`ZhTe_WQTmP!u`6gJa!<sLQnuRW`-
zpVK33WN!+`$H_Ts`jMtL!-bp1q-BZ@jS8M(qje%pWN^ct*SUkkyu7%{=A-j)ornv(
zS*Z=ME^m18y5&`^)dfs^C+!Elp2A<Z?jnZoSmR^$QS<@VAZ#yeJ(%#t;%$o@*$h9Y
zaVJt32cjQK`Cf1+w`dXefBTj>fMWJSW3G-*_4^JMGj;Z%OP4j-xghLA;zp6r3I2c^
zFM=*F@D%vwXZ?p(%s^kO$;gw-2?h-N7DCe1;o-v)G{2NDS^<Vghlml9LfJqA+X;0-
zE{rk`<yz2*f=8eU>n5aCi^wO#%x8U;Oj=g|U~4XR*zDgxg5M4D(M0|UH7>KWIm?)D
zx4Hdjv71B=U-<S1@u6|%6?|aUM#@DCDLkBwJDpr)0>HK^0DX7&n3YHoZ0yyCqUk00
z8B~t+Fg0AXsQO1{H8|YZj~$209j)xGl!SPVPeML{A=m{^!n>v-yC)$6P$FEA0SJWU
z2M1$7P;70T)u-*>+^7De8CLeNte5QL@h-FbMJGe!x%>-@^v%nlJ=FRTKR{s&G}_k?
zQLVtY^z8wkip(CSMnXNfR|$Rw+BZVY^DJ6Z<}|ra8$Z3ZXef|Hqy`f_hXBld6UKY^
zA<zMc`T<eLjKd)p9PUOoi|dCqISR^T>fQ%*Zy#e~a0l9wvQjwTz5GwQA^_U)sGG05
z2!)#YZ%a_<eFd^F`2i5GC+x*1B7e+bv(~C_i9N(QHl)1J8A%O^y<ApS1H<r$XNXg1
zwKn67z~J7M0s^iPheHn-6|w8~sKACIUJZ;o7F%IIqz#M*%p(aWRDf^1gKZhGI;H&a
zZf{UjvA=CDRFK;!G7KER>axTtI8#ulLAuPQMW#ki{QUw6SUkmpt&SaS!Z(J^gd(IJ
zH@C2|FD2SPv08#@=_1p#R{<1c9m2N^gIa{Fu4st<81X>Qd^ITEBtNr4R6J`P^xk6i
zyOV&P?K9$jKkngSCZ^av30Ss@+t`jD(65zk3&v=#!rAK1)ND+MXy_s<uUN}SZqeU;
zLOz3~v@NxK|NeTrvJQW(m4V(xUj71~%aFIeX#x|<M2C#CW`he~4{O%WAJ6uuca7yk
zra{kdY`Sk)<(W{db^ROC8fZ;M5iA4K6(Tp`Vz^&v(Fd9a3Z^IGL4#jQt#-@74>|IG
z1IPxsq*~1<g4qP6%=goC`jWp?X&?V!>8dzX0MROL<p;^km1_?hFkW`uzvJU3X_oj2
zRadvc&y*h~T23-ap8$@VZc+p@w&mg?P)R-lVTc4!?7g62aaz2oz$O}1J5wqX=EQo1
zcfe<>S+0d<l(mD}pMZd0NxFLCh8ig$Mlu`##HceNno*LP{U5aR!UMu%f=fK0c<8vr
zlSWEg;Z^bX78>|9F8>*CrC#k<bvRRS4yDB#SK1cAOxBOI=cG9#=+y*p1tQTmN!GIb
z%SBpPs4(~9fxb?oj-HT5f(nX@CHutcvmH6L>`is3Mq)yO8EUnC@v`~y<2vV!)i==V
z0%21XrTdO#^2ETXZ#%HpmjkHJZ7mmX^2YTn3o%VNUN?n?hOT#(t5D#LzFm6}bwD0J
zC)z9aRoBxPcgOAZ%DQ+tI4?FvsC{86OuVcw%**@?|52;K?M?^BJtbPI{hu|0${rtH
zu3hIOO6cA(|LQ11bz6+)zn=`>Z#SjRsG_o&W;Ad}PANbNNPNdZ^iJFSjVF|&Fdi@4
z^~Oe{TphBi+RjP~+MIPAoX!|k{6l-F6*sQ^*+)r9N!GXES~jDkVBZ0W=56pJAL!5R
zbL(OaZs=C%Yr$mYLVC53hS}+e#v@R$Aru+%r(m?y+QjCF8cN)XV7*W9<)Lbv<5P&r
ztrBbk{@CA4NVFX0f17qwd##!h=;*(nPLugcrUQk>hVAG#|MH$7E4Jc(9bgkcz#ZxC
zf4*y{!g;xQ<mjEJn*=Q#?+7{U3FdNAyir(6DQy?(JH&B~*!h_aFYp`*1KUq_7U6ez
z=x+hpt5Oavdt7n1?h7PcbD*j<&&JjsjBiFu%zYf<M$M(ForZ=BN3G5|B~SLE$KM~H
zsM@JVd;^k2yVJq-i<Zis88@Q$`ihQL6Y{o3E-gG{KsXSsJOCl6Y2LC{@okJbT^f0k
zI)W8u3D_)cQz4rQH+%;ePR-+=j9V~ozKnzGBa7qH*B&zaq}im~W-XsLBWLnm5<<);
z=S-N)FjLnTY&MiT!2X|S>Ox*JB}f5l?3B@abqPp)Da|SgRHVHVhv~nz3yN3i>_SDi
z?bWtC2#)rLYjp)i1yDd=$xM)k*aLWjyFL5p-*{LP>027-2YZLVwtPeaEl8Hj#8;5~
zx^I3Tn3W0;{;vn;WhJffU<XG|U*@;Rvn#uOqvpprZY4)Nz66jf$6|8Du3MMtUVEvH
zP7zn19dU1Yv)e#FvgwQdl`S(4^{-N`k;y&L)*Y1uj<St_oYn78`d=CVN%NmNcjvt9
zFafsg<R9Lk&Rh(Kk+)Q0Y~rSIra@J%-mW}*P=3Y#>XiGhJ<atjOAA8`XxUfm>6yRt
z$Wu(SP9{wH$EFmhE6VF@OY3S!2<0ABg^Pl}QXT9)hZkzA^YJ5j73xEp2LF+1llM-P
z=>j!w*r$H&Rpq{O%viAGNwg+CA?W;^#YM^)#*`dkJzOx=l})N)fZTtD5j<`Y?Ayg_
zFRgb%U7A~@&TwvzL9T4;Jk7o3wfeyK<yU--<5<|!5f?&Zvsl&ZOG~B6`Fb#jeT|_k
zci&MlTD<U;Co<X`(deejDcHz2?)?o9k=aDJO)zWV%%mg!Ut5|Y_lBZS1KQtI@b94>
zQ}kc{VBLliw57+85_iIB3L2LEVR?Gb8V6OmvVaKxm-_;4U1jY1>{LLnShO(Pr+k+%
zS{5&02C#W3%34dwAYs9blfK`Yf5yLogR7}xS5~+YwCc$(a=UV4_v)$`t<ZzliuuQz
zSKecCNh*~|iv=hiw~hK!^VgtxRn~sa(@e?rFihg&@>{!xiRK#8UqgMfHSy0AMsty9
z$f+Rj5&zz=0+k#uzzZ`unV*?E>IDPXdR{6OfwaZ`PE>P;$_XDErEnIv&H{)sK?E2+
zb)RZ;gnXDfXF>1|sfcmp%CX>yWlMx!e!`o)5%&}iPqD67Yi@p1LGnbV+<-5Yl)`t6
ztqOKf?O&q=R<`_`>wB}&qdQ<KEzkzZEzf>8lx*K7`etCiDdg!$_4~lZM>SHU-1mW7
z7fAghDM*2@{tC4kL$ZaU_Ri)3o&NBbju(-bdkb|i%?#SdcdlUxr$p0@?XT`GMlOeL
z$O7s5wI_mbyknlFM*h(b+gFYP2Ok-YrfAp?4(j>x>{<JwaV*y4RKajHGnH*WSH{$h
z2qAmbbV;0$(9Gtm&II$d*r-OpKNzr`@K7i2FFZq#uW0CS>k^k_m44GJ3bnh=CH{~@
zlV$C}(@3P<yo*ksiX3vuk8HB6XL)-y^)ssHa~P4NGKE>o<ARzdWsNiD%id8Rp9ZH8
z^L;|G%ly>ee$u>B1Sv*6-#dhaQiL(c3z^3A^;zh6a_nz-8bl#)pnK)RcL#P1ikw0_
zV+8UOGHxnFg*i)?t-P2%Mhw3kB}e;1Mo5NZMMo&N$x;%(#9K-7=w1N>x}eGBqR+V7
zqh*G^U^F2R7i_1*x5#J6b4}poJYWr6(gYZkl~~wWyj5Y27>HqT;pdf~E=}rZ?O1&x
zWc{NUp`GvlZBRY6pqo)jQIqZ0{7Us0Le98EMa?o~w@()S)NT^m$wW-}*Zwlah883a
zJA@^<jU85Pvgc_6t}hwsEAJ0A)cAIy_ykShNmN6GXK;CidqEKMm%Q41liM)_+#>~z
z(mczdyu;bIK>(KRNYIt+2r`N!4)cAKYY!@%G?XaLVIpJ3+rOCOPf~T&-@&0Ne_Q?{
z3p<Yj`05UbaPSmJi223OfQd0B@hZ(31B#4AnT8YmfHX*|Ci1SB<}tt!3K_(qrJ$K6
zY$urZ@9mOnDnz0}QYMD|ZRk#*PfKREjZ|4`Y;-b>SY9@pj;kT&b9L|}9n8QgEJ&9s
zT@<C8v&LGem_Y)C4%!ql6Lj+vH?6YH?>yztBJ2?-eu#$sRt%gX=3`y1_E7e%6x=7p
zP#Ou!`1<uhwjlS#0R=Bw_Xv$(L?KFcB&ztQGLwOCCK#{?BqK4h2td=u5As~m5mw0Z
z0OW9#-y*-+tYITA=BRbFv~*q<q-fZMBwGIGrDtfwB;g-#Q>sE(w5?NOsp6INl9X|_
zo~GtQTm$=d{9CFgDG#EE=$nd<NM)v$T5vl^Q15LUX~{ebVy|Lf&uiCI_3b<fR5j=E
z9PLKwco|r)RO0z`JN%LSyt+=QX4J(6*j2p$d<5l=Nq@B^J9hv)g>WrPIe_`(gMPs}
z(SBQWr!o6L`ps=eeH+r+7BX4`{3vm<+#7V(C<U@Kss%=>KkD7TX!xcrjwr)cR;Pan
zmJYtFpw<lt-{1r4d>zF5#G+Cb-U)4gt;*5I!s5T|GfU}viUh`5B{<f%hpmi*hlQ1N
z_u4|L)15xS-vEQp`u2i?-;j!$0r4N-VL(`gAj|oduoX-7=jVpXkK#lT6Njm#-OI#{
z;}THASdU~FAnFT6FGTbby16B@sfv-DU0d!IH7m2zH1}OQyKWC&;JU<6X^M6K7!?)8
z4PN}&TUa0%(#M(~_xe3y0ER1d*r(-cP5{RC1w>?i$oO7AH%$Rtc$uQIR!8NQq<%i6
zh~1wu5x>vE|I{gQJr%%xhk?ld2M}rlvak0WPAz3pV>jTV%2MMPzZc~oMz_R`=py?U
z+SUTgFU%Io?z0QuR}p+Ex15n*Me@7qpRnSlpOAjtV`<Z;?ZNED00}J5Z+Nj=U`4rI
z*lA@{48k)=4@-9d+7h{2K&#BSXh5jrMYKXxTh-7zm2Z;16v2#6;pqIE-h{7v*24Zx
zj-lBR762*f0@~lH4or@EBEmKqLA{eYAgccc^xao1caC7}H7CJXb|w~b`xY_aV{mIs
zqCJB2(cxPVA#tnzv+%QWl_0DYJAh>S6aZ8lJImY@zGI63(rS~yH3GB#>-$1=-nx7}
zo+ojSXD+x7DI*huboYb6grKh-4}OxaUS}5s2j%u+2x!3MO`h18S=K4x%Ol*Y*w3_o
zn~7cLQ9gW5L^C23?mqtV2SyAiubklxJ;R~wPBuKhw#m%^0o7N|Pv7Xffeug357;S5
z`Na=ufFD)$JYy$phbU41*eS7NA6RU4h$~?;#ePRr8^loDOzBV@wHEZ)A+ZKI!-F#-
zdiT6>*SQ*VT53UtBxMPrA1xTdiLqW}B<NbD!gLGt!z8%rE3nGIgP1u9g!ZaJueH!b
zdwK`uzr#Izbr)#RAWzdOe-%MMU~yqrj?OxyUXuyF$j^$jD@nRN3WUzSph70`YcE=+
z?WRxum`3EI!YUN6+CYvba+S8nWl7|YAT=`^SHE%k8d>|SQ9s=YtZ5gU>}xOE3E1V@
z-P397r<MSSGEtzuJDPHSTvAp9dVTRT8p5_vcGH_npAYKlrKa|WK2RtA*0mEn&gzb;
z%IkZr&JQE5E!mnI0%g{G3P)4TrPop6Jlz>Y_YR(IB$uM-aohmx@QnQywQS@B5XzuH
zIUT<~pImtVj5L0=2!Pl_V0rL?P)Nq|QH2+m3)X#YzWPq8k%c(HMk4@JmujtW`!fT6
zvlPg45xUlKspe-lyTc1lUoH%>P)u@;&|`V6{cl~gb#E_beV&M;>UVJ}*w7FkxCib8
zsy%q!=63<Uj4rb=gYE_#x@F(*C}PwUZm<qeX5^r*zCTY?CN+nyPV9%Gz5p<z<rP6w
zcv~2P=H`uE=J~y(hnu$$+&cOg2inuFbS+gM$7<m9d$jr{4z<>9@xZGQV(oUqO9kHU
zDEr^&MC$6|3Ps7l)9Z;pV*lZK62%#{D)OY!4e|JhRt>&BV*;$O6JuQqa4}=SfBP-D
z1?i#A3QIW|6GFOahxsXH^q*vJI}693ziK_+{q5qR&dl1;TAvbn5Y7>^!RY{?$L33M
z`)5L<uCuRElGDh7ztCOYb?J5a`+Z6vyEwVD8xjR^pw=xZ$xi%CPe{$Yko1wiIIx=A
ztS@ipz=N6SBap*;K#|~W<T|=irdDAczh~4P8YC(6{OP4uLquHZ<kws|0&nO?Yijcw
zy!`9Q?O!4hw~H$^ij$jwwMKmYbnQi{u&yr#?k@Z&V{A*kaLV0ll64;a7p(xZpR&V$
zat_xd`iJvPVN{n18Y!XAexPM!=MJ$`;7P-RQL=9Oj;yTxc|}?0k>m9x?X(Keob>K;
zR`7j1HZ}FxW9&5uQ~=|G^(b32Wn5d3Rz|(9`p*kqPls4@gO}!X;obQcewZPMQzS!I
z<UxQ&l<_A0%hmXrRoC!{Yhe~@pC=f2_0&2yyRf5*Wt5`>A%X1#5TUWK+L_|EIIq9a
zu`rrry8JtLeYYodzBs?QrJE);9FKybzes14Mdlq|)p;F~eY*ZfutggVv(@URAieWt
za<9(;bmlhFJw%WW>do@(Hz_IcJ4|xpqfdFRtF6~o7B+xHCwwepB`mPo0#zQbN6cFL
z%O@e-bU(k9-_MqPx_{pUzkj|L3j7de{XieBw0Etic^eOTG>vl{0U0%xda2cS_JXPb
zdO;pGUZecV65WHu$Pnh6D}0MXSk+UReWyV@hGe1LX`BEoL4C0wE25bQtSMy63KVZJ
z)d`La=$%x;qBpnrEwF8@00N<d|AUFE0n}12S{z_xWL6uLbh8D?AWmzcO?iXfieIVn
zUJi-#xvl}~Z#Mmz&nP&oe5!W8XZz<^7N|uXx9LT8s>UFfSm8hC;J2oLAGQJXfak~5
zHun?lH&!81t2FH=zM|h$o+5so<xDS55ZGYPq&<j&AwFPXR{U=}#847pov%vgL|lpF
z!zeOQEQKB7gRbOQ#4m0K>rb$2_RV)h52}cUw$Om1Q$<y3@C84p{F2=f;?!hlkJH4%
zrcb2sxph{dLzYle0vLD}vmO#(u}3Qw?z46xW?qxK#FqET4JleiS96vp*sbfNF%vP~
zT70d@wx9EZV`<qF+sGm<kdFqQPG+V=Z_G2BH*MWMd_Q%k+htXL0QK<|0Yf0XQp6M4
z0plrvl#62>fd-t(4dP%U{1896oM-z>4qZ|WKliq2O8Z(O%vG|uV?ys}nCG=HR5!;{
z9nrA2dYtbVP!lRms;Va*SlM|I*wpyA?L*)z$MW;G{*Oma&8ygMMo7J0uU1AIFnf3a
zy6s79w5j%;y>gqj1)^Jfv7m7uM)WHE3e|q_7CYV}DAc~OFztQF{SN2bDmu++_aGJz
zPPH5Lp3Zo=ItV55SsF`L0}6yYV6|_L9|tBAe~)6d4GeuS%4Jy#rCpD(_RWq=N&}FX
zx$zFl&kJwJY)&jADTp5iiu?mt63uUv>XzbGju5<yeQdEd#8na`W~{FWbO3cU%6ARI
zSU%u@VI8oWsw{Hlx$t@yj>Zguy4;o3B;CGD=#tNLvjchiUilTa5fgky;Cl-g%5Il(
zp!?@G_f%SU@em{peuo)8N+%9xYG={n;LBJy^!{ARExPKwCUrS4Z>_d*?PlK<;Cd7I
zkekS0k#nyNjpXAYHtA5d<cAZ%efI#4k^D&7>wm(|0C-YQZ=SL70c0b@2nkb<)Gl+_
zwZ&0=Bhu*$Yqw0PCwUO{0B0%htlr*fb<y|#ZZ<%iatajR>0}TW<4*C1Pt)6ZZ|K1<
zLfZw$ruUG2+N8s;8|*56>pW-XuBBtk^;0}t=980!E^X@>&6_;|ghABBci*SXyP9j9
zf3#X<226B(7_{?D>i^xjvgpcw;G5laU5yJl90DD+(1u6N+8iD$vjrOo=1?KrIpu#^
zY!|C~Z#j_5vU^2-&TpWK-^5a_0P8%PozJi}N4dRC6bw{*7KXrlEM~kyw=xEiCp7&M
z9^FB7HKFm4yVwM{UZ;1%P+QA2EhMEgH*3Jwquv6VEV-j%_?fZT*&T4H0k3K2zZ{bw
z8FjbUkTj1cYaLT&dsZ}EQcI<qv^H^W%r%g6VnuBXVo`(a!;pdV4os`FVkh@cKo&Qi
z!1|C&m)p+r@Ik5wkce4InP%x<*>;8>&3w!8baN#XhHtWv5=l&IhB^&;z&3V_mEJ3$
zJ;_InaD@MkJOgEbG7jcx3WX9g;msTtmY<D^L6V>49dSQZ1IiRYBJ4;sTb4(h015c^
z;kRRF;lb{%n68~gAy%y|#RY%C-k@FRjFU<t9$aIfMA(f5{*g&evV;IoIH<Qs?abzR
z3xilyZHri(K!_+Hc-osu@WQ4H)V!~N@-pU7COm6aZSR%}&7_aKO_K7%4Ib`F%p{5q
z;A0Jz_<~E|<OXtTS<bNvq9M>q{|vGt`<OVDJ;Uk0uBUlnfSL9FZ|8ggm=rslzTV7~
zj%Oc<92iyK_wREBSR6E11iHh78l^Byh6ilos3(E)!N;`^8YRwvp?A8)FwCr-w50L3
zqmGHm%36kbTW(hfZ{6{eFu=2L<NYL@2%?sXem*=M-3q<lymsd3(u2B4TGN;IGKb4|
zum)S@GI^+6o;SK?HtnWcRP7mYnL~1v=*KEi{X{mxaxA6h(Pj4?JBYME!>`1uEbn*k
z%<G`AO(fbY-g)Td6zXZCV$LT-yHz6KLmgFBcWzgq+?4Sv-q`PmoamGHlNW=6jfAY~
z=CxI8++2g-Ac*5HrTO%J>Br$6P%mYEa?56Fa+q?fFsnu>VBTr+I9x!ahV|H5Sy|a%
zdt*yZ%!kP~l6Pq$2m_-#^i~alcaXAb0JH0SEQqi~b3Xpdaj5}*nHhPNtjRO|*kP)^
z)S?m4;kA|u;!l!;*_7gV0$3uyrFUf2h&hg{s-VY(H^L6esxs;F<}ckIG2}^dZk8W)
z76I*r@8d|KZE8n^Env}|->c|Y;IyMh^hSXXT2ZUQLm(ZT({sGFExEf>u&(5Jz@EkH
zV|#LDzC0@B<Ub)_O^W+~I{d1$N1NzHB=5Vc0>J>NG3%u|jS+t1lQMBBKwFn}ERoq`
z3oJCbdIq+Y{A4c5nOV~alcD6eU-<8ZI`;p%e!t?w-0#Hhu79Arc`Dq3F<q~oAv0>&
z(0a})yGjG6cO_*vF0FyCad}H$`OS9Q#y7ik`)=RYYt|Y$y#J^V4W(W)oLFnr@r7rv
zvO+C@*D3WB@PF3_Or>iXx_Z}*DgVeNvaL(y2?I1G9RPb>d4b|_Oi<<WR(1>iuX(b-
zTUwgcA9{>0Bom60*MI;0$kuH;mMIM+P;-WIPIn`dABJz46NxNuU-?3y>2YR&f#2>x
zJNftKN?zoDCTT@bCxEVQ_7bFdo7R1Ccwp;I=CanDVK1?ANQi1L%|)OTe!0fDj_87k
zwKe*UhD1C%LNzeI=}Rc&&!_W1c*LoF&@z6mPFY0wcxMZ((E=CH<Q7Lh;X#?WM=J0K
zRvV#9kG6AcdX9ovSXq0(_>xZUqd=5<3Yw~aJ<1)Ezb<#teu*8GIFVfn`G<}`!q?_}
zJY*w_El^ZWRT1}}KSdIfcAQSQSndSn)lL<X-KR|IgY4M5QG?(6%uNVxd=5;7JZd2f
zlGo_`$Oz|4Jas!I#T8V{K{35}Tqo+L!1EK2PGIcIK#IXvcyWB%()d|IZ-Vz>fnQcv
zf{ePbrNR2&>FSVD&=8)lDlW&QqvuP)7aO(!pFZa`r%%vzML|)NGpBwO0w@VAgkj?g
z8DQ+pTBJ{Gh(!)sq3sLptERSx-b`g_-}v)>P<({k#zH16$@C{)wwu)XKY)(GAyU8_
z$o;oIX0Hh8XavYM&VTA$VqB`<J+|G1>RRKueiFVyD<;qD>rePd#0(nKgTDLQB|b<A
zS_1Q6RHH}gKyAS#*UQ`YfRj4P^MXN+O_Uq5C=GEcqZtSF^TaR|jz9IiP>3m#x?f&~
z6<n=K#~Vw=zp8f35RvG=>_aw`BWZt8Uo@WGHv<KU1`C~e-n(H!Gw25H?Eg)#b&JE%
zbaN){>uF0v*e`FNo3Y>oXd8L>W)8C_w-VO%-82mVH9)dfUTWy!>G*pXc%02uo!%cm
z>x10gI-HJ9JHx=hpjzS@QIe8al4_M)lnSI6j0}v7bqx)54U9sJ46KZdtqe@G4GgRd
k43_QQz8OVBZhlH;S|x4`*Prq812r&sy85}Sb4q9e01S-1RR910

literal 0
HcmV?d00001

diff --git a/docs_new/logos/tinder.png b/docs_new/logos/tinder.png
new file mode 100644
index 0000000000000000000000000000000000000000..28d0d0ddb561bca80a58c74ab15b79e93a6941db
GIT binary patch
literal 28182
zcmYhicOX^&A3uJsSyWtEA#^Lr9+9%SsO;R!wa2x}PRPC}T`OcnvJ<Yovu|W%R<dUy
zD<gaRz4ZQkzrXM84>|5R=Q$tG`3lj}P&rS<LIpw4`TO^f4<U%G7=mDIXDGmx!8Y|H
z@Po`oUR@r7N+PKb&B?*<oY;F0)gj1>3xaUZA!r|5!p%dFn-ByonnBRr1PEesepjO{
z1^z%`p{jy}PEP-&)n~_mE0j;~t1D6dq9QvZeEl2o@@oikV7ZT!M?D)^9`}4^gx;fE
zTlKaseVT6A!FAAZRg3-sll98`Pmq;ZX}NtUKR@a-x%Q9Av@<`>!!0ikD+&E!Q0(hU
zPNyc%id7<$r$y4H-z@)n{Y#gpzyCD{owrX#zVPi#?D?f+(^x(ctzPwBk#_A(^6p{K
z8tw}Yrd7VMFOAZK&4``(wcAy=^FYcm@iOAS506+gJ+1ONnM<KM5G>7GZfob+T*Tbd
z{lqrAxiIAaKG!c=Q=K8yI&tfqg~-z=<^BqNNd8++@G<rCwxPbLg!I$a)2mXvn>;~D
zY=GR!gy@O*LMML`vp7Nn^%MtmEbO&8ugAj5D%!;c2cPKMe^*!n{bs{O&1QNWi|oCg
z=LnF@JltD0))c}bXT-L)w$Kg^4l1NMdi@6Q;p=Tx7?@cD{u*k_Z_sVhTsH&{K6OFl
ze1RI0u2OuA_CB|4omB2q=T{>gtFdQARghVX5z0FvtzD@8o6zI+&B{wqAz?xUP(k_3
zhVhgH?S&GY(pN$MJyD?r4{>RP7r({Lw8oeHs~!A}Cp4n>kS&a1E!*h+RZ<D^7skDR
z#nAPSH|M{36h|qY$^><n&kRma`_B9KW2irgouQ^W(@G3gsKj-6v3hGDGVwlwr~FII
z4H}b*$Neo(17%tUzH#3Z5;Y$5eO(~;?+HHO36NZ`^u{W*y|CB*)n27*6lWBew0an1
zUoH&SK1ODKyQ(y!@l>G0<aOagWMM&me%;61+}sF$g?|r_3ci`vV}5Dk@W<fzJ;9=)
zD$=yZwoTFBOXW8yg%XjmHgplnjl2kcZsWgH5E9)uFfdS=Q(5V*Cr<rO0)LBfO2*A1
zp4gZ{FJ6l+N2$+Jd2`7w5j}53!tY<t2zYF3n-i{_=}#NY3q6-EHN@suXtr3C^p?Ch
z6+b>SNaxI1C~Yp~uezLH!`@Y78OI$~6MxF$$=D=6e~vO$<Porn+t*^gNTm@}JfAYj
z!lPs8LOx4XG6Dkt%=N*}Z7m(^p-@s|V`GOhhOhCqI_4j65HwIn+1Gq1r&N^@Yvv2L
z%N+C`^OGcw<@IH!L=;PF`nmQRs)@k2`V)k)JW?igB)D|K{yktX#;U5F3Y9ECw{GJl
zwK1kUjhUI#o)C~Td<&;@=20?DdDZNot0B!&->5(A(S}E|%$DTHS6Yj=A>2vNq#T23
zIQ(9=h3PGwD;<yl!rQ@io(c-iaDf{0<GyGolqRQ?{|PE=%IP5A?=O>RznanQ^(*9t
z0J-zqNO#Xzyp)jR*hl^$uT%9PUwk5Abj#f}8$V&l!IR6iHd&_hFWC6J@xa2hHQCw{
zxe+IG&Bl{?1i~U?CFehlm;;R@qh#kzVml1sb~<r5lYqC2H{3=Zd5iu<mTUO>9tlB>
zlv9HwJ8#(99+|1qo!&LeL5(4UqDGxesi$Jb!Z7-q25aAYo1e!ohUZRxd-ksF3Ov1_
zU|9qjJ*W|6sQv&fK*BKQRD;}&7>Tr%c(?PPo?I$)*qt0(9U%n&;yK*eVwCpi*jAI(
z-V1H9pd|T7Ty@D=tQ;6o9m@E}+FlD_jn(sKkUAx{-f^@W7mC5z(wpxkrFgXF+LL97
zEc>CRp(fZ_8{KPkm%4SYpF&Ls2V-&0=ovZ=husWl$=t#3;3w`_zBkD$QzJ<{AwnJ{
zlksid4|+wTbx#p#$!Hk=k3@eAG2=X$Lp?)cDC#o_Zb2yTgwU7m7k^#>B`!cUG8c_k
z3qvBz#rgjU&<_ZZHK)?!*dx;%Q{}X??JDh*>I%0D^1brRAX~FODnfzYKIVodJ0HT7
z%FK3@8Ov4j;vbr<12i$vR2mWY`F12p&TV1kOM{YM;{_HXAUz#^#r|@JnL)kR)_th*
zjAbg}?hAgkQ%S^f@R?~>eTy%5I&|9++}HZtvEZ?FIFUqON2Vi8);qsvv)o33pfan<
z^fJ=IUW~rqzFboHPlP|HPo-h2Ix2qI__qk%_<VP1qe^PkV14u#zJW_Nw~C|jf&EJm
zo=l>l;IW<Ej*aNWQx4CwQDe#jGNcbMWp;hP_!k>`7`5s(0cQ_bSrqK~5GP?kVbG9B
zyRd;>TUpjW)+k<)^DPz@wCSmyST>!Xr1zoB%sYN>D(g*_6xP0T;hYx^QF=Qm@l@Ql
zQ$XA+Bd+`+acOvR?gm`NLiW><M{LY=dkKUUG=*?!>c6PbK?<m^?dXB)IbAJTF{LlT
z_m|bvQb}ite)v%~jQ6qI5y89M5Enfq`5O@+Ei3+<!O2M$50GnQgbBxmcD7_;D@NAH
z_DJJb;e0_tRc5OH@n@89Gi{5$KUs;(c*m{xwt~ave(*9H77bffj9jfx^gSy-GkDjv
zV`Vq(i1qtWg1O@N-vCOEkTd5dQ$7$*G`M00DKk2?FB?$29CJ=_`zLGCHejwC8W$Gx
z>1GW+cWhU-S8&H2Y)2KDKkZN1pB41K?^wTut-QX~%3Su6;Mdl6a6*q;BK^)5)G-w`
zt)ugwfe15NP}yD%$Xz)`3&>V$+-G!~T-#^T>W}{6syX_$;QkDCYOH)XU(P>1W?`p#
zdfe-wxcl?V^Ci<k;hZ!40}I)!pD3=c*D+YpUZuwTqs@~DXuHjAs`TMCL&@Fcz$pE+
zxnwV)q}cXa<V=vrnGNa6TmM&z<8BiHvkM4a#+u_sL&c^IxBVaY?R62o`;^#gr+Ud~
zFoT8FuvQ+QJV1Z$;K}yAL=|-B?#5=!P&slUlOXs;KD?;p|1k>3+6Wbs@kVYrIPa48
zjFl-N?RVr|Y=^$R38L3id>$6s-e+(9@4*3BSQrG22)T0O_{EP}l@&`&OjcI~yZ_c7
zH2R*28h!dZHqrhOV%6Dr=Q318Ch@sa%x-|s`hQ9VM*HJYSjFh9+j$;n!c)Y2u$5<U
zp(~fEv?N3->pHQ7-=OZ<UlwuU|I2Ce_l4d~2SpB<MCVKdqhFS2wowPgjdcpRrx3B^
zDEj~LzQJg-N^&iH23$(}^T@7^;NT^l#%|>sJoQLKG|8GWq-aO!ENhtX4@_i%b-TT6
z3ML;rdic4cg=@7y)(Rfhei4OSIb&u-!$lXOD=v6SgZ3$qKu}(dLqPWGl^uj`rTMDd
zj*^Y5xktu#>X*%fuw4l=^P_Nn)l((L%Ki)GoTEhtjayXG<3%*y=qX<~cVYh}o!*~t
zIn-U(xgO&SQ~S`1%^QPmsaBo;0XNNQSSc9{7Nn}9=&w;%ZOoB`gmK5ZHfnd0DPh_R
zC=}|Do=6Ehlhw@ZDr^}0kB3GK5n(Ik2#sUSUZmMKzMtFKz6a+en{rR~+y5rTf`5>?
zB)~Vx<cHc?jE|3hGyMM}Htghhrw5{Nd(}|mpDEFIRHyw}fjoSyNK})xmozZvmeF86
zDN%~!<yvHDXedX?_5aL1lAMo3q*Qaj;c(ruDS%;$LcaT|*YTr+ziFI2XV1TE)`8V6
zHT#;{h2JTtuJ&X|ivCCC37`_%o{D%GHQu;35_9&TYC`wIfe6*WttHRwm9FW)oZYuc
zNt1g0eXysyj~dZihk)2Dd-;FH)&#6LO_ye5vt`BfSphzdKMl6pX=A3Kes_8^TIw!7
zR4`FZCY_K?(!Oi%uQYS=`pZm#{hj}AH3xRup1^J@pbgKXIWXMy)&6rn?xmZqFGD0v
z#xGT_7#fm<7h@fRS<ViK=&2gLe@;l=N}`+CSL9<wk4x4fy!hemMeQV&NK31RY|CNw
zjHQhWUaGl`Z%>3qq-(;;j0cZc3`0-Z%vr+_c|0y}pUKZ#p?`9oXJ&hXoyQ`%K=gxO
zBqf#hqib{_wMsJ>sDOQ>pn|Z^Fz6;|{xA1}Gf+}vta=EEQQqb^s2imci+f8JXMyW2
zNEzTCT#dfhRW`s*W)?u{J|||kj4AZ|FX=SmGmE<u@XB<@zxwcpm!GG|PvXER9aJ^H
zDSBZRu83mEMJ>t>7*G<cU_M5H#`DmctKISyaPdy%nSW;OQi4ObJ{>9l1c@6H$9em9
z2I`I4b0d8HSLKDa<)v4{3!cr;#wqi}J*X`_*wErGzm2Pf-iLidhqhw}Mih+(kKB3U
z_Mh|1pNb6yGT68>v1K@SjkHw}UX!Iu56+uyK9Ae6qQv`Zv9j=ZuF=Nz)Ol`9xEb%L
zVY^HqN3!A?sPTg$k{xokMMO)8$2~ZTNW5{XI-X=99j%6ZF_n<CGOe<Pl(Ll<nvE0c
z+Eu+7y9Y8C#294B*WB3Uky-uUzpqo^E={*a)P5{2T}MLMA??u+#nuMDBx>ss6hZLP
z<cTwA<@<-jo}@s|YGkIE6Q$fx%<#^Wqe=oUwl0keH>Du$)It4ZdbJua8coS(P(F`^
zJV-v4H8eC3K)NJ^zHAdd%t}ll)QI9fLBVcE?K}`KbVx&#Mb22+EsGNFrs|OWbCJ3b
z@_dmjo8*cl>*Bdq!HnV;oR;2^e^!^15%DJ3ec1xhNh^LSDi@<YZi$Iiz!E06B2SG6
zPZDTn7MJ&Gjmg_bnu9uXr&Ums>yKf6klChRvX~V!4cDny0H`-v!1B~?Zid}qa(^;B
zaQda4i`3+hy9&$g4cq^|gW%y~EbfR(@!Ia84PQ7!yy5uMxx_ADVbx#cdq>GeP5Pw@
z!>IuPWcyi+L%&i;7rbI<m5o%&tz#%DZBUWfQ7Y<N(Al01_-AN9x6DAdSIB#|pJ;ce
zE;($u9DEY`sM#x1`&NbNe-Anawq(6oE*ME(tU0F#UN&1?;Dw#sQE}U+VmJTiS4iYu
z8}O`>C|4gBn>f=HwH#Ilzw{K3;O5Nme<C<dL5Vz9?^<;T!QXw~Q`%W2!`7t6H$XO=
z+q%gELhn>q|Igc=2eGH62yJj;YHvqrbYxZ?7S$!yE_On}4)-q9^xKdv@!n#spO8+R
zKTtfP*qk_;G+~JHsq@#M99CYfzo-y52uUQ3|L1%F-pRHv@JU;@ws*~3YDDMt^32fi
zFsN?rCQ8QIDcUJ2-a|K3F5Is1k5;|j_EM9ptJ%xYckE5sX5GyjrQTQmlX*v^9Aok3
zZeq6W8Goiv)fhcrEqm&IY)54lZ!iZ<i-51Dg{jGoIXn+j!cr~Y_+2Q}_}qWYd?S)S
zrb-{^9WiN6O!~3-O?uo6OFrvO$@u4b{e&ATW5LgJRHh0^Z!^hYm`T{Tff|||ga4D4
ztr>|cWjy=sO!?--IY_!@C!*N+j(BOxejGM!y}FQXGd{_&&fNxG33>aizLrhY+G3OT
ze-J+Yf8V^@fXk}r3%21(^^YMO!sWVyo$PB0($uIi2-(f`r;UGnb}bc_N4|EP_tI|W
zbQk-d9*^dhNaEQ=v4OfTP3ij|v6_slxul<4C4Sdw@Z!Vu+H|W&7BQI}<gZ4b_<sA|
zT0=RY>zitP{(s6l*(=3m;pT-!8!{0BUn>gHrlHh<%bmF{VstclpZ>h2&x*?4**1XZ
z`Iq<^_DW*?-EaS&_fFR;ja9ghI2T+vx)2p}`-rQQ|6~1&fnF9a^}#~Euz~UY<&Csb
zCl1%LL?dfE`dth7Gql`;i+1G5A{cS27<ifG`plOg*~7JFLPtl(r`80*)4LAg!WU6)
z?}WKX&0Y`;AdXPHft{>KYrby3Kk2h<P>{*0tE)S7U7J<BMweTdlf#@v5Q83zB&tQ=
znc>0%BR|K*)x+A`k_VM*#+Ww(gsq41@T>7+;^LK}F-U1vQERmpa6uWC1CQ2DRI`r7
zquhMGpA!l>&NE|02sLyN@_n<xfjKw5fdHc5r}M`+U)gUIpy|;(GCEpSGT`_qeE1=n
z7v)BaThw&ePbs3pcx`qCq1J@xB@2o;W(>@7dh97V*U0&#5B|VwbdQTPtG{#rJKZ$N
zf7K(ADew3(FWQ(M9xWIOyNa~D4K1Bq!d7bc^TX`Vn{}e|bk7&&=aWGgJ9D+Uh-C!H
z(T{>gLd};YP(nMi&G|zUur&_o<Y9kfSo>tSuwX@bdA^yd)=n6Gn~OWTzt{b8q-5S=
z59Cf!d=YmA@YmdG>|bH%@n>r{*tpG-D@NYVj+d>?(-b+O)qfkRyMidP6iJ+e)xCa@
z_1&I?suYykzk5>~!|-g7nI0tRmW-8jJ7O&<q*-04TRe8#Y497b958_7xMBjPCGw!U
zJc1MKnJRLEY%QM9Kp+TaJ<Qb{LeO73fBt-2XFR{89P$FJ_y%qq_6KHGRasfNsA(#@
zwO(D(ej7;0`uBaR>WjoE7z1;@Bjwa=8V9GldP}6z)2EZKsr{L<o_q`<IC+>eJf7UL
zx$7{3eX`M=H~az)_f&CkSbjzD!dGlh6;kH~eQORfMODx?`YjduzkJ$gRyIJGmN?Ps
z&sLgwgicOX<1)3di;KJ;!?hg!#j+J9agJyM1N_<UG@)SjSp-9vgwOcuJWp4l8{1q~
zA=V|m!d)K3QX0dMLms-2=gmLYU<5Cu-lYH3;oim$xzKyLxq<BSu;Q2aKi`g&-~sp?
z@H&z;5?o{Mq9_9cLs$~XpKLm#O@H+FJH%;ZN76^*3G4$SBVIT>>cXb4v?=kSob%@1
z8^SM@D7UNtm|B3&@b!7hu#~$XxJi~(J*Oz(uBZ;mKG8W#O}C_Rrl^5g?wv8{p4_Jt
zsC}uBeiglS4^lC(^Dw&YyhaNvuEPI0Kg9=^=VLq&#Cbw*d3bqG#(w-Le_e1XC+x`h
z`@*KMG1I}cNMe^b`$+v`2#az%hf8c_mUy-C0e{QTjw3H8=iaT`w~H)zU+eV`4(ho*
zeX3SGpbQ)l@6%%-`8mk<q~_Rvu+aYo{>=A<Tfl<U(W&&&O05Y<(AZ(M*Iy3}4i46`
zvs;otOPe5HJ;?4mAc^u}dfw7`|7-BGEaR<D4kYg;HrSVbXU>}&zBokOn%j3G-y(i{
z)W`Gmg>hg{<M-KERwsMYI^XE;pKB=2tn%KTIOeLY{*G_4xRJA{=r`^U_vDGb7jC6W
zH~Ul@z4!AahX$(u%fQGNM8Gaqmla2bQ?^?^dgu5{*0?XOHLSga;B^k?>&w<-fL0G_
z=ZFxdiGQ-6H!a~rV%!<A``5r^Yt2$*?^bVAz*0@vUuviU;v_M1CMx(a$i)!Ea~Ch>
zq8{@bCWrd?GCX#0hzM!FNw|B77*Bp27bay${JyW>{9}kRJ=I{>YpUf%Th}|*SLYYy
z-)F1UqSD`Z=W+Ktct<k)cz>;=!}m$Eady0`4O3W4p;rHD=_pO?Y~j+txlMJ@RdGW+
z`4~ybE~~~B<KNSKM}(0$EF;5yoO174;myBFXe%I6?fJv2L*Jla^4&wj>j)ZryOW`T
z0b{A;dgZ-qPojjasc<?0axz!E9R}0(SK@f}dh|;z@2d#=-=W6L7J!x#2iifap8}fQ
zq+dfrzjGGr`}_MNU88qpwitHOT_aQ*YreRRPsZEwRTWii^T87X+2f74Ih_gvvQ&D<
zl>22UmW^mgQVY+Yvwu5iA}@Wm;^&w!)zMGrNzCMO^s;p5N5K~<8VNf2``4vU!t-Wt
zd;I<Nd|sz!Adq=_`l#qjpW$*Yq0i^t%WCMsLn27vE;|(@e;*kcfzm#xX(V4EczqzG
zEJ*Z}5xlzRzh$n@UYwN@%V94h*mxul9#BY)J%(<Cwudyor19@{c{d5$Em*2?`g<=;
z0gptCvSNEEoI@+zPejz4X`$@U_FRCJyg;ta(8>d@7=iucTX|2)1P_g84%LMH^%WHr
zZHG&(+DH?!2+<GK>$i+xs*!K2YZczY{ya?k)goXZoA=;t;dvC=fYtf}y6-O~H&2+}
znT|@GwH7bKdd89`k8$s|m7+@CkS2WYJ&(u;Sn@htzeKp(Y*1^Om_1Dkug<Kf*iv~l
zZG_g*apk4f&(bk8^agNwXVO5j!VahJSP65!iBkn0ghJbFtROmG-n&M3;e|pQiG&`y
z5#s!upMTp3*+b&2J6NONi0&VKnEONIL-4*!;d!yvcY^>!0`ReiE4tHJGxJOWW2zDN
z9m>DfIDRn0t&j)FIB_Ustz0T6?R>C5sP#vfIL29J^^Ia}B4tQs6sF(q8*Py6V$ENR
zL}VO;Zj|@kSFVbNhNF@Mugkcjgt#qt5@$tVaCYBpH=jQ98ShIAQO=xZYuhWU*d~W3
zzFOxqabr@LGuGF<^D5>ETXu@d*)0`g<8fW^eGzQeC}Ce~aNmW`WXq8TzLzWfTlUrT
zVMAr*w%Ld0`v|&_aJ+J+tWay+*(sZ~_8<1WV*y#FADso2T3hxFe^Nc8d{u#k>X>7L
zR5kDHsG+58<f`Q>4}V`-H5q14ax{E><08Q%i#YeZYKsm&dyO&7_X5!awS|24C(?zz
zJQII7;qs0;zrs{C>tV0Z^%p?whIR!<GPtXW5hj<<PTgyBNuiHc1x;>}56?>hUk4%W
zgMAB|D0r^5@)KS1si$_!3$LbcktQtH6n2(>%?3_#{)?r^B&y#DVqU2@+Caiomjs4t
z^PW{W2!}H13R4i5<t2!lv5#Kn^s{P@-E?Lbc(5fq@QbPB>A9Yy5ghC4@zvS~h>SlL
zB|{kWQ3;_nA>v(xSgX{mL{GonK>ps(UoBySLf1Vn8~HO5dkZS+YHL<sY4z^`sSPu`
z-Uq-Mrgql%?X*NH<(faV_(}5qhHfUL1z;qlWZ>9ThFwPScyJ5(Nzg_lu@mKHShQmE
z<>Ya$N@`<va1=@LvNAdJR1n(Syj$8h(OHR*xOIjyQM_wRxj=@>Ol!N&xyIDAaxkNf
z=glLjo?1#gUi>O!*yW)Lzn&NfQ@7Yii*q{9a%+CCuy)9HDrjk$t*si3OksdC7VXd3
z+AWs>G6|;*>ppq8xxTxX+dS!see*uA`NH(}*i{MaJc!6UuQrf_eUZ6p{R0Ea28M>I
zl+I;u!t~@qs4)iKKVQ9#_SCYMo(hzAY<bc9_!|56hWv+^?DM*o>`gUoy8ETvv)x;j
z<vAsb6VFytO)abLBg=C^5ch3$c_)c>yNUbKb#LNNI)VCf+a4q3i=)1;UumDP%`HJD
zFy{*rac*f%kU;rB!!L=ROKK{o-E!|%i;{x~;oj0o!M7oVeM>$@ZwZ!&m6oTm{d<)a
z(-I!~cE-Dwm%qLC9eR9}&o371$wqwTsSo3>`qjjF&S7g=p$JSQzCW&J{o3s7X@efv
z;XZ1xbHN^^Jb~nq<tDzqzUW&lM~+-Edz3g)3>!E80VM0nHg{Z8Q*%X4W%AA|*L%Rb
z`k+Dzq1a1b@*JQs0DGxdquaLuf%67J`%HxW3_{8|6ILew@LaDtd>t(a8B=o#Qz6q5
zrs#e>NNuZcPxm#oLP4sST7fv}Q1&Y6kq?ugNa~JN8wtMRs-wbOQw4d)`E{Kf`H=0b
zuS)rcZTtFN2_20>aJU;I<syJx6S`<#KwM}HNC%q0@0SxpyHE4t7t<1Y8Lkidn62NZ
zEk+0n0&uvX1Gfj6TsnqNe4KfMfcGIyZgJyDet<r5yqkRNeTzjpWZ_IoVeRM%*_1VO
z^|10pNzm-kHFn)FS&}DOH^Z117kB^S%1jRz;}Gj{3ctCQmXz$XzuTJzV$cFs9uJVS
ze{wQku`om^J5dTT5gohk!Hc?4G^GOqAWV$E9G>3_^pE`6{8Es_`K;6?{VI+XHZ_3Z
zbC(@#rTSodYkwHb&9c#3%8;SR)s)gqycCcZ`>PfUL($8hz<)KmEky(~T#fIT@_4~I
zbp;wJLNi14!sk7`10?*CD&zKk>G|_Tt<qn+t-n{#-QS1_tIDbMGqKUg#a2B)THVFp
z$purYRm2m<GHsDjL%TVZOzjE%&6}^L-^F~fOd%(}SXGrnYxY=*sP0V#Oo4dR50LN%
zVqI^|QMoA9zb$WZ@^ly^erWe+s3ZsgX#wFFM&y=ld-#JSUZZBpBA|y)4aPaQT-D?v
zdSOjf(W*h>h)R3U&*VwWmNeDdonMZrFEL@dpF0v%hwACH?_9d=IL2tK>Ff<bCryT4
znM#%>e2kJ&xtK=AQGUpgJi9DD7XK%MK%Mz*9wh&hAP`DmyIc|L(hx%zURH(`I{fDH
z&bu{|L-zH;+M9dVN_W!agHkA*9ch2vGyni+t*#r*OKkwn+of9^?X+w}{?e5bxuJQI
znB!NaDZKSpJ?zbnS$-8d77=fY%GNyMtJO#FP?liq?a}D|hfifbMbu&66U5&f#N;4=
zU4vk(nQvMdP?z=)={GIWvxRL9H~-n}KZgRIXTpWhgr;Iv4j0NnSMz%Gy&0_!O&oya
zFRclUBodt5!VFostdfl#REx3naRZ%3fDW;boX^2Fr~HO`qbknHYEnYelW$NgXTP|f
z@v3fgQAytQ43Up%HTTe|&|u9vp#XMEJwYg)Z}5Fciv<?AJ|q$TGVn$>vy9K~7=}-`
z8MUhj;+H^c!X9E>5xQdixe;l}ggv(!yeY2(=BXT^_pXw-Q@VVj#(0gj80JL3gq_>@
zUM|E9wIB;p(X%bV-+~`Ci-{8Mlj-cgdjl29&nDft_?ur>+8C6|yxY}CdSvuHOHdx{
zgtQBO%DtMx@g8)JrBJ4(rf3ZpmCSF`Xfc?^qb(nu+(i-@>AnNDwmbaE8nTpTd}VNE
zw6TZcgsA8G%qO*%Rca|xp2grnghi{h;kT2%+FwcHNiysPCSVD^rHtpE_)PoIUrie#
z`T4YdkVQ!|p)$!pA0qx_L9J=Rd1Xs<N$(~b3&P2rY0Hv?e?NjuEyR7WM&8~BMV;4&
zfPdH9Nmt6-dzlX1i21@d(<fSg6I4C`S+C=~p>tSel7;gAJUv-q^QPwxiB!Eluln;%
z>hHp5><XGD0??963eb(;GG7HvgqsZB{mP>Fb}`#uUU6N|w#t)W6hs)asF$<p=E-Y(
zc2T<+1cO(UYxFqZ#Pj~z{f~e%F>Xo#d!<2bDLZ?`6ye)142+QbD5O_`F~PUiglb5@
zLY(B}l8dbbo8MiQh$ojPznj`E=Mbn-TWg$c`o+MN_V?wfsylRA@a&aL?Un@x15vM2
z4{`G$bH)TBRx5l{!BM=b%*1K1!70+O#b8Qlyh>qoJ!bg)MQz&87A3g@BaBaFTJVPp
z9{?U22eB9pHhq`15IxXU9}J>w6=FzXMskc*1ix#xkEMhU%Oe?;GrwgfM!fcYwQN&h
zOvbt9@t*NLF3W9_CFAibxz=ouMH?+DKhdZ|`D<|EC?G@n!`Dt2v&K%4;J_%4EnIab
z?@wq*`zi>ja_CivSo`3}Axw>HLz2vyKcetgKh5yxhT95AgWad)-vJU9R1AE8{(QEN
zO-@Iw-vbUGmdm~GSrv&dhlU#1e(~Rfbv`~2n?HVriqxdm>3^+$^~c3*j-p2gBsedP
z9bUQo1|N}hh!}m>b#dj->yY+H6dL3tk&sasLi%q4l`0Yel)&^pfvqj1>=^_1)o|fR
zQt2+sw7@XsOw-B%@qi`cxUyox#G=H76I#=2C73^QcFPx`%A0QzHvF`6E6QHMTlA>V
zLfq};Lh|+OqM11y?gp}UnBTh_j=FI#(}zlJOBiph&g$d(nm+Ti&%YVwzZk8lgxaE1
zQ=Br=*VALd-<WlmCOJMCdLN)e%BU5pUK9GMk(Ac@Zbu5tSqGH0Tsqp$2OiR59-U7l
ziR;RC+2{4*SbZh<^QodT7^pXQ7h)<1(@Jok3bGO}4{wguTE_Q7+;g^4P1G>Ndgg-<
zoKeBqeG8pO>_`x_^pSquzpt?ffh=$^u*Gcuhf6v^Hc~m$=yPLn175_)^i~l$UtW-3
zgqy3NuQ<Qj6((CnMfp^z%p2;X1ML1LnF{RW#n@a6{=o%58NPhSppzA?-UcwGm73x8
z^Baksmbj=i<GA)H2uVs8*{06NSgVcfu@TQ{i;ip;1o(E#I&l4_6FFam-?A;NXY()p
z$K(>PB`j66rNc+tD0bOt@3+DyayD7zKS~0Z9A;$)n(|-vH$OE34QDzNNfxZY#`G%{
zM07H2j3(`E_q`wff1izYrOKW#qODv4#Z{0qt+uy1thBlPnXc?TyLj3A8z)0`Wt+U$
z_VnG8mKT1GMPFBEMXRasr@hM2!DHiqRH*cYp2hOy_Au%8N(0l40TTL>R|6Zp`R00{
z>1>itXgggl*?V-Ft7u`nu5y=EDB#!m*&~z4sTs#8rr4~5_C<!6qLYAGm8g-=Dw<GX
z+_io`gld1I=+~_O{t`--DiaW7@@VUpP>xJ0K?cYwl$23o$DbbjIf^tLt;i|)U5!Tv
zl&262!Str?7AMR(H+B0qJ=OSyDLL`W--uFwKVNLr?H5fv+aZrZpJ;)iN=QeZRBVbQ
zURiKVm2<Mg&L#bMUlXd^>aAf`=D}#FMv6<Iv}3C%)!|v<CzDrh8>>Hmy06RU@9|qK
zyct4+O{X_Gi!&es)^cq3yS!V?S^Q!zmJtAK;*Co;?_zfPz{nkd-QFJpDJ-VQ`94)Z
zgmq(xd$?Svzu9H8OUJcC{u=uL)+Qw*#zrcesHhcw^L~E2ux@Xqj*vySA)NZx{GB=H
z4pPvrT4edo?3zwPG4>L+3r`BOH3MRL0RoP7Aca}27WXZiv5zi>LJU7v!y;k*<On8q
zYO~jRR20Ko%k}7)Y=fPGEBe#2f9nc16E01D<Cl46Ouj?%#LOJqU8#7zLlS%049;tp
zZ;tw~_XdkVtiKh#p_w8zlqacYXeeofT!CqQ`hZw$1Zh(!!K7R5t6c7)JQ|Iz%19MW
z!S5Rrzu%L}4;+RiGgj6YU@ot24XUNMJhNWktFn8!R~2eR-hX+6&Q2GgcsO?Umcr=^
z%=2^$*8gx|@+)qqRf!pd<Y)qh=1^PWjBMPI4a{LJ0v0Sevits&@%k<TINuEes9h-_
z44&~Mnq71Qzm?Pv+Ut=io*|$C*hwhGpn99+K4@5vxBY7A*s7#rwM+Rxr}1xcZw&Uq
zJVOXx>wo7#X$DC|@FQ=fZP-epeVK`z)<Nrs!!Bq(#w4PeR^dLH_}zucDV^|!+{XZO
z?~2EK$(;WnyYD9>v2Aq*qqAmCh4L*J*9b8hrF`0JM!f=Iz|N3JE{(Uh7DP=(vOzXL
zI(4`5g7%4<I*lJ0(I(+M4~`1jX|WJK4oG?Oy-~85N75QL^f*S|W4rU+Arli-b-jg0
z#Yd71H%LJrJIjI!cM~0@sl#Wr6?cL$M3J$oft}{&_Y1j;MdC);blU#;1{I53X>TZX
z;Ir996O6=F;zDk0r;<t!rGe#dE%cF>h7wF8U>ZEAA}^Djco6pEk)q=DT7Dq%A<mg=
zMSoNQcCipql~md%pq3Emy~^`_u1y#uUo4{XzWtGa!iEcGsv6X!)Xw1*rlT}zc(HR=
zDI>5|z=_#KOBx#dFu5KX>FMbp8Eg5?x!4G-UlwPVaF&M=ePqSQI2x0f*3h1d%i1an
z&h1dj5`dhkj1pTyr&<U{uj2a7gy7p<o9qS)BVxU7#Pb_&t3%$!ckuiF%$FIvLMfCm
zJamtO$chkQB32-tz;KyLGiT7!?aXCK74DtR5>h|hjKPxJ*{Z6lNV;gy>8}lGheHiy
zQwXS$F1ih>Yc4OmjX|&4+uEKHW{BMiyqaAw7~B(N_JfP`>4J>hAiIemaa*6k#guso
zs_JbLTt>aaM3S%ZyBi~sGuII!XNj3v`$w;)?J1|c?ayW$g9WY{3Zh9DnJy{j43Ru4
zpFr?x02CAy$&L*TCBW9qtKLk^KdL=@k;`6RU*9u|z)q3INW?(1HAd~`oo@sip}K>}
z(s@^yYMWY~R+#?vn~>$@dCHt2?f8;-n-Xn?sIeU@!K01^$uFxbo?+umaP3APlBr3=
zdS%7-RcsfPox@AK&-bk(a|9iURskUs0&RZ7rvg+xZfCsp7TeNI$<4Tv%;iR)e3CWo
zLC;dcQo^JHvU2_|2O;5uS*UVm8zhk(uCoSKB>54M{Xc$4^$!gVnarDhH2V}t19dI!
zdw7C9E*^vm3-*;R=<*rH>yYLwMgpKFJ)Jm9UCq7i+em8Hp4JE1n|@VLFH#W=)Qh#q
z=M&FJRpPt&_RN%kyD1W>?>)l=J0vG=Al7|AM`)5!z-ZT8c2X1UH+hr{tWcWCLX@O<
zoMx?|&14C-?ZW9&$Xkn(UjS9o2q`^W^D*L3Z{A`jVx0=-kFhg7-kpv#H9hS^R&u-o
zeL8%*JLFCm?TdEvhYxcQCm@NVw(Qd~<RQ{gpl`$c6=6agtbkS<duYU|f!;eA5eu<{
z?J=g%d&zh;m@`QS)?kjYHNlirBE;tU)`wLOvyf_hUlX(46t>&VHuk0OmufsY+gV01
zR`Oa;!Am_ZP=DXIU^!@7wV?#sr4w><sHi(jcjrKYcZ7C(O-P}da<zJQ5W2XyND7lO
z{$rX?c#~;L!^J~QHAN0>pen59zot|F?n*54f=K`DKy{R6XK|E+^(LmKviV>L?F^~p
zJXS-t29onn=H%xeis5==wcB|iBYwo-b+D`=xMBuI^`yI^I9U*?;jIY^P=hYT@>2`o
z!vbJOqC+&s>w7kcUJdNpE%NUeTQQ^@b|Kr>+Yq|XjhW!zyjDp+(h!d>-q0{&j%;%f
z4kP8~=NHV_&?quhMnu;~2;0CUXoz?b`!HedCoo>R5Vi|uBAP+!bvcA8P-2XYg1t?@
zbboyvoxTT9xADoZ>9$A?eQj;w`$!GQnGJWvp3Z6U?OllqwWcFxV$ta6sJ}3-)+J0J
z?tw3vnY*H5b3aHtvq?WP6J_KbuEF=Lz@_Pygx34VeSmE8o~4a$o>Q_f*)>-7rsFG;
zz`n0a^<IzvLB%v^PpKnWR5zEZNS&WY9xUs@q*gy2?LX(pMIos;Zz66-vWP9FyD?Q4
zF2#FA{iVXKf<oRT?u%su7lH}KtS=B@5q?Q0aMJUR=F9U(AoI*7RK0(goBg2rs&Pwz
zEXghwRBcgeT%yAU-tqV!0Un*6twSYP%8>S=PEHi{l$YJI73kh4ng-49=hsTpg^W3b
z-(ezJo%wxhdmJE_y9_SRGk$;F$on!;6`FY1UllHVLcSJ)&sWdxO9ca;bSUbJB>`zk
zk7I#f$|6XHwX+jyC~<Em{Nx8fxsgzjWpF0|SRXG|X(1GobXb=RP3`zC4}*0VX=vqU
z%PMF>oK7udj?EILYXnk}_h$*D#CpPA$wak=nf_$^N00PjQ*CZ;Znv}g7VMyR1{WeS
z2}=BCZsWcT(~koB=j-xjvQ6sp-k&b+;JjgDpq*KpSK-crD}tE2Kkya+yH$n$u&fE0
zGDN#x4EY?GM)pIjeyw$Xf8QI}cmw6s(wQMrM&!(a%4N47mru8xdPMvCWJrqTz>Ym5
zXc;oq4m@Fv%6xVAF-k?HZ3R(!0|(rb13DmWw`?R;Z8C*vwYVh$7PlV4E~qrMk@F!k
z*s{=D9*8GdOEnFiRCa?u3i?0^LrGkQl*#Low+o7jiaKUpHb~9G_uLaxnJy#NN4K}P
zJDK<}Yo@Y)oCohdc>_jbM;|jYv2XD{7n41db)uCJ(c?-pzo3^1Fz2neVs?I*(3h(8
zii1DKRY3<~JY8#E04>ce_cLN~etuU5=d@<O^^OVcb^)iN63Kp~c7@&VEBfQ1>R{s*
z3k!?FPd==*cE9W3Bwf=#+PlpU_$b8V*3N7vBT-!JsQ%UKF<<<5#h^jAIZ$3UhYP!Y
z-7Bw5BVfv`yDvb;`tzprV7VOCf3aihJ#BO~LZmE@b4JqG!^+s$)YRV>V51)Fkd_$#
zXC<_a+4Veb5f{9ZPDqJY<C=?CH2m-|w^8iYEw)OZz4ea-lkh1iIdf}ZtNq>G-A)_f
z{MCc7F^7VBNw8Z}hLZdSV$kp*+h+UiHhIe`b&7AS49s3$zR;+RZzN)c8f%qL8d`C(
zS}<HSsbV93&dT%w61M%CplmcKJ>1)p=SUjv4^~M@Fl;l*)Grhy=@;~TFV0rSf9HWj
zFNPrNO~drY#lTuib)mO0=+3M_0+huRDCdfb3OJ-fd8<wjIkWcUrimNtP}L2(kkw0A
z#JtZ#TUwNrNTTYq)_65mTrJwIZmGumV<D@c)!)a--F|DM;mNO0NaO^RB5?GNfo&|e
zHCW!Xxb&X#@@Ky|Jc+FsPE|J_<Kirm=!3Wy0H(o_U~V1>TT{?t?KLo5BN@<Exh(bo
z-OJtVI(^@6S-x;-A}X7gvDfzTV}pg76&i>wjSyMJ%Ew(Y&~w4oUcEJ;85noo?8rv|
zf7B7eFFz&G9Deb|WbMa+<V9L@G{cLMc>VI_%aR`py_H20jb^fX-<eQjYE0`M;)35%
zL&FDZCk%oexlkKMx?(kn-$GwH_J|iw-qnYAQoK}C_J&XYJNz@m1mazK#Wpwk{7mc_
z_|0dqy!!ehJ}9W;&)BY#x_UQK@Nla(%4pD6I#F#PtUVWs4_FsqiAW)v+JK!~{>i|r
z%Qm@18&Ns{HlUpBmY+e96rAsPuw5R{elxbn!h9+SgvYlB-E0Ig#Z#n$E0!RBMeOg|
z=6m~?-%+RT^m_z=lozZ?28P4n+CuMPheQL(2DalTDG#zHE=K3h8(B(dp(?(T^TDGR
ztlS;Do*M5MzQAw~=7E8R@4yHtbgRU>D;12A7^73k_yype?>~I_9h-LiGcz?+G*#`L
z#9t&-Mb7!o^@dtPF~4Dj*#xS;?D6Br+wR6Y@fOnftVS<>GyVSHKYBN1Zkjk7bG0sR
zV7zZyfc;>k{F(qRVX5Yghcrta-^W5pQ)J;81Ebc2K#eUhoP|T)abL{Wg>ln0AI`;3
zLx<9XZb0E;(DEN1N&$MRWESGa+&Aq^W#8PAlIXVf?9bCKk}khKyTZnnH?#=6JUrH=
zN*gP_&_8m(%+6zfHu9P@WAW3jaw_-_SA&GMDoQN>24?5`O`YJ9)$54Hhy5rCHkl!}
z@x58*lN{|#0hPHJ-@qRq$m>Zv4GaZ5O!;rOvMtm&R!;`JFI+OA<_zs|)6Y3UqJt_q
zw@nB0^<BYGuRg48x)KP<9`fJx<n9BDePV{U1ivHJ?d+CKz+H1y8`*P`#e`raEPT@k
z#_27)-1v)pFwG+0w~@Ltji;RBzzngTYp7djq(68XY^%A_N=izMk%8iAY$iFk3yX?4
zxDah|pVI$^5ZK4Re=m(4N1m>j(nkv;*6GW>$%0<71SDVM=yL-|srcDn3z=$>0~>Yg
zmQV&ASl5ONvth78`c3<8n4x|6oomtej1nGX$*f}`ARGJkckPeNP4IDcTEYVnRg>@W
zR+diN&Bz&19>k=J@lKvm5g9NB8faG)r@%AxYjJsb`37P)tg7k~Rs)QhHOumbW{mbG
zr#=ONJj+`A$qmFLwU&fke#MQhq}!{;mG+Z-`las`ks73HADjf)QG#7AUzJ$%DV)E3
z4!PJ?RbE~<KPTtKXBIMo4SNW+FW|TC{|zkJ{wd9DHxY^<0clvfCLb;_1o<mltwp*1
zsECAK^9r?EHu)i@J!f$NI5WQ#zn8d8&%1sAQKn?hq-EbcfI#k%&3_3Zc!Bi`uVDV9
zBqZY7;{s_o&UDp4lIZg8to@gFN94qS!lh##_DI7<%Ch?)GdC{h;2fQ^`I_7*m#Q-*
zRWHq)_h#GI7p_`i#8`Sg13K5EG}8%yrX6LU4-^iA?>Wge-Mk0mWejsDAe3-&P$Jfo
z!72b}FTUH(-o6VUiA*4Rmw&T_;XQdVhdNhGRVn)5M%)8wu)S$_mazC{IMi434gSy`
zXlmMc@O+n7Qqn|#m6~xX<@j)I=Xa*7666B1C8b02{WmVIU*JQm2VH<CzP@)2<b+;q
zQB`@5wJ0oq3S(PQ^k!e5(BYhMG(jo=<^(;(H;<ygo;Cu$cS)zj^CFkp5F;@Jc6N>w
zp4h5;nR(D{m1Hf7W5*cTfH5`geTlKEdWgOOz&k9c(x~}^hnc=)u0N^o2?GQONGq{k
z-4CQ+D!l7pKn{vsUVen>gKKm<!VF0a36f5gBaidyCSPO`<W1_1x<QyOi%#`!?tFlV
z{>TiiE{=seOZW)j<cxR;i{VJbU)jpXCS4@O-(Ma`zX$0rz4)Z{Q6IDGGXV5__)iWf
z;`gIpaX7WjEnJGu6tmqW23E8-tk&9NyAL@t8#V);Ak<gvS-2#C?ym}>?F50IVs7K8
zHG;M31s+K{vbhXt=znzn`BI8{KN~2C1n!AT970{AJ4?FzSgtAYjT+Z*&)z|orwY1p
zsm+%Ou5tCx_%ARvXf`k~FuCv;$2u7xc9<y)1v?usgTSdY!tNOo2U3=Gf~j+NUc=6M
zYr-A*GuPNI=>6U}A`l1-P#RC#@-cSM?WC!>@#ORXTwGk74#a06E`Djg4Qcrjdj49s
zKWr#t&uu1JW#SCcLsJUf-)sj8A}eag*KguMJ^N^7thzIZZLXD^Pk36&5Ne-LYA!qI
z!E{Ot*h{O+Y+t8k;zH6((P#ktwfJx)JFNGGJigf=X|?SgrQ7~g9%#cDHQ%Ip`}XY!
za%LuNtR+DZAtJ_^K9+F0vlRAQPl~+?u}+%mXW|#l|500py{NEIrV^xt=&#uEXH-`#
z`LbUK>$bja-X0dbOw}OBSR-BQfLI?P`gFe)(+OJp_&dWS_R_|!vai#WkBsH8NuYrI
z(aO~mSy&ORK}LK>l9QLmN;35vdS2RL;C;!k+-$B}3hvqAbkI#Qpq#>X_v!qX*>%tL
z1*NezM*FYHeQ^B@3o;<y>ej^&-`jPhVfZ?<3+>+Eze)DR>PL?rd6E(jx;|5YB;$Z|
zhzFNGzFOlbYADeNGGjVuRrsPD$Tj~$!TZt|(DgUt9`Iws7++Ovx5<nxD}tJ}$6op{
z6GYyjJFdU;1>WEWnTj{@PAl(29X3Sfw-tp0mqdgIaG!agRU!gG^=d9LFoL2;18O#K
zD46w<L1`hL7(h+77o$x}#i)4`)sFR%DSt<=IELTcsV&ykw`;^W(!lghi5)p?-#yc#
zraAAv$$<We&;Y;eNy!M;c0P=`IlnE;)I~DTeOf=;8}q~#E(BHhMJgg!PESx&+WyqH
zb6Wd3*DP!`VLz#n_0XTUDy=*C0!ixfWc2g~iQE7fkcIbO9o=!1>@Ikyf<#VSF0Sw4
zdkW6%oq;7V7Z;D@S2UU}J-h<)tosmwl(xfZ=xvL(v9Yf#WU79|6zu}`r$;l4M!T3d
z4D?&;e?rBSg&eZ32F$0yXCka(XQ#0B&RqC!6+yT*XECndYHxk^0LBb2E?$`#*F%xT
zkkU5y0pukYw{ih%%>`s@_+IzgjGCIetpmJ*tOLY{cb;Vm3unJ50ZEZBecnK^<W<DF
ztlhGI;)|_n2`RHHRto7~dxuI<_dsKr%=~d2VWMk*&#Py3`=<;?OFXY21@u(SM+lcJ
zF3l;W3$<r)6G_;x{Mnd@%U@Q)@$Dx~Ki8gEr!U25&bY#;b6qCtJc~-Ka-$geFhUf>
z0T~)>9N{>+K;MXPmq{+K*ZqLnzIy$;C#CLg%y_zVWJ6qE69$LPpNQ;dIj!ddoz1P)
zc*}}S(pHEM`wfTQqihF8?g2eN$a<JsEEdKa_aF<_wKMFv$`*ojw_b|-=0*ouD!4~>
zpU(h~Vz%qidlX)gbKn?zTEalglJ0+0Ap7dMD|8{}z=E{aFYW%SNU8PqYmv@8yHh<0
zSC80(@KRmzeEOzw>!UJY-_FIpT4Zq=lDk!|QYl~l!~k>Y^V<RyY(D!dH?mq$1UTB&
zHq9tUN%`})9Fz{dc&ev2O2?oEmW}SAk7N>7+bu;ROT<(Mc0L?d_QE87-+Y5Dx09-4
z&3+LlpRZpsG(U<VKlLSipk*T`c_zW5<bLi~ncb<Npr0qQj|J4S9&&)&L55Oy#w@Yd
z-CN4A(C6uRK$hg@a`WIEIqqHIEfQwlu!w+&egZfggG2^gK^#zWzsuIi{lROG=Gi6o
zB$W6?%A@b}SAJH9%u0H<JwfkJBv4(C>&R$OyT-wk@@vN<^Y9Ah(V&vw+Y@)g$1lO*
z<kn<aHU2mY&*%IPl<p3+?n(!1aepS!U8r9yI}-cpysIT{=qY3z^lSOaC1+mXMguUX
z)a8o_CBG{LAASOcfG4?j&%;ct>`pE-a)vYYWwG^4d$hl~YJ5ClzX10_jmuTYuuzH_
zOj60sP=!IZv*kjqchfNX_FbzDw%1PuOBMJR^exDW(Q-|DQ|(tHjXHF*6hK?3AKKpX
zWcCNq+TO-4{-c~(m!03}97OMwThw+#hP`fUAMTnShN+$cPh5P1kuF}%Xd(g4WO4;1
zrItb%l?{8U%sr)2S*UR``h+d1WE1Q{*vX7JfLSJ2lkF<=S*$oaz3Ps9`SzMuRX0Oa
z!$kP&X>agpWl6xe{_MxR&;i|0Zf2zAAk@;d5QNIdI!r>(qYFXL#L5ov7fSd5d+u94
z>a79BwrlV1xayXX2l0m&pL@9sP{(`ltUuFZW;-El_ke2iNsDsx&F4rYeL}#l0D%RM
z0o%-hry_USJ|o90nqMJ0^!5i$I+(w%Qce}ADHSJ%_rrZBM&4+0&73xi`36RRT?R+i
z=+6v{q>Tk^`Tg5njSZDY@0c;>boB;|C%%{x)e`07Aa$IrQ45bNVU!t|sO+|7_$T7e
zExE#yl-FYChTfw15KW&X<7%Z714TY6)3!wUoc|bR__@O>ZFNV|nc2?FLMgz4L7Nr`
z5=U^n7$uQ{wR;Hvdx4K<r}sh%sTGm<RJy*)PANaX!vciiE#v%QuK|+fu*a<b-h`K<
zFAf)09ganv1M#65b`Hyx^X{*}=XEovpnV5tQY~;1%Rh=yEf++b9t@6N1*P<JDEvaZ
z*`jA9H^TyEG*qASmoa{p2z|()4RSoSmkzMBv5Hv!(}Tyiuf_|S%CN|BL+Ry_Y<Jet
z;GUogcUJ3wGnt@W5-fE7)4=itJ`L(6k9?z`(31FP-e@b>soR+?;XeO$#A30*-S?%O
z@S!Sb*=9bnq<|OQk7$z>wdL-RzdB8{Em}HGjc6~W!rfAiLi@xVFdun)B0%Yfr+|GM
zxB>S5YE>qp3Rd;Rfcdlsf2L%S%H0NN25JtXeU7i|{%|ejHnIgYx}H($5-;H^nY@Sv
zpow;hp7!~Of`44>Fnwqx`*r^@qDitiC!vjX@aW2-6cdI>weLn^KltNnBYjCJN2TYa
zM`0B?Bdaufp0;t%sk0q|MO{A5^t$GO_ZHgypSnok5U%eLX-|&4VF|@1aBm|gY8j^+
zMXiKjfP-Kq1b!B*w4o=h02iRl`w?>bWvH|?2LOu9Q5Xg?%ioKPYia$pT>H)HlrnB8
znBv+i&7oKn@}0uHACJg%Xzi8xWx7_=Bs(Qi|HJIPNtvr0r68Ul8ZSn7$QhZjmIW8w
zZ;_fwAVJ8XRfMQvtACofHncT@=Ks85*QXr&M_1nb=+&W;2f<-2bZ~823XY$gpTVX!
zI1y-(a{CfcA1z$t$~lRk^1U2-ovheXNSCSVok%9XGvNNGX32jz862L0$TDy(Kw|A}
ztl<1+va%34Y;{$L`^UkrUT_-gN;Xk=U|m4Gf&cGED-UHP5~c-@3M8S%pl%U5mgF)t
zeA48Jm-L_Yzl3z(JjNBCb~as)W3BU3y;iG+;=uFe$bMbmgX?#Gc#n7>N{)m`)>UX1
zjpqXWzi;K37J;KWrtCAmrV`{AjGXL<qnSA+Jg3J<>1hc(@_3Na7!SWiA%Ru9S-^Ow
z&O!JN{|6@jBIiwDC$Xb8=rP~hp|^*@o8XkcQXY1L*DRV*;h|z2!?O>r=lE7*y$G?u
ztgu_E`6;|P3!Q&DfJV+BuhEkJ1(_(m;J(t29(r1s8H{(W<^yeqAt?uv#^*mqF4?V&
z8hVBppJF<alL|9n0FDUTBA=7<WoYo3ksHlj#t=@rfT^o$69tv>|8Sgd8D#>Vbx<cg
zBK-cBn7b8VBq~56o0e|2-Jla-Rq{C<EsFDO4}Hb5Q`a|R&Q?<}VEHpKO2=1jPj-bd
zA`rji#uSmyBcEjdIzJ!&2+N4i;oFx4#|91wgB0@}H=R%Y&pHAf&{)EH6~yH8pY_PE
zT(DcFt>j|(`|JR(&zVvBQ3HbXy=EzCedd!8y1GVhLQ3LQ@|9<m+_8WhdcwX7?JwGB
z2O!ta)60t10r-x91#!mXq#M*wH9h&qQ!J^)SP9n(gD~<{&c{klVfYm=#SB7cB01Pu
zTB62S%noNTKu;E<sKH=Y{vRg%wmCRA!jo!#`_x2A5uD^NE(3e#JB`jAN>!4@Pahpq
zj8~uAls>dn&<N=K`#7S6(PMH}=~3PB#sAU8f|o2tT6zOcRJrLb0KRlI?UU%<TmCLX
zlTqexKj$k~^p2zrxcvCM8@YsU6u=)h^T%ph6u-cM^>3j0d?HMZL@^GYT{>|Z`r1f}
zNji{gd`9XuOE1AV#k>?8qxbh#tq@>}4t};n&Zm}5@F9M`AUF5XbrO-I`2Vl2?+m2!
zjr+e34n>m8RMIIU*$T;aq#+}Yy+TrUve%L1WTvu;jDFU!lf4cy$}AyUWn|0Vp6}^-
z-uz$uUwL($`?{~|`d;7bvkt+=MIt94IwS+Pq?ga|Ct7=~4CZEjsJF$(4V+)h=K(8Q
zq$~jAs`T;ab4{#im~94a1+3zI367ga0;>&p(<^A9xWXrelS?D{pG``4LrvIW+1wzL
z88-Av0)WSIL0t_c*0c*8)9QkY-<_{P<lRD7>RX~=+BDGTt7XnvDGbk-si1aFrzep7
z{>ZhAz_(OCTze?Xz$)bgpOS1Z?&FTWQJ3%K=_<QOyTz}2smKeZh0tJ&wV|4g65sVm
z-Gky+s+O1k9R*rkSu@_45B`YKl{i@GtJa8x4Q0Wb+c&V){h4^~Th{q4Vda}D2o1|e
z&F)#2TMmo!<rrO*<vmnHp_{jCc!}mq;N}?7-;ERxDPf_oq-yo(VSNsU@H~C48dkQ)
zwH(zL^4o`P3Wf&9synNvubr7_nXi^zBF{3Nb<UD&j>fIu?LObYtc70r^*v6K;*RDI
ziQ~`h8sH5kN($3(g*EiQ#D46rS?Ctl7meJpH{tn53A^O)lqY^{j`55GR)-NCO{AVY
z%iR>mXY@Mc=le)GFM5UG&W0agZ=P%GJ)deSuP48!q#r9xQBbczx!n<(cxqIBc5&4q
z$2`ZEaGBKDr1s}5neAlOA#Pv8TU-edyxCY__4(ydq(<IdDdpH=Jto!E{toFSx1{@_
z7vuj9NZI4FO@l(wI}Uz{|DFLSop@9Y)yl3N9FU)I^Shq#mMrq`AoMygo}X5^xXtUx
zihOYOXdGfw53tPf9pc;y#o=lC?V$?35F`VbNNeD+P@==h@M{cK*>^+<EiaL$TCFkX
zQ%6?eoQm&iK)#J!L_F-Z+{vW-&M{lL4eoOc1%n`95nH;0x5%<|xf8W}84;ZWG4r~W
z_Hiu@z>+?j<=L?`GNA>L@%rxn>uqcbYq&Osm&M4QOm(d35lyM)LX|9bU^w1=?*5bg
zou}yX^Gx3KKIgJ8dw%a&)5%qWT0l2oq>fDRe+ej-5CNvJMGEXumf3ew1dYp<l#zo~
z?^8rC8A#gj`euu2)p-+)SnbLPs_lsh=8QuJ?Bzdt6<P6iGXS&YQS7*59j$7I1&lLu
z#*E@v6C<!TMunH3|LF$E^mDh~t_HKiDE0P;Ai|eHTAQc{DfE)nZA_YADBX=D3mBp)
zZRhC_JA4T3eSkh2V$zhDqIfBeUE@nsPx=gVt}!@AlXbalxQKGwxTJHM_8h(t{K})v
z{KwSKoX4bohyv7)V(B4XX~6SmeGpUs2OTO0g<n#UDrT7NkL_->DPPuuk(?zzl=;eQ
zv{ci~qW-Ar<oo6yCNN!Bw0bfEnKi(E5Sx25XtH8q78#*c;Hzv;lCO;syonvQFTrz%
z=9^Of@#PjqtR@<GnI2UOHVrgy=#KF`facmAZW9w&?7O4-G8w=-4wJt<pqwrgcHVHj
zoSl=A*1u|7wNBnBv#k>+E!*wQ2I%L)=ZN6JJ|cU%z(c7i&hdjmRu}d>;|oomzTV!2
zFRa|Dt{+$n#`}}x(n#Wz{4n9*ogHT`ldGb!JpHF8Gg)h)dd<WQ2NU1`e8OK0UW0<Q
z_sT1t+uKKsVrl1`k!j=;qi)ZErOWZJ(!R#*8h+hVb>0Cqt6M*yfJGEe60tiD-?nYt
zwBcVwz*X)W%3P6DPXuF96<sJT7>PboPYh}GTmbciSqVQQ%2&WV=oxO5&fH+pXfxj<
zC4oYZy3l$DyXukir19HkWo^M*SiO%}lCw02E#E9ib+`dEu;jdcT|wz^*33OSEeyrB
zKIx^7k05jDW*+jw9t@Xt`Ud!m!|N2~A>jC9jNalJqBqx9MvU8g)0Dozosfzz;T3=$
zDRwxFnD0o}vwX^~$t?<nqp!8c^zkbWIi0^pMmz4Zk*C0Jck^#26ku~3`iS4(3xie?
zzCU|*^D*n8R#SjVnMmL8PtP^Qy~RXbj}yG5GqP<=dO794d{k$V-xtM2S1TX?V%8ol
zt~<_}&F3eu<tYgQmC=yq-?}&>3XYLta;THt#O?!o`$T&4OX6Q$gZP(skRTTV!T6Id
zH@_oKMtHa>dVb<Ls~qexIKA{~s(e|?t2agq7u^^Nz2U#nem9g$Dw!H6wiULfn^Fal
zA}c8ffFJ;#H;@*pOku*+h@z85F_}ISI#8YDy-#=1JdqMqBTtrt!_M4&)sb}-D)|_X
zp%3JLN0C!}8<Bp({+6)I1>AOrE@vF@J*C8v_}`62W8@-KOOzH?I7?6M<RZf<X=kzs
z_Q)4Dm!zM?CtI{^jZ0eOBj$nd|H+++N<_2k{>FsoT+#5E{>xvjsjsp?cO3mnhkuO`
zzlEaDuV;5zBok91g2VD;(DfVp3Ohck?ONmlt-egh6Exi~JKe}j&wi^>i<n^-;$=LS
z&8?r(3QkeP$n31JA=UYjMgSy<B+viD<Bu;?X8OH#H`8M6QA06Cr>Ih#8dQg4*B$T9
zFkwg5dg<uuf(Jv-`u%<7;G>{r3RVocffJs8BX*M;*he9%rDM`(5(=h>xNN_U9L<!q
z+4Wnm3S%j3rvBCQmGWviq43f+%Tla?(7+~}HHQBx9Lp#~ai6lY5Y(B2cGER*Nmd5=
z3#T?ty>2Ls45v{$*>*J*5WRT&8u(9u%&k_l#|(AI)qAhX=91T?Ze_IW){g6^nxz$g
zR9U+>8X3Q5825-4ZYKNdx;cK4Cdb$6BFN);>$x&nO~-X&oE^7D+w9EoW1v6$q@+V=
z-h)7<xNy00RNX}+l4?mLW0X}7AP_#*&|z3?f3noLFtzYhKF9f3!ln-xW+IWCb#)Zi
zAWqR?F_+vAvWn{BhJ+Y)9WGR!p9xHir@-*5LSg_(`3S1)D)@Z@a)<vZb&PPhH7fk`
zwrithc{>$^OK6dgd*I};1hpQYQTY<E#$6MwvGU18vynDy$Psd|F3Hpcq1($ed0#ft
z9JG8N7PUq~>AnC%Z0l@ca`(K?h01d??{j%?zgazz4vl0uQSOkrJ*go_r-z0tp)c7?
z2ZdV=SUWW+q~iKp!i(1j@auHJCoMUjV=>0(KH^U`_kU(Dx{ixnY?$OFJ12^+1&Ji9
zzIgL9(2sV^%u^T{zr##U+aRyUc2V8JG98Y5|Hxs+H8_nqTcMRIpZ4PTne>J0dmUyO
z)M)EhO17_2{-MVC(pE8*o4x#OC61p8^zs0mRW{J<7PqF%KwK|~uSKLpg3nZ@=jA!D
zSw}A0E1(5#H1_e24=N_XX-Iyn0=fniD?h{MLikiwMbJ<UvHuyTr-y0wfC1)e%FtjH
z2*SQOCzw%y-h>t4clr7|Jr^lXjXHzU%U1?$TjttRssO};T0KeMc8cN$fAg8qR4_(#
zfaUX;_>C<q>TT+uX?n6;__u=!qyCL_f*BW_%mIRr4dd&0)zoyP67$6walZrmG;JuD
zOnVe?6Ec4se~zWY`^VfUQ?5v7GSz9K{Idrlha?Io#z&RtcV13p)SY208ZkSg5++Iq
z7kD>wTL!gD9K)^b;%3;IzR@=UgfO!byZ**Do(XMJY6XjX)Lgc=fOKb;0wMl;eU>sU
z&a&wfZpz;ouF8Xc^y{6yWwLgOL&EMyQuzoKV>F(kEGr?eM`b(~sHX}b%!Fc-0cMk8
z4K=X~^Q1Oj<ahk_w+z>Fk&<(AfBLbP^ZDaf2!ih_$zE4*`jij^&>~~p7qREzJUH7X
zFfp$zdd3mcX|}~r6-(o57)L$na1I_Nz}LU~V@7SOQ08V%v7%?#<1E(Z*D#<#?0>!N
z;?Gr?8KaZW)Sj+0YuoD%B&{f?bb+p_#IeiL_0;`TQR^&X2Q}$(_;SwdgLnC8nYQ^b
z(NMC#qr_tr#1jp%!e;xg36MsUEK_aIjjwEqIDM6HBHgEDkeOLL$4i3hR-oL!Iy-Am
z@8f;Zq9aPxaWBq9^9D7_mhyOWEbNyBZg&0M>sa;afnX*NETdK-5cl_x1)+xeh7@82
z*=zUydv^<63VD$=JNq$kq0Rj{r_opazYz_O`eSohqej6>-1WaDboAdQJsH}#Akrk0
zMGR&g)0?~Mu{4$)_uxx#mG9;a{&x~5ygBh|T^T_aGDwApBwnAzD}2b`GGtie0iop$
z{^=b?+8{sEHZ%6{Yn@b14@&F4I~ymwBHXMq+@V8Cjwt8r_Lza-S^mW9%0k$fy<oR|
ztDa$oa*{6V#npkrg9ara^vX%Y#?jHg!#P2^WVZ=MNdU*tE1-<}bk^FG-gy(HfdKs5
z;mc7_wZC@l-NCe3a@j)?8b4Q?HQspW-)(D~CA-G1?ZvJBwbzhdb@(QVj_BnB6aOfR
znw`x}b}!qTMq;~xK|X=>K7<e=7=5#vV(hD!dL0($Tj!iGNG=K)fm6e$o6Ajc2k0@1
z$Lg0;F(yjvmS3!$?HT(~Bx9(b<)o{cASx5!2>0t_U-9b0%qxd{Jwo(4x%ljTz?Yr*
zW}c6`V#uv+V=`PYHqFZv)5BbL%3{LHX!!A=akU3kUi6jkUYdT&Ijk?!cQxWS^UU5q
z0kCtzNu4#lKX(O>+s+@BArdmNd!rhyGU&6&eag+?-%_>2?CilWvsX|3lc>LigJ(gq
zsfxm!dbzas=tOc_+P$-QnpS~1;q>n5SXgiFv$M0U4!*^(G+jZ`o(e3WAY(a&LSg8Q
zLGX(Yk%}UzF^gycgGVmGqGBwh8M!?KP#4|WPtv3f)<2CtGs!f0>&hv~EXT4D8L;I&
zN3&ZcR6Oh<vLOkj@wYXQ4DVxFhas4!5iz!=z#&3v8QqQ`bk*oPb6KWGnA|yXgf`Yl
z`tR@d=@c8_G_Ku8MckY{iReZfv?*i9Lm*#358ie1DOR?~MvEk{-bvtmi){Cb7`cZM
zg+aoeC9{@UOZx%&PoR96g<v{2_)F!e3cX<E1t+-(7Vhx1voanlm**bIbWNXFxq&or
z(2k*Cxy$<PZbnghz9&P`k5Aq2g%E&)bfP~}U=(wv|FiQcxnBu{n@ZT4PJtXf`msyM
z;bv!3a88Wo!uw5o8aU}|S@@$7j^+H`fj-<CFXDDj+%p(F%+NO8<JgZt@=jl}>8gcu
zrWBgNE-p0NbL=(PPPk_0{ZyjA;?(e)9@8I^xay94?WC;L`XI00T}4IymJIvq*t2GH
zr@CgJp{iH!ATJisc#M70-Z#Ek?`(=YA-=+jIoyLAKA#Lf`1X~aPq?DDtZn!nCH;$V
zE{5FypRO~qRYux~IXN=2a7NRPtpiEBVJA;ECly<_zX^mo#7OkIgny3kKiVDUrSOP>
zfEgAeOqBi{*cOpft#)=+l8oh;M0K;DQXTTBHcF9B!?$F%g2jeZ<;}+|t%f^C^m<KP
zAXzWffH8gl;pWoM`-oJsnsz}kVOIxRGbPn=1{qqYSTNgVZy_uAaC&-u&s)Cd$rTQ~
zuaq{%IYu`BE`(4DPu4p^^+7%^56(P}(9IhL1?f9JH!H!1s|VGxdY|~sbKVV(3>KC=
zf2JPO^ftDr^6xYF5O`AANF6Bp!fUu%`mw=?-&G7_Cvd)l>_x`7K^3b4Vg%LF38b>5
zxOo0ySeDLdP2V*tM1Jh63N|g7ILa3FUMo2sjVkc&t3A;a7xr7qpJoZvt(lN)vwt_t
zQ#|UrfNG3<$!#*i53O1Wu4f)K$Bpq0ZK)wjQ<MI<CUIjD3>MgK#RyD;^;sNi<rkl%
ze6FjUXYdFtk>gQ=I%}p7n)-KHz1gCW!**T^^9Y0YV!p4q9?*9fi&>fKJfOxWtqA;Q
z%v^$ByR*wAyLmOku*Cui4AHInqpfxBVCT*fCMMWI@P^ix;KGnNy782gF$LLw6G|C}
zD0JN+*uoH7;Gii=Qo=c|CtZ5*jdNwP-8(#(6{Zo0dpVBqeMM}#$Vr2sF3*g1lu2OA
zS`xiUC#6Bp+9DHj#%-^yB5T?x*Rc4Z%|_D2u_;=XjZ=o>QXQAcmMjF>$(rdz?G8C7
zupyUDSaKdLH88LVf^6jd>#Gvvv`uPyj>mDu5<Hk<$m$?$7B6{t3b5L$pFb|V^x}OD
zE!#F*%eVf{42BA;^g|A_W_2X9#?Szk!d;eGQVD}cPU2|AfQ(0j;GDJ0w0$~PEyP$H
zM*N!8j%i&Ww74UuB0DalvJRd--Wv?b;uG#200mnFMS*ZpVH73mluhnl$I{ORJ?V%Q
zX6e}59;~FIyxE!n3PC35JD!EK_+cI)cxB`i%G}@9KGS7rdgqrj@SGs>J19?2PnY0)
z_?_)%>h8;_yb4*DCHJGq_{nl+WY@6jPcb2SZ<#K@G?18JQM_E1WxQ_hCOYyT@!;p`
zvnmEo4JyEWbujQ{Okm!fGvrGkJbHaItrwa?YJA3v9mBW@3wC@(o<D*#JIfu9($iV9
zM-CXHN^^=Gvu7$X-{9<%yVUd;{B_%}oTRFFK(df)QrSzm?#jo^L1F%p@SWZATK3F3
zu_BZ(mqV-!Y3Vq$|LPrsex=(y3sUpBA#@vKoMUDcb9B3e^P`(aw4Bg~&(P{-`&;@N
z@QY{M1e(Jn?0$6z^Z4*x@@>bh?g{$f9zP+I8FEAXzxMX7-?Q@f;%8hVxenh43m_x@
zjQB}k2_+l@jgwTDr`~HqPLj24Zcs5RC?;lWO*als((}p{`x#%^OTl4lI71uvUtd4}
zQYL}#ZEAZg%ctC@kNNSjlFW1!EQr^|Bgx**DW3+X6sk)W(P1oM#XGYHuhh2LA5?jw
zE7=zB;N(Q}ktAN|wVTE1u{!+ZxF6#7W$P!I2R!Nt9%$(#!y7suZLizTpN58?N?Xx@
zA;w}~%s|XL=*Dk-WYInyd0|_ZEHXIt6Hxb0N~V@*OEmMv##GbEp=Dt=!I6@Q7$Z=B
zB*wQ?GLi2XX(0W%kGP96VfUCoX$=Hn7BXTbk!j;f5p_CDwDv9!j6=Rq!L(|I{#CEE
zg8UcdIxcy#MsvTm)GG@KK0bfT^HYUx*eTw~k+RprAsghhAxe+u@CAQ1|Nd4LgJIl+
z@tDyp$-^-GCp!`=jd2?2gCDAmu3pfDM?Xkzzwr5-<V0h?uwgV4>LY!!dBN7ai`H0u
z=yn-06*}#gHo6gZDE*7Fq)eSqqr6mjfynPBP_ev^r^fKFjTvEu%y|gg<m^3qp`6=|
zwQ{v`fK7I)X*#N9`8hYb=%s!CC4j{HF(27>T6I52T@7NJcwC#DSDHrQgX66Ei*n=L
z$6z`|=+Na$QUQRz-?;@!A<DAgc&-`wKG%GvsM+|#{HalxM<<^gtg6iLYJV}~^xG%y
z=uBKh=6KM7a}AvfDo;(X-5D7$ZFQ?KD`Cucw8SyCy;OB53paUw0I+s>o08%w*?(?U
zO2+6n#ctAgy?8KboV@(-dmqwvmnTyn`e6uhh(A2@{KDGNr%9E|M_I10&4%h%v(kBa
zB!af9(0AQ|k5FjH=%d0}9uP{ut>fadQj6@8C$kP*Ng<-}t{;MMgUNQD+3EC|R6nwy
z>tk{8ZPb!Xk)q0D$gZA8sZsjA`CG64G^u)7g3sDDhW?2xTHOFor<8L-OZKh1=~fbP
z|G5F*A-V7u1evp-`+83ty$|o#@oTEenc#+>pOFy@iOlOnoi*|m=w|wp3USl^s4f>h
zlYIVe)Q@D10B!Y+l@%|e2U=H@Cx=>8AJgCoGED-^pPZwvpKrT#EQ$@Hmy;;VV&V1H
zPEJmiqCO#2t#+D#Fe(CrEN{V+dG&W%;$eCB2}c%bw1R3^2I)e^&Q>8(dNY~#OTz2{
zDeoLRWIj{a+~Zl!gAx&WpM%hH@pm$aw(<xv%{s@{l4dLNvmc($Ps}8E(6HKn@~{uC
zzjYrnjd>~bnN}}2eVO+2<0ln|C?>X*M`?FZoF}U&mK-XS!dS2FIAbsxHVzJCNxnMO
z5~@h?yZ>jaWp3MfHpJ(Xo#t6#Zj|w_iUr9y+L;ay=ZyN!cE2`-TBKzebDM5OgR!-_
zaJvNI-L9~)&XIYv_i9e;NwcB1_PY+LflSA4CXy-_Q&cbr`X$3fc~SZecE{-d7<*AC
zb{{r4mae3qP1H8&OKjKa$~cj~5GRl*JIWpw_$z$wTw1u_huFUA%IcVI4xRnbwR}wU
z?{N*u>ZS!u0o&dD%o(OO?8rP^8W`@v+qebFFjfjFw8ESIe~UP&xEIE925e(CC-Cpd
zN_{o-+9wasDXY)X`EfV5s+3rB{iDd@wWt!k#xvM&jGn7+4hpjmd8pd7-ELDtZ97FV
z#i23FV<hjR`1m%euOt|PUQRI+Qq^In$yo(QPQq-(N<8h1>(JyDste$3`uG`Xg$1Jn
z3+RMl0zDt>eB`TP^qSl5zSjX+R=24}LB7gKwd81xSu8us%INuWwxX6L>Q?-7(lk;K
z^MLOiU{uz8%x?~vA1+d)6v@cVwKoR9?^RZEyQtrf8wSVCP8)jlWF3o-a4q%m&e9Cr
ztt#D4;R}MA^w+|k#in|9YdgDN_H|@aQ<EEV(5yB-?Dy5?u`r+<ReT^RO(C+0GX;DN
z=$KBaa;a&VY=66vSL{|r9BH=Gj1WVF-by+DIvmO>xE(j0Y!{Z~Cb2q)lQg>6jOZWb
zdNC_@p)Z-=BqyXcB&zeXrR2u~TRwSktW+_PR@$)d<gLQ4yffg!<b$YJo^r|Fx_(0%
zNq}Xetfuk+UR^OPsY^(#=*jP@X9(A;XDj|f&lZ1lru#>|<5lCoySR29ZRbo1f7G!S
z*Z9}Ank-j!abx$ztXltyO;z|88u4j>JSn7QBmQ~SZ!jh<U+Q!9A93YnO}aj-0V+6d
z)_fU7lfH0DDH@*vUi*>Vh=_<zAX+@ZpjyA_oG}s9dceXVXLAZGL}?yaCiE(1gF~k=
zY>l!|N@8_t)L0<=ubbQ93BjoAyEky!ZCK8HE1bOigKz(JiZDUWBs~0mqL+m9;`rGE
zeK0&IVWK?ZvX+kaxjO3g6e9bd?Yc4^Bb$D=Z-*CwSL;CR^y`bi`%cH|$ei|CCzb&N
zxS2c^)qPF=E5t#L$Yp3@$YT1oXs1oYb63*eT2EmXd^|oUkg2h$zb-tlK09%oW%M)^
zi?7ng&7yLwBz%6#TB?0aOXF_T@mTips)761F(kZ=+I6dv7y+rCyy4;DC(;AAFBdF%
zBR|uM-m38w)$(!9Ia8j=Z~bqggc|pE-8V=f{Z_Z*%*0pQvMyeGaGPwIVIDDxN9pkb
zR5T7=M+yZU+^DsL%skq5{s%b-RxQX~ketW~yqQ+d3To-|$V4k~MmV?F^5y6ANxC;E
zEuI{ETD7#Y(f+NUH#=GwWixC=>7Up7ee02Djv}tj_n^pX4)yKvaC?2RP$t}Uez{|o
z5%DO=ujKoc6>MAvklE|v7*JzkC*Q3?<R9@c6!hEa^=y~o)l413t|;Kx#7^IvQbCwt
z6H9`|$uK>T`U};CxFr|8%N#wF=(wT>ef@iMo%yXF|4!CHqozOT&(+nf*6rGSePwMo
zoBtjQZICs^CWCks8w+SKRiV)sX9L4fe+bLP8f47Sx<NdLunQHm8N|uXZ!$VRg5WuY
zsD;}*Th?oQMW24`B`E&GKxlm<J#e{^N|4K7yW73Js3=l<ObR&|`SVup9Yuz*V|MMJ
zQP2=KX6N(o+Va>8nKc|7S36B{?@>@i$I(gGu48LHklDLF*IFJ-#^0jk@e%Ew3b==z
zTRQ4B3i`HbXih8{uN)%zSf4L=1G>8Xyx9)Z4_1dmqhIUsWCjE!CCWCpB>N)eM~1U^
z$=@23u)y|dt%O;F@a1eubH1$ho4h@I#?=X8TH1sP<CVh-CyPWydIvhi>9FIf@b*m+
z8k9cCjyZ|Y29d!@^M*-bPn^~FhFymVgoN-B&L$PV)#Ld0CVjFM<kg`@!}8?tBc~7N
zYowtRIG<xbr?6rzxqc(%TB~Kn=p@K8mK{TX>2&*;U?)v?WHekj##D4a%u=+9o!nbb
zAtaFC^fN2ZgJ^}@x0y*Bb@C6HTlo<)XgAitBeN)eI^aD#R0O~@)wnguv5!jf-Ws@!
zzVF|CdR>Qk(gSHfzgB+n!D>BpvcF%|Ao>Pky1F-f?TYUuDv{oi(MlZtmX*GT@PnEK
zxpyP4g~{V%jIUR`Ra?c4VI=1}^AlP+$zorSkbh!#5%G}JO(_Q{8~B0p(HTdm2ZqXB
zVj%*94r}F~9Gmgwm0m956kqHXX(i_ALKX;JNKq1>V76=gmTlNppcsruF`N|anEQIc
zsoMSv#ZboR&N-AHS5Kf@RC!-VOjCB>()V<qS$>P#N*Wr{X(MfmU}7?cN)eF7m6~h=
za@l^@;d3Y-7E*BAbyHt1%v435QR{{4_uLcLh_e;E*)twUXdI))&LPr0U{Wf<iK79c
zM-grvf%GAY7;-IfTfc1`s_2buO5-;wYKrc@|E3Ypx;-)zd^uBd$=4w3$;JOL)!CQK
zyvt>_yCXesY6l4Cu@b;{YkFVeL&aT?uou&@3XBb~LRQI4#e!h(fGo<GM%|wy^Adg*
z%sU&P8Ly>Wtl&5LZ1pH=w&FMRD(B-FHMWi^3>j92Le&nUWBY*DP+=45BXP69FCB24
zW;)1i95{jZqkPesb5NzTYV0!b$5pYHoEEozFhazSe-JKT@UFEK+^nfOm)b`v8Md1o
zsd|zj%AyVL<}wgI{Fpx1+ZX=}hoFdj(6^6@vtXGY!wx-xHvlBX5SK_lMrF-?-=8&V
zaoNFSro+{Q4L>W@!H5W<H6w=>4khn;;&Ay{nVDtp$qu@1#WbrgvMBPZ{WwlEiz^Ov
z)2cJTSAE=nE2kfLv-M%Mpr;m9cbetI!Z)A!&i{fMFY@=}C*E6$U11@`A_A7zdMjHE
zD;YxdfRL<6mrvxws1jN%$4+B?3rXzab*{sf`c<hTz(@QCMT6aP9U5koQ<Dl~r37QV
z47go<p;WC2m`1TZjenxPe5dL(DKTVoI9^%rjBQTaov-X(#8tL!_4L${=g}(Dmi9DD
zgv?#m=eCAV*`AK)h7y~A{35Zt$l{q=rC9PQ<G|b^9!ZPgpZ#izDP$)Vf}D;fdnS$O
zh1sO}{f@e^2;2lR=})%FoBkd$)z{bO?>;|Jd_T112;wa+?z!=fIC6Nb7m7<{o?p1Z
z*-mrRtTvT%==~)=yz1JETW+NVQ%X)k_&4DatG6Jnn=#gB=5FU;Ex)d#O4)CT2Q^WV
z>O7rLpB7P47dXNrn4<cTrf99r1D&3WdU?j4eJ{ChmF?3petlNAeg`p2AVUw+kGEwQ
zuh5WM#MEzakd*&<A%%W2D(!(N{o`>`K57>fDW`-m+A8DUZ#O-BJ-gNaAF)=ur}<8n
z&YKHo^DR)e;^Y0<`-jJl!vfQy&ORA7b_AWOzY%*WlBLjVJU3ORWZsBn`-N6H-NI!$
zk5ZPc8(;{dG6%#~Rvnv%6qw-~e9<J9-U0a~OfVtSoeYN1lOzEx-3<hFWl}oP4DF<U
zB8}|dzKYvRoUK?~A|(63Ff~vf#FUhuCn3@_n=PvQYq@SOYn!_6JaU}qRVz9fg~ilm
z*u4@4F*XgXF%HscWTC4;Xk;W^6#YaTnbUijsv|5@n&~}Sa&SE$IQ)<Arj32QKzY2o
z^NW0dt5s2LS-H942DSi>332((Kg9W7$1WsDIa}NU2LO6o_7=D%y6r)g?Z5I@j<}7%
z_xrU67O&&hsVSFT<Tp<C8s=$jU2vIfGoT-HNBktAOOXXh>olTHQ*a+?*u&z^pqx4(
zU2#f6Lc-dEO7rah-chMY=l<^peIf`6$;4*Vk7p+g9l515mgK$x4q#I*ucC=_WQoe%
zt2lu9Yky=XDfjs<7tP*Twna;Nbqc5BRKbsT)Yp&P@-LYmTz~m{oo`@$w$xV}9eJ|O
zlKt`r1CHPGGH>`Y0SZ4#(!;H(G4?m!uS`M}T*obD=H=DY&X)MX?~jb$|8y4fD2K>N
zIi|VkUY!$-5$iPEX--iN-p|T5SJk-#ogO&?goe@yk-~_fNH4^02B=J(p`Hq9F}9t5
zRps(-Hv*j_^R+>^zki_3N-IlSk&37+`SvFTTlmd8pJrT*Oz2vC0%VTP+r0eln!iLy
zdBpr2yAD1vx-oqunDUUPlcX7CS1R#2BoU934}qcy0)ZC%77p2JJlc-A;UO(jNTs)r
z`K~0=j=TX;LLuJv?90CJ6XNL>AZrdbux+N?l)-nRe#j}D&7)q@!<krx8Js#AoHXuw
z3QbR^k51lcyx1Kmccq>^{=cs06jzuP)+UP<i!i;DB{+fLmRG6U=aniB*|wS8``csd
zVRg*@7y!YcEm+I86u=V`{XNg%#)myJY>5MOO>t~Gt|@>a7rGJ%Hy!Jr9v;-y(Fs$X
zdSgK2ws*1TaV!IF%@gq#Z4$7twKdXN=#J!*nlg|&wO>UzMP+V=c&DsPKf`g3uIVD3
zlpb1G(~rG!rH>}Qu<2h<Pqsv5A9}oDfMdVn5yo<wx2upAJ`;P}DEX60&i;IBux3^A
zwFB!MA@ul-wu7FF#`r`&cM!!?LX>bPgwircoyNI(lj#k0<k{I$l^c3)d)}(p5y~ZS
zddM<vMG_za5-s%O26kK+<QKLG4ROZdADT7fdb|9bbW**<i{kEH3J)66N#gc1(Mfuo
z?+Fs!?r-YoS7hc5GW0r#6|salLPzCCH=Ul-SgWgu5sKY(Y<QE3v}4le3<1ghxjT-0
z%-d7si2HlDw5~u<J9afWKT<)Z@xI_{k%+Z_z(NX%6YmgP?7D68X~sPu=kybQ<-h41
zp56MsECs^J^}Dx+vbV4&QZYmzT(=?6fCldVXCV{%(Oy5X0{D-1aINR@jZw@(`ovwb
zqB*ztJqpmdXIT&RaixE%IA7QG8nQ%PCJwOD2I)fnkOnkxJFt)T;8u57{Q7rESk?mi
z1?K+|BMN#Wjs!QHLr8{|XbI+*!S4Pd;iBnES7YNpXU^P29x!g{W2Ld<fNMOmo!>!X
zfT1gH6Vr-$gpPj4xtV)5^QElV*<nS<Q5GFkdVH19&Jy>KN}A%yPvjcH{(VjRt$lkQ
zg)R+V13_c;6w|70cP0=+7X71&7nyUZ^^d-9HE*wtDm<wF_49zmN#Wlw=s?bQ+TVi^
zr6#+n3_mLP;WX~YD*BJD?moUJZTaXP{6WM-M5P2pMFd4I>4->4U%DbKE^$#rL|R1T
j`EKd*|MLNRN2>?cp8x+3Y^E@P+!H}=s6eGwGJo<vW@uqj

literal 0
HcmV?d00001

diff --git a/docs_new/logos/upenn.png b/docs_new/logos/upenn.png
new file mode 100644
index 0000000000000000000000000000000000000000..36fb621841f4f89ab280f4f040738a312174b601
GIT binary patch
literal 97063
zcmXV11ymKy+uuufH%Pa1cb9ZaNjFG$BVAHb3P^}j(jwgr0wPFCN(e|w*EhWX508h#
z-Fxru&dxmZt0zuVT@e$F3=M)H%x6k++7JX62L7|4Ac0qSmg5k?F9bVjHE9T{O-8@9
zc>J5%N=aJ{f&v&JC@d0!uE9%TI}qf}1wp&!5G0ZfLBwvkEm~sW0FtGOq8#+__^+V-
zQ#yDj^h{1#*Kc7z&)Z9PmS~_;l*ktw2k%wu<4;3_&`bA)sTiFlmV>H()B^-e1Uwn)
zgM*EWJMMHGWt5y@deZsMFUtXZ!(W0Ag||g-a@z&rWn5R^JZa29WK3f~c|{$&9t|TJ
zmBqS~&YZ;kCdD{O^&4YPMV}fRLRk+y8bS!*zY9RuKc3*>@DXYc7MYI(hnTfnl1$^~
zB6T4)m7mlwr8fw8ahy$XP?#>GY&#jHH9q)ok$n_GvWmi_v^}c9dlbo36lyYPiZ{|4
z`|;SNImro>2GJ|ym_b8Bc;O7LV$wJW(m3=GrX}1t5%k-GD96x}2j%}?JbTid>;LZV
z@Smy?5+k^>c!ygUJ1iZ19!nuV9WDACh6g^3APqN(v+B7F7jMnwKo6mU6TPB_*+U4F
z%81l(P_3Nu`G2?d3ZWpp3+_!?`hS1ng&{FR&;I*@{zwz`2rN{oR|wSqf8sbioG|Oh
zp?F2D1d(M!^oUB2mx+KXg%|etTZ>+k*8asSMaD@xSm`}#)nB2+CXH_EXW$d!gL5Nz
zcU!8AvMCV#KLto87;&f7mVfNdEQJAF+CLIu68fJGa48U<(BlXp7LajBH5*_&C~~ga
z+9~MMqr937#O>;tdmzN4EkY0FiOvjr2H_deLNoTACLH7EPr@`0x2@&DnPf@0n#9EB
z>e4Z!;Zi24V5Kc^)=egF3>(mRK575=4*0NPCE43(Li0@z9wbyLnZbAU6q=GOfxOt{
zb0|7PiQrbESGmm%m45u-r-sV#P*A0c`rdi^6zj)mv^L68QmNR2>*1mwf>#1(2u8d#
zL@HNzb|W!Np+@PMM8X<*q5@vWhn3orWnccrr~|_nS%^sybq;A0aDnrQA{4x>Z<hOS
z%#dO<nHlLG$3zw?_pNzz9u7v7gu>8}mIn^JRT7~f08JAsnhgxJg*`m7+C8}IXm+R@
zd;(ko-*29U|1QG&>Uk-Hd8XLoV~7^?(){l{gBEDpFzaFX$Itq~f!PE;+p_W+o&Oyj
z?C~`>ctwjdDp?M2C34S-nGqE!rNH<V6$FP`L`c~mftkD`G#w3|4u&R-$Y-A#n;KQB
zr{D$xZ<0a2x$oH%B$H5r@I-#0rXWAw`f5u38{`xOYKztXU0u_#<kcYt8fA`@<RUk}
zt&DMMM0j=F)BJZC@(3$rq$YGiD%zS1Psx&aCAzg4Zx4j!Ft7_18c{YDsb9N>3*{rO
z`)}cRT)Jg`C~(v#AhSd0*3OhiVv2=B?~=hH^C0tnJb4x6Fnl&SYOj1v7glXh)X-{J
zRD}>J<vPNgGVhQS>rwX*uHrr!ZJO&@|3rGz$0J^!Zuvd3oz(q^#<IiiVJlHBwh1`m
z&ra)64x~yXF8Q-cTW`Bo<_!I-X6>_f70viy5Du>x0{J{`xZP`nD$+g*%Kef*%na($
zG0YMZuIev7U?9`LiOQX_r}ptjrS!(+m|&1QAdSSWuQ99rkbx<wbb-a!t1zml0xptb
z0uC?wRDO`dN~k95W3T$#)I4<Km)cE?evbM?5{x!D*4D@ll$H^&RYrBG`hUsxY4@l>
z`nKlyh}0(L5!O{Ki3Ti@7})1i7){D)r?)qYM1<~3jS|1j&7J!cJ2^1FSNZeL8Rcw5
zx#D2ub|e?n2<0}E#eA}*B*{M_vFUpzJg9ieUi-659Gnt@@U38u`Wz7>`Blu%f1++K
zQJ6cpyxQ*Q7V^86kcb*s|Dg6l?)Wp8*cx}76g)#ZT^C)w#}PMsl7_;$f1?e=m7bs_
z$V>27=Wdn$ncj^Nr|Pexu}U_$B#I7yb_m>G@w02D1|{riG7bK#L;F%Due*}WCzL2P
z`Je7{UQ6WdRs`L$eO+Lt+A8LsptJPzb6Q6;!++i;=%YY8!Dq%<Qfu;R*D%#S1vin*
z5En6jCV_;~^lfJN*Bs<qwOpDBZ%l9ltQnrt{=Q5o@_22>*s@A;LF0$))-qli{z*g*
z;bdpD_D@m8c=u~u!_Dbht3$R~pAD?`*_ARU4qwf)CnyWK7*nx@`Ftz47>uV{HC<R!
zIZTv`-!vY#n&0xoi7QzRlW)}xZBRxJd$aaCwi8tSxfU-Bk;YOOCFdbbpDoq$`!%mY
z%avITTSCY)&Pke?WD#U|Fjb<T>dzd{jcuU2);8yGi~XWl(;uNEvK>G{iSKdh<cej`
z-Ygg@Qnw`$DQS~6bno%4rn3<`;%32g<;8B}caP$m?bGerE>zg$ar3SV&-)_0j;C{c
zuR{PAA^8;(_Vn?pywv^C>GX|?1f2HB!%k*cOM^`QDCppjrXyt9Y*YU`Lc6~9f<J-q
z%qHNzSR3K=e(igb5btNLk7Y=4Pu#iTAzZ^Jc(rclA)N50K@~+fNR0Psf}O361>HBf
z`hQY<qn9Y;4z`r0Y)DjzPmYHwDNM7QrTc<h@F;C4xIWn7dpYW%C++&lbUoBHUv^**
z7S>}QO%#$VsUyCg6X+8^jVDRl!PL;oT@Inrj*fVV^%;k;29`V+RUbj9t<7s;WfL8n
zO)u9&=9ITkXnaEMrFPsOoNyw%?rPn+HFtL5=hkdC_#X9A>{j?QHqx1W;`^=ro)@m6
zeunztSvG`jpadY<nC!dl%>}*j^c=gOL=2)N`bGjl26cZZI1FaxA`)>7MtRD@_k<lx
z=uGO$!vnVnJ^HdM(lHgzoWjRHI~BVBI$3?NTTgkLAB8SxYWb^b)bCsb<#87GOk?WR
z7fmwE5EDk$n!R8GYcFOi-_LUEHO6bL@rf4{pSf!S>Hhs$f-s%@d*)&v&T6&dgaYny
z8=tg=YZTTtIqN-H28%B95&?XV@|X`#&f9)?50e;-D;)1(LpjFD-Yg>}?A5V4Z%vwi
zdU3xV6zPy3?=0Ej4~542qzuxQhN>uzmyU%#Gj+^|2MBcCvrn9~m5~ktZZ>|)yXLn2
zL!dq0Hf;aIOli}Py2^6V{J|~hczKBQnb?oqjla%g+%?8l7=M2esO{-NI$>{w%KjJ{
zbX=KLE(RmYq#|I1ef&f9{pponI2@T*l2fLh`@%=8G1VluhiSO!^7YKq)|6O3)$Fh2
z&kx#=AR*-+ID@>%=t!~w0^Rpt@J|Rr{V|yX)tCAg^0JNvs3{;cwtEk{)Dg|%US|XZ
z`C_MdPBFP-KbvxcpUE4BVWT4!45Knhl}dagm;d_HcnghVNhzz0^#}Mz%W+tl5@()f
z3{dIZ6li?9UinA#kVTHYY}x<wv?*joaNABjO;?wozekrt0d1BfT;X->cFy+ga?cpa
zU8_Lf<np)47WWLe;Enl}?bmc)j|R)}m6V4e_yYH9rhjT7F!Y?#xtDXh8Ir6s<sVMm
zOUTp}be(w&i0W|zc3iK?z=lW<>7B4%WxpA~9A&+<wtjzl8o)2c<)Dj<aqyjfBF8JF
zkLjbu!=IZu?|($MaCG%0{m<WvClAVJ{rtjJXNLaI-fa7+hFN%&lM=?q+h$J1TasGs
z26c3)(FSPl&$Ak(*e)WcTv4gMqZ+Ss>am<jI*sGTpgn&!qfjL9ARw-W=Q{hwr=H=+
z150vpOlz`kd?U@mJ>O>3H#cSvJL1iVD4~ek`e~jKhV71LoS2rRW@af|CSp9CF)I2q
zsbhMoTO*lugGHq9kKg1bw2SGvEJ0Q4{G;>o;yS@qVNk$T6V(c-u_vqw)568J@|;(0
zu*Cd?D$xCUL<b5hGfAWzuKv!-USQDKpUHWr+3aJ37X~)0UxZ=35)gB!$0Faqze`fy
zupLukLK#z{tKfJNO1Pva#{(Y-^4~VoMbK$3N%j_~>=|9Y40_(+pshzY8*s~m<4bw7
zcD1J=<Zt+W_%-Py-e<Ds@}KniSFK%3YPUtauUKq3MPF$Y#@o<g8BC93$QCD3>*5V`
z8E||bK4guc>!O`?0UL1<nEp+^CmPu~WWlG0vUK=0hrh+?89!Of+*&xsNva)zidf$G
z4iQ>CZQ7Zsr4;#E6{Wms0zvmwbhQt^#h1>`ks-cul%CSdm}y~0VXdI2k5#U9i!fzC
zFP4-1DeqX`G=93Gx2}KB<+o=Va$+3IK~bSt%U$m-EJ_WVF&L8%B44vRmGaXJC-Xjk
zzcw3%?0HW7Zlu4IEm0w(lDR1^-TbKvr}l(-0`Z|{i8|pKB6YSQif~g}hdN$Z<ZHAp
z(gSZTEj5_m&lZ7ORg;bzUk|R>b>rY#i-oail!z9=0ZvIT=`Zmm34H=_!+&b3@9Jd=
z#1_J)I2z?Wp47_4i!&Wqo_07L1ykMkA3B$ix}|+&Ezam9Fd(I%vvxzUf4`eNK>EI6
zw{^j3g&_Hx7#QKGZDQiynPCN{1^CgrAw1=y?OlBPl_DgOujM)0E1&26NqMs<kVflv
z-dtk^r<rQoX{)wzdH3p7IIi4@vh>1|Gh~O3z<x1y)tnK<!ieD-Ln`SV@%f8?>e5Ev
zRbb8ENz7ZZ$|sF%q&n7(LwNDEmV5q{x7)OJk?YkVu6@cDH2AM3MySng%ni;fcGW&l
zEf+hq?A|G!Zo6uIsoG6<!3%3aMGVLBXI47fMwIwig4@s8B=~J<jN~(!p2CAm(qy$~
zgbaJJJwBWSwj}1ugk#x-)q#kwbN>H`s+K~MrvI_gpNypp@s3_xDuwfF%2W&to8TKf
zFtN)Qq{8QC`$JG8J9!1~y2i0EFBd<q`SDn|-<81SyU>e0>=)D-BV2Y@TA<oW|M-{e
z_z9jymoCpK5+{5&z4n@4oU78;Gx5nZ8d1QMZ#D_~(Jbd3*XgH&rmG|Ns$anopf;`R
z*t7NFu>i12oA00M=fj-L3rG?{fsx`f5|{DHpznz=x0)}PyRPTCO%Y6-l9rkm1KDCN
zbYwH<Y)4zpzow@!76hwbpMR^6rOhbx+4es|PwWbFNzxquhWm$9M(G2On=r+kLC*T0
zJ&mK5?S#uil)fUV0XMgS7AzNIUAet~P=m^GL=qdm2bYlTBE%L}_0qBUc}xI<<>(k2
zcasU@#OnK3kAN_}<?2&Y_L-+|Zu>PxvKi8PTn=&o<Bg`TCdFSl7O^Fhzg(^!)nX4@
zyLq!I^-@_s`MWngKtUz+%k-Wyeg{@vq2f);&i4)ugUtT8gk?~XiRrEd6unI9EbB7g
zUiIJim7$-@vZeMImD@H179tkp63_#(Bxga`UBBCC<8ADQq?nQ}dibgErN#RgsBQa>
zQo}Sv?#DIZ_sS6YhyeB=N<lj_?XQjApq8Fi-`n1-y?eanNK2wH53V_@6@m%XQ8=i|
zST{O~4?RzNrF*mvtp`B>=g8mSblWY-RFH$4u-oDo1uGdqKd~aqG~BH0__<46P~@n=
zAwwDNG#y?~gf9SefHE0l%p%&<yin}euWhf}Dh8mVR$y<aixGd-t3gh9+`;;Sjt|-H
z9yL9eg)dKUe>rt(*06P41qN?zF5+P1x8aIW5-(;YkDj_l(xQgse<g=HPFC>&`z;Fl
z;eLjsrB!7TykazxF<LgQ86qTV01r(G_qXPH_f^s4=VT)Ga^V~`;XM1`XRZaqn)1r6
zFT;egB})WZ1B5DqT*`E#lM8G8HmIV%!q)uAaj+heQ&G?oWEDO-k}nu$%7bmI$TI!-
z(q@&ov%Zc?zIS1B@`I_om#Z=InPU?%bCQPXEF_iMwGvp3f$TtG^}#)<+s>vY&8AzI
zL%(`-`0By6vpRN%Ude_hqQTo|S*A=EDbwckwu-l2;w!@oOg=imrWg`idMS0FEV`H6
z!&FaI<URHt7&|zW4|1{w#ZOP5u(QylzMJIKxwaMbT7atE?6HV2M7~tU`KJdE?F4Zj
zvot;MK^JXz$E$`=&`QkaZ>!s%tNH^oSO+D4wMOQNuo{Wn8os_NZfBryX18!~fz!c<
zIaiKR6FkGApp2OuLWq>7r)Flu;#@(Zqh=;!amWF)0lhk~!=UMaPq}rj$PP@$AI-1@
zzA_{8L-X^xqU2{RxnHYLpSyc>gdTFJPrl=mT%j{C2bAXkh08}Kvg<m*Q5Z63!7obV
zB7;xQFC+N;_C-tTw@(*wvCqaXM5ml!Ll13g3asg&8AD`HAD&@V_>1;90plI4w|LS(
zQ}LoW6AKF(>c~#o%7~U$?;kj25JfIbWXXTSVgW}^QLpzqK9{c!o-uhhEKwErUf2s?
zI|qX;tRs*7_q#4(UX{cOU$c#omsCrqA=u!N`rqfht9Vkb$CgBvF<PJ;rG4b(DK|fL
z?&jT~#UigdgI3aUZ3uu~e%o~{6DPd;T?Z#+`(@ZsshqR0+Y@6flNdWgrbMa*A}geZ
zF;6EM7o-d4SHVn&>uaajP4;>ltru2oS>{Sbf@C%bdw+fZxb7$N3V3&rm!^*J>!Zlw
z+l?i*panX#D2uCyz|qr+uuPO1MBs!qBfK|Y`=$eJVZFlT5i6|^4=*7V0s6!28Tlw&
z;&qbYx0ky5Q*0!)@yuU{JG$*x+5#w_V6^Ao4w>Q+4kaxAps^8pT&~*D2?O3ER%cYb
z5VOF0U&j~t#I9|T>Fo@!U30E8E>t<}-0#9UeV()@B?y|)^6rg>&krx7eIwVEJ4MRS
z>imdY{+t!_Df$oEKq9a?d%5W0=NZrRz4{Z3`z|FuwB2-d>5E;H&jqGLbRT&7bOtqB
z^wIyEUtQ@0gw_|uM&)p~VSVfG(7ABC-0uyNS9A_y?J2^mneAW^g2;AqPzZR%C>op)
zTzP#VD5keSubB05Hj&M;Bzw4E)qD|%F&U>762p2h4GlLVrj%nHR4UnAb;d5ZCFe%q
z$)Tu4ElU~xtu%_WtJdcl&J_J>LiUT?+Y6qioHAV1L`1L^rVj8&%fk8vf%HTzk-*1K
zXpT2Qz_VG4<&(+Rs~Zt%E&u7IHb$TyhtZ4_{R-JfM!0=5(X|9LW@F-Hg{VMR9Q^n3
z=~-=nT@>fSt|#Y_4cgY&BoGusi5`eP?ks#HTHyN}y@knIRQJg#ZsGg)aXd*3mn15r
zuaB_Shq68CiJ}Xw7%T20IPd1g<|+?GAA3JVZlf|E>fIJs3rS)-#k`OCGI}8bz-RBL
zm)(D!YPioe(y*df{j-G+;h80P_r#b^)f_(BVzOEVo-Ichhwx+$d1Q@?d3!7SM~6y6
zx*lFwq4LD~*D3T8J2aYf7JckAZ5|AD$iw3^iuEE5=f3k8E@ein*Xz{l4)`9oj;B1Z
zLR8atZ+eusU#IYx=vEk9CtTM&oKfJeY{{m`=n|goi>J9}r2F{}k4ZBh-rI;IO>y{;
z%vU2q=75Bo8xn<tr<ShV`6ga_lY^m%nAj*X%R*|Zu|u*nb?z?2Oh4*f{mH;_f1i+0
z%zvM~KrVeTyVm#9q11356R!+IG<jKJQ-Jr88kqTWp~u(aQX&6TlLDH7wfkk0mDV%P
z4q2L*WVP2X0YME1Wc{_v+r$myk6SGF9-iB^R8T=0SBrVNLKRD+NMjFs$**hgm)T#r
zh6(?f$EFW0aZ_E1ceW>4t_kX>!AiqLOR(}0s0#(#zxfWWqSTCbkTwc6Du8k2c>PnT
z)_Yg*Azv))>Xcgu2{NbQG~%2>T?|NxknT0g&*o{d$`9oYl)*moGr7EsvRu37wV&MZ
zZ3`fVgT0SoDHf&O3766IeyL;p_j<S=HEUEF2tG&aQ+O1MVu&-WnbAsY{b_33n#CK=
zxt;eipHJVCpmTYv+Cg3DKX^nQFzd`~esv)BApQirzx%IMr<K%Kg%xpi#`KK?e0kya
z_@3;;`#XGIUpBA5SXAj#<|IhG1HQe?n;#+&Rzr~-v6yCWuFg+IS)ZG}O5C4m!BU%i
z)Dx>$=X$)}Ad0glIVH}FKRMQImbCYKZ|mpz(!Ya}8ZYeXM~j2r=Bfc!>;Ah6UA>Zm
z%k0eP_qFIG(cLYb3PHRgjk)jWz#4{w9;Z4_eP5~0w<pzAZ2i}r3cI@eDV)eUNboKE
zJr$$z+cjX{(=?z6xT`@b4*3;5?KRWy`;IaGx~w(`J(l#>vf=4u8^v-?Q`q3f`dlVU
zsqwFNsloO;oY%jnRo%LZUNH--bhbNb!2<ujqe<vh1XEy1Oawc0g8bw!<=+H}zMAdd
zS8)w%!Gz87hX%S`95=Q&<=JXs7NsI=K>3_Rv<|=DeFHSHlh5Wa9pXJjL;C?6?l6u?
zSH|3&s4-RJZc@arf%4yG^y6h!39}gSg}+%_edGNC>Kf|!_Pt5{emh^ZjEMv(Wq1>m
z{hqvyPN!dZ)h%w#6sn}Za6<&2TiREDgu>uI9To#v*!1wPU3uH1?yI(Xezv9BTCoF<
z5M1t;DtPG9#>CFc(XhU2&~~=1G3&#2YYab-oYDonyWvO{Dp*Wkm`3X-l@l0blUQ<m
zUrKMdZJD%ae`mR+@e6`9zAe=*ltnnMSXm%a|0n)Ns_Uc><@T>_#7&+0`fXSPPW#n-
z`{MzPnV%R3mTPY0=#5>W^k<J09tw(Dr&zb}FAC$Ky+u5hS43QbyaQ>5@SQgo8R9Rr
zzP1!0)imeSE>Am-CjrQXRrNj=-}qDa_y9%sn!CNIHOtTaHj`D)ScWN+yz+yP--2uv
z)JaZ}fb3jaIZ|c3KEu}i@Zg^M;rESi^4N&<*pX(j2kHKcJA73&1TuW7Yf%QyY{-@b
z>FE|$_wdUf<}S90T;Z7N3)Aq@Kyv02Ku_E<4o8FsvLHYuTB|xb3@NX^m)WR^iQN&u
zovB6<CTY{mw10}hVuEkdsQqCpN3i*Mc$S;*{T&@iba=8EN?WkR?&ohe-VR~Uv$6_b
z{wPmTw#lx&#EI5e)GmNhfhrOeX1B?e73XzwWrKly@l6~C)rTAu*3BiyJx7r=HoSjU
z8<9u>L14<Vx;=4QfG>Ru9V|A(e5xs`ahA3=QwG=#G*P}yJT4t<<7wW_{k9k7F855>
z=G|)^!lln;*Me0xh?Y-%=F3GQDI~;3Mb@(1^U#A#KC`;mubAL+x4CQ+wm15-*pZuv
zb0OC={eybBgn|xe7BEZ`FZ&vTk^}#(dm6T!slTlKUW<t}#}8B%)<<O(wRh&)F4RFA
zsmuf)AbShB729$Rt|@}Ex1Cu%L$P|d65_I@KXa#XT2nYPyU?>$J3*CotNZ$j0^=3p
zdB69k>E1^;WGo)C0&mZxhMb&1$H0ysGcKVk4QW3VpyAp0GpW=3&Ftg<c6@Mqrea2d
zs=N3}AlQT$>KYzhMoChg3|wn>KBuNIk;ENP2ogKHiWG>*qNWx3^KNkILSL>^nH1|b
zDew4aA0hUbI8dZG1VkFU_oss0dp_ptBgeM!>3nb@n{KZ)de*A*gP@yNVFfoVBZI|K
zV81zjaioE}9EJJ>A~N9<5j7|i2>(Eh0~mO?PMAcO45Q4cSJbT4ZEhYA>9oSt)f|v?
z=>msy6MES5%)XQ(vGmH@XgmjskKE>l=-2W%pooJ9Kj(XrFFhBcG{M-^7D`mfruwLL
zeY?5ChDrNxe<ha$6qx-<!b_FG^+SKzM|tOw6$5CBrs9cj?Sar_VMtn$D@q+fH`!um
z0}JBP)>Zhk)RGzSO99p|d)grN2ESjHIjvx?w6k_&h&^NcWo&S)h&0!5S&rWY4IHGY
zkgJ661>85Sl<oraC;puuh?zO#BQ)u|Ct2QcUnRHW#M4iA+oyc0mV`l-C8Un;>EbTt
zZ!T`8msw|ieG`Y0c0MU_yZ>I2ji?Q+P|$PR7!ng&Fr=lYrDVV1>>K^s?T_~RSa#9!
zcCs;>yMug7btjx%Wngt6)ju|%k+k)BQVgK~GKh^$j(RWl@_FC7M(I{@lSE68DrC%v
zDEe^4ME2f{7xY?s?0=@fz5KECf<)4eNp9Ws;t7T91yoKGw#^}l7N?6uZpdiTiJx8E
zO#1Xco3g<|=DlmrWDxQV-2IdH85WLXYJ9psC4YIWijZ9xr5tLLmV~nisR-otqOX{9
zLA$omWx@GRGiPXQno6~o34v&$e<Ri#APC>fb@V`z^9g0@`L4e2U@vcIcYP(kCBF7|
zpCfItz$dz^8)}J&B)>Wt1UM-%p(Z<1)4lUrL{@m-@;AQUo-tOoK)MKI1lBo=x*P-5
z+@|(9=P;<|bmyfPN#pH1+7#Asg4?U&4Q(##6UJN5OGc>R`P5T4^d{Go|IEkeDoVW_
z1cXWgs|_l1xgIMZ{I#89%YXg;v~&JhD)P}y8zze4O`*y$Kps%o>#{@hYFmdU`g|73
z>@0ZbR^m&t(Q>CVAmNh@rif72)gCA2ZXVFG`#u++YR&&31ng&r#}U!=CZ^o$^=o#5
z?<a;75@~+6MrT3CLGq}zWEs=d)|M{ib9Rhus}hQoPX{OEToaZ35G0=V-hDc(4H(=!
z+{KgKm+?4+-DV<d(q>o;fAlP!z80itJtBc7#Wyj5rxp&ZTs%blFh`0XY<G0F3TC6Y
z;l6Je3YDPpfm1<g`{_N6rJzeJ3<F>{B$z&{o|ahA8+9Z*zPI_IZ&=oMGou(0C|L5s
z*?lit27!^jq_}r-6VO-F)eTRkXT%PMTuN}yX_Q7Y^(Bp_pOLzyjOnn4eE`0YmqrM9
zB<?F0<Zzm2%m%@PVJo}RgA>bflKOd4M$tIuQ(mg^?Ri$SuVg8Ky&M@{Y|5l*l#Ae0
zMoTDXBIB4JDl1KW<+(EkZ*3c=Wt5vJto8i6=vBr-T+|O?1;2u&8^KWJL)Xk8ov3Sm
z!sR}uYB)^oG0}PBwUO|7H_pFdO?|pv6de?YlJbN*24$+56&o`giwvydX)Yzx#9`o$
zV9d1Rdi}~Dl~B{)ocwLCnx~!QQt95LLP*P`>Fch?hM?ed)Aj11)6!HzEjVB0Ct^Se
z_`qVqzDk>mkOt2cDiOuvSo(7uL*CGRZR+z<Z7B+BZxmpENIp&<Qg56LjA@wh)22Mr
zsL07&s(hM)Z+)Axe53%7H^MeCq1Gr+J@;R=HDD(UtMZ8=3LADr>4L-5A3?=nNPl>{
zTGNz!SRuPBdNh5OHuc{KS@cfwNX4b5e3;v_V<aZt6$7@5C{g@ffX;vyt6P86;hME;
zh7Jv&2j9(C(!q7OQ05^g^I*lsX-K!ysCYWAg#<d=c!&w}D73bi*79h1TPWXUXzOP_
z+a@D?w!XKI%5D`#%Wz?x-&<($4!FYvdUmpm%-neC^@!$cBXQ!~r&6r52$@@qmlkeL
zNz27~9-PLO9?{aBPcNphjzsu~0S%!OpdtK>%sw#hG|~BM-h<>QL*}-{h7j0{4O@o4
zN1mBkG@@E_Qq@Cv@zg^M#_*&cp)qmkS;q?)CY1<6vZ_?j%shbw^1EP1G^DFcXiO7L
z>P$S0EsQ)f-o#}C>5G0vxvX>>?%QisAkksH4~&iYbVvsj1yEFL8u{eswa@LB3-AJb
zN@IZ5rz>J-n@OTLA=y)=vg~)pLBDMxiRpwsQKwE;FE1>pxFjM#l^0#z6|-vp?@1@o
zuf>_}`6rzr3?GFsBEo->udU`F`<WlaA#dd~v#TqN{)6~5(AmBRhkI;}vL#-u0F$Ch
zT_85`nGD|w+IX6-<0P4jXEbXVDwxiWWMSuvf2il*t$wrK4TM&5nyAxUF)5Q1%B(~k
zZYfJf+7$*A;`=%QO%tT*=>Mx2Qq#aC4Xz1Db22s>S%>ix|1f9*9-5HA;hPaGX$$wJ
zyE*Ix!F+iTIk^mWbp7XScF_vII;^sM`SLqWq<rwQEGgl;L?Cq4`pnuS0!_Z+fdmVk
z*D7hQx4i*A|F#uJhj_m?HI!<tCn?L|%N?!l-O}CYc;;Y#2ft*ocY66;z^X`oC#p<S
z7g@|p^7XPD(U1L0nxTi8WMH?8-7==lv_mcP7;datC{p%+UIH~t2Yvdu78m1!+EvZ;
zU7k<=lX^Bt2h0(aAWyo{U3E0{&PuANO}sLo)kEo$VS}yIY5F}}d%J|ymGBuONT@K6
z<CYO^4a^0eaNL^&^Gw`@0RfJkOjqXO+<UUqN%oEVl}MNQ_(?GkXP+PhQGQI`OSC<E
z+S9Xy0BR>FLa^#qSGU<ks)-eZ!ii)jt?FoGBGto$gTK2$-c>*gOMG9CpXGW3L6z)k
z+%2Ih4UXD~G;7Cmp}Ul(mq=FI-xnXp^x@zd6QEmuOJXM#9YEJ%{H(1OKzz8gnn7)F
zG(k*yBhA~~`A`dQY_Z>^e~CkoBtCd({_EiBC8W3{$eH}p0VaR_i@{UvXsC@MNOgp|
zZ|J<DTr%pHnLeoUf|PCY7!!3UflR&2@TD`9QvXY~AQbY<-#ie&X|^-0%H@*+QUgK2
zj3_-06!SbIt&6k-Byr5x9pfw24cPqIRKIul^8*h&2qO+KBsMuq78mh=7~|w+i}AW*
z67d3tEwzITI^~t%*r780n$trn?lz5j&Z?OTS@E%V)A%sy?*@&3D6sMZW4B#jO?oal
zPFMd4=Y}^fG+q>Ob8i^P-|&4D&FMeczUtMAtt_cW>R$dyyB(x6=+#q;p3LojPl<pu
zSa3IK)8y?X!RjaEG$zq`H%B;o*cm=6Ziex;3C79E8I`2Be6t3QTK(CVj#X3YKiKe1
z>Eh+AbA1}^8VCHW<LNPRyG|+y<x;FEqb@>`6xr{M`&is>y)0#A>CX<>1%<ApEF7?*
zDJ(azhAed00->_vVOfY(%!#6gD^+D$WKg!=1MumLlPY`t&LV|F4f;&;?zcJ+1oYL#
z<h!m-4=x*BMX>o^5XZgV^dJIenfvk{qPm%X{KfmC8i2$0Mg6hSL;fV>Y_wW{xdj9t
zdy`LM(Pu8#-}dzXxeescNLn2HHU(@I<Mr(g%7iKM+Wa21YL6+(WgEqW16ZQ_ru>xb
z<XbN$e1P&_YSs(?5}8xpKo{<DwN7wygxBJl(Fv@(kx&>2;(igGkhiyn3whg$KKNDj
z+eNl>Yl)j0-@V!}2^NR3elKCgZVD@1^3x_!E5DVkou7>Yj?dXEQg2h{b<e%D(4;D8
z_hm6A;j<N~lyRaCs)rD+7tKmVooww-%~4UMqL3%$IFO2?rEI%b3we%Do4Q}}g@lGy
zNyH`7a0@@Y80g2E#h&PYD;X0?b6*#wxgoMyYyDHIt^HmQ2zK$i8OlK1x4d5&TwC@M
z;aKuRjkc(99xs5y+$9vxtquK)a2fEq%T?0no1r$5$_Fl9{+`PBts66&B`XA>^Pp0_
znq6zI;o^-pGq+YB7FS!dw~G_62!#WT#ebMw^!3un#o=FPg2vHR5{G{hJVFWvhjT>4
zl}<7vZkk_PMDKr{k>9$0?xWHPEOznsO^TO<O0_!(Ao*rdwJmtQe;j3VnPKR`E@WZ=
z!u0vt9ugql@}nxl?5k{{AyIr{U=mTUo^lnr@+K`WMv-#@L{yUd;9^t_mv3qi2?%mf
zXZjk~&dL`WQ&Xe(eqQ8P6C1QV0;o`p-&Pk~9-UyuNVECoZ6S((110YAyGQ&iC;lsO
z(-VnUDEWmvwq50QFbQOS8p7Gf;3#1EMQ8NwBQuA(8amInjRh~GT)Chw!?#7Xdv+cO
z27i}eumN>oR9*%y<A@1=#R1z1i#80%;*56l;=1O=d1O%t)TCKpW5rqM+olNH{rfP=
zar3iPM_g#m(FOY_0PG|p-k$2Dov>eIYXGIYHR${$GvPo!WjyOkqC54#(^)hm<}k|z
zG8kYX<@#_NpeA(}C&B}}0%(Juk1v{?$Ibc5|DqNuy|U2xO3k~scR>cUy3lrlX85Oe
z@=m{7fa|h5iIy7l)W9%%eYacqkr*^Bt^YZ20@Kb<0257W0p382ENN>Kn8ZaWOxa+}
zU&2b~;O)z!A=M4C?!gaykogkzH};2Ud4__)xf!Ji&1uIwfCwm-xlk40&MwCIfhGN8
zfO+c2m|+OPcXt*BOIMDp2vBWckjv7=%;nXU7{~eNJjj#R3P9+=>IAHiGeP{&75G#G
zG(`upayxm)-z}z@dPm)JxWFO&^i@M~Z$K}8rb?b2!feM=N>z|fhX*;d!addsw?HT^
zjiwAoPYP6CV4`U*^b`GktM_B*?t=}=cdIc+xdG-(C%eSM=+MlhAF)h-s?Dwoomm1w
zrCZN>c*4y6POwUcGR%^+k$FJ{JaT6fAvCDw?nn(_XrXOi-}XEA<E^KV6vYI^&Hg*;
z4CAJy$TAZvlbR}2=+ZO53^?Ln=3W|Al#@f$y5`38+k$?u(q9uA{{<^A^|$Z5V-%Q&
z&Y`yzG?3ZbOc?a<kFFUV`Avy7DU!xQcD?W4(9ZM0mD6Z@=Ob>gzP9>{VbCCl*nn{L
zHiM-nQ=oi<qp}fEe#d>}-bJwcNt)oJgFSm+sHFal?H~VR)Ft)qEj$#G+l~*QE(|gf
zs7#EpyO$Y_&gg#CZN-cIDEFD+pLyahH<ls*fC~JDvFkG9xVZ*(eqTrmZi2#);+x!a
z@&H$@ptu5nT@a)RIw4gp1=y&rG2HlXl%r*X)X^%}EF`k#tmlU^?t7_4B7;B_Iv$)}
zCJS8#=A;#%(84{;v?675FY`Z|EZ7<nB?Yoe4X%xHyR9SrBWXngWiTT6@B7JuL8{B`
zBemE*yf4-D{@+P-Sq|d?&H?gEnrD^57jgHS%R`~ojpx9{qsg<1WPknHv#42bGcKpd
zdU={V&*R*X=yBC{^%+1!Z(kq@Si?U2rIdO*36zTlzn`V-)`w<=SajGRn)zgPma&A|
zM#!z9UhecdgU^HW%qd`Xp<DRTSRBins~^huKV9d3WD1vke2KR<Ppx_UYZux~F@4$c
zg>h^@DuAU7<PjNA@ZWee!m>ugO~%)+-cdisW-tTqQ9Gs1&i~}(kgd&am2q(st%5^i
zFGBVcRxb86iGjc=yRFYpgB%gzleB7AY4mE>zJmeJ&laez!J>l9wTIcGn*|CCQHmmO
zTq_DY#=0V;3h1=D-pD2bJz%OGF=OMqJYL$=8#v7`46DyaHdjoa>^bmF)%!;%XG=7C
zrb^U*DPei-l8+&gobU((w$gh09X;t_RG?VM?I(O5LYqq{wR?x_=Zo1x=^Dvp&KXq2
z)|-ne{-~)5-dorUS`QO9<LDIzluQ|%-DpBx*W%CJ&lgUso!wn#U3cb+kbaxH4OQXN
zt9G~hS&WGLZjOo+zNuG>3mNq;XMvNZjqiiEj}3(QKYd4e)G+9W-pn9%=;^#E3hPw!
zNqV7I6lTZ;EIPO_vAvC5IFO^V>~Eui!n|9nf4_-&st)`Vv%h)2fhMNajFCA`8A(i>
zRR9*|V|-<4?1^LG2M|n129+?}#_*QQ30Z~?zvHr<*5lI7!1yg~zTImOWkU}-kh590
z0UU?nUVTs37g_U23P(7-vx(OW+uR53V-ZnT1CK1QYNkOxG*4%uL7z~NCpO-i6IP*(
z#1Zcrr}|o^&A#QIu{%$&v)wIU13P5C`~Jplqucm%OXtafx={TFF#uv9W-_?BDB<|t
zX?sPos3?XPFZ~C%TTgWLIg!Y`v4wRf5frxl5s^47$3&h17g=-^@9Xn+viSDcRJaN_
ze{T3Bf^?FJy4U_1Y&P$sP_wN3dNbu%zg>L|Gm9iNL1*W1IEw=#*i+M=kHiS>49xm2
zH~?>GM_NTU7dt0aPxnYe8n?GnUeE>c)NxL_a|iXCn>D_<d%ij6m4u0|SooseF-6IS
z<ZiLias~TR96*Q!*b8zv2<v@t>A64KSfaaqFFgtQ@!H>0+#e6uBADunA7|dcb*}Pz
zH&^zUYL``-r^n1$?~8h4hgOP&qn`R$EtuW7ul~EsZ5fGroFiZ{0N7<MF+pw7BxF5{
zg8gk221WlHX=QBCJ-^EF6N4uROPc^cbWNWUD_)FGOw3JdIqX4?M%gsuE7zoq@h7+w
zyfNwRgg%v*H1AOv5yOAa;|OE;RFh!?Jq>8?HB-&_KJ$W#JhSvG8+ChwjzuTPqjist
zIjDSVHxg}gh$Qa^`>QfG2v>BywBs0J$CE>pVFm`+Hjk9ZZe9rLl8kY+yf_v$yOn5_
zsA2Vo`B%!&f3QP44IU(R?T{ITC1|#MCrG^dOT{+th4nzu5tkHLZ82TFs9Ew?6$qBP
zrQ5uCM<B{6Vds%e41%SVUI(#(Tk$4|KL&ji-m9OT5`{I(u_rtEtt+JMU0vN)zh7CH
zBEVN@K!!}}E93h)Uf>ofP%1?8JX*`KuhItidl)|nD6PFMNKAO*LiO+j5RG9j$H@;a
zh<@+QBGuWQULhV$SF#bc_zVVGeR8F=V%Paymhf7IWBLmf@P)~qY!~QnH}41q3Ge#6
z!=51OTs?eA>~YRW#1H(+P7Yn&Q2^hg)7wjawioZS5(A4qs&O>7<J&Rt%%HI0(EWZO
z@(6dH9_#!z^U5S>Q>cy5O;uoC)mdiP1WRhtVvgsrK3}}=0ApQ~Zd(vyi(>Q%vA{oo
zI&l*ZA*#P{rc3^)<$2RCGsq8jus}c+H-2J7r)!SRt3$VMc!&S<R`~YacG}k4M1tqf
zT9;OA4u(_O5E!eQ#lu$cx||*;UaKW2^iC}ya0jVWx*vp@(}@x?g-5mP7{M{Zlt?hX
zva!;|IiV3NIxUHwF@em+aAzRcFw_2=d;lwr<M4e{LZDp_5dL_C;heqgqo!T`;GQtV
z53L>T^yMP-mY>|J;VfXki1wwTt&4#vmGDF=Fg}t-H&+qNP}kOi)RJyjT#!BhTJ-9_
zj8Cf`!gDRw&T*Tyqscr~?~$iV=d1y-ij?c4qhYyANV-O}+33--*Q}rbsOtKmjq*4y
z{=)z7ffE$=iMgt4+ar@%?4$IWJ5GGZ((@hf3F0pz-}=AB9CpDH6iB5n#dq;C<#PM6
z-7mO>j#;9a#JbH=x7Rq-HJ2wuK#KPHqP|TAWf%SO>{&j1VlECxO>|itgsslm#k`q6
z=!z!gaLfto1fa-lPrX>CVuj1)=XtK_-cBd>cd&IDbl$K6`7>eDv@Wa5U+h2-znvE&
zB93mMcYBv-x@-#zx3|~*-E}#t6dE@e+a0k=TU$@W9r_GC^exqWRKe$>pLt}QWmv~e
z179<|8@K3b`_<&JeMa1A*06^0sImtxAExFo{45VVoN6J;&o_$%VLR&IM<D1d?h0d8
zlSQ;(1HCF%cFxZ$Jo?8NVp18#?rzPn;2cP09<d(3=y-R89gHk^93+l!o3H5|ecmSA
zXsiHB*K%>Hc4xk@N?)LIDZhZ|j<}~`io}oa9afBpWN+h^AJAX)8OTXdFM)|lw*QhH
z2M;&7P)CrCpZ%#_TyubaBpMtSVURu&pm{M9lD}_nkD6n4By1V9r+C(wpY<m0IFaB#
z;FR>6F)cS{+E^HkVZKSDr(>6mwfG@Wkq0e~55Pew-MB$_%Nh7&K&~n^yt;u|fjKJQ
zj6i{#7pb?%Sp*hZ>34T5qAJm()gz&ONpC9D5$&Y1KQdyW_>mx}8D_juwsX2+l-qIA
zpc&WGso02|VixvH3y4laO(C?<YI>@8Bw6`@%np?p4@`?hTR2`^c_n`NCLpt1w^Y7B
z-3njZ9qcVjO@|3!q>pJXM@woR^M6|!FEXyPyFtL!!=JFJRIeZN<vV@(?Vn#J+oVh2
zyc^}?z-DhGP@V`VY)TLB-PX$4vyiwi>VJ=S`jMv(#E4t<Uvs*NuL2}>ARhdX*Ngzq
zB_*ZFCxT%dU4gt+Am7`=kb6%+mhckU)qJ*gQCOu6)FF5CUy->e+ArxZOe-HaV4ptP
zt?hr=7TP|83sv8XK>Tq-w37>TOQAY<9+6~|a8QJ~ov(?kt>euZi3`(eR)vwTDltHQ
z+Rttst~$_lve7{=_cY-3_s%<b@RXE$NNx#yFIpX<Re$3pCb!0vaJ)Dgki;NAH|bU7
z9~m(;WbK%*6hkmKtfjq6dKgp!8-Le4&4YOsr+C=gmsvov5DGja9XAa|a($n?^6$5;
zbx7WO9k$f<+*wa>o=0WN%RvXdU35kfWOdDLEXt1$s|teCO0KpE;hWKJuK(9_75i$P
zT1^(DG^+1<C%>ruwQTnRn&3tL^~S*6*T=w{L`dNknhHo_%Gyx|Z?26{ngb222UJai
zs|k67Y|`T3)7_n=79x;6jpN~9*!ZRTJd>3-Ei~}zOD;1dzLl$0maHeJIRzU>n+6ZF
zN~)AIYpvTFqc|Z<a}igS@Gwwa#%^v%4)ShwO14u*4KG6z`;w4d-A|_Q#r%j;cqbn*
z#QmFP63GU&?;o2cA?-2@e%35$fZ)HkSujjj?jtb;s({#F+Y#o#$M#vbt48emuLpd&
z6qQG(H9m+UuC2B+HRb_kk1Os#@(d(mH6BnTp6dL$FGndp8t+5guI}tgF;(Mo-n9kw
zQ5RRiLVN$^(HyY-gq~@C7w#wCu5@~Ah{)z@douPdQLlyAQ>T9=A*Xt!te&3m^%P&<
z30sX{FR8N;pGatH=o*aewr6HB%q_R+H^}*zhSXm0NG2lO9yXS25a+L6MlpdR`zWeh
z`@W34e?d+p!tLurmfZ#?)$w#!cg{^`C$3rKh8hJ(JTBVuBuS7JQ{^vrQo0*~n2L0@
zaNF$Nf4pD}KrE!m=Mb}M>AV}*9W;0qjhrvdlNBxl7o!fnBS1=LeQd@~2&iR{yivmW
zxANi_r@QvLi}>e_ME#a4fT<8H6Bus!uJpc7C3f$s!d=dO(*u}$F+el+znYXg2FvjO
z>}ht{0UZl_8$^%h-)VppDZS#DtxkO%Ipb^R<h9_pz0-09_$#d-u@CU`=&A1uqgk29
zXt;N`g!967)VxZvIUqlX=LmN^Rp=bQf}()_3UYb&dFQ5^Hx=(l<3T#>=s*@h1k)@W
zgf~ZG(~ms&f59bmIxl>>-uNg_NdE~Ac|wA@Ci|`u@%N?W`4N@Y6x|g+07AV#BLpW~
zp6UI2u3!*a@pl`;h&2G><&j6hhzp-+)m#eshc!!e3FLk)n?*8ow2n8d>BEn~JW4V2
zF4qL?qL3sH`^5&<w}O?CteYy`AaY(}PNbI^2BYSZtGxynS*}L=zZN%d5`8W4x*aB~
zmpXUuvqf<0c;PO^&mZLhw=pK%mp%x2v%dtDR3>BsqPgt+!CwA5L7kgUC$_<hW<7lh
zhcW}Dvuj8OnU;jG6#t<G|6ec_*2=}2!Pgr!sB8Hr9H=!Gu2xuq#^?swIa1V@OSzG>
z+;x$7B}{!NK@!*>85JJpow6j!FH9rM%;}pdDZF_)J4sixGo#{aeQUneG6bnATY)~D
z6~eG`Z4jb$B2HwUToI@?!!2_#oD01+eSKN6dXOlQw^tlGt59n8LzdHcW#mfMefLC1
zuAgzE>kl&GNgrI-8FGi;-|E==jEHW~=(Ll2LPSBhP@BgGKT}V8yzqg*a$U!~Px)B(
zHkhQg{42S{Q`z)HjV4<|M7t_6JwwP=<D`RuW!Da2Bg27IOwhYG_K$HF7sj=>!x1L-
zoE<JNun|9ew4+gBzGt0dSVYZ8%gi|qRFkhIC$IRLrOsjZh`xay8UA@2#+)YnE{Cjp
zq=!HOZdtIs0+=d~2%Dv!@BPgkji?9*;twKb50#%o@tTI0_>WSI`K$Ph_3b%PgN-E^
zC|UN)hoMb%p#q?pAf(t+83|hQUyl*hxQ)LHk*EFUJpEcjwf9bLMw}1PuxEIP;8D+>
z1O6*ipk@TI*%%_G-<y9I-ed<OyU|?_4OyQAiDLpY0|cApOe}Z5eyQ=Grux}RDsbyc
z4+5NMKbw!59wX9tX$0@KA99z8e++oN%82?SNJsi0y3-jz{;Dg5L~f9QPM87S5<FLs
zeFu*fv^GV*s}Khvho0eOWzmJmuw?fLBdg<qdLgf^kZ?rvYOJ%P+jYFedvEf`KxF7=
zr?t;<PeWo`vB?d)BLa;+1a~>5YeQj$sLO6?rBr1`3{UZ9*(z3#6QnYd)t>1r)gKG8
zE^<K%{gN`}>D&@sxDiaQeQzbBtwwfU`v*+d2?uXF)OPA{WF<5tvL)qdu%*i6(&`!J
zo*-UbPyuVed_VCo@W;YICmQIJQ0cmo0yz_m=(=?|)uiS)K*VrKC?Wt_Ss%Hd2Sjyj
zw%$J6vMjlG?Y9vvYXAbAM4u(g{r$E~-2Cv*1&V<II5`EdF|YvKq28OdajeoGwkWiL
zf6dsg`A39f#EaT8j-W6V_3CHzrW;9<kWr{HR2K#W86Y$=2Y8o$nhJiTzZ=U=!Vm3k
zzZ}?1e90iqk$I$i7j=7+uVNn)58l)(?gHIW-8ASG@&x?RzITc;rH-Pg%5?c|^fmuT
zq0PpKy=f;N==D+l{o=yjj2JjZ{hll4Kn=6d>bmm;;>qgruw<>}BI09<3e{Q|*q6Bl
z?qTI9=$Tue0`%$kKaiHvCrTG57RZPr6Q59USF^g05U>!PxH5ets?k=BYmRz?_06-9
zBT}Kys(*>bOck&0M*$ZGycLW}ruKtD`tYbV4`^U&=(}0C%?z6WSTh`7P>MC9TOl)!
zYU5z@A0lE@iQFWZ)G~Tv^6uHV!I<O}i?pmnAV7o~)UF@2nNj;yk3Q<}R0|_WcLU~!
zQTC^Y_He9UKD%g5MY=z{tQInd=wR5418{Ca6kw0;o>PK4yKa93><f9IgCM+2m+~Y0
ze*3Y)G?m7el$lQYUQG<F%PsMFgPg~`He0Mb!l;{c97~^M`4cH<rF@t<q%BIv0Z6cQ
z`Ti*6443ZuOV>|$Q@`_TmF^y!O}956yOlE$d7DWv47k|PL7)UlF^r%J70=%mwCnUS
ze(2fD=WIK;vYgKh?cR3(4hWv=lSR~Ekr~dPD1#z1BR#+^1Ta}F1UD9oy_~pBXUm;C
z)PvK@Uh4@wld}}?2n0mFNKaT(c`5tf{GxdaOOSZRsU;glz_UUt8dslC{AFZNftg7p
zlgHvdomW2|wbs{dx)F9@q}#RoB+J~At*d9{diz#w=A@*?Y17Z7^8)G7deVP2w|Z?L
zO75w9;&l}*m0>loE(@YBrTK($+{f7FjDlk1xb#pQUMXVkU;np+o`e;7BGeM72K5_w
z`eRQ{`d4m+KClDm`3mho{c@${htjjrsi&i(n<u&2CVw2A#aqh89)5Cf<{7PHL$`@}
zo=Ay@h?@YY+*^hc7R<fpbC(MQW{hv)9{Qp2DKE&=pcg8^0k8$K)PP@6N<4(wyc?%s
z7Q+HY*V?(2WQiXPWK}iUpW<!G9lH@T5!mAg!cAb{202BdJ0A@X!9cSjO&Z*zE(?r5
z{->>L$qvDjpLFiv*d>WutTx?2NDcWw?6}u+g*pSlTma}!>sS2@B;3L1RZIR!^*_Jh
zj(_bQiX6lVvePiT8^cKwQ!Gfee5&|s_kfjZ??G#=){CtRj}MtGnlYePo1+i0g6y&5
zP~prKHc=PPDzs%gK*C$VFo0o?nrXB4z>St;Sj3sBS>IVLw#YC_D<4IXLuyEy+jfMI
zY`WF?PrSgGj<52sp*gmzse$3!SH-!xk5kq#NPW%Hl1&Q4yg(yC@7&H}AWlT@#t{o2
z5yT~M-|ee&7vy=OIrp)95)ZVL_n6nF+7>yqSUzam%k@0(JOn;b%a_lX(9UEL(lS!y
z_O2+2f$<g<w}U0y&{w)!P75v6(qI0j_seg051K;k4?-g;-qpvHQ3EV$joLpnh9^s(
zNiD8}5GNP^rh#o}(}TXmT57ybNp19%9IZSe_O{qWdb>4?-L#4~&@e#*Aq<KL<hwdp
zSzI(S2UO1Nw({i<N6@+Wq1Lqrh_Uz^oW((p*aYptWb@VN$afKTi2a9T=@+|i>ThA9
z4oHry?WP`FC<1l8@TVt!&RgL?!Y1Le5O>t&h>-kvF1<328E=B}vtsQv*u;+HlI}d4
z+a-w4$NCPRCR_D4qWjtE69EC?mckd=XR3((d)6@SYDpRGGtUd3a)zyOy!c~odIe!_
zw#%FS{97^!#u8k{nr`80;>)ub4u5+LB)t*h;O1w5`0ZT$XbF1w_<S$`Bq<%@E?S%G
zX|f4=P8InVKxR*F4Jl~z<6&&}hoFb?xmdH`&q6=5t64a`d@W}~^w4Jiy+AuZ9MeZ&
ztKX&789TcCYNf8(QV(+tj+_M5FI%GO)c2u7TR7`5^YAJ2V|T+bx;%jK=)XZS2attg
zJW4`V0~*d`ZDB1j&_Vv4p`ZA9W%*jFp)hD}?(K>r*KF5zV9Q>{4i+W;HRV<@pjTtN
zZL<`zk6fnv(Fe{7E2evDK0>p;oIE7^3%o^5x5qXI303H07pwd^yJ<stI2#rf*v3Ii
z&uisD3I8)(yfRwEXDrXsl<~9WY^F|K5z`2XLF8U8y@l{-r$0t4)U|4iupHC+In5e@
z2Qk%O7#~|HL2ih)bIIxnZQu`I6YiPeF>GK9dfZRC-gf}sf9$rb^=Dht?`e!LcE|Sv
z<4xu-@S&R3tPtT*yZ-)Z#GCH{tCqJyf!n?h=l?lmvINOymQ_x=w|B&W4kC||NY7B{
zcQJ7s6a7b+_&5vR{<}v9IftiddyOvIguwFwW7p)jr&l;32qL#jqE}2HMwJ4@G^0=F
zO)iGbN(Xm2hK5VJ`8UAM3Eu3!qnuZ{1_gt)ga#FKg&+;0D8NxJ5ZAm_OI5Q@LKNz}
z5lq~ly{*`;24eI0-&_8bH_!XyR6p+Q3o}gr(LuxoROGyinapO`WXa^?(Lo<mi9-}O
zSmQZ5uKeW2PM){I?VOZ(cmGG!dk1pazW@I(BP)dLJ+o&NviA-d5i(08v#e~gS0X7p
zS=nS{Q$%LALPqx9S>NO8et&+y|L$A7#&uoic^>2Wd>r>PEB5RSP+Uo=`oG!|ny<HM
zh5muEXejxxpzHH$%Z5#xd*?m42TMZ*f7~z5>t8dMkKZB9{ZxHNA5K*<xl*IQ%1#-b
zeIve0x^DDk%*wCq9y41%h-yqF^IU(SF%>-C9g~;nFM9K7L2cW~DRbo$lS|+Mdvxc@
zB%SfRk)xsM2FRA_KYuVs+glw(fgDEnnD$`_2T2}<u!~sr7N$RV84FB2-*bT1p<`-T
znq_7D0p^rxQ1fT|t?en}`ua*zhZMcx6knEGYAHdjHU`>;ojzvbAJiD@FnQ6gQnTOM
z9QEUte#LrLCoFUFXSnpRT@r~b{_Jzx)qlvlrlM(x`6apYGJO>fb`aeGZ(Ofnp!Xwe
z==>SwaAAbl?8m4d`7Gd~n)k|5cZ&e~a@$90Vc!3-NvAN|3SJ8_ZtQ^6@RbYW(N^vs
zLP&B~zj1=t=k6Ze^NU^&_zB*72*RCK2W}yL!;y`;K#PVH%kg(D(?xzKytL~)&g(O@
zkW7mjY}N!*e0XwXJ@0RUdmE4Jm<E=+(dvYUeb(6h(M-jspyQCW(V3y3e3Q^$_e3dA
zt;1JX_dU<bulz=XDs^?&1pbv(JDTxCUA$E^o|sv*-~$<YnTB8syz6hdd0ZBrM&1b~
z#pJ$xM}uY(w}{4YKy4eFk-b2N>`Dfh(}5PzqSD;9y#6E7VKYmQtBwEi6?5pofR~}|
zM7XU0oKfCUcbUH&iAAu@a<y{2B4<xbE;421JmTEod!Tq<QPBb)9Crh|vD!svJQ{y}
z_`&)mveRQKCO|X!`ojPJ<f$w41@6hFGt`xyy+1uoP9+_2dN+<PX&o4Vb=JJX%Kov_
z<4E)Dz|kVDnK!@FJUw~bH=@}1Oc8}Gew+Ekm=>5QK6I%!p-R^?m;-7*bD+fx29}_g
zn!614>XqD00ViN*%={wxAcjiL4E|+c38YY9b>SF1b?Kn0{PkYQ`T=L~ec3~l8YN#b
zced|E@(4v$VG?hWcXW(8B2slZ@Y19oMq4G1`hK(AfbXH;lRt#uY(Szf*5COCW51B4
zZ&f@8F@3M`s1oJl(*2pIYtlYfAbV*2p8lDlCe6`o3%;2^jmTF5yMHeIzZZmvL+s4Y
zEVg>|f=`cbo_>9Gwx<d&`Me23bZ3*L{<?DuH2;B*p`$~2*{T-vnv0;uU0kqh8quOb
zjTlq`L_DSPT%)3+V|^J<=eX_)TcJ`$3NuE|SXpzdHp4%o{6VtAMfbK>r0bi7(bwS^
zaZIw#;#aMl+Z<+U7|<^guoV)G(7QI0^uQWF@!GUTuUSf9ZyQE5E!ckop%nKadRdR>
zu=Zle=k(JNBedyUC3Jpt{9$l@w|;!f?EsAuB~OxGFi*!tSYV@=1#FYFH;G@GMyiNj
z<MRAE<v$mDe%1BbLHql7Iz2VnSEDEE;J{oN-ja!qaYgE$ZHx4?p)Ez1pH4$51F0;2
zh*=K?*<>?RjflThE3r=3$#i_KoyhbXeR1S$J1M6ee9^EjF(?#V3Q>aZgHp8)7uQRX
zqz@ArePq&pIL4hb1I|bEqa|)lu-xD4i8O<kjjtzZ6Z)GF)|e1N+cSn@-<x%t^yDQ-
zcPKo^D?kQ+D&c;!c*-Z0g_iK++c0th?geiXYYh^lQ9H-G_I3|lb4bI*be@@(huRu5
z`^f_i#6!qdxT2zh{(w&`ir{m}Y{~r-3+w4WArn(GE7n28K5wns(h6(5I&%Bb1>U}>
zl5pN{ro{~H_zm?U9K2{n=yL1+AiKDJW47&U7yb={2TspQ<<{*sprmuBh>O8<8>%Q!
zqsBumrR*r>EZv{!9#XE|oglunw5|V6)b~WfS{-`VnP_NB6qErwyna2$o@T1=jwPPN
zE1Wos(cU$lZ+LV|OhkSmMvFCRo~N%s&d!+Jn#<!lRJ4a{A6+6u|8|g~lB2jN7_P4I
zrNEERvO8e4vy(KS%|Za8Se`c`OP|Szv`l>?(ihj|<aEWMP$7S{VSwi>`ykTU#&YG_
za=KGnxlpl@0U`$a+VeAJc#_}p+s6Q|QKeo@Ivbnl<{8T9>CdoXe*atRafe1VM=0ve
z_7xCwbwykq^6HQ<z`}~4HAq$BL4?^+(MQJ{^1>p-%_<p7cR&$_qxjO&=ZXI0VKtOu
z0`)$edzRZjh}H0T-+1(+6F=ggW*JK}Nf(8u`10M3LNHz$kuy24Y%cx1r@S!T;K}d;
zd4Gc~04Ow1Px6`nXtWo@Ehv2D{^ejnI>YEJ;+Xzz8dE!z$gE;rq?4g8l0C<e6{u3(
zh|OXUQOwzh=%-&770ksbUaCJK3Xqa6R{Y^bR{Hp0X46Y%m)ZNaMGCoM7sU&*Nxqr<
zKK8A<RFIdWD-(#da42R3|0WL?7;&dWErd8*|LYj%8JT1fI>w))RSC7ZfZb0UT8m<D
z>jV1I<N(Q2kIwH+9*JXRW4DldJziaJDzq(#7X5gUjUP$_>iwQ!vpNRv!idJQ4rA?|
zd=K$KNX9W#5vt5)xj1ldtg??gH&6`<^N0Z#iQUiLne*<Sgj)8uM|)hT#$yyaD-mI_
zN$GP|MSrs3qp*+}bgk6f?eJVp0SZU}v=}m=?}f;*$j?U@;5eo@damE~+43Nme!41$
z4=KSyn*ieP_*3fQm!wsF`-bzL)=p8-Vqgri3g(Gv{meeU#l+vjMP|}z!ODMHE{y$H
zD}_fhJ@m3z)L83J7WteRP&Ii8QR&_9AS=HLE4LNv%^rC~_`vp+vcX!GDW5Qz^<Gy&
zXKQTy)nE9-+zaVyNYKu8gNk@S6(>@yIr3&p&2IY@5U4HSV?xk}h(e?2#tLL~ej%Mr
z#Ne2ujeoTkvyCJ@!&fP6YRvfex97nzoW7#=cW6p9)qc<R7}mITV?&bK0TxW1Y?n8i
zX}P*$6g+PwRTkH`=ZBT?-_@Xj$4)<ZU@30J(Dr14DTw-Mc+c5o?s86IVX(zBMi5Y_
z;`v6`#+V-T5GA609P^^*{*)F91ycywW@c@Jpp_EFhPNr>Vs~xc#ZzlIzCI6%b+{~(
za~&-Xc%&vT0ndW4UqEJvh632h5z~egEndy5B74~41qBao0B^{tx9)}%<*ixh--&v!
zQ$bxhOgeX85!|$Ue?RZ~FvQSX88Wfl!My3r>>Kg9aAc4_ldk_C1uK|P*Z5uVvCiu)
zAlz&i>B;0ZldE!4A#1?Csr32v)3+%&+8~pIo13MRx&x8h2eo5NM-W#sJXn_jGa4nz
z0r&=vgO?MtWuC_GWQm@V&t3AfdI+xM@K_~6QQzOVT4p<6XXrJz7WW$SmVSS9@AH8s
z&)$WnB+{H>)a;layGLy_0X#2g@^esPndi$?XNo*9h#_AN$V(&(5X(7{A;h933~Gb|
zrhH@y=$J#mrg$X5CrV5BZjBc)R*xjDu!(8)5m}yh&B4QcnWwu|gYRota7SLz&x=PY
zdgMykE^fl0BAp4;*94J3$8Xmw&TFk!@aT~4vL7m4k|YlN96Q|-?D?4~zH5eAwFN#@
z=_}lPw3Tn2I-S?!SmHrXHK|QAVguE|yhN&)*B@-4UE)AHEu=X&D*EiVAhNB4O&KK<
z{SgOW-2>g;C`op*5e~P42Z}AD?PQ?qV+LdPS?<Q%R;@DSNFk?h`UFjyO{&|@IHH$0
zj`b1sGK;r+|55-;`pYl<{KrfP0T0(9?UF8Qa_<i=UKGD(_YCq(!cz^Th=ESe1|KQQ
zsR%1X|BGtY;FAN3jBP<qc&d5?aHEhK255YoTI#OJSw#BeZv5`A4W%3_B7UH`6qAzp
z%k7UR(&&U9xT1Ab)-%`c$}5z~d*2z8jtL%2+~!t-_ioU3!XIm`*;PfYh$n^&z}-Nc
zk(4){UZzh!V*o=sx!Mu0i96QoCk?vKEzJic&RD$I8v@^$C}3S+iBvFavv8wJeXwzA
zDZpyA2tv;7&ZgM@oTt2}JMt7D4HKEB?qd4}!aner2|SV1#qm`)B8Ln}w&x>`S1l0y
z;`(oj&Z7-oesRr%rTq(eyr4|v`DJIX6SqW{;hcv!$SqtG_o9K4)}PaQxYk~5CmA0_
z1ENggjw$^o-dMn$!n&WJTk}Alj3J8Fd3l7}q;4M%Avk%-a7WSiqXR;zayBi!fv|OY
zvUd*F4>X_z%m+jzan4?vHNV4bGHaE?Sdd?+M8IyDbCPIP7Cy~tNg=9b)rh}Lj(K(<
zZxkI-lh1{IQiLLjCl+o4Z2q7*`aRO-zEyYlCsK_JfhqAW5J)KMImnnY-IZS^KDC|q
zUj+#tl+^=UXG_g{RC!GNsOeG+m79tQ2C{uj!Kp;PdKTVJ@~-wT{w>1kuI(c<8R#$n
zWtUVoa%OH&0GOeQ?XjNL8yw??aoTmXAqHrk2yp+UJ{T8i!M0LHN_`+P4NQ#|<MOU+
zZ5cri|3(!SqpsxPHfy*a)3d-Vtae4NVy{_F--EHS{SGO1UybN{m+0-WkF@kD-ptfV
z(EApFq7VUy6CoTKfc|Brg8@Ob-Rsc&aGAM7ro=9%-GN*0>gJ<=7^X2ZExS&A?P|4!
z`dlPetF4y>a{C1EZ(qWd@o6<niMel8k8q``dAXSS$Ix(9mp!(47-QW_R9s9{DTBI<
zbn>{OvZBuw!*Vpjz%0}+xKAfVEj9xx`{x?!EVv&$_GU2<6aX~l13Kmqh5b<BMu1(S
zS~kcaATUEj{16jdYBWpIjw;Plfb#3yQd6u`@9Hr{R$o|RmaLM1@dqD&8qZVwlFeK-
zMtHt4HZeOf@<Oxd^IlU$c)iI4%Q6t?Vm;EtK#VJp7-_-V9h{?q!>-D;zRkr1#n>&<
zFBaOfcY^WOpl?Wp_O%yB*u`G~52;SjT$H%-q`akS15-LgXCV8<;y3iMD7zaG#Y3$7
zqdYG8<$1_h_lv$RZGFlRajML`R5C5xRNTZ&rXa*IZ5d!(A!T)qA{%PVRC1ADyo{XD
zA@1pCTi~_E$NmikKD@`ocGjVkX>4FTg$xI!jk)^%A&$OyQzFmhXM=9S-*M!!qs&Qs
z6b^csU@y(YwhH^Z@jRuMIC!MlW`ToSOApiHy<f~%azP2!ZH@R|K{@JC%4J%`o<BS>
zj1Tc1ephP-Z|6&Ob3Td|gA_Edt?{m2RTSClpFtMl`ZbLp3|ZS-N*_tV%OQlBS=*TY
z&L8ViD0otMQIANJ5fmm|)6^eMA#ZxZkH1>WBA~=C@uBtD$d9jm&mo}-08FBP2I5S_
z47mGtuyY=#6ZtbN7L-UlFcg6uDjX7l{s1i)=kd1)_y^3)TwpoIv)}&S_4}iygAn{E
zIL(T;r*-TIn_Bd0P5=|gIE3;2JsQR2FrQiD!#hf{2PJ4J!AmX0zreT?rrAK^B#>3W
z22V(W2twLSX35={O?L1R&gDfwPi-$Q5m{P?{$atI>0ol`WQ2F{`hHOy6|C_g03C(T
zi?&KaiYbSD$bZ`r23{gg3M0Y*{fDo_lwIQdSnjX<dl%L92p8NP$glnT&40gxtWi?3
z$C+FK$d7|p{&6YcZ|PWhxB{YNC~3QR{pW?@mk>fFnAXt~jr)*tbtfezV(4{q7Vm)i
zrfgpG`JKao&Bc9iJSNx93y?dAsw*)e=Ow&oW4UJ&#n<mg*t+>Viw`f4ylbx~>*y2>
zwm8Cf8EhuqQE>T($V8ZxcgyxA5J=ZU)c2m9qD8^w!-lv!cGnS_DTz3me4-36UQeqw
zg#Y-qb-(64{zP97Y6XWM8mW5lnR$V03U@CKlK3V*@v!~RC4a7m{{0S=1t1EQIj;9X
zgU(te@aWn+A3I$9BUC;u=C=@)B`OeN%`L9Ar(Jc|7vESqRtCZ=XhJy!Zp<by;Wh-L
z?H|8@;>zm7>mNAG1COpi3kaH2---PS>bQFE`vv$)7s)4SYkXzhvu(Ytb?e_{_fi9m
zf@DYW*Zl6=(zg?ak_}B8*Ms<2E|iR4#6<x{_9lsL0^Hm{#>yb@zPnME(T;93aQ+N`
zmG%<)7q-%FoN(lRtNC}80j+ui{%^pu9S`&d83Vgg;LMF_d(9#x7#*dtL{RAv28(I8
z6pjD^4=6+oc=cO)oIK|Vj3@E<O%r8+LBS++Ej&tdX~Q=2o8rCy`B!9LF(RICF>_)7
z!B7n3<sV5mHLFz8)nh{K+1l;jHOG{FMG9Sf+W|i&0D6#i@1qB14U}IngOwz18$LWN
z2M#^lU;!1xbyOG-NknW~-(Ai`ybUkQ1ZC|X>WpsgGZ})gf53xK?A&)EIYD4#Y#BfW
zs&Y`Uay~zmTecw*4|=eR(q%&yoh;FpPEcwtNw{3n{ZXN|C-t9OltjXFKi&)vS-;_y
z%4L`uHa<8QGaw{EWJkBS+)8D1)s*7jr@=G#?_ouo0qYkOCNUH&^T{<E{9g$x?mPla
z&E89#U)!{#URN3atxO=|wm+WQ41tFzNRR7B*GczxlI+DAebCUnGb7Ke(&jaX!eW}{
zJZXg(Y$)kSaD3<dC$?vAn(I1kJ0yNsnm@yU2eO<)jI2I*hUuU#^f5<S!2L^E{T~VA
zaHt~1alS3x9Fp`32=;|X7#?L1XonoVg>nd(xUzbGeb%_vF-A36_9&gAK}9ZdpC<d+
zB&*w%e^^#$3+n5#w%PuFEk+1a53<U5A|Ls!hbnM=;h(GPS7)o7`^<Wkl4l-`6-Ta+
zIGl}A)H%qEC$pR<a+;j3^<pIT1BLn$(<>a~*8qo&?}*YKc=0U&_Ash9%zI~#il%bz
z979hCR_)yz-b4-7fsAUHP(;GaOk~Ynap!ATQ+Y5Ut!R5$c=;v|oI|mt5zLJ6l=bUF
zk*^;f$A*%^&#<^zb*Vn_&|?sWPH^j>w}o6|XA~mrdEkcg@1^t+Y0a>|BHs2xREjO|
zqbQHN4T>~vHHiSs1VCqIHmDYX0Y_FwyK)vAYxXu`mVco+iH;jClZBb7&jNR#x7xA*
zQ*McVQ%WI2@gn4il-fS7&u@12>fW(4FJ#t#DZ?z~uL8)G7Fqzt2WDclUJb!dpt(w9
zOkfh+naqljRwG#Bwn~O@ph0BiLmIphuA-#mjSzR)BwulKTn+5^g07Q1*n#8YUOidL
z=0MFrH_MH#fdt5md(Qv1OzpD<7s`FJ5|>L<uTld!><}e?akk3Z>1)u)W&a`~`SNeQ
z>LoaJv6+jkit94lW>F&4$-gnVl8)1}$24EcJ-|Ln6~rMlUmM}>g!gOt4mBmveR11_
zLWun_b%qep8jIh!l1|C=?_Z%fl{Dp~kS$u{XW1%6n!JquijO>x^@euegKw0>^8(vy
z=*|=w5rB)Z=;$;_d0lEH7N<#j5WaE3p<48t{u>65C1dk1=bgSOzzpTyVe#$kja4C{
zW_DNSsr&)t3=2eg|GU8j9?J=#&QhKSJW|26+%btusKc}ENY0nLc7cF>s-!K&aXf8X
z3St(<@<b6(6kPh4&T)u(*~qR&|3)JQch?;;<9*>cPUfcxB7ttDaLIk#r}iGr6(3nZ
z1dSLrC~=P<G3q(~40Jd|uc>&AQGgYTAuwxu@FwM9Iu_u-_6v8(5e2NRq^%&L8-<{5
zd5&3_5#@R*p#KvS@*v22qi~C4!1FZEf6w`}HDNaGVqQBp<rYF&;sa!;@fY1oJ1%!q
z-Zq0^HRf4To?+WuMWwY7Qtwc&e(nTjd*HR`uHDd7`zEtU%<bg7sUe(yFF;Bt%jpN(
z4R=x}i&QL^f67)wNI(~VH;^geCZenR;_Co1XXvI9_)-pHK^xvy%UB2>0Y%x7sapZL
z18UXbeuTf8i><2T<2kuLEvjqvD1e2+`B_Ja?dV5ppU1&}U-G_wkWFTeL5deVm6?Z$
zZr7vmEiR2oA-1eA()3hp%orgR5dHPQFz96{(Aj=Xfz&L11rp<U*3e&^R7r!8Fml}p
z`gBYCZFVLNCn-{*at?Rr&Csu@%D^p;JXE8X$zp}%81ZfZo651wGy$}mE6G|Oe%@pc
zqBgOha&GDK0O^PbR^3G3EN%f;M)Lys?a`DsY)U20_D+DE{JVuyl_7})A0+<i4}%{+
zV<TF@#~DU^$ZuuUw~lPg%S<cG2(HxEmE%M0_=q_!o3*q;u%qr0i_>8{1CWUB@{E$I
zzo!&_#Bz{ccbfsyX0*FUeCdPDa8WB2P=l*%5wTFDtXyltDLIpmWK&4$UNGNiPxJ8j
z!w;5{0{u_6y4C1rCdPpv;Ueqaw-yrn+A|UJt7Ef2R3%U3H?}Y_rX0MM8nq%d<<1Q8
zm-ds}I=16}#8Q~tSIH>K7@9%AlKV*$vAIAuziRLr2z17po}t5aHRe&bha+FQcPboW
zgh6aoQc65@BusmKbO$p)+x}Wa01)16Z+YMXnkr(!3o0`;uBTV_*0c@T*LEjkO`Y9-
zkb)LJ^3<58d-in*cB#jl(O-e^K%eU=_~@BP+#ddNp8t=l6-anVd7*=nxu5YREJCxF
zUNhpAZtsm6pEdWPqK)D^&sM<Nru8`0;PF(Kv|&0=d;xdxhO8_S_&5B%5lVeshv)u;
zm2@RT9Vw(EwGt>rcS&|{ZiFkc@GI4A(CpX5$k%wygcj~=#5KTp{BHIvSV}Le89mM*
z8rE~2hDfLhdYl=*j<8A1Hi@b7S#%}7MVU$8kb*x#jIj&R6?dox5Ssb7?*U()H6&h`
zv;#h}k#ZiF`R0AB08#_gV5ky$9EX_Dcb6&7o`tZi86-X4%zX+c#upb!=qiFzVW`;H
zfPd|#)pJ9zMm3EUR|9@B>fC|my)Kncef#Lk=zoPxuR!6T0~CoO*EJ=LX>?V=i10^N
z;ug?4^0JbL*Lk>q+sU<x`;$Gr#4d{EgCp<Aq0$l-$a^a>HQ>RvVf_mKL5d5NsgC#4
zhvcg+mN_nbcZKXTrp5<hrN1SQ<{Po6|GZUmSF$?z3FAYNCA9Y(aCj@=TK42J5@o5h
zPyZzAXxt(v^;~SXDD@up0Cwcw;0N&d$zBu_MDC|hJgHZh-nNftBmJSQ$$`muDFB|Y
zJ<K;@N&dLeE#*AErXHmVx)D)JKWH^q4EpU)_})t9jb#NUvPeJ<8qLQN$?qT9e|{tF
z`UChwc$@@ueTx~1?7k~7{UPb<B3XTQpYy<;2G`^6miqmM>M~Fb;ja(B7ghE~H3K#o
zc6v<Ui|%F8h(b6IUzr9iSysJSA(I%^k&w-u@ZqF0XB8R%5D4Jh>q{LfR7jdiYao;h
zwde+<q8dXzTJ``V5!E^xy!ERYw-{wvMukk!31lB$v3J^e=^k#9Zb7L*|LdFinrZ&K
z%PAw=Z*n$XI|d=mUWKo$1EG$2@-K8zgXhw)%@$j6<w*JmPeArI;QEuxq;|KieIrwT
zcLmC0kF}pMjF%{en@;izknsNHSAZIu2uhyR-$g8%0(o*}lDaWeu8YfELSk01CjgzI
zSYgv@En@`h;fb#)^{bz#?<lat+g`U7L=UpcW1^%S=Wb4+;?MgB^2f||{n>}CfpByZ
zy4Qh->i=6|ESnd@b9#zY<6!~OlS$*Z*#?Ou1#Vfi<$#>+cWIUbWqvM}g^K)(cmF+X
zgDH^gXd^0cIJt=6<Wj5mHHU6IsEqC6HW6-+u^=%LWkUd(Z$XJOF3Tuv2Di<vQRH$S
z+0{#5FRWON3GeQCE|&yTI8BHYJ}BMH@-1>Pc4h#^9P-KOoABRbS>niJ{indZGEbs&
zAU(``+-^ad9`f#!>yiln^PmB{X3Jn@HoG3l;lke*rJHR(dQ|N<*u+Gx(uMMO6e(}!
zWiE*c8T}4|;;M%0-Gb{duYjX=<a>A5d2Zc8T1}Tju#k-#Wu_m%31gqhC)t@2)`$=P
zBsfTY<?cRMaH85rl+`h&4-7_i0;Sk^HL^2_*lyl-F>syr#=lfVf*ZxC+&`EZQJ5K0
zF1x>PxzSXRKx^Lk>3zaaNw8(#96Pjhc!X!g(rEDO&cBqUs1?W$`M;MqjhMfP>Lfez
zcVKj~A%g|7Fpt#<Wj+%5uT&}Wg<@tj?v#vPc4Md@SvJ0lJ@Tz(aArC--oNDOAuEZx
zpzZ7n5_NnnQeL?Q(M)1A6@-7G@a^^rit!GI?roF{nKv?Lqe&EcZGUgyEjgI#&-ywL
zgP9TUpPP8d7T+ac-F0nE((^U-%SHlnwJRCnWimx`MoDLDJaLeQL$*VP4vd%Z$4CaK
z)AZz|-cb4S?_Ce9i?kXOhqagE@6Hhq+E~pYwFrkKv$tGcF}8M!_q&;Is{h7#Bx}f&
z`6<(t-^8%5xXXyw8}VPRY0NN8m;UO()}|Wk$XBw;zJW}hZcFHAn)mc}50&mMyybq6
zdZpqLeN~t5zi}SPa-OJ)*O`@1^W!)~{fZ<*AaYu6Z}`ggp8E8JB)qDr2Lg7g$Scd=
z-4&ldEY;pX({slaz0Q4Tz@kNNr01R?LUG1Q=0Bm62C5`a;m2~Xk>r&}B8<Shc;&lf
z$F=uz*AncPg=_~{623jP8?1s&9@Shv9+cOzEI`%eYi5d;oRX%#`)1pgJXbyrTU%Nc
z)Bio`K4!+7@c1mNPFA$MLODyg65?#<sF03hd8f*j#_Fvlh15y+&+u9r=T?qdVuEsq
zDq7!As4?TWWCp0NZC`hBqgPPs4iu2l-FQ({juHm^!=45c)twHae}{T@uaZ5r3&8|T
zFG#K5<i#$YTlD;#?-CdD@;xp4<Bxc}s2|$1Usxe}P(WUw7{z)e;|at>a(!im2-Z5)
z=Y^r8JpRF%4}SQ-xc9X4JUJW+asgH^I8Z-cp~qd~y+DSo(0*`2@7Jrve3vGRu5Ac@
zS%qBUZs+wWnpEGI)kA{nwXuZP0pdOv%I7}8zS1S5;WFg2_Gq?6A5a?IT3mhqcHvi|
zJ91vAiXa&w4jct`KpH4cwy1Ay9b>2ZB>^WBPURb&haQLFO})}>0Ql*5KNYXKPDI4U
zjuI(FP9AvF2psAJ(EH(#LB+jGkxd$MK~iiGfIuAi@|Nk_-qvy+XtX-SZEnjuMQupq
zw9V`|Fu2iG?NmGr>$iJ(hpZ)VB<Vv4NgV3sw-h8cQ2NS>CSzIdg=PfpVHmQ*inlcQ
z0E(=zkuf6{qXPCbk76J84XU4?C7vGV`zj1_rT^}|D(yzou+vS-6m0M^L!<eLy-o4R
zwxBcKua43%CVS+1QS)8JKtcFyN5J*$MnnuFlm`ZH*>WNYyqI~W7B=DYNYG|MnEFV2
zbGNfL)ay_KY-7rRmLX18Av3E?9NPOxd)omdZB@O*bZ=ko-Y(~Wb>BYCb@dWQmzvGz
zL;aR1rQc|;RBS4vvO^G44N_ZR7FTslcV3LYhqWsu{i1!QEC$<u9yj~Q;NBJE$E7FY
zRjHlHeLi6zfb>Ea&|*R@(Ec||hd$w>Ee8eW@Uln6Yk#!<HPh~z)pbWAvqV3_X&H^@
z5wi6^L~ik|=H{9nt00Nr`V1RRtlE35k2cn~nH`FHhKn(GYEy;&O-*6zoxEzvQ5cZN
zg5vGbWSrqHz#Np4SqG|*;gZmfW6+pk;v2TF)A#t7&RoB?=Po@Se4_7h!HLI<%`@1|
zd$~^I6L$abH{*3z1@1#sVB!x%vnQt>rfY~ouRTD!7LTn>1AQy8zh~BWmqn0G2WAL(
zokew5pV|v-3X7T$SMEV~Tg2Aks(%b%0j`F#seDt#)clA2dDdJSLgEj(2+CZb+iBu^
zz>h=3RUOu@8|3u_jQ)^3tvWCc(p(5Up7Q%u`XHz58&Vnptdwrs2zq&g`^v^eWR&Xp
zPpJjVe^;(j{z7n|d6CFRpL5;vGOE%sH+UR?zg7=UtxDNGAsSY|;d>pLz~<GJeuBT#
z0N*&HCxdP4;8=yAKKR?l^l&takgzZRe(H;@tHO6-p;XGF3B{v^Vn}({VTT7nsd^OI
zy*?}WbH>*Xy_@l`v-J&0A1<33CyU(TpDV6zvhW@I$n5Eq`<r_&%PX31S5TqsOKV#X
zfYd3|C1)_>MUX^1kaj2T;(Nbdt3D^HEn5{uesa`Vo?Nwu^QTz!;J&_N+yR(CEc@RR
zbaoEK4hBP$#B$1Tb0qFsZp4MU>YbsI8}onr*fbC5?*Z{=1-H3$Y)vZ?6V4*JNHYhb
zj6Cv8!b(gNp-yX;QwhhL1`bUsUpqrKPn-}CskKn8mb5Y`9>ArK6CS{~eDNA2t6ov2
zAwa2(6O4FM(?0Yr3Vi`@8O<8?kEqX^VuOt)1Ma%PVU<57DH8r)Vn7*GV>DUT=Kqqh
zuoXzG+jDEF;xwc$-=&1}WuGg7=C|Ry3EW7e%bLq4&1OZ)GTz+;(ls@){Az50$reWJ
zs~Y5N&<DB9!;k4ti-LBHK83&96?2du07}!kkC_3J?f=rGubrU{?3?Eg@4tT%m&Byd
z=Ulk`RXWPdk{*VUAav^!tzRrwq;FAr)(+16>nJ`$NzS;6FJUZD&4hT7I`9xL^ZTVd
zEj|(O2?|76R>X*-zvK{&uqhn|lg*pVC%bpTVXwAo4&E0}kmd)Af&&)sS8(PmFDwK>
z399DJWZCMNgOHNo7N;K5)R5_U&vz0(J~nn6sdyxmFusO?wts+@0=9?o7CaOYrN(k<
zL<Ca3q%Ho5(<ktPdZ|$8VX>4vgrt>?Uq8Z)5MDi@08mdt4L=H|m~6W~vjFOjbrka-
zd2LcuMqp@#amR^Y)c1pSb&GeDGXE|VG5!`VDsUJdlXJ5;4F4o=ahOsKPf@D!i^rba
z8@wR4{am~&>3%Xry{=U^{LL@&PizMtwk+aOHcX-a7o-zK-gpN$f@yq`xWIrTXJk8_
zyRI~y$iqnC-|PQ)oMI!5E+Z4?MPI+uHH2aa%5RPRfE%qL(|sN}Ge@D8mMKS+p8P^;
zZgdt-wW<$Ko_gppBR<k>zsV0b^X=6FFWGDTiN_1?#{`elIRja(NTDHD_K4W2)ayNO
z<$rJTYtrTvZC=<-pEH-#KY$e_(v{nPx00%olA-+d$hYfy98e@YXuGMC--8e3M}5m-
z5rAPOIZDPBtUvs{m-v&oXfJTs21abaVhNi{!r<YQ!t9V?O$xL~mh(nY2{?Ysf=JsT
zDE<U3LWTD!vnTAJ81gD8y9gw66*G^YIHn4BOh|qg!|AsnzwqgLnP-z>nPN%qLmY%H
z^~&*>`%jT*lJW$b%QpE7uufXeo=S>S92fN*H5&3<BsrIbROLL%#LS@m$Y7**^LIS%
z9_0uJDz0<hCFz7ol)9>*CjS}=e6WzigRcz~VC9dPJXUsbO_eXsJo8_&IcSqU$BX{*
zO8U^^cinYFqK_14u&TPqzMV+8Y%meNt1S`f?)%#Xes5;UJHe8LUxSp-w12Z_scnDQ
z*eL;qV(uCjW88lm@hZ*~x5BuEEUK&NgebEd_3>`+CHyvUBrwgucnN(T%^Ug&aI*Eg
zuGzg`I3SEa;?{*}Jjgc<aqs<wYdSfk1W3ezEDJW#G2?>gjqUSpKlgoOZ^cgk9@I?l
zQ$JEDb6Q%%P<gKl#Gm+U*&$@e65Ur@*<W_VXssyRQDI6rPC(YcM7Xs7&gSqKwynhj
z{KLTm0Km7wBnnp4@I3Urk?L#je9;eONNY+h`t%=<xq^<bxo4OV(sk{YUsN_~<1MY<
zF@Y)#o(GmY97;+baG<;x7N*u7nqm7KCq@Y%`MmN;0v-GkvhZ*0pR|I?h|}@2u0b+^
zh7!?t!?WJqd-`lb`VU>}ZR`J*B@V_>nT$KoEYM{E&K(3nS^ecOMyrArkS7YPymrt#
z(T{|%5~}hz;b9%o*p+^!6zqDxluFubQ=3|Ueow~FV%`paB<3AQGo<He-yV5WwrgAQ
z{M`qIy33aWmc(3MY-6}AKGuzfRyaDBMUMF7mFf1aNB2}tiGyF%?k~~RMH__=xw%N?
zt9?4`?P?Wf@4sz>mbb^))#54pZ#i)DJVz;VQ9jLLF66w4dx905z^jj#WRhj#BQbTX
zk>ut;{<SVDrU1QPnF+Nzy4T5hn2!gImd4G*IkYw@mPm+iD#={q`B7yW*FqirSIp|E
z)zBzO$=@n_U1sJ-YMf9P&;4lw1$_WZ1{1a=G4Tl|M(N0;aldvPHa*|>q`)gvz(Waz
z^_cy4C%YhGK>4D!b{xXGLJfD^0{r`eOnP5ZY~6xUQBXb4TDgR&69qwDpaBLs3$u85
zln&#-$1T)I(eKQWZT=}N?u#$=it`5=NRT8KX>qq6)8DYMg^Qtk(b-k^l64<#cuD-X
zc|dMbfDTsUD`S}D|FX)k+j`X!_o28%MD`Y6rVR5VVQd;U$5*l$P9*HkT<C0R#2ng1
zp{KvbOFMqfZ1pDqnh5>*Z~yJiShBA=c_@&<0}#mOx!Fh|+<Fwk5b%B!DwitXTYA{}
zZ*vG;fDZ&BeY;`RBG`q}U8RY~fy6j)5-K9<yp2m0#HjW$_Ri^jIK&{+JrYw1I0`&b
z+a3W;X;n;Yl<HrTue#S7ZvVT1w>|_KqUHg0cg~QlCne#X-wgS2j%lZ&*9)Qs-@DP=
z?zRlVvbL9zt_~#CW(}8_tWHC{Z<C>ydbX*Bh||lEw#N9d{=18*ho66-Fj373h5=j*
zW(}5)pb*3Oy29g41=?7t>VK6>&UgV?TP7sahAJFn+N@R_sv26M_cNFo9!FvU+_VED
z2W3t9qTUa-LDK-V6C4zMlXipr{B}*AX%!DjapKaB!266IMrsorxwop2Il!6Z1o^KJ
zmjCCeO{Uisps!X2_0f>Ha__mc*!l$w+*`tTA&*6y8Hl`MM>n03WXUjNB@#PIfhZ`T
z!T{vZOO$_$;pM=MmkB*0P(MevOwKk|OQSLrgVbi6^fv1Ow7&JHLFhG4n?yzlL@Jrw
zr9VSaNMX3$dieU{O~;I}>n<0M{9CSx9$Zj{gD=fDM(DjnA7<u0a%*7!dRJ>VrpR3`
znaN9IXmH|TW0TDq9QJ&vn;AYIo<-Uw>U}BCWk-?0TI<Wr$w#Em1)7B~!!U;IVgD}>
z91jsn-o7s^OG*g_9od8%volm}ZYGN0%kJPOMfH~{5>)j%n=UKbP<o)T=QWGLbL)Pj
zPG~c$^)$&jmF}bVKpPdx>2~caa}MYqBG;^{;3sL3)K%l4nD0tZ&qKPn-daCJY7E`h
zH)vp;KsD2D^J2Gd^D-)~B8tLi?=M=chPkK!K7BdQMB(QvrHXA`Fj4KR&*mZJ&Ackv
zv9*zG_ocHo>i1;Mbk<OX*5g_Xgcsphe|(u8yHMgntG7HXMpMbjL!y=W>db{8iClcJ
z)RV0Chb$mj$Xi>+gV+Wk&%v;pG(W!hE+Vo7&{ALqE3wX;<*J0W4_a#H<3qI=u~$4=
zHcA%%kkC@k%%5{uY!yjuPC1+#%OGzlWoY!f^Is*vjX-QH6rWPUix>wuxKgo&gWEkc
z2CY!ScOI3I@JD>t8LBjq3g%+t_cFn9g0LxJOpL;%9Cv+vII|(RncMD`F=^b#tENBB
zPXxl0-WM>1O5Pe$oP03cU!&!zx94P_ZUP}3!inEBOm~`V<S+i^9mJDaW2?tSD)QOS
z$b$$Zdf;N#-<-bMU7mi)yfDQsb<UNDfW|+g?UlJs)1N27vNHnB?*z$M(PyMZ8&FVz
z)L!0#t={9)(Iphoj;&-@&De5LJ_6JP|Nb2u!7skU^<W|t$hz%{zcGy2v!ajry6~8e
z5_WpTat(9PWW{K3quYS5Hr~kh75{MvSPz<2nn#aBVC-ql^GWOzt*N%eTT<fz*ov)d
zxRM|fWo+<$2nUf))5+$H6QzRmBsy4v6Z;0BAdI{wW$D{_rqEn%_5OqdjcWpovs&-v
zYp-ah{=5r%CYKduv!;?J+_-AGx`6nm$Gq>^*DH>=7aj*SEfUC%%|(xuTjK-!%7hh+
zXuyxynv1*P;p;3|%#aq^d+}!b&K6E4Aw<^KqSsItZwsBgPWZtj^CUh@8ig><VLXqM
zMIV0s=Q5H6rDy(SIs+#5Nc>cBnrGJZc}-1m-_9eC?Oef9RjbuotlKPm1nHOPP|lE}
ze)_&Qa~T3pA2()3+Mi)#&>&pvoz5mM34*B+4{k37xZgCe{Ce~Rx>vj-4<CUen{WMX
z$ZF_f<^!ARi%#7@4oavZ2b7HBA?z_Y=_B?h5c;0;Zw}0#X3+p}E=X+J@IGx=N56K~
zSgj%%&GxO<K4R}C#3cr+NB3e+%hmlK$l8Ik7CNGG%JCI=Xh$<f{>I*u6E_WUWJ*pN
zy+$p1fhFl`Nc*F+*3124=oR-7^YJxV(D3jLEZV_GN;AD1Oy(LY)12$jq|gGs-Ij&3
zimlD!=^3Mue1*-N%1B;>g^iti0dP{o=ybo;mephdzv`*|N6=ps8-DxOt+-@F?7ZQ?
zWcv#FC{e9sf-y3B_2IPW1sW<-n>C<1cBQI*@@5GksI36Fb#%1cVP>9Wv%T6HbhmN;
zd1jvt%%(Ku?wH#<xIc3mtuplzDE|o2ersv%K475}C(yR14Yt1c?Jw&wl<@H@<|c_$
zt?<zdAGGgAMhZUFV6Q^g+Og`0zeg$M5NUQvClweG;LtOhV0uU<Gu*h{+VeWX&o$X~
zNb6s3W6O-GSpscn65AFg71j;^LPMDyK7Tqz3qg6d-w8HSZr`f~clGRc?S;j6|0Ls+
zdXNFnZP$j}UnhXhUP!VfCD!E+hhKnGKQzH_Knv$k>5~>5+F!y{;7g58I{_&ORNk+)
zljNJ#4?o^Cy)V+>2Vz|t_NLbX>~%8h!LY%vy80wF;N~^OmdGCnmBwshkU|q4+FsyD
zy}#rtQe=o0+e_ZF^XfyH!1b?9pMBe6<tMbQ7+pY%GS*-x0kqRX=S^cB0dN_@r~?MA
zgOvTs^tw}0pN^6r$EPWXjSA=nsQzqu(|yeMqq}Rktt^;nVc7;qi{NR^8X3xCXC>SC
zX+VXm>aShj^aioJi=cVW9IX+d&c8ANhq=iuQ3@#bitnBx(~**5z6AeX&zIyVgBcS8
zns;XYq;NOH2Lx-n*3u%}_w2t_{rM+PfWlB6dr`c-P7caRxkEv2sNM~ZyL&AsU4dN>
zmS=yxFB=>t2FtS6Ot}m9p=V%5q_<7g-2Tma#{^gEveu6}GIM(eLO6&DyGSwWZd~7x
za0xN)JHM}ku#?_R?$f}?0P%;|(EE8bz!8+c@{$3<V)!{X)-%W8TkPjDApTZkS-z+M
zLJn_ahdB(x5cj-5?9U@*7ZJ^^+=p+kOlPjgvDs<9tO}T0WWYdqeJUw5n8^QcK2!0K
zU-@-mCkPz(0-NUX#|lwS3!h@Uly^N$+#K#)_RlhY!50r1l2)7HOiP8oTvT$zMd8x#
zDD{&ENgl5pEwC?&)W0GVXs{-!fpsR@<gFA#fQ<613UUQMgcVWW#bTwDcoSU)2(I6a
zZ*xasO4#bwO8<Y_9L?OdcqkNbyOa8y%%o`o`cyE-A2Tu+=hgu+ve2dDFkYk0>=cdK
z<wz|}X_e*95ArgkhNnbEXezDBMh-{#p7X;%QqhJp3Sd31oK`o_IXhebum;;Xjma&T
zumz+YF5pQ1{nE$nAvy=VNn^5arliPB9zTusY>K|Yyie>mc&UqiQ@zZ3sf5ULs1n{^
z|0L?OzzNC~81OVuSLT)Q`7v6j%UCudvGNMmcz(&B2j+6YF0U_DZ~xX<fMhEUF49TJ
z_6M@7|G!?ye*XC_gVV4AGPUKN#hEqpkf2remG(b=nh{M2lOKGp%RYhjT#c7GAq<-#
zlkTrL^GC+3ls>VHXEo&G1glWdI4?n$`vyct(AIdM+-@h?(JKvdwv_=K2{QXejEJS_
zD=h<f_XA$Jzr)gi$4iRXmzmiACd$OXifZiVchQI6)&i6GF-|TiWg&_XN|NmO-Tlo8
zB8<;B2S4A=(o86TfFqIKZ!evU7NBF>ymtl~OjDo7BD6Y~K@?I{RXeqgZBlLC{?NHl
z8S(FfTG;Ni9vF;|Dj7JrAb|u>V!AG>p!(GNRIYbt(<gW{HUljI7QK5LlQ(WB-AW1d
zc>bkc6CbRkVXebX1ja7nM9iEeQt{jh45tJaG&$AN?K@&YfK|=Uy`8J=1T4^#k@Kju
zOB&HSpY4o1=u$9Xic&+FLiMMrr#F`7yx~S$cFzhbx@;ibaoSRjTcUE@Vjxk6;Ol=N
z`r(dCNj=2=JV92-E&H2?8o%yu8LO_MP~R1L2wN-Y&>=36@HtamrQb|D^oGWwUN%k5
z;_c;}(UaE4MLmLkYL1}GhHc!(-uJ6HnIsFcFSe#piPLQ*{8)CS1oyz*lFy%j!c1yJ
zmH>DyM1=xv`l_ArEAMpYMd*gI%Z%@dgIq=#k_VJZ%iO#70WaaY#0JxQUY?QcF1K4~
zbA4Ja3!yq1ik;ufemMJ|{4LwSR`|9B)R$onL+X?txisOUgsW(Uw<o7WX|X;V#&q<X
zX%34ws{X>OsdOa&{?r7MaBYMVrjL%CY<j@dKE+SDB7i4{vQ<lh;RG4&$qhyc_D%;q
zzkM!}bunM%?JHQXx3*<?2wp*2AQ@tjT!C^ri15OVKXJFmks%iRz}wxtSf7@?c~k|R
zQTv(4>FJMp+taA|C3B*op}#$ORNN&f<Jjo>*h~To=IHScdBRYVu<Mh41LK?i7L+il
zshjcdCjgW8SNQzO5)-=&uIr55Fu4z}1_j1vi=0z;>@MZ!0EcPF#&Ll~$lEp;X#@+p
z2s}j1oL3xzxLL;(^~?bWt~~^;MacvV(Cx;!_Sj$LW4>`39<DX89pl9f*fIs8Rz{f(
zwM2eV_|Z_rwBDAVw3~kpSkyMao5MKU64y@%UJyzj4g%BL!SU-ltk<a)j=`=l@TlFT
zC@+!^q|Zk`<x$PHBCGKXXNNim?P6#Tvebp;kp1w@491i>@wL|#c5F|d0U@*d#dvq~
zZM8j}Pqa0NppyG#eQ=wX$Z+J@{Yzg)CQd&Sm)Etzb1di_{6Kx`DG(2Gt1q%#naAU<
z38MNNjh-UjoxukDH%69wwAFZv$NG2AP$++ts)DTE5dMcBq>GOA3T`HNu#HlV!FZ6x
zG4I)8-D#nd?So)SWGgJOpn4&RjY2RjtC$$CICWx7J@DUXm@!{LMpq(Ts^a$}DeK!E
zOHaLC)l*G;ego|qFjB+d$0x8>dW-C4x4wRxDU5$u#0ndu*PBL;X|$L&(#R#U@El5E
zXBKdM8C&+H@J@mA1jk-|`05bU9oy$#2cS^3Q91w4FDAx^V+klHaDKG2eJJ2VeZCRs
z-zTr&M)E}JxnccEqK%;UIPrJW?divtQts_K183!odI5jqGaz~jj71cxLKLrKnfC_M
zD|$u4<m@^4JQu2;3R5$$!wlxu!Jq9jjdM>NB0CazlT-vfc9a0sSTdF_$iv0Np1ztK
zu%c<;PS=onb|xi%lz9_b*;q18Io_yNHeD*MJmMbvhgNBKXpWkfXID?5cnE$~%LQqQ
z$UX7Yd!s{zcDR?2tYg2xduhJm0<6w|Zu6y?heES`W~MH@w!QqBFZTtn`$G<potQnY
z<+TAhP48Gt&q+hXK2<eC9WL;`xE4$Y3Uq7|WBW)V$`0SR26YO(OBLpT=LP=-oS&q)
z(y`|!Dn-|$wD~Fkpho(Ls?~DU_U2lx@%*UW=*PMiUn*W|gTY$w^YxD6BYUrD%Z5MA
z8_}g!U#=KS>OkW3Qr5Q(Hm6UTg=DMkZYzOjz!y?0GXbQFW$vK&O$!C=D&1*X$qP!*
zx}f3$dHyg)hr>F?kn1g1M!^w{0<U5I^XITtS~%JxEG8{8u8DX4%w?EC*T|-I)z|Hw
zV4au(CS&-o_xFYwaUg<zd?#J0uwmtPku=tLid|Hc`|);Fp?;21qR^KU&3}>(qRXS{
zoICcZ_4Cn?PA9*mvQoIef3W`KmIoMC1ja&>F{p%@k!uj%-`0m#Z)HRd=aj**6Ab)n
zktB_Oh#jGwY-1gP5ZT`~>PoyV@|55M1ft+rUzVlV!=k2+$F(9hZpt5zT|L65y2k>F
zkPx$|A|CufCbUHahA0EBJIteJ_Uri)2kiDqMcAaX4j!8iO3Sq+kLDI8%_a7A^juzc
zFHAlXfUSg*k*Av;d+0tE{rK&8@0H%G7tEZ=Y440r#j0U~xdrPC&Dzc1%YWWqJ<Qi^
zRi_DF^5Zo8s$(yi@CAVDY>VH_{?k{NsE{l)9bD;ARXE$l*)Je<nDFL3Nx%7OI1@p7
zwJ<HxqN5m^w81oR=XeWtJ8i&e-A(ElPsrl@KRNF6^US-Nv)USOJF3XkFR#Z^4}Z-W
zeQ6RgGJKfVx9$PRZH?q6cEc2|o>Gfy%PVd$JrG`H<(@v6ZJqUMXWIhD)q<dt)2fXt
zO^GwZ6jw<9KH^lgk^qJ9?TEtJsB4Q5-QijjQOQ&YskeVlQPmq+KavO6l#JKK2LH@N
zPW!G!NJolMvNavr_FtuPO(5wTtpV;ckbn!nk_>mW=g#Y$1JR)Zwi~Z4aZri+D^kj}
z_6;yXw?x3ad{vrW)b&0u+Rggg*J0celawIAVsQh<-WkHoM0$-l5xtEqmJFi!bwM)%
z#`ktUf?M}OwJFQzR6lCjcBid9`HFmpTJ(Hp${%1EQcC$;W~C+ypm}ybM4J^E1hKcG
zeS2X}(GvF)$B*Gs{1J@moPX;Ys0AGUd%^GBKu@o`uQk$28uZF0>=|vXQQkdIm@5RD
zn)clOF%qIyJRu>)7m`foM$w)wBC0?y1q$r@3s&<HEf}QkKVD0s`&uEzwoCy_p1px`
z@D5op(AzTLa3hKsax?^q?_<M*lBV-9{UX(V(C1gIHsuCzfB1>i%!1fA6vW$RS+{B{
z;|ih}M#xc69sAsuc8kmtCalB!4<&02-SA{4Z&^XkFQwO+t#R3?r>ozetLxO0DdK)o
z6~7ZmfJ*A252l*??6d7a#{NPo{@ZsF3N@!ke{VH<?2^^R{JrO(QwOtpz_Z;3Xr=5S
zj;9^vJR{t@q(aTlo{PuJ1u=mB*Z-Z(GmK^^Gk_sd`}hbHpGaT?l_i+*4FCT>E&S}&
zA8J3z!@s=%Wx@533D@g9kukh>GmvSz2?=?B`Pq^znkCuLLGq|$*?6B0qj|l<cS+of
z4HV>h;wD^)7H#w)9&h1SzH8`DPkRCn9OR&ADByg24p5Q=)p;i4+Ki7s0a&X*e|x{^
zjt2EG3>rE)qD$loGO0htGgUe-a>>1`_Z&ZZfd0x|Tq5fD43WdZM$e}Z_(zUxsGib;
zUmviefB<yZzPa3qNh!n-m|eBgxFPL>2zMSB{^lTpIVMUVmI4ou*xAQ|kK@j&+nVU?
zTaWxW2)y8)%<j@AetCi#^PDG#=4BT<(lvRvRuuO#$!I2L-im@AiCZG_gn7Xz`QE^W
z)Q5?YLydbQ8;rlAXMJvvzPe4M!N-935&qPu%<Se3CjHO#=3k<q6IawwjSP;kn_&i^
zn%|)#Ojhu7ymbWyL9z(EiC*Ira6ci{yzML{P-t5N<xryg%r@?5L@yND=b*L0#qfl6
zX0C%N;*|hk-p-0C)|;140LoLSlr<;Z11%7&6RPJpT%LWZ048yIG|dBB>8f&WHz@mt
zZzG@<!N_CVptG3iOTqSsXJCy*R0qVV7io*2o(Dn{EvUP$`>uQ#sY-e*@m7+qHaISb
zA7Ocg@rOx}^vMw8zcjT|(Z3dDdj+;$@Iix3&VjfVB+yD12eDswtRAI!-##TcxDM#2
zGHbq%QLkXQARX{${6o^hnjV8q{-EesjNP(P)MIBSyT!M|y`id!0BJtnTt(|Hzgwpo
zU*NdspTu#m%WT0O0W}VPb}a9!euX6t&qqAFOJ_BzxysvdXonv1FJv+!${$2Je~xb=
zlEk4)ig&K5-m!L1e(?f&#?Vt$|1-;@4No#;-`F<{{=SKN3j)`RI$S`gJ}EQV?45JR
zI;IO%WNYRZ&}H8c5jBZF6z1<cX&SvhQ{~6uA+bU6C`8W!n0ezq>jc5{UNYIVuv$9b
z#<9erBpw=(PhR<@y-NDyE0@sF{ie-G5b`x8_w<^$)O23Xu@8g{7)PX8bMLm9j~=JL
zasq;&ERlBw3{LUrjX%3JqK!6OfFClQAFYF1UUXJp*Sl~}O8jXPu+5--I6$!=YLwJt
zR7W>ahzli@g5lWIrp&wBHdM#V)H85KC^VLuSEDzBV<!Ghj+c|tSqi=FkM~@tnav^!
z6bcd7_WTMN?otuvy@^82^c%Ik<w>ubH)S3z((AosNcjClYILpam{IfM@1vxA2%zlZ
z&DcQ{2Q)<LpYG4+f3l>0UWDVglBymP<g;if#0L7XBnTM};{MVr(FXiQ**_zDekyOp
z--12$C2rXB_+$pOxL2=U1_S}~2=}_Zocq72cF5@N(noL0U?ckF;k#f~X{zjKzb{nv
zAzsZ!)T{9byElTqx5m>=>K)xj{DL=cV4(PBF%^QBj5k~KPkEH;mgt{hn>u#!ngZ0B
zOTw072%Yg_Bn=A{wKho$<%qoJ#ek}G(iO|J665i_K&}p2*w3_Ys9R`?8yrtvW}n&_
zU4_U@=2}AbZWtN(jl-Xe`=d#Pe4YIx-!?A)bfI^3O-Sd9(tpz~86Z){8Xtn<h-mfd
zP2nQL%+6xKx?hKR2h>rT5F_UNxLS=g-6=oeLr%s7^D`iytLoidTuDf8@RJQK9%xpA
zb%Pgd^40W;3@uE%fl0mvPh(WA#L-dEbb8tc5-*lp=zT<Y_xrwBHl)E2m!G0SS3u^B
z^=kw{jU$ryj|)bui+$fotl%UMDzyQ*0!&W6vzX5EIoR~|jjExElg0I2LH0z7?oQua
z;%LN1_HOHp7~hE&HL_OoafH*$Jui3nK@GGApGXe+@z21P{H5;kpFeh?G|GQ60j6Cg
zf-)fgfDG<i(R1a0xmz;%w6;CaJK`xx#aY&f9_b*aK6jGAYE5l-EN4T;ceaTWW%ycp
zO1<$Vn9Wn3v)GOd!@J?mwX|Eu5~<Z~;3?ay6Q+fwT>8hl%(UcH1;G>AH`oA-8tQbh
zfD)lS_F~2Zo14+;w*V9U_R*WY2ffsZVVWm<4|9HWwPO0^m7T6W?Yrqc*j3)lFjS62
zLMxbbo&+=3P#>qCDlW8nnh^4ByxA5j!p?sC0IF^*pYq;N-lGt8`ir*VNpX8a`XM&A
ziL}HOa7^d<EChh9wU{m?2udW6PJedaNSD`6K2H*Tx7T3?1m3+rp66*d^%khdKtg_~
z-%)X~OR9f{ga>wQfu<NJc0SKS6Nb_DyZ$hR-s*SM`8fLzqdylu_!c^UUsDKhvSCyt
z(eowQgJYm|wcn(QnVJ9>Ac9B<5gY?{d-aft0hT7#Y+`LJq(jut)UPrIK#snYKaF{B
zas231F;dc+l1BXOIC6ydv1YgvyIfrU=6A&gO)fdP+z+wS1YyqBdri>$dF^}o!SEYs
z#wq&J*{gq{ueP(>cqF7i?6E#$678*40^;ZOP{)<TBIuL}F7)|$?Q3`&oZ-CCarmYH
z!+oakLJngs%C5F3pXRe6s5&`LM)JE@DCi>{Kl!~=50y}a9z^GJ+LyYOO~D}Gek|DF
zSq7Nlo^MwTULj*nCR%D?H%~QAz=J^!Ej`)0GdfBY-fX!So6QttUT$xAGAYk|jYFX-
z-Qc+itr3~GABOd$o}6tIju#3qyAWRZ4Fw|i6YIm+tQCrekwm>rsYi_-$`2U#?0=PP
z)`3z0^0C{C=^$;Ae!wfK%QL&V(lQPt(x8cW>Q~NXL8F<q9P5^a;kC3V#fW&|A(BTf
zR&&I68v!Vb9lAiPJtu$u4GfZ9m;>GO_Ti_?UDe-t<hQ15?Ie*D7vo)AnYqC~z<r*G
z6DTHtN20?25m6M0g2=CTswwt9hgbByJM-P+FuMX4|0r{}<3kQ)-e3>QF3j28uDE$(
zrGk${R=IaRXqc8>3GI^{Ds-28U~XTvbAIOd>k;d<TnrgZ)(@+B!-&(h$HMU$N=D^D
z(%qskbJUTAw!P@Ys&6enWP28OZqC6JF>&`zD0pRdcao{7a!+&-Atl&pNeo^ZgAbpw
z(IT=eyI{iO-nm6|uVOEl$^dV*WNO{`CY*SLpE7gD`J92JY<-3x#DVg=GIYZ21d(s*
zDlyGx5zvGNb9YMC#y_aR^xTgpt~F-KFsdl4@2G0bOWPfb37F8}*U*W7?R_0)Z~W|?
zj;Q}yl&J$9NW&}N0~(4#baBbf-vLw*De;O|(DFdq{I}Y#v^=?RgcO^GJSGr4-V1Xd
zE}o*Gd4@`3scA*C<5Xyz6L{mVefh1IBV}g?D6`ZF+cE$v%$1(yl_NCfdP=A}=q7zn
zy1Tr_ZI=w#pT;+f(Fv8kuhOM89MD25+jKrn8o-nfpPdF0^4e*8bE4C(^A2Px61=Nf
zisp&G%d!|1rP^9+-#1XJuQ@hu>saag+>-il)cr`;(X`cH-bJ&^5p>+X+Q&6uIY(HM
zWWR{#&*IBDql|G8ld{COlJ}*>oYwwN_@_h6e`bsb+!QfrsNmiNdiDOp&0Yu{em~tL
zn7C1hMZ>+0=LW~r-J0J#*DlH^EW<yll(5b0I=M^=Lk0Wo|6}Q@!>a0@E+8PGv~)-}
zNOuazrAxZIJER-wxO5|pfOLs;r*t>cp@4w&xB0!_U*ZGLJ@=fm_sp8L*34d#!Q47F
zi~;rZ^8_AHGKGqe7a&o9Ls40VQyM?S44c)TrauqD7^l?OXfVL<eCKs3Oc2%uPBrdX
zn^P&;1|`vR{~Q?6*%h!l5=OizR~h^q$VqlGe;Bh<_fTP_$H*~!C{gG7C=%Mrp`hy%
z#49ATib0wk9&AC{IXi?1rd%uYeb!KZ9+2z>4$u1hQy5;S7t-9Kiw33+1%TEPB_eu6
z-E3Ph2Kfzf#~)3TE<xbxD5(;Z1g0u&&I)*J6R`*p3t)Bm@7Rj#w_MD!tm``qVwLaQ
zm28sZbZXCR8+kaB5lUE9^h;Dn-lDu)rwkY05O(G4VfO}5`nj(OXrE<1D4=4UIbkPb
z`dig+AFRDo?X+2n73Bv!Y}>#6Sq%G9{K>KKe7CFt_x4~z`#kh{hC4F-c~Hy+Sp7dS
zK+tTQ+tuW^dFw|!RoqL%@htv=X?U1em{m=`e@?C7u(5LYNatw}H39Qy>6#HYn1896
zk{&w|d{oATfpAzg1TXPL`rCb2%`RHf@UJTnf?XXnKJ2K10g;=J)7M^FnVjr2$RO^8
z`WsRWd-L`ze)&X^%Td?6;M0lvzSy=K9DzliI23iC0ceu8a{vD7HTg2VBp5Sr0x9q=
z5qJE9$9m0jW`W$-?6^(MvVIj<7yO37ix)2nBWXk{3oKw@Q45<DV0m+sGYj4Aoh!4R
z^(AHVWxR)@ghh0~z&d+@!BjFQiu2s#89HG1Pc+OZ?#xpcoDn99fNqhjh?yFXGLM)@
zo_OT2-uLdw;&(+){=yQ>W2PP<GPWdwnr%&MYb(OCjfaPae(NP+$m;59j9y_7#cFly
z-rb8$>_jZ(y=y_?>)KWq^Iai}anC&=2Wiey{b+ywnLMKy<9$n;qJMlV>$kO@^OEZJ
zd_%41fo1LvD51UITZ9hgiof0q-Wx0Ex)$GvM1_Y(KhmG1(%JR}d~>r}t!9!6^P*pC
zukuG)M$<Q&ol_O`JKL?=uRGj9buwRWmLO}|o^V@2*F=09BK0-w^Mm9Ka(u!Y(e#;F
zJ0^iOMCJUC!R*JnT)dE&zr<}hua|#5yn!1@6DBz!Of3<M*-~TF2yh+JlotBo!{@h_
zPV)Ji)9x@OXU3C5QIy)JL28Kc?5O?U+Ddv6WUG)Ki_^)k*IP-7f0!{-o%<(5Q|Dd}
zPgUq!S*tUJIPpJF@!a(<=c=ZX)&5>!j?HO(8kk*tB>m2>{b^M@u*_*Is2?t<n@-mo
zNB{SWc#I=1`&gl!1XfDM0QR!DBD>ef2d_Fqv$4x`Y5emu!_~`%45y<e{tR}=xGrhS
z3@29?+t#e$jAmw2+_*rODL4q^W44MCmCI)SA%^_<JF7z~eDP~j6&;)gM_Bi8F~aE|
zL9NQqkta)XLm+K(EZ~&s{*cnY8+un}X+>YAG?j5qT|uA3`jQ)b8l%N7E5#;dLH^m<
z>)nI$CPA;qFknS%alKe%CF$wNFq>|j>QFtAAe;ACk(%Spf&D}qLKogz&>?teW{RM+
zJO7GCwzWX?!MX3k(d>5;)eUlGt%&3ka;B+~AC<{1s8I<p@QK5!{FhLVGljmB6x?wq
z2r$Id?k&iJR~O5mfGqbq)6u8zo$9UY{=wm&E)m|o7ixYUk$APg#uL&bgimkVG}Ns5
z5Mrd>QVrp1c++ZSu-C%TT^tP>Sy0lqpwfAKBa|yggvh>twoYN_Zs$LQ2+B><aY!|q
z={^<LzMbL~KsOk$|1+$ZB${3y(3pbNR4zN1vB;@BASqc+xB3YVJ=Zg6quJ4;lvP%h
zhJLVU!&C)L@udYa_lxMvR-qvsMF9=;N2-qUiu)qT;#YRmjQ$<Sc7(|$Ct0Lry4g~L
zK~JzQ{q>6V=P%#c+&tp%pm)RP=Q2Uk3n@wh6CjIL>!rPR5c1o92qBghoLu=n$VU=Q
zM)oN=BTf2$mLA%eCO8NG!H%WzS0poquZM4iG7Y6r3w{2%B3r~U%xb@0ph=F(&3v4=
zzkMp~Ty#{;Mb_lkXSqTgH$0UDVW>q=qA@#LIhTTJ!n|`kv-i1&s%!dPXNc`vj{K&z
ziuoc$_|8{?NDHHb95_@A$Bs#@^4Z7MqhlL!ONg;^&#ASGitoZ3Y6KWH9LH#BeNrH}
z=O~PcTj8z<-(*qD2Eqp|r#tiBWz4Xo?{mAl)lbe&@6Pmnt`(g|3jRL3y$&HU6T9zE
zVloabt05owDH0!g#u#td^Fv>aF8yD2!h!5{&GrPBs00c*7mZ{34@rvWHE$K2H^4B3
zcL~kAY#XqjsnU-=*5aW0o-R{#ezqQa0sr{LO~jwptG;|Z@<7k*Xp&6y25Sz#XnL;~
z!9okaOs-N!Qj@ofdvajae@kKBs9f2G6RD7`S%5gci2i?{p9r(}0ctgQC~kQ9ua5}S
zy(v^XU(AcTE1Y;KM=tQr74Hr!vMC)%9wFds;)RNLwUo!185LsnOWm#ru%i~SJLPC7
zPcjgCWdB_Dc!ZgEj5<V!+o2;O(1J~Z^x(f~qom(I*xV@@G_jJZ3BrxA?`BFwj?@2f
zV<RJ8#KqtAbrgg*RlXoK8m?0HZ!p#kVW>Pg!`=Ueq_K#X*~0`=w6J`S|J?{IxE#}n
zHpg?mMqW6jlvSgljt>?!^&`ze8^<!{>fB#cfQP59OuL)&zq?P$C}3YNaqFqANKNh#
z$*}6t;b_dw8sD>+9{GklJN!oauT!eMu*v3kN*JU<<!1h&rSEc)<T@~Behl5cyBSmv
zz4zS|(+HnKrmM}*B$~bysj0ln<y%JAPejO&sjo0VEHQgAF$FCN7s=L6M!_b<?Ay`#
zX=Zc}@4PcOihRcwA<RNYmF1PayG4eN5`^6>@G9lMhuwi5IF6ZybD2V=z2>(sp3T7h
zYcz#zatvqCWIXgp(djbDL&T4e%=ey9jkP~E`Nibq9MtR_k3Nv>KB(npqk$<JM?yz-
zD7&EU9ov-R-x{zYHtsJ=U|_{{R$-n#2i8iMsbG1cYVPiBpBkcP%~`@f(Avs4L%s^x
ze9o&XVIH}#p!XM_sFrfN`bHG_EtYAoqx_CKMgFK;<fMG0TUtzg>{wykHsWVM_1(UF
z<8eED0EUW#zTE{_EQwQtxFBJWn?%?^F@)ND;Tq(kUzaJdJLG3hOwaaHc+*^wbWo36
z&?=S4U|xMoPVVT?Lbg?_ga4KzyyCybf=ZRnS*9B4X|hav%iF|5X1QVqi<vMfZOFPp
zVm>&P^qh{?@t*G5(-tftP_>8Q6TF3n(vXM*vHVPW^@*T54btta-C(9^oI<Hk@nZuK
zVqYN=EqwnLtc5~%i2ir4fw|0N2;8oDvF<t#*JN3u8UvTSH`A}sD4w@%SWBEeKlxNr
zN=JW+uEC{1gk10UV19~=Q|@!PzlUZnHu>R)A%3qFR1J?TI<NJAc=zhB*^VM&k_>{*
z<B9AY%^OoADzl7f6=%CAPMpYI)~ZCjlU?u)VU)!w`TtCjOM_2V^qw=r94sXCW1UUI
zZDA&N^FFvzs2xp+0AUJyH9F@Tcj(1#d_j<E0LUqPjc(Cd<{xO<%|yv3>+R{YazFh!
zZb)T{51|O$u7!m-JMnp6vibHqNFZS*p7*U9H8CkY@2DFWB%l4Vcgz{cbRL_xs$ly-
zsdfrPNnPv(TK}x{>7-a;(d%*I4V(-ZW+@ophK;611&fPC-4?ZM9}mmUdpId?v9A}G
z*vU&>+d=oY*L;x*?FabaKTE(6_m3IHGs&mtKIYR#iPAJoN)W}u9whr4(zPYlyg9D7
zg(>b2s})TY1)G?7RgfRzm&Wdo{+)m6SdS+uj4LuYzC6zNsfd%v#@tQu-lr7PmGnD`
zL#}9mhxnHR*un0Un;TZoS|n4I?2YBgv;{#Z6~B<jSqh3~rYe;|tOQR%L8T;prmpR(
zHzdW1qMANFQ8jjZFsVx#+(aB6a#7IkrHsqYm5Z*>&PI8aQC`(ZL88l~tT@JOFM`10
zk0Xiz89E7U-K<}5WB*HfFR}NJ^TC56p>um&!sc(N5dThPi?G)?*u8uWyW-@YV3q-M
zzB4>bKj4&R#-0#Lo@iL#kX&!229izA2<E?BatPKyV{~y|gJ*u6kk2mZ3mQ9dBn)X2
zM>j_(HAABIpiU=y$Ez52YpRXMF4nZ1f_dte?BwZI=m0|$(*#-kJOiUelj2()TvaW6
z_K1E_gU=5{Vane=k!s=RzB99Av{s`_+uzm|^w}45eM1-n)8)5vjQ9T=p(A)+;6<p^
zA(ACpZp#3FNgc#ms!YwX-%60*8E)GF02|bj?EazQT)EnJ3MD@$nG5L65*c@cI|Nw@
z#bS)*h-U;R2PCB=ryJL&()c?zXgHNu6CNH-RS2G=!#vjCPCX+PP9dzFpL*Wn%l$ip
zh8ZlAV*r4}gHQI3T}?8(B}iH1aB{!<#5Bq=+#L_$_WHh5Z?BmGtVDmt_kJ@gW)4uG
zYFyC>FLH?fW+UQtz7iG<5!J(sBdJA?lKV>e98LKML$5S9oTM$`^*&dLRnAeBku00P
zsj<;l#@(m=HJB2p8f#KBnghWNlZ#B~;v}IZ?pLDnkL@^rjCpn0;%146BDs~^;m<?7
z(~*q_g8`jEZ?f~)!KA)mjAssD@5l-8W{|V}CJUVz+perq(a^|bFuR~FE2C+Utv%`F
zx`z{g^s31)W8X~L<~`pTr-@@r#~DyENL7D5;{5Aa$mc%jDdTQVg`wHglAkZpP+7!p
z6LwXqLJ}#EN|R}Ycz2Kzva9i&7N!tSW1A3+2|JVgJzP#sN2s_8DvN$~k4O0X$}f)+
zMDy+Nazhd#*x8Y(Wu;6M1Y!zAb2?12AOStKk2vF*pGNjv@ErJGX_D6UkVhmEUh`1y
zU{XCNUv?P#8}mTk0+lkWc?YeEVAuSijKvTV#+MJiXoqBagB)h>wvx6rQT9g2<LJ|K
zW3_0oW0&J-G{#5$@7rPz)}y&!k5Jz>=fyi}U+T+=D&Yr2g<~~3Dp1FfrW-{3KwwG?
z<E4Ga;H0e*&LT#>iCDF|K3%NbLe-GOs-b=#+})fbA-nb4I!D^Uc#wXe28Yp-9Gx~@
zt*3V|z}!u?=6Zn{F1Y7u(A!qKgz9Mb9^fiYLb1$t*=C<3KGAB74DC<+EkgxTQtB{f
zrdw?c>*)RU3zW$Gp!kp)w#rGs3ay0BsI`6N&T9UHq3F=jVU8)>c$tVp!p>e64lq!R
zO|t3Fn#>w5yC-}UO^dzz*a%4EKN7Byh_O~yILnZ{*Qc?mJChWNQnrZf>{jTZMXM;V
z53gRkwm2DjL46myy9);OJdu9;20vPQ5)d3Z*&bOkB#knH;Xnq%b@Tc&mOPY=jVBI@
zh$ioKzp-@DAty(2|8AFV+O&bekS=hn%VmwjNaw5@AEt<hL|M)#Q^q@@y^bn$%k6{z
zOf!~mm{-*^>#I8X5S%DMM7Zy~-;q#zPMb(Gm@7a;OVSVJ3qEu3X9Xc}g;$UnGQ8f<
z%6h=z{sIQeHaS`3Ia((`ka7fiLS^dV;BtTX$nKGsV^CN{oV)Cdjt0i`Bs_|GH?`-W
z-OZybIxi6ur00rl`tb1{@P<Vnh|st_y|&S{q<@pO&K9`BH-_a~F@FcM^jA;ZGzn#+
zQHlhVGH=+eUN62>qB+6(b^O_C(gLgA_i9O2#-j8?pUMq*@}yBr`v(V(w^KbLQi5Rr
z9_xriFC|9}nn==&8Kn$`fS;=Y;-tvW_b<0erHT-;2~(+4*_0q7A|@1Djdl5Sy7l^x
z)Z5uklgti53R6(SNYSvt^NV~eWGzNl6ZXDD{y2==$?#!nN?wg6J5J;Dx96YUhbFuz
zfS3i;(ejOAw$AYrAa$lA#xelw91HqnhYzkxASXRz>jiFEpO0TkE#EMEu|&rFdyP#f
z>AP~0J3=W&!%X1-J5uP$`E70+BTK1lrv7)wE5)k95101(3z@Zj=l65x6LTuHzB{4-
zYty5nrE@9{QK%E&I!Hk8_x%tO5u5TX?g%4<6_CW9LzCwjd4RSwl!R2rVTe-ne~T45
ziW8&6Xa_rg-7Z#t(v(!XU$~=`OcfE;N}E7w6Sl5`rL8SPoN`s7m927K9e-n2s>BY8
zK(ql-5oV2j%)hneLrLV7P}jd{e9OkuZAgLpG35`!Eql*(o)k^XkHsxUoP3h-K^qSP
zcH=OBFs9~IobLu-fwk+;h)I`>!p(a4{wU0B=j`ppvT~)a*i|J_SdRTt^q_-oIaRrZ
zyW&($`LQCK|JY`nbs;AK&%kTvxRJseE=k<LQHS>d7erIb|74_N>{t9_%wpwmSWVLl
zi(g&3S*f(bA2nr57C`REBLQ;|cGCvVx`hg&nfmEC>zIBilxJ}#Yr#~-=7~{i{6wn2
z7b~xlzxP1vFm2bX9fBcUt<cOj$mRFH+t{V03v9wE=1zH=%<K0E``Q|ZRtI;QipvbF
z9Ou4DrxkyMgfKTz2vn1;fbprmoaMS>)Im|dydsGuBUA6gw-RLPiKXFaozLcZYBYZD
zkGo?^>i2J0Cnv1~-glqKTR6@RwlV!L3W~rQscnQ5DsiinacTY3^$gLwkl8<?9&4hD
zD<YUvH$N8g6BuLgA1QYG;xMq}JTP+4rm2&zJi#yE9k+hl3&$#bvZcm1#_?zW*AFV{
zNdJGMi2!_KVBsP=ev}~MrvBtnxYGIz3Men?ki04Y#xJCTWIw=(UplOKOMW(gllT`A
z-B7hN$TjRX>&LIGU3<6jw5wMK$_A${=_0Y>o^}4k@G@mz;XyLlJuAB@lofZa{hZ3$
zz&{$v%yuJJn$+Og$|imK@qc^%W)YtM&P4mZIHcO`@;>kHrz6=5ucE{i-98mh<X32W
zORK3?ySRI?C!M9jF+1%kHwDR}<;3ZHj^|e&c%o;NA8SoHw-#F@)9?>yZ#ADfm8_S?
zcaxU|U|7k^Y&JGGsIQBoD`18P3p23(b%y@!6Qs6ue~J(Q3-)UYbNsK}!;g#^=2()7
zL~9u}4;jP0Ocw*H6?q7_Af5l2$yPh-8?(%awb;WowziGmHwhN+E?c)^xh8Feg-4xn
zb6mbsJC*z|cG<y^uBfKr%)ZY*Zz)%#Gqh7k@Zp25=tg-JwcHKHSXBA1qCOOMF{g1h
z7p*k0`@il()q)DyAfjcr->LpIv9|3@<hU40TBDmB2n&;(YF_sFw&n?o!J@?3H+J@Q
z#rwZE5`_^sZh=!Qd4XKb=X1-##ehpw;@MBtG+(UDr=aJRRuM@FF>_s~8O5KJI#o;W
zh=od4Jm;&L3v@r3I~u85K|T($IyssMVQvO($s=J|Uf&x`(IkxeuGPop6`TJqQl2bY
z0QZ3^w6<8La>#RIhD3XY@jc?SE5us%OMJDRdV<h^{aF_)@9e3OWgOxt(Y)BHTzW@}
zGj#<U=R^#1Ns2K(!Qt7jQJ-2fNbbH#!AEQDYpw2tx@(vjacL4Kj|Il-l>c4|LJ(>B
zJEmF#6$b6l&=2O%!KZPIk%mtwx-kWcCeK*?Fu?^E-0gR}0}j#-E>absrN^v{9?g{i
zAzRv_bYLio9y{5gv?}Uis0Bcze|oSD+V4xq%NM-w)uYc}W`k?dTtol67IhF=n348z
zVMz7e23dv2!zrU}CasWAG{`9=8R~`uT{iHcAy4%qr)D~dvvuD`hI6Ckvar(d$l>>@
zXj$94C*Dfk)D0)lPYzgf27T+9-*DYv2*3yuZWgdj@&53?KsiMbh%Ox8zTk#zySnD1
z)P8nVptY(3v1iE+Bn9_Xdoke#7c8O-r%I=%H9z-Ji=*`=N@&IX(4ddZB}!;`boB02
z<sO88jW?S#PGNk2-Fu7KUmWRH@n77NyMFi$HsZ!<wbbdG{}$W`yr5&>OuwwSwb$<&
zB5|az8gxla)YTt*YL9tcV>Twj)6i8#4eD(4BVsPRsTalRDKAP?%bhWU-gmcn(K_Am
zaikS)4=vuLWfmz;d43pkHeTGtCQBCU2N%&ief{6=Btdd*FPAouBP;cJyj7Kpc};K7
z%|A4jj!!W0U763F^m`74jA1#V*~0f`R&DKbN)R-J4%KS9Sl@;aT^rEuUu&@wUL3E$
zv75b<nYxks>2BP$M9DTxHuY(RWOuOb?f)WL2PVZvmzCtL@nV1qEemfVr4FJDZjDUJ
z|7ult(iSmG%?hlvPt4Vxt17lSY#4T0H;D;1<WMT*4lgzzpWv3ujXRmV7A>!wT!U~>
z0z=2StTtJ8D+KZH{CM6s^T{Q8|Ch-8WU1dhDG#Y>A<?=1UGcR=(NuaH=LubYCU51f
zQw`@gk3zzz+fTi>wW$|stTFhPw4Ff<1eG|ltx_BQse_M6xuZ|u$JkLp0RgfXzN@`I
zvBlq+9;)M32;D2+B;o?{tbZ7q3W{@r^!1bhxFeLi!s`7g!kEEUfDVUoe#SHuX+d>H
zlh(rdK}WCCKSE5*No3(~`m_1RS_1HTBz7%2rl!z5c<yf>gXT|vJNX@A{%#fb9P@Ow
zCgq2GY7D`6mGL8VCsVOM$KqT6m|s1UQ@Y4wI#lxQg@)nUi0BlW09t-+U^^P^@qkK=
z+L|es=TRW4o4`2DTUbvx8+E4Mt{6&aR;rSe6irzC1QN<DVMGb&I;4)$bl&carjnd0
zpCx=GXov-n^3mYNrGbo6R%_Qm6ip=QM&MH`w;UZMQLrlylVGIz6VtD=D4rqfTGvV4
zEO$b@JzEyvZQrlfi*SH#&{cl~_c8Rc!|>28BMMvrOR3CN*q&4_;~cM2W!h-tTysYe
zsc;hj#0#G1qT!L{+;y0uj1w>5e%Jh<rP&6n#o1gzJyFr1ttfWYcJ0$$fs(W(C&g(i
zpd6fCM=YaXU79YA3{t(oLnkA2PagmPcK^__#?MdDHd9?r1@01HrIR~Wll6#Yw&7fi
z_l_RA46<7P+og}DOaeR530=6+%F5B^Y8Namj!dd@2>77Gg|Xd+JMSbE1nZbE0G%9H
z{9K`wSZ1cC#?78jT=L&pH`*L94&+}gw%XHu9Kn@V(b>dQu0L)_+r}dmmGf-UZ|1X2
zDq^~5A_ti4-nZG#4O3hd?~m`nZ7ku5`Jbz7|JsbvWDX9C#u(bFkLojoua}mzqlp&e
z{(D|$sOY730#N{FX##GnXf-jh;yx54gZhE)Cq{qqLl#DVc7H~|%<lfCPPNm5P-er(
zp=f8iELr)~G6${+2hJREzT7Jz#os@w_BeTWs76mQqAf*(Y`fx=BfQO>uToY6n1Y_C
z7^)AAuOQZpZ|S1@S9bwVfEqv-Xd0FIV6Qd&UkczcHqw!S7GjHr`LxtJFIfpL))gzN
zCkp$C%^dNjddIB-7R!8KSd8=ANAZ+~XgN4c@pnd#3g%UUFujFJOEuO?P+u!Rds!`*
z>D!`p(UC(=92?eYWwm!5q{d&+U3dV#K@u_31scBuq=tPGK9{(Z%MrsQ1(6|n!o#Df
zMkmj41Mi^aO!Q+x=V0PRc9s0<A|?ldbTSM~0(nr9Xqx$>1lZes-KwtehZgMTwq50$
z-=z!sUgEyhM4EiX^O>@?@IPSYK-+oqx;Ru(noj%f_A@@Qpj#^LrHqMfk}I{6f#PR(
zc?QSrHCg?Zw!Muq0RT3?rZB!8(UICE`r*5$7(9N#>;4D}lNeTSIw>KLG+fwHzW(DW
zgWui7*w@E5RN}=7G3y{SCE<U29v_a(D{*Z%BxI(a^=+@=?=v0Favc;^lO8Ibvho7S
zyW%&Wz)XQX3(~B%+cXdGx~lY<Af6+prjMoon@ZX(FFxOZke(g)=<mJ_`%7rjF2cr4
z(Pz$6ZnCF5B^bOfYIB`^kL^rW1pf=q5sCpW6<7=WZ|O!2*Uy&0#UlUGKs4ZzWe}K;
zQeh!m_YJfw&hDFC*d7^k;4t{c`+q}&Ft#GDG3sr2+jtl<jZDSFtLzzokfLpAtMq1>
zY=2_!rYsQa24nvl9GABkWVQdRSKBM$&FS35vFg!Yp>ZtuQ{2epQ9;yL4dVaOgaJ8w
zbXvoK&-aFW>@cPK9dhWAndu`=2!@<qZE9Cy69!3T@h5Uw0G|sS*N!3fE<^p>e>4A(
zSvMeVR47T!_KXdENNJ27wXjKKf+I+;iZ3@|gHS5KraObm4-mw^YB*)x>j2MExV@1q
zAt7aAX9gw}7F0K~C+;v`%wvM`3)G4W{^#5<K!LO9x!1?vr)6GvUx4{4*XSoSX21lO
zr^`J?S<Qm<BpO=8O!>d06QKkrfRoHQtooMSV$8PtnG{l=pjaHs0k0;d4Hp||OCY15
zIS{5(oD2`%v;1)N{cw>J_33`=H7FIL6YL09OPrt&5Bd2YT}c|l>|R#&GudLva((ms
z+>H~%WcS}2f(WK$qsWN65JzMc_alPHnDxxMv@X6>T(z2U6L$FVz{Wn)V`Syer;#^m
z&=nCizJR71XdsHzfADt^<W!KA<ea_zV9D!G>q3An00xPi4Nc=S-}B<0g2JV>#vn6P
z64Sj|R7S;C+7FgO52e8W5+Q*RLM1Ln5K(xEJ1@WHj5l=ed|v;PzATA00Pfa{-&RNX
ztrn^})2V~6)ry@kkg34ARcSy&qx9l}f=Buxe-g=@W&LM8wYO3^7kK<8>)*fNjHgl#
z*$I1`6+@a3emGtthT}r_L(8{mw3gV}JJeup3wl+`nV_{fR1$o7bcQ*DQr!Xb(mzr-
zf6PfwF;X>feEGWk@Ay?ht0K+p;3Q3O;S3+Vni9jCik#VO(vz>?nE+zO_T{&Y!9{OT
zxvAxkb>9KWNj;*w8`m{20;aWUIQ`X?!%exJjjVNZ5@9b^O#Q{py6PiUz9tKf)`R<l
z7sTBxkd;!z3@C@ye)wbscbtq-%Ovx+qd3sP=y>*m_{ZD@m2lD56iARW#F;9wYb|Sn
z4i$aa`eQYvz*{mUSKM$pgFDFXtA4@)S1%PW?DxoT`L*Shq?>7YIdvtZz81z;hm)B-
zsJ&36?vjt`mFHaP-5%59ErEW>9#!nANc>V*z8E7R1c9b;6bn&$)2<trsN1Af8`8RH
zkih2|m;C@qK8cAS#w>m0{%XZ@Uj~pvj%R<wpj1+qR->Z{c8STTE<IN+e{-uNK>7ki
zytKUI@~uQlquf{3<`^`i?+BBYk#tm{F&GO0bZ}GyHOn7^WC^?c)WSZ1<jiEp7@&Sq
zS!*uWaopyhL`iPEG&`lD@}-aqlnKR}VyKb0gOFHTf{EG5M<W-9^_z>O2cfD}-%lrJ
zq<GiYKOE0?AF+8bNiiQ9j?mzoq@=S%D}@6Gt~AaElWqvmTPZXxqT(%IzAM!cdd!|-
z8yY{#t#umkyp4H^1K$?p>)p<<&7mbyfX!xY7kwkzw|THgF(v!|9sqPmc3J*L-r2jx
z%A}Zi4f}4=><h_kP^!QVCo6kduqW$_dT~ZNDrNbynXhG-8SI2H|IGKlM<l?4)Ev)O
zz?RN|pK!As)fU@`6(fa#gEbD+esWgbwX(JLi}|=nb>AE!CJAxv5@kl*t(G163>BDA
z43+s!o4FG~pEjFM5=mHjay2V|Ej6+mV@5!&6`x<LQ8pv!brzO+ZQye8$Y{UjA9KNh
z{zctJxFdBQx)CA;MN42gFU&#_rP;A8P}AIJ!o%`~c^2SqTZAU3Wc;+p0P{Wl_1bvt
z4l!X|3H`3`!BV@VkjXiHJrZb}h;l2v{UqRUXS2kp>~ch-WJ(~gze0}5o2Iogp;{<~
z;LU2}ESveUsN!h;D9sCyx<%x_q@c);vI+_3S!Be)YGJNbst-VrJm1Tce#~0H8I1TC
zD~HVf$#<Bs$s_Cg-~F*&Jv^Up>)y#aNSLi}aLC~ko#M2S1ZF#QVgLSpHOy<f>woP*
z+LAz$r?hrFO0pa%#T8XzeH{`a<#@V;t)TC^3q(&~Rl2>Wg@qb+Ai96q_JF2-pxhB-
zPSnLoai2zzgR#_a!nu;YrRDpq^@9hj5zdpE_yN|nJP<Af8ejB0TD?Wp-EG=fv|$BY
zTX&B|4$P~4mCp#+$IzYm!_6Jp=~L8}r(P0KN_Z0|k2G3Yo!t_{*^lV*u68ekPUgfX
z1SuGneoa@s{_9q%$pGwzxOrYTK}oek$rLV6dlFKTFXeTc@47pV(DpQJB=lk6@j}E~
z3BF~*`K81CdwjTaeNkfXvh1|)6}Ir%XN_0quWN(9=Xy)qO=kq^IG|AjC@p^bwwn>t
zL5w>PL$SZj(f$%tW*_SAc8{lmJi$dyNxIbdKRDK!vB|sF0m|1EzglqIs@*pcr6PMe
z*fq?ncGPT&DfX^v0(caH-cvr8oJ)K{mg|M$wM##oo&X#77zK#N03~t=iPJN_(V>wX
znv^1AJ~2o9&fhr6Ag`A!L(=r7yr`(qdhvedt8R$(4vI>#rhR>bN>#kJTK!Uyc+?Q>
zZ(-LHl>LCeq7vmcXLASdzcL1hBb8k*yhfkr%Vvgu=VAL9$0_WIAS6uE`QNCJJ4ws<
zCWnWJ2I6Wc)x&4;nuK|sl^t498i?jnV^Vx093lYGeAPz00M}EuYBgy0G<z;<RbL^A
zBLQ~2>y+#LU@BudlS?|Mq@yML3tBtlkyBY1oO*7H`9pH>kD_`j+8xWZ_c0<MjbYYM
z$TPN1Kk1}2%b}1bj5J2pw2M^@mhPtjA^Ty0b&B9nx$5g{`J<$K<*_Z(ck<WeRUg|7
zPZHh~nDe>6hz@X}NHo&3uZmIlh~Ri1lucyLy1uT5)Ky5%oX%dj<1jh>#VCQ8&pYnq
zWfkCDNH8l^2G%+dyCASfY?1tL?~(?oGISfwsAKBM2CYp?KTn^-4l$TN(Z5t!`!Z5J
z3y=8O$h33-aCerg)fLb8)K&SCW#P3pW@c2J0yATsvDdt*RyiOc!W5e<ZgvxE7-!|b
zsPhb%EbC}(u@BLFw$m39I<W^p{3#?vvX`$o0~EJk#R#-#Sl?I5giiOy?plQR5ydf+
zagVj5=P{GxTCEU&F3Eu7b<Hr@tn?li0I*W}*|)KL4oi;(o4)kp`nB;<j%<YoKjZWM
z`ir=RM4kd`<K*!-H5Vs6M3Eu8PF7}241*ddrl_<FY^6GFI99hUQ9PA*OuL&Lic}C|
z!};%{WDc2MmoeDc&(tZ%HE-9ELG?@lcoHXmU-j|hFuS_QA?x`8*6^!`RF|v6?FGqU
z{esxT(7ka-twVN*MI>Al-fQ`ozXR~0QfgS@Fl@`MZsiLTHD=Tqv-F|uISLv6f3^Q=
z->*MnH+_FW_$A@%wvj*EN1V<j5_2JNRJt-O>P7U50JlJ2fsa4JvjbrJ>cUpw(?@Jf
z0+P(&cM{Nr(#de@W`Xt;GSYPTgcf;zFex0_^3}vhBh5&O8zcc~6I7Uw1Z}av0rf(u
z!4|f$5?^|5l%{sYOeNWa4DNv2=QZengUQo96Xaz&QM8X@8nV7_tV6BzV(ovZWDSr(
zVuSS<XsPWrgGPtidHMQg>ubyGOn^_VTZoH#Q=jK@#NcYFeb$uuJ`@ACKh$dl^>y4=
zG(jwT<vu7DK^7gtg!@D!A+RC{6-*)l5nrZ`zv!{lX$uddn{a@GY4MZX!{d9U<H$<H
zE0j=keijAvUnP*;fqG>IrUpFJLRgr~uefG+!-4oE(~6it@^|}y&*SitAvq&!k-gSr
z_U-VSpOjI0G#`0C9ij<IRc;3xypt&kI(w19W@LC_+VQKxO|U0#NXIx);0`IGv3V&s
zk(wou%6Vazi<>PC6f(2q{MwbT^MIBM*zJS$BD6O$yAPEPzd>PL5;;Tg5rImNypW}N
z+2O_$#c-GVAToK_`=|@M>3&C6(*>BKMBG;-B+DIF{6kyH(WHdEcis>o7bZ}-TZI@$
zW;yB8h-ogOC5Ld5Dd93&b^#4#D4p3PlG7K2|D|h#e(&|SuJsLRd|#PCApfKD-DL2d
zG$ix`n^z+?V^fYeQedd3R3ppjhveFu#94b&HBJyH%pV$g^?uiiS_fygr82pt_cHpq
za}#jjkH11fv#FRwLhf8)5^A%xnHp)_C2KG-)^jVJw(-bC>m=FSjkk=_4bnjE33{4A
zMnx5Yt+q09c$VaidCvZ%(}J_%H#c{pq$q(ztEh1C3b;IC&93A+<eJ5i6!g6Yn>_`@
zPkCE=!(hs&9cL`nRClZ?*+tah1m;ph7SQl5h7Ciurygg%=~78%T>U*ScnC1lZFztT
zK%9*_Tkl?TM&?V$**na}b0umI+hbu1B&{S$F$2m<ArG6JdMg0pR?{BH^QZ06TJcm?
zmHqFW^N7P~6D^$=Q@A*|()YH>beZQms!TB%0ag<S7J(;e{{tO#eKPuGSCto{Sk{qS
z2OIaQIGybBG^fqrXj38x^pAHnB6h*An@5DukUGZ`BjPK)&4#->hesA5bAz?>ey0|p
zRQbzWY80IZEHSc<KLKA<j~{Z|{W2f7KTKj(%$63=Y_?1*`rVqgN0{AVVGa8Rgr=lG
zJ-7Ma6?#A5SZNojaClDI?|u^wm!cd@=}TgMx++pH>IORuDi2<m>f>rpP{U{CaxSof
zEhY<NT^)3o7wvLe&hcBt7(ZQ;W4slYRuR!Xc}gr|No>ApFd2%b2^)EUd%uOHpw5sT
z=ky|DGS90<ttI47I9+VEe`oqg+`kPxIYUSMt}mD&A0#j_aHQ1gQLrgr)Mgfyh?o7C
zDKmZ_3Z0{`F)83?ns|VGl2F5hrF3-B{0E*5*6RV!HSb?3t1}BQ9WQMX-TFXoeP<l@
z0#Fy7Pa~zKJR<>(jZm<`^_#u6kI2sZY9vuL-~GzHb+y-`*6Qm`90t*f@}=C$uczH?
ze_QrOba5!nhIN~Q0t54fe67bgofR0_(Lhd9bMd|9$<Q({lVU6v6m$yj)z7Z7JBX>7
z5-9{f9Rh2YO*6=@0mEkJ7wbX#&lRS6t2okbrEqTuf&h44FgfBX_5E%~ZXlK-)1+3K
znX4w6wLgl*goaO=0bZrM8Fma9unfQIih~Rq#Y1c|)NcZecvOcAD%>q;f`!yTu9cqS
z0vgWmY@isW!k<b@?{V2*^{A-9;)v5-sFRW23}0rm0~RPJTPOZkRCe9r+wEQZmz}`y
zWpc+;BwyfcP?@uFurmOJ(iZnU1Uv`fZ4ev@$DZU{GjXrJNIw#n37>{wR5%H4v$-Cs
zKMc}+JSM#)A;%AC?J>I<O1j~nZNN#{bwQYV7XVx}i*4pe&!k$yobP?mKDQ6^hw-%+
zdA2ozyezVQrp2S{p)y70z~v>UU!Mu^Jg5`p2v>p<+K6cHik6RjVMtz9Ma!PS5o0eY
zXM46xe@OWu7X%8q_uO@(F2es_dT5k!$pip&!|p97Fq~Sg&@jRaECOu<w&5|F1H9S7
zm*sfW8ski9_(X=*wgF$-S-=&5hXVM$-?ESjXSJP&%Qoc{sPw^~Tpg)p_9Y5X^MzY4
z7Gq-um)K55sVc_TmXtnKVB|Z88MlR)TG7gI5wb#ukCR_M|A$IeVA6A+0TZ7%rF7e%
zwFB>ciT#`Jb9&Fedxh$Xt?=?&xPc@n<-P(x@Hgoc_XM)vvIW)kI)PfW?{DanZ)nsT
zPpD;EF@~lTOqu3%fWiE2Uw@%g0Va4;Sp1&+sL#KkTarEIcYY5T;gHWi1tkvpvA1>)
zhAWFB@mSXAr>@c&<3c>sA8YJBn9Mq%$SDtE2g!9?7I4x}xrB&4+Hclu*GXqjZULI9
z>@jOpyT*Xd#SN#%<sZl^5|)~#L2>?@Tg?2gwFs`WW{-G3l+I>Q>5UZY;TbR0)J8#>
zPT+CfNmSL2Va2}G;i$9WY3KEaQ|T2pDF#f85EV4P&iJ8a$(ccg`UKpz`KCQ*;<feI
z!@4AWVJCZj2Zf&3HAlYN;!nEaCvrpGr+g_S41?$uqdY^iUrJ4_u4TX<YqtnQ3_YDH
z%yFFY&Hrx0@%;KKCe!fyD;;XMAaS$TA9l+@w&|^Sa~JB@{_XkGZ*RUwdX}RjC(C3X
zH-2PjNW1@S|0TIq_4w1@&iwago>=jacess3>~DTwQn(9`=(`7K?zwVH_FEVeu=SAJ
z>)S1veED;QEbXjwlr`Q@P!GM$Z}+)CCf8a*4mBINW_xN_&QPkM8O-tD>`z>tT^glB
zW&{0oD>~Wx>>^;oLviB?)2qL3E$#jcp%>H+N8A+b8E7w3>Yjet)z>ExFLJW?<ufKq
zEA1JT6Q5k`p2DQTSVP_^n&5ElG(}r-{5A_v4`<<@)RwR!^1bBI)@D%qYREZ>>74Rd
zXlEvkLh^y?5EgixVlOYejFY0Vv~#iWiTeX0GMyIiwv!EwauQ-ty;SRr$+fQ7omQHk
zBJ~S+{%lyu@%?=+EUb*kP<F<0jvLoM-I?%MBH-Bz#ysp-E77<9F^{TZ@=Pa#cC;$H
zV?UUI=$Fk!NPb{x!&d7-*?;8_7U*}Vah%q4i(#^B`@-<k&oEHGkuTa~W`|XD8?Bp#
zxmNwmheYtKy-xq0T=YaTz?67a=Rp>99{h`PrJ|08ie~e(2204qa`l6c($C4G`s8Ou
z8?j{Ph2?y+Pd>{z02nxB9E1U{FT5u!<!4Pjr0Qkl6YWuXohH)(LE-6a?pHgFMU~nb
zx5}Ml#VOn*rL?{MAvH85pSr5C{V6muOc*6wSCsoL5j06MkVZgS3*Ho7PUJ3iCc3Xn
zly~FhKMIH+RO%naW_iLN3Cgv%^lN$Q&6@<L*-4{2{%^5x8w;o)83=S0&ZhlVHvu^E
z$3HIqxZNtQ^a9BMz+a`N@1umXryg{kR2SID$T5an-Pug*K3EOf!9cFVe%WLw@iUBG
zrDgZ;&Ieb*T6XRQMQSQ5ns0N{zYO`(Lk!iLy;X|}tI=Pf@dV+EbR=`MDo-rkr|SOf
zBl=ltCV-kOGgj?SSZx@jSG$ZOKGC?Y;!f7q^(QALO=OLG(r#gfq^DpnCwhN&(5`E#
zhYvPXlp!?pAv;M4rW`HP+~4o4-vs20ZJ;v`<Bs$sx2a+k-5|n;Ce&y;gkY%AS&aei
zUKPLy;0KgEjP3`~6L^|U&VDI$_-vq%UuEK)y75sl`5@-KXhse#K2WTrf)>OG%b3AZ
zRl=c=qrkQx{G!-3nB)BCQx40s(L!~mOZh1_KBqugu-|1Qv%|izuHKHG+zwt27^lHk
zj)m1bEY!UCEet;eYBsd2=yYt9Uu+f^`NB!`DLVRAkK(aeZ6APl-qHI4HUKb?b>{Bq
zD8C?&TZC^aji<WxMh#rj;%*jJ5HhZs`Sg;hSufsK5bf`drdgb1dcbCfVwesa_Wp1p
z*`NA4%qr_zrBX&e=_nYN<E_%_j2UZ3kWHZ_z4qgR)}kd4hACr)ZStVYUdZu_?R11R
zw70Gfm2T1U{vm!M+!)I^rGL!HM}{t-2MG1rpLT~$5U(vehsS6^bApyoxd6b9H2=gs
zh`%e`!HyV@(;14U^4|-TVau4E`vx=}w@b`#mIIZ!Z+oe>UM2UR`Hs4(1;Ba6B=HDH
zn9%+3^6f{{rmWUF1j3KV2r3YAI%FI~roi7Q)1n4E6a{5Qrl0`H$l4`%vpB}SZ_2?j
z3vtL$ure-;0}d;|(C!~t4<s{laB=@TeUAr-WmhnvZ(bRs`1VI~klr_r$b$S0k^Er4
zG<Tf!ot(UEm>2MuG)+4EuzmW%QBnMRiI@v|k_RB|IL&|Ecl|oGB`=#@4nTn!J4`a}
zxz7RBajJ32LhT(Ez|s`FoN&y{)IHwZriaz)prK_5^`}86V78(&c6#T8Qnzw`^O-D7
z*S(Ly(bCOr_mUmxVr;pXxZFmeB*B@Cv_DU&lIzwF5qL8S?fL^Io;#uisJ|Yr_EJX&
z<mk05E2H|r#QgN;WLb;mj&%CyYrU|_SZ`b&=0o|tlnBOZ&~{>cMcN@5gA&Sn!Wj#j
z$YY@0a${<?(ctJ({WzJCj)CA9r8VT24Xl5>%}(&oCE)4TU~mfqZ3T6M3C!<<Sn7c`
zr(obGQAVM@^2p1_S)~)n<>D33_x_v#PojKAQyo79m^uN2WVBpitW7Zf4Rwy>l+U_&
z8ZHk%4-0CUO-DUZkx1@!SSw16EQx?&7Fd<?L79#UbTJ6$OScmPE@FPn%&2$}XlCks
z><P;hvKfo#-oJ@L26PObC7YHgJXYNwo{vjW7FBIrATzG37(WCs!W}R`Z`qZlcsyIV
zkK;07P@5)|L-xlp!^s0olPNit)n5VNleCQGck-BHPdVp!*3SVWgVSjI@WbCv3+k4h
z8;21TBQVZkor3)s_(fEXW!RtAqn9ZW5?SBjd_*rb=1d_-f$~-tNoxDF`aaIzN!IKL
zf0BUX?F<3N@GDWTByqBscVb`SCoJVU9L|+PC?u%DVL~Xd{(K<!6-71B-?{n2*t9n`
zf=0ZtM^?|vE>pxgF>f^48w+fa4{~1{vd2N7!e3}$ku7#rNZ>AES?ihU^WJ*22Yj6O
z9Gd)}Gg=n@>AVZU5R5Em;tykyp`W%-@UrUe-@X?m521@`-99m;4**_rMdy!%DlnZV
zhgj?xgvy2EmbyZ|bXlw~_dMkOeF@##eV9*j5GEjp=RuR(1E*vH!6Xh;U|igW-Fdiw
zYs!<EN)(m@85i9DSVKwZ{dZ-rTNvYx=8xamuwc0bA_eAS+UxxE(b=3<`zXo`J1fEv
zYmjCEw0@P;#-So6t=`Cd1ZH}}?3nwOx-aDg4>~^`a`gRxzkM8d@mKth`^WC83Rc#9
z{tb6xB)rwb63>#D%IdURZWFQoQnNoH0mXTOC1|KmiYA(fh6HODYM#5qNcoIjHbw!4
z6=eA6{l)l*V4cDCK!1D>((mksTgA=JHn_V4T5JVGG1j<>HF)2F)7a&YTGq%U<(K$l
z$C?gSSrN>n!=kZP4928i2$J@p7`xKHFv5>zLfJn{g%gX1=l*D;z6tmgMb#oJzFmw7
zeJTBo&NFPnE9|H^Jl9F393@!2we9i*K?&G|Vn-0~*$p~=f)w@EcT&G}me8I!P2nKF
z1v;w_>aXp+6T`{d0ygy@h-Qnve-VLj^k7nljq~>CVCO?9ZldS*iQ>qu*+PSt<BA90
z2)fr<|1xV8`~{aiS^=~I28G-;E~^a>c!A(5fzI}C&EQ4n1MR7rgFtZS#yf_T;gN)~
zg003^9a0YpE_@($1B;?#x8Ez6&kE~i6Tg_7z50JWae{I>af!N2`x^|xd723@X#$!S
zl=;Snd9tZwGo++!J%N%(aZ_z}TysB<(&tF+;!uGZw0zC)kUtL#gMzBz^|xhRSmcqb
zKCww@ab+E7DP2f-Ngu)fky5cGDMRwOYzx+05)(A)SZn%;>tYLJ@NfN8FXDv#9>pd=
z7L@{IzOJk%YZv+JNQ7oP-}4VEpWIuCp((%M^VvmDh9!YI;1H>4YwUS@$lTU%yD>F}
zZ)4c5tEDAyO$JO_Z&?RB_<#Z?r&}G>pkM(MDbQ?2FwufG9oD4aiu>6CoZa{0dz|Jo
z5eq{OVVbtQiP_qjlx}(1{tWJ4&4f<UMKVyyV<Jn6F*BP5S|;<iVo-EV@_;TtQ2h5U
zNRt8M1%Q!T57#bmBHs_-!$NGt#VI(AHkAoGIx);Z2`m1yjJ)6BE}aT;%|&>cH@VU`
zeP#wVXs;yN&&b^6+t}GQz{}Ig{92K<$ShC54xC$JMJcLm6Z7Pqo_e)#h2#3Zqkstd
z^mhe2;4%BC+SS?uZ1S$6%%8wcg?UH)(7r6CB+dI#M{%!9)njyQ%H?R{Rjv8sQ^q@&
z8^5LqRCjHPKNzQIDeFg>Wb^^lARxGSA8rZFl&OSl>kRbzdH_y`&^1TaG$}G@ZueM+
zC;CTAt>3JiR>!n16=dr?XXuO_R2formJiCf--L*pn9sPtWw!m?4m>K!YP=8Vt@7oK
z^ef%?=bL{nTxN!d*M10&%4w_(EsW2Ql8_p3p#$XihX!YaQmwLg$ufHd5=M+Ih4Zwa
zBH;BuIru!QqbMZc^_Zh9QTP?<>DqrrTH1!wJ|}o1>+rf``A{1r>UcV&*HSb=$i@q(
z+mfpc-1P2<<nRFavx6uL$aiU|!~kycKM$1woJ)};2uTQr?W#+<avibfZzA|!cu%Ix
zxzKzBCY4{rfG>_t+w3G9XpVWM7??l8<H7x??%c#CR-}grVW1U*go-nKn?z!4->%B0
zE?@(mNZ<{Fy<j>R0+8KSOKvEE>qR;{enQtw{Hx*v6<hi6=*Zn*1EhA?t{)f3-P6nH
z^E_m|BlG%3RV(WsTD|^R-uy!#=ugqP%TRz@Nx?|v#7donemkEc-~UY8SS{~hZ0pEC
z$NK~LB;`ovB9UKYmb9=4f{qkmd3^5Q(r;3y+D`*=k?O<}F$C)9e>4KZyXJ3KZkAaU
z{`W$=F-M{f%NYPFO}bn^%fd26aC_J;?F7*~WcV^sIZw`;{i?w5!rS3O0bGOqcQGKT
zE{>p0Ad%yUE8Td{4*3<rxH~MFY=*rr`OV&FplTj#P4xTDWRGmZsM8s^(w{AW&kVVK
zliJEPXc_~je{SE;1OqXw#_p!GDu)&nPQfLoj`#O3P6b_e8V0<Bf=s4{4-TwNmVZ!O
z3@rL_^vP;|daej}RzHhAdwt(e*QW;au`^_VQ0^^f@O}QI;jXndLHR2$YKl0y@)8Zt
zr43Gjdr@~1qY*Qs4>Ik2(NQZ`^0J(V-*4=R7p4E4Ovy05dFtG=(bGCF&L(r(BGiH<
z@HF0#F3|Ws(tiGX$7Rv_zo#vkp&iit@CLMtg3cw+qz9T6cXy-CUU}V)Ym@-+Ho>32
za35nV)`Z^y>EwXZl=`RM4--<Z2ol1W{ovMuCj%Bg<S4XICFzfE2+Oj>>67P?5`No6
zCIX8g+}tkCP-<|HXI&zXhI$3_*kC5(k%z~$-y6_uCFp(9SEr#b=y?<79k<eaH^}wv
zm9`+cG!(HJeXCHa2if(W#n%Y4v5{jAQ`@58<5wu~i&f;#C&n@}6b&h~T}&iT#n9$e
z7WJ-suA=MBf-YK_%SMj&+iaDxJgXBZQ<~wmB=W@h7>&#JBjXY$!@$)WgF_isz4gh$
zJ2;>mh=cC6$Jae&29^67KUIHpY5shvDCjH=tykZ+UKc1s*_b*ud->|G7g;AzbS#yt
z{^E5!$I!7SEgT+#`cjPy(5(AxrUL`IBTqWt0EBtAKeGb=26NsL{*AWW?sN7BP0_$&
zArcKkD6|}LV}q7RAB@S$$|S>q9*+?O3<uR_^O9HD&GLqhnu3f5VApb1Xk|SsiTjOf
z29fCR6lxDcLNpO9k$Xj&`RxChoKAd~H>cP3W|kg<1s-Z6P=ly3WV1>tnsi*V_Y^N*
zbg|-xJ1OW37IA<)y`!9>{<6aWBN6x5c(t+4@_&H%MeB^&`9ummH<?uV43oA;2n5vw
zlSFlZQyPg4%VVIyzp^{E=Nl5rS6OIjaeUG&m-t2Bd+4}Q+!S{UosJ+nH-fXV>RVcn
z?Vd0buAiF$F9ws#n+zO5KNN_}5NXTaJsHJ9xC5UgsRr~0a}lj*-$L{+1A&BqkpdQt
zV+}l~e_--n#$QQqOhVkD*yLa!EggIKuY=d}++kFLm=iR&Xr)^Y@5?{e&3>n0!pjGo
ztC##^K+3l#$slE=a~Sw?^f&h(87Ldl2#KmqcJh7?+xFK!V1hfHc~7BKT2Sj-uGHi4
zrgM))fBr&)Oc}TH;c&`9>i+g+82hJT+~ezhO+wJvgFjPAs`2~B#%S^Yz+-*G>8n^&
z$Cu0FlQyGXp!jR{_d&aHHhcJ1%L@_(4eJs@=Tw=V3a5v*iqqL2ql?oh+eJ6F@d3<E
z&{^KkB|Fgw@jTREWDr<>BEG`MJKiM3BISGKh-Z(M;=e+oy>t~H|A-I`pp_i{LYC7*
zX0i17n&O+tFRCZN!A@;|980#4W%R@uui9V8HZlslHEDI{hAXJ%ATelhSE2&7TDs=q
znF=ALobS6~A$L+nlEhnS{C@utuE{<zoi5u8B?1vzSbE*kO$)&ay)Y{gANrBRobwrY
z<moN#awrbQ{a%7z#OvH<AZnQ?==luxW&SiS=-xi2kSXf^P&sEhslw3Pe%Vt9OiLv<
zcZUM@ls3dal_r+5j_yi)=RX5qbKiuOj#k4ZCydV1$!EQ&5Hp&xStDGFD&b<i?Q>pc
zvVf9QGSHtEIkKKHFclq4^x2=spaghsiii|DXD@m_yib}z^SC34P=^8YEO8frCu%$7
z)EBVkX@1*0{aP>F>XSitZye;kwo-w3_psteE1?(=-8u3#C_yOyGq%3!xj$U06`KWk
z_5i{+C{9yGwvmz$8r&Eu$K!inbz9*f;PjiYG`?p6;P!c_0f-orrREU^D($-#D@Py_
z1Gmw2PtzBJu1pWVRF@PYHCQDiF{2d{(kCYlI~E;TNWsFZd_vJhW6y<q2eR|M@s#`$
zV8FrzW#x0Ty{H`A!OnE+@XjwkHgP0(;`Z?czU6IwcIHtIK+k$F@zCcQFSmghH>%|p
zzc_L|p#9JtilOTCljo_bY~KiY|HvohzNm4}SaMQ}tWL#z{0I8)Bxaov%CuShYo{|F
zAb|UKI@{%l&2IbIj@`6PAR(GB4zc0qESGMj*_o<e0h1Y2n&+g&dkW~{2eLq0!@VQ(
zZy@+(dKgeyG)Dhu|LYK>4`=KP>*iK}u-?LLlRb`wMHom00Va(Y&u!a|%stHlOwNP$
zfP%eK&_Grcp<agbmITxODh8_vAY=BJ+@>CjtKk=%U$wG=i9S$CD<vpTR;;_nBG{k}
zn5Dm@Q)S`-X<@QnVaS0-+2?MYED9xf!V=yiZaiO8EQ*zL8RW{%`X`hD)?S=hB|02S
zPQw3V>8!)5T%WHGD$*fR25{&uNoheqLXd8d?vidn8WAO=yF{c@DUniAx;9<X-Sy7?
z{@#Djb?}0{pXa`3)~xjz7>)-Vgg-45<%By^t9k2j(U#Id*OoqF6=}ne-}vfctGt$A
zzw=dX-9abB_+ak}BUpm^os}T;(d~JU`hlgv1n15DwAvSWDvN7dp%Cp^F&E>Fc&cV(
zc5*k@XV&}T?Dog9e*!OJc$}XNj*JZ%4bG9g-<%eZG7QRAw+4_!dt0#3VvYO5G3J(t
z2l{`KUe1sx+JQZ#clQCF)2As2-s0?xgbnyUEcjRQ>&UzGpQtgel0g$lixsb!>kquT
z51LyiEi7zv^Sqg9g)4drC@mmR)c94ZP_vP-8u9|FcXQO5c_^FzYnVaL)q165?F9Dm
z3nhg4*ryGEG~5I#0U9ECjOFF5Vs)WG)9d7hW36DkpY;BUHSB8_`ezqtK_7Dv;?&eR
z5X*H5ymdk$>{{S7hSuP@N~tS_Zo|So+Oz%XKbpxe+KQ}N-eh?yEt>;dJYWW3LW%!a
z<-G9Yj}RBu>v%B48VLAbVXB5!n=fRpgd6a{e9ZhT|Nc!s*3CW0coagweOXQ?7NVh^
z&SrELH8`81Ua>V-#x=?M>})IG#4Sb409wPO&%m=MeyZPORs!{ldLbu;fuFGdR(o7l
zF=Q>IzT$?|?p$i0&pK&LWzaLyntSWZ=thE2^0!%DA9|>ZZgkskXCAXZj0~?tA&7oA
zZPf`<i*+p|k#J2zC<`e7dyUxsPHjoV_iNFszlecUXPMuW_FcE)6MdGg0_;Bx>gdS!
z>u(vE;T+h80b`7&Gs<zG>eF52?B0|&YAJW)<oDLF`mP%*J!Uy^0B&1uBvR}Gk5BGD
zD`<dU{~Ipe6uK9yI}C%8MRYz35I~!Kum8-Vf-AYBce?LG>)1UBu-%*%=)30Q*u-aZ
zB!Tl-;mn;sa+Ff4coZcFBe%S6qI8Z!ZvC6hw5OC>)sGDL{okxz4i_23foV?VTV^{4
zi!4cZ2@SvHlmpX2hwJO-rP9rLK9s_)oldjFHRN><ITCXnn;svoX+?r_qM-0?B6#w8
zx^ezv6@JpkIan@57jiwci){j|=yP_XlYx|3*LU8FTm9fgnW(s+*0gU9yFfhgrUwaI
zZj4X;{*v3lPDl<;#x}}NL^(OxKoQl0$fG@646I(pk0FcZ&5SKruN^I)?|jQUFTul(
zUZ@(XyW=qL7X=ohxw@SvQ~3JP7IYVn$;s`qv$G-IFHt<urYO4iJ7ype^36llFV&wU
z=I4k<k##TSs2-@$Ips!%;#X>JT7diFxpcef1FmjUiHr}`|Linb3`En1=MdEl;L7L*
zvp95q&*3?+jF0$R_QK#t)8adSPSn1QrnO1}5{h>3S@C_1M-)mkn#YRa2^Fi(&|k`5
zt+!iF)gjzVIYwnF!RM~<7;y_8(>-jAdC=2gF!T<Ns@Sov_t2{^^`}MgR(j`#airMK
zuq)-{jGzT&-rvSM_HU}r4k)1(T*L_=yUPd(w<OVQT&Q?nI^E>-*9IdGXgiNb5Wqg{
zChZXW1Wz2Ro_i48qv(6!d;S*F{5sGsLicS6!tYu1CmYY>NtWZB%3D&=&3$9|FElfC
zUt4;_?X5`p{u{2?R}vW$%Ncdrp1c`K;DbwM?{<p6vq(6b4~nDt)ai5W-Wtju9s4gC
z7-Kll>FnMNfY4~jYyo%F`o5%w-P86gZuiAeDjG=<!J01$a2Ti#XYf94(-}WJs=|y8
z`*6`<iHbeR!!DCIc*bZK%VOWCU$G}3SzwpvgK1N;nI-`O2pmkyf4XZC4JQ$?2^80}
zrN=aRkt^8=c5E>CPyMf538MuDcu94y86+^grbbOizHOu6q|}Uf?{zF{2Za9WZJ}?$
znuCX*eDX-w8eCLuJJ-oX@{iZAM-!Z3|IPKJNot1Y?%)wQ>a*FBe|ccC{=P6e{6#Aa
zP!p<IrZaVDb{!SI3OaR>&^p$++m0%2o$}9I_vqC`gT8m4m^|lHTQ`ZTL6hc3)z`pu
z`*j_0_-tDMkb!*Bk*`g588Mu>r|2R3|GqP<mWYH`)}&2w&4-5jMvhUECgd=}ZC0nR
zzdt?efN-LyUm<R`ISVh&NQBI=J!N`$R=_rCGvqjsL12#y(YGw#`MKloN?=HqDNUmg
zl*zY^S8$W8-d!PyhS~0wNpRWl!~x!HRvrI1@N{(8VQaEzehj|9#i-qem9%u9fT-hF
zHJI`Cv=EiHP~^f}oC)a>Ca)4ugrNNEjbLLiA)Vu-YO8AQ8rxCDsvY<w;X)m2qvruq
z=Tsj%&o|}Eq60%#cgpV-9<5w8+uz&Tp7ka@sT$!o%m<~nj8TlS4~5I>g?v7aRnnY|
zAXr<F_=Xu1eiZX4`_9CsmxDp$kZyWi-1>meR*W>E^~rIt#X=^D8dJ(mg2C(~2mgJw
z2Xbmv5nvv8b3LgFpXWUX^p}NdA9*V=Av-ZJI{K@7ijo#HPi0{`^gp$XO=CanEgS7-
z9Q`S^_I&ZaJBIzM&#2RM)y;wS;v_f-AcmzPQ9YEHkIymn$amUx8Jb6S?JN=qxv;~`
zBAEQkM+5tcnTNisFMHgcTr8$!t`=m!G5oC10i*&ddFe$5QH*-LS(HrIoD9;iIZCpw
z9<vLr??-BgsoIwTsmMYXZi9TUzp~Fw6&@sk(@Qx|0Z=Ka7wVliaQ=P1c2CvrK5;%e
zj-+IM`#=G=`TS<K3Z<s_czOq1WgY#qc;X!OI<u*lwpdF`o5z@O_QF)Kt`JUco`?4i
z)>()BrPV*)>CaTu40e4N6Y=kG5p6KBK3cDug(u_J`1Q*NQ_WTpQoHLLYn*=K+&^lI
zJ3#9lFS~lYS}K;M5F3sJ37jq3Emv{L(n9jaRna1;(bykBZ3$`y(x8!}e?(OKZyWN$
zRI0Td8V4w{sYI4u^EJIk4qNYP84@hD)hMELim7zs%Y61J44U-n4`PJy^}laz{<5<m
zg?s!DI7ilrvHUbqS(V4J7G2JvcRa7}qWmdX;QU<b#xT6jk<Cd}4^%{C&(nLY-Ig`v
z{(#?IXvg=KHfPIr=tdG$)ZiwUwfG``jhOg5Iw&{5`xh<(&>^~)ve+j47x+Nvd9yy)
zTIUbqk;!`5XfrDYoA>Bpm4i+?TRn%<`5;Znb_&wlDzI7q3v4EJSAT=yXolp^LC3|0
z6&GY0z`f7`We5b_o=<<}3&)y?)F=dJ1fAkVX_@DV)GwOskF?=X8F4$SO?jAfPIlqg
zuvFGUKW(OoA!UJ!kpC<D&^3*cT6;1)OSLwl>Nng2$T4<IUK)^mUq;kYk{1(p^Jfqw
zMVx#PQVMuT6EqaSB2heg8RLD7jd71ySK%J~zyAbntHVBMT<s<@aLk4on2g@+x<~!!
zi>~i?L936P`?zbWztR3kX>oGOZ+8;kzu9kUf+%(f20Fv=!yj2U5jWN^sVxeqV}hrT
z#{J3)g>>Mo*)MqC3t!4}5C|5|5no>6b9oN1gPnoilu+7KtqPvk+}{Rn;;D~&yAMf`
zzavz;On#zD9AZd(Se>)%XSXJ;@ndRK-h~^NAOg>qST{6aig4QN7`pSPQhld=?_3f7
zRCBruEST13E^dtXl7JRJdle;@N1{QD9_i#<+nFYrRFt#pv61rcI^%L@+hc4J<fD5>
z?|A8`pU4SpH&er<NbgcyAJ#*69>(6P5nkkOa9Opw(9x)wZ516XVKl<Etz+Yw$i%Ma
ze=eV{3Q;7du{UR<2ONhX@W-`51T?6%soS{ybF6Vxs}=Ab)tw-R=d_v?iGdxhu>XBW
z^bPUdv-<Xp+CW~Jh*;9J#jEWtA^-UeO4c9(<2C=zE3xZi@;tBhsR~selidwyw2U!c
zQQ_X<_K5<QAu5;m&TMyhEK#4Hr^4RdMKS>^Ze$0Zl4o(%kS>ssC<$;HKiRKIa9CPb
zrTiHY;$SrVmF((#Bm1#bFF<&OHEHlH^1o;fHo+T8Qji%vye@nXYQcc=zhcL3XP&_i
zm$c;O6u0^S-&%wr<Tk_3bCefYPRFGN?ZP#pMBB>5XvVYz$KYp+*L`Vr>Vh69salio
z9sWxrby_t2Y9sU$be!OC%4eI9PiK})s$ZX5{r=NblnT1m&(jG2a^nFA(E|L<p!bnY
z*J3L5vuFCR3Nj^oE9a?HI!@oi8h+f8Zo-cmbmvj9AoEUVZlA<9F}2A?ar5Pe+4BNh
zy}ywK7Au_x3I02e+RnlH;bCj3MJu(xd}<VRB5ARmU0TpuHDA0~VFhuzK1xn5y=yWl
zibt$9@MZkK(<M*%wP^!Mx}hv9G`t)B7|Zx%s6p=BZQq>q?X?i2L$+0bU|8%3u;MAy
zCzOp5a9-vbnN%#O+4|-YaKG@`r^}}c%#N&D&Wtnekg|rUTK2Oe<8i*C7tM;*YmlvV
zi|X$DZcWFaA_g}nw2@iGdAV?f+3F(_&^s+U<*~?#i?dxb=;ExNe1f3aKg8FHLW!)f
zI~BWB-?s6Ie-=i%kDVN2K98YgPWY2SpV2qIHOnW4A@KqSzG@fB^3Q^9m7bRiFA;oW
zL)-^zGc`I9^pnAfbBU*(pPJ72<RCn@;ByET+RtBX28Kqivz(i1MsG+H0Zr0eq~?8b
zr$c`4fLAmPM`I$x{Lr?b^j!4v=8{QsY=8XhVv_4VP3V%VZNWqSPBaOOM{{vH#q+bc
zGEA>b?&!K1%I#J17V5{y6s+v;<SLr}v3w7ZL6cSc0`aI%w&Q-=^(J!;G;oH)CG)9%
z|J4sX!(KS?a=C|1Y@74z477hir^G4u?Br3{9&gdGH8YKcV8=Oe6OqL5NZr%JpUwpt
z2o@(FMh#B-(93t7m{Tv1W=T+xk4q^@I0P&%8^}*jeVci<m#6xfKB9=V@*WqgnQ7k7
zQ+amccH`bU@q@vse;=FF0p5@I5>(|jN3YMvM<lX7Itn_*LqPEI@9U4CgR)jyz<*2R
zFS!jq)0;!nu{_&1;FIUvrbv@bkzQUPY5~s^GL%SJBFpqAY=-O87`o#08PDtWf><tJ
zLH5nmu;v575B3j&0Zm}|?-u+OZ;tw~g&Qy7E&m0$6tLYjMk=2^uZIg2Kcm?W^%bLo
z8`6dEth54WvQyWqDQTu06hZPjx%tVQ@v4nm5*XlGZeJWaNEFR}8F4Ny3fk=W_o?!)
z0*TbF(nCf+_bEkf@h(c*@@Bx6H%Ho@p?a(N?Z+FIU#Kk}UT41U`=|F@zmB(Q%LNl4
z%?_8m(q-UI5C}caXqj28#lVdEdp8g>T9moegwn~~hYj&~9V<8eZ7Ix@;7fW5cn=ky
zTSEh*#*Qam4Gm>sOp>}rsfv{N9EM&W43q}EOeQ=}&)ti0`~|@m*`w<lhKLX4v`rR)
zl#NNMVg@K_`{xVIJ&#9@Q(XIFBfu8|N=?4)kSwC%<{`l+i7+OKsZxyX4+Tr;w4yPz
z!P?rU$9ES3=?LYJ->w50NkVcbWU56?`29{24>U9213TWiMuHCd4E^OgSGahcy{n2S
zP8K#n)-Q4=w8yN6c!C55R1?yB5}#o%6(#Cc{Wj5BkH%c!x-1GOKO~Orb7;eC;ph=J
zqt4I3Etb2G0OkRd+TEXUyO$RVg*+|7BYZ%7HB_1f2(dVpbo40c-JSDceRFv4ZhC2l
zLCP@&e;x{J`XCcKx&G{j>K>FkJgr}gX55LzN%?p731TfciuVxhDNYB*f!J^o$Q{pN
z<KX2S{8a~kehc`IE6n+fhJJ|x6zedB_ks}IzsMM$>!!Ktmr+K0z)kFYsD?$GT(znC
z0NFirXS3o?*owTU*I3RNVh|M6*(<3v&DjIH&O7BS3dV1O9Ty27F}Vy&zV(nz@LkvK
z?;7pKu@RMsFdKBei!*JlaW21KcVARUaw(6M;uinKkwjQSOq+_h^ZKA<#3*^|$3ora
zFY`%y5NY7Da>O<;&}svlP-a0<FXS7o#{3_&KYDLA)l%diPyKI^v@NdP8%fjwkJjS<
z#e$P~a_QI;NV!c+r-+o8S%Y3ftb*v<)BT%!O@ySuALSOd1>%tjo6dy|tz-+qM@6jd
z=6)}B%O|+Y?b_egCCoK;I+Ty@_3-=qukW|Q%fV8cnr(LV#L0vf9K#PsSSfZcg7!~M
z2v_bkKI-)4vvH}fWBUilC$p{vky$&MCvsAiKk{N|rvMl5dS(CWPYnje3{MQmV@SRh
zRn|S=PLikmFcEM$-&Ufel)Sb=X)|pj6Y%WjATNUJ4%uoN<c!_^o2O2E?)HuR#5Z;_
z&!xJ$ni|z)`0I~ZWt1t#jQXZ-v{#kJ>nv|!Kni`4-=61lQy#Fc7J#{NQ|#XT{Zz!i
z*`WkVBUN_=J={hANzK(iNoRZ1aj}vBUBVtC&w5q->D!D~GICtK`da}IP{`x2=fO}q
z5SpCUV-EPI>fBDugt2>T2^HywyGXrDGY&kH-#wfD-`B#|x6@uS^6sgTScuI<!9tY}
ziYIF?p-U5&Zwgq-GdNhkKq*=XIqJg5lgq#7Nn}6`6Y%EEUr&)9yqx%x5o>{aI94Z!
zLqZrY{^L=xO66hmB2h6vAA6d(Tnvmqu<6J509kNtMQKv)aoTeq3=oVe7Cgt^95(~H
ztuS0#1I<~3IF!?FkHY6l5~yoo4``+q-gt7UhEypyMH+4*luIUP;oKm9^m3CfXjrOF
zQ0;=In<l?5?~ip2XoB#;A?i+;2Rf0>9mwgI&A&KZS^AVjnoiT;Xqef#{pWS^c8VR+
zHx3CquZhw3C!YOE*E61{f?)EZS3vdITZ!y{KeSfqDA8HhT);NddK}$!Y^aHnW2IL4
zJbQT`Os0SVo^QpkX}>R*licp<Xv|kBQ%jSUMS&bJRdq@{bl3f9b8^CTL&A0kv~ly!
zfQ@^ET%y%1d|;~@-sxW$?6ts2tP7W8D~QQ6HI(jip#`KT+|=LFr$uHQkeMyzrcAN7
z5>%6Ie%E;xPV8GM@Q6)90^;j3R_?iEt;>*;Ir~&7Bvp96D5pYZ!V$<T&YBcU!9^zp
zH8|Fq%rQ%_+|q?|==0{YoO^Ix_Dnv#`3s6I(f3f$fXaM^Fl)9OpO@nCKHcKEPU&l8
zTTkc^znNM?MG|RmsCdu)%nzYOqYybkia9^$906m7fg8P)7s=nvmzD`uc%)(*xme+=
zkz8!E@fGI4^O3XhaPsLr<@O)5K4XG5GhZ>D8mXi`a!-vuz#X={qe{9nKhKT7nQ$oo
zNsiElyuVE=jyBcFsey{aNbKjxd~)W?2du6aa$#%Wu*vWBX3usiCS*;0ov@%Q^t95<
zbSu@Lj6d@M@-GDxn+J^K(n3p&Cu2n3Z_VO>LZHesH?Ik0u`^8KdjeG57g*B(AIDWE
z<9ltTDI%pWFWNrE@DU2~O3#z_anf&=n4hhyk~-us{_~t;cps5>AV69ovfjE#>hL73
z`I<-edY4<KdinB*0WOndaK^xbj=7G2wY~k0^rp+OOV5fcNBdY>0T@Wvf6!wn@RWWy
z%J|_x6(7(!NEH=gT*#3G6os+5D7UKY^nm`r?g(``%1Wu|^Qq`-H$zo2?A7k{f7RtB
z(~lts_uDF+sr_6|Qzm7=H|6hlb>fbmiIA(AZe)t9Qj@a4(z6F<&zw#ziU3+fX%qPg
zgV8&8LasT(80e5NoOs<Eha(u9|IV~Of=_CKRP^X(t3h0?O3>|59n&bTi>D{geq_?<
z_}@$;M@NA?yEneI|91PuF<NS_G%;}_P1A<7N1wLiFDO`i2@Nj~SbfCN{P@MI>Rui=
zBU^rpMlop$D+Lz6fsZ|c`yM}NUtrXwtB;G3rh%i69onZkFpfHXF4l?1_IH0(bryiX
z*nQ`8$~WRu76r1K9Nl9D;6W#3aT?zKPd7WN>zCjDx9;bCk6o7)e_#N!b{sfeUC9$7
zWQayl(6*hHHXG}9z(pEf(kF;~az0;5O3-MPFPrjYvJn(5Ewf~8sx5ZBF&c4Aka*sI
z|L@-4wfe!m1N?uZH4lemsucIAx4_|6X|3Vjfot2*pdN};0FbY$$&wB|`}iL~SFhRW
z9u#P``|&~Vhm;$MQ<UByrRsk9i6vvL{*WpwExz`5e~f|O{B8VA1k=qnJ*NQQ_a*_F
zATnA+8pR(Ik-Zm=1pnOT&+m8OTYmnxSX~__z`CaNM|u$)Omh%Dqwkq*3B+`BYJTdv
zHt$YcsD<tO+EJjEq9NaIaeuhjTJrPzj-{I45+2WqRK@Ja*r!|q=Nlgy8y{{}3kg$4
zj8V+iyl2apAx`;`g9(Bzq_(f8Jm(`DN|9QIWk6eeoAcgo)`#iMI>cwc>lUUN!-<N-
z3QuK9IOB35kpVEgkaFi*=ktA<(cUYDKlb|2q#iz~-5aWwk7_fbB1&<sB6^gQtxnIt
zA(rdNtij5pNsUjgtCs!z9*f+>5)y)3+9YiF<LrIYA|D}fG1OAIgb$4rT=M09CUMlf
zjgJ|=m1(b@GVDK28=exHzgYK8toUJE3&fUS`n%MXlmxfT-qN8@;5!qny(BX}ZIQ!_
zyA*IXfBEb4WLoCim63I1Ldg!}m<u7;tfZh$hOh@Z|A!Z!1d|^|g}_vDpuu3xV=B3T
zLqx7dL9~6bcosweQ#lKX4Ev>#cK?ZHzj2GvfVU3JA2J9@uCJ=dK@pKZY$WPRNt3*Z
zdJQq1y_r{!PYmbVq*hJ;SVavbeTi<Nqzsd*)i(G+XoHJbDt2=I$4Zp#wOO;P)5%Tr
zl{dmHMX{}@B`t=e#cb%u?*1=ou!u)z1+YW?$!)*w#>#?c{(HE%&g1lwV&UL@;gF58
zA19c}7XYsXri<2x?Qp<;NI@@vwl6p%|B@VD<OSRgE|<wst>u1oaryZbEo|*G`6p1V
zy9Sck*c)G=n3UH%F}e1bs%yTPc6<*U^LAAfY#o1HRD|E|L$I_9uy6DS;I^)fXA^2a
z>1c=pGs#e~UA4f52Be2l+zc+akAQY@T+gJ7<UHiI+?29DIb_it`*2b6s$TAw`<&=0
z8dx0l)_vY6NDOVzLOkWb32afoVhvO&L}D?BiugZQf3P6ja;pNzKLqBA&%Rt*(K-^!
znNiKJ|HY{-_Rbz&Z%2t7C70bNlf^QNTsG=%_I)s!g9`AL)35uhAR~|ZekOp1Qw7`r
z%*MnRnd15^E6sb-)+s#!E%Q<+z|1?6rx37XquWYRd%gO##lygmq~JGeIwBk*u)i}u
z?h2#KP+lg0U7daB-D0ICN5)KinoNo6l-nqOcb7h>$Z&z~{~AewslelL&eT7WI{8|P
zVqpA*?)cpG>&yC%WwCO{=~E-WTqQQNjDzp0r<Ku1n*O<WIzpECzbOl+yB?0-PH<>4
z^XrPdqeGFSzFtR!4zpY2YZ0oPC;4N{V&%&zDz&bgzQX5all5$oXIv+Jax*2ku?Z!#
z3kx34>LwBhC92R6zE_fD!e-cfuF{nUyGI#s<eV!H!LEI;mPH;J1B6!*X0R=THy>h4
z&mAX^b-N(ZqB?BILkFfi<k_AtLIR0CGG`6)woVQVm<&LX7zG1UBuubk{mV@lT-txZ
zBmt!bL47ePyX>IeJrc8bDXj*c;ta7b)*>WghN1SEO%=K?0fr}@8!}F+pv={Ou8^4h
zH8K<m$Rj4h!ShEUXf2n#p^2gmt_`lss=!q3cw_?I_2qox**(b{D}sbBATk6(pTGFT
zp^dqCwut!VdwaQOzmyu2?%R2hN9~ng>4M=OyHB)M1jRcv)dI~X)Skr^ghYhUj6DMC
z1**d=U5}vy3<o~3d0(*vO8i3d7O<2h7gt_)Z?N5*1gJ={qe+M~UXJ_i>Lfq-0k<SF
z;|@&=w8zFL$y8Ne3_FO#|MQhXRa@hMNEh_4d2vzd$=)Pm0;%D>?3Wmt1qCs*QtNAL
zhg)NXZuP2oCX$AA9z#tx6&yq*FywzV<wgQT9ps_Y?M**aCcT&8mIftGctHTAz>#mK
zSEp}6bD`srrKU)F%#W?P>gOCm#SgE=v{okhRnXc#f4^|qr40f-*z<glWjjE~dHLSh
z&h@y_5BZJL@QlSXclsDl8@vo0%*fw&f^6a<i1)w7+uqs!AS9)wW`t>uVg$7;va10;
zRCp~xF6{$_AuceF>9KQ|LsN!C4(JpcjeUAFT$3gRVkn4J|G{+IztAzg$*UXL{OH;V
zMAfs+!A)k#69^t(E^ikhy;4@A_V*=X@7ERZAXT}>^@dGqBPzU_9uqS)NP=j%ZZi54
z&75kOyRIa!1mbOCan(|pTc22>|6%p<F=j3x<Mo=3Qek}4+5PC0Nhn&c^Q^FwpSVrS
z+ApSjahqmr$K2XuKm690(+-jnuyBo3)^`?*0M+!1$)khKi&M1uh56Y3Y-)6hTkQX0
z!5m-vNglU$<)we^)>i#n+CN-S|4bekb^x9BrrBrS9l!$|;0b`pi5DXgp-lgD!Si=}
zz*pF<@5D{7I6BG=u6GF*zg4ajBt-(dpMQ~x9U|5ksIG5ol^r`Kdn%I)SqSm(G)pUK
zrT{nmBlr!gS%|~8ss^=Od7}zY<N`Yg?yJWpHeL-IU$hiWq7WMQTac8JFUx6>8Vapu
zkQT6unVvIPse-XcX39?<Zq00%tAU0Rw-7HWTA<u}8r;0q{3(|kWFs!;n(~f$9PWgj
zi*<+nl`gpirLU$9Z@fGETEin9zyanWY_m9~;^32v&Y1<Nh(P#(Ph3|wIb&El78V0P
zXv7c_*ND)vXM@E1p|PafU4-21xi1?g&nc4PS}ZKW0BW86ise8yrs{LFm5ChRW2O4*
zZjZHd=Nd;y3WE~&<m5x&!uaR_PmwAX%=NJ~9Xk9F6x)J6yagO$a@9OfK!W>*@QT3d
zWk7cDfckh>RDciFAF`^foa#IPC?K5Q`>MXAt-o>-;eiUajzSj6#ydAZ2-F7Zo{ldp
z{G-Q5#7Rl~c+5R3qNE`NPR;qUD#KJ2SHKXvK`>;);WQvLMCytC_HlH?<Vh^+YB#Z1
znCYauN?qIe+%hoYOR#>AA%)$Un(AvY@>4YBDK|I7dF)5B0Kfa!i^Iz@n_5pXZu_Qr
zG_A~tqROg#uBwjHO0+PYjA_(TfBb$rlX!Mo>Rlq^IHVArCaoTWQ*Cb#JNsFJCm0=f
zY@EZOWj>fUOZl*R{l$Tm)c#DzcPCHo-J5%vZa{2v@V?Q^9Kawc5&E?d)rq1dFz|se
z*aXK5U<?2X;duBeTSNWo=1YZrX3?t`N2m|!sj7s&`*1|MLw5|##@zKNzs&nGVna*$
z7}`eL=`!1_kder=q;eMy_>heOXjZj(#PLa)`D+5lZ^#*`(8+azx$B4e^^T+`d_?ID
z@0(q{0s983EYj1z)SZSK>Z{pgg_bIk@0;3>{<CcD*a_cB0!*P~{A+ja1cUy*k)RQk
z;BX(24uV%l|J1YIV{TaoTQa4r(XpWwuq4A}aIT-xqqJ;fZ7U8Q%~;wED%GvD;fF7|
z#EX?HUl^45-AdP;vDDl0#t#01q6aX2*p7iz1OQm~z#-#j#v2{km@m5RA&na$3x2Ll
z(2KVI63!<Y1l9YC3I8K7=a4;dwW%oPHblng;j~9i6cQ8G`XO&ucg;oqBCBB{2^d8b
zV+j>@XX0PuSx!~!#X%GIn|caki4IT$P14y~umddV5RbbSw;i=9zKB^pi?73>wX|52
z@kxKU0_T(($IVeW$mpL0D93XE4x*{Ozq=oGasnQv!cn(#P5%v2f>`rntF#y~mw*V@
zKc|ODWIUeSa(N4(>eeeIg1><-&YCgG%icR)AoLP0VZ)ba$26C>C7-@IzLsfMOyG;^
z=Jsw!Wv)s51rwj9dJ#R~)q*GR*_E{v<cc9&8wtoY$Tk;D1uN>*c_Siv88-aXu19SU
zkI`G=b}pR0+qr&vE%fu7=?cFBGlXt(l*i6YS*9~Sag*PZcs<sj(%|zJdhGr7;&>FS
zJjSD&ncDSnJwoGBDEDS+H99I^U)kcN|Bx&amABMx=7E?y4Ir?AvgUYGE;P2&v;=^q
z47|sh5Xet)7%lo%Vx~N^^6$6kOKaQ=dk(3hVI<ZIB%|DQk^jGOW#PI8uyIh?<`fh~
zK{pTo)~hahQm_dH5%6q2p{(rbu+<q~E!xs=*n2-&Q@P-C1f}>wZtM7+#^5uwpw}8-
z4H2Ur;1L9Fo@IlH4UAsSl&p_yV6cfG(Z;|F{|P4iDCu|b@KFfw{X9#OL74K`dDzhV
z5NhjXoh{W#RY2NV;lX4Cr1^c7Zyfit_2Es#hpJ3S12<bcE%z*SeRac&`H7tRqjC8D
zCx3UVGru;{)s4wL6W<%J-3ZX%;QFexm_hL!1Rv+QC30BT@VngkWbv$IrMgRQNRZA-
zt5QI{Ye~uFXD~?mpDVH9Wn1_+;jPzrK%MT2#2VD}G7sjt5@W5(r&)?o-Kqb8cn#6H
zjPucROsbWv#bzzVrBtk?vR)RZyj2ub{4w88TvX-YkN>UfDga|wOB*Ik-yeIC<?A5P
zf`ijIvsB)t3d_!p4O~d_FkEduf8sm0War{q9e!D-m`wI4Gfxe<P{c4*$a@0y^gNUk
zs1kSb-iJ?1T~wvl0te(4MajhqH1&wraJ2wXDgwS77$)QZ@Q0|XPq7I!0zdbi@ox_H
z{{E3At(u+j<yCeXT<q^4Ek>|fB5Jf#uDjG98wD+oud^tj1gFCh6+P=XT>Df22!sNj
z5^S~2W`FK3j^5JvFO4(_XWh};?w4GomK!r;mL~?I;&^npW(Gem@K#BH+HuzyQ66$)
z`x~rR3-SLBfWlyTRSu2+Eo1l8Ta4P1+qVkiVI%;f9)5L>%OSig5*@i<>AX=!+BWK_
zl*QehtK%iQ)#+ky4yy-Hsf|>i2{c%l2FAmu4H0qBhf==@>qFO-2S<aG{|2(5c*!e$
zWOq68>Tu)4V8HV?bKh1X!AS>QP*^Ro|3WtR);U|G{NBgTL7dLKD<)Oe6aG@9Rtsdv
zNKMaH*fP*ruxetIelJS8tm~R6#a;jBD?@xcJ36pn0BP?2V?sp$W|h@+$wA*s_V6EJ
zvC4e{vWTKpbTH33njZYFi@OcYI8mwC2Ut@9j@x-~yPTp28yn{;oz#qc(JA>4337A4
zBmq9Pa#Ij5Y!8ZJ(TvrCA4i(#q6j3Yg7UGp6%#(n(&aOjc^I+CDxkgT)Jc{(w(7|&
zIcn1W%1F0&gl}|_Wh1a%-vblNMeP@ZDP$^nwtig+J@UX>00%pFdXdDY>A%aKcVOiE
z?sP&2MwK(_upSt}gAq*<a217G%!UzYQFMpx9`Mj2{0Uc`A5c-TjlIy}cA6&oFS<o7
zTaCkRP0~^zv%%X%qatRnbAPrE`;hEA2{Z}N^c)X4Q1QLC0>K}OBLaPLHXPhT<bZN3
z<ud~e@@7S@!6E_?6w(Asofz6cN{7>RXTPf%mi+7b6q*zn07ZnrFg>;TIEY)5LbWN#
zwSk#Dlt1b{%j#zW_8AZ9BRB4a^dtQu^r8Lx8Qx(Nx6&H-QLx7-^5r*})k<H5J&|$_
zcoXG}g5pIl`}~<&=b52jSD`$9`J*aA@G{cMMvP+gx?l<h#>rX^^Lf~Mr;t6B|Li6d
zFsk0Qf@I|v3c`ueWCH3v^3?#aPctW8<KA(B^u>c4_vw2JPCd5fb{5}W$?gOZc!h6$
zj9WMjMz<Yz`v5zN=NCzlnBaCSo!8*GA(mgwYkf;>O%=2+gMAez<=tH}<+eUK<&LTX
zMEbfq8<p(+2YoPP!ky}SbM>=oXGKNsx2{1F{r$yXtFBemj`crim6<HI#hh-8D?GOs
zdJ`uUH^n^BsqSC}KSzRKCbH;9x=G&;9YepSRkJ2gRFmGpTFC`6nCf?1KjdlAq&K1B
zN);)Ur*d%Yivm{l_iICR@=XMlQcvFG)BKl?&EI)zwg#*FvZwuC#z*@Y+Lw2GP@h<E
z<~yuT;%q64P57q10^8x-Tw(oVLD$fOu?&I-M`i@Euy(~fNdW^d3LJm?kv|#2!dEy{
zdHAbYNb_a4Q~*`ZG9u-*w!#fE(U}$DsU>EO)hmS#yP`<pe=3xJ2Dfovr>kt_=oDvZ
z!FwMZ1o0xPuZVX9x0D>jZry4H+3*wa*0}wy-Kyf`z#_Kax=7jVAvwY~1SUEpt^8Z!
zz)|EZ$P};2+ld}09Nkc#i_r1}0+z1)=zhkjk`2L4*zYfvUA$?JNrTLdP2Y|c6Ys@X
z`*&p@$smye(gV)OG^Am4hd)H?pm=QqUnm`lEDebz(L)(T4P8^ZUde}c+Ne707prax
znc(8HGQI>(bAd)^lCb+;2V<IO3q+9!cZ{o0<fkYI2hSl6zt_;TS+M+7jyc6me;6jr
z(x@KTwG9)^jg*G^MdDF|+rX|d<K~y|hah+vXMf_{0dfN9bWr2G_|pYRGCp=iGDG)!
z89Ez_6(|SjTmC!;28R|{)PcT#x_$!c5|T2S;2?q=4wvn|cw^cv4wF_UU15Icm*M6B
zi_T%>)|eO?a9dzlORP$BrGAY+b|SelW;D}{lJ@sm%xj{L6#v0uG@P}8x5zwrGBpO5
z6$?yy+eW51>bzFbYSZ+a9>m4Lgf;@$NkHWM8qe{fdLJ#eu{}Wz>3Yiz7JBMTS0b09
zb4w<QK-CfEp!gMrg(z+f_}TwH`(OA~$Ec`~nl-;A0H@Ps-pR_=ec(A^ax3V;y$=j|
z$DMK8eP0bGM&QaKC&a~eyl@+w_L5c$tA})a*Fdu2@>J_-{~<u|H|uJyU%WODtj50m
zWB9D3g-Wt~MvYDPEl6OeSuOdUzg?AqLAK#j<+wfkg}Q$3JGbzFNqd*QSYUoNGw})(
z7lxv44iGYL{kaS+kPP4&NtCt~5{!yoG3HvbS>_CDFv}DUKLktq$CWrSl#6N7+i!6I
zba(oJJKwWVLLLG(UHLi1?djE$gx@aDAa>z5gQp>sDjk!{ao0@+E(`vbHVLLTf2L3b
zHXBo4qwO}6lL~kp&+rqAr`&pqB<TVi<*GWiqQ<R9i_I!l=w>Z+pZZ8X{7<8RjH!L)
zW8Z0t7}^fFxJ}c*;dr1yo@=-&2xjkp^+$@TSRkLkOq+OB^Q%hX-Z5s|&{Q<oQsBh^
z$~Lfk4vVHF3VR_T2pEXJ1-d4;1Xy8@@w~CWA%&h5>L<Jx1N**6DDTLi(JXaV7p3^-
zsQKnYB<?w&#R!V}au}XOEcNfbT*f!nV}`6>=Q4`Vc^AvGRa_iAlRuqbG~U9-k*1={
zXYVhNQ7GVTUX}Th+pz)isYtadQ)(AXkRcLbd@(AW5=)?_krq796C$uSe}R>p+I0Cs
z&~?q|>&2~(hZ(>Y=mg%L-c{l`oAX^0*tzmyihN*D<pmaN^;FStvJX_N@IdCD2hTu>
z(%R&JUKR`x!ub_$8kDFQht3+^@sa&w+)ZpwBsqrU!pW-cZ6?ypD7D=sslDYWn12xT
zy1!hP$<6=%UEb)6n}X6AdPdYd=q#~EDOR>=YD?`^|M;Ye_ZnLhv(OgCu}1L(_V*DW
zwi3e1zBJsjBHCpKxd`l(F-3y}9eva3bF|->kRT=C3crF&87N;!Y7%e<kGkWV!NjYY
zcf|9xCh59x-(Bm!coKWOmmlMstDY4aO!~MuaCVzsjzuo2W*kKFXH-~kFu{zgvpq!#
z!n`-r?<vYTES<VkZVmlpz1n%M*H%U=3jvSA&`ys={MP_X7CUsfjDzRq!BuUhdl>CK
zw76KF2o^2Z7YNak(AJlt`ClJ#TTF5M!4`1cqsV8UxnFR}3ovk4wV~Ybk*<Y`OZw$)
zUwv6Bs+V^xhsqwq<WuE{y5GQ8p`AqapP&ye%wbo-6##nH-tN8`G>bTZ2%)<nm&sGH
zxs7VdDSR;!@}_EFY;5Qn0Nfqba!8WYoiu;gECX-RO}pC;0FW@wo@sx-(@)O#ec_D4
z$6)qUl_xndtQjd}!tSOomLu->y{@;2IBYS0f{aD)=SRg?@q93m`k%R&q2+}(7M0iS
z^zw+;@bw|(CH|co-}+<;3}jPsW*&<SIKoL6N)|awl|`@|lp<+@*xdPjn-O=OROQy+
zAwZUX%gkeqRw4tPwq65&1-3PQF>o;JpQJh@uidL(ycr@gpWUCmhy(FeU?T(Osesd7
z9L|%eqAYt4=H6L-ZyVp3|Aew!tn}MY-w&O>fY|SUA9V>MBDHFi)a-?`h95Entix=g
z<|=Ugzv?2QtT#!QklakJm$7hq-MTAKv;`}OXk~x>7T+nR1sKg+$kcuUr22Y(zAP}e
zhDvUp98=uXwD(R9DjW@NrG37m3z6-#=Ho}f!PBf%T%+LPk5=rJFVhy`VSU0b=y7Pv
zuu}T;@yT?UK@|9EBqI>WoUi4vl0t+aEM}ufOO6DpywmIjdPs%Of$Zqhw3eZ6{9p-4
zs>lOvO2VOD2UNnRBj_KcX)ti`khcU}gCK5&MQA{gN<m8Z7^sOOhrC>T;#9Jbk6xVF
zg&01yjrn~DiTQW&BH<*}YGME@W+ii|H}@!&_FZCn8f82!PO5&s9rG~+b+p`J9{_=z
z-z-`^cqMM0Q|Dwmfh$OA<HPQ*w(T>TVZzP7mPD1i<IF-_xyt{?2SsXU32sWC(PF|}
zkDS?Oy%gg|hk#O`Rw;na?>8&vFdA`NA+h;(7?a(@jed8i2F**ym7sW8P~nCHchIiw
z;a%6Fo8HXmp#9BiP0K$ai-F+ctVBh<^->a%T0-xE&>Al$B=m(O;r!m3r(yBP1<QEd
zqOh*Jp^x+4N~KYZsTJwQ(w77l3v&=^QV$E?#5{CpgFslBX=}L$uPJxFlFagxv!`!X
z(gk($G_Yi?1B$MiI3AX5#>p2F^z%N<I-)7~y93-;0Iv(NlcvL|AYacwXUX!Es!OdU
znP#yoZGyM-IW&l#CsbjP{ONB_6%?2X`J%c8GNrk!)tMDmaKs#ns5k|1!0-!h)x{&x
zw-7k-Tw9NVFanaY+aeb9_b&zylj)DXkD}x&kUzHn9PlmaB_K3zpB*Hs3K+T}K|!o&
z<zUl<?JbdDkGjx9@G?aTU*)?(SKu?C5I8TM%I^!y<Z#vvGLBz%-6%hk2Q3y($K-fj
zMZaWDAss>i%c{7_#Sx--+1N$(3ZJvb&(o2lxNOa?b3K`&VTd-j0f>e@U9`b5JSnts
z=HfmX?D2c1r9}j6plWH!=g)a!XtQ3uU?j#mJgj<Qkr@gAUg^7DiOCPMq49?TSbFV{
zi=Vq4BF#@L%zv_0$~FYvDnxo?);M3E6<9?f_t0H*n){EPC)Dr%t&>x)AHxaPoWL@6
z+Xo)*^P~OXx~)wK4^twJJ`RO@t?xQ@XLo<ON?(KoNiU&D1kv)gj9lvl;?=Att841k
zNl6vDtVGvdl{lDM?*SzY=~Dg87*BiWrKmJ&EeBPWFsf-+5*v*SU$jE?^MVY(I!b_q
z2EQa^bxu{gwZmS?n4Rf7SF9Pla`(KrQ&j=KUcHPt9ZMr}TX38uC%b(%JEYU5LM?KC
z2Ugh*Vcu1m*l%BjImqBp4gGDwF)9d45ycmPCj<QQ$edKaLF4k7jROeaj}p3>*dW}S
z>oR}<>wCrz^+=Usy#SAJ7XLO#cYhgm6O5t##O)oDY@d*j9#2|4dxxcdKjyW--&?OF
z6h!7qii`eEm5_L<GLmg*!Zt%vdA`Rc-RY4T(*L2$16%W1CE1u=$v(;{?(d26!d!w2
zWarRWL92{<KjZ%6hb^!~bKLXvE{tY77Na-<sEpbgEU@QR3swdPMkAClD$M{S#|0B*
zlQb3<n0&-=xNE#I?><#k1@^;pr_IaTHQP<U!H}z5II^iD613vlyxk$<yC#avkhU!~
z`g)mPYmTAU@>4QSIQsP1Pac*WZm0>hmz)7w+67#B4Pbp?h4rgZ*=!WY=!9>+0r?CY
z+FBfbp1;d2drzB9`FB(yHuDHZ3}`eT4KQneUuLM?T6-l<7@;bD3~*fm_AMn?wBfhb
z?nYjtQ)$*u7ff3&xw!s!`(S4ZUrWR5<kKU5HS4G<=aJVsyo|A%&gpgk6k6F27Jd!i
zuO~)iI`@zqo_mTc`ijE34|AVy%C<=&_j@<uv$dxO)tF+%l(G72F>oW7R^E*v#qg99
z1E-TuMG(KgXk^gFZO~*UnsF_f9M`M6R%782UXwyou|DsLYrz1E#)H9}7R~$S{!t<N
zE65M)N#aA%{l@0eyH6?Sc7lWL?1KZJNxaZUQ4O1*DAE7U2|?g|*<wPgCqgZx_`{v1
z^)`?DAnSQ`qj~S12ZpRp$dfm0rS5mT$geN{)1FMbW9@FM9L{z_-z1KlC_#>*`?G!m
z^nyq7w{8(48CS??ER^UGBvtQaKdFuHhnmgCNPy1@meSG45~A&!Mrt0S80-1G(WD?u
zNWE$fZd?O46aR8*?K-Ezdt8;C^OGbeITKZf_n~9;DDwZwVd%_nJN^qr*Wv%$n=~*v
zH8j;<1;!WH8hajjp#ZC;8!5-XO!?4=Q~ku_NAEn<G@f8e_#8-M9hLu9rSLVdX>c~r
zovGiASogXUXMimjN6NcMqLsQTh*ofJy>9DZ7R*9xr;tC{Ab|y~IjYX83S}A0Hc<Rw
z3o)o<ng05!j!d=e9n>7m&m;?oA(^^dK0j_BeVydA7w1Nrd8x5x&o-2x@}UA)+H#93
zzFJ0H`*{$-9OOdrc_oBpc5&F2od-9*w|%aI*Kwp}eJh_IFZzYhy-paWN*`Od#TgYZ
zT;enbgS$nJ?_sn~@w5MQ=2z-u41V}Y_6R=6W)OT4)+DaiTLiF@p>ioR2wEwCusnn?
z^M?8i*y|7vi@capvZ;vBl~7@&&S@#RlMovtE6G7mjwK)S>%m7zDad_az;s7HpCj%0
zt^EPw3h(9eWLey4gEH;fpLlXF^;v1?8)_DLK(Z|zhLElXbsql8XbdFa!kuq;`USZr
zJ8vI8kQpNR_~UA9g!=;D3B~s}u!RFqN_WHuZjm6KJ~>Dk{WC4&FbHTe*f<-LJ=&~<
ziG^7JeRb)yu-MXFi68=!I-zUMyk{Ftky9`ju?Yq9Bxf_}^rYI$qYwV)UETGT(_DfO
zhPmzdqV6fre35ht!cJjujcf}aH~QP%zm<}>&@jbi`#pv6RX%i43KWAeqqyB2)ms3a
zgt`vd)q#WNU~*C9KPhaHwD-bOWS5>o?d#E7S99SP)&TuPl(M}nuO)(|OT~DV3s3e|
zA@(LDV^06#?*9)6!esakt_>0T>(%)Ut(D(#w*J3l`b8%E_wg9SR*u6XuCphfkp=3;
zNc{LZU}>K-Pmz9RaP{S5H9@vKT)zCi>+g~@0IN>xJ%)Fc_2#Dki!1Chp&;m&%6E}k
zLObE_{HppCZQW1eg<U`$dOa}SU97c(VR4j?@+4bzv2Q`CuyNYr1H9CR1KF3x_n4@*
z1}t;cJ;n{=g!WIQj7FTI2(mR{1vKW@xO?63w+>gQM9p?q(OF@aE7dLA5ZBWKVuRAK
z#@gHCY}9LhLH~N-fPXDD<A*cc1+X{4kC#{*1i~-s<(#}M#QD%fFr}+lfyM*~FQNpF
z-C5Z*@CVJfM^Au<8MaKFT0IM;#O0ut9GICOV2%$UP|-(hDbX+0!N82N89a~E;1WOx
zaI&d$mw+7MR$j=h!=jVdhgpXiUqR$hbLyv$dMES8XYD_MH;pH_>lm*D4DnJw;{qan
z=>iRligC=Au$H2(P21~@X|MRJ7aplO7CyQBzk07X7;5Y@QxvBV96YSQ{>vkG(FYKe
zh68I0<xmUG&csIu7&UwMK(GKXxJ(il|3?@kk7X-@HBOZ~r)<X*^?#g6cl!Ma-_xeu
zf2c^)MqEKQBOkFq69g;pW<HEnYEry-<Q(_B;iVq%YyfMsS0CE2cij17{gta}oE#JF
z`NDdcj=`^Mls8p>-l~jgNDQIn3hX#Q#|frm>F&E`mMZn8*<B*^p%7Dpj=Sera%|U&
za(6A$mo5kbsl)yah&38_goEO7-vi=FYkaD|?mD7xAx^;XuMF_C-->YByBX}Z53zZr
z_FNpV`*bo<ux`f(JQ*1Hs?yglj0Ii{xsCJb1js)<VNcvsMMe`~(1YBXvJk97udWDy
z6NsFezyp(B?@{SIlF8B4oyP~FkK?u|bw}?D45PF;ERm$CTy9=&-v?xm&CNj#Bz03%
zh+lsJYt`!!V2SsU2<!Rjo%QkSg7MAfP#&QEzH(dFrF4dR8qI&g0@<|V(54GQU$NwU
zn-E1qCcUZ-Zat&gTNDIpo6MhI8gNyq+<!=sMG*5Zq-ZhUp*p_FvNQ=bAZrnX0S`ZW
zqlmjI-<$>I;R$?8>3~Klbn_Orxxij+DTrQ-HyIP475B!v7@X8L!4if~kgd9Ic07=Q
zHh@A}s^}O9p|N$|`tO{-#?_7K%Qytpu2nwgU{llqkH6Ew+#%(4$xYBp2gp8j;WZH1
z<DGr<MZIBV6(rJNBuWd;q@lcj(82tUmxUx->gox6bj<BE<Y^gRWjk$lsmF%%YegQ_
z^Eq!NB?xVtdfoSBKD#B&Vsw9KrEU8K{?GmOh3q>QnFe5riD1jm9jW(XLOkB5KmLy@
zh~o8aBJ}-67hlc8!{tyuO^0SqA>ZXs3kP~WP{a)VFw%yjI@Q>08@h}>;5#tc#ilT|
z>RiQGt^j&Lzj)Bg>)B7Wz^I4(p%mfi&9v>jsB6GxWP=4RbXVW>-|vQf39Io<Fc9)N
zHXt^>wTTG&%kyh`ZD6(L^Nvc^NEM-<Uw}n=^D$pAna$kgUd?c+2G0cP1LEQZaK(YE
z^zWcb3`9DSXr}!IN!V(2n}BXud14o_c0!IF&=em2X9|p{L+Rw!W#(?tMEA1^0z0Nj
z9ARaN#C3y)_S0IETToVZTug&u3n#TIX`imRP=6u>g5`vb(?uFK$O#J21|Y2)D}eq6
zt_ncvQJqkD3rA~6=?^{+Uu$qZ|2TMi1qjw0EH#NS48?tVu+;c!cdLJ^H&p05GcFUB
zC8n%0#aQv!a{Q)JZd8n(U^UT09qind{}v=RXl1iKvJxLBSF?1|O1^wiF`*O=&?8uf
zj?{mVB!u+`{h16O>u5hdekHYixs9Mbn943&_mII*HU>li%-)z=I;ZPSBKCWS%uB>E
zQ+e$1YMAOotx>fZ*b&;-m(><gU~W+dSX11^ka&VI8r#3&uUJ4}F#i09R(n;oi1%BC
zSZ^S2DAMrgD>U*!^-e2IHZpipw#x`UsI-MoJs=BsYgPB+R7QWUVvrzLe=6JGB-IIZ
z^zrTU2SWweDSQpp22e5*g%@h<i$L*_1lria1Pgv55E;g+ELZQuu-P9Q7(Tc!makdL
zJBd1}JcXapG(^pp_Y4?$$m~eqI~qdClY2JE0~zBO+dXte?7Y$7n%(WwJ!Wr6Gl30m
zeHlA{yc*7~0?S4&Co@Z=J?+PL5x|ptfWk~N-+ulo^gntW>u2pM;R#aEalj5yv-XwP
zJ-L6tD=cs>Vunj7si3ssGZ<%h$;26R6#S)W=eI)HN#*#HU(vZm$w$TAfe2!Em#x8J
zt8o7qu?G2wxYYkJ*^C@!{sB)c3-#4e+0az5|1*dS15qBRPbx5Rg_O2(mAGPsMU-p+
zEZhcY7GzuYPTL=qOnbH<G0{MWbZ%xqMcgB-f5k}d+1RfC`GQg*T~V7hzT?ryRHYtS
z7df6QpXspoa^CE}V_tvgm^Ru}5()^mJ16F@hXroLfA<h0ola)I|Lc&Ot#Nt^j04ir
zI6V~)2$LmFOqI>LC;5D@FouN-;67$l9*@{!L8sRpxJ*?|plm7~b1F$E_(kRjw9XOJ
zUQ`Y8B6$e`RHsioPb<t`X>u68(*CaC*A-tz4+#y-5B0md`^%GQ3AU4-^H;F>sKdIN
z-i?7n{O29SV~zc&>9Xg=`#Gz&rCu8|Ma34SA~j(n``yI-t#cGdR@%)96}2I^1Al34
zc!YovBNB8U$Ok*?k)JRujVrW!XPg}27bZjM;p9WF+(2eLa=De6mQO;(O|^Tvk+Pm0
znP1-~dn5naOD|O>W{tKPlO4K_8TsM1?ScVXdg+(%r?pVLcNzmCU`L?xGp`-;nw~Si
z%%Y;V*n!!7zQNs_n?0LJ+!p5Pp8dM(YgXBH&%^S#FZ}*Ktro6$gSFe4uX<`RhGsSK
zluiou?ul&(*E?S`>OScn#Oc9yYXA#s?x5C<4rcR@HyciwJ3ongH`YbID_L-WLvTy7
zXlq#;k686JBh_hI)LurnNN*Dp#<!3k0cKP8<tE={(xEKxYY^>v5@2{Ztp4*(Fh)sW
z%wrIrhz*-fX>;A-4?N4<4&v;@b>aQ!<KP=#DX$iHaO5ocY=vgjL2B!0>f_U(x!RiE
zG^dsu3*l(RfSXU^n}S4)CJ4*PFQVpATOl)P!eY$MosFRbi4BhEgenzfmO32oDaP|V
z*!(-$gTl_=>vmVL7Gca`66eqmUfmb<>g2mRe311^L5>qc{JUZ-eaPe&2LpVPe6Lp;
zDsKbdqN@|r5s}3I$Ht9s^}y5Pl3F=6J?W%mw<|m75_fm(KXVzyW&R#(PS5$0mfF_z
zQY_qPE<d(Ayd&y=M{23OT=xT&RAs)?94^FiJ1zEcKNcaCP}0N6%agcw8zs(|ec{qS
z;ilLwGl?+gFk6jLB~{=X`L#A*g*3WE3TpF<q9ByFO&{(amt^Rf6%0278edi);f&e}
zy+d3+xsjo{&89#$inkbY`dJe-_xXt5l3{z^LdG+v4;^RjU!!bJGW1C)l(g$wF<s-*
ziaJfSc1S37>`+afR-wIBtoEy1{Boi7LG9K(zPrKbk`M5))=__wl*TAFOWt~0Pqaj$
zg}BSssA;g^`N07tlIlKYS=09da{I@6YQ_4Ojf=v5dfPV|E*@zFZzBmdYQK?O6qBOj
zr99f8URlvuGya?>;KW3<;coCV?`LOb>3d~%H*+n)BK)R}Ib}O4#!qDg5pC5fBEN4e
z3E<Ict6=R$!N@l@7#ka?<ehS8{U%PwMkb*%STXm=8%6ykUaG=d7FM$p34I^DXnc(2
zi7z_z_5Po6@lgo=n$Jbh2M4z|i};X?3G+9HQ95qY-}*4kT7Qzf-r71ss9}Eh@1R=n
z=MXHikpH;rQM4z}!d~_;;{7;zpKHFv)JP|d`PLNcoFgutPpwc{*?aLp=MIBp7~7*u
z_AFl(%|qKAWs}t}uBzyTC@${hZCNjemKD|X=q;J9g&fJ9{U+XhLsVE=@DfY)WK3dn
z`~6c(Iy!W7D*S>eE*_9l?Yb2r{C*JfJPTf4$%))qk3x>m%bfP*{tj`X92iFBzuJ~P
zj=Eo&bLXck%Gm643s$4o_q6fjc<&k!wU`k8%q^haS6~`5dR)3a%8U|_EO%tGkpF)3
zZGcF!aSJ87(^3z)sVS+Ne8osE3*K#OjUr_^;Te+`v|&np5^`Q8Rn}Oo1n?IGq;HAr
zE=f_JM;S&ID9d8_Q$&j{j{cmOA=>ziQ>qcf5#1@Cq<OI5;QZ4JWY07eth+y@bZ9q&
z4{xJY4?2|B=b%2X{ArJ3<~k(hHJWMsh_fvORTV$(>pKEbM)ljTF#e0mqa_cD3Qa^j
z%l4OWrKYekcX<%49*~(C`{V2WT#Sk`dbyU$$ScI1TY5&%j-M$lUv@4zav!I<h&~l5
zXiWbg_)D?$ZTNiWDudU9gphe|zB%*X*f?7wk5e?RU*++CDC;aI6JDllNU<mlh(!yG
zW-E4C7v>c?(WEJr9YSSPer4Epci5+pd({t%{_9ahj7nvZwtZH^qmdM|<n=C2uEF5y
zB+<vV3a>59Ft};bu<i#`#z`pYl6wiRbMg*<8vAzh-h%6pnT`zozt0_58@JKd`!@K|
z(K{0YmghbN#oQns_c-0y_A0;<hKHfu{FF4iecVC95zl<ytgL>)B7AOgr8~TWPY_8J
zj`IHYb$U?1bYr*vs>)xJD$0GotJ#`f{j$Y=C*J*@Yckk98=x#^@8|0GPnr2Q#c6fZ
zfBQD39I`2qD3ANl@oxWDMq!Wr`RFwx0#zsa)+EZnEk=BC`mEn~o;}1yphuI2CDbqE
zCP-|`@X_UzHGjyrR*=?cS_w;f<T$gl8hwoIaW6J7^xoe8<LN7as(ik`FWueU4N8Y}
zBi-F8AR*mdlA^$uknV2jE(s;2y9Md)|1$4yUT2(f#+iHXv-|8h=M(3wu7D?GS9$-<
zjfZ(HtYq;^3~2e5`?cycz1(*l8t{$lzlZ<wSg{<@FK3`de*}(5#uN)z-*2j13roCL
zgQ)P?yE#rnl!zC*R|d|%ZGRvUb@HX8BYcqe`MuRofq)=;wQ7?V2^rWTup(l1YY-99
zl7*3V<#y&Cilul513^Ysz{E<99eRs>D#T2bkTpO<&5XokE5OM5Mw-o5(n%MsF;Ec0
zFb+lR3k$B>&%dKLEVw}q?U|;QIakW$O6%&X{Hz%&eX*07XG<zDq6)(eI0TX$KEG4u
zHXkYDl!!GoN@owKx9x=(c_L2?2>y$%s$r#L*+uVG(ZY1VQ~N{G^Q_TBDCl|U>y5r<
z?Uju`e`ZsdSSf%!+c*AM&jSJlJ+BM8uvYbUL-pCiTI5G>V(w2#V6%fWLIqlHc_qpR
zuRr|h-^dRQE*rUDA;?w)Q4}50Rir`Y?Ck94ikbCi0ym`)zMKGU;8HD9<ZprkEvw_Q
zf5bO}K4*@7tln&fkmMSt7}lWmIJjDQt^wLFgY*rhJZcvs7^+HINK7}+;n6O4TvlAq
zO(gbB+2ziBAo1|m9?5dM{kXM!%1TVe`0v^^F(nNl=zEb}VaN7W;TEhZSzp0$QrB1Y
zx+0WYp=JJt{ru6IVFhz}UXa#81u^nOD8G9(`@1nM8?mw&GSCToR@crIWI>Xc6nv;x
ztF0B>Q&BnS{FY%<y^Yd69oeBpFjmQbqzbfhy=$_)?a}XOyv|c*+O3%XcD0u>y;!a9
zl~RV}8*oIx2KJlHg2M|6O3%8a`ZL3S+WvL&Tsw)G22rka6r0-v?B6@CPcx(SPLh<q
z9_Fg>*3<^z*caCFX5x8cuW+Xn5(_?)CT#eRWPgf}eIITxls&@6#o)jI1%LG+{wOf^
zj#eg|N@LDq{LBFIMb2AdZ&9vEd(O0T7DfRzfV(X#%Q{SswI}FvAh-atYj)u0_0yc`
z9@k;$CI+M^;Xt$W7AB}x)M;2ok2pq?u;hE`B)mbqCbbahea@|~djb`#a~(%d1YW-C
zqaQ-mu|W{|k+4X{t-|~9@RM#6<mY7<eFO+q7-<f+nUnJ4C$PWV&~@mPf0_Qa5OR5r
z+Bii2C}M{9xQSR>{49rdN1$`_wtMD8u(Pbo;Ht$O`5n}*Wbq$)3!3pH^^_Gji2V<(
z$Dehz6_MEtSAO|sqy9zEP(vZ6fBfE&%gi_){IVdBFpd#bjXqloxh*d^?82^y%qV9;
z%EH4<VQjSD2;mak(<Jmd<)EdtF&kU1<oA5aH9TC>C#L>42HeS8knTEe9o0Zx&O&F(
zZf4%uK;<~*zzZrt#-1pQ%Aq}u{pkUez&C{YTX0?Fp<vbGVdnlOjKauYpu{Ar;&avE
zetI75P7c^wUoT-|jqsq$VSs!I*XZh-O7lO|ywFYA99%*CCm2A|hK`U|(k%Z885?;K
zrWUH$zTM0k=73A6$5o!m)#Jv{?ZaRFV>?jC5PPKG7&}{fpAv?={5psc*lL@*<q{HX
z!hCbIM=ijH0NbA(cSUkviTY_($KP!M!^NF7Jv=!HIZ|q%2$Vb$ZTN3^?s0l<@+TZm
zk%hdN*GQs+#{15WM0-aRyQZ#q?@e0$56=BnAk-uOp^AHW8OQM~`t5ULK5E(u!uctZ
zF5*WvjrITRtjwt@del>~Dpd_zi42|aT~1ej$rty_l^a9Altp?VGl%gA4h;>>ryuse
z4ViS)fE6D-?7_-`=<|KLW$i)Q5l0LzEEQ?};SdHMp{I8YazimBcQG396JAWLRkFAT
zYg?#dVcEkwp`OTrPe*>^668}EjE@?Q*=OEDUafi-bGal_Q^k1VOwM<^IEF1(ghore
zA!X6#aXg+=A!LZ{52z6J($qP!JB-e>$L#j%CkV>d<4$MilLBzu*ZzTg$G?=;J3Lz;
zbKP;%&vr@vK;Uv7y}i0S4ktNTTt9~ZFDU(AV!6}+orIA@iD=wvef@6#8YY8;b90_+
zbx!wC5*FDf=PmBb!&z9nGWqPQ#{mYL%^NfZ4(D^9$>(oP-rjcueh_I`UCGa;h~{>0
z^Ws87x0WJyMOw!`wETs9a5=ix_})y&T<WBjP|M3#qH$zN2p{*Ijr}R|_o<yKMN=Pj
z;f?}}c%mwIXlWmMfoSPu1(Ig*1`Ihn2NPRlZ7p|LM!t8;VH)vYG0OEVko;P%N&ZXv
z99n#mV}wFrOSYb{ahd#(TJJJ$pLi#X{V35{c&5~uywzYY%ylZQXvf8)MYk-{dYLc1
zo^7CP@jF^rYh-68ze6zoW7b&@pIjG>xNV=!M|KEhuBjnb+k<VKd)UAi97t<Nt74&~
zn6w@iPGn=>7>7{-l8`6?T_H88fG36^#&QlpD0#5e-$V3KhA>APrV9h_@Ja!Getx^H
zyB=LsJ(=JI2ri_1;WzG9<d1<bC+UH+ki?PK4HJF&fgyfloWprSi2RPE7AkP&@`PoU
zqE?&{KY?#%PU<#%jd?7urw%t$_Avs5uXoL#RV%8f6&V!lvGJA5Wh<Q(-=Uk`ABo<V
zE<l{W4Bp>7?80C3xfa|%&BU^y8PgZ<>@VpNQQzLa<dhnBpm^SFh{b+yES29JFAwH<
z12^`NBdtW6(mn5L?qjW-D+0AJICvjdRVz|MS}LV$CW4q4q51bi+2-D`Fo(4&$p^nd
zTjS4R1C8s5K#j#gP==dce%x#tvWA7z!+=@|mwOQc>C4&APBC(k-LE1i>AI&JA`b^e
z;gBJ?)fS(1Ta_>CT__|fxG|1h^Kk`>^qngKtAQii7e)PET}pRPvtQRbqGWh@e@Zsh
z6KBDZt9eSO#j8sHniXT+kwQsNWpup4f_Te+x0dk7<IsoYfG-(2tSU<A%X1}~?KvA+
zMe7rz@5s8EnObR%tgk2ODtG?K`@FZyh_X#5I$EWJ`I*U7j->jubHB_MT4@GibSTu}
zRTT$n$`gC1A*Q18nMsImAq_~74|a?e1K*(txoXgEvzfl~u@plX-iB|?hpWP__(n9K
z5;^=QD~4X9SqJmUH>4ZR3q11-Tm`%VJL_P`7YauIZ_>d-78AYG<Q?z4^K=+gQi=Y~
zmRidW6<gf3=Fj`z!uMnAQOOatB2Gf_m@v$nf1rT&!GxfHOw}%Zn7#m9$-zUHj`!`h
zUr{;p#z-d|#aPMN8w8a#@t4nsBk`FNCzv?S{|rb+whaW=@7ZinW8cfI_0{2W$$Ab_
zplpv<eor%kyr%bbs=%=3Pr?qc<od7*8c!Hi51en%&^zm;&1<E^kanpC1Fhmn8{9>e
zsMYvkx=A2p6Dw^qfk#q>36)>e@ubX=j+Ya9!Dbt_+!0bG<guJ)7Qo@^!Xz$^=-^p{
z_nZHK^2w`4^J8gY5qFqbYVbl!_>_RM1fq~3+vGwq6ex5;{>$JI*?IcG2@2-3rLLh`
zrgh{n>NLNJNT%$xppq3^o;%;gDvS7?ZM>@qO`V>OZ=<Sfd7EPv;OjO7PYJdOQhs*4
z0BF-3=mB@|cX+QgDLoMdJe#ex2NYG<vAG4bj9)oTs$UGk(X!+50mEBtYe@I~4h@Id
z2f>yf87~h36^}a4hGSQ32OF313pI_LkL#_h3L9LeE$pYqTyT(t()5WA8J9&pkkK{W
z@WgN=<o@==O(XldU(Vd&K9LfWhbo9Ce7c&&G3>$#?xk*-#+74a$E$sQals%$M~{EI
zp||L`Uv0##iA%5w%)@wO+4SGarBr0eGWFYXT{9x~C#LXP7wxh;deRo)bXzuHXHWPs
za^)_|9!cL9haCl_20idxOJ3*CG~SNYFNc`Z@FBPwiZ<7%d*;8kW4CdXtEs|a@o%^7
z&UaXc7#>~|Ul7C_Awn^3GSN6oj>VHILjo+CJyLx|9`z)AMY1C;3~ci1RYEZwEvD`0
zjR%KsTxRNOg~8iJ?^!T)LZY%+Y;pq~lp5j?J}=y4x5-ZwLzThA9)YRuV!o6hZRAHX
zQ+Dfn%8d>T(}SJgns$!MqjM-<F*qtI_+x^Yq#lQK(c~l;z|o-NV)xtknMa5`lpgIc
z|Mg)ff0I?ATW2?Rc*70c+Ak4RtbNL5q!BcM6C>U^Yd>|-`DJDDgLu1$R<&J6pBdZj
z<4y$5l8x4ZkK4<8*4Bi3n9ocLE=O5I7MM5<`1BFUdu3pf6?|uQ<KI8F)MQwDo*Nkf
z_IDl_E0Io1BTs6UAIm~p%do+rhUG)JkZ)3TB@NHp9Eo!QwWlkC+fbfu$M$t)+XW>u
zo6k!Eo&qf@oh6g`W6I1=s&ou;8N>xR2#HSY2$A5tPC2j7xveEO80j7E&)_(kJX3E+
zjX8>S*!9J?;`M-gv%-5)g|R|~6F}HebN$=6=~u};z}TL_g>sCwi%<2ud}8C*<om0u
zWNaP(><e{+g=h<{6)FtI<glXq=_#AVEy5gvV0co<Wd;H?6V!808AsJQD?CviZ-$!X
ziwCATFW9Y*wO{k&+_TE_4H&y<tGjV?Hk{_{@PhM;g^?ci_P@BR?8{0FopU9Mj$fE)
zIAoP21v8|=Tr@_@?^+9{><EMdZgD`QN`4^g!*sJ>LIg7qFww8#%3UUV>X#&V>4YPp
zoWBjy|ENloSfTH8M>07_4RU;ORGPYZ=92EBZ&RjCdA((Pe%}*I$4RvE&vZd~{=H2E
zyDz!UE=cD2$F*U>dX{{>6W)+?aHT}p(OltY=f4#q^4GI6Q#^Vp)Z1#@$|fxd8vpfS
zo-@kOXM|P#Aa1?qmAdgD-=A5ml))h%?VG%~LBAqzD?)%k!KH=hb_LqGyy!dT7w_%j
zzraCg9)9%Bt_}H3@VT6PHqd?CqiT<OU+PQ-UhLxk0;(svzF0B6Lwi9Zfa!yx{F=g!
zW-oM*h7)8$E{{&brUe^Pyx=V@fBYt;`A48xIDwI!%o{LKEc{}w4@&j)PKevBP7LL(
zFY4jSX6PT`)xr{8#I@kA_A0XPQhJ{WHgD+v^nc>mmE=_oIeVvHCe0ykuVu%a>nd)B
z#0zIQc)(Vi_&w!f`1iN;X!YQ*DR=4K$F;8HcX8lcxD!4F*s0>9(uXM`tn1Q0iut;q
ze`S5Cr7$6v6q9hlS2jc*sCA`k_IW5(RE_wW2RYJj&IxQPx~&QwDwZEH`8e_6aX0o~
zh%Kb9RP2Mv^k210;y9DkwB#322;p))(T^7nd}Qc=!lkGI?C<S<6bLGCTl%LOw1vAt
z>c~*p7ziXZtm(6DQ;1tTffNIIq8QLYVd@Ps-vcrEf&`!tA?zRcYxRuHHm=ljQ|EFE
zS>hcsz`a7^mCr@0w1b}cI$Qr5BO$61M2p0eP@Z;{d-yOUSf|I(>vz=|VdBE=qCWrK
zh12P_J+aO@ipGrn3sG?~{>X;5acmb3HoE}5M_wKojyiE0qgJ;a{S(x}^8U{H`fY(O
zYP~sSnxc3BBnJuNN1i}7pU)JKHS_IL14q@Yqw7VC^Yx-6sUSfJmktFb--{A9ES?qI
z<%hlCVBz-3;i03@kiQmk<6|$4Sr1`Ul<YJ)<qPvcdOS}g2g>=oc=$+xxk<fEX*lQ(
z>Bfca?Eo!AD@eJ}ZM@!<_YT$Qy<Q5o{9o*TIMy9_4hD!7`$ZNfqX5s43ANe#gzIDe
z8Z%<p#v}?A))dpznJa#;d*uZZq@R%fd$O|HZ+0Bp)ImSkNCaziZhqN_6nPX%^_o9C
z+v~R9Q>stdubmc^bj#Xg!xJvH!2#A0DT3<f2vhf(;P0*M%)P?RZ{rrxOofL7tlF+x
z;;*&flb)?W3104)CPetGEUBodHQIUPCpt1tL}O@Tuy!0VV%17~H<&Eze0oV{xzq0b
zmv`b%_sx@m#idxbyhfQHtlm4vEqstDpJ8&s>b~Cd@Nix6%6qIO;K%f5(6Wxs8<1>t
zN7y>o@>Nc*$tjPD_x%=CnBuzd&f!<t`X~sapzFx*3Vka4ed!n3#XqYl>(61*C(6e2
z598a!Dqjtman1G{0zd7K89{PzEm_T<e$38e=%U-oGde<bXKszI|FzaSGFoZAK-IuM
z>hFtc)<5z@<FRu4cob3988?yN;T!JcUR|91vgotxen)=l6&p3ab@T=+W{zs{1d1;j
z)WdeaIoT%&g;o)JH+?bZJB|NY+EZt7v~mUh^!$}up!-lS4W{k%1Eu}bqd8-=!iZrm
zQACcBD?u#9EV}Sw%iM>op5D!=w1fD32B-#+T;1WpyYcem(0bx2@!$;#tj`0?Wz5mK
zUH^PhPp0!^JlL=m0OXUu9?V397o)~QB=L{=Bf<}*3wO|zerMrkxX3JDx~yF0(xttF
zn=?1x<qY_9h$&MKZU3!P#Uzd$$1SDU{cD2aYTMPcr;r|y7&B?a2n5+XrSMBRzXjg<
zh4-mTz04s!`l-WlthN$q8F!%?SjW8{65YnXxOwIpg`;qO&qg5G^AU!_L@lWG##w`s
zq09jKo+(GHRuJlmT#TW;?wSKN*JzQuw?U2U{nGC5>_S_RJYQu`-g~3oCx%+5ECc*2
zc5f+!Sc(G!jbnvcWL}zQ%ow^rn4dY=xdPbexN{@oY3Pk=>bqob@uJC~a>|_~KN_C(
zA!xK}!WZHKsi-Fqp$aA_wXUDwEP$`M=ND2phWU!iVBUUnLm}j}4?^+r`3k{rjGI|C
zWv=9Ap68>Y#5))<IV7^+)SwQY2VR~Ma<eoqNJxF#UY?}F9P`Phhc0b|(i$aAg`N6o
z<F$vBD@Ggfrw=J|x}^MMv4NgOQtK-=Q#(TB-}>a|9V)^cB_z609Hm?VI=FonXn>;0
zrD5t(h4@VE1*Zjj&4^JvM~SeB$|ucm0~A<0c|r{a$Frf4*RoIceMefcE9kcY8y0n6
zX<Y@A^PY()n7_-Y(SyC{DajkUu#t5%9$w6?@lTDPM_?@Zx<+LKzrrl(;L9xO8=TOv
zSlQS$ddDDFXjRA9cKS-W_oH`2ue?r}C;@tD3SvQ$c_^58@R&`(g8P{WLX>1qx%v3^
zk5@h${EM8JwqE)H&H4QSv%woakO5|-1<N(}nuW&V+&4Dt!e!p1oZ6WhS%g$`_GZ(W
z=C4WLk|9YL2VY7jpL=cB>5e%<=R1egBfB^||1{+t3pb;EVv-AlTvrpYurSp;wegyD
zQcMtodO>s%qMFgqU}JujGCkjVQ`WHfHKv1UmX#c^i^>%BhhkYf6Op*7i0jmGS577m
z)P8aZ<Ue^+3X7@}w&t?oLA?jlGXA>B7MI`Ia*{%Vs!T0lc3ojIae}mDh8+G;fYQ7k
zVk(uLGJn3b&)lIUoS@u;A?k>a5H1wR)AjyD4c;G%4AX#xVcJ{{GsqgHI3N<Jb|W1|
zfbL+5@h}HU<2f*B+m`24mPTvGk*zOu6O&hIpE@^&b;b|c2-?7f(;@PETy!vDN0^;6
zu?YNB-%Q}&_>->aJkry124}`GhNU@`jX6aTq)CM(G0~8S_S6|@&za+^Q4*Nmz``Tr
zqfP$2=mKy5BU}1?DI#Q|?Hj@D8zN(e(VhL4B^k#xPb&mSioDmZ+<da`A%ZzTYfN~*
z5!f@;%Mc-GnXK$K5`33O%gs|ZHEW5K0OWqQ(o3NVq98dGHN5^j(1X;?oMS0yu2-ts
z4uP~bI>5fTe-rVt0D-b6IxxO2<FQ(dHIW_C4|^L!7b4wg9bP6n7L-`SloYZo*g^^l
z3Ks`$O;)M76P*oxysrWVsrzQC7*3b_>rij`4#!hko+X#jWBRRMM)}=#;}t)BiZD}W
zWI;0RxuuzXn?$k$3TFXrNs=I|HXMojP>Dgl!M<P7WD}K4)7ADkx~cpAXOZ^+oYe;+
zbtk&sKp^+;i0AL-yxi=Lw0Hb#S=-{Agxk>|_Q;C&1y>7f2>P5!Hz;uM>&X;Fxt;V8
z*LJ}h{`w<QsP(J^(N{yX0NGlOpa$~G{s|Xu-h}!NsR$0{_kxD0%TmXbC9xn~<|o_P
z@o9ab#W#R{Q@?a=Vihhi%u-H)l%GTv+8p#n-y#2Q;yG->fAFdSE6_L7aQ8f<7rTL^
z{oI}UU<3h~YX9FfWIAv^3}~f?mq|a{aQ{Pm4cgH0tNktZ7*<~^&k4^J$La!dE)h4c
zg?8Zu5r%YvY(=e3as7$p+thMc5G|39d0l?wJhk!Afxne5&@XSpV}dsTBg$fq!`00W
z>pJ8)1W3cO0}|be9ZqXcWAQUxOwwYGXJOI1;8e7dam?qB^6bz-Gxz3JMG3z}(AF(&
zqK_V`QX(RT+N;)`Q<3Z7X@I!pN#zV(QzT`oBC>I(psTH=C5`op43ZMbUKIq_Ir7c<
zJKs&9|2T!$eJ!ag`1S)ToB9NE&1K&XsgUd5?YcJ@ZF(C!Py{B0hFfSi^zg@N-4F!|
zeEz4<dy_}ov9UMZ!P7m6{60A1H(8@eX4GNruvQ6mVdS#0{9iKYch_%2fZ&3zJr%C%
z2D6SLc6UUei=m@%d9WQGV3giHShvbINv9_ON66#L6>q|rvA;jLO7<|TQmS5XTGy5E
zTz@+r`V9(Kl_&SuroK!seg;|!25CU9ow?-(-5j|Ux7LHD*)R}naU4l3ndE|GQsJ9q
zH(Sa=H=IkY8!F5CeGML^KHJO8$6oVlJXeJEVx$!oqyN;Sv+{Lxbmwk)M&yxi{Z}a!
zNoS8^>TZV>KAOV=dGmFUHRd({H5o?~YfzRYlPjro;og^koS-%cfc=;NuBsU>rb8{N
z!#k{HE^4W1TM9OMSOA_s2vh^6CDnII9yo=#<xN>V%gMgob+|_UB%42XiChezQKID?
z?fB!aVROWgrg1siLPa0M|4@8@$C^&VoI=V;huvmUm=DCPX9#Iy2ie-!D;CV3`9MwO
z1~sA%7X<~Y@x>J)5H-J|A}U)yLn>p1i~rvN`3fhF@M4=z>+#nv^o>CK1Wo5AMiM+3
zczym(2YeyJ%FfTz#M$3J;{f^O@$rF20EwzFX=*fS0tNG1jNB;(!k+F}&zWs4X(=r8
zrL~wJ(*%AH0VEAJZ*~{H!C*<IpxlY3!Oq;|kWRRZ_Avq8Y`a3~i=3_^`6T&aO8gg3
z4eeo%ZHrvFfAYb@QCBA8SM&^!u1|iq?0meD(8FQ{fvbBCqu3EPcy#7^DJ{O{dWCsM
zrweWSd}f)kjS4b%P!oT&`~LW&tbeq*SISxdLj3oZd&toEe}A^ehFQLLd_a5;Q%ImE
z@g{_!GpeNHZ279YN3AdQS_59~5;oxnC2l7qeXG*e2r{9nQ6#b&Ul7;C_3np9xTOEq
zyUEL$Si^og>)!HU_soN-F3<14#%cBUQTgKw7TA@M#!Q`?@Hq)`=_dM|m)csW{N|ov
zB(eC`kTXhQYSSqw%IE7fyog7e9o+6oPu{8ebC%`g=sIX1blj@ZZFkZGny3mv7F-|^
z*-S_nABpv;x-7pvc<2iOtHOAFHyvVFuwe@odc2(}#|)wv7yJih@eZhF7*#J{JYWEE
zIJTAb%Mk2qO5+kTsytA{hpyEPDNcO<^Vi$nnSr*6u|E74Rua9vR_}@Jz?ze72M{VZ
z7XFBCm-TY{4rF~1<;y;Y^q>)Mtwobmm>~F7p>)1sR(O7#uJ&cY3xw3zJ+Hr`dImpH
z*%-Mm{4n7Z#MtO)X#N?#kc1|j#U|&{GvZ%(RO=TLKU%nq{|k^GAL-l4P;q8rCnbQP
zprNNC!%kU*uA%FujlvaUjTAvpaH<P~hi=b^S|NjS97zq`LL;|;`^oAIPxp(kI|KR`
zlwG*z8(e8Pkt?@)dJ!eEjiqIx)#D!pKkQGm;pCanvpXUIb+O(X=WLs43|T)Tl%xHa
zsqn4QdYJutJ88sxKk6-*ly~yruRG4;%>9<DQa#lr^yV<Q9uk0-iZM$cW(g&>T<Olv
zGv)1++}$=l(fV@kQWUT@QJYMneDeKi0Y)TChTd#_yX%ZG@1u32LPkU?UkgJ!v2G|t
zZhS<xyEbEgdI2S+D4Q=_(o&K64Se{Px+#CvYCtpqb>~vvyb4%y;!YA0Az}(AfAo!%
z(Bt2+T(9>8o#iC?1a+f16<&J91y{j4?3yX0`!4h?t>BYarMOQeyagmF5Lp0B2!9O0
z22i`CoSBM)j)d=6#`N2ol~8O{-?iwVx#e1`4SPnX8{h1cCwk|v@6H^is|a%#ofOg?
zT?UE}eR&I*q!t2wWr)iR9;+-f2PT1r{MqSpH763M$+Z6r1;6`UKzJMGskfKcvkrhf
z{<D(r<$uX+k;i;VN=Jjdl5I8~zr0nF2I;MHKp_hXl>V<o{-%Z}7jydsNz~NLL`@PV
zl>5hA%c$PLHgCVJb?qZ2vWE#-(ft#~M})(SHs64GGnKj(Gt`jHNf?ZSI31Nc6UuoC
zDRhLW`FU!QCj@`1L`rX2>RswX|0VRXE9#OpKsE$Sg%tj0C`=9z$W9dvzCw-H6_rq&
zV*kc3u7hQc*YCpjPYD^{qdGa$_WN0Of9nlxDEO%9ksZ^x^T}4b6TO*fy*fCRjPBq~
zzU#$#3{U>1`Wj9UtHP(d-EZf#F_&bf|4i4$-#iD^A@0lpVR8`tjBSqE)+{e_)ORm;
zYhulGZC7@TXpIaks;QF$tT5Tt_yw;tBB?{{#s`a7wI>4@#thC*JnYXIRO47~Cj#Ka
z=ND-9e^Whxh9-cHg@xpm{XCzEl$NKxe1j4eLW#S+CJWd_P&m&BzeciJm?#uUT>|7y
zRAe1YU6^$!Tupu6uR%i=uiu8D!_wTAlpOnU-%{<6r`JARwCVdGAR(wh>5M!;0RivB
zBWJNFdS;6OLqgxm05*IpF^GAR=JNOl1~U)q9Y<^eKZ)Wyv72%3l1=wH&dls&st(KU
zLOJXe3DQpLTv2kDObx@A%1W83&hPCA_wD><EkKY6N!+8gB^!-U#&33g3{1X%C@k3P
zZCN1&{Dp+xd#Ic7ilEA5vrJ(Xi9|w->_L#giE}qNz#!DY{3x#^<bj>4@huJ}Z0+-?
znr#jgA*fT>*mVvDQ~?nr04=E2mqK<HLR1AKdc&cALrBk*g9Z&OuFElfPh40o;q<Qz
zB0weVY&+iWWJee#st{x_G>6|U1?8+YW<+Wt@*VD!g^I81^c`f5#FKgzCkVV7riACl
zmk0LN35VAUU|E3Kg8_u8bxZs69#afSe<ekdKS&DC@4}v5lq;vfF5|-b0qJiTb}_0S
zTPPqaXyTMGEPRL(&}`%~{ked>`W(gs8%0I;;jO<xKK1D#hKhyWHCEwA{?Sb&vApj5
z#Hb%Jdqp<<Wf2$@MT6SwJ@~NKs8%1DO2{Av5NlI$$o;)25v$@QNIb|o2weON>D>dX
zxbeL7B?`gBJOk(MvQ&vR6LVTGyE1a-cC+O#G=IO}ZoyK@lI80%3cL?`|6d3>c#CLV
zVYD$%CntvUmg}4V|C=h9Ymmtg4|)JldOpYG%E|p^<}2B8SZs3Z0SUV2Pfx|vRaU!P
z*xH5|BNLMt*eNBFg&Zr3hw*@Az}$jzeqv%6+Kb1XHL)ftT0!!*-b+e~ZtJYx3!JIm
z>7P<y>i__Thb%%DRJwUY(3w_H4+vR%RY>)U<mV)^#>in3!$o`!k+<d6A$>9^puo8D
z9>l~#3DON_PeC;m7iWV4G!9FfnQBlc%>^PAJKt=0Wir+Gk4mo#LEx(alu`*vrMsAa
ziLzK{&3N|`_X|M2`z3tDY-qxWy){zF$(Q~`mcvv}-*zy(IxaOVmNSq_!lL=`Gfohx
z`-!IU&KYfCxO8DOpeZLS9ca>=_EhfBpXe)+YoXsQq*I!%(6*e5`6C0{k142L_iyXC
zeE`3<FW|%T`(L_7X}CIAlg0G1i7C}mw-@Cf^n^3@M1x~1(ZGd4xqeaOB|{PWZwWi%
zUf=y7;`bIuO+9DWiXgrc5a&jlxx++Nx~G5A3VCM*fe431^euY%k-CO;r@o><B<S7>
z!wEtOQ4mMQ(+mD}ikeb5iJcM$SBS)FwJG)mB7sebH9)gU6<$vn@c#e1#Y#8`2g9IZ
z#Y&0bSRv*}-Kge>tNf1>E|UCL?%bv(Ylt4&5jC}(VU<=3xAj0{*<|y<Mv4J-Oh3J~
zY5j}U#}!XJU=sola_g;*;nlXJ-RiGwoap93&CJ`#ntNUnzc$8*f*v=9kR#4W-Ez8C
zjqn@clqJMQfIFzrO%<>r&Fe20zH1b1m?2~j)snO+IYtTupk}T59o04Wsw2Lxt)4Yl
z%dI8rKe|?50bU{WQI=9WBeBp0&LPEzTpS<I^Lz73Vj{h}*LTWN5joF%8WS%gx@2Bc
zYOsKsjZ0;fBt(~Ty|^}MFnZvt6m)v8`d9P`Ty{CQLR*L^9QBSSt;3tms#bk$_pAI%
zO7knR&u<9J5ah4KQ+-fNsOZ??P$V%^q1nIONE=J*L8TPkS3h5xI_4puXb9_O5OvRv
zoQA$tHQ}uGpVv&nqr|VDQctkQ$LO0m5oO($52yaX9~U9V3Z<Ij<nDEnLO7Anq|f6K
zRQI)4N0|DGByN0o?9D`8PCgb%O<bFd0Mt*lxf<fdpqfh&$p0{^$+j7*ZX*SdkpI&w
zDy<6n_VT3rms?3|;uFb(@n^mqR-f_dKgdvlD);{c;O)OY_F@B*1hohgPVMJ3Ir(7;
z33+?~zP^0<9^RkyG3|cHrNj9&6<Z|XJ<Jz<X53;dN)j<%F<0I5`<FCQv9D;oqflmb
zJ#sE0fq(l|!gl1?Fmvg}c&K)5Lni7(F_nToaD>@~sFNY@L&cJdT|x2;uATlI{{jd6
z>XOIIt^@`Iyi}sY6r^^Yz!YnriMh${CV4=$Z+_+6P+cRu3m*V8IfZ{4`Ue56+u3@K
zwB>rzeVZAS#_JtB$j~B}?tjcIbKs2F+%0#vL9GTuRQ>h%@bCk$lwtd~@eZDfuu<Jg
zz5iz){h(s5ck||g2$EEXt<is?v`PNsJ7l{g^prKy*o^6_bR@2_N}Oq+&Rw?F?Uz|v
z|DS_0dHxR9-?A~?o@=`%TzM)AudWAAiuV2zs7f>If`?zRB=St^Q?7|$4Czs~zHufd
zCMOikd@Qipp<$yT7`7FV8y)+GG*kKpRWh0=g2@VTxHt!{A`D+%mlpvyJnV}(=5}x%
zCai=ivo#Z&Z#uRN>`2Ovf~6dj?!ElLG<J}ta0@RLv})P4cHc~QU5L(uO}F|la6Ks}
zs+xCW6*S^x9;+54m>Hr4K?VlC5=hXyAmaCmrB!9H{zho~hb59gfCS*(OjCtq5MH0X
zsu&GN(t(iZLw;7)ppDA<a|`1eENUJa0wO&)7afj~u{S}P<E*%N|50D`SeGyc&W8|W
zqz0$>zeI*@SJ*W*Cf3u%J7s8@wft_CyI57ocMR_=!H7QO{@ulMle7c!fzg1UoU4R0
z8RTLudy73!W%TW}ss{yne`<??6{`kV&Sno&Grevpo)+9o!lTHFBt2e_nzy`Fx6W}F
zo7hw!8A#FCj5>`VJvyADD2#-eS(8t3UVKSUjQO`ry}gZ3%}|ZG&{F*p{pHC{&wF;(
z2?>zTAIi$bG`9Y99WVp!I0Cq+F5PBw%Z=zyFWqNytIMH;9o+Z-mi!rJCUg%5ha9{(
zm^M|*k)V9#tsvC05*=9Rf?XeHEB^G<EzxlL^(`T*hyYGKxKe8W4l`=aJ2>e-AZu<7
z_DE+}eBT*zX1SDB2M-?y!9*@%Gxs#vm7!pO2pq7-4y<=E{j?-w;}LrI!V-(xa9~LC
zYC0(1=%QlqLz!V2#_-sX;rB^+KmjRX%Bo1_U2~u{oLj?8+0~Nho13A7N}GagVCly6
z_U8dKrU;|o^Sm22IK`1b+Z>B^uf3a_{4iA`!nZF1DZET3WR)-$=ZIDaV`OEP2LF<m
zmu6&S6%R1`?B@Mkz60_IWTfIpwLh<dy!2Fh`2bz-4eM9wfkM1pEk6=CC=>d$>ejNa
zm$_pWgda*+x_D$jrE`&+TZIY$SyD1?PEi>wR5k(|8XhC#X>M5sq86TX2_uVyMu*~l
z2fFgTcBLzi^rz35b0IX&*9S1*E?3A~b9D&hoxYtM;&Yq6IP>J$-<pfG@U=a(Uf|Z$
zv0WPb2OTCH&oo+FR*BBS<-h--l)5^H5a{7iDkJ^%{|$THa-$c&S(Ei}%>AdZpXL2w
z>qxW7<Pur_{J>jL?kyAE9|Y&{!TSdnn#9-qn(5Q0`5F5GGpMmGhOXmBAXO=={nOt1
z`d`7@=OLw^f#+s{n}?p%6J|fFIa9h%9vXXpXjzm$wy`Rxb?u;nlqB~bTlNaY1^v|*
zUy29i)xn(M7rB8gfv%Ph%RxB7q@<0&cT4{UPIz!9XXsQAoAx3T_%@HTjnt0(JxQ-w
zxc_L?ysVWrG87;yu-m6mfKmL+M%p^|uw6wpCESyF3o~IA-+SpH$R%N|G*s(we8gRx
z7Fcs|CRC+uC`>(=kO5GH7Cc}CS}V}Hv((GF`L)hKUs8c!-wjG$tCS-V?6BMIG3J-!
zJCX#;{_OY^VqnTvE%Zm-*T%m?NYZ-ySLo3btmwud8E{UTS{KA5`Wu+dnc%Sd8qVGH
zn@J-b@ALp@28NwgV0Fz#tLkX=7A1?mx}^Scic}vfzeI?}bQLc-e;i~vV<^H6<v$w&
znE&l(!k49_FFg=i9Vg*Q$7PF2CE%)7)k*hc*%pATpX%M&St%1Oh|{CR>az9>bPs7g
zudqSYy<we}aW8X2Zlzu(U8BwX{hLi{F0tIwju5c@Kfiv{3H$GpEroj~04GNDrZm90
zmype%=^F^><DYL|Kju_l*aO^h=zg9Q@T@v@uksn#3bVGF)~5r#b5&k;5LVtj|D(i}
zCx$Q<<-lO>1%4&z0MYw)Qz~aoP5!SU1ksmtT5-u;@izFIXRnB7J@(i7g1@){cMqw{
zPHO;zTZt4G>}me>q!5bht3{!j8n?3AF8gmxOT%a<^T)RN&zZoTA2Nv2<`pXe=)$tq
zgQcT4Ewv+OTG${2eiC&GbqxIk5-3WBHka9=CGsV4{LsFD0p_<a#)VIR?aFhV{u|^R
zRW>YpX8Se$ix}d3bUm$q5;LdW1s}irVg1AN!nhJpp`V+s5y#S%6T#FO^0^bK65}qM
zNrE8aAO_s8yHRNTV{x0+1mxfq5ok#W$=yj&4pOGB?SEYVYnuNHN+<8AGLBHtK7U+2
z^XV)GHZ@~k^1WKnS8S|a@}cgg7A{;o=5ZTeFyb_=bT+^!qkH;~{P9|5s}}>gs0vtA
zhgHgk(+XpzW?&!2%>Ir9GNx^qJs|o}(HaUclI7n514mKZ$MqfSI0|d0!88^O=lc==
zetN4QGeW{YwIj5oq{*KiB^a|h6bAv<2?PpcV4Li{J4~oK?sDDVTtXC4$~3hr2;G}x
z0(}>+spa1%RuA&_;P}cFKnBv?I3dCM{S)xexEBC%+yX)8t%$gh7Lvh@vA9_+=Gx7F
zqJ8fJK;r+#S&kd^!}JdnIHT+9gSOc`12dTjYPWYy{)ECdC>?$m989dtRuEh$%pmF*
z>^ewD#blI7dUxEvYLa85i%L_03K}yy2Jt9Ot2C0sqGyr78s)X$ad6oRAAuPo|LIFI
z&}jPstNbc?or2b_B06k*ZMLG(?R6iI9gO)-4dzPDv5w-Aj}!qLd;kq;maj+_KgA4<
z<AD0ZP?sm0KJMKg;8WU)>FR0Ddi3rDD7vrwp_i|CM3cfX-kNZ-2@i<DXU4F$;CYhp
z;7p{5k>cNGZ(?sCuo&H5n#2=uFagaWQE#<(lza}OooGp|w)B*&Sz<77?pF_~nN2{V
zpt?RIcHe|Wo}yiD2}xkjNQKYmCEvI^6ar`ps=Z14mMW%Px!50}+Bv!3;GUn?qU7g(
z0TMbU75C$H7b@_i0-q4|y{G;8@OUoJ_JcKgZf%CVVb>CT9?Jj2y(N&ULDQT^mD=Dm
z_egTie_jg8fPU7FTJVpDR$@w>gq=qdBmFN)&uP7&3ywaHfdHI}_}L;FH)+DT>A@q5
zm%f@+Hg=l<^5%x@7&3%vg=xQbYapNzo$-FsZDao21*)L5qd!{7fELg^@>oNU2?OB^
z^@`{0p`8zkx#dnewbe@}`Fdx=j_yiwW?F3sc)pz7Minm)_Vtb>Cr{)k8&(GMeg&DZ
zpEUXN2T*Aqt2>dBQb@3-tb4m5UuQLkbGSD3)-Q!%IJ&yOKnwA=VFQ--1Tn`iZ5+6~
z%%08@IY$9Hrs5WoIlg8>yxm_|m9$Q4@9f{%*!>WZXHJE^bE%|>d1~h55QY{&aCQ@a
zak5e##JSe51ohv>ya#Xkm41kQkIJXjmtvJS&+1$*XG;+{HMTrBeML@Q29I!R@(<JV
zQojB7FP~G%1i<0_^|{LlMPv?c4f7Ke6+}YUltdBXt%X;rq<BQJ^>*2>+cn3>>yxvo
zZ2Buz(m!21Gn%3+|7t*P56I?29JRUW3n$r__db0Wpx;8!@y@4rb`dclYq00!gWf#?
zc|s59060BVfhoTvVOTrEv0GYhF;vJG0{v+{0PDHgPD8}>g5;fbX`zjk6v#w;31ok-
zNLI&ZxtwUIsRe-=Uy*3xEOc5Pd~z^mNDE-|fNvx%vb?DcEuFhAxTy`?={3gBLM0ns
z!#YqpYBpcDhkSk>lUrUHQ8LdBaJM)kwq*C7nb=Jc=Ue{d+~WNBG>cZ#^IkGeiG8DY
zjUJ)tW!T8bYailBJhOZ9k&j6;m$>_XOY#+DCzv4iRB#xT{lUPmmKQ0q-tO&{QmJwP
z9cb8Q2^Jg<!*9lnS`hvp7lhD<X=zPBNkxYOfMWUqbE;?y1L!Y#et%L5B;m80z&&SR
zPO!aam5`%i^r)o_r17}XI7VJyhh1-y2jVh5AxybAikLSJVj-70rwdQf+m$%Da`c*f
z_y%k1*kjfkv|4&qKvICVxmQP$ha;6^M0@uNf|{x(uIaL=$-F*Mt_{MC-JgAb9f*hH
zbj{+CkZU1pP~o`2?|slE6a=Vk#6ay`UM}}D<y&4?pSQXN)G-Q<##n?Arc!GdI*uGO
zbN#ne4jZ2Qn0%6}mNrq*O5uPjJQj@~xLQY?nVuGNed}ApD($CUq^Q?1=fzWk0o2{)
zt5dW(rgiK$tB6C5ODcs(O>4UoMSnpvuv+r<0{FG`U1?|1@;+}s_oSt<-K`&ay6^nP
z@>S%N6}8%*GCOWt+6>PNj}Nrx$kDmEMTR>8CB-`0M{n6A16W#t&#3L0I3Xf)FSkH5
zthqq3pDtO+{_sO-t^=1%gY9nECrQ7ZR)iT?a4zS;xD8$e!<5<*YzgB!*0wbHFoQ47
zlrbcCq#)fg()g$_2hq)VO^C(@DHc-V03F9|4v@Hqo<=$T?|c(&^WTAz<N6z~f`Nt>
zGrVZZWrfi9uIK$D66#jJFo=-v6%>d9lT)MOt%^W{L*P@E-8x1|)KQHESo87mP~`Dp
z^oC-2ZS>X+kX^Uj?qMype!n-cky!9Ds?7*Plvo0Tayxf@4~wmKH)eJoZk=9uU|0R=
zjXIs4aF2;^p+VY_U^OdGt}CIYM|Frql3d#8A5gs{?YQqa*<nk#=EbM?VdtFqdIr~i
z!$?e~-#St;_j5sl8IqigTPDO~XO@Qm0BJ$NUJ8<2Ev$~Cu(y+)F8TJbv_AQ7kD^re
z1~z3AZ-|11u-k{72$0y;cTKN{nY&cwvAk{sKnWNEeCX}MWJ=zsu4-dbjogE~s5N0b
zRn+mG8-5Uh5xK>0J%jM9c_FMg^23M_@L&qp*{CuAP;4wXKzWcMhn1vXy%l8uWU^&h
z6t{z=#*z6y;KhHnopu@#cM%lY+gL>z1uA`6edGh#7wEjA&Eru&X(3|c=;ta#T&UjT
zd2<E1N-cmlM0Mji_3zSd4lt#<e%7GBpV3U7sBI+SyFO&3u7%~3GzKH+gxSx(!z%4f
zsx}ZoB~+oXTTLnhd$Tc3*YuyJNndC*ueCrG<JHP&*N*SFEH~NxDSDK*La%N`yI<Vg
z0mZOe;Z4TpWxW^w4MS8b)56^qWGMkPYlxZ}2^t)7gY)kgumXYYV4l>O2u*NY^6+`I
z=E0c~z!p_C^`ARuscWTS^PlrQJ;f^4m_T2QCaewA_7^&!*ajKl&ZCX?kRj4dL%E@%
zB~a1WIP`~zURz9^Uf-GpVE~kO__!fQ%gTG)Nn8J~jrn%BV>H=7Mb(fxDec)mgZyZP
z!`D#fv9p%d^B_d)AveD;`b=_Bn-!8j0&Cia|J$VrOkQ>8u}Uj#yRJ0cS^UvgekFr(
zS~{DQ*8VC~T&>$+ejaGX!4EGw{;@w*B%U1N*$Ef0<O~f8C5CaF#AE{qaJ82bwSdwy
zrbjp<*H_RJbXN7RJgCk4ZsLQgjPl<n)Y+5Ppjjo-v++7C&i7O0AK^?914!ra_MCnq
zm+arV%`}zh>^!x+OC8=`67t`HN*O=K$f=0GsB4scb;SYG_-2>rkk%4hF39uoCWy%a
zA}+_|z!Rcv)dhx%P>AUmp#AosclY@Y;x+&ZH4Z0uz&=XvUus1taplhASZuldWPHnd
zPy0$pYrP)gTS7*2=-+DVq>2bjoPL~X>W_fyDZIaMdU(|<F8{3ubA&GDIJoY;)A%O>
zdRc)96mGOmRYzkmZXMBf<96=ySkoU$clGk5?bX_NECPmlHDy5{;5q|11_K=1yJk=C
zQ(Amrch*vVPZKaiyaWzd)e#`O*nM&^0@9|tw8!C~#JeXc!JOrouj*LWneA~zFH30l
zt}xuTx)fPx6&(JW#R=qs!2jLD4*TZX8goWsLII#n6=T|0z5-xIRZUDWB`vJ0OYX_;
z$;W&chrF%nPX{i9*cz7ksu{kEW?Ru}Q{$mNUKYb+BDZGO)}-$=@UMK>ZkgJB*?NM}
zG$w1ct}0sqFnCR6u@!vUU7&8|+!@^-ILrLx2YkpWRRXc|VW{6*{V4wGdDn3N8CuU;
zZFb(m3ON>zEKNiQEj?auqnkO2`G@0gwi`&lH{)s-u9w3UhhX`MB8U3&mUG?IcmiE&
z5OlP^zW<YFPS`f~wk5bFrHpq&#YHdaE+=b*$B6@KSH9c`k?pCdbKzo~DK)7OXj?WK
zu#&MPQ}}&8O+hIUBQ{RE-v~bI-VQ;L3R9e~a@ztnGc2<-;LK}jwtk|taS7ofu+VC^
zC;!mzujI<&G6dsyRajYc-k8t0F@0ZSpUSCoZGUCHLir3n&yX4;bHKL{>g(CX@pRK4
z)M3;d=vkei;Nn;+7TF)f8=sFv1Cp|B)a`hII2HD9S%Q)I6ckP9DmwPfZTh`^e~>{W
z{DU11R%O2Pg!oldT|szjWr(;c1425`IYIqPckk{14JUrJS}JXhMF0B*W9ix``HgEW
z-$0*hSISa0xe>w7Bry#wwf>`g1<?^~{)3@k^Pf&zFE9Sr@S(Lj-WSEs{f}G4df`wE
z#mSQd;GNtC%`nh(+DwMe<Qn(GaWkq5A46KQ7^7>fJys!9sXUEy6zocWvj4@7-CoNf
z#-!(?J|hbg<?Z}SS*C0aFGvoX^Wm7cWaz!%f6rP|+JeUO(Y~{0?GpUQ8r@vt6PB^M
zb4$oiA%=eUemNNc;biXLvtQL0voKvCLAS#(Qi=w3D1*k|*iEO_8jW=>Bmp{D2YK4d
z*@G^^Z_bc@MM2-19O$Ub9l6@!;DrL&Us_av6x`Xy<80c;T}tT%8jt@NjJG$f(swBU
z^<sa2I!E3iQBD{4zbH%Cpwe9BOj2GzYrYOe=)y<yKWWn<5v>~Azx7@!ER~TDH(&NY
z#SV~##=^-)Pn3f$iq_5qJEWM}M7FC`Yz`V)PCc3i;zDD-d<vwQClrm!^Vk?JT&s2b
ziwJ&dPE83e&>}cK-oxwiW<(f8LYKRfR%=wXE7|!Nhd&8i`7U5b0CozDHQ1BAva)wt
zP><=kp($^q@U_pssmKE={O4`o!%tops1<rbs3h945)&H(QO)hh4()4-P$nn=(Yqc0
z$ty@}M>MuTrTZ=pv$o+s;<CI58$0$t*qdWdf$my-w})qp*SL0`b>}YeSj#F6OoD`>
zPJ*#Q4B$S}D}MhS5n60~YngY_pkwHSmvfc_m?hF&`v<fl0Yh<C>x?fJpU`Jl1K0PJ
zgOX5*sQ_IP^p9~^JCmUPJAPWI3;|$j-jkDX&Ph~S^`|QstRMk+&;Wael*313tkGGr
zfmaLno$0q_tHtL9Zzv6d|AbQk@pigBb@Eyo==&>@sEBJjLgm5Cb<fBJK>hbU1?Jn}
zx58I;?#t7M3xp;?-y^C;UeBuJQgv_G2FVPHH3M=tuTPD6W1gVz`ozZJt6ou6Sy&Ah
zOBr|aZTc301jq57gWRv5;lR)bJQuZ7>#!+>w`;95x}C>SifH?e4kTtb=}212FQ0((
z_cFU#`CiXN7N#`!C8g~coUX}&J?e!f?+D7nja|D3cbUP0-rqJU{{-6Z89uU>A0y1?
zOpklvV!eG50#a#3x@dE6z@y`sU7Ds6vE2w_*Xa^M_2(SlanMck-%ZFgv4BCINfW)Z
zXKh$j3NawC&(>{8!KLJ3IbZfbqdK@}L(uf`%MXeQ7xksEh@tKmIZOb#=x~H9bBh!+
zc<a2Y%JZ)kx?6|aS|ZU9Ug=N9Cogt_hK;k-zpM&v?c@g0$bJn@f1}Mh4Aw!*SGKfQ
zdmP|%G9T2M@kxX6jXm)0cRCcbm7zZB{|d{GLkdBETQ%i`o_Sr({od`y<Cy<P<HyAV
z_xrdqMS~7D?M{*ldUr4YUObYZPt^qHeW^fAZ3P015n7r<zd2Hc!EdqhLG)7t!Lx5S
z{<rn8pcn3M$IhDjX*Uqw%NnYiiUa=Fp1pU=+~kn|LZ_(LfP;;M0%ltX3Ei-`HSUyO
z_^fNP!z<m%%E0=x>TUe1YSAjg^93!TZWLFMhR~<?nTP<mzp1Zyx3zDiNUV~AMb-5r
zgjhf1UU?p1{XC8VEQ`qjj=N?t-I91Fe89U~<H#X7C~fLIZtW5`4f=7}!Kq8@pE}vJ
zw1>_vig$MBj*N`K16@+0<>MxRX#_VMMP>phO_OJ`_O@fi!S%p^kbFi{?e1=zE|ja;
z21VAP2QJuo1`~lcs!tlp8LKGNfry+)l8_TLxbpIQGogZ}kDi}mD3ZQBLqPQq9ntw~
z0Xz~Pj%{yNXVk1z<vFg49exh8g#0M7OAfp^qMc-T6JZXT0YM`Kbb#zgS%xOAnjw+r
zYFbu)Ql4Hl9p@}1i1`g4KVmpx7y!9$b0Vz`sJLYwtMv)eCQXkeXIli4XWX1f@w|gB
zCx=I+YfZMhW;#0_M<S$ApD$j%VF{l{i$eD28j;AQ`z`L)WD7@+g=1wBD4zR)&uJSS
zTTMeNsq3w~<$!P>=2*t^MgF*&tBN40kiYD`fZp%(4<}PU!ee3~sNT47>bZ!~$RRFs
zbE+DH%{`sEcPAMQ?*U+OdC~?QH(uUbZ$F;4g>|=Kfe4E(k`SDnCIULtCkL9a(!J*c
zt_3Z@+jgCeEYm%if!KOu_)bX)5uo0<W-XjZ1K^e*XmtAPbA@~(wy_RNIYjVYFF8>l
zXk17))x#1|2GU^Tq95N|rc_*GH_ZgJcr&kTZNizpHhGLCm0?g^zGnQ_zPGV^pbkli
zo4rpSc~@G-*z0s8<>hXGn5Y}_SHZIn5QAXFLoh>5-f0SZA04slLCp^?ApsEdExHHz
zoxtxR<&(^^0P`_G%bQ$rr2(J&BkPkMGb>5~5T^&W6UkmV7l@TcP}j3Bsi4lNP0JE!
zi~<=r6vSIl?pxmXjy2nfvybO%u>Y-`vSJ=V-ZwxK@AUF<dR^4?DFnzX-UX<k7L2U9
z8~j}NXmMbK^)elH{9$csCx1LKvOGB!!pdCyyHcpZZW98P&5)VSh8K-ROs?@)v#UHm
zts=F$S0NqD(!f`KssbDX=pJ_7G-m{+j!;mJj=<swET!eSX292G`V)V&+d67U!BlVf
z>g(g|C(%ra8Lz)xMpzqAJWts5%&LIKl#6pGw7Y7LGCVUNXg1}==>r%p-}p5NGd0^x
zJO9X8$C}y_1fWAmhu3))0xbC&8R?B8gqj6XE$%TgAxUt=4Ma7hpbNLTO;>yP27?g=
z((<rgI98`;(I#e~f%ht(PpV6Tw+b-NXtGUXQu05$bgwN$|Kb*o+h7MF4O?lIJkBcj
z1BPN^yHcw2SvJ(tZ;+XqiHC&iWe-7Oq$~@3^oXB{I&+zN#%|R-4VU};)F_mEbO<vA
zNy5<j-h1IygU-!t3%Re;(e%~TAI;R5@Si^cKIVNLfU9427anw18^?y|w4LD90pF=a
zyzrB?4Nu=&bvVD+>30|4O*&8J?qk{ytt@vefW><MoeLhSFui&ISR7!!aDizBjBQBe
zuP{2^_%+QZ`tGIts?0l-%|An+nl=nKICKlAPY)s>f(9CHAiPmQXf=epKUF>W`)p*8
zrOo?5<Z=Edfe5Pa31C6EJekD>8{_A(_Yb&M3PeHxwtV#^Jux`ZfG74JsvZDN`!Y>+
z$H|7(`9My>+RsFCqpUHf6?T<c5U)#CV|u6tVEkL;t>*vSQ1=yjSLOIKh*~tlTqaex
zB#k}xCJbB_Mwk*cS2rA|2c4K;X8<s1U%T1pzouYW8S007p#tQ8{(X}S(DS`cxvLhT
z**WR)7~X`BuyRdImt3Er{o9BG#ptl25%~Qd=l!up1ODCg9Q{&q>)GO%LUgZ{3IV_g
z6s1_%uM#zDm`F4zvBiI~Fb+O~7G+1dCFU+M&B3-7Mw9*mZC?{`x-Cc0SJC~Ua$H*u
zckyr)G$z<ld+x1hSSsu^2;KwpH*i}!-;R+PqQrk0=FF+i4?aFx2?yAq`s#4XGC7a>
zh5G;M`tEot|M2hQIJR`MvO^gW*)t=1MD|g(%w#7sTgaAC8KGnyvSn|g%w#2-Ldhz-
zp6h&jp4apG{qggUzs~Ev&*yWW>vLW2b+d(goF$+DMOuJ=*JfZpQm*P-9m0sznLdre
z#-~XZd?9hCTwxV|Z3M;)Q-`a<h@b%u1$>}5%cAtPju8*lEmyMA;<RpD%iECOX}ezb
z+u^7AN^NH2Z=Kbuf)cNY7DLyUd9R|<;^J&a7XvgDE+}gr@6^;delD({6Far|50=|^
zKh;)>2$2MCO~PG}{L#_#(Wp`Hi{>SZ6Rn_&eAX6M_t}UU6~H2OEYdbhRcDj-Q<jN4
z9f1QnfggGrUTM<`b*gZ#@?@Uw5!=@9@>1PxqLfrcxUH1iOLwS*VSL6KwfZ#cQ*K&Y
z`qT#o!%6l{I*>2<qX%wXf`SfN0O$F1>OKAH1{1)H5bW4JkoZ;TPG6S}qs(W2wP_L0
zmM`4}?YC`hBbLGwN3CYQya)rnjtSu=AL@{4AF_+}Y~6l``PPOeXAR%$zj{UYmoy~`
zT=&D{2HoDw*M{Gi=mWBL^xb;v2T`Em+Uy|CM7!{v_uN+p{a-u!_GRx@)t-p2^!5D>
zJi!`$O2i;_R!YUl*B9b=#V_D~K#1TMMw`menXaffut?C5Yg`5o0somEU-1gDgd~d@
zSs&gbJszyAkWjTP)$u;c(o~O=QonMgSkC#*t(!au`bYEQN%=TsnxBzuL+pVW!$C(S
zHDm&G0ZLtL*jdX?_oywmrDD%Cx|<RRxKbl@cjI*#Nnd3Nn9U)?gHUI+zZ}vPt{m;M
zrce_Y%Qnh!QaE9_lKU`SG2c>zM!bLX!$FwVC+^@_oM4`H%`wu*B_W_W>n{~CU66vB
zl=q9Ta10Tt+Fp}oMY$ytXm@zLqI7WdXGwrp(Ts9Q*XbO#;!kYPg>2~P8U?!w+wc7t
zMS=EtVl7fx8EY;=oi6o<Jjqzk_T00l$#EIES8dXsWvutrIBS{26f1zU0tRWM&s@Kl
zt&M6CqKTO{y3_t=suP}+kJQ`m$~jI?NR@pIu^GA*jw`jhcgpMd<&qC!`Z+}9D9bpq
zd!^;;80v<@k%JvLGhg?B9W1+wcOCotFUz-*Y`87W=iNQzGYYTDiZ(zX+km?HL>att
zEB`BYGAuE21n9mHK)st}?z!8}A)hjCjQ2Si0<I5UB5kvl1d#s=ogxuNc1PdF2eTeH
ztWl0n6|@^YvZ6pMydC4i0T}azw1Uzah|EL;myYYC5BJB-@Z>+_ic;rk$WqM<QgzW=
z3m?bM$w>R<s7jx$XDRo>i#+OiDpSS_O)Q_c!rH`1{o$CuXz)NW_jl+YL1|)Tdb3d_
zLJ=oWg*cOtJmg=Qq}6pxLg>@Lt9coL1Ihr4&5aGl4RoNGjdw}nJ2f;mg`~4@AgnC!
z6|6)`i*CG(ugdeZvUld5_&rWb=t{yZ`yD%bZvt%aX@lQ+&mi#nvBmKE^sZG4)rrY^
zzX_g)-}abqT=-O^aZC(7POG_L%fha&UWh93b(g?%K=%@3h^nM~A9Ci~E$yWeNqDZv
z@}nNLrdteyKyMDbrT6cx`8C!TWL*fp6P{?Z*)z~iB2N+j{F0Y9TW~Y=B(ES#2{Iy&
zf?U&|l4b2Ob#%?poKN_!IE&RzM;6YuHWbeS%*c|A{qKgJovG6jruxg2nr_Q8&HHIz
zS228V=Guk51@2F`s<qVpc|y|WfA%FWjl&a_`-!A&z~_Ru`__p_u5_@F#oLf4f7{>4
zoe6iBxs~*Hv$LK@zYR`0HLClqt6mK+*4fcbt|Ovl1!j)QNp%%M{udFN5!PKCw*5D@
zhcd?S7tU=@dhgWUpBit*j0{gvLc~T%lIJg%OIrhp=c8iPOZAi!ld8l3TPBPYAChNu
z?d8EJAEp5Ow<#MCt)JcjTqQohNxZ@}CD>O8*tc)P{WXPgbsr7IRd*l0QL^Co4qL_j
zrh2)2MdwUW^W0=i1V`JJMUy+Jbl<xx#})^}xMzIP5r?fmBO#&%7!5_$3*yzztETx)
z553|_{oh!_OF(er`=rOH2X^WK;c8C<xHL|~{v<(&56+`py;U7_6`b0Xzj&Tk=dF{;
zd~WQVF+T6E{tVGST0s7NmijJ_hJ&09N41CnBxz*q*S_yh(jqiVU8lfr5byqOfwb|c
zvb8%u54R!`$Q8VB2Y}9#wHvcpujQcH1HYak*<$9zDc?~kba(`BKOte}Gb{@F(i`nH
zMXi{kSAC(9lCSCe^~=ftLJS_C8m!5>FyGQEcvBAnA(^e)#06L9dZ(Pu&S`GgfMl58
znITV0xWH!4Zjk7HwZU0@;!ELbZSDU(K|_6bB-U=J1ZBvlc)QmJ5o;>MYZN)Rzz{g~
zEr^8i^eQgf0@Rt1KBKT-8#lkX<NIFTkOl@SU{E^7`_a%>ys}?qla(YJvfu=z9xIz9
z*(q097$oF@483!t2Zako6CPK7=!uVmXg~=v5q7cMYG3&kAp7Z(r*dn8O}|Td0nX>9
zY~?aI3}ERw;~WFKL!@Ij#R!rxz>uS&G&)YJc;Cy|3vN69B$hw?Y?D(Gg5$pbIwXGB
z0gWKW1VNkr?g(AX&1?CA)#8vKCl`xoqB$~EK_G1d6Yw@Tfei!54lm`9EdFW5IfYzQ
z=W%tR(PygUDgmc@If_iuM;Lr35s%s{-ZH)?E*sAeczI|S?cRBslKRDh^d21`r_qzT
z_)8{F^Rr5|nef-B)$H11`bU!2iAfeVYnUu}P5CV2-<*BI_Imz_HtHUi!qeX=di46~
z(Yaz$H7>yvrKTe-`5Wiy)sP5qE^o?3cSc~+V(qWKpGj1*R?qj+m4>6dh`CD6(WDoo
zpU&WIimuIYdhc~K^sYa$et&EqDM+jJGmhh<e|qDb7=B=T&3wV=P{D%_D&R}@sN)4S
z6aSCZ{$)c0J?i<9Lkr!xJoD6qUBJQi7<SY5c3+T1TK1gHTA4Lxx5~Zux7i>Qz!@Ak
zNIGG*Gr@*E*?%h{5RMK#eBy7tP;Kr(L<grHQ!Cx$J6{#&R~_PIvzqUUR33P;Qfi`^
zXrpmC87r$);Dbx4*hAVL9!k}@D;}qn5rEa*yttsi9oeR?pvD7cXuNmr+xZ_8h&-<=
zB7T_PB5K|?B47e#eRIvZ+dfk-0f4@G8{Oix4JEhxN4eSz#lw*KV_hGHXDBa2j%y8D
zydPp^oyTl@ty&ksNde+=8<|Oi+z5wdY}}+5x6aYj>@@NIP+6hCP_BRa;V9}O0w*5?
zFRxL4t@D&8!BXHXXBAda6O%bD*w`;g6{&rUS;tI5vBs&~RvsGr+*E|*J;o%z<DVM%
znS*9QQIM#o;&FR0BhbB~YjLd^M7=Ms_D`P*0{qz7-R6dSCUz$-G-EW4LHRjw?fY=g
zXWg-}FJjpI^v-j#FY1X|Q_vkA`x>}L(k)=XI9>#$#7wUtr91o^VUvP>F{HReIKJh$
zPJF?YOJuh^>~UoW#{C;K7C7jrxn_jx^vl8Wjmetq2)$f7kXZY;)h<3RurHI92dz!N
zi+1?@Xtr-^TA9<NBMXa4QxT!Z{|(Vv5#u-Wq|=;WO8Y0dJK1TTL9Va=C~y$>AuvwM
z^P*DCoRw_s_dOr;I0BFhv)h*m5rK5`VIbd1UX{f<G=gQQbDhsVG5eRDi43zB3}Dx_
z4A+{8*RRrp-Waf2pyA>{;_kLGP6ZV1JM2r=`In9t84tv~Uu!Xp2hA22b-LRsN9=O%
zuJA6nJo@h-Kd@Wt4SS|{QGoIS^7R>c6FeP%|1;p5c`i~Na#NTK_&=p3!e6Zn*Nro9
zwHexekg8f${pE0Uiv&SIw)vmt4g__s8oBdC1%SI?EHRq=(>qBT1Qn<4ohN0b$LzI(
z1lY~e9X~|5-v<4MzYi~HM7X{t^{@8s?V*tiD8DiV6;aS(M-DeXyju9Nb==f$__(|U
zY>^W#s_<g1jJLb2>jP(=qeWaSLE}w+<+$R4%Gtf5`*J{0<K%CW-vrt3=;uej1dtcH
zp~dsk*}%fOOJAV`ifw+i`77YnhuInUooFZ{;jY54v(=g|Yb$Ols+zy91Z##TUfC>O
z1_~P*K7xZSE30LKt!ebI7~(lz;<K}>yS@(JY~t-$9zeIQEDV5yY}OIS)~3PcXPg$}
zEt|98a--dfk>K_p7s0&6(XHjhpTi$@uVM}&<YIcCJKI+rmw$J~<$$P)4HZ79Zv+{v
zpzi_&Onp!8nP@5!g!zdm&XvV2Y+7H|a!dvVD-ie#t)~-iJ9>0bs+WSQj*NUs84zh$
z$=JWY&B7vzSU}yF_yiM0baPQ4sNegkP-UmMcpzwo(LVe1E|Dtikf{Ue^TeVzWGVHo
zZzrGSMWo9(&y)vd-qi1XZ`ad+L6}|)?{dem5#hi?iQ-64IRa{u=L(@>;R4QxX;hp4
zpcCjL9sX#b{Te^#B~=j|eB!3s7rmHN4)_)F#Za*AWzct7+F?!jg5JZK(AyuBfH`Oe
zO(!uMv??vrKbq-xkQw;=3+tr?dK<RB#?RMZKGZyqYm<MNG`Mj!Fi_KwxdY3CW3Nnd
z|MSAY2ry@I6M_?QHU#yHj8^zuyXQge&ri4JWJZYYx^?icXz4bFmg}@L=gfvD+i_Q+
zTQJ9(x76hs#mm{UU5wnkYeEZ3XzpFvcqlx0P%{7yljUWa>lPU1`onIxc_nT0a~`|1
z5~oxO{R51~%n;eT4R*9c0uCH=4SN5h1_pp&53P%b#%Lh^pv(N-OFEcvRn&Y$z|+-?
z`9jDY6YR2JOC+uuv!}wp+Fr^ioo!@5jY9f-AaI6^8f0SF(yq1WB=U;mV6dNlq8(yJ
zM>|jpl(Op!ca^~gIlp<A<m`EEYL9?Z>!0l(B_LsxpUOYMdh3!<g=s|r>wM?H2zhqd
zc(?X%&Y=w7Ny2(*lPr9<7<BUBFV25Ewk`(QVma(vNH#5!2rk4~1no*`aar*PfKJcd
z6z3g<4h~33w?fyG_PZ<)ixLOy{<XlQVDM4+>|DbG;iJz<UO)xc{O&w-*%<#ScBl7#
zx)kFob>IWc`t(UJNS@fY@Bdw`g#vr*sPMs?z=wrAm!<j!wN#q->9!BKsNh%<)S>M!
zW~KA@oRGknNlZTtwtD+Mb!iu;INY*_tFh}I?Q^rU6w$AwcdzXFme*ianVoNV@M-<B
z7^@R_`(>*4M@s&)>jG?-A~$p?{uO4E$-R$}YVA}xET}C+6eFVKahW=Ya!6N8OVVU#
zx5h}{Ah=_CaCVD`XqNMC9lA>p^1G;fpA>dzp)UpnCY>m_xSIM<sT(6mXeW)wdBsFW
zwM0e7eeGPGsb@-^!#rtw2y;zmW&T~#%^)0p1IxS4c-cgqZP9w`<ZUFuUdkC;bdN+u
zhN-7-GYCBxS0M#Yj$5s82uxmAw0=pJfLaGUPK&mt&-Z@TC|?!g{MF3E;0lGeb4;yZ
zvP(zWP}M|c^i}pa^F!u1-cu<FQ||{4Xgy@!|JUTfoY3Zz$b#ykvCO1LPGCBQS%S@N
zGcNOxD$(}%alki>ZH}vW91+bA0SsaAP2rlk#d+_3{lMIS5prAonVMRk0tg86to63p
zISG;ki=a~nEwX$5X9#G+7A0x7K&Z*%?0CQM;^IX!JVpK=##s-Dj60MUAPWJW)3Bu~
zOIc$^@*xUrS3Fx61$Au-S~_@oG3zns3S7z98pg7*=^HO#{q(CVMD#?&M<U=pVmBN+
z<uPGZtmC>s6hZS1QOojxphHPOzetIG+eHo`3T$kWB?_%g-pbi6Wq!434H|imA6rx7
zkqq&1H!z;daJ!)EPleDatAj`qXhfhBv6PxYU8`W`zDM6N8QK<FIfg{hBG&t=$R1f4
zYk93u!l9R>F5`Un?yHOC0-yAumFX519Swn+1NYOeV^N}EJ0!uw352=hGhc97Dr(&3
z2iX{S0wTv^uf2Nxm0gDSEJ@!j5dMV}W)gqi(b1z$DJ;2cU&O=Pwd;FB$^3fX>DY(R
z$oUv^H}-_E|3DL_InuO?slj@h=pH6YV*%T|;mlCu5Ir(99e$ynO>|ypj)9q>dSPjP
z|N7k$Sd^@SH>HCBqv?!|7}-~g8iMM95Qc<r4r!+)d06984OXu#VB9@+V-hngVrQB_
zRZr8bZj@ya-h}BZfhZoFg<-imQ)ndp8CPAQ|Cxl|s1$0HBSd(pU1rqRG2V&Izr=zf
z$+^drlDf2Y&mZLIu!8@V5Dy%J!XgN(ScEl7VFNE#5Pv*4^l8qAPzO{SI(Vn|&vW@U
zDqdm~CG>9<m6b??6ckAIZ9jdY_T|wajjesnpo`Hrf@MJi?7Y-in1>}+ju7{y5*0OQ
zYtkqiSP}$hvkF82_<PyRad5pMPH7Pgr%8{}(c@79aX`L;AZd&SE48|gjxiD-_mfcT
zC=Tr2T+@m5G7a@b@m6Av{JaENd~!unAMz(WR#OnJ>xfdbBT*;HUt_&=jxrh1BEd4^
zuUEefX%)CJh$8-G3<Tq;rNp-D$!r-j2an+8OOY2+aF5{OGi&7HzdLaZkY-lW#7ia{
zW<n>EJvLQP6XPDhQ{&g1A_v2e?Pz<l0e6%ho34iH2MT^?Qlfw5Q=)|vmP%|C`NAPs
z^%EiwMIzEBa2HX6qGe!>Wk7H|lE>fsB=zcsgr3GBVr*_YR|t_`r<5)wFKenDk%OjQ
zdBvAY%_^B=Nuua+oel3g3pY(zVgSxXGC*XdlyeV{j<`J!-Gos%dxM`_?z}t&*D~zU
z|NmP@{UzdW<4e}Jsh>e05FGcFjqlr7-M5vx<6#T`5JCb%LVN<kd;*dNf|61qqEf<=
qyaEDJ0s>^2V$T2L1J15Cj&{EP`v?5m<452F2z6y`r7{J}(EkJXLX}(q

literal 0
HcmV?d00001

diff --git a/docs_new/logos/vllm.png b/docs_new/logos/vllm.png
new file mode 100644
index 0000000000000000000000000000000000000000..1ead9972879c29e17160f1d3ead150482467b1e8
GIT binary patch
literal 89976
zcmeFaWn5Hi_dg6HD4_y^q9QP&gpx}4C?e9`Eh^I8ISL|@2GU_rA|c(3prTSrcS%YO
z-SAv{XwUhb`~UKJ;kiHOb8q(SeZ}haUF+J%S58Lk7{OTrEG(>J;<s)nU||vFV__ZA
z!aEG^)UozI!NS6(Hxdz%GZ7IJu`so;R<zVNxO?r6#qGNaV%M+m^73F|F+4TU(^I&`
z#PC)Bgr45l8pcZm)(#4fANwoldDNo5z(2uj;a}m=VOm<fr-^$juyEMUo^&nxg2H>v
zI`*Kt{6!^UzJh|^bL1Q~yjF&V;j+G-UNu%`NR=c3Vw?)DdGQiOwGP&8a%@VxI~Ld<
z8qSZC>Twb;P-a8~!wuo9hk|QDaqh7r$NMFHGH?;5NhNUjNZ7O37w@oUN*=w6kN2kY
zb%*0gC@@MOvTfY2-Ic%ReAD;A1-7>eCk1Ao&D)asANBV?a~jciTz;LG_*{(tZA0SQ
zc%H<O7e1nT%U&RngGAS>--aUfk;e}m+Enrii#UJip{LNHic;Qh@uGL-3{Ho8+Mnn@
zgmvCkLN!)GxCc<~#$7e>d(zTam%%k277Y6Y)**0(4gQDaV}yl+xyHhx1OJbO1&f32
z+#$@5JG66sNDKPWSYt>HJc4Vaq-LuoEyZVGVaB2lQJ=-p%o2J4OTdv2T$<gr)rUKp
znVQ@1ISSHXp5O!5&}~*4IOY*s6G0j^X*sxvh4o!H7YiE;8;uYF91a(-zH^UH;l|DF
z*TG+cH1}<7E%{hk9UL539N1YbtPNQKt9e=3uCiXe$_$=hwsA7I)pukzx4E!2$j&%7
z?%Ehw8(G>KS(wA2arJLo*x3rw&_EOI{<F1CTcdk>Gnw0LZwqXY75ats3JV+SzrA+X
z(dhqp4f<v4HO#)YrW1e$<CC>Ex(gNzElcQ%0A_*%_x6@20FA;YXXJR-RPBb5*<EuR
z=pA-8HZ}p)fBorSUsd0Gm51xVoBMv*ep7%IBGf*@ZE*#27qCT$K!A0Z#X<x<eZLH`
zun<_{H?Ao;VvlzbW`{Jn!O=FyUz2=*NxyG@b>(BD|1mCcvFv-I?*ghc*p=kHE=F@j
zzioIsWtAh%ZkCwN;dbY$Swn(&Km?a$)nKV3&IPg?@`p)~^?EOED6w6W>xKv6Eq1mI
zKQfV@8gm`)oE2P{djGj_(Y42YZL`d6DsOqR%`jHs+e0jD7!DpW9O3!*;x7^YO2S_)
z_^SnfwcxK7{MCZLTJTp3{%XNrE%>Vif3@JR7W~zMzgqBD3;t@sUoH5n1%I{RuNM5(
zg1=huR}21X!Cx)-|CtsHNX12EIeJSTtY^L^S1WOz5C8QgAd=%=b=cI@#6<LI1l^0O
z!K{jw>AoKIzSi)*u`_+G6y=MCULFP&vmg4F!_f2D==6m}PYE6!eBD*5KCKC-`4_`E
zQ>(E)iX}~Q31>?e8cUmNz0(EhgIY=#sD<5UR%48F+FdfGCbVQzCoIQrOsq=8E10CD
z*Tlrcl;+C398Fgo`xP=J($+1Pm|mt)q8QT>xR_)2eR^ZK<RgQvda91)N8BFAhRAN)
zq*D)%K*iYs5tOT8X;r1{I|h7W_>X*4q21V*7^k6d!&n}x{wPZi#Z~8p)(-}iJ}yyK
zujaGLnQJJ@r%P9h(Q9*x8}nUb-qr@?Wew$mO=0b2Xt$nqouyLyx)z5`^3nxC<KdOE
zsm-(B=X+M?Jtk743Gc2>wi)-deD@NYph}%`XNz(f)OoLxZNGMBxzyz&@suyR*nU%$
zU#BRP(C2cctkIkFAj?vfQv0$c2^acRhNj7rBGDW6tCJ)m{BHBmGAf=An^YPnu1ub<
zSakcaNKNRo+CJ(*HA%Yqy*aR@)Tw5=)L<P~hJPZvTcbZGJygs3+auzRX}NBwG+l7Q
zaiPtyg^JI1BvbK0ns(Jn+Hl{PFFM_xa<XVBwwP;SeSLC(vpBu;+muPm!bZWX%@wVb
z9%1e|XeZz2^PIltl%<bfnrLRpTc4QMZZR2_iJ9_CR_T3F5SK1>KhL8(IV5?oezCY&
zcBz!?d(fTI(V~r(6~4QrqrxgCc^0>Md3&Oqi%%)^yb(@&F1q@?QD&}XG}dCJ&PUPB
zraHOO#6~Jr#w=#kLn=mK`J4E2I?ANaz1Oy-@3DH^!s}cN7x$IegGrIhnz_Zpzo<8t
z-bHyuhO<>%_hP)x7Icpej)a~=1;d7l&O03zG%BTd@dY2qz#SEMwUg>2m*UOPHZ0tQ
zC#Xc>_0b@2`pu>9*gfDC_z{>`5LF5K5cLx3t`8Bl5NCTk#an@^mn}QY^0X<hyRg0K
z!}r@ZRzI#HTS`RA?h1}Q{=<AFX~M^cX5H*m6_A);2Bc7~p;_3ZERY;tJcVb|?e!@{
z*)$oUs@l@i#MNeIl*o4h;fZ(%^CM^7NNHg*x5yA}0#dr&PLO>Q1zT+98S_3k4~Hn+
z)i{;%<%wI5f;O1xlM)D^@!ww5E0sKKQ7So<a0>igPljg`gWF(q96{-pu3P3l{#;hb
zNs>~Tj22Az4%R2lYcW1K853em%$KeimzY@Db7f3LackYg46i+LCq>UG^OEc=1zHnA
zANO*|_pow0`YT{QVz{LO%5_)_^U7@!w&6jOw^IefF&z~T<QE5*&vN}d@mNJ~v_8J9
z%w=)%{%j}@v`Wl4rMOI%b&E^BF$eZVLyA|l>U3T(^GYa0u-jy8J)e1gThoDwqhR)y
zbuIbn+C>ZFi80Q<S6oGzHK0fEEWVuSX=_`ixH80IxIH`;@;07Yua3Xf3*4O<G4tG@
zkz)yEwhX%SK3?1t3mJx6?QkWiv=fk<I1~{!TdWhAbv~BcqB~l@FUhK^EgwG=VG0Wp
zS{bR#HS0K~***g(x3yq~U{8uE*s%nq1Dzjm0;DyWIv4M^3`Uwfx=7<d@bDX^-oj|C
zR@YE0>5~wTn0gf@6{tLMy3#dres@l}4)d$hCcT6Br$Iy9?PcSvso9~M`<6X=R|ber
z(_#~Lis7qd9gG(AIB=NUPajh@I{|T~8K>Gooq}W<dJV-u;HcQb&>wiSp>Ok2e1_#*
zT83w1Yrx?QE_~wqYI+R`(pja`0TyaRjTv|NjwxL}IV;IG=e)hUt8~4>*=Hz-`}S4<
z$pG>dj;rgs9A7OY5eRuOo#+{HC(Lw~u%Y5UuS_TR&reQTv}GCAd3?KP`<ey;zlTH*
zMM{W?Rg&N43g+M81>y=0lO^jHp6?G1a-Rf`|8qg;zjDWumo@g6TmL8D`pyEIscb(N
zp-S0CoF`1^@vORyfe22MS{Ca-YoJM6BR&>E`Sn}*XEE(*5?E*)9A@yv--<O>kE!A{
z3H3xIYKtYbKyB#usSVq3d~%Ok&*n<{CZAnTshe=U8ly2Oc;5+@!`)@;;NX$$#BA)l
zyLf~%OovRiPN(i5f8xxM?UhGcw8e2TVj0YiJ*!=NSSlZSDp(B_6&l7e|D-aIlF`o2
zt}DbD4ud!+j_fLOC>z^Yo$_EO-VtYPm?$y&wIj(*`QrzsAb#!{{ng;CSpkp$xDFV;
z*nrgry!44queNIXagWPXNvOB?q@s1ygCF?B48f?i<=HYvKt<SHGQ-P*=*R3f>R?|O
zqUkW}Ay-ob1$Jq;bzhwv4HIh?yX6W^{NpH?gPkk~W1XFdfx_#*IP5!(?yHCo3=CvF
zYTd36X3v(=+uT@JG`1)-QupDb+2IMC?K|vLe$ds7p#x0uM6lz&YkKFhj1w0k@%i;r
z9P@X8Rd8l7(^V(%FLw?^t^^<e((6NWCHH^6HVHb%I};nx5TOG7BBl538SN&$mftgF
zl^y#G<nu6YF+<G9Ya}Ojgl>ylr=?J0jmPPJVJs1uILcM^?avp4&@)GA^=c_3=&6Ux
z)4s)Ln9S%7T+om|0hhah4ApTSJu*}m>tR4*^PF^NJVY_rO(p-rRr=ixf(zt1g4S?O
zpNDjJAl@Zuf(f|82{0U2#J16qIWozq@8ca$4czgMf^^>TOJFz+8uSklqV?|X5*N+-
zP#LhP?GdvjfbmxmAM@FjG|UCZwYa>zmJ(OebDS7Pw4Gw>x$EM3Lk!|$4Jq3&XWiAB
zm-Z9w${GdMgVYIU9+u)0KURcSUogze%iGk5Y9nVK1JRWA753*Rx3?1~gw6V-WjLN=
zXM0;y-qJ)7p2NQhFK(@5Mn+x+`?w{Kvg5}tU1q_K!2GET<apyW2kiqO0J_I*ro9q4
z5=g+_{pEW5e5f~?COYzhGO}+&6xiC;%mqE9i~ARew|lps-{R=>rXR*0f~fZR8p@Tt
zf}!duv<72fS{|&O6WCmvjbEKA+jJm&7RZ4|{1u6CRIC~8tKp3Oel6+x#a$lA<pIBC
z_dI#`#@0nWDITj<L8xqyE=Eaz&;arBi?b7fNR~_(HmgAWAmQ+j?(Pj=wuE01?MYw2
zn%EwrTFN%&iA=kT+j<C&Ok0I*5d}zR4{YO2<kU2&-~n>opp#2h8lNO@3zG~5k6s}N
zo4S3@ju&VcTm~@z!<Ti?G02<>?0(CWkPuU_i`NK8_L?xK*&#O}zYo6qEDAfYkQR?#
zbM2B1bAj-~*;19}QvnwdicnB+2~6^=FLLTxL9!?iKMve#C!^m%GClg_i)DC(vUR7U
zn{y<<4^#L8>vIh)LJO@1hSLl&2ws6Fp&|Y|!_N~s$``w{bME|okT=^#75*c`Q`Gbr
z>Af^$O(b_{bGq&(VK|`_0tb%_NQiGjeOsH8q5g?yc)LVFVuNi^V|_jkv2{nPfzB#p
z`%!1|UG0Z#1M(i&K}2#g^VTzQO|aRanpG_|Prlry_1{5QNFE&f$<Ad1A^|iC*Xj<V
z<jz<~H4sAt*$oUz@2n5QYNwNEk{l;WUMtA3Ap}7j=$!om@@z-}z~>nO=ZBw4t$uy^
z5Xog0!THt52N<JZHF&X8f`G@yn#X&U6yFXxjO1-moC@rqu7bgL?-oV~qR8<0+Y3Ww
z(GA2pb6NK<ZYzf*i1wtqT|=v-q3yiAq*p7c=wZ7W!|yOTp@05&*v|>kF~~eyqI*C3
zO%_d+I(&Ya=elbT`$>V7cQs%!*xBPQB@bSl7Dd{PHAm;%t$G<t^_hi3+7sag^E<`7
z5@}EwC>rJ@aLI?etS#w6Oy5h$z<`~wc>%Q$>E!`3vVW+n486F4M~^=VhBg~F3W$~P
z+&CC|<mzSMLPk&=7v6g;hk6t+j`vS;F5lBLeu#yv@2l{s)6KVh#a+9kzRT6e*HJaL
zWED)}2iDH;1d*d*$Ey3{;|oj>fAq56PDD=35a!97u6Jl9!d4|+XZ?#=N3OI-jlL|z
zkXw4oGlL$ky(r`jJLNqO63h*()Ues7Pk$t*c#X`E<~>o%i`cTY+(5m%Z%P$ITbHv%
zq7Oe(aB*1;tzem)6!#Nqow)#fQ6HbRjG^H$YoIlE#GVZgJ_BMcc6Kt@`7OhNkpe)F
z3c?o^21^PO7%@h=5{6C92)e(9-a|eGa&^-@q~#f-%6rGz+Gu{(DwDN-Is}{oMkKV<
zYw^fLoa+I}UL=Y<1%lIV*d6VCDsgZ!jJD`uVW+no_f&W*A}A;r4{Su&UGhbmfOlGW
z*(z`?;cjA~WBoxI9Vy*AO9TSP*h@q=x-u4hfXR4`jJ-OgI#&iEIfDlQCJO{cgl!2O
z5*zm=v*yAD)?#!y|6zeqIAHxF616h-^*Zym%)4O<GUTGD#3Hi07YDafGLFl=q7QE3
zHtmrNv-Z@J40*Xlx99F97T@-N!vw6j1keS&{2B-RA4-6hh!tSS=PLCYFE}m_hDs8i
zVp^(Ca(D|2J^?-PMv!pAn>W8~yd|Zs?^%jRBpE2T)9}Py_1N3!0m>mRz!tk4dIpai
zKbN2Ml&g^z0T+TS)aKlyV-&UitTtzMVjR3rEWr7U3uDh5SUpk;-%C5PvufD*6AnhQ
z#6fT?-q;9<08)SeSPdTIZ?}c42X2`Gt17Fa36%RbSyBy&wzjszr?8@^y%jpjch(&w
z{(+@#{dD>+Qq=F1cAjBf%(pk40U#b?c#K+@=wS6&>hq~%zQtF+t5tBhP{^NIU6MMe
zxb-KKlr9h2UifM_)npu^DdZ#5h=X?nD3>G`dZ_4*xiJXR;nKkAM69Gvt@eI=7!PcP
zTf)5H*i$TIF}_z<M^#r>moC9(Wu4|oznNUzIjUn^ulTlYLmV-PZOCt*rrk-y_ttYz
zP}RxRGW0%<XIM4`R;yDpOq1MFhr6~{bcC{KuUxp~^U7t<7$oDX^~Rhn3EZO<L>a_7
ztWVT(ab`X%88a6zDO=B=Das4O24CRJd`GP{uP|vaGi&9egKQ!_Q6PcwMfHF7+gun7
z)ioE_<=a!#U{PQm46T$t-8e|Bf5?qD*~stTE3F&rtMsR?@`{rH&xU$>LCoFq=WEdx
zuFl}Ydv6S=ff(!@Q4ENm_YnaufwX?1&K+hOuh7>V_v$wg+WQ|o$LwL2v)V8YW$=1%
zDTr9tuiSUS0k)q>aAo98S=UUU-7QZ93(iX>^%;7f<CNb;-#^%kT-Y?A@c89>`vWum
z%T7>OFiJ|w*zt)8N@3I`AE0V#aSNlzqoboY1#iQWcZHpJgfs)yMFiV#@IlNG>glrV
z8W4IPng$;3xy%wCsR1cnurLspjW6#nkyv!(rp}=&h&G)S&o%&W?oFarWItgs(VmmS
zwtVZYJJLAkcZ2qCL&fd04-9dh!eL~Va`FF*?ro>SKz1GLt=-uZJi#C;=7~tbF>ZN?
z9zDL9qP#x37v~~gLioVU$m4*6%2a$42x>a-z$hNhxGM+jqA0ZP^T60=eq`&RwNZo;
z03)EV^^LM+Qx<Mxn(<=PZgUq##b{tKoJ=^n)gWlSyg5}fi}=auJs!V>tio>*IM$xD
z?pdL|$A}-x$CZ?<r&pxa7#LJ**W%!<`5^L-M>RJ$FSKl0eR9})ssB_4N^o*I^k96r
zUxFaMUc&d1?Lj-6Ni3+3+Jl_jfZAhix=r0z>>M!V7tjOGWj9(Z1$iz+ch7?mZkvs8
z7ubg0vyv+$BnSC=S_#$fbIm9Ul2RlLKWvJCBTQjiHy)kLU!5!}J8f@4#cp~H7mt|X
zEn;as-AmRcza;pZ_&al?wPWK>o(hx_8&IFMH%?+NAl;hXF$WcP#h{pN&iF(fM3*>E
z;7nl0-@SnT0N}?l3a+U68VmQ?b@_{oAP~@@M|axPX%{%p|K={8^g6UhAR353{U1Yj
zEY8;U24vx;y}xvXJSymJL%c{*1o;zvSU>b$uqF^sbDy+PWL-4>Vp=;(t+a*OLjde)
zz}LD%iwX-T@pRcK6MOggm92p~W}WhR9+MaCO}+QP4nJf_90#6?(Fx90v#Z8hf6W7y
z6rSNQ)h+pwRoC4LCs|@I_(Ny|C2jc}&vuab`%J?&qwc36%oc|h78Y1r>}|}=t28bk
z;8ieSb7O_w*BoDpkWuW8fK&o8rmVMi;=wh-KX{TClKxOmQ8vf##JJEWu(S?*^dBA-
z&Q9`PtH62&@*j0l2oSgEDd7g5<Gt!M-T%2kEQd5;&meqF2F`?+l1G2<)<0Jc?jWpi
zoSmH|fR)+2L%!JWpDYE^W6P<M#SJI50U;%lVHyCtY306TEVl~1y0s6qGxp{s=Mvuw
z2RE@RwmW3;_G~fHd=>RZ2!tK>)1_{!2}2BbUY&pkBjCkP?<xRT)aQhf@*doYAOqrl
z;xn7%vmFJ%j6`q|gfxhUmmo-u7>K?rH$wpF*3^inxPjEdI>sg@MIhc2rMg={!Vm(P
z+;4&|3vv7Qt4EB@dg3VT*PpzUJBI`%2NoHwt6S=G2n2hF?X6y57BC<UXL2nW2OgDt
zklB3t1f1<E2rpk+0a}%DX4XsW!KIHCf#=|<)NcMyio9i#J0A2x<cHf~+(0Tpjo-j8
zxgHiF1+axQv0B!>8nM##2@CF5w9;hqsLf+fc-~oWhr`c65E*krn)0B{3V!BE&ZF?4
zg&E>XHlRz3f_e1zn$D1^6+2Gn=)YPey359K9(X-ZL~DtgDF7S|Dehla-Wy3Cj7{|q
zn*URAW@s!&e!QxGfJ?~qpr*>x_c;m=m&?4<@fWGAz!iJ1&nNf5B;r&cdOwesd^+ge
z5CX6!<D7G{xmPhMni@c4KLi#RLl4Cz!Ss=<%=&r5)C|V%Gc?4Jh$Xi)fWGxmX?E`+
zvGCv<fX)|<fA47X)-sVVaLUU$7iTWDjlF~zr$Mh*Y7(6j1N{ou2BxHNLw(9^s=FjU
z<McvZCRO4RkfU%E@D1r2fb3pRLsXvZN*tmS0DM6%Yj<V;l;?fo1pRd8WSFqWhFE!8
z9Vz?Sm<%xFE5wo@i-Wy=_m%SgC;Nm$2P}HT0|)Eis&{R|&IRHWu8w1$cReVM<;dQ<
z!gYlN&Ss4y*LAh}y}sBZe0PBJg}*ZJi64eoEVlWFN(Lq>_Fdn;*b9=tU@Ihg<q?3{
z43v_UirHTL2lg2oTMF!{JOB8GtGzqTDUd>@pTvv(g8mFZqZX=#s9m~XA+O;SGYH}N
zSNx}pLnJptT6nB>#GhnTS-aV8H4=-3yaQWZm?&a%Jy8)vpOOj{EVenv1LgHLop%rB
z_5YlKmt<soe61nFZ<WIOL>e1DF^?#m$$T^AjaX2O^(SD(cWHZ+9$=f6*K7aV?CII!
ze1FsMUd4|t0|d+U7lH(4e!d>0rG|-;<5@^K{Q_cs@1&sY@*WrO=K`UwIn_k9yLimS
zX>EXSoLG89PFg@~xsB4~o71DXduX>n8&HzG6$cU?iwai{=fLZyE+a>9PqFH(eOq5$
z6q~}*8vshtw2uB-@M1eWBwzv&3i61m@xD#%?rz_O+)-pDwMX%c#pZ%VmfzC=COk4-
zc=7qWt{~Ah`&_5=Xz%?9RxsqObk42+6lvsBJjW9IixS_9X^8EuHY+pkFJ!A8UwVXv
ztR@hc`yo;`w>Dd6$R&i_Bg)rI5DngOvi)@PKSu_Xk|B7nnKxNYAL}}^jsTHqF2eD`
z_tA-o36Bs=bhsmD@FF15h7P7T{uuG_2}@St;O}gemqR3+!f0e5_9AjY?uP%RVyE5$
z1YZPMU^DE_%M>%e8GkM?I`|h!YrfSl=AM=2ytSAw>;o9ngJInLOwUaz?lVd@dt}-Z
zqJ3mW=w;(HzgK7P3pb1@xR>4D0;*L3_+H#`{`f_C9v?oaGzR{^DdoXz3YrzDX>M+=
zN3NgCmm}hiKM4fp8x>X40hz|{V_%fnrDz;JAPG0N{I<{Ap;VkQOsCj9B-<TYHpu^G
zwIfzmK=}*{sRE*6jb!cjrBu!%uLyH3dfc5pWIj8FTqE=vb1ulrnhjhtq`|<iTPAIk
z0JxX5vhIWC8Vg_-b8~Y`6t@VY>@W%2C_wM(=UcXFJNM%O+QXa!&}n-s*1SwzH(;%@
zQGt~F-8t5S-?@pjzpL+PIiU9=Qgr|1|I|vbc7J>(-O@EP*3zZ^_%no(JUFjUCYAKP
zEh<`FXs3`9*vF##QoyW!S!;y0jN#}X_Ia1D%;Y>E3`++4mMO~5i_-DIISfJ?9;dNq
zw36<osOQI_n%edd|3?bGZ`$o=GtbLuO1S`!n0NQWS$X2eno#Z2W43F`o!cCzjT`dx
zk0^fbEG!NSu#jAjIDU5`%*8YwD95o{UOa7XzKxtfVc$1&!ESn!Z%>r&(dSM6@<9U{
z_X4xkc<ToU-6Fprt=o=s^#XDCee20`Z^E9~aL=p6Uq6%UMj;uzb|(o|#rDAeI`-8!
za%Uks``mp5bAe0F12A_<Rb4l`k6ETnC7_nH_#2qk4LO6<Bmg*GfC0E`>EXj8L|pYZ
zUi}6&lK6Os;$M{B-^H))k~#Hc2C5{d+j%bzI5^4ngGfG{do>X`se%1JPr8l9he`lo
z#e-bC{{&qy39`GpJBy!>&z<u^-yIr%O^cHG`*bi8oE`Qe{7sHz?|ciu97-D}e&FwO
zQ9IA=-|c5|$1%;Gl`2&&tIw1Dh7kf4-h||}c!XVblHCs?K%xGmV=wZE`@A=~&Ml1+
z%lYnw+&tw(>Bt4f0|*r}w7d9DsDQtJbYKZ6Uv*Z|!vUB)h|>gx?pPoksdo6*TjTJp
zvk?SvhzNj&k&Kl&Rx#(T<&lI2X7mCuLQ?W2)AB5yV(3+*5^lNOnm<S%WaH1?_`*j7
z%nFlN_9X*-{M{rOZU};sx;9UIsMfGc@qb=C#h-(iQ<W$a?aYx`JSLF`6sJP}!^GRd
z%K%nFh`>x*(xPYQ8e}zHY!s|!2EN6QHYD~?#w>8*P&nMBu=^yqk-3eR$q+2b=2CCf
zGY@Wh|GZzb5d~U8S6mjxgYNNteetj($3)c#WFuvv)~5^Ui31RhBHzMlYisMoo}=+j
z9E;7JF6wihsJV3^L&0ljFM9U@4(qdEg$dsNscO4-uEN>O+OumfDJCef6HD9^0l-cn
zGgF};e-<`@KLUwUx?E$SQv?A2EP@vbBFV+0(iWGsJ=W(>t#Q9i+Rj}7Kgxh3Uq4a0
zPwO$;^M+tHq4vCtH;kBuhY@nNo_Wb7Og1ih60ZnWCSX__)x%`$Uh~xigfsDjRiQo2
z{<+!LJW;){zJQB^fQ9@=SV?-q!`Tft-1240$aIIU?wY65$Dqo-h;rn9^6gok9TK0`
zL5<~9P?Ws;G&kBX1=YfI?W&)v6586^Z^)~;x(uZRIp2f|srKWo(*`LjsfBUjW<v$3
zHnqHowzD<dn^OYAaQhu0L7L((JrvLVBEoTSRy`i<#hD6xUQCcHO3c=CQlD&aYY8{6
z!N-rg-dPxLi#_X|K-e(AUMHYPKxnZk2Gx6jfo`<7fEp(#`%J#jNU}zqVZT>R*fQ)k
zDk>KhQ&AVP+Sh+HQEH26$ip~miAmT3%73_)KZQ6h2d1jyG-yaK9nsh>W7DkJ`gwix
zk?6C;VJ6eQ6JNbiY1&0B8o8!zK|?gHr*L@9JM-HoCpNQiwMTdR3g9$=z8B7#CLd%h
zC><H|n|zKa!Xz4SD&Qvt444%Gfj+UYS@nNfXE&@}lNdiPGPwDPR91=0v=#mF)ps6%
zS6xM>yU(Y@@&o6Q`5pJY-THh>k4Lk*;RJ)8vBGYl0u)JMi)7Pw|0{y6Aebx7{dA1G
z6V0!GZP0|s#r+8avg8#My)yiR!4?4Pl<SQT)#0-M+9TYg$L^gh1Xx@j#cShk<>{s4
z&_xyJFS5tQ!@wZ9$(^O&)sBC^fcwFIRR($JfAK$~TXox1-S^UXultvwnn=Stv{Y8a
z3=dGi3vOIwvSz)0S^8U_+muVso8c&%+UD)r?sg=U2(m<Xf=6JL|CHE$ci_}8&@|DI
z9_79=x{Sva&oEWGI>`@Ga|4CnI(aw*v=G!{Q&?1Vq&>A*&MA;19Aw{xgD*9OG+06>
znuQdyNnQ}f43dy3?idF-N5EA=;ZG;_z|$=;EoK`q#F*mSKzcv3eYd`dh`SF<>k0@M
z@xmv~)pqEz8>29ZZ_Q~*)8^N&ddV!TKF)MkERGX4)||RJuU%xHlaO_8&bmoiRZ!4r
zX24<qu5^A^bOz)ASI-@f2tKf3(0AcS-B0Nf4x&#u9grlcH_WOSRb!(;Vmll;zG*m8
zGW~@@SgS*{)xZ)H-5oE=7dadC^$e_nz^msKcQ&ZXf6KtVa2frP$Q><`Glqs=BC>Z4
z?mxblusG&;Mz*TDIQ|g=|AWE2BuNugt~H!30+G)x74dDF-YE|YE_T`o$4CtG(aI4c
z#i*PHL^^WK7;kX}BwhiwlN_r3RGzgEC)_7U0G@NxBNscc(*sTp^DO)qET`)suz<_2
z{C&cW9YHEjW*?I_(XS)JNgPf3H5swg#hVHeomt*v9!ljgf-d%EiwnP#IMU>Ij1e~M
z8Vcqhe0Xne%$5*+xW$ikm(IV0LuLU7dp4SSt2aQo_q893^V}20v{nj}jSJsgK+BP0
zjK{{ZL*U!@An+*G=<V5D@A3FOT}dtct4Zva+d{2F=`6Z2S&{m&s-7)7k%h#}y|+JU
zo429IU{WAVHdp1!{}iT!q(!`im52;#cZfRHM)BHPQNu<*`I5#@PENAAh;$O)k|8;b
zfUngmL2V+#Sc$I%A+zhjqvhc82>bL*p`euZzA{$p$FBcbBpT4{eJ07GPGwIf6fwng
zlY1pzGIV1EP>_YHOD9_<P7S7+XC9pkkX0MM5ak(D_az^Xr6o`jDP?maG~Sh2fN6MA
zc8U$XCHvF7{=v;VO7L+y*W19tQ`g*g496QQ>RC4j<*p8P5`Q?}03vR|_eh_^P}(`K
z_DtdmyM@q-`31)+zQ$b?8j%gcSuy^(69-Cb*f4o~kS!=xFhi054Mx3EKl>3{8BatY
zu2xYbk0U_6$D*#@mB{gxK9!aX#h4`!Dv)96EB+Y8fCoa`U;V(RZVt=e=L?mVi4iE6
z>s~BfoYXqZdVfcRke!e}n|vgNIv_;E1)$LJ)dTA-@Mp+dApClIG@An>&;X@!7E@pu
zzq~dBSyG#ecMX_r2EIP6e=X{_!PUGrC>=+IL_Z2H6SnHVA*f8Cz0j;_v3?`JmWt8#
z*|Xhf@2kHjIsVmumWmIo0vIObO`HVTSoaoOa;+b?NWIfno6_CwzCcBER@c`jF8=UD
zO?DM|SVAVg4MFPt&Uvz6_+|I{q{F5NrFg(dfVf>Yx%5vo-2-cmZxUN5<o{XFKPa>y
zm|7;m8-o$mMR1NaEc@oqKc(Waj3YZWxgk2{(tH8Ah6iO<iaD|Z=%pA`bvXMb3zM~n
z?_R$xQU#Rlx-T28|7qis@#hv6+LswP-6%0>cM2eI>a%7dNX+hN_X<uq)fv)BN4%J4
zQw*^9_yAYf0kk|!$#UWWvyg;>DsQ#1!kZOmj-?OkONX(&B0H78sb9t4e11nx5si>M
z39dc+L!SPiE@BARdCb=m^t7#3+^$Y<RHV<%EOMA)N!2QtA-Y|qS&DO$2L1IyP-iai
zGEd0@URV2{VIO)VTiL2=m}#e-H)g77wJS(?0EiFL3y%LEm`E6od4Auc9^SF{S+^w$
zCbjta`8_VU!m1aWwB>POxTo;SNDXs=O?O8szGut<v1f>R;b=m~?{{zapk^c;)C7QY
zL;6p`9^7A?4l*OSek^DHbwI}2()2G~yrSc*l={;Qeu7X(6i^M#{Ii22lFn>MA8Pld
z;j-E2siXFMg}?f0u1i+T+Zx*XCWcwIGM7c&UM$=RJMY{?QA_=c;8qZ-t+5sVM9^Li
zc|!i^@B=G<AV6nMg321j>77nIL|i^hzojoI8hG>1(i!;S-qG(i+17f-s_8i5@nHqz
zcx8l%cDA;zEQ`JzmyqrAHUKE0Hc_yN*sIh2zb4eor$(gfdv-nd^TvSqS^&^W>R0~O
zeL}12bmFuG)XF;5Q|vU``s({9+sSs5bZy<DK)Pz)pLYEEy6<*TDKf}+?^LS(Fn#+t
zKwJbemr0DfJ|{GI(f-N#St5QF53X@7vy|2IAyG_)h$YIT?3|n~^*I5t2;GT7UXbO)
zrfZ9dX^8Re$PwtmBeSIMk&-ssoqyjPTcOI9m%E2UCmXXCV3Jla)K-o2;v=FwuqmXH
z%4J;S%u(rEI2v^5YW6y<GwwumIqxg{9XQ-p<`V1pr;6#XQYD~wb^ZTB2T_ZlUCcOq
ztMi1R&~rI$<N}j0w8~pFavg3<Ly;C6Z5Qnf-jM=m?|SRUmFzI<myAevBsxp?c-bL`
zdEvzzr3erUADQTmsg=E+pPu!Sk<z(j+q;}0MaRDOKla#icOn0N0orC26QtB_KA%(_
zH`DR^Y+?jXkr=wJkLi|ua18(U7=?y`8hP=+$0O9P9cbEL3?ylxtfhm@7V6|;xtY6e
z`RhB7qp4G#vHD*`t$^v1_Zzncw6aWA+yd=5i)k6p=fm`$@Jh#7!e)o-Hm}Rd%J$7j
zMEv;H)sZ4Tma<nMj5`CtY9jI!2et4Y0u!u`lw>Km(*~l7pnF`Xqd2+l>nmA#rfoB&
zN4;opV9caE)&pM$K+$!ogn5kZ7iGz6*Obt?v{0gZ0#bdu8@K@n7%a|1Ceb~~fhGo0
z7_V@t(j4vq-DfBTj9W5b62UNz1;DVvB=)o$p9h#f$<M~POfE~be(vIX!1GA+<+V7F
z+Oq~x@_*@ni%QMwUa{$OHsW^?CfVIiGGYD|X*WEaeL}dCaUa;E2;SUrEAvl@EJfp)
zflzWSrggsjL5ga+&?0E?FS=UZ7=P{+3VG-V9n_d{h$KNs?>hwW2l-HTk{@%-+;}Km
zVBXS`dUkhaQP9m(!f@dd(ZRj`V<D$OAR@6gGv1bEL6PlA8lU0flAM~Fn(J32dwq6R
z&JhWu1nsam)uX!vG_aRBuXAR)JyF{~yV&~81>!HLc?aA3P{$+WN5JU|*kaVNFw6fH
z$A+V1*FMR1T7ASNznF4PJoN~bV=ptm{kWs*#Zd`%1tgj-&<bEuWt2uM>5{9U!fiA9
zE+wsS*pB|(UZw+TZ@80}K5o^U_Nl`CXn=~&a|btDjbRGv3Mbfl;&IY7!2&txdUfKj
zloguz1Xyi3X**T_BuGAux{IBl;_KtXg5^r8X<Exq(KiF}*r|C2Gh9~b%jc(CQHGRI
z&1}nG{&2?jU^6NypgOoyfYMD4Qv6iJKtwY%dZkrWw|t*t-pRfhmlxjEX<2bBHk`vq
zfjyE*OS3AJHMS6RCFPN`T)`*7gZ0%=7<6=W2owHtfelAH=!_PQrLFqVxp2%Eb`7+0
z?CMSMAy7#j&<pnZZ^sGjA_-m**;FjA&5&wDgT@t2C#V+;G_L4gzGt8LL|botWu99I
zAl?QB4kLA&36bV2Qu!>UjhmfY85L%`>cpURfbb3c@c$5``{4+H7?6TY*FxRHX?z4^
zoi24h--xkR8;swkV+#Yd&yxC|Kdn~vh7I_IK-RCI*1IwMVK92owIcG^clAJ7({17X
zk08iVpMlePD*_?}wM*tt&6tkr0Uc0@>fr^a4dM%A?~p;(;1-bLiP<-;-U;~GT6&<z
zb*d{vzryRKL)6dX9z%xiLot@rA6WMtGD5&_L*NCK>%NYjJz)b!1vU+*gir(TY9CJb
zbSa0bFkU>515@GXI8BcIg`-2SknJWs^&;v+Ky;l|P3Tl?{^vl+6P5P2`F#y2D(xAp
z7_Mu>Z;(EwalSz+N_PI;*`&D4ckHvm51%r7Wji=<^4nC&)>XXea_O<DUEQe7aIl*S
zXWyX*8QzORn#v55X>ymjk)J2E4HN22(n}Vq=f_&6qQf7MJu^W9we14dk+p`i-2U_8
zL<ctkVP{!k*DDG4Hd^64#_+h(T@85#P9C^q4vX3E+KPX__<;+PmX@vq-T&4)<hIe_
zEFg=g@D`mE{w8&e;9CwV^KMnQO(mtnxnL8oi6@(_l|uF+%MDDq`~$l^!`LuiV1@**
zD4vIU)c@TnZ)sH0O9W8(OXE6g;VaHTH62%Dxy;(>%WGR6wm-27M4ZI|!R=hp?GIs(
zl!^o@L0)!~nVDI4d<J=RS0@p1&tV(!-Qqkof6RoO!+Rn8fKRaP`&Kk>D}$O(voY6z
z`~spBcL6jrR?6^=z7<_oxnHyr!l-()Ma!mUu_Ms>B_b17B~_!zwc;4}x#)4Gl+Z4t
z#OMW-C-xzor0&Nzjnfg1vmvWt6Q56PE9V^zU;?FN)BFEe!%mpJZ9jsYkqWBHa^0WZ
zs>7~$G%K$g#`k=CPVYfm9G>Q(@>^bP1E)Na!@Ur+p{|`Qeg1^v?Ah1gz@(1a#&X{C
zh7i8S5I5D!VtZM_MW}5KM>eqWzo$jSncyDh8;NrX5?p8v(dL&ZbkQxeHL`7x=5}N5
zQvNEG4l`+fe?|Z((#zuJGJ`Jq7{`8g5uf93)|J0X_cDXvi$GPLb5L0C1(|zn*s+n|
z4)Ps=m$ttjunBp7F8(VZh2LZu@9(Ckt2To9l7@i>k-X+K?a2IM<EL^<PUPC)C~53S
ze^-Zze8iPq%X*a#$m<t6CEh<I2-SgGE>cvOM{TzSRNaw8&9i9ECZB~wEh!BJGV005
z9+177B~2!co}B=d8UM-rUxBO%j3#YE+EB~$dV*B=%;sR1zUF%6Ih*SG^@R?L?%b*M
zJs=SdRrl%b?hX9W$^8#HE0f?^kae5|k)4+EVaaJYDCEqSd1-gLB;?~H4@>p_6r*1E
z%lIm2s618JIg_aD73=U>TZ~811c_#E)=1NkeK5VXW8@9IrFaIC1~_z#@;_B39wjeE
z$+;Z&@OUOVD?r>49I*WjGWu@WT#|k=<R?5Y?{YypnW&}bpDrr>x^dJ0^an0ijn8JB
zG8O2(k=tnjb&JAqc7vpUpAvKkmZyD*ET8=1@qF+sqkP9(@#icAgg!;8uUXxQM<kE{
z1lpto$AxIdZg<}hnkt?ZqI9o4Sy->tvp3ge+FtTy3Gv^%HI;2qx%D35=r~Mu>$aiT
zZRlyaOV}+%JI~E9S@!$VZ<xorEqr6YW;y{P%Bc;IVMb9<FnA)4;g(0Tou!isL+f%%
zotGD(2IW@GK%V}`;l2_QdxbHac4D+;-m!Cu|4C8M1TzGSH@-{U&6exlnQtZhuG!Iu
za`-!Adhy$0PoC7}D*jcNtibdqUvNnAe$TjSdno}pP`0{^o;8zPea(16IVZN`<Tn-i
z-7WiqmQX<}MVj;fi<coNP}^Q+J#^&w`zCbq>BUz*RKHh2jVSizlRLEG{NP-lUm=JK
zwIYhj9pusA1lSuQw-FK5iT><;y&xwp0bbH1iYn~R|L+%agvgAl>EFOP>+1M@Xy6z+
zsj4--ni1R@@sLAWL&n+-R6B20NAo+BMZKWVytkzEB;7mKgu>tu?#1hIYD~?vG$i6X
zVq(nHOC8A>gH^GIodRGs05+%{oFJ%lR_NEqa`O7HvVA?<m^YD1_%Gf?PgDvmd`Jgw
zEWN!V!MGUUQ4VMT3JJY<_m(!&2|43dm8_Vkb8e&7HFe?=)cf@FQexE1*utkLdxqy<
zZ4sgz_dC&JbEn(0Bp!lRV<PO-%uL4JS*0%JiZ7{<+&JIUwaGBFfset`9Mog==P}NJ
z6Sus7A>I&(+I!=`@Z1mI2_Mr-g#k0&RaVfz<#eXj0T_)V)iHvr=%b;nuj&D!KU=52
zxG+y?tQ4*{#g^p%MrTLD3P5o{Fi$bvazE>{b>r6JtS8yy4N8ULm#RaE8tXK(K1{9;
zHXk|eYi{nrO0#SBcE%Yc;Q@GZSv@%BId@U(p_E10`oa*c<jZsN4__S<*Gq&!&Z-XN
zLYWOVPW(KUy{HjTL|GdyZc5eJnHp&db*}|DDeaw#01m0}MqF`*TleewD?Q=zTDGHI
z-YzO>td!YiaSs>$Uw@(cY12`<$o+@sLck?4VL`0bmou}ON{aC&mrnT&YIkmSuuO)$
zY|=Hv0T!Kq07S0*DePuht@j#}<U^01X46wXWumxunr{KB9B-UU!5n_s*(&8J2H_X3
zrsAVX)LN=qjz1~AU0xRJ7<YeoF5=T%Lr85=<OI2);2TJ^#=O<2c+_dkacP7ouGZA$
z@5$_#?Gk&Nm?X!?53<kCQ>En1TSHz(pHVeUAUpN+FrJc#9so#sG|SxGI3G!@odyxG
zU#-yUxCO0{JpQB<%B?y&nP(gHACN(neX`*mB-jCRAjw6g;eIFVh*j6&TchVTcbn-k
zzudXG_M{%W=!uZ+R6VP`UF|NwwL>}tF0RDM@e~p5iJTUlplI8AaYFt^2E&h<xu(8p
zr7OQBtyOZZN$H3uO}Jztxh5p>B~$K=M%gv1vsch8-sO};?L%7ap*F`ZoW=tscXeV0
zR#Ojs(a{Xu?!x(})VjKk=;XV0Q>H0t=6$-)SBdWDmq5GUQQDmgBXWkRej)IE3E9Mw
zz_i-Um9cbT0C6`Rv7hHzq0eAMQdksEOM=FO)}~P{%BiLlRSxI5k@Uv=k%*Ej$zSX!
zcF&W?T?c+r6;FYCzs0$gf|_ZO^jR}E?~Cz>Qg<Bh3mxn3h~Ycl#s8$-=2>V5DdVH<
zfM_>-2)>7W2tt$?0}!I*aJ2krDzNCxzY@V`_ghq3FmMBZTncX5{nG{jkiy+<1p-vy
zGYk!r?ciyhuLcvwyPWzO2$I9LaLJJUTaFWfQE<IjC|fJ-A1{lPcU&nQe*K0&GrqQH
zA&?4>Kg_JF%@Z8!`aKU$ImcvK_LV0@M>{~E-2P3$e_TTbcIJ-gfzNRCQQev@o~_OD
zb2-`s&|FyyGc(Flq(dCYCxo32J#N3CMt~=akzbA%*^Qlw233<-NrOq~eEs&A86{wq
z9VSZ=`{&>#R#UxvELtF)XCgh3OeZWW&e8f6Y0zULN)3n4flB&`Z{79fR~JEioUIS)
zXbqr{m}g#jx`#9=1o1)5mgud@(icp3=R5rTZsmY>a9|MAl_f$XIG3<kGUY~Lt>#=L
zsZ(Tc#sS8N%_%}6U?++37B5mjel(K6HpG`jXys>+5a<0Lak$|OH}~+2f>UTq?v{ph
z$9*K3xtNilbk?@(g7@aekg_*zi?E7-q0K9sro~Ar(bc2x%RDs4lv7lWIRc!0V=y@}
zMxzMy`%RTOtwjK=*3R0?OmOud%`w48ku&sd0o1}9tk0fGTW5hj_%BNty^faF>@U=t
zaX?Y+sUj6<-|tpWNnxM&idB?f%Jvl|^l-Q>`MlP@^<Lz~fgkrChzp)^f22_r()jHu
z`4$)Dl(EamltVi7_Ompp6p??vdvl0BZ=-UN?vzgCish&xMeexH@hM6bay!4*-TJC&
z4#EHdCJhB2!Y~Heq;AKn7D@Z1?)`0#+m4O!oL!?lDRdO=_B+3k{(*{{c#wVk;OFyG
z5OO_)$V91d>obAEc;~<dSSRv&+-Dv4#PsU}2zLt;<=ndkN9fVR`mkkVi%y*?-3++o
z-9q0<y*?Nbz>oj?)2oL^EPNqB(*)J~kV<azP9A$h?YZ%`+0tlA2Y*B=PI;}eUC!AK
z%t3fWoCa`sP0zTFL-ZI>hHIH{#{$9hc;2$op}7cB$TH$dNzNQ@opZ=q0xy>rVx7P%
zi|}^{nM{-?N@<;Tmuce_G$Bo)+Li!>3}neiKQZoOA#$9+>&Rol;S#wO58ztVJic|-
zFRHvxH%)cREk2P08t$5mtd`JyRUHJD&^H_Q^8y}x6G`N;R8iIi+;DxEbCdP)*adQ^
zg+w>K-ShDGctrjY4Q~B%oLHLWp>cr0_oehK*E)2+tRjE-de?bgg(4xkL}mGXLl4Nq
z1sX>e2A<V>Z*rvMFTCy`*vuW6|IT=Hq=2{mP0P#t!1Sa1xXbyIJha2RBW;;VWFx+p
z3sourWx9hDLLEa$V+5RL^1yk8?)2{!XIaEx>A;MC%<Fa@bFopfj*NdK8@lB0l4*8=
z7kv0)YlFd8LC&PerC+~?A^~p_2-DYhsmMwXsa+qH_U7y>lO!)u`CV!t6<=y$-cH=1
zEFw~|`ROqJek1}j;9I3$=yc6D017HbxY^aagFdd;%d^9q(sno5C8%%{WTIF6uD=Sd
ztnX2o)%T&yHTv|JguTbU9H~U$#eKOk{;HMuPRMo{YIpU%X15D4L8+TDzL)pJe0s48
z-(aW>L6wKCw*K!2@$yq?R7C~Joc52w@%cKxV<!-C-(iD6scAF8kG3jx%H)<MDSQRk
z5C!3hutSHBsx{Tu&#Xqh?3v62%_Vc|=vDD_dv4-eanVp800kdNIU738#4Klk4A=eq
zd*AMe<?APRp{-y>KsOuCt|$BG1}B6KdP_>`vQpYich=AL6s1|tR<MdHcU(2?`4V8|
z6$wrv=3Zs_>U2=FK)QbfM}qQks>I0Grx(h8g0Hl^^ABmhc8W#obeBz?V63dk#*9oF
z3Y4teGxSB)=%wlM9(=!R51P|%0-&Q)Ib=*;c2A<7od;yQI&|&S-YsxA>?Cl-x&`J9
zF6sH{OG^%hWWekd^t(5?78a@Y1jzSHnt6;{e4WnI&x<ooI5|eQBOkj?LCy%7k?%Ij
z3tB)7;#X@lJV@zf)><sflEj*SE67gZsHADNL6r&NbnV06&N=}ALHF3A0XHpU3U{Ve
z5D)S>4egF@MnETB8ce^1>?s+tlfbl7jqm>Vhu$uwf+c-Fd1XV6GdK25qtF}EMKtpx
zKMii1(~I}*-xGwBq~Y8QI6EH?ke8u%LxHDOmwpAc2(O_l$ZfNu9brW$pc7A<Q@@*1
z9PId1D;L4Z`;kUP*>$_LCn`)k8jdgr>#3_zV%({!LUP7Ike&PTQH|@Jv3cES4abF*
z%GYnW^GcA;p)d126irMwfdn~naNWW}w&Ud=QJ`>m1Jy$ib|B_^H)h5GZS`vbLQ%2d
zH8@gOTg5A4x(wysqTocn`3EEn59Glu^6B6^N9!U|YUVO|+1YCh<K0vq4P4wbZ44aM
zM%rAL<#*y@y+xeP@oquk)cyMZgBI4OUlZ3xND%6=+FoAOKowZqnh`Jg$jBklPX1DR
zV$YZSr4W$1X;F|52<&Wjhvab&VVp}$MN3C3&VH%~ZcM1*#}8uas9%o(wll8-j&rv0
z(#bh{<3P}aER#-g{`HdhqDV_>E<RB2+@J-h%x;hc1)ofoJvjk8$RK8}q(X#RYqw)d
zmgeR!3Ppl+Z9CmjQj^;)FI4G2udi4<cIL=lXW|uTTo+ybXl{4)D(2fo`<+YN0?Y+O
zrUvvNGJV}{_-y=WT7Lf%zX~_jG16y((!RAU#qNdDQ9RSCzA~9lzD2uC7KD1cd`L;*
zxg+Ld9h?8=cK<H9x@mf;gDv{d$M%!K_yRH2k2!bra=k+-`<slRXGO2ev_6z_IgJ}P
z4kawf7EZKf4f|{7I2;ISAqfl?Mb1R7gD*v`e`9J^`4$14;l~Cc8ejY~x8IDqFmPtz
z>ls<$b=x{u<&s%r3JQuU(7I_eydSF%hMNAet^~c?XG7ufklhWc)Vbuke5B>1hKHR?
zAN9-PnCk0~Kt<)(1_w7c>tjEw7~nmBkByBPbLLuMwus2V0b)mMb`GOqXqBy4^;aOj
z`JiUdin$mPoQ{a^(+T4w;4fbiQi)qD1Rd(SXDPy-djg%l2hMrF9DJ*Fceh_;KnW#G
zWy3f9eJ>_N#O1>JUP<M7OXlgIvm*V*Hq7itL+jF$%!JGWx=Tt=9~&)7P7R(Z5Y?~D
zTvA3`mGJ!In9QwkL_WR3Ibru=?M{Osv&qhaRUlbQtzm45_{Gw_Ksg0hkVmLaujP{R
z3j6?NrV((gVAJ}zr0QOaW-X+vZC;G~+Fg<D1TP#YG-jMft97d{y<h7K_<2&6<&s^x
z+0E%UL>9&Y?H?BB>wdYM_{TZsuL&rCXgD{l_H1snQo=IT-B#aSKIQjQSykK4C|`I;
zvAJE0?lRsBUDoIHvbip+k?D^b09(_5lPMed{8Sq`!|pY%l6ZB;u&v^*DKxq>KHYDn
zMzRsh*Q#eY<B7Cg4W~n<q*%2Uze>s*B=V<NR}6t<d{WlH;MXy$EB`!~l6e`F`Fb$m
zoyu5>k~ZYX&qFT#vBZ30>^LEldgg$3UYC6bf0nM1t9MtSo!`wMs)=DKfe^gyC1D|t
zLOC{4TxtIS;BOdjkdwyI|77)v?0VN+dGsRF<zvjW++ET4xQdM`WP&U_EJ4QQm%G{B
zg7b_{Zo}h>5%0GxgiJ{T0M+%cSGF80CFv6Ch!ER!??WOM`}fMFuR+d#0#Zu{cZ+W=
z#Zxjlpg6y3=d|eI@p|O#a~WrA00!!ycAGOj5nCVEU`p9LUx|f`z=Pbuo!<{XA1UQ5
zLH~$Kl<*;|&&?D}njg*Ogu)`8GAAEO`_`d+_@7Y`HY}R#)@SVyAoK~~-xfY~1M`S3
z$6o<rX=)$J13L@Cr)$aE+uQlZp~Ujltl*u*yKZi7bH1{!{m=V`r7pWu-4oNH-DP6V
z2~e|CnZaZ)+x7#V1xrnhcQX{yC>~8w1Rq#&es4Q4blq7}&=O%{I@9;+JL~O-IQy*`
z#-CclEz9xg*X}`};kEc^R*+}}`~=OstiN4}G=@a=zQ4F6+?{QtsPo+{s}@pUtJaeB
zk6aI*+0|cPyHn!>J{GfOEs`Ms&rXp}e}~OrUC%BP3u!o^$8o*!drEWB_1Qj863riX
zEeRxttX)b61`T&IARxeIqp@6dcWc`h$Tb450IX&okPq;d!6s5%hpNM=wR4P{&<mn|
z-`}8s|6Ihg@LI|eXfDl)0G+(TSuU%!cdk0UXL+uCY3GF8k88l9@#E1QFqixUAYNOa
za?eHkSx8+fc5ZTupElu|JpHtLYSGM}m2UjpoP;yu3*P;P8QNuBCQLtZ&;v9o>D}e8
zQL_MGTu?d9VyxkNra>jS-rNrJ)!*QQUzb}AoQqa~0A{R2gRiJfjHV)n_IAey9qF#D
z$=+vTJy8^pjh%90B`aNwKI7fOTADH!a&jySmpy+mA<3gI)k?I6fd9zJ{s&R%Nr(@x
zO9m6!7ZEe;EtkO;B`83a5*;blQdC=D1bt{ngCKt>N4Bi7G~>>VV-T7o3H~dB-@ein
zr~~wi)6mCz(uJN|B)mx?&b7x5aJT1O(~>yFA~0)ZY@HA7N^Z4(`nn4{$4TOLj!!D+
zMDGN^Tb>5(8DIH4{YB}_!!d$X@-CH?jb|-cYIn&vxy>FkgQHXCWl+5PXT0_`90diF
z;PX6XQOaE(tX0!>D4}#kctnlyrxS5L)LC0(J6bRB;;iN`m!8#b>AteD+lQW$2fyQ{
zqYaY1xl6@U`7X!yOEQR$0Ri_MfjeIj+HY7jJ4;`p;t_YwmwVl<t&Xy$!T&f%kEq3g
zaLhM10}9U@-gU_D1SqPzeN6=CL%vSAht(jh%jlv9jW8DB%oz{811R|yW&w!V8q_wf
zmsh^)%JD8jhbBMCKE>I)MVyp_a7<Dt8<?;Ede8;f$!H6^Z^c6TPedLab>f7=Z*gM0
zP3jif^F+LujUm&>EK9~DwB0a-M+Opt0tp7$4CJns_XA-YXCWxTl!d%`7IO9i8W_5m
z9Du@{>+}df;w2J+!qrHaL0~vB^(7ifSl_W@lLq&Ut~XuZ;)pF&d6Q`{+D#llkk|~c
zI<J@Mm)u7B`Tf#e4W0FyFR+RRsHZz=x>W!Y>2y8mDD7(M0pZNVhigT~<GWu>+>PF%
zVe$4~XfHFF9z|<`!`Q6y;<Rkx0Eu6-%O`3PMkb3VgHPMo$qDP1dr7_%Fum<#Vq~-&
zX}M{qc$?Ie|AW#-r+Lvu!(9r#gA!BDoPd4hB*bNa33RBYGL~!}a1u2gT}wj=&8-E%
zD72jXa1FL&lXeA$_aq*nqNG0Zw;`wxT3`NAnyCvQfDx$A$awX&=w0>RZ}hQiKPrfX
z1>jYjxp}$z6^}%r<9{BxAiv)=1jzT$w?cBxC}X+_w<WNM7QOPN{-Yu_Q|DspsJG9s
zfHL3P;JlCti{Fqw*c~M@qJHvk>hJ-y0l005(?@4BRN_or798*2h|U<jF=dZS`28)y
zso^P_542Q7X7K|gk^0eU`gBq7otdUMhCKr-e+OBlioLr$aPa_PVq(-<z5JP?R4@l!
z{T`$UT;wxV^>V8Ij~2RG=@5>++3uU`i@II+f}i7@xJXQe*G$VmnY<-*EF?haPN+m<
zqAq0D3A~nJOdy`489fj;4e~zn8ql@m$IGDXTSN?fE~kf7=g^Ufe%rL;L{tJiD`+3#
zwdYzVQ`PpC$Pavd>7mE_d-1=9BGi%uN1AcHtIFKW`UZ`NPRaIU(CW@ly>FkHoanl}
zk#m}=R!7MlQIxe1_|J0Nns_hj5K-|eNCPyN6EBpuhoeA^&Hcn=(|l_iZ)GMcvAxBn
z^ar(S_g521Y~$VT-$37*zRgp6HT>OT$k#MP^)v7BQ)xSa5+uem%*0eLIlJdg`+CZ*
z#G~6!E`|twNSHWQW#iq{DZKIAVs&nm{gNuf{senlr4-<%C*(|lK-<zs4kxsM51Oib
z05W@A<;G>43xWYuy%Xvh;adngV!b8`zC!qhy_V;fg^g73YJEi(p-F-D=4GsHi}?bC
zYy)=t%qK`G%WL{{h)UE}njP+uHEUU`j;5=;uTx&0xSU)3(4{k$O3+z=g#6b(QYudr
zHBR?z%wi{cW3kMGG$<FhoTa;rU~#H2hjP%g8EXIC$YR!X&A#WPBfK#kq}$NinlE5t
z^L_`k0%W0t4wLQATtJszJi?K!dHvhx_`3Cm>hcxc+DmH<&sr4KZ-&x$rMx0OHd(@J
z9`zZN!{?hl0Y*ehOKe|dLv&G|o#qTA8Y2XtY?&t4M2`{OB!5J{j73TB141Fy@b1ix
zDuBN>+H9;X58unX4i4NBPj(zHTxd7Zv=@k}oB|(grNp3}vuZ^VyVe44oT0aGGLU9p
z=^X)A1W=%O1$$Lq?%G^Jd3$WP*3!i55TS4HQt^fK5GG2hEh?WVhTW+s&j0DE;Bvf3
zg~7si+Okmd>L-!oyl>Ttj@Dn9X;Rf)@0bxRvAx4s)dkL3EqIuK7`LcHnywwH^EiO6
z57v9H!x+AQs+;hy_P)AfiIT}cbQID+XSAmoIQ=-~A{i^HDxosVO;P<O!lgnH*?LBe
z>}dhB@{OI1XRCfRHgd3oe15-zV+KTN;*_&YMt%vPSj@>gH&@evGg-E|flWxxA~0`8
zf9`7(BHyRO;df)QM|43spS;}SWY-&T<Z5`+_H{Ddu2IDK7wm6_k(la9W66T^=!oC&
zDczNB|MD855~Z`jx&eWzr53*)wcZ3=)jmye@gF2Z^qRrwqb$oU0Hm~#`|EhIaxYX5
z&MroN1{oiV2@KTbHab(%OCxeJ`IPG*gS*u2GPSOhc-G{_B_#S@(o&sTjne1cisiT#
zN*K>!vY%n!{x6YYK%?Bb7{*CPp^a1%g2nDxt^bd%_l~FffB(l{i83<EDp3>-L@9)D
zN~L6HCM!us_MV+el8lV3>`fUVdn6LF$=-W!$M(CP=O{kkcfbC6-P~S|=XpFIkLz(=
z_v?PYU%Ds5(Cuk=JEK~LpL(jP4A50`&hXFx#DI<_@^!k+eBtaWF2ISA0MP$H&I!8)
z;q!QXg_UcNj|(dnju<Rix2Rf?A}UXhS_S#UHZ{f#XqO=Ela4JVwW8#d6we0_9ze5R
zC_5&kg%+Lrck^`v?_#~<&K|}c*b3ks7mMjbE`_D3W6rB)p{xuK&c%Ly@p!PKQ`ix+
z{QC9l?K`)>e%YG`A}`vM*(>2z*lAIZjgjjw4-7qB<5$y5Q+fye2IB<YS&L4N3m1^m
zzP?s?JD;0PvqWeyg1jJOduv@d;6skSPH5m-PFGEk^WUf`CtWN|!}=zHZO_I_WJ@(v
z)g86)G$>fsSEQz?wq@tHG>ga`(0P`D+@s!WOB1{T|J-;S5J0Mvpqp#9%CY~WwTK3Y
zP7YpxqK}ZijS{R^o)YF|Z7VLbPCUGYK)`X6-Q9cwpZ}Z!{4Z>9sauXHXz$sCu;ZRj
znpF8X-zzdx)vRc8iAv`&s4z&7zW@;B3PUIOQLUNE`*5{(;Fo;o2$PxRp+1OwAfrCg
zkPenvLSAigGXBE_OBSEa0Uhv9Zvl8CU47J0E*snkpMr3dwxi5}mx8i=5y&NiCJ<U2
zD1+4G8~MlMqZEL(`OQksLi*+3HKVwLMhKGu%;*2Y4-|~E2hl)k$K<WIl}HuxWv0$o
zuHU0gZo3mJizl*f_-@VIQ<c|2%X+pjQ&wx-1?5N3xj%s`yAOHN@H6nF+?u?&jXau-
z0_$G?cXZt2KRmL2DY^v%34ar@=u<3Atg*}-;$;_y-e!Y=V%ygE56_d^x<;0IMT(NA
zCQN1Hxc+RqXT;fUU1vOfb{CS6`8wmZ#nGiYx0sZz{%N#6Qbshl{AOWFG8n3Zc%mYm
zqE_H|{9i{mToq5}qsb3{;EcB=x8@&!Zr6FFlbEvbz=>wU5n(xWTC}LHYo~qn_`O(&
zDKNUfVx%D2;kt+EbP!6dg^N{se|gc2Ub5)a)PnUFEHSkkf3Atga}(z;FcawVVtOco
zwZnuQrak_V8!^^#pqWHoU*iR`vP(Pl$AR#P<h<!u&{jR^rS;;MuUdD31Jo|GI<eth
z?2kc=6Sm+QnCBY*w`M~<A;{JK%C&nA03myTo(+%E?0Y*pAY71OJcK~>O*G!#r4tik
z_1&91KeScA6H3s(oB$^@-LtY_Ff0ULB{KuK10=vFY!@YVzOha>bTKhAlRF3+ZbdE)
zqC=@0;(qz|tMmJx)P_1($G#nZe9490^v>L9lxJbCPc`>neGa3S%|3x_9mlu3*gc9-
z5Gi3ctCH@W-PDuRT=h(56Hm`B;5W-sA9x)vDw0tBt)%nei4(ar9fwT4Q7;H81zxdS
z6->ZxTJ&2&h-<H8aCDmRMBtfYR+02IGLbB(<XcJbe>G}=s0p1HSFP+ILh3-_9vC%X
zDLTKd4duL1$>#6sRch(KtM?!JE5N5l9p#8N3L7~gG9@xftG!tmUu!tX0UN6mqeEqA
zGaB<$gZedhX)s>3#OE{`z}i>CS{e84{u1uD(aDFOu7|9uUVb8Pdjb0BkkCrWYo}6?
z&jZ{cifCG8-roD^Fb;Zeh9kqXpE^bGpF3c>c=_i_DV>5%+<t9!PLJD=)ZY6vG_Ln}
zbEg<hpqRZU!y0?nd-u<MIWMi&T4;HHtd4DlO=x~9!TI2{vXX&iOWvAUS7!#3WnFA+
ztj)hF1blvWW9VLVJ8aCAr%W1u|HvHxed9R9OyjDKwoYtkYWC)gipl_Q`%m$2&!`%p
zQS)+knC9rSqkOv+3J3;M_FPu@dZ(Z{l3>&R(wU1+cKeSPKu3{K!ffehDZg%&bHAD0
zJIULzSY>*Ggv;#hS4-e6G|uz#<L)u~lL?OcqLfA-93oNUoRk}&=YEj(=$84!=PXBF
zQ5K!pCS#{GleNV$^`GL`^JIPk{r*@843ZG;a_6|R+Yj<0L0UwYs^9K@x)(iIwu;%Q
zVxF3`x5QG17EPy(eh((8sJP%6NPjV$J26E+Z+_Nmsb$Es%21%Pjg{D1;Lh39@A_AV
z{kmNxxsj9_SFOHEo=`nocLjccnj(0(peKCL^=EJDcd|T3XsA7#x#9IjTtB!J8mHS_
z>>mshpV8c&i7)=7oL>0FG=}{AxjY0eFCr<*D(aRIncID*^9ZZ4`^m$C-5=>Or|QY;
z7Ehl2y47Np%*GzFCO!9&R7KDCjb_iOIwVy$2C<&N*adfctda8WZ^C|*mO-cXT1<du
zNm#vJ@^Wh%Xo@3|7@TyB{X#m6GcWVvN9b9SUslmLNK6vueJlhLJEs=(nHzs%FSLUd
zpoH#}EZBnq$cZvuD;_n&l$4M=+3RaI%ah-9Dtw}4@q=4z9X}|Tr}O~C)&<vpoQ{2X
zc`U}@^Co$9&`ZQJ$@AJ3GPjs*j!v`55SIt((gTP(xVwt8=%cf&W!~%bS|K}IfM@5s
zrtciw{bHtwVuS3Avg_`T^q8c(14q7%kT`929dllyt-Y{0r9Ywf-uY6;k<ktZ{S48R
z*3XNy<j1&5Co>7gcFWwIEgX)A#1RQiTB|&tawQ>X>eoYSZ!QL~aY{yt>BGDa9hRDl
zBvGu~QV~*NLiNfzAxnzfQaABjX$f0i<d&X2e|;x>pr`>NPr>`b^f#Cg4&G(FmNB~A
zTBRbq`CYjzvnaP<F}*BhBw@|G*|zuj@77n)O>jB^mlcX1zjOcH@@o}XqcuyZ;w3}O
zMh318z)idIQ&ea^?E7odB%2d=r8F9%?`|KN7FoVy8){Q)1Q&hW1*Ozt!}UJri*S_)
zfD`M=qi%J|@!fl?PHi7=UI!b?p7uJLhoEGC%$klGoBLW@wR7{f<vsH@vlPGFthVBc
zAd)M*Wt_(3Cj$l2hh2B=4|nq6pLwYBod0PKbT5J9@#bB%@hO84DZwSVO>o89PN@0X
zk~~3BiE;ZrJ>_q)d8guJ1k^M2NVCg-LtJ|okzwRR*t9wvk9RN|B;WDF8#+kM`I$9j
zHi^$!TevX!;${VlOphFm_xEVbvrY<?rMd_a+8_UDfm6f>U|Ew<kW$KjD*9we1?*ex
zY~HpDNlP&W$>90wz?e^@d8d6M{^nNPN7KrUZ!ZrDtGX7-Hi;T`2lz(vy=UFo6pZrl
zU~zi6%)3s<@T`4(GF?eYyeCa6RLgX>w}MKZkrz;rN=jOCxl*>K*T%-Di865*IJR;d
z+<)k`u)9V7><2$eLX4Jzb5|&5%0mFx#@>sb7(2FgJfwQOK~;XW+r)Z*3Cmb=Bv)HY
zON%IYy%b*DaC-L5O8d^z-}Tv%OV?NKhBoQDXzt_pX7cB|`}2*|RfK1zon!WQ2Bt(C
zTCC|!_s|JQ181OC3sCVpuBg>-dSbGEXRu#$G~@Y%YTVcr+()^>@Q-Y#g0!?tCN?Y=
zB8Vu@Y$$@*j!D;MNHDBL&2GvIWFzz+tXxbtPp<wHRns#xbbLdAUM7AOru*A8llI=u
z$@0<#Mvb<uE%h#Y03c-m?Kfk8v30KjTZJG?NR>Zd`jf%7ZBi)8tnqd0x03SV?j!$5
z+3J??ayrx_jCu<1LA%*PtEDm4?iDmjxrM+@JuW8i?WDFa7zMZyJGV-C{!;4`NO%f7
z-LPL{(^EO?IR!prR(~hNXz;@6>><4-u_wOW(s-7TG4e^&&8P5E)NKE!^0z0u^0_(v
zbkb{b<Ecm|vMzniMoQzMf5tP_Qxr7WPB{SZsy0}C;4c_YOmZ|T9~k^K3Y7&Wlw1FI
zMMXutWgx=;VMF*o0{Z^w_q%^SS`otQVok`e-5;6f0OZiL{Q*M$S{@}yRopc>1KqC9
zArZQ8V!ZEkytf7O$_$H~21JF|WQ?@qOYD#CwsUq0=+w(-xcx@}cX{VxX<3sdOw<WU
z#!K1@4;CYY8Rq0K5Fl{KclYB1ye;wC9?f)iAZ(?5-PNIGRI0#sa^O!Q1o(#!)Camj
z7m)E5JLf*$8|f{NOzV(r&sHZmx1`uJu9eW$o=#*|577(?exP-bIPc{q57M2dB6{WG
zI*pEgS{RKDA0d>0x_t~*z#a>r(f75h=>d`?E+Qf#dtdyfTA$jJE4mHQRV6(Zu6NFr
z-~IF;Qx63W7n&YKTS{wh4xi@92gAU;Ezg7Yd)kR87yN6TQxyIhyD3~74Iw$&_S(vg
zG0q}^sMM|gJh2{|XR2Q(s`V|ZH+f4xbjJa63eH2*O~wtU2Z<HM2XLu_Jb=+jr@lrY
z$2B|_FWH8BNUE$C!LAp<3s%kd?PL9bK~SudDSdfdahO~DTitaHXr(&AU@kqH>l%=s
zJ00$h?x*A`p!;*;@d?dk5g58ech@)q2(x{fdRgVP{z7x#C)_3+EmZo1X^KK)#+$wr
zpt5FNvqjc_7*?ob({!VrnwI|!B+Y+$nu<<?=I%=8s3HMzSZgp2Yvh>b4$9T&5txkL
z4xu9ri`VR#^RS}wgc{HlVZbLGCKSeO(%-8#xVuBT7!`wLDNj~$4=?`+2gM2gCgGwl
z=EJpyte-Yof=T>^F?tCITe{nj_zK{?bdR=Jj!uVd3vDQNR!3afy=eaKl;_3!4$>R9
zfDc5`7d%z%TsIg@Q)D$%>F4q!_a{EB70h*VV4aK%>phg@{{Xr->MrX^!AZQ4?|0Yk
zu@vJIWszlKg}*Kbu*YTOx+|W}RW`UdTIg$52&<`NPglgqoc6aTyaGKCns(@3S7C!)
zAao|4jzsxZ2bvR6{499R+hNhq(rep7bVUOGgtXnhmi+eG>;0DrTUH(^U9{8er~GX4
zBAD06qlB!#=(7o34!uyP(KOAZ)BCD^t8=zMvv!jNu7A&zaqC`|pE)9JCVN4&y97X4
zoiAA)nh-n9k(0O5WR-HEbG5Fj-s?)sX^aQS6^y5P?+IIn)B!>LJsTN!dPlj?vi?gS
zy*<bUCQ~OwL85V<W$ZV5wQ~9sh9Y3%?$`f-k~|0`Hy7b$kS4E~*InH1t3J1VwA;5K
zIXIQiA?MsK?+%p*5zmO$hjI68jL}+eE;TSIb}QQV&csrm$X#!1V9Ie~4cIzAWlT_z
zM;&>+AS<JdBcJHeCd|Mo%B^NRyBnSUgoZ>`3YuSr4B(_YlEnX%_+SU*|3DqRBH*fM
zGF`G+y5>rGR!z2yiBh1+i%BF>%Y$!N$5kRV*VP7JM%5_l4+6mjwvcwiv2xdUH+VN|
z-2+zA;CYR?Sfd*DYYr3lV@{I@t*m`U@vTiikNq8=oILdcnj4PCP6XJl)baSF@BS`s
z-=Gso=z$DMNyazmmFOA#bstj5URF_2af!JwlQE2FC{ALEw%3dHHwHPeEC8guXmkMb
zC0myDCPfZ{N7MZy4L9p?pNb+wxQu<#Y=Jwm6$NHG`eHHPS_B_t&mX5G=By1;zh<%U
z+rH^@r7Fo47lLUMC%WUu<4Vh}N`dB~>`}d-*A>Z+Sg`|;jgzey`j6n8<YmuN@UXNZ
z35wqMkrdMss0rPHcRlCBN*-L*S28_i0mO$wcK@;&=|xbNMK;2GIROFsWvsF6&U*-<
z0u76ganPUjgovF8d%n(k!Uw27c9bM3k`D3ajuXStPgqm*NG3b7nJj<KhA$%7c_z&)
z-=WOVeFJ-1o>0+b64<wQ2KNIsW7G)I!Rf8X8Rml>S=h;GA2LbR{`2>&!BcOh5L)D9
zV#@8R-qZ0iQL(S9MbxIeYZ}`U!aBx2^8TG^bmTO=;774HPVK%VODV={sbdb9bmtj+
zGd;~}YK*8;&Xa-Zr8tYP_y?BT=}AYMmwC1{P1f%r2yEErO8Nh(ZGL(b!?ybEew?>v
zdhpWyT<|tL$8Nf<CLYk3cOKgnS)q#Pnq=9mhE4{h`w79d?U<q?)PHs_EE7TouLvXw
z-2D+!b+&d-eWSgnzA^po#zdKw-&S9yMacN=sCSSGXZ}!Vt*e_|_-7u#O$~SLWydSP
zjt(03di@QY-q!q#XEM%{=EPnq2g}t-D?~3C_M?+6WC3J7Hg-ybTe30h%_n!VO(-p|
zkwD9Bq)l;mNY@$sP-TUq7)XNM;Bs&^rK-P$hg@IF^e0MwA)PkhEcH<mkIM$-W?+vf
zDi0`d?Gh|F2}Se+#)k`vnHd=mL&r}gdTlhs+>1*6Z~eIZQC5EuEH=`SZ)3)x-%~gW
zXTaCWXxnYdb2X^gkXWfIGx&MX&c0|v&<CDj8S1~UkdH<wHDjxU-alp0lE^pG;rIY;
zHli{k>1Ynw6KQt-e{>M>I~W%=0R#x{Y9`W`QZQuD2+XG(8ocy_Odo>PZ#YO-Ee(Q9
zrxPDa@eTQ|f}^2hY|>f_=kDU<BW9Q_fr@{%C#W+3k_Jvoaa;%ua(zTquZ?Gi>7rVi
zRx?Q7o_%0@ZgNYzDu&1Jf6)bIk?_B{B-fX_hBRPm`REEbD5Ws-em8%@1IVBe_?q#A
zR487YSX*1eW|L$X`Fqrp3VX>L_;6<y9NN99qOn8W);Vo2lo8x1V8!A@1B>I$6T-ko
zVZOn5?V|pA7qRP4*(2C>0b1e18OvHJeIzp9@qDhYW;@L48u*%ETAd7I?j7(T(v&za
zd;6Th^nZyr(vPsOO9pZcB~g!aG<QMkfru0I2twp>xr7zFr}s`g>}jHqqq@7%;HhVE
zV~T~a_Pq$f_mAe;>i`-0x&egE8dqfZbm2jAX1DbiOBS1g<V-)z2iJ7<BEn{nw>{`r
zrO&(_D>;SC>RbcZVx&>^>doZb!lmCT!dU<TK5WiUdW864NVOF>DDr#kK2572l$%yN
zUl5l)a#LID=`Q93oza*rC24YzD(}kc`O$g}1I-bC<i7r!WA6Ni-f8K?>ePx8nfa-f
zz-bNBefso{WhBv05R7iQr7l=w@T0xQsa|M$BAJg)0!QSHij!u)F0qkc?R}mDOXb$J
zlGC-S{CUFu{)InRSR0}~7BEXV^5;fGe}pWRNRXEO0z+W>7r&0{=*@MbH@Pk4LiYlw
zPrO-Ue@@m;ynD0nBsWbDG=6$9d4}50SGehgh&Ggm2|GDrVAO%Rc*BCYts^okDF;}I
zUuu}eo51T<h+#T`#z#nJeh)UPuy1?z&cv^Zzt=S*1}~Ub^k6|Y7v|7^kMM)mvwqyX
zDFe{)zp-pa6Fj5oen-vzBlk3J_^UgvIyp;-7+CdG<ecwt?&@%AsTdB_vxG#!$hy|9
z+e{?(t?Z%Q6ZHz|@TU^J$FlyhOAm$IwKR7A^rLx~tnBP4{pHbE;u46Z@_Cb%lH7_d
z@r2r~1P-Yq#Q@;W4`;XpGKq0tD31o#eE%?_zxZ0x%3Xh4=LDoEg)n61=rPLSMGkyW
z4ZBj7aICoJT@6Y44oYoElplWKR=mZw>n{nt1VB%AdI`94Lr$#Mm8FbRE|GKd@%?a7
z5vKO^+TrZ}uN!MEMUGfL$=|5-C)_*+*3;KAggAaCunj2C(3+cXNTtlNIKuGpgN6dH
zcBO9>e=ECB-;8Zb-dj-yi>6X>nz%h=1EM3NAHp9-?-bk_OtbBiKAvyC(qIyj*p^tE
zn=$98-2^Ab$Mb0Km7UPE`)b&Cz|u1}H(o_9QJ;{PF3*2Fa$4rRUcB2lUF2FkQ>j%_
z3W>UC-vmX`)J)x1p+q;j8+mKO^qkEtE;BvZ-;N(3bTDVTH56Q6(Pz9OV1LqiprfY9
zg#9C1m}mF9b&+E+6x+9ZhIouR*~vCtOu9F6IW{(jTbO3u7nnSSOQ}U3uGW4X7;@ZN
z9C1$(`P^~RZooiSCGT09fNPs&WsfRb9j0qry~(a9{rkP*u;e3z+vLlu)9*^It2P%o
zrZX9%8ehk6e9!(Y+_z@Z26@ruBb7@vEce2d7TZWeJ{}IUsmJN(rAeTdEt!<pZ4o4i
zIH5G%F6K_@noc&kBsb%T$<uK^dg;;q)|h-9@u$gBJ)m`ZiKs}(55DR!Bd!0|uMQ^a
zhs{Q6x>8<Na~(U;Ek3W<7pgazd@7J#O{V#Xadpy%43nJ%?j~{-j*h*@;R!1~D2D+7
zTWYletF>jTE0yX-Dp^Wfw(Djc4-B0QbdqSZgo{2`*YsD8SkmOBYjv#YPIP_^nD5D>
zGg=Q|C^9ta5%>xNVe!SB-J6e{E04<W&K;^NODURiIKf9FvsMG3<pV3rl?&y7Dw_?d
zfVVcvlHluRyKfh<Jg#QjG7o-^<AeVUdhXO%_UoyVah?VEgQyddbaJ{ARWFUDn%qHB
zxv*I)vx^Jy5ezA>HRLff+gme5Y#l?JQf6jmi+vzs>CYF4FRB92IQL`gf4ChBP6nUe
z-2Kj8f#YD}qXmVsTU7f}VlE;z3XQd?@^?OF!crn^)zA%`GM9lhpdsG|u(0CiO;+p#
zn0BaJ^?k~5lfS&4p-+R}IT4!q%pJEB!gB?tv$cv4eW)@z*|ffFeyEB(dv!S2#<`6Q
znwSpOTBWZaiszrwKSR+i-rr-v-N?J+k_AHAA}SVLo<ZDMXs3uzT6aua4;6Kabf*B}
z^w0OjFIejRP34kh1(@nm)_*@mrF?$&^$@w10{PL4y-WW>_nXv;f7ee1yfjYHX#_vE
z*-Xn<!~D@o?6-mU+dNYx;4?xO8B$jJk#AuH#E`l==9b@q5|J{R(I~mzGN};1hsRO@
z<WNE71*YTJSY8*4YCP!Vph9T<8C$aa%tu+zHf#qD#h?Q@<<I~<_dZ`roBA$NtNwcX
z8l#8b{}T7^xD*R?&v|SkO1;MS2BhivK-D8t1&32jkdjl2!0cPgpCH(@Q(*O;>Sr#8
z?yyX@e{QUsONA_ET0lpa=MVLaA|C0ia-(>eaX8y2RJMu2q|k~|UbYpDv;f|*!`2qh
z77MivqjVng<R!mDIPaBHEh#;V$yKQTXNoBj37ct-ay|Z;3#8S}jP`-pXZiCUgJ9{j
znvyZ3I%RJJW@ts~MM93J-StM~r{+rHG;Skyp_oL$<hk7xWl$OF)N8ec?Z{;aD653%
zHrx-te)%Pct2N{tA<TQnJDAwZO0z!rrGJH$Oefx#c(UVYQL&)!@?)_oA09Lr$tzys
zwbnVZBwg%1s0%s6y^ePkI6}C&d+{-yk8@D~sai>l92@6m?-s&JN44Xl-+nkUJ;(c+
zcb|PkWJqjAI#AzbS5!OBvd5WkMgR3y@REe2s*%s=3j%th=}BH4e_A#zGT9$G;E!Sz
zQ_p+yECX^5PnYApjy<1d-w`zC61YXx%4Mr`X^U&z36qd1I^mq1NHYbw(PxwEHD{&5
zH}X*PU%YYPLHs8GJGBLuuxGl0+PM1ObZ0bQ`fTn4KB&8@8Xt;qoAzXzXkD=wZ&H`n
z`gFzPaQA<+evrLKUs`%Q+?i$6b1q+Z(1^7?p@uMniNf)9M|>3N$sGd$aPm_!_6^HX
zz|Q-7d?$Q+;cPekw85tu>APYT0jP=Bzn_8?Xd_HcPAGvv24WrCm=Y3BUoHj1%&KE6
zzvg9o^vLz{&4~bYPf>xf80)o-@_q(?@|N9euAL1*_>Hdc@O<UHmSuwAvSVP+v_4Fu
zLF??|u3!rujjJx$YgQ-3;`c{@!V7$pBB}S_*!vL7UE@eJE}7wHmF6{k7_r_vAXHy(
z3Brv#MaABJ_3xAVTR7s1q^T!Apv<;O)28f}#w|qt!7cp{(T}Od(S`x@hBuCDmGw-4
zVOQ?c_6*(a2W3q6FEv3G>9FO-Ss3v4VIEfnq@TBumVv#ZN5r3r8Ax$$354PG8dZ86
z66Z=g>a<A2+g>I(*T{SwkDAu{k<HwF>KQqgz$5W)V)dy%fKv@XkD6~n$%4H2CG{bN
z*!D+hqbKfYpTlA0L^QBDX9w@#23R$rG(dK)P8>|%q%mktiWmc>AzfJS<WG4V9pIId
z1$rRWCX3V2aU=JkkMSN15`}|9-)ZkQJ{Se?PE0@1;gH8hLq5v*q~nTDkDC9AGXt&m
z<x>=fUM?7&l|~tKufldITg-w$yQF;~{%dZ-ceYo2+u8Z^KTHFp^PMN_*dQ({0LJ(g
zue(lO#VHEOn^8+Z?P%BOTCVn3UV5JwpL)@%4|v2UD^>d#*sWpj)JyPoEs`@Bd}x~m
zSzp*12(B<S2IwjFPZ1GJbYS@G40DVfw=x>VkM8DqVo*7$-Iahr0LYnQB{_6;IC3H~
zKI1E>6g@yeMRdq!IrHbtZ03Sg+0XBQ-!4ox#?^BED<3C}eh{92F$<LR=|j%BTNxKF
z!bz!hkzhKI!Qi)<J|v+4sD;0FQNs$cFo=aY=)ku5^xXSXmQ20w`|_T<3OwA}zR2<8
zj|30I2OF}!F3Jk0k*9_6@SPty%#Yh|oV~0PUP4G=?N(euU7t3=TL3=Hg!vNTIZhjR
z7;t&(+WPvM?0<>{!pD*twSpM?K?JA`$SGlcSO*tf)I>_rRxtH6k3|<^*C^VBNLN9w
z=}bvh+Y<@lR8<4BO#FR(w<!jCKz8Gq|2{}#FBLEr*eAMe`sAm+!x8N0VLxi?alaCE
zxv!!H+7jV*4wu@Kj5hKZ0YA#Vs-KzUg8GQEW>lvALs7hw2QjBzM|ul;eu0QJM34);
zM&X6pe(R1@vsK=bnhJ-3)9J?!>!k!kGd8B}Q?iyk15`0pF61X2oys4$?$rT#Ba(XA
z`|7w)JAl_0YaY=NTHaPxaA5g0v$O<y);OZCLc9K0Gkv4d80Aw#J$Ha<{b-Yq->+)%
zSd>4jh5c5zbH9^p)e!p9j-RmVG^rA3<dJ%&=D0dEiny(ch3a(|{zk$JL&TOymhgv`
zlyc9L>fHqe0WE9Ie`7WZTxl-dJL&BeNx$=fSJC@UZVYiV{ot53k*=Y<lU1BLq9$Rr
z@5_uWwzgu3hGT1~`Bwf2Xt`#k8w9=BB|ilrzubEP0J2@F*laAyCc;oDb^Sc!9MYjZ
z=~8{$woLR1=ObYJ6?NxY=r0pbXxy4{+Foxd0(-y0&o8TA?N*?S(wh6ot=JAF3Shr9
zSezJ}<<>dOrqTX>WT16~QJMiA=P!EJVBR2tzxDhHZR^%=g44A?p*1mj`0gLQyS5|N
zEs*_IK_kMzm-6ywU?IMn23HFbtr9LP{>g5kNyMD(=`iuc&FdqO`<|_;?c%SOqr{^=
zOqf!X`_?>~a%c5Xl~CwNt>TM?VNpc`%BTFclZ&$!8@~%W=vIOupb}l%66-Td5L5ah
zPdnQ6nSyjOlf|%80x%Rd5JyPmqt4GA6Z=0-8L%bp0D{Gt!_D5gDeNl78c`t*7lAPF
zC%b?H<{xbVmQI`2-w${OkO?l<T@a-=DgDK9{rApzTI`l!K-4xO4`1NGBvXR%aiQGI
z^<pidkj_KeU~jJj7Ap5?*xiLX+m_phM0$(&O+;losFu9^<@Vi)Ms#F=B|;2$Y%KQ`
zrAIp)7{WmdaM$XGQN=CuWgfGDdQ{4xT%cRV;PTIkQRofK2pM(fr`bd(K@;UfW1Q4>
zdVW)fLI?!t)koP*ba`%fR2;5iME412ZQto5`Lln~HHdrbG45yBO^3Qc5ahwYdm;LX
zoBOL8&kVxJ*Tu6VQ#l34?N^_+15cvd3FtqG_D|!z!6WKzS)?7Kk&X%0B#P@yudRO?
zkvy&rj!#7?W#{n$z@P|0!>7v`WI1dG*BZPk`Yq4tL90g{k>-$zE?}k*#~n}m!PAO^
zDW=rByY5cxbC<k|kF?6TfGbIqpXRn6t|i=<@PqC!3_3=;sI!JDq^8I|8aaF3b*jUI
zU>YiH`zz;HD|d!D?w}vY7Ykp*gK0t=ETE({S~@7%_D1s{rd#tY=JNdymM)E?ZYy?D
zL08697z_ix9YEaANR05>0=_l)?v8<kG!NR?rtQqQ18aGq&iq&F`96aUp2EE85+Erb
zIWybzWz>!6-d%no%A>9NW=&0I`VSRSNa?Q{Kl^77yA3o3)+!JSwmsWEr<(df-Y)O&
zOwb#scHac?D9pB=>#vN6$%u;(ak(8AV%csa8&sHa`Ha_6n82}Mj-j2<t$NOD%;}v)
zL~t^0yFDVP%*)0cweQrS3=S^^?z9&w`+b}KH}3{b20iAhR(Qzeq;L52>58`XbLF?B
zRC}Ws_P)Cl`rM9=`lAn`6Vw3^F<Apz?BmeHIGGKa5y~&hs#MN)wOwby0BhmX@0Rzu
zh*^)1@l4Y&Bg1t!&m}W~?mN3vO^(CDWuXUfw%vD<P}@;Xe0+n2=fU#{PN*=EK{q|a
z{CuLY!wegq0EF7N+r1u;G)VO*tCSNC-eSBK{i*8S=W(R94jQlQ?jfsbdEEFQ{!<i=
zI@Senk75{feOg+9nCs-?+#8r5hxjDhZ-61jcxX#bmx9$wQx(>_r~dyqkiifHxTB<g
zT>FV@{mRL~)!E&CarS6lwxXQ84!+*$*`NLW%3t{s+#PIhU%2(14iH+PzYHcUHZeDT
z1v+-~BbAX59CV{hVO&sE!k&mmVp03)-k_EfESFT7kYQ6mztK`O4@nRjsR_yJ0cMW9
zT<lxovR7e(OFm~La<A%JSj2|PMBP>u(kT9~<CpmXDO@RreKV@(0Q#rxAICYFG|6gs
zYH<b~{Q<Ub7tVfb#b;$q)llxDWvg@N;9^geDsq>ZF-)5X2_rIcY!A>p5VH$n{y7ik
zsBrZn$10$Ae>~1#$Lt?b_UKa3Ma>dz&Fj{&mL^VMb|z#FS_JZ+L(ztc<}2Hb{OgnH
zMPIv^u!t2l)B-z_0%{COuEPzukM*MgzngBYx-=)XF-I6w2}xLqh>xQb2gkHx!C`mb
z+i+!-&vbVKr`zeC*0ViMEq}J(C_mcMX=6Zr5=fgZq+ahsVLr}p5pV@r4t?i)AFfM7
z$b|j1V?w(UPHzYsYbgyDnp>^a?ZRHbDg#imK>y*+jL{=Zbw%qsQ~jZD0=@9jHAJ7<
zC<^RW+iNnLS%tBOz-_7{Yo;0@_dQlG7WA5|O(VuubZ;Nd<P+~)!j>f9u+-}eQ$ozt
z@{SK)F$Q0SdT`t0tf}Of9x7YYtl!$l7g8hqfq{p>wFc%BaBj>!=Okpl8D3Z@bU*Cq
zdkO{Jor8Lw$~axX+2@QlHul!3p>xUYHST$SD|bjPm;+bCJ0FU*Qm91FvYl>YmI#kX
zrkd(9x^M6R85TwIPX?kX3xyETS95rg_PLO`Dq2{@wgwv&uEqmjg1#y6>s|fJ0MhkP
zyOl-9$S>JT3U4mjRT|C5uN5D_5mj*M4>AZe%k3@&?(C<)0TsT?^!Q}a+NA=P%8#fs
z5Ks0A=3H5mJ>zy?t_**9riacViL-3UYUE(a@lfO%^+dIPE2sRqDDE$gEFic=g2eDu
zBi7$X_H1vEfA0CJhb7R3a_INHFR6^nDnX1;m6xq9*Jqg2-IXYx^tuNeTtyWfdS8*Y
zepboC>ZTXwgRdR=5M+*f8Q2H32A3u)P6FKZXoAkoB{vx384ak_$rm3w{4)~O*jOdz
zuFNhj;Rmq<Y(A6`+?V<vgcvV_eQwfuPaP{s>{WoKe{5?y^6;0&DPS#IjlZR4TfS#@
zE8=ySAMM1*X07Yl(3M}|*I&{Mc15^NV9fH|&=!BB;XiW$q4?u-_8*$@VsC(O5m#sM
zSGyF&{lMK|S7mTlN)x(8bT9J^-Vh>zU!3k*ugN|-LNo{j?08JH5IN=N+wg@3AA-%|
z2<Lx8*jo@S3G@Ko9xLFN7X7Xm&e27jnMZ;Dy4{jNLodW_KAWN2csq`JTWm<R@D>$F
zm~O6g;7y7}>@fjE(IU4yL3}Rmecz=-4(Y|&)}lKP)aGHnY73qe2U}>?M0yRyo5)a7
z=cjkVs?(?-<g3yBW5rO6S$LWeMkc^=3Hx;cb&xphlGg32_Dp>j^YFWmZir0xsHQ)W
zcr>3Ne+u#Rz<NHaeg47AfxZfA4J!Sdey@)^9RO3NUjN94;_0Vwca{w;rdADjgDLoU
zV-$qEoD-oBWpghwv{imlKu&5@_GbL|?-fav<mdGLqHulzJ7EYHJsOy1VK#=@a-Yk|
zLMDc`-yBsCL_~SP6PR4j|N6*nGR((y326Lq3bgR6e6E2$3+TqC1E1(W*zJjWA=2N=
zR};i=!*2wzYo%3pDXa%HVsJV?7KiKjiwaWu`@K5Caxwq>mW>vm3!L}l8oU$UK(R75
z;KOz*0h^lM`2y97w~nYGa=@hT0zE=ZZ(QFBzvM;FXOK&;3+FY8s<k7AWM(kK^n`Ar
z6{sTAALKS26nxEV@`==FI9pJ{?Cc(^U>n5pF6DBHA+qm(#-OwY6amhu3RE<2$5&IM
zb}+8~k)eH|2Wd?M&{U9+Fy@pL;y*{oEYHYJXF|@L;B%T2`PKLUc(Sr{7P!R5^gNbE
z>IiA1rp`n_o6}OWViYXSf{|9q5i|Y`k5;UlMNIq#xA-99V<2aJSv2DH`J;PWwa}Kt
z*4Z7B=QOdw1ngrhz!4{><3zdRC6df|G`L%yP<fbrv1%#d8HV7%V&@Uk$pV(-C{h>@
zsZRY5QxCZ1zo3*>@>XAvaI838=_f#Z<ug-_^VRJiPl1Q<>^BoA7=CTGdwnMpFhGik
zC%+wUKmK!zz-@%G!#Cab3|MA##i{p{k37{|o8N9~zLl1ADqSx90b%sr@ce=P7iIMU
zSBBVI#z`i7n&<?c0=LiUwTpL5E9!sv=n#3jsjh;L2uC-V1X`C8g6G3kWU|kek+@Q|
zl*@WGDdidQm1otIcTsp)P$pYFmwR8r9W*K!G^`I=o;?2}yg+|M%nW>76>!YK)-$tE
z-|hEYZ@{J|sDAv{9QGIH?QHvFS(J%LXNs73BEYIa>K*A3FNJ+4sY8DU`e2*4<~PB?
z;Pit>*c+=FC{|O!f=2yDVU32O&K7>3!5DV?o1vBOOESc#R7!K7O$xJruEw_)_mPHD
z_uZN_x*P_W4JTv0@I$`&Xw^|Cs#&vjPpZN{SpvoAyXVfALu49pK623y!xOgs9gS!)
zzEY3Y8nf0*IfiH36r;$W%^oAkt240<E-svcSOZ+k@~>s!ItG4Ro%`o*ARmXm25N&L
z%@55ux@xXc@5_@Ly|o9(Pjr?=MHj6Do%9ye&E;WdGri~IoDtmZOZ2c}bv}|;HqQ>b
zyCY>dc7NfzS(Bm9gS+xZ;<#~FR!s3pu##V{Lf}_?E{@sK{t_K7wa9tP`v7JlzCg}A
zc0}YUAIOezJys3AF1bs>`}<{EU5lY$P0K}cNddm0Ix;-XeXB*zb|zg`S3kZn^5;;s
z9qi08b2)P=knT)ct<d_-H~=7x^a5T)U?UhV8jZe!s~`l`xZ|38J$x*HbK6Spt^LL`
znltxF0}m6D-^jh7L>KZNtfd=RAhhw6E_VR>=H&S=O$$nc7B4Q$aSc%?S(*Iqdxi&n
zz|k$xvcBtCc97j7!vsxWO%B~|d7_cr!wLEj9fz3Db4uFm<Q%YQPvhXBrd}PvwP3!S
zkyf;79iD#axp@OjkI2!!B$IignZAmQn;f*QMNayYM?p{U>!0ild`oCfs26D?kT_}H
z)u~xaJV;;Hp4lvTPmwimR3T}HJxmKasjzTU?_)9eXGK9$yP5q%_h=}v3rlJ#^>g@{
z&jjEW3#K5bzx~3WK8}mGQxvIgb^>0nDB>fFwK!=dyJh41pT8ChL8-2cx@)>0*IGEi
zZ$6q1iBJQ(w3yj^W~Ja{le}V5nmZPd4LJgwNc`acS5Uy!>RjBKrd4#5N4B+0*(HnU
z8W<}TjMRp0r#<UA+cyL9{Vl|F?(UtKSiV=hx(`n{imItLG9TI<Q2HDdhag#0!8khs
z@J|5Gx#I^I)?*ChqKNUvXq8ewO;?Sj+#parpK7#CD&zUOb)<XYyGXEvYM@Kup6#)-
z+dywjlU0O(&j-J#k4-CHMFj5NUnb@eu500NkiKNsW@1B^2ow#AbC41zjV0e$-rfE4
zh$PN&L~RKBqh3<9pvUHXjHXEMDKb2KJrKt^E&{E@-1=GiMcW2~*H|W1$;<Mva$r*J
za8LhkduU%sG8#lm(*bu&2XRuQ4O(A7dhW%@zKgp192z3W9kT(3W{#F^qz6k`&o1m1
z;Fc)q-F+m&|LYI{(?h$<3f6nICh0=l$K-CkB`Aq2JF(JVNkz-Jl!zr4dNejQncw>d
z2#(6de<a+%p*j?Owgc#96LX-9wvn;rN>+^p<f^B|mKJky>p9}rt0dt3_#HjZKG%}q
z%Y|O=3?A>H0HSn|SN1N52m6eST6hvKg$-#j@2V3_8_v#_HjKWDEH@tu&Srhm2~wJ=
zW~=*?ZxAyqzJF;L|3h06`dXP70mb&_N*+V~Dp?NT2pF5K)e&K(d(PWo1?&I$Onf4D
zc6wrWGE-Igiy~VeU69GQTVMCRJpMK9f=addjY}F0y|29n&+yuKliwgB^5UfJr^sIY
zS$65bI}xM48&6N}yJ3+3h4kq0Nx#R+9Nw1AmcyYeR23hVVWdD{qoeH!;r5UP;hFXA
z?e4kI=0K~hh1FW8EelMKlau%P=tt8eF#2o)?!|lo>Iy1eq7-W?tErtT86v=&=cLo5
znpX(1;Sz=ukF~mC%sEsygA5dz?C{!XvY(j^HMb?BmaAz3#=5&{92w>s=iJ=agG1cr
ze5l_Dn8<rqGX|$CvQlGEUvkW$trQ*(DgM%h{0UVD!y04^s5jg|cgKVC6wqmDX-1*o
zu-c*L&Z+lsBTva3C|mi$Tig8hv?)a3>RB+iv0XqUqNDfI!^0zRdWy3eQYaSV=O}wm
z;rC%o2cvucxZHz(Lw<JfTWv2Kns7%gJbyN8n<?=&&o7oDDz0<k#lJ|%2Pm7hNR)ED
zM;I^iT%mq9JuzV?b@3k>=>O1sgr#;D*pBlUSbw9JJZ~`SU$_`EkY(7l7R~%$;nHWL
zf~(VvVW2k(S_rJQA6@|$ex4ZC^A!{UFHsv$0zLR%k>GkNSUJxDOee=@fY;$H6?fu`
zm7mCqb69!T^RYuY5{4bVf|-u>bCl$~kC?Rvl{K_GzX>kt)JNV@Coj7iN`juvV=$G8
zJ%#*scrUCZffL?21j8hN{ykgWBb}V40Z-vAGq4Tw9?{o03^PJbBE~1%b>ke<bM_%~
z`7?yk{_EeLJ$ojz@#G!t52MQX;X);XVP%brwPgvoP7P$9N$}n?#awat1ObIo_DZHx
ziksa&ng^~4l-b(%hUS}CSR&+z7A+<{X_lr*hUCU(W{QNV2CbI~`f0~4b&B5JWHlyd
zK5Xd&cV@T{5Es@u`_muJjK`Xtg3aW87@B`Sx;~lFzTrJ21y{qCwwZ9cwT}IAdYIr4
zXfpU>{rq}tCqCEaK(55E!ly$w@&Mq;;rjaendnQrSdZbIME)W#@SH}5S#YW4-~$vb
z<{@WFzbt#I?kQXCw~amzi1ANkF?JTfzs=3dCuu`on8)nN`gT4BW@O)&<Qw{t?UT1I
zh0g4P30mT`7z}<1o#oZ(?(SuTJWxP-{NnfhWnCR%6DM?E(@Kx;Mlv#r)Nh;r;ix(A
zi82|p=Bk~7<b)ZrjE-}fvmT$DPe0&`%lUFPHIlx&x4WZb9X%<3W`Li-zSK0tSQ1%j
zv^bc<YU}0T1TPB|#?`?GL+`7wQ5ey8+d+c8ZK<zQJ!}xRF?hDNtI%Qnt-|B&fs@-S
zA5$Er^hq9EQVbe-)AKy-wB!CC{*51_I``DR@v6lKFzugjViOqj4MJbB^64=;zJ)Ae
znNO#PqEQ+X9;2>vPFoZK@*3vGm`~+8WXl+g6YtS~i0_{lU&B{OQQPc>cVY8R`rZAr
zXU|4Ct@=ByCOuI+&MST1Ves@x8}$vV^f<B26%))dZK+X7fTRP=d843ss;|#RaD)C&
zONHtNJl75OooWt_!&lEw>+A7!s{!>nX|89a{fnnmSSQB$>+d=rR_e?<OE=zqGSR9W
zy0<6D2>2w!f8x!Nd{jHiI!}q3f_HK54HPNYX&waPqW5+53dX$xtjvopeWT*Kyj0EG
zV?PAbko<Wr_aP&6?y9ep&?<17d9c6REIlh^(sMs_-dE}|lOo@qL{B;xISYKQxMaim
zh*IWwXP4?}ot1$48=rF@`A`ImC7X3P)K_^2bbM?(eDrt_@tKbL`nlwlvRluBUdZC@
zc?~<)=P;Y`+E`qXg|)SH;rGF<4VXvCyao?Y-^9D$e&zegt#ya><)Lmaxn3xd)erZa
zB()?Wav#`hgfn(H^_!reJ`=8?;l+xcIz5rnSSBcc?^K{YF?+VzJ&TKBF$sw>cx+WZ
z-y(oUDRkt}|MD&1Z0^H@=x_fNyd`FgiluW-_D7xE98D>Am!#o|s!K>n@Pi#z9@hRY
zE&a+cXxbS*s}-!hRIXG`-rf6zuLul`$&b6@zMeBxU0H!wRq`keN#{Dsd|sq)xbs3y
zrTN&A_}ijn(`2!Tws}s&>v<<(Dd$@S)BEpk;DVth5HiX34`^a5)es<LR^J=<nb3>X
zljvO?3EO`C?RtU0jdn;fP{?qMO^ozI!EykthONXUUk7V7UAFvC<C>RM#4+7nc=jG2
zEwiF<7Ri^ozbdUO_h7F*si4DgB?eI%Ph;Kkgkb=Yghu10-w5wB<NYHyG**kJ<Ks?e
z`@_5`0{WE4LyyX+>&4M4+msko5Rria@@#jUVVALfxQ^qIhEw3}-@ji4<hu)FH>d<^
zPYrNPr?~h0YN)f?4{{VOI1H;lwHL%MEW~lD7hbcS_X*pu3go=G51)+Ujgwlor_FO0
zPF${s?k8S}iO<$gF-86iBL&sD&-_+a<=&T_1dg7A_`pX}%6sl!mc-54;00I7bH2q&
ze4YJ8Fz<@N`G(A2N6(#j23IR9et%6{%J|sW-CG>Gl8+ueIt8=Sgf~^(f6Y~?Z_RJF
z7CP9ScKJ1^G-oy|ar*RWfy;sU(R<R)0C`w3WgGe(%#Vg$6?1|ZKZsFUuOvYExar_Q
zmBbL`dk+dfQ!DXN8U8w9Rn>4lDYNPk5BBl*hFG)`V|S+`hQ<Hkyabd9h?%ex4X@xB
zsp#v|PjbJN4;1jz2hO|!iquWS>0f1DHD-MG+)ydPbMz+{RAQeEC4qNYwKMO-r>&5n
z7X*hHrPGVaze<D^>Kj$};gOuHQ%k%ldT5eNp%5Gy@?QMIq6uEPQn(zed<r;f`dgWs
zU$GwCAhp7DS&6EzRw}Bszo{%-YfdQ~J}bO>_w?(^#iIzdbDfss;fi#uo7Z0F8y9K_
z`l(m`^umiMj`D976|B>$BW3Il!cM;W-(Iwr>3rN}8qTFYddfOZxZ!qT`uAe>*=v17
z-K_6_5EvLrHV>19)z-85iS=&B+yDN|3XAtx2}*R2GP8oD{nh7@9j8{z&AH0$_3aw|
z2Ph~fY(_%j_KA?b-rlI)-tbs)R8u64a3OWAohgK@yW6_;`3w4^^z`pI3Kko9srGtX
zQk;u^PHRPyGx-)5fz2q-`6-O`3-NN*<3FK(qyFk?5oB~<lwZ=b=NBv4us*G8X~FP*
z`(K?s+~GmY7aPkqVFsA3<A#;blb^MzP>PFoAi55t0I6Hxs)h*_5E2p!V=^_F*7jGm
zTqrIs{_v10>aRT`d<fZjjhE|8%GY-EsMnC&ThwhO6t#(W|Ni|dCu^t85rOohcv?dh
z8WD5l!rNgQ|Ld&=J6?k7=<(z0Kfpz^;Ir%f!6%~#?Lg3EzpsZaaS-3kkK;xgVM`x_
z<IHSrZ6isK^K9G>)srPb?<eTrZkZb5rzqV~c%d}G5=pK#MiCcm1A>bo3GgV>$u~6A
z0U|)JGz{u(vZ(3abm8y6gtyI7*C@n5ikT)gMTUqL{^QwD%LIvSoJ>X3rCl9tbFT+z
zhqFq4$S;R&uZ9(5b(&ViEWNWvdUlI2Q7t_$)NH}L<`%8$;!<~ax81zleM$W}Gl4jg
zB=(&ng-XIVFgEhbu)&@V6d69v$GNql>g&hE7`?KKSB6LZT+cG0<w!m~ym-u6C?F*E
z{U0xY<{TKl^<$;%`<NW<DZc+bp*INW;r)dPtPc9b5z9<obK8h>T8qP!F$8staQzY&
z9r1KrjG76q;<I@-H97eXiFXq=pClyqTm$s%@G`%hD8!|W_)iYkl{wJRyAU>=qbQ=N
z&(SqAmvrwdA!bcAeafY@+S{wqr(Q>Viii<1_>I?I4f}f5u)=G?L0N#a{i_oHQsGxH
zUYp>7=wx%oTjFX3SRD$G^czl&)Udb%-XA)bvJE5YhD{|;XaWDEa}J#ByF$y`n|6D!
z&x^m|@noNz?86ZeP>hV>Zg&6u)Wq9|7L>{8FDJ);LC=w2EF;qQL~(O-bK9vef|!`1
zNajcRI*s|1A9QZfryw_@*3E1NHWK;~N%k<c-R6RVL+7_m5$F%wvlv-$M}=m)t#8wA
zZ>u*tx?fFRd=I_p&Bn&Et&8O=loz5#U>+>zn;!ALZl%kizCU*RC|+FXpsjDdUp~g{
zsBD<e&8Blq->ca$B8t_}!waNoHJ2L7*X5jhY)cv(vhiZNpuQw@GeUQ{l%cag{#5kf
z-oya+I#5;k{qE&<yfC%B{ewh6pKIaPQp!mah;d<R_lRKBB#?ZBS1YviU8~K0ftubz
zy|Z95&n8*s8U7P|doKTxKUQzOMX=v6;(rns_SD{=BimCSrA)z5RhY6hgFBao0NsZF
zqq1*$3jVc>$z2JK)PiH(d*bJxW6cL7#j3U!tT?a7#}d6DohUSfxT~+5b*(3x&CVP~
z$k&>ia5|qU0#<BFM}eKyId)7rV)i;xhPj#PS+Oa`w#&6Bb|T7dn+n{yUw$cvU&*yi
z6rsiN<?*3W8;@>bR&FlcWhVS}=AuIA$En#`kENw0)sz3ue#-FX4h&*b#I&-;C#Jca
z4rBL3dw`=P6?$P6vY#}>uHjePqqoMB2~rN@BFIQ8#kbN)bbGGJ70!LfI|V6m_Z2!9
z-`pg{t@<tjZEFrM|8Vy~5|8kW=O}JEhC^1yGvniHwZ*7CL(O+q7GnI(USK2IPt495
zpEA?YZ5SIH8-0D;(7MM!wNrR|wRU?6%*uRYz65zsKm$yjzR2cL?)jI;_u6*L5x4e+
zR%*ZB>6jNGv1^(|XUlNyvL6#nT<a7%ESMJ*YW~>!20CWbF0!zuH*V5U1(dWkRI!{f
zG(<4fl^~4yf>_xV>^2J5g$h$5iW^Htjur>D`7nWr%I7DEgZl^Ef0BPz&09X{UCX?A
ztn_%RRQuOw9ceC==H(s?r%$fcxg=k0N;@UE9DP<$S8Ai@>|21(_pI5QQ?O(tf0m5k
zfT#V|whE>~l35ys?%+{$N`J1+W<!RjCBkiC<RrRrDuUKxq?gsVxE9Qg=H!q?L!Sa{
zm|D|s2OH*Mgh6Gr8+~<k(g(OCc*-DyzQx;ZJ<!9z>V~l6orEr3GKm*)wl#oEx1|&#
zf}u?uwWpdmVJY<cud)hV3RdHi@48E|U!QRuNi+m#!&dFuQf?C92XRt3{W9oTaqz1j
zu6_KPo#dV0*_UIOr6mXUf7VH;7f5Cp+yap@5dZ$qUrk|aO>v<@+}xS{#jgg|E8LQc
zw$}Vlt|%QY5*iNBibQ<+)=vFS9_}YDH5yIzPs*;^$plfo{{au2;d$Q<zM(OVHxw1R
zotzUkbO=u-;pqYL88^pg6CRQBpSyDiGs+~|{K+TDLcf5;sldyBW<bpB#$Z2Im>3s#
zwvW9**daehaIN>CuyvjO1*q7zM`{GR4m<=SpPABN2TZgR^&Z2B=cp)1_l+{*al~fS
zH`wr;tuNTJ4AC<^_j4kreL|E7(NPEDB9oVU<Gy;7bB?~U`bK+0{Qo#Z<MzES%{l)n
zVdzu46@?7bH+pxJgEwH6<#eY_skjh4h%?fOV{7zQk+ne-{PmYZ=iW_t2leFdxpkP?
zG+x?$*?NL&5av*v3!7U%XhBnPOXe+O9SXfd;M%Zqb8-9tnQlBi@jFrpf~4{3X-CnE
zcTm#=tbSs%Arv{H@lnMqI%#%{7TJ8xkWJi$VBdUp##S*bI34z=S2PqO-fP2#Us0X$
z5xGyQgZ#}QeAy-b@+^X$4Px%;xFF7i?L@!;*lt?mW_iVeOO9kyl^gx6d!SPY4*A(C
zw81_*8-G86SWJCFO<%Lo)!9qxl#Iq+#l2H1iEd=JrM&njSRj7cH=Xv}{F;(7W9I45
zbb8OX{sYrDoKJ<ftb!=O@)`nFbHA*?(8uKx;K{4n4!HBu6e?FjL@}>m@y};bWn<n$
z55a!`xtASVW;M=L$dp}LXo_^^mx+zo>5H6b*Mm`%SreBpbAHbc>ugqY{jfYk^ns6y
z%k=*TBJ(BuWp3grU?MEMCBSklhSwb^5~2t(aC-=94O3em1DPj^Fbzexv^LO1<w}Sy
zVYFEBS+f{DKIXsI9z4#Al~`P6YICHnOi(H*;lpi`PDdwghp6%^YK8beUZ08jTC8tB
zN5ZfEX1gOS<sZ`2_`Bf`?)!$z5x;YHcNl=Xru#$XpzMbx;O<h%v-<gJH@BOECtf6i
zd}34tlG=h8dq*rOrAhyc1v%#p75}XV&oXgn$r%z~QT6%cGg6w?c!88}1fKaC=B2DP
z{9@tg`#_9@fQ6;S#e4xI1faje$)zC5cOOiR<Hy}kT&IEE78v7=b7VW`i`ozn(*_e6
zPIkHax1Uvdzf^JL*%%ue>v!$Hy4M5zAtA(aH*L8c0MiAfyPDt=@;#}?%L4#wMs7T}
z=}x+(*fYmnqh&R;ZCC?Z`pew5FZ??Y5fgZT1f@0v#BzPX8s~qhihZUWotknFUHCY%
zaeLy)6-r3OaeHzfo-PzMf}h!NZC^$&nSH9AmvfC941<cLzKX_8w{~Upk$&8d-m2Mz
zG{hl7j1@AbTpDLzVpIFr3FYfz^4R_)6;i?!yQP@J2U8(ehX#pSuj#<Z+7@Vr3-gQd
z#Zs>K_PcRzjl*!c=>#S@9L9_b3w*Tj2qPq)u3?|5P7fVsVWMvO<79@d+QMO&V6`$_
zNzxD$oZrBdnt1J)cr*Ze`_*R2xc+w}3&owZVg%J8X8>VbcXn7mw-{GuO-;?ysrF3H
zdt3+2DDS+4(ZrelY@{Si2XTwQXo84t3h{S9pjFCqDb`~G^NN0N34f_-44)Z%hEYff
zT@S1|=UuEsdc1SMb9l4-R3MCc@`28!?vYQL$VCVCHOW>956CKZPU=D8Qd0WmAF7?|
zR=rw<XJ$#zI)W3*18%gOg`2cJd#5cS`DL5x<dUKw_N_d9`<!E1LX-O`55aB6)C=<h
zRy-}q+RC0}5(Fbbp`lxn|HiC+Nswt}PQ3j5LG$b}EqZ7e@q=LIZ+h@$o-K6_?<DES
zTW1Wx<LQVtQ+<q!1YDeu9S{3U?<1wiHq+0r#S|Vm^!QCDG5c5-C8?!^%XmlHqq4Uy
zH?Hpyo9?DDd8mA}4cB!4R&2(|`J&&%A03w4Fl87N_v+INd2N3-89q?ip{SSR^c=+g
z2n1h_zsnq~mrg#(!VBb84sRNqyMJk8rF60C{`)t~vTPR>$gGTJCMP#kLfmBYUfpGh
z^^upIA4YI>v;vU(GG#j+B3t-Rtd_jl3%moTDr`Cxd?f^LF8D;~#EGZ;Ew^#@C)lv;
zeFQezd7A~?Nyj5ZPI?Q$x%;k=!AgHt=-_?k!NEUxo&3m3PH}79<GRSKEX)sVm8W;u
zm^_+><r4O9S^&QNDJ6UZo8SNmpx^x$E;}NTPCTMdaTVsFo#2NnzED;qo&wJqaG%>W
zk$2^oXzuUaiN*Plc!bi1`CvccKgg)fO~NF~)9?uCSEJTRq+^au(44!~JR1M4>)Qe_
z+bmW7@zNQ4@c6q%3>-MKdtdZ`K|`god}V?tGEkcHEoMPsoUes1fo9fBpt<VwpZX8A
z23NV%HeCfSyMBz6-{PEMgCk6{n|5)vT{9mxuHi2@jnAXIWUNGI@0wa|^h;(}+im5@
z?Nsc4?jdC9Pov>qHH;2g755O66=T3dM%9KQg+|z?0lB!n|En3K%w~nfmAGAamt?d*
zbdfkf_&j>fbu=h?n8%0*sVN*Lpo=y0>o)h7-EwIs?WbV8$cw(mS+LkFC@5$qaMwH;
zAH)u`^V#3Zw;oc*9Nmp)_;jb>Vi7q?idBu^LtOwD9r?EH*8z{iKZD`$Pqk;T@Lx8v
zNQg@VoFSq2IW!7#X$ky@kM#~XOeig{l*wXH<Hlylkd`BeKh6H0rDU=z$oFBi0gBFG
z3+ez7k#E&d9tUJHMrW@S=u*;LU2fH4Lt6c$L~)w7d!9y?wYRLxm*ZUqZr_LEi^UB2
zFu2vOL|to+rol@mnskj!v-?_0)2CEVz~9bV|7EM|`fI3ZH_!b$xE~<s=N*wpvP^Gi
ziyeBxIkZX)&iK=rb7qW%A=6+WevsIDU>;3ZxP!XCg)-gly`nmTm;(Upj7L(l5|x!@
z%zyvz9l73hD^Xg7#cnZk(gvZ`w-E{Pf*UDmMrFKTFtG(RtN%j^eZbgjs6yi96YvO3
zDY8PXx^0BT9~qjRIeU?xpH6V#SZy{(&Y@qOc)%W4d2T?wvk<6If)W+h`G*DV*j4xs
z{eft-zne>sG6tp5_@Po?*6THuEIYL2j{P#!(CAhDHv+}Q4hWP363O3g*BDdFh9#k(
zjIABuChuVmnJDhZ9VToL{pk*OdXBP~Gj<cA7fINamMV+HoIY?dUlc~`Wwu`yk`Z)t
zjJhf?gX{qfSJ(sU#E8LJ_!IQXNV;OB{_i8$Ew4-s)$DlXj%b`pMWgI3wYo4c8@|%2
zL|knqCnpz0THqaT@283Fd4PCP2zRzkA9s?SCKhS_{!&#{sPYGLB`uGj-w{r>A7>Dl
zGrsNZ_v92_0YhWGcXz?J9Po@XSO3Rhhp_jKq%<U-4Z%gO|EThEbjL&TE4u-^u<6K_
zQI_2qi6|L3<ecM`NjRN5GxNoa`Jk+`S2A@QS-d(pa<e*BY$S8zwLxF#H&k(M(Y0#M
z&6c-pTO#g))oQee#_+8k^T(%f#Q(Z$5bB28=&)`q4J^RWOkVPmFvUY?dpd}B#BBMA
zGyai*$AMh@?Zqfmv>PuE9Mawk`I&OS_N$>)vcYaYurJQdj(c`%u<<Ecg0eo-{Y0bn
z>XD;vyxA^Wix^B1FV+9)dtiSDzI#4lsnK$45$J~$2V+iH&_BQ?Qo4{xO+<Lw`u>ML
z1}{~Q(?9Od<QGPo*?aHfc3_P<Q%Gu5Bn?Ywh)Q`G4o<hqs&+??iTlW)>%SSFPK=0%
zFnXMEO_%ac`g1!*T{VUrWZR>npg%#!qKlg|gbE+RhBs0$qJANjte<1Ae`a!G!Yw&>
z{Qu+ZzvH?7zyEQ(9uX2rRw`7sj7TWs(Lj+AB4qT;mQ6N2(v}e+WMn3@kR27-p+ebv
zZ$jqx+#f@p)${#%{eI8C9+#eVzu#}Sv(LGnb6$2{uf1-BgOK6QIEu2oVeC$fYsMgg
z^0T+=c6x;>VKVWlb90NSC(U-yswrdcNVvulODvyhoP1EgAVIam_Fub1a3@Dc67}#+
zU=us}pf7^;$i$bw*R9DFR<3dsyOeI8k3l6`2t0o;2iIi*#rzn=k0=$k{|A+0`Mvn~
zt4f7uWSCdatrzCuzlD$d1DAv@`m=80S)5T%-EEy7C;aI_oo$n*u}%({{?b{0^fluA
z{CuPQoZ`Y)v5dU^C=bg-H%uI1-W+sYQB0WX9mMGi%(ES!-@n}9wbpDj%g!>584n_c
z{aO^*tgf1;=Q;@lsXKU;lhg`D)rO33B7&>D6BzBEluT~qdsrS`nH47|-G78i#N!*S
zqmAd6cQEIB6ooYTEQB1<w3Tfd^_SY@!=~=Zcvs}mWP~f)K`0gNy&Q{n3wOZG9MFF0
zrOAJKt(FGLQ51Ji=JIR58FQ_zdH;=Q$B{0ka~OKPVYq8fe<<=`=(Lk?6ii1UeH7aF
z?%kW!((N8GGdK5!?IP=^qqH=0<29^TQENIH#Sucd%fsv^gLUDu-bplAb>}aP3FVsr
z6~7Xg($A}hx<|b@e5OH_iE>g$@@o<k&;63!*kp{{ixvx;l26M1y*OHL6mpOsGu((k
zku<02Ec-d9rkglOjGHPgpKtrxTOggRy=h`8X^j$Snx8+#@KY_GCYIc~)Ldv0c}6q>
zDniE<)FQ(lo@ITkg?Tk&Adr)mmgYEOKx6Qi+MvH3hMX_AYXO2KTV2DWRRv2U$x)M$
z#z&j9f_{A=BN9a(7A9=ZsL=!$LM%}1;2jju`q*1B7n>cbfhrFnbbHUcw!FM76&&*<
z9&d`V${0^JmfcwILF0-|S>m41wm6DCqA*`DCtWLhq?7S@GzMoVuz#8J5pdTXO|v^V
zw`~sLu8wTee~jYUeIrWi<vlXuTS7}q2Wsx3wUZcmncEIrxM5(>=L<E*_vEEkFO%q)
zo=a{4n{H=E2M?r%T@=aT?LX2iPjJ8!S@wcWH)yL3qreP`4~jL<&&?Se*3@KD(FX)B
zzi&Z~2BmYo*RKm8rFtq|Qahn(y&unQT8F_b6`3#ANLe6ypxz0?EWse*(A@FS&%AVn
z^bw_^bcDq_iOSw-Z;5W5sh?LblHu=QbE@FOjt!53`FIJW^q;|wV%Ny2+Z=A(jpyZj
z(I<#!npn9bs$Sz?tHKY27zsId3;Su&3@BIiR9Eez+?NcQh(CI2xu0hmRZ;B)w<_@6
zyZnNJ&fR~{h_aw;gzh8Tg>*kRG+gw5ID2UTsY27Gcdxl!if7#9$!M{`3zOq&@g}W?
zTR_luLQ$v9n9Wuo9yMKlF2|zfie~udxws%|U^X>@z_)MT){*>W&I}@MB`_VyN&+Cn
z6Wq#`d7M9>w>Dk;N0L&}LHaVzO%oO|XCWi4bh|wcM}d;(;3`W!d076LPe7QKce#7j
zv`)q2{0{qUCE3I7uD7@zJ$lr3068kU9c5E4090?nLQqHzq@l{d!^5V8J5sbdBBeGQ
zQ}m^t%`OjfX-IKSRM@z}DQ=aG_y+hR7Ny%(JM{iMPQdyu0>g`^nXQ@Uk9H5G6iFHg
zuaiso?a=tx-r9;vFD~PscRRRNCuGX>Ns)j8`o;GZi6zi7@YVxbF|OAcJN!)M)=eKW
zRdQ{CpBJPzF<F)hNVU3IxHnEKkOI#$u~I|rpq1lg`Z-Ox&*wvt2t^<^GBVN;x#Pt=
zG)!D0-i~Fh3Zx6RL*ueL^28~aBT}e$j%Yn?`(dI!br#E=7d0W(M{cXF|7nLe45ZL3
zH0lnI5lhw<^V<#B-lMEc8^?d@4Weh)K#lKZKc)kRPk?qY3?l8<z6?}lGK$7Mx8)R|
zbF|*IIy^eMY{vQVkLBBN25DLZ6)ovn#g%y5DvOqulg)u&C;wRf)O+%ELAGIt%GtrM
zz5lH`s4^(<7*0OwO&~BA*81v%Cw(FDU3H!1H_p5MxtT@&<a4Ax@6U1Hm<DDkW%;<L
zF*|nJPx#TzkC6xIX+nxS7^_!nl%iFVa!CZGP8(x8zPE!-7tA=Pf{!mosPBuoO~4F-
zUSWrbvAAHeE3mw$xy<~Bbv663%@YvwXs3ECEko4t;PnDj{9}%>vCFmtQHBt|(o1*w
zMkz|!<*Pg~Yt-~Bc}d)@)MH^~wT@ni@iSJDDEYL>JHQJ)=c5v67IFGO#__C6QcF(5
z?cFWn1_bwTk4r&GYZE6JTmVjZ#QPK`Dk5IVMIt)!iIyCicy$LO4ShGe;g&HY{#kX>
zM?oL<wYj>wuD_zqd-`W%<4wTV_ptAF1#hA_yq|IA*>XRWxqG|1x>m7epkL}XOA+4J
zq#6DI$9Ye^z>Tmgp$@x`gief6FCw2*^s(1IDjDih6KDn<K1l@JCwe*BWhw_>3Cx3_
zsk;9)H+EMV<-x8nM^c(yyG5TNtJOL)!J+tYVBYC83f9myWJaxyHM2tmg|nnM=6w!4
zCw4`R4I6xb>j&ySB&f#t0MxN0mO4mgHN#&r9u`30K|^7b6mMd$@JZ1i-V^37Y$yFp
zCF6y?s4AG3r1BR18K3wV;GlA}>CA@pOCohvG{z!YU7zLoSEV2yvvxkVT$l-S2og5I
zD+Ouy$<vK+Dx&Ks%7e>g=K~Ua(2|9GNpO&V`E#P<7NhyfuBm4#oyO=9w?@y+`CIYH
zlPA}YdtCgaI-GHufDzb4NVudf7^>nrJ2f@+Zr2juQ4APkPLD6|c&Oz|a3u&FZ$LeC
z`@$8z<QVLG(H@}j>gu!xQUj)IgF0)p6-G=`kYc@PSABiG&)-8O(iIhOABcCsV6}N>
z&!0E>sc9igKyJBr|NcUu2G$FkQOgscVC&igxN&29`{Y5y-BOKpUp|mxdhO^Ilh4in
z<3;Q=D*grI0bz!YuD1r20j-Qu1-mzqjG!<2w6R64Gi~CBI;D-G^TE;N^vHbY?Ivfq
zolgGYugef*2V3%y?m|BHkD~8GP(W|f_(f;-Oke|<wqx8$S(qf1;b!glwNaOX&Bk%%
zeVqYkg$?Q+6{%z%rHx`Y^TTyd9$KcY2$Q~s5u+OAE_oe|m;uBEW95f9WreYV#R~<w
z$~{P!WGwj9B~8U+=}VNy0!%4bKdd3mh3i+q>Sl}@l78L5kFp4a1Jh>?iw55=NnN<a
zC;yMGosq15zp_)Oi?J~P6jUE$``0$vZ-jH><hugo`SRZ6<cR2;r{xR})rDR6*PTwS
z<cN&Wn6wAAP;Yh3(|Ug$KMIXy9N0^4c}JxmPSO@b;iZwJrn_ILC&w#<$2P~YF+0bp
z>Zgd)UO3`0Js$X1QVXGu*Sj!@%S2z{Tj%WPTN^LmKZTYjLVEbdlhsp~%;edH&eIXE
zw0-J%S&0qffUEouK|V$`i$Mjimnb0xxZG_a=o0s&q0sm<Q|>i+hx2!5LtFxhPqcD3
z>|QPK$~*#ov+Q3#frxa!ihhFKBCW=i>CpoR4%`!;csH;k1yG*S(~N3iC%qdsu?pu)
zkZx%mX9ryG+D^oodna9ewQ9k_FEYzrgOXfT<pYuWP6g@lZckHp9*=2bs}JnL&~%th
zdV9St#gGP@gF}Ytr8y4kGJnUXC@AO{_U_H=ha_Ne@;lfE&SKs3fJ0GY`hn7lJxk41
zoJ8XY_LaBG$g$mC5pM*vc(MLl<%pEkkHUe&*3l*7t=#6jR$v(@6E9MenLFe6^DWe5
z#{Hb@xBcSbUn;ZayncfB@+A$OAMY0@#UePIJW$B~<LYa(tF!yczHj~~zXfUMo|n}3
z>aGsZB(TNbB2`On{JiMis6hu9o+d85O@$ojX~d@9+jCz}e6FGH-iK){0H99>{23AX
zBr4=+q@=k6#5eb|D|=!6iyMf4k$zC^+%_Ub(eB$-$V#N=zs2RG`_)0seE2Mi&#=WI
zj?J+&x#zZ8*5$!3a+GO#ZYZMM4=l%)6}fs1!t{dOyAV}Rko8SJsA!%#$5tS2Bh5~l
zdMXU0{MT+=XL?K4;{Sc;>1cn{ye+NmcUt@i3vsOyerZ3tX<WLS#oq;jbJgOrc?wgJ
zBr&QtCMqqQcbMNSshUxvSvuASX>7;QPfc5{pP_mhi(XleaAmPOwg;=fz9T?`j<8C6
zpNPpLblatsFEC~PVL^^hxA+(*EJu5Ug|2?sPB<k_+_#fC+2ow^zl&#oEe?p#M^kF;
zGmuilETu=OBM~JX1CJ8ZO3A^uqr<~LJMTGtJF}UNgj@K^MOEBjwGJH_db)$dipdvQ
zCZ4P(zH(S(H}UBu+mv*_S<(%9R_fWKa-1-LhYy@T(qCz43$B#K%G(igJI!Y?jO+p@
zj?wN<E-qeo{SRTQIo}1VWtRqX16>UjImps9yY3(My$_Mwt(<1JTj<T4UULRW?A#@W
zUp{_L5lGxOw(po-(y898>rZtGV!D~Hd#|ju2~LkGWMvpzsZDf9nfFZZW4pntm1R<%
za*C3glw9q@K`a9OCVBvwL0Eiy2XcB0BeIGE+J#tHsT}D#3m`YsNl_-I8&!e`!o;`5
z$4j4XAo13QoD6`{|I}7^7AyH%QCvvLPulz9O*M~vp%d2@-lAw;mgW<y2Y-mFK@hb~
zi_$FoSDI7D&Z2&ht064~BZGF>Vv_;1wDBD#XH&pE9B>QJ8s$+3o^=@5UlQG1>#O##
zLva1##Ze6r3>|r{-}is4zu*<cFb#}XJ<%+UFAJb4!1grXWR1Iu%287DEV}Bomoo%_
z2&(V$U0hwQusK1T+StVK=x9DnZeD2`Aj=B#(gGvwFJqvmsg0r4^5pfK|56bq8KCYc
zwhQAY_19}5nW#~nuisxbnk^^r230@QD$Y0G4Y`R%z!t(Zg2JNUhSJTiBD`EVsswQl
z2Od9uyrg)e=^D}Y?b{(Cv;OHCr4!-rU`1Jh29hBUa_{2PzhOvuXV#~dN*ar>vzH=6
z-J`$URLn6l{$jr*izKz9=+gO>&Itek{w3><*4IT%OXOk`5g2ERo<6-DS8OZU)>{B5
z_(e~$YhykLLg*|sup1pqI6izcNh|4yP4|$uGZt`ZAf9k}v0%Dv+*xqNKIVOMag6xh
zV?!e&E~=31#4r*iBqk;c-zIa0f@#kiR~A3PX=Lp3=ObX*J4c{8KRAFu;#j%Bvd6v6
zn>~V{KMLPo;Ty~=4gT8H-hKd(VH07&7mV=66?Y^1Ss?koO~4Ib>NQZ+cUl}f5aC_t
zbnSR%hyxwbThGq9<k$X*`vdn}7CT1!Onkz2?)zV0eTan~;H~fMlzIhlfVvvj=v%w2
zVR@B@m5Ip^39+%VBqn%=TJj~Z@7$JxIJRU-MeO@%RRunVdm32s_FCiuz1@jZ0w6p+
z`yNDuAD1<5WZ+A=i5ibL%K~>NV+P1u@K+26C0o~*8p`;dSDM6pd3^!q;OiUv*i22K
z0Eg88oAGq_d&>E*uT$3bRdf@JTTxo|Zj@Mo$B!dCj%zoBabw_U4C}?p8eVq!x{$Pb
zqb|~<OrljaJhD`UMSv0NZYJa%$?I|}r|Z{f;&^Z^^c!^;?2kKC2$18UNS<c_RoU_9
z6|mlFA+5W+ds6tS$r)^#1~2$5oZ#sB6S{2tX@UJSU#ontnthWRG@JYmfMPGtzwx03
z82wI42FHg>;n>R$<Auyz<{ckwADSu5%aH%TU4fmKfxxj#E(}idq~-QBgTt-*SSdK*
z{&)<4e~Zp7%pXY&yYkD~=#z&;jazD>8P1V@P7KJExfzkvMW{LnE9`44$)jT6xhIl~
zV0zf-%eXhC9q^xj6MdIRMn=X`3`tZ}DOV+(`vEiH+uoQpzD!RzwD<fbK~+8nKGiR%
z?N3_JZpG%|XAMEU8@t6U-DVv)ZhV|5zcn%T@!O(z29U>mEes_dv=0i}bAlMp;8u!Z
zYM{Dlcw?9qYvSx^pey9?Q)`qR&C_AD2ZA%PPB@$=l}#?bkXIKm+%dtVHg0g}*O94y
zQ#k%9K<mn-L5q1E*VtZmrcc_JMoi<d#?{5*h{yA#s2+N+gneS^uQt^5rXMd9u;q@E
zNV%h#j(3Dh$+_yQP5FU^y#R)1Jxo=c&~-qpvOb((UqV6IacNTx<F93w9uT2xcX=5O
z?hzTZ%*%P5MogDbkpaPr;8S-;$AxuV#f28eB4?fcz8|p^{>Xxq&pi4bco}j)2}M*O
zS<s%ZY{8#zBQ-xkijyk&qMz|;F!(phEYJ}|p7l)*mv4~<9BL6KPy5P=a*h~{W1e(E
z2AcZUISbzvfs5=KZ2n|p;5W|i#u~E}a%9SL)QSEls`Ca#zD9%41tIm8`eggJG3Q5{
z+aWRjaK?iN30C|9;=fiJ%3u22Xy!2FTif_pjg2i-jcn(Qac{c0v0eRH{2g6u%1G?;
z%Mfnda2N!otewn`HpL0fLVefzfQwhvuV;%eH@?`!i<l$i5S{cl2I4>5`Z7q-F8N;I
zoQ28zk#6gs7`^V$0xU!fnx0}~dk;*L!%AipSAHleQnmK0@NtBh<B;?hFVDeU$ak2$
z0fZAGUtNPJ8{?VLUd0JomV0!Uv&}BqL)O1&p<Y(v*Bm~u^-TaZ?9j5x#M@9_r03(#
zH?e1CZt{j_D63Uewt?MFycPQcx`?oKON=@h#Nynhvj{67O@EHi^oLJTpUXzK-T$W!
zfi*^ru<p449UN;MuP!bo7~c{L3q$WF>QQ2tpUgjWBnuC_#6&{}@U64>)*4I-a_5af
z@{*H_16NH~-VU)<8XxjHM*fKMj^5}4J*p*-FmwNX1PR_m<U?dB(9O~g{j$>Bc?nuj
zSf_(klFpu&R5SiHM*TNChIgBUBxhqzzEN8i()F4P=M?yy!!a`XjIuVv0xF?_&sp#s
zvH7*PP>uSxx1fYcODVW+SbWRqP8GgQ$K40ZOIKRI4Lm+goBGX&97bkezq+v#;NR7<
z=piI9Ufrv#x4Ne_7D@;rgK1XQ#{H-DQ}%b{bTl^p7+91(C^BvLs^=lWBmJe4D!kEc
zNh?UFhclOPhr<!AEKZjJp{Soy0}}pQH+T?!)T2Rl?YnP^pUDjP5UA1F>c}P4j0~P{
zf0I4){@5T#P9L%Q&O4o3j~+WV_n=QG=oAz2%jqcbrg(NNXIn?#9Bs!Jonf?o?m)k@
zLu0TyCYP(s3+6tqqP1qt9dwa&|Ggg9&{i#p^q-qQ0U~Y+B;v;JZ0S1<o~jz(HCOJ;
zhk+0E)UFp;9MXzvTImZaL~FM_S$*+tYWqT&H~qT6&+CTJSh*Ec$bLLtvLY2!$VRU2
zCi*aOk3xGoe@L~@e!;+N6qJ-QNQ1|ufuQsGJr(IUJ)8R-E%0-PNo!6)Yy#OXJo6i9
zM7z~ktxDg>3~}}*mYHZ%8NWXLETP)J1VKn>xMpDH$89)X6z%1ffB6^j-f@_!vmpF3
zOx!o#p*nSSM^RsZ{99<TaGbnK>57k9aAbQx1`^It^Z=*G0parGuF|)LB7+~{*e*h@
zsShMELL3(Ip9XreIY*&v+jXJ6y}d(ikJ?`fIcOF!jFa6$Hq^8$+5D$!05Q>E2iPwS
zwr6*TH<=0+Ys6&js_Z$1(@iXXC@<OdyA%_{n%`7hRK!FB6&4hWd%9%ZWV`gCatK$x
z*y~g5G^I&^{BNcpnneLT#X@Esm*MzMMODSrnto(5#{^`%b|Ht@{CEG|a7rY|k3`^h
z(|fII+z#$A34=_dMQ0XU{tMGR45G8B2#cVh`xuFLC+PUDIN^%psdNCxY3#wGnlP-8
z)D2J;!R>rz@js7INm46XnW@%m`Tjk$a%C{kURG8%3x?u^)G*W-?;qUkpG0dytyV<7
zDHlgRG&gZo`AU;soZ!*(@zPxt%$5rB+%xB5OCILBx@vdLlC-q6So-QQv_ySe>3_0s
zE5G=F5f?4Cyp;*~<l0-k*y%`pa?rq|QJwTOg$(KuE5>b=D`_eH5}j^QKWmITgjJW_
z%55Y!Yl<+WwvuUp8&LqtS_RAg@@4_~*UabUioPVBmp#Nq0}&~4CiHhXTm2Owuod-y
zyvAOjc&3U+k47Z*oOp;j8&$LTm!j1>`8hdqZQ?`&>{RQ;n$MWe*~RY0N8<q2<CiAZ
zw(iH@%EdVE0^@hx;2BTO$?tNk>4gpf^<WM!Djgl2PEssFq92m~jS~LB$>CTSkrpp(
z_%XAY#oGpFDux3GR!3lHSf<3T?5`>xAnTixdihn4d3Nc>24<{PP=UgVf?Nk4+>8DP
z+vgRM;-UHv*;COV)U=HMr|H^;FRG!>;#3aSH@}k7YAfJ>B5w%#w)O95G!}nKOs=+V
zKhmXSbK<qB2OW}!xv(%VH)|}oZS#eX0OYX9<D@bE)O+a|#Fs=M4?49ZeE5fijon0n
zg}mi<3(JpJvZ0k&VvV!6c(zB`GWg2DKjoZwH{@9MNIHxpC4S6#FuOVk6Hz-HU`~nq
z@y<&xe@!|?=9_?+-@BxS{}cgZFyw<5f|vJv^i;{y&ktJ`q!rzTp@rZx)78aAk^$;i
z`jM#VeO&h6!V;f<L1dR@70G+)9o~z4Ir<t*ux0Eup+U{k-VawF^Mg5ALY+3BJ=W}^
zmfJ$gvK130l>P<cDcSB=hrfsD3{dkN`E&7SjipNaa;zTMuJ@E3teUUiPh|GJtE01X
zo!O2-1%Rxnv><HR9Y#nt17m;)n~ew$?jio$uwB$kddt~*)fU$<9Fo%yf)E54^QU(z
zTo`m3@yo`SpkZC^5O&jj<%WSyTJfcLZpqsNZnC4imgA}E+V_#u4@Lq<m`Wbx(O>?W
zDT3|(wOa<j=#AUA_W!XG0=W!#a9?I6kSweFlA!qo85p{@T$}sR-Dhtt=Vn5gq=9*<
z{ULk;|Jev2MD;D2SI(-rW9XYJ^YsRUgi6FMHV~MEq_yKmeyU4e7`(%=nQsQTyCSQv
zl$AEKWx%LyM?V-j<6s%4bYf{JHWp^T%`$(6PHbBaKp0O1Z){om-F@Oh4sA#tpnM{y
zT_f*nMAPjlA;ED-=|45KvYITv^;ZZ#)wxu3Tgg8fZDp$#NeUUCN}IqdF1NAh4u9t0
zJ9!MJ!f|2mF$8NV<IRXqELS{(7$8*p^<9oaQ~BXb6wF_5iXqLp6nFKOlhl)Q249%O
z$7z?i+Kya~T9m%>{{+7gQBsRSJk%iy@bLO;nrlWsXZjm5BY&h6f3l^p-9)EoCy;R7
zj<96Bt#f(;AL55Ht<8yVy09mbt>cNffQO#pvHe^zFHqIn=g&$U2-2+f6QN!;0K9!6
zxkuB*zvd)ZOfiePonAKOv%6R$WMc>X%zuXGCLHi#tP_Q1PYo82_A1@90qT7bINz`0
zKDdv9ua!UJ^BZA2kP)$S|02G0Shb-yCqFxCfK5(3=hgE{eM-=B|CRG$*@Hsdm(l8e
zTXx$B-b!#kWJ7dahNEiedKotJ$Y&(xWz>*X#M_i=h&!#soXyw^e{@s@C7qW!ltgn`
zlhVq1a&mJ03U^=#u7T$_2hNV}{>exs*kU^%WG8vlkfU_pk9D!Me$112wga1qO@uUK
zQxl-j4-diFE~+nSIDBNDy_|1hXJ4DnG);@~B0eD(Wb(aa!i!nF)Nuh8-L>OVnsM?l
z(a=8i-gU%2M~i`SskITt@8pE$9H?jXwLo)Xjk=_$D1*+8?}q8s*SAt{CUQOMz@0SR
zX}q{$wh<ZS=St1^n0ICww>EpM-@NA@YK6t&kk9DEj0_w`5}XdVb4cAszp5&=i$pm;
z`!y;3S0W4MyHxeh_%)?!FSPS}OnY<0^JKQX&KrvHmV{NEj#d^8xbhf|#4P@N;L3c-
zW_pcJ=~Bahk{(Mh*Z1D_?}tUCXe11m)94gUUe=JM+()oxr8>_#=pNcUL%c(!s29g=
z#jjyC66P3$-|;q=`mo)EwJ?;y{O-%6WXEKngSfWMODykbY_`Xfg`!FK)TpnI{-t7I
zd@B7_FLpt~&KDSQhBPiC293_P*2y>iSfA9}_}51%oGT6%VX;uQHuQ0x<bPYMO5g<|
zq8^dvZ`HH{ws;ZZybHYi@YSc6cwQc0K`@b&-KT}PsT|9>eU#Mno;A$Hx1%tT^TQ@_
zR#w&t&}Ds{40#C)1+stWDs3M&g;NXT+Rfi&OmB7we<gXMhY`Oh;u{g3hntWhWXr?G
zh{>LlZ_c(mM2(lWfwN)7Pb_I34V|V@J(k>G^2#9;-^`ncna308*N<pv$VTOmw&0r1
zZld2Hk8W~l3>;Ttex%r%ml!D&p-KbT#3cPBaB|mq{)&VB^RY(20p<xf)Fw0`AT=4+
z!5*QG%W+}e>Z7R(3dOl5!pV-7sMs@s@Al9gxjZ$FB%6HbUmW*1cNpskV7@{pP`u@N
zRxI@uV91I&9^Sv{7??4n@2)*N5Wm&D-T>oXE<VX0CNBN>9zUC=8+fm+_Imc00>qx4
zY)AD^xMTY_Sr>LvP)IOf3)NqtP#q8!?~l#U9lHpv6TkAcSU(v}P6VZB;)(?P1%BoS
zt2vkccqcwUxJ`8TUR_g}#KC2{3$CRR3$h;cmRs%B6Cs^dkOMvLg~@8%!>@xIrGqyX
zc>Qr<gd_ROZ4Gx($vYB&&!~JxREU;(%q!=WuC<{tiFX5SJ`rC`zEdL^(hIpznK<+A
zZyn-3RP_eYyB~gpPezXtt+ZK2WQP)*{yfM@OKi0GqN6W@>medHbF}vf)9g8N=G^hO
z!y#FKNH$cAX7r)Fb22w4C#N$w?=vfIbsym|o8@hTwa(=sStmV~VZ2LGi4;~=7>@~h
z0nYaobRl0nTz*XbEs@O9yo`H9kn+XWyZ7!n-T{xIo^@-x--wsPt~f~%WBAuBTmV8`
z>eHFTBgFzB`zEi_+4kFOAxMC>!T#6humA_hNCtvgGu86B+h_aLZ^EiVbbDnIGOMAe
zC^>nwHeRKYyk5+mk`tc>-uUi6f1E?07ky*3_(L0<gD7ApuR2Skw&#h}lvN-E=G6XA
z*(KkUoxGNz_|IYM6D|5*^2S02)#goM(Fp+4Y_K+j82ECZyO>1_by4A4p@>UFm^Zmx
z%a6SZyy9VVu9wt*)IJD&Ly+jFnQ;Gh>~?tQt+E>UR{Qd2;_v#Z_v%CP)7;I@-UVp{
z=d|wYQbP**mn1#LUX@Rf!AO^z=n-TIcYDA&Ec^g^vs{{q)c%!1{`EezIb6JDY7V}w
z1hW&Nba&aZ^T^408NUK!*ZEQ9Q?o?s&?-YR2~B_U{}CqrPEQ>qJzh4zio43_7r>mx
zuKk|5V)-u^uYZadXEc<O!q7s{y`|swr#Hg<VVw0%P3opSQ}`-N5|X1X6Fi5_+5tvB
ztQlpX`n&cE1(d3Uojm(h+<REg!$k#tjaqNZu}CX4T8m{kn2ypJhgf$_;@!-zSogQD
zCCD)}h0fd7w%6>%{eM`T3CkTmNMEAHX<&)d0ak=r0QFbkP2r=hDGkYuW|Pn^J!Ir#
z)iy?+!VvbTMJ}`XB6pA^(|V5QJf|!f)wv`9L38Qgw%Vc3K*f)-%2eR*Tz2j>B-mNM
z))ML~T8PeV4rug@96l7FB$?rpJ545;#dE@=ryCE3gnmr#;qmbzzW-6FTY^#_#SM`H
zC7^78d0t+$R4bh(<(+NSqLL@>?OdJbfE{{M2>|7{e+N1K;voFxDdzyoI(!ksci?fp
zMw|ALopb;*B&6>KQ>5Mq23znDFP((h6fO&Q5>k%n<lAc_Ib2?<7jN__ag@cx!QBMt
zZk@#<%aS1y(_*t=`t71v0H2X-Y?Q=`&nU$yfJ|qlR&>vT=k1kamBeJTNGkh~Pg#Cp
z9!lM#a`7thL+T?O{+D0~jUG%<%XgYV3VE0?c$wLTb;Gh2tbr*_H_h`4tqQ{TfV0$_
zfs|)!wYlM$!V<BqpHwW{cT3o`ej4lJ`O<`KV-5tA>&oPR6uN%XwistfPD%1ha^UN>
z(va1c*jIP6HAk(<@UW^TI~y|VMAQL*-Jbse5Fn@ug37BwX7e~ti{bGoocW+Y9i+C&
zIYfz-@9FikJ<@(dZ|=2NBzqwm<?Xs~)=THh^M{U*z%-Hlk4*aS-~1E#W+^Ew>JK6M
zA&B?ZSua~J4~*!KWIoaV2X#%@9%(Rhb4^lv$sQc{fNVbtRk2OE-SieeF}Go+iv1b4
z!6#rb1(T4k0T%k-XyMJ-aYW!CeyW|YPxl+ceF$bJf)$&P&R@U`^~v5y#NwF%OR|%_
zzP;TYsd?h0^!JMay6Q8bY%&kH<_&ZYs^Jxd6ciGx>a<=G>UYY(LqyZtRIT!hXI9Yn
zKM*6$VkY(iE0p(h!&}CQja2Y)?ytwgIiie$q2~&j9%r{_%FfkN7v|653Bm;FCj>B0
zaKvB8B7zK)h}Kgg?IOO2W1LiNTP^l9%!Y}0J<jCIxUTQ`wf!Pl(DVlw3YPSD1p99k
z*@~KllnYdYWC%>KazX!HZ<;&#!!TkwVcAcItu<&UI$!0ri1{6S7G{TeWeaF44UJ~K
z{%5D0iFO<mm`zALP4n^`awnWfmFRhh9wF%X_hmx!551#Z=93-xSRjB77<QP}X>4Ur
zD=cDfS@>2iak`5h#dD9{bGrZPIuirKV$IcE_zvDbzWZPLw_X_((!LFah)vKZj_$Qy
zKHDn)CFt?>&#d(g>JNe`#wZhxiHo~A0;M7PSJOTjfiqg@Ll9v-$Oy;IaiE6qLVip8
zlYhMdyr+DmOK)MR0-rjncnabn%zrs)pyI?f(BQ2JwL#S*;_arykZODOu;s)0p(XVs
zwHdK3#7>f&o4ShsbuFyC0<%l%pT)2ofV|OvJK_-#!U337C<kn1e`N6M*RPMI+QNLV
z7(7<Q;<#S{j=Oj#HiFm#1K7@thn`alz<x9EYMSG0i5)&({KalKjT<KeV*d5WowInM
zC`6jZCmAJYm1b3x)JJodpQqW)D;f*CI_E$|LwxVE1n9TqLG0`QO$6+FD3G3yt53EQ
z>VPY9`gJcItd<)c-^!iVcav|032n{KzEcg2KlQ!qy0AS^2SEk@k1FpjsE-Z-xovot
zlE{M%!0u^)o$WFCsR2HXom-5pWi`j~t47!!brsWjd-?d!B)#iylqF*qeA!v3)*mIa
z3#4*kMGP$u%3G_-0t*Z{rwqB_Q*NqwZMt#PB%}{JTmJA-ywHz=7gCph3zJ%w8M#k9
zm*b&&TM;h<dvpmlbo}Av<u$nrOA~pB>I_V?`>-nM2ZXPUJcGDSYCJvS&@7gPy`2tR
zmr1zr5?@IU9V;+l3f~~*uZp-L+TRow)INnbb&BQ@SCrK7+)cQp31b1*UH>n2^~<(`
zuK;r*I1}Fyh(`kMnMa0f^4r!dHkV7u1A<EPM0EC(K1f}ZO;Q-@Nc#VyHvfo}A1^Cy
ztTCsghUmWI6|`K44V~(&A$<~72JO2A<suqN*P9kU*1pBL-_R%S`H^|sh~1Xg#ckXK
zy+crXgwNcd0MWlusW;K^xheSr7T>E+6zrXB+jn4A)jcP(;3jW{+jSE`(P{usQy|P5
zuxDz(`@~qUNNO&L%Dr$^c(6ZI!*mWTra3x9V<${hLP3nJc)Q^5-=M;dbcgcZ$1g6x
ztdKqN6}1WBn@H7eZsmwE&apf$WT-@liL$fZrChTKc(>o51@qo>Zx&tItA`%O^OnzU
zfI3~<?}M#Rz%id*0twf2W^|c)^Pnnu--+mVwrm+$o=RVvHOU&Y91IK$jQ>}HKt;O4
zNfj40gPLK&^}tkydagX(#$@sW!HVyTdXZ)}6urGV2-@b`TMG&b)&*hx>?YXlzx2~4
zviKhaB`cJ5Sp`?nCwk3`ZDBMnxOLY^&uqv<3Pw9s4mH1-F!@;VvBn$06I_;4QmQM*
z0%s~>z=74gA}og7Ldg{ny#a3a$iEw06%K%cjs%}Qjbj3|E|RKStMc%Sv6Y@RK6*x-
zxJABkr~=deZq9L{?dOO8QGFypp1*@e-3yykfKyWP0>1XS4&$^*wia*u@!duFyRYp%
z9EFladV|Irm2G;<VXAl}Vn_pM?t69+M^&63FviaCXQ?Lz4@8EO5sT@64WUKIol61O
z4Q`um4E(-*ZQ8t+k~rlNU5_1QLl-qIdq>7ZcN!ecG(NMADVb0C&^4l#aeHE7B4_*G
z!JIgy5q4OgU&HQ-58ls#CIo)zz1jl9_!E7nzij=HxkzI9Qe65iWOfMivkQ8<f^vQ!
zcq>Gj(7Q(eCGgb+KpJEKF$nm#IAqXQQ1u}$+*kfm9XR4Zvq^aSuVxb}lbh(IEH@SU
zg%@A9RC(Rab(c3>J+*7`Zpl^%lfE>3UXvgZw;kf|rI=V?TDlLR%bybQ`=4kXgw!n0
zlUl+EXnaOGbT9{$x_#@wi2R_~qc3(yD>7|1afdoSGLkCaDYqHRiTz%KOVMGCrTxgX
zuv<G8^K|O(Xkd{R@EMTbI1c%Z&ybGBeiTd;Ml9cc-PL<`qS8ZSFwNr|^^Z?XJtmKd
zkp`_9q;W$M0R!zIYH4ld;W~F_2lk&vCxL-g=oiK!e&f!8<&gG1Nc-FM!E})0xrs(x
ztav%6QHT~_3Ml5&swMflZz$d!Z9t*7X5eZ+@<~FNXa`qH$pw;y`lvhP^d1t}Zre~O
zrm}vi=D_X=e1T-6Wbj)W@DzfO(6c$-{%Ra<U<7HFI&eekkK+W4MOO$azCUrKXHDMc
zr~A>jc{8_u`gDznlVdw_U|ev}<-H6H-{}6@;rfSMM&}`S*W5frxI0ba9UW!v>fu_u
z)RP<Ev`(bz>X;Yi_K>9<QQGc3F*KB$2E!e1{jFH@r-l#X2KuWyQxN(ep6YZN>$HVD
zn<mMW6=S4g$eUfOn9SZ9cZiuO`Kp&IV3bI_ugBf^?ZSU&T;MXa7SVMgxoWuzeCrKo
z3fj3$zVqU@o_)%T5!$hhuJa=LENA-E^~*s1-mI;yts})$e<qBXi%O*$`*>)T`fKFG
z2;k8UNPy2sAv-}>f7IfOwHq7`k_6K<_`Kur{B?+ICB=JrYD9!4X&|>rC3fUM)Kd@V
z*pGsWYed0w6A!lMC!D!6ei_zs9sVm)g>X{}bTEgivg9s4zyBK9zH&V!@CZ(}Fwe-J
zD7k((X{Y;-YE`m&@_6r_25E)8ZP8d99V8xa+<03AiO1kzZ(0r9B%9v_knkepsQwP1
zv@+k`n!&%#_MkGikD>#g<m`g&#S7-%qBCh0sS)+pzf&XdK=!sTUuIwZj|w;#QNYwj
zfrJZtj~k$6t!8;h&gVtn@`7SiLGkDNj~_jj#i*$H$lQB>pe+REwQ~9WNPe&dQx5sr
z83yWI$Uld~%9Tmm+!t~LN0abO>`~ASFv)|j<zX^z%q&u*Gqul#j;SjV$Y=|($U_jq
zR+HS2)d4kM<`^OuKgG9L&-u4eDt}znv9mzry8?`T>u($<kJ{=iT)jEj{(XVZG1o1H
zQ~I-t_0D+VmV_h=UarHu<V5EW^kLq=iV%IZ@1%pdyh(T#W&J$!bK-xZPmL`?W47q=
z&+Hf<8q3`p3M$_?1M<zfSR6)3*vcM8(s9WhVtd%uZ<}Z5%2%mb%jY=oX{>OfZJ9!U
zvPPnKcG2B9*Z7#-0Ab>6v<QmrWLXhU7VCL-B}w`Nqc3*{k@IF;sqM&-6chWId`HIr
z%0c4ylf^YwmNbGk=Mf(ookW!47$0U?3HBrUE2^nbMX9*&H-(L1M+6^ZLugVF*om2=
zSx7olE-CQnElHbkn}z4-#jKUD-z~K-7FVFE8yjop1_lS$IfeeOXj%ZWWE3AY`El9=
z{fW4cv+5_&UM8IH0k~cF&e_a`5dP6mLTgQd4ZBW$$hFIP{q-qBvb0E@PSu44SUad>
zfH>OGoT!d%TvZovYXh~&E>CI=bN*o!u-t)pOf_Y+aPC6jDZITjq&}x)!z^V#rWeFD
zSk&h-V?Vy?nv|$0Mr5gd*_(UsdiI_*qQ$6>HiZ`{cWV;dhRwqATWlqBhUHpyxe4o4
zLhy)*AsZ_m-KJ<`bu{VfwVAmOiGjsUht6KCBo3xm7a|@Q7?6Cn7iL!dm(TU<`TEg7
zhh0y!jd54D1v5|7v*;_|OQNx6=-SZlo$n;(GM*Gt9g_2?mHu*Cn&I}kPoLJ(Yrqqp
z;qvdZyDWn<6E58PMWbAaACXhAlmpH8EBQVr1XP5heSpE%?xGVecu$YiV8@u=&wdg?
zFBuV*38SITMa%K>{G@ctSHwo;>1Xd5@fou_Pj$Wf5{lHwVRF|3`r)x~TO`oO7@?X9
zj0Dp*$5r}JKNFqrbT}-abQeFsW=^lfJ7#U)C>~ABlEHy+S>>5>PaCBma^2y+q@l=L
ziJV1zM4!L3_C88ihOuI+mQI0AnckLags=RU3i$0xhrA%S=#Fk=#ou(a28tY3>IN2>
zC3B9mo)=uN814&>={xM#L4HI-Lug+_RMh$vqrXnM#S>tIWEc#%-ICU}2M7$(LVtf!
z)W&oLw&^u^_&k14BdS;6%)}q*x3)x19~Q!M(0cj37W@0QwzeokgoXd#<Y&l*uiwVe
ze30<<LYNTCr6eWk^J=3h>Nc^(Kih(GXhcqiI`Cw8j@YV2<Oej?EWLmKJ}-r(AEp*i
ztp!L2G{5S1yV9|(5-(?r#3_zGj0f6tIvh&bkUHiE`iubzPs&HXe@;czO_73G5e^w_
z`3+A0jon&Sw9MVsH2!{Ku2pqr1WEU{;>B*&YwKANmyh<-h-%eJ78R)^wK12!qIlo$
z_ON2p!l{jG`}dCuq+BEx#qxn5x^XCStCe`AlVz(TcF&h8h7F>OiGi6wzZgktH!qC0
zF5F2anfPy#+a98#bbTGd!(n%o0Y|YCYtS-`X~i7`4QD}CI&I#}5#qZ#`{ow;{<#|Y
zhJ~nNwJxIgrg$`U`x}4vTg%_Reap!oJLmBK_BFO(B$4z4$K_fAAGq*lOnuZ_@j*rO
zFin;XcB^JbeV3PTs5)qEU?~x{H{FoU)yULz__VseEebq9WiA7Ce6_C?Xwxbo{#V<!
z1+yE<cJ4$Rwf=bfZ@VKdF;}V0ex<|Kh*`Q@N$?|w65Hev^R<~zbf5{0wks$otZMy_
z%={cl(H?U-KoHmy^e_0<hOFp@XciQ=Qc5gcIo4>(Hr>DbvhL;T3pZR`mgEtk@6%(c
zT&MEE4Y0E31r-yfjP1A)I0PzYcPQtF-@910b3k&lE9D1)sfM&zJjTPDIVe-xzBruu
zV8t?FPwMHw=klRNCNaA-L^_(fIcD;^BbQM7VOd}C;Qvc$3J_sLeaHJ5Zx!(2Z)k<g
z>Ee@Ov7?9ZWJJqjMyiV9#miRvgevQDZeP4?hSB~oFhBjOi6oNk{7mZ9$0?Vj=d4Q6
zZHRSQSgA%LRm0sG9YKZ({rI}dzj+UXB6rFxC<){d#=Rr@DavLgzdFdPyhEzzyMz5L
z*Kbe7Z_KET^B#`O>|VI2pO+u@eF=0x`v0l}kQU7|{muCLHd@~b<y}7U^ZI_-eXhw8
zF4;{@G2zUv9&;ZpO1<_V2H?gUtIoISSR?s|NXMc_rP@Qvcke@H4p#AYE!@Z()FW&u
z!lXgUXcA9AM2Ej42L|(F7L^=H%*~-#SgdYn-xqFikNikziJGBMzUd7Kp;1@}=P3d<
zAG4c^)U)yZy)zryy&Y_F{;Vn%W-cRJ5+-uO&JaI0Vna!nbU_;)^Rw(of(4SGeUVnU
zo+GnJvyi*zk*p$HasBziZ?<Gd_*102J5N_@?=RnVcU{MHZyb*2{!I*=ZV@<(nd;A}
zqAF;{Z5jiECjv%v(1?2{K4;HTA6n9=+D7kJl%A;-OMjrFVxN?A{`{F(5!=H|PfBbZ
zgfo*vLqnB}|3@{QkB}$|wT+<D$8@5#iUy^8r?@68*gUu%TDtb!d2^aLNDZhT9Qw1T
zr{@gTWD@@V-%f!u4|ic+({g=c$ri!|xzgX)TLAvrR^CytKsTc|+hBZr-GjGsUDU@~
z6t&DQ5>qzplk&-Yc6RnH`{Veyzl<^3xPAN&fDKOXfAo%u@Cu@Qs&B<J6P+H@@9s%P
zur{2fukCv?Qu$?C;-bMlS4$3)if!hDbMtcXGohls=4Y5*PLR#1l<=2Hm_(j%;9jcI
zc>B7R?|eWD=uo4gyt(Ng%O^b(LPncMrhk<Gh`LzhvFfO}YNM6cp|sj!c{H`^#>Yvw
zXY;RSwcrn%QpP_%U0RKJ<@4i9!sqht8^V(=*Ik}pT6NJ(y)kZX4o|r8d8u37YSxUI
z;;2a0fq!y*dd<e?NAyxPm>sV%|2RI{eELF>pM}Ts!nOV9(yjKHOBCCSpKMK%N?Qor
z9Z1()o}cPhTU(oIGk!f&GV<e_@a({kw_DTPukhrn(?u4vE?JGmMlAhUjF9>wdAe0&
zu&_f`f}Phqx7gwlMRV&*J-_tzo$2InRz{j{tX)`6FCRaWEM-X>7~I%<<BeTU=>w~K
z&aP}eHq|uW!jfywZPf-p$lTm{b>Z4#j?<5Y3#&FeE)^GRYp!aMCQGf4n15-E()=-M
z(cJ8{)T?Zj?6K;xx^VVP4aK!qG2g6p>tw&-5r*)u#ailhiAUMLcB-dK`et>1TvRT1
zflZtRjS@NE$JNyrQr0hAOL1(DUol~`{*l+njo$ljGYkImbr69Ni;w;IsuGPNi?3d>
z&BXJC)m|R{wWvBDWm}t**C?1tl{yxd!#xXdCOtnREOc>hZq5oT3ei(2<GM&6$;EfO
zCTfIMt8c93w=2^6kx>nWmyC>zv_GF-Z_lATcEx4usLpT{Y71%K=@OR^Z&LyM2r_+p
z;zB*Kd^_VloM}ey0@FNlFd>1!dLO8P^CCOR+|zd@uY3G(?S9|iEN+*kmuj9viptSw
z7X0!wJbc;ZmrX0#LiGN@M2Fvjo0)vbL-cduzPp!3@nFMW0;Wu~mH7qDRl0t>+x0}3
zMWW1oroM`whN7pNp4~IoM65E=5p~bFDK_sxZOV2%%_~>Lqi8n-NPrZhl{&Ec?G$<x
zn^m|A4)s3n;!gyxV#7VWLp6nwJD6}0x1j{oD5n;KXWgLJhZoO!m~w=SIg}bFEI1S|
zh!F=TCx<@p^YdGG*<XI%D0&OtO2L^vzPH%YK7{K!+IgwOMMTXbf?$~mcLi0ru5Cf5
zFT}DE{^TM}@!F4lLECsH`5sCKunKV~ainv}NzM!vflA}eYyg$UZ}dy0?L5a)`0<Cw
z*v%9N>=dC7donsG>U8TiPJ;h|i6LD?$8!pCVY`*!O3_r8#pg@t-}{Z}6d#WMv0^so
zZmhc(gJJAZhA)3UmNQZ9@v_slns?VO!<#4A>SZw*nx%SdL$T=za9j0CxQjfFNp-Y1
zjdW#0P@m!$q3}>RU5LNox3{Ckh}@!8e4T^Ya2%~n4Mslw!hZ4j*~9*p&xAk9My9Sh
zsXv;T+0iEM$rYAjFfr{ez7k&1L9C}M6N<mIZFCIeIa?i*a4QG|thbJ~msC37gAf||
z5?7M4Fs1K!HS9_=6=ULO{gMzJ&2TI0{7hQk|MB=txC^u7f%b5XYrTwbOB(Jp)||Sa
zsOYgN=GDK6I=(_drH!2hCvXhjQ$OVJ2W$#ALn~D70YUMnflK;;XAg;bN06OgFh}k^
zsoKKH$6qR+KO59{Izw|g&B*iKH4l%qUH3%K>JS*glIlgH8Lu8mr}8^oSE#@(VE*0D
zj-A0b8x!~fs)@+hnr904y9qCJIjF4TIZA)#B)w}xlR%5p{Pj7JPln9(5|5q_3z4Aq
zT+1FB_$eQHSj=s4YR@KHwT*^UuP<e1#z`IQ?O>jhUY@-BK1~(Rx*Mhi{TvKC|5tlC
zLV>|rfko1$r)uu=D@Wg#<%f;UtoAm(+UK}VjD*}DU)M>xbNcrBtoh08BD#CXe`mZ#
z1maGkn^D)gn;Z`Jo0hm}MCJu^!Ll{rQU~$pO@+Rd*irPG72QKVkcgPNj1}C;F`swT
z6nC$!GfG$T#BQfj{Q0oh{e<)63!hde5)SLEF!#dsv9U2(5|Kj^1X{Q3Ax(EEVs2#$
z!nR*ATm=e}Dn&lW-?=l@RL3bMqjwDW7d$6-x_E4kOMoD{XiuW|!;Rs+_a@fz*0MU9
z)fc3|b8c>CIwtuURozXrkfQyqYmGzmC~lh$T$dLeX?jBu^GOr?Rpc8=N<b7gMSL3`
zT(mVxQkkpmi5wQ01+SlaS+Z6jxfXvBLZu}7*kU`I)yDm8hwC{7L~Phj9COW^{+zb3
zi%Hzcr5Qy}P*NXgU+m0rEyA7#g8|xYf1iTeVue?W2!Eh$zz^h~XsaSjv80Mm%+;V|
z@|30I>M^=@!Z|;<eERslw9##JCOALf=Ukpt*YjaQbA>;bKkI{qYeGTT`4lBD$S+UI
zxNrACzz+YCZUHyY$vZcNp7k1-vY_Gx{cj3uDtae}FZK?TL~}$bbbOB;IHn4_2|!<J
zg6(!cj&(TcNcDQo$v7?Eh5Ru_h5UANfnt`$>|XsM^F4A11l*Pr;*skrDb5kjusvqu
zCb2)NkkW(LfMzi6gU}C%dlpl}_mF6Kup1K<M0R;h|6{+%ALUmlQ~TYZ-&Az`Sr^ms
zTu)x5YscyEhd}GFtOSMX=#ey6ui~N{X{roZim+i~c3vzQG&oaA@Iw8WxAH%n5u%qN
z6ac|!zW>6p#yJ)z)|x}<@^wmxJ|^e~kA&F;p*wpnKYP2M_kyA1sU~_lx=*xx&SZEN
zy+Z}6>@@2kGVEv6gMn>E$!?yGg+Gxt@QbN8em^`I#w_BVoV%)RyE-FsRNJi3?tN-A
z6Z`n`(Ovl^4I^d=Wu>L1+q&a&S%s7dWYQ-^g@=a+9kfrzZ9WiC24O(wIpDhj391A#
zKXKsy-g%~uO)et;+LGq7Q*Zg2Z0^!~AGR}j)Ra#;J>&!8Y$4HBo}>5Uen~Q_n%q2H
z(!|!v%xr{Q;o%hmeRk@wxYwSXJH>$eh@s4zKho+-!Zsn?addow3^$;@@S29e*-2>8
z#{A<p*8PdIFJHZRbE8_X-*DMo_pzKr8f$*455+`aU|{DF;Xr(z6LW?9(H-qH5)*6e
zbw!ha=44_q+tlZa<5gu_V9_TlWYF7!`-@jljX@-Dso5Gr#f?@M<%OhjHLH6o9)WF?
z9cYRjp044j2#t@ACwFE@KT0)}iT*q|FO2P2cBi(Ev5?JZD)7xE0{LT+f;ajOz7K0Y
z8j7nf{bEpb10sM+JFD?6BP{hS{*gVB9zzp3!dDoZ+MSHcJ6&$fx}QvUz7WlqaYZM!
z(_+aEZt{!*IW2)KGY6F`%LN|cPBaHTM6-sj4_=KC(?emQYtIo0{GlX(Gh6Ja@dYI+
zn1;3%bjEYPqK)TqC9*q#MeIyq^&%3F?v;N_<ja?0NNGYy0N~$t2F!vexDr&W4%p&|
z7lh2e^1K9w0x>;A4z5H+CXWbSh)VcSHg0;Uud`<6bslbaHf<$?O`r|qN)ZPP7M2*<
zFw?no?XUS=y_O#zmVGNE=e!r+>IA9csWQdM@?fhLLolzH&k$_NiS$zw_=s=|q;pqF
zNVp|98d>nVhh-h*y}p`|e)mgO<frO$#qJ67XU+$NG%(0^5LxQv()bv!hg6Haw{wxO
zx+X%v50+43s(j0tun8U*O?G+mp=h%0A6c56$}Fzo6<bJRdGLpW2%5BXp<Tn9{+`Ff
zW%bt_Q9cvxE?tzE1E?JSl!;v1-Vd;7#;?t(krq`(Lz;MSNly7_qB_h%d4gtfJ6N~;
z%HmVzk2D2r{jcIVh(hN1R|5`&ZJQfsq~g4HbC+7x(YlRE-EAz^UEfxB-SPJ8R2Fzj
zZc`LKb+BWu@=N9Dh~Fw}hH96IgT-L{J|-mv1^1tKz4-`_n&Id=+qzQwU^k*{(34<`
zibNU@;YEQOGnE(Gtil#*Zi-KJr6L%U@mRtd9OcUx>o9de=dQGzT_@jV(#JGEb8-F_
zTkok$oWK*_@OjK-V9IyD*vy4Z{`(ZM+c}99F#!~EXG6ts#`TXb@bX_ggS?#v>7hEU
zl-@^hX;O*leBFD@5BBhD%l^ojYZx?>YUyOklB@zNJ2&q%TDaaWM5fo|B(O4BH4&di
zNX^-@>3`5mOZcFVJSSQSL{RdbS@6H#1;AE1D1}uEewiU8uf`4`@@HexpXkWN_$@xL
zu55uaMd*OfGbB}DS0f}ORQ&i?W(wlzsioGt#rtc#4k8ze_Jn&}{n7Xb3@~~p5|dpG
z?k9|MH#w+WG>VbhRmP!xdoGppd_dkkrAEt9Z4yQ%XG|;oYmr<an8dM}2XZ&j#;Ehg
z-B{2di!3}QM`~|V!8~@!D2o-G>yuGwSaWEI34s;y>>o{pyiz|Pe&dyFSk<zq;3Zim
zp;$im9-Mi8-Rr_93=QcnKW)2DmzZv_`)hD;aAe&M6+)r5!VA54D{<l4tibM%SW>yu
zW_0`r(&)wN$eZYQ#P0`hbe!Z|+%Jz;VJ1qD=%0Ufmf&aY;LTX~VP5s@e(|E1!9IW^
zDrxn@sfDEChTY)nMs?^WsY&kX7)$k0RF}bn9-dq&dGLnK6Lk)8BiN9XoIy<dA1S$^
z5HLln0gZh4^XlIw`mEZ<BVXQmwBq2sFNHmUiLnd2UMso#1cTYr8n~I5nCQvzt7gT*
zf$DZxFi^bv4=#Hgjojti1~Yua&o9PQW&|FQ7iy#?R3WnHM5pfhd7hMuNze8j(KOw!
zrRB6}|GL<i*X)Hp>b(8(>3OmVJ3G7e9Q_CKxNTfioEhlFfI%nBmq14B_vS)OIeYfk
zO9)~xgbW9>ByJ>_L7KBj@H$daRY$=ckyt0~IlB^n>PJG<#XXn9iuTtke$}EpDQ>K%
z1rcng;5kC|0QltU2R$~~4@YsMk?ZLpOxL?C-)XYcVK}^d3qc9seOqeC2<ygF1{2_f
zEb{uDqf*l*_kWtHU*WG}7SnNc<!@<_IZE6jhNS)7YVL)XJ@eGerXvlPc6-qob`HFC
zB3mAT<HwD>T*EW3{g+>cX#lT!)ublyB0)@NkRkbz#^v7#;=YaUelKUJ=uCnmMRe;b
zlIGrvmda<ii0KC!Ru~%_ugm|^DQI6J=rwWjy?YJUbsEi&&@jUcuckQt0GHJtY95Mz
z8!u5Irx_-e;8!clU{=%Dzg;^x$-f}9P+Qy5%H+-$He$@&!lZ+FHASCj)gDsxZWY!J
zNs8rbGUC&+$}(FyT7qqw`VnI^1IDOIWKi%=V}z(NU2ijjU02?MnJ(*fm(W?Qw3|NJ
ze0;|iwmWZ0+XRs~mQ~Y2ifOIXbe?Z9Mg0lha1(9k1O>hIpHXbH%UyOzK`<8i5_}+>
z^nrNXfjVB1fGXE+y!4|diRdQZu2*IhExzl>&Mk@$ObmE1t|5L1f{2EZk>p)rZQ-!Y
zL^<$K2;o_Iw4{3V-gYNNx8W)?64W5!q$)xi@4)}SP(u|=x2H0MpxvT_N>(rh9jz}5
z!(YQ!gpZ8SUFn*?Gv@UjAyY=z&d-PI)mX@K67K9ksLL{m>UCz;*ZpW*R9$_6fdp7)
zo)!G|o3&xhTZ5H<S6{~v^=0&ujez(bx}ZP5$Wgq{H)N&glbNBcUXpafwL0VS2b_r|
z2Jv+5kP;C3`QhF!sm=OX3M036*WfJn1kr=Q9Fc)$f25O@ksiis;m+3t0k>xra33O>
z9Di|o-#evZ@x2z;-|Xg$dxFXt*Y^4NUW7-`a@QRx$G4An#}br<zbr7tfQ|Yi?DSR_
zq&$-SVjeGEf@I&SFC}~r#yd3iYM>w!j$!WvYy1W@BxkNr%N94vKj6>}8!@zN;eLvw
zafD*OfB){cDUH*oT5`A8?UVOKtj)GUZ1#_@IDN87lSJDiH)mPJMo31$%`MeW>y4zy
zj1HC8L5sQ8i?(=Kh_JR?^;(YFR6giaW;sv2Z_QSe|1P#@Q-tXHCROu4I6)9_p}HS6
zLxk|UH+mc9PQirdQMnkU8_MyfL;ba)OA|bS3e~hvwDm1}<m|1D4)5vidgqlW2;eu)
zH)xfyIRhAC51V*PZLV-x-jJxPacQ<wp#VYE7XUK(Mx;6)2uvbqMZ)Q5kp0@fqW-NY
z2`YxDb?x<hlG?LMQBe;9SSO@aTR5XjZ1*MRXmAJB-P=`VPJBU64tD}H8PWSu1`e=_
zk%(2qyoPYKYHcTBq<y^)+KJbYUqj^P2wq{_**z8WVzYFu6VWNTO~hvit!fmEcUR74
zwjY1)fq_zglVB!9P|Mdh=_W<u0M==b>xzRPGZ03Ic7$_vXqw^Y`x1@>K8QLJcsL9o
zh&Jqf4k@#QO07+vNU=S$sM<~BD1xdbY7u+q4q2hPutrBh!V?@RMQdivdBL-Y=nQFr
z-!J@pJ3)_b(5#;hWV3CS-<6;tT2b;GOu1CDmBkhX?|n-eeO?XyY^*$>A;D^`&{Z<m
zZ_?RkL|5Jdo6kQP5pxj?QQu|YJpz&dT?eodiP*2$p&OfxN%S*>_v|_h31|5hSki*o
z!`z`-raU}6o7uhITcoa$-i1QgZ?EEo3vPnGy_*72mPFL~fNxVFU-FJ(SFsL7fSc$T
zto4Tn{sh^X{Q6U2Awi3|Qss^>-fMnO_rvqey_XGvgU0hJ3K0RMZErJzLx-UXM~`Ja
zOv&P6F2BnYEC;|ATJUP^%r>%<k|0tCYP^Nhw;g%Dae;W=F-6i`wQBMN7r1^O#>;V~
zG_gmtWhg6~L+{PIBGt^-rR5i?voqhVCWKoq8EIbV!@Dgssw^wQjQiFeie<k>&Vtzv
z26@}v0{lxbl*DG)TJaya-xH2=2N@>l|Am_X*+j1s^)$CTie;@uD!PakOEMK@Nq#?1
z`-+WPRIr#y!KYcvX#ZN_&*7=9Kko#e+Cg~CPh`W_x}Cvo<VE#GaG9=WcsCOa%W`tO
zV1$zr^=1g4Q$)7^+3JH%3Ss^?+RK%*idI(5FCW=dlP8CoDk&+gZ_3J}*NCq#R<G~h
zi*-Hh`9bk@D3jx#gVzplTouH0j0}emwmL&;bNaz@TB*&9?+tS)Loe3?wfILTwJglN
zmP6zo95%)AYP2${oBy+u)Ei|X?!Cw#eN!lL$q9e>heFzlia-+Adw5@MyjdGX(zy~?
z%*<4n;qbgHB2hQe;^NB}HK<<5cLqbglStNg|4&6kT-kP?d!NdhBo_$MCrtyPdw8Z9
z!GW3Kf&=O6rZ^waCnHz~!u81=hFddD(2614TE~vkY4_|GbVY2*jhW_q8tza;nHz}(
zWv?{3n_pyGE|z<{ThUXSKlw49X)!&dw<)ggDXWxW$38;e5wb25d>bK5%kHUW>z0eu
z_`&~APO-T1oZm~+M{wv749yv-`9aV9+>%W%XMp9kXnA<@N@m_`C6quVM;76Dyq=Lc
zQz9ZIdoA)2cJ*&Ql35*0B*11jLa^vLM10q{wh;a#!x3Zs@N7Wo`D)d+C>Q4&lLuaC
zI)N)@X5<S~E%iSA+HJr<v31{?%>U0XsE;V=7V%u=kKz3JrCsZM0H;e8*+fA@E*+84
zN0He0@71AnKqzPSUw*s(F(P)EYfoixl9dn6ibGkV;0CJ^en4;#`D-Gm4q2(qlSdjc
zIg?Elp^cS8pN$N24mFdWh2UJK`)8oTve#4yD$M}Yow6Wi>c>HB(t_b481*+&HvfM`
zTSGY|JyMHHd#*;!OpcEYR)!cm>drhZ2?oP%G3t?dGBYzX%1nKL;7jR$B?@4*i5Y*P
z5H^h%A2a~B3jUJ!1{r{S<}xMKE`o>?&9e&}m0VRC5iT!JA8&OYS-G$p5JhB3b(xYQ
z*SI|)A%Ust;YC6`$3Tmj3NbBYKFJ+u_zdB?3}axr$YC`?08C&QQmR}?vC}GTOrgX?
zlB#+3J<mK(vo6>2t#r{TOVc~XSohVtB92_*>v1QtcaKNC4G7Y-0Km&n1eElLCHnJE
z$`J2W)Q*dgm@cU&=KI9EvMlO*{QY^Ih9vJV4KlUWwa=C06iEQ+h{YVa9TyLIuff*#
zKRlkO>X3kL@Y!E*j(ShbNyU{qdE>R#T`SjzhG&(E#^&bNy!|Y9$$|+dk&=SV=U6-O
z9KDz`JwDa%SFEihaX!E)XSw_TYwz4+n!3UOZV)GP3uGA+){1UIhS>(JQwl=?2kK;O
z3_(F4B817K;IQDMl+ub*nM!$thXxAFDKrDy@+h=CYS9S<gKcbw1zNdlZ~_G?6|Z2s
z+wQnoYQUIff1mr;&3ALp`}dviob#P?f;V=?ewR==_&y%5Yz;}EMT?`9Omgc;1Cl|?
zf3(vS&x%2QCQY~K=n3wZrJOFA5#M*uQL$g#6f!%i@!%ki3W~UjT#KII3q0F;+i(du
z9veN8pOu=%B!fk8cR3Y}*$Uq9V{+onF^*rLDniLrR5A%1<XTAdQ4wt;-?M6zhiDGx
zb>P(_dok)l`ygYJ5*O)veZAtEH1Svf#yKu0R`htrrmqVYiapEu^f3x<il6Dr*`V1{
zz;2oW$MlSl1{p0)r+s2(>d@afIx>4+Fqlr1tW6k>VD#<(I3=E%dQaObo7Uy&V-9;3
z)ZW*B_(KC-h$J5solEtvV+(v+1!Y-{2fC>a)w|UZk#9=xa*&nGc{~F-JH)xfuST`|
zTAWN=0Sm|T>Zqfub>)nol=X6o=&o2t1*KNSpP;s*+#3{j6s7A=gd)D%PBzN)cW&dx
zK$$4Noyr4?^le!=F<37hXGrft97`f>1&LTXhrm3BQ+7m%Qb0ZfF;3glF#jhl`xt!`
zo}GHcZt6a<U|{uRg@dtjasfDA?!R=!8^dahwmj1}<MnjkTYF_wXsSA?c0Jyw{wf(w
zos>=~xV?FODV=3{G_>64*MSQcuhp=n7aIs_x;~pUMHW`mxvcU7hh1}6_ue8*O1Ef9
z&(NAltU)1v+?$$GsV0K+&2FL1R7^kiB#AnsaofgTcQUkZ_IYpO?DB`_jnR)TC`VPl
zzp!A=yEdYAX}YkI#d-Yh&j9(jt#eE%Z9Ml;7g_+A>d$e^EHe3(OjK{6Y^$hB#)cor
zWHRH*)wW$M6D@z)2xh6B3?K&Xk1+`C(Y_x?UhKvMllOw{-U@Ep9=xeI$-}#<StULF
zaM<HctVE?+A7Z4YkjXmQTtulw16XX-r4iWg0p{06Nj=L<Vc=RJYCzCrH=JR%DDSd?
zwdT2#<yS-kL8oi%_w9e%erXV|iY9bZt}ft^^FMw4^T=~mHmc*wA&){4Uti}^Fd7Or
zBJ{qOmwntVom~1@GVdNcFaQqh(qDLK-Ez5>u}Idq7HK^)%npdaBJ#?02}+c*zL-0g
zIW7@xp7}l_<MR&MvkXKpKSHq9cFjqKkDM|@6Pk-xnLCz$ZS+?b&Gt^Kgqb(aFNXh1
z&-lcO75Ma^9bfMKpwS_>F(~U8(2g6<)Yxfjulv;6QwAPB4U=?UUDV~~;<O~$EGY*4
znf0PL$HAtutfZ%>M}Uf)o$LeR5)&UfpV_^`?1}K&#5w_EZ1m-|i0pT4*usO3c)X!I
zR#PADo(qn|w!2-{l=dIeWv#IyV+A#fDYMqQ6}j|Jj}`jUb>PMaO#8LL`O_ogRVPBu
zhMT;0d9MmIk0@T%0L+bK=*r1=Rr5CN#lh|NajSUwW0H3d7U{$Lbyt-2&CCZ4M8U;F
zZ5Hir{&{)b2=~PO)f>#U#5hsY4SZ)ALU5<Ec=Di~Q|ol;Y-Br0eH#0O=eJHG0{#<)
zw(KO)i*cYOcjoGVlGktmKP_*9Py&&GTtEsS1&{(r0i*y@04abJKnfrQkOD{nqySO?
pDS#9}3LpiL0!YFCrl8h-!6cShZH@{^lCJ<y(9Vz@+<>@Ke*;LjI~f1~

literal 0
HcmV?d00001

diff --git a/docs_new/overrides/home.html b/docs_new/overrides/home.html
new file mode 100644
index 000000000..9e7fdaf0b
--- /dev/null
+++ b/docs_new/overrides/home.html
@@ -0,0 +1,155 @@
+{#-
+This file overrides the home page to use HTML tooling
+better.
+-#}
+{% extends "main.html" %}
+{% block tabs %}
+{{ super() }}
+
+<style>
+    @media screen and (min-width:60em) {
+        .md-sidebar--secondary {
+            display: none;
+        }
+    }
+
+    @media screen and (min-width:76.25em) {
+        .md-sidebar--primary {
+            display: none;
+        }
+    }
+
+    .mdx-container {
+        display: flex;
+        justify-content: center;
+        align-items: center;
+        padding-left: 1rem;
+        padding-right: 1rem;
+    }
+
+    .mdx-hero {
+        text-align: center;
+        margin-top: 6rem;
+    }
+
+    .mdx-hero__image {
+        margin-bottom: 1rem;
+        margin-top: 1rem;
+        max-width: 80%;
+        margin: 0 auto;
+
+        @media screen and (max-width: 40em) {
+            max-width: 60%;
+        }
+    }
+
+    .mdx-hero__content {
+        max-width: 600px;
+        margin: 0 auto;
+
+        .subtitle {
+            font-size: 1rem;
+            letter-spacing: 0.025rem;
+        }
+
+        h2 {
+            max-width: 460px;
+        }
+
+        @media screen and (max-width: 40em) {
+            .subtitle {
+                font-size: 0.8rem;
+                max-width: 70%;
+                font-weight: 700;
+            }
+        }
+
+        a:hover{
+            color: #A7623A;
+        }
+    }
+
+    .md-buttons {
+        display: flex;
+        justify-content: center;
+        gap: 0.5rem;
+        flex-direction: row;
+        margin: 0 auto;
+        margin-top: 2rem;
+        flex-direction: row;
+        width: max-content;
+        font-weight: 700;
+    }
+
+    .md-team {
+        margin-top: 4rem;
+    }
+
+    /* New media query for smaller screens */
+    @media screen and (max-width: 40em) {
+        .md-buttons {
+            flex-direction: column;
+            /* Stack buttons vertically */
+            align-items: center;
+            /* Center-align the stacked buttons */
+        }
+
+        .md-buttons .md-button {
+            width: 100%;
+            /* Make buttons full-width */
+            max-width: 200px;
+            /* Limit maximum width for better appearance */
+        }
+    }
+
+    .md-button {
+        transition: background-color 0.3s ease-in-out; /* Smooth transition */
+        border-radius: 6px !important;
+        padding: 0.4rem 0.8rem !important;
+
+    }
+
+    .md-button:hover {
+        background-color: #A7623A !important; /* Desired hover background color */
+        border-color: #A7623A !important;
+    }
+</style>
+
+<section class="mdx-container">
+    <div class="md-grid md-typeset">
+        <div class="mdx-hero">
+            <div class="mdx-hero__image">
+                <img src="assets/images/logo-light-mode.svg#only-light" alt="Outlines Logo" width="600" draggable="false">
+                <img src="assets/images/logo-dark-mode.svg#only-dark" alt="Outlines Logo" width="600" draggable="false">
+            </div>
+            <div class="mdx-hero__content">
+                <h2 class="subtitle" style="font-weight: 500; padding-top: 1rem;">
+                    Structured text generation and robust prompting for language models
+                </h2>
+                <div class="md-buttons">
+                    <a href="{{ 'welcome/' | url }}" title="Get started" class="md-button md-button--primary">
+                        Get started
+                    </a>
+                    <a href="https://discord.gg/ZxBxyWmW5n" title="Join the Community" class="md-button">
+                        Join the community
+                    </a>
+                </div>
+                <div class="md-buttons">
+                    <p>
+                        Follow us on <a href="https://twitter.com/dottxtai" title="Follow us on X" target="_blank">
+                            X
+                        </a> and
+                        <a href="https://bsky.app/profile/dottxtai.bsky.social" title="Follow us on Bluesky" target="_blank">
+                            Bluesky
+                        </a>
+                    </p>
+                </div>
+
+                <p class="md-team">Made with ❤️ by the team at <a href="https://dottxt.co">.txt</a></p>
+            </div>
+        </div>
+    </div>
+</section>
+{% endblock %}
+{% block content %}{% endblock %}
+{% block footer %}{% endblock %}
diff --git a/docs_new/overrides/main.html b/docs_new/overrides/main.html
new file mode 100644
index 000000000..94d9808cc
--- /dev/null
+++ b/docs_new/overrides/main.html
@@ -0,0 +1 @@
+{% extends "base.html" %}
diff --git a/docs_new/reference/advanced/guide.md b/docs_new/reference/advanced/guide.md
new file mode 100644
index 000000000..08d2f7f0f
--- /dev/null
+++ b/docs_new/reference/advanced/guide.md
@@ -0,0 +1 @@
+::: outlines.processors.guide
diff --git a/docs_new/reference/advanced/parsing.md b/docs_new/reference/advanced/parsing.md
new file mode 100644
index 000000000..e9662999c
--- /dev/null
+++ b/docs_new/reference/advanced/parsing.md
@@ -0,0 +1 @@
+::: outlines.fsm.parsing
diff --git a/docs_new/reference/available_models.md b/docs_new/reference/available_models.md
new file mode 100644
index 000000000..cb9497be1
--- /dev/null
+++ b/docs_new/reference/available_models.md
@@ -0,0 +1 @@
+::: outlines.models
diff --git a/docs_new/reference/core/generator.md b/docs_new/reference/core/generator.md
new file mode 100644
index 000000000..508ba399c
--- /dev/null
+++ b/docs_new/reference/core/generator.md
@@ -0,0 +1,122 @@
+---
+title: Generator API
+---
+
+# Generator
+
+
+The `Generator` class is the core component of Outlines v1. `Generator` accepts a [model](../reference/available_models) and an optional [output type](/api/types). If no output type is provided, the `Generator` will return unstructured text.
+
+!!! note
+
+    `Generator` is new as of Outlines v1, and replaces previous generator constructors:
+
+    - `generate.cfg`
+    - `generate.choice`
+    - `generate.format`
+    - `generate.fsm`
+    - `generate.json`
+    - `generate.regex`
+    - `generate.text`
+
+## Basic Usage
+
+```python
+from outlines import Generator, from_transformers
+import transformers
+
+# Initialize a model
+model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
+
+# Create a generator for plain text
+generator = Generator(model)
+result = generator("Write a short poem about AI.")
+
+# Print the result
+print(result)
+```
+
+## Structured Generation
+
+```python
+from pydantic import BaseModel
+from outlines import Generator, from_transformers
+import transformers
+
+# Define a Pydantic model for structured output
+class BookRecommendation(BaseModel):
+    title: str
+    author: str
+    year: int
+
+# Initialize a model
+model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
+
+# Create a generator for JSON output
+generator = Generator(model, BookRecommendation)
+
+# Generate a book recommendation
+result = generator("Recommend a science fiction book.")
+
+# Parse the JSON result into a Pydantic model
+book = BookRecommendation.model_validate_json(result)
+print(f"{book.title} by {book.author} ({book.year})")
+```
+
+## Parameters
+
+- `model`: The language model to use for generation
+- `output_type`: Optional. The type of output to generate
+
+## Generation Parameters
+
+When calling the generator, you can pass additional parameters to control the generation process. These parameters are passed through to the underlying model, so they depend on the specific model being used.
+
+Common parameters for most models include:
+- `max_new_tokens`: Maximum number of tokens to generate
+- `temperature`: Controls randomness (higher values = more random)
+- `top_p`: Controls diversity via nucleus sampling
+- `stop_strings`: String or list of strings at which to stop generation
+
+Example:
+```python
+result = generator(
+    "Write a short story.",
+    max_new_tokens=200,
+    temperature=0.7,
+    top_p=0.9,
+    stop_strings=["THE END", "###"]
+)
+```
+
+## Return Value
+
+The generator always returns a raw string containing the generated text. When generating structured outputs, you need to parse this string into the desired format.
+
+Unlike in Outlines v0, where the return type could be a parsed object, in v1 you are responsible for parsing the output when needed:
+
+```python
+# Outlines v1 approach
+from pydantic import BaseModel
+from outlines import Generator
+
+class Person(BaseModel):
+    name: str
+    age: int
+
+generator = Generator(model, Person)
+result = generator("Generate a person:")
+
+# Parse the result yourself
+person = Person.model_validate_json(result)
+```
+
+::: outlines.generator.Generator
diff --git a/docs_new/reference/index.md b/docs_new/reference/index.md
new file mode 100644
index 000000000..320d43af4
--- /dev/null
+++ b/docs_new/reference/index.md
@@ -0,0 +1,3 @@
+# API Reference
+
+This needs to contain a list of all the misc API reference pages.
diff --git a/docs_new/reference/models/anthropic.md b/docs_new/reference/models/anthropic.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/reference/models/dottxt.md b/docs_new/reference/models/dottxt.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/reference/models/gemini.md b/docs_new/reference/models/gemini.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/reference/models/index.md b/docs_new/reference/models/index.md
new file mode 100644
index 000000000..64f70dbc7
--- /dev/null
+++ b/docs_new/reference/models/index.md
@@ -0,0 +1 @@
+- [Anthropic](/api/models/anthropic.md)
diff --git a/docs_new/reference/models/llamacpp.md b/docs_new/reference/models/llamacpp.md
new file mode 100644
index 000000000..51b62eca8
--- /dev/null
+++ b/docs_new/reference/models/llamacpp.md
@@ -0,0 +1,230 @@
+# Llama.cpp
+
+Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library][llamacpp]. Llamacpp allows to run quantized models on machines with limited compute.
+
+!!! Note "Installation"
+
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration. See the [installation section](#installation) for instructions to install `llama-cpp-python` with CUDA, Metal, ROCm and other backends. To get started quickly you can also run:
+
+    ```bash
+    pip install "outlines[llamacpp]"
+    ```
+
+## Load the model
+
+You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+
+```python
+from outlines import models
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+```
+
+This will download the model files to the hub cache folder and load the weights in memory.
+
+You can also initialize the model by passing the path to the weights on your machine. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
+
+```python
+from outlines import models
+from llama_cpp import Llama
+
+llm = Llama("./phi-2.Q4_K_M.gguf")
+model = models.LlamaCpp(llm)
+```
+
+If you need more control, you can pass the same keyword arguments to the model as you would pass in the [llama-ccp-library][llamacpp]:
+
+```python
+from outlines import models
+
+model = models.llamacpp(
+    "TheBloke/phi-2-GGUF",
+    "phi-2.Q4_K_M.gguf"
+    n_ctx=512,  # to set the context length value
+)
+```
+
+**Main parameters:**
+
+| Parameters | Type | Description | Default |
+|------------|------|-------------|---------|
+| `n_gpu_layers`| `int` | Number of layers to offload to GPU. If -1, all layers are offloaded | `0` |
+| `split_mode` | `int` | How to split the model across GPUs. `1` for layer-wise split, `2` for row-wise split | `1` |
+| `main_gpu` | `int` | Main GPU | `0` |
+| `tensor_split` | `Optional[List[float]]` | How split tensors should be distributed across GPUs. If `None` the model is not split. | `None` |
+| `n_ctx` | `int` | Text context. Inference from the model if set to `0` | `0` |
+| `n_threads` | `Optional[int]` | Number of threads to use for generation. All available threads if set to `None`.| `None` |
+| `verbose` | `bool` | Print verbose outputs to `stderr` | `False` |
+
+See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters.
+
+### Load the model on GPU
+
+!!! Note
+
+    [Make sure](#cuda) that you installed `llama-cpp-python` with GPU support.
+
+ To load the model on GPU, pass `n_gpu_layers=-1`:
+
+```python
+from outlines import models
+
+model = models.llamacpp(
+    "TheBloke/phi-2-GGUF",
+    "phi-2.Q4_K_M.gguf",
+    n_gpu_layers=-1,  # to use GPU acceleration
+)
+```
+
+This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
+
+### Load LoRA adapters
+
+You can load LoRA adapters dynamically:
+
+```python
+from outlines import models, generate
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+generator = generate.text(model)
+answer_1 = generator("prompt")
+
+model.load_lora("./path/to/adapter.gguf")
+answer_2 = generator("prompt")
+```
+
+To load another adapter you need to re-initialize the model. Otherwise the adapter will be added on top of the previous one:
+
+```python
+from outlines import models
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+model.load_lora("./path/to/adapter1.gguf")  # Load first adapter
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+model.load_lora("./path/to/adapter2.gguf")  # Load second adapter
+```
+
+## Generate text
+
+In addition to the parameters described in the [text generation section](../text.md) you can pass extra keyword arguments, for instance to set sampling parameters not exposed in Outlines' public API:
+
+```python
+from outlines import models, generate
+
+
+model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
+generator = generate.text(model)
+
+answer = generator("A prompt", presence_penalty=0.8)
+```
+
+**Extra keyword arguments:**
+
+The value of the keyword arguments you pass to the generator suspersede the values set when initializing the sampler or generator. All extra sampling methods and repetition penalties are disabled by default.
+
+| Parameters | Type | Description | Default |
+|------------|------|-------------|---------|
+| `suffix` | `Optional[str]` | A suffix to append to the generated text. If `None` no suffix is added. | `None` |
+| `echo` | `bool` | Whether to preprend the prompt to the completion. | `False` |
+| `seed` | `int` | The random seed to use for sampling. | `None` |
+| `max_tokens` | `Optional[int]` | The maximum number of tokens to generate. If `None` the maximum number of tokens depends on `n_ctx`. | `16` |
+| `frequence_penalty` | `float` | The penalty to apply to tokens based on their frequency in the past 64 tokens. | `0.0` |
+| `presence_penalty` | `float` | The penalty to apply to tokens based on their presence in the past 64 tokens. | `0.0` |
+| `repeat_penalty` | `float` | The penalty to apply to repeated tokens in the past 64 tokens. | `1.` |
+| `stopping_criteria` | `Optional[StoppingCriteriaList]` | A list of stopping criteria to use. | `None`
+| `logits_processor` | `Optional[LogitsProcessorList]` | A list of logits processors to use. The logits processor used for structured generation will be added to this list. | `None`
+| `temperature` | `float` | The temperature to use for sampling | `1.0` |
+| `top_p` | `float` | The top-p value to use for [nucleus sampling][degeneration]. | `1.` |
+| `min_p` | `float` | The min-p value to use for [minimum-p sampling][minimum-p]. | `0.` |
+| `typical_p` | `float` | The p value to use for [locally typical sampling][locally-typical]. | `1.0` |
+| `stop` | `Optional[Union[str, List[str]]]` | A list of strings that stop generation when encountered. | `[]` |
+| `top_k` |  `int` | The top-k value used for [top-k sampling][top-k]. Negative value to consider all logit values. | `-1.` |
+| `tfs_z` | `float` | The [tail-free sampling][tail-free] parameter. | `1.0` |
+| `mirostat_mode` | `int` | The [mirostat sampling][mirostat] mode. | `0` |
+| `mirostat_tau` | `float` | The target cross-entropy for [mirostat sampling][mirostat].| `5.0` |
+| `mirostat_eta` | `float` | The learning rate used to update `mu` in [mirostat sampling][mirostat]. | `0.1` |
+
+See the [llama-cpp-python documentation][llama-cpp-python-call] for the full and up-to-date list of parameters and the [llama.cpp code][llama-cpp-sampling-params] for the default values of other
+sampling parameters.
+
+### Streaming
+
+
+## Installation
+
+You need to install the `llama-cpp-python` library to use the llama.cpp integration.
+
+### CPU
+
+For a *CPU-only* installation run:
+
+```bash
+pip install llama-cpp-python
+```
+
+!!! Warning
+
+    Do not run this command if you want support for BLAS, Metal or CUDA. Follow the instructions below instead.
+
+### CUDA
+
+```bash
+CMAKE_ARGS="-DLLAMA_CUDA=on" pip install llama-cpp-python
+```
+
+It is also possible to install pre-built wheels with CUDA support (Python 3.10 and above):
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/<cuda-version>
+```
+
+Where `<cuda-version>` is one of the following, depending on the version of CUDA installed on your system:
+
+- `cu121` for CUDA 12.1
+- `cu122` for CUDA 12.2
+- `cu123` CUDA 12.3
+
+### Metal
+
+```bash
+CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
+```
+
+It is also possible to install pre-build wheels with Metal support (Python 3.10 or above, MacOS 11.0 and above):
+
+```bash
+pip install llama-cpp-python \
+  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/metal
+```
+
+### OpenBLAS
+
+```bash
+CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
+```
+
+### Other backend
+
+`llama.cpp` supports many other backends. Refer to the [llama.cpp documentation][llama-cpp-python-install] to use the following backends:
+
+- CLBast (OpenCL)
+- hipBLAS (ROCm)
+- Vulkan
+- Kompute
+- SYCL
+
+
+
+
+[llamacpp]: https://github.com/abetlen/llama-cpp-python
+[llama-cpp-python-call]: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
+[llama-cpp-python-install]: https://github.com/abetlen/llama-cpp-python/tree/08b16afe11e7b42adec2fed0a781123383476045?tab=readme-ov-file#supported-backends
+[llama-cpp-sampling-params]: https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
+[mirostat]: https://arxiv.org/abs/2007.14966
+[degeneration]: https://arxiv.org/abs/1904.09751
+[top-k]: https://arxiv.org/abs/1805.04833
+[minimum-p]: https://github.com/ggerganov/llama.cpp/pull/3841
+[locally-typical]: https://arxiv.org/abs/2202.00666
+[tail-free]: https://www.trentonbricken.com/Tail-Free-Sampling
diff --git a/docs_new/reference/models/mlxlm.md b/docs_new/reference/models/mlxlm.md
new file mode 100644
index 000000000..d435b9c1f
--- /dev/null
+++ b/docs_new/reference/models/mlxlm.md
@@ -0,0 +1,93 @@
+# mlx-lm
+
+Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
+
+!!! Note "Installation"
+
+    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To get started quickly you can also run:
+
+    ```bash
+    pip install "outlines[mlxlm]"
+    ```
+
+
+## Load the model
+
+You can initialize the model by passing the name of the repository on the HuggingFace Hub. The official repository for mlx-lm supported models is [mlx-community](https://huggingface.co/mlx-community).
+
+```python
+from outlines import models
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+```
+
+This will download the model files to the hub cache folder and load the weights in memory.
+
+The arguments `model_config` and `tokenizer_config` are available to modify loading behavior. For example, per the `mlx-lm` [documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms#supported-models), you must set an eos_token for `qwen/Qwen-7B`. In outlines you may do so via
+
+```
+model = models.mlxlm(
+    "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit",
+    tokenizer_config={"eos_token": "<|endoftext|>", "trust_remote_code": True},
+)
+```
+
+**Main parameters:**
+
+(Subject to change. Table based on [mlx-lm.load docstring](https://github.com/ml-explore/mlx-examples/blob/main/llms/mlx_lm/utils.py#L429))
+
+| Parameters         | Type   | Description                                                                                      | Default |
+|--------------------|--------|--------------------------------------------------------------------------------------------------|---------|
+| `tokenizer_config` | `dict` | Configuration parameters specifically for the tokenizer. Defaults to an empty dictionary.        | `{}`    |
+| `model_config`     | `dict` | Configuration parameters specifically for the model. Defaults to an empty dictionary.            | `{}`    |
+| `adapter_path`     | `str`  | Path to the LoRA adapters. If provided, applies LoRA layers to the model.                        | `None`  |
+| `lazy`             | `bool` | If False, evaluate the model parameters to make sure they are loaded in memory before returning. | `False` |
+
+
+## Generate text
+
+You may generate text using the parameters described in the [text generation documentation](../text.md).
+
+With the loaded model, you can generate text or perform structured generation, e.g.
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+generator = generate.text(model)
+
+answer = generator("A prompt", temperature=2.0)
+```
+
+## Streaming
+
+You may creating a streaming iterable with minimal changes
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+generator = generate.text(model)
+
+for token_str in generator.text("A prompt", temperature=2.0):
+    print(token_str)
+```
+
+## Structured
+
+You may perform structured generation with mlxlm to guarantee your output will match a regex pattern, json schema, or lark grammar.
+
+Example: Phone number generation with pattern `"\\+?[1-9][0-9]{7,14}"`:
+
+```python
+from outlines import models, generate
+
+model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+
+phone_number_pattern = "\\+?[1-9][0-9]{7,14}"
+generator = generate.regex(model, phone_number_pattern)
+
+model_output = generator("What's Jennys Number?\n")
+print(model_output)
+# '8675309'
+```
diff --git a/docs_new/reference/models/models.md b/docs_new/reference/models/models.md
new file mode 100644
index 000000000..eab70f2a1
--- /dev/null
+++ b/docs_new/reference/models/models.md
@@ -0,0 +1,56 @@
+---
+title: Models
+---
+
+# Models
+
+Outlines supports generation using a number of inference engines (`outlines.models`). Loading a model using outlines follows a similar interface between inference engines:
+
+```python
+import outlines
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
+model = outlines.models.transformers_vision("llava-hf/llava-v1.6-mistral-7b-hf")
+model = outlines.models.vllm("microsoft/Phi-3-mini-128k-instruct")
+model = outlines.models.llamacpp(
+    "microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf"
+)
+model = outlines.models.exllamav2("bartowski/Phi-3-mini-128k-instruct-exl2")
+model = outlines.models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
+
+model = outlines.models.openai(
+    "gpt-4o-mini",
+    api_key=os.environ["OPENAI_API_KEY"]
+)
+```
+
+
+# Feature Matrix
+|                   | [Transformers](transformers.md) | [Transformers Vision](transformers_vision.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md)* |
+|-------------------|--------------|---------------------|------|-----------|-----------|-------|---------|
+| **Device**        |              |                     |      |           |           |       |         |
+| Cuda              | ✅           | ✅                  | ✅   | ✅        | ✅        | ❌    | N/A     |
+| Apple Silicon     | ✅           | ✅                  | ❌   | ✅        | ✅        | ✅    | N/A     |
+| x86 / AMD64       | ✅           | ✅                  | ❌   | ✅        | ✅        | ❌    | N/A     |
+| **Sampling**      |              |                     |      |           |           |       |         |
+| Greedy            | ✅           | ✅                  | ✅   | ✅*       | ✅        | ✅    | ❌      |
+| Multinomial       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Multiple Samples  | ✅           | ✅                  |      | ❌        |           | ❌    | ✅      |
+| Beam Search       | ✅           | ✅                  | ✅   | ❌        | ✅        | ❌    | ❌      |
+| **Generation**    |              |                     |      |           |           |       |         |
+| Batch             | ✅           | ✅                  | ✅   | ❌        | ?         | ❌    | ❌      |
+| Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
+| Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| **Structured**    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| JSON Schema       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Choice            | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
+| Regex             | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
+| Grammar           | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
+
+
+## Caveats
+
+- OpenAI doesn't support structured generation due to limitations in their API and server implementation.
+- `outlines.generate` ["Structured"](../generation/generation.md) includes methods such as `outlines.generate.regex`, `outlines.generate.json`, `outlines.generate.cfg`, etc.
+- MLXLM only supports Apple Silicon.
+- llama.cpp greedy sampling available via multinomial with `temperature = 0.0`.
diff --git a/docs_new/reference/models/ollama.md b/docs_new/reference/models/ollama.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/reference/models/openai.md b/docs_new/reference/models/openai.md
new file mode 100644
index 000000000..0777ad8b7
--- /dev/null
+++ b/docs_new/reference/models/openai.md
@@ -0,0 +1,204 @@
+# OpenAI and compatible APIs
+
+!!! Installation
+
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively:
+
+    ```bash
+    pip install "outlines[openai]"
+    ```
+
+## OpenAI models
+
+Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.openai`:
+
+```python
+from outlines import models
+
+
+model = models.openai("gpt-4o-mini")
+model = models.openai("gpt-4o")
+```
+
+Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.AsyncOpenAI` as keyword arguments:
+
+```python
+import os
+from outlines import models
+
+
+model = models.openai(
+    "gpt-4o-mini",
+    api_key=os.environ["OPENAI_API_KEY"]
+)
+```
+
+The following table enumerates the possible parameters. Refer to the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list.
+
+**Parameters:**
+
+| **Parameters** | **Type** | **Description** | **Default** |
+|----------------|:---------|:----------------|:------------|
+| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
+| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
+| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
+| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if no specified. | `None` |
+| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
+| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
+| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
+| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
+| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
+
+## Azure OpenAI models
+
+Outlines also supports Azure OpenAI models:
+
+```python
+from outlines import models
+
+
+model = models.azure_openai(
+    "azure-deployment-name",
+    "gpt-4o-mini",
+    api_version="2024-07-18",
+    azure_endpoint="https://example-endpoint.openai.azure.com",
+)
+```
+
+!!! Question "Why do I need to specify model and deployment name?"
+
+    The model name is needed to load the correct tokenizer for the model. The tokenizer is necessary for structured generation.
+
+
+You can pass any parameter you would pass to `openai.AsyncAzureOpenAI`. You can consult the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/lib/azure.py) for an up-to-date list.
+
+**Parameters:**
+
+
+| **Parameters** | **Type** | **Description** | **Default** |
+|----------------|:---------|:----------------|:------------|
+| `azure_endpoint` | `str` | Azure endpoint, including the resource. Infered from `AZURE_OPENAI_ENDPOINT` if not specified | `None` |
+| `api_version` | `str` | API version. Infered from `AZURE_OPENAI_API_KEY` if not specified | `None` |
+| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
+| `azure_ad_token` | `str` | Azure active directory token. Inference from `AZURE_OPENAI_AD_TOKEN` if not specified | `None` |
+| `azure_ad_token_provider` | `AzureADTokenProvider` | A function that returns an Azure Active Directory token | `None` |
+| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
+| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
+| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if not specified. | `None` |
+| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
+| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
+| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
+| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
+| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
+
+## Models that follow the OpenAI standard
+
+Outlines supports models that follow the OpenAI standard. You will need to initialize the OpenAI client properly configured and pass it to `outlines.models.openai`
+
+```python
+import os
+from openai import AsyncOpenAI
+from outlines import models
+from outlines.models.openai import OpenAIConfig
+
+
+client = AsyncOpenAI(
+    api_key=os.environ.get("PROVIDER_KEY"),
+    base_url="http://other.provider.server.com"
+)
+config = OpenAIConfig("model_name")
+model = models.openai(client, config)
+```
+
+!!! Warning
+
+    You need to pass the async client to be able to do batch inference.
+
+## Structured Generation Support
+
+Outlines provides support for [OpenAI Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode) via `outlines.generate.json`, `outlines.generate.choice`
+
+```python
+from pydantic import BaseModel, ConfigDict
+import outlines.models as models
+from outlines import generate
+
+model = models.openai("gpt-4o-mini")
+
+class Person(BaseModel):
+    model_config = ConfigDict(extra='forbid')  # required for openai
+    first_name: str
+    last_name: str
+    age: int
+
+generator = generate.json(model, Person)
+generator("current indian prime minister on january 1st 2023")
+# Person(first_name='Narendra', last_name='Modi', age=72)
+
+generator = generate.choice(model, ["Chicken", "Egg"])
+print(generator("Which came first?"))
+# Chicken
+```
+
+!!! Warning
+
+    Structured generation support only provided to OpenAI-compatible endpoints which conform to OpenAI's standard. Additionally, `generate.regex` and `generate.cfg` are not supported.
+
+
+## Advanced configuration
+
+For more advanced configuration option, such as support proxy, please consult the [OpenAI SDK's documentation](https://github.com/openai/openai-python):
+
+
+```python
+from openai import AsyncOpenAI, DefaultHttpxClient
+from outlines import models
+from outlines.models.openai import OpenAIConfig
+
+
+client = AsyncOpenAI(
+    base_url="http://my.test.server.example.com:8083",
+    http_client=DefaultHttpxClient(
+        proxies="http://my.test.proxy.example.com",
+        transport=httpx.HTTPTransport(local_address="0.0.0.0"),
+    ),
+)
+config = OpenAIConfig("model_name")
+model = models.openai(client, config)
+```
+
+It is possible to specify the values for `seed`, `presence_penalty`, `frequence_penalty`, `top_p` by passing an instance of `OpenAIConfig` when initializing the model:
+
+```python
+from outlines.models.openai import OpenAIConfig
+from outlines import models
+
+
+config = OpenAIConfig(
+    presence_penalty=1.,
+    frequency_penalty=1.,
+    top_p=.95,
+    seed=0,
+)
+model = models.openai("gpt-4o-mini", config)
+```
+
+## Monitoring API use
+
+It is important to be able to track your API usage when working with OpenAI's API. The number of prompt tokens and completion tokens is directly accessible via the model instance:
+
+```python
+from openai import AsyncOpenAI
+import outlines.models
+
+
+model = models.openai("gpt-4o")
+
+print(model.prompt_tokens)
+# 0
+
+print(model.completion_tokens)
+# 0
+```
+
+These numbers are updated every time you call the model.
diff --git a/docs_new/reference/models/sglang.md b/docs_new/reference/models/sglang.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/reference/models/tgi.md b/docs_new/reference/models/tgi.md
new file mode 100644
index 000000000..3f0568103
--- /dev/null
+++ b/docs_new/reference/models/tgi.md
@@ -0,0 +1,3 @@
+# Text-generation-inference (TGI)
+
+TGI uses Outlines to provide structured generation, see [their documentation](https://huggingface.co/docs/text-generation-inference/en/basic_tutorials/using_guidance).
diff --git a/docs_new/reference/models/transformers.md b/docs_new/reference/models/transformers.md
new file mode 100644
index 000000000..3649b9b13
--- /dev/null
+++ b/docs_new/reference/models/transformers.md
@@ -0,0 +1,148 @@
+# transformers
+
+
+!!! Installation
+
+    You need to install the `transformer`, `datasets` and `torch` libraries to be able to use these models in Outlines, or alternatively:
+
+    ```bash
+    pip install "outlines[transformers]"
+    ```
+
+
+Outlines provides an integration with the `torch` implementation of causal models in the [transformers][transformers] library. You can initialize the model by passing its name:
+
+```python
+from outlines import models
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+```
+
+If you need more fine-grained control you can also initialize the model and tokenizer separately:
+
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from outlines import models
+
+llm = AutoModelForCausalLM.from_pretrained("gpt2", output_attentions=True)
+tokenizer = AutoTokenizer.from_pretrained("gpt2")
+model = models.Transformers(llm, tokenizer)
+```
+
+## Using Logits Processors
+
+There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
+
+1. Use Outlines generation wrapper, `outlines.models.transformers`
+2. Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
+
+Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
+
+### Using `outlines.models.transformers`
+
+```python
+import outlines
+
+time_regex_pattern = r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?"
+
+model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
+generator = outlines.generate.regex(model, time_regex_pattern)
+
+output = generator("The the best time to visit a dentist is at ")
+print(output)
+# 2:30 pm
+```
+
+### Using models initialized via the `transformers`  library
+
+```python
+import outlines
+import transformers
+
+
+model_uri = "microsoft/Phi-3-mini-4k-instruct"
+
+outlines_tokenizer = outlines.models.TransformerTokenizer(
+    transformers.AutoTokenizer.from_pretrained(model_uri)
+)
+phone_number_logits_processor = outlines.processors.RegexLogitsProcessor(
+    "\\+?[1-9][0-9]{7,14}",  # phone number pattern
+    outlines_tokenizer,
+)
+
+generator = transformers.pipeline('text-generation', model=model_uri)
+
+output = generator(
+    "Jenny gave me her number it's ",
+	logits_processor=transformers.LogitsProcessorList([phone_number_logits_processor])
+)
+print(output)
+# [{'generated_text': "Jenny gave me her number it's 2125550182"}]
+# not quite 8675309 what we expected, but it is a valid phone number
+```
+
+[transformers]: https://github.com/huggingface/transformers
+
+
+## Alternative Model Classes
+
+`outlines.models.transformers` defaults to `transformers.AutoModelForCausalLM`, which is the appropriate class for most standard large language models, including Llama 3, Mistral, Phi-3, etc.
+
+However other variants with unique behavior can be used as well by passing the appropriate class.
+
+### Mamba
+
+[Mamba](https://github.com/state-spaces/mamba) is a transformers alternative which employs memory efficient, linear-time decoding.
+
+To use Mamba with outlines you must first install the necessary requirements:
+```
+pip install causal-conv1d>=1.2.0 mamba-ssm torch transformers
+```
+
+Then you can either create an Mamba-2 Outlines model via
+```python
+import outlines
+
+model = outlines.models.mamba("state-spaces/mamba-2.8b-hf")
+```
+
+or explicitly with
+```python
+import outlines
+from transformers import MambaForCausalLM
+
+model = outlines.models.transformers(
+    "state-spaces/mamba-2.8b-hf",
+    model_class=MambaForCausalLM
+)
+```
+
+
+
+Read [`transformers`'s documentation](https://huggingface.co/docs/transformers/en/model_doc/mamba) for more information.
+
+### Encoder-Decoder Models
+
+You can use encoder-decoder (seq2seq) models like T5 and BART with Outlines.
+
+Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
+
+T5 Example:
+```python
+import outlines
+from transformers import AutoModelForSeq2SeqLM
+
+model_pile_t5 = outlines.models.transformers(
+    model_name="EleutherAI/pile-t5-large",
+    model_class=AutoModelForSeq2SeqLM,
+)
+```
+
+Bart Example:
+```python
+model_bart = outlines.models.transformers(
+    model_name="facebook/bart-large",
+    model_class=AutoModelForSeq2SeqLM,
+)
+```
diff --git a/docs_new/reference/models/vllm.md b/docs_new/reference/models/vllm.md
new file mode 100644
index 000000000..8789b588e
--- /dev/null
+++ b/docs_new/reference/models/vllm.md
@@ -0,0 +1,235 @@
+# vLLM
+
+
+!!! Note "Installation"
+
+    You need to install the `vllm` library to use the vLLM integration. See the [installation section](#installation) for instructions to install vLLM for CPU or ROCm. To get started you can also run:
+
+    ```bash
+    pip install "outlines[vllm]"
+    ```
+
+## Load the model
+
+Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
+
+
+```python
+from outlines import models
+
+model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
+```
+
+Or alternatively:
+
+```python
+import vllm
+from outlines import models
+
+llm = vllm.LLM("microsoft/Phi-3-mini-4k-instruct")
+model = models.VLLM(llm)
+```
+
+
+Models are loaded from the [HuggingFace hub](https://huggingface.co/).
+
+
+!!! Warning "Device"
+
+    The default installation of vLLM only allows to load models on GPU. See the [installation instructions](#installation) to run models on CPU.
+
+
+You can pass any parameter that you would normally pass to `vllm.LLM`, as keyword arguments:
+
+```python
+from outlines import models
+
+model = models.vllm(
+    "microsoft/Phi-3-mini-4k-instruct",
+    trust_remote_code=True,
+    gpu_memory_utilization=0.7
+)
+```
+
+**Main parameters:**
+
+| **Parameters** | **Type** | **Description** | **Default** |
+|----------------|:---------|:----------------|:------------|
+| `tokenizer_mode`| `str`  | "auto" will use the fast tokenizer if available and "slow" will always use the slow tokenizer. | `auto`
+| `trust_remote_code`| `bool` | Trust remote code when downloading the model and tokenizer. | `False` |
+| `tensor_parallel_size`| `int` | The number of GPUs to use for distributed execution with tensor parallelism.| `1` |
+| `dtype`| `str` | The data type for the model weights and activations. Currently, we support `float32`, `float16`, and `bfloat16`. If `auto`, we use the `torch_dtype` attribute specified in the model config file. However, if the `torch_dtype` in the config is `float32`, we will use `float16` instead.| `auto` |
+| `quantization`| `Optional[str]` | The method used to quantize the model weights. Currently, we support "awq", "gptq" and "squeezellm". If None, we first check the `quantization_config` attribute in the model config file. If that is None, we assume the model weights are not quantized and use `dtype` to determine the data type of the weights.| `None` |
+| `revision`| `Optional[str]` | The specific model version to use. It can be a branch name, a tag name, or a commit id.| `None` |
+| `tokenizer_revision`| `Optional[str]`| The specific tokenizer version to use. It can be a branch name, a tag name, or a commit id.| `None` |
+| `gpu_memory_utilization`| `float` | The ratio (between 0 and 1) of GPU memory to reserve for the model weights, activations, and KV cache. Higher values will increase the KV cache size and thus improve the model's throughput. However, if the value is too high, it may cause out-of-memory (OOM) errors.| `0.9` |
+| `swap_space`| `int` | The size (GiB) of CPU memory per GPU to use as swap space. This can be used for temporarily storing the states of the requests when their `best_of` sampling parameters are larger than 1. If all requests will have `best_of=1`, you can safely set this to 0. Otherwise, too small values may cause out-of-memory (OOM) errors.| 4 |
+| `enforce_eager`| `bool` | Whether to enforce eager execution. If True, we will disable CUDA graph and always execute the model in eager mode. If False, we will use CUDA graph and eager execution in hybrid.| `False` |
+| `enable_lora` | `bool` | Whether to enable loading LoRA adapters | `False` |
+
+See the [vLLM code](https://github.com/vllm-project/vllm/blob/8f44facdddcf3c704f7d6a2719b6e85efc393449/vllm/entrypoints/llm.py#L72) for a list of all the available parameters.
+
+### Use quantized models
+
+vLLM supports AWQ, GPTQ and SqueezeLLM quantized models:
+
+
+```python
+from outlines import models
+
+model = models.vllm("TheBloke/Llama-2-7B-Chat-AWQ", quantization="awq")
+model = models.vllm("TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", quantization="gptq")
+model = models.vllm("https://huggingface.co/squeeze-ai-lab/sq-llama-30b-w4-s5", quantization="squeezellm")
+```
+
+!!! Warning "Dependencies"
+
+    To use AWQ model you need to install the autoawq library `pip install autoawq`.
+
+    To use GPTQ models you need to install the autoGTPQ and optimum libraries `pip install auto-gptq optimum`.
+
+
+### Multi-GPU usage
+
+To run multi-GPU inference with vLLM you need to set the `tensor_parallel_size` argument to the number of GPUs available when initializing the model. For instance to run inference on 2 GPUs:
+
+
+```python
+from outlines import models
+
+model = models.vllm(
+    "microsoft/Phi-3-mini-4k-instruct"
+    tensor_parallel_size=2
+)
+```
+
+### Load LoRA adapters
+
+You can load LoRA adapters and alternate between them dynamically:
+
+```python
+from outlines import models
+
+model = models.vllm("facebook/opt-350m", enable_lora=True)
+model.load_lora("ybelkaa/opt-350m-lora")  # Load LoRA adapter
+model.load_lora(None)  # Unload LoRA adapter
+```
+
+## Generate text
+
+In addition to the parameters described in the [text generation section](../text.md) you can pass an instance of `SamplingParams` directly to any generator via the `sampling_params` keyword argument:
+
+```python
+from vllm.sampling_params import SamplingParams
+from outlines import models, generate
+
+
+model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+
+params = SamplingParams(n=2, frequency_penalty=1., min_tokens=2)
+answer = generator("A prompt", sampling_params=params)
+```
+
+This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
+
+!!! Note
+
+    The values passed via the `SamplingParams` instance supersede the other arguments to the generator or the samplers.
+
+**`SamplingParams` attributes:**
+
+| Parameters | Type             | Description            | Default |
+|:-----------|------------------|:-----------------------|---------|
+| `n` | `int` | Number of output sequences to return for the given prompt. | `1` |
+| `best_of` | `Optional[int]` | Number of output sequences that are generated from the prompt. From these `best_of` sequences, the top `n` sequences are returned. `best_of` must be greater than or equal to `n`. This is treated as the beam width when `use_beam_search` is True. By default, `best_of` is set to `n`. | `None` |
+| `presence_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens.| `0.0` |
+| `frequency_penalty` | `float` | Float that penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens. | `0.0`
+| `repetition_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the prompt and the generated text so far. Values > 1 encourage the model to use new tokens, while values < 1 encourage the model to repeat tokens. | `1.0` |
+| `temperature` | `float` | Float that controls the randomness of the sampling. Lower values make the model more deterministic, while higher values make the model more random. Zero means greedy sampling. | `1.0` |
+| `top_p` | `float` |  Float that controls the cumulative probability of the top tokens to consider. Must be in (0, 1]. Set to 1 to consider all tokens. | `1.0` |
+| `top_k` | `int` |  Integer that controls the number of top tokens to consider. Set to -1 to consider all tokens. | `-1` |
+| `min_p` |`float` |  Float that represents the minimum probability for a token to be considered, relative to the probability of the most likely token. Must be in [0, 1]. Set to 0 to disable this. | `0.0` |
+| `seed` | `Optional[int]` | Random seed to use for the generation. | `None` |
+| `use_beam_search` | `bool` |  Whether to use beam search instead of sampling. | `False` |
+| `length_penalty` | `float` | Float that penalizes sequences based on their length. Used in beam search.  | `1.0` |
+| `early_stopping` | `Union[bool, str]` |  Controls the stopping condition for beam search. It accepts the following values: `True`, where the generation stops as soon as there are `best_of` complete candidates; `False`, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; `"never"`, where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm). | `False` |
+| `stop` | `Optional[Union[str, List[str]]]` |  List of strings that stop the generation when they are generated. The returned output will not contain the stop strings. | `None` |
+| `stop_token_ids` | `Optional[List[int]]` |  List of tokens that stop the generation when they are generated. The returned output will contain the stop tokens unless the stop tokens are special tokens. | `None` |
+| `include_stop_str_in_output` | `bool` |  Whether to include the stop strings in output text. Defaults to False. | `False` |
+| `ignore_eos` | `bool` |  Whether to ignore the EOS token and continue generating tokens after the EOS token is generated. | `False` |
+| `max_tokens` | `int` |  Maximum number of tokens to generate per output sequence. | `16` |
+| `min_tokens` | `int` | Minimum number of tokens to generate per output sequence before EOS or stop_token_ids can be generated | `0` |
+| `skip_special_tokens` | `bool` | Whether to skip special tokens in the output. | `True` |
+| `spaces_between_special_tokens` | `bool` |  Whether to add spaces between special tokens in the output.  Defaults to True. | `True` |
+
+### Streaming
+
+!!! Warning
+
+    Streaming is not available for the offline vLLM integration.
+
+
+## Installation
+
+By default the vLLM library is installed with pre-commpiled C++ and CUDA binaries and will only run on GPU:
+
+```python
+pip install vllm
+```
+
+### CPU
+
+You need to have the `gcc` compiler installed on your system. Then you will need to install vLLM from source. First clone the repository:
+
+```bash
+git clone https://github.com/vllm-project/vllm.git
+cd vllm
+```
+
+Install the Python packages needed for the installation:
+
+```bash
+pip install --upgrade pip
+pip install wheel packaging ninja setuptools>=49.4.0 numpy
+pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+and finally run:
+
+```bash
+VLLM_TARGET_DEVICE=cpu python setup.py install
+```
+
+See the [vLLM documentation][vllm-install-cpu] for more details, alternative installation methods (Docker) and performance tips.
+
+### ROCm
+
+
+You will need to install vLLM from source. First install Pytorch on ROCm:
+
+```bash
+pip install torch==2.2.0.dev20231206+rocm5.7 --index-url https://download.pytorch.org/whl/nightly/rocm5.7 # tested version
+```
+
+You will then need to install flash attention for ROCm following [these instructions][rocm-flash-attention]. You can then install `xformers=0.0.23` and apply the patches needed to adapt Flash Attention for ROCm:
+
+```bash
+pip install xformers==0.0.23 --no-deps
+bash patch_xformers.rocm.sh
+```
+
+And finally build vLLM:
+
+```bash
+cd vllm
+pip install -U -r requirements-rocm.txt
+python setup.py install # This may take 5-10 minutes.
+```
+
+See the [vLLM documentation][vllm-install-rocm] for alternative installation methods (Docker).
+
+
+[vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
+[vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
+[rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
diff --git a/docs_new/reference/models/vllm_offline.md b/docs_new/reference/models/vllm_offline.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/reference/types/cfg.md b/docs_new/reference/types/cfg.md
new file mode 100644
index 000000000..12e38fd50
--- /dev/null
+++ b/docs_new/reference/types/cfg.md
@@ -0,0 +1,113 @@
+---
+title: Context-Free Grammars
+---
+
+# Context-Free Grammars
+
+::: outlines.types.cfg
+
+Context-free grammars (CFGs) provide a powerful way to define complex structured outputs. They're particularly useful for generating text with a specific syntax, such as code snippets or custom formats.
+
+## Basic Usage
+
+```python
+from outlines import Generator, from_transformers
+from outlines.types import CFG
+
+# Define a grammar for simple arithmetic expressions
+grammar = """
+start: expr
+expr: term "+" expr | term
+term: factor "*" term | factor
+factor: NUMBER | "(" expr ")"
+NUMBER: /[0-9]+/
+"""
+
+# Create a CFG output type
+cfg_output = CFG(grammar)
+
+# Initialize a model
+model = from_transformers(...)
+
+# Create a generator with the CFG output type
+generator = Generator(model, cfg_output)
+
+# Generate an arithmetic expression
+prompt = "Generate a simple arithmetic expression:"
+result = generator(prompt)
+
+print(result)  # Example output: "3 * (4 + 5)"
+```
+
+## Grammar Definition
+
+Context-free grammars in Outlines use the [Lark grammar format](https://lark-parser.readthedocs.io/en/latest/grammar/). A grammar consists of rules that define the structure of the text you want to generate.
+
+```python
+# Grammar for a simple SQL query
+sql_grammar = """
+start: select_stmt
+select_stmt: "SELECT" columns "FROM" table ("WHERE" condition)?
+columns: column ("," column)*
+column: IDENTIFIER
+table: IDENTIFIER
+condition: IDENTIFIER comparison_op value
+comparison_op: "=" | ">" | "<" | ">=" | "<=" | "!="
+value: NUMBER | STRING
+IDENTIFIER: /[a-zA-Z_][a-zA-Z0-9_]*/
+NUMBER: /[0-9]+/
+STRING: /"[^"]*"/
+"""
+```
+
+## Advanced Usage
+
+Context-free grammars can be used for complex generation tasks like code generation:
+
+```python
+from outlines import Generator, from_transformers
+from outlines.types import CFG
+
+# Grammar for simple Python functions
+python_grammar = """
+start: function
+function: "def" IDENTIFIER "(" params? ")" ":" NEWLINE INDENT body DEDENT
+params: param ("," param)*
+param: IDENTIFIER
+body: statement+
+statement: assignment NEWLINE | if_stmt | return_stmt NEWLINE
+assignment: IDENTIFIER "=" expr
+if_stmt: "if" expr ":" NEWLINE INDENT body DEDENT
+return_stmt: "return" expr
+expr: IDENTIFIER | NUMBER | STRING | expr "+" expr | expr "-" expr
+IDENTIFIER: /[a-zA-Z_][a-zA-Z0-9_]*/
+NUMBER: /[0-9]+/
+STRING: /"[^"]*"/
+INDENT: /\t/
+DEDENT: /\t/
+NEWLINE: /\n/
+"""
+
+# Create a CFG output type
+cfg_output = CFG(python_grammar)
+
+# Initialize a model
+model = from_transformers(...)
+
+# Create a generator
+generator = Generator(model, cfg_output)
+
+# Generate a Python function
+prompt = "Generate a Python function that calculates the factorial of a number:"
+result = generator(prompt)
+
+print(result)
+```
+
+For more detailed examples, see the [Context-Free Grammar guide](/user_guide/structured_generation/cfg.md).
+
+## Notes
+
+- Context-Free Grammar support in Outlines is experimental but powerful
+- Complex grammars may take longer to process
+- For simple structures, consider using regex or JSON schema instead
diff --git a/docs_new/reference/types/index.md b/docs_new/reference/types/index.md
new file mode 100644
index 000000000..d3d1d3632
--- /dev/null
+++ b/docs_new/reference/types/index.md
@@ -0,0 +1,36 @@
+---
+title: Output Types
+---
+
+# Output Types
+
+Outlines provides several ways to define structured outputs for language models.
+
+## Available Output Types
+
+### JSON Schema
+
+JSON Schema allows you to define the structure of JSON data that you want to generate. Outlines supports multiple ways to define JSON schemas:
+
+- JSON Schema objects
+- Pydantic models
+- TypedDict
+- Dataclasses
+
+[Learn more about JSON Schema →](json_schema.md)
+
+### Regular Expressions
+
+Regular expressions provide a way to constrain the model's output to match specific text patterns. This is useful for generating formatted data like dates, phone numbers, or other structured text.
+
+[Learn more about Regular Expressions →](regex.md)
+
+### Context-Free Grammars
+
+Context-free grammars (CFGs) provide the most flexible way to define complex structured outputs. They're particularly useful for generating text with a specific syntax, such as code or custom formats.
+
+[Learn more about Context-Free Grammars →](cfg.md)
+
+## How Output Types Work
+
+Internally, Outlines converts output type definitions into finite state machines (FSMs) that constrain the model's generation process. This happens automatically when you pass an output type to the `Generator` class.
diff --git a/docs_new/reference/types/json_schema.md b/docs_new/reference/types/json_schema.md
new file mode 100644
index 000000000..74d068262
--- /dev/null
+++ b/docs_new/reference/types/json_schema.md
@@ -0,0 +1,81 @@
+---
+title: JSON Schema
+---
+
+# JSON Schema
+
+::: outlines.types.json_schema
+
+JSON Schema provides a way to define the structure of JSON data that you want to generate. Outlines supports multiple ways to define JSON schemas.
+
+## Basic Usage
+
+```python
+from outlines import Generator, from_transformers
+from outlines.types import JsonSchema
+
+# Define a simple JSON schema
+person_schema = {
+    "type": "object",
+    "properties": {
+        "name": {"type": "string"},
+        "age": {"type": "integer"},
+        "email": {"type": "string"}
+    },
+    "required": ["name", "age"]
+}
+
+# Create a JSON schema output type
+json_output = JsonSchema(person_schema)
+
+# Initialize a model
+model = from_transformers(...)
+
+# Create a generator with the JSON schema output type
+generator = Generator(model, json_output)
+
+# Generate JSON data
+prompt = "Generate information for a fictional person:"
+result = generator(prompt)
+
+print(result)
+# Example output: {"name": "John Doe", "age": 32, "email": "john.doe@example.com"}
+```
+
+## Using Pydantic Models
+
+```python
+from pydantic import BaseModel
+from typing import List, Optional
+from outlines import Generator, from_transformers
+
+# Define a Pydantic model
+class Person(BaseModel):
+    name: str
+    age: int
+    email: Optional[str] = None
+
+# Initialize a model
+model = from_transformers(...)
+
+# Create a generator with the Pydantic model
+generator = Generator(model, Person)
+
+# Generate JSON data
+prompt = "Generate information for a fictional person:"
+result = generator(prompt)
+
+# Parse the result into a Pydantic model
+person = Person.model_validate_json(result)
+print(f"Name: {person.name}, Age: {person.age}")
+```
+
+## Other Schema Types
+
+Outlines also supports:
+
+- TypedDict
+- Dataclasses
+- Direct JSON Schema objects
+
+For more detailed examples, see the [JSON Generation guide](/user_guide/structured_generation/json.md).
diff --git a/docs_new/reference/types/regex.md b/docs_new/reference/types/regex.md
new file mode 100644
index 000000000..b3e9c94ec
--- /dev/null
+++ b/docs_new/reference/types/regex.md
@@ -0,0 +1,72 @@
+---
+title: Regular Expressions
+---
+
+# Regular Expressions
+
+::: outlines.types.regex
+
+Regular expressions provide a way to constrain the model's output to match specific text patterns.
+
+## Basic Usage
+
+```python
+from outlines import Generator, from_transformers
+from outlines.types import Regex
+
+# Create a regex pattern for a North American phone number
+phone_regex = Regex(r"\(\d{3}\) \d{3}-\d{4}")
+
+# Initialize a model
+model = from_transformers(...)
+
+# Create a generator with the regex output type
+generator = Generator(model, phone_regex)
+
+# Generate a phone number
+prompt = "Generate a random phone number in the format (XXX) XXX-XXXX:"
+result = generator(prompt)
+
+print(result)  # Example output: "(555) 123-4567"
+```
+
+## Complex Patterns
+
+```python
+from outlines import Generator, from_transformers
+from outlines.types import Regex
+
+# Create a regex pattern for an email address
+email_regex = Regex(r"[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}")
+
+# Initialize a model
+model = from_transformers(...)
+
+# Create a generator with the regex output type
+generator = Generator(model, email_regex)
+
+# Generate an email address
+prompt = "Generate a professional email address for John Doe:"
+result = generator(prompt)
+
+print(result)  # Example output: "john.doe@example.com"
+```
+
+## Creating a Regex Type
+
+The `Regex` class accepts a string containing a regular expression pattern:
+
+```python
+from outlines.types import Regex
+
+# Simple pattern for a 5-digit US ZIP code
+zip_code = Regex(r"\d{5}")
+
+# Pattern for a date in YYYY-MM-DD format
+date_format = Regex(r"\d{4}-\d{2}-\d{2}")
+
+# Pattern for a URL
+url_pattern = Regex(r"https?://[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}(/[a-zA-Z0-9./_-]*)?")
+```
+
+For more detailed examples, see the [Regular Expression guide](/user_guide/structured_generation/regex.md).
diff --git a/docs_new/reference/utility/application.md b/docs_new/reference/utility/application.md
new file mode 100644
index 000000000..47f455ff2
--- /dev/null
+++ b/docs_new/reference/utility/application.md
@@ -0,0 +1,76 @@
+---
+title: Application API
+---
+
+# Application
+
+The `Application` class enables you to encapsulate a prompt template and an output type into a reusable component.
+
+## Overview
+
+An `Application` combines a prompt template with an output type, creating a reusable component that can be applied to different models.
+
+## Basic Usage
+
+```python
+from outlines import Application, Template, from_transformers
+from typing import List
+
+# Create a template
+template = Template("""
+You are an expert classifier.
+Classify the following text into one of these categories: {categories}
+
+Text: {text}
+""")
+
+# Create an application with the template and output type
+classifier = Application(template, str)
+
+# Initialize a model
+model = from_transformers(...)
+
+# Use the application with specific parameters
+categories = ["Sports", "Politics", "Technology", "Entertainment"]
+result = classifier(model, categories=categories, text="Apple unveils new iPhone with AI capabilities")
+
+print(result)  # Expected output: "Technology"
+```
+
+## Structured Output
+
+```python
+from pydantic import BaseModel
+from outlines import Application, Template, from_transformers
+
+# Define a Pydantic model for structured output
+class MovieReview(BaseModel):
+    title: str
+    director: str
+    year: int
+    rating: float
+    review: str
+
+# Create a template
+template = Template("""
+Write a brief review of the movie {movie_title}.
+Include the director, year of release, and a rating out of 10.
+""")
+
+# Create an application with the template and output type
+movie_reviewer = Application(template, MovieReview)
+
+# Use the application with a model
+result = movie_reviewer(model, movie_title="The Matrix")
+
+# Parse the result into a Pydantic model
+review = MovieReview.model_validate_json(result)
+print(f"{review.title} ({review.year}) - {review.rating}/10")
+```
+
+## Parameters
+
+- `prompt_template`: A template that defines the prompt structure
+- `output_type`: The type of output to generate
+
+::: outlines.applications.Application
diff --git a/docs_new/reference/utility/template.md b/docs_new/reference/utility/template.md
new file mode 100644
index 000000000..3a9cb72a8
--- /dev/null
+++ b/docs_new/reference/utility/template.md
@@ -0,0 +1,116 @@
+---
+title: Template API
+---
+
+# Template
+
+::: outlines.templates.Template
+
+Templates provide a way to create reusable prompt structures with placeholders for dynamic content.
+
+## Basic Usage
+
+```python
+from outlines import Template
+
+# Create a simple template
+template = Template("Hello, {name}! The weather today is {weather}.")
+
+# Fill in the template
+prompt = template.format(name="Alice", weather="sunny")
+print(prompt)  # Output: "Hello, Alice! The weather today is sunny."
+```
+
+## Composing Templates
+
+Templates can be nested and composed:
+
+```python
+from outlines import Template
+
+# Create component templates
+user_template = Template("User: {query}")
+system_template = Template("System: {instruction}")
+
+# Create a composed template
+chat_template = Template("""
+{system}
+
+{user}
+""")
+
+# Fill in nested templates
+prompt = chat_template.format(
+    system=system_template.format(instruction="You are a helpful assistant."),
+    user=user_template.format(query="What is machine learning?")
+)
+
+print(prompt)
+# Output:
+# System: You are a helpful assistant.
+#
+# User: What is machine learning?
+```
+
+## Using with Generator
+
+Templates are useful when combined with the Generator class:
+
+```python
+from outlines import Template, Generator, from_transformers
+
+# Create a template
+template = Template("""
+Translate the following {source_language} text to {target_language}:
+
+{text}
+
+Translation:
+""")
+
+# Initialize a model
+model = from_transformers(...)
+
+# Create a generator
+generator = Generator(model)
+
+# Use the template with the generator
+translation = generator(
+    template.format(
+        source_language="English",
+        target_language="French",
+        text="Hello, how are you today?"
+    )
+)
+
+print(translation)  # Output: "Bonjour, comment allez-vous aujourd'hui?"
+```
+
+## Multimodal Templates
+
+For multimodal inputs, use the `Vision` class:
+
+```python
+from outlines import Template, Generator, from_transformers, Vision
+from PIL import Image
+
+# Create a vision template
+template = Template("""
+Describe what you see in this image: {image}
+""")
+
+# Load an image
+image = Image.open("example.jpg")
+
+# Format the template with the image
+prompt = template.format(image=Vision(image))
+
+# Initialize a multimodal model
+model = from_transformers(...)
+
+# Generate a description
+generator = Generator(model)
+description = generator(prompt)
+
+print(description)
+```
diff --git a/docs_new/stylesheets/extra.css b/docs_new/stylesheets/extra.css
new file mode 100644
index 000000000..381f5aa11
--- /dev/null
+++ b/docs_new/stylesheets/extra.css
@@ -0,0 +1,22 @@
+@font-face {
+    font-family: "Source Code Pro Custom", monospace;
+    src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
+}
+
+:root > * {
+    /* Accent color shades */
+    /* --md-accent-fg-color: #d36237;
+    --md-accent-fg-color--transparent: #d36237;
+    --md-accent-bg-color: hsla(0, 0%, 100%, 1);
+    --md-accent-bg-color--light: hsla(0, 0%, 100%, 0.7); */
+
+    /* Primary color shades */
+    /* --md-accent-fg-color: hsla(#d36237, 1);
+    --md-accent-fg-color--transparent: hsla(#d36237, 0.1);
+    --md-accent-bg-color: #d36237
+    --md-accent-bg-color--light: hsla(#d36237, 0.7); */
+
+    --md-text-font-family: "Inter";
+    --md-code-font: "Source Code Pro Custom";
+    /* --md-typeset-a-color: #d36237; /*this is the brand color */
+}
diff --git a/docs_new/user_guide/chat_templating.md b/docs_new/user_guide/chat_templating.md
new file mode 100644
index 000000000..e3f33fbfd
--- /dev/null
+++ b/docs_new/user_guide/chat_templating.md
@@ -0,0 +1,38 @@
+# Chat templating
+
+Instruction-tuned language models use "special tokens" to indicate different parts of text, such as the system prompt, the user prompt, any images, and the assistant's response. A [chat template](https://huggingface.co/docs/transformers/main/en/chat_templating) is how different types of input are composited together into a single, machine-readable string.
+
+Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself -- if you do not apply chat templating on instruction-tuned models, you will often get nonsensical output from the model.
+
+Chat template tokens are not needed for base models.
+
+You can find the chat template tokens in the model's HuggingFace repo or documentation. As an example, the `SmolLM2-360M-Instruct` special tokens can be found [here](https://huggingface.co/HuggingFaceTB/SmolLM2-360M-Instruct/blob/main/special_tokens_map.json).
+
+However, it can be slow to manually look up a model's special tokens, and special tokens vary by models. If you change the model, your prompts may break if you have hard-coded special tokens.
+
+If you need a convenient tool to apply chat templating for you, you should use the `tokenizer` from the `transformers` library:
+
+```python
+from transformers import AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/SmolLM2-360M-Instruct")
+prompt = tokenizer.apply_chat_template(
+    [
+        {"role": "system", "content": "You extract information from text."},
+        {"role": "user", "content": "What food does the following text describe?"},
+    ],
+    tokenize=False,
+    add_bos=True,
+    add_generation_prompt=True,
+)
+```
+
+yields
+
+```
+<|im_start|>system
+You extract information from text.<|im_end|>
+<|im_start|>user
+What food does the following text describe?<|im_end|>
+<|im_start|>assistant
+```
diff --git a/docs_new/user_guide/index.md b/docs_new/user_guide/index.md
new file mode 100644
index 000000000..bd7da2bbf
--- /dev/null
+++ b/docs_new/user_guide/index.md
@@ -0,0 +1 @@
+This page should provide links and an overview of the user guide section. The user guide should function as a complete reference for how to do and use structured generation.
diff --git a/docs_new/user_guide/migration.md b/docs_new/user_guide/migration.md
new file mode 100644
index 000000000..f931641f7
--- /dev/null
+++ b/docs_new/user_guide/migration.md
@@ -0,0 +1,244 @@
+# Outlines 1.0 migration guide
+
+Outlines 1.0 introduces some breaking changes that affect the way you use the library. You are likely concerned by all of the following sections, so please read this document carefully until the end.
+
+This guide will help you migrate your code to the new version.
+
+All previous functionality will be supported until Outlines version 1.5.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
+
+## Removed or modified features
+- [Generate functions](#generate-functions)
+- [Models](#models)
+- [Samplers](#samplers)
+- [Functions](#functions)
+- [Text generation return types](#text-generation-return-types)
+- [Inference arguments](#inference-arguments)
+
+### Generate functions
+
+The whole `generate` module has been removed. That includes the functions `generate.cfg`, `generate.choice`, `generate.format`,`generate.fsm`, `generate.json`, `generate.regex` and `generate.text`.
+
+You should replace these functions by the `Generator` object along with the right output type as an argument (on top of the model). The output type can either be a python type or be an object from the `outlines.types` module. You can find more information about the output types in the [output types](TODO: link to the documentatoin on output types) guide.
+
+Associated v1 output types for each deprecated function:
+- `generate.cfg` -> `outlines.types.CFG`
+- `generate.choice` -> `typing.Literal` or `typing.Union`
+- `generate.format` -> native python types (`str`, `int` etc.)
+- `generate.fsm` -> `outlines.types.FSM`
+- `generate.json` -> `pydantic.BaseModel`, `typing.TypedDict`, `dataclasses.dataclass`, `genson.schema.SchemaBuilder` or `outlines.types.JsonSchema`
+- `generate.regex` -> `outlines.types.Regex`
+- `generate.text` -> no output type (`None`)
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = ...
+generator = generate.choice(model, ["foo", "bar"])
+```
+
+You should now use:
+
+```python
+from typing import Literal
+from outlines import Generator
+
+model = ...
+generator = Generator(model, Literal["foo", "bar"])
+```
+
+### Models
+
+The model classes found in the `outlines.models` module are maintained but there are a few important changes to be aware of.
+
+The functions used to created a model have been replaced by equivalent functions named with a `from_` prefix. The function `outlines.models.transformers` has been replaced by `outlines.from_transformers` for instance. On top of this change of name, the arguments have been modified. You should refer to the [models documentation](TODO: link to the documentation on models) for more details, but the overall idea is that you now need to provide a model/client instance from the inference library the Outlines model is wrapping.
+
+For instance, instead of:
+
+```python
+from outlines import models
+
+model = models.llamacpp(
+    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+)
+```
+
+You should now do:
+
+```python
+from llama_cpp import Llama
+from outlines import from_llamacpp
+
+llamacpp_model = Llama.from_pretrained(
+    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+)
+model = from_llamacpp(llamacpp_model)
+```
+
+The `load_lora` methods that are present on the `VLLM` and `LlamaCpp` models have been removed. You should now handle lora loading through the `Llama` instance in the case of the `LlamaCpp` model or provide it as a keyword argument when calling the model in the case of the `VLLM` model.
+
+For instance, instead of:
+
+```python
+from outlines import from_vllm
+from vllm import LLM
+
+model = from_vllm(
+    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
+)
+model.load_lora("path/to/lora/file")
+
+response = model("foo")
+```
+
+You should now do:
+
+```python
+from outlines import from_vllm
+from vllm import LLM
+from vllm.lora.request import LoRARequest
+
+model = from_vllm(
+    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
+)
+lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
+
+response = model("foo", lora_request=lora_request)
+```
+
+The `ExLlamav2` model has been removed without replacement. This inference library is not fully compatible with Outlines, so we decided to remove it. You can still use it until final deprecation, but we recommend you to migrate to a different inference library right now.
+
+### Samplers
+
+The `outlines.samplers` module has been removed without replacement. You should now use the arguments of the inference library model to control the sampling. Depending on the model you use, this could be done at initialization or when calling the model to generate text (so when calling the outlines model or a generator).
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = <transformers_model>
+
+generator = generate.text(model, samplers.beam_search(2))
+response = generator("foo")
+```
+
+You should now do:
+
+```python
+from outlines import Generator
+
+model = <transformers_model>
+
+generator = Generator(model)
+response = generator("foo", num_beams=2)
+```
+
+### Functions
+
+The `outlines.function` module has been removed. It is replaced by the `outlines.applications` module. An `Application` serves a similar purpose as a `Function`: it encapsulates a prompt template and an output type. A difference is that can `Application` is not instantiated with a model name. Instead, you should provide a model instance along with the prompt when calling it.
+
+For instance, instead of:
+
+```python
+from outlines import Function
+
+prompt_template = ...
+output_type = ...
+
+fn = Function(
+    prompt_template,
+    output_type,
+    "hf-internal-testing/tiny-random-GPTJForCausalLM",
+)
+
+result = fn("foo")
+```
+
+You should now do:
+
+```python
+from outlines import Application
+
+prompt_template = ...
+output_type = ...
+
+application = Application(
+    prompt_template,
+    output_type,
+)
+
+model = ...
+
+result = application(model, "foo")
+```
+
+### Text generation return types
+
+In the previous version of Outlines, the return type of the generators depended on the output type provided. For instance, if you passed a Pydantic model to the `generate.json` function, the return type was a Pydantic model instance. In the v1, the return type of a generator is always a `str`, the raw text generated by the model. You are responsible for parsing the text into the desired format.
+
+For instance, instead of:
+
+```python
+from pydantic import BaseModel
+from outlines import generate
+
+class Foo(BaseModel):
+    bar: str
+
+model = ...
+
+generator = generate.json(model, Foo)
+
+result = generator("foo")
+print(result.bar)
+```
+
+You should now do:
+
+```python
+from pydantic import BaseModel
+from outlines import Generator
+
+class Foo(BaseModel):
+    bar: str
+
+model = ...
+
+generator = Generator(model, Foo)
+
+result = generator("foo")
+result = Foo.model_validate_json(result) # parse the text into the Pydantic model instance
+print(result.bar)
+```
+
+### Inference arguments
+
+In the previous version of Outlines, some of the inference arguments were standardized across the models and were provided as positional arguments to the generator or through the sampling params dictionary. Additionally, various default values were added by outlines to the inference library models. This is no longer the case. You should refer to the documentation of the inference library you use to find the right arguments for your use case and pass them as keyword arguments to the outlines generator when calling it.
+
+For instance, instead of:
+
+```python
+from outlines import generate
+
+model = <transformers_model>
+
+generator = generate.text(model)
+
+result = generator("foo", 256, ".", 10) # 256 tokens, stop at "." and seed 10
+```
+
+You should now do:
+
+```python
+from outlines import Generator
+
+model = <transformers_model>
+
+generator = Generator(model)
+
+result = generator("foo", max_new_tokens=256, stop_strings=".", seed=10)
+```
diff --git a/docs_new/user_guide/prompting/applications.md b/docs_new/user_guide/prompting/applications.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/user_guide/prompting/templates.md b/docs_new/user_guide/prompting/templates.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/user_guide/regex_dsl.md b/docs_new/user_guide/regex_dsl.md
new file mode 100644
index 000000000..f42b6dd34
--- /dev/null
+++ b/docs_new/user_guide/regex_dsl.md
@@ -0,0 +1,344 @@
+# DSL to express constraints
+
+This library provides a Domain-Specific Language (DSL) to construct regular expressions in a more intuitive and modular way. It allows you to create complex regexes using simple building blocks that represent literal strings, patterns, and various quantifiers. Additionally, these custom regex types can be used directly as types in [Pydantic](https://pydantic-docs.helpmanual.io/) schemas to enforce pattern constraints during text generation.
+
+---
+
+## Why Use This DSL?
+
+1. **Modularity & Readability**: Instead of writing cryptic regular expression strings, you compose a regex as a tree of objects.
+2. **Enhanced Debugging**: Each expression can be visualized as an ASCII tree, making it easier to understand and debug complex regexes.
+3. **Pydantic Integration**: Use your DSL-defined regex as types in Pydantic models. The DSL seamlessly converts to JSON Schema with proper pattern constraints.
+4. **Extensibility**: Easily add or modify quantifiers and other regex components by extending the provided classes.
+
+---
+
+## Building Blocks
+
+
+Every regex component in this DSL is a **Term**. Here are two primary types:
+
+- **`String`**: Represents a literal string. It escapes the characters that have a special meaning in regular expressions.
+- **`Regex`**: Represents an existing regex pattern string.
+
+```python
+from outlines.types import String, Regex
+
+# A literal string "hello"
+literal = String("hello")   # Internally represents "hello"
+
+# A regex pattern to match one or more digits
+digit = Regex(r"[0-9]+")     # Internally represents the pattern [0-9]+
+
+# Converting to standard regex strings:
+from outlines.types.dsl import to_regex
+
+print(to_regex(literal))  # Output: hello
+print(to_regex(digit))    # Output: [0-9]+
+```
+
+---
+
+## Early Introduction to Quantifiers & Combining Terms
+
+The DSL supports common regex quantifiers as methods on every `Term`. These methods allow you to specify how many times a pattern should be matched. They include:
+
+- **`exactly(count)`**: Matches the term exactly `count` times.
+- **`optional()`**: Matches the term zero or one time.
+- **`one_or_more()`**: Matches the term one or more times (Kleene Plus).
+- **`zero_or_more()`**: Matches the term zero or more times (Kleene Star).
+- **`between(min_count, max_count)`**: Matches the term between `min_count` and `max_count` times (inclusive).
+- **`at_least(count)`**: Matches the term at least `count` times.
+- **`at_most(count)`**: Matches the term up to `count` times.
+
+These quantifiers can also be used as functions that take the `Term` as an argument. If the term is a plain string, it will be automatically converted to a `String` object. Thus `String("foo").optional()` is equivalent to `optional("foo")`.
+
+Let's see these quantifiers side by side with examples.
+
+### Quantifiers in Action
+
+#### `exactly(count)`
+
+This method restricts the term to appear exactly `count` times.
+
+```python
+# Example: exactly 5 digits
+five_digits = Regex(r"\d").exactly(5)
+print(to_regex(five_digits))  # Output: (\d){5}
+```
+
+You can also use the `exactly` function:
+
+```python
+from outlines.types import exactly
+
+# Example: exactly 5 digits
+five_digits = exactly(Regex(r"\d"), 5)
+print(to_regex(five_digits))  # Output: (\d){5}
+```
+
+#### `optional()`
+
+This method makes a term optional, meaning it may occur zero or one time.
+
+```python
+# Example: an optional "s" at the end of a word
+maybe_s = String("s").optional()
+print(to_regex(maybe_s))  # Output: (s)?
+```
+
+You can also use the `optional` function:
+
+```python
+from outlines.types import optional
+
+# Example: an optional "s" at the end of a word
+maybe_s = optional("s")
+print(to_regex(maybe_s))  # Output: (s)?
+```
+
+#### `one_or_more()`
+
+This method indicates that the term must appear at least once.
+
+```python
+# Example: one or more alphabetic characters
+letters = Regex(r"[A-Za-z]").one_or_more()
+print(to_regex(letters))  # Output: ([A-Za-z])+
+```
+
+You can also use the `one_or_more` function:
+
+```python
+from outlines.types import one_or_more
+
+# Example: one or more alphabetic characters
+letters = one_or_more(Regex(r"[A-Za-z]"))
+print(to_regex(letters))  # Output: ([A-Za-z])+
+
+```
+
+#### `zero_or_more()`
+
+This method indicates that the term can occur zero or more times.
+
+```python
+# Example: zero or more spaces
+spaces = String(" ").zero_or_more()
+print(to_regex(spaces))  # Output: ( )*
+```
+
+You can also use the `zero_or_more` function:
+
+```python
+from outlines.types import zero_or_more
+
+# Example: zero or more spaces
+spaces = zero_or_more(" ")
+print(to_regex(spaces))  # Output: ( )*
+```
+
+#### `between(min_count, max_count)`
+
+This method indicates that the term can appear any number of times between `min_count` and `max_count` (inclusive).
+
+```python
+# Example: Between 2 and 4 word characters
+word_chars = Regex(r"\w").between(2, 4)
+print(to_regex(word_chars))  # Output: (\w){2,4}
+```
+
+You can also use the `between` function:
+
+```python
+from outlines.types import between
+
+# Example: Between 2 and 4 word characters
+word_chars = between(Regex(r"\w"), 2, 4)
+print(to_regex(word_chars))  # Output: (\w){2,4}
+```
+
+#### `at_least(count)`
+
+This method indicates that the term must appear at least `count` times.
+
+```python
+# Example: At least 3 digits
+at_least_three = Regex(r"\d").at_least(3)
+print(to_regex(at_least_three))  # Output: (\d){3,}
+```
+
+You can also use the `at_least` function:
+
+```python
+from outlines.types import at_least
+
+# Example: At least 3 digits
+at_least_three = at_least(Regex(r"\d"), 3)
+print(to_regex(at_least_three))  # Output: (\d){3,}
+```
+
+#### `at_most(count)`
+
+This method indicates that the term can appear at most `count` times.
+
+```python
+# Example: At most 3 digits
+up_to_three = Regex(r"\d").at_most(3)
+print(to_regex(up_to_three))  # Output: (\d){0,3}
+```
+
+You can also use the `at_most` function:
+
+```python
+from outlines.types import at_most
+
+# Example: At most 3 digits
+up_to_three = at_most(Regex(r"\d"), 3)
+print(to_regex(up_to_three))  # Output: (\d){0,3}
+```
+
+---
+
+## Combining Terms
+
+The DSL allows you to combine basic terms into more complex patterns using concatenation and alternation.
+
+### Concatenation (`+`)
+
+The `+` operator (and its reflected variant) concatenates terms, meaning that the terms are matched in sequence.
+
+```python
+# Example: Match "hello world"
+pattern = String("hello") + " " + Regex(r"\w+")
+print(to_regex(pattern))  # Output: hello\ (\w+)
+```
+
+### Alternation (`either()`)
+
+The `either()` function creates alternatives, allowing a match for one of several patterns. You can provide as many terms as you want.
+
+```python
+# Example: Match either "cat" or "dog" or "mouse"
+animal = either(String("cat"), "dog", "mouse")
+print(to_regex(animal))  # Output: (cat|dog|mouse)
+```
+
+*Note:* When using `either()` with plain strings (such as `"dog"`), the DSL automatically wraps them in a `String` object that escapes the characters that have a special meaning in regular expressions, just like with quantifier functions.
+
+---
+
+## Custom types
+
+The DSL comes "batteries included" with types that represent common text constructs:
+
+- `integer` represents an integer number as recognized by `int`
+- `boolean` represents a boolean, "True" or "False" as recognized by `bool`
+- `number` represents a floating-point number recognize by Python's `float`
+- `date` represents a date as understood by `datetime.date`
+- `time` represents a time as undestood by `datetime.time`
+- `datetime` represents a time as understoof by `datetime.datetime`
+- `digit` represents a single digit
+- `char` represents a single character
+- `newline` represents a new line character
+- `whitespace` represents a white space
+- `sentence` represents a sentence
+- `paragraph` reprensents a pagraph (one or more sentences separated by one or more ilne breaks)
+
+
+For instance you can describe the answers in the GSM8K dataset using the following pattern:
+
+```python
+from outlines.types import sentence, digit
+
+answer = "A: " + sentence.between(2,4) + " So the answer is: " + digit.between(1,4)
+```
+
+---
+
+## Practical Examples
+
+### Example 1: Matching a Custom ID Format
+
+Suppose you want to create a regex that matches an ID format like "ID-12345", where:
+- The literal "ID-" must be at the start.
+- Followed by exactly 5 digits.
+
+```python
+id_pattern = "ID-" + Regex(r"\d").exactly(5)
+print(to_regex(id_pattern))  # Output: ID-(\d){5}
+```
+
+### Example 2: Email Validation with Pydantic
+
+You can define a regex for email validation and use it as a type in a Pydantic model.
+
+```python
+from pydantic import BaseModel, ValidationError
+
+# Define an email regex term (this is a simplified version)
+email_regex = Regex(r"[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+")
+
+class User(BaseModel):
+    name: str
+    email: email_regex  # Use our DSL regex as a field type
+
+# Valid input
+user = User(name="Alice", email="alice@example.com")
+print(user)
+
+# Invalid input (raises a ValidationError)
+try:
+    User(name="Bob", email="not-an-email")
+except ValidationError as e:
+    print(e)
+```
+
+When used in a Pydantic model, the email field is automatically validated against the regex pattern and its JSON Schema includes the `pattern` constraint.
+
+### Example 3: Building a Complex Pattern
+
+Consider a pattern to match a simple date format: `YYYY-MM-DD`.
+
+```python
+year = Regex(r"\d").exactly(4)         # Four digits for the year
+month = Regex(r"\d").exactly(2)        # Two digits for the month
+day = Regex(r"\d").exactly(2)          # Two digits for the day
+
+# Combine with literal hyphens
+date_pattern = year + "-" + month + "-" + day
+print(to_regex(date_pattern))
+# Output: (\d){4}\-(\d){2}\-(\d){2}
+```
+
+---
+
+## Visualizing Your Pattern
+
+One of the unique features of this DSL is that each term can print its underlying structure as an ASCII tree. This visualization can be particularly helpful when dealing with complex expressions.
+
+```python
+# A composite pattern using concatenation and quantifiers
+pattern = "a" + String("b").one_or_more() + "c"
+print(pattern)
+```
+
+*Expected Output:*
+
+```
+└── Sequence
+    ├── String('a')
+    ├── KleenePlus(+)
+    │   └── String('b')
+    └── String('c')
+```
+
+This tree representation makes it easy to see the hierarchy and order of operations in your regular expression.
+
+---
+
+## Final Words
+
+This DSL is designed to simplify the creation and management of regular expressions—whether you're validating inputs in a web API, constraining the output of an LLM, or just experimenting with regex patterns. With intuitive methods for common quantifiers and operators, clear visual feedback, and built-in integration with Pydantic, you can build robust and maintainable regex-based validations with ease.
+
+Feel free to explore the library further and adapt the examples to your use cases. Happy regexing!
diff --git a/docs_new/user_guide/selecting_an_inference_backend.md b/docs_new/user_guide/selecting_an_inference_backend.md
new file mode 100644
index 000000000..2d52941fd
--- /dev/null
+++ b/docs_new/user_guide/selecting_an_inference_backend.md
@@ -0,0 +1,5 @@
+This guide should provide a general overview of the available models in the [API reference](/api/models/).
+
+## Models
+
+-  [Anthropic](/api/models/anthropic)
diff --git a/docs_new/user_guide/structured_generation/cfg.md b/docs_new/user_guide/structured_generation/cfg.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/user_guide/structured_generation/index.md b/docs_new/user_guide/structured_generation/index.md
new file mode 100644
index 000000000..541a7e1f4
--- /dev/null
+++ b/docs_new/user_guide/structured_generation/index.md
@@ -0,0 +1,13 @@
+---
+title: Structured Generation
+---
+
+# Structured Generation
+
+Outlines allows you to generate structured outputs from LLMs. There are several ways to specify the structure of the output you want to generate.
+
+### JSON schema
+
+### Pydantic models
+
+### Context-free grammars
diff --git a/docs_new/user_guide/structured_generation/json.md b/docs_new/user_guide/structured_generation/json.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/user_guide/structured_generation/multiple_choice.md b/docs_new/user_guide/structured_generation/multiple_choice.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/user_guide/structured_generation/regex.md b/docs_new/user_guide/structured_generation/regex.md
new file mode 100644
index 000000000..e69de29bb
diff --git a/docs_new/user_guide/unstructured.md b/docs_new/user_guide/unstructured.md
new file mode 100644
index 000000000..9046cf4d1
--- /dev/null
+++ b/docs_new/user_guide/unstructured.md
@@ -0,0 +1 @@
+Brief overview of unstructured text generation.
diff --git a/mkdocs.yml b/mkdocs.yml
index 8ec7ac838..20001418a 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -2,7 +2,7 @@
 site_name: Outlines
 site_author: The Outlines developers
 site_description: >-
-  Structured text generation with LLMs
+    Structured text generation with LLMs
 
 # Repository
 repo_name: dottxt-ai/outlines
@@ -11,171 +11,228 @@ repo_url: https://github.com/dottxt-ai/outlines
 # Copyright
 copyright: Copyright &copy; 2023- The Outlines Developers
 
+# Documentation directory
+docs_dir: docs_new
+
 # Configuration
 theme:
-  name: material
-  custom_dir: docs/overrides
-  palette:
-    - scheme: default
-      primary: white
-  logo: assets/images/logo-square.svg
-  favicon: assets/images/logo-simple.png
-  icon:
-    repo: fontawesome/brands/github
-  features:
-    - content.code.copy
-    - navigation.expand
-    - navigation.tabs
-    - navigation.sections
-    - header.autohide
-    - announce.dismiss
-  font:
-    text: Inter
-    code: Source Code Pro
+    name: material
+    palette:
+        # Palette toggle for light mode
+        - media: "(prefers-color-scheme: light)"
+          scheme: default
+          primary: white
+    logo: assets/images/logo-square.svg
+    favicon: assets/images/logo-simple.png
+    icon:
+        repo: fontawesome/brands/github
+    features:
+        - content.code.copy
+        - navigation.expand
+        - navigation.tabs
+        - navigation.sections
+        - header.autohide
+        - announce.dismiss
+    font:
+        text: Inter
+        code: Source Code Pro
 
 # Additional configuration
 extra:
-  social:
-    - icon: fontawesome/brands/github
-      link: https://github.com/dottxt-ai
-    - icon: fontawesome/brands/twitter
-      link: https://twitter.com/remilouf
-  generator: false
-  analytics:
-    provider: google
-    property: !ENV GOOGLE_ANALYTICS_KEY
-  version:
-    provider: mike
-    default: latest
-    alias: true
+    social:
+        - icon: fontawesome/brands/github
+          link: https://github.com/dottxt-ai
+        - icon: fontawesome/brands/twitter
+          link: https://twitter.com/remilouf
+    generator: false
+    analytics:
+        provider: google
+        property: !ENV GOOGLE_ANALYTICS_KEY
+    version:
+        provider: mike
+        default: latest
+        alias: true
 
 # Extensions
 markdown_extensions:
-  - admonition
-  - def_list
-  - attr_list
-  - md_in_html
-  - pymdownx.highlight:
-      anchor_linenums: true
-      line_spans: __span
-      pygments_lang_class: true
-      noclasses: True
-      pygments_style: nord
-  - pymdownx.superfences:
-      custom_fences:
-        - name: mermaid
-          class: mermaid
-          format: !!python/name:pymdownx.superfences.fence_code_format
-  - pymdownx.tabbed:
-      alternate_style: true
-  - pymdownx.inlinehilite
-  - pymdownx.details
-  - pymdownx.emoji:
-      emoji_index: !!python/name:material.extensions.emoji.twemoji
-      emoji_generator: !!python/name:material.extensions.emoji.to_svg
-  - pymdownx.snippets:
+    - admonition
+    - def_list
+    - attr_list
+    - md_in_html
+    - pymdownx.highlight:
+          anchor_linenums: true
+          line_spans: __span
+          pygments_lang_class: true
+          noclasses: True
+    - pymdownx.superfences:
+          custom_fences:
+              - name: mermaid
+                class: mermaid
+                format: !!python/name:pymdownx.superfences.fence_code_format
+    - pymdownx.tabbed:
+          alternate_style: true
+    - pymdownx.inlinehilite
+    - pymdownx.details
+    - pymdownx.emoji:
+          emoji_index: !!python/name:material.extensions.emoji.twemoji
+          emoji_generator: !!python/name:material.extensions.emoji.to_svg
+    - pymdownx.snippets:
 
 extra_css:
-  - stylesheets/extra.css
+    - stylesheets/extra.css
 
 plugins:
-  - blog
-  - mkdocstrings:
-      default_handler: python
-      handlers:
-        python:
-          options:
-            docstring_style: numpy
-            show_submodules: true
-  - search
-  - section-index
-  - social:
-      cards_layout_options:
-        color: #173a58
-  - git-committers:
-      repository: dottxt-ai/outlines
-      branch: main
-  - git-revision-date-localized:
-      enable_creation_date: true
-      type: timeago
+    - blog
+    - mkdocstrings:
+          default_handler: python
+          handlers:
+              python:
+                  options:
+                      docstring_style: numpy
+                      show_submodules: true
+    - search
+    - section-index
+    - social:
+          cards_layout_options:
+              color: #173a58
+    - redirects:
+          redirect_maps:
+              "welcome.md": "index.md"
+              "reference/chat_templating.md": "user_guide/chat_templating.md"
+              "blog.md": "https://blog.dottxt.co"
+              "cookbook/index.md": "examples/index.md"
+              "cookbook/classification.md": "examples/classification.md"
+              "cookbook/extraction.md": "examples/extraction.md"
+              "cookbook/dating_profiles.md": "examples/dating_profiles.md"
+              "cookbook/chain_of_density.md": "examples/chain_of_density.md"
+              "cookbook/models_playing_chess.md": "examples/models_playing_chess.md"
+              "cookbook/simtom.md": "examples/simtom.md"
+              "cookbook/qa-with-citations.md": "examples/qa-with-citations.md"
+              "cookbook/knowledge_graph_extraction.md": "examples/knowledge_graph_extraction.md"
+              "cookbook/structured_generation_workflow.md": "examples/structured_generation_workflow.md"
+              "cookbook/chain_of_thought.md": "examples/chain_of_thought.md"
+              "cookbook/react_agent.md": "examples/react_agent.md"
+              "cookbook/atomic_caption.md": "examples/atomic_caption.md"
+              "cookbook/read-pdfs.md": "examples/read-pdfs.md"
+              "cookbook/earnings-reports.md": "examples/earnings-reports.md"
+              "cookbook/receipt-digitization.md": "examples/receipt-digitization.md"
+              "cookbook/extract_event_details.md": "examples/extract_event_details.md"
+              "cookbook/deploy-using-bentoml.md": "examples/deploy-using-bentoml.md"
+              "cookbook/deploy-using-cerebrium.md": "examples/deploy-using-cerebrium.md"
+              "cookbook/deploy-using-modal.md": "examples/deploy-using-modal.md"
+              
+              # Root files
+              "quickstart.md": "getting_started/index.md"
+              
+              # API to reference redirects
+              "api/index.md": "reference/index.md"
+              "api/json_schema.md": "reference/types/json_schema.md"
+              "api/regex.md": "reference/types/regex.md"
+              "api/applications.md": "reference/utility/application.md"
+              "api/templates.md": "reference/utility/template.md"
+              "api/guide.md": "reference/advanced/guide.md"
+              "api/parsing.md": "reference/advanced/parsing.md"
+              "api/models.md": "reference/available_models.md"
+              
+              # API subdirectories to reference
+              "api/types/": "reference/types/index.md"
+              "api/core/": "reference/index.md"
+              "api/advanced/": "reference/index.md"
+              
+    - git-committers:
+        repository: dottxt-ai/outlines
+        branch: main
+    - git-revision-date-localized:
+        enable_creation_date: true
+        type: timeago
 
 nav:
-  - Home: index.md
-  - Getting Started:
-      - Welcome to Outlines: welcome.md
-      - Installation: installation.md
-      - Quickstart: quickstart.md
-      - Licence: licence.md
-  - Cookbook:
-      - cookbook/index.md
-      - Classification: cookbook/classification.md
-      - Named Entity Extraction: cookbook/extraction.md
-      - Generate synthetic data: cookbook/dating_profiles.md
-      - Summarize a document: cookbook/chain_of_density.md
-      - Playing chess: cookbook/models_playing_chess.md
-      - Perspective-taking prompting: cookbook/simtom.md
-      - Question-answering with citations: cookbook/qa-with-citations.md
-      - Knowledge Graph Extraction: cookbook/knowledge_graph_extraction.md
-      - Structured Generation Workflow: cookbook/structured_generation_workflow.md
-      - Chain of Thought (CoT): cookbook/chain_of_thought.md
-      - ReAct Agent: cookbook/react_agent.md
-      - Vision-Language Models: cookbook/atomic_caption.md
-      - Structured Generation from PDFs: cookbook/read-pdfs.md
-      - Earnings reports to CSV: cookbook/earnings-reports.md
-      - Digitizing receipts with vision models: cookbook/receipt-digitization.md
-      - Extract events details from text: cookbook/extract_event_details.md
-      - Run on the cloud:
-          - BentoML: cookbook/deploy-using-bentoml.md
-          - Cerebrium: cookbook/deploy-using-cerebrium.md
-          - Modal: cookbook/deploy-using-modal.md
-  - Docs:
-      - reference/index.md
-      - Generation:
-          - Overview: reference/generation/generation.md
-          - Chat templating: reference/chat_templating.md
-          - Regex DSL: reference/regex_dsl.md
-          - Text: reference/text.md
-          - Structured generation:
-              - How does it work?: reference/generation/structured_generation_explanation.md
-              - Classification: reference/generation/choices.md
-              - Regex: reference/generation/regex.md
-              - Type constraints: reference/generation/format.md
-              - JSON (function calling): reference/generation/json.md
-              - Grammar: reference/generation/cfg.md
-              - Creating Grammars: reference/generation/creating_grammars.md
-              - Custom FSM operations: reference/generation/custom_fsm_ops.md
-      - Utilities:
-          - Serve with vLLM: reference/serve/vllm.md
-          - Serve with LM Studio: reference/serve/lmstudio.md
-          - Custom types: reference/generation/types.md
-          - Prompt templating: reference/prompting.md
-          - Outlines functions: reference/functions.md
-      - Models:
-          - Overview: reference/models/models.md
-          - Open source:
-              - Transformers: reference/models/transformers.md
-              - Transformers Vision: reference/models/transformers_vision.md
-              - Llama.cpp: reference/models/llamacpp.md
-              - vLLM: reference/models/vllm.md
-              - TGI: reference/models/tgi.md
-              - MLX: reference/models/mlxlm.md
-              - Mamba: reference/models/transformers/#mamba
-          - API:
-              - OpenAI: reference/models/openai.md
-  - API Reference:
-      - api/index.md
-      - api/models.md
-      - api/templates.md
-      - api/json_schema.md
-      - api/guide.md
-      - api/parsing.md
-      - api/regex.md
-      - api/samplers.md
-  - Community:
-      - community/index.md
-      - Feedback 🫶: community/feedback.md
-      - Chat with us ☕: https://discord.com/invite/R9DSu34mGd
-      - How to contribute 🏗️: community/contribute.md
-      - Your projects 👏: community/examples.md
-      - Versioning Guide 📌: community/versioning.md
-  - Blog: https://blog.dottxt.co
+    - Home:
+          - index.md
+          - User Guide: user_guide/index.md
+          - API Reference: reference/index.md
+          - Examples: examples/index.md
+    - Guide:
+          - Introduction:
+                - Get started: getting_started/index.md
+                - Installation: installation.md
+                - Licence: licence.md
+          - User Guide:
+                - The Basics: user_guide/index.md
+                - Selecting an Inference Engine: user_guide/selecting_an_inference_backend.md
+                - Unstructured Text Generation: user_guide/unstructured.md
+                - Structured Generation: user_guide/structured_generation/index.md
+                - Chat Templating for Instruct Models: user_guide/chat_templating.md
+                - Generation Formats:
+                      - JSON: user_guide/structured_generation/json.md
+                      - Regular expressions: user_guide/structured_generation/regex.md
+                      - Context-free grammars: user_guide/structured_generation/cfg.md
+                      - Multiple choice: user_guide/structured_generation/multiple_choice.md
+                - Prompting:
+                      - Applications: user_guide/prompting/applications.md
+                      - Templates: user_guide/prompting/templates.md
+                - Easy Regular Expressions: user_guide/regex_dsl.md
+                - Migrating to v1: user_guide/migration.md
+    - API Reference:
+          - Overview: reference/index.md
+          - Format types:
+                - Overview: reference/types/index.md
+                - JSON: reference/types/json_schema.md
+                - Regex: reference/types/regex.md
+                - Context-free grammars: reference/types/cfg.md
+
+          - Core:
+                - Generator: reference/core/generator.md
+
+          - Supported Models:
+                - Anthropic: reference/models/anthropic.md
+                - Dottxt: reference/models/dottxt.md
+                - Gemini: reference/models/gemini.md
+                - Llamacpp: reference/models/llamacpp.md
+                - Mlx-lm: reference/models/mlxlm.md
+                - Ollama: reference/models/ollama.md
+                - OpenAI: reference/models/openai.md
+                - SgLang: reference/models/sglang.md
+                - TGI: reference/models/tgi.md
+                - Transformers: reference/models/transformers.md
+                - vLLM (online server): reference/models/vllm.md
+                - vLLM (offline): reference/models/vllm_offline.md
+
+          - Utilities:
+                - Applications: reference/utility/application.md
+                - Template: reference/utility/template.md
+
+          - Advanced:
+                - Guide: reference/advanced/guide.md
+                - Parsing: reference/advanced/parsing.md
+    - Examples:
+          - examples/index.md
+          - Classification: examples/classification.md
+          - Named Entity Extraction: examples/extraction.md
+          - Generate synthetic data: examples/dating_profiles.md
+          - Summarize a document: examples/chain_of_density.md
+          - Playing chess: examples/models_playing_chess.md
+          - Perspective-taking prompting: examples/simtom.md
+          - Question-answering with citations: examples/qa-with-citations.md
+          - Knowledge Graph Extraction: examples/knowledge_graph_extraction.md
+          - Structured Generation Workflow: examples/structured_generation_workflow.md
+          - Chain of Thought (CoT): examples/chain_of_thought.md
+          - ReAct Agent: examples/react_agent.md
+          - Vision-Language Models: examples/atomic_caption.md
+          - Structured Generation from PDFs: examples/read-pdfs.md
+          - Earnings reports to CSV: examples/earnings-reports.md
+          - Digitizing receipts with vision models: examples/receipt-digitization.md
+          - Extract events details from text: examples/extract_event_details.md
+          - Run on the cloud:
+                - BentoML: examples/deploy-using-bentoml.md
+                - Cerebrium: examples/deploy-using-cerebrium.md
+                - Modal: examples/deploy-using-modal.md
+    - Community:
+        - community/index.md
+        - Feedback 🫶: community/feedback.md
+        - Our Discord Server ☕: https://discord.com/invite/R9DSu34mGd
+        - How to Contribute 🏗️: community/contribute.md
+        - Community Projects 👏: community/examples.md
+        - Versioning Guide 📌: community/versioning.md
+
+    - Blog: https://blog.dottxt.co
diff --git a/outlines/applications.py b/outlines/applications.py
index 8d4ae149b..b43c20625 100644
--- a/outlines/applications.py
+++ b/outlines/applications.py
@@ -20,6 +20,7 @@ class Application:
 
     Examples
     --------
+    ```python
     from pydantic import BaseModel
     from transformers import AutoModelForCausalLM, AutoTokenizer
     from outlines import models, Application
@@ -41,6 +42,7 @@ class OutputModel(BaseModel):
 
     result = application(model, num=3)
     print(result)  # Expected output: { "result" : 6 }
+    ```
     """
     def __init__(self, template: Union[Template, Callable], output_type: Any):
         self.template = template

From 5b9faf139eecba1382cb33ee3aa8a042359d5c9d Mon Sep 17 00:00:00 2001
From: cpfiffer <cameron@pfiffer.org>
Date: Mon, 12 May 2025 15:14:33 -0700
Subject: [PATCH 325/505] Fix whitespace

---
 mkdocs.yml | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mkdocs.yml b/mkdocs.yml
index 20001418a..cefd6f923 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -120,10 +120,10 @@ plugins:
               "cookbook/deploy-using-bentoml.md": "examples/deploy-using-bentoml.md"
               "cookbook/deploy-using-cerebrium.md": "examples/deploy-using-cerebrium.md"
               "cookbook/deploy-using-modal.md": "examples/deploy-using-modal.md"
-              
+
               # Root files
               "quickstart.md": "getting_started/index.md"
-              
+
               # API to reference redirects
               "api/index.md": "reference/index.md"
               "api/json_schema.md": "reference/types/json_schema.md"
@@ -133,12 +133,12 @@ plugins:
               "api/guide.md": "reference/advanced/guide.md"
               "api/parsing.md": "reference/advanced/parsing.md"
               "api/models.md": "reference/available_models.md"
-              
+
               # API subdirectories to reference
               "api/types/": "reference/types/index.md"
               "api/core/": "reference/index.md"
               "api/advanced/": "reference/index.md"
-              
+
     - git-committers:
         repository: dottxt-ai/outlines
         branch: main

From 7b6ca4d58228c281dfe09b0595a2266856591d10 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 15 May 2025 09:26:57 +0200
Subject: [PATCH 326/505] Modify Application's call method to include inference
 kwargs

---
 outlines/applications.py   | 20 +++++++++++++-------
 tests/test_applications.py | 14 +++++++-------
 2 files changed, 20 insertions(+), 14 deletions(-)

diff --git a/outlines/applications.py b/outlines/applications.py
index b43c20625..63eb45fef 100644
--- a/outlines/applications.py
+++ b/outlines/applications.py
@@ -1,4 +1,4 @@
-from typing import Any, Callable, Optional, Union
+from typing import Any, Callable, Dict, Optional, Union
 
 from outlines.generator import BlackBoxGenerator, Generator, SteerableGenerator
 from outlines.templates import Template
@@ -9,7 +9,8 @@ class Application:
     """
     Application is a class that encapsulates a prompt template and an
     output type. It can be called to generate a response by providing a
-    model and optional arguments used to create the prompt from the template.
+    model, the values to be substituted in the template in a dictionary
+    and optional inference parameters.
 
     Parameters
     ----------
@@ -40,7 +41,7 @@ class OutputModel(BaseModel):
 
     application = Application(template, JsonType(OutputModel))
 
-    result = application(model, num=3)
+    result = application(model, {"num": 3}, max_new_tokens=20)
     print(result)  # Expected output: { "result" : 6 }
     ```
     """
@@ -50,16 +51,21 @@ def __init__(self, template: Union[Template, Callable], output_type: Any):
         self.model: Optional[Union[BlackBoxModel, SteerableModel]] = None
         self.generator: Optional[Union[BlackBoxGenerator, SteerableGenerator]] = None
 
-    def __call__(self, model: Union[BlackBoxModel, SteerableModel], *args, **kwargs):
+    def __call__(
+        self,
+        model: Union[BlackBoxModel, SteerableModel],
+        template_vars: Dict[str, Any],
+        **inference_kwargs
+    ):
         if model is None:
             raise ValueError("you must provide a model")
         # We save the generator to avoid creating a new one for each call.
         # If the model has changed since the last call, we create a new
         # generator.
-        elif model != self.model:
+        if model != self.model:
             self.model = model
             self.generator = Generator(model, self.output_type)
 
-        prompt = self.template(*args, **kwargs)
+        prompt = self.template(**template_vars)
         assert self.generator is not None
-        return self.generator(prompt)
+        return self.generator(prompt, **inference_kwargs)
diff --git a/tests/test_applications.py b/tests/test_applications.py
index fda179195..490dee438 100644
--- a/tests/test_applications.py
+++ b/tests/test_applications.py
@@ -53,14 +53,14 @@ def test_application_generator_no_model():
     application = Application(template, output_type)
 
     with pytest.raises(ValueError):
-        application(None)
+        application(None, {"value": "example"})
 
 
 def test_application_template_call(model):
     template = Template.from_string("Test {{ value }}")
     output_type = None
     application = Application(template, output_type)
-    result = application(model, value="example")
+    result = application(model, {"value": "example"}, foo="bar")
 
     assert result == "Test example"
 
@@ -71,7 +71,7 @@ def template(value):
 
     output_type = None
     application = Application(template, output_type)
-    result = application(model, value="example")
+    result = application(model, {"value": "example"}, foo="bar")
 
     assert result == "Test example"
 
@@ -82,7 +82,7 @@ def test_application_template_error(model):
     application = Application(template, output_type)
 
     with pytest.raises(jinja2.exceptions.UndefinedError):
-        application(model, foo="bar")
+        application(model, {"foo": "bar"})
 
 
 def test_application_generator_reuse(model, another_model):
@@ -90,15 +90,15 @@ def test_application_generator_reuse(model, another_model):
     output_type = None
     application = Application(template, output_type)
 
-    application(model, value="example")
+    application(model, {"value": "example"})
     first_generator = application.generator
     first_model = application.model
 
-    application(model, value="example")
+    application(model, {"value": "example"})
     assert application.model == first_model
     assert application.generator == first_generator
 
-    application(another_model, value="example")
+    application(another_model, {"value": "example"})
     assert application.model == another_model
     assert application.model != first_model
     assert application.generator != first_generator

From 1cda909736846917e12addb8a6de3acb92473039 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 16 May 2025 12:05:05 +0200
Subject: [PATCH 327/505] Update the Gemini model to use their new sdk

---
 outlines/models/gemini.py                | 110 ++++++++++++-----------
 pyproject.toml                           |   2 +-
 tests/models/test_gemini.py              |  31 ++++---
 tests/models/test_gemini_type_adapter.py |  10 ++-
 4 files changed, 85 insertions(+), 68 deletions(-)

diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index f3939af08..e2f197c75 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -1,50 +1,44 @@
 """Integration with Gemini's API."""
 
-from enum import EnumMeta, Enum
 from typing import (
-    get_args,
-    get_origin,
-    Optional,
-    Union,
     Any,
+    Optional,
     TYPE_CHECKING,
-    Literal,
+    Union,
+    get_args,
 )
 
-from pydantic import BaseModel, TypeAdapter
-from typing_extensions import is_typeddict
-
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
-from outlines.types import Regex, CFG, JsonSchema
+from outlines.types import CFG, JsonSchema, Regex
 from outlines.types.utils import (
     is_dataclass,
-    is_typed_dict,
-    is_pydantic_model,
-    is_genson_schema_builder,
     is_enum,
+    get_enum_from_literal,
+    is_genson_schema_builder,
     is_literal,
+    is_pydantic_model,
+    is_typed_dict,
     is_typing_list,
-    get_enum_from_literal
 )
 
 if TYPE_CHECKING:
-    from google.generativeai import GenerativeModel as GeminiClient
+    from google.genai import Client
 
 __all__ = ["Gemini", "from_gemini"]
 
 
 class GeminiTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the Gemini clients.
+    """Type adapter for the Gemini client.
 
     `GeminiTypeAdapter` is responsible for preparing the arguments to Gemini's
-    `generate_content` methods: the input (prompt and possibly image), as well
-    as the output type (only JSON).
+    client `models.generate_content` method: the input (prompt and possibly
+    image), as well as the output type (either JSON or multiple choice).
 
     """
 
     def format_input(self, model_input):
-        """Generate the `messages` argument to pass to the client.
+        """Generate the `contents` argument to pass to the client.
 
         Argument
         --------
@@ -55,51 +49,66 @@ def format_input(self, model_input):
         if isinstance(model_input, str):
             return {"contents": [model_input]}
         elif isinstance(model_input, Vision):
-            return {"contents": [model_input.prompt, model_input.image]}
+            from google.genai import types
+
+            image_part = types.Part.from_bytes(
+                data=model_input.image_str,
+                mime_type=model_input.image_format
+            ),
+            return {"contents": [model_input.prompt, image_part]}
         else:
             raise TypeError(
-                f"The input type {input} is not available with Gemini. The only available types are `str` and `Vision`."
+                f"The input type {input} is not available with Gemini. "
+                "The only available types are `str` and `Vision`."
             )
 
     def format_output_type(self, output_type):
-        # TODO: `int`, `float` and other Python types could be supported via JSON Schema.
-
-        # Unsupported languages
+        # Unsupported output pytes
         if isinstance(output_type, Regex):
             raise TypeError(
-                "Neither regex-based structured outputs nor the `pattern` keyword in Json Schema are available with Gemini. Use an open source model or dottxt instead."
+                "Neither regex-based structured outputs nor the `pattern` "
+                "keyword in Json Schema are available with Gemini. Use an "
+                "open source model or dottxt instead."
             )
         elif isinstance(output_type, CFG):
             raise TypeError(
-                "CFG-based structured outputs are not available with Gemini. Use an open source model or dottxt instead."
+                "CFG-based structured outputs are not available with Gemini. "
+                "Use an open source model or dottxt instead."
             )
         elif is_genson_schema_builder(output_type):
             raise TypeError(
-                "The Gemini SDK does not accept Genson schema builders as an input. Pass a Pydantic model, typed dict or dataclass instead."
+                "The Gemini SDK does not accept Genson schema builders as an "
+                "input. Pass a Pydantic model, typed dict or dataclass "
+                "instead."
             )
         elif isinstance(output_type, JsonSchema):
             raise TypeError(
-                "The Gemini SDK does not accept Json Schemas as an input. Pass a Pydantic model, typed dict or dataclass instead."
+                "The Gemini SDK does not accept Json Schemas as an input. "
+                "Pass a Pydantic model, typed dict or dataclass instead."
             )
 
         if output_type is None:
             return {}
-        # structured types
+
+        # Structured types
         elif is_dataclass(output_type):
             return self.format_json_output_type(output_type)
         elif is_typed_dict(output_type):
             return self.format_json_output_type(output_type)
         elif is_pydantic_model(output_type):
             return self.format_json_output_type(output_type)
-        # enum types
+
+        # List of structured types
+        elif is_typing_list(output_type):
+            return self.format_list_output_type(output_type)
+
+        # Multiple choice types
         elif is_enum(output_type):
             return self.format_enum_output_type(output_type)
         elif is_literal(output_type):
             enum = get_enum_from_literal(output_type)
             return self.format_enum_output_type(enum)
-        # list of objects
-        elif is_typing_list(output_type):
-            return self.format_list_output_type(output_type)
+
         else:
             type_name = getattr(output_type, "__name__", output_type)
             raise TypeError(
@@ -143,14 +152,16 @@ def format_list_output_type(self, output_type):
                 )
 
         raise TypeError(
-            f"Gemini only supports homogenous lists: list[BaseModel], list[TypedDict] or list[dataclass]. "
+            f"Gemini only supports homogenous lists: "
+            "list[BaseModel], list[TypedDict] or list[dataclass]. "
             f"Got {output_type} instead."
         )
 
 
 class Gemini(Model):
-    def __init__(self, client: "GeminiClient"):
+    def __init__(self, client: "Client", model_name: Optional[str] = None):
         self.client = client
+        self.model_name = model_name
         self.type_adapter = GeminiTypeAdapter()
 
     def generate(
@@ -159,15 +170,13 @@ def generate(
         output_type: Optional[Any] = None,
         **inference_kwargs,
     ):
-        import google.generativeai as genai
-
         contents = self.type_adapter.format_input(model_input)
-        generation_config = genai.GenerationConfig(
-            **self.type_adapter.format_output_type(output_type)
-        )
+        generation_config = self.type_adapter.format_output_type(output_type)
 
-        completion = self.client.generate_content(
-            generation_config=generation_config, **contents, **inference_kwargs
+        completion = self.client.models.generate_content(
+            **contents,
+            model=inference_kwargs.pop("model", self.model_name),
+            config={**generation_config, **inference_kwargs}
         )
 
         return completion.text
@@ -178,18 +187,13 @@ def generate_stream(
         output_type: Optional[Any] = None,
         **inference_kwargs,
     ):
-        import google.generativeai as genai
-
         contents = self.type_adapter.format_input(model_input)
-        generation_config = genai.GenerationConfig(
-            **self.type_adapter.format_output_type(output_type)
-        )
+        generation_config = self.type_adapter.format_output_type(output_type)
 
-        stream = self.client.generate_content(
-            generation_config=generation_config,
+        stream = self.client.models.generate_content_stream(
             **contents,
-            **inference_kwargs,
-            stream=True,
+            model=inference_kwargs.pop("model", self.model_name),
+            config={**generation_config, **inference_kwargs},
         )
 
         for chunk in stream:
@@ -197,5 +201,5 @@ def generate_stream(
                 yield chunk.text
 
 
-def from_gemini(client: "GeminiClient"):
-    return Gemini(client)
+def from_gemini(client: "Client", model_name: Optional[str] = None):
+    return Gemini(client, model_name)
diff --git a/pyproject.toml b/pyproject.toml
index 4c3d6b287..ce8534092 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -63,7 +63,7 @@ test = [
     "openai>=1.0.0",
     "datasets",
     "anthropic",
-    "google-generativeai",
+    "google-genai",
     "transformers",
     "pillow",
     "jax",
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 9dfe7a70e..329c02c83 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -5,11 +5,10 @@
 from enum import Enum
 from typing import Generator, Literal
 
-import google.generativeai as genai
 import pytest
-import requests
 from PIL import Image
-from pydantic import BaseModel
+from google.genai import Client
+from pydantic import BaseModel, ValidationError
 
 import outlines
 from outlines.models.gemini import Gemini
@@ -20,13 +19,12 @@
 else:
     from typing_extensions import TypedDict
 
-
 MODEL_NAME = "gemini-1.5-flash-latest"
 
 
 @pytest.fixture(scope="session")
 def model():
-    return Gemini(genai.GenerativeModel(MODEL_NAME))
+    return Gemini(Client(), MODEL_NAME)
 
 
 @pytest.fixture
@@ -44,16 +42,27 @@ def image():
     return image
 
 
-def test_gemini_wrong_inference_parameters(model):
-    with pytest.raises(TypeError, match="got an unexpected"):
-        model.generate("prompt", foo=10)
-
-
+@pytest.mark.api_call
 def test_gemini_init_from_client():
-    client = genai.GenerativeModel(MODEL_NAME)
+    client = Client()
+
+    # Without model name
     model = outlines.from_gemini(client)
     assert isinstance(model, Gemini)
     assert model.client == client
+    assert model.model_name is None
+
+    # With model name
+    model = outlines.from_gemini(client, MODEL_NAME)
+    assert isinstance(model, Gemini)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+
+@pytest.mark.api_call
+def test_gemini_wrong_inference_parameters(model):
+    with pytest.raises(ValidationError):
+        model.generate("prompt", foo=10)
 
 
 @pytest.mark.api_call
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
index a2d6b4a5d..46834eae9 100644
--- a/tests/models/test_gemini_type_adapter.py
+++ b/tests/models/test_gemini_type_adapter.py
@@ -1,13 +1,13 @@
 import io
-import json
 import pytest
 import sys
 from dataclasses import dataclass
 from enum import Enum, EnumMeta
-from typing import Literal, get_args
+from typing import Literal
 
 from PIL import Image
 from genson import SchemaBuilder
+from google.genai import types
 from pydantic import BaseModel
 
 from outlines import cfg, json_schema, regex
@@ -62,7 +62,11 @@ def test_gemini_type_adapter_input_text(adapter):
 def test_gemini_type_adapter_input_vision(adapter, image):
     input_message = Vision("hello", image)
     result = adapter.format_input(input_message)
-    assert result == {"contents": [input_message.prompt, input_message.image]}
+    image_part = types.Part.from_bytes(
+        data=input_message.image_str,
+        mime_type=input_message.image_format
+    ),
+    assert result == {"contents": [input_message.prompt, image_part]}
 
 
 def test_dottxt_type_adapter_input_invalid(adapter):

From 9c6962199410526cf79847045e840406fed4599f Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 16 May 2025 19:01:39 +0200
Subject: [PATCH 328/505] Standardize the use of the model_name argument

---
 outlines/models/__init__.py                  |   4 +-
 outlines/models/anthropic.py                 |  45 ++++++--
 outlines/models/dottxt.py                    |  47 ++++++---
 outlines/models/llamacpp.py                  |  44 +++++---
 outlines/models/ollama.py                    |  45 +++++---
 outlines/models/openai.py                    |  46 +++++---
 outlines/models/sglang.py                    |  32 +++---
 outlines/models/tgi.py                       |   1 -
 outlines/models/vllm.py                      |  17 ++-
 tests/models/test_anthropic.py               |  56 +++++++---
 tests/models/test_dottxt.py                  | 104 ++++++++++++-------
 tests/models/test_gemini.py                  |  12 ++-
 tests/models/test_ollama.py                  |  67 ++++++++----
 tests/models/test_openai.py                  |  69 ++++++------
 tests/models/test_sglang.py                  |  65 +++++++++---
 tests/models/test_sglang_type_adapter.py     |   2 +-
 tests/models/test_vllm.py                    |  55 ++++++++--
 tests/v0_legacy/models/test_openai_legacy.py |   7 --
 18 files changed, 485 insertions(+), 233 deletions(-)

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 503586de9..3dffe64d4 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -16,7 +16,7 @@
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
-from .sglang import from_sglang, SgLang, AsyncSgLang
+from .sglang import from_sglang, SGLang, AsyncSgLang
 from .tgi import from_tgi, TGI, AsyncTGI
 from .transformers import (
     Transformers,
@@ -42,6 +42,6 @@
     Ollama,
     OpenAI,
     TGI,
-    SgLang,
+    SGLang,
     VLLM,
 ]
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index a225b1a87..6a3a01a81 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -1,6 +1,6 @@
 """Integration with Anthropic's API."""
 
-from typing import Union, TYPE_CHECKING
+from typing import Optional, Union, TYPE_CHECKING
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
@@ -36,7 +36,8 @@ def format_input(self, model_input):
         elif isinstance(model_input, Vision):
             return self.format_vision_model_input(model_input)
         raise TypeError(
-            f"The input type {input} is not available with Anthropic. The only available types are `str` and `Vision`."
+            f"The input type {input} is not available with Anthropic. "
+            "The only available types are `str` and `Vision`."
         )
 
     def format_str_model_input(self, model_input):
@@ -75,18 +76,24 @@ def format_output_type(self, output_type):
             return {}
         else:
             raise NotImplementedError(
-                f"The output type {output_type} is not available with Anthropic."
+                f"The output type {output_type} is not available with "
+                "Anthropic."
             )
 
 
 class Anthropic(Model):
-    def __init__(self, client: "AnthropicClient", model_name: str):
+    def __init__(
+        self, client: "AnthropicClient", model_name: Optional[str] = None
+    ):
         self.client = client
         self.model_name = model_name
         self.type_adapter = AnthropicTypeAdapter()
 
     def generate(
-        self, model_input: Union[str, Vision], output_type=None, **inference_kwargs
+        self,
+        model_input: Union[str, Vision],
+        output_type=None,
+        **inference_kwargs,
     ):
         messages = self.type_adapter.format_input(model_input)
 
@@ -95,15 +102,23 @@ def generate(
                 f"The type {output_type} is not available with Anthropic."
             )
 
+        if (
+            "model" not in inference_kwargs
+            and self.model_name is not None
+        ):
+            inference_kwargs["model"] = self.model_name
+
         completion = self.client.messages.create(
             **messages,
-            model=self.model_name,
             **inference_kwargs,
         )
         return completion.content[0].text
 
     def generate_stream(
-        self, model_input: Union[str, Vision], output_type=None, **inference_kwargs
+        self,
+        model_input: Union[str, Vision],
+        output_type=None,
+        **inference_kwargs,
     ):
         messages = self.type_adapter.format_input(model_input)
 
@@ -112,17 +127,27 @@ def generate_stream(
                 f"The type {output_type} is not available with Anthropic."
             )
 
+        if (
+            "model" not in inference_kwargs
+            and self.model_name is not None
+        ):
+            inference_kwargs["model"] = self.model_name
+
         stream = self.client.messages.create(
             **messages,
-            model=self.model_name,
             stream=True,
             **inference_kwargs,
         )
 
         for chunk in stream:
-            if chunk.type == "content_block_delta" and chunk.delta.type == "text_delta":
+            if (
+                chunk.type == "content_block_delta"
+                and chunk.delta.type == "text_delta"
+            ):
                 yield chunk.delta.text
 
 
-def from_anthropic(client: "AnthropicClient", model_name: str) -> Anthropic:
+def from_anthropic(
+    client: "AnthropicClient", model_name: Optional[str] = None
+) -> Anthropic:
     return Anthropic(client, model_name)
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index 61fc9b19a..eef41b83b 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -1,16 +1,18 @@
 """Integration with Dottxt's API."""
 
 import json
-from functools import singledispatchmethod
 from typing import Optional, TYPE_CHECKING
 
-from pydantic import BaseModel, TypeAdapter
-from typing_extensions import _TypedDictMeta  # type: ignore
+from pydantic import TypeAdapter
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import Regex, CFG, JsonSchema
-from outlines.types.utils import is_dataclass, is_typed_dict, is_pydantic_model, is_genson_schema_builder
-
+from outlines.types import CFG, JsonSchema, Regex
+from outlines.types.utils import (
+    is_dataclass,
+    is_genson_schema_builder,
+    is_pydantic_model,
+    is_typed_dict,
+)
 
 if TYPE_CHECKING:
     from dottxt import Dottxt as DottxtClient
@@ -31,27 +33,32 @@ def format_input(self, model_input):
         if isinstance(model_input, str):
             return model_input
         raise TypeError(
-            f"The input type {model_input} is not available with Dottxt. The only available type is `str`."
+            f"The input type {model_input} is not available with Dottxt. "
+            "The only available type is `str`."
         )
 
     def format_output_type(self, output_type):
         """Format the output type to pass to the client.
 
-        TODO: `int`, `float` and other Python types could be supported via JSON Schema.
+        TODO: `int`, `float` and other Python types could be supported via
+        JSON Schema.
         """
 
         # Unsupported languages
         if output_type is None:
             raise TypeError(
-                "You must provide an output type. Dottxt only supports constrained generation."
+                "You must provide an output type. Dottxt only supports "
+                "constrained generation."
             )
         elif isinstance(output_type, Regex):
             raise TypeError(
-                "Regex-based structured outputs will soon be available with Dottxt. Use an open source model in the meantime."
+                "Regex-based structured outputs will soon be available with "
+                "Dottxt. Use an open source model in the meantime."
             )
         elif isinstance(output_type, CFG):
             raise TypeError(
-                "CFG-based structured outputs will soon be available with Dottxt. Use an open source model in the meantime."
+                "CFG-based structured outputs will soon be available with "
+                "Dottxt. Use an open source model in the meantime."
             )
 
         elif isinstance(output_type, JsonSchema):
@@ -87,7 +94,7 @@ def __init__(
         self,
         client: "Dottxt",
         model_name: Optional[str] = None,
-        model_revision: str = "",
+        model_revision: Optional[str] = None,
     ):
         self.client = client
         self.model_name = model_name
@@ -98,8 +105,16 @@ def generate(self, model_input, output_type=None, **inference_kwargs):
         prompt = self.type_adapter.format_input(model_input)
         json_schema = self.type_adapter.format_output_type(output_type)
 
-        if self.model_name:
+        if (
+            "model_name" not in inference_kwargs
+            and self.model_name is not None
+        ):
             inference_kwargs["model_name"] = self.model_name
+
+        if (
+            "model_revision" not in inference_kwargs
+            and self.model_revision is not None
+        ):
             inference_kwargs["model_revision"] = self.model_revision
 
         completion = self.client.json(
@@ -109,7 +124,9 @@ def generate(self, model_input, output_type=None, **inference_kwargs):
         )
         return completion.data
 
-    def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+    def generate_stream(
+        self, model_input, output_type=None, **inference_kwargs
+    ):
         raise NotImplementedError(
             "Dottxt does not support streaming. Call the model/generator for "
             + "regular generation instead."
@@ -119,6 +136,6 @@ def generate_stream(self, model_input, output_type=None, **inference_kwargs):
 def from_dottxt(
     client: "DottxtClient",
     model_name: Optional[str] = None,
-    model_revision: str = "",
+    model_revision: Optional[str] = None,
 ):
     return Dottxt(client, model_name, model_revision)
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 93f8f5eb5..93aefc2e8 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,6 +1,14 @@
 import warnings
 from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Dict, Iterator, List, Optional, Set, Tuple, Union
+from typing import (
+    TYPE_CHECKING,
+    Dict,
+    Iterator,
+    List,
+    Set,
+    Tuple,
+    Union,
+)
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
@@ -21,7 +29,8 @@ def __init__(self, model: "Llama"):
 
         self.tokenizer = model.tokenizer()
 
-        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
+        # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613
+        # is resolved
         self._hf_tokenizer = None
         try:
             self.vocabulary = model.tokenizer_.hf_tokenizer.get_vocab()
@@ -35,7 +44,8 @@ def __init__(self, model: "Llama"):
         # ensure stable ordering of vocabulary
         self.vocabulary = {
             tok: tok_id
-            for tok, tok_id in sorted(self.vocabulary.items(), key=lambda x: x[1])
+            for tok, tok_id
+            in sorted(self.vocabulary.items(), key=lambda x: x[1])
         }
 
         self._hash = None
@@ -45,14 +55,19 @@ def decode(self, token_ids: List[int]) -> List[str]:
         return [decoded_bytes.decode("utf-8", errors="ignore")]
 
     def encode(
-        self, prompt: Union[str, List[str]], add_bos: bool = True, special: bool = True
+        self,
+        prompt: Union[str, List[str]],
+        add_bos: bool = True,
+        special: bool = True,
     ) -> Tuple[List[int], List[int]]:
         if isinstance(prompt, list):
             raise NotImplementedError(
                 "llama-cpp-python tokenizer doesn't support batch tokenization"
             )
         token_ids = self.tokenizer.tokenize(
-            prompt.encode("utf-8", errors="ignore"), add_bos=add_bos, special=special
+            prompt.encode("utf-8", errors="ignore"),
+            add_bos=add_bos,
+            special=special,
         )
         # generate attention mask, missing from llama-cpp-python
         attention_mask = [
@@ -65,7 +80,10 @@ def convert_token_to_string(self, token: str) -> str:
             from transformers.file_utils import SPIECE_UNDERLINE
 
             token_str = self._hf_tokenizer.convert_tokens_to_string([token])
-            if token.startswith(SPIECE_UNDERLINE) or token == "<0x20>": # pragma: no cover
+            if (
+                token.startswith(SPIECE_UNDERLINE)
+                or token == "<0x20>"
+            ):  # pragma: no cover
                 token_str = " " + token_str
             return token_str
         else:
@@ -163,8 +181,8 @@ def generate(self, model_input, output_type, **inference_kwargs):
         output_type
             The logits processor to use when generating text.
         inference_kwargs
-            The inference kwargs that can be passed to the `Llama.__call__` method
-            in the `llama-cpp-python` library.
+            The inference kwargs that can be passed to the `Llama.__call__`
+            method in the `llama-cpp-python` library.
 
         Returns
         -------
@@ -173,7 +191,8 @@ def generate(self, model_input, output_type, **inference_kwargs):
         """
         if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
+                "CFG generation is not supported for LlamaCpp due to bug in "
+                "the llama_cpp tokenizer"
             )
 
         completion = self.model(
@@ -199,8 +218,8 @@ def generate_stream(
         output_type
             The logits processor to use when generating text.
         inference_kwargs
-            The inference kwargs that can be passed to the `Llama.__call__` method
-            in the `llama-cpp-python` library.
+            The inference kwargs that can be passed to the `Llama.__call__`
+            method in the `llama-cpp-python` library.
 
         Returns
         -------
@@ -209,7 +228,8 @@ def generate_stream(
         """
         if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp due to bug in the llama_cpp tokenizer"
+                "CFG generation is not supported for LlamaCpp due to bug in "
+                "the llama_cpp tokenizer"
             )
 
         generator = self.model(
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 04526f3aa..c2495351b 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,12 +1,16 @@
 import json
-from typing import Iterator, TYPE_CHECKING
+from typing import Iterator, Optional, TYPE_CHECKING
 
-from pydantic import BaseModel, TypeAdapter
-from typing_extensions import _TypedDictMeta  # type: ignore
+from pydantic import TypeAdapter
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.types import Regex, CFG, JsonSchema
-from outlines.types.utils import is_dataclass, is_typed_dict, is_pydantic_model, is_genson_schema_builder
+from outlines.types import CFG, JsonSchema, Regex
+from outlines.types.utils import (
+    is_dataclass,
+    is_genson_schema_builder,
+    is_pydantic_model,
+    is_typed_dict,
+)
 
 
 if TYPE_CHECKING:
@@ -38,16 +42,19 @@ def format_input(self, model_input):
     def format_output_type(self, output_type):
         """Format the output type to pass to the client.
 
-        TODO: `int`, `float` and other Python types could be supported via JSON Schema.
+        TODO: `int`, `float` and other Python types could be supported via
+        JSON Schema.
         """
 
         if isinstance(output_type, Regex):
             raise TypeError(
-                "Regex-based structured outputs are not supported by Ollama. Use an open source model in the meantime."
+                "Regex-based structured outputs are not supported by Ollama. "
+                "Use an open source model in the meantime."
             )
         elif isinstance(output_type, CFG):
             raise TypeError(
-                "CFG-based structured outputs are not supported by Ollama. Use an open source model in the meantime."
+                "CFG-based structured outputs are not supported by Ollama. "
+                "Use an open source model in the meantime."
             )
 
         if output_type is None:
@@ -81,23 +88,33 @@ class Ollama(Model):
 
     """
 
-    def __init__(self, client: "OllamaClient", model_name: str):
+    def __init__(
+        self,client: "OllamaClient", model_name: Optional[str] = None,
+    ):
         self.client = client
         self.model_name = model_name
         self.type_adapter = OllamaTypeAdapter()
 
     def generate(self, model_input, output_type=None, **kwargs) -> str:
+
+        if "model" not in kwargs and self.model_name is not None:
+            kwargs["model"] = self.model_name
+
         response = self.client.generate(
-            model=self.model_name,
             prompt=self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             **kwargs,
         )
         return response.response
 
-    def generate_stream(self, model_input, output_type=None, **kwargs) -> Iterator[str]:
+    def generate_stream(
+        self, model_input, output_type=None, **kwargs
+    ) -> Iterator[str]:
+
+        if "model" not in kwargs and self.model_name is not None:
+            kwargs["model"] = self.model_name
+
         response = self.client.generate(
-            model=self.model_name,
             prompt=self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             stream=True,
@@ -107,5 +124,7 @@ def generate_stream(self, model_input, output_type=None, **kwargs) -> Iterator[s
             yield chunk.response
 
 
-def from_ollama(client: "OllamaClient", model_name: str) -> Ollama:
+def from_ollama(
+    client: "OllamaClient", model_name: Optional[str] = None
+) -> Ollama:
     return Ollama(client, model_name)
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index c4ff7335d..540b04bac 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -52,7 +52,8 @@ def format_input(self, model_input):
         elif isinstance(model_input, Vision):
             return self.format_vision_model_input(model_input)
         raise TypeError(
-            f"The input type {input} is not available with OpenAI. The only available types are `str` and `Vision`."
+            f"The input type {input} is not available with OpenAI. "
+            "The only available types are `str` and `Vision`."
         )
 
     def format_str_model_input(self, model_input: str):
@@ -95,18 +96,22 @@ def format_output_type(self, output_type):
         """Generate the `response_format` argument to the client based on the
         output type specified by the user.
 
-        TODO: `int`, `float` and other Python types could be supported via JSON Schema.
+        TODO: `int`, `float` and other Python types could be supported via
+        JSON Schema.
 
         """
 
         # Unsupported languages
         if isinstance(output_type, Regex):
             raise TypeError(
-                "Neither regex-based structured outputs nor the `pattern` keyword in Json Schema are available with OpenAI. Use an open source model or dottxt instead."
+                "Neither regex-based structured outputs nor the `pattern` keyword "
+                "in Json Schema are available with OpenAI. Use an open source "
+                "model or dottxt instead."
             )
         elif isinstance(output_type, CFG):
             raise TypeError(
-                "CFG-based structured outputs are not available with OpenAI. Use an open source model or dottxt instead."
+                "CFG-based structured outputs are not available with OpenAI. "
+                "Use an open source model or dottxt instead."
             )
 
         if output_type is None:
@@ -130,7 +135,8 @@ def format_output_type(self, output_type):
         else:
             type_name = getattr(output_type, "__name__", output_type)
             raise TypeError(
-                f"The type `{type_name}` is not available with OpenAI. Use an open source model or dottxt instead."
+                f"The type `{type_name}` is not available with OpenAI. "
+                "Use an open source model or dottxt instead."
             )
 
     def format_json_output_type(self, schema: dict):
@@ -174,7 +180,7 @@ class OpenAI(Model):
     def __init__(
         self,
         client: Union["OpenAIClient", "AzureOpenAIClient"],
-        model_name: Union[str, "OpenAIConfig"],
+        model_name: Optional[Union[str, "OpenAIConfig"]] = None,
         **kwargs
     ):
         # legacy mode
@@ -199,8 +205,14 @@ def __init__(
             DeprecationWarning,
             stacklevel=2,
             )
-            config = model_name if isinstance(model_name, OpenAIConfig) else kwargs.pop("config")
-            self.legacy_instance = OpenAILegacy(client, config, kwargs.get("system_prompt"))
+            config = (
+                model_name
+                if isinstance(model_name, OpenAIConfig)
+                else kwargs.pop("config")
+            )
+            self.legacy_instance = OpenAILegacy(
+                client, config, kwargs.get("system_prompt")
+            )
         # regular mode
         else:
             self.client = client
@@ -218,14 +230,20 @@ def generate(
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
 
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
         try:
             result = self.client.chat.completions.create(
-                model=self.model_name, **messages, **response_format, **inference_kwargs
+                **messages,
+                **response_format,
+                **inference_kwargs,
             )
         except openai.BadRequestError as e:
             if e.body["message"].startswith("Invalid schema"):
                 raise TypeError(
-                    f"OpenAI does not support your schema: {e.body['message']}. Try a local model or dottxt instead."
+                    f"OpenAI does not support your schema: {e.body['message']}. "
+                    "Try a local model or dottxt instead."
                 )
             else:
                 raise e
@@ -250,8 +268,11 @@ def generate_stream(
     ):
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
         stream = self.client.chat.completions.create(
-            model=self.model_name,
             stream=True,
             **messages,
             **response_format,
@@ -294,7 +315,8 @@ def __repr__(self):
             return super().__repr__()
 
 def from_openai(
-    client: Union["OpenAIClient", "AzureOpenAIClient"], model_name: str
+    client: Union["OpenAIClient", "AzureOpenAIClient"],
+    model_name: Optional[str] = None,
 ) -> OpenAI:
     return OpenAI(client, model_name)
 
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
index 8679cda66..0d3fdaab6 100644
--- a/outlines/models/sglang.py
+++ b/outlines/models/sglang.py
@@ -42,7 +42,7 @@ def format_output_type(self, output_type):
         term = python_types_to_terms(output_type)
         if isinstance(term, CFG):
             warnings.warn(
-                "SgLang grammar-based structured outputs expects an EBNF "
+                "SGLang grammar-based structured outputs expects an EBNF "
                 "grammar instead of a Lark grammar as is generally used in "
                 "Outlines. The grammar cannot be used as a structured output "
                 "type with an outlines backend, it is only compatible with "
@@ -57,11 +57,11 @@ def format_output_type(self, output_type):
             return {"extra_body": {"regex": to_regex(term)}}
 
 
-class SgLang(Model):
+class SGLang(Model):
     """Represents a client to a `sglang` server."""
 
     def __init__(self, client, model_name: Optional[str] = None):
-        """Create a `SgLang` model instance.
+        """Create a `SGLang` model instance.
 
         Parameters
         ----------
@@ -102,7 +102,7 @@ def generate(self, model_input, output_type, **inference_kwargs):
         for message in messages:
             if message.refusal is not None:  # pragma: no cover
                 raise ValueError(
-                    f"The SgLang server refused to answer the request: "
+                    f"The SGLang server refused to answer the request: "
                     f"{message.refusal}"
                 )
 
@@ -140,18 +140,16 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
                 yield chunk.choices[0].delta.content
 
     def _build_client_args(self, model_input, output_type, **inference_kwargs):
-        """Build the arguments to pass to the SgLang client."""
+        """Build the arguments to pass to the SGLang client."""
         messages = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
         inference_kwargs.update(output_type_args)
-        model_name = (
-            inference_kwargs.pop("model", self.model_name)
-            or self.model_name
-        )
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
 
         client_args = {
             **messages,
-            "model": model_name,
             **inference_kwargs,
         }
 
@@ -242,18 +240,16 @@ async def generate_stream(
                 yield chunk.choices[0].delta.content
 
     def _build_client_args(self, model_input, output_type, **inference_kwargs):
-        """Build the arguments to pass to the SgLang client."""
+        """Build the arguments to pass to the SGLang client."""
         messages = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
         inference_kwargs.update(output_type_args)
-        model_name = (
-            inference_kwargs.pop("model", self.model_name)
-            or self.model_name
-        )
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
 
         client_args = {
             **messages,
-            "model": model_name,
             **inference_kwargs,
         }
 
@@ -263,11 +259,11 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
 def from_sglang(
     client: Union["OpenAI", "AsyncOpenAI"],
     model_name: Optional[str] = None,
-) -> Union[SgLang, AsyncSgLang]:
+) -> Union[SGLang, AsyncSgLang]:
     from openai import AsyncOpenAI, OpenAI
 
     if isinstance(client, OpenAI):
-        return SgLang(client, model_name)
+        return SGLang(client, model_name)
     elif isinstance(client, AsyncOpenAI):
         return AsyncSgLang(client, model_name)
     else:
diff --git a/outlines/models/tgi.py b/outlines/models/tgi.py
index 3bd2e6e33..3e135890f 100644
--- a/outlines/models/tgi.py
+++ b/outlines/models/tgi.py
@@ -141,7 +141,6 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
         output_type_args = self.type_adapter.format_output_type(output_type)
         inference_kwargs.update(output_type_args)
 
-
         client_args = {
             "prompt": prompt,
             **inference_kwargs,
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index e199fa76c..c3a191169 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -138,14 +138,12 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
         output_type_args = self.type_adapter.format_output_type(output_type)
         extra_body = inference_kwargs.pop("extra_body", {})
         extra_body.update(output_type_args)
-        model_name = (
-            inference_kwargs.pop("model", self.model_name)
-            or self.model_name
-        )
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
 
         client_args = {
             **messages,
-            "model": model_name,
             **inference_kwargs,
         }
         if extra_body:
@@ -244,14 +242,12 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
         output_type_args = self.type_adapter.format_output_type(output_type)
         extra_body = inference_kwargs.pop("extra_body", {})
         extra_body.update(output_type_args)
-        model_name = (
-            inference_kwargs.pop("model", self.model_name)
-            or self.model_name
-        )
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
 
         client_args = {
             **messages,
-            "model": model_name,
             **inference_kwargs,
         }
         if extra_body:
@@ -260,7 +256,6 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
         return client_args
 
 
-
 def from_vllm(
     client: Union["OpenAI", "AsyncOpenAI"],
     model_name: Optional[str] = None,
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 56ed242c6..66b3c74db 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -2,9 +2,8 @@
 from typing import Generator
 
 from anthropic import Anthropic as AnthropicClient
-import PIL
+from PIL import Image
 import pytest
-import requests
 
 import outlines
 from outlines.models.anthropic import Anthropic
@@ -18,6 +17,37 @@ def model():
     return Anthropic(AnthropicClient(), MODEL_NAME)
 
 
+@pytest.fixture(scope="session")
+def model_no_model_name():
+    return Anthropic(AnthropicClient())
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+    image.format = "PNG"
+
+    return image
+
+
+def test_init_from_client():
+    client = AnthropicClient()
+
+    # With model name
+    model = outlines.from_anthropic(client, MODEL_NAME)
+    assert isinstance(model, Anthropic)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+    # Without model name
+    model = outlines.from_anthropic(client)
+    assert isinstance(model, Anthropic)
+    assert model.client == client
+    assert model.model_name is None
+
+
 def test_anthropic_wrong_inference_parameters():
     with pytest.raises(TypeError, match="got an unexpected"):
         model = Anthropic(AnthropicClient(), MODEL_NAME)
@@ -34,13 +64,6 @@ def __init__(self, foo):
         model.generate(Foo("prompt"))
 
 
-def test_init_from_client():
-    client = AnthropicClient()
-    model = outlines.from_anthropic(client, MODEL_NAME)
-    assert isinstance(model, Anthropic)
-    assert model.client == client
-
-
 def test_anthropic_wrong_output_type():
     class Foo:
         def __init__(self, foo):
@@ -59,18 +82,17 @@ def test_anthropic_simple_call(model):
 
 @pytest.mark.xfail(reason="Anthropic requires the `max_tokens` parameter to be set")
 @pytest.mark.api_call
-def test_anthropic_direct_call(model):
-    result = model("Respond with one word. Not more.", max_tokens=1024)
+def test_anthropic_direct_call(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        model_name=MODEL_NAME,
+        max_tokens=1024,
+    )
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
-def test_anthropic_simple_vision(model):
-    url = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/assets/images/logo.png"
-    r = requests.get(url, stream=True)
-    if r.status_code == 200:
-        image = PIL.Image.open(io.BytesIO(r.content))
-
+def test_anthropic_simple_vision(model, image):
     result = model.generate(
         Vision("What does this logo represent?", image), max_tokens=1024
     )
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index ae32bfcac..0bd1277af 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -10,15 +10,20 @@
 from outlines.models.dottxt import Dottxt
 
 
+MODEL_NAME = "dottxt/dottxt-v1-alpha"
+MODEL_REVISION = "d06c86726aadd8dadb92c5b9b9e3ce8ef246c471"
+
+
 class User(BaseModel):
     first_name: str
     last_name: str
     user_id: int
 
 
-@pytest.fixture
+@pytest.fixture(scope="session")
 def api_key():
-    """Get the Dottxt API key from the environment, providing a default value if not found.
+    """Get the Dottxt API key from the environment, providing a default value
+    if not found.
 
     This fixture should be used for tests that do not make actual api calls,
     but still require to initialize the Dottxt client.
@@ -30,70 +35,97 @@ def api_key():
     return api_key
 
 
-def test_dottxt_wrong_init_parameters(api_key):
-    with pytest.raises(TypeError, match="got an unexpected"):
-        client = DottxtClient(api_key=api_key)
-        Dottxt(client, foo=10)
+@pytest.fixture(scope="session")
+def model_name_and_revision(api_key):
+    client = DottxtClient(api_key=api_key)
+    model_list = client.list_models()
+    return (model_list[0].name, model_list[0].revision)
 
 
-def test_dottxt_wrong_output_type(api_key):
-    with pytest.raises(TypeError, match="You must provide an output type"):
-        client = DottxtClient(api_key=api_key)
-        model = Dottxt(client)
-        model("prompt")
+@pytest.fixture(scope="session")
+def model(api_key, model_name_and_revision):
+    client = DottxtClient(api_key=api_key)
+    return Dottxt(
+        client,
+        model_name_and_revision[0],
+        model_name_and_revision[1],
+    )
 
 
-def test_dottxt_init_from_client(api_key):
+@pytest.fixture(scope="session")
+def model_no_model_name(api_key):
     client = DottxtClient(api_key=api_key)
+    return Dottxt(client)
+
+
+@pytest.mark.api_call
+def test_dottxt_init_from_client(api_key, model_name_and_revision):
+    client = DottxtClient(api_key=api_key)
+
+    # Without model name
     model = outlines.from_dottxt(client)
     assert isinstance(model, Dottxt)
     assert model.client == client
+    assert model.model_name is None
+
+    # With model name
+    model = outlines.from_dottxt(
+        client,
+        model_name_and_revision[0],
+        model_name_and_revision[1],
+    )
+    assert isinstance(model, Dottxt)
+    assert model.client == client
+    assert model.model_name == model_name_and_revision[0]
+    assert model.model_revision == model_name_and_revision[1]
 
 
-@pytest.mark.api_call
-def test_dottxt_wrong_input_type(api_key):
+def test_dottxt_wrong_output_type(model_no_model_name):
+    with pytest.raises(TypeError, match="You must provide an output type"):
+        model_no_model_name("prompt")
+
+
+def test_dottxt_wrong_input_type(model_no_model_name):
     with pytest.raises(TypeError, match="is not available"):
-        client = DottxtClient(api_key=api_key)
-        model = Dottxt(client)
-        model(["prompt"], User)
+        model_no_model_name(["prompt"], User)
 
 
 @pytest.mark.api_call
-def test_dottxt_wrong_inference_parameters(api_key):
+def test_dottxt_wrong_inference_parameters(model_no_model_name):
     with pytest.raises(TypeError, match="got an unexpected"):
-        client = DottxtClient(api_key=api_key)
-        model = Dottxt(client)
-        model("prompt", User, foo=10)
+        model_no_model_name("prompt", User, foo=10)
 
 
 @pytest.mark.api_call
-def test_dottxt_direct_pydantic_call(api_key):
-    client = DottxtClient(api_key=api_key)
-    model = Dottxt(client)
-    result = model("Create a user", User)
+def test_dottxt_direct_pydantic_call(model_no_model_name):
+    result = model_no_model_name("Create a user", User)
     assert "first_name" in json.loads(result)
 
 
 @pytest.mark.api_call
-def test_dottxt_direct_jsonschema_call(api_key):
-    client = DottxtClient(api_key=api_key)
-    model = Dottxt(client)
-    result = model("Create a user", User)
+def test_dottxt_direct_jsonschema_call(
+    model_no_model_name, model_name_and_revision
+):
+    result = model_no_model_name(
+        "Create a user",
+        User,
+        model_name=model_name_and_revision[0],
+        model_revision=model_name_and_revision[1],
+    )
     assert "first_name" in json.loads(result)
 
 
 @pytest.mark.api_call
-def test_dottxt_generator_pydantic_call(api_key):
-    client = DottxtClient(api_key=api_key)
-    model = Dottxt(client)
+def test_dottxt_generator_pydantic_call(model):
     generator = Generator(model, User)
     result = generator("Create a user")
     assert "first_name" in json.loads(result)
 
 
 @pytest.mark.api_call
-def test_dottxt_streaming(api_key):
-    client = DottxtClient(api_key=api_key)
-    model = Dottxt(client)
-    with pytest.raises(NotImplementedError, match="Dottxt does not support streaming"):
+def test_dottxt_streaming(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="Dottxt does not support streaming"
+    ):
         model.stream("Create a user", User)
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 329c02c83..5df7b2669 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -27,6 +27,11 @@ def model():
     return Gemini(Client(), MODEL_NAME)
 
 
+@pytest.fixture(scope="session")
+def model_no_model_name():
+    return Gemini(Client())
+
+
 @pytest.fixture
 def image():
     width, height = 1, 1
@@ -72,8 +77,11 @@ def test_gemini_simple_call(model):
 
 
 @pytest.mark.api_call
-def test_gemini_direct_call(model):
-    result = model("Respond with one word. Not more.")
+def test_gemini_direct_call(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        model=MODEL_NAME
+    )
     assert isinstance(result, str)
 
 
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index 9f4050e9e..408fa2ceb 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -11,69 +11,90 @@
 
 
 MODEL_NAME = "tinyllama"
-CLIENT = OllamaClient()
 
 
-def test_wrong_inference_parameters():
-    with pytest.raises(TypeError, match="got an unexpected"):
-        Ollama(CLIENT, MODEL_NAME).generate(
-            "Respond with one word. Not more.", None, foo=10
-        )
+@pytest.fixture
+def model():
+    return Ollama(OllamaClient(), MODEL_NAME)
+
+
+@pytest.fixture
+def model_no_model_name():
+    return Ollama(OllamaClient())
 
 
 def test_init_from_client():
-    model = outlines.from_ollama(CLIENT, MODEL_NAME)
+    client = OllamaClient()
+
+    # With model name
+    model = outlines.from_ollama(client, MODEL_NAME)
+    assert isinstance(model, Ollama)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+    # Without model name
+    model = outlines.from_ollama(client)
     assert isinstance(model, Ollama)
-    assert model.client == CLIENT
+    assert model.client == client
+    assert model.model_name is None
 
 
-def test_ollama_simple():
-    result = Ollama(CLIENT, MODEL_NAME).generate(
+def test_wrong_inference_parameters(model):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        model.generate(
+            "Respond with one word. Not more.", None, foo=10
+        )
+
+
+def test_ollama_simple(model):
+    result = model.generate(
         "Respond with one word. Not more.", None
     )
     assert isinstance(result, str)
 
 
-def test_ollama_direct():
-    result = Ollama(CLIENT, MODEL_NAME)("Respond with one word. Not more.", None)
+def test_ollama_direct(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        None,
+        model=MODEL_NAME,
+    )
     assert isinstance(result, str)
 
 
-def test_ollama_json():
+def test_ollama_json(model):
     class Foo(BaseModel):
         foo: Annotated[str, Field(max_length=1)]
 
-    result = Ollama(CLIENT, MODEL_NAME)("Respond with one word. Not more.", Foo)
+    result = model("Respond with one word. Not more.", Foo)
     assert isinstance(result, str)
     assert "foo" in json.loads(result)
 
 
-def test_ollama_wrong_output_type():
+def test_ollama_wrong_output_type(model):
     class Foo(Enum):
         bar = "Bar"
         foor = "Foo"
 
     with pytest.raises(TypeError, match="is not supported"):
-        Ollama(CLIENT, MODEL_NAME).generate("foo?", Foo)
+        model.generate("foo?", Foo)
 
 
-def test_ollama_wrong_input_type():
+def test_ollama_wrong_input_type(model):
     with pytest.raises(TypeError, match="is not available"):
-        Ollama(CLIENT, MODEL_NAME).generate(["foo?", "bar?"], None)
+        model.generate(["foo?", "bar?"], None)
 
 
-def test_ollama_stream():
-    model = Ollama(CLIENT, MODEL_NAME)
+def test_ollama_stream(model):
     generator = model.stream("Write a sentence about a cat.")
     assert isinstance(next(generator), str)
 
 
-def test_ollama_stream_json():
+def test_ollama_stream_json(model_no_model_name):
     class Foo(BaseModel):
         foo: Annotated[str, Field(max_length=2)]
 
-    model = Ollama(CLIENT, MODEL_NAME)
-    generator = model.stream("Create a character.", Foo)
+    generator = model_no_model_name.stream("Create a character.", Foo, model=MODEL_NAME)
     generated_text = []
     for text in generator:
         generated_text.append(text)
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 992276054..455087707 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -17,7 +17,7 @@
 MODEL_NAME = "gpt-4o-mini-2024-07-18"
 
 
-@pytest.fixture
+@pytest.fixture(scope="session")
 def api_key():
     """Get the OpenAI API key from the environment, providing a default value if not found.
 
@@ -31,7 +31,7 @@ def api_key():
     return api_key
 
 
-@pytest.fixture
+@pytest.fixture(scope="session")
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
@@ -46,49 +46,63 @@ def image():
     return image
 
 
+@pytest.fixture(scope="session")
+def model(api_key):
+    return OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
+
+
+@pytest.fixture(scope="session")
+def model_no_model_name(api_key):
+    return OpenAI(OpenAIClient(api_key=api_key))
+
+
 def test_init_from_client(api_key):
     client = OpenAIClient(api_key=api_key)
+
+    # With model name
     model = outlines.from_openai(client, "gpt-4o")
     assert isinstance(model, OpenAI)
     assert model.client == client
+    assert model.model_name == "gpt-4o"
 
+    # Without model name
+    model = outlines.from_openai(client)
+    assert isinstance(model, OpenAI)
+    assert model.client == client
+    assert model.model_name is None
 
-def test_openai_wrong_inference_parameters(api_key):
+
+def test_openai_wrong_inference_parameters(model):
     with pytest.raises(TypeError, match="got an unexpected"):
-        model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate("prompt", foo=10)
 
 
-def test_openai_wrong_input_type(api_key):
+def test_openai_wrong_input_type(model):
     class Foo:
         def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(TypeError, match="is not available"):
-        model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate(Foo("prompt"))
 
 
-def test_openai_wrong_output_type(api_key):
+def test_openai_wrong_output_type(model):
     class Foo:
         def __init__(self, foo):
             self.foo = foo
 
     with pytest.raises(TypeError, match="is not available"):
-        model = OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
         model.generate("prompt", Foo(1))
 
 
 @pytest.mark.api_call
-def test_openai_simple_call():
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
+def test_openai_simple_call(model):
     result = model.generate("Respond with one word. Not more.")
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
-def test_openai_simple_call_multiple_samples():
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
+def test_openai_simple_call_multiple_samples(model):
     result = model.generate("Respond with one word. Not more.", n=2)
     assert isinstance(result, list)
     assert len(result) == 2
@@ -97,23 +111,22 @@ def test_openai_simple_call_multiple_samples():
 
 
 @pytest.mark.api_call
-def test_openai_direct_call():
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
-    result = model("Respond with one word. Not more.")
+def test_openai_direct_call(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        model=MODEL_NAME,
+    )
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
-def test_openai_simple_vision(image):
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
+def test_openai_simple_vision(image, model):
     result = model.generate(Vision("What does this logo represent?", image))
     assert isinstance(result, str)
 
 
 @pytest.mark.api_call
-def test_openai_simple_pydantic():
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
-
+def test_openai_simple_pydantic(model):
     class Foo(BaseModel):
         bar: int
 
@@ -123,9 +136,7 @@ class Foo(BaseModel):
 
 
 @pytest.mark.api_call
-def test_openai_simple_pydantic_refusal():
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
-
+def test_openai_simple_pydantic_refusal(model):
     class Foo(BaseModel):
         bar: Annotated[str, Field(int, pattern=r"^\d+$")]
 
@@ -134,9 +145,7 @@ class Foo(BaseModel):
 
 
 @pytest.mark.api_call
-def test_openai_simple_vision_pydantic(image):
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
-
+def test_openai_simple_vision_pydantic(image, model):
     class Logo(BaseModel):
         name: int
 
@@ -146,9 +155,7 @@ class Logo(BaseModel):
 
 
 @pytest.mark.api_call
-def test_openai_simple_json_schema():
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
-
+def test_openai_simple_json_schema(model):
     class Foo(BaseModel):
         bar: int
 
@@ -160,9 +167,7 @@ class Foo(BaseModel):
 
 
 @pytest.mark.api_call
-def test_openai_streaming():
-    model = OpenAI(OpenAIClient(), MODEL_NAME)
-
+def test_openai_streaming(model):
     result = model.stream("Respond with one word. Not more.")
     assert isinstance(result, Generator)
     assert isinstance(next(result), str)
diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
index 2a594cc5a..e04fb3d10 100644
--- a/tests/models/test_sglang.py
+++ b/tests/models/test_sglang.py
@@ -1,4 +1,4 @@
-# ATTENTION: When running this test with an actual SgLang server, use the
+# ATTENTION: When running this test with an actual SGLang server, use the
 # llguidance backend (--grammar-backend llguidance)
 # The outlines backend does not support the EBNF grammar. The xgrammar
 # backend is slow and buggy.
@@ -11,7 +11,7 @@
 import pytest
 from openai import AsyncOpenAI, OpenAI
 
-from outlines.models.sglang import SgLang, AsyncSgLang, from_sglang
+from outlines.models.sglang import SGLang, AsyncSgLang, from_sglang
 from outlines.types.dsl import CFG, Regex, JsonSchema
 from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
 
@@ -22,7 +22,7 @@
 """
 
 
-# If the SGLANG_SERVER_URL environment variable is set, use the real SgLang server
+# If the SGLANG_SERVER_URL environment variable is set, use the real SGLang server
 # Otherwise, use the mock server
 sglang_server_url = os.environ.get("SGLANG_SERVER_URL")
 sglang_model_name = os.environ.get(
@@ -32,7 +32,7 @@
     openai_client = OpenAI(base_url=sglang_server_url)
     async_openai_client = AsyncOpenAI(base_url=sglang_server_url)
 else:
-    warnings.warn("No SgLang server URL provided, using mock server")
+    warnings.warn("No SGLang server URL provided, using mock server")
     openai_client = MockOpenAIClient()
     async_openai_client = MockAsyncOpenAIClient()
 
@@ -120,7 +120,12 @@
 
 @pytest.fixture
 def sync_model():
-    return SgLang(openai_client, model_name=sglang_model_name)
+    return SGLang(openai_client, model_name=sglang_model_name)
+
+
+@pytest.fixture
+def sync_model_no_model_name():
+    return SGLang(openai_client)
 
 
 @pytest.fixture
@@ -128,12 +133,40 @@ def async_model():
     return AsyncSgLang(async_openai_client, model_name=sglang_model_name)
 
 
+@pytest.fixture
+def async_model_no_model_name():
+    return AsyncSgLang(async_openai_client)
+
+
 def test_sglang_init():
-    model = from_sglang(OpenAI(base_url="http://localhost:11434"))
-    assert isinstance(model, SgLang)
+    # We do not rely on the mock server here because we need an object
+    # of type OpenAI and AsyncOpenAI to test the init function.
+    openai_client = OpenAI(base_url="http://localhost:11434")
+    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
+
+    # Sync with model name
+    model = from_sglang(openai_client, sglang_model_name)
+    assert isinstance(model, SGLang)
+    assert model.client == openai_client
+    assert model.model_name == sglang_model_name
+
+    # Sync without model name
+    model = from_sglang(openai_client)
+    assert isinstance(model, SGLang)
+    assert model.client == openai_client
+    assert model.model_name is None
+
+    # Async with model name
+    model = from_sglang(async_openai_client, sglang_model_name)
+    assert isinstance(model, AsyncSgLang)
+    assert model.client == async_openai_client
+    assert model.model_name == sglang_model_name
 
-    model = from_sglang(AsyncOpenAI(base_url="http://localhost:11434"))
+    # Async without model name
+    model = from_sglang(async_openai_client)
     assert isinstance(model, AsyncSgLang)
+    assert model.client == async_openai_client
+    assert model.model_name is None
 
     with pytest.raises(ValueError, match="Unsupported client type"):
         from_sglang("foo")
@@ -144,8 +177,11 @@ def test_sglang_sync_simple_call(sync_model):
     assert isinstance(result, str)
 
 
-def test_sglang_sync_streaming(sync_model):
-    result = sync_model.stream("Respond with a single word.")
+def test_sglang_sync_streaming(sync_model_no_model_name):
+    result = sync_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=sglang_model_name,
+    )
     assert isinstance(result, Generator)
     assert isinstance(next(result), str)
 
@@ -177,7 +213,7 @@ def test_sglang_sync_regex(sync_model):
 def test_sglang_sync_cfg(sync_model):
     with pytest.warns(
         UserWarning,
-        match="SgLang grammar-based structured outputs expects an EBNF"
+        match="SGLang grammar-based structured outputs expects an EBNF"
     ):
         result = sync_model("foo?", CFG(EBNF_YES_NO_GRAMMAR), max_tokens=10)
         assert isinstance(result, str)
@@ -191,8 +227,11 @@ async def test_sglang_async_simple_call(async_model):
 
 
 @pytest.mark.asyncio
-async def test_sglang_async_streaming(async_model):
-    result = async_model.stream("Respond with a single word.")
+async def test_sglang_async_streaming(async_model_no_model_name):
+    result = async_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=sglang_model_name,
+    )
     assert isinstance(result, AsyncGenerator)
     async for chunk in result:
         assert isinstance(chunk, str)
diff --git a/tests/models/test_sglang_type_adapter.py b/tests/models/test_sglang_type_adapter.py
index b9630ba15..c7ac2b6c8 100644
--- a/tests/models/test_sglang_type_adapter.py
+++ b/tests/models/test_sglang_type_adapter.py
@@ -108,7 +108,7 @@ def test_sglang_type_adapter_output_type(
     assert type_adapter.format_output_type(None) == {}
     with pytest.warns(
         UserWarning,
-        match="SgLang grammar-based structured outputs expects an EBNF"
+        match="SGLang grammar-based structured outputs expects an EBNF"
     ):
         assert type_adapter.format_output_type(cfg_instance) == {
             "extra_body": {"ebnf": CFG_STRING}
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index 7e1f72696..76a9da814 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -112,20 +112,53 @@
 
 @pytest.fixture
 def sync_model():
-    return VLLM(openai_client, model_name=vllm_model_name)
+    return VLLM(openai_client, vllm_model_name)
+
+
+@pytest.fixture
+def sync_model_no_model_name():
+    return VLLM(openai_client)
 
 
 @pytest.fixture
 def async_model():
-    return AsyncVLLM(async_openai_client, model_name=vllm_model_name)
+    return AsyncVLLM(async_openai_client, vllm_model_name)
+
+
+@pytest.fixture
+def async_model_no_model_name():
+    return AsyncVLLM(async_openai_client)
 
 
 def test_vllm_init():
-    model = from_vllm(OpenAI(base_url="http://localhost:11434"))
+    # We do not rely on the mock server here because we need an object
+    # of type OpenAI and AsyncOpenAI to test the init function.
+    openai_client = OpenAI(base_url="http://localhost:11434")
+    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
+
+    # Sync with model name
+    model = from_vllm(openai_client, vllm_model_name)
+    assert isinstance(model, VLLM)
+    assert model.client == openai_client
+    assert model.model_name == vllm_model_name
+
+    # Sync without model name
+    model = from_vllm(openai_client)
     assert isinstance(model, VLLM)
+    assert model.client == openai_client
+    assert model.model_name is None
+
+    # Async with model name
+    model = from_vllm(async_openai_client, vllm_model_name)
+    assert isinstance(model, AsyncVLLM)
+    assert model.client == async_openai_client
+    assert model.model_name == vllm_model_name
 
-    model = from_vllm(AsyncOpenAI(base_url="http://localhost:11434"))
+    # Async without model name
+    model = from_vllm(async_openai_client)
     assert isinstance(model, AsyncVLLM)
+    assert model.client == async_openai_client
+    assert model.model_name is None
 
     with pytest.raises(ValueError, match="Unsupported client type"):
         from_vllm("foo")
@@ -136,8 +169,11 @@ def test_vllm_sync_simple_call(sync_model):
     assert isinstance(result, str)
 
 
-def test_vllm_sync_streaming(sync_model):
-    result = sync_model.stream("Respond with a single word.")
+def test_vllm_sync_streaming(sync_model_no_model_name):
+    result = sync_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=vllm_model_name,
+    )
     assert isinstance(result, Generator)
     assert isinstance(next(result), str)
 
@@ -176,8 +212,11 @@ async def test_vllm_async_simple_call(async_model):
 
 
 @pytest.mark.asyncio
-async def test_vllm_async_streaming(async_model):
-    result = async_model.stream("Respond with a single word.")
+async def test_vllm_async_streaming(async_model_no_model_name):
+    result = async_model_no_model_name.stream(
+        "Respond with a single word.",
+        model=vllm_model_name,
+    )
     assert isinstance(result, AsyncGenerator)
     async for chunk in result:
         assert isinstance(chunk, str)
diff --git a/tests/v0_legacy/models/test_openai_legacy.py b/tests/v0_legacy/models/test_openai_legacy.py
index 13e3c750f..c07443dab 100644
--- a/tests/v0_legacy/models/test_openai_legacy.py
+++ b/tests/v0_legacy/models/test_openai_legacy.py
@@ -111,13 +111,6 @@ def test_openai_legacy_init():
     assert not hasattr(model, "model_name")
     assert not hasattr(model, "type_adapter")
 
-    # directly through the OpenAI class without config
-    with pytest.raises(TypeError):
-        models.OpenAI(
-            AsyncOpenAI(),
-            system_prompt="You are a helpful assistant.",
-        )
-
 
 def test_azure_openai_legacy_init():
     with pytest.warns(

From 0be7d8b38608d502f1af5a2f320e8f58c2a44386 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 22 May 2025 10:25:24 +0200
Subject: [PATCH 329/505] Add script to autogenerate API reference with mkdocs

---
 docs_new/api_reference/index.md               |  1 +
 docs_new/blog/index.md                        |  1 +
 .../{reference => features}/advanced/guide.md |  0
 .../advanced/parsing.md                       |  0
 .../available_models.md                       |  0
 .../{reference => features}/core/generator.md |  0
 docs_new/features/index.md                    |  3 +
 .../models/anthropic.md                       |  0
 .../{reference => features}/models/dottxt.md  |  0
 .../{reference => features}/models/gemini.md  |  0
 .../{reference => features}/models/index.md   |  0
 .../models/llamacpp.md                        |  0
 .../{reference => features}/models/mlxlm.md   |  0
 .../{reference => features}/models/models.md  |  0
 .../{reference => features}/models/ollama.md  |  0
 .../{reference => features}/models/openai.md  |  0
 .../{reference => features}/models/sglang.md  |  0
 .../{reference => features}/models/tgi.md     |  0
 .../models/transformers.md                    |  0
 .../{reference => features}/models/vllm.md    |  0
 .../models/vllm_offline.md                    |  0
 docs_new/{reference => features}/types/cfg.md |  0
 .../{reference => features}/types/index.md    |  0
 .../types/json_schema.md                      |  0
 .../{reference => features}/types/regex.md    |  0
 .../utility/application.md                    |  0
 .../utility/template.md                       |  0
 docs_new/index.md                             |  4 +-
 docs_new/reference/index.md                   |  3 -
 mkdocs.yml                                    | 86 +++++++++++--------
 pyproject.toml                                |  2 +
 scripts/gen_ref_pages.py                      | 53 ++++++++++++
 32 files changed, 110 insertions(+), 43 deletions(-)
 create mode 100644 docs_new/api_reference/index.md
 create mode 100644 docs_new/blog/index.md
 rename docs_new/{reference => features}/advanced/guide.md (100%)
 rename docs_new/{reference => features}/advanced/parsing.md (100%)
 rename docs_new/{reference => features}/available_models.md (100%)
 rename docs_new/{reference => features}/core/generator.md (100%)
 create mode 100644 docs_new/features/index.md
 rename docs_new/{reference => features}/models/anthropic.md (100%)
 rename docs_new/{reference => features}/models/dottxt.md (100%)
 rename docs_new/{reference => features}/models/gemini.md (100%)
 rename docs_new/{reference => features}/models/index.md (100%)
 rename docs_new/{reference => features}/models/llamacpp.md (100%)
 rename docs_new/{reference => features}/models/mlxlm.md (100%)
 rename docs_new/{reference => features}/models/models.md (100%)
 rename docs_new/{reference => features}/models/ollama.md (100%)
 rename docs_new/{reference => features}/models/openai.md (100%)
 rename docs_new/{reference => features}/models/sglang.md (100%)
 rename docs_new/{reference => features}/models/tgi.md (100%)
 rename docs_new/{reference => features}/models/transformers.md (100%)
 rename docs_new/{reference => features}/models/vllm.md (100%)
 rename docs_new/{reference => features}/models/vllm_offline.md (100%)
 rename docs_new/{reference => features}/types/cfg.md (100%)
 rename docs_new/{reference => features}/types/index.md (100%)
 rename docs_new/{reference => features}/types/json_schema.md (100%)
 rename docs_new/{reference => features}/types/regex.md (100%)
 rename docs_new/{reference => features}/utility/application.md (100%)
 rename docs_new/{reference => features}/utility/template.md (100%)
 delete mode 100644 docs_new/reference/index.md
 create mode 100644 scripts/gen_ref_pages.py

diff --git a/docs_new/api_reference/index.md b/docs_new/api_reference/index.md
new file mode 100644
index 000000000..b0d5c88fc
--- /dev/null
+++ b/docs_new/api_reference/index.md
@@ -0,0 +1 @@
+# API Reference
diff --git a/docs_new/blog/index.md b/docs_new/blog/index.md
new file mode 100644
index 000000000..05761ac57
--- /dev/null
+++ b/docs_new/blog/index.md
@@ -0,0 +1 @@
+# Blog
diff --git a/docs_new/reference/advanced/guide.md b/docs_new/features/advanced/guide.md
similarity index 100%
rename from docs_new/reference/advanced/guide.md
rename to docs_new/features/advanced/guide.md
diff --git a/docs_new/reference/advanced/parsing.md b/docs_new/features/advanced/parsing.md
similarity index 100%
rename from docs_new/reference/advanced/parsing.md
rename to docs_new/features/advanced/parsing.md
diff --git a/docs_new/reference/available_models.md b/docs_new/features/available_models.md
similarity index 100%
rename from docs_new/reference/available_models.md
rename to docs_new/features/available_models.md
diff --git a/docs_new/reference/core/generator.md b/docs_new/features/core/generator.md
similarity index 100%
rename from docs_new/reference/core/generator.md
rename to docs_new/features/core/generator.md
diff --git a/docs_new/features/index.md b/docs_new/features/index.md
new file mode 100644
index 000000000..bc09a026e
--- /dev/null
+++ b/docs_new/features/index.md
@@ -0,0 +1,3 @@
+# Features
+
+This needs to contain a list of all features available in Outlines.
diff --git a/docs_new/reference/models/anthropic.md b/docs_new/features/models/anthropic.md
similarity index 100%
rename from docs_new/reference/models/anthropic.md
rename to docs_new/features/models/anthropic.md
diff --git a/docs_new/reference/models/dottxt.md b/docs_new/features/models/dottxt.md
similarity index 100%
rename from docs_new/reference/models/dottxt.md
rename to docs_new/features/models/dottxt.md
diff --git a/docs_new/reference/models/gemini.md b/docs_new/features/models/gemini.md
similarity index 100%
rename from docs_new/reference/models/gemini.md
rename to docs_new/features/models/gemini.md
diff --git a/docs_new/reference/models/index.md b/docs_new/features/models/index.md
similarity index 100%
rename from docs_new/reference/models/index.md
rename to docs_new/features/models/index.md
diff --git a/docs_new/reference/models/llamacpp.md b/docs_new/features/models/llamacpp.md
similarity index 100%
rename from docs_new/reference/models/llamacpp.md
rename to docs_new/features/models/llamacpp.md
diff --git a/docs_new/reference/models/mlxlm.md b/docs_new/features/models/mlxlm.md
similarity index 100%
rename from docs_new/reference/models/mlxlm.md
rename to docs_new/features/models/mlxlm.md
diff --git a/docs_new/reference/models/models.md b/docs_new/features/models/models.md
similarity index 100%
rename from docs_new/reference/models/models.md
rename to docs_new/features/models/models.md
diff --git a/docs_new/reference/models/ollama.md b/docs_new/features/models/ollama.md
similarity index 100%
rename from docs_new/reference/models/ollama.md
rename to docs_new/features/models/ollama.md
diff --git a/docs_new/reference/models/openai.md b/docs_new/features/models/openai.md
similarity index 100%
rename from docs_new/reference/models/openai.md
rename to docs_new/features/models/openai.md
diff --git a/docs_new/reference/models/sglang.md b/docs_new/features/models/sglang.md
similarity index 100%
rename from docs_new/reference/models/sglang.md
rename to docs_new/features/models/sglang.md
diff --git a/docs_new/reference/models/tgi.md b/docs_new/features/models/tgi.md
similarity index 100%
rename from docs_new/reference/models/tgi.md
rename to docs_new/features/models/tgi.md
diff --git a/docs_new/reference/models/transformers.md b/docs_new/features/models/transformers.md
similarity index 100%
rename from docs_new/reference/models/transformers.md
rename to docs_new/features/models/transformers.md
diff --git a/docs_new/reference/models/vllm.md b/docs_new/features/models/vllm.md
similarity index 100%
rename from docs_new/reference/models/vllm.md
rename to docs_new/features/models/vllm.md
diff --git a/docs_new/reference/models/vllm_offline.md b/docs_new/features/models/vllm_offline.md
similarity index 100%
rename from docs_new/reference/models/vllm_offline.md
rename to docs_new/features/models/vllm_offline.md
diff --git a/docs_new/reference/types/cfg.md b/docs_new/features/types/cfg.md
similarity index 100%
rename from docs_new/reference/types/cfg.md
rename to docs_new/features/types/cfg.md
diff --git a/docs_new/reference/types/index.md b/docs_new/features/types/index.md
similarity index 100%
rename from docs_new/reference/types/index.md
rename to docs_new/features/types/index.md
diff --git a/docs_new/reference/types/json_schema.md b/docs_new/features/types/json_schema.md
similarity index 100%
rename from docs_new/reference/types/json_schema.md
rename to docs_new/features/types/json_schema.md
diff --git a/docs_new/reference/types/regex.md b/docs_new/features/types/regex.md
similarity index 100%
rename from docs_new/reference/types/regex.md
rename to docs_new/features/types/regex.md
diff --git a/docs_new/reference/utility/application.md b/docs_new/features/utility/application.md
similarity index 100%
rename from docs_new/reference/utility/application.md
rename to docs_new/features/utility/application.md
diff --git a/docs_new/reference/utility/template.md b/docs_new/features/utility/template.md
similarity index 100%
rename from docs_new/reference/utility/template.md
rename to docs_new/features/utility/template.md
diff --git a/docs_new/index.md b/docs_new/index.md
index 7ce2641f8..ab0a36515 100644
--- a/docs_new/index.md
+++ b/docs_new/index.md
@@ -10,8 +10,8 @@ title: Welcome to Outlines!
 Outlines is a Python library from [.TXT](https://dottxt.co) that guarantees structured output from
 large language models. It ensures LLMs speak the language of your application by making them follow specific formats such as JSON, regular expressions, or context-free grammars.
 
-[Get started](/docs/getting_started){ .md-button .md-button--primary }
-[API Reference](/api/reference){ .md-button }
+[Get started](/getting_started){ .md-button .md-button--primary }
+[Features](/features){ .md-button }
 [Examples](/examples){ .md-button }
 [GitHub](https://github.com/yourusername/outlines){ .md-button }
 
diff --git a/docs_new/reference/index.md b/docs_new/reference/index.md
deleted file mode 100644
index 320d43af4..000000000
--- a/docs_new/reference/index.md
+++ /dev/null
@@ -1,3 +0,0 @@
-# API Reference
-
-This needs to contain a list of all the misc API reference pages.
diff --git a/mkdocs.yml b/mkdocs.yml
index cefd6f923..e09fa3cbc 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -124,20 +124,20 @@ plugins:
               # Root files
               "quickstart.md": "getting_started/index.md"
 
-              # API to reference redirects
-              "api/index.md": "reference/index.md"
-              "api/json_schema.md": "reference/types/json_schema.md"
-              "api/regex.md": "reference/types/regex.md"
-              "api/applications.md": "reference/utility/application.md"
-              "api/templates.md": "reference/utility/template.md"
-              "api/guide.md": "reference/advanced/guide.md"
-              "api/parsing.md": "reference/advanced/parsing.md"
-              "api/models.md": "reference/available_models.md"
-
-              # API subdirectories to reference
-              "api/types/": "reference/types/index.md"
-              "api/core/": "reference/index.md"
-              "api/advanced/": "reference/index.md"
+              # API to features redirects
+              "api/index.md": "features/index.md"
+              "api/json_schema.md": "features/types/json_schema.md"
+              "api/regex.md": "features/types/regex.md"
+              "api/applications.md": "features/utility/application.md"
+              "api/templates.md": "features/utility/template.md"
+              "api/guide.md": "features/advanced/guide.md"
+              "api/parsing.md": "features/advanced/parsing.md"
+              "api/models.md": "features/available_models.md"
+
+              # API subdirectories to features
+              "api/types/": "features/types/index.md"
+              "api/core/": "features/index.md"
+              "api/advanced/": "features/index.md"
 
     - git-committers:
         repository: dottxt-ai/outlines
@@ -146,11 +146,18 @@ plugins:
         enable_creation_date: true
         type: timeago
 
+    - gen-files:
+        scripts:
+        - scripts/gen_ref_pages.py
+    - literate-nav:
+        nav_file: SUMMARY.md
+
 nav:
     - Home:
           - index.md
           - User Guide: user_guide/index.md
-          - API Reference: reference/index.md
+          - Features: features/index.md
+          - API Reference: api_reference/index.md
           - Examples: examples/index.md
     - Guide:
           - Introduction:
@@ -173,38 +180,41 @@ nav:
                       - Templates: user_guide/prompting/templates.md
                 - Easy Regular Expressions: user_guide/regex_dsl.md
                 - Migrating to v1: user_guide/migration.md
-    - API Reference:
-          - Overview: reference/index.md
+    - Features:
+          - Overview: features/index.md
           - Format types:
-                - Overview: reference/types/index.md
-                - JSON: reference/types/json_schema.md
-                - Regex: reference/types/regex.md
-                - Context-free grammars: reference/types/cfg.md
+                - Overview: features/types/index.md
+                - JSON: features/types/json_schema.md
+                - Regex: features/types/regex.md
+                - Context-free grammars: features/types/cfg.md
 
           - Core:
-                - Generator: reference/core/generator.md
+                - Generator: features/core/generator.md
 
           - Supported Models:
-                - Anthropic: reference/models/anthropic.md
-                - Dottxt: reference/models/dottxt.md
-                - Gemini: reference/models/gemini.md
-                - Llamacpp: reference/models/llamacpp.md
-                - Mlx-lm: reference/models/mlxlm.md
-                - Ollama: reference/models/ollama.md
-                - OpenAI: reference/models/openai.md
-                - SgLang: reference/models/sglang.md
-                - TGI: reference/models/tgi.md
-                - Transformers: reference/models/transformers.md
-                - vLLM (online server): reference/models/vllm.md
-                - vLLM (offline): reference/models/vllm_offline.md
+                - Anthropic: features/models/anthropic.md
+                - Dottxt: features/models/dottxt.md
+                - Gemini: features/models/gemini.md
+                - Llamacpp: features/models/llamacpp.md
+                - Mlx-lm: features/models/mlxlm.md
+                - Ollama: features/models/ollama.md
+                - OpenAI: features/models/openai.md
+                - SGLang: features/models/sglang.md
+                - TGI: features/models/tgi.md
+                - Transformers: features/models/transformers.md
+                - vLLM (online server): features/models/vllm.md
+                - vLLM (offline): features/models/vllm_offline.md
 
           - Utilities:
-                - Applications: reference/utility/application.md
-                - Template: reference/utility/template.md
+                - Applications: features/utility/application.md
+                - Template: features/utility/template.md
 
           - Advanced:
-                - Guide: reference/advanced/guide.md
-                - Parsing: reference/advanced/parsing.md
+                - Guide: features/advanced/guide.md
+                - Parsing: features/advanced/parsing.md
+
+    - API Reference: api_reference/
+
     - Examples:
           - examples/index.md
           - Classification: examples/classification.md
diff --git a/pyproject.toml b/pyproject.toml
index ce8534092..ec23577ac 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -75,6 +75,7 @@ test = [
     "ollama",
     "dottxt",
     "sentencepiece",
+    "mkdocs_gen_files",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
 
@@ -154,6 +155,7 @@ module = [
     "tensorflow.*",
     "tf-keras",
     "tf-keras.*",
+    "mkdocs_gen_files.*",
 ]
 ignore_missing_imports = true
 
diff --git a/scripts/gen_ref_pages.py b/scripts/gen_ref_pages.py
new file mode 100644
index 000000000..48c226442
--- /dev/null
+++ b/scripts/gen_ref_pages.py
@@ -0,0 +1,53 @@
+"""Generate the API reference pages and navigation automatically.
+
+This script is based on the `gen_ref_pages.py` script in the
+[mkdocstrings](https://mkdocstrings.github.io/recipes/#automatic-code-reference-pages) project.
+
+To exclude a file or module from being included in the generated API reference,
+add a part of its path to the `EXCLUDED_FILES` list.
+"""
+
+from pathlib import Path
+
+import mkdocs_gen_files
+
+
+CODEBASE_DIR_NAME = "outlines"
+OUTPUT_DIR_NAME = "api_reference"
+EXCLUDED_FILES = ["_version", "v0_legacy"]
+
+
+nav = mkdocs_gen_files.Nav()
+root = Path(__file__).parent.parent
+src = root / CODEBASE_DIR_NAME
+
+for path in sorted(src.rglob("*.py")):
+    module_path = path.relative_to(src).with_suffix("")
+    doc_path = path.relative_to(src).with_suffix(".md")
+    full_doc_path = Path(OUTPUT_DIR_NAME, doc_path)
+
+    parts = tuple(module_path.parts)
+
+    if any(part in EXCLUDED_FILES for part in parts):
+        continue
+
+    if parts[-1] == "__init__":
+        if len(parts) == 1:
+            doc_path = Path("index.md")
+            full_doc_path = Path(OUTPUT_DIR_NAME, doc_path)
+            parts = (CODEBASE_DIR_NAME,)
+        else:
+            parts = parts[:-1]
+            doc_path = doc_path.with_name("index.md")
+            full_doc_path = full_doc_path.with_name("index.md")
+
+    nav[parts] = doc_path.as_posix()
+
+    with mkdocs_gen_files.open(full_doc_path, "w") as fd:
+        ident = ".".join(parts)
+        fd.write(f"::: {CODEBASE_DIR_NAME}.{ident}")
+
+    mkdocs_gen_files.set_edit_path(full_doc_path, path.relative_to(root))
+
+with mkdocs_gen_files.open(f"{OUTPUT_DIR_NAME}/SUMMARY.md", "w") as nav_file:
+    nav_file.writelines(nav.build_literate_nav())

From fee8a2321b189259b454931040005724dbf7d203 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 26 May 2025 10:35:09 +0200
Subject: [PATCH 330/505] Update the utility features section of the
 documentation

---
 docs_new/features/utility/application.md |  85 +++---
 docs_new/features/utility/regex_dsl.md   | 348 +++++++++++++++++++++++
 docs_new/features/utility/template.md    | 123 +++-----
 mkdocs.yml                               |   3 +-
 4 files changed, 433 insertions(+), 126 deletions(-)
 create mode 100644 docs_new/features/utility/regex_dsl.md

diff --git a/docs_new/features/utility/application.md b/docs_new/features/utility/application.md
index 47f455ff2..4faccb10d 100644
--- a/docs_new/features/utility/application.md
+++ b/docs_new/features/utility/application.md
@@ -1,5 +1,5 @@
 ---
-title: Application API
+title: Application
 ---
 
 # Application
@@ -10,67 +10,56 @@ The `Application` class enables you to encapsulate a prompt template and an outp
 
 An `Application` combines a prompt template with an output type, creating a reusable component that can be applied to different models.
 
-## Basic Usage
+Applications are useful for simplifying repeated tasks where you have a well-defined `Template` and a fixed output type, such as classification tasks or data extraction.
+
+To create an `Application` instance, initialize the class with a prompt template and an output type. You can then call the application with a model and the variables defined in your template in a dictionary.
+
+For instance:
 
 ```python
+from typing import Literal
+import transformers
 from outlines import Application, Template, from_transformers
-from typing import List
 
 # Create a template
-template = Template("""
-You are an expert classifier.
-Classify the following text into one of these categories: {categories}
-
-Text: {text}
-""")
+template_str = "Is {{ name }} a boy or a girl name?"""
+template = Template.from_string(template_str)
 
-# Create an application with the template and output type
-classifier = Application(template, str)
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
-# Initialize a model
-model = from_transformers(...)
+# Create the application and call it to generate text
+application = Application(template, Literal["boy", "girl"])
+response = application(model, {"name": "Alice"}, max_new_tokens=10)
 
-# Use the application with specific parameters
-categories = ["Sports", "Politics", "Technology", "Entertainment"]
-result = classifier(model, categories=categories, text="Apple unveils new iPhone with AI capabilities")
-
-print(result)  # Expected output: "Technology"
+print(response) # "girl"
 ```
 
-## Structured Output
+Instead of providing an Outlines `Template` instance, you can provide a `Callable` that returns a string. The parameters of the callable are used as the variables of the template such that you must provide values for them in the dictionary when calling the application.
 
-```python
-from pydantic import BaseModel
-from outlines import Application, Template, from_transformers
+For instance, we can create the same example as above using a a function instead of a template:
 
-# Define a Pydantic model for structured output
-class MovieReview(BaseModel):
-    title: str
-    director: str
-    year: int
-    rating: float
-    review: str
+```python
+from typing import Literal
+import transformers
+from outlines import Application, from_transformers
 
-# Create a template
-template = Template("""
-Write a brief review of the movie {movie_title}.
-Include the director, year of release, and a rating out of 10.
-""")
+# Create a function that will be used as a template
+def template(name: str) -> str:
+    return f"Is {name} a boy or a girl name?"
 
-# Create an application with the template and output type
-movie_reviewer = Application(template, MovieReview)
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
-# Use the application with a model
-result = movie_reviewer(model, movie_title="The Matrix")
+# Create the application with the function template and call it to generate text
+application = Application(template, Literal["boy", "girl"])
+response = application(model, {"name": "Alice"}, max_new_tokens=10)
 
-# Parse the result into a Pydantic model
-review = MovieReview.model_validate_json(result)
-print(f"{review.title} ({review.year}) - {review.rating}/10")
+print(response) # "girl"
 ```
-
-## Parameters
-
-- `prompt_template`: A template that defines the prompt structure
-- `output_type`: The type of output to generate
-
-::: outlines.applications.Application
diff --git a/docs_new/features/utility/regex_dsl.md b/docs_new/features/utility/regex_dsl.md
new file mode 100644
index 000000000..740a0772b
--- /dev/null
+++ b/docs_new/features/utility/regex_dsl.md
@@ -0,0 +1,348 @@
+---
+title: Regex DSL
+---
+
+# Regex DSL
+
+This library provides a Domain-Specific Language (DSL) to construct regular expressions in a more intuitive and modular way. It allows you to create complex regexes using simple building blocks that represent literal strings, patterns, and various quantifiers. Additionally, these custom regex types can be used directly as types in [Pydantic](https://pydantic-docs.helpmanual.io/) schemas to enforce pattern constraints during text generation.
+
+---
+
+## Why Use This DSL?
+
+1. **Modularity & Readability**: Instead of writing cryptic regular expression strings, you compose a regex as a tree of objects.
+2. **Enhanced Debugging**: Each expression can be visualized as an ASCII tree, making it easier to understand and debug complex regexes.
+3. **Pydantic Integration**: Use your DSL-defined regex as types in Pydantic models. The DSL seamlessly converts to JSON Schema with proper pattern constraints.
+4. **Extensibility**: Easily add or modify quantifiers and other regex components by extending the provided classes.
+
+---
+
+## Building Blocks
+
+
+Every regex component in this DSL is a **Term**. Here are two primary types:
+
+- **`String`**: Represents a literal string. It escapes the characters that have a special meaning in regular expressions.
+- **`Regex`**: Represents an existing regex pattern string.
+
+```python
+from outlines.types import String, Regex
+
+# A literal string "hello"
+literal = String("hello")   # Internally represents "hello"
+
+# A regex pattern to match one or more digits
+digit = Regex(r"[0-9]+")     # Internally represents the pattern [0-9]+
+
+# Converting to standard regex strings:
+from outlines.types.dsl import to_regex
+
+print(to_regex(literal))  # Output: hello
+print(to_regex(digit))    # Output: [0-9]+
+```
+
+---
+
+## Early Introduction to Quantifiers & Combining Terms
+
+The DSL supports common regex quantifiers as methods on every `Term`. These methods allow you to specify how many times a pattern should be matched. They include:
+
+- **`exactly(count)`**: Matches the term exactly `count` times.
+- **`optional()`**: Matches the term zero or one time.
+- **`one_or_more()`**: Matches the term one or more times (Kleene Plus).
+- **`zero_or_more()`**: Matches the term zero or more times (Kleene Star).
+- **`between(min_count, max_count)`**: Matches the term between `min_count` and `max_count` times (inclusive).
+- **`at_least(count)`**: Matches the term at least `count` times.
+- **`at_most(count)`**: Matches the term up to `count` times.
+
+These quantifiers can also be used as functions that take the `Term` as an argument. If the term is a plain string, it will be automatically converted to a `String` object. Thus `String("foo").optional()` is equivalent to `optional("foo")`.
+
+Let's see these quantifiers side by side with examples.
+
+### Quantifiers in Action
+
+#### `exactly(count)`
+
+This method restricts the term to appear exactly `count` times.
+
+```python
+# Example: exactly 5 digits
+five_digits = Regex(r"\d").exactly(5)
+print(to_regex(five_digits))  # Output: (\d){5}
+```
+
+You can also use the `exactly` function:
+
+```python
+from outlines.types import exactly
+
+# Example: exactly 5 digits
+five_digits = exactly(Regex(r"\d"), 5)
+print(to_regex(five_digits))  # Output: (\d){5}
+```
+
+#### `optional()`
+
+This method makes a term optional, meaning it may occur zero or one time.
+
+```python
+# Example: an optional "s" at the end of a word
+maybe_s = String("s").optional()
+print(to_regex(maybe_s))  # Output: (s)?
+```
+
+You can also use the `optional` function:
+
+```python
+from outlines.types import optional
+
+# Example: an optional "s" at the end of a word
+maybe_s = optional("s")
+print(to_regex(maybe_s))  # Output: (s)?
+```
+
+#### `one_or_more()`
+
+This method indicates that the term must appear at least once.
+
+```python
+# Example: one or more alphabetic characters
+letters = Regex(r"[A-Za-z]").one_or_more()
+print(to_regex(letters))  # Output: ([A-Za-z])+
+```
+
+You can also use the `one_or_more` function:
+
+```python
+from outlines.types import one_or_more
+
+# Example: one or more alphabetic characters
+letters = one_or_more(Regex(r"[A-Za-z]"))
+print(to_regex(letters))  # Output: ([A-Za-z])+
+
+```
+
+#### `zero_or_more()`
+
+This method indicates that the term can occur zero or more times.
+
+```python
+# Example: zero or more spaces
+spaces = String(" ").zero_or_more()
+print(to_regex(spaces))  # Output: ( )*
+```
+
+You can also use the `zero_or_more` function:
+
+```python
+from outlines.types import zero_or_more
+
+# Example: zero or more spaces
+spaces = zero_or_more(" ")
+print(to_regex(spaces))  # Output: ( )*
+```
+
+#### `between(min_count, max_count)`
+
+This method indicates that the term can appear any number of times between `min_count` and `max_count` (inclusive).
+
+```python
+# Example: Between 2 and 4 word characters
+word_chars = Regex(r"\w").between(2, 4)
+print(to_regex(word_chars))  # Output: (\w){2,4}
+```
+
+You can also use the `between` function:
+
+```python
+from outlines.types import between
+
+# Example: Between 2 and 4 word characters
+word_chars = between(Regex(r"\w"), 2, 4)
+print(to_regex(word_chars))  # Output: (\w){2,4}
+```
+
+#### `at_least(count)`
+
+This method indicates that the term must appear at least `count` times.
+
+```python
+# Example: At least 3 digits
+at_least_three = Regex(r"\d").at_least(3)
+print(to_regex(at_least_three))  # Output: (\d){3,}
+```
+
+You can also use the `at_least` function:
+
+```python
+from outlines.types import at_least
+
+# Example: At least 3 digits
+at_least_three = at_least(Regex(r"\d"), 3)
+print(to_regex(at_least_three))  # Output: (\d){3,}
+```
+
+#### `at_most(count)`
+
+This method indicates that the term can appear at most `count` times.
+
+```python
+# Example: At most 3 digits
+up_to_three = Regex(r"\d").at_most(3)
+print(to_regex(up_to_three))  # Output: (\d){0,3}
+```
+
+You can also use the `at_most` function:
+
+```python
+from outlines.types import at_most
+
+# Example: At most 3 digits
+up_to_three = at_most(Regex(r"\d"), 3)
+print(to_regex(up_to_three))  # Output: (\d){0,3}
+```
+
+---
+
+## Combining Terms
+
+The DSL allows you to combine basic terms into more complex patterns using concatenation and alternation.
+
+### Concatenation (`+`)
+
+The `+` operator (and its reflected variant) concatenates terms, meaning that the terms are matched in sequence.
+
+```python
+# Example: Match "hello world"
+pattern = String("hello") + " " + Regex(r"\w+")
+print(to_regex(pattern))  # Output: hello\ (\w+)
+```
+
+### Alternation (`either()`)
+
+The `either()` function creates alternatives, allowing a match for one of several patterns. You can provide as many terms as you want.
+
+```python
+# Example: Match either "cat" or "dog" or "mouse"
+animal = either(String("cat"), "dog", "mouse")
+print(to_regex(animal))  # Output: (cat|dog|mouse)
+```
+
+*Note:* When using `either()` with plain strings (such as `"dog"`), the DSL automatically wraps them in a `String` object that escapes the characters that have a special meaning in regular expressions, just like with quantifier functions.
+
+---
+
+## Custom types
+
+The DSL comes "batteries included" with types that represent common text constructs:
+
+- `integer` represents an integer number as recognized by `int`
+- `boolean` represents a boolean, "True" or "False" as recognized by `bool`
+- `number` represents a floating-point number recognize by Python's `float`
+- `date` represents a date as understood by `datetime.date`
+- `time` represents a time as undestood by `datetime.time`
+- `datetime` represents a time as understoof by `datetime.datetime`
+- `digit` represents a single digit
+- `char` represents a single character
+- `newline` represents a new line character
+- `whitespace` represents a white space
+- `sentence` represents a sentence
+- `paragraph` reprensents a pagraph (one or more sentences separated by one or more ilne breaks)
+
+
+For instance you can describe the answers in the GSM8K dataset using the following pattern:
+
+```python
+from outlines.types import sentence, digit
+
+answer = "A: " + sentence.between(2,4) + " So the answer is: " + digit.between(1,4)
+```
+
+---
+
+## Practical Examples
+
+### Example 1: Matching a Custom ID Format
+
+Suppose you want to create a regex that matches an ID format like "ID-12345", where:
+- The literal "ID-" must be at the start.
+- Followed by exactly 5 digits.
+
+```python
+id_pattern = "ID-" + Regex(r"\d").exactly(5)
+print(to_regex(id_pattern))  # Output: ID-(\d){5}
+```
+
+### Example 2: Email Validation with Pydantic
+
+You can define a regex for email validation and use it as a type in a Pydantic model.
+
+```python
+from pydantic import BaseModel, ValidationError
+
+# Define an email regex term (this is a simplified version)
+email_regex = Regex(r"[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+")
+
+class User(BaseModel):
+    name: str
+    email: email_regex  # Use our DSL regex as a field type
+
+# Valid input
+user = User(name="Alice", email="alice@example.com")
+print(user)
+
+# Invalid input (raises a ValidationError)
+try:
+    User(name="Bob", email="not-an-email")
+except ValidationError as e:
+    print(e)
+```
+
+When used in a Pydantic model, the email field is automatically validated against the regex pattern and its JSON Schema includes the `pattern` constraint.
+
+### Example 3: Building a Complex Pattern
+
+Consider a pattern to match a simple date format: `YYYY-MM-DD`.
+
+```python
+year = Regex(r"\d").exactly(4)         # Four digits for the year
+month = Regex(r"\d").exactly(2)        # Two digits for the month
+day = Regex(r"\d").exactly(2)          # Two digits for the day
+
+# Combine with literal hyphens
+date_pattern = year + "-" + month + "-" + day
+print(to_regex(date_pattern))
+# Output: (\d){4}\-(\d){2}\-(\d){2}
+```
+
+---
+
+## Visualizing Your Pattern
+
+One of the unique features of this DSL is that each term can print its underlying structure as an ASCII tree. This visualization can be particularly helpful when dealing with complex expressions.
+
+```python
+# A composite pattern using concatenation and quantifiers
+pattern = "a" + String("b").one_or_more() + "c"
+print(pattern)
+```
+
+*Expected Output:*
+
+```
+└── Sequence
+    ├── String('a')
+    ├── KleenePlus(+)
+    │   └── String('b')
+    └── String('c')
+```
+
+This tree representation makes it easy to see the hierarchy and order of operations in your regular expression.
+
+---
+
+## Final Words
+
+This DSL is designed to simplify the creation and management of regular expressions—whether you're validating inputs in a web API, constraining the output of an LLM, or just experimenting with regex patterns. With intuitive methods for common quantifiers and operators, clear visual feedback, and built-in integration with Pydantic, you can build robust and maintainable regex-based validations with ease.
+
+Feel free to explore the library further and adapt the examples to your use cases. Happy regexing!
diff --git a/docs_new/features/utility/template.md b/docs_new/features/utility/template.md
index 3a9cb72a8..631114f95 100644
--- a/docs_new/features/utility/template.md
+++ b/docs_new/features/utility/template.md
@@ -1,116 +1,85 @@
 ---
-title: Template API
+title: Template
 ---
 
 # Template
 
-::: outlines.templates.Template
+Outlines templates provide a way of creating reusable prompt structures with placeholders for dynamic content.
 
-Templates provide a way to create reusable prompt structures with placeholders for dynamic content.
+## Overview
 
-## Basic Usage
+To create a `Template` instance, you can use two class methods:
+- `from_string`: Creates a template from a string containing a Jinja2 template
+- `from_file`: Creates a template from a file containing a Jinja2 template
+
+After creating a template, you can call it with the variables required by the template as keyword arguments.
+
+For instance:
 
 ```python
 from outlines import Template
 
-# Create a simple template
-template = Template("Hello, {name}! The weather today is {weather}.")
+# Create a template from a string
+template_str = """
+Hello, {{ name }}!
+The weather today is {{ weather }}.
+"""
+template = Template.from_string(template_str)
+
+# Create a template from a file, assuming the content of template_str is put into a file
+template = Template.from_file("path_to/my_file.txt")
 
-# Fill in the template
-prompt = template.format(name="Alice", weather="sunny")
-print(prompt)  # Output: "Hello, Alice! The weather today is sunny."
+# Call the template to render the prompt
+prompt: str = template(name="Alice", weather="sunny")
+print(prompt)  # "Hello, Alice!\nThe weather today is sunny."
 ```
 
-## Composing Templates
+## Composite Templates
 
-Templates can be nested and composed:
+Templates can be nested and composed to create complex prompt structures:
 
 ```python
 from outlines import Template
 
 # Create component templates
-user_template = Template("User: {query}")
-system_template = Template("System: {instruction}")
+user_template = Template.from_string("User: {{ query }}")
+system_template = Template.from_string("System: {{ instruction }}")
 
-# Create a composed template
-chat_template = Template("""
-{system}
-
-{user}
+# Create a composite template
+chat_template = Template.from_string("""
+{{ system }}
+{{ user }}
 """)
 
 # Fill in nested templates
-prompt = chat_template.format(
-    system=system_template.format(instruction="You are a helpful assistant."),
-    user=user_template.format(query="What is machine learning?")
+prompt = chat_template(
+    system=system_template(instruction="You are a helpful assistant."),
+    user=user_template(query="What is machine learning?")
 )
 
 print(prompt)
-# Output:
 # System: You are a helpful assistant.
 #
 # User: What is machine learning?
 ```
 
-## Using with Generator
-
-Templates are useful when combined with the Generator class:
-
-```python
-from outlines import Template, Generator, from_transformers
-
-# Create a template
-template = Template("""
-Translate the following {source_language} text to {target_language}:
-
-{text}
-
-Translation:
-""")
-
-# Initialize a model
-model = from_transformers(...)
+## Custom Filters
 
-# Create a generator
-generator = Generator(model)
+You can add custom filters to your Outlines template to extend the templating functionality. To do so, provide as second argument a dictionnary with filter names as keys and filter functions as values. The filter can then be used in your jinja2 template following the regular syntax. When rendering a prompt, the function will be applied to the associated variable.
 
-# Use the template with the generator
-translation = generator(
-    template.format(
-        source_language="English",
-        target_language="French",
-        text="Hello, how are you today?"
-    )
-)
-
-print(translation)  # Output: "Bonjour, comment allez-vous aujourd'hui?"
-```
-
-## Multimodal Templates
-
-For multimodal inputs, use the `Vision` class:
+For instance:
 
 ```python
-from outlines import Template, Generator, from_transformers, Vision
-from PIL import Image
-
-# Create a vision template
-template = Template("""
-Describe what you see in this image: {image}
-""")
-
-# Load an image
-image = Image.open("example.jpg")
-
-# Format the template with the image
-prompt = template.format(image=Vision(image))
-
-# Initialize a multimodal model
-model = from_transformers(...)
+from outlines import Template
 
-# Generate a description
-generator = Generator(model)
-description = generator(prompt)
+def uppercase(text: str) -> str:
+    return text.upper()
 
-print(description)
+# Add custom filter when creating template
+template = Template.from_string(
+    "Hello {{ name | uppercase }}!",
+    filters={"uppercase": uppercase}
+)
+prompt = template(name="alice")
+print(prompt)  # "Hello ALICE!"
 ```
diff --git a/mkdocs.yml b/mkdocs.yml
index e09fa3cbc..a09f7454c 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -206,7 +206,8 @@ nav:
                 - vLLM (offline): features/models/vllm_offline.md
 
           - Utilities:
-                - Applications: features/utility/application.md
+                - Application: features/utility/application.md
+                - Regex DSL: features/utility/regex_dsl.md
                 - Template: features/utility/template.md
 
           - Advanced:

From 962a847f84344448c0bc43cd0d92536a5a1c1b59 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 26 May 2025 10:22:25 +0200
Subject: [PATCH 331/505] Update the advanced features documentation

---
 docs_new/features/advanced/guide.md           |  1 -
 .../features/advanced/logits_processors.md    | 98 +++++++++++++++++++
 docs_new/features/advanced/parsing.md         |  1 -
 mkdocs.yml                                    |  3 +-
 4 files changed, 99 insertions(+), 4 deletions(-)
 delete mode 100644 docs_new/features/advanced/guide.md
 create mode 100644 docs_new/features/advanced/logits_processors.md
 delete mode 100644 docs_new/features/advanced/parsing.md

diff --git a/docs_new/features/advanced/guide.md b/docs_new/features/advanced/guide.md
deleted file mode 100644
index 08d2f7f0f..000000000
--- a/docs_new/features/advanced/guide.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.processors.guide
diff --git a/docs_new/features/advanced/logits_processors.md b/docs_new/features/advanced/logits_processors.md
new file mode 100644
index 000000000..cbd7b82a0
--- /dev/null
+++ b/docs_new/features/advanced/logits_processors.md
@@ -0,0 +1,98 @@
+---
+title: Logits Processors
+---
+
+# Logits Processors
+
+Logits processors are objects that control text generation by modifying the probability distribution of possible next tokens. They do this by adjusting the logits (raw model outputs) at each generation step, effectively biasing the model's token selection.
+Processors can be used to:
+1. Generate structured output (e.g., JSON that follows a specific schema)
+2. Prevent the model from generating specific words or tokens
+3. Implement custom token sampling strategies
+
+## Overview
+
+Outlines uses logits processors with steerable models — models that run locally and allow fine-grained control over the generation process. When using such models in Outlines, the output type provided is turned into a logits processor that is then passed to the inference engine.
+
+There are four models that support logits processors:
+
+- LlamaCpp
+- MLXLM
+- Transformers
+- VLLMOffline
+
+Instead of providing an output type that will be turned into a logits processor, it is possible to directly provide a logits processor. To do so, you must create a `Generator` instance using the `processor` keyword argument. You cannot directly call the model with a logits processor.
+
+For instance:
+
+```python
+import transformers
+from outlines import Generator, from_transformers
+from outlines.processors import RegexLogitsProcessor
+
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B"),
+    transformers.AutoTokenizer.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
+)
+
+# Create a regex logits processor that only returns hex unicode notations
+logits_processor = RegexLogitsProcessor(r"U\+[0-9A-Fa-f]{4,6}", model.tokenizer, model.tensor_library_name)
+
+# Create a generator with the logits processor and use it to generate text
+generator = Generator(model, processor=logits_processor)
+response = generator("What's the unicode for the hugging face emoji")
+
+print(response) # U+1F917
+```
+
+## Creating Custom Logits Processors
+
+You can create your own logits processor by subclassing the `OutlinesLogitsProcessor` class. This allows you to implement specific logic to modify logits as needed.
+Your logits processor needs to implement the `process_logits` method to modify the logits.
+`process_logits` accepts:
+- `input_ids`: the ids of the tokens of the existing sequences in a 2D tensor.
+- `logits`: the logits for the current generation step in a 2D tensor.
+
+In the example below, we create a custom logits processor to force the model to provide a response using only binary representation (so only the tokens for 0 and 1 are allowed):
+
+```python
+from outlines.processors.base_logits_processor import OutlinesLogitsProcessor, TensorType
+from outlines import Generator, from_transformers
+import transformers
+
+ALLOWED_TOKENS = [15, 16]  # token IDs corresponding to '0' and '1' in the model's vocabulary
+
+# Subclass OutlinesLogitsProcessor
+class BinaryLogitsProcessor(OutlinesLogitsProcessor):
+
+    def process_logits(self, input_ids: TensorType, logits: TensorType) -> TensorType:
+        # Create a mask for all tokens
+        mask = self.tensor_adapter.boolean_ones_like(logits)
+        # Set mask to False for the allowed tokens
+        for token_id in ALLOWED_TOKENS:
+            mask[:, token_id] = False
+        # Set non-allowed tokens to -inf so they are not selected
+        logits[mask] = float("-inf")
+        return logits
+
+# Create a regular model
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
+tf_model = transformers.AutoModelForCausalLM.from_pretrained("NousResearch/Hermes-2-Pro-Llama-3-8B")
+model = from_transformers(tf_model, tf_tokenizer)
+
+# Instantiate your custom logits processor
+logits_processor = BinaryLogitsProcessor(model.tensor_library_name)
+
+prompt = "Write the number 47 in binary. For example, 1010 is the binary representation of 10. Answer just with the binary number composed of 0s and 1s."
+formatted_prompt = tf_tokenizer.apply_chat_template(
+    [{"role": "user", "content": prompt}],
+    tokenize=False
+)
+
+# Create a generator with the custom logits processor instance and use it to generate text
+generator = Generator(model, processor=logits_processor)
+response = generator(formatted_prompt)
+
+print(response) # "101111"
+```
diff --git a/docs_new/features/advanced/parsing.md b/docs_new/features/advanced/parsing.md
deleted file mode 100644
index e9662999c..000000000
--- a/docs_new/features/advanced/parsing.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.fsm.parsing
diff --git a/mkdocs.yml b/mkdocs.yml
index a09f7454c..f09d31747 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -211,8 +211,7 @@ nav:
                 - Template: features/utility/template.md
 
           - Advanced:
-                - Guide: features/advanced/guide.md
-                - Parsing: features/advanced/parsing.md
+                - Logits Processors: features/advanced/logits_processors.md
 
     - API Reference: api_reference/
 

From 884e9838fde026fd875d87e67fb30de5607a80bb Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 26 May 2025 14:13:15 +0200
Subject: [PATCH 332/505] Remove forgottenly committed print statement

---
 outlines/models/transformers.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 4bece9546..785e2c983 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -230,8 +230,6 @@ def generate(self, model_input, output_type, **inference_kwargs):
         prompts, inputs = self._prepare_model_inputs(model_input, output_type)
         logits_processor = self.type_adapter.format_output_type(output_type)
 
-        print(inference_kwargs)
-
         generated_ids = self._generate_output_seq(
             prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )

From a99321060ab67c1c80678aa3cb861e55b24db5e6 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 26 May 2025 14:03:47 +0200
Subject: [PATCH 333/505] Rename SgLang into SGLang

---
 outlines/models/__init__.py              |  4 ++--
 outlines/models/sglang.py                | 14 +++++++-------
 tests/models/test_sglang.py              | 10 +++++-----
 tests/models/test_sglang_type_adapter.py |  4 ++--
 4 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 3dffe64d4..6ffc62320 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -16,7 +16,7 @@
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
 from .openai import from_openai, OpenAI
-from .sglang import from_sglang, SGLang, AsyncSgLang
+from .sglang import from_sglang, SGLang, AsyncSGLang
 from .tgi import from_tgi, TGI, AsyncTGI
 from .transformers import (
     Transformers,
@@ -35,7 +35,7 @@
 BlackBoxModel = Union[
     Anthropic,
     AsyncTGI,
-    AsyncSgLang,
+    AsyncSGLang,
     AsyncVLLM,
     Dottxt,
     Gemini,
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
index 0d3fdaab6..8a4d90ffa 100644
--- a/outlines/models/sglang.py
+++ b/outlines/models/sglang.py
@@ -11,7 +11,7 @@
     from openai import AsyncOpenAI, OpenAI
 
 
-class SgLangTypeAdapter(ModelTypeAdapter):
+class SGLangTypeAdapter(ModelTypeAdapter):
 
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the client.
@@ -71,7 +71,7 @@ def __init__(self, client, model_name: Optional[str] = None):
         """
         self.client = client
         self.model_name = model_name
-        self.type_adapter = SgLangTypeAdapter()
+        self.type_adapter = SGLangTypeAdapter()
 
     def generate(self, model_input, output_type, **inference_kwargs):
         """Generate text using the client.
@@ -156,11 +156,11 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
         return client_args
 
 
-class AsyncSgLang(AsyncModel):
+class AsyncSGLang(AsyncModel):
     """Represents an async client to a `sglang` server."""
 
     def __init__(self, client, model_name: Optional[str] = None):
-        """Create an `AsyncSgLang` model instance.
+        """Create an `AsyncSGLang` model instance.
 
         Parameters
         ----------
@@ -170,7 +170,7 @@ def __init__(self, client, model_name: Optional[str] = None):
         """
         self.client = client
         self.model_name = model_name
-        self.type_adapter = SgLangTypeAdapter()
+        self.type_adapter = SGLangTypeAdapter()
 
     async def generate(self, model_input, output_type, **inference_kwargs):
         """Generate text using `sglang`.
@@ -259,13 +259,13 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
 def from_sglang(
     client: Union["OpenAI", "AsyncOpenAI"],
     model_name: Optional[str] = None,
-) -> Union[SGLang, AsyncSgLang]:
+) -> Union[SGLang, AsyncSGLang]:
     from openai import AsyncOpenAI, OpenAI
 
     if isinstance(client, OpenAI):
         return SGLang(client, model_name)
     elif isinstance(client, AsyncOpenAI):
-        return AsyncSgLang(client, model_name)
+        return AsyncSGLang(client, model_name)
     else:
         raise ValueError(
             f"Unsupported client type: {type(client)}.\n"
diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
index e04fb3d10..60960f463 100644
--- a/tests/models/test_sglang.py
+++ b/tests/models/test_sglang.py
@@ -11,7 +11,7 @@
 import pytest
 from openai import AsyncOpenAI, OpenAI
 
-from outlines.models.sglang import SGLang, AsyncSgLang, from_sglang
+from outlines.models.sglang import SGLang, AsyncSGLang, from_sglang
 from outlines.types.dsl import CFG, Regex, JsonSchema
 from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
 
@@ -130,12 +130,12 @@ def sync_model_no_model_name():
 
 @pytest.fixture
 def async_model():
-    return AsyncSgLang(async_openai_client, model_name=sglang_model_name)
+    return AsyncSGLang(async_openai_client, model_name=sglang_model_name)
 
 
 @pytest.fixture
 def async_model_no_model_name():
-    return AsyncSgLang(async_openai_client)
+    return AsyncSGLang(async_openai_client)
 
 
 def test_sglang_init():
@@ -158,13 +158,13 @@ def test_sglang_init():
 
     # Async with model name
     model = from_sglang(async_openai_client, sglang_model_name)
-    assert isinstance(model, AsyncSgLang)
+    assert isinstance(model, AsyncSGLang)
     assert model.client == async_openai_client
     assert model.model_name == sglang_model_name
 
     # Async without model name
     model = from_sglang(async_openai_client)
-    assert isinstance(model, AsyncSgLang)
+    assert isinstance(model, AsyncSGLang)
     assert model.client == async_openai_client
     assert model.model_name is None
 
diff --git a/tests/models/test_sglang_type_adapter.py b/tests/models/test_sglang_type_adapter.py
index c7ac2b6c8..7530174f9 100644
--- a/tests/models/test_sglang_type_adapter.py
+++ b/tests/models/test_sglang_type_adapter.py
@@ -4,7 +4,7 @@
 from dataclasses import dataclass
 
 from PIL import Image
-from outlines.models.sglang import SgLangTypeAdapter
+from outlines.models.sglang import SGLangTypeAdapter
 from outlines.types import CFG, JsonSchema
 
 from outlines import Vision
@@ -27,7 +27,7 @@
 
 @pytest.fixture
 def type_adapter():
-    return SgLangTypeAdapter()
+    return SGLangTypeAdapter()
 
 @pytest.fixture
 def cfg_instance():

From 5bf0f9315b10aaf3009d21b64fa5f69f30f025e2 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 26 May 2025 18:33:14 +0200
Subject: [PATCH 334/505] Improve type hints and docstrings for the processors
 module

---
 outlines/processors/__init__.py               |   2 +
 outlines/processors/base_logits_processor.py  |  87 +++++--
 outlines/processors/guide.py                  | 228 +++++++++++++++---
 outlines/processors/structured.py             | 124 ++++++----
 .../processors/tensor_adapters/__init__.py    |   6 +-
 outlines/processors/tensor_adapters/base.py   | 219 +++++++++++++++--
 outlines/processors/tensor_adapters/jax.py    |   2 +
 outlines/processors/tensor_adapters/mlx.py    |   2 +
 outlines/processors/tensor_adapters/numpy.py  |   2 +
 .../processors/tensor_adapters/tensorflow.py  |   2 +
 outlines/processors/tensor_adapters/torch.py  |   2 +
 tests/processors/test_guide.py                |   8 +-
 12 files changed, 546 insertions(+), 138 deletions(-)

diff --git a/outlines/processors/__init__.py b/outlines/processors/__init__.py
index f0f0f829b..f07ce0cbb 100644
--- a/outlines/processors/__init__.py
+++ b/outlines/processors/__init__.py
@@ -1,3 +1,5 @@
+"""Processors and guides to control generation in steerable models."""
+
 from .structured import (
     CFGLogitsProcessor,
     GuideLogitsProcessor,
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index ba4df1036..bde129b2c 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -1,57 +1,91 @@
+"""Base class for logits processors."""
+
 from abc import abstractmethod
 from typing import TypeVar
 
-from outlines.processors.tensor_adapters import tensor_adapters, TensorAdapterImplementation
+from outlines.processors.tensor_adapters import (
+    TensorAdapterImplementation,
+    tensor_adapters,
+)
 
 TensorType = TypeVar('TensorType')
 
 
 class OutlinesLogitsProcessor:
     """Base class for logits processors.
-
-    This class standardizes the shape of input_ids and logits received and
-    returned by the logits processor. At initialization, it creates a
-    tensor_adapter attribute which is used to manipulate the tensors
-    using the appropriate library for the model (numpy, torch...).
+    This class implements a shared `__call__` method is called by the models
+    and returns the processed logits. It relies on the `process_logits` method
+    that must be implemented by the subclasses to do the actual processing. The
+    `tensor_adapter` attribute, created at initialization based on the
+    tensor library name specified in the constructor, is used to manipulate the
+    tensors using the appropriate library for the model (numpy, torch...).
     """
     tensor_adapter: TensorAdapterImplementation
 
     def __init__(self, tensor_library_name: str):
+        """
+        Parameters
+        ----------
+        tensor_library_name
+            The name of the library to use to manipulate tensors. Possible
+            values are "jax", "mlx", "numpy", "tensorflow" and "torch". You
+            must choose the library that your model is using.
+        """
         tensor_adapter_class = tensor_adapters.get(tensor_library_name)
         if tensor_adapter_class is None:
-            raise NotImplementedError(f"Library {tensor_library_name} is not available")
+            raise NotImplementedError(
+                f"Library {tensor_library_name} is not available"
+            )
         self.tensor_adapter = tensor_adapter_class()  # type: ignore
 
     @abstractmethod
     def process_logits(
         self, input_ids: TensorType, logits: TensorType
     ) -> TensorType:
-        """Main method to implement for logits processors.
-
+        """Main method to implement for logits processors subclasses.
         This method applies a mask on the logits to bias the generation.
-        It is called by the __call__ method that standardizes the shape of
-        input_ids and logits to ensure they are 2D tensors.
-
-        Important to keep in mind when designing universal logits processors
+        It is called by the `__call__` method that standardizes the shape of
+        `input_ids` and `logits` to ensure they are 2D tensors.
+        Elements to keep in mind when designing universal logits processors:
         - logits processors are only used once and never re-applied for a new
         sequence generator
         - Some models only pass output_ids, some models such as llamacpp and
         transformers prefix with input_ids
         - Some sampling methods, such as beam search, result in unstable
         sequence ordering in models like vLLM
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences in a 2D tensor.
+        logits
+            The logits for the current generation step in a 2D tensor.
+        Returns
+        -------
+        TensorType
+            The processed logits as a 2D tensor.
         """
         ...
 
     def __call__(
         self, input_ids: TensorType, logits: TensorType
     ) -> TensorType:
-        """Entrypoint for logits processors, this is the method that is called
-        by the model.
-
-        Because different models use different libraries, we need to standardize
-        the shape of input_ids and logits to ensure they are 2D tensors before
-        calling the unified process_logits method. After processing, the logits
-        are cast back to the original array library type before being returned.
+        """Entrypoint for logits processors, this is the method that is
+        called by the model.
+        Because different models use different structures to store the
+        input_ids and logits, we standardize their format to 2D tensors
+        before calling the `process_logits` method. After processing, the
+        logits are cast back to the original array library type before being
+        returned.
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences in a tensor.
+        logits
+            The logits for the current generation step in a tensor.
+        Returns
+        -------
+        TensorType
+            The processed logits as a tensor.
         """
         # if input_ids is 1D and logits is 2D with a single sequence,
         # reshape input_ids to 2D (needed for mlx-lm)
@@ -62,9 +96,13 @@ def __call__(
         ):
             input_ids = self.tensor_adapter.unsqueeze(input_ids)
 
-        assert self.tensor_adapter.shape(logits)[:-1] == self.tensor_adapter.shape(input_ids)[:-1]
+        assert (
+            self.tensor_adapter.shape(logits)[:-1]
+            == self.tensor_adapter.shape(input_ids)[:-1]
+        )
 
-        # Guarantee passed as 2D Tensors, then covert back to original (1D or 2D) shape
+        # Guarantee passed as 2D Tensors, then covert back to original
+        # (1D or 2D) shape
         if len(self.tensor_adapter.shape(logits)) == 2:
             processed_logits = self.process_logits(input_ids, logits)
         elif len(self.tensor_adapter.shape(logits)) == 1:
@@ -75,6 +113,9 @@ def __call__(
                 ),
             )
         else:
-            raise ValueError(f"Logits shape {self.tensor_adapter.shape(logits)} is not supported")
+            raise ValueError(
+                f"Logits shape {self.tensor_adapter.shape(logits)} is not "
+                + "supported"
+            )
 
         return processed_logits
diff --git a/outlines/processors/guide.py b/outlines/processors/guide.py
index 983050c09..8139417e4 100644
--- a/outlines/processors/guide.py
+++ b/outlines/processors/guide.py
@@ -1,7 +1,12 @@
+"""Guides to control generation in steerable models.
+
+Logits processors rely on guides to control the generation process.
+"""
+
 import collections
 import copy
 import warnings
-from typing import TYPE_CHECKING, Any, Generator, Union
+from typing import TYPE_CHECKING, Any, Generator, ValuesView, Union
 
 import torch
 from lark.indenter import DedentError
@@ -28,60 +33,106 @@
 class Guide(CoreGuide):
     """Base definition of a generation guide.
 
-    A generation guide defines the behavior of a finite-state machine that guides
-    a text generation procedure. Unlike the DFAs built from regular expressions
-    guides can also emit a `Write` instructions which tells the model that it can
-    append a sequence of tokens (or token word) instead of generating it.
+    A generation guide defines the behavior of a finite-state machine that
+    guides a text generation procedure. Unlike the DFAs built from regular
+    expressions guides, it can also emit a `Write` instructions which tells
+    the model that it can append a sequence of tokens (or token word) instead
+    of generating it.
 
     """
-
     initial_state: Any
 
 
 class StopAtEOSGuide(Guide):
     """Guide to generate tokens until the EOS token has been generated."""
-
     final_state = 1
-    start_state = 0  # TODO: remove start_state, use only initial_state
     initial_state = 0
 
     def __init__(self, tokenizer: "Tokenizer"):
-        """Initialize the generation guide.
-
-        model
-            The logit generator used to generate the next token.
+        """
+        Parameters
+        ----------
+        tokenizer
+            The tokenizer used to convert tokens to ids.
 
         """
         self.eos_token_id = tokenizer.eos_token_id
         self.vocabulary = tokenizer.vocabulary.values()
 
     def get_next_instruction(self, state: int) -> Instruction:
+        """Return the next instruction.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        Instruction
+            An `Instruction` instance.
+
+        """
         if self.is_final_state(state):
             return Write([self.eos_token_id])
         return Generate(None)
 
     def get_next_state(self, state: int, token_id: int) -> int:
+        """Return the next state.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+        token_id
+            The id of the token that was just generated.
+
+        Returns
+        -------
+        int
+            The next state.
+
+        """
         if token_id == self.eos_token_id or state == self.final_state:
             return self.final_state
 
         return self.initial_state
 
-    def is_final_state(self, state: int):
+    def is_final_state(self, state: int) -> bool:
+        """Return whether the given state is a final state.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether the given state is a final state.
+
+        """
         return state == self.final_state
 
-    def copy(self):
+    def copy(self) -> "StopAtEOSGuide":
+        """Return itself as there is no need to copy."""
         return self
 
 
 @cache()
 def cached_create_states_mapping(regex_string, tokenizer, *args, **kwargs):
-    return uncached_create_states_mapping(regex_string, tokenizer, *args, **kwargs)
+    """Wrap the uncached create_states_mapping function in a cache."""
+    return uncached_create_states_mapping(
+        regex_string, tokenizer, *args, **kwargs
+    )
 
 
 class RegexGuide(CoreRegexGuide):
-    """
-    Guide to generate text in the language of a regular expression.
-    CoreRegexGuide with outlines cache
+    """Guide to generate text in the language of a regular expression.
+
+    This class is a wrapper around the CoreRegexGuide class that adds a cache
+    to the create_states_mapping function.
+
     """
 
     @classmethod
@@ -91,6 +142,23 @@ def from_regex(
         tokenizer,
         **kwargs,
     ):
+        """Create a RegexGuide from a regular expression.
+
+        Parameters
+        ----------
+        regex_string
+            The regular expression to generate text from.
+        tokenizer
+            The tokenizer to use to convert tokens to ids.
+        kwargs
+            Additional keyword arguments to pass to the CoreRegexGuide constructor.
+
+        Returns
+        -------
+        RegexGuide
+            A RegexGuide instance.
+
+        """
         return super().from_regex(
             regex_string,
             tokenizer,
@@ -103,15 +171,25 @@ def from_regex(
 
 
 class CFGGuide(Guide):
-    """Guide to generate text that is in the language of a context-free Lark grammar."""
+    """Guide to generate text that is in the language of a context-free Lark
+    grammar.
 
-    def __init__(self, cfg_string: str, tokenizer):
+    """
+
+    def __init__(self, cfg_string: str, tokenizer: "Tokenizer"):
         """
-        Construct the PartialLark parser and set the empty initial_state (PartialParserState)
+        Parameters
+        ----------
+        cfg_string
+            The context-free grammar to generate text from.
+        tokenizer
+            The tokenizer to use to convert tokens to ids.
+
         """
         warnings.warn(
-            "Outlines' public *community-contributed* CFG structured generation is experimental. "
-            "Please review https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
+            "Outlines' public *community-contributed* CFG structured generation "
+            "is experimental. Please review "
+            "https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
         )
 
         self.cfg_string = cfg_string
@@ -144,7 +222,9 @@ def get_next_instruction(self, state: CFGState) -> Instruction:
 
         Returns
         -------
-        A `Generate` instance that contains the model and the allowed token ids.
+        Instruction
+            A `Generate` instance that contains the model and the allowed token
+            ids.
 
         """
 
@@ -152,17 +232,21 @@ def get_next_instruction(self, state: CFGState) -> Instruction:
             return Write(torch.tensor([self.eos_token_id]))
 
         valid_tokens = list(
-            self.iter_valid_token_ids(state, self.tokenizer.vocabulary.values())
+            self.iter_valid_token_ids(
+                state, self.tokenizer.vocabulary.values()
+            )
         )
+
         if len(valid_tokens) == 1:
             return Write(torch.tensor(valid_tokens))
+
         return Generate(torch.tensor(valid_tokens))
 
     def iter_valid_token_ids(
-        self, state: CFGState, candidate_token_ids: list
+        self, state: CFGState, candidate_token_ids: ValuesView[int]
     ) -> Generator[int, None, None]:
-        """
-        Iterate over the given token_ids and yield those that are valid for the current parser state.
+        """Iterate over the given token_ids and yield those that are valid for
+        the current parser state.
 
         Parameters
         ----------
@@ -175,6 +259,7 @@ def iter_valid_token_ids(
         ------
         int
             Valid token ids.
+
         """
         for token_id in candidate_token_ids:
             if token_id == self.eos_token_id:
@@ -194,9 +279,10 @@ def iter_valid_token_ids(
                     pass
 
     def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
-        """
-        Update the state of the guide.
-        Decode the token_id, and calculate the new parser_state with the token applied.
+        """Update the state of the guide.
+
+        Decode the token_id, and calculate the new parser_state with the token
+        applied.
 
         Parameters
         ----------
@@ -207,7 +293,8 @@ def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
 
         Returns
         -------
-        The guides new PartialParserState
+        CFGState
+            The guides new PartialParserState
 
         """
         if state.parser_state is None or token_id == self.eos_token_id:
@@ -219,7 +306,8 @@ def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
     def _get_parser_state_token_applied(
         self, state: CFGState, token_id: int
     ) -> PartialParserState:
-        """
+        """Apply the given token_id to the parser state.
+
         Don't mutate `parser_state`, copy to protect
 
         Get the token string
@@ -227,6 +315,19 @@ def _get_parser_state_token_applied(
           - else: normalized (with possibly leading whitespace)
 
         Don't allow empty ("") tokens, raise ValueError
+
+        Parameters
+        ----------
+        state
+            The guide's current PartialParserState, or None if complete
+        token_id
+            The id of the token that was just generated.
+
+        Returns
+        -------
+        PartialParserState
+            The parser state with the token applied.
+
         """
         parser_state = copy.copy(state.parser_state)  # prevent side effects
 
@@ -250,12 +351,38 @@ def _get_parser_state_token_applied(
         return parser_state
 
     def is_final_state(self, state: CFGState) -> bool:
-        # TODO: remove this method, use can_terminate_state and must_terminate_state
-        # here and in RegexGuide per https://github.com/dottxt-ai/outlines/issues/885
+        """Return whether the given state is a final state.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether the given state is a final state.
+
+        """
+        # TODO: remove this method, use can_terminate_state and
+        # must_terminate_state here and in RegexGuide per
+        # https://github.com/dottxt-ai/outlines/issues/885
         return self.can_terminate_state(state)
 
     def can_terminate_state(self, state: CFGState) -> bool:
-        """Generation is allowed to terminate"""
+        """Return whether generation is allowed to terminate.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether generation is allowed to terminate.
+
+        """
         if state.parser_state is not None:
             try:
                 copy.copy(state.parser_state).feed_eof()
@@ -264,11 +391,32 @@ def can_terminate_state(self, state: CFGState) -> bool:
         return True
 
     def must_terminate_state(self, state: CFGState) -> bool:
-        """Generation must terminate, no legal continuations"""
-        return state.parser_state is None or set(state.parser_state.accepts()).issubset(
-            {"$END"}
+        """Indicate whether generation must terminate as there are no legal
+        continuations.
+
+        Parameters
+        ----------
+        state
+            The guide's current state.
+
+        Returns
+        -------
+        bool
+            Whether generation must terminate.
+
+        """
+        return (
+            state.parser_state is None or
+            set(state.parser_state.accepts()).issubset({"$END"})
         )
 
     def copy(self) -> "CFGGuide":
-        """Create a copy of the Guide."""
+        """Create a copy of the Guide.
+
+        Returns
+        -------
+        CFGGuide
+            A copy of the Guide.
+
+        """
         return CFGGuide(self.cfg_string, self.tokenizer)
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index 965806621..f1a55b902 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -1,4 +1,4 @@
-"""
+"""Logits processors for structured generation.
  _______________________________
 / Don't want to self-host?       \
 \\ Try .json at http://dottxt.co /
@@ -25,47 +25,55 @@
 """
 
 import math
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union, cast
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
 
-from outlines_core.fsm.json_schema import build_regex_from_schema
 from pydantic import BaseModel
+from outlines_core.fsm.json_schema import build_regex_from_schema
 
-from outlines.processors.guide import CFGGuide, Guide, RegexGuide
+from outlines.processors.base_logits_processor import (
+    OutlinesLogitsProcessor,
+    TensorType
+)
+from outlines.processors.guide import (
+    CFGGuide,
+    Guide,
+    RegexGuide
+)
 from outlines.types import JsonSchema
 
-from .base_logits_processor import OutlinesLogitsProcessor, TensorType
-
 if TYPE_CHECKING:
     from outlines.models.tokenizer import Tokenizer
 
 
 class GuideLogitsProcessor(OutlinesLogitsProcessor):
-    """Bias generation using a finite
+    """Bias generation using a guide.
 
     Attributes
     ----------
     tokenizer
-        The tokenizer used to convert tokens to ids.
+        The outlines tokenizer used to convert tokens to ids.
     guide
-        The `outlines.processors.guide` which is used to bias the logits.
+        The outlines guide used to bias the logits.
     """
-
     tokenizer: "Tokenizer"
     guide: Guide
     _guide_states: Dict[int, Any]
     _seq_start_idx: Optional[int]
 
-    def __init__(self, tokenizer: "Tokenizer", guide: Guide, tensor_library_name: str):
-        """A Guide-based logits processor.
-
+    def __init__(
+        self, tokenizer: "Tokenizer", guide: Guide, tensor_library_name: str
+    ):
+        """
         Parameters
         ----------
         tokenizer
             The tokenizer used to convert tokens to ids.
         guide
-            The `outlines.processors.guide. which is used to bias the logits.
+            The `outlines.processors.guide.Guide` that is used to bias the
+            logits.
         tensor_library_name
             The name of the library to use to manipulate the tensors.
+
         """
         super().__init__(tensor_library_name=tensor_library_name)
         self.tokenizer = tokenizer
@@ -81,13 +89,15 @@ def process_logits(
         Parameters
         ----------
         input_ids
-            The input token ids.
+            The ids of the tokens of the existing sequences.
         logits
-            The logits.
+            The logits for the current generation step.
 
         Returns
         -------
-        The biased logits.
+        TensorType
+            The biased logits.
+
         """
         if self._seq_start_idx is None:
             self._seq_start_idx = len(input_ids[0]) # type: ignore
@@ -141,14 +151,15 @@ def copy(self) -> "GuideLogitsProcessor":
 
 class RegexLogitsProcessor(GuideLogitsProcessor):
     """Bias generation based on a regular expression."""
+    guide: RegexGuide
+
     def __init__(
         self,
         regex_string: str,
         tokenizer: "Tokenizer",
         tensor_library_name: str,
     ):
-        """Compile the RegexGuide that drives the regex-guided generation.
-
+        """
         Parameters
         ----------
         regex_string
@@ -157,7 +168,10 @@ def __init__(
             An Outlines tokenizer.
         tensor_library_name
             The name of the library to use to manipulate the tensors.
+
         """
+        # Build a guide from the regex string and then pass it to the
+        # GuideLogitsProcessor superclass.
         guide = RegexGuide.from_regex(regex_string, tokenizer)
         super().__init__(tokenizer=tokenizer, guide=guide, tensor_library_name=tensor_library_name)
 
@@ -171,8 +185,7 @@ def __init__(
         tensor_library_name: str,
         whitespace_pattern: Optional[str] = None,
     ):
-        """Compile the Guide that drives the JSON-guided generation.
-
+        """
         Parameters
         ----------
         schema
@@ -185,48 +198,71 @@ def __init__(
             Pattern to use for JSON syntactic whitespace (doesn't impact string
             literals). For example, to allow only a single space or newline with
             `whitespace_pattern=r"[\n ]?"`.
+
         """
+        # Convert the JSON schema into a regex string and then pass it to the
+        # RegexLogitsProcessor superclass.
         schema_str = JsonSchema(schema).schema
         regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(regex_string=regex_string, tokenizer=tokenizer, tensor_library_name=tensor_library_name)
+        super().__init__(
+            regex_string=regex_string,
+            tokenizer=tokenizer,
+            tensor_library_name=tensor_library_name
+        )
 
 
 class CFGLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a context-free grammar.
-
-    Attributes
-    ----------
-    tokenizer
-        The tokenizer used to convert tokens to ids.
-    guide
-        The `outlines.fsm.CFGGuide. which is used to bias the logits.
-    """
-
+    """Bias generation based on a context-free grammar."""
     guide: CFGGuide
 
-    def __init__(self, cfg_str: str, tokenizer: "Tokenizer", tensor_library_name: str):
-        """Compile the CFGGuide that drives the CFG-guided generation.
-
+    def __init__(
+        self, cfg_str: str, tokenizer: "Tokenizer", tensor_library_name: str
+    ):
+        """
         Parameters
         ----------
         cfg_str
-            A string that represents a grammar
+            A string that represents a grammar.
         tokenizer
             The tokenizer used to convert tokens to ids.
+        tensor_library_name
+            The name of the library to use to manipulate the tensors.
+
         """
+        # Build a guide from the CFG string and then pass it to the
+        # GuideLogitsProcessor superclass.
         cfg_guide = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
-        super().__init__(tokenizer=tokenizer, guide=cfg_guide, tensor_library_name=tensor_library_name)
+        super().__init__(
+            tokenizer=tokenizer,
+            guide=cfg_guide,
+            tensor_library_name=tensor_library_name
+        )
 
     def process_logits(
-        self, input_ids, logits
-    ):
-        """Same behavior as GuideLogitsProcessor, but uses rejection sampling"""
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Same behavior as GuideLogitsProcessor, but uses rejection
+        sampling.
+
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences.
+        logits
+            The logits for the current generation step.
+
+        Returns
+        -------
+        TensorType
+            The biased logits.
+
+        """
         if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0])
+            self._seq_start_idx = len(input_ids[0]) # type: ignore
 
         sequence_states: List = []  # vector of states corresponding to `input_ids`
 
-        for seq_ids in input_ids:
+        for seq_ids in input_ids: # type: ignore
             gen_ids = seq_ids[self._seq_start_idx :]
             curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
 
@@ -241,9 +277,9 @@ def process_logits(
         for i, guide_state in enumerate(sequence_states):
             first_legal_token = next(
                 self.guide.iter_valid_token_ids(
-                    guide_state, self.tensor_adapter.argsort_descending(logits[i])
+                    guide_state, self.tensor_adapter.argsort_descending(logits[i]) # type: ignore
                 )
             )
-            mask[i, [first_legal_token]] = logits[i, [first_legal_token]]
+            mask[i, [first_legal_token]] = logits[i, [first_legal_token]] # type: ignore
 
         return mask
diff --git a/outlines/processors/tensor_adapters/__init__.py b/outlines/processors/tensor_adapters/__init__.py
index a61fa1ea5..8e3afe7f6 100644
--- a/outlines/processors/tensor_adapters/__init__.py
+++ b/outlines/processors/tensor_adapters/__init__.py
@@ -1,3 +1,5 @@
+"""Library specific objects to manipulate tensors."""
+
 from typing import Union
 
 from .jax import JAXTensorAdapter
@@ -11,14 +13,14 @@
     "jax": JAXTensorAdapter,
     "mlx": MLXTensorAdapter,
     "numpy": NumpyTensorAdapter,
-    "torch": TorchTensorAdapter,
     "tensorflow": TensorFlowTensorAdapter,
+    "torch": TorchTensorAdapter,
 }
 
 TensorAdapterImplementation = Union[
     JAXTensorAdapter,
     MLXTensorAdapter,
     NumpyTensorAdapter,
-    TorchTensorAdapter,
     TensorFlowTensorAdapter,
+    TorchTensorAdapter,
 ]
diff --git a/outlines/processors/tensor_adapters/base.py b/outlines/processors/tensor_adapters/base.py
index fedcb6924..ffaac58df 100644
--- a/outlines/processors/tensor_adapters/base.py
+++ b/outlines/processors/tensor_adapters/base.py
@@ -1,9 +1,13 @@
+"""Base class for tensor adapters."""
+
 from abc import ABC, abstractmethod
-from typing import Any, Union, TYPE_CHECKING
+from typing import TYPE_CHECKING, TypeVar, Any, Union
 
 if TYPE_CHECKING:
     import torch
 
+TensorType = TypeVar('TensorType')
+
 
 class TensorAdapter(ABC):
     """Abstract base class for tensor adapters.
@@ -11,78 +15,243 @@ class TensorAdapter(ABC):
     This class defines the interface for tensor adapters that are used to
     manipulate tensors in different libraries. Concrete implementations of
     this class should provide specific implementations for each method as
-    well as providing a library_name attribute.
+    well as providing a `library_name` attribute.
 
     TODO: Update the version of outlines-core used to receive plain arrays
     instead of torch tensors. In the meantime, implementations of this class
     must make sure that their `full_like` and `concatenate` methods can
     handle torch tensors.
+
     """
     library_name: str
 
     @abstractmethod
-    def shape(self, tensor):
-        """Get the shape of the tensor"""
+    def shape(self, tensor: TensorType) -> list[int]:
+        """Get the shape of the tensor.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to get the shape of.
+
+        Returns
+        -------
+        list[int]
+            The shape of the tensor. The list contains as many elements as
+            there are dimensions in the tensor.
+
+        """
         ...
 
     @abstractmethod
-    def unsqueeze(self, tensor):
-        """Add a dimension to the tensor at the specified position"""
+    def unsqueeze(self, tensor: TensorType) -> TensorType:
+        """Add a dimension to the tensor at axis 0.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to add a dimension to.
+
+        Returns
+        -------
+        TensorType
+            The tensor with an additional dimension.
+
+        """
         ...
 
     @abstractmethod
-    def squeeze(self, tensor):
-        """Remove a dimension from the tensor at the specified position"""
+    def squeeze(self, tensor: TensorType) -> TensorType:
+        """Remove a dimension from the tensor at axis 0.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to remove a dimension from.
+
+        Returns
+        -------
+        TensorType
+            The tensor with one less dimension.
+
+        """
         ...
 
     @abstractmethod
-    def to_list(self, tensor):
-        """Convert the tensor to a list"""
+    def to_list(self, tensor: TensorType) -> list:
+        """Convert the tensor to a list.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to convert to a list.
+
+        Returns
+        -------
+        list
+            The tensor as a list.
+
+        """
         ...
 
     @abstractmethod
-    def to_scalar(self, tensor):
-        """Convert the tensor to a scalar"""
+    def to_scalar(self, tensor: TensorType) -> Any:
+        """Return the only element of the tensor.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to return the only element of.
+
+        Returns
+        -------
+        Any
+            The only element of the tensor.
+
+        """
         ...
 
     @abstractmethod
-    def full_like(self, tensor: "torch.Tensor", fill_value):
+    def full_like(self, tensor: "torch.Tensor", fill_value: Any) -> TensorType: # type: ignore
         """Create a tensor with the same shape as the input tensor filled
         with a scalar value.
+
         ATTENTION: This method receives a torch tensor regardless of the
         library used.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to create a new tensor with the same shape.
+        fill_value
+            The value to fill the new tensor with.
+
+        Returns
+        -------
+        TensorType
+            A tensor with the same shape as the input tensor filled with the
+            specified value.
+
         """
         ...
 
     @abstractmethod
-    def concatenate(self, tensors: list[Union["torch.Tensor", Any]]):
-        """Concatenate a list of tensors along a specified dimension.
+    def concatenate(
+        self, tensors: list[Union["torch.Tensor", TensorType]]
+    ) -> TensorType:
+        """Concatenate a list of tensors along axis 0.
+
         ATTENTION: This method can either receive a list of torch tensors or
         a list of tensors from the library used.
+
+        Parameters
+        ----------
+        tensors
+            The list of tensors to concatenate.
+
+        Returns
+        -------
+        TensorType
+            The concatenated tensor.
+
         """
         ...
 
     @abstractmethod
-    def get_device(self, tensor):
-        """Get the device of the tensor"""
+    def get_device(self, tensor: TensorType) -> str:
+        """Get the name of the tensor's device.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to get the device of.
+
+        Returns
+        -------
+        str
+            The name of the tensor's device.
+
+        """
         ...
 
     @abstractmethod
-    def to_device(self, tensor, device):
-        """Move the tensor to a specified device"""
+    def to_device(self, tensor: TensorType, device: str) -> TensorType:
+        """Move the tensor to a specified device.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to move to a specified device.
+        device
+            The name of the device to move the tensor to.
+
+        Returns
+        -------
+        TensorType
+            The tensor moved to the specified device.
+
+        """
         ...
 
     @abstractmethod
-    def boolean_ones_like(self, tensor):
-        """Create a boolean ones tensor with the same shape as the input tensor"""
+    def boolean_ones_like(self, tensor: TensorType) -> TensorType:
+        """Create a boolean ones tensor with the same shape as the input
+        tensor.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to create a boolean ones tensor with the same shape.
+
+        Returns
+        -------
+        TensorType
+            A boolean ones tensor with the same shape as the input tensor.
+
+        """
         ...
 
     @abstractmethod
-    def apply_mask(self, tensor, mask, value):
-        """Fill the elements of the tensor where the mask is True with the specified value"""
+    def apply_mask(
+        self, tensor: TensorType, mask: TensorType, value: Any
+    ) -> TensorType:
+        """Fill the elements of the tensor where the mask is True with the
+        specified value.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to fill.
+        mask
+            The mask to apply to the tensor.
+        value
+            The value to fill the tensor with.
+
+        Returns
+        -------
+        TensorType
+            The tensor with the mask applied.
+
+        """
         ...
 
     @abstractmethod
-    def argsort_descending(self, tensor):
-        """Return the indices that would sort the tensor in descending order"""
+    def argsort_descending(
+        self, tensor: TensorType
+    ) -> TensorType:
+        """Return the indices that would sort the tensor in descending order
+        along axis -1.
+
+        Parameters
+        ----------
+        tensor
+            The tensor to sort.
+
+        Returns
+        -------
+        TensorType
+            The indices that would sort the tensor in descending order along
+            axis -1.
+
+        """
         ...
diff --git a/outlines/processors/tensor_adapters/jax.py b/outlines/processors/tensor_adapters/jax.py
index 73b4539a4..30cd2ecc9 100644
--- a/outlines/processors/tensor_adapters/jax.py
+++ b/outlines/processors/tensor_adapters/jax.py
@@ -1,3 +1,5 @@
+"""Tensor adapter for the `jax` library."""
+
 from outlines.processors.tensor_adapters.base import TensorAdapter
 
 
diff --git a/outlines/processors/tensor_adapters/mlx.py b/outlines/processors/tensor_adapters/mlx.py
index 0e9de15ef..5714808d4 100644
--- a/outlines/processors/tensor_adapters/mlx.py
+++ b/outlines/processors/tensor_adapters/mlx.py
@@ -1,3 +1,5 @@
+"""Tensor adapter for the `mlx` library."""
+
 from outlines.processors.tensor_adapters.base import TensorAdapter
 
 
diff --git a/outlines/processors/tensor_adapters/numpy.py b/outlines/processors/tensor_adapters/numpy.py
index 9bc2fb159..831220444 100644
--- a/outlines/processors/tensor_adapters/numpy.py
+++ b/outlines/processors/tensor_adapters/numpy.py
@@ -1,3 +1,5 @@
+"""Tensor adapter for the `numpy` library."""
+
 from outlines.processors.tensor_adapters.base import TensorAdapter
 
 
diff --git a/outlines/processors/tensor_adapters/tensorflow.py b/outlines/processors/tensor_adapters/tensorflow.py
index 4d2f93a6b..382f847b6 100644
--- a/outlines/processors/tensor_adapters/tensorflow.py
+++ b/outlines/processors/tensor_adapters/tensorflow.py
@@ -1,3 +1,5 @@
+"""Tensor adapter for the `tensorflow` library."""
+
 from outlines.processors.tensor_adapters.base import TensorAdapter
 
 
diff --git a/outlines/processors/tensor_adapters/torch.py b/outlines/processors/tensor_adapters/torch.py
index 6570b2f5f..4b624e960 100644
--- a/outlines/processors/tensor_adapters/torch.py
+++ b/outlines/processors/tensor_adapters/torch.py
@@ -1,3 +1,5 @@
+"""Tensor adapter for the `torch` library."""
+
 from outlines.processors.tensor_adapters.base import TensorAdapter
 
 
diff --git a/tests/processors/test_guide.py b/tests/processors/test_guide.py
index 3eaae2e32..0238e37d3 100644
--- a/tests/processors/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -34,7 +34,7 @@ class MockTokenizer:
 
     fsm = StopAtEOSGuide(MockTokenizer())
 
-    instruction = fsm.get_next_instruction(fsm.start_state)
+    instruction = fsm.get_next_instruction(fsm.initial_state)
     assert isinstance(instruction, Generate)
     assert instruction.tokens is None
 
@@ -42,9 +42,9 @@ class MockTokenizer:
     assert isinstance(instruction, Write)
     assert instruction.tokens == [2]
 
-    assert fsm.get_next_state(fsm.start_state, 2) == fsm.final_state
-    assert fsm.get_next_state(fsm.start_state, 1) == fsm.start_state
-    assert fsm.is_final_state(fsm.start_state) is False
+    assert fsm.get_next_state(fsm.initial_state, 2) == fsm.final_state
+    assert fsm.get_next_state(fsm.initial_state, 1) == fsm.initial_state
+    assert fsm.is_final_state(fsm.initial_state) is False
     assert fsm.is_final_state(fsm.final_state) is True
 
     assert isinstance(fsm.copy(), StopAtEOSGuide)

From 333f7050350bc910ca78c7601776165de70fbdd2 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 26 May 2025 21:52:34 +0200
Subject: [PATCH 335/505] Improve docstrings and type hinting in the types
 module

---
 outlines/types/__init__.py        |  3 +-
 outlines/types/dsl.py             | 90 ++++++++++++++++++++++++++++---
 outlines/types/locale/__init__.py |  2 +
 outlines/types/locale/us.py       |  2 +
 outlines/types/utils.py           |  2 +
 5 files changed, 92 insertions(+), 7 deletions(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index d653f2a06..8673b4e47 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,4 +1,5 @@
-import json
+"""Output types for structured generation and regex DSL."""
+
 from dataclasses import dataclass, is_dataclass
 from enum import Enum, EnumMeta
 from typing import Union
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 66a0d99df..14b0c8c60 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -1,3 +1,17 @@
+"""Regular expression DSL and output types for structured generation.
+
+This module contains elements related to three logical steps in the use of
+output types for structured generation:
+
+1. Definition of `Term` classes that contain output type definitions. That
+   includes both terms intended to be used by themselves such as `JsonSchema`
+   or `CFG` and terms that are part of the regular expression DSL such as
+   `Alternatives` or `KleeneStar` (and the related functions).
+2. Conversion of Python types into `Term` instances (`python_types_to_terms`).
+3. Conversion of a `Term` instance into a regular expression (`to_regex`).
+
+"""
+
 import json
 import re
 import sys
@@ -200,6 +214,14 @@ def __repr__(self):
 
 @dataclass
 class Regex(Term):
+    """Class representing a regular expression.
+
+    Parameters
+    ----------
+    pattern
+        The regular expression as a string.
+
+    """
     pattern: str
 
     def _display_node(self) -> str:
@@ -211,6 +233,14 @@ def __repr__(self):
 
 @dataclass
 class CFG(Term):
+    """Class representing a context-free grammar.
+
+    Parameters
+    ----------
+    definition
+        The definition of the context-free grammar as a string.
+
+    """
     definition: str
 
     def _display_node(self) -> str:
@@ -225,13 +255,18 @@ def __eq__(self, other):
         return self.definition == other.definition
 
     @classmethod
-    def from_file(cls, path: str):
+    def from_file(cls, path: str) -> "CFG":
         """Create a CFG instance from a file containing a CFG definition.
 
         Parameters
         ----------
         path : str
             The path to the file containing the CFG definition.
+        Returns
+        -------
+        CFG
+            A CFG instance.
+
         """
         with open(path, "r") as f:
             definition = f.read()
@@ -240,6 +275,15 @@ def from_file(cls, path: str):
 
 @dataclass
 class FSM(Term):
+    """Class representing a finite state machine.
+
+    Parameters
+    ----------
+    fsm
+        The finite state machine to store. This object must be an instance of
+        `interegular.fsm.FSM`.
+
+    """
     fsm: interegular.fsm.FSM
 
     def _display_node(self) -> str:
@@ -250,6 +294,13 @@ def __repr__(self):
 
 
 class JsonSchema(Term):
+    """Class representing a JSON schema.
+
+    The JSON schema object from which to instantiate the class can be a
+    dictionary, a string, a Pydantic model, a typed dict, a dataclass, or a
+    genSON schema builder.
+
+    """
     def __init__(
         self,
         schema: Union[
@@ -257,6 +308,15 @@ def __init__(
         ],
         whitespace_pattern: OptionalType[str] = None,
     ):
+        """
+        Parameters
+        ----------
+        schema
+            The object containing the JSON schema.
+        whitespace_pattern
+            The pattern to use to match whitespace characters.
+
+        """
         schema_str: str
 
         if is_dict_instance(schema):
@@ -302,14 +362,19 @@ def __eq__(self, other):
             return self.schema == other.schema
 
     @classmethod
-    def from_file(cls, path: str):
+    def from_file(cls, path: str) -> "JsonSchema":
         """Create a JsonSchema instance from a .json file containing a JSON
         schema.
 
         Parameters
         ----------
-        path : str
+        path:
             The path to the file containing the JSON schema.
+        Returns
+        -------
+        JsonSchema
+            A JsonSchema instance.
+
         """
         with open(path, "r") as f:
             schema = json.load(f)
@@ -484,6 +549,7 @@ def either(*terms: Union[str, Term]):
 
     This factory function automatically translates string arguments
     into `String` objects.
+
     """
     terms = [String(arg) if isinstance(arg, str) else arg for arg in terms]
     return Alternatives(terms)
@@ -530,8 +596,8 @@ def one_or_more(term: Union[Term, str]) -> KleenePlus:
 def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
     """Convert Python types to Outlines DSL terms that constrain LLM output.
 
-    Arguments
-    ---------
+    Parameters
+    ----------
     ptype
         The Python type to convert
     recursion_depth
@@ -539,7 +605,9 @@ def python_types_to_terms(ptype: Any, recursion_depth: int = 0) -> Term:
 
     Returns
     -------
-    The corresponding DSL `Term` instance.
+    Term
+        The corresponding DSL `Term` instance.
+
     """
     if recursion_depth > 10:
         raise RecursionError(
@@ -722,6 +790,16 @@ def to_regex(term: Term) -> str:
 
     We only consider self-contained terms that do not refer to another rule.
 
+    Parameters
+    ----------
+    term
+        The term to convert to a regular expression.
+
+    Returns
+    -------
+    str
+        The regular expression as a string.
+
     """
     match term:
         case String():
diff --git a/outlines/types/locale/__init__.py b/outlines/types/locale/__init__.py
index 511631d84..97c9abff0 100644
--- a/outlines/types/locale/__init__.py
+++ b/outlines/types/locale/__init__.py
@@ -1 +1,3 @@
+"""Locale-specific regex patterns."""
+
 from . import us
diff --git a/outlines/types/locale/us.py b/outlines/types/locale/us.py
index 0bda82d44..275aa492e 100644
--- a/outlines/types/locale/us.py
+++ b/outlines/types/locale/us.py
@@ -1,3 +1,5 @@
+"""Locale-specific regex patterns for the United States."""
+
 from outlines.types.dsl import Regex
 
 zip_code = Regex(r"\d{5}(?:-\d{4})?")
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
index f46b516cc..64537643e 100644
--- a/outlines/types/utils.py
+++ b/outlines/types/utils.py
@@ -1,3 +1,5 @@
+"""Utility functions for the types module."""
+
 import dataclasses
 import datetime
 import inspect

From d83c1bf2a1f51f34947c9c5a69e8fc4975624651 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 27 May 2025 10:21:35 +0200
Subject: [PATCH 336/505] Improve docstrings and type hinting in model
 implementations

---
 outlines/models/anthropic.py    | 104 +++++++++++++---
 outlines/models/dottxt.py       | 100 +++++++++++++--
 outlines/models/gemini.py       | 117 ++++++++++++++++--
 outlines/models/llamacpp.py     | 123 ++++++++++++++-----
 outlines/models/mlxlm.py        | 116 +++++++++++++-----
 outlines/models/ollama.py       | 111 ++++++++++++++---
 outlines/models/openai.py       | 127 ++++++++++++++++---
 outlines/models/sglang.py       | 203 +++++++++++++++++++++++--------
 outlines/models/tgi.py          | 181 ++++++++++++++++++++-------
 outlines/models/transformers.py | 174 +++++++++++++++++++++-----
 outlines/models/vllm.py         | 209 +++++++++++++++++++++++---------
 outlines/models/vllm_offline.py |  93 ++++++++++----
 12 files changed, 1312 insertions(+), 346 deletions(-)

diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 6a3a01a81..4ed3828d9 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -1,6 +1,6 @@
 """Integration with Anthropic's API."""
 
-from typing import Optional, Union, TYPE_CHECKING
+from typing import TYPE_CHECKING, Any, Iterator, Optional, Union
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
@@ -12,23 +12,28 @@
 
 
 class AnthropicTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the Anthropic clients.
+    """Type adapter for the `Anthropic` model.
 
     `AnthropicTypeAdapter` is responsible for preparing the arguments to
-    Anthropic's `messages.create` methods: the input (prompt and possibly
+    Anthropic's `messages.create` method: the input (prompt and possibly
     image).
     Anthropic does not support defining the output type, so
     `format_output_type` is not implemented.
 
     """
 
-    def format_input(self, model_input):
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
         """Generate the `messages` argument to pass to the client.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
-            The input passed by the user.
+            The input provided by the user.
+
+        Returns
+        -------
+        dict
+            The `messages` argument to pass to the client.
 
         """
         if isinstance(model_input, str):
@@ -40,7 +45,7 @@ def format_input(self, model_input):
             "The only available types are `str` and `Vision`."
         )
 
-    def format_str_model_input(self, model_input):
+    def format_str_model_input(self, model_input: str) -> dict:
         return {
             "messages": [
                 {
@@ -50,7 +55,7 @@ def format_str_model_input(self, model_input):
             ]
         }
 
-    def format_vision_model_input(self, model_input):
+    def format_vision_model_input(self, model_input: Vision) -> dict:
         return {
             "messages": [
                 {
@@ -82,9 +87,24 @@ def format_output_type(self, output_type):
 
 
 class Anthropic(Model):
+    """Thin wrapper around the `anthropic.Anthropic` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `anthropic.Anthropic` client.
+
+    """
     def __init__(
         self, client: "AnthropicClient", model_name: Optional[str] = None
     ):
+        """
+        Parameters
+        ----------
+        client
+            An `anthropic.Anthropic` client.
+        model_name
+            The name of the model to use.
+
+        """
         self.client = client
         self.model_name = model_name
         self.type_adapter = AnthropicTypeAdapter()
@@ -92,9 +112,28 @@ def __init__(
     def generate(
         self,
         model_input: Union[str, Vision],
-        output_type=None,
-        **inference_kwargs,
-    ):
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
+        """Generate text using Anthropic.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            As structured generation is not supported by Anthropic, the value
+            of this argument must be `None`. Otherwise, an error will be
+            raised at runtime.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The response generated by the model.
+
+        """
         messages = self.type_adapter.format_input(model_input)
 
         if output_type is not None:
@@ -117,9 +156,28 @@ def generate(
     def generate_stream(
         self,
         model_input: Union[str, Vision],
-        output_type=None,
-        **inference_kwargs,
-    ):
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using Anthropic.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            As structured generation is not supported by Anthropic, the value
+            of this argument must be `None`. Otherwise, an error will be
+            raised at runtime.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
         messages = self.type_adapter.format_input(model_input)
 
         if output_type is not None:
@@ -150,4 +208,20 @@ def generate_stream(
 def from_anthropic(
     client: "AnthropicClient", model_name: Optional[str] = None
 ) -> Anthropic:
+    """Create an Outlines `Anthropic` model instance from an
+    `anthropic.Anthropic` client instance.
+
+    Parameters
+    ----------
+    client
+        An `anthropic.Anthropic` client instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Anthropic
+        An Outlines `Anthropic` model instance.
+
+    """
     return Anthropic(client, model_name)
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index eef41b83b..62c40a7c5 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -1,7 +1,7 @@
 """Integration with Dottxt's API."""
 
 import json
-from typing import Optional, TYPE_CHECKING
+from typing import TYPE_CHECKING, Any, Optional
 
 from pydantic import TypeAdapter
 
@@ -17,17 +17,24 @@
 if TYPE_CHECKING:
     from dottxt import Dottxt as DottxtClient
 
-__all__ = ["Dottxt"]
+__all__ = ["Dottxt", "from_dottxt"]
 
 
 class DottxtTypeAdapter(ModelTypeAdapter):
-    def format_input(self, model_input):
-        """Generate the `messages` argument to pass to the client.
+    """Type adapter for the `Dottxt` model."""
 
-        Argument
-        --------
+    def format_input(self, model_input: str) -> str:
+        """Format the prompt to pass to the client.
+
+        Parameters
+        ----------
         model_input
-            The input passed by the user.
+            The input provided by the user.
+
+        Returns
+        -------
+        str
+            The input to pass to the client.
 
         """
         if isinstance(model_input, str):
@@ -37,13 +44,23 @@ def format_input(self, model_input):
             "The only available type is `str`."
         )
 
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> str:
         """Format the output type to pass to the client.
 
         TODO: `int`, `float` and other Python types could be supported via
         JSON Schema.
-        """
 
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        str
+            The output type to pass to the client.
+
+        """
         # Unsupported languages
         if output_type is None:
             raise TypeError(
@@ -92,16 +109,51 @@ class Dottxt(Model):
 
     def __init__(
         self,
-        client: "Dottxt",
+        client: "DottxtClient",
         model_name: Optional[str] = None,
         model_revision: Optional[str] = None,
     ):
+        """
+        Parameters
+        ----------
+        client
+            A `dottxt.Dottxt` client.
+        model_name
+            The name of the model to use.
+        model_revision
+            The revision of the model to use.
+
+        """
         self.client = client
         self.model_name = model_name
         self.model_revision = model_revision
         self.type_adapter = DottxtTypeAdapter()
 
-    def generate(self, model_input, output_type=None, **inference_kwargs):
+    def generate(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
+        """Generate text using Dottxt.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
+
+        """
         prompt = self.type_adapter.format_input(model_input)
         json_schema = self.type_adapter.format_output_type(output_type)
 
@@ -125,8 +177,12 @@ def generate(self, model_input, output_type=None, **inference_kwargs):
         return completion.data
 
     def generate_stream(
-        self, model_input, output_type=None, **inference_kwargs
+        self,
+        model_input,
+        output_type=None,
+        **inference_kwargs,
     ):
+        """Not available for Dottxt."""
         raise NotImplementedError(
             "Dottxt does not support streaming. Call the model/generator for "
             + "regular generation instead."
@@ -137,5 +193,23 @@ def from_dottxt(
     client: "DottxtClient",
     model_name: Optional[str] = None,
     model_revision: Optional[str] = None,
-):
+) -> Dottxt:
+    """Create an Outlines `Dottxt` model instance from a `dottxt.Dottxt`
+    client instance.
+
+    Parameters
+    ----------
+    client
+        A `dottxt.Dottxt` client instance.
+    model_name
+        The name of the model to use.
+    model_revision
+        The revision of the model to use.
+
+    Returns
+    -------
+    Dottxt
+        An Outlines `Dottxt` model instance.
+
+    """
     return Dottxt(client, model_name, model_revision)
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index e2f197c75..c66c984af 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -1,9 +1,10 @@
 """Integration with Gemini's API."""
 
 from typing import (
+    TYPE_CHECKING,
     Any,
+    Iterator,
     Optional,
-    TYPE_CHECKING,
     Union,
     get_args,
 )
@@ -29,7 +30,7 @@
 
 
 class GeminiTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the Gemini client.
+    """Type adapter for the `Gemini` model.
 
     `GeminiTypeAdapter` is responsible for preparing the arguments to Gemini's
     client `models.generate_content` method: the input (prompt and possibly
@@ -37,13 +38,18 @@ class GeminiTypeAdapter(ModelTypeAdapter):
 
     """
 
-    def format_input(self, model_input):
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
         """Generate the `contents` argument to pass to the client.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
-            The input passed by the user.
+            The input provided by the user.
+
+        Returns
+        -------
+        dict
+            The `contents` argument to pass to the client.
 
         """
         if isinstance(model_input, str):
@@ -62,7 +68,21 @@ def format_input(self, model_input):
                 "The only available types are `str` and `Vision`."
             )
 
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the `generation_config` argument to pass to the client.
+
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        dict
+            The `generation_config` argument to pass to the client.
+
+        """
+
         # Unsupported output pytes
         if isinstance(output_type, Regex):
             raise TypeError(
@@ -116,19 +136,19 @@ def format_output_type(self, output_type):
                 "Consider using a local model or dottxt instead."
             )
 
-    def format_enum_output_type(self, output_type):
+    def format_enum_output_type(self, output_type: Optional[Any]) -> dict:
         return {
             "response_mime_type": "text/x.enum",
             "response_schema": output_type,
         }
 
-    def format_json_output_type(self, output_type):
+    def format_json_output_type(self, output_type: Optional[Any]) -> dict:
         return {
             "response_mime_type": "application/json",
             "response_schema": output_type,
         }
 
-    def format_list_output_type(self, output_type):
+    def format_list_output_type(self, output_type: Optional[Any]) -> dict:
         args = get_args(output_type)
 
         if len(args) == 1:
@@ -159,7 +179,24 @@ def format_list_output_type(self, output_type):
 
 
 class Gemini(Model):
+    """Thin wrapper around the `google.genai.Client` client.
+
+    This wrapper is used to convert the input and output types specified by
+    the users at a higher level to arguments to the `google.genai.Client`
+    client.
+
+    """
+
     def __init__(self, client: "Client", model_name: Optional[str] = None):
+        """
+        Parameters
+        ----------
+        client
+            A `google.genai.Client` instance.
+        model_name
+            The name of the model to use.
+
+        """
         self.client = client
         self.model_name = model_name
         self.type_adapter = GeminiTypeAdapter()
@@ -169,7 +206,26 @@ def generate(
         model_input: Union[str, Vision],
         output_type: Optional[Any] = None,
         **inference_kwargs,
-    ):
+    ) -> str:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema, a list of such types, or a multiple choice type.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The response generated by the model.
+
+        """
         contents = self.type_adapter.format_input(model_input)
         generation_config = self.type_adapter.format_output_type(output_type)
 
@@ -186,7 +242,26 @@ def generate_stream(
         model_input: Union[str, Vision],
         output_type: Optional[Any] = None,
         **inference_kwargs,
-    ):
+    ) -> Iterator[str]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema, a list of such types, or a multiple choice type.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
         contents = self.type_adapter.format_input(model_input)
         generation_config = self.type_adapter.format_output_type(output_type)
 
@@ -201,5 +276,21 @@ def generate_stream(
                 yield chunk.text
 
 
-def from_gemini(client: "Client", model_name: Optional[str] = None):
+def from_gemini(client: "Client", model_name: Optional[str] = None) -> Gemini:
+    """Create an Outlines `Gemini` model instance from a
+    `google.genai.Client` instance.
+
+    Parameters
+    ----------
+    client
+        A `google.genai.Client` instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Gemini
+        An Outlines `Gemini` model instance.
+
+    """
     return Gemini(client, model_name)
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 93aefc2e8..92d5fe15f 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,10 +1,14 @@
+"""Integration with the `llama-cpp-python` library."""
+
 import warnings
 from functools import singledispatchmethod
 from typing import (
     TYPE_CHECKING,
+    Any,
     Dict,
     Iterator,
     List,
+    Optional,
     Set,
     Tuple,
     Union,
@@ -12,10 +16,12 @@
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
-from outlines.processors import CFGLogitsProcessor
+from outlines.processors import CFGLogitsProcessor, OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
-    from llama_cpp import Llama
+    from llama_cpp import Llama, LogitsProcessorList
+
+__all__ = ["LlamaCpp", "from_llamacpp"]
 
 
 class LlamaCppTokenizer(Tokenizer):
@@ -121,7 +127,7 @@ def __setstate__(self, state):
 
 
 class LlamaCppTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the `llama-cpp-python` library.
+    """Type adapter for the `LlamaCpp` model.
 
     `LlamaCppTypeAdapter` is responsible for preparing the arguments to
     `llama-cpp-python`'s `Llama.__call__` method: the input (a string prompt),
@@ -133,10 +139,15 @@ class LlamaCppTypeAdapter(ModelTypeAdapter):
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
-            The input passed by the user.
+            The input provided by the user.
+
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
 
         """
         raise NotImplementedError(
@@ -145,17 +156,24 @@ def format_input(self, model_input):
         )
 
     @format_input.register(str)
-    def format_str_input(self, model_input: str):
+    def format_str_input(self, model_input: str) -> str:
         return model_input
 
-    def format_output_type(self, output_type):
+    def format_output_type(
+        self, output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> "LogitsProcessorList":
         """Generate the logits processor argument to pass to the model.
 
-        Argument
-        --------
+        Parameters
+        ----------
         output_type
             The logits processor provided.
 
+        Returns
+        -------
+        LogitsProcessorList
+            The logits processor to pass to the model.
+
         """
         from llama_cpp import LogitsProcessorList
 
@@ -163,30 +181,49 @@ def format_output_type(self, output_type):
 
 
 class LlamaCpp(Model):
-    """Wraps a model provided by the `llama-cpp-python` library."""
+    """Thin wrapper around the `llama_cpp.Llama` model.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `llama_cpp.Llama` model.
+    """
+
     tensor_library_name = "numpy"
 
     def __init__(self, model: "Llama"):
+        """
+        Parameters
+        ----------
+        model
+            A `llama_cpp.Llama` model instance.
+
+        """
         self.model = model
         self.tokenizer = LlamaCppTokenizer(self.model)
         self.type_adapter = LlamaCppTypeAdapter()
 
-    def generate(self, model_input, output_type, **inference_kwargs):
+    def generate(
+        self,
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
+    ) -> str:
         """Generate text using `llama-cpp-python`.
 
-        Arguments
-        ---------
-        prompt
-            A prompt.
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
         output_type
-            The logits processor to use when generating text.
-        inference_kwargs
-            The inference kwargs that can be passed to the `Llama.__call__`
-            method in the `llama-cpp-python` library.
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        **inference_kwargs
+            Additional keyword arguments to pass to the `Llama.__call__`
+            method of the `llama-cpp-python` library.
 
         Returns
         -------
-        The generated text.
+        str
+            The text generated by the model.
 
         """
         if isinstance(output_type, CFGLogitsProcessor):
@@ -207,23 +244,28 @@ def generate(self, model_input, output_type, **inference_kwargs):
         return result
 
     def generate_stream(
-        self, model_input, output_type, **inference_kwargs
-    ):
+        self,
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
         """Stream text using `llama-cpp-python`.
 
-        Arguments
-        ---------
-        prompt
-            A prompt.
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
         output_type
-            The logits processor to use when generating text.
-        inference_kwargs
-            The inference kwargs that can be passed to the `Llama.__call__`
-            method in the `llama-cpp-python` library.
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        **inference_kwargs
+            Additional keyword arguments to pass to the `Llama.__call__`
+            method of the `llama-cpp-python` library.
 
         Returns
         -------
-        A generator that return strings.
+        Iterator[str]
+            An iterator that yields the text generated by the model.
 
         """
         if isinstance(output_type, CFGLogitsProcessor):
@@ -250,7 +292,8 @@ def token_generator() -> Iterator[str]:
 
         return token_generator()
 
-    def load_lora(self, adapter_path: str):  # pragma: no cover
+    def load_lora(self, adapter_path: str) -> None:  # pragma: no cover
+        """Load a LoRA adapter. Deprecated since v1.0.0."""
         warnings.warn("""
             The `load_lora` method is deprecated starting from v1.0.0.
             Support for it will be removed in v1.1.0.
@@ -268,4 +311,18 @@ def load_lora(self, adapter_path: str):  # pragma: no cover
 
 
 def from_llamacpp(model: "Llama"):
+    """Create an Outlines `LlamaCpp` model instance from a
+    `llama_cpp.Llama` instance.
+
+    Parameters
+    ----------
+    model
+        A `llama_cpp.Llama` instance.
+
+    Returns
+    -------
+    LlamaCpp
+        An Outlines `LlamaCpp` model instance.
+
+    """
     return LlamaCpp(model)
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 238e16ffc..bea4ad293 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -1,28 +1,35 @@
+"""Integration with the `mlx_lm` library."""
+
 from functools import singledispatchmethod
-from typing import TYPE_CHECKING
+from typing import TYPE_CHECKING, Iterator, List, Optional
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.transformers import TransformerTokenizer
+from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     import mlx.nn as nn
     from transformers import PreTrainedTokenizer
 
-
 __all__ = ["MLXLM", "from_mlxlm"]
 
 
 class MLXLMTypeAdapter(ModelTypeAdapter):
-    """Type adapter for `mlx_lm` models."""
+    """Type adapter for the `MLXLM` model."""
 
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
-            The input passed by the user.
+            The input provided by the user.
+
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
 
         """
         raise NotImplementedError(
@@ -34,14 +41,21 @@ def format_input(self, model_input):
     def format_str_input(self, model_input: str):
         return model_input
 
-    def format_output_type(self, output_type):
+    def format_output_type(
+        self, output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Optional[List[OutlinesLogitsProcessor]]:
         """Generate the logits processor argument to pass to the model.
 
-        Argument
-        --------
+        Parameters
+        ----------
         output_type
             The logits processor provided.
 
+        Returns
+        -------
+        Optional[list[OutlinesLogitsProcessor]]
+            The logits processor argument to be passed to the model.
+
         """
         if not output_type:
             return None
@@ -49,7 +63,13 @@ def format_output_type(self, output_type):
 
 
 class MLXLM(Model):
-    """Represents an `mlx_lm` model."""
+    """Thin wrapper around an `mlx_lm` model.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `mlx_lm` library.
+
+    """
+
     tensor_library_name = "mlx"
 
     def __init__(
@@ -57,15 +77,14 @@ def __init__(
         model: "nn.Module",
         tokenizer: "PreTrainedTokenizer",
     ):
-        """Create a MLXLM model instance.
-
-        Arguments
-        ---------
+        """
+        Parameters
+        ----------
         model
-            An instance of an mlx-lm model.
+            An instance of an `mlx_lm` model.
         tokenizer
-            An instance of an mlx-lm tokenizer or of a compatible
-            transformers tokenizer.
+            An instance of an `mlx_lm` tokenizer or of a compatible
+            `transformers` tokenizer.
 
         """
         self.model = model
@@ -75,17 +94,28 @@ def __init__(
         self.tokenizer = TransformerTokenizer(tokenizer._tokenizer)
         self.type_adapter = MLXLMTypeAdapter()
 
-    def generate(self, model_input, output_type=None, **kwargs):
+    def generate(
+        self,
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **kwargs,
+    ) -> str:
         """Generate text using `mlx-lm`.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The prompt passed by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The logits processor provided.
+            The logits processor the model will use to constrain the format of
+            the generated text.
         kwargs
-            The inference kwargs that can be passed to the `mlx-lm` library.
+            Additional keyword arguments to pass to the `mlx-lm` library.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
 
         """
         from mlx_lm import generate
@@ -98,17 +128,28 @@ def generate(self, model_input, output_type=None, **kwargs):
             **kwargs,
         )
 
-    def generate_stream(self, model_input, output_type=None, **kwargs):
+    def generate_stream(
+        self,
+        model_input: str,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **kwargs,
+    ) -> Iterator[str]:
         """Stream text using `mlx-lm`.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The prompt passed by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The logits processor provided.
+            The logits processor the model will use to constrain the format of
+            the generated text.
         kwargs
-            The inference kwargs that can be passed to the `mlx-lm` library.
+            Additional keyword arguments to pass to the `mlx-lm` library.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
 
         """
         from mlx_lm import stream_generate
@@ -124,4 +165,21 @@ def generate_stream(self, model_input, output_type=None, **kwargs):
 
 
 def from_mlxlm(model: "nn.Module", tokenizer: "PreTrainedTokenizer") -> MLXLM:
+    """Create an Outlines `MLXLM` model instance from an `mlx_lm` model and a
+    tokenizer.
+
+    Parameters
+    ----------
+    model
+        An instance of an `mlx_lm` model.
+    tokenizer
+        An instance of an `mlx_lm` tokenizer or of a compatible
+        transformers tokenizer.
+
+    Returns
+    -------
+    MLXLM
+        An Outlines `MLXLM` model instance.
+
+    """
     return MLXLM(model, tokenizer)
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index c2495351b..862c71f69 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,5 +1,7 @@
+"""Integration with the `ollama` library."""
+
 import json
-from typing import Iterator, Optional, TYPE_CHECKING
+from typing import TYPE_CHECKING, Any, Iterator, Optional
 
 from pydantic import TypeAdapter
 
@@ -12,24 +14,27 @@
     is_typed_dict,
 )
 
-
 if TYPE_CHECKING:
     from ollama import Client as OllamaClient
 
-
 __all__ = ["Ollama", "from_ollama"]
 
 
 class OllamaTypeAdapter(ModelTypeAdapter):
-    """Type adapter for the Ollama model."""
+    """Type adapter for the `Ollama` model."""
 
-    def format_input(self, model_input):
+    def format_input(self, model_input: str) -> str:
         """Generate the prompt argument to pass to the model.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
-            The input passed by the user.
+            The input provided by the user.
+
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
 
         """
         if isinstance(model_input, str):
@@ -39,13 +44,23 @@ def format_input(self, model_input):
             "Ollama does not support batch inference."
         )
 
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> Optional[str]:
         """Format the output type to pass to the client.
 
         TODO: `int`, `float` and other Python types could be supported via
         JSON Schema.
-        """
 
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        Optional[str]
+            The formatted output type to be passed to the model.
+
+        """
         if isinstance(output_type, Regex):
             raise TypeError(
                 "Regex-based structured outputs are not supported by Ollama. "
@@ -81,22 +96,53 @@ def format_output_type(self, output_type):
 
 
 class Ollama(Model):
-    """Thin wrapper around the `ollama` client.
+    """Thin wrapper around the `ollama.Client` client.
 
     This wrapper is used to convert the input and output types specified by the
-    users at a higher level to arguments to the `ollama` client.
+    users at a higher level to arguments to the `ollama.Client` client.
 
     """
 
     def __init__(
         self,client: "OllamaClient", model_name: Optional[str] = None,
     ):
+        """
+        Parameters
+        ----------
+        client
+            The `ollama.Client` client.
+        model_name
+            The name of the model to use.
+
+        """
         self.client = client
         self.model_name = model_name
         self.type_adapter = OllamaTypeAdapter()
 
-    def generate(self, model_input, output_type=None, **kwargs) -> str:
+    def generate(self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **kwargs: Any,
+    ) -> str:
+        """Generate text using Ollama.
 
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
+
+        """
         if "model" not in kwargs and self.model_name is not None:
             kwargs["model"] = self.model_name
 
@@ -108,9 +154,30 @@ def generate(self, model_input, output_type=None, **kwargs) -> str:
         return response.response
 
     def generate_stream(
-        self, model_input, output_type=None, **kwargs
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **kwargs: Any,
     ) -> Iterator[str]:
+        """Stream text using Ollama.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
 
+        """
         if "model" not in kwargs and self.model_name is not None:
             kwargs["model"] = self.model_name
 
@@ -127,4 +194,20 @@ def generate_stream(
 def from_ollama(
     client: "OllamaClient", model_name: Optional[str] = None
 ) -> Ollama:
+    """Create an Outlines `Ollama` model instance from an `ollama.Client`
+    client.
+
+    Parameters
+    ----------
+    client
+        A `ollama.Client` client instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Ollama
+        An Outlines `Ollama` model instance.
+
+    """
     return Ollama(client, model_name)
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 540b04bac..2112e1974 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -4,7 +4,16 @@
 import functools
 import json
 import warnings
-from typing import Callable, Dict, List, Optional, Union, TYPE_CHECKING
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    Callable,
+    Dict,
+    Iterator,
+    List,
+    Optional,
+    Union,
+)
 from dataclasses import asdict, dataclass, field, replace
 
 from pydantic import BaseModel, TypeAdapter
@@ -22,7 +31,6 @@
 )
 from outlines.v0_legacy.base import vectorize
 
-
 if TYPE_CHECKING:
     from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
 
@@ -30,7 +38,7 @@
 
 
 class OpenAITypeAdapter(ModelTypeAdapter):
-    """Type adapter for the OpenAI clients.
+    """Type adapter for the `OpenAI` model.
 
     `OpenAITypeAdapter` is responsible for preparing the arguments to OpenAI's
     `completions.create` methods: the input (prompt and possibly image), as
@@ -38,13 +46,18 @@ class OpenAITypeAdapter(ModelTypeAdapter):
 
     """
 
-    def format_input(self, model_input):
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
         """Generate the `messages` argument to pass to the client.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
-            The input passed by the user.
+            The input provided by the user.
+
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the client.
 
         """
         if isinstance(model_input, str):
@@ -56,7 +69,7 @@ def format_input(self, model_input):
             "The only available types are `str` and `Vision`."
         )
 
-    def format_str_model_input(self, model_input: str):
+    def format_str_model_input(self, model_input: str) -> dict:
         """Generate the `messages` argument to pass to the client when the user
         only passes a prompt.
 
@@ -70,7 +83,7 @@ def format_str_model_input(self, model_input: str):
             ]
         }
 
-    def format_vision_model_input(self, model_input: Vision):
+    def format_vision_model_input(self, model_input: Vision) -> dict:
         """Generate the `messages` argument to pass to the client when the user
         passes a prompt and an image.
 
@@ -92,15 +105,24 @@ def format_vision_model_input(self, model_input: Vision):
             ]
         }
 
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the `response_format` argument to the client based on the
         output type specified by the user.
 
         TODO: `int`, `float` and other Python types could be supported via
         JSON Schema.
 
-        """
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        dict
+            The formatted output type to be passed to the client.
 
+        """
         # Unsupported languages
         if isinstance(output_type, Regex):
             raise TypeError(
@@ -139,12 +161,11 @@ def format_output_type(self, output_type):
                 "Use an open source model or dottxt instead."
             )
 
-    def format_json_output_type(self, schema: dict):
+    def format_json_output_type(self, schema: dict) -> dict:
         """Generate the `response_format` argument to the client when the user
         specified a `Json` output type.
 
         """
-
         # OpenAI requires `additionalProperties` to be set
         if "additionalProperties" not in schema:
             schema["additionalProperties"] = False
@@ -160,7 +181,7 @@ def format_json_output_type(self, schema: dict):
             }
         }
 
-    def format_json_mode_type(self):
+    def format_json_mode_type(self) -> dict:
         """Generate the `response_format` argument to the client when the user
         specified the output type should be a JSON but without specifying the
         schema (also called "JSON mode").
@@ -183,6 +204,22 @@ def __init__(
         model_name: Optional[Union[str, "OpenAIConfig"]] = None,
         **kwargs
     ):
+        """Initialize the OpenAI model.
+
+        To provide temporary backwards compatibility with Outlines v0,
+        the class can be instantiated with a `OpenAIConfig` instance as
+        a value for the `model_name` argument. This is deprecated and will
+        be removed in v1.1.0. Please provide a model name instead.
+
+        Parameters
+        ----------
+        client
+            The `openai.OpenAI` client.
+        model_name
+            The name of the model to use.
+
+        """
+
         # legacy mode
         if isinstance(model_name, OpenAIConfig) or kwargs.get("config"):
             warnings.warn("""
@@ -223,8 +260,27 @@ def generate(
         self,
         model_input: Union[str, Vision],
         output_type: Optional[Union[type[BaseModel], str]] = None,
-        **inference_kwargs,
-    ):
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using OpenAI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema or an empty dictionary.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The text generated by the model.
+
+        """
         import openai
 
         messages = self.type_adapter.format_input(model_input)
@@ -265,7 +321,28 @@ def generate_stream(
         model_input: Union[str, Vision],
         output_type: Optional[Union[type[BaseModel], str]] = None,
         **inference_kwargs,
-    ):
+    ) -> Iterator[str]:
+        """Stream text using OpenAI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema or an empty dictionary.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        import openai
+
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
 
@@ -318,6 +395,22 @@ def from_openai(
     client: Union["OpenAIClient", "AzureOpenAIClient"],
     model_name: Optional[str] = None,
 ) -> OpenAI:
+    """Create an Outlines `OpenAI` model instance from an `openai.OpenAI`
+    client.
+
+    Parameters
+    ----------
+    client
+        An `openai.OpenAI` client instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    OpenAI
+        An Outlines `OpenAI` model instance.
+
+    """
     return OpenAI(client, model_name)
 
 
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
index 8a4d90ffa..0c2a51da1 100644
--- a/outlines/models/sglang.py
+++ b/outlines/models/sglang.py
@@ -1,40 +1,60 @@
+"""Integration with an SGLang server."""
+
 import json
 import warnings
-from typing import TYPE_CHECKING, Union, Optional
+from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
 
-from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
+from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.models.openai import OpenAITypeAdapter
-from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
-
+from outlines.types.dsl import (
+    CFG,
+    JsonSchema,
+    python_types_to_terms,
+    to_regex,
+)
+from outlines.templates import Vision
 
 if TYPE_CHECKING:
     from openai import AsyncOpenAI, OpenAI
 
+__all__ = ["AsyncSGLang", "SGLang", "from_sglang"]
+
 
 class SGLangTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `SGLang` and `AsyncSGLang` models."""
 
-    def format_input(self, model_input):
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
         """Generate the prompt argument to pass to the client.
 
         We rely on the OpenAITypeAdapter to format the input as the sglang
         server expects input in the same format as OpenAI.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
             The input passed by the user.
 
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the client.
+
         """
         return OpenAITypeAdapter().format_input(model_input)
 
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the structured output argument to pass to the client.
 
-        Argument
-        --------
+        Parameters
+        ----------
         output_type
             The structured output type provided.
 
+        Returns
+        -------
+        dict
+            The formatted output type to be passed to the client.
+
         """
         if output_type is None:
             return {}
@@ -58,36 +78,52 @@ def format_output_type(self, output_type):
 
 
 class SGLang(Model):
-    """Represents a client to a `sglang` server."""
+    """Thin wrapper around the `openai.OpenAI` client used to communicate with
+    an SGLang server.
 
-    def __init__(self, client, model_name: Optional[str] = None):
-        """Create a `SGLang` model instance.
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    SGLang server.
+
+    """
 
+    def __init__(self, client, model_name: Optional[str] = None):
+        """
         Parameters
         ----------
         client
-            An `OpenAI` client instance.
+            An `openai.OpenAI` client instance.
+        model_name
+            The name of the model to use.
 
         """
         self.client = client
         self.model_name = model_name
         self.type_adapter = SGLangTypeAdapter()
 
-    def generate(self, model_input, output_type, **inference_kwargs):
-        """Generate text using the client.
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using SGLang.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        The generated text.
+        Union[str, list[str]]
+            The text generated by the model.
 
         """
         client_args = self._build_client_args(
@@ -111,21 +147,30 @@ def generate(self, model_input, output_type, **inference_kwargs):
         else:
             return [message.content for message in messages]
 
-    def generate_stream(self, model_input, output_type, **inference_kwargs):
-        """Return a text generator.
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using SGLang.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        A generator that yields the generated text chunks.
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
         """
         client_args = self._build_client_args(
             model_input, output_type, **inference_kwargs,
@@ -139,7 +184,12 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
             if chunk.choices and chunk.choices[0].delta.content is not None:
                 yield chunk.choices[0].delta.content
 
-    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
         """Build the arguments to pass to the SGLang client."""
         messages = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
@@ -157,36 +207,57 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
 
 
 class AsyncSGLang(AsyncModel):
-    """Represents an async client to a `sglang` server."""
+    """Thin async wrapper around the `openai.OpenAI` client used to communicate
+    with an SGLang server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    SGLang server.
+
+    """
 
     def __init__(self, client, model_name: Optional[str] = None):
-        """Create an `AsyncSGLang` model instance.
+        """
+        Parameters
+        ----------
+        client
+            An `openai.AsyncOpenAI` client instance.
+        model_name
+            The name of the model to use.
 
         Parameters
         ----------
         client
-            An `AsyncOpenAI` client instance.
+            An `openai.AsyncOpenAI` client instance.
 
         """
         self.client = client
         self.model_name = model_name
         self.type_adapter = SGLangTypeAdapter()
 
-    async def generate(self, model_input, output_type, **inference_kwargs):
+    async def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
         """Generate text using `sglang`.
 
-        Arguments
-        ---------
-        prompt
-            The text prompt provided by the user.
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The inference kwargs that can be passed to the sglang server
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        The generated text.
+        Union[str, list[str]]
+            The text generated by the model.
 
         """
         client_args = self._build_client_args(
@@ -208,23 +279,30 @@ async def generate(self, model_input, output_type, **inference_kwargs):
         else:
             return [message.content for message in messages]
 
-    async def generate_stream(
-        self, model_input, output_type, **inference_kwargs
-    ):
+    async def generate_stream( # type: ignore
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> AsyncIterator[str]:
         """Return a text generator.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        A generator that yields the generated text chunks.
+        AsyncIterator[str]
+            An async iterator that yields the text generated by the model.
+
         """
         client_args = self._build_client_args(
             model_input, output_type, **inference_kwargs,
@@ -239,7 +317,12 @@ async def generate_stream(
             if chunk.choices and chunk.choices[0].delta.content is not None:
                 yield chunk.choices[0].delta.content
 
-    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
         """Build the arguments to pass to the SGLang client."""
         messages = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
@@ -260,6 +343,22 @@ def from_sglang(
     client: Union["OpenAI", "AsyncOpenAI"],
     model_name: Optional[str] = None,
 ) -> Union[SGLang, AsyncSGLang]:
+    """Create a `SGLang` or `AsyncSGLang` instance from an `openai.OpenAI` or
+    `openai.AsyncOpenAI` instance.
+
+    Parameters
+    ----------
+    client
+        An `openai.OpenAI` or `openai.AsyncOpenAI` instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Union[SGLang, AsyncSGLang]
+        An Outlines `SGLang` or `AsyncSGLang` model instance.
+
+    """
     from openai import AsyncOpenAI, OpenAI
 
     if isinstance(client, OpenAI):
diff --git a/outlines/models/tgi.py b/outlines/models/tgi.py
index 3e135890f..d9f3ffd75 100644
--- a/outlines/models/tgi.py
+++ b/outlines/models/tgi.py
@@ -1,16 +1,27 @@
+"""Integration with a TGI server."""
+
 import json
 from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Union
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    AsyncIterator,
+    Iterator,
+    Optional,
+    Union,
+)
 
 from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
 from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
 
-
 if TYPE_CHECKING:
     from huggingface_hub import AsyncInferenceClient, InferenceClient
 
+__all__ = ["AsyncTGI", "TGI", "from_tgi"]
+
 
 class TGITypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `TGI` and `AsyncTGI` models."""
 
     @singledispatchmethod
     def format_input(self, model_input):
@@ -21,6 +32,11 @@ def format_input(self, model_input):
         model_input
             The input passed by the user.
 
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
+
         """
         raise NotImplementedError(
             f"The input type {input} is not available with TGI. "
@@ -28,14 +44,14 @@ def format_input(self, model_input):
         )
 
     @format_input.register(str)
-    def format_str_input(self, model_input: str):
+    def format_str_input(self, model_input: str) -> str:
         return model_input
 
     @format_input.register(list)
     def format_list_input(self, model_input: list):
         raise NotImplementedError("TGI does not support batch inference.")
 
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the structured output argument to pass to the client.
 
         Argument
@@ -43,6 +59,11 @@ def format_output_type(self, output_type):
         output_type
             The structured output type provided.
 
+        Returns
+        -------
+        dict
+            The structured output argument to pass to the client.
+
         """
         if output_type is None:
             return {}
@@ -69,11 +90,17 @@ def format_output_type(self, output_type):
 
 
 class TGI(Model):
-    """Represents a client to a huggingface TGI server."""
+    """Thin wrapper around a `huggingface_hub.InferenceClient` client used to
+    communicate with a `TGI` server.
 
-    def __init__(self, client):
-        """Create a `TGI` model instance.
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the
+    `huggingface_hub.InferenceClient` client.
 
+    """
+
+    def __init__(self, client):
+        """
         Parameters
         ----------
         client
@@ -83,21 +110,29 @@ def __init__(self, client):
         self.client = client
         self.type_adapter = TGITypeAdapter()
 
-    def generate(self, model_input, output_type, **inference_kwargs):
-        """Generate text using the client.
+    def generate(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
+        """Generate text using TGI.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        The generated text.
+        str
+            The text generated by the model.
 
         """
         client_args = self._build_client_args(
@@ -108,21 +143,30 @@ def generate(self, model_input, output_type, **inference_kwargs):
 
         return self.client.text_generation(**client_args)
 
-    def generate_stream(self, model_input, output_type, **inference_kwargs):
-        """Return a text generator.
+    def generate_stream(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using TGI.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        A generator that yields the generated text chunks.
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
         """
         client_args = self._build_client_args(
             model_input, output_type, **inference_kwargs,
@@ -135,7 +179,12 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
         for chunk in stream:  # pragma: no cover
             yield chunk
 
-    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+    def _build_client_args(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
         """Build the arguments to pass to the TGI client."""
         prompt = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
@@ -150,11 +199,17 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
 
 
 class AsyncTGI(AsyncModel):
-    """Represents an async client to a huggingface TGI server."""
+    """Thin async wrapper around a `huggingface_hub.AsyncInferenceClient`
+    client used to communicate with a `TGI` server.
 
-    def __init__(self, client):
-        """Create an `AsyncTGI` model instance.
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the
+    `huggingface_hub.AsyncInferenceClient` client.
 
+    """
+
+    def __init__(self, client):
+        """
         Parameters
         ----------
         client
@@ -164,21 +219,29 @@ def __init__(self, client):
         self.client = client
         self.type_adapter = TGITypeAdapter()
 
-    async def generate(self, model_input, output_type, **inference_kwargs):
+    async def generate(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> str:
         """Generate text using TGI.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        The generated text.
+        str
+            The text generated by the model.
 
         """
         client_args = self._build_client_args(
@@ -189,23 +252,30 @@ async def generate(self, model_input, output_type, **inference_kwargs):
 
         return response
 
-    async def generate_stream(
-        self, model_input, output_type, **inference_kwargs
-    ):
-        """Return a text generator.
+    async def generate_stream( # type: ignore
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> AsyncIterator[str]:
+        """Stream text using TGI.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types except `CFG` are supported provided your server uses
+            a backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        A generator that yields the generated text chunks.
+        AsyncIterator[str]
+            An async iterator that yields the text generated by the model.
+
         """
         client_args = self._build_client_args(
             model_input, output_type, **inference_kwargs,
@@ -218,7 +288,12 @@ async def generate_stream(
         async for chunk in stream:  # pragma: no cover
             yield chunk
 
-    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+    def _build_client_args(
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
         """Build the arguments to pass to the TGI client."""
         prompt = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
@@ -235,6 +310,22 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
 def from_tgi(
     client: Union["InferenceClient", "AsyncInferenceClient"],
 ) -> Union[TGI, AsyncTGI]:
+    """Create an Outlines `TGI` or `AsyncTGI` model instance from an
+    `huggingface_hub.InferenceClient` or `huggingface_hub.AsyncInferenceClient`
+    instance.
+
+    Parameters
+    ----------
+    client
+        An `huggingface_hub.InferenceClient` or
+        `huggingface_hub.AsyncInferenceClient` instance.
+
+    Returns
+    -------
+    Union[TGI, AsyncTGI]
+        An Outlines `TGI` or `AsyncTGI` model instance.
+
+    """
     from huggingface_hub import AsyncInferenceClient, InferenceClient
 
     if isinstance(client, InferenceClient):
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 785e2c983..06ed7df7b 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,15 +1,22 @@
+"""Integration with the `transformers` library."""
+
 from functools import singledispatchmethod
-from typing import TYPE_CHECKING, List, Tuple, Union
+from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
+from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     import torch
-    from transformers import PreTrainedTokenizer, PreTrainedModel, ProcessorMixin
-
+    from transformers import (
+        PreTrainedTokenizer,
+        PreTrainedModel,
+        ProcessorMixin,
+        LogitsProcessorList,
+    )
 
-__all__ = ["Transformers", "from_transformers"]
+__all__ = ["Transformers", "TransformersMultiModal", "from_transformers"]
 
 
 def get_llama_tokenizer_types():
@@ -122,15 +129,22 @@ def __setstate__(self, state):
 
 
 class TransformersTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `Transformers` model."""
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
             The input passed by the user.
 
+        Returns
+        -------
+        str
+            The formatted input to be passed to the model.
+
         """
         raise NotImplementedError(
             f"The input type {input} is not available."
@@ -138,21 +152,29 @@ def format_input(self, model_input):
         )
 
     @format_input.register(str)
-    def format_str_input(self, model_input):
+    def format_str_input(self, model_input: str) -> str:
         return model_input
 
     @format_input.register(list)
-    def format_list_input(self, model_input):
+    def format_list_input(self, model_input: List[str]) -> List[str]:
         return model_input
 
-    def format_output_type(self, output_type):
+    def format_output_type(
+        self,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Optional["LogitsProcessorList"]:
         """Generate the logits processor argument to pass to the model.
 
-        Argument
-        --------
+        Parameters
+        ----------
         output_type
             The logits processor provided.
 
+        Returns
+        -------
+        Optional[LogitsProcessorList]
+            The logits processor to pass to the model.
+
         """
         from transformers import LogitsProcessorList
 
@@ -162,19 +184,21 @@ def format_output_type(self, output_type):
 
 
 class Transformers(Model):
-    """Represents a `transformers` model."""
+    """Thin wrapper around a `transformers` model and a `transformers`
+    tokenizer.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `transformers` model and
+    tokenizer.
+
+    """
 
     def __init__(
         self,
         model: "PreTrainedModel",
         tokenizer: "PreTrainedTokenizer",
     ):
-        """Create a Transformers model instance.
-
-        `outlines` supports `PreTrainedModelForCausalLM`,
-        `PreTrainedMambaForCausalLM`, `PreTrainedModelForSeq2Seq` and any model
-        that implements the `transformers` model API.
-
+        """
         Parameters:
         ----------
         model
@@ -216,7 +240,12 @@ def __init__(
         else:
             self.tensor_library_name = "torch"
 
-    def _prepare_model_inputs(self, model_input, output_type):
+    def _prepare_model_inputs(
+        self,
+        model_input: Union[str, List[str], dict],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Tuple[Union[str, List[str]], dict]:
+        """Turn the user input into arguements to pass to the model"""
         prompts = self.type_adapter.format_input(model_input)
         input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
@@ -226,10 +255,39 @@ def _prepare_model_inputs(self, model_input, output_type):
 
         return prompts, inputs
 
-    def generate(self, model_input, output_type, **inference_kwargs):
+    def generate(
+        self,
+        model_input: Union[str, List[str], dict],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, List[str]]:
+        """Generate text using `transformers`.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response. For
+            multi-modal models, the input should be a dictionary containing the
+            `text` key with a value of type `Union[str, List[str]]` and the
+            other keys required by the model.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        inference_kwargs
+            Additional keyword arguments to pass to the `generate` method
+            of the `transformers` model.
+
+        Returns
+        -------
+        Union[str, List[str]]
+            The text generated by the model.
+
+        """
         prompts, inputs = self._prepare_model_inputs(model_input, output_type)
         logits_processor = self.type_adapter.format_output_type(output_type)
 
+        print(inference_kwargs)
+
         generated_ids = self._generate_output_seq(
             prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )
@@ -241,8 +299,10 @@ def generate(self, model_input, output_type, **inference_kwargs):
         return self._decode_generation(generated_ids)
 
     def generate_stream(self, model_input, output_type, **inference_kwargs):
-        """
+        """Not available for `transformers` models.
+
         TODO: implement following completion of https://github.com/huggingface/transformers/issues/30810
+
         """
         raise NotImplementedError(
             "Streaming is not implemented for Transformers models."
@@ -287,7 +347,7 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
 
 
 class TransformersMultiModalTypeAdapter(ModelTypeAdapter):
-    """Type adapter for TransformersMultiModal models."""
+    """Type adapter for `TransformersMultiModal` model."""
 
     @singledispatchmethod
     def format_input(self, model_input):
@@ -298,6 +358,11 @@ def format_input(self, model_input):
         model_input
             The input passed by the user.
 
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the model.
+
         """
         raise NotImplementedError(
             f"The input type {input} is not available. Please provide a "
@@ -311,7 +376,7 @@ def format_input(self, model_input):
         )
 
     @format_input.register(dict)
-    def format_list_input(self, model_input):
+    def format_list_input(self, model_input: dict) -> dict:
         if "text" not in model_input:
             raise ValueError(
                 "The input must contain the 'text' key along with the other "
@@ -319,7 +384,10 @@ def format_list_input(self, model_input):
             )
         return model_input
 
-    def format_output_type(self, output_type):
+    def format_output_type(
+        self,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Optional["LogitsProcessorList"]:
         """Generate the logits processor argument to pass to the model.
 
         Argument
@@ -327,6 +395,11 @@ def format_output_type(self, output_type):
         output_type
             The logits processor provided.
 
+        Returns
+        -------
+        Optional[LogitsProcessorList]
+            The logits processor to pass to the model.
+
         """
         from transformers import LogitsProcessorList
 
@@ -336,7 +409,14 @@ def format_output_type(self, output_type):
 
 
 class TransformersMultiModal(Transformers):
-    """Represents a `transformers` model with a vision processor."""
+    """Thin wrapper around a `transformers` model and a `transformers`
+    processor.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `transformers` model and
+    processor.
+
+    """
 
     def __init__(self, model: "PreTrainedModel", processor):
         """Create a TransformersMultiModal model instance
@@ -345,6 +425,14 @@ def __init__(self, model: "PreTrainedModel", processor):
         most of the initialization and then add elements specific to vision
         models.
 
+        Parameters
+        ----------
+        model
+            A `PreTrainedModel`, or any model that is compatible with the
+            `transformers` API for models.
+        processor
+            A `ProcessorMixin` instance.
+
         """
         self.processor = processor
         self.processor.padding_side = "left"
@@ -356,7 +444,12 @@ def __init__(self, model: "PreTrainedModel", processor):
 
         self.type_adapter = TransformersMultiModalTypeAdapter()
 
-    def _prepare_model_inputs(self, model_input, output_type):
+    def _prepare_model_inputs(
+        self,
+        model_input: Union[str, List[str], dict],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Tuple[Union[str, List[str]], dict]:
+        """Turn the user input into arguements to pass to the model"""
         model_input = self.type_adapter.format_input(model_input)
         inputs = self.processor(
             **model_input, padding=True, return_tensors="pt"
@@ -368,12 +461,31 @@ def _prepare_model_inputs(self, model_input, output_type):
 def from_transformers(
     model: "PreTrainedModel",
     tokenizer_or_processor: Union["PreTrainedTokenizer", "ProcessorMixin"],
-):
+) -> Union[Transformers, TransformersMultiModal]:
+    """Create an Outlines `Transformers` or `TransformersMultiModal` model
+    instance from a `PreTrainedModel` instance and a `PreTrainedTokenizer` or
+    `ProcessorMixin` instance.
+
+    `outlines` supports `PreTrainedModelForCausalLM`,
+    `PreTrainedMambaForCausalLM`, `PreTrainedModelForSeq2Seq` and any model
+    that implements the `transformers` model API.
+
+    Parameters
+    ----------
+    model
+        A `transformers.PreTrainedModel` instance.
+    tokenizer_or_processor
+        A `transformers.PreTrainedTokenizer` or
+        `transformers.ProcessorMixin` instance.
+
+    Returns
+    -------
+    Union[Transformers, TransformersMultiModal]
+        An Outlines `Transformers` or `TransformersMultiModal` model instance.
+
+    """
     from transformers import (
-        PreTrainedTokenizer,
-        PreTrainedTokenizerFast,
-        ProcessorMixin
-    )
+        PreTrainedTokenizer, PreTrainedTokenizerFast, ProcessorMixin)
 
     if isinstance(
         tokenizer_or_processor, (PreTrainedTokenizer, PreTrainedTokenizerFast)
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index c3a191169..f787e7072 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -1,39 +1,54 @@
+"""Integration with a vLLM server."""
+
 import json
-from typing import TYPE_CHECKING, Union, Optional
+from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
 
 from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
 from outlines.models.openai import OpenAITypeAdapter
-from outlines.types.dsl import python_types_to_terms, to_regex, JsonSchema, CFG
-
+from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
+from outlines.templates import Vision
 
 if TYPE_CHECKING:
     from openai import AsyncOpenAI, OpenAI
 
+__all__ = ["VLLM", "AsyncVLLM", "from_vllm"]
+
 
 class VLLMTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `VLLM` and `AsyncVLLM` models."""
 
-    def format_input(self, model_input):
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
         """Generate the prompt argument to pass to the client.
 
         We rely on the OpenAITypeAdapter to format the input as the vLLM server
         expects input in the same format as OpenAI.
 
-        Argument
-        --------
+        Parameters
+        ----------
         model_input
             The input passed by the user.
 
+        Returns
+        -------
+        dict
+            The formatted input to be passed to the model.
+
         """
         return OpenAITypeAdapter().format_input(model_input)
 
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the structured output argument to pass to the client.
 
-        Argument
-        --------
+        Parameters
+        ----------
         output_type
             The structured output type provided.
 
+        Returns
+        -------
+        dict
+            The structured output argument to pass to the model.
+
         """
         if output_type is None:
             return {}
@@ -51,36 +66,53 @@ def format_output_type(self, output_type):
 
 
 class VLLM(Model):
-    """Represents a client to a `vllm` server."""
-
-    def __init__(self, client, model_name: Optional[str] = None):
-        """Create a `VLLMServer` model instance.
-
+    """Thin wrapper around the `openai.OpenAI` client used to communicate with
+    a `vllm` server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    `vllm` server.
+    """
+
+    def __init__(
+        self,
+        client: "OpenAI",
+        model_name: Optional[str] = None,
+    ):
+        """
         Parameters
         ----------
         client
-            An `OpenAI` client instance.
+            An `openai.OpenAI` client instance.
 
         """
         self.client = client
         self.model_name = model_name
         self.type_adapter = VLLMTypeAdapter()
 
-    def generate(self, model_input, output_type, **inference_kwargs):
-        """Generate text using the client.
+    def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using vLLM.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        The generated text.
+        Union[str, list[str]]
+            The text generated by the model.
 
         """
         client_args = self._build_client_args(
@@ -104,21 +136,30 @@ def generate(self, model_input, output_type, **inference_kwargs):
         else:
             return [message.content for message in messages]
 
-    def generate_stream(self, model_input, output_type, **inference_kwargs):
-        """Return a text generator.
+    def generate_stream(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Iterator[str]:
+        """Stream text using vLLM.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        A generator that yields the generated text chunks.
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
         """
         client_args = self._build_client_args(
             model_input, output_type, **inference_kwargs,
@@ -132,7 +173,12 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
             if chunk.choices and chunk.choices[0].delta.content is not None:
                 yield chunk.choices[0].delta.content
 
-    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
         """Build the arguments to pass to the OpenAI client."""
         messages = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
@@ -153,37 +199,53 @@ def _build_client_args(self, model_input, output_type, **inference_kwargs):
 
 
 class AsyncVLLM(AsyncModel):
-    """Represents an async client to a `vllm` server."""
-
-    def __init__(self, client, model_name: Optional[str] = None):
-        """Create an `AsyncVLLMServer` model instance.
-
+    """Thin async wrapper around the `openai.OpenAI` client used to communicate
+    with a `vllm` server.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.OpenAI` client for the
+    `vllm` server.
+    """
+
+    def __init__(
+        self,
+        client: "AsyncOpenAI",
+        model_name: Optional[str] = None,
+    ):
+        """
         Parameters
         ----------
         client
-            An `AsyncOpenAI` client instance.
+            An `openai.AsyncOpenAI` client instance.
 
         """
         self.client = client
         self.model_name = model_name
         self.type_adapter = VLLMTypeAdapter()
 
-    async def generate(self, model_input, output_type, **inference_kwargs):
-        """Generate text using `vllm`.
+    async def generate(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using vLLM.
 
-        Arguments
-        ---------
-        prompt
-            The text prompt provided by the user.
-        logits_processor
-            The logits processor to use when generating text.
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The inference kwargs that can be passed to the `LLM.generate` method
-            in the `vllm` library.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        The generated text.
+        Union[str, list[str]]
+            The text generated by the model.
 
         """
         client_args = self._build_client_args(
@@ -205,23 +267,29 @@ async def generate(self, model_input, output_type, **inference_kwargs):
         else:
             return [message.content for message in messages]
 
-    async def generate_stream(
-        self, model_input, output_type, **inference_kwargs
-    ):
-        """Return a text generator.
+    async def generate_stream( # type: ignore
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> AsyncIterator[str]:
+        """Stream text using vLLM.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         model_input
-            The text prompt provided by the user.
+            The prompt based on which the model will generate a response.
         output_type
-            The structured output type provided.
+            The desired format of the response generated by the model. All
+            output types available in Outlines are supported provided your
+            server uses a structured generation backend that supports them.
         inference_kwargs
-            The additional arguments that will be passed on to the client.
+            Additional keyword arguments to pass to the client.
 
         Returns
         -------
-        A generator that yields the generated text chunks.
+        AsyncIterator[str]
+            An async iterator that yields the text generated by the model.
         """
         client_args = self._build_client_args(
             model_input, output_type, **inference_kwargs,
@@ -236,7 +304,12 @@ async def generate_stream(
             if chunk.choices and chunk.choices[0].delta.content is not None:
                 yield chunk.choices[0].delta.content
 
-    def _build_client_args(self, model_input, output_type, **inference_kwargs):
+    def _build_client_args(
+        self,
+        model_input: Union[str, Vision],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> dict:
         """Build the arguments to pass to the OpenAI client."""
         messages = self.type_adapter.format_input(model_input)
         output_type_args = self.type_adapter.format_output_type(output_type)
@@ -260,6 +333,22 @@ def from_vllm(
     client: Union["OpenAI", "AsyncOpenAI"],
     model_name: Optional[str] = None,
 ) -> Union[VLLM, AsyncVLLM]:
+    """Create an Outlines `VLLM` or `AsyncVLLM` model instance from an
+    `openai.OpenAI` or `openai.AsyncOpenAI` instance.
+
+    Parameters
+    ----------
+    client
+        An `openai.OpenAI` or `openai.AsyncOpenAI` instance.
+    model_name
+        The name of the model to use.
+
+    Returns
+    -------
+    Union[VLLM, AsyncVLLM]
+        An Outlines `VLLM` or `AsyncVLLM` model instance.
+
+    """
     from openai import AsyncOpenAI, OpenAI
 
     if isinstance(client, OpenAI):
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index 4766139b1..97164171a 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -1,18 +1,22 @@
+"""Integration with the `vllm` library (offline mode)."""
+
 import warnings
 from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Optional, Union
+from typing import TYPE_CHECKING, Any, List, Optional, Union
 
 from outlines.models.base import Model, ModelTypeAdapter
+from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     from transformers import PreTrainedTokenizerBase
     from vllm import LLM
 
-
 __all__ = ["VLLMOffline", "from_vllm_offline"]
 
 
 class VLLMOfflineTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `VLLMOffline` model."""
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
@@ -29,14 +33,17 @@ def format_input(self, model_input):
         )
 
     @format_input.register(str)
-    def format_str_input(self, model_input):
+    def format_str_input(self, model_input: str) -> str:
         return model_input
 
     @format_input.register(list)
-    def format_list_input(self, model_input):
+    def format_list_input(self, model_input: List[str]) -> List[str]:
         return model_input
 
-    def format_output_type(self, output_type):
+    def format_output_type(
+        self,
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+    ) -> Union[List[OutlinesLogitsProcessor], List[None]]:
         """Generate the logits processor argument to pass to the model.
 
         Argument
@@ -44,6 +51,11 @@ def format_output_type(self, output_type):
         output_type
             The logits processor provided.
 
+        Returns
+        -------
+        Union[List[OutlinesLogitsProcessor], List[None]]
+            The logits processor to pass to the model.
+
         """
         if output_type is not None:
             return [output_type]
@@ -51,7 +63,12 @@ def format_output_type(self, output_type):
 
 
 class VLLMOffline(Model):
-    """Represents a `vllm` model."""
+    """Thin wrapper around a `vllm.LLM` model.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `vllm.LLM` model.
+
+    """
     tensor_library_name = "torch"
 
     def __init__(self, model: "LLM"):
@@ -83,22 +100,29 @@ def _get_tokenizer(self):
             )
         return adapt_tokenizer(tokenizer=tokenizer)
 
-    def generate(self, model_input, output_type, **inference_kwargs):
-        """Generate text using `vllm`.
+    def generate(
+        self,
+        model_input: Union[str, List[str]],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, List[str], List[List[str]]]:
+        """Generate text using vLLM.
 
-        Arguments
-        ---------
+        Parameters
+        ----------
         prompt
-            The text prompt provided by the user.
-        logits_processor
-            The logits processor to use when generating text.
+            The prompt based on which the model will generate a response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
         inference_kwargs
-            The inference kwargs that can be passed to the `LLM.generate` method
-            in the `vllm` library.
+            Additional keyword arguments to pass to the `generate` method
+            in the `vllm.LLM` model.
 
         Returns
         -------
-        The generated text.
+        Union[str, List[str], List[List[str]]]
+            The text generated by the model.
 
         """
         from vllm.sampling_params import SamplingParams
@@ -129,9 +153,7 @@ def generate(self, model_input, output_type, **inference_kwargs):
         return results
 
     def generate_stream(self, model_input, output_type, **inference_kwargs):
-        """Return a text generator.
-
-        Streaming is not yet available for `vllm.LLM`.
+        """Not available for `vllm.LLM`.
 
         TODO: Implement the streaming functionality ourselves.
 
@@ -140,7 +162,13 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
             "Streaming is not available for the vLLM integration."
         )
 
-    def load_lora(self, adapter_path: Optional[str]):
+    def load_lora(self, adapter_path: Optional[str]) -> None:
+        """Load a LoRA adapter. Deprecated since v1.0.0.
+
+        Use the `lora_request` argument when calling the model or generator
+        instead.
+
+        """
         warnings.warn("""
             The `load_lora` method is deprecated starting from v1.0.0.
             Support for it will be removed in v1.1.0.
@@ -160,15 +188,31 @@ def load_lora(self, adapter_path: Optional[str]):
 
 
 def from_vllm_offline(model: "LLM") -> VLLMOffline:
+    """Create an Outlines `VLLMOffline` model instance from a `vllm.LLM`
+    instance.
+
+    Parameters
+    ----------
+    model
+        A `vllm.LLM` instance.
+
+    Returns
+    -------
+    VLLMOffline
+        An Outlines `VLLMOffline` model instance.
+
+    """
     return VLLMOffline(model)
 
 
-def adapt_tokenizer(tokenizer: "PreTrainedTokenizerBase") -> "PreTrainedTokenizerBase":
+def adapt_tokenizer(
+    tokenizer: "PreTrainedTokenizerBase",
+) -> "PreTrainedTokenizerBase":
     """Adapt a tokenizer to use to compile the FSM.
 
-    The API of Outlines tokenizers is slightly different to that of `transformers`. In
-    addition we need to handle the missing spaces to Llama's tokenizer to be able to
-    compile FSMs for this model.
+    The API of Outlines tokenizers is slightly different to that of
+    `transformers`. In addition we need to handle the missing spaces to Llama's
+    tokenizer to be able to compile FSMs for this model.
 
     Parameters
     ----------
@@ -179,6 +223,7 @@ def adapt_tokenizer(tokenizer: "PreTrainedTokenizerBase") -> "PreTrainedTokenize
     -------
     PreTrainedTokenizerBase
         The adapted tokenizer.
+
     """
     from transformers import SPIECE_UNDERLINE
 

From 891108ec6123b933b27b6043430ee0a6940427ae Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 27 May 2025 10:25:33 +0200
Subject: [PATCH 337/505] Improve docstrings and type hinting in model base and
 generator

---
 outlines/applications.py    |   2 +-
 outlines/generator.py       | 283 ++++++++++++++++++++++++++++++------
 outlines/models/__init__.py |  24 ++-
 outlines/models/base.py     | 202 +++++++++++++++++++++++--
 tests/test_applications.py  |  49 +++----
 tests/test_generator.py     |   5 +
 6 files changed, 459 insertions(+), 106 deletions(-)

diff --git a/outlines/applications.py b/outlines/applications.py
index 63eb45fef..1a7dd087d 100644
--- a/outlines/applications.py
+++ b/outlines/applications.py
@@ -64,7 +64,7 @@ def __call__(
         # generator.
         if model != self.model:
             self.model = model
-            self.generator = Generator(model, self.output_type)
+            self.generator = Generator(model, self.output_type)  # type: ignore
 
         prompt = self.template(**template_vars)
         assert self.generator is not None
diff --git a/outlines/generator.py b/outlines/generator.py
index badeccecb..1623fe85e 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -1,12 +1,24 @@
-from typing import Any, Optional, Union
+"""Encapsulate a model and an output type into a reusable object."""
 
-from outlines.models import BlackBoxModel, SteerableModel
-from outlines.models.base import AsyncModel
+from typing import (
+    Any,
+    AsyncIterator,
+    Iterator,
+    Optional,
+    Union,
+)
+
+from outlines.models import (
+    AsyncBlackBoxModel,
+    BlackBoxModel,
+    SteerableModel,
+)
+from outlines.models.base import AsyncModel, Model
 from outlines.processors import (
     CFGLogitsProcessor,
     GuideLogitsProcessor,
-    RegexLogitsProcessor,
     OutlinesLogitsProcessor,
+    RegexLogitsProcessor,
 )
 from outlines.processors.guide import RegexGuide
 from outlines.types import CFG, FSM
@@ -14,10 +26,25 @@
 
 
 class BlackBoxGenerator:
-    """Synchronous generator for which we don't control constrained generation."""
+    """Synchronous generator for which we don't control constrained
+    generation.
+
+    The output type provided is not compiled into a logits processor, but is
+    instead directly passed on to the model.
+
+    """
     output_type: Optional[Any]
 
-    def __init__(self, model, output_type: Optional[Any]):
+    def __init__(self, model: BlackBoxModel, output_type: Optional[Any]):
+        """
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        output_type
+            The output type that will be used to constrain the generation.
+
+        """
         self.model = model
         self.output_type = output_type
 
@@ -26,18 +53,67 @@ def __init__(self, model, output_type: Optional[Any]):
                 "FSM generation is not supported for API-based models"
             )
 
-    def __call__(self, prompt, **inference_kwargs):
-        return self.model.generate(prompt, self.output_type, **inference_kwargs)
+    def __call__(self, prompt: Any, **inference_kwargs) -> Any:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate(
+            prompt, self.output_type, **inference_kwargs
+        )
 
-    def stream(self, prompt, **inference_kwargs):
-        return self.model.generate_stream(prompt, self.output_type, **inference_kwargs)
+    def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate_stream(
+            prompt, self.output_type, **inference_kwargs
+        )
 
 
 class AsyncBlackBoxGenerator:
-    """Asynchronous generator for which we don't control constrained generation."""
+    """Asynchronous generator for which we don't control constrained
+    generation.
+
+    The output type provided is not compiled into a logits processor, but is
+    instead directly passed on to the model.
+
+    """
     output_type: Optional[Any]
 
-    def __init__(self, model, output_type: Optional[Any]):
+    def __init__(self, model: AsyncBlackBoxModel, output_type: Optional[Any]):
+        """
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        output_type
+            The output type that will be used to constrain the generation.
+
+        """
         self.model = model
         self.output_type = output_type
 
@@ -46,10 +122,44 @@ def __init__(self, model, output_type: Optional[Any]):
                 "FSM generation is not supported for API-based models"
             )
 
-    async def __call__(self, prompt, **inference_kwargs):
-        return await self.model.generate(prompt, self.output_type, **inference_kwargs)
+    async def __call__(self, prompt: Any, **inference_kwargs) -> Any:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return await self.model.generate(
+            prompt, self.output_type, **inference_kwargs
+        )
 
-    async def stream(self, prompt, **inference_kwargs):
+    async def stream(
+        self, prompt: Any, **inference_kwargs
+    ) -> AsyncIterator[Any]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
         async for chunk in self.model.generate_stream(  # pragma: no cover
             prompt, self.output_type, **inference_kwargs
         ):
@@ -60,20 +170,32 @@ class SteerableGenerator:
     """Represents a generator for which we control constrained generation.
 
     The generator is responsible for building and storing the logits processor
-    (which can be quite expensive to build), and then to pass it to the model
-    when the generator is called. The argument defining constrained generation
-    can be of 2 types associated to different methods to create an instance of
-    the generator:
-    - output_type (through `__init__`): a Python type, will be used to build a
-      logits processor
-    - processor (through `from_processor`): an already built
-      logits processor
+    (which can be quite expensive to build), and then passing it to the model
+    when the generator is called.
+
+    The argument defining constrained generation can be of 2 types associated
+    to different methods to create an instance of the generator:
+    - `output_type` (through `__init__`): an output type as defined in the
+      `outlines.types` module
+    - `processor` (through `from_processor`): an already built logits processor
+       as defined in the `outlines.processors` module
+
+    The 2 parameters are mutually exclusive.
+
     """
     logits_processor: Optional[OutlinesLogitsProcessor]
 
-    def __init__(self, model, output_type: Optional[Any]):
-        self.model = model
+    def __init__(self, model: SteerableModel, output_type: Optional[Any]):
+        """
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        output_type
+            The output type expressed as a Python type
 
+        """
+        self.model = model
         if output_type is None:
             self.logits_processor = None
         else:
@@ -81,44 +203,99 @@ def __init__(self, model, output_type: Optional[Any]):
             if isinstance(term, CFG):
                 cfg_string = term.definition
                 self.logits_processor = CFGLogitsProcessor(
-                    cfg_string, self.model.tokenizer, self.model.tensor_library_name
+                    cfg_string,
+                    self.model.tokenizer,
+                    self.model.tensor_library_name,
                 )
             elif isinstance(term, FSM):
-                guide = RegexGuide.from_interegular_fsm(term.fsm, self.model.tokenizer)
+                guide = RegexGuide.from_interegular_fsm(
+                    term.fsm,
+                    self.model.tokenizer,
+                )
                 self.logits_processor = GuideLogitsProcessor(
                     self.model.tokenizer, guide, self.model.tensor_library_name
                 )
             else:
                 regex_string = to_regex(term)
                 self.logits_processor = RegexLogitsProcessor(
-                    regex_string, self.model.tokenizer, self.model.tensor_library_name
+                    regex_string,
+                    self.model.tokenizer,
+                    self.model.tensor_library_name,
                 )
 
     @classmethod
-    def from_processor(cls, model, processor: OutlinesLogitsProcessor):
+    def from_processor(
+        cls, model: SteerableModel, processor: OutlinesLogitsProcessor
+    ):
+        """Create a generator from a logits processor.
+
+        Parameters
+        ----------
+        model
+            An instance of an Outlines model.
+        processor
+            An instance of an OutlinesLogitsProcessor.
+
+        """
         if not isinstance(processor, OutlinesLogitsProcessor):
-            raise TypeError("The processor argument must be an instance of OutlinesLogitsProcessor")
+            raise TypeError(
+                "The processor argument must be an instance of "
+                "OutlinesLogitsProcessor"
+            )
         instance = cls.__new__(cls)
         instance.model = model
         instance.logits_processor = processor
 
         return instance
 
-    def __call__(self, prompt, **inference_kwargs):
-        return self.model.generate(prompt, self.logits_processor, **inference_kwargs)
+    def __call__(self, prompt: Any, **inference_kwargs) -> Any:
+        """Generate a response from the model.
 
-    def stream(self, prompt, **inference_kwargs):
-        return self.model.generate_stream(prompt, self.logits_processor, **inference_kwargs)
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate(
+            prompt, self.logits_processor, **inference_kwargs
+        )
+
+    def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        prompt
+            The prompt to use to generate a response.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
+        """
+        return self.model.generate_stream(
+            prompt, self.logits_processor, **inference_kwargs
+        )
 
 
 def Generator(
-    model: Union[SteerableModel, BlackBoxModel],
+    model: Union[Model, AsyncModel],
     output_type: Optional[Any] = None,
     *,
     processor: Optional[OutlinesLogitsProcessor] = None,
-):
-    """
-    Create a generator for the given model and output parameters.
+) -> Union[SteerableGenerator, BlackBoxGenerator, AsyncBlackBoxGenerator]:
+    """Create a generator for the given model and output parameters.
 
     The 2 parameters output_type and processor are mutually exclusive. The
     parameters processor is only supported for SteerableModel instances
@@ -136,25 +313,35 @@ def Generator(
 
     Returns
     -------
-    A generator.
+    Union[SteerableGenerator, BlackBoxGenerator, AsyncBlackBoxGenerator]
+        A generator instance.
+
     """
     provided_output_params = sum(
         param is not None
         for param in [output_type, processor]
     )
     if provided_output_params > 1:
-        raise ValueError("At most one of output_type or processor can be provided")
+        raise ValueError(
+            "At most one of output_type or processor can be provided"
+        )
 
-    if isinstance(model, BlackBoxModel): # type: ignore
+    if isinstance(model, SteerableModel): # type: ignore
         if processor is not None:
-            raise NotImplementedError("This model does not support logits processors")
+            return SteerableGenerator.from_processor(model, processor) # type: ignore
         else:
-            if isinstance(model, AsyncModel):
-                return AsyncBlackBoxGenerator(model, output_type)
-            else:
-                return BlackBoxGenerator(model, output_type)
+            return SteerableGenerator(model, output_type) # type: ignore
     else:
         if processor is not None:
-            return SteerableGenerator.from_processor(model, processor)
+            raise NotImplementedError(
+                "This model does not support logits processors"
+            )
+        if isinstance(model, AsyncBlackBoxModel): # type: ignore
+            return AsyncBlackBoxGenerator(model, output_type) # type: ignore
+        elif isinstance(model, BlackBoxModel): # type: ignore
+            return BlackBoxGenerator(model, output_type) # type: ignore
         else:
-            return SteerableGenerator(model, output_type)
+            raise ValueError(
+                "The model argument must be an instance of "
+                "SteerableModel, BlackBoxModel or AsyncBlackBoxModel"
+            )
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 6ffc62320..39638a2fb 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -8,16 +8,16 @@
 
 from typing import Union
 
-from .anthropic import from_anthropic, Anthropic
+from .anthropic import Anthropic, from_anthropic
 from .base import Model, ModelTypeAdapter
 from .dottxt import Dottxt, from_dottxt
-from .gemini import from_gemini, Gemini
+from .gemini import Gemini, from_gemini
 from .llamacpp import LlamaCpp, from_llamacpp
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import Ollama, from_ollama
-from .openai import from_openai, OpenAI
-from .sglang import from_sglang, SGLang, AsyncSGLang
-from .tgi import from_tgi, TGI, AsyncTGI
+from .openai import OpenAI, from_openai
+from .sglang import AsyncSGLang, SGLang, from_sglang
+from .tgi import AsyncTGI, TGI, from_tgi
 from .transformers import (
     Transformers,
     TransformerTokenizer,
@@ -27,21 +27,19 @@
 from .vllm_offline import VLLMOffline, from_vllm_offline
 from .vllm import AsyncVLLM, VLLM, from_vllm
 
-LogitsGenerator = Union[
-    Transformers, LlamaCpp, OpenAI, MLXLM, VLLMOffline, Ollama
-]
-
 SteerableModel = Union[LlamaCpp, MLXLM, Transformers, VLLMOffline]
 BlackBoxModel = Union[
     Anthropic,
-    AsyncTGI,
-    AsyncSGLang,
-    AsyncVLLM,
     Dottxt,
     Gemini,
     Ollama,
     OpenAI,
-    TGI,
     SGLang,
+    TGI,
     VLLM,
 ]
+AsyncBlackBoxModel = Union[
+    AsyncTGI,
+    AsyncSGLang,
+    AsyncVLLM,
+]
diff --git a/outlines/models/base.py b/outlines/models/base.py
index aa0f736c0..08beece18 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -1,4 +1,7 @@
+"""Base classes for all models and model type adapters."""
+
 from abc import ABC, abstractmethod
+from typing import Any, AsyncIterator, Iterator, Optional
 
 
 class ModelTypeAdapter(ABC):
@@ -13,7 +16,7 @@ class ModelTypeAdapter(ABC):
     """
 
     @abstractmethod
-    def format_input(self, model_input):
+    def format_input(self, model_input: Any) -> Any:
         """Format the user input to the expected format of the model.
 
         For API-based models, it typically means creating the `messages`
@@ -22,21 +25,40 @@ def format_input(self, model_input):
         This method is also used to validate that the input type provided by
         the user is supported by the model.
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        Any
+            The formatted input to be passed to the model.
+
         """
         ...
 
     @abstractmethod
-    def format_output_type(self, output_type):
+    def format_output_type(self, output_type: Optional[Any] = None) -> Any:
         """Format the output type to the expected format of the model.
 
         For API-based models, this typically means creating a `response_format`
         argument. For local models, it means formatting the logits processor to
         create the object type expected by the model.
 
+        Parameters
+        ----------
+        output_type
+            The output type provided by the user.
+
+        Returns
+        -------
+        Any
+            The formatted output type to be passed to the model.
+
         """
         ...
 
-
 class Model(ABC):
     """Base class for all synchronous models.
 
@@ -52,7 +74,12 @@ class Model(ABC):
     type_adapter: ModelTypeAdapter
     tensor_library_name: str
 
-    def __call__(self, model_input, output_type=None, **inference_kwargs):
+    def __call__(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
         """Call the model.
 
         Users can call the model directly, in which case we will create a
@@ -67,12 +94,31 @@ def __call__(self, model_input, output_type=None, **inference_kwargs):
         model("prompt", Foo)
         ```
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
         """
         from outlines import Generator
 
         return Generator(self, output_type)(model_input, **inference_kwargs)
 
-    def stream(self, model_input, output_type=None, **inference_kwargs):
+    def stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Iterator[Any]:
         """Stream a response from the model.
 
         Users can use the `stream` method from the model directly, in which
@@ -90,34 +136,86 @@ def stream(self, model_input, output_type=None, **inference_kwargs):
             print(chunk)
         ```
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Iterator[Any]
+            A stream of responses from the model.
+
         """
         from outlines import Generator
 
-        return Generator(self, output_type).stream(model_input, **inference_kwargs)
+        generator = Generator(self, output_type)
+        return generator.stream(model_input, **inference_kwargs) # type: ignore
 
     @abstractmethod
-    def generate(self, model_input, output_type=None, **inference_kwargs):
+    def generate(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
         """Generate a response from the model.
 
         The output_type argument contains a logits processor for local models
         while it contains a type (Json, Enum...) for the API-based models.
         This method is not intended to be used directly by end users.
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
         """
         ...
 
     @abstractmethod
-    def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+    def generate_stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Iterator[Any]:
         """Generate a stream of responses from the model.
 
         The output_type argument contains a logits processor for local models
         while it contains a type (Json, Enum...) for the API-based models.
         This method is not intended to be used directly by end users.
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Iterator[Any]
+            A stream of responses from the model.
+
         """
         ...
 
-
 class AsyncModel(ABC):
     """Base class for all asynchronous models.
 
@@ -133,7 +231,12 @@ class AsyncModel(ABC):
     type_adapter: ModelTypeAdapter
     tensor_library_name: str
 
-    async def __call__(self, model_input, output_type=None, **inference_kwargs):
+    async def __call__(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
         """Call the model.
 
         Users can call the model directly, in which case we will create a
@@ -148,13 +251,32 @@ async def __call__(self, model_input, output_type=None, **inference_kwargs):
         await model("prompt", Foo)
         ```
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
         """
         from outlines import Generator
 
         generator = Generator(self, output_type)
         return await generator(model_input, **inference_kwargs)
 
-    async def stream(self, model_input, output_type=None, **inference_kwargs):
+    async def stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> AsyncIterator[Any]:
         """Stream a response from the model.
 
         Users can use the `stream` method from the model directly, in which
@@ -172,32 +294,84 @@ async def stream(self, model_input, output_type=None, **inference_kwargs):
             print(chunk)
         ```
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        AsyncIterator[Any]
+            A stream of responses from the model.
+
         """
         from outlines import Generator
 
         generator = Generator(self, output_type)
 
-        async for chunk in generator.stream(model_input, **inference_kwargs):  # pragma: no cover
+        async for chunk in generator.stream(model_input, **inference_kwargs):  # type: ignore
             yield chunk
 
     @abstractmethod
-    async def generate(self, model_input, output_type=None, **inference_kwargs):
+    async def generate(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> Any:
         """Generate a response from the model.
 
         The output_type argument contains a logits processor for local models
         while it contains a type (Json, Enum...) for the API-based models.
         This method is not intended to be used directly by end users.
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        Any
+            The response generated by the model.
+
         """
         ...
 
     @abstractmethod
-    async def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+    async def generate_stream(
+        self,
+        model_input: Any,
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> AsyncIterator[Any]:
         """Generate a stream of responses from the model.
 
         The output_type argument contains a logits processor for local models
         while it contains a type (Json, Enum...) for the API-based models.
         This method is not intended to be used directly by end users.
 
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        AsyncIterator[Any]
+            A coroutine that will produce an async iterator of responses from the model.
+
         """
         ...
diff --git a/tests/test_applications.py b/tests/test_applications.py
index 490dee438..5fed7bf23 100644
--- a/tests/test_applications.py
+++ b/tests/test_applications.py
@@ -1,39 +1,28 @@
-import pytest
+from typing import Any
 
 import jinja2
+import pytest
+import transformers
 
+from outlines import from_transformers
 from outlines.applications import Application
 from outlines.templates import Template
-from outlines.models import Model
-from typing import Any
 
 
 @pytest.fixture(scope="session")
 def model():
-    class MockModel(Model):
-        type_adapter = None
-
-        def generate(self, model_input: str, output_type=None, **kwargs) -> Any:
-            return model_input
-
-        def generate_stream(self, model_input: str, output_type=None, **kwargs) -> Any:
-            return model_input
-
-    return MockModel()
+    return from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("gpt2"),
+        transformers.AutoTokenizer.from_pretrained("gpt2"),
+    )
 
 
 @pytest.fixture(scope="session")
 def another_model():
-    class MockModel(Model):
-        type_adapter = None
-
-        def generate(self, model_input: str, output_type=None, **kwargs) -> Any:
-            return model_input
-
-        def generate_stream(self, model_input: str, output_type=None, **kwargs) -> Any:
-            return model_input
-
-    return MockModel()
+    return from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("gpt2"),
+        transformers.AutoTokenizer.from_pretrained("gpt2"),
+    )
 
 
 def test_application_initialization():
@@ -60,9 +49,9 @@ def test_application_template_call(model):
     template = Template.from_string("Test {{ value }}")
     output_type = None
     application = Application(template, output_type)
-    result = application(model, {"value": "example"}, foo="bar")
+    result = application(model, {"value": "example"}, max_new_tokens=10)
 
-    assert result == "Test example"
+    assert isinstance(result, str)
 
 
 def test_application_callable_call(model):
@@ -71,9 +60,9 @@ def template(value):
 
     output_type = None
     application = Application(template, output_type)
-    result = application(model, {"value": "example"}, foo="bar")
+    result = application(model, {"value": "example"}, max_new_tokens=10)
 
-    assert result == "Test example"
+    assert isinstance(result, str)
 
 
 def test_application_template_error(model):
@@ -90,15 +79,15 @@ def test_application_generator_reuse(model, another_model):
     output_type = None
     application = Application(template, output_type)
 
-    application(model, {"value": "example"})
+    application(model, {"value": "example"}, max_new_tokens=10)
     first_generator = application.generator
     first_model = application.model
 
-    application(model, {"value": "example"})
+    application(model, {"value": "example"}, max_new_tokens=10)
     assert application.model == first_model
     assert application.generator == first_generator
 
-    application(another_model, {"value": "example"})
+    application(another_model, {"value": "example"}, max_new_tokens=10)
     assert application.model == another_model
     assert application.model != first_model
     assert application.generator != first_generator
diff --git a/tests/test_generator.py b/tests/test_generator.py
index 824463c93..5ce54dc4b 100644
--- a/tests/test_generator.py
+++ b/tests/test_generator.py
@@ -207,6 +207,11 @@ async def test_async_black_box_generator_stream(black_box_async_model):
 # Generator
 
 
+def test_generator_init_no_model():
+    with pytest.raises(ValueError):
+        Generator(None, Literal["foo", "bar"])
+
+
 def test_generator_init_multiple_output_type(steerable_model, sample_processor):
     with pytest.raises(ValueError):
         Generator(steerable_model, Literal["foo", "bar"], processor=sample_processor)

From 7eb1f36fa865e95afbcf6010fe4d7b8c403a01eb Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 27 May 2025 13:57:46 +0200
Subject: [PATCH 338/505] Various docstrings and type hints improvements

---
 outlines/applications.py                      | 50 +++++++++++++---
 outlines/caching.py                           |  2 +
 outlines/fsm/__init__.py                      |  1 +
 outlines/fsm/parsing.py                       |  2 +
 outlines/grammars.py                          | 23 +++++++-
 outlines/templates.py                         | 58 +++++++++++++------
 .../v0_legacy/models/transformers_vision.py   |  3 -
 7 files changed, 109 insertions(+), 30 deletions(-)

diff --git a/outlines/applications.py b/outlines/applications.py
index 1a7dd087d..4b4b34187 100644
--- a/outlines/applications.py
+++ b/outlines/applications.py
@@ -1,8 +1,14 @@
+"""Encapsulate a prompt template and an output type into a reusable object."""
+
 from typing import Any, Callable, Dict, Optional, Union
 
-from outlines.generator import BlackBoxGenerator, Generator, SteerableGenerator
+from outlines.generator import (
+    BlackBoxGenerator,
+    Generator,
+    SteerableGenerator,
+)
+from outlines.models.base import Model
 from outlines.templates import Template
-from outlines.models import BlackBoxModel, SteerableModel
 
 
 class Application:
@@ -44,19 +50,49 @@ class OutputModel(BaseModel):
     result = application(model, {"num": 3}, max_new_tokens=20)
     print(result)  # Expected output: { "result" : 6 }
     ```
+
     """
-    def __init__(self, template: Union[Template, Callable], output_type: Any):
+    def __init__(
+        self,
+        template: Union[Template, Callable],
+        output_type: Optional[Any] = None,
+    ):
+        """
+        Parameters
+        ----------
+        template
+            The template to use to build the prompt.
+        output_type
+            The output type provided to the generator.
+
+        """
         self.template = template
         self.output_type = output_type
-        self.model: Optional[Union[BlackBoxModel, SteerableModel]] = None
-        self.generator: Optional[Union[BlackBoxGenerator, SteerableGenerator]] = None
+        self.generator: Optional[Union[
+            BlackBoxGenerator, SteerableGenerator
+        ]] = None
+        self.model: Optional[Model] = None
 
     def __call__(
         self,
-        model: Union[BlackBoxModel, SteerableModel],
+        model: Model,
         template_vars: Dict[str, Any],
         **inference_kwargs
-    ):
+    ) -> Any:
+        """
+        Parameters
+        ----------
+        model
+            The model to use to generate the response.
+        template_vars
+            The variables to be substituted in the template.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+        Returns
+        -------
+        Any
+            The generated response.
+        """
         if model is None:
             raise ValueError("you must provide a model")
         # We save the generator to avoid creating a new one for each call.
diff --git a/outlines/caching.py b/outlines/caching.py
index 5e9d5b679..7cb2ed169 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -1,3 +1,5 @@
+"""Caching and memoization of function calls."""
+
 import asyncio
 import contextlib
 import functools
diff --git a/outlines/fsm/__init__.py b/outlines/fsm/__init__.py
index e69de29bb..4e38718ef 100644
--- a/outlines/fsm/__init__.py
+++ b/outlines/fsm/__init__.py
@@ -0,0 +1 @@
+"""Finite state machines builder for CFG generation."""
diff --git a/outlines/fsm/parsing.py b/outlines/fsm/parsing.py
index e48fb69e4..3831c57c1 100644
--- a/outlines/fsm/parsing.py
+++ b/outlines/fsm/parsing.py
@@ -1,3 +1,5 @@
+"""Build a parser from a grammar to create a finite state machine."""
+
 from copy import copy, deepcopy
 from dataclasses import dataclass
 from functools import lru_cache
diff --git a/outlines/grammars.py b/outlines/grammars.py
index f0c122964..e2050148a 100644
--- a/outlines/grammars.py
+++ b/outlines/grammars.py
@@ -1,10 +1,29 @@
+"""A few common Lark grammars."""
+
 from pathlib import Path
 
 GRAMMAR_PATH = Path(__file__).parent / "grammars"
 
 
-def read_grammar(grammar_file_name, base_grammar_path=GRAMMAR_PATH):
-    """Read grammar file from default grammar path"""
+def read_grammar(
+    grammar_file_name: str,
+    base_grammar_path: Path = GRAMMAR_PATH,
+) -> str:
+    """Read grammar file from default grammar path.
+
+    Parameters
+    ----------
+    grammar_file_name
+        The name of the grammar file to read.
+    base_grammar_path
+        The path to the directory containing the grammar file.
+
+    Returns
+    -------
+    str
+        The contents of the grammar file.
+
+    """
     full_path = base_grammar_path / grammar_file_name
     with open(full_path) as file:
         return file.read()
diff --git a/outlines/templates.py b/outlines/templates.py
index a1df14aa8..cd6a4fbf5 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -1,3 +1,5 @@
+"""Create templates to easily build prompts."""
+
 import base64
 import functools
 import inspect
@@ -12,16 +14,25 @@
 import warnings
 
 import jinja2
-from jinja2 import Environment, StrictUndefined
 from PIL import Image
-import pydantic
 from pydantic import BaseModel
 
 
 @dataclass
 class Vision:
-    """Represents the input for a vision model."""
+    """Contains the input for a vision model.
+
+    Provide an instance of this class as the `model_input` argument to a model
+    that supports vision.
+
+    Parameters
+    ----------
+    prompt
+        The prompt to use to generate the response.
+    image
+        The image to use to generate the response.
 
+    """
     prompt: str
     image: Image.Image
 
@@ -29,7 +40,9 @@ def __post_init__(self):
         image = self.image
 
         if not image.format:
-            raise TypeError("Could not read the format of the image passed to the model.")
+            raise TypeError(
+                "Could not read the format of the image passed to the model."
+            )
 
         buffer = BytesIO()
         image.save(buffer, format=image.format)
@@ -45,7 +58,6 @@ class Template:
     template can be accessed by callers.
 
     """
-
     template: jinja2.Template
     signature: Optional[inspect.Signature]
 
@@ -54,7 +66,8 @@ def __call__(self, *args, **kwargs) -> str:
 
         Returns
         -------
-        The rendered template as a Python ``str``.
+        str
+            The rendered template as a Python string.
 
         """
         if self.signature is not None:
@@ -66,7 +79,8 @@ def __call__(self, *args, **kwargs) -> str:
 
     @classmethod
     def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
-        """Create a `Template` instance from a string containing a Jinja template.
+        """Create a `Template` instance from a string containing a Jinja
+        template.
 
         Parameters
         ----------
@@ -75,16 +89,20 @@ def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
 
         Returns
         -------
-        An instance of the class with the provided content as a template.
+        Template
+            An instance of the class with the provided content as a template.
+
         """
         return cls(build_template_from_string(content, filters), None)
 
     @classmethod
     def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
-        """Create a `Template` instance from a file containing a Jinja template.
+        """Create a `Template` instance from a file containing a Jinja
+        template.
 
-        Note: This method does not allow to include and inheritance to reference files
-        that are outside the folder or subfolders of the file given to `from_file`.
+        Note: This method does not allow to include and inheritance to
+        reference files that are outside the folder or subfolders of the file
+        given to `from_file`.
 
         Parameters
         ----------
@@ -94,10 +112,14 @@ def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
         Returns
         -------
         Template
-            An instance of the Template class with the template loaded from the file.
+            An instance of the Template class with the template loaded from the
+            file.
+
         """
-        # We don't use a `Signature` here because it seems not feasible to infer one from a Jinja2 environment that is
-        # split across multiple files (since e.g. we support features like Jinja2 includes and template inheritance)
+        # We don't use a `Signature` here because it seems not feasible to
+        # infer one from a Jinja2 environment that is
+        # split across multiple files (since e.g. we support features like
+        # Jinja2 includes and template inheritance)
         return cls(build_template_from_file(path, filters), None)
 
 
@@ -231,8 +253,8 @@ def create_jinja_env(
 
     Users may pass additional filters, and/or override existing ones.
 
-    Arguments
-    ---------
+    Parameters
+    ----------
     loader
        An optional `BaseLoader` instance
     filters
@@ -345,8 +367,8 @@ def get_schema_dict(model: Dict):
     return json.dumps(model, indent=2)
 
 
-@get_schema.register(type(pydantic.BaseModel))
-def get_schema_pydantic(model: Type[pydantic.BaseModel]):
+@get_schema.register(type(BaseModel))
+def get_schema_pydantic(model: Type[BaseModel]):
     """Return the schema of a Pydantic model."""
     if hasattr(model, "model_json_schema"):
         def_key = "$defs"
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
index 83008fcbe..1c7657a19 100644
--- a/outlines/v0_legacy/models/transformers_vision.py
+++ b/outlines/v0_legacy/models/transformers_vision.py
@@ -9,9 +9,6 @@
     TransformersMultiModalTypeAdapter,
 )
 
-if TYPE_CHECKING:
-    from outlines.processors import OutlinesLogitsProcessor
-
 
 class TransformersVision(Transformers):
     def __init__(self, model, tokenizer, processor):

From eb6efb24387612507c87a137fa66892589836f99 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 27 May 2025 15:08:56 +0200
Subject: [PATCH 339/505] Add the section for output types in the features
 documentation

---
 docs_new/features/core/output_types.md | 205 +++++++++++++++++++++++++
 mkdocs.yml                             |  12 +-
 2 files changed, 212 insertions(+), 5 deletions(-)
 create mode 100644 docs_new/features/core/output_types.md

diff --git a/docs_new/features/core/output_types.md b/docs_new/features/core/output_types.md
new file mode 100644
index 000000000..3a49c2623
--- /dev/null
+++ b/docs_new/features/core/output_types.md
@@ -0,0 +1,205 @@
+---
+title: Output Types
+---
+
+# Output Types
+
+Outlines provides a simple and intuitive way of defining the output structure of text generation. Possible output formats include basic Python types, multiple-choices, JSON schemas, regular expressions and context-free grammars.
+
+## Overview
+
+Outlines models accept a __prompt__ and an __output type__ when they are invoked, as well as additional inference keyword arguments that are forwarded on to the underlying model.
+
+Output types can be from the general Python ecosystem, including:
+- Most native Python types, such as `int` or `str`
+- Types from the `typing` module, such as `Literal`, `List`, `Dict`, `Enum`, etc
+- Types from popular theird party libraries such as Pydantic or GenSON.
+
+Outlines also provides special classes for certain output structures (more details below):
+- JSON schemas with `JsonSchema`
+- Regular expressions with `Regex`
+- Context-free grammars with `CFG`
+
+The general idea is that you should provide as an output type what you would give as the type hint of the return type of a function.
+
+Consider the following functions for instance:
+
+```python
+from datetime import date
+from typing import Dict, List, Literal, Union
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    birth_date: date
+    skills: Union[Dict, List[str]]
+
+def give_int() -> int:
+    ...
+
+def pizza_or_burger() -> Literal["pizza", "burger"]:
+    ...
+
+def create_character() -> Character:
+    ...
+```
+
+With an Outlines model, you can generate text that respects the type hints above by providing those as the output type:
+
+```python
+model("How many minutes are there in one hour", int) # "60"
+model("Pizza or burger", Literal["pizza", "burger"]) # "pizza"
+model("Create a character", Character, max_new_tokens=100) # '{"name": "James", "birth_date": "1980-05-10)", "skills": ["archery", "negociation"]}'
+```
+
+An important difference with function type hints though is that an Outlines generator always returns a string.
+You have to cast the response into the type you want yourself.
+
+For instance:
+
+```python
+result = model("Create a character", Character, max_new_tokens=100)
+casted_result = Character.model_validate_json(result)
+print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
+print(casted_result) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
+```
+
+## Output Type Categories
+
+We can group possible output types in several categories based on the use case they correspond to. While most of those types are native python or types coming from well-known third-party libraries, there are three Outlines-specific types: `JsonSchema`, `Regex` and `CFG`. Their use is explained below.
+
+### Basic Python Types
+
+The most straightfoward form of structed generation is to return an answer that conforms to a given basic type such as an int or a python list. You can use the basic Python types and the types from the `typing` library. For instance:
+
+```python
+from typing import Dict
+
+output_type = float # example of valid value: "0.05"
+output_type = bool # example of valid value: "True"
+output_type = Dict[int, str] # example of valid value: "{1: 'hello', 2: 'there'}"
+```
+
+You can combine types to create more complex response formats by relying on collection types and types such as `Union` and `Optional`. Let's consider for instance the output type below used to represent semi-structured data:
+
+```python
+from typing import Dict, List, Optional, Tuple, Union
+
+output_type = Dict[str, Union[int, str, List[Tuple[str, Optional[float]]]]]
+```
+
+Values created with this output type would be dictionnaries with string as keys and values made of either an integer, a string or a list of two elements tuples: a string and either a float or None. Example of a valid response for text generated with this output type (it would be contained in a string):
+
+```json
+{
+    "name": "Alice",
+    "age": 30,
+    "metrics": [("engagement", 0.85), ("satisfaction", None)]
+}
+```
+
+### Multiple Choices
+
+Outlines supports multiple choice classification by using the `Literal` or `Enum` output types. For instance:
+
+```python
+from enum import Enum
+from typing import Literal
+
+class PizzaOrBurger(Enum):
+    pizza = "pizza
+    burger = "burger
+
+# Equivalent multiple-choice output types
+output_type = Literal["pizza", "burger"]
+output_type = PizzaOrBurger
+```
+
+### JSON Schemas
+
+Multiple different common Python types are often used to store information equivalent to a JSON schema. The following can be used in Outlines to generate text that respects a JSON schema:
+
+- A Pydantic class
+- A Dataclass
+- A TypedDict
+- A [GenSON](https://github.com/wolverdude/GenSON) `SchemaBuilder`
+- A Callable (the parameters are turned into the keys and the type hinting is used to define the types of the values)
+
+For instance:
+
+```python
+from dataclasses import dataclass
+
+@dataclass
+class Character:
+    name: str
+    age: int
+
+ouput_type = Character
+
+def character(name: str, age: int):
+    return None
+
+output_type = character
+```
+
+There are two other JSON schema formats that require Outlines-specific classes: JSON schema strings and dictionaries.
+
+As those are contained in regular Python strings or dictionnaries, the associated output format would be ambiguous if they were to be provided directly. As a result, Outlines requires them to be wrapped in a `outlines.types.JsonSchema` object. For instance:
+
+```python
+from outlines.types import JsonSchema
+
+schema_string = '{"type": "object", "properties": {"answer": {"type": "number"}}}'
+output_type = JsonSchema(schema_string)
+
+schema_dict = {
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+output_type = JsonSchema(schema_dict)
+```
+
+### Regex Patterns
+
+Outlines provides support for text generation constrained by regular expressions. Since regular expressions are expressed as simple raw string literals, regex strings must wrapped in an `outlines.types.Regex` object to clarify the expected return type.
+
+```python
+from outlines.types import Regex
+
+regex = r"[0-9]{3}"
+output_type = Regex(regex)
+```
+
+The `outlines.types` module contains a few common regex patterns stored in variables you can import and directly use as output types. Common patterns include a sentence, an email address and an [ISBN reference](https://en.wikipedia.org/wiki/ISBN). For instance:
+
+```python
+from outlines.types import sentence
+
+print(type(sentence)) # outlines.types.dsl.Regex
+print(sentence.pattern) # [A-Z].*\s*[.!?]
+```
+
+To help you create complex regex patterns yourself, you can use the Outlines [regex DSL](../../utility/regex_dsl).
+
+### Context-Free Grammars
+
+Outlines allows you to generate text that respects the syntax of a context-free grammar. Context-free grammars are defined using [Lark](https://lark-parser.readthedocs.io/en/latest/index.html), a grammar language. Since grammars are expressed as a string, Large CFG strings should be be wrapped in an `outlines.types.CFG` object. For instance:
+
+```python
+from outlines.types import CFG
+
+grammar_string = """
+    start: expr
+    expr: "{" expr "}" | "[" expr "]" |
+"""
+output_type = CFG(grammar_string)
+```
+
+You can find a few Lark grammar examples in the [grammars module](../../api_reference/grammars.md).
+
+## Output type availability
+
+The output types presented above are not available for all models as some have only limited support for structured outputs. Please refer to the documentation of the specific model you wish to use to know what output types it supports.
diff --git a/mkdocs.yml b/mkdocs.yml
index f09d31747..c423c9f47 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -182,14 +182,16 @@ nav:
                 - Migrating to v1: user_guide/migration.md
     - Features:
           - Overview: features/index.md
-          - Format types:
-                - Overview: features/types/index.md
-                - JSON: features/types/json_schema.md
-                - Regex: features/types/regex.md
-                - Context-free grammars: features/types/cfg.md
 
           - Core:
                 - Generator: features/core/generator.md
+                - Output Types:
+                    - Overview: features/core/output_types.md
+                    - Basic Types: features/core/output_types#basic-python-types
+                    - Multiple-Choices: features/core/output_types#multiple-choices
+                    - JSON: features/core/output_types#json-schemas
+                    - Regex: features/core/output_types#regex-patterns
+                    - Context-free Grammars: features/core/output_types#context-free-grammars
 
           - Supported Models:
                 - Anthropic: features/models/anthropic.md

From 9fd653d7e5b53acd99c6a3d013e036c2c067b15f Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 30 May 2025 15:38:46 +0200
Subject: [PATCH 340/505] Update the model documentation in the Features
 section

---
 docs_new/features/models/anthropic.md         |  98 ++++++
 docs_new/features/models/dottxt.md            |  74 +++++
 docs_new/features/models/gemini.md            | 144 +++++++++
 docs_new/features/models/index.md             | 108 ++++++-
 docs_new/features/models/llamacpp.md          | 280 +++++++----------
 docs_new/features/models/mlxlm.md             | 186 +++++++++---
 docs_new/features/models/models.md            |  56 ----
 docs_new/features/models/ollama.md            | 103 +++++++
 docs_new/features/models/openai.md            | 236 +++++---------
 docs_new/features/models/sglang.md            | 171 +++++++++++
 docs_new/features/models/tgi.md               | 138 ++++++++-
 docs_new/features/models/transformers.md      | 225 +++++++++-----
 .../models/transformers_multimodal.md         |  89 ++++++
 docs_new/features/models/vllm.md              | 287 ++++++++----------
 docs_new/features/models/vllm_offline.md      | 170 +++++++++++
 mkdocs.yml                                    |  32 +-
 16 files changed, 1702 insertions(+), 695 deletions(-)
 delete mode 100644 docs_new/features/models/models.md
 create mode 100644 docs_new/features/models/transformers_multimodal.md

diff --git a/docs_new/features/models/anthropic.md b/docs_new/features/models/anthropic.md
index e69de29bb..f9b648ccc 100644
--- a/docs_new/features/models/anthropic.md
+++ b/docs_new/features/models/anthropic.md
@@ -0,0 +1,98 @@
+---
+title: Anthropic
+---
+
+# Anthropic
+
+!!! Installation
+
+    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines: `pip install anthropic`.
+
+    You also need to have an Anthropic API key. This API key must either be set as an environment variable called `ANTHROPIC_API_KEY` or be provided to the `anthropic.Anthropic` class when instantiating it.
+
+## Model Initialization
+
+To create an Anthropic model instance, you can use the `from_anthropic` function. It takes 2 arguments:
+
+- `client`: an `anthropic.Anthropic` instance
+- `model_name`: the name of the model you want to use in subsequent model calls (optional)
+
+For instance:
+
+```python
+from anthropic import Anthropic
+import outlines
+
+# Create the Anthropic client
+client = Anthropic()
+
+# Create the model
+model = outlines.from_anthropic(
+    client,
+    "claude-3-5-sonnet-latest"
+)
+```
+
+Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-claude/models) for an up-to-date list of available models.
+
+## Text Generation
+
+Once you've created your Outlines `Anthropic` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+
+For instance:
+
+```python
+from anthropic import Anthropic
+import outlines
+
+# Create the model
+model = outlines.from_anthropic(
+    Anthropic(),
+    "claude-3-5-sonnet-latest"
+)
+
+# Call it to generate text
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
+```
+
+Anthropic does not support structured generation, so you cannot provide an output type. It does support streaming and vision inputs though.
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+from anthropic import Anthropic
+from outlines import from_anthropic
+from outlines.templates import Vision
+
+# Create the model
+model = from_anthropic(
+    Anthropic(),
+    "claude-3-5-sonnet-latest"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt
+prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+
+# Stream the response
+for chunk in model.stream(prompt, max_tokens=50):
+    print(chunk) # 'This...'
+```
+
+## Inference arguments
+
+When calling the model or streaming, you can provide keyword arguments that will be passed down to the Anthropic client. Make sure to include all the arguments you need to configure the client's behavior to your expected behavior. Some of the most common arguments include `max_tokens`, `temperature`, `stop_sequences` and `top_k`.
+
+See the [Anthropic API documentation](https://docs.anthropic.com/en/api/messages) for the full list of available arguments.
+
+!!! Warning
+
+    You must set a value for `max_tokens` with Anthropic models.
diff --git a/docs_new/features/models/dottxt.md b/docs_new/features/models/dottxt.md
index e69de29bb..09f06e5d5 100644
--- a/docs_new/features/models/dottxt.md
+++ b/docs_new/features/models/dottxt.md
@@ -0,0 +1,74 @@
+---
+title: Dottxt
+---
+
+# Dottxt
+
+!!! Installation
+
+    You need to install the `dottxt` python sdk to be able to use the Dottxt API in Outlines: `pip install dottxt`
+
+    You also need to have a Dottxt API key. This API key must either be set as an environment variable called `DOTTXT_API_KEY` or be provided to the `dottxt.client.Dottxt` class when instantiating it.
+
+## Model Initialization
+
+To create an Dottxt model instance, you can use the `from_dottxt` function. It takes 3 arguments:
+
+- `client`: a `dottxt.client.Dottxt` instance
+- `model_name`: the name of the model you want to use in subsequent model calls (optional)
+- `model_revision`: the name of the revision to use for the model selected (optional)
+
+For instance:
+
+```python
+from dottxt.client import Dottxt
+import outlines
+
+# Create client
+client = Dottxt(api_key="...")
+
+# Create the model
+model = outlines.from_dottxt(
+    client,
+    "meta-llama/Llama-3.1-8B",
+    "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b"
+)
+```
+
+Use the `list_models` method of the Dottxt client to get a list of available model names and revisions for your account.
+
+## Text Generation
+
+Dottxt only supports constrained generation with JSON schema output types. You must always provide a value for the `output_type` parameter as unconstrained generation is not available.
+
+For instance:
+
+```python
+from typing import List
+from pydantic import BaseModel
+from dottxt.client import Dottxt
+import outlines
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_dottxt(
+    Dottxt(),
+    "meta-llama/Llama-3.1-8B",
+    "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b"
+)
+
+# Generate structured text
+result = model("Create a character", Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+## Inference arguments
+
+You can provide the same optional parameters you would pass to the `dottxt` sdk's client both during the initialization of the `Dottxt` class and when generating text. Some of the common inference arguments include `max_tokens`, `frequency_penalty`, `presence_penalty` and `temperature`.
+
+Consult the [dottxt python sdk GitHub repository](https://github.com/dottxt-ai/dottxt-python) for the full list of parameters.
diff --git a/docs_new/features/models/gemini.md b/docs_new/features/models/gemini.md
index e69de29bb..5c1b1b949 100644
--- a/docs_new/features/models/gemini.md
+++ b/docs_new/features/models/gemini.md
@@ -0,0 +1,144 @@
+# Gemini
+
+!!! Installation
+
+    You need to install the `google-generativeai` library to be able to use the Gemini API in Outlines: `pip install google-generativeai`
+
+    You also need to have a Gemini API key. This API key must either be set as an environment variable called `GEMINI_API_KEY` or be provided to the `google.generativeai.GenerativeModel` class when instantiating it.
+
+## Model Initialization
+
+To create a Gemini model instance, you can use the `from_gemini` function. It takes 2 arguments:
+
+- `client`: a `google.generativeai.GenerativeModel` instance
+- `model_name`: the name of the model you want to use in subsequent model calls (optional)
+
+For instance:
+
+```python
+import outlines
+import google.generativeai as genai
+
+# Create the client
+client = genai.GenerativeModel()
+
+# Create the model
+model = outlines.from_gemini(
+    client,
+    "gemini-1-5-flash"
+)
+```
+
+Check the [Gemini documentation](https://github.com/googleapis/python-genai) for an up-to-date list of available models.
+
+## Text Generation
+
+Once you've created your Outlines `Gemini` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+import google.generativeai as genai
+
+# Create the model
+model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_tokens=20)
+print(result) # 'Riga'
+```
+
+Gemini also supports streaming. For instance:
+
+```python
+import outlines
+import google.generativeai as genai
+
+# Create the model
+model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=20):
+    print(chunk) # 'In...'
+```
+
+## Structured Generation
+
+Gemini provides supports for some forms of structured output: multiple choice, JSON schema (with caveats) and lists of structured objects. To use it, call the model with an `output_type` on top of your prompt.
+
+#### Multiple Choice
+
+```python
+import outlines
+import google.generativeai as genai
+from enum import Enum
+
+class PizzaOrBurger(Enum):
+    pizza = "pizza"
+    burger = "burger"
+
+# Create the model
+model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+
+# Call it with the ouput type to generate structured text
+result = model("Pizza or burger?", PizzaOrBurger, max_tokens=20)
+print(result) # 'pizza'
+```
+
+#### JSON Schema
+
+Gemini supports only three types of objects used to define a JSON Schema:
+
+- Pydantic classes
+- Dataclasses
+- TypedDicts
+
+```python
+import google.generativeai as genai
+import outlines
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+
+# Call it with the ouput type to generate structured text
+result = model("Create a character", Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+#### Lists of Structured Objects
+
+A specificity of Gemini is that, despite not supporting regex, it does support a list of structured objects as an output type. To use it, put any of three available types described above in the typing `List` class
+
+```python
+from dataclasses import dataclass
+from typing import List
+import google.generativeai as genai
+import outlines
+
+@dataclass
+class Character:
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+
+# Call it with the ouput type to generate structured text
+result = model("Create a character", List[Character])
+print(result) # '[{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}, {["name":...'
+```
+
+## Inference arguments
+
+You can provide the same optional parameters you would pass to the `google.generativeai.GenerativeModel` client both during the initialization of the Gemini model and when generating text. Some of the common inference arguments include `max_tokens`, `temperature`, and other generation parameters.
+
+Consult the [Google Generative AI documentation](https://github.com/googleapis/python-genai) for the full list of parameters.
diff --git a/docs_new/features/models/index.md b/docs_new/features/models/index.md
index 64f70dbc7..2c1b3323c 100644
--- a/docs_new/features/models/index.md
+++ b/docs_new/features/models/index.md
@@ -1 +1,107 @@
-- [Anthropic](/api/models/anthropic.md)
+---
+title: Models
+---
+
+# Models
+
+## Overview
+
+Outlines models are objects that wrap an inference client or engine. Models provide a standardized interface to generate structured text.
+
+!!! Warning
+
+    The model loading functions have been modified in v1. While they used to be called `<name_inference_library>`, they are now called `from_<name_inference_library>`. The model classes' names and `__init__` methods are left unchanged.
+
+
+All Outlines model classes have an associated loader function to facilitate initializing a model instance. The name of this function is `from_` plus the name of the model in lower-case letters. For instance, Outlines has a `Transformers` model and an associated `from_transformers` loader function. The parameters to load a model are specific to each provider, please consult the documentation of the model you want to use for more information.
+
+After having created a model instance, you can either directly call it to generate text or first create a reusable generator that you would then call. In either case, you can provide an `output_type` to constrain the format of the generation output. See the [output types section](../core/output_types.md) for more information on constrained generation.
+
+For instance:
+
+```python
+from outlines import from_transformers, Generator
+import transformers
+
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+)
+
+# Call it directly
+response = model("How many countries are there in the world", max_new_tokens=20)
+print(result) # 'There are 200 countries in the world.'
+
+# Call it directly with an output_type
+response = model("How many countries are there in the world", int, max_new_tokens=20)
+print(result) # '200'
+
+# Create a generator first and then call it
+generator = Generator(model, int)
+response = generator("How many countries are there in the world")
+print(result) # '200'
+```
+
+## Features Matrix
+
+| | [Anthropic](../../models/anthropic) | [Dottxt](../../models/dottxt) | [Gemini](../../models/gemini) | [LlamaCpp](../../models/llamacpp) | [MLXLM](../../models/mlxlm) | [Ollama](../../models/ollama) | [OpenAI](../../models/openai) | [SGLang](../../models/sglang) | [TGI](../../models/tgi) | [Transformers](../../models/transformers) | [Transformers MultiModal](../../models/transformers_multimodal) | [VLLM](../../models/vllm) | [VLLMOffline](../../models/vllm_offline) |
+|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
+| **Output Types** | | | | | | | | | | | | | |
+| Simple Types | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| JSON Schema | ❌ | ✅ | 🟠 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Multiple Choice | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Regex | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Grammar | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | 🟠 | ❌ | ✅ | ✅ | ✅ | ✅ |
+| **Generation Features** | | | | | | | | | | | | | |
+| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
+| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ |
+
+## Model Types
+
+Models can be divided into two categories: local models and server-based models.
+
+In the case of local models, the text generation happens within the inference library object used to instantite the model. This gives Outlines direct access to the generation process (through a logits processor) and means all structured generation output types are available.
+
+The local models available are the following:
+
+- LlamaCpp
+- MLXLM
+- Transformers
+- TransformersMultiModal
+- VLLMOffline
+
+In the case of server-based models, the model is initialized with a client that sends a request to a server that is in charge of the actual text generation. As a result, we have limited control over text generation and some output types are not supported. The server on which the text generation happens can either be remote (with OpenAI or Anthopic for instance) or local (with SGLang for instance).
+
+The server-based models available are the following:
+
+- Anthropic
+- Dottxt
+- Gemini
+- Ollama
+- OpenAI
+- SgLang
+- TGI
+- VLLM
+
+Some models have an async version. To use them, just pass the async version of the provider object to their loading function. It will then return a `Async<ModelName>` instance with the same methods and features as the regular sync instance.
+
+For instance:
+
+```python
+from outlines import from_tgi
+from huggingface_hub import AsyncInferenceClient
+
+model = from_tgi(
+    AsyncInferenceClient("http://localhost:8000/v1")
+)
+print(type(model)) # outlines.models.tgi.AsyncTGI
+```
+
+The models that have an async version are the following:
+
+- SgLang
+- TGI
+- VLLM
diff --git a/docs_new/features/models/llamacpp.md b/docs_new/features/models/llamacpp.md
index 51b62eca8..d269c4111 100644
--- a/docs_new/features/models/llamacpp.md
+++ b/docs_new/features/models/llamacpp.md
@@ -1,230 +1,168 @@
-# Llama.cpp
+---
+title: llama.cpp
+---
 
-Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library][llamacpp]. Llamacpp allows to run quantized models on machines with limited compute.
+# llama.cpp
 
-!!! Note "Installation"
+Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library](https://github.com/abetlen/llama-cpp-python). Llamacpp allows to run quantized models on machines with limited compute.
 
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration. See the [installation section](#installation) for instructions to install `llama-cpp-python` with CUDA, Metal, ROCm and other backends. To get started quickly you can also run:
+!!! Installation
 
-    ```bash
-    pip install "outlines[llamacpp]"
-    ```
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration: `pip install llama_cpp`
 
-## Load the model
+    See the [llama-cpp-python Github page](https://github.com/abetlen/llama-cpp-python?tab=readme-ov-file#supported-backends) for instructions on installing with CUDA, Metal, ROCm and other backends.
 
-You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 
-```python
-from outlines import models
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-```
+## Model Initialization
 
-This will download the model files to the hub cache folder and load the weights in memory.
+To load the model, you can use the `from_llamacpp` function. The single argument of the function is a `Llama` model instance from the `llama_cpp` library. Consult the [Llama class API reference](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama) for detailed information on how to create a model instance and on the various available parameters.
 
-You can also initialize the model by passing the path to the weights on your machine. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
+For instance:
 
 ```python
-from outlines import models
+import outlines
 from llama_cpp import Llama
 
-llm = Llama("./phi-2.Q4_K_M.gguf")
-model = models.LlamaCpp(llm)
-```
-
-If you need more control, you can pass the same keyword arguments to the model as you would pass in the [llama-ccp-library][llamacpp]:
-
-```python
-from outlines import models
-
-model = models.llamacpp(
-    "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf"
-    n_ctx=512,  # to set the context length value
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
 )
 ```
 
-**Main parameters:**
-
-| Parameters | Type | Description | Default |
-|------------|------|-------------|---------|
-| `n_gpu_layers`| `int` | Number of layers to offload to GPU. If -1, all layers are offloaded | `0` |
-| `split_mode` | `int` | How to split the model across GPUs. `1` for layer-wise split, `2` for row-wise split | `1` |
-| `main_gpu` | `int` | Main GPU | `0` |
-| `tensor_split` | `Optional[List[float]]` | How split tensors should be distributed across GPUs. If `None` the model is not split. | `None` |
-| `n_ctx` | `int` | Text context. Inference from the model if set to `0` | `0` |
-| `n_threads` | `Optional[int]` | Number of threads to use for generation. All available threads if set to `None`.| `None` |
-| `verbose` | `bool` | Print verbose outputs to `stderr` | `False` |
-
-See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters.
+## Text Generation
 
-### Load the model on GPU
+To generate text, you can simply call the model with a prompt.
 
-!!! Note
-
-    [Make sure](#cuda) that you installed `llama-cpp-python` with GPU support.
-
- To load the model on GPU, pass `n_gpu_layers=-1`:
+For instance:
 
 ```python
-from outlines import models
+import outlines
+from llama_cpp import Llama
 
-model = models.llamacpp(
-    "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf",
-    n_gpu_layers=-1,  # to use GPU acceleration
+# Create the model
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
 )
-```
-
-This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
-
-### Load LoRA adapters
-
-You can load LoRA adapters dynamically:
-
-```python
-from outlines import models, generate
-
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-generator = generate.text(model)
-answer_1 = generator("prompt")
 
-model.load_lora("./path/to/adapter.gguf")
-answer_2 = generator("prompt")
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_tokens=20)
+print(result) # 'Riga'
 ```
 
-To load another adapter you need to re-initialize the model. Otherwise the adapter will be added on top of the previous one:
+The `LlamaCpp` model also supports streaming. For instance:
 
 ```python
-from outlines import models
+import outlines
+from llama_cpp import Llama
 
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-model.load_lora("./path/to/adapter1.gguf")  # Load first adapter
+# Create the model
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+)
 
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-model.load_lora("./path/to/adapter2.gguf")  # Load second adapter
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In...'
 ```
 
-## Generate text
+## Structured Generation
+
+The `LlamaCpp` model supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
 
-In addition to the parameters described in the [text generation section](../text.md) you can pass extra keyword arguments, for instance to set sampling parameters not exposed in Outlines' public API:
+### Basic Type
 
 ```python
-from outlines import models, generate
+import outlines
+from llama_cpp import Llama
 
+output_type = int
 
-model = models.llamacpp("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf")
-generator = generate.text(model)
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+)
 
-answer = generator("A prompt", presence_penalty=0.8)
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
 ```
 
-**Extra keyword arguments:**
-
-The value of the keyword arguments you pass to the generator suspersede the values set when initializing the sampler or generator. All extra sampling methods and repetition penalties are disabled by default.
+### JSON Schema
 
-| Parameters | Type | Description | Default |
-|------------|------|-------------|---------|
-| `suffix` | `Optional[str]` | A suffix to append to the generated text. If `None` no suffix is added. | `None` |
-| `echo` | `bool` | Whether to preprend the prompt to the completion. | `False` |
-| `seed` | `int` | The random seed to use for sampling. | `None` |
-| `max_tokens` | `Optional[int]` | The maximum number of tokens to generate. If `None` the maximum number of tokens depends on `n_ctx`. | `16` |
-| `frequence_penalty` | `float` | The penalty to apply to tokens based on their frequency in the past 64 tokens. | `0.0` |
-| `presence_penalty` | `float` | The penalty to apply to tokens based on their presence in the past 64 tokens. | `0.0` |
-| `repeat_penalty` | `float` | The penalty to apply to repeated tokens in the past 64 tokens. | `1.` |
-| `stopping_criteria` | `Optional[StoppingCriteriaList]` | A list of stopping criteria to use. | `None`
-| `logits_processor` | `Optional[LogitsProcessorList]` | A list of logits processors to use. The logits processor used for structured generation will be added to this list. | `None`
-| `temperature` | `float` | The temperature to use for sampling | `1.0` |
-| `top_p` | `float` | The top-p value to use for [nucleus sampling][degeneration]. | `1.` |
-| `min_p` | `float` | The min-p value to use for [minimum-p sampling][minimum-p]. | `0.` |
-| `typical_p` | `float` | The p value to use for [locally typical sampling][locally-typical]. | `1.0` |
-| `stop` | `Optional[Union[str, List[str]]]` | A list of strings that stop generation when encountered. | `[]` |
-| `top_k` |  `int` | The top-k value used for [top-k sampling][top-k]. Negative value to consider all logit values. | `-1.` |
-| `tfs_z` | `float` | The [tail-free sampling][tail-free] parameter. | `1.0` |
-| `mirostat_mode` | `int` | The [mirostat sampling][mirostat] mode. | `0` |
-| `mirostat_tau` | `float` | The target cross-entropy for [mirostat sampling][mirostat].| `5.0` |
-| `mirostat_eta` | `float` | The learning rate used to update `mu` in [mirostat sampling][mirostat]. | `0.1` |
-
-See the [llama-cpp-python documentation][llama-cpp-python-call] for the full and up-to-date list of parameters and the [llama.cpp code][llama-cpp-sampling-params] for the default values of other
-sampling parameters.
-
-### Streaming
-
-
-## Installation
-
-You need to install the `llama-cpp-python` library to use the llama.cpp integration.
+```python
+from typing import List
+from pydantic import BaseModel
+import outlines
+from llama_cpp import Llama
 
-### CPU
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
 
-For a *CPU-only* installation run:
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+)
 
-```bash
-pip install llama-cpp-python
+result = model("Create a character.", output_type=Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
 
-!!! Warning
-
-    Do not run this command if you want support for BLAS, Metal or CUDA. Follow the instructions below instead.
+### Multiple Choice
 
-### CUDA
+```python
+from typing import Literal
+import outlines
+from llama_cpp import Llama
 
-```bash
-CMAKE_ARGS="-DLLAMA_CUDA=on" pip install llama-cpp-python
-```
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
-It is also possible to install pre-built wheels with CUDA support (Python 3.10 and above):
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+)
 
-```bash
-pip install llama-cpp-python \
-  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/<cuda-version>
+result = model("What is the capital of France?", output_type)
+print(result) # 'Paris'
 ```
 
-Where `<cuda-version>` is one of the following, depending on the version of CUDA installed on your system:
+### Regex
 
-- `cu121` for CUDA 12.1
-- `cu122` for CUDA 12.2
-- `cu123` CUDA 12.3
-
-### Metal
-
-```bash
-CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python
-```
-
-It is also possible to install pre-build wheels with Metal support (Python 3.10 or above, MacOS 11.0 and above):
+```python
+from outlines.types import Regex
+import outlines
+from llama_cpp import Llama
 
-```bash
-pip install llama-cpp-python \
-  --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/metal
-```
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
-### OpenBLAS
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    )
+)
 
-```bash
-CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
+result = model("Generate a fake social security number.", output_type)
+print(result) # '782-32-3789'
 ```
 
-### Other backend
-
-`llama.cpp` supports many other backends. Refer to the [llama.cpp documentation][llama-cpp-python-install] to use the following backends:
-
-- CLBast (OpenCL)
-- hipBLAS (ROCm)
-- Vulkan
-- Kompute
-- SYCL
-
-
+## Inference Arguments
 
+When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `__call__` method of the `llama_cpp.Llama` model. Some common inference arguments include `max_tokens`, `temperature`, `frequency_penalty` and `top_p`.
 
-[llamacpp]: https://github.com/abetlen/llama-cpp-python
-[llama-cpp-python-call]: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
-[llama-cpp-python-install]: https://github.com/abetlen/llama-cpp-python/tree/08b16afe11e7b42adec2fed0a781123383476045?tab=readme-ov-file#supported-backends
-[llama-cpp-sampling-params]: https://github.com/ggerganov/llama.cpp/blob/e11a8999b5690f810c2c99c14347f0834e68c524/common/sampling.h#L22
-[mirostat]: https://arxiv.org/abs/2007.14966
-[degeneration]: https://arxiv.org/abs/1904.09751
-[top-k]: https://arxiv.org/abs/1805.04833
-[minimum-p]: https://github.com/ggerganov/llama.cpp/pull/3841
-[locally-typical]: https://arxiv.org/abs/2202.00666
-[tail-free]: https://www.trentonbricken.com/Tail-Free-Sampling
+See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__) for more information on inference parameters.
diff --git a/docs_new/features/models/mlxlm.md b/docs_new/features/models/mlxlm.md
index d435b9c1f..67d5d0d5e 100644
--- a/docs_new/features/models/mlxlm.md
+++ b/docs_new/features/models/mlxlm.md
@@ -1,93 +1,187 @@
+---
+title: mlx-lm
+---
+
 # mlx-lm
 
 Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
 
 !!! Note "Installation"
 
-    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To get started quickly you can also run:
+    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To download the required libraries, run `pip install mlx mlx-lm`.
+
+## Model Initialization
 
-    ```bash
-    pip install "outlines[mlxlm]"
-    ```
+To create a MLXLM model instance, you can use the `from_mlxlm` function. It takes 2 arguments:
 
+- `model`: an `mlx.nn.Module` instance
+- `tokenizer`: a `transformers.PreTrainedTokenizer` instance
 
-## Load the model
+However, we recommend you simply pass on the output of the `mlx_lm.load` function (it takes a model name as an argument).
 
-You can initialize the model by passing the name of the repository on the HuggingFace Hub. The official repository for mlx-lm supported models is [mlx-community](https://huggingface.co/mlx-community).
+For instance:
 
 ```python
-from outlines import models
+import outlines
+import mlx_lm
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+# Create the model
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
 ```
 
-This will download the model files to the hub cache folder and load the weights in memory.
+## Text Generation
 
-The arguments `model_config` and `tokenizer_config` are available to modify loading behavior. For example, per the `mlx-lm` [documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms#supported-models), you must set an eos_token for `qwen/Qwen-7B`. In outlines you may do so via
+To generate text, you can simply call the model with a prompt.
 
-```
-model = models.mlxlm(
-    "mlx-community/Meta-Llama-3.1-8B-Instruct-8bit",
-    tokenizer_config={"eos_token": "<|endoftext|>", "trust_remote_code": True},
+For instance:
+
+```python
+import outlines
+import mlx_lm
+
+# Load the model
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_tokens=20)
+print(result) # 'Riga'
 ```
 
-**Main parameters:**
+The `MLXLM` model also supports streaming. For instance:
 
-(Subject to change. Table based on [mlx-lm.load docstring](https://github.com/ml-explore/mlx-examples/blob/main/llms/mlx_lm/utils.py#L429))
+```python
+import outlines
+import mlx_lm
 
-| Parameters         | Type   | Description                                                                                      | Default |
-|--------------------|--------|--------------------------------------------------------------------------------------------------|---------|
-| `tokenizer_config` | `dict` | Configuration parameters specifically for the tokenizer. Defaults to an empty dictionary.        | `{}`    |
-| `model_config`     | `dict` | Configuration parameters specifically for the model. Defaults to an empty dictionary.            | `{}`    |
-| `adapter_path`     | `str`  | Path to the LoRA adapters. If provided, applies LoRA layers to the model.                        | `None`  |
-| `lazy`             | `bool` | If False, evaluate the model parameters to make sure they are loaded in memory before returning. | `False` |
+# Load the model
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
 
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In...'
+```
 
-## Generate text
+## Structured Generation
 
-You may generate text using the parameters described in the [text generation documentation](../text.md).
+As a local model, `MLXLM` supports all forms of structured generation available in Outlines.
 
-With the loaded model, you can generate text or perform structured generation, e.g.
+### Basic Type
 
 ```python
-from outlines import models, generate
+import outlines
+import mlx_lm
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-generator = generate.text(model)
+output_type = int
 
-answer = generator("A prompt", temperature=2.0)
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
 ```
 
-## Streaming
+### JSON Schema
+
+```python
+from pydantic import BaseModel
+from typing import List
+import outlines
+import mlx_lm
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("Create a character.", output_type=Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
 
-You may creating a streaming iterable with minimal changes
+### Multiple Choice
 
 ```python
-from outlines import models, generate
+from typing import Literal
+import outlines
+import mlx_lm
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
-generator = generate.text(model)
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
-for token_str in generator.text("A prompt", temperature=2.0):
-    print(token_str)
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("What is the capital of France?", output_type)
+print(result) # 'Paris'
 ```
 
-## Structured
+### Regex
 
-You may perform structured generation with mlxlm to guarantee your output will match a regex pattern, json schema, or lark grammar.
+```python
+from outlines.types import Regex
+import outlines
+import mlx_lm
 
-Example: Phone number generation with pattern `"\\+?[1-9][0-9]{7,14}"`:
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("Generate a fake social security number.", output_type)
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
 
 ```python
-from outlines import models, generate
+from outlines.text import CFG
+import outlines
+import mlx_lm
+
+arithmetic_grammar = """
+?start: sum
 
-model = models.mlxlm("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit")
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
 
-phone_number_pattern = "\\+?[1-9][0-9]{7,14}"
-generator = generate.regex(model, phone_number_pattern)
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
 
-model_output = generator("What's Jennys Number?\n")
-print(model_output)
-# '8675309'
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
+
+model = outlines.from_mlxlm(
+    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+)
+
+result = model("Write an addition.", output_type, max_tokens=20)
+print(result) # '23 + 48'
 ```
+
+## Inference Arguments
+
+When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `mlx_lm.generate` function used to generate text.
+
+See the [MLXLM documentation](https://github.com/ml-explore/mlx-lm) for more information on inference parameters.
diff --git a/docs_new/features/models/models.md b/docs_new/features/models/models.md
deleted file mode 100644
index eab70f2a1..000000000
--- a/docs_new/features/models/models.md
+++ /dev/null
@@ -1,56 +0,0 @@
----
-title: Models
----
-
-# Models
-
-Outlines supports generation using a number of inference engines (`outlines.models`). Loading a model using outlines follows a similar interface between inference engines:
-
-```python
-import outlines
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-128k-instruct")
-model = outlines.models.transformers_vision("llava-hf/llava-v1.6-mistral-7b-hf")
-model = outlines.models.vllm("microsoft/Phi-3-mini-128k-instruct")
-model = outlines.models.llamacpp(
-    "microsoft/Phi-3-mini-4k-instruct-gguf", "Phi-3-mini-4k-instruct-q4.gguf"
-)
-model = outlines.models.exllamav2("bartowski/Phi-3-mini-128k-instruct-exl2")
-model = outlines.models.mlxlm("mlx-community/Phi-3-mini-4k-instruct-4bit")
-
-model = outlines.models.openai(
-    "gpt-4o-mini",
-    api_key=os.environ["OPENAI_API_KEY"]
-)
-```
-
-
-# Feature Matrix
-|                   | [Transformers](transformers.md) | [Transformers Vision](transformers_vision.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [ExLlamaV2](exllamav2.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md)* |
-|-------------------|--------------|---------------------|------|-----------|-----------|-------|---------|
-| **Device**        |              |                     |      |           |           |       |         |
-| Cuda              | ✅           | ✅                  | ✅   | ✅        | ✅        | ❌    | N/A     |
-| Apple Silicon     | ✅           | ✅                  | ❌   | ✅        | ✅        | ✅    | N/A     |
-| x86 / AMD64       | ✅           | ✅                  | ❌   | ✅        | ✅        | ❌    | N/A     |
-| **Sampling**      |              |                     |      |           |           |       |         |
-| Greedy            | ✅           | ✅                  | ✅   | ✅*       | ✅        | ✅    | ❌      |
-| Multinomial       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Multiple Samples  | ✅           | ✅                  |      | ❌        |           | ❌    | ✅      |
-| Beam Search       | ✅           | ✅                  | ✅   | ❌        | ✅        | ❌    | ❌      |
-| **Generation**    |              |                     |      |           |           |       |         |
-| Batch             | ✅           | ✅                  | ✅   | ❌        | ?         | ❌    | ❌      |
-| Stream            | ✅           | ❌                  | ❌   | ✅        | ?         | ✅    | ❌      |
-| Text              | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| **Structured**    | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| JSON Schema       | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Choice            | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ✅      |
-| Regex             | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
-| Grammar           | ✅           | ✅                  | ✅   | ✅        | ✅        | ✅    | ❌      |
-
-
-## Caveats
-
-- OpenAI doesn't support structured generation due to limitations in their API and server implementation.
-- `outlines.generate` ["Structured"](../generation/generation.md) includes methods such as `outlines.generate.regex`, `outlines.generate.json`, `outlines.generate.cfg`, etc.
-- MLXLM only supports Apple Silicon.
-- llama.cpp greedy sampling available via multinomial with `temperature = 0.0`.
diff --git a/docs_new/features/models/ollama.md b/docs_new/features/models/ollama.md
index e69de29bb..2f75a7621 100644
--- a/docs_new/features/models/ollama.md
+++ b/docs_new/features/models/ollama.md
@@ -0,0 +1,103 @@
+---
+title: Ollama
+---
+
+# Ollama
+
+!!! Installation
+
+    To be able to use Ollama in Outlines, you must install both Ollama and the `ollama` python sdk.
+
+    - To download Ollama: https://ollama.com/download
+    - To install the ollama python sdk: `pip install ollama`
+
+    Consult the [`ollama` documentation](https://github.com/ollama/ollama-python) for detailed information on installation and client initialization.
+
+## Model Initialization
+
+To create an Ollama model instance, you can use the `from_ollama` function. It takes 2 arguments:
+
+- `client`: an `ollama.Client` instance
+- `model_name`: the name of the model you want to use
+
+For instance:
+
+```python
+import ollama
+import outlines
+
+# Create the client
+client = ollama.Client()
+
+# Create the model
+model = outlines.from_ollama(
+    client,
+    "tinyllama",
+)
+```
+
+You can find the list of available models on the [Ollama library](https://ollama.com/library).
+
+## Text Generation
+
+Once you've created your Outlines `Ollama` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+
+For instance:
+
+```python
+import ollama
+import outlines
+
+# Create the model
+model = outlines.from_ollama(ollama.Client(), "tinyllama")
+
+# Call it to generate text
+response = model("What's the capital of Latvia?")
+print(response) # 'Riga'
+```
+
+Ollama also supports streaming. For instance:
+
+```python
+import ollama
+import outlines
+
+# Create the model
+model = outlines.from_ollama(ollama.Client(), "tinyllama")
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat"):
+    print(chunk) # 'In...'
+```
+
+## Structured Generation
+
+Ollama only provides support for structured generation based on a JSON schema. To use it, call the model with a JSON schema object as an `output_type` on top of your prompt.
+
+For instance:
+
+```python
+from typing import List
+from pydantic import BaseModel
+import ollama
+import outlines
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_ollama(ollama.Client(), "tinyllama")
+
+# Call it with the output type to generate structured text
+result = model("Create a character", Character)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+## Inference arguments
+
+When calling the model, you can provide keyword arguments that will be passed down to the `generate` method of the Ollama client.
+
+Consult the [Ollama REST API documentation](https://github.com/ollama/ollama/blob/main/docs/api.md#generate-a-completion) for the full list of inference parameters.
diff --git a/docs_new/features/models/openai.md b/docs_new/features/models/openai.md
index 0777ad8b7..c06acbc96 100644
--- a/docs_new/features/models/openai.md
+++ b/docs_new/features/models/openai.md
@@ -1,204 +1,134 @@
-# OpenAI and compatible APIs
+# OpenAI
 
 !!! Installation
 
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Or alternatively:
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines: `pip install openai`
 
-    ```bash
-    pip install "outlines[openai]"
-    ```
+    You also need to have an OpenAI API key. This API key must either be set as an environment variable called `OPENAI_API_KEY` or be provided to the `openai.OpenAI` class when instantiating it.
 
-## OpenAI models
+## Model Initialization
 
-Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.openai`:
-
-```python
-from outlines import models
+To create an OpenAI model instance, you can use the `from_openai` function. It takes 2 arguments:
 
+- `client`: an `openai.OpenAI` or `openai.AzureOpenAI` instance
+- `model_name`: the name of the model you want to use
 
-model = models.openai("gpt-4o-mini")
-model = models.openai("gpt-4o")
-```
-
-Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models. You can pass any parameter you would pass to `openai.AsyncOpenAI` as keyword arguments:
+For instance:
 
 ```python
-import os
-from outlines import models
+import outlines
+import openai
 
+# Create the client
+client = openai.OpenAI()
 
-model = models.openai(
-    "gpt-4o-mini",
-    api_key=os.environ["OPENAI_API_KEY"]
+# Create the model
+model = outlines.from_openai(
+    client,
+    "gpt-4o"
 )
 ```
 
-The following table enumerates the possible parameters. Refer to the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list.
-
-**Parameters:**
+Check the [OpenAI documentation](https://platform.openai.com/docs/models) for an up-to-date list of available models. As shown above, you can use Azure OpenAI in Outlines the same way you would use OpenAI, just provide an `openai.AzureOpenAI` instance to the Outlines model class.
 
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
-| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
-| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
-| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if no specified. | `None` |
-| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
-| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
-| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
-| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
-| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
+## Text Generation
 
-## Azure OpenAI models
+Once you've created your Outlines `OpenAI` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
 
-Outlines also supports Azure OpenAI models:
+For instance:
 
 ```python
-from outlines import models
+import openai
+import outlines
 
-
-model = models.azure_openai(
-    "azure-deployment-name",
-    "gpt-4o-mini",
-    api_version="2024-07-18",
-    azure_endpoint="https://example-endpoint.openai.azure.com",
+# Create the model
+model = outlines.from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
 )
-```
-
-!!! Question "Why do I need to specify model and deployment name?"
-
-    The model name is needed to load the correct tokenizer for the model. The tokenizer is necessary for structured generation.
-
 
-You can pass any parameter you would pass to `openai.AsyncAzureOpenAI`. You can consult the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/lib/azure.py) for an up-to-date list.
-
-**Parameters:**
-
-
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `azure_endpoint` | `str` | Azure endpoint, including the resource. Infered from `AZURE_OPENAI_ENDPOINT` if not specified | `None` |
-| `api_version` | `str` | API version. Infered from `AZURE_OPENAI_API_KEY` if not specified | `None` |
-| `api_key` | `str` | OpenAI API key. Infered from `OPENAI_API_KEY` if not specified | `None` |
-| `azure_ad_token` | `str` | Azure active directory token. Inference from `AZURE_OPENAI_AD_TOKEN` if not specified | `None` |
-| `azure_ad_token_provider` | `AzureADTokenProvider` | A function that returns an Azure Active Directory token | `None` |
-| `organization` | `str` | OpenAI organization id. Infered from `OPENAI_ORG_ID` if not specified | `None` |
-| `project` | `str` | OpenAI project id. Infered from `OPENAI_PROJECT_ID` if not specified.| `None` |
-| `base_url` | `str | https.URL` | Base URL for the endpoint. Infered from `OPENAI_BASE_URL` if not specified. | `None` |
-| `timeout` | `float` | Request timeout.| `NOT_GIVEN` |
-| `max_retries` | `int` | Maximum number of retries for failing requests | `2` |
-| `default_headers` | `Mapping[str, str]` | Default HTTP headers | `None` |
-| `default_query` | `Mapping[str, str]` | Custom parameters added to the HTTP queries | `None` |
-| `http_client` | `https.AsyncClient` | User-specified `httpx` client | `None` |
+# Call it to generate text
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
+```
 
-## Models that follow the OpenAI standard
+OpenAI supports streaming and vision inputs.
 
-Outlines supports models that follow the OpenAI standard. You will need to initialize the OpenAI client properly configured and pass it to `outlines.models.openai`
+For instance:
 
 ```python
-import os
-from openai import AsyncOpenAI
-from outlines import models
-from outlines.models.openai import OpenAIConfig
+import io
+import requests
+import PIL
+import openai
+import outlines
+from outlines.templates import Vision
+
+# Create the model
+model = outlines.from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
+)
 
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
 
-client = AsyncOpenAI(
-    api_key=os.environ.get("PROVIDER_KEY"),
-    base_url="http://other.provider.server.com"
-)
-config = OpenAIConfig("model_name")
-model = models.openai(client, config)
-```
+# Create the prompt
+prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
 
-!!! Warning
+# Stream the response
+for chunk in model.stream(prompt, max_tokens=50, stop="."):
+    print(chunk) # 'This...'
+```
 
-    You need to pass the async client to be able to do batch inference.
+## Structured Generation
 
-## Structured Generation Support
+OpenAI provides supports for some forms of structured output: JSON schemas and JSON syntax. To use it, call the model with an `output_type` on top of your prompt.
 
-Outlines provides support for [OpenAI Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode) via `outlines.generate.json`, `outlines.generate.choice`
+#### JSON Schema
 
 ```python
-from pydantic import BaseModel, ConfigDict
-import outlines.models as models
-from outlines import generate
-
-model = models.openai("gpt-4o-mini")
+from typing import List
+from pydantic import BaseModel
+import openai
+import outlines
 
-class Person(BaseModel):
-    model_config = ConfigDict(extra='forbid')  # required for openai
-    first_name: str
-    last_name: str
+class Character(BaseModel):
+    name: str
     age: int
+    skills: List[str]
 
-generator = generate.json(model, Person)
-generator("current indian prime minister on january 1st 2023")
-# Person(first_name='Narendra', last_name='Modi', age=72)
+# Create the model
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
 
-generator = generate.choice(model, ["Chicken", "Egg"])
-print(generator("Which came first?"))
-# Chicken
+# Call it with the output type to generate structured text
+result = model("Create a character", Character, top_p=0.1)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
 
-!!! Warning
+#### JSON Syntax
 
-    Structured generation support only provided to OpenAI-compatible endpoints which conform to OpenAI's standard. Additionally, `generate.regex` and `generate.cfg` are not supported.
-
-
-## Advanced configuration
-
-For more advanced configuration option, such as support proxy, please consult the [OpenAI SDK's documentation](https://github.com/openai/openai-python):
-
-
-```python
-from openai import AsyncOpenAI, DefaultHttpxClient
-from outlines import models
-from outlines.models.openai import OpenAIConfig
-
-
-client = AsyncOpenAI(
-    base_url="http://my.test.server.example.com:8083",
-    http_client=DefaultHttpxClient(
-        proxies="http://my.test.proxy.example.com",
-        transport=httpx.HTTPTransport(local_address="0.0.0.0"),
-    ),
-)
-config = OpenAIConfig("model_name")
-model = models.openai(client, config)
-```
-
-It is possible to specify the values for `seed`, `presence_penalty`, `frequence_penalty`, `top_p` by passing an instance of `OpenAIConfig` when initializing the model:
+What we mean by JSON syntax is what is sometimes called JSON mode, meaning that the model will return a valid JSON, but you do not get to specify its structure. To use this JSON mode, provide the `dict` type as an output type.
 
 ```python
-from outlines.models.openai import OpenAIConfig
-from outlines import models
+import openai
+import outlines
 
+# Create the model
+model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
 
-config = OpenAIConfig(
-    presence_penalty=1.,
-    frequency_penalty=1.,
-    top_p=.95,
-    seed=0,
-)
-model = models.openai("gpt-4o-mini", config)
+# Call it with the output type to generate structured text
+result = model("Create a character", dict, temperature=0.5)
+print(result) # '{"first_name": "Henri", "last_name": "Smith", "height": "170"}'
 ```
 
-## Monitoring API use
+## Inference arguments
 
-It is important to be able to track your API usage when working with OpenAI's API. The number of prompt tokens and completion tokens is directly accessible via the model instance:
-
-```python
-from openai import AsyncOpenAI
-import outlines.models
+When calling the model, you can provide keyword arguments that will be passed down to the `chat.completions.create` method of the OpenAI client. Some of the most common arguments include `max_tokens`, `temperature`, `stop` and `top_p`.
 
+Another keyword argument of interest is `n`. If set with an integer value superior to 1, OpenAI will generate several sample responses and you will receive a list of strings as a response to your model call.
 
-model = models.openai("gpt-4o")
-
-print(model.prompt_tokens)
-# 0
-
-print(model.completion_tokens)
-# 0
-```
-
-These numbers are updated every time you call the model.
+See the [OpenAI API documentation](http://platform.openai.com/docs/api-reference/chat/create) for the full list of available arguments.
diff --git a/docs_new/features/models/sglang.md b/docs_new/features/models/sglang.md
index e69de29bb..beb991b6e 100644
--- a/docs_new/features/models/sglang.md
+++ b/docs_new/features/models/sglang.md
@@ -0,0 +1,171 @@
+---
+title: SGLang
+---
+
+# SGLang
+
+## Prerequisites
+
+The Outlines `SGLang` model is intended to be used along with an SGLang instance running on a separate server (can be local or remote). Make sure you have a SGLang server running and accessible before using the `SGLang` model. As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed: `pip install openai`.
+
+When launching your SGLang server, you can specify the backend engine to use for structured generation through the `grammar-backend` cli argument. Add `--grammar-backend outlines` to your command to use Outlines instead of the default engine.
+
+## Model Initialization
+
+To load the model, you can use the `from_sglang` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Make sure the value of the `base_url` argument of the `OpenAI` client points to your running SGLang server. Consult the [SGLang documentation](https://docs.sglang.ai/backend/send_request.html) on using an OpenAI client with an SGLang server for more information.
+
+Based on whether the `openai` client instance is synchronous or asynchronous, you will receive a `SGLang` or `AsyncSGLang` model instance.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the OpenAI client
+sync_openai_client = openai.OpenAI(base_url="http://localhost:11434")
+async_openai_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
+
+# Create a sync model
+sync_model = outlines.from_sglang(sync_openai_client)
+print(type(sync_model)) # <class 'outlines.models.sglang.SGLang'>
+
+# Create an async model
+async_model = outlines.from_sglang(async_openai_client)
+print(type(async_model)) # <class 'outlines.models.sglang.AsyncSGLang'>
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+# Call it to generate text
+result = model("Write a short story about a cat.", max_tokens=100)
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```
+
+The `SGLang` model also supports streaming. For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In ...'
+```
+
+## Structured Generation
+
+SGLang supports all output types available in Outlines (context-free grammars with caveats though, see the subsection below for more details). Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import openai
+import outlines
+
+output_type = int
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+import openai
+import outlines
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("Create a character.", Character, frequency_penalty=1.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import openai
+import outlines
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("What is the capital of France?", output_type, temperature=0)
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+import openai
+import outlines
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("Generate a fake social security number.", output_type, top_p=0.1)
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
+
+SGLang supports grammars, but expects an EBNF format instead of the Lark format Outlines uses. Thus, to use a context-free grammar with SGLang, provide a string using the EBNF syntax to the Outlines `CFG` object.
+
+```python
+import openai
+import outlines
+from outlines.text import CFG
+
+ebnf_grammar = """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+output_type = CFG(ebnf_grammar)
+
+openai_client = openai.OpenAI(base_url="http://localhost:11434")
+model = outlines.from_sglang(openai_client)
+
+result = model("Is the weather good today?", output_type)
+print(result) # 'yes'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.
+
+An optional parameter of particular interest is `extra_body`, which is a dictionary containing arguments that are specific to SGLang and are not part of the standard `openai` interface.
+
+See the [SGLang documentation](https://docs.sglang.ai/backend/openai_api_completions.html) on parameters for the OpenAI-compatible server for more information on inference parameters.
diff --git a/docs_new/features/models/tgi.md b/docs_new/features/models/tgi.md
index 3f0568103..331b68931 100644
--- a/docs_new/features/models/tgi.md
+++ b/docs_new/features/models/tgi.md
@@ -1,3 +1,137 @@
-# Text-generation-inference (TGI)
+---
+title: TGI
+---
 
-TGI uses Outlines to provide structured generation, see [their documentation](https://huggingface.co/docs/text-generation-inference/en/basic_tutorials/using_guidance).
+# TGI
+
+## Prerequisites
+
+The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed: `pip install huggingface_hub`
+
+## Model Initialization
+
+To load the model, you can use the `from_tgi` function. The argument of the function is either an `InferenceClient` or `AsyncInferenceClient` instance from the `huggingface_hub` library. Consult the [HuggingFace documentation](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client) for more information on their inference client.
+
+Based on whether the inference client instance is synchronous or asynchronous, you will receive a `TGI` or an `AsyncTGI` model instance.
+
+For instance:
+
+```python
+import outlines
+import huggingface_hub
+
+# Create the inference client
+client = huggingface_hub.InferenceClient("http://localhost:11434")
+async_client = huggingface_hub.AsyncInferenceClient("http://localhost:11434")
+
+# Create a sync model
+sync_model = outlines.from_tgi(client)
+print(type(sync_model))  # <class 'outlines.models.tgi.TGI'>
+
+# Create an async model
+async_model = outlines.from_tgi(async_client)
+print(type(async_model))  # <class 'outlines.models.tgi.AsyncTGI'>
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+import huggingface_hub
+
+# Create the model
+client = huggingface_hub.InferenceClient("http://localhost:11434")
+model = outlines.from_tgi(client)
+
+# Call it to generate text
+result = model("Write a short story about a cat.", stop_sequences=["."])
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```The `TGI` model supports streaming. For instance:
+
+```python
+import outlines
+import huggingface_hub
+
+# Create the model
+client = huggingface_hub.InferenceClient("http://localhost:11434")
+model = outlines.from_tgi(client)
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", stop_sequences=["."]):
+    print(chunk) # 'In ...'
+```
+
+## Structured Generation
+
+TGI supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import outlines
+import huggingface_hub
+
+output_type = int
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```### JSON Schema
+
+```python
+import outlines
+import huggingface_hub
+from typing import List
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```### Multiple Choice
+
+```python
+import outlines
+import huggingface_hub
+from typing import Literal
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("What is the capital of France?", output_type, temperature=0)
+print(result) # 'Paris'
+```### Regex
+
+```python
+import outlines
+import huggingface_hub
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+tgi_client = huggingface_hub.InferenceClient("http://localhost:8080")
+model = outlines.from_tgi(tgi_client)
+
+result = model("Generate a fake social security number.", output_type, top_p=0.1)
+print(result) # '782-32-3789'
+```
+## Inference parameters
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `text_generation` method of the TGI client.
+
+Common parameters include `max_new_tokens`, `stop_sequences`, `temperature`, `top_k`, `top_p`, and others as specified in the [TGI inference client documentation](https://huggingface.co/docs/huggingface_hub/package_reference/inference_client#huggingface_hub.InferenceClient).
diff --git a/docs_new/features/models/transformers.md b/docs_new/features/models/transformers.md
index 3649b9b13..f03baa991 100644
--- a/docs_new/features/models/transformers.md
+++ b/docs_new/features/models/transformers.md
@@ -1,148 +1,207 @@
-# transformers
+---
+title: Transformers
+---
 
+# Transformers
 
 !!! Installation
 
-    You need to install the `transformer`, `datasets` and `torch` libraries to be able to use these models in Outlines, or alternatively:
+    You need to install the `transformers` library to be able to use the `Transformers` model in Outlines: `pip install transformers`
 
-    ```bash
-    pip install "outlines[transformers]"
-    ```
+    See the [HuggingFace documentation](https://huggingface.co/docs/transformers/en/installation) for more information on installing `transformers` with CPU, GPU...
 
+## Model Initialization
 
-Outlines provides an integration with the `torch` implementation of causal models in the [transformers][transformers] library. You can initialize the model by passing its name:
+To load the model, you can use the `from_transformers` function. It takes 2 arguments:
 
-```python
-from outlines import models
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
-```
-
-If you need more fine-grained control you can also initialize the model and tokenizer separately:
+- `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
+- `tokenizer_or_processor`: a `transformers` tokenizer (created with `AutoTokenizer` for instance, it must be an instance of either `PreTrainedTokenizer` or `PreTrainedTokenizerFast`)
 
+For instance:
 
 ```python
+import outlines
 from transformers import AutoModelForCausalLM, AutoTokenizer
-from outlines import models
 
-llm = AutoModelForCausalLM.from_pretrained("gpt2", output_attentions=True)
-tokenizer = AutoTokenizer.from_pretrained("gpt2")
-model = models.Transformers(llm, tokenizer)
-```
+# Create the transformers model and tokenizer
+hf_model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+hf_tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
 
-## Using Logits Processors
+# Create the Outlines model
+model = outlines.from_transformers(hf_model, hf_tokenizer)
+```
 
-There are two ways to use Outlines Structured Generation with HuggingFace Transformers:
+If you provide a processor instead of a tokenizer for the second argument of the `from_transformers` function, you would get a `TransformersMultiModal` instance. See the [TransformersMultiModal model documentation](./transformers_multimodal.md) for more information on using multimodal models in Outlines.
 
-1. Use Outlines generation wrapper, `outlines.models.transformers`
-2. Use `OutlinesLogitsProcessor` with `transformers.AutoModelForCausalLM`
+## Text Generation
 
-Outlines supports a myriad of logits processors for structured generation. In these example, we will use the `RegexLogitsProcessor` which guarantees generated text matches the specified pattern.
+To generate text, you can simply call the model with a prompt.
 
-### Using `outlines.models.transformers`
+For instance:
 
 ```python
 import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-time_regex_pattern = r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?"
-
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct", device="cuda")
-generator = outlines.generate.regex(model, time_regex_pattern)
+# Create model
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
-output = generator("The the best time to visit a dentist is at ")
-print(output)
-# 2:30 pm
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_new_tokens=20)
+print(result) # 'Riga'
 ```
 
-### Using models initialized via the `transformers`  library
+The `Transformers` model also supports batch generation. To use it, provide a list of prompts instead of a single prompt. You will receive as result a list of completions. For instance:
 
 ```python
 import outlines
-import transformers
-
-
-model_uri = "microsoft/Phi-3-mini-4k-instruct"
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-outlines_tokenizer = outlines.models.TransformerTokenizer(
-    transformers.AutoTokenizer.from_pretrained(model_uri)
-)
-phone_number_logits_processor = outlines.processors.RegexLogitsProcessor(
-    "\\+?[1-9][0-9]{7,14}",  # phone number pattern
-    outlines_tokenizer,
+# Create model
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
 )
 
-generator = transformers.pipeline('text-generation', model=model_uri)
+# Create a list of prompts that will be used in a single batch
+prompts = [
+    "What's the capital of Lithuania?",
+    "What's the capital of Latvia?",
+    "What's the capital of Estonia?"
+]
 
-output = generator(
-    "Jenny gave me her number it's ",
-	logits_processor=transformers.LogitsProcessorList([phone_number_logits_processor])
-)
-print(output)
-# [{'generated_text': "Jenny gave me her number it's 2125550182"}]
-# not quite 8675309 what we expected, but it is a valid phone number
+# Call it to generate text
+result = model(prompts, max_new_tokens=20)
+print(result) # ['Vilnius', 'Riga', 'Tallinn']
 ```
 
-[transformers]: https://github.com/huggingface/transformers
-
+## Structured Generation
 
-## Alternative Model Classes
+As a local model, `Transformers` supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
 
-`outlines.models.transformers` defaults to `transformers.AutoModelForCausalLM`, which is the appropriate class for most standard large language models, including Llama 3, Mistral, Phi-3, etc.
+### Simple Type
 
-However other variants with unique behavior can be used as well by passing the appropriate class.
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-### Mamba
+output_type = int
 
-[Mamba](https://github.com/state-spaces/mamba) is a transformers alternative which employs memory efficient, linear-time decoding.
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
-To use Mamba with outlines you must first install the necessary requirements:
-```
-pip install causal-conv1d>=1.2.0 mamba-ssm torch transformers
+result = model("How many countries are there in the world?", output_type, max_new_tokens=5)
+print(result) # '200'
 ```
 
-Then you can either create an Mamba-2 Outlines model via
+### JSON Schema
+
 ```python
 import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from pydantic import BaseModel
+from typing import List
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
-model = outlines.models.mamba("state-spaces/mamba-2.8b-hf")
+result = model("Create a character.", output_type=Character, max_new_tokens=200, repetition_penalty=0.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
 
-or explicitly with
+### Multiple Choice
+
 ```python
+from typing import Literal
 import outlines
-from transformers import MambaForCausalLM
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
-model = outlines.models.transformers(
-    "state-spaces/mamba-2.8b-hf",
-    model_class=MambaForCausalLM
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
 )
+
+result = model("What is the capital of France?", output_type, max_new_tokens=10, temperature=0)
+print(result) # 'Paris'
 ```
 
+### Regex
 
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from outlines.types import Regex
 
-Read [`transformers`'s documentation](https://huggingface.co/docs/transformers/en/model_doc/mamba) for more information.
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
-### Encoder-Decoder Models
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
-You can use encoder-decoder (seq2seq) models like T5 and BART with Outlines.
+result = model("Generate a fake social security number.", output_type, max_new_tokens=20, top_p=0.5)
+print(result) # '782-32-3789'
+```
 
-Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
+### Context-Free Grammar
 
-T5 Example:
 ```python
 import outlines
-from transformers import AutoModelForSeq2SeqLM
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from outlines.text import CFG
 
-model_pile_t5 = outlines.models.transformers(
-    model_name="EleutherAI/pile-t5-large",
-    model_class=AutoModelForSeq2SeqLM,
-)
-```
+arithmetic_grammar = """
+?start: sum
 
-Bart Example:
-```python
-model_bart = outlines.models.transformers(
-    model_name="facebook/bart-large",
-    model_class=AutoModelForSeq2SeqLM,
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
 )
+
+result = model("Write an addition.", output_type, max_new_tokens=100)
+print(result) # '23 + 48'
 ```
+
+## Inference Arguments
+
+When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `generate` method of the `transformers` model. Some common inference arguments include `max_new_tokens`, `temperature`, `repetition_penalty` and `top_p`.
+
+See the [transformers documentation](https://huggingface.co/docs/transformers/en/main_classes/text_generation) for more information on inference parameters.
+
+!!! Warning
+
+    The `max_new_tokens` inference parameter has a default value of 20. This is insufficient for most tasks and will result in the generation output not respecting the output type (because the response is truncated). We recommend you always provide a value for this argument.
diff --git a/docs_new/features/models/transformers_multimodal.md b/docs_new/features/models/transformers_multimodal.md
new file mode 100644
index 000000000..b2b846f6c
--- /dev/null
+++ b/docs_new/features/models/transformers_multimodal.md
@@ -0,0 +1,89 @@
+---
+title: Transformers MultiModal
+---
+
+# Transformers MultiModal
+
+The Outlines `TransformersMultiModal` model inherits from `Transformers` and shares most of its interface. Please start by reading the [Transformers documentation](./transformers.md) as this document only focuses on the specificities of `TransformersMultiModal` compared to `Transformers`.
+
+On top of the difference of arguments to provide to the `from_transformers` function as described below, the only other specificity of the `TransformersMultiModal` model is the format of the prompt to use when calling the model. For all other elements such as the output types and inference arguments, it works exactly the same way as `Transformers`.
+
+## Model Initialization
+
+To load the model, you can use the `from_transformers` function. It takes 2 arguments:
+
+- `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
+- `tokenizer_or_processor`: a `transformers` processor (created with `AutoProcessor` for instance, it must be an instance of `ProcessorMixin`)
+
+For instance:
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoProcessor
+
+# Create the transformers model and processor
+hf_model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+hf_processor = AutoProcessor.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+
+# Create the Outlines model
+model = outlines.from_transformers(hf_model, hf_processor)
+```
+
+## Model Input
+
+Instead of a string or a list of strings (for batch generation), you should provide a dictionnary or a list of dictionnaries as a prompt when calling the `TransformersMultiModal` model. The dictionnary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
+
+```python
+{
+    "text": Union[str, List[str]]
+    "<other_keys_depending_on_your_processor>": Union[Any, List[Any]]
+}
+```
+
+Some common keys to include for processors are `images` for vision models or `audios` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets or if you are using batch generation).
+
+Here's an example of using a vision multimodal model:
+
+```python
+from io import BytesIO
+from urllib.request import urlopen
+
+from PIL import Image
+from pydantic import BaseModel
+from transformers import (
+    LlavaForConditionalGeneration,
+    AutoProcessor,
+)
+
+import outlines
+
+TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+
+class Animal(BaseModel):
+    specie: str
+    color: str
+    weight: int
+
+def get_image_from_url():
+    img_byte_stream = BytesIO(urlopen(IMAGE_URL).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+# Create a model
+model = outlines.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+    AutoProcessor.from_pretrained(TEST_MODEL),
+)
+
+# Call it with a model input dict containing a text prompt and an image + an output type
+result = model(
+    {"text": "<image>Describe this animal.", "images": get_image_from_url(IMAGE_URL)},
+    Animal,
+)
+print(result) # '{"specie": "cat", "color": "white", "weight": 4}'
+print(Animal.model_validate_json(result)) # specie=cat, color=white, weight=4
+```
+
+!!! Warning
+
+    Make sure your prompt (the value of the `text` key) contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are images present in the value of the `images` key.
diff --git a/docs_new/features/models/vllm.md b/docs_new/features/models/vllm.md
index 8789b588e..ad7fea0ba 100644
--- a/docs_new/features/models/vllm.md
+++ b/docs_new/features/models/vllm.md
@@ -1,235 +1,186 @@
+---
+title: vLLM
+---
+
 # vLLM
 
+## Prerequisites
 
-!!! Note "Installation"
+The Outlines `VLLM` model is intended to be used along with a vLLM instance running on a separate server (can be local or remote). Make sure you have a vLLM server running and accessible before using the `VLLM` model. As the vLLM client relies on the `openai` python sdk, you need to have an `openai` package installed: `pip install openai`.
 
-    You need to install the `vllm` library to use the vLLM integration. See the [installation section](#installation) for instructions to install vLLM for CPU or ROCm. To get started you can also run:
+If you want to use the vllm offline inference mode instead of the server mode, please refer to the [VLLMOffline](./vllm_offline.md) model documentation.
 
-    ```bash
-    pip install "outlines[vllm]"
-    ```
+## Model Initialization
 
-## Load the model
+To load the model, you can use the `from_vllm` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Make sure the value of the `base_url` argument of the `OpenAI` client points to your running vLLM server. Consult the [vLLM documentation](https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html) on using an OpenAI client with a vLLM server for more information.
 
-Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
+Based on whether the `openai` client instance is synchronous or asynchronous, you will receive a `VLLM` or `AsyncVLLM` model instance.
 
+For instance:
 
 ```python
-from outlines import models
+import openai
+import outlines
 
-model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
-```
+# Create the OpenAI client
+sync_openai_client = openai.OpenAI(base_url="http://localhost:8080")
+async_openai_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
 
-Or alternatively:
+# Create a sync model
+sync_model = outlines.from_vllm(sync_openai_client)
+print(type(sync_model)) # <class 'outlines.models.vllm.VLLM'>
 
-```python
-import vllm
-from outlines import models
-
-llm = vllm.LLM("microsoft/Phi-3-mini-4k-instruct")
-model = models.VLLM(llm)
+# Create an async model
+async_model = outlines.from_vllm(async_openai_client)
+print(type(async_model)) # <class 'outlines.models.vllm.AsyncVLLM'>
 ```
 
+## Text Generation
 
-Models are loaded from the [HuggingFace hub](https://huggingface.co/).
-
+To generate text, you can simply call the model with a prompt.
 
-!!! Warning "Device"
-
-    The default installation of vLLM only allows to load models on GPU. See the [installation instructions](#installation) to run models on CPU.
-
-
-You can pass any parameter that you would normally pass to `vllm.LLM`, as keyword arguments:
+For instance:
 
 ```python
-from outlines import models
-
-model = models.vllm(
-    "microsoft/Phi-3-mini-4k-instruct",
-    trust_remote_code=True,
-    gpu_memory_utilization=0.7
-)
-```
+import openai
+import outlines
 
-**Main parameters:**
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
 
-| **Parameters** | **Type** | **Description** | **Default** |
-|----------------|:---------|:----------------|:------------|
-| `tokenizer_mode`| `str`  | "auto" will use the fast tokenizer if available and "slow" will always use the slow tokenizer. | `auto`
-| `trust_remote_code`| `bool` | Trust remote code when downloading the model and tokenizer. | `False` |
-| `tensor_parallel_size`| `int` | The number of GPUs to use for distributed execution with tensor parallelism.| `1` |
-| `dtype`| `str` | The data type for the model weights and activations. Currently, we support `float32`, `float16`, and `bfloat16`. If `auto`, we use the `torch_dtype` attribute specified in the model config file. However, if the `torch_dtype` in the config is `float32`, we will use `float16` instead.| `auto` |
-| `quantization`| `Optional[str]` | The method used to quantize the model weights. Currently, we support "awq", "gptq" and "squeezellm". If None, we first check the `quantization_config` attribute in the model config file. If that is None, we assume the model weights are not quantized and use `dtype` to determine the data type of the weights.| `None` |
-| `revision`| `Optional[str]` | The specific model version to use. It can be a branch name, a tag name, or a commit id.| `None` |
-| `tokenizer_revision`| `Optional[str]`| The specific tokenizer version to use. It can be a branch name, a tag name, or a commit id.| `None` |
-| `gpu_memory_utilization`| `float` | The ratio (between 0 and 1) of GPU memory to reserve for the model weights, activations, and KV cache. Higher values will increase the KV cache size and thus improve the model's throughput. However, if the value is too high, it may cause out-of-memory (OOM) errors.| `0.9` |
-| `swap_space`| `int` | The size (GiB) of CPU memory per GPU to use as swap space. This can be used for temporarily storing the states of the requests when their `best_of` sampling parameters are larger than 1. If all requests will have `best_of=1`, you can safely set this to 0. Otherwise, too small values may cause out-of-memory (OOM) errors.| 4 |
-| `enforce_eager`| `bool` | Whether to enforce eager execution. If True, we will disable CUDA graph and always execute the model in eager mode. If False, we will use CUDA graph and eager execution in hybrid.| `False` |
-| `enable_lora` | `bool` | Whether to enable loading LoRA adapters | `False` |
-
-See the [vLLM code](https://github.com/vllm-project/vllm/blob/8f44facdddcf3c704f7d6a2719b6e85efc393449/vllm/entrypoints/llm.py#L72) for a list of all the available parameters.
-
-### Use quantized models
-
-vLLM supports AWQ, GPTQ and SqueezeLLM quantized models:
+# Call it to generate text
+result = model("Write a short story about a cat.", max_tokens=100)
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```
 
+The `VLLM` model also supports streaming. For instance:
 
 ```python
-from outlines import models
-
-model = models.vllm("TheBloke/Llama-2-7B-Chat-AWQ", quantization="awq")
-model = models.vllm("TheBloke/Mistral-7B-Instruct-v0.2-GPTQ", quantization="gptq")
-model = models.vllm("https://huggingface.co/squeeze-ai-lab/sq-llama-30b-w4-s5", quantization="squeezellm")
-```
+import openai
+import outlines
 
-!!! Warning "Dependencies"
+# Create the model
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
 
-    To use AWQ model you need to install the autoawq library `pip install autoawq`.
+# Stream text
+for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
+    print(chunk) # 'In ...'
+```
 
-    To use GPTQ models you need to install the autoGTPQ and optimum libraries `pip install auto-gptq optimum`.
+## Structured Generation
 
+vLLM supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
 
-### Multi-GPU usage
+### Simple Type
 
-To run multi-GPU inference with vLLM you need to set the `tensor_parallel_size` argument to the number of GPUs available when initializing the model. For instance to run inference on 2 GPUs:
+```python
+import openai
+import outlines
 
+output_type = int
 
-```python
-from outlines import models
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
 
-model = models.vllm(
-    "microsoft/Phi-3-mini-4k-instruct"
-    tensor_parallel_size=2
-)
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
 ```
 
-### Load LoRA adapters
-
-You can load LoRA adapters and alternate between them dynamically:
+### JSON Schema
 
 ```python
-from outlines import models
-
-model = models.vllm("facebook/opt-350m", enable_lora=True)
-model.load_lora("ybelkaa/opt-350m-lora")  # Load LoRA adapter
-model.load_lora(None)  # Unload LoRA adapter
+import openai
+import outlines
+from typing import List
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
+
+result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
 
-## Generate text
-
-In addition to the parameters described in the [text generation section](../text.md) you can pass an instance of `SamplingParams` directly to any generator via the `sampling_params` keyword argument:
+### Multiple Choice
 
 ```python
-from vllm.sampling_params import SamplingParams
-from outlines import models, generate
+from typing import Literal
+import openai
+import outlines
 
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
-model = models.vllm("microsoft/Phi-3-mini-4k-instruct")
-generator = generate.text(model)
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
 
-params = SamplingParams(n=2, frequency_penalty=1., min_tokens=2)
-answer = generator("A prompt", sampling_params=params)
+result = model("What is the capital of France?", output_type, temperature=0)
+print(result) # 'Paris'
 ```
 
-This also works with generators built with `generate.regex`, `generate.json`, `generate.cfg`, `generate.format` and `generate.choice`.
-
-!!! Note
-
-    The values passed via the `SamplingParams` instance supersede the other arguments to the generator or the samplers.
-
-**`SamplingParams` attributes:**
-
-| Parameters | Type             | Description            | Default |
-|:-----------|------------------|:-----------------------|---------|
-| `n` | `int` | Number of output sequences to return for the given prompt. | `1` |
-| `best_of` | `Optional[int]` | Number of output sequences that are generated from the prompt. From these `best_of` sequences, the top `n` sequences are returned. `best_of` must be greater than or equal to `n`. This is treated as the beam width when `use_beam_search` is True. By default, `best_of` is set to `n`. | `None` |
-| `presence_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens.| `0.0` |
-| `frequency_penalty` | `float` | Float that penalizes new tokens based on their frequency in the generated text so far. Values > 0 encourage the model to use new tokens, while values < 0 encourage the model to repeat tokens. | `0.0`
-| `repetition_penalty` | `float` | Float that penalizes new tokens based on whether they appear in the prompt and the generated text so far. Values > 1 encourage the model to use new tokens, while values < 1 encourage the model to repeat tokens. | `1.0` |
-| `temperature` | `float` | Float that controls the randomness of the sampling. Lower values make the model more deterministic, while higher values make the model more random. Zero means greedy sampling. | `1.0` |
-| `top_p` | `float` |  Float that controls the cumulative probability of the top tokens to consider. Must be in (0, 1]. Set to 1 to consider all tokens. | `1.0` |
-| `top_k` | `int` |  Integer that controls the number of top tokens to consider. Set to -1 to consider all tokens. | `-1` |
-| `min_p` |`float` |  Float that represents the minimum probability for a token to be considered, relative to the probability of the most likely token. Must be in [0, 1]. Set to 0 to disable this. | `0.0` |
-| `seed` | `Optional[int]` | Random seed to use for the generation. | `None` |
-| `use_beam_search` | `bool` |  Whether to use beam search instead of sampling. | `False` |
-| `length_penalty` | `float` | Float that penalizes sequences based on their length. Used in beam search.  | `1.0` |
-| `early_stopping` | `Union[bool, str]` |  Controls the stopping condition for beam search. It accepts the following values: `True`, where the generation stops as soon as there are `best_of` complete candidates; `False`, where an heuristic is applied and the generation stops when is it very unlikely to find better candidates; `"never"`, where the beam search procedure only stops when there cannot be better candidates (canonical beam search algorithm). | `False` |
-| `stop` | `Optional[Union[str, List[str]]]` |  List of strings that stop the generation when they are generated. The returned output will not contain the stop strings. | `None` |
-| `stop_token_ids` | `Optional[List[int]]` |  List of tokens that stop the generation when they are generated. The returned output will contain the stop tokens unless the stop tokens are special tokens. | `None` |
-| `include_stop_str_in_output` | `bool` |  Whether to include the stop strings in output text. Defaults to False. | `False` |
-| `ignore_eos` | `bool` |  Whether to ignore the EOS token and continue generating tokens after the EOS token is generated. | `False` |
-| `max_tokens` | `int` |  Maximum number of tokens to generate per output sequence. | `16` |
-| `min_tokens` | `int` | Minimum number of tokens to generate per output sequence before EOS or stop_token_ids can be generated | `0` |
-| `skip_special_tokens` | `bool` | Whether to skip special tokens in the output. | `True` |
-| `spaces_between_special_tokens` | `bool` |  Whether to add spaces between special tokens in the output.  Defaults to True. | `True` |
-
-### Streaming
-
-!!! Warning
-
-    Streaming is not available for the offline vLLM integration.
-
-
-## Installation
-
-By default the vLLM library is installed with pre-commpiled C++ and CUDA binaries and will only run on GPU:
+### Regex
 
 ```python
-pip install vllm
-```
+import openai
+import outlines
+from outlines.types import Regex
 
-### CPU
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
-You need to have the `gcc` compiler installed on your system. Then you will need to install vLLM from source. First clone the repository:
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
 
-```bash
-git clone https://github.com/vllm-project/vllm.git
-cd vllm
+result = model("Generate a fake social security number.", output_type, top_p=0.1)
+print(result) # '782-32-3789'
 ```
 
-Install the Python packages needed for the installation:
-
-```bash
-pip install --upgrade pip
-pip install wheel packaging ninja setuptools>=49.4.0 numpy
-pip install -v -r requirements-cpu.txt --extra-index-url https://download.pytorch.org/whl/cpu
-```
+### Context-Free Grammar
 
-and finally run:
+```python
+import openai
+import outlines
+from outlines.text import CFG
 
-```bash
-VLLM_TARGET_DEVICE=cpu python setup.py install
-```
+arithmetic_grammar = """
+?start: sum
 
-See the [vLLM documentation][vllm-install-cpu] for more details, alternative installation methods (Docker) and performance tips.
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
 
-### ROCm
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
 
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
 
-You will need to install vLLM from source. First install Pytorch on ROCm:
+%import common.NUMBER
+%import common.WS_INLINE
 
-```bash
-pip install torch==2.2.0.dev20231206+rocm5.7 --index-url https://download.pytorch.org/whl/nightly/rocm5.7 # tested version
-```
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
 
-You will then need to install flash attention for ROCm following [these instructions][rocm-flash-attention]. You can then install `xformers=0.0.23` and apply the patches needed to adapt Flash Attention for ROCm:
+openai_client = openai.OpenAI(base_url="http://localhost:8080")
+model = outlines.from_vllm(openai_client)
 
-```bash
-pip install xformers==0.0.23 --no-deps
-bash patch_xformers.rocm.sh
+result = model("Write an addition.", output_type, extra_body={"guided_decoding_backend": "outlines"})
+print(result) # '23 + 48'
 ```
 
-And finally build vLLM:
-
-```bash
-cd vllm
-pip install -U -r requirements-rocm.txt
-python setup.py install # This may take 5-10 minutes.
-```
+## Inference Arguments
 
-See the [vLLM documentation][vllm-install-rocm] for alternative installation methods (Docker).
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.
 
+An optional parameter of particular interest is `extra_body`, which is a dictionary containing arguments that are specific to vLLM and are not part of the standard `openai` interface. Among those, `guided_decoding_backend` allows you to select the library used by the vLLM server to control structured generation. You can use the value `outlines` to generated structured text with Outlines.
 
-[vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
-[vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
-[rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
+See the [vLLM documentation](https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html#extra-parameters) on extra parameters for the OpenAI-compatible server for more information on inference parameters.
diff --git a/docs_new/features/models/vllm_offline.md b/docs_new/features/models/vllm_offline.md
index e69de29bb..178f4789c 100644
--- a/docs_new/features/models/vllm_offline.md
+++ b/docs_new/features/models/vllm_offline.md
@@ -0,0 +1,170 @@
+---
+title: vLLM Offline
+---
+
+# vLLM Offline
+
+Outlines provides an integration with [vLLM](https://docs.vllm.ai/en/latest/) using the [vllm library](https://github.com/vllm-project/vllm). This model allows you to use vLLM in the "Offline Inference" mode, meaning that text generation happens within the model, there is no separate server. If you want to use vLLM with a server, see the [VLLM model documentation](./vllm.md).
+
+!!! Note "Installation"
+
+    You need to install the `vllm` library to be able to use the `VLLMOffline` model: `pip install vllm`.
+
+    See the [vLLM documentation](https://docs.vllm.ai/en/latest/getting_started/installation/index.html) for instructions on how to install vLLM for CPU, ROCm...
+
+## Model Initialization
+
+To load the model, you can use the `from_vllm_offline` function. The single argument of the function is a `LLM` model instance from the `vllm` library. You will then receive a `VLLMOffline` model instance you can use to generate text.
+
+Consult the [LLM class API reference](https://docs.vllm.ai/en/latest/api/vllm/index.html#vllm.LLM) for detailed information on how to create an `LLM` instance and on the various available parameters.
+
+For instance:
+
+```python
+import outlines
+from vllm import LLM
+
+# Create the model
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+```
+
+## Text Generation
+
+To generate text, you can simply call the model with a prompt.
+
+For instance:
+
+```python
+import outlines
+from vllm import LLM
+
+# Create the model
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Call it to generate text
+result = model("What's the capital of Latvia?", max_tokens=20)
+print(result) # 'Riga'
+```
+
+## Structured Generation
+
+As a local model, `VLLMOffline` supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
+
+### Simple Type
+
+```python
+import outlines
+from vllm import LLM
+
+output_type = int
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("How many countries are there in the world?", output_type)
+print(result) # '200'
+```
+
+### JSON Schema
+
+```python
+import outlines
+from vllm import LLM
+from typing import List
+from pydantic import BaseModel
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+```
+
+### Multiple Choice
+
+```python
+from typing import Literal
+import outlines
+from vllm import LLM
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("What is the capital of France?", output_type, temperature=0)
+print(result) # 'Paris'
+```
+
+### Regex
+
+```python
+import outlines
+from vllm import LLM
+from outlines.types import Regex
+
+output_type = Regex(r"\d{3}-\d{2}-\d{4}")
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Generate a fake social security number.", output_type, top_p=0.1)
+print(result) # '782-32-3789'
+```
+
+### Context-Free Grammar
+
+```python
+import outlines
+from vllm import LLM
+from outlines.text import CFG
+
+arithmetic_grammar = """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+output_type = CFG(arithmetic_grammar)
+
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("Write an addition.", output_type)
+print(result) # '23 + 48'
+```
+
+## Inference Arguments
+
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `generate` method of the `LLM` model instance. Some common inference arguments include `max_tokens`, `temperature`, `frequency_penalty` and `top_p`.
+
+See the [vLLM documentation](https://docs.vllm.ai/en/latest/api/vllm/sampling_params.html#vllm.sampling_params.SamplingParams) on sampling parameters for more information on inference parameters.
diff --git a/mkdocs.yml b/mkdocs.yml
index c423c9f47..5419389f5 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -180,11 +180,26 @@ nav:
                       - Templates: user_guide/prompting/templates.md
                 - Easy Regular Expressions: user_guide/regex_dsl.md
                 - Migrating to v1: user_guide/migration.md
+
     - Features:
           - Overview: features/index.md
 
           - Core:
-                - Generator: features/core/generator.md
+                - Models:
+                    - Overview: features/models/index.md
+                    - Anthropic: features/models/anthropic.md
+                    - Dottxt: features/models/dottxt.md
+                    - Gemini: features/models/gemini.md
+                    - Llamacpp: features/models/llamacpp.md
+                    - Mlx-lm: features/models/mlxlm.md
+                    - Ollama: features/models/ollama.md
+                    - OpenAI: features/models/openai.md
+                    - SGLang: features/models/sglang.md
+                    - TGI: features/models/tgi.md
+                    - Transformers: features/models/transformers.md
+                    - TransformersMultiModal: features/models/transformers_multimodal.md
+                    - vLLM (online server): features/models/vllm.md
+                    - vLLM (offline): features/models/vllm_offline.md
                 - Output Types:
                     - Overview: features/core/output_types.md
                     - Basic Types: features/core/output_types#basic-python-types
@@ -192,20 +207,7 @@ nav:
                     - JSON: features/core/output_types#json-schemas
                     - Regex: features/core/output_types#regex-patterns
                     - Context-free Grammars: features/core/output_types#context-free-grammars
-
-          - Supported Models:
-                - Anthropic: features/models/anthropic.md
-                - Dottxt: features/models/dottxt.md
-                - Gemini: features/models/gemini.md
-                - Llamacpp: features/models/llamacpp.md
-                - Mlx-lm: features/models/mlxlm.md
-                - Ollama: features/models/ollama.md
-                - OpenAI: features/models/openai.md
-                - SGLang: features/models/sglang.md
-                - TGI: features/models/tgi.md
-                - Transformers: features/models/transformers.md
-                - vLLM (online server): features/models/vllm.md
-                - vLLM (offline): features/models/vllm_offline.md
+                - Generator: features/core/generator.md
 
           - Utilities:
                 - Application: features/utility/application.md

From 8ca7588763f08d8ce4a1a90ccf4a751395aae1b9 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 2 Jun 2025 14:44:59 +0200
Subject: [PATCH 341/505] Update the features documentation section's index
 page

---
 docs_new/features/available_models.md  |   1 -
 docs_new/features/index.md             |  18 +++-
 docs_new/features/types/cfg.md         | 113 -------------------------
 docs_new/features/types/index.md       |  36 --------
 docs_new/features/types/json_schema.md |  81 ------------------
 docs_new/features/types/regex.md       |  72 ----------------
 6 files changed, 17 insertions(+), 304 deletions(-)
 delete mode 100644 docs_new/features/available_models.md
 delete mode 100644 docs_new/features/types/cfg.md
 delete mode 100644 docs_new/features/types/index.md
 delete mode 100644 docs_new/features/types/json_schema.md
 delete mode 100644 docs_new/features/types/regex.md

diff --git a/docs_new/features/available_models.md b/docs_new/features/available_models.md
deleted file mode 100644
index cb9497be1..000000000
--- a/docs_new/features/available_models.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.models
diff --git a/docs_new/features/index.md b/docs_new/features/index.md
index bc09a026e..00df02ba3 100644
--- a/docs_new/features/index.md
+++ b/docs_new/features/index.md
@@ -1,3 +1,19 @@
 # Features
 
-This needs to contain a list of all features available in Outlines.
+This section presents in details the different features of Outlines.
+
+## Core Concepts
+
+- [Models](./models/index.md)
+- [Output Types](./core/output_types.md)
+- [Generators](./core/generator.md)
+
+## Utility
+
+- [Applications](./utility/application.md)
+- [Templates](./utility/templates.md)
+- [Regex DSL](./utility/regex_dsl.md)
+
+## Advanced
+
+- [Logits Processors](./advanced/logits_processors.md)
diff --git a/docs_new/features/types/cfg.md b/docs_new/features/types/cfg.md
deleted file mode 100644
index 12e38fd50..000000000
--- a/docs_new/features/types/cfg.md
+++ /dev/null
@@ -1,113 +0,0 @@
----
-title: Context-Free Grammars
----
-
-# Context-Free Grammars
-
-::: outlines.types.cfg
-
-Context-free grammars (CFGs) provide a powerful way to define complex structured outputs. They're particularly useful for generating text with a specific syntax, such as code snippets or custom formats.
-
-## Basic Usage
-
-```python
-from outlines import Generator, from_transformers
-from outlines.types import CFG
-
-# Define a grammar for simple arithmetic expressions
-grammar = """
-start: expr
-expr: term "+" expr | term
-term: factor "*" term | factor
-factor: NUMBER | "(" expr ")"
-NUMBER: /[0-9]+/
-"""
-
-# Create a CFG output type
-cfg_output = CFG(grammar)
-
-# Initialize a model
-model = from_transformers(...)
-
-# Create a generator with the CFG output type
-generator = Generator(model, cfg_output)
-
-# Generate an arithmetic expression
-prompt = "Generate a simple arithmetic expression:"
-result = generator(prompt)
-
-print(result)  # Example output: "3 * (4 + 5)"
-```
-
-## Grammar Definition
-
-Context-free grammars in Outlines use the [Lark grammar format](https://lark-parser.readthedocs.io/en/latest/grammar/). A grammar consists of rules that define the structure of the text you want to generate.
-
-```python
-# Grammar for a simple SQL query
-sql_grammar = """
-start: select_stmt
-select_stmt: "SELECT" columns "FROM" table ("WHERE" condition)?
-columns: column ("," column)*
-column: IDENTIFIER
-table: IDENTIFIER
-condition: IDENTIFIER comparison_op value
-comparison_op: "=" | ">" | "<" | ">=" | "<=" | "!="
-value: NUMBER | STRING
-IDENTIFIER: /[a-zA-Z_][a-zA-Z0-9_]*/
-NUMBER: /[0-9]+/
-STRING: /"[^"]*"/
-"""
-```
-
-## Advanced Usage
-
-Context-free grammars can be used for complex generation tasks like code generation:
-
-```python
-from outlines import Generator, from_transformers
-from outlines.types import CFG
-
-# Grammar for simple Python functions
-python_grammar = """
-start: function
-function: "def" IDENTIFIER "(" params? ")" ":" NEWLINE INDENT body DEDENT
-params: param ("," param)*
-param: IDENTIFIER
-body: statement+
-statement: assignment NEWLINE | if_stmt | return_stmt NEWLINE
-assignment: IDENTIFIER "=" expr
-if_stmt: "if" expr ":" NEWLINE INDENT body DEDENT
-return_stmt: "return" expr
-expr: IDENTIFIER | NUMBER | STRING | expr "+" expr | expr "-" expr
-IDENTIFIER: /[a-zA-Z_][a-zA-Z0-9_]*/
-NUMBER: /[0-9]+/
-STRING: /"[^"]*"/
-INDENT: /\t/
-DEDENT: /\t/
-NEWLINE: /\n/
-"""
-
-# Create a CFG output type
-cfg_output = CFG(python_grammar)
-
-# Initialize a model
-model = from_transformers(...)
-
-# Create a generator
-generator = Generator(model, cfg_output)
-
-# Generate a Python function
-prompt = "Generate a Python function that calculates the factorial of a number:"
-result = generator(prompt)
-
-print(result)
-```
-
-For more detailed examples, see the [Context-Free Grammar guide](/user_guide/structured_generation/cfg.md).
-
-## Notes
-
-- Context-Free Grammar support in Outlines is experimental but powerful
-- Complex grammars may take longer to process
-- For simple structures, consider using regex or JSON schema instead
diff --git a/docs_new/features/types/index.md b/docs_new/features/types/index.md
deleted file mode 100644
index d3d1d3632..000000000
--- a/docs_new/features/types/index.md
+++ /dev/null
@@ -1,36 +0,0 @@
----
-title: Output Types
----
-
-# Output Types
-
-Outlines provides several ways to define structured outputs for language models.
-
-## Available Output Types
-
-### JSON Schema
-
-JSON Schema allows you to define the structure of JSON data that you want to generate. Outlines supports multiple ways to define JSON schemas:
-
-- JSON Schema objects
-- Pydantic models
-- TypedDict
-- Dataclasses
-
-[Learn more about JSON Schema →](json_schema.md)
-
-### Regular Expressions
-
-Regular expressions provide a way to constrain the model's output to match specific text patterns. This is useful for generating formatted data like dates, phone numbers, or other structured text.
-
-[Learn more about Regular Expressions →](regex.md)
-
-### Context-Free Grammars
-
-Context-free grammars (CFGs) provide the most flexible way to define complex structured outputs. They're particularly useful for generating text with a specific syntax, such as code or custom formats.
-
-[Learn more about Context-Free Grammars →](cfg.md)
-
-## How Output Types Work
-
-Internally, Outlines converts output type definitions into finite state machines (FSMs) that constrain the model's generation process. This happens automatically when you pass an output type to the `Generator` class.
diff --git a/docs_new/features/types/json_schema.md b/docs_new/features/types/json_schema.md
deleted file mode 100644
index 74d068262..000000000
--- a/docs_new/features/types/json_schema.md
+++ /dev/null
@@ -1,81 +0,0 @@
----
-title: JSON Schema
----
-
-# JSON Schema
-
-::: outlines.types.json_schema
-
-JSON Schema provides a way to define the structure of JSON data that you want to generate. Outlines supports multiple ways to define JSON schemas.
-
-## Basic Usage
-
-```python
-from outlines import Generator, from_transformers
-from outlines.types import JsonSchema
-
-# Define a simple JSON schema
-person_schema = {
-    "type": "object",
-    "properties": {
-        "name": {"type": "string"},
-        "age": {"type": "integer"},
-        "email": {"type": "string"}
-    },
-    "required": ["name", "age"]
-}
-
-# Create a JSON schema output type
-json_output = JsonSchema(person_schema)
-
-# Initialize a model
-model = from_transformers(...)
-
-# Create a generator with the JSON schema output type
-generator = Generator(model, json_output)
-
-# Generate JSON data
-prompt = "Generate information for a fictional person:"
-result = generator(prompt)
-
-print(result)
-# Example output: {"name": "John Doe", "age": 32, "email": "john.doe@example.com"}
-```
-
-## Using Pydantic Models
-
-```python
-from pydantic import BaseModel
-from typing import List, Optional
-from outlines import Generator, from_transformers
-
-# Define a Pydantic model
-class Person(BaseModel):
-    name: str
-    age: int
-    email: Optional[str] = None
-
-# Initialize a model
-model = from_transformers(...)
-
-# Create a generator with the Pydantic model
-generator = Generator(model, Person)
-
-# Generate JSON data
-prompt = "Generate information for a fictional person:"
-result = generator(prompt)
-
-# Parse the result into a Pydantic model
-person = Person.model_validate_json(result)
-print(f"Name: {person.name}, Age: {person.age}")
-```
-
-## Other Schema Types
-
-Outlines also supports:
-
-- TypedDict
-- Dataclasses
-- Direct JSON Schema objects
-
-For more detailed examples, see the [JSON Generation guide](/user_guide/structured_generation/json.md).
diff --git a/docs_new/features/types/regex.md b/docs_new/features/types/regex.md
deleted file mode 100644
index b3e9c94ec..000000000
--- a/docs_new/features/types/regex.md
+++ /dev/null
@@ -1,72 +0,0 @@
----
-title: Regular Expressions
----
-
-# Regular Expressions
-
-::: outlines.types.regex
-
-Regular expressions provide a way to constrain the model's output to match specific text patterns.
-
-## Basic Usage
-
-```python
-from outlines import Generator, from_transformers
-from outlines.types import Regex
-
-# Create a regex pattern for a North American phone number
-phone_regex = Regex(r"\(\d{3}\) \d{3}-\d{4}")
-
-# Initialize a model
-model = from_transformers(...)
-
-# Create a generator with the regex output type
-generator = Generator(model, phone_regex)
-
-# Generate a phone number
-prompt = "Generate a random phone number in the format (XXX) XXX-XXXX:"
-result = generator(prompt)
-
-print(result)  # Example output: "(555) 123-4567"
-```
-
-## Complex Patterns
-
-```python
-from outlines import Generator, from_transformers
-from outlines.types import Regex
-
-# Create a regex pattern for an email address
-email_regex = Regex(r"[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}")
-
-# Initialize a model
-model = from_transformers(...)
-
-# Create a generator with the regex output type
-generator = Generator(model, email_regex)
-
-# Generate an email address
-prompt = "Generate a professional email address for John Doe:"
-result = generator(prompt)
-
-print(result)  # Example output: "john.doe@example.com"
-```
-
-## Creating a Regex Type
-
-The `Regex` class accepts a string containing a regular expression pattern:
-
-```python
-from outlines.types import Regex
-
-# Simple pattern for a 5-digit US ZIP code
-zip_code = Regex(r"\d{5}")
-
-# Pattern for a date in YYYY-MM-DD format
-date_format = Regex(r"\d{4}-\d{2}-\d{2}")
-
-# Pattern for a URL
-url_pattern = Regex(r"https?://[a-zA-Z0-9.-]+\.[a-zA-Z]{2,}(/[a-zA-Z0-9./_-]*)?")
-```
-
-For more detailed examples, see the [Regular Expression guide](/user_guide/structured_generation/regex.md).

From 39b20b73ef3c97449f8ffba658c2b22dadc40dda Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 2 Jun 2025 15:09:59 +0200
Subject: [PATCH 342/505] Align examples lists in documentation

---
 docs_new/examples/index.md | 14 +++++++++++---
 mkdocs.yml                 | 17 +++++++++--------
 2 files changed, 20 insertions(+), 11 deletions(-)

diff --git a/docs_new/examples/index.md b/docs_new/examples/index.md
index c36b98969..0181d137a 100644
--- a/docs_new/examples/index.md
+++ b/docs_new/examples/index.md
@@ -4,15 +4,23 @@ This part of the documentation provides a few cookbooks that you can browse to g
 
 - [Classification](classification.md): Classify customer requests.
 - [Named Entity Extraction](extraction.md): Extract information from pizza orders.
-- [Dating Profile](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
+- [Dating Profiles](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
 - [Chain Of Density](chain_of_density.md): Summarize documents using chain of density prompting and JSON-structured generation.
 - [Playing Chess](models_playing_chess.md): Make Phi-3 Mini play chess against itself using regex-structured generation.
 - [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
 - [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
 - [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
+- [Structured Generation Workflow](structured_generation_workflow.md):
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
-- [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
 - [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
-- [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
 - [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
+- [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
+- [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
+- [Extract Events Details](extract_event_details.md):
+
+Run Outlines on the cloud:
+
+- [BentoML](deploy-using-bentoml.md)
+- [Cerebrium](deploy-using-cerebrium.md)
+- [Modal](deploy-using-modal.md)
diff --git a/mkdocs.yml b/mkdocs.yml
index 5419389f5..e5d3d5776 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -84,7 +84,7 @@ extra_css:
 plugins:
     - blog
     - mkdocstrings:
-          default_handler: python
+          update_features_models_documentation default_handler: python
           handlers:
               python:
                   options:
@@ -159,6 +159,7 @@ nav:
           - Features: features/index.md
           - API Reference: api_reference/index.md
           - Examples: examples/index.md
+
     - Guide:
           - Introduction:
                 - Get started: getting_started/index.md
@@ -223,20 +224,20 @@ nav:
           - examples/index.md
           - Classification: examples/classification.md
           - Named Entity Extraction: examples/extraction.md
-          - Generate synthetic data: examples/dating_profiles.md
-          - Summarize a document: examples/chain_of_density.md
+          - Dating Profiles: examples/dating_profiles.md
+          - Chain of Density: examples/chain_of_density.md
           - Playing chess: examples/models_playing_chess.md
-          - Perspective-taking prompting: examples/simtom.md
-          - Question-answering with citations: examples/qa-with-citations.md
+          - SimTom: examples/simtom.md
+          - Q&A with Citations: examples/qa-with-citations.md
           - Knowledge Graph Extraction: examples/knowledge_graph_extraction.md
           - Structured Generation Workflow: examples/structured_generation_workflow.md
           - Chain of Thought (CoT): examples/chain_of_thought.md
           - ReAct Agent: examples/react_agent.md
           - Vision-Language Models: examples/atomic_caption.md
           - Structured Generation from PDFs: examples/read-pdfs.md
-          - Earnings reports to CSV: examples/earnings-reports.md
-          - Digitizing receipts with vision models: examples/receipt-digitization.md
-          - Extract events details from text: examples/extract_event_details.md
+          - Earnings Reports to CSV: examples/earnings-reports.md
+          - Receipt Digitization: examples/receipt-digitization.md
+          - Extract Events Details: examples/extract_event_details.md
           - Run on the cloud:
                 - BentoML: examples/deploy-using-bentoml.md
                 - Cerebrium: examples/deploy-using-cerebrium.md

From 4a28911d925493c9d203e3ee6af2ad18b3339221 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 2 Jun 2025 18:52:21 +0200
Subject: [PATCH 343/505] Update the guides in the documentation

---
 docs_new/features/models/llamacpp.md          |   2 +-
 docs_new/getting_started/index.md             | 350 -----------------
 .../{user_guide => guide}/chat_templating.md  |   0
 docs_new/guide/core_concepts.md               |  13 +
 docs_new/guide/getting_started.md             | 363 ++++++++++++++++++
 docs_new/guide/installation.md                |  73 ++++
 docs_new/{user_guide => guide}/migration.md   |  10 +-
 .../selecting_an_inference_backend.md         |   0
 docs_new/installation.md                      |  86 -----
 docs_new/user_guide/index.md                  |   1 -
 docs_new/user_guide/prompting/applications.md |   0
 docs_new/user_guide/prompting/templates.md    |   0
 docs_new/user_guide/regex_dsl.md              | 344 -----------------
 .../user_guide/structured_generation/cfg.md   |   0
 .../user_guide/structured_generation/index.md |  13 -
 .../user_guide/structured_generation/json.md  |   0
 .../structured_generation/multiple_choice.md  |   0
 .../user_guide/structured_generation/regex.md |   0
 docs_new/user_guide/unstructured.md           |   1 -
 mkdocs.yml                                    |  30 +-
 20 files changed, 464 insertions(+), 822 deletions(-)
 delete mode 100644 docs_new/getting_started/index.md
 rename docs_new/{user_guide => guide}/chat_templating.md (100%)
 create mode 100644 docs_new/guide/core_concepts.md
 create mode 100644 docs_new/guide/getting_started.md
 create mode 100644 docs_new/guide/installation.md
 rename docs_new/{user_guide => guide}/migration.md (85%)
 rename docs_new/{user_guide => guide}/selecting_an_inference_backend.md (100%)
 delete mode 100644 docs_new/installation.md
 delete mode 100644 docs_new/user_guide/index.md
 delete mode 100644 docs_new/user_guide/prompting/applications.md
 delete mode 100644 docs_new/user_guide/prompting/templates.md
 delete mode 100644 docs_new/user_guide/regex_dsl.md
 delete mode 100644 docs_new/user_guide/structured_generation/cfg.md
 delete mode 100644 docs_new/user_guide/structured_generation/index.md
 delete mode 100644 docs_new/user_guide/structured_generation/json.md
 delete mode 100644 docs_new/user_guide/structured_generation/multiple_choice.md
 delete mode 100644 docs_new/user_guide/structured_generation/regex.md
 delete mode 100644 docs_new/user_guide/unstructured.md

diff --git a/docs_new/features/models/llamacpp.md b/docs_new/features/models/llamacpp.md
index d269c4111..1abef52a3 100644
--- a/docs_new/features/models/llamacpp.md
+++ b/docs_new/features/models/llamacpp.md
@@ -8,7 +8,7 @@ Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/l
 
 !!! Installation
 
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration: `pip install llama_cpp`
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration: `pip install llama-cpp-python`
 
     See the [llama-cpp-python Github page](https://github.com/abetlen/llama-cpp-python?tab=readme-ov-file#supported-backends) for instructions on installing with CUDA, Metal, ROCm and other backends.
 
diff --git a/docs_new/getting_started/index.md b/docs_new/getting_started/index.md
deleted file mode 100644
index ca30f0708..000000000
--- a/docs_new/getting_started/index.md
+++ /dev/null
@@ -1,350 +0,0 @@
----
-title: Getting Started
----
-
-# Getting Started
-
-## Installation
-
-We recommend using `uv` to install Outlines. You can find `uv` installation instructions [here](https://github.com/astral-sh/uv).
-
-To install Outlines with the `transformers` backend, run:
-
-```bash
-uv pip install 'outlines[transformers]'
-```
-
-or the classic `pip`:
-
-```bash
-pip install 'outlines[transformers]'
-```
-
-For other backends, see the [installation guide](../installation).
-
-## Quick start
-
-Outlines wraps around a variety of LLM inference backends, described in the [installation guide](../installation). The following example shows how to use Outlines with HuggingFace's [`transformers` library](https://huggingface.co/docs/transformers/en/index).
-
-=== "Transformers"
-
-    ```python
-    import outlines
-    from transformers
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # Create an Outlines model using transformers
-    model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(model_name),
-        transformers.AutoTokenizer.from_pretrained(model_name),
-    )
-
-    # Generate a response
-    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
-    ```
-
-=== "OpenAI"
-
-    ```python
-    import outlines
-    from openai import OpenAI
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # Create an OpenAI client instance
-    openai_client = OpenAI()
-
-    # Create an Outlines model
-    model = outlines.from_openai(openai_client, "gpt-4o")
-
-    # Generate a response
-    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
-    ```
-
-=== "vLLM (offline)"
-
-    ```python
-    import outlines
-    from pydantic import BaseModel
-    from vllm import LLM
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        role: str
-        age: int
-
-    # Model to use, it will be downloaded from the HuggingFace hub
-    model_id = "microsoft/Phi-3-mini-4k-instruct"
-
-    # Create a vLLM model
-    vllm_model = LLM(model_id)
-
-    # Create an Outlines model
-    model = outlines.from_vllm_offline(vllm_model)
-
-    # Generate a response
-    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
-    ```
-
-=== "vLLM (online)"
-
-    ```python
-    import outlines
-    from openai import OpenAI
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # You must have a separete vLLM server running
-    # Create an OpenAI client with the base URL of the VLLM server
-    openai_client = OpenAI(base_url="http://localhost:11434/v1")
-
-    # Specify the model available on the VLLM server to use
-    model_id = "microsoft/Phi-3-mini-4k-instruct"
-
-    # Create an Outlines model
-    model = outlines.from_vllm(openai_client, model_id)
-
-    # Generate a response
-    response = model("Create a character.", Person)
-    # { "name": "John", "age": 30 }
-    ```
-
-=== "llama.cpp"
-
-    ```python
-    import outlines
-    from llama_cpp import Llama
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # Model to use, it will be downloaded from the HuggingFace hub
-    repo_id = "TheBloke/Llama-2-13B-chat-GGUF"
-    file_name = "llama-2-13b-chat.Q4_K_M.gguf"
-
-    # Create a Llama.cpp model
-    llama_cpp_model = Llama.from_pretrained(repo_id, file_name)
-
-    # Create an Outlines model
-    model = outlines.from_llamacpp(llama_cpp_model)
-
-    # Generate a response
-    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
-    ```
-
-=== "Dottxt"
-
-    ```python
-    import outlines
-    from dottxt.client import Dottxt
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # Create an Dottxt client
-    client = Dottxt()
-
-    # Create an Outlines model
-    model = outlines.from_dottxt(client)
-
-    # Generate a response
-    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
-    ```
-
-=== "Anthropic"
-
-    ```python
-    import outlines
-    from anthropic import Anthropic
-    from pydantic import BaseModel
-
-    # Create an Anthropic client
-    client = Anthropic()
-
-    # Create an Outlines model
-    model = outlines.from_anthropic(client, "claude-3-haiku-20240307")
-
-    # Generate a response
-    response = model("Create a character.", max_tokens=20) # Here is a character I have created:\n\nName: Ayla Samara
-    ```
-
-=== "Ollama"
-
-    ```python
-    import outlines
-    from ollama import Client
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # Create an Ollama client
-    client = Client()
-
-    # Create an Outlines model, the model must be available on your system
-    model = outlines.from_ollama(client, "tinyllama")
-
-    # Generate a response
-    response = model("Create a character.", Person)
-    # { "name": "John", "age": 30 }
-    ```
-
-=== "mlx-lm"
-
-    ```python
-    import outlines
-    import mlx_lm
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        role: str
-        age: int
-
-    # Create an MLXLM model with the output of mlx_lm.load
-    # The model will be downloaded from the HuggingFace hub
-    model = outlines.from_mlxlm(mlx_lm.load(
-        "mlx-community/SmolLM-135M-Instruct-4bit"
-    ))
-
-    # Generate a response
-    response = model("Create a character.", Person)
-    # { "name": "John", "age": 30 }
-    ```
-
-=== "Gemini"
-
-    ```python
-    import outlines
-    from google.generativeai import GenerativeModel
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # Create a Gemini client
-    client = GenerativeModel()
-
-    # Create an Outlines model
-    model = outlines.from_gemini(client)
-
-    # Generate a response
-    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
-    ```
-
-=== "SgLang"
-
-    ```python
-    # SgLang
-
-    import outlines
-    from openai import OpenAI
-    from pydantic import BaseModel
-
-    # Define a Pydantic model describing the output format
-    class Person(BaseModel):
-        # Set name and age fields to string and int respectively
-        name: str
-        age: int
-
-    # You must have a separete SgLang server running
-    # Create an OpenAI client with the base URL of the SgLang server
-    openai_client = OpenAI(base_url="http://localhost:11434/v1")
-
-    # Create an Outlines model
-    model = outlines.from_sglang(openai_client)
-
-    # Generate a response
-    response = model("Create a character.", Person) # { "name": "John", "age": 30 }
-    ```
-
-### JSON
-
-```python
-
-# Define the input text
-person_text = """
-John Doe
-30
-john.doe@example.com
-"""
-
-# Apply chat templating to the input text
-prompt = tokenizer.apply_chat_template(
-    [
-        {"role": "system", "content": """
-        You are a master of extracting information from text.
-        """},
-        {"role": "user", "content": person_text}
-    ],
-    tokenize=False
-)
-
-# Generate the output
-result = model(
-    prompt,
-    Person,
-
-    # Note: transformers has an extremely small default
-    # max_new_tokens, which is often not enough for the
-    # full JSON output. You will experience errors if your
-    # model is unable to generate the full JSON output due
-    # to the max_new_tokens limit.
-    max_new_tokens=100
-)
-
-print(result)
-```
-
-Result:
-
-```json
-{
-    "name": "John Doe",
-    "age": 30,
-    "email": "john.doe@example.com"
-}
-```
-
-### Regex
-
-> [!NOTE]
-> Insert regex example here
-
-### Multiple choice
-
-> [!NOTE]
-> Insert multiple choice example here
diff --git a/docs_new/user_guide/chat_templating.md b/docs_new/guide/chat_templating.md
similarity index 100%
rename from docs_new/user_guide/chat_templating.md
rename to docs_new/guide/chat_templating.md
diff --git a/docs_new/guide/core_concepts.md b/docs_new/guide/core_concepts.md
new file mode 100644
index 000000000..7f53983ec
--- /dev/null
+++ b/docs_new/guide/core_concepts.md
@@ -0,0 +1,13 @@
+---
+title: Core concepts
+---
+
+# Core concepts
+
+Coming soon. This will document various concepts at a high level, so users can understand Outlines before diving into specific implementations.
+
+1. Constrained decoding, tokens, and the basics of logit biasing
+2. Different ways to define output structure (regex, JSON schema, Pydantic models, context-free grammars)
+3. How finite state machines are used to guarantee output structure
+4. `Generator`, `Application`, `Template`,
+5. Prompt engineering vs. structured generation
diff --git a/docs_new/guide/getting_started.md b/docs_new/guide/getting_started.md
new file mode 100644
index 000000000..80a6c2456
--- /dev/null
+++ b/docs_new/guide/getting_started.md
@@ -0,0 +1,363 @@
+---
+title: Getting Started
+---
+
+# Getting Started
+
+## Installation
+
+We recommend using `uv` to install Outlines. You can find `uv` installation instructions [here](https://github.com/astral-sh/uv).
+
+```bash
+uv pip install 'outlines[transformers]'
+```
+
+or the classic `pip`:
+
+```bash
+pip install 'outlines[transformers]'
+```
+
+For more information, see the [installation guide](./installation).
+
+## Creating a Model
+
+Outlines contains a variety of models that wrap LLM inference engines/clients. For each of them, you need to install the model's associated library as described in the [installation guide](../installation).
+
+The full list of available models along with detailed explanation on how to use them can be found in the [models page](../features/models/index.md) of the Features section of the documentation.
+
+For a quick start, you can find below an example of how to initialize all supported models in Outlines:
+
+=== "Anthropic"
+
+    ```python
+    import outlines
+    from anthropic import Anthropic
+
+    # Create an Anthropic client
+    anthropic_client = Anthropic()
+
+    # Create an Outlines model
+    model = outlines.from_anthropic(anthropic_client, "claude-3-haiku-20240307")
+    ```
+
+=== "Dottxt"
+
+    ```python
+    import outlines
+    from dottxt.client import Dottxt
+
+    # Create an Dottxt client
+    dottxt_client = Dottxt()
+
+    # Create an Outlines model
+    model = outlines.from_dottxt(dottxt_client)
+    ```
+
+=== "Gemini"
+
+    ```python
+    import outlines
+    from google.generativeai import GenerativeModel
+
+    # Create a Gemini client
+    gemini_client = GenerativeModel()
+
+    # Create an Outlines model
+    model = outlines.from_gemini(gemini_client, "gemini-1-5-flash")
+    ```
+
+=== "llama.cpp"
+
+    ```python
+    import outlines
+    from llama_cpp import Llama
+
+    # Model to use, it will be downloaded from the HuggingFace hub
+    repo_id = "TheBloke/Llama-2-13B-chat-GGUF"
+    file_name = "llama-2-13b-chat.Q4_K_M.gguf"
+
+    # Create a Llama.cpp model
+    llama_cpp_model = Llama.from_pretrained(repo_id, file_name)
+
+    # Create an Outlines model
+    model = outlines.from_llamacpp(llama_cpp_model)
+    ```
+
+=== "mlx-lm"
+
+    ```python
+    import outlines
+    import mlx_lm
+
+    # Create an MLXLM model with the output of mlx_lm.load
+    # The model will be downloaded from the HuggingFace hub
+    model = outlines.from_mlxlm(
+        **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    )
+    ```
+
+=== "Ollama"
+
+    ```python
+    import outlines
+    from ollama import Client
+
+    # Create an Ollama client
+    ollama_client = Client()
+
+    # Create an Outlines model, the model must be available on your system
+    model = outlines.from_ollama(ollama_client, "tinyllama")
+    ```
+
+=== "OpenAI"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+
+    # Create an OpenAI client instance
+    openai_client = OpenAI()
+
+    # Create an Outlines model
+    model = outlines.from_openai(openai_client, "gpt-4o")
+    ```
+
+=== "SgLang"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+
+    # You must have a separete SgLang server running
+    # Create an OpenAI client with the base URL of the SgLang server
+    openai_client = OpenAI(base_url="http://localhost:11434/v1")
+
+    # Create an Outlines model
+    model = outlines.from_sglang(openai_client)
+    ```
+
+=== "TGI"
+
+    ```python
+    # SgLang
+
+    import outlines
+    from huggingface_hub import InferenceClient
+
+    # You must have a separete TGI server running
+    # Create an InferenceClient client with the base URL of the TGI server
+    tgi_client = InferenceClient("http://localhost:8080")
+
+    # Create an Outlines model
+    model = outlines.from_tgi(tgi_client)
+    ```
+
+=== "Transformers"
+
+    ```python
+    import outlines
+    from transformers
+
+    # Define the model you want to use
+    model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+
+    # Create a HuggingFace model and tokenizer
+    hf_model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
+    hf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
+
+    # Create an Outlines model
+    model = outlines.from_transformers(hf_model, hf_tokenizer)
+    ```
+
+=== "vLLM (offline)"
+
+    ```python
+    import outlines
+    from vllm import LLM
+
+    # Create a vLLM model
+    vllm_model = LLM("microsoft/Phi-3-mini-4k-instruct")
+
+    # Create an Outlines model
+    model = outlines.from_vllm_offline(vllm_model)
+    ```
+
+=== "vLLM (online)"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+
+    # You must have a separete vLLM server running
+    # Create an OpenAI client with the base URL of the VLLM server
+    openai_client = OpenAI(base_url="http://localhost:11434/v1")
+
+    # Create an Outlines model
+    model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
+    ```
+
+## Generating Text
+
+Once you have created the Outlines model for your inference engine/client, you are already all set to generate text! Models are callable such that you can simply call them with a text prompt. For instance:
+
+```python
+model = <your_model_as_defined_above>
+
+# Call the model to generate text
+result = model("Write a short story about a cat.")
+print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+```
+
+Most models also support streaming through the use of a `streaming` method. You can directly use with a prompt just like regular text generation. For instance:
+
+```python
+model = <your_model_as_defined_above>
+
+# Stream text
+for chunk in model.streaming("Write a short story about a cat.")
+    print(chunk) # 'In ...'
+```
+
+## Structured Generation
+
+Outlines follows a simple pattern that mirrors Python's own type system for structured output. Simply specify the desired output type as you would when using type hinting with a function, and Outlines will ensure your data matches that structure exactly.
+
+Supported output types can be organized in 5 categories:
+
+- [Basic Types](../../features/core/output_types#basic-python-types): `int`, `float`, `bool`...
+- [Multiple Choices](../../features/core/output_types#multiple-choices): using `Literal` or `Enum`
+- [JSON Schemas](../../features/core/output_types#json-schemas): using a wide range of possible objects including Pydantic models and dataclasses
+- [Regex](../../features/core/output_types#regex-patterns): through the Outlines's `Regex` object
+- [Context-free Grammarq](../../features/core/output_types#context-free-grammars): through the Outlines's `CFG` object
+
+Consult the section on [Output Types](../../features/core/output_types.md) in the features documention for more detailed information on all supported types for each output type category.
+
+In the meantime, you can find below examples of using each of the five output type categories:
+
+=== "Basic Yypes"
+
+    ```python
+    model = <your_model_as_defined_above>
+
+    # Generate an integer
+    result = model("How many countries are there in the world?", int)
+    print(result) # '200'
+    ```
+
+=== "Multiple Choice"
+
+    ```python
+    from enum import Enum
+
+    # Define our multiple choice output type
+    class PizzaOrBurger(Enum):
+        pizza = "pizza
+        burger = "burger
+
+    model = <your_model_as_defined_above>
+
+    # Generate text corresponding to either of the choices defined above
+    result = model("What do you want to eat, a pizza or a burger?", PizzaOrBurger)
+    print(result) # 'pizza'
+    ```
+
+=== "JSON Schemas"
+
+    ```python
+    from datetime import date
+    from typing import Dict, List, Union
+    from pydantic import BaseModel
+
+    model = <your_model_as_defined_above>
+
+    # Define the class we will use as an output type
+    class Character(BaseModel):
+        name: str
+        birth_date: date
+        skills: Union[Dict, List[str]]
+
+    # Generate a character
+    result = model("Create a character", Character)
+    print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
+    print(Characte.model_validate_json(result);) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
+    ```
+
+=== "Regex"
+
+    ```python
+    from outlines.types import Regex
+
+    model = <your_model_as_defined_above>
+
+    # Define our regex for a 3 digit number
+    output_type = Regex(r"[0-9]{3}")
+
+    # Generate the number
+    result = model("Write a 3 digit number", output_type)
+    print(result) # '236'
+    ```
+
+=== "Context-free Grammars"
+
+    ```python
+    from outlines.types import CFG
+
+    model = <your_model_as_defined_above>
+
+    # Define your Lark grammar as string
+    arithmetic_grammar = """
+        ?start: sum
+
+        ?sum: product
+            | sum "+" product   -> add
+            | sum "-" product   -> sub
+
+        ?product: atom
+            | product "*" atom  -> mul
+            | product "/" atom  -> div
+
+        ?atom: NUMBER           -> number
+            | "-" atom         -> neg
+            | "(" sum ")"
+
+        %import common.NUMBER
+        %import common.WS_INLINE
+
+        %ignore WS_INLINE
+    """
+
+    # Generate an arithmetic operation
+    result = model("Write an arithmetic operation", CFG(grammar_string))
+    print(result) # '2 + 3'
+    ```
+
+It's important to note that not all output types are available for all models due to limitations in the underline inference libraries. The [Models](../features/models/index.md) section of the features documentation includes a features matrix to easily visualize output type availabilities.
+
+## Generators
+
+Generators are an important type of objects in Outlines that are used to encapsulate a model and an output type. After having created a generator, you can call it using a similar interface to a model and it will generate text conforming to the output type you initially provided.
+
+This feature is useful if you want to generate text several times for given model and output type. Not only does it prevent having to include the same output type at each call, but it also allows us to compile the output type only once instead of doing it at each generation (which is important for local models as this operation can be expensive).
+
+For instance:
+
+```python
+from typing import Literal
+from outlines import Generator
+
+model = <your_model_as_defined_above>
+
+# Create a generator
+generator = Generator(model, Literal["pizza", "burger"])
+
+# Call it as you would call a model
+result = generator("What do you want to eat, a pizza or a burger?")
+print(result) # pizza
+```
+
+You can find more information on generators in the dedicated page on [Generators](../features/core/generator.md) in the features documentation.
+
+## Other features
+
+On top of more detailed explanation on the concepts already discussed here, the [Features](../features/index.md) section of the documentation contains information on additional Outlines features such as applications, prompt templates, the regex DSL...
diff --git a/docs_new/guide/installation.md b/docs_new/guide/installation.md
new file mode 100644
index 000000000..c626841fd
--- /dev/null
+++ b/docs_new/guide/installation.md
@@ -0,0 +1,73 @@
+---
+title: Installation
+---
+
+# Installation
+
+## Dependency Management
+
+We recommend using modern Python packaging tools such as `uv` for managing python dependencies.
+
+### uv (Recommended)
+
+```bash
+# Install uv
+curl -LsSf https://astral.sh/uv/install.sh | sh
+
+# Create a virtual environment and install Outlines
+uv venv
+source .venv/bin/activate
+uv pip install outlines
+```
+
+or with basic pip:
+
+```bash
+pip install outlines
+```
+
+
+
+
+
+## Optional Dependencies
+
+To use Outlines models, you need to install the Python libraries for the associated inference engines/clients. Such libraries are not part of the general installation as you should only install the libraries needed for the specific models you want to use.
+
+Outlines models with the installation of their associated additional depencies:
+
+- [Anthropic](features/models/anthropic.md): `pip install anthropic`
+- [Dottxt](features/models/dottxt.md): `pip install dottxt`
+- [Gemini](features/models/gemini.md): `pip install google-generativeai`
+- [Llamacpp](features/models/llamacpp.md): `pip install llama-cpp-python`
+- [Mlx-lm](features/models/mlxlm.md): `pip install mlx mlx-lm`
+- [Ollama](features/models/ollama.md): `pip install ollama` (after having downloaded Ollama in your system)
+- [OpenAI](features/models/openai.md): `pip install openai`
+- [SGLang](features/models/sglang.md): `pip install openai`
+- [TGI](features/models/tgi.md): `pip install huggingface_hub`
+- [Transformers](features/models/transformers.md): `pip install transformers`
+- [TransformersMultiModal](features/models/transformers_multimodal.md): `pip install transformers`
+- [vLLM (online server)](features/models/vllm.md): `pip install openai`
+- [vLLM (offline)](features/models/vllm_offline.md): `pip install vllm`
+
+If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
+
+!!! warning "Hardware Requirements"
+
+    If you are using a local model, your model may require specific hardware. Please check the documentation for these libraries.
+
+    Some libraries like `vllm` and `llama-cpp-python` require specific hardware, such as a compatible GPU. `mlx-lm` on its side is designed for Apple Silicon, so it may not be appropriate for your use case if you are on a different platform.
+
+## Bleeding Edge
+
+You can install the latest version of Outlines from the repository's `main` branch:
+
+```sh
+pip install git+https://github.com/dottxt-ai/outlines.git@main
+```
+
+This can be useful, for instance, when a fix has been merged but not yet released.
+
+## Installing for Development
+
+See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs_new/user_guide/migration.md b/docs_new/guide/migration.md
similarity index 85%
rename from docs_new/user_guide/migration.md
rename to docs_new/guide/migration.md
index f931641f7..df766d8c8 100644
--- a/docs_new/user_guide/migration.md
+++ b/docs_new/guide/migration.md
@@ -4,7 +4,7 @@ Outlines 1.0 introduces some breaking changes that affect the way you use the li
 
 This guide will help you migrate your code to the new version.
 
-All previous functionality will be supported until Outlines version 1.5.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
+All previous functionality will be supported until Outlines version 1.1.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
 
 ## Removed or modified features
 - [Generate functions](#generate-functions)
@@ -18,7 +18,7 @@ All previous functionality will be supported until Outlines version 1.5.0, but a
 
 The whole `generate` module has been removed. That includes the functions `generate.cfg`, `generate.choice`, `generate.format`,`generate.fsm`, `generate.json`, `generate.regex` and `generate.text`.
 
-You should replace these functions by the `Generator` object along with the right output type as an argument (on top of the model). The output type can either be a python type or be an object from the `outlines.types` module. You can find more information about the output types in the [output types](TODO: link to the documentatoin on output types) guide.
+You should replace these functions by the [`Generator`](../features/core/generator.md) object along with the right output type as an argument (on top of the model). The output type can either be a python type or be an object from the `outlines.types` module. You can find more information about the output types in the [Output Types](../features/core/output_types.md) section of the features documentation.
 
 Associated v1 output types for each deprecated function:
 - `generate.cfg` -> `outlines.types.CFG`
@@ -52,7 +52,7 @@ generator = Generator(model, Literal["foo", "bar"])
 
 The model classes found in the `outlines.models` module are maintained but there are a few important changes to be aware of.
 
-The functions used to created a model have been replaced by equivalent functions named with a `from_` prefix. The function `outlines.models.transformers` has been replaced by `outlines.from_transformers` for instance. On top of this change of name, the arguments have been modified. You should refer to the [models documentation](TODO: link to the documentation on models) for more details, but the overall idea is that you now need to provide a model/client instance from the inference library the Outlines model is wrapping.
+The functions used to created a model have been replaced by equivalent functions named with a `from_` prefix. The function `outlines.models.transformers` has been replaced by `outlines.from_transformers` for instance. On top of this change of name, the arguments have been modified. You should refer to the [models documentation](../features/models/index.md) for more details, but the overall idea is that you now need to provide a model/client instance from the inference library the Outlines model is wrapping.
 
 For instance, instead of:
 
@@ -139,7 +139,7 @@ response = generator("foo", num_beams=2)
 
 ### Functions
 
-The `outlines.function` module has been removed. It is replaced by the `outlines.applications` module. An `Application` serves a similar purpose as a `Function`: it encapsulates a prompt template and an output type. A difference is that can `Application` is not instantiated with a model name. Instead, you should provide a model instance along with the prompt when calling it.
+The `outlines.function` module has been removed. It is replaced by the `outlines.applications` module. An [`Application`](../features/utility/application.md) serves a similar purpose as a `Function`: it encapsulates a prompt template and an output type. A difference is that can `Application` is not instantiated with a model name. Instead, you should provide a model instance along with the prompt when calling it.
 
 For instance, instead of:
 
@@ -215,6 +215,8 @@ result = Foo.model_validate_json(result) # parse the text into the Pydantic mode
 print(result.bar)
 ```
 
+The [Output Types](../features/core/output_types.md) section of the features documentation includes extensive details on available output types.
+
 ### Inference arguments
 
 In the previous version of Outlines, some of the inference arguments were standardized across the models and were provided as positional arguments to the generator or through the sampling params dictionary. Additionally, various default values were added by outlines to the inference library models. This is no longer the case. You should refer to the documentation of the inference library you use to find the right arguments for your use case and pass them as keyword arguments to the outlines generator when calling it.
diff --git a/docs_new/user_guide/selecting_an_inference_backend.md b/docs_new/guide/selecting_an_inference_backend.md
similarity index 100%
rename from docs_new/user_guide/selecting_an_inference_backend.md
rename to docs_new/guide/selecting_an_inference_backend.md
diff --git a/docs_new/installation.md b/docs_new/installation.md
deleted file mode 100644
index ad5443512..000000000
--- a/docs_new/installation.md
+++ /dev/null
@@ -1,86 +0,0 @@
----
-title: Installation
----
-
-# Installation
-
-## Dependency Management
-
-We recommend using modern Python packaging tools such as `uv` for managing python dependencies.
-
-### uv (Recommended)
-
-```bash
-# Install uv
-curl -LsSf https://astral.sh/uv/install.sh | sh
-
-# Create a virtual environment and install Outlines
-uv venv
-source .venv/bin/activate
-uv pip install outlines
-```
-
-or with basic pip:
-
-```bash
-pip install outlines
-```
-
-Outlines supports the following model types:
-
-- Transformers
-- OpenAI
-- llama.cpp
-- vLLM
-- dottxt
-- Anthropic
-- Ollama
-- mlx-lm
-- Google
-
-To use these model providers, you must install them manually, i.e.
-
-```sh
-pip install openai
-pip install transformers datasets accelerate torch
-pip install llama-cpp-python
-pip install transformers torch
-pip install vllm
-pip install dottxt
-pip install anthropic
-pip install ollama
-pip install mlx-lm
-pip install google-generativeai
-```
-
-If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
-
-!!! warning "Hardware Requirements"
-
-    If you are using an offline inference tool (not a remote server), your model may require specific hardware. Please check the documentation for these libraries.
-
-    Some libraries like `vllm` and `llama-cpp-python` require specific hardware, such as a compatible GPU. `mlx-lm` is designed for Apple Silicon, so may not be appropriate for your use case if you are on a different platform.
-
-## Optional Dependencies
-
-Outlines provides multiple optional dependency sets to support different backends and use cases. You can install them as needed using:
-
-- `pip install "outlines[vllm]"` for [vLLM](https://github.com/vllm-project/vllm), optimized for high-throughput inference.
-- `pip install "outlines[transformers]"` for [Hugging Face Transformers](https://huggingface.co/docs/transformers/index).
-- `pip install "outlines[mlx]"` for [MLX-LM](https://github.com/ml-explore/mlx-lm), optimized for Apple silicon.
-- `pip install "outlines[openai]"` to use OpenAI’s API.
-- `pip install "outlines[llamacpp]"` for [llama.cpp](https://github.com/ggerganov/llama.cpp), a lightweight LLM inference engine.
-
-## Bleeding Edge
-
-You can install the latest version of Outlines from the repository's `main` branch:
-
-```sh
-pip install git+https://github.com/dottxt-ai/outlines.git@main
-```
-
-This can be useful, for instance, when a fix has been merged but not yet released.
-
-## Installing for Development
-
-See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs_new/user_guide/index.md b/docs_new/user_guide/index.md
deleted file mode 100644
index bd7da2bbf..000000000
--- a/docs_new/user_guide/index.md
+++ /dev/null
@@ -1 +0,0 @@
-This page should provide links and an overview of the user guide section. The user guide should function as a complete reference for how to do and use structured generation.
diff --git a/docs_new/user_guide/prompting/applications.md b/docs_new/user_guide/prompting/applications.md
deleted file mode 100644
index e69de29bb..000000000
diff --git a/docs_new/user_guide/prompting/templates.md b/docs_new/user_guide/prompting/templates.md
deleted file mode 100644
index e69de29bb..000000000
diff --git a/docs_new/user_guide/regex_dsl.md b/docs_new/user_guide/regex_dsl.md
deleted file mode 100644
index f42b6dd34..000000000
--- a/docs_new/user_guide/regex_dsl.md
+++ /dev/null
@@ -1,344 +0,0 @@
-# DSL to express constraints
-
-This library provides a Domain-Specific Language (DSL) to construct regular expressions in a more intuitive and modular way. It allows you to create complex regexes using simple building blocks that represent literal strings, patterns, and various quantifiers. Additionally, these custom regex types can be used directly as types in [Pydantic](https://pydantic-docs.helpmanual.io/) schemas to enforce pattern constraints during text generation.
-
----
-
-## Why Use This DSL?
-
-1. **Modularity & Readability**: Instead of writing cryptic regular expression strings, you compose a regex as a tree of objects.
-2. **Enhanced Debugging**: Each expression can be visualized as an ASCII tree, making it easier to understand and debug complex regexes.
-3. **Pydantic Integration**: Use your DSL-defined regex as types in Pydantic models. The DSL seamlessly converts to JSON Schema with proper pattern constraints.
-4. **Extensibility**: Easily add or modify quantifiers and other regex components by extending the provided classes.
-
----
-
-## Building Blocks
-
-
-Every regex component in this DSL is a **Term**. Here are two primary types:
-
-- **`String`**: Represents a literal string. It escapes the characters that have a special meaning in regular expressions.
-- **`Regex`**: Represents an existing regex pattern string.
-
-```python
-from outlines.types import String, Regex
-
-# A literal string "hello"
-literal = String("hello")   # Internally represents "hello"
-
-# A regex pattern to match one or more digits
-digit = Regex(r"[0-9]+")     # Internally represents the pattern [0-9]+
-
-# Converting to standard regex strings:
-from outlines.types.dsl import to_regex
-
-print(to_regex(literal))  # Output: hello
-print(to_regex(digit))    # Output: [0-9]+
-```
-
----
-
-## Early Introduction to Quantifiers & Combining Terms
-
-The DSL supports common regex quantifiers as methods on every `Term`. These methods allow you to specify how many times a pattern should be matched. They include:
-
-- **`exactly(count)`**: Matches the term exactly `count` times.
-- **`optional()`**: Matches the term zero or one time.
-- **`one_or_more()`**: Matches the term one or more times (Kleene Plus).
-- **`zero_or_more()`**: Matches the term zero or more times (Kleene Star).
-- **`between(min_count, max_count)`**: Matches the term between `min_count` and `max_count` times (inclusive).
-- **`at_least(count)`**: Matches the term at least `count` times.
-- **`at_most(count)`**: Matches the term up to `count` times.
-
-These quantifiers can also be used as functions that take the `Term` as an argument. If the term is a plain string, it will be automatically converted to a `String` object. Thus `String("foo").optional()` is equivalent to `optional("foo")`.
-
-Let's see these quantifiers side by side with examples.
-
-### Quantifiers in Action
-
-#### `exactly(count)`
-
-This method restricts the term to appear exactly `count` times.
-
-```python
-# Example: exactly 5 digits
-five_digits = Regex(r"\d").exactly(5)
-print(to_regex(five_digits))  # Output: (\d){5}
-```
-
-You can also use the `exactly` function:
-
-```python
-from outlines.types import exactly
-
-# Example: exactly 5 digits
-five_digits = exactly(Regex(r"\d"), 5)
-print(to_regex(five_digits))  # Output: (\d){5}
-```
-
-#### `optional()`
-
-This method makes a term optional, meaning it may occur zero or one time.
-
-```python
-# Example: an optional "s" at the end of a word
-maybe_s = String("s").optional()
-print(to_regex(maybe_s))  # Output: (s)?
-```
-
-You can also use the `optional` function:
-
-```python
-from outlines.types import optional
-
-# Example: an optional "s" at the end of a word
-maybe_s = optional("s")
-print(to_regex(maybe_s))  # Output: (s)?
-```
-
-#### `one_or_more()`
-
-This method indicates that the term must appear at least once.
-
-```python
-# Example: one or more alphabetic characters
-letters = Regex(r"[A-Za-z]").one_or_more()
-print(to_regex(letters))  # Output: ([A-Za-z])+
-```
-
-You can also use the `one_or_more` function:
-
-```python
-from outlines.types import one_or_more
-
-# Example: one or more alphabetic characters
-letters = one_or_more(Regex(r"[A-Za-z]"))
-print(to_regex(letters))  # Output: ([A-Za-z])+
-
-```
-
-#### `zero_or_more()`
-
-This method indicates that the term can occur zero or more times.
-
-```python
-# Example: zero or more spaces
-spaces = String(" ").zero_or_more()
-print(to_regex(spaces))  # Output: ( )*
-```
-
-You can also use the `zero_or_more` function:
-
-```python
-from outlines.types import zero_or_more
-
-# Example: zero or more spaces
-spaces = zero_or_more(" ")
-print(to_regex(spaces))  # Output: ( )*
-```
-
-#### `between(min_count, max_count)`
-
-This method indicates that the term can appear any number of times between `min_count` and `max_count` (inclusive).
-
-```python
-# Example: Between 2 and 4 word characters
-word_chars = Regex(r"\w").between(2, 4)
-print(to_regex(word_chars))  # Output: (\w){2,4}
-```
-
-You can also use the `between` function:
-
-```python
-from outlines.types import between
-
-# Example: Between 2 and 4 word characters
-word_chars = between(Regex(r"\w"), 2, 4)
-print(to_regex(word_chars))  # Output: (\w){2,4}
-```
-
-#### `at_least(count)`
-
-This method indicates that the term must appear at least `count` times.
-
-```python
-# Example: At least 3 digits
-at_least_three = Regex(r"\d").at_least(3)
-print(to_regex(at_least_three))  # Output: (\d){3,}
-```
-
-You can also use the `at_least` function:
-
-```python
-from outlines.types import at_least
-
-# Example: At least 3 digits
-at_least_three = at_least(Regex(r"\d"), 3)
-print(to_regex(at_least_three))  # Output: (\d){3,}
-```
-
-#### `at_most(count)`
-
-This method indicates that the term can appear at most `count` times.
-
-```python
-# Example: At most 3 digits
-up_to_three = Regex(r"\d").at_most(3)
-print(to_regex(up_to_three))  # Output: (\d){0,3}
-```
-
-You can also use the `at_most` function:
-
-```python
-from outlines.types import at_most
-
-# Example: At most 3 digits
-up_to_three = at_most(Regex(r"\d"), 3)
-print(to_regex(up_to_three))  # Output: (\d){0,3}
-```
-
----
-
-## Combining Terms
-
-The DSL allows you to combine basic terms into more complex patterns using concatenation and alternation.
-
-### Concatenation (`+`)
-
-The `+` operator (and its reflected variant) concatenates terms, meaning that the terms are matched in sequence.
-
-```python
-# Example: Match "hello world"
-pattern = String("hello") + " " + Regex(r"\w+")
-print(to_regex(pattern))  # Output: hello\ (\w+)
-```
-
-### Alternation (`either()`)
-
-The `either()` function creates alternatives, allowing a match for one of several patterns. You can provide as many terms as you want.
-
-```python
-# Example: Match either "cat" or "dog" or "mouse"
-animal = either(String("cat"), "dog", "mouse")
-print(to_regex(animal))  # Output: (cat|dog|mouse)
-```
-
-*Note:* When using `either()` with plain strings (such as `"dog"`), the DSL automatically wraps them in a `String` object that escapes the characters that have a special meaning in regular expressions, just like with quantifier functions.
-
----
-
-## Custom types
-
-The DSL comes "batteries included" with types that represent common text constructs:
-
-- `integer` represents an integer number as recognized by `int`
-- `boolean` represents a boolean, "True" or "False" as recognized by `bool`
-- `number` represents a floating-point number recognize by Python's `float`
-- `date` represents a date as understood by `datetime.date`
-- `time` represents a time as undestood by `datetime.time`
-- `datetime` represents a time as understoof by `datetime.datetime`
-- `digit` represents a single digit
-- `char` represents a single character
-- `newline` represents a new line character
-- `whitespace` represents a white space
-- `sentence` represents a sentence
-- `paragraph` reprensents a pagraph (one or more sentences separated by one or more ilne breaks)
-
-
-For instance you can describe the answers in the GSM8K dataset using the following pattern:
-
-```python
-from outlines.types import sentence, digit
-
-answer = "A: " + sentence.between(2,4) + " So the answer is: " + digit.between(1,4)
-```
-
----
-
-## Practical Examples
-
-### Example 1: Matching a Custom ID Format
-
-Suppose you want to create a regex that matches an ID format like "ID-12345", where:
-- The literal "ID-" must be at the start.
-- Followed by exactly 5 digits.
-
-```python
-id_pattern = "ID-" + Regex(r"\d").exactly(5)
-print(to_regex(id_pattern))  # Output: ID-(\d){5}
-```
-
-### Example 2: Email Validation with Pydantic
-
-You can define a regex for email validation and use it as a type in a Pydantic model.
-
-```python
-from pydantic import BaseModel, ValidationError
-
-# Define an email regex term (this is a simplified version)
-email_regex = Regex(r"[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+")
-
-class User(BaseModel):
-    name: str
-    email: email_regex  # Use our DSL regex as a field type
-
-# Valid input
-user = User(name="Alice", email="alice@example.com")
-print(user)
-
-# Invalid input (raises a ValidationError)
-try:
-    User(name="Bob", email="not-an-email")
-except ValidationError as e:
-    print(e)
-```
-
-When used in a Pydantic model, the email field is automatically validated against the regex pattern and its JSON Schema includes the `pattern` constraint.
-
-### Example 3: Building a Complex Pattern
-
-Consider a pattern to match a simple date format: `YYYY-MM-DD`.
-
-```python
-year = Regex(r"\d").exactly(4)         # Four digits for the year
-month = Regex(r"\d").exactly(2)        # Two digits for the month
-day = Regex(r"\d").exactly(2)          # Two digits for the day
-
-# Combine with literal hyphens
-date_pattern = year + "-" + month + "-" + day
-print(to_regex(date_pattern))
-# Output: (\d){4}\-(\d){2}\-(\d){2}
-```
-
----
-
-## Visualizing Your Pattern
-
-One of the unique features of this DSL is that each term can print its underlying structure as an ASCII tree. This visualization can be particularly helpful when dealing with complex expressions.
-
-```python
-# A composite pattern using concatenation and quantifiers
-pattern = "a" + String("b").one_or_more() + "c"
-print(pattern)
-```
-
-*Expected Output:*
-
-```
-└── Sequence
-    ├── String('a')
-    ├── KleenePlus(+)
-    │   └── String('b')
-    └── String('c')
-```
-
-This tree representation makes it easy to see the hierarchy and order of operations in your regular expression.
-
----
-
-## Final Words
-
-This DSL is designed to simplify the creation and management of regular expressions—whether you're validating inputs in a web API, constraining the output of an LLM, or just experimenting with regex patterns. With intuitive methods for common quantifiers and operators, clear visual feedback, and built-in integration with Pydantic, you can build robust and maintainable regex-based validations with ease.
-
-Feel free to explore the library further and adapt the examples to your use cases. Happy regexing!
diff --git a/docs_new/user_guide/structured_generation/cfg.md b/docs_new/user_guide/structured_generation/cfg.md
deleted file mode 100644
index e69de29bb..000000000
diff --git a/docs_new/user_guide/structured_generation/index.md b/docs_new/user_guide/structured_generation/index.md
deleted file mode 100644
index 541a7e1f4..000000000
--- a/docs_new/user_guide/structured_generation/index.md
+++ /dev/null
@@ -1,13 +0,0 @@
----
-title: Structured Generation
----
-
-# Structured Generation
-
-Outlines allows you to generate structured outputs from LLMs. There are several ways to specify the structure of the output you want to generate.
-
-### JSON schema
-
-### Pydantic models
-
-### Context-free grammars
diff --git a/docs_new/user_guide/structured_generation/json.md b/docs_new/user_guide/structured_generation/json.md
deleted file mode 100644
index e69de29bb..000000000
diff --git a/docs_new/user_guide/structured_generation/multiple_choice.md b/docs_new/user_guide/structured_generation/multiple_choice.md
deleted file mode 100644
index e69de29bb..000000000
diff --git a/docs_new/user_guide/structured_generation/regex.md b/docs_new/user_guide/structured_generation/regex.md
deleted file mode 100644
index e69de29bb..000000000
diff --git a/docs_new/user_guide/unstructured.md b/docs_new/user_guide/unstructured.md
deleted file mode 100644
index 9046cf4d1..000000000
--- a/docs_new/user_guide/unstructured.md
+++ /dev/null
@@ -1 +0,0 @@
-Brief overview of unstructured text generation.
diff --git a/mkdocs.yml b/mkdocs.yml
index e5d3d5776..13ba556e1 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -98,7 +98,7 @@ plugins:
     - redirects:
           redirect_maps:
               "welcome.md": "index.md"
-              "reference/chat_templating.md": "user_guide/chat_templating.md"
+              "reference/chat_templating.md": "guide/chat_templating.md"
               "blog.md": "https://blog.dottxt.co"
               "cookbook/index.md": "examples/index.md"
               "cookbook/classification.md": "examples/classification.md"
@@ -155,32 +155,18 @@ plugins:
 nav:
     - Home:
           - index.md
-          - User Guide: user_guide/index.md
+          - User Guide: guide/index.md
           - Features: features/index.md
           - API Reference: api_reference/index.md
           - Examples: examples/index.md
 
     - Guide:
-          - Introduction:
-                - Get started: getting_started/index.md
-                - Installation: installation.md
-                - Licence: licence.md
-          - User Guide:
-                - The Basics: user_guide/index.md
-                - Selecting an Inference Engine: user_guide/selecting_an_inference_backend.md
-                - Unstructured Text Generation: user_guide/unstructured.md
-                - Structured Generation: user_guide/structured_generation/index.md
-                - Chat Templating for Instruct Models: user_guide/chat_templating.md
-                - Generation Formats:
-                      - JSON: user_guide/structured_generation/json.md
-                      - Regular expressions: user_guide/structured_generation/regex.md
-                      - Context-free grammars: user_guide/structured_generation/cfg.md
-                      - Multiple choice: user_guide/structured_generation/multiple_choice.md
-                - Prompting:
-                      - Applications: user_guide/prompting/applications.md
-                      - Templates: user_guide/prompting/templates.md
-                - Easy Regular Expressions: user_guide/regex_dsl.md
-                - Migrating to v1: user_guide/migration.md
+          - Getting Started: guide/getting_started.md
+          - Installation: guide/installation.md
+          - Migrating to v1: guide/migration.md
+          - Licence: licence.md
+          - Utility:
+                - Chat Templating for Instruct Models: guide/chat_templating.md
 
     - Features:
           - Overview: features/index.md

From b2aeda54db912ccafe8986c13ab77624014c6adc Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 2 Jun 2025 18:58:11 +0200
Subject: [PATCH 344/505] Modify the documentation home page

---
 docs/{api => api_reference}/index.md          |   0
 docs/community/index.md                       |   3 -
 docs/community/versioning.md                  |   4 +
 {docs_new => docs}/core_concepts.md           |   0
 {docs_new => docs}/examples/atomic_caption.md |   0
 .../examples/chain_of_density.md              |   0
 .../examples/chain_of_thought.md              |   0
 {docs_new => docs}/examples/classification.md |   0
 .../examples/dating_profiles.md               |   0
 .../examples/deploy-using-bentoml.md          |   0
 .../examples/deploy-using-cerebrium.md        |   0
 .../examples/deploy-using-modal.md            |   0
 .../examples/earnings-reports.md              |   0
 .../examples/extract_event_details.md         |   0
 {docs_new => docs}/examples/extraction.md     |   0
 .../images/chain_of_density.png               | Bin
 .../images/coding_structure_diagram.png       | Bin
 .../images/knowledge-graph-extraction.png     | Bin
 .../images/nvidia-income.png                  | Bin
 docs/{cookbook => examples}/images/simtom.png | Bin
 .../images/trader-joes-receipt.jpg            | Bin
 {docs_new => docs}/examples/index.md          |   0
 .../examples/knowledge_graph_extraction.md    |   0
 .../examples/models_playing_chess.md          |   0
 .../examples/qa-with-citations.md             |   0
 {docs_new => docs}/examples/react_agent.md    |   0
 {docs_new => docs}/examples/read-pdfs.md      |   0
 .../examples/receipt-digitization.md          |   0
 {docs_new => docs}/examples/simtom.md         |   0
 .../structured_generation_workflow.md         |   0
 .../features/advanced/logits_processors.md    |   0
 {docs_new => docs}/features/core/generator.md |   0
 .../features/core/output_types.md             |   0
 {docs_new => docs}/features/index.md          |   0
 .../features/models/anthropic.md              |   0
 {docs_new => docs}/features/models/dottxt.md  |   0
 {docs_new => docs}/features/models/gemini.md  |   0
 {docs_new => docs}/features/models/index.md   |   0
 .../features/models/llamacpp.md               |   0
 {docs_new => docs}/features/models/mlxlm.md   |   0
 {docs_new => docs}/features/models/ollama.md  |   0
 {docs_new => docs}/features/models/openai.md  |   0
 {docs_new => docs}/features/models/sglang.md  |   0
 {docs_new => docs}/features/models/tgi.md     |   0
 .../features/models/transformers.md           |   0
 .../models/transformers_multimodal.md         |   0
 {docs_new => docs}/features/models/vllm.md    |   0
 .../features/models/vllm_offline.md           |   0
 .../features/utility/application.md           |   0
 .../features/utility/regex_dsl.md             |   0
 .../features/utility/template.md              |   0
 docs/{reference => guide}/chat_templating.md  |   0
 {docs_new => docs}/guide/core_concepts.md     |   0
 {docs_new => docs}/guide/getting_started.md   |   0
 {docs_new => docs}/guide/installation.md      |   0
 {docs_new => docs}/guide/migration.md         |   0
 .../guide/selecting_an_inference_backend.md   |   0
 docs/index.md                                 | 151 +++++++++++++++-
 docs/stylesheets/extra.css                    | 161 ++----------------
 docs_new/assets/images/logo-square.png        | Bin 3323 -> 0 bytes
 docs_new/assets/images/logo-square.svg        | 133 ---------------
 docs_new/index.md                             | 147 ----------------
 docs_new/stylesheets/extra.css                |  22 ---
 {docs => docs_old}/api/applications.md        |   0
 {docs => docs_old}/api/core/application.md    |   0
 {docs => docs_old}/api/guide.md               |   0
 .../api_reference => docs_old/api}/index.md   |   0
 {docs => docs_old}/api/json_schema.md         |   0
 {docs => docs_old}/api/models.md              |   0
 {docs => docs_old}/api/parsing.md             |   0
 {docs => docs_old}/api/regex.md               |   0
 {docs => docs_old}/api/templates.md           |   0
 .../assets/images/dottxt-dark.svg             |   0
 .../assets/images/dottxt-light.svg            |   0
 .../assets/images/dottxt.png                  | Bin
 .../assets/images/install.png                 | Bin
 .../images/logits_processing_diagram.svg      |   0
 .../assets/images/logo-dark-mode.png          | Bin
 .../assets/images/logo-dark-mode.svg          |   0
 .../assets/images/logo-light-mode.png         | Bin
 .../assets/images/logo-light-mode.svg         |   0
 .../assets/images/logo-simple.png             | Bin
 docs_old/assets/images/logo-square.png        | Bin 0 -> 2233 bytes
 docs_old/assets/images/logo-square.svg        | 124 ++++++++++++++
 {docs_new => docs_old}/assets/images/logo.png | Bin
 {docs_new => docs_old}/assets/images/logo.svg |   0
 .../assets/images/normal_computing.jpg        | Bin
 .../assets/images/readme-dark.png             | Bin
 .../assets/images/readme-light.png            | Bin
 .../assets/images/use_philosophy.png          | Bin
 {docs => docs_old}/blog/.authors.yml          |   0
 {docs => docs_old}/blog/assets/4000_stars.png | Bin
 {docs_new => docs_old}/blog/index.md          |   0
 {docs => docs_old}/blog/posts/roadmap-2024.md |   0
 {docs => docs_old}/community/belonging.png    | Bin
 .../community/contribute.md                   |   0
 {docs_new => docs_old}/community/examples.md  |   0
 {docs_new => docs_old}/community/feedback.md  |   0
 {docs_new => docs_old}/community/index.md     |   3 +
 .../community/versioning.md                   |   4 -
 .../content/docs/guides/migration_guide.md    |   0
 {docs => docs_old}/cookbook/atomic_caption.md |   0
 .../cookbook/chain_of_density.md              |   0
 .../cookbook/chain_of_thought.md              |   0
 {docs => docs_old}/cookbook/classification.md |   0
 .../cookbook/dating_profiles.md               |   0
 .../cookbook/deploy-using-bentoml.md          |   0
 .../cookbook/deploy-using-cerebrium.md        |   0
 .../cookbook/deploy-using-modal.md            |   0
 .../cookbook/earnings-reports.md              |   0
 .../cookbook/extract_event_details.md         |   0
 .../cookbook/extract_event_details.py         |   0
 {docs => docs_old}/cookbook/extraction.md     |   0
 .../cookbook}/images/chain_of_density.png     | Bin
 .../images/coding_structure_diagram.png       | Bin
 .../images/knowledge-graph-extraction.png     | Bin
 .../cookbook}/images/nvidia-income.png        | Bin
 .../cookbook}/images/simtom.png               | Bin
 .../cookbook}/images/trader-joes-receipt.jpg  | Bin
 {docs => docs_old}/cookbook/index.md          |   0
 .../cookbook/knowledge_graph_extraction.md    |   0
 .../cookbook/models_playing_chess.md          |   0
 .../prompt_templates/chain_of_density.txt     |   0
 .../prompt_templates/classification.txt       |   0
 .../cookbook/prompt_templates/react_agent.txt |   0
 .../simtom_prospective_taking.txt             |   0
 .../prompt_templates/simtom_simulation.txt    |   0
 .../cookbook/qa-with-citations.md             |   0
 {docs => docs_old}/cookbook/react_agent.md    |   0
 {docs => docs_old}/cookbook/read-pdfs.md      |   0
 .../cookbook/receipt-digitization.md          |   0
 {docs => docs_old}/cookbook/simtom.md         |   0
 .../structured_generation_workflow.md         |   0
 docs_old/index.md                             |   8 +
 {docs => docs_old}/installation.md            |   0
 {docs_new => docs_old}/licence.md             |   0
 {docs_new => docs_old}/logos/amazon.png       | Bin
 {docs_new => docs_old}/logos/apple.png        | Bin
 {docs_new => docs_old}/logos/best_buy.png     | Bin
 {docs_new => docs_old}/logos/canoe.png        | Bin
 {docs_new => docs_old}/logos/cisco.png        | Bin
 {docs_new => docs_old}/logos/cohere.png       | Bin
 .../logos/dassault_systems.png                | Bin
 {docs_new => docs_old}/logos/databricks.png   | Bin
 {docs_new => docs_old}/logos/datadog.png      | Bin
 {docs_new => docs_old}/logos/dbt_labs.png     | Bin
 {docs_new => docs_old}/logos/gladia.jpg       | Bin
 {docs_new => docs_old}/logos/harvard.png      | Bin
 {docs_new => docs_old}/logos/hf.png           | Bin
 .../logos/johns_hopkins.png                   | Bin
 {docs_new => docs_old}/logos/meta.png         | Bin
 {docs_new => docs_old}/logos/mit.png          | Bin
 {docs_new => docs_old}/logos/mount_sinai.png  | Bin
 {docs_new => docs_old}/logos/nvidia.png       | Bin
 {docs_new => docs_old}/logos/nyu.png          | Bin
 {docs_new => docs_old}/logos/safran.png       | Bin
 {docs_new => docs_old}/logos/salesforce.png   | Bin
 {docs_new => docs_old}/logos/shopify.png      | Bin
 {docs_new => docs_old}/logos/smithsonian.png  | Bin
 {docs_new => docs_old}/logos/tinder.png       | Bin
 {docs_new => docs_old}/logos/upenn.png        | Bin
 {docs_new => docs_old}/logos/vllm.png         | Bin
 {docs_new => docs_old}/overrides/home.html    |   3 +-
 {docs_new => docs_old}/overrides/main.html    |   0
 {docs => docs_old}/quickstart.md              |   0
 {docs => docs_old}/reference/applications.md  |   0
 .../reference}/chat_templating.md             |   0
 .../reference/generation/creating_grammars.md |   0
 .../reference/generation/custom_fsm_ops.md    |   0
 .../reference/generation/generation.md        |   0
 .../structured_generation_explanation.md      |   0
 .../reference/generation/types.md             |   0
 {docs => docs_old}/reference/index.md         |   0
 .../reference/models/anthropic.md             |   0
 {docs => docs_old}/reference/models/dottxt.md |   0
 {docs => docs_old}/reference/models/gemini.md |   0
 .../reference/models/llamacpp.md              |   0
 {docs => docs_old}/reference/models/mlxlm.md  |   0
 {docs => docs_old}/reference/models/models.md |   0
 {docs => docs_old}/reference/models/ollama.md |   0
 {docs => docs_old}/reference/models/openai.md |   0
 {docs => docs_old}/reference/models/sglang.md |   0
 {docs => docs_old}/reference/models/tgi.md    |   0
 .../reference/models/transformers.md          |   0
 .../models/transformers_multimodal.md         |   0
 {docs => docs_old}/reference/models/vllm.md   |   0
 .../reference/models/vllm_offline.md          |   0
 {docs => docs_old}/reference/prompting.md     |   0
 {docs => docs_old}/reference/regex_dsl.md     |   0
 .../reference/serve/lmstudio.md               |   0
 {docs => docs_old}/reference/serve/vllm.md    |   0
 docs_old/stylesheets/extra.css                | 149 ++++++++++++++++
 {docs => docs_old}/welcome.md                 |   0
 mkdocs.yml                                    |   9 +-
 194 files changed, 453 insertions(+), 468 deletions(-)
 rename docs/{api => api_reference}/index.md (100%)
 rename {docs_new => docs}/core_concepts.md (100%)
 rename {docs_new => docs}/examples/atomic_caption.md (100%)
 rename {docs_new => docs}/examples/chain_of_density.md (100%)
 rename {docs_new => docs}/examples/chain_of_thought.md (100%)
 rename {docs_new => docs}/examples/classification.md (100%)
 rename {docs_new => docs}/examples/dating_profiles.md (100%)
 rename {docs_new => docs}/examples/deploy-using-bentoml.md (100%)
 rename {docs_new => docs}/examples/deploy-using-cerebrium.md (100%)
 rename {docs_new => docs}/examples/deploy-using-modal.md (100%)
 rename {docs_new => docs}/examples/earnings-reports.md (100%)
 rename {docs_new => docs}/examples/extract_event_details.md (100%)
 rename {docs_new => docs}/examples/extraction.md (100%)
 rename docs/{cookbook => examples}/images/chain_of_density.png (100%)
 rename docs/{cookbook => examples}/images/coding_structure_diagram.png (100%)
 rename docs/{cookbook => examples}/images/knowledge-graph-extraction.png (100%)
 rename docs/{cookbook => examples}/images/nvidia-income.png (100%)
 rename docs/{cookbook => examples}/images/simtom.png (100%)
 rename docs/{cookbook => examples}/images/trader-joes-receipt.jpg (100%)
 rename {docs_new => docs}/examples/index.md (100%)
 rename {docs_new => docs}/examples/knowledge_graph_extraction.md (100%)
 rename {docs_new => docs}/examples/models_playing_chess.md (100%)
 rename {docs_new => docs}/examples/qa-with-citations.md (100%)
 rename {docs_new => docs}/examples/react_agent.md (100%)
 rename {docs_new => docs}/examples/read-pdfs.md (100%)
 rename {docs_new => docs}/examples/receipt-digitization.md (100%)
 rename {docs_new => docs}/examples/simtom.md (100%)
 rename {docs_new => docs}/examples/structured_generation_workflow.md (100%)
 rename {docs_new => docs}/features/advanced/logits_processors.md (100%)
 rename {docs_new => docs}/features/core/generator.md (100%)
 rename {docs_new => docs}/features/core/output_types.md (100%)
 rename {docs_new => docs}/features/index.md (100%)
 rename {docs_new => docs}/features/models/anthropic.md (100%)
 rename {docs_new => docs}/features/models/dottxt.md (100%)
 rename {docs_new => docs}/features/models/gemini.md (100%)
 rename {docs_new => docs}/features/models/index.md (100%)
 rename {docs_new => docs}/features/models/llamacpp.md (100%)
 rename {docs_new => docs}/features/models/mlxlm.md (100%)
 rename {docs_new => docs}/features/models/ollama.md (100%)
 rename {docs_new => docs}/features/models/openai.md (100%)
 rename {docs_new => docs}/features/models/sglang.md (100%)
 rename {docs_new => docs}/features/models/tgi.md (100%)
 rename {docs_new => docs}/features/models/transformers.md (100%)
 rename {docs_new => docs}/features/models/transformers_multimodal.md (100%)
 rename {docs_new => docs}/features/models/vllm.md (100%)
 rename {docs_new => docs}/features/models/vllm_offline.md (100%)
 rename {docs_new => docs}/features/utility/application.md (100%)
 rename {docs_new => docs}/features/utility/regex_dsl.md (100%)
 rename {docs_new => docs}/features/utility/template.md (100%)
 rename docs/{reference => guide}/chat_templating.md (100%)
 rename {docs_new => docs}/guide/core_concepts.md (100%)
 rename {docs_new => docs}/guide/getting_started.md (100%)
 rename {docs_new => docs}/guide/installation.md (100%)
 rename {docs_new => docs}/guide/migration.md (100%)
 rename {docs_new => docs}/guide/selecting_an_inference_backend.md (100%)
 delete mode 100644 docs_new/assets/images/logo-square.png
 delete mode 100644 docs_new/assets/images/logo-square.svg
 delete mode 100644 docs_new/index.md
 delete mode 100644 docs_new/stylesheets/extra.css
 rename {docs => docs_old}/api/applications.md (100%)
 rename {docs => docs_old}/api/core/application.md (100%)
 rename {docs => docs_old}/api/guide.md (100%)
 rename {docs_new/api_reference => docs_old/api}/index.md (100%)
 rename {docs => docs_old}/api/json_schema.md (100%)
 rename {docs => docs_old}/api/models.md (100%)
 rename {docs => docs_old}/api/parsing.md (100%)
 rename {docs => docs_old}/api/regex.md (100%)
 rename {docs => docs_old}/api/templates.md (100%)
 rename {docs_new => docs_old}/assets/images/dottxt-dark.svg (100%)
 rename {docs_new => docs_old}/assets/images/dottxt-light.svg (100%)
 rename {docs_new => docs_old}/assets/images/dottxt.png (100%)
 rename {docs_new => docs_old}/assets/images/install.png (100%)
 rename {docs_new => docs_old}/assets/images/logits_processing_diagram.svg (100%)
 rename {docs_new => docs_old}/assets/images/logo-dark-mode.png (100%)
 rename {docs_new => docs_old}/assets/images/logo-dark-mode.svg (100%)
 rename {docs_new => docs_old}/assets/images/logo-light-mode.png (100%)
 rename {docs_new => docs_old}/assets/images/logo-light-mode.svg (100%)
 rename {docs_new => docs_old}/assets/images/logo-simple.png (100%)
 create mode 100644 docs_old/assets/images/logo-square.png
 create mode 100644 docs_old/assets/images/logo-square.svg
 rename {docs_new => docs_old}/assets/images/logo.png (100%)
 rename {docs_new => docs_old}/assets/images/logo.svg (100%)
 rename {docs_new => docs_old}/assets/images/normal_computing.jpg (100%)
 rename {docs_new => docs_old}/assets/images/readme-dark.png (100%)
 rename {docs_new => docs_old}/assets/images/readme-light.png (100%)
 rename {docs_new => docs_old}/assets/images/use_philosophy.png (100%)
 rename {docs => docs_old}/blog/.authors.yml (100%)
 rename {docs => docs_old}/blog/assets/4000_stars.png (100%)
 rename {docs_new => docs_old}/blog/index.md (100%)
 rename {docs => docs_old}/blog/posts/roadmap-2024.md (100%)
 rename {docs => docs_old}/community/belonging.png (100%)
 rename {docs_new => docs_old}/community/contribute.md (100%)
 rename {docs_new => docs_old}/community/examples.md (100%)
 rename {docs_new => docs_old}/community/feedback.md (100%)
 rename {docs_new => docs_old}/community/index.md (94%)
 rename {docs_new => docs_old}/community/versioning.md (87%)
 rename {docs => docs_old}/content/docs/guides/migration_guide.md (100%)
 rename {docs => docs_old}/cookbook/atomic_caption.md (100%)
 rename {docs => docs_old}/cookbook/chain_of_density.md (100%)
 rename {docs => docs_old}/cookbook/chain_of_thought.md (100%)
 rename {docs => docs_old}/cookbook/classification.md (100%)
 rename {docs => docs_old}/cookbook/dating_profiles.md (100%)
 rename {docs => docs_old}/cookbook/deploy-using-bentoml.md (100%)
 rename {docs => docs_old}/cookbook/deploy-using-cerebrium.md (100%)
 rename {docs => docs_old}/cookbook/deploy-using-modal.md (100%)
 rename {docs => docs_old}/cookbook/earnings-reports.md (100%)
 rename {docs => docs_old}/cookbook/extract_event_details.md (100%)
 rename {docs => docs_old}/cookbook/extract_event_details.py (100%)
 rename {docs => docs_old}/cookbook/extraction.md (100%)
 rename {docs_new/examples => docs_old/cookbook}/images/chain_of_density.png (100%)
 rename {docs_new/examples => docs_old/cookbook}/images/coding_structure_diagram.png (100%)
 rename {docs_new/examples => docs_old/cookbook}/images/knowledge-graph-extraction.png (100%)
 rename {docs_new/examples => docs_old/cookbook}/images/nvidia-income.png (100%)
 rename {docs_new/examples => docs_old/cookbook}/images/simtom.png (100%)
 rename {docs_new/examples => docs_old/cookbook}/images/trader-joes-receipt.jpg (100%)
 rename {docs => docs_old}/cookbook/index.md (100%)
 rename {docs => docs_old}/cookbook/knowledge_graph_extraction.md (100%)
 rename {docs => docs_old}/cookbook/models_playing_chess.md (100%)
 rename {docs => docs_old}/cookbook/prompt_templates/chain_of_density.txt (100%)
 rename {docs => docs_old}/cookbook/prompt_templates/classification.txt (100%)
 rename {docs => docs_old}/cookbook/prompt_templates/react_agent.txt (100%)
 rename {docs => docs_old}/cookbook/prompt_templates/simtom_prospective_taking.txt (100%)
 rename {docs => docs_old}/cookbook/prompt_templates/simtom_simulation.txt (100%)
 rename {docs => docs_old}/cookbook/qa-with-citations.md (100%)
 rename {docs => docs_old}/cookbook/react_agent.md (100%)
 rename {docs => docs_old}/cookbook/read-pdfs.md (100%)
 rename {docs => docs_old}/cookbook/receipt-digitization.md (100%)
 rename {docs => docs_old}/cookbook/simtom.md (100%)
 rename {docs => docs_old}/cookbook/structured_generation_workflow.md (100%)
 create mode 100644 docs_old/index.md
 rename {docs => docs_old}/installation.md (100%)
 rename {docs_new => docs_old}/licence.md (100%)
 rename {docs_new => docs_old}/logos/amazon.png (100%)
 rename {docs_new => docs_old}/logos/apple.png (100%)
 rename {docs_new => docs_old}/logos/best_buy.png (100%)
 rename {docs_new => docs_old}/logos/canoe.png (100%)
 rename {docs_new => docs_old}/logos/cisco.png (100%)
 rename {docs_new => docs_old}/logos/cohere.png (100%)
 rename {docs_new => docs_old}/logos/dassault_systems.png (100%)
 rename {docs_new => docs_old}/logos/databricks.png (100%)
 rename {docs_new => docs_old}/logos/datadog.png (100%)
 rename {docs_new => docs_old}/logos/dbt_labs.png (100%)
 rename {docs_new => docs_old}/logos/gladia.jpg (100%)
 rename {docs_new => docs_old}/logos/harvard.png (100%)
 rename {docs_new => docs_old}/logos/hf.png (100%)
 rename {docs_new => docs_old}/logos/johns_hopkins.png (100%)
 rename {docs_new => docs_old}/logos/meta.png (100%)
 rename {docs_new => docs_old}/logos/mit.png (100%)
 rename {docs_new => docs_old}/logos/mount_sinai.png (100%)
 rename {docs_new => docs_old}/logos/nvidia.png (100%)
 rename {docs_new => docs_old}/logos/nyu.png (100%)
 rename {docs_new => docs_old}/logos/safran.png (100%)
 rename {docs_new => docs_old}/logos/salesforce.png (100%)
 rename {docs_new => docs_old}/logos/shopify.png (100%)
 rename {docs_new => docs_old}/logos/smithsonian.png (100%)
 rename {docs_new => docs_old}/logos/tinder.png (100%)
 rename {docs_new => docs_old}/logos/upenn.png (100%)
 rename {docs_new => docs_old}/logos/vllm.png (100%)
 rename {docs_new => docs_old}/overrides/home.html (94%)
 rename {docs_new => docs_old}/overrides/main.html (100%)
 rename {docs => docs_old}/quickstart.md (100%)
 rename {docs => docs_old}/reference/applications.md (100%)
 rename {docs_new/guide => docs_old/reference}/chat_templating.md (100%)
 rename {docs => docs_old}/reference/generation/creating_grammars.md (100%)
 rename {docs => docs_old}/reference/generation/custom_fsm_ops.md (100%)
 rename {docs => docs_old}/reference/generation/generation.md (100%)
 rename {docs => docs_old}/reference/generation/structured_generation_explanation.md (100%)
 rename {docs => docs_old}/reference/generation/types.md (100%)
 rename {docs => docs_old}/reference/index.md (100%)
 rename {docs => docs_old}/reference/models/anthropic.md (100%)
 rename {docs => docs_old}/reference/models/dottxt.md (100%)
 rename {docs => docs_old}/reference/models/gemini.md (100%)
 rename {docs => docs_old}/reference/models/llamacpp.md (100%)
 rename {docs => docs_old}/reference/models/mlxlm.md (100%)
 rename {docs => docs_old}/reference/models/models.md (100%)
 rename {docs => docs_old}/reference/models/ollama.md (100%)
 rename {docs => docs_old}/reference/models/openai.md (100%)
 rename {docs => docs_old}/reference/models/sglang.md (100%)
 rename {docs => docs_old}/reference/models/tgi.md (100%)
 rename {docs => docs_old}/reference/models/transformers.md (100%)
 rename {docs => docs_old}/reference/models/transformers_multimodal.md (100%)
 rename {docs => docs_old}/reference/models/vllm.md (100%)
 rename {docs => docs_old}/reference/models/vllm_offline.md (100%)
 rename {docs => docs_old}/reference/prompting.md (100%)
 rename {docs => docs_old}/reference/regex_dsl.md (100%)
 rename {docs => docs_old}/reference/serve/lmstudio.md (100%)
 rename {docs => docs_old}/reference/serve/vllm.md (100%)
 create mode 100644 docs_old/stylesheets/extra.css
 rename {docs => docs_old}/welcome.md (100%)

diff --git a/docs/api/index.md b/docs/api_reference/index.md
similarity index 100%
rename from docs/api/index.md
rename to docs/api_reference/index.md
diff --git a/docs/community/index.md b/docs/community/index.md
index 9519f5b56..517bf59c5 100644
--- a/docs/community/index.md
+++ b/docs/community/index.md
@@ -1,12 +1,9 @@
 # Community
 
-![Belonging](belonging.png)
-
 Outlines exists for a community of users who believe software doesn't need to be complicated. Who share the same passion for Large Language Models but don't want to compromise on robustness. Together, we are bringing these powerful models back to the world of software.
 
 ## Connect on Discord
 
 The Outlines community lives on our Discord server. There you can ask questions, share ideas or just chat with people like you. Don't be a stranger and [join us][discord].
 
-
 [discord]: https://discord.gg/UppQmhEpe8
diff --git a/docs/community/versioning.md b/docs/community/versioning.md
index 023b92537..a6e12f218 100644
--- a/docs/community/versioning.md
+++ b/docs/community/versioning.md
@@ -13,6 +13,10 @@ Each part of the version number (`major.minor.patch`) conveys information about
 - **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatability with existing use of core interfaces.
 - **Patch Releases** address bug fixes and incorporate low-risk changes to improve stability and performance.
 
+!!! note "Breaking Changes"
+
+    Outlines v1.0 introduced several breaking changes to the core interface. See [the migration guide](/user_guide/migration) for more details.
+
 ## Releases
 
 Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/dottxt-ai/outlines/releases).
diff --git a/docs_new/core_concepts.md b/docs/core_concepts.md
similarity index 100%
rename from docs_new/core_concepts.md
rename to docs/core_concepts.md
diff --git a/docs_new/examples/atomic_caption.md b/docs/examples/atomic_caption.md
similarity index 100%
rename from docs_new/examples/atomic_caption.md
rename to docs/examples/atomic_caption.md
diff --git a/docs_new/examples/chain_of_density.md b/docs/examples/chain_of_density.md
similarity index 100%
rename from docs_new/examples/chain_of_density.md
rename to docs/examples/chain_of_density.md
diff --git a/docs_new/examples/chain_of_thought.md b/docs/examples/chain_of_thought.md
similarity index 100%
rename from docs_new/examples/chain_of_thought.md
rename to docs/examples/chain_of_thought.md
diff --git a/docs_new/examples/classification.md b/docs/examples/classification.md
similarity index 100%
rename from docs_new/examples/classification.md
rename to docs/examples/classification.md
diff --git a/docs_new/examples/dating_profiles.md b/docs/examples/dating_profiles.md
similarity index 100%
rename from docs_new/examples/dating_profiles.md
rename to docs/examples/dating_profiles.md
diff --git a/docs_new/examples/deploy-using-bentoml.md b/docs/examples/deploy-using-bentoml.md
similarity index 100%
rename from docs_new/examples/deploy-using-bentoml.md
rename to docs/examples/deploy-using-bentoml.md
diff --git a/docs_new/examples/deploy-using-cerebrium.md b/docs/examples/deploy-using-cerebrium.md
similarity index 100%
rename from docs_new/examples/deploy-using-cerebrium.md
rename to docs/examples/deploy-using-cerebrium.md
diff --git a/docs_new/examples/deploy-using-modal.md b/docs/examples/deploy-using-modal.md
similarity index 100%
rename from docs_new/examples/deploy-using-modal.md
rename to docs/examples/deploy-using-modal.md
diff --git a/docs_new/examples/earnings-reports.md b/docs/examples/earnings-reports.md
similarity index 100%
rename from docs_new/examples/earnings-reports.md
rename to docs/examples/earnings-reports.md
diff --git a/docs_new/examples/extract_event_details.md b/docs/examples/extract_event_details.md
similarity index 100%
rename from docs_new/examples/extract_event_details.md
rename to docs/examples/extract_event_details.md
diff --git a/docs_new/examples/extraction.md b/docs/examples/extraction.md
similarity index 100%
rename from docs_new/examples/extraction.md
rename to docs/examples/extraction.md
diff --git a/docs/cookbook/images/chain_of_density.png b/docs/examples/images/chain_of_density.png
similarity index 100%
rename from docs/cookbook/images/chain_of_density.png
rename to docs/examples/images/chain_of_density.png
diff --git a/docs/cookbook/images/coding_structure_diagram.png b/docs/examples/images/coding_structure_diagram.png
similarity index 100%
rename from docs/cookbook/images/coding_structure_diagram.png
rename to docs/examples/images/coding_structure_diagram.png
diff --git a/docs/cookbook/images/knowledge-graph-extraction.png b/docs/examples/images/knowledge-graph-extraction.png
similarity index 100%
rename from docs/cookbook/images/knowledge-graph-extraction.png
rename to docs/examples/images/knowledge-graph-extraction.png
diff --git a/docs/cookbook/images/nvidia-income.png b/docs/examples/images/nvidia-income.png
similarity index 100%
rename from docs/cookbook/images/nvidia-income.png
rename to docs/examples/images/nvidia-income.png
diff --git a/docs/cookbook/images/simtom.png b/docs/examples/images/simtom.png
similarity index 100%
rename from docs/cookbook/images/simtom.png
rename to docs/examples/images/simtom.png
diff --git a/docs/cookbook/images/trader-joes-receipt.jpg b/docs/examples/images/trader-joes-receipt.jpg
similarity index 100%
rename from docs/cookbook/images/trader-joes-receipt.jpg
rename to docs/examples/images/trader-joes-receipt.jpg
diff --git a/docs_new/examples/index.md b/docs/examples/index.md
similarity index 100%
rename from docs_new/examples/index.md
rename to docs/examples/index.md
diff --git a/docs_new/examples/knowledge_graph_extraction.md b/docs/examples/knowledge_graph_extraction.md
similarity index 100%
rename from docs_new/examples/knowledge_graph_extraction.md
rename to docs/examples/knowledge_graph_extraction.md
diff --git a/docs_new/examples/models_playing_chess.md b/docs/examples/models_playing_chess.md
similarity index 100%
rename from docs_new/examples/models_playing_chess.md
rename to docs/examples/models_playing_chess.md
diff --git a/docs_new/examples/qa-with-citations.md b/docs/examples/qa-with-citations.md
similarity index 100%
rename from docs_new/examples/qa-with-citations.md
rename to docs/examples/qa-with-citations.md
diff --git a/docs_new/examples/react_agent.md b/docs/examples/react_agent.md
similarity index 100%
rename from docs_new/examples/react_agent.md
rename to docs/examples/react_agent.md
diff --git a/docs_new/examples/read-pdfs.md b/docs/examples/read-pdfs.md
similarity index 100%
rename from docs_new/examples/read-pdfs.md
rename to docs/examples/read-pdfs.md
diff --git a/docs_new/examples/receipt-digitization.md b/docs/examples/receipt-digitization.md
similarity index 100%
rename from docs_new/examples/receipt-digitization.md
rename to docs/examples/receipt-digitization.md
diff --git a/docs_new/examples/simtom.md b/docs/examples/simtom.md
similarity index 100%
rename from docs_new/examples/simtom.md
rename to docs/examples/simtom.md
diff --git a/docs_new/examples/structured_generation_workflow.md b/docs/examples/structured_generation_workflow.md
similarity index 100%
rename from docs_new/examples/structured_generation_workflow.md
rename to docs/examples/structured_generation_workflow.md
diff --git a/docs_new/features/advanced/logits_processors.md b/docs/features/advanced/logits_processors.md
similarity index 100%
rename from docs_new/features/advanced/logits_processors.md
rename to docs/features/advanced/logits_processors.md
diff --git a/docs_new/features/core/generator.md b/docs/features/core/generator.md
similarity index 100%
rename from docs_new/features/core/generator.md
rename to docs/features/core/generator.md
diff --git a/docs_new/features/core/output_types.md b/docs/features/core/output_types.md
similarity index 100%
rename from docs_new/features/core/output_types.md
rename to docs/features/core/output_types.md
diff --git a/docs_new/features/index.md b/docs/features/index.md
similarity index 100%
rename from docs_new/features/index.md
rename to docs/features/index.md
diff --git a/docs_new/features/models/anthropic.md b/docs/features/models/anthropic.md
similarity index 100%
rename from docs_new/features/models/anthropic.md
rename to docs/features/models/anthropic.md
diff --git a/docs_new/features/models/dottxt.md b/docs/features/models/dottxt.md
similarity index 100%
rename from docs_new/features/models/dottxt.md
rename to docs/features/models/dottxt.md
diff --git a/docs_new/features/models/gemini.md b/docs/features/models/gemini.md
similarity index 100%
rename from docs_new/features/models/gemini.md
rename to docs/features/models/gemini.md
diff --git a/docs_new/features/models/index.md b/docs/features/models/index.md
similarity index 100%
rename from docs_new/features/models/index.md
rename to docs/features/models/index.md
diff --git a/docs_new/features/models/llamacpp.md b/docs/features/models/llamacpp.md
similarity index 100%
rename from docs_new/features/models/llamacpp.md
rename to docs/features/models/llamacpp.md
diff --git a/docs_new/features/models/mlxlm.md b/docs/features/models/mlxlm.md
similarity index 100%
rename from docs_new/features/models/mlxlm.md
rename to docs/features/models/mlxlm.md
diff --git a/docs_new/features/models/ollama.md b/docs/features/models/ollama.md
similarity index 100%
rename from docs_new/features/models/ollama.md
rename to docs/features/models/ollama.md
diff --git a/docs_new/features/models/openai.md b/docs/features/models/openai.md
similarity index 100%
rename from docs_new/features/models/openai.md
rename to docs/features/models/openai.md
diff --git a/docs_new/features/models/sglang.md b/docs/features/models/sglang.md
similarity index 100%
rename from docs_new/features/models/sglang.md
rename to docs/features/models/sglang.md
diff --git a/docs_new/features/models/tgi.md b/docs/features/models/tgi.md
similarity index 100%
rename from docs_new/features/models/tgi.md
rename to docs/features/models/tgi.md
diff --git a/docs_new/features/models/transformers.md b/docs/features/models/transformers.md
similarity index 100%
rename from docs_new/features/models/transformers.md
rename to docs/features/models/transformers.md
diff --git a/docs_new/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
similarity index 100%
rename from docs_new/features/models/transformers_multimodal.md
rename to docs/features/models/transformers_multimodal.md
diff --git a/docs_new/features/models/vllm.md b/docs/features/models/vllm.md
similarity index 100%
rename from docs_new/features/models/vllm.md
rename to docs/features/models/vllm.md
diff --git a/docs_new/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
similarity index 100%
rename from docs_new/features/models/vllm_offline.md
rename to docs/features/models/vllm_offline.md
diff --git a/docs_new/features/utility/application.md b/docs/features/utility/application.md
similarity index 100%
rename from docs_new/features/utility/application.md
rename to docs/features/utility/application.md
diff --git a/docs_new/features/utility/regex_dsl.md b/docs/features/utility/regex_dsl.md
similarity index 100%
rename from docs_new/features/utility/regex_dsl.md
rename to docs/features/utility/regex_dsl.md
diff --git a/docs_new/features/utility/template.md b/docs/features/utility/template.md
similarity index 100%
rename from docs_new/features/utility/template.md
rename to docs/features/utility/template.md
diff --git a/docs/reference/chat_templating.md b/docs/guide/chat_templating.md
similarity index 100%
rename from docs/reference/chat_templating.md
rename to docs/guide/chat_templating.md
diff --git a/docs_new/guide/core_concepts.md b/docs/guide/core_concepts.md
similarity index 100%
rename from docs_new/guide/core_concepts.md
rename to docs/guide/core_concepts.md
diff --git a/docs_new/guide/getting_started.md b/docs/guide/getting_started.md
similarity index 100%
rename from docs_new/guide/getting_started.md
rename to docs/guide/getting_started.md
diff --git a/docs_new/guide/installation.md b/docs/guide/installation.md
similarity index 100%
rename from docs_new/guide/installation.md
rename to docs/guide/installation.md
diff --git a/docs_new/guide/migration.md b/docs/guide/migration.md
similarity index 100%
rename from docs_new/guide/migration.md
rename to docs/guide/migration.md
diff --git a/docs_new/guide/selecting_an_inference_backend.md b/docs/guide/selecting_an_inference_backend.md
similarity index 100%
rename from docs_new/guide/selecting_an_inference_backend.md
rename to docs/guide/selecting_an_inference_backend.md
diff --git a/docs/index.md b/docs/index.md
index 3692d32c9..ab0a36515 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,8 +1,147 @@
 ---
-title: Outlines
-template: home.html # Note that this is managed in overrides/home.html
-hide:
-  - navigation
-  - toc
-  - feedback
+title: Welcome to Outlines!
 ---
+
+# Welcome
+
+![](/assets/images/logo-light-mode.svg#only-light)
+![](/assets/images/logo-dark-mode.svg#only-dark)
+
+Outlines is a Python library from [.TXT](https://dottxt.co) that guarantees structured output from
+large language models. It ensures LLMs speak the language of your application by making them follow specific formats such as JSON, regular expressions, or context-free grammars.
+
+[Get started](/getting_started){ .md-button .md-button--primary }
+[Features](/features){ .md-button }
+[Examples](/examples){ .md-button }
+[GitHub](https://github.com/yourusername/outlines){ .md-button }
+
+## Features
+
+<div class="grid cards" markdown>
+
+- :material-shield-check: **Reliable** - Guaranteed schema compliance -- always valid JSON.
+- :material-puzzle: **Feature-rich** - Supports a large proportion of the JSON Schema spec, along with regex and context-free grammars.
+- :material-lightning-bolt: **Fast** - Outlines has negligible runtime overhead, and fast compilation times.
+- :material-cog: **Universal** - Outlines is a powered by Rust, and can be easily bound to other languages.
+- :material-lightbulb: **Simple** - Outlines is a low-abstraction library. Write code the way you normally do with LLMs. No agent frameworks needed.
+- :material-magnify: **Powerful** - Manage prompt complexity with prompt templating.
+
+</div>
+
+## Supported models
+
+> [!NOTE]
+> Provide full model list with links to docs about each model
+
+- vLLM
+- Transformers
+- OpenAI
+
+## About .txt
+
+Outlines is built with ❤️ by [.txt](https://dottxt.co).
+
+.txt solves the critical problem of reliable structured output generation for large language models. Our commercially-licensed libraries ensure 100% compliance with JSON Schema, regular expressions and context-free grammars while adding only microseconds of latency. Unlike open-source alternatives, we offer superior reliability, performance, and enterprise support.
+
+## Who is using Outlines?
+
+Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) use Outlines.
+
+Prominent companies and organizations that use Outlines include:
+
+<div class="grid cards" markdown>
+  <div class="row"><img src="../logos/amazon.png" width="200"></div>
+  <div class="row"><img src="../logos/apple.png" width="200"></div>
+  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
+  <div class="row"><img src="../logos/canoe.png" width="200"></div>
+  <div class="row"><img src="../logos/cisco.png" width="200"></div>
+  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
+  <div class="row"><img src="../logos/databricks.png" width="200"></div>
+  <div class="row"><img src="../logos/datadog.png" width="200"></div>
+  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
+  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
+  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
+  <div class="row"><img src="../logos/harvard.png" width="200"></div>
+  <div class="row"><img src="../logos/hf.png" width="200"></div>
+  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
+  <div class="row"><img src="../logos/meta.png" width="200"></div>
+  <div class="row"><img src="../logos/mit.png" width="200"></div>
+  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
+  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
+  <div class="row"><img src="../logos/nyu.png" width="200"></div>
+  <div class="row"><img src="../logos/safran.png" width="200"></div>
+  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
+  <div class="row"><img src="../logos/shopify.png" width="200"></div>
+  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
+  <div class="row"><img src="../logos/tinder.png" width="200"></div>
+  <div class="row"><img src="../logos/upenn.png" width="200"></div>
+</div>
+
+</body>
+
+Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
+
+Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
+
+## Philosophy
+
+**Outlines** is a library for neural text generation. You can think of it as a
+more flexible replacement for the `generate` method in the
+[transformers](https://github.com/huggingface/transformers) library.
+
+**Outlines** helps developers _structure text generation_ to build robust
+interfaces with external systems. It provides generation methods that
+guarantee that the output will match a regular expressions, or follow
+a JSON schema.
+
+**Outlines** provides _robust prompting primitives_ that separate the prompting
+from the execution logic and lead to simple implementations of few-shot
+generations, ReAct, meta-prompting, agents, etc.
+
+**Outlines** is designed as a _library_ that is meant to be compatible the
+broader ecosystem, not to replace it. We use as few abstractions as possible,
+and generation can be interleaved with control flow, conditionals, custom Python
+functions and calls to other libraries.
+
+**Outlines** is _compatible with every auto-regressive model_. It only interfaces with models
+via the next-token logits distribution.
+
+## Outlines people
+
+Outlines would not be what it is today without a community of dedicated developers:
+
+<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
+  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
+</a>
+
+## Acknowledgements
+
+<div class="grid" markdown>
+
+<figure markdown>
+  <a href="http://www.dottxt.co">
+  ![Normal Computing logo](assets/images/dottxt.png){ width="150" }
+  </a>
+</figure>
+
+<figure markdown>
+  <a href="https://www.normalcomputing.ai">
+  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
+  </a>
+</figure>
+
+</div>
+
+Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
+
+[discord]: https://discord.gg/R9DSu34mGd
+[aesara]: https://github.com/aesara-devs
+[blackjax]: https://github.com/blackjax-devs/blackjax
+[pythological]: https://github.com/pythological
+[hy]: https://hylang.org/
+[.txt]: https://dottxt.co
+[vllm]: https://github.com/vllm-project/vllm
+[tgi]: https://github.com/huggingface/text-generation-inference
+[lorax]: https://github.com/predibase/lorax
+[xinference]: https://github.com/xorbitsai/inference
+[sglang]: https://github.com/sgl-project/sglang/
diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index e71a6124f..381f5aa11 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -1,149 +1,22 @@
 @font-face {
-  font-family: 'Source Code Pro Custom', monospace;
-  src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
+    font-family: "Source Code Pro Custom", monospace;
+    src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
 }
 
 :root > * {
-  --md-default-bg-color: #FFFFFF;
-  --md-code-bg-color: #2E3440;
-  --md-code-fg-color: #FFFFFF;
-  --md-text-font-family: "Inter";
-  --md-code-font: "Source Code Pro Custom";
-  --md-typeset-a-color: #d16626; /*this is the brand color*/
-
-  /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
-  --md-mermaid-label-fg-color: #000000;
-  --md-mermaid-edge-color: #000000;
-}
-
-.index-pre-code {
-  max-width: 700px;
-  left: 50%;
-}
-
-.index-pre-code pre>code {
-  text-align: left;
-}
-
-.md-clipboard::after {
-  color: #FFFFFF;
-  transition: color 0.3s ease-in-out;
-}
-
-.md-clipboard:hover::after {
-  color: #D8DEE9;
-}
-
-.md-source-file {
-  text-align: center;
-  padding: 24px 0;
-}
-
-.md-typeset pre>code {
-  border-radius: .2rem;
-  box-shadow: 10px 5px 5px #D8DEE9;
-}
-
-.md-typeset p > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset strong > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-content p > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset td > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset li > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset code {
-  font-weight: 500;
-}
-
-.md-typeset pre {
-  margin-left: .5rem;
-  margin-right: .5rem;
-  margin-top: 2rem;
-  margin-bottom: 2rem;
-}
-
-.language-python {
-  background: #FFFFFF ! important
-}
-
-.language-bash {
-  background: #FFFFFF ! important
-}
-
-.language-toml {
-  background: #FFFFFF ! important
-}
-
-.language-text {
-  background: #FFFFFF ! important
-}
-
-.language-json {
-  background: #FFFFFF ! important
-}
-
-h1.title {
-  color: #FFFFFF;
-  margin: 0px 0px 5px;
-}
-
-h2.subtitle {
-  margin: 5px 0px 25px;
-  font-size: 1rem;
-  max-width: 540px;
-  margin: 0 auto;
-}
-
-.md-typeset {
-  line-height: 24px;
-  font-weight: 400;
-}
-
-.md-typeset h1 {
-  font-weight: bold;
-  color: #000000;
-}
-
-.md-typeset h2 {
-  font-weight: bold;
-  color: #000000;
-}
-
-span.md-ellipsis {
-  color: black;
-}
-
-.md-nav__link--active {
-  background-color: #ECEFF4;
-}
-
-.md-typeset ol li {
-  margin-bottom: .1rem;
-}
-
-.md-typeset ul li {
-  margin-bottom: .1rem;
+    /* Accent color shades */
+    /* --md-accent-fg-color: #d36237;
+    --md-accent-fg-color--transparent: #d36237;
+    --md-accent-bg-color: hsla(0, 0%, 100%, 1);
+    --md-accent-bg-color--light: hsla(0, 0%, 100%, 0.7); */
+
+    /* Primary color shades */
+    /* --md-accent-fg-color: hsla(#d36237, 1);
+    --md-accent-fg-color--transparent: hsla(#d36237, 0.1);
+    --md-accent-bg-color: #d36237
+    --md-accent-bg-color--light: hsla(#d36237, 0.7); */
+
+    --md-text-font-family: "Inter";
+    --md-code-font: "Source Code Pro Custom";
+    /* --md-typeset-a-color: #d36237; /*this is the brand color */
 }
diff --git a/docs_new/assets/images/logo-square.png b/docs_new/assets/images/logo-square.png
deleted file mode 100644
index 9bbb39243415b826a71ba65121b20ada5a7b9503..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3323
zcmc&%X;4#H8V#<vfo)fVB9L~likU${76E068zM47+wj;43J5eIgguf#5NSc1B~AAz
zvg<H(h8MyzAwh%$Xl1bt1SJ6_K%xyqLPFS*K*EEGT~^OHwk=a-)%>{is_w1(z5AW-
zob%m#;|n+E9b5Ib!eFo+E@w`C34>`IP>;WAZc@(>$KYu7XG_c(-*^}dfq6YNt}w}f
zdU8*KlTU&NCM*GaG42u!i^ZCL6^)7yxfpZF6cZO-JZ-NBgKfX=a_YF}<r2wA(DiX)
zVArfPej>`+eAUmyrKmLb{CB12HfHM?{V9U{^z0`Gzi9{!uN0hf+E4jogvp<;o;har
zjV9&Jt#3=6Dv(xJzd$v3?(?Mh&?eoi>9-4aoIvl|I|$jt8=8GLfDA3NUvJq8MIp^@
zH!sblC}Mi?a_lUtVH6+eVx3oOKZ+Zn4GI@T7{?a*Lu2PKYj1Bu3J8Bzagzypk6<sO
zA>!$<UVZ;1$*4%Eh$PiU%G%~`$P~n!k|TsM5*`B|BHvsIQc=fg9td-b`dD#yDuwnv
zp7z9h0c<_mLSGrkFU<Yfgg=6VAi;Te(^?Auh%FTp=3xe#IlDz?=gGP&yRUA1xbwj0
znHMz7H*CAU&*#|o_I1IUzb-|n5|@(W6UX^vk(}X3GMm6O)g{bY8d-WPo_CQ&W$hI@
z8T=s1!O_GTk(qN?jc-F14uv>+*)wf+>a${62jmYG-F0oDm@c0nE8F_Q03lg4EMPry
zGFZNupY9vDJPWR<GK4u(l3F&Ko)#54CUU}~XmV1sE&7$ipda+0$*%<+I-%@nZ#JX3
z!r}g8$7Xe6k_^jc3qJ8AvcXd|i@0A&na7scLgV@QJ=_XjM!6IWT=v&NPrfwJF4Y@G
zUqUUEfRSbdMlfSYjz+D{iMv^yq9^msmq%&{>Vr;k3p%m>N9-2XiM^&$7ont-@wBcW
zI;fYzmLXlS)wvS+!fb^usNLFFu&Fg|(0<pt-$d&zAaeo(38MiHN;m*@w%6~CoFVX@
zM$HE5PWYF@3k!o<L?Hmp!m2-9YMVB*+;?qq#5QIY)RuvbMAiX61~yjk()o4GZwK1X
zOi?4HhDC)vgb-A$Ksz&R=$Znk4Km=>Fav<q!?y8(z=)Ma9~nxnC3hk-2-8{Q9s)Lk
zsyjV{w6l7mGNx-G_3no<5@J^-yr`SAdXrxu1&JA!*&vP(dZ!VmoVg*h*<MEFJ!V6l
zpx%ywI!Q;m5VP7nDLfsfCTaj*`7AmcSrEP2t()2VJU5qHF-ULkGn0uLC0)^7M8j%F
zIxiS+-YOY;Ryo!aVj}D$L;>7w2yyi8GCdAxYG}v^O%-=1@v_(_Flf|*18#z8<ey4c
zG;_DPN1hZmAj7fvoyRu3MZLGce8b8RX4CySj_E|EG?6Can-^mcmKO6$QgkYAp~T5w
z*-pM?s=vE1Sw4JtHF@gQ@#aHPh$Lzou*qSul!-w@<W#1Y)SHp!h)@+9S%x=;KfvlY
zY9W_r#4iDDVy`4iLWo@?@RpTSIAz{DZ5}C}rwuhde--q&#??Crw;%|n3!7!GR2$PF
zaejM!F}A;qt`;40nW9{(>^RqGpQ6lG#p=|hsg*ok1-eVRd=0mxe?VftmFI_r`O$aX
zc4xGW3s|BjhyYJUtjOW=S5z-;gQKyc418*DrzjA?QNAp24#RQz298ZD>G_2{s=O3;
zn%v*Wk|Im9Ymh%==dc29Y#@U;?h`M_jx?HKu%LWM(6=Oa^f0)JymK^RrnX-s-zQaE
zg0w8>&YjWR@X7m~=S_mc?jTtCT+TMtakX({Y&JvJGawG3q+ckD;7ubetf*WqjG^6}
zO@=dsD(3bbLF45t!5;7Z?I1XaNKlslFq%Gj5OI(21K69xiqf}>3!=$y-OcbTMI;16
zD|h#RKN;=YdFq2c7*kLOS`%xR-DFC7l|oQS8(1CdyJ76uGAJzZbx=atVENv{NuZqU
zi|^`YMb*oS)FL6~uo{?)!*h>}39+FQXZQO60AnPOD=o$CNb2Jur2WUU{Bt3lYmwBW
z%Y9XMzl|~sO2$uxw0`!tA#rggb9L)@{?2zT`JKw&bM*ljI;|W({0Ij>^Hhgg*{_Q2
z9r#kktE4C&<rjMK_ErcNB=W+6&fZB%YV1OZ$$6e_(%+?B$LC$Mi=&el8*~k74i%G4
z6rNm?{O1G2um7tHyz|R{H;9kE^oEDC71AjN^ODHnf-!}wt5s?%Yp+t=F1i2!QduuZ
zkaF;CZa54P;F2AE)~2!MuK}*#D#J;dW-c?V``(wFTZO%+ujvIC9IamwaO_}aY&8}B
z56XK3;e>}-%#mld%()uhT7S~$`lNqG$QE9FI+bQGGxYeh8jNHx8ogVwzaY}s1&RRp
z+6wg`@JA3Fg;#kI-5Hz2|4<l{kT2ps_W@`raZq{{7KSWyKJniE|Iy3)PCyn3DTu49
z$Bn7{P_@N4?Po4h$Te%QY;lZ(|I4{L5af!}MbK!K;!0^lk%Je2DZ&w}lJ{@;k4wrs
zG2hU0dY2bKTOo3%hB>IVJik>j$0R^YZBVz|tN93-R@>wGGn&75`FUU5-@h`gbyv<3
zXIp@lwKqQ|oJ>KzF8EUvAt_lh+wUPYKdkmIup@&p#PGmq*!kQ4l6Ut1kT<}u2HEb6
zjpAyI8>4;vSRELdjD|`*q_7tIkFIa?S};7xTrmVj8m;G1mr(2$RZT}X>n_GY76tji
zi!_p?{n92?+d(0F^pJ%m`Tj*+_){dn&eNies%Bi&&cg<{xeqcP+{!yp>p;Ls*%}~d
zoyT9I4}`;cSrsdI_1O<soLrO8J5xj(nPn0QWr@U(SEQ?n<DFfMqFc$dg4mY~sw*qy
zi#@CJ%N!J}V9T122x-J_X(UG?R{lV}^WzGS$d=2aGm~Tm<%-7>qMs(g`ynC+0WkJx
ooF=m)A%l%qHm_eBt7g+qwHY~_RyAv@A1*K#C%02ICoWw57oI%f82|tP

diff --git a/docs_new/assets/images/logo-square.svg b/docs_new/assets/images/logo-square.svg
deleted file mode 100644
index efc68a642..000000000
--- a/docs_new/assets/images/logo-square.svg
+++ /dev/null
@@ -1,133 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<svg
-   width="69.82"
-   height="94.089996"
-   viewBox="0 0 69.82 94.089996"
-   fill="none"
-   version="1.1"
-   id="svg39"
-   sodipodi:docname="Outlines_white dottxt.ai.svg"
-   inkscape:export-filename="../projects/outlines/docs/assets/images/logo-dark-mode.png"
-   inkscape:export-xdpi="137.93103"
-   inkscape:export-ydpi="137.93103"
-   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
-   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
-   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
-   xmlns="http://www.w3.org/2000/svg"
-   xmlns:svg="http://www.w3.org/2000/svg">
-  <sodipodi:namedview
-     id="namedview39"
-     pagecolor="#ffffff"
-     bordercolor="#000000"
-     borderopacity="0.25"
-     inkscape:showpageshadow="2"
-     inkscape:pageopacity="0.0"
-     inkscape:pagecheckerboard="0"
-     inkscape:deskcolor="#d1d1d1"
-     inkscape:zoom="0.27368421"
-     inkscape:cx="171.73077"
-     inkscape:cy="45.673077"
-     inkscape:window-width="1416"
-     inkscape:window-height="820"
-     inkscape:window-x="30"
-     inkscape:window-y="38"
-     inkscape:window-maximized="0"
-     inkscape:current-layer="g39" />
-  <g
-     clip-path="url(#clip0_56_2)"
-     id="g39">
-    <path
-       d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
-       fill="#e35a26"
-       id="path12" />
-    <path
-       d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
-       fill="#e35a26"
-       id="path13" />
-    <path
-       d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path14" />
-    <path
-       d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path15" />
-    <path
-       d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path16" />
-    <path
-       d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path17" />
-    <path
-       d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path18" />
-    <path
-       d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
-       fill="#e35a26"
-       id="path19" />
-    <path
-       d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
-       fill="#e35a26"
-       id="path20" />
-    <path
-       d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
-       fill="#e35a26"
-       id="path21" />
-    <path
-       d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
-       fill="#e35a26"
-       id="path22" />
-    <path
-       d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path23" />
-    <path
-       d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path24" />
-    <path
-       d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path25" />
-    <path
-       d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path26" />
-    <path
-       d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path27" />
-    <path
-       d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path28" />
-    <path
-       d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path29" />
-    <path
-       d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path30" />
-    <path
-       d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
-       fill="#e35a26"
-       id="path31" />
-  </g>
-  <defs
-     id="defs39">
-    <clipPath
-       id="clip0_56_2">
-      <rect
-         width="347.60001"
-         height="94.089996"
-         fill="#ffffff"
-         id="rect39"
-         x="0"
-         y="0" />
-    </clipPath>
-  </defs>
-</svg>
diff --git a/docs_new/index.md b/docs_new/index.md
deleted file mode 100644
index ab0a36515..000000000
--- a/docs_new/index.md
+++ /dev/null
@@ -1,147 +0,0 @@
----
-title: Welcome to Outlines!
----
-
-# Welcome
-
-![](/assets/images/logo-light-mode.svg#only-light)
-![](/assets/images/logo-dark-mode.svg#only-dark)
-
-Outlines is a Python library from [.TXT](https://dottxt.co) that guarantees structured output from
-large language models. It ensures LLMs speak the language of your application by making them follow specific formats such as JSON, regular expressions, or context-free grammars.
-
-[Get started](/getting_started){ .md-button .md-button--primary }
-[Features](/features){ .md-button }
-[Examples](/examples){ .md-button }
-[GitHub](https://github.com/yourusername/outlines){ .md-button }
-
-## Features
-
-<div class="grid cards" markdown>
-
-- :material-shield-check: **Reliable** - Guaranteed schema compliance -- always valid JSON.
-- :material-puzzle: **Feature-rich** - Supports a large proportion of the JSON Schema spec, along with regex and context-free grammars.
-- :material-lightning-bolt: **Fast** - Outlines has negligible runtime overhead, and fast compilation times.
-- :material-cog: **Universal** - Outlines is a powered by Rust, and can be easily bound to other languages.
-- :material-lightbulb: **Simple** - Outlines is a low-abstraction library. Write code the way you normally do with LLMs. No agent frameworks needed.
-- :material-magnify: **Powerful** - Manage prompt complexity with prompt templating.
-
-</div>
-
-## Supported models
-
-> [!NOTE]
-> Provide full model list with links to docs about each model
-
-- vLLM
-- Transformers
-- OpenAI
-
-## About .txt
-
-Outlines is built with ❤️ by [.txt](https://dottxt.co).
-
-.txt solves the critical problem of reliable structured output generation for large language models. Our commercially-licensed libraries ensure 100% compliance with JSON Schema, regular expressions and context-free grammars while adding only microseconds of latency. Unlike open-source alternatives, we offer superior reliability, performance, and enterprise support.
-
-## Who is using Outlines?
-
-Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) use Outlines.
-
-Prominent companies and organizations that use Outlines include:
-
-<div class="grid cards" markdown>
-  <div class="row"><img src="../logos/amazon.png" width="200"></div>
-  <div class="row"><img src="../logos/apple.png" width="200"></div>
-  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
-  <div class="row"><img src="../logos/canoe.png" width="200"></div>
-  <div class="row"><img src="../logos/cisco.png" width="200"></div>
-  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
-  <div class="row"><img src="../logos/databricks.png" width="200"></div>
-  <div class="row"><img src="../logos/datadog.png" width="200"></div>
-  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
-  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
-  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
-  <div class="row"><img src="../logos/harvard.png" width="200"></div>
-  <div class="row"><img src="../logos/hf.png" width="200"></div>
-  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
-  <div class="row"><img src="../logos/meta.png" width="200"></div>
-  <div class="row"><img src="../logos/mit.png" width="200"></div>
-  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
-  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
-  <div class="row"><img src="../logos/nyu.png" width="200"></div>
-  <div class="row"><img src="../logos/safran.png" width="200"></div>
-  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
-  <div class="row"><img src="../logos/shopify.png" width="200"></div>
-  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
-  <div class="row"><img src="../logos/tinder.png" width="200"></div>
-  <div class="row"><img src="../logos/upenn.png" width="200"></div>
-</div>
-
-</body>
-
-Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
-
-Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
-
-## Philosophy
-
-**Outlines** is a library for neural text generation. You can think of it as a
-more flexible replacement for the `generate` method in the
-[transformers](https://github.com/huggingface/transformers) library.
-
-**Outlines** helps developers _structure text generation_ to build robust
-interfaces with external systems. It provides generation methods that
-guarantee that the output will match a regular expressions, or follow
-a JSON schema.
-
-**Outlines** provides _robust prompting primitives_ that separate the prompting
-from the execution logic and lead to simple implementations of few-shot
-generations, ReAct, meta-prompting, agents, etc.
-
-**Outlines** is designed as a _library_ that is meant to be compatible the
-broader ecosystem, not to replace it. We use as few abstractions as possible,
-and generation can be interleaved with control flow, conditionals, custom Python
-functions and calls to other libraries.
-
-**Outlines** is _compatible with every auto-regressive model_. It only interfaces with models
-via the next-token logits distribution.
-
-## Outlines people
-
-Outlines would not be what it is today without a community of dedicated developers:
-
-<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
-  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
-</a>
-
-## Acknowledgements
-
-<div class="grid" markdown>
-
-<figure markdown>
-  <a href="http://www.dottxt.co">
-  ![Normal Computing logo](assets/images/dottxt.png){ width="150" }
-  </a>
-</figure>
-
-<figure markdown>
-  <a href="https://www.normalcomputing.ai">
-  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
-  </a>
-</figure>
-
-</div>
-
-Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
-
-[discord]: https://discord.gg/R9DSu34mGd
-[aesara]: https://github.com/aesara-devs
-[blackjax]: https://github.com/blackjax-devs/blackjax
-[pythological]: https://github.com/pythological
-[hy]: https://hylang.org/
-[.txt]: https://dottxt.co
-[vllm]: https://github.com/vllm-project/vllm
-[tgi]: https://github.com/huggingface/text-generation-inference
-[lorax]: https://github.com/predibase/lorax
-[xinference]: https://github.com/xorbitsai/inference
-[sglang]: https://github.com/sgl-project/sglang/
diff --git a/docs_new/stylesheets/extra.css b/docs_new/stylesheets/extra.css
deleted file mode 100644
index 381f5aa11..000000000
--- a/docs_new/stylesheets/extra.css
+++ /dev/null
@@ -1,22 +0,0 @@
-@font-face {
-    font-family: "Source Code Pro Custom", monospace;
-    src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
-}
-
-:root > * {
-    /* Accent color shades */
-    /* --md-accent-fg-color: #d36237;
-    --md-accent-fg-color--transparent: #d36237;
-    --md-accent-bg-color: hsla(0, 0%, 100%, 1);
-    --md-accent-bg-color--light: hsla(0, 0%, 100%, 0.7); */
-
-    /* Primary color shades */
-    /* --md-accent-fg-color: hsla(#d36237, 1);
-    --md-accent-fg-color--transparent: hsla(#d36237, 0.1);
-    --md-accent-bg-color: #d36237
-    --md-accent-bg-color--light: hsla(#d36237, 0.7); */
-
-    --md-text-font-family: "Inter";
-    --md-code-font: "Source Code Pro Custom";
-    /* --md-typeset-a-color: #d36237; /*this is the brand color */
-}
diff --git a/docs/api/applications.md b/docs_old/api/applications.md
similarity index 100%
rename from docs/api/applications.md
rename to docs_old/api/applications.md
diff --git a/docs/api/core/application.md b/docs_old/api/core/application.md
similarity index 100%
rename from docs/api/core/application.md
rename to docs_old/api/core/application.md
diff --git a/docs/api/guide.md b/docs_old/api/guide.md
similarity index 100%
rename from docs/api/guide.md
rename to docs_old/api/guide.md
diff --git a/docs_new/api_reference/index.md b/docs_old/api/index.md
similarity index 100%
rename from docs_new/api_reference/index.md
rename to docs_old/api/index.md
diff --git a/docs/api/json_schema.md b/docs_old/api/json_schema.md
similarity index 100%
rename from docs/api/json_schema.md
rename to docs_old/api/json_schema.md
diff --git a/docs/api/models.md b/docs_old/api/models.md
similarity index 100%
rename from docs/api/models.md
rename to docs_old/api/models.md
diff --git a/docs/api/parsing.md b/docs_old/api/parsing.md
similarity index 100%
rename from docs/api/parsing.md
rename to docs_old/api/parsing.md
diff --git a/docs/api/regex.md b/docs_old/api/regex.md
similarity index 100%
rename from docs/api/regex.md
rename to docs_old/api/regex.md
diff --git a/docs/api/templates.md b/docs_old/api/templates.md
similarity index 100%
rename from docs/api/templates.md
rename to docs_old/api/templates.md
diff --git a/docs_new/assets/images/dottxt-dark.svg b/docs_old/assets/images/dottxt-dark.svg
similarity index 100%
rename from docs_new/assets/images/dottxt-dark.svg
rename to docs_old/assets/images/dottxt-dark.svg
diff --git a/docs_new/assets/images/dottxt-light.svg b/docs_old/assets/images/dottxt-light.svg
similarity index 100%
rename from docs_new/assets/images/dottxt-light.svg
rename to docs_old/assets/images/dottxt-light.svg
diff --git a/docs_new/assets/images/dottxt.png b/docs_old/assets/images/dottxt.png
similarity index 100%
rename from docs_new/assets/images/dottxt.png
rename to docs_old/assets/images/dottxt.png
diff --git a/docs_new/assets/images/install.png b/docs_old/assets/images/install.png
similarity index 100%
rename from docs_new/assets/images/install.png
rename to docs_old/assets/images/install.png
diff --git a/docs_new/assets/images/logits_processing_diagram.svg b/docs_old/assets/images/logits_processing_diagram.svg
similarity index 100%
rename from docs_new/assets/images/logits_processing_diagram.svg
rename to docs_old/assets/images/logits_processing_diagram.svg
diff --git a/docs_new/assets/images/logo-dark-mode.png b/docs_old/assets/images/logo-dark-mode.png
similarity index 100%
rename from docs_new/assets/images/logo-dark-mode.png
rename to docs_old/assets/images/logo-dark-mode.png
diff --git a/docs_new/assets/images/logo-dark-mode.svg b/docs_old/assets/images/logo-dark-mode.svg
similarity index 100%
rename from docs_new/assets/images/logo-dark-mode.svg
rename to docs_old/assets/images/logo-dark-mode.svg
diff --git a/docs_new/assets/images/logo-light-mode.png b/docs_old/assets/images/logo-light-mode.png
similarity index 100%
rename from docs_new/assets/images/logo-light-mode.png
rename to docs_old/assets/images/logo-light-mode.png
diff --git a/docs_new/assets/images/logo-light-mode.svg b/docs_old/assets/images/logo-light-mode.svg
similarity index 100%
rename from docs_new/assets/images/logo-light-mode.svg
rename to docs_old/assets/images/logo-light-mode.svg
diff --git a/docs_new/assets/images/logo-simple.png b/docs_old/assets/images/logo-simple.png
similarity index 100%
rename from docs_new/assets/images/logo-simple.png
rename to docs_old/assets/images/logo-simple.png
diff --git a/docs_old/assets/images/logo-square.png b/docs_old/assets/images/logo-square.png
new file mode 100644
index 0000000000000000000000000000000000000000..0592b8fa9a2d43e08f7367e187514beabc685e2f
GIT binary patch
literal 2233
zcmcgue>9VO7=PXM8_DUO${Zw>>J%v?8>;KK?o<p-k(Mas$8e|ld9J&pj4P!(W4j2M
zm<WY6+G)EYn%`%b7Ny24v5n2RukJbBQqmvy+&{kOJ?DMi_kBLk=lML(=S@6<uv1+H
zT?GJus)N1tF#u3f1<!#M%fLJN?c1xt%gSK;<DmesJ!S4Fwd<OIPXHQWHXdOvL1)9T
zJ{VsBi^Uq83k(QF`vm(M1Y!J0qi`qytgv;kKIj^eDeP-WAtN=|d^udCvcoYr-k+#z
zaJy~CQPlBlPS#4N_0Xp`CiRZG7i%iko@ltrT4R*4!Q9p30n0nYN3d$!`L*VG&^8U|
zMx)f<wwYxOJ$SAfk}+{DT3(AEMzv&x8fG~6mz!NeiRip%m0ew4G^*5@zWlJy1~yZU
zphs-YAa?DkgQfWc>KtHGj*_)C<jR$0Yt&UYZh{`LHjcaUPr%8=FnAYzqOM+1Q(nD=
zNJM676!k?_#P(Ae1yoD*5{cAHXvkM->TBK|HTY6i|Ax(k?SPtS#=>2rO*I~#^P5WV
zA)Cu(9Wx4vj4h&aEp&t(a+*7(d-|9$q^n*c;}WT3<TEVslCP!z6jxbkkh6i-{s?a!
zR!tf)WQ1|CnErcHtwctGRrwax*Ki-(qB99<t~4~~`7wF#(*Y8L)aH2LAsNv9R)t`S
zlE$7iC%e_WeU)QH4NXUw6r3jGX48q7(5WZ>M&~0Hb<!fK=Mc_**w5Xx+3I@I1)Y;M
z43_L=ts5#V`cD5{a&HtyU?IVr%(({}UXG##ZQhF(#zguKOryuSiL?G4I)dUribz-Z
zmU=aZgVPf1lJ6Twv{HF@sQ9GI!z;gF{nN6+c5IG`#JtUh1k|7{Qk(8my$L40q^92c
zp;M<THG=F!LS=M7<2W4c52NMng~1xFG}5uLMk&#OOPp$8Gz6V_b5O0*K*a04%n#nv
zN>%Wi3G2y(meZqdWHTwd5Zm2oM=_UUkaW0Md)lyT?dJ-ZhXs_zU#$`uDrM_s{?tvv
zGt+|TxTQn4zEsY<_cqUtI&{blN_3_tO!^0x8EMXn>q9(Z42b87XW+Ex8Sfr7?a9Gj
zt90X$+RFEnc;$vBZ$^|fA>maY>~%$3jwclB)rS>4?re`7c_p+)U%WF$40zO_nxp2l
z-fdrVpNUwz%P~N4>nCv!*eQy|Nu*R?udxy4EkgE_+H94BGM^z_JH`s7-PaZ;d{9_x
zG36<iP0c~&pdx(*S|xd2!ZCrOb3X-RO7;)WXURD5%7^L`ZUqHAz9Y9YUS4i^7iqyL
zyLZXopa<yc_@L}+ePW1((!%34kEDO+?*9nP&GwdC$x#=??@On%3#D<}cM@Q<QCPT7
z0I24&QqBgya?$E&xT#osH$KD^b`;4j;Gqpm^gr1TU&Wb<kB)3DLR5K)vT$mD9104@
zf&88E$-Rfej5knm;H<<|Vg5T>Qt=9p)=t@s$W_%Yp9je|BBT1uX5Y1Yj`;fIa!@tp
zI4e-uyAwOtNy?MH4}BR1GaiA|qTtkC6i>H)V?Tbi{slMc2c&s=%SFS?2x)s{-H%+e
zOi<MLerGz{Yn>@{;6l`&I4#l~X%vD!^}KE_ANRuooB3nr<K0#I%+Yt{mPvjEVD-;`
zNx&<r7^R#^kun1xRSFTEUA*Q65zh~lmFhTgEvqH=V)Hp!B)aLlAp*52I0Ah#Zy^u)
z3@KfD4ap335h<m;8F*6Y{@^;Ez!I>g5{d-;LMgy|y0{N~=1(8LJ>R&%09O((XAoD5
zy`;7YD&ex^G?LZNElMRkC%djkt!&xhJ#8q9G;(<uoAnyqP?x8N)e<n@G2h(arBfY^
zPmyu*sEV12Gq(u?xgQMkdnS2j9%NutbBM)<aN;X0lM*UB)!-!Z(i3!3m6T-k>_OxE
z@|4iG`$LS~$XuBp(nhJ<9QVSdSjL~&`9dJJOvVOglH{??i#IGa7NYgUCOf3JwIBTL
O032)()+L8d{rWfXa;-}M

literal 0
HcmV?d00001

diff --git a/docs_old/assets/images/logo-square.svg b/docs_old/assets/images/logo-square.svg
new file mode 100644
index 000000000..29b1c20f2
--- /dev/null
+++ b/docs_old/assets/images/logo-square.svg
@@ -0,0 +1,124 @@
+<?xml version="1.0" encoding="UTF-8" standalone="no"?>
+<svg
+   width="95.38842"
+   height="113.55765"
+   viewBox="0 0 95.388419 113.55765"
+   fill="none"
+   version="1.1"
+   id="svg314"
+   sodipodi:docname="logo-square.svg"
+   xml:space="preserve"
+   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
+   inkscape:export-filename="logo.png"
+   inkscape:export-xdpi="270.25565"
+   inkscape:export-ydpi="270.25565"
+   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
+   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
+   xmlns="http://www.w3.org/2000/svg"
+   xmlns:svg="http://www.w3.org/2000/svg"><defs
+     id="defs314" /><sodipodi:namedview
+     id="namedview314"
+     pagecolor="#ffffff"
+     bordercolor="#000000"
+     borderopacity="0.25"
+     inkscape:showpageshadow="2"
+     inkscape:pageopacity="0.0"
+     inkscape:pagecheckerboard="0"
+     inkscape:deskcolor="#d1d1d1"
+     showgrid="false"
+     showguides="true"
+     inkscape:zoom="1.0083333"
+     inkscape:cx="280.66117"
+     inkscape:cy="25.785125"
+     inkscape:window-width="3440"
+     inkscape:window-height="1416"
+     inkscape:window-x="0"
+     inkscape:window-y="0"
+     inkscape:window-maximized="1"
+     inkscape:current-layer="svg314"><sodipodi:guide
+       position="-146.75933,64.2455"
+       orientation="0,-1"
+       id="guide314"
+       inkscape:locked="false" /><sodipodi:guide
+       position="275.80377,3.1942902"
+       orientation="1,0"
+       id="guide315"
+       inkscape:locked="false" /></sodipodi:namedview><g
+     id="g13"
+     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
+     style="fill:#e35a26;fill-opacity:1"><g
+       id="g314"
+       transform="translate(1367.0913,152.50405)"
+       style="fill:#e35a26;fill-opacity:1"><rect
+         x="28"
+         y="644"
+         width="28"
+         height="28"
+         id="rect1"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
+         id="path1-7"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="28"
+         y="112"
+         width="28"
+         height="28"
+         id="rect2"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,560 v 28 h 420 v -28 z"
+         id="path2-0"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,476 v 28 h 420 v -28 z"
+         id="path3-4"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,392 v 28 h 420 v -28 z"
+         id="path4-0"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 84,308 v 28 h 420 v -28 z"
+         id="path5-8"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 364,168 v 28 h 140 v -28 z"
+         id="path6-5"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 336,196 v -28 h -56 v 28 z"
+         id="path7-1"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="M 336,112 V 84 h -56 v 28 z"
+         id="path8-6"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 364,84 v 28 H 504 V 84 Z"
+         id="path9-6"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 56,672 v 28 h 476 v -28 z"
+         id="path10-2"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="56"
+         y="84"
+         width="28"
+         height="28"
+         id="rect10"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="112"
+         y="28"
+         width="28"
+         height="28"
+         id="rect11"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="84"
+         y="56"
+         width="28"
+         height="28"
+         id="rect12"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
+         id="path12-1"
+         style="fill:#e35a26;fill-opacity:1" /><path
+         d="m 560,28 v 616 h 28 V 28 Z"
+         id="path13-9"
+         style="fill:#e35a26;fill-opacity:1" /><rect
+         x="532"
+         y="644"
+         width="28"
+         height="28"
+         id="rect13"
+         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
diff --git a/docs_new/assets/images/logo.png b/docs_old/assets/images/logo.png
similarity index 100%
rename from docs_new/assets/images/logo.png
rename to docs_old/assets/images/logo.png
diff --git a/docs_new/assets/images/logo.svg b/docs_old/assets/images/logo.svg
similarity index 100%
rename from docs_new/assets/images/logo.svg
rename to docs_old/assets/images/logo.svg
diff --git a/docs_new/assets/images/normal_computing.jpg b/docs_old/assets/images/normal_computing.jpg
similarity index 100%
rename from docs_new/assets/images/normal_computing.jpg
rename to docs_old/assets/images/normal_computing.jpg
diff --git a/docs_new/assets/images/readme-dark.png b/docs_old/assets/images/readme-dark.png
similarity index 100%
rename from docs_new/assets/images/readme-dark.png
rename to docs_old/assets/images/readme-dark.png
diff --git a/docs_new/assets/images/readme-light.png b/docs_old/assets/images/readme-light.png
similarity index 100%
rename from docs_new/assets/images/readme-light.png
rename to docs_old/assets/images/readme-light.png
diff --git a/docs_new/assets/images/use_philosophy.png b/docs_old/assets/images/use_philosophy.png
similarity index 100%
rename from docs_new/assets/images/use_philosophy.png
rename to docs_old/assets/images/use_philosophy.png
diff --git a/docs/blog/.authors.yml b/docs_old/blog/.authors.yml
similarity index 100%
rename from docs/blog/.authors.yml
rename to docs_old/blog/.authors.yml
diff --git a/docs/blog/assets/4000_stars.png b/docs_old/blog/assets/4000_stars.png
similarity index 100%
rename from docs/blog/assets/4000_stars.png
rename to docs_old/blog/assets/4000_stars.png
diff --git a/docs_new/blog/index.md b/docs_old/blog/index.md
similarity index 100%
rename from docs_new/blog/index.md
rename to docs_old/blog/index.md
diff --git a/docs/blog/posts/roadmap-2024.md b/docs_old/blog/posts/roadmap-2024.md
similarity index 100%
rename from docs/blog/posts/roadmap-2024.md
rename to docs_old/blog/posts/roadmap-2024.md
diff --git a/docs/community/belonging.png b/docs_old/community/belonging.png
similarity index 100%
rename from docs/community/belonging.png
rename to docs_old/community/belonging.png
diff --git a/docs_new/community/contribute.md b/docs_old/community/contribute.md
similarity index 100%
rename from docs_new/community/contribute.md
rename to docs_old/community/contribute.md
diff --git a/docs_new/community/examples.md b/docs_old/community/examples.md
similarity index 100%
rename from docs_new/community/examples.md
rename to docs_old/community/examples.md
diff --git a/docs_new/community/feedback.md b/docs_old/community/feedback.md
similarity index 100%
rename from docs_new/community/feedback.md
rename to docs_old/community/feedback.md
diff --git a/docs_new/community/index.md b/docs_old/community/index.md
similarity index 94%
rename from docs_new/community/index.md
rename to docs_old/community/index.md
index 517bf59c5..9519f5b56 100644
--- a/docs_new/community/index.md
+++ b/docs_old/community/index.md
@@ -1,9 +1,12 @@
 # Community
 
+![Belonging](belonging.png)
+
 Outlines exists for a community of users who believe software doesn't need to be complicated. Who share the same passion for Large Language Models but don't want to compromise on robustness. Together, we are bringing these powerful models back to the world of software.
 
 ## Connect on Discord
 
 The Outlines community lives on our Discord server. There you can ask questions, share ideas or just chat with people like you. Don't be a stranger and [join us][discord].
 
+
 [discord]: https://discord.gg/UppQmhEpe8
diff --git a/docs_new/community/versioning.md b/docs_old/community/versioning.md
similarity index 87%
rename from docs_new/community/versioning.md
rename to docs_old/community/versioning.md
index a6e12f218..023b92537 100644
--- a/docs_new/community/versioning.md
+++ b/docs_old/community/versioning.md
@@ -13,10 +13,6 @@ Each part of the version number (`major.minor.patch`) conveys information about
 - **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatability with existing use of core interfaces.
 - **Patch Releases** address bug fixes and incorporate low-risk changes to improve stability and performance.
 
-!!! note "Breaking Changes"
-
-    Outlines v1.0 introduced several breaking changes to the core interface. See [the migration guide](/user_guide/migration) for more details.
-
 ## Releases
 
 Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/dottxt-ai/outlines/releases).
diff --git a/docs/content/docs/guides/migration_guide.md b/docs_old/content/docs/guides/migration_guide.md
similarity index 100%
rename from docs/content/docs/guides/migration_guide.md
rename to docs_old/content/docs/guides/migration_guide.md
diff --git a/docs/cookbook/atomic_caption.md b/docs_old/cookbook/atomic_caption.md
similarity index 100%
rename from docs/cookbook/atomic_caption.md
rename to docs_old/cookbook/atomic_caption.md
diff --git a/docs/cookbook/chain_of_density.md b/docs_old/cookbook/chain_of_density.md
similarity index 100%
rename from docs/cookbook/chain_of_density.md
rename to docs_old/cookbook/chain_of_density.md
diff --git a/docs/cookbook/chain_of_thought.md b/docs_old/cookbook/chain_of_thought.md
similarity index 100%
rename from docs/cookbook/chain_of_thought.md
rename to docs_old/cookbook/chain_of_thought.md
diff --git a/docs/cookbook/classification.md b/docs_old/cookbook/classification.md
similarity index 100%
rename from docs/cookbook/classification.md
rename to docs_old/cookbook/classification.md
diff --git a/docs/cookbook/dating_profiles.md b/docs_old/cookbook/dating_profiles.md
similarity index 100%
rename from docs/cookbook/dating_profiles.md
rename to docs_old/cookbook/dating_profiles.md
diff --git a/docs/cookbook/deploy-using-bentoml.md b/docs_old/cookbook/deploy-using-bentoml.md
similarity index 100%
rename from docs/cookbook/deploy-using-bentoml.md
rename to docs_old/cookbook/deploy-using-bentoml.md
diff --git a/docs/cookbook/deploy-using-cerebrium.md b/docs_old/cookbook/deploy-using-cerebrium.md
similarity index 100%
rename from docs/cookbook/deploy-using-cerebrium.md
rename to docs_old/cookbook/deploy-using-cerebrium.md
diff --git a/docs/cookbook/deploy-using-modal.md b/docs_old/cookbook/deploy-using-modal.md
similarity index 100%
rename from docs/cookbook/deploy-using-modal.md
rename to docs_old/cookbook/deploy-using-modal.md
diff --git a/docs/cookbook/earnings-reports.md b/docs_old/cookbook/earnings-reports.md
similarity index 100%
rename from docs/cookbook/earnings-reports.md
rename to docs_old/cookbook/earnings-reports.md
diff --git a/docs/cookbook/extract_event_details.md b/docs_old/cookbook/extract_event_details.md
similarity index 100%
rename from docs/cookbook/extract_event_details.md
rename to docs_old/cookbook/extract_event_details.md
diff --git a/docs/cookbook/extract_event_details.py b/docs_old/cookbook/extract_event_details.py
similarity index 100%
rename from docs/cookbook/extract_event_details.py
rename to docs_old/cookbook/extract_event_details.py
diff --git a/docs/cookbook/extraction.md b/docs_old/cookbook/extraction.md
similarity index 100%
rename from docs/cookbook/extraction.md
rename to docs_old/cookbook/extraction.md
diff --git a/docs_new/examples/images/chain_of_density.png b/docs_old/cookbook/images/chain_of_density.png
similarity index 100%
rename from docs_new/examples/images/chain_of_density.png
rename to docs_old/cookbook/images/chain_of_density.png
diff --git a/docs_new/examples/images/coding_structure_diagram.png b/docs_old/cookbook/images/coding_structure_diagram.png
similarity index 100%
rename from docs_new/examples/images/coding_structure_diagram.png
rename to docs_old/cookbook/images/coding_structure_diagram.png
diff --git a/docs_new/examples/images/knowledge-graph-extraction.png b/docs_old/cookbook/images/knowledge-graph-extraction.png
similarity index 100%
rename from docs_new/examples/images/knowledge-graph-extraction.png
rename to docs_old/cookbook/images/knowledge-graph-extraction.png
diff --git a/docs_new/examples/images/nvidia-income.png b/docs_old/cookbook/images/nvidia-income.png
similarity index 100%
rename from docs_new/examples/images/nvidia-income.png
rename to docs_old/cookbook/images/nvidia-income.png
diff --git a/docs_new/examples/images/simtom.png b/docs_old/cookbook/images/simtom.png
similarity index 100%
rename from docs_new/examples/images/simtom.png
rename to docs_old/cookbook/images/simtom.png
diff --git a/docs_new/examples/images/trader-joes-receipt.jpg b/docs_old/cookbook/images/trader-joes-receipt.jpg
similarity index 100%
rename from docs_new/examples/images/trader-joes-receipt.jpg
rename to docs_old/cookbook/images/trader-joes-receipt.jpg
diff --git a/docs/cookbook/index.md b/docs_old/cookbook/index.md
similarity index 100%
rename from docs/cookbook/index.md
rename to docs_old/cookbook/index.md
diff --git a/docs/cookbook/knowledge_graph_extraction.md b/docs_old/cookbook/knowledge_graph_extraction.md
similarity index 100%
rename from docs/cookbook/knowledge_graph_extraction.md
rename to docs_old/cookbook/knowledge_graph_extraction.md
diff --git a/docs/cookbook/models_playing_chess.md b/docs_old/cookbook/models_playing_chess.md
similarity index 100%
rename from docs/cookbook/models_playing_chess.md
rename to docs_old/cookbook/models_playing_chess.md
diff --git a/docs/cookbook/prompt_templates/chain_of_density.txt b/docs_old/cookbook/prompt_templates/chain_of_density.txt
similarity index 100%
rename from docs/cookbook/prompt_templates/chain_of_density.txt
rename to docs_old/cookbook/prompt_templates/chain_of_density.txt
diff --git a/docs/cookbook/prompt_templates/classification.txt b/docs_old/cookbook/prompt_templates/classification.txt
similarity index 100%
rename from docs/cookbook/prompt_templates/classification.txt
rename to docs_old/cookbook/prompt_templates/classification.txt
diff --git a/docs/cookbook/prompt_templates/react_agent.txt b/docs_old/cookbook/prompt_templates/react_agent.txt
similarity index 100%
rename from docs/cookbook/prompt_templates/react_agent.txt
rename to docs_old/cookbook/prompt_templates/react_agent.txt
diff --git a/docs/cookbook/prompt_templates/simtom_prospective_taking.txt b/docs_old/cookbook/prompt_templates/simtom_prospective_taking.txt
similarity index 100%
rename from docs/cookbook/prompt_templates/simtom_prospective_taking.txt
rename to docs_old/cookbook/prompt_templates/simtom_prospective_taking.txt
diff --git a/docs/cookbook/prompt_templates/simtom_simulation.txt b/docs_old/cookbook/prompt_templates/simtom_simulation.txt
similarity index 100%
rename from docs/cookbook/prompt_templates/simtom_simulation.txt
rename to docs_old/cookbook/prompt_templates/simtom_simulation.txt
diff --git a/docs/cookbook/qa-with-citations.md b/docs_old/cookbook/qa-with-citations.md
similarity index 100%
rename from docs/cookbook/qa-with-citations.md
rename to docs_old/cookbook/qa-with-citations.md
diff --git a/docs/cookbook/react_agent.md b/docs_old/cookbook/react_agent.md
similarity index 100%
rename from docs/cookbook/react_agent.md
rename to docs_old/cookbook/react_agent.md
diff --git a/docs/cookbook/read-pdfs.md b/docs_old/cookbook/read-pdfs.md
similarity index 100%
rename from docs/cookbook/read-pdfs.md
rename to docs_old/cookbook/read-pdfs.md
diff --git a/docs/cookbook/receipt-digitization.md b/docs_old/cookbook/receipt-digitization.md
similarity index 100%
rename from docs/cookbook/receipt-digitization.md
rename to docs_old/cookbook/receipt-digitization.md
diff --git a/docs/cookbook/simtom.md b/docs_old/cookbook/simtom.md
similarity index 100%
rename from docs/cookbook/simtom.md
rename to docs_old/cookbook/simtom.md
diff --git a/docs/cookbook/structured_generation_workflow.md b/docs_old/cookbook/structured_generation_workflow.md
similarity index 100%
rename from docs/cookbook/structured_generation_workflow.md
rename to docs_old/cookbook/structured_generation_workflow.md
diff --git a/docs_old/index.md b/docs_old/index.md
new file mode 100644
index 000000000..3692d32c9
--- /dev/null
+++ b/docs_old/index.md
@@ -0,0 +1,8 @@
+---
+title: Outlines
+template: home.html # Note that this is managed in overrides/home.html
+hide:
+  - navigation
+  - toc
+  - feedback
+---
diff --git a/docs/installation.md b/docs_old/installation.md
similarity index 100%
rename from docs/installation.md
rename to docs_old/installation.md
diff --git a/docs_new/licence.md b/docs_old/licence.md
similarity index 100%
rename from docs_new/licence.md
rename to docs_old/licence.md
diff --git a/docs_new/logos/amazon.png b/docs_old/logos/amazon.png
similarity index 100%
rename from docs_new/logos/amazon.png
rename to docs_old/logos/amazon.png
diff --git a/docs_new/logos/apple.png b/docs_old/logos/apple.png
similarity index 100%
rename from docs_new/logos/apple.png
rename to docs_old/logos/apple.png
diff --git a/docs_new/logos/best_buy.png b/docs_old/logos/best_buy.png
similarity index 100%
rename from docs_new/logos/best_buy.png
rename to docs_old/logos/best_buy.png
diff --git a/docs_new/logos/canoe.png b/docs_old/logos/canoe.png
similarity index 100%
rename from docs_new/logos/canoe.png
rename to docs_old/logos/canoe.png
diff --git a/docs_new/logos/cisco.png b/docs_old/logos/cisco.png
similarity index 100%
rename from docs_new/logos/cisco.png
rename to docs_old/logos/cisco.png
diff --git a/docs_new/logos/cohere.png b/docs_old/logos/cohere.png
similarity index 100%
rename from docs_new/logos/cohere.png
rename to docs_old/logos/cohere.png
diff --git a/docs_new/logos/dassault_systems.png b/docs_old/logos/dassault_systems.png
similarity index 100%
rename from docs_new/logos/dassault_systems.png
rename to docs_old/logos/dassault_systems.png
diff --git a/docs_new/logos/databricks.png b/docs_old/logos/databricks.png
similarity index 100%
rename from docs_new/logos/databricks.png
rename to docs_old/logos/databricks.png
diff --git a/docs_new/logos/datadog.png b/docs_old/logos/datadog.png
similarity index 100%
rename from docs_new/logos/datadog.png
rename to docs_old/logos/datadog.png
diff --git a/docs_new/logos/dbt_labs.png b/docs_old/logos/dbt_labs.png
similarity index 100%
rename from docs_new/logos/dbt_labs.png
rename to docs_old/logos/dbt_labs.png
diff --git a/docs_new/logos/gladia.jpg b/docs_old/logos/gladia.jpg
similarity index 100%
rename from docs_new/logos/gladia.jpg
rename to docs_old/logos/gladia.jpg
diff --git a/docs_new/logos/harvard.png b/docs_old/logos/harvard.png
similarity index 100%
rename from docs_new/logos/harvard.png
rename to docs_old/logos/harvard.png
diff --git a/docs_new/logos/hf.png b/docs_old/logos/hf.png
similarity index 100%
rename from docs_new/logos/hf.png
rename to docs_old/logos/hf.png
diff --git a/docs_new/logos/johns_hopkins.png b/docs_old/logos/johns_hopkins.png
similarity index 100%
rename from docs_new/logos/johns_hopkins.png
rename to docs_old/logos/johns_hopkins.png
diff --git a/docs_new/logos/meta.png b/docs_old/logos/meta.png
similarity index 100%
rename from docs_new/logos/meta.png
rename to docs_old/logos/meta.png
diff --git a/docs_new/logos/mit.png b/docs_old/logos/mit.png
similarity index 100%
rename from docs_new/logos/mit.png
rename to docs_old/logos/mit.png
diff --git a/docs_new/logos/mount_sinai.png b/docs_old/logos/mount_sinai.png
similarity index 100%
rename from docs_new/logos/mount_sinai.png
rename to docs_old/logos/mount_sinai.png
diff --git a/docs_new/logos/nvidia.png b/docs_old/logos/nvidia.png
similarity index 100%
rename from docs_new/logos/nvidia.png
rename to docs_old/logos/nvidia.png
diff --git a/docs_new/logos/nyu.png b/docs_old/logos/nyu.png
similarity index 100%
rename from docs_new/logos/nyu.png
rename to docs_old/logos/nyu.png
diff --git a/docs_new/logos/safran.png b/docs_old/logos/safran.png
similarity index 100%
rename from docs_new/logos/safran.png
rename to docs_old/logos/safran.png
diff --git a/docs_new/logos/salesforce.png b/docs_old/logos/salesforce.png
similarity index 100%
rename from docs_new/logos/salesforce.png
rename to docs_old/logos/salesforce.png
diff --git a/docs_new/logos/shopify.png b/docs_old/logos/shopify.png
similarity index 100%
rename from docs_new/logos/shopify.png
rename to docs_old/logos/shopify.png
diff --git a/docs_new/logos/smithsonian.png b/docs_old/logos/smithsonian.png
similarity index 100%
rename from docs_new/logos/smithsonian.png
rename to docs_old/logos/smithsonian.png
diff --git a/docs_new/logos/tinder.png b/docs_old/logos/tinder.png
similarity index 100%
rename from docs_new/logos/tinder.png
rename to docs_old/logos/tinder.png
diff --git a/docs_new/logos/upenn.png b/docs_old/logos/upenn.png
similarity index 100%
rename from docs_new/logos/upenn.png
rename to docs_old/logos/upenn.png
diff --git a/docs_new/logos/vllm.png b/docs_old/logos/vllm.png
similarity index 100%
rename from docs_new/logos/vllm.png
rename to docs_old/logos/vllm.png
diff --git a/docs_new/overrides/home.html b/docs_old/overrides/home.html
similarity index 94%
rename from docs_new/overrides/home.html
rename to docs_old/overrides/home.html
index 9e7fdaf0b..de5b2258a 100644
--- a/docs_new/overrides/home.html
+++ b/docs_old/overrides/home.html
@@ -119,8 +119,7 @@
     <div class="md-grid md-typeset">
         <div class="mdx-hero">
             <div class="mdx-hero__image">
-                <img src="assets/images/logo-light-mode.svg#only-light" alt="Outlines Logo" width="600" draggable="false">
-                <img src="assets/images/logo-dark-mode.svg#only-dark" alt="Outlines Logo" width="600" draggable="false">
+                <img src="assets/images/logo.png" alt="Outlines Logo" width="600" draggable="false">
             </div>
             <div class="mdx-hero__content">
                 <h2 class="subtitle" style="font-weight: 500; padding-top: 1rem;">
diff --git a/docs_new/overrides/main.html b/docs_old/overrides/main.html
similarity index 100%
rename from docs_new/overrides/main.html
rename to docs_old/overrides/main.html
diff --git a/docs/quickstart.md b/docs_old/quickstart.md
similarity index 100%
rename from docs/quickstart.md
rename to docs_old/quickstart.md
diff --git a/docs/reference/applications.md b/docs_old/reference/applications.md
similarity index 100%
rename from docs/reference/applications.md
rename to docs_old/reference/applications.md
diff --git a/docs_new/guide/chat_templating.md b/docs_old/reference/chat_templating.md
similarity index 100%
rename from docs_new/guide/chat_templating.md
rename to docs_old/reference/chat_templating.md
diff --git a/docs/reference/generation/creating_grammars.md b/docs_old/reference/generation/creating_grammars.md
similarity index 100%
rename from docs/reference/generation/creating_grammars.md
rename to docs_old/reference/generation/creating_grammars.md
diff --git a/docs/reference/generation/custom_fsm_ops.md b/docs_old/reference/generation/custom_fsm_ops.md
similarity index 100%
rename from docs/reference/generation/custom_fsm_ops.md
rename to docs_old/reference/generation/custom_fsm_ops.md
diff --git a/docs/reference/generation/generation.md b/docs_old/reference/generation/generation.md
similarity index 100%
rename from docs/reference/generation/generation.md
rename to docs_old/reference/generation/generation.md
diff --git a/docs/reference/generation/structured_generation_explanation.md b/docs_old/reference/generation/structured_generation_explanation.md
similarity index 100%
rename from docs/reference/generation/structured_generation_explanation.md
rename to docs_old/reference/generation/structured_generation_explanation.md
diff --git a/docs/reference/generation/types.md b/docs_old/reference/generation/types.md
similarity index 100%
rename from docs/reference/generation/types.md
rename to docs_old/reference/generation/types.md
diff --git a/docs/reference/index.md b/docs_old/reference/index.md
similarity index 100%
rename from docs/reference/index.md
rename to docs_old/reference/index.md
diff --git a/docs/reference/models/anthropic.md b/docs_old/reference/models/anthropic.md
similarity index 100%
rename from docs/reference/models/anthropic.md
rename to docs_old/reference/models/anthropic.md
diff --git a/docs/reference/models/dottxt.md b/docs_old/reference/models/dottxt.md
similarity index 100%
rename from docs/reference/models/dottxt.md
rename to docs_old/reference/models/dottxt.md
diff --git a/docs/reference/models/gemini.md b/docs_old/reference/models/gemini.md
similarity index 100%
rename from docs/reference/models/gemini.md
rename to docs_old/reference/models/gemini.md
diff --git a/docs/reference/models/llamacpp.md b/docs_old/reference/models/llamacpp.md
similarity index 100%
rename from docs/reference/models/llamacpp.md
rename to docs_old/reference/models/llamacpp.md
diff --git a/docs/reference/models/mlxlm.md b/docs_old/reference/models/mlxlm.md
similarity index 100%
rename from docs/reference/models/mlxlm.md
rename to docs_old/reference/models/mlxlm.md
diff --git a/docs/reference/models/models.md b/docs_old/reference/models/models.md
similarity index 100%
rename from docs/reference/models/models.md
rename to docs_old/reference/models/models.md
diff --git a/docs/reference/models/ollama.md b/docs_old/reference/models/ollama.md
similarity index 100%
rename from docs/reference/models/ollama.md
rename to docs_old/reference/models/ollama.md
diff --git a/docs/reference/models/openai.md b/docs_old/reference/models/openai.md
similarity index 100%
rename from docs/reference/models/openai.md
rename to docs_old/reference/models/openai.md
diff --git a/docs/reference/models/sglang.md b/docs_old/reference/models/sglang.md
similarity index 100%
rename from docs/reference/models/sglang.md
rename to docs_old/reference/models/sglang.md
diff --git a/docs/reference/models/tgi.md b/docs_old/reference/models/tgi.md
similarity index 100%
rename from docs/reference/models/tgi.md
rename to docs_old/reference/models/tgi.md
diff --git a/docs/reference/models/transformers.md b/docs_old/reference/models/transformers.md
similarity index 100%
rename from docs/reference/models/transformers.md
rename to docs_old/reference/models/transformers.md
diff --git a/docs/reference/models/transformers_multimodal.md b/docs_old/reference/models/transformers_multimodal.md
similarity index 100%
rename from docs/reference/models/transformers_multimodal.md
rename to docs_old/reference/models/transformers_multimodal.md
diff --git a/docs/reference/models/vllm.md b/docs_old/reference/models/vllm.md
similarity index 100%
rename from docs/reference/models/vllm.md
rename to docs_old/reference/models/vllm.md
diff --git a/docs/reference/models/vllm_offline.md b/docs_old/reference/models/vllm_offline.md
similarity index 100%
rename from docs/reference/models/vllm_offline.md
rename to docs_old/reference/models/vllm_offline.md
diff --git a/docs/reference/prompting.md b/docs_old/reference/prompting.md
similarity index 100%
rename from docs/reference/prompting.md
rename to docs_old/reference/prompting.md
diff --git a/docs/reference/regex_dsl.md b/docs_old/reference/regex_dsl.md
similarity index 100%
rename from docs/reference/regex_dsl.md
rename to docs_old/reference/regex_dsl.md
diff --git a/docs/reference/serve/lmstudio.md b/docs_old/reference/serve/lmstudio.md
similarity index 100%
rename from docs/reference/serve/lmstudio.md
rename to docs_old/reference/serve/lmstudio.md
diff --git a/docs/reference/serve/vllm.md b/docs_old/reference/serve/vllm.md
similarity index 100%
rename from docs/reference/serve/vllm.md
rename to docs_old/reference/serve/vllm.md
diff --git a/docs_old/stylesheets/extra.css b/docs_old/stylesheets/extra.css
new file mode 100644
index 000000000..e71a6124f
--- /dev/null
+++ b/docs_old/stylesheets/extra.css
@@ -0,0 +1,149 @@
+@font-face {
+  font-family: 'Source Code Pro Custom', monospace;
+  src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
+}
+
+:root > * {
+  --md-default-bg-color: #FFFFFF;
+  --md-code-bg-color: #2E3440;
+  --md-code-fg-color: #FFFFFF;
+  --md-text-font-family: "Inter";
+  --md-code-font: "Source Code Pro Custom";
+  --md-typeset-a-color: #d16626; /*this is the brand color*/
+
+  /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
+  --md-mermaid-label-fg-color: #000000;
+  --md-mermaid-edge-color: #000000;
+}
+
+.index-pre-code {
+  max-width: 700px;
+  left: 50%;
+}
+
+.index-pre-code pre>code {
+  text-align: left;
+}
+
+.md-clipboard::after {
+  color: #FFFFFF;
+  transition: color 0.3s ease-in-out;
+}
+
+.md-clipboard:hover::after {
+  color: #D8DEE9;
+}
+
+.md-source-file {
+  text-align: center;
+  padding: 24px 0;
+}
+
+.md-typeset pre>code {
+  border-radius: .2rem;
+  box-shadow: 10px 5px 5px #D8DEE9;
+}
+
+.md-typeset p > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset strong > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-content p > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset td > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset li > code {
+  background: #ECEFF4;
+  color: #000000;
+  font-weight: 500;
+}
+
+.md-typeset code {
+  font-weight: 500;
+}
+
+.md-typeset pre {
+  margin-left: .5rem;
+  margin-right: .5rem;
+  margin-top: 2rem;
+  margin-bottom: 2rem;
+}
+
+.language-python {
+  background: #FFFFFF ! important
+}
+
+.language-bash {
+  background: #FFFFFF ! important
+}
+
+.language-toml {
+  background: #FFFFFF ! important
+}
+
+.language-text {
+  background: #FFFFFF ! important
+}
+
+.language-json {
+  background: #FFFFFF ! important
+}
+
+h1.title {
+  color: #FFFFFF;
+  margin: 0px 0px 5px;
+}
+
+h2.subtitle {
+  margin: 5px 0px 25px;
+  font-size: 1rem;
+  max-width: 540px;
+  margin: 0 auto;
+}
+
+.md-typeset {
+  line-height: 24px;
+  font-weight: 400;
+}
+
+.md-typeset h1 {
+  font-weight: bold;
+  color: #000000;
+}
+
+.md-typeset h2 {
+  font-weight: bold;
+  color: #000000;
+}
+
+span.md-ellipsis {
+  color: black;
+}
+
+.md-nav__link--active {
+  background-color: #ECEFF4;
+}
+
+.md-typeset ol li {
+  margin-bottom: .1rem;
+}
+
+.md-typeset ul li {
+  margin-bottom: .1rem;
+}
diff --git a/docs/welcome.md b/docs_old/welcome.md
similarity index 100%
rename from docs/welcome.md
rename to docs_old/welcome.md
diff --git a/mkdocs.yml b/mkdocs.yml
index 13ba556e1..3ff548000 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -12,7 +12,7 @@ repo_url: https://github.com/dottxt-ai/outlines
 copyright: Copyright &copy; 2023- The Outlines Developers
 
 # Documentation directory
-docs_dir: docs_new
+docs_dir: docs
 
 # Configuration
 theme:
@@ -153,12 +153,7 @@ plugins:
         nav_file: SUMMARY.md
 
 nav:
-    - Home:
-          - index.md
-          - User Guide: guide/index.md
-          - Features: features/index.md
-          - API Reference: api_reference/index.md
-          - Examples: examples/index.md
+    - Home: index.md
 
     - Guide:
           - Getting Started: guide/getting_started.md

From 96fbfa66e505b63e807e7ca934c21439c2e8aaaf Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Sun, 8 Jun 2025 09:43:07 +0200
Subject: [PATCH 345/505] Use the examples from old_docs in docs and delete
 old_docs

---
 docs/examples/atomic_caption.md               |  53 +--
 docs/examples/chain_of_density.md             |  51 +--
 docs/examples/chain_of_thought.md             |  87 ++--
 docs/examples/classification.md               |  61 +--
 docs/examples/dating_profiles.md              |  30 +-
 docs/examples/deploy-using-bentoml.md         |  22 +-
 docs/examples/deploy-using-cerebrium.md       |  11 +-
 docs/examples/deploy-using-modal.md           |  16 +-
 docs/examples/earnings-reports.md             |  24 +-
 docs/examples/extract_event_details.md        |   2 +-
 .../examples}/extract_event_details.py        |   0
 docs/examples/extraction.md                   |  16 +-
 docs/examples/knowledge_graph_extraction.md   | 102 +++--
 docs/examples/models_playing_chess.md         |  28 +-
 .../prompt_templates/chain_of_density.txt     |   0
 .../prompt_templates/classification.txt       |   0
 .../prompt_templates/react_agent.txt          |   0
 .../simtom_prospective_taking.txt             |   0
 .../prompt_templates/simtom_simulation.txt    |   0
 docs/examples/qa-with-citations.md            | 125 +++---
 docs/examples/react_agent.md                  |  90 ++--
 docs/examples/read-pdfs.md                    |  79 ++--
 docs/examples/receipt-digitization.md         |  99 ++---
 docs/examples/simtom.md                       |  52 +--
 .../structured_generation_workflow.md         |  62 ++-
 docs_old/api/applications.md                  |   1 -
 docs_old/api/core/application.md              |  76 ----
 docs_old/api/guide.md                         |   1 -
 docs_old/api/index.md                         |   1 -
 docs_old/api/json_schema.md                   |   1 -
 docs_old/api/models.md                        |   1 -
 docs_old/api/parsing.md                       |   1 -
 docs_old/api/regex.md                         |   1 -
 docs_old/api/templates.md                     |   1 -
 docs_old/assets/images/dottxt-dark.svg        |   1 -
 docs_old/assets/images/dottxt-light.svg       |  32 --
 docs_old/assets/images/dottxt.png             | Bin 42419 -> 0 bytes
 docs_old/assets/images/install.png            | Bin 12669 -> 0 bytes
 .../images/logits_processing_diagram.svg      | 157 -------
 docs_old/assets/images/logo-dark-mode.png     | Bin 7425 -> 0 bytes
 docs_old/assets/images/logo-dark-mode.svg     | 207 ---------
 docs_old/assets/images/logo-light-mode.png    | Bin 9174 -> 0 bytes
 docs_old/assets/images/logo-light-mode.svg    |   1 -
 docs_old/assets/images/logo-simple.png        | Bin 2196 -> 0 bytes
 docs_old/assets/images/logo-square.png        | Bin 2233 -> 0 bytes
 docs_old/assets/images/logo-square.svg        | 124 ------
 docs_old/assets/images/logo.png               | Bin 6386 -> 0 bytes
 docs_old/assets/images/logo.svg               | 134 ------
 docs_old/assets/images/normal_computing.jpg   | Bin 64918 -> 0 bytes
 docs_old/assets/images/readme-dark.png        | Bin 38912 -> 0 bytes
 docs_old/assets/images/readme-light.png       | Bin 38943 -> 0 bytes
 docs_old/assets/images/use_philosophy.png     | Bin 16463 -> 0 bytes
 docs_old/blog/.authors.yml                    |   5 -
 docs_old/blog/assets/4000_stars.png           | Bin 57307 -> 0 bytes
 docs_old/blog/index.md                        |   1 -
 docs_old/blog/posts/roadmap-2024.md           |  91 ----
 docs_old/community/belonging.png              | Bin 142520 -> 0 bytes
 docs_old/community/contribute.md              | 181 --------
 docs_old/community/examples.md                |  19 -
 docs_old/community/feedback.md                |  81 ----
 docs_old/community/index.md                   |  12 -
 docs_old/community/versioning.md              |  26 --
 .../content/docs/guides/migration_guide.md    | 244 -----------
 docs_old/cookbook/atomic_caption.md           | 193 ---------
 docs_old/cookbook/chain_of_density.md         | 103 -----
 docs_old/cookbook/chain_of_thought.md         | 127 ------
 docs_old/cookbook/classification.md           |  85 ----
 docs_old/cookbook/dating_profiles.md          | 232 ----------
 docs_old/cookbook/deploy-using-bentoml.md     | 232 ----------
 docs_old/cookbook/deploy-using-cerebrium.md   | 126 ------
 docs_old/cookbook/deploy-using-modal.md       | 206 ---------
 docs_old/cookbook/earnings-reports.md         | 276 ------------
 docs_old/cookbook/extract_event_details.md    |  34 --
 docs_old/cookbook/extraction.md               |  94 -----
 docs_old/cookbook/images/chain_of_density.png | Bin 515603 -> 0 bytes
 .../images/coding_structure_diagram.png       | Bin 73586 -> 0 bytes
 .../images/knowledge-graph-extraction.png     | Bin 33127 -> 0 bytes
 docs_old/cookbook/images/nvidia-income.png    | Bin 185563 -> 0 bytes
 docs_old/cookbook/images/simtom.png           | Bin 109657 -> 0 bytes
 .../cookbook/images/trader-joes-receipt.jpg   | Bin 430638 -> 0 bytes
 docs_old/cookbook/index.md                    |  23 -
 .../cookbook/knowledge_graph_extraction.md    | 153 -------
 docs_old/cookbook/models_playing_chess.md     |  90 ----
 docs_old/cookbook/qa-with-citations.md        | 256 ------------
 docs_old/cookbook/react_agent.md              | 233 -----------
 docs_old/cookbook/read-pdfs.md                | 367 ----------------
 docs_old/cookbook/receipt-digitization.md     | 287 -------------
 docs_old/cookbook/simtom.md                   | 111 -----
 .../structured_generation_workflow.md         | 211 ----------
 docs_old/index.md                             |   8 -
 docs_old/installation.md                      |  48 ---
 docs_old/licence.md                           |  34 --
 docs_old/logos/amazon.png                     | Bin 19489 -> 0 bytes
 docs_old/logos/apple.png                      | Bin 21938 -> 0 bytes
 docs_old/logos/best_buy.png                   | Bin 22328 -> 0 bytes
 docs_old/logos/canoe.png                      | Bin 3719 -> 0 bytes
 docs_old/logos/cisco.png                      | Bin 5784 -> 0 bytes
 docs_old/logos/cohere.png                     | Bin 113536 -> 0 bytes
 docs_old/logos/dassault_systems.png           | Bin 21865 -> 0 bytes
 docs_old/logos/databricks.png                 | Bin 59004 -> 0 bytes
 docs_old/logos/datadog.png                    | Bin 32779 -> 0 bytes
 docs_old/logos/dbt_labs.png                   | Bin 13418 -> 0 bytes
 docs_old/logos/gladia.jpg                     | Bin 11198 -> 0 bytes
 docs_old/logos/harvard.png                    | Bin 31910 -> 0 bytes
 docs_old/logos/hf.png                         | Bin 91529 -> 0 bytes
 docs_old/logos/johns_hopkins.png              | Bin 76335 -> 0 bytes
 docs_old/logos/meta.png                       | Bin 20763 -> 0 bytes
 docs_old/logos/mit.png                        | Bin 1483 -> 0 bytes
 docs_old/logos/mount_sinai.png                | Bin 23749 -> 0 bytes
 docs_old/logos/nvidia.png                     | Bin 15146 -> 0 bytes
 docs_old/logos/nyu.png                        | Bin 30682 -> 0 bytes
 docs_old/logos/safran.png                     | Bin 24435 -> 0 bytes
 docs_old/logos/salesforce.png                 | Bin 20712 -> 0 bytes
 docs_old/logos/shopify.png                    | Bin 8003 -> 0 bytes
 docs_old/logos/smithsonian.png                | Bin 35574 -> 0 bytes
 docs_old/logos/tinder.png                     | Bin 28182 -> 0 bytes
 docs_old/logos/upenn.png                      | Bin 97063 -> 0 bytes
 docs_old/logos/vllm.png                       | Bin 89976 -> 0 bytes
 docs_old/overrides/home.html                  | 154 -------
 docs_old/overrides/main.html                  |   1 -
 docs_old/quickstart.md                        | 228 ----------
 docs_old/reference/applications.md            |  40 --
 docs_old/reference/chat_templating.md         |  38 --
 .../reference/generation/creating_grammars.md |  99 -----
 .../reference/generation/custom_fsm_ops.md    |  41 --
 docs_old/reference/generation/generation.md   | 195 ---------
 .../structured_generation_explanation.md      |  54 ---
 docs_old/reference/generation/types.md        |  69 ---
 docs_old/reference/index.md                   |  15 -
 docs_old/reference/models/anthropic.md        |  44 --
 docs_old/reference/models/dottxt.md           |  51 ---
 docs_old/reference/models/gemini.md           |  96 -----
 docs_old/reference/models/llamacpp.md         |  88 ----
 docs_old/reference/models/mlxlm.md            |  70 ----
 docs_old/reference/models/models.md           |  50 ---
 docs_old/reference/models/ollama.md           |  74 ----
 docs_old/reference/models/openai.md           | 110 -----
 docs_old/reference/models/sglang.md           |  51 ---
 docs_old/reference/models/tgi.md              |  50 ---
 docs_old/reference/models/transformers.md     |  68 ---
 .../models/transformers_multimodal.md         | 203 ---------
 docs_old/reference/models/vllm.md             |  51 ---
 docs_old/reference/models/vllm_offline.md     |  68 ---
 docs_old/reference/prompting.md               | 395 ------------------
 docs_old/reference/regex_dsl.md               | 344 ---------------
 docs_old/reference/serve/lmstudio.md          |  89 ----
 docs_old/reference/serve/vllm.md              |  73 ----
 docs_old/stylesheets/extra.css                | 149 -------
 docs_old/welcome.md                           | 183 --------
 outlines/release_note.md                      | 392 +++++++++++++++++
 150 files changed, 846 insertions(+), 8726 deletions(-)
 rename {docs_old/cookbook => docs/examples}/extract_event_details.py (100%)
 rename {docs_old/cookbook => docs/examples}/prompt_templates/chain_of_density.txt (100%)
 rename {docs_old/cookbook => docs/examples}/prompt_templates/classification.txt (100%)
 rename {docs_old/cookbook => docs/examples}/prompt_templates/react_agent.txt (100%)
 rename {docs_old/cookbook => docs/examples}/prompt_templates/simtom_prospective_taking.txt (100%)
 rename {docs_old/cookbook => docs/examples}/prompt_templates/simtom_simulation.txt (100%)
 delete mode 100644 docs_old/api/applications.md
 delete mode 100644 docs_old/api/core/application.md
 delete mode 100644 docs_old/api/guide.md
 delete mode 100644 docs_old/api/index.md
 delete mode 100644 docs_old/api/json_schema.md
 delete mode 100644 docs_old/api/models.md
 delete mode 100644 docs_old/api/parsing.md
 delete mode 100644 docs_old/api/regex.md
 delete mode 100644 docs_old/api/templates.md
 delete mode 100644 docs_old/assets/images/dottxt-dark.svg
 delete mode 100644 docs_old/assets/images/dottxt-light.svg
 delete mode 100644 docs_old/assets/images/dottxt.png
 delete mode 100644 docs_old/assets/images/install.png
 delete mode 100644 docs_old/assets/images/logits_processing_diagram.svg
 delete mode 100644 docs_old/assets/images/logo-dark-mode.png
 delete mode 100644 docs_old/assets/images/logo-dark-mode.svg
 delete mode 100644 docs_old/assets/images/logo-light-mode.png
 delete mode 100644 docs_old/assets/images/logo-light-mode.svg
 delete mode 100644 docs_old/assets/images/logo-simple.png
 delete mode 100644 docs_old/assets/images/logo-square.png
 delete mode 100644 docs_old/assets/images/logo-square.svg
 delete mode 100644 docs_old/assets/images/logo.png
 delete mode 100644 docs_old/assets/images/logo.svg
 delete mode 100644 docs_old/assets/images/normal_computing.jpg
 delete mode 100644 docs_old/assets/images/readme-dark.png
 delete mode 100644 docs_old/assets/images/readme-light.png
 delete mode 100644 docs_old/assets/images/use_philosophy.png
 delete mode 100644 docs_old/blog/.authors.yml
 delete mode 100644 docs_old/blog/assets/4000_stars.png
 delete mode 100644 docs_old/blog/index.md
 delete mode 100644 docs_old/blog/posts/roadmap-2024.md
 delete mode 100644 docs_old/community/belonging.png
 delete mode 100644 docs_old/community/contribute.md
 delete mode 100644 docs_old/community/examples.md
 delete mode 100644 docs_old/community/feedback.md
 delete mode 100644 docs_old/community/index.md
 delete mode 100644 docs_old/community/versioning.md
 delete mode 100644 docs_old/content/docs/guides/migration_guide.md
 delete mode 100644 docs_old/cookbook/atomic_caption.md
 delete mode 100644 docs_old/cookbook/chain_of_density.md
 delete mode 100644 docs_old/cookbook/chain_of_thought.md
 delete mode 100644 docs_old/cookbook/classification.md
 delete mode 100644 docs_old/cookbook/dating_profiles.md
 delete mode 100644 docs_old/cookbook/deploy-using-bentoml.md
 delete mode 100644 docs_old/cookbook/deploy-using-cerebrium.md
 delete mode 100644 docs_old/cookbook/deploy-using-modal.md
 delete mode 100644 docs_old/cookbook/earnings-reports.md
 delete mode 100644 docs_old/cookbook/extract_event_details.md
 delete mode 100644 docs_old/cookbook/extraction.md
 delete mode 100644 docs_old/cookbook/images/chain_of_density.png
 delete mode 100644 docs_old/cookbook/images/coding_structure_diagram.png
 delete mode 100644 docs_old/cookbook/images/knowledge-graph-extraction.png
 delete mode 100644 docs_old/cookbook/images/nvidia-income.png
 delete mode 100644 docs_old/cookbook/images/simtom.png
 delete mode 100644 docs_old/cookbook/images/trader-joes-receipt.jpg
 delete mode 100644 docs_old/cookbook/index.md
 delete mode 100644 docs_old/cookbook/knowledge_graph_extraction.md
 delete mode 100644 docs_old/cookbook/models_playing_chess.md
 delete mode 100644 docs_old/cookbook/qa-with-citations.md
 delete mode 100644 docs_old/cookbook/react_agent.md
 delete mode 100644 docs_old/cookbook/read-pdfs.md
 delete mode 100644 docs_old/cookbook/receipt-digitization.md
 delete mode 100644 docs_old/cookbook/simtom.md
 delete mode 100644 docs_old/cookbook/structured_generation_workflow.md
 delete mode 100644 docs_old/index.md
 delete mode 100644 docs_old/installation.md
 delete mode 100644 docs_old/licence.md
 delete mode 100644 docs_old/logos/amazon.png
 delete mode 100644 docs_old/logos/apple.png
 delete mode 100644 docs_old/logos/best_buy.png
 delete mode 100644 docs_old/logos/canoe.png
 delete mode 100644 docs_old/logos/cisco.png
 delete mode 100644 docs_old/logos/cohere.png
 delete mode 100644 docs_old/logos/dassault_systems.png
 delete mode 100644 docs_old/logos/databricks.png
 delete mode 100644 docs_old/logos/datadog.png
 delete mode 100644 docs_old/logos/dbt_labs.png
 delete mode 100644 docs_old/logos/gladia.jpg
 delete mode 100644 docs_old/logos/harvard.png
 delete mode 100644 docs_old/logos/hf.png
 delete mode 100644 docs_old/logos/johns_hopkins.png
 delete mode 100644 docs_old/logos/meta.png
 delete mode 100644 docs_old/logos/mit.png
 delete mode 100644 docs_old/logos/mount_sinai.png
 delete mode 100644 docs_old/logos/nvidia.png
 delete mode 100644 docs_old/logos/nyu.png
 delete mode 100644 docs_old/logos/safran.png
 delete mode 100644 docs_old/logos/salesforce.png
 delete mode 100644 docs_old/logos/shopify.png
 delete mode 100644 docs_old/logos/smithsonian.png
 delete mode 100644 docs_old/logos/tinder.png
 delete mode 100644 docs_old/logos/upenn.png
 delete mode 100644 docs_old/logos/vllm.png
 delete mode 100644 docs_old/overrides/home.html
 delete mode 100644 docs_old/overrides/main.html
 delete mode 100644 docs_old/quickstart.md
 delete mode 100644 docs_old/reference/applications.md
 delete mode 100644 docs_old/reference/chat_templating.md
 delete mode 100644 docs_old/reference/generation/creating_grammars.md
 delete mode 100644 docs_old/reference/generation/custom_fsm_ops.md
 delete mode 100644 docs_old/reference/generation/generation.md
 delete mode 100644 docs_old/reference/generation/structured_generation_explanation.md
 delete mode 100644 docs_old/reference/generation/types.md
 delete mode 100644 docs_old/reference/index.md
 delete mode 100644 docs_old/reference/models/anthropic.md
 delete mode 100644 docs_old/reference/models/dottxt.md
 delete mode 100644 docs_old/reference/models/gemini.md
 delete mode 100644 docs_old/reference/models/llamacpp.md
 delete mode 100644 docs_old/reference/models/mlxlm.md
 delete mode 100644 docs_old/reference/models/models.md
 delete mode 100644 docs_old/reference/models/ollama.md
 delete mode 100644 docs_old/reference/models/openai.md
 delete mode 100644 docs_old/reference/models/sglang.md
 delete mode 100644 docs_old/reference/models/tgi.md
 delete mode 100644 docs_old/reference/models/transformers.md
 delete mode 100644 docs_old/reference/models/transformers_multimodal.md
 delete mode 100644 docs_old/reference/models/vllm.md
 delete mode 100644 docs_old/reference/models/vllm_offline.md
 delete mode 100644 docs_old/reference/prompting.md
 delete mode 100644 docs_old/reference/regex_dsl.md
 delete mode 100644 docs_old/reference/serve/lmstudio.md
 delete mode 100644 docs_old/reference/serve/vllm.md
 delete mode 100644 docs_old/stylesheets/extra.css
 delete mode 100644 docs_old/welcome.md
 create mode 100644 outlines/release_note.md

diff --git a/docs/examples/atomic_caption.md b/docs/examples/atomic_caption.md
index a2b43e5e2..08782594f 100644
--- a/docs/examples/atomic_caption.md
+++ b/docs/examples/atomic_caption.md
@@ -1,5 +1,6 @@
 # Vision-Language Models with Outlines
-This guide demonstrates how to use Outlines with vision-language models, leveraging the new transformers_vision module. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
+
+This guide demonstrates how to use Outlines with vision-language models, leveraging the outlines `TransformersMultiModal` model. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
 
 We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
 
@@ -11,17 +12,21 @@ pip install outlines transformers torch
 ```
 
 ### Initializing the Model
-We'll use the transformers_vision function to initialize our vision-language model. This function is specifically designed to handle models that can process both text and image inputs. Today we'll be using the Pixtral model with the llama tokenizer. (Currently the mistral tokenizer is pending support).
+We'll use the `outlines.from_transformers` function to initialize our vision-language model. For this function to return a vision multi-modal model we need to pass in a transformers model and a transformers processor that can handle both text and image inputs. Today we'll be using the Pixtral model with the AutoProcessor.
 
 ```python
+import outlines
 import torch
 from transformers import (
-    LlavaForConditionalGeneration,
+    AutoProcessor,
+    LlavaForConditionalGeneration
 )
+
 model_name="mistral-community/pixtral-12b" # original magnet model is able to be loaded without issue
 model_class=LlavaForConditionalGeneration
+processor_class=AutoProcessor
 
-def get_vision_model(model_name: str, model_class: VisionModel):
+def get_vision_model(model_name: str, model_class, processor_class):
     model_kwargs = {
         "torch_dtype": torch.bfloat16,
         "attn_implementation": "flash_attention_2",
@@ -31,28 +36,25 @@ def get_vision_model(model_name: str, model_class: VisionModel):
         "device": "cuda",
     }
 
-    model = outlines.models.transformers_vision(
-        model.model_name,
-        model_class=model.model_class,
-        model_kwargs=model_kwargs,
-        processor_kwargs=processor_kwargs,
+    model = outlines.from_transformers(
+        model_class.from_pretrained(model_name, **model_kwargs),
+        processor_class.from_pretrained(model_name, **processor_kwargs),
     )
     return model
-model = get_vision_model(model_name, model_class)
+model = get_vision_model(model_name, model_class, processor_class)
 ```
 
 ### Defining the Schema
-Next, we'll define a schema for the output we expect from our vision-language model. This schema will help structure the model's responses.
-
+Next, we'll define a schema for the output we expect from our vision multi-modal model. This schema will help structure the model's responses. We use the `outlines.Generator` object to create a generator for our schema that will then be called with our prompt and images.
 
 ```python
+from enum import Enum
 from pydantic import BaseModel, Field, confloat, constr
 from pydantic.types import StringConstraints, PositiveFloat
 from typing import List
 from typing_extensions import Annotated
 
-from enum import StrEnum
-class TagType(StrEnum):
+class TagType(Enum):
     ENTITY = "Entity"
     RELATIONSHIP = "Relationship"
     STYLE = "Style"
@@ -86,7 +88,7 @@ class ImageData(BaseModel):
     short_caption: Annotated[str, StringConstraints(min_length=10, max_length=150)]
     dense_caption: Annotated[str, StringConstraints(min_length=100, max_length=2048)]
 
-image_data_generator = outlines.generate.json(model, ImageData)
+image_data_generator = outlines.Generator(model, ImageData)
 ```
 
 This schema defines the structure for image tags, including categories like Entity, Relationship, Style, etc., as well as short and dense captions.
@@ -95,7 +97,6 @@ This schema defines the structure for image tags, including categories like Enti
 
 We'll create a prompt that instructs the model on how to analyze the image and generate the structured output:
 
-
 ```python
 pixtral_instruction = """
 <s>[INST]
@@ -131,30 +132,34 @@ pixtral_instruction = """
 </Examples>
 </TagCategories>
 <ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
-<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG][/INST]
+<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG]<image>[/INST]
 """.strip()
 ```
 
-This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required.
+This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required. It's essential to include the <image> tag in the prompt at the location where the image will be inserted.
 
 ### Generating Structured Output
 Now we can use our model to generate structured output based on an input image:
 
 ```python
+from io import BytesIO
+from urllib.request import urlopen
+from PIL import Image
+
 def img_from_url(url):
     img_byte_stream = BytesIO(urlopen(url).read())
     return Image.open(img_byte_stream).convert("RGB")
 
 image_url="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
 image= img_from_url(image_url)
-result = image_data_generator(
-    pixtral_instruction,
-    [image]
-)
+result = image_data_generator({
+    "text": pixtral_instruction,
+    "images": image
+})
 print(result)
 ```
 
-This code loads an image from a URL, passes it to our vision-language model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
+This code loads an image from a URL, passes it to our vision multi-modal model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
 
 ```json
 {'tags_list': [{'tag': 'astronaut',
@@ -184,6 +189,6 @@ This code loads an image from a URL, passes it to our vision-language model alon
 ```
 
 ## Conclusion
-The transformers_vision module in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
+The `TransformersMultiModal` model in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
 
 By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
diff --git a/docs/examples/chain_of_density.md b/docs/examples/chain_of_density.md
index 10bda9771..8a822fc2e 100644
--- a/docs/examples/chain_of_density.md
+++ b/docs/examples/chain_of_density.md
@@ -26,42 +26,12 @@ The prompt also asks the model to return a list of JSON objects that contain the
 
 ![Figure 2 in the paper](./images/chain_of_density.png)
 
-We can now implement the prompt provided in the paper:
+We can now implement the prompt provided in the paper. We stored the prompt template in a text file, and we can load it using the `Template` class:
 
 ```python
 from outlines import Template
 
-
-chain_of_density = Template.from_string(
-    """Article: {{ article }}
-
-    You will generate increasingly concise, entity-dense summaries of the above Article.
-
-    Repeat the following 2 steps 5 times.
-
-    Step 1. Identify 1-3 informative Entities ("; " delimited) from the Article which are missing from the previously generated summary.
-    Step 2. Write a new, denser summary of identical length which covers every entity and detail from the previous summary plus the Missing Entities.
-
-    A Missing Entity is:
-    - Relevant: to the main story.
-    - Specific: descriptive yet concise (5 words or fewer).
-    - Novel: not in the previous summary.
-    - Faithful: present in the Article.
-    - Anywhere: located anywhere in the Article.
-
-    Guidelines:
-    - The first summary should be long (4-5 sentences, ~80 words) yet highly non-specific, containing little information beyond the entities marked as missing. Use overly verbose language and fillers (e.g., "this article discusses") to reach ~80 words.
-    - Make every word count: rewrite the previous summary to improve flow and make space for additional entities.
-    - Make space with fusion, compression, and removal of uninformative phrases like "the article discusses".
-    - The summaries should become highly dense and concise yet self-contained, e.g., easily understood without the Article.
-    - Missing entities can appear anywhere in the new summary.
-    - Never drop entities from the previous summary. If space cannot be made, add fewer new entities.
-
-    Remember, use the exact same number of words for each summary.
-
-    Answer in JSON. The JSON should be a a dictionary with key "summaries" that contains a list (length 5) of dictionaries whose keys are "Missing_Entities" and "Denser_Summary".
-    """
-)
+chain_of_density = Template.from_file("prompt_templates/chain_of_density.txt")
 ```
 
 ??? Note
@@ -84,19 +54,26 @@ class Summaries(BaseModel):
     summaries: conlist(Summary, max_length=5, min_length=5)
 ```
 
-We now generate the prompt by passing the article we want to summarize to the template. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use JSON-structured generation to generate the summaries:
+We now generate the prompt by passing the article we want to summarize to the prompt template previously loaded. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use the `Summaries` schema to generate the summaries with structured generation:
 
 ```python
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ")
+import outlines
+import transformers
 
-prompt = chain_of_density(article)
-result = outlines.generate.json(model, Summaries)(prompt)
+MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
+
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+)
+prompt = chain_of_density(article=article)
+result = model(prompt, Summaries, max_new_tokens=2000)
 ```
 
 We can now check the results:
 
 ```python
-print(result.model_dump())
+print(result)
 # {'summaries': [
 #     {
 #       'missing_entities': 'English mathematician, cryptanalyst, philosopher',
diff --git a/docs/examples/chain_of_thought.md b/docs/examples/chain_of_thought.md
index 5ac18af25..42d6c8550 100644
--- a/docs/examples/chain_of_thought.md
+++ b/docs/examples/chain_of_thought.md
@@ -11,20 +11,23 @@ We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-
 pip install llama-cpp-python
 ```
 
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
+
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -38,20 +41,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Chain of thought
@@ -77,30 +69,27 @@ class Reasoning(BaseModel):
 json_schema = Reasoning.model_json_schema()
 ```
 
-We could generate a response using the json schema but for a change we will use the regex:
-
-```python
-from outlines.fsm.json_schema import convert_json_schema_to_str
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-schema_str = convert_json_schema_to_str(json_schema=json_schema)
-regex_str = build_regex_from_schema(schema_str)
-```
-
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(user_prompt):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON "
-        f"Here's the json schema you must adhere to:\n<schema>\n{json_schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + user_prompt
-        + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+generate_hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ json_schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    {{ user_prompt }}
+    <|im_end|>
+    <|im_start|>assistant
+    <schema>
+    """
+)
 ```
 
 For a given user prompt:
@@ -109,11 +98,11 @@ For a given user prompt:
 user_prompt = "9.11 and 9.9 -- which is bigger?"
 ```
 
-we can use `generate.regex` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+We can use `outlines.Generator` with the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-generator = generate.regex(model, regex_str)
-prompt = generate_hermes_prompt(user_prompt)
+generator = outlines.Generator(model, regex_str)
+prompt = generate_hermes_prompt(json_schema=json_schema, user_prompt=user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
diff --git a/docs/examples/classification.md b/docs/examples/classification.md
index 023e4543a..18a933b47 100644
--- a/docs/examples/classification.md
+++ b/docs/examples/classification.md
@@ -3,52 +3,40 @@
 Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
 
 
-This tutorial shows how one can implement multi-label classification using Outlines. We will use two functionalities of the library: `generate.choice` and `generate.json`.
+This tutorial shows how one can implement multi-label classification using Outlines.
 
 As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
 
 ```python
 import outlines
+import transformers
 
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
+
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+)
 ```
 
-We will use the following prompt template:
+We will use a prompt template stored in a text file:
 
 ```python
 from outlines import Template
 
-
-customer_support = Template.from_string(
-    """You are an experienced customer success manager.
-
-    Given a request from a client, you need to determine when the
-    request is urgent using the label "URGENT" or when it can wait
-    a little with the label "STANDARD".
-
-    # Examples
-
-    Request: "How are you?"
-    Label: STANDARD
-
-    Request: "I need this fixed immediately!"
-    Label: URGENT
-
-    # TASK
-
-    Request: {{ request }}
-    Label: """
-)
+customer_support = Template.from_file("prompt_templates/classification.txt")
 ```
 
 ## Choosing between multiple choices
 
-Outlines provides a shortcut to do multi-label classification, using the `outlines.generate.choice` function to initialize a generator. Outlines uses multinomial sampling by default, here we will use the greedy sampler to get the label with the highest probability:
+Outlines provides a convenient way to do multi-label classification, passing a Literal type hint to the `outlines.Generator` object:
 
 ```python
-from outlines.samplers import greedy
+from typing import Literal
+import outlines
+
+generator = outlines.Generator(model, Literal["URGENT", "STANDARD"])
 
-generator = outlines.generate.choice(model, ["URGENT", "STANDARD"], sampler=greedy())
 ```
 Outlines supports batched requests, so we will pass two requests to the model:
 
@@ -58,10 +46,10 @@ requests = [
     "Just wanted to say hi"
 ]
 
-prompts = [customer_support(request) for request in requests]
+prompts = [customer_support(request=request) for request in requests]
 ```
 
-We can now asks the model to classify the requests:
+We can now ask the model to classify the requests:
 
 ```python
 labels = generator(prompts)
@@ -69,15 +57,6 @@ print(labels)
 # ['URGENT', 'STANDARD']
 ```
 
-Now, you might be in a hurry and don't want to wait until the model finishes completion. After all, you only need to see the first letter of the response to know whether the request is urgent or standard. You can instead stream the response:
-
-```python
-tokens = generator.stream(prompts)
-labels = ["URGENT" if "U" in token else "STANDARD" for token in next(tokens)]
-print(labels)
-# ['URGENT', 'STANDARD']
-```
-
 ## Using JSON-structured generation
 
 Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
@@ -96,11 +75,11 @@ class Classification(BaseModel):
     label: Label
 ```
 
-and we can use `generate.json` by passing this Pydantic model we just defined, and call the generator:
+We can then create a generator with the Pydantic model we just defined and call it:
 
 ```python
-generator = outlines.generate.json(model, Classification, sampler=greedy())
+generator = outlines.Generator(model, Classification)
 labels = generator(prompts)
 print(labels)
-# [Classification(label=<Label.urgent: 'URGENT'>), Classification(label=<Label.standard: 'STANDARD'>)]
+# ['{"label":"URGENT"}', '{ "label": "STANDARD" }']
 ```
diff --git a/docs/examples/dating_profiles.md b/docs/examples/dating_profiles.md
index 1cbea556c..5852e6df9 100644
--- a/docs/examples/dating_profiles.md
+++ b/docs/examples/dating_profiles.md
@@ -3,6 +3,7 @@
 In this example we will see how we can use Outlines to generate synthetic data for a dating application. This example was originally contributed by [Vibhor Kumar](https://github.com/veezbo).
 
 ```python
+import json
 from dataclasses import dataclass
 from enum import Enum
 
@@ -132,20 +133,21 @@ samples: list[Example] = [
 We will use Mosaic's MPT-7B model (requires 13GB of GPU memory) which can fit on a single GPU with a reasonable context window. We initialize it with Outlines:
 
 ```python
+MODEL_NAME = "mosaicml/mpt-7b-8k-instruct"
+
 config = transformers.AutoConfig.from_pretrained(
-    "mosaicml/mpt-7b-8k-instruct", trust_remote_code=True
+    MODEL_NAME, trust_remote_code=True
 )
 config.init_device = "meta"
-model = outlines.models.transformers(
-    model_name="mosaicml/mpt-7b-8k-instruct",
-    device="cuda",
-    model_kwargs={
-        "config": config,
-        "trust_remote_code": True,
-        "torch_dtype": torch.bfloat16,
-        "device_map": {"": 0},
-    },
-)
+model_kwargs = {
+    "config": config,
+    "trust_remote_code": True,
+    "torch_dtype": torch.bfloat16,
+    "device_map": "cuda",
+}
+tf_model = transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME, **model_kwargs)
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+model = outlines.from_transformers(tf_model, tokenizer=tf_tokenizer)
 ```
 
 ## JSON-structured generation of profiles
@@ -162,9 +164,9 @@ every few  months. I also like weddings because I get to wear those suits, and
 it's  a good excuse for a date. I watch the latest series because I'm paying,
 with my hard-earned money, for every streaming service."""
 
-prompt = dating_profile_prompt(new_description, samples)
-profile = outlines.generate.json(model, DatingProfile)(prompt)
-parsed_profile = DatingProfile.model_validate_json(profile)
+prompt = dating_profile_prompt(description=new_description, examples=samples)
+profile = model(prompt, DatingProfile)
+parsed_profile = DatingProfile.model_validate_json(json.loads(profile))
 ```
 
 ## Results
diff --git a/docs/examples/deploy-using-bentoml.md b/docs/examples/deploy-using-bentoml.md
index 2926df0f6..5d000cb52 100644
--- a/docs/examples/deploy-using-bentoml.md
+++ b/docs/examples/deploy-using-bentoml.md
@@ -119,15 +119,22 @@ class Outlines:
     bento_model_ref = bentoml.models.get(BENTO_MODEL_TAG)
 
     def __init__(self) -> None:
-
         import outlines
         import torch
-        self.model = outlines.models.transformers(
+        from transformers import AutoModelForCausalLM, AutoTokenizer
+
+        # Load tokenizer and model from the BentoML model reference path
+        hf_tokenizer = AutoTokenizer.from_pretrained(self.bento_model_ref.path)
+        hf_model = AutoModelForCausalLM.from_pretrained(
             self.bento_model_ref.path,
-            device="cuda",
-            model_kwargs={"torch_dtype": torch.float16},
+            torch_dtype=torch.float16,
+            low_cpu_mem_usage=True,
+            device_map="cuda"
         )
 
+        # Then use the loaded model with Outlines
+        self.model = outlines.from_transformers(hf_model, hf_tokenizer)
+
     ...
 ```
 
@@ -142,13 +149,14 @@ We then need to define an HTTP endpoint using `@bentoml.api` to decorate the met
         prompt: str = "Give me a character description.",
         json_schema: t.Optional[str] = DEFAULT_SCHEMA,
     ) -> t.Dict[str, t.Any]:
-
+        import json
         import outlines
+        from outlines.types import JsonSchema
 
-        generator = outlines.generate.json(self.model, json_schema)
+        generator = outlines.Generator(self.model, JsonSchema(json_schema))
         character = generator(prompt)
 
-        return character
+        return json.loads(character)
 ```
 
 Here `@bentoml.api` decorator defines `generate` as an HTTP endpoint that accepts a JSON request body with two fields: `prompt` and `json_schema` (optional, which allows HTTP clients to provide their own JSON schema). The type hints in the function signature will be used to validate incoming JSON requests. You can define as many HTTP endpoints as you want by using `@bentoml.api` to decorate other methods of `Outlines` class.
diff --git a/docs/examples/deploy-using-cerebrium.md b/docs/examples/deploy-using-cerebrium.md
index 6fbaab390..bf010d583 100644
--- a/docs/examples/deploy-using-cerebrium.md
+++ b/docs/examples/deploy-using-cerebrium.md
@@ -36,7 +36,7 @@ provider = "aws"
 region = "us-east-1"
 
 [cerebrium.dependencies.pip]
-outline = "==0.0.37"
+outline = "==1.0.0"
 transformers = "==4.38.2"
 datasets = "==2.18.0"
 accelerate = "==0.27.2"
@@ -48,9 +48,14 @@ Running code in Cerebrium is like writing normal python with no special syntax.
 
 ```python
 import outlines
+import transformers
+from outlines.types import JsonSchema
 
 
-model = outlines.models.transformers("microsoft/Phi-3-mini-4k-instruct")
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
 
 schema = """{
     "title": "Character",
@@ -89,7 +94,7 @@ schema = """{
     }
 }"""
 
-generator = outlines.generate.json(model, schema)
+generator = outlines.Generator(model, JsonSchema(schema))
 ```
 
 On first deploy, it will download the model and store it on disk therefore for subsequent calls it will load the model from disk.
diff --git a/docs/examples/deploy-using-modal.md b/docs/examples/deploy-using-modal.md
index 15e200cb5..23a846a6d 100644
--- a/docs/examples/deploy-using-modal.md
+++ b/docs/examples/deploy-using-modal.md
@@ -68,7 +68,12 @@ When running longer Modal apps, it's recommended to download your language model
 # downloading models, setting up environment variables, etc.
 def import_model():
     import outlines
-    outlines.models.transformers(language_model)
+    import transformers
+
+    outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(language_model),
+        transformers.AutoTokenizer.from_pretrained(language_model)
+    )
 
 # This line tells the container to run the import_model function when it starts.
 outlines_image = outlines_image.run_function(import_model)
@@ -134,19 +139,22 @@ def generate(
     # so we need to import the necessary libraries here. You should
     # do this with any other libraries you might need.
     import outlines
+    import transformers
+    from outlines.types import JsonSchema
 
     # Load the model into memory. The import_model function above
     # should have already downloaded the model, so this call
     # only loads the model into GPU memory.
-    model = outlines.models.transformers(
-        language_model, device="cuda"
+    outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained(language_model, device_map="cuda"),
+        transformers.AutoTokenizer.from_pretrained(language_model)
     )
 
     # Generate a character description based on the prompt.
     # We use the .json generation method -- we provide the
     # - model: the model we loaded above
     # - schema: the JSON schema we defined above
-    generator = outlines.generate.json(model, schema)
+    generator = outlines.Generator(model, JsonSchema(schema))
 
     # Make sure you wrap your prompt in instruction tags ([INST] and [/INST])
     # to indicate that the prompt is an instruction. Instruction tags can vary
diff --git a/docs/examples/earnings-reports.md b/docs/examples/earnings-reports.md
index 26c2a9ccf..cca1e9513 100644
--- a/docs/examples/earnings-reports.md
+++ b/docs/examples/earnings-reports.md
@@ -26,16 +26,14 @@ Choose your language model. We'll use Phi-3 mini, which is small enough to run o
 ```python
 import outlines
 import torch
+import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.models.transformers(
-    model_name,
-    device='auto',
-    model_kwargs={
-        # To reduce memory usage, we'll use bfloat16
-        "torch_dtype": torch.bfloat16,
-    },
+tf_model = transformers.AutoModelForCausalLM.from_pretrained(
+    model_name, device_map="cuda", torch_dtype=torch.bfloat16
 )
+tf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
+model = outlines.from_transformers(tf_model, tf_tokenizer)
 ```
 
 ## Set up the data
@@ -156,7 +154,7 @@ year,revenue,operating_income,net_income,basic_earnings_per_share(
 \d{4},((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),(-?\d+(?:\.\d{1,2})?)){,3}
 ```
 
-Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.generate.regex` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
+Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.Generator` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
 
 ## Prompting the model
 
@@ -234,12 +232,12 @@ def extract_financial_data_prompt(columns_to_extract, income_statement):
 
 Now that we have our prompt and regular expression, we can run the model.
 
-Construct our regex extractor function. We'll use a greedy sampler, which samples the most likely next token at each step. It's a simple sampler that is more reproducible than multinomial sampling.
+Construct our regex extractor function.
 
 ```python
-csv_extractor = outlines.generate.regex(
-    model, csv_regex, sampler=outlines.samplers.greedy()
-)
+from outlines.types import Regex
+
+csv_extractor = outlines.Generator(model, Regex(csv_regex))
 ```
 
 Provide the prompt to the model and run it:
@@ -247,7 +245,7 @@ Provide the prompt to the model and run it:
 ```python
 csv_data = csv_extractor(
     extract_financial_data_prompt(columns_to_extract, income_statement),
-    max_tokens=1024,
+    max_new_tokens=1024,
 )
 
 print(csv_data)
diff --git a/docs/examples/extract_event_details.md b/docs/examples/extract_event_details.md
index 0f87d9586..063433ad2 100644
--- a/docs/examples/extract_event_details.md
+++ b/docs/examples/extract_event_details.md
@@ -8,7 +8,7 @@ place, 3 avenue des tanneurs, one hour will be enough see you 😘
 ```
 
 Let see how to extract the event details from the message with the MLX
-library dedicated to  Apple Silicon processor  (M series).
+library dedicated to Apple Silicon processor (M series).
 
 ```python
 --8<-- "docs/cookbook/extract_event_details.py"
diff --git a/docs_old/cookbook/extract_event_details.py b/docs/examples/extract_event_details.py
similarity index 100%
rename from docs_old/cookbook/extract_event_details.py
rename to docs/examples/extract_event_details.py
diff --git a/docs/examples/extraction.md b/docs/examples/extraction.md
index dfcca0cfd..cfb39ca5c 100644
--- a/docs/examples/extraction.md
+++ b/docs/examples/extraction.md
@@ -7,9 +7,14 @@ Getting LLMs to output the extracted entities in a structured format can be chal
 As always, we start with initializing the model. We will be using a quantized version of Mistal-7B-v0.1 (we're GPU poor):
 
 ```python
+import transformers
 import outlines
 
-model = outlines.models.transformers("TheBloke/Mistral-7B-OpenOrca-AWQ", device="cuda")
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name, device_map="cuda"),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
 ```
 
 And we will be using the following prompt template:
@@ -17,7 +22,6 @@ And we will be using the following prompt template:
 ```python
 from outlines import Template
 
-
 take_order = Template.from_string(
     """You are the owner of a pizza parlor. Customers \
     send you orders from which you need to extract:
@@ -70,14 +74,14 @@ orders = [
     "Hi! I would like to order two pepperonni pizzas and would like them in 30mins.",
     "Is it possible to get 12 margheritas?"
 ]
-prompts = [take_order(order) for order in orders]
+prompts = [take_order(order=order) for order in orders]
 
-generator = outlines.generate.json(model, Order)
+generator = outlines.Generator(model, Order)
 
 results = generator(prompts)
 print(results)
-# [Order(pizza=<Pizza.pepperonni: 'Pepperoni'>, number=2),
-#  Order(pizza=<Pizza.margherita: 'Margherita'>, number=12)]
+# ['{"pizza": "Pepperoni", "number": 2}',
+# '{"pizza": "Margherita", "number": 12}']
 ```
 
 There are several ways you could improve this example:
diff --git a/docs/examples/knowledge_graph_extraction.md b/docs/examples/knowledge_graph_extraction.md
index 6a6877756..f4b21c84c 100644
--- a/docs/examples/knowledge_graph_extraction.md
+++ b/docs/examples/knowledge_graph_extraction.md
@@ -8,20 +8,23 @@ We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama
 pip install llama-cpp-python
 ```
 
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
+To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
+
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -35,20 +38,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Knowledge Graph Extraction
@@ -92,17 +84,24 @@ schema = KnowledgeGraph.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(user_prompt):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + user_prompt
-        + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+generate_hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    {{ user_prompt }}
+    <|im_end|>
+    <|im_start|>assistant
+    <schema>
+    """
+)
 ```
 
 For a given user prompt, for example:
@@ -111,27 +110,26 @@ For a given user prompt, for example:
 user_prompt = "Alice loves Bob and she hates Charlie."
 ```
 
-We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
+We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
 
 ```python
-from outlines import generate, models
+from outlines import Generator
 
-model = models.LlamaCpp(llm)
-generator = generate.json(model, KnowledgeGraph)
-prompt = generate_hermes_prompt(user_prompt)
+generator = Generator(model, KnowledgeGraph)
+prompt = generate_hermes_prompt(schema=schema, user_prompt=user_prompt)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 ```
 
 We obtain the nodes and edges of the knowledge graph:
 
 ```python
-print(response.nodes)
-print(response.edges)
-# [Node(id=1, label='Alice', property='Person'),
-# Node(id=2, label='Bob', property='Person'),
-# Node(id=3, label='Charlie', property='Person')]
-# [Edge(source=1, target=2, label='love', property='Relationship'),
-# Edge(source=1, target=3, label='hate', property='Relationship')]
+print(response)
+# {"nodes":[{"id":1,"label":"Alice","property":"loves,hates"},
+# {"id":2,"label":"Bob","property":"loved_by"},
+# {"id":3,"label":"Charlie","property":"hated_by"}],
+# "edges":[{"source":1,"target":2,"label":"loves","property":"love"},
+# {"source":1,"target":3,"label":"hates","property":"hate"}]}
+
 ```
 
 ## (Optional) Visualizing the Knowledge Graph
@@ -142,10 +140,10 @@ We can use the [Graphviz library](https://graphviz.readthedocs.io/en/stable/) to
 from graphviz import Digraph
 
 dot = Digraph()
-for node in response.nodes:
-    dot.node(str(node.id), node.label, shape='circle', width='1', height='1')
-for edge in response.edges:
-    dot.edge(str(edge.source), str(edge.target), label=edge.label)
+for node in response["nodes"]:
+    dot.node(str(node["id"]), node["label"], shape='circle', width='1', height='1')
+for edge in response["edges"]:
+    dot.edge(str(edge["source"]), str(edge["target"]), label=edge["label"])
 
 dot.render('knowledge-graph.gv', view=True)
 ```
diff --git a/docs/examples/models_playing_chess.md b/docs/examples/models_playing_chess.md
index a0df2eb2d..b8d89958f 100644
--- a/docs/examples/models_playing_chess.md
+++ b/docs/examples/models_playing_chess.md
@@ -1,6 +1,6 @@
 # Large language models playing chess
 
-In this example we will make a Phi-2 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
+In this example we will make a Phi-3 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
 
 ## The chessboard
 
@@ -18,29 +18,33 @@ board = chess.Board("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
 
 ## The opponents
 
-Phi-2 will be playing against itself:
+Phi-3 will be playing against itself:
 
 ```python
-from outlines import models
-
-model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
-
+import transformers
+import outlines
+
+model_name = "microsoft/Phi-3-mini-4k-instruct"
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name),
+)
 ```
 
 ## A little help for the language model
 
-To make sure Phi-2 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
+To make sure Phi-3 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
 
 ```python
 import re
+from outlines.types.dsl import either, String
 
 def legal_moves_regex(board):
     """Build a regex that only matches valid moves."""
     legal_moves = list(board.legal_moves)
     legal_modes_str = [board.san(move) for move in legal_moves]
     legal_modes_str = [re.sub(r"[+#]", "", move) for move in legal_modes_str]
-    regex_pattern = "|".join(re.escape(move) for move in legal_modes_str)
-    regex_pattern = f"{regex_pattern}"
+    regex_pattern = either(*[String(move) for move in legal_modes_str])
     return regex_pattern
 ```
 
@@ -58,13 +62,11 @@ We update the prompt at each step so it reflects the state of the board after th
 ## Let's play
 
 ```python
-from outlines import generate
-
 board_state = " "
 turn_number = 0
 while not board.is_game_over():
     regex_pattern = legal_moves_regex(board)
-    structured = generate.regex(model, regex_pattern)(prompt + board_state)
+    structured = model(prompt + board_state, regex_pattern)
     move = board.parse_san(structured)
 
     if turn_number % 2 == 0 :  # It's White's turn
@@ -79,7 +81,7 @@ while not board.is_game_over():
     print(board_state)
 ```
 
-Interestingly enough, Phi-2 hates capturing.
+Interestingly enough, Phi-3 hates capturing.
 
 ```pgn
  e4 e5 1.Nf3 Ne7 3.b4 Nf5 5.Nc3 Ne7 7.Bb5 a6 9.Na4 b6 11.c3 Nec6 13.c4 a5 15.d4 Qg5 17.Nd2 Bb7 19.dxe5
diff --git a/docs_old/cookbook/prompt_templates/chain_of_density.txt b/docs/examples/prompt_templates/chain_of_density.txt
similarity index 100%
rename from docs_old/cookbook/prompt_templates/chain_of_density.txt
rename to docs/examples/prompt_templates/chain_of_density.txt
diff --git a/docs_old/cookbook/prompt_templates/classification.txt b/docs/examples/prompt_templates/classification.txt
similarity index 100%
rename from docs_old/cookbook/prompt_templates/classification.txt
rename to docs/examples/prompt_templates/classification.txt
diff --git a/docs_old/cookbook/prompt_templates/react_agent.txt b/docs/examples/prompt_templates/react_agent.txt
similarity index 100%
rename from docs_old/cookbook/prompt_templates/react_agent.txt
rename to docs/examples/prompt_templates/react_agent.txt
diff --git a/docs_old/cookbook/prompt_templates/simtom_prospective_taking.txt b/docs/examples/prompt_templates/simtom_prospective_taking.txt
similarity index 100%
rename from docs_old/cookbook/prompt_templates/simtom_prospective_taking.txt
rename to docs/examples/prompt_templates/simtom_prospective_taking.txt
diff --git a/docs_old/cookbook/prompt_templates/simtom_simulation.txt b/docs/examples/prompt_templates/simtom_simulation.txt
similarity index 100%
rename from docs_old/cookbook/prompt_templates/simtom_simulation.txt
rename to docs/examples/prompt_templates/simtom_simulation.txt
diff --git a/docs/examples/qa-with-citations.md b/docs/examples/qa-with-citations.md
index 79a2214c3..213942d02 100644
--- a/docs/examples/qa-with-citations.md
+++ b/docs/examples/qa-with-citations.md
@@ -11,17 +11,19 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -35,20 +37,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Generate Synthetic Data
@@ -74,26 +65,28 @@ class Users(BaseModel):
     users: List[UserDetail]
 ```
 
-We can use a `generate.json` by passing this Pydantic class we just defined, and call the generator:
+We can use a `outlines.Generator` by passing this Pydantic class we just defined, and call the generator:
 
 ```python
-model = models.LlamaCpp(llm)
-generator = generate.json(model, Users)
+import json
+
+generator = outlines.Generator(model, Users)
 response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
-print(response.users)
-# [UserDetail(id=1, first_name='John', last_name='Doe', age=25),
-# UserDetail(id=2, first_name='Jane', last_name='Doe', age=30),
-# UserDetail(id=3, first_name='Bob', last_name='Smith', age=40),
-# UserDetail(id=4, first_name='Alice', last_name='Smith', age=35),
-# UserDetail(id=5, first_name='John', last_name='Smith', age=20)]
+response = json.loads(response)
+print(response['users'])
+# [{'id': 1, 'first_name': 'John', 'last_name': 'Doe', 'age': 25},
+# {'id': 2, 'first_name': 'Jane', 'last_name': 'Doe', 'age': 30},
+# {'id': 3, 'first_name': 'Bob', 'last_name': 'Smith', 'age': 40},
+# {'id': 4, 'first_name': 'Alice', 'last_name': 'Smith', 'age': 35},
+# {'id': 5, 'first_name': 'John', 'last_name': 'Smith', 'age': 20}]
 ```
 
 ```python
-for user in response.users:
-    print(user.first_name)
-    print(user.last_name)
-    print(user.age)
-    print(#####)
+for user in response['users']:
+    print(user['first_name'])
+    print(user['last_name'])
+    print(user['age'])
+    print("#####")
 # John
 # Doe
 # 25
@@ -135,23 +128,30 @@ schema = QuestionAnswer.model_json_schema()
 We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
 
 ```python
-def generate_hermes_prompt(question, context, schema=schema):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON with correct and exact citations "
-        "extracted from the `Context`. "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema><|im_end|>\n"
-        "<|im_start|>user\n"
-        + "`Context`: "
-        + context
-        + "\n`Question`: "
-        + question + "<|im_end|>"
-        + "\n<|im_start|>assistant\n"
-        "<schema>"
-    )
+from outlines import Template
+
+hermes_prompt = Template.from_string(
+    """
+    <|im_start|>system
+    You are a world class AI model who answers questions in JSON with correct and exact citations
+    extracted from the `Context`.
+    Here's the json schema you must adhere to:
+    <schema>
+    {{ schema }}
+    </schema>
+    <|im_end|>
+    <|im_start|>user
+    `Context`:
+    {{ context }}
+    `Question`:
+    {{ question }}
+    <|im_end|>
+    <|im_start|>assistant
+    """
+)
 ```
 
-We can use `generate.json` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
+We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
 
 ```python
 question = "What did the author do during college?"
@@ -161,11 +161,11 @@ I went to an arts high school but in university I studied Computational Mathemat
 As part of coop I worked at many companies including Stitchfix, Facebook.
 I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.
 """
-generator = generate.json(model, QuestionAnswer)
-prompt = generate_hermes_prompt(question, context)
+generator = outlines.Generator(model, QuestionAnswer)
+prompt = hermes_prompt(question=question, context=context, schema=schema)
 response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
 print(response)
-# QuestionAnswer(question='What did the author do during college?', answer='The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.', citations=['I went to an arts high school but in university I studied Computational Mathematics and physics.', 'I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.'])
+# {"question": "What did the author do during college?", "answer": "The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.", "citations": ["I went to an arts high school but in university I studied Computational Mathematics and physics.", "I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years."]}
 ```
 
 We can do the same for a list of question-context pairs:
@@ -219,12 +219,13 @@ for question, context in [
     (question4, context4),
     (question5, context5),
 ]:
-    final_prompt = my_final_prompt(question, context)
-    generator = generate.json(model, QuestionAnswer)
-    response = generator(final_prompt, max_tokens=1024, temperature=0, seed=42)
-    display(question)
-    display(response.answer)
-    display(response.citations)
+    prompt = hermes_prompt(question=question, context=context, schema=schema)
+    generator = outlines.Generator(model, QuestionAnswer)
+    response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
+    response = json.loads(response)
+    print(question)
+    print(response['answer'])
+    print(response['citations'])
     print("\n\n")
 
 # 'Where was John born?'
diff --git a/docs/examples/react_agent.md b/docs/examples/react_agent.md
index 15b6ed369..bd3d51d6c 100644
--- a/docs/examples/react_agent.md
+++ b/docs/examples/react_agent.md
@@ -15,17 +15,19 @@ pip install llama-cpp-python
 We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
 ```python
 import llama_cpp
-from outlines import generate, models
-
-model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-            "Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-            tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-            ),
-            n_gpu_layers=-1,
-            flash_attn=True,
-            n_ctx=8192,
-            verbose=False)
+import outlines
+
+llm = llama_cpp.Llama(
+    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
+    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
+        "NousResearch/Hermes-2-Pro-Llama-3-8B"
+    ),
+    n_gpu_layers=-1,
+    flash_attn=True,
+    n_ctx=8192,
+    verbose=False
+)
+model = outlines.from_llamacpp(llm)
 ```
 
 ??? note "(Optional) Store the model weights in a custom folder"
@@ -39,20 +41,9 @@ model = models.llamacpp("NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
     We initialize the model:
 
     ```python
-    import llama_cpp
     from llama_cpp import Llama
-    from outlines import generate, models
-
-    llm = Llama(
-        "/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf",
-        tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-            "NousResearch/Hermes-2-Pro-Llama-3-8B"
-        ),
-        n_gpu_layers=-1,
-        flash_attn=True,
-        n_ctx=8192,
-        verbose=False
-    )
+
+    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
     ```
 
 ## Build a ReAct agent
@@ -115,51 +106,16 @@ from typing import Union
 
 class Decision(BaseModel):
     Decision: Union[Reason_and_Act, Final_Answer]
-```
-
-We could generate a response using the json schema but we will use the regex and check that everything is working as expected:
-
-```python
-from outlines.fsm.json_schema import convert_json_schema_to_str
-from outlines_core.fsm.json_schema import build_regex_from_schema
 
 json_schema = Decision.model_json_schema()
-schema_str = convert_json_schema_to_str(json_schema=json_schema)
-regex_str = build_regex_from_schema(schema_str)
-print(regex_str)
-# '\\{[ ]?"Decision"[ ]?:[ ]?(\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Thought"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Action"[ ]?:[ ]?("wikipedia"|"calculate")[ ]?,[ ]?"Action_Input"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\}|\\{[ ]?"Scratchpad"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?,[ ]?"Final_Answer"[ ]?:[ ]?"([^"\\\\\\x00-\\x1F\\x7F-\\x9F]|\\\\["\\\\])*"[ ]?\\})[ ]?\\}'
 ```
 
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic:
+We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic. We can load a template from a file for that:
 
 ```python
-import datetime
-
-def generate_hermes_prompt(question, schema=""):
-    return (
-        "<|im_start|>system\n"
-        "You are a world class AI model who answers questions in JSON with correct Pydantic schema. "
-        f"Here's the json schema you must adhere to:\n<schema>\n{schema}\n</schema>\n"
-        "Today is " + datetime.datetime.today().strftime('%Y-%m-%d') + ".\n" +
-        "You run in a loop of Scratchpad, Thought, Action, Action Input, PAUSE, Observation. "
-        "At the end of the loop you output a Final Answer. "
-        "Use Scratchpad to store the information from the Observation useful to answer the question "
-        "Use Thought to describe your thoughts about the question you have been asked "
-        "and reflect carefully about the Observation if it exists. "
-        "Use Action to run one of the actions available to you. "
-        "Use Action Input to input the arguments of the selected action - then return PAUSE. "
-        "Observation will be the result of running those actions. "
-        "Your available actions are:\n"
-        "calculate:\n"
-        "e.g. calulate: 4**2 / 3\n"
-        "Runs a calculation and returns the number - uses Python so be sure to use floating point syntax if necessary\n"
-        "wikipedia:\n"
-        "e.g. wikipedia: Django\n"
-        "Returns a summary from searching Wikipedia\n"
-        "DO NOT TRY TO GUESS THE ANSWER. Begin! <|im_end|>"
-        "\n<|im_start|>user\n" + question + "<|im_end|>"
-        "\n<|im_start|>assistant\n"
-    )
+from outlines import Template
+
+hermes_prompt = Template.from_file("prompt_templates/react_agent.txt")
 ```
 
 We define a ChatBot class
@@ -175,7 +131,7 @@ class ChatBot:
         return result
 
     def execute(self):
-        generator = generate.regex(model, regex_str)
+        generator = outlines.Generator(model, Decision)
         result = generator(self.prompt, max_tokens=1024, temperature=0, seed=42)
         return result
 ```
@@ -194,7 +150,11 @@ def query(question, max_turns=5):
     previous_actions = []
     while i < max_turns:
         i += 1
-        prompt = generate_hermes_prompt(question=question, schema=Decision.model_json_schema())
+        prompt = generate_hermes_prompt(
+            question=question,
+            schema=Decision.model_json_schema(),
+            today=datetime.datetime.today().strftime('%Y-%m-%d')
+        )
         bot = ChatBot(prompt=prompt)
         result = bot(next_prompt)
         json_result = json.loads(result)['Decision']
diff --git a/docs/examples/read-pdfs.md b/docs/examples/read-pdfs.md
index dbe4ccb02..1854aba42 100644
--- a/docs/examples/read-pdfs.md
+++ b/docs/examples/read-pdfs.md
@@ -46,17 +46,19 @@ We've tested this example with [Pixtral 12b](https://huggingface.co/mistral-comm
 To use Pixtral:
 
 ```python
-from transformers import LlavaForConditionalGeneration
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
+processor_class = LlavaProcessor
 ```
 
 To use Qwen-2-VL:
 
 ```python
-from transformers import Qwen2VLForConditionalGeneration
+from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
+processor_class = AutoProcessor
 ```
 
 You can load your model into memory with:
@@ -64,17 +66,12 @@ You can load your model into memory with:
 ```python
 # This loads the model into memory. On your first run,
 # it will have to download the model, which might take a while.
-model = outlines.models.transformers_vision(
-    model_name,
-    model_class=model_class,
-    model_kwargs={
-        "device_map": "auto",
-        "torch_dtype": torch.bfloat16,
-    },
-    processor_kwargs={
-        "device": "auto",
-    },
-)
+model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
+processor_kwargs={"device_map": "cpu"}
+tf_model = model_class.from_pretrained(model_name, **model_kwargs)
+tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+model = outlines.from_transformers(tf_model, tf_processor)
 ```
 
 ## Convert the PDF to images
@@ -159,7 +156,7 @@ The structured output you can extract is exactly the same as everywhere else in
 
 Suppose you wished to go through each page of the PDF, and extract the page description, key takeaways, and page number.
 
-You can do this by defining a JSON schema, and then using `outlines.generate.json` to extract the data.
+You can do this by defining a JSON schema, and then using `outlines.Generator` to extract the data.
 
 First, define the structure you want to extract:
 
@@ -170,7 +167,7 @@ class PageSummary(BaseModel):
     page_number: int
 ```
 
-Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers `AutoProcessor` to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
+Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers processor to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
 
 Images are denoted with `type: "image"`, and text is denoted with `type: "text"`.
 
@@ -191,16 +188,15 @@ messages = [
                 """
             },
 
-            # Don't need to pass in an image, since we do this
-            # when we call the generator function down below.
+            # This a placeholder, the actual image is passed in when
+            # we call the generator function down below.
             {"type": "image", "image": ""},
         ],
     }
 ]
 
 # Convert the messages to the final prompt
-processor = AutoProcessor.from_pretrained(model_name)
-instruction = processor.apply_chat_template(
+prompt = tf_processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -209,10 +205,10 @@ Now we iterate through each image, and extract the structured information:
 
 ```python
 # Page summarizer function
-page_summary_generator = outlines.generate.json(model, PageSummary)
+page_summary_generator = outlines.Generator(model, PageSummary)
 
 for image in images:
-    result = page_summary_generator(instruction, [image])
+    result = page_summary_generator({"text": prompt, "images": image})
     print(result)
 ```
 
@@ -225,19 +221,21 @@ arXiv identifiers are optionally followed by a version number, i.e. `arXiv:YYMM.
 We can use a regular expression to define this patter:
 
 ```python
-paper_regex = r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?'
+from outlines.types import Regex
+
+paper_regex = Regex(r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?')
 ```
 
 We can build an extractor function from the regex:
 
 ```python
-id_extractor = outlines.generate.regex(model, paper_regex)
+id_extractor = outlines.Generator(model, paper_regex)
 ```
 
 Now, we can extract the arxiv paper identifier from the first image:
 
 ```python
-arxiv_instruction = processor.apply_chat_template(
+arxiv_instruction = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -256,7 +254,7 @@ arxiv_instruction = processor.apply_chat_template(
 )
 
 # Extract the arxiv paper identifier
-paper_id = id_extractor(arxiv_instruction, [images[0]])
+paper_id = id_extractor({"text": arxiv_instruction, "images": images[0]})
 ```
 
 As of the time of this writing, the arxiv paper identifier is
@@ -269,7 +267,9 @@ Your version number may be different, but the part before `vX` should match.
 
 ### Categorize the paper into one of several categories
 
-`outlines.generate.choice` allows the model to select one of several options. Suppose we wanted to categorize the paper into being about "language models", "economics", "structured generation", or "other".
+`outlines.Generator` also allows the model to select one of several options by providing a Literal type hint with the categories.
+
+Suppose we wanted to categorize the paper into being about "language models", "cell biology", or "other". We would then define the output type as `Literal["llms", "cell biology", "other"]`.
 
 Let's define a few categories we might be interested in:
 
@@ -284,7 +284,7 @@ categories = [
 Now we can construct the prompt:
 
 ```python
-categorization_instruction = processor.apply_chat_template(
+categorization_instruction = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -310,14 +310,13 @@ categorization_instruction = processor.apply_chat_template(
 Now we can show the model the first page and extract the category:
 
 ```python
+from typing import Literal
+
 # Build the choice extractor
-categorizer = outlines.generate.choice(
-    model,
-    categories
-)
+categorizer = outlines.Generator(model, Literal["llms", "cell biology", "other"])
 
 # Categorize the paper
-category = categorizer(categorization_instruction, [images[0]])
+category = categorizer({"text": categorization_instruction, "images": images[0]})
 print(category)
 ```
 
@@ -332,12 +331,12 @@ llms
 You can provide multiple images to the model by
 
 1. Adding additional image messages
-2. Providing a list of images to the `generate` function
+2. Providing a list of images to the generator
 
 For example, to have two images, you can do:
 
 ```python
-two_image_prompt = processor.apply_chat_template(
+two_image_prompt = tf_processor.apply_chat_template(
     [
         {
             "role": "user",
@@ -355,17 +354,9 @@ two_image_prompt = processor.apply_chat_template(
 )
 
 # Pass two images to the model
-generator = outlines.generate.choice(
-    model,
-    ["hot dog", "not hot dog"]
-)
+generator = outlines.Generator(model, Literal["hot dog", "not hot dog"])
 
-result = generator(
-    two_image_prompt,
-
-    # Pass two images to the model
-    [images[0], images[1]]
-)
+result = generator({"text": two_image_prompt, "images": [images[0], images[1]]})
 print(result)
 ```
 
diff --git a/docs/examples/receipt-digitization.md b/docs/examples/receipt-digitization.md
index 0fa256db7..8ecb07969 100644
--- a/docs/examples/receipt-digitization.md
+++ b/docs/examples/receipt-digitization.md
@@ -38,18 +38,20 @@ If you want to use Qwen-2-VL, you can do the following:
 
 ```python
 # To use Qwen-2-VL:
-from transformers import Qwen2VLForConditionalGeneration
+from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
 model_name = "Qwen/Qwen2-VL-7B-Instruct"
 model_class = Qwen2VLForConditionalGeneration
+processor_class = AutoProcessor
 ```
 
 If you want to use Pixtral, you can do the following:
 
 ```python
 # To use Pixtral:
-from transformers import LlavaForConditionalGeneration
+from transformers import LlavaForConditionalGeneration, LlavaProcessor
 model_name="mistral-community/pixtral-12b"
 model_class=LlavaForConditionalGeneration
+processor_class = LlavaProcessor
 ```
 
 ## Load the model
@@ -57,17 +59,12 @@ model_class=LlavaForConditionalGeneration
 Load the model into memory:
 
 ```python
-model = outlines.models.transformers_vision(
-    model_name,
-    model_class=model_class,
-    model_kwargs={
-        "device_map": "auto",
-        "torch_dtype": torch.bfloat16,
-    },
-    processor_kwargs={
-        "device": "cuda", # set to "cpu" if you don't have a GPU
-    },
-)
+model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
+processor_kwargs={"device_map": "cuda"}
+tf_model = model_class.from_pretrained(model_name, **model_kwargs)
+tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
+
+model = outlines.from_transformers(tf_model, tf_processor)
 ```
 
 ## Image processing
@@ -166,7 +163,7 @@ class ReceiptSummary(BaseModel):
 
 ## Prepare the prompt
 
-We'll use the `AutoProcessor` to convert the image and the text prompt into a format that the model can understand. Practically,
+We'll use the `tf_processor` to convert the image and the text prompt into a format that the model can understand. Practically,
 this is the code that adds user, system, assistant, and image tokens to the prompt.
 
 ```python
@@ -194,8 +191,7 @@ messages = [
 ]
 
 # Convert the messages to the final prompt
-processor = AutoProcessor.from_pretrained(model_name)
-prompt = processor.apply_chat_template(
+prompt = tf_processor.apply_chat_template(
     messages, tokenize=False, add_generation_prompt=True
 )
 ```
@@ -222,17 +218,13 @@ Return the information in the following JSON schema:
 
 ```python
 # Prepare a function to process receipts
-receipt_summary_generator = outlines.generate.json(
-    model,
-    ReceiptSummary,
-
-    # Greedy sampling is a good idea for numeric
-    # data extraction -- no randomness.
-    sampler=outlines.samplers.greedy()
-)
+receipt_summary_generator = outlines.Generator(model, ReceiptSummary)
 
 # Generate the receipt summary
-result = receipt_summary_generator(prompt, [image])
+result = receipt_summary_generator(
+    {"text": prompt, "images": image},
+    max_new_tokens=1024
+)
 print(result)
 ```
 
@@ -241,32 +233,31 @@ print(result)
 The output should look like this:
 
 ```
-ReceiptSummary(
-    store_name="Trader Joe's",
-    store_address='401 Bay Street, San Francisco, CA 94133',
-    store_number=0,
-    items=[
-        Item(name='BANANA EACH', quantity=7, price_per_unit=0.23, total_price=1.61),
-        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CHOCOLATE DOUG', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='BAREBELLS CARAMEL CASHEW', quantity=2, price_per_unit=2.29, total_price=4.58),
-        Item(name='BAREBELLS CREAMY CRISP', quantity=1, price_per_unit=2.29, total_price=2.29),
-        Item(name='SPINDRIFT ORANGE MANGO 8', quantity=1, price_per_unit=7.49, total_price=7.49),
-        Item(name='Bottle Deposit', quantity=8, price_per_unit=0.05, total_price=0.4),
-        Item(name='MILK ORGANIC GALLON WHOL', quantity=1, price_per_unit=6.79, total_price=6.79),
-        Item(name='CLASSIC GREEK SALAD', quantity=1, price_per_unit=3.49, total_price=3.49),
-        Item(name='COBB SALAD', quantity=1, price_per_unit=5.99, total_price=5.99),
-        Item(name='PEPPER BELL RED XL EACH', quantity=1, price_per_unit=1.29, total_price=1.29),
-        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25),
-        Item(name='BAG FEE.', quantity=1, price_per_unit=0.25, total_price=0.25)
-    ],
-    tax=0.68,
-    total=41.98,
-    date='2023-11-04',
-    payment_method='debit',
-
-)
+{
+  "store_name": "Trader Joe's",
+  "store_address": "401 Bay Street, San Francisco, CA 94133",
+  "store_number": 0,
+  "items": [
+    {"name": "BANANA EACH", "quantity": 7, "price_per_unit": 0.23, "total_price": 1.61},
+    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "BAREBELLS CARAMEL CASHEW", "quantity": 2, "price_per_unit": 2.29, "total_price": 4.58},
+    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
+    {"name": "SPINDRIFT ORANGE MANGO 8", "quantity": 1, "price_per_unit": 7.49, "total_price": 7.49},
+    {"name": "Bottle Deposit", "quantity": 8, "price_per_unit": 0.05, "total_price": 0.4},
+    {"name": "MILK ORGANIC GALLON WHOL", "quantity": 1,"price_per_unit": 6.79,"total_price": 6.79},
+    {"name": "CLASSIC GREEK SALAD", "quantity": 1, "price_per_unit": 3.49, "total_price": 3.49},
+    {"name": "COBB SALAD", "quantity": 1, "price_per_unit": 5.99, "total_price": 5.99},
+    {"name": "PEPPER BELL RED XL EACH", "quantity": 1, "price_per_unit": 1.29, "total_price": 1.29},
+    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
+    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
+  ],
+  "tax": 0.68,
+  "total": 41.98,
+  "date": "2023-11-04",
+  "payment_method": "debit"
+}
 ```
 
 Voila! You've successfully extracted information from a receipt using an LLM.
@@ -284,13 +275,13 @@ class ReceiptSummary(BaseModel):
 which gives you a result like
 
 ```
-ReceiptSummary(
+{
     ...
-    roast="You must be a fan of Trader Joe's because you bought enough
+    "roast": "You must be a fan of Trader Joe's because you bought enough
     items to fill a small grocery bag and still had to pay for a bag fee.
     Maybe you should start using reusable bags to save some money and the
     environment."
-)
+}
 ```
 
 Qwen is not particularly funny, but worth a shot.
diff --git a/docs/examples/simtom.md b/docs/examples/simtom.md
index a80ebc24f..53820258e 100644
--- a/docs/examples/simtom.md
+++ b/docs/examples/simtom.md
@@ -17,7 +17,7 @@ SimToM calls an LLM with two consecutive prompts:
 
 To implement SimToM with Outlines, we will need to:
 
-1. Write the prompts with [prompt functions](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
+1. Write the prompts with [prompt templates](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
 2. Define the JSON object each prompt will return using Pydantic.
 3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/latest/reference/models/transformers/).
 
@@ -30,29 +30,8 @@ The authors have shared their code, prompts and data in [this GitHub repository]
 ```python
 from outlines import Template
 
-
-perspective_taking = Template.from_string(
-    """<s>[INST] The following is a sequence of events about some characters, that takes place in multiple locations.
-    Your job is to output only the events that the specified character, {{character}}, knows about.
-
-    Here are a few rules:
-    1. A character knows about all events that they do.
-    2. If a character is in a certain room/location, that character knows about all other events that happens in the room. This includes other characters leaving or exiting the location, the locations of objects in that location, and whether somebody moves an object to another place.
-    3. If a character leaves a location, and is NOT in that location, they no longer know about any events that happen within that location. However, they can re-enter the location.
-
-    Story: {{story}}
-    What events does {{character}} know about? Only output the events according to the above rules, do not provide an explanation. [/INST]""" # noqa
-)
-
-simulation = Template.from_string(
-    """<s>[INST] {% for event in events %}
-    {{event}}
-    {% endfor %}
-    You are {{name}}.
-    Based on the above information, answer the following question:
-    {{question}}
-    You must choose one of the above choices, do not say there is not enough information. Answer with a single word, do not output anything else. [/INST]""" # noqa
-)
+perspective_taking = Template.from_file("prompt_templates/simtom_prospective_taking.txt")
+simulation = Template.from_file("prompt_templates/simtom_simulation.txt")
 ```
 
 ### JSON Structured Generation
@@ -65,13 +44,11 @@ We will need two Pydantic models for SimToM, one for each prompt:
 from pydantic import BaseModel, Field
 from typing import List
 
-
 class PerspectiveTaking(BaseModel):
     """This is for the first prompt."""
     character: str = Field(description="The character we extract the events for.")
     events: List[str] = Field(description="All events that the character knows about.")
 
-
 class Simulation(BaseModel):
     """This is for the second prompt."""
     answer: str
@@ -97,30 +74,37 @@ character = "Aria"
 We load `Mistral-7B-Instruct-v0.3`, create the prompt using the template we defined earlier, and generate a structured response. As a reminder, the goal of the first call is to get all the events a character, `Aria`, knows about.
 
 ```python
+import transformers
+import outlines
 # Load an LLM from Hugging Face
 MODEL_NAME = "mistral-community/Mistral-7B-Instruct-v0.3"
-model = outlines.models.transformers(MODEL_NAME, device="cuda")
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+    transformers.AutoTokenizer.from_pretrained(MODEL_NAME),
+)
 
 perspective_prompt = perspective_taking(story=story, character=character)
 
 # Call Mistral 7B with the first prompt
-generator = outlines.generate.json(model, PerspectiveTaking)
-perspective = generator(perspective_prompt)
+generator = outlines.Generator(model, PerspectiveTaking)
+perspective = generator(perspective_prompt, max_new_tokens=1024)
 
-print(perspective.model_dump())
+print(perspective)
 # {'character': 'Aria', 'events': ['1 Aria entered the front_yard.', '3 The grapefruit is in the green_bucket.', '4 Aria moved the grapefruit to the blue_container.']}
 ```
 
 Not bad! We will now generate the second prompt with those events.
 
 ```python
-sim_prompt = simulation(events=perspective.events, name=character, question=question)
+import json
+
+sim_prompt = simulation(events=json.loads(perspective)["events"], name=character, question=question)
 
 # Call Mistral 7B with the second prompt
-generator = outlines.generate.json(model, Simulation)
-result = generator(sim_prompt)
+generator = outlines.Generator(model, Simulation)
+result = generator(sim_prompt, max_new_tokens=1024)
 
-print(result.model_dump())
+print(result)
 # {'answer': 'green_bucket'}
 ```
 
diff --git a/docs/examples/structured_generation_workflow.md b/docs/examples/structured_generation_workflow.md
index b9d7c28a4..524283f84 100644
--- a/docs/examples/structured_generation_workflow.md
+++ b/docs/examples/structured_generation_workflow.md
@@ -2,8 +2,7 @@
 
 This is a condensed version of [Coding for Structured Generation with LLMs](https://blog.dottxt.co/coding-for-structured-generation.html).
 
-For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this
-with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
+For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
 
 ## Unstructured approach
 
@@ -11,36 +10,30 @@ Before diving into how to use structure generation for this task let's start wit
 
 ```python
 import outlines
+import transformers
 
 model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.models.transformers(model_name)
+model = outlines.from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained(model_name),
+    transformers.AutoTokenizer.from_pretrained(model_name)
+)
 ```
 
 Next we need a prompt for this model. Since we're focusing on structured generation, we won't be engaging in any form of "prompt hacking" and will be leaving this prompt untouched for the rest of this example.
 
 ```python
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-
-messages_phone = [
-            {"role": "user", "content": """
-            Please generate a realistic phone number for Washington State in the following format
-
-            (555) 555-5555
-
-            """}
-]
-
-# This allows us to properly format our prompt for
-# Phi-3 Mini's 'Instruct' interface.
-prompt_phone = tokenizer.apply_chat_template(messages_phone, tokenize=False)
+prompt_phone = """
+    Please generate a realistic phone number for Washington State in the following format
+    (555) 555-5555
+"""
 ```
 
 With our prompt ready we can now generate 10 example phone numbers
 
 ```python
-phone_generator_unstruct = outlines.generate.text(model)
-for _ in range(10):
-    print(phone_generator_unstruct(prompt_phone,max_tokens=12))
+phone_generator_unstruct = outlines.Generator(model)
+for _ in range(3):
+    print(phone_generator_unstruct(prompt_phone, max_new_tokens=12))
 ```
 
 > I'd be happy to help you generate a realistic phone\
@@ -81,7 +74,9 @@ For a simple example like this, we'll just be using a single phone number, for m
 The next step in the process is for use to define a simple regex that we feel correctly models our real data.
 
 ```python
-phone_regex_1 = r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}'
+from outlines.types import Regex
+
+phone_regex_1 = Regex(r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}')
 ```
 
 Next we need to validate this regex against our real data.
@@ -94,8 +89,8 @@ We'll start with a simple method of validation: just checking that our regex mat
 
 ```
 import re
-re.match(phone_regex_1, phone_number)
 
+re.match(phone_regex_1.pattern, phone_number)
 # <re.Match object; span=(0, 14), match='(206) 386-4636'>
 
 ```
@@ -107,8 +102,9 @@ Now that we have a match, we can move on to generating structured output!
 We're ready to see if structured generation can make an improvement over our initial unstructured approach:
 
 ```python
-phone_generator_v1 = outlines.generate.regex(model, phone_regex_1)
-for _ in range(10):
+phone_generator_v1 = outlines.Generator(model, phone_regex_1)
+
+for _ in range(3):
     print(phone_generator_v1(prompt_phone))
 ```
 > (206) 555-1234\
@@ -135,21 +131,21 @@ We've walked through the loop once, so we can go quickly now through each iterat
 We start by improving our structure:
 
 ```python
-phone_regex_2 = r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}'
+phone_regex_2 = Regex(r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}')
 ```
 
 Before rushing to another round of generation, let's validate this new regex. We'll add just a bit more sophistication over our last check:
 
 ```python
-re.match(phone_regex_2, phone_number)[0] == phone_number
+re.match(phone_regex_2.pattern, phone_number)[0] == phone_number
 # True
 ```
 Now that we've validated, let's generate with this new regex!
 
 ```python
-phone_generator_v2 = outlines.generate.regex(model,
-                                             phone_regex_2)
-for _ in range(10):
+phone_generator_v2 = outlines.Generator(model, phone_regex_2)
+
+for _ in range(3):
     print(phone_generator_v2(prompt_phone))
 ```
 
@@ -192,10 +188,10 @@ This prints out:
 Ah! We were missing the last digit, let's fix that and regenerate:
 
 ```python
-phone_regex_3_fixed = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]'
-phone_generator_v3 = outlines.generate.regex(model,
-                                             phone_regex_3_fixed)
-for _ in range(10):
+phone_regex_3_fixed = Regex(r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]')
+phone_generator_v3 = outlines.Generator(model, phone_regex_3_fixed)
+
+for _ in range(3):
     print(phone_generator_v3(prompt_phone))
 ```
 
diff --git a/docs_old/api/applications.md b/docs_old/api/applications.md
deleted file mode 100644
index b44ddf566..000000000
--- a/docs_old/api/applications.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.applications
diff --git a/docs_old/api/core/application.md b/docs_old/api/core/application.md
deleted file mode 100644
index 242caa135..000000000
--- a/docs_old/api/core/application.md
+++ /dev/null
@@ -1,76 +0,0 @@
----
-title: Application API
----
-
-# Application
-
-::: outlines.applications.Application
-
-The `Application` class enables you to encapsulate a prompt template and an output type into a reusable component.
-
-## Overview
-
-An `Application` combines a prompt template with an output type, creating a reusable component that can be applied to different models.
-
-## Basic Usage
-
-```python
-from outlines import Application, Template, from_transformers
-from typing import List
-
-# Create a template
-template = Template("""
-You are an expert classifier.
-Classify the following text into one of these categories: {categories}
-
-Text: {text}
-""")
-
-# Create an application with the template and output type
-classifier = Application(template, str)
-
-# Initialize a model
-model = from_transformers(...)
-
-# Use the application with specific parameters
-categories = ["Sports", "Politics", "Technology", "Entertainment"]
-result = classifier(model, categories=categories, text="Apple unveils new iPhone with AI capabilities")
-
-print(result)  # Expected output: "Technology"
-```
-
-## Structured Output
-
-```python
-from pydantic import BaseModel
-from outlines import Application, Template, from_transformers
-
-# Define a Pydantic model for structured output
-class MovieReview(BaseModel):
-    title: str
-    director: str
-    year: int
-    rating: float
-    review: str
-
-# Create a template
-template = Template("""
-Write a brief review of the movie {movie_title}.
-Include the director, year of release, and a rating out of 10.
-""")
-
-# Create an application with the template and output type
-movie_reviewer = Application(template, MovieReview)
-
-# Use the application with a model
-result = movie_reviewer(model, movie_title="The Matrix")
-
-# Parse the result into a Pydantic model
-review = MovieReview.model_validate_json(result)
-print(f"{review.title} ({review.year}) - {review.rating}/10")
-```
-
-## Parameters
-
-- `prompt_template`: A template that defines the prompt structure
-- `output_type`: The type of output to generate
diff --git a/docs_old/api/guide.md b/docs_old/api/guide.md
deleted file mode 100644
index 08d2f7f0f..000000000
--- a/docs_old/api/guide.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.processors.guide
diff --git a/docs_old/api/index.md b/docs_old/api/index.md
deleted file mode 100644
index b0d5c88fc..000000000
--- a/docs_old/api/index.md
+++ /dev/null
@@ -1 +0,0 @@
-# API Reference
diff --git a/docs_old/api/json_schema.md b/docs_old/api/json_schema.md
deleted file mode 100644
index 471cb3a80..000000000
--- a/docs_old/api/json_schema.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.fsm.json_schema
diff --git a/docs_old/api/models.md b/docs_old/api/models.md
deleted file mode 100644
index cb9497be1..000000000
--- a/docs_old/api/models.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.models
diff --git a/docs_old/api/parsing.md b/docs_old/api/parsing.md
deleted file mode 100644
index e9662999c..000000000
--- a/docs_old/api/parsing.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.fsm.parsing
diff --git a/docs_old/api/regex.md b/docs_old/api/regex.md
deleted file mode 100644
index 5ef91db46..000000000
--- a/docs_old/api/regex.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.generate.regex
diff --git a/docs_old/api/templates.md b/docs_old/api/templates.md
deleted file mode 100644
index 6150512d4..000000000
--- a/docs_old/api/templates.md
+++ /dev/null
@@ -1 +0,0 @@
-::: outlines.templates
diff --git a/docs_old/assets/images/dottxt-dark.svg b/docs_old/assets/images/dottxt-dark.svg
deleted file mode 100644
index a95ff7d32..000000000
--- a/docs_old/assets/images/dottxt-dark.svg
+++ /dev/null
@@ -1 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?><svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 129.09 196.26"><defs><style>.cls-1{fill:#fff;}</style></defs><g id="Layer_1-2"><rect class="cls-1" x="39.12" y="3.91" width="3.91" height="3.91"/><polygon class="cls-1" points="46.94 7.82 50.86 7.82 54.77 7.82 54.77 3.91 50.86 3.91 50.86 0 46.94 0 43.03 0 43.03 3.91 46.94 3.91 46.94 7.82"/><polygon class="cls-1" points="58.68 7.82 62.59 7.82 66.5 7.82 66.5 3.91 62.59 3.91 62.59 0 58.68 0 54.77 0 54.77 3.91 58.68 3.91 58.68 7.82"/><polygon class="cls-1" points="70.42 7.82 74.33 7.82 78.24 7.82 78.24 3.91 74.33 3.91 74.33 0 70.42 0 66.5 0 66.5 3.91 70.42 3.91 70.42 7.82"/><polygon class="cls-1" points="82.15 7.82 86.06 7.82 86.06 11.74 89.98 11.74 93.89 11.74 97.8 11.74 97.8 7.82 101.71 7.82 101.71 3.91 97.8 3.91 97.8 0 93.89 0 89.98 0 89.98 3.91 93.89 3.91 93.89 7.82 89.98 7.82 89.98 3.91 86.06 3.91 86.06 0 82.15 0 78.24 0 78.24 3.91 82.15 3.91 82.15 7.82"/><polygon class="cls-1" points="105.62 7.82 109.53 7.82 113.45 7.82 113.45 11.74 117.36 11.74 121.27 11.74 121.27 7.82 117.36 7.82 117.36 3.91 113.45 3.91 109.53 3.91 109.53 0 105.62 0 101.71 0 101.71 3.91 105.62 3.91 105.62 7.82"/><rect class="cls-1" x="43.03" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="54.77" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="66.5" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="78.24" y="7.82" width="3.91" height="3.91"/><rect class="cls-1" x="101.71" y="7.82" width="3.91" height="3.91"/><polygon class="cls-1" points="39.12 39.12 43.03 39.12 46.94 39.12 50.86 39.12 54.77 39.12 58.68 39.12 62.59 39.12 66.5 39.12 70.42 39.12 74.33 39.12 78.24 39.12 78.24 35.21 74.33 35.21 70.42 35.21 66.5 35.21 62.59 35.21 58.68 35.21 54.77 35.21 50.86 35.21 46.94 35.21 43.03 35.21 39.12 35.21 39.12 39.12"/><polygon class="cls-1" points="66.5 50.86 62.59 50.86 58.68 50.86 54.77 50.86 50.86 50.86 46.94 50.86 43.03 50.86 39.12 50.86 35.21 50.86 31.3 50.86 31.3 54.77 35.21 54.77 39.12 54.77 43.03 54.77 46.94 54.77 50.86 54.77 54.77 54.77 58.68 54.77 62.59 54.77 66.5 54.77 70.42 54.77 74.33 54.77 74.33 50.86 70.42 50.86 66.5 50.86"/><polygon class="cls-1" points="43.03 101.71 46.94 101.71 50.86 101.71 54.77 101.71 58.68 101.71 62.59 101.71 66.5 101.71 70.42 101.71 74.33 101.71 78.24 101.71 82.15 101.71 86.06 101.71 89.98 101.71 93.89 101.71 97.8 101.71 101.71 101.71 105.62 101.71 109.53 101.71 109.53 97.8 105.62 97.8 101.71 97.8 97.8 97.8 93.89 97.8 89.98 97.8 86.06 97.8 82.15 97.8 78.24 97.8 74.33 97.8 70.42 97.8 66.5 97.8 62.59 97.8 58.68 97.8 54.77 97.8 50.86 97.8 46.94 97.8 43.03 97.8 43.03 93.89 43.03 89.97 46.94 89.97 50.86 89.97 54.77 89.97 58.68 89.97 62.59 89.97 66.5 89.97 70.42 89.97 74.33 89.97 78.24 89.97 78.24 86.06 82.15 86.06 86.06 86.06 89.98 86.06 93.89 86.06 97.8 86.06 101.71 86.06 105.62 86.06 109.53 86.06 109.53 82.15 105.62 82.15 101.71 82.15 97.8 82.15 93.89 82.15 89.98 82.15 86.06 82.15 82.15 82.15 82.15 78.24 82.15 74.33 86.06 74.33 86.06 70.42 89.98 70.42 93.89 70.42 97.8 70.42 101.71 70.42 105.62 70.42 109.53 70.42 109.53 66.5 105.62 66.5 101.71 66.5 97.8 66.5 93.89 66.5 89.98 66.5 89.98 62.59 89.98 58.68 93.89 58.68 93.89 54.77 97.8 54.77 101.71 54.77 105.62 54.77 109.53 54.77 109.53 50.86 105.62 50.86 101.71 50.86 97.8 50.86 97.8 46.94 97.8 43.03 101.71 43.03 101.71 39.12 105.62 39.12 109.53 39.12 109.53 35.21 105.62 35.21 105.62 31.3 109.53 31.3 109.53 27.38 109.53 23.47 109.53 19.56 113.45 19.56 113.45 15.65 113.45 11.74 109.53 11.74 105.62 11.74 105.62 15.65 105.62 19.56 101.71 19.56 101.71 23.47 101.71 27.38 97.8 27.38 97.8 31.3 97.8 35.21 93.89 35.21 93.89 39.12 93.89 43.03 89.98 43.03 89.98 46.94 89.98 50.86 86.06 50.86 86.06 54.77 86.06 58.68 82.15 58.68 82.15 62.59 82.15 66.5 78.24 66.5 78.24 70.42 78.24 74.33 74.33 74.33 74.33 78.24 74.33 82.15 70.42 82.15 66.5 82.15 62.59 82.15 58.68 82.15 54.77 82.15 50.86 82.15 46.94 82.15 43.03 82.15 39.12 82.15 35.21 82.15 31.3 82.15 27.38 82.15 23.47 82.15 19.56 82.15 15.65 82.15 11.74 82.15 7.82 82.15 7.82 78.24 7.82 74.33 7.82 70.42 11.74 70.42 11.74 66.5 11.74 62.59 15.65 62.59 15.65 58.68 15.65 54.77 19.56 54.77 19.56 50.86 19.56 46.94 23.47 46.94 23.47 43.03 23.47 39.12 27.38 39.12 27.38 35.21 27.38 31.3 31.3 31.3 31.3 27.38 31.3 23.47 35.21 23.47 35.21 19.56 35.21 15.65 39.12 15.65 39.12 11.74 39.12 7.82 35.21 7.82 31.3 7.82 31.3 11.74 31.3 15.65 27.38 15.65 27.38 19.56 27.38 23.47 23.47 23.47 23.47 27.38 23.47 31.3 19.56 31.3 19.56 35.21 19.56 39.12 15.65 39.12 15.65 43.03 15.65 46.94 11.74 46.94 11.74 50.86 11.74 54.77 7.82 54.77 7.82 58.68 7.82 62.59 3.91 62.59 3.91 66.5 3.91 70.42 0 70.42 0 74.33 0 78.24 0 82.15 0 86.06 3.91 86.06 7.82 86.06 7.82 89.97 11.74 89.97 15.65 89.97 19.56 89.97 23.47 89.97 27.38 89.97 31.3 89.97 35.21 89.97 35.21 93.89 35.21 97.8 35.21 101.71 35.21 105.62 35.21 109.53 39.12 109.53 43.03 109.53 43.03 105.62 43.03 101.71"/><polygon class="cls-1" points="125.18 11.74 121.27 11.74 121.27 15.65 121.27 19.56 121.27 23.47 121.27 27.38 121.27 31.3 121.27 35.21 121.27 39.12 121.27 43.03 121.27 46.94 121.27 50.86 121.27 54.77 121.27 58.68 121.27 62.59 121.27 66.5 121.27 70.42 121.27 74.33 121.27 78.24 121.27 82.15 121.27 86.06 121.27 89.97 121.27 93.89 121.27 97.8 121.27 101.71 121.27 105.62 121.27 109.53 117.36 109.53 117.36 113.45 113.45 113.45 109.53 113.45 105.62 113.45 101.71 113.45 97.8 113.45 93.89 113.45 89.98 113.45 86.06 113.45 82.15 113.45 78.24 113.45 74.33 113.45 70.42 113.45 66.5 113.45 62.59 113.45 58.68 113.45 54.77 113.45 50.86 113.45 50.86 109.53 46.94 109.53 43.03 109.53 43.03 113.45 43.03 117.36 46.94 117.36 50.86 117.36 50.86 121.27 54.77 121.27 58.68 121.27 62.59 121.27 66.5 121.27 70.42 121.27 74.33 121.27 78.24 121.27 82.15 121.27 86.06 121.27 89.98 121.27 93.89 121.27 97.8 121.27 101.71 121.27 105.62 121.27 109.53 121.27 113.45 121.27 117.36 121.27 117.36 117.36 121.27 117.36 125.18 117.36 125.18 113.45 125.18 109.53 129.09 109.53 129.09 105.62 129.09 101.71 129.09 97.8 129.09 93.89 129.09 89.97 129.09 86.06 129.09 82.15 129.09 78.24 129.09 74.33 129.09 70.42 129.09 66.5 129.09 62.59 129.09 58.68 129.09 54.77 129.09 50.86 129.09 46.94 129.09 43.03 129.09 39.12 129.09 35.21 129.09 31.3 129.09 27.38 129.09 23.47 129.09 19.56 129.09 15.65 129.09 11.74 125.18 11.74"/><path class="cls-1" d="M.31,196.26v-7.35h7.35v7.35H.31Z"/><path class="cls-1" d="M19.53,196.26v-44.17H4.81v-7.37h36.79v7.37h-14.72v44.17h-7.35Z"/><path class="cls-1" d="M46.63,196.26v-11.02h3.67v-3.7h3.67v-3.67h3.67v-3.67h3.7v-7.37h-3.7v-3.67h-3.67v-3.67h-3.67v-3.7h-3.67v-11.05h7.35v7.37h3.67v3.67h3.7v3.7h3.67v3.67h3.67v-3.67h3.7v-3.7h3.67v-3.67h3.67v-7.37h7.37v11.05h-3.7v3.7h-3.67v3.67h-3.67v3.67h-3.67v7.37h3.67v3.67h3.67v3.67h3.67v3.7h3.7v11.02h-7.37v-7.35h-3.67v-3.67h-3.67v-3.7h-3.7v-3.67h-3.67v3.67h-3.67v3.7h-3.7v3.67h-3.67v7.35h-7.35Z"/><path class="cls-1" d="M106.9,196.26v-44.17h-14.72v-7.37h36.79v7.37h-14.72v44.17h-7.35Z"/></g></svg>
diff --git a/docs_old/assets/images/dottxt-light.svg b/docs_old/assets/images/dottxt-light.svg
deleted file mode 100644
index 567e239ee..000000000
--- a/docs_old/assets/images/dottxt-light.svg
+++ /dev/null
@@ -1,32 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-<svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" version="1.1" viewBox="0 0 129.1 196.3">
-  <!-- Generator: Adobe Illustrator 29.0.1, SVG Export Plug-In . SVG Version: 2.1.0 Build 192)  -->
-  <defs>
-    <style>
-      .st0 {
-        fill: #1d1d1b;
-      }
-    </style>
-  </defs>
-  <g id="Layer_1-2">
-    <rect class="st0" x="39.1" y="3.9" width="3.9" height="3.9"/>
-    <path class="st0" d="M46.9,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
-    <path class="st0" d="M58.7,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
-    <path class="st0" d="M70.4,7.8h7.8v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
-    <path class="st0" d="M82.2,7.8h3.9v3.9h11.7v-3.9h3.9v-3.9h-3.9V0h-7.8v3.9h3.9v3.9h-3.9v-3.9h-3.9V0h-7.8v3.9h3.9v3.9Z"/>
-    <path class="st0" d="M105.6,7.8h7.8v3.9h7.8v-3.9h-3.9v-3.9h-7.8V0h-7.8v3.9h3.9v3.9Z"/>
-    <rect class="st0" x="43" y="7.8" width="3.9" height="3.9"/>
-    <rect class="st0" x="54.8" y="7.8" width="3.9" height="3.9"/>
-    <rect class="st0" x="66.5" y="7.8" width="3.9" height="3.9"/>
-    <rect class="st0" x="78.2" y="7.8" width="3.9" height="3.9"/>
-    <rect class="st0" x="101.7" y="7.8" width="3.9" height="3.9"/>
-    <path class="st0" d="M39.1,39.1h39.1v-3.9h-39.1v3.9Z"/>
-    <path class="st0" d="M31.3,50.9v3.9h43v-3.9H31.3Z"/>
-    <path class="st0" d="M43,101.7h66.5v-3.9H43v-7.8h35.2v-3.9h31.3v-3.9h-27.4v-7.8h3.9v-3.9h23.5v-3.9h-19.5v-7.8h3.9v-3.9h15.6v-3.9h-11.7v-7.8h3.9v-3.9h7.8v-3.9h-3.9v-3.9h3.9v-11.7h3.9v-7.8h-7.8v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8H7.8v-11.7h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h3.9v-7.8h-7.8v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8h-3.9v7.8H0v15.6h7.8v3.9h27.4v19.6h7.8v-7.8Z"/>
-    <path class="st0" d="M125.2,11.7h-3.9v97.8h-3.9v3.9H50.9v-3.9h-7.8v7.8h7.8v3.9h66.5v-3.9h7.8v-7.8h3.9V11.7h-3.9Z"/>
-    <path class="st0" d="M.3,196.3v-7.4h7.3v7.4H.3Z"/>
-    <path class="st0" d="M19.5,196.3v-44.2H4.8v-7.4h36.8v7.4h-14.7v44.2h-7.4Z"/>
-    <path class="st0" d="M46.6,196.3v-11h3.7v-3.7h3.7v-3.7h3.7v-3.7h3.7v-7.4h-3.7v-3.7h-3.7v-3.7h-3.7v-3.7h-3.7v-11.1h7.3v7.4h3.7v3.7h3.7v3.7h3.7v3.7h3.7v-3.7h3.7v-3.7h3.7v-3.7h3.7v-7.4h7.4v11.1h-3.7v3.7h-3.7v3.7h-3.7v3.7h-3.7v7.4h3.7v3.7h3.7v3.7h3.7v3.7h3.7v11h-7.4v-7.4h-3.7v-3.7h-3.7v-3.7h-3.7v-3.7h-3.7v3.7h-3.7v3.7h-3.7v3.7h-3.7v7.4h-7.4Z"/>
-    <path class="st0" d="M106.9,196.3v-44.2h-14.7v-7.4h36.8v7.4h-14.7v44.2h-7.3Z"/>
-  </g>
-</svg>
diff --git a/docs_old/assets/images/dottxt.png b/docs_old/assets/images/dottxt.png
deleted file mode 100644
index 1bdf8cb61ff415b10450a9f4a80bb235f048e694..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 42419
zcmZU*c{o&W{5L-KY?Cd9jG{!v*ptyRg`t$}`^b{Wl5h}`C2OXUQbV>G(;{Tama!!J
zZYFDl%65=^5F*QS_kFJ4bA7Mt_eaguoS8G{ocq4t@7H#hU~Ht%&m+NuKp^-p8R(iK
z5G-`?r+ObJc&9wy_%H$yfw-i5!Tj#X&sy&;#aSCE`shls!07}Y+1u`S9x7hkY#Zm7
zQ;#%I@$N18-gfkj_9u4Rb)!aUmUICf>D+tKeF6m{hrCb9HJMDdZMVG@e$|%joh)js
zV|el<#o2gDK)7{TU-C)LQN6>PTtZ7WrFXOE)Hc2z8VcxD^L)24)@zV7UVigk&{?(8
z!7h!Vu<IKE*;;+%y5Fyq!sZOGZ~3Ju4tce72A-+&{j^?pcmL?$ZI02uilM{fJ~qX*
zy%UY#%}?w1C#5PjUVsAu;J{w(^_eZdDy0pd7AC3n6?pg4S+Au`-inQ1;B^eV_JCI|
zF|Vap;B^Q5dv?9JHX9~a)j^a6+)!j8+gM@5uf1QmA@cuwhrmS#*Te>1kkDD^OJpIq
zy#M(q!3{#gTc62D@PcIo;Q#k`+#qld|Gx46KlHy&_wV4^E2aPW%>VuU|99?x-~C^w
z_}`z}2Zsb-xW4to-XZW_WKOHn&S#~t^&ah|GtR9ud(SiO?!QsFan&cJ?~q%?2046_
zO!iu8eLwnF?fj}*(#?t>6u52X&-L9uNqc|(z2E=loJMtRc*fZHL7$rC$e-)?H_73_
zfuZNq$HuKbZVha>Rc?Igy&o7Dx~EWG%iB7%X6S$27BhFJZD(U5{P(|Kc&ehXcHR5l
z58>?LHNsrrz5gB@Qm-@{9e;mscQt#WbA3E)V{GQsv*wZYFt4RM7X~V>jo#kOdUJkf
zPJ3e9kc_DsJ-IiyRJGD^GN`WB=icPOfyE1?)i0%P8!PG7%7t^}nL0^6M!Rq?Th05Q
z2jQazcSrB<!@WUI>TqCvLTf+9u|gwPqbu%D;N*IbFIIm2H%7PiL4(|{3DNsm;4{x(
z)&}k)V(3rFOQV{!L{ib$>@ZeB)3>r|fd~={{O?KQDwUzRu_{loA3A&l+fnTHS#?Tc
zM2Ph|H{=jMK<CD?<Ktpu!O3zKQ%3*&02-Ev#k|v=3>iBYiJ|kJYm!7aV5d$U5@G~P
zSLkpP@ZWq5h?tJvTbj6kvm~<c0xZ<shn8UI{YVL8BZ%ab)o%(k$xBV9EPWAU1Rwss
zbCw-H8&f`gE`YKCrVgIkDSt0`pR)))-p>Hn8XdQxh%dRAKdu*?jf>NjnVJk^$qaG`
zZvDVUxQ=CL>?7Fn81fq+^#a5%mG0+&Wnwbd(UMUp4t|bb+EAKFh15k@h-Gf2=Kp@&
z5wf-JrsED>hCiDXp>0XX_U`^pX=SW>cA3wB?32o~Ws+#qV|~r?U=Hd~bbiI}f5g;t
zCh0gqU4*|>5#kZCaUE9d(Q9%;hc3dZBja1aU|K^I&wWt5$gc4`%=kJa9gPt@6=Xm)
zER$Lkn1_w=K?YjbL20m({#~yKO}F0HAOK3PZlkJTmCpOiYhhd7O$sc)kH7w)o~BVe
zzW&bwXp|&CV&ULm(Fe<F82z4OSXvGH()iv4)i5?-a5P*#&#);i7o2nJeH4fF)HmOF
zTu>X4<V2DC)W6A4sJG7U)KLq-D|`YgZCYJn;)8hhwMW^zb8@|WVk=F5K8w}5hekTL
zlEWG`C0%E?3%R6D(>KOL%_d`%?m2mcX0H{vy_&tZ6L&6Ud`Ky{`K0M+HOHf=45fQ8
z^>>{^_|DqgV%SpXMEZ+WvAxxBqooD2+jUyCwc(46hm5*6gARqC3A=ZFEzjyzWstkZ
zv$R#h-meLCW^m&{zsW;8G+Eq+Tf6574UIHf^26*r82KujB|Zha@_9L;rc2gFl-~y>
zG>+Zbsamri$rzYeP3LX)*%|clZmnl@xNO?jog(_r_PnatxOFXbWPP>6Z7}Jbi9>kf
zp<Qa-^|j@K>-ifSuavs4%NHz}Nx78}ly;w<_f2{mz!%#(<Pho|61r5qbS^+@>F)l8
z+tqu!e?EA$-bJ0?qHRuU>&g}^T?@Tm^K1V6U$xarv-Lh~ix|e<z-LK&V$*2jUP;!+
z*@?ZsG_~{KwgWOzyU$I|ACmIy72B2)CH4lLhH<lOgfP<%Dk0tuk-2?uzoVuLf(~DF
zC!olPOpY7O8aqm9L<?KybDSI6Zu;*toZFbXyUTG#J8zAqPw<W@3w9ysXj0HlmPTaH
z>0d%{Sy*|N5(jqiPrLWy)_d&O{ogvU+_7M3`Q5>}5xi;>mxU;+GyWLS7$#M0Vk^v0
z9IYD1UZ2biZXLvq8%FkQ?J5KY^FgBV`t?H(nbLfIEpW}j*X5|p*8WHSHxbwc6#m=&
z<sZ$>;DXFN$Hf^`ni{=MQk@~KkblOZ1C}}V#=c;I@k(rqiII4R$`lKEvS5yni&n8?
z%{-NT>eNyn3N0aZn;SAoEju&8ni=SSLr;3X)l-@7C_>TE`Svanxx-KXh+$mflp6}7
zEOcIAA+Y<2PI`QTbE=(}tvi`E-;N!s>|h3q@dig`DbwY-P9*szGF}A{pP8Ttv7ajT
z5kfRijChtxW^Rl=OR`hL^4g;*s}cC99#cx9-^j@F7;C2Ji<Drk#Wk>YwEjq<1!g=j
z_>x~;`LnY6Zn8^?oQ%uFA2Dl(@o_&rJ0&w4$IE#A(exWSjfyC=%pVhRg2<`<`<;pr
zBtbuMXOV@CBiWU0%wVGZC=v_VJowiMC6Cfpqvb{>EOBDf`FXP%4C(J>mRfX!7<c|>
zOzQXVvN-)%g|jM`*=Jk{hAKQcNJ)PvVaJ{yjxsRlh99TLZ)pzlz^Mn*xEZ$;dRXZ_
zU;vA?u-8;xvs~+B$Q}+OYhfR?2a6G3;bxCd6;yRgW>yvw%Oj1ndNdfpRGY=rB58aA
zrb$?Z=jjsR2*f=D%Qz*KPbgj_8~h<~V8x&7t_1v>6~(x?wwHEOumI;=KU4alO0d&y
z7RbR!XH=@R2JvO$6;9ti;s`~j<9J{*d^6ARrLJD6_f?A8pc>I=x5xG^hFLuQl0@Ps
z1cJW)fRB)4`un+aO0F{D3lemM`$kd5ohwlWQRf9|cgy`9YKKN}Z@k!eppl8^&LU;*
zVKVp&*7mh^0kBZ6T?(#Ma`O2!hkJ0N0wjKbxFJcnj&vYKISAMdGWGo9TeId>okKjQ
z;F`oHKKOA+#nxHH4qgMj965YKs^v+1To-RX=4_0oAw}o@$H;lrkv+?4j%0>h0d0ZW
zS(o9i!eeS`{ILQbe4$=f0Uwvqq__YxqU$_ZVq*{3y~PN&IH+KCZSo@T8B!7(8eY!}
zm0Np?({)!LSdfs%X^(sVU4`rs4>Td=4ZuKTVK5&XMhCU%F!HN00bijvc&SE|5nQj#
zf~}xbS-`417p#ixaB=ZIgAl%?LjM5mZfk{9wE!(%>#SQ!+iyz1r_(|lzKZoBq(UV*
z4mfKwR#P%_XM5Ic@_1vy#uGAQGSOlA;qQ#1d%u~B&5CM;)!$>itAzq<zUe7l?hT%7
zJ@<TkXm46BTs7ccTyo~>mRy#>(!%CJpDXAIr|pHkt?Op~yZtkGnd+kOg+aT%){TWj
z_mnke&u@3ReQw?O!FlF=>xP@p#~rw0<72ke`3;}FMP}OfGuMb8*D@#4#>evu=^mpz
z$5mnrxrUCm4h6_-y~1hs&m3O6Z0kOm*FE+&|ET(}*F|?%E)8Ec9KSVQ#zG>#xPyAY
zqMUape`B+5V%OWDFns$D{rs*=+wDu~8>6Rw^5l7&bK3nRWPQBle!lUKrsWbE$ECtE
zn00<pX;;QF=Y+y7#QqfN_92JeKa;7-23n#^o0${78{MiJcW0M&F>m<Z?K*rdHW++Z
zc&07{5zu+>i;GuLc-RX<?e?_Nwb>&+)r%Fh?1`>kQufx{Y}*FTrJu%r4<ku$_01xY
zi}&tq*Rev$52`Ep)hMf&?H|D`*uf4@UZ@?pUyu<ToU6#7iaBz@IZKsU;k*!$c3Fdy
z;xgYjKPyZ*a{B})m|B|3j7TkWZb&}&S_G-=)>B-o$huD?GC<u%92VF}ryl>saJOVD
zeeV!j3&1c|46Zp{yY@g`oPa>EG4z5OpDR0o6Tc`P;pB#LyZhVV$6FUm&ayG6UPg6p
z4RDTRx^)sy;M4>+jqOUs!kPwy>Zs-f6*RWbELFMS<F@qkaB;bE2PitXk<$?oJ*Qhl
z8Pb{ujxb81p5Sq<757dMZzI(94^RCMsFufz4U7<pJ!X~S3t-O2U%82kLtHyRzcFJf
zjs`auflLcRl`>rWuPZu2`w&-AXfwWd2($<J6bts^r_+4g5b?y|4X7T=3+JRaD>C$u
zIL={$GAJTfB_-+hEU?V?#1pD{7s(O~8?T~&_qz{feArWf5`I57L!pH}=N?MOajAbT
zh$P+m>cNV3{A|yLK6%nA62l|ML(viMiDpWRg>u|jkR<dET-eQ2!9KRhd>4T%8O*D6
z;f30|*|?#s<WKvcT%O92!UouQHA<XeL+oLz!-K-AtU)>HRR}=Mi&#vu_VO0jJZt9D
zUMD}uZ-*Ub?B3>KNNbgHGlIX(|AB0Aan|HVxVTe72Z+cj-^ZgjADTqeJOTL1+ijl8
zO6ZT=No7haNXz1V>wLJOw#^(>>{6)6-;fIkw8U1BKa^~;%0o5mJ^B0q?Ki`f3;Ti@
z9AquAugecgcJdXV>!wccgDC&$&A<W_6AN;(JbN2ckBH<U$iy&NVdI;cEOdFVx00SG
z_2c@{MiIzm_5)Nyd?bn$EkIYqwFXun@`vK-=J!5|U7$Q@lGjLImB>BCiS6e@vtYUX
z8dv{Oq+Et;Q{gc!?6H?_DPp<^bz$0J>LFh$ql33*N5O?1`-4?71|vA|mB`eS{^5Zh
z{e|ylN>AkH&~<NHH!x)P6~3=Q+!_SKFrsULKr7-eLC1=k<>PzvzMc|>0|n^Bi^F-7
zSCy3DPTK(J=bgwNtz=o6ORSAPm%Ta<&1KfyAEL02n-XLi6_gMjg$${%J81ObtlUWC
z+;y=$OyuLTW>_Frvz<xRydt9?Cpo#bd?g?uqz>yWawB@0L%R@gii3@?$IX8X63qwY
zDu7P(?e>5f0Uw;n&Co+LnpeSNtzxaZ-vBet7*Xg~)BxtSeq?aTqkHbT7(3OLA2ybq
z0`1{hp7}I~3Fh-i)}>&UzBrR9*w{Eb8k4-*jG*Tp%6ZO{nu`enDEs;?a|ZzydqvxS
zTojo{TaC4Gz<qG*&+F{U{9vn$To3LuC&$lb^FV?O9u=O+?pt{n?$=L`D@8QP_i$*P
zKy5grA0X;~q~$rD;dx-!XimUSAH)T-JfpF#*6F}P2gVs3lcK}#?DgH=TF_QlZCr-@
z!AO}>4UhIa#$q05TPQ*a@+B>jXlLBJf=Y#Y8LPxJ4%CPs0f7#^=SJ<!4JT*_T&k?-
zbs-_=VlfVt^YG)0&-<X{<+9T)wrKR0U92C}5|q!9DR#iPllSp|uI0ZjA|R>wpXp5L
z*}$_e)h>6B9IaoI1gOJ2m}R~_HxI+Y@_<ZBY?7v;nv*?;dgQMNt`V7n_sLlt*5+)k
zAq3X?2jg0Cd4{@cb9V`!iKJV%%)z{wxKx!TN;6L?UtxW@upQ0R6ZLa;*P`!pmvbP8
zb}&_D)1T4$8dCnT@K>F@pq<J*5fJObLhZ5iW1u9l<hMQ_a)my3ckmj!iLgLzzFBG5
zBUC3>J4Us;*2gp&TRjDVcENsA#U4J2S%6X076pBMUD+EXq#~NyEI;61LAu=nTU({H
zIE%b4YZ0QkZ&o28IiJm{?hj3*lSMTvd)0HV(@Cn6KL|nIp*5~8p67O#75^)qjFSqH
zKRdhOW3=9t(TK9$zA<yJ*(K@K?Dp*b^BLnqhlaO(0>^)5Ol-6rnbX>vn+Y#n-TiMy
zGdXRQOY`HWpBY8A|JiN*>h_UQo89ob|KN*OoFpOediL&M<-Fbx!T&TY2{~txY5`ze
zje1rRy2PNh0joIa^9#s3r-LtEh+E!5r&ssoGVId#3cJUYy_95U&_=euk+yZ`zV1{9
z%jp=a7W1dXRH0j##skX>AB&x=+{38b$XuhDfxodW`NYY4JAX#_L$6lldF>(##9fD<
z^5zWQJ)rOi6~dv~JifNwdvE1Zm&8>fz43IqU5)XJ+Nf!%rynaPdQ_@fI3}-jA7X<t
znS@xJn7F9t`3eBZjpWj2K}(U&3-OA?%zNC9hSVuK7$WEri(#Zlq@cNbuRcX*1_E_x
zra*tVcI<;dCvwstwUX}ujAI%B2(!BcsKskt3M+Q-1A-MhTa7`tr1h{#Go;5~3c(D4
znTNz4UakGkDQ!dVXJHH=zl&v4TZwD!!qcNo3o%S-=eVs+b~yfrCpYBYX!wjN7&F5G
zr-mscRpP{gPfN5~yFg6-t2RkI22;($&QRMhD|)rovM89?Gn&Os)2SBNM@dYj1tv>H
zA~(k@M=eo=8PxR@T%5=Ya+L|b@<EjE6er#VG6@Hbjx0#i`T2T&?F5ZYJH`+{`$cJB
z;U?A4yrobXdo=pA3uF+k8N<}$m>CJY8qqVV@2Y*2A^pfsj@WZjRTGo)gsC#tE{79a
zd?QT1AyQ(8kGs6q><R8tMU@V}*4wm0p-*(A>Rrcu@9+xcq?CGn4D@1&jUWJ^UI!)x
zMhzb~xc(t1;oDdVC$w*vL`0^wwlhnD_Jk=q)$TsK&n3}j$CO>5i1dW76)dno1d|ks
z!UTXWN=RMTJ{XCtfkUZ=O{SjtJ*$4u1%tXCCiUFQ={0d$BF`8aFIIm0M7UB)x+&%7
zMGkCY?X8BCl@Cs6^k*{0l(7Db0}UVMl?hfQ6kL;fjqln=klB>9AhO1e9A&5|8iIE5
z<D&g6JUs69R1(co$jOkdt!ZJ%o-j3Tk~g4Mw!@E~u*WQ3!iyciCyD7MGFcR^b%L%@
z@`9GwagMY%NF=onT#a_26uUpCtwwau-(WCU`KoO4LD5$jk1R|yIb0c=%oIEq9ssNQ
zjWRH%9wcPfb$(pyhwG_Cl6x491wF4W6pKOE_gy5%mk&G#<FMAU9Y8|zzwP#e43MT>
z*VV9Pu1_WCN4+|aGN`qzazvyfX>HpQ7Fs4)tj^U3V_hJL`giP*gH!d!L7fFSBt5J1
zLk*%^?TS%WQ*#cIpn7|)c;2+?v>3qkxRqrhGx8e-6w9WqqfmSNZFZW@i_addZ1Y26
z(HN7Gqz;&IEUz_OQRTAv<FHGym{0aMz_S^R!6{=ay1qG^JA6Ctu)z+&kJ#pkpS6iW
zln4HNYATmsimKteVS(N@ahguLkp$K7Sko(3bZ$*SvpoE@6pJ-ab&AF?Z9j3-T!#HX
zH*I)*f<b-Qdje|jFaI>n(Xuh78zBDnR`VpCgrTdL`9rY16eIX@C;Bah=9l@4_&E{-
z;79=EL_+@7?2kbD&Sy`-kFi)BF>0#lCdcIA!~qmKMDv^=jl%B|Gyg>6K~{>Ga?B(v
zoVx2|9Lz$c>!$U1gK4|%7{yG(>2E)u$6}klS-%6*sJVPK7gXP$UvnM5n#tQGCB7}4
z^#mfV8ur7)sNUNtnJdw+)Odajs|eB#&z~t#l+_oT)+`G{M`HY1#1_g{Sfee{w6KiS
zDrI`^bE8qwGB)}>sRWan>N^$z3DMVtiMOBAbnGuS`CP`*4risQ4ahbO@>5)hb#Avf
zkM7y}v}R?R3|m^*e)^yu_>0=vS>q@<m*o2J-w9|nh<8fqD%ewy%`_}<P<cQ+M&9{3
z5$WxB%Qiyr6^9U<y4i@u#&PI6i!3#Gf%Yb<ia>XG-JIqS)X}VHe~-{vx$J^Q_j!LR
zM(B+Gy5ixG_L+4*d?t1X1*f(}58W*nBW|p)jvU|4Ik3U?^$aM5Bk^74WD_w~6J^rx
z(|zk>im0GpT{J8<iSPncC$AGgRD5IR!#28hc7?8sF+ahOmcB)EJoDsLz606B`!y&N
zBRp?=<Z*ESD*eh>N={U&JoJ{^#8Mume_4}<VY4$E^69P>!8Ep&I^=rT=xt<={|Oaw
z^beNf4v9_jR5hNy0*iL0bk@_?L$d7|<qX*^$@#_CKdYySQMof$+)_AR_b_WJgQqzt
z^G~=3UO0;wf4ecjl>Y01zaC-q(*uqE@%TLuEyXox3l3y7TrtU;wNc}pRM~|R`pWj-
zjh5B(8&&-o*Z1CpzsT};Uw%3^L>;od*`s~_H}!YM#9EW^T-Z))_AV3x+Tz~Ikwcey
zH(mChOU>HQ3g4Z5tHtHm(-YhXTIj~w;@;xNS2Hzh8{y~A&XSQ~f=YyrN3-WP;4ZbS
z3rR(PUKDB37YeIQC$mC-T}rbWA{M4&gD8KqCx6;tt&APWgv2!MV(&?de(Q>m>_I2V
z+R1|li*nvM=4~2IS_EE^D3S@dJ3yx{d(7x&L#S*0->I2Lr!t-*C*}M0gM&=Y%Ek}E
z&lMh7(J-SDo};H6^z6(NW?wCz_&isAPHMCz-+?Pg>9dM*Kv`18PpzQl18)=e)Ur3b
z7#R_Yw=U`ufY3WVJv9M(1YlqIB6f!Ke7hhV!FxX}nSmu9;c5B4Bno(&=(KYrA~SF$
z564yU;5aVsk9kK}77JSXKSzb{^R0H2)rU2i0ib^+I@l})qJmc5urYvO*2#7iOQhY;
z=V4Gc=Pt$h_I5XIDelL`4K}oAj9#SZjNYLJWG8>WV6N{bem00woe|N~sgoe=4lt@#
z{u-SJieNQjMSC1R!;1b)(qoBch1`4VxEazrEh!&CQwlV*qp+=BUHPMSKVxk{7%x`k
z6TcLUz)m?AdB<@>pRSs*VDTz86|OfR_pA>RYz%3!z)kIR1lqLJ_M-z^=Irep7AGA_
zXCQx4p<1rmLy@=8z%-yKS6CKqir`yMe8N+7`lE5MkYY=gGPcaJfR)~MNpEyEnJKLi
z>ud+u?$pYN=MjdAQxgx&7?fYT{)`#fF)RV2Hd4V=KAf0eD?LDS`7+f8BLv8;6i#eQ
zw<{mvCbr&_h*U&eM?lG~CrwA1r1G))aR~bjdX0SqviRp`BL!}VNf`de5=(eNQwT`q
zhIF0$&~#mGw`wiyH`Z4%7-qu_GLwU2Hq9>{paG@20eM`U@};WVyil$<w3byG#iUNW
z<$-jw>{!tuYM@c`nUQgE>nV0^aO$1&qtC`ee7a#e>J{6VBy2GHMu@KaWOHRxml!p;
zE(zL=ytRYEO)czi)_@FJ;>F{}BuvL8%U(}>+^Me_%GjKUkN102A6+70`uK6&gq%Sy
zPyxzGJOB`^+mJ}7!xdC`nqcFIr=w;ONGIP%k;vm3PT@Y|d&(S>0EqQa4H;TgQQBRt
z1B+Z*+yt53Sn!x$cn&TAnZ6!?r$r)DMJs>4`~|b*d*&rr$V<p>SCw?g1AZdtb{ZFl
z%_QRDDvV9d$zFv<9N43+R}q<=aJ<0cQRvq6Y>g57Nj0o)^pEM>={?FLD{Wf!R|_)5
ztd0|kul@rbYGAM>3Z1u}!9swXgv5kEC5*^a#?}+RI799q)hq(IB`%O7Pv^R?zQA$G
zpD&k0hs++Df%%M2pEM^E(-Ys)*>i-2VB?^tHt>@K5lWt+_fChi$l$&+TXpJ1hA#=%
zxagg4NtpL!(+*z77R?40Qehrg$dSzC`T!nNP!lKcZUltkGt(IYw8M{1l~0}pk05B9
zg{JfC!<=NMi9^JUhq(%mpz2x&uRA_&O6Q~skMR2k{*c3;em43u9kO8riQ9*?NY+IY
zDiI4fO%Ckqz2TDiCsltfo8{fd%$mVE>;pGCawuW3xPv!Gck?KeV6B|&oKj3eYOHXe
z2<-Qy8Co7!0JxFzRKx4D_G;-7qt<)V<4qn5i9cXUS2zh%T$;~i0g>;wM*-l+{l=EF
zY1oZ_#NLkhBW>>CkO#%(>$02{_D&%kFtb34cV`<cQ2yp_wg`>VJpXiswM=D^bzqtu
zPOUk-4>BepZ&8mrlmb9=+KivT+VI*xV*c|BGgruw#9*c8rZjXhr3=I@2p*i>{1?eT
zA06*g?eeI}E2$iR{bm;au<IfNC9F@sKoD}O-W1SWJVyIhqS{*8<&|)rp}8xpt*m^{
z#(E==@&e=q6XgI4vTpCWTP!WZdT#?Dle06+o^CN5)<^H=#(A=cek};b&5kj0d*t0O
zJ-*2jYaOY9pv%`Jl@&M_e7;OFqG#Vlw^Yxs_;oW`%CQm*YLL_I;K-+k0OWYIbnPO!
z>+7?5Ri5^aFpEy!oF6|X155+jKj?nonEbR($!xcvxF86C;s>M1%*f34-s2pT217`L
zfVr-BTC~G_BV9~Aw1zr<_PFA~5Q2<?qH|z;{Z-FCm+4t7ZFFy?p!+mgqBnG0*p4C7
zDh5Qm(e*=VDu}T=E3Pe-w=qp=S|2z2vd@LgYEPs|XnytYWu{Rz44t<VH)`ZGX6=@0
z)XvUU?zZ^6+*td1$H}lKJ#)2XqgLtcZ2qr0?Xyc}TnK}%&GL^Mp>q>U8)vrMw`w-S
z@9%Ax`PT*e9|0<<&u-kFGM8%6nvqUXr)*14?&DXx3f<xyZ}S&$mDSUxTR8H4vm4hH
zi#&ZH&G3>eVT(13M^;B}z8q`kS3D!Sjrt2uhLgxO>=o5d7hKO{%d~PwoFuy{pWTPx
zGf&5pKdzjt8F}v;%wqfCp}LjcUuJ+x$1QfO`OYg)mWxj;WtmghJX-(+x&H^q@9sQM
zbf{t2uYPok;PsZR7H?Kax6&uLP_^!{7!TuCu$Ge5(C~tg3G+RW+lwsphcZo!glRgR
z?7BnFGjUU@ScZDlFXf)xzreOD!7^Cttw3$-zS0J>*7x;Ts1Hr6(hu;qlu`{|$0UE<
zkbrZtpxUdwm$>LKY58v3)41c8>#Tbd)T0Wr#3(uoul}7yjG}hGk&U79?#BEEAdgIe
z%8Yym!xIaD%rOc%PCgeqENf6ue{}{9fmh1nFp>1)_dT6VJ-NPSRX63hb%_pm>J&%`
zxIB|YyI*({gR$W#=7u8sE`^zO^lQHMgHA%IUUke(hoOsPlSO+j`cnnGw8VuFA|~&R
zX(STzkOzu-p*aaSGX*Vh#L9CZFr02n1@@?BP6M2J8zIKIR>XNTt9_McjMusq-2xU~
z%}6J&5id<=<o!OpTZv}Tus<i1H5vQf;i1Xn;ZTq<C;*zTQ-oh*?psXq?F}I1J{*;0
zs0>;g!=F3*7VG=P7rQ=TY1aJK)u`#R1d}M4J6GFdG75ksQ~drjMg*AAuAG_7wGs)y
zsC-xlT&D&nRk6mWTSV5<0SmS(Org1~JTysTX@J!=Q;^V`XG|@*!Z%Y7f6R%M*Bt&#
zY7wOAF8q7AK+yTD$=F9w?RIv{>x65#L8&Sj^?!DNR=}t|?0h@7B=Gl(1N5I9lW!FD
z;~?q4hsI3%M&NeXX9BwcGYzJUbtaR#?sWOnX&1t&hbC>ee_EbHw+E*I)1{0e$k5XY
z*L9D2|7pDTo1YLZqWct$y)wux4{)3;Hw{{1mqTdqg+toHv<Sbw+?x8%w6}SkZQm|r
z<o~`phk%}^`8WM15kgQ^49XcQm{o=5Vorvl$8I68y(VYx;4eIBE5*{~hm5-yr{Nmy
z5C4Yf1pb1bTPyH#8Xx>CwlR3%o+F+zgB6U}KE;mls$P_M`S-%ki!wxFMByxK<VQks
z6QY6=UXMmuzLes2;^S5RT?lmJL)!T?|Au&n;0rGuEi+9b{98*G5*tDR<SEf=-<HV7
zFe}X5<<08kR0DoZbG%`JQA?NV&-WVgr4unBm?nARCr>W=IZI(0g~z}I(6`veC|~9N
zUV!MN2nDs6<T^;(nDCxIka>{1o0adl#1bxuHR9K{&)#~4x6w^|d*#4$vpXAy2p{f9
zYdHLO<h-i4$wi{)_0<>sz<wFX|LpIuGkHF1e8_FD>DtTL?w-xX^S^cGv_3Aj9PC@Y
z|GjtD%cI9_{QlnrtQ+RupGLd-4x)ci_zpA8v+&ZRh3lIg{TqIhGup0(``+)84Xed|
zGp4y&=#g{E$8ufAUZx!}6@CMu+@7>fFINC$yHDM{=3%sU2OZ$%GGnW*>p>{D40KCN
zzKNW>E8iolv~At96_2cvdxko2I=U-oUE-Dv?5UEI)a4VV<^_v;FTpw3*Z%n4s(HAb
zHI?BqX`%j`8$u=B_`OuK;b5d?^YL8q&{VlAyx9s})IZ~?_EAMW`mIfb^cWwOk%^-u
z+76Scfr^|NS7S}UPro=p!p#Qgp6U?gq$F+)+AN0)!x5*BziRw2JL1erFu&px-3q#;
zzvvN$^yLNw%t-8h*?8KI&7{_Wjc`<lyr54x^0N$x@krdrDv)`<Ut@)9e#|Q}Ze2A^
zZdi)Ji3ODFcfX@~3XROSX)CF#T<&FN0`%<1sfHIbNH4FRWPyhh!-KtVKz!+EIO(|y
zksO?M6yBQiik64P0eU&FitQb5cYF>UoQcv7xPD3Nshaq%QACgDmzNPZ7Ocp(c@-9{
z`Q!4<&_O3Cd7ytEBvmglYQlzYbi4v8Mm5%4J6ignD*Z~9VkxHazcakb-eF2gr@vD>
z<6UEj7&zLI1?{(?!G>kG=Ku)j>6-sD;9K1I1lJV5ICDWCum97f8)g(2d<DwYkt7uf
z=O?Qq@ZB+agn~G_&&OMou4~?UfR>0U8_Bo{3+)@Q#fe33x4|g4rw1ZQJ6Xn}2LsF!
zZr7DX`6f0GENC+_1t>Zhb=@G;@!bMv%!(F1h1MaP$X_k!LlLHWwc431YnMj?m5H}G
zAp>{IU`!;kD)&kRh7Jqk;--=*j90$G8i>p{zt`4-St0gsDUyVy@v=ozpy*r8e+1aB
zg--?hPM5;MQ29P}#l`yN>gg?Uli&;BxjR!z?_W{Fa|K!#0<7hk9a4K7&s{-9poLoh
z4P;`IQU^fd$G%ulbQ(eiwm@MG&sndH!_9gbz7D)4O8;#ioHFzZ#ZmhwALs=<*B+Xr
z4r&50HYx-3j!i5p+HSulF_uWuc;2&{HQf8mSwvB8mc@~aE-XUVeOfJypG~`<?M1O9
zybmr-K*`R#SdF2@R4$hljnoJOneCt@5qZmaK^4e)*4{~P=tQhy5@77fCyLzo0^Eqm
zxog(7g}}@m`YQ+<pL{o81uW@Jh9Dd-s%;BDaWw$vk5MWgl3+ngoYRlsx45P6%C!<Y
zKg~hrW>C4E)bzQc5DjnvZ|n;i+scV@z`lc0okei*GCJfdk;TXHEGb#Gu#mTfDkwEM
z+_F%@cvTV*zt1*jfN;?AIWS5FZD;`6w8zY*Ih^nF!ZIeUd~jKVkKwZ~2w1gc$t!4?
zct>y7XrA$^d8X_f@ZgOO({%XF8X2#iR(_5^)~W$v2uEtpqv_C}0Wh|2_&Bh{dsTQ~
zzHe_MNozal(U?m&<Jy|bgXNq>e57;&(tL|Fo8==W@&Lm8>HCDXO6<xE29knMg0Bvl
z=?>_gk<_D_!?UXMvL4fLNZhhn3M~;E(ECfs_0#FR6yZF>V?G+Z1Q{ZbeUH8oQpFa#
zk99CF;J7-*rb92os%9D50Ka>?=ea~8($jzj*kh7<8L0n4CjO9nEyYxSHYR%d)FTjR
zs8<9z4@0>C$X5K*LtNZ_F|Tv{?{-|zsMJO+WAIur0g!><jBdas$hy#i9B-hrtFZt_
z@y5yG6K=E$yNMhPKPm-?qgG4l3Tp%ZF&;ulPz6wQEfNfN_0s*p8v42!DM?WO2yC0}
z*j_LZxS!@x4S%ELbjYPqzVGlos|EJQp){3iIDcPx+G+>yysg&DG78%mw-PGp*KRi!
zi@5ia0~I26Q<Z+${6w>S0sk-`EaU39X0M7>pVBBnxY(8=sD?{TR{oG^r@(``lAF)P
zB80-lftKem(mbGm+JD<&KFT-R1l&r5cNRNVT~d>co?G67j6^zZGlNDiQf+48$E?fa
zNxxEnMW^?*R1)p~*%F`^--Rw#zII=EoLJ1F0;eM@y8dDTmJYvEi$t0eZmvk>fpCM=
zC0=DQ_^~MxkQMvi1YTuj;@bfx^<st%3Hi45sCZ_^5IeS4{MDJ;a6o%3!NjomO7O)g
zV$=x?s2Iirvd*!*F_=vI=Rk_9T|DM2qD0KSOUU`IQ-y)>z8HHAnv4Cxo3DFJl6}h^
z@@YCt9#@Mx5mV|aJkF=j@DhZazPI&@i_#(*#EKjx(N7t&w|;u5&<(s#uhNMxatD(R
zs?e|8eNtlWvaRxNqdx$;l5l|wdwcV`=HzbIhdbXJvcLMaqV?03x37hV@JjvsGuX}H
z3-go3bvl!QFY_md{BaxZ`cnG)Pt@$$B{(m?@baUgKcV4D8nen-YvtELHf+fuEFh#u
z>2ItLNHX-*EraHIbGGJ3Rqxf_&1WF^Iu|^#p1o@}zwy_9|Cz_TJ9RUYhpuI7_U?V1
zHT6FiswVfY0(jEzW4^cX%h+0ei!oVX$TrSn8NQtwUmgVXUDdpRL5*at9S9h5HfO){
z)k!kIhrFOP*zHhzn@hqUg}GC^>2`>}1;jAP9SdJ8oaTz1O>2s7*US@&%U3UZ*F4<k
zev0o<Pv^UOIQ7h#k!wg3FWS(t&VF&HKs3CLfJ}HLyzNblr0Nx{15x@<cRk{Kb^EE_
zX8*7WC%rBB20Oh?P#%G8!CqNZM%p%9jW9Z)L?eCY@8o^Ys@^mfr(wW2cWptL1W7L8
zD)sSkM`z@4tp=G%;^?Q_MXjRn<BR^@cK?7RD?_FH*$G<WO7j6miBrwW{R6;@i06RS
zCuh*2M71mY0a{L1GHAWbd061-EE3XYrxQ~^9IWiys1sx|>?BG*`ohy_yf3masRh^i
zP@6N-#9G&<xS#<>sq_Qyujlj#zXv~L`TY+m!{mA3e*MdrQpW>qCkNV^suE3*5loei
zQX3LxUZ@a3Fi(8Fvrmp{ShRb!+8scDnlyttKaqgF7K@=BRF{R)Ze3+#$*e%DLgH09
z1>XYFmQ(e@oa{cl6af*+rwh2C!ywambfBB5_3h0xY%KOboFUzil0{o^k)^o!8Q78k
zVxc*4S@~qa<9MfghqDv#*UK2PwhqP4lLG~tr#!wl7q8|z+Vk4k&h&s`0A9Sg{#ugL
z4Ty+(1$;|sGYeuAv{<PBZxld{O!7HNHGHWDEJ|k8YS)&%o<v^`ZZ^Txa$dQv?CdpO
z!Nh^uh__Tg2i+(-7ZFb-2~)1Q%Gj4}_M;>f{WuY)jg_c#fsbiAHqQWiOTY=zip6F?
ztHN4MRh<MSk!=v)BX1DGCiuEomg`FVun@=#Oh?9In5VSWC^}!S<~=oVfds#Stc8b!
z!nJ>?+Jd9_3;c22E3A&zTG*EduBOsT2Mq^Y`|p->HXc9`j-7As|E#AGJh*qMsIvz)
zGC-qEo5DDRH>^<TNpnY4Y=377fY2v_9G`Km4X(#*0G{%*@dlL4k;MVoo;E8Sja1|D
zEbIA@M|u4@SM9P-O^Bm&ib3Q{RczDC&AYmJhWin6J@O_W-l);ft>@DFTV6>3<KYyD
z2zuuLrH1FdB{10|0%CwO{2}yht(aL0<(U|T#x|DK$!pB<`GBRM2jXr3iR9~?W)q>;
z(o6{W(n><1ZTkWanmXV>z;&*g7cvQzzDkUWHwAg%v~+HKaJ!Vw(u-UbZ1k4^%4i8H
z#zG|qH#@~Tzrs40-(x`t362Ayx8V}Chql&|&%{m_bea+x0wDKUwtUC0o$a?AhYg4g
z#V2Q8KRABs`U~0u$7CoceJOlrA?epv=ES8(MSFec$D+{Mb({X#hP_${!p$D1d!GV^
z(3w>$mqWH6746Q`P8NQU(Wv{GUv~o3pxv2?-O;yNAN|zc`=47|+Rpr4cYnDwJ2OjD
zdsZ!Lt>AaYwTXX5vUd2YrrE~{>O150jlUytQkSzeuP=j)Co`=#JnK;SpE!ND`@i);
zit8YLgXnhVh{pIbDM&Kx8{47bxj&C4{_xwvYJ>VxWHE{qknRKdAU6!*$Gwfb>QB`v
zQdaT5hZgn4qR$(~EnoQ%;8pYRx1iY;le(F_RqlNr+yBnN>j6;FO`^U3GFgs&8VRz~
zbZ`wj(Atteeg^NJ)}Gn5#SC^mW`rsDr5|U@ct$<t@QCgne;*<q_WY^*Yv652@}uu(
z^g}xaifU&x=6IJi4pI^y<|>NURc37-Dzdh#3wJ;JT+!mch{Ajf$VW5&quo6NjH47R
z3%%_`SFY>{9C>-xT${+2B+H<V$Ce@|Z!~tXz>l3Rs`!E3*?iF=ZrT0NJMR||O72B6
zNZp)Xy8?9fHTRrlk3lpDsk}@NS*v4#RU^)^qC*Rx&+;(_dJh6PHmKuOhAJ*yaUJpJ
zhXpPeIzeEjiqSxhJ*WD9N#RY1NX-N3Qs0EnPLj*D`4bC3hMyS#^1I&@upe8?3&=#w
z`^&&bT59kt{>=)B$8D4}o*pI_-zR=pb`KgnLesH`1}T=Wr)~lx9(y{5St7J42&?|h
zXF-2XaFk}0L_|jj$^7|m+_QUHg)YAztYtV5OihrdTytyTwot?3^|Vy6zc1RKB?Irz
z_Y4d6WqI#%t~i?Y=h_6ghA_Z_boeXXfa?j{>}Q7y7Sq6eJpluwiyrw3WJq=OcRJ+U
zR?lE5RqUj;A|O?Vi`RoWA>CUIe6Ya44Hvu>n`B>%wpo1U=Dif=+rirj$jP%%l@BHH
zYkyK47nG$QBt$>@<rydB9`{WU&MD4%LhF1{`+{nC*rydiI16&N_S-C!)p~cJcMs#&
zB>9Egy(7dk-xmeIpP!a^{;lJ{zQkN50`*+wVFM6u12mT-dRzh;=?73~Q-eoS<F*f6
z_#uP8#h1d4GVYeMjY$Jbu5^e8e$1P*oqXjY*}=f@3rPaDKA4?=;t=_04V)5Q5>8B`
zTKykKkin<LU}_n~{uw5ez6!S+kbI}0vq;NY!wrc2!)1z&KO{^^%m{8;kkxOUIzpi6
z#M}VM*!$0>H!o{pIf=vTng(#63zX}N2OjLSur>n?OfErzcuBO2zIrE<+6V+?uiHZO
z8yC-CBt`{FVv^i&`jWFxXfAf~ABdRYS=)ei?By%w<d<X``Q+*@3D^233kz%kef3C;
zF(HT8!%uiWYs3#d4<OH2e31vx^`y!2KDH>)BV~TjbX5z?Fo`f?P1g+XWiA3=GWkSe
zt0_?^MgffJErg8<u&Lv^_<K5daT?Bw_zQNiT^GsnA2>l0>0>EqTKG{Qf86p>H6w6l
z`$RsG`@VkvIsmYpe84~beY2YbJ6j2y_{Rr%KtsyIEX;}mN2;4Q;+L#aaC;sVulp(C
z`Ei2**@L#fwCH`D5QtFqr?>GA0g3roN-MDkjA<0B1~m7Uje|k;wz&zIkU)OTu}<Ea
z-~op0XNv}443Cw9MtmyzZVlQ!Nka@aICd)rLsJHgiXx<XLg~NtQ4tzq`XI3A#kCQ1
zS)6q#?N!7rz?Vu)fgSn10|~_ea$s~!8t;TLz6q<Ihz;Or5gc@RXmSRj%+tc$lB=va
zoU4(qL+)yxxB*!rFNqM<SW(ZIdcj(*+2YqPm<3#F+kO}2dySa$Tuir0bvZLj!<^h!
zk_Glc1bZ6>B%~ZD$e?wK2J3&n3K3q?iT77m)^ZOKmXz-s<QY;(SFtqR@7j;Ts<g20
zNk}IPCs@_7%#;}Q!3N0LCJB);_=K7M)G=KmYu6DSa@zJRADp^9FrzdgaiRdwsC_9>
zzAb<6fb`gl3$JPriesrN*rwkRu8;%*a|1F#j|sgIel4nW!l#(0#e~X0Kxxh(1O6BI
z#`BN!T5Uhw?LXJp#f%g?o=-wno$k#r!L8Y{{8R7w2%OTF69d{n2eecM|7fj3pgk8}
z=V20RZuE<G*y^j%>sb{%&>#@Nv;q?CT}>r5O5*&vcN0M=w9?D`5<^Vsv!gOCvHL8I
zU%hUHWeh7WS@Gxud;x;)MKs!2nKT|bv;?T%zD8jCxcfJ!sa=MgatAMgr7v_CKRXu5
z#W88}P8>LiUJqsh5@J2C;<<)vUB}?71F{rofTyK4oeTh(fTvB5e8GRsT}+h|x5{d8
ztPV6uw8!E=PHIihQ$0uP{rNSBO~t{yP6z#<Cmp;IN2@LM<Ko4DV$cY5n~&)<-ZL7m
zX!PilW^;nf%|YOJaclzKEz)LHbP0q{<MRNk5Ia$!IBMr5L?`mKqg8lDZ-oHW+J;Sn
z-tUn9OiZ~@voV-7H6<ZNw0j?7xY>5s{{312bMLUI<KTu!FkziB4FxaWtjUrJ-&@-I
z_{CDfZDMbiwwf+1uXZ1%##M;<1=jq-_5WYb?xiKYp4tA@DrMNanb*C3b7PH5v-P;z
z?0%m*GH|bdChLFxyA_tc*Zt26;F7w38h8Pkd+C`Q2ht~WZ}wQ-{+H<b82-s;&hU2a
zGb?)xJAI+(<NZ;q+vG{G9n5RwwcN^~0Y9j==CH%L?2+%(PttP?<H7xM#z%)L2d|sZ
zMfJ-~oey4)^HrL<HP(GiqKNhTK*-k;rzdH~-|FofT|6Wwzso#GHTINa-q|=rzb@bE
z<aqE(Tgm<1Sn3|iY}M+~_2J@^uiv`voE!N(f4AmtqfF_;Vb8eNIYesk+DQEi!GPCa
z?rNAIodQrBJ<OGjw92nh5A2Rms_%SfwSc-MQawQky1QzNobHllBQ=qc4`(FLzyG%f
z4+)oAvSkK6v33EVfIZ{0(;2Kyxl<r2$)o86*z0g@WlfP0tR*McYsH3MxT6X>mgQv@
zx_jFuNIM<pZa&3_6HB;UkUYn(y!|ec<Q-PZfo-_-N{Ry>E}j*I852+b^k`6Y!VKJS
zaiSzm1or4q;#9R@WIPw-8nzR%^8o}@bFQ95xnUK-rY06B;?nBXR=I0a7H@aH0dJXb
znsi->h^#&G4!G4gmMsKC6mDUMjnC_;1hLTL(Vnd#oRE~6C~mgo)c$WE7j_{1;9n3B
zc9s%gMN>r4Kn?Lt9tBrjCdJQ=T}VwLk{sU&!6@3|%H2b_pyeNWcmJY_?Dq@sGM-6)
zDmYVg{bZ5`UMw=73$ipSnT4MNYgk1D7dIZ$xdv8a!y6F9En#bd5f5fwfXG(^AXvXG
zg(tU{+aF{UuiW*C?9qPSq5|Tq+@HEo1d-iSGfKGDDXn9m+h<s@5L(Fpd<C4r7qAy@
z_~0>3i2G;h4XEwW(+(JQuozDy9hjhGDyUPQMjT`5b-pLcpgNG#vO>`CdtBW6%*j>|
zc`B(+n+7EA#_|vd2J3aGQP}7=!FCRbP%ClteCKf#dgPZ2C)CEDI7|Y|tL8}gcU5|X
z$`PtzqLazkQ2x+6Npy~7<`mra_!d8OMaLnYDgN;!$iyNpu@EMKTKKDET69Dn_a-up
z<1Po^p`sim8~p5nde3s8Guh6-8Hw@0y<G9o(B~R|1^RIfha8R!mnjoLYl&=O$2PQC
z0U_v=88AM$)1q*3_Nj6>t`CWTmAq!FIf>_9p#To2GP||34~jncm>t^`9U^0k^M^Vf
zbv=^Vikcd@rE(c$)S+3DDe6jOSSD>XHKW1^H11n*21I9-8YJ{CvgZSC+<Z(Hk3_5_
zX6IreF^Cr^jw<DWBoIg>jIx2?kRuSbQo6Z8Wl6gSp29q{Dda6(_u5DrisP+d{@_5K
zVOe6JJf7RH%LM++W@8q>9q&`Xa7{UXuvz|wL$xwi=_OYjhK8Jclwp#}kUe>&<eU{N
z0Zsp1Z={N~lztR%f@PG0h@FBV0zE(eD;9I)dh<i3-fry3rL%ytR;7f!YnE^0RzI%P
zHvAIJh88MR5hom#H=Rjw7BOEbVnxGG%*bMg3v3D>Z0kQ8;-$HajDN^8Wc1quEvP=s
z^T4ZVxZe030BLebg0Moq(!%mZ6&?@Wo>bgjFH1J;4UjLPYqw8I0W~i86M)Env@CIP
zzqKAjAU`;Q{@)klC`xB1Y#PE(uKFbrk;kcbn&jg{iQJIEY|tAZGrm(;0yyhPdcFsN
zA2LzL6}tUwC|F@t9|9_l3}OL+K9QXb+{V4*q@73j;{v9_d942eLpk%RecwdDJ`_i{
zvt(xx^`b(+wkPM1kf#b305Z{l=hPV(@hXFlpnCqc8pOR%Z(b7J$SSY$@NRGLn|%kI
zq4+j^B38fkJ^BDFu&#hYkJ`%jJ<B+b8j1+F1g`!EJrYT)hYSRz97)R#Uer1u4QOB(
zZkqO6AEx{Ep)wCN5F?NpW&Z(Ck@y}7eUU^<S5`L3C%a!fKoh8+1VOk%!_mxSsV6Vy
zg}Bdm^Ry&OQo$af^faS+cweJqK8Y0UrHVzkgQBF=W{^6@itRu7nV*m&ZTkvr(y>ZY
z#twdw%cbeuw^O5xK0w8I|Be2c&Ivi({cP&g!I*x6#lGe+;i8LD4Jqp<S+PsHu^A?O
z6rJve)2DKAanb21Sh~UU+gO@_LgxvT{_vOr#65n+UDVNJIQNXDKV+~&c|y}+mj)>I
zXAdiYuS&-l(miUeOUWFQ@||?^Sr}&w>gZfH7fYt>MH^|TJz9n~L0SuU2#(~uXY%Q^
zN13Avz0I1=HU=)Gf58->BBX5L&?~Y&4~o!l-tP(-s00}+|H^a}E;5X^uqcwoxn=$V
zIqpAMAeR06rt?h}QLUHkqK>(DWquz0ZXKc0+TX<#oIVJ$k|sa9FOu6HU2$ZIjov$q
zuViIVUBz-eo*f4_lF<sMW)!`l|D<*j0QZ=BRD-GIedWnA<MjZDg&q%Xf~{#hzpNvu
zHyaC49LDKc`#t<?p1L={29(ahlQBK=JwXbQNMCx|08=m^RD!1CPkF7zgL?iPRJLRr
zG<yE(LMqk_*9u|Wn_$(L31aR(jezxndg~&?;Bf<dTrN;3lPB|hc4rp3Q^a*fAAgT}
zA+DoT+?8?uZ#~<bR@|wxU&40cvNx}I^n7=Ik?HSV4l-*jkgEUg!udZBerM?Pmj0}B
z*aN|};8Oqb?QajjchYNq)=aE?>3uyTxwQi}rH*gZF2P@8wQjVC-S5)`yGMKc;bjXY
zu=`ymOFU>Sz>4n4!V{eq<vZ9fe*E*$$ZuWmP%OBtpI=IiTUi#l)RmEVz1#ql{#V}l
z67L!bjal_!!#QPAEP7#4O`FIyY61djQR3x|cT@D${Vp^%KT@g5RCmz3=-UzKj&3mt
zeFq4q(8z(-BYi!tC=g+L>DM=}Oe#()Sw1nBLep^28~C8~aT57a`$KD2*6#2A=HK)A
zMMnAW)x7Vf#&ey-Ckzur=v~$k!B?%d8Hxz{zTqhjPtbDi5$s<o05qA{4>H<hcoO_A
ze<5^?11nc_8iCGZq!d-*#G;=W#`zxYwdg$yq{pK_fB5o4xsP9NXClzXVwN9Ce=I!y
z{$t#1rQ+`!PaiRV!Gf0Y_)>7*js<O3TF!yJhWjzlrol?Tq9{2^h+@%ZhjXylM5fC9
z3<B}J8zr%W6HiI}wf=0D1AhGc+*$=lC>abje+41)>HU5?`yk@Ua1O{slWWo13*^gt
z<VSN@2xVvh9pfJ~(SbhTx8ECl%A_{iPf|4?wh^Sg=eo>xFXQ7R#IB&wpNW;W=fRE$
zL8WjvoLm}MvivCYLSf^~x97j}N+pUy6hnqrSdThKm(9&hb3BSbD#m7kj2M#rhptov
zCXcrd=Lk9l5o(TMbAm98$g8>wCKt)#hVOu@GiceG#LG}Q7g5TPb+NW*L7(KX0`)vT
zPf9IlZ*3CzM8(3O!;6{w0H!6Qe?X9aV{{<pqX&pHV~;Q0LO_DmH@}=V9ZFE1*xGSV
zmwGgC_V%C8G9NFcFxzXut_!deX$JwK<kf6eM&Bf00W<g#iL}oO+*;uLbAMprn9S*E
z1o1=e=iI<rQ!LCg@q==$4)a0C!*{jlx$!}~P<;Bs_j_J&{uyvnpNvFhp>}!Hzr7)9
zA67sbKzH5F6>?|!BjpDo$vz?p7Ys$pkdrUomaf_J5_109M)Z{K&1=y?{68}DB>n<n
zGU+-XXjf#ecQYe7%7oy@(Y`jCAk$6N5Txry;Q7Ipq#@@$044aK3-0EjNx;UinvKOo
zzNq1INmZst&w2of7-g^h3;4Mt3@ePXtFR&UsJagF)E`vA#`;QkofqEH9dlE_&z8d_
zbTQ@Td6V<l?pc6HFH&6Kk03ejhqQo&YX8Qe2uGjiUL@xpa;QZEQ_+D^0ElYmZLrPl
zM_>uR1TE2)e5FYq?$#|g0$Gigchgnva76k;qR$l2puf9MZ<xnnlI7d31hlJ5o|BqI
zP)cXe;%Hkos|RCXLlc{W7dbe3hkE2aqOTz>S&c`?`S}=c!JSw4u5n>O4*KTPl`vqy
zuz;M`XSN(*qErbaGGy~q`Te2sm_v?`;1OkUv_M^fINHxiMTZ>CYUTpD!=UQb7k&|?
z=`=2Z(D&5Z36M~HXsL|^q#3rd2GKO$Ua;ruvnj9lDIwZnO}5<3>*l8#bHjnjl}kt(
za5|&<q;Su+N8Vv03i)puOR#B1B9&tXQRpA1&N)H2#`djR!2Cz4tY}nVUMH_WM1CvC
z*4P_a=(m<dq#;17cIjv8o#-_VNEq(r$TR%iGj2{`)d&CrYkA&_x0p|X=V$8~uPS;A
z=lz-`(fJR%M8e%JzA%Yf&dtEZD)Z7hy^4U+e;5281`qxZN#`9%)&Ku-BV}j1#x0Se
zxK@-K*`ufrLe`bNskn9unYod@b-5)oo2-(Zd9OXvvM!?RkovuRe}8`ZB=?^4KIi><
zKVQ%1<3$v1^*N`kb%DA$^uf=U-d<Q06Md-1tXRfK@;KDRIG>|jBHV`-$KzNsQ^)#&
z%c!sr%YBAU!rtX4?u*i_1WC;kw?159|NA0JKgwp?3y;va$|W=UDy6Jy!}sDCDOZ?M
zr-e;6oVZWAZu~l0zl3?#X+YAJaxBC{g`Ci=4Hkcq#$v-%%czlK=&#>jKLa<Ny*xDb
zYo=Md@_aO{ckXy=s)Q*q6VJT44D~SYFPC|mg2r2jhGS?~h?-8_r6uFO)Pk6=^Y>^s
zF3=bd`Mg6sTJGbm&;9AX=3<y$B05}q*$YwW_e;m}UsKP-1~W&;XVT2(Tp)sL<>1Oq
z`pyrQp0NE2j-*i_TkjV^8Nl)w9!5k~8(DBr5&SBGLQg&j;h5hLsNDt>r082u2{>|4
zhh5UtG7mlz{yoo1fp6_eh{xbO4Ns|ZzQe7+7mIKPN6CD}6r)Oi*-J>6h-yWCQhAo!
ztISo(bducUt{1}WRtwF*iRV3iJTyx-?jxk!u7DJRkjuN5%WO#SeLwFSGE5~)85^a3
z^(pVaGKxl5Rvr6}qcRT;x#@3DPFTNfDBiL|ixpY1*|_i2%?>vHv}{Mp&XHkL<weRb
zVB(UP#5vk^4(IGtE6w19{ud0Mviu}H%NeIwtLQ8<J8&{cM0Rod(a+r>FJ|S`neltS
zPyZ8=UG)0<XK>i-?=Q8JuXer3s?XCmn==y?uP**uFZ>TI-x-4s&ne#X8Mu7ymvX?s
zi{E4|@VLw`ES#5FWVm)Pd1N4e<GUXIogrimnt15_{x5&}*<~IDdS&H&dU!c~{a4$`
z$iMYmQLY0buNATWjy`Qan|CcuT&_l+99Mip5!VxNcvExSW{x2Ti#QR3w~u^a>|L{}
z6-n!u<PWCsF+JZOlDGPd#*!f#C5rWbKV|)?%Kqx3*d4;~=UF++ZO?khTotUblx`-j
zG<6n}>SaLWreEPqId>ID@V2irJ%sHh^W;McN9?n9kvkR-!udkg4&F%p8+??1?wi)k
z_2ma$5mGWtxhuOL(P-@7UNC@f&*9_<{rt)7d-+A_LN?!w-a`WRdss4rp3qG^#E?_5
z-m^xXj`!S}=)*i^%atTtywV(E!JfU5;)#L2c}1C=r9Hat&L;TicIzw$c~6WyyTiIv
zU+V#KbwS7`NZ>Aut28922e5%6%8Mc9vuiZqvJ(hvdJ__S!jBOlNMYN5l>#|iYxv$+
z20b`iNu5oJ)h=wAn^XB25%+Hl3XB);O(WN%NYT3-uoZ8dfxYwGX=>zfct9ta4U;=~
zr$%1;zd9bsQPL)@4h5r@YbB}Cmwp<SO2D_^QgVTdmRV1h9Qe&X5X_vWHl5!}eom`$
z_)cfuLc@b`zJZrq?Lm`sVnG!jgZdQpE7aPh?;u6;79F;PGkek78Top6DF5|!azZ`J
z7vmQSthC2)%u=S{EMCa-`vXeA%N2@aSmoI<UUJhA12>t)+lc~Oa7cm|`7PK)o|k9r
z<UoK}aR#TY(oZ(It7h`-LT%}BbnP40W9SjXrhvQP|2B$%9)eE5C43dMF~0S}nVOP6
z%*bB#MZt@+eDul6T}fjm10p0j%N+>uGUSA-<!R>i1j435MFB!<M#fmK&WL(l$rmal
zl}FDX@axM-f+d-z)hL$P-QvRCCu*tRy+4xET4z|vYzuwxj^{_I;$?#CC@@=%BxNlo
zOz7&fCjRW=fs4+^H94UJkn!67>@8<#_NDW&qi`)O&Xzq9B2~D7=Hq;aU@s4+#c292
z)R}&J-qKb9@zWdwQM`<0Vy;eS?*lO9pIBLL27&w9Q+v0NSnJ;71U0_bDT}sbw)>X~
z$2q|)abtcK-0gqllZp2EmwAj*tRV2yS-71>iPRs4mxmtR{*QbP(UP*$uJUO0pn7~D
zWhS9F7|%Ur&L*hAWA85p@!f3`_vI|QF3T|}zrAzj$qL~)o*VEVJ5_w(M)B%j_AhNv
zS3C=$DpoBKZU`WABNw4vX;AgF-k!{6O$lYu<ZLNqsE*gSrOlIR6YrfQ3sQd}#tdIM
z4%r*tH*~~!uU>v+#_3C(Ln4H#;G0#)%|)@%^oqpi_L^5(t6No+8Y!}UnJOQR7to|-
zRi$zV_uXjL!F{^M0X7lZ_!TfnT~RZ^F-J!a8m4}|=2FyKAn$f~3RLhd-{N4DXABAT
z)Wu(=JkFb(urzA~&Mju_Z4y*TPkk6MmDyj4`H@5Qg2&L6&c)y^wz}q%2b(FqkF-|Z
z6Lf<au63bv%;uqZyBGi>Y;7H);dsvq<cEFF{08;q91e(KIEE=N&YR*OglnBog=8%P
z*U?BV2$+m$?Oxkb$iCK{fW+LCIY^x6CPeYKYazx>d^iWOb2dm8PvqF!52P=%EwPJ0
zw)IP=RJ$-iLf2ykcK?>T1ZZ3up=dgdy_{YYhNJYaqD$9{I8_zJ2SobYMG7hkiW<}#
zP*8SwI=V!l=mdTtkQS1$-Gd!orLoNV)<yE@v)IBYeU{+cJV;7E;}2JcJi3f9$GIK;
z^X8#My;3f*++8LMK%-pZ_TPXaou3CxE99m{5rV7Ja!E3`-O~!0mbC2iy^ONp;s~$;
zRRF&Ni9af-=mUaF%kn+A^~nJj2x_Q^<HX!INPr%^T%;$0eP`;CLnk?}{sTY8&_)up
z|Lu5H@94M9#okzT1|jgC_Eg=(t2HFXY)~0CoBum194rlsP$UbPlILT!T^A9l;(>Fs
z$-N8;taRW1r4{C?sbB2-1NTpqYdlL|kGPhr6Zt`9o<@1Xpa_-z=r+XTI36-e(PucO
zgQX-buK-0Fpe_Q9dr{)li#rWm1Wbk_&_%YN-M0K19&N->PQDT|F6Nj-`=gd-wqSz@
zamqe_P^m0@K;a+rn_$EU3ahXZHpj~BC{tLUsXKJ_1#noz@9N-f-$Z8=%2f7aa8&Jo
zjfC&usgll@z+S%?zCwl%Fr+?Wh>?D()5=2wRu^RH;KfoUB44^B);U<e*6Lv&iL#lt
zX56JrM?Ko){c%p$i|<`uHno_&=E4c(+Pt#j-)3)qnO)2?Iv=Z>?FykuHv(gEc!soj
z=3-C;gsxjo+G98T9kF3&$^ufOOteI~th5y|JmAzoe6En8?<wgc3JadNL*JRMkiLRr
zi&kYE6-NKl#j#Cfo&kYh=CkRCE5j<&Qr@xqG4;8<ctg!9iWt>)hYwdsxN#dSt5>t(
zYaZede{rlw<bViow|IY*)Yq5cr&G)2-S6s2J>5BYD%Mis{!T1pZTs{-aYwZ+mhZBa
z>=tV;?(yGWZH4lY>zMx(;xjh?{h2#I{Qp(K_#x>uM7;JsoPeUBO6%11pDVqW1=7=x
z4&m9G|Mq`(4Ik_uHM$2uV2}8p4&NMiEhA_5?!zrYq5R#w7HTcW<K?|La}Rm+=1^*W
zr_Q_CtFdny*U$l<?zb^%{q*a)>T~V$?z>&&40vm<BH!=0zSpd$9W7%5oDri*OTZ-6
z3{NkeP<Id$CZ?*?$^rJy-t&ZYmzQ~BgV5Vn);aCZ4K7%%<^1Jc9I`WhDA*;)xv3{f
z6#`_MJ%&Qu5ptH{Xq?uygs!r2WT$CYmMK_o06x;N$oq4~DD&AJ%1!r4KTi3`o(U)x
zZm?Ca@(8kV@olyyv^8+QlMK>-2$=-xI<}`9ZwpHKrc5oa-uvt3^~U9y!F*Ikp&66O
zyd&e|)0ZDQvqm^>6Jkt$&NE=$SeK-T_MfKvnP;Px*AQeD)@X}58uVP=OE<Nn)Bg;|
zu(gFp;<U8_f|lrfph$^6ek=jm*}D%(vC208{6H{8D6j9qF8QE=Ux*lzjWooo1wd#1
zG8G#8m~~Wa>e160IY4L7R2C54Ztu_DxQ3J)<L0t*$|(n9?dqFb`Bd1ba@A{}sgO1f
z#n1p}?q_(0WA^qe{Cp>Pzzb1viwL_*zz#Bwl1!A-+&K1v0yBpa{CpW2`T>g8SY2vM
zaFx$Gxwmzb1v91X6_=^bKDHGi#=Cj=f!yQ$0l|YbILN2W?wAv7aqTAS8Fxknre)61
z{()?QSf8!Slq>n4qNV6dQ*SvJ$}t{LBZKkOG!5hvS18eQp(f9G$%5NJYoOzm>Rmj_
zndA!*a|L|=1b|s~3^~8eN`CH;5AP)}@WSjXKJ^E;)`c~w(1$b2d2vD-hf^R0+EW1>
zsKy}^&ZYFJHdr>=J62ecU*#RY`_^%)si*MH-+F2Pl+e&&%nnNA30R=dp;UAPK@z{d
z2e$df2Tw>EjBu`=Bj?4Gzcm?Cj!dE*u?4Nb?G2=Y)leG<-Qd_Hq7M9AswQ5j!6Ye{
z5*hH&Q66n`x61dgM=XlM2ZE`mxi4dvG&CXS_qA3v7R8>T$=T-A&0~*cZSnu=P$3`l
z%^$}M2Wu{?Y?8)d7eD04wv%j+HHgd+ED-jw0}Wq$bAeI9+d@y4_I<^!d4eJfPG}ti
z@ZftZU9NpSM@E$EgXr9^o(whxM+fqdsAI)(o2Mm;aBpXe3=Wf_jq)}!SLfsw)J7IJ
z&wedUgHUr<o>*MHeAVCcR1Mm03I6|PKN<+|G}wvtHY_KU-z_pq#i?C8h8<|4AtNp)
z^AM<c16*=-j5+e)8U=47Fi97)3Q^o}@QytRMS#s)j8sE`QFP6}LjIX`t87mzyjRPu
z+n7AxO?J&LS_6>Sx9V7l_AfU}O$Dd~rL|IxOf*$FILLWHR%qPa9hzB3mh+7E&7gZr
z`x#>LG#hUCQ}Sl$b%+$GRKY@BFS?*fiKb9jj=VsEx8Qvtp%7--4A~Io?ikpQ0?ynH
ziKSi9=<C_{xQVAeW<&$p^yi~&*~RCNn~ATI>!wv=nVZvXR9Ku_UBgj4&#OMKB%ke;
zT$pVQM%I5TZbq?AgQT25q7>v{Ag8Hs0P@1OP?jh)!=)=l|22^(ncMFGeNlpTdBV=0
z($-lP6cNS2_9i%0FV)CJpjeIY6N|Lurpy9xHdHu*Ep{ib0K{wEF)$9uK74}^E~10(
z#qGN;eQy9a;Pwmg!eA>+Y8<RmX@4?z1GfBe2>UU|@(Tz@TtOp(#=DmrXfxTD7&=Lf
zk0F|^%BJV9a{t3~9O{CT9292%%An`qyFd1TIb>ij7LtcpyE`l7?pjqdf68ovZG+E6
z@coW^7<pe0i((+2LB5>dg>t^yv;>rP>81#<I`b!d%*?wiLh|nEAvu4wk2bPXdHWFc
z&jo^#f8R7%m_dNV!Y51rxnC0l_CvHqj6TcVx2FD-0aAw0kyilyc6h%N9wrJLs8HGQ
z8yQe8p*o9}*$QtPwu|U0Cijq)AFxk(V;-G=iuI#=OjnM~jbqDDPm|K9u(<#1q0<x0
zWKb4;$x-v!@O(g7lYpmo{7>5+29_@2;+rSGM-4S{ck^TLu|RxNt(gCE9ddN%N+GhY
zI0z)M_T##sH=jZ5`N3X(p?O)#g)-kA6gcI4i%67E3PCXk!t`89H$M0o+6X*$%v~O`
zu8AtmG=(H-v1b-suc(J3P`o6?K`HTTp57*@B#_R4IH-PR&J|uYC=p(BHDw?f<XU5(
zd8EULCa9TJAwcKrWaPy0#fjr%65fKfZ^V=remxaEbi^*{Yrxi-Xlakd*A*<_LjBxT
z2Shs8!K%z?ZVY$v4TytX7C1+nB=R}&0<d1*guLckQ7Da!>s4$=U)>Q*T<3ds<L>i^
zcq2=e(H5xzQP$}gPiz!|W=Yzz?57RKOXll^1lzQ(5-WaZ+Cs4AB1w*xz+*G$TLRnM
z^fxr&N*5Yz&o12n*(`6I;t6MM>uW*@Q7^f`!JkX&;}LkEBQ#fMHIyCxu5y_?0q=hf
zdND6<%4yXZAdwjawUMN#C(P#R9?9J5CAiZ1;>jFXFLFAy?IQY*n}eul>fAbK3fBR0
zN=^r;VGr)b-zQ2FJ6kqgaf-jc?g%FGDE<2XG9eDIqa#^gjwFFa{;4gy-R<*A`=pU9
zrNj8Rmy3Vi{jW~g{$DNeKfz#;=yJZXOsSG``$yBH{J)P2-;=U88&3y{DgNX;?I$3+
zSp0AHar?%DsePy#pJ(zMub%KYk+D3~^FeArq`)0vv+TA?7;XMaTp?Sg<op@9Dr%2W
zT$}k6O;~=pF92P4!5=D1k^|LyqH$0W=nq_X)0*+)Ocovl3)oGSWaim|;qI5~MdQxX
zdx1GlhhTi8aU;#Gm_<IQrtq2e<Lo=*^6U<YKh@$t@2NKtKO_f|#&M9Lc}|$PSGs;7
zPp)b8@Mj%HFRy|cZ_AVs@5Fpt$e9x-WSMFSJ``VAQub`H6IuJbK0!vDY?|$&A?GoK
zJ$z`sWQOaW@9+Ycm(ITlYXufhNnn($4xn-DLGvKfMukt(Vl6M9m~Fgv`bu74*Irgh
zvHAuE+ywb?pC_#O3%_w3+=a^Jp!o1r(tBu7vnDAFfUB;X$z{|!i!xllT2lR<p4GGz
zvEo!*d}RV_l^+xU+Cyx)_>NgXe|2fF6Pz#|Crg{GfD~dO@YCqlsX3H0D6H=+Y+yqc
zqty*&cIN<@fB0)v$ut$RJ}!ZjyRBwF1O5GsOY1W^WoLW4|BlUCylj<40X$HnNVVRB
z3X2aQ*<kw~L@aA1d&?5kbURkOW#B+?ia+ikiOgVlPZrv&CmjKK&`mm1W8<{O);y`4
zo;M!HOsB6&GYc=s{$f{#a*@-%Q8r0(IOgkycL)wxr}%|q<ls}Do-<$txbAvr3wa4L
z&GPRD{ba%P%1^uTB#k1yc^hFKWJX*X9Lrv_f?~apxH51(@t$6xStN=y6nC90Z7{?8
z{tFeBw!6ElT44FNe?NJGVGqpFmAv+UAuiax1Rki9zl?uqb@7_Yi083{3nmN>w~JV^
zUSWH)wGo6nk}!FR*~dobrSWRKN*87=h1cRNvPOu)M8C=~R8@W{m<40rUDWKVT?ZXc
zf68cuo>)<ev%oWWxC~tV_M9Agf55YcF4or?8+Bu#pA;W_l)LN>f}E$6sAV0bHWDF4
zo}_`!)d?HzM<DZREk3TW)laIc#q027{b2+Xj<v6on(;G|)a4I}e26w+uo|WI0-(+g
zQ<?~^*~wzql`RZ9$P+B#5V6}pX1<5Bl{oeV?9hhzfKo1kWL`MDPYYa=Kx-h@ey;U0
ze%$&M0sNDmj+Na)<r5K6An>=+<K>fQ@2aNaxXEY14;m4@2i@u(rE>tjVf>=vPnqbb
zwpwl?<XARVdmMd|$O1%^{I8VAA_a{`ZtbsE4{^8TAAm?;<OOBr(unv1M1)IkW<r?U
zvWOYufZ=w)p=@}cLZ4MOj|ypW(E(yo`jSSeAEe4cn8nAk!%^%=k}KZUMcxP>qx#ec
zw%WVUiO79(;d>)+J}cU!VIy=_X(I|3m)Oy_vl^Y78e|1#8GO9!<-xJr3rR&fm+tza
zgPFZsS#YuSS7o6G`&I{FU+0g3{epuQ)V4N@?Uf6}5H|rT^vc=nFcd*f^64_0RLEFt
zM}7C)^aPiwCCmJc<!u=$`gpz(IFb`0%sam-0BBMq25J*L%PWvR+&ltRN{muV>b-37
zw!&NKB8&&8Of7>s!i8!)Sp|UK<2hTLeHBUtQGO*PwECOsP!h^(!Md7V=(#(fR~xv1
zp<dz-D}ex@xVk(mY*b>*Lk|E(;)e7wg4{`78N&Q5ttDPqk+jTt8uCqZh_L_qR?wi%
zm-=*^1z&||$B}5hyn-G07?S;4S_+}6UG!om+A&Ly#VRgCS?h3349=R-=JqrxztzES
z(<1`Hr!3htFKHOST5EbgIlr+O=3WAi?#^{<Y<k!RqQQ>usWC2=DY8HwS^*Z&1n#pR
z-3Tq{|4$asq^j+(&&z6a(ac8KeJUlW-Oy<F0&{l8$rjCNl@C{_*P!#!el{EtE^03l
zC&7u|Y)u8NUt@_fe?huvQzPqkiC2pOCeK48C%$Ss-6`_H>V)%vh=H?z)FF9F8m(UA
zaD^biT?nj%0{3BhXeNvRiRQ9Flz!R{5=u?|JI2u0X_Nufi#xT~pFd4bhpgMbrB}#N
z19Qemq-W$RpwsqBH$p{DFh=X+<eN7_$u|5?^|QPHFN3!VEOfwXXXcnaA9}m)(m58h
zBo-3(=Q+Bez%E!6u{)AxsnkGD<HX0btYRSx;rl&yIf0A7^5aGoh1!881m;Q5lIrfb
zGq7S7Ota8u_Th>Rbv^`rFx5v({b3Q!EPk@~oTkdAI27twpI>4vX_DdwDK=1a@-!Yj
zjb)m!ke#GNC4SMjiBnj77~y%KtkPpGos;%?e`-Z~*c>diTmsjaFT6DJEh<{SupH|N
zOPIp}j0CuszEB^t-#9^L`9+tlpB0o0{4N8fW9UV}S73$T#^GXF2?)4lFUEn6hOK@M
z|NC}d`nd4KwlZi^otf;8N$S51&;-euFBpdMmJQvD2rbV<hY!DV#9Joburjvf=Ap~l
z{N`M~M;qo1QMIo~-zZZ*B$qtcZx>NM$PQ%s=!rcsPlahYt60`6l5z@|jLZC#!4~%-
zP?Rj?VlBYD?JMMGh$CtAjCr|D=L<l+<xYzokn^LNCHW}p#=`qvt%id7L@Q9I%BeRS
z6p~nFEitzK3X;C}-`^>ZT^pg8;rsmSc+aJpcOnl9f0*%`3K$)d(;5r**C*nb3me(v
zK;%DrDIZWSn43k5psQCO+%vtrLr`lM8{I5ecY$5;fua7XROZ<qzE_tdZ=quu56RMx
z@Q`DzR#)cV)IH@0MmL)<diK;qGBN2xZQJk6&+W;_lOB9i-EZ|-tr^cp<+BzmTkMz*
zdPU8~UR}dK3QQ@K8UFB9of>I|75NV^pA`!qAUv*bTJtHzbk<0c(S)~5+DFA<)~I~p
zKWen$j>3xMEm~iIe?G;4`7KFVWMu_9TWhvdSSyd4J_yJEHar<SnP<rb3mW(M=V}X{
z-T4s90KM_g&MZX#fphcg9WgN>9(0IYEBnG@=6#bJRf@{i%ag#j!e}a#_4Wc<`CA?x
z@r<lGo#qKj<OSs`miMF@oS=DbZ)kIC(mE^MmkL`z9chJia-YSK&ZL;Z;WH<G8#<Pf
zc99PZ%M+;@Tm+sPx;FxR@2P357HH~}h(7WhsP2sp-jjos6g>TR?OP@SXNsIJK0bk0
z3wvOhdFns!!b3%G@DxRzg5>P?Q+GMwY>^!_&;6FBp&&I&>K_G`aX1Zj837Yhz;ccC
z-2p<QrYb}>@+*W0wgIJdc(sz!Wq0V`Zy+y}T?g;X!^rpSc(si**dn)0pk+LB5+v7A
z?~R|w$W3pk11TC9XM01rHy$lirMhXg(-XWwcE2H!$2{x#<yKRnR}x|KRvv6+^3LFD
zcFEnZVI^zFeSmHJ$Imd5RIrx<U}$vX)gXDw3Cx!i&5b}=9_0Op;t&l#s)x;q2<6C6
zlSPoLi(u01x-<{&IO6?M3iJp&2R+72n3^dhmNIONA~k(F@I4I$ZLu3u1EAneWl#NV
z^U_=o{@pIqfP2yc#IWqD6L?#G(PfE#5SfDm`5V+{G?w>3324_LfwQ_}5gOPH*4>~*
zw^S>Dzba8?8kYeSnj;Ic0b@KNW^R82Bn`di#bdcMXXn-dW;%C#;rBi8VJvxflk>Ws
zA&?1QY))$l)NEDBeC>NAC-eT{y>@a_bqKuocXkZ{!SiGXf&gBoT0s1YJy+DL&ErU6
zDXtma^aGBtW}A1yQR#VOV1v8=Ep;0F4NfHA!v>y)W4alx@XebqK?<}AauclO&eP!J
zYDr*x`Nn|SN*45&A`m#}pPd7vzAi=zE4q18e}H}|E}~)0u{F|olcf8%RO5}FhZewa
zhlhJeXQqRedl{;|t^Gr!?8s~tU%+h^?{oPEkQzs_!QtzUE7}Hw*17@62hlm6vc!s-
z$FKz2HAVqN`n>)de|9AjwApv&U{>d}y7N<G5mxP9_Lxl)ZuXt1X0%%~7Zpa%jVyJ_
z@T|>Y+3sH)1r1Z8;ERtG_HeQ8E}HbZ!9H#qR0K}wQ#wFX38gA+qF5zg?Ip2hw?ghj
zY^a|^Q_p!38$D+Xstu2^q1NKeYQ21eK+|g}w2Ali3oKA|><)u9YPu!NsUIxRzcBE@
zUop!5zJ<tMygdnIu;)UpPxY=}yEr;nbkVf4lj`-ZQYC=dNKbXMS<4mIVL_=Cggz+C
z>8C(fv)?ahfbjVWKXN(nvNhI<Oa!K2AO}0SWlGM}vSrS*l|S1{Dn%ZB$qfyjj2SL<
zGTSOAfCr;ft~#OAnJ{da6L2MKUL>xxZBWtq%>Qf+eV`*Pp+aZx8P)dg@_AnRXD|Pz
zZ#-et!O?e4X55G&`P*a$%oN&UX{)@vu(#IKK#H(VdrqR1b_b$pc@`YxO(e)bTuJcb
zBva(i9-a3lO9>1u|D|x_&ovn$vu|6rM0|zi`~8ZRz2w+U8xt*Vu$y&_iAyN+VAQ1O
z&6QG0-EBSZYe%sxH9ZTxuXD~3{*g{TGbQw;(`KE8ppRTaIA9|XG#{P2R{>N_)shIM
zju0^QqJ1n_aH!$Gwj2Z&)zj6n1JJ^6O9?}COy5f;QRaWwM`B*8%D{4D;^|45&jJWJ
zy&n{1ba#$q3#qz1nh7&mk944(@n)Q~(>Rshlnu#?cUZGsD!ae8p<bJ*f#=!OOCAyq
z1!yl#m4HUNGM!r2Iv2|0(pOC-W94v09gqfi0;K%2_nT34<?&orjO7FEA}zZnN;^#`
z!eSz{1Q(72#l+Gfx0lw68uq$BitGgK`<pqSu0|A=>N}2quVSt3Ru~-laRy>#)jcjv
zyzZNYK(+}QcYpvN?J)QVQGZ?hDTfrzD=1<n2$=+11-ar4D(t}iHc)4mt3-Pk1h-!~
zQ$}oo*oyRIm4k0wmJbmpU5$5g)Mz})pqLPIg8{!m6+HiuAE>+7W9a-x9vYhCd6uY)
zw5G0W#<3R|*|)$VH0J{E$rv{ptVA!axa{423itcp9gwA+dc1kB_WWH3i_P>wx9QmO
z*(1vL;9gLiLPCZY;(zG2{=LN0hfV<l5PUt@N50rMbQ$=!I{f0+;2!n=@TZyoqkbQq
z!4(&o;uHZI{<yN<x#!3&uO)YxZBqU*xbBBGug&-0TL?k{$9M3B(&Me#y5`k(oNx0A
zMhmQS<Iz03pS(_v-XC)veiC76PKjBjez!XPj;SbccnovCkxM8qX0{8cY~S0Ju7{;O
z0xE&7!fC__gD3eFZP%61tsh@u5n3D9HjYYwz9r{pcP3*;;FKG7a%m5rPPlUws)1+p
z2O6_R*fmh@RWIGv1bBrUsq_R#fZ+n5M<+eFY?aV+y9CAe^IX^5>h*A*v(GN)ztM3c
z^R4%3#BQILTPzy;YIv=dx5R9wqvm+A@vzZL71x=#*u6urS#^IesVUsF!P_>hGS8w*
zT`P9qqT|<SGUSWgKf!B%IkjjUHbN^Q$k?)+275aE_OY4K!8+y0BHE4}M&(p}jv{0y
z^~KT+=ycb{Y#t-Ce9)l5PTm?iN2Kg(1P0GoLd2Mx=EQBN4N&lh{@>;+Jt)v+4j$oa
zG&E#|)K>*@r))|iKH<*fRq>J)%1OaxC2Gny-*E*+Y`=#q=gEHNMOAL^6F7B;`%z&_
z<`b>3eaNI9vfh`Tr=(oj`#>3`rEK2^25cJ^f?ii104dBO>^!nlC<XQTHW&<MxaPiq
zRpC7W{)&ON62ypWIoZla!J^r)VO9_!8w;@WP6jL$_NDkVBUw)?>9~9YY+&#3PD{rG
zTdYfW0d3mSMa6e=M4?61B)fyHced+Erhx=o=nlcWm&YEC0q=gTR41OGTLoA2w5@w5
zx!VKGLu$MJ(3@@aV#5CT+kivSszFr{;s~<ZD>@s$++K18J+@^%7PzZiM|SnQOk^S-
z{RG>sJ~)@^R!wa9wSt`;km1hqxtGKwAWp~LYS(U4viV;-xkT?}9W&E^g8w4R+o2-(
zM21lbQ|aH@!%)qqal8%3UJXiXMN*wShV4`i8DPYm<Fy<EtNQg425gHQ2j2F-4RGQy
zrfSNg?^$w_DKt}g&JZc(8~Sbvj^0NAM(?{4^gza0U}TH(INr*F3X{+Nu-#z7;LN>u
zYz-qO$W~(Bea9^xOA|u~31UWO^5nb+{JhqXWR@FcLEpwrzp;FsE9F667y_*#M0GOY
zkOKQ-rVdnYh9^3}M^me^<O_*q*-KiiLC|dX@be&TVjKWS+hq-5V9%)KAkbA*zic-^
zknhl_Rt<;}8WZJA*^s?beE{dLq?04HYSazgqOT~4XS^$o<=Jfw`trG|$L+{NydrP?
z8T)(srB>`>QEa=~w-?#@U_OOE0FC7R(~0wQ0t9gSN}&a`QnO{?C5>aI*?LZrKZp%O
zS!<q<Ud(Ig3w!M(_4EUIVpf+khyfq75jv(}N`o$+cu0liN;JO?RL|!4&iW`a?m=gK
zoGg5t7geh9#v(EXw$dw#n%vp%f%a^u<EuAMgS)M6V)->pjUMp!tuksP(4GXn)NtQF
zerg8nN;R#ysD$#=Op+7;a9>;Zlfkkv<x>EfbJ@)Lmrd1595zS-a-m7yz!mvClPX1x
z4!zC{8ro<kF1GyVP;hL&)J-sSv1w~|ZRJ82N^q8Lt8s8)h4?Q~IWj`D`F&KVbRz^2
zKhc`Z(=@s?HC(|)BLc=AL%QRVH0OE4(gwQ*N*oTw8nF)VYh(Nv$^XrIC@4)=+}EO)
zr4#;Kh3Y6`74YKI-(Ummg1S2%p*cdbmoNr0K1U}97%?Zy!JRu&qJl(DaqDvu1pLQE
zv>d&~O|E<MU>>HQt9820x_M0$SV$fPQ$vCO89uT){0|0(G_**%2QZ^RlJan*nN<!-
z_WAZ4h!q|;#SvO=X+=QZBbEJYzztX{RW`M(IIKR^Ck9;9@{X1)HT72+(J%)m|0&vD
z2EmrAk1T<kFL%@%(=4PyS8I&CYzp3@EL|Z_7#tVD^2`|e;ocm-vX=vE-B7k~L7b3<
z45%^=9=OP&)dtJ{#@4B4;$}rUbH8(*0=aIUDU09*pmA!k#Vk%7f!kJJtIOnb0gm~Y
z&n2zb+3Fy+Q&NH+V8IVARCVyqh(O~+*-SRtI|cHX+;B)ClU_H!Ir>rKqLk8rH&STq
z`@+x%;@FR7!KL^Hb(Pru!Q7Q5KKf!SekLWs+#AQbiT<bWgxQjrintrL0ydD)eM_j{
zr~y_khh6e=f`0K$@CdGT-;W*l#OnXmhHH$|0_o&T-@fc9RiEvZ(G_@zm_8H$^p!#+
zw^#ejYeg{2#;?tFDM#`??3LHT4{KV<c|_~89Pi%|-pbM^5~pg+wrHd<&uus^V(UW@
z;WcHp`$bYSLeB?~DL_IWUN8--2#%hJat4v<*xf~P8Z+Ao@JnQElnuLhW$l~=VrJlT
zs3SA<F68Pg3?PaKGODSQp?~1i(E|AP#PgYjDE&ZLK=bv-qz#CC=ubQf<`oGzCJA7p
z!^G!Ztv&iCm`{>Lz;$p4a*4EUIM6^&X*fUEeejmX2Waf=-}QdpID0zMEJV>m<R#(H
zIUCAapU8EGqgDYORRbQbyaHVy<~W7G&Fu8wfBLAztPH`|dbMp|k$u$un1>C}g5NAB
z7}Z?JA8-rI1&>PKp|oYH=L;ivUX`fnJH<7nfNEl{K?hRoS+TC83&Ip$<qhdAA-q~l
zLwY82`;T<GY*;nbA_|^5(aL>Q0>v-PEmJ=~3V`B^z4{LRpl9XHbk4E#^(UwQ{jMD}
za(KA-`dV^^*V{CQ#Xqm14gtV=;I9wR7}(c(oCLJlVQ}uJ25lsQ*Fa8tQvUAYuePNi
zq=zba`P;RWGIBNpPX}FUF8gvgVS8leXD;=Dy#31iKdA&6-!IvG|Gu8)&z28NId%7N
z^~<~FxAvW7e|pOF&fL({s1HL5Do09Ry`;c1I<E0z$sdzf^TaLoR`290CwZW0>y^o{
zc6fGk?y?y!tZ=yt>I?i<Js9+LsFxG1-I{p4KO0-}r7^Sbh{Sa<gRY|Ys_n}6cFXtI
zz~qlLyo1J7v8_RGUOnBcv=|_@{!hTqpO<7<Ya69)hiEHVFY<nec<&tH-p5;(r%$t&
zLs_1O`sRt*JLtHZ)li#t)t4FNG`Jsu66e(lIwdq;4eZ4*0n&4!+rHeQXa-_#y_>&V
z6LJ1S%Fz?F<ejfj7tEa7dWOiu4MelF(}9YHfP2*k*IL$h0BEgT^893-wIrhdUXBWB
z-DBa482tVGR&Z=nbi8t;#}l9qDh#}8^Xuesf>yFdG9VuBl=RDk`Tt;X)ao)tw(;(Q
ze2&<$hAKYeF8;giTqmUjtnM+!-^Xm70t=j^P}>jUu6p0lMa0R#&>VVz%i0(Y!iS`9
z<RNTw;5r+CwiogA92iz#g9^eMo$?+)u%}|F$bu)|KPQFIcT!>9{yIWmSM_%KV+wS>
z$XkUR3JmesoxkF^Soay}KQmOwWt9t-j(GwoV(xdg7!m}%8F-L_jMm973qHs!KdKJ9
zd816Hx`TNtEZO>rw)O}lzbBvck^H~<_`zs0e&AZ4&V>5sJTqAG7m*>%+Xw8p*qR#v
zxMe+;43Fe7u`>4%<{2$&@cCVZ@+ctG7~0#-U_MYkCJd<Vy!=d0k3<q>m%{8jC56+F
zy}w`t;-qLH<rgZ=)cW|S*7@$cBebE`X->PL1;?5DKT37XvtCpOas{_b&62zmgCz6?
zi~cBpA<|fgu(`q4b6q?nR@CoO2(wipw3emk=m-q-Q}YK+z$Gf<ZD8cclO@`pJz@<g
zN+%PA87d)3r!Vz%9A>Ws{{RDpcM%0zuI>^QlG=lp(0fS_16_Qt<{vki2?cJ}+w&O5
zOAL~o88NIC_>Ki#sCOgcq`a?P8J!O!`TtGo=}=oBBT{6g(fRpZN1jt45D*_uL7>c%
zliZ^EjhCz^Uwu)Na>9rSi)&8KAPAXwWfRm6?4Sc-U?BnQaO>oiqc9_*B_)!NCS3}h
z|11HVQ6XiN1Y4X4aq2A&AMQR7GnVcQMbdZCi#)+N)$}2r4xRY!+3)_CQ*sbrsA&&K
zC)hqMH^Ngd#I=ijj!W(#@xa`fFfh<7R|KBbT!*m^Cw6`ncZleEJ_2E*oedlwxl{Vy
zAFFS9B9LCgasbrzubhovN_PN76FA5IZnx--dFRL`z5<;W;+o}c-n-6q4lP>-&kdvm
z+Z-pCpwX9JT2N<?2OBM{GYes_qaT5pFj)t1s9S$7Do5T3URgXxpic6H&bjBVtu(rY
z(-zL`doMw&4dP%L6~2uxPT^)jozFpZFl<6didK!=hNIm`?HaRD>Mg`IHXD=6>>yk@
zENs*pSq2AVghqv3-{E}rm18ma?IF=_ez`jT`L+HwaoGQ4`j0yu9XOr=yqOi$qY0ZN
zir`D5!bQdz$=tyNQ5f0tq!bQ_k3_J}jIiK+TP_gH3!9W_&_wHY$mV$!<CytH2v2$#
zqC$!o=fiz~=Fnl74`xh^oE?_uc*j<<+Av+o6`J(HDhgW#6whmph5Cp6U;&FYx5g$^
z_}}PjD<V*<>PlnN4^?hKGh;{wI?)M}Ixr7V?&^jeAM~k%IDtn%_e6GA9j8j6f{~0(
z;f9Pwbbv}>gWcIdAZ6tp>M~(V5)WQl@vqX~|C9EN9$@I&@eSuRqn8(q0bV93k^aln
zu>Dy>fn?uQV?Ow)^S=M;{&4K$=$Uf^mP+V+<_1ah;Pg<yqltN%TYvk&2<c{Jtaakt
zj=tvWXI!}0RXu33Q-c>bE_Q$f6VlF^o}|Rw7ZNr~b&BoaBpjP&D?~Maser7<Y*x){
z1GQ8}fG{U>&AMiRBln(3+4ZnzG6LPTRtaEfJ*%uk%`z6jl?!Mv&KKA5o1gSxPbC6J
zSTZ(L+5hw$E^kf{;KS!FCaAK<={xH0kHQjfhkZ{CWdNSWv6YXzx)9hXHQ&AhL*snk
z0=jpEo)OXoN|09lier<dFy=vi{Gs>iRS%iM%YECAFi*o<<!WIWOu;>U<An8D(_99s
zN~aHR$K3@COdOP^SG2ki{b0yl{b|jOlL;@NAh_nhO-XguJW)-s_QIYcvepyxi*y9B
zALX(c#an`BSN$;&xY$EvjWH(SjE!8K0ti{h8Duma!$t0tQ*0Wh>Rw}_)Tv#Fm8FaF
zXaZ7Vi5Od&dWjgltziAt5h~rQJ3w~Q+f1AJ26zo)(a~ApX>;hqWF<)^sM30p9Rc-w
zve~5@!Y!XN!co!pON}IB!>sy<_AF~d0BLTDpvP)wzXUB=aqz5Gm(Tgj;L!{$xc=da
z>6_5%Nz%e4fJpF~fH<7YcBzj930pfQ*83`|+2ngCysxXks1_JX!#g!YK_h1vWGVx2
zlnY;kR>*>PlG?1WcmHia9=&%#>fP`Dd)ES{GCwErLjQkX;;#*$ygnTfb0qe{?`rq&
zjr9M6lwK=M*KXPQ#Xqi@K;w;(o6ejWe|<LY=P(!l7YP%X!@|e#NM}<1MdjDuIeT~h
z(HEz?I#KRqEh;RMZ)UE(<L}J8{H0$!bM~P@C-Hdn`oAqtELY4`vC(<27ct|bcAE13
zC5cWS7frpYGm5t#l%vF^Hr1vc&6$`_H~&2R6<g->=j765+ttW%8^>pd+Eg>icPy8l
zayHq_s_|~uYS!39Rx3@(rFQv#dZ5+db;6a>Cr!C-YwoUReC@*S#);C9w3&M=g6}!X
z1zGIsIGg3IJb3fmdp-r5eAyn#<Y|BPSlr6KPKsU_HTk#iA8fvTG-!kM3JOBAlINDP
z3WL8AL$LmU>(X=OE|Ws3qTaS+Xw7I&MJ<WvKR&*l^O|JOY1XcJcV8OCCBEQHGoT~!
zPRHYoPrrz{WGk0gPQ{kZ;$6U7nV#?>)fqlA?M$Sp6uY>Muqj@Rt91pe7*z?gGg#b$
z53J|NXNmT0aXfk^Y?Zc?><qD`ttF*u#-|wEJCaDY)>5V^+izGq?b9xJ{gLi^yqXy=
zOc7Jw15vgL>7>!>Ke}Nz1LEQbV|S>n)y3LkYQZ2sf>D@m{`$)hIQsUKc}Cq)bpL6d
zm+!qb=3ryW(i>4f*8QxOnbUFN0<XLY{ZsK+hJ^om|KykQDV)7mH`ize9~<?F>p;0J
zkx`&Bw<U^4gYCdcsp=q2k(rAq#pEtqrgN6#VI}Mu-31*q-1SiaQq_M$*KB3qT27kT
zH>mwYMG^myG?aA)OfK)QCUFY=IKKBuSoz7{9P_NqAA1d<2>a_3p3dE#biZ@U>|3*3
z!<}jdOVm=MCL_AY#2_>ye)!hDSf-kq8tzKS<U8C1C8hR%@{^$*P6v^e3aN8Wm3t}?
z+4h@GC4S-kqE%fHd;f$FIHlx|<Eq}j3Rk`ZBApYnIC@~>mQ*Nv%!AzNcx!vE=Ta0e
z=w7+x_IC4h*g|f<9VUR7KfO}P_>3~=)yLfu?M{YtaaNL~HRjS>a`Q*!4Q0mP6?mJV
ze>`V2xGd5#6B!e=(!~Q46GgY97J!#8B2dxXkNDafwj*8CA4RhF^eOqyW}g5GyT+y&
z)F?3@;Lb|>M~?Y78ovk${ZCE-fuTre@YkWFWbiDtTkEE&_fEUimY-%D@)VnvCF<p-
zUtW{P|HD!9r#qs-Lhw>_THJhNRJr!qrl(D*Od$@n;=K+liw54&e!XUvkVd)&Z|f~o
z1t-E`ZEald${NAL`FeDRnJ}A<=YKja9;vJmH5xaZDt3o9`)s30j+gDQHL$6USLCYX
zzhkCsdBrHsa7AZjtI0jqq!cvW0TykghPHlyk<@W<t1fH;b?5oj#aB4nk2`_B^XcWo
zY$|k_yFDZNm0qZiscSBQX2F5yPQL5xpYQR;Hr`iP*}$Z)e)PE7s95Ss$vP^$94pw2
zB4sUn8z<E;SK0scF;5aJVU6uyh)OB?X2#v_SogZ4!s+9q)XKpi{)3dRiqYMsP=qjF
zkNT#6dJ})!*{Y5O*Y}Z%r;d#F3AEx8R}}EBXKa+SxKqZ2#Kl?ed4KDXEaU1}E8u_c
zEw8QF#Qt$Tz3Dp}g@<)7SMs&$K8>>EX-r-bhQ#yAsKO^g#a@SIObm6e;=k@J(6enQ
z7u@e%#2;9>SOjfH4qCjuk6rlQ@t$}7LGu7#+d<wx(wO<b+2%U~rjIobKTXN)o!T>D
zp7X<ZeVq3B*-Ta-IcjeD9mKpl74-V^uG8W1=AT0Q81*lWI|sjFl>#l|2Q>yx-aCjN
zT#rSTAAZ^Tvv|*ZjnS4Yb?20;tzR<^=5Tlf>hr5&+P(yIdrt*cTQ}b`xqEWQD{_6T
z?%g&SXeUi#od7;7Gg$2`d4y%uuA^%Bu~A8rxJ%MH{rcf2-!DN+YhSfr$XDOB$X~g#
zne*jv<_lA!j~3I-(l=k7{>l5~kpGo#o545_q_E3ymCsScF=&g_bpiAmE+PDPC^|AK
z%>LqN+1D-k{<<hT$|^kz?so}&`cA?n%cWU@su|Kjd@MiDwO>Sh35V$nnA1U96_8Ir
zdj;N95HjWp;haA6iQ00njw1Vnpg#A5u=?Du3l(Z7e_v`z{3)v3)j%)-c%R|R|NY>D
zJeW%RR**ss!R;c+ACxk`mWvs2)m*GN%k)Q3QkC)&RW<EdeYtdCjgT~Cg8-gQCd9EP
z5^O2x)^FK)IpZj$Dv_4412#?xqW-!l_Sh}e*R)r0wZ+UBvoDd}S&9sjW!*5v5kBLC
zUqCXO5%RwjBPy^WF1Hk}!%F$=SVNSQ#*Yg<^UlHTlFD)R0n@kta!K953gsg24pO;s
zA59E9`jtj7gX`Mr{D4>|O9<yiz0>a@+axwJOT5-u18Wm&j(v5{5T~^_%$Pju?;CfZ
zDcPC=9>f(w+_hUc1P#sQg=R(psh<8QrW=}zQu3O?p-)W&Yq>&%H!a1rxI;30`cB|`
zw|>A(QmxZLSqMlKF~oa@6x5<FAfqUltnu3`a8gmW2^SJN-d(ADjGt8Q3@nV+24k}H
ztkM(!Ag}FT@f#JU*QJFawbBw2-I6!dO9Uar9b$KbvC^I{r~E8^g$sr0QLVF+L1O|K
z#eMRuAtmPAR#Xaa`HK#&D5#i3^fYd0X^KaWko;dUiEP4nsAjt7W#b?$i0l3TPpAW^
zP>!BP#1I)xuV>OI$pkZ8Yj9L-Fwc%Fe+$p}k&N)oXeA5!CcSC+arEWgK3t))(<Vz%
zDI4DWD6Mp{$&4DwprAj)d(l&6WkY&ToeuKQ*X13+&vx_PRpK_R;qwusT;)2$o9Q3d
zrDT2M5%$sXUY$ui{VVnx;cz-IxR5u)0@TSV(L=o+hIbA!mS4J5Vo%yVMqGxV0_oZ1
z&l6gwW4g-z?UxN}bVT*A^~0M@PcBY`2RXI<YhsmxLfGTFx{Bd-fu3CdIM(xtx_%tq
z*6{-(rvg*rgcu|*{#SPoCKi7~DWx7-J0Zi&nL7<r>kFi~i@8?KCJr^SmGu@9K3=p}
z=Q)MIGm5&-hF8j-b?ZF!*ufHP+pfccuWT>Mb&l+qPXCZX8{%4hG=IXR+0o!w=QnBb
z!9s^!EO!E{zotF-qAZsCJk%y7SKS*ERZdmkwzJSzZDqZST-9XpmW*2H9UHF@<OTy%
zxmf4d$pc+Y#q{-2+|(9ck=h=aeGW~^fpO%O5~(nfL*p23LwwWBH6nfQ2cxUYVY*1$
z)8*d0yt^Z8F_b)Eds0Z6G~;V-vCON%R{ogEfwLvw6P7GkCIh7y<5&q!hvc(()6c1q
zc9dhOx9PDwo1b()huhQR>Lq%p5HwZ`s(|^pWRXV4o>yvU68Hv(7lJQ+<&|XYc_$}C
zeY`SX8H%!vQO&MHl+`>5CR=Lg$ux9#?-8fAw&s%}QK@su(!K*YBlbXRmk%X~lO41#
z&E`|2TNzR+s{8JpTJ{m!6JsKq_(Y5~*F3ztAS3I@u{Dun=$Q1*;vATtlr(%tqvBD+
z_Y>*#?cH9dNK({(7Rk6gbvfxUPhMw?o2_3auOD6Y>X`OFdG+AX|B>WeCQ?!mlB%39
zCHdXki@YV2H@kU>&boj@JgD>nyO~w`+JIZ+qAnBHN=C({*HLKq@>~UVz1FZNRb9`v
zpY=(~ZY|hfPaG9?51BhSONjaL&P2#QUzc+ItvUm#W)~2@=ez|DNY1T^u*`c0LSmWz
zo#8%r?nUH|xb51$j4&gvH|Y2u7UT9c27;YkTLA9cgW>?$opW|{lG0yzhMEPIPe)c>
z?$nX7MLbj3h4{<+;clwF3o_)RmEwgpeQv&=qx+7jnJ7M4C#-AZj5*DoXXpAo%{Hpx
zfN<u1^(88MGwnU<{X;=PehIC}oltY$Q1u3G^=Ia+4zJVEgc4KNrl-F#KCq4lJh9^8
zr*+pQdMAie*L*TlBK|Xv5qQoZN6n~Tb~!@4;*8qMM+tkk;#ciG(@6pUdDp*O68)|i
zT_DACA4frS9cDkQg$!3WxgKV;`m~bA95h|8Bvn(LGp3w0TS~Vd=HV1gd~a*vVsbk~
zxOqvs(swJr(@Qhj(_Dl?e2F#|t7{cFRB;HNd_5gp-rn6)358yFZ-Eo`%*7PhGtZ>c
z8-r?>&fS)7;mVXSDs@PUIe$ZCM%#q-#xV!ochVQ{chfquKpJRN%S-bk1<U4Rkzh@M
z#d+qMpdmOpN1fkj_~DHWn)D}oM%!w|WV{xeHf<v$%F<gtp^>?u8@6Q6mz|e8C(H51
zZI2=&&~*vdN_$;l(=LJ}E~*nXlB1n@YU`?26H1l{uijL6$9s*(`YAaiNATS3()}k@
zdqvSpK^mW4Q}GnRzT(3sKZc+FI94uoOU|@d+E9Q{2(zk%hpsDZG^%@f7o-W>JfHg4
zy+)%eVs4Ag5d!k%mc1-KaAw8Q7uzddw*@b^uYSGepDXPzrndR*Hhy7xb+Vf8xQ;iE
z|6RCOI$rp|RoA<jl}k&rc1+cK^Yaagu0MvICoW5TXFpu!vI>%CRd%e8A)IgK*S&Ya
zeXXylPou}n&=KW9{*;q=b@+YE%y{2!y&dZ3<z$+~zH62LdP9w4e%jF~*G`S^S1g3b
zj41^uFJ@1}mR5OzD8)@C2SrV}EYWmQ-Nu>|53NuFIw7^A9atuobySpYl*`A5r8}~f
zxn`9pYH3|&*N=N0k7CEl*kpZ9!-SceHb>v@9uO<kQ+TQyey!PDmG0}i>d|;b?sNE1
zv9;ONxv>QMRly%=G;t|ks!lu-L1@0Gk-$b5Sk;v00~PhuAQT?*QT*AQpIsk#I<|&4
zz7?pW5?py}kN9dJm*_9ii#B)9bfoW#sGSVzt(4tL6SH?TU5M<n23zs=J#o$Tr|qz1
z*gGnBeO@Z{&owunyPwcO70>XHs|3-4NqFBghca(Yog8#~BYQ9%Lm}GuLPzy4%-48z
z8n}X#)B8u&gy-|3-V}jJEx!BFpX=sqJ_-7EdYs<p>n)eKl*BiAN1Tp5@I|=#iLr)t
zVpgJ_)=S(njLO`vyp1<6s<cb|o6d3~;Cyd{j}PU9n2g%SB^_0p^g%n5kGy?xOf=G$
z?b3fnQkDtpN*=s(x+(~MjZbkV!On5;sps)4<dg!w$t<gd@8O|dy$SSOh_OCW>4PV%
zGwRjH^-}BuzEGNdW}RuuwYEO~2hmx}(b9aF1jO$25HUnt?1Mqlw!IO)>d}wx8H5m@
z<I7Kzyk}G9?Y)rVCWbqCN%g;KN|t&o{ZhmBrR`5L3Y6b4TtE<%r<Zr9mJugcUV(K~
zHkHe&?`Hu#*T$R#JC}~M@{^hBcN>tUE{kRr#bUSZBo?%Pr#1Il(t8vrQ&DP02v=1G
zxTBJU1sByLo|%$D3bq6eFoD2muzox0$7$cDs5PG29Dy?DY^i-CioQf+S3LCON1*w;
z^X9t!MGE6$*2=r?&t_Z0&Y5e;nXQ@|YH=s?L37#Cl9>90l^|9)8LfT>XLXkSmTlrz
z6k`S}&Mw(YhhImUyGgdr_2Z&LT~MKUpIPgxSO2D4krDEi_RVgyoZg|pflRlZ33(&Y
zAF{TPdGA`o_Lw<L_^IT(_xNBs&g1>z&b6#{NG@`(F_smgWb|LavU6s}nyH8k)rnQo
z*aRd>&2*%De9@)Wf4Q4A?iIH*JTE$y=dOOWNQlXL@_3c@d&RlBgr$;!9gZ#D!Ibs6
z#Q!4t4Vt$<+id%(HAZXAbd$Qey?!dbS}$6RfP<kf5_1Q$1M4#KPZ}Y=K(}s{toqES
z&w;mHGpIW4TtUv@>ERFjG~b;LCQx&vYkvK<r2LxC4^-|M=B;Wd-wgZ@@U7G%j6{u@
zCPiiSmfizh_eU?~-I|j+i=7B&5?bR7od=b7@V2O0Z7sJkX%U<IegCTP6-ny?Vls`T
zPAAK|Ex0MqZKXL{s(p^5sJYV{{?~GO7kaOi*-;;JTR0Byavdx6X0!sQ`RCtbU8(;Y
zzO!Empr2qawj6bo<&W?*O1dqZEXsLZ+Hu?NzJ1!Wc>{ywM-|qCQ6afMIQAIN&c%Gf
zCU4V_#TK3fpQ2LeJ`fYIZ7t~Y!Q(F3U_WmH4)o{}NGR7`x|jHI9nVeTnr;mvni!|d
zEDCo$XELokV?RhI#Ga$U^U9advJl+7$`?F9j|}jyjuQ75;0Wg@vApDdqpmF3BMNhc
zdQ#27-xKoaM8mrhW^BoGtI`T}h04xFsroMAmhf72Q>w8^eiw&8GdEvj6ZV;~5#>~-
zi)PAE3j=!3EqJ;5=eArlx#!Kq%X)FtH;?|26oTu=2o?>?b`AtrON2C>%32fyB&ZNk
zjsl+F1|OX)5uu^`=YTjLz*yO3FdavQHMSIE_ptjQVdLfWh}q~~z^04WI80qox+va~
z769XGgC?EHm7HGA+wV);9yLyMj^!F~l~hrTpv$_$ICw;_7Kx3Xa)q8}?P*bNf<ahR
zwiNCeoqe9b$~<rg#d@j4Due#F1@Rzz?x*caaqDZ2RM^cV1gVdNrw1vvdq)uqI(+-2
z{TJ$H5AXGprL9%ebz4I%ju)h{9hcC0-TWU618}_%k4B3>hRNI##l@guC=>5nFQ}Vk
zR*38K=TXJtrX+9c4KCbi`tDaNa7ur%a#?rhw@Dd%4b9Il#6_(-&XVO!pNIz3SvrC)
zh*$g32b6l)z|7e?RhJ78_JSFXYLh3vdxWCPC2xr^#`FdA-Hu7)^pL!vBljuo<BG$s
zM&JuZjOLGs(rjDHn4|<<6~ANUpASHaPb7>3{_dXxT>2jnVh9T*-oIO`e3gC2W^#y)
zbI~BBE6N*jQB01bdnCW6Af$I>Ors`sQkoi)Vhd#osqt=a6Drm!C(I+f1FTHjMM}OF
zDWrguoDWB=UKe@v&l9~Px6fR-Hb@F(l@Tj_WoA$nAw4Q$msWb-PVKGkArHd^9&Z$H
z>j1q!OBdo9sitRAO5;;}Wg?xTAO%<QWpyz9?Zd7@NO%Hw(v*`dO%Y%E+Kh@>Tg1M9
z)g<q1AH9Xie*?l+0~u;Ur(XWhNMP+Lf0s5yh1)LeOH<IINHNu0_(*|&2s5$Mw6Axr
zR2~&e2?TS&HLIpG;^$ID#T*`rO+G{r&Z_UYe`KO@dCMQ3S#{=irwOl-yI@y(v%d1<
zNMmxs0}X3W>jItY9Jl)B7=>b_21n9mKWW0Op+~U5%wtir^h1J%QMA@(VLfIebPi@L
zrS{piFM}UP+neUQ>vV3(OM6(dO;t~AiMiJ#t~XY1{9j#H8P{anw+8~!B`qPs=#miG
z=#*xZ)aV95VWgsnFuGxcjP7nEMHGoqqgxszmArAc@?Jd8$M=2rar<q%uIsGhf1XEp
zGpo3M3x@<JI^iLgbAY4fZe4ww%iJOd?!QlMB%5DyEuO`j{CqSbzArbhBbi~A+pvP4
z`GWh4JIx)pVhw8qS9--O#{{Uu{^2__b;qz!j>UikSZtwU^&|CL&SWtF*3GT+>Ei={
zhR|bLNm(|DU7j<P*k&LdG8K~I@)<gJaajEB_~DDQPh4D)=40OBaQhOb2tU#2>4c}N
zLQAK-`yUved#jHWR)}D~#{;4XS&`2DM$uP*!ej$fswv=3Ia+=PTUt=A0A2(p9On^s
zs#)9nCLPWef5rY53ahHe4MzC2MgthyPP@2;a>Hf_l3%aKOHP)4V-yQBdztQtU@3PS
zzr_has0r>gd}L<=g`oGozOD}PBlGjvk{fKr43P$-%Hhf5_8|H{UQ+4<;b5B2TBf~y
zvps4xNVB?8z>=EouM63?&N(L=N&&=zvf-y<TUjy-9k><GNx%O?n?)3o<f;btZ;zyB
zmuY6UT|ZzP3I{c=3@lLde%Uu8VxrIDs?<J^0(N(;zXQYb6+ox)XlD}bC(joV&T)1R
zHFJYHc#RIPN%+Mwn{1|7u{Kfuc>^O8+k!R;^;&$46nktLN5N!RN<G>V4s$2m90>}|
zgj_*@;biO=EwI)&YiT}Z-5QlCET|Y<=VUQl-#v%IC$^l*?4}t^dr{BEHIW;nO`w;8
zZ8-9HNNw-2lKGR{@I&Mdi0g6}==b06SZDyD_aO`Si#!Ut9Jep4skRO2cPGgLopZr5
zB_%PVaTJV;1^Fe)85FTD<i#zJ4k5lr$cY65aP|iW+!*HyS?0#Szk}c!yBEa=1(+UI
z#~>vPN_r2j;2_+h-t$7!i_E0Dp)fNpJK5=rHmkdt`UsF#m0puxgt<(15Nbma>oPUD
zlse#B=y(Bk<ukQTc+UKf{)6rS_L=>=aM6V>1JjTwhKwS$QSG=`3Bp}r&#^`%CTsOg
zR^M@pH_D*j8L*<(B4vVUeH$X6&lc-rD(3h6dJYJMeg9su!fo_C_fkgtTa53t8f>+Q
zL=Gaj20#83ePl2em?A=x?g&_DbGVc$pbxD>iKS)hhlrI>Vo>8BStO~m&x3&i7&zmU
z0vuR>H{Qyu+amacl1s~&WfUyFMEEa_`3u^5j9UzzGk+%EREQ2GUqUd@r`Ml}lK6S}
zsuFHK-85>(x^0>EO@HU~;w2VTi<uH86XOo1H4pVO+U+?^#}@ebUWWBFUnZm{yD;@n
z*0X%sM=`{-iH(kJ4TvLWwL8*W&}L?2^*XLJ{DaQ>$Kvtk8QD6o@{Y0~8s9D1I-2QF
zhH;a(xPWwG<@MR9M%!wz2S<pN|K<^Zg0M$z-&Z3gR&lFWNIOT573z+2fMWSz{=b;1
zr7omK@aFlWiSgQ#{n^a-K~zl?QT;IPCg+4iFh;#kdG#{*M)1>pAZ3jb@urT;?@}7s
zvDcLM<N1AiXVwsgvjyV!%9kw(|EO+=UN0YrtuCFPJMAjYZS;A=(@<d?WYiu#ZD5ed
zN`h2t#PuP72v0WU+qhDO@{mX6CJRp~rHeWa<A6bd2RU(!>Dn*6;$#ds@p|<<6tg^9
zg!|0834d=b8Gwp*zKoDTnO`QpnPq{6f`Ky~WO4}Le8+QD17rd^ECY7~bwXJ1HK^_B
zCYY{cO_zp{5skR-5Xr8(aZj4Zp7dVUTwBWz54*O$#gSaM%G(I%t5}$t$B8WV{CC=)
zCB+4;VM&u_hecZnh_({^YwAVC?*<RBg+#hqQw*sI4~+`T&1TJ6iuZd4gPe8;cR7)?
zvyiX&I!mRuzd!+K)@8rw%;zX0)*V%HarWk&AixYaQKj3?EI74+0E7x-!~bXSYLA~D
zdtc2sj+Tu}A2_@mz4*%4ZQCV!7p(s4c&{Z@g!H=pK|2|>vTp*PN=xff6`Ol{<ga<=
zSDenR5u+Nl^o3T?txOq|2hwOEN5W@$e;~>?#bk*O*sqH|zTHA#UJ*GyUFe)3SMh-N
z)H^?E^P6urDX&g{U?2UH_x|Y!@<YZJv}k-Nl=8lpr(B+7NFu}>5JeaTPcg%7E{PB|
z(>_tuB{OQJ^tPOF9*n);W=);UiAghwEmh6Hkhf&pTzwNw1CfNk%jJ`tTo!&Sq+y*_
z0)b!#=8)SJ9*kHJaG553FhS;8)n$rdku;`D`6JX^o+<2VeX0QC48D|}iP3zx%>ZM{
zUicK3B7L@y^??d7svM0xfIuGKy`3TNk|ApP!aT9pwEn1-)e_DwsY?(3^1#A8o!rEl
zDa*c}O~YEV1tl}OG@Ef@JC!tZ^yhD3GvJLNbD`e(7w*7iu&VC)#CO!F7~VA4C&RaX
z&WkP3=3~SM4lhMtmNk4e$m+6<4Y3TiLQVI!5ZeJ%DPh*ScOqxFAHWktKk8uc;udy)
zW6@IvK3lwag$i>$!2ikMThN>2r0H?=O#2GWKq=>uTLoA%%0~hR%Pn%i+%U<yXMNQ%
z$F|~w5eD|e+kT<ix}08JeIHnl5BP*zG0wm(k{QLo41V6W(vIS;49A1+PqGy+r-|u*
z$RA6+RAY&?$lI0*8+~l&lwI&}DN6E{TE_=2UGZdOOYZi)Nv^T%-M_dKDyY4Nt~z#X
z$#lwiM{e*2NObMqskNS(E>Hb9OGWj=UgV|?piI40$d=HO_B+NX;h_UwYJ5(74jHy8
zrg_`F=POKYJqdaYS*J;|n2gT(F*%0#qu!Cygr0AiBb^9h^eK*ZolfplD>AC{_(*8V
z{vue{sHQzDVP|jG5@!vhl|V;}6z71x;*XDjx%N5m(-=5e(I>@v6EF~m7BU7h!~s<{
zXx1T<olNbL;|EVd{kqOGpKBm8L{_meGz^cV7$=P(?LL}RC!(vFtGTA;+4qypIGF=v
z;L_PJHZ-{RKaBuF?;0kE)$)zJYosE0jU87DC0=Pe{Fg7^*&9FUr_d{xgt2kY&e!^Q
z9VG=%)6#<qx``i9KTywNzNhS{C`kG;_|Wi;7aoLQYwOBmd+(u|_kTL4#y}X`@G<oR
zyVK`S=Nym)s=7Q2KW$Mmts^+6gsI8baDpyzjRAb-d#JO%Hw*@W?UDq>-39`8!my8v
zr0?zl3o&K;J3DCHE#w&k<UmyRs<c_-PC5<?T`Yav61ea;A_Cvsl0P^wv2uM?0;T@(
z63%x}d0jqFK%yD!n;!(F%9;4d+o!b<F=Q^T=mL0&Kl}#>5Eg~vb&RM98p^335;+;0
z8`$l{D`)eqJ-F7<XZljk1N%if+*oM!K)_WgK?*oN?d7$%e;P$aV)sy{hVL2UHjCUg
zIJ?R5GzuFZPFExbBL^}vUoPA}MiambT|?~D$5<rKNp68m0h9JHN%4%4*)&>UIjKv)
zLOfZf_UZggFi>?Sf8^qljP6SyKTuYp43e{ti<&1gA3&-c9JPGqds=SQ1UiU+hV!8Y
z5cf-U!lA7xvll7~cVQ}p!&;`+<HJ`l7zLNVl5eD|oE6H?a-h#1;Yo2HAIY~?zx27j
zrP_dqB)u>{w##M~`>3TDAe|re{<p$#DA|q@VxwuUXWY{%c;W-Q+)+|aDA(bjsH*+-
z`o~S75aW?$X*A3%`L0C<5~-YZN*L}YOjFxDUpC_MV4HBzJAV#)u?^%%()y@?I0a;E
zGl^J;9Pwg@z=fwFj`zja*GA%rSaRXC?9K<5)i*PMy#)#xB`v<ME_u^04mEZS^OU%A
z8Rtt1I76oodIl-ob`%|e2--ifD7a&vX9spA7Na%E$0t?ijNaj5PUo?>OnzzZww{o}
zFI@4?tHP1p$!SE^ymdPo)+nH9{BPq|9$mm!Q@RESU0MnNlAUZ!Dd6JQ{E}B+r!&<k
zdE=rU9+3BipBP*Y1iBR_@K>)z;;mHM5)oSK`b9HHb8#v)v)4vMFj@HY{$s2jmu_O+
z*e+Am720xA<%`gfKEJIuNs2JPd7P_DGFi}IDF+>55zJWTgdOU+lb4X0nxLiCxPXxq
zS$*6?>PM!|a1)_UVY)%od~(AQy5T&AyllG%5c@{ig#jTG!f8M&l*ml1_>RkIaNy<#
ze(yPWZ>Hw`%f02`22YY$3(FR63JnvhsXkXY*t}+Z`-|~59F^+rKVOW$Dq^sHz(81i
zr`P#m>}ip?!!pCk=rVKDLDN~$YJ~YxrRN)xhGhGZ#N-WE0%~d+QX@zxi-730a4TPg
zqM0l5Xa<w2FHf>N!=y0FH*#yI&IT+&q=+ms#Znuy+46vlrM9{dSCVS1+GBey@+Spx
zkw5MX5Nn~PD;osVcixJJTv{wIlLPL+!X#NPX>Ark?@Sa|Z=nDl5RjZE90=do@8w?0
zBBQS~RWS+i%#+yKh_RYak|j$Ww`0A|A``bHj&Ec-huZphV^7{y=wm>S?-eLoUZ(j0
zBni4<9V)e>I>2zhFYR`LC)~4WNMzvDY9=Dq?G>enX7B3*$sVxui78EYkhTvtRi_iO
zG?!;TN_NJh$Lk$&X-Fmv$2Qb7B#r;lK|_lUFxG%r{*I>U$~SS!St)iWEk3B@eRpT~
zQnSZHR@52w65#+js_44B+Cc&TcUBJn@JH+!5rvYvm+&A(LMz3sczsTZVn`UpWFBg6
zYI_Su8Px}72@3jav%;Z18Hq?ziJ-c;7lhB*zVvXk_j0Z8>3k$7%&#uy5dJns?12iW
zSjchXsqR5|iE0cW%Ko`og1%aQa?BKiqbBnF*pRf?E%;eJ?Df5w%8&@w-WR(Tp)3Iv
zhD<Vwly+MLG2355>*tw!t_IEH!#VKmFCIamHkd2k>!+1X4>N{+6Gn@wlrsVmgPNgZ
z1>R8fW_nTrz}38Xyi$Qp*VtuInlY*|-PEK&rM%bb${RbY_aCi#H7WC!7lI|s74h|b
z2!G-E)<ba@DI~aNDIl6Z2X4GCu9b1L%M=~W(^(n0A;xXuP}c~h-|vDdtFL^!ejV_%
z!jLC9B|JwKWfZfv?6ED27AwVQ>B^3j{NX?`7TamLRL_JY>J4Gnv7EeTdpWP;Jov1L
z(0%^K8bwd4`;G;=AsKW(bFMmeE5wM<tmXDef}uJLhMIp(FF;cfCXEJ&`4KI>OD1Nu
z=e=-0H4Ds?Fn!#!Zc#jlqmj5)hrP~0hv96211LodW-(Dyy_j0KH#C9LcGT=^9BVP~
zP1XF>y#EWlu5x8==o=&Q-qdb}T~O3uskcH-H}S}$BBAzl9}uw=JJLnY9wA?sn&RBX
zKhr%m_D65dl;{6opz4$nO<$}{F<#aabmyC*Fb67}vaMbU7XdQ1j0Fq1&QGCU@h>*$
z5~|ue-PzgRKTEy~E-GGC>r+R>CPN1{4I1x(au#{odG4{t8pf6~sFcl>bdDD<KY*K6
z%vrR1>G4ey?&H|ekZ46aOH-U>Wl>chAW&!w0;GrIO=^2%L+#w5Kq!}{kA)u9U>aK%
zg5?J_=khU_ElhVeFeRtP(mHu|NJdqNH?3#qGsoH9@-rvpZ|gUJRe%p-@YpOPy$pNN
zu}n0Hw$hyp4BU8<a`1RT@Z~V><=r=sZ4NI=6AHYJAZR+lmKb>Dr93(cQ=svTg4-Wj
zR^37dJqLCi1x^H~$fz~O-t6I2$H&KS-dr88)KqcEzy60Lg)ipV@>2+|72;+aZ*aPe
z$;iuHU~^r0D>vQwiR16d=0Kw1$CNR?LH@0}t#`rKrzOi*V{G=>Ua8~|zXX{h(D!zR
zF_{{_lz`GQ!Dip>e=6;owI%&#Fb4dmTDh7(2nu&vOOa<C?T;diH^{$rPXVI3j4hfB
z^LA5}y9cxj;0gY9@-jseDS?^GVdKLwvy@8EGpD8sbu&Wy2;FC+Oh&wog1ePrE0e#T
z>@H=!unFn8Q#@RJwSP7CYOaTC|2)j*(|5d;hM_y|x>k+Cijjv5d(#W1dRz;|4!_d^
z_I7x9<S=v(QWxyCR?@^Hb-;YabCb*@MaR$JN!56z_{F4a{4H(*Ycf}o3&+R1ICeOF
zu(9*F$=MzV;@DPzk=mTN(Bn}~jCR=#q^wlZem6sb_N)D{U&Dj*QE8*c>Dv)P3RVz!
zg_Eb9v8;Z|frRrhylo5}h%f0PXSDe${)C8H-U6oDl20afEO!kmmy@Pe;G?f}SY9)#
znsDU?Ynk*6w7b;q76*&&<L9Sdq46;SLe<DJXl$-Z;@A*IILOESDWEC#*H^Fbxg!!`
zKAHu_`O9@wh!q8URoWRHpdr?wq%oBZe!U{#zfTCbGCa4=V~q3AEhl=}(8FSKN|g-l
z5l!%wX^<bewHiCy!6b=jyN#=qP%+CRDTOV3)Y9Xj(dQjR)4$d#ff%)CT8jO4FrE`X
zO^Gq(0nW51<N7T(lB#u=HL|Fp52Q{aGn05gr&P|Mj1PL~+1&50AeDP>cb+1)-SexY
z!a}!(=KKPQk$|0dq;;Ob1(bbS^;(IS>La2Dn34itjI{#e;(!(2XZV4L5PXx7Vpz$F
z#yl|o;wE5(Et%(CZ{^-yv_lFQ|7k|nlFsNQd8N>jw~QA<KvFAznI=r^hD)B^<pVtR
zV4xKOOL0yE?uZsPeG$21lEQF_Kjuendt21#(|01c=?Xt%g}0OMwR<Imk4+CYtPHUh
zY;bxXV`y%<BH%XAiNW$8_6yvsqxcZ>o-xuTJLRRFJypWlRz*J$nwqkFkq!G>tR{`|
zXY$hL8LX+<H-@Kg-8C|rB};<g0!+_TQax<j&H;jIVp!bdXta!7Z31ulR#->GyjKPT
z_-=Y=8fiX!yjpuMz`~t>-R*93`p+Uu+ggSRDEYYFaeJ`-4E`ZCLE@idB*0yFykbYE
zA_Afa+;!2klNIvgc0Cm%swJc8OSMB4EryIVZpp2d%K6Q!hQ3)MKP!T)TpCJLrKWR>
z^t{Z5WCTY>OLq02<L#s|O-wTf5npdcz6rSKk7c~Eva$Kj6YT5ji@P{@MyD1~n<pRA
zxW9gUFf%iwKjCw6dV2b}LG$L|BopCiI5ufAx76l+bi5<*9^)nsg)<f-Tu!E0P2k^C
zJ@&)XClk&sm#fZ${Z@B9C4%yxQ}{f&xDtoT%MgEmRIt<cM|Fd#ppFu=9*60BUkWJS
z8lO?`?<Kc5yrgK5Sn~LB27gg|S%9o%HB_<4C^=MVPt>cexZ^w%R5@3m@}o{(!Tbpb
zLClVx=c;R(d`3PP{xb7?t;2f!JO~lD#=vD|72N6j+5=~t%P}Y6)Ai)y$s`XZVYifB
zS}}4BOP~Ah_jYAmys}3w5!HaDpB#iy2x2>Bjws%3mkRW5r+!*nb3m^z9FsBzKE43$
zQOiOah+7;SQOapXcDarha@+KJcLpt<w%s?ZmPHP|E}DXUx>sn_H!-9<d?YLW%hNda
zizI?-f)ljTm%$_TKpW>n-mo-TzS&aEcA)AdP!d{oT~qPz<n$xIQ%#Z=#qC-b=JNd^
zIb%}ne{U`h=0I%t6EZf%(X3y?vA(89=w$<CI#^#>bb+NsTyZ!p;TCi&aVVsYUZpIY
zKK*AgyKlCoc8`m@JDxZx{shT3;SedbK-0TDgg39ekq&L8wU`tj5LZDf&#F>QYs}hf
zF#pjKzR4}Ho}>!7GILtpzycRaBd!%m_(U^*kNX5|;=d2)u{IeNGG2(OpgBxp&*_Oq
zAI;EG^yB;zz7TTFNLVgjEq+f;0xL#>5M=D|;SFE+;SO%3W{PTL=-SR{8iO9&PHuMO
zK_F{1Z_%-lTyDi{$AJ$%x$(`v8cFwO<bhyp5S&XhH&E3`sD*MJd||-Wm_S<SmWu_G
z`E_q@E$Y$R>ZifDuY4P=?5k>JhHrlDQHOd(%}5W?9U}f@607JjMCDa_J=fm(j>@a@
zpViMoyTfp^;{*Jk?nf%zK@cnQhDT4o&^0tTY8-+b<)pA`mpKd=!>+kOrR@Iu$$Gg{
zTYiX0lwO))&O_-CPbi>nP;7(klHO%)`zZ@)u}__T+g6foD7pN6YUf-aBr1XD-ik_T
zXH&VIew)pqX)i$f($~#Zz<FP$)_AXWU7u}iY%r){%c|<EyE{mp6BPL;@p5L2pMH+v
zi;<@cNvu!-LFm+g!WeJQfR1pFnjl{V2KaqYhf=D+LMfI0_lN)IoB!`O98KQf3K)UE
fatrxO{cYRE@}5!tR=8Ib2zY6#>8aK#+lKuQwEX;Z

diff --git a/docs_old/assets/images/install.png b/docs_old/assets/images/install.png
deleted file mode 100644
index 2b89b72680c6dff599fc96dc5a2cbfe014a48903..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 12669
zcmeHu^;?u(*Y-8EAYmenq=<mh4GKs%(y4&bokI*N;H}an4bl=r4-JZdbTc$acg!%r
zFvK_NgZFnF??3P!-}msthHLL@*V=2HYwhb?{G*{JPfAQp3;+PBqJpdz0N_Bde;*JL
zU`vvp2{YJVcy3P=b%?M(engh<u+JbjE%_%vNgwSx05AfIvX6DVO}A$V{dBCFF81eC
zsBPZ99l%?^_nPr_K7{it@rEHFrvVSCgS;o*ZN$u7%wmiBqo!<qpsein_mC3ODMNr&
zP~#?Enbys_l@vn?iHR3TG2yLjm1vI8o+ZD&&1GdX>y^aShIBr@IiE?FD}*>ij1b^g
z0m;HM4i`02yA8slkOqDg(nJtryuZz@ctj8~J7+b@%XUV<mGnP-0UWX$jDTp-R}quT
z&Jf^-z@LG!qYJA*fVEPBRrhOuMg~AQjPT#H{wn4lf^afgsZ)mKeou=nuoL6{KR&S$
z&V0sWABU#n=V-}L1J}CC?Mmz5Z*}QtUV-Gg&PXUW@u^z)EO?MoChQVj({t6qzjyUg
z{LU#5sFq)e+4pVC0DHN3W*4e@-ExYSLq--<YS!HF@w?94tp!&T>-)S+LtT4*m*)Zm
z5#S@*R*j5Jb_^>mS`pFC#ox#v@)38U5{;vdUnbNzk{q@9(En8n=`|e5{oVqVf^(!0
z8r_yDHT7{g1j(|ER#rfsQU2til}LMYd6Kf+8FvS?S5Eq!eQ{ceA-JI*Q=W(T??yCi
z#Qs$UMv7`kzmUiFOU~PGPfjCoGq{iU*t%j?f<GZCzX(RiU#h@o&dA-8_-+}KJJ($S
z+K28-9^C1hnQLRIL;ulE?2F&7?bZ6hp+;2;KQ+>5u&)WbFFmiI_0hV=^X<i7Rl~Et
z?u<?|thWu``7yM_d;kcuIK>!<4mEZ{TX6q<YwN~d=So#N*Y340lD9Vj4<TEp&rpbJ
zn#|pmvdhDPq~bJnRTb)GzeNz=#h2#tp1<N_&1_5duW<FR>DT-EC?s-c6W#z`<RljS
zHLyJ%tkUb@UJCaAOQr81XkkfEn8S<Wz9QM_!C;S?F^x)>;GS5!R)7%Ljf=G07=>x)
zxc8MrPhZu(TWqHI+MZJ-I&3N?r)#9bAVM}>$D`!BdboT;G5J`3qtxT;hinQ%)htmP
z)+J^Y3wsNH&gpwKWD#c`A{+IO4b$0-X@P=5Y)>qUO`Lgjh6tIrxR$XdD@U5G!LCR5
zuocth7jh?DvNf%{328bH&P3@Vm7HE%X4kO@ESrD258YE#tW}Smp`R#ij1xyH<`!W@
zmP>U@zZm7jM%+tK((ada`dp(OX43Uos^FnT=`{1`j;*2~N~u>pE}N?}v?`$C&VLS^
zpL_lj@M-a|GnI;aJ#H;pkjf@HIgQQH>#R}76sqo-5Z~Dp&5O>QWcP|YwuHiG>5Eg`
zBYN&=-;<cqQ}Wzov4OX9IqF58p#942()Eas^g1KNVysqZCdy$4aEdYS<WLlmQ?9&H
zKTp(`@X_d4tJfx;?`x);JyXZ-53Zt=o>>Iz+=^2k3YS$3WR;n2I~?OFidj(nBQjx~
zmTMpl@Y(sNpE3M5PU@;VuND-*b~xjCZxCD<E|eDw{c3ccbyQSWj&`I|v62z4zNw>Z
zV-R=hU{m7I7&N&F3$%}7$;(r6%{!p5W!I{0)VZ(Ror97pUn9M$7*5Fi8v)GIr9Y5E
zsL#)ne#Uu<IIu`mJ=2fQSthiE&-x=eu6vEhWP~&~8`mgL!c?O^R_^(BDs_pMyQv4C
zuIG4ak_S<rzh~Qu8JT{ds6y@}467TtO0&xyUG#E_0;jbO$LIPVVGrJVH3-rO^{1@7
zEyByrxwW^tI!19kJj;|=ZuVJ_wy3oU2_3&{BE}L*cYcZ*rl_gP(8}f7xKs7hobHe-
zr9sco!<NqT7ztOGnGf2=uwtTUL*8Uz=yUo182Xdj*zxxQRgX8*6R8oUP7mei$V|<Q
zxi1`_nDd8UP&FH!c%eeXw`jJiJcPZ^bYZ6*$lAgDw3}wSEJMt9mn`PYc}iNpmUl8I
zeyI4YB}M%B59$T&;ThTkBbjqjRi2%isguF%H<k?M9OqaACw4!0-_s8bxbQxwU!US?
zuvj-|uX5=Ip*J}lnCXefGUpcd+oEUr*uMniNbSV+tnyJ2hkv;reD}B1DUzf=kU|3a
z8UtmfvXwDC1Gl;O+hG+oI=b7OaZ<A5B*$}$p4|hIlIxLlvklGWs0a5Betg+AMg*F^
zE24D1$YydusYWkV7lcK<Jvi!;SYDXYXDv?uO!<&SnEvB$^q0PgZ)gdOsHUa&Bb4B3
zbgvyz*Gn=TuT^+}gFf=OhBg{hIwDS|V;zPXm2C%#O}~|-PcT^{`<GoiU-$iVy6qt7
zGwVOa-6>q|@T;{r-SL_>Vwk=&^@XqX!~;>|r)i;o=8?tW;srMLb}xEey9eo*H}Of>
zFumb3h>Q0P7UZ}LwC6K+A%WWy-+dP2iTSRk+&Fw#N_G)%YRpt1QO;af^ID?TPdhW4
zr_=n-_-zy7KM_ZUOVDj9pd<TuX{EY&>76||ges8ZK{V!&Hd5&Rw<%iR1IYE-^`h(R
zLy`>+&#3Lj(!F%MrUR{iByaLkgI_fUN;x;w4tMy(MGp;r%$jo6*&65RRF26RO<E)4
znbLUnUiSC-X2=m&{s6q?zNnXUswSk2{lw2`H}u^-zwTqMS0D~%YyiB`rA)QfBOse(
ztDQu<*0Q{OC3<i6XK5Jz20%}+T=q*pzn6D+_|B=~MS<?kc%=mzZpCg`o*#_@$MxTF
zLC|ehY2a#>$Q6i12EKSOa3_(W|LO0^8MS`xkqXC}NO(wyp#P7b&0r<LuIL97#qFTV
zB_XcWkUuic{w3pb(yLP~vM25H=syrqY58k_s;}cjqUBu9<3x3=&D$bEzYQXfq;eei
zQ9REy^&-;ftH@ba5c2o<uR3wK1OeNuGw1|<PN4M{cZV3NIKk;UsQwsbLU*&gH;r(h
z)%nX)$Mv~U&Kc29sfAh2_8j@~D5Dt!L%;kOW)-TmZzabhah;`B+r8SZVA}NlZ=>0s
zNPk59_GZxU<wBV84Bdeq7mQ`ll_2>QOGRJX!gth_m8X!?N|-|Kr#%>)X7)U=iZe21
zCkm~%6<I<T&C1MG4#r7{^|7>joTFkiqInqqR5!jkZB`}XrrIoLv1nEN_k$mT#U<Ru
zvA9!5%7@G%3*6yqXH2#lX@3+R@&+F}>xk!Q=6tiYcaDi#a_KMZG65fTGAK`5%$;+M
z3*^FrG2OxS+5S_pzZ+A*_m|Xpsz@-0PG2|jWWLJXI8?NC)E&R8(Q1a?Ve0g)CWQsM
z?S+!&h5y<ag%}Q(Byb`#&%oyw9iMg^;ap0FiI0o#G*dR#Bo93Amc};qO>wQ;DBdL}
z+$?qa1ILadt?A7H)Hp{GT+vGhBaxwRJjP+=mBxk_%pdkB#dg8AuYZj`?+lX1lXiGP
zN7lsG%G!9{WQdJf)m8P5d;KOu7s}VaeClnJKdTi&rb>P1{=S^UxX?H%xAiMu`%R_(
z!?@@fQN_tBmk(37vI8X3w<cHPzMVZoEcRI@qMv$>g!sQ1RS$imeCdng?ki4kEV$Nh
z`q%Vt#(nulTyXz>LN?t$(dSA8txaSX89xtx2V&2aTONV-G@O#DuBUynz8zZAs1+4?
zFOfrW=>E_7luBIsh?{ykitjrf<nGliuk)4_!+FN7G8%*ar|+hC@IQRFLt_9|>Wx9m
z%w0J{BYcC&_jYt%Bp^0ZnusI3yCt-?OX8Q(wOIt_@Y=Yddg(qE!$_+B0CR>FkuW!K
zlB_E9VaLu@IRP)d^1Ld`p;>+DWrVVzuk4qeK%v#Be}k8?0@K3#nC_YK;)jZYj_UFE
zwfn6luKR0^)+&jMupQ*k6{Xkra9si4V6icZR`B41Lk(QKUH<r%=SD4Mpoi7T!IOWm
z8w=ovS15q(iX6+!D3g&5p9u13;(-qF<z1ju9Rj>c=VJZiWU*ZyaYzFnJFPRd{zk6;
zIC#<+dk(70F7FgUIOZyXBymsv_#P167WN7^`BJ&_vWv6|F$M6V+#vtve?k~W^K<|-
zAZeuZ2T`#7o@+sXJ2Aaaa{k8T80_8ca6*gsvI|8F4qnSA``h<}f7>dy7-yFTD*NN3
zFC*`uPd6EXcFPJ4j?4YENMi4U{?)O63=P}mdo}=i$d(+e^k0!&O7XuU`QN1_AvVd8
zk@sG7n6fOkhsC|`A~tZoz(wqLFoYTyA|<&bBusKuliAo)9?E)djdSz8?2AtiI4SVq
zy!ExvZi<DKwJQjhkm+8|y1K(~ZrHR-$F}Q;ru?YSP<t4S)!~+imX6Mdd60kj2D8_+
zXT9aBCERB^*{P9LUHrU8M-QB2%0W*xj8kk_75(Ll?Y{RHn&Lb<_|pXL)LsE_D6Px)
zYY6nq^zH^UGCMoFa<Fa>z@vC2g0;+tQ@oEJJr3|%j^jPqswgH$$X}d$`5ZMKeP2Ze
z8$3r!aB*@fJ2syQu1~?BH+Q&V4cx4|z2*Yoez%B`aWUG_R>!43d94p#p?|#H*E98b
z)?kW}xdGZUBaM%qRC;P6Xex@@VY9+0FU`m)TOO%5kJt=m>vFhmuu>|(1N@hy6%!j^
z$M_uMl>$BS5$Sn0d!iSoB-z?|;k<$8IGcoeFGA@+rFu|P*VT_yW8~pZ&1bsSy-8e{
z!)?2w`=@juJi|yVHuSVt;>_;`uxF``4*)gJKbgzzClJD<kYZJ!;mi!=eE0ypQ@`69
zYtxiRO3km3JLeaZotx7c&8ocAMrkJb091NN_0rn9YdcrIbna1yf;JeiOBl?0Nsfkz
z?Z&30(EgK0FO6fqw`m4G%T+-)UZoLK2{`vjp7wzEu;ELKm4ul}V5dTw1E(uAq>u{g
z?HtD@FF`04!90m50A`I*^8ON&`uBNth5@H8F|l!RavFw6iP2@(MEZZ!hRR+La&U9U
zt0oJ!Eju%VO3@{EFF0=ii_E|V|35o3^@_T>b`!O3dG=F>%Ess34(FU{)Ql9(*z-Gf
z5VJK^h5+NI33#A~k~ro_Gma4~Fee#d>!!I{Hsw{Z2rbZ6O2PzO2U?rFy_s-`Xmx*i
zewf(TqOs9i^@}~u@<CPiD`Y~$T~F5Y6nAGES-H4Isc{T2%Z5H@VQ{OX9SM}+Yi`~e
zM}!Bj>r#Qh`Rs@sqm8v=Qb{$P=m8#jtD@iXCo{?NQjB_hOlazZnL5vkuIUD=3O__A
zI|3_tMIWD92TB=7N25fSj;n*U)<N>xltBlZbLx)uC|yNG8&-13V@Q$dSz?N77g1+-
zH=m61OwQ-e-+BTu?I|5%h&D<)nv@R-&u`<UxJ<6|+8;kGG;t?Oo!XA>>QX|tt7w?P
zJd2BZ`V(gAJ>mI<eK|>@R`1u4$*ybnJHR`&@?%K|TLwDO%7nSo6(L^lJqqnmvgoKN
z?q}Ao0y_@35v>tYKj1WihJ_8u<>jA5AZ1lVECK@05KqHmGiB6bxlQU7bT`M!Va`zd
ziSlQX=W8#}d#&WfcEyL+jsxb(vRZm3>brYpFtzQiI8sJky*DGlyUBJXM?6Ix*)J`B
z6c6Whw=5C}E+Z=GW14x*=>66*37{3utDz@z6~(=1LV8sbyz9M34Rxfn!qRFID4k8@
zlf3j-G;2G!W=_Y3@lgG{&8J#k`x@qrZsl%E4amMY5H8{SfV6c-0^ggI2|rFyW}Qj>
z&5`PQ{`YfoiX0sg_&;#zSm$ein!t;I%@2Mf`NJySiY3G2Cm?>f-#e$-@|>?cZ(F27
zZU91wB5zJWLiWeboLeK3lIR0%pQWFnz2CRas*NKDi6QIjHi&2g`@t-Eaw>64=hV6{
zy+~|ir)TD0HR}^B`Y2%NyYljbiloVHPeKW}*Q~!DJ@}xE+H7Qy2qJp%$jLIz7(7!I
zof3QV0OjFvM9ARB7$I?}pEYHLjW5^%dY`2z*X<>{Vtf%4-DF9jZgKf!)gieQj=Aci
zK;Ywn1oZK?7BWo=`H}B+)2gzvlI38GdKY4O7miuflxogZO=UZa*k(A8FmecJk+7Xa
zrNZVc0<yT~m~K+<;)IhxsCgSqp(qmYa#x)7etSf;y>A&oqFL0Na0W|%#$tntr#sCT
z=TBSVPHG}@j>l)D3$C@vm82@><QZ_VbNwd?uxB@BH;_?`*i*qXnU=~Ep{NT@8PkM4
z^4Y0<>0lO_dE*;Ty?r#wPIQD`dRuiMGASy`db)|fG2iSx+^^tJNaC$bC|TV1@3Iv`
zWzs7>X(sGC^&e3p*0)gE-P{$2+no2tIh@XrT!nB|axP5GVtCBg4)S<%gx=T{NY$L*
z&#NFpf))wzp?>2Ve0+SZN~nTiBA04iqMW2m&-ptCw@E_>#KH%jP7)sRIrC(!P_e*q
zC@z%v>?dxx#8{n<d)4N~XkpQ|+9b@$3DsmLJ+gRqG$+J~zA(*#OAG<}n3In=&su!L
zY{{Dl@-@H!vrj)eJ9~GwVocG@je}6=;YQ%WzB~e{J|&w+ey~2%eKM_4=pP)+>$M|p
za5SnXa1Z4KH+Bo7a=2r27HDGAyw$j@Wn4aYa_X2cr52JVtmBK?;8K6!`x>~j%MpAs
z#Ww!~(AU(o3D}pqzuO4g8sL$kV;dsBMK?Rz%wt@mq5H|=UC1Ho0}EjL!8W3G=4jgF
zMSo+s^|8xI@vt1sru01Eh;|BQ`42x}Xn|VRTS8Ns3XjO-nWgEO_{62Sw6C*0c-DPG
z{Q>XKo%9912oysQjmBo^nZ(fyJI$C=zFXNQ^sH#i!E$(|mz0b(^ag-xbH}eFL!a-p
z=K6NSOCO%CB+DJW7`A9NsIWG)5Ey?9*q;OWU(XMQ)nW=uY9}`)M%U?(EB(+icxp9<
z3#J0|g;sUgl%VS!-)~6~K?I<CEq9lk=a8ajR$Z|L)-{5?CqL_74KfXCQe9+j=YsY^
zXdEAEKju{-pLaEQb*Nt~tl_fuC@5Kf)AQL0l|lWHZlE#!M9zna&xK0W7bkCqs+}~6
zpE9c9(^3IaLK1$+)m~V&_!aT+N3F83Pa<M<jon|K6&9Pegseo!UI$RP2g`b<@vQw^
z1%t`i43L*;KDov=f{AJh&dIa6f%+HYwV5v2AdOYI84q}oC^(5O@O0l5sVZdXmO1OU
zNQ$)c(+qPP<+^u@ArH7XBj>eSd0^+LTO(fsDlLN#qm?L4y_<|(J6Wv@(_L#<P5B}Q
z@Q88DO@j0d@U9;(ek1lmu5_-ZLkl#2PTtiFC1&9BGT!eeWB~}sAwL^dt5kD8iPW<v
z&5HP+r---}ol)2^Oe%Y$M)i!`Dm{`u?j+n1W_ajKE|bc6ww4t!CKg&zb|ere-?IM8
zq9^D3B{4+}r>-;gvTFnF#jxq;q#xD@!yUs7=LlmjKL3NA2BS4p?m^_;zU`&pSZ{3S
z(&jhJXR#`~qJTU9{?=|cvo-Tpi`lv_y=W~)ts=4PaDMdY4&C*1QI9XE;EYD#g!OEh
z(k$dGt&$V4)xf&Yo*NCG<=ztX274_y70&a_IOY;KW!HhKgjXT8!@-A$b*3}}uikhG
z@Fw@pkiR%8{#6~(&t#{jvfJ;_yYC(YSLuOPpZ?9&uOF$r6zV;<mr}iYQcU@zpiON;
zlVGnMkrzca9LvkMtSx#Q1tcjC4m}LjR{K)rR<-6FZhLNzmoabklcv3cXyemT0v?**
za92pgr7l|Qn#FN-(Hjlb^}9_*g7*1hEL34YjUAjj(I#MG3^vZ<9264Z(<^wiGUKq4
zH{pR*4UZbP^`RWZh<5)-Ewms&V{?Y!V2;86AnJkso^;q~*vaF-LxY!{xAcrHdO;)f
z6MhwwdaDOpd`EqwKXpmFR3(_MKt%D)-x{=u4;3kmJZmZTV0C@+BsHOgUq*>s;y6;^
z<%98S{W2~8<kC%g?yx8py9epav?&Iy7BvoGJh#dn(JNBVSpj<ggTb+FQV;qJ?tDY%
zi`TwFw(!}2M#2iRU8kw)-f2ScSmrDnj`<nr;q=kzKkRAugl2_%ljh(fGJ!g?&E`QU
z^iQculM;bpkEdbOR%bkP3!~_1BSELCCr#W-g*^I#C;=Z@9JOoC#=uA$gJ_vP%%Z>_
zZ+2GoO^5%AOC2N2_%%iH46>lL1XX{eHcseMRI3WZ%T=Cf1swCIPjlB*{U;vi8fZBI
z2NzmU9^gwM0d#+{qw~|F7e)BY0=OaYo^Fh$DcGT4g8JpNRD>N&8EYIb`_ruZ22o?i
zrR`*-5E2H{SJeeX0aMN1WDtYx$u25AwaBsBi7?Givl>vp)1hR<_684GZ`>&G$qOwk
z@>%c5x{-3P1%$hFyW+>=RnHCaa`~Q=StULg>@YL1x$WSDoc}|MyJ=qE?g#_DF9opq
z$DSztmwrU?qU$<IORAwWlA;XtO$=M2JVzUa(CE7|AtN^0{(W^Gn@ep8rm?%dv|qg&
zq`+BDlJpSio7gKm=$$UW->8|TddFr5@Sbvj{nMvtzYfwsSFx8u*UP9RpVw49*ACr2
zAZZPJwdOlJJ*5YxE-aLI%!6^p$q1>k$<=#anh4G3tY{XLXksWjZ87xLHV=tAsKO0}
z_%AR6=aV>^q!4vU9{l9M_uu3vzcTRyYC>Q8nj<p`K+Y>YiRB)gpwgz0Jsun^rn4m-
zVYJnyn4w8WZu6f@HE7IMiEV+AG%Qw0gTc7ubX|e-UkQ5yAa+DSnF|PS`Cs2>LU{3t
zVq%-mPc}0spRh<kROxZfA5+SYNCk2j>b-J11jf9`3=9lhhZNZjXp|X7^CARp7MhAm
zDZ{p}c)rDY01=_&TTO)E_qzHwy6l`<ia{PdYZbj3sO5!6cgt9hfTzBytajw{-$`<i
znvL1h_CGK?pIOW8>}E*)V|rF*nAwdP<>`m>zIFr~G0+XAx?r}4*u?5PWm-h%^eAb!
z#$B9gKIa|)8{pOGD~rpokvy@8@*!%P$M6!c@&o}}3Xp&MBrV_|(;EP;(frB<NE3AU
zG7%`yHc5G~o|o=Wsn=Yvz1y5#8^D%sSv|DzTo>5fAFV^>s_xjhD?@Xe$RcDiSg&A3
zVY?_`bxmO#EsYX5AsT*x!py0c>26ByAKxH^i97<&<BQ=Of$TS!C``hNK1g@<m=J7<
z9w^%{e7}O2nmgBh-}C%q{p$3L`QE6n>=4bs2y$Rsj5}3WvrlA|b{MJ0N}e26=CS$C
zy#4%Po|mV=bKk3G{4_d3mA-~=X4VE@8Z<VYw3??GP@NYf^5|j)e2ra}cZ<;PPilg#
zzjl{fk&=TjNBYBQ-wi5W8elj>T=BRw3Y>0iX>^IEM5ywb?c{xw(yJpEeo7V_+X?<)
z7WrD05Ilkt5nHLt>$)H}RI9Ge!7<IgE8Zka?1eXH$Ufq>HeMDZaG^u_=5(MXBELnB
zakJUbiR<2;lk#T?6+IIVXHI5^=0t06eZ4}sF!>EjXtMZ3T~=^Y9_}ulR()R;ZpNWt
zNQImte+jAG+SJu3JN1IJ;Vpx(iyznkl%b|Z6*e{K;PoVNacaa)I#7D!Ji^zpzE`sF
z1h+1eAzYO!q2unr5Z*6j&Kk(I#~qFxh54?2NOo)(OzJp{2yA$@<GugnO;Fa#(*e~f
zf)zO~xIOOTzRr~A_PQWDrHOlxvEP21mZpJ)Llkm7uaMfWQK9*2k}t0>STHsHr9-vD
zp`#zXi2mwrzynoH0KHnVtx_7y5d<v+?ZH-4U(WoK{rL;;nvd7+n5v@y39+ch)~jus
z<Kq_*-BNP_Vy`4BQ=RzXZC9*u*K<!J^ZS{cL1ERwj*6D65s=(!`4<t&4~Orm-Epfs
zpn0StZhb6%(I&LiAz(IIrX*wpqy{Xv4IRD<FZaTz`5gVTszUL=w&<Ac!URS?#P!ad
zlc^DZzQM<hW{x0ttV`$2g5TmBtLs<%Rih6HD~ZoGPToA&8}Ju<`6;7-MoH*4fQk=}
z=~Z%_vC|Qf9y975g%K5%#e8h9%1k%V&QmM%UFLsld&l8b2X3);_O6dZjE(hZa}$sI
z*Mz>d!#P1{nH$;i?%2EbW$PChxBe=@n6W6_#vB$b(LQq18OuQ_ehU<;KPOaMBA+HQ
zY#m!*PmhN8%zBXaY7Uw^hfah`v$JeAIUn_=8okgSo}@G~2S%FL6Iy1h)oXW~&UOZ5
zio-H`S1KEQQRQ8AM`_^At+!4ca`q_e**&l#ZzH{OLeu?zaiq!{<Tle8?||C?&v&wI
zu}{wure4po?QGbUL~HtFkvWbS^_w~d-45E$l(>qG6Ych@9YpQ3SN-3l!Pb|-o4rC-
z`@`y^`VTU|3G4T*W(b}#K&M(B1Z>zCoJ`t1hLWZ#i`ooXx$e}i>RYW>KVnD(_l59M
zNjNJrHO6$=4}VxQOEH(n$QP;zY>4DHUu6K7ti6A}eP*;j&8bCAvBS?o_JfA|jha|n
z(nalhH>w;F49~7Bzi&8N+SNFJI@gtRgH2zXh<b|1P%nAI?@=R9X>;m<v6`4etg(Y`
ziY?Dx`Eqwreef}5%qy`7Ix)3&Q};|1br^n2OWGmsEyAU9G)WOAAk<fOagwO$Lt6hE
za*0>l9|r~msKbsL9MRce&<4hKI9FAdsK`Ii3bofy(%0Kt^$|x8WlNqO5Wn%Q=47KV
zy`!!)4^~^KP<lozBe1$pR4EIRCy;hfn9<vHsMy~;nhSi1x|8bCo;XGrLN98ZVbsF!
z4kPqWAzNPyM^Q6XzEqHsh8LlyPfCZP=AR%p-H<C>p}zQbgwkBCb!UlJ=<&r!ing`-
z1l@XFr&bJ61C7-O;VhFF^P|kK+O0)o|E%|p%=ybZ1U!U!AoXF=nB|0R0bM5f#ct^%
zk%r-|(Xr8kv?77R!vp%>^PjMrZs4YvUgNY(_7rpf+Kuguo%$n5z65q}o~V+vv;ZWR
z>lWmEOk;a0H^H$W(QahmwrJhFER19jImQ#yui`ucKG#j8|7z@}%+WX#TXOx7;p9%E
z+#Fv&zj&Y`d){U|<%wJ#eAo`!=(uv5p0*vz|M+D^ZK2{k9fd6nI?5?__@vhR)SJ_L
zW8k1G4y~(RYHBn|lW&AK3M(JovCAn#X!p|5BDfo+MU&!MeAkRyifNkSB&IwSl5DcL
z<&P?>lLnDr?g_yx`9b@6VS{GH1@MDj>la=Z7(<x5p~VhYA?)h#(fIm`Y2{3vN7Y8Y
z1eiPt(63i@=T0oKgm22uE!HKO$P*q24^EdYELpH8B9wB%uCRvTOH0gY_s~{tLYrsn
z$ZNUxe6b&sicL#HZVOF#IM52M2_pj%9eV>u{nh6l?fNv5el}gKJGS{y`xJLqjvlKr
z=L1cLlXgXS-6Y&!oi$o@7d^}$7H}VLG=lB8p3X1I1rChDz!2`6!#_8_XXBxl7S+S%
zl4L-dl&{zPXE6us4qO|)XDIkG)r+FlRxC3p@H<X^r_D~V>Gx;`TymIc29Qe}M52NR
zV(DpS;sQ>%gc8Fn#YQ=lLz7l#g^mW;`hq~t*pR!t!aU)Y?uMBAD~0qzv+JRYrA;YM
zKStWIo7tV!k%=F?sXWx>89e6QBa}`Z9vR3am^Atvl$R;fZ*jb5F&5H$l7sbmnJwTR
zTAs!^=*7V5RT|H7l{7z-#{R($8|YkUE+N!wcXlk~E9o}pDvusmh}~3lu;-C@eIMzr
zM^vZru+bLz^{;)3yHet)_JYtIE1KgP=g@faj;d_vV3QrCh{pncbfSrymG%54xaa*&
z|C}K=jpXUbq*i;l4CR|?%h%riyXqI4>*REFv!o4EZ!erCwR4A=+aqZ7cAFxpXAey4
zC{6gs-tOa)_JKxt1u|X%BNhWE;eIy|4E*U*VP|U}uyAZ)*MJUkX<FwPEqD|ZdiZXm
z7>NVEm$jb;hYy&SOnqG&6C<TNK^J<RcyGB>Sh%;`mc(9>IU8t1-8a57M~G`J9tL_E
zzli-ux`7xQl*A{e#rPn16ZnzJ2SpYA@!Y8qrH;>X8$D0j&cD>=IpiZ<58%1nx@&V#
zvoP|Z5NupYeFmWu_b$E4CzP}MY9(Qvu$*Avs3&xm#;K7C@hwP!=gGE*D0=BeZ^E~>
zu_d1i3q4?DFUKv9-<#}?aLWd0lpaUopw%qYJ3et1w4FGxHx?_?y){~xhyF(I)J3b$
zyacBN&a7-B83Ol(k*)K>75S@XQdjx&?IO|-)olCQN-bKG+!O9_>GjKn26;rt%s93O
z@(TbPivld_HI}b$X`xQ`7|Q5@sx-$6;f;5*64`Sc?1mVejnXvwy0sUrp@qwsQ8Pmc
zh$jPeyqXY+2m*m9svkNgr=uTf6R%S@-=2CxFs3-QPwV!Q*g>8ypvt#H&>?<#7aAGI
zP*fS9q9)LqRHsppDaUxuTP5B&v-k>$&f2E&#xCON=Hx)#M=HL_Q@8L}<aHONe87g?
zT2)cHzmK(3%L_tUqs7Uw>f?cfN^T(sWY#)Wkp!uH=xoF~_+ttpSzATjjG@8>+#$2;
zc2;(ndgZ{W2^~_z3yMsWs-H87W$4EVzY0-BM^<NP;(`J4j`IM|6kIE&-w7oU=~EI4
zfcyo`b%2~%kmr)ydyM5JL-!XL0s5!*rkCU-5oEF5gk`Zu)-e_L2gSo(qW{o*&gFmv
zB(w~*4if)3-f+n=4&W+~Gw$MCQi1as5N2P%#{Tjd2nxiKp|#Y1`NmM}Bi-113il=9
zC=D!8UnZI2(XvSc?)q=s{$^JKph&w*zLUcF6PC`jAnjZK&!zw@MOt#nk4m>#U>`#T
zg!4XJa&Fj(3I03re?|Qt6`^G_w*zST??J<a2K5*hlsrFi{;MnB^C3{z?TuR;nL2&9
zy56A5a6^|It|z#YqkP2a|J5-9P`bS-W&%N<8B6O6(2Ba29UI3y@wrC-SLqndV<6o}
zBa_cKxk+*kB5~&-naof(v@v-wf=T?}T<sQmP<pr5S=}>Eda_@yWh4@g&m`YdPokv4
z=CkhW|7$?`Bun*$%Ltw({U(vvXc;ny;c?4d=Dc^^9)b6GzW-lov!5T#BUMc>6Yude
z<EeFlEjSMk^WEMfGj5le+Z!YZ?|+B<f$z7FyvX5pwi-g(t^9<Gs|d*FaXol<EwS^~
zKY?s`AxeFVf9d`oc>Jf&a&<1y$)~3a&Pv2$z{Up~b@S(6c8U3Di#XLk-MFJhvPXM4
zdd4~qdI8B*!tKRHoVVJ4!Fvr~I`D4Y!|yRmV#7D%*-hcjF6eOcinUxAOT?c%E3AQU
zQ0_fT+CB=n2HTZM|ABaWA3l}IeVh-S&c~BZ`kO*6%@~0F_OU;#J96{#NZ!52(uwy!
z-u#^+^9$>~M_?-gCf_Ih1qsb-jKI_2s3(_D5Oj>KXjeDV)V^FZ<9#eOm;X?U=n|AD
zq<(RAHk`KaFR?^c4$I5kdU~1p2f|yB$5u!=$f^DnR6dqD%=)Hq_isMNDYjxUuQ;FM
z@36P9IUG87IPm_io&j5NT6s(0Zzla`CTt=IF>~<WME!kcY(;wY=iI-;>R@G{X^~Cw
hclH0fg>o90?Hy*n8bj!a{r?F-QBF;^<cay~{{vwg+7|!-

diff --git a/docs_old/assets/images/logits_processing_diagram.svg b/docs_old/assets/images/logits_processing_diagram.svg
deleted file mode 100644
index 92668e6c6..000000000
--- a/docs_old/assets/images/logits_processing_diagram.svg
+++ /dev/null
@@ -1,157 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN"
- "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
-<!-- Generated by graphviz version 2.43.0 (0)
- -->
-<!-- Title: %3 Pages: 1 -->
-<svg width="214pt" height="784pt"
- viewBox="0.00 0.00 214.00 784.00" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink">
-<g id="graph0" class="graph" transform="scale(1 1) rotate(0) translate(4 780)">
-<title>%3</title>
-<polygon fill="white" stroke="transparent" points="-4,4 -4,-780 210,-780 210,4 -4,4"/>
-<!-- inputTokens -->
-<g id="node1" class="node">
-<title>inputTokens</title>
-<polygon fill="none" stroke="black" points="48,-751 48,-772 159,-772 159,-751 48,-751"/>
-<text text-anchor="start" x="51" y="-758.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Input Tokens</text>
-<polygon fill="none" stroke="black" points="48,-730 48,-751 95,-751 95,-730 48,-730"/>
-<text text-anchor="start" x="66.5" y="-736.8" font-family="Times,serif" font-size="14.00">7</text>
-<polygon fill="none" stroke="black" points="95,-730 95,-751 127,-751 127,-730 95,-730"/>
-<text text-anchor="start" x="106" y="-736.8" font-family="Times,serif" font-size="14.00">4</text>
-<polygon fill="none" stroke="black" points="127,-730 127,-751 159,-751 159,-730 127,-730"/>
-<text text-anchor="start" x="138" y="-736.8" font-family="Times,serif" font-size="14.00">8</text>
-</g>
-<!-- TransformerDecoder -->
-<g id="node2" class="node">
-<title>TransformerDecoder</title>
-<polygon fill="lightblue" stroke="lightblue" points="190,-690 16,-690 16,-654 190,-654 190,-690"/>
-<text text-anchor="middle" x="103" y="-668.9" font-family="Helvetica,sans-Serif" font-size="12.00">Transformer Decoder Pass</text>
-</g>
-<!-- inputTokens&#45;&gt;TransformerDecoder -->
-<g id="edge1" class="edge">
-<title>inputTokens&#45;&gt;TransformerDecoder</title>
-<path fill="none" stroke="black" d="M103,-725.64C103,-717.67 103,-708.77 103,-700.58"/>
-<polygon fill="black" stroke="black" points="106.5,-700.38 103,-690.38 99.5,-700.38 106.5,-700.38"/>
-</g>
-<!-- logitsTable -->
-<g id="node3" class="node">
-<title>logitsTable</title>
-<polygon fill="none" stroke="black" points="19,-593 19,-614 188,-614 188,-593 19,-593"/>
-<text text-anchor="start" x="22" y="-600.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Model Output Logits</text>
-<polygon fill="none" stroke="black" points="19,-572 19,-593 85,-593 85,-572 19,-572"/>
-<text text-anchor="start" x="30.5" y="-578.8" font-family="Times,serif" font-size="14.00">Token</text>
-<polygon fill="none" stroke="black" points="85,-572 85,-593 188,-593 188,-572 85,-572"/>
-<text text-anchor="start" x="97" y="-578.8" font-family="Times,serif" font-size="14.00">Probability</text>
-<polygon fill="none" stroke="black" points="19,-551 19,-572 85,-572 85,-551 19,-551"/>
-<text text-anchor="start" x="45.5" y="-557.8" font-family="Times,serif" font-size="14.00">+</text>
-<polygon fill="none" stroke="black" points="85,-551 85,-572 188,-572 188,-551 85,-551"/>
-<text text-anchor="start" x="125" y="-557.8" font-family="Times,serif" font-size="14.00">3%</text>
-<polygon fill="none" stroke="black" points="19,-530 19,-551 85,-551 85,-530 19,-530"/>
-<text text-anchor="start" x="41" y="-536.8" font-family="Times,serif" font-size="14.00">foo</text>
-<polygon fill="none" stroke="black" points="85,-530 85,-551 188,-551 188,-530 85,-530"/>
-<text text-anchor="start" x="125" y="-536.8" font-family="Times,serif" font-size="14.00">4%</text>
-<polygon fill="none" stroke="black" points="19,-509 19,-530 85,-530 85,-509 19,-509"/>
-<text text-anchor="start" x="49.5" y="-515.8" font-family="Times,serif" font-size="14.00">.</text>
-<polygon fill="none" stroke="black" points="85,-509 85,-530 188,-530 188,-509 85,-509"/>
-<text text-anchor="start" x="125" y="-515.8" font-family="Times,serif" font-size="14.00">7%</text>
-<polygon fill="none" stroke="black" points="19,-488 19,-509 85,-509 85,-488 19,-488"/>
-<text text-anchor="start" x="47" y="-494.8" font-family="Times,serif" font-size="14.00">1</text>
-<polygon fill="none" stroke="black" points="85,-488 85,-509 188,-509 188,-488 85,-488"/>
-<text text-anchor="start" x="120.5" y="-494.8" font-family="Times,serif" font-size="14.00">11%</text>
-<polygon fill="none" stroke="black" points="19,-467 19,-488 85,-488 85,-467 19,-467"/>
-<text text-anchor="start" x="47" y="-473.8" font-family="Times,serif" font-size="14.00">2</text>
-<polygon fill="none" stroke="black" points="85,-467 85,-488 188,-488 188,-467 85,-467"/>
-<text text-anchor="start" x="120.5" y="-473.8" font-family="Times,serif" font-size="14.00">13%</text>
-<polygon fill="none" stroke="black" points="19,-446 19,-467 85,-467 85,-446 19,-446"/>
-<text text-anchor="start" x="47" y="-452.8" font-family="Times,serif" font-size="14.00">3</text>
-<polygon fill="none" stroke="black" points="85,-446 85,-467 188,-467 188,-446 85,-446"/>
-<text text-anchor="start" x="120.5" y="-452.8" font-family="Times,serif" font-size="14.00">17%</text>
-</g>
-<!-- TransformerDecoder&#45;&gt;logitsTable -->
-<g id="edge2" class="edge">
-<title>TransformerDecoder&#45;&gt;logitsTable</title>
-<path fill="none" stroke="black" d="M103,-653.83C103,-646.79 103,-638.04 103,-628.47"/>
-<polygon fill="black" stroke="black" points="106.5,-628.25 103,-618.25 99.5,-628.25 106.5,-628.25"/>
-</g>
-<!-- OutlinesLogitsProcessor -->
-<g id="node4" class="node">
-<title>OutlinesLogitsProcessor</title>
-<polygon fill="lightblue" stroke="lightblue" points="206,-406 0,-406 0,-370 206,-370 206,-406"/>
-<text text-anchor="middle" x="103" y="-384.9" font-family="Helvetica,sans-Serif" font-size="12.00">Outlines Regex Logits Processor</text>
-</g>
-<!-- logitsTable&#45;&gt;OutlinesLogitsProcessor -->
-<g id="edge3" class="edge">
-<title>logitsTable&#45;&gt;OutlinesLogitsProcessor</title>
-<path fill="none" stroke="black" d="M103,-441.93C103,-432.71 103,-423.91 103,-416.22"/>
-<polygon fill="black" stroke="black" points="106.5,-416.17 103,-406.17 99.5,-416.17 106.5,-416.17"/>
-</g>
-<!-- logitsProcessorTable -->
-<g id="node5" class="node">
-<title>logitsProcessorTable</title>
-<polygon fill="none" stroke="black" points="30,-309 30,-330 177,-330 177,-309 30,-309"/>
-<text text-anchor="start" x="37" y="-316.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Processed Logits</text>
-<polygon fill="none" stroke="black" points="30,-288 30,-309 92,-309 92,-288 30,-288"/>
-<text text-anchor="start" x="39.5" y="-294.8" font-family="Times,serif" font-size="14.00">Token</text>
-<polygon fill="none" stroke="black" points="92,-288 92,-309 177,-309 177,-288 92,-288"/>
-<text text-anchor="start" x="95" y="-294.8" font-family="Times,serif" font-size="14.00">Probability</text>
-<polygon fill="none" stroke="black" points="30,-267 30,-288 92,-288 92,-267 30,-267"/>
-<text text-anchor="start" x="54.5" y="-273.8" font-family="Times,serif" font-size="14.00">+</text>
-<polygon fill="none" stroke="black" points="92,-267 92,-288 177,-288 177,-267 92,-267"/>
-<text text-anchor="start" x="123" y="-273.8" font-family="Times,serif" font-size="14.00">0%</text>
-<polygon fill="none" stroke="black" points="30,-246 30,-267 92,-267 92,-246 30,-246"/>
-<text text-anchor="start" x="41" y="-536.8" font-family="Times,serif" font-size="14.00">foo</text>
-<polygon fill="none" stroke="black" points="92,-246 92,-267 177,-267 177,-246 92,-246"/>
-<text text-anchor="start" x="123" y="-252.8" font-family="Times,serif" font-size="14.00">0%</text>
-<polygon fill="none" stroke="black" points="30,-225 30,-246 92,-246 92,-225 30,-225"/>
-<text text-anchor="start" x="58.5" y="-231.8" font-family="Times,serif" font-size="14.00">.</text>
-<polygon fill="none" stroke="black" points="92,-225 92,-246 177,-246 177,-225 92,-225"/>
-<text text-anchor="start" x="123" y="-231.8" font-family="Times,serif" font-size="14.00">7%</text>
-<polygon fill="none" stroke="black" points="30,-204 30,-225 92,-225 92,-204 30,-204"/>
-<text text-anchor="start" x="56" y="-210.8" font-family="Times,serif" font-size="14.00">1</text>
-<polygon fill="none" stroke="black" points="92,-204 92,-225 177,-225 177,-204 92,-204"/>
-<text text-anchor="start" x="118.5" y="-210.8" font-family="Times,serif" font-size="14.00">11%</text>
-<polygon fill="none" stroke="black" points="30,-183 30,-204 92,-204 92,-183 30,-183"/>
-<text text-anchor="start" x="56" y="-189.8" font-family="Times,serif" font-size="14.00">2</text>
-<polygon fill="none" stroke="black" points="92,-183 92,-204 177,-204 177,-183 92,-183"/>
-<text text-anchor="start" x="118.5" y="-189.8" font-family="Times,serif" font-size="14.00">13%</text>
-<polygon fill="none" stroke="black" points="30,-162 30,-183 92,-183 92,-162 30,-162"/>
-<text text-anchor="start" x="56" y="-168.8" font-family="Times,serif" font-size="14.00">3</text>
-<polygon fill="none" stroke="black" points="92,-162 92,-183 177,-183 177,-162 92,-162"/>
-<text text-anchor="start" x="118.5" y="-168.8" font-family="Times,serif" font-size="14.00">17%</text>
-</g>
-<!-- OutlinesLogitsProcessor&#45;&gt;logitsProcessorTable -->
-<g id="edge4" class="edge">
-<title>OutlinesLogitsProcessor&#45;&gt;logitsProcessorTable</title>
-<path fill="none" stroke="black" d="M103,-369.83C103,-362.79 103,-354.04 103,-344.47"/>
-<polygon fill="black" stroke="black" points="106.5,-344.25 103,-334.25 99.5,-344.25 106.5,-344.25"/>
-</g>
-<!-- sampler -->
-<g id="node6" class="node">
-<title>sampler</title>
-<polygon fill="lightblue" stroke="lightblue" points="136.5,-122 69.5,-122 69.5,-86 136.5,-86 136.5,-122"/>
-<text text-anchor="middle" x="103" y="-100.9" font-family="Helvetica,sans-Serif" font-size="12.00">Sampler</text>
-</g>
-<!-- logitsProcessorTable&#45;&gt;sampler -->
-<g id="edge5" class="edge">
-<title>logitsProcessorTable&#45;&gt;sampler</title>
-<path fill="none" stroke="black" d="M103,-157.93C103,-148.71 103,-139.91 103,-132.22"/>
-<polygon fill="black" stroke="black" points="106.5,-132.17 103,-122.17 99.5,-132.17 106.5,-132.17"/>
-</g>
-<!-- sampledTokenTable -->
-<g id="node7" class="node">
-<title>sampledTokenTable</title>
-<polygon fill="none" stroke="black" points="40,-25 40,-46 166,-46 166,-25 40,-25"/>
-<text text-anchor="start" x="43" y="-32.8" font-family="Times,serif" font-weight="bold" font-size="14.00">Sampled Token</text>
-<polygon fill="none" stroke="black" points="40,-4 40,-25 119,-25 119,-4 40,-4"/>
-<text text-anchor="start" x="58" y="-10.8" font-family="Times,serif" font-size="14.00">Token</text>
-<polygon fill="none" stroke="black" points="119,-4 119,-25 166,-25 166,-4 119,-4"/>
-<text text-anchor="start" x="137.5" y="-10.8" font-family="Times,serif" font-size="14.00">3</text>
-</g>
-<!-- sampler&#45;&gt;sampledTokenTable -->
-<g id="edge6" class="edge">
-<title>sampler&#45;&gt;sampledTokenTable</title>
-<path fill="none" stroke="black" d="M103,-85.91C103,-78.33 103,-69.13 103,-60.23"/>
-<polygon fill="black" stroke="black" points="106.5,-60.05 103,-50.05 99.5,-60.05 106.5,-60.05"/>
-</g>
-</g>
-</svg>
diff --git a/docs_old/assets/images/logo-dark-mode.png b/docs_old/assets/images/logo-dark-mode.png
deleted file mode 100644
index 24c5f3bc36cd271902376dcaebc7c667f9e909a4..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 7425
zcmZvhcRX8v+sCP*cJ0ltTB}y6J^M3CQLFZficwNxC#0xq)d;G#7FD}NY&ApeQKQsM
z#HiX5lpw;B-oN|xyzb}q+<)XG=OpKxoa_6&KJU+UedEkb^ciS*XvxUP7#<ktT9A>E
z|0aEoy+%daGXKr}lJrL7XJ8jdM#irA=S7}pUXw~X$Q`6-8}tb55d?W0;7$gCK%~8V
zL4m->e(uuX0MB=O>O5p*d`S;<|F(p_-$ldT^I7HM_s7cfxLFLYM7j8Cpo~kXuNesF
zz7eZV$V{iyv#xVfH9t0gc@(2$4aG(2`#Et?vgKRZG7EW#$kJUI)=M#Bvz^s@Gwdp-
zu`HCG_tKB<&70wopj=|%r^z2bnzpj{HVY2Jk<LWJ8H7o0pohp^WYwcm6K;@H;SbCY
z|6<u+5^F<W`ZR)&?jReiy2#wVRCIF@N_Kkr%Q2ux;9C75q%UiGNPr??zV(qr?ZT$p
z%B`iO13;99=pq_o6#^ozj8axLFlA`HOjyw<Pk=PmxsCFnX6Id)`-3?~>^puH9QY)}
z#=`R^HyC4MV+S_G3rM^2<ibJB%qb!+x1DrMWi(IjR1hMEtzA*CC4Uv?E@Tla*IG}D
zcbr@n>FdwM-i+=l!YH(8t=eg!!jNmpz>&LiJ~Oucw$?9`PE9{(jkhvT9aOz`*cp31
z7&Xq?Il7Hu42cGvAkO#RNkl=e3PoNC8Kt??*XHdnIG`0%*fm055A%dUFc!MHy51G8
z+JJoHXPvKFhdFAG1y?k}F$<w1ADa)I`cHwc8ZT8Ab5&&^Bkx0I;wnG;jWq1<O=YV!
zapc&xGyuPRm4)=RK8ob0FG`KejPfdGyn~zF*_&$B=~t)Yvn<K=Tt%0wblr=rJ(P_m
z0-|yd@eq98qXMLmkdW2no7UJ^16DI_PG)AWE0j8W+nnEW^Zi{3l(cPh-A#u~brA01
zq+Ex-=9#Z_QAAdL5NEm%*LoMc0KCrZHt#wuDGh;x-s?w|)-Kf8_66S^eQQ<&e16<{
ze3Q}fjSUm(+79=%H76bFy2Y>=Y#=ib(I<EF`j*1<uQRMN!gd(B9+`L?u#_7$zRMBl
z^I^2>q;?9C5MD|8Z$F;UP&@XYe(L)5wn$w4$pB;IS!dq5qDK?)m3su2*g?-uUdQPg
zj|oJ0^sb*#6}M;iPTo{SZNH?+`?zoXs~+#Wpb+mSm+UAv*XCD~6a{$8_Y}|DJ668R
zEtgYetkuEJLZl_@sz7@MN&b1#7)DXvoxvzoXX+Ab<@>t2MM4bC|1n2Cyp$Fa`hL~^
z``~Dkc1`2>aeHMAjUXM>ic--*(wsW#_*=M0Pe(d>kpp)3QTcLWswhAJqgWq!BfDT1
z(VU(X$28Is0VIGl=6oP45|m04O+6ydLiGD0IxL+2j(*qZ!{9rX^DYd!TKC%YHpDp0
zb=yU~x*c=iX$^L?xX5}8A!O>%?d`B%T&O7qo>k;EH0kXUpcDB!=UH7<Pb~@40TD3M
zt@!&5GD6><NVz~{06lIc7@V8(=DLd8+GKR6ivojS=(j?hPBVr6YVRMoAhFgraLi0C
zv2sIb+XP|O|J$$A-eEKAqBpMVt4ho%CUVA~R^)aEKNpHC3-`LK2859X4`biO8dN?R
za3aNrM(jUx=)-oBMMG3Czo!F8GfPmxos;sD+Kkg%6Op9R1n#^f86&5u4+0A5LA9&$
zcLv4W9b!;*o)wOS0IB(1tl|53ojEJfI=RYdPkuk*A;RW%(~Z!9i*AuXyEIk)8OOV=
zrT7jpcRikmG)|v9DpgpND<_Bpz!7y9XXlZeaGuHRU%)%LiYsH!U=?fI_a3se0F`^}
zIqr*tVnib-%d*CuMC&sB3x9akKqw414jMn)fc(k(e~)R}xSp<V&hx4QBRh)09u_$I
zCRbfnbb|UzL>fqfvwR+NeN#Mmuj0?cy)FdPH#)7$#M~14E}u4DEJ>H?dgxhglQI$q
zDIKlw&oqDp^F_-tv$SM`q_k-CMH^I+MMLQfdb)jT|7|)2F*uRgw);6Rq@f$JoJqW{
zH7`7;INTdwd~-1FxmU0A*`13_<n_l#QR3!|;NV97EN3}U@I*E`#2D(2NjC+Tn=`rd
z@e0&N*JOp>FtppQZeAoZm$M58k-HcU`-%GQ4Em`D+;2Rtf|`08TPG(#@I))+3};j*
z_jDfb+6jDHog-=Y39s=L4TX(7cpw3>;XuJRz7Y{SW$(mtcBU6Nf43%sZp$3<L(;Sn
zg69|sV&vJ_a(t4gAq|No29neZAMO5qut%@6BWgUCBucpryA>pT9hlmrM;qUr&a=vI
z`3#h8`g^(3D3lMTo-yXB#PExGiBMpXe}z%VX2a?p4QVd*Q9g`k{i8pI8>GB{_(57{
zuWGV|@?Gc$fnb|4O{emgbYt8j-E%&`r+~=X)h}i>qV%JEjtv!`odhu<7x{_3zO+=X
zI?-VHPj|_5_LkP9q*#3i=opy<-Qmqu8M{TCg3q|7odk+2WRKhChR=Kv?u5|r2O_>#
zSkl$LKH7ILuL{sR;}pG^w(W2#G5D!J(Wt!FzUk&#W}%Yc^kWf?$D40grbkUO?N!U-
zZ6*vC=vL9;DMKZ#EO_8-YruoUZ72l1cy%<BbM#$(Vep1l;ZeE8(fOSY$)m}fDXZcj
zoAqU`)&R-*0Ql)hg?sy@EQiml;?!IW1<Vl5E2l09$T%~YS_yuzvA_2N6y>MQ*EFA5
zEKMyv=8FKJnmaqAyW|vCN8=D1;UObxN&?)|O_=JI73$)eAeXT~-l)Vb+pxi|Nk8<w
zV%p$N5|A;`Vyb1-xxuHc)A2(9KoV?qd&Q;H-gu3PdV<fg`5M>AJWqp7t~Er+_a&D1
zxdJmw{g~gKJCy-7UsK`Vo+dO*uvD6+Z#9LcLi#g;MP0p*FUu<63aNb`b6Gr?*Sdc{
zv2*w#))IsISLU#^{0BmUn`0I)%XNNnD+NL{wyH9Mn?JXU2H$U%qexIg0-ajx;2@}+
z2kN*@JLzyZ$j0?D_vkD8u12|pH_D0GD}dt{9un-x4d0!f{_X*)H>;%+J%R-Pbkc{w
z?UY*8g5Ar3_l0te-(US{IvX(Z+rH5}33xZ(RFqY8owAbX)K-cuVR;QVP|g0rHhYsT
z0wT2__HE~#iAV%nQlBIyZ!R@Vm=9s3oy@?{(2{hwkTk!(>1I#4mbzx;q61u9j%yn>
zUE+e|WGMpN;g837jxBUU`1$*?l3YGm5At<(UV+=avC&eeosusAD)S6;g6uunD`W+w
zuJQC!dgqKsjV)_Pd}Ka;@fEvYy?sF@X0X8%@Gr84DJR6@`dk)*S!b;eF3uVoGZ2rP
z?^w9^sXX)Zt6ouMU@i4UKoKpmx;FVY@YZT?{7>(b!k4p_ZKwxd^8j5*eiUQ}u#Gk9
z+TjcF|8RqjCw8j)hD5%=RVt%hK?9KxU4C|f(dyF=>0^pP1vgTh%t;CSn>Q~su2WlT
z6<x76rO~OeHYD%8Cd@o1Oz$v)_^dJDl1yD<DeZ;lvz^+XS>E8yZfqqo1-c)}JZRH`
zYEvxPs(y?ZsPjt}Ka8rmq60l}+6-ab0esJf_@D=?-9t>V;ps_End;9z`BPy4Mf=kS
zJ~;~?e^&|t*SUtf%;QyM*LyC6b3^HvcdN58UL<N1?Kl3B5tD;Yf8`phGxC`?7iEL^
zX)|sts@1LXcMh{Bhcxr;n0MJK9G=X(W*0?%v}yQ3x8Cs3i+`TExS}iXh0)^Mis{*g
zd-`!YA@SKhG~238)(gm#b5#5TQ8gsZ@pmzIDMujcJpZ2LDqq0^|37;2FOB(MUODM-
zGIKj{e`W2VchI*^pQ8$K5Sukzy=ZiB`SRBnG#tA5-sh3%thhr=JbGpJh0^0Haa+)9
z_>J>A_)~nN@qLnDxzTff@ehG){kETC$L?t;@@-=cpUrTMBJ_wqYP0$mT4Bi*?z#HT
zha1<&Ig(0K<J6v~de2$kJT;0-_Ez!#ZzT9PghYDdVhtX0?VzIER+@lQsB?SkMh0us
z`d>YMC>|Kyo#rY3O}CFLwY4c+8Ob4W`z^PAb7t_#)%6Bz0g_w#8HPWJZBrz~?r4iR
zoVXOK$+84yRED#PoxIC45ux?%raMi@3F>T!<nm!BX-#Ff&RvqhvewuVP~vlpKgW@>
zndk4Be{lFuQkM<CA-qvo0KWO8k6w+bJRkU*nK=|zmnD%{(`qB>`K^EoW>_2`%pEw=
zxuRRdei36&9U0JhG=>yNnh`F*B0qu|#_f_@NEhxigh8;v=0haad*JeuhCit;WA!5&
z+y`Vw!*(m^-yuQ2Cb_JJ14|Jg=wK<^URJiQa3Y*b;tv)qBo>H9Ro}@BH!KvQ=a7-+
z>#c_`6zj^vd*?lQr-Z3E#u8hzXW=d*vc6w)N<tlLW8t0|Hi^qM^On6gI(yzyy4Hje
z6I4s7bau0wE{a}CTh&cFyYNX5FsUk0lq!_yyTpahDIcTGv)(qx!V{sUSnzYq)W@nI
zxqUW2+j1fm%m_+_It;<zlovPo|Muq}dr10<EEb&Ad^SetP`}Fu_TmlW8-}~~|CD_%
z|HinDi{36eQO1aUO2vM&D#@xZD_Upyfq627Z&Kw;;aXzbk^q>!qD9`j&pF*h<U)n5
zon-sZTorzthc35qOI7OhdW>}df9A55C!8)LDadquqH}vQZYK)0SBHP}LK9C@LI!y9
zNH>rwv-EX&u;TOurSLuE4oV(ow?%hTvx6#aytj^I{m9L530Wp0s91#^kH=xoJ<}fm
zy$>axe5UR8T`;1qc;@{`O3m`C<z+gFZS#iqB8<IU9@dqr!WR0Ky%9<di}S#q!mgU~
zs==q+r&zV=4({SCwWs!pK>dlNIZ#jkv;j$awq701A7|ZBZxskm=2UD1P8w+W$H)HT
zE1YZiU$=0tCm)r;P<{*2RMOr-N~9v_Ug`>IoAZKX#3+xAY(;-dy5R3}1!|`iP7Wry
zD`4Ns?#BwlYJLfJcS&{cIo$MZkvfL+*Ba46W<N)1+_rX6DM3oq)PGWBV9)PW&#PD7
zOW$<U@5R<M{7U`<n7)sua)#56*No{XO}laX2K`NpPPrx_^hrft-9@&)zaL9iq~ewO
z`;3b~8zPejd>&tBvB$|#gY3!u0uHT6NJ%@&8Fbpx)y4+zb@kaM*}n=X1cUfm6C~%a
z?x$VPV_e&)@hN7G3%gmlNgY3@1vtEIX68vkQM?!pKcjpM1!%6+f5TEpvK2%RBY7P#
z#YJsN==TKHz#3*&)n{@;mvOG2-Z^X+-9J+k1f2c`v0b7&-d>|Ye<gQe<a)4R6^M3N
z)yvB@&wS}y^7LW$w^l)iKS{3>S=joQkkDJJ%$J&nIZ>6J-A82(q)^g&S-k&4=5d9f
z@|=d_8*neWK6nE`FIUWliAo@F*)&uZGy}II#H>|)15}Q75}rqUn-Q!N+H3xzExqpi
zrNkK?K4c%LQ$}o$O6@NS=T6S6zJ8%4%b@Hf$<yS=RINF6aWnXd_p4Xp1tA<{*a{to
z$6)&}!hBCbb?*|`t+g{hHQ;|cF*ypti&ObanL|H1?%DyZ8)dfJE6jH}9S$bm?R$H_
z8Q<OBMVFg$PUl18Weqp#GLO)VC;@mSKpxVut!%i_T1Kh~zjb~ic^vcp31tss7FA^o
z80&oZpxFLqJfyUPE$K*<RZ9u@c6_(*c(3CHhG-%G^IzrKzg6`gE&CG&24Lgkx$&}J
zm7pg1PrKhoM7W6f9#`7D7wa%_3t~*B{=#clxyni%kyP&e8BZyMwKPx(Xk4TylGSCL
z7QnsDsG^**RX@A&nbw{l)JLx;c4GfiRll?Io$h(0(ui+Lho`y9?BAJBv=9?=Ay9&C
z{iAxH9g)7JZTYN3t8^@OtE~EZ@~KZ-E55sMGYv}eoE^TB7#W8FzZ|VO(vaPf`)_dA
zk`tG1|I^XP0BxJ&p^yDDNfedx3DP2+@xg=!&8?W~)GzP{0Cd0+EH9dxRNYAap9`6l
zzH;A@u#K?KZ|SG*Z%I~cVIq;yjs$_a46R9~XH^2hq+-|L#=lRhT$BE`^~5FT@<PSs
zAq9b?z7WHWw+htITh+vfB~ykiBPmDP_e)45-^u;e79n!MD~r3xNRAVC(r>?8ri}|r
zP#H7$7`fB=p=AizrWNH#n}T!X8(x(YaX(XZSO%EJulDi+8HI(Lw>~Z<1N_EZQV9p8
z<d|!7awi&lM8fSh!0j}ah8u#ED;ESSxRZ81lN+I#DbZ*{$U>Ba1G$W8^2py9gpZrP
zZcZX;eBs5|QyJ!RS#+bwpE^A#MtZHDEaCaBJDgcppLVwH!5B_;N)Gou?syXEp~c2%
z!fCK4k#}*|-5kwW&k={AIj@cn3&T~3XCCk`MZGB7zL4nSdnr3fF`-ed2ziLg3dHTP
zofs`$o=k*Otc{UtE+ZqaeHo9u40MUPywHaNK*AXEqyr3w_M8h#I0R`aDw&07p3dxP
zU7^=|a&z1A!{?C;g-$);<8bnmFP9h00T_}E5f%;-cjv*umS)Y)`8!*m%aCJ)l%Ad4
z#wGvs=m(cRb0uPaUTh#K&u_UROUPLFHPzXSHe|>rjdtjAuI`Y(kHssbAAs)-A(S?_
zDl{b75+3iysWrjph9i|G!J-wDk>pPqdXlM+@fZyCc^4SNQ4YLPEsAq)V`3x=zd)MA
zh2R-`3zMc)A{-8*Of$ss*KrhW2rvC70M8G*-kQ3x29!JjnHE)|nIiy70g5vU!ou_6
z=ZD1I;7%|FA=9!<E9<)N*LODrdeb`-(oAL(h(x;EQ-KA5Dn~hL>~w9g-U`#>5j3>v
zxjO@J1g!?B9!v(4?HvQUt~QBWk5otOp?YeJh!G~YCy<lp#7G+Y)o<^|M1qveF6p9x
z&U8K4<6%A71+C++v};L6!{Mr~?Lrn5!78M7-TS5{FJ9GR5?**?L_IErj@$}c&GiZ7
z4$tdwd+ehp`>VLO#J?|&d!cw%tt2J-OJ=QzTYg-pG&0+Fmfmz8f4@fbqBg-Nf7_zi
z6SKfm4^7p{_)#@4dp(z5%!ZDu?fc47fTxxnmmbs#+VibHC39?|1GW$Y9Atm#u(gau
zM&S#@U<>nBoj>%3Aj?uafsucOFjn)NSp5|u?WgNPkP7hJT*@W)Bm`38tS*VGKLG-9
zD{Kdm9%M=@!h3$0Izi1v>?MqU@ihZ;d11?~Akju(<cz8jDZ!NkWETZbJf<m89P#Mn
z%scs4MOKM|XCltE?flrMAzH|%EXoLjb~*|ypp$nV;l{Wadymv{oYAfsGuJE_le8GC
z)7v`bYJR!6A{Z!jT6p~^?JHr)pK384lDQNP9TW-DxtQ6fIIubU+<-`N2$k7pSt>W%
z>Hf)b=6ZQVM-t5~6P-PaAtIZ|Va56&QZwTAVgrhPaSk@k5|2>E$8{2lbp;O?RB2CM
z3+=zdRqfvoZ`%cs=S#5w)@A}{uAN!a$t*uiqBBufX9V}GBE^l1@Z`hWv-x$njHe78
z@o4=B+Qsv0%-ledn;WWRKc>#yK_#NDA1nME?ej>{x=O!z-9JPqt5*%i8&^7jlzxA7
zdoi1k5qnfYl_$;;If+<5Vm_)0xex$%d{t~?P6>_Tgl5vt$11W$$e-bUnO_AEmYl+9
zIu@@e&NRyiS?D^&peliq-!yxE?X{JlDsF3Jxic(YMWD}jLoP84P@QhU!=}uohiyZ&
zCNTt!MZc!BjD3K_KFcox1;Lt(Y4Xe53WV&nQkf4kO}UYFJ{pa`nhxfDyI=Xe;t&oH
zfdPVXUD7jej5bP_Ney)k5S_^HC0tt?glZ8lpEaHe-wEh7*nO~9MTPry^JO?Ec7#|V
zyO|QZQ$!dqPD9^Jo4bkAR2~WATns*@Aj*|4J27d#MO_hf7|5@YT0?95);^xjg%<K%
z%HI#qdZ9#K`(AJreS_4$lItV2kPNdO9$%9G6eTi5;iVVhxQZQ(&91t`bY0|VyTm&<
z1abZvC$dQy&ZRt$!C0?H{3c_fY8>k_LM_kqsp55X4b8MOEi@e;UzSJVxoN@KXWY%8
z*2pMG2zMS@lLvcKsag67)#6i}#?9$Nfp`G*Vn4zYBSh%?z6N~|Ahf7~VY=$P#Lx{c
zIv4}hd%V@s#<46~v@g%V_V2suy1_|>pP>pj7LMB&A{<}sV4l6;$K5X*x!?A9#dIay
z@0@&*n-5wyB<L2j*Q+*sxnOo)2%W(dwlITd&t96HheL_rG&?AMdyK9}ppVe}93mKo
zI}1Hl**4D{ugqM7N({vfa6KYDf<2C9UVjj9&r`%%h~Z*F`bc>MG{8_7seb)ZrFTYH
zrj%2e6|6FWKdmVo5S*truq-RZAl6xbj(W4|ZPDZ<GOnUEryUCR;=cG+3G|<e4vnn<
z;1De(Xwv}kva<uSA{F@VNbdkkXoLd#Vds%r3HXBBhT2+L=Lr1MHbf2oOuO><CE{(X
z50<u#om}RqNh?9=fThtUB`X$AT5Aw@S97+2(SxCQ?zZI}rX?w%)P}V02?*3{$Yp^3
zU%##>!9Qs6;@*zK28Dl9M2%V6jfo)h7z1d+@4pI@V-q=dwbwavG(COU@f?+N*py<l
z0Wnj4y-Y=m#ddID$;CX+7E*EdEi)c{;c*_XFw;Ul&m6~hW!=59Pd6a^djY2z<i>gy
z#}<i8h?S2exoq)#aw?p4m>M}2zT<X~8mhGBdDnr~vle5t|2kLqNTm&iF+#(_`4>md
z2u3?&s-$4`(yq1OL`v8>6Dm9XL%OptmSv*be+|hzr1KMNetF@J%rE8u*)3y{bXd>e
ztIZ1JFGi1s(_uwE!bGSJaZMm}{(<kHBQv*-lr0SC+>=!}pwYryk*oP;X9Q-s()cht
zGCy#A9tUI=c5|}nrv~=XF#a7Q3Sa>GY^YXpmqCz*v&I(`h8GkF<oYrit3I5}wCohv
z5&zw&o9b1*e@BpI`-oJ1z|Gzn)f1Vvx@p3I#zFmT4q-0L_r|5bS@pUY(l3FI-gTGA
z#hTXukPKy9PE<s3#b4gQT_^^m?uSKn#qYunxoSNpli9^HiDBSkzppMKDMb%$-#L(v
zXZZ-vJL)WVKdn_<Nj#^-Q*%BJgQIRoo2HfN=-%(NeHSTyH`f?-@0C7)ajt-y`dGbm
z;n8gaQC3rn2;oYp{XNUgI^@2)++z3uZ@Djfr$iLu1dHixr2^bGC$&wlZSu6}He_$j
zY-qyIsq-cE1|4;aq)COwwB{=%LMYEE>qg@;!(>9N0l!#*Q6+y;(;js8l+D?clR1Tl
z+guy0P)Xc$D_#hH<#b5#2x=Rl;YZ^aOmQY8R6ES&^>re3swfS-DrKX}k99sivP8zh
zV`DA7Y}BLE&K6z>v<V5xmDE?0jv26uDeFKS*S%+EF&P>|vyok`qIN^Fj&RU)#=H-|
zgKrlL$zcxAa%9AjdT~#CU{rN!%U?!c*bUF9Pm#LEu?C0#cU|db&YG`erd@ZDJ)~sr
tDX<JV5q_OH#i^S#3rV(f;N(E7$@ulK__9wEY59`ufu4zOmA1=^{{gxFrcnR@

diff --git a/docs_old/assets/images/logo-dark-mode.svg b/docs_old/assets/images/logo-dark-mode.svg
deleted file mode 100644
index 004bab98d..000000000
--- a/docs_old/assets/images/logo-dark-mode.svg
+++ /dev/null
@@ -1,207 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<svg
-   width="348"
-   height="95"
-   viewBox="0 0 348 95"
-   fill="none"
-   version="1.1"
-   id="svg39"
-   sodipodi:docname="logo-dark-mode.svg"
-   inkscape:export-filename="../projects/outlines/docs/assets/images/logo-light.png"
-   inkscape:export-xdpi="96"
-   inkscape:export-ydpi="96"
-   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
-   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
-   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
-   xmlns="http://www.w3.org/2000/svg"
-   xmlns:svg="http://www.w3.org/2000/svg">
-  <sodipodi:namedview
-     id="namedview39"
-     pagecolor="#ffffff"
-     bordercolor="#000000"
-     borderopacity="0.25"
-     inkscape:showpageshadow="2"
-     inkscape:pageopacity="0.0"
-     inkscape:pagecheckerboard="0"
-     inkscape:deskcolor="#d1d1d1"
-     inkscape:zoom="0.27368421"
-     inkscape:cx="283.17308"
-     inkscape:cy="45.673077"
-     inkscape:window-width="1416"
-     inkscape:window-height="818"
-     inkscape:window-x="30"
-     inkscape:window-y="38"
-     inkscape:window-maximized="0"
-     inkscape:current-layer="svg39" />
-  <g
-     clip-path="url(#clip0_56_2)"
-     id="g39">
-    <path
-       d="M101.92 79.3501C101.92 83.0601 99.78 85.3201 96.88 85.3201C95.13 85.3201 94.03 84.4601 93.43 83.6801H93.39V84.9801H91.07V69.1801H93.39V75.0601H93.43C94.03 74.2001 95.13 73.3801 96.88 73.3801C99.77 73.3801 101.92 75.6301 101.92 79.3501ZM99.6 79.3501C99.6 77.1401 98.52 75.2801 96.48 75.2801C94.58 75.2801 93.41 76.8301 93.41 79.3501C93.41 81.8701 94.58 83.4201 96.48 83.4201C98.51 83.4201 99.6 81.5601 99.6 79.3501Z"
-       fill="white"
-       id="path1" />
-    <path
-       d="M108.52 85.69C107.88 87.44 107.15 88.83 104.98 88.83C104.32 88.83 103.76 88.72 103.32 88.61V86.84H103.52C104.49 86.97 105.22 86.91 105.73 86.29C106.15 85.78 106.46 84.9 106.04 83.84L101.95 73.7H104.45L107.61 82.34H107.65L110.55 73.7H113.02L108.51 85.7L108.52 85.69Z"
-       fill="white"
-       id="path2" />
-    <path
-       d="M121.12 84.98V83.84H119.98V82.7H118.84V75.82H119.98V74.68H121.12V73.54H125.7V74.68H126.85V68.95H129.14V84.99H126.85V83.85H125.7V84.99H121.12V84.98ZM125.7 82.7V81.56H126.85V76.97H125.7V75.83H122.26V76.97H121.12V81.56H122.26V82.7H125.7Z"
-       fill="white"
-       id="path3" />
-    <path
-       d="M133.7 84.98V83.84H132.56V82.7H131.42V75.82H132.56V74.68H133.7V73.54H139.43V74.68H140.57V75.82H141.71V82.7H140.57V83.84H139.43V84.98H133.7ZM138.29 82.7V81.56H139.44V76.97H138.29V75.83H134.85V76.97H133.71V81.56H134.85V82.7H138.29Z"
-       fill="white"
-       id="path4" />
-    <path
-       d="M146.36 84.98V83.84H145.22V75.25H143.5V73.53H145.22V70.09H147.51V73.53H149.8V75.25H147.51V82.7H149.8V84.99H146.36V84.98Z"
-       fill="#FFFDFD"
-       id="path5" />
-    <path
-       d="M154.34 84.98V83.84H153.2V75.25H151.48V73.53H153.2V70.09H155.49V73.53H157.78V75.25H155.49V82.7H157.78V84.99H154.34V84.98Z"
-       fill="white"
-       id="path6" />
-    <path
-       d="M159.47 84.98V83.26H160.61V82.12H161.75V80.97H162.89V79.83H164.04V78.69H162.89V77.55H161.75V76.4H160.61V75.26H159.47V73.54H161.76V74.68H162.9V75.82H164.05V76.96H166.34V75.82H167.49V74.68H168.63V73.54H170.92V75.26H169.78V76.4H168.64V77.55H167.5V78.69H166.35V79.83H167.5V80.97H168.64V82.12H169.78V83.26H170.92V84.98H168.63V83.84H167.49V82.7H166.34V81.56H164.05V82.7H162.9V83.84H161.76V84.98H159.47Z"
-       fill="white"
-       id="path7" />
-    <path
-       d="M175.45 84.98V83.84H174.31V75.25H172.59V73.53H174.31V70.09H176.6V73.53H178.89V75.25H176.6V82.7H178.89V84.99H175.45V84.98Z"
-       fill="white"
-       id="path8" />
-    <path
-       d="M180.83 84.9799V82.6899H183.12V84.9799H180.83Z"
-       fill="#FFFEFE"
-       id="path9" />
-    <path
-       d="M187.59 84.9801V83.8401H186.45V82.7001H185.31V80.4101H186.45V79.2701H188.74V78.1301H193.33V76.9801H192.18V75.8401H188.74V76.9801H185.31V75.8401H186.45V74.7001H187.59V73.5601H193.32V74.7001H194.46V75.8401H195.6V85.0001H193.31V83.8601H192.16V85.0001H187.58L187.59 84.9801ZM192.17 82.7001V81.5601H193.32V79.2701H192.17V80.4101H188.16V82.7001H192.17Z"
-       fill="white"
-       id="path10" />
-    <path
-       d="M197.96 71.81V69.52H200.25V71.81H197.96ZM197.96 84.98V73.53H200.25V84.98H197.96Z"
-       fill="white"
-       id="path11" />
-    <path
-       d="M15.18 39.46V42.5H54.64V39.46H15.18Z"
-       fill="#E35A26"
-       id="path12" />
-    <path
-       d="M39.46 30.3501V27.3101H15.18V30.3501H39.46Z"
-       fill="#E35A26"
-       id="path13" />
-    <path
-       d="M48.57 27.3199H45.53V30.3599H48.57V27.3199Z"
-       fill="#E35A26"
-       id="path14" />
-    <path
-       d="M51.6 24.28H48.56V27.32H51.6V24.28Z"
-       fill="#E35A26"
-       id="path15" />
-    <path
-       d="M54.64 27.3199H51.6V30.3599H54.64V27.3199Z"
-       fill="#E35A26"
-       id="path16" />
-    <path
-       d="M48.57 21.25H45.53V24.29H48.57V21.25Z"
-       fill="#E35A26"
-       id="path17" />
-    <path
-       d="M54.64 21.25H51.6V24.29H54.64V21.25Z"
-       fill="#E35A26"
-       id="path18" />
-    <path
-       d="M15.18 51.6V54.64H54.64V51.6H15.18Z"
-       fill="#E35A26"
-       id="path19" />
-    <path
-       d="M39.46 24.28V21.24H15.18V24.28H39.46Z"
-       fill="#E35A26"
-       id="path20" />
-    <path
-       d="M15.18 63.74V66.78H54.64V63.74H15.18Z"
-       fill="#E35A26"
-       id="path21" />
-    <path
-       d="M15.18 75.88V78.92H54.64V75.88H15.18Z"
-       fill="#E35A26"
-       id="path22" />
-    <path
-       d="M15.18 6.07H21.25V3.03H18.21V0H12.14V3.04H15.18V6.08V6.07Z"
-       fill="#E35A26"
-       id="path23" />
-    <path
-       d="M24.28 6.07H30.35V3.03H27.31V0H21.24V3.04H24.28V6.08V6.07Z"
-       fill="#E35A26"
-       id="path24" />
-    <path
-       d="M33.39 6.07H36.43V9.11H45.54V6.07H48.58V3.03H45.54V0H39.47V3.04H42.51V6.08H39.47V3.04H36.43V0H30.36V3.04H33.4V6.08L33.39 6.07Z"
-       fill="#E35A26"
-       id="path25" />
-    <path
-       d="M51.6 6.07H48.56V9.11H54.63V6.07H57.67V3.03H54.63V0H48.56V3.04H51.6V6.08V6.07Z"
-       fill="#E35A26"
-       id="path26" />
-    <path
-       d="M60.71 0H57.67V3.04H60.71V0Z"
-       fill="#E35A26"
-       id="path27" />
-    <path
-       d="M15.18 6.06995H12.14V9.10995H15.18V6.06995Z"
-       fill="#E35A26"
-       id="path28" />
-    <path
-       d="M24.29 6.06995H21.25V9.10995H24.29V6.06995Z"
-       fill="#E35A26"
-       id="path29" />
-    <path
-       d="M33.39 6.06995H30.35V9.10995H33.39V6.06995Z"
-       fill="#E35A26"
-       id="path30" />
-    <path
-       d="M66.77 9.11003V6.07003H63.73V3.03003H60.69V6.07003H57.65V9.11003H63.72V84.99H60.68V88.03H9.11V84.99H6.07V9.11003H9.11V6.07003H12.15V3.03003H6.08V6.07003H3.04V9.11003H0V84.99H3.04V91.06H9.11V94.1H60.71V91.06H66.78V84.99H69.82V9.11003H66.78H66.77Z"
-       fill="#E35A26"
-       id="path31" />
-    <path
-       d="M102.09 60.7V57.02H98.41V53.34H94.73V49.66H91.05V20.16H94.73V16.48H98.41V12.8H102.09V9.09998H124.21V12.8H127.89V16.48H131.59V20.16H135.27V49.66H131.59V53.34H127.89V57.02H124.21V60.7H102.09ZM120.53 53.34V49.66H124.21V45.96H127.89V23.87H124.21V20.17H120.53V16.49H105.79V20.17H102.09V23.87H98.41V45.96H102.09V49.66H105.79V53.34H120.53Z"
-       fill="#E35A26"
-       id="path32" />
-    <path
-       d="M149.89 60.7V57.02H146.21V53.34H142.53V23.87H149.89V49.67H153.57V53.35H162.79V49.67H166.47V23.87H173.85V60.7H166.47V57.02H162.79V60.7H149.89Z"
-       fill="#E35A26"
-       id="path33" />
-    <path
-       d="M189.18 60.7001V57.0201H185.5V29.38H179.98V23.86H185.5V12.8H192.88V23.86H200.24V29.38H192.88V53.34H200.24V60.7001H189.18Z"
-       fill="#E35A26"
-       id="path34" />
-    <path
-       d="M206.38 60.7001V12.8101H213.74V60.7001H206.38Z"
-       fill="#E35A26"
-       id="path35" />
-    <path
-       d="M221.2 20.1701V12.8101H228.56V20.1701H221.2ZM221.2 60.7001V23.8701H228.56V60.7001H221.2Z"
-       fill="#E35A26"
-       id="path36" />
-    <path
-       d="M236 60.7V23.87H243.36V27.55H247.04V23.87H259.94V27.55H263.64V31.23H267.32V60.7H259.94V34.9H256.26V31.22H247.04V34.9H243.36V60.7H236Z"
-       fill="#E35A26"
-       id="path37" />
-    <path
-       d="M281.73 60.7V57.02H278.05V53.34H274.37V31.22H278.05V27.54H281.73V23.86H300.17V27.54H303.85V31.22H307.53V44.12H281.73V49.66H285.41V53.34H296.47V49.66H307.53V53.34H303.85V57.02H300.17V60.7H281.73ZM300.18 38.61V34.91H296.48V31.23H285.42V34.91H281.74V38.61H300.18Z"
-       fill="#E35A26"
-       id="path38" />
-    <path
-       d="M321.81 60.7V57.02H318.13V53.34H314.45V49.66H325.48V53.34H338.38V45.96H325.48V42.28H318.12V38.6H314.44V31.22H318.12V27.54H321.8V23.86H340.24V27.54H343.92V31.22H347.6V34.9H336.54V31.22H323.64V38.6H336.54V42.28H343.92V45.96H347.6V53.34H343.92V57.02H340.24V60.7H321.8H321.81Z"
-       fill="#E35A26"
-       id="path39" />
-  </g>
-  <defs
-     id="defs39">
-    <clipPath
-       id="clip0_56_2">
-      <rect
-         width="347.6"
-         height="94.09"
-         fill="white"
-         id="rect39" />
-    </clipPath>
-  </defs>
-</svg>
diff --git a/docs_old/assets/images/logo-light-mode.png b/docs_old/assets/images/logo-light-mode.png
deleted file mode 100644
index ac04c20f52f01fcbdca2913c22df1e0e9f63d13a..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 9174
zcmb_?2UL?=({2bL2q+ec0zyEl^e!Di=|w^59YPU8?>%~yqcn*adQqAP1PQ%IJ(N%d
zga|>vgMf7DC6xO{Dc}FEyVkeXT^AO~o5}3iduGo(GqX3(?&_(Zqhg{0fk5XpHEtV#
zK;RnSxsmb=@M?lMUkSXC_!y|GfKWXwv%rVHpeCA*y1Jknz%wO?oa7RSl*j`7f=HM^
z<iDRmAT1K+lV<}G-Xk6`2o&W4B0J(S1Kx>0FMvN_^wB$MHp!{PZ16ERxF(zQ_!-nl
z95yV<pbxyA_0%x+0f88}h<_xS2AsbDX{IjsO?*vsb!6>5U_!PI9(GWnK$s_S6i7Z$
z7I=g~eQhCuFgJG}*+2!3-yE{QGm$LJ0r}11>#D$EqI(yj>fsHANC}AuiEt=VK_C!$
zZwE(NgWGCH!+}o<9L~PJp0dKi0RaI*0pda)-cG`zGBPs4B4WZ~VuApNpihvyuWg{9
zyU+DMLXPF!hWglhyLkG#c(_A|a&7HA{CpKSIEWLS{QOy`uZ!d9Ozu8MY5@ug6TcA_
z6%r9Xkqr!$CsJkgy<MQd%tZN$qVm5v{|D`89(iHn<o~glKbrog0;(!f$qS#DO_A!V
z)prUIi2bSN?OXQ)Nfy5;_>8KhV!UkR;o{;A7>igizN2eFMJ3CI-o4Bs4=W@uTOX6N
zAm;Nan!$VuDwC*0^7_}Cv~CA}5jU9U2Xf!(NV26M4Ggtpy=Y)W%1*!o*E9QLN-w_T
z^yv(21?w$02Qvg}dRw}tw*0%ZQtffQyq8ekYgk^qJ|fe@DCQq*HnR1p!??{@$z4%k
zRo*lnX_-8=<a*e>GDr(IwT@uveYvq6-^y3+Z<1{984ChaLLz~eE=wEHt#j|g$u1~^
zNdCM$2b0y@IhT2z`1Kzit4L75X7!76zd!wcp(CgCjAoKoKOS=-xe-LNwBW(cd?Mxo
zFO~c*XLwovW&@VzKw*chSq7Pv&*PFBZjVf|Bs3lR%@OeGu!oub5c|EPn2m2aUP}n=
z-mRw4`%5=88k<Ad1be#Y)`d(7O<Vl}0ak>uh>(MSmAZDh)uiSLUY2U<C>ylTu_EG(
zCg;n-=M${06<)rqo6FeE`%vOnT;A&nQw6sjDjeWx6NYE^hZ&{OyyaiR775aE8)Qp*
z$kZ=~i`yPZ`#z>C!$UdZ9ZEU^mks|qR(v4~<Z<t7b4W+rv31~&86X&ipZMPCi${vW
z_b)>*-~KC)XXYD`c9*lr0eY=x*LX;7>Kl8WLvkPTWpNRV%j?Pq=)2%+uDC*KmtCHn
z34HTZ_YDQJ1*VL;iPV?@dJD0{qus##LoTau=fmdNgDox|lCL94LA82d&kWQAleV7r
zcDKKoic<l94foRA7xJs)j>e9QW|yY^Y$7BQ({%P1!;`T3oCS)boxX6>lLS0UdI+{5
z{0Gy8i6#6ssL}}qPJLm5b(;}{KiTa~U7MZGVJFvd^I|p+0ZRp;df^?@1Db1o@y|Ux
z8Ut&8St-18HnO><KJ!-GVvg(e@=jllVsy)-n9C2hL+8j<@Kv6ws?R|$|1o<|@c#A2
z6Ti6nAFJPu86!;QL$Esq#>-V~NY<V~r5;5&&Bm6gbpOzwTW)$=&nr2IKIwaX0;hDo
zE5avqL-lj_dW=|A<mh3=5Yzd+Z4sG5ju#Oz7C15H`W5(idD%`18&OxAK$5XHA`LR)
zzXkmE;B{W*Nc?pn)l)a85mt_r`+wkhzu@KeRLuqnrd6NW;m7|wQ138Bs7jqhVpDP6
zW@<K;uUYN|>@DKk&D6galc6swK2nQj{WYJxKyby4c)Yx`y<T6vg7UB0UHAuL{=mg_
zE^mbmHTEPg_U&yD9DDY{gx(VJeRdf+YtIuseb{`mxo1(x3(4k+#i^!+?|W12rur?R
zWC+d|Emjg{zap(EpySr{LroUJ2^G%*Dt*34qZ2lo5>`~&lIgBTVs4WMMt{8RLDQ*n
z61vXZfF-(Ot&TSw1inlU!DRf5wKY8@b(2wgrq%XQAA63n%^LtFE;ot_oRYFrKqT(s
z(!uN}$f0vT(*7fNu)T><GuWHkaW3<9ZpXLTrq-?fQp54_CFdcR%1j^gU^<f>H%7bf
z@`|(Db8RUZyXH-%4}ZtkdDTeJmdKKLJ&T6&WBls~uSaW1H)$%I@^AfKz0E^#!yVg)
z0e@|rB2&dB2nN=o?sNiZ@Is;q7vtWuo|24!wU=!Fj5!W;h=?Q*2!-D+4n000eb$OJ
zq%&LT3wj4`P7kZvY-|~6Y?@YwW0OlQCdbBt%ZgCdt1TR>h+Mb88BrT(M#U0<F9}7l
z%Um;AR*lT-8WK1&UiIupk!ysiVvjj4FH;3SN{EOqFq6v|qtT!NW^3=&OkROtRGEKL
zxvlij^8~7Ldn^ZT9Xc~wx4Tz93QHDJl-+fb%VZaU!3)F2^~|FO*+nAYec;iogLSt`
z3(k3&%lTGJhu`NcP5fwyDWaVp1|J@i)JwUjRIGakb%Ih9#6eU2=0VSJ+s%3fo}^RF
znDKVmj~%W-l<sVc!;(Gcb9ASIT>d`RpXCLJO-wF3t=s5ChUx_X`~Ir$Vm+ad0>)PO
zUp!Y96>6d)I{~JNGR%Gy&xXSGIRxvcO~A{&9-}h+u6pnLO=~S9UM)-G9aDpg2(%xh
zeumfY!3YvD9!$R1Dzpdocd*}d2PS{bjVfKvS#WLau9=pH;-Id)^U_9^10ka`!R3bT
zyXkMr%L;ztre`{xNl*u$HP{q$2Zy__hG*vvUCj<QQ`v-z`4k@VsLQ~*K1%2w-ifR4
zr$r|`r4OvsFSy@sm2TVnNv3)!2eWI_tfY|lokAxmnjrqSbXalGT(T;vxyM5HvH7>^
z^)No3hKAyYcipNR_P1*-)LYP+iu#Co?)i>`OsdrN5}(7amCmLie2i<4p}%F-Q*?OF
zpkzavLMd)%t$HSCo>u8rvy_sx8ntJ6#6^mYWlU&@gZlF))929>n=7+at7t7#PrJBJ
zl1IrF5*azO!fMe(b)0fY$Tqcr{9EeIllI5!iLMX9Ou3!}u(C}l5S?3d{^+ZxF5ai;
z;2+V2<uL8tVfa}{IZV-Z12%a;85d=ZbFDhOuh_P?T|uU_|JA?lk^fkM{!ZEzixVh>
zkpZhs3O?`nDd*9MTTN%vaDuh%lqlOHym|qbvu6kZyf$$@Y!?c8a$r)9A%DDVQ*A=U
zYU2G~nCQ;~Eu6?YpAzbF)EW9hnUrD~!CA4N35mneZaA*_6suP(mfTNKey?^{7|0o!
zPx2E8Tuukz&J@hP?-X~0NWd0JH2;H#=wRhv?1)U5Nbqo!ru2h+x1EE3`nS*zQnI9o
z<V3PS@&K)X$Cdwq3k+8l=|?9$#?{cl%^T+(C6qIa?S0LooF#JbSc4;v20$%`yyo(T
z%@OkHSWXxRg*oH9IR*yuCmlSZ%lw;l8NLrJkF^eUAU<06D4G*8o6*uMq`mxdew}0d
z{QIU||MoM)-WYi08`A%_Xof=QQ{BwM%vw@<^>SVSwxe_d{bnlkAoNWSD`90|>f(gI
z|H~_szfl#=PPSLFxUkMGt{pje(=KsG$ik<?bSvRY3o6*SHeI+G3EBK&HSSnw#p}pS
zsOi?mH*W2RXALzDDLN^QKU!-DJG<;kUcF15$C}lfr9UueZyjEzIQAGnXlciC>KK}o
z037}+^`7#SgaX{B9^vr+h3O`DF~4T(A5YDegWn1dL+^n(X1?EB?rr*eRNddNy01sg
zOd!*#8B$oz9&qmdCUrVm+@&-i)~kI38V<7zB(E-k>>oo62^lvRkd_PaAM8#8Vub3#
zL=}Cm(n-i*asdr^{Nzk%PDsk;0M1rDT0bdTE+i)cC2FoaMf7PHIrC#gZDHiI@kgjP
zNj=Jy<Xth`y-kfZ@=Rh0>QVg~Y&K<gF-X(yn!Ik~tK|^kl2^!&QTG1|&}T4k;U|j%
zW-`vFoRcz(Bn$-mRR9o?8?SrqW3p=2MgV30!o}a!M=Q8#D=!@{mvN5f+Gs~l`Bw;<
z-C<_c=)7l?3~6n%^raYvi7|?5&5tDQRiz&vLa&F;-#5Qfa;z|1lzf*uqivMl=OjWY
zS&9K=K8yuR#T^6oLIFStA2arUa!mQZ@nD?U#PIrhr%XEf82u%81CN#b;g0@S>mmKc
z9;T(f8wU$(x~9}l87$9?{NH?WCRtistu9H={WV(adE<CpBxJG|x7X*s8J(d@r9qOv
zX4KtX`k8afr`(IJhY)Kq8P71Gc+OrSX1G=a_lWOwBb2jsBDW@xvv;G8R|*l4gZEi|
z$jG3hR{H~Ma4c1O>6`tcy&EHeXzJp!_g}O(?bsx%MuQB&qdl5eFB}yjPZ@qGnf(Q2
zlVu84=#oYO_4*ZD*6^wK$TCdOkH+1)>V5!T9k(djvR0tBkW0bylR-mPmJru_0~Qm#
z{{%ul-~0LHhs4}0Z6x_7vLR|dFNsrF8!1xyLPQgl5t4$_OUGw}iSuSF!6%E@q{Zr{
ztp*6?!W_NY`b9cuVD#9VyLS?Z2y~=a`9%QGx=ZX{pOPK`o?67)cjjpQKy4}o6eH!e
z4+Q8=NZHSULW-Vi<U|f-ml}`)q^h4kIW1uCUz|vOe?zMxB@1JB5qr-XI^V=wBv2G`
zv-O{`k6VZ}{n9bmV&#6~f%oR{$QFa^*3ZKFhXLLE#xL>sd*NKuW_zTBRK()SV7*U>
zWn(`&L_qy(@m#E$E7^rBw<m;Z6PLd+<1JQ;JU%H|G*ZL-U>*L4dobDH@+>DqbNzvD
z0e$hHYU(?!$?3|QomV(;AIGx7#%^K+<#L2ALw$dwmX~^uuNgyKABubUZ`!7@*DVol
zZ{8S2*6q<2KAaI5F}igXOc}#D5n|oZOELdr)kh3%R3WgiJmc|a2JJDUZ)s<%QV~6;
z#-m*40czDO!Rx1i5CahzF{eB5+8oxL6BJVax@M&l1mvgpjLf!|j|VEZ5n+96`GF2%
zAa3iE=P?}wrVzrI)1BX=j$AyENZ=M3;jbZ_!G=$BPc(a#-c(4a(;wYfN?jL9urF4m
z9TnVty8=}t@I>)<ZI;jMcs=pYkko9HK-zbA>@gqo>eXE$=sHm=yoK*&7k-1GDYL@2
zq!f9Dr0_Y)=nZ+CtH+l5x`K;T&|2xr>dx-&PSN=d!`WV(gu?2Qe(GB#)ZR7m&%HRa
zHWB2({+&qt55)neO)zxXHvA@O<*?|zrsYN!Rvwr*l$Rk*iuB83TE?BLb-S9`gTAV}
zjN?ySEjl{MmYT=(JI!k5klMnCx~~uPWx#%99X;rtgNsAILK=2(fjhect8=R(JrM`U
z{m`((!E)uEsAeH+MET~kFA^02al=(_zVvSN=2pGz|J^_x1?4?5wGeXS2<m?C?~P1Q
zoX+!;0DfOhH1Tjql~f3SXX`0Ks9l^8q{?|O&UXsT)_}i<Va0?_C>UOV^1pd<8Y!wv
z&xs0wVdSC0GR=p&AxQY1go5V4I<xkmLv-l)q3lN3e9H4(+g_|-Xj$Nr#ek*v@E(i$
z%-i@I9bN0ilU{L;Ln^vxsn+F37TQs?9UFFW*46P+w?2HJP7I43pOjDn7b)QpSx%~l
zdySR{RhDDn3kN&0x0=yeE%kepo&*Bg@ka_(Xz)f46V06e50+ap%XNdbQICs&5(wz$
zoPRxI)-$OxlF8Z5ZxOD1+wewATqault^DW6bLF&4tyO$8+kUQi3TTwz-Azf9*lMGU
zOg0*rPI%MhN@s5`!KbLiP`|Hq!Kuve@)H#RJ1`TcX2R?NM=UcDw8JDPc2d3-3963#
zV0fCicYxNs=a3amdP;f(KnKa=f>Re?W1OISvNNvgFOnU}s*Sb?EIpHs-nd^MXW*jn
zA(AxL>e6l|+lB(PaG03p`rXcs1pM`71_vvA1ucaw*?y~ynPm<e?+CWHPoGN?-!tQL
zcmF5kfrVPeOze_Z#@Wq=#UFbkp}hoefq&H8zuwtMn{JkA+GT$`mWecEctNmU?&6mj
z@X%_@)EP!r9<_#;o5aSACBX5{NfG-~6Ht!F<moBpiaLOK|CPt4!jLX}cc6UGJE6F3
zcyrH5p=sAk{Lu3KiA}OYj(Z=?i~;7l@x3nf(XP;affL96c_*@(cl~2W_NmeAfu1pr
ztGD5l(g>7IoBx$3W|Li9skR<CwMet_Z}|^_6#ODUi7-1@U-FiMg1@)*G(!awBc$=Y
z81KnGV@!Z}u*ZrEpHP&A!8wh0UMGDM0l4nd-%NO_t~S|TGkxiI%7B{!^-#}hRoPD}
zu&YZ>ZL&)3XPwms?OQh`A)%A0+JBtGkq>}6#~{3kFFE{GMIQ*2Tp0tLbaj(U^-D^6
zLvWk`n~ec5^UJzx0%w4OE)mj#xA|)L1IYr{(NT*6tCzo{S{A6$7cabgeH!Aej1bJV
zL)Q1;Qztx6zWu^9Dw&2@iY(MB3lPL3NT<3pms1wmBQZ39V><;7n1})Di6E4D*B){c
z-Qpz!V277Y3IgX(^B@XTIx$3_v)^<!x%n3adF~G<uqhZFa-hP`myUy0U-~>gk=OiF
zd0cK=2~kSD!xO}+J*F|_?`Wod%faDVk)E0MWP7_`+lwW(TxFD7n|=y(1f?t1#$P|S
zkgPEuz27}upw?J*jw)7awW0H1dizPCR@RwTS>N>7AM)g@xo#OFC=-uO+%7lY3%%(Y
zC@ziar8BcnOty+rDltJnrus#-ls&2=MFLkI`enAt!{r*p;m3&R<!XMxqS<38ps%+U
zy>PEo8sUMSx1^O_N>a<gwMr%(xz^{qk;b0PANycPFuZuz%-ht{d^EXgU~^u@mb3TJ
zs-`CkwZ<dIYHZ+frh0_yixN|A<@z&DYb_L_!jf|*S8KAk+l)0wpi_0K0*SnboU&!$
z+cGsaDXDtKX<V6<|2Vg@>yjp_9+kiEaN`ou;({bVp*6H`Be_qNT-q?_rlcLm#im(B
zo@ubKe`i&yGq?o)7)d{1CiISsPWX8SdS<+HGB&S4t51t*cynrQek7-;5n0NKOHm~_
z2P38k)SYZ)iS+U5Vl&$h=Bm}2-xwP2$2-K0az(_%1i_N{`6O_~hueLUt%$W;9F*|I
zza_XaZX^Dw3)g;^--D|Uar^JY&cFb7f@@L&b?`xj)ips6CB>A#KGHCH5n=XeXJ$V_
z@~wOl3nN1%VkQZ3Xu{i>^>mf<FgQ%GxAGegAQ+x7<go^Bn;bD<HS6%FFUk9$<qa9u
z33>rV&iiFxoCFt*5u{khT~jK%a*_I~?lSD0N>B`S?}n_t%K>M|{Z0%oztvOk*E0IL
z1PU>XpQ6U*TPQQO?8jYi(UK`UE}0i@`96Gq$Ef(QP0LmW*{r*G*J5SKkv*u=XV+cQ
zC^mwsU&#+UFz(p<V@ZitrX^wyVM0o?tNH%)YpK`=y8|Yz0>&%GrW+&ZRg(@YjUmVs
zGqxr77RJn*cSmj26)Vie2w1(Ml)ohRBMo=>E3s=u?q=FihUz?)sz2VAlN-n7ZxQ^u
zFKQLU=H$TQqKjdi@m`yu9kg{G-5#7#h-eo<Ml<QkL}vWZ)NU$e?n-6f_h%L}_OKXP
zApmaRrj!DXWm*i+lmo!KT0_BYo=FCWjd{9dSyiF~)a96(s=+8VW}zs?za6ypDRLIs
zV$}^;ub6grC2qxcQT^%hpE44g=<n4XF_fd!ZZ~Apwd<9bVER*LkJFg3=2#{HM(cz@
z<0dPOaKT~Z5c1<^gQ1FchU^skDHuoL=lw6by?Lq?Ifi%{e;Os{%CYhJTTX6@rb|{%
ze`zqy?5_vCEIm7`Mm(9!=fd#U4Aij|=Uc;NJt+Q$@-bmjXLIj*ng(Vn4T5k;BLfqf
zr<JjB&4WC-SW0J5>9GHD_nn7n4UML4zmHNde}fwy$#Eze(hA*`PL6SS;&_)yBL(Ri
zEVyH^$`hQB$G95Exx+2Yo5*}8ehlw2<fG7_oT%$Eri9n3&>b5yQp!W$n3qrKe`HDx
zZ0Kj80I!Z^4V&S9Us5{}s2+08(@Hzs)Vbip5;fNLwN4f)*0qsF5K5=3s=Tbr3HZxF
z^*%FSG-n%P2GOO^1^rfh8NfAXP9P1WUGtfeHj~V_;LM#(;*wEO@_E}}gpSl=(e6wQ
zKcBeu*#oPNIh1<ci!+xf%Y&p=ozFxiB)aU}H-u0RSV+2HP_wtaS0b)2QSXofB9n74
zKzqhgX0rC-EX#whD@G$fLqwxV7TBhOxnz*Vvn`)KQA9jtv+xp?gjWhU|8`|%Dk+3L
z_Vtg#JY@#od-nH6tPDAaX{IaFiB}(8dXcm#-Al-_2?p>k)q{-tmHT<n&vtj~rRE}D
z&EwRrCJkIN>GYP+s2~qD%lC7|8fw7>)Yi1b&H}5s=?t{pkDK->s(0RzxSqFnWt(~R
zME#tRq_beYpMM#m?cbt$GnP3vPBK}B8sv&ZvqbC0RGk2GckOLp+oy>~@P{Go=igDw
zC*&8+buU|=6HoK#2G=$v6yI_eb-5eD5o9UQl^KN;d!En!qC-p~%g7|I3X3;(*1Js1
zKk(N;e-(vr%q*lV<`=q+53{QnG<JcJprq<0M&_nuJ9p*vcf&1lSAx>=<AdJQC(^vA
z?wcsqtV8O_w&v&h;AOO)vNmWf<9x2W4mag0OQH3*T&<U=34mxMIJCW(3-Zm0$CI5u
zT(DN4T^_Nk-Pxm}ZEBS}=`vv$D)-S0)eVtJ6zi`b({qriB9s(0B(TiBROuHJzhnBL
zR^Gi^>nM_dBlc@UX~X&K7qDVrY?ZPD*HkH28oZEB<}svXz?W}Fc&nH7;C<{xhB>_a
zpce-g{=*P}i}@Np28mEst!51X-_q?fUUjPGt>18^H-5PP`R8p`j-35k3y)Ya9CYwN
zT`UqfTQFeIjN?k<OMC<jkXcSrnlsTgaPXy?GV{q4ntHZWko@T1C-S|qmGai{QqR_D
z)gRVbwRr+J1t9PMaLSm`8fHWPiOP9mM+tFwW2f>5Gj^A8pBlg_e1BG+NYw+n4wayA
zzF~&156hRi<}EC9MVK9~$G6z1MK^y@lGU)4YE@(HmdVsenSSSDSZVt@LpF77F6>>Z
z{|2oCq?;e>_&nFfp1~1kyIGT<?i1$9N6o2jij302vhaunXS7!iw>qy9f~D2)mit{^
zHN?XKN1~4q%-heHQ8pwtr3P??&41WIshNiH#vd*@7x`Hhn4I}y%q;n<=+f6YD|b#;
zOfHGdD?07#8LEaEHnUDHY_0{*`Lc3)MxLwU7TGjrqW=Z0%5Oc3^~UE_^~pqUwv-lV
z%(?Md&uHDq`pQ%x5S`ati$b%nZAkbDGpETC%OpB@3vam?A9#{-&osaLP^`2L3}d?w
z6IfJH$6M#^Rpy{{y?8T7u``GwP8g%cn2}3*<=H9GcC`F`z1I~qXaQO|Lx`XR&NNNp
zP=S0^$7kT6e`D-k8ztAW)@un$&LWo#It}~o*FO{ucpslZ6rr^758WrZziARnA73(t
z0aGgeTiF3{<`ni9ds^r-=}^u2N=t|-;ejp+!d@&3T0APxvNIwN2eL9tklG&5^DoD^
zT*ph-$E%wIfFil?aQ_j{%S(X~n?Ytyx0;qCGIkGgY!>}@@IZTlJG(ungNNc@(d5@A
z+tq$9p4C`2^M;sa&F~?45`N$iX$wAF`{4+;`~rguX-}?M&9my=h^c-hRVNG!aj1xo
zam7Pz_<a*w3{bW1=n_FTdwzZQRq8hT!2|x;(%R&AG1hLSz44s$=`D1fPo~io2(9ia
zI_=lf^23{SQ?KIG@pW#xYhLfV$`B+Hi~~kO?$618-!`{0{IaBe5$Sy0j}aAYE<=LS
zg)%|?NG&|d0>g>L0#i}e;Ef(=<@go;gL(_sU&=@KTyPRH2GXt%bd)U5BhQpEcRC6K
zSbfxuDC>jM-+aC-@QL*=jG6ycC%G{8cvtK+mMko+Mj#DIaa7le0ClVGMbyQ&Lhq6J
z1#+qLMb*-BBX^H)ps!H@x2xwxuTqhyDvyWMi+0oy5se5*ndGuz<kg!LDZyhbpRS9%
z!7!(ok5`?n3{Eb4&XdAQZh&2+N|~4Xb2S?aJ;SBWH4Sgxl;f-D>p=Gr7uk~-a{WSw
zp2VL@9PmRGa05kviuB{W)03hKpc;pHv{N3Hmp~j}1}<qbK6l&jost&G)0aOq0w>wt
zL5H>}TIJg=!HE3A9eyqTjl6q@;6Kh#HK$YEIew$MM3(#5oj`Elj2guq1RNM(v=Q^D
zR{Qiga)#)cw(+J|S=LwIkGzg}Z3!3y`|T3Lu>z+V^tAKg1`c!NYRx}BIvz$sHUj8b
z)t7VH`CxvKr5Esiu`CRq3lAc!tkvy1E5(2cIA|uj1dj9(X~G<m%1L`AZFc7VP5kE;
N%{zLxQ7Sf%{trs+hC2WN

diff --git a/docs_old/assets/images/logo-light-mode.svg b/docs_old/assets/images/logo-light-mode.svg
deleted file mode 100644
index 9dc4ae1de..000000000
--- a/docs_old/assets/images/logo-light-mode.svg
+++ /dev/null
@@ -1 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?><svg id="Layer_2" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 347.6 94.09"><defs><style>.cls-1{fill:#e35a26;}</style></defs><g id="Layer_1-2"><path d="M101.92,79.35c0,3.71-2.14,5.97-5.04,5.97-1.75,0-2.85-.86-3.45-1.64h-.04v1.3h-2.32v-15.8h2.32v5.88h.04c.6-.86,1.7-1.68,3.45-1.68,2.89,0,5.04,2.25,5.04,5.97ZM99.6,79.35c0-2.21-1.08-4.07-3.12-4.07-1.9,0-3.07,1.55-3.07,4.07s1.17,4.07,3.07,4.07c2.03,0,3.12-1.86,3.12-4.07Z"/><path d="M108.52,85.69c-.64,1.75-1.37,3.14-3.54,3.14-.66,0-1.22-.11-1.66-.22v-1.77h.2c.97.13,1.7.07,2.21-.55.42-.51.73-1.39.31-2.45l-4.09-10.14h2.5l3.16,8.64h.04l2.9-8.64h2.47l-4.51,12Z"/><path d="M121.12,84.98v-1.14h-1.14v-1.14h-1.14v-6.88h1.14v-1.14h1.14v-1.14h4.58v1.14h1.15v-5.73h2.29v16.04h-2.29v-1.14h-1.15v1.14h-4.58ZM125.7,82.7v-1.14h1.15v-4.59h-1.15v-1.14h-3.44v1.14h-1.14v4.59h1.14v1.14h3.44Z"/><path d="M133.7,84.98v-1.14h-1.14v-1.14h-1.14v-6.88h1.14v-1.14h1.14v-1.14h5.73v1.14h1.14v1.14h1.14v6.88h-1.14v1.14h-1.14v1.14h-5.73ZM138.29,82.7v-1.14h1.15v-4.59h-1.15v-1.14h-3.44v1.14h-1.14v4.59h1.14v1.14h3.44Z"/><path d="M146.36,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M154.34,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M159.47,84.98v-1.72h1.14v-1.14h1.14v-1.15h1.14v-1.14h1.15v-1.14h-1.15v-1.14h-1.14v-1.15h-1.14v-1.14h-1.14v-1.72h2.29v1.14h1.14v1.14h1.15v1.14h2.29v-1.14h1.15v-1.14h1.14v-1.14h2.29v1.72h-1.14v1.14h-1.14v1.15h-1.14v1.14h-1.15v1.14h1.15v1.14h1.14v1.15h1.14v1.14h1.14v1.72h-2.29v-1.14h-1.14v-1.14h-1.15v-1.14h-2.29v1.14h-1.15v1.14h-1.14v1.14h-2.29Z"/><path d="M175.45,84.98v-1.14h-1.14v-8.59h-1.72v-1.72h1.72v-3.44h2.29v3.44h2.29v1.72h-2.29v7.45h2.29v2.29h-3.44Z"/><path d="M180.83,84.98v-2.29h2.29v2.29h-2.29Z"/><path d="M187.59,84.98v-1.14h-1.14v-1.14h-1.14v-2.29h1.14v-1.14h2.29v-1.14h4.59v-1.15h-1.15v-1.14h-3.44v1.14h-3.43v-1.14h1.14v-1.14h1.14v-1.14h5.73v1.14h1.14v1.14h1.14v9.16h-2.29v-1.14h-1.15v1.14h-4.58ZM192.17,82.7v-1.14h1.15v-2.29h-1.15v1.14h-4.01v2.29h4.01Z"/><path d="M197.96,71.81v-2.29h2.29v2.29h-2.29ZM197.96,84.98v-11.45h2.29v11.45h-2.29Z"/><path class="cls-1" d="M15.18,39.46v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M39.46,30.35v-3.04H15.18v3.04h24.28Z"/><rect class="cls-1" x="45.53" y="27.32" width="3.04" height="3.04"/><rect class="cls-1" x="48.56" y="24.28" width="3.04" height="3.04"/><rect class="cls-1" x="51.6" y="27.32" width="3.04" height="3.04"/><rect class="cls-1" x="45.53" y="21.25" width="3.04" height="3.04"/><rect class="cls-1" x="51.6" y="21.25" width="3.04" height="3.04"/><path class="cls-1" d="M15.18,51.6v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M39.46,24.28v-3.04H15.18v3.04h24.28Z"/><path class="cls-1" d="M15.18,63.74v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M15.18,75.88v3.04h39.46v-3.04H15.18Z"/><path class="cls-1" d="M15.18,6.07h6.07v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M24.28,6.07h6.07v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M33.39,6.07h3.04v3.04h9.11v-3.04h3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04h-3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><path class="cls-1" d="M51.6,6.07h-3.04v3.04h6.07v-3.04h3.04v-3.04h-3.04V0h-6.07v3.04h3.04v3.04Z"/><rect class="cls-1" x="57.67" width="3.04" height="3.04"/><rect class="cls-1" x="12.14" y="6.07" width="3.04" height="3.04"/><rect class="cls-1" x="21.25" y="6.07" width="3.04" height="3.04"/><rect class="cls-1" x="30.35" y="6.07" width="3.04" height="3.04"/><path class="cls-1" d="M66.77,9.11v-3.04h-3.04v-3.04h-3.04v3.04h-3.04v3.04h6.07v75.88h-3.04v3.04H9.11v-3.04h-3.04V9.11h3.04v-3.04h3.04v-3.04h-6.07v3.04h-3.04v3.04H0v75.88h3.04v6.07h6.07v3.04h51.6v-3.04h6.07v-6.07h3.04V9.11h-3.04Z"/><path class="cls-1" d="M102.09,60.7v-3.68h-3.68v-3.68h-3.68v-3.68h-3.68v-29.5h3.68v-3.68h3.68v-3.68h3.68v-3.7h22.12v3.7h3.68v3.68h3.7v3.68h3.68v29.5h-3.68v3.68h-3.7v3.68h-3.68v3.68h-22.12ZM120.53,53.34v-3.68h3.68v-3.7h3.68v-22.09h-3.68v-3.7h-3.68v-3.68h-14.74v3.68h-3.7v3.7h-3.68v22.09h3.68v3.7h3.7v3.68h14.74Z"/><path class="cls-1" d="M149.89,60.7v-3.68h-3.68v-3.68h-3.68v-29.47h7.36v25.8h3.68v3.68h9.22v-3.68h3.68v-25.8h7.38v36.83h-7.38v-3.68h-3.68v3.68h-12.9Z"/><path class="cls-1" d="M189.18,60.7v-3.68h-3.68v-27.64h-5.52v-5.52h5.52v-11.06h7.38v11.06h7.36v5.52h-7.36v23.96h7.36v7.36h-11.06Z"/><path class="cls-1" d="M206.38,60.7V12.81h7.36v47.89h-7.36Z"/><path class="cls-1" d="M221.2,20.17v-7.36h7.36v7.36h-7.36ZM221.2,60.7V23.87h7.36v36.83h-7.36Z"/><path class="cls-1" d="M236,60.7V23.87h7.36v3.68h3.68v-3.68h12.9v3.68h3.7v3.68h3.68v29.47h-7.38v-25.8h-3.68v-3.68h-9.22v3.68h-3.68v25.8h-7.36Z"/><path class="cls-1" d="M281.73,60.7v-3.68h-3.68v-3.68h-3.68v-22.12h3.68v-3.68h3.68v-3.68h18.44v3.68h3.68v3.68h3.68v12.9h-25.8v5.54h3.68v3.68h11.06v-3.68h11.06v3.68h-3.68v3.68h-3.68v3.68h-18.44ZM300.18,38.61v-3.7h-3.7v-3.68h-11.06v3.68h-3.68v3.7h18.44Z"/><path class="cls-1" d="M321.81,60.7v-3.68h-3.68v-3.68h-3.68v-3.68h11.03v3.68h12.9v-7.38h-12.9v-3.68h-7.36v-3.68h-3.68v-7.38h3.68v-3.68h3.68v-3.68h18.44v3.68h3.68v3.68h3.68v3.68h-11.06v-3.68h-12.9v7.38h12.9v3.68h7.38v3.68h3.68v7.38h-3.68v3.68h-3.68v3.68h-18.44Z"/></g></svg>
diff --git a/docs_old/assets/images/logo-simple.png b/docs_old/assets/images/logo-simple.png
deleted file mode 100644
index 2c14b4d9f9d8e38431c44c004594d83ad6a111e8..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 2196
zcmb_edsNa_9Ea7cY-(l8=XAB1bt<(qEfs4EmmU{A2rTe<GGL<Ot28OJJhRlum7B#3
zoKij@P*d=Mg3G2k#fE|S#7f0PiC7UnLiclZoSkFY&gqYP?m74OyXStspU>y}xi|Fy
zY_Ea7xjqO4GVu0-9s+^D2Ef?8R1a9m7jLWvre%>{M`J*sjTtioEMB`E_yc4X>v<#=
zPQb**1xFtT#l^+h{SbkVITjpw+>Q{9&3@``4gxJ*>ka+paQt;;uP~j8HEUFlkTB=W
zS=G1G?0H=;ojO{)hnx<JsDIsDo@;m4<bZFgh0fTo6Op`nlm5EG-byy1@~5BWHyn*m
zxRO!+<=LnetJ6@9&wC=OH^P3~C2WdmBYBKeRTg`W$$QeX3pBD3;b;&)zKS99&{#R7
zmfpwu+No^5s%M-8%dSBeUAUmPVx_^C>&$mUosur#Kr2O{b$MXu>kpmR6V6%;D~2a`
z`B^BrxWbCsP+f?FI*U<l>q=J;5SpP#JtUh$*7`$YoF~-JQqmea6g){|hUXsXR8)6Z
z;`E?&a_D{pwU%C?HtNJEP0D(Ga#}G#8bf({t!yNATYthQZ_aurh@;&Vwv*`eB`LWu
z8%LI;m?0@^YDx7CsLBf0BShdW$hn&HjEb11?+MEOGOCXtKmtn|S+-);yNY^!&_I3-
zq_Yp@Ub0m@lu0Q~=`&y@+PLEA(_1Hqiia(avsns%raN8Rm7}z?_Qn{>yN$T*9vlWW
z0xYN@V#mkCK9)oI&|^3qTh@IwkD>^qDmPoBC9oQX6ibvSH+c-EGSc!6)^ZaNF9>q-
z&1oj1eVaMkc0Qix)_!?R*`!kS@}J}AItR+Igv1x6OkSY~r&L2_!m+@M;bQ|BN$M}O
zgVHH0BPy=X;!;WuJBW(D^Qd90P3-2k-;zv6+A>3FE#(1%D!jIc`eHLuC2oZas0ru$
zTe6x>n2JA&v=0SUw712c3*wCnoQ>q^R{K6v{uvh6i7YoNk-uu2<BoYY0Jgc?2Pr}l
zcnR`R%IBleK2O3=$03yK9a$4O$nw_d@adM$jGl`bHo}u9X+9XkHR)Qzz4ep6oCnoz
zJ*I)K96!E;#;UQ2E*fbqZ7A?9Y2K0tap|s&Z4{$>HtexZbVl~_yWq~WLdFvMWxOu^
zn1!J9*E8*XksbDjCF4cf9hU&@=SMc{^sf_%hcwc5g{{l!O!VE3$-sQsqm)ohGE&1v
zSc!<r#yzZ;D5mNbz$%otL!&2sK<n&k<(D8&9Ov?SP)*Kt`%X9njv^$(Aq|?v@9m%k
zkj&F_A8SV#thP=wYQM(1$K71!2(<o>^<?!;c=O<LxGL&8hujd^A@FFAk5E8@^y`BK
z2M|LAvi=<(;7`sOVjGHg1VNxC7+w4BUw+ECqDr7A;t>!aW9sl9;H(9LXO@I}U8K#!
zL3lMMR6{A!CV9V@5QP!OpNa97*l_wmR$?JTSP-yiy=bC~lqFhNi`!hnehlyGecu_Y
zfC8HB-yahHN1mhI88%A77Aw;f*X=WGN8QrFiOse%q|Rm(Tm^=kq5!V(W(HE@cZBYo
z+W8PO;-uO^JS8U9t!^k~DFr$}IHvup_P>8Otf72#p7}a2gqsV1o=ghn631VO(Uy%1
zQ*^J34=Iw-ijh2cF*x&ya=gg5k_O-ego;E!+&{7-yoeaSZZPCwA;X#Z)Cbd0>R*FZ
zwQ1?Xq<fBNZDNG!#NJFwJ)^+oPK-0wt-0~a+2jY^=rg!J7zU*gGFBNJ$wlgnYo#Vv
zDCE*zuDs||_LL|H_`geUmjAbI!?65CD)&>570tyLMadjGsd6R(ZeYR1fyHVOnLEx3
zk$+)(o;kcT%E{x|K<4x<#s<)m=Sm_F#I#RqNGy(;O)Wr9<&-TPPXoK`aGT5*Q{+2=
P7X`@M69#2}8+7g;FTJ7w

diff --git a/docs_old/assets/images/logo-square.png b/docs_old/assets/images/logo-square.png
deleted file mode 100644
index 0592b8fa9a2d43e08f7367e187514beabc685e2f..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 2233
zcmcgue>9VO7=PXM8_DUO${Zw>>J%v?8>;KK?o<p-k(Mas$8e|ld9J&pj4P!(W4j2M
zm<WY6+G)EYn%`%b7Ny24v5n2RukJbBQqmvy+&{kOJ?DMi_kBLk=lML(=S@6<uv1+H
zT?GJus)N1tF#u3f1<!#M%fLJN?c1xt%gSK;<DmesJ!S4Fwd<OIPXHQWHXdOvL1)9T
zJ{VsBi^Uq83k(QF`vm(M1Y!J0qi`qytgv;kKIj^eDeP-WAtN=|d^udCvcoYr-k+#z
zaJy~CQPlBlPS#4N_0Xp`CiRZG7i%iko@ltrT4R*4!Q9p30n0nYN3d$!`L*VG&^8U|
zMx)f<wwYxOJ$SAfk}+{DT3(AEMzv&x8fG~6mz!NeiRip%m0ew4G^*5@zWlJy1~yZU
zphs-YAa?DkgQfWc>KtHGj*_)C<jR$0Yt&UYZh{`LHjcaUPr%8=FnAYzqOM+1Q(nD=
zNJM676!k?_#P(Ae1yoD*5{cAHXvkM->TBK|HTY6i|Ax(k?SPtS#=>2rO*I~#^P5WV
zA)Cu(9Wx4vj4h&aEp&t(a+*7(d-|9$q^n*c;}WT3<TEVslCP!z6jxbkkh6i-{s?a!
zR!tf)WQ1|CnErcHtwctGRrwax*Ki-(qB99<t~4~~`7wF#(*Y8L)aH2LAsNv9R)t`S
zlE$7iC%e_WeU)QH4NXUw6r3jGX48q7(5WZ>M&~0Hb<!fK=Mc_**w5Xx+3I@I1)Y;M
z43_L=ts5#V`cD5{a&HtyU?IVr%(({}UXG##ZQhF(#zguKOryuSiL?G4I)dUribz-Z
zmU=aZgVPf1lJ6Twv{HF@sQ9GI!z;gF{nN6+c5IG`#JtUh1k|7{Qk(8my$L40q^92c
zp;M<THG=F!LS=M7<2W4c52NMng~1xFG}5uLMk&#OOPp$8Gz6V_b5O0*K*a04%n#nv
zN>%Wi3G2y(meZqdWHTwd5Zm2oM=_UUkaW0Md)lyT?dJ-ZhXs_zU#$`uDrM_s{?tvv
zGt+|TxTQn4zEsY<_cqUtI&{blN_3_tO!^0x8EMXn>q9(Z42b87XW+Ex8Sfr7?a9Gj
zt90X$+RFEnc;$vBZ$^|fA>maY>~%$3jwclB)rS>4?re`7c_p+)U%WF$40zO_nxp2l
z-fdrVpNUwz%P~N4>nCv!*eQy|Nu*R?udxy4EkgE_+H94BGM^z_JH`s7-PaZ;d{9_x
zG36<iP0c~&pdx(*S|xd2!ZCrOb3X-RO7;)WXURD5%7^L`ZUqHAz9Y9YUS4i^7iqyL
zyLZXopa<yc_@L}+ePW1((!%34kEDO+?*9nP&GwdC$x#=??@On%3#D<}cM@Q<QCPT7
z0I24&QqBgya?$E&xT#osH$KD^b`;4j;Gqpm^gr1TU&Wb<kB)3DLR5K)vT$mD9104@
zf&88E$-Rfej5knm;H<<|Vg5T>Qt=9p)=t@s$W_%Yp9je|BBT1uX5Y1Yj`;fIa!@tp
zI4e-uyAwOtNy?MH4}BR1GaiA|qTtkC6i>H)V?Tbi{slMc2c&s=%SFS?2x)s{-H%+e
zOi<MLerGz{Yn>@{;6l`&I4#l~X%vD!^}KE_ANRuooB3nr<K0#I%+Yt{mPvjEVD-;`
zNx&<r7^R#^kun1xRSFTEUA*Q65zh~lmFhTgEvqH=V)Hp!B)aLlAp*52I0Ah#Zy^u)
z3@KfD4ap335h<m;8F*6Y{@^;Ez!I>g5{d-;LMgy|y0{N~=1(8LJ>R&%09O((XAoD5
zy`;7YD&ex^G?LZNElMRkC%djkt!&xhJ#8q9G;(<uoAnyqP?x8N)e<n@G2h(arBfY^
zPmyu*sEV12Gq(u?xgQMkdnS2j9%NutbBM)<aN;X0lM*UB)!-!Z(i3!3m6T-k>_OxE
z@|4iG`$LS~$XuBp(nhJ<9QVSdSjL~&`9dJJOvVOglH{??i#IGa7NYgUCOf3JwIBTL
O032)()+L8d{rWfXa;-}M

diff --git a/docs_old/assets/images/logo-square.svg b/docs_old/assets/images/logo-square.svg
deleted file mode 100644
index 29b1c20f2..000000000
--- a/docs_old/assets/images/logo-square.svg
+++ /dev/null
@@ -1,124 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<svg
-   width="95.38842"
-   height="113.55765"
-   viewBox="0 0 95.388419 113.55765"
-   fill="none"
-   version="1.1"
-   id="svg314"
-   sodipodi:docname="logo-square.svg"
-   xml:space="preserve"
-   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
-   inkscape:export-filename="logo.png"
-   inkscape:export-xdpi="270.25565"
-   inkscape:export-ydpi="270.25565"
-   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
-   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
-   xmlns="http://www.w3.org/2000/svg"
-   xmlns:svg="http://www.w3.org/2000/svg"><defs
-     id="defs314" /><sodipodi:namedview
-     id="namedview314"
-     pagecolor="#ffffff"
-     bordercolor="#000000"
-     borderopacity="0.25"
-     inkscape:showpageshadow="2"
-     inkscape:pageopacity="0.0"
-     inkscape:pagecheckerboard="0"
-     inkscape:deskcolor="#d1d1d1"
-     showgrid="false"
-     showguides="true"
-     inkscape:zoom="1.0083333"
-     inkscape:cx="280.66117"
-     inkscape:cy="25.785125"
-     inkscape:window-width="3440"
-     inkscape:window-height="1416"
-     inkscape:window-x="0"
-     inkscape:window-y="0"
-     inkscape:window-maximized="1"
-     inkscape:current-layer="svg314"><sodipodi:guide
-       position="-146.75933,64.2455"
-       orientation="0,-1"
-       id="guide314"
-       inkscape:locked="false" /><sodipodi:guide
-       position="275.80377,3.1942902"
-       orientation="1,0"
-       id="guide315"
-       inkscape:locked="false" /></sodipodi:namedview><g
-     id="g13"
-     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
-     style="fill:#e35a26;fill-opacity:1"><g
-       id="g314"
-       transform="translate(1367.0913,152.50405)"
-       style="fill:#e35a26;fill-opacity:1"><rect
-         x="28"
-         y="644"
-         width="28"
-         height="28"
-         id="rect1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
-         id="path1-7"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="28"
-         y="112"
-         width="28"
-         height="28"
-         id="rect2"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,560 v 28 h 420 v -28 z"
-         id="path2-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,476 v 28 h 420 v -28 z"
-         id="path3-4"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,392 v 28 h 420 v -28 z"
-         id="path4-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,308 v 28 h 420 v -28 z"
-         id="path5-8"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,168 v 28 h 140 v -28 z"
-         id="path6-5"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 336,196 v -28 h -56 v 28 z"
-         id="path7-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 336,112 V 84 h -56 v 28 z"
-         id="path8-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,84 v 28 H 504 V 84 Z"
-         id="path9-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 56,672 v 28 h 476 v -28 z"
-         id="path10-2"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="56"
-         y="84"
-         width="28"
-         height="28"
-         id="rect10"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="112"
-         y="28"
-         width="28"
-         height="28"
-         id="rect11"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="84"
-         y="56"
-         width="28"
-         height="28"
-         id="rect12"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
-         id="path12-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 560,28 v 616 h 28 V 28 Z"
-         id="path13-9"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="532"
-         y="644"
-         width="28"
-         height="28"
-         id="rect13"
-         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
diff --git a/docs_old/assets/images/logo.png b/docs_old/assets/images/logo.png
deleted file mode 100644
index 005b7c6d2195cf0af295c15fd2e167f3f30b1fad..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 6386
zcmcgw2~^T~yQb+Iota#QoE9~8(oD-yGjheWnMlngQ!8_W%F+}G6&1~enx+YJD?@QZ
zO0(4^H`G+riL4MY_bnwcB{v8a6<PfL^v!qfnR`3k@1E)0!-4nkUw+H`d*1hbo;UI6
z5qrh&)V>3OK#GSQVD2E$Kj0wHx&z<H0avcxuB`;jx1kPR7!YVzn(Vu-XS=B)a8Whl
z(8&l7bU+05boe<C7K=3sx`4#^pAJ1|gbokP<yxzOKwEK#VLu;>&YK&Fzd;I&(RwS2
z>F^r-#o|D-N&1Oaxx)r+J0n*1MyNGscd6#xaCUfh?Zn>N@AbQ1XW#htp#0E*so%0Q
z<D4BFHI!S*tHRHuE0k<FokNK`yYh|yZN^qFgUc;L8K}z#^9BaeqdG*3X}|AUe^a74
zH!8}{D4pSn>R9B66Z4(uWlN*#J_`h?mg|+9X{V<(4V7Pf_j>V-I3xZVuoDZSX}mmO
zK^C67fW3o2>H2@q5WJBr)f4UIjnF82jSI6@P1}cB+zHCb3&|S=d~i9Y0TNb_5+0Wa
zxltwwNNFic!<^Yom*jGkMo1aofIx43A+ylt;NakeF?<xZUpS$!jLwBN1my+(#I0xR
zc2dY&InIU?lPGzmCKNEp+(5pDr&&>>e;x!ya{-6NAJGPZE@9Fh+sQU__y^-JVSbO7
zx489uKNRlk2!V_~4EO0-&E}2^da}x8qg&+;u#4<4@vw#{U=bRAky9;&2$$>s0B-Hs
z?3p5a<nKE5&ofczXKzmul|~;<9@;-n3T`@DAe=pEPj8I{KEgR^Kkw=LcRB*|3UW6r
zb~y8jJzAZ^VCR$2iTWK3`o1a0;Rf`K)=11<-Zo%O#f+REjH@Z0nH!rdfIUh2>V&!O
zUxC5k!lH>3H#vW*(KxcOPEWgaH@tvAJVa}A|Egqt<ZdOq-`HP+K&MEUUf{SuKPJll
z{cZCq^v(JgOWXm!=~HVKfjRnG8i!)I7J+ilRwhrjY#r9!FigdSw8i^38U+@bP2`s=
zn-=0mC?2j3_)4P?Ytt+qbUm;N1M?`pd%@tb0-IRIL2}b#^O$j}=%@5}wn`h(&iUQU
z-o`Z~V&>zc#e^@5kfcgTWZrsR_R>_HFnyEEGjDyz^!{Eb{WTI5I@pZy)OI*O@3?El
zk$tEbcNCr2YUvSNpOq1KsI)F18gNhtJ;518-|sq%dp{b+{E0hEZA9JAzmXBth4NKn
zAmU~;lm#TfRhKd>bJ^#xDsyv>^UpS(r{AZvYuM;rzppa9IiX5baURiTHmU*Usqq+Z
zu{q}vndyNV7Lh;us({I}P+7V#7=Hu?%z=&1H+%t;KIWAtjOtl=`AE-8FW7@4Vm|Sf
z^LS>TnK_d(*b!79VGmpjmMq{avc3=W({~DUWvK$Qo~*0f1?T45Sgf@kcl{$LmM*Vs
zyA`Y=^gg0xJujs~fJd4g>w;x*cmu}&N@m`Ll&Y-*9SB^eE~yezGw+Y)R5t~+mH1q!
zTd4FBtvjVF%NKXm43)`hZmjg`CZ%Dh&M*A0nqlFe1G{qn*Pu)G<xj92bcVP4s5udD
z974)vk;~ODy}BK}4!GwwG(V|tl2Klt8GQx@94qs*Y#kub4z~v2_rDwd*i#TFk((f3
z8pvIl=N3!mZV%$7+OeaV)mosFs({N66a;*~#h%s?IA>hgxkVdTB1naPTn8BH(vIs}
z{-4Y<DgdPB9rdC9bt{)W3e(GDEvn<J{Eb1k4$8cp#@?l6Eq?2<s<f{8>BN_W%jZCs
zewww8`EGP@#MS5eDuIeQpk-#HIB853q+VNP{=7yl=S&fu6?2kSndO`13R~9aDl=Iu
zY-%OH8j1A{9lJF{DVFFVg?&rw?bnQuI4v$tQkrxmA5PtYTst-x!P7OzP1GBVU_zoM
zr(7W3W9E<Kd`{$ycM{z58*0&_rD93<a8rV(*c(;3R3^WnY?Pn8_qbClx5!QOc0!(K
zb+a6*eP+zGo<bO#AKx3LJf!F(t9>JpQd(rkP}Q-fdP|+Q;2ax)08J~2?7^lUCI|yh
zcm1SnpG1s*udXLzcHWPLSH}B5lQ+orVDx5>HJ_lYnXm`BM%!%M7P-j1M?;F3>*D__
zUM0D~_eQs%W~ZPpyiZ=3=D}#j%cSuT-w(gFKnhE685CQ(@gbUg#nSTFyN4@-53blU
zhMIhjH15`xo-K;a!LO;zo$7Xsy1TZasCBqN34oz=#jidg1W7{#&LNZ?yW>oUvgiZ*
zdli)IXw>8pFu15k(y0Uo5MgPDwWoF0Gs&fT2=Ad5a==+H7gn-C4SaJLf@QhWO?4j@
zrmimAh}lnyh#$b@DX2X;SoO_jlyxO<8*=PsH~)A1%F(`rFH=VqL`tAg8?U^PMx8Ul
zY(I_@>K^i8_RXDk56_9%vrNwjl9qyf4O;rFOskFYkLJRQZ!B<VqJBq*LZ!EydS;nl
z-iYu3ad3zbH7pgToeiXivm!kkqGR@54DT1J746Etywf3Nwcz<6i5Nj5!>v{k<9$|i
zMwR=9vO&R{t}0>_sjNu+QaFE+=uNJ)s9!(>u7Rsb%I>pP1<Zw8J`IG`R}Us4SSGiX
z&+Xz)LDyRbLoUVywc)0%arp!z9B&98j_MzyXu!3==6aLs-ntR+NjmgvB4+uJV~M$4
zgqRcHV+StZZhEw!^8wYoz!yMPoX|cR1}oufwF)<}M0X9hpi-kPOuFcb)5eDkMNd=7
zu4UMsXbda{)6d@^-;U8?xA}4ODA9vv*mw8vwGTO4Wl1WqJ@8Y>Dtn*<NOA9dQfkY0
zg^hiO4M|nAt@qAOIGQ+Lr!O$@5?zaY_NsUUaQji<^}H5Joe<$Qm3noMWW3w0>pIbF
zC@lmU`Ys+wzn3z?{+Bd+@^XFIaHQ~dN!g4QA~@z}So?-!?B!rX7Blt$)}?2}l%3%v
zLaN)=)M6)9{W!#-qO~|-OZ}0v6YJiZjCd7_`QElwFj5L$<8XcM^F|1dYmFFR0hkY~
z<K;$hSR9K<epGGUv~XFi+bz7J2amQVu-XP~sQ3DCa3|l4YMk)Sizfdt9wWSF>f+ks
zWgQ-33X;Yf;s2`E-u9GfZkk`coA-;8_=R`dv&ZGTom9eHQdZf;5>&Ui4VZ2@lH}O{
zDsKL|ur|b!O&W3Ck!b|(q|7N2HLX{Bd#*)P0g%`P)-8!n#FG=%Y`Y_)cbkcWWEEr8
zo&A!s{u-;YFU>EH6mxpVCS$N$LF$6dj~CMLRqP_v_<cuOu9cR<Za-;)K+=stu3aR*
z`eQWPz^UB;?noQ1Z)!KN*FN+7xs&0Q$Au-NCh3V8zxN*^M7gewg7>hvLZ0*HWu`)*
zoi~3Y7`>Ha-t%SZiL6S52AHyI3U`sGt=4K4ulQ%jw3-C}PDU0;>Pi=iN{rL+xSLzQ
zy>$E-Qg}jv?E-|Q53h}ZY2{=tv1-nJKv;F8u@AaUt3V(n2*H-%?w_#PdBNi?v-|v2
zGnRW#Z=1D(53P>k2Qsqouq$R4yfg{MOQi8wpq8r?!-QRn(r0TWylsKwK_cZ5bug*Y
zm^&O9Q)Oxb20eb%V^5hQlULU4Q2Rp{7seUJOs-W%#<@nQ&|FYBY#SBrCzXcOPkfjg
zXK0a0!s9&0zOr_Cj<P*%jEaf-x6IhNqzB&|7&v;ou%Uss?e#Uai4-Dp#L~o84IoR+
zgV}&U=&LOztj@{`lB0*w^uOpT=l@z2Wz`J`#jH4Wa*ud|KbSp)O|5=yH@iFjTC{>w
zVK;7H$UU3kNNUdt792L`NQUO9#fevD8S{?IXM>&UP7hyxhI@@o<$a_bAF_gJm{sPJ
z6FVCt67jX-w|;usY*@eWY9j>acjAuOW4E<Vdd-f<e3mHngT&WV#-JPufM5sA-g3_r
zibfvJ57ton@ATmMsr2wkg7fD^)g=vJEYQp9xy76PTGtI&S>>XzT<WlbD|+QZE4OkH
zy=9rvpo5H2rS<wOci@U<8a|2*W@JrYCl~KV=Xz$guzQ$<-oQeb1}yRCA?ZX^@$kV<
zax1>7ObS|4yfxoL|ERkSlMGWKPbeStq4m96%E^?FwSi8Lc+Mq3>*s)P<13?A7O|do
zDLP@bysc>uGw<6yboHIeiF)lDwfq>HKWkO?(MQbDvfd~wHl6Bko%wt)TV0#2XaGMp
z?kabsE9TATrce=$Jxz*@YN#ay+>?Q6A|Q0Qdr9kQ&7ZOjDD$Q{Gdk>QV6ip|1)7Q9
z$Vhmz)#uQZo+t_R18E2MjOdjP3vP|L`hM^v0zClUr0z7A>dCFG^|go&;H<CaIwV(?
zX5h2Ks7jh4Kq<5hUF7YBNN%h2??-yj9zL#T_CKh13iuIpV&rRAHR?NBWPRpShEYx7
zwHi%}RpG3&pP_G!=#gTl^d!qt`vh-!GQU5BWtOFOgeJOgb$zv`?iu+#^u=BG29WcU
zjr^q|xA@q8A;m^K4(9jL97u+)8DzYhB$Q>zR1_FG6-GPtPo<bj55wY&OkOKnYB&*z
zZjz+Z1;;Iy1ovU0@*=nXsYxcjcDD&LdZ-gh3>N>wGLe!e9bmrYU50a^E7zlr;V_%N
zD)BW$!JGve{_cNO1b+&HznZgW*2?U3hm<qU+YDos-aiW`2tUe+C*dkS4bEEDs}J!j
z$5@s!@G>($sUvqTLExaiD-!`RcPLj_Tq{=3{H*tMQpNnfqQROyP>eQmEt;N2WC%>9
z_J&~(W}>0zsiaLss>xg=vRgQF8&<cwK-yhwU9AAG-UD6uwOF!t!=3KYKOYlE?h@1f
zm}TfACFh?_Df=$nLFX}sl#BB>@@<hc9e+lXm3F#RYGnHqc(2W}{zX4*(5-NC|7`-R
zq}m9pmNH+`8+{fvY-~>m8@(p2^q%JSMlUJvs&?+0iQnmB=-FHG9YS7rGB}0FGu&hh
zGT0*%%}IE~lvUFc^khj<N)=mIJHfPiHtWZ8-`6<v$W3*;ZD6pE@;1L@6ROhY<@Sc{
zLk<RT?P;Gr@yzX=)a(hj3~!v3c7hc^p<~C|!FQrBN1nRKeMvq<BS(^KZR%&Png+Y`
z-HmsFd2NIHsXbk~+H$#?V)I<Ig*0^fU_vvcPKTU|cg2zr15k{|!<!xrFv8)BjU8u?
zI}-8d?dH&^x{3EhXh#{{LzbRaw*O5LBr95ZuE8kV(S7X_l5*8Vis|$Q6&;sUZ)t(T
zbW7>@oVzHSVexRN<-Q~2DRm|qHU1P|yVInT-fIDu<1<UgQHE=^J*!Cow|pM%m(SXh
zH7&i^A05;)a4Y-UAGHfISM%N3?9v64x_RcpJJyXN0wT>lEsDevn=j|l8I@{sZWoGO
zBrHD;(q~k~?LGwJa>G{NMSQM*ENbf4`WIdaaPggWSw9dpEiUS7D@BqJZ=efZ55nL;
zODBkzH;Q{VNyQXjovJ%%uT!v)TQ?pXSXvrckwkQ-ulHHj;)+)dxg))Hv>3Mr--HjP
zD&S7ZazIqSiq0eU*AC|Kr=%S(vJrWty6pyScXxrcSQWSgNy06T){}|mu?`rFgXre(
zhoDPznMv#-vx&-X|AsVG;wK~Fa~-&keYLCmA?6ty%jST-WI<4A_9%{{CrSlQv6;S-
z_!lXw65IRDC49>%p~l?&TXR4&?39}rj<?V9or`rMsEdy9&pUc(ss>U^)NQQrw>n<Y
zG1H6l<`q$a>OB{q7z(=MIV<@kY1<M`6+z|FaPL!>?zjY%VfS6RS_LQ;w|qo_i}3ez
zkgf50{=DTLO1wWrf6KYq3r_n<<uz!OMq|WL`Z4-^=i4ztBeE#-vC}JI<(?q9_N(?H
z=Iffh`eoNCBLmn})3D&_las3zPXYrq78rBgn1v3f!wd0ots6MirYp%r|JW$sLBeEe
z7?Xg{r7+tO<cx1+y&?>}Dnph};$<xPe{=*sSw(^J`PSJn5#6VckTrGhQ23|*?29BC
z!tZe=S)0JTX-iovyV>iDHtW7j%kPt;ym|rydcEtbVSaBQ3nG6~;%XvFKJCq3_`Xxy
zfEF7sO>IjA3GKK*Dqj=#M<s~Q0=wgsjR6e4+>rnfmktBnR;zA8#Sh@?TV4uQ^~=MO
zg$I2&lV#3MvaZ*^5bE_gO#bKhe)rcJSbXhC)LgyENyP+wFNoPtSx>d#R$k(oRS9xD
z+E3LjT&@s1oY@8}QoEqXyIZT`33d{Zq1Kcu{m~*#QFHC}y>|0$xZ00}U5xuAZ~5fW
e{|6U^O+Qpov+E8&+y~4KI(+B|tn`;tzy1^S?}<qO

diff --git a/docs_old/assets/images/logo.svg b/docs_old/assets/images/logo.svg
deleted file mode 100644
index 440b3f0dc..000000000
--- a/docs_old/assets/images/logo.svg
+++ /dev/null
@@ -1,134 +0,0 @@
-<?xml version="1.0" encoding="UTF-8" standalone="no"?>
-<svg
-   width="397.00375"
-   height="113.55765"
-   viewBox="0 0 397.00375 113.55765"
-   fill="none"
-   version="1.1"
-   id="svg314"
-   sodipodi:docname="logo.svg"
-   xml:space="preserve"
-   inkscape:version="1.3.2 (091e20ef0f, 2023-11-25, custom)"
-   inkscape:export-filename="../../../../../Downloads/logo-square-side.png"
-   inkscape:export-xdpi="270.25565"
-   inkscape:export-ydpi="270.25565"
-   xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
-   xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
-   xmlns="http://www.w3.org/2000/svg"
-   xmlns:svg="http://www.w3.org/2000/svg"><defs
-     id="defs314" /><sodipodi:namedview
-     id="namedview314"
-     pagecolor="#ffffff"
-     bordercolor="#000000"
-     borderopacity="0.25"
-     inkscape:showpageshadow="2"
-     inkscape:pageopacity="0.0"
-     inkscape:pagecheckerboard="0"
-     inkscape:deskcolor="#d1d1d1"
-     showgrid="false"
-     showguides="true"
-     inkscape:zoom="1.0083333"
-     inkscape:cx="280.66117"
-     inkscape:cy="25.785125"
-     inkscape:window-width="3440"
-     inkscape:window-height="1416"
-     inkscape:window-x="0"
-     inkscape:window-y="0"
-     inkscape:window-maximized="1"
-     inkscape:current-layer="svg314"><sodipodi:guide
-       position="-146.75933,121.72017"
-       orientation="0,-1"
-       id="guide314"
-       inkscape:locked="false" /><sodipodi:guide
-       position="275.80377,60.668961"
-       orientation="1,0"
-       id="guide315"
-       inkscape:locked="false" /></sodipodi:namedview><text
-     xml:space="preserve"
-     style="font-size:117.641px;font-family:'PP Neue Montreal Mono';-inkscape-font-specification:'PP Neue Montreal Mono';opacity:0.99;fill:#e35a26;fill-opacity:1;stroke-width:0.612713"
-     x="115.73952"
-     y="91.977005"
-     id="text314"><tspan
-       sodipodi:role="line"
-       id="tspan314"
-       x="115.73952"
-       y="91.977005"
-       style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-family:'PP NeueBit';-inkscape-font-specification:'PP NeueBit Bold';fill:#e35a26;fill-opacity:1;stroke-width:0.612713">outlines</tspan></text><g
-     id="g13"
-     transform="matrix(0.16222521,0,0,0.16222521,-221.77667,-24.740002)"
-     style="fill:#e35a26;fill-opacity:1"><g
-       id="g314"
-       transform="translate(1367.0913,152.50405)"
-       style="fill:#e35a26;fill-opacity:1"><rect
-         x="28"
-         y="644"
-         width="28"
-         height="28"
-         id="rect1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 28,196 H 168 V 168 H 28 V 140 H 0 v 504 h 28 z"
-         id="path1-7"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="28"
-         y="112"
-         width="28"
-         height="28"
-         id="rect2"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,560 v 28 h 420 v -28 z"
-         id="path2-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,476 v 28 h 420 v -28 z"
-         id="path3-4"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,392 v 28 h 420 v -28 z"
-         id="path4-0"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 84,308 v 28 h 420 v -28 z"
-         id="path5-8"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,168 v 28 h 140 v -28 z"
-         id="path6-5"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 336,196 v -28 h -56 v 28 z"
-         id="path7-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="M 336,112 V 84 h -56 v 28 z"
-         id="path8-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 364,84 v 28 H 504 V 84 Z"
-         id="path9-6"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 56,672 v 28 h 476 v -28 z"
-         id="path10-2"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="56"
-         y="84"
-         width="28"
-         height="28"
-         id="rect10"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="112"
-         y="28"
-         width="28"
-         height="28"
-         id="rect11"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="84"
-         y="56"
-         width="28"
-         height="28"
-         id="rect12"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 168,56 v 112 h 28 V 28 H 560 V 0 H 140 v 28 h 28 z"
-         id="path12-1"
-         style="fill:#e35a26;fill-opacity:1" /><path
-         d="m 560,28 v 616 h 28 V 28 Z"
-         id="path13-9"
-         style="fill:#e35a26;fill-opacity:1" /><rect
-         x="532"
-         y="644"
-         width="28"
-         height="28"
-         id="rect13"
-         style="fill:#e35a26;fill-opacity:1" /></g></g></svg>
diff --git a/docs_old/assets/images/normal_computing.jpg b/docs_old/assets/images/normal_computing.jpg
deleted file mode 100644
index a71483a68ebea1b02a8b3b167fa0efe57b498f54..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 64918
zcmeFa1z1(v*EhOpknS!erMtUBx}_ULy1PSCKtj4xMM031?k)*wq&pN4;jRsyb3Etx
ze?PzbzR$hibFb6=?7j9_GsgVQF@9sr<y!3Psp~}$nyjRZBnScm1d;~+fv#skVj$?7
z0tXEZcLxp*?hf)Dpdh0mg9{oK+D*awUhr-T5x9{4S-_n@z`(%3!@;A#!J&}hVBnDb
zZ<p&<5IQ_u09*hR#9a_1Is_Ct#C1D}2m}Fwgn|I;`$vI+gMx*Jh6JA_1mxhef1Cn7
zp<yAQU?AbHXFy0$z)3VHG!O_PsAILG?}8|hzdw*)`d<6*xBmqO9nHrDn)my)9mL-R
zk^KPf|KuTr4T5@IC62DLjbZoyWWPW5^$=3sw!9-5>meTNkK+HyW1Qw^56f80I&eCX
z{QoEW{h=?I3QCPFU@;_K|6YtAVEM`6@5T5DC_feb_Kcsx@^9g9n)){;KbQQbsh{KW
zZ|QHE`Zp#&m;9!wpX2f`>2I3)7bO2d@SCRo12F#*|E8&bLGm92ziH|}0P|1rZ<_ii
zBLC_8Z<_i~$oy0Oo2LGW$bb6&o2LE~GXIeOf0w3E6`#+2!Fl5XDTSCkC(PqB4wZQL
zv1;6R(LD*k0E$YWk~e!!_?h0PFG?uN=V`%_rbu4E2||6Bt<=45qf$5_(+r*SW^Oz4
zB+HvAT!jugA|Lnp&b#!t&Ia^Qj32%&_Z>Gpqz9UyDtIXjB(wk>2}pu_GjIX@@S$^b
z>WkZ??rfX)(o8tZe{2Fg3_i<l>ICc-bm=}*oxkJPVmPKV&$^gHE@AYq!a$E8sM*7@
zueIL&Td?yM-)r1odc7S!cW_JwJ}>0+mE-h%bl+aKO%U)#^qzE*X#G{ri)dnd@M+J>
z<jiihm14rPhWD{09Q=>hU*oicPyAJW;voz(Pv&O(tGYy!K&1TpBRW?vW4W5Hc)lo^
z0vDF6t-UPX&B61kq4AVr*kY!hpdG&hfgad=5szVf@eX5aX<G^C9Em4HUF}r=Jz@R8
zrh9mjeOCnta;5SS9aoMNa>dQyr~{-P9&`QZa{@!dFXF@$)j)*iMb%!>!$AciAlRa6
zUytctgcMu5VJMPjpaEoO-CLi$97*H~VaTMH^qwmj?#NEy6C~?tW$s-GMZ%~$!Wu~i
zVeOfErd<O~H`fiBcv+7Zx);Q^yi{j7gfKjIob13&{whBSIgoS3FXkHh<eh0PdFzFv
z7q;4wPGayZI;7N3HMvCJw!wdE27zFwOLj9?oXIwogx88?-YHlNiw!P<bTa7*)txCP
z{qL4OaIz)lP3R(QH1D*0;Z)=Iz}K_d^`*C0<ivt*ht>Y_v|?=IrT@M2<NOupliCn!
zeAqR-6A&(OnznZ19qz{g=TE4gpPYJ@^uAENDCx~BO)upw+fNMB;t8_;LUL+duW`j~
zp>tn7R<~xQmbS@dm0$j^g(0XB&Kfq<<4v2;)jGa!v?rjg?0gMkX|2A5|CSBR9n(JP
zbGK{k=18uPkc*8?akWmXFU(LLYsMW323@VwF5Uxzb$_5<C7oc*T+dYSg|h{Vb1auw
z)#EvsK_C1#Xwu6{br1H%o%V9d@_~Ag$9YTwm(eE*MQ;TvRS&5%yErca@ek#{vaw|4
z2_KIB;m2RO#Ezl|wgeKXGzelo9#<wzVtWlX!DTFhplrrJ4>{DKITw>TsZo5SX4@cN
zw?~fE-<WHd{um~j2qoDi*|~SY-_9ouT=Rygn!kL{_5SJTE7ZEj!fes1nHNn-k%lj!
z<+wq*u8+;kPiU5^p-DMTia6S;f^81~{bx$_S6qTrXL+%dPG1f6&8ZHJ)L;0AS^fnn
zF>Bz~pl+CD+Qy*b#^u(Oc-?YHQ&ol@0hZHVGcZvf=TEf0edW-l7xA?`fn07wc4`82
zWqF*`YF5_ET`ze)_<!IBsv!sE)6sFdaC03w&%iDelYKPh4NAC({77n_tlQB~EF&+q
zJK2g2_79kKNqw(#!i@|?YcKPnWkrFQ$7g`}hf)n)I;S1UNK?%Wc;#(*_!hgqCe<n?
zgkeyRP+?<7LokaKxxUDB<I#Ga6mp!m?uT(+*q9C?vf1XJTQEIhoBot1yWunpl)w-{
z9l;uAmMg3<&^eU-`*YEwXL<|4b)&ozvAQKsT3>n8EFL!?#tf8~&xa6kHV3UWC9b`5
zjf<WK>O(d+@C@Zqjf+a*%jvPtXBL=~={7%po;kh8C5rM4M%2K?HnTv?Orz<Y>&pV8
zfzj+H!7gSg9TyAE3ld6TxS5WzM&rJ*^E1yeJ(V=U*MkkRCv$`zGE2}glt0?2;1!pj
zj_oUt(ety8bY)MeqJjPk$cr4C=(v{wf+>@(=A4e6V2%O-{PW+%RDVh*Il}vlr>0|H
zT*fA7Y>4_*jug_e|5%+xnxSq5vVaok&ly6aSO=#CulYdp@7WN1E7>D{$EzmH!RTE<
zcZSP6Hh!~Qm=*G8Ti2jD6N}BuOSV@k>w_;oy>s<rjyC*@Dap=X74W<y&cEm8544%C
zyPs5Np$6qgEUB+MKw5)A{_w9MVT$(9_1%ui)_foj7tZ%{s7Kq}SP7r=wq-6)xv*gL
zgFy6kV|g`>uH=I0ff+?_`^({Yzt~S-n6HoIhIXsHbujrrwB7tD6Bwt-N<Sl^xtmIT
zMq^vTdSfFf%T>CI#It@|W!BSA5LIWZF1GMI3Az5jatBcBze%zPXMTJh=}v;dFi)Kw
z!$kYaAB(E{XwHz5P&g(SfAIKF(b@c0V1Rm_L$$<{C4$PAd<r!d&#e;pSpEDfYOL<|
z5|5u4TsELQ2!yCJlg)HKzPQliv@fFS$$Od-Wd9!^B{k4geCLJJ>TJ2vdi+ue6YcWr
zyGNOS5e`#UPOp&QH3*Jw<7M(%q@o|y9A+splwaGr-~C2k1+z<h5OP@F_N`Al9Al4&
zi&a6}UV~&5-37i0Wx~C1{sC7A^N*gN+rVh+>l@T=(r9GI&Ovt6oj)q3nqZ9};u1W;
z_9`lo4##EwZ`&#+1LnE~YmCKkvw}$FO6k$~=Pt=`l9fm4n-k+Fw%-iNXZ|vrA8hqv
zYw+Qgl}q+BJ*%&wA<KGrJT=WXkb<Z@4U#V6^$aOBCtULMtg~cMd=~%m{g8Eq0;|oV
zu@aqi$FUMA2Hr%+4-3jD4zrBZ?}jTCIrlUR3M~oNW)jty+U01hH~&WI{QMuOnKoqb
zRu(kAZ4$nh^(Uzs<9H?I6onXWwGHZt^*Alu{6fr>72av}m}z@G%@38Yu8d{kycz2U
zY&nTNSEgYHhzXdZyGq5BF3#6{KrB@-$6|I~9yP@qDw%dNnVQE_*w-M|D-yt&5C@|g
zz8rCl@MeyG9#Cx0^dg!CBEb*kXZdiAHJzIs>lpDJ9V#8_!9g>kp&Wl${n)j}GB>82
ztgOn-SfT78*7k*QRNg&;_38mdn=gbpaI%4L9HI|eodg3(+&nLmpLv$*SyF1Me+$Ul
zP8ExB*|{AQ)j;JKPVpv$G*oQsIh8yeYzT0ee)}wKI4^VxT!$U`>dgn<il|Gcy`@-T
zAedCD&I)vHh>T_j3@9borw?U^Wz4?#(ksBl_?n7R^W&$Hx79H(n8`l+8zdGf$h`U%
z@hn8s>R2JBojJ%3svJ&p9i%eR5v%L;3gV?Z2qK}V<l(ZJF2w=1PtG7O@61M!0|?fJ
ziZ^S_KudrZ<T+C8^hMK&LXO=q(g*~?&A|8AU$x715h&Sxo)y9WnWttEtmK~satfo&
z#hm_K?o><McOiuBxW?YRSs^c0m3)<AMtUz#mKX@cm`0jZr}9`*TzZA3Z>Z&^9g{$&
zpv$MX)|FRP&f%NGpg$9=b=9LT$(}_UV>7)MJDGlbJbI_c{TFz*y%RD?jJlc5giEj7
ze`p;wHa;ATnDSRTb+~loq#skGSQ}85H@0U;EOgq}R(f>g{py7t*SXvbVXgjIM!+7&
zV_oO4!OUcu@{3mDeE#mA2FXt~Dso@L#+ZiYpP{dwzRpx{YqV^bJ@Z(woV~be0eiPj
zmcSJDC1Q%z7Q`e}*|9(EUnu}w<J$IIli-Cj=(G?qjaN=RdZFNhn4&_m9^}mnO8WhF
z8HOU<iS4ISLQh29K^vlNVDTF^`iA`O`CyjpD%y-+%B;U0c>gERHwQntAb{waTjP4n
zz%pErYnJQ1v!;9l)qIua3jxDXH`{0=0qa+^jIY+Gk{-)Ve3Gbe2_!KoIj$JW{4;Xt
zslEJCli?W?LV@Bs$#~x_!vopLA3PjHYGi5iktcW-IRgSP{SG!p7~L>k=93zBqA`Sq
zBF6@@PP%;3lKa3uU`pN~t%&}ld@DG7q5Cr1?wg&aSbCS>-cQZ9*%@a}Qs3u}@Z<Fl
zRk89;{B*`TRIL-E^<lDsN1ofU7_|M=oR>%e|AZCC#blK5oDh_uc!LHAD*d!R>G`_8
znC>D--`Q;a{nMt1@y|uo*1i<SQ)>%1len!3ydxUQzL-dPvm9x1lj?xwF!jn0>phZE
zHdIDfM+9A+Xx4m?fTTvTfXBG+X?v~s%yS$oa1u0I!t<d%W?=1QLbCP+;W^O+0vAI6
z1-!?{lk+GSsoZz3FVxV!koNRvJkXNb8EeIUcQdFq0j{jwigd&T`k99F>JBdN3BlT5
zu8!vYVKn^%$|!n$GoYJBY1S9id@0#*IQG`5)2^HJXqH2u@}q4q@qPHe287km{L~jC
zg)x9>?U(*~g6IQI3!KAP^5DyuJ?gRab=PC(hj`*|KyX?puLeN{WsxNB*7c!u4?t6O
zyj5@3RfGH}I#kb|*_Pr)a31C0{P;SSn&%Wj)oI{wn@D80y@~@u2ZYB+a^{uf1-0r-
z=5)zf!jMUt@6F8qd}9c>U!!5%_k<w|{|_0?u7vIGnvO}1oOqx3`E-9?>Io3%;Pw|c
zGP2U@IDMGN=K8w^1t`m{Hjm$aP}Woan%115^r!PLQr2lcV<;h_(|S<a<QoD}<RXA@
zXSO8#TQ)EN>H8X~cbRjm8!Awfg^{&RwlP)G0Db)xp#GMe`gq5b4!qpaQ_O^Q`Jh$=
zL>fYSAd~qStdPDCS;U5uuC`f2ACHH}QDUmohA$6CF6#IYo;+5`DQ|3Ie0;_)145}F
zHb{q%)5H;Q7;J#8U1T+x_pz#LERtJaO1C@_Jk34+zE^@Cl5-4`=zY19t7nzBkJqsG
zNvfu!{{qDN-`hkbMI=QWSEfp_Z5RS@Fw`Y5y=&A6dA>0MCFni>G|uGDw8`JJ(<Il`
z*I*-J{5;@&d{>;BdYFcU<et;r@zFfDMN$JnDv)P&mX32znu*s!$3*q0y1l?7Y`+LJ
zm+7GfU^fQwXH5E=p@Q`JzDd}g&_dK?#U}B-%DCrtnu9X@S0eSp&JCnhV?ghG*;|y`
zRPpef137Eo<dOLN1}fpn666*L<l?_|xp+tC-Rtcbmqm^t2%pgmKC1VhsWeCw_1u;v
zSi;v90>BXk=dsznJtoB_-#Rri+UFl~YbRma*-!m{ba%<~xqiBwJJWJ61Ag)<%yDrJ
zB54M?0?IFkD<JOHWY>!BJkx&-UDplzYeNd6$kRR^`HD)X%XYt7?D(@38i=_|?)4t#
z2eM8-Aw`{Li1Vd_(}Af{QRH^OVnB?1oNp>UvFv9dQJ-I^K3|#<<#GoUL7x1&=9`kH
zhCsx@;QWENj=gw&&i`lo3J}b+%+6ObXX%ZR=EyZU<H8_YWa2o`KLa~XNDGtzI4UW4
zH}J$T1RaQ1`hzD-3{^9#c|+GcWd+ua#GX+i@BWH2i{L6CkkdTCnclMp?(w(ZK|hZ(
zbDYBOvE$8j1|mlJwWig=afQLvz4*;}1I4nw@Qi=u&=#<)RF8YHBya_fE8N)(O!N74
zHlfFy!Pf!8kD18t#$5#(hO|zLp^oTC3ZCPnzVsP-I3Z~H?UhAT>@1j;0klGAsm6CE
zOuwKytko-&M;js&`-6ZBV4iu;UI<)1bkU|bu$gmG?}~l~3UCH??oNVL&UJ4TinL3+
zH4ZYbLC}cVCWau0QPpvh1>P+4D-DOHs*TrR=`aXl@=nii-gIfV$Ol5nS9fMeD61dX
zIea0w25HE=MdZ*%ed244LtMcDY@&c7peNRwq#MJ~TOQBNX`>(IS>#8NXQ&57lAM8W
zxVtvD)x%h6UPk_(UisEe%)_?i>)_P}m!Di5xjpg2JIR768}Hhoc*;O<XY-%?{U0j)
z>ful9{8yX)a=o8infT?lzgY1n@#Pm={&KaSqy@j+_7^MuUW{J^>SsCHuQvVFdcPOr
zSCRVd8NZ6uZ<_j5q<+VYUqtFRP5mNJzhlO)0tNa_Q@;w;Z<_i=q<+)XF9P+OrhXNu
z-!%2hNd2a%Uxey6P5mNJziH}Mf%<=rrow-?_2-wh`^Adjln{8@oEZ&xZXN~>8vI=O
z?L+6#kRae`b1YaiGG<nGdkjn*4s<qQC2SV*yA<~ffk)8cfk)FJU?8>4TqY7H%s8|k
zXsD|HKKKp7kCrg^r0>HZ;_ui0HUbw<Vd*vUX-n={i$0Gl1Rt1DO8Yvh6qMe6vldO=
z^EAjn!IFLzayO;J($!2pjgv5S(pPuw8U)2)F}>+E?P}^)CC4uJoW0Np;{Bv2S?81S
z5DxMO(yA(0KRaOSib+2ykW>FhjUlJeYf$UcaMd4l0Y<l97W4H=|I+k_I$$m74sxO0
zV{Wxh6^jeL22pxmgVa+-o(#)<jXTcw9fR)kOQA>~oq8mQg(J=`svvLQLHVT8$_oSk
zE*)$CH~ikvkScBYIyR|FrTGpyBveMJU6vX2dWkD|=4T}A+*To|*adSI-IDtmPi#17
zr$4(Hc^^;wIZVW!(;NrPVnW)z&nZ+7FSDB|J_S)>@t@|PtN3I(O<U@6m}zI_9TMNg
zOQO(PYVUsfRo*G5ROAxJ-gb$5mvLKAn9Gn}BhI#%=N}!%m|`g4%cx+d-gca-`_zn`
z+kP?cAcK+U1P{yJcy$`+H<2EboCI3w?h4Rvg#UQRdQWEo95J9TGQ?_JVl6_uJT3-r
z{0SZJ;0{x!=lj)H23wT`OwR36&jN@0u1LRk9>JehMr167Kk@=k+HA&h-(m(oXttY*
zK26%`IdP$ra7N!sW#%V7gnJ^4=!+7Nj1s9A%q}@~!*iClRa`oRFOVlgx5ap5d_@Aq
zIHBFqVWXc&?J@%QHV`b!4@BZ1sda_;<P(148C@#vZ<g;5OXtWBPZF&go=kShFz?)o
zc8EPg&R0B3T@WO!aCOO;GT=Y;>(PHdVa~3{m4;JXq7ql)P{RFanVdz*-5NT*mb9vc
zbzvo5wtl}=IWIJb0fHq}R%Ic|2?2d^22UA($T7qJwGlq_YhyT4;?xVKR2x>KBsD6M
z;WUK6v~XKGaa>(`6(K8^9Ktsy+O1we^ICf}W;~m|kJAX3a0g`yAqs-j(_8=yWkU;K
zH*@REc4;)_d(Y427Awp5TF=+Zu4GTI%2;x@rKww=t9i9h4mr(vFWwn>vwCA}PqZLo
z$!TRV1Fd+3ALFoVZTBS)nE2;J1nSabT6x6~?|EfC>Hmh?UyUE{(VJL~ZdAz5T`_To
z3ZJy9&Qr)6HZeig72#kmsFG;CWEo{ivS|eKBW?NcH3+_Lbg$CPL_N*>+y`S4-RI3K
z9o;H)n3b-9%6i20%va}hA1=x}BV$O4<;pMH>QSN3P}*%uZopb{$}H-Gw7<KUK469M
zrEz}%dLvtsC|p9M#S&syqQl%ba_u3~G|ww$dCYFT+898US{cVN0>Z@(D{T%V13|(j
z6w`3q^ADX*QkZkw)hweK1LU!R<x4Z~+zjz4jJyW*F6Dj+YZ^r!A?5s=sig@QrLg0b
z1z@qp8Gd;R0o_939@y0bSSu#O6`>%hdpahj3TOo9UD;4Vzdn{XOiX-Nc!-!jA=`-o
z>y-I)_Z;kERmnAIB2;RZDXV2yo?swhgyfW`*TJMg#c>tNRTO|{7C#_J$?S8oOG;v+
zH33v8$yX7=I-gPaxoMwchf<Nh!?VlI6#hYm3M<$!vv14LwEJ5jLeez|sa!|6;=#ZR
zrh{(eVnUc)+$xRlD)f?$v9?$>U9t*U`4om+g9frCl@4=WuqXp2y`(KBqKrPA*d}r8
zXdUy253zh3aA{UyF7NHhB69THCiX|=AAJLv<t6jkSpNn~RruDObNo}W#q<D4N%7bj
z!&JKi*{%{k=#5e(Kdh=!HfDvc{l0_gtlaaYHco}ik|ad&ro4mhZw1FwK0S{o-7`0m
zGdZ+UHEe8)KlDfBqinOGE@J}_q>>X>ARjODf#xaPqU5rUws&b3PW%66AnOrZN&4h;
zDE#o<cd^&ugQOyn?Tt-tmjsaw?rMqwow!YiDE)~)lZtMBXXqJAh4!r;1I=615F=Z1
zZTCJWr^hV1$;cD}8e5yZF%we>^2w(zr5wd_?z5|^FMplQACHkt&NTz+)wvk|R0(D+
zR*mW6?|Q7ooF^~F$@3{Q>wtL2Vx)GN<!jC;nCl&PpgZ=Tp6&G8owx9Z%%&Pv3P~I@
zgqozE3ThF<Cl3zT=Ndo#DmTBMEU<uDr~g1J8KCQ2Q`yc^-c#Cc8RdByT-mV^rl!%o
zH#<{v%68G%%;e%ax$P|$2)hY^t{gn)u7Brn@0JC!#cR+5TETaZ=GSS*nDr63Zv%%v
zg=<^GSw{AIico6tKA1wOc5fVujV7~-K3QyZl?V<-Ncdp-dr{#IctCv!g-xn$PQ}(&
z;lUkYBzMB1V+NwfV9(D5{9j?u-7~Oq({_Aaa$~Q!@P~=~rnG$$pTlhzQ~#lj$5$fO
z6*UAJ9b>BVY({R2+D%`;9|6XjU-xbtJaajoaPIS?sMUirtm?T2eF_9~iSC!6shGL&
z-M_!>EJdh_CH5Ove72^Wl5x-BJvP4wnNWy66mbJK@qtS#+M$eWdj5iC=SFM<t>-v9
zE;8i#roi#5u-vb5b-mA!)m#y#IpCJ#(C1;#D#i$~w5)?L@Bht`g-o6dk%kUK43`5Q
z5@`0m32rWML<qR4d6gKrEyA$s<u#}a(2+Ed!cyru|7Qc3Csi<=7A0H0w^qL&m#}c>
zJbS{(I1YK9^BCSfPhem5a7J%ZJ&mvs`BNA8yVs%i5g*lJm6&T}@82wWXlhw>0@?8x
z!~R1@X!!5|x)FNdjTwDu{(7Jz<sEQN=5xlt*aee%!B(`|MQ~@3<?+=q#SKk+2WPfS
zQRS&*go<++67JFpA#?+Fn^>@|G)%<&Wk%G~XeIQmXawwilh!6z4q<$azuOG3N{&Gr
zGA%ri72X)P6DPbdHmpK8Pxcia8`npazXn}S(GQJeST#jWe_y{wCZsSdj?YoH@9?1g
zk0PAXlMR@u${Ql(>U!dk7r3CupXa@y-xL^O@@mpG+{Y?U<Ev=$0Pf<CZ}cnew`}+p
zBOYip9}?&Z!|t{E>MV5;APo^sN91E?H_G?%3NfpFFh!b|i-5cAd&O_P)+yI0!!GTx
z1|K+x3B*@1L1wq8ia+aHuJYpKP3bW*GLyb#KbYs#S4^=8EQ{@;IpR$N(MR>`;$sjA
zfW@++*Mk;VsS@c^@ri>Dbd;O06D(YDhXf3cG+&)+N8@I_?he2==|D@wg*=iHQG)k>
zou@~qDc=SpU4g2Y^^8sTiE%JLp1ePtey1m~Sf$Jx({dYZ!lm$GLaR)3g85Zdy*_^8
zwWD;VHvO&+2znIv>OcqLE42X~NebaEg23p}W5#VVEmVr&g6;?F{t<!vmKujS4?9Eb
z2(TOecIc9h&(y+7P)z?X?BrvN<Vc|}3>Jc%VLs=9vD1FnD7u&~ydy6}oDndK3f;-%
zy>si{y<zC~$+XYkts15Wh)eR^#?W|7739w`s9J6ikokQp9rExjckk27r^(fISDLs;
zTqYQmD0KhBJU_`$VVf+bLx?|UmfR2V20$5-+9eJgMV>%B6cf#D4?tiho44^avcDyI
zMkLFi4y<4tfn#Os)~?&ckJCG|CpinT0ejLBS2=lx)f284u+s>`@f=?{X)ax)$bZ_1
zk{LZKS%<j0W+s6GzNH%F+q8S7$j1y<@GRJ{^fVFCC}#&M;&=xd{040nVtZ08RtTFx
zp|=T1l4aI;3rm2!E$!`$0p~L5ltE*967x}`lIASHdkQen7mV9P#nyT6X_<A!VJJ5=
z`2MMYURH|hvSAgOcWpt3a?*OOLY>+BHVe5u;TB8bZN<d^SaNg;d`RK<-`KlL+n8@>
z&Z<dR;fpATn4b=0W&n$HnD=Z2c@GG%!B%c=ChM_L#nIQx)peP7ZUlIauiopKoL8tm
za;Zsp(U9?Yb=!{mp$G5m--Pt(esPh3zFdYo|EAX(fj#fybs~-}S&h^<S?Hw7b+_y}
z)`cbL0$3HCl<-C%rRPfG$E#h~%^&UMcBd{JL;Rkx#dv)@d$cFkG%C7<iFFgT59JW@
zInQUCs2)?gwZGYxR#ejk6m1VKu_n1A72tiD+0FCjh*rKuh_C_YUr3o&O+d=3Ju5sf
zM&Y)+Cv}(pjl6#r!0u0m`h`}YEtyx^Qe|`f%e#2CymKEYhSH0RggQK8PuI8!K?8~H
zdvdRyx9;^Fe)k)q?C06GSzc7nJlAv*kfA%3S&o2o1k2y{@Erq$z=~4=v|51Y=(GHA
zCK5rqf3Di&%oOi>NOIB8uWJu4Wq_1hw}ra7IlsUvMf*!_VhFkRU64#^vMd?Y6A!jx
zeZD{Pvkc++{8}p=17vs88(NJS#GIwtjCv}Cxq@hvJr*<RpV{q^l<e|E6ku7f)6nVU
zGVu;P9V3LvC(Qk~8rC?)a-jBfP97jB!^`i$V3us<MZplot4c6W94WA9wy53nIautB
z8m3@_?`O!wK%}G<s~B9H!{@vOrIJNxKhO^GEenvNfH!3J#-&JF%9sQ=_)8pc<~n+l
z=!W3TgpzB9mCtxRf_Mc3FwiDj4kd0Nw?FEX2|62ou_9U%eP^DX&i#LNC=V-b?i;fG
zO~XWDea9((eFQVc?0q-ZyOs_}6VRFacrZgI=)T=yHmf)`KAY7-r~1uqgv}J4mx`N#
zUNFAmd-i#EFySL1L_W8&{+Gsc0Ibk~=m|{dohLy+1dV&{f0I8VXH5CZtXF+Qyh-wh
zrq;Wuh|jbNA6}+77>nrE`Jr$vk)=T9n(;m{*Sj6Y?2u1#u_~@E<U14*BJ4$mRr%`x
z2>6EM0s2NLSelLdev6gCuDvZ*lIIG9d=-}oLsl2K->H3wgts$_7M>9nA6N`}+7ct?
zgo=oPNCGVARLkL{7;$6G<XbZTAhdMw;bWNh46dslkf;NB+&Sy0bzA*6miCF{G%GhP
z%6sYn>m?kb9yC8GMvtp%mjLoXy19c(95<tL>ddhdCWEcS0dz*i1YtWW|M>?H;Z@8g
zsArU_ntii9*<61}8=ga9<F{izXK!vHfGJQF&6LZn;y$>$9Byx~!;u3^h;A7bK4Kyf
zRX(jYNz_F8N3$nYXwrn-4@FQwK+ixQ9_+B{`NUSO5R6FgZ$`V7-DeA@rc&p)29fOF
zNF%FgV>mVc=me<qKvBGv6G5#LsqwbKktOo#FoEI2LApZ-LFwFccGw#UVNW)TgHabX
z^cyFn@5B8e|8XXO7XOtP??~6kF9*1mamgz~j<gxEXLlr3Im9Dh{1msn?0@^(&hHCi
zU<62ki_ZM$z!RV_P>4`az@I#Y0so~_2uLUpGz=OhI*T#}HW@h^yNHUR!)s>N0Ia(?
z_k=}_aF$>x#FR{)j@(>;1};<!K|uGJ4amRaU|k&V3sPl7{(bm=9|zMQm^nvWmu|fe
z<Zikl|L-4`UL1;$t3gzr*Afb)Jq14S2R2ABiwPdlK42pS*<9I54y~=44kRjZ#+RWp
z_bS<#t&}9Ih&-YtaOapwF-C>D><IX_D3tR0?hxgbe3ZN2Uh*|4YVY#NSC%oaD12-Q
zGKPVyf$r(ckI{eoh|&U{q8Fl1d<^CFmIaHpjV8$UNw2DWSMzr@z#r+x3rYoSFqu)3
z_fD!oFu4qW^)OhbA&o^M7<8$*IbNvr_PhEyhe9RHaV-|vS0{{{;~=mYA8$HQDZi(|
z&j#1?wa@qhi8J}DEQP}Qum{J6zHQ5vHhr@aSBz@!A21mAmb*yNiqy!%b~;9-WEo$y
zgxhnQ;q2lS9{EXN4@SI6AS58gr0be`EII5wMOtw-ETJsFm~C%Ngrdab10R2y%u+?3
z_L{xlkVc38ptv-<nBRP(5nHw>hj8x7biIb!{`jk5{Y2zm;n^{@7Fl|#b*#BE3wdy8
zlaj?uNVZp-%%2*G(qZ;|gZ4GwMpa*#B1px<keER(*Lw8Y_8yYJM?Zho3QZ?37q)(U
zwWj4q2yM%-t{g$T#wken62njIs1pSVPbDzTv=&yDhR&`*kFNxYu0gS%|NIJURx1y$
zrnN9~b_liAVp3L%ail{rVwANzwzn`~sLm^dj1H;cEIfsC5WNJ*B~+<t?se*@b%mx~
zoG7!b#vA7}2Xtf-&29STxPWBw>oFBr8`)XKqOu__%t?3iS}vM}#&hH*goR${I5aM-
zDR?=$p$b(aHfl9(*~PL3HmFIFxwBjK(FkT|u}Y{?*ej*u&pDQ-(CFa~1bjt}$f;zq
zYF`k{5UM^*XT?=LO{(<=alj@@;;6SV2)QM(i02vh-uY_yjpvH}Zd%VLi0QPZgs~D5
zgg94ia_QyNce0Q#!dRVd)+Z6>k+f-8kW|LEGB_*IOj<^qkGxeYWiy82+_JFyiP@$6
z1f12P_z;~+15?i(o^3J36H&rY$u+*K^3H7Qg>90}jXNBR-UFU;b(&3h=kcvVZ>!^r
zs<%fXy6&?h2O~t$7R#Wbv9g4=;2hK{wZcbX5`Qe<j05iTZG?BS1nZ}FrnsO$+GWGh
zwObc~-ra>5F-mN27T=gDaSzfqWGWG-)^Bzp7%y@MXU}%unxv8oD+!3jqKb~fbV{^`
zhe;cwB-aRpD3OVSt;QPEsr12<2CG4bx1bDyIg!nkDxozy-Z=Zf1W%qvjM}t>O6|b1
zPTO|n+oO2dI~Gu3u=0zJ0c|mfc5dehnIALShIRw@HJIwsR{GPC;}1lxKA}0o1;24W
zFoJ-nFuUXPGE?RiGTfglA@o89io`ycHXEwl<>EdXiZc1THziLjFi;VMHL6185IVRL
zdZ8-0e7I*)<yIFl0;sK9y&>(=pgxmgxC)9EaDAbvY@}=#UHU+S*jag3;KYP(OpJoN
zmt<etI9?g^6F-p|(p*FVv3@Hgd8bm<{RX(xmy$GitP)Ams@#SAl+XqY1SV~=T=kbd
zL$(S_>z`ZP%5di%9Ii<tKf&^=?MyExEvD|z&_@dN&(Y?95Dn<FANX?(lvk#n%7u4_
z^fF^V!GQ_blvM3;ufZ2VjXyMJ1G@>Fx$T8wY_J=n@+&heY$R$f99r=<Lw26A&Cwph
zwHr5Jjm~)_nXz>*iHKsSA~#131ql)oCp6g1GjcTyK{#WFteoHU^$VU*umLH<Y%`k3
z!iLHk7ayiOAUzXNTA>j=`2Zr}t;+Ly-=F?@b&upv5;Bc#s8sTIj6il0s7I^|OaV+f
z-mXZ9u<&MBc22WhrgP9XnOlOt&Vj;Jy0_pFQcjRb4>i`jA~jxIU7{=unA?N^LhLyQ
zI9jpK;@FY(_1l5sx|B}V&?=Wmazs6Y6$@9#P^$ee`n;&%6~PoU*Y7tD2%^w-4xxl$
zS)$?V5NDM<NNa8WZa<unSh+@!L1b|hDO5xhtf98;BjK+kpa>*rQiQX+^AX}OQ7|LP
z)j#^ZaD=GTVtX?Ye?MiT^NU|9&u-60{;Mg*Q{%5CP5eEi9?o}TT*j)J7UPhgx8w|E
zDLu>c!xywco$`Kwoa!ZaCE_7yhDu0+urL-I`%y+r>?%J2Sdg*MM5F^@99mG*@aT>I
zJ||@Toai#&2HDc7<o4vzYIGa-)@Lkq6xP7{@rU&8mTg?dmR*`vi{bW(_;(^dp&~;d
zy0DG1tFyYxLN*q#-pl(p?^5`D8j0dj5m5QaYVU>X;`q+N5Dl#ge}h4M=^N7UbaGYR
z>Q{KHyWT;GXPypiGffH*G%O4r5UmamR*%BOL{@W+IHFS!2B-bgcT*4tyZliP`1so<
z#kAX0u}zX^==brog?uY4EaDmEt4>Hc;dTV6Li-!@FZZ=EF@!mhvj?{(<V#Ix0)bFb
z0PmLX4?oX)Vg!T?BtsfAM8=ztVTR}<GRM+KY&`vG7H5cYs@lV~qq47>0ZWfE<_;w_
z%*IRA|A4usco~{18)HN~ccCvvVABk#f?p_F6}?F^MXPc=7)tf~F>6<E2Z<Wh?%0V-
z4%nQ$RhKVh^{3p^Tn5k2e`jDpTy#!!!~!Hdfv}4&Mp8_VhZ3~b{4J~<49&9bi6$Ci
zm~OW9r@Is_S=faUvDD&p_tmB#2SoA!H_pM=e~wSyH5SIC4VfnT6QBLsSAgeOEZfn*
zFCB(~hos@;0T<jNY}wby@uqbNtYKV8T}WmemO56al0}`h3P;=eNR+^s!c#j^%&__q
z@qEH{!2W4a`sYrvMoI%BxhSpgs{ZBg4qe_=0ZL0W5KYI=@+4o+_nSeueW4eq$^ihr
zhd6FP!7Vd4puU@r{G6oLk8C<;JNfA6he&VW#Ha7T)(R(AEQcQs|0Dv$3m$ED5N*&}
z+2V^X(9jya#EGJbrVol~!T@}1ESl$2(rY_}w!6Z`Dw=ca4|We?R(shPYm>N(Jw$c4
z>5?7=iT=U;rSd@$%!$bI`K<Q%O@DaE15t=x3N9RV7Wri{drP#dcb~68hk0!ml&E_O
zmdes4n_^tX50(sN*l76iTJLg3<w&czq3-Ladmu|_@yW~?$%qyc;pIZ&1h4#zc<St8
znh#P6=^YV}DA|J_H%S92kSH}O7IKoJT|}K^jN;qJM(3sm!<sWz#Rt!QkF|!BTSN09
zC)G_#ZQjD^W@N<F9>EXpeG>Lh{WAooE2T0~CqgJTw5R=HFV(>g3zWiY+;30;G_7K)
z-F?*UUn^V6HLADH2I;4>xl*i@O(}_U_e>D<KH~&cJ~+J7oyyblp*U3fARQd7YMY-V
zt}T-N!w)j`UfUe)Mp`7>Jmjx8Zs#V-c04>4{T|Yuen}pV=FuTlQp|t`I5Zo~cKyKr
zB(nN>^0o<=;x&yJX{~rv5q-pRC#~Mmo*0k$)J+wo>cOAQ$L2N`6W`%K6`XV5lt1Yv
zgn<cj2A$fo8{Q!udx4xFs7nFA9>W35^SzKEg{*C}BM~_=PO!Y6U<}A@0#>-^_m4Mg
zcMzudJ#*a3Fr<64jnO!&A9<{r=ERcOqf=*-LE<3ti{ji5M8FxgnLO^EezE@aDI&ZQ
zF?F#q-8Nc<w)@@o&yUQ|l!ZKOQl(9-X;_(yLWH6-rf1+&IZ8PzKp-swWq8jmMw7U@
zG>(QNcH%j4X+tIYu_#_*<3v?N8*ryRVwvIgf)ZmMG^zIz-yxdUWJ(U$@kfPgiTZCe
ziYCE4kgiLqC<;J>phNQ)4G@z8`r!n0w7}|r4nw}5Ft2w{1^Mywc@IM11Y1xemaNEo
zfHLt7X%mnQ!3sy9c+v#;M=7I&&=b`$OJih>EWva9V?LD+hwgnJGx~ck;O@bv2RktG
zMbkjQYF(1$i!=}fdS^G|=#3ivAF&Vmb1x<YY?|rNf%~!`VW4k*IRw0C2LFl(Iusf-
z1{n)0CKkCev*=wVBN5?dm76^>BzTV;pI>rc_W!gu#jC*5$NfgGZ8N)JZ)Z`hc6v=7
zil^vRe6BazB^JhfV+(8zij}N4xkfi@M%i^@2uEGGxO{AKQe<!HKdL7AU?cEy$_w1W
zZj0LuTHLN6SDx`!V6>~2m#Ek#o8&T!dRw!o8p#`LV+{l79T>7qiY6;CjSUZIoDmkl
zcEL+@qHb>VB0XfT>A5p?k4X081GOAMo$%<BkwrmVcH#*gZSM^W@|;%jyZ5?UWGd#c
zJxBv`q4jqA?=7|+Ri0&y46j=(lQjzurD+zyF^><5?kz|V7?`+?BhT@n`E^XVy&ZR}
zqN`<n$2b3izI!9DS*JiZvi6E`{j5G~dCxE0815nf?h*R^Jl2$Y9z->#!aF9F2-V~j
z7J8g(&RQuL&%Y=}s;{~X+tf{Ew6-0-HE(;nbgq4=xUu2%*0rrbspORNt~l1(Czb+W
z4vm{^ER8xNOLdzXeT(yrLKP!Y{?*hz@*Y-k0Z%#GV=-MFaoV~1%lVqB%cpG{_<SeZ
z`P%9E%*L!6V(O4XXKF8#@b(zOZ6ZE|eX=+d%F7v?3OE}mdC&aT(~-i;g51r{^Fi;d
zy07xV(!wd1mt#sJ3FK79>C(`O^`-5g=tTQlmTCF<Jjayr5l9C0>GX<~!!$s7cA3a`
zJ<o*O%EiOQrJ}WbHT*CE;CEFc-g157dZ9?0fwaoFyDKl3dz!&`CK5}7k7*-)%$3*t
z5VNU)`|@$P;N6{7`X#cN$b#y2uHC{Qv&rI}7ilVJBY@iFeBNtsEyQfc?miB;;dL?k
zj!k8C<DD#3iyG!&ad$4T;x^pJoI=D<XR~NRgd>X#5euq#x=QHF69y@B>7h?h!X(nC
z(F=o&s@wSnA96hclvc6T=y;8CjN4Y<u2X4NS8-W>MSbGTGd-(|ca*Zz`dB|}N)HTq
z#lmi>gAfU!iFLz+jRTw#Io{p(Wz|<NXD<Nj#B&W<q}H%hzn^RgR>C|L(|R{dFIQf#
zQ+}hYdo%yzV0~0-Bd*xX&n?m1-*MU5ty1<?^(^nT0gP#W%I@<7cUCO&)yykQPkz|C
z2H6aGFj`@5lM}NHO~mgXX*zSVPpmpG&8~2i)e#oN01M5vTZGy>wCpD(mXT5Y9Q7&e
z1b&J{YPa6(ERy$A3k3ATrLGn0^+RCwCahQWdQR#SHr=R4@v@9K^Yq^MXNVM;uHmHc
z;sXt4(%#hBF2OQ9sF-?dAy);AQzaa!QMgWunEh$kaf8R1sJN<1<gMhIjn$v-@VMY*
zuPIfXc%(7@XB=3Y-CgmvSLzk}5zzrS?HM5eM(z8h(em@irIpOX;3=MDFjSNIs-#dh
zYwz_^tLsZi!1fibluo%u3DF5fd?QA2b)6|z3xaJ|4E}7!T+I6r$Mp-~ByU}%vqFO?
z2DbRv3Qyb-jol_53~x4ypa90gxVs<bj8{!<SnOC$zM!Y>L5xL+T$xE&xBc2>s%}Tm
zm9ZXIFXAQ25kt6L#K<4EJ~*p~yM|NV-=YL-sy#a&v<MidEPu3(k;B3BkGU{qLNw7>
z;Bh9AA^?3g!3%0{(hDnIE`=RBKjdn{TYAK0nSfiq>Ds{7RBiv1fca0Gd1P{wDm;{d
zvLJE4wt=w8pI^1)ot^!tt9;uv=)@nYx$+|cMG+rMLT)UCjl#u7s%15WJ1lq}Cq{G2
zn9qq93&)VBHJrKh^d-++&j&2(pf8bJRk>bHxg1=UFOh}ZE=3)R9vov(vPfU{s<`_#
zs5Fd^&n^g6b<3I02UO~^-c}{r%PyF0?bR2R!496thcH0~Ejhq(0nVZ}b2&swmaJ=t
z-mIZBo=W!dB~VD3sW+?i7GjFNjQyq-#VfnbHpB6d@RFf?>Zq>&Gw_PJNL3$r(9XUW
z6fkoXZ63TM!uP01Ij2HXJ9EMfj+9`7pUY*^qRP4=Jgw6Hk0#pMuRjqCh!155n0x7&
zd#NMUNu*+aa!K6<UTUruify9Vc7Z_UXbmsPab2kVmz4tl#j9w~&xF!k=z<p<BZa|t
zGXuY*g#-%)y1knj0uls8hR(uf?+}kcewSIqG4Lrg8mq9VvQa?J>o+Z!Sof4vMvR@7
z#8eLrbHCr+3@ZdFr0#SjQKISo_WnLjf!^|XS%c0qw@8(jht}%z46^kdzfb)}<F{?x
zx<fDD)T}<$%pvE8i0Xz1Z<imx!hGzW_Ne{;ze10NF%h*4QH@-^(Q&ytod`EMWyVcw
z$~gw{`hp4#3W3^7ECfZR?}FlF=eHPpNi7<8+0Gc_)w~D^L^&dTG-rd%7%ZIR^X5MW
zHPlGE)r@)O*o2)d$SkYdQuR7ZUvXcik@#v_-ml;Q8vksd@BRMmlueN}@$^py|ErQ=
z89GrP0^_(u$j{nakx&%Q8<G88^V`NAF_=}rZ2*61^AN#y98R0mn)A43GW^PAxL4|a
z|JN6<WJ0cL2bX00W8SjA@-5j-b26-#mfpgD^>T-9$jiu{qvulns|^WSU>5!MoKaSv
zG-tJHHiv@;djtLw@e?ckJnAF7&c~A<^4Fja*+!Omp6KSf;NvqsWfc$P^31$!=4a#W
z-5((srqWQ_Un4Y!s~61qoM*@9V;LjbaWAgi>YGKzV^4Xe(?L6!?v4UU!ztN|Sl*$E
zp+$Z?;SG9yd&02kCnQK1>_oG3M*Y1XzfpdEP}_ug((QfBSlvj?T*4#WOZ>4zm!f~i
zxN41f%j+05+`=oiUfgm?EZdZm#HiZPt8(jh(orogqj;lUS~*(bS-jhiMveSlI*D9^
z5PICQt>}sC@MsC(Sf8#@T52Sn*@c9O3<yjTjV%db#8NfuvuDcI9ispPvexiIbXYKp
zD$;68y=UH6T@VaI*u=*pXG2nc(OM@eNu#!c<7d<S#bR7={PDWo3pDRs&HGE|>!RLr
z-`4eRXKHHy5|71Mb<gD4MeAcLVi^7d<dY1Fhp{J96>v@p83T!lBr9L-%GVvNm=92o
zI<DADI`1+XAc)w&tjMgEy<+{MCB|*lra)jn7Mjk!RD>M9h8UL7Rs1GhAcLiD<|LDY
zcX979zvuonNItpD-BCW<@?(#8L?UGX5d$x)Vztj~21c(9TEh|%@se~hR)$E0yAE@`
zDj&>ggnP!8*ZPOnSTVG!!zQ5?eO-mG_6^x*ffluT7fxS)SiBVjXvI-k`LOZ|9S8I%
zX0Opyv1-U&W2JJ5#80ZuiduubGbbqq+v!~c2LW+B_g>O-Z6y?2F$A?V#d|+QFjw(B
z#(1Z#XdXRqaAw#n6qj0&5QQffR!IuWNV0K=A6C~2cM-;MGG8LFdur;*=RgYE3<u#C
z5pLhE5GDJD>!E%^*J1FGRQR$ctlD0GZX4uBk~rgM3T%B#)?@Jki`qm5t1(v;y!@#H
zLDBO?O+G07>oha&3#2PlSJc!sgQ(%u&SraKS+mRn)2FC!2|Z@Fr3ko<omb0dBN5cB
z+h5Jfl0Cx;@64KX01UhJ8l>A5Fs#6V11}thkaNc9n0W@vWI%@?#Y0UypLmrbXm5tL
zI(jeR!V1aGB^dP-#o1l~$1FMimC2n9Ml%F_g=f6-PXRzOIcOC2>AtjwWXir#^rl=~
zZS372%V@z&5(^k1VMalra^+XV9K>>7jr9dw=X{lq|7p||nbu`EkwT^mJt^KDY!(dG
zk94`FLyuEPb}qtZq8oNk^P*Q`UO$;5@}oNWbT7f;9Cd&XAC-DX1MHGu%{$1_E8Ir8
zCv1cm7r&_j_r@0R?1%5!#%&v(_ax%^KA!6F=$*U<nRu0x`4u}gze<XF=DTEY^i&zF
zFl&gs18c^8=Xmb8gQHMN9Wo|3+DT2<^bPIR`rdECMSZDvHB=`C^O!J;bWFRfR@_SV
zGY#$t@IN&S7|VR4ULQTX&_mXoi9w|mH^VmjgZ%_7i?=VKTv0T+MaXpa`0gtP1SpWm
z7pjIO6tUlWrtx`5RI=(LQT<G{9^!I+e%9iler8<p+OK>@4?|9Za35t*iDAweL46a;
zWUS&>3=f&98`;MM=*++Dsu_E?{pFpK<R~37TD6X(#S_I)r{&c7RIe)GOeAV|hbit1
zo^1_2GC<u7c=>fq@>b8g(s$r44N-lWqZxZh4jAq_e?r;$JRPt%RP0kcJGh`j=KS>t
z{AY&eg@O;hab?22Bu#Z?ojZ#fyCu*FlnW_ZjT9M0a=e7EYX%VBHTp_1XsMa8=-+j8
z^fGcO1FIJ>c9pC8J_2=wSH7I&t{H+j{F2pJyicX2edg1rftUbo!W28I#`C?49BL7C
z&jzb_7HXlbcR@y{)-2oST{14SVgM(=!sH&Z0{AgK&s-(cT$a*T<(iD9D&c2WbU}*^
zLdpx_VaQOM4V`>^YyeGC;a;ra+V+q3dgW-K(ymFbS^DIfX&D*lu!cD+-dL<h_inO%
z7~G*S$syQtas2R6Nt@Z~1YzM~*q=W6ICG}FVT_YPB8b^TR(06uH7PuI4XWT9|1x+f
z#`p2%E*ciXyy?}V>!q`tESXztXH>#&#t+UVj_O4r$V~aQk%T<{e3HEshTFhli(Zc(
zgO)qS=)0Y^ehDU4-iNmbI{~|E9<QUKGYpk+pR*@L^l;-On$S4Ur}u1RGV@06Uyr!U
zM{b}z;vIG$W1{q+Zo-!7(O23ZAfHbEL?{ox7_sq*(0$bJu3{C+slajgEws!XaIKiU
z7;dRc;+1(6UxW|4R#_hA<ns*t2;f%@(gaUO&7PWXVe1)4R-dLnq`17@WPbmFIseME
z6gT!gCFTIhflto<ooi6ylaRfNfUBIr#UgxqViM)fGD8}on$&yqV1j}UlauKTyO8d(
zfw-`6_C_&7>2IT%0v552kl~_ED!TiMY(D_tCI@et6Gw?c7(v}0j5nE#F8K2I7uPI;
z+C@vKJQDfV2Y+z&T)1n{<SQC8<ohM)5eDy=)uV72FW?AlqW^$!lUN!}&_D&1L-;ZP
zt<GaImCJ93>1KWH$stvox%v;DKY%KQJU@I&9{K4DN)92|c(**0yw5(gU>DJNH5WZf
z6&3AsEEzTJX9^+7BsUCNikM&C8-vO}CG!vlq8)dbxyL{uXi?$d3_k9hEgQH;E+xVc
zwS8|W;3t{}E%l^|y$&2UgTwRBD`Y901ahjo-{F)V6>ge_i{)^9wMOpvGS>*;A8)wD
zhjA|kx;55T>6{#~YisX({1{?uD_X-v{xFSlCf315L5-nfZS){H;9d#$>;^Tdb|z}6
zV-<zMl%fNLd}etYcs|#l&vZl4s9oX}D(1}8Ayj6|oZ+pQJnh}~nc?w}0X;4WUjsP7
zbKwj#H%hcYD@Mpzjdj1m&lqXh8-ctf>||c~)W#pp;u>@@@rks%QK0yn*zsT}rQ_~)
z#CO!2qmAG{5KT;9iDHqi?*Q}QiF&1hY4!At`XTC}4KAs)do@N>{@QKq?<Yk%$}s(6
z8DQH*V@5SX+QG)#wU7(*UI2Y1ry^<{oKSE!VR^cT1OhU&(UdFn)QyDq;^b$kKv?{@
zMtC|%!6rzg2y05a{&nt@vafMeLpD-Q!@my(KZ#cKdhv*Vf}j=4&+v%3aT5QFS?d!s
zUP6YWArBjmF2GI^zd3<KRvUZvNAAqU$n@gi1@`Yd)>jF~z-RDC&9VXtv7@C9Qi%nh
zh{xRvj7d`7>!)0%mPuqa0LaTw8MD5Z<ow@?rS#_MQ~EBoN!)8Nsd<BYI}W7!xDj=U
z*z-C?ow0SqoXf@>uX47%w=o{yk!%U3MhaWW=XY+-#7do8=EPH4bdEHP2u3B{pvF#3
z214SiaC4W~_u@>r*`|@OrtiQ6g@DM=wd|ms$MHj=7fyN&(rKs^qB^(@iTyxG<VQlO
zzl>r!?WZwVg$Ov<Np=;qUUluO!=;%NR00Ubhz<=|HnOEe*`4D)6KsLR;&^cb!86<-
zW6$WNH`2^z<kYoq(u?YG6XaSZmAo86pKX~bu2K!$<~b>vE!cTr|Dj72l~jE@m_Q!j
zO}_yIlKToOM^e64fS+5^4}PJm7e^(DSdG(hl7Y2I6)M&{cwI#C)x5eN)zwFAS90F<
zVg%|Mw5^$B#*}ds-{ZSyCR~I{rAph@GuF9!4PxRp*iO+cD`B<mfXv8@yXR$?+QqKA
z&?0Cb?2AD9PJbJobc<*R*q6+Y+9!qul^ZkAg_w-8w~%+@9lV5R^_1DFWW5cpMy}T&
zVeXyL(QrYU)Q#jOINOsR+Nx$EDhsjpH6RCE10xSzXauR(pezlm`#}0VY`#H|3R6o7
z?&5glfwjaL$K6YUr@qL*iB?oxnm4#HRXsFSKbgs+MSh_N>j7bFb5S);pNlS6#D+SA
z_kb-&h^-%tN9%zlcv}#1cnyj>gV%4k($@lP0-T5#tyBi)<a4!oMQnL{8MV4e9^>uo
z^nXV?fr)lr=CeQWD*-R`;iAVf-<nj)-|?=Er!*7L-C|82YgF{^1tQjJq#}n7bZ;;5
zdB(qdIqgglb=+;4cd0v1pZ)6HoWQnXc)(R{*ivdH0^gF<9YrzpievVA@hw(Z;Ax?!
z6sV+E_ps~k-YOU2z0sZj#vrc%a+pV^_dy=>_mt&~$|6_ckR4U;H2<(<e=&<2vn7GA
zjkdyk4h-qe5Kq;yntcv0QfOQ1-=ZUU9@<s!B_qCyQ64u~X>enGQXNbf*Q(cwD?(p3
zfMLsDLPSD6?Wgrj9W^s;6z|3OZfVI0`91xyh-!m}h(=N{n8izk(6{m2WgVKgZ65b!
z71Wcq7;rcO+#Jn4V$=*NY%^6<3v2(7!a=}mvIGbah+3;f6mB;vq6IPca-!JO+}T9H
z%TbNej{X;H>F3l%l-_AZ=j&px{hBBQM_9|P#Ji%eYQqoI754Ee+Kyj_vM;M=cy}=B
zIVSel@0qsHnl)|or~RSV=vUK)Cs+cPW1nt<&sA--H@0dXUPbenAFJCT)HzKiM<ZrG
z15X#=?I8zwW2}hKFxDvuI>GwZ=pMxqYdN^j|I?uvuL?6fMwI=f2-`=W3#T9VngQ3K
z)3QDu8pFIk1xjJ1a>K!n58^8_JIm)%=|xV~*PwK}(|&1Rc107IJUs4p44gJM@sfHz
zwj<cHp%TqS@-v!GjU97^TsQj?J{r>oVsd(TuQR&A@6ZyEq!y}&!Zpr|MGX065pdu4
z#2YB$-dCMx4jy!Xz2Q!87QQiR*Yx9>8Hg@ZAvA?;QrGm(9}`T!Z`MbHVU3s(K*Su2
z!q09Ys#;OpY(mVBFGS5ur*47*P>x;I&rlqv92)^{s1TzaNi}HLyW<ew7JN|p0AcGI
zWW=P8kCASc_)W*0SoBs|{!a#-52mucvsM^)9O`a2Wvy5qGCcwxA60lb@JuYzsr6ta
zJv?=ess%7Vyi{e&Ve$?4MN1t{c(34dhph<6l$1(h?)KTgjEEQbmqBEn;}*RoEmq5?
zfXt)a1w1H<0aldC%5PPXMpWrF4ZZ#Wk;%a~PRRCPOP&%UYMg|eigP<5rg4M+i@o;%
zi(*;VMtfjjhCJjU4;dxR5G2D8Bu4=OMKTBoN)`lh$T=edf<(zEU_wM4f=EsR5=04-
z1eBx#BHkWcd#&X@|2bzr_nv*8`#ir*(_LL%RsGdl@Ap-h?yBSOzf1LB>9;>>K@Bj3
z9!E3eBrtHsr3?OKYc|-m8z2?d;(AGg_SWG`(idxB*AoLW5@yK=Es}h&VT`oO|7Z~q
z%dlt4r~6`V$s}{>bxVr2J%z@xSsv#&s^b5Hv@{5(V;wBw8H(K;T4=C$v`D-ph}^ss
z|Ebv@{8}+lY+6(t8uaGwlyj5P_(85ZG&$AW&Z9KOaqk5E_NL*?xDXv=ly)Wrx!`ww
zbRy@cH5kbE$Jt+z?`Jyj{%N1`T5Wm!a;(HnBqHk>lJ%Mk-uk~SZ`SsbKN)U{4pU?Y
z8&A}BG4sV|Ov*?~eKIl?N~X}l5jCBRqfh@2_VqtCygzT;Xc=@SOAqE>68Nk7xF(hq
zc>P<i%4@mQJ@9kLGiDB!`ErSMRJBRnV$p?~i=BR94w_CI7oncEt5RPru|ELgrSNNL
zt!A;ehdLR4hDSYyC5b@!@_-OZ;cL5zr<dIf33<?))hbCBjXI_OiSM#6Lxe(e#yZ7I
zdHzgGhbhI;j{~K?8HACYpW2okQqq?7@#4r2Ano~Uxx&q%7USWT*OZ;M3xyz&+e2#*
zFJ9%v*a*A<2X+X(IG^W>-0JUM*!%;;Zib-aT}m4z;ewXm&VzMQ`})u=9fLAh%78}T
zvYQdc0w$@*$-^8X;@YJ=<ax{e>-k~1uA@%SQDd!_?)#e+r{|X?iosEy>|?Up2YPlD
z2h#q+U-R2S3ezUX+2QtFoqdm=@qT7xpZ_UBri2Z3Y8I^P1Kr1$gboH+P4#iF+;ViC
z&un-Hwn1_w`#&qqs!@Usd2THd4v~`HpKahd7vs!-bSKd3e~MBgN@D4wKcf5g=RUw5
z<BoJY_O93LIwha5)&&#c;J@#Rp&0ud-*>9B#8qCyUN?Ncl<&N0rgGWW)2)$s@tLy}
z9C2(7auGj~)=Ay5HGV16EZDb4ZO}Z@x&3cXfH~R6NiO?5a>^uQjXOJew2v1ynZ+<{
z@<s~j(u)aeBx%&H|6^a<+<g)4+$Tz*bILr0{--peJ*5oS$@u$*{w5y(9Gz)6-%Ik5
z`;5nI4E_-tOlGZon~FH+cH0Oe12~>~>CfHynIBp$u|a)<A-u-t0o|kj*L%aRnnu_u
zhdP~pgXagD>MtnB-2Fd^8a`c5+WN%QV{&5j*zY0kUx)Woz@ObjjwTTzB?W(SJ9;q%
zd^Lp@;4W&}K%p_JCSLa5VQKY42<)cu?4J_|sc<v!(yBf1K5<82w0*MirUkwGoTs%x
zb=+^S-$?wooOm8Gw7!{gDKR+2acZ9P9rDZW%SRnb&oKY@dVU#84zS|LZOpKsbDuM?
z)~Sw@_7+e<TjJD16@xaw1vU9b+mG%#ZZH#3#6F+#jm#hog~}BHw-W+mNanOR#y@YM
z6eW1JIcp4QN$Wn|Xmq7*rVQO0l51#T=o`pD=&<t~*i#^DLcoOoOzp2{{psa8h;)Dp
zSzn4tQ3jWv`uWhmf+<0~P>m_2)YkY2V$wh39NmvjrWn-T-T|UZ8-ZWrU1iTwOO5^G
zx!KE?;!xL)D}7mMVhchbG_ULjE_{a9tX|2bXgn9qpKh`Ig#I)G@;&%cMCQCxAZJT_
zjFcbE=SDb;?!Ee5*Ie5KRYl;T?Bow1?goXQ*=4eshI)--6019%hJkI9Xrq#7Ug{b?
z#6_%sQ@?m&wwZ3k)b0D((aZacT<V-LG)uIc-~6Vj8n`ROP<2)Vfmg0d2+L2;Mi2V)
zb+)RDwzVuC>{uH!NR+6@Cx7@lojlqZ27glQw$onnvH3f4NtHdqI5dv>+|>rIlZcs$
z<wuNpwyF%SZZ^}NMK^XO>$i6>%U*}xw4WXs48E~@lKGT^lFkb7;a#D%8e5t1_EO@G
zW^A{+%7@Rs95FsK?!c}{&1`0%T;%YIILDnreV6!c$ua`VO+m{=+aI_OFH(3$-6AF|
zdv~*#3)Q<rscejbK`v1fZ?UJSwH1+Tdo%Cu$X=6;{W?{)>_H4SZzEw5d*~yCqw@KH
z|IQ;U|0Yt}WV@*KowfCLr-PM7hOuHwgeFr6@(S^rg4kVpmT2|F27yi^{g|GK7)r7*
z<e6k0;)=P!L*Q%0<6<|vRH7R)pML$5jMPeNgf?zV!0mx(^-r}!WP<1hxVP^V69T6t
zq(Xd?rkBG*VCl=krg-q@)MM>2qZu=uCgTuI$~^ei6ExZ`9tuItG26!X3-7qhZ0E-m
zq_K1kg`jqzF0FWD?TD$`LnUgiejWlpxKh#D0O#!BhGRnw!bzR%Begc4Hxq4htSnOy
zlc*N)j&{we7xx{7H}jHs&0#CyxK2usgoL=yJ>>x^lI%J<jq`6E#GDq7GhFQ5?nSQ<
z<GE)hD5`P3BC+<%2QZOvF`-<`niW6uAd_+zKA5P+j$@eV=zFgdIBBFJjr!tpq-t@e
zrq-;vv6xn0V%H3H4DVpw6*X@5lUsg4J8g4@V~pX!Qp~d_gpjkJV*1WA?n|kj<c=AT
zd`6R;84Ry`DE&z=*zftvqPD$ut!LBV@~ODRdwMhaIpUN}NFbh~g^4jNq5XArkKOzZ
zr~n=s(}QPxMNU8pDRB*yB?U4mq>_}XGU4Gr0OtPi?y%s9BGw6cL+=ZBO%V{dH|aFJ
z9%H^ZI?#DEPgz)vO<Y&@F;lxU7y8uu$dPj5PI=gYu!f-4U{OP!H4n|jM<v{+Q0#in
zwh=EeA3VfEUI({^@Z2cSS2lbnooCVYScg5CNtjJoJZ0cGUZJvGmgyz_psV2Q+7)({
zJdNP>SAHto{R--H1k<3^p-LioE7!ie|7v_*Xc%L@Cnc3Oj&yDy@TvWIfwTP)1xm=+
z4|-iPZPBIy+QEYj>RiTtiYGI_w#Z7GviTW^oc4PrNgnEwe_A6D9~`#H%t!Nrj2r(@
zrEl4E)<TGzwxbnuiYJL{__K0_`M`IT+lePJe&-k1tp{2^AfR7i10d8}!9hWWOD`fH
zxD7@O46L9<Q2VN{UVQnJI&5St@{k{Z3dsvy3OfyPjuyL+idR`NB~+JVn-WvbZVAgz
z(q?I(*HbPWYuC%(VX4<(E%jAV)p%-n!sZPda{^LYRvy%`mLN(Pt$W<<`2HI&WMa4A
zqM41R^}$WtysfUA8<lx`{qrwy8-;4JO#RP^lTpDo3YqJI!Txnk-sfIt?phOt`ElAi
z+>75auEbLF*OhT$dOx3#Fd6j@+Otlii0F%KELr_V!aV+$3ZsJVo;s+D^^<qf_E?|@
zF?q#Ckn~h5QG~uoX4ti<51Tvjj#Yqn&lL5T@d+O(y3r#NBokYHgg2;)yk1<63h$L?
z2vk98h&lCCGb|wTDAokr>h|tzT5<`Lmq7gxtiY_jpZNI~Z|=6vhjlJ3pJmLms0j1t
zow^<VF@|=Ajg9oP@KzIhZa!H<I24&`Ge0<cFvt7u;?1JExH;-}uB{GHqC6k{q&y|j
zVyk^y(5`h`n=HRgX0i#_+*Ep&V}{uSzc3MJV!(xNQT&9igC*@UX=l4+xc+MmciNaH
zB-eGHSzfYnrV2SDN>vi7VL7j${<gj-<6C?j^?L^ediGW?u4FS0D>aRq8x8d!ImS9n
ztrYJ?07j6&x<VF-y}RCj8rku=6beVClE9DDH8!agb+h5NrFK>(Kt$|Qo_n>|RqbF_
zTUOjL`>&w;@%CC{ZP2<+H23?n?8^IkQb$BYjZqomCXC)v4Eix^_;_t=g*^4!>74Lc
zSx;-yj1?+H*F9lZI?WMNko#!>q=9g+XH2K05-~Xg9D=tDxv)q^a%1<|V`v?3;}@;=
zb~-V5^I!UIPty^s+#ev5e8w&57Axwt;K>!ne2))u)GITVe8BDG0AsypB`?}u@L#XM
ztBxq2a3z?dE6PM_0EVybGr#wh))tl*Gl|o8I%zk;kkN{&MmX)`NLH@}*>Npo+ayLW
zbGa;Di-I(YNPB}SZgYS;by+#^?xG+_xlQTDkaGLQ4ou=crqxs~>Wi5T%CvamRJb|&
zr@v~baX>S?fefX=9ZC0Ier42d2abry77=rQ#{Oa!lyrM-TWeg;TeUZm%R-ati!ILp
zW9Zz3iB!GS@Wx;=mWDqQh<J7F39b0BAG=lOJluJfodFYaXJunW1j;X&YCZnqnbSaR
zt7K46|I&*n)1oIvJFAZ2F?E(TbIn?v4sw(tq6e>oyU*Z7wgvRhjo2wtMT*~AN8Voi
zON|+n0&dA#w@5G|QN3)y$LiV6`eg=g7>`(I&8l5EyN)Ih0j!eM-$L|!*gE_1RQDwF
z8=BJ~j>I2yQ9hR@W!BlgBz@ETUJ^R=-1Xgk$GEPi`VJ-U9UOCsA79SG7F#|OdsZWS
z9aD9MT~DMYDyDw^ifXuge$Nt5U|L0QfP+(^aUdI6TAxeFi~7TIKsKwTFyy7=$hd74
z)b|HqxMGK}L<QDcy$DER7}B0~b`cSt@GLn@-C1$u@VI)Gp(dp5&gNfKEn0n)P{W<r
zRtu7W#p%O#|5|+iLx47yDl@V%9QGlbly~5KYWw8|6RCaFV2HSY8I)92A}l(MXp!sP
z4Q$7P16hzv`)U8WTjo3!E5ofG#>N+-9$i^&{&X=*k~4N5r%>Nz7H>rF4a$@F*2eeo
zrlaCoav;2(b|k%YDl`<e7jsrfP;MNp*ZikdKOuBwy%G~n8GAm|5BG99QUl7eY55?L
zY6vWTbLj>(X@qjE=~+HRgPDPNkCLpQWKIXMmW^e)?&dxpC`rwm<4Nbny8YCKUC+5G
zKQm`lsf{ua;E)C_^x)np9J`)9R@pJ<gNMl18z)=d>^9133!;_kDAmVWH7qS|zx{hT
zUxt#&Qs+~`xz$_hK6y8ad>4zYS#^wota)q}cVDvzyU=L0NUu~=k!w37weJKzTf&Sd
zqWwhq<}6NO+$cpzO4;N6Km^|93MmUY1ej6Q0ggGiMr6+VTD|a#mu3!nJ1tUp+W?v@
zb6ke|E$vxA5^rtU^CcyG^{+-@Jm-_VqQHBv^hK6#hM2r&qwuZx%9z{#@R~RFAw7FZ
zdx1qbx_!cD>lDN#?smi$EL;!g?%=6!P#O0CbD~UOq<uUc;oeu6Yw;eZ#^k$=6OFiN
zYf2+ecE~g2_k6x~55%<UiV>NJLru>e4|Hn3oqy#1?COn{#kN;Q9#Ny`8Hah*kHfjf
zT8EA3sK=+p`B?V~E(q~Y?}kj@ssB5T9*eu0b2dGMsdvPnJ?s}9i#m7g9Aea8s=r>k
zJ~ZocuQoUJ2XH-+HhG7aJqjLEK4MW9Z<Ez+mHy$hJt#U@n*ph27_X0A8UbPs#dRWu
zEc=l7*%iAPyJXaiRo8qQGmP>s2*1;<7Q?VGNGw$hYGv5_>sV_ZVy@HPKIptc2bsUs
z3*YO9a+CvDd>j$LPLde@(8jB;T8Z`E(0@1}{p1)p)Ng0<iRZvV#D2+2Q)`Q9>y;Ug
z1?{#i72ouUnpNizanr(Yq+PEN&)nj_>o1Ba5D7cW94CWxwUguuWC_ck%V0T<VHt}K
zL{i5kARFt3(hP-F^oGHT@?<XlsR(b&7CpnB7dgpb2vNuJ&L&K%>Eo!c=6M#!^4k^^
zXjIGvxSN!tPWJyozQ9x@DKzGLeL@05J9q9*;2yZfDUDsbB`D5{(hjGfQ;Cr{%lO{D
zJ(<QL^9i;~_Y{D>gqirC)h`pkz|@9hl#)x1ZgoFH9P1C7c~3D8#TV%M`~_z8CfD^<
zqOW`#4;e>^7DLXBd0Usi$FN6pnFISK<9m<)sxKU3?tGF|$j85UK~5E8;zvx>GpPaN
zQ6|+}hdD03IDb}FynfkgmKetCNEc!=4&zxGR@=<jQJk=RmU>xP^vEvn6y8j^qbDS!
zW@H!E4Ay`pvf=cx!S=;_(j3a#Q0`0k)bf|mJHzARF9fy5pMAU+KdW$TMOtisT#+k7
zeJG}zA$3mdZ}mgbYI85H%;(kX0R^FyF#M}yG%fz`Oj8~r<ik|Cr1L%Dhh@tl6GGE`
znnCi4KKtBmx)JX*TN_`R=B(PKA^{1scc9qrWU-sWz1u6-Q)!iL*PWSJIE;pvBTxH3
zthp7`|FSP$i0<H2`%1_i4GbkmeRIuRBUXI-yzgAgW%bw%N;7{$1O1(mf&cWNpW08R
z`}Ms@1r>?#RBgvEg^JTq<D}rodzAdtgA&04_>#mgZPV(998Y_Fj#K%9{A6G@ABP^I
zorO@hK-G<{Rh>k1-i}sT%+R&-=Wi!+`;PT&os^Tjeep<T3JqhABar|{iw9qJ$)HkL
z;K}o6FQien(s+1qca*=IQ}YsA3^jih(GkFTNm#Zv>a#42Zrp?Brs-H~n##l#|7WLl
zQA@I|-Pt44OTm7PzZCoD)fBm4QdtyQ+l49uz0`d5W~9QaTZyaOSuVMs_rSh^76Z9<
zhrJV5yBdXZWnMD7%_S|YHkOpHF8xOFyp*0$EV(4EEmxE1XQ!Z%#bCIt_7rV(6j+U4
zWIPpfcMWk6M{#}w7v&ei;D^oyw#G{fGE|4R?!`~lZiBs)hRu+!!+i14hJ?F^h;fP!
z64WI1yDBGQYJaIJK4o$drLN?)cGqO$wnoNlRy_l#w4Kz2!B@`MsnQBr;Nnn>r?z&p
zFBeHt{7VSD>FT0$TW=lO`czMbuv$k?O(jJ<L4$3N1ttyRu8y}-p|(VmgD6VwfYrHI
z1;*?i#g@KMs)?n)`}v2M^T#*Lc7vBMw!P=LbV0VZ=uvk#>~nqm6XnFq46WYc&#gah
z!@TFZX-YX1&;EkuUa&UUU6fBKL?*=dKZ38dvyjF_kmpmg>@%fLAnm9%lyK(ZkCZSD
zHSWi@oaj_p*hm;d9A{&RYsgjgPzd&vNc}qPWIPeXjcs=pJAdcK&!(bwUP^Jqd#mA{
zX-T#Quy1WA7g55+kCdN0SGCj|Y%}J4rU#s}5eIhDDO2Gvh`rh3lV_&OPPG~BE(++d
z5uyltn2`Ko%SjMY5l#td**uLTtL4#K|H2_Lu-`FmR%RmQYSYE&7c}SgG{{-XwW6Vj
z;_35eGjnKTJEcULhE>{qM!Xw2BLW$#<tEZX*wiw0Z~&?iE|f=dgC?{up|9_#vrx6f
znPmce1gG&B3IHj5{d;<~d4u`<2z{{w?=f+KFZbey&wKeBDaG6rYOwW-|IV0QcUPk!
z9qbESd61$WvmN_~6t;t%#(NLJ!4u>iD3VSQvfv&;1UKFY4TB$88*$%My5%qE1&*A6
zs{bWCN{%j?>6C}*GAI5CXqCEG3<s7$PE_mMt(L9jfo|KcM<vdrqvB$3(QWUT8|poo
z^&qtKYaRdWeAKJI-XSCB(BkAcODai4?RdMY|54o2v&p#<?$6U+!3bDvQrI{p3hf9E
z#eUZ0mtJ&|Xg+%SCL|VVX$Pw9g_8=mCqR9;;Zp_=p5M|Ynjw`bHZ6QsGuM#7u#=m~
zWcH+|n4Nci#jZ@N1Zhs6YHJ7cUQ3CK#d0Gj{vg=oJz5eVwsu$ZfH#yjIg(O`TJ0*`
z`FE*aBLeRtFSW|dvTY>oY9fS$BGTW<O2hEhZKx34lk4ei^QpA#JOeg*fmUG4KkG)c
zwoS`gBQy{GX&JBcO)hHiBCkl+o)qN<C2=V2bNIVRN8)(C)P?7qy*0EMHwkGT@bY`4
z`x3i;_9_z>`1MKClxsK$5Bs-A;L{WLslwT*(N75S5z?`%VSn3uQ*6^1jzR_h8BK#P
zmMoRaK&J_y4S#a)XUI#fn6amcaUCZND6;A_UH_Jyf28$i^j|#^uQa}5rn{Q`^*}@9
z6;yRpGIXI@Zn~iPz2Wfx?ZdkNVMzN20Dh!INm#7`AO=4HMyyhz+x|U!M<L-G<A2L`
z9QSWYmXK)Gzb$2=iho<u<c0sXq<=&GE406%{uPhkto{{`-v;EL@c3;&eg)*W0r?e=
z-v;DYJboLHf5PLp0r?e>-v;DYJboLHU-9^DK>i7j-v;DYKz<vLU-9^DKz_yJf5(96
zZ})<8{{aAn;xggH=U@nh!U-<)fa3OmECdBQMYad9;0XYrz60_&f`j^AKj8!b)*C<1
ze;)s0H3<%&qWI}w3cg7U$ANALdmMPAJn|*t#S_kuuA?h3r0~(@3;_O=4{jvDkr)K$
zAKPCBgb92-8F}POlnMZ;04d)fJ{`~pebC4O#d!pdP6Pl))^Jcl7(^IcwT9qeBFYAi
z91H-8aQ(OKui++3APXXoCEOy%0w^peir^HX!~+1n5Ok3p<%ldAFa|GxWgv#=2)#ri
z0pcScV*p3E@z?ejt08s_+Q5-Sz%ej!1~efMF%}F!OoSu2YoJ&FpfzJ~1>LNv#D(N|
z(4mHLgMu>zf;eG`z~kSyze9}xx`}28H%m$I1WX7U3h)9b$&Z`}0+v{E{0Nwi9rGVh
z+d(1ncp>n{e+{)4_@izQ7>X#B7~+Ko9pEMe#Q|tSR7jl`;RMVAU>=BY%S3nr!sNgj
zAOc`P*MAK)9zu>Mgd)c}5wdu+71DuFgF~TkxPdmogB)5&4nTO3soVspNw9Df00-B$
z1;7%(5$^xB{T*s>P!)WK9S49=<4FGqLWI{i8f*oisGmNi#5i^UMhJBTH3SVn&{(Pe
z4r)p;G8!S|1(ZT??1ev5A_AqMq~7Z0MX2+D$$?lg1439hK}?X<uviGS?0h7|>%V~-
zAW%XNh>RdgWdz6x>3{@*!obH7Fg7&C5Io4S0Q#6Qv;=g5P@56F!0q^d3pE%h5?Fzg
zW64n<0SFr*BR^RUas~&wj{#IbwjBa=flw2n;Ii|{vU7Ow&VL6r06B*1LV?kbN)rI!
zXiiY_qcn+rW~3CPqjMArOqv`I0r0V)1E2%2xc{2fM-nnp!huqNxCVMz5e4N$6M>$8
zu)2^NOa){+k(3uZAwdvo{22m<pqf#DtcHT}gF7KZN88^F$Pv__P7$I*D6rsAJPu&N
zf}xI}2G1JcKL|Ax2Pp(yAk;*J;sZAVix`eve?R<XK&&x1!VPE%<&mi*rNaaBCWQPF
zse>Q@05269#ODL(sxbx+#MFflN(sUZ5Ftt=p#8`10Ifg0z-Sf!aQ(-#o)G>&dj9J^
ze;JT}J<h)v@-L|W#R&g;u)m=G*W>(aA^$6=|FsnU(@?*m{uP?vQ2&a@|4w7?pPu>u
z`SLfbe}(WjtAEAgH>>{%kKfheuYmkEAiv`A+kpIv$8Q7jPk8(`Aio0g+kpIv$8Q7j
zD;~cM$Uoun+kpHE$ZrGkD;`8gLvz3S{SRGH{IU8C0D}`|82Mw45eR^w<;K{jshZe%
zgLC;0p;|W3v)~*fN0ayyR)|HMSHPz4kct03{3{snb30~RE;7kQCENe+IrgtNBVM{-
zb=+^IpD8cRs$7A+$)@$*MkB_e!zx8hhxN>Ml0pZmv@%H5Tt@^PX5W#~B}P0f4ZSC;
zP2zCl9$Nn5GTY68qIC|Mi(*nM&+Jc~EbICsUizpK-vc9^pNT}R93rOJ<f4;qwo1;3
zlROh3VjUcm=C8W)$CD*F2Su(Hcgt+vH+~5D%Uf0w88hQeuJ0DY=1lifK$a@3l(Nhr
zQw`wa2Byu)gM{DIQ|Y7lykEK@d$vJ2IZqk_l?u9gp=YXblgFj3Wny33$KGfuC^eTS
zI@-;(|Iim}dcROBdD%;aY5T?yQ(-oL7E5xo@`F$C<kVwuA_(gtrE#OIlY*%Hu7^^A
z1aYI1=xLdvLaQ3AZJPH4A@Pl;)vMM&04ed;AJPoT_++_|bXUm>F?LLh%6f4rOPsc6
zw!VB+q|`{cFOr(Sb9N<aZua{SAhn?Sc6o;M?m`7wbUMT<FU3<;OEXE{9r2P;lSShk
zW_vIsiDYK?Vhlu8Y^H(giy~YNQ4a*9*suDz&l;#M+S8CgT(OJF7pwPQXWE&j@Gshz
zUGzcrMNL>W<O&3>**^)Hm<>isicQKNmw!(LXfRbUK?1KZNq5zK37L!zZaO<w<tQ+}
zpLnUI%XG|UTiatmnfbxh3k8c&``=F;n^l^oR7#>9R#f1RF_bXD23<9a&KaLn41KV$
z{zi22w3<b}oSKg4aU@PxJm-d-hAT0!g`z*<lDEnOsf(@XTVh_ke}R2_W15=i*-8*Q
zF|Vt{-U=O^L;Pu|jL8omz;>*;dYn~4WvB^j>Z7z)p6@!wcCYUV`(|%z#B~Y!s0hiy
z1!NjkF04RzJXqONU2fWQ^`zt_vd=4A;f~=-l{M@NDf1z_Hlv^KT+3x%{j6Qdu05hM
zq?%1dLT;C|CpKi1GUg$VU%8wz*8KEIR5r%Djr)yR=hKv;yHr?W!~v0I0gd|da#4C+
z_+ZvanMeK~V<5JE&Fv>z3)9*dG%$@D&ufD$T%p15QOWHw=V_Y|*$%k+S#i6VMfMQg
zWISh9AKUE+Y*{R4Q11?sHW4eM(jjcWwr5#$N&QqLiJQ0(gg=ZAMddlow#g)v++Kxx
zPeliy;IDh4)@{!$?G=SSH;eB=RX?q$unRI=OBqxzLPseFOJeMIu07)-FovP4xP!9i
zX^r5?>ajP6+%5RFDoys-;#`ulB*%``pSZ?y+ert_L@{;0QR%q5YGYPTZPvvk9=ugX
zTD$}sRXp>vnBBsL<fvz>?mf=Dd>iDjz)tW>Zt`0Uy{SvN>H$I>WR&uy=e47xW4Fhf
zgE_6m-S=Pr)S%O*a%U^N11?b4N%o!g)UXkSUzsG0`cBX#Q^p?v?yzFcL7Nd9kGTAP
ziVfJ`sY&O3cl#%f@tspduMV1vs@()K0X7{?zq9?9T|q>#^|LHrTV>DR^uz4<nLjqD
zoghsd3}KD-ZJHX3ch+i@psk!Wy`RzakaHjdBSy_aJztemwAhk*_%Oj4MzPJ+aqJGq
z4?uMK?hT5`XAcIp)xL*TGB&Z-$li}2w=yLTto!bMOTB{a!IJcmm=>+pWm0d(A8k7Z
z!5qJv`b#!)2N&T@H^Rs74eK6s@rOF4KbKH<l0HPq$f+By&$6y%oQjS>+AI!D$fSF8
zS`EaEH8;5N%~Gljs#8~T_?*`=E;Fg|yI0F>xA~Mx7r{~bl(a`jgtZi-S}<*9AbPrO
zV$*?tp-pnc&&)x`Zj@Ud5N)CHE)8&xRYR$U)jxf*`*&+`wrC*Y*#r7!e`!P*CxR#{
z;KFB%Bf1jrT{%{gu}y!BOV@*JUwE1tmYv}0=G|E+05~Z=BdFl98+o(6Y9&z;NshJo
z?hg-s7U^uB=tyI*eBhc=0n?7}OG$ESGdz7%6aI2J4#-55dV0EayTcWo9G?i!w|==>
z>w7Tx+_viE#TaF=TZdZTSS!dUTEyCoUDrtjVjwott-0rP7`a~PW2Ho~XEygRt$6n{
zf!wEr412#k9h?y=$QFrA_ldZlSGX=%p@3~u^QD}*MOUUezgLZzGL+!s-k^||Uaw$3
ze@2vae)jWS@r$$i4@F#i6;HEP%t^LVdPc8a{9uQ^K&m?R(i{!4eBkv<`?X!f^_B(E
z5t%9y3d)I(Xv8P8yds@$hNFXusEEBUH%)2TjP)3M!o%cSihm>PvhO_Qz56@Zx*tH&
zs8QVPo+zt^&LdQ>u~1+qolxzQD2$pS?L;-b_Y8WGquKue3oXn*t<_8z8KB<p+bnxA
ze$@k4EP4-6CL~R;Nt<X3bneT0y`!lO3cgI?)I<7wtg_+E`17n1idd2y{KaC({pZ2~
zSRia3G^eabdiMOLmpHXt^8DSHU%0s^e{=5#KwFi<1{Wn`x_PqW-5s^e(@j%vtEL)_
z0sJwi`he5mYI=_(>Mcz-*?eES*VXpeCxEjONc|M5lTxO>1ikSvpN`m<TH&Yqodp!%
zr5)b$$28+yuT{KjCVqEYk=)Rz_5jqI2(SLjXRi#YcUALj@aA3(re7T|uxa1_BKW#q
z`Dv^}*VckeNmvPnHCa*iFY8%5xYXBwmdr+~K{q%{(}F`Z{lq0>+Do>u>Wi~3$WK0p
zUQ<~oG@&$|+gvAlu7(d^X)fXp+6&VJYPV0&oc}i4#7=T`c#FGIF4DNNFd$QW0j!m7
zJo0MvfWXY+F|-;^a(h~MOJgEEBBD2ty&L(UpxtY%GDVeYCHHSsZqfwRoH&0IiwI6z
zKOM(5twzF8_%!5~vRz`jN!2m1mg-#>-daqr17c?dL^HTNdUWpkh%p~iBr8l|{);4*
zddl0o2z#bzoyqY7I2*gT7(?6yB`cYlWA8Z00*QL`;`UEbw;jbrD@gRk?O2(D!K&!#
zlN8c3Ue7TH&a`9t62MkiU!gMKMO#d9I`!Rv{aWpF&dpA>@+fwy+k0ml0Hoq4T#O5+
zj2d#T%#HJ7md84C>z4{Ogb(y5P5&b7kt-4)k`VuHPcZn?)Yo7d8*5n{D5sw06CV|6
z@PiJB$Lu$?9798Ru$?8bF{U%~Ij%+DB{ZdbCw(`6TKRh_yY%CxEkr{mQ_KW4nr}&q
z;gsZvj(UjO9+wSP4yJMTz&@^FfnDM-!y|kpw(kk=HsF_7!MQ&fzOsvqC<e3Mjrz0=
znsrOt>MI&A-Ray*?(;+wi83$ka!?iC9)sRjd&oDc<tA-1$s`YWzt`yJosP*V(Qc6N
zF51yOS%a_XQI(+5%YNEd(!fMDRj8mMM;eV6+u~2xK;O3($S~(9IW_0(*F5D;+eC8X
zJ$PN1X+%tG@43omz(*`WZeSJw`V~1A!Z^iMJ&O_IP9EWs#bgQU<qW_@NQ4nfRq`+>
z5+V`UtcukbY$LsMF-_;EA9{2?=L2h~sDkX^P7vrf_%xS7{Y7qJtUV92j)^sKIv6jv
zVqKt0;g1Oad_npO4(WyLgNtFbhbAr~Aagh@F)>-|OLTp&h%W{E|E2@{Sg;)sB&-`s
z3<W<K{JA0;_?imXbHdo!qqxzkCLnp!>NkvAj(SamHA4}~&N(htS^s}KWM*E<2)X(G
zoxq+<*E55y|Cx)e)>}A8*#+E|7ZDk2GG15MH<IHrRw%^DR{Pv2cMU}HoE-q!)YV+R
z4Jjk;!4hA}nSDd&AO-Lp1j<y#5Wy)o$i3}|{6Zaj)K4mx*p74FeJb!FQh*qzL=<b%
z3)q<iRF<r<5;?3R;=E2$D?+p34%D~K#~#FNbDp7ugD-=;Mae(If4;h0F!woNxyhcs
za69$$w8u9JoRc|(l)6Y3e*N(FwMyF$7W<i35z$>}BJzyO<E$Z>#-bBfS7YsZ&mG@y
zR`627c<{cF7!G<DwMVn4c%RK9pjf57O@dPW)fZUx`sxnNO`h|<6;Xp>P;IxEZFdI$
zSY*V*Os+4av5KQUZmz3Wy@?EvG#3VC12%%)N@iQ3S7t-i_fkYjZxE5nP+VJY&XoEd
z_TgKg>eQvV+w_WDyS-Q*@-!u$<i|xb4=#Tr)-53CAC|}J;7Vf3@{UVv%?!0n5<y7|
zw5{u6&lS^(!QQK6Zw=a~**`i=m2C7ZQZHGfv6#}54-A~;2GoFn!RYLgJ7htk$UT*D
zR{rQ@SlJY$+)GJn9%aqyAeL*SfaYU?^H5LOUB(|lM`gaq)!xEglIBZDM?VmQ<);@g
zVi?*OHtI;zGIKS713Xx|(;tO?0Cqm8)^{BDCyG%mLyer1Uewtr7%9%FRY|eqs?^Zt
zAL*HNVi6rw=uH){1RruFykfu7*kd2<hvy?h7p7|S(bYErh!Yg|3iO-{hrYln!cd*+
zh=8fW_4%0lP&hmlnrX@$A`f6tWI|<PZ^acb^!2B%eCoW0StuNp@+gan&5xhhzf`AW
z>#0Y2)!NSoEkQ}kYg$bZ-jM9XntV{4z!b-(YwQ!x!pnt7W6xOJ*w1{L8f8V$^lKpO
zio>0hY_CAL7hXb}vtm0^Pk}F70lXffL(*!Syk#(g*opRZsLCGzX}r_MW*d$VuZr-)
z8;^@zVC|?c)^w3BNIi7yQLS=AF+eah6U3xufV%mY!Ml92*z2&G;^b6EuaW@LoGq!V
zgo2)*Euj&dJE@bu8?dpOtnL1j4VBd`CvkL|a-y5YqgU%NjmA7v0VOe)2Nh@B82nGV
zV+dsXYU{xk>*A>|`*0@Mb)}|Yvg`X_he8f2G|ibmB)Krrr+%@x$irp&f$gO{_VU-?
zFC?LWj>*aqq19L?x-GEmt~`6w4xxlh-d-o&j?S5Non40z5gW{V21h=<zZno40n^*!
z>J4(0HgG|rph$R$`PX7W$4@sI8Fsdgy-6MCfOJ(b1VL#);I9Zuu0VAv8I?FT-C~>E
zlqvNd(a4)ZH|lE$_z(`U8jp=cNh@0*hoX#Emr)XkrWjRXW1Nr_&D&-o5)NNvkN$Il
ze$z&M(Xu|noeHpHoFB$m4~z#H3B6Pedk+KXC?mb-!)MsjSRbl3+5lBAQO7O(m9s`4
z(CK7U*xjsy3)Y0)B&L3eq6}$-@oA=$1{Z3C$7(#l@n9a@>W9vFrSeaN5s^cYR6MeH
zaO4c7ueepPi!&4ExvBp@YxaJocI{~5UI_TQBPj`3JNz-M2BGB!P#6>Y4YaD4cNhYj
zR%Be?vN`m(iF*;mXh4!PB2&n?NG*z-;eYj%a(MgS_)N{U!zus1)#TNiuQ^`J+)=o5
z$@;SO+s2QNpL$RGUYT!N{QUCw?f-39zBR-b9*-c4puc)l^6i4lbOxGqc|ltq5vn54
z8t61g!<ejCcS;R+#cSKUBWz6embOAP{3Vj|wUA;*OyJG09<t+IHXF({GshP7Ln`eZ
z3o2FaQ6PGK4s~0NHHOMv6Zf_~3|cc;Mj{%mW`!cPwUQ&t5RU25vfY2NyZB<3o~~k+
zhN}syAZy4?)Sw<j7@|mW#DDc_s5B{$H2nnirL+hP>fMoIz-UyD+oQ;XVb~2a^Fm1c
zH-AFd;<{kw4!+2n@eD$7$MJ4<*MaLPFD^cQ9)5vQ0x+xlv~;Lcl2k3j&>LbwR?vZA
z9Y+-~b}Sd@_9C+-Hi#o@=}w}gaJwE8-kTf{!HlL?$eLqYr`9EWzvd3V(xBkf9FDK)
zkZYw$s}C%M_nQ>f&t<}-g=drs7*1vSaPRSr7C3HtHWZ(uJbMxwphi<N^`tkgmtU*7
z^VvpSegKs+@%BLAwC^?O$RG!ihHG5L`J}U!?s?M+>H`zt%Hy;1C6WEBIYVMuvXh<Z
zK~xm-jf?Ixk<x4&@gE)+&rT*W*}N9gEcxWG%3U#zF>o>;ZuO0JotP!@Zq_KMLAtF4
z4$XGgIp&`=^ay5sseL~^zxuLLTVt%UcNFyy&;B#hi2h)7XyDhR*D@QPR<tH^{3IJH
ziV4K`s~FS-9M>8r?|CIElo0&@?gTbJK0%Bj1AYK^Wf-g1z#h0=F7G!^?%D!}=tAqU
z^O(xbPY>7_qYO>cjbfm>R>wwASQYh_ij=xp1UxmF%<&4#-GLx{Tg5kWD%R{)?h*Bh
zZ<n3MjPqW8p_qJu-onF)>ujSPjJIh9l|`oVMKmHe3i5+HR&Rcq41ZC4&%`COhgF@D
zs*^QV04Wof_|Caw^k#(o0?AiMmW%BDk|>HHhEm*3q6nApY)ErQPcO5uoP2~Ri|~%<
zp~B%rl|aQF_h;u4ExC86I-04TuGW!g8Vd!*X>KTy(B)-paqV{Tr{}9diFXuPV>kUZ
zW{%eqy<51p{=EAKFm>(v58&YsK==pXd1!Gsb#V9lqdnVim$sgN=@;t8tK5NXMPRg_
z-H^-?RM0Qq%0Rfl4LF&p=wer-m>y)?QixyemX0Rc60u6i=Y+OZM$~ri@|88nBLclw
zN?By}-YDJ#5XMSoWUfppC~!f(?$j*J+CpGxZ((FD0pUyxkiz(KZ@ritH7G!hC49&?
zyMhUOdMt65=HT_Q#yWd)j`z$<Mj(Z*Zl7VL%<@IzD%3Qqow#)9;^<B8+nMz1R$QCG
z#J)iqCd1XvxQFHRhUDdIjf32pv8z-!v|N1i_&#lG)Ff9Kc}d#nr?5Uzp;j#ZLCcg$
zZucdA4VPLQXpkxD0QaPQ!OjA%|1diEfutUJDm^e0r*W}GUuQ<QW}qR7%`8c0PMLn^
z6@A<tuw0__n4)}+e-aAGg15vXTvZoN>s!K4UZnJU8mW?y9XZPTFqxUybP8#_XE^ul
zdT-+6YFJjVn8u)xo%ANLGwM}69qrm?rQ9*U)S-x`$7+tX{5Qg5WOvBf#YHf?LBWE-
zO!9LvCY+oPJ_>TMogXMIYp?M>yn8d#&mK9$QTz`)PL(39!^?e4i<Vo3((8|tUVK`*
zWT)><Mq3dV`{tOTpM&En!{7y{Q;1BhaqkB%&w105y8Wm<X~cy&+Zh2A^^1vzP)cV^
zvYcXe;bW@m8+CRtR8ddttVm^%Z22LLWOm3gQ|jP!u6-S<kwQq@qJ@yD0J}hE&N3oS
zLkl{)`oLw5JTkH(D^-H6I*r|eQl~^xMsJ*}gz~}}OVZ>oxA$i@!w~FV)Wab(MyxHf
zJ>~dw#r(99LSJ(m%SKc&r+|->{8jYOyV+;2`{c88Gq<x!E}PZyCL!-z^o-|sx&~p?
zi&KPCdEo`jJKy=rNS_dWw;pZLK;P0F70rKdwuZ?60mu?m`<--1_~+(o`f<Z74R>x<
z9-pgxi-zO{pSr3(oS>g!(0!Wda<9B@-n}nK909GWgXB)6QVI8>^P)J)^M|}7k`a~<
zwtS9n-N1e(Gq0+S-^{(IasMK<I)1t2oa5E-_nAfugDJ(N2qT<eeeY4W<0VW)&mpNX
z{7*0=7c<ayBQ)uXoYK{DP{FoKneUlyWbbT`Gs*eWF8F4#Ah$Tj^NjURo**myBpmL|
zQ6Ncrha_Wyt6Id>K#2&D8OZr!GC%%AwX)=mW&$&yeu^cu`x<AqSv~9g_dX-4ZWZ1B
z@u%Zlc}nyCeP7qL>JwGbgF5rW%JW@1U2<_6i&j#x^xX`h16?!<>4xUH?b6uXEFAF`
zYe`|`51={pM9u4pXCCn?ZTVDVLdkmzd|R$%$DCc|hp*e(d3uq!g)7zd<xs$(X%#oP
z;U;OX+o|qL`-lj3wZ3-m>%dk(7Mia<(iA^@(&|k<6Mc${ukX`J7b^K2tosd_W_~y#
z#pji%abx+UdV5@r)4a8c!(5&TvD8NHxdLn{m3wH$+BsX&tm9aYu_3PtL)CAvKD@cq
zN4m;)#8V^uP`R#M9@vFj=MYK8WAx9LI5(RXzkMlFXAFq^dgJ9CXxMAPGq?l9){{ii
z(&PRgPfKMm%!7K}e07Wnqt&1&W`BBgwBdck7o9a~n{Hr1&*J4qR_>ilRotw`TIPfE
zQJ|(KC9lpr-kJ8EWR!CI?$t<_Cf852`SptZ=BrcHs=OthdS(%@*~c+0nIncPG%)J;
z?Zku~{!M#IH9MptN%uMjiwWy-DI+yu^9NyHz6iP_Kg;UI1f_YGgvUCEkUXO_KEn|g
zF&3f0EJ@tN`5IH=-o|9eJYzp{?Zgo+A4DPVo)7(qUW>DiZVHX+Sd{YkB!rce_fC$M
zQ@jY4j8Q|B7G?-`eN>zenHCDJl4j(Rds0}QZU7h&LD6XzTi0f)NPgYOUAJm{RdhP>
zd6>U`N8)oVyBM!p2zw~>^W7b%x$b93SCTe6G5u#JPH`}GCecQEb27S>59D3YoXeuV
zi&Si;aI_e-KwC7Yl-CEOrI;oP1;jBkOez}apLf-rGe%h@$SR+*;w9w{m!?}3(Q!0s
zMsJ0Beqy_G>A5?r)28_Fn_xi>eC9b=o3)-QpKXq8vi6W1f2TVYw2^mb@tE_$xR3Uh
z)3$80r`4TXe5OzKLO%ARTHUC2cCQxqWADh4jplsv!la9|($9_O_?q)C%F2WJ8Q(ir
zsqL5crm;{DTPId^)7_AnghJ>W)i+p8aCQh8{f6$Az}mOe5Hifh86V!L@%q;c^06C_
z&EPo`Vhu`9&8irb9aqQC-O5pOG3r$oQKk|Ba0(=?@EA>oj5!0n`uG6_=%N348SYl&
zJn|TvK}r-nna4pT_h*&+IxuPe!9-)al~#E)4dqmQ1|aD8IBZys-3BZbmpR496z?zc
zV2RA}B84LQTWX4kU@f*{`g=0d<0RqKU0W0_pmw~FT$DG)5i@!st6FP$R6Ajj+m9jJ
zz^LIm1GQV*+?~#I>1>XynJ$=RvLP0F+J>AAQro9GQhL`H@{ohOPJ^5@@h`jy7A|6T
z7*A$VcP0eFJH82-AB)o(z8bl%S01OYc|rZb%dB-o4QnIQ4eqs3ThB>-^>a?tAI~K`
zXY0PM^}><w;MH3h>vy3eqN%pLT+Xj&!e}G)Y3r76_Y3WkDZOvhHI~4t31UQ!*L<oe
zr|)y!4ce5QymwB>)oA9qTjO#ebB1#Jmx><FQ|DKn=sYffNl)JM<`L#M-Xf}v7xJ@W
zh_Xh$;>~=}=n9!e!lXM7&otP`tKl#71yxVQbx8P%xH5h&$8!4tqx`dvIn^fOpWSo}
zc*s*qtZJ$irdQc;nKbikB6k6DUa6R|C4*MINY%*hlfuhRZ}o~4ip&$`+r`g9BH!YL
zTp0P!@N=-a&lIz8@%l86`m(A(bLscWo56-`)P-RkA!a380f918JEfRdIzgYAXY`L!
zF>+ENiZ!do5)5jBe1Rb#`|MdG7wbJxg8|39ZPmZNTAi%gA<B!eHF&icNbSo>VNqyD
zS5rX(Hg&e9uI+g~KX~}=@a!e<r?^KyfUKz>K#wQ5xNs?5m(gehF{cwONZ7%t^JymV
zMa08iz4>!kDU}#Tu^lF6G{~Z-r{%!1X0*20tZFEZOjf2#DIasmR%a~F;SYm!w2;w1
z_9;kL7fYC>Se&tPsDWL-fB7{G^*v&7Nx*Kyz&0p^S;&(V9m_EPBvrk^A?75fg`0kF
zW$rV_I@_#<DA}q<t|ngoh8J5-N`q5V@;<IA(KX!`Cp4oeUzhF|DIWd+4wxNhP$$s}
zNSpo;8Mf0GBJr^$;hG<rj=f5<+%Mo9m5GEIPY3ubn_N?Rc>nMf>S_X4-?!2OE$91l
z<5`Z)G!V&q+HPlJav#_bVrPq`&L4j^#4}HIh0+^4Vsw4CGGpR2&3I$-9^FiVJ;_Z|
z$`?8oPu8Yyk4z{;=!&+x7>iMqERf{ScUDy&|N2}Q;=ORL0CB-zvrFg|p-<5J6s)Zd
zByEy&Z}HO!-c*wv@)-&6a$S;vZM!b2lnZ9K$&#zG-*^Y(Gqe^1RZiyMk%l)JxTfQi
z$xnM1!%c)?vi#q_8CVIs!TFeACS0cy6rC}rEeOrJ$C}*<)aTz3tqKU)y@=3J6hD<=
zr5!PeXe?Blmxc^EvvT&0Wd}KwLZ&@QKdW}C*ggH|DAoCr%<tqUyk6=TU7Cgg=*qGt
zhd$f&FD`xz9A)*{q0R(rH#@mw_Woq2A*tuCO#pO;PdE0B^er^`+^W7|vE0Mf6j)NV
zboU*`ujEwkcARZ57_XEznw+Dfa#o1>y0O37F`7H?j>?YX3GjrEXyBD`de@CJa%RLM
z74GyqV|^~nF}>U^0Y7sc|Ez(7R^u!!t0TmdQg4`lxj0X~*9QM2A=a#}(8qdgYi5i|
zXvtzGMokQAH|V7qJKkp>IX9wMvQ0xE0OECNX5}jFs~%Xi!5d`^ME@!=MHkK*^Gbm%
zf4W+F{ABLk3;T^>w`tO^2A{%e_ZWS=VQ{7J+`GX*`Mmq4fI~I*d5LBctCU{ddr3Rf
zfn#=S`^6AC+7=NJX889_$8^2ClpE)HSySpeKZ%wJYqN2Y**O=!N&$2CQJh>x3Kng1
zHv7;sQ4LZtGB)^LDZP5_apWlr11fD!3n|xt`;u~l58l7Le>uMxxy4Z@CWMy3yi5@%
zj<R@vgVIPSChH@$T2Xd#C3V}Mxv^N-!^HP87&<yX&pwkct-hU1)w<=_MpyBn=>C^1
z5mN5L$f3{D<N1%Cg4r8N&*aRwb~dF}mNs-<WkxW)Ncn0p=MCbth!ub3ipr{#mr&3!
z<Dtumaj_FXGgr{{O?M3soZS7pkJ3A-$@b0*l;*x6pTnVsXp2wSS^Cv2l`eHw>i)F)
z4$!FcS-zD94lj*?*v@45o9rkxgx&XL;S|}Ipx-TK;hO-`WEVA5;IdZQb8LE(Vy+!I
zhKJLTyu2;61s7o_Nj?xf=l>eq4fYdzb?EgR)E+eYqVH`)w6%CfarGSOgq}?{IkeHv
zt)Q^+Tx{kpd6ccRQn`_pb4uB@1jkd0qq)mHw2P(`h0O0aaqG+}u4UYn{%N^1OXqfv
zs!iWp;n?8A?S`t<QcC0Z2gppj&sc6=4bMnjh|wzwdE~+iu$g<MD*B<2PLRFgP@|8@
z?Vc|-H3{l$?l;hU-^e0cs>5naOBJLorq-qx4PcFZc1qQ4t^VUuyEH!qG_}c|Bz@2_
z7TWy?4&`%AD!jmpl>62v&Av3UI^V?A{P=-O6B$xu_6LB31>b|G<)lvwef{FP&D5v*
zWeUbvPY*v`ZE<?8KI&T}r=^v7RyConY{wOC4f_~6n8;<7$&K8WSFsmkP^+ZF*99py
zg6)R8_U&D*#X*X3K0JIdqt7Xp_t43jTzc_-np(N&UB@Sr3XAmzULSbdCSU8S1r0dS
z^xOj5iigMU>Mqvl5jlQEcg0*UB$5}*NS87+5K})j!nH}Ib@E=vOO03~eFidbJZTug
z%vG`bjGeL{iuL%a{he0u%-HN#xjG%yijk4~Pg8?1;jY)Eu4c?}2oP5<?$HPrMaIb)
zQ^yICfz8MSY#Lh^syRK0t(;izy|E$Q8>~calAQ$kiMP%~5b5Ob8<-`RSD|RR+39*a
zRCup(#*kTE$mxR1U+<F2=$Uvj<IQgA7tu3ivhSUv+^#J0iVGY72o_v8%7*A&of<px
z6H3j`N5|OV0<wzNpicXG0%*w+BdL}O%S(vJ*mU0XE9+Kwb<GF8&scuES18swsAhI;
z<7wX0c$tQfWcN<GyQl{)g>_LMnkxNT=SKjrmlOX`V=;28NJ~V|eD(s^SNn42a^kJg
zz$V&_+Wz`oYeUfN?)l{Omkw8ZR|jj^8f_tr&N4>nd$%pju5&C!!TJ^I-nfDlvMt1o
z+^w`S;o9)o!}NDMSwDc%N8kCMuZaEtbS~}d{muLSht*az(E~EBygNhOy$n&*+Z0<b
z2~~K)g&|@QV)%91y693}yCMm26&hjO`tejmT>2SoRk^fk2zBYqttk0^J*z^+%8Ab)
zBMcejUR^FOD;zJGorA``X>_Fy<B`(ny7#`arqBHOMVFG7^5g)KxfjtZB<r24hM&2J
z(;uPBmL0A^dlf2UXwFs3FiuCSr%Rk@#ea_t{isStG}PA44ko;~QHpGv=42+dqSCDV
zc9qhqHlX`i^8*p7aEe-6mjI^I>|?~;lMUO6EnoPzeQvZ1%9^bMjqTQ4=j$>@>f-n{
z7nIn`WMQsRdya&bSBLa6nJaznC}np6QZ#MS9L1-_*GOD_>NKOFX8!Dg9XUB)yya;;
zj4}4U{SJlpV0gSHGamDK@}8|ftEPkFJK-Who{tWiC$#5yO^!>8e9F|VZ55HW@1mkc
zkpa5#0ORTIoAIeVZ}Re<)HSNy)IxoK%s!=CCE#iKxN4iw8J)MNOjx>T;*=n7Wkd@O
z8sU_dNzu%<sh{l-A|qknl;A{?7W}4z>ga4zBZxb%CMNtlTrpV2u~g1EGW5k5KP9#0
z(u((6i;Xts)8~WJ8|oIp4)!be{z+^ogzVkcT|%wH`np__GJ9_GwVIDuqcjtJyu%}q
zJ#Xzr)#B*y*(b}gTBKcz5Dbn3)MFiZH6Da^9cFiwzHhX;{~n!~7CEaYRR060$)#Iy
zW%EhSIPQ}>x5UuA==abxoipT_o{f4tp|5$J!Co<AvKIf2+bUA!^f#BB>d$SUEyI`a
z>3A`gcduRP4aNAicVgBr?PR5w)^k(p>(Q@J^gL&ETP(XZxK77%LNUcvZ60_bS8plp
zC0g5%W_Hkgxir_NQz64&V})hjW{sx%NZ+LIY9%Gd(Vd9CU3hJdRUi%zjsgn0%?cO_
zI6U8}upFg4mThj8BP*Etgl8jV(0Cn_<E+s^%0*0*_IAj5r{7n-*Xpw{D(=liw^z^A
zJT|xlFx&a91M0pwrG=1)>j}N;kW>W8Y;oyF7_;2tCz$LxQ2XW$^+_^^{9<E?&J%s3
z=3+a2k;tCooiBAV)opeab<_vD)hkIcOzFCxq}Dk5CUTgxF1L#(5D_Qg5HN=YB;T#!
zy=$;>{RKY9h((>nHZQzg9M9#))2=ZbRCWp*Z)dE#Bb~If-+vKnTD|j+z0p+Swi23D
zPN{$O=I-KC)CGtvXvh5(=kTc-X=uG~r!P{N_$e;IvJbLTCs9C-?afMjP;?%Zv~fZG
z{q6@<9S7-kGsb<_vQD2FKZ(NuI5fov=zR*Kpc^YW)InPxDymMMuG;hmI}GIlBq8cw
zFKcKBN)5cG1`DmF`u*Je@d8eWSo^_!dE#zCJz4&Ws#>tIGj_&rxv~a!p~{~}gkN$m
zxn=Ew{zQHJ>)y+3*+UP^BLvAL!?H^Dv<Oxs<-!oB@raoa$FhQ^!S`wzvb6Sx^gjSe
z))QEv2#xVWbe$lC$4qOSjViLr$@~mSGEpf#NkyA__trcCalJ(H`?mH^n93})j3~kJ
zubI4YKYoZG!}7WCM$6tf!L|GX*Y%SehR8jRsCF!!7<th`RI0SQY_<8-B&5I%ORdY(
zByY^pJ4ynOK`KIo4*ll(74;RLX!1er<q40Rn%l?4stN*aKmZuSbE^d9=I3?VJ*ar;
zBYcYQ=g3*o!P5oE+Anvx(5J~ZF*eZPqo~|GNZ+(W2PMd<0Hx?Br!Pu4!+F!~J>&|;
zV?MrT)#u7FiuBqA+2~by>Bh@>nwW?W2Y|U3G2=Z?v#<sIWHb7jZ!#gLWlqKzr9M$F
zrM)8-Z|hWp74Sv+d3o{{TE%ij&*F9I!z?!_1e^+8<20@b5X7D?tvm@~%v)S+`f#&X
z#PGBn{hB?0eYjtxZe|~;#RIUx^8d_1?^M7uk;e+mpRIR1DrUTP_}P-<xAfbyefU-H
zJ;-I?DEPN=S0@8!IK#$0$NYYtzjWcuhpb%({%HGkDv11+_q=IrZueNJz{KK2lOTga
zTVD3MRqUINBn3R)bL%NbqHqio(B`ubB^L5uuG;owg+@kH;MBXN*F%`zEagk|6Z{#-
zaU}Glyxgu=oA1SGWNPa)@A>%kdTbikB+e|&d#?@ePc-y;mVIOQx?QgxYJCbm;J0So
z`px|dmWfT|bF*dz8WgtZ)1xb4L7V)(he`LgFbaJ82Fn7_ys%}vs*AT*(3-u#oB&)y
v;}D!x&gXgVQljCinpL;6>T`{BJRJBN#Pa-&P2;sM#g0zY4YmnN`TsWo?Y*>R

diff --git a/docs_old/assets/images/readme-dark.png b/docs_old/assets/images/readme-dark.png
deleted file mode 100644
index d4526256d462dde768cbcbde886afe07bea2769a..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 38912
zcmdpdV|Qg;&~4aJ$4)x7ZT9KdIk7vo?T&359d~Towr$&d&vVCp?~k}^jLosX>{?Z|
zs^*+4OhHZ@5grd73=9lWQbGg>1_luiy7!0u20B0VP|t%daCQ<Jj$mL&eg7TcL$)PG
zpqIE#qUugcwkA%l`VPimuCA_(<~Ej&hWd8KjJ6J@8JGNcU|_^xk|IBq-7?R!T-`8L
zUWbHC4;R<(rT&1y5fh7wKw*j~r{_|a8(dX-+Z$t68)zf^wmn;|uKFm>uMebj;RO~@
z-jf#`Q8rCwZV{yfl9Pj@1K{u6v-y!*91k<OQanvN@3Na<X(+z^zu}8nSc=H}zr!pI
z2Qb_He@CwF|0fT*ai}ohrw@+T+K`|Tfc~esxw*Qiai@y7AW_yj!0+e3Jnj1=-zuPz
z(!bE|D10u$2I}*bI@2Vkqk8kX{T>eP=HQzH4rXR<Mp%-54yG$YZOv-isinEe<(2tI
zc2|)q-Q?xW+0^y*EDdos!B(AHjJhz=oRsrayN3jbV!;+>%6-`u{eY?(YisJ@SaxPJ
zVn8VE)?bGkbw)f#M@PE;zjG)279_Y>SgI`M+S-|cWxKW7>JWn|%iZ0%GtI9-<KW5F
zsmgijQ)^#$zUUo%CT+>|8oTBeZRvC@EINHPuQ>~-H$iV~Og!n125S^u_c7A>REHQA
zCGhnzY;_L*_&oKpeMb&AI-S7>p5#V4z1^d%J-wL&jUdx98O0!@iyHUT^I)ZlkP{Hn
zSEG}KjlnfE_>&Qa9A=(J>8b*quQof1H*Bsdsinfp$}ax65d#%5_;Mk5w`ZI?poZt1
zy?u#MmQSAEnoLYhtB8bryecP55>PNwlY)#Xeukm+UMu6_;QT<R99Q4sBd0yFPvE#G
z4OG}@(W-Sgl+`8=R8Pvz{xncpTUv*eKF7m0PmynkQL4gaBOZVB@v!&Nwckb-g_TWt
zHK94+8sKTePSN}Ocg?9)3R{Yg<o*JPz0u}c&7Tx+rvJc+JYqfWB*Q#MH4rRB{9+_&
zKa<nu?TTe;>I3Z5HOi11fj0<O-rvkGU&aMi-g6CXS;LNeW2KNa0vnOF9;yLIaZdc8
z@#JhEnD_zS5HW<U>qb)R`}KLGNuPX&A&N^Lc#M>aSDg*z^7iq$<)ot*pOBCs==1(8
zU?-uR>`7ZJn?V^>VGDAzVCKs(J>HO^Z1xGGh-LtFc+2}kK1Ut($}@C&5u$dz+01QG
zeN78$dTji3al*k;S4_0bBrR;U#lH|6NvziThV5z^?hW2EHUY1<y1OTXl+1u>u2bNO
zr-kWR?`BrHzHS#eC!o;KKdnRolO%V#w9ce0Ba6p}k)JHGDmm%+pQ{@a)9C0kTCU;I
zq!7QI@p-I}tLF=I$8n{9dDfSCcvlAr7Z)fT!X1!}oWW0=cn1>X1#w*_d5*NLI5U{k
zRGg`Q3nWWJ5=@vlif`W{5%3<<sg|qnJ-u}e4-Y?|r(0I8q-12!?d|UF2g0HE8^#Dw
zBEw7Rhf0F}{_nx=e9oF!zpFQS*@rn7P521#qnXMQ5je{KZ=to4=G@YPbgf`Lzsw<U
z{=gGjgYQUHa8Xdqdp9~FKmKX|Tb4g$+uF$RZB6TVzyiJ9GG_3-%k&rdTes^WHllh{
zpQDn!>fr5ge`3Zt*3Am$FS5ky*x#RZMbd{=)D%@BeFrl%jC1a<3H+BITJ;7j5l#7P
zltY?MPa)B;Ar<|FdpD8=-7fE2am2ymwSG8+Ba8An5Vsk&ohR9;sctD{Wolrk9CF)q
z!T!z)^>CTE3XL%9Fx`6J!I5(VnEhxltQ3FSLgo>WypxBXBF1g_4_OXYnXk6b&d#dA
zAmZHzfZU&khKBdY>LX`k8{+rxrv_H1v&+NHmrohzajrCFErSXS8(QO=Rh42EeqYk^
z@?UE=jWwlVW(oXuZbC!0doUq)uddM9&Yn_hd`}m2kx~*TG|KK9Nk_q7XNedY%Qag=
z@0f-4Q5-BKrm)tAG9ui{jh7EMbHS|iCPBt$W4fbj881^H`@m(P%i2br6XAW1iE(wr
zSA4(3^e+R0RfaAI9JeB3@>LQJ;bu!<jM9JeAri86a`#y|eiRR))!#u%8vn@FebH#+
z<Q(ucbR$RnZjq-goE$Mc3gPhi{!%O=Dk@M=cB)>MG6SF3^Eps$3vTp<f)NX>>LNL0
z4#fwvo=U0j0?U_2g>hX1A1^D5jtULMX|vgAO=3(rD`lJFcbx2TlM>+INE#yaEPuaZ
zfii|fL8mOaIQNsNn<6NpvxuO{GP$J#cTA;Tf1hlcENUFcb!YL|9onhU36Y637y2D<
z(I~dSv6xUt=!B?aI&xE6Rf-Cr27Y7O4DTkiP^;F2;*6*+O+WGLF?GV<s!QN#kq?jJ
zRHXC*m-I>u>*@HloN;#aH)vp;l{X%uc(9XS=;+ZoSK18SxV6|bhOr}PyC_n|cgS7y
z#IPwkEWMd;iD@%SFK~er^hIw0d@TY*`O0OHv=#z(HmaCda?CDk)J;I2_0f~_vO_q8
zQ)E-@tNx`<ch*6pc*6xdVLy+k1|O9qO@o+sPUM?ZA2WY^MZ&4>aG`Vxtz5B^*Za-T
zMVr&fZ8OOK&}8D_NFl442pWCCJUd~0&G&GzH#OoN3UkLEP&N;$`+%c4ptYkol2E|I
z{QyU*kI(%cmP91v-yk_1Ff$<5l$Zxv0vrIwmXzp)APmz^T!&H)phc)Ty<^?VO4yhW
z#$`wRE8b-gh3kBCK`DN(ykKSP$bN)D@0&DR;aISAcJb1OEcCvz%*n~Y5=J1uSa{o&
zpDFiIsEvjK2p#U1({r&j#U;Wv-<@aVl(R4;DW?9jMz@l*XWA!+obmL0>>D4|nPp8K
z{@b)M#x7(Ox;TWC!U+4@@F;RM)7Df|7xz<%4-Ye!tLfrMhn$CzvhUCyq~|iKu%D8l
zZDr2WY&h<6-{f{S3B86|^4*@t62^9>LYu>$64$_9HFS>olBtKA>(3juHmk;vgJ8tY
zl72ZRd2Xs*BD$tlQ`;5n<_2~uvf`Rex?d8UD8D2crD)^LM>}GRvpvdQPKt9n*9>a-
zX#-&^a88)~CTx2Tzy7e@ach+V*7T<>;|n(I6lK@Ue7bxW!!hXyv$^dJ^j}<_j&gi;
zFY!1!pE;&aRu#j{ivD3V14!k;^#yM=7*PBWMt+DcWv|?vv}4LzNB7+1eOcS4w>@xw
zSxB3$8<q{s+?2>C#(!*mWC@RS3Y?sr5NN&inaRj3*>Q-xm3qb(J`wSK6k}g()v&h8
z9&N_$(0jkPKAR;-R=@qpc{hkU?*;qU)Ka5?;X-g-b#Li~Z-b3Y)l{SY%L>9*I35La
z{r$r_VNFBfv`b0cMwMfL6q9kLKwT#@>!lP0Akui^)X_(2BOl4~#QnET;9Xepk01>v
zTE2`|YyEQ*$Jz%v)a!>YAKy91f*#a&$9@T`8ZIjpq$iPtB*_YGy-!_yO<&LAD+s`y
zr8dS$T}%Td5&crt*A=v*ae*|xZ2G5aLh0`$E5e}T6lI|!cFXtcL(<;fzD`+%K!Jr;
z;TyhQFyfSd(>5#{1SeaddbbJ0Ke%z)5_oEuha4z&0tnt`_Umum7ceQv@8UIds9a#L
z0`>|DgaE0QsS~G7mqiWJ!l8kIlVI0cllZD1OK|TAsM?VAC6#E1KcyhTx9k2WQ(b-T
zuYJf`$NbzQ1}Y`d^;P1T!bM1PX!L#FJ_snJPkh?iik+^)fNa0S6&xii3k1e@$TVeA
zI_p5c^bddP2guTTeQBLQ0%9>y%lLg)R(596k@VgbcAY)$Z3AP+;==zb;m_@*<e>Mw
zCl!{D*~O$#f4n)>@^hcwE6AMYN#FW|UT+?H%Te!Uv|81q?u)zXU~DL==<DhJS?S_U
z;nC47%P;GM<*U2I!QXJ!%g#Zbjo3mDzm2XuxfrL^Z9ecX=^s`|!qi%)*Hn7EOru&G
za*?X;$Kd6#Qteq1;?q;Z_}II3R?ZpbUiCk5)`uPG>h^XN6<66|FP%k@R0Ed;-b!@K
z1W!aT5i6=l!kXlh*Dp}ez<Lm$P;FyCg&6m6aLFl?Xt>k0IneLd8jn>X!oj)2aUTaI
zR___AvtVk!S>#Mx_)GAKEECVeBf^#5ear2K9I&B&{G&VM+W0|@^Gqt@^v_LFiU!iJ
zTbilkezwo?aQp_(;1*@Wwv&7OXr@SNM4&$3qxykZTydX)PcZ$?TM6I5&Z@=WH5m2r
z*d^v)W3}Hy)G+l%=O6zqU6G{+OKe8%=NT5x8V!_9#zL*;$wS5lD$YjQ+8J)*hBky+
zm(IoM7P&~*^6d4us8IH`X0RYdh)ES$a_~^8CLNVp(?j=KO3|9KTc)RCxR5e7oFs$-
zqaV|GEY@7y9u{bEhJ%qcn~qZgQ|S|jYjodu*|v|g^uO}csen-Z+{BBh7%@Ze|KN53
z0Yd3WK{mq-LBOjW!>7rR@=m+aU(=@W)Fy!_y;huGcY0s1YF#^hA$rbt6I?{wze%GL
zZvz?76CEB>Nr2%^ewDJ;(rm;(Pj}Du{|(7(zI3%PWrya3EDi>yoqb#GTN64lTl?_x
zGgFYMqVi;b%8}G^&X>i{bNJ_Bz%M_!ATXYE8NP+@Km@AIIVdO57g2&DoO?KuQN-*b
zPr(J*0LkD8f1pcC4dUNqzMa|Cp=by9`_L~kKc6XM$Km4Qjz;3L^@bo3yq1=g_0fL2
ze1`SuK=7gnkv9MmlYHOp^Z&AuQKq0)HY89u3^+%lc2bFQo9C~7MO|it3HG6flSYO4
zS|ss)z1MhrSklwuw^?syXlZGQ|4ljwM?ooCrod~1>sTesKT=ciw2~k98xU@qQ1!za
zBHP@;^ZV%#rhLV@anxxlUb5XSH%a0*svt?7zKl9TnuDAOY$h9(9prpVHA^G&-${uM
z`TjU!5v2omRww|OaGN>zG>lhER5$`5u+7RPeyZV)z0h?oH78}+^`LDr@)~i^?EHlH
zd+L_D&ND%7`F?X+R|s^!<JU>Ya_+dsMoY9f+l~iS?X6C4Qv?z@Na4F)TXg-gj1c);
zSTl4a>t4J@=M4pI!DS_k_0P)SdJ&Sz7vXp({R4a{8#%d2QILOV5MW_Zf=pMIL+#x*
zCGey@H`y4y6#Jf)oaFA!KO?N0@zTYU^TWO5r{m7-d=Bu(%?2UI3RxvuxW{;W|9eoH
z(L@V6&A~rVZu#O?z2k%lM9G&0I_Q@Cs-YN2zo6iQFo?(V*hkj_cXUQxHK+7>{yUCM
zq+e@v_*_LlIr&~M=$Lg4a2N&5{5s{QhDsqTy&I#-gBr1bI`gvznpWl8q5Hc?$N+ay
z{k?uSC0YTY=*|bqGhNc|2}Ytz5kv#$pRxTV+4aj8!$`?N_PA=llk>_E^gcu&@c`eR
z2ngCEIU4Gt3a$P%RI=UZ=$O|H-G+cHKZy!|pF?uutruh$55ywmw)=gb%3`Rhf|XFg
zV^aHN-IR8h^(F{G$=<-g@CPl=0+LlQvk8P5u#{v(5NRf&%v3G~n0W4205}NlE-5F4
zx{B-;GLBmXZFYKtoGHWV*-3ZXoN$H6jn1Y8k#2IX;HMu?G9>`4e>)~)H<4Oht63i^
zGE?fFjMkNMTGD59uAgtaN#jU4fPQq)B@NH|<kgzJTmWWYQ&daAu5t_gDH`)#9wTPU
zHV2Gwl%9kQ)I=FcbaQGiG~n=4L)hp^pApOTabn@FtQUX8Z;09JRCcTdLsC7rouFYs
z@7M84l(@n+5^D?$6}prn*kcsTNHr@Pkr3f#L>EjbQDYup2<8)lwkHK-Ha09YiHNe7
z*ZWhQ=@g-$%lR@QK9>y!$SOcs5?)>R9p69_(wJns($X)?($b?b^SNmE7L<Ts#~nz=
z{QR}242(7RX`!Hzbu(;o`A0mkm=Pn*MRiMdOx1|bdr)0NN-sh(U#T!2_Mo`cI5ibT
z#t;}neScxl#zR@NhyP#Q{juOK$4$cHoEI7!FCM!tdvT)}Bw<qO-;{#~>fDS3;C8wT
zkF`&ui7hSCpJO(9|IG!D=BE!%;m|%osLALM3nu!1L0sV1cK(bC9VK-*p3ZBmo8&!H
zA`uY+8+GYxzw2)Kd#b_7<`f3)$;QZF2F!J&4@~;jk&yJTJ(vQOjy>rW6=1+h$-;88
zrcq~-<v1tHbx{a}kV{zEl5lD@7&nS-Vs9W|h|D8sq9=m!T9^;UQ8u10V9H1LjC@sU
z2=3=P<OL#ckF5Mjbhvnbxr<RaDmO-rOJ)5n!Sq(Bs874~aaz!P-;Wjt<_#=wzm)}l
z(!uH*{>+O`(%L`<SNj4PAaxo@5}2CKSQ*Eg5%(UEE>b{@oEjMt%_cVbhDS^KkBmgI
zw@XKmwWuKMO^GW^a|~RTh~|iz<R0~Upr*d(5A8=Ss7y|RA@aHwRZ*FO?9M{Gn-tO(
z30~XY4?JWbr9&rz-`L&P9}&v<LKyo}A;Abjk?in%`h34^yK*ryGNOhNNbpz15L!Mt
zMVP|$r}_^d2^DsiC@9KiehNzlhya>?Mlq0rd%0)Cfpc~D4(gt{5x2TD3^rcOlkv7X
z2KxWt02K!=2hS^wTSE_Iq<GF3htEzv4lr~n-{JS}h@_5Fe;0|1pH;hDtOkjYy9f%9
zc8={>&sppQUQ2A|P{tS2Z+8J%^OH$@pqSo9-FAbdEgmrl4fFq`8GE*v-xAuNjmOmn
z6Z5-yRMSII%E+R-O1OcO_0|7Wj*g(mjvGs1>YB0Q-1-i+hF28j7k{B5ABp6bh=Zs8
zbMG151pQtj)=H4LtY6l~OA%$Zk`(0MGheHgRo*u%ra2uPTwH2a>&>YCOC7hs6NcnV
zF1;U=<X5BU`<kxjam7Ve8qQePN=jM|^b#EvzQe{x$I|0vkqd>)l3`_MCF#-d)WG~>
zw4mQfT+m^$Fux)TxH*y4-fo{=&b|LtSq=Y4u5P}I31Sf6{k=l;N7Ev^;YT2fl-Jw7
z&|oH}4^U3AyU9I<ONeE)f6Wo0#lTD?E5((7I*#{kM!n3FN2z8O4pRk+%ex%#`)4kg
zAo&YS#!xD%TdY+wIMxXvqtb;)-n3ZBvgHf6xp4Bl9dT;0r^27j@H7dG`+z$wRC0Js
zM9IAQIE5M?miVMu@zO-?yj3``XkNiqZ@xA4*2?e2(o}^}6moODE>9O$5gHvE3}kDC
zk+iERaV`Z<jVgNDK_~{d5TBLZb+wH)nEo7qqrnrh{no!}j;8?BpMTg2r1VPXbo@M3
z(9=A?Qke5{Q8E9>3G0#Igirnn`0C=Imvx1bE-DH$g*SA&kD4p!DzcvwA6_ah3Ug*{
z9&l3jRzyT9cF-=EJHaqFHy8K?0AwQ*&g-u@$9A=zO0pnHM4PvK@?>aC3xIPTDa-b1
z@Q9+=#~1XegXfl=K6HJ)p4Ean?K?7mt@V5$rEA2%*$mL{uPwBx9S+)ulxA3L=HsB$
zQZVuxLwGm}GJ6u@8-k2?Fe3hAm1qQ(fL;zW8cAUDPE>DR)bCY&x7tPGLMJRNEJh*1
zGY(3ilt81-?EqlsDMMT|XzFDnjhZ5{*?Dy^RlUBbe9>`hX>=HNp$9___$F7N&6#a1
z^a-$+gbUA)U8eMVI9o&?UjHQ$Q&}&65=aV_UWAT{I+;omF#Pq*=0iu_y}LHjnZ-y*
z(LT5~ik*(|VT69h`cXp3r{{Ak*u|hHX>uq>uZn=Wv%{^!xaoG}&Fg$3^Ry{|3dl9b
z!-@!!BjB^E9O?HY^6oWUX$y-|;HDaq@jbXkoYv!N!JXca>#B5VJbyqtQRvSy{A+-O
zVKJ_!!FymPLSDaHe({zZuD&qECo#C_r0tL4fAlNd?OH-(A)yodVqCS7SY5)4I0@=G
zZuCn=lYTV~avf|PntZV}bernrs5b4du}g*ToWYFZ4VxOQT;<qi@N{@xqv*l+iwMTn
z;n(B8F`v|d#|Yg-bNa>&p3(&bt^Ez!s2aFULe!LyNxOgO<CtA|0@ddC7$=8^ahUac
z1IH5+F*@4c5E<s_R$jtcs7}Jwkte$TDzljmP1-oh#m_~1$7~jDClq^7WqQ}T|C^9;
zQz{qot%!O`;Z<+8%hSy195STzouSMd67|GfDCZ3JJ~-|KK6*QIcy#j#2?%`n4{D#e
z#V5w!4jL{EY_-}s;~IVoyj?-2xIg}VumK(9G%w3xV%??BxVO)CuKoI?=JwWs{IdVi
zgoLjI+}$sC6t8`JGES8Vc{{f0G$cj8YEF8}?n>VJw6k}$k%($aZmMTxY^Q}-`Q%N0
zzeacL{45{rGMrzQcE4-DK8ba5!yE9wAHYaStp0t8eIK?lSQI10BVO8?$D{Ui9k3v7
zP#*?DR#rN<LNtPGBvq3?cO1Oi3hBj6ObB(AD!*wOpO4MI7*wAc%3716g-OS0vluqF
zS23dU@aTL`&!qgEveKuDH}(7ZF3NOycQoS7CN@m<q3?us4W^(OAsY}Bwsj@Xm&&0H
zIh9T2%Cl_+*Y$;oFXt4<U2h%-!Z`T}W8Aj=U_MnyNRi*a{UfUIi0?y_(KJUSMf0!8
z(%CJunR*M4m9F~M(ThqrQ)4)k5)l!Rbn7Nr^7`38?i3WlEV5sd>v)@7V=rsnU$Z+3
zvvh=!ILrlFOkUA84-*~nv_v4>9Jxq1h$<>FqUQR_&i%nOpusy|O9RW_!LAB1?LSK!
zs5<|-{FF?SyV4b?PPI%1+u6}~I$JFiLxtOR0l9co&&2`@9Yn>T;%{__Vy$A&rUW}~
zfJH;8&RS#OvgM_5{uys;B|2_KU?ogbhoyn?mjV(7^s|8m3kWlLG5?!UjS|GrCXf-W
zDgvOxh*z5J7b&eyU9HW6c5mHoKOs9fYM5Z*(Yeoxr<@Jkrvl&O{zd{el|!=u%2+4i
zL7T!+EOEjx&_1cP3Ym&){BvTX5IeuQYn`n)2$0Kc&8O1Xit7dp(jbJp5yvX4=!?3`
zzlo~#WdPWjddsIt%EDb0_bdK9%+a-zg1|!UHkS*xm?1I;M@LZ$3yX;LhGVafWh7!e
zj1Ut$SegeyS?j(UBb}np@+GEBS!;3c>SfX?0m3s-jsnXyA62xrZNw-z{kk1`s1%U2
zrv#M4^N505ne_azQ4!V1o_7~bzRmB5<ji7thCpjt3z$+fgu(*@l<ELi$~oz=5Wr<;
zD=X4m$#mMObe?06yRlm{Vt#FuY$%UgTyOq0Zy1rCc0#OcIoy|5QMqyBS0AU`B~#%f
zw`jL?t^rh?a@|?R*LeiT>g2NIW)R3GSgjr`)WJXCHZW>KR)|Ry*`Ottn(q>9Q<M}|
zHxrAeWWJ5VYWx*vaxR6<N~hEnt3|`n9s6OkEl%SuFSWU-EZ1dQ!+nZVUK}!-MS?Y%
z5T9a!i^kxDHBUil`aNC_fXncGY9E#6sVK?R+>@Vhrqlbac>A~3sI$EA$YL24g?K>W
z4AFB&@Uy~CCkYkf=10jwhmP^s*w|t!xlA)x-mLsW6MIX+4=jr|B%F8S0zeyxA5(X@
zsCS0Bm43EWT2mwnv><fZS40%u9np<QWVJn+z5hE(WHu>6K_wpk`7r@D;Q5GAzN+9c
zSSzGWCyxBx_%;xSUvOk>YhR7WwBg=<vV^&46K|;SSMQ9|^@~>^H(+?ZB|^_v?`ImF
zdKH&Y7xA%uWEf?x`T6kjNEyeOdW(nbfS2k&Xcd*~SMroh%t4jCd^;*QB#A7mQWDS7
zbuY*7Tv5>gDTP4)3A}QqF1Hib&JQ2PEibN+_V?T@n<o5(k-)VJ{hzP9w3cb(lukP0
zu+VKPJ49;zc54+i?>1Gm{=O^w20f#|Z;#b%pc>7?xpvg`5HphgKiWhXHczbaB$YHi
z;f&sCi#Bi{(osZtg%C@kxinf|6kxi~Mz$ZEjwGi+{H?jYJn0%4JfswsWEp<lKO+#A
zY(F-Q>>!Fwe6$N;<U2?AB#pVjT4K;{9*s~$dV&Wni*?EO@gFo_G*C(?AwxNmu;QkZ
z^U(Q=>I1j|1xkvz0sY7PiZDra^(l-xv>>j}A-Jpz8&aVbk&|ETA@I5S_fq<z%#fL)
zD7B^T^GEh5g&(i3=S4$Pk|#w*Z+|dxAPT@vMQLrOHFl`oP)e!JE8>c$Drc3AKApty
z>qQi10c*9TW@$$I1xo4>jNW|2**FM=wA1^x+%M}K+PfkCLosaR2yEG^*3Rf_nmj{e
zCqplD-CL5~@SAFf^Annhj}+5Ei-vo9jw%!y003Z4lWD(s^o;B@g6Mk3yFZXgzeOSP
zzJ<<1f=Nqj<Mv!O+x$94&rbdPk<nxY*=@zDNJw$N(`z8;(e}rt4HGR$%3s0WYTEsy
zkU#s499cF+f~H0@AdV*@nSL*BzE~l!x1=WJ_UN#JIy`)&VAr9gdFRTt*61WJO0I51
zL_z`<C<QOY4&37DuKSueP<CCISh$o_C2VKKYEm-=5gTOk$e3{?7UF+Op7%Q?X=wPS
zPEVHOawBvUZ$`W=J#)0$V>9c8Mb1ooQ2g^6kkFo;M~vsh7c<iYC}h_of8+Z1b>*Vv
zB=Rfc$X0>omXB?4bhYtUYwjJCu~jn&^ZGEe$`RC_6p7ay+cBeg{_JtjUtBIIo=kc1
zX-2VI;5-y5s^ywJaSCVDX{DZ-nL$>%)Y0PMpw1tyNzL`%TST+ON=@u*Y+gxWT;}Vl
zsq&lpUoF5cN85(rNe>K>lk%ffW1{+qoLZD1w1l|@-}iK`CjJ&ds;!mU_IUlQ*DNZL
zy{$$8p0l{t;L3!};|LB?MQgtAcx=q)gIYr(VPPpELz*(XDl`9522f4B?Idzy19P~V
zA^tVB5xMXDdmqB8r1=MYs;E$CQF66QVGG=7^CCvF=8*R^gJ)rx5fOvB!XC-ubr@M@
z0du;D2P`qt$fUbb>3QD@>hMxQqD-?FBocR@%r+|`?(h$UOHDg+h}GDWk%*YUAXveY
zOyB>WpSivZnXUPSX2{h3XXdY1_5D!JGzX%nVt|W?80-G|&!lqflCp52+@0IgOjQGu
z{H>l`3plO}7xyfA@A$8ETi;gO=hLBhJyL*j@FRk3l!R2c7=^Q&kocPxW2ixI?6q4d
zYhmM4%K*X%tPmlbq;pgp?-V~9a8X^sc@*M?+rO!#Ma&S1_6QVuIb{{>@`z5%H1uA-
z^SB=*AeB<!@uwRW!O=3bO8ErmNSnb*1W2Et{+qbHn@|u%^Qff4(aez8h3Dd<^AK1d
zp`i7ywtE~3k1DeIFMLD=c}yzH4s9?B-&M)D%1GK?2DYDBb6M5PQGDjC(rN~%(c*X1
z(W!|>oIz4Xan5aJsa7wVn9C$cct&8EkCwnE+925}E?yD3r_tURnO7WmAI?af#dcd?
zT?%%m#-m5Q-gOt&TRuMjPU<@`(jpux`cnnUqz*Ts%%Ewlc|U(<Tpl@j>E6S;e_(L?
z0lJGa9OeD}?CSk!!j(4OR%&c?8X5kV6Z3*Y7!Wr22thQLFtTOdnIMELP?)&;6*1#@
z?jO~sFxbKGA6*kB0|q=8Bj~WtiWtycLVxrU8ccp1Vp}2~U5zzYG1J|BFnIDWeb)CJ
z5W1B8)hXX#_5JPob{xfr%%chAM!HI_hntk@<t@wEWLBg@tpne#wS+@rQAx(dqvIrc
zEg>sV5%fC;BU8XI_5Gr1m7GL->E(2@khHdu<*7K#EhR>DelvQQjy$D<8ECS=IkeAi
z|I!0E_$J<c_7ML3{M<%^jboXUE2Wa5U|D_bCJ>9J`fk1#G?njmo`tiSN$~UV(-1$q
z_~g1zlW*#sDSgOf;QBHXw)Pf36>?FqwrqV=udv!xcc6+h<6`n>z_&EbbTy#Vq=f~(
zCO!bQK{rH`m1{!1LuhIw1EdI;Ar>ji*R=YD=S@<&H}MHcv%+FD96K(iSA`l2yqc{`
zco;ZFIVf%E;uG?=w8`uA$QWAr8<2DsERM+6CLe0v6Am<W$@AhU(kV9ecsy;D6zA7o
z^9lzOMOCi|d6)HVGosJ&e&HxIHVzDdX&UI*%)Op<mP{K9TJk)8kzR6QmlX#CROxt0
z3eQnUQB2iM9U1QI5*=^&qNGIQkL9w#pus}XMz2{n{vySsW?=ZG2&J1tBn!jTz{*}|
znZj_9prnc*YW?+bH8~ZCvd&BSdXFH7%yBVsXUXK7#nh#)T02+(Yi*Vj`&A0~kt{Sz
z3m=b=l~UNdF;A#1B<SoHn#~DRJ;CI4Y+mPd_ghw0(h;N4U~>#yUj(a*XYk1`4sM=!
zn-gJzHh~=B$fEv9+TuJ&tzl{MI3HvX`3Jd#Bzumyx#DBlFh<45qU&kXlf|U&#a6i5
zapH1CMcz!Ui@`cYe5Z%Y;Kg2Np?S2)RNSHO*x(XJZGJnL`(=RksRL_>Xzd^G#8xIj
zj}R4x!^_Kqj9^0sv?2*FbmcF#@~A1vc(Ig}56LtZ!@=sF0GKJE<0LDZ9bhqg19%0Z
z&Qc?hyYiTk3gy&5pku~F<lj>3(ughryF<zF6<tS;c*D@GmK<RXB;pjA49*JCMI?sN
z&4%m9xweWdFZXftDWABm?WcxKvyi?ct$MKy*Q+gqd?z@$CPLo&7;iPJ=FyV5Uuhrz
z>V|^#HaI4asGi3eL=@)1zw>{vzxxP%HM_!df$&726a-9DC#2>dIwpL1b@jq#Y4Fxc
z=%#Y-20d{BU_Pv$0+f&M_O=&Z-=qLc!~hVLOXEe8KCpz89#iG#F+|Q(2y=>QGND|C
z*<ABbWlZ16touQRKT+_LW3_Qi3W#WZ?_7x*tI%t*>2kH{W<swLVsa2bgW%RH7nF~9
zl7RR@5Bu508@#M@{gjRb%iUgU7K{4Ch|++KA_QNs9}I&Po|mMT(JmYb^w%M^z=(sf
z_Aok>c=>5P1cD+E3MNx<IBqQp-=$SMD>kb1y2fab^_Fibhoy`$e<c5Ay70K3&IpN+
zfb~GxdJ**hD&#6ZtSRd2U>YK#1R-NSJ#1eQl-BN$m%i3yWuxomh5ow}m|QW_)6-||
zx>eLqQj5NMd%tD_;an`}K8zLgYF%A~3|*D*XVll(tsg$y4HVbBb60-L@}Fpv;B`@I
zxNC{f7)PuxGMdU~`>J^rgpqgyY#8assW<9OW7evmU7w*)h|JO6Zr@8_bXXchIV8KQ
zq{-2`gcNp+l`(>93>k^NZ_vIW9miCjVUo`6-NdI5-D`QjJe-GzfdWp^pKxTgTJ@IS
ztv$pmjD*A`+v6QI+xv!NRGV?!#Bsp8vbqFTmXEy@1;P)k^zYbe(+p&5!FpVp4$koV
zgZ$HE|8cJG%u$Mj=!#Zqn#MfOqQ}Ry(hmz8`#-|M8(QAkswCv#1%q9Ft9x!x=*WZQ
zs@@t0I=Mxkt5^}+mBGv8!pFQ-n~hSb;6?N(t_VG@!WEZHa8r{wBst|5&xuYfs=Z6X
z|Hkt97ZW|nI&`b5SFF&%roF2>(#7Pt4f4IC%2zCmS5{lu+PB_0AyS$?mp+5&epro{
zf4KjYayuB4IK7~*&w$=({ADUl&^S@Xyllj(!we$vLPwn`GVUH54O}P;+)iR>Ah1TT
zPSbcblPfXVIzIQZ>YtbO@T?mH2~u9;Hp?5p1ZC%v31c4cB9Ly9?fyK&Dgo5n5+@}k
zB_A7S4w1n0KGXqH`9W_?5ylIuF`rj4w=Ujyb$4GDSu#*r=*Gszm}^U5YI^RtTsppp
zoc}Oo=Rl3Y`P69dg$vtD-q7<}($d{WJ9H0Cps-Y!B&M^N9UlrTYGeexp5+L#(9KY=
zY+z83!<EqGc0h~}Gz}aSI~J2paAJIp4-ZCAu{j=W^;)@`7w@NLdcpywxt_c?pXnai
z6#@75QoW_`$jzhM=f8JbhX@FeUW}|@6M<MzjGzQoB9g^W<>C1Bt{#ts=MJI;QqlCA
zu2xEn*1s-p?^OD4NRA7H`<HK>*fogk6{bdn2}K1FYAEdhf*mLbT{jc*1WP_a3R&%_
zeQ681@2uQ`C7DKdATF&58}J>B*b=VU5h;qD21TW7d03C@M~jR?sn$-+#lzBjSGmwF
z-rJqb`KXPsjEQzbG!kwOG_-MS?A?@*Z}XC;1R~ebg>@cHTz&qT=ACd=i?|luZVprV
z_$UL`5Qpk*%}}(@c_3UVR-PuFfFSS`FB{UAm?8tzJOC}*HV@_g)4pva&<B9kB0y57
zw+~fhP-buNk>97tGOiA#h1pTJ-#ZJgJ-u#f*11R*3k#=d&Lzo$sjB3jpnz7G#wzgF
z$CHV@7$~hXi}c{7v%l5`q#29vgQi3iyL(TY`x{)k(82=8x1gX<=uwMkMmO=xhgzb=
zj-QhK;-d4_l@W2{PSN9y=KJ(PC#=HG*&1^H6>@#Fx|POnZZPn}$CsIF*`nb$3!IdN
zx-<kJw$pMbT30Z~rBXxXNW`P1gIG=bYw|sx;KE?yUZzntT;|-({Lts>M7OKJ))XJj
z{?|i`<^r^uirbIjD#PJlrFZ>1o>7F^Fpo>jf&h-Yt8e&}Zg?Xb&WLf>{gH9+{~C$w
zK^?fDKTDiA@cR`c(UNk`tw^C#u^buASNT=><1{X&!#)xxBNIl7UvXJqTl4vGr=U1r
zCCSku_v8`ss?K4aT#9i)$7!eBiOtW?^wGjUj9_MPYij^AlgP*mLrF0w_1X6#&Rw_b
zHO>o-n5wj7;2Gbm(Snt{*}3t15Z`o%T<#$)6@%aO^FAkpO*oPg!wUr-1!EQh4DaQS
zS*rcgW{ISQHU1L#f-1(nbbVA}|As<Ll_$$2n~Z_u_Og`Q;D*RlZgjst{RTDjcy~6O
z8UNhqFiEO4Dli+!7?QUq<12q{UAmZ1xVa5z59*fYLDt=~>PFa+c$W2ex$9JsYW~~f
zRgZ0_^xtrWvS|XvunCe=@Y3GKcyfq=Fr<VNZuy|XZza}2M0n&ZH_3Yh#hTl?3eOJt
zuj<N3xL>_|u5$H0_7l|-;;&`E-g9`$1Xj*gouQtjB})~+ZRwqHCx%Rgp+vj%&de#(
zBo18`U}qC=3T55HH=Vo6uV}&9@D8$R;s*&)sieq%$*#B{Vu^zP*T$C~FE27E*V(4B
z0Cay<vrQ2pvb7xmt+n2^zj6v3R9O{Viy9gAZ3)lE0w6`Jsjwa3>5!UY5axYsD571o
zqmBlNUbMmM|8<J_z@77=)O-5+OuJMhZ%B&5l3J?>d3qkHzRwx4X5nLF7nf@`NP)3G
zU04Sgh9-XLz6|1vYrU3>I=-P>J{=I4R_)IiAX+=>2Gl)9``<oG<PN|~){PD`hrHU!
z>Hi7(Cygt_9T_nNUp&u{q;VH9O!GZf3Fo(24$3=tn4}6CI0VpH<q^$h8q$gI_g?ne
zUbDRsIi*UENnL(f!x44cItSGm7$)u(xGZZ)aTcApSnT@Rs0nU2jk6a4#$^X%GTkJi
zSZA6tdS3ME-qr*qI;Z7CMnZZ~9yNptI0(sD-HBUg=Zv@B3AqFp6#m%^M%qEzV<G13
z?sVBkZ=$Rnc+!)@*_zswpQ!lp>m<vnZ!f`W<!qo}mSNl^UDv%sm4Y1nM;Eg+eDWf%
zF!RmlOTPR=N-=Um4kU;54mG?N`-$fa4nr>zHM7Matj6BJSr3t@wvix0?7B~&yOd(H
z)gbkY(+dJy@R;aF#0eF76GTO=lejD_$*9&VI10$TC1rT@FiN$EqmMuE%4vH(=Nc5Z
zp<Ek!;V2mGzA<6-$+_xtsx0Vzw7R>!KJr)oI3%Z)`RfP*q<I$<y>?)=?Zj})AU613
zo1bn0gq?yp*Rf^()xOBcHK<A1+8dp@7fCm-Bu&kaL1nIYE565qq=9B{pkBG){(L3j
zaPve=(cW(-N(f8h*}VAhxw`a%JG=ay?K-m<*QHI7{&7!P$bkWk-=^s;Wv@opkT|H;
z{ya_JYVGK6i40YL*)}p1U%R%W50n6}I!|#-0L9s;8NMMUt8x@XXoDrirPO35MqaH&
zONO`!!FIqsX?b#PirrP`$N6Sm%KJOS?d>fM1L9(cvPzO10WVljYkXm0VPbO9jzimv
z<Cv<?hc~pw3XcR_m5)~USBF-pXJO9I@F3VcyIY$eHN6SrLveQr|9q#d&JQmkSMZD`
z8^f{23#;{eDtyilNmUml&Wxx(MR6^T*lB#O_QYH%{=EQ~tCg`}S>IaEDP6!h7Hge1
z|GNiLRae~9%0LJw`v%ZIwkN}1%qUfdqBpNrzc_`DF<cH;fH*N!0W(JEIr%@RU#Jct
z%}w%qYhsAU>8#Y~m$Jz!d#QXjmCa9ML<J?RAPvP-=QUb;lLyj0zz!#QvDAp<!lWd-
zRw|phANSRZ1x4U#MBo4x`jUWs_PW~K4JgOb69{68Q#T?R!H+bb^dViXI<6Q4+rDo~
z>Cg(=yQf9|Mq|H1!yXF1eIAYxe5HniHQ73P*LGGYN#f19)yZDfL#|7}LEnCq4tGBo
z``t2K@C?SZnGn%e#AYCj`=qP*r!s=<GIL3E%7|&~vkpJwAeK`qbSCu~zz+m8|1!N@
zSP+bFSk(#1t;lZ3ZX!)hEr4n=L_<T9d-CX58*6IHup}Uc)}TgBw+dQn+>`0Vf?cN;
z`f7jdXiyzyU|YP<`PekKbq!5;Vt>>W$ERe~2b57b?RVjIaF$)V?v1TM06!n+<b@XM
ziqz-+BAnd6QlK-1nH(OyoKHO5&52>BAR|(-Di4^Tw&&>fKlki3^YcPa3TN~q(NuZ7
zv(EY3ouX8Ng#ixqQG*Wyn;u(21y%<6u6)O%mUIV{*m9_ty1-~C%F6Po;pK`GRQ@fo
zdX2m78W>)`e5hAoopa(#N?xRW_OlcET2Nk8Kxk05C{}pT5$3Dol2&;6bVrT-y?f25
zYd#n}qh_SpiKyz#D<4;>SxmNGdp>A7_};dSCU>z0BdTcEqAfua79LrkHbwYeqfgfL
z^~!dXS?*Yf)=QIbp{z}V7AA2V8(%8||8WFTBL0O5taY=-HBmvaP{R@8&Rptb$W^1l
z-}*S*kB-PUBeKk&S8C|f`*$bN+tZIPse|o5M|y{BEBjv!m5gT?x4&RFNn_G8`*s+P
z7m-24;ZM5FI8^n$N>aq<CP<;UC<2CpUH~4IM^A;Pqj53jKn9;y+B7mT=36dI{1lHL
zKWA82Sh4bX?_ou`2lGAbLPDZuCt;Fqm%bAVvQCbCjGSQ67Ai*EK?!__JN?-&V^lah
zRVFUiuZqym4S{wJij2g@1s6igI<1PmPGeQ*MybJX>;An_m|Y;Df2W^tu7)QWIOz{b
z9^kxUN6+>!tTsTca`o=zT~&*(mND6ODf{NR#5?Aeo6Ix%xGMD`YO`%2r3-rF%Hqb@
z7c9R|`CK&!#EP<7{zUcrQ{cF^{UAs-7K0LtQh56?B{zAB%X(QUoqxreCX!0jvZZqI
zSOQUoWRFT~JLQqXu+}(NsiOs|w+2Qw+Tzwr$u{Ip0ajMut^ZQM*~eurUSY@y6msdb
zIZ<*_2G7m{u?iUa&u9uA^P!8IqDf&PcFAO$Z@KQ)<nPp5?Ksoc7UMfy+0xd&i1{qJ
z;!sgDXEwt7c7L7XVR8JKuo$$c*>-4A?{vTnf`(S7&g(8k1^#l<Vc`XuoxM3X<hV~g
zr-pnEUsqEtO<nsS2}&^K>y*QZb`O{a-JD~)XU=Z|Vvtlpzk57gHW05_7z>rk0_>D#
zz9}#vS1MveF+}3CO$3=1T39xFXtOL#MQN@d^C1-V4%SUH#VLaY@M*fr?!HWAKI%1N
zZ45P5urGNkQ5Kgg*;5~j8tH!OrJ?4ah?$K~@-&V^!<52|HT)6r*n~qnB(7Ppc!>!E
z&!tRo%<Uf=S<>T92NkFm_3wwT91W2`y698%kbKuVb89LW<<WGZH{y<^{$u)p(KE;Z
zBnz7{y$_gb$_XR$9k+R-JCGA$Nw)1f;<Kb}_Us7cj3fh?xH*$@mKI9s;cGeSZCApm
zS0Js+!Lk8wrZ4~EjZp<S@wERQ8Mv~JZN~awWRc{mNJeOc0#Dag-Y5py8y~D${8AU)
ze&Qsj4G!u^`#~XNjNS)w{6<j7R{9OEGr3*giwOX}Yp7w*7sL7m$9gM8{myi6>w`xc
zo9Xv#Aq({Y-HZL@8HVjg>ia~__QoBe$D*s-qnCJ0Y6obEBHON9>$%?Zz=mvH3qxD0
z_FUS#wj~E_$@rV&BRyTw>)Z=~)Fo6rtSRFUj?`;)U<_1<AHsO7T!<7*QOr+Mnd>v7
zAUqZJ-%klFsST~n1^0`Xvkjubz|wU##`Jp?WBRS#!+RT-U9x}-UV_~Tk)YDZbE5FI
zVlt@vQ=<yB)1|T|-*^3WeV17T?u+UG2vK;!^j6{T*@8fuaxb$%yB3F2Z5HM97bJIe
z&jB2`92wniAwmi4DbfIEAxm;~P78!<7Vyn%7LX=NPJb4=*CzTmE4&hqgZHoiY-JXY
z_NQH-eM7#)qr^`d6tyj$bUL6ErA(^AO<P%67M%~xCeXS10~YRD-({LCAdD^@$ppP$
z^v}a`qY(dLs<Ih2x=j7eYd}6LE&H<)_QF<-)KyA8^$1v!hb@;@Rb!$-14^wah#!F%
z!3uBZ-`l%N;b<U{otVmKz`cn(o#2$Nko_5rAHL$EeJbuzyF^k*KRPz4>&(As$zFhF
zQb%1<_Ittw87Tt*;-%lwqJ|ZM8d*%jrJOH2xzk}o049ww<?nyK#xIjawVpiqe%M)(
zhm{qY4N3CR9kyN*-;Dj9xn_o<6c*hUL!*R7P))8iE{?H*!MNpeLlK7ZXC=q$=^sXc
zLKnSQ+C{*`PxJjoy|({Xdlp`pOr2E%l2<w@rE_#>2m3oUTVK)dLWwXDs?;8Q;$s?q
zd3A}B$EQRR#b1Fhk8}u+hgd<!cWTKFAJ-B#KQKqyx1{a00_aRJp(G+XXTHe!a5V(S
zDStkZ)0rdy2Qp^Mg7)uBrJnjC1_nOCx)ha~5F`sQ#<N-#G6lR-%V%5kU(H%Rdx!>3
z(leg?5G+zZ>2+>>AzmS@{jzz<sdG=`NiiBr;k$&qJI6~VI+X)~$&&|%u-OICa$HI0
zA0qa4OLGCnJE%j8p(sjNcG&)D^P3!{qDmV|MbuK)Ujo|@PkvG`A__Q!VX)%Y;wj>c
z=<}J?8o9r`rfn2m!Pnu-L2)~qmE>!w=WVG4J40_s5j5@l&E6U`mHx5B$;6Ocrth|!
zlSQ0GJl8R}XOERWU0Zwv24{?Gi_4Mimm1=0d9K+r%V)=%XrLAh%xAS)|5wC!MAZB)
z9)6Op99kowWSW>;Xz!`DO)tSu)CRgCarMwBCSUp;y_CUi1PQ58C!ekJ!_F}SQc%2;
z>SpK&KSuWn#9stT@hLtrIC85lGCr}Xs!J%IGF1Rj?lBI@bn7C*ok9Afsv&P0X$vU!
zdhm^NI8>yhV!;te@U_QR90BIUdX>WR4OEQY8dO@HH4V3qHV7L+OlV;_{cg0hR0kTM
zmQG>^-Cq?1$v;$+32&G3=-AZN7@?*55-_E8W~TsqA7_@A`4v_*;z1d?J>G5e(F9If
z67m6g7(<LB6(LpprVg5MMS>O?Y<}Y0CB@8IqGroMiI<6)O2sUiJ5eqSU2!|@Y?}=K
zfR6?`G;1&n(J+|qELlvdbd3+og>PE%q2=RxHEX~MlE^oO+Afq{LB~UxA<%9y4Tng2
zOLhow>$nl4S!Uk#ABZo&aPyH3=`D3uYsrr@<pzMynn^k?5o+^@Ns$i$LNs`s_h4pe
zdU@36^wo$)%siw7%^bdG#UL>a%}92l)#fx61^EDWtdYr^Ig#n00a||`3=<?z5S|8f
zVI)3Aa9Q`O4D!*veJ=dZ3^Vkxre;XXMo+?vsF%{$<=;P<g?W5*BFIEP$k3l8qgZ1;
z&xazyr`nxm=BIshtc;FeLhY!_*H2CQ&GWn(>r*#JI2$h>?17#L1ID~zER^fI1SI>_
zam75r7lPY!WT5S7n-+%Tba&&hEkEa4{-kS>Ksd$m3d!jZBdsANNbnd2dg+B!LRrc#
z@_~K$LFbPpvfrp1QLz?m@BKxI1K&0V)$)<359o19*x{ix2>nhNi$BjW<2dZ61F*aT
zZ&eZ8IkHfQKDvS>D^7LFU$es!6BA2rk+>#WnBAErMKY^7lw;yq1Kl*r{5JW7UM@L)
zi4aa^$<uE+=EiEsng=Wpu1w<%ii^5`NXX7<8s*D^nAmO;)#RKtuD~^>NYXCyXM4hB
zGN=&Qyb@(Y-%g@BNfd1T-MxW<d!do&xe9a(wdJ&nqk#H+jb2L+>y&xT4_Ez^c?FcN
z%9X>`x4A+J%_%w6MPoOKP}aD#3)3fXp}%s-#uF}FT2HUcrT!pCR+sTV&|25#;*YN}
z4brcp<-_Q=t=J^fn5_G|A5L>(z|DJyGC%)V9f0x9PobmDW!Rjl^RdlFW`ui98k%yM
zGld)8i?eOj5$zI`a1_o-egA963jA-ucH-Z4BdFAWeTk-G!-(R99^9qUK@$fDE>OtC
zp;c2qD!@rZrZcvgDm;Wa0{8Vx7SM*@RmM(&f`U{=7R`g(RzpzQWA^8ZIRCzotmpcc
z9m{OpoIGh|tfn7Mjxvn0-6(n_xKm&zJUo-H8dZTWLt6bG8cwxR`ve=HEnw+`{nn>8
zq?3tZY6430Gdr}cg!c-SW7o_!cHDQ#DQK8wyVW$~I=WIrZfv#_bmKQ|5T!3KxD<-k
z@m%}T9kw!%{5}&|-*KlusJr;;Sdm01irN!#tkdhAb7KBb@_Z@7O!=lw1*i=G599W+
zV8{q$Xq(%4J&~UO>occ(scAG*cJNhWI}D%T$jC`jMI80gxwSCIf5rp*Vqs}sOdQhP
zRC*y{1`Ih0yi72<Ai5xAB(eEG5?E$papAP@@B_@b70l?-{Nz7kN{FF>5?CkkL#EM*
zLW@VXGCH(?F|T)>Zg1GtMbqfTz{FYlK-fi*av8FhIZh+>mCmX0?M)j4vKVW)&Un@!
zqNSCvh<O}42Zjbxe82+$YGIQtVZ<o-_rHG7Hfob{Qj+Z_Uo+M;pr061#w@@NG6$A}
znQgDnT`QV<73_~zan*XqYlqU-1q#(A@1|u;*V|cY$($I`BLAK`I`tTZLM^h+lCOGg
z!GT9nVWNkK*U{Tim7Om8<G9^H!wU5Gv!m{l#Sxl$<_xHo0dSMA)BVznJsLX;`nSK|
z^Fjq+ZW|)t9bHqJBtP{tzTZco4%ueBx&f)3Z>PVUOiT&Es7EI=-@66~CiGn%E?dT=
z2bS6TKyCbPK)nIr7sliMtG&CkOxbo{(15m+7|z=rB)zWEqh?QM<i>RP*w_4E9ofDK
zM>j5Tas5LK%C`6mrTX*JiAC;5DU*r<^&of$`FE&qU;=+Y0_$5?H*zni@2P?Sq^g)u
zFK$_KSy^=Xl$T|SZFi9T>!-i}5x0QO=UWHP=ZDRAY*`!e({db_BX<5iz9xx-uNA(l
zXX>&JWFgbWVlIh1ioi|L9te15AG3=8)dIMn4(Kv2qYP|TlIkjwJb7F-bUx~$h|(`J
z5@7rc|Dzm7NDNo-)M;%p`8HHw!$~d-A>uBWBkXcGKEASXe*W}+I;I!scw$K4e%MD%
z==+94vG(?Uk|9&{3qy#nhWq&;A3*c6?KZUsmHT~!FPsska6FxBry!Ky%V&l5>+E(K
zp04Sa;n})lDsx&!*{=P(vIbckkkR_WvVoJL<t^TK6ZoQIUO4zb26(n`r&$T2J$cb3
z^_TUk&P}}}sDRyWxI!3Vm{~N&E?dGpIIk)lpr7_#rO*<Sf1{&tBCp?%a!pZ=t8MI0
zNCYa$oFWv;eQdN?j*cZ;mu`;>f?I7K?KbJiL0G}JA-SnHRK)9MmG_ILzid?%7S2|5
zsdz_sC5qOje?`W-OO6c$B@uqY@ki(6dV#|MZa4O%)!zb1v)A<c+pDS)7|ac3{Tk3Z
z=l=&nLA|~`eDTs{(q9quLvsCWVd@&nMG__gmjUJzHgN6uVYG8G`P`57)-kA68aY0A
zRKc88K`dAk$VEddxn)!--8&v$GGuzLnVCyMn6oknx3~74b-D$TRxqj>zUtc3`n+Zy
zkLSpHJst-vn*M@nR&EC(j`v)SJt9{iKkz8}yXi9*q#y6uv}GRozbzS?J~`#1mbOYZ
z)o#=h6nRYC+HNC0QUjJ9i|e&Et}^Yx4;B=kcHb2d@wZ<R#!Isc5CVqwbl^?*Q0~-u
zW=5_wn}oP=WXS31dEQ^cV!Ga4oW1`804wfgJEh5~f#XQB^>U?7E*hm$@-;4cOB1~@
z6CxC&^`Vqhprc}BX!jQu_DYpfBnn3TKZg+L5}Zgfq(*vFD3$pIoKj^a9Vjex7Dtv#
zxJ`PIQp~S@92S#!L6_)&xx@Pf*ni$!!OW#4T9;^xDI`F_Y8G^EZ(+ud#f%sbz=#2X
z%vn>WuHWOnvp6rcCL8lFDI@`=j+83!<P$%M1ybW9M1!`7)tn)9euxj{;N-<(7AAfO
zFtefR_+}$P3P_RgD}tq7oyD(QERU!M_M|1Ep9#h#gmRPgSJvnxbtwbdVF{_fRcjsN
zT11FOW`6C*O4S4l)&#O}Z6K%ibaC0R3Wg7G5+3Y-O;YYuShLN+{9glE{f7hH&osk>
z_15vst9t#@T0-kRPD@Mk2{b*iN&b34Ue~dgO_T8Y@t(^uLITMM`IZi!rF}l{2+M?>
z#Fq`bOfR$#)o(vmbtQ!m;<opTN3{Bdq>d|RD^kk6OP4N9YxV0L1M>1xvx3?uocVv>
z<p2BZegT(71e@B21}dD<+d)C8#+LoAzbisqdeb?<7mOGX2%x0p<W2^|gOQz`T@fCa
zuw<`PujnZbUn0G2s^FZ{16jGt%9vpnFm=k~{IYU2PNxfp!%j$WFoA&qXqraafpngI
z?j`<q{q>AGXE=&Up}4r1(z0@p5}VCNubw^V)2jy;j=q{*mE~M}hMVu^l<=_DL|tG%
zc#!Or-lCFrfwgt0h5pFc^CIlsJ6P!uW#*M{3$qU7dOriseCVo(kxq{!qrgj1nO^ka
z!lE^6cUAnSI4iAQzu~{EkgVpp3*@S#<zirko5vM}NmAnX-4W0U!DMtMS9&OUVu@HU
zVVe-l(Is&Lf(4i*s7NYNC=`-xBnNbbJ!XY%htq||mnWBsKAOrk2(DA0{|OzSb^@X6
zdV$Gg>WbaR9qer#A(;^CW!wIS`BzfHhFx|x?6PB0AR*36Vh0bAAsRwJQJIO9EDL+n
zEqI&gM3~^)60xQ5`iNV5R9y?D{{8pgw>e>wl2!3ed~D#gz&XczF0F+GRUlVOcthAK
z=ci0xcFf(de|C&m$x}b>+)f^oSjjQ>e-W!#?9C|+cT{+b)6gmlky5tlZZ^J@=cZ%>
z6NC_=N3U)rPyQ_;xGF|c8`{J1)R}!95y|O>hp5u6Q|{91c?5+nzuti7ik_y!mq@V@
zW^5M0dvCtVTkn2Ar;Z)C>2KFJI)34jpO`q|R=Os{H##OGGm8QJ`_ivZZ%TH)N2o<H
zY{-YnAG#Me{2=Sp3KfulFwaZ(4p#C@G|uef=r}0R{*U+P7nRtp3Vpj+$t_a(VO8n<
zh1sdI{GInNFC+n`j+8FopN}jSA6SgqjuYUJS$d%oE&wjjBwb&;NVa%Owg(pqn?o=h
z6Qh93k@RS-1d%^8QdU`g9qmD})+UgsTB{TpAJpB=+|`F(DQY_GO*cg2kzS)!@c7)B
z0y@^bI8w@jdGqFZ0|EjF4i1)m`}S=UYZ<w0nc8*S=v$-|>-qC_aw-*q**d(FHfzDA
z6R|k_pNS2-#0%}i<oIKe!5x@5bc9gMtzHM!mHj(aVdQ@bAsG?Z^=yZYDVm1M>(!=6
zDYr+&CFFZG8EiHSqQlLmtA+<he6g&w{Ez*vpZ8^Y&M7F<(u*s#d+MIYtXIz2B9zkW
zkRjH?_~O+8hgFbWs1hCB4v(kla3;6gjmdQAx7zFVVm6xz4Gp142P}dzmyQj=;mmKH
z^dhC4S$CUa@=lrcb7|LA+bRcU<a<{I*p$#N(U!CByfi$tuuP+S2RN9gqDv(&JT@U-
zQFz{NQTkQ7^u@(Fk&~s|;5{iM>rmxte(=eK;$NDHi5P;i>5elkK+<wyuJkfUN=6De
zH^9s}QWA9HHlk}AXquL0Hk(11AgBwJCL22>;8G!=W!oJNALv9;@avLpDMk#cY@HLk
zd@P+Ii8U5UUDwkTMUgg}t>!OM2ytSU0IFjD=kwFz$6a=VFiL+lTCS@Nk&NIYsi>c(
ze749~D1CmMXY?0f;&U4gZ2b7tPr5|Qdt2gsWo2cJDgcW;=oqy{+zg;=nzsB{T(32W
zjO-rLJ;5^f;cFs8JzmLkU*-)AkBk4i&_!}ik(v-6ZRMKbA-wi|;r{GG_57l&l+?z$
zbLHmdmWRj1f4xQb-qD;Q<MVVEJ3}=>gB6DM3m`eegV}7xY%#uHG|lKyttdv)r`zqu
zW;1R@UauEbJ#?g;*=)vDb&Qm7)`&oUOmmZ*;$}iw08!0ICM}XWJEeDjNlx1Gy4wrV
z)6;>og%FCAQWrvqnH63BdFSYer}}g;=9pdRl_**$D0A)~6VbnKyp6Z!6!qCkK`QY2
zNg?Tn|BRA(z`v7{#D6Q!V3-hGjbs#po-K0)Ato&$UeP%Ph0`RQE?@|TdmJ!ap(i;y
z%x1H6yWPpvA6UYA*9u7hrw8n2$SH%uN^q-&g)-0zk%15%aM&3wQK1^c`a4<qd+UT!
z_2}fGSC^ycU{>Oi<d;LaTcGM-HAA=qVnU$8TR+(*w0F(#nx-ZDo~kW~$ZAqQpSLW2
z?C5a<F_WhI*`J8G1^lE--m$dPe@cFRL7Pf^`m?Zlk9a;hoZIbI*{Xj-La|)Xn&$_Q
znj3fCPNbCUmn~b?ShJ3{gVocid&iJan?+!=2)aaD<If))^xe`;<yvvMj);NeOb=NF
z>XXIUDX9Q;MWnB&l*o5iNcSB>X<A6GlO7(rJd(&zGj@wGgrvztfW!D6c>Dbi`QN;S
zm=pz@&Bo4Md)Tpg4Hk=qtn3^%ZP`j<*IJv^^iMx$)8^k%HH~*(e;J3vj;5+uEfzE>
z>D1nW5JoTTf-;?`W)$FAB|T{HzLd48<U$BDpi3$BI@^`voV28+n{BtB(c2LR;Py&!
z)v(a77yec*x^=K(HVOK6v9Wc(vm2arZ6A`UQ>6~9tonIm@{b}8jWZAo5`tb563hx=
z4=xn8$lvf9<Aq3Cj>%;vSaTAELX4u)359rFr@Ju3XdeYatKa*R9_!XYRaH?|R<=DP
zq$<%Sau6J-Q)|B13E8DkS;rEmRc$YJ8iKNGB-nkdmR=1iJcinLSQ_lVWmGAvwmKSJ
z4A_8kB<vATH@f{4jADO>stA1RrHynbbgRbm5Eygv?VIh<Af*6GUG@LmTwciXLO~^H
zKFKzbgK7`=c(wyncea7HW09;Tb=LfE<HwFhNbxD+__e(|OfrznRHXcP-<R`G77#zq
zJowy(0~@Cgp5CQB_cT5F$Qb?DcU@xpk7;#Jn5SSbx~J?<QT7L*F#EvnwL20wca624
z;_^tgrnq(sL6=9bi>eX*nFT6RS2u>mcAey~iNV=Hj$7rnZJzMBt~ZCq#7_ZwRgW#s
zOxqY9({*)<q7F-FBI#zV;$Ys)67r=+!XP6KUQrZmHbaC41_tuOS2Gzmej?XheFb*A
z4U5G>MrIZnnOV$Tu!!rfz5;+1tA1t6)@@7}e>2mbeI6+#4!Z*om`o;g9p3n+fYXXC
zTp}zaHX|X(gJhNNuFw_FM|HP{#U%U#sb3ru6LWEPc6LR0Y{G3F61qN_zJFhjjtPm|
zsVXBl>G~+v?Q#lDCvC=(;x1?O31I2wa+fZ}=WtR;9v07EC^LcYs(t{NyhQAGn24`D
z6ZhmLLPMaSW5ZJweWw}e+Q^3Wo=?*HEL^y7`-BM-yg~>IQK#46S=fY&xWWTFbD&3i
zLsY6d94Laqih6Ck3A93&2;+ZXa1R&fpH|6F>l<p7izGZEp*=<l|5yb)KsVqm9jw5w
z0-hI;;dAW3Zf-6b;!`8vr9xUEl=|dXggPKT+^{&hRoImao<q1#jD59#p#gCa*s11W
zadGixT{nbgSy`FbzJ0r2m(Yn$>a6+SCXBr_ACy^~xV2V25>_b44=T;-{OnH``)Q1D
zlH<k3-O6*(VfuKxwRJ4pp&bUDcPzdvs@o#+Cy9@gQmPRViRZrgL%}#HaTPnnEcece
zu9*eu!pS#8bq26o1d7Rgw!<#e2^WXm_}2dl(;{QLd^Wgy!28i*W<R+O^2H^(d)Qq~
zB;E8D7J3T{I`D3ulMXQ!gb;YET-LYWG7bQz%Y_hvYp*gy<Dwsz(zj19R<2ruuIng@
zLRNMTJ9q6SF(H8;zBa-c%@#8jiy5H^bX^kQ7M!m%KT2nf?x7rgdTB;Ry`-YJD5dzy
zGsA{`vZUn2@b(Gs59#B0KQ7Xg8X4QAPk5*$zIxoUEfsRXZ)FopauPn-q^4ZIagVFL
zW>yxIW~c0dlR~nk5LCO%TBX3>Nt%O{lBA?0z_6UDsw%F!>Z*${w3>zJvp_L}ei)UT
zR2`D@pj$h`qURe&2!`O?m1B$vHU11Hk1OQ&WE*+KMi8~Y5Rl{QNUsE7yo65#Osmst
zbC`sg5*|^Y`c@z4(3yIB0ZLsDMx{rxJJ)bg`%>bB9UkZw1t|p*PZO!&zQok8Jkq{>
z`<uV|>Z`ps-+c1{pOIC|q_#uKtOdU$UO(nc-J*Sf6c-)u`P3m(B0fi#)6>41zw5As
zdJ@XU(3Q~=(NjT*M`-ON1c9g&$U>>hG)$&lWkuf1kyA&u+3qDL0u!IxaA4yngForg
zj=Njq{ANecF=~sr1u12j$z=NNSX}Q|P*}LH1bAC??`{jL^!6!?4BjsD%f`)LSGMAs
zvqOZUfWs<Y>fFx!(kb2Sbnj?=M3?fSyGDoG7yMRUQzNcAUa&YPb?zTA@mtfSHlSlO
z>NX0DE-$94yRGW2|50<h>PMvP*twfCh7MuIo-f$2X$wP78_ZQ>Ml)yLLWZ9|ikF}J
z7Z;p25~)ipRq>WOU^|Q?Q7@7orQ0JE(~AJ2Uq5Sg*0#$fBTnt%pu#05e)?nah^vPO
zTXv?o1EesG3$!WSRZWspq_TKp*@q=LX&;BRSNbaooAWbMcGkVdU#_#wN&d8mq@*OV
zefxG%R#qmaOqt?HO--E{85#Mh5W?sYZ4HB{)BGPDVuw>YLcdsu@UbL%HQ1HyUqn`;
zOW2=ZlldGGqVdXuxda6mBJXw|5n0Wh66WYeE2MhU0bbJ~9<Z7v&)t(xmzcUPcpdiU
z<5xu1o%hZhaQTSHP&@RAfx#UivJnq&hx+=;YPH@tY}l~%n>TO1)a7!Cyu3V-l$0dE
zh%IV-5+^46zMgX+GAQRVkaq()$8joNAy|qOH#*JU9;vhDKA!sJyq(Pvk+ZL#ZQAh4
z%Ap%yTlLb$uvJ?TV!t9-hR~;jctIeZ5%3&{_l2S?7E<mF4m0oC{QBy*H@?35+>P&V
zY}rh+zYyzp%jb`_+l|ZRYVf%cB1uuqwpOnnse7Gx3U&jmTfF$sXYmfL@bpV)Z{2iR
ztjw~R&6oYQ%el|vm1GsDj5#yJ92so744`*soAr{@f=@A<1i=oaLqu%+sDz%~ZtdNt
z+eR?jE=q-|bM@xG)rz3Tn51d-D`7|^PN$QqbMw>B%UQO31wny<%>M2N0IXIkGpE1D
zV-MfY%@gin#j0O1nG`H$v*Fo<`UZt2SgCs`lJbMXjC}_H!U`&$7&)xh{E!Y^dq#$u
zZw46NFTj-8!P=*Nn3=F(Q(Ro6Y1kEK1+#Xilizl`&^1vpa#+tj7Yzy6qU++}2G6<q
z(>cjWPE4wW%=YcuMP6Q>xbemttvBC%^Oq)*>1qI^1_%BJYaawxBc(he1Hr5q@r8*|
z5a-jP+4pPRVX3h`)}ZQ;UK_LCJKn>)kL7SyxW>drWPt#p8hKtpz%+->D-Y#zRzGL`
zomr(4w~s&!wn0!6?v8drwXS-T0#S{WDoXhpe!U0we+ingiWWkIgoTBDv1ZMh(Rq1!
z%J%KsMN(3dXoDVZ05-h0!J9H`{(DxGejuI$3XkFCvjx1Yqnwg5YyO2PU(Efgxi0K2
zuPzVU@Y?DL-G;@jQ53O8z@q~Cil&o@ECLdxl=leaQsH!N+4TCViECb2b4-eC|CzY^
zznfF@ikaTh=e_3Xb9$RB0UcWUyuP^ewj(t!z>?O!*0Dy2bIm1L2ev2*dvCiqEaItg
zQIpMP(QWa@a{6?!(LUT15*2Db3!tb>qkn>ptU{IEoo(2yrXL=-JmQnyU2J`XIL!i3
zoSFK+?b5quPcwRuNI(}=ZW<}keOe?cih{eUB+%~Uy<B?ng>>%RiN_wfpJl%oosrl6
zwt>{N1Kd3J2JRStGppCEtA1igB#I*NXwAFlRw7xfyUQ$^e71U<-9qd=V<IjcnrPe6
zy@NF|uUN&aNxH{bDXGv%h_Q<HVdfG2;%#h7c9NXoX5#4Zhab2yGBG09gcN)sgb*<?
zF@ZJDt68%-$w^LNs$AM4FE3Bz<>e_4KKS4>CX?x60Ll-*+%w>(D`CeQ^|Q9?pQQfd
z4xtbg=znam9s0%?f~<3R{e6X%^$L6Uj`#5Tv`oyyDh}_{CV)Xs`X^;^Mqi)Ru#}QK
zpC&W0A<!!tdPg_3s8$ndH)1(`PUd%P`a;<CE-e2W{4fxTcU14O1_uXEpEGAp{Qmv>
zYs6$*v`SX9d)A!%lv#8CVOI5C5*`Eopk>?JR^yWJlMwP2v!eIhKWpA&X|v{RZ>|e_
z-ODTcY<zvy%epY{R*)Y`<PeVZ9xf3;A0+Qt6>a^-S5{m^8-5Ru2kZ987u;T)O^<9o
z_NZ<ng=pd8wtjGB8U3|YRsB=6MXV;gL;NWbv8g+PJ9JG<&hS2+lI0;b!px)BM)K39
z3VL_8(W{dUt*S+?UX{DBFVjO-q003mLabe)Em%xg;^ML_U}Sa+T|WA)=4^K7Y$2+u
zTGaVumz9^3pI<;$b`DZXW_<Pqy?gZ_EHsocqeqdFdVqNg7W2op9o+NKV=P?!6I*`&
zgL6lmjizbnx{j0*i>YDnh{jl;yQmP_^SMTWb^xqalS2_O{^BsDuuLO3z{JKqPP%ol
zvM<X^QJKcTu6E2Ofs}A&?*LBkZLBrd?{O{;P&~gy#wL8KTLNCIKIcgxd6G7b5Tc$-
zd%=PQ-K|#ZBLJx7u;dcR-_Z1lO8mGkx<nY3M_=MXWKA%9ba4GKQW_c1b>|qY!=3r?
zY$jY(3hfWEvJKbZ+;Vs(3m(peC_iJ1ow8m#LyM`t1=ZJ@3Sc!s*Qkc^yT#3L*_7;t
zMHhmz#seJMrAwFRs$JaBwn_mtM)KDS@=|BbdpR|$VzAJ~U|=e+0w_7cQ|cfN0<)!L
zf~tycsk7!?xc~F{pC!*+&`A1yeU>LJw{LlU<rSM=U%l93RyGUC-!XEvS_}bwg`&*g
z^xDc-PFg3M&)xssoSI$CjFvp7*&1++>PK!tN|~FNm$&s;T<=(+>x$*?*G6_9*2DI@
z!=_C9eZO&kt90qaM_Y&pHX(!&8I+viA=sgCFi*whku3kcf+0Qa{IsbYi&^0D1lI&b
z7UZO^Nk^NpLO*oXErF4E;#H+_PposlEhs4D+N;J8-=#ApC8Y!f1=6=qFG|bGP!xr!
zPd-k1MkZH|8O>`i{+EobES{Y<g)@ephR5qg)3lmpzupXsWQNpOp?mfg*~H6rx7*CZ
zoRP1x<A9sQI4fKBxk%3R5Ep4CyHI7`+A{ja8=kGWNDE;>3MCa9xy33O`D%|Y9fDFP
z-4J=x(4Mspkdw(aM~T##A{2s;_dHq*g*Md&uR6gdFE3BHTrLqC8+#96D7&pMK=IB)
zpP-teN+Urfw9faaB8>7sJ96q9?gFg!<ygBQBb&RLoKk2X)Uaz#oEOFx89!wv$TdTS
zVys#NbZDnSSTmNIS*7(CfRO?9ujQ~9kM%lpj*m24_jRfC=;W3+Z!uY!2^*h;;YRfc
zlgTt@&YU@2uD|~J)GDha)22<MEoO52B9|>wf#0is{+TfTsxaN-O;jZPq$JU3mntJc
z1VeT2W$6M+B_bO^DpKw?>HMKPDh?b;%fiiXu8xzMep^VG`F35BcntEH%t#)xDcXR|
zZ>+rGln2hvYwZjE8CbX1{GaGBb-dkj_=WE4meK;a^wKwM(O2Kr>a9FAXA$1wG=N{W
z)oGRYXTanZyGzP7&z+Zs+cjODJ7-N907c~*B^5?d@XvlKX6$*P{NFbPoY6ag>_U|l
zTPrwsP#{Y;S1`1fgLONd=(<+#{ZXZ9@;u#j*+5~53U8zUxDM!wVKMZ*6UKMx%(SPT
zsChhP@}od?%%j4C_uW;quU%9mJ$rPkJ+G#tX<Ch#lqSg4bOMDS&`)-+4$RUkrB}%N
zWlBoD`;W~6o!eXZd2<EVjR@i8Zwj!Q1z~|E7OgMmj!|K(+E&Twy&MEN6gKQ}(Lce?
zuiGnm=B9QD(cxx1s^q=7p5Lp_c~VFoB`;hezc}7=_$yP_8tHbsh1qN#W$cl#=fnDg
z!yFJ9AgS~U3Mv}51l(T$y<?0b<<$gOAua?m>KLVb?+}rJ#(u5J@!`=RzZ}{hVto;;
z89#?hu~Smvsl9Uupq<ZkWX-?ExOPDe<2k*WWN*G`@LbG_@q7ZUMoFRkI$?YJzJN1c
z2Xi2RDK<9t5|_*MUX@j{%?rrJNSXe9F|e`fr`4D}d$uVtJMJtbcOm7KAR39SXjRUY
zy7+11t1B-#_2IL#j>Gla6!$*++riB<27lZs>gelA2Aw&ypULWI^){aFoUI@=bY1-1
z+SfYP6c+9)x#avlOU~#MaG5~h@k(?F{Sxdf*=XF$N-K2i7KOVm3+L<QrF3j(A-hQB
z)E*9A{wANx&Im?Pz+x$@cY`j=-kTc{7ys}qy>jNgrl5NAZKL$CNO2PqqR_XagMv~O
zkj=U)+HAHaJKLDdCR9yBRaJBeXPy<vf^-jQnI5jFv@=N5_niH#yGWJVjFPOB#lGiK
z6r)RbNT9;518(9X%_L`d_;FJ?H;xQpKvz54Qr)PU@!Fl$*H|m3=c}C3&G;VKu*c~w
zEq2$GK|JZ&K1rL!H!kn#r=JcILUaZUH*95{??JO7NsNMsKtaba=pSdqH=62v1XLZ8
za~lSr#D>&&Usw7p;?0U7*6Na3dgzSmBSA`8H<<@8y6@P1)`KpujxLSY#%h92VfFXx
zx+FQTezJ}tjJVJNRj*5Apkc+Vdv47MCAH<!6-DXecDqHLx{@vOH2-Ssczb!!=GRtF
z=$aYx8@gPMl;eQoAtI|8AcXjS^$V;2Jf4<TWK%Lnoxu%Y3yV4LSk#WIYKtgU)%~X^
zMDEMlU9LB_?su^-%fsfqE(%Ij1P~gipi8h?1iRDSv<o$JR^I>?tuLoroR#cC6}P9h
zv5mvw&;UAh>f{KEiN7N(rt2LJ;r)JxRKN3eaWB)|jK40N$8K%MrKbngL_X=wxF=d|
zR`e!%>IxySn9YXZlyJ^~0LGpd${GCv$d&cKuQH^@eBFIOTJ^t&$HiYA9+U7;`}PSj
zHBU^U0ywM+>-V@A*ws#9na069m9zQ=pzE-8zl)GS6D1WI9ixozzKnb?^VgN*@<^6!
zsQ4y7|6on2<C7N2leBrN-P)ZycXm+oH|_HQmOgiAMHumo5dn}|Xmqjxqrms>TqAL%
z+7&DW#77vVbO3V8>$|`KVuK;Cyw)n|)gYtTPjbsqkX<HmdTW)hs1kp^<d)Sw`A$A!
zS(mUUPp86DTNEwW1__Z3yWtqruZ&vfWmOdAc#562&+OTzO&J|-MwpX8b`~dA0ov0Y
zcHT)!O1fq0)Tyo5Ilk>P;n_`Tn`aE3e)v|&?)4rimSa#qQtG&}w<4uHxMIbMy{&$|
z<3Mr7{w3XebzA+!4N=1qVytZ4@8<kd%ye#NA*V>CeVCbVSC;aR8>1MMXeT1XM6XUZ
zb{{mvC4fgoOHGZ`0CcnEt9vd9yIcaF{aA8^=5>w#UKc~V33JB|RkbC@C=Ql<Q9`8&
z;!m|<w<(mCm9cI64yJ$lIo-P@q9_WA5Lj(Cq^{#~yNQg9prD`-o6W|mU)Q3lsxhwW
zM+iY-Q4w?JFJfOxDyI)Q6-9v$<`rSvs}m<ooLQ$Zxf{vHT7~98-lBNLJ=Y8m{dbpW
z3$uPMd89*!SX4(o001BWNkl<Z4t+8*GAf-O+yJU3sdVe~>|{gN;ljbj?fl10<y5*Q
zL3RbNDhaeJlvV0v6{_5JaTqgxELJn~J^!lvnx7PsCuy_PEkF*iZr!@F5hF%;gb-F7
zG2n=<O%Ty^NH3Oj4zK@hZ81Ujc7`}SSY!w-r?*N>wyzaoK{g|cJHNsZo%t2@v%1YD
z!v%h@j>Ryy9Kr)6A@-wJ2u`mgv!woc+6Ogk-&p3-$*YQc47JzK;MS#NZyuDnYpbNk
zg+gqoe?v%iiHb|Dy+@er5L(-AUf1<PRaHfGJ-no(B++K#B<~s%Kv>W)6wXB#13`8M
z90;K!$VJNCpsYn`Kl0x%Qd|1+>)u%2(W=LOEEIWx^j+kfFvzi^!_Hp<e0scRa-y>-
zndjSuaKzniE`9I~Tg;%_&ur;=`J$F(p;qJo{M05sr$<0!$L<|<O-p|D`+{Ku66`Eo
zUruC*iJqNoY)f&G5Mv=Z!_DH26<j(rh<@=l!h%hBygHrQn|W_R1?5_}sa>b64}xR5
zJwCL%?UErq4a322JDry<Tiv<X6q9<>$6Do2lgvS8JAxj<OpmJWHy@=s>6B>Y)i>T{
z?TQ~+y=EO-w{6Gea--`yLI{KqC_->BBa;=Yer41Z*D>zK>#$j^M#&%vs-|I56vD&9
zxckoA@p?VnanD2Cc$UDOuXCHJVReqLG$+~U|0>E#`K5c$#OFs1H9Tc!_6g|n!wT=9
zu$Y8S;b9>cy}QsufLUWef{h@%LXbltA=XM>iORNAH(g^blvV141S)J#b+dGH1)bVi
z_+fP!8}>L4YPuKz)FdFB6p|<TQ{Z;H#hY)w>6tWX(taUCFA&iERwLo2nw&DD3rv@Y
z+SUWrv=1`k0{7)t^{w{6?mXz(u2xKT41*%4AxcvVjabRLF74=GLtr}V6b?@*fIcx2
zi|+*w16`Nw%csU&4Tb>rwMG^@b@mr1h7D2FFHIz+WM7^}iOcxRn@x}yX%u4jMM}xR
zVx63_`uB(UYrs<bd|fVA&5fkZ#7R!Nr9CF?A;N>l12IUd41smt9!Yl)!$97SR9rkU
zZW&U%OhET#w19PP)9b5FwMg^#5;`^7ko`nJA`~fK-u%YOg(s754$XvTH>LePWAG;(
zqMG0BMpd0Pe{_NEP9Lh6?IDL6-x9cTeg~<e>*7z*7SSKN>U}Og);jjVD<cRAFmd7F
zAa<v_nY*TptRj^OmxM8}tDUqQ51I~9AtquY%<RqZ&~xZG?wW8LQBje~`i+}!9eve}
zv3oP6X-~SxxP@kUR!g}|1%3&O?fT@WTICCOn(C*!=vC)juyzHbMqffq`v^jV9GFZB
zss^eCQYe5REu)Z;0|Q9O$>XP=mJsAHVYQl3guw0bqH7wOt}*uqaM%=L0}JUG6+(GM
z!`sj=x|_Aqv#=yP^;v*|Qf<ZR?Ukp;M_bsi$BAy@_WLK^G2!VaCy}3Dz}j`|dFPFP
zai&LSP**#OfM2#&GHOT=sW~2IuP7lsUnM-q#7!535)o`dN{9$HCCpu0_CjRG?yC#Z
zlhfg3jO0n$IMs<Z0E-td?x`qBA^@rra(-=iKtTm;&uZA48Vn)XC)#lJ0#vwRPhM^I
zv!%-F2!@DEErguXS|w2J0<U{l@o0kNJZcJ;9WGMBz5;N1>kH8?5%pcx1r-{}c}k6I
zy{@%VO3A(gNpY2k46qo_)1Qc}^l0qJk>r#$><3n`8Qir6-2(#yFICmUY`u8ANJ>)p
z$4&R5s5^ltfmoUskY)!)A><OAd%okbab1pP8e3lbb%=l;flf^hJ`oTiO4r5Y<2{YG
zz=mX=bE~5Hv6At%zuF@^p4aj-YDiVtAL;eFmLH4j9S@4KQ*Sz$<LUeIHw8<w3sph_
z6^8V5aQ~GN+<r+IU;k3d_B1y`dOAqU@o>T5ApS^k(`(p7rc8c}>u<iDoZLJH^zVxh
zoPT<6hw}Va`O9ATx?pON+kI8_bxN{RUfQC2XU+5JdQ)Di5nxxi={y^k4D`?=B8zab
zk7&zYVyt_KRQ3=qcVTyJ=V01it{xsjzb*=qru`%Yr_#x>mkzdMx&)@sIp_eLgEEM<
zrxUImAU?*-!r#iV`sC*9k}BWo&OKQ5aZ(yVlSSE4Q>HH}y8oF^@_rrC(*aW4e8uIX
z&^3)m9)FUUm}p*o<5OlV6I^qS@x7)9*pu#I{caaMI@);Vrgog&KahxE6JAyF@uH#w
zbAK(lyeKO*zFI`WNg;WXKM8yH?iJm;cYlED(mz|0AZL~T!S1R!$drPHi8W>u^lWb=
zn*fx%jaG!!GO&Gc{aC_-C8A!LogiDoI*t_{L2{nftjO!SNML_~q}W+s@U#npaGw~>
zgT=LD%u4+^q!sBDRaUvX9Y*}8Z#<-?OAZui?8s55^!UFo6z_nwkJMhqY`*-*AAjst
zw~WaN=`EW)A(+ZN-y-Fk2yvuAs|s)hX0evZV=p|C6Kr^ORaYJTe^L_O^ym`@0e5YD
zd3j?Em`+l~gl9LU<rm9Onmaze&hLuZ(dzXhrLN-6`2!)u)|D$)W;cCA+a*6Ub?5eN
zdq=%BxA2|pLY4dyjZYRAV>1gLy*85eVP+~^I$l-s<?>R_x%}@u`p|t?EM@`&1KG0m
z58nUq6F?U2&+;OKTw0Ws^3T%r^qQ5llu`y;)H@dIuC;6R+62r|s-4aMD`)BKQZ{^F
z&b9>=9Q?^i%6umU3tg1`?51+L8}k~KA#MvF&o3e^*@Jhzm&`>@vKKfxu+T~RVi$Qo
zx+q@i!uhiY_fijHg_lm7Bopd{r)NrykF<(Hk0Qntr==Cu&ToUaFe~*fp~wX}MJgb-
zj-PnHzHa?SI>g43*foKEeS7ovyPq+0g_)u<jUFAWY)^A>&*kBq*3-evrNwMYcB1K$
z>5EHt{ql$N^up}4dA?y0Co?Ti(q>6YN>a9O-!As=->+D$R`crBs}BkxBGKINMGScA
z#4B1%#*fQezrLbbfevAYIP<kn9VmhvqYfEyp|!5$YH{h+4B^--y8f{e*q3Lx!Rrzp
zpwlg?`HQ!@*RLkXX2fp#HX+%UFDa}Pm=z;&Cag+m=9cTE6&lx!41|P;hQ}6EXdEmS
zc(wY8JPwQDhpAg8Nio9>=|=o1;BvX1>fgWrOQE5mde^R9^{>DFTIWP^aVL!rqtdeg
zK}*J;IT7+^{yXd2<{q$P_71Bnv+x(-j24{7F+~1-@Iz<6a=d48qVs;zX`M$8lDlnY
z*=VffqAhl2EDxljq9WkeUw?HDpYTF(k6{nwDrV~u#aA{LPvKs&7d(KJ@|7Qd{PFLt
zd94!#A%ywVQ+lla$KRrcR=RcGo?FDFLxSiVZ==$sGkawPH{AXVV{f>&=9uE*5@vq+
zH4)+A^zGA&J<0pn`o}iDnEl<)dAY@Rm*noN*B&i6rdxYE`l>t40p0tF!*5^jLSjZ7
zQJi6+)cHTH+*vH-#f6#sSNhH?G|2wi)mMzZZp7KcOzongC@U}LV0s2uT|S!V=-R#p
zsc8q9IPq4l8B#$+h?z>4&RcT|xq5gAJvv%hw7y)My}azyqMZG|H<$z`Es`f`qg0pv
zsm|D*K7D#;Aw(o#SkCH-NUI6@#~N)7Pw8kB84t7?`&1p$3Snn9RQQzj=^Spjs!fU^
z=&I#WkPQaL!N3kiOF~}(Dlodp1pCw|E~-?>DQjNw$fC-IBGO?o5`+A8vP+B-4C-La
zDXfmQGN(+S2qV6;Yh?XnU0$8txf=TmOf@3XZZX{Er*tr0uK{sJEzN3l74p~DJT#ll
zr=GCx?WF6iI6Ysru!sb3AjJ%x7}xCL(Ulnm57TBvBnTmHJl=C?3rw84;Xq!gJggqv
z9ea&Av$aJe)V4*`H2#!jfBYe(RQG4OlU8l3B+#yK@8#iqzq*vtN}XBDE1B@%^Yug|
zKv-BP_uqXdeR}oeU;llHO`Csb=k8?gzGM7(XAd9pTWD<Jkm|7!5s9H@wD(jYCw`<?
zWF$AEAGa^|BNUOP=)9>_c#F{QmAd?UNQZ=8b?4PBzQgior#^XOVPUap-h#z^_{nED
z>~`+HV|+ak2@v0<6Q6uCi|;pukz1_t?W$6)JS&(U9jz3WY5csU;+w`qB($Amo3X&?
zW{#DB3n&0mfDOPV{P)hi3b@?Q_H2wes;0OXIL*)Y_keXk0Pv=t{aL`2I*)GvE^gvl
zI^f1=*OrNqG4&fzcBq36aqbU*HBB4~5DvU}i2WMy5FC9SwZp+TV>>i7w5CoLUY|bH
zR&#AUp&|^+Tx6hOF-$KutdtcVLrg}87?#KCPBHC*jLc`RUSBW@VF=X5P=q1+cIVa?
zsOiND;SL?EZ=0DqXgZ|%G#Ij(4VU{N4#VPSK0VqAOiKOZB#=|;C8NZQE{z>llVO>S
z3~cy3?dJNmGreVXc8ln`mrF`Y5>uy6JuxhjF6%QOFKVT;2O^otq+3qo#ZPk&HE`pr
zD`QQHc%m&kf20(rZF_B9biapA&uMz}Uz;5Vc)ooww^=PTtKFm}AncgMM*@{ns;iI1
z^-dHDvr-qtc1f75>GHV|0|Lc;V<MQf+{NQhzs?2cj_`X-dPXMy`tOU(`0#Clf`TY0
zDCBRq-pNx>JQ|krz@u*gLjka;{RVdrxN>WXtEWrlns;Tz|L!pbMt2jfE3L|qysnmO
zIYLfQq_}lxAN%=hce=g@b^+)%{jNtJy07<)nP2ebrymg+8G+mF;fbfG;qiF5<f022
zoHH&qhR;5o#rO%gaZW#-ZgEypvORpVr1+PTLi3%6nonC1l7leB<dKpNyaBw1A;zjH
z2FCjtuUCyd)J;<Xo(5ukwrkXiP-`RS-$i{PIi!(sP2mRS1OLVl%l>HW+&|MiAvq41
z(8QSKz%lLcBZOeuv}rYubzQGbl@9jFhp6xv?&u&N3t_ke!t9V!4w+S*X|hX=M4awX
z606yeR3@c<VoP%|DbTGQ?8vT-#nh!Bvs4m)h)})U^7^rx)uPm&P;3g{SV&cq?9G>y
zxh*y4jtMb}uKTuh1f#B|Zy)lYgqmNaG?w7iC4f#C9v?Rdg~wa<8iG!k)c?}d=R=5l
z+ge;?L!{!7?qR@pO%DBwGj__Rv_EDJ{wOw_`<pY~WU)2;R&QaX=E0r29Vz9eWy_Wo
zH+{sPi>$Pi7s6r^(pLT9d^IE_DCR#eyvJE*4D~y<veL=Jlm5XgFaC$1pdfx)_6zTQ
zIGqW%-NflbPoqbV?gPSNx^ykcIglC>8z0=cy@i2^_TJN%6y4$0l?&g|Dt^4n6d2v3
zmBeHTI&Y}uS{Cwl(A$kZBjo((a1-lxIs=6eA~>pRcf~AZaBwiAM_tUl4?o86v(Dgw
zdnWSYfBwm>cic<&ZiyszO=xiLc2SXhHe)8^C*010b=zHkq<G#gw8<$lFZqy$qQ8o3
zyCd*C@GJF&-%7xrH9-NIvSge}LlFtE8pEP_EU*DrV^}7qG%@x_xuY3l4!K%sEw*pp
zUL#&MY}k-WwevZ!6GWa`J5Wj^`?_wsziNblm|(;8U7cW4>4BX&qCq0gk>IbhOhJV>
zWGjO_AB$>-(E6Qqjuc%=cI85u+xXmvIiO!`eG6<|ik*^OIkiRB!FX+>YF}Gj*Y}-}
zu5Cp;OLgK^t0Knn#7+Gh+q*nzxm|(VT7MqL0%gGQo<Un-%Rc2rk4ojC#&|p)G3K#(
zQ6@`3zt*0QJ8vgab0ekrv9+&tqEeil@>O<;^7AWiOm9>~>bm6NN&n!{hwi6yr;e;y
zx1L#F&Su63@9@Li`KA3%J!|>0l`9_#3o^ekXi)dfCMmn+mud{}7r^koffJ;nJ?KKN
zep{<Z-zvR+uW>ZwOPyD=a!(f8LV>)jR|k92n6rW@F4xhe>@a9Rx8GtSmCu6B-Z#hJ
zdEc8~eErS-|9k%5D2l?rpMM#%*~}|1{)fp=B~f1C=lc){4+~|+^y!pn(fbRtQYQKK
z>h{N9MMzdN7?_7)DFje~IgO0Hq=|D~*2wnbv0^s?&(b`1{NXSZ7}$()BY_?*eC@;H
zzWeSg(sjMcDl4G-ZS@Cex{;;5BO3}U8}`e#n2Z?0-qD8T(xX9zyJeh_k%9GF4eC-*
z=Jr3<={2&-t1&Bw?%91fIK2X=SKzQfk9LNY)7N4tffAP?IFs`lis{60pw#NKva+&e
zEi<_mWb%!@K{oaIZ7FQPeU0r54L1#wf)2-MF0D=1<2`}4!p&1R94IQ|<EF+1#U>6H
zCbO-naYsv)TN@$dpRT|5;V}4=0V7Ts*!X(rh5vhvbB3SAnWqn7@4o%K^7`An`O1r=
zrlqlH$<lAK^NMdMu`$DBB5U&XQ3Foz9q@X;cpIDdy10Hsh=>X?-*Lmp&?{6$yYWM<
zV)ts@)9AGhmjhDcWv#3tUzgX8IV-qV*Es8Se>*=UXj8J2zVSBtC)l6-$JlnKIcy^L
z;K75=qU@B1ckkW%_;)|d<-WV_z-qPf&ifzJzFibgJo*riJ@Is-RW?FHLKr>jqCO$*
zx+XSfZhti)35F%HZY!(Cgx^IN0<AIp2}%7B=Q|QAFx>COKqXCQ!~PThF`Gw<KWpTA
z8pcD4M#dgfq?ENTo78U_4iFe_?tS^NHB*vNQeU)zAtHN6L!TH3akQMJEzBoes?zKK
zz4Yzl9b)%yEj$tetj2Te8{4qOqgRt;l}ff{!k%0~nah8b#nf*>)><f4RXqXSWTwAy
z=+ABoBd=uAE&kg@N^wDh_MB}Mg?7h#3T=svY5H?s73ZPGSR4_p-xd+0zqXXJSXI>x
z$KrY?DyphAnho>A-1(H3S8&tct|z~sfJskI<;_=LARxd&?_NE5{@H1_-+t3|Ib%m&
zSW@Y+T$fp>5+7p`F%f2TT~bn^bMC-EE;=>PG-^o5Gg2{K_=Q&a>;H7;QJ1m!U3z&{
zE6*vG^8CR)>}QWUEx6mzo(>{IOca%Cv=1{A6K)o&2I+a;Q~IAeZ2gV5+);7s*c)e$
zz0pV}n)Jv6>`zVQhq((l?bK5^eaNZ2`R)geUbCd6RJWlwm4gAaofyf+_yQw7ueu+w
zz<5qwfop-zjU0C$@Oss8KVt|rAO9m!81aQ`{T$x`IK|KQ^}xZyc>+x^4)|Zy?*fPf
z&H)|<&h|6jgb}xSMGIUz81R>=`9{_GukqP_EASN1a!ar#m_2(oBS(&`d8{ak;jd6D
zAg4||xxk2347P*QW4O2rDg>D&kW~U9c1d(Fgw_#`fps#KTY`qMivGu`J|1uX{jZM<
zLBlxFa<@))nV{Gykh1EpEYN8G7-+3+7g=30sWchNQy|z_s0IfI4*^cBE*YXpZ$e8G
zjYS`X5!O-<DdkYVS@vxay?MzaCo#BX%9f0svrc`#U8sD>-+q_Nr6?A=|7?#IM0c0s
zE!>X~V&$@B%Z_PM<cY?{En61t+LL@&uO9XLoNw8>jVGV^7fa@ShtuWazK169+`p#M
zzMZjl5<)O>!mZWAt;u`$Ss#C5GJpKBg`OR4^o_Uk@uDKSb+D3Cq|&d8?V{v#_Xo)Z
z+UO-t&zo6NA9t%M0LNjjDJ8H-cQIdgZIy6EWQggG^G^wwv^Ud()gt(AWf{Z!2Oy>7
z_x&z3sqoPK_jGvTu}3;stqmi4p8MBS?zrb6Vq>BidCqVyzjiDG`t@bR*~99MOFM9o
zx$_q;%FRt_R+wH}w@9uA-p9y5ejg)S`Yqs5pcjo>2fG8^s(!CdSn<cGMz)W}zrCh4
z-P#h!ue!!aps|{6mtqLcqvb{nF<F<=s_TE|XMB5FM!+_kY15|FEcE{U?|%mfA&mZg
zo>Hh=hQVYqEP^Is2&wo8=pR?x-mt_a*pnyNl_MKiC|e56XH6;^Q5U|}NA0LP@-;Fa
zU6Pz9*_oqIRB4FFhydspXSn!7>_)OskgZ`qZ+C4*xyfXzm$iM|Gq(SJy#r|Vn6m~*
z={M*~isKvd(UJ&p0@Vj?BWz32Ui51C4MyvD%7hrjZ13CJb8zKuL+W0nlqX&o$}z_2
z)~8Oq_o0JX*|qv8CnuNj6YtxZlbL<i4L9AM_P`^L^T53m>Di-OgX6!SGmkav*3-RP
zBHw)VHItushItz#S%qGTD|Aw`y+ntbnX|T>znvR)aeJF-dT~y~_^nc%^rBj>X2|Bp
z-B%)cU#nog?)uuSRSdIQ#I?6v5ISZ4x^g<Tw@_53vGbst=rA+ub~%ZPO=R*vKjxVy
zAIG1F1TdLQy!pxty!guNjK2EDt(oZu```1xqrascsI`A(XXkM9gu61!$}5}Q?z626
z$wT1;ZU>4Q**~bx?~Mt-#w;8D+}aXoc7fufL<c4ToBixF11+)2_2;Udjp1%zh#_iy
zIW9>z02?-J@KWviw?~1cjs{jWBNMtVR+Awp`^MI~oy**UgRSJLE%%9uY|PT=yS==X
z47?d7)O5@Vu^TS$#HjkQoM4nls_SkSCJ3r$=y#jV<{E*Kl$3OQ&2MH)M`~lBj~$W*
zu|%n#EVqMb_5R7n3MnZ(-g9VcTsLJ)Mp>nN-`~FA-or);)1m5lwG7qqk*ccxWfN_x
zDbLxTvvu2!5ocd8dg=WSPtvD7`#htryk1K=aPZZ_+_ZJO_av{r=z{YYewP1|uZxx}
z<>{pVuxIZ+OeTdp$KS$Nvlo+RAC0P;sdVe8n#5)ntlQ<}fvX~~Oz2+xfk1wfjrPGy
zT3Oag-QDDhY@bwlPAl{LF1=p~!E~=F@S5=#g}tEZkecHq*r8Bdp%LHS!pBP`{my-a
zUw)j==|h^1Yzqtw;LTTFz-G1DHH*y5%*uWGyi3NE|KsU@GWB22alz;-H*Wo7``HB<
zdz<ZD{WA+mFlsK&Yh+*D0Qg1VH^c5!Xb>0V0Q}Yc`W@JJxDz}IGz=@OKL$0jy}1ma
zx=vPoKB@ZO1AOgg|M}4H3$-=y4HEwT`|rI{%2dEe2I=f$98D`UYDW6Dysj?tU7fvL
zZKZ5UyjsKF-XV>|*9Thb7Y(oU81ChkCL%x(7iz?U_H5tK<t;HvA?@>7FT{)hdu=gW
zuh+W=z;3sn7}hGY9>Fn<kE}w{poL&*euRnU_495MNMy$Go<mz?*8%a8s)+`%X10K+
z)~_EaC2pg?cJ}h+%eS}s_1XZ%*{Lb{1x4?D{q46I%N9=862kKt5O2FXET+qp$&Woe
z_WEnC^n2W|Yky<T{6)O`#{bcyoAEWbYj-kX;bA=Y>@?=iUBVe7Z{WklP72C2wxzg;
z3N<mjU%;(#k>;(pUmSLi8(#ZDcP@BOtDw|RX=@djtGjqjuShP?^c&6_6fkP!z`(hQ
zvDOcU_Hu}IyPP<zf}giku)iRLQP)1q4|9HE;_bJRo14evr>1enh>N)3(kuCF)>n-t
z)5Jta^XW(L^oR(I{3b5aH0G3euYJqvFSXA;`_P@9nX|6Mp=2J;Yx-vwlF)d4#k#@q
zl^9~u7ZVN30G;b|Y&FNq8d;ad>OgjBWP5Y*XN^pzehipD_qIaQH78q?cJ10p0JGUF
zJswZ(ihFQUeXAruRw-=Hg2GDKU`gn(7;O+kTcN>3R%v|!>#*q9EdD=xCZ*vVGD}*1
zPC+(9IQxqQEph7X$cBT(^|RX*3!L`fx+C>Hd-l}pl5yN!+s6yP?`HOi)GYF-Q84X1
z-g9V+jCp)hW`#?Cr@;=Bxw%a?ju^Ev4{vD(z*67-$bVHtLV{fpe0<aSp>cO#7Gdfa
zZ^tCW3zuAY!BdYt?7#G<S5=wx<TE^Q&z*EjtgXMeJ9#go#$3yeUAyVju>&vsCyB4W
zTh6@?JVCn-J(%@#32v`sSl@u)Gy6EOn#H)1?9|J);=SitwXAHd?y0#&q)L_N)lz+-
z?s`X2%LhwAuU?&OAs3z+XbyA;%Wu1!q~}Xc9eNHI-uNFjty#eX_fEv&aPa;|GnoGA
z=ZwAKI%a(E7MEUpA&*a)#y_6^XQQt}|Gs^gI%V>?xuxp%$PiQJ_)Ef+2d|8<#Dpt%
zhq=Xdhx3~L6@+9~Binr~1<iGDH}>H_9%X+TE0`X76z6mRH~HB<gZf{K>wrCe#@)s-
zsNYvN#IV{LmY$ye0fwa(g1f@VFTvzvX>+NtFJG`N6LQOBtHmHzQ|W2wR@ZdFzI;h)
zq2xf3<UkP|D3YWUNRsow>66i2SP8|4O#VE2bScR%mu%0L?8#FaWRD9I40|8i)f%c*
zRb6z&6<4I1&1M-K94!0x?R)ZzstyAlyQl(j=vsnp2~~Li(iy);LuQV6Nte{D2u+S~
z>>+Wj9`@kQ*$Pre*Yy)CA?`RLQ&L<~%Kp<A6-}J{e$J+>0<XR&-JL(;oMG0+Lf7B>
za5^O=rKF^$v3t*60K8r=&%N|2=L|oK9^Gr>DMLd;xciRrEL%LEm)@SkS(n^Tsbv5k
zEOMdivQKz)m%o9wp-jq)KGQ0Hd{rxFrdG+zTKS$Vsa}E9U-Fs)h6ur>E4Ehh{dz0i
z2VTs;i>9!0#jkw%*~eUQ*`-)47V`58c;cyPy!GCPO#1tS4D8>Ro;|wr_lNG|%TGV#
zqZu>#dHG7e&k5l2(HA47GBh&8^xLyD^9!ean)~0ZLTy-aR?6px^P0BxtnKFDzv^Z~
z=3^w@`1%|##t_^(hWPX+BzIK(9BUd|2po+&?HXv%g+l{A_1PwYnHVL9>Jp35cCuB4
zb@l4{`s=S3U|1wajvTpT@7}#1*=)9n0Cc((F8K+TkA(`Ka1WOXX+=g%WO#s)jVtMQ
z000+VNkl<ZsT^YS-^cn$@o2C&zhTWvk0vN=pf5xtzZ;|!K(FYQDNJ6C1j^l#!U`y;
z6pe~%dlU@+2s)3cyVaSTocy%eY?d~g?WBNg1Yv}`L^nC`ghv((i5t<FcjqKcaP|Y6
z^3uK={6?USrvU6R16%f9(+E}Zky6+7<*k0PHbhZYN-FTN=+|%N>=mW|K*(UH)Bj>U
zBQuNVUw(}b-+7%gPand|ufB=RX2WbY<954w?BV+{n`)zJo|^V=?!9XQ9pYl?(Jhf4
z-4c1|zPs?MDtq?qHSO57`}364G%~ZYDJm}3D=I3{b)5^Xmc*dIpgHZLB1OlJadhk2
zmA-v?(Xm5Zz4@f4XK)}rgHs0&A}T79e?2pW<ttY6#@p{x*P=S&>|xwF{uVy?cn0T<
z^jG8stQIQ*tnbV%`a?>6a&dOr{v&xk{|Z8~xhAJ#U@1n{b6whD)Moq*=#PK5abu#X
zwZC9HHL|_Dk?qaFU&74K_%CZT6G)<=@VS#?WTEZbx3A30%abmbORB0W-+c4UDGxvV
z@F0`PWSGvPhx7mKop*dw<^IRt&q>lGEv2l=6rqS9vMb_cPf-*F#j2oQC#WdH1u`r+
zKtz@ba(`UB0!0KBMN|-FiR_J_MW7TYvy|>Ow8?pXf1IW*Y0}b2OOfaOdgb+MPR@Dq
zO`7C<&UeyhFYJtj=vm;|AXb-<kZ#<}no)RDArOMCh0nDvilj`4J<nyCjsr$ppg+$*
z?TSWg9)EG^hImZMG;GXr`97No4eC7#t=<EB_56>>#Kgp*-MV#)4h#&GHk(blQw)R<
zKJ#))2a$~NoiE*Jixl3gv;@15w56SASQA0FfDu7JI!F@)LlvpgL5fBxqSB;;bm;-<
zy-Jl5gFpyXFo1vr5b2$OAXS>wP(m*u^j_|IzwgKWfA`<)&dff~o;lB%GkeZ?`SncH
z2|=lQ_u=d4Ku8kyy+#E|ni;{v$>K|l#L>$q8!{UBr|K+CfN`<rxsEX4aGz@yw8*5X
zF=h-pg!GAP%Cj@FWy1MH@|UV$uEV(yR)0@%9pkP9zYh#bWms5Oj-}gZ;e@kak($R>
z#=YNLnton}*}$R4qAhSfVgDETs7AuETPWe~iZtjNpFB9T{qw-H`FfX6|GQ}?aQnv|
zT+)zGF!1^i>?|elc`hI$nj<9Y9N!RKUhm}&Wj&Y?_iSl^4vx3%^p*|fn`ZlNsOf6s
z^(;-vb8*=Dmlw6u1S$CrQWi^2{^&X6Nt7>M^ArKSgj&QS4v>$vx-?mZSwKWE7)(q;
zLU+np%0Vh)`KQCkFq5F+`<9XQzFrQ>Zv-`((dWl)uQfmXR%J)Nk@kPZ{xKTkR(|~x
zbZfAHMQ=PZ796X6$Afn)E!#u1dHyOd&0IQ8bpaY6H;6BcZ|ptz9zHt|cMM0Gf9fb?
z$=z4dK%?k)3>xR1(|7neCB-8Yh>md)R2~Y-NO69ChqZp`xyJkV??0(AF0He~b$6@z
zb?%p&9?>()Zsy0cZ!U)EDOQ{7sdZeVeo`7detm{}J?aJFBz4w%89Zg<t<Hvu_aJb}
zPv*%)dw6jzA8p-?zS;=z0diEAl`dcDr!KrdS!uu<wOEQ<t5UbvUq{9@7Z?maUZ0i3
z-)Rg+^X6U4#k?3Hy-m@FajBX1Jv(jdJHH^`#!W*%L|aePLr4TdjL-!_)9xaT^aDAo
zYFpX^zq)Rg&&P>)i~~V0617E&CF55;U&1u^aoE!iw%B)r_;Z4_KPXBd23=6Jw7@oD
z!_)`UdDsqs8+f=y55FRwC$e4}VT-tOt}(}W;~h1|XuPNfI2+%a#hfke3f>bfZ-~I{
zsmRU~b3ji`vW_KY(O8dNA>O99$_V$rcNYuq!JjAE)5_NQ`M-)^Na2ul96jyNke*X`
z*OR;}l>s5xCNJ1K%*BK({M~(TLB*DNlV9;Rjz=Z<s~zeKZ@k+_-hSUg?x@yYcLw2&
zcJ@c@XcoTv3RG-%AORLrDr;IxLJt~4fv~2#$qhHQ#@qgmTh5U%l`VHmJ^a%~3N?Ba
zb*y_I#thnUhCllF2%5KO|FmlUSIDyub_uJ;PXJ5W1?AxQQ}4Cu#Tch7ozeeqv57pM
zRkOfm>Z%+lh0-cKP(W7qZ`x)ARdx3rNbSr!TD?4(uO?0}75E4(m$kDoDhp{dMNQ2K
z#R$^*Omo0X;wdQ}eC5uL!7$vND%ZZ|%MSYPzT86_gvVnR#&ShIQkK9!XJw7cS}%(@
z8CH4U%NSoC^Zo_M!etaZ%XTUUs3RTCCxedK<6|ZJu^+@}GykyBQ3+nn^E7HcO{7<}
z%f<PS>WRUoJ^>L4iN|;lt>CE@`{d+gJhOP>v`I~#8AdLx-_}^d32caw3v3DGp`|`3
zu!FkPn)(fVteItdu}OtF+1qrDU32mU9bYs<AG>#;l9L8kL~MS2pKR&sV8?KK1*^$<
z*i0_ho*01ltgw&dowKu>^_geoT!q|Hwa%&#Am%qgfsxfA`_}yg^KO9b_htR)o(00c
z`8-f<?2d^crDbEwWW}0QJ0q-rC-IcI%r^Qt;HIkh9Me-*3CrJ-^v{MCpcl4o$utrq
z29qP29pQVqz2Vjj<j*RfRBHbTj9-8WhJ1Bv&~*vm2;Xe29&cOEcX_7X?Owl1#-OzR
z@C)bYHu2%v(?e90^pE>SRMFN007w36pc}^ZS%@*trE>D%(<d4Q2?>dDR->qg)gc8E
zkx1@X!Cop?>PTu*%~{#Y8s*kjoyyf8TwJHBV>#xs#j_QBL4#L0Q8t+qv^>a4BL9SC
zn470UD{<j|INp)GQ@Nu{wUAIe3;`NLyO<m;?p^nty;t|}G5`JAr25g;A&i3Mz|>{i
z%3b~TAa!u0{({itaaT?<dHu;;z4f?DyUE=mitU-6Py+HM|09dVl~RhJp|$fS>n>sc
z!>y;K>5?)#<Deem6zkzpXxibz(o7Y*e>rtr^I233i9Z*O1SE5OuvvXsJ^r3#Cd_&+
z7m_h8>n)$i^Af(Rv1C7xnXp<V{YxhGa%Ut4DPF;0w%rY*OCmmv3I`7py=rSXF6Bl~
zjdgGK0`X_<_EJ}#1J<(>2_ct)_|7y9kGgNMFuL+BF_aNOje+|vw$uOl^N3q(fR*Cq
z<8vZ_Ojlo40k~S+>Oj_XA+Vy$07*(8KDXDZYI1A1<dSKiO$?$g)PIV_f4Qcl6ztPF
zXSY9P)V+9|uF~57;mhB*<xYZZMieM=W%lI&SxWF-aDwprO9d5kHS^ry+D|DdRrbf*
zm^a&XtJ!Ek8Ed9-^}rSi1OoXR8m4VC!rryBuB-eYZPQX0@=_JF+qe$&{vnM(VBPnK
zh{;b;w^Dzy4F@l$NMUn24lET?0kH6?a>>}9(F-Mb^42+k3Rrx3=64m#o%3<h|MGCW
z#;<&iUfeA3YLd^4T}(~Fbt)~G`mq0SPc}>AR*P78(dS?EI?eJNug=rlmxTu`=bAm3
zn($4mZ^w;lg1PADVu|Ka>==Z&&kd^Rn?NWk{AH>bx2o<HOHgRV*XaEzR@c4pTy$NK
z!1+-#+zsD6_$9Z@ZAXhjlgbwcEzm&fD1TL~zD84v*}2>~KE!niJTtdxH-P~;KMbxd
z(tdBi7YI8#O%5Sgo&^s#`5vWcLO37Db&(EgeP<#0PUVG+ZBowQa%nd&tkGyLQu{;r
zrPPgETJ2yl63s_svda1zg>SYdE4wC<FFuz&)csa&bGR|~wph1FtktSw1{BeKe&NLU
zOjoa$<#uuk>F)G>OTNrG^4~+B{E2=9t`pw~8_RVkM^_rGUJ))>@-Ih5L`3BB^YfE@
z2oFb*lai9sf{D^!@rutU>DbD#C5{3oARa|UMZ1)&Qh|fcQa;4--YWN2o|e3L?@5+<
z!ZLshjhER;xw0XPa`(EAyrJuOJH_NKF2Ffk;YeZj6~CsfjZNQlt=nSZ<@vEvzC-#?
z<I;c9fBN)keZTEzv*$J}+OXCRViT4jSMiwP2kvelQx2x;XO&NJS5vY<m(Db^8KWuM
zM44^eW?(-qk;m!32tgf=1kqoYewmvHriQ#aFA`|mUF;GMD||RgagLpSrxYx+S7iy*
zfxi-CK_J*3vv*C?^A0sR+1-_NCV0|y1evH17a0+fm5oeGGYmFuPWnBfXuGvWE8#Fh
z8anL)d13J-3($JEfK7d5exb)R2v_%s>cI{6&0`wqbNSYDS%yV|++4l2iDn^6m*T<E
z7mF;4Vo67D1H>T1zV$x1_yPxtkawhS<-9F`A2tUY=kMDKN~cx%az17Wy?eAG&PWf4
zoi2KaYu6lJj;9lDde|sonC2lM>=ou>Rvi+Cm+!mKZC@K-pTwMLP>48Ya!4`2FHV!(
zudgmSd3kR|qe1_QZOE`0J31CSPySdIsYjKiD}@ACNkOKb1KCK*OA$>XROBSO5$Qh6
zH5Zz&tXqk`-5SIOskFDNiCFy-Qoj)PE&+`BHgq&Kwf#szzxV4Acg;TD&-}-(aDlQD
z<#Df}YRGi6wzf8q_YfX6e}$SV0i#zFSeX~L`$j%x5u%($jap3yXyQ`$;4@aw)e)&#
zSxxdF&_W!OnXly`z{AZvB}Pt@|MP?QqbqweV?4Q-KP=g)&kEF01fctsamW|r%y7{;
zp?JJ8LRCFeX9F5dy{`40jOpV`>;7~}jYgk#t>mXk8Xe#7B(O>cInK|A<@8;7j`n^a
z(ZdZjk-y<fSUp&8O#aq^v?08St2_4Do<&nikJ|Q}<K0)Af?lW(unJp>A!@%m+3A;F
zrXZJW&Su~S;k1|IPZbfD{0{5u7v)wwA<sFw8A<><YP<In>bHJshmdoSHy1YQSy~sD
zx#JOWMXjKQVRhWUn1_o9URs`hcNT8Uau*ZMMb2^PRpGK;QW}^SIYKnKqmK97F80~w
z6OISl^YuNlx&pk4S*&OM02*qEPp?o-cX_!eV6$|o$;gHPD6wf+SeS5JXP${5<7H7{
zwwzH*@a6hgv7S>6Pz`A3=g*%N8igL@THW$s%a~vDqTfa-z$Ps0-Fwn2x`s@5AUZt}
z|N6K#9qMNQjsm38P_X{DO_IxFLPA1!SMLUeaYbNq*QuZU4Q5ULFo2+U&tqkYvC;k!
z$;!mYOnIS>2Rb|S6Q?o~wYdve*rCZvJB}=XoSRVPIObmrpp2Cun0m=x$Ex}*?}lF6
z%JqK=`#2fL2(#`=O!?I-@PU)gqJ6a_ED0r{2;KHi>F(wf;mx4NB_m&yD$S}^qzVxU
z-}ulT?WF~RL~@eUxti-^*VCX!E^~%>z(-;ETi4ap)ogU8AnCN^RYh8E*xfUrGM<nV
z0jS<k9s)Ks+AxmkOE|;kNd_FP4gBL<wph|*8a#V2D6#}#v-3ON;d6}d7i#+1ob>aF
zEKCCpeL%8oPCuWdPyz|Q1<EiNr0g5FLJ{oM5qX61znYVtyZj>x+TQt-T${uMBgp%n
zO(O5h;oIMP#1zJ&%r;H`b+VSvI@BkS?G3dxw}qu0c7%Dhx_zb-#C*xG5PQpIz~}nr
zd<yvQuJY}u&<v{Y`xVf#Dh~A8O@~`nb-qV3veO%G>+IuvVVX_OO6ipvMTcML!ArrY
zt^`&-AaG9vySNCDS&QGxuwmOH$=laGZx@d6d%k6|@V)8vmI>gkKUD+hmBt3&4J)d~
z94Vy{uU<#e=si#Y0r*hK$BzSo)YW@x#BKWw&wlgeATwU^7CEGQ1{~jZ!s4o_<Nq|G
zHrs)Odj<i0iDL7Xt7qN~O@##xln-VjEviS><+nWLd}P@YA$*$IGco{?Z4EePtd)5E
z*=6usVFIuh@b?E{L@JpFgWfS;d5Us+D8$&p0e|<oJyZ<jsO^?Gb-EBm3POm$%xvGJ
zlKr5iaFVTktKCv^yHhd&a8AL+S$KNm{Na7T;n<?teZqnJo7(@oyPI25l7XyI%t}JC
z*kJY2r;9E`I4rfs#|Ax>^X)TzyT)Mmd;xEbn9U!L#;ZLG0w77l*YCvMHb{avbmD7Q
zZE2A7${<3y29>2IpVJMjg4+E=7d8IVl?INJz+JdqHEw3#{AIO=ZQw<P4wTbW2>gTp
zPiybd+tysp#8RrC`#*v>v?jJD&L1!r_Yd-aL?u=e%<{Gty6-yUZChU}e<%L+`<(Do
z{E&cHY94WVSyhcyOyeGN5QhT4%axtz>^}DmeXGV5olc<#|HMepNTx2@kzV+Xh^tXB
z|Bkrbz;`-!b~6zfnOg5BPo8*HPMW?TrA0cVcTQkZK>6};1j0{VQBl!<^qV%xOs=IX
z;jqv^;AB>M`a^=oShJe>EJ-K*<-esS(p0x^7u!`B76a8KY2vM@`13v2>a&xwvhLyE
z%V=aO2A!$t>W;#Is$sING;yDp^hI9DhYuZG7Snw?Wuet(6)Xw0O739_>mbJJ@v6Mu
zU){;#aEbY(6AXA-$ak^auY341s3F(56shOAwdvoqs=O5}#gyEwG4-@SS$unU)$xIg
zl0(P@PHK2~_|Gd|LY%j4h#d57z|r3J#E|ZHdrzW(t)kVmouiC8y?LcF<7cDoB8W#2
zy5$~XP`=Fqwv3Fjv>mb_<YxMQr+33k(EK#cy1#x--^Qk3P_tNf`#f`H!yuwWS?%GS
z;KkQkgiA~*^BDRyGp>*;PG(N=>!86#)f$<b@0q^&eynVF7VeAN2o4_1R3DUVT(+vO
zse|;2u;17V;JAS@zJbVrb2-vk7=<4nz3t?=2{N++Crjhcc`B{XTuX@s@0SUn8PGuN
z#HJ)D$3d|c-6jBmg54zF<T~Q~(`}o3*=_LNPH`)c3~7i?zR}1p3CD?0IcZbqm?MvX
z0BgzE&YP^4YG2fsdqTB|gXiZ#?HhmFTq<Fy*#o&~)d^tZzco{(x;@i$zJ@=)GF=>A
zc7Kz)UKkPaf!|;!e<d#hiA`@4U!o*d=3&l$A{xQNoHyL0a@C^X7AJl@r(dGCJ=bFH
zy)pV5F7s?@>5)2res|p>ODaiBZ0yiX^)gfAtn<gXxUc%9#%2)0_t9hf0V?~`tJM{K
zjKe4B+ORO2f~Eu*)uOqf(GS)GI|NX-XIa$Y_-7S^XP@;Db4-UFbpf67`#a3=q_xaB
z{vRU@X{1Q43lCbbeVbmp%jheA@-S!T3S-w|y|2buj(R13zthm^$Ub=R$FXuU;>8HS
zwXl(66B81`z|t~3w?}`A^J!$s(l2_}=fT{|eq(2{X_$O|2ww3DfQo8)`PFsXJ3+R`
zZZ2-?33)`uZXJjkdCeL~Gm)#Ml;6~He|2E=rANpe75OQmCAIj`l$T^en@=Rb8l5Yd
zs!P~jcEu^hpl3H<_rPhemY-5Yj?c`x!7)OFvL-c?EmRydW=G_qcRq&PS0S&Lvvrgu
zcC#g`(BNOz^LbkM?>qCg*ng0>B=f`4?Gf2p`bMEJg#G6DZ0WX)ITUsxc@EWUIjz^f
zqw19FMy=z0Bc1abm3>r1sspJT0l!pkEJwvJ2&en&*z4M|th#DbKX{#Z{ThY8OicZg
zm=jivB=;I2?BN1)caYo}V<m;*Sl}A20zT*-33e0u8i52agxSTg2s64xF0fQLXA+i*
zwwbVHggR-;Fm|?vF*<S^=jLfFh4jQnyWnc8Y{8e*0VEk8L?Wn3MUeR^;9RW>h2&4V
z4}@>j)b>G>=|nPKzDD{@dFJDQN{_cYtD@<WZbt&WfJTefXULuZHiC13H~;V5$aax`
z<Nx#EB=Y}u2kZQ&2Y7)=9Z<Rd47~u8z4-5hk>(FTMgL|7+B@Vt+zkK!$+6cr^0}YK
UqCg!_2!QK_vc_{1*dqLY09x|HegFUf

diff --git a/docs_old/assets/images/readme-light.png b/docs_old/assets/images/readme-light.png
deleted file mode 100644
index f59fe8843787c25f5cd994e7170616561a481df2..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 38943
zcmdpd1CwN3ux;Dsv~5jK+n%<&+L*R&+qP}nn)bA9+qPeQH{N|e;_Zm4I;Y}9RMgJg
znJaUx3|Ej7M})(L0|5a+l#~!v0s#R_0A2^cKmp$$`=}RyAFy^38jc_!NCW@9Ku2uL
zjDTO_IEkq{DchPjx#~L@gSfi7GMd|1IvVQR88g~Cm}Xw_;(~w>gGh=BskmibWV?C+
z4nIe>r*T}3uj2$)q){{q{J<3g&^gM~n-?uA7h7ul03<PRoNE@<l}oC=i>ggF!F~ax
zd0~FPjL=p8>2kEcXB|I}$O@y`{6K|)J^u7~y*tV}&N}u)e$IL4+J1zgp}_ioD>QO$
z2}=EcSCc$FiMjs&U9+_Rzuh#ou>bAf6BUIaDN1U%`i3C^Df8WQ)7o@dRc+DypFL;_
zs`JUEr;i~lkV9PZoen~oNMjbkzsZA<$>s0fja-v(($v$1hmSNIE9yMr-1c~*wbfdm
zos~FScqCs8TQzT0!(k)2tbJ+gEHLmi)>?O_Ru`V@--A)UHiq~&Icbd&eZEv}>87Kk
z<nYhn7X?M}t~B?bzkVUh(L)q}yyF>wKaHDsrFrD^c<P;L4l&a0;#ms6zLv9}9TY`q
zN;0WT_jDxSOuet%@!;kysyD)=<pM1g`edJ@eZ?N3YW>8A9(Sa@SZYf{x32BoWNpKc
z64zA62_}jb``BNv%i6JN1G;nnI#^lTld;n7cpOz$R)(RoydQ)jH_r;7DK4ju;P`={
z>dBikS5Qz8U_M`xjD?Mz3Bt_GjP?{h?xC;2M0?)9x0Jfp-@g}T*O#5Uru>H3F6lhH
zSzFd3<9k@si{bM!e(3Ml@Dj;%bZKGv9>bBq?M(0o?l5Kko!hsi_-}7v73~AD@Q)HB
z5o7(Yr(whWP8ZoSC5+mQp3T&i@^x#D?A)6TzZ#??`Wl0c9cr`!fJ<tKdj06RM#t8D
z!-EAf8+KBCw&zoPh|=y>XZyiQ&%F2l>~bkdUb&kUJ-+9|2}zFoQMzt@eSM*`jsRf3
zDKQwu+yCXw5IF#1LuqU>U%@QeJcDj*d^~m2_xtmFxk|&36)PpBtf{V!QwgE>k)r$i
z?|sdzb~ti*N2BTIQ-8ZF^asT`Uvo1Xj&v$xv+zdr$wKq{1MGkyA^yx{jm;MNpGPdB
z&3Wg7EI6Ck(IY<p#aFTX5$oTRR1}Je`EI>!RD8PL3UDs3;B&b@3bPjYHh&vLMW84E
zEYW_Ei+FG;K=6Jm4!zysa@CZWX`u=#m0qz-aSyR|=h3Tl+RK~C822q<THp3C01v2r
zTGN3yJq-=2O{<o+_v2!VmTk9l)%amX8TA+hiRe;%i;$EC23+vb`{x%j|L4Pemqw$B
zyMcj$k)eDs1vSvikg!P{k}Y%IR*hR$l5jF6+3IfjSri6xTuc6hd=O3N3Vau|G}Kp1
zNS<P&&wzbaCTdq6t~L7)zoW5q>N&29h1)(%%xH8C55(yr?G*wd=tKtFR1ypU2<TW<
zRr<5f%_fS)w*~<HFM@&4C!f=d`rF=<lp2oo%6{v*M$RxA{ZVdYMWkTV&#bN+MR(o0
zmZONhy}b;fKo8q2dvA+ef*~P7pX!&{i2e`g*i*>z0ki}xtk6(1?Vt1Nf<4b$o}csc
z^R`NWM+*{Bi%8Qr%9n_I7mgPxVg<+$<Tzs1nZmvtm?=4wMM&1;KC21@sKD#C&8yyA
z=Oqm~0;JH3*}_m*Sx5?1!s6gyF?bk~!Gl%HVV*s1fr`~ad9e$kpSM?EAqiqEKX-E=
z-Q7cz-j@}_CMRU3q@1epw?<ZtXV2DbU98sGHKoi!21{KpneWJ#v3t9S>+i1;h&!c4
zthV+`Ei2?wRS0|)8W^{ycXE&nwBeY1mFgWF6)zzc7v!%pO1^2=%5_({G}I_#qWbjF
zUjh>j@wx^(FgA1(Wjn#ERt}8&Z_GlTo$qcN&h{49y=)<bqb_GhU8FRe!^Sk-UBvoY
znAZ2!)GvKwe$F37jw67J1A~z!M&DaH`oXic^~mk-ZtPzi9BRNIRwyKxl9WU(okEX;
zj)&*jiekVU-@mx1m3Jde{e?TGCUgh|vxvVp2q$VR)IS=>i$C{U<V<2lJ6e3`yu(M?
zz+g|-a;Zu~Q%mc8p-3{3pOKM~U0O<t{0|r~GH%|<0mCP&5DW?Br~P8txgNE*sOxYv
zpTcl-tsUSc*nCRh_Yna{S%~$+i~g+66#hxUl4*|W?kO`XN%zR9dH9*wF;X<we!zTz
zj_0_D)g~Ts@6WLLnmV`Y)JQpHoP_~@CgN9z=ukn0J@~ZdjPhJ8Q9Z(+gAt%Qm-`(h
zr=v{XN<yg^(fZxlQM{w!6);j2!bz$)MHFUkpOop6xUJerX_`8gUss#Oi)oZuEaVZ`
z0*sb>UH_RBO3UFf=MpdGskxpvK|C#sR}rG=BAWZl-AlEltG7CubIR;rpw)$cmVocq
z>Fw=pEF8&{#V^eag*!H@TyP!%h-dbgG8P4A;o%>W%XE3?#$cX%J|Z&n+5q5I&HRfe
zyH`?Dns~#P8s}N%Zm4M|7m4pv4k~yMiwh=4C0U;aFT_w1hW+OiS4klExsF)db!}tW
zNWYdM>HFbQ$5QWEa~0JIlaFuYk2bm$>m!?IAbuMFc|Xtd23~%$xlX8u-qB)k{iFtJ
zFy~qPB%!RDy)>Q9{N0JmO>xSmmW7JG*`0{!_ebJxUrs=J*;6tGAStU5ZeQUfxj3q&
zDDqTimawk^QWVrqXW{lP{XCx>ynY=nb(-RGT6QAO#LZ`7XD=4^$n%Mzh6^E*aLw+i
z7_`&1tN8G*j@ef1a;%{u&Q*}({bbZ|4+aMffZFvTIkZ)r`Ds|;YP$&$H&eden2pw~
zhmi!yR8GH9zf-1LiS(txd897fus^c`uc7DPZN1KJJe9?TqoJX}0sv%-?l0zh71zBI
z)!ky2Z8=#*sSw0D!#?yP&P^A!a|lUswbs?$)I>7%xJi2>0fQNGOd6}?WyNAvHH<wM
zHF{CfCTMf@CAR?#uLKtCL9cdx+4EA*qw3BMw1yJ~qvy?i4l@pwY->AOtx#DgMmj5R
z@rC>I?;;!z%dd~u#MqW-Ygu8BR&lEihhCwdl}^hwe8MnnB;D)K9N9#^eSfcz46@*o
z*`VlJUtaVxV;^}wf@IiXB%{Vje(e{ic!7qG*gDL=-ja(}T5X}nbxH+pW2Qj0s=60d
zZ&>Hh>^_ZkuWv=jU$2p|>H4~g3LNKEsbs%UCs%ud1E1dsxxKZbf%#GEt9q+Yy#7MT
zr|8?WVf~cD`76Xn$7)c#a!`!MtD~#LVXO7(GziSMmhD2>!B)bjylCuaN>#P@Yv^UZ
zK+5H+gG$8(1BrW3eUQ;eUVB&-vZ3$urfYBA`eiJ7O$X9GjUQX1{{A+Jil-U68Qco%
z%;JqDNk6WF)KBst1ek?S*Rt)|BxQ=gD{Lbj;l>Zj*y&@>i?8?V@A(QqWk5=rRr-}c
zFFm2&jcU*bW>^irrm?rDX`^3gYpAyESnJlWwY49Sg(DPP_bdjTIUmrOEW4EX7Ec3Q
zEiUiEV<{8Etdp78^y)WbvkkXy#QDn4c=A^^zQz(vp?A}efa8&he5NiX-_wNn+2v?F
zw8;(IO{-1xl_SJ7I#xjaYvA1+&GF!E@6aOKy35P^Ace(rHV5xp!>)Ih+Xp2DfCK@Z
z|E;6MMBHDXvx?dmM^HPJF`S!I+Vye`oH$4U{g+1+#@vVJ)Z(ajAc>@E#Z>6{hNpnu
zWp|I$^b+Kz=d&jF`O!kW-uKC2t|Q%j(Sduu-J74y)Alwy^Q<bs@}~9`V0`ZSS}McO
z@?@RKSm5c>aQj-Z*1)vv!B|DR@iiIWqni@P+YfZ|&v|W8oAnl1_t(QzWhg}4hk}ZV
zcUsKA#hU>c^=E7tcTN6$QsQAGW~8tNe`)ATlx!`f049iL(6U>RX5^22aH2$q45Pus
zfF_oVY$?Xo9`E;QD%p%AB6M`JktiJHTpk*(z9FMj$;7ORqju@PI)A^~!Kk3%i%T|=
zzSp4Fk@*)2_?T~VViWc|kvpYbDK<z{_&V5*skNS1o3)OmX^d-M?~iSH=Ga@<!qfR0
zW#+x4i2*1gjY`f#`y4dYzt};i%u44d*mMgkmh!6BZos#t<G>!&Q-85i$Q3qmivptC
zaRp(ZOp8y=#(GR=X=8=Vi|fVVT!{W*>7a{rl$@~p?Ny*$X*xDyB!Qbn{Fq^ojeUZW
zfg2O%p}2rVnW5uTly+Ff6b>;mwXY*xDsD?K0C-9!3n^8N#T5K$5AxpqMUq=Z|Ihch
zRF_9uf>58HWc3OaB#}~93HMMq2aAa4I|QXbdZ6smaVK*-&{Y1Qks53-I}QV~KIUdy
z98M2qtXn}9&u4RP{p7K+F3+im)hbyjWqDy?<EP8TiaV<CCe^`Y-Z>j~E?FECj8JV$
zOK~--0^Hr2wXK&|Kc=#i=<pUlRfl8m{PjX%!j%%Mr&6Unxh&k)J7ix*|0FzC)~x*4
z=@BaHeH$@*qxklf(&`O7MMbqN7bSgyPtQ;9UYhO9?+tmrR6w(}em%0{Hg*GlQn$fS
zge5-G2)u$?_%%}?MQ^}ID_L$|jgSQbxp=V1k~nxM5)QY~V8A-Em`&LOYob3|OBdw|
z3-a$?^3e=xkRX(BASS<N)MzkTtJ5FI0iAF@Uy9Gn9FM|bkt>V1+HFNzmad%tE`zx$
zu&+6XjS4bCA#&+_>Zf_N_4-KU!^E;4D3WYvUq8{?*zi7i^;K{(9hT?Qo6!+0Yk~Qr
zFOC*3jIg{vH)b2A^l580ms3F|fl4)8O)Lc$XAp-XJXq?SStg;OP2?b;O;6$u;kM%#
zkHdOGMWKV~23rOx>_<tWDx4ALgbDXqMBW%Z*&G&SmV?~xa5&a!#lB8dKAlLX3k6Rl
zVjbjVoC$|xuMdio|GlP2Hca?6jUB`XP}rg$ykZD|Oofwsi#}MOf~-IkY+S#TuhML;
zkalr#k&~2^G~4WO{twxFl_4=vcFJ%wDbCMrV+GV(TaVj^=@5Ty?w^HUS%1G0*H%Z@
zb2<$lG0y$k<JLn@_FhI<JUX*4nmVV`M)cFaS&ipDzU^}jLD|fCknna8ZgM~iT}iGk
zoZ%Dc6^pRR_KPGpfcpbGYUnQ*)_MF(8^DD;st#Hz6^3*T4nCLjFJ*{<Ah%I;Py(4O
z_&^&L8~(&~BC6FaoN{vLT=W1TbCSJUe{i^WpkckwN_C?GMj|#?u3nto%e#%!=_DP&
z+-F4*G-udFO8vVk&nnqyg9pr%^ImxQk>~#IuE4haFlGJz{$AuUXY_QwbeUj*d9>pw
z0YO$-01~RXnPG(4r992nDcfN-jNtpLY~XOfnbpD3F*$m6S>_^xnOv<=oalDC+4tJQ
z&#&MgQ_MsAdO}nbg7|ce%Lol0!LiLMdY+Mpbh$7=;#Rr~n&CoZ=(GJ@e<&smv;gTZ
zsu1L7n7Bs4tQ;0SUf*r62te?m1R>P$Qsh4o<0Z4x96sHn$+48@$*Gjj;@?aVu?nd_
zE$t`F6VNrNNQtaa;+6Vcp9d1Op0ML*U4FzNrQ2K6k>a6ShbGSDJAQnzV|r7(xFr>N
zI1W(iEc@QhNi%(7&PYi~4R$!6i`v-KpzNu40AdNeYqQg;;0*a;q(up{$i?oS>)Q<(
zLI><84pz^Hk4!&Y>gn1?I>kjqrRM*teSj-9`1A!piiy8{Sci|y)DjBNukk-6QG?|p
zy<cpQjxTwpImq!2ymO99&D&xOgq}kSSoi|H7{n%*tl&D6$|p0CirugQVYD`moM@EG
zBsNjYG&&Wzz2Ou=8?uT&;keZ--J?D!GjqMId-{?+^AcH5U6ipCvBx)y5{X-A1$>o>
z?E;o}tE{!w7nd``-0!S9Kz=GgrOJAx==@Fg{%CTXfrdu$Y^W$HA%+s@y10iAhCT3E
z3rj6cWWz_cQ5kbfztu0Mjbj-v-3KBM9Si5*eM;60e-l(EW&9b!lIg{Z6K^hmtf#Dz
zP92=U_&iN)xWGsXUQu*1a3=#rKtJ+T3q-Ly8m%F*_muKo&$w5Y&p>Va)`>R~b+nbi
zZzcM-Pu+)72Q^9yi-&Tdu#;TwPCArCEm*CvLidM-k`BD?<rKDocdosJOt!%t&&|^!
zr%hcdM|9@cta}7Kdziroi$SxfZPVuVaam8#sC7LT1VZBFjiYWixCFWatb<CVUue14
z##6hc&ecIHy=boUT5VXFI?WR)>pPgdCl{n{)#{l+ugB{>43+E)G_Ae8U090dyhgdx
zR|Q)*tY%sJ1sl{9->XTCh4Ad_zW#gj>r7Zr3(ufF_9V!r^=zsQR@{?tdkbp&cPLr{
zX_>Ce?J2Xpf84=!u8Q4IdcD`vXn6C=JVl|wN(0qj4!o``=Z47F?L}Q^UiY?RASfbp
zzz#Q)_bV5*$3`<9)RUJi58|Tk!Y;l)d|+-Eon)hu=OrU)-rZGLdToF>A$B|p2mHmW
z7#8?;i=mHX9h$Iu-7VtUOSGx~_zgn{E*IqkF%mcIO}c5*>UQ1z{3b3g-YfNaMnOju
zK=i%zK&%tYPowX@q^p2=l1#s%K|)irN=4Z{^pi=;ZWb*J&XvufUE5cq$!s=+*>IR%
zr^yV}ZxvOPT(H%WDJu$t!+B?6x=<3?Yc>_++uBkj2IBsU)lE>T>xXsFOhCvCf^u1i
zc|}M$HCnj@4aPuH1gm0ETt6!&Nf8E!5`H?=<9W+Xt~pbc_P6p>{#gyPcz;UZE}eFC
z(o2`_^_(H|_N*`r$=-aY0Ln31>GF$F&p?S2&c=LS0H8UF0AEb?tgld7h<v?Bh$x?6
z3A!)RC>&?UYAD3#hxk<jlnX?8sD`92r7SJ;DGd_+Ee0ru(a8P6dU4j;$mn02r!UF$
zO+vZo`RSS`Tuj|sx4JI7^NherQ&>r24mIkW_zeVJIrzyi3a=V!-b>z8`BE~i^)Er<
zA8-h4GKABk=(g0f$O1kw8%*h+)s;n{9~7oNX@Lj1e_vFm%<BaT>>F>P3EhZk$;kE*
ziLv}-zx@_t+4%ZPgtzxule~Wh`PGDjk@QywoRx1v-c$7l>je;k{{hym9-oJ)di<X;
zJXh2;<Ed89Lmb^yI&lFdA2>zHWAmCpAMGf6)iF1JP=6Ltu}HeBs+D_^CM__~4fXV%
zqvA)$`ZTLmL25FPPk&B2O6r7eu^zx=V4$*`4C%F?#_9P!E~;H_=^QtpJ1!?nVvF)%
zZoL)Y>Y!JRKi_wbqtj4Zy61-ie*L3m5rm~Imc#2t+na%lP(+m=M}m+eIbmdN8AXAd
z69mBp>-Br_|NO^9YhZ3q!z`SSjEhYh6XsYQFVuquY6WGbAk#2CP3b#i@iA0P3{f%Q
z4-<TX%X&*flSGMD2U>wr25Udl%(7jpJ!zMZ4n|2G6K2rVF#J~^{b9PHvHGW=W`IfO
z@z?=$XYsn*;e3Q`@As^h2let}5O#vAFKvifp&O*6wBFcjgwWLf@i;)-$sinQ;}5QO
zY9f^`2Hf*BubL_w5xZp<JxIg#IJzs08=P6F)X_^me-YxYZajakytdnJFbk#_4$9Vx
za~pT(ZEWaPjF^#O16u#LEj{J9+JG8|<nWhTkSAK$6xr{n7cC2mvm)#D7SObdZ|U_K
zm~tt`xTPhRCda8Wuf-q;Vgm~~&%Hg*p6#!FJP|$l@cojOhTTfp`fvBgEJw!Ii^gfq
zWZ#TA#}5tR`PjE>*E}ciUngQxudlN?=Vr>!C{Eiq`rh)n++42{HRjxXapYqpH!tK&
zKH9SeC$_LY4Ix~d#(#@3uv2f==%yQ;*Ig<qDil?cADUFKm^6k7Tn}%2YkMTNx4IQD
zAP0ViMMsm6@aO1l+GCd@;f)>D<{Sl5*i6-iYZo;PBG|nqxuslQiKz<F)b@OE6eYRw
z>N@ERm^7loKyW@rIl9@c@kQY|VuAVV<E}X2F*DD4wcJSeeA-Y%9!={@q{KSszDZ!d
z*M_~`-r=Gd0^VF&8XKv@N`zBYvV*sm$80$uN>q{$h?G!oVBdpsc*Iv6?gp)7^~FR{
zX;w%urZSTm8j%5_M@MjwLa236>PTM57^Fxq;8kF)lA|eEy!PhOfhMNorkKI=<V|%+
z5kvViyi5+adqc-39beKQV3>6`7C&)~>8QaxN}GI9HRM&n$U8D@*7?P6_U{pM6#)RM
zp3kRsWiS3*Y8o1Iv$-w8Q?A3kEf}BKB^M`<tc;QvGfT1<j7-;vbRlqGtx2XJl!&=E
zRpbmlmyP(QQVOr&-CX_w6atgt!fMY?cb*9Fs_os<GE&hbk=<INB*S^luVcGm>J!|L
zuNXbDco>qC#YX6X5G0d-G*t66e9@9>+GBP3fj9y-2pT%v^IR-}H5xh+VX@G8YnLYd
zFZL>)GINCsSVL+_C~e)8iBvkWsf@cV^TIF%1==fC`cxg0CWbqHUs+rp7~!z6<m7{i
zSxKT+RtUl{0e!^X9#47SqhzN^Mf$cRSw0qeRJLBalQ|8<V#y>fSkS8ge4|a5?W>Bt
z>P#|141aVbi7g&AguCHeK07ATem0aYl&Yfd3CF`O@a$vcd(qPJ`#~mP9^nAm$YlBy
zLUy#hvEeXJWA&~{W7XlN%4(J7hb%^d{oh>wgfJ?2wZPs?7Wu9$9&u5be$Nl>_A!K?
zKSu8cyr;1Q3*%bWFR@xT93tuX`M)&IdZDFlLo*0uV^88Sbcs62X8B+XtYnF7db-6G
zn)CHDfqmh}kNat=_RTAoO|*Y;8_*Zu)HbuYP_50N;t}D>z3WOE<Gt%^F-rv&Esilm
zygmKbptKuT5!W$8UL#=h81O9^5Jtxs+1LU#JC)6Zcktla_9j+|_Gs#KfnMfsu<a0i
zDAW-F<}F#PCaujMTKSgT9X0xCXti<lzdjGfy>}l-&Fg+c(WuH|fVzYd4E)sO%^g1#
zT_)q<3?jXG|6I95vdZ|wLb5bUoMXaY&7W<JaHfCU8(FcciL%o<$v=nA)`A5gU~Rt#
zlVvCTi%Nc@oVqG59?#zsV>G#OcwA4wJNe-^_t%tsFrTUa%j*K~t+RxFEJjz*4sM>B
zs3IIMpOS~EY#h9)@%VP9l~GziF6b*NBF4r=xZ!EHJL#Usro1k>bT_3b$;(dMR^5=#
zLTHn1*X{5&Jc$Qb-97lqlgYgQX?@&Igk6YB3~8R4LmNtIP8@bm2dvRRG~MgNynL=M
z08pHiloSMyxTYbUr@JgKfhjC!W68sfpyXt0i5^3K?mC{{hvFX?%iGiTQp1&A!O(F%
zn=^6)K&@H{NpiT2z@+1rvCXRJztlVS)^dDK3GjO_xkqnv(Z-*Z3qHil_&V@4gi>E!
zC*K8wR?Ec}|4Cl{=r+@wQmD+#%+Q+LvTkjb5yC{=ZZFK6&E$BitLaB-t`I!RlMPme
zz4m(excduHjM~g%d=;LJt@(t(hy(ce_O&<wFo(uEm5m>1+couFF000`^${4VnLApR
zn6<&q9B*UJ&E%!cOr=0~W%-{0ehS}h@uIS^$$ucfz@y$FKwhxzHjPmI3p*b_VnmC{
zawi=rSMnPbAkPZn{ebH^H8eU2zrFjWO*{dgh(kGzH%tTDmd&K{5aS+UIhOu@l44A~
z8Qz`p1)Cqldwn~vb)9qmv;h%rzIn`+6&GL4bJWI5bC0Ss54Z}KC6bR!fn)zH)X%Ir
z=Qun(e53=y=>MSTLNf&SQAtaRf<!NxqzteLu81N`-rS5Ro<Mh{)7(D6u%4~<i{0_x
zkNh(6&(?LlE-3SmpR#Mt>O3{dv8b@8o4Rr*3;TT&6EfI;mnOBolQ1f4&QePDpVA(t
znv}0>G_4F#e3qWxD3Y6@lmcXz?Kv+#^xq1zJo}2hUIA@!-cR$>H+>_y2}yB(hbOh#
zo`&^)nU77UyB#*SS@xeZu35Ek$CenKZxKf>FYM89*unTG?L223U_-8G;EPK2i}A+T
z2~8pPc17zK4&U8t`3`vdrulCZfq-UI)O*T<l1=f3nj`aX{OPPQU$NR1l=&BTK?|#-
zVOTI$k}#x?ZbW{T)QPC=5^OJ0u+lX0?dJZvL`xEK4x8PvnH{(F=?lCNAQv5<9FvD3
zaktGmsgY2XYO4w+c@^Tk47X}$pPR~R-*b@nvxYI569#Z7Kw*T<<m=31hmP)4S*6pF
z7y9+Xh;V1t(VNzUy&SwQ0p`Lqho|GZaLjv$z80ALPk^N2Uw`mxEa4y`qGkR?)*=Q`
z&X0AUEZb_bJ_-pZHpP&;PG|`KiZ^S`R7M>_;HfX*AKZjeZNWvgxv|xsB^5Zhcl(!j
z=&-quGKuOK)xWm;6|aN~A3NkF9(ZuTm=4}xl`HUg{*Zay#k#u-@?;7Eq@r@{r*_I%
z;OT_Dnx~ZU_f=@$ooq5dSv2eIhkqkp-V=YP>p_y@uR--Ks_9HsTrbC^;OYWqE^A0w
z3ehV{W;;!h*=^SP!RN1b`^=vebag?4?(xsZvDB>Y9JW-m-$j@Q^HG>EuBsvo2H}Fm
zh&u69R~655uqkzXXv%45y59EmXz1w=e6M}OlJY8OvsqZm%Bl=LxD4^E)KoA#H$(K(
zfMuEbb9(56G39cG>^o8Eprr(jrC`;ookp{*0y4CS2PY9om^59s&r@C*)%SKr?$!Tl
z0i3knj72<o9QNnOuWV1bUEN6Bx4oD(|Bj_(cuKDnT4vp<v}`PTZ7)EQGMeVNK?l3~
z$)O_Y(eB{Z3T*(5#!IJV0vEl=<!&&);J!EgJc-?nY_eSi>(EY#6Ja#(!vi@=?Zxcu
z|1ZMn;^A~|Lw9l-#mk}2UCs^{?)QdG>&tTQS8~m{N1MUC8kl9qqn>Oms3|l}4c8A1
zSXtqXel-NkXB}kOc}A_eKYe~C!5kyqc^A%~hvgF7GWKCZv%721s@9CvDc|OnJV=-%
zK?+VIt0M-C+gpngG_F!K9E6dtF{}nVf|=!&f#MI2zlX%jgAl?8D1B0Z4x2FhS@^ZZ
zGC;|%ex}<k=IG?^y48{NoYL;Q_<^qzJf@JpFUb0W8HAF>dqY0;!<m~-HDsto9~Xhk
z7$rnMj~Wh$rUpgD=NrXcJ4}M#A3VBsCpxD5`|`+(HFyXyC9vRbof~9O;Y@TOsA<Q<
zWMtY(x}=Y%fCx=EUm?2=GIUeUYoiBW-aLakLii_!&+jt|);Cy0(?+dlHOsDV-3BQq
z%@(hLyx>Z{UK(on1U~w+yTPcd-iS!0TDtOIVa_yB{G*URWgf1qYoE{9OGMyDv)T;`
z;=CA%dxA;WD6`&AINS9o9*FJ+1?P<r1IEel)rfZjfKcE8SX}0sPNgnJKf&g;VtlBk
zS^XPDkBtayYaz@eey*T2l^BhI>70+7n0^%$MVRe&PG$|5)rJ*)DABwq+`{7m$_cdm
zZ&LLy=TWQ{ki^vn#^GDgX!7f{4bKU)HLqu1J<`#hl7fzJ-h2y6<^e*bD<!vQkJACY
z`Qwop@$vSip7&BA0shLUPJbbmTL#GQDzq=NiSC=l14-7zkamiZz2-#fxjG3h-G!Tn
zf;K!suo$%4>U7+#K=)WWbmEl&TK*V;SheL0lT2_87L~EkwbJ#*C1%&UytFOX&+A_F
z+bZvwD+ad9ZA~>CQZ%kDPW+j)3Yd_k^Zq5l%T;YwdtzPpSI`Tk??6nktd(j?vl~B>
zyVF1ZRBZO{<s&m73Q~daoE8<NyL==@T_8ja6(H1NwK<&o{RWoVVzs+As)hm=jGGiZ
ze_CY!)-iG}rik(3Puv$*RD4HF^TMCEC?`2_!d`6Vx@SqmP^qz1+L+xa=fdZ^AM{ca
zLbF;oKK+r*rqus7kypQExyEndlY_J=u2g}5kpP~&@pU$8)t>)4oaJsnAAl@&fN6{c
zWu^U9%G_|52Ne`)`(<ZKIBL|mJuhb{Ls0rNybB_oT&GO}gp@s~SuNg#7L~7ridPs0
zSm#HDZ2YUN=@`>*T_3po1Gf0i<e!GR-UM^_Q@v?y$nkoxv0G3atb>xvGc2`}QY|j!
zW&U~y?d7J@O!td8CSdMHZCO7t%hiLyRaZku+llV|pmYc(Vk03BwFe!t0s>Dj2_b_L
z#7QUjVK0*DYj=w0n%@6r?O?<+(IG!f*<AcJ>oQ0Jx#KdYlJLc4f6W*9Q{IDDz|v9+
zDTJkbd&MxG=yV;==6&AgUHt6sz7hJjr?$)ONvPiP#@D~^<RY^28Nltq07sDpNbOX$
zovVDSp6z^+<um~r(dX%1261rwbw1QCu9)m14Aj8$oit;QpE)w$M5-QxvEa9l1P$yW
zKm#wN??QlI#>t5cmZU#Nnnd(qER36X4npHNzKTch=MM-7rtUkYEuZ^b&g>~l%uRM3
z+Z%o`2r$qf6o?O@qC~9VA|0~lHT&Bsht@u90hKM?Bh{u!PcUr({v05KZncJ)c0V{Z
z7cG4A2y8_N#Ny#^?yo-6nXn4ThdxPUa5MFR=CD#^YpvXa#fuj07^X&Fh@i#GBJD3_
z@y;6DuX`Rx#49zRg(V@IXC#cWOh<n-Ioak&uwn8;bE}Yk#FuR0y>@PYUU`v-a)W5;
z1_~DXO6de2q*PuZ4e~@9v>n&dSXH9OgY?Dy>1$=QabxWkdiliAb>qm{@Fi#CN@XH;
zyB;^Q$w-$2Skv`RMZkpK<K6E&{$bxp%OVqx=~}yfj!Z~Mc-l=)x+`j7&o9iEK<ZG@
z^fcadxpnirky^uNHU#2z=ht|WUcILAkAGtXJhUmcH#J>GDpd|10>?=a-Rq_*7&hPI
zTMaudFyhXPIGE^(xTTxFZ0hJdB$%KLp*uAvOyQ(pC1mEwQ4p}-s0K%v9R}@c+K!;6
z_#Ii5#<=^S`xH>jGdM3?Hm)*4?tg@YqoBYweZ<j;a)!`A>=xkm$JgkKK;Pep%)5C#
z<$~aLC4=zy_6j)(x2^n!5u(NrgbB0gdrwa;;hE-L9U6>h=DajGcr6`9kPSof)P^4m
zkT|%<41PskBd|D&UWM&uzlQZoip<O{1qMy3k$*~{wXK|jc9unTjVN7<{ixeAL}Xg$
znjS~~(_$v~(ZKu`QE&tKs{tkHrgIhjU_;`)#G4*XrB66dr1_9DL4v0I`S;tW4=N@m
zpfw>VD4L0lAPWl}3lKV*gBDG-En_wds;X3e_j1m#Xhwi**}Cd}SeR1zuNC`*mxU_-
z5xwYlV9A9ibh9Uf=Mnpdb?bU9Vz6-kr3*ZaCxw>4Ds#T27T!!B47JD2c?<iV8S|_E
zwGX}H_`V0jZM~r*n~L{@Vy;BSOQesWR13W>Z|{)Z^=-^1r5A{Iqg9~+kE<M5n0>j6
z-$y^lw$+V;rT~@R$IgH9n0wWQ-%&79f8+F-)y{8s?k~>`eRp41AI{e^r_yos<4=(l
zbS_P7^-<k(84eo7^c+4X-vi=H%nNAq2zRbDnBHV)ZeIr+tv8y&5%JiWZ~O0Ia7g&G
zTS0V|nGpiBJ6#$}^Zcw|-@%|taBOG_&v#u!7`OymFA-Hft}89R*m@JX5A(~Xy%k(G
zyRI#0i-TUMF-elm{6KZAu*rwZR-FB@emz;HFXji6`_<jTFz_UF9#8BC9N`uBMT$z%
z#ch0`OS$Xxp&;6;kOZY=P!WQD#ipe4c^SC0c(KpLnCpc@%nxHL8My0e3c}190a8qJ
z+$mK7i#HjE#@ix88N-1j?ZuD{NE1}!1uziEAN4OD)rkdg&46QQH(&b)A+l!2-k5$O
zv%?W(#X%psEi3oxSsUYjp+r6_-90BzoLin!zb^(XW-U^zy!n{!4B!ZY*t8Z$0`d*j
z`WT9NtFGne(oG*1trXw6$YNOg#t43F!3nmkw+H|+$+}-nOdm)FDTe-8>9F&e`x%h_
zt!l6c9J1uoc9r857IdzzKT2)~s9n+`%7(cy3M%*aPhCHAxDkNbv|VX_eOj!Lpuz>;
zK!Z_7?3Aa+?mc>Fe176OTzwIsAOXNnGB_2C@JdM)lI`3!^r*<i5r~GbuU^<f^l$d`
zzC~44O}+Su`u>Co68~$IpDr+s_5LdQLk$6@EeAaZK_-lq03j2IL&R9i<$H@OUnA&(
zM3dj1TU*inkZ5Zm{Mm3ZI1)BWL-T3S<>LtR_pgmhK!u0Ny`!@m$?MCHbmWi1;!mX;
zg?sD9snpK!i{yz8TiwCEOp(Kaa`(Lt9s$g!_oLgCAZX1d_?+Bi1~PWYZVvgP48jy*
zu2-s+xn1*~lBbi|!V!3q{^<6*V^#wEW0=qut-hyj=-L~0iUXqlwv-iByBkAq(7vsP
zTohqj!lms`$+i)bAG3|7)9`rMoA}3k6k2-B#;D=Gqx&C3AEgeH514pTm9cuYMP%*A
zZHBW66KW59&ZpD41dpRJXt#%K*e1nh8YMxt9B&e_YTFk%C0~vYg4oyeHAex;h3?)$
z6arTdjg{mtwf`wBQhuAou1eTR?ug?UB_kB9LXePku@)q~d&y*!u&Wrui}hVUqZ1HJ
zCa20C9OddwZ}xU)&x<2Ko0$<Zg0M;5^qd}QvCI(DwvM5|%L;TBp3!!K3xPQoJWkA#
zweQVE+H+?Q4iEm!^A-5;2MIBssWf%1>AFCZ<8f9G)Oj>elS(|`OfLTQx3nad?;R`a
zFGNv5Z^fT??aeZ82T{Fa>r$t14XU^r<X-#l>L%_%NHT*OK(>Fea99rTU(B;jhSb<g
z-=A;p0*@9BmQ@axOKh8M4rlG_Z}S@vRNi#qo0lxEg2fsNSYO8qde9o)zj?h=Qz@-t
zYtH}Gc7HhpzrkhD<3{XB2J4IL)z?p*a@Ok%k3(gl*`J5mMlSschU%P|E@*=P_LM4x
zUd}K!-=mC(@h<t81+yY~HgbLR)pct<hg`afYuC}pB+34<B^M2B4hZn3QR=9PMGh7o
zX~dq}Gb3KTS`4slZN?c1Sma3>iCOMRsF-xNdK16WLHrXWoX-U8ABkrKMcJg149Uo1
z(raDsexS~7c#AoF<Xy!iVtw!QVJCInCud&l&PL&(MJohB<*gRg+9zD#;uz^4qx{l3
zj$pQzpt0k`H`6=_bp|NjdXO!TgCR~1i9l_534xXB5{P__1~AD8ET(m(UQpdDQE|v1
zl~S)-v8xmthhrf?*>hzxT!H&fA5-ut?muNFaJoCNUW;BgGKGms>_XnGU$*F`yV!4X
z)tjS>5BdB+NZ-r7rv=i2L&2qD$RD4Wl0X$+@C_|}DTUM7meFnsy08R)JVRtLdFv2p
zfPkz8C6Z|_Q!AVsW&g?6_OSi6{LfOnpBEtSrV|;&*I`X$ra3?I|F$|Gn}wr66bAQG
z|Iw+ewmN1-whR>}$v4+*`_3o@p<QMpM6i$VjT2P&!c+|!Dl`um0cte{Tp|9r^?uqC
zvg^$CYJp@DV!6!9CGu-j>qT3uZ+p9Qol25)6kUC5EpxGnY6=%#l8{-Iju5r*pak_h
zl$pPphzB&5qcNl{-?WyPlQFgJfK}VekM47HF~mqS&mhH@?Bw?^A1xAknY|yIB6~d>
zN&+Ie?yN&l+9L{9r@7beB?e-9jrf$tY_S=&-PFa3N8slaEQWLT&Jv$>5YO*Efvu5v
z{L%)KPGn`_;PpRuY3e<?*L}w5qN`$1eh-`*J-_9i^O7Kh6BMYsxTAEiSFBsGl%s+`
zIyVX3sT2S@(fU7tBH&R?<Xrk%{dpt-4dJ2S09EizSLl@{DsAHZ`}bL=?xE)H!omVg
zjOG$^(qi(l!@P>6ox)m6^|lVIW|!Ps9~z-F8xV#frw9jXgF`H$(r4ES>>Mn|_j2#-
zF8<zxt#tMvKV2Mllh4+u`M1u3UkS~WaxF-kK)|l(lJf-7M)+gKE7#9w)36=<#c}@<
z6O=G)zFj-4B>Y|Y=EnP8bn61v;SiNY0Zs5GQe!nf^<r!kT^T0y1bJb~Ob2Gl*TSo<
z-Gk8}jw&`Yo;QT}HEeY)rVQD#7uXmt7hb5w)NcBL8%!0}5S;i&v!KD`+d2g2Nr`gM
z?y8*d>()N*(s|c3lhS>6i&XCPu(aW~eqRK&(q5FvllJeuv3eUBSdsUa9~<kj=5P)<
z@+P}u-Z=|;y4s7ZcYsS+_<s`DRBNd21?p57>}S6KFUI@b%s^@EwI5T@ivi#>WS?_Q
z!#|nobw0kUbKWjnq0;oTE^{Qouw80db3#YpNc(FiO}}FD5xwFRDOzw9Yr+q2&0bv#
z5EXF18STGG;Puj+8AtMJUz+=+;W9;l8%#q1G&}jQ8RsQj2C3DpxdxwXXo))o=$VgU
zC@s5Ha6mz~PYU_sue{d4>h=54?fOJKUw1sYQusJdd>5d;_hhvjPd7J}>g$2NxY+mw
zTwkb+ns#W2!p(y3<FZ&y0i)0Ly!)e|Kb(LkrX%yo_CpcF+w>;p^ymWKV5P|&-ljD(
z_j6B;x<Z&3We_G9C4!idJ%)hCmHE89x6b6!gLtD!d!Ch7m+9~jQ&e1Bzc56n2c_i^
zsxT;8re6MJi4?U2P3mIdQdoqxSV?M%5MwBkDB1Cn%U;SyKW?3p@~*eG;&gshu6>=D
zPiViQ(ctxR`w5G!=ZGpQE{p$nNQpPCk4c4>bX&L=deBaNIsD)^E)Xx?lAeyHGu1$(
zkm7Bskav*2ocK=<mB(kKw+2{)G3KT-^xF22xyH}%g64o-@y>XVqHV>(*<C(F8i~O8
za?g=xrd~~=f2;n%wL_k{BU6SJmKp_0Y;kMUr*rs1m3m2~M>%AnNAM34elQgM4(%tM
zERoFVDGb{F;!4gI!)?{k9i{2+Cmeh+OQahSwfc5fA8eDc=_sq%wFM;eKHUFeziO&)
zmDxXvtZXxLV(D8?*krJV4yz9)sy>G_IP?(8!?oH?8aOyO$oJ{>b@qOIx{cI;iBH9D
z<}_Mb4|1sVI<E#_gmU66k<n`RnLHOcc9J?+1L~v*)ikH=M+|bj$4SRr=*UGvu<FhO
z`uGTdJuf%k^2~q+l-?fBssWBmI0a4Xn;B;stF_A!?1WSu)?ZgCK*6c*Gj0-uO1ilS
zv0q~F)F_zC4@#uWomN-*!|LyMXn#S!yNC$8MEiNsdDgA(8}y!UeTG8uoMGo>ng0F9
z0$R8PyGml6Wq0CdW+iR7Ib{1uuzuP&5|v-U5hZXY{I8}t@Av5BR1|0k!7vbD60qeo
zqZbVHdjqljMynj#4&UCc`P6WPvCuRWz&Qa<0wf>x#uRafYcg{;R`4LXWfL6tMTsKe
zkAFTj%N5-wLt~TtzjF=>eARe-y=$|83dF>eyjh<4sgdorSD(z>cS;HpP+34m(6ZbQ
z64;xUR|mip7&Aelc;<EDPXEE=hPM6D!|D4507zJ1a=GuN{uRiKzQ#ZR3b;%SHujJI
zSy$JW+UPv>1{<_LpT;}%9yFUfig76GBmMmFJN66BN&9QaKrtW?&tT~f?Qj&?OX7~3
zhHtJdE)3HLP}33f_9+_JT{#YC)Y^2idk_jL8bIoUp=p`&p4zQyBDS5N{^Q+~t?!MD
z!p$Er_&fb$|4VqR67cpLE|cd%Fet2OA4JYt43(2DytD>k6G7-uQ}T=&^iS(L^b&IQ
zIhA`>2l<fdP8vPMvd}6nTzRUM@A>Ck3wsfRdb;hGy>QkqN#)qsZ3!{QXug-S+?ZNB
zHYF)nW8*^(Xd-W~f|K7)$4PbD$6dR#6K>UMk#ey@mFMk=XtfFWrFWN<OqH!KZU-R?
zO+%KvLBdDpFvYgXN$E6`zg)U}5`nT#2$3P)hxbSr2#P)h&W-8{6okIHM}8k)chcO+
zZ;o;7Zb^@>2+@K{Fj-6pM9C@<eF<NxYo@BL=_`*9CA3Qwx04Qx7W$mg-y-ZAMu^IL
zx03vC!;|332^w7{Y_T%$INdDPK&3bd%LTZCsGCmo`Bb#&A{J2~>4Y+gEX~JWpsXjD
z0I!4LS>7RDch`eGq#?9;{Id?jCQoqfZkn>T%%>-FtE`M5pdS(a-L;vG#)Ox#hkPVK
z<M?hQi+2i?)SFp+G3J1|f2-4b#Aw|2@E}~}k4xfgKut(r#rsiYrfF>I<d`3F6bgO9
z&HL)AgLBcRGIp4rRA26{-*{ItY~O*#{-J3j^6s(MlA8vIwzqALn<VV>@&qS-<~q7F
z;b=%N0I=D*<Z&1)3r%FDb9tbC1KIBy>0ywCvdLOIXP!DDBf6A`W*djL9WX#W49g~0
zX8sUWU>WBW7hsZ#$QwO3Mdur|pa?=8F-M012b(*+JR|S$_Ha5<n08S&*%-?XwiF_B
zKc{2s-PBxdZBA$7X7|`}{mu4pIbl>|W2r^Cm2oabXg^(6$QO(AfI__vb2ABKv@~hu
z_`aA$RXqYi^rfU|5j~98eypv}DmZ(OmZ>{@>54Wf@20?`-$2O#=SEiPg61IWk`3bK
zCbU)tD(?<_rR+H|FevC3E|l`XGEg9b2KV9P<Es*L@*1ARiJFaR`Fkbl;n?QL|Af5r
zFn%_Ku*hgdV$yzMW<MXRpe+pakrc||lm1m9CZ_;^2C}Vxx_u;k5Pe>3>_96%tJDS&
z(2$VBM{8=6{8P<zIe-BLq<tNzJf=k8I_~eW<p=?15(?BpqOtjDj{{I=Pb&Um$8lN(
z#8YueMgUfI(2Arh<Seq5if>=#I7wiYpo-g+u&batShwl^oGrK;EKa~SNcKNgS-px*
z@$N$9RBFA|`FPPMv;rAm1O2-dp|Mm8)DL+@x2IdjgVuPc19}W^*t=hLN1!6B31S_?
zi?P{ck(z{yq}#;m!53}94ekS0Vv7ClHU6|WY!;!VFm(h)FnY=)bZlAZIKTP{yUVZ~
z5MnvFDTJ$;=!2XY=zWlvoHn;ET6VVPZj=<0s@|5JKgo+xgz;^w^<{=WRGE`I)@z4I
zNaMZo;17?0Ldlb$2T^ttYyQ}xVHU~dwHE5dS?j&>2%A{0mw&k-y>0|^FklA+m+di>
zXpQVB34_#dbnNbm$#vp(1B?Qn9V_Xb6(OUB1~~2sk#`n>04*<>AB3Cc5=<kN*9^|e
zO^Zdojh(}g5rMFT&?K?s5}e_4OTe7$PZTue_)SeVgmFO4YzAFX&HI<iqHYF<{^8Se
zSORy6yGlux|9eaxS)~SG@N^ma85?<XuS$!UfX+H%C?KD$v(NT=|2@WeZ-gzExwBQ7
z;BGBMLH5SIs)2t{RO15W)Og?bi-YVv^i#7huFArz2fe_elS^Jc?l7d!T`-6iE-qjh
zNX&oorZiIi!w&jGU&qm}Zz>hsZV|>pp-oT0C1FD1F4J%QBg$ztESF7Jmq=1{>pos0
zhg<$5nfToA6W8a8CB`qP6Bwb}mcssJ$q(8=`UpDg?h>x+merEiFg3vj3k$=b>PBvK
zV&gu}^_LK6#=46MYOAbdF6)XK@hQgCZPwR_XtkHN$5m}^-YCJu17~%dHnus071wC<
zx~ebD#FGL)oPG2!lkv7+c7m|uL6l2swe_@z^u;n_pq|p2>PPn6|Ec|`$!9?*+^nu;
z2tompn%6QX3)b~0V|s97{iw;-o`DNYuWH(;20&{eW%idPYJ0WLN_#a4>x+fm%SThL
zl{a+%yb<w_!rxT{<q0qoMMs6*_03ien(7vHEJSl&;B&I4)-K`X(&19Lv>wO!qo9JS
z`LL)d?;IsfLI)TjO1fY~)K5#G+23>bO~-@4PzyOpC}=dQP7V5jNwXX3wcQRBgWaQZ
zJ7=P&t~EZxp2Q>EAG|tzF7VptCryZ7sKHu`+}r~1wkE7ri>#bYT0VXqgU_mQy2<p{
znjbA9TBrbvF99%qZVVn^21Jm#Qoa1pltP*bgz_s^+8tiqKRjB6##zbgOo1)EqQp+_
z<N_rk4no;wgX^BeI-ZelleV{WY7VE8_mIYwzD%R5?V9@<qpQVVhWz2F`|)@Fb1jk|
zy2ip%E5_DHf%K~q=JjujuW#4>&b~bT(v<?CbR<Wu*|@f(SYx*5jiLa-|8QB6z_qTU
zDIqoHS9W0ZZTKuJ6@!G6Q0x0(=imxG@WzQEI1`5f_<^DNLx!6t1(-pABVIu$qC60k
zH)ExT=hBvkMLIw5*4{qxtzD$RX$CXDb<{t%liVI(l|8su+Qz5r_~o{=)rSrm4RKZ`
zr?whUD3VF0a52Ps>G34%X|<lA+P*ek27$9s;kiCqiH}=qY<f8}y#*|dhpEE9P{R&_
zT%<-nvUuzD_HQ)Mi#<(Rx|EksD0d~vLmwe>J0&<Bjv~e#?*>EDE`;r{q0!<4lxSXF
zFJQBnJ-ar%I2~bwfCQAX6Ni?+Ctm2GJO(4Ml#G7Up8o@w&pv~7<wA;CTd<}H07MTg
ztpPb;<hkz87eT38$`Na3g>_P5@*P95Ult>oM0&R2|NhCi+YM#xNo@f`O5wmux#OOS
z)<YaHXY4XisW*e~L^fM4xAg2<<0)!)wUo5+a_((XTqO*0*J=Q&g-jY^D&Sl&p*4Ev
zr}GG1o|2$mkHndPfoHXDULL4V`DUy#(ebg%==}431ZIu_$^BwYriTj?X!~>}<#u?T
ztNN9el2`B0h%ENkug2x8vn)f|gL9>8WI81P1neXU!zV?vr*C|86CWr;Ivy})>#e5V
zc)EZ+kFPR4a(B?3;<YCXh&9t)L3{eso6Mp&v7tK}XsffhaICZZcD%Kr9ly_&z57=k
zb1FN9ioHU?3irheiak@VA%8}M4JT>`frbmbcnaQQCoCA-t)nC%eOiaY5~(<^q|b>2
zm?9WEkiTFMcuznH(yFTHliwaEqD_W=hN7KWLq6gJidX8i(hLw~7drR=SSJ~*Tqe1l
z&kjJ$npF1e-Iv*-8rm~b>hK1A+GnVKhWNEp?$s06`aCozDLDu`hx2+ZT>!l|oTYZA
znd~Z=_oS7BqWoGy3@DKc%&a|uj-xE0S<_MP%$W=LAVO#9{Jd)FiSf;4o*7!#Q+3=b
zpM3#OxC|DOIumY+?bR4NV6K;d&8@DWU=T8%eP)wa#W?Z>0nU28CwQ4_Z;3m}xMPSm
zD2LF}M&2(=wNL&x96i?CqxRK4Ksncf4aHd;8px%ma)0@oNv^ZMW{)_81{J1f^Q2h0
zAcTbC`+Y21t;2M^(GDzxlZ@FiK%1JpP_$@6JY4WswI{yb_I)`g&D`yLskGw~M7!4#
z1eY}2mr;KN@dc|&gfm`e@t<bW*IW!<a((R}1rKxx73s9{r8#5O7O$Q5N~1yTDT#@6
zvW54f2+OO?5o9#}9SH3BM^^g3q{!CgeYs=TUkOrA*E}rukD@ME1noV;JaEp0zMtQ#
zJl8kW6`#?*Qxi`QThy?C2*kdK0K|(=1CiQGU0)A|W!?6OfJa&bv37P*1r<RBjQ9f9
z;%7OyxUI%fll6($PiFH@Nf?U!^k?@c|7<?Tsr1RI^k=W}_h^%Uy*xK3p7zyCwjf3`
z9mK|e!KbgEiQM$GG5=Q!fO$TAMI&7y^txj=y$|D$d?2GU3vK-HaT_rq$&yM~j^{^)
z$Puev-;<{;$A@p2DYA!btRcF2{fyuI77tR)RlhBxwKov~qj;KNwYv-eU;tiLbxz1g
zRgCvG9J+iYb5-yfQ@AuTuNc4Fi_fnTHYPeOoWZzkBY9om4UHubV-GlJ5C_!lZI((t
zfbn|5>l%6AG+ENLdFpQu66FL<Y2HHYGSevCZKB?p>1m0@6c<U*_EN_R;KH}CWx8<B
z-;VlbOrc%1-jOsxY?Ae~wxY^&5Bqkj*EB4MJw3>X&90RrYH&H^U#V_~Cf4<Nc`PGJ
zz5EAhjt1%!VgAo>7cdRz)^&#BD*ZSqlyb{L@zXPFRViAPo^+yN=V@{l126h?cW60k
zSCdTlr$Y5FcG7g)<50yj6^+L3=meht*boCaH$5HTv?#Sd;{hk}HE<h$7HIa12Jx~v
z-c`{xQch{o=kkblmK(yH|9bd805w6%z88J|M)|O93L%6LCi~zvA4uwW^49|F`t7&h
z_A7MiNQ0CseA}~qkIh|OCC@!E1e-;$d6$QbTrVABO>Et5luZ0`ZW&V-If)KeOB`k{
ziB9QSr$(V^+U!m8;FmZrlFzg%c1S<LH=G~vAnT4xjAUz_z(<k<iloj@x|_wi7Z(!O
zK|ZuLN*_{TcJ2G<=&lh0ePn#3>G=FoU)AVIg{TTx%&;TJM^dzjAO5JM#Hn$|MUfPg
z`T22K<@e<|8FOm;I5us!2Z)!>mW3lnib1Y3c?QXUX#D-Cz`a%*|9E1y_(}-Iy*h-Y
zvf1jJEkL46!Ez)s{4nLsp>jvdC)H!vH$v?a!4TjdDqIej+Qm)`9yD@feQEWTD_0u#
zEm$#y1U7D_j_JX|W|dUcc!V__W-Sfj=RZO?`y>~`&U12VFE>gH+c-^!Wt;3wnHR>Q
zwRZHNUzyF)MtmT6AX8|-uWI-1-A9eaT-KLAl&0Qv*^2_+<*y|LR0?u}s+!*2@x}MX
zy3%Cv)pqO)Zh3C)hP9uc{9Q_{7}j#@*1_ffoD`#&t!*A3>2v1bugXW)^`qJvS&OpV
z^l#!*x+M5D`T8{j!)=d`&&OgCbWgG3@k##JT)hzRoqfFHyuV)<*}cT6bsssgFa)>|
zpuAxBj_9PcO(jzA9WD0X);2+NQ5i1TQ|>42gub{2oWl6u7ooa6SZy}U;bE8@c6^#f
z-kxkicjr+S6-LLN-BA^Vs<Lu?&Pp^1(rTtl&mLTK!MQy2_%nR1Rdalr6<dbREy{sK
zJgn1wx~iJ$l7XdF8-3H@$lK04IfPluD*0|<Mb=lpRV=kz#NhKzawLX_sQk3dkW9+R
z^YYb#if^kcY_~V`XtWc_Ak<W~dGX2FVyYDU4>-MvW@ZGpNVw(kUnqPWF<a^^GApdX
zs5}%Kl+@xCP%L3Luw4k&3BfX(n`I448txl}c=8v{KyXu91UE>CInZl8Rw0Do`RAXn
zd%u4DdL5_$!m&mNHukp-8B2-vQ(PV>;H69WZMB2nRy&A~(inI`HT`<HNlW$-9c~yq
zN-9)#W}8{P-NvG|b_&ay*SJiMZ6o(f@ORSSJCjmYFJHd=(5kK<Bx1wz#+Oti-XsNQ
z9qzdt0_+sX4^_YXZ2Od1g)KL4$G&1+rs>7_NbOp)sVQZ?)<<a1{(}_N+~#9>^48V{
zb<b(3?a_jQ?DTiz;!?l+WkuDHxCqlnm!2MKIX>0Ks_oTUs9o81Tt~}o-BT<G0Wsk!
zabXP#Rsa(D-D>IUeZJWLioykvk{gCb;EK79o#`38GU~Zz+e#_9;F7DD_{OXB>3v+Y
z?Fx&Ex%|?LhzJXzn`s_j7nB(p6#I!iQX?Pz&$2x`>owh^=+QCW@Rs%HV4+jIxpP#g
zD%YgDGY9r{BwEZ0Zm-Vgzm&f(b?v?4+^kK({k9XyX7I{Txe+*{CTsg8pnEe-;{*&B
z3d3tm(+U5*h{F44VTrLLHWuOwPSV+5j9egWRc_7`3ReMF0ddE8QdwIdALxK1De$Pb
zZM;>zQoXF(+593@P^vQZw@{}37TRo^me4!3jgl+9K?K#F*zjHHci(;IcQ_n`g@wsJ
zefmftL?dtbk%Sc=uJEPZbj4kuF9D(s_k7v_I<QzE-}U)JX6~Lc<p3Tr?Pv+NJhyzq
z+6gCrmlAVmtdgoN?2y&huCXpHI27ZxC4>;Bi1@T~6%}`BQP!fuLiu@O=Ps`upK6O&
z6sU0N>O<GW+*wrNXWAks4y(fMLLZq0zE2zWqaeRpF1^p5k9EJj7>S`OE4Nqk?46f0
zV&v$S=8MDOAS7h()<u`gh0SWEZ=b$oFP}rGJTT|V3h6}{Y8Lf#Kz=rPUik$>`iF}7
zYpP^Kh$3zq8kzd#-15}Ia*dO^nyIRmXj1>RG&gHgL{irQ3VL5%^Ditf*u4YViDZj-
zb%^{H#CK2p!j(uKM(jl_yaqKlrRPA$963@b?z1>i)3^}9PyqvlviAg?cI+LxuJ2J*
z^%QLJU~+)J#t`h+3?FhU%K)9;r<$+lwrw$r)BCiMVbBvqB*W6`vbc3!->oRhk;@vJ
zLB^yRTe{qM`9F0bCyQfe-E#upXi|Q#`?ynA$>*MHk9}-oY{*haB}D4iw1%#vlzZ1)
zP!EBAq||+Q3N`?&nLBrGL0ga4mJmXSl+-RWE;}vs;%cvC*0R(Ogb@FVOiaxxEceI9
zN1C|e%rI7LuhxHFUiq5er+oU(^wM*rBKK6}WzKHcS8;CErs$-!WqYJ{S_jdxQ8-^}
ztd>5!NTglC>j$gVvQM%anznDo3|-f;+iivoDOur*oi4(q;4;O&PbP8=SSEcQO*ID)
z$&$RxNfC)@=^y=6dXg&jg;FQ-<;nh!@4GU(r?c8vQnHJEAjphLN_*~<Zno!pcd{~L
zsk1sFsq3Kf+?~shl?AnpA?2$>m<~*T>SvLv32qXCtAT#BLc-EBb47wC>5tCo63*fb
zPD97hR?;zlMfm;xbX8T2WSh`bD66r~OTg`i5DWM;D0dq~N>w1#0-^iw+5m&(MTY5=
zIol?mEjnCdK(B)!k`k!$NGd%7O&Zl~B5g)F$lzR>@a}cLL`~Du0jyT*QAZ}zC(ryg
z^`=WZfq0#xJJE29lzglR^^5dL-ycYc)OH*oZXU5>{kjSLrYFbHNME=$amfQyJ0$hH
zyjxq20pKgmM01xRg_wV|XYq$6r}iA!*KzTX{)U%lPm%AQyvS}tvWxw`!g7s-C=)T^
zDj)t>`gUnf##1$?ZELPSUCQx`bkAuws4d4Lexz4%)uj<6Mwv*8F)?|51*&SoW;322
zD^{&x?fMPG#l~PVnOMDMJ&!;90A{m^@`?%yi;76?T$k^ItGb%qyR)gNs9-?<epszm
z+-^5kiv?Xu?z}t-uU|50UOAgL`RFTJPUKuKeUwRkT1C#*M!MJKc^Q8If7HH@OzeEg
ztYwv3{&7i^8K9_KlhGmSxUx!p!0kgL39%`7{nGx;@8!1u?L=~a@$yA76T{X0(i1a8
z7&cCqk^u<%D9}}v#67%JI3v!$^E}As7ofT-uxp`23qOg9QQtca!&TlJBfC1{uwBq!
z1a7x`i`7~~3Zsv&Cz1eFUU0f0yS#CtiOmchA|Pu20-G&n$u;LV`Q*pF|7jIyl+h<>
zIhp+0!IhZQ16=n@T##KRar*<eDvE%V2yg`v$*6h(0KebA`M?G}GLSWS)|k{AFEarz
zgFJf3q{M^5&qz5g#2+$m!;~pUooMv2f^}Jpii_atwv3gG>T<#$p|q5ackrk!izqei
z*R~$3?O`d(FZFw*#Ol{2MdccQzc8|ES(R4ja9W??K??arbV}EO`2DgZKWi~SSYo$$
z6WUFsc^Plj_7<Z0CNI~;t5;x;akP-9myF6Eo2!WkQRv^pjv_!+6&!Y>&d68ad`I_g
zT^Tw0HBLOCFT3|-^UM<u<MDdQ$lS$`KmEdhQ~CjrmtVjgci%@!auUBSTEf<~%Wyc1
zeoQ74x(;72C}($u4_mn;RqWfdX}<13QS^5k9*am$8!DuTD9cTs0uY+eCBdX{Ygt~_
z>s@+uTLGK_F!)4=IQ2ODsUs&8vNOla30-XD6#GGnop9{9cn%9*y+S%M0+j25l%M!T
z<eQzuyMBSk0T+M`)MHnc>Zm|rVd1)nh#FEC(;w2muKzH*tofuHw-0s{K|&aGY7kym
z>4AbusPY)B&u)geP>8HAP;~Rq3cmX-jPlCH&w~(PxP+@Dd@JDNnsv$ne56BZ&9OC2
zOer!<=jI_5jXx-@hN4Qtq9}k6E5wJ^TU8~nyHt|z6f~8=tV`HkM!;vbSO`n2ds|*!
zz8=8i@f?-*aw(-0LcHGj#!D*&jDaIPWJG3ym;mNaGrpXbdAMiO4j4XS)!KCv`b}#~
ztE6fPKQv_#^VR^k6h&EbD8_3`RAg^0icU<s{rN8nNBRBIom=XC?zJh!=Uy`~?B4!8
zjCI-TlSrTHRIK`tQ;rK68J*Jkywa$!<w;igV_c*;9;hRd+1c5y$iyxmF4R3w4^{Rq
zwRWc(LV1=>q12gOUP)4{+0ZVoaVtLa_#;F`Me@U}*<5h$AZ{FH{B#yC`2(NN$Lu-t
zc>dWZ0hlv)KK)KQkx|b-!=KAnAcVkbv7#snilU(Fuw$!-huxw3T#zD5YV4H$KTC6W
z1^k`Et;$JBF{YOyl2V<~HZeab(lk3Y!4wI+?$cyltaaa18?Q_*T3+RmZ@&9|$>VX6
zrntNk-?XyW&<~)UNVek`L__%>kH;hYe!p0-U_nN=Zr!|qg@mBw-F?a2&UYF_taB9X
zDTCY~BIof#Sv5pEAT0(IL1d`T!`GMc+{b~5^?iU>b*N4I`G<u00@8u|B=iE1aE64h
z1dI^`2-rPxLkVGy26L8Dva?uF8YBSV_Cs+MBu5yQ#$6?ng3A3PfN`fogBT%?$5X27
zx_~3UvQ2ZSneY2uZn|{4M0^UQ9PT*<!7E^n6mmQk*X)c>=O0iE!^)3VIy@DgP?Nn{
zi`F|=cDujZ0mp5(BM_^ylo#TSSjmI28a8<^3Ncc-ZTv&$97MnS1gUxP6l_9DxpK~&
zIpr;%apa>UFJm$U)N1-RIw`HPL#+AnzG+rs!c`PiPL2sPeR=DU2+Ou?&&T$v>aQRE
zTXg?zd%TUjEq=+{^L5XIgB5FtwFRqW1Pb9oVUFfQRg6R$pVx34L`6jc;H<1fQIt9Y
z`1|6e48L(06TbY4g2E!=<6=3re?Ojiel*chk=%dpKLHSmz+y3CE_Eq!Aofek>}TDL
z)OjU%tq!r{er#B*^5GAq<5q2VeSP-@VJDVWYTKidQXlCMXC4Fw`5yN}$=1qp^ZDh4
zTb#gG!Tq-r$#xu55JCtkB^x(x6i%m80IXTFW=M2&^n27=pJGn|OE`F&u<Dr<hz~P}
zVTp?XiO>S4VeND7<+)zsGIpU|4Dxu{S(U8bZe{xKM%+$cV4O}sVRbX`vJMLb3~Ml)
z0B)2J35*a1kG<|(CzqU26*y*kF^x!ML*$ejb9M$<5fx#$?5(O{U2Td`CkGbR=-j#U
z-1PMH58is~t!KaZ;*09S!a^}}<Vf+{bI-w1R@XXx(vNex+;rtBQuxOr#V`)P&Sa*P
ze1Tc{H1ms@nfpV~SAMiILRPs3D++@^IR#`lBw;o)Cj3@eA@Y4I$L19y<z}IX<%%wU
z)f|pxr~LDjqefi*NWyJKtmM?Tuu5M2)PQcP$)4KU{RYERxEZN=kWwBqvOl#5Pf0t@
zuJE2;XQ5vapZvU{@~*yV*68$HF9UkoQdNbncB{h4-R(UcRxu(i(Za?|PnYlx9b(MY
zcAsX}t}n|=A77cDo*$LeW!!JN=aC^TSS4$r)Egr4`vVhcq%Nthu77@P+_V{o!$E9p
z3^RV1#a*}GO3xnM`Elx3{Oh5|m@(^TrhWAVAt4UTCKF(&R-5;uv~QNwSg-rmloe!t
z4-lEsWr#^=TG5`Yxn@;4OI6_d!Qq{@?(sa<ue+V)TdF&RJJb$6QmpEpLLZs=KJrWb
zuPY}1EnRway{fcA`?EAJb40*3@AGWh(T*b=wIp)n$dRJ3uu!<&Zn1jx>f2&sV!jhX
z7*;2#L)9LLI=<;nQX&oOU6&Z>lV~KNG{oLWWG#xT0*Kwyx0G=4=~aXQuLY6FT66>M
zl1($2izWO`8l?2`n@a*eEOALnTrICE&LCry)nr^>ml)%klbea<v?qcyy{@#R5JI(D
zt@l0p=%dr0e){Q<TB~GD>%&dN_8aMwzR%C>G3zF&D3=0j55p9Ez-&;i##+@YbMg=W
z$(%4FV;@PRSo!wifvetHI6+ibXDW)A4&ot@=K#q<5Qgz{b0B4efG$FaOOZUU3$akC
zc5hoXcJb&x-dd2_mKS)eVqF%ad|K;QyRj$s9?T($k2C^jbX`Ac>uWa$si`IXy4wf7
z_K$?;hMweDy*=AQW}c6jaFxEP)=0ZmxdXsv5ez;tM0kCY&Ix9#s+qpN@C@6kGkV+J
z6v9v=_oKc#SNBxxYekv5YP?dS2Nk@KNIV{6EiSL9ATPgwRjb$1AvuZffA|T2>gsBG
z_v*<nKTO4Lv+?=F$ta47qN+#<iUQhxY}se%)ga`vwQGu5$cz7db=;h^?%k#*MVqd3
zdnGa9DkpWd0g{wBlRY-VbkD`7hO%I-i*0+n^z3MTb;R)a_wE=Pbs>r<2_B=JNVemc
zK`n`_wMH&nxbVV|kdXHP6(nr^2)^lRyz&<}-HEk9{Jh-^J>m^3Wm7P#&@<j3WcHw}
zx$JsRk*Z2w_(u`rt}P?MbO1&>v*e*+r96K}abuODoFI>s!(z;*=RVBa)?m(-X7UCW
z+-Q7X-*$&>pHORbR8@8GuwlbKj*pL5YORuBkoo<;z{3mqoYXTX&iE-NEM<TY3=@*Y
zw7qpT8_BDR5PdQy|9El6q#4uq)rGzC)n6l4ja~SURbv;g6jJ_%5Vv3$uJ;E*B2z&=
zZ4su8E8kl9_PYNriQn(5_zPqBh*fKg%0;W!;5J)A&u#5Kn?g;ZO{tV}<)IkwP#`R^
zTYPj<>c5<MI|^3s^n7Yk1vj1>VZVNGMD(~HN;sj5jjl-+H7-K61C&>4M2D$t$aIq&
zW9Fo;_VfQXD7<T!L$#BdY5{QOZZE8o@|9V7%eSYFQT|X7#`7q6Kd7oIo*H-eZwnXm
zkK1qMv;n8^{@6EIvu*>GRaJaG@hd+3_%m!)E2j=Ph0xGYq|}YRq{O5OyunGK&0vG{
zu|;ZgOY^d3)gH5%mGEAjtm4j#Bd?5)GKng;PI|7Fv_vx*HDtSI2a7d2OeLq-N0nRW
z;p<{zIwzP}w6Ur{@t8i2Nk~02I=S=k+CJNfWGh2L2C)68eM$DYzW0%YS^~3i<3>?f
zSSUu19vv1H74@MIH35EqJP3>LhRV#ncjOMrin5uYOKj7&wv-46w;ANETWp|aShdf{
z2Cso4;J8c6VN@=hRR!jKSP0A#23EqT94r@CG!Cq;^5|6i>W?bIn1{n+%qyh{ZXqzN
zkinK#kALqIa;C$gJ7M{gb%#tQ)79U6^UV#lR>_egM~b5?TrvPFKK#&^{^bwfWE?l(
z9HEPoLA(NNq4h{*Ht-3^i?LR9%b5JblRLkhxqV+<*el*%(0A3?h3^QPeJ6;IK%PwN
ztl<tJ#QlD=zIN5yi-+xJ_|;z;Ycu6&zs3QkQZ`nV?tk}3W>m|dgPm7P&^;CSO45;1
zE}TFAz+!NZM3_}y_p|zjyqnmm%hw*S{^#l)#-*KiQV5ZuDxKoZB*mD>F7o5`Nvsxy
z(n<|ywa)Y<72JG&1UquPm{pXxs*t)xGQ!T7g}P^FP7Ad~6pT<MUDAj|ZrHe)&6~Hf
z^p9nfl$J4a^sDH)q{>x|#bU;0vvSux574t`ceZTZ&b2q*N^x-sH{Wm_KA#U=(~wf4
zDgu8C%Qb)x$#<IDFLWMmxW-s4SoRe8=n!kB!lmQaB<s>Wq$XI%DfTmWRTXFSHZsBE
zqfA^gFbq`)JU+?V9qvWC#s5gZ<Ltlo>|h^S+h==>WGmyfA+o8GitT7))LJ9KNCFu&
zXwW~YO`^#71~$F3Zyn`3jqK}&F(?r>NC=0#pqR=A8QR#)5M?)FBx_ONG!kzbBGCu`
zD1w_6*l2@vD-@cbs)i4Q=n&_Jt{&*)2K!zTf%1bAXQCW|1ToSMSwZI{g&QQZAvE2X
z>%L(9>yR)II$aKc!{HdwrAwD7g@uK_jT<)t$eIv?qn1(D2z##u)}DLrxg&NZho)(|
zJV(H(2#yC*fq2-f{7V^#T_86i#Gg8Hk)>4Muzl97{YrS*G;)*09Z`B2$a@rF8Vs1)
z&{&#?l#^GDU36ytJI*IAzW?I(<eU~Te8j4?>nHY`o)pzOv67vV`}Q<hg3E8UKHLQx
zkh%{Y=C}1&hZ34k&p)G=oy$)T8@6Vrdsu#{!SNTacOe8=OoAnwT%6g*&L^|W=on{W
zcY%+LTpt#bAiva)%cEmb1)Aor+oWY=WVoY~QXlwAum0g7HMCi`i^SL-xIH!Nc7uj}
z`}86^eG@_m6h&dh;`s<6m^tfbA|k@szJ3K#N`w%cd(Oase#<K=&~zPL>gc+}ybpW5
z*}8`U^tVg%)7LdTZZRoJQnZP0f3M(zej&_Q>ZIBuxu9PN-z};jK2l{*p^xZrm6&jq
z`D?1U@~m)n75KRDl(1_~=wiL5OOl0GzbeYE?X#Uow&R#Vtu?Z|yj-zbtv47)C9EA4
z*iaQ<GfUh)(J0u&<A=<Wx@4I;q(ne@b-h@(5RF~Ou_1v(ve*Sl5r$PW7!Dl<xQy3<
z!0m^Us=(DGHYBi(NipVG%jq1Jreh&9OLRP1W6PGR7;~@oHPF|Tu;#@&A}NIE`rdo*
z4Y>H?i<id5#mO2nf7Fu5V9cLC-%q{8{$nA8=<~O~nf+lNtJ^7K6_YPaQ$!emq6xn=
ziwdh>DfC*Mm6@pvyz)7DK-Ff}j#-+btNI->Qr-o0YVm~DgM#G21R0*T?!Uj>bi$)U
zs#-qfuZ^`CjEW888nfx(t;6=H<n#Ag)8!!Gsfmx2ns!Vj+Ef(o*||MCYUgpCtX(hY
z7s9u{8=L=r-EGWWZTys%Z>{FRt7FK_^HEu?Q&g_8WRr`hZj5E-(n{{RB8uGNN)#nf
zsJA3Hb5>+h+SkA9o*M=!frTonkgU+X_>h>)0--crHda;Z_nx&;Kqiw3m#doSs3@9?
z3=%@%_iOn58d8EU62=u&5rN=R)kb*GPcK_){HnV%&00vKLf>W;-J}ZqI;e`EuuQ}4
z(<!gg&@{=gK@t44tdgV{KW0_XG2Tp=T_H2i%aHydb!8>DWqazX0ksp!b{zE(LWt*|
zf4<Hd`N9h?gb5)G6>V28l&o*q7<wgw-7L_hkr<Ox5m;ZkxXQ3FCWqG_Q-on%+?-dl
z?Udlt17iy#?BFoh3wldwXjFEMZT*YRE|bUx+tONhc0>3z9a$r~Ed-=RH!j(gU9M5$
z3bd?QO@`$%Cd9aImk&1O)f3X<mEbM}TO0tPp`oX{-R?gM3kwBx>N>|JYA()O^V&x<
z?mk1T9{c;*pxh%Bu0%GOmvx9Bukf23lYe^Wr@ysZCijjSUCI5=;BJC5m@J`fuOBHT
z?!2`~Df8#gpTDW)6OJB8DP@<Q-9H(8Qpo61r$%A9pK6azh)p3e+Mqj{24$5R7Yzty
z(!6s1+*(a!s7hprN_MdivnmYhrNvq2$F2>L3Qg?x!1ucQ++M<*5HIS<+ZURPu4mfl
zoM7h6STi%1I_n-2Ex~TL1rW@pU{X!!x`yBHN0;#UO|g_$>C9VG#cwOTT;3p~+z(99
zT)J24ca;}ztP-NXS;`f4>I{k^`~Xp*3V&>NadI~svzApdX<j+^UlonjEXd6FGO(`$
zmq({xcf-wHP^Pgy!^3IE*_pS-wY4lS>$kc-TfWS8wBzVNZKn2#C!SdEc61jKU?8)b
zWL;v6A`Cb7@ks`8YPhgxml={c4H057q=9PDl;lpZ#jnH#m7dmRR!(>0Ozo5=ERChr
z_2VXl8Dz5|@|_wv<qfyU#{Byx1-kbg=0KOZ)WKa}WlK?%B#*}<f-<<<v9B13vU2RA
zzpWg*XcegQCEP&M$^C~Gg3A-Mus1jgv+)l#HtrEG`Zcj%8Qj5N>abzMRMliXyA`L^
zB>b*Id=+~D<{pc?z7n_hgDDFt%3>o_wr6|E%r`hxRH#CgTPG?^W!@UYI(fwzVeBsO
z(WkSObs28dnl(?;(Q+IP4InZx?fU+w^eE_ma*qN<`p@@Dao1;hRY0_4ID)KvA3L(W
zI9)n6i{ZX5Dk{R~+q=S+K(yFq^VV%hDe?RLC<2ymbu(>I1)1AD^c8{jz?phAd!+WZ
zP(?}aKHavTeMV(@baL8DwV!+a(g%i>b5oXwKB-n*9-RR_?bxjf*+o9K?Di5Fs!-w5
ziH$HJC2Zd1!KA?I9c~t_t9n(+dT&d+MY0`7JsJv-19-jO`n%Ge2tuiouc6v6aCs#T
zOJI^phe#vFFukZg;U=pDR3V88sc)MSVUR<gW{8r<gao>{qZ~#hnaT!6oQx7dpTxkp
z%L9cjC7C6XhR>leYhb)4fTru@Rv2w8YGa~VSP8pI>q(@|3|(UylT84ZSI6zEzeu5i
zJ)vO>pWp94ddbGfyB)Zz`XKc*z$jlNlrnUgL8WgK?@g)Pdj|)ORqrqDMx=k7K%BDz
z0&D;PAOJ~3K~#s4N!?aM9T0>N_pE+v(c+W;J#cc%t^c~X@#&T8)=%s=Z9iir1MV69
z_y_tc7Dt<x%k|{11F56y90O$$8>1{Mt28<}^}q3v>ZsutL^5x66}>uH>62<>d$yO9
zSQFD1J2|<VjkH7yeN(M;iZ@eOu5oHl8=H4|NYC{VS?=#8gjnDIlpgOrd3|g&z{``1
z-e0+TTdYj%{O<R9_5If<#=3ix>Yyw`r#wS4BfEl*9V|?r@go-vxq!dlbuU5)ygn~T
zu-dF3!Rz&6ci367^iRfrI)P1_w<3g*=~n9+H>oD1l!S+e5gs1S(m$7zXc8>_qmtW;
z?AWjd{E9Y6A9Hoj>M}b|C8JD#>c-fv4y(eKbIYHKO-}uy*eWs-brHRGryG+BX^9q+
zV@#M-7|_#>rb|AbQ%-!OiXsFSlRy!$Y)dt1iDn!&h4*KcZZEL~h`zTI$#xtYsFnQj
zcs$~hPd>>XJ9ex~2;sn<05K;+QH{cdl(4-}&^@j(X$rAG??fX;vNqN+qu97jV`}0L
zRRuam8Y1964R#bl_c#J{c{)YayTHLHv$(rdr*rhddjNKog4^GK)Dc*^5kS)=+Y1Fg
z&6s<~s6Y~1ROu(Pr2gKI34zYhM(k)XTwa}Rg@MQ7P6KOn-CMWYozd{)8iDd7ixCB5
z<lPS3mBWzS2n+=50e*ieK_Ce;yLs~FB}gWtt6%Z@gjS0gTs?NlwV?kG#QsY695P6x
zf?({*w-zoq<=<!L9`5-Z$=I+<ycicQuQI6z_FT0`rk&q%tAmF-cQt_2b^kGyXj6-*
zq_mNp6D&_y%z{#<Mtqcs@v};?m;^pea$&zvhW3~Iw5*bQuZX5&yg`r>qD(B<62hsy
zhx6#rj;5JE{=D*=snhS)b?V8zFf#4SLd}oOt@NChDom&L6NZN?QYb_M!UGnw;G3z_
zxoOySTz1J&>~<SMC@8{+pHzf`Pz34enar8<Gc$jf#j{U6j;bn{Os1NuI6nNE;m!u|
z`~5uf#IuaJISQ>pVhgfq7fGEjbXS?Csl#N(4tK}2E&(n4Qua9|xyy{Z@4Bs<NmW@r
z?OPH;(=n-nq-ZlCHiab{U7X&_PJXGM^_gz6i~PJdqlCB!6Ma%`lvQe+(aYX_?&_*d
zVTs*NcjoOVfOaC;j$;r51`MzPZj92^Mw!2&22yEW4L0W+1S>u?@MKjnVj4p#j8cda
z$S5&5Mzql#F(HQAyV7F_eQ(Y;^d@WLC2L*a`3*wo3Mv&G7M-}z11FP(m6F1sOznj5
zKtk;CYiui2aM#EJg<0$84N&6J$t*D@Duh9b6T)Z$l6<Gmo-zeJNTwvdX<-7l+bwG2
zB=`5y4jH}~rn1s>7l{7=$t|=N7%t}X5uA^zy~vZpM^hT|5g&edP!(ZTja_^n$T1up
zYjB)U#Ph)2hkGJNGHR@n2b{sJm~CyZA1QU*`RkBUZe6fo!LBwR@9<!LcA-Dn+fDv`
z$SI-b+U;R>v7Z@BoXlBSiNmJQG0seBrN&8JZTS3>UpGhb`RAYU`NXg2)2k<g&puQ6
zcG~nYHj6Ngo>ZtSiYQOd_Wi5IrR#}F>2}>j&9yvU4Nd6Kpp2}~CpqmnJJDf<WUu;w
zLbJeJW7+fhB_5wd*TLhHOkGsL^S8t^dqo*DCOk`rI8)7@zjhfl;x$bw*enX8Zi~ij
z5>$Eg086AB$%k5{UM00#$_vtW091PQr(T*|)IB;(O_Lx5%17hgej}=Pub#a4%76Lj
zgK>;~Zydj_n#OHIjlFDoo{wIgEbPjc{P2g9E6xa`Z>p_si@jx+hpF40^QGj&%2-=j
z?f6H-BG!&}>=&-R_S*jwLKr5qtZA_2{l<-SX}GRSU6M{wqLH9@Vz}Y%O)oOa2&5Mq
zBr~EWsJUaLK_J1%YECbb9G6gkYj=o*a<^fPtVLF-f=Tfc<=FrGuEZtDDh(ubc5`Ew
z_0A$uM<N9VS*}G{wMIs<f<)92=7=WBepUN4vP!_|X*vRY<6Dq)E_A#Mfbj6}$Im<O
zyorDQ`DdL!|A>?TdSZA7gn0sRpbzXMkna-hyq>3Syp@+HXC6o=E5<ImRe__0L>iFX
zy7HYRubp!LnL7{nRE}h9%4SqtI9Hq0y-Gs?AZ7gqZRoB44M`bvZ<|ZU`O0^rxrzYh
zH;i#RmQb3Tu>e>QncV5a%4&Vu4TB>(6jx~Em-zU<%c5AD?&gbM%eeJ|$hvAb3w9*&
z#e`3Y2oGn<chh;~!GDppYc~?%ijP!77p(FAyDUHBeJSfhBa5?l?2buDz2-wvHTO|9
z)D}^bauQMwW4<lMViEZKl4_5{?~{aC6|6cqJvzDsMFYQ->@M`-^5`tzSdGIXuqpzt
zl=yTII-pBT2uy%qgHS0rw}DGr2R_wYWTQV-mY4NI-S?24xmpO(Eg~`X@te<!cx7Xz
zC*zx`Ga_Gl;aQ<53Omv>c<F`zF#6RHHm>}F)C4n0(I$RgUdeGCtvq>QEPD$5tlVBr
z{~mVM>~w1%&nkT=XII9%!E5``vye4|{i&pJ#68&B(JH9bj|7N|i&N}&`>lWx7rFfL
zy{FDqxCI;YATi8HG->Dpj<myZ39z-$xPx~VL9YaZz=c{NHWZ3#!slE*LBq(!T5<5r
zdG+<Iq=cQtCVDD5VYYp@Ojfudqd4$7RK@7O;e4f(xEsVO#)TTzQGjZnPI{4HT{NqP
z1=SHmo;6*PQ=wDfY;1*Wh}!S9<E0ooW|8{ROD_!?IB;N{l*kcxt37`6NhtCMs410s
zFp%PG5KDOC#w&Pr@{0XAb=}xSrwJIx(YN;6kovvABZqq;M>4K|V%6FMh?OL=@5QGo
zW@~86Ee{U0Wf3))+ty<pZj|NhS`nL^I>c%g%dQz19=bhS^3ji_+&(mtC$5jBs9ZzS
zCBLP2X5xhZ5gB1@%AbAW5&m}b?I?=E4cA>AUsY8_?4A|33!(1>W;Gn6I6reqbYj{a
zAIPeS|2BmxHUzh-4%|*jwUpR}KnJV{gexe!Sv6#X7S43h8-<&daFm*@c<t)hXb~2T
z4^GisY}CEu%k#5d4{jcwm^Q#_F<rT<sxw0_jP0{-!v=1>^Byi5ay|pkH1;jeKKUqb
zzU^Sz{Bc~^FQl$q;=DCgWajy}^}+~t=6d<x&t)TWveKIrWIoy!vqnfsO#DA7L=z^b
zCfX(^>p%sNju3z7x;`a0H}|)OZBpanl6+?K?j}y3mz|R{q=o*c1D6FI-kzP4)2rt5
zpMd~*2MM!ta&j(h(1(Y;3mw<WkCXrz5>`klzt79d`@K1$oRXZp4d@bZP>6Ba+1dB)
z?Ro%GQc~6-W$$J_-jJP>b5cuvv?Xe!f9f)}Cr+H$Q3zpNYv~54*iqkD5k@SaMK!9<
z6jvG6L@ABLnPL}os!6)3g~I~9;$eG{VO`8BF~}uAY7~T8j6R|rjqiH9*{F)Mqo|$)
zN(noP1bq^9EX~z0YygknAg~RIoe)N1j4g-=3L)qbXT*2f%?8n}MMg1b(y+XCk851n
zro^SOt5o3A>VJE7v*E@Mu^98nDy_ecvdvJw6(WuUAUr&L5HQE-bRKz0H&5Kq3q_p`
zqBTnZCLxrccyjo8yfSfp%gtfI`wK$DF?}MAgpBM1fRNm<<DH+L?$#~`v{!7&P+o`+
z*RN64eOE=Xghn6AvWOnP$MgH481HbSC?|7sQl~Cs&N)70#2LLEm`#FrzAxdytD`9@
z*VtU%hl!t$BP67L&A;xNE4lXSE73GRpH7&>AIp|=)AiS#?DPA7OiJrHA<v}$Q*Phx
z0ps}-mE39fZYdx8P^)_5UNr<W0viI`KG?WI5L~1*_6qImB05udvru|xAefw-oNNc`
zq(q{VyFPX2tv8K2@%TPUS$R3#yLGh=IsaTtrpC1I;Ro*J!w?5QPJ4r4gTk4&ri%1j
zFLzuNg%B|7&&rJ4%ubDnYAyb1@;?ejP1=4)d9NbGf)2^a%Q___bggZWSy-40tivF4
z4PS%O($bpn0~+YxzrPt65`3P7Z-d4U_*^vp8w8IV{4!z0{&fJ(7VwZF#De7H<h3a&
zDJM5MIivV|a6c{nR+E#H`y=J?dzs6L9XfP4tVA-HJIBPtbfxy%g3BR>+YHg}aGN3b
z(>Dov#MN8vynci9Z7($9C~ILe8}+L@L>iXJT8J9?piWVN#~UKjZit%)L_O<*^hR~>
zn?_~{_#1ejgfRMxXoib~T8;U{g$9lpW>x4AX(SUhB#~af#<oI@okc3X8d364t6{P1
zn^a#3CoUAigRaq;S@&Oy#Zq?<9Z?cMPY#Pfs8ca2(X=*VK~ClAzsI!P6vF)Ghk)ib
zWk&;&OZ}nkmAhKN-$txhU+fg$?aw)iYHf4<NU8hr6l_3Bx&HUxe=ltLjAH{X>3i#w
z*`*t++&cYx*g30@gB9D{Y_IOe=O2#`Ad&TuJoxBSL_~!1{@DL=;rZvXWy?0+cxCjh
z(bf7ZwXLEPJM}Ut>Oyp)wn^WM<F%>-t(Upq_(gZ~v+kZJkiS^X%3POL{~L-(P8)jH
zZNo?B7ZfO7pN~OjpG9F|F~e`ao6@qfCi}bl&fB>B<`GPs=On_RaQo0m1TbT9g=bZ|
z`yMIhH+yM+H6qyv5(a3hx-2m<vEE`Q#FQqQTHJ1Tv%ALJ+}y!5VQElxy{W!hA5tEV
zlz(*S(4o~b|AdtHv^cD;-`iT_9SDLW8x%#U{~MTtl1SYC#%UlCc7t4XiK$PPDXWGJ
zxrW8Eu47@4styN`U>og#4ndjKPPal?_1<HYN?nFqyCFJ81}4w!D>_CQ^Ka-1FRs$q
zkS8dwHZI&^GU6tCB{Ytml(KOha&u^dkEQ*{lmN243LMvl19U_8H@j1>#p-t!#tA7O
zYQuRP5||X_yu&?*qXC<98KvnhatZ-(^+VGmRg0s|H{{;pZRp-Aq!7ncHJg@DzI%7c
z#$DQZFHI_1vUHORO_x+S&f$OK-p6Wf+5qs@yW=?d#1j~H?N#WyWX|0AAOxokIEAy%
zIO9$s4Dm)yQ9ImtZaBR<Tc1HHD>h00YvZ*ld~KX>_>1o5d)>WI(dDx#vF88&^U^3G
z6<L=&s7lVe?2@6PxVVJe+<cM}6Z!k_8~D%DkMq#u&!Fi|KYE7WIE>+Uk78Y>pKW`*
zyfwAB==-G=mzC#a%x?088vNynWDTNKRh?A(M$@z=$)BPq!EWn@xT=X`S-bP{_Q~zl
zO0Y;>pOzXIchI$36B83d5aPz5!^JeFVmGIxq#OXVX&p!@>nva%k4Ly%uKJyJ%*miO
zAVhAT;o{YsxXEKe;P~VkVRB*ApH44^?DE!G5fZ~1Ydz*x?7i>2pu#Cpj^;9~_aD2<
zbkd7eG+jZd5h6c6scDj`OCxjL)8M&iazJ$bvrpIcLciZH+-~=gLnfcOr61^b9)gi>
z;ql>T?X7{-mD{M*s62`xkpm9*9F7KD`{>Gb#T9buzO=O^b~{T^&21j;P#Yhq`~~78
zgW=5EQLtrm#NdyADx0FmUBdfg-^AR6ix*(l&vVGl&Eu}yZ=u@l=8n7XWAHg=bKn2n
z#k3hevNJt>dsJfT=%}Q$zoTh-W}c6GE{_)RQOY=}F?oaZO&zCI;b|i>`Gf9ehVI^^
z3tqB2lri^S5pByU_Jc0`5s7IxMkS_>*M&aiwKw0cyymLQx#Xe?xaa<d*|K#zy?gcK
zic2rzmDk5K+yAv!UBTTCz0R9cOV_Q*)OwfYWwod`dMp#kFA?H%g!o*-1OXEzEWn5_
zYz!$+OG!x?2#}YTmkw-hqRpj^j@5s5MH9zb#TqGLlu(qDYhGsxDgO>kqiLCyFrV4{
z!oiNID9TL)Sy%;p1S|<^>Huy!*zpb)wXu0KX3W?G7^Xx^By=4f*s!<?Hs(QTb>nZZ
zR+zkRV!i7)r(96wZJiaNV^pBKxzeNJ^*6mW^!Nis&1<E4S_73HNnV9<??>99Pm+=S
z9W0nGG2$(Qt+N2#Zws(Uy4~)TCX?yN%COP(CqW#FIB*lHe6rb3N1`D0TMofI4h6a`
z7`fo6kF7rx$jCLlsA;q)6J-gHKO|P(8vRIDNlD4lLowcw1}Rtgx_9e(+?%hCLQ$Hg
zJ{vb};mfbT;iVTwpzAshKJq00`oFvA+9j16ZoO;TsMlT_u%RUPtS-ss_y?|v{wzAo
z{ANtJiq9{(^}@&`O#G3WU_N6jzGug3m9nxG$)pZ^rMdV?ch5{sH0Oj`lo?l@6@Hr8
zB#4hx3AKyI9=Ixc(#hRy$z?fND}P_MDE0AYo_}&@dM5w-;BB6M?gg@T?dF;*FURBc
z^21Nf8yPRVcqmt1dPy&#w%|3Z!7)oDHC?|uJ3ITf?Ck8@a&mJ1z9%Q=mYkfNb2VM>
zh!B4^b&ino+WI%*n<kocPE1Tp3qF{foZJf|W-$O#h_4Ug5*lLno;`ck)V!AN$<3XR
zos)BoghAAjra;`>y?gg&g7)?e0vfrs{kr}?g!rUUV<``tdpp>*{mLt^xHL_x&!lZi
z;lk}Rgv2-GLV;5f%;jumBX+V!ykP}2wMI!Az-EH5AeV7f)BBnL5(!3}VsL`Yfx@f`
zq8-pP-blQ$1@-IKB?V5&#$3tHqCiVzEo5Ldxa{ogg{?EWR>BiECL<dqzO@}B7x2`L
zsZD*fENoFXgy?n1=F)}`5SQBd=ugr-6u9)!<r_*W*-NaXKnNielXY-g&qeoE;Vsz?
zuz1Ca6;0H3{c9j4IVq_*p_iiKVn&P{&AV^DhRI~&xlylh!Fhujc;@MZhK7>Xxl=zO
zl?NiLc_}@|*RykinXAqUJN~#%R+ewACN0sz;1e9y2d{}q>Yi-P&yw1mZ}=;H`EtK1
zXEq5BlJQy<3#IpyD7zSS%LS2x&N;zhKIeD`OE$Xb(ZS03Cx`UtlWJvsrZ*xwG3`c0
zGe2gx+0KoPi6%Naig(An#=jqXn&OfY{`1shd^Ke%+jcas=GCcV2Q$(fw=eTMW{G5T
z$ji$s@cDdKAw;8{qJ;jn|JR#|ZB$jY(Zfjyab=^1Qs(65=B_%(>FyOdIXMdv;@?et
zWUsEQJo7;NO-xMe3j`+a2=QlLUS7J-=bHiqXKkL`F(KiE10An5(YtqVMNyPK0BF^a
z5ya&q?S}O&(q5a1ELfXuxUqw4Vn^5viJjIWa8RGvQwxv3=`E`_Xz>YcqIBVb!depx
z86HGxoo-2bkz{SQAgfg1_KUjjD>%7l=OobUNpO68eBaiaU@JgTFXGT<N86Cf#Z6Cz
z(>VqdG5{n*X1b1zYdI9i%;iPDMsrL;TzvnOSjB8Rv90IfDcX$md(d_Ln5<^Auh_D6
z`xdXa@#fv@^>W|8ALr#4pCc+NlK*`=fr#)hhFxn!tVKsfvSh*R&bzj4xMjtXc@N!n
z`>kjE@Mjh4Gu+rMf)l&iShlsA4sm7_0au<G_P)&`oOwdK{7tR0e2YBr6r2*NGsa)3
zufsb+NNvnz1H)&AI}~$bl!+ZVUOL8`v6=;2cYDY!_7fi$f9Bg`UY)e;_g|k_zhd#w
z^UoawKyp$NFO3|*y$?Q$-|uJ4>o4)xQ~$wP*|-9%uInsVxH#9D6SGgcv41%t2~bc_
zkcX7>n)s+=?SFD}b2kE+jao?Al$E0~krd+FL!n!llauo~4O0Fapx=S^sj6yINjCoj
zC@3hXM2H!|A8JjXb4X#eTIC~;JQ6L0hy|cB3;YcXwic6NflH1s{yibNKIX8v3O45p
zHswhw+->)FYs!7x6oP*``!1Kd|EQ?e*_fxZr9e>ZG6qTrgA<bjzb|tGRVtqos`k{q
z4G9V9)^-y-Ae?i^M-<}hrVcBTK8JcL?Km=U@gu7?Rk&ocb;_f=^dBS?Q;Xu|2a6iF
zHoC6=($-@g$;izs`0uN4yxr*dV^2S){W^ca)M+z*kU#%2mkpb?@W_LK>UgTEl981~
zT54xrdub#Oj~Gw6mdv(2Ug9E5%v|bZPoWPXVAT${`MR^izf^U_TrKo7-q$L({;KbP
z)Z|v_=XI^pxl8zmNll+T@c6LVx=wtFQ={6e^TEtgIwzQ^a_eN|`Pkqbz;APY6vJ;E
zMvv}YiH(U4=&#ptJ-GMoyLj-?r+DSRWBiMjEdBg}ho5i<Q<c}pyu;dc8=jPM#XeQ=
z`b!eY8mw;OSYW0>)9&p-4c*3RX=#y2d0L}Zt>@O>E2NY@q+HwRICMUc{<?SXZbi!A
zcrLfk=liO`n2p@q0*3ER9;vM$rKC@vK6Q6NSXfxSOWzcfIO5lh`q;IwsD@Nd-y|d1
zMgUb_!PWvnVP)Hp$>7ZA=G^Aypl3@{?&a1%fm3H|p@Pe+AgD<?N;1BehI8<1MgmeD
z1eod#enN<Y3Z-auNO|Zjq`H25lgEWTpnLI%guoqU3Ogn%qO=t<i`3E0WpJz3kV6n3
zDe>g3MM_yDn|N0Il~9_OK4Z+g?>+U%lOy!aTeneET*Aw*k72^3uO2Jfv-8^jzV+^e
zFD8G@>o2{~C~OK~>7UEF;<}sIu``{4XP(Z&c|Xzh^t<`+XAh}~W~77=n?iPxpXe~v
za@iRnQ#2_ObdPe{G_Csk@mdvC`#>jk;3wV9yISQorxa)U{bF&y9*)_UoEDlqV~LZr
zBnzcZjZ=@avnk8NH_Jl#{NwR_JnlVWWB0B`HfZ3P3_ANPKAkZ6hvMw?+osQ)b<e|(
zKcnx-&Y`fdh^PNEO1|>O*b$|<=@a(#d;QB2NeOvP91i&9$flEJtS*;pQ0*J9*Lz9u
zc|cK4PR^kVXjYWsCXR(4NLN)=Rl|Vb)ZnQF1qF5X@7&znIT#t-4H4>dyKgv<F<S#+
zVPO*Bk3aq>p*Eo^q$7ki2x--ILnG1`<Y}^+4Edkli3X8U6#^>6v=*dnEawyK>aIm_
z$;R^Py;tZtP`LdHs$yJspQIWYq#)0wSA*?^G9cSMyc^i-#R0rtZ+`1f?x4_89qrbK
z5OLgnx4H3grOhjr95P5s?78WON-FP|z>xb^Y;ji0uY!+Di4Z1BtI4%B2d%mkZ+Rxb
zZ}aCjuLby5K}mMTE1!Hm@w_t!58c(h?<u_Y<~x3*yeTrd^C^A%95>{Hv2S2DH;&$@
zsBkjooe#L?%FF4JmI{E`Wa8iV-NU4>XLIHSx3FQiU~Re^yH)W0QYTlP6=v_(-8Nlu
z=wo_zw4St9df$CTb9%Pz#noLRb>7x0nJL|qQsT_H3DM&7xCr&fyDpCII&V!CiUNMW
zWb-Z$`BhOIclN`~{Qf5{8gf2ayLR*CSKlyw<}6&U=0jm`y8b#Ie&C)fqY_i^5Rx-K
z|8jEm2?GYw^Mq4<<HrB*{?eTEQTy|q{$+_I;BKiwZf@?%CJCpmH*{||lKqkJ-61W<
zsOybm!F2tAB7B?DjucY<ui^WedZS-9K3N=AO-X>a-+tSrY1%sDoq*mC1~#bj8c8-q
zm9o*s&|xu%RKE_bmm%E{MOA?$)~rhGmZtB)5KEv%(5o34)UAz9QO5OliD^t;rG&yt
z$;Mp6Qri%H9}ak1T3Y(2$z+l?o9#%++&&x@QiYJxV?BOwSMTSC?UTE!9rf5<tdG+4
zK*bQn7T(2VY5R~6PyTw)HFRCHDC6>1fDrx7LNNBy)520l-yW|$d~LM2eQ5N|PRSiV
znfS>^N#Ws5ORT;0`kQp>kiu=Z+*p_OJ!kF$rhfkeF)`7+_@Ad)wrC*{J@25z6NAgG
zlbPq^#4fgobB+(W{a;r`bygG+Uo5mU-_|NOexp|hNCZh>vF_z%-C4Q=-*AEa_Pmor
z2HkQ&M89sy=6+VQAg9=ml!C8UCh_!$SD80w9(Ub-3(m?)o_T&0FTXy9rGGBx>E~W#
zz`*lay=Gmr-^0_7KO(NU?BWkj>SmpL=S5N0*Z&c(-E~Q%5*==OJvzBdlVLi|;jc_2
zTYTV2DW#C|+eRM<ahdUtP?7SY;Exq44~bj50UBq7D#|{VPN_%7jvc!r#2G<{v$ON^
z76i8yit=gjF$n{bl9CQv<&Ihs<8ry`zyIF<f|Lsb8~Sy~C=mvstZG_UvbCY<kW(Ib
z%wdtZyb_OJ;?*QRP2%-Ss(q45FWAk26Wdc>U$?t0p-Y2kZ^)HomWW1K?Hw<Jp8u$C
zuIrmGzx?twweK5-8ngq1l*de#tPo=Lkyl9l$j7<&tle7eVPeB$_Jre)M>L~D(~77m
ziztQAo5~CL3u2`%rTleMmV3+MjV>iFQq^bv>HPhMVb`9}A*E?Y#<~p~`S{Z>xct(K
z1EhoI%v-=)?|i_9O@>-vbaWICKlm?JF8!VH<3DE?=doz3mGVlB%sd|^RSYf9$y!}%
zqyK#Ad1bWU>02v(<VcONS|yWo*G#wKJ)tBg^Ghkkg_<tOFY~iA$H!+2Be~@2e=_^0
zpP2UTciix|YcQKld^7cXZn*VM9(mwjjC<!z-W~HAD;Lk_^i%tD{VjJ;RNUBM37{x~
zM<2T1vSz1yK(Ef0*&DMw%CarhWmRr{mV#ciKi}xFzI#4sNFk;O2(C$;mXwr~nv|5(
zk;ZP|vh2LPLsrK$IXSt%5TenqUr4!6*;WmqY1;oaE)p;HBqt_5nv&Fn#<>vH1ZBo5
zhHTLN`#bhQKmsBzE>1d~PGK^c<Y%9K_W6DH-S?mnq65MN=S_tLH^A;61JA4#DBn|N
zxOQVh8e8RB3t6R&GqKC770Rpkc2?k@!|R7#C6KmH75Vlb)jl({abW-e7ko)XK~zam
zrIC2#)0=kfcfJ}1O$1SIEiElAeYvEhM30DwkhLKO?Wl)D?%DV1xkom1Sai7OaI~Vx
z=^yQ|nQl@9HULL*(-m!78}7U{NGUU-qoeoxnQ$zk#-sMA5aO|&`Kdz|ZE$^7RV_Ql
z$Hg|;Ue|RVeeyrt{`cX8hK6w8Lyz;sBM(wiQp&{1Q<(GPRF3acZ$DYJW*up%ofvfX
zSqwV+EOcFG<K`_a{c}0jDC;A3q-X5T$jsbP;wjxzjK1h2t?FC>=&%LIR)y{DlTSOR
z)8Mnu>>ZlclT%JUi6>q?mE@$vMsq7EEhR2ChV>gZlAe)CbhJ_Pz;3tk!8>nq!6jGm
z?g!&}{@JEwZhLg=ilUlRU!GhPT2$sA?NEI$=VoW`_r|xKNVb67+}y><Nl7I@K>CxS
zD3?ho8)-@+#PpgUYMUb^CB-3Nd{duEA(rh=UqT2|a#GT5jXRR3Aw=V3qUJlgwR`vO
zk8ZCiYJVv3+;h*#kt0V^SXe0CZnu2xwbv?#4I6fELPEkcK*byi7km#J--1>D-`<^v
zMOE%`0RO&c=FG5)h@)&O;SvgoVrni4u9=!k?t2KiW~sShg=l74wtKH-R`;sc)Vdmq
zWs>Fv4b8o9R}>evVP9vMbIyDJI1CKKfG~ohzMtpeQ4i;wc^P5m%e(yE2tY6^cVG~K
zlwA44P9^W|;%ry3zM?fG3f#)7SkT*A=>_r%?&i)FSAbhXXQ0R`l84O;iz~k&?+66E
z3$&XHkh0&%i;IhY>eHvsF5|GaMx!wX$%h;$Y4Jk{ZxWm;mU<x^iMLeF^pQiL<BQqw
zsI{IA{wpCVveUOJrCfK{#<k9?{y@l2fsl=TbyOAM)Bj!?=@5`k=@jA8p-4B<DJmk3
z)TQK-B7M0CNav+X!Hbl1H%NE4pme?a{r-OEy#Ky?_K!Vtc4nV_&a*qu%zS1h>j4nq
zeK>z9znIMUkYju}(CK$ss`y4xe-@`Ueokv5Z{xPh-LTjE&j30(=WA+OQP+-mYO7yg
zJ+oFvR%o!aQ*{KMFV?4p;_F)}X?PgyR0jtuO0Ufg{gWn9D^?CvfYj8~fY@g|T}1XO
zj0qPPH|xdFkWMM@PzE&?q~%_a>iSHX?d}G9051SL&x$cN+fqjOi71VWKmZ-?$H>r{
z*WHW{huNeZg%@LuU)skv`k&IW>!LFgO-R$<)DY(uX0GQPyw5?Z0X)v0N;{<Lr))cz
zG{2n}e%8*mx7$LwWQcx^F_!PBs;b)Y(^n!RO84+m!rt&c1+5es3xe}vv9(>vm>18S
z936}Fit`~52=2F<noSH2m;L9EizfW&M!fI%5VV>vBSF*#V_!?+dOb4I#J#6vr^G(8
zFXzZ#B)j%%Yh1*sxlu|x{9*fvFPt?GP#MU-kV;)WdL1k^lRrk<TO5Q!VX_HAnq80t
z3AtR1x7&Gfg?4Y>s-2ykA*#jQFHbg3vyBk!GNHJd-6t6jKTo(}NNDNW^S3Iie!tIU
zob}7<GI-1xSC&f|rY1#nfy3x|=hIaqmwU196lHAj!(~aJ5B*_lsG_HdDe&^tJ6_`~
zXYu2*cxX!l&PqR3JJefmd>Z@U)<0)_hpg^h;7&buSeTE)CC|I@ahoG+D_`lu3w-G%
zZ_+OyLj&u1u-*ejJbDKDk;YU`|4e7hr8?Pn3nd)$=4oHTdyvSQ)2JB<3FB{vPxuH4
zy1N~WY<AjUJ^`(I`Go}#fJ)E@Q=%Sr@qWCeFTkk@(@zogV;+^KsofpS;ew(Fa}Sse
z2vrF>RWT%Cky5ZPni*}*p18T}*VfXiIAt=%D0BDrwrYc9BWHu-3yXG>0mVBL@}SZf
z_jx{o(qR3_jf@X{6_{T!s>EeptY<janKtJ-r<|&(0n1xk=SScyys1Z{{gv<%jH>7S
zFl59b4DE1&Jvi?YhHw7|6`j}da${;|jfSo#BR^vFEN2FI0qg`nJnp0|OR6Y$n!zbE
zVpv^;e8W`GO`3eGg?us&<rYRRF_7f_Z|wk58|7aSLcKhK0c{}X5;37Ef6}8gQLph!
zfoQsv%RmhS8v<2i(ZxvUYbx)=2-5;S6JovEiT9jb@D!t!aQ)#&R8@KrjvC+bs<3GA
z8tRInBRJUKXSa2EeVfXpDhPD!nwI=#+2Usy?0ahN<)LSZtB{5}mpRL!x*CrA^V1*U
z39qr<^@*aO;R($wd|p;hOW84~+?bBJxd(cPX`d|FO?;)i;HXSx`s8+~AO!c!MZ#_V
z-M`%#hDsdm7eBUsA~TaS6W~duQ>!79yraK#iIWlRHoC^px6QQCR>U_rW@$BQKEcdE
zu2|r@vDweJS84Rb&5-i!sQKe9?7t0ShouNYN%ITstKCm=Q^Y*UB|ZOjBS&UF&DJvT
zJd6w1(((a|`(JEYrICHSW4$`XZ~H6rbER)`@7-)Z{{Ft>@yc!zUgj}vI5CWDP<ms^
zF~gE%$xF7tsduQiHI~VHZQknm-;X5{Ce_}aluW<#7bXhUyw%JmMBoS9($G-9k#pm(
zqCUJ}sb+6c_HKeyp7BYex6N)8eRp|r@8H71f*C1u&%V%0t|SPUJXum>*S)YH6AhI1
zGqC5Up-^^dQr?H=#i~+eO56pZq~v5|a6~=ApmIuIfwZWl<vz2gr{{v0LDIj%fX6S{
z^7Rt^_6>xz3Pa@vA6@rIMG{Zt<;{5{f}V8nhhUii3bn!Fg+e(5mYb21%24(^N!w&k
zLQ?ikrUu%kd`vWOiB8ss90tID>%%wO0j18GY^0u(q2X^4CZ?<d>0x!ABsIO#$%)a?
zgsZEoF~lNR1z(IS61z3uFUvRzuNJNOdgO$8<=>Wa9h`L)UWYOqrMQ$3y|VeS>XuEl
zjr(eiu@a~fI&M7EQ4N)S%7`3OSB|BJ(a8LZLn5pYVq}WVF)cm?<J5Fw%@TW)@xnGy
zn`EQ;;m*bv4ezFY;|h}S@-Ch%&c==AnHh<C*<(pb&AAh}Yh_48*<nh`=jdb`4Ie_?
zb_q9!Gr~3pr4x|pdXl*>@Zys)Vzd};b2Nv4^^V(BIdzlxyg9)FP8z#yp5fSZ$2E-J
zJ+k_K3ZpGItK|zwzY7Q%sq|*13U3NwFl)vxpxl!+iQ^zx@8eKji?uxr=13-|px9DZ
zf__QzWNWC)l%irUh3oP7pc{^`g>w2R@T<|~t09U(6v%i5@K0CajvXOEoT9)R9wT@7
zkQE}2$Uk@xYv!~<@eLTL^if(6IzyHjS7QMAp`lnn5HsHE=_(uip`oE?B#hD*h!=UV
zQV#+pKxZ0bacv_m*MQmW$4vsc{+PNk0hIAiD!zv)_8-Hfb1d3A%<v2>Edk65e<GDy
z17a;30yK7C)H_o`T*t4*(lL6bMdLeS&P&a=YgJWM8nveU+(~NhU0ek1MQORYxKv7*
zp3QRj>`>#Ghjm(es*vd36x3+*;NQo<X{ktm^|=H^e3B1+o{Ym!;=AfV`a3+lefYIi
z#_0yY;`!HwK1XwJA6|r|n~V)OS~?IRqc<}Ep^5~D{&Lh%`pi|F_LKz77rh3mV0sDK
zh`}X)6I#Ig22JPN#E36We(dm%1uBUYr1RVTXA=GBDMpYq=kaBxyA4tS^V)U+wWmIO
zeRSBAA|1#FEUy;urica-%3kd=ff#p#9oKX8sG#j`>9^660&%cl=kpd>xT}-t+1DoD
z4XN!De>sh;CnTWaN|CnVYs@-n8}l|)No$_FDbV9GPESDSsZ*TwS23Gdoc!^6si0?P
zXJ@oSr^-OC(1Q=89X{p=FdAoZQvB4mFv#C=1H~?RIVsD*u@N2-(Y=)YeZGQ`(x-EH
z*hLle$#%wQ?%}auA@$zVKK7KXM+tCpMqFKAA03~Z%u+vkB+=8^`6gDcc;Dm@r%rR@
z*VYsLMGT<>5@rsn5k^drA>fME(7^o<MkX!W((rd7NkEq}E;6_1B^M>FqRKO%E|io5
zfa{sEtLjy-tL^|IuuLWpH=6Sv&*L5?J@xm*=vPd}iCFy894`nhF7S0Ls89y}cmdLU
z;yZ&Jn`moma{)12Ta{Y_sd*bP%T-wB?DVvXmX;PGFOThMxXRjJt8X}y2``tHx)6~q
z!3Jf2O_hn)LJpEDK0C?%*vlBt>T<?nAz;$+dr#8283b#}+?M(3U^Wq+dK_@IpI&aj
zg9$;VB^%8>*PS=C6aP5yrqDd@{q`)_$FQZ9ZnH(!?^kqu@u%x02s=lCJZgO9B2xe2
zV8q7prW?f^y1ZqT=XPp3(;vk38T#SS#bak4U+!QAx7IW!zW7AGHz=^_xT{UbTpJDl
zvZ%fCZr)9UB=rG+_Sq3p4B=dB>r|h<$vW`yoTUWhrv!%EhK7i7T7yW8D&>31ri49L
zp?sEqaC|KBy~T|*GN8l^Qs28j<%Cq9Z{6ahN}RWW&N{m{92^rSPg9j;RotbS?1g9q
zQkb+gK|i*X>Kmc`N#3O7{)MPD9)>9oFRwceHa51RF)vs1#U?|Z4-YyO6-d)H<G8u`
zleM@;BJ@l*bl6#*2RF8O`MSHGSg4gQl4U05;|H4zN9*ONU;x_M+DTksTAOrNamk^{
z;T^rW9%bi)T6_?HVE`~1$BPoXrG=}O8>$2Hl~WV|I^bID^!$8@S5?oz!^5LN%xPAm
zuDbe><~j`Z#z*<r8usP0<_2k$@D_P|GV{GfB{tz@kTKu6WC?9gLld+XhgOaNS{I6K
zfEo{AASbU8m6V+Q<z`;cs*YH{jJ6xjT-e#%JY!&aUU5C1T_DZ31mlnQBT|vWUbn>l
zjDW97!+b1r5G^=qT|pU~%iBbv{~-JGx-kjPA+6TX^rWQPH#0SFuFv*o^atGNpPng#
zdSZ8dogL=qr!GIU`qh_$RXw2=U<m&CF@r)r-%pEkW%9MBoA3+r>{qoBzMp|Dd{5(`
zD<O*&;SV74z@qAVd}6UtjXpUA(73#RczW{t$h*P>dQ|&ckc2LHV1Tc~3jK+jqC4qK
zYjb_QZL!(WVUgFU621m)6!l?#p^ev7>Gv^Ik!3i>X@50%C_{2}d4*>Cv~zp=USnrV
zz@q+h=0Zd6+y{$4z0~&|`Ny`_4i1=#cwN!-v9{gX)>q$)KgG_5*+1EQbq^0okzvCM
z!6lZTUcE%S9*ew)&PZqQ`xp0eMu6CABtKL#fa&q+VQrSpp}uNL2Uv=@=ko7Xq+uxs
z3<i6S8YgSA)b@~M;8Z0iQ%V5jmzo1!?)@$&0GAMBv0<qdACY5iwBFbFuU~)W2V9?B
zOT^|q2n)stDiE?KOz0AGo~h}S5fXBZQ7fR0{aIYcj2G<6CntB0N=ZrS85tR&w50<%
z3L#^oqtzZZF}<!{xBe_9lE+d&<-~EGWBaUAVu^my^#Drw3eNmlHa)}py&A7tzW3Ip
z44(k&s}xS6<G=b6Z$QV8%beHFz|MH%i&&ZmwIyW+an*byaIVYqkdH36j1A}e?b=d5
zRtF1?eibA<`NA(S_t5b+yprs?TlnI#PHCGrab=U|+%7tc{4;0{jg};dw;46Z;oB;|
z8X|->9f7+yGN`rC&@Bv0Q#6L<GpkmXM8xlM)78G(y^Ghk;cb|sxZ*UwDtnv!>R*I)
zmIqoI$G>K@#gZ?8>ycuy;$aArB!T_TEvaK(jGL7kp6l==G1%1cG=&$jtD-=tTl0T<
zn#UFuaW&aPF@d$LLM(;Fy}4;wdEz8tVPOaup!){P{1{2<E_Z-nSN&;eZcYzI=BYni
z{8@SSVRK-V_E|YB$e?u6Aa7KqkoVi(f@>W|f8kGg@>obwY5BHuo8jogesR>x^WZ00
zOeXT&m!#RJ=o{KzW@O`po8>t=Vu3uo*aZ*v#oHMn+trlXy=SW&yDegW^G<qW5@TjY
zV!Da*0p`^yJW(GrbjTNzzp3euWyjpb+H@9P>Z#=2)e|ad!Q`xlOjk@eX1MiyQF68O
zU<0sjPt5YovfJCL_b;dS(M&{V|ChCOrikT<Y{&zGC?SstER9Sk6<Y)mWVn>{v_H_l
z0cQ4gzIx5-VbuD!n~Bf~H99NmVda2866=&}CABVkR#sL#z~CVlq@HtxhKECyRY5_a
z<j)^hCK8f3I|H7VW|>=N+(~_JLrIN*q0soTU?hh=p~!RHMh?}nPEda&f*m?80>D`l
z=RZ9xp4{a*FBBU(`cCN!(%;G1kL6o}+GGKo>hLx6m(cjaP*SNxGaWU1#sX0bM7~ro
z^Zdx(sSy~MTeG#b9Ue>*7sTY+BP%Zp=(>uJ(-6$BFG#$F>@Ww-9@-CKv0xtG{ib-@
z7@K~zzb%C&N-w`K6;xbr_MDYXYUs%A#6~>SX15{ykYAEJ(%tk&g+&+PqZ`GwR?O@6
z@peoY>=JrCwWOjs{>u8kx$4J{XZNjaLZE2@IslA)Z$H2>t9VxV|Lrm00q9g8g1wgc
zP|u@EEau^F`%heHKr^5t{ZUR;weB;p;O}rEs+!kbaN+zf+?KnzIBjC&K&3b&Omwm}
znEmf_A!2XLH!?`Tn%e;5p&v%fxiBLppvDcR&Bpi5&82~uON!jw+<&R5sReR6g)3x$
zqoX6c`pUrfjOOO%iRrq&S-Zuk_p^m}x7U9`3QtI@JXo@@R9iy>iN~@ri`A1e`fsUl
zr8>WS!UK<vD~-V9$d85JL^Owva=oRf3BhZ>jMg=7>U`Ck^`6;IR!%ZR&mT*t!7u9$
zAk|Uqb$OsAQGPpLac*H2Ym>>|F|0kw{4JY=#M#*IV+(wt!RvT;w467SXdBS%J&Tg4
zuZMa-sTO{E30OGYE@horcZ{$sLy3c#m%A7Ivd4FX-*@Ohk(6R6<m%g1{tgB?_;jAz
ztNjGDK4|YG<mO+na2{MuMw}@nTPq-7xhc46*=5P(Ky(ff*B-v8r30ZWeYfs-e6bE!
zk?=$+j&ioYd?`=5$*bZ3)ABD*1n5PGlDZr;=4>44m~mg&&cz&v+}U$WW4ISRby&)~
zeO2vat`0i?Nj(}WuSwnM5v0+zbApR5iCTf=b78LQysuypCAi@j#n~R%`}<;^Ul)r7
z-mi!ZP=t^w(!P^6b5GBuUO`t_4yNILw_oVn;;!5z9!&FV8=QO?V?o<GItr;169Zp2
zmn%ob2n29R=cDYFo!uWeKVDR9i%FZI7PN>2w**oO621oc{u-7ilfJd>lmv(Yihd8>
z*y9vf5mq0GFx~JIqLj^OmO+p0S(=f+ixpg2GBVxr%dp7<rE;w_=fy@<vswfS|0}L?
zcA4FT{~enSlacLG#KTsxk)5p|L?^&UJy@1b#k{9ueuFK$Xp(1I{xb5KeaKKLR0rea
zF`ntf?|@S7RGk|92Rl)*AF?WqT(k)PLRmt&%%$hhdL=FbkCwa0DiS(WiHfl3)S%co
z2o?Fy&=K=U1R=J92q%kfX;Z4<cugP)Edhcsq|c8hCQCeAJ)6(ku+!f%=NSEzz@X%P
z8f+UJZS(;~AFgtuKGQvQSXJauHVtmPDsTW~*Q{(P0w3Yaa?E{4)mKYN`zYAGmQ!>M
zTi6h$f>3xc4w-z2sb!8fH&)PN3S!~-Z2`I@kL-p0Y|}-;Oq&!=eDKM(U7~0R*OV<j
zk|ZTbm?VzlygX*HX{@`B(x-GhQ}X>(p2H~evCf!Rd<W{6TZ~%lTZLS4X6~W2_T9>!
z3Vye&j+qjl@60&WPghhZ@W1EOp05I!cs0x{sq*N@vl5GM+NTKHC#a~{A8j&5^R_??
zPzKpMo?Q;R>k$Lf6}yXz>-|18A#CkJg!&Gk#-Rzf0nakMID%CXLXg-lbovM!<$tfD
zU;gh6ZNcpSr4iab1V#UsiKeuQx%WRDgzd&**sA}J*2Tl8^8W=H+9)9YBb+9@`oaHM
a<PDSNm$7e|PI{66c&I9AJ+Dx-4*nnRE!XY<

diff --git a/docs_old/assets/images/use_philosophy.png b/docs_old/assets/images/use_philosophy.png
deleted file mode 100644
index 23922d4e9476dcd0d3f457828807271ff58b1b01..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 16463
zcmeHvXH=8V_U=m&P(VbQf{02JsY>q%0s;ckOXwX!@4bkEN>M;Sq<1MvAdt`rih_VN
zsR<<@ozQ#d#-B&e`G33j%UyS^d-yP{@aCP_v-h6;?3vjQubyfsQ(j`c1ONb~%Hv1c
z06-WH00dXbNbrAgiLu$k|04F1SJ5TI{{)lSyu^Pd@X}V62cSJSmH^-upz`RUuD{vp
zgrB#`Zsz_btd!jd&QrJlnyHo{mZ^v>x8|ujXGhd?8yYgSbb8W+>|RGl)&+Ta&s*vC
zCAq25R)HSl4Ib-F21anba%ey^mb)!hi8W{~)-G5%5bNcc@}kZ!bo;2g>F7I*<02^`
z*{yKkS0h(;^|6y)b_!L@KOcWJw}|DMiT^(5rnv<m;g=R>?)-iMAjCf=`tRfaQ?=Z`
z7~UF1sfMlfu%>JLqdtHzf&9xo9bq-#hWIr6p|(NLE6qfXn1i-gn)RxQvx;XP;xq)r
zSLA?Sjhvuff?a9r9To1xFEvjSb?jh9B4BX!1FZ`!O9RElH~EuU93pP^JyU;)6v%k<
z*PQMg1XwtmN}vosC@WpiUPHY=pPIC#Kfl<wR9*ehK)nd{dZ$~Q>hCbO2qV}I6;#z%
z!(hh3=Z3`g;Cw6MXv7I!GKcg$4Yrj!qlfF?@@rjkP@7`ehKHPARZ+8eMBx7AKrQs1
z40b>(C4GSUx8Qi`^Qa%I20E;6p?hnap5p*9=aU_1i42xB_RveM8~5jX5%csLyk#n?
z^`SFc;u2bHC#^ns#kd~53AuAla|YT_{i6yX!tkQQ`ylu3*WzJbULhm+VQZJ%7MaH9
z=hgjr|Lxr0i6>1Ota?^mp=<YmcxFivhy<5?T=S==qHdqTe?}!Bw&JKzA(|2pHSnVY
zT!fbg%X^YtC&NendOX<|eL13>vL_46U-i9L35Rco-}tu%%&2bxfrT7*|E{SkoCJgx
z894z=f23{U#;+OJAo1K^b^K(L17c}}+#>&2$zL%XSWNzq^S58Q3}Uiyk7x5P|1v&6
z03*8sG-=bkBl=^hmq`GDfJR><<)8UZ(Gd{l7&txi{MXtB2tBUe0@^NqsCoF01L2#-
zJ95B|s`}H*e`LF*K}g&p_HbL|&xFDa@B^EbT<nbaBik31aNruJ!z9U{fj3?Q1by7X
z55@n;79n==3NS<5J@$9t|1G6JM1UgpR}?DLOKCro_KtdeGd5nWz|<4F5qE};dKb1H
z4|q~8>9b33l@^BRn}!@}uy;k|3891UnwBCm7Iq%*s`$KdD(QwL*CTi5)cDj=-hHN3
zk$lUc?G@kUr{Ur6^?Q<yynV}rQyaPo*NyJ9CgKY2u<K!pidB?KIUY*ib8;#yFExJP
z1h%m@g*W=D+L)>bbG!TZs@BbXzIq)4k-rDlVdG|1c`$%$)eT72)X+O9P*;UGOT4=c
z<L^%JIUf6h2p<Y5EeQ6k({&Wr0qd}f3o0i5z?}x`F_LtF3pomvejrTJ$OYM5Gy_e#
z+S70Uj-xH;3&!fecg?*HSXGjZg;S#U_l_U>2p$!U%M+@41`7kjNFD8`o~p=^gV{~(
zY}(<kNJVax3q&f=Hi2T%ggw(!hn-(kF)``G-RxdYMNaVCXa^(OL&!AiKuGga*Rq0P
zd3=3_F;^2**QtK^v9VcH!JOS=ZBNf{HQPf&D!Uz*O883O4z`vgN(+J#)vQ3xh9R!d
z-zXrbM(aRF&}V!an!cJuK^YRV?P)SJvzB+Yq)*_W;YY!H1c`AO)(y(__Is?*(ug`a
z?&qk_N2N~g(z$&WpI9Ss4&sOo>#hAmP>{Vz8N^UXQzG${LvL>G$J&I6F_CG<rqyrL
zSaJ6KB+7Y#!rG3gES&5mCZ|y+%$96j5;yjj4+VV~U%B1)85=g~{-oa9nbid$DV&)9
zq`~wJt!`&nBen#sfPXcqCiCrgja4>bZur{#2c2Z?&~ightg|LrY{4^e(tlhw!j9gd
zpP;Wk<Q$_>01Ac9DqTA}ndB5EzIjiIOO`9e4tE*piY;E3;hB&uR6Abz_Bk6B%&7{V
zo2tAQus8?p9%go)XjYWvdH0=9JTZ~R=wLe)hJ2(jm2kBqNF;l0%C7lU-HqR?(Dj0`
zH*msF8OAzhtNO;oZ=zZK_DwOPJK)z{b66o_k@n5wbaERTti%$!!?eatXT9m%Fs>Y(
z$%6_gw)Udce7!Ekr9!05))dp=iY=OKTz;H(;5(?=VBnQr7;HlPuO)bg#-S2Fq0sYu
z(?p8pVXg;h96gx8#rG%nmHf+43*6D^s!zM^<Lqzsh3Q96)Ax11t=sv&tmhuB4^?yJ
zOF*7(rw_rL?Q>@vlN1cxt4clZ|F*~Y#p+SuK+<I=Jp-%1Rh3@6+u3VuXv{%Q-Zfjp
zzLpbdej4U_?`=mUu0Y{<-tlzX{QTe*qfsrp(b?(5qA`Kmni!X7XlcR1(g*Qwe_gsI
zj0BSt+k%fVL3%>6+HcRFQq55$kps;hW{yiNO+eI=)S7CwFZoyUQdjNwQP&;Xcz%gd
z9J|x&b~+orvzmsU)#j&8)l7YfZbF&3o`T<eV~816%p$)j&Vg={5ULwgE10Wdw^H3_
zPL0=e>f(t}RPOv1@H?74V(+^{lrwNTCSye6^9o8=&V`03Uz(NG98?^qbe=$+hlb=*
ztT9h~d0o_5W9wFUl1;V6x<-Or!@@zT>{aGwy5`U#4L+<Mn+vw5h)Im2-h}qIM^%wm
ze*^q|n6gjUWEBFJR<MRQj=e=T242u@IojI9D3}Z9>E3KC*FEx%>*l*2blIRo(A>7m
z9tUSVI-s-mE`C#1TA(hPsYnx<zxA}TAS{FLdf5YOcOSl2|BAZyG!6*_8}Ss`QloW{
z>fTp8KqlqtBb5><uj(h0<u7@F$xzEmDgS`FS;T;}m$f@Bok5qNxm_oy>khelanU=K
zZoT?B6W8td?47W|&LrE!BacKI-q`M4Yq?JL-^k~ng%AB?rAfdw0^=4g;+Dk|F{a<M
zbGTq^0~C<I=bLhHqv%oo2Rw4ve@*!t+ZB$8>e=*^-l%^Bm8&M?eE`5ID7so61zEex
z{9XjSM!!s4CP4oz-u2=Q`e&{kHPYFXzxvKWkmEh<o4$KL<UYQIoc6aY{L{Zf2O+Nr
zU|e{b3VrqfZd$bSZ75{=!?c(?{LwINY8Q%5wDs%GRTQdDPH7L4l$H6xUb6PI-N~kV
z&E~LMvLA7FC~LZz$;s|0_qxE(F>B7aH2&QyH)Tk5kgy0&P(%~rPWRizjhp3SE{7j>
z$Q+hMwYb*b@^-aT(nSQo8|5!A5%$UUZx|RdppU{nUHIk(;?~557}oo#rtYD+w362!
zIYMqo>9bQcjq6?N%4<kL-SQDfXz#SjcdR6`a*6jN?hdEusviCm70eUWJ`@LQMR+{;
ztrCb~jv}dCnE~$7nW=B4caIyt5p`6H4&zxoeb!rS3WsvzDM$x<e~f-j)=iAVR05fR
zOiQ?s58pBt*M7=%z8PBlvL*~%8azxFyoIDmM3-b1*eTKzf;AiuB=vdf+>rLiHTowN
z)A_8w;Kg4#=PRjP7}<#nQu6^EUf`D1ur@WJ$gb_N{O$xQRDy2JJ_lRe-Vhz3eQc-n
zVFx!4mGmFUW#oc=Ou>c?R^p$dg6Rok+$6Y`($W$qTTT7khaw)7+~ZI6U7x>t*64-R
zx%x|=5&C?6_Xu-_Dt$=LB_vfyia1o-Gk6?fmIC79-UhCwnVK37i`;j`*~h@aTmB_5
zbbUzrQL~ZXM!o@kM|h9k4=z!?El(Xb@sd|thr!z3aQ*nq@&sn(fw@<j+aDlfH6>d{
ze(p;?h1vrMr5{@l*my-15)+dYn#T<cIEOg73+g}0Yf}}g?tOQz4*?xUM!oqb@txw0
z(5C@3YmkmaW3fC5mK{FY35oYy1_szz5zRP6UXgxhX(@X?Qb}Z(y$f8-W(NsX51}j<
z-iRkxvvH`oR8p><P#(+H@dQToY)(%`TrW3hr)aRVNa6a{(T|Mxh$CS=889VJwh5n}
zanB$|S7q20SFjT(Ypw?wboBF9SB)!JcVoqazT@up@TrF5??lO+*w_@<8EN39mwx1>
z)=QgReO#0a8RdkdL{vmZUhCpyv%6T;yMzls$vfZ5Qhd{VG_+oKCC$lceQ=g}$2Vl$
z;}uwNR8X&s7@P+$c9rs*Z<W}na6L(=CUU$ZyigmwKpQRkd$qQV5#qC}zy{;H%zv`O
zE0XxMbj7*uUk*ujYf1&5E^?N=kpEXk0uWZx;M2vP6z+d~=vOY6MT^gE;l_7q|J%ub
zb@E?7`M<PwMDn;fiwnNJ$GN_Hm{j~YU{fXS_~cx5`s|e2w5%jaVsD9CXFWjff!COD
zV8`$aa+|sD`vp+s@NUrF3R>Z2z%2C=22s|y)yciJWS<=agZAwDh<ySPs;Jy?Tq@Bv
z)nk>(uIU{9pq7LwiW?jneB=!a?C8!4=qM@C)-ejuh)GUNnQshp@!1vUZLbU64m~~Y
zXhj7ZZf$MNq8GA~1H2ZNLVDxEpg4B=bxk1lg%j<_lR@D9hYxpZCOykc{7!w~UJHBy
zTOG2GAAZ`bU^}|Je0QYh8h6PO*Vf?}*k-1b=tr|c%D6w)e`%CmIi+{(_O{GKUf3d8
zl4XR*Kn%~p?h5uHhX%`C+rs3np<`TODV$@aOb2GEQx|C2z-MYRy3<u)TC4#-h#M&=
zn0jBf-Lz#zZW5vtpYl+rsACdQW;>7(mXX8~dtDs;z~P7PQotv7qtgy$Pxk>O*p<vA
z&(W%ry)m~`FtkqJ*BQ$fb~3=y8k&-wM7Ui<O@0MnEsE)WwMiEKiM@9nt-}{OrJRW7
zGq~eIAm!E<&;x7RTlJcoGcZf(l{vn%5-tAlwg|S+0fvz|ProuLtc)y|zMI|DlmcAi
zb$Yh|D#LyOt-qaQRaV(dV12Ov!k$J1@MV(H<-i8i6xiBz*7wRLQ=T8(Td`Mq7h9~U
zsj88ZJHVfuU|_~OL`Mfd*fp4KX`}4+Iegv}kD{y4DZaclpYGQ2J&i@hH1pUj_xS|}
z_6wXqV0j+ReW0jDCG~i%&VcW{XAU0Jdoy6`mFzdm(;3>ehX94ay=E9I&^98O;l~O@
z<<p*l$-;9XKM+29@z(`sOUt0|P9=z!Hrsd8LJyuER5U)gjXw;ySyuFjBAkeX;;s48
zDC*=gv&}py%Qui4id|TE`rMce2wYJ+Q1>7UpDbe_wAdcLNKoO}JI!En3w1EZ&!zx!
zFc02TOxqpeX(JR_A6d^Zj|2RB%5p3TuJQsHq8$tvnsD83oX@tnbFQQ~D-i)6sn=!)
z>VLi-i@6x?k$?F(jx5|#=f$o(p~b+HfWv^TZ>>~aQ=cMjzrKdeEfr^{z~S%;lK?F%
zvg!li+O?0)Ih&J1W!_P%x*fQFwM$fMsgUjY$Df*|du~ll<}0ZMrc<D6EHj-mc9Y-H
zM`eBK6M;2l`1R8+J=l<{BtVxexfE3A$l2r8l9{_wIulmMdJfdMRFp^jt_wZxNf$Gc
z*jeLE6JIUZIq(pNTQ@j^s^v&b{Iv|L!d>Xs!!I(Z81O2)ra}`Zkr1!UJEtbcLeE!6
zdZ~M2kq&JDvb!fG`+Vj0bwRsWXKb^=S<VEiJ>4C=obpl#_%QQDjkI%i{c<8ryJp96
znECXRo7V#@S(b(c4KOxE9h>>Ka4=5GsLmqwT}HXX7IY^%^Kzz_O@}P9GtX#@U65P7
zO=F2uLD?rYW1Oc13Qfo-)#!{$Cbbz@CX(PrdT`roxJJd+e}wJa?0YD&JX$wcO<U|7
zy`TNor)#gHIrBoC-4EZTeht_9uxAB}U{s)6ruW*gMpWF=QgAd9D$T|{zy9t-1O4Fa
z=h=@Y&1PC=?&P_nbtF`B+`{UKSv%5mrfG%^<%Q%qb4o7<RhlblF0-DJ`uU!jel2kl
zZ(M&~Zd!yL=vHP@FI~!^amA0*_+C*=8J{Q4hr(5whvCmZuw8R(irSvs3{@}KZ&dew
zzi|pG#MW+Vdp63=OP`l#NO{&i+~j?3;(q8i_iQ~_v$U(rcDjteE0;GtW2C^|w&R`T
zdNTA#S!~?8w-jc(B#F7iB2b20IAr*y=r_)rIGe;S;W0bdLG!?n)7*P`#2k)UP%+ZV
zrCph12ho2|ojKl9{dj3r$|C1U)3FS%4hSuF%x$#))jrPdRt;{@sA%Gx$Z@1DX2rD1
zIHhs-XvuT5_Uo&f?!E=vKtq<otwAoMolZ4R2V6>p8B0&h%KE35>!u|$XM9NoHkos@
zZ5}`S9;WU8^y;Y^im#5Pu)kpkqh`%aW}HXuJfmGfGjFNx#WG6mT3z?qML6AqFVCdG
zeIA~c8>95%Ke)W#EnG2a)+JfOCPNjX9z&1xmKOL&Yczd%&GUM<cI`7)Ge?@z&sO_T
zTp>q~p=N?y>_)HOFeiO(3bNTB;htUOYtU@X?gme^_)P{Bn|RF8dFCxX*H{VCMmkS0
zMqT4qp^ea|16HHpb0mZ;a&+v4p4#mq-Z#abuGsCekczk@vb#0kOD-(L4i}#7-&Bb1
zsAY0XK}smf3~H5iJ#g-ST4(0A__T|0l*pm4fJ<WWfKf#SX6}1y<F1UaE~yX8$o_TA
z&I1ACcqma%(#4u_4m4&JGiaG@dD!6AQIeM@7|N1=F<iv(U<4JYE8)9**K7W}PYUJ7
z(sIw;(>FqQiAh>?3iaT&-M6oMSn_jvUtqZqtu~uKqTf;3`t^1f{U`?E+3!ke5Irj?
zdO7tnV+P}pRzFw3{tG%vsI7TsXrx4!pKHAwX70u1rwCymxbxa&BZ;7($c;s+a2GGk
zV<VM9o0=|%jg%liS_j~Je;9Ut2UUZv72&eR)khYe>mL*4To6yMPRLB>l0FhT%fdJ>
zD^?*p;Np^BI%j3c!f!V-YE5pNk%xCkv<`$8?A;Lgelhfu@?6OSBj~ipx`fMDs!Qa=
zEpCj&JCgS;#)+@{4RrTllMvY<<q={#6!RjCQ~V!1QoUIod_N?pnEJX<_?W1WIl?jk
zixKXDn=l{mfOR>MY$HmUx2$jXpEVP+?1|b=eTms|>qIbWphA?0n|q*@K4(%!$8+|U
z{596SBaQ<_oo_;SSXKH34wt!M9qrknjQg)7pl9zltPBVt_ka(vh#uxXa~;zjrrGg#
zd)$4l@h)ogDhX#>tAchJ-b_M<_YfsPpSbxu6I{kLBIoy)A-IK!;%R=YcstZf7qseB
zTFO=f5?zwCE4Q&(Sj3nh;70>gwlS0LpKw#??=>n%at`?rpS_z62GF-Oq*SV<XQeAL
zHXpr*_K~fw%|*imD}4=bp41(i72ohZjZ7DH#7&Eo{J`1zt?)?iOlOTWMSEp3AC+z4
z_Oa44>MZI&#sgdU$w51n>W4GS00sM)5CN|u9O<KxtX13P<Vbd2UOoB{grct-ulwho
zuLe{DQv>?LV71}puR)Bbfs_>Cd%b%&WT`ew3im!(ZA8VifzRlOa*5}3H*62<=TtVr
z=>|#gH2uMLKt>VXfV`26Wc*yt-eX)B&wMXsMp>1OmD{wuTo`lp+6{pB%Y=Kc80U#m
z3kEq|u`;2x*od@daOr0nQ^45OBw&nPgpd96%S~#JcNR8D<;X*aRcq{)Zo0UXC--9#
z9di5PXOz`!(JRb9_+)k3o-A{xxePF;@?$||>sl;735jx&C5=#Hr2DiT^psWfZ+X9G
z$s&u9c6MS_0kQ+^emsjJLwB#ZFt``&E-zoCpJ7lA>Q!a}B4y1AOk2+A<;cl2xZj?w
zx3ukg8iGbvj-7hn$>g_-$;R`3<Ox3fs)T!^^y}MdcX|nX`q}-9CiaPn7rvg6@c19H
zm)zxb+;LeM3=4nKgfr`K7}4;B7(!*gVSUfci1Ei6G(k#6YCmNtvB`nFU`0fVH7N5F
zBTc!UU+;1<Wn^8}K-h{uXRPrA`pJ9~Zur|~6XtbU7!wAEEi`LY#@~1$i;VTdP7<Iy
z0<tUi)sVG~MOM^R+~xQk($H8pEk2ePa%Gpo=Qv|*&#>8kuJ9If^N234QKB$R2-Rby
zu^UE*(?M{be))x{D$>LEU|8yQ^=^;raQC)d7e-UoaU<qC;hbVk?^_WJZ=gJgZUME^
z_|Ac*@w{T~29B^Y+QUb&kSAzC9bIKNw8VrtJAR(-mORO-BwqadQuy8na<^tk^YKI3
zUikTb3KWFMeJBQZcbw4JV+^{>8$7qJvEj0e2W<|{;&1Ar6SqXM!Bt2Hy)MD9(=r~j
zMzPnh{05y!B8#}FBG@E|ABURslNc<hQAzHCqU89ljp#&bLxMM<zUyE)ip$awmhHs>
zr5Y?*YKRxkFQ5+%jW$<dt~mX*;fAIglR@(LI&loK>gwt#OCuBmYs1YN(mdG>I`qMn
zg%5GY*iV_W%VSDv`o8ss-~t~52D-sF#TO~U%c^%hdX$+ma~9Vpk#N3);Ftlc)Qm(n
z26N%Tr4KR%w*KeqU4_-PGJPZ@a>wV8xsCi012v<~8dGc&sqb5`U989FkK0&t)4B3Z
zhUJ$cKz*6|dQ-jlS^$is1tKwMv#TtqNkv`hy7)<_SX`QBIImi-;h90$mSyB^Hhhdj
z<#S=SL{-mk8oFsD54fBs-kAJRQ6^CV#zy_*3BkpY^d7gM9{SpnCk~&w^@f~V#!(MH
zbD$*e2^+Bw2<WJ3x>hGa?AR&2FZpm#9-Am?4i@^76H>GGKqYm|5sOHtbz+S^rQ{jl
zn{bcs$&~EMM}_JhIn``&tZD_~Hy@oHN!S$r4Dbfy`laepz_5VVW!t=zrOPrVOr!2N
zbg3NGx6<|jhVq9rh)SX&pA6mS@vSO9-|{j#*E<dL_03mrSdX?soj6mxiMx1_!cYHe
zM3}(J$nXeDBj;GE!>uC@mr<fY<Fyq#Bu5w%i#2ML@C}rRv}3km%|qViBVc&0cE(qk
z0x(5dn-b;vt}c$&WIq`oRY=dsAjxJOG`Rl*cn?L1tv8&@)SbY)3Tn6S?9&4P%j&OM
zfLg<>lNlD<nWzKaJjNS<pR2?@7F6Z3x4?FoK8~bmZOucYED!NL<Y<1GB)7+G<QCu*
z0x_*OINO|#8i0W9>tFEtqp!(;Av}F4pZCp=M8*$$$thk5V^$^981&3)f??Z38>B5f
zl(t^7!}op2eI<%1nJ09b{kF*#j<*jb!I>V>zOk9(nta18j@ATNa(`G0;J_Q4y|^$A
z%@!3{G2TC54_x~gb0h`{ALebg)>&i^G7bWq<AWw+2r3@6bMqe@dr9cmPQz~+)vXFY
z884oA8{)Gvs+9)otxY265~l{3WG>X>+;dcpcr?1cz3bk#@A{xA<GR6vtei;N4R8yy
zzO=ud6o9#PMeGKr0ANg>eJVX#7REs6W_WxS+jPcJ95cd5nbNrVdIf~);d*&tcborM
z<8)N(R1!WLXOLO>Ebg)kyVQDKM}6xcrN;W;&k^&>r}bB;>TaAKTjCnb1Wtm*-KWC&
zNVlKg7)0gY;8n=cyC8S`!sb)=Grf8Zm+XSbyaupe3Gay`c{qk&9PC<Cbb4ael%X$k
zu0Ek=@G9HCJu|T==}|ez=WE0I^9&1Wu|711cX`HPXb5*MCCRXPOg^mX;kgkh5f0{d
zZZXvYOOzHWnhZi2hLc|o_icUCxa@TO!Rpl#9xjj24(72Y$yw3+YVCK;g3qZkp%&=}
z@Yz+-4=W_`LV%qkcvds)h$c<I@)<2Xedo*@>D!XNYcIz#sjmSzP=^SZ+h``DM<i_q
z)6k}x9l@Ll9o#oF39aAlHXm)Nu%0LsG&j3_>`NE5YeYh^ng?+biMTtIwzxS&ipv~s
zo3ifPaP^#nux5@JsbQVLO9KM<lqk(*>>E!eF?=uGLTuxN&vSP*KeNlQVxoEaWk{UO
zmxD>qw0KP(Za%j@I_7Ht?@P==^~Jjp(V4O~HdJ*#Q>YX3cvHzy;~Z+<QWhUSO+%ys
zrI23kMh2oB^(Ze6dxsn{sg=VA%cWr3jp~?PBdwvkcdr0Ull>6{u*H_pw6ZV|AyJm#
zxH4$+hwv>7A17G$yx^f61x2_<{%3N`$19CxF_RS`%5QuELlS*nKC#GuE_~I{IzpJy
z5SDT(U165{G<u*S2aGT_8eV+=Rb9mHe(c5D5xxHD<MzD=nUc7$98b39Um$ZK$ae{b
zVaVX{PVcbgbsWw^hMgMQtFa%`%@J1|3xzLPm@u{0Eay{YthGeYaCV5DvfBxk^;A1G
zod#z4!hCPz-cubH+MzmE>U&!C8VpFxrA*tyb(dRrg%?61sle!lr<{H}*M*C;fa*Rc
z+qL^E!3J2VbS(!{a*>N1gVc_Iw;gbPV*kKy$HBfVr~PKo6JJs4$EI{*-x)|r<jA$N
z8|&KAPWry-f|{DW@_mT*J=;Dv!KT;bp4O66NQzjG7;p9nfk)48Mnrr+Lu2)rmZzOe
z+eMef9x|ENx|~`TzvwZZ)*zK*wmv8%Y1!poZm1xo7_bOvcUZ`l4Rk3KjOw3km5Ek=
zDz2x`zAl4$62E*+Q4*hh#O0rCAYhsIUwl%Ag&u0eeDqPy=Bhw010VbPC%5O<5WxG=
z7$oUgh0DRrtqydG%Z~8KP_N<XMhH4j)`}=dATRC0jT7l7v-iXIjLKvDM%hz4Uvfiz
z8Hvh|hp^8Wt`%{sb!G7VoU#;)AlTSILV9Y|(hqM0k2=xlc;Cpq(A?5S3X1p~I=kS#
zLOV~RQ>NWc;`TvMt#z^3K2y(`p}P;2L?v@zohIa%m?7rz*=Q~zt;NzS=<EeV5`Rlp
zTy>uB)6N%)6JB8OumE=4V(d&?=*N|CFzbd0eW$zQc0`T}C(qy|2Rewd(fUM}Cod-s
ze09d&Qx|ic=c)9ZyNNbT?0izTND6Z&&+!3iHZ2g?FOi70r`_c9&3s9e6EA#{#z!FN
zyJBd+l!~{TiSLq3%ccZgbLwXs(FICH60WXCLknH+xk57KcBXOMw)-hj$HgR?gojTX
zl(<pXXZ(tKts}l;*}>+~Xu*-LWp)wE!c8M2&v*5Dln0cr%8Eu^CeXB@3?Zd{q2)|L
z5_hx1I%7S<XJ~EEi;&3Tkc%>$pC^fxl#x^}eSEV-!~Y_}&uw9?ZtIk^O|mLZ6ehmt
z%^w|T3ceiPv&H;tcS0F1;{C<$9f{`CD*jaUyVt_Dd5<E6^@bO(*^`r>jr0?a9%LbH
z8;h2oGNo%Z>DU|ahrycdzXe(pO_WKi)OXM0H`})TV1CJ@Gr~75z_n2_>JvWZz`Y-Z
zKHg;-nct>)AltqnI7plU$}pPS2&PvX9d;7v?&qVkVDFUQsu{YK^CUw|>-D9W*;les
zx5tn!RQs#`ne7+I!ly~(R_QC(sE=(|H~_vrJmk1h66KuT88ou)CvKTSl;j;|DKY}9
zBh#{{1G2BWiDej2mnVqj(zTpvI6vdJjv5<DqRXI#9$&*-O!?MZMds4FC`J`1VUDa#
zRt(^PJ4HKN5LzrU`Vv~a+Yh!PY4KnL+}MGv)wc6_RPl28@~fJ`^=@bD&n_<$d=`$P
z3DJ-0S`g=VE{9W{^j!?Zo8T!}HtLYN=Kzj8Vm)7w9WcwPw4Qw*DubItS1H?H^QtDp
z-?UsK=vyLecF&7#nm*Pra2+*mA>fVa9=fo?QsWS#+dFcf^-Q=h?@&aE5zz5jO1&<)
z_O1-ptg7Mo?X=7M><+iVa{0U%GqoQ9?mL>Ofe9LSZ4@&n0+<PyHQ2aaTjSxPb!DjV
z(HoNpwO=NTFS_uSZ|W}U`S67sbJ8rXK(&4A#=7b4)*^Fgt0E&GG4lymo+(M&5@A-E
zb-JvJnB}L*_O;@QhVX?&`MzvXGcpMS&R(>9pE$FGh%uSOK^0pdL&bK$38V;%`g(oL
zMa^~9>=3cHH@?WT$kMoHu0tS@$Y+s$DA~P4xVFfX;u2#Qy(}L;uTe2)?tJ|k=&~Iw
zJ~!-SvwiMcLVmwVqL(zj;)#M2SpE<uvewgvMT<4Mb3_;#tkIipPNW#S+bxL<uw6O7
zwQ6jt*4f6pHXxjxC+(c~Qu?v82X0l3L6WM{-S<kR>^1E1n%`Vj(7H#xl)3+OK$<Xj
z<EO9vDd=<3vT1ED%O`}D9n`n&mhf7I+&~UQsS{SRTYXL=PE5(lYp`EQ49Ir%ObL3W
zNNzEjB-cDzDw*v}C7$uAz7*1hsfX4q#<L+^EyN&s-C8xdP|FD|Q*Wc!))FzoH$mEp
zA-%Y02Y~_w(#77OCG?RnhVU{7<<dy<G8A*_)N~<2G9p75Xg+YxA0m@SUw{%_TVi{w
z5ppb{@YFfDiuVn|X+9{9o@323E6$*y*Jwe=pY!cgg1$7i+}*$;mZ8ccnZxYc3jB3X
zl*~Mmb#YHj(kUF02q$jPw-haZc?c5!T<<Bm*h#^#v~dU2&hNj9Dk8KfNRo4Q&WWS0
z?qJ>aPH-F>%n^T+g?Vv`3p+%3<1l)|P$TZJz-3MCQfo$W<0yQ+*muAI%%wiGOR$;>
zscOrfbp9lWWaJ~#+3tQI5Tikrx7Yi{DIdfojG&0!28JNnA52K*pHcU;ex>)GX!0zq
z8c^QQe&CAw&|^ruYE^O0B<8`Mfs4H%tR1@9i+YQ6shXX6Ba<@X?Wq>WCw%1%E8Ny1
ze(`~TL^HpIi!3)-h@cPsP$4qWMZq3?(zLHVgRY6_Y8ib_2O?>c9TTM|5E!?$kQw%P
zIVKuqM$%UQ(yi97kuNlpyP<|TiOSpv?eHqof3V6Y-!!npyUx0eqNCrVVwi4sXQ206
zTU@27=ST6i!y`yLG%;7g=+@wO<LxT@*I`Q~Rd@)E5p^^iyzs2cJ32d{_uS2}35XyN
z$gs62m-K>0sexFB!ieMDCf#Vpy?;bPGNxo#LU*FM(r<YnfyF1T?#PrbCr>NYa+G-l
zBumEh%9y@d#hR{YH065bE5U^&`h2>t>~OzR#^T-4wl%(hffQdOW>2v6NLP!lzR-Iq
z2Q2TH{6_F;cHBTUvx-P}Ce7ZR-s#{rt$d}W?bC(C!U&nk-hKDZc5oj45;p2OkO@r%
zx%%yq`;z={679tFelR1$3^P{0$cec;44bfvhMe2u0F1@5Q&U=M)3$G9@`TG+&c&mf
zO;I?ETcA+gmFiW6k~!p@?J#awCVl?<W%2wjc^lYvv|7a41xIU=bcq$8;xHAAuVTHv
z)dr)bdL|gua(dLSTxgC8g}a^x1}tl8ih~K;X2=j%s=M<hHg;>srAPRQLz)G>hJB^p
zLmJ%6J$<Jg3_m*%<{%c)fguXYyY9_QNZRht{OM+7M~`yXK;^eLM{|!diL9uB)%u_s
zmXTK++?lI$C(2XueahbE)?i9r@y%71O}S(Wkc@qsVI}0qx!(z--<Hz<YJWOb0gj%p
z$5x-emkvTo?VoHzYnZ)t_U3CBAah|Y&tKJt%&JRV7K3Q5e_E3LS)Y~uqNwVEoITe^
zZXd3XG5j9-G=^JkILlo^q0x(vB44eNP}lZ(;YX%-kVlxHiN!&jeH^)H!foX73rOy=
zEYB^C`fVBeZrz1rhH3`)ijb_+kHc$L=>ql;SEQA^m!D;n4i@?I^NX6vwd!Ha>@|=R
z`yi&&zCHydIrh0c%;3#Y6d?#xcx$%g6Zyk2o@vE;CG#n5z6kP%(Z)_~c7&j^^0YBz
zQP*UE!XAj+?xm3fGRlR)^s*glP3WrANaB<{#_#754w^LP)-Z%7m7XL&jn?sN<dDGQ
z$$Nsn_YKQ?9>|SipNLT+E7;o4%CN|_&sreXRRq(J-2q17M`)Zl)m76+^`9aionJF_
zuX<~5*}FF}US!IpSpV2THIQlIwPU$Ai`486$DF#$g#{^EZ%T!9I%lf%$aE12QG{u4
z{1ekYgxC35J!1T@{<%7Y=hCQ&%;3mZS8vc$?*l)W`&pGKc)v9HNZcuN;c;@j8k}xc
z!H0Y-YT9qYr#N{1HFv>d5ZhS}Q<U2nW(UgJA7aXs#e4%;MPa_1roRANpFHp}q62M=
zrkGc@ruZFB%_VD>2OW~I^b^Z9=&utn`*fse^<3qr7ZQcYEb>%1k5&7uP4=MKzaV|8
zcro<2`XIIQ%I>8z1gkzlAJheW9STB@q@7qzyM1{+*(32h7g<0qqSB8Z(1G2!GgGwZ
z*A|UlsJtJK6gIo2_EH$6H)d=jI}our(-tQk(!Dz7C#I%7y0UF}QwzIoEM{Mx;!BNT
zu>igFTQ?4J9poPc_ePr4Z)@ruaxIYlL=)y*P(81AN2b=kh($<b^&%ssbR6&pEql#-
zu^0$0^5*DM)W?b!w}RrN$$1|01g@^fag<y@ilO&E1<E0L&F&U5y2E#?(9mncyXo)V
ztvIXsA_0e9SchOWxlWMuO8bg$QwqW&Q^I4usVlvqdD|ytRoAq>E2?GG4#3%m+?Gt`
z_Zq6YdN4D&ImGlu|8>Q#nZha-47+$ad~ZE|S)x=VeYEYFvBJaiiwqLrUJ*poN7>pp
zKSIJ?vhBz0Tx*aP82Tc_ke)_}F@ul6X~T~?Z1@0quqd4>k&8<jEL(a<$38iFB4}^%
zbdPxE{KIjVUp!11@WwCS+!4POvFpW251pAmJ&m3?Xh50WHpd;je~Mc}D|o-;<F~Zz
zGA4dTl|Jqv526gBV1lB4ytak<Zqlw{u~W9bltBWq%}4bdUDL9kLe=@;!XJC!a#gE@
z0-N#Id-C!=DR1_(_S`8%O3Hn-u7(mFY-fW-AKFDVp0zPO*lvCE0l}1462v9|=o<`m
zX02Z6@s|Qki5*nzo;y>F^>i!CN2NLt<#dzWl&Ft3fhIk~*c6ByPE-tcgay}De!41H
zNfu7(O=ZEYth}0a*=H1b2kDX(CfQIF%iCxM0aHiS$G<?=WiR%yZ@X-+j!JiwXF+z?
zm1U1|XBQSE-Kx#kU$uK{VOqPT=D7}Zghr%gSAKNa)~=0OguRWV0u>L?d)Kg7iXiu&
zSOcpj#|e&tKD-q#B;>ovcJWy`v6bvK$>mAZ@D@w3VV~jo&;s0Jul#MqkI&_+u9h<s
z&!~bX(FdYqoj+2P#X&%|z?l84gFOrA_~DIULvW6|bH&BWl~Uz%z4hAB8mPgvS%%=}
zRX?8%;rf;&QZ_ovGFa5?Di;w3LhnVTpWGGzFpLOf{>#HC3FrG^AS(`kuIoXNQ(pM}
z*x?H<>mfz0&Gs>7A@|E6QzTT3Y;UBtUd|%j_*rl_Kf_SauHG+>{+c|vbZ{8A+%L;B
zi!^;I$639dbQT%HvJmqgFpnX08uaPK(#x|QufCAxSxH5?Ok2;yu2h)M?p2Hm;J6p2
z#gCS82t}BX4X|U7jlbtL_nTgXP-!tK&&rnbwENf|FduObbG#Bl0NxvM7GQ(8!A;e@
zx|Q5T*&cep%O|AN{A8OybM@J9xfyHNFuH}kUPq-Iztu1_GMiaRNr2|se=i9Ak?Vyt
zWGGtwW(odjRI^gqs_v(bKl(-;I*HM6DB<7&3QFFx%)}6xJ+NBF>RF{i`xsi^Jwlm>
zaLUdl!lZyV4!NU_4>}nL(j=dhEssoMYnrDb<rtP>5kGwmCUXC2b|TyNBqr&LKV^T?
z0YngvC)fC?q<7xBs`&`;@rHN%^p1JnRwy#ziS)^}B{xIr*(A6^5MEzBT`G-odCwIx
zLq6lG*;kwi`Z{e!B;9`6EQs6i8bq|?0|{i^pdzT4#(yhl=f};sBbJNpJIX(mCMrJ<
zL%`Ka7A;JpR{O)0F1)(FRK}WLbGy?sRosRHMVESl3eu5cnZiJ**MN7QX)0gsNzf0y
zAd&K8bl4Ye0CQ}#QT5jOp)@S0cc4jN4~yRt?MZl5t{H-0z0Zc9p>Rb+1jk1tvQ$Jw
zIYvfr`lq;b3ajGds$D&*5ccbK6K^f{N4heHn~E+6xsZmO6G-g!$wbEq?z8OGd<(T%
z=}Y<A%2<@{lt9sljB99`j>63jgWQs>LQ>9`G{auwM6%UGpPngHUBchRkCq#o@Dzk!
zz;HtkpyMGU6Fr^+VQ1UsHmjJnh72K5b9bORli&F;X*iI0jo6A{qsnjiB_bC2uJ!D$
zsl9cgD@t#^I~`^XWDj&mQPv28xv~*;vdR|ZIlKhcprlN$S>gTTngg68P?ggAs%B#P
z#Q*X*98%JI^9GQ)dg#poXNZqymiX3+Il|v|64`G4{JK?xIyGNF_w>W@g$=xAGQH^H
zr5}Y=Qd=9WKVV_9X@d5yCTC(a0F%C-8dbP-(B9i<mQX|IpZ(_9hb%_*-j0UHX<1e2
zpB3Dm0I@E%VjzE~n#Q;pq{nk4L3muN<iPY0KAsPC@ipkKHJpb|xWb==^_?3H!~AN5
zG@KE2fnIo~M7kP0*}UDDfWKkqYTf?iae)b`!1-ZGvDH~0o@<o_jH)$UbfK7fdt1Tg
z)3X;_<6gaAZFyT#JV7<j0f&i6kw;(avofWrTcw->7{>Dbw#UP2vnH74L1MOWx|_}q
zUQ!akg^hbfcG}7aDw^UCkLUP9Z!1JoCikKa<jn%+lHd0rsYc$92%`XkibHkKDxC|s
z7Jo;mVqhxd9Vg>H0ySkAaOEJh=%Nbmp^8!COK4cH13BMbwCzgi+U}L+f1&sSvL_QW
z8~IBf8buCVeT0bM*s4d-*W(GZE;Bc45|=sB^tVY_$y|{>JVmSB;I1Iv5F90^dzOpc
zI|CvM15<4~%vT0Up6&Ub<W>UkLnW)P;*)gvTEiOGyJ(|drjZ3OoHy<V*P4XQP{RHD
z@1&tD2xiXi_&=zA4qW@x9*^cij>qv@xx^Te-jqz?^Ux0`v**di_x-)EwqgO_Lsr?<
zOSDk@41I`O+)-4GP)B?cwH+$?jK~6^p@eDH9l1Yvp|jP<2Cb?9I8RDUJFVeQAJ)3X
z{_<-cJK*1tH(+`3yQC6d1wD?h44O#yAo)`b`L+*VLH!&mbLlVJs+QpvP%k<|@w?Xe
zirlSSe5V&pLuhaQQ(28~-1+fA)vfLNAHI#8Jib$!`lGv2e|nuV!*hWqV+DT~jpA!;
zpVQw0IJ|eY{QvaY(u61AG+np<OSiemD+fGTT+ZM4(`$GTzP2++u;(x1=&Erz@V;+I
zdHOHwX`hf-Zr?8BFU^Oo8oxcvEqn9%Z>8gZ74lz&{MSSNKVC8s?@3xLTcZk#?R=o#
z<~;w7CRbi2%RM2Q8m+V*C{6Mlc9@QCTPAWh{S%)60t3QyA>gu?mr`On!x&gOHvKH>
zEPHG`tNkz66VJN5vr$CSGB#m?{x{2W=6~`xor#6MOV`?c%|O-F)&B==^o|P6cq27T
z@ei%>ii0bj<ocnY04K_q?KRq+Y)UHEaPV?7^+Y$gxp?A#e$_6bpdfmfFZ(ohtW1d(
zDAvr*GS>aD^IkKUsv$(d>~Ac;b)`uzk?r(8|JXi1z6z<tBM9-5FsG?wv0q1WsHo_q
z>|oZw75i@FZ@eUXbo(t^*yECt({;VFd6t;jd(@<IK~+a7a)%K}VwslrEsqcWMM?Vw
z1guk_`J{pcNt-UOgzA$+ELSP8i2C;JVk%`~Jw~lsE=e(XrdqF}j7=pg{Y~qC&{dxi
z2WuCksdD<WK^|Yl%ByIkHQAxdA^j3~9&s%4%71WwugKyd?4&$^^N(d(Xz)>AQ<vz)
zUtsvjiKn7w3$H)_L-+KT$HUJnr<^N)ute?z52u?|^O=7Di9i!iMU~c~q$T_VYA>_#
z*bzG1Nb?uNF8<;clt#$^g{V#Zu~u8}2S52gHj}!Ir=mU*eoFR7wp*ciG(9ld*7!4U
zIOi`M|44G_FGhXQ#8Xja!IKyN47~9ikAyTsa<~7;7IE;4ifZ32`Ipp?bp@YS<T2hQ
i{v+G}Q%X5Fzj6GX@NNZs9seHz02M`zM^Jf-Xa5V;XD-12

diff --git a/docs_old/blog/.authors.yml b/docs_old/blog/.authors.yml
deleted file mode 100644
index c3fe2ac10..000000000
--- a/docs_old/blog/.authors.yml
+++ /dev/null
@@ -1,5 +0,0 @@
-authors:
-  remilouf:
-    name: Remi Louf
-    description: author
-    avatar: https://avatars.githubusercontent.com/u/3885044?v=4
diff --git a/docs_old/blog/assets/4000_stars.png b/docs_old/blog/assets/4000_stars.png
deleted file mode 100644
index f3a596d7e13bd919064b87b463d2a4fc1fb5dc5f..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 57307
zcmXtgV{~Lq*LG~%=r|MGwkOHNwrwX9+qN;WZQFJ-vF$H&Ki~VKd#yg}be&y$S6y3I
zoiGJCad;RU7$6`Zcu5HnMIay$Cm<kT1Sp8_D?l9K58vO=b`t82KtKrn|31LOwxvej
z7qOj0)tr=UO`Kfy9gKloU0vzSZ7dxP_3e!5Y#mH9E_rZ(fQW!3MFf@IGS9PI-S8Gy
zo3EJH7d#t{#+$TySCoIt_S^#j%Yfts3J5_F6Xt=Vphytiqnt{^_X@ayk`muz8gdom
z*2vm}Kjm+oEu|g38M>%e(5=U-G);SJ$E(ycAMr5PKfd2L6o`OLh6)3R2`CaFQ98A6
z4hkSEn5nDq#TF!vK`<pMmI2~HNP!7@;5qsXp^!iV{0Pk;HT+F-e~O&?Y4kIc`@!VG
zt$-;1QdVxj#K7?AH#B3xi4rTo?xmzX-jicefRllEtf8?i*T+Kj10at=qU`A<V3vZ6
zB`R9_{pykG-u}uO-d_!;g=78qx7kRfhFn}7GzNl|?Yci{Qe_`CL0TA1rq+e~kf%wC
z6*S`-7Fa_^J(xY@4HWVPvt}&mEqUOZGkXw2z#*E+tm0J83<^*j`<6?&C?HMLMSN_s
zcitiWwMd#f;Cq(I-w!ZjX-CwUfu}34tfb}V<aB4V>j<3yma64-Y=_0DQ4ztw31%cz
z16#=Db^A{sIPCAc5y+B_KLn^)F;k(92G(<tML=st>i%HjBMgJ+S$4qB{hPCisxWNZ
zg)p#tvT>FH){r@X3ALU2*F5`051?qF+dsumqv1&IKZ)_hVB4%o;5l%f(?aoNy{Hvj
z(btj`ZGq%LuJoFfz9syAa6P`6;4(i<s)_m%r&QUz|LhH$8KzNz;;C+r(h9BOeTd?J
zLs<*Jwr$yi41;2ZkLa(+da?>f$WuMAT_atS>kX|E{L`qJ3rs`qnH+>&ai~_nCW(z}
zxdJ{FRE-vuGygT=c@ph^=9;|^z;MDx13C&D3F;|z!z~D_Q>S_$x(0K3L;Y_M7VVh;
z1P7Y&HW{i1X(IluAv2F6Evg3w7mg;r?;rmLVUAzP0@qG9R`32Pr44s5w2n3DSEekr
z3HuIj7T(PGKl6gQhzq8Gz2iM6gdWeFPK{#}Ej;uv`?3H3F;L}QC<IM`D1ev4Mx2~#
z`>XsYi;_W>L11St<o;7jMqN8pTN%Is(vD^)w1O--=nyfN$0kvKU@=iU6zMu#@ie9<
z{agNjTAZ}2ulNz}G+t+e5jWU_82UR^iBT8I%InpfkN;kQjqetzJtUa`BeN$)gmkY`
zI;@`ee-{7IUY^VBXDDhEr+8{{-qG3u@FN0iO3v%x(_m8#H0<KK<qj<j9eDuULfe78
z?pb^bX2Pzo3fKA<s5yLu(~o;=$Z62Zy6EPtT4LQ2qN;(#X>B=nS5;~f^b?NdTQC1<
zG&o*cat6D{9&k&ffMRR-DwY?tah;o}0r`&O2I)QfZ#&dS)Kvs(G1HE}5+b{YA=CdA
zI1Aqu7LgJ);<JN{#jy4K&vw`_F5rlpz<^s<y(>3dCWQuH8#|;XPRUEyqJ@{9Wnul8
z|CarYf$8H6r$4Odfwl{P5mIB~f-(*SwWjG|22{j|*pA@FM)!YOBdV+SIIg1|Z_}l6
z=HhOq#7FVC)3``<<$?fiQdW*nOcikPz@cm>j$CueDJe?yKxx6#X2|=u6}A1ce!|Iu
zZvB+X8l5u#LYOV9#@9BE6awf{?~R^5<5+<j9APCR5QETnJo2e-J`xEc$6jPFBE>n1
z9$4$m<*t?fGn*{lN*P$3c{)E@oU5pjSci2OWbK}4-}W!%a%jDL2H^2%v2dqP!0j3c
zV=q%LoFL`O%Q)PsZnF7R1UjEs?1)-&E{$N=2YPl*0_*P!RArdI1g;!8H0t9N(yDO5
z!5j2=0lYRL;{R=0^5}CKVpUALpwJB@pU(A}1e25N@URPV;q%-AHNA8R2ERi~iiI_y
z(n|FCvq2cJI0d9y8ci@91kV%3egRH61u>4mRcE=d=olsi1jw;N2zijILZeK{(b7PX
z#ChmoL2)P+m)zdFEp>pF;}ySJlLP!TzXNJjuR0$s5x|!LZXfvEc;G)Xu9jil1XShc
z1E~1<zkG@ordLOK!q=s2_bS&hfCu_+-Jx08k;^NIm38HFD+dvRxU^g1mFx<O_M~L{
zC8UNVBnEbG{raKBI84^6cTTTch;s)(m;=%84Z+`2w_JoSF4T1BQDAlzU=Ho~WhDFK
zvi+Gk(HZe5SzDT|FOnUi-&P6TkBi4SEFQallVQ>#yRR@rZUcAqV*d|mcJ&Kk+q95h
zAnm+5;Ia7tPb8Vggt7t5jJRj!+a2J#?$}-F7Fxl$rY_`F0YWG+Fm}a7`lRIfcnE+n
z8<Zt$K(6viK?&)e<C!X47!Y+P8KyndP-?k>5AwXoyAKGhY#<J9?V2WXkAy5SBRip~
z284kVE!{;;zhVFi7mrt`Uf3Y`@`m)UUnJ8X7hGBhhBYEtK=e%w4VgK~$twID!vD6R
z81sFkk)SF7Fb(=PxE`A~6n@Yr8PbA0E~KGprrnIO9SrX_K4wo!9<Qb-)R#I-ih*7b
zAK#a!xDHy}^lMS#UK5hH4Vq-D<CD94RXBJU4S0OO!g5emDQI;g(1FX&#CU)7hd?Wj
zpAgkx4_Ptq@UV1><C3xK?K6J&;dR<PW7EsQ+ksNiLYcse9zywlAf&*k2)?O|?2es^
zWh)HvyGl{GtqTVB2Xq=(`496i6VA#CRb-2&I2IqCFeUPMCLa(n3z+=KXyo?d@+cxe
zuZCbB0$&iNze1><Hm!df4h=Gox_XP_?N2#OYP_GAi>gQ~p9D-n1yfz|@%aFP?hnPx
z-wD!~<5EQnl09v`1Zw~JiCz;@n6{6;?A9zj=<2SM4UNJI*B^LtjWq8`=qxlEBI7C2
zy#a|Q%p)tl{3t__O|VE1%Wn%Mm9-)?1jiJGXI-;qmEj}psp=xJ3Vs6K`2K4aiMvF3
zrC6pGIC{B#APgvbtZhRcKR51H95<e(!8>kZa8uQoU|64*PugorY*zIU|NN&7J*GLZ
zO+mmdo&wspHWE_jKoTiyXeNR^PWfMAi(~l7cHm<<Ju%Yi5eAevV6^BII*sC;JGINF
zbv-vabBJA1rwB%}#ji@|8hyjJr8Px4RV-n%JG?epU-Ynk5}SHnoG)XzI8O}K!9;Vt
z$-mzp6tq6>%`-x@J_9DiiWaDP=JK$#{-+W>Qs}m;AYMoJ8#W`&6YBUj&+_wZ>%F`A
zItOg$JKYld2b?<7OYx;y@dPGgm_0j@n2UpvaP0cKB|amyY3&fmi6*Fy&?8yrG^-sY
z=|5GVG+vA7U>4=}#wPI!3UCOf!xl*8`==Voc2ifjN8{+4FXut+5|a0DK`98iLR8Lx
z8Y4=G<$tczL>!wcz$Q7g1L|OSJ=U5GKu;4$U`b0$tgI?Lsd=UN@wi?ZND33pAN=PE
zKa!dEcc|AX=`2L9Bw(<bTH}uZi>kGN6IJmAm#uG3c<L>@2(;8|m24Dm_eyPLDzEZ+
z1^^7#;zlKTcWi{H){X^B&XfP=Z1~C-om<Wi8~1?rs#ow50^bn@_Y<Hd$|9do4CjB2
z5ZgZzOt4wE1<0TTrru@p_!Lj2u|oMH>DNNXGsvPXJ*nZB_bBIy$Xu~pW5%_MU>``4
zWml%t3xT_8kZ(dws1kCJX;9A<#ln_c&8GCoQ9m3QgD~+Q1-A!CFk>t%ISParNL8V4
zyd=dNmHfs_yEcF#?J3i_(J!~rY6QkogEi=<FZWa1nCLCn$8wDCwg1kF?j{yBl$9No
zdyYXPAghfk{(k=?L%<$wO886?oY$toooWRK@tdM`(<V!)<ey7I7S#gRCPG@7zi{a$
z2N>>|47ki$Dl;4HZPvuq`SJ?yo8~0A93UsYA5gg6-sMMep6~bX3ZK};gvDQS+GK60
zW2g`be51dwF_|?hYhbztk`hCro;n64C<!qz2^qGy0YI)9H)K4h|D;YdlL<(W26mhU
z8vHldhmQ~+zltYrax*$Kggww$0|O%iiWbM5lH};*U{*>9Gb?_70@Bo7M4*siwG{ud
z8E&kEgbIz9CY!#8rX&tK@S$8CJz-yUbyr9;pY{-L%y!Q}?V!Fak5Rr&r{XD=(?i8S
zw=RZsgXTmAxJ6LFp$COPpYLn8p-uYo3V$8cP4J#Wq~v)BadU9!>x-&6s~WdQwpG2)
z2=^EsIP2H^DS7sH@*8w(WMGqG{eG~*%li;ewYqc6D-mV-Qt%$clo9AWh{WUs)O7CM
zp_*Ccc9tjiEHO<almGEUJapKn8X;+Q0to{L-?_anJbN`JN@_sq;!EBZe=Bz8m{+_$
zNbytT;s4U6`#9roy*1_EnbUnovUAdbGQ3tt#&B+v&-zb(FneUD%F0T2Dk`cwQKK*Y
z9IqKDU3Tf~U&;!cSI`!Guc;e^c+;izQ$0yp5#fL4khBd1`|^L_5k|40V2GfN5$4)|
z;H<G(4cx!jbMTU)9mHud;kcTxwQM#$NlMD7+3P*g*X^A}yG<I*t)Jhlt>38pnyuTi
z@jCAt7#SmXTx0phJA~>B9LC%V@J(K~b-+1@^RMUfF$uaXsf6i+TvWPTqYbsc{64%k
zC61-xBnB@M--*D?52=4Um;7=-c!n%u#Mgg6*mJM9uSx^FzBMkqB&X<h5Oaj<R3hBh
zg{>+ez-+Qas3x&5^klG@D@Y+oL5bJ-hbmI(AK;XP`x{gV9CiY6#r(|-k~!bXY4W~J
z;l_{gH2=zxnKzd(u?Uh@$7HHN3rn+i59}>k&E5-Q&+QXZ`OHv<_6u)MWDQXCbdZ%C
z^&qhKhRP2daTmU$&0A7JTJ34eyDH=A(fwfI8Q_l6#8+sK3)O$2=Rn%!>Wgp$tean!
z$ixFBlL%6GP8<7n@%ikqM7F}%LI@$Ld0*hvn>d&jX}3(@7S%XFK~uV)eH@yNWp8AH
z$g;ZAZki8F8PU1U`!qh9DXdD!jHn2@Gl{X3j=%j9^`Ws$Z9Bcs4(2id<n|U8NN~-2
zfmTOoQ(Ru3lp>D#4G2YgXqDJt@$UJ<^W^-j3=FP0q@nC>zQG-znF6iG?ndKp_ZyX`
zEWvs|HXJm61I?LDKLZ`aL|(Scf5^RsWrMi+3*dmtVAM$lkr2I8O6xOFXQtT|Q2E3g
z*qxW2A3;z47As6tx6=U=B7XuP1N-Ab_^|C2=4;70H6b%sauGzqzl>qOspYzP0fjh!
zu^64D?W@^_n$=2ph@OE-DNx#!-6oH*tKIEi7`8vkd9Dz7a6rfD>e4$TV60}vdfql^
zjOQ?fE<;WCx6;(6_hRD_oy`Na=0lIFhz|5}yLj}!kPcfUSZO8;0Ql;AoV!pCg)%Wj
zO}Zl}(eJf1dVgktndXBvm69?-<Z8q~4ueV8GbU@y@e1ZZ!)AUYH=i&|Bmn@u-HAK<
z(#Ll%dCNm{SHI%W@5L!)spN-BDWD@O@v2t1Yym(~cu{`Xfz!+vdOx@>bUzWt@0cu%
zQtVBPUS7l4m<E;8Bkip?Rv8>NYd-J1bThec9M?eF@Az`x6Fgq9r(XP$`oK!%w~JuK
znIuwx|4whAqD)8`w1}6zMGlZ%#27K*W}iRbtXhwJc=m^xZbm62uz1teFn^T{clv?J
zrl-)C7#~vOfrF#H@3<-FQ@rwtzz}mSs5h&SN%@yXf4utySYM<@O}bV5xLHy*Tv|l7
zvLc#UiCkSmxX>9m5`PyQAnbvGxKOC>@3CuR;Gv5ILnI|PC?wsLlkZQ;3W`e*7ZT)?
z3<~0ypSLR}&C{|0BhkpkwbYQ;X{hRU4OB23BSQOjCRCo6f@9g9NTpSI=dG#?a1rnO
z-7)Q@BYZ;6clM7F!-q!O-sE*w7b>4uMa|g0y(ca4@#nSYV~Kn*G3bsU5)2Z#wb-s6
zIAa%J<2G6Oc+uH3(r9t+Qe`B5zt)Xf#0plCH_ij6(W6W#u@DfoqeDAj&si5xFQByh
z)5ZeytXVa$-2IBHQ<R=FbmLsl87V_LHT<1;HPH4DXRZ$oIx@y2)*WHWl%7O5qv$+c
zLi21=bV#)*X9KIPn7ddlp=~xIZTrOnl74~V;6OJu5*z5Z@Ny_A67zy4rH7MnVyj9y
zv~boy9y;Y`A}F`aDa*h@gT=ZKnpkq@C96y%tC(5$TG;?*HUgPsQ4|&JjE#3GDSMPO
zg!Tq*I4I<Z-0#gI2aHJPJn2*xDFwr33tVfa<x`+Akjl&QZWmRNQDKg<Lz&qji@2nu
zAp%0L`)~>Edvgd6jz6T_k6ld@lo?=5EvR{=Y6wCloTT;cco7vp)<EXudEz8f-ow*_
z2o>Hoq!YKEF?rN^xxo?@Bm7DH{{qW?Cv+}}Vnn=KVrDEP{&|lx*nnFk60HeAaipKn
zq2`L+*TQ30ZIDk3qcZO2wr()6ZHY({;+m~`q3JPvq{^DYLtT)utd1CC`hmkqrK&$r
zf|FA96U|ZnROacgvNzCt#5W~XF#Mp7TT&6ia=V4GOF>Iuv1F4pl$M~S8&Zmm`PQ7N
zV!k%71r{Umf{p(|>0)hmT_`aMjQ$G<*!nu7rKS052YAIFysI^57J3b-7f1Sn0KrUi
zNzy3X2-BRCLc$ut6VLM?QnI{6{fJdMPCw9i1^wB-W0S`Btw$&;8gi??+5OrI0=>|~
zt9{DOHqLXHo>%JO^M?l!#%Ge0G=1&PB$A?rjEYKc_{2>}BET1#Yzao-H_1u<E9jGd
z*REm}YL*#AjC_k0I(z8Dv={$}al|GPyiUo++Qo@&rX%E7R$voD5Nz8|3=f1}60?Jt
zv95Dka3p&_6!8hzYWp3cq?6~GOlZ>}r^)T#Ik*j_2VqKUG9yG(X^9t>r0#BQ&7SWa
zHM*WbuKrGP{4Bkfi7&noyd_YQ%-{-Yph{p=RP3<}N(w$Tuvy7ixuQL<6V@FBK{0H{
z7776F$Y`EvFaX~&{sHCFiuWA{70(k{fBT=wY}qB^`s)MqHR)h@0ndj-POJ~C@R=qr
z^k3Ce6i3JXlAE@=>zXCXgyQQN?<TRUB|1Y(w2K|jGruObJYk$Im*W}r0l^gZ8R4`Y
zn&YKjPxy6+S;7{hSjr3nHMPj9ayAQRtFXzTT4q>7B%qn!wKFUJHQ|!kKx`>Nc1;l-
zNW~C~fZl!t=O(Y#Oxos+;4ulqiJU5c`NoQwCBhG_lkx!_whHG5DUEIo!LfVnLYWL4
z3_Sd;cC|_3WTZb8dHOdg3*OhjpVi%g!D5-}oTTy@LSqxZCAK_m<E4lE=F!o$HkEGL
zgFQS^*yD-;++L#j`=Un}J$2VfDf5pHOd6Q9eeDO*YnEqr@>vo?j3Eim$NHq`=ml`%
z4c_By)s@r%hlmOvV_qOp@MxJX)Km9PXaS15WOE)PVW_nnS7Y4nCm+=<ryB_^d#WY5
zB4%2R0S#313|ev!C?n_tQ)#eiJIa*CNVFns2Gx?)ER;pjAjvlCVU{fE)*`T6QREW}
z&}}rx>La0ujlqK&L?VmbpTdJ36r;~KF<6eD^me4=D|@tSGpaRULg+~gU0;0AMMXyE
z?bTT&OU69NZ@adQ%}QUZM+-D=Nsoilwdre;SlTkD#E=P$N1lapX#sKRI2d-?XVP!g
zDi>#&!SNC!YQQF%xU@f1=k7nnS$wXE^VT3xo>GPWjt<9p4#Z)FNj9zE`_-H&ry^w!
zwCj`nn(mVL8y6DS^u;&JMv*mqTH4lm6?W8e8Hbt1T#$G~*oM)++hb*0;=HdhS0Z=C
z(S-Pm0$}xg#9!v|A$uFcoo#QLBRa`_HB)oVJEUU|ei<1l_xR{WC*=){7c^BeZx)O~
zd(4u~ARH%bs+R0rp=fas-Wm<zm?}GI)p978-mN~mqFGI`zoU*dhEOS07!^WEqwCk^
ziiv`f@SY$)Bq0O$=3#Kk2hKnXVys!}_F{~q>y2+=Nz5KIfqE40IIvrI&Io>ZgLI}{
z-zX~HD)RODXV;u{-t(|rn}r&Rnk?-;4KmiJ2RO3p`D^xmaap<h6(Xs5Q~VJ0PJ;iF
z5=NdIUEaq(i+Oo`)Y%#v#auU@!v>G6yEAs|NGIR<g&Jm>3%>vG)JCgi_4>YS-=13S
z=E6i{&8x7aXh%YdoT@suY4S{x8k?}OyZP*oh}o+hW7Xo>O|<s!y6@bbm8GLQG(G~P
zDfQxrJZuCQ;E<({mGabC@?XV|drHCW_;2{gou6S1<kn9va7x;|k9Ah7=W$sH3z#DI
z_q4$Ld^u4#sdHA9b(QM;le6j(&U?xY(U)0^sg<FoXAl<O#=vmjkC3Jq%<o!GJkJ|Z
zov$5;dOadYN`)%$EHN~RO~L1~G!}vfsuO-SdTmlA$|w^W6zh`G!<<*St|z{)JE!_E
zf#?l(Nt~KSxT0800Hw#0w!E%Ywff!O;6mpU@!(PR?{B|T5W+ZSdB(XDlCop{#QZhe
zZ8y_3Z*@gI)8xYntsca|#B_dbox&XX`T~}oLR)azi8#u0&M;SU6;@piqbtgJ;wg8g
zuJ9-$LFLSIz2x_}WX7@{{y1aKq7R`ta~LsVDln>tz*QRR2jxuY)3=myMdbbn9QpEb
z*b<+>6~YZJWNoe7bw>~e9%OXcOdmXdOv46N*co*i1Y72Q%HL3_B`frkc$FeYVCrUg
zmx{WsQ_;Kix5tG4nYrik(;Kd--Ldq-gH*`e;O$H4$GK8@wpa2atK}%?I}(h~BdX?#
zK9SUM!;ZSG2HQL5Ly~IU!)OWB^KKZaY+B$cnh$ya#zdwkaR&8i>*d!+Xv8!xJl^i7
zU$SU_`Js_a8b+1SB{vjarP$@!<HNM8RM~I;wF^R_g8!=aII^j|XimY#>M>MBK74AR
z2z&vlLX&s&N}oObMmbN+Vp+}*2ql}Ebf-V+bDbBdaa4Sydw+`WPRU#>xvO~bQS11m
zjhTK%S*?l8>3GJy&F93skMv@VN)AW?F~4UNIL<f#Jw7&V_lRh38_+D_u6%5M(j4Q4
zAq!J)SaAyUALS@Lj2ODHTakSfUMYEPwYD(VTlKQc7`KTqtgo#)eUTX9IYVJXo)f|=
zi?2;(I-gi~eBP*LUUzyw4|l5O!;xsDSv9fE>im*)I3Wc@83*GayEAlbS-x*p;Y?Nq
zVL1J7h7>`<mHmyc18l@lZlG2kJIERsG<plp%dyTQ=VtNVq`xQc!Am{PytPhFjC^g$
z#Zp?bSC`}-0@x{i&ly)V#OAu@sd&Cr3#ZeA9PVjWm76O++;{t~s9_5|BO)(uW}RJV
z%A>FI>-rYk@xAYwoX@kLt~lfuYmuST7P>wVSnmkPW@cnQt9>Q6HWC;%E(yZ%J#e*Z
zgvbvO`Q^q4P$fBU8SwP!)4{bIkNDl6x4(u1`|vP6pKRh$P`FoT8$P*F9PaFkiV=Tj
zf)XJibapZG%$u{c!)yGvNJVE@Sy_#Xi;BD<WH#&*Tq3hQGoe)+e`wltqh2kKV>_Qk
z6<#(%Lb#w?77E(smS@+wVk$Dc5uprYIQy);fi84>@fpWV^=oxvOm8t`Ie)eqO=TND
z>7LD(U&2dx_|ThJ4wOoC!-d@xi^|uH*T#9SnrAJA21aHrZ&5Su?hshOz1BGB`EWyD
zEzLJFYHF6fKRGV&Kd@YmCQ!b9*h*5<UZ4nFSlab62`ls7(4!2a#l3!z*squlqE2`9
z5?*@2r9kC$-ZB)xI@Y{Jtqe&p2+XJ3O6D+G?_~Gln+7SK3glY1kT00IfHvgLoAtpZ
zhg`ZKw>mIDtRWkG=rWqyfLWfc+C4Z+IFgVxZ%3L=9Gm%aa+k=>)}3?Xu>C1`P9?D;
zC3AwafTkT!sXy$3y!L(s&mKjkYT5B(I+_q{)yf)Ba(#tKdAx!~MNTBs(@>kFZCF&^
zsV)<<D8E@=Fko#K(Bz(53V!xhD>i?Ua^zshlv5T0DJ&GQfT^00(Q0kQdI2xQz}Oob
zg$U`_F$SF0o~e5|tMMpo>z#ADVECeFMiK?M_Cv63x)XMxSB$(&R%1Q%w?5Vxf1J{2
zUTYLfuI_&kS-*oRMrZ!c+xZZc<~ZD2Cn9}+^Zo4k88UH>MNgC1iEd2U3@Pt&Y3F{T
zAoT&>qMuqkZ{`B|Kdl7kW)RQ?N*udR>*rQnQ7Y(+w*B}5lcO-aY6%M5I6o$k5)i`p
zb>$uTgDfl(^r?S0G!!0cak1_sPu$Gin2(wysYCZ8{=L<)-&*`U3*h*W8MFAYCN;3W
zB)YsR`4;fJ29=yLf4kY-P^}a6op7wNS&QIgFa{9A&1VlQ=4Q&Zdn}za^miikV`wLX
z=OKapmLLj+0k;p;OLf6oRwSpUN;|kgWsn-)PveTaH=nmpNz@}H8@?YrD>c42usE~N
zd;8=Qf5b`-XED26A<rZFHQgz)*e@U3@&z}f_2IdpuFk7c>yq^|j3^H&<7kq^I_E=z
z-;DehYopBDr@LPSzpfihN|q}*rKZ%H4;OH)nMP~oJ#Wddni<><k4xxxzfeZ+K|wBm
z_FT+N@H-e>{V8~a7WdM__}ggOWq$ICY-hV83BJ1>MqCGjta&5y2XB!fqBO(029@s{
zsDsJ5?B25hH&}$wyP(CESHdkM!H}5^;30(Kr`MJ{M}2*}MAH_J5^v`+@fQI4HB_&u
zX|~CRHcA?E{&f+i!?LS!Fe8rvC7z+KP6Ce^XZYA4K0MqG9b;!~f|QsSUsZ}}ECGm)
zg0deWk2+NOHrza5J7#q&y1F(@K$s-&6vyi@DOIy+N3$7%x|~5X9Xbp=EsazrL+Ew_
zbd=AGJgi_CKoCQFtZZbi$m$ns=8K*-tRDJ*T!1#-4tk;WwMOL)w;z1AevHdHvSSTp
zF5u@t#r5xeY}@vTt^Az$-!hrtIkGv^k~_w5`H9=MGtnS$ZxibN0_VVUbGU8GT#pTJ
z!%R}-AF4|Fu8?vE{|B)6=uRbJBg_M>M_Tesz1`-pXLOw?%g?is?Vt1<it*O$Y8=YT
zIzJ`%A*DxnJrJtS7fMHNt*H#>FI@NEo|Uxb$rYUrufE%z9*B91TQK--UE4PqLP)R^
z#Sl0c^u?F`G3Gk&;ELKI#_BD%x02~hafv6MaZSr`C{p3qMu&d=k_{dtOg5x1)rQx=
zX^T`JQw>n8tUI(nN{n9|G*W!RY*(d}MOh7?qV35Fr#Mq13!=%L$Iue{v8lA8u$#7r
z2tfxPwIDJwDBF*G(zm`61Ne0Ww-`g7P?r!0X@zQ<vcFADbTv25aX5GO;<ac}oE*9L
z@wFAvA6j`Fg=aL|;sJi5R2+1K<-iE(V%wJB|02vACsvP}3!Wv0Y9bD%T^H(waLyR|
zXvNGM*4i^f`=@<R3U7Z%(c_~LOA=b2%<yWl%Y=8bVkzGBJdT5h4v2<X<l=p~tmDjA
z;ghY-$DZzmpBL9u(O_Z(v|VL3v^>x_)6Z?a?8}Rh)pnxf;*MTHwo*a}VE#5i^pdM?
zG{NIHFe{3c4W>}8)Rhi1%^RC*QqA^Y+3+!W`)a0LpB#+HSNPER6f2u?u5f(-ZS-|a
z3KPSROu|WqmwaJs9dXI&)QL@ioCZ@906g~fLgVb~8}QDrCAPIe7?^^1R$`m!_cdN!
z=JSScQ+{*%-bZgw^5^3kJ05j|cWt%H8MSJ`>D!ljXAbAqJIxo%LdM{8FR^Q(TClWJ
z{^zq?L}s+<0Iu-GjO$W}&KY3rHizxwKJr2*SVP+mBx4uA#?`NS&@8mtVXKPLOZtv>
zl$!BiXQwN6!@LcXzW7gDo(z9aD<spHic;8c<|;(Oij#w2X=`LUGT4+}ZebU?KOq!<
zg)ox^6@r?^``jZROr=L5W1xUm$Ky*WQ>>N>VbE5nD!r|Nq;I(+{?dlWJ~7JlLUAw2
zj5JlQrL9vgjCS!9uFZO?8LScA3}9b%_~13tw!;`a$_#SA>4cF8d(q*(GQadWbNuY8
zf)l3Na@!Mo%wj7V-v5)E(-f5<k;Y;Ee1Mmm_{w@c!XQ^RdmjEj#L*E<Ap!jG>+pHP
z6Uy(CTE(GO(;02<?gVkQ0WNCZ^dljoI)A7Y&T7n|6NCGGuKtE>PPTFXux#~m7nRDI
zRG8zAGuZu37M*C62vrPHprrHbiqoOC)iW@Fg;rhW&^M>#VP3RUz%S5zeuU#u%^^DV
zH#d9ClyPQfkmJ5^!wOKoOFHf3cc~9~@yi>*@x|fK0GMLT7oR9UKcxK6TJNNMidw_I
zD&YlaXIRX-hiPB3xA%H1=P7vwv8z5tE(@ZbO4W^ttQ@c3{dV?WIa5cGt>OWvTl4Zr
z8p}T74w6*CmTH=mHov4Upji-)3UMas{)Z74-T~S}P)Lrmm>?#G%$8qn@Kp;eLz`$*
zX{esw%E+RJkXQSoQzVAfWiH)JN~LbQWT#i2NRkf>ub12-nXEsk^=OQjd4?ePNb;tm
zuNWVpzGH+BiX4T{Zuwp9CY1gQ^8L4A1-sK2iEI{dttl)<lrz|fYQNu8{y)J(^SIIu
zC1gpi2KGmJFU@fHzXHaFbMkLmRn0u^Wb39le$SfbM3JzPbu?cLHYDl|JWik4H}P5h
z3_tHj5Z?T=;`Kn=QeQvHySri*+Enrr*sh(5^RM3Xv$PDHO7(gqI94Z%iM^@GZgn*o
zakpQU=!mccqaCr``iy+hzdn+7)&;UtfyhChV6%~*6Ks9BUSm_;XUwu*Fi7H-C*;(9
zLrsWejZEJACZZ%Rck)6Q8J0jzW!7Z#R*-8iF4mij@%tqE9u=j?AkS4h%+QDzH93vB
zXI-Dr<-Dj>_7m2paRmd@{3!Pk9E?0KysN$-1v2IjqhRI%b|kK_w(R?4kVA(A_v>CG
z0w<yeT^f@!Z^YUAV}3I)XJFCp&eZdM#O4T!#DNC#RRm3CQt+Q~tfmNBm`42%&HH1M
z#2>k!9<-vw9lZ1;=!wvKMi+mXL;|<c*WH<z&2{eZp<iB{yr6<VFGEMjq{@c<(us5a
zt3tK2tkWF-$N5R(L>sb#&9`Et=pxFnZ4VI;#YQFUKy+2%B6{>Ju5sQKH<H1UUTa|l
zdj3(a<~!+}y6q)A7b9;lL6)LNO0f<i_%Or8u7QGcXBikPIlru1dlKjV+^rTs!XEMy
z^CP@ZNv=1|TsL38Jaa=I!bV?hf(y?G--#&g?Ob?pyjtPzn1}l7*)%=ojjw>d6HHYu
zy0iLJ*5$!bo0KLyiFTkanh(a-oh+leJu2+(lBCv5SuMwzfE%TJQewQP1`JNEx%N*=
zTHlcYheEKB45~c`L@X?a85&48gbiL5^nY9%%Q{(3C!OkM4r1b6X(H2OnbS16?AB0>
znm}1Z9+~k9SjyHXg!re_fOtFiK!F!e!UR5hnzBLBP{jsX&C$OVRpL11Bc1Px4szKM
z4qScuO%gFJpM-w_NR-uj{l5|CUYCRtheiLUl_+@)owU?(TW-ev4Z=n1+p0fq4cl4Q
zm@58i&6)AM$`?atKAl!gU~~LR#h~#1)%Pe@I3p4!6yUS8ZO3e4jau}y5goU4vMNl`
zclw~2dbj@-a$*-}PRXZx&5@Rdt)2XOdfo6!{o}j*vnMr;B_{}}ke8V5|2SvPvi^H2
z?>qK(nnt+-jTv?XwnuWTT|r<59?ySVd|~-~au2V1AsJPo5Ff<DQ1-c|Iw>M6jHH9`
zFGl1ynHNnF%ni!t{*;=c;;&QHAc|);&TZ3*9x>buXDLHP%E<};;u#Om$rN7Kyoz@?
zN0#w^l^pu`yZpWF@_XBMq1OhXPaJHd;uzXON1HcAE`O=&v$85ZW<->Eotpnq7@>O|
zH1Om}qTV|YR-j`iQ`Hj7&GY}xObtEtJ-NiFP*8cXc6kJ<m&e<lwyqXIz)4L^_p7T-
z1*}ydQiR6qrf{eJXUMZ!+5oz5B4BT<`W)mYP6!kW?Pq-D(}w&bcsJ37XMGqO9dfBo
zUEn}INW8RI0ZqiIDyD+V;IGmRII(SC`V&aP(7ZO7>Py6%5sd!23M0*K?a8qSy1etp
zcKc}j!9Z~Wqqzt6SPm1JYQuB6(0R!%DhlSuE5toYZ2A3*ul`_Mn_dgsh>MfCL(5!B
zFgp_?rbl^Ig{KAI>q_12vn`tx)RcLDAVIy0@w?iNu4_tUZiCD8`5>_0BJU0TVTz|H
zL`B>9|Ir~&f~=99_`w}Q#$vZZY2`xAKy=l?acw_-l!Z*?Ur#XSFO~unnytvh446)e
z_8Dms#!PccUV}5@Y74MDuXsu2)8L$FWsl#0ZC*aCKDK>tH(xM!wcqexQZlEr5x=4u
zyq`|g9+g4;C`HE(Mk9UieIon2#CKZ13@@&}WJ((wAzzwkG3CyMpoxsvbc{ZZ!Nv$Y
z(4d1n5`%0?zGtDc+ULYkCTBN)KZxmMgymGFi*fGbF%}*{Ne?y^Iy~@?MzG3?RhsJ<
zF)}D@AxWqPrDEi*m=gofz8y_M8#bGw>(A5&TMpY$r5S?`7uR`&h9<N)KdtCGtc|5K
z%yg0mDS^k!wWeCt&rTw;WUQcF<PX(V2cx2W19?i|f%P=ND3it>Gt1A9_}xGCp15vk
zS)mHG&;-ByLqoo451UPxC2-t{sO~tJ6sWOQ+qXE(woaK1SR9gKJl)7aS?dV<8-po`
z=$<QA0&2=p7`>ZYqz5;t>+Nr>&iLDzZ|a*=b?ezYSu7|dLI<p6pY|Vfgqs%>-H8Hp
zKX@)STVXb5SGBn2X19A^vpcwUS?l)?TCxpOgtdwa^RhC4>vpFCK$-7>=9lY6)I{po
zz#CytQik-Cg6@@~3{AuuBEQihXE)LHAqvy<2bI#-BfFxM^u?3w)$yqx0}3i-=c5IR
z96$Jr9dgF!H3dzzEhNZ1{vnK*1GW3nXXIGP=QsV)$r56p0WMPxvq^@1Wn(3>|6-ur
zGPgap!S|`Kc-PWq!{*E-mRENE-KT^6AKPSdYEl3!ix|*tl8XOR8xvoF@|6EIG#FDS
z|DZV~PMIru<oAk&WuDC<HH-;CpwQ2cjfjN}PxhfLrL_BHZhI4x*#vgVy|^>#I~rfV
z${60!zuUU-l5|5`&J>0|-o(M{)>kEMD&S^qD4oSNm_@#?_?N8|JDyi0Y@3e!fqy;`
zt_&CK#&L-5YJ`c83*2uoAJ@apKzaKWIPefW>#?lLEl3TwLCv4vnLW(pKw%;SNoj1o
zc%1z7v4*NwhJ?ueb+FC!^bljPdwPYfC2U1RRB}54WPLkB{_IDS-=!Ah6DvWdDbwl5
z?D|9MV?nUOZsngww<{DGrn((%wm3Kz2rf-USUZ1Wr;S@!AGV*_=Dp^L+g-B%=KW4b
z4VP$?-SpkiwGm|IG4Yqh73eVct6o4){ji^N+_TeDbm6<_D<`-882a9D!@>?Sy_n$X
zJ55wOhFF+X#(Ege_1Y-&>RQnY8ql_bAJmeuNp=h!PPF6_R~_ZH80^s)IDO_V|1o|{
zi*yA|)t6fRJh808P%1;<Tw{@ZW1U?$g185TXUQ;E&)lg!C9BivG2(c=A*upJ(lNco
z7U9_q&<_|svpwyPj$A7EF3OhXwuqjof2i5dUYU8`!OzL@r<J*GNB!XM_O!_|5fHgO
zp+TEEASK^&Au-WM_%2L~o~<?RdV|#u8&ka!Y9<2iuKUf_YU1=nMUOlyl^c#Sg??_Y
z$T<=PlJo{WgUpC-Ra`ULcP)l}lbTUgVgGft>y_qma>rTV_80VPNrUCGas5wIbctJx
zpvin+QKu4>0cZ7W-t921E(Fl(oDG%g{lyhkgEI2@_f_1I)O69<mCTzkN}^`E?a8+E
z9hp1LKfawNid1z6O3Cp2{!O~IfE8j|P@lG%#H?;rA5y}TO$)AV#`2~4@_4+~M;?&4
z1`KCCV(#ZAVomP8OAxP%ODJB4KN9PMY@tkcpzq|G9f}cVPG$IgZp1b(JKR3SuV24L
z?U@DsF$cKGykWTJSz=J*`gWO$e}FXDZb%Gn6Ocb0nI<$Wv#e{9!{{2be*VQ^B9{dy
zEA_wf>qJa%i;cpw$1}M*P|PkV73hY!&o+2*ZtJ{z1~eu`ww`!|PxGMm+a1F!2@2?k
zNLa_<7~y*{pi>Y`PfD}h-bY-mZSJ^;@KcGtW(u2sctNY-x4J(H0p@1G)Tx?OmSTC_
z5!X~IFF9inf(BHli2;TV4i3?gIpF_6C-#Z-3{^ja51Uo82s!*0n!q=Vzn*20M$-~s
zQSgU}4G01A!&|DxHgA8b@eyuo`@(iTSW>p#!Z0E5FyINC6$G%}Dg_SFcN=ULb&Hes
z476VX-a<(jl+Sh2dx<>Q;RYjNr%ITZV?uJgLyBE&r?={W0Rn3}o(4<GNET*w%|RI`
z;l|bQ`c&X!xpl<^Npb!bHiM@ZpW&Sdc07!5`$zVy|I;kKH*_0d=X#e3jH`nK2uiu*
zo_gzW9Q$LKUyszSZ#^izNO~l@mVb<4<#vW=jsHFNJ||mlca$Qe-?g({-C$6wKYv=?
z^(86xIRAdDU2@x2_aY8g2;fRi&!5pcJXxII71YfC%5<K8)}16{8@et#Hc-~!)xJg`
z*wYZGFhUwnXbPOv_&`~L8lr{-^{A)IagOtq%`y!1&F=dyl$$|Fb$lT$-4lO@vs_8~
zHcl2tldFW`dZOS-@;q|8@>8Vtdx6Iu*|fOqv8f5axmlxvpZIbg(|q^S8wvO6U*Usy
ziCmngxBCAG&F>Xiz0d0sar#5MZi=2RP-Q*M!r_ngVh>LaIi%94J^#Y|XyA1f*iGw#
z%v;<RTegHxBM!(@xc5d=US3+yrzHcgC@M<H6AR3h{yv>h=!W}`JN-etTN-R^ZOctP
zsW)4Zt0a}hmKnSfc$jKlYreJLqNFs+eKuiN$E9|mvsN)CL7yd}cc#L&?Ex8|HgaHD
zHJE006!&ztjhttFNX_=Xi<3CsJHUNK>O59xCi>Uq?~x^hgpp|UiKmR7l_1-dcNd)Q
zOrfxk?HPPc!a+Q_ENM$49y*4hzW|ai(SrGC%UUqGh(F1X-wMKr@WHN>X|{HZa3ly{
zpZ}n1y@n=2fwEz9&^o=hfL)LsdBwFQQeoEiG2(ln6`nM{py@pSh+J5>=BmV4)b$pe
z+Jf}ko*v5hl&YzUuk~$>4&2MXuH9Tum^sX9lFd+a99&dGG1}0zB{LoyFx%_+<Uqg-
z7m){nyx@()$jPZrslKq!@dpU>l?rt@N6q#@=@N-Bppl?~pW<rP(DaUsk}EII6lLs;
zAUo=t0F(XC-X4Ixr1s0UvzFuhh&}7*U>&e5V|9KYo_>9SBk^(&Uq9OKp#D@4GVZ2R
zm}J^ut;ve0)r<wf6$Jh3lYVO&GtO}-&B*skkE^b}Ee0AtZ$m&9{_M++R>hHI-?l64
z{$#%2U5Vt{zN?}j(7Pnr0LrV=L)&fL1>VS4UJ7hnKn4}r?CzC;7TQzGdfL`om^s;4
zztGfxm33G|h3s!gBpSOlj0@dA3S1;5K!2&Ji7ngfRx;b?)f@*0XU}@cMqWNZj;bh_
z-<RTDrqcZlf8wgkaKe^Oq!o`y@8!KgRW*Uh1Fx&mvh?L`f8@YEvULX}k?Gay9oK#M
z{#DKOViK(DBl>+?TWshL=*5vs3xFv*=DMm^Y;k<KDiDM`5lx>m;X%`a>yXc8UBE>!
z$f3=Syd_Qn3MpM<?Kp8ZFSo1_f4t7ljfa=!rYL_qiIJ(QDRjbkx&7gL2~dbr1%+OZ
ze&_B#fBH`!8TJN%->_LWmrw|@!>stWz$5oq*c6=-%E$9Qf$<a-JyIhJ>^F~(t{8HO
zF$@&1Uy&O%F@MIW?n}s?-I-%1K@|@Uy%7vOtpXgTcq95+DT2H`;to54pVC%;6*_A-
z3VEzRv~MNwax;U604K3V@rIrP!|>Er7c#-Yg^+BNgPoqHE6m0;^CmA7lA^)rb7ZQi
z*zU(rzIZXjqT7;uFWTZC!TQ&?<Mf8<<Wf{RZoA+z^9zH>_JtX_%(8b1M9v&hNXFb6
z<+>mILXC|FIgT#(X*d>thObF?q-2U;i&U@awPl7*_{pt@?z@H~yO7Ftyt9VB_+pN(
ze^Yhss8aU#D59j0b%M1A1#3sG{%UiUXxReW3XQ|zjs&oBVu#EN3M*M6Qn1x9M~%@N
ze?g*#6)Y{X3sA^s>>#E+gZK>S#5O}Vt;=&L8zBulCw>S2#S|ieOG_$SUU!qnSFo6L
zg|)Ser=9a~{8Lr9iHgGr%LlU)?7$VRY2_tfqEp{|FUpZ-!a2Z#t(x$q&KqiZRsMUe
zZ?4#(lEY`U0JcPm&sQx$HI#KlSHXF<XKnaxn%m-{3qLos_-U{bei$ugB9t2W-UEu6
z6<}o}^6-I}7!?lm(0*4!ZYUuwEG|17OOCtbhvEq?uW4gK1b6SDlL4hD5n5<vSQthb
z+(9>(=5cWK4+yMB3q;n3sDXh6U<T3IjWszKgv1KZ;vzMSF*~@iRd~QKNcBI4nOH$s
zZlb|xJek#M9U1jbdPau()21uuVWN`Lpk?MU^)ZiB8D~GM`#EU)-r{nNm(H!vyCFc`
zeydzc(E*kr(cljCD?(j;XocHmFfJM4zH%6>Mq_=|ZCn@G7B}Il71UJDMMm^O!Dpwi
zpd=egr6Og>Sa<$*5@xMiTYmiSq*}44feF`l`3f%^EV17o1@`7Gb_pY3k~S4Mp|FqV
zo9VWb#+LL4^szrFDT*m`msRQl>=Svx+7MSEP>5KBN~_AoSJi4?0%jZ~5hbx78VY^U
z`B68FdahZdhQZ9tikzPq`n#>q0s@7?^iz&Ce<K&*&~SR3T=@<4BmpV9K4}Uq%cg}A
zdqpWadx=VNf+7v25un;Rg$dYyRiHs;E{h&@r(^Ga?|=6eR&3SCFIdryE;vE2nzv`4
z`y$*-2u1!Piv!=|&_&ra23E%K$>|L;-O{(GCi)09{1#+UqZTrBguU`<OIO|g44}$b
zcAUF6m@APX+K&~Mz|{<pk_za`o=^7U_iO&gd}4J=p)-8g&KXL|6q)ONOL+Srxk*ga
z<FK-CX8znax3(N+p=Fvwaiz?p@e3591hwNpuBEIlhxN28n$MoByihZN5_fP&;Y3ev
z#urySMNnfRl>r(8KHIkr{RTLohUVtG*RyKSEY(oAZ!4QyS&=d`GwT#zEIFY&g}@IN
zW(gI16Uw`zSXdK$Ps?Tn3^adnjs^{4fuJMu2T(@TcDEuOp1R+JSH<ikS9cd*ruoJ(
zSdtgW@xu$lI}c^j9<ZgT-gVP=PalmhF<ZVG(eYk^J!W!1loU8F?;+e^*6MOW^Vm~L
z+{QB4<AtFP_tuIlp~1~{`H;w7(HnfQHt$Kuk{>lBjBwo}!Ie<d5edPOtk#mSur(7G
zkCvX_uS~A{gXCZ)r2+-}n+H=?{JCPv@Mc|G`%v?D$!u-i<YW14!a#9!Xdd><`^)gJ
zk;h_Bu6SMJbfUOaskwadk)Dpj_q+r8cPfi}quBfya4cf<+*i39Qg-`h?sn;}h>#Xk
z(K6z59X_L?DgN1j-QZQ5nEXwEyBOG)*m6wwmLc5YT;D%3yvushBWuI*e6jrrot>)q
zyZ5OpG<AWk$?l3@cfUSKljG??xb=J^K~mBVbWF~_pkC!?XO}`2yZB!Pm<^lwVi|z>
zEh&tz4-UL^SKe*fQJMSu_Cc06m%2DXFo*W+mL+c#vjNhIh7PO&V=)INx})kZqY!@t
z(jEHhfvgPpe5iB`%_bEWeph6TwT{3@pzJUg;gjPdB*Ax|#Bp9I3icMYWY%GZ$l-nK
zknhaDGwfiz^LFgVtequIvhmH>k7_ePObP?=EYMWm1ynJ@y4@q^{rIAWDvi;?ZC&6o
z0V*@9b*a-*fbvM8*A`DeHdy%d2N@OwFYhsl!KhqxNadGz@=keD<m-hoX>xSTCF}8?
z7?^I)v{H4)boaK}fO`qSBr9fi`0z=zitdm~`6X1=DJ%Mp9FgS`u+h|7&Qj8k)!SxY
z#g}_us@F++32r^##9c75&XNAO-tfi@1*m!QsBL0cbeOL-R#KTXFc|7`3v)|b$Yxls
zQIX!WCt@lXaWRjHOqb6P${?(v&~}Y)E=KK_clu3}Xu<4xS16QDb;`@#5vWsz*)OS^
zsc+)a#)gcB1`fc$m2d_QwpP^x5X($cFP5EJEcy^p9m`>jQ>Ik1qJskLtHdm+&iUOf
z34gj9G{W<R%Z`t~&hzH60r%3RF8$axwUfz`>7^<WdDA$T=xIEfGU}_lT5^0nbgii^
zjyApez&G{zB0|Zmy8fgthH$kb6Dln3=Og)@$Xm?#u%$`)J>3<jZIVCjmntCW#w6@(
zFlq+1POBDZ#3tvQWf8EBxa%RIJDTkX(+8uJdJpMQrN=qc`0*N4*Dt?~rZ~~W4Gf-z
zg1wu8`Gbd<>)K|V8DKi$QonJWS&Ax|R&!{r@zEhldp!XwSWr|HJOYC@JjA`7w>H4W
z*Tc+^X$49z?4#!~#k$CQmf|4GhnfQq3dUC$Ka6u#FMWaU0nSGl>+Vk1iHhjh?cBxf
z2uPaJ(}19N2x3Uf37P|bVo4>@c&@%J>;i1LG0N67gS%)cVz67+J7(O*H*@>r7R~1Q
z0o#a-f*5PJv`S=HNtp5!pv{>Ga%dOJc}G<I@d1ny0ju_LCSCT~y^J((aL@>ba*C6#
z-s%H8Po5H=oY?}przGf!`^|3qJ`t$USKgSxSakF*=t~(Gm%33FDgtEd%?q8TKKsZ-
z5-Fl5))6)H-4saC@d(pnSlRb1HaAEor|-B1Nj5(5aq5@(<Cnpyl@QE|$XeYUM#7sl
zu^NJe%(K;o!!#s2&~8DtC7Kia@eS^nm=D^>IDBpHhZgBN+5)$`Aqh=iKDmK<!X%%v
zxD2$Y;)*EHtDIP6<}MQP5{v<XcV%25>!g1@2iTAAJX$~Ve1Ga}_H<vh#cxDP{3<-P
zV^W`$b$M}F=<B6ZJl?Bj=}C*9IKK0h{7W2CbF##3_F}+j=;H}}`DeoB2n7+q8Bgea
z13#%lzs>Jti^r{ySK;@!x|`Ja%DZM<OIF3Jnb89lz)e+)iVfOJG+Q_{vJRp`<abNT
z5CAJ_0zyBu9ttH@R^Zp{;r_;k+-kvbv5NguDhlE+&>Kk6O7;p`54=&ZZ=1hLF8noU
z)NVC=kHe0kUKrebf)W)gItSPaZu=`2rUdMITJXm(q#vdk!uke?a)|XO%(H@s3SvlS
zvT?1^C<@=_vtcs5?o=>5u9<~JeqNrx0Fi*EChqsKk(SNx`WAayTACXVL3~026*YB!
zLqnqX(`E{%<6i|WEgn4b7_$x?dRPpa<9hd+(CZ~l+u5}>DT~vZs-HFM^T??vtF|rV
zzkoi0oU6MfAAgE}GtWVhENuI)N%Ir)Ut2L5WfS)5r(j?!v+HN`HDx2hyZK=dAPs%p
zt2z;AcBlxV^05Z`{+$uj4zkjU$)kx}klAQ;FFin0K<HC<8-O8H(9>vGR6%v%c=l!I
zn@EJ-cfT)|w&Ttc3DRO|<;$F-7jk0!32qQ5I*0@dA*c`xFg`^q=5yfBAn;0DJklOW
zt)Qpy{_<AX1qimcvHc$xAnKV|4TQeyh0#WsV>k3Sw|0EhWd0igbWPw;41dw_hAxm^
z-m}G_s}uLAT_zV61GjbEFjO8M#2A#Pnz+FXdho^e5zsf;d0o@LJ$ODgE6T2V8rT1d
zu?A7o<hg>KjjHJK_3Qi9172v(pI<pSX`4p#sB<UyXd-9~FyVlF1M!bF-t{uqWYqE1
zt3R0^C8n6`tI_@8etzIc;J8Fkib5d@ASb-5OgY<x%=Kiet5)%wNDTA@u;X4PTRBe?
z=Mx=&wf$9#5_@GSto?pM-U!`0@{vwV;swsAL)F5#sY$&^elrzeO6Er+=(|Q7r+LK%
z1qJKR+O_B`EG$59Cg?SrCf>Avq76mkk0(+oVexXE=KaOTI9skZ-E4E7tZFtKMo6a9
z9(%soUl**9OJ&fm8a;lfT7wQKXRWwKyCi;CQ0hQjO5YPgVXB`UlPddk;EVo$060O%
zzLBGF^dI1Lk6GRMBv8F-HMKwfK=LUQh#N4#H<Rg`R8*9tWa9e`AYu4O;to9mH8F|$
z?OUl{{xkLKenTtS2OgJB<D>wWo0{*IkpIx5Y+bdU>+XAj@nc5tj~g!{#%@Ol;d6U7
z(oYD1P=xPI#-kynhX9rrX<{dgY*^#F5*JUO+i#-V4Og@Mp{LpV-iJs{Ycy;7k{2Oi
zM`1DF{<4Z=jyQzg-E#d%RLhTtvtT>eV-V@-fMnO+ee5f(whDa2?e?J4ZsB#UIxEYG
zO-;i*ksXoFNht>b3Hw_Q03pP2=$55mcu7iG+@AUzNDx9i2rNXmaz*N=)(dzZm?wlt
zX>R3~KnM{Fd;m-%d@>gfh5;W7A>I%|M6^9iQBe`&#*O3l+iwqx%!LC@)A;t=Z#nL`
z<9OzoKi>O_={}t`YZe}lr&(93neE!?a&2GO0Gv)|ce~x*ZcV_2l$6Z-nuLM<i0{|W
zs7I$)>Q`f8*!SsIc&aMMKI=@dnXb{`kmzJ-_xt}vDhl2HbeY~lEBMU5?d&nwvU0I!
z<sdZ=o~lafb`{{<uog8Y4q045#m660_WM?1?zx*O?=9q_i%($Esbf)u+Q^a@YG*+R
zJv&!w5>3+#A}9>2o>^Wb0{3>=sC}sbgi6Lq<B0Fyo6UDVNW=CWbie&3l(=}qh^4io
z16GeMd~5nloWjkelO~`V8Bl@cf5VXkLX#BDn?vSF6A+3(YLeg9Zlb|x7TshEDeyFI
z94!Uz+8W|>(>o-Q^Ft20d#h%DD+2t(VtF<2bbBgtAVCOm3-CY`r#KP#KnQVW1YNK%
z0T)L|Rxn%x6a#lhLYUL(WYVNb{P4pMkr3Aenx=8b9e0qElf$K#UK*7-Iw~3(8p0-N
zLI}S4>MLq%Ye`H@YzOT+0fvQ6Rn<dihlB_)k|JtqsagCZN#n<Zsv>}f$_ng>DJUL|
z;#srNvNP#&ylw;touMkI-LQe`#ov;2(up{F_i4tZ=&R4BQ!?Fpr>3NlFmMP+2fH78
zhQhn<r*h>w;_kei=RWzGD=!$wIj0=2_xkiBV&Ov|6vas4mS`R~(&NUPI1#DaZ(})z
zg)+W(Uk1PaJnoWG*57&$?vhd?St2lLBf6wUu(!CBX|v`txK}O{#*ESxZsnv6yLO>f
zR1tf~AY;EJKP_A1moZ|@Y8(?G!0piqo!3?GcA}^@qa;p;Mhq}`e=8w`$O9%=40l@R
zEFgsF4LsfsQ=Ja{DH4K&5GSG&u4v=$LWoh35cK@>&$DF7l1PXOiJNY^iK3#SsL#_8
zV7Bqvym>RLR;>zqyUm+7vtq@HC~DS;#;^ehAr5WNNh?;YK&vPx{e<JaG@`z|l$7qh
zs9C=rEq^n~7oLaG46`?5sP@d5h1TFA_mYdlcrXX1J(%3gRa8X5Yp+r>Yc6(=M6<`^
zxcNq&{dy@APd$#2X?CiA{E39YgHXGqflyk<!Gu5v6(JPe{ps?6s(}({iPX84G~({s
z0IAO@0efm1eII{-oiDyY{w;UX|H+4Obj?MIM)O7i5>TYZE7LxwqOP9DuQ;Drv+G9-
z_ffabX%x+xLH4Bc5H=e~y%fqXE7t0B`MFe$EdX$N0!z6yl$R2d)^5h>HbK+0Lls4t
zx1ZIpvd-1e89NsW_oBy@g@Odu0r}_&1r>k`JtQU-=!PCk8QYAUM}!a`NGY3Eer*}1
zgt3krJ<(wsKAi=+&M81&boXi~x2l33Q+R3{bfvDYj>jKwo-@bA#W7;U2y%0CNli^9
zB_)N1h6XAtE2*fcVEy{_tY05GXQRBloF|@mf~TK8_zluR2s|DSpMU;29>e?5oRv;W
z-5d4UXP<HGvB!Fsb=IVdE^Gjbq6~_*uiJD`3Qbb=<B!Dk>w~RpSFg287$>^ytFMqA
zH<_oMZe%0+9>$F!0a;#7&DUQOcf=9I4;q9SR^L2DFmhNmDEoF1yPtjrb!Q<qn~K0j
z>@|Pl-7Q5NG4e>xy67|n0<El^s^5RZU0y}fu%Xy{^+X88P*I`@Kq@HKDhD1ARY52M
z&FukaJ*wS~R2A=VTVl%^@2aVe;)tWi<iDW4`#(1P_3!k4>OqnQ4)CrJ4GW!_=&+?=
z4<F3<n!^t5$C)RN!6aHO*+UJT<kqZRgN%tIre8nrEV~O!$lq4r?Ud2j`wZar5;_S|
zqPd)i*f{)2Uq^+aD1-Lr{6dH$fYU67Uy@S(F-#UMgb*SXxFp0X8_)^Oyyh&NLI?*i
z2DlCDbI}8s1bozTZN2jx2^<|_l_kJqz#=JSbu+S((XG1oV4d7^vJm1BDP{8!_ZFBw
zeL5v2C6?popMO5L+;R&?9d#5jF`+xc?A^PUS+iy_Wy%z`ZQB-foj2cnlZPI9h=ha=
z5!xD6nB*)RGUtyubLQ~qqmSZnI1b2u;|CfPK#@`oXm-su!fPvOXrN}<uQ<l!Arzeq
z)mD~~&?N)v8>m_ME%yGsNg6!JTglDpK`RiHeD@ucm5_PPIS8A**=MqDwb4L=yS$Rk
zk3GrGJMKp8D%2ASq>AT+u`G$tra`fD^;Hv5gn|$PwM!Z)$DKghNheaXdoN|5eop;{
z^+=BkDFuoqP>^97ob;7ELLlrmlo&g%h6dcVH6S&8znj!j6MmnT!>p$#Y<9X{c?p?U
zUC7qS_fq-$YU3i6A&+;fAwhF{c=)yVscmTBk-uGoU9}mh3E|IWmb;O)HSC=6Io&S0
z&=)i)B};!>OLc=2YkDZqq|YGetSl!cHO;VscKqO7Ddj-+FYCPDV*VOBW9LHQ46GAc
z<^o4ZDc=vrT`i@omr~A`Ql1Xn8DiD8#=Tt-V(j0*SSjTf&5=0(DP^UR&3ZKOw#8~@
zCT*+n;fEgv7c2y?zWOR3ee@B>9(!!FWDX!ZJDWfM`Op0J+ix6q+;KtIDK9T)?%cU;
zQTc$umMvRYwQ5z1$ef}mUN^1jw(Uf8p`h0d>Kgs2^nm!ff&v<L?j&u@aYzZ85V&e;
zQ5|vAuFuC+u#?mY6TohCTrgszno0val8UdthO|`D#~fpfgr*1r-6~hLW)1m&xe?dL
zA7hgqgd$J`)c5E?^*LuVW6?_Py7e+tp$5_@0ZM!VX~*P|cE)Kq%PT3KF_Zc=zoWTa
zNU0}%_-Ppus=1Fs!4?;X8WW4NvJy{yEmAZtajW^c5&dJ|!n;uX-Ks)S>3Yr?<lKBU
z+wOdX%9XzvF?A;Gi!8qU?kB!j_yZT5IgY$xg99S=LKSsx0!^~}<LPug<4lB~OI}L&
zX7N%`jZjdFL*l=SnzIgDd?I=w=mX+r(~EYBb!`kG#87mrPcXdPX`KrRD@$4-aIuuK
zwlzYflurSlTMSPyGE`fO(=C?&2;3&6)GWp=VYq<*0$BMo&uFPk6ScLqEL^xS=-Aa)
zU(NN`U*9UZ$;rt~pFZ6(uz&XK*{za$pke0BnU=+btzZ@zo;h=7i!FPdXdM*vH^;1K
zP9lvzNT^u09DA28IQsPVx)sgBvgJQ4MtY#j2`3ot*3dQ6w9;a#mMtgYs65mz>5c2f
zHkn_d*Q?w2#TRV7@fMW)^?)>{(rB<d$hhS%Jp0y{+<xQvq$b5Rk>w?KLLg#dNgg$d
z^fOM!Ra-;Rr=L*2b~RFS>%BpxKDTE4X-ry73zeDmDuh7TY{Vp|Al+{2OZMs3KO~yh
zUg=NrT24aDd;#5?tB^DC6mlnD!?wF0qH@*mUTWpDj<y}8<rUoh-0Sqn$>6T*E=LhQ
z658A)ZrTma-%t6^KcU9NlQ8VCfVFBH>iPcX-|-Q8i<bbQyR==kRoD}fz2~6gb#M)e
zuIR}T`vaZUxmLg^$#Js9@S9S~iZ)L1l*RHy;Dpu)6hia_hFT2&yESCagp|^WUScH}
zjub+)s(SDI`SYo-uMawQ>#eu8Mx+<Hxw%|^`Q<^!=FFLc$8&JESoU~4%$PBw)!WaW
zJsYRf89|LYVFiVDcu07k0-8q6vZcfwHUtqD4}hzomY9S@q}xU94@*#ccEjGMkFi0B
z$GwDoi@ztf!AZ*TV;j}$Xga?DsI6nizh9*2;m5I88nL@V(aY4K5_i^_e6g*ZVMh+(
z=pzO;vGp~X!7$m0SO-a?^2j`A63&V$_I>z2oZGkPra%dbG-*aSq(LVbwuB@cNnNNb
zF2Yq)sh4#zWQazz%NPF@<{VPjMiUZ*LiWig)9u#3viY8WQoVkIF;`?PbskOQ{{J_X
z;)-gf+<rYdUD8@borDDE_MMb2Sxok%bG*dN#ILK?vAeh=WJa_^5~J$HgFSWCsBs5e
zI(;mV-QJT5AqD~GTMWM<rQFw^`W#>w4WwI)z1LRZQpzR3T8rV<TD+{bs{~sAWcT5H
zAQ<}+z{J)FoHJ)m(B&gXj%3J?wwKSl_S$QME-xx7V(HSQZI<5=p&KP6+qZ9L<;s<<
zcA2hOvxW^DHbhvvP6UI39zYvq=#A=SZ5?%M){rpbNZlF*xSUQ@M?CK0A{z3y5SKR!
zB_@_o_8~L7wfeh7xZ>kT9x<|64O}H<<llQgweS5Go7;)*I~88Ht}C}YvFDz}{6#Cc
z;i^+nQNT-}8W;EQeHTJ{&0ZolmgHkblX3QW)NI&5(MKQRF5IoRq4MJyp)6klJ;qB4
zz-}WxGY75SMctlVNX=uWa#$UeupxzCB|Ra?IQb;HUi~LF-*F$cTel&k@FPd9wO&a+
zm@$`6=l{et=by^SV~z@NwT7*?1W!c;JEpx)udA;`#Ks1m4=I`X^$$o@@w&wu>uk>A
zh;!)Hxx~~1HKt>K^dGvH;8}__SX+oXp~I?FexZ~yKSFaXuvmUlE9F^j_mh;ep_MYd
zkWy9wzqD}fkoe}CZ-OpA-Y9+5Rtz6LoGx90XTw{~eK5c*CR|@%-zJx-+1s@f(S@RE
zTDSI?NFb=+z7?&inuK9~ea1Z+LRBeWz5>nTCh6EQIynj<OQHwH*RA>;b<jX!(yd;A
z^*0RXu0l3Uz8(49VuX-LA&g?bMq+}rlXm5$yfO1fF23{>l9Lkf`Ffj4&lY~9&v2Wn
zadD(gIEB;`Po!eW5-R7;!BbnUzwrQq&?b5__tTUY+a(=)Y8q8rH{q$PM#Cx|AzZ7P
zo+&D%pQBMx$UXNgvMxA>t#|wbXYoEi;@R>GSx8vECZ9)Me3v7K_Tld9|I|7%heX!Z
zvgPe}=yB1-ddV1kMVTewbb9z=;Sw)F4<=hkKuTh53R3f+I-0uE4+MBTo_2_L^+5>H
z7wZhoPU~Fru<DT=$zo|%vIYtvTGLF^O6N#c>s@TIyu~f<N=r-Gym@obu}HP)5<)O$
z%$T5K%a<>Y$Q&IFQc64?51)PZS)0h5l+v3a+llBxQ4}R7+LM@p=vAxM;K<G)u6s8<
zC|!bLxWHV?mqB6@$%h`&yeyz*<0h!6B;lyMMh|jxJ)O+eZrj4fn{Pv`Sz|2MWAn_~
zYpRsg_a8v*(8JiWb2n$4I4Z!^+FXi-kipjw2x2m_NI&;HZ2kJOciKnPE?;KE*M-ii
zRQzW(E&_X80!h94QoCy>4SV;5(8;Gk^DbB!xT>Pk?XrtVJpLFq-|-Nx%E~5MhU?fZ
zDlOy2e>_EOQVcIY^mpQ7Vw*XZ!L^eOF1Eh;E;-}I5!<V0BTK0ODOs>&8GFmB(2YpJ
zkE<Dn7aJQxY+`bY^yo-XRkgicu6GkW5u=vQ*tu|EMNoc=;2c(rLIH-iM34|72ff~1
zFsy8etY+}5#qz#JnU}D!eED)q!ZCb!i*`RGjy&?npv!;#)oNEF3%A>i+wJzcLfhk0
zP*4!i)3!AbLa=u2TJrPr+f{c>)9`pa-nM0b1lTl9%WfCuEG6};R}(v+KO#03DL`{M
zQEYaE$4$-36~y%JjV&W9pwMkF2nqFTe@D1Hqz*rvpleI0-mr<Sx7~%ka2KEi`aKP}
zghI}h7xT)eUvkAIrx0TlyA7mV`v1bNm8R%xD^dU|2?Gxyp>Kc67cHXl-S^0xI1x1~
zyP1v_e!F6<CsaF0{RUFIa|cx$*OSzLASkM#p%qCC&!y?rMN#Q>)Aek4_zAW?_!#{k
zdl1FX#pw^L8SComxaR)HD6OvK-De)6cWzFQ?ZeQ^>1O+X{zK+5c_bcwn0G6m-B5y*
zy!Y9cMnUP|=g5DVBQ{2ln>;X*Aeunaw5;|j&S?b4Utux)ij;D1yC`yiLI}|f>x8q7
z5u9o}dV$|y#QQMd`xeLw#ejmF!CD6^Spq#F!^(o!1S?kt=b3i9ot`~=wp=p{^zR?M
z-T96kJ18$NCpEQ2JTxqr<vuh`qq4FRx7*!rmM)LS!?bDBsH>}MlN^&wz4zXGOrAU$
zMNzyhFrx`6C90~DmX`L1DVt&wLS(edOqzykZ9a*U&cvi29+#6CM*^O`g}8U^A?bp#
zpoY<>SxV~iH{w<lQicSv=t)V<#w~1_{CDj83IRpWOwa?y@aAx8P~W`|3?0Iz89#IH
zNS!nUBPuQ7_d1a?S(x#cEr!(bC*dmENAZ{Qi0jddq%orrX14@m)?h@#EHx~J1QZgw
z_rh6PLh-5<bUE}el$aR5BnHul+=b~jHJFkFglc2JJ-3p7>%DA${uO%P{x`2xvpFi{
zY-r%NCtqU4y7fH!kK4&R%xps9uT}62R9|X`L^inC@y<J>A8`~(BM!$uqR(G}?K=xu
zv}_d!RnLp@76ELkr9coD8yhx}?x3)WFNg;2B6z}7J@Eh9QI7)^p<1;?jL=CbokEBm
z=v^5aqhF9^p<v}A(TFUe+eqO1nLql|0zzli)uC_SzSwM$tMX$-=s<6CHDW}Iy~WMK
zBq5cRm5dxYlFG_TFTraMLR4MdY?;BZplKRUKm9aQr%sKCyihPpqzEBMN=jnk!i5nf
z_8l2Epk4YN$*KzKcNdZ|Xi#8=fe=(}$j9b#lW>@sjbwG5qt|)UJk)R6L`-%rB0kZK
zjK_4r)ot6!roaE4n7w;Yga}&bS0eBT$e45*pD$R-DHD#t5o?E#d6I2_p+ih&7Ficx
zO4TnvQTX;dWSn_Awwzo9ni1D$y|I@{2|-MkbkbrSlq~&))ZxRi#U~nxCXLfJ{J%F(
z>dcZRiX)zWkKV`HYyU>kC$q>p|EvIN=!=K*$@F+M9((OQX3k&C{Wo64r18f$*4$DD
zU8Y1x@YL0_?d`Y8JZdyaBM<j(9hmE&6Z8*fFQB%;g-}$z=x~sP2vbcVAR#U`U_TEu
z2qD^q&>60`t1O0Jl2Yz&4?PZWgzD_EDZ+DYAUL73W+B00rE`IlvTddsKS(KSgb;fO
zUIVp7B-Dz~4HyvFdietf4zwJL7NIk(TZxH@Tyn`JJp1gk)Ya8_mmerrCn+Uvx0}+^
z(x}g69utSd!If8DNmf==tFm<{DCj)`qmG&_+t8$&gaP`ss^`5bNC|c8@{u+>@dF3x
z5ARUiLrAomT4YfX_I`c5EyK)>oP|Z?-*qSUox2c<F3Ud{U+5xGG(mk_A{i&2$owV0
zaoS0FP3TX=FBKv5gbbUF<YUH=dH#8neleG-#or@69&fjXaAnyN6393zkJ4o;aF!I)
zxY<?H(j#H)mGu1+lb*@o=bm8ao9|Hh%d&uN&H8XAG)?2*?@njxr?a`^$_u#ef^)o<
z#c*td!XRkwswy^5eUqGVCpERqH9}2oE&rXlK%c`j!214bBwwE+B_Xa^T{<}0tMTle
z=tajHqce6cd{~jGy%DHC*BDk7zhERU6s&ZPGyw&jga$*42%S|zM5GB_mo8mMw@AR-
zvSmv|=Q80<v{P07`Okmy@y8$2w{KtX@=i2`qA2v}(S!Hidygqori9fQ<Y0k<-b6XN
zkl@<B4W&yeN^&Y@_NlPxLF{tl2HdI1IJ$P#Ej3bvDD5F|S5~7{Rb%hb&1apH5@&f8
z`FG!oeajZVAG;r&@LKXDXc{rYhqJc0l74-=k=Z2~L2wJ143N`$cq^QsU#Rx9G_tPz
z6PnAzzIWckQ&#Fnl){MF69UC%Cwt5>RIXi5-KNcG(hUFi5P}tl7K@KlJ9_q{-#_nU
z^8-)Puzg2RiJGSI+VmMb^Pdm7`O<T_`N|88V$nV~t97{*9|v>oEn@2%Z`14CbBI4=
zpw~6(OAztjNaKrdeq`I;Vg#TG4Nd52@xD;WM*Arx$%*k%VCwCPlxR;NgvbJ{O6$KS
zr7VoL4hIP`EtW@v99a>%mL}0y=^V+Jzvg6yt+S?S6c-2Ym=<Y52i9)a!ouiY5<&=K
zVq%y$aUu&BE@aZANnUrds;Yl@a;GQ?ilSh%*_be40^fY|4QHQyHZd_Vf2iYQc8Z7!
z{12gXX94kjdKtm%0*~8;%^riIX}C9U!_l)37{T|t*UW;9NN|^zAzT`2SHEN|mz%Bs
zdIGU#wKs~)KUm8Dn~>lZDqT(*$J_-!GvS0hFPRJA3<0Dn3N{@fP*fYK<4z!T{5T5#
z`(NtU{BC6F2=uU(=4A-gM%I|o)D-Qba?KiJE0atGQmU5hi@9a$F-MVg!9+Iy;}Nvl
zx<E-D4^uyx&Eqe>$K=b;<Bn@C!=|X-`9sa=rL9s@o4<)2AAU-oD=sItM|Z<6=f~(o
z;Fd18Wa|5$f=wWWf~w)mIJUZWC8Q)D@Qt+FqZ;yUKzxwp^>kY2S^_Kg_^v3-WtIA#
zE`(@FAeoiU(QvU_$qd^^R#sLftU5Jx?Hc(CKUOmL?Q6g4JzcwY<%17C;Mr%NB{@0y
z50k+e3Yw-78ym~R4?oN&pL{~^-o5`YB+egZFW2ZoN?dyiv31M!ku0MQoTsW1SyF=9
zr#I3shRoVN2kw#*gpk-WjC+qJ+4k1kI2SBHK`~lO`Nv!tHa4Rc84{d{2_%mg&hl0H
zj2>|aKEe@<{6y!1(af572x5EnCij|as9g0s<uhj^T~2}q{{@2p6qT&JJY02klrR5P
zZ$jV1VcEKU(S)Ge<(Faak;9G`Uk1(NO$&eSy^neH#do>!igUU3>Pt}#!XD^m4L+wp
zB0Q4PAC^$E<R|)GdnL9mX~q@`vI&vHIDe7_KQ86h&0A5Via-z_ucE5@K1-?i_SVn5
zC#4`IIVp(H0}4@Pq(X=;zzr6|uXM)FwFp+^C>nu8Rx%Z!C5dlVF8z|0sS_5;ES85&
z=t@dTEH7p&_h?&SRb03|37x8{#Ky+*m%sdl1q&8%*kOlx$;%&+#0Vi6G-wd>=g;TP
zJMSbmHWoFU<jR8(3VJ)Ch$A4HOG;e33W>?>idi&RC<+Kc!@eRswRObz?TZNZ01pCB
zMI{~t_LLNmP%{5(D&Kww8`38us2sf)swTiTs4rWqYU$N8m$+EFG3rk}qW{(-qyQoH
zGA%-&I^xN^@Ivfane2J}P245LVZMDCfr5%4bL2>b+fCWhpV6e&XrEefu?ypv+U)eb
z_b#eeE~j|bESzo^PrvdWPrdyqcV2ffw_kG^HpT3W>JMr8Kds)$&W}FDQ&mUzD=$Mh
zVvRFo##s8TZ4f!o)9IgkbsE@oi>@l*-kYyt#3B7c+B9u;h1BHM)LHFNh|2Zqhg*PP
z2dR$Et`IGSRXB?kk!cG;k;Jg#E`<;=Sij&$LuV_QVaJJ<mzP`K)Y8(@TC7lWu##C)
z5)Hm<4KuE8*sx(NSg?R=ue}zB!|{i4FPbrgHk*x0FTIpSix!cWmj~$XS0M!LRo<y1
zK>_T|uGLn!H9V!|C^_AHRx=~b(6hG?&E>(~HJ7Gtx?qra>g&;LD)xj#>bGxa*VF&T
z?)Dhl`(n5B)%{6<?&9-EiQ|Y7EM2vMqmCTfXjU_PKKjrTH2jMQgCZ0Xk3O1=NoTQd
z+BE9cuJMbD3r8ZPjXDZdu~E8osZI#J+qZJBbjFDqpGfcf?<fD&*Lm=s$9Zkee4e}e
zdTzY*JXGT%YV5LZs-Ye~*-*HbZSTBK>d?c;I%NWiYIepjT)p1E@RB)AH#@T6hovlC
zzZFHZ0TPeUc;vaaS^E1HV_WH+%V#MRBqqcWYmZzRgZ+f4MfeIKQh>i%48JC&45y-X
zCw#CvSCOd&5x&fDUjjn?w%Su?n-C%_6HBP3?}>?#k4~_Xxj&oFCnqQK!V53(=9_Pl
znVEUeyI@Vinvs#g%P+so8*jWpMn;Br>|o#jP*4coqp{79ejly223c7}e2z}qz-YZC
zz`dssx64Ib*PO85EToiZNEAh|dCKEBib`}M;I+&JAuq;%g#=9~qz)U(udDMJewdl1
z-$CAL46+-W+m-CAuA=&vWmL_chvxBEW^e{>FF=9RBacMYBxS$+toM>7FxxlqTL^Lf
z0Ai7U001BWNkl<ZZeIbm-o5##>i~XU^fT{0a3>d>HVzfPkVlh)StUS7N%=3!D4aQ)
z9_OD+!jM7c1_AT<$HGYj)msY4f9Nq9+)kc*<6~Tk4Jo8i#92`1GF{Ai$wA+QhVc?f
zYEt6CV2Y3C4%|fWtoJ(L$xs^#AyR}8eS>`3D=W7n!s=z&6@|G%$qe@;Ak^DVd+Qt)
z33a(#mbbA;)}S+Cw+BC^`-{-oY&LH)zy%jvz&GD~!-*%J=(W_DR;q)+YG+#Igb?KA
z<uPyGJg&Ls8tis^z}UgCudRizp=p#L0C#mYo(3neIeM2diE(lyxQdH$A&AKeH?c|2
z?6gZk>ATZWR{rj7{U!Y4<$~^(KtsV9V<)~}AIi$A=$UJk)V2R$CUn>mlF0beWoU7B
zO5S=0t**uh0SuiiVP?^G$;(4?d8k;v0?k{FL^fGB6#RMGp2B_n-}QI1p}v+o#~(w^
z{BMv(JhBx2BwqS6NPS#NwCXx`fAAsB;xc+(b0zllG(U@Nu*r3coDdM#Et_r^oy+|1
zmhtoIt=<;`;e9#KS)hO1T!4F55oD$%88LnbG@^373Lz5E-L1h8j&<&8^un`^KX*n<
z$pa5oM5?0_x=`Ml_SQKp&ymYzEg~A_DmogWgFv!2XwV=&`|LCBzyE#`67+CeMNtkK
zt5s}lER!csX7=pa3>`WY;15sY{Gkw5DrL*ze|X(>wV+5uQi2~r(ARL6l_BC{5pi)r
zwhRu{;UGp;X;}CzHq-tjtlXqcBf9L_In+o6ha-mgxM+tYwE+SY#YXCh<B2<F6#L$M
z2iKlGNNJjtgJh1HrZQu<Ge;joLsbQptJmQ1(yhQ$h0vjk5~RlJ4O_VAhJTQm(S`S4
zdXkgwnM}#bm6R-8Vz{+^pTKL&=&8GY6WiZ?pY&r!lXcq3#+#$G&igQKK<56a4hJ!P
z`||YEkI+;FU)lqW%E1IVh{oP!rKh5ZaNPPG8d14k(Gwwpz1?(XwzeKtq{JPCxehi$
z7p{fQa_o;_p)(<+^jhN*5)!!YzWey>v(FefaNt4H_Nr&ko=lrIjmI8)jMUUrZ>(G=
zd{BVkIp1wWK+b4GJ!)(W#Kd@Gq)fOgDzGOcf*Rh%Cnc!KNl3dL6^}1ypKx7>5ZALO
zdrK?H%1XDWJlAm`jr~g;av167pUvKB)2YwTw~UC>`Hw+j1V|ys9eW&g1>33Lw%u#L
z^vd_&HkdqM&Y~Z=_~r*W>9{;ze)2xjQ&X{}Br^D(+u8E;i_{mF1uu#sCDQ4laOPYp
z*R7-Hm6sFSw+A56%sO?!cEay%(wmu#<ipwXS-EMOk^L$SBIi#go9gb5+{`X|j9ri^
z4<JOG3<x3OfZHvGr%EZKQD)<hjA%TCLdgvGHWx}phtxTg<w?^b|L$WoUo_@w2}~=M
ze?7MG<Hz&OH{Wo<1s8aiA4nl5CJ|K>g;P#Bg#`;1aORn3dfm(i(!R786d?Rp3#s90
zXh3>2T4KS~YHCs9;*CD7epF9_tg1pyi~}{idyYU-SFj6vVuIJ@27#6t-SaBFNf6SX
zT_T9-n#+zo`{>a<@^y+L4Iu<l(J$7RoNjcx{?8OGSV+xJOOcwU*R1od9fZ^fKvBs#
z<y4A)T#j>ZQImD7sF26w;iY#!<<7@n<e#@(!F{(}L7dH(5u7+|2$>VdQSkJONVg|Y
zpeCu?v6F&#KOp&#!Q@Oj3pL*1i*xW!C4Hn$>KAdz`t`VKY6D&hib~3v^71r{hOIs|
z{gn5&G&bHR0d&vF^4q%u7SS*TTu1O?uJtq%JJ*Sjup-~VYW%eop=5^pn+f%@&>?lU
zvX}vFZYm!PwY9Y^)?<G`N=a5$7H_}(HqSr*yw?SLAR#_Jo<|;egpWV|_zzoKt|d^=
zZ`#d50{g!Ik!5F{!=|U7!s8dO=0os<E$eC#@o@+>u(i9860NQtTYOwFUs+IiD#|H(
z?_F&14y0Xo+nKRbrUnL=3)D|~9fHK<WV5fhirmaBEKEdU>C!Hcek4VJl8{J`Yp<bZ
z%O=V{pM}Tm0jV{Lt81jc5TGjbJm)luzFCO7qB7_n`V+&-s%mb3?0G(%_bt=@{Rsaz
zaXhM`1{`Y?g<jWOhPz+~MPDs23~3UN)5YE|zoG2MU+8+-CBzTt8{+huy><;No0OC-
z{}pFxnO7&NNgjRv9f~TdjNP%e#+mfiw~FyagKjw)VgJrLFe3Vt5<<iRf43O!w9d5+
zR{p?f)cy?hv<}x@8fv|05V|(&9EwO;k*~JkKq7QzSr3Q9!FAVN$D&1x7(RS>hmAS&
zPvTG%g+mTGgt>F)a`)YLlaP?mne-3_6!hMV%|b{?^>3@Nmy}VobR~)=gIMW;KWJq`
z1JdEZXGJ$`DN;i^UC6k&@J^DH?0)}$#Oy0Z*i`IEiI#JEGdhjTP;Vg%0cv^%B^7ma
zNey;C=|E)8gy}{VdJR3r7DM($7a<&WN~Zr0&F$7ZOUNc0c_D;AIO6Df+Q}5on~&yl
z1-eOvZ@Yp$d%66!2dJ!b@!<;(F>FwOg9kN@wiEFQbi3n5DqeXV*PbG3H*8?f2k(>6
zy*phmI1ig6)~rxtv15kw*fhE+3jHs+g!mqM7Z4#O3w~O`v@gE}BN4*X#H&NV|7<vH
zf^kO;LNnwFsUgtl)h#Q?kvZT{6JdE*1Kolw@3hXf9af~Iy{%mxmC&_WXS>~Qc{%^l
z37tO_MPb;mVSM$~SKN5xjoy}69SLU1k(ihmF2DS8zWw%FjydL-P6F2gD1=#GNFwNV
z=>>GT;jbL}$b$$)H1>E|Z|ibHY)qhW#LPr=yAfg329rSDmTgph{4t<+DX`^qM;JBA
zM5A>Ze=~`LM8lq(NNs&B@$oVE{B|AvF|TVc3{omomDE#C!_mJVdtZGWX}09@`Y9V9
z5eY(~rlyjaH<G>c780<5VO{(2*Wa1+*Sk33xTAUH!CT2r?-C$K*8w924x|D}hYu&|
z#IfwW`EOLOUPsppFCt;+Apxwb=_M(B(@O%mqT+iCE3c~N-Y2Hwa^tgBnj4y`8-!7j
z0#px)3GtkC{ITe<diqT9F?O;u&C2ly^bMjaQi6mKcC52dU+=WewHcj+E;J%R=q$%N
z37rpmJ4s1ON(xUu{WKqZ@BukS)rpRTjEoFkdg&#mPMu0-X67HVIB;8_2q7xMSy2dx
z?UGLKtFIvS5WRMo3^!v?lL)^byw4aU5t>GGH6$ex;nvvm)?3(WYLG(UF0CYPa4+13
zw8o`#g9YhJ^b8PNTpXI)g-!1j<!9mP*rZO8ikJ6D^mCG&cQk3^kEh^`skqC^(9qK$
z&Fs*|dn5%odUYqEM-F8_F2(Hi>UO)C`oYKi{jq2HfA>x1mdnp0*4FeUWs*5Xw|}9y
zCAAwj;qKEDw`#}UB^?px2&f|$6cQSWiYQ;Zrjc}?-42h&{m)Hh<IX*Lc@o7FK<dn_
zR?`}rot{K;jE&-odhUDj4Uh_o)L%x@QWA}}Uk5zOBPi=~pm&hv4Rl)PA_1$H<@P2=
z;T8pMpkt8GdMRaC$$CjiNgcvM7mXsuk%XB<U?xmVnly<83l?zl$tMTYs%vj+n13S2
zs8OT%>Z`A~;)*K*YRh#Zd?-@NRy`m>m%3=WvhX+WHZmTC3hFnRLG`+|)P4IMieZgw
zC@aO0l8l<!g{JhasT_Y&Cv2)7Xov7O`5aIXLMLYef#Z-N<eq&td)|K+ci~>Iwaqkq
z1#D;hm3;V6JauK%Y}rgjRSh@a`xrm1TE|B(KgrmUhoktpH~k^eU6^PRt*o3Kvu06O
zT1K}s&ZPUjcd_&3*YQ+VH<~b*6sXM4r*ipHLtX#aN_~|X3l}qO#v+7J4KgQCZ3^9U
z(+ttl*jA_V6UL5Y$bjBxxB$gCM!HS0dsZ3=jt*R+CQL{vBhm-dsKkGd#qjG=%AJvz
z=MN66eo)bfP7g)6!fM+k^%PF4b=y<tuz%y}>FMoC=xS<$M?6^dc<qQVNmAdwefi{*
zPk88|hlr1l-!J6Jbg?$BE!T<GL6K5MzX_chqJFU=z)1P_U73y6^=Rz<?|-p1)B^&k
zfFdQ8J9d*W>L@&>N3qfN!D9RbfaXC}?Cj@E^ESiRw^k6-t2en9U&_8|AK)t3iRL{T
z!H>HT3K_?pz;8b+<*zr~$FLEH^Xj8_k&~GoG;zS^8I=f^hkf6F&&~x4$sT(gT}F>W
zsDk*z4k2aOPzv9l)+qb7$?PBmnPZNl$7NR{gigNv$^EA71>E(o*U%I@2u+WDM39l5
z#GQY=5(R-r_{H7<?|!y`AAZ}2kV;^pjDWtqdmJopd&Lowb1~3A$npl@i3rW{M~BtD
zHSFf|Er6B(OrvFeSld%)t7`IL|Hd;iGNjdYy?5{4*60)xRx-`jQwIY4OPn|y4(__^
zF2;`^zu)vZ>ZqfbGGz*BX=$D1NZJSmXj;c2!Vq>lQU+m9iU6Ap+^$9s|G@0j`fXdO
z`ThsYwpIc?cB^XnD!Pn63C-aM+{WzX+2lIcovWVu2JAK)PNOL0foGosq(n30>Ci7y
zdqyT*uf3kKnKNnFv_&sK9OTGK37Y2Si+SJj-?@wV%dw-m?wm8Q$CxpNe(M_+Oo0SV
zqh|9qwtx6BaXH=Sb?!OXyQKOY)`~*z^;c2y#eC{EZ^h(W5}G7ofZ;Ax1ZGKSspoxA
zT2;vn_dP>VWeq~n>+mU3K?+IHzA|om@L4n|^_aOv`=)pEl}KSyT7i&~!F{?0PiQ{y
z5Gg`;hsE*_I<0dNgO#PJdlcphme>MNCWGF5Ezu!#4p)H9W+OQ{Iq2A~U6Bs~wUU{Z
zmKO23q79SSx!rD-En9|Wl)-9mtX;bnr_<>z6nv1pc<&b|ilRjRHO3DT5RO=crZtUk
z69VjZM16xLtCAY*{p3?(8|v{-!w`U0T1;I@35jEm*NKa9+p_#k3cXOFv!;&3gg9#I
z+<wvyGzb*gsOpLUB{7lAtFNMB;R0$_tu{!VnSvk@c&MqV;;F}8VZ++3JT>(xjym;t
z_AS=q=}cFyuXwK>OzSGGVE61fRPWqHj|<KxX+VFU$J(#5vn?xw?ti_UUC+LPTWV+$
zw5n>BUwIww($Jn10n|I4+&$$5)@<7eip@(_%?xEJ6g--5S-_bSz5^o6pa20#jsAUl
z`v-5f{;=<7D2lQ#LZYlJI=dqzzY|bkvAmmz{8P({a9QbKg&i%F-xl;}p$@@?^aw6y
z=g!E!mv}rLcJB^;dGzerGoo{~2c*=K3l}e5%)Wj5_FDo%Nl6Jm{P4qmnfqXajdr+!
zs|g7REl6ye5U36Zo|0m{Ds*G(5E8ATlB%!2!D~D6ayo&Eq;%#?awbn^+Yd{y)#|ZZ
zrp?Feax>=AvtcFfsw&dclPNdqngtCuI{+b#j7}vkj_f~OPVs;Lhp4V0ZcH930!?ac
zT(_0CUVEQ&|L-i08gr!Xnv9Rbws{9N8+MQ|sHdJeY|3<N?ET><+%>giAAdYHGqFHG
zz!?)>hc;bKJ(c2F^Qip(CsL0)1`+R|-{c!mQ&NKKt#6Lo<L3TnUZ=9o$!RB!=F@pg
z@J4Hj#wV$1(ud-;NJ>2>&oB|j*cE#B$l;(k$<SO+)wbeNA;i(G*v{&ScA*ergvHp;
zQp)>VG1q~DO%}`J(8)}Th)XjVXt8{A3*}g?KTrrEWQ0sOLI{Q6E`P8-Zb8wnUq62T
z{r4uz^Yimtp_w(dZry6xYOG(sei5Cky`ZWpGiJ=d<B3YcsIc+rr=K!m!i3JO);7Wh
zM7=bDl9-4Zr1z*40wpO4XH^YC^BC&*yITcCUww_eqS9-f^9Iiw_Z8Qw-*N6OBIWFJ
zshajKNS_POSTW#VDJ5EYDcOU2v3^UT9=9b9jz_&PQePz?VjX1u=`zYbcn42)3GpYM
z%#7)uuzAyV?s@PwQqq#WB~Sn|jvK+w8FPs1nu#qj4m6GObsMSNu$hd!k;HY)vN%Q(
zAq_c3!iv^VPwwAtWY?4@Nj~Zb)K~{;qeeC%>|VXmTjxCU`ZWGC<7=-x-PJ1e_GkX?
zW1pecPYXy)h^4-v4yW5^%N0OMVmv*&<p$gAKt!|&T_58UDfA|ICN1dkdIvK?k*Og>
zh|n3<zC$gBH?~lmmCh;XF49Puy?SGvXwhOq7Yu9GtO>fjwFo7ze}5}(+ObhzU(dXG
z^Y+{PrlETF?Ag5V!VAR2gxhww6TzS~G&Ho-B;WE$;^I-zGb0=M$WYVMa97tN-JYf~
zR!%pib7!NP);r;~*BC@ZplTk9U!O|Gxs!1B>Vs%pV7IYnLx8)mklgGp?Al{`n-7A=
zJb*!Eh3PI;ZKPdt6(xnGym`}oq@?9=*8_KulIF9rfqqd*yGqW9$FlGHpJ~`r$nGz{
zLGx&IKW8HG7GzF%ZF17Eu<E5o*57m!NR7B*hf_Rbj`v&zg9J29^7hBGc;@ZT{G{q1
z%{bWsv4H;JlHO$+_up~_d50f@$E=YjBvQbT0ey+J+YXkq9QEt96A?o`*1fW=2v}>-
z4X~=ov#y1*)`l31q|Sp|sY4SC88Rg3^0Klr_Uwt=`_KCI>w_-0+wBY(Fd$;{v=^k5
zEL*mWf`a{bOEC?;g@uJIUc5M9StnW!B{nv;m<SdtZ43dZadDsu&^(Q}lfa&tiFCP;
z)m8XEwk1%rX(QSO<At$_%gvh&fVOcHrE})c{nk5hIqcv^2~78_|2D3|Lef)`DA`+d
zko|b{<L5C*k%otDTQ;+(xRP-v<#9}}95mEMYhzf}1R^dTXW3o~XU!w)gt4Rz8;sIi
zsE^c*rFtH-Kqw0R9-Tts;6dbGc@ZU_e2%BIJP4hW-i`k^`%C`u?E6LtqzGusEdst-
zS@?AQAAIf|zW;TD;WE|`QlJ1w96InI==cRd1(B^o*NG@XN?C(7p^9XEhQmUPEpMT$
z->|Ml*B(0811&PGjTm9&(I==YwzaU5HDt&T91cgs=4meoA(%OHCT@4A7QWGkS@d|u
zj2UPLdg+P%472rDR1+g?aq$qV1~za9Kw{6x#3sO5w9gyhuh%-0lzh7gn>S`5ka+oG
zw**2FRKD{TN}_|bE3QUURT>qz^NYBW)b7|pa)JY=N1?Xv;PLuMp_g;9Ne^glzMcOA
z>o;uYvYW4Gz@$mISN~4!$~9=xuSSRjG)eh}jTA0jN%m={64SpgLQydj5rWQz^!Chc
zWOT*Ry&H<%Mr?LAY2(MS_q`9%WZ>3jU6Ie`eaGJ)dz}VN1B_mM-u%>H2u2Nc|7>k-
zy*Cz9GwbL{jZwn}wVO*_dOty8M%xG>L;;~2sCS~{VVTA9(Gi^ESc~P=Kyb5?1*DV>
zz-o)(V_G7s8CdE3o0Kx#&K3X+9Xgcw_~0e&A{lJ|{rBGoT|RQ;$cW9+o@i)j;EON5
z*xzb+3Ek}3vn^v3I}sif^f#hN2In?~!-1NRh*lHWS6l)yx!KrbRO<E=`aS+6T7!$~
z@4iRDKd;pv0x5h>Kc&IR?)(2q`st_Qn0Th%no3GQ!CPWifTBVD-XfIRYLZiv*i%$`
z@OWIM9&1)yRL-nVzM^0MzDzuGJhm7cLRColzezaPuBB$>YH#n?`r-<TmaW2`oJ!ZR
zqlimMCj00-iho+FlOt(yc%)$ayYEr*-S@~w<}dI66hYR-=TW(MF%1Q}>%)|#X&Rr+
z`<Ayqn@fXR!^a%0{x#<fvHxb<2y+uDQH93H;X|zDMG(?(Jr{4>WP}hzb}lL9Rw75}
zD$vVAMCucGar;moVw-R~7_=BdhzudbYt4Kfh}Qgm_|9Va03k%AYP*fKSiU6Ojth-o
zrSs@WWak=1t*r=caMlz8s;V-2bZ~KzB6QogZ)f}V?Lo(mJMOrM%+X$uQtu43YSk*X
zY}vBk%Y>MvN_Ott$<n1u4>GS=`$7SbQf`RmBvMhZXJpb)QPw29P07kaiM8R{xyLZt
z=vfL4+qUA`z75lA<F}QcU?Z+hC?IUucI;r+Bmbi74cDWdI032X4oIZXYp+Q?uFB?i
zQl7t&LkDzY<?4+G4WH9A$&&AvuwwZNCZ0Wk9=*HzqTB?as-#_bE|nWMQ2Fa>O4qHW
zp{kVhk;6#HNk<X-@loRJ#P{w-<@!xXvj$xN4C_<96U}RN4enG`latB0<U;mNebZ}B
zcY8ei`^^uz{n3{>{iGATUUA{gW`)K@<3xbL%Q_y<)Xmg+=)m5jCM7gkD_W@Dxd+89
z_0A{}zA;7eI+aoml2Uew)CaiFVyb_K*zTf8?N1OS1J^b4`M)U6Z{iEQ1(gCIfD79y
zTnNz*YjQNZm13>7>xwmbYzfW;tXeC~YK_3tPd`2A@&yYPL?n*xop;^|x?ELNPCM<i
zHp_2MNGVYig_$#Fb|{OO$K&Di&p+SaY8=ooqm^13I2#uH-J-UPESzQKO=6ksS(y-@
zK<$<tNHb?bO3Ifn!{)a7u?bVGSH3_Z6$Q^vKT+`F3-r3{@2JO*$0JmIOc>TCsn>6-
zSewt`hxTRp+N~h9gM!o*7Zs76o5P7Gk3)^I`PF&R-IzjvP*vjk^kUbWui?nfA*FY3
z6ve;5i-IN;5_|L@CMCg-{{$Ai69{@=|7WsJ9M?FqG4Q*9l+#Y8VRJq;zpX_|ji;wh
z<MF8<Qtfhc|5LAmqIi=lkR}x~-L8I_txb4bXs4_q1bM@UG#>{TA;<<NrTH638{97{
z1u~_SQMpb#5j{vL_W?^ShOaZqceDjJ6A*P8fY$oczC*X*2E$)kDbpV}Su8IDT4ho@
z<BT(cE_b`#y!F;wtr6)(Lqh|vzy5mAvAn#z1FB1e38&M^tXVpDjzrv=|NX8l_q|!O
zW_2i%d7wc7P!uJy`Q1Sv(lZbhO}ok{i76=QS-7`v^b#v6p=#L*y>Kr00+Rv+`dnxd
z!vbX#n|05gL%|Er)9e0wNjhf|PMd<FVKl54C{j?fY9)vC@4<!*+whpdU<Wudv$N@y
zlg;Ya|Kmr5B>n|=kx==?9Fj-nvFq#kG?Z2n5GW_4p5UM;#OLN9rNmiQg4W<{oX-;Y
zZzFiUo|;4om8^>{X6GwYdH9*Pc=qkj@CZeBiz_zo+OkQ^p70yf3As=NjyOHOFK}uF
zyCPA%Hc5fdJdDjdvR&fmq$Vj?{4?G1M(KTUqX1P^qkf%sA}X-5cnt*}X{}Hp#0kI+
z7Q??AF8UV3h++H6V)#NKM5`^ULWsWrR(+=CNhw<uQqaGDe+CZ@o;drjfBlQ4ORYF{
z^SJfaTP=$eTM@~Qj@7GIlb@g8YWJquI!sX%1`HU$z<~o%Rkcl|PD;tPZQEGByrp?O
zoiJfOc{ZUZM??n#?Ah74t15l*Y+elnF?|M5yJH79ojS2{*5g{U2DR~nF?cXn;8Y5z
zfHq?`JDz!(oSXha){VE|N{H2oi`1<^b!*lU+u)>Ib~@`e@3Inpz`#7JYJ431&p8XE
z;w#}|d<6RO>HWY1^tkd*^ga75iWhy0v!XgskTk>k1;XiO<pmeA>%-|S)(4<u*$=FJ
z;xQO;6tC|n;nn}m!i`D56)$JAFa!pA*pXy)N#P&2T!U2foM?0P)Z{pxedrbx@b<Ay
zNsML0&_P7&+?q#8lNzOK^GWWWi<%g!M|zu31+?k`-HH8<PeY8|D}<OOgb2HknGhmT
z2yqWE7qDBb_DM?=Y{B*dcwY$dq!1$Pepf<>Y$3#pz{{<eGc25U-g!Zn*Vfi@!U-qv
z@WT&NS7#OK*%bNt`J8?B*}U?~D?!JF5S)ANxvkQ5f5Oaebvm7Gl4G~qx$?>@S-f~L
zKmGI*lP6CmA;GG&VDnHE1((a!>2_@cgu&2bfP5)s?}!S=$D!)CZKeAA1!P=ywecV~
z0^ubTzwsaTOnr@hA5AB&R}X5}t!Kv-f5t8~yj|P90)6tOmwXTq2rs3ACNz*>%Ns?v
z2me9ChOO*)@=4U4+wn+21A<=9Ji(GWl`XsWa@%$P*KV%4$iSq=O(Wb)>y<QXiumnU
zqeF;RQ%~_%3+Xc9SZs+&ph@a$YH%bb=qqcI@})l!*S!z1-MR*y&00l?v#Ol+OV@DM
zOCRt{!9M*o=`~K8>Tam2pHiRhLQ0>74v3AlQQP1~MS97bf*?IDg?*)^hH!<kNA%~T
zS0Bd~rRw9-*vE#FQcBjWqw8^_5kl`dYgI(K4Ol0HI2`Me68jZGh+BdG7i8!ODdpY!
zL7%=f-z8y<lroZ~I6~<&A2>0@DiXb1$M@*PfGdDXz=>YeHw8V0?O0$G5UNgw3+OGS
zY;_kPA%q>+X3S{?H+or=CBSw+o$=GTH*g#<90=8Qy#VOj3JYDiF%Ja=1@!IPH)I)(
z`1pA8^70rxdNjGYxum3|ker;1)9Iw5qJoNw3i9*w`R=>#EbGphIQit0`SQyzTd3~-
zM@>x)W5<qV<;s;|Tj->e0dCdo>}(!;>@hC6<PvN)n|HmhzWR#ElP9xb!v^m*Ewus;
z9z2-k%a`MDIK1(God^%+nQ%%e*Pvf~5rOw{W}M5I^mLq+)nJk|Ki`^q$N+3k4>eo1
z65q2sHEY&l$NzyZd<hf6=Z4i|<ATgAG!!Y(go690U)Xlt4Rrt4gA9J-Rd&AkIyLiV
z6YF$Q^8F8-aNA_wykQDATzwAlcDujk9UQ{zq83K5nrWr;T0u3*u1}|v(4!A2d3ixY
z&_sfRE)!2+-`oXs$s0w<-rZyk9^@s`LJ_3qjrPU3S)xe+6QtDmxdCfF001BWNkl<Z
zaqSlV_SEa_DXl<J6f~pjgMW-%0R1t~t53y%TnJ4<U4z>oe!lHABt@m=_*}6X<3^7_
zRYWjJ;Y0c|q#N%ofadW~xMV5aCyoc97<&_WNGVqfAvz?P>!1k7R1np2O^xsnp$VNp
zcfm%2V7}fKG6%ZbRS5Af;KdND*?>Ib(^gDrt-IAkj~+d^`s%BhI(4e$I<>X6EL^yd
zg$oz9Wh#Ua{PUmxY?F-rg^e3GvS!Vi785NgC90}&?6Jr4{PWK<WXO;vvBD>vbQ1ID
z&*!eY?qd4%=}0MC)K*MN$>z<QS+QaTd3kxAWUhtq9xW+l>w^q}pz%*dO-RHR6N9I+
zqRDt-|3PRmDz(3@0ts~+HhSv_2R<By--m!Wx+e1r#Ygyd6tMl;n<-l`pI-OeMvupy
zMD5jw@<j_tz{BB(4C4Er*LdsU1f8qHz{us)G`7Dom8vb91M*OP|LO;@?)Npg%1SLn
z>Ey#6>mXrpZ)$h$pzF{f*d4J(LW7x55keb}2hDrI8opgAaW*vY^lQ_(=(fk%U0#i*
zs37zrGK$a;NEGRpktzfCAdujh<X-4H2@&uE@+bWWkV0TrVBDBdUfS3an3lJy>S~_<
z&vX_qThVkJK$GnJW)V4KkH%)V8>&U^Hxtq8^ACVQVXR{nCl+x?DZc~$(+*P=0Fh!E
zuK_<tNLDEP0K5?iL61KADBZesi-f3<`17Ct%$PA_qCQWg;7{IW&z{}l-mO9i;^X7F
z^Ugc@{PWK_^w2|t#>ojG$jQlJ>eQ*c^wLYDr>BP<PiK-tm&?V>nKPr3Saz_(OXw6u
z`Gp7v8#KLRSmoXOK0H&CU6Y%3uE{ngd*aS^lkJ*p+pfvBZQHhu9sc|N{+{PWFFKB1
zeAen(7tiy&prB_E2v$}&rNbbtxFF7P&0`WJFVY^iXUvvOw6WyAy&QtNqQAOZc??GP
zs;EleV1N^MD<8@t7Y^tyuEhFyIsL8CL>cZqvm&2II_na3@ZR(6^^q@ehO`((2U31{
z)9-QN66YI3ib3*v-H_2SPNO1oV<&u%ozeY34Z)2f2t{#1$&u5AbN3gg49tbEfms#u
z*IOs$9zCw1$?i2fZx?ONMz^HA582Un(u^QySU6n5NLJaq#U_kSjKeuiznIQOgAcN)
zPnWslW1R4ZJH8xMRp`fy1tVSpiIc-MUGIbXva6$q6Nlrp^dHq7jE@aw2R396$C8$R
zq?T(n;bavOehV!zXUvYfB272m4<+qYbuTF-bYXKUNv*Jg$}t&fO^(nDsbQhLVJ0n%
zJK8CR$H(`<Nh>8-2n&7J&tSJd^uJz;LOcM;EUk#LnJaVCpm!pcTvS0Gh$*UYJXeg8
z?Rg6q&3g+AW*u&R3mIe{7YSWnR;z41r3OEVAu#_ba>B&IncruY%jHt*^z@Y5VzvSG
z<8fR!RM9g^lI&dD!44CX4L7;eZ*Q6yj2@3Du8nKeZu3MI@$uI(%+;T)@Ij*z%HJ8v
zUU@9Gpn5&Wc-`{E(1!joI4%6TU2d)?Eog^h52Zlqs_3UFjPV-1PUhl{n7VQd4~A68
z1No+e@6dkvOc)um;Tr=aa-0vU$ebnwVFeAAwE8!#noGSlwrQUfg=VjFjSvArMZTJI
zt}cUGlkJc6SQTpE?17~U7FDRJKa(JNT#XHRVi^UO=r^UEIO2u0Rz_Hm-wVdW*ad7!
zypE=t{cUr84b2DZBA&Oi5PW#}%bI}hcAjQYwf-71;qw~i{biCgScgRN;Bljw`c#su
zUU#n2F;<+7&NBF3JV>GWvD`!@ox0G{oh6K<AmO3=q`U7JDw#Bt;V<vc1Dn|If10g_
zzSy9*XX#kvepT`{EKCSQOBf45BYl23%H{$KhmihE0>>vp&uc-G$qY$LOG^=Bas1NK
zJafEscB|c{EpH2j^K~$&bOJoc24i%E!Yi!$f{+X+Gq~36$LRXO9kYrnaC=C-&Dp9r
zurjdv#cL~^Fap1Q&JJI1uye(}P2uwlzGJDV$(I}UrvwgvMp#v6uEB)+IU97UnL%L9
zuf1P1;S~jW7|<1!VYf%R`38%sjxCq)tr7S`382cHB7XPK&fMt~6bae<`kcS@n;l?s
zbrIIW&aax6SCi*3hvOlaDbST-OUGZqr>$U}WR$DsPu>QeCZ+p3hI`I4%LE)g6GpeU
z&`naT^i1g{7=SQ4Mk^80PKhvza6F!7Bev~l-KYM}6iv1H@9*T$6Z3sUd9aC7r=_;|
zfjM5UFCqV^^>=TE<eQs9(psqllw<Qw97LoXrqZ))fUuK%2hUjdHi?|tzxmVHz`U@J
zzj?Lr-~6N}d9x&oqYz}0V$ogP*hI@zW}fcBgre#1U7I}$e^CwS<O#@cVtWmU)NX}6
zn$Pj})QFU^gWR`IoM*|uh)Kr41fVkHU}F(}CGguU=Vy~)`hD;5;C?Um2$24%Dk`Y`
zr69q$J3A0HI`7Prv$fOzukBo`#b3|8I`kC!!gWF@r?3CLGl!$egU>pb&9BbDgH7+v
zhN)DRRQ?mbZ$gy5;MA@o#Ba8eiUc03GQ<@V(^BgzCOsCdTDM;wRoEtoI3ViliPRO@
zHz9XsNs198id5%GovWvz$|ISP03#E{#r$|K)9j8V_or3i!{nw4Y+cXTHS4y>C0(D)
z|5){4aiWKo*1FElEWW4pcyJ;cjLTwTyQpr;WVu}Z!Q7%7`fYu~WlL6ETs+!u5(fjD
zSy@%p;UH1f9*p1Vhj4cIrjqcfqNLRK)cM2*p6=uD-=$7fd%oPdMrEB-eTQFnJIVw*
z|GtA6D`>^qoTakUKuwQ0&Vi2n%mtUt$!ev^U-k$3;h{wRi;s&Co6p<o%Gur&A6|9C
z&+PQB$B(hWi7{!FyYpn9l1i#Yb9LYG+uTJk%co*cvPS~OIt9hWVTA2lwu9-~3$D1-
z(y#DTLhuR0rQ{i7)-|qIsmE+yNu9(Aejn&4Nnj)P?*emQ3bB9IJ1hDu6W<`Ki3%~b
zo1K)a)3$wZKYosFQQ>o!ukq0EnXIi5$M+PJrhf0HKGV>6zNn~sKG40=H`Juex5~O3
zLfiw5;h8^;hTRddS+eEbz=~!M>6-vpE!hb+Y@P%lXMRw~t%X!?SsSLjA7$@7^bAy;
z-p$k4*e*od?CMsvi!Q^(ZVWEocGE4f)E7}RDPnf<O|7}S`)~Z_CuAw$9cMdtN~04K
z@&4jj=wrHKhZ>vxMfdn1;SBhY=UZD`6(T{$81Rb>i0%kL_3NTYhb5uH|C>ZdN3@I=
zu=aNWZ8@EEIg?cQI@`cEXW=(bPou5us=!TM^JoLZcX%ib4GlAM^IUnVYU_>GW;~YU
zF$Dz$L1Sa`gM)+l#YYe7fxo$>`*&_p0w1nfuDhW+QxitaGMU_oXH{KU{>a=^mt0Nc
z4#KF9+O<Y-*xHV0b*58;Pw#}HnIYAmAD}yMMQYJ@7lcev=yypA>~OM>%hT^oP#bH7
zxy(2CJV}Uz>s6~+{4;e2HZ$VPoJgdeE%*+%<5mpEZ(V!RKEM2wtWF!CwUy*O^S@J5
zFTXE#U2kUuw$Drtrw&aS-7fD4qT_}Bn`;)T6ZdCF2(;}|9&q<d-kfEv01ElyeV+Xb
zheO};I8UYFqeO2D{LQ^sMaf|)!m;OQ)szcwqx%Xpx)q`DdCYTNkDBE5X*(kJerx+Q
zRNv$N?*Ff_*O|GlJtLSK{@Hcs#oX-&e|9qw$%t13Jl8x0M5)Ai;<}xS4;qd_r}s{9
z%QFVxQronowzMSG!rK|6X%hr%1r@Fa1=c}%dAq>g+PTIB1!oOxd(x?4Z9SmoN2Lh9
zRKI>jfC(7}xgBYLx=;(n)}W>=_x*I!d-}LX=7LU`KzHP!Sp+W{RO_BEQ8Kr=;#$>J
zQ5hT(_^_+&eEoBJ7SHalDLgH@OP4+6u;H|HvvK8}oRpMA=u$d-V9yH<6=2xm#gD0u
zI)&E+!h8XcR%ou}Ip^(Iewce?2O~vtuEF5febw@A>G!P58&wn9M|aNr<gEsHK|_tN
z`P<+XGp|!V2g{@;*W39HoCo1cb~{hjJ2%)z3RL?TfLI}hkG!}M50qX}gG$2<Zj~-J
zfaUBp$?dC%zHG!d(84P}jdg3{gB%~=$P-F;M=AswD(A-!+KxTsG;Debd-sxnoeN9c
zgFf+ub3;IdQ4SILV-A!q5JooNc~w~?hcx@BaV6UBjgtN#T%b9Y++uMrcw$dEz<GqJ
zZKL=vcG#T6c;S)^BNdnMvLU^C0=m~`@xt#a_eG44y@I_sQlSJj><>u6ljHO4q_wMq
ze|7v*vUP;HbX7R`mZnK)V};otj4UYY0;rbyzq1-fcY_H{m+w0un3lxT8zC~k?Z}94
zafW0K5Of1FQXI}LaSV5DEkqKIurQe;@HwKuXzbYR_j6M)t7~AWiFH3UYv_)5bZo3;
zKTbLYtV0F5Xwu~sKM{!G;o(UI|Gv07Q8~Jy{?kj5#<wC(89|Vf>Dm#t#ct1Bw*jvq
zNkT}ZHt)4Mm1xKR%>Rj$POSAMhZ9CFEP6;uq8j$>vNe2sRE++0IMmTm<MMOKA_G)`
z&voM<uszhYK7=2xx~-YcIicuR?CPi3*47q1%|kA)qdRo{v(4bk#d%IcZZ6rnbgf(m
z&S>yLTuEt2hw6<gP9rdI@?oO><HSTG%Zp;zAnJz)9@k)4XMx^~pWHrHJ$fj2PYB~}
zdGz*$50OvjyG|77m|W|6OTgr^*VqnrEnrj)$`f)ig-j|{CwSf-siI=rWl1OK=x0q|
zB`d%hBd@(Je`uUH<|be4d!SH0W7TU?v2JFcdG?15zt2OJdV@2cAggo@368>V-ap*g
z@5{kqw}<8MV_B|YqcwZ~(pZT<)Cq~QijKV4erR*$Cr^29gCLDV`f4h+LHyt}#{+xj
zduWAW6ju-oz%!<9ioeYaq3T0Og!8<*uNlk%ds!ah7h_MqU@k7TKgx8OU0FE^BYl-H
zAI98H^S(IOp_+c~?wAL7(H7NGGkrTYQycL^$&@Xs;2SrjL0)hY^nh(i(u$p2nWgfC
z`xfcu?^aHIV45{jP^Hz&rQ4Q;y{UsXS)oRI0R6aT#Y;)Z9%+cO8m>(lF(^?(bU`5?
zQ2XVJreYx8_ke<WU}t<1x%9u(c8l{LQbrayS@YM48njq)?PLi*^?G?5-ddfEO8*e0
z3x*j1?POGaU$#K}t(}QKm$Y#AF(GD>kizltUhcF}?9Vbk^;S-co(0P5<&jHyF&;AZ
zzc_;zfyv>G((eVa*xX1dx9@7TITq=ydM$0V79o2Wv>ddKaqU#&j)<V{6SG(Dtm<x!
zR=5ZViJKD62fSCjZ?gqy4=pjDH7zl_FHaxCd;=EVoM1CHM&t&cb8NST{yRMwOmS)7
zI+_d$87G!lzKRkkJ#B$BkIvt+BVN)8=}iT*+f7Fke%5e?eGpC^*n1w(=<du)GZ+uY
z5Q{E77KcckMG570gLh>2g4M)ER~szQ(8?PG+iGOMYd>6l!N~^3)R|R28jRz^^w;&J
zDi#yoQSrz;Lu2qFQm9%|G5N-!taMD1vv^7dudoD9)G!LPzSg$O0*4(E`j2CekrYNF
zvtJR!RlN>2^gC%=TO%ddpMP2CKfVgQ&uxI`)SjydP|8-%!f9%isP|DAJ@AgK?6YsL
z_|JLIlBA8@(o}vw#}w+$tf{8&Ps`-1b4PMH=)YG%A6sfn)g5^TCCZWG<4Q}7o16xd
ztjO4Ulq69M=nQmYs5XGUQ&U8SKPYTAtv@Z-dA?aJ**<gWZm(`_4c-o#^y9E<$@m-O
z5!rD>H5B|NheMJ!D}Znd<VY?TW9;!VFR+}0cZrv4JYLwBG)(0({8Re@pBc#8^?F7d
z1sWb$_XPPnI%D}AS+TC?(f9}F)~+96t7_wgw=h~ou`#qqSxqSU40}W3nnW@UuJgdo
zSto}Kciz(;(q4CbWgXhVQI<-FJGy(>2ke)Esn@AppYK;%&MU^s%F6vhHbi)@MMu}X
z(O_xA<dr&8hwB~~?3F7Jf1Sz1sj8+kS?k?2u<Vo|XKU5Ug}prIy~0iThn=I|H(wXH
zGTP6BD)aXZ03*48+y+mll$YIx$FbjDXcNPvN;m<pzOYuA)L}~g^4{u|!$`FC!DF$`
zkdkPmE-fqUB$BfPmAr1&P6*`VVF={YZqB>2>Dxd3{n6e`NJZxEpd{R!I(;{y*4{t5
zf^igb7Z<echj?N-uEd!6;zJhSR~Y_IV5@m;fuF4%XKE{1u{2bKEO1=I_2z7=pu8x_
zi=i!H_Eo$00i4tvv@C6HZ;EFhW{H1Hj5bW@`884~RO;7PV!rUQj3yl2O`Gt>NZ2Rw
z{sfl)t&!j93yCm}g~G(B^s3Go6f1+k0W2U6$klKd7wa>H9PnWKWK3l^Ktnt)-p)Hz
zT{i&UgH*S)us#lAUGJ4?{fIQg=;tP*89pHZyf4ipOiho49A?@Ladtz1K&Fw!Kaj&~
zB((s+#2iD6kYOK4J~x)nqEb_&0=}rPJ#NSD#MC1H^%7J0JlIR6DB^eS=j4u7Evk&c
z`SKT0Pz`du*U1$f9i8W0`qn;p!&CQ@mq-AZJDBnxr@)e(hl<@=&hhEHrt(8J$?eQX
zioEan^-j6ZiD)S|H&L>5;6-=Yzy3d{I_v8`WRA%}3mg!T--JPia<gw$#V_mPn|=8V
zlf@-<{t*+~(qAzL{=o?~K%_>%;An6sdSqqw@qK##=rKJH(RfT{j)HW_9<T49s5@JZ
zMXahCxci171)-!3W#Bg|6~76ooYS5msj7pUlEhH`agbcXwf!Z+bk~6_U4I|6g#z!q
zC&pf>CU{WUMl2>&gPuhxwKANW09@*af`PAha;()mvCndIBa9ZfgO}^L$17W6)kmF&
zB-NQ#gx)2A^M^>vefG@zgJ@HsEm=l3_49`!`6D%i0d6X0)W08tc@2?0kYNny?dAG|
zn{7AaG+><$;<$hTrJ38TEY1+d@}e-7Ef?M|;i~1kr`eyOYbFA|4FP=75`i_C!l))}
zI+WPn_m2G6n}+e560jD!S<)VTQiGYOAuaoz(8$2|;1EuxB^US6Y^Ci}U)KH|97__W
zrRC+}csc&iQblm#^Pe~qxJZ|yh{wGrw24iof7Q{{9Fz0D#Q+OdNbrNDv*y5m{-2~M
zSW#%jqADxnDk{X<xP|G^8ORsx{rrQ1%7{53`iJn)t?@p^mDnx5CZq7Ef<>%%bTYP!
zWB9&sY8CD03&Zl=pzdAh6QI|ri7zTFI!v!TfLIwH+dU*`sQxdSJR7a%!gwvFEbsNR
z#oxjr-m&UdDd%j)qz#HFgDJ^993JOydt!*i)XBL;jnv~@(&!^H|Ju3J7i%7qyK-K|
z1p-e?)%FYcd?KIT{0{_P(LkF{JS%a7$s|D?^qna|iIKc-SIH@Niv~I=JlWVs<N3LE
zR;UY!+x-1BrdN|HVv<2V@IAjV$y=<T`${8r!XuIDjac+y*|ns<)|4b-_Ode&hI<;1
zq&57zX?>HAB+!^prRPkq<98u<^R-lCT5B?e6-z|v_*4s8RJjZ8n%q6r-H<Yrh9MC&
zQzLd|7xDTKQm~?>1hDxtc)jMkQv6XMEqB{7Ud3BvI_xLSpJem*#WTFbsb6Vg5i!En
zb4LM8olXt5ArCc<8%Fbqf$KyVLRTV=bz8`}VwlfV%d$Xdl3p-2YZAOyqXpatb9>$i
zcH8!a$;df_+z8bAdJ24?(>q?wVnkPPes|Z<@r+%~j}nuu&F@ZYyV76q@d>YuP$i=T
zq1#8cb9o`O1l5wl_n6y84g5VU={mkc6nG-_+%H?YJDmmoTQy3`>*U&5)M8(lCSxay
zn#v3Cnyk(VoK{lpR>I-2CyNKecgI8ih@n&%n`yH(05Ru*D7mG7V2xV{QZ1v(k793l
znNtyU)DG#?*vmEhZdG;i4fkCaS(~a$Yf>?3ma6S9xRdB~N8%2U%5BbmjTm&)AHy$+
z^lEWnQn=JB(g(GyAhN$MDZOC^mRh0aA@|k94zre?N6GnL&5@)M)l=^<e}A}=t{~r5
z&;9ikFpPeIh-URtn7e%OarDb~G-D{f={;@`7@2WjVl!<$$75_(I!)TTswcwfcuN2F
za-^!jMka1_ylSeO=XD__1`xp5lcavJ=~ky80ezY9X9hV3g?Pd}>!Sd=q=6ec%{&zJ
z`>P10qkd2b|K-iuZ~K%brKMq{X#N+OXGmL5OoP*li-UlVYZzpKH`}G?N|bxD<Lbpt
zY7$MpBeEo5DXYMXP0)EAO6u$Cwk`s^Y&2ib?UYYw#!(<J<m60h@U+s`mSbpc$Q{G6
zCp@Q_SXaF#tLvT5Mj}W~>jq!rHgKB8#J|7Z6&}fk5?A073H=CW*$BNMbQ2@8)dVbQ
z4!~v)I0|df^GB`12^}ba-}$J(GE8j$OG%i5sA5T!+FQd51qe6o!A@T;^2Po=`|xkZ
zJ>WB3XvYo**n=8z1v8<7epA5z$yOf(TC6xgN+&OAfrANMux?-Op`?z$RUf)djKm>Z
zXQr+Gxd$hf!vcW%t2CwX&2T5)lyI!6%ke9=_g?V&&Gp<rCf6N%@m@AM*hsX;RNk;I
zttYy{C_H4mH>jr5(Bjy3#`9x;e{Blr8$fmGtES=ji@#~lV_*ocSoMMG&!^JU5lE|O
z!){x@$sO1eg}ky;U65h#vr7SveLTf<ixFC=)?Z8QAJi0ySIYL<Giq0F9F>ope~9>*
z!Gn@*m@U&!?ZKOpxoG0h4ycAq(KtJY^N6&q0oiLmhQ#eLJi$moJ``qLOZM=I#M~9y
z#W6h=b$A`Dt#4KDGEqX61f!!jb-13w)!yH7^+c#7h`c@H$MFND_}C;;0H(^ls3lOe
zdF&(p2pyKjxP5^~Sz;8(p_IUKtY}?Uc9toI@FkMYyEyo5Si>Hcq;UR6K!OLMiXU6l
z;oG$N)nR^dN4jH!<?b~`X|NI3XJ0YP)rIIfXc&QrV*Fw57nyhy;y+g$!)1qcTkF>k
z;|N?X-s#50+I>KD{73?mCqAmfx1Gf<q*ACgG!kZb&hp^sPg5c*J5GYAp1h%1tf{+8
z)djX+T^C{*_-tv9P1u6sJZUuI@}s9`#Cw4}u-`q5q)wPrJ5{ABrQ&8o(7XKBA^bnr
z9cVxADl&W~gP)(uY;LMrq|xXKzG>{bg@O7D3k;SJJ1VP4DD_>AinlN?w($a)LW>!2
z)%V^Sy5|xfttHPzX~*0$UOEk%ofZXd5Fp}){^>0k?k?jc(b_~LkrnABa$r{!Y!{Q;
zF*^(L5}7ftzeCREUHwV$tJ3=D!~YrJK@GhGlRec6%9WQ%Ml2F+ekq@eGVE=h(RyA0
z#0;Sfq!(m~ySn2nZaK;P@M*q;r-()=ZDXyX?F0$_DUvC0-r{RhRz-m(sYP@(s=io`
z!a;d!Oy5H)J#}8O<s{&N>VTTj1U`Je8qV5o!IGtub2xv$7)$ypcumjaNu{BH=336q
zr8s3^<52$79h;psi4n2&RJE(%dX={mq-)Al>keyYt-7yIgI5S6sX~HYurGgdCTkw%
znk0`VbsJY%mRqO_H2?wO-NSr!Vf>8c_hG9O!$Fy(6eSR*#g5GzOvWtR)9ZyI;BQmN
z>5eyvMk~F4a<=KbaDN10aP6OB|A)`7ZFO2Fr?7&>%MMivhy&D}5Iyq#d!&=m#}^G=
z_PH(5uT_1J+9RC8+8P92-x{Yq6b<KhRPlrMZHouju6uL-qRA?R>x;#Ou5XlV*728L
zz-*-jgw~f<gK8@Wuyy$6EtN}HTuhMRp1%uOue;vTsQd-EtB%$tMC0R=|8i2`*ceIV
zN65*NuyO~u6|mmh8n3Rofp}DzxFNUR*I|a;bYj{Hg*4(V@qWILOv*`T*Zd3{&>At-
zj&j<Ij9_|(pX#_bIhC-DBYAr}<S`ptW9+Irb0jILQR<~1A*Sv4o3J#Aa`!FYNluPP
z+sVK^;#!0>Y0B#jF-S$#?qTXP#D@7~uGbiuw?d4!=$tu2vkHXeAwz%wp9xm(yS{p2
zJnR@^GD^OY>=5CAaQ%A|$;5%OxQ`s%x3kQD0-QExPBlFV?$8cLq++1dRF(4{2qW^+
zMqK&X1_gM$=4qe=%rEAu>B9+`e5BJpU!ri?pMLMOU92HeFIN+hl}mmw&)Quixba54
zA>s>7+ifa=<f9kSz{N(8PBd!@o*&4QSk=UGADm|VPF;u_GO%0O1j=q)qW9cspSxCq
zpnqPx11jE%iKH;wAp-c``PRKZ4SDZY*GleY&q{G1JucW2G<`g|lqjO0Ff|V8b9Q!$
zQyd3DBS;m<WJb06)~BGHkk(?~C1^tf@QG|KyDUqsTa-ksPvsz$1J?IL>lxI)K>-AZ
zFFJ-OKrvRYs#X9aU~E?0e2pWg{7;<dDA1L`i>%Lfd@<nwuJ}1j9NYzicCP`7oGMKi
zrd;cj7W#m!UsU1aBUao6U4o{kyo4-1LDYlHcA}Dkh$U4moMs9v?v`*XnpJEa(^=vr
zUhJ@PJ{T}3_3=`b+zV=`S^M4Bi#Ve^BB^00MjA?NxLqpEkdvRE33z*5+%_d|2Ove(
z8K5!|uv@8GgI0htIDFLqjJ)2=SCcsyKgOTpN~mbyuaY)nMU^i8KtrRo#Oh(V{dZE%
zR~cQrnYXHt@)J4Ic4DB>>FEfA?pLZrrQ$e_+~5LfE_2=5@vzv-O{+g4uBAjTn;e7|
z-go$Egq|rj+O~K@gp9ZV3^DAz^6os}HVmVUj?0JkXD`5@+>Z+#^r8-*Rb%t~cietF
z2-(>4ChU>Qa0X5cQeQP15w!mw3-Ct$!)t29@c9|$w@7Fs)7kFtf||m}<RiL4;a^B0
z(KC8&wHphjO)`}KlGe=)$`lUzC;@#tFqeH!T_3th6*_=db_|>Rf5B)6_wgt@asr(h
z`c6*YHF(tA*B}#6Domg4Sp2aVMQL(GeW;~Cw_+Oj>8;ua-Ud+HX>dFQ$iaqnC*e63
z;7nmr7y6qHrv7k&WPKHQWlp_P!<tcjB?atFp?4}}m$!7EVTZ?geX@(FxRjyC42^;a
zZ$Crz?x{s})zraixIl0Hh#}E;Ck;nC$(kT%7S9pP!fdO-HF}r1r5KwgMOFReo4%Ls
zP;sXhf<PET4|-ZX3Cbm?-m3~xq;>lv;lDwkIjkuRGhN^6ZSp^S=&p2L6o?~=hNBSC
zn#Z>^NcmAe6u=D>&h+G?_7U07tIq1rY|#OOfiJisw{J`0a)+%|IwHm|sN6~H)w`7-
z9H;x-`SUgy3f8WBI5R3yU?O;PgzVng9aUA?>D(V{B;^N{9No{GeXxU$`Vr6H5ccTx
znS5`EBk_f$!0)`;ojZwM*xoonE>mNks2SwG*g4j$L&@MC-%m0p#EhB+y)oVYR^{VA
z9tLgw{K|FCWbInrCc>){hp@z<zfiGywm8U%x_;K3zS9vQC(8zdW&Dsvk>&GVkvHcD
zfxMH*bqBZMrakymt+1g2BVV#XTya<aJdQT5izEbx(E84;hJ`9A1Yb&;<jo$p`7{o!
ztBTi4zV(81!44ZCBwrAx5M4|Ce8R!7TZ=GNxSlX`{~TZ6{>r^K$(gtw@@lk6C6cb@
zOSSB0{E{!XYqLE3yl1j&3~ecV>})t53CWiFpY1!-zLFi+uAJX>`@jDQg%uu@%d2)~
zTU`_zAclboz0-D7N2w5OG&;T9{QiS=2fS)gF9uka@;Shuu(dd7L;uOy7X8C_AiRvd
zuwYrA#pLQELJ`|IyZ+A>#_Jvy;M=UVp*UqYb`{BL4arTB`HkxQ6j}Mt5HCA)tENKo
zyYD^>4U2gGm{sB1kUf3&o47-Ew+dIgtGb)L76{nExhvgOhxlL?h=%p&31)`Vt=%%)
zM0xwgrvkerS)UyDSBA9*alVy6vgT+{3!3n&EZK*CNkkr3ydt>@heo`oFI6b(j{oz|
zvB!5N)+D#_AFSVgjgAgS0mP5;no78?f8DlNk^mZ`oX^V0MJ}XC>|&qS{YbuQHzqW7
zdcu)<jYTGB4_O-q<?hDDv887iN}sSbm)-z2JQ>(Y={g_;Dbz54T}hKEw^UO09SlS`
zk1Jk6rNMePlIY=GiV9g8ERqi~2&UhC-=CHPZM5#;E`9dRt>4xx<q;rM{Mea`30eHx
zeB9Vzt{Si>ruc?dhq8_OB`K9UrY?I>+)%7#;mTY%=q>aH&WCV3?st^0nE8iMBYa?A
zvwfeh#=!qR0v`JGI-V>4Zijbc*e?(V`9tvx--lIDh&ha>$VgHhBQG*#Ta{apXWz+7
z-r61O8GqxmJlk~kv@>|c&g8w_@O+MR#t&`p=}U`BaRaI(D(|6kn<1KxKoXZl7PUSu
z;5f7Qy~?H}T0t4t{4V0oGw<5yt>qg!XL?PBeeaB*|E4mhsxTYnUnc&}6hb@{Y2Hey
z=HfD(AfT%4I}m%W*_tm|>rRvV%~KDOP?*}@^Jv!L<pG$?SljLw#B=%Or{R7HWqe8H
zxpzgr5DHK^aqvjD(0o%fLf~a1VL@f6cf(!YG5>uCV<-7PYbCe}nLu;QB(bMU#~AaV
z#1TG7=r}gh5?oMV*~<&D&aX-rY>Hi@SLm6AB$tZnHz!p~pmn~a8GpYIoNs8do5RfB
z4Jm2+s?*fr6dr9AdQ{B8v;<n-sLU-8xS}S6^eRG}zzOm&{vO&rUpDD?+i~_8U_(x>
zCA?LHHwoJIc<yZhNE{3&F01dOVSJ0ejkw(#;~#5Q=zBT$`Ah*$ocZs<+@v&Kxn6!O
zBzitVUgdjDYN>3x0Q#c&d{-fh^=0nc!|tAxS+6fk6wR)TMzm3OLx+|Pqx+5gt;QhY
z{oC8=Rv4P2)hsTaPL7~A|7W^2D+Gctg3^JiuYEDBn=$areM9mrRsi@G;uJ>DX7b6n
z`KtR(Iobsg?c1F=`HU0@ey+EHyHwCh_;7{w_G?vkA52tc6d1Jg3$8=Te^bQ2layFf
zOJp_}t#wXi#sUsr=<r2d=Tg}(Lr5|*OPRE|X^WarxaBij?rzEBLo&L@AoMT{FIKJd
zeuRbVX4wDvQO^4Cr3Hb>fU1gWFFcz!lgH;l`|$^V2GC`^7t2#D+YUN)wKOg}-ayV`
zKhZ+FB`mP+(saMANJ+ped!NZB?^9pkdK_HCUILCKe}K<eV@SN3NmJs&Us}pBw>2VK
zTF*wS_tNAO;&muGFM>Y~1FQeryIxeyQ(G0TV8^xR?0D{yOh-*0_i%)~s9;}#*R=k^
z%Moe#8cnA2bWM3OW%A(@#Mt$GW_j!VwH|d`MVTt+dzIJHlbJnj=TtQ+ai{n~-d(uW
z?=U>66K3phvW|VA=t1N<=)DIidg(<GWhOj!Vz7&t{X=#g=DVDunt^%UT0=wU9SLhn
z@I>$ZSP}lx_XEZDMeakQN44mZKj^b%U%+jN-=pRumU6gCSMJ?zh)k$5J$v*yG+4U+
zbtTzcw<F4KXsVxv;mpJd8j*&?6v77}LrM^JkwTH<Tpel4*$A~6nln~%%F;dWU0n<d
zMn#@vwW+;lUKw6P=*JxXfNQnxwJkIZ(r;3u4wRy&J0^SQ`pHvk9Pxa*gyXUba(#Z`
zcs_OVF?3R*5%cOs0g68Pm23JImn_9)REY{zK6@xPgla@UNAlH^iB}%K&NIPvS&Wjd
zin<m;(aQse>paBV3cqx^+UW}uJ(77_r?pbWM&f0~N7#sP*$-~6J6;&`e$0<$_zX;y
z5bWLz-2L8D0=Nn|pV;F8y!>)mOcnW!ldF%32{hwOF%Np-!b<hlR@Gj&4V7gxB;M;g
zV&d3v%|!OMx)pwTUzdZ=Ymt%Tz$pYiG$f}I@YBTtf9^+PMIspY5KWsBttT{vQ{#QI
zd$b}XNGY}LbS60+i_TxTWO1C!EHx24rN3*+-#7Y*-1*k0`SN6e;LD2(WR*z{TlEvk
z2d^sp+16J&S${0GVSDOyijbW@ny|D=S4JKmfT6TGx+@CXo^!Lj;r0}^NRr}F?rd)&
zOTNU>u&or7CNaJ8tDKlf&?8o48Y-);KQ3pu=b!yHA;$BblWMi6C@J7|-G%G*_`T)a
z4~3~QFAa$Dj#1T~u`~1|GI=5n`7J3(npB~bitY}3=6Ss=g6~7ZzVCGRr41oKn%Tf7
zkK~qrd-Ef6aA!H!?!VT8Td^N~S!pyxu(h#AYvBz5W%@dDY(#3rS!u)dnB)d%znTQs
zYF5d~x)Pp)5iW-{^=gEY(z8!P??03O{C*LAdj}Rtys*lYIA#flQscGeX6Q5(Om3Yq
zdhGXt4FJ`Y$|W|`hl&&5i-i`6C}S_BL&Z)DCuil4MGhYQ`>QrkhqVPO#zyZN!WW1h
zI$8aTq!mPE`WsUY%D)!$tp9d^y(NgipZs~uk>PR(;`}rC!O92X#mAyI*smFi!RI3x
zH3_({i#QC$OcnK#?(37wagE?VWt35;P)omr4{j?wE;fuewSh(nAHHcefDJa3{$E1{
zd}`!vPGARM*izJ^b->w?@hNYLn4!xLd%>>1RjjR9xz+Vpk)-O0fZJe4WfdrFC`efY
zdxlq!D1aV5K%^3z#y#kDfgY6U3gLjoc71_wy?4^0ptLTeC&{WK$)Xd<XT^um>gcUT
zcFg+SF{egrTDh6MMmMamB8afnY~T7@X3$Y6v@=MIk^MU0g+gPcAKI`GTw|#P!R@>W
zX?G$GFj_CwtiR6D;IRB5vuSTg(H@M>t2W}2m3s?M!<{SHbg_x}`A|}5nIItmT{gQp
zJzVR<-@Dvr%51dTygkWYd@ML80HfKk$;Ln;{`GUyU6`AIhy3?67Xk4mSi<z>ca=7x
z4CLeHpT+xtqgY(7u!lO+J@qy6C-Xaaj5eE=SnmqHh@XY=L9YeJ^KrZaYcrFrE78^Z
zCk90XT=6V&d}+CfnC-WRiW)o!IAuMHs;Cp|{u`@%;)TKC?|LVOaBUHQ0q-_Bqnj``
z(T3(3_;Fq{0QVZt^7LYAEgg}+{Ua#)U}8Y}_#V<GKkXL@USt?CqA+|C-|(D4E|Jbn
z9yF62qOIT7q#Y<?k7vm{{>D%oQhsGoYEOVrY`}0HwAB|gL&}71dWgA3BZ-_vW!7VQ
zus1c`?g|&b`J^)%v_RqJUTdyjL{)*XX>br9t;Fs~l^RBUdzKC!6-pQ7)f!R8Oo)qo
zAim|9Df~bl;@PqI^`eIlE;|t{t=HmMJ43YSNhUNW(y~x<Kellb0Q@ZFDTrPYwQx+y
z8LguCykbQ0_2Fotp3Hf!gQk04Pj{zWN40+zH$2}G`cj-Wk~xp~g28J#sEWugRd)Fz
zf@zRh#g4GtV3l-3z0$bF&W$Y1Ef^jT2l9WCo}MvfEe))}=`EOAjX~dfZ`X!9|78TP
z8OxKx<Oh`8{ksNJxt2AtVf)a4xSL*^KGz5MmBzcxHIA(1CMy>{#cr4Q5j+QEVn5tq
zai_kCQsojMATG>5bqiBrt>WRSv7YHr4@Dd9N>Xko8uzk&SzGGla7%fLq%nJ2bWy)=
z1FJ`4y}ybbdY(eOAys(c=bJ7<Yz+o#{4X?OuEfPy^gjU~W-Ih%s;KJeJ=2@`o%maj
zD*gu7gdw>Qh4lGaQtxY=VoCC<E8I7N@<Ns);soYOj3s|@Bw7busb6<oI|vU%_@~(3
zn3inX=<0Yla0$Hj@x9#zwx~S_!W)(683@%F?=jjkic)_|Z<NX>QN!<&!-kTthu#4V
zZ&EeUAI#I>@i%nC|7+;c2rTR*k(5rQqOBmo*@~lsy;#vh&d+s`&8g)b^YK33$nZHd
zkHBM3Xe=Bvb3N8)7Pi23yJs@8pqA9Ya6wA^W2vE+@PJV6dcGCH5W#(`?b=dHfAq9c
zbx2d?^J1`??H+VmCopQA4MIe6S)UB++8X)bc=MAn7S$C$^$XDwBswNQ2BUU`Dv`y9
z7f~G-N?)0)$CA?LcuK2x54YwV-;rF|{X^uqSLlADNlUYn|89?tziGqgdCS)nRJ{+@
zlSW_)v!uo1z;he%Q`qC>ilt?v>`aHtevz}spu9b|KxnzSLb;oEa2P(~@!N%f&(YUJ
z$!l-A77M?9W@RF~6o8RTG*<&^u6j&TU^bygQ<OAB+(Lz`0{1>#6=0R;W@&hNsPt;^
z&JhrNVQluN@%SBQXS5t2J#bVMB>R-6s?TeJaXUdeGQ0=h+rx<5(ai=53MJ3Aq|rZw
zP|A(>@(Nk`K_sLmgi-^`vAHaFnZcrTM43=#=7WpMzR6HEg8mjvn&|eKq$Dtq_**r0
z$ibwM7>@sU8D9YM3tbSKi#X*RmL5C~^zE=$4XR8U1WcDI(}J{q=uArsx)|E&SdKHP
z+Pt1=t`SnI3@6y)Bc3BIDjn0K5OHM%A6}F!m+jH=?A|1I$(6edF7M$ygkG@=XG_GV
z{;So@>_DzkybftG6^Ju7-ns%<k{g3T5cl7Qy)%FAxqJ5lYqEyQ(Bx{?ZD~~R#b3K6
zzC<mPwNnW<8J~cworFE)LVnf3*`d#Y41x84D-gl^h6N6@z7VZf{B97Hq=Hg3ZOhVI
zP-MC5y{-{W8Ln4<PsmZ><OYObag2?I^jex2^^jcmSfY}dc@LzFlT-HXb!!iHtdgyj
z6hTx)=>OdVx?^fJ!^VlGVYQCJGNTLq9m6Xv_`TrJh;G(9x-vUasaUe}-%fth-p;K9
zhY1n})tVS-(7Y#U2okT%BDGFs7#dg&`X7o<46(=g<Z6=hrPsaa#}z>Cwih4a1_%*t
zWM;XKOJk)QR~bQWEn`U<sO-s5D)UH}m~~yM3mcB0h+bi~6taIw5=j}m#L}UHQ{98}
zz-F9byDaFr-NC|T@aaW$=amE|SG2zzbxm`d(l=ajrwCWI1g%%U%1b3B@b+U;=QbfD
zS=Ls+t{j<b*Tyjo5r|o<X~+IOdv)YMc}Ps=e>d0A@s8Zde*5O3ZOeb>^TvW_x;1q#
z@E%*6D-3zZUZ$t=X&Lc?L+>)Vvz*2$cw`%3S&>8vh1MbhM1c{rEM#gEg`$31^;f|5
zPGVd7fMrU`tF|a-Y=vJq#`e1Towt#V&r=jOFM3}(&yYw#tu>u6Epr%So6b0eBAh3I
zsxxk5w6JAk2>vN{e{~egyd&nf@7>Apy3M84YK;RMO%zO)XCMa_{m_$O!(Z`~V}KP2
z0G^h{uQYNXsYb624y{Zr$W>4>h|SN8s7xxEK`l6;8si!<qO{87APwo;g`0ne>KU~I
zf&DgrhW)zC?+iDh1!HyYB)03rDxaY^9-HFnYV-)k`{q5){n9IW6(DY~g}(qPis?1~
zaO}M6TIK&-&>?QnB~RoI3?Squ-PXmca~+yOyB60Vj^~yiRS|9b$#cd8o;`S~T*k?C
z^j{yy+rYx@n1nySv`cTS)D#qe1^-{K)~xN$nU^L6HqTGAEdXMUc<I>-8%5iGJ}E$=
z4WX>4Ls2x_YL|!X#WbLCP=Ec|p}JF%-<?)i&YdHF^HV42oEvKEPOaXN4MhM1kHY^B
z@35uSn|2s!ma=q&Vflawn_~p3EKnx>VUHwgTn}U6D1iPDji5q&Yo@{ED?SMyLJryf
zv@Gm)o4S<3>5LdmP|Y*cZh^;AbRj9L9^|>Fk;H#PHb&T-JZPFtytV>|Q;sv3<qpqM
z*-TJX9_*z|eq$ttS^x4d0rtC+%xn!2{BNi-i{V$|EA9fec%w=_ebC%0Y;pKJc=0si
z2vFBNv9-j4$l6;_nG7V^Up|@>WgIyP;3kFtGa@U&<gb^?3s2S)f*!qF%cJX4KFDql
z57mCN2&55`c>>>qwMf6l#AE=K4@SYsYhH^;7?iOB75hpJapDTJ8twLrpSyaHOxE~x
zestdBwQ$LXjSXvheqlo<2R8j)%hKk4B1iJ|;BF$VdJXuBD)bNyzIjTc+RJh%|Hf7o
zx$-+zn}UZ*C<!#7Fuq$i4UTWZ(hWQvC6>Wqir}M+HaN|S=d$hMTk-0p3R+M$^fd14
z6$OvHq42UiGa4B95YE#3yujWLNEN{aqAl`y8B3|vs(w&GABkEb)qc|B;Zc8h?)qtI
zY1$tfte~P>H$d3jIk(`k<}c91VT}=6M|4hd0WM!C{+G>ZYEI8=Zt`;_kMThwu9ZSQ
zv2NWSw7c#2#H!bPpYK;!Q~BE9`_GqXPFvC3n5o1We+P!$OHAuHb>DR+Gbj8ErrvB^
z4>0E_jiRV&5lgTnF-le;;wdA4Q5yCjI6&ziJdcupxqUKE;P^$j7$9LiE5e#Cin+SR
zV{W<G-CIbb*81m%v}z@oWKIEQG*ml#vC!eO&#tci1|!yV5n#7*qb=EmMU9hIN8wLT
z&z9Q^U@^qzXSf{^F@_eZ*Oq4@a5SVv>7}7Xxo`LJRccQURMkM|4&`14ZmY=*;<qoZ
z+Ic}EBtrc0#!FM4L0$qck}`wgsv-YEghE6)s7^H(!N8M=yDgLbap4ji4sNZ(iQ6bE
zCiAbV_b36KHHsodkTi{zKyZxehvN<=H7Y~k{<8HX@F(Jh4^sZ26rRpza_@q{JGFM9
z%<r%<U{``|m|Pl2?u_XuugoV!bWDbBKM<I;gCps!b>&BYkSm>8c0$*vsOdW3OgW1}
zGb0j8ITe&x{%`iHs}q~@1$*G=NIWjH8*`uYR^;uB;m@*?fifz887y8DvTHpJr1Zpq
zy+Za5*e2q-ZBWeAUQ^@D%(~Fo>_L5d$KS@6JGIZpwR&1D5pg99+)(Qc)EACDCuZN2
z2~`z`ZjLIWTS>|QI?)hp)xOb%e}92q(s%fG>Y&rIdD(QScO9kPMsi7iqq*p?%6Z5q
zWL|qTD2t0=-YF1Bg%RAUbBz8kI!ScFa(CA)eyE5{o<;Yx<&3^2>+D`sd1l@^Sx6@T
zqn3wt7;7hLqrLwt+QC%NERZfn*lLTIdwi!BfC<;Cyv%N;0`@yb)^-vc=zaq`+c^%I
zL?isQKlM{DrIW31E!a>k({6?r=VK~L$TgLOiyG6kwAplIF^lrhM@LZ^FvgEHElnvK
z8zMc=pt;l1Pb~iwRad4)6+P^yr{_%j7XPHwNNAwgNs}ri1NIW|-<aWZg^_lB@`?T6
zq-(~WsrDp6kHrfE>i4bL%suap&0lf{zP>++wpXUwl~B%AS^CC6{sLBU48CDkqW10_
z%l)s$f0dpVM78%)20oos!G*~ezPw<&f-H24%S(Qq>nFMTdRpAn7)A8d`C-NdHE7VD
z{)awLj?FpKxF{8klQHap?11HtW3+rtiF|kB3mJX(Qq)6p`ylZ(V>34^M-LH+WLV-{
z#|<s`r@_9=0Yo>|prXxZ3tgf>_Dz#7SMP_fFjDJUGy2_iDUO)QO1#V^ly5;|Ocf%u
zc5BSa`@#y0ch3FzTjFYrOV;qQM$8a0Z1=6p5-f4@3FZ~{q0MIdA?pQ|@*^rg9FZ}V
zvUa_pBXm0B8b?~hX>lM0gfU1JKcJ!2;_xD^x`_x$3F(ef@Hu`*s8y`eBCt>FzDuyj
z5i@b_pdT%Zzjxah#|U<sa}XN}A3~ai7FGzSc#4}mncMxdmN%mWJyZyf?zW?tH93qC
z8l!3k?hrnYP}-fcsVcCOVeiidl5jqHz9cOUgqyox8Z(oMGZ`HANDKyuqN4|-_5HeZ
z9^^*j!gDpQMzZ5{hWNj$*PQ;U^{-UG@XPJV7x}`22QJ!=t1%{NY=6vh$<=?pv&2%H
z)ap|)K)Aq+e&XNhtDQ7xg674`Ca|sOS6X@oWVK;c;k`|1{COuA^>J=<nyKW5j&#?@
zwMZKG{Fr}!A7MJ9IhEK+hPE29F!#W4dm0Y}MOu!#61*fzbM3#$#h_t`n6Q!WmlwWN
zys&mKC$Fc(MsPk8wbxcGAlps<RI?B`aNC^lw*o_z<G262Ze~<Jdc%tM#1PNgF3K>4
zLc;<0&5uFj<s}`ICsC+8)I6Z4Z{*-^xT0WExop}JwATvR!NaNObYQOp*{GBB#X%an
z;|P$?^;l|Cc)pp;N}L3zEx@<UQ?6g*Ce{EYCt$W*V!n_!EG{lGFN}fWJ`HNAUd>q2
zgGQq|B8xXLztal}BZ`w%Mj=9j5#gC%wraA_ZIhQ8EaJ6X{~n{o?)-cK`|orrl1aDw
zV1kPPGVp&%=Py|9zkh%oFq{TN6#l=4pU;+VAD?_=c6wm0;d^Y<Y503S@_eI?S*K~S
z0J&Jv<+9Q=aL&{rfjYLdXjXVZg!z)ff|P=OI?9Ra^LT|K5rv{Es=nCdnThV>LN!2i
z?~C8_Kgsjxu;kEH*C?{ul+D<=rJ*b=t4(&iu=}eiwnkptMQ`~HM`>_9jvbMkU)K<E
zSt#sBp*ANUk@2$KlNOC>{%!eCH+WopO|9;TK*v^U&)-mJqB98}Hve^|hrerkF7m$|
z5iubLa#<7z_)391A3!y|G8rcDnK-z$A)#);qXIRcC;Jr$7fzDU4r`%s3w<gRWG}6S
zVL^JcNM1~v87(G-BS;oS!k`mpDi-ZWpG7jyZbz?6jm^l#LEf!s=tv1)xL;?0q8<7V
z5oP(htvJaK&GW`=&{Vr;=rHjQ^L$=Hax&8iDZnT?oFCjIAYt5q&$L*oIb1m!wy`wc
zZ$mEf>|Lf`Gal%51YtYgiR`=To&4|Hjou*s;aGy(saC$e-yOdM2WPdWeN9*Rs+n^z
zBd%`0!4N|(Ei{FyPb@^`CKQ{5Eu61PR24*h^G|r>nO$Z!`cX-uDmCH0Ogd0YvbVN>
z>ESRJ0V#eSbbJ>#^c|Dqd-OtP_Zju-Y_>G33O5;z;2Fl2kr`Iu&#So;8cdwi{_Rk?
zN$LZ`q1vqjUDN%YU{V9rvhwUtEZM3FVkM=i>k}Ax_WGn{$R=%zMnZK}s_(ytP3opl
z8j1hc;djF|6NIsiD*Zq{)+e2+koI(0jMJ_EYuwcmnzVwV|K@p`(M?_mfz?yztNMqo
zT8B2PcUT6Os{sf9un2C)yF@!2e>SV$KoFZyjsxqrc5;BDlHtIT2q~E;Mhz!zX#`!z
zr-+qUBN_4TaIgGwX%B}`xNG?}&xodR2z=i_{<uc@E3A|Xa|G9g(IaRUn+m`Uu{JYu
zTAW?zC5w^Rb}6;7*=9Ie+b<gc7jWIJJb*~`V{x@k`%dq_={X|1Kk;$D2FMt#B}ngT
z$(u5B&t$18_vA)g-Fpt8e}mpHr<x~LJNj2Gg5ep~YS1w7a0EYB3sHC=UkDQgKUgyV
z2a4Y(C9w(}y8cc}>>xGw;(~=~jU81j6&b!}`tv3_6He#se!PU*N&+3VFUK^AV_5D;
z_p6^AjL;+l<EZO?`9SGFsX=*MjXa+d;IVTb0WsT~v0^M~hS`NbZbZ>%+7aA@U_yMJ
zn$wHLjt%$4*E?!4{~yF59Wf{e{93(qfnA)_7&KvxO|vzw#7Hgz-LL8{@!DO4<o~w6
zTM*Nl#8mItD&A1D;^(4et4uNW3sq1m?<tsgiC9TByZzNYcS4{s4!=968tstI3L*nm
z-lUYYNtDf#*yzrxaVELx1B21`dmB80OaN)$HAHXb6-^Vl1Ik~&`R?f<*GLVjwlEgx
zefa_)k`xhCwg+odtrZJQRO7OkX|^M#Pbu*`(=K?(_P9b9NVs1f|1157W;%rL23Q(`
zC-;3jIJAN=F-T1?pygk3<zE_q9i;&s$34O^eY%1pqXop&O?gy!<)t;KpmroAjpCsr
zZ8RhGBdPJLzsuw(p3Ude@5MeK;568Qq8tTU$&~TkZ-s2sPx7l32518U(F%7It3zq%
z4{#CBfxdpO^=0}MVCW+>U_dFG8tsof_52WNLsMDM0D%yZHBDKn?P)4vHaq6eu^4=&
zu+k1n!_hJSM{;zmBnfUMn;|iv53cnY5XVyUe;@w02h;HjX6Y-;HYB3ZXnU!_8oQ#I
zniMFHoORuf!|KyJNTxGix^6?QvR~7&FpO+oPt=T~SA*8VQeIxKJg%pB^zPp+uYIYJ
z0`J+-t~{$YW9&Hg4}?cax}PQzfo^nuRr0@%<;`yBDpXFs`(Hs34rVkUvJnCoDKQRv
z26ENE_CQ*4MnhqG{xp29^kBR>uN%4K&U$@+6dQHWCXD+6<vad6|KJ`Aj1BxXq=yey
zp7>ApOb>zkuLkfP<NBi~g?Cu@uEs*eFnwvJ^Pg`J&y~U6+|+J^4dOyPw);$8rL_XT
z24|3%W{9K~>rQyC9->uN?t;cDJQVkLc(GxRo|inKLBHZK?8dxAWzllkK(h+yuUgX8
zH62JDf5b2iiD`sX0*D9Yt%=I8rkX-tb78|@J46i_@1Nn+@^LJv{`Ev;_dmyPP6Xsf
z=Xy_V%+y-WxL>3MmU;>Kej?x-+2H(tCL&Qs2aZqNz;_zk-CuG99)rq6TY}yf)CCp(
zkAn^6A48UD08dn#_t$xoM+!8N@7P+78zwXvg5)=)fx1b0W3;+{WaWM-Y6UA)kiY$Q
zp@mTw@4pDkJJFPe7SjP^V(U+k!gyFu^^;q-7|9uBKG3;&g*SSF8Bgw&aDDn?U4e-`
zlh`|KCwiy9X^*Y3WAFYR_OSQ=H_weyY{V3Vk*_NCf!^eaiKpOiBI}xPuHUWFj?@4&
z5P}VOq!k3nxr^j7f)CVtg0-Nc{u9!d$F5=}S=F|?L1_71id-sC8eA+xsm&-!Mgu|{
z8Ijiu>EP1#Hwr;Jp@W;9-R+3Rdz8$ld{-vBKd<ut%zo=nab*=pa|vk4qWdrL`0qP5
zYUtp&VR@apV165;I*DqLuP#dEmY_Anl#~}aEl}eD+*1z40R!|}At&Rc?DCqOrV4bA
zCIf^m@}Zbra!3147ksK|EZA)eRG0e2xqrjAg8ilgmX`lXEn%0diaF1IJ2fHfY2J%A
z=>G~5q=2+M0<qo$s7ZXF_#-hK{ucy-Ru;eg_|?$xxI}Wh)5A>kBM6hWvF?X<7xMqB
z>8it`+P?M~W*Cqdx{(+ZL_{g+lvD}nhM_wIM35RnngJA~yA+US2#Eo%q?B}rNOyw>
z-*A85J%66_Jo~J(;yv%)YwxuTyV<C;mGn{L_xtSpH#rS5{Xd}_+$*-m7YojwjeWL-
zmwr--sMo9dZl)LE=Q-?OVzBb&<yr2CaA|6I>!|<N2MTS)tg#4ellLh<=Fma$ka)#`
z{CB<1O8TKSB!j7oCMvCEo2OyL*8{Xi-iTj_8En6Btf#v;Mg_sL8PC5IXAZJ>ZI(XB
zjaYd2NyQ%|C}1%oi1bQk)qZ7yS{NVbyg1UonLVm$y`*PY9k}0WO{LFzy!akRIe1ot
z7d!DkoP5qGsp4}M3CVz(X+|wyuwRYB;d<|JLSYU(Cnuvj4IH(2+k+M0vm(z}&V;cs
z<U!EEy@O8qTpoNTF%2x<&C<Hx84{W7#EON@7IUJtu?ca*Ixek~5@}COjx(W+WjvY6
z3)tjj^ull`FB4xrIcK?aGqyMgrOwyfjea3J9|zD5A<gm=Q-zsgC;>M^h11<|g9^#X
z3S*@Qm&>aiuX7n8--)4A!PermA3C%qG$h@K=FbnbR5Inp%0AKxTIss6DQ2fQ77sr$
z_AGE+{;De~w=#oyduMs()_AUtM~;0#u-l1UQTSCd#-UCn&R&mv8#X)@=!s^_xEqMm
zZOuF7)}6p=0%cfjjVM3=%CK&<AZO#we|>39ur-05ikgdwR=yXv=MHbP<Wmxi>iONR
zqq96VzBo4eq2x<JDH__77@VY3{lJ#t{y~JM(_To-`1GK6=39Q%A0k-2ScsV$(s+;+
z6>qu(E(JR&aT*-I0LL0i#!`#6J%NP9gjrTCdWWwCdgUxXoHx!9BmT59lz2XVZ_N3A
zi)3*xsmANE+!uN>=}~1qI+5KC81R3gk=Az#Uk3``H*cQOJIV^mzwRSr2yc4OWFPGG
zT?;}6?nFZL9o03L)_qu-UX1GAo7Biy!fCS(ZpVMRL~9&eGT<=5_9UuXA2a)3%)f0v
zy|)f1lx(^n&nh#{Q)-U2w;#PMHn`LJc8a(bn|RfK{lRN3&XsU%g)z|s+ggE3k%sa-
z0Dh>CIQf4Z?h>=|q3{5Hk56;St@Js9zmxA#>ZbFodlYXz?M%LD)2<TzS0^?;SUR};
z=Q@Daq6|*~hUP`>pnk&-?G~9{r04$6bg_=*L66NRTx}xQTwzbT3e^{7$V8a+1pafk
z+R1q#JTF5zRgJM5&@^<0wo6c@<TgE3wD?c@EqHkHpx=@hhX7s0o8?H(V5ew{Gz(O?
z5g?8tto&CBJWHRmA^8XUap8=Nk)lZ}yGX=DSXN~BH)NNq`M@pwh&>RC40{{{Qq>w`
z2KkU+#gzGFMRoz-S4)q>Zk>g7!3I#vIWH{&|HxSu7Ml>(kf3EypaN~}a`Je#z_9;D
z0g2^eM*;9~CW<$Q>t9UXrF;3R?b2#7u}!e^_BE(LOU2`%0`*0a#W}beQw1V!=k+ld
zepndpt0$-09wBn;%iAH_>r9Orp@t4p9B@-$6WWJQd=}xl+s<96t{p%wm~0~S|El5D
z+jb0sAx}?FjeC20TR7osKto`vE@=4GK;rhu=H#Pk`4#p>d&Ja#MnhERg&BP-XlnA!
z;a$~)^BHTqm;r<w2D*=I#w=pZb_O`|>u|op-9CmZgC(Q>2aK+O*q7{=yj`?M!_RVf
z_$fXAHqC@o9cSWs{Y`Zk7S#3qM(3zCwtzjoCyP|Wae<*_{%6*g+X<`_Or!`GhaDRE
z6~}D1Ewvm<gPCrjx>u6#5(w*^Q>nF^MY0BO9wqhlchs_A_pigk@(i)(STZXyMcOA<
z`zJcXCiFKO1n_q<f%oBu4vPwi-z+o#(ts{O>}pQ4CUJUBF%5tRo{~Y^C>CnUT@zoC
zC*wLYTKe1q)zYN1YfzbsB@&E`E!o5ij1P7kM9Bpg&)Y{g(0S4A<IS+#>RT8sl%74~
zYmfYhbk{n8Ai7y*swJw_<(L3x&@P@JQI*91K4?N%rp_w*(C<de;ztT52Vy=rETIBF
zytf~K$&mi_pJ1?H3J=5B*O-IicLc8Yc<keLrl|GGietZQW@N5O;0qA_zkasEg=HqB
zMgcqn2Q`6C#`0o3@|C5X^%eOQ{Y9BY&?2sp@YugD<KK0sF)#-v%qaX!n}jCbPpcNT
z51lIGZo^AO>S~=l*Y=w|YU02as7L?19}SfO<5-7=(IZF}&Pevf_oNZU#j5Ps6PHn$
zP2w_A^|-~CRmGBf-t$oy@pL@rAv&KQ9Fn2KtUyc8e5Ko|i|6LNABC#0ORCH7SHSm{
z%H^CiW;(~TTidiwQGf3kkmlbnUGp9c-IJFqMlN9}9d?*7lw>K7;1pVK9`OWSpR7|M
z$1rYW`i@_X5z!YnI~m>{j3@D3ttVT<8-%~Wb_!Lmf85=rN&<gU_yD?tPx8$A&fuxi
z>ONR-8|Mz_(<9z{mii`weU+eA+&(^#C-Ks>>jixZ*S+KlMnZke`A;QOB?b>&wqd%%
zJU24;!KU&43_aa9$-NB;?Xx{y<2imc<Kj#0zmmbfYy6<uV4XL`S9ld9>eLzf3X}#A
zC<fbF=Q3gmZC5{>>df9r0!UXjLXpBAK`h&6;j(ou7yvyZ<&dE9Qeusgj8iXbj8<>B
zSsmzX+YA#@b&$~Ed`ckAi+uL=4O8hHx4<LCuY?#?U?xGsG^KFLjJA_qPj%W;54I}k
zQwq&a6dkI&nDdb<W_ZHZpXig@p>=tn9M7(j^RDYVd+F8ki6VNh=S|DTt4G!$4d>f6
zj^xb(>8JQXEGq=B*H8TfE=8@s<&Rv7naT_#oB4iB+VumQUGq|?&Qefe^W+~lgn7rl
zrS%R+`MAaW4Rr159swi<sBh{dhHE^9cgP?z-BjeEWFf4U<~5ZN%pIT4))st~i+`{2
zVq?A^g345}y)y8cWV;L|L~ss3H{uW(Vb?hP3>G46+3;kIFRlF^s0bTSJ%)M#UKr3X
zplUdiw^QZ$x^H#48d0zSTC5(o>d)FHPHXLntBoSW>XAH5l{5qleZ3Ab!>{C0&HCZ-
znB1v}H{&txQpC8NX6Ngvq14R`da6#)UQ<FkzfW%4<~>rN6D`{^5MWX7M1Cvq^3a`j
zo>!uN1CWAq%DDSnf_OV|WR(;#=3_#0qHj89dBxs|k_aC5$f+-_t{fe`-Hm7t8#VF9
zk7qKol??B`YNWX^<X~(%DIuYGvO7i5YzHfEEz2(P+49OCZA&lucBEWur;=;ehwun_
z_Zih=pwJ$2zeJ~hZ;k=ILxQz~@|D{UFI@L@K7H_9+Q*14t3MJlld*A<9zi@rA$bQW
zkawfhQY!4!OuWe#1;d@PR9-RMyWMy(dv`V>cs6Vh=_DonF?c45a<OOo1K$MiQ{$b&
z?!gIkpBi<SmfsuNU|~!*)1yI30bZqF3%q3tYh0PrN^Eo=>eT744srB+Ds*6b>Ru?b
zm4)*e=hdW3uRTuc3pCADpLBY)l6FTxAZhing=&?RnWlnR;b2PkvrChTqv%Gxt+Z*K
z$CtwE^giH71z9s?&tjLTD21$>ji|FXc>P?l1ER_uUZg;B1IM6pztP#Wh3Q3%P)~g*
z?Hd3E`FSHLoD9t4E)D;5!X<%9{HjV787n>E9#`4#>Y?hNz@1|J5-A1EjQ}tPrb#6>
zkMQx}<T(@}X2|f(ImfNV5>4V&QBTuBu12b?7o6{?F_VvU(|<%89yY8i0Sia9-1Nlq
zxz#j0M~`x^C8e1OKyn~CwnSnx;O!-`&UAQS{di-ziMGy|>r?zQ);|k5Ux==Je|2Y?
z12?pm?#b+f74p0NKg;qTO?&0?9ake#9b^7^r&BMSgo1LAe^IJg@&MRC*=JkjFz67I
zfLg~nG135F8M-@_E&|V0vg&YB?o^A2U(5RmyaFl6|M|Qtn(MvByHHj=-M?A@=w~sP
zybaps{*o^7{naL);jI5Yt8q#2Dt=JKAEm?+mKEB_??gQ5CU&VmWX@h#JZsJ)ITlIr
zf@1V|S|Go)p_%a2qBgJAJ@H1{7c~bgQgddK$M39De~Z-MMz7HcFlT?0@s$ES30%rI
zOgZ0_&XeFYhnus{lJup_5UmQec`-%bN7S6urr4b5)<*Rwdx#p@oMxy*6vq7RO3)Vi
zC;5UH<F-m`J8Ym`^Hq;_J)H=o+E^`SX&LE%%E}ls2#NF5lY0D^$xnB@Qv1Ls`5cxV
z%U?&0sU{MKJflEbRlq&$24dtUj6SGvkdY;eb6zhJn0L5#f%UM9Ek{8tmI`o>pg0S7
zTBNGgvqs%%6pkRY_G4Rut%WEb6hQ3Pveb|ZXXc+gafiomx;NWi?E#Go)SLy!_<Dc#
zciThwpD-t8bxFN}kuHpMZn^ya)FKMXE^S;Ax4xREu9kyxw<lJ|vKJHz3-UX&F3RfE
zS31oO{BH89{AHxKF@KH&5TE9lOU->sDMnROhLiSd#`o5~<y)eTOZ2?AxG-I@+jX^B
zc9zH{3NR7?`8^XkTqQBFY?D+&M8wM_RRy+e+A%ap*Sxi%F2HnbylOq>n)BfKQ({dO
zit->*C}2AZ=%o*brcQO)QCOkt3yjNs`1fJWFqW6WHZrvK_nvX!izrZx`TX-ws+{Ho
zV)g7nCJYg>HdJ(x{n3D;zjbf#2MDR*)yehtO&Hp0#5gXgk=B_jq3#1$s!MM^1t3bO
zS?)b{IaPY8-~e=f$ju$5>iJEW<dq=;AgsrKm~_Ei&j{H)Grt^ChDVoViHI82u8Od?
zP>s`9sY0#M?9?V%J1uTt5i(JY?Jc6}LZ3JjiZ47;DYpI+c?zjT(pN7BbF{5nB=6te
z8M*8{>7yxwHY^YqN-lhnwAE}_&~N{lZS&Ua?;Bj#n+=3~^DTaovXP0(bo0)e*wrxX
zC@AF(c#HABOFlUpFEvVNhL6TTaWK%<oi^bRNZMxrd6h8r$%zr5JfjVJM{U%+CG@Ew
zmLg5-eGc<E70xTuG^yheiH+^P4%alHfZ^2SOF<xEtTXn3fb1ZS@U&zdb?Osm!>Mft
z2WP}Y&aZ8<#;Rg3N;zV5@*@EP=N>^pa%1;9{4GvVGb&##7)<qRIwpwP+3U+p#&HOF
zCa`U;!-FS$cV`>?M1R8Izk<g<Euf+OWDv7KQtZ#ft@|Hai|0(PZU*xi%?c^%drd}x
z?|YEAe+vsoGJZlknXfw8lLX80D*UBAlkHPS6k`uOu@(RGD&xJ4d_QSgJolXlO=7<O
zsU}n@3ic#OjPP?3)VwsBq)L~&;yc<AHIn$lXnq2j)!OIdJO4Q@=iUW4J9V=TS$k$0
zuj0>PgMc-d=~<&_;`#L0%Qct&!jE(l-*A{_>)nP^9WCZWDA~L6@}dW=Zv&sSu%PG|
zc!Ia7ZM3$_`)ET)rW+S)jAyVOyn)}xgsBJWK7EU)^ch+sJX9ebQ55Wadv{W0mB&!O
z3V5=uzfnff-A{{()ySp@=S|J&;<3FKC{DO@(%{-5__BJVH!qhSxIj;;8wsz*BfuUp
zC24$xv>7=+PV5!}^_yuU1h2xgujzECIKb!HGa(s5XKFhzu4vTp&c+XJGi+5t9V7|>
zaQr74b1xbkH{txt9|6(+@wt(g$6JBPImo<p;O<6re4j}tQQT44xqD&I$Lq<j8jswe
zucTS=+cVi<87wQL>}7Aka9WR*_EDp?^Wf+@p&Iealm)MEYbl!QFC+c(xZ-Vpsk8XW
zCPzI1^WZ3T#djY(lb9Z9^))sU==_#hR!*F{ITbS?ad*`Ht7Yf!1wLuT>E-eL@t{(&
z#>mJ9Z}DFH-yq<}JL(1p%C~@rJm2tfmj46;)0JjfL9IJVJ=Nb_YMGmL<SHjikneTG
zQ7n$xhID#a$5uHPV*iU>ut#sV-uj|(RyY3j!1T@wRF(X~pc-jNa~P#O{xejztpLQ+
zKn*#%ct(G}-eH*1s!Zvays)RJO+~EibY}0V#@xO!Pb;{PD{z}|g<r3BcJR4{Ti(+}
zL|RP3Mj0>U)5X&!71VAz6rpTzFIQ8%$fxFcUvQa_c)dK?x-O-RT)y&O((Xumz7Z7%
zQ@_zVwYv^=bu7e-J|IA>u`UCe-95hl{7(sC%jEYLC1gttWnY3FblWi+Q<bu$Zoyav
z8prlxY{d@vDY1xV7T9%I&@vC`E&3igIj!E-zIX#mm7Ug%mYy>0EEpqRfd1l)8-ZIA
z2JTuR*m-sNbK3uay-L`sx46q4#bMG7+|Jx{Cc<&l+$LcdA$b0zYleLL5wNf+BB4Wf
zne=U7nr5g>cH+r1U4toCTx2zXl;)bzee64_HgXH;xNxqYokI#PE{kV4KtiWHO1cI<
zlKj*5$`ZJbo{UeQyeN|XUfQztIZ+}Mm(@n8SDVxg;{CG#_(Xypwtv|A`Gp7>mG3Hz
z)60%b>lOo>x2W&~U|XRE(v;CVvF74yHCz~NVV6muTPD)4zieGx`mVO^%`S@edRP#%
zTiF#}c=YWdi#$>NhP1gq@6_Q}nhtG;RUDmy$FwoO$Et5EIwIu*jwe1EJ_v!Kb&o5H
z9++b(?!DM^XT$BFoN|I`4enNqIC%QnP}OVDMT}?dQb6rJ7n!@HY`>Z{eUZS}x4(bS
z6SLoh7mTbazk*Io!q@EhE0E8yz7j0A4q?^xGI+ZW7)Tu}a;j~95O(GMI5+T1&TyOP
z!~<<fuWP~}l{wpOTCS6`XXCyKuxK=w_1Y;A`o<YGAY{>%+s2G)ZD+H@S`F&Y`~x};
zS1r+%R>LV)0IK@CDwB47(1oW;HZ#t^PpJZZ326bBtV=`1hbv_{N0jm_fdRx~@&o_8
zg>~Ca=m?Mi_{W>lWu_Vz7b2O1t`=_QHl5x?JNbrqhZmgRliG<Xuhd=@>tV(2k~~pV
zB|X*c&E@ZtMjci9#OVAH4r$xkYuXV{7O7n7|DKW3^`&0%%6R55Y~mu-Y%{A~<231N
zl+L}2bU))*fGr;<@996=PSV?dq6OBxyULU#r2;sB8urwA9NYLcBY8WIekz%Xf%~{Q
zWX@5fXvW5$#a?%jFLx~V!s@7?Iy_*Yt=c*6;s(m6NY>!<i(D#&^H4{UTL<z=-oTyU
zdW-C<b^8@t7o}1o)>k5Ky&by>%VzxGQpUmDHAbh;{+A+<Y(Kc557a>VSx@uX{2tmz
zoZ~3IvOOH3i|BNa0ndb?N4Cq}f6H3;*-8>$p7}h<*EzG>-WovXUHs%vImtGzGt?r@
zhracJ)LA4(Leu@dv7Gr+FZSpZ+L#zExf|y!!NfLQSW_~RqbOh`@|s?tv9$Q$kL%r?
z1Q`#P!q?W<;gKF@P~>3nLz}lAPR(Q?&+Vr6#5nNV=dsONWV%5s`h94z7@gV;l{%;k
zjgC66Q0x-=h^I<e(W>^~H2`xj8!4&U&3(m`@rG7f%?suf;mFn7Sj1tscZpu~$S7bA
zo5^+g7G+^k=?iSs8X$Uo+I;*qbM4s3cJ~?ST|A5Y$t-12&x?#>#2+`(U?{O97R>NT
z2#<z2-)zvNuC&ijA-74;Up!bsp89z5@;9Nb4Sflw>z$mb&0~$bt#Qe-YyIa{PFDkU
zzB|+qvvN2VwTXDqM@y$KW?~dd{sEWhsEib4;QjjO_{7QSkkaenQ>6p-zg@}^f=rZO
zP^Niir8qqEwQN2w2}<U0y~j;Aq&allZ81BK(H_y_6uX(#%$6@Gk{qAZlVD7Z++SPe
zTMzX7?INaG5%mCu8MkOmCW7zQ5o-+7?3UNJ2uyp~-mu<k#%k<<q2{?uAkTCoCE9M$
zW@^;F4$c6eILjjxFJn`%f5fL)@kvrMp^N?|A;zccR1h=U_=mbOwC0c2WQ#Ilxp&?f
zkAK!-e=(3E(a!Qzf;-LHRVL(T?;O()4*2H><w;Y1d*ojF_onw@$QXbM?2C`IXrsib
zNDM5(uZIAOnok+N*b4`?ZOEUQ)|&YtYC^-eX6JIR@{&tyZT>R0#@YQS);{uKdrBNY
z5~I5<3$5`-hPO~?iKMa#o=IiX>AKkKU!5Qlc=q`~0T#~%^m)rsat!m}AqAwl0z^)+
zLnQ!=yp!wY1OKCXpJ4q^JipRlq{wQxQ!4NcUU27(JXIJ7LfVYuN`Uc3<{RRg3I99K
z<}Iz|Xbde#SIUR-OO48^1Mku22hykO_ks+imng`T+1dTDvd8*e00V$48D<Sr{DBx`
zJE6>vPn)7Mv-Qiv?r@gh_W6*+*;XZuk#rmFt3V~Ui}w3K<_V)Hn-QJM{#XXvyO<^g
zU>C;lBet$ifMlBNApzAaxKnWe6olhSjy$6M3|{g&ezPh_vD9vvyP<!MAeR7ls_qrg
z8uiH01oPbWEY3tG)jY4pgd3cJGI~dTmdTO2MPuWaXOxEuq-T9|<ykTERDCi<&Fg{f
zRsCg3-ju;n4N<o4;hR$(>K}CuY=fWS%ZhE?qs1#vfVm73QERrZel*BC2)g97E9xGK
zl3`STbXSn$YP-53;MI{ivaKs^=0d$-9qX5qF@EXHZDsEVo?UAtS~Q6K9?g;#O3yaN
zPwZ-d^00Nf>eyY^Z}zwz$Cg^9`%zDoE%h!+M%))P<PK>RW%lseihNI(mufOiq2iF^
zDkwb&S~q2#!`-UR_~>XSsTQvfZBMIc$s@d)p2rX35g2NT_a=pynP&SMl!hyy=x1b`
zzqFn@p1V~H%b6jRGkD)B>$;pNZkV{rh1Zqo)jzPEmlhGE?pr`xjh@dx-km$2UV*L7
zoMdjMyRQDuJB-q>@X^dLb$&tgnEMfa`_L_m!V-}?iVV!~eIrBVnp%O0ZMLmTh|GB?
zoKshZG=!17%w9YSqYDCP#faZ?eNUl)n5mIrnJ|P^k<lc;7!=4HH1{wR|0dyUK{N|~
zJK~llW1nrW&w4W8cj=M0+MM-LKx8H@EQxR51$z)NAUTdWLCw*#;})pCd~CYEwuF93
z8uAw$F?1&giw<%}32g!ka}0_8_XpBh@O>z-t2Z*Fp~<elryXPj`l%;UFIF=;?BiJQ
c|J=PH3DqG=n%uYX1+dq1c~!ZxXNaKx1MMff)&Kwi

diff --git a/docs_old/blog/index.md b/docs_old/blog/index.md
deleted file mode 100644
index 05761ac57..000000000
--- a/docs_old/blog/index.md
+++ /dev/null
@@ -1 +0,0 @@
-# Blog
diff --git a/docs_old/blog/posts/roadmap-2024.md b/docs_old/blog/posts/roadmap-2024.md
deleted file mode 100644
index 2832660fc..000000000
--- a/docs_old/blog/posts/roadmap-2024.md
+++ /dev/null
@@ -1,91 +0,0 @@
----
-date: 2024-01-10
-categories:
-  - Roadmap
-authors:
-  - remilouf
----
-
-# Roadmap for 2024
-
-Outlines is not even one year old and it's already gone a long way! As we just reached 4000 stars, and before laying out the roadmap for the following year, we would like to pause and thank all of you for supporting us, using and contributing to the library!
-
-![4000 stars](../assets/4000_stars.png)
-
-## Thoughts
-
-Before delving into [the detailed roadmap](#detailed-roadmap), let me share a few thoughts and explain the general direction of the library. These thoughts are informed with my multiple interactions with users, either on [Twitter](https://twitter.com/remilouf) or in our [Discord server](https://discord.gg/ZxBxyWmW5n).
-
-*Outlines currently differentiates itself* from other libraries with its efficient JSON- and regex- constrained generation. A user-facing interface for grammar-structured generation (it had been hidden in the repository) was also recently added. But there is much more we can do along these lines. In 2024 will we will keep pushing in the direction of more accurate, faster constrained generation.
-
-Outlines also supports many models providers: `transformers`, `mamba`, `llama.cpp` and `exllama2`. Those *integrations represent a lot of maintenance*, and we will need to simplify them. For instance, `transformers` now supports quantized models, and we will soon deprecate the support for `autoawq` and `autogptq`.
-Thanks to a refactor of the library, it is now possible to use our constrained generation method by using logits processor with all other libraries, except `mamba`. We will look for libraries that provide state-space models and allow to pass a logits processor during inference. We will interface with `llama.cpp` and `exllama2` using logits processors.
-
-*We would like expand our work to the whole sampling layer*, and add new sampling methods that should make structured generation more accurate. This means we will keep the `transformers` integration as it is today and will expand our text generation logic around this library.
-
-Making workflows re-usable and easy to share is difficult today. That is why *we are big believers in [outlines functions](https://github.com/dottxt-ai/functions)*. We will keep improving the interface and adding examples.
-
-Finally, *we want to add a CLI tool*, `outlines serve`. This will allows you to either serve an API that does general constrained generation, or to serve Outlines function.
-
-## Detailed roadmap
-
-Here is a more detailed roadmap for the next 12 months. Outlines is a [community](https://discord.gg/ZxBxyWmW5n) effort, and we invite you to pick either topic and [contribute to the library](https://github.com/dottxt-ai/outlines). I will progressively add related [issues](https://github.com/dottxt-ai/outlines/issues) in the repository.
-
-### Many more examples and tutorials
-
-Let's be honest, Outlines is lacking clear and thorough examples. We want to change this!
-
-* How does Outlines work? What can you do with it?
-* What can you do with Outlines that is harder or impossible to do with other libraries?
-* How you can perform standard LLM workflows, for instance Chain of Thoughts, Tree of Thoughts, etc?
-* How does Oultines integrates with the larger ecosystem, for instance other libraries like LangChain and LlamaIndex?
-
-### Simplify the integrations
-
-We want to keep the current integrations but lower the maintenance cost so we can focus on what we bring to the table.
-
-* Deprecate every obsolete integration: `transformers` has recently integrated `autoawq` and `autogptq` for instance. ([PR](https://github.com/dottxt-ai/outlines/pull/527))
-* See if we can integrate to a library that provides state-space models via a logit processing function;
-* Integrate with llama.cpp via a logits processor;
-* Integrate with exllamav2 via a logits processor;
-
-### Push structured generation further
-
-We're just getting started!
-
-* Improve the performance of existing structured generation algorithms;
-* Improve the correctness of structured generation algorithms;
-* Add ready-to-use grammars in the [grammars](https://github.com/dottxt-ai/grammars) repository or in a submodule in Outlines.
-
-### Keep developing Outlines functions
-
-Functions are awesome, use them!
-
-* Implement a CLI `outlines serve` that allows to serve Outlines functions locally;
-* Add more functions to the [functions](https://github.com/dottxt-ai/functions) repository.
-
-### Serve structured generation
-
-We want to make it easier to serve structured generation and outlines functions.
-
-* Implement the outlines serve CLI `outlines serve`
-  - Serve local APIs that perform structured generation;
-  - Serve Outlines functions.
-
-### Improve the generation layer
-
-* Use `transformers`'s private API to prepare inputs for generation inside the `Transformers` class;
-* Support successions of model generation and text infilling for methods like Beam Search and SMC;
-* Differentiate by adding new caching methods: attention sink, trie-based caching, etc;
-* Differentiate by implementing SMC;
-* Implement Beam Search;
-* Add token healing.
-
-### A more seamless integration with OpenAI
-
-* Provide the same user interface for OpenAI and open source models so they are easily interchangeable;
-* Integrate the function calling API.
-
-## Last word
-
-This roadmap was influenced by the expressed interests of the community. If it doesn't reflect your needs please come and [share your experience with us](https://discord.gg/ZxBxyWmW5n).
diff --git a/docs_old/community/belonging.png b/docs_old/community/belonging.png
deleted file mode 100644
index 7346f9aeed55cb0b72aa6274849599fde0c153fb..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 142520
zcmdqJ1yod9_%|%JA|e7(5=uFwfPm6ml^8&9kS--fN{~(glNN@KL1Cmpy1^i%r5RG`
zMsgU2VZJ@ytJi@0{@44xYkhIoa^16-bM}7X_tbu#bDpc*kR>@zdwk!%eI)X7*I@hh
z9R~mQe?58_{AIplngaZf=mAXjuYK9AXJ_{9W7;QwP5PFz{`}C9C|#3E$*&{x)~?<Z
za!1eEpF4H%ps3uDnMSn=$DGQ<UcVJtDV6MA9}Tix{cz6nuR0FRR35tqz3`HJ|D%w<
zTnARXT3N{1^7G5hOwAk}O(pBi5b^OYZEnM>82XFj<x5yo5%Ax8>4*D>_8&ZYib?8^
zi%;@Q%yP@OcKhk?oBfE7!o+BHe;C~1&2#V|gn@*xx+ZT(=6ckZ-8OscN+j*~)O!~y
zyop#pQyn{Ut8|FVTOVxYs+4+Q4^pKPZyq`b3Dnj?U;l%Re;8ZSmI~>x_=|o|b^<w=
z<jlJo%Xs(Pqj$d#T%UY)uN_X^y+tGqO;YdiAlpTn-^i#ub1Hc)M&-yqGV|6rBy#@z
zfk(eFf0r7p^8wxwxF@aQV4Yjqx4ghQ7i8b&yopSjJi80oYE<C1_`XBCs+XW4SISe#
zT^17ovA<F9gZ|XAhfK)X8lHcI_$NVK8E^NIz$|4{a!506$$vce%Y}Cp1$Ck{)Zlda
znZ2;4k^^MxniaR#5=8rd;YEu6m?kbS$j>D8H}pvv96bu#`U@qvC(C6NfMpZZiI}}u
zo+bv;H@HxKcu!g*0j*_RvC@06OSIqP9FyD(u}SYQQuukfS(sD^EYd{a?k-CE#!(_K
zupN(=^R&P5_k(CDz2iq=vw4<>mHvQZ$Wi-i*A5^0wqNQipg+HoGPb=GII9CF8Ce%_
z-IEeOz%9q<?4I&UwE^<-=xv)m_9r^N{l)wI{_uWTg^<zZ^b#o~L#rRD;?v#Ovb^w_
zcU6q+pexYOnp`r`&jR~doflPrtsf{{S=<w~(>H;wGru^z+b#r>-bw-7pXh2VxA})_
zDQ05LsnQo2bHhJx^eF|fE-~TAx2J3pfR@&8aT~GjCEl%DfRdC!9+^M1^v@GrVL(9%
z=_Z*!(Rxeft%OLpZ5Qzm33wARG0D;FyYA@!(`~<qTgnB{x-uBIr~f;#k4U<4--SFK
zmc3|gBtMl5pPJ;_i`(1#fNdR?dNFkV2VK8is3ZVp#DvZM@PD(qiOlEu9Xv^&D1Q?~
zIxmx4T2(IRFD(4PHn?yA&N!{}J&n+BJN%$s#RH&jJy37A!13*wR8??8{F&A_|HAST
z{GrJfU3#4LlfSpRaQ!XgBWQa^gcLW>W1)(Z(q6155<;27O*Y>>g?A5_eA?qOi_?4A
zyu-Oq)jkwWL_M`vjY0@oRDlrT>op+qxIcuD&I)dY-y1@D{!9DsH<<ukG)g@<{<nNS
zz1`1tT=2f_5bdw>x}XQFyDVDZXH@?XA-M1+IuCHg^w=?WPh8CaTy?a>?H0PfN%6!f
z0<Ith1AB@F3Zf9)dh*0x!oB^FfGfF|Lu|hi&-CV{l<YO*#P6)To;sE6d~GjQi1zD&
zWyJP$m;nBMW_wv7&_>Kd8VJ(G{7-lKkGxJW5%}aN(%IA2p#=MxL<;^5KThwUXCJ$|
zpOHlTjBA3JKUD)T;W8LxuI@$78Z*GIB*&AN-&sG}75mV|U|pSW?C0I3(tvJYwFT@y
ztMq3X9u9cOWbzi?AGjwfksyjdDLWH>r}5j09z3_TP!jA;();HC#SDk`Qs6@=s{K9`
z!9NE~oWfgPosAq<YLXxHcoXpetbHbZIIt(ZWgx9_7Wp4)*xQrywnvhu4($QvpZWJD
zVrG(Si0C(forq!|-iveZ(|{w^+vd={RB(a|sQYbtarw^z|3ykr07v*kB;*H!gbT|H
zw_h47AN>9|sU(oM(pJ7-4F2WOqo-06A8x+!i7Gq)jdTZs^YG`VVfp^tZ+A1P5ONak
zF1@hdfB4RMD&Yn_cmva39>ZH3SjWAWi!pl|I>CHqUnw4z|5qeZACC3qlzoBDKKY>&
zT2&Cxq8uDF_FU%>kSwKC*WdZaF5d7{6hLiVW=8qnyzjCO2MBg)`(E+<*z~sx?>dkn
zTr$-gqTEwnL_i#ePC4x9wJyB|k&16_Hv6}we=O*I9EgMCVJh)npikr@R2*6wPJ2e&
zTT;OJ5?#~#VZ^^*1VTW?(Og!>x#u4HfvAo%&HDUMyOjBUda@Ug1&XX+6rIQhZj5<T
ze&GlGzg+;I<A?8vWkPzXHQe2kXHOOtF}d6hDM^kO|GJ*vsV+Qykr}DJlCAJBPJWR0
z?PARh2)~;51<{^fr|>k$!J)$rzwq$Wa^9XGKbS$Y@=N_DnZ%q!mW%_vquH|G?%v7)
zTEzFf=3t(imfN{7y$jlafq2^;;*VA=O%cgC2~_5wSzz~+>l#F()nymx9X-{jA{E7d
zKjA2F_Nj&YO!iTrxOd}2g`CJ6s?>*FdqqeQ7i0bP+-JgVSF@aIJ_I)hL{7H)Z)|K-
zmA=XNU3}j*z5VdmB|*X3M@seeH&mUT6)&t$I@jNPdh()oU9}!aTe?fiQLK@=8^N2k
zIbMvst<gGWP9`P)(o1`dNsD8k=n)A_vG_a2FDT#6e;!+xBDwtH^47;<CzQ7q*959(
zmP;C%HQrQVts&j$h5wpjyK{&t9F@_zI|vd(_v1b1MaSq2aPI9qhalnbY)=Cp_$-~q
z{PEhq;Nz7`@wO4CkcN#SzgVxg1m(rB4CPY200IKG)R~T{R_%n3z0$mYUU6nXdu{Tb
z4}5r9ZFz7p;%g&*zN0C}Pqhk%;|RLe`!7>JQ>2ppHjSO0QI_8J<5p)mvC1RP2@Vak
zm^9SEqiI&$>RG70XJ7kLAv;v7@y0Kde8;`i`k_#9)$EdSWjbBCD3}KtsxKV^lW#e`
zrUrdxC<Hz9c6aCC%|T4j3YF(!S4`t~JycGWeCt$xv7STW5vT^vbDeI)Z_c^|Ssdvg
z@qMGeA+JM3_iSyTgPQu=pD##%eu&!$V%$L&mx+B~;9$MolbFVYJUm^?qecD7URZ$j
ziX>+Ht5Zc=o<1(w^T~G={(@ic?1P~bl{=oxE<>$<$+vQ!xHTW$A(;RHnL+B+E9ap6
z^Wr0R_X~=P(Y~S=5B_8*jZ_!M65U13zQcC*YGPU%WU&-CJAd2lxa=bcF6XrmoVJ%m
z@*2d%3vS$(<JRdd$WYm|dp#>f>pkP7Kai%??*Vf?ZqT7?5W52GXkuV}t6BFx?B2sr
zi|YDV(%E%t`|r`{7e0vgcT<(1gDmo>AP$mnUAH^Xb&IZU%7YNWyd}sr`TXSyCZtU5
z1FfGp_!kNA3n9cXyOi+LsYg|zHQDzcNJAe>ciQ)M+UbigM2BL<P|UTvN{?ti&LA;E
z9M&=aJanY48D5oB$b<}TDJQ<^8uJ+Dyf`=FD#sW+Jn`>VKO1V|S?_Y2@_W7}m9EES
z%ZN1}K_+e`Za*M!9qw+b(c}_No5(_Sqt)euz74O-i5tH${12wRtEe`ovhbRW$@Y}5
zlC>6__zgS^M5UpO(J3T1%XE4;F<jLM;YE+TjDf=@fDjcoZ^hH+dp)UT{t)B^h0obK
z*@M6<TFXF{n>>!GHgZU#*pI;)Un_VS0)mi3#H^B1^4T$>$|0duhqb>e{}=ueBlgRO
zgr0p&1l4#c*N?2^T2;fPABUw|>fU$;bH4G#GGegJ#&zJ18bjsc?!Gzk^nRyd_u9kK
zP|t^<j%hgs3mkm<9u<UYZM0LR)lMHXwMLOz;ez;(aoJUG4?rOgE|$zJ#rwc7<*Nb`
z(vHJNf|-!+<<@Ong4bpsmmu>wm6hVdWg{=tXe+JvAS85u@z8xO;6)vdtA+}@#>Agj
zOfzPME-8S5M=RLOzA#S1&M>zPdhDhh7RfS#?#HJx)W6!}P@zlzbI%-a%V9Q*!Q*q#
z#VJDtXXH@aZ~`Qp0Gs!(kxwpQzCd?guhd3tLQP+e_p>9GJl?ib0y+I6;2$w8k#>Kv
zu2aL-Db^WDFGuK?C#$7zSS9byEIKG7)W$py!w?{_tBXorYdFM!8KTlz%+S0IqfQ-f
z`>VkQUvyyP=DYSezhA@ZcYLO^-d`=Ga?*aE(Kin_Y;s?}bQO6~4zlr}aM6NkjZ-d7
zCA<@mOQZmNgy+Y+HtD6zPty*2)EeCsB{MPpc|%k@OKSI`LLHy;-hob}L*&vPnt;c7
zR3Jjw<gfYfX&|KtwCMN<$HgrU`xo*;zG;6|mk|m+Z=!Rb@iK?6MGyUwyB>~vOQM0x
z8aRc#e<kOAG(_*v1#{t7jQ2lNR@?!J*~O+~FuSSoCAUyEDf#{>ANYZj9!`g3a(+#=
zQe}y%-Bp6s1WXC8f%@bAUOi0Yx->LT66z$NdK|6TFrv)LcZm~egJ1!RRwxYJuDUOu
zV<<@#`#6;G6$^sVxi6QJx^j;^GO>SuufFZO!(Zefcc8f~Q=LyBj|J4zJrIF?)lA6!
zYD(nE6xNqUI-jgOmWJMW7V7T#>B~~u+v$Jf#wYy2?F1Km#&fcN4?L#h-wJx+AEanq
zFn<qU6hiREm4?#YaXP24v}kxcfm<$3YN-4;%&*eGQa)!p1PQdhslPaNHOSrOkc<I_
zCx+SZo<hx;7J5DrjOfE>nuBC~<=o_?#WDqIJE5%WeS;Ky4jt)HVx~#~3_Vqj?wm1(
z8}Fr;KN4SkWwoFqC@K<jUYk89sL0l;UBZ3gc1(ou_0{dkyV2s8`2`saT5jt`{Rt{5
zdSYD%6e16B=?vpiY^aPJwBE6V7bv(NgK=OJ!NL0sDi(F&aXzf$$%RtKrwq6q{h+G6
z6lII&o5e5bbKMkV3xr7h$gXM|YAT%0xR!a>%HJeI+el>~CTck5IkAT6=)0aq{pytA
z*Gl@7^%d6g^Fk)q3mgh+p1q8w7BaT#$ahzh<<dcsaYy%**z~+9xxbYf#pfoctf#)V
zDyXVn-zc2@Oid>zU~)scT`cno|JB|thfGWVB1)_F2rcrf2{g1<muS!Unv!HsuRHvH
z)V3z`HlSm8`by+msZ%8@p_198utpNqGM&+ka-}KmrS5We#SqwQ17*5qGyF<<lF@>J
z&y>fKIZ)3<WS=5@8-3EP{z&6l4-;j&u;-iap9S@q+*egRD}G+)1S73d^y^nt=1X^j
zO`0_7;%b{l-*qq8POd0&Wv-m?=ld`)T5^D@jwei`-gHdcaQy4!-Fw=?YOim@7Z&Jq
z?Ok*#E>{aO#ED;SEpt@l(ybC|PVH$I$`g-ZEMw`hMA7=zkp@aLKe_(ef7zkHsV2sR
z)3Bu?oNpQxL2X-`dRJS0uuOZboA;Mg<R3SIB-XWEE~T>+8qtgD=nge#P+JvFYs1Ci
z#l4J&9&;i`-n|vS8Uiz2D35&}YOlq0t)bGO1TVkhnY$19weme8`10gclb|ZQ6xMu|
z8?-chch<gA3*0l7r{~qpyF`~7S5@Geulm}misOcup}ykT-qS|Hw6{7o&v0pzKbS18
zn~4k#rfJhKbA<8}n=JGB8+~HCYV@&=8A>~nt`Tv44sfwzlVtP;1z|;*EB<hsKQ2I`
z&r|&sFQnPSc}!vDK-l2idvvJn<WqKJGqfQ-wp>F%wK1a%95xoOOUv_|ZCHApQF(La
z0DRD;!A*~Qrdur}v;yHyzn&cozp~XB7xC!A2n~uMHfZRw%n3Se^^_)!%z#p2{pU73
zR_U8C1Nmom{i0KKk8H03B}L@b6<c1{X^t-DzfpC!)9)!n{#o9%II&URYr5-AUsrr$
zdL}XqYX`6MZM7Laf|#iF$?uZHZpnP^sNRvFiyJ?}_V}f|6BbEfEn>i>0y%NxAS663
ziwUV%51kK$T^3-aile|L;PrXrT<UWfxgV@mN^)zp&QZ}_SELaYc@jzUYFy^UBS;&M
zCIv!Ex74I6bfThl=W|@2UH&^hL(Qd#N$s)<LDLp3as=)}jUV@Fx6mu`Gah!*C!#Oq
z&7WeO>MgTW%XGChx7^Oijlj&}lCqBMl5DAwm2CQq0PFC33>+b>OnW#Z1z9;C4#TW~
zG~*JtZW(8<l1Bc`(6$G6<W@weUfCCuLU}GWv}iZr;;^30)m|0N?dY8XRT;+`Pa|1)
zMVy9)#*GL}*`0TJmmSQPD;1#h%kr`<{<xIFaKewh^592LnFsFJj%DOY<XtT5l0F4}
zxnLyWM=IqAOygkhnWUY(o7d}7$af+5AP49DFPLf+NQ&-(5F2VBe=(FXoh7NUGSV#`
zU-{KR!-~VrDR2l>z0F%26v_6*?d}=^03~7Xd#$XjP+wV(-3;w*Ks!|~g#oBLr7U!u
zwr!kNw5&$4RnDb;p077>zj8xwuzilTir~4}83iRB`|>I#>dMN}qT8|qI(g`+1WY4~
zPiSP17|Z9d*7H0k(+SQ>9mz|fs!1i5sb#_<Z*{4e{fv(3La7x|U;4_$>))DyDpUKt
z#FsQsQpu&IE*=*{D_S3)AL7QSSq8&l8_))WGnuV47X0qg(CGPEANZna62%kt+6|eR
zdxxPNLR1OkZBQ39R#I(p+RaIupKc;?Imxi=*_UtBvo=N43$8}2xvDod0IB;rS_QW=
zHw{oBGh($X3IwxDt06b$CNkm1LeTm4)4sB5JvQs16J+S3m~dEo3hZyHnUvoe__&<?
zp5&0QehzjbQEM$LiLTyWj*K@=rZ^1I19-`YxoQt{5;Yi%Dui^D%U&aRR8Xnpo<jC@
z=!<htgiF$~y5JH>JzqDsM|x@J4SXCCvJU2Ew<37Z6;Td5Q#V;RG3@)n&0>w1S6R&o
zv9afOR!~!hb1hG0RaH$Qjx~`WD=V23JUv}1u=CunvMMWx94}pO#y2YP*)_7gNc*}e
z(l2T_lNo6=NJD;Jb#k+@ob&HSgm>ZoovQ^??A|xqWVF9$QAa_y)>{()WophBK9~s~
z<k!cH5Syji16Wh08sq7>S7%3pkHcKiw*dIp=6K}f^D{Lx+>$7)IS8R!H*sZ{)GPZN
zGmct0fiXtd=VOga$*}^A2m#5e!c=b4f{0!Z@KX|1%}G_);%#*e4KK@{^vmm^xhA|z
z;aqctYkf#Q_c^|aq1&Wp++JJZ)GqIiHo2)?*PuSR^!)oQK+4bH`KQciq}uGfBB@iI
zsgcWC<+l(GEBWHCLHB9u2z%{c<kC=#LsP%wc#dp$dDY=&)DZWVa3-K7`L8eBs_(Nw
z@7xd7spIsLguh2t+BLaXE-}}ymttOC6&WQSds3Io*Ny7he*D;fv>C%W)~Gho!sdeI
zQ5ai2!W-BSdotUvW%YqZeY8e!fS1_iyv!Yq#j5k_jJYCD6wm7QV8r7K?zN6myJJ36
zZ{^$(7hJ!xTWIpQaDRBIt^al~TsJE8(OL5iC5Vs!D~)G|vRyKX%8FU~+Lwjib>u;a
z2|NeR1iT|lBt(c9nL#p19J$=cjSVNh8M?%{IZf3_?Xh&jrR`JDX7g)i!{*c#pGOJV
zRreb@HZg-H%TwVKhamM;cWm7!-%|29CP|CyrDwKLfdD00?Wn2ywQa>GWWq-8jMY<z
zn$Ll<a0gt%4Q*Uw-rcJ6<R5G@Zs5iBMy7U={!t35%jjaWY*Ix(?gh0?{amH#)yS1b
zQWZZdF+{+t^8450JHV-LhfbCaY5)h^P59!*b;#gt3CtBqIn1Z}YRx)et$s|`fuXQ}
zMD4P};n#2RxXF6DgOJDX<a6&MyKu0T4`JE*u`gg-tL$=+q0G(>J$Wj<78s<c-F*iX
zT>>z>m(VZsVT1GV?Nh8SOCJ!MFq)*LA;|ihf?leV`n0hc=H@vlpgLvlh3@jFzieyC
z6w+M3bo;NrgYTt^w^yon5F`;BP@k~vuPiNx*Tk*Fc;(VsjacL$9|XNhvj=Btg`$)t
zI0qS#8K+pil;PV_tde37cN`;P3;J!_PVMtm06UY)dR=EoHw^PJszLgva?(0XR|U+n
z{5oALQQLXcRqJ(?oY$z#c)eWj1XSEkZkfz@m7%(eqn{w&UW_y|_6^eo*|JD6%KeMX
z+T++yR8Jrc&&vAQk@q*`cTV4)7-@Jbsu}{5X;C179XCT#4BW2D0BP$N`JFaK+?qxc
z0WIc1&#J8k8FX&+Uy9+k5avtoY9DTAXsfy?yt5Gg`YvaF=j7{W)$SXmeXgrt+PiKI
zOK|GxmHlryo<yhpp2WpYo~{4%-e<49WptfQR(S#r*u7_uZEsLX&QrV^f69R&f_kQQ
zbm~J&+rnN@Ej~Xlm!>Fw2x4;i%wfm`*PL2Ovc}4TorDXIo17;$WE(0+`uk}a@$<_j
z0afUfIQ~55leCf?GJ({?*wz(S%{)sNT>%^UxUvE#qL2SNejN^xs#*n|u^4RZXn{gK
zI(>u?ZF6N6PB5rxRd*{y$PNzB5`t|`-IVcfI$#QXXwMZGDs=E@1;^Le{AxySZtl5A
zG=~r~@_t$l0NgPRI|1m5!IX{ENay5=X)+vr?KrIQeg+5wmpNG%06@JP*(K2H{4yh0
z)x0LRz}A|LSgQ^Kz0TXUkz-x8^d`J>;dA%;$JK0CZmt?V<$kJe?=FDQqe;;d;W2Gy
z=*aNwRn($a2+}*vT*NSt;dxwHe`C#u+Tq=WCRepvzsA^q+$0C{*ExiT4X#7K^b~6g
zzkxXS0xJ-$N7>x<5ZEi*FZvf6X7m<<JBC6D26P-PJ7X;b^$b~Q7^}KDI?dTD79E~9
z(nrQ;+V#=Tz6`jP5|GL0x5+e+OeQBy*6wun8_vu4a`0Cj^6#lx7l;`fsEi8?rBmeR
zL>d)OQs*r{^9!))=4?9uzyOp_l<i%|{`7|4jl|otnRp)BWc$JjjYV8`#0G~DCo7GE
zQ<EEIY%UBVYE{8-2*TP^06d}a9E+q%gICUd1NivVqEKWT>CI4Nr3CIVUV}idyCT~t
zNE$8q9PI<Bd&;932}!ON@tOJ*Dp{_~Ps0pGPl{_%jB(_uY$yF*d~hqS>CL0GTAugM
z7|ZvHq9%@HFRxj<pB#l60QsXnrpAZ54D<0W&2TvlPgavgZ8}v%U0pRTol+A=xs8%?
z@$7@&VDim>bR63r1wol-fe`T*CRZ`}u;aq9Z3c@9A)(o4kVU7H;Y{k5yzue0?0=Gn
z3m9w?#bQ_}V*-d$2uvHf0=36PESrM6hncaPB6Yd#30?t3`M=<iXDUv!b6d@>hw49e
zx@@(oW=lxfYnHDX)x4n~k|TSCY|<G+Vlqa@{e=Q)+R!{-FU(75jasx<^9J0BwY=pR
z)j`OEon3AQ?5svni&@;C@XwUpjR%hNUsn-v!zLl2jH%-`B8UovBLY5NQ2-9T%KZj)
zeVxwB$!mQExW7UON)Hf}JageXRh(!~r~I3C0;*-lZ-%17XdnpOsIY2Km1|4r#fjt>
zWaKiWye`o>##eH#_(#^xC<suJ-n4Uydjw%;;NxWrvVi>a<EqZsT2sA-*tLd0-eam~
z81pY5w&E`s|CaV^Ph|zX@S@|smNv|sZMF;Tlw_B&^&>NUa*#nw31M<pyjvPw>r})l
z$C-q2FZCC(mY+cN0U;jfwuEo1uzHbiE&9L>=~|qMd7&K=$8H?=4c|GH%z{R6t*$>N
zqt9*n{5tBg>$L!C3E(bkBcty9UFn)0fFIE)u8(dKpwwy2$bhjmxzjs+e&1)ZGTUdV
zs}OK@_^gsr!<J5b@Ww`vW?2Qde%uHxX)D87Gd)qZAv=8tBFV6&U#6&j<wHbG3?z&1
z?)E3@ONN@pWz+_;ZnBwLMdy2(b!dDOte%r;1mwkO@Xp8o-l;vo;!QD^OUiN*-9H6R
zHDM*08hKqB)`y{A_zf@us!-H@iPBu%zEHP#ul%7yoE||ib}EYGWUgDvO|Gm%$;V(T
zk?6+9?6naav)1xCgJuY9hde}fM7YJ#{P|_myR!4c9b{Q8zJ@g+5!5|-j`CTQSys80
zY8C|rS<k%+O-J8{cgH^N=pMQeTfq_JG8Ecp)#!J3;qG&?s{(xPAN!?DYSV)>i}ILb
z1hQ;LGVj=aS-Z(+<Y*GPGMje1VICfCqf3@iq)*eb;QPu&^MUb`?DaWK)EO(z{v_^v
zktbpTm!1p@avj!c%wUHhb~#6#7Tzz3McCw!!pizNk$rWmrQgyyN^1!OvC?snE$j!G
z;7sr%CRU^t(id*1;6}+a6CV2<cA1$~SahO<7ZGQ-<}tv9$-t*^lj&S?s+o2mmptQi
zt=YP%=0g^lm4|5D4GjxxzbFxNHARMeHW^Z#@^aiLr4papmqh3B$LHGBUy>{}Rkj4Z
ze%8~Fkz}Q-Q$T7U8ufZYsa>r4inb|fkYZrH#-hg+LhpNmcrz<^Wg+gT<}&yz?)v9U
z#`p*l*veJ-U;sSZ^tF9qfyNi$HSR#|sxdFYNo<@i6LO9HuaMA=%oHZ%{SO$HZMh&r
zx1{l5&J(J30!n55APBkS?9e7Bvvhy!K7FrMek4{0WSFRu$0NSZ-cyI}zzl{CK_0sy
zSwe*3a$3h&<m6|xtwTbq;Tm6tFO*6l?|T6<UXA-iu&BsaRekNYL7uHN{jt<97b*(g
z0$!V0^OkT!BIuW9_%`mQU6Thv%modkss|z68QDPkR)Rio-8UUWGOHLj<hW__coi`W
zc1BrX7vnv6v`dqT0L%>zGX&f)wsmU%88GB@oJ&sLTqlc()ypMk`%^OT{@bVv4Rb7t
zn6{kw7`wHYe@`EO?n6j%AM#xBRqbh$nXy)fTF>MRD#8nz+>XK0H9(6II$-AmACDW<
z_;T8yRw%X&l>L=FaW4p-O|0b)5fb=TG4ck|5@lwkdQCxm^LXT1<H+4%21qv~4{&J#
ziapPVFzRH2CnvsDHgH$MD-5%Jal*eI5%HfSZsWrITjd}^qR4WRsRs%!6JBkks&+YH
zZc6%7UQ3G(+*VuI!b1=U=dBhdBxvXm4tYRAK{%ZJFEL8WH6WrxV?BTirx>uT>f5=1
z$yYu8fkGNOaU2wL<|qwm=uY!DUH2dT&y|kBj$b7t!Y=WW<gCTA&@cKB)8sXebAYCn
z&V!IrAtWK#F(WfP2swnj=T1j&b^L#^yffv-?x5YD(U+(HIJPB{9Qa!;%jY?R*eg#5
zI}8f#)1F!Yh^VKjOG8gbv0zijk3dR$ec<rg&fW1Xct*|#J{YQBx{p;Dlu1&oUe9x2
z5gVtIZx!`{#}W!pz*f-L8%8*<hlJ)i?~azly8Rew=obbEXbe)NJZ%C5^rUD@L#?%P
z1~bFP-F8;=A7hip+Y~V?!c}3vb0@WQh@Y&z9c#Tt7{uOY?nn)HFgSShmfUAqJgx<l
z5*(mZ07vtKmf`QUAm$r#@=q+ybB<H;<j#LoKFZN+^I`o~AQLO-N1g$vG${9#XeQ(r
z1_9w~Kh6XHy^!)f4)TPd7O4FDux;>+B3bep?f}58x&Ct|)?*5xo}o*%Rd(2y6g(5*
z)o~gfY*cUlR|@u9co7PYK_#yuHA6xNm{~Ci!~lX1@|BO9y&M+`Ya6*$mC-9WInx_S
z1tN&D*HK`j>(KX<($L0H^WV>#znw66(-H%<J@y7g8Gu}%zuJ8laI-5Q<gdRU5GU3Y
zEi+^0{?*i^e2;-4oZ>{Fw!#d@kWiggPOp5-qEL)5Gpos$OTP>{eNQH(Bu`xslxyI|
zgjXkPEEXaww~7eQ4|~ugk8eOBx{Xa8`VfC4&G-9%X#i|ROXUK`0#q}(Qo7^=-%*tU
zbZXz?l#@4&C@5w(pnnk?9*VEL4_s&6Dp1^ic!z8s_<P_2iJ&Xu!UtIKg)NiVst`cQ
zWH$+{pV>>E7p(m1?(X6eAp;X3@-O{Pr~-^SQj)?%`jaZN#6^k0866>EdAU3Jvnq*T
zSHhs(IlJ&Dz)v32S!QD86(a*`^4Z;Or>c6F9f!?!m1uuBlbj4+095TfSv9!wWeuLx
zySsrMya%~o=cF89%#0aKxKP@=A(v*N)}f~p650v@#J${IFZ;Z2fT5f%j|<=y3wA)-
z9x4EHMzqs3rvSyTnOIrj|5813Ff@=ZjN~l+Pf_Q@Q6d2VkY)lvEasVNy=qh3Pwc+G
zlor@NuWzeX9^7nlcOERhhe()IBETFCSCOY}C=;s*&EG7km=LX|q@T!i`y)_!rvsAH
z-}ke}ABVyCfv5Nf(m#>;&na6x7zJC$ssPmY+z1Ih4gERj(cXBQ`S}TF8r?tR8}|r9
zUqr5tc_2VkgnxU9%bqYiv~=|xz#{0SLoe(et-$P6hg91`zz-jA)O)E23@RNEKLF%Y
z3sS>J9lNMss{7FVXEg^ys??H)$V!eI{buxlYuNuT*FTwis$NJ3{Q2~=kH-9{obity
zON)h}4%|#~@)-Mn-Nc_VEhL7mya&0R3BxtOz3hKWU<u(JWSo5?{U)H9@A`i&>jP&%
zc&|yz1na4ldIPAu`4x-5&;h2_63wK5p+*78sq7w59s)#pCjj|28iKw5K*+2jlNP<W
zmTt}8!yD;;8(2g{#9eFaR?oJ|R$QDr@AkcW_x#2=T_n*Wm;}r72<#OXz0L4J*~QAi
zxxvZ_p+sKy>KL^Ck_XU+y2|eW`$^Q(EC5?0PeB%$=lH)(^7r+8;D=og0~_)GFOcqK
z)BmUC_+Lb&8hI;Tv+r8oB#GbWy|#bHMcD?)`cD){OgOaFiNLmQE<SJX=s4Zd(t=tl
z?JY%wEj>0+Qc`;1&bl<E&psj$-Mobe>wRn>-(t5kp+7YnRbepiAl_TrZv-MC1#qCi
zXrzz_w>t4nYO@Jze@+=(ybWfB)^%I~5c)yJ_6HEYGw@%o;`=>7cwp}HlK`+0Eye!?
ze&{5PoDheHLsP^hv#w;2TCkRbe_ds!ZuIN#qTWn6w)gbVMsAN*zXDOF;A8<0$EQL5
z3;o)?pPoOPl=gjPr6l6t;uuAX^%zyXK9r50K=Dm2*F%Nqt0&ZLKiT4=hYIGry>!V0
zvMbBRCF1@V*I~MN{~*_bT6H0iP-gt!TVt!{jQFhcAt(fo1a<%=ReSTb6dSp!z5sX8
zzykO3T7cW=%+p2g{G__eR$*Di&(pB;bQbE_exj{-qx9HxnGk+kY;VmDV#{)4i`2kN
zZ_ugMB(3fh*&lKJ?OVbtPBx&<`8)0uW}_tP-7nSIRGDm~__s=M_YXA7L6p9B_k57G
zI{VafTmdEg+Q4h(At@|KVi``&XR0VBjvFo+tB_1KYBA0ekhQCSWL4ykldKdv2yv2W
ze6&y;d_ZEo)RK`@&r500t9GO_@cOSqtkXA{FuWzkg;YlXd9?Ci5VpD-xY+NqNU946
zKKDUZ#!bh9D@vfeAgpj0IPAPH2X3l|z`iP4l`S~hel`ngs?xCz=@{%dJ(glgLm#ck
z6)h^P&ZRZq@b^%5<>{|fGL@qb_<xAt`{3xT-9upKh2#W$W{!e5xI>eCtKyT$Z|^AJ
z^xx)qu3=Bp+^#0Pm^Zg#FM9b0P}Ob08v{(pQPS{}zl|VyAEr2Ezm9%&9Jb~5xQxMI
z<PfCceMFA+O%*3YA9#X-<0q^nRZ_imf7%`8*cLN>CEI4(6yvt?$zZeEqVmcsuPr6H
z`$hYWdXmuf2a0tmlj^NbmoGiIW+`?DmZ@FDWqPBUHNo}kB1v{Q7rW!vBNEj`h*i%s
z84=vi=||oeDE7P7Q}Y`dXX@v2>ANmAM7f}qa_!sSGuYgH54SSPh(|wLj#Ya8?&?!n
zi@90NyxaO>_?oNB<G%QL+mXntlG_v160S+kpCi0##?V7#SDyqwTX1qyQ0zCYr<S;v
zq?zyPz@<~cr0>4+BI*jRXnl+(e<ryPFOHbCb;pooZw}Gq8k(<1M9_*^YR+#Kc^Iw6
zr__GT9ZO0>LezP*PV<Umu8BF$wdCG*>}@Kh-zl`Iks)WRo2MbIYpKSMyh0cD9+5b!
z>Np@C!L3Uj8Rt2B*eVE@vn1Q}`CXd0wdB&Ev+0)jP!X?_c@89JNN{88lb8Lh=z{5L
z>1vp`Z!5k|Bgy3B=RWh114fP6<)&y4bG=G7eb3qFCHO{8!*+!St`$=ao%(b-OSxtN
zh&;6uG<@fKGW99_Co8D)7h@Te2OZ3U+wR)6o^)Fx4|9y8O{%krp;w+uSIf13w;hFU
z?nK2@Zf+Wp(W|p9=2FxC+cW&9z9zhh(zV3g;Uo>k(VEEe$jy{K3_l8s;Bvkq6F6k-
z>o4wtkLPp2>g7pa+p+C@T<!1DCYisJwaqlKXu%T@KT_bul7HoRa8Zw-E!JeJb`YgZ
zS)HcH<~TC-wa+xN*U6<*;Y_Q)4v$32z}NFs$zpbBrRBlqBZf;(?w6kgo*i-#v$C>X
zMx|tVRaYN^7_nI~U_Dkk`RvD2Uop@x(A(lAd4gFKFE}2B>8mx68uH+GoQ#%8y)FsG
zTzpAB5TnL9HjCI{q4T2<Jxt2T;N~?gY53l8zzGw;_N6DV8e@{FTb6t`Hc&X00_(Ms
z&u8ED(Hj4y>T^ww*#}C3MgQcHfk>p5K%Q5*nlXU_Mz`eyr%;JjWWg&0(Hm4d&9Bwr
z@^ciTU*x$?_?4i`6kcUHILsF<k@49j<)S3ixpYpW_1$a>3JdG#Ms^mdMcj?WRzIzk
zptlryNVz2CfND8nj*tYUm<^KWBit|a>kOQazOxb|zTr1H8KIefPfgcjF*?@{({V$%
z=tF{AeSt7zl!w>MD-7Rif{x*`Ph@^Rtf=1{-RC|KpNHU%5oXU8#YeWcx9d}|Qzf$)
zQH$F8_ubuo<cz_XjUvqu0o|ODCcZD9s}{Aj=E(<?iQFqK(BLPZc$O1?>Z)vT>yzio
z!!EMTuAi&!>Q#!=DWUishw+K)USwWPSE~J|7%Hket=5F(AoFit$F{JyIZt+{IpB9H
z1@a^#MVMJjQ&^^{1wF%@$6wDG?c}%BguF;H9O$j)W3#LlI_vVK{pncU$aZ>|nd*ZF
zXSXrDb$tyw#*(oPc#q}5fFnkj_ZU9Qq(Gz9G|_r>xA|!L-X>*9%H9NACyCEz-KS4I
zF`Lu^=CF<Y=;A>NUa^(?j%PGhKfmO2?NkC*`nq;J-Z7{;&fbTKwYGZa(z)N`)*8r@
zX5>@xp5TN;iw;ECREL(=Ri22x871TqNU#zZR;pNtvE<cJ9uUwk9(jT4#g{FwtSh*%
z618**&y(j=Y#Zu2?eKonocwT{E!Kf=bvRa-dec;A>6FctQ4bV4G@1uJ<UVm2LCPS)
zbrKt#bL~M>qGplv9bNpCfz0W}amMn+2bz}Xeqqy*7*8B>dF?9tOl`v<i2U77u~gO_
zm+q0gCLc<MSnFJ+W{+%t6dxH%OOO!7#PxFujw3B92kyw$-M})2Sh+8;Qq?u<2(6tJ
zw05t+DHsGwUwvvCIZ?J&ePeDda)X32nc__)`UqYXl-gWN4lT-FXIIT6i>rEBwl9C~
zMED!e<c^PA@e-Y=(X4dG9M*_;{B&-)|LVz>DyKQd*5_Jie1<At%6ysI{Q5gmvyo~H
zx;hXgz|fc2GOfg3LspLn@vx~A)Tlk6VYS&Noox#nK@wA4lXY8ub8$#D`Omjkq`=#F
zGx#=X4N>^_n(5Vxitg{T0QdCr?R?>G*`*f~JnR=TpC}dei41Q91sfZTtQC*cRZu0g
zR1MqJ>lwBA20d46imaW$V{z*Tnn+=x(QF#qi?Z$Q!#(Dozb7KOuC*&ATvsG*@J~Tg
zVhAmM-t9)7m}Z`=!k2b_XSuUG?K#(*5~ca@wOY1g2L4Of7sRm8RYu%Y;PNLuVfd4h
zYzJ;%xKo0i_6a&ck%;D4sz|S;wO0akq19c=J9d&+zU58qC{^aAQ^^&6FRBZARh{qx
zi`%O%@d?kf^oz|ZCOjBFRw0_%qOo0Ja|0~-H-}wT%uzU%P><W8xZ7nSCvlC-e5U#G
zd~;P^wOLaM`O^8@;)d239NyHd;PS6th;Xiv&IuCzY#Tv0kezl>g6OZ^n>oldvC-7#
zPlZT`UN9M3CLeo<_GH6~pu{|vw@XpIrCO{D3s$!6ZTdqxERu3A8!xb<2^PE@BGdHf
zK~YEjV*B3isRT-eGgTE;Lvy4OY_I3vSO{G#nb#O;u^!f&moqJ78Iv=m)ZO;)^gk?A
z%EXGX+G+B_`dZ#sZM~Vn#2OuVfBe`;^GQs5tWCg7q`t2GrV+bxP_fr?2C6MInxx^x
z)9tGvE9o+xA$9|$c&fHx;mWX?k_Mfca|YYc4V(`CD|Ws^uhgj1swnV)Bo;FtEPjZ|
z#<|zcuw1zEN<=e)D$)ZjN~dd96P0Vb_MF{pWRBw#WJ=57<e#3J@Ck!050URj5{21X
zrP7U7kqLAU5<J8Pnx=DOYowPXd9YjQl5sQ&2GIx)aodcAu`JnaTp+GGem!d)G<tpE
zG{t)K0x2Wv_MgIv;%RLp=11}oL(AyZvBxP4S|iqYBS{(~HcN)4NFF~<FLV-*jcdQo
zNk96HK8uCEou%e^eU8T<YRYfGN;i}xc}XtjIAr<2<wX}Mu%6pts277S?+~MDUc+i@
z3^6*|qvCknhk3I$MFc}J0_{atK%QLD_<}K67uU&HUtxxrF^;L8?}?wn^Cgr`Ev{QR
z_PHVY#-^Pj*0{2?N;!5`T5@chZ?KUc@z_|+ZRi;=9x=FxEsa`#dw_2#I>qqIYMX2x
zs?ztDb1?$667ztt8eKO`Y(QA+C)AB|F%Fc;ECcrJbIlWs(Tb{P9}GfJ#nzSMq{|(d
z?0Gq>zArL%zJ_UTHTukB$zNMixzXd^&Ce0Hi8$a)moN+E1LZ7U4R4GOV9gj@9D<);
zmDqetUAkPNd1(n*vY~7S;=X9bViw;*))6yF<&B$Km(h=JSb=vQ_&=~ko2xDYIGHpJ
zp`R63d1XDw^$=QFni3zuNnawR1<ro^{@~^SiLe}1as@{?kPaq^vF1Hh!qS?iC3SKf
zmGaPP+q#uLcH@I#zqbXVyg*jR!eSdkXJK|?g7HIr!1Kys{7dTU?jE{|k@jIS|H;;d
zCx*ep_8yKs;k(EN4P-M+*OPt(L<w-TNr)7U@h>`_NjJ{do7L#yHx<kux@65k<Xm>X
zJRN;Kvg7LOMhWzGvw>;il?~!7=NWDuY$^1HqEIoS1ZQEm3L=wI5lToX;WC6@Nq(?&
zL(a5W7rmN){+Q9(-|`KhsY>v!?OJP2gc&IPX9<TfSr+PZOi1O(jmB8+upBrTmC2EY
zUQuup|EH3%{QBnXMt4qq9l5YSaM1y=Uij+kjLF5g`g)_4)=opz`^UBoGbxVPH}oE(
z5I)1^s0)r+1<K|-rgOdcmHDXEO(S#Vp~N9KudkA*i=$uWa{RH`C7JaSu5`)sC1{D|
zG6w}W+*rM2z@dCIv~vrL_5_WDn9zv1wxZ~C2=x_f-Hk5nj;Fbp`-gR;eG>QIrm6Bo
zDmQw-#D0w8drsx;$pvCSeeKMhyWcAq?OraFBnwKG<q8jG!t1&PRcsMELc_>Bry#b)
zV6{FUBsmo=lcbp9sFj%BjS|$hPO!6q;-Kr(yA~Gy3a)*YRlU-U(}CxYnS>^lvo=99
zRNAs+0~r}*J(*ifS9NLzdaZ)=E%PzPV{66-)~Og{X@s}aH}(0xXA_M~6XSA13QI~M
zq4OkmpQfrpu9qweSNl;hc$AohoaY{19g%!C86mwq-6CAKrog+H8$5g%f+6}OPGvwH
zJmmTyxR)#CUC0aOv1en-&p_E{Y;~IXf1O1V%+0EDm0`II7u>jGzg{XC$VxorfRY9c
zweY6My6|dA=Kg_xQ9Y*-Cl9VVD<M|(9Br~l+?Ly%aE3Cib$QdAQL#tx$AkxMf>d?+
z!NVgn+tR*IYwj@Ov<ptqQmh)d^H^`s+4v3XAWf55yrMm7>Vw+!$97_K0@xR;A1th=
z<1GB670*6sG$&-%n*28Ux;vGf`9!K;UR2x2o`$n{%kSj2$x6P^6S;@sBpw@GvQMz$
zcTWmrjy4J9f`psOG=4D<JqMERa|yA_w+B0FPedjp^aQdg1{gQwj+e~wPSuAwWeF_I
z)cCZGxU8h{S+2=^4|Sbms7ewT!TrUvBhL6fB@M3St)$S{p2n&Td|){iBZM-IjlvH;
zu<Z_Rn_eEi^mFU)Z&x<0Pbj;UpMf*b1;UwbkKllFXUphAw7UqlGe9&uFd9ABVREGV
zC@j=p+`hhW`R+Rhc57UW^BmbyThPSwBa41mn=(URTM`<Oso|zIl1fWiBe?aWhU7fg
z8!71u<UyD8zI@P(u<jY<wvv?TN;M9-2IKQVJ8Q!E0*t2O8&ZQF-6K#=8D8xr&+(gl
z4hUTD@Ej{uZBp@IfwWChU0`sr*Ng?<-RTg`T<4p*+Y2@h$905`YJ1}Rooy$RayDKx
z7z>Lb@&*=B=#GOBa2nH@i#jQ4>Km<aLo(}bNoOawSxn4m-;z6T%n?!mNYBgG+g2_D
zJR6wl`h0~uS^8CCJX;%(_99cI;>bm-kWl*_Li)y28Se39sf>AXXYSB-lF&~kvBDLQ
z2yV?!=5<P^Sp)@zsfOkyW}K#{+ikmqt7y1=kJRVm%Y?15V^5k2*T?7noP6<a{T5g`
z2!XY?1T<}53!YuuX{!YRRYph0Ld+gpxV%E#mWf}$s5M2@q&pB@2Qcmtv^bCPA5Sp_
zwGE@&#r}Z-!}n(f>Q1^X#*?u0WNK4tAlyt@1T>BZ(;MK6`Y=jIj7Rjh8n>O-7Gv|8
zyhs>?#EN3J%AzW8t*Ev$EnbknOPgW!HAwm`0<%5LlBdXv`z5eclcXaqdKF*0l2Ex*
z2~C0x9p=|f{gd4*$}J=2qir{r*^SMR_TU`@a;M=lZNpv$1=;%av5KI$k@y-yEppFz
zV_OG*|3-5e-XMO9R=_;p85<eYw#eF1<O&Xe)+uxEZ_I-;iEEbFI^p`7g`!{$efM*9
zmJ(OgLiim<a7Jk6o>oU!)2%h<waK~sDTjibcn^3$R%Mp6(#jF^YcDSY%ls+2JR4S`
zoBSCpE<B+|Eq;`_?(Ll7D@rJb;=Vaw#)cxdndkLm!1D%9`1e}v8);|o*LOAdmEa&t
zW@%fOuE&`L#8+G|TEgTb{OfJ11<$lZhj}o_1kuxx7Y*^ep5G{wn{}GPOx3EQw;%Z0
zT9yvXjb<@!s&F}5t2cXYzKjx{i`FbOwk<e8b7rh4>a^%$z731O<Iu5w1cLGb!u^#j
zG0C4PXBqgqMb{EJp>bYbM*@TGhqqe}n(EXHFKylzD!CxWP%^=|yy@X&Jd!OF658y8
z0`-0U*B5jSWsW*?F(tUXRl~*k6u)6L{sYEnT#%o!OI~&m(tom^n<}2XDUxe8Pl?qR
z&KG;b&tqpH={k29k?wN2fMIQigO|5~Ojenw$^3Xj?nu2PeM1#ZZO}L(@W{?G+IhOn
zqI}-8&1jfD$E8U})FUBI6xZQzxLHSMW>%FP{i1{4UheExZBAY4ax6vV@WNC)U7I4W
zgu(^<qdQ45_2=*yBz4tt)D2v+UOML$+ofQ4uA-<Sg%pXNOkEl;%w+$>2Rm6-A9!+(
zv3U+i9lk`Bd^kCzF1lQ~j`w726EFF4+M$#QSsoU4d)6RQufzNd>@g$n&>y1;gill;
zJr%eK;wvM8r3-&y3D?c5>20+clq_d&NIZA77oobqG%;+!8$^Y0;c|2xtI9?7h}MPN
zVK(e;oFF7#hLc0WwH?=q-Ol14K?Z7t%I9f}*cDq_W#mmoxLi!Y@z?l=<*oxxvZp`V
zJKf@!lg}1#oo8DHX9z(`LL>Z7PB2n}1MoRsgBB*%QaS$oJ5P%jsA88Y9yr!rRehk4
zshh_#DX|XDw3=E_T4Y1?4VG~F4ObmY>}%orkG5k;4ik-0l5Xl^u493GQ_t)3&O!46
zG-JTLUPAna<V7~6A&5*6O{Cr{PN!rNSm;K>4qG{cqqAe$o0rToCTMf{Aeax__<3gh
zhA30f@Yp)vd5ioap6(McJG6-TY**hsbZ`A;R#5;8H8g@tr_f%jf`95UAt6~FJb<q6
zIJ7Dn&PFr#Xh9#GWexHXP9*4{z%_7*hMA$>A6o7RQ?3~+9*KyDjQ<%UQVHVc4=S7n
zvjMP5YhO4V-Fvh}CXj~lh}Y*;ln&?}nla2&G~Z$#OmQPT?oGs@GG-7&oj7a#%jcIU
zK|*#bAg@jIG#JDI)z_o+I|9;Bl4ZT!z8mFjO=t>WAAT=#)1S0}*MV!C%)e(rKNl_s
zAtYNTz@2B*L5FAWJH0)*Zx?><kE{=X_xC0al7S~02@QwtH(;3H$tKTkA9j%{{^qr%
zzL!I4A56jg!)W=XpZEW6@}TeILFoGsgWmXmG$+Bw5j+`>|2Re7wcSkww*7yyi6r#(
zAeh(S`)WNvzsqPD_$-p?D)R1nP7%mJ*Fj@?*TyIz#}FY*E1+M3TfQCa|J1wwC1d|T
z&@cQ+3IE?{zWrfp|AUTT7eb2dyvVFg-}<Fdfhsq{f`Wo;4<0<&a9ea(oEmrgqM@p)
z%3{!fEd?(vDYE77OpQBy(HISN<3)>@g@0?zX>F`)g0@fXf3&>+?@lHB|5USy(EZWR
zrc+@fWxIUvN5#tff2<EX-cmG}9YRME6jHcUoqz@2Ff-vl+UcLZwZ}*w6Y}(X49Nw>
z*XLq-?pj#|78MoEqjTc!yEe94?BPcW39UX@wNo!>G_Ga1gyivjn|+A_QzTuAZ}}|+
zlbr4VrwXpCu@*8*?|*2|fzT39GNe2K-jw{$rnnS4XyW4}D85&tyADa%98`2#2uyf@
z%?06v5e$&M1g`hbnUC*Q45;`&-;&=l2<adrYY=&cn+InD`99Qt9)A&h)~wvy4t<AP
znLaN!z-!(JZP3IszjMO>?~>^stKG_ah<k;#GC%QcAmsn0lPUYWV~($x2B#0l>2$_s
z`}*tjXPorwsy+3+?fXBNpDC9ig|W0i7!U3~L!<HpX!AsYr=1^w<~dq%Ir)af-RH25
z!oe4;AP2G#afCeh=zsL)6A9lt{C_Y&;wrRWr$L6XI7Sm$%;#?WM^Ck_zJ)i;MkzSb
z3;hd-AiSM@L9xbG=oY7HqRXemyE^%7-Kg%=xlQh3Ki?p18yMGFek+;%jHXvV{vH=?
zH{z{@6N<IUr9(LX0KO+JrS1c%xbAZ7W!ze)5q6<?u4}0|T3{?W>_+zH`c~P8cUKSB
z&#Gl$xAI8bDzl0P`G;MaMU7^=()e80rtjXn*El@4P&~pQxwEWHyHQ1lt2-&N`FV4*
zUh<5x<hs54*dy}u__iF}`tZ|0>a+FncK02fg>|=71|Q6I;v7=NgXRv@FHpuRp-NQC
zrx04yQ}G@;!BZOP@3L+B`A<riMp`;eZEsFSVc(BJ9rVyR#Tb6JVRUJ+F@D}*PEBAe
zXpSKVB`zpva-Ki>u1y94FVyGyRB2R?(aq>)KC!2VnHYAnBF3fcBCsSV+2xgw%wxWc
z+friS!nP4I!ZC5n>e!n0TLwraTc*|#+y%LrXU`N{m|V-=Fd^ArZf)v@z!)9UH!cXH
zTw|lk+NyF*C(=vCwk2Pe>4JLr`9L+>2ENNIzt?MXYi_iPlSe`DOyI0CILGj8aU8U*
zzdZWvgHf$t_;Nl1m$@B8`?+~sS-t@g;p%2tzEEU4Yo1c~S=v{`fmLoM%X*=e@rvEG
zIG0IfX3S1}r*iO-MK7lH+Y%1%iTGTb-<05D>PL$#7@88CHr<?p;~huD5<Cp`b-!ka
z3Qt>f$y^=lWM_yf!@qwbY=F5t=<tCYRg6^#v72Wfq0q4wuIdzL#gn|TyACE8$}c4(
zToGO@w@}a2EK8rRB!)q~SR_`E#kFj)fkE%z%t%({DF+}fatz$j(UIMrFUT<J_G_))
z4n3&x+-=c$ZHH<!Px8i+9g~i=q(l&|CN>W-;_xBNb~QX0E-|g2z-N3{f#pu`+;Ex0
zHYU=k=DkakE6Qjjewfh*PL}kf#!OixH96LEF%4t8!@Jn`IWEU@>8oX;x^qEcQN7=U
z#pnbTRc-STgMg<?@$nvqGdu!@o$t=)_6p^;bRB!KEh>a@YU3o1s1GA;dG2QIBBFlA
zWhX=Ywf08!n01(8q<BgVTbNAkr#rqzwRvjGqnR3`JLL(1DP~o=O)t9_3hM(H$rvR)
ztJlY>asBmUMHyyiDiLEh#95w@&{5<k#8G+eOg!;AJbjW0lP|rp%~mLa&eRbVT(Ht3
z6?ZI-Dw-Pqh&S6Rw23<Xv|_Fh<z?!*xn7-;t-Lbz8L_>_&SyPeve}^A8ZclzS8m`e
zCoV@qr%pK9LN0*uH@pl*^4ii?GQFO`K?q7RWD57vJ<|8&bMqtgq4ck4Am*HELyoct
zXJ`Pm09!(Wr_G@4Fe6Malhz^r4Q|8sZHh!><<8Y!19G&%iyczWw8L39tI4MQK^q6H
z!?r|q`i8_ravvzeQqeRJGyJ?6pJ&*zZBtfpopRf@d3Bn(I4MgseYjW{r+W6Y5&UAD
z^Y(K6tHMHVHw8}h%xBmA&vdUYko4tNFL^BaJ5`KLJFhgb11sJ!kU!}Ie+H&k|N2PD
z*;8XOHLSBHyc8zMIgYQK&qcUCu*_ZWyKzohz7~>XTiGnvfLFlyx0Z?~I8P(jonjf2
zS#0!|r?8_h-OQXi`p{qFN7u2I`8&7-A2?(n&>!u-(t_HMkBWTc>fxa{VB~*!O}yNc
z7aO}Yq5{4I){k#?5p?e}jI~91Iff?XfOq!DlbcJ-9CcmZdC<YjGMJEGU82w(p__B~
z9f637yjS;vp(}0gZ!-2VA0{=2xCWZ|Yt$Kdxj8j^yj;fPOG~ei%X-~y>vNoa#8bIc
zy^1Q?dVeHCJy>SJxy_^KIqu|$Ic}lF!}Dg`YM74Su=OI<0sZJ|F;86Ooazgf{Q0Ww
zp$~cZAZO>T`AsY)XGcIItxO%1>ejQ*n1i2@dVKZeGiy~?uI2TF^bhpzxb)q;;tj=}
z*PM{)l5{<tycR417XVckRFKe%@Ep$M1%ma_JL%~L66SMqGmNBiSEQkWv;Dsh6XmYS
z;GO^Cko2oM3Og=~%fhz}J3PGZVBbzIa#W7;=Cy;A*DW<~a`hR>U#c@OGL)Z>={s#P
z&q}K^Ul)4K$nerV#d*E?N=}XoXF}Mn6G33x)uMABipy=B7|V(8@uRUThn%Kbnnj%^
z7Z$}IEJgFvczSxq$?;W*j?6G9DYd-u#EC>jN8jnq9?~qf9uoDCbYD4sFf%JF_ud3l
zQ(MchAryJpQ&;5Tt7ADnry-`3Y^j>vdU>KEH}yJ_F1rQq*F*8zb@kVZ-;g%cO*dS8
z0^i^t&3VB0BG2dp>!&n({^LhcecG=qUCeHM<!#uB$DOB=Y!UbAGK<!Zyj|?=ohT*1
zyTD39PDXppQbp$Q7YxjP+JFhBZ#?nl^Gc<T?dDwM>OK1Q{z6Xcz`;S?sOZ?gs9qSF
zn(lW}mQ_FMofnXh0yU$eoXnrlV}sqYg^Y~_SRbIMxpk8aGXBA@I~gH4iV0tJ@HLQv
zL^MJKg}vooO4)wQ?q$`0P+KnDJ}`1`5rf>i@OgV~bPQ_IrD^df@1W~Ri<%xxli25e
zjBG%vpx_~n&GIQpi%uASuH(C#5!|abMe6YxFI~lz|1YM#Ixfnn>sCRer3Mg?8gc+Z
zQaT5uyHmQ64k@J@28M1Jx>FiSN$EzAZjo-dkMH|^_g?;n-wfxRz4zK{t$mJrq>NJ5
z#%T$vv~SXL1}(j5o84fOOU*+eCyR|v5}cmnqYG5JBaOv6xQ4a8?O=-``v#SjeTi<X
z1NU#@-(A9({@3o`${Un)G?4q5Yc-S(I|HXsjgl3)l8mAi5aU-xGi6#`-Q7sT$yK7#
z(qU_HXB50h9Js%s5m<36lk^$Vu)Wuch=CXTZ$I^i!@n(yNpiJ+3zmlDEp7F*p`bZ}
zCH)!xMu8q`5?Us^Ch*EM9|u-+>$VvVSH~Jsqay#N?^U>!Y3@#a!>O$`%rWZp8yK0U
z_^p*`E*1{5_gJXWB3CcY<CST1-S;ySoU3ffFA4fzFMz!nF3bel)CV2P3-=a#?xTTe
zr-CO|!Bm^PWyT47a9v=S-2r<&o^7@NFt@(Ym2^psC_-7Z84B?xGAq2Gy^K_zjg)QU
zO{|t=*OSYHM&Kn!tr$3gB|q(AxEfz|VXG_-!|<}0I|Yh~(1^m|^Tn3ho!Xf`g}v#l
z4~EJ2OZQVcf{a1&s+2xN&sJ>;=G%{cITOx}NizMi%;!l|?o1U&K2}J~XU6^elFQw_
zu4jiP?Y9sC6}vEY=Ucwq`W^eDig9mG8Y!glx{ou5J^Ui3SSdd{_f?r7*IV#es=T=K
zfwcN7Q~%@D&)8aHOd|e!{m6RiqRPxhAzi1|sDJe(WaQYT3x@e}?{Z!q7G#pKZp<eB
zIQ(RKJ2MZ(82!{6wuXj<)e8tG?4PH1{S#MjMj3J6tv8e9iw%W?^2KyE$$|fteiPV4
z^l817C<4f*=D-N6&l%DXaL&!Un}56a)b~C7zq>axLpYrD*8Ee2elvdPt3l?ukIOoL
zSE8c7XYj4|3=d;6C}nl$Nk(EU3o)p|<-D!Is*EWjb}isxZCMvFLQh46m0<+AFt%vH
zeu1{5!;D90+!+(HbdxxhDx+iODHrn0zL&`yWisw(KK@R%-152H&QCdhkr=?2`TO(z
zeZYGd3~V`T@)k`2QN#)k?mszk>G{xpP)*%Os_MK=RTLdvf$~zxsxJ&#!8}P`Ug6#K
z(sq<!&Sb!F;i082?vs@5!~`WbnP-ij+S@CJ<xfXB&S;N07<UYKNW_ZEPw&Nz&%N_@
z-U@jPGGYeH4&qG|Lzb$XqbIFa+v#mP{Z7z(rG#eq7!>ASPgMnrA*M^z$nvG4O|ITM
z9C|KCHWi&2dLhSr$pzM|Px-~_@&&M{;lN5DzTy)Qv~K?l_>^DBs7Lc<CdtR~M3D99
z8ub<3FXdJ>V%|hZ;K(%BvY-#HhmtMPX|2cGud>nU-4)`J^B}gEWG^qzM=p|yZ{kjO
zR$)I+Kd7^om?)ciqD%+0%b{;>n$bW`sU0Zz!x(V>Y$Whu&X4M^b|ya*4VE7stu0mK
z$!c5%RwQ7`kI2Wt;sdT*J)91-(X#sb2Xuwn>XVV^CGHm;rnNyH%;Cjkb53?gO7Tso
zzO$HF+&1f8&K!@ad>SRAPaWyVoS2w8k9kr>Joh1}$*flPRaHhpWf~up$Hr!ElbCso
zTBm*#&%Z;fSu^HsbLKnLH&7|?l8bX}*I*YgD=eTo-Ns&?|Awlwx5t0H(9=meG*gOA
zti+~G#2j)+tYc>{(&CKnF_tA}Sb51`vq`)vrJ&rsMAr8v<D-d11o7UUv=@GCllRr%
zze98j-jX|JlGrayz-ux{^h&7#&yi7c;;YzHvkE>Q80qOaHu*i=?#@;)>l?^XF)`uW
z&Q<nQ=%q=TmuZxR5WP+7`Ig6acz6_-ni~8gmU3fcZ))EEqeyMtj1oiS&f)LhUlhp&
zz&V<4dHML(ymXl=_1hyx)4zRhH<O8DMOlECsNsD+b3;d;xvSJGVkPCa4vN@f-W;Rh
zIK@-U6bzOKh0N1dxmQsav<nIfjyKxtsaNWwv$9scd(SDkc(xN36C-7{`<A}{HA#t&
zNiEgm)U5kMRcGxf)LB7~B)=;VHQ#!Qp4a~oY5Qj;ri%+F7p6nVZ;LKR504Lqh7`bI
za@y!uuv@4lv9=xx=-ixYaXZM<Z75}tR(PU+9t1BMn~yCgU{G8?T&N>0lC^c*9KahG
z9TmI!lif8mghA<dO}#%?g-WcY)i&_?>-c#3e3KJFUf#6rQ=~v+U7e|LiB8?i^wxp`
zx?%pX*w_!++OL0R3dA1FRh6M=F{zh-N#(Q{2qtLh-Ihya3cfsA%$3#f=M~^%$0EST
zcifw#J6vwYRLZt_%B`qU!Z006K4T3{daIWwo5*x%=zQjkM|rh5(|!oGDa|{F(-TN{
zxY^Db&;9wuXt+#qwAg_9K*i9_Bt(OJ;^YJ@g5cwgWvkiy+gpugk_rW|>VV(1F|}g0
zCXss5#{IW&>1NMUVQJo^$9xv#S_bwKUSelr4+$e%E2|c^Z<CkB6uT%&pDgJ;)0408
z$KpA(t9QFk(4vcg`hXZmzksbaprG(TilJR}_7-R`6KngJ#ppJbl`-sBKY8p%&Zp%C
zTav@%7>)yn@~t@xNFme2nE_SG1+>g`^_F$@k7yT4=B`Mjo?We~;rCu~hRCCVf91nw
zI+PIGR>xL>kEVe-sr3YR4jbpWtu~{*(4m`+tMP(QzCo*DeXE@vkAD3_pZT=cex#yK
zxOkdILopJq@^W%Z3C~?c%#~&^m_nU9=zeeNVr>%@dcqPTEoJt?{8EDck~w*LUWMy!
zOJw#7TP=3fW>gF;T20a7*6P%9xKw+_uX|vj+}U!opb~osXzQlcU`3@2Sn_&ovS#%%
zRS_@^@2EA?4rw?Cw}v7gl2?co)H1rToXC)Nf9hXfO#NWfn^ZX1Spk*ZxMI#VtOiGJ
z?A%aGN6D$RqVPc@u7e(KOs{Zj)y}S@7hVx@(@xl7U?l+k#Y~F}w-kN6y|!O`A|`00
zIp}T2bs^+#hYd3(;L<XXsDsg{4;#Ud@OY0x5hT);DzRo~zsKQw)mA-r>U>uH@!lat
zFD2JOjWWqz;($d!fLH}DoVVjj(a-wBOf%N&IjnP^oYV07j=h@-8+amzRY0B($h=%-
zfkz`>xR|Zh4KtQnm>;E*?nveRJSv5uA{H^&DPeum@n!^>&X<0squg#prQJ~HT3z+f
zTaFBK{LvdfGobolqRELpVpMBp_be~!8>YAWXHS8yu7cM~*{=4o&ioWXLdIjK_3%6h
z8zX!lJeqB^O&J_#oSKjsNtbX{a&!|<@oJZyE9@7MkjLJQoYdHns$tCG@$#V{T|j|h
zH4bTKmFFz%Y1eQtYD;Us>|Gd$lT)qSpfxG+YO&dr!OwtANkU`JaQ)cjMLqTRaJYwu
zFHms{k9lEWY<P;aOn78qdwMOS{(QLU-*Sp)%mRHl#$B%}wIi;*_eIi%W-0?e8?<x<
zMRdz*yKB;8B7OvSR(^vxSXi3|S|EphuTi&dC>snddb%dZZt?G5nUX?y&Xo6U8`~JO
z^5N_sbT?L@i4;7bav79=`fjj2>BEplvah9xXmhU<MF>Vmcip^AAFn^AN|Ae!ncdfC
za@>70Fq%}EZ)a>8C?zBks&|7yI;X6v4icX&hgzh$#`QAi@a%g$KsWpfViNMe3xZMh
z;>wABANzWt`L5PmLY;eKA}5`!2Jjau*&_ZiY@PQQ=l(SlH~}5Lw>jozNu)(GFS^gI
zIpb2`j0^osOJoELK9MGkLu33l8EI)DqyCTKY_~pLl`~Mpc*APcn{cMb2sL+Acae4B
zj&yFhq}NRyP+6cTzw`Vn8kZD89~l<*eZFQR>PU0e$QT#$?bB+%$LeTz%0xc*`rT0w
zIvS4m!-9DBCx5&7??V(;b!A$WCV$$zJWs^iePE)9@e3NMsNOlF*cUMMQmMRSf{x5$
zSjOr3d2eqxfliKL`o-abTm>S5JseK2Kx+E?`*YH{m%n~2^SO|)7Yx1t7vu=DZ)Y)N
zU+zv2*e`Wdok!u0rCuIgt@@8iGbzyHLiD~p_HoJ3E2cfOWUy@5hIaei?iJZok*#Yv
zSyK8=qyK7gGiAf;Ku4b(gV@{dR8XtB2j!{jgx+(!FLj}>vBHl|;=s`5K2i>z87Gc3
zj$Ddw{B9(YEpeVdCnmk-3&blgFR{DO%ct_rVmGCIIfOE#EcR~JwD$`f@5aez^{wXi
zhoO2rNd05DI#oNd0d0iQ=%m$y6L+Bym@hV&&FSxPdXw|*SIsx-t9FmZgfF;Qegwoq
z1I#Rn{k8B2AbS3Mp<M}jUCYa$VZ+2L{Gre2x150t4J_*5;DG9SDaJt2bQyC_!j~{{
zmlP3)_oNLclE3JM<JICZMlfwvmxKfrWPc=aTc)~4vwD+-gBG8VJMEwAdCT=m3HRO7
zO?J+0;zHIr?BF-9d+J@YEv)yw3>Q!OZZC0;>7|eNK4(!yqu2H<eOk|B#FUUn#5H;3
z&k~Z5<l2DVF=SFJm@0cqT(+fD-+LLN2_LFNCc3I}GYwwc)h5``mN$)TY~0DrSFgjR
zt^wHCCB0G}Wyg}$rc{x6t)tMS%3~ZlX7lyMm_~H%O0iu!91r%ZQBOfxhed&i;0B&Z
zrPdJgmN`#MP7)%mAf<hHyHQ&%fz&b^-*{!Mgq6CCG0U?Z^-QFe2EO+cmm^3bWJ=@E
z$LuC{CFq(4MpkRvmTg+fp9||hbOVp>rZSj;Fb-0=w38GCzArM~T!?YdAaO7J4u4BB
zSTI=SY5*MG>ErXgskH+=JQbB#cIQoLiQn6E&M(Y>v;oq2*Pa*+<cz7e0OoutD~}k5
zoY(n3e$60eQ*L!3p66?=j^t3yS{o^h9vB^z>KXgoz4(*j>StEh!YElHzuU#Bcw|@4
z->gCJfuAGdyD;QKr~r(q;Nku^*>uoDW<p|c=9vYJkj}#g$0@eke0VnH!TWsx&k4Ci
zWX5=0G9o`ERT$eiyyBGWi&a&CVwhub9ZsY_ppAJZ1uXp#k9%|Xn$6|x_wPS3veAJm
zBNMR9TWp;=40ezDZ28+N0!aVgvRr7y{=RR1LHmlg66E{T5#aD$MH!X$g2BELlk9n>
zUg3X>p%ByUCYWz&>Sk*qu`@zBtvp<e8K>Dd_qL*JxV7QU5FC1EMPbk$FhDj3%gR7I
zSbxQ6(2UV$Gh=b~*;xb$HJMFCV9UYY=nZ95`u9t9O<huPiB~jh6B3(dg*|yGZ{Pfq
z%eRY8%dvzS<3qUohp@Dt4n8re<h{7PJix?sL3#f9p&#3Bu?gKZAS<Ago}NA)JVGwu
z4h4`^wXKP%X|?^52uonFVgxEGYQACn=fieX9Ey*J^Cv7DigvWBO;&3-R7dFinVsHa
z6#BgWUyX@yalgDeT3VBI=NXowN&u_w(oRQ_aEb<Dkz_I!VFd;;DE;0%9J#e#Zo)f5
zN69IYPX&9)DhaarhJhG9Ft8ZbhGh=O%E)wG9tq%5P{f>{?_l9n)f%GNjXx_aqz{%t
z6#vnZko86v^B%fvayp}5tali#wfUJP6w08qZftHH9MM6j^Rf#us0I%M5fG@^EjK-v
zk&)r`xMcYg6kK;mcv==2iD@%e8XO{qz2K^p#TzaeNnGZu91#uzcgeOS_r{P5;1UtV
ztoKFIG+Dp9FJlU;8%}03oh@I8r)-2Ps5O>bKHj!C?{Q=m2nHr5a_g-&*vyC@=|10m
zxFP0mU_XC*w53-AE>ukK4=ad@!n%LBT4fO!oE`x+AR)bmh;5RY4@TYJ+A24u!1oRa
zoV)}juI`DAd}4U)L0!-H557MlhlPg+H#873X`V~AHrnfGg3J*<oHn5Ngz3&t;|<-5
zf}X#r0Ngs@10p4jVAiUne#?z=*>|LpFNN)Scm2Cgi7=G6*Mg-}@%mCvz043>DO*V5
z$?t*AY3{<LRmrCIci(8F%Ap7c|ENJ@0OmS{NY8}9sgf~S9!z`z`zPI@<HJ48ay#LV
zS+dMW;-Juy^3m8S9O<m2Pb(AI=8ADDjRT3CII#OoK9=$k<)JB<xrg%0=b3FrcReZ%
zYTup1;F^>gC@_$jY70dv?|}qDnO-)#D!rPDTIf?{rf-nIoM~KU;xLlYMY_fS&Ncx|
zA}FR5=&jL|n2uCyN?R52g9GxWlY~8Xd(VSS@jUVGu+H&z*Bv7HqlW{7f~{lp{DnjB
zI2vfm%SUzM(tVuvdXe8O`bju@0h@^7+_<Mou5keGidki2CkN%0Co3<0AcKk1rl;tW
z@s+*3^AO`u{TyqBq~#&t-U;fBX(Eq3uV*33p~JJw9&%ve3{rHL+dByRG>IDxI%z=t
zX7C|WQ6u&G`iMF9OzEZ-UT<kW^^?5)N4-z4R138$rH8}@9wF=tB4{USyteL}cu8#x
zn$>P!80S#!jkO^fLE5avCmNVI$+OJP;^D%I==*U=L`qs}&JV3<kl$FoT`X?llCk{3
zWEDjlQ&TOgmcPHYX@6L96?4nEiVtPG&n@-9RX}(+NY}gmTV+=6juh7!<%QAvCc$CB
zJgl$T+Z0`A>K80%5?5_0(z#>UOjT00@3z;}=I3>CZ0wahleFIOgf$VwR)}P(1=PCg
zd~l`AyEY{J48ee6Sl23ra+w_@Mj>5-nuya=r1@%@LD9NeNSTr7t$2|sp@S>}g-r?A
z-bBK3^2xNSR^-0Zt*p$gN$Gd^b7Ml#-b@D1taubTTh69rr;pjFL44+(#k$J2Muqlg
zyAEJ*_c{zWAz<YE&E<XJSo0Suws`^Vx(<H>SNW<#;bvQv3>`YFhjACD0r+fnBhQb-
zDI6!!`obW;&W!2!M0ozr6D%{r?zYrzIA`?7%vtE8{T_O+psE>-|G5|WB?}16;qz#U
zU%_J-Z{dR|iCA|=o3C7sJHe+$rT0}TW6JS?YR4<>3m!VP!oIirBeF?T44-=N#zUSl
zsIL6F=n4m=c5FZ+xRg!6TbcvL`YR=j1}ehU&-dLQDevx<XGd=e+DaSzvhhd<WAm$b
zLcL@&8!7maerB2LKys1gh`{W!&35vsr85)7$$7Ms4|2p#p97tT3evyfq~Sb9v_I_*
znP*(vXPy!>&Eb3c_ZHu?5J13v8AFv(*)RAjlr|`JJA&{eO!B`w4O2_#;PUhFlX?@4
zOteHqq)F%EYf`?t*4!S_ba3Z0V*poX&mHJ05CdR#|H?=bQT*&;aGCNPtXiC2-wQo7
z;L3&t2k0y=<~{75<Q%dLF)p+!du6K#EC$vS4=R=$g?`XtD{CIB`>gQdVz<vwr18=c
z`m3pMy3psN#Prc}aJ#$J7Du#>iTY&pkPpdHuh`5&oyEqad7ls#84ngV858ss1o?}s
z3rr!EAH^?`*>tDU;M(n1)y-5cOI)Hz<;PwzeSN|H7_!hE>miZ&J~1tbLO82|uv9WD
z28(8uPESCsNj_F=RmUyI6CDIxaP}i};=(_|a=SYi(`g_6**fbbHWItfwd1VC%JL1b
zhv5xsF;o97E(D%9RrSeecBbBXipzpX`EM3lc-}ixTnO!Vg{?vN<AunJ&WC<ksfg%5
zJCd42`1nTIn?+wtdqTu`olX(D-fN3ufiF2~qdRDn8>bQ&l;F371Q7KV+wN}(@O{=P
z%iYazf;FwaSQxGE?WLNYKivMJ2PYyPuJ_Fj#avHDN9O(eUJ)5(np^cH0Beq9@DUI)
z!7a^))Q<!Esk&ht4g)dSykF@elcj&EQK@3=6PO;|3<1ltKMk>SxVt?5F(&d*E(J<v
z*^4>u%tisrTpg|i8qHQFwh<fK7Zw%uOim(260;?;lu5dDx<9y$NAhz5L)*93*fjhN
zFek|xtzdR6YSm(#nW@=xQWOMIIx`339+W$A4_`0US#kGL_IFZ1ROn+NhsThIf>?W2
z4{6Q!Z|cvC<PmmHM<&8L5?zHReJm=?NvKwDY6rZ-S+pgdzq2pIfr*=|6boXIz_7Tq
zYE)9IfIqAwn1m>C*4QMqgf)U-;$!ufzhtI$--nC)UQsT+QJ)NvE{PV&(quM}pvw~`
znZ~6hRVtVZ%ZN7>=Og*-P?_c%l-LC0&;SP`nP=Wj_hs6$Q$3-Xw`>07kD4wb83#)D
zCEJVfEqutj=*3>3F>j3#zgo0ULVQN<Sx#V07ziWc#|M#~MmOSlbcY%k4!MEdj4L)H
zLHH20HQ4(eNl=vT%Rs&P7j)BhfLAaVSN@WrILwyqq-@G3Voa{Q9gQ1TaF^Gvy1ej`
z8@c`WZ^XdiIHC@0IcjS=<qb7$YGJNb;I`(R@P-%6+BShPMH_lEJ$)zdR8yrv%Y{_$
zNfHzI&tsjpi`fByElqvsZ{#j({eb7S(BSVoo|gO2`H!wI>@@sV&m(YIKDzDsX!Sze
z-I_I!`7uMM^>>Ocx&S*&{4Quf{If+_64-3(n@<MG0yPgf|7DbY5}Qq@^$bbUFN6)0
zf40~t$=s`0m*kUW;V~4rt!Bbf^%$n`y31`toc8kGYi4{4aq@Z1C79!bHc6FTU+z1m
z6X}9EtADgT!d(cj3Xu;M^eFl2nDZrz`rM0&p4mEjQkQTwmi<B3>e=(0@&js#kZNj+
zk&JL#twjGjqV4&%<0pEg3~4WqhoAp{+B@%Gb$SW_qaADJBBkLd`{0smcf3$CM0ZyS
zFZ%9~%~1jGEMz)I8aR||%6+hsOic-Mv2@xV6QM=5W4~sG|GhL`c&o>t^l<?aFmp6X
zxOIS1%Vx3EnI5w<C}ZP0{pT8QN{JdkflS=i5no=BFWd=OjEUL=oQE%RJZel7dr;0g
z>}2RjhtKKP-F3<%UiVZ5u`{R|vYF@B6Sq5CY`lc(E8`TZ>F{TNzJMhsL@I22fS`PZ
zkQGcJtmScnT|Y|WNF}sCU0Ct9B*)Xz_Z@s5?sdw+O3mYPQr6HP3K>8H(}l$yJ?<6V
zbVZ2;qSj75!S}>Sk$`d<SRj|YHeX|E^G&R4T7y}umSErG&vI99cmjNp#jp(*<Vgp%
zn6J)_e-2vP<TEIiMJV%N;BkS_3Qk24H@Sn`BbmrmwdQX7+SN{l@*&yT*_#>M7fM06
zcoJsW{*S@BB=vdcHSdzx!$AAY=d*9+GdX68kt#MZPtg}@&4(S&i%)NXev_K|Lygak
zGdEfvbZK)?ZQ9i*%q*3|i=)+HrOk^)Z4h@VR2mQUA@&Vl03ov<JOc?21XWjlsktib
zOIK(5--diPi;~lVYxp;{Jqehe+xitc13`ni7n*ItA4#%7I+UcEd)cufw*wGVNm?2$
ziOtA(;Pl;4LkKPe<d~<nP)dttovBQOyb%82;dq<l9ImwnViAxOxeAjC$X`I!Na;tJ
z={I^TJq&Z4G672x;JT7Nte>Fii%^zuYAgEgN+8e*6d0T2uD}Hl^664=*(W8b+h?HH
z#OX{h@h%B`?7}apmClCKN=N-GXbx2ae+@Zz!us)*)}N6io=&*wW+R0l_ka7yQ2WU!
zygvpznX8fy$RVx<A@X+UV8_Ns9|{B1Y0ILVJuWUCpo=l76)h*sISyeb0w9&H$S7YP
zjB(t<Pok*trMZ+rCvJE<e?aRe0Pht<zErU5ljRLwjp|uF;PSV?fJX3SPz2Fc3n`I+
zf{h>gd4?Hmx1$#z&8yb>liRzTOeqRt4WL8-Mad^nb_(=L(`pa|tiuz(*)2XVIF{(2
zKQjRb%f7?ME3)x>UNBLaF8<wTBSI~#?dkTaC8c*tN_(<_yahE!HCnZtl8J!mL7C4*
zT^^yFu27~}2BTZNmsyC0Z%Mm<T#80YW`nnChVka@F#5Vzq*WM7iaLn2EN+P?^<;lM
z+n+NZ>0O{Z@Z$}+er>K$aP_1G+_x~yC}iu6W#5bDIy90$rZ!~qt6gU8fUSs$LqF!Y
zJ(@F=aKTa5PY4u4aQDY5gVkJ2l_!(FbH7DNVx?%raOf|c-hVX8&ega-*9~u;Xa1AI
z7@(i*kpZpK75l%k4&;NsVc?J4h;tu|4prPFyScXGv%W&1((C+_9bcU|myMLFGl7B}
zHWdHCn1|Ys#I1M^_6f1AkaZ(Gu4zrQMv>_7{y#U9I$+v?Uv}R=&NhHxlRm$33rS#a
z2%?2ZZ*y#A!XEQwrnE&`f(9egq`i)X2;ddx{J%7}@hwO3O=)0hED98a2+G_F_Isww
ziVl;rQWt-qiB_jKm6jT6EJ=uim9p~aHOp{LAO3w#W_<uhFF#NE2!-NP%eea-DU~CV
z>C5h%wBS{gVbi@Y4pRzPG>;5gS@%5-ezML$cX#V7HVtWBm0s>IkRe7zr63ESXLP=S
z!oDCg2AwfOp;Xb!2HUxJru`%La}c#4q=oY+hV?IsNph*UKBI$!VzCtM^CdnUV3A!e
z-h~w*@4shklMHh*onKtB2GqkrC+J=VUr!$&?(!ugx5J{Nd&YAm8oCQ`V3YOMYFHG4
zVHa}GbudOctT}<>Yep3QlgtKV6htr^otQ{BTlo<D&~DuG66hF3WPe45!vUMg+NR9K
z=T+eq;S}yW0ryO}a#PCnIJ6(MnRQOBn#PUU|NSfS!aJAkk*;9O$rjwBrEd~rSwcDj
z5Y;e4Vgd-tNjOl>wB=o|{s_ryt0!>V&Bx*8n+3jnl{EbUf0o|2Xm8G6^y`r5>)|T6
zLWAQPj$M|3$Di)1&W@i=%mXd&Ec7J8;ANH+CQn9_*>XLWZDX>t#YqjhUm!D1MG0M?
z`P?WX1<htDYbnM!7!PW=GbVkgbZ#S*DCiK&ydcw62hhBStajf^QOB`N*CdnnRMZO%
z%K|ka2@nr%`+lf=v-Z+SW$f05s(RiwRp*eR%yE$<Qqoi))f$ko(VWMFG1U{ocO(n!
z)+x&TL-8s#=u~&zsEd=&H~BDG68^qN|B^sR4+Fm%vcbST(8XN8ik3v^^Ey$gFODdy
zz4g7^OBQ2M{`{mU*jNlVQYAmIr%`3BmJ5O`tRApZ-4}{e=?wvloGUqAqWYzDjD@_I
zFT0hvT;Z1Zi=DO_0d7OZRCZ`f9Nvv6ZO}&`{Vm-hIMkvS)lHI1@U#D$5u{bYavBOO
zA3Gkt$u+qgxux&B2~F&q2EW2G_r*)i8+@_m`q_VKh||+m>E^1;b1Z&b5dA0ck_6YA
z#LGN+qM-~t4qlAO(85M|szi~Mq{07s0lZ_%WtN7bnhND%l>Ns%Qdc&<VKulSoF%!F
z)WN>roNnWsA)>~kf=YvC3kK(20(Me*?$XC|l0Q!R-Ye+HbY<^Qn$uo|rP?oCo2-%o
z3NO-lnPN%XDWIawcQ)EBM2K*keIed{<GK3u_60BO^v`WN0ffW%M>xnjrx%K`7Ytq1
zMEoZ5aFJ+LiNH!AUmb()mpL;OFlk9m2_w%RkV!><aXc)PYQ;%uM85l-$(@<V63^kW
zI)wJ8sOfQeTcOGO((NYQp>C(AcLcsCl%dT={)XBhR578j_MjjP9$PdP0MHyG8o5^P
z)_1G&A0!h9xnZg~gT&6Jrq_vu1DvXc0<r_rgQJ68g*B_vpOy994~|pS63>Ht$hgHS
za0s9KHcO>RrZYsT*f^M5MqcLk*Kv3vro_EPZZF@`3i^BxQ^Q~(Dhv-S&Gwz8G1_g_
z0Bjjc-*B?%8S<Z=N&ZegtQ!;VL#6cjevVlvuOf&%&~f)i0M2$C7}Wa$ivlr_mr(oX
z7w~w{51a)(iO^S;x=dyWGIrm7*UKv<TR)S607SWpGHwTHDOEKP{9o@OXod)ti88HE
z70Ezdj7z!cru2K3C^$Uq9nPH0R#$6nytWqR14>n*A~dTEY2+q?<$Vgl4sUpm>Cpdz
z(ZE1o$08ms^RSCg04X)O6sKIb1709X-mnUD_Ih*MrI1RI$OH+X#KXhmb=iLNXSq2V
z^@?&cA<WrnHdLBA_Z&w=J+98k_kwV^Bc5XtnMB0!WTDe$i|LIZ!jN8E9s%lQ0FLZB
z7P#3{rFFuzf4%GB&L$ZE3g-z@v6_BT`}?cfeiuwyuUtB`tPYeH-Va52O#w0@SR-Dl
ziIQ5mQWr^*#YZpB+vwY0Ugk*J6#8h;X`%u<Cj8Xd?3Nbvu`p-na<$+>rq_j$5e9}=
z<q2OpbPRDAA)+`0q<o6dw1W*}`YT1?oC3W;D2dy7zv)O4!Frw7ubKM_MuVb)0)ES^
z_o;5#2MIGeN;B-2WMGE_kK@+!01&c`IazdPjBt_okdO1u(QhH68=xcpWMkPqn<Sx%
zjQ-qTZ=FX;OU=qJSwtVP?@Jmfz0mka@XyM024NMt5k~oJ@$ujQ2N=P^GMf4-sI&wy
zm)ru_D%V-mHB+Ak&+g=tll>eFjYo)pQO~lk{y}&6*m_br_hXH{lWpPHbdEU+sII|`
zpR1EiC|9a5?q`UcjGLHS37>_9Ss-+}&*Cr&p5OXuO+PF^U!B|OWvY!u;(r<m*bsj1
zO_&Vj%%oFrA}kMQjn_&XbAC~e-QSv^eXb$Ix+0;GV2@#fzoxO~W5EVOo{1*BUQ{|+
zIi^ogR_b$_vy5{KYmPA=*ORgst8Aj!-TDE&qkEm&5I<gjK8qC7lVr9RX2-A5OU>9^
zv9Ji~UP_Uy{x})hkDtEwC0uc+hJ+DvfGfkkgUeKhk?mWVhERxTr@|byez7?_j(@|p
zwr4dcWt3}m`Js5<V%Qwj*70z9z>AayxS@txTY3Ib@E-p|!N1XI=P>d#(4Umi2g}I^
z-Lro-^BbgI+Z2edHJ|jP8$~@C_?eYuC8Vwr_;4C?)a0fPLy~KuGq4;r_g@uqA>C=T
zi_fTT$^>Ac2)uPvCx=!FV(F#7&`yvWieqDUYecV#by4<=>0blD_EmcD=Sc51x+&dU
zW{UKU<u+FVhsVrfeXxI|rgM_JADIFI+x!!X`@t(R)ZqF2kJG{@6B0q0f<MDXyWbV$
z>=$L%sEV73m$2faQSvIobj0X!sY?r*iQej>d3bmT`dp&{qS_>{1_oH|U8>#k{K*YJ
zF}Lk(Z<7;-2#aq0Yry#fKcvY49%S>KQB0KH1FUlV9*@PM@60WCHpf0p1q0l~v=-qL
z(`yd1clsFy@RmaMp5qYH)fL}(eq5B6uSi@xfJQ_j2X9u#%^wRI6FJUgldU{htk8Sv
zE>sxG76Py&(h<l}i<=#m2f-ZZn?dObc@<n$RW-Q<br<q`6TI<5i387@T1|FI|AJaI
zNS<;3>WuBSc&`a<0}mpzNxN`+(OJV+k|v4(P5+B)jfV$zr%9+mrMkBf0Xlv8C_pT%
z-yJs`6v$xPV09)z)Lx$LFxOW%+TW4+-%}fqHhh0?{~ZPB0|FeIrvFxjgSO(?&4*jR
zNXG@F+2kn2%nN2V9dWmzV+yNzy&1B}P3TQ4&jAEDQ>i)YTRhHHOC4U2*~)FCbE*>1
z`MTs=qC9+fF_Mr1DmQW#&ky4jy6ecQz%^wHv!>}PrgXy9!WEdE?F6WW=MEUlDhm+&
zx<Z190HAt{o6Za9=07kzmuC~T1AUZ_Qj*-=`oZhZB3_G#|4%GWB+k$Tv!QUyY<!uG
zVND=_0tpZAz;`zdxxGq30P54ut0u?oqG0pJz~syo8|BeE+NxV|UGAu1QEUtMo(WlM
zt{wOJ2NDw#&PJ)B$G3Bi7t1t*C2`~4bB|vS+lA6A*ZU<=RBX$I*yvSzzq+270qViT
z8})3w{%L9I2Qqv)wlF*ug)zAr9%-@4qQ?54kh)+r5<Z?!BdpxrakXz{I;CPB8ZHI@
z1eX$&8Qji>YDp5K>Clz>PTUr_$&b};ab7aR*w@=-dmFx0G3&aP4h`^wkN7Vt15f=N
z=9^Br<Q3${3yg$j+W~xFh_^=sBkZk8n~i(Ja&En`rb3EI#K?vIv~u~LrwwJb=)4)n
zZ?rY=ACF0HRPq!4f6F$8B0^o<QKS#zBoPjm6L{)6ilZrK%g2S#(_fp#toI~v7%*=1
zz1Extj2V|qJ5|@Us><H@7tcxthfD@hjwF^1DZ!G5TDfe1VlFk40v&4HONcv>chSM-
z>v&&R549En?orn#VKv<Pc!2Qf{(E_n-*i64!)HKP9L*k3%9SGw9Rk#8ZOvz?HQ{i;
z#02J(Y5wW*^$Q;NY_fDNYk7HjN(@v~RDr0iNKMydsloRBY=vIY@ic~OCeSF`^Ao4h
z)3B<!?I^x({Pcvnw&nqgk56OUb+EAs5qZ4MbUT?dTB>&1kSh2SC5ccjat)YR?PRi<
z!w@YPN?hC1(cDuv_0;Aql+Q#0x7F(ac2(E-SjF@I7b7NM^ON;HUMKOiYj~?@aDx{>
z)UoI{N&hjIJpz(XT>_`(eew88?&lTtxjTiWbXrwL&Oi>Xb9O!`#swMfX7Dy8qs`92
z<S-)!5WbcmQ38N8&t-v=cNzp`2D&|#m|-wQ^fmO&O{U#{b{DwBM=^uEv&Fh|m6m`r
z9SnqJH0TZmqnpa$LCz^Mbs4}1CxNufooBZKw#EbyE*>X2ApW|4RJ{-!;_GQZw5!si
z?r|xLvTI>BOFg`fl!oQY(Oi<7)B9eU<EC?l%VqO{ehz6C2oEa{E$<2=sAZGgdN)u9
z=d1PaTu$x5s=BV{A4Xpq9wvpe!0}OuH>iPvcS-I&BNfa<r<3EgH|?7<g5QNjr^@*E
zP@^l+?Zd6I5<!97h^mEnvTy`d1CtCbJp<5aL^bSS)2UjGPAt-3wu6EV$hax{@v|7F
z-S_H|2c*3#(q*f{hjmn%rJ2Zda5>>X0}UGbo?mOPPIS0Dffi&CYeAyGw@@|c!PFqN
zRX4C5`>UX^-JB67es=Bb%$$-oIBbJQV8_M4R3u~0Jy@KfpU7{n+~A&iKi2WNYir{A
zL6*1WfQW;`@ShGagGT22Aj@xou+&tiXKi57sQgyE-Sy+<nChS3WNr>9S9=1A2x(*u
zNAQ-LH7k%`mC%dp(%-LT71wLcEu4L{v9M6Ku43i1=vf|$ux^jrj$|({4hI2#Kp$RF
zYPhCy>zi5vyg6(F4x(6VtWH&b`%V(%(D*;&g)s3&0YulCARo!oh>t;w3`U@Tpp@)a
zPSxFydvZCd`SfU>8jH5a-CoiW5lp)r6N@Xe3nYfm8^rkEZdb`t9qF|_3y(!8{n!&|
z`WG*aeg*9z)I6FhF6l$UqZzs5u@m3wwE)d;qRfMB_D4#MsgsC!dr+J(FNrv!IA&*C
z6G<$ZbS>}|x#;?=*~X99AHP~Bw`k5&s|sybK(Jb$K8~bBi5GpnG6`pc1Sf(epvC=}
zvIV(d1w`?>sjVIs(A$db{LB=4q`m`W$&)dWrH1Y*9AJU0s}h;D*2k&c)~4UfAUwUi
zjDKbvlL#Z|#M8hbDL`yWJjky0@Z5J$K=(<+cV7nt?l(v(Ai|VJ_FYc3Q#$p*1$HRJ
z_f^?oWGOB3C&1Kxn|S#-%D`V=(>V|bkt;^M0O>ER5Bq@}p1U5{;nN~5%m%VrXe+;3
z2}3*m*zdh_Kkza5a1=jJr$PAukuDwP!h@t$GGl-x-RzicUEo*D;op-8Pf>ar5;gHF
z{nelmIm{<uX4HSa_j=GCRe|z;QFR!|pkr>w7XyToHIA>zvrUCf-XdZ^pDc}Ta}@b~
zn^4=sznnLiG4WKtrGMg5IP*^xiFOiqu8>S`5?rHdCtNd=@Rk~ZEcsdapWwCt&<^LO
zQ8oY0e!2bIlPyHjVWVq>(4+-8y6}Ij71$g=roEpu<QRZL{KF-exy>WS-QEh^<V)oQ
z+yk%`OGnT3E#4RCiy>|%o$ImB=H!CUZ6~`*=jsKSRT_P6p9?CDiUUOO_|6;aU$p%N
zxnjc+VNue({gg(@Mp~z3F{;)97PvzBL`sXy>R*!ud;=F}O4I~8`M)+YA-}}#X4^&!
z`!7~|^MA40kTx_zsa<sXU@{RqPs6Um`W@0Dvnz&r-}aK95b3Kw1s~()?|)bQ-ah*0
z(zQ$HQE?YCv>y%|VpQRyt5w*LqQ25V2dBzaeIH$+2lu`9qkE={^F;cS!W@^t9rc4|
zMO^RZ11x_{yA&=DHph?P<6nC$h~#Y->%1-MKee4(a51*A`045{{pmw8Ng7PGyGT$M
z>oy+SW$m~V*FEp!dSuHsL)-sw&Q>w{3k~GCAEuE@o$umpELi~VPfofMnmm6$&u)sV
z%={(u@_6B_VO|ggFo4`x@<{sjZ9kM=8xL&`mgC@;75POOt9OSf{w7!uLxi%Ataf71
zE2Q<d#1quN?@s~T-6$wevH#I%2pb}|(mKSvfuC}nd@5<_xQw7dx;f(po0NBAV)}B4
z8XOi<5ry7`&5wV3aH{L-`nHCX*=>`=*(4J7BoJJmR<Mv?GM*3CM~r${#byb8u6743
zqpMJ<h~{RJ*-8V{+ouJWop=5#KFxA6wf_t*-J&Qb2W?(1YkPa)cUQ-zBd_=3V_*KV
zdd|jb1C)>Ap27xW28W(cy|oFow<-Dnxrk6M0o8Jb9<N`oERlv1+(w1_M64!djPt#n
z#Mu+zPidH<=TYK+D-^<Ga^xC58abS=tU0tplU}XL4u^@RQiH<OxxYTk`+Ns_mr$mZ
zyZ2J$w(XS<Q!Dj_4*eAO;P8z8D}~+scj2J5`evReH-5S4FG#_@9o87OWG>L<@E@Aw
zNJoziLVq8hKxxlyx8FeG!kSZpAXZWsd&J7)21~-Zr*1ukonssnlp^x|A?Ayms?zTZ
z@0jDG{-pnTQqTsB@Z)P!&v>|wWERs~<XK_j41e}Eg%-AFEKjJGJs0jw{B4+HKY$Nu
z7Zs#*#Gzbhg}_GN=gZUTzcW4bk_ZQ-4S#VO)gckW_hJZu?(2x@P=Fj6Xx78g7&U;x
zvbS8s8a6+xWoJk=z%f-<V-~?bK{KK?@R;X6M)^%|woYZPg{X72P|mGzLcD<d>2@na
zw?A^6r&>}H4%z8F&q`8UD1ztGmq$;9cI#Zq?@wWsXB1o;FbJ|uW|{w9YlkLPI3FF-
z94?h+4Dx=|=S-sOz2~>C+7#^di*8FEsJ)Ar<agcuLLnq<DxunJ$#gdoO{1IwP^ks)
zCf+W0@WH?YM*nJ$ISaUNWfc3!D12&Fz@NzSz0BOT1D_eA{;*dd*Xu^bM<PC*DkasK
zn_W|Ns~og2hbM=Yc-EkIce<JUJ>PQILyO?^ZANw5#S@Ll-cQpZMkl|4S3ZhHw(_RS
zO?s%5|EAmhC?}CtfD2nAlvY38argga{$-lLcAHyoM2HYk=wIcWu2Zm!5cpCnzryLI
zL<F03nC1lmWX@*v0dI$9CkG=Cszowt#(;Y$wo<=8eAP8smIEl=<rM%y!w$QKH}M`C
zzAI7=T2qi$m~fjF+$Nah04vZKY;PO|HXbdl-<<6nx{9Fp+j5xVK`Qhb#Irkm=idPl
z*E5m7-y0ttjjj23O+!FZgs()vz+}t9qWzANf`Z~7EBbe>ht~fM+E+!=4;nOm-OF#7
zwH^mvtmNlWm`)c@)Yh}QuJJi<qA62}utV?wvT}AtfA-FGA1GP=rq78cJE_RA3^Z{&
zWiRM{d;eMX8c8o24~Vn91*Ao1nMJhxT!~OHqqIs$KrI>Lu$Y~s7-HOWs!+k)<opW!
zhaH>ppZ*E1B=e=+MAKw>QgKEzrhxKZ?oFYlrl#5)onM64T;5-wY$yYMn$6A4kMq@Q
zo~O{KFWP|ofO5hN(}%imZEfdSR;>WGN&`F1KUwTjt$FmSRzE4PN``}SOQHx-3f1BZ
zhBvHQMVg_qfIrY`vXbLqe|+`5;~u=_687#VYW55g3URn7l@OtUlnzZGeWTg>tU6MN
z;!LMciC^M%P@DQktqfS6uB9Q%z4yV8>EUut#;ADy4CbZS#ued}m2@*#OIAP|G$QU*
z56mF9$x%H^Qw;cm-PUJITKP2_9zC50Q@;4Aq6l%jX|-~r@=Pu@4BjV4tP}Rr;c`-O
z#5vg%7GMh39i+gRRU6Ne*9eued5D+hHm{ER@5~Rx^?k(R3v#qu@+O;EYJ{N1)4fV&
z0+6L#S=*Q)93uWYjv(}!Ds(i7<GK`b7}DGMM=B?W#i~*Z6Pp0B)y6Nt^a=+@ZH?WK
zADy!rpo1iRA3A=;gwh@S*n;CQz~zUyeb=A${v{0jbA0|kT#fqNeKDc>t5L(fNvp{j
z;4tDPI|^1Gy{ycbz0XM2MOI&rs8tQ))#h@F-E_&21>rYCc2x#vkm6gn=XVB!l5)7E
z%u!{C<r8KcHlu&P@;kJj8H3LPLuI-`N{xRYWRe#y`8(B8j?4xbLO-OV44Ifqy*~KA
zXk%*qpc4Z>Dz-5hk#Jw>zA+8XzF^G?{M#Dd(R#W6;MI{-rQhZy5F;vv_ZuoJhUe|Q
z=mMCU*?@|R4F{C#eS-o19MSJU5sU+s_+Wgm`;7hKLOBdYq|F(f!C&o_b~F4!ysRq3
z)+ZZj$3Dcr=UjaN0SMl#{=9zp%*Tm|k8eV?SR|aITq};&>3*%LeargEf~3X9R)4R`
z&W1s+rhL<B7IK?djgN_-)OnyW`4gt--*H6|Abf;AM#@c6?kwvKQ;co>L?<;8_qtV>
z;cfKR9KF%GIVla><W!CQtyw0CTnIh{Yi#zor_<avgm~|H=IUt??MrBj=UheXj*LuL
z2YG3|9b{uuNB<3n8p8j6N@$`Tmbp|5jIdaVXh4gql38>zjCf?Vh=OYZO5Gb9wQEj4
z;HKhHQpU2I{wP7WCYTTp!dxhwDRcV!#ra*i=?}%-mG<=ht8^~=0O~*^0+LcFhQW2`
z(*q8r|8nYu)8BFGa-DYF__*AJ^RBK@?Z3&jIQ2S3isUDQ4B6H6ApIh$R@VHegyJ6)
zq%tg%3M8Bs_3MUO72REdq4n-M%AK-qmq*tp#fJI(`1k~Wx8m5ivkDVhG*@$k@T6@0
zZ&G2Fh)53Ob!Y>n`o;bXGQg_(Jp)F^UodMj5@7XHsfHmqIXM$RrA6NEj;_7&l9BWJ
z|6>QVy$BbEVz}|zUV2ceM!h&<kWUHnzJkxGef&rz6{YMmn<?mp5lJLt8?t)<I6l0b
zGzow8M5V5l9C<w#ofLKUU@Kv#8N>zmrT7t=ia2#cm(ek&8fGvr9^Q)~!X)7`#9w<y
z_Gqi&6A;r6b0)F{PV+8Joq30ZsfAyNGE)lVam56+wWOY9t`v}7oBeksY!^E?_PH*N
zM-c%*AhcudHumC6UtK_v4xJY6X~2U_SLNJ@`V+v!zp~e;cp)B5)9XeOqt5N_n@{We
zKZ>Bt?)IL+$cI)5f*ymum}pwOqDk28t$V{P50)H=%jK>$*?vwpid52!rz2{z0oqOC
zU}B&C<oa%kU>vF#UPV*!deW0>>Dv}*r8O>UGU+fbSaN)W%IIEDslA@q9O`>(jyqgV
z?VSdXnW<}da>xxN>?!<ayndl-a_H=1Z$&m#Px`KrZ^|#%*)3gjP+*WXRQlvx1%8^}
zAye+%Hxekt_I|p?(*EB#>&N-l8>Y%^E&BZ4-n|7H53y8c69NLrd5gMKOl&rp@V-H)
zfz&1&?0f&p8L*A|zmDe5qC52H`ipAhg-<vGv%1<t-WwzE+Z9P1NXBMxa9+zI#?Km_
znzrrO#(C2<vtT}cQ-Y8l`5DV~%cohRpEd2&JfAhkTk<TweiZTZ2s8J7JuI8#BpO>4
zpzQ%ya5h;bk2XkHQ9Fi36CP~*^BO*W_MNuh52Uvz?ByqE|6u=uz?NXb*1rrMoj><F
zSGX3k^Z`16TS9ZmEF|HL{i*jA9#BWvIpR3Lk&}P`|B9H{Qoqd;Z`EI<h%t2A^x*@L
zSB6>-fPun+bMk*mEC)SoegCLO_SFP&{NV`8vS1sflgDv@Vq<cpK1yO5m%0B=$3TDO
zO&u($*l?~{<ye?;!zfYVa-xB`$zz)fQ(3_KbDmAvm9dG#OBXrj%MO5btMb+?iwP0>
zT<5j9ZANoHEdyO`4lb|XiTdcor1p}+?~!rsh=$op>7kzT?(JcQfp*;e=}-T^n_YqT
zEMKDFmUb~unw%j<<Z{jVqhS*m*=+;;GY_q>H-sp|CViq?J+etv-(_3)k}l)65{4Ag
zb1m+0UT?zPT<_+M-G2=WN|uBJ?rm2KYd~r=33&I%fma-d651lF*fwHn7Cav%hQJk@
zJRLT6JZ0=@QH+X!UY8Jf5|IrDus13yI-Wsr#z>!6*VlBVMw8(>tz}FK{`2Xz9n4C|
zYri)6vn=j>Zfa&}_PsRa?J~2Lc;9!acMb0>gEoIIaKZ{qqo>fF_#v!3Ou2M!d6|o^
znTut8ea}@GTn`GV#7#_0{>~ZsQ*&@&xL>+k5xv2}P_4Bb%Kz}1(;`E+5e=o9nDzcS
zyiXMR7qBHeuB}VPv0&rjkPBPI5-%)NH>z4iO5;JGaq`*1<SQ#H<=8NayDM^0b#+>r
zSJH>=vV<aBR^zzpigq`rglOpK6Swn>d)xinl4A;B!`X@l!Wt(GO6EV@q<nnMvcBiX
zb2we2Y3ut92TjgfOccVN-#LOJmX`9RMoFz^O1W9yzKsJ0M~r(yz;BP5XZb+WRjQf3
zf4fkx59^c`^k`ponC`ry6_gQODSJpuO)X##`~YOcXu6EkqRy(-rmw$R-D%jVpF>~G
z;9x=PEV>R;;V{_qA>9`D3sS%_RN{&W#1tuy-51~!*w+JQH9S3evxIyxY;DQlS0`ev
zua4#yr|KKk*=Lfasqqvif8qV{T57a!@;alR+D1b|6AcA^V^v3o?eFM&BN|r4mbS=_
ze={Rb>ycv<g$iUZ6`WN$0OMYCb+x0Jg?V>w6EYKUqe2KavPvyNKwLZ~z)xQ;l^Pns
z{$OKn9t<dU0gq2R+5RTONfo+R=Ej{j^kJYDhLdouNxvpfa6o4Fs2?oQq*0S`YYUC(
zNuq0=j5x1)N6DPp^uOcsbX|^MqGefu9OnyF`M|GHt{>DJWYlZuCIeppbB*#k9HHMr
zIoS7-X?3Lzj%Cso2c0H=U7M(@5h;V64$3Ske_uAdHnZHygBKwezraMn8)QrmsAW=D
zAO<WZ#T~Cp(4<tYtBm}33yHqx=P)M9<_Rf%eajE?1tjOmsaj)e`<L%y*Nc@Guqjr4
zO3M_gvhDZsX9g9j3aPsv=?@6GtLjt&6!^Mc1xWZ>UghQfMEi>jAs-y;k_fDqw^0b)
zCzCSud=S4Y&d2pFk57Fx+bZ}|sMXaRp}?@4&HKv`1E;=YOr{SK=GOJ8+GBwM!5`e0
z&CKgx;GHFVv2o)o0{p%ow_N@&hKRrq@DtZ<N8YSn#P{iG5aL0tsOCLO#V857%t6L|
z$KR!2{NaKB^_mc_LfA5zVg<DF&vuj!A~|X?rldQA23~S3&eeBrr2vCOSy&4fX=sQC
z8&liCbIP#aIsOTR@bh?1R2vB~3$PWWNzwZenDqC2{rCcyE3orfj5LF0R2n$KVd?*}
zIsCcUR#Ih#SCD`f1r5Xd*zg6lkJ$sP;0ea5=-H}Ll9rul0JUcXIjf()-}|L6>_rI~
z<9Mz39J@XBi$E$6asZr}7D=QzoNm1=OA)1@5e&zbksCD`6TRYiRjfHj(A?~GZT_f2
z$4j1J_qVV3qtthNV^Qz7OyF?eV^*ET=Wc1Y)nj1es);^8(F@=GkXCoHRmqTRpj;Ml
zcoq;9M)i5tnp4bDHc?b#Mn^>mMCDhh{FGMPD7_I{?>7)eOOu&BZAs$DWE{CDR+XI#
zy~aIJgKz)&69bqwYN;UyX|+|&PA@RXo+~0iJl(-_A`Gf#Cd2=<^(YxcH5-fDey&4u
zWsYA#l&lxI(G};^P#g!Qtt}TNZxG6~MzoLT{vgTM%>VTQTwN4DO#EQsEf1H@JrFFL
zgFZaD1*R6PN2bxs#l-Yyp|mKmjPE`C7DhHUu_&FLX|70jx~jjt?cqq?(_0!*?wpLC
zc`U|c;!)06At1ZCn7p*Fc5?iPwByX7I#SMZ^7!}D$DMVi@P!iW&b2POU);0NTUzYT
z8METxIc)&@M;cgL1)$B~_>Kt5thA`VLMI2hP{YpB>7X$9=~s+3Z3U^#!SDtVYDFpf
z7rtK$aylJbp7y4Sc2|vZ2dT{au}UX#Hai|}b0yEVNq<BFgLr_U7ZyiF{^kUbn-l02
z&1geQ@%E}MM$_)6ZdC!KC{fTZg+&%(GebznU4LWN6+j8xMXOVBw5z}_gIM(J4TE{!
zYu|u4@E~t>8^;&fU<~Re%v|cX=QY>s0k;}9AMw9EJ>KcC6u&i>OJeCe+4w=XI`!Ua
z>|Uk(H85d~RaJ|Drwv)Ew<g%7_&?0ORajSF_wKDogLHRycXvrQ()k4`>5vo<32Eu>
z?v|7eK@gB`q@=syo&28XU)O&3zV_kv$dd(&xn_(x?t6R&A@BPUsMA-lD4I5wARbE7
z2>qVNVcY6{H2sa{oLaqjW3{jgk1d;&ayvgFbnL^~*2X;d=KSn><LzzltoP%jil}s}
z`@z{Oi`0(`b>66cx~<x$_Q0eJZrD#RALckCF1Zd4c&#16<#IS^G@U^UNL0a#(os#7
zMRk3UokkMHm|ki>K}6!(tNp&Ui9V$ln~-ah%AvO?zmTVp<we#lIXr81)*(pXM&=44
zJzAJ|yHrB>8S76j7uF3Q`O#G>L@WDijqB&WCTo^Ii1WwD9l+QlFnnvYnHG;+-%|I*
z1sdB%Slx@>4UOy+o9i!1Wa{;YGi^zHEu8V_2+k4Eg(VTO@oFai7grN0WMIeON=!bF
zA-EHN3)NjIy`6Lwm$xSIY0k<RrfXT;$wbPopDBxm+j#!USC&ye5j{*SRm8b26PiLg
zsZ*|KqxX?bl}FALKLLeuU47Ch&n}r+G)va!oU?&G6ja&Z+CV2JoJ%XavlN$4^t7hF
z`ulnR14toAaZNm84VSzz^}866+<)3@k@3I_k~19}Vxa!jsQ<cROiYR3t>%2$h3s!`
zKN9q==8dZ_*>ScOcTEO8zo9IOcJ;orf9945o9LU{#uA=%cJnc6I5NUau#LA|dK40+
z%2F*Xmz<cU$m^gp2PLbL`D&dtqqBGCbM=nhM&o)K+!P6*q>eD2xL~N5Yqg_7$1Sd@
z<*+^b2r1#^4x8%J<HH;|{d=N-Zkdmm^udviYZqg{OFQdKAx%JM4%8dOgw@iiW_tPf
zsuyYNed0y66JNig`!ZrRMUjg4a~`;T+O2PQ*f0V0stAHsI*D3wcsFRgD8#*AL%q?Y
zc0mZwg{q*huW5${?O|*@`z`TfaKme7<2ErHGYSk+VX5<qw9pP;1*@&Vk3<P1W%H%n
z2E&QwDa?TnxS?3UF9WsTAs)G=A&S=clVxydM9OT6auEeVe1C~eCKq_+0zhK*W~Lw3
zZ&WW^6Cye*AW*lD(DeuGqRV)G1=;z=jxO<KT5x^o>7Wrr@SA-2E4@-@2zc37xK*{)
zjdfO3G!SFq$VeGGPdth$LutJ@fe&K&kP>;nh)^eng8;LrpKNvYV#?D<`Bc4>TiSHp
z%EYpkrKcsz_R)*(_6zHIKQm%8GozlZwoEUSEJw0S*<c|m_3wTWVKZu)%+aObr+5+(
zM{-(?nLTwp2)Mx!Wf1);(V(3|g#VbAM>(Oc;k5Zz>;BsZP7Y3vQ|Vo9boy=~D4S_1
zkfteqTLVQzL}X%P1HiT=Oq8G5_+7S;86|0EQn{=nc#fLH@uNAat5ws@Xh2RrR4DtT
zOuKBj)|_tVy|r~%c{!8RfYiom0po0w18e54C(4&?kX^EAp`|DmaNi%~AO+Zk>T<j;
z#4mC=kIYQW<2k<Gi3yq$OBo3t^N`7hk8{30Nv)!FNPb2(ox~S#<qLJi9Kn+cPf4M_
ze<<THj7Hr2k(clN2CJ?{T=_jqn(WS~>i$A!Rs||QaMwsYd^JKuN7E|R4ES1yX|%Pq
z_C^A#(x4-PxYeCQ;!RrnYpHOyP7cU$HslA+@zqsGT^*<X-tI1bL`1XO9?nl=zUX7N
z(ISf1K&Ly<AtfapNxdL$ZTQFE<NKk@fZw(-+a9Px6eC{)MWm=aZ~5v>C^Alej*N(_
z`w$4C`|xrlLQ0EMX()pIUbPL5n5H7R>hYbcW91&wV#DK1P$RtCfx+b!bZ0PNvoGkP
z;|%5?rrVJUV7O(<*7&1#h3RzC`>F!c4QzB=a2P^GwOkgmq0oGVmBtbl%A?<F>2!W!
zM^!df*YcRZ+_QRvj~pGU?-bg6CZ$A`0@eL$fY%BiWrkJC>?ms(j>GlkTfljNx%kV&
zjC!i%Bx@Uk0+RwRFZRMeTG@z61cH@mhju&YlCPQggCN-k{10d<W}CJd<8QI(D(Ai<
zQ415jg$<3lQB9UYjNHRjO+`r3h|z*yw91F9&ELqSFZ#o2T840h4gP`kC|NYj<LN-A
zOfyTo>DuU~QKz&0)LT9D&f##rINiF?J_D4b4OZ2o#9?G|t}HjB4RLrPF34H5Yg0O0
zYch0kr0?9p9@Mq+nmevtKio%9|Namf7XNG|SkCZ-ZYY02+8M`GN3-z(=P*S0nGV22
zshqUix()n^@MmXbx`<jfN#Y1s>vo-L2vy007HDD4Pf_l^oM0#9!u7(L6k*mWflfED
z*B$C9FDIhv%_rOUV$vr~9c)$4j4CGFe-&LDRb8vLkL2YivlwNy`8%@+!j2J`d6JWq
zmqL@kuE?~=(zHmN@GjqutTwOp{o10MG=dH%Cnj>%jfmj><+V5m50j4Q3kHk2{1~(c
zkNF>u>aYq4^CTlx(>ypgV~E4Ucjh-k6trqPB)H5*&{uKPP9{$;%Z=*#@mN-U9id$*
zK%s61T6@FmXXS|-$gT)j4*lq>@Q%=YP(&}IUdVo=RQ&@#lFXiwK)ujvh3fYxE%2pq
zkhTx*evCqlO4ZR@)Lr5x31%P$&q^rZ@QJDTzMmfo1HKr8o|lEew%l=?56XcPidj$#
zqN{0<q`W+k<NqFK*Y)B4{_U<F13qx^$Q8U&lNLZ1ne|5jfmT~9%eGXw;>t4tNVN?g
z{QmF%^*Um)l&eR^Fkfku%mlcD;9g7Eq~1q-r^sR=JvNSf0nr|(w;hJ}(t!H&Wb^T6
z)CpP&B=v(?Lp<<bGK|SW4&++H_s_f7wB@Uc?IwEP99QWD@8-scAo50usXc=ZawX=k
zH-Mv>rttrFAdY3w@B}T9p=@AG?m*fKz#vx!iLSGWUNWd@UL;(5&PL+Jd<H-4g>1GX
z(DUTiR)@Wr1)z(4%(g`?R2$on2+s#E(`qjYky8pB`ofb%%mw5&9ImU;<>Xo3&M8aw
zKbKBFn>Z`MIQ)}Duz>1IbNa6@@jMc3=+N@<bOAsgBNM!X0b-gUV_9$HzX$2D9sI$Y
zD|-u16B)5a8Wq{jii`|Ln+gW}7ccO4=YkPm5(}N2Tms7=db8u2Sx!`+;5K|Qm;-?E
zp!6S%2TZ!<i2pw5&vW2Ezd?Y6tQcnODai}v0$hQ6pBMkkkZfE~m+}j#v1y&+*@ylL
zgvtNwi~si-wF7?~5ib#vjQX#i8xaL;FcB!$v2KIlG*$(9+5h#^|Btu$*Tj5&xcFay
z3bp^Mm;bNFz{|v*vg*Bh_8BOGcZLq<-T3cw{y+cT|M3V|5z6`ZU#7v^S_pgvj-7z@
z(f?c~|Jz5NhX|AYDVG~8wUnQKz~s*M<osvc{*RA5kGa%v=eyU>e;VhrDD{8(F8|{_
z|KI%6?vv0wQNRCqns|NYz4`CK_C-$?-O~q{y{@UL_y^-_{cGL_;|oztOw4ir^Oj@Y
zho|<d<sVPH4;8nhqiUW?*ZRr4?JO5f9ak5|%~9YBw2y#66~k}Ef0D4VlltKQ?>*~(
zKeFEm#lY8ac;QO<dg%F45c>UlvVncB?dmU<gwgxT@)~_q;+IxqwKn#cB#Q$;6tK9+
znWt-B5o^BFmUyqs!{aZA<Gh_aV{%Us^_@`aJ0Q`q8nt-_akBj!$4SKTjJ-Hqfy*)7
zX^#hvJ(q2t=>2p9A8uDV;94ex`_q)2q-~56^hJliS9-B0XHy|Ugz_KHHLqChN)KZa
zC67?~cXPde7=zS~bVfvzMM?t~guIB0aE8Z~w++&={!@0V!Q~@FOqumMRo<-SUq(ih
zV!meRyj|McZ;CIXY(cgDuB??g>j77U8obx`nHOvvw(zdI8mLAcm4ib=JpfG?n^d-v
z={<ROVc*pbjL|#4HaHhs+>k(p<p=%3zU?lsy6y-5Nouyf_zz#gzfT7&e5fsL?P9dq
z7uZ1r8{^>xa!PVB-0nawm#B`SskYwd1w@BfZrtTi%3~HXF$+i!Y{0vj4q_A%zEJS1
zI*}$5)AQ2w5$bIhaJ0&_dP31DSLW`r!*ohD5uVE*WMpUnoT>1tT`@%n-%snl`ryd^
z+Pp`m{q6%Hs}F7%R(Flk-t=|u7;b2|S436k4J_*!B8lA8Ox0t<@ajY5?-ar9sO6!@
zqaAf*Us<@iPF`Rc8Myj-x^(dOeaJFpYPEmXtUnl=VWdMrzJ=_SWXt`QI@!au*-ZWY
zqo!Cndc&UxZUo*N(v~Rh4lSqsiC%MvEZ4@Jw>HV`H3%mG{7}Ik#r}=rySVSUUW!nq
z4{hIlurVBg(t;9NH7tQPSUfPXw~yN27ozGdpxD<#u1xpKKFdVBC7S*m9sG?ShglaI
z@Kye8RAqusZ)5&vqou=we*6u*ecyZx*~?Xry}xEWPtn`|I+4KOFE{Mv`ZXQewGr1@
z4brH;C00~p9@%;olKrU(6?sU`8Wi?T!{O7ahua|>2ZY1F)X~vl(r=89Y;@in_DGe(
z7xH9F<*^S0{UDHu_=-c(n9@xXS@mIYb#RG9ouCwO_LA(MK?D(`qEQh1*92U6B_$=n
zoO8TW&lb0HU5d(z{VhY1QCJj9k1_`tL_>vus!hhkp5<i{#>A&kN^)7mPub*t%<o=|
zjVxmISk;^I8HnPBwd+?7r<N5(BekHY#z48gHdMIe!JvQD&mGd9$1TAs3S8wZKUzAw
z-~G{D#y7l!tF0U(`S!)!`#o+yTbl}4XVR-*0|JhZ-7Q_-vib2kBgy5r4L#^pRMG|t
zUUzj*C{liwDz-ED0E=&}OR#`-&R_n+^gys)c8o+AjgT;8tw)xORmBLWv(oS8u3!42
zjvFsaW?NGeVSqn(?5Vq?!|^_t=9m)ybsU5vzw`~m`i;OtdXK$`Or^eWsPKJwndAdM
zzntfF_8{BpD<|l1sORhu%rFdT51!O8s1VY%`~iN`@kX>Ljt_>z4R)|{e(|C|Le}X!
z_8o=GQgV<Nc>?NKwnkAD0GSPd;-c6jTI0Ssw9dbn-wChnk!G=*x`V2r?TlPxrb~<e
zaePmk*z;EM1rO@zvVemYIV&XvVNsZPj(#BjnoCIKH&JN9CiL@K!=ZQvbM@_4;00hj
zRIe7$tWd>H056ATsS%<zeF}Ul&X?rkqQM_OBoREkHOoK=Yf;S7H=Lk@PQ$6t4v{fa
z`^x&p#(v;86Wi2HVl@tkiHT_%Kl70X+v<0Jgnz=b<*ol2o~c^Ew<$@<$1ZV^f;WJD
z4t&%CjH>HGL3WRXq@c#63(DOEDEB>)guGw=%6zd)_ehaLzq9(MM*K0IX%TofMP3RG
zGvy$E?Y-s#YLwAOmg)~M&<&T6t9qY)@#o?ge-INSrhT9U$EMgTD2F(?ob8#fvzpHZ
zjD-hgfNEycL(nzcPlTnfE8r#yTqV>;5?R~Tl`!mPy^heZu<KntA$Y<z^_5%qzW*G(
zs>RQK57jm1sh`hBtI_}X=-`9mPx)j&@jaK@3gUjuk&>@6Y9FsO)Tg4QHXmOVhZ*+t
z6hv3c`rQhsHcKNZXToi5ZO@dsk<5p7rrFoee^J}NNvIZ`gMHly-WH=8|HRiYqz=8$
zVO+2jq85r@3H66+!zO}bq+=6KP>_MQo~IEWAy=YN#3K_)(AQ7M>n?OCy|`f;`cDMb
z#{L|(^k?&Wnt=EH6oX0M5`YodGMADvD`d*lnhk6t*E}{*a2_7X`ftgFH}OHv0f2LY
zJP9zF=SoUa@O}RBezf{+LuWKWyp^auIP4kK8-nU|wQnu8g7}+1MLpquRijEI{YQL2
zLWQu=I(6-Y`U&vB*O;1mdfs4gsQ>P4o=b9+y!*BB_NjHtH+m%A>eyipmgG^(6k-)p
zzHJ?f&>A!z1Q(N9Y&e@{1G$^3h_2;+wYqAANQImUB=R@0v7vnaD6uo#Gvk-#+Qj0*
zqOwNsS$LuzU@a4Ij&0|mGJlulCFfB(8(D0;=`}Q1sv{g*H)~io)h)+66EBfZFx^%U
zyhs4>pZa{M!Z38V4FR+6o_fDM?$0W&0tdv_Cdi!=6^FWPxI@39e;?|#!H(H2ehfvb
zQd!l0KY`3G=tD+Gua_riIXXUPv@&6IAQLI=)D&m)@Z+}}dSz#$2kv_!p^tIc47tve
zCVUAjuQ(YMgj5u}m(Hvlx8K1Fek&X;E>Cvok)hL?Lv2)MVPoeMmyt(YKU(gqFee9E
zSKfseriO;+q+f;6D}d581yHiKbc1|R@9_MASWhB?;Y+*sjpsxB8Du}~On^87P&E{6
z)gtMJ1~DMBqMws9+j7Aw<aG|eH(ddfKs&CLnEgJZ<=s?1IzwXB3qfx|3rkD#|AacC
zlGDnz@SF$2oUN^`1%MT{Ue(!{qxZWsyJZ5%Sk<+SGCl@ED5U%mhOO><SpGDTL6*#V
z*?vncbQ`*#M+KRgUju`-)6Td!Xe-_Rvm<-c383&sG6ZqAwk+S_EFyP}2Cf|*{*;%d
zqM@l}3S`#jp#+i@#xMwSim_FZxNWp$KV+SfaeTU!Q+m_4G4PAeV~q%fnC}xVhR6lD
zX_U{c44tRu&SeJ##U{rWpAh)^s%O8E<027sM}~revhc_&>rD{x<L2fD;`^jnGltBH
z4=2Ic<OJ0=m)(vk7qU%gdODGhpefX$?BbY1)8mY>sWpWD&M1*dHyoH}6&n4=dZBH#
zt3GFBFkc?5L}yLkQ<bgQ*w1%$N-JFQKdw%|K>LiNyH{3%j}!96-$A^)Cj?Adg3VWi
z-@7LWYOW5~7q=n@<v!5Nbp|(ZGjP8TpU}EL%Q8$UDG6=$z%`|fBBy7N-);jbd$Hi#
z*K2QExD-c5j+sPI){2zL))#dLsnsJc_#H+jj9fP|1pPl%8n)IRN&4Pi%rw7;jyb<5
z!?W-WjTeaY9k2I=OpnG5-NmHjT$h7nN{GuF!AlP>8=N>yBhey}Lk-+bABAc$3K%RH
z4Qh3{oesacQ#$=FY@WtM&&<@Y)*L6kZ%4|He5GGL1*SmS43lT;LYgp%48pr6CXjVg
z_~qUx4+K@yHoKmROsMB8n7__C5gB1tbgLB3>8gxOLic*}AQ-uiQqOf>u@!|XQdTg%
zz8~w&*SMXyPY&3$stwBcnx%&*I=q5!=igRio|TMopyOU3U%ek72Z;;kS|k-g0Ovx%
zepp)F-9X&N=rhq8!T84iZ3HzIJ*?UN65T8>bGU<6k}z&dG2tVJUF7V%F2_R~Ku3VW
z$z=n;#>oXXp7={=D1?Yk&)6RM%3#CD{!LC)>IPvtq%XBK?+rpFb~fpto34WL%9t@W
z)SY&Y2t)niBVd^sa=jjEtLr(;UFLv=ff1WJnf*4q<4;T?S2REsK_siv-I5PTYY=>e
zkX62Cm5qsIJ;*=!KDuExH`8Zgn|-C^aoF)I;J+d!9{DWna?KKT^t)2{yl+NH%$A>=
z2fLGrPcIun7yp?1MSFy=_>Yd0?5$9Un^Nz{2o^xKLN?po4`;)SXR2o05d5?L*^aRY
z!1+MW^$w}s{CPW0{%1SRidx~@veoqOBxc~Xr}?m!uq24?b)7x$9d%eRe*MKSf@$31
ziZy*8Mf7PLueWUS8xjLF&y>L00=cZC^|q*ueXscbr2l<Z``(K$>wj})U+&5u!3PFX
zm>tb}V??IDi=a<^YsNKQ9M(QVb!mj5U>3<-bG3|Q=70zNvOr$8_kP7UEnsd**FDit
z>VD-xMAY=Ef&XfB-+8R{>OLU5@xa8|wrHAV8^QyM#W2`Wj}(4Niq96d-*$KX(BBZC
zaQM7_e%rvrmu&+D>;^}f!mlE|d?Go1{<Jx$t>bWVT=R((H=}S&y*fbUv@P>FndR7m
zlc|(TA@|Aky~2vd`FzeutNtQD?wiLJmb#g_+F6gXd~RQNrKjsI6pP(*vC~&iovN@F
z0EjG(&poUy{dIPC8fDK^O$U*3<~ahUg&$s<t?0X%{$8!e@=vp~vnHp1!W^BPwDe`Q
zEpaSIGlPK)RCfiO<%!zbcp`Rq3~<`h_sFlkBX&70SNo7IzE_In@2_-{s{LZWF112^
zlNEE!i6Can;lRooLBt=25MWyiIJ9QF6A%IzQ5cF%P#l^p;tT7C$FIKOruO}z>j6$)
zLn)o_CHQx9dp=jUoy=oTU}hGNU@Zw8@wZYdMS0|5Sibl2ceh1zb$IB`hl~7R$5c6(
zAmA#Fe#Ump%v3!Q`#(K2sS|=n9v>f*IUpQ_{-A0l`+;r%nj#p5GFd<AbL^XCmu-lb
z03|B}j}~-3wD;o<ISQ*`K2KXknqcb10URge#VS7c<JCoHb4OMOD{-umUKmF4)lgCP
zk+Hh_(}?x#x7MDnR-(*c;qF*!B<cnHQ<vp&DS1^@wEigKo@piT&zH)6+Nz!5159!+
zs(c_;u7B0iUe~!Fz7dMnCYmn6H$WK`q{|$LH9eVaW^#CtSUop-hxK^-U~`Dk+qGn0
z#=5$(>_#1&_NMy7L!gFd!h_!HxX%^$<Jk$u9dFx5y!t+EgvG0I`@X7kKBWF;k1P&A
z<{b)0UXIMRUyte+W(>M15rqC!a(B9Cc3kdUDn0mw<Q_3%4T|`#W|9|9@xdYlbUs<T
z#iOb|El4a{FZ}%i-+E<WPek!o&%Hd*=7QVbK|WNeClPxaPmP)<0Qy6042puS24319
zba&0IFnNjJ@7MFDXxG(*c$mjpJ&@U9l}zL)g5`pP79@~dJ16wQL8y4sdgTD`;Q)d+
zh*@6*faJ)hFDY#N=!#77k2>lIQlQ<ma^q`>@5c=aG_^2@Lk~%l8k793q#d(eot|bn
z98V7t_L3HVGyq0PQ5kpn^S>59;`8Fidb={1{d_g!q*bqfMNOm5;;<5~m(f~{9C2(r
z6LYWVM9>ydM%Y%WJ#erg`#Thi(c8P8P#_TUe!S%9&Ps)xhY2~w+M4cQ_2cOULAA+?
zxr)3HN8clo2I&vbusKd+Rdah`ztr}r1T#jcUfLlM65`M?>(t>;R2t97=UZ4=*6c@y
zY>%zKc?a)YYuxpFjP&KG{`bLI==cwH-qWr_BZ-#29RX#X>lS7sL3i0Euo*H9jUFbn
zhxSX@^rKi}9ku3Z^KL-V6Nt1qSPd1VjiL~_Cr#GV?P+p1<kKnsek;Ay9=TDbEhkXC
zR2^IG=3-uSbCe;_Rr>+Nh9;WB$ORect0ZXf(I1Cr1>VO~#hSrqy<6e$D+(rm&_kn;
z#xBWaIgp;?clC-bcmARcZ!;P7P;M)O2q{7jdvhfS$!H%L{McRu$H425p%6v%mMsKQ
zS3<DF__}jvDDVNFHO}ekj}ukDP+up1l`f4zU;ii`yTa9rCE8zCQz}){_tl*Dg!GSe
z`N$OmNvssy?C7puzbB?iST-4}M{4T`r*R2lCHAtotSK<tY4S|gPBv&`QWJ-3<wbz4
z`A?f)<V1l&-Bm#-mVh;{+h5H@VTuT+lXdFezW(l?%b5^ruJ?<frqX|&$*RBQ@eu&C
zsI$4Xb?xVYH=px4+r-4g+J**7hJf4W5^H*%$DLn!C&xLrq~kw-DkpQ>;xI8W9iMJ0
z5b-&F98F5^fyHC}4=+GE6u4{G+x>#V+NWBmae?D`KTCCkh87w>I4HN+=7wayP%o~j
ziN#{j96CF@&Bn<cl+{UEtl4vofC(lMVx0Gf4{N8JjLp7xIJvobCmf7O?v0nlt2!?M
zs~S>-@n+C#XFO-{r(eN|spj~E(dXl^oU9nAN(2OiotY#3xgNsqs?k*Lo~D&ca~GC>
zN@cfVaoro;uQ!vj#CkP$A_<du+)}3%xqRil-S%e@mRdc`ZGkF=mno9qk!nPyUQ9~7
z@9B7mEW`juk^SL^0n_f;<w?Si{X1C5r?`W30fgN*A9B0m!%n+G)w1mbyhMZr{G)tl
zJTFeE?<&>5eJER$L@Szpq)F+M578?1&mL6I-=>4y754V^4F4o~IhtB`MBpmdJv*gI
z8EdvP_CXSpzMj<YPAitV{^|S6`<B}opDR1@2L_U<g)!g*Nwi|M*UdqTYGSYtziQF4
z5+!}u!x0Q)=JBRw^bu~O_qM)sNIKq$3QR{!s>!JZy}1v@mS%#A>LJiW-zZVzp|&L)
zlQ<~l6dx}3{GE0!@Z#1dP0Ss_T9ka%`hNyS`kNgQ8wgcC5*3PZLTZFQ;6+#8jiKXa
zM2wv4R97>=c+=WW!C%T)6T1BQbN|KdOr_K?nqg47<XiT2Y0h<}IFDVw_r8ZA2dL(Q
zx0q49T%88FHjDsV1bjMcO(-6B9J)Q*=<wf*;wjw0()pfRj$$(+7ifeTfgw$~=x66J
zXRMG%5KB>1H_|S)VZja2k@-Ia-g;#E@OF;y0&heYJn@xtc#v6rX3W`I=$AK^-W3rQ
zFbvQFJyqe1kB=J+m@qmGEu+JEPJ(?`*u%9gj{D@Hvw7w53=KlM#vH?q6t&U)*otK0
zJ6HNT6zI+@L{d@{$$^Ia4|q*gGPM}0QZjMOX?~v|KVtCBkO>B2q~x__cz&QU4nD}+
z<_I&}owbbqplohf_l`@Cz|wlc`r}ujK7eU2S~>igA}}Zvg+32CS1;0c*2QkOp7Ptg
z@_b`fMjrKoM`FQfGRyGM7iC7sPRS+fpv&6@f}Mv<Vej?!D=h!2^S{3d>FI~ZocYa<
zdH%`men$fr_dYZ_oKZZm?}iEfv+tt(#Tbe~yTB^))9(w+HKWY!o&SY^fIjR`R)1~z
zf)cLyp$o<(mZ;1Rs?>cuS}jMMjqzKZ?K~AX2Ich`37vk00G)>U%+wnj7bIg%`av?S
zRE2b)S4v(Ub{*t}#80dIk^EZaT{ly_-@lo$kg!Y(6%2J~fzB)8WV7TT52-8mk@;~1
z+q2yBHG^mAVbax|z}ODdhI446FsFxH4~8z%oI_frJ6~phggF?s$L#;8+#Rmu!_13H
zEVlD<H+4zsFrS7-&hi?sn|9tKY70mF^+L8pB&+~xY)a;5;WDeeMb=J~&zF0xV)>yb
zh<0-X;`S@%th@60?^<*WhVODdbUxaRw9-ax{9O_Jok>j^MKU(*`0GKusDd7zo=1Rr
z+2%Xh><3Q@b_T?|t>U9EKQW7{G1ta=Na&R#6+oY=?Zlcl^l#Ru&<HD(GX(HN?!9@Y
z0P?>^zF-6mBBz)pf=#WEXkniZfFB}{@L8@0!^>H`4kSM_wM4%$i=41hFEat*xIg#{
zF8g*8zkJs|LqGr?uOCsG`PHE?=$?}?zZp%+OA?^PKnZS~=KT?nKOv<Thf7Em`c0Ob
zX}4FTAM>d$)M+HXi!$y}D@H_Cd#w1CKRtLfSS6&T5mL&nuO~e1&?-7CHU}UljFFi#
zgR3rpX1_M+WHG9Rrj4>?QBhL`(?W#3_Youd`M;`u*u6$WtShLbQO&b_IAR2x?#PVP
zNpW$*Mk?|nl*C-7*{HD`cC)HeWjd(!ogMe5fSDax4GY@fj4g`}M47Gd4hlqPjmV)h
z5Vn|~LSGW)jNP$lSmor`pesK%BE~ko-!hVKxSFpE!ne?Iw;yQN7{+HmpIda_addB{
zVrED8SIR1{m!Xgqh|KC8&H_7|>C<AC%Z!BiL|xVBs%g0<c}igfmZbkrLV${FVAZQf
zQ&v`P_J0!TkJ1wI=h78_C_VC(p)C(l_C=+ZhYImn?ro@hswM|@@vw+w>X&e_G@7s>
za(KfrcwFC@ZR;Eq;2B-0lF=3mR=#3P=#uF%4cfhjgd0(&b}ueMk7agzZ@KxqxhLei
zADbTkQs9CWqB&f&T}{Kr!xEO3L|N_dWy`s4s*@~0*O&eGl>c@4l^c!vZL|Z70;7hM
zkn;B}h!wTU!dDeK!N9@DZnp09(DI@uInjlP2>I-RnErALj2rvDk-Xp<<SFaFUucf#
zQM{KrF0s6MgXKh?goft=cA|@-G2^GjCDvA=tcR-O=Ch`nr3b{O-er?j7BlJW^tE-~
zKx2H$%t7iWHMxS-eX3#BjmbDR7Pvq=LtoMXyY(y6I`h$j#eI6o2yqvJIJsZjep+r^
z2_kv6RNfj=h{9R05e|L#UK?i6mro^+ie$M(E6LgsZwnQvhu_AT*B6CJ<vgf7F1M`<
znNchoAPAliqycZq%U4({4}kUid3-O&nI)iU?H_Rp_F1)q4LOf|{U4&FAfSv=GYS*8
zS}93tr#g>%DdmAhm3<Jiwts-6&%5>urVl|(i?g9xZ2QgREU7APlYL*0i&cCoydG+N
zWbF~|(yLvw(||y7seZYHWel6B$qV}|KJ8spw1ANu;X2z}i+ly>pis8lRTzA0$rzbW
zvjE}q+f=mkLl6PZGPg~R_|@3lUPuy?g}6ygSP*{Ebh%9LDB*ktosB77)(%&R{08Vf
z!Cr679~-AL|K$*S!Z}BszjWGf>TTlC#s1nWBHhNfy9eP)skt(DQTq*@R?3hrDf%<y
z@<173@0zA!O{3GZHV&F|rd=Ly6z{@u`cuYM_nkTj(&AMJsZ1^~_9<3vq;;&YVMzS$
zc=E&hJ6-X^NE=k1pPAmAZdNxoTBf4qzs?!5eBvV1%-sc$qGVo&4t34|Ir1zZG=xHI
zi{}GjkXF~%BlLM4OeZX)t*N?(Z{mB(Q#j|UK-Jl+U3<Bdh6HTwO1+`boP;v%wQ+Gs
zcj_(tOaI%O%M8I}E~mJClO_Y%`^$x@>iYUmjp`5>a&_cQ%x^OkO@`v2#Q8Si_;Gi8
z&av9#JMrmW{+o9*jl^E(`&}#n=Z<o?vQeN46rN(MrltmuZB>wevW91U!U(L?xP>an
zQ-I^?Nnafts-wk+<+{?I9VaHqKNFXt3ByS4scb0;uzA}!zQ%DA0l7?~6Z+(WEj|c8
z9I;1QWnOdbJW}6X+!u<Xx!=Uib9-tdm=%Z*f$g&UI-M1n0s;sH$7g3fUvP(ZCW}-$
zIv9$LAr2pX2>dup`Q;b%{#eSZnmCXD;ZgkfxE%S;UETjFaj)5lA4+q*aGMQuLi*fi
zb<8Ei+RQ?uerewkm8NrwG66q1IpicDu8dE8*1xaygC*WYn7Fe%7%_mQHt@19MJ{Lc
z=!z5j4-!N`Rp!@%&APhdk0Me_NFuWJr>>?AJq~qxe)d;o{vwriG!bt(EyfDb68Up&
z>aXsiZTo8NL;2>bsg91gopf2)b<eFkZ;y{_taLTgbp8-S@7|faBnG&>8&LX`u7MW+
z884`xohz2Yty7&vi;(zeCYMYd9+f;Osy;qZQ6yBSD*gv7qOR7|Wz|ZRoF!w)A&eJ|
z-=m?Vc7G)NT16R4n;J3M?+a+H*WI)1;m)zH@4`@6o&Kf9O9ud{JsirdXXII{<pV<4
z^BUYdEWvCpl@yDEg27_TMvgqd9QYWDr!@T-!mc=*Fo;88*sFc?_3*gF5RSjfh@IWR
z>AU6~_QwQVjNUD7bUZHQO&n&mh9x+=!6ngAvD3P??{))Dn5Js|uwg2cH4cYkag!6I
z9Hc5KUCOEQFrlE4SIWu^xe#cd)5t~!sjtV+<?HhEdLR(`Mv}{gYu)O@CzDeYv(oUc
zyH8alHmJ3br&0Mmoq3l3gJnral@WyURD5~J*ms)joco`(DVwi)iJvg^wYJdsDp+6a
zMeD$h5hfu7!k<+2efWA)HhqHLNq|||-M!~Xkxb8yG%Dr$uz$PfCGy?&gTYSVsrn2(
z^!cs@QAW&H4kukIEs6s50{L<-yty&aCuD}*f}z{XgD*foC8OEGhkZ?uKcxtIS$(kp
zB|>qM@Tr9A7X>BIx-*rAx~f$wN6T`*$mo<`jb({OhxOZH@B-jm3-ruc>k4)OCzh_T
z#bgnqd`|GEPi(}15eLelRKXE@rXVEqueCNUF{_x&_EOVq>o}G&d!G!WhE^gnl=~Yf
z&a5@-bB<v_51V@eT9|Xu2=L(U(+IOdQ`RQRErgEmg)2@yl`oC2mG@3x__jLjg-Q@c
z4VqGf4HXlu{qBsfJNVRLmK#(nPv5Fpt(PJ4*|^8#a<B0ukGhjrO<R7(x=svgUtLPx
z)aNHIZ0zX<s%SQ-uGrUWb=ym1wN_Q_9A18L!?izA#3JW&r#z^!zwG!L;!Auy#Cst|
zG38My3W2!Lp((yJ#Rz!Xu?=n<%^eDfiA_y{A*WGKROF{hAt%Ra(mBXlX$buRr=|^+
zu#;m#pW!t&7S0(fF+*LR1wA^$%vZtr!6ft{#~Qgt<(;wX-;(k9oC(=qD#+e8D}6nm
ztu<pZL_A5iLCQjS%cEVQI^&VYh;Qu$-3|*RqqFq&iu_N+T8Dj+wx4Ryn=lws+-dS;
zmxn8%{goFw&LR8k&FJZiKiFQnPDHa$;c-Rfgkec8RY^<d$1^)btK2fU`p7kfou~Cq
zSK4wzPoKU>V9&GFqSnj56i+g8TRS)!Jp2}^FosPZCb5@y<ilyj&~fx|v&{pVG!?Kp
zIEe84BR7wSJ}<GOEzmjdAJf(I0+Vj|6x_Hh@j^mh@Z?UDB4`Vp&^noGudMh+*{!Qb
zQu27eOSrHiMgn3?C4U)u2PlUz!v1TQ?7Q_|C-f>#jc_o`A}J@Bgr_G73S?bgO4F9(
zs@iZFuEg$&YG@<?%Mc%gv`x-p0Ho$gJ!+M_M7jX&2e|H<bO}~+at<gJ2}$%T6+DbA
z5iN$76b&+xH&VD+9ToOSJyq#)LzL>dM@xM*2@~C$5;UrMW>D@*8bAHWIO1GkUD*-e
zs;{csnT@`hV1q~f*3Z#~K|z675ufDgN2sq~{<gCDQP}&^3|x}qlKvcRhtO+a)2?cT
zVU=Dfx@EbsGgM6Fg*cOf<RytS@noUGSAF0Hl?j4<P39Koe|@}GtRagI9cP<w%HiGU
z1)BqyR)5Q?sOwzoN5!G#erPkJz5Rv=S}~fwf8P~&RVsteib+csN=MHKnmEyAy=#>*
zTww+OmFq;@EBNRM5FXA+DtdgPk9^^~f8_{on5%w@w4&m$eh$ierJNxW8H|)4JjK0M
z=YBwtoR+q+6{(+mpS9CQ#1y;@<6A{`K+_f2xf52aRYrb4kDY#F?t8dr4~H6*VWqps
z*PQi0=~9OGv9SHFHfzjb4m3gpMG8T_92(-&_bwL<Ki>~B7Rau+?Fg!9+b}dLX~uA<
zPd^USRJ%LIP^lwsxO9N7Kny(+n{Nj@?rY&VIJvVXSc{D}uZ{H65;ID(y?j}<5v_YB
z>WU|Oag~;<dINuZNaQAemRGdEm}cdBy{0J&lY<+fvyasEY4!cjvYF((fK@ar7^W<z
zPY~Q<0tk#|$pOO33oE|nw>^uw@qf;uZy)Y^4t+W)a6Krb%S*(m#}196CH`zcYa10;
z%YS%tfIA#@QnvnC{`aC-?h?+lWqDg_yuFbhi3-UZt9hG+FR9C@r5ZT)A}D4KUmsPj
zC6@-Cd`#F3gzew%Hg8TwzcY~Ea8FQeN@4Z(|GjVi%?mG(HJ38(_^x~M+k>Ew5$2<K
z<u|GP<9^&%L2K*QL3JUA8zbZ}C+4B-?}(lgzspr-WM6>$|KSb3SiH1^fJ*)q6<)%9
z?u7BV!8bn<BDf~oDpxV;?;XSo;jf);&drIuk!rh&6{!-wP%S)My)W0#y82)^eiwfA
zud@wC9K@F};s*yiJ>YEf1MAQI8QGXIkw2O09La_J>0{@jNNQje6@gO20R1&H&xBFV
zezd+U3cI@aFaU4%RivdHH^hf2xMZx<zcfw4<~Hl2T`5`S|1CUYfhCn=f@m#ZhcB#N
zl9DX<i+1(BsqL55v5rqcO%lawY{T~-PJ}+gSRQ8bG3a)T#E8W1O*S<wEo|w$EXu`6
z!@W}=jl{z66#XVcPE&64K}w!tQW`_iAWk947qixb_dTgrIwEdFU_yhSM}P{k^c+)7
zERXn9hJLft@e2+OB-xmWo*rVCEjp;$v?jZ*M$x7>Msd&7IM87?Yc(%gU=IogQ|Aew
z8{==`f!9Sl7u+8)L$M|rcb<$htaFzB41HCt$(8u1U1?2W|LLQ@j3um>PYK~0DHn~^
zgy9#l*=;<n-d8OB(OKCwy3a?^jnu-tn{OwXR5<wqLwS?M8%lsGh_*J4QA%cuO}L%t
z0|X@PDx=r7wv3Cd9;6%nn^mpkMj9ls7*Vf~wzfK`g~~{b+XGz)PoX%wa}b>HRjU4S
z|2nQ?eY^72)=&D6(kS=*t|P2(R#=NK;efwkte|n4c#ny3Bh%#DJMTnS971lKP@<BE
ze@A@M<G*wEa;$KKTe)=o{IrVJUFYoQ7p4(eLM(|H>YNod6H9uwxmQX5^t(cntoTO-
zU@m-bj%VMBUI>9@en`e^6Dtpu<&|=r8LHs=;aS6AU+B9QAvgEcVYhc(IUh||TnZOm
zp;YD(u@w`{R1z|!czh}(K4OB{M!9X01{dk&1~+E9U$HCAQg&!v;+K1@v|{)IEqX_>
zqhemHRO+3Um6rDGH8@DIA@wsu&TnF9-!SdTR9|NO7V9g0uh;|`>$m+@<^O2NX7FQv
zbuku>59I5+`m+O&ovJ|oM~f5-8>Cg=&CZj)*3&k@mr^Jsk&#JVcYJm8*+=SfV4X7*
zBY|?NK4h`o{#U!d<Yi+iPJHFvPt}9Il>!eSqrE2NC9ruXg7U8u=-F4Y13jbv8z^WT
zOI=}D2V!ACEu{j`^Ik~)BSpP^>qjZ6ub}4p(U*By4PhZ=&;iih2kmVpHr6Q%37`Is
zy*H>*M^rf3c8f)$*VTPiqzV)`f#Ezz{rg-&AnD>OhMr#oxI8TH>Bt3QmF8Gv6IskQ
z$9pZ`Wer|rG-qUdfwLpRmiX54R{2ul)5@qZpH|WdqEtRst+DmP(~Ok-P~5d^W1tUN
z_t#Jr-_hw&Y>>&&{l?}!Rx5A8WKBb?TMMuGd1Sm$D&KB7!hRti;q5o6Vcv&Za(3}n
z)ut&oyl}a&u(1AYamn*QdH8GEcF%wgU#z05&D5%B)<UJMer4l<z5&y_#sfu95fKrR
zDx=LmD4dXCtn0hGwdHniJYHpfPF10#4ykYvky!6h=LK+!!RxN-7AVogOt8b)A4#c8
z7B$wblU?9`zTfDH#v^drCFd#@5-~DRZ0;WD^PTJsFH(aWGy9YWtDXTc>TL`}%9<n$
zD@y)78^fA+1XMu3gbi<ddmje=wY;jo)f46}G6&LGq^UXe75q3#LZKrOcJS2i>52_F
zX9-_&3r#7nOcLF~TJHApg)L56J2DixoJiq;;KdC8WmWAN7b-zR#o!=Ot+C(YSBIf`
zFl*sETwEJug2i+^Z|3Wm-~=G7c#uLnJ$D?0n$vUQh6A|YSHbE!=eHG%v5Fssd1zo*
zIqO{N0-hPWp_B`?r$UTKR&r$v<w68a8bt1%byCChy;2<}XU-Ixba9y6N!2tFK<yzH
zCh&6qwGu7szTEq}F1!@J`>|jf9)dT-u43#>!=ch>HOal$ajU7^Bw<g<(%J%8&7d`8
znd<gNkjp<#Yo%vn(DmVKAn88Wao4|kCO<b@v3f#li@RImG)CR2*&?tUl%Y4!sX<*H
zlF%7<k_$R18zYxs!6I~N@Py7Hs?RI<q-s_`PZf%4GRFQ<PST%eyn%67)D}Uomjjvz
zqXrUEGaNL>OMM^*T{dds*GXK2DZ+Iyo1MWh98j-;rn-QvKcKV!mm5N$#_U^KqC)f2
zJuYznDMgN4s6{bj$n|sPaz_|u=kSbHOx**WLf=pm6qH{5r@H7|F6pVI1-=f3BdO+e
zr$6=RNtg5OR;bezql_z*hgrtfRaYzCtUBds=Ol7hIb?;bu^eW-@KcqBNSZw5R%2D?
z7$Gwua`=45f?PIJp%vzX<2eLj)N5WyL2~TzDv0X?_qSM|<k<6#t?lc#{cmn)-|FiJ
zaikRs;ilLWCN$cDhtUJiS}@!`gI*e7Ah;(BbhG>%FRF@ltgHTY9hOFmi3SCat6evk
z^)MG@jqNnsz5)d*MHfTUKZV0cBnj)X@Be0Di4VX6rLj7J6y>P-<x}bV<MSFF1|6!A
z9M)I~iu3#diALX1YN3{lq7B-PDD4a=wTcd2@7<?hIdo%DNQr88(jd$!jP7jvHdD~x
zxTU#D*Y1|xtO9MfqtBTA5Pfkyc0w}+BQk|_Nxja?FoGa}IzxKS&m&7k!Y(UrOE>gp
zu4d~+`iz(RzPNL{L3YECTDjA*p~RvFTi@<<l}Ib?r&HT49GV@`N^E%oP)bz#%RNiA
zgpHC?dbcG#eP*7)&X({e+0mqj=B=_M4``j5&hOGsws9){k7xX2C75W^I=#!8m(MPA
zB+!T$PM(yaKz^?=+uy?5KP58%9mGlSpzepQc;KZJS&W*!OQ{n}V_@j$(Tz2Yk<gbR
z;=;-6H`x(*l3}BJ8uN^znu-<W$6bSJYz@`%$Y`DPS$~@Pn}X$XxCy$e-X}nf*C(x8
zJb>~|w7ck*OR)UW2(Io^8-6M0Psy|c*NdY6SbBJTb6|#n-Ex&E$Sc>}i5r|5FB=xT
z!ya#gVJwktgzTewCZy1&mszJ@$=-)@{A8h%vt}izlYR<__%o>ZrY~w};~a6`R*w?B
zipoZ%t1CTeZdS-L-!yFbi6)VnXQdv$1DJF!RA2MKX`+1ZML?9eoLQMRdFt@}n?#jY
z83{TiCb1Zf7yKBoI|Me$)v#es?IdBa1SJY4=z_M_`xLHkwuQtUPwop86-<{>x#_Ge
z@*CPPVQ%T)sWgHv3`G1C6<O3I#)&GfFMGx8@2;M3l+t*lB_yD3FErkTTXTX8(P0U|
zosbpFW9JtZuHCJz<ju^qCpV&48%Ge`P|}@AM==|<!N)UeqD976b^Z-&JeaT9hP6J<
zFWUB(c>6Y*90V#%<OdkF_kCN!zuZG9^ktOnI?S!EVHjN{5gB-l@v_b|3llg%4$G4U
zN{=+JymrO!E^>G7nHM4Ybg$6SgFRJsTduaFU`s6UAe@Ncae4eon`C=y6MlaocQEGS
z(nTz3`C`@<D?k713LvA&<A*7&d@s-b9hZ_B8H|bguC4y?6+!pXkHq$;AI0t8%vRNj
zJMas}l~kEqE#MbjiG$c^F#epDuo2zqcRonFXjCUbH->5EAd_Io%utjR@5Skc@iJV<
z$WSeu?N|)4_v%XM>ll3(MKyB+4N6ergC$cbEs9W=d%pFcx{^FSsFYD=w+MQ>Z!z+$
z)bUJilOM!<*tjeU8<p*Zk&>T8Wyo<iH~#ZS6=#m*4HJ;Kq#{jV)~*Xfv78VZZ#FJB
zAc<SfpM#w`VW-cfT2-(tm>^j|<AO%7dg_2p91{wRr*c!G$>3E-cYC+eA>45-5j!XM
zN#3*BnJGIov9IX4THq1&ncH*XCID(LZh`ZEK41dyvVXY^kCSm~rjV3iYg8u(_FD|-
zREbRyZ;E~gp>)FO9JgCszMtIQ5iI9YZ27sXT^5@R{YOUlpA(rkwzpU@a*d^7GvcYt
zGgtKfA6Q2suuk+oof@9RqVut-@bjG%0#A!oJ25lznvP&}TXr%{z?uN5f8gl_mi6&`
zC>91!p<C?19fSu1Nlpybe-0B8R7(CJjp45Ri>Wyyi`HjvT6T#AaB!Rz?sgUOLx%`4
zOld~5x~sDrj{>+%2Ia=&xW6j7DVv*9y6%j>G%_-3_Sxk?I<?_saFQyBDs~;^T{)N3
zvnkcXtx^8S5UW1i3k}-X1r;P<Q&47^f%56Wp<f564ymff6a^B&r^!4@YMCb>5%adD
zZ53Y#(2XTNWC8DzX@5jY+;J07n|(Xg(9+_&BvXV^%N0{5-hNY8)3yde@NBD2V6pnH
z9bHy&<xqb=Y=-DxWCaC<{e@OsR8o<N#`gH-qor0{y{7k(_4QmbWP|LIx1h<A;2)ba
z;Cw*^S6TBE`p%IX7#_&*s@g1N1Tm-4^dt7At>34V!aZ9oqKJSX*xIHd<&^Mo!uepA
zk_xbf!WZ2$+3+p3rz2jh2-~Jop)>hi66x1q{kp=wePOY_{q#-AfAwTjNX;ZCk`GB;
zp0a>VWIK$g$VzZ7xFYR%r~3oHzM?rHDk|!w0mWC1&js0jzinf4e+K$V87LxRdWtqP
zW6iMALAe!L_T%x@Fk_f$m~5E9Q08I0?L6kf6v9pG0D3-v@5x{OASxW7kVc&4*IVCP
zQ_@Z3Noh1Wm|5WsXs_cYS&ZoV>6Y{wMkk>E#@$Y#@IfN*#Q(`OdCa#4PZ*v4QwBM6
z_GiK6BkhdW^qVeT8JDjPNGZw(<#y7bJ6~JJs~sOCCjyN)_)jl#`Qr^AWwdL}-<hNR
z>HqKoT(l=IBhM|5l7mlD`pbwm*6FLdj3+^6=<6?M$`74N1?@K6C*pNi*mE?DRwUBu
zfgCx?O#?GMX#t10wY<{#p`0Op4>x@4S!MpTBV$(JqWXt-SS*JlU>jvcb`CKr!j;-A
z8eYUO;!tZIwZ$li#bsy09dyMh)~O;9Z2lF*QDnTGAUFjfT2B1O*|PHmWQ5M8t9Ndn
z3$eiA^a~{(pp+t)YoEmUZmw%hpkQK1jnz;QRRxFh;or>@d8iP}_9Ja#8fIz*qPbn6
zIM>!u;~M1o%4*t8#J9_?ms3qJ@Bq4WmGMdwVN(<xlE&9*xvy1s)$gx<rgtY04|k8R
zmj{KW@jBWZRLRrMmRP8k9l-_#{&lVJ_AiXU*%Xa`HWsl3DeKY7yqU?PUW!H$jS6-u
zQHSqzhp~Db#2mn8WB;U5f0Vxds~gR)0K?R<a75~jIc4v|8D0nut14pVT+6qhzsr+%
z7GT}Nu{zX~6w|plo9<Z(wK!IanP|WNg`u_J`~4OpN2WKm(sFw(pivTwdMF=ysi1!{
zNsb<38h|PrZME56J2C0#x@#6`p^lHMb3I-1rOf|8Q!hNM1V7!!_at6UE9@O6Oo|>V
z(Xdno=hv?_zO|ObuT57~T0JC-qB5MfbHqeo&d)e+^P(P`DFr=`$+Ro<Dy9}&4fsc>
zMo}sImSY$KrW$N@eC|(V_kAhYwJQwhb<dW2d-1?Bm9?`gpKk#)gLj8vUuh_^)u|hs
za9NE4LCKE1k(ad>tlPV2Vec#X#AUEI;w#{zii<P-Fxja#wb^KGX)*gdRibE@4pf<K
zJU&%)%_I&1(+>9of#lT03HrM(G!Zq!v8fV`)$Q#8qo0E#-+^WKWIZ_dvW;kOwuZ9B
z<0+n-nCl&znb|1^!`Yq#eCO@;waI)vJsSr{KX9y9JFev}(%WiQMUl$VLmjueKLi8Z
zv*g}p1iml<DDr^_@U6VQW-*S0``raRRYBN{iFIP9{gH$<<{_tb;*)=1$<<$II;^}*
zNTrj1Fj#7M<)K3Vg-3ruF#BK}9r&!JWloV~oWFE_nb>|)bn;bgc#$aIsn-9Hb@1o#
zh#^nA${f?V;m3oMYFWNoi=24~{;f-g)KJ`xfE!MPd<xm$OM9ZHhE)p&=Ef8{Ocmw1
zZh)BBv-~=@nrGy;x^Wc0klNe)CSJd~s>1;a4oL1R^yZ&9WP3q`EYU(6KPT3B&;;CR
zdF2Pv;uDp=f4dGk?XYZ`?7SX}MqqrC{^TQ<Cp-17;uCVLK}Mk~V7u*i>m}t|Sk*UF
z*^+COx(|iE8t@9g61l4+2T7@z#PlKP#@F^(p9+!5Y87Zn*RBr*Z8SnTNq0tt(<B`4
zN(J8J9L@I%(8rtBduW7G#VUAl<daLLP@3g&s66yHn(2zOVocP$@1nmDUMxWjbtD&T
zet@maLLDPs2h`}cVZm(M6)!t`+$iDRAzzP#N-af#UxWZ(;pEO5$qL8Ffnu#x&to`(
zc4gaIsfGUb2j*@GEB=Yh<<;l1W<rWi8))<vdjTEku(<EPIisB{W|$3G@>J(pbZene
zP%Z)12ocDEhIB5+;{4Br!XJr1X#xdYK~o**LS%6p{LcZMB&P=b`g2`2-Q3r4v`@yZ
z;h@hU9zJ1CeH~I6+#$MZFifh4Qd)W9!ATct-W<YMQPcbNxt=@ElY4$6E32OFX_QTy
zX){{Bp-fS)&>n>|D8BCBfaxpXrW_CoxBYYd&7c-d|KZ>5=wW8PnwdII&Klr(Q{4(L
z<=(qvwpGA!IU=Dn;E4Nb0yzr(Xvr)lZdD5&Tb*dEuz7eb4s14M0iUz{!#;)5hPL@4
zIjKB~Ys+&)D2=AX@Sw}%K<s80Kf##h+5q>|Wrl{2-p-y*%Y=+@2Q0duor6$Y^j(p1
zEixR1Gy}JN)SvA!T>xMNJ*)DGkn(lkR-bC?Jk19*ANwD3NIK;nuFO0!z)W+IEBO_A
zhd`Oci@QN)l=J%sHXlthFzyUEI2GK40`P`DBt-6cyKmfV$PV79%nEYe!2DRncs{2h
zrLW|BSiNdZ=tl2W_wl8cT7$}`xGtG$&dxu{ijh%T@9Ne9LrVcSFqX;`A%t7K>5Wts
zLPe7FQiqbyhAz)YUn5=Qn<3>zIfQ$dw2Y4xr@e;xS6Z@4Su=bf=Rze@2;uQDe7ipk
zlh)~aoIJm<fSvsuxMMk)><_RYZvpI(LQYO7Bk;F{Ywgo&fMjSygvm=wT3TAoN}`WI
zXX+jvR(D30M5AZ<`7@c{mWGK5tH|bASDGvuBV%Dfq3Q2;ymFKUopx2rw3+b;3C+KM
zJi;r<vjKt1y2kD=cJ|p0UqMDrOEbGb1}&(QE;!Qa2$bRb943-s=-c%()>CFwS#dwa
z%TB)sf;3)-e$f3bCi$Eq=_BQY%)Mbi<p7!p`i6R`CjI?m>#0#Hzb#IsQLI11!psaV
zr={Z95+$ME%E!<mmDq<Y%eU_CJVvEEA!_En4*(_pd9Kc?yRQ%1$>{`7#HU%0ze1$l
zJ5Os*8(DT3<D1i&Wk(Ys#LxHfTW%ZqK~1N}(<qnDpsYnoitdeNqbk1;LznyB7PDfi
zknTdo=rS+JJ$XTnqKW!C>c;DC*KRoBJ4w1lP>Ftc<v9P~VY`do5p4Hk&NR@WHGcqY
za7IzF^O+Mc3y^1~{y*~GGN`Vt+ZGK3g1bv_Te!PB1W#}a794^*1b26L4;owp0fG}8
zg1fu>n<P8?+vhv?ym!v4Ten`-tK!EDHOZQ5%s%=UqxIHW_j6X81#GU)5{QQANK}Iu
zrvMnLXUo0SYC&iTrVns5LbXMhRS&U^oW;SRy;CelAYg??@Wo?pM+EYRIypYvs%qoj
z=#(5je4#CGUNh0=_MAJqUshAd$TeUjNE<`dn$HleZ(!_QA;D^ZgIE)ACP7HVObZ4;
zPG7qhUe%`F`kZS81u=eM)HA;^7{w9glbUdLaY(t&>AQ=cxvQ@bTRH6ul$7fqBxeBm
zt~m^+a?R4w$6l-};?v`M_)VZg21|}aTpRZnUeT4w&hEIPUblI%1n2>jlL%CD>CWLk
zR5siM12WjzWRa_6tMKe_{x#6cAZENG-mwd~l?E?TTpUvd#hyj*$z3;Zzs+ubkmL*P
z;U*Bx#<=H&T`(&(3nsu$s~YWdPCsLy;Zo)Q3WHJFXz-l1uu62ysystap{;U>_9Wf!
z0o*dkD*>EkKnc3!pieqEdNgSdGfm|uo;6^xs0Mhaf+wgHShP+L`>(AGVaKu3d!NRq
zp}hMJ$0Q$Xd-Xk)M9G8>eCKv|(md^_8x%_9nmP<KGs*on&^Zv^rrkB+i%7jS-nyI+
zH}e|`LWnYn^3jZXCnbEfY!3nJdH!^kto32?!q=~Jk_Eg7%$w}{nDKeae^jjgwpaK|
zH@F*+{46B50c7EhS<fr`D?C+>sOw@PpK?<b*rZ1`gAfgISV-SDJe%ndal7)jG~ZR|
z-tN7zjPKgKIt<YD4XCVCgrM&h*MOn))b0acQ4$A3MfIsYdv{+&YXp##RtIm{!m<?E
zfD*=yOT<`m#Y5|x1E9}*WE=lzn_Y-J9N@#RYId;QTP?SZUR6Ra;MHT+?}W`4(V8DF
znaz%Y3z6$+>`&oelHc`x<p}4k5g7hVH3l5Abn?8CcEsYM$>Y!qan_QQiC}YJ>a)rN
z?xBhm<{a3m)Q`whhQW%8GaM2(=YWN+$x9ILI`y#Sy$EJD;H5_d0YxL15mTmB3mb??
zFvbpvwwwXbbhWh6DSNxcb9MkW2vqzPiDru-j2ylye}^2<k)6^0_&tQOa+E_#82|X@
zU~(uNR@@s_Y#&KIevoFuWQL>T@RN%S#495oGKwOoV*v6uLGZy0qJS5!2xN|tpgLkR
z`4Qz_%N3MGQXSe054x+<_3EPRDs9cLkbeiIRc8U(-`}4ZSpR<1{*)1KH{RW4wEFJu
zHJ>Xy@HDaba)j-$k6CQ;8O;9CSS8|3Yyv|W2(XT1<P@uSE2L2|F-|GEwR}4t22<F4
zqgj-I`d{2lPTw|v{#|@fkM*4wO(T->9WGEpve{7(`?C0;R!|m)qLO`gRb<q&Sx2V`
ze$iy_E1%vJ^Kg{Og>W{OkK<ASc!fm-Jf;<HnWc7!khcq<6x-fj)mGP)7AJ#%y<s<V
zN1gSOH-K>T?W!WT_v5OGlcJVWfSFex2inY33pduN$N$VQnG?(AL&nGm5)dVi^jo<A
z!VLdJ>SM7{VM}k<l9!gy)nmv=iLY0{>aEz223XwW$;PsH*r(`fzzJpLQh6amgn4;l
zCT}gVR)JXI;fs<qs}^8BH6%fKaAkoAVigXmhHtto&U~X&Tpk!yR94E6r9Ev%Woa-B
zUCM{I5xf4gT>hT_=8MM_?s0M4lks_-E(6Jaxve-_xCnWLh6lP|i396Xx(LL;GtYm8
zq7Gm&WjowJHRe1Rv)K;J2Tx0rp~lna(o$6eeGL>a(ToNZRH8~=-|b=2XL^1f-3VyV
z?fc#;!5$dix~ec3K7cawl(`r?-+FoD)bWw7ln8{;bqc0#*f4+MU`c5RV{Ftay42<&
zv{VMTy8&37H&AFo1Bv;!Jf4LBOz@SgqSyi1*FaIiWC;U6F-<V3wb$M@Bl3cmejOfq
znNLf1a@1ZC{kRH=N=iCwde)1V?F>5E_&zp;NZ{@L@sci*x!kI<lKxw@s!l;6DH8>(
ze0d<z{wid6hXQa46520xvL-EZ&}>je^h?fXs~&rc;&4Q7MaT~UtF&co-;`-j+EOz8
zDLfR82Sqv=`~)T1+5HAwGn9e_B3hV}JMqd_vpXb<dwU8og4X$on=&xH&FgZTn5R{~
zD~5giJLS(FdaI+oHPsaj(@e-(OQ@|=3^mi*^DZ(+)SJtbZ!&qUd*`$J4nQ}Q@Ew(6
zRBBcecsC^ZBn^$qK~!e#Aapa`aurJo9gDU79MJ$k+hzg}3;(Lt6A5vn9~3PH*h>9q
zK@U1NM}~jya?Zn@9oGnHs{`P6?!2p+ex$p@bSjsTm+d7eR#sr~LL_}gY3X`q+0bSm
z3Mfbqr8qn3wE=+Uf*C*MvfQ_o?-c?8gtfJ`R+Br<n?zbYO7D%N#$+yT$K9aYi@nap
z#pv{zc7WtwXSLAp(|Wc!Wb_RgMwsjqMebnmvUvFNmMU0Yv7X^O)IJ6!HI-NCr-f}J
znWaz?@M3vCnj~ogM27(UBit5HLeg7?r%gYR5c7jY5(W~Jt#hMAQj{TrfWuuj6CRtH
zAHbJk)K&PAPRqrGBN0K+F-50|<3b+P8=C<aFFG(HS+VjBNdRbjLk>uis>9Ga00C0x
z9ppvut}7rw>f!Kc*TZ4saXfrs#Jgj59%eFLfZ&72{z4l~kJN;GT>@zDRDpmy@jZ$d
z7WLy_oY)@nM24f#;<p8rN8q;7)(Vao@ie81xv%hFr>p(YX0#NAy|D(|MrlRN7k)a-
zv`T1LsI{m8X`GcF<@_-`wWHP6NNsjDJ1?4v`K{71?>_N>THIT+Xl4Z(R+C<66zbH3
zbKX}-BWXRkZ?F{h1Qc)YgU~esvjw(>-Xsc5UM(6Ycbz{Ji%NAbVK*Uz?4J<vuV1@a
zAz?Er#1zH-l!QDfEx(E?xLkaU$g1Wxw3mK5_vY2ca~)coh1K<{pkjJslv?B-TY}>Q
zjNn55&m1`xU1#Un&cV*O`04CH6xcArSr;_ZV)gAa13QIRp=nvI$&F0A>=iFEL*zb3
za(=xl0NP5{6_3~OQcs@?;%M3_d;oj4iK=LJo?#sO`fV=a=IgwY%`pibd2D=nlM9z{
z3?t)28*&fG(jcH|2wFs<;7oXr0(@h3+z@}!1&t|M%4DvVx@6*1)XbJ5L6%3W$Y`SF
z0bijp0BXMVK+q4(15x3oE=b}8y7gSN>~kVD!N&b-#o7$4d0xx+R_d?}Uu>h0Q$yT1
za#3vAwV3agfj*PlDMk2%FfBZ8GnIl~ooDa-blf{1xtiyeLxno_#{hY~@EOVt=6n+g
zQ0mk(%W#sE0*(M4LxZvk^)4>0hpJVMKN8}%P#%e%+C7x7prWbr5C8#h*f+0*v0~b!
z>?1y0>qnoP{Yuo=RyiVa^S>$sf{M?MJUODe?Ts$tAu{*Rt^`H;i!$+&FwO2&NxCSy
z{6OLKxSVMD7loy;av+&ekK}M$@UVpj`*?2|8XFf(Xgo-=2LRBqU7h(e2%w?Kn~Ek@
zR=YzTaD(rZDI1&(eyQNI0zPF)#HFP98;v7kP#3Gt9791sykG4C`mt_B5q~=!5~Sa+
zJwEw$!h7-Tw6ofb^tTnrJ0|O%;5hw5!8Z4ZuRLLb%~wBHOEYw<2IGJh@ga$1&Ve37
z3<pERg0obim~{Fn{J=RtP%>!KVw9tvn*VKg2!=X{uU{O8*^o*<gKr!L{ZKW1c{O}V
z2!aYqCcCI3oHUCEJdWvNXd9CXICat!c=@5ZZ5V&p<JVld!WOUs^6|S5O;mZ2oxkmm
z(Al09S!PJ+?w&yHuIt$y#+!O`h%;Jg@O5wE2ntmXv_f(*fR}u#34K=<7YqZ42_b<-
za+urspBSE9o4ztayZ?B6>3-9V<$QgNA|6R#nIv>W<x0%xIIjkwT-@1TrZ`*cgMDQ#
zq_F1k>s7t@(IR3r0Z;#3I=$?1hg0&bpdlq{G(9;vnAkUvgqSJ2K?hcs<N0kN{u;i$
zXS1Q_vZ5il@?feBNq!Pf6hkJdLKq5Bdf-%<#!6RTEHnawn7s^qaqzc+IXG>BLmyaA
z3^geqo3)-5z&5@6TKZ;;3)~tDJdfsdCGa4*truvO5EU19?nrn1WWn~~!v~Y4$Cr+i
zZkPK;8j6rihtob;J$KLUOHYx|*m#M4wc%Vc(l{L)Um+1RmEWguU0I77OyAOU<Hq2!
znSY)Bp15<S-KN!2k*~(%!;k?$f(p;eW4DtCP_8OC`YRa2fZxn}yWik@>xaES0WCCp
zm7Ks9bq!T-(0l~?F6g_*yYKwcYXDze9A%j;W07j!ojF3#M)YfI+Uq8-IYHAM0C1>s
zT$IYAFvk4;X4QQ27wy`b4IEfvvP5CxXcc`>R?j14E{wtOS3I(ZG!cGkR&xU5XAkke
z7(YNFDiqrjzM?uxNMLnH^|pW&D54Yb6NA-TEp`HWlH>E8aea6Hzn|)lLy$EdIKb*A
zWrK@6`o6mU=q4%E?>v{#Km83bY4`}S2m=8dD~TeB23SH2tE>PVIj#SR6Vt@A8^*~3
z`g4H~XlM05v6*TSaRWx3IsKPO(aDFu#`4co{}=xRe7sDMo?p}d)GrG00NC#I+EY9a
z?0>Qx`-gMaKR+MnT$j!F|Hjjbrv~adi6lSkCjXP$+&@3z7d_;+7ad^45qJsU(*I-M
zt)m|V0JTcS_+LIxUs0&^A;M&20KX*duF@Y~w?O;4T&E))XZ%ZSfe2j4ldY)7T&!Ve
z@)zg#Z7}$0Wha1IYhTN)_#*}Na#0V3xO05^+pmkaPQSSbrM<6yHk{Lv5K?XnP!W2Q
z9utCVyH5Jox0C9$x`sbjzZ{15!1W9)lu#?G;il>Z(c1FUf7V+^GPGUeKewy*gF#yw
zQ$98NH={Rn4hYX=rdK^ZJpjO}{{$xh*FOGK@+6%=3v(+`0Vgp2L@piI`;hLb@*pt-
zT>&Ei=RCTnB!I=iyR`259Hf^23^*~rPX=6s@65)~XOq=A>vtkaH2n1nYOAZe07Aad
zlYBZmLiV(X5z_o`E~%iX8lb4<`F;yI31F~<P)y95TZETX;mV{lcz~r{^)x#jUnq2P
z#{;zk@95=P$>*g^Kr4LGs&;v%YK;>utcysaY$rA(&fnnq2qVnEA03{bjrXelMB)Jx
zV}g<8oY%eONgkr<w=UQ5#7^SK$S;Kc1At4Ug!gxp`j7$@jRtycgt4jlYIN%9<4^Kp
zl$;G3-vA-Y>}&<f6)ohI2(U^Y5|&M)tBs=_@GJ5rdg3;wK3MO)4_+iqt;*z}<qIuj
z&a}ywkzw>&ai5u+H2|C~fS$WyLgleYoxf)H?-c%=ou9&i*b1q50><rEuhhTeG6dQi
z>ATFQ{4%=V^sp(cQEGS=%d5kM(W<pT=ksk8z^J8cJgW&61Z*ZwY$}>8B7U8kCJU=K
z&3_T3s3pDQzeJFL_|qTm>rg=XmUPEMMVl6uP>=9*8YKC(guC_U2IjL(yS$H$gp_dR
zi$%vMSzpwiFdtd{xxW=Y?0)<9J!6Q;_*`dwTQU!NBw}J+hBn=`L1DUtn=`s~=qFTG
zTya&8OR&Oh%{?!((ihe`CSYF@`)4f+3U&zqROzc1dpQFpZn=mnvHu=ve|!u)e%Z=A
zv#|ac+X<o&AD1J8Ha9obs@%C`hGeQfu8^d0J4jQCI*=<_?9=_&f`T0VXuG7X#d)bP
z<!;%mA{MWa@!9>yJk#CrZ7m7>P%!$?vARZb=kRQZ*nTc)Oi0aO5-sJKcsDE<kHj0b
z1a<=m>37aK!{!XqB>BLBPob&=FQ}`L5Z8CRm#hVV`sB*<QQnP&K3f&Rh4<*T>yA2G
z22dz%yR-lxbR4xZ>~C*qYyP(q#uBH9|CSno^#sTrr3<k~tAFoY{v;jX<vRIGnvOTI
zxn3OlfBu{Zs56k5Sy&u|->CwPymS##7ZJ(BrAgMcI1vvTh|Sx|J@z_ETVA5Bja8am
zR?YJR8SIyzkMkZf74Y;%-vzsR@|Jt@q1i04`l<H-0{0-TjG9ZQSS%t5i?#;Tzz<Pa
zF4OHF#6`O8{Esk5-+KY$Gw${i$bWw$3;>$0)N)U*LkFb1=W_SIzmYG-BR=)@9K&f`
zp+FC(y|99EQCL~QTd+J*WXCP8x(7$askD?^U@5?KZ$Xh*E)&V33PvAmyywXLX?EmP
z8t@610RBtuncS+&yJ|8|f1$i*1`w%u*!#2aTN_pae?Hee_wQKGPDOuZ|4;nTEpTEw
z7mc2asyN=AOGv<8tIh5C$)3B7y9mbHIC^{KQUD4X@BHxB(ztjlfU`l2LFU?Qt^`2E
z4#8p?y+Qn7XaT?Eil9&@jojbR(AhNsua09HgoO0{8{8JM!<OS?C`b{!t#<cop_Ru>
zLjySZqmkVrp9`vdxvt1D7N26B8Jnn|&i7)6c2KyHnQ3q&g{(n^_ODV=dQ@8JSr>@$
zOSSk6(8vzZO!ZtmtFzo60zO`ry$EGCQnNgmxP}m2X!LKUj3lL@ldZO%%_@O@Jf&$F
z2D+0I27a5VK?hH%6P+zlPJY!@nm8MT;1;xZce&dA1-oqcxdsSm^gCOX$m1OC!5<!(
zSzb!X=OWvXLg*zjtV|9h;(i~UqtsAU3&RBrI!_&TJL;`G;v5}2+bf>L^zG3P(>Vir
zH91DFOBaue@YR5+s8SfYGR=t1gmk#JeM(;GXfL{fS+=ayf%G9^s#iu<ysIp=!R0++
z^%zD9#!WK7*Xf`93>jFxnDvnPz#O{3ATM0Ly5fIkfn>&<so|hdq3HxT0_g9}5UR6W
zed^!GdB3&`8>{p|Qr^@1no|BF(8PmGOag=;mW(Xdwy~na^df@4t<v_RAc%#fj$<4Z
z6~*HT@i3FiKMZRO>`VE$@wbL|Aj`z0yT@6C_O;B?MDn?D4fnkXUsb@Wz{bHhI2i{h
zhejczbhzW#tTj!!K#)i{J0_`XX6xsEIEIGz%P)Hkc!osd5(f4H#ofCZ#lDjuL1Ger
z3!vW@r;{8pW<@B5pXAMUDfe;ly5G#U;pb#GW~U}*N6boUgC0-$Fj+BIvqd}o?HWf$
zcBW2%#ci`+q1osG^l7hvMJ!sFfucV&dawV%2McBgIMYF>%JF|F9LOJ>$s=>8Y0T0R
zlW1_A$pZL`Vc~B`NDs(Y4c?wCtH2A72t=c2IG$8?0FDn1-tYqjuf_Y}7MQ)(<pe1Y
zN;!^I%pbRQ;hfzWR1O)5h!0abHWY@JVBU51<;9`seAr;*&Fw@xUfyK!69^mPz@~F!
zrb7s}FI8EB)7f~f__HUhjGXR*JSUTE^|H3m*_no7$-!Cmqo~3{^~b9^gXs%ExKG9_
z-~F(s*x<;orKG^Sa{Px2Fh|-DF0fl)^a&7T+T3Z7rFbB}(3GI*5_r=Qoz8)BL56~$
zahq+%$B>kQv(}gETo5dhp98_th0U6<JpOr574XGtC)22j02*dhUL}lhhOCQjY*jIe
zU;A=esJmpKBwwT?8Ves`suW1R>FfG7H}Y}wE1%&9Nn_m~Go23wP0(EWXPq7lPP+3A
z*xnCbn|yHHB!^p({9VNlPG_%o_l{D(F3#fGFoCWuBhIgAWM4w(w{LrlS)<%3&lduc
z0f5T#S^Y~u{MFB2jCb7@wubfF)cSw(R;79_7AkQzv;`|ZZ&Zw)sc-$w;&c;t>u=*&
zna5^00Tl=!cFH;N&{XAM1zL@f!m-Y-4J3Z_91y?1yRFrTMJ0ceGn&XJ)S(B0<Eum}
ze~e$=KF}@#Y^RM%owlK$mV2@?^TT{<C{-#%wXOxstl>mqTdUkI;aPW2*!nqnN^5h!
z?sxc~oskz(73n^WkEp*_i&f0J`|86LW4*VBcED43U6b`729ao-MZ#HNrs`<n>vX_f
zd38J?nYZ1{0YO3zqQFgjoDVB#ZTAmOF8-+z^gh>(j#88Rg@mC-E<?MIor}L(LmauN
z`091Z1~8qY-F1nSwufW|+#^d47A5ivV9OAc;bxoh>LaN`u~bbLFwYqT+~VW?0ow|e
z80kp>!$QWCb-VEQO?0i)Okz3xKqXh9o61l{h796BBJ2`SrOyR~-<4DbxpjwWsuh(t
z-&bN17aHw=jyO}{t<&vZUR~YYd+xpfU(<9do7oI9Q3Md}W6E$w@fd$+FwYjwg1dW|
zU%BJ-IOe}g_E}nzH5u~cZ&`a)o(?y1KuV#M!5b!Wh6olj(t!u)gS|pog(k~%J)9pY
zQ_5gI<@wp)H|~Pd&{5PmMN8V)gAl0mKj_5^*l>Eyd$tU^=AUIm&w29?<JOP}=eG2W
zu;cEno=9+3T}Aw?oM<=tP95eqY6w$hI#a!xhFe7*MI?%^u93=&D$huLCQ6=d#&M#E
z%l_Ngg_IQGpT;gHHb+e{!h2+uZJr&VmVT9|VG(nXG>QM_(PBF>fWrW_reF*fM02tq
z|Jmo|$0McfL~jEbN#|;nY4Qi3Mj&W4&H$Iz<)fp(-rhoNc8jEW&47Q8*uRX1SDQh9
zZ6f*zsm{iXe&|i8VJ9aM(Xe6EM-qz7$9`<-myrOyp_yrF`c?|0WFif;XniB=NJkNg
z%1;82UPqaH{xg+~h6VZAHNG$Vr_EnL9obqC*GCX^UbMPR0Gs1WK8U(^5Lf7h%aKHM
zOkh)p_Gazm{hZAtTgvP_-8xA)AT}Nkh^K&X^u3_4uYkNb3Lg&+aGc|Gju+|fk)6S%
zA~iZnQ#YeL>;@*o7t;qPIeanqYXQdE>HzBlfbEh6{Ym1Fr%P{{5;^Nl34?STt@m_=
z?eX?1m*_&PZ&)0<<LOeHh5EaqCFEIvQGYfrQ_BRuf|gvbR%L=Yap?h3XVE<+owuC6
z_B|>E8OhuY>YddrtRI%qM+MCI0t<4Fs{p~gmciby17Y?FD0W+2OnE)so&6msfU^f+
z@R3o*WH_Q5lZ+(lSVzDK;Yg72c_DU>IH<wp<xcPkBZI-qE9v+Z?Jt;caXtKq-&O!`
zBNiw?^RjWEsaR@<mC%jid4L;}31S}UZ)oZ$HFLO>Q6N`>mTzLn<a7U2<mRN-(?4(y
zpQNsy6BbKzkH_ZYA5J&de7E`VnJu*}E59}6;wAv)vtith9pIv5NY?(|vuvUU$6^6v
z!|*Vj3zB4jHJR}V7+k%h4FpXN4qiAnzwKtH6J|P$xNPCG85dXmv&ax;rV@6?L2N*|
zDm)g7wLMp8?R^-^E$a?Q?}BEE4Dsj6dY8{moP^6L_^+?nDjg<WYHdoVuY?EWV0{$u
z!;Ux#$Ui>~NV>QmaXo64QQeUvhK!lIaw{_=A<vNe<c5`T0f)-f+wPCzV-3!6Lamz^
zw!X`ISQ+9nqA@>RsW|tx&B6ogGfA0g5n*ghdp(n1?u~u;cvdYI9<XQu4_i|4c++6`
z%9oL9hsK>nTIg6YKzO^G60);4D?Nx88QP)$D^X04Ok!AEhDa(S?{;%<{CFga;}zG~
z+b?R%3&W$ZZSiR<c*|k_Xo%?400&;1e8O^bU@9{@#h=FbNoYbKIBO7|goEmisz^?K
zz6Ia=8_#6l$S;?X=8(?<2a#~!34DUUNj3pU)=w8h^^zdolH_#xVy`8P+#4eM%33?v
z(3O9KhR3%DJ*=|PgDochQ_lYDYNPQ;B65)LgT-IcD|=mq9vJPt{8&B75P9G5R`Ol;
z3rHisXcUmxdR*M-BI>aDgryEaN7;iq3l|QkWDO&)Mm3&0!{p9JnH?PkU({X_#|U}S
zZ2#Nk@tCPZ6h%LJ|IuR3eyP>cM4*u=OF`)AUCy}KdT~R0edfzuR^|O*IvVs5?$WUW
z=L$`S%d^pTtEw<;?K0z$4CkX($IqI88)$S)&K-8saNpc-)xa+}^%=WpN&TfCxQ+k4
zGmg9W&N)?f;NR`UeXZVxJwm`6wJ&ItUo@B)SNqGFgi6r;V~19sV=a$Y7%YO1`g@1z
z8mV5%KAIZ?Gd=H!HOdaa%?T)@ayGn2++MGL-}9j$tSEb#N?i{HB*))oGV4W@y7oBB
z($KOdU)#AP8dmd~In4-lp2q<daHUZ75e1rdzJL7D=@J%&5(Z95cZ=}eDB8r@DwQ!H
zUI7gcg$*zZL3&F=1MD1EwnMtZw3OIL+Jj`Wl?+M>D&!3pb#PN5*(`iv*>JacWj?s#
zUVhFfvunTSPKu`=I?iJ@7PWe<!76X+70=8e(X2;m_oWyhiiqqZ06rSE1wG#lp6+$y
z+~pr1->;keco5v17dSGm9e}l%d?jRw4o2KxofYF$qEct(0O2+2+v;vRRFM@1uM9Mo
zdbdbJybF?hs0&&6rcY`+@!QblZe;d5mZ_9fSXc~mll&qcf6T%-u=`%Urx-FiFY*GE
zU?l88lEk46C?!-^l++wzk5)Ge?-Ptf^Ck~73rNXTn($*81@wfjFpS2KrSi?*1%yN6
zfizoPF+P`-*&O_28%)ZUv%<B18~Q5ug)`v6Gegt<XzLUIrn-h%?CGRT?R^ILCE%p5
z#;O=yCJeSO$7l8>GjW8|XT6IzGgQ7;paI8Mdyj!8U5V1$wZELG*`1j-aUVq&&!_{L
zD;L!41HO@$ELit!)p3>A7m^~0QZeug4=r^r=p$F!Pyc*;aa;gYM`$E3TrKzMk_fxj
zeTlC^&DHVT*|V~J-iOQeY{p|@C+C<=v2W6*20<fKMJrs_8)z_thgAm$Q7K};TaPj!
zvpv%I7D&6okU@KF(qU&xU|$f<#X!!ED^*w8?-JKdQ`zYq$2l@GQ+vNQr>{a)H~b~y
z2T+4GsTI9{Hw`JQX^NH39A%4X1(En?AiV6U*Y{MWdmyJx3d-;_&|nfK-Cwk`xQmRM
zbQF5`Dohf+zbnVMV-sVGJXqCo4SV&0LKYX``hZy!6^|DXpEuna=MZnM82lTo*mC!K
z;yl9-z>yM|IH-x+H+J(>wSpz=Nm6h8f*`AZyKG#bjh_=h2J*<K^nh~#CL2b{1U~T<
znuHEb?wy`|wU-=HvM+5?J4uvM+!yc;!MtfXTL4~Ic>X6d&zWFTDTUcQz%c+ML`I>+
zYQ(Mps(rN}VEjO3u-q7O4KnIdxO_RAf8e#+F+2WqKI^;<Fg6_bLk(Y|Aj#K{Ehrjo
zBh}ofiXTAnkZ-jJ$F>8t<LFxh25KajLGKl6pM4GMt@y+_tqAypi{S~3a_w;il6SJB
zbf9+vOsXB1A|c`(-2e_xp_gvyl#EPX6v4ou`Z(+EBf32U>+Wo5yi)GnPQ}jK)7J$=
zqSD7(d<u0M4&suBDf~D)!NF#B$Q^I;NM?t=1tK@;^5I1+qHP)<AijPPk2=vkI5boR
zXoaUXu20s~stt?j7cR`+;F2wcnRWCagWt!k{Ae_`<ZXHg+rd@H@&ue(XyNk2zuB6c
z=ouxzFgp$Ekmra3&JE)LSXFgdGPM=BY0*L%xmxMccz&ic2Z@j}@v2~xqPId6J6v%-
z7#)yPVaia<H6Ahtr2)1=$Dt{-27dRqb7`?*765Z1h1iYVzX*kn6%%sf*;0a4HRCZh
z2;t2s|5~?EqX{AG<!@7o@3Da=Gc6<)>u;9`Q@?~oi11jEtf%Gp6%hAi*}%)@?a#Tt
z=BQYIoq${(&df)(MoPJ*JBas1l}1VhT@<!V8QG_BFpoSIsl*g3IQVC~XS<JE_Ko=l
z%+<lx5}RSNoM?lkwgVVb^zN*A)4q_$Po)APBlwsyjapnNB!UTJ&c(!Er02hbarp@w
zfWpyrx1PHL1J#Zg0fp61B?Ze?o}|!}>pNXnm%#?77=2bx->@kvWewq?>3t@TE^+!N
zONQjatzTb}N@@yC4i3bL7uv+m4gkQ4LR8bhMDjlCZR=agk4Dpo;fV*ztrWPPmym8=
zTXDgXUeQ=Y$Y>yAPu;-ds8X}FA22uUB5lSMnL6#_2kUOJI!lZ;E~wFkVHmO+I4PJW
z(3!n0t*q={9Bf5!+#D>iE!Wyy<HLKQc8!c`E(VSsdidNrNGoB##rjR=v$_`5s0J#+
zDgm`arin%1V|XaSRr~u<r*w;@>;ST@_d}EAs+x+GJfn@uU@=>%x)g(~%GR2^LYt09
z-fzOsWJ>VSU2~z_C($Vcz0dGtUbSW=?Mq;y`ifHgvKy6tTyVm9_V^spJ4s~x;azQw
zyR$s3>}6YPAQrV*A!-pv!&&{l0rbtn)+&!N&U7hq4i1jNRF1^(R<|pH&zkka$RGQ0
zeU}Ni5BMBoGS6|#b)9_z${t3uj3CqaKPJXU0@7e&=%4Ek(pjkbCnIl9_D1?=EdXy)
z4XjcxZVIk;B3rsi-Ep68^6H`?<0Z&1N_wj?21*{GJkcaN0!E7puZC0Eo1k9FT73&b
zRPGc$)bM-H<B^$zYKU&1SUs#e`60g_+lA4QP-ly(MhF5hVc-$@ancCz(3+=o`(9Un
zUl#vp%kT~2hZOXJg24|?0C@!LA-b5&-jxBG7uE+sMaPe$>mo5Yf~)-hvS#RQ10pfR
zEn87Kkt7a2m-&NzS%r?&UTYAK!;W%dJe7NVy6+!vhSQUij=^M>8PunHJG?rJSrVtc
zGsk~JG4a93pO4J$9M7e%XFTxcrC9Bg05%gY*-shKS38bTm{2gq?LooAs#(e6m~!a@
z_E6^HtK*bjn(=e_Ls{9C414|n`&Q-hO{_yQd#tG%`m1Q+!mAzh_HuDBE!cHf3JN7e
z`ZP+Ar0&l?9KtaC?xwV*bEr;wU8R&dKxIaBfIHT}S>x<o{)Rb7-ag};?vM!CIapD?
z%6eFSj0PYtWe#2k$3H2NE5X-3pN<7OY;B5L)5j@5M(PTf=td$t`tNY90BmZa3CNaB
zSoshuwL=CLX0S`#7t1=0%obW&dS-fkHF0?s--5Euj83}_?8%jaj>){D`Pyt@w%8xD
zZ5lhz$k)V}9c6GBX62FumiS_jh!!{c^)6|}xGIJ;>)y|0rd+7itY}1C6}zLyM{_%^
zgUJy#1V*-%zXV7J`*H29xftUL#NH;TAy!Z72=6}c20+zk=xu3i7)J``-P<u>YO3l+
zX@&d{;9`0`V7zXG6@-W8tx%?;9(UpJVuvh_nN0{LxD4fl6D&0$W74^=J6)0(d}rJf
zLC~n9SD!@plcaBv7wR!Bb)y2CB7}nFYuZsaLzc_ZN&))?gG?m&fYL&Qm{{gY_gj?k
zg>G)z(hr0KlivW}$L%1p`!+^~Z`FV-e1&j5gojwZ%1uNkxs&15t8d-ouZ)ovz9)->
zD7pSwsMWRY#E;^;MAEu*HH76$L=;g#H4*@m@~~FIDMRnh+GPZS`bI3jJ;<;5w>bgD
zk-vEk>ieY#<?U!~{s-*uk4^b|jx5e@0D~83LJ<|-e0Q{<R`1!`v9<gX=tK0U{nPWN
z_TG!=x%j_l43w09i()sl+Q4_6_J+H%yT!qlt|<ZL=4M=;&Wl=0w)e>R12-m_q7x*y
zi`w{_jkH(sNS@dTWABMQ;ymnF_YCaudA!T2fP;79si(<eMZ4UzP`Gj*>N7XW4R)|3
zg*iG$XLTnw8T|)G2S6N8(<!9F$_vHmq-+|zs;fWX)uTp(qkJG70~<3WluhfmHwExO
zkpp1IN&FhIKP;LV=oxGN(=d;mVx1k%?2Z6jvu+B;N;WS%Ns^cjbIfqG+hSpUw8lwS
z{JO6yoHRf>WD(8-sAfoB6`VPX1*F8OL_5pY$km5{(UO`Dn3z_-OM2QlNaZ{+C+MTu
z`dJiSSV-x4BNmPE1?Z1%G2ftQrRrO=H<Qaw{m$`&@S+<>;QO!FQ;hYjdameXo(l$&
zq!eVbK4S||tbWajmh@F5I-dmdB%^7Cn|VpdeN8_XX>6nrm@G6(ms5FoVqKrxfU~!J
z1G+@;#OvHGkK=y4Qh9WLxMD+o;)SfQ;4>255WWl$V)rva9+SAaZ+?FX8{DhQQ%V(|
zABut^;wNKa4?8u;mtAF^UeIQ8nn40*@ZdAvm<-)W)*DMk!CBsv2<^*<cLu;73}u=!
z_t^UIDbrDa5NeYayImI*$p}zW2RRRpgaAB6HW(a>8mL4WzOb;GM2lk4hOq~^U+R8X
zRffzWSJ@iCQ`)*$XR5$EZ-8K;uq-f;2ViA>3z%s%pk!}Z+4{n7zUA?oJo0xbcaa`Q
zWT%cdB8ihgb%H_zD4{^1r_)n<AX9@jmbDR8951C`v&BE5_*{Z)s;=EOSzTg_Yx;+<
zgCaWHbVD*ExtjG}KQ@-=#Jm~qZk7F|h_@@dsjad*`4pa8I{a<juwC?eVUPQ2pu$%$
zuM6Pm=ME&8SWwptL`setTVn%;z=N>;l~hGHQzgDl3r$}Rn~g^VPwo7(zToE?i`_*V
z2yR>!?a!yLNuU7OeqWW1w9$4+of~1iCY2n8H9nV2bzTbN5+1Qqb#<lD;L{8b5*1fS
zspZlT*Vg1AwS*rsiIicq3@SDwHdM?@Y>QZ#S1`U`$-M3%G%36wyqMoJf*Xj6Lx8;^
z7KBERt{I~!W8Z3daxtH1=40VwSs3SQ{!&^$JSt>TQd*i>vRhtKf7rS>#Z~0%=%veK
z1a8yUKN!<~aHx6QjwEm|lDsSoUG9fmZ+K&;+b70>PdXVddphxSC)mi*Y;kfT-P>8f
z;Q2->l<?!rf{H?IV)FLhT2rJwG&KhI+hGa}=eEWw?MAn;7>Z5jp0Q&?_Axr8nyrrF
zr>a<6!__b?5N;+p*G;h%CihI1=jDq0yLsHgw<|O77$yX!CjN0q@sNlyXe2>sMp^Xe
zN-b{KtE2a~KW}!mPll^&t6l473L4EyD@-u}kWA{fLGa1t^39D^?Uf!X%LZq%@$LDJ
z?EszEJ286grgsT+ZnJ%>>X<N5FaG>|B@q22+}~?y0Sls-ix3WxQXrT`W&sx(MfpNZ
z!7ul4+XKKv2bTv{T(#F|{r*2X_wJjHPL$EQ;DqLpHOpI3(Kf;~pYp7{osp0|!^V>9
z7xFW{4|u#?!RA`&FV&uYa!dZ~wZ!N6Nj*dk8kAyQkz%R2G1NY9eL3~=Cd~wGfa9Wp
zD9vXG3EII^-;}64P<~nm@dBY)xWDFGL=;!Ewi~2@SdKMGt5}FKvC*oK-9{G6h)Q?s
zQ&+CiV|vwI!b`cQXc_()2h>GVVsff!9og{Hu`&)kN6ThmSgjPghjwE!YHAI@irAp>
zh<|TwYB*_bD+`tDFid?+VUFUcy-p5UOY0~E3T&jR1IFCvj9x&?GD&~U^ZQk;Redc@
zo=FvJ6Sv#Du=oUuqZyQj6^6%;euf+Q&XD9+hKw41=p$Pp@<Gv3ET=|`Rk7rJETe1*
zI+~Q3W!H;x9E>gOpS9fly~X-R1=2uvhy;bJE!Wi)vrm;U%w8Sv-Drn<yk$~p$OgQ4
z=A3j&3W;C5SmZ+}4TCp5>4=%c>D8S-i;9aib(g?Q2od9oi#6U~UtC^FJ!N^~ZjVJh
zczy-D8<<pW-gLTJdiwNPtC898O230l5d80-+|(ETanW;a=t?qmul><uv)l~_QhC)Q
z+P3R=fj&-WA?{Zh&GPrHkvuxbP&>6xjheXB3oqkS8t;e&1XO#5GuB|)y~b<%AtQ}p
z(JlD7*;R&bV+T?;_BJZ|>yKIvlwYlWy*H_!Sq0@0vG+tG1Gi{6x_=Spci)9!QNHS5
z^NY%H5;LTqarX;?KUCoyr4Zt4In4Qps)X3Stg2KqE%Vf|$?|1A4lm4ih-Z>x9~RVz
zM%l-N^(3dLQ@gKQeUgjhV@1Iu-2YgZ33`jjFWU6;=eoZb$MxrLRqf4Mwmso*+XdVX
z2=99|xi5OeV&bAfnyOyJnY3(M?K)`q>G14?pT%G^PDZ@iB&4OK4S|Lqwg<I6<#Rq(
zl`pW&3cu=&R_=Z)v;A=&Y-QzuD4>^u&*~N9e!eNrb>09fscB!htgB&lXo$B^l^oTt
zwb0NcAIQ)#Qrp{3l5Dg%fkDN=A>PQps6MY~<6v9C+{@XERM}`b>BLn(OKy%Lq|OlE
z*wr>{Jzw9dzTr7j(Q4aGj>4yGZ`qaXD7Z*Km^@uw%j9~LwCdOd*v?x7j}xbSa<_eL
z%mzzL{Uv?Yl{aTdQNb*yaYjQPaJpBp?7BFwghM{GLc$fhRXjx!hvv#B199ch+8iJ)
z*^wj`hLd-P)n>)zISgZbK+1B(pkjb)FlUz1s9TkGyyh`sRLPIS!zB^HXIbzSa^ag(
z8|e_jfl4TR>OCALZIs>cO=AfXL_FZQ?H1{WfUe>PA^GXg4-pkOx{^;PCMK4mB-1QB
z5K%s-(R9qz(w^?$Gs#Jqg{NAd%H3~{ROf9;Bzb+fNG%_aC{JevGQYTg0|gS$-#s1M
zJ1hjsKUC~r_srFf=lO2-c^f~fZ7_vI`*;cLT{c5l1-e6K5;Pxq+V+kND;hb&g0Aat
z+S?>y3*-{pGIVaPwCkI@3mNYefK2ihiAU>_XR>?k%&V0sm7m}oWL-cYR}-9k7Z%4m
z9+AdiC&#a>eAOyJn9z@~*HVJ0_X1pq>MlT=7}96hG}H<M+DDehso(r+@jx~0c$(kQ
z4_z(yE45fRzIQ3DFX{In7yB``X_^73Jg-IPbbF^J>z{jjPGghPZ!~l=_SNpf-LLbw
z>eT`>?Aqe?Su`yV+fEC+1J;xYmp12|7Go+B@vPH+)=GAK(Xt;EvppY`C}!unYB|dv
z5JX0tOp=ldPYku@7RS(iJ4rix12SVlr+0~>kQ+g3*zs<znD=?NvvU}KXQgpCY47MC
zr>pB0muF0VivfAbRVf;AH%hBqk-iardj3Fo+ZB_9U%J{#)Oy!nf7&gfDdNxgzR`66
z5n>^rI}&ZD5Ubf|R5hNhsDUI~Qh?XmE4f9r?I-%v($iJj#Gyr<eX7S+c)Wq!oTj!f
zHg8b(TJq{@u5#2#(hL=Y_aIEYmRy`S_mNi@M6y@Rz|Zwjo~8Hz?yDJcXQPMJlAD<1
z*t}Et_$cPRuc$X2hd_hdOjSushtPa>pKT_$Ma~_IqXZ}`&u?S3y<v{Nkg(tbUZ<Dh
zpyae%eGp*3{^s#!+lwwf@CxPDchb1h9t|$JqU97ena8Qz-rl~~WlMK?dMM9dKRNzS
z6awM+9*Inl*l3-d8^~vt62#pCRBCCghx4G<vOOo5$Q&C%y7$r;r?YS)eVg(EC)CGD
zZ#)?q(I`p~hU7D~OZ?z-{R9_TB6y{zyc#;`+Fm|B)TCNP_ij^Re0&^&I>dnJn^KZ-
zZ%O;uJqhk)Alc@JPNxIs<3g(QMq*(314=*V<Een8v||MIH%ZA+&83Nh8Tups+bC`a
z>p0{VZgNTqcIO6UJ7K_#C^Ye{l@^svuoQTrDRflCtuuv%Ql!AL;OuGg6-D3zI}=xn
zxG1|COS<dBQ2C>>4uv(pbEf7~NJL~TT1CqTB6@|Q+vj+#G=5lyNOsvrLU)2r%xphy
z(L;6$qDMnaqbYN6P^x<5sc@_&&D(c4XR027p)6%q<$Vl#_t}nta7s@?+rG*$^0mOn
zC8PQU5=(@(mlz{Y7uI?2cUj}NHs-7^@RD?BYChh=bDbt)H7&v*zke@QfadUZd1gtQ
zB3Z1vq-FZiz0h?Uhn&(c4(R9JhoP!f9-`|!E<9UlvlbPOWbvruIS2C+Y;kU`d4r*8
zXu>Z~?Pg(!UI@HT^~XnG6;E#Gm+X-pR<w%Exc*Xa&JctmaJP`%5B?G+b!KMpGX8vI
z11HM8clodlW%u+>GeQ-wYmR+C#*3f8*$(G$ASZ_;uc`C;E+r0^p}p-x+s`hYE9J(P
zCVR%^rr41{1e>mWn_7=p-mBiimtfr&SKb3vF;cVDC#ll)bGfaacH>p3uU~`09V}1J
zUlK}7Xx3{IKwCLOy!yihphMP*;`OJX6h)@&7hLF(>rePXpH{kivvkEi05ve(%onJ`
z+KD7Wp0^$@6>rDDqcsWXAiVIIOf9YB6kduUM*}4C#mv-v2Q?^3@=2#jW6H~Dam@MX
zW#|S};Zd&si!z-?<wcj-nTy<b*V&<F!#>o$;o%s7l_-4JG~k5K-+Z-l@%L&CT;yi{
zTCjWmTCkt85P8>VlRqvCDK*;Uau@6BD$*|;9H3^#(L#<F_QkmbO1-lPv#+b|FLhd^
zPk+@ZEUi+9@71jgUIlJ?mr{{;J}^nWDfP}iE^l+ARCX)}$E|e8!XmD2YPwNbhA;`&
zMy<``%F@)lC^jyQH`!&dDkdeB(qn<IgdPa|R--p>w97ubcr?~<ut`fh)`eW4;=HB<
zn!ccB##m70WeMM<I|j(`turcfK%6ZekHXF>Yf!gpr=w=3QO~aR^v%rPQ495$d|QAa
zhP(>+<Zuny{*LmNnA+oMNq#cG!ObpqHBDQI1PzH>?bg+8UEpedXR;Sa;(f&IB-J~2
zs<JyO>UWo^Kuhz|u1a8&BBW$zo2PpVNs^%0%stAHhOBhH`eOH;W}S#SEt`rmsEUY-
z<?ShjTujWY9T`27B222Lic2#bFgpZ)@ZRj=O{fL@xH@RI?k}fXD)mg^Do`sRbctJq
zMM0fnkwmUAldiwnr)l`G6f3XTZmro0yv^fsSFiJ{Mjz{Mw~z@08K#f^SZ_Mwigz<h
z*)5_FXawRe3tmYL_ebNLMJ5d9kX+0a$gA{kF+7fMJ)|Qmt*)(YL`x80-e28pWhP?j
zINL~m@3~RMC+30#?Ph3BsSbq7H`L!%lqWMFGe=b_KAr;dj>(boDDW%!k1zcp%#-Ao
z7V_8+b8~VG_t!1BmXBdj6OnV-SJu!B@}TImtlWpJ@%>pmZL(yN5=w8cB&;o>WS3N=
z%<Z<mvT46w5Wx%-V!(92kt+VIsZEidPCrU`VPf{NzA1&hSpoCjKH}YhNz~9@NxrRH
zdWoBbdl));{=j3Pd{i9kw9`1%hw#|fl6I#4%vFkQt{a%3IgJX@iuE{7Ab*p8L8L-~
zR)Lm2O5S|UCx!UnZJtYDdDq928fWyzUFq{6ca~bgF!j6k?AA=r<y^3OookNYlS~`=
zY>k71fKIBfe8$@3`;d7l88f#qh3rVII#fh_$>}Q)m7HojQhbz;o9P<T+UmFIslJG^
zT5v2w$uim3V~4~#J9AX+RNBLh6Bx0uQ?i6DbvkAxnsFC6sPko}URRS9A>WPrXR!!*
zk6CVeVl<cDtxZ^r%>x@++N?v(2Hs_lp@i2mZ9YEWjZcqv6Zu}6r%^5StF7-tR@k0r
zVrA;5W*#GbVlY%9Fw~hdXaz}}MoT&3Zo@f6$rB`4;r=`qUpV93d1QHh*HfF>#%?E~
z`Ews(p(wi{v#x1vH@BvZcpl7V^*ka5A2veALr5jo@eTeimNJ(I0p`a~z7XKv48+~F
zeeO43H;dcgEw0<HW<;u&kmOap4?HXrL>iJjPQjBMXWTc22X%iQuRxu44?Niptj0z8
z2#0<cj&U)5lF?t)#^V=<$r;B(nm4~reO^ksfThIi0q!VhM8{d@!^5((l%(w2&X*9-
zBZl?XoWyxqa>$g353?>Y0+ZxK*V1NdH#?LF+8eVdN*@IpF1e4x2=DMoEyEE`M_J_i
zY}%V;l#%G4-V3eKbWm^m4OHM(VxJ9Oos6{))J-ljaS&m4q>z7U=rW!)S}Ko^AK~5k
z;J=<c3ruq(9p_Ph+rd$ZH!d!XS~lJBBHZlA?>+k$A$K#7@heBm<TZ$#?^8#m-0z!#
zwav(+@TUjgeO)+PwT*1|C$|d0!mYE=v2{w|{k^wbQ$G}`Z>JjyZ(Hb@wSrj4fT3ma
z^xV;V3ATD+WPR&cNbA<Wfcq9r>3%Z1qC$7k>w-reL%-grvWRQFp_uuqNa`j1RoJfo
zrz>dpqlZT{uFEiW@M()SREUHv!kex)3bw)!HaYNRi*JoFdVMm7nAR)xWb<vcU#U7v
zih%OI+Con;k1{Fh1TDF3%zg4$kIOuwbY{;zA2Q}yzrbo)DA+yAiM-#%O?2bWvsO8X
zn$EoP3g2q=d~6nhJn@SH&W-EdV%AC3j5U069_SyZ)Q$ThJ?#VR9Uc4pv2>^o4oEqs
z-V|Y1X+6i<?;mVj@LZ9wf+%LzR&sQw>?r*hYJc>nEPXUo2&<AxP%2k}WDIjP88^L(
zD|d5Vae)T)v<D@gz@q|p%FDYwao%_sDEy+XsKHK>U|H#y$47osbBnm!YUbjZO1b_v
zbyDp%%xV-A=AEyN{lXM{?SkpdRz<Xl4o4MH)wSk^9Tyxo7)r-^&IAye42y0F)kHfc
zmI?jgc<aqhiwT=4+P3Qq^l_ayEbg!%+_IW`aB{{C`4rk<Q=AZ(-Gz7BWpYc29(Um@
zFm9Co=o655-LXjg?_djtq9*3?gTv*c$|Dd>i%Q=4j3tab(eS0G^W)7rrY7MVlak=w
zP7B#Vo;bcIhcpjt3b9qcTAWj=vPw^LJS&TdilN3+^&~6RaqZY5g`DF^LshB?!H~<O
zU%TWJto|OZd=a2FF*`<$pp2nZYnr=DSUJcn?dXplm(1swHkjlUw;q;CR23{G{my#>
zS}VBhb~?1bb`dnq@NSj*vkbq^ZOxLEPaFSQkpTpKa|}y$+(mMNI}GUh^$6cdmJ74v
z5?4${X2s`_rk`_pjuOch+84UBj~A+7&4q7;`+>Ga8xcH@d52CzwbOldWg=MPB5+iZ
zZURr+g-T7uJqL{_0$?`iQ94gs)2BmQJS`2+)A0w6T~86h)jX3a^(MHI+nd7AH|~L-
z-R*q8=Eq}-J`Q0`^-hEN^nMeY5|WUMTZKuEi9=;j9F2bIfYcbZDg=`NKQocZ$S1GP
z_SGLHCOQhdUgu*xo2OZM^~X<Q$h5Y%u6KD=ZQ73Rx-_~TO4Q9IVcK4h`wc>dz5TKY
zvv%L5`toR{h9ngZn|^9E1Npzzmf@8-xo@S!OD9|bHi^q*9u?ToaY*_#Ac*s=R6{M)
zG(ox^T4p95%YM+SqByQvXa$?F;4NDi`K95HO}m)tXS0qMl-7e;4&N1EK}CdG?yp_v
zS#enCMJ_Ptl2Me1E|1)fIg|OD6&OG0ujRbzB4H7RJ<PYcydZ#?*dWSDv%hRZB_yP?
zZa*y}uYd2|kGOetN7$eXdwRA}#<3J5LFh)CMB<F_2e|r$V8J72MKRm*y!`7ZKvskj
zH53-b3{K#y3@-vc^ND*eqB(xAp3RyzvDR80X$mhOvYQ&=rT3hbChgk-nds?6hD6H8
zt3JIVFl(u2*rWsH6Zvr0j?38C1PE#2aVn~D2(G;Du_A@ZAmIKalH-l8Zl;#!O}vuw
z2EiRA`juhw-@CVL3P$eFxWcl*8-WqPChkWSf<qs$jYBPn(np<u?iD9Cf?yqkF4Gs+
z^G^G?QCo6(Byds`=f!akW-CF}&)4>o#rTjM5b-IM8`3vk>zi7^SPa#F7a@ItO609a
z83q5Q3NIp|w-@?kW7Tqo#xxr|6q7zC+3hHQHHU~F7VSTb%M7_7aW{R(N^=H)=v_Xa
z)p5aQsq(@;iNlUllf04;_42cK!u_t4bhnmCU2hK^^*FYgnqS|{Ba`Vji3<Z!-h8Q%
z{ACH>Wg}j7{mAR7JzBwND*E~CMfq_8<ksgI7@=A|OwexZPqaA%EHs5O(cSfQ2@w&9
zjrY|wdF5?ZdHk?@vi}02$dJV$sPg<24b(ov0x#<;RbCPV8+bJ&Y2J@xI|$Z&<aa11
zyqizA`k)81+BmOzm0cQ|IJIXu=PeSKOwZ8~EN#o}WTMeS$W=g=cuThlMO11kjRd|8
z@I9RR1{m6zKX)WZuo&~>!9)ojLm(4UlJg1a>*oUasyo;EOdoXt?CN^U{UBBc7XB+$
z;ob~Ry&Q{$CNY&V+}RP-SDfmQ|4P9`-V4C9hRv4)WB9!4Gr^ca7{ZB5;(!6cg9ru&
z4f_7@Xxr}e^L|Nw{R}H(tKbT`2FhCK*SBSCf#?YviObY+uhb;}snN7ve8(S^Kw+>q
zv{J@&B=?UI5d{DC27q3VPwVZR0PuC^uMf%v!WS1LQwlw;6N{Ke$%RHC?JDF=R*)~{
zCn{}hUOHj1M3E}+phnH{8EO3I7mdH-Z<!XXHEL{e45Z0%7=xs<L<9gk!;3-r(SIYo
zz{OW8;$KH+vOn{M>I3PYh3D@T=${wIB+vId!u|~o4~qYXZ}4~U|G6MGLcah5?#7FK
zhLZm~cYE^*OYUGrL#Kv|6L(Vx>iowc`t$$(^{H-*U`>~}jp6MQ&>%ThB!6^Ds%rRu
zyt)7MWoB4lo;Pp@e{JBHU_xg91!?`kwl3d;A47q)Bg_2#X&Ju)fM4iO0Q|51>Yx8l
z%dc%>x{q*3UtKOR-_gJPhjIVsJNR`CJd>EmC*ba$y#Ej0{k4b`nV94u_pta{@b4k{
z=Y`0dPaT_q3H7l~z?Y4o0HP@?Df}N0&7Xn&*ARUT{dKMd0rTF12Z#?LCSaoer(yZW
zZ~WsFC!n71migax_a`y%umNKf>@Ts;Ga-mfa9W47diGsD=)d_$pRj<yCar&~0KlJ)
z&nZ=({r@iVh=99ui;$l0PWtEFsSw*!9sfm~>q&jW1nJGv7Ep^3ccW9F{D-d|1`Zzg
zMO=vV6y;yT`g@>>27&N{`SPkt;Wh9Vde4dE|6c+7UkB3vw<p^Fx#tFg2Zdxz05-!n
z@=tkaMP%9Qut6~=H<z56TKYltuKDumx?(F6=#JSzc;9^Kbk9LTqQCrf;xr=QY<93z
z({4_7d?fCLM5wf)iUtJQoZo*V$1jULmlqGvfhq920`>opD1f`Ec3-ZL=6%(7{*gnq
zkNg%eWT4RV|7XnKqhzTPTq6C4*#}Jn?iL^>on(CuHA~tlAqjH|7O=|<3lb4ax%z#U
z|L3~)uWRp7iT5MOO5(cBE5Xm{M)g0;xql?2f26BFnIMdwnA}fA1YJqoiqd#U>OW__
zzkl<-;dvQ4#Dso)%zYyXlOghz29yUO+6DhtR;`pkem@h;my>)G_0K%}e7(#GB$5cy
z*VW)aq7s!Q?k4*`yJ+hPezLmm%tBH7=MD*s1H1-UR$u?O7d&T_UmJDE|L(>ExI5(b
z1#ow`)PLg}|8Kn;8}-*zA#+-l{b$gn5q-r5$tmDxFau$-4(uoX=kGx;j*gjGIK%VS
zG&npw-g>G$zPp~DVCrW!fws1`&~$b~gW6%^9sSPXPW+OtYccD@G?X?b0~r{fHzXWp
zFW1(+7NDZ1<XF0w==`rj8X63U7UECbqGC$GDv<I#L4SR!bcTmxami?bZ0bf2-v7nh
zTSmpXWox4e?ry;$xJz&g?jGDdxCOUhA-KB-2*E8l6b?ZG!J%*q?i4QHOM34<ea`8=
z_Z#=e{ez4F1+~_ibInKQd>x6;z>*SDLlp31iHS<WvdjMa_FmiB9>JA8WR4&m%L6WP
zL*2`Sd$Zj-(bOJf+0Mkbb_!98Ppc0J$?4R8Y0U|7|Ft^gD2Tt(bqKhT$|U|S4~+wX
z-9xVKW?}?35mgLWu5WBZaF`p-R!{jh-<>z4ns>j3iW|0OWf9-(8$EqFx;yDXhi5Jv
z$q~|XXj=2O7<W1COO<hTJ&zMXcI0hA%xYUT=ex>sEiAPu1`rtT#O$m<3t5`7^xuE+
z@4WVp01{UKG6@E>$I0_m-Tho;y-Tym6!c2y?d_K{6E8?gO-1)Q1c(HOb_iZNw|z$2
z-I34PFL89#8`qOh1-Wm<#DWBGcU}+SE+<e2B;qn^Mm>i8#!mQ?vOTd^0|PFR{(Ay?
z^}O=pL}x+IajNj=e7L?|4T)2G_eScd{mM||sG@P1!FJYPeuxSK3rb7hxf$?iDmNsj
z3~^~0VJG-htNgbY{AZ3ru!71L)RUl~&eB^BSo*IDDsw|@2-r4S+UyTsre>#y`%?9p
z{QMpY6`&|6oc>`AFsc{=J+=^S-Z8I)Z3;6b{rej2QUZ{6r5wuC|54f82YYXhOH&Ck
zE9Q8v2W&<2tRy8W35xz6%VqWQ@|u{Rr_bcJPtY>*<I3ae1L$q~RDGA4@AdVAH`_*e
z4*XUfd@k$Zv~set;Z(h}G6sIxz8yC{It_N|b}OKz3R8M|`m^heSTBtkJK%xm5M+E7
zwHl2W;afVs+ns_UCxv_onUJKElpz4rF_mfht-0lfW77<gcsvf8SGicUUh|~N*c()=
zaeY2Nn=6O~ME%<J+iSB}JqDumjf4dAJrik}wePzL3E7O}&NUO0^YqMa^=ftYWl0qe
zQVcxyIJ;M@zi@tLQEyx=7}d)6mvFSU@)y;Vsx<`E>vBI`sqgMX$kbmuc1xst_xcXL
z5>WcGu~Asmpd{C!PrcH9RjI)LzVaa8Hhp4#hJlr(W&H4!a}}pUv%$l{UaC<7gbd{P
zWA#FWCFOb(Xa4NQ>TK6l-p?gMSuwdJjF={L(n`j~vOz*hTFys3@txx`)4s5Fe$F6W
z@(5TOP3Q^=k~L=Nk-W~k@x!dpJlBtk{<XMwON+&9%Fi>%*=(V+l}4w#*g{NhT`tZT
zSej!XiSBEA`_e?4ZR)#nBic-n8*MNl(S3?pqoN=(@BoMe+FW0?<|}W=O4ZQhm>~8E
zUs~Y5KLj$J#J~am%^PaQ0$hQ5p<)pMDj9KSXTw!8&Z_ZPB2g0)MPoy?7OXc+3fO9K
zYqTOa>GzwoA}Q)he0|b7Pb#Jn!hH0Fi$2FsiGWhbpd~G}TP2_J<BoBW2a=7a_Ejkx
z=)hblb$FWXc+^-9c!Ew8nn`AE&r^wMIl&57mc8^@I_EJg>%}d=$xNdQ%k~M}RLnT2
zi)yLm&^ev!nHZn2OS8JglpZ`t#FvqPiF&=HzSWPaW~52O8N<)NDK^yPca<1CDya9x
z)C^@m+!?f2gl{o9K<4#O2~c@!)2+vGJ)?J_g<D%*Tl+@`aXe&gE5+o&$bhD_b%N|q
zU(&ohYYTgMdGSNOktwP2(!I(D@T-6bT~F8hINbJd=ouMP8LQeR)B?;b|9zm6zg+o)
z8<+pWji=~x*lOT~1z6FtS>N;Mw$t1FiC@3S&E)%sURmW(=%nkE#t9rXsXsp#w+9L$
zhE&e#e^X{8=js|+e(zb7bWEo{+sU!D8oWI?Ty?)7FVl4#AFm2+vGQRfn#1+Dor)kb
zsY0(|XwG?Kv;As0aZJmPWn5L2!{-v@GQuuhcBZGRyTyZVKxMGBB)m2k9e>tCmO~@*
zusdIR&sf={HZ(k}c&V+`fL+-p#eswj`{2<Nb}Zl5a$Hzfuk*!pgPMy+tA8YoUXt@N
zA%LZASw$usc>2qRe?YT29&XuLL%>7wvlHU`^sJniZc9`;C^c%qF@QSm0Nf(EY<QN}
zk}3?a!<5nwzp)_7hd&k1ei98QJ7HxeF(q!3FiCP=;eAGDcRr=SRU;1}b{V}XtvrlY
ztbGi|RWs6Hd1&`bFjOI*M9ZyWl2>`cjTNFyCYosm1X-Ctr=*1J*Za!>S~l{HvidF&
z3WBTTgP8|dnG7=l#Y1dz;nuq@-N{;v$T)>($Dm|;qso+-7TuS}Q*FNK)lV-I09Lro
z^E`TD9=xX+g>AoHg`FMnc=MdKzA<&&Z&ejG&ubIcO<|^||A$z4`^*n1{lyRc+2*+Y
zvMNx5z;W9obBeUPd&;8gm{3Ae5|x1aL7d3_8pEIcUG%YEXiz+A-X5FSt4YsFOQVB@
zg*|(?dNku@(*Le<nZw6PM*)i)l36fosCT3}OZId<;L`u9xGbP`zp`~JlC{mi*BXBg
zSFQ4Imjs|BG;inZQ>$Iu;%W<I`Tm4341*h&<G37~M8e$d0X6vh)U0fPQLdIVXY51Z
z$Zepl`a^+LPQsy)53(`!-2o%uV96Oe?@CmbnFn(@Y!L^{CnxCWQwkXv)ZQ=q6gFF?
zE*$^ua(U4Ertj0&J1|_r!-S>nGL>=|=aC}#ZLdICm7U$?z>!tsi>jYwC{xFkL*oi?
zo)>-EA*+6+Ow6Kre78Iu$&2BM?*v3)c5Q6WO}#g(042eibE+fYa{YWLG*vaj_atBK
zb_?GSj~>u~JN#XHF~f!(1Mc*aw;C24Q@M;<8;d(^Z~hl?koFp5NM-c5O{f0z)z3x@
z#2^a1o7K^C<tU9r|0uc}np9TV0%mkatQ5XGsN1@jR91&IaD{NyR)is;@BQrs_&!yp
z(jEsD89U6<Ix-XPDIg~2=2oxoy<241wPd1m-WZnhtp#|r$oW|}GV5)KnU+lR-BGLg
zVv}>N?JUd9p<{qmZDUKViTG1mq`|E3_2$M<La$uVRu5n|(n!Hx(5y;`O?_P2-lDx<
zBOIioOFA_2qbGZKC<*kf25UQIGyWd@luR~+&+1h-v5g3rY7RfYwipc}EA_s=8(k=D
z{@Q<{6Jd449G3kg4<TCg;L<oI{!xst=G$lTipMH|@48q-YqpcW82nn>N7p|v>%(;Y
zo`?38@5M->W`=}L`J8>?Tb``?F{VIj{0XyIAr?o_jb)zvwZOkXn}0>59}$7FK;bRY
zBopw78|7JsdP7lUsTHkXd-tYw8}~2PSMv#_(EuUmDE&fgDB$rtLl82x`n7HFtIM#Z
zw2aqXMI#`AlX0~jaw>~3JD)nggzj9xRU`G<G?7(S#&3GndfBqiL?JWuBT3y+zpxS!
z=i}11)kQ??Ex@-^<&Q+wOMG2GHeA?zUfA3q_2@TX58c<bG%YIf7=9bj)DkqzSO8N4
z8zlD)@NY$s_sjcPgfcymdB|h?`C)v1ypXI+dxH@}PHrvBT3CL6H>q|zZBr#Z$iR?7
zrvC^JImVDBPuZEgcpaPIr|jN$?`1m}TstiO9&xHu>AP&65AfVo(WAC=v}9CDie53A
z|CILcfFn0&#j8&ZusyN><^Ftg2F=la4m3?lNKD)er_E6j=`z4uwDh6@4w#m4Zh&x8
zynSuSVlAM*rm#TJEUR!Ug^nX|8O;Y1I&R7{3@ElrigVtvTg~{NS1h&&?6g^<VPd)*
zP<NzeonlW>yIBaB-(_$CWliK%YHC>%)#JnzV3&PRa?`;`S6m{=J{!x_$HSMYQ`XN5
z`tiCQ@TN_uY1wK|VRl?6N6ca|P;5y@9>>jvz|_U&pH=u<q(>Bdp>B^m`yG5GtGLTu
zb#`$yCG2<kWIkLvwsbafx68-jKk7dOnLVF1Vn(2Iq~;r^2|P*#IQW2@;b4Y^vL}Cy
z9RJ_xZ&Ob^{?jU91FMww5$CJUa|uiE9GKSK-O^+gN${keooue=<{^SWbnZ3cGA^f5
z%j+N2TmYNa2?d=(;GRGNEC_9G!c*I2#@Zo#D9*<AFBU*(5PZA_YQ6*}B5u2+P-J{v
zsra;bim=B6{XtTm<wNiz$ltoWGQqw~YxwqR1-JM9M4>K5MAeeiHLJ)+jKK4#rA)FH
z@O^c1jwZ;SqiF^Nn2e)-Vi!MIXv@6^R0of@G9fs<EPMbo(1`P5+%20mOg>)r^I64e
zj|<Xfh2U-(_BA8Z%Ub}ew0Ji11d&u@b8A5&k0oHxR6nsdB~JXY^C&54wz=$ctRcee
zW1Y_QFd$N{nRxS`$O4>~S7|EG#du?0oN4VLUEwu6E)vC71Y>2=nalA{n?%L0wb36a
zV86Wc_s>a6P999wYSEX{QN+MFqg9S)2QKO?L~LrI4i!pDcu81o&9?K#4o@BF9iBrY
z-8LC8QD*tF?weGjrjyXnSsz%XStOK_IILInBFzibD0Y7vHt&g>IsscN5AbF(UW>_9
z+hgsl6)WemG%lV=C)@8%6xG_z4+pZW9sf_R;%^8>1MrdTFX<KPp4m?}9JTx4+C8xC
zjqUJO0CdykmpIDW_iL2t%~ysKzvk%hA8*AH+ouB>ntCO+q-=lJh{SD{GbP`YJ+)Cj
z*V_xBlV}2BzYp9GA-8@3DL^?me6`}yMbGi@xnRmowDnR>A&JxBA(6ohGR!)7i@w5B
zJ^<Lzso77RKb;znwjC;khsUO#9nTEiDGOTO@Hquyi$xgB8a@Yjn8@nKhpoLb6Adbt
zrw8;OL&=QL2Hh+FX?GL=xzQ9#EzJ_p<!eozH^+i&?jAc<grchB^pjS;H}I+m^gN4B
z?awe4n5WAyYe>Ya?IR}_A^6y2Td;&UYdFt0U%Ocqps^)%f6{X{DsW^fnRj#*i9DLR
zq0B4VH`TNnWF(`fr<W(M@I;!L#YAi5+TKe|Q$l`%09U2_w>$b_vee=!A)3t3xW$h7
z^`duIJHA?V*Y5%|v#oNGW~knP_InNBVeX<HdSE|OZ0s%R*1i<x+yI(f;l`<oN@=4r
zpTSHvRsIaU%_2Q~h8)dAY_XxJo9EhEzw7r3P`ka;L*+hxn+fB6^f&il5uB%{+amOf
z`ab(VR7IUI0D_<)54FYvJ`sKTD-Y^x8MxKktsJ%Acwf$24eu7W2BjLg9=g@qmwDV{
z7l$UT7~T5t9sCkk)ce58Tv%9CG*l2PT$*zOJjOH$sDm782byd1k38WD7w}eSY(Y2u
z6Ocd-vbzO&4YpfwEy-EG-_u5l5CgxilHP|hU<Nw4S>3Sp%ArS_;qfTmP=${6ag8w9
zqGt$wD>BvaVkjpe@ilU#TOTqz5_jw$<1yf-P-^f$ZT3BjeL80_I{@7cQiuphKeWcj
z1wC6wVlPf7iZ$#A$_5cBs0*s!Dm9(yy6qnt1yGX-ofj6_05BoY!4r^5e=Ih;78MkT
zj|x0N?H=zVe?SbN{a=#QKX5&pC?NkJ%9vyTKx^<*e%v})lA+AaZCMrIB{+@)fx6V6
zfT18RV?rMCYA&@ao@7eM^Z=Gi$nrUAE<NK}{*Y(~ZW=re23AZ$O6rHYvJkVgbIsYB
zG}&M<y5RfgH7hDA8hjlqSPoV6v_bY1mujeMd3*c4%@-)*IZz^E#Maf-x03c9u(W-X
z{BJ_b?>ns8hH@am*h$E2RK55&_o3UU6@1lQB5R$^6S`W(<$4#pA&iv?QS6cvlhf40
z$L)F<wl<6YA_t4eg`P9Qdn<rtB4c%OX`JE-bnd?U-kOvSnx!?S;~pIV;{}gDWn=;g
zKJ1}JCi$=~ZY*^?XaE{7iq6+9{2#s0=n2zRO?B>KGF_;#w~)RB5x3HoD=SuERshk?
zRg+TRjXc}9VaLjqpG!lf@BiIMX@S`LPvYtC9Q0RiG6oElICm&vWiY^*98?D^TmUBH
z7vQI65muTh*Lhvlewp#+&6}aQgc_IMI##7WbqfCrk^!)_mZNU7t3!}{e0*F@U7c}E
z@H`61xT=18w3K0Y?}+qU<fIr966ol5-+UBwCkMC#1g{od&j7ohyq|c%4~Hu?r`fHs
z>8ZbYHITtw#iGl81&vSnkGXl*&A8lLMmK>5`FW2Oxu2Kz6Z4By<Iac)8v2xzF}!M6
zlM@<Vj!%cIY7Y;VRkajx#CNU3RCV99AK-?Q70CjqqBS&1M+ei>6L_2xl16e~e{Mh9
zt^3DyG!*#$#p#QZf)d%8wi^Atp<yZ(dY!MtE}tW2Czj?Z*qvG8H~J)5_Hq0&Tz1xK
zp8x~e)0CPPqphJ4%b{X4uw@md)c{W+|L#LnZ!r}S5`7bz*bj>BtSX6;<s66ki&MF0
zPZ7VKZgM987yvkDDR&eSKA~&G%DUnosYe2EYY7|^Ti1-0nk>_CLm&gj=L$FltQ#H?
zlQ7i=^-9D_B+#}V-w~t=`*oI<7Tl{v%RomLcX!;O{BYk$jXvnZxXih<k_Pw={z;_&
z>ld9OK)QL|81{V004(2q@I*)wy72Ye3AIV{+<6vc=&|n~cI$w<N?sV*Rad4fT8aNu
zg75hL9Oc;SC)eyygZ%=WKXPashe<a~Gh{}Oiko|;8V&#HH&#hS8Dwx1AOqv&S=-;~
zT<iM|_Nf_ZIhM}ah(&aaI4xXkbr@Z~>Li)%&e-^so6Tb^xmG3^6(5`9ujcN1Uay7c
zXNNIHX5uzJGew7u&ikV{ZXFAqB^7XpR@Jr&@NcybkS09%d$yS4;@;t?IhhO!-wiF+
zjcPh{-xu9RAXx|%C@Ue-({VDhDx--!X+bqwg4B$ZtM2zxG7HPvmlz|l?M*+d)Q$4X
z`(G4r7S`3tv$UOvS@SeCmW&SkmyG>C?q@@w00Us_zp@jW<(@{Q-a96ITS>4k0N#>U
za;WT-vzYuEeFi*seF$8?a?4DIBZlPQdD3;hxd|{YK$dZ-D=NyzP;;sR+}V7PDRqnf
zsNIB|913>!acDOLszUN(V2nHE;Gpc;g}U-^+0#vIi6L^Mp;zKpfSI#S(BRN)21O{!
zs5mR>?(ds#Uu7n1NCl(gYvu6JX9*khANqVWk0oM6q%Xn3d{t7;M_9^GXfr7_U-R46
zf_a^BOEHIO5<QbuXQQ9N!?dE_?d)OuDH%dmuJUr1mh{bRaa<6G>eX{gQ^Zvc_32>A
z#5kBKnTY3y$8j6;Pp;O@A4d9gp`4c2xN0g&`YyE!eAg=UY5W>$E-uvE6sq2>EdbKc
zv}sJ0>uYf=GJW8$@?=v298K_GtNTZInN!2Mq_+S=MdjBnG@-9ouvlRKXkPSvhUePe
zw$*-J=lTJCkrnpl=4>X@aj+m($g(t-eKJRhTucIPiT7yt&Gr5N*p=sRS7ZOk<05~N
zyU&mnr&rt9C8bl@-@75}vS-r#?()!OT0=`f&yyg@{a3mbJdRp-UPeI+5~t-zoXF#;
z<yO!UK-MLdmJX>twE6P>`zoBk2EcTmGMZeko|Sv*P#hpDwnjkDw<&w}5N@#s+(c5*
z&<y21ehBl0OlzSn()vaK;>x{!&0hr@0}Il60o#8vH}C?6mWVpRu-Z4k%UY|;JeNsj
zAxsN{$Vw@$V|>!wj*o@_x&J9{$%+>V@Hi4r!5{Ov4dLkcE|~7_?8FVTwu<fT?SE<%
zYx@tNG)n9bRb2q68_zKhtk(ep#V{T&E^~2C%iEKc#$|7LLc;rjF$5h+F%`DZB#gK9
zDw|mkzkl>1Vq5IbFxT2l<KtDDYySJ?A%9u-3{t3Z{coPP0#G>t=<AMai5vv%4|(1x
z)_?f5K+I>E^Ro?0<IunUXCrve6_M!M4Qd~pm&ZS3eWo_f0geBE<ul`Ve|&72@&C8}
zk$;ubco;%(NUgc&!T-;0n^CO1xOJwne5PnX6ap8B$*|v)|Igo84SRu3SL5@qswWIk
zr^sRa>fXe9uIT{1@Bh`+^Z!lpeWn=xZ;T}W(1-v3`e~^bfu9lLzc^1)QBtb$gY}A9
zYTxYW7DPQhg7T`+!;oFi<*9_(bxv{?YV6x(0s?s7MY@p(vQc4BlvwK|BqcD6kJkW6
zj8A1?h-5|H!^4VT#_Qr8y%49e+GHIH+55W`H%bndWdm4AY-$>AC9rO0k>z%b>ZFX4
zzmYU5jxYjZ$j7%$(e>kOWN4zCum<K55+NC>obewMVb<5Ns2xXmHO$TY>fg-F!Wgt=
zv*%{yFz==}?=1^3Hp*K0=>3AMh89G={dP%xhBP3n@B8J$hwrj(eGSzg`;E^{#!*By
zo;TDESj`}CHZ!%5JO{^lVL`iMDI>hmJ4v%*8S{7F-)T9d#$4jrhU$)xkYF>0_9enq
zc1yg(hNHp42#-rZoW%WH83SmGb2hK!)LcrbiY<WlgW&IPru9I=SuDS3@V-sPRO~l7
zAs6?X7<5OxRWhuhqADVo2^W*w$&3I#yjti8emE|B;ID?kn!`p28PFzDJYRF#;hNi#
zR~$<ME&G=&X^^66Ahn>?)Bt5p2F2;AEEX=wYm6|UX{4yNRa{puqp+eV-Zmz07tX`P
zS1e%@FG>TShighsd`W2-mXH1SB!??Cvyk4B58Y)blL!?n0XI}WI5jVha*rfRL<Su2
z@ZdMc--Xb`{;W7lJd2e}^R?#@)a>+<g!>*PA~NjkHNXr}SrT_vMg!IdzkH#-s&GEt
zpRN1?g{8ic_3$P~s`7c}-6#;d>}>nmx<znF$j%M{JyUvlWmQ=T@r{^V7jL+{%)2ph
z3X3UsFX1#rDCbke1T9{}<cZ}OCPhd+pl5Qb*mx-OR5aUhpnc%`{rK=u3Hw6LTdJYZ
za`RiqBqcRQd;Z<h=#|@!&7eSh=MP%8MKk;FKf`+RrK9N6%O%2QQlv^MNSnpvq})MN
z#pXjtLIUF7Uq8I+$?wpv5cd`PvWeHkq?SQwQ<y&rrcf{L_w|+<?YysODSgKe)~ygm
zI*(O@@}hWN7(lQcq!0%dLJ<gi(Zf;AH}0iCD>8kmtW$PKQc?no4qqt8yC5mUE^Xj#
zC4iPmEntJ+8Mdf;`@r*n%7t)UV`-pjoDqkZm5r2REj$F36e)W`X(dK~>=4913hy4g
zM^Ck>A*I65gqh%QfeDBklH>WNo~+O#|8|82xS<-EFFYF0{ZMm@jTL3c?-5b1w2AD2
zQ}O<EoV?I{3lFgzFOE$s2;#4$6?OrE$bV4y?n0Xt7sq5@p{Q0@$3$Fx%<k4nf6>$g
zvm5(4E~ipy_>NL3`sUlW_u(<oiJu?+cMlE-P918m($L<(pc9t*AwK8&tb^xP>4`t4
zJSJt+4H!_zTAA-a11fY1ZF+NPLRJ>bPenI1BXtW<C%KDhaGU4iJAw*XzeA<LA3^A&
z2qg;|6wE(SMU#HW%NWOfR8N%Qpni!(zz4(OibTyvZQ$+WQ`GjM=SRskT_rtWaa(Ni
zeFNBgt7$uLs?QbQOoZYbu2TK@;MOo)S$b%*yqRvgv7VeZF^AzBg|@w0*ZUTR_ct`=
z`th3FM=`)hOMCa6GT#7#GYmE5ApNU^boD$MWsR-i8d{7h7-xBmceq^|;^|n4YSaQ%
zGD}dJbk_pDqunIHtB+$(aV4SXG&n+KNz{78b>>LFz)PMZ|ETdA#Hi4t45ZnU#+SB0
zX;Qt7wdy-}(?B)|=u;&9N#vw9N&GT8str2cK0AorH-;R()Gj0FWIoIDq`+l|qjmrx
zzbXU%rNTeX4a~tI=A^>S57sv2$)9w!UT`+eh~;@H&LPzMR2(d_C!}T3EE#00>0Gv-
zbs-vMy&tHx-w$J}HVC0ADnts&o|}W2?Nz%iDlU@`=Vw+#omf~n#AgYiPm><#?Qhjr
zg+NMQ>{c+N3d7PLimPfFP++R6CS>GL$79A%%#YL5w=%pV;ix;KCKXM*yGtfp{sd;g
z?dVU2gVqi*lG9D0qAn`kk81>37Wek|$w_Gjc2>MtEMAj{SN-LtP1_}hxc*AwO#w_k
za+J4EOOfv#O(T;roR+hU0AmBgtU51lPF6;~LQs)WVQZQAjhDTn?dhPJQH*e6y7Bq@
z)_Y1VJD9qXwhxJV3tvF#^RRno2Ed@8_9YJklOsXPD^J8^*wh)8oG~*lDLDm&EOJuJ
zkzR_VZv0xtIP+8>RhPU?APCEs)Ba9z?cpm{TN;cvQrl6bwbZB>U3W`CbWTuJ33^r5
zuc@U~{O0&5lkbK)Aug?4pV*k#o91D?u+}beZ<gSSpyTv3d`~FCgI3{GFJs>?CLm41
zY^1y11o#VtvjlpFhd(D4RkmlNR*pO+1L){JNq+g}!K%N#_7It-^G(Dw;(KF#=_z<}
z3AfZ?P1AwxSOq68)9A~ha++d3z@T`%Tt=+k?7btWW>zQB{u~pIxf?7W;tuS0e9UFT
zM~yyD(bd_)UYtfpq~)?5brS#OeBCu((Tt=F7H8)j;0d!N;vQP5F5}}9@3vjbL%%oz
z@|F`N-Ub(o|CKgod`_jq?t4c~*EE$D$8%y^gcp%Dq-7w8>cr$^1-P&|f5xhH-H(VU
zmnz)z$)s(dSH|4D@Zu68>&TN+APIR@5Y6P<dOpwuvUO`^m~Qti`tgGfqA2!c?NF|f
z0=hNdzKd}A-ou(!utpTptYqWeaO=^AKR&=Opj(vho=UHk<-XkW5?=KqiaWUXsNK&1
zJ@+~-wIFuYASlEU3N+3f4VjqUS}C$(Gc!b{iaI9w9sNqj3yTE17&Hv2Pu{E6u<?!|
z4A?Ppf<`HGez@%n=UBY8Hcz_crlFyh|L}py9cSu+pI^7As79%o^fLMTDOg*p;&*FT
zlol;`_DH)<pxE71_+YSW)Mu{5D-nC8-*~&HZ{ur;N6!2^b%V;^GizxV@UJq7_6tGW
zp6EE9tlDH%O65k`bkIIz!TVaioTrCx<iNyZ@9m8}+H9^S8zVtzmO(!S_<nQNB2!Uw
z^&kgN8=v>^8D<lBD2;}3t!eC=sj3q2%-ghw^O*p2#>K@4Ss&ZFymr4L;~6<{%go5j
zx{Go_S;b*BQd=Ntd;hG;2B*p>C~4HBo33%QP>Qv*2uusV=ug$5fZd;-`mTWA6WA=)
z_@SY(q&_h5j_}EKAljxP#=R$?zQgt@_R0IPOOPz!<QTkiY9v^7`J}7j^$;+AiNbx{
zi{PFNU{xx4BzlmjHTH`{x~1AfaI<*(T(*)~zhd&YAKzZGOLv&47}PIKS;jB%n{aXp
zRB4AJfUs_HmAJ);PZiPxw>Fazpos6}4((Z1YJ;IZGZtUwcUp4q<tuCRnyEt}oB7$4
zuTWpt-A$8IV}C8{AWRM%WDj9z>x5WaS(JP@iU}`xXkCidd7#1?ix?yqPLL{b*J}RG
zO~lG_z$BcVCL#7itfLnmg;LwT{z|IlQelqn=Z_(+Cq0b@u2mXoHoxg`MC98SO1m5B
zzC|hn5<|?y#S(A<h(rmk4DuRAXa*4~DUKK@QCQqv*y<BWb7oR9qP``cKhq745}EzL
zCtCJQ_4-<v^k#76a*4+8!A~vNBQI9bqjf{0gZC|&^6-c%QGO2Mv2c^mFVpv-Uknb~
z#8q>hu+^$xbl(wo=c|f!svx`rVd3FwU%a)uNIrS|5boQm;f8YfGSDE2ASznf2+^#(
z<;&E`$G3F1Qq^KehB~|HUN(vST6>54$r(c=6Z5aM^N_wi?F|reyEr#>(RGo~bEico
zJl^<Czl_`g?{V}5!-Y~OPG5&qqX({Nb^A~=T1DOJ2>i+$vwaS)i%dZbs#6xO41u+W
z`a0jTT*>l>D<?Mll8%0d%(?asy;cbcIZeZOFv>~l3CpD$2K5|!RXF)vXP_rGCVP`=
zd7nZj1&UwyJ?6O#GoR`wO*}N}O}r!&1lb%xwpfSqm0i~YS)wWon$sL)%QpV;2KRxC
zCBAy>bCc*`>|>MmIIM#<`frW}(q`av^jI1SNr4|b`+6qsC@><1@RD(>vsFq-Yi)6+
zR;Yp*AxUebRhD<v9lp=OKmqH+i=pj`JRtWj>9hqIUPCZ3yIw7?IsefRc*99_Yqk%2
zy`T_0sm{-SuIlYrTC4eRJ|@$+3n-a%^n8k|L-Q-kQU;32{Jg&<zQnD=^sUhc1NHV1
zj+$>U5)TC|s|*67Ovd}5!P~SN_309xZxe1@$uxSxO?KBGad8w?idEyt@z%bx;3ZXG
zcESbnlMv*ta=vHB^X0XO4f)ysyWhK>uPB>)0Fj02btJ4JU=M+~2^5tTC{9ec!YOGL
zh?tq-XD;r}d)d4uEbd`+w_ASb3{I3`^4zy!DD0>*;;21Oo_>j?L}pxoqsfrQYnE^K
zea+0nOtWK0+>gE4!M;+9NMvp!QdvP2k3j~bkX&JRpv(su&Ea|*FA6tn;*DAS`aE(F
zm+cH50h<Tw0?3)%KhEmr#*795j34=ND3BYmlJ&M@sP8qikbJdCp!p0X>|wMlCbzV4
z{M3{RZMK0ulsrZhCsPBv^Fa7m=^1nWw>AcBKN3Hv%(vNx%o9tK6?+)CxMlrXa=M>N
zU-y%8@AzJ!`e;mX`>G(_(kqo4E10ycn{)M}_9MY?0sUnC_THb$6ph>AB3}yP56OOn
z;Bl7=h2RJ}I&^tj;D!i)Qm^CfW5$fPdCg4CrLKrlQ-frSnI&_n;6701|F-Uy>TfTA
z^GE*GB<kg6`-<oYp=Ik;c9mF?OlJ6&Hw$X+%lT4Q1T3?LU>?nl@Z*;>CqhC(L*5k3
zoIb|mo3`RO$L055!uJZXLOw9jofvfb5c?s~j-MP#1Q9^1_H*VhO?Y(B{%RcWaAH6{
ze%}^6?HP8xJWY{o;j~|(Zj2$KXjQTKZUsS8rmGQU<0T@V&`yD;^s2#2v=Rcw>KBKH
z$wMwtxZ4Ja@6Aw|C>*RcMVqYj_1kJUC-7W6!w9oeW|swau&-%?{EQrP`8QuI4_c?U
zGV|X~`^=r=kbtodBN`Z18dS+iUr{T*lgKTbI~uMqI9btppozU59q29f|J>12iuZuM
zVmo_REg4=dH5h(XJ%1Hh<?FuCd-MQb6&u&qT?4(*QZ{KZIJdNvUrt?<A0L<Y1?{VG
zFq-SsYQHbglU45%?d7NUHm{FBUdE;UsXpX}5ftW>{w;bS;xIFb3oikfW<@+Aixf%A
zP$ok)%Tprz)0Ta<TYC}>?#4x=JX$M?U3&(G(=gjkuY#$}RmW<3>dP`KhIxE-14pgG
z6=YGVEevvnZn?bK+c>^^b0Xt@SQL$okZ7?m1IxADz+;%L-NCQ=>0B%#(bo8|y6{yb
z&e*A~)kqmcw!^j(BbF<&!ALzkSOX-)_1+C(;FNI0XKC=J`V!RH%EQf~3Y6VRDlfXA
z&SiZUj16MN`=rDovg@&a51tUN59{if72vNv3^1bk8&PsG>rH4VY>FyVFD)(EL_^CH
z`0m#{_VNEnjr)=tX}XNgSX+67=4Skg5{1dnm_5jRI$}smBtH&o1OX}5_L*dYf4<(D
znMn0ng_#nZhOmjCZ?&oBZ~AOZ7F8P6!Fuo_NkZ*&7YCg4NWW{>9-U&>TsG=O?FlI?
zw(&|}U*o;k{uat%{WT^K4{xv}eedF9u_O{V(W!}e5CZftiC;H7aw2+`!@zKRtzBJa
zCb6Q-_&~|ar=WG&QUWIO8Yiisvaq(Gn!RIz)k}r?fmJFj7BfN{gG?Lmt9Q4s>bg>u
zg%;*zis)mCuB}9MImO3suDHEPZ-2hI5U>-3m#gfvnur6m0kWgW&EYpjBF){Nr&gbc
zc3%B_jA;-jDiyAtUhI}$!7pWOy!r0*wgb6{8bN8>410Xm2PTMm5tZ_$PMc$Rh>ncg
ziY;D)rZ<~pueaebGf<KuMB^t`XH;C!AZboXOX-yDIW{t;+n4^HQmPK8w%-1Y>|d+`
z!;M}#99_C_nl3U<^=f(sM-W~hl0i>joL(~`+AVKww(gnCLV~N@>+aCF7(AMQres89
zMLWv<3JmrVk8mng=p!bf{aigZC3}ety#l2RB5f-R%WY(k?2gE_)Skh?s?fpS1_pKO
zw9oS7VY0va$a`^S9KY8D6p{gRP(f<#U0q0a9A+b?@dTy9r6PV-7umA`tiIEtBtbj7
z^L{ctt+S=ubUELa6`7e)p*(VLVK52DcD+S3zbV>}6&lg)Z$m#Wr)zdp-S(ALUE%+x
zPpcwNPIfri3s7ngJM=|%s#xN#@d-He9tmbAySxGoI67l|yOnJApQdd!Gh_`)vNvF@
zrV$IuH2SGnglCKNP=|07O-VU*r&t#QG`26Q>*22+J8^}a7AnuZ?|O-6S8Xh)anu<|
ze3jy4+QvWKaO+YY!&SUJ)#Xsi4<Vgd5lQn(TylI3?tYFbJhg}cWvz|5QQd`QPdC$t
zJ&9MN$n<*;^$UGvk0cl)giQeBq0y^?On>X$O5)2!Ul_9CiTb<!Xwz5Q$7uB-tR}2J
zd%O!_gkGrq1#gXJcMI0Oe1Mr@2agtLX`v#OR=n&E>)IXP)BzK-WQIpzxB+E^PlbDi
z#mS;fho1<MK5ajSKBpWlZ0#d16S2@0UMCN_?O~H=*IH8P*>SM7e<Kh;b5gu^uz&;<
z4)|{PR5Fg`v%G`H?5WxotKW*1&VlB)gH{$vptdLQ1ac}FoYql^Ok^}HR?6mJ^JP3`
z3(#b|ySj(kzo$_&!Lf@+DaDwnqhKRZWubgpv1GOuizp8L%<XLRZG~E~b;;8NyRy0$
zw*#(t%79iAo;LpWr`3ayho*x%h%)}o<mj@Yc8Hym&pRMYZY6VvOO5Mhq44qEn)SCR
z^;$ocL7z6XTu-IqA9Q0{yvNRoY0#{f1x!;!@M(4->r~%og@S=`CfFVbDN;wfQ1x9Z
z4~B0X8o4sa@cG>LzIgQg-VZ@iS}-kmH3-Cn?;{Mu*zF$e7HDXI%L{g6sy}_Wf*kzQ
z;75aiAq1i#>`I~Qb0o97h;GVBn)yYEw)I%0;5H;U?n~e3>DBVdJ;<axR#A+~D0Ax?
z^@wfFA7fA*VcZd?3k&lIQ794Y#G3x47v65@=z=X1U2+9MN+$3#Jp0(}m4pwAA_2$7
z9bejG*-9FOi-&<Jr%-9qt3|NZrK3yIAvGh|i-x%R<Lcp<LEi6~;Z%z#tLO9WP0=@d
zhg|7uJRKea+G;*Mp7NPPS};C`vq>}_o;E7hsWo}j&M#-LAknQA_ziJrzIaf&gIgni
z0dm%zSoGg9u;f*R8=BQN4lXJk@j(k0{9J%Aj5FKVIAz;o{jI-jx}lMOCM!oBUVzMf
zPpG`3qwLU^?BLXVzqb+7f?tS?v_uSvn=x{rb>@y`cxttL{4iRJcIF?I-})w@ClwO?
z8oyx$s4jlx&o%J2d2XSLG<JToR~NCex)$!YWu<mQj&)~rdDoh8L5nV~8iqVCW1|f2
z*8VRlG{fHq!V2%H3U{{S=4{X05Z(pQjfc+xZGR(|Gbh8qj4Ja?;Do{>(God#kK3OM
zvCF!g50nA`c9+w&NN21lSLO9Dp~e08^l3lc_qi~}Gx7HN;214Lao+kp^rbpAGiH7b
zVHi`yQ<OOs>?^gW+gJW>Jc<116%Xf^i>^co!6Rj}?H#l#6djjLS|Hb6dmm^D8MFCQ
z$enMeD4w$)?1$9Q29!lAKR~)zh*dxN*Bl`={66Ba><oMF6UjkD5@GJv9Ry6TDn|B=
zr(8?oR1yf&BK*J^tWKBqapO$@U7Ag%DQ|PF8Ii=x*n;x)qsQ*3I8!&wQv8UAo7P!n
zoM`UXhcQ}pK{3mUNgar`)qX=q0#iGn9_B)0@pp*oaEJ~|PI5?yczB-ZHuL}!+x#Oi
zM?<xxtN4y)bC&x^NYb-?<y>@F@;KssrrDlewS9Z$XS@iIiH1y<$E4et(8Mq1uaDsu
zkv(T>+B=@D-N+3A>n$&}ME&)7ZCL7qT+esB;VFOc^tX>+%|B#r)~-Wvv}#Uzp{AA|
zil1m3Z~dl`=|SaOkV*=h(&562z9rmb^<xtOd%|m7S9`1`;wQcf=a&$8tMVjUj}cqS
z&)b5UwzOlxQW^5<?YITl7H>cI`k*4(LY1RNCHN<f?z~mn87NDhc{O6|R?a6>jAki(
zu<4^65KwkbLko(2k666j^IAj3*@I}7MPtk9)S7wzFI?W^1u&~zCo+I~=hmjXzgD>j
zAdP238~M@W-gd9<phdl4A6Jz_xw>ZVZQ)bHZN`e%%=jKC#66B^E6I`5aJwnQkQgtL
z?ZaV_j*y3&n@lnpCIMc(ry*6{*hUp9T_Fr-Q1=EQK-(?^ipd$PRN~Wf(;0hl>}+T+
z@S@G|zdp%^ckLu=!h12Xvsu}IfkeadY2N#7@&Fu4f)0=DZ7qb>8w!?sR{y5c=H(2s
z?nDTVa-u<Rcwm)6dw^UTeK!t~HJf&uKNjDPrKqEQuqVTuwkYsJ_kMRN$BQcSOKH=}
zBM-3ZGfHf)npxRx1b|cL_o|DDxPE`>p3kQ#?U&;x{!?e}1*-}rZ55VVRnc)O3#dp`
za7=@xwWFhkGd@5dpjKdv%naBFv-R!HUtd!Jl&yv+_>}Za;W21fr@dIXqv3}5J!Vz2
zAREvq$ZFdji%M3|vS&cCv(k~{1+iSLCJce(3D1*MSsG04D#P8J$&%Rt!-dhfS{iXF
z`d&-<RL6d)oLshlG!@@r77^O8v?fZ`WLWcJwNK=XYel>A-*E^e{B!)q$>6HHp(7cX
zDzg;8+UFLCz*wSU*av6ng`QA~ejLzvr8YVo>)YUC&ZB%seC3sC2Zwgw-cjfEyHb3$
zsGxbDnaQ2reBi9ng69CZTM^t^M&Y=R#JROVat@M=qr_DnoNO=L?-UOrH)eSx2k@=G
z|K1p-vIxF|y>$ZECiPB_nUBlxXLV!3z@GlhE}t9Qxe4t`&+VOFHV@yay@Vd$pX?Ul
zj{=*4ez&)AIZ%eZuSGlLgs-+|;~^J$<W&54Q_q((4({ANEnV{XE7P@A)wcmBul_1@
zntgjyS2nYZkmTCMwb`hpD@i^8xR?200X8wN9IbdguC(fnaiBehouPo8U6GIwy1ibr
z=CC{kHG$2VGb%tpAHJoaW}~w|+ql7+#4&u`(HkgVp8}o2dKxY(Y*bj8nEp8zs>Uj~
z6E}HM78#6NK1|p?LGpvA6)&>S4a@(bv)~$N5BZjN;@!x68-AQ85ZonOxU3QX*i$7U
zS-Fz61hU9p09Xs)T9u{ZPN`Ahkb9*0B6U;2A`G#BsdJh#ioNDODLCTW9UNg%33{Of
z({rfy4XVKBteW3syfUoA+m0nTOgMj;;iIcrc7CoV?EvxOP#_UaNMJ&IYyR8Wv>O9#
zIw6B(Ps6gCPdu*+wf&4Nf8&&XT#=DFh5p{MqYU>-@dc3Ht`v0~^QFqdmKF0R@f4NX
z#G}DW5pkwpSJHlRo;YyMElv)7bkY`F-Z}@@1z4{@*IkJVkNxzsPvnnNHuls1R9ai%
z;Os3n&Qp#=f7|q0)AJg<*r@i&%COZ8b*l@Cz!lAR_9xX?FdCQ(nnhIzs(If`p$q>F
zctxT19|>%6wi!YO%_vj6++&$DW<**H?o|@H*kIG1#daUK|A;U#&Z2=EOgmYRiqA@6
z;St5!UZzVs^h+=0u#_S3lG{U`M#!p(x?8_nX7Ph*8`xUiXjYfRR%xsxI2~Q!hV;^d
zmmb3r$X{TgtSt1{cNDNp2mLZNB7cJ^z?WYcDE%c4t8#<umrwtW2kmwb_zvpS1acY<
zJv}^FpWqLu6i}L(@~aDxj@9YrZbqi7tI{5+t9eybg(P!q{Mseah@6ngvulA>+XiAJ
zdP~_*BjM*%1+)_L(6!s*B@)2bQq19UA7(4?zWf~eCgdgO;O<X>!<-Dpn0y?y>Qh50
z3Q2pKn@IHhV|-2g+*{jV9p1gF+18f?PVsn{Yy4gSuFY|98@aF&@G4i*zPRdc5^sBn
zP=fFwX0`v;91@b3y#28(wT*ov8G@)}*TciK%<R-vfIj>(Rl52Tu|016sr{?{1u(EC
zD#-253_B`L&8Sg5gJ>~Swlz~yi3P3v0=7}OgtL{KQOXA-9(^PPh&GIs>4Lc;-xR0M
zn1_UlQh*&kK%nP!j!E3^Nj`|-YV$F{CBP-AST}mZH6`n#SMb5ZtvUPzP-=s22ottN
z;Sc9AiJ)eBD(7U0;7tv!E5R57s4EP0gL_xgIuw1&K63}0gj=q)iU(6YOK$5|6R3=;
zk%LER9?#ROqKvsM7f|n;7D%e_PBTXx+{!~U<4muF?5u<<?MJ{j+@DN^1rtJX?6Zu~
z3>)mTx1dd)Q^fN;X<%r;1y&uQov{9ZZQk~q+kYV7>b)Ck0-#*eMTEzw4h!UqTMqr2
zsaKV4@fBT5cPX`+wGbu7%P(u&7(BDbg`-xEg`9g2h-M8yZ?B`u8ciV;)mJ$s5aX|;
zbaP0s*;j?lS9r6Aq^DL=8Ilo}I%eirdlz4nZKa>Dg}VivrhH=R+e5UW@@AvS%(}4k
zi@J_STet4@DyC{#1=-Z^-<fbf5DwaXG_5mJ7%skX3X{~7oVN^&6euWUJ`kfWL~-DC
z`(_16E^oT5WhjK%0+)SNTXogR9kIi^A*$+1D}(2!wyQUSv8Z&Q{50!?O#juu$LpHT
zJ(h(1ee?TJTQ&!OL9>;RFOJ%{YR1eTwH;(MD?sHZ9Er0d+UMIWUCb;Kqa;B;jS7;|
zvss_<HLa0Y#4>C9Z*KYH8Tn&}!I3(h2M|9+AuoiK``j0F#yd~rzQ?X#*?wWRjELDA
z&JpB_J4IJNy_M1t{n%fYeR<(gF<tZfB8xxmraJL>&G<-!Eo_7l4+ib>`<_Q>MNl?+
ztT}~b2OM`q5B}$hhts`3^OnA#^+&%7OG8O>yuSRF*LRJw3AR;klyNp?*{^zn6)=Q`
zSQo5hXQ9J07V@@hPV|G8aVy;8;P1Yjq<r+K_!EjJ7<DAGhLue{n#?{Ug0UTffL4b^
z4B7{@I&)=;KB9N#=2H$vi(ks$p7krdqwO)V4nlRNVp_)B#IZcPT~!CzXJH;8MecAd
zZu>;!z??Eh4+1#feo0<0#74Pq;x`Qt2in=-_Qps4cKJ|ans_>S*DX_?xi*ieq<oJ<
z*vLRYj0J9Z;K2w%#oMvzbT!V-r7NCr>h9j-#r4U*gMN~k89h)i@Jq2di`KTRRV4E@
z(Jxe(B3W#;+<y{q;~14Qnz_;m2rR5fD)CsF0?3hH_}NMsRg(=fmZDjx1)DAgJc@lN
zI|uLaKfhC=^}zQ51^pb5GYMy0wWniJyCAai5B^zfB`al_QyMmYH+bZU%JR|FiT3M5
zeZi8)H`RHdT7@C86MijC&P#sdl8N(Pp@{^e(_?bLR0dh$NWa?a$*;$+CiJ-ZNs@3G
zzS8&{Q_gd4Yoi_Wa~O69(A4sMH6yUxh$~WX%Ij{%O?X2~)VLe2`1cGQJD(!CW|@DT
zSsnRm`@A{iZZ9Nm9Y8>T!n#E<s6sxkbJudHUf_2&hG-VuYZ@9&Hm$cVlyhwzH~l9F
z^wdySEtcE0|4h^kuM2lVuZ=sRE&CIuZvgwC`4-3Q<U@A452^<#(8agnd9I;+w=w<a
z{7I0|1j__%4#URz3!)XpeIYqO`WhYd>RjqDI!)WDK4jBz>5BmokkZqh3jum<3RsA{
zO9hN7p(%dCZ?K0`_A(Whz|Uq#H$?M_nxpn?+W+;l)tsww-{}thYEL{FXt%<(-adYI
z3#dPLbx?F-LkW3-e7~nrOR<<MeBsGfyd(JF-Y__NW-nYNuQ;4=!89vH$$eSVb!;9)
z{=(!1y>?luon0wyaVdu{Y|P9Zfx{UB6p<1hJ-9!k%D7ool|4ujTA~TK0KO{*8Di6p
z3*Es($#*MatNAG=lQ-2(!?2^1*;9Rf@eN%mjCaj((#GH-^Q7xm{i=l)CdzqIqZN)*
zmFFBK*k4O;ifWOLy4YOCQYr9u@IWd!-Dm&2aHkj1Pl2LQu;f*cB6pz*0{v}YvB<0w
zxV3Am+bTF>pg{SVAa?#(qfz5mkgr`aT~Jk}HPpZ-Vp0zfOMB43$;nd!ar=BnmEJr>
ziK;XzhqJBC`Lf?36$o-RvgoQ_m>8A$jv~48YZlo*o<ag%4ooA`{{jZeH}li$d0e$O
z9{!I?>ia9mp8d5I`_mLy1%*1VY-!=yKRNNw*}PCsvj69qQ8ID@PD~}TerU?-szOqR
zmd3%gaNds{FV@z-63&S}`1XaVvBtb*kK;2~`J?^8Ev39<`OZH>jIS1yg)RUg1<nxf
zsr{SjO1$kJ%iE0fyKY51{bP3zk8eE+;+P(ol<C8L%1IeU>#b39C7KECf90V{+Ws!o
zEEh+6>p1Wp5-W<gmL)JL@qJYAe6f=$DhkH!O?bL^_%oKiuxGAHJU0<|reLXO=W_R4
zj;B=^?U|z!Iu@Xa!%}>86p2pjT3GXQ#cy>Bc0x%D2<W{}`zwcK8T|lT60YuAaw_15
zy98`&WJwJ#oS(=yHjDQ|bu~a5E&L0L6Ad{(4=)TJSOSTVjAR|vNPh%?AQ?@SzhJ2D
zmsU1ku=*qL<rMpDU?;*_M#xANuOz`@K)Lq!l?g&SbGvKtW`*E9tbcE@Y2`mGRR$vp
z-!A}T>v)?#Vx!1t8d!@G)2UI4hE!miflD|M;?mM6lT#hk*o8lFG{cftmEj?{24X4A
zxw*OB=#!hA!ad#Epd6H{I2&atVK`E(o7QMruY!?}Y>Z7_o_kXnndU5cQ%H%Lu=jQ%
zP|xIv>~_4GGw!!`WcF8}1JsFNOT{3c9&qn@HnKBbq9w%8+B%64eU|$AFPYCvs@qsh
zE#hspXb=J-`^2vK6^YHeTqg{*lgGt+*VY9oa_)pxZ)75Kkg3jl+KLl>?GZCo$D~zG
zJ}x1k4QZ-%sE2@oSU?hTnv=r<Q4FIF5w<QqolfI%BEpXV1<pxd<CTImTh+rG0uvrT
zu=OK=<a5TpB)_D1xxr0PBd8)YVwZ9wuvY0&WpiQQ=L4Ua?Ccy4t#Rtc2L0+JD`Q&U
zPqC=bYOCzhp}Y(a=0#&dxxW!G{Wi-RUC&EttWBvgT;1)|qn0ErK!UOzgzrAk_NV;#
zNP(ujhJl+9??XI-i`1{6Vp?6*!vzFdcEdmJ$nSu3=Q~$&RZT(xtAFTEr21*{+~lVs
z>hJPo0dCkWJ_=!W*$T<}&E0(RcMq6bdpx4-w99}a9!=Mqrd)_E?CKsgM^vTsnH&)K
zyun@K;(mLpaLcT^YPP}b*f7Urh0h7)+A8fXsSH~A;cKfzwb}AH9G%sOcIvM}oo={N
z3Er65w{F~j3L_GVzTJ8?%~1kbIa&36`*KcpTHtY$@K@y*>;PKweu)oX>XfGt@E)PV
zEE@+q@+Rgp&^_Y#dgg2=w%xD27I}Y%W#ig~4!%;q=BPRo;4P_neP@A*3xj|91;Js8
z0sKY+e8x1Ns|GDk(osd#)@qUeNYD+9x6En+;h3e-(>puw2N^SXs{@f&F4M?OppjTz
zKl91TmSCmE75YjM+ONOQ?A;4pnJMLVsqTvTiRPR31{0PYt=hUsc@1Evm_M+^5bl*t
zTXN#<%9iW`@Q%tLJA2RTlbVdPPPYQMoqg^#s+$Ib46A+*#hUoRaY0oxrl_GEUQhAY
z4D#cg$n`?Myrh&I<~h1w3G?CRc`m^>er#{eJG^s7V`c)kkN0;~6OsIe02G#p0OJux
z@L7G>uiZ}$qq~<0klSkd+>#e~ymW+N)TPI4FKC;cCcc-Lq=IE);$*^=6_S#^GtHJX
zEgRMkmM@}q)hi+-2<LH@uV?Jpz9WkOOk3YCh-by<#i%$aIjHdHJqw#cBnBCsbofNG
z?Cs{LNe=BxNyzO?dSC;tJg@*S%QP#R3Oo}|X6D{>2;r5YjDp%RtF{sS%Mmk^UGvE~
zuQ0-CGe8iCrtK#RLCZReijGd5iqh}mfgeoufw>T&Za9e13lT@3)9(|&_oL*eQPLsH
z%Uykn=3ayxGhWzVutxEGW*r>dh1N_O3d;8_OGPeGxj<#=#dIPWgKsp{p6Vq08gLa!
zaBN<1YzsK@yrn5b*=!t2lSfD0xg5TH`2u^6-)5vz$+|Y_U{v4A9*HW^q^M%%edQ@l
zOczxLIG?r%UqWV=18n@?ffP^t&yrUyp~osAT$^QcXl2=*DZ;lL6@LQFDS&pl#~sXt
zV~_R0Zz?de23kxP@Wn-hhn^x96z<Fs$SNqPZZue0HJe~)3;cCUbfO;%Md3!-AK*7|
z&&t}`*6nRO2}MGJ-P2%?q)+@W07-QJ$?BND6<apN?gf@6c}Uyw>G9#pg7O{oCF>E9
z$P)80stmzPuxmk*np)Ce=@3?43~%w+#Ty+R9Ui|$wwgQk&8}~*Ke^+Be)kEXFKCA)
zV$pM7N-b=!f!C3yQ?pyU#P@SDJW^w}3tNngp~C1ukT$Pxw9K(~9cv{-gC^+R!r((U
zC*Gn({hTqP&*iIInDKTXFuADR)EaqR^;}+m<T&Z&_`*Iih;el6XGc~jr_=TpWLqd4
zXQ>*A9~;Bo@8@T9t5))Gf$b%?JviP`gSATt=vt6iWhME}g>c(8tA6RX7cXLb_x@F+
zD&-@o#>(cSyU;*9zjQyUDuD)TAg9nJPW^Hh9gZuS8JsP{g#x*j6Z#dU+DVo_UY`zO
zOQTY&WY6PAj%MOnRThmK5DAV0-%uI`B{h=nq=x2aoSsJv%04nWp_vL&cZud7CaS_B
z52l|#w3#wsvV;|`w$XN_;CdfeHS$Qe;YEqSHV|sg4o&oy8Pc|m9&Z+X7!bvhrOKZ=
z!Nz!lfV}m%Y)z^5%l#p5^AT@zRJ2V)8yQ3!|9?n(%cwe+EekZby9al-1ef4JgKRXo
zgy6x0yKLNp1b4RtcXtVH!QI{MedOG8``qr=J>D4a5BtyFl+9Xe)m&><`F<51Yq`MO
z8eS}`aPFU#^LdNip=-GIF{?L8g4pw@pr+@OqF}{AkMsE*2`iGkHj#lNE*1VvLOz_k
zzO?LX3$2o_ORN&1x`M(JUE4U{&sL*R_O>OixNvg_)@XEgnx$F|S2m{#Md)pzctw{9
z*2B&<P#fJ-a|ZfSomA(J96oEjkr?g?S{0u~cF`zUn5BONx>lQwiz7Bk+0F6NFf>y!
zF_FQ$YmmEMf^|vnwr*A;D+NB2F|%=@vy3VTSEQu7lQceLxoa9hO}7;jYBcurHl<6_
zg|Ki<7iwy4o%o5w#BUd&bTjoG^bEuwbGO|<h8S#C+k4oG!XIzZx;it(NoTo-{%D=2
z0JW*n>VRX*Bx~-eq}_x+m6sjflo1qt5rX9EPpK90oh=aapd7KRxPUv|rxmd%2*3TF
z!|In7{$^&J478R8wkql58u=i74M5ba!pHaTS^$IEeBZ|7)ep$)ab#ni^I6&{=<@wN
z7J;uB2{W&ug{UnaZq5CmE0w^0Y5cxjeIG4&K0U%W35o3dS)GE7!y`{FcJFWigvY*0
zn|y8uk;_;QXG$Ly+{520Sf;&f3$<gJKCv6j`Qt-(+qRxOCF&ww$jjDESrRY8poaS2
zMC1&<q|v_lHY5^>YR1l6Ghgr_(n%h1E`Rn6AVtUEcLb}FiHDG1i00ft<(6bDRvvw;
z7EP%@4_xf%MjwHw%pk953B;t8vl;6)M5=G*%fD|lHDK)6^LiAFH7HhdPG9*I-Os0X
z>gO3L>|1qIE{G#<eQ9Ev2`jrS@l_H!6g4IDw>RjplJg5E*d#w(s@>Quvt~#}w}yxW
zwP4{;@%$(Rr;x#V*bR%`ywg8G;A9HN)<7IFty=_s)DTrzq)f<KG-t;H#$qpfwDrEH
zoSVcR0nspQI$@elkExOtjLAv?)7p8KqgbX$uJ=iSUi(SkXWVV?hpO!U)4)WkkO-C_
zY4<~cT7vFYqL^j7Ch2O|(U%eW<YZbR4swH#t{kgQd^i{HN(eqYwYK2}*|g&UEL1SB
z|A=iUO2DhgL3f|hLxV%OAE@!6k0R-OXdfCR)9aX;ejiG`(YqXk=*#N;8oDRT=x#Io
zdE5CB64#RcQ`9JhnU`?oYZH`9*W<3>xGiQKos^1vpR;&x$0_CfVK@A%RyL7kHQYTF
zU&)QZG=GzdE5nj3YxG{8zHHRs@ZwO%8i#E`E1sNu&xB6Jnxil7_Ut*lTsEtaN=dA7
zk}N_7wzcf)Y<i*Hh<Y29fvXFvsqn`USL=dqbvfiU*gto`^Pu%yp$ar!60;L+-vqW2
zz2z9ZWh_IX``{MX)7zc9FlN<tM3x+y`(i<gZZoqON1mhFz0tV?EW$&PL{fYree2$j
zF#8ltzYT)JJl-%}d`aQ8{)5v9Qf$hpJH&Jy8{A>Vz){515hzY;>w=l{^!=JG;5O%r
zqE3ghT#=lUddhYeo_;>d{oiq)tmW1P!YVa8=L%IEdy0B)lF{jB;p5e7ZVq5rIRyMX
zo|<8nqJ!RfH${z*B(J))E7x{zpax+Jt@k>gPr)J>W=C-6u)o74EIv35y>#L^P`*EH
zKh-|IPFxA_83KTV51LK4rRmZU31`Qmpx#ZH=_O}tzeTz6-cb`<j07Iinc&jP9$`%0
zUd<5?nfU7FME#8swVKbMS$~BHhdkdMyV>6r$^nvRX8U7#>}KO|=OaqGqYCJbi~{CT
zpNK+%Xhi;G7H$=6y(_d@Hko(TDOha)kAL0C!Q(F#%*~azN>2_#b^{)_7R$&<BqM`Q
zq<l3P-ki>RFnO;Pn#8RvLnA31NuR=1pVCx)YGI>?y5d0@gA5_jmWyM^@Ns`AJ-*T)
z;#-^?oR76Ju{@k$p(;8$G&F>WFz^vWK?Zai-ZN!8Kq_gm1)GAp;J7WV#oryB!uTs(
zeU&zPq8hXVqju}B*Aw$!&D1%dFMd<vnZQ|IK4VC*z+Jhi24UYP$^h>pLoN^SJy>0p
zxmN>SgopI?CS!Yq@6$26yWeTnq!S--Be!}M6wSdmWB}}eB1jHsJPh62_~35(`k=;~
z(I9CoG*^)Uo{UOyE6H8#XJLh+RD;XKCED_dWFf~9G2^#EACHOvsmjR0$s2ZzD90P4
zt77+Wf46m_gdWv6F1g#ku}m&5V~0$u>W5Ar*oPEt2p@-y1&me?8odL_mg`h+qe7TS
z0e)?k`?wsYp5VQA?3&fV1k_sIJ1)RmT%mONzt>YQV)gLwcs~ry3ztVsQBO2`Oim6u
zFk;~Jl!N|6{FL~}phx}J5CQjr4ZTrzq2&7J)(Jj3hNx3|>c!rj0Rx4k?gl2g4bP`L
z7e`ZDlkirMX*H>hmk?gqr?%x%N8d>618sj(z;6d-q!4Y?1E+9w`YZNtuZ3tJf1r8|
zd44O${Ckn!Awt@z-_Z#sJ@ETWNk^>am;=5W-335}PmCcj(daTc!qM-D1jfYnBaXhX
z%-V~5s=*MJN-}p0KI?w7c@8r$JbVsE9;aTukoWmKkT0@y?n%mTdT_JtRZfru05bSQ
z?B~xfbDDhUdTa9s7(mzcX<p`3&S9dF0~8eJ<iq<d0UnsoVqjv0rP=@f78wO30R<)>
zJB2Ao`@eAYFL;$ukX>UI+sGKdgNKiP(<mke1RgC|K<DY`Y5tzUKUnz}Wd8GmEzldC
z{wF`3zZClc|G!83-@c;?9~5w?;t1w=|6i@Q7(n|2a2~;<=&m>hk%Q8}96M;HJVa;d
z?fPt2`OC+xPlQipPo&R}$IrKVozt@a9?i?1_*dWm+lP-d1)2B&)tJK77oaPrvH?;q
z<O(l6gZ{sUQa9g^5Bx_T0tw-uID$hS3e1--0Y8J52|;+-B!7)Y5L#SXA5FGU`hPa>
zgFp$i>(5Wtz8?VmOyS>V;D4XDKmW50A5t`D=qLk#%71GxC&HVDJ_oh3>3<n)MFcSI
z)8rnth(Lq>n*aYB{6Bjdedf~xv=^94|1Z06lZW1?Xl0wkF!g$Pc#}SMxW)6azg_R*
zVzN5oy^&)mZjs|_<U5CpO{e|KWR=AJPo~$6UT(O}jm<fY`}VQs?<|u<Gv)Rd(r@ko
zSG;?)b=K)wybJy|z*)r1GUlEsHD`HrS-C%mWezB~)!v?OnqT4+aZlgykI27Mu6^X;
zT&`-l3`PWQR^y9w@O<E{&`Clb8_D-lqKVCwzxNqYXp!6CAOCLA4GNg%y4JhfMRHs~
z!@6zBNEqM~NMiYDSu$qm5uo>UdHkF#M`_qH)x#HNo4*`v$Ys1ms<kkiZRp}t$#e6-
zICg$M<Q2OMH%Yj!Qq1ax`+Rrp6|}R^-hutoIyWUO+&Vjdg3ql#X?p=*=<J&{e30b8
zK?E#0NH@YQh?|s3G)J6pW9se=Gv{>SglEoqEO)@r$kS-t@bK7ezxrO4))+%-(s)d>
z{^te;12Od(r$UuN*J%oI8BrUP8(SR&H!nAd+8L9#51jqUEiY%5vp^Gx<=4?!j0X8u
zq+y~t=*FhUZeY1X$9V_W2Zb{q%<RCbt?%gc?0oBlE}ztb<8t=dq^32SMLOHx>gW)8
zP&OYvCz&^k2m1k+Eq#cV&0>1q0=Kfhdc)NcUKu?-CmR%JZS>Hmju?H0@n^}K>0Xwc
zwOei4sZhIcd~q1&*=J~U+syu2l53vt;fzrb4Lrv1ZQQpFn_Q7D-qd5C|BKRnrSvto
zItS5}p4QEzmrS>)?P~#0N+eNkI`H&?04OLV_f4r49|`?+ZGyu+irY?`tBcz+-oWRd
z7nA#M3IP<k5eq(wJSWd1?a3CaXlZU%+Irpv@I_9n@1loGk<Z^hwchYIEK2h6BVOrC
zTkOhPlnp~hQ|#MlW;4_+eWU_emUJ$-2`PQ~l18HG<lx96K5(~1%MdpppwYrf0Yefz
zI%;jwb~iq--$Fw|ANw&OZ?p6bS6(jdxRs=)N%@J|;Nn{%N_uwLm-S@2oHo@0cG+L%
z&6U0es{12L@9I1BD=1Lmw3&-zsOutXD#;^psVF@&yc_{!sBE#nP*f*t>zECja{{hB
zTJ)Q{dYM87aYT$bwE)rJw6_F%*9}C0hg#nZE9ES-PzxPgz6wX84O`<NJ{@U)iP$=W
z84wUsu3gC7-ri-W3zEs%K%z!9;W@ZgMldpcqX+=|r&vxOpXD2Q%lIS-itZjhCh>tC
zI$v9juV2!3T%;D|7q-xo>l-f4E70lR&Uh4ZY4p4=I*~6qs~XW&5|TPfQ^JTnn|LwZ
z6VsGihZumhXM8c<zT46*!pZzXHovO;Mq8UT3Av|hObs&Shwm*K>IFE<6=e`1m4n&{
zGubn;SzDX6kN^wIU9c&Ei(bIQG=4VEHRE7+gXfkTBu5LW6dbqvfXH78SX0yp(#rV1
zmPItjy=%;JJ6ZhW7r5R?jnJ_0$l2nF^ADmqateA<vTNjI;xYoi{~1pOnPxVKhfjUj
zb6jw6y};%6u(+1|o;(_=`hzzPWEX<r?S#`*vc$jJh)%C8K$l}?1n(^X7d#?9(4I~-
zE&GWN1D#^CR1feqBkX)ArQiRaE{?F4T}wZ@!Tg_kd6}xf>SyA;`$Fh1bzA3gU+SI&
zIJJcdAVRoGeQ~82V`uZYs<?X&UQQCV>hdwhREd&VE`4JO%gRWOQ0mH5&n)q{JXNS`
zyz9$pV5k~@Pj^eu{?nF6jOgb#Rf?s`xgaA}Vt+m{&qR^uVClTxUfgQZ@0po8X8{z<
zlko+LCBJ_ECKAL_YP}Xu{Tj=Dg+yOn^+S4U9uur47Q`8)QJss!@QopEc*K^JthRUY
zk{3nnY-(r}uRv4pvz!u|;mTUGV#rs%$cxzqq-qXHAr*636?%2EcX|$@9msG*n3+>e
znIlGn?orfq9~qbxbJg-iGatrj#a#C;cTQK+G_Pk`jEt;|B11M|*122Nf82sxe2PLD
z<5;nItotXF6G)Ps4sLO1xbCU?4Y<Cf>&ugek!>8@kdnwI+9$HYr~CJxP43&I#niOl
zKq4FnByT@(HR~pPXB{p3a)40*ydge5)qRRzwJ8%07?Pf+MUJ_q;oCc|89ch=VP5;k
zOlc?XZ4T~){P5J{`qW;%=ZtrDw;FMp9szsdoa?VUV}^T)35y_PZk_BM-w!u34s6;9
z=>1BtKR<7}y1W<cf%EtlSkU4)WA)vVgd}ZNtU`^_fh6vs`tHr93-PcmsUeEh>#Ei!
z#q0oczcYB&_<qIX!a+BR9!LD>tWoYqExDLsO61v|npow%yEXrXeOmmV3-?0y_rFYb
z_;q4NI92gFN&v0Ekaz2P%E|1PvByi8V|$C_`;mFOvw#Lxa+#^0MGEYmz-=3QxAYpa
zuNjkD>{0GbTQoA?j4<XRMoMmAzD5RfICqdNuQ|H^;zmPBG$m$XtsozA&+Ky+uK2og
z;D0RnC5)pJdo49rCR}FoO*%#j*C$h+9c4U|ylEYMQ(;aP8(ql}#1!^ltvA>8^)Bx$
z9Qn~4+#E4php?oxDp{E&Md=f-W7s2`_}?cZjlfiD=oz@TJUx1uF%gtIQn;%)+RhP>
z@;o=EkS`q4a9Wpp+;{fyHQn#(g=H3Mr;*x{-b(-eSu*U&`H0(UXU_cf`O9}o>VC~H
z_9&`{3JtAAu2~F&7NLwPdS6EnfF;@xoK>(n>bWJv#3rdy=~=3gYU{7JM%gV~$n6F5
zFIlrWjblrw?YTi-#haI|{)W(-#_+H0Ny0W!W^MPVubH(Op|f}C3dwC>32r!)Y&lPI
zzNGMiSw;$>9{3!UUM7i#R_p%0fSE<v!zJ}NOH%I{-^{nAHjU`%aPIwYIBq8lN8g>h
zheXZN1!0Zs2ozO<B|q)rfN{vpbC-Af0uyhK!HXl5)kj|nW{^+|O^OvP!w4@-#vi9f
z+=mr&HP<~b?ukZmT=P83d-;5(ADJV$Su1U?evFm7F2tm`SY8xQlEYKIM1(o;@<%Um
zd?gU6lG4!$Oy(}2kIWbX2dt8|AJjqJks7~5e34GO`x8`{vu$s+muI)dETf=3gyNVk
zd-(LS9zU@W{o3q@puKkt1zoKC^6+qxvmm@rOE7$hNa4K0pV?=i<QUz<WvqUnVmd~N
z`>H7JH+vQf-dYUx<!fF*jnh>8a*Gg@dA`n?2JSPX0v%bin3)-Op4(^K@CgZ?Q14k@
zOaJb)jCnVP)0kiac4zm_nw2eMiGVj6(t0E3yE48-4KJZl#_}dQkt;gJp~djAQ~Mm`
zZQw43Q6Osdfj>ItlVyRYiueHF+!=4DYp&=M2=%~mXH(O<IWtC~Qa&JJk6DeYwDVGc
z#&C;rFJfkB%A4WlI<`*K_F2Q<xVb%bKD(h3M_j6}zQ$s#S~$ejNpZgnXaIGmt^4~a
zL+|#b#YI(#>nQFSDs*wz!`$yZUh-xx*P+m3&^P$@qao96mW`l=2kB2nPxFN}WcoMO
zlTcS%jwCD|KZS;K_(H?t;hBKb?&pIrPgYoUCo<a{22Wh1OeCLg7^uFfSF7{eOt&u~
zc4A6>7f0L@G3tIk2so}2c$7)Ky1BLf4lJR`N3BP5qY{ARTPxeFnO4*{F!EAOe6Je6
z&f?unVCUvB(&d|8;IU7TbD6N*bHt$0sXp^H(eeUoXd!f>yb!a!VP<B=>q+dghrx&K
z471hxJ^VZ8RJg_3GFFATok0a`rL{}`KK<W5N}s<6V_$6Eg_4W?YU-{>^?K-fpFXR>
z-Ldt(|LL`w<gZ6n52&EqZ<jiw9!lTZKjOodvpG1(f+)JxTK5h^+_oym%0XZ<eJBWW
zkYxr>5+863>Cy?p_R7c$mhzEcW+E`~ShA|!DSYly%d6K;L+r*>#Y>=z?f-F!B>?--
z2i0k?s%2^B;#U-k@<iPKjZ~|4(#zG?mRbwHJqsTiCa%#_920w^;*)RNYb>=+Q|@^*
zWA)N$n<hS3C_dghiy`yNVZ*O&a1<Iw?G3fhllD!@dxH_xSbCpE1WGGKqO1-d`;Prp
zcODNp9o-_s=u3+AhkgBUsz}L<nxI2|hNqIwj^pAQD$ZW=MY?B-@Ml!DT?^YJb-+-f
zjNu@)y^3E{)Lh4eWrcl0wG+UXN5zO?N&J=MV9Lt*)qvN*Y26(OyKBmI{hORW4oOMt
zV$A#WF5V%ULoA)j(9kzqPh#hRIMu&;aRyb7wBWa1vB2rt>!M9{O#8B)LnokCVXR~0
zYE>MZQ-X-XmaDwomY?s9<@8P;5U~i)apJ?Y5-lvlZ{>y?Eo#`rif7;0*(PnX4~PKt
z?eAp*?i{)C%FOx9AncO#ThS!JT48Kq^3ZB55{3%r0361WkO_1;%=GlsN+qUvcz+?P
zB1y#Qy!zXYVm#7$Jv!_;QC)i73RENHMv2;2YPy;V3JN2-6?H>hka1TU`Hw-^*ZAOG
zSSGy>0-zHnfp5RcHMAn%J<F3ZHHk-0KYFG|v_2h8P@itUz8XGUB-Pu7*m8qg$|T4F
z70Oh*`UQ5H^gzbEjvm)A8ijQZHyjpiMTQe2iLIE|*05}H1__)$nHvTZnGgq3U+)F3
zf0)^KEO2xiPLt6PTcL&;rS-40tUXeu0=@C)#M<`lofdRE<Xf^E9Z4!|5>|X8x<X*@
z=*hvw;y{gd3<sD`<&|TCrR@agCQ+OB{)SV`8gSFKE%r@cz)%g1Ei2nU>zRzaoeErh
zO(rNnHo7_s4)YK&fTKIf%an^}p0efYN|Drxazm6EBMWXs-FNIkIsA6PWRb0RShhbh
z|5HPcCWV+xN$_Pa+D`w9EpLrX{cLqujyO{Qu)duFojgt>zn8Az<h)ax({JV0@wLK7
z+=e1AOOIWieZA@0QW2e~qc}B2MWjSmx@|nqcsWP0Jt7vh)7UT~1yT1uK6l?ko2Z1D
z9PoZQlOo+dX=w8(y`EgX>ga<Yp*fN9yGbQpsjO(&bw1W-4p>E=c5t>=DVB9dRWWHZ
zqXDUy4(ry+ruMo$!KT3dm_r^O#&Krune}ONhJH__8&joVSwhHlr49h<x^p=7ST9_X
zO@YV|Wl5iaMjibFbe(U-Md;V)jWnZTy-Uwv`@c6d(#wV>zkRv{=$W8;bowXZ)gMJ%
z(;#D?X1a`m79K%o4^v~K1lDRrx*F0YQJzaUGQPmTW^`R-us9v{!(JCTL}o8pJMP}<
zF(N}|UoIK&J|VB-kZZuetIAg(c*r#fuwAWiy|);~o=?Zc)}BL7&=(k#0BCIXyY`@O
zWsPJ!b|rKt6rvZusOr3(KfYVb-O+0O4QG*@uHgca3$5kjx*=s+V);w7v^|87&dCal
z)=rHq^9n3Z*bBLwp?B5Ufs*-<6O3I2fk&Y&yy#3n%B0;UkNUiETOj+|1M`y2B3t1%
z?%4^EEtq41^+qFBm}$q?WhE>ec@#>cZGT^I3vMgvEDw<ev}CU4ERJ#*@Oo`TJ+&<W
zjLIV2;4Sr0Q`hS;-;5e$8HREhEprPOKzWCcsKL!b7m7r36;UOjw~&oVZ%WA0dh%&f
zVsv}88qXVjcK&$|;q-Je`)6k>!t{Ym1q;PN{EsCxppm5mbb1^k3K$uxA!?In!Ox&^
z=Gjs7GBgiIFDL0*_0=xd$@;u<1YHZkFe0^SqDXuxO(5yin?ZDjWYOvNaD)~H(~(aW
zCX*<x16E334_9Ajw|&?|Q;=1OgWBC%lI@r`y5q;r${jG&H>IoI&XO;$mQ@!wL=ob@
z6TrQLQ5zYsTa^8n<LQ+4oZQL3mt_}M=p`E}H8*<~A^82V1!^g_f+_F@L)NvCRw8GX
zXK;Q;eZm$$_xIX3A~Q=j<;&jSnpRA{FYaJOjm}D!d^pSZ;qV7DLwDu#i!+A70|(=W
z#+f9z=d#H3P-W-q@U+mZ3m3(Y9ucce5S!1NyX~{CA%dvlynTk$sP)fSjP(XQV4I;M
zH>N9yC<U@`VYU>KN6zV#@I%sB4gcmN^zVa?SH0ge1<Lc2q}rDfk_Ritf)eyw+iKT_
zd>^q^SDUD$(`R4fNnfjLovQ}W_Bg>T+S=lQuWJ${LkMFFOjxxZi)txHl6uvaWdwb_
zQ2=(Y(r4|{-U|)tL=GDA1`W6^Vv@+Dq$)hGA$Az9es8A!SeUBN%I1QO3rl^ZWz!}W
zga{wkpm_BD+f=v?!?$AAw6|V&nu_$Bt?C{2sZyt*M#XWE@*0F)E~J+ozgQ)~>0WR7
zG<KImpq_iaR;(d!+6M(c_xC#*h#HD_E7SUyjYRIYQ|7TUo?Ll;HbjiKFd@mby115y
ziuPq=)OEo?1ZW6rK5?Qm!!{!cY0ZD3*ny(^^bKJ;V@1dOl9Q-Z0TGs~P=`G-G<26j
zCCa`Yj_9FaP@MBQrx<^_=hvVy`DFRXRn^USWTQ+`c6r8VLnF%qR&{a2C(`MpS_lKO
ztTIO10a|L%!_}|SMbeLB;T2+v5Bs2x1!HI5*KJo!gkMqmPvLc20`OaEt-q(lIzEBt
zyFP*c)#-ZcGE-+8>6)aj;Zq^Nd(&y}rHV-3#K=t1{po?OTRrNBR5|S)UziEbDQfp}
zVY+dhw2?9kRI#OV?OnK$j=>(GuM%Z{MXyo0y00U~dp2~THN+)Rd4gU>auf=QQ4#d2
zW&Oxr-D7{l_S>OlIx~=XNcd07anBDH8Z^|PwUWO9$OQm!l)F_?eJ=u%vO>(6);w6=
zW@dirKA<0t&NDi@S0M@yHO-MN>(y#ZU>L(8oP+y`iMQ&rDbj^~0coK>tpns#Dp*Nq
z;hx~c+gaF|ua5(Zp5`$-eQ>oNF&ny;Y8NY(D=|*Gp>;m7+_iNf1xjgDm&Q5i!eabw
zq%?m{7}<#~Kuw?NN-cOr0;loB2n_mP-1c>-lBZ$M4bNpyD9_z6a9$U!>(6jd{yKJ^
zoy&_}Hf)N|TN*5Uz#S1!*Ut~-2K&q9_$40<*mD^zLN*b5DXu|Rt$Kq?9o}tcxN!1R
zb^&V~C`tqTw9oKNzlWQ@*zmqOg=So3qkzq1QEBnZW<A+(Mmo;GqK7xz*xbzQ9Tyar
zn{3|^Q4#Y0PKD!O6kI_K@8NA`_pr&mtjOXlPhUGuR;RCn%xKitzOCmW7(bbt9!yPz
zj@l>Biz1H5Vy=tZ;okoH84sLi3ucopR>)#FvAhVD&mnWblHy@ilD2PdYK+Umoc4lC
z_c}-34OBD@`he<!MnOH|KE<r+XBg2*iI(OP^_RuuL|qDsGqIcIIko|r9YE<i(vXAn
zbr7jT6AElJbdBF%`Hn8_vt1=G9t(dq5fyc7ZWaXMmo)!Nwvr%`dH#)uohA4;a5<>c
zoF&#Yn453P%L#Os`1d?q>stHK=|_@aNIoh4NZL9HZa`fpHfQbR7tRq740kn1yb#ee
zzNm0cvJ9+<7i;Q>-a~Y1?b?QyY^XdZp0RcD>@#qR3)3ir+c@^Ctw5Cg4jb)T!MTGK
zlz`J>US8(z7%O`8ln5PdxX+Yar8n_9Svn}L&Z0h?b8_qHu0fZj6tfSzZ>AN4VuEQV
zlLAVxd9AuaSN9sygB3+aG?I&6lxCm>)A9F?J+eXlm$DAAh9W0Ag*vu3gd~&vte?u=
z!Y2gXSHicy@nan;yL|neAZ2^i1FApUEA@A)Ia<xO;XK5eH01)aiy<5L#o@O_;ID%G
zw{kj>O`2n^2h6m|h1(t{{6qJu<+LM7v?H|xZ+(R|0Vu}c{@9&i1c=d|wP;M#III{b
z=(Yl1d3t$EO%dp7SQ!Nk8|+Y=?}OX)vHfjxs|tVD0=UqyaaK2=JfEFnHekZ4l}DmE
z9b_TPo>o+-HcT1EIf+5|iq&D}3Lt^ncH&ti;R`oIexmFT4ik69NRGf{jTptwAHt2;
z`r_{yv<EtJ+f7)XhU$E|Yf|TR?)DZC(-U?%poNzO|7MS3J1*?Uq90|&Q#&rVh3_1n
z<<6g?RYLp8btkSK0vip-FqxL{t&-vda@_%|Y3ON~{OdRk=*1zcpnw>5D&zCNNma6Q
zO_n%QRzs~-&;n07yi4|%Iz3+##!|U~FXu#MpTOWJ2D}TEkjo|RsWjKICktREv=$nT
zk|X^uL!ww25u<%?mf<5IMf&8TigmL_^EraCqZY}yL`d4rvE9l@{Iso;gU8<i2JG&#
z?vAX+Molb&myY{U^y_<~tDf#LA^DW6H{0O$#1l;Cj~xpjJe}1D^l)7RgNxJctC8`U
z9<pK^!QFD{bbPM|+4eE}K2p|t!yXIzcwEy%YEnqM%yz8b3N#`)vwVXzhcg8?$?*!J
zIbAw~V>LMz1;dZ31~|QxFr#V~vMI7%y!Me1jvnrMDejvBpoX3J{XCB?{Gqn{rIsCg
zrm$JtRvihMpz)%G;<qE!S>kB?_@-=Eb|i@~6uAJWBx4-Cg(Pv6qgKE2o5$sFj4n1u
zA?rdyaW3ZtQ;2@D-;9Ph(qsFtC3)S#52g?RI57f#a%16tS~kQNIE5JRt4-6%5^kJ$
z!<c^etstx0=xE@Qlk4CPm|DH&M|CrWV?tVlw59;WJ``=CDr4)5k}j=UwkPpTu2SH4
zcyNHLR4$@O<B@`*<8)|G5oTJRUaU45fry)8=Kj!1z2j@932saQ#ou@1sBMa52?s;)
z3O<-HwKV7+0})<Vd4NQ1I^8CHj1Gy}=J;N00r(@KM-~=+1@N{?Mgkp0cPRAS(sRJK
zguZ7#0bx%_Mqv%8ybnNnlfa~fqKw@RMZ}kB#1|XSc3b_Iy)CvgS5KgMECncvP;zwP
zTTkux$s~U%7k5)kY!2P5AZ*snzI_RSu0gTr^1S!3wn5lc#F-Mpuj{1k1;GSstwzmU
z(R2wef?My-HpyE36LvPoDL^X{uo88M)p%E>G&1)E`xUOVO@vW>^B!M2-a+k8HEPlB
zx_-~vrbluc1t^WxNS1~U;`&bEh(=>^1P{+4@}{^Q=)eT<kwyI`r*8mjSB^qQOZQ`A
z89Iv0T`~B~mQi)lvF9Ur(NX*Zm-QdZl)YxH`3@l((}`K#O7=(K{BJ!~w>3=q?A7KG
z`1nJaNrL1~*GdR)uN{MlcW*Bb-j*t`=8v*ZRK)GMRDY3lDPM4wN`=tB_<Az>e5+Rd
zqehI3E)NK^=17(b2hLfM@iyX42W%67ek>7eUQ@lmCd3{#NZN$uc9}5S9c$n{uUlqQ
zspr%J!SxW<rB-eeSrjg$T2Nd{tP{bVB);h?+SAdKpS9w1-cwnNEN#)F)P5CP;*{%4
z+JcuH)Z;T=dW~%0&KwAZhWd^RhMaJ^w$<nuOJkFQ<lT~=jk>EpjYf9I9{UFh2@kxO
zCm;`Z7156uN-2XFMYr9X-plmva&UX<mR!&@lQwI`I8<u2Q-N!cS1UtX6jFsMY@&NK
zT!;Q}!Xcl#ooZSHJNPpcw&YLIoI11-vrFm8&XVwVm9ja|<<cQGNeC3(IQL!~Kguy!
zCiebBOV(-wt=DHk$Xb#9Qsk(IFbb-)wVlW1^|dOd68>ERyhI8)`(UfM!G|iKXheDS
zFey11xHkE^c4pGEDLY^Hm{X3a1#bg(L=!zWidSjIfLNK1GqY;Tqeam$7mRs<BfxW^
zs9|6vIB@Km$lx5qam+FYT*_vthf|P*eCTm)U3PVJ6=Scv8>6!Nv?Ly)$CJbZ{B7X~
z7|jWeqcWMxd3<e5^q!tynu5CjwK))kId@ZoP9Elgj8go7WuoKP9J|KmP4{!8g`(E;
zo&95ulfGz-T%p)_;#4Cq3Ha#5%;w`%#}A{dud4+^01p)Mio)0fJaY>iHYY#6T0Z;L
zl&!WOnAWfuEs)xf4@QVy8==R-S<cy9KPu<iDn@+dGy6jic|>mv#a1b7x`?#F0*j9w
zE&$|h-j^p%k&A~A$W!8Jv4S?WhN9nCiHC$xBF-GHcM5r+O%25C3kG3(IQuLzV>1G|
zv`oXwAt9H5%5i&Qs?)Vf>72TJgeK*IDL!Tvme#<flOJubhlQde_?!E&Bi3INW6wh!
zr*K3=ZpLJ5oZ}kn97S}es-=qy4$bdi5sD8iQ0I!_cY2=bW;U<6=Imq!`L6AGP5#pL
zJO_$SvT9{7HpJWvJcoz3d1-OQ1JuVmU0qxVp-az|g0S5(o4MdKjeWO@)BJ`$Xhv+b
z!}s+<u)cLp<{hX!`3<Esvr49Ikn*FVP8^9ft@?!P9pB9acW+|9?t%LI?n5{iUw2Yn
zjQv#c<G%G+)aW-8nawN;@Q5ZtNq^!AM)3jXa+;4~Rw>QDnPo(C^gSS4;Fb12RW`NV
z<94xH#-?FW6~RfxR1a82NRWe4niA#ZJ8#l}TjbqT5uBid($)0<`RPkOB24;dsK7{!
zbYXab5mNeq`mTWuNj&79Q0^LF0UTe8y3HM^6U!!>q;X8c6e^>`ivgrDn}6OlaOy$6
ze=h*+zJbs7tfx)H+?AKWX^RjV?)5hdD8yG9uPL;#q#&p}w=*60$#Iqi`;18re?Ox3
z%e=Kb46@NUW^OaJGD=5(GdMU*WaSVuY*O&}D+eiEk;ULpK2}QUI1q=Rq@PT;U+)sC
zYB^KjfZy`rom~*J4A*naj<Z_K@I8GbRP3;Yw20M}G`2C#-6TLol7ZB?uVx|ndFI^H
zYOaX7%L^q#D0I;~UmK`*Q0=-;vDW(ZZ>iTJkfW3TwLR&YfRa6*Tp=lWDfHFcNIZ$B
zUKF^WRZZ>pY|sWz=trZ>r!m9)s}urQFks{xfDu6mQNtOiY^sm~N<;jkXOD`qhdAkq
zD_TWksFKTB%dvuDG8EbQj*jZD_?E>)t?1A+tFHHjo)Ly+aUF?KwGvJm2kgPE`I!}y
zSai$UX`u5Ze&~`fKJLfSVzh-F#4hIO@`2l1(6Ph&<m$liW)jSHsj+w6sFKw`;dQK%
zGwCM&<?$=N_B`sK!4Ss9c5G?B(^PMI2`Y=`-|nTJGrG+s$*?46A1Vl)H_HVSDCfKq
zxQ+xExTFXSCp!O#KyAHLyTny#8TUx+tm;w)OMN)BrtHMb1c$YEJaZ4D9NgU7yU}k%
z^=~(lMB@Uc!inLW)SbmyAAmfBKe5;i`;s;4bC&KXe^>rd>Fzf%xKY~O*z4D_27iLg
za#aI2iMZnkyQ?2_<*tMiJm=!=!UKqq(m<w?n%+0oO&2KVI0%qJfR)j)Sp*BC!*<&R
zFw5wFjLD-LJj?jBi8MuRUcwiQvQFm(h!bh?n^^Q2(&&4L<7D7faaCg&K8G*C8SY`l
z*kI8LIxvPg<{DroIA+&CV~?MU)?p15_0Ceb9-CZMu#?Zx97l_`oEIrFOx!WHJIX8a
z(E~@3uUB5Hifoak{0$t6KW1O{xs0mv`_y#z+z7@@dYg#_bD5liCZ;^{T+C5*^a{7W
z^_Xnk{15qQRJMV@Z`ZV0|2~#1Exy>()EtYj-ShKv7sy?-delk^bw}WQ=^&A_e5YR4
z^fBxEmXs2nJ6$7s7GntAzSn_m&>TuP0{n-K>R0WA5?Ldh0kop>K;CD!Fh^M&ZGP1l
zqqLmR^IVF+$$6MFmkwFRvm_E{()MPw9GWtY3^<}BuatlZ1%)05HWEJcZ*G~nJ=2zr
z50Z&BTew((8IU-i^jI;O04VLRagDqp@Sm;E-49)_9Cfhc%P7^59`BFNx^C3tKly@_
zH-Qj+P3R_h8;lZ(9qZWl=t=nE4$SRK_8fQp6U=7Gu#o-Ebasy;8~=(4-rJS0j+{+~
zLN}g$cQ`VBW~+qsr4BlPRRZ6JyhU@<r6kHB>E#6>#o)<jRn6cm+IFt&^2SD4z*-b)
zT~v27Rwd~6PIa>rqSi}XyR(ynvLa9x3pyg}dS5r!-%nT^$#1m!aYt&!Tu7f`7^sQO
zKA9V}Pz>GlT%r9Yo8S+Y-p@{Hw!A7{s^J(Bpjl=yhKA4rjw^;p{nzD^0H}ukfC>XD
zn~+0*(bK@mm?+F`v4(v9G^B82zD`b*cP%0TsTQ=M*uF{xX$OgjzTo>wA;2!6S^7b|
zevd*WFjm}F#X$jW1$Gn0FxeF$PNDDPTFshVNh0-HU3ar?<S+SSCv^mN1#(5cmGNU-
zW3vFSRz9ITIOh6dU0j(k&E|GI+J(x9v2Cq;#i&mO%g3kA4(o*{Yu4<FHUKRH3A44k
zy6peFmt}O6yCIvcpu0`Vv^VQ(!@*wQru_$rh<4bPhfiHbPzlb?g}YX8m9V)*14y3$
zKx(w%o6tDVCYImfw+QfWAjPsEg`$`<@gwY#PsrrV;uG!><_}X!T;nP);TyOqf;I^m
zQ?9&doTEkC131Oh`_`Qfq@`Y=u~UOCns5`j$ud5`oJ+g70KBrIhl}4f*_kfcvjlc9
z-^>NNJU|-?)WoY3AL9q%0mV2%O-+;zo8TGkiG>vjdHqEn%mzi94)5<HGX?u>djYiX
zw@=$=LfpH4(genERIwUFRX!sTvfo6v@r{A+nx+j15uJ=5`~+7H@Lk5=mu3=LXjyKS
zP=$XwFCa9T63y&#!(a+p+%{QknCWtEhmFpPTjb$SCwef*`B`HdUp-cyX~gy(LT8s~
z1Exql)q#0tKjdF{yod@OD5gTpadj@=4~<m^&fp%#wr1fgr>}e}aj!7TvuUs5WgLTH
zA4$L}@?tS`hqm$Eyx7;q&SARxBlpgR`swdYrC1eAI9n=OjnbX4G+jw>SSci~Db5kS
zP*_3ElSSfPG~zOom6-|G=R4EaR2R9+i*H_tLdw60!b7W@vJ8Hf>4mS2#@HQi2j&Jp
zx9)jhZ4Vh0Tj!Fz1TvyIv%<D-9A9X0pn{`an7+G&BwplW3tTrkxBqfMVjAVEY3b`)
zJpY_(f&oyaHci(=00lzG2O>>BQ9JzRU-?l+l{as(I2AvAU64wgr*s|prTuPnRoDi?
z54zh|Pt(@0LQI&6<G7rMfRah$6m=!k8dhW@SEV8|(W=Dq%%^$6bHxv*BmfN6BqV7A
zA$f0;(bN9Cu}_+Y8*NG&=TJu`?2RBQwt^mMvb}|5865WM>gKsBt}*lyN7J&zme8iY
z&W(VS%*JLm&=9f%cTCEeKonZ;yHxv40H|j9cw%vwH}dlLZ-FUu6jVaE)1O-atQ|Hf
zdk_VF=^~_Uo7CVt$d13O6v@f>;2Pxp<vJQj@4AEyZ6Z}^=xy@CLVb$L$rBGL+u5rj
zg~#=MY)>{JxRb>L(hujD2qjZ@umKL+r)d)?aSWVlJji!Y@-`^a7G=B3IO;wL4weO6
za)f}!W7l8({3(z`tG3)?``o6@UK5XSfV*UBR+)?pIr9OoFsxEXOM6uiVE`}8R+$1O
z4Vp}h$hSeEb|ycCNoTXRjR=uKMU{m1z)UfV-w2Z=ajYaMdxW#e!r~fE^nOfjX!uGj
zl<s`!9(@FH^kPUuSitDwC+CGi6W+O|o0nKBn5M>sA)&aWs#7=X%Jo`XEREf{<Q2cW
z&=n<JnKzzcey1pb2T+~;J0g43|B_y6t)$ZwZW>kp*RFnguZyu%eZ2q}!ru_a*AZDG
zhn&Zvt%KC}JkUsS%;=a`<44Pu83F1o6U5r~gF_WF7xqzu9T6-QPqrgJaZf&|_${t^
zT1KfEkZD#gd`+5qMHz`_62@o3s$9NVw_mX|qd~hjI6|<O>8LQwUqY@>zH6D7a>>I@
zE%-voyBU3*%vUNM-hqpvUqpqM?hX1hdmI^{R8<+jzl`aGQeE9zR8s+N8;NZCOkRQ)
z_dqsX86j$ed+`>w^SCZ~btz|BtYplu9gEllf$Qsf#kkgTbJ@>2N^<omKE2`LQTslo
z5^`s7@S4z(hICQYOfPQ_df563x!?gBUF_f~cf71>$NY+fZ^mNKD~N#l4kBIpW_j?B
zdZuFw`Iy^x=@5^5!p>rJ&r_Pe8(O!{5wKUMx^dnS7JCdskMbCax!4;r2y$58ECm50
zO5gT?RARem9*gXgo2Rwy^?gbSf1K)Zm*A<rOeNI~mlmO0d*^qKyM-b3LngQ9TGAN-
zZdm=&10Z8-?o@lZ0m|Zis|>NPH}ea%AFy#hLkTcoTYpp|S`|CIcyYx)W&&o=ull-w
zPx~Gr{=y~dQ*kcmkx8z2PftNzLZpR3;ou+HK>{8H3f5FD220q`Fj)iYQMyaYexJBK
z+wf!zPC%)(+#;ZwR%rX2(Mhq5MxZ@a97X#&L5k`7o_z82Me4Zi-)5J{Q@@qQqHop(
zlN?#=eak-L@*NUxy`g_&b&YVMqvIrcTi?xkp#bbEi5zcgTc*pD!E7!4W|(sRp+mL(
zc-D0q-<(ukN7v{<t}VuSf7xI2cSpsBn%1;&NOe5qtxy{~j9Z!)6+Se}WdmKRE7Zh<
zdyf-RIc*&_sm&=pGh*cebTx6-=EG-uLg*0o^}N0Ondeg5DJT5e7yhkB;|9Em7bhCd
zT+(&JboMC6qVk#5r5<W!;ic-G5L#TtJcpkB5suvnji_Asrc2GPmaft9Y#<=_c6)TJ
zNY(KZIYbnmsoKDin96qAKoB}fZ;>ouh1Y07OQjj@(fbJ)C;~EUMnNglWIgf8#YA^}
z2l1nC{Tj96WDV(2;Zsa96*C3Q@z;_IjF9R3S3w>ddk0bB+8tC!d?ho3OGyjM{DCke
zzENpBGY!<cAE(t?0}ZgZo#+x~xULa+^M2d?sjwQ0>ZzHwyDY1-C@6hYeg5oBTKeTx
z=;^8X+GZ)^iITOV2J2%p4_sNvk#m~hqK6$&z&L9(flesn7Ut+b0^CaDdFY@2dv{DP
zO=#bs<2MYOkcd?~{3t7fpONWh>h@C5TdEFIdTS6C1&mb?AD8f?H@fJuE5Z>cUcO#I
zWyA0`9eN{t@D?tiM~p+s?5v*Eu)6bGgS6Yy)f+Whm5pQn#^zvgUbnfunP?CCp2dlc
z#{WsC&C`JN4Y=CU+r60uPvE?qo_)UM-*G?aJPJMCFF&8R29_$+CjLn2%yYZBt({3I
zq#Cgmb)+(5XNoMEqOM6)QY)+YBr;2{-Sso5w|Tp3ol8ls%M6K0M)n&F1NI;hiv+zz
zhT{7mSh}|#76(rSm~2A25lQh+KeWA8%dXJ35?XQ#RH@8NlxSRK_Ej_vaF6aJTz$lE
zZO&)W3INz=7j(~+4iR)jbd4UoI_nguH`w)LFy3$-nMou|YZsyR8G0FehN)tC87MdL
zfC`9Jy|j^Vohm67NKvzpT_{u^X~eAuPI}qJZ7S9U1ccTs($)#j>=hIBk}R?Bi4w<)
zk$*{_r39snL3iV9%|E#NuqNXt0P-f1WCmO|>=DHu-{DUr;7P;Wh$8`Jwkot4IbN+m
zMSCTb?gVUzm&R2=1|@>XUd{LRR;~4oG>)@vZ=m@o$4d@YiZ{x>f(1O)JT90kowL$x
zxu~a=<=)aZAOwzTUVT&w7z`54u}S^*I@10?wE>ekOsd^85PDk1_sqyYyq(Ve!xT)h
zc8R)%*cj;vtgB`1&e5Ikb6L#zJqGYj(kMeDiF5Xan@&v-Y#I-o<pB>>Q2I6ARVE;4
zG8VGyfw-cg>s?(9t-p`oSkU^o(L)0;$eyo!V2x)U+)s^5DQcu7oC(0#h|3BccXS6%
z`HK-#90IDukw_Xpg(|Cs;=CI!3m4FP@;Zp`>3cXvY=|(?NP0^7uXKGYy=7$5X`&&U
z*z#9C1S;CiioKHZ`ekyQ@0QZ6y<}ll+F7Ja=j>uluj-OMw%te?os|~u7GFJAZeYYV
za*WS-naeqVCrdf7eN6_6Da!;RAZ}f<_=@sPq(pA(jTnUJtZhZ%W+gFEBW4!X$wEb@
z7H7oI&L~g_$kr?+MoP`Z*<o;6GHxkw=&v#Nk*7uQbytRFy*hOI$?X;n!NTP(C7UNC
zB3>1}JIaxo*L8T>_OmkjC?Nmy0cZG;w$N}@t|1>2nj<qb9x>nn?WZcBw9g%n1K_Gc
z3XGsguxHaT9$(%6JZp2?`?_;?7FVfwtNCz*Uvoz_z+;n8UEiviR1R-nThX^f8<Ex(
z{429#sU52S5Qq5U5%2+AcbvOPw9;M!KYhKeUOYhf&ixcE-~py<g0WKtk{p3NujpSg
zi#9n5@6ZmQ9KUyDmF)v*DQ$t^&GyU;o7}|Yd$Z8L<v@R_Ds-S#<byfHIOnY@C;!Mz
z+InA#?&c1)Uf2`)e+U~$k$v1Vg$WmgPSkZ$h%2@yO9u9OP`?dSiyOp!zFf#%{I^mA
zXu-W0{@~zwNVlTT$l2cGs<eY_nfXP`(xyTNXn%&k2n$3<xx?e?y$m4n-$Ec^^h9=O
z^enP-xfJnPSrvS4|6pNV3@YVRH4#WF|Bllbm#&tjTOUD-7=NYj)7A-J{~-?{J1M0=
zcYvJ~(~&?)y((R!|Eoda?=SfxnugtJW}vB-e>LS}jR{e)tluv283RHdzp-+ZJ!2vA
z!_{=#_sH3vAce@f2Xu|&_7Lmu##Z)`KjwH{O&3A6kA0^XA}M5_zj~OZM&Ut<5B9e&
z7j9F2|D2Lz^&exWf{tp1=2eNp+A3VgYqt@uKOp+Xs9MBadgg+zJ9Scqi}9C!CXoq<
zzU6SAsjPw0Ms&fK0`NatB{KLSd`L3TbZtt`FVAsvKIyIK2WOxwMv<}hcm4Wo8KD!b
za=5|&DH(A`hW^?Lyj^&UFye_vtCy&c@&Bb2_(RoF7Nr)AJogr1pr-W+(*F#o^PaR^
zQrxcO#L`oxiw9H(N-c_6x9({_=7<2X0iX`Z1`=UM$7QBGDNDaTy@mL}0s4m|$A?}l
z<A-9tVWDW>VM1~z0o!1Xbn8MUc$&Cc(~08}S426mkVEq22R>Ek0O7*?X`);mfLYj!
za_1j{pJhHs%)?_XQi!w1e@^bd2JQSJG&(><{_i0DfBEWvP1FA__xZ2y{H-?n$1eFF
z>YM*f{`8kz@I@096o<wE`yzaj`^TdGpN#-WaL-@|_J)B8;=cuXfPO3y6`E@G*5Sgh
zN)(v0HA=F54flf$>A#g_f8<F%h2p|?IdaCe_?$99*p!Ju|3f?VXEZe!=%8+{EKvGC
z#s=(2XkZnLAv3gGb6+a$|D*B$YoPcqH0vBO_P=ZNKPXko1)6S8v;x(Tp!`SSZCX8=
zdA_RkTNj=&F87=SlQ&5i3YB!rTvkSuG7GsSWG89xB~}MY03mFDxN=@@Ga7d3np9q?
z9}9-|NfHyjni*sHSsgg?TlmDvCeS=|P^EC%^b3nZlI2zP#$RHmN1pUMdA%6cI8&p0
ze$Ry$aZHkGbmO?z=zl7iz<(*3hRpF~8!5>YnV5ggZ>4Qx{jv~lWnYLPH&mE#b_$()
zx~iS-<>j9GOKK!~0*I4dO)5GA_Z4_bm$ryr_Pl;RpX)aP4uWC}r_6XS>Q6Vlzr}h5
z35NgF8I6{s9UnZ)&HW~Nq3z@waZ1Pga^F^6KepZ$zEE|&&*Ek6Jy72QUNbYXOo*MU
zp|;?@V#Xd+Ddslian36a%}(T$@0|^ryFt6vVBwZTs^<G9qxwBLJ>QZ=(r-2DPC;i+
zpsHZ1DYrf2nY`8GyO6oZK9<3V2<p?V!lyIP);`B|S=9%0DJY0JJx7(;vP?6e@@!vQ
zXE_&?yEg|-4qCgE%T&p{9QN?6b8(a4o-sLP_`afSr~Vqo%b-#ud=q7q1cSoZ7wR%q
z?@=ij8+`{5L#W~r0&1^-v^3Y_u!E{*mb?1?7i};qcT8wFr)Ej{iBM}~i}n278%;Mr
zF&Q|+@eLPH)^WeUaW8jv;Ev@R;}phV!R@Vp)zMSmdpNM3XYRJa-4X#%DT-HHu;7vN
zGn-o!csr$wRgjxoLCa0Z7RK%*JTm)f<7!DEE;%_LFe5stz0{x}8bPjuj8k^=j~WHa
z9>#t@z8e?a4+i?ZBlf&~G;%*d_UQhjI9kJ^f1_to@H4m2*P9yJz{gSf?^=LQULH_y
zc3+zp%I0N#7p)ZE*NWH{|64fKLLen@ncm_uxWPjV;|GKOHR`@mwxO9XQhB&54FDM3
z#mUMdkb@oVQfOvb)J?C<A_%;XddhfaaoPz+fp@7yf|xoN?*=7J>q<?0ypn2fF0ZWV
zUUsoN$w#Nh)?tNUjo;p`tyI#}%$y%*Ffs=caoMQX4yCN6Ej8;87f6mICW{sfi!eYr
zhQsNHpI-qhGKK^A2modLJg>Iwh~#DKQ`L$bHFX!WbZf=(CA}nz**=@qHFoY9%pNOH
zX?ts<8h+JoGXkXOu_T07(MiLv@roP))*5V9^79bvW?rksL4*vi%TQ6pq;AgPPBK2W
zz5Mx{(xh&RtnH`$x%jBq-q-o)R`D&1PkkGJ&UFhT>hZD^2>{YtA^a*d4Q);82OlEO
z+wg0b=nZ(`(+FP4BxJ@V1Lrrg6i&HgOANS!Rac^6I%&ke*2V1O8<gk2Q$RLc1G*l1
zNI{>X*tFz}$Ei3eTd%xuJ{za>K|H$i2f4*Bk{$ElFlurT4ci=3GYD`CWG`+ZBxvg#
zX8!lBkI`Ea_ou@tbJ%#e3SrYFy@}z?LYIKI+8{t?%|9A%IPE!@(^CKs2y;x0w%vR&
z^?8WcP7$-Fn0ew>ZQ#mTqUS_~xArYB6ROny<$=7%pu=ck{UC@vFgk;(Y|^S-H_JHY
zjlKm)XzDI(c?s}7g!Z#zh`a)#2#WPKGzdbfpm4q%aMp`XilZs75XgxJHUkMjaz*hf
zMV+jkJ(|k~D?UNjLD?VG_mn{5ZCC3@*)3ZFlsvS_JJ@k1`j9z}F^}a2b?&c)DqjtA
z^-?@@uZ-qhVwbM(H>||p=ow@G%TK(5aL(=IC%zqtpi;?44#9(dH>A@Kij<k9loj@D
zGY6DX-l^uhv6fL}TzeH4yvuT&fR^=(Iw0qsb$SVN0z$|CshR2!`=gnHXHhVt4u8+X
z?vG<1YWBnB@%HDs7f&2Y$u_RJ>_Hy<o7KXZOyG5Hg<R`*dRgJbv`;?^i9BxoVEt&9
z*KS3lm-Xn>Ouu}IxDE3>4(2<gF(Sf#Ez+8jLi99jl|G7BsHR6#7m=;^3-3`kL322n
z{o1x`WC)K@g@w5S2bmUP^fNEQ%?g5*X(f0p!y5PfEAiJYxf<FT^|!nOpMDUL67mSu
z90@GWI0-$5cK%En*u2G<*^JLI5-UghM2=|$DT_+XO@!VvE4I(|3x<@u2v9Te_~T$Y
zt<M2kB{B8P@-$xeai~RI9sj_r&TAD56dDS2zAm%gD1c&h<d2;6G_?&XF~jlrP=Rq&
zuA-SCQz{<-+2a^d>HcpaR+{v<s%R5<i+TGuP!yYAe+#Mgyr`YZMxb3>ewH}=*`6rx
z<ZjLR)8h{%xI0rb<5(=#EHpGgc{FEbDrsaK>&5?c2X~2~po`@U*+F@7beD4Ug|$$l
z=tO(;oM+h9Wyqc?8wFD}vN25h$E{gPx%9MR4_WN0M+T@pg|#V+H49Lv_4chBh0dc5
zJK5?*{tRpZ9{;Nr*j+MpBL+b)4o?wq9+%O9FKuiCWQ-4(&jljy9!ahTn|qzX30j4q
zY*78sKF?D+T_?G1!;W(Lqm&2ud$|VIiZs}>khxT`P$koHdQLqW3eq0K3L&P}1kX8j
z)r9oBgOk0yy}irT{CV<uk*@QlDj;EN_slqyoz-IB=BJOH1kO{*ZrS0L@eW=eegtMc
z-f5Cv3GOVMPn5TGSIkV~+WtShy=6d@Th~6W2+|=SAf3{w64E_@k`huwh^WMn5<^Ir
zGzdz|ASlRyNOzZnG!jF1ck|zPp7XrtJkNRG5C0FpFFK5P?|bjP_FC6kdtGbk=uK_=
z@8el^+nz@EM7`9nME*AKL<iW&T1=KCwTFw_#6OTLGXn9P56L`NoM>~LUDlUxg*lsU
zo_ae{P1C$tEF=fap)@L6o*p)6TYulq-JszHO5gKRXK(Yf2@TvDw~I*05I%K2y<HN|
z{4e|cxoZa(Jykq29!}2oYeq&ts`<k!HJ{+2hEyvLlZDso#M3AB?s?tkT6{hrQd@mV
zGVicEP*t0egq4bF8}xB}UV1KN$qIctMyEq`$h?LvTnkO5+4-gY6$wh2FeCh3S6;gr
zn!#{MsNhT+dWnY~ge~G{qav5ZV~<}=8lSG*pK`7(55uDA{^i3r+h*!R{cVY9+<PpN
zFi&fV>aKyBGJ;uxFzEWmE*y}xS2|Z))NVx1`bm0rwyu-R-aI00a&0YByiB9<Ug#ZY
zJI643tAEJ<+Fq!sI(0L4IJ*i3B1GlwMrruO%(gTukJfCp2A=IIoeU1qP@vvCz^q6;
z_N|~J>T%p-9THtTFL8P=gJ4dH|MvyvHl{L734R7O`)m8xbvQQ+Vw|-4(psnSlT6(*
z7DlUb!<gJ(L~yLx7!ydD0fM!M>uxOYRuWh-2i7BCy{-zZn(1rp&m6vgtLe&!8=rB$
zGc5F^R!~aG$=Qoy<#9pTcZDz&P9G!u0P+})6@ka0Pv1rk?u$#@KUO;ZDRMNMzD}0K
z)le&I`lb_&r^!}YBC06U+v7I1dfYTU=DcZ?6a(F-rlJRjABQp>?oNwj9YyK~aS%FO
z4`r2<j_C$;D|7!WM(K3BfQD;Lsl?Qw_8PfEii>Hd4KFeYs`^2kz<~rct~a^mj~J}7
zkZDtkVY{CFvgVhbth9PQY)wnTr2B^A82r61s>uj67wo$_K@={wr{dXhUtEO9AGzn*
zn2IOUapps;4hbt!=j40sKEC4HzQa5AjB;JNhF|h|BVXE>qhhyOH`}nY(-`&>8q!Uj
z&3jHqvuP(#hE;*C+l*=&x)RW-hYoEPb4<dy!X3R%hAOeC)z<6bf-UdE&mXi?hQbQu
z5%h9Iq~gYw5&?tz^Xybq2!@wL?DmUQ{YW@3V)<xz=ErLn#|7!Cnp^B6x*1D9v)o)B
z&j&OY?oyC;3iBcu`lS%eqf|DN@i8+=POwKOUShCVNXWfK-J#b<`KuAkw+?JL=s1Gk
zjIHh#%(T>UGGYW?>%BFQP_1<L?6tjDFnU<3^VkR#=b-$nydz{r8PX`#D=Q-Cqamv%
z$kBdTSlDssP&(yt4e1fDC5XiaWn^J?@DJZDu*Y;K{jP*yUXEo(y|KI<?aMIiy%I4o
z4TJGuwGTP_6wYrCOO7sSCLk8?9+oz-NbkAN44L{tlzt&s#^d)0s6T04Gv<R-*)QkL
zL~sMl$tZNZc;yz|V9-4C`EVB$zL`&Y-BkFFy{Ht_H#bP+uSNb8dZ5=nJG0+gW!mXA
zpwFc{Dsaaa|JLLKFC5}HrJ4Mi@89ZaF^Tlr8Y=K{plop5W&v=u(~4`#mmZs0D^@MD
zcW7RZ<P{0h(Ce6ab0S7vz{1Vth^a1EnMWB9Ix>oyT#I}+ZPG@rKuttWM#Y;-+LOn%
zQ)G?KfZplaV$;iND7O7reMi>Eh)Y-4TQ%(Ksqd}{Gwy9tZ|n5+GJkTF2xk_)my;6(
z4#rQ_>-vI(^wPwnD!v7`4tu0>M{gZI;%OZsq(G|D6(5<bvw(1kdd(COUjbV{1L3>J
z6jWO|ZxV30c!0eSES3xPT6!pXE|b#gii3HMhJ$-o58ag7jK@45%@M0*qn%;sO(|{y
zCF+PKRN+S<psMf|EB7YQKao1iInLp{Z!_TIGRmyE^_2p%gObKdE<^p9ig1>CqsM-S
z;;!Iia``ts=<`mF;1I=7c$KQd)yYVBCzMmMl-F3W=V?8mdAS@?dPmP?RuhATGT1pu
z;kqp~z;{Z+4rB1l#Rjrzy1}xB<Sz1#Dq6s7?$#Hk*rLTyan%bpACCylqzHI0YQCBp
zuNdJnT-2XEkKrp2Ivj8X&*{R8*KBGCj=0{aJP<Q4xLFva<T+Y+!D=>pjTcFwhp=5g
z+nSl)`*F&6@cc)WW>$v+=?Jd|!Hhh@BoZY>Gs;CGQNXt0OnW>w0{@Hy)3H=5{JKWD
znSXSe`?Z?CfoO;O73|6aJ!ebWw>;rf$T*wx8my|_kL=?Q80KrS+{6xn((K^V8W!XS
z$`w<FhdIXd8Y?Y`iD{I)<{wVpY_7O&b+|ZvZ+bkiTI0^Aw?TAQHE~Pc2sb;9K>1qf
z6JhFc4JsBEHc36jrZ;iz`S;+e?9r2*Ow9|EiF5gJq^$_JdEX&ioKEPw?zjjtZoI6A
z*6SjYj+ejh$v9Nd#e2@X$OiH;IBsz|YO|c4FG(n20KT&GA)r|0&Ud#vQyGszA-yVd
zn_W@C3Mm7-TB(6+9nWZZ?~FH!w9ZC}t4?L5hp8fRVD>|)qC+%O5+SE8Ry3I2f%@gG
z`;bC=+7gkW*@$g~XUrZIX|%6<dTUwb!`Y(AzLka5bDR^ON>Ea7mPwMYiC15NI3@nn
zU?zs%UEF?kN(9Y?f2o5HRw5z)&|qXqw;1zo`Hh5*2l^9s-8$elN56EqYQv<aVT|S$
zZ93tL(v1L!(!=)>b-MsgzW#W{nGzX#+C78K!b~cme%CO$ru-e!dTl2VEdWi1fPI-T
z)j0Mi)oXc#S;x9CL9{P9xVrzQISz}h7n0eYX0_+VRQJfMwU+V99~vnWip~*DvcUM!
zHMp#IZ|`t6J;;wbZpgOb1{M9}a!HO)v2`uV60KFG48KxxYdT}FB5QTmp3{Zxn_x+b
zo#n`Qw`$Edhc@mPoDB&KVmeXGT0d$AshzE->Tnq`8g$h2d$H@TVFrP_`-G_2M#)yg
z;(Zl@p}3;Vo`Kb3MF|&A$Fzx72H<~E;ZA&_`(gSOZ<V~K=6#PS4l0f-is`XjWJ7Ke
z+~3ZSNK$3Bz(uM&Z2KA-cBp#PjSY2|llgiY0%=xhg3IH0lczVODP(RRs5aXxUT)C;
zFuZ#&rl(nf#x?sfV>9gS3i-9#2aVw(Rc%w)*WJ7sSuQ+^ivu}iud^-pL`uR5l=A9F
z9|`WEQ46M?dzBUj-Zko+c;~2e*ixpKvp68}#uML>_!%Z?rfq$qi-DV2EvGh_xUAHd
z#rTF{m}g&z39>jyiI~l{NaQY8p<ubZu3SUd&j~Lgb}XTsCaV;~v9qQ*(KhM>vpp=r
zKE2WO0DV7R-<(DrVc&F#w}9_4LKPdqE|%7E@xck-*_*suc~umpwP9h+fcJGQ9n-4s
zU_<RupY-UB$Y;ryPW4q!Q6~{0{>J2~THd2?_zDO>Vp6gQlH{i_r}<?+hJRbj!Qmsw
z%n!K3H70iLnQ%yXZ6#uivvtS#kg~6bti!Qap{<XNW(OpBc2)Cl7_s9>cv-}5gbV(f
z%4dp*5ou2jA||oAKDW1(RtOIn7HSwVakEMC-l@oCu9pXQ6E{w82Q^c^7&_Cd6qMwD
zKU0bv9ib8`kf^C@T&)QtGU8mVJo{{{Kz`!oE3oy;N7|2v&qKBb65L)cMgS-dV#x?K
zo457@(cjf#yg;$MK1aB(RFKAPIYJHW>T5$3l75)g;u<}VH(((uzDxU{-XI<9lLDpS
ziS{A-L0^C3xX<9P8=iIB36d>U!K6CE_YwnVaj6^u1tG9^?}7(}?oc3$;%1yC554X3
ze!%S0H(}VMQAA~eC+zcGlSE&i3_MoSCCf-*#q^GNx<K!NHPbHBAVb_-O`XF$?q#FS
z#>P?zol!kMcU?G}E6P5#zyCy0WtJ$Uk@;8Xg>#q}Jwd3q&96YETkIz;)Y5n6zlq75
zX7n~pr?j@LvqTaaIax$Q$6r!6+0pyaUvTWCEPbT=*vifNSR>g(gpE56Q*E5+Sq0;b
zp>$(rYMum9xa7TQEH0nV#Nu2D60eVabsU1_f81dKL6>jy(X==~hI3GY(KDq(w(A#d
zI^RF>e~sn1&iVjqlb4}h#)MAV)YG{p{yb*=!2q&L+NylLnxEbFm0?CIy^P2P+@z3D
zjWAf3LM;X2bNxCV_PB49@>Zy0Lm0zmG_4nU(^?O~&zV*Jr0Jr-*jkHZyOxs>=Yf~A
z{P1o!Y|@RfwB&Jn)587PnRzdsmC??Y$rqXHFV~Pt!@bd)eWbzkq*A?ZMhnyOm>&sz
z9rMWR15r-$>EXLi_2@u?<;)MatUXME?4a32_kj>QuB}HtZ(QA(Oi&;Vm%FzIZpoSk
zHJ@S<Yu{NB#J{<vf60wUY*ky!8f^xLy@l(3yJ-xOH$9{tTMw2+Fu&hj>H&<o=b7j(
zIg(Mg%MK;PofY41MQx&`a+#KM&6s4Sx#c5OMZBX8oU(Gp3L0A`b7JYG-Fyc;ub+x7
zRMM(GMfz<nLoc`RWN$Vqv873q`NZMGfDGsGUAts2JomfZ6s12|m1<?_Tw0;tmNTAC
zmv#>69E`>(H?(Wg(hn8Ih00gIU6sFN;)piOg*r2`Y|!Z4f^I2@h*Jx_QGV!El}gUY
zg8u}StqQ_K_8{Vh$L0#5rygpv$TXZjn)d4OD*7-RA6mh_+`(9el3AW&;UJXKGpi9!
zxUndO+j(f6v!hyywWF+$tVLnlJag2ttBNw=54%wX<2p%rwr!i`R7J~jASK1+S{udr
zX<vbcisO!XR6N6pE&LMth}NC`i?c;^S`w00^knBOO15X|<KasaVs$P0yH8Xlx0<${
zq>7wz(?LC(6Zg{3D(I0K%nsfn(;UQ26WJ?$YZ6BsJBp<oMH-krOkwQ{8C<?QKUALl
zI7t^caa4+b6Pt(>99AFnVjafds`VoY%H~fQ@G`f;r-7Lt>D$6#k3##YqIVtvO-{Ef
z7p7H9`++{Ti^wj>HDIAHgw*w`o15Dbe|?Hzjhop_vj1+3)@51zmsgj+%gH3$eYA~?
zYd*mVnrL)*4VYsx-S;Im&{yCARH46><LqAf>HGT!!+@|dsjyfz@2^xXEN@$3&&zf8
zNz9-z?)p};FXr7Q$-Q9wXDJT><Mro)7#=3y1%tFseZ`LR0AwW`^0T_v`P*%>-Hv`L
z2Jfh%w#`S@gk3*B<dLw$CADn3B|#yUYW{8`c5Z6LReTgx{|iNG7RNpj!bsAnyBp#%
z`$eLWpENJB?FoN|ka|b%<E?uwx`xcA`k%%SQ^7M+TQv^F><&v+P0GVjk8U@)2hY&Y
zP;EwlggY_9G3r(Oma=dOr4#&z(H|8Bq!NWFN@u8F?;VHZY56lXWI~f&gI0BHLfvg0
zyu<K`n-pk3q}Sm5gz^_>J&j(9MuE2<<hoCC3AjNMY*-%T1!1ld8HE3h=69%5;eoft
zNAd9q*i>{E?_RKcv{kP7#s(?R$QwIrc=}nVrJQ;xp@<C{hC!F`+-UERN#dYu@CBnw
z?1?mvRJ32>I>c_dDf%E{;v9t8sXl8)W`v<<9K}Alo0-_P<3odn#B`NB3UN@Af3%Yc
z2<oK$@$y)$KGbBkca&6J?VHL)$3OD$9l1nZOElOj0dBf+qZUiE$ok@~4Sa?$dV6tA
z(n6o!hh^5sA+$rXrJCEU(yxU}LzD8fY=&CDg+J`IO#&(84Y{))d2&n>KU>BC)y5CC
z$<J_K5;O@M8A>eOVswopQW6~QW&W+(mQQv*HHE?GuVvjj5Hvg~(4E1XWpNQoaB{h~
zVwAIB74_}OVWK0Pe?@^vIyIoexB08Oraz37%g+HO|Cu$<dY~u)fY&^uyR~DJP=K<(
zbh)o>6!0+s600dJo6Q04+#)CY!ORa5@`@A@jPIcvwMa(#4Y;LAl9wPiN*5XWGzD_p
zalA_J-f%eUSE`<Wayo7~bHA`y6#*_<C+)mK9g^VY$(1pWxnkGC1;;X0LNlMKK3q(o
zTz&GD_Z|b584((xwX2!z)XVxwV%|VVM;#MBZ|g8M`&!p>xBWq2nZTqINsHh?%TIVC
z9k1I!3*3P4^eu!sxGh7jw&!?WS@!8TQWq-YA0ST&N@D$|e4(kCTNnmxea6bf@w-FO
z&GwlWsbOOu;COT#fT?4VnVK|?GP|l@s`%uJOz}mWubqa1q4k!#;a9gj=*v>Q;hDwJ
zB&SE%(J`d;?c!IWc5o$XSv-UrkL8}W1I`Bl^1;)PoTQk1F0|!qJM}80(ZtU~NERK^
zVSO4klPdg@G1ikzJWhJ&1s~*%iEisN=T)8d<+L6s61%x>F~=*hJ$KISDxfGSkZUrO
z5_354>Xp+as~5$>xn8=239F6b-)i#_3RGsyTd6z$bU9#w_U1MtOu8nv!S>wkfsls%
zI5m4m$3w0fQDVv2^qW6tC2M2o^Lrnz5DO_Fs+v@W8cRlVZmfS(LZGuGX`=#rov5GF
z%l%uh)bk(c2#S=a8%*_~+^SgFe)ThITYvq-+cF<1!#y<$=bU5f-z^Lf&-NO2WDO9X
zwLp(@s`;WKH2F0M!=W@w-(aS=-YR9__Jt`b-HWio<in7u=9PIIntah$QXo)r2ox3D
zuUuBujn}t6?x;*%)Erim#PEfp-ysAtn3IRk$g+|=tkg-*OZ$DS(5|O%Nz9E&bI*12
z<muy5#^Kw!(91w#J;G&6-1cqv$s|Z&xll_v=ctST2tDS~02d<ehtgsd&Tf!Z<1~3|
zqrsgUZw#;7=ZNt*MJmtfG~5~gRF$IDir=?)m{Xv?v6~lOtr767;ndfmZYkB;R=|Bu
zVv+Ib-ezwj`pLxUlj9r!19uC$BqHv$*z76w!6;ZXi5fbeq6Q7^6`>l;KBC{h>xzB^
zg5x}LRW<w;phONL@50l8;21v#w{dA0PSQLLAl3-ttBHDrEFhCrX6JnMs+gMNZ_X7r
z3;q@7YNRn(I=Mt{BF>sKY{rma?=dBreZku>PWwsZBnIx+t`xe_VmdPNe-oNu%!TZN
zRBthYf%LHQoec$(mP{&#K&}sMopF$n=ja(vjdMBf%3lbv6ns$hB6K1CRc79P;K*@Z
zU<D+-18+X44H+ztpguo$zxLa3-=Ir<!XNg~nuc*d0`dhN(0O8<@U`{ka4~%4T?_Rx
zJZ?FzufGWoBjM`mE718?82C%C(`GX~nzeHE`h)x#*NT-4;R01bz?<sC62Ui)PX)ZG
zYQ1due5LY3D99_1G}K#h+;^~Tc*-8u7p#%?-TPx|HK<opSY3P9NOm8w6`VoH^tOJA
zJw0ra^wPbi0pa;-)(wF70TH?&$eqVeIOTA}+|HU>uH4(0XNFg9K^1CqbV+$(&e2$i
zLTgSUE<*UQP6ds@Azs4=r`gZ{hay(4)qBqG@>hYWSvD6~eov0u>GL-tfc*4zw;-IK
z&hstF=u#}g24EBFdXSJa?$~ueLvgc7IH7@$?50iG4R#-640gQnjN!)UZ5~jN)^P6b
zWiq%;@dP*f!s;Ncv9gNuJJ>HLl{AF07}%^|62wJ2Yj5ld_lS)cHtuRFH1xSLLxqyP
zn40fqBnI@|cX+|eOg1nM-@f&dBg~G0E5YiCRo=IK4(7IaF@giZw4aTW^wtD<Cz;C0
zy~Y*uIlnF5k*x*8rhFZ{;a)PXv9Xv8LKre?C|ljaOqvv67Vt`i^9k3xa)>@dtaV<7
z;LuK_9Kw42yvd5WjYMz6{pnjK|M`QO*mQAxMvEuqr(`^B;**bL;BIb`3Pvh6327iJ
zCtNY+`~iD3TBPt=0?`B$V@^+$C815kO0_ARx@<KGJ(`&J`|wrRr277)`7QQx>xdFb
z5i5hJ#cG@h(gio<q5XI@f4dP&dtz~+&|SA$R1r9bf(um@(CLj)kaa3f^bCEcaeeKw
zp@lKnD}4y@{VgL?(7_60Pzp%|-bWcY%P5t309Az;!<?=*G3$5T#9_lHk#XO98tq%X
zllg5k)n-a`#aZQ)g?)cp{|_xRdY;@n!#rF(KO0n>(Bx1eSHv$ND`90A3%J(x9de2X
zcaa5>{*XJEINJ#+9(M{NdU{U|2(51ss*#k9%9uRHUR5h=pEdb}uEW*F5M4sgb$fix
zkJx|Jrr;}=OfVMpBZg1*&wR3CttA@gnwv+ZtrKGmRgJlAF>kQ)<VWV92OJ%9nvGUG
zihW&kN^Ws)vVC~gBT>_rMNN>-#yZE&wj`#D6?^@*KUH1_2)EE1zU$ZAXjv9IyjO5A
z^6T{dFkb)ha=tlm9@C@z(eEZfQS%f{HSG)8^xqSdArx_rrl3O8uvQ!V7Q4INclx*I
zEltb|xm^$LzsNIR%ePr`-08Vbg3D?oYm+x$a@c~ff*3}UPOKa(?8JRI98P`wbCwx}
zGIo%0HHvXZ%<anm;!_f=t#};&rKlyGB?+p};&1xqXq{EbKH}74ph))v;MLaFua*m+
zTE#qg!aDwO?!Cd-NfR?~RbB*IH%zdQ|FadbT&Su7>tn&oTQYD4a#9F!FjQE`vpP}+
zk;6~q414RSD{u#9YGJ(=Zn_z~3#Vo&3Kxlts-F*#cb9XW8ngN}a=bB!3xZZb9k>r3
zo}E$ouqi@vK_R+|;;7TUaE-ZVyM%{MZl73^N(+QmT8sh9Y_-$>On^wVec=-4k=<9$
z&wju51<BW!NoU?#)LPo*KQ*3s-OY>XP3h(|1`aYRiYJ=*G4J%nJB`fn(lc(2no$KB
zcZgXm|EpYRKeHxjAjjsti_0frcDFFhU7j*%#-t04T7ui#CTMV+NqSsgz2{56n9q9o
zQ3<%I40|D=?DDyy&$FOeCGqkV+E!70Efcg^{&i)I*Ka6A%tDA6<;#0-KHXa)fAW2E
zD-GHjlY+Dr&YzyrRVDNz48Dtc@lE5y0Ex48jtct9LFaM?T|reGY%_~XlL1;WA(%pQ
zPJ)BU#r<XfCTpz!x0~<CHW@h5@FoMqFbh({8Qm|g`4ExK{Fw_-HY#y{vxPRO&ji!|
zvEZqh&qZJmK8%>kIsGSO=QEGF*Q^z&x7v`rOXYXhv5{FKkKm(Mow!J_g!kiL&-xZM
z7IVKoxSwa^E;cPJr=@p(Bo6Wl%(+Xeg+6$=sJ}0hQ|1v;8#|4#myK$j-1$YIP4>#L
zo0C^ZlMw??V2EYvm*cH+7%C+r<vb0m1CA6yLUpov{QGg+ZQKvsk=6XmWsSQUhr?x7
zY<cg80dcq|M)fqBNA=d1{xw{uo#!7K${+bcuA6lb-6Y%34)S)N$wAJ86tK?|uVXq$
zJh>uCJwWje*B29ggU7BTJSlZyvWO`fjf0jl7nY*YDQE}K``_`6ds;z-q#tg#4~UJ~
zK&81wZ1rQI_`_~+18Yj#XwodFM2Sh40x1u4`7u#e@8^YbSiD<UD712DU}C>cGO4<b
zS%&oc-Xg9X0usinw~kgfQ>`Mu+z_FVdV&#R%qp{C>J``yPYG)Lm2sKBdEAS?=RWQB
z+ImkZPS8F6F}Tif@3Ak1F=x{{DXLYsrCDVXRHr?98*H<m+wOU9g=Pdlee!ZK8#<~k
z_8GV%f5C|8H?jzRWEO^*Vr=6=(Wc49S4;4Jmgqi;4A_W&J8^z@tb<X{V%_6Jh)y`u
zRedRB@eLK3Ra46Wu|IcWQYT!02-ND>i()iR!XWJzPQHz!`hqGr0ZlJ*J92IX+zpD=
zV$7mbkrG>Aw*P9OUBtLv?&Po)RA^?pR#!MeuXiN-sO9Elh>L7hjN7XH4~+$MMD(i3
z3)!dsviIZcj6+tojoBtjT)dyn)IIyLJiqN>xEUo}MApuED;0NS8tpnv5Q%QPWn53`
zcjG<JR>kPAAYGNf6HnH&H}F)p0M=ATx~1PY{8D2L5aDyXRiWWNVYvwAzA6b;B7co%
zX1m2tJ{5PVjSm3n5L?EX4~6N;4t#Cs^zjgL3fKwr)dFP@K(CDA@q@`k(xfsuoT2WC
zbEvE59uwb_KBGYeUU@S!M)}S#|HHZuWQ(juWMotWuA`Y}3sv7_S~k^#W%yTE^cJo5
zZ^p`OPUBPWuUVMG9zFcDRkeP88<6y63$sma>S;rYo7=<8Y^UBFR}`O;q}5_JUN@rl
zKG1$LO`{$obLV}@qfm$jZQ79p(oV;5*X>u+Q*_tx0E=^1xsIfn*KIBOCnIP#>(COV
zs)i)T;O66%ntj`^z(BulA{S?g*nzcEm73y;;GbeUNW)&VBZ@h4$$J@<&b+2%2i)Q{
zT~<Z!NJ>`*oR$UFhS6Z`w@B}|ciw>Dp!73xoNm!6o0%c=I$2DG*F>({Z=UtZvx;wR
z8q=C=y@MK|=3A#2a5Fni+U;^)uH90fnen2%TjGRcx*4)l>6QOLqa?Q?_TYbn)YhR?
zt##lr&j+v4@P`Gduuaiv`Nyp};a<h|m@OW1-8yqO4`QO*i1^z%T&iUdi4on)oFK{7
zGz}}77%%Q&!IClgiWC{?E{f=4q}jx?qWEw&>|^o2s6Nedm<Gs08_EvGi51!)Hob}6
ztoi0|L^#iSn%yn@UKQUYZ<PM#<}JU)_a+F7;Cmgpj=AvC$<wQZnJq4^DSN#}(M?<4
zoH4g&Ra5oXRzX#UP^)<V)wQL2&XtkoX_-*7>EeNk%9#p2!Kfd!8TLdjJHWSuDAO?)
zH{brki?qKLr})w2U_`^!Z1*1D*xQGKTQ!4S;=YHs(W$rzsv##hjJV_i&gD{X4NLpz
zdPdaA>ep#E4uyLqB}#IJJ?C)c?#9mEkQ7tdvLTAnnRsnS?BZW@gW*%>^deL-?~9q5
zlpp54A5+WFY$FZB#*W>VfMBRaOPA6s5!=tjdQgU_J48P$-)a%*sZ-AggoY8IH}zvz
zIesrRy>c8eh4^Bp`Q9XsMJ=4;JJWvHOJ)G1rYE0ywuN%|4*&tam&{pM{?Y`0^|f7_
z)7|j`l%;%1+9QXWFID$m7$iS<!cI_7Bh@}(!^Geg*~{*DEXH6Nz7ol1>Vo$vVDP~}
zi+*a$Bn2ks{v&I~fTagDqR83xAaaL{R;!ys!^0G9jnkd-TcEUg?c4jqH>1%w*MQ^T
zcG=`1!}G3tR(#U}l#L}aGwr3@qQ&0+ush<HVb>aP+t(fkjb>WA(50t~T=?np&xZ4t
zhr6#;`pAuY6egfHz)`QMKLD&~QfK@@-c|ZhJL8>LOZS8Osz3Rgj~d*v`fYy}NT!23
zMJwI%Cs{I|f6QWZw0*bGl*3`YjXIA$)U`Gu?6PvNyLd>wQ~tT+`w8V2*W`gSSAZt^
z8nEUMPJB06b{%&;-Zkq>zvnWaaju1!Uj;dp1)hV!l_ob2t*IF?oLN$L`R*U>;WF?t
zH_;gDe>FeCU$mxpo5kQBLF(#{&!+(6$+;Yu7kTT*GQ{0>zJ42?P|@5n8SL_=z$XXJ
zS9OD5vLJirqQT`^+S09u(Y}k)nlsjfeHcA)9q|dr@8(4lk_Pq~OF$0XAr%Dln>qBH
z$M&lv*~qe0a;Ghs=&7?gZRR9*_ZiF8ilE&3LgMB{8)oMt$h;14NA%rj;{9>ZHHm9X
zSj%k959``{v`6*|Z%o1u?{#Q91E1S@S>bJWVy!1lW>4eweWU4fo4s6o#%2-LIY8L|
zsG2>$gqM~P+2)vNlvhZ!q?2x5$nba8<Av1n_R_xaL`N3|j%6uv1JP^F<}YEL-@mUl
z@j8cz=DUIuX{cj*o<;C!Ll9&`HSYHMby*h{st*1VSd7LZM)oqXrTIhZcMCw}^SXes
z({<sz>Any#_Ndh`5}Dvzyc#%*>6=@kM5{ezU+2g7>lv|vo{)qH2&&WZ%&J;X#lZI4
zQ~-yEPp0>CgJ7Jz@=ua8&mJd7=Wfh3zIq?P^J^V8`;?*AuoAu_Dw9S(nyh)#U3}GY
zJF+n3#+b~YQp+j}y<WGPL?q4lZ)>qdNw=|Sg7I=ULPv<Yv-pr%(<H;+ezQ0bgC$p9
zlvU&^0;T+vRPfy>@hww2Jgo783STd^Z&%k({lNJ%iS~y`dzAv(luRnIWO=2?V&@Bs
zjBe)fpTj7{&|mbL>I`qOhK5|+Qp+$xIoSA3gRl8~G?-88?__+>HH5KtDDCUGwtant
z@6a03r>Fs{HNwqfN^0`!c>{6ko!@UFzoh0^<?PJeJN`r;rZyG2B`*!<oc!s?VkPJX
zOBMjdYI=2&n|*R|;ui<KHj^3XA|dv0^;JEo=`c|PKn7O&M9;`#jVA|O&&l|PUqlD+
zJ`h{G5UvH8p<DR~$5e4~V+f9ak&GTE3zlRU-}k682Gpai-o{NS>@Xs8;|95&<BZL#
zdj+h<P>`}0bhJxMM4@XLq|a7Ys+rlo(1YzlgY=|?-)lhB7c|P*2?k))>KbK6OHaO1
zai<l0sXHLhmN}+p&nR1laaaN}dc?g|`dBr_mUsYv#n(k%M99q(w<=Oc$PP+Hhy0W(
zDYzol6uh6Z>DlrkcAwc6BuJD5lirpYgXPub+)6bm>F(Q9Ks>u=ofA-)(Gnz~dUlcO
z)a@|`I?nK~8C@kyXmWiMEy@1IQ&A~lsnV&0L-4NPn+(<s&kR>WMhsMVBk3ezp53*+
zu_;9aTUGCK>KZ_Xf<e^Fu2~h#0jZRnENb>iU}n)&XO(dod`EePdPHX9!sSRLNl6P)
zuc~E<t6^=LJs_E?LuoqHV8quyL`xMcvzs<}<P@!rGjPRB`tt3YO6xsPRqFMJ2>f+n
zW!wj@54m<+WrA<)VUADS(wQ%OK$$Av+7S2+>xeuQ_sJKo0!t#nGi#pCrpF-DpcS2m
z)+?j7>f|P-&{T+5b94J1AIwp;kr=b=i@LM=_`}&M75Z%24?@Yp!)jAcwT$v=-}X~`
zcWTvknvUqZ2&LLq2ttUcoR?7kn?c1iaX4EF%ZC-{gsg_of1)54*lucobtem7QL&2S
zeTNnLf(5j&>UjLRhPbC6KqJ0ZLs?(7)J<MggfpSBsrp?LkD3JLr?Rnm^K>B)X{c$>
zn?lT?XRbEFI#~rHa5E(5nxebFz$~&&`+yAgi!9o=#DsA6Jg=;7lAXdpJ#ZXl_)XEs
z978#IijQN2T<WAQHcOpNQz5+nf@I&FhBj0E;jvrYrw3k-L6H)WGzWu#g2(HjpmY(V
zrt5iE*DV=YMkqteLsDw>i|MM~)#A_Rc4{_kTaO5!*wU%Y(H{>!zVd7qv7B0i2>Ts;
z?srRHSjV5`=earw1#Lgg%;ULN?RoDIoDO(-OgUp(rBM2L;q2Z&Xkm8AVwSzv4k9_#
zUOfQW#&o+;JTQg%m$k!<Lae{Uh5JQ{MqY2Q@vXbgk0wu_7M9dqF8t`@M!33UrIyC@
zn(o&JM;jj+CvG2d=TU)HTeY=2+Bqw79s9@p;w1)8T8*KfH3h>&gr=(RFgRu|6ir37
zc3M%hTz6)#E(e4xJ7ou2g2sp4n|fi=IdxL}34YQcve(dWu^)-srZv6Hym<5r)LFD}
zx7|4<qp+o?eqEzZ%l|Q0z|v+Qqrt&EvR-7)=B+ykyVB_(u(2Ti!P<s6K4$E3`A${A
zF}H31mNG*N@-SvH)mw;hKI;acK@DFeyPEyk80D{w1$urojzG!obj-`Mm5dpsGn~&?
zYzc{LXjn>UjUV{rS1^yB9QzYSNcH*m_M=56L0%ru@g==GP!<Igj;Zsjz6Iy}rTYrt
zRxO$o1Qq|YA5Rp852!rC5pT7&?drbgGIN4J1GJ2zM65Nh6LNUo>HN}ZB9t4tf9vW=
zOb1^)Vgd2O6(kOW7A3ggp?_mn{YhMrVnf4vxfD-9M%Yl{VZMk-AR73rqK)A$<L_2L
zR6fyBD%Z<#FhrzH$gX%BWbR78*uXytf*+c1hQ@xL1#mf7n>;2;6uSvz14^r<-u_#&
zp59w}W8<?&3vI|j8%c+8XCl_%9z&>pHfVo9!VPYwYvEob%j{Rg6A5&Oa_ZKkAo@9x
znKC9M#!%?uBEw>_l$gscX%ihVCkq|8>aQzSmj?KOg~P00>jDhUZq6a6+;L51|IZm+
zF={+yvHV$X)8XB8vOfush5tcadc)jsG|M8CzzQ-lZ$$ylhXP=fyx)*`S;dN0a1arj
zA-3VzhY#p(pch1+*6cdiga7E#^RG;4e@4*x<p1I+Q3nn&VO(wMm{i?k#FQ7#$Nh_q
z7?58JyMcb(K^P83psRq-&w^i(`h(U&&>j*OC;RF#?sve@Qc;6kiNtTN&HwsWqBtPs
zy|0$g;#O4!J%vgYUx6ya(Ni%Ee_e<Fg#twRyA6@uf43p}Uw;4hBj5g(Bm$iOO_KhB
z$po4yftkL*18E8)AXoHX?YI8#c@cqW9ZQ)@0isLyKb?X<XoP>G6oHZDIxs13T%`?t
zx*`?*PbaMZV=M+uiC&7UCAbHS-d~wu|HlIUzolkX-b)K%yEoZ*Mb*-vo7GaDryL|p
zcu}K{uogVjZ`_~9F5ZBd4X{NfGX9HWWVyW?%RBH=GEs+R{+}ht=gF-%%Gh66Zv7Uo
zi`r_@pr6bychP7E4?R&6j-Emk5DI#!)oXnXW+o<#$%8mw1FjQ8Uf6sm#^Ni`c@f~p
z$Iqvyhi#|xrcSECwDvsWJu(}9vax)ccSF8gPR+^b+4Cw`wRXJrzRG89ZJ7LXg$qVY
zmrs`{61B%uX_p(D{+_1?0oCxs*U<jM<;A1ovm(o$RHDnKp8mt_)xN&I%Cy<T`83zd
zy{9J<UF=@Qr;}gyaj`ThQ3TS;OarBj5or&pRMW~SjH^9vZA_lbOtcS69BYgexj@U{
ztM0D*JEPX*6CZ!Js-|LFb{l`0G>ZDR9mWV5bv1VZl|8OqM?=RV`uB$s*Xz<9EmW`p
z8g%yXZ@zH~Ik2FV$|DV7hdo!<v;FD~HL-(DNg^KaJ->nyvW9TzxxJo&QOf#91HOl!
zigYE9?GrX?Ja-?8xx2Dfd7ds9{p{Q6*eP+1;7&K#;;vO`C!~=e<}tiLRMv#XVce^v
z{`>IYJ?a9nu`~tWzHa2@g1lrVL)@3~6kXHDqdVB}+~lo@q?NqgomxvgS><)x7Iijn
zO+NndXAl7mHhI3suZi=E6yC<LMYYlwcWsl<31}rKd2lqVi2gdkKV~}iNGj@E7a3Pl
zK7E$`I+&_@;m%xLz1{Jz!%8FR+OwJLO%B0s!!qIs8qwRS5^fYVHQtryz6QcOONFu1
zYXP|vc(xNq^1AUeBl=*$hE6I)|9c`HXufgB<Dcx_fiZ4hjltk=-l`UKZ;>}h2nh`w
z6O#0HnTe0|B05-?EcnVHYPLmkBb^6wyt6n_VjS`HHW9-5Yi%XTzxU?P4?kGxA@6%^
za4fMkDRrfz`mk=x2*an9h27VlfHoA;u6t9eE@swzbW~lfvkS-z=Da+vQm9=-bPypO
zuA*8p4kjV(atVcjfdM`YRfWx)Dv8~bDu#FLzK|ekzhWrtydbknBlKnTtdJzT@x5b0
zW5<ee*>SU4<He1jhW7r&hI(1W##C_?)8qBAt(oTjt-aM#Q)ip1;UCjXtPO>;tc^Z?
zLAATC=MSq~FNDS5C(gsA&zFMFZ6eadsLX;3VW=A0vyPpyt(|%Be}6XC?^m6)M>b4W
zpH^@n^unFa=HByIkCl3R1HoEFD`kz`4ASL=-Y-WAK|(fXLSii&rkPqBYCJJz^z?B!
zIbn^f3A`09I)fVh0u*xpDbMtLnQ!PV6bi8M$r1N?0_n5pIL_tWnC9V)8xzL@Rfcw`
zXbPz&`j4Bz+AJ=$Aki#gah|`Jo~3(pV|{h{tl7%+<;xz-T_a;Js*UlpDxtPuhQ4Fm
z;fiaB&0uk>jft8`cz9;3THw9KeE+PhP0E_Z;ZW)B^RZ#LU(}wLnA;wGlhnf2&)LDx
z;6w<CxjA|7J<wN+A})O;>R5C6bf$}LvUVcChTw09@Q;lSBPBv`X*xP0Ss*X>$Pjz7
zkJ%yZeo^zKu!dR1!%7@K3N~s{FAi!E_mA+XQkZdofzV8ed1DrZL2<F}7f6Vsk%1Au
zl?sE{@rjq~`HI^}(RU$gk=FuEZ<uJ2&}uxFniE^x;_sULhh?tw@aejG*$Kr$O4dM~
z&Q4gvB<PJZ-Pl@WRfXjO^W2uU=%|+sY7FR1YyKE+OZ)R)Hv`xDGRMP+G~s>>uH4*Q
zBf|H3S(>!V=fR7->!Tz<7foW^?(FQGmdg7s;>fc$QCTa>ee)qNRE+8@+hLNPCI7rU
zY`DS-$;BI#9ZBae1cN&FcuavKcs5s_|LZQT3#{4Xg)43NWp*RqKP*lP*wxb-EZ?}6
zViK@4%^ZG2HJ^2aJ@)i~{F((bv{a8lYebjFvmz^A`707``X_+SxwtFy@uiy|ehTr(
zZx&FzGdw>EGMZP&8z}SKO)zz~n6KW>?lD@q*pZ~=9&>oi2zeUjU3lq6o;cUKT$E5@
z^KC{WcTL3p#6#Hm#N+Vv)Wz0GTG?%Gx}kbZsG*a5GqT)izQ}H>ciNWoyQokt7HqOg
zGKfhrGRTo!a)sygY;<;OYNLyABO23&KRMV`sU}_zkC(^j)N7>35=QRcEcJXTkrfhh
zy?ZdyYXsvJ5?YJr0y|&vJLI`3Lq5ApD@F>m9*r%EI<?gurMvhS#n5|~=+esZ8Q6c^
zJMzDeK#3kuaLu>{5o5TTl-ua+HtJw%1g&bx8z;76RW7-rFMCqOr+;NjF4tVHob*(N
z28Rppa719b)Z+A{iJwdfZu1TxJ<l@p1!neh$)~$@p?InmIMT{_j3LrWZcaTr?U5Bz
zOU=94(_Vsib)vvhwi?-O&5X|4)e_qwR*fnSs`CYyZ*ue7DYaglV>Zr5)}YRh&XzN`
zrY3c@bHfrO@i7VMmK)wB)t<c6O^Km%i*i_P;jMg?z9a0mAS3LyVtBZ{PW5{cH;fY8
zeB&sDo2S@mOKZ;Dg2yTcBg7~w_UCyU#!An3M}l2O9$^2E4f~g#e4T$KBl%zd(C{|u
z2}n<*)!U1y$*!%l<=G8{tYSnUZi2+JyIqsQE|qg~rCssZ4hJpNYrC~No@bHQlazo?
zWb(Y(n_6h#r6-PbM`I;4z%pK)O=9u5lP;(+C^8^EKAxuu(h3w}Lo;XEYsr>3EH}A5
z71!<%mx_mJ%RxPlcU>AI{xDg6=VIUIb!au4OgL^nIga*?D_bA#T9wRIWwK=;%axwl
z-aj{(UXDe=7e&>g9JY5@jnY8T4C8<?A+J&R`be-upjOJ*M3NEK{GWE<FEcK{0#=BT
zv~+<Im~0p64&ecz+r*N~Coesm&n+ho^tXe@RGMu*b%X|Ob9M{P^bNvJe40*MP7i79
zSAm-EeqBqO8}(iW=q0u={Anx2n(E<jUb6>nLI&@f*E?qfpBGUB5&=1vC%kHNZ^HfE
zu$a@4UYXtT?8dBK0BLUY4ig4@;h_@Vvsa47#d~}0ja&QYW7S$|tdR19<;cpfjn74B
z#mh)i#k}}B*;|PJdj?Ba)(({*8`W3#+BeQVXT;;|EGx2crNf~tT00f->J5@^oIs6E
zlDf7xnJM1`pbyxk&(8}oez_0^X)#;W(V1Tj-SK)h5dtqjoE;JB*xi^Lo81$1Wce;L
zxA*w$vRafy`%4p{Z4-B8sMo!hCU{7*;M<Y<M^V0Ui#@6SK=8*Na-|vIp(IY$)LiKB
zZJ5ZsCDo#KwhwE{54c)0gIpS86%Q>%n)Uuc8&v~KzkccfK-sWr)DFWyh}L4m=X>Oo
z7$~}fmHFNL=|jfIa^E<gSCPV{DMX0zstZ<A3pcmtD?qN1bd%U>bk)%2!hPuu&7&i@
z_ZRp@l}pbLdR2P9$T!_NyE!63e@AI+5U|?^Mmo~Uv%j!4KhYt66crDv&u)K4XYVqA
zTWB-2s-@y*_jnmN-y<g5KWE|_9x^_J1z5^^$G$H;a6jW~TFI(0PYsQ;O1ofgT5mM2
zai!dtxcIi6=HB=hEcoXxaj61bBHY<k6GNY+VF>2;WKE>dWmCwL_1(L74b8cz&6CU~
z<=7FuDqKd@WbDE_$GPM^lc$~AgT;g8Uzh{mtpdO_{X(VaIE*f!QPD-G=(ucMXWWaE
z#%*M4tA2gUI)ZyJB1M>{JGpY)$f02L>~qodUa3KrjdIcPA4e;qo6}pD$ME<*^+P9s
zXY%}$-fPWDUIgCy^BsS%q~CYs5oG<>7os}gl|2)5{tzeAoU6mSn;>~H#ruWV@PwKA
zp~L-w5|391813t`re(X6vDzbVyUpjW7kk^7i#))9<pbSER9oY<FED<*vRl8ZW@?2@
z=t&ENyGZ1HTuPmfr3=W)(tEk>#dJtvZFGL5XxcMbJ-t3AG(8j!^)DzHvc7}4fdYvU
z7fYBEB0Z@RIE0&?+TxOdl$eAx*xPB*6q<SY`87g3+K47X*u;^A>*>NE=CH`0y6+h1
zK0FwQd7skhAst?V<cSx^IfO%<#kE@^s(4k``iW1A^A3YErbA*E8vrn?P)=Cvk?=qb
z!3qCj7)JQf9Exd*;%NP^1ZaVr2=N0}I%tUlaHB~p20QLW!pOk*S@SOI^hL^mx(B=A
z>0ax^JLOWC7cDR~qdDTJf@=6dAg5750qMI0l+z&>6{0nr<tERztKMkfm#SS--?&8Y
z@#=GV*Yllbcz>8SeXCl^x<sY-V&~S(asQ#0_s(8`-OirSwA>p04$yFeTrBQ1;$9E)
zxN4Wt-k+7Z?ibl@O-=Vwd`SK?7l4O3Wbp1hC9TM7Mxc|wiq1C|wN775B++ph(;}KW
z5OA0LTA@hbG`V!GRQ4`*t-@r|>do0b_D0nzoFj%S;io4{xFkr=`AoXkn=@D7l{{^}
zYjwr#l#m)vF)|G&l6L$~mF*fK=_z`adCn_KC*PZYQKkQKiPo+hp#vkct24G<MxV7^
zZV;69H5T{*qGFyq1)}S9^^MbggFO{E0aVa4`_-(W2Ru3tnpOf`xoSO0XeBF#&*#rl
z0j6!)-sZiSy_L&oJ8|amUD=#ix?|;)o&nBud*kW!V6LU<%brPF+#=u^IQR7FKJB+D
zeSL)@rZQkH)n23fydkvPBLEDSP7Fqyv&xQtTA!XkHWXQI>bg*ic`zGRoplN7Tjl1a
zEan0jOf<U{Jv&iivweEhqqrtibhiG^_L%V#JxeNG?hE4@53*DVFA5#qca1&(%ZK!n
z0~mc;p{3s#e2+ZeY4uLz6%?bG%2G}oj2|rcw$nA_aBM%f6k@kABb1?*4D(QfR(p`?
zB<MEuO;)G$M>n?0dMuPZm;iQ+%7dI&eoQt974CMjB(3ZAPgY8u<VwjO%KULEy>gDo
zmkZy!oPB^z#Fc2d@6WiE!B2I=`&OyQN)1A_Sz9B#wVX9Y&FUNuhsaGq9{JvLEiBEM
z1MNaxr9JLQ3d8a8V>i90pCwiDnuJ|0^h>LMjcDY)3R!HbTljH2!a3h~+KOtuHd~)U
zWhA93su{I&Hkw)B(o<u%)K<7s22f!jYve}c<D%1Z;C#sBy$c!Sn&0-)9yqNHH|==8
zt6R|#moeM5Iy`;QHQkX|M4lgziMr{yt<Qa1^)aC((SLh<5dgAVqbzPC?&qgmxf<Hs
z;E?-MK|O}Iew(u^$7L2+lG&D8dYECvPtza*I$WTvf4gKH!e$Se#^#474s?5Hy=jJq
zCO2n)y416yvqOz8pX%~E%<LWK`!ee|H(g`7c~h;nhE9hnl_kG1+B_S$WI+`~jDd{y
zV?}!JNjM+P6wkbV;i)*jv2ql0Z@PCn(-elsJ2cd=)8XK{fA-RKXHHu04p!DA2D=(v
z0Op}>07XG_=m*bJLMKP>(-jNgMtTWFK8-k{X6-S?DwG6R@l>YceV}ig>$zSEX{k%t
zbYIOmrtCN|PoQ)s3>64~N+3{={t3Kkkvg2sM2kS<Z@MD@(sUdWyk8`iU-%C`xZ#zK
zCRhJM8Wh+X$r$a{nwIl%SZMhnpKab3M^V`{$WIb#uHdf<nNr$H|CetkWuk(C&Ti0A
zo1X+w?(vn29mp*Qo=T)Z&hTHM4SK#i;7C7vXSNUm40Zeequ*{JdMATrg<J8i^S^Zr
zkzNphs5Gty9$x@Rz5f5>js9)_{`T-X%-?uK^A9}Ihh;(Who}~qqP^xC=hG*@`<KP3
z3k1O=!DQD)TYAV#1Q1=BI|~23(pTTmVPR>OY^pFoUK)ym&{Lf0fA8tlH}sHE>CHD_
zLi`w48}iSu2<DAheFL~}&7}Xwh`w<&djwz><XRxg@#ieSyZg{E7m&*6jY;>+YWy3p
zl&DHY#(gb<Zl+igYR-7eH(h=hu?V8KNZ>AOW%<j5|Ba>U(17&dHjU4zbkQQLchIvP
zD%o$cUv9a034vp)8qg;8?`Qw*Lpoyth|q2IT(nCPoF1&>!VfprYd&rE_iS4ov!;^&
zXXH*e;3{(WnDZ)f*YHQ=E&?DmPEJms(?pt(mwV6)<1)`h<5?1le^_}Fa_o^zM{@3i
zJ-YqA=G|$!^K+CJ`Nz0vCE71@rlIr}lvbQzM}n?i{X+80^eFT4qLwnP?w`D7ustI<
zKG0JYZ;7+7wa2TzabN48b>d*3g0=88y-tdF8#d1`YHd8w{u2lKbdhgKhM+tgU6?_L
zZV%n-wDdytXq!;-HsTTN<%j3Yke6mSrlHcx<1?$!vs@Iad<=f<($!Mh|F`$BzH$9B
zycD0To%G6IGhsrXDKWmB+B_eEpKDmJ7vJima(OJRoVl(5+=-_iXC&~{CdutniYNrq
zO)YMKiska82x>@pd4f8tY{}+(`|qv%+Xq3=z42lJMId>0(ONC7Jo!fAKrTcw`{u4i
z?D4=o`J&@4!+Iic*d~+iAQ0*-7<btpea2wlE|NS+yJS6U1^Ypn*7Ud6=8}<et7jxN
z^7~_)kx?}7S>)y5mX-H(&+BOjZ6YExuBzSRqMF`9e=`7xzPR8+NIR#-b9_y+#2rb{
ze+%DxEA7&1w*Q}Y`08v?v2Rg%uOt0yj}|X}R1#SX(X(iOLU6>L<%Y(E8<rCyxj_&p
z)gdZ|Lnj|qMshJlcsbU+{=TnF=bsOlNiMpXS{&+Vzm{5}=Jj?p2dQ+2&eyh+O&}}x
zlVzd=zBlVms?0BZJ)O3!J!15@Xx_Iwob<~_h5l=fe*)`26$=?J5vM#)vLuzn>E7&C
z<EA%dzOe-P&hs7F>m9vA#vGbtwK}`T8c}_0rNQ(p#!P{f%T}*=(kQ#L@9ugUz~jRg
zt;Dr<_?I6Z4-bq;Ox(MqT#Ua!xP4kwO;gN2D{4wOokMu034NpbeFO}63dm%F;Zvf2
zJ9uyEoG@l|bc>lT#XQrzxoaacAZJ|XpKWsEDeGF1#gfHSh_w*Sh<3jOr*i>_Qd+i6
z^$zkl`81XHc;L~e%5VLN@4?>C2r@$0Wy1x)AJIpr8C%SkMl0npGI37GvqjR62oqL9
zwZ6mXY+gBb*_e!8x%9rHkN>C<rHZ7WZ-{x-FL(m*k2gDdcIa6G*xV#zs<1uo<RGS)
z2$l3@;sxY@Xe}4ruieR|&pM8OZ<S7;6>XF08z=VYvvw514w0wu*=9FBvJ0tw^FF(j
z(rtc=tVhQ01)hXx<~~xYS7FNiXDzFjBgHMn-WaQ&sHyWYE}qUUF`7feEIk_56R(@o
z)MFqMuOAfym8?b5TVx=WcbkE@KQXw2m>XoV&{lq$A+R}Q?3So$yBXgQ{XrU<7^A2C
z@qUfTKgtqcY2C<>@qCFf2tP_9ZC15)GJ&{%6Uf3-@9XiDR=zk^r-(%%^P&oihsTkf
zKUjf#(O$v<VL#29t5<>sgNd}49(0_3Y+CKnjP9H1g}nN0j~>ZL5ymNT`6>M=V!!vg
zP6p$|H`XcEK?UgmxC1+xc{uEx3MU*I9iM<4dG7BU+(K~e6(^pWfk;Qi(6x*bZXNC%
zw#K*mgs~Ym)vpZT;o6yVUF)3+hsB|Bdrz6a*lfp3^k7>qC-p&K|1yOCVGV9#oRsW#
z_{Hv=ialb_)#$LP)_ef5fg^_kcaO<TJi8W7-3#U(ZyJ2)N|4rs`W<J`-ZL)WL<egM
zYkZEu?v}2p^DeGl89IBcpDv&it!X8k79`{LLcj_}ItG7G^X#9OwWw~S#JKnxi`Tm#
zeDJDbhA+QD?%=5Gq_P*+upO7rlYtq&hlT~D0=qmn-bSB=Cj!>M9DmU&e90<&wX@~G
z&6U`DST~+HWM<RKRQckm61&t(k8(Vw|Crv@x|&R&ar-Dg{e0_2DCw8I2T48fScDYI
z<uCncP1{}V1C4fMtnQ~d24$-vx7+NiUMZ86z=HpDd`HiMLFgv0h{Msb^>KVAm+Z~n
zzS8p8B&TQUhk=Ca$`zJ1)w&W$(ujc)c1}MM1Gfh?QvW#Hf|qEr2A=rZ9}%ed@QF@i
z7K?P^T&0dr+r<jIo~9k@k#Pb*VOn%>x9^y+VaOd1XKV`D*q2O)gGCM#REyHQ$>BDr
zNSi)auYmJd6RS@*rB63YVCY57TDWF6J5rhbM*P`g)6k-(xpqe1(YJp*uyr|gdft&4
zHT1=NECG1namSvMoLR-wXL_)>DuJvarYv3vc#g2QbjOm_z2R%p9YkaL<7}Hr*QstJ
znpQvAF<t8CK4OP|+_PEM#?{iN>`KtS=f)eS4~?t3?#Zq9el^^Q_`jTjE0ZzAuAU3|
zjKS&Fcw7|%KdB)@ynT>NhpYN3e4!H@B}HY4GzU0xPgJ_&A@qY!Kd{FoRJRa_l}NBC
zrE^~b7`Kb`nNDi1WswcU-R3X8U~_}U)c}D|f0A;)=AR)Jz<L&uQLdSV^jRg3b?%r9
ze5gCL>C_fXP#bffmsWWi9#evqJCRMcfuLv8{+I(9P1IG6>RWmU14~URI|iR0bH~?4
z%!zBAlpJ+XjU>eL6x%dwIx88+$1Ko>==;URXchDw`Tp}lkE0dc|Ii~FU&X5Yb<~3!
z!cJA6@OZ}RHmhfbGPnPoSNBo6;W3hm>&Jm5B1u%9<SA?b!K@Vm1&TlaTF;{K*zE)n
zcdq$rkZoQN3RKMK3kjG2iu}_+DtF;aLDUi2KMiYwiPfZvRmjvZ5c%{&SN0SGa-*%b
zr}6)5?%D&HUgLg-4#r7HoJx|FxkO7!Q6yp%#@I%zG-^Vn+j7ZNx=3!nxmBb#w_z@A
zPSKKz%=L(oIZ1~}l3Y7Ty^qf6ZBFa``}G&|?0LSo=llJ9FWWErm<NmCm<Zgxo)fR&
zAtrGeKXV6KV%%p!GU%zrWu)e^#(WzbLEw7uMiVg{gL2G(%r+w*8Zz;ioP@ItObmG+
zT4MpNO{d_2v+EsWuIE1nP94&Xn@AKC#u8tO;NW-5sRSLPRy_v)2sa<CPB!rle28VP
z%bl>d^9Y>?kma|pL|kJ@n@!!Q^FuQQN_p>YY}z;Cs0E_(5FGP*AZ+)GIwLKXP3_}o
zLUa7ZRU~UPm+ihSFsC*YNK)>|C>k=Z8TVvwLr!f-Zrc6?i8~bf;$&^n(C9<24n$qg
z`hV)+)PPm5x^HowqEfa0h5m<NK5Z%nVHjS)gV()vUfLu5D*Jm^)!^dhLd~QNvg{Xo
znjuBn;a-8ezct$aL}Qjgr^$nO5HEY8I7M%n9dF<mo+6_wklB^QWO$?TG!%#J4rqGO
zgkYJ~O44?k405~xpigd?2c5n-cN=M7GsiOtSeutO!jd$QVXM$wb9Zl-DJ~isr&QRe
zchH+gKhWoHAe~EwRDO%uu?&E;4HZ&8f)E-NU2rr6KO0}&tI#t>WT@VzH`7@dM!s+E
z-jJk%2U^Y=c~H}frTM*gcR=c1##TRe17s8rDyhl2OzW;E3%qFp9~u@(iF01Kf-+(2
zZL4HOYlV|^Y;*T+&8TyVO90;6cE1NJ%l*)iIM6+OOt=K@@8x8Jb5P0axqwvmH%~GS
z<#btadO{j$)uuDDg?%LjpXW@_btY8T+XihG>jhpat95d86_&$YtMh;?5djlz7#b$9
z2g}4D&!te9&c3pk>O4w??=Z;7d_z}*KM5gpy5gPu<?3>nX<aNiBbmC}zybKlQOd6L
zlHV8zZD%U#uV|9k^9EH@vU@@c214;u@l}x7&RWT&;-UN1;oecGsV4T@lySWGohcd9
z*3~v~C;@YgZE6E^+09?z1O3NS%?={ZK^1RT2bp=pw~Xc`B&8z(Ho@w$7fue8ZjJ>}
zvY(a?z6_Qd6jIufIXSkxb<e7zty28gU;G3sV&5Wv5dSiSk=sp-;I*#w;-h<zX0;Cu
zmsMnEjZu{UE?FPiS?#ZHRvU5~>Tc&K)jdw+f7Thn^ZHe9*MD9phw<IyVH<}DWYMZ(
zN=8BpbQ$vYJYA@h@5K&y(&87^ro(p1k-r)RQ^q2LD?Imz$@?vc>!{oaV$LpG4La1e
zA`FWR1HR>vtwt7I14*%(Mx94Wrh!AJy8L_WU=10k*7fc^lb}hucEL~&GisFCHSM3{
zh}JkDk7bNcm>rUJ!YI1H!9TJ&3ivaHTn!^q4odaL)!FcCj3aX5G+*vY%rCjkVPiE=
zt>BS3y5^$*sI20F@7E;94EO_o{Bc0gskg|HFnx#aw=!73U_F231IGwY>wyGf)IzJf
zg(;639r~S2Lx#1@!3{wRQ#Gq~2ZS20?Kf{-mL=~h)I1^fXem`)qq}YH1%_x1W60b#
zstV%fB>>AA?k5CK9dAhbtaJ2bBy@qhWu>P>9O~jQE~n_yHY6!0;iH8*+?A)eW;iC{
z>~`0$m?n~j+N{nrR-c(&ZvRR}OdCW><~Yn}iK@Zw_PT%vJaTLW2C>ML`~WjrP+rX$
zdAs#tekk7cETn44YP$$36s9k)SW6=T^a+I*mL$*(9n=Reqg>-rtt%Pw=cgl~8%_sN
zHSj*MC#mW>!QHFV8rrL%5=ZolQo7r2lxjL;X)m{X)QRCHkti%nw;2QT7hW4IdCLPS
zI3txj2)m+>kDoH}?)%6N;PL{>M|n_Zl9!Kyur8pQ2Ry2(FT4M+tlN40;0mHRLn`Wn
z*{?d!HpqC|EJe_BCiht48FP5g?fm!zI(K3#a^|-D1;f&Do7%E5-t}i6rEO}D)d#dh
zbFCAV>GcxZ?c<aTS4u{!Y=M@{VUUH|xdcR5hC2<C;5m6ir$H=d!Lkf3ayO>3%tKaY
zzJ*M^)@A?eKxbjkzYK3??)qMusZ^D_DlJ8-H1bbL(6!`|<JBPgEe6XQ?QJ!6Jn-IL
zoP7{v{Dlj^LWC>aUA5eG{qg88e{eP|vgF+u+~RgQAu<tVF(5$eu{3HWV^OXHgJ&@_
z6@VSM3)vEV<CMmNz9yIO3>oRQ!CZ$uBA}9ZAGV`wpfD!&YiWnNn2P6oypo)UJq-yC
zDQvvi#I%b#lMQ)Lo`&@PRX6mQ7jD3Qo4WZ6!3erU4%=U0F$8pflKIi5u*?x{;AHn`
zm&QqGpKU>RZL=TSpV2hR8lYe<ILO&grw8j8FV3l-!kEU)OYr|m3aKaH^4P3i4<fC{
zO2Y)oM0>?x1`{!E^zjS2R<lmRp=cDRz}?^ge$(hWfM!XJ(isTbTC2+y(|7nA7<@fM
zcRKU1O-7J}NvksB-Nlx67U|$t7IzzRU`j|o{u{W+90#J`(wl+m%w;0I_PO#?tqkx5
zAi3!_@h16vEVf|0yNL;Wft?8}fFTxPd3h-%S`0%S^+6gc)w*+`@?v`ZJzJlK_8Hja
zP7?wp$C-(?vqY<>^-9<e&5Txn2-tapJjRk<Lz%dYhD^QJWQ!}B6&;}h>>4H@ws9Kr
z0t&DI5wj92wlIcgj)%YWX7JA;Zf-Yb5deO(<xCQ~2NAZZ&x7;<#}_jLN(p(GD9*I*
z<AhAEPJtylEyO<s1IGyuoS5Ttev&;WYSlH(g!qK2OW-*;@<!w(NOl2t^hyFsAIZdw
zj@1qOx9J=zG}tXvrfZVISdb=Z&%;dg)yxIAS<`n^k3oXa32EK1edul5{w$LRgT}u~
z>sSfS$YO+{mv4wxxTp#$IJs3%3LI8^C4smJK0aZ8=mlSIk^VG{##N-VUTZIMCie@M
z==5>+v-1G->1;>stgU$`JheaFZz;Y0j_McE@uBjwD-6o+=FJw$JG~C5sfsB!csTkt
z5oL@llc|$T%rsg%yRO0T3n4t?jqjf={``^Pb@9)?@@x?s@qV+Hf~W)z`%$?^=)<W|
zd6>~7`|^KdFt#u_=I(eZ>h1-11GC!g_G1nG3EQovW;=#CV<APU!GMfhbN;deNqNxb
z3*gXmv1V$y*sO8vUHOePmbC8Cb>OatrJo(TqYV8YwtQ9EfE`dct?N2{Cb2cGxh!yo
z7SvAj&NkbQ&88dCZ`f-j$OrsM3%U9tf7#iKUKVkv{;FTGnRI)86@;SCCBU~5+zYQ_
zb0O@ysY#*_O^1ZT2pXA&+Nm4C!zn)9sOvQrW%OUavQnnw^N|^@O6fugA5f;*w^QAA
zTpCFlPT(pYNP3${$Z1)cPHzI~CHH>cSkv3Gux{vSqi0;no~9%NZWN66>6F#1JqRm=
zjp%~gAc3-z;GHR(P`Z{sMmn}doNd9C)tFCmX1SaP?s9tMOtkmkf4JYoOj;u)Q$J}8
z{KlSXhe4q15>di!N5_?Hd-(nx*N8Dmnbwb5Ly3sz1}V5qR{_U`)N`$_f<95Agc%*@
z*TPRzJtM{W2!$nY;Obx@;gU0aYt*)7a#&0eCd!zth-gHFrwgI80kh#&@uxmK1>I0w
z*MuvB{%ZeWK>;3?^ohV8<ApFxdxL?_cajdGf-a$kDR7)_GA=t}vVKtN_>b&W-j@~@
z6sie?dHDdrU*;Qd2{2l*X4XlTadIc$XUa_ZJ#J&;{iXYo_ku1(gmjQBW|XMa-N@|P
z<FDg%NY-V#v8icyaBU4=n0z)VB`Tdw!^E}dl75_$1iUgTwIZRUfFiOWl$Jr;-9H0M
z?)*I|*u@sbS7!84D(uDpj1-oIoD$*~HiN3(?1($^HUwRV>O2o}&u(?7*XP)6uHw$k
zB@s(&5uh-<5Z1YPPf&b*?KuC*RZ*8<w1UcQ1}ZSfe|GLFs#T4NQJ<dGUxls+8cgy!
zbZttmewo>Bza?eiYs$@R@%1*=V^T=c!GteeN;!9<5MhGTH0=6;^Z{D^+3t^Ng=@|5
zP4_&$#Kx$9OASQNDp~>?l$DNZG$^a*NnS<~2bivS$D)Lc>+koLJEB&0-uFKYFD<~8
z4TU%D)u8JKP6>g~s;+kOz1bEA4wn~Z20R{pMORH5$I>BcU4A4C!au#bR!l+UmP*GZ
z3&bJ^o+d*zzZI5*a;UY6ugIet9v^@^lgEJ0x9xypru?kqN-Wy>MvNdPE0f6JH__c_
z36a-;%N_;8x=)&0BuVD`o}Fpj`wDYUMVz?{a70k;BWV8Nu8wj|P54F2st0Y4K&z*;
z=J|p8U_eEe$F>ueyvSst&ZRdqe>R(VfaksCTMWQ=Oh+bhN2dBsEen(wh))OAONr!U
zkti$z1a4W_zrrwIt4$x)!41IOG0pMr#n2L6iW;hwg)sl%G+|d49CD%)bUTLJ8dm!$
zxbQ~$+&n-vjkeF$z#%tgU`F5k6^_)Pc;13BwQ+r%*o&l;g0NK+{K}NyRSk<x(oG~d
zVagpTtS5+_FK#W(?gz8N3ZsP+E9@j}!t07ZYlBQ5e>1^5B)^Jb!Wk|#ZpTX|;A?`K
zQNQbp6H~b@F=+u}5%SpWPVKT6oCOQuS#~=uUHSMvKX%5&YI=7cR%dF~PZp@=z2Lq7
zQiIv3ZF!m`Ny;2IMt8t76%fXD?L%<V=!|f=&%S5P!!?b6TAWN4?R0%(i<m4YFOBqB
zWk2+B%Fhba0CZa1sTi-bI#2BEJKbYYlcD9~G~6?Qz*Va{&c+vWbq4A)kAm`5%mr=s
z;3W1@z0&Pg;e$ueI>+xy__jGf2lu21R#8a?rM?BYR7M(N(E;A@#qjp&(BK|ZBxtqu
zAX3&k*h$!_Lr_K&YTZhYaPRU!y_mO&ZpgnxufjK@?Kn)uTSajLO2}=<5&U!{t#|?A
zrr_AaPur}zgE&Kh_cAm|iEd}o1GwZzHBKAty4{HE!QR*h>B7;0EmhF&Gx1$f->Lgw
z;ZuFVwjgq-&D;y}FZ@)&9eOk!nOYn&xwl;XX=N%fmJNvwXLtCgC=DCAJw<WM+nbZT
z`X>Ch_d6oBQ_nWswEKbiqQ}xnuf2ssAN;y{VmH})(=yYMqy;FxH=_7gY2Ch!V;TLj
zmrwqU3GDbw7L0gv!h5(8bi?|C3a-4di+K*yu$#tzQO+e&ngqRwBU{A;E2)#z$_h)X
zZc3?1WX7iBxzhTiTd3jEp-579+%6URVVS`X))#@ZzlGvVxY;x6IyO%4Q0aN5TA>;E
zBekH9l_e*?F|}!JOUqTR{Eq5E^U07^yezI<+jUtC=b(ZxS56&n6;)_@GSEA&#M&&-
z{s)fdz@PyLXlUf)1VRT&HLmsj`jg%?9gi_I=#^GKC?CY7B*fNczfihTPnZBh_f<<D
z6@1T*^Ia<W??fWWEk!lBG5L|*r+mcnL+a0!>T!1)=4K|+PfYCbq1HH6ItxzvAJi+H
z=<<(F?MfG;RSDR2t>(^9?g`Jd@d^pFw?!#ZEwBYl)UWxVb=Nhe_qpfr-=py!l1Z{o
zsj&aG1<_wjJa6%JxWidKf`FKS5v<oK{>ZXxJqshD+zQEPxm|>`$A;v3*LU^e`$W-z
z1MAo3{0+YEJXi6w+>Va4JH?lretwU&dC+a|h)uM-HKB0h$H%7eVu=^n67|PzV_Q&x
zhwJkN;Q_0wSFFcc@}49Wvuo^3hbTMZD2@0hVQXyyw}?TlGVhL1i#6}`jr8BE&JT0-
zmGvVC57+Mph5G2Do^P;11(z0WzH(fCf7b=A-^I5AIJU&}<^xHJ2ht-*27!7UYmM?*
zP9$`hD>10xLXJgfr@q;LT~$<G->$4+5AkIYDP55y=y_Kp3&(K&h-`drMM|nszFU8r
zB-ZI|#S>rNqWj+dg*O@QB83zjp7oX8>Z@MfpyrS)Ry>Ko37<|Sy`yU!OWtEM_X&pZ
z^dkq-gE{Ka17e0MLD}D4N+@eBH1)IBXxzOOj6eJuy4Y7?kMD|ylC@^yYt?zM(kPA6
zYc<=#!1D3-gMV&Nf%306OV9C1Tq>h~L6Lv*QSh0a^R6c@KjPJa;&hjaG)>YKW{G96
zQqEz;{5qu$B<!;)B|D$q|KEP1iKRtEiR?4wdrvP~qCaKquYTwd4CWg5@veEgE&6AM
z`I}__(RAe+IT>e@#UZdVOgxHeZ6rO90Fmv9X&}EmR2YpHmqdMas>+IrbO+*M2dY>h
zgX2=NK)#qZxZh2jk<G!DN-1De$~Y+(k#pxehb$!ckcRMn$`Wz5A|+Fg1PxAH+L&kd
zVjbXWU&{WBn+TAjXg+0%8GiZqJVpBHU*=0nl&ezN98lP+J1DZfuVT@r33*(Uh`|`y
z0bOe+Kc)X{BPtTieq<HN3BLuUANY2_fQ+P;v>qvP(!AY>azHJ}{76%Au<z@(iC}ul
z;TYcGiYNo38b|>jL?g*DiNA`zC`w7CHjrhar)r|s&*;no<0+|Cx+#+4)uO2R|56mV
z*h+%TzC-3qkF5JI-LE_(EU&o#)RHxuS8LjikiXjOeko{ye)S*Nd20D;MQad1Dy_#-
z<o~<-H=8y>00V*C9((@y-hW=9+Reon_VP7W-`?_H4LH~V17WloE&Z>7RIXdx?0Ym>
z|8C9PXvweOZoZ4K9+v?+#^pqONyF=8M=$O$llbx>FrB@1-kq)tR04~$+9vbA8$_0t
z=FD{-I^`D;R#w9zRai12FMMA<NByl`7IClJMD1w40m@)1B<H_Fgh%QNDOY;EkYxO`
z?Tey;t0+HMY_9vW6-L*AS(_N%`Hl<bUlHnFZ7@@19X;gy(-kkk8Ko5U-NgMc|5fK1
zmm6x2i`7(7Fj%<N&%j*vXT%f{fu<uZ_g<SX1H}%qMB9wV1K4}V2#bdiU*LU1nnklx
apSPng?5)|4UTVAm{McFV*;%-Q82LY<tWGQd

diff --git a/docs_old/community/contribute.md b/docs_old/community/contribute.md
deleted file mode 100644
index 78c713b18..000000000
--- a/docs_old/community/contribute.md
+++ /dev/null
@@ -1,181 +0,0 @@
----
-title: Contribute
----
-
-## What contributions?
-
-- **Documentation** contributions are very valuable to us!
-- **Examples.** Show us what you did with Outlines :)
-- **Bug reports** with a minimum working examples in the [issue tracker][issues]
-- **Bug fixes** are always a pleasure to review.
-- **New features**. Please start a new [discussion][discussions], or [come chat with us][discord] beforehand!
-
-Note that the [issue tracker][issues] is only intended for actionable items. In doubt, open a [discussion][discussions] or [come talk to us][discord].
-
-## How to contribute?
-
-### Setup
-
-First, [fork the repository on GitHub](https://github.com/dottxt-ai/outlines/fork) and clone the fork locally:
-
-```shell
-git clone git@github.com/YourUserName/outlines.git
-cd outlines
-```
-
-Create a new virtual environment:
-
-*If you are using `uv`*:
-
-```shell
-uv venv
-source .venv/bin/activate
-alias pip="uv pip" # ... or just remember to prepend any pip command with uv in the rest of this guide
-```
-
-*If you are using `venv`*:
-
-```shell
-python -m venv .venv
-source .venv/bin/activate
-```
-
-*If you are using `conda`*:
-
-```shell
-conda env create -f environment.yml
-```
-
-Then install the dependencies in editable mode, and install the `pre-commit` hooks:
-
-```shell
-pip install -e ".[test]"
-pre-commit install
-```
-If you own a GPU and want to run the vLLM tests you will have to run:
-
-```shell
-pip install -e ".[test-gpu]"
-```
-
-instead.
-
-Outlines provides optional dependencies for different supported backends, which you can install with
-
-```shell
-pip install ".[vllm]"
-```
-
-A list of supported optional dependencies can be found in the [installation guide](/installation).
-
-### Using VSCode DevContainer / GitHub Codespaces
-
-If you want a fully pre-configured development environment, you can use VSCode DevContainers or GitHub Codespaces.
-
-#### VSCode DevContainer
-
-1. Ensure that the [Docker](https://www.docker.com/get-started/) daemon is running on your machine.
-2. Install the [Dev Containers](https://marketplace.visualstudio.com/items?itemName=ms-vscode-remote.remote-containers) extension in VSCode.
-3. Open the Outlines repository in VSCode. When prompted, **Reopen in Container** (or press `F1` and select "Remote-Containers: Reopen in Container").
-4. Run the normal setup steps. Your environment will not complain about missing system dependencies!
-
-#### GitHub Codespaces
-
-1. Navigate to the Outlines repository on GitHub.
-2. Click on the **Code** button and select the **Codespaces** tab.
-3. Click **Create codespace on main** (or another branch you are working on).
-4. GitHub will launch a pre-configured cloud development environment.
-
-You will not have access to a GPU, but you'll be able to make basic contributions to the project on the go while using a fully featured web-based IDE.
-
-### Before pushing your code
-
-Run the tests:
-
-```shell
-pytest
-```
-
-And run the code style checks:
-
-```shell
-pre-commit run --all-files
-```
-
-### Benchmarking
-
-Outlines uses [asv](https://asv.readthedocs.io) for automated benchmark testing. Benchmarks are run automatically before pull requests are merged to prevent performance degradation.
-
-You can run the benchmark test suite locally with the following command:
-
-```shell
-asv run --config benchmarks/asv.conf.json
-```
-
-Caveats:
-
-- If you're on a device with CUDA, you must add the argument `--launch-method spawn`
-- Uncommitted code will not be benchmarked, you must first commit your changes.
-
-#### Run a specific test:
-
-```shell
-asv run --config benchmarks/asv.conf.json -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
-```
-
-#### Profile a specific test:
-
-```shell
-asv run --config benchmarks/asv.conf.json --profile -b bench_json_schema.JsonSchemaBenchmark.time_json_schema_to_fsm
-```
-
-#### Compare to `origin/main`
-
-```shell
-get fetch origin
-asv continuous origin/main HEAD --config benchmarks/asv.conf.json
-```
-
-#### ASV PR Behavior
-
-- **View ASV Benchmark Results:** Open the workflow, view `BENCHMARK RESULTS` section.
-- Merging is blocked unless benchmarks are run for the latest commit.
-- Benchmarks fail if performance degrades by more than 10% for any individual benchmark.
-- The "Benchmark PR" workflow runs when it is manually dispatched, or if the `run_benchmarks` label is added to the PR they run for every commit.
-
-### Contribute to the documentation
-
-To work on the *documentation* you will need to install the related dependencies:
-
-```shell
-pip install -r requirements-doc.txt
-```
-
-To build the documentation and serve it locally, run the following command in the repository's root folder:
-
-```shell
-mkdocs serve
-```
-
-By following the instruction you will be able to view the documentation locally.
-It will be updated every time you make a change.
-
-## Open a Pull Request
-
-Create a new branch on your fork, commit and push the changes:
-
-```shell
-git checkout -b new-branch
-git add .
-git commit -m "Changes I made"
-git push origin new-branch
-```
-
-Then you can [open a pull request][pull-requests] on GitHub. It should prompt you to do so. Every subsequent change that you make on your branch will update the pull request.
-
-Do not hesitate to open a draft PR before your contribution is ready, especially if you have questions and/or need feedback. If you need help, come tell us on [Discord][discord].
-
-[discord]: https://discord.gg/R9DSu34mGd
-[discussions]: https://github.com/dottxt-ai/outlines/discussions
-[issues]: https://github.com/dottxt-ai/outlines/issues
-[pull-requests]: https://github.com/dottxt-ai/outlines/pulls
diff --git a/docs_old/community/examples.md b/docs_old/community/examples.md
deleted file mode 100644
index 813bd4834..000000000
--- a/docs_old/community/examples.md
+++ /dev/null
@@ -1,19 +0,0 @@
-# Community projects and articles
-
-Publishing examples and articles about Outlines are a meaningful way to contribute to the community. Here is a list of projects we are aware of. Drop us a line if we forgot yours!
-
-[MMSG](https://github.com/leloykun/mmsg) is a Python library for generating interleaved text and image content in a structured format you can directly pass to downstream APIs.
-
-[Multimodal Structured Generation: CVPR's 2nd MMFM Challenge Technical Report](https://arxiv.org/abs/2406.11403) shows that Structured Generation can outperform finetuning, and maybe even multimodality, in document-image understanding tasks as part of CVPR's 2nd MMFM Challenge.
-
-[Chess LLM Arena](https://huggingface.co/spaces/mlabonne/chessllm) is a HuggingFace Space where you can make LLMs compete in a chess match.
-
-[LLM Data Gen](https://huggingface.co/spaces/lhoestq/LLM_DataGen) is a HuggingFace Space that generates synthetic dataset files in JSONLines format.
-
-[Fast, High-Fidelity LLM Decoding with Regex Constraints ](https://vivien000.github.io/blog/journal/llm-decoding-with-regex-constraints.html) presents an efficient alternative to Outlines's structured generation.
-
-[gigax](https://github.com/GigaxGames/gigax) is an Open-Source library that allows to create real-time LLM-powered NPCs for video games.
-
-[Improving Prompt Consistency with Structured Generations](https://huggingface.co/blog/evaluation-structured-outputs) shows how structured generation can improve consistency of evaluation runs by reducing sensitivity to changes in prompt format.
-
-[AskNews](https://asknews.app) is a news curation service processing 300k news articles per day in a structured way, with Outlines.
diff --git a/docs_old/community/feedback.md b/docs_old/community/feedback.md
deleted file mode 100644
index 033e91870..000000000
--- a/docs_old/community/feedback.md
+++ /dev/null
@@ -1,81 +0,0 @@
----
-title: Feedback
----
-
-# Feedback
-
-If Outlines has been helpful to you, let us know on [Discord][discord] or give us a shoutout on [Twitter][twitter]! It's always heartwarming ❤️
-
-
-<head>
-  <!-- From Marvin AI's documentation -->
-  <!-- Their library is also awesome -->
-  <!-- https://www.askmarvin.ai/ -->
-  <style>
-    .tweet-masonry {
-      column-count: 2;
-      column-gap: 20px;
-      padding: 20px;
-    }
-
-    .twitter-tweet {
-      display: inline-block;
-      width: 100%;
-      margin-bottom: 20px;
-      margin-top: 0px !important;
-      break-inside: avoid;
-    }
-
-    @media (max-width: 600px) {
-      .tweet-masonry {
-        column-count: 1;
-      }
-    }
-  </style>
-</head>
-<body>
-
-<div class="tweet-masonry">
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">I am once again reminding you that structured extraction using LLMs is going to transform every single industry in the next 10 years <a href="https://t.co/xQ3tcWnrZ8">https://t.co/xQ3tcWnrZ8</a></p>&mdash; Sam Hogan (@0xSamHogan) <a href="https://twitter.com/0xSamHogan/status/1780637917737816323?ref_src=twsrc%5Etfw">April 17, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">outline&#39;s growth is insane, using is an understatement! <a href="https://t.co/rHCNWhZdCs">https://t.co/rHCNWhZdCs</a></p>&mdash; jason liu (@jxnlco) <a href="https://twitter.com/jxnlco/status/1780618454040797554?ref_src=twsrc%5Etfw">April 17, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Outlines is an amazing lib and more popular than <a href="https://twitter.com/remilouf?ref_src=twsrc%5Etfw">@remilouf</a>’s modesty will admit. <a href="https://t.co/DfHbMPIlX1">https://t.co/DfHbMPIlX1</a> <a href="https://t.co/mDHIWJrD0C">https://t.co/mDHIWJrD0C</a></p>&mdash; Delip Rao e/σ (@deliprao) <a href="https://twitter.com/deliprao/status/1780780217180598377?ref_src=twsrc%5Etfw">April 18, 2024</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Impressive implementation of a true regex / json / grammar guided text generation <a href="https://t.co/RX5RVYaVIx">pic.twitter.com/RX5RVYaVIx</a></p>&mdash; Rohan Paul (@rohanpaul_ai) <a href="https://twitter.com/rohanpaul_ai/status/1741099984299135403?ref_src=twsrc%5Etfw">December 30, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Most underrated Github Repo in AI + LLM JSON guided Generation: <a href="https://t.co/lSB8KIet1H">https://t.co/lSB8KIet1H</a></p>&mdash; 🎙Jean-Louis Queguiner (@JiliJeanlouis) <a href="https://twitter.com/JiliJeanlouis/status/1736857292581093706?ref_src=twsrc%5Etfw">December 18, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Nice and useful. <a href="https://t.co/LX72AE0lgt">https://t.co/LX72AE0lgt</a></p>&mdash; Dan Roy (@roydanroy) <a href="https://twitter.com/roydanroy/status/1691556956941525458?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">HUGE dub for open source AI <a href="https://t.co/bYKuiEUZ1j">https://t.co/bYKuiEUZ1j</a></p>&mdash; kenneth 🖇 (@k3nnethfrancis) <a href="https://twitter.com/k3nnethfrancis/status/1691304781732843521?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">This is amazing - glad to see more outp guidance modules! <br><br>Will try this out soon I&#39;m wondering how they translate from regex automatons to token boundaries<br><br>Also why Open Source will succeed. Even today I don&#39;t see any guided output functionality from the big providers. <a href="https://t.co/Ity2H25Klf">https://t.co/Ity2H25Klf</a></p>&mdash; Hrishi (@hrishioa) <a href="https://twitter.com/hrishioa/status/1691181499671080960?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Outlines - a library to help LLM developers guide text generation in a fast and reliable way.<br><br>&quot;Provides generation methods that guarantee that the output will match a regular expressions, or follow a JSON schema.&quot;<br><br>Need to check this out. Reliable JSON output is a common use… <a href="https://t.co/Bkbh8vKogN">pic.twitter.com/Bkbh8vKogN</a></p>&mdash; elvis (@omarsar0) <a href="https://twitter.com/omarsar0/status/1691179888214966273?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Woah this is cool! Makes open source models more usable.<br><br>Give any LLM Function Call capability (and more) with Outlines: <a href="https://t.co/PtPykR5ZGR">https://t.co/PtPykR5ZGR</a> <a href="https://t.co/RRQjWHnIxv">https://t.co/RRQjWHnIxv</a> <a href="https://t.co/BwNnH8SMwv">pic.twitter.com/BwNnH8SMwv</a></p>&mdash; Yohei (@yoheinakajima) <a href="https://twitter.com/yoheinakajima/status/1691231912466223104?ref_src=twsrc%5Etfw">August 14, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">This is awesome! Being able to guarantee the output&#39;s structure unblocks so many applications. This is a great milestone and a fundamental building block for more advanced AI apps. <a href="https://t.co/WdwMOc7hE8">https://t.co/WdwMOc7hE8</a></p>&mdash; Guilherme Castro (@skastr052) <a href="https://twitter.com/skastr052/status/1691239359494619136?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">Juggling with the unpredictable outputs of ChatGPT API lately while building my product. 😓 <br><br>Tried prompt engineering to channel its wisdom into a neat JSON, but it&#39;s like asking a cat to fetch. 🐱<br><br>Luckily, stumbled upon &quot;Outlines&quot; – looks like a promising way to tame the LLM… <a href="https://t.co/oYQ6q8exAS">pic.twitter.com/oYQ6q8exAS</a></p>&mdash; Charlie (@14435635Sun) <a href="https://twitter.com/14435635Sun/status/1691439342689095680?ref_src=twsrc%5Etfw">August 15, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-
-<blockquote class="twitter-tweet"><p lang="en" dir="ltr">A complex system of LLM input-outputs interacting with non-LLM agents and models benefits immeasurably from structured outputs. The outlines package saves so much time, <a href="https://t.co/NhVQ6NpKDR">https://t.co/NhVQ6NpKDR</a></p>&mdash; Amir Sani (@amirsani) <a href="https://twitter.com/amirsani/status/1728734266568376433?ref_src=twsrc%5Etfw">November 26, 2023</a></blockquote> <script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
-</div>
-</body>
-</html>
-
-# Let us know!
-
-We highly value the insights of our users, and we would love to hear from you. If you are using Outlines for your projects and would like to share your experience with us, let's connect:
-
-- What are you building with it?
-- What do you like about it?
-- What challenges are you facing?
-- What do you think could be improved?
-
-To schedule an appointment follow [this link](https://cal.com/dottxt/outlines). This is exclusively intended to share your experience, please go on [Discord][discord] or [GitHub](https://github.com/dottxt-ai/outlines/discussions) for support.
-
-[discord]: https://discord.gg/UppQmhEpe8
-[twitter]: https://twitter.com/dottxtai
diff --git a/docs_old/community/index.md b/docs_old/community/index.md
deleted file mode 100644
index 9519f5b56..000000000
--- a/docs_old/community/index.md
+++ /dev/null
@@ -1,12 +0,0 @@
-# Community
-
-![Belonging](belonging.png)
-
-Outlines exists for a community of users who believe software doesn't need to be complicated. Who share the same passion for Large Language Models but don't want to compromise on robustness. Together, we are bringing these powerful models back to the world of software.
-
-## Connect on Discord
-
-The Outlines community lives on our Discord server. There you can ask questions, share ideas or just chat with people like you. Don't be a stranger and [join us][discord].
-
-
-[discord]: https://discord.gg/UppQmhEpe8
diff --git a/docs_old/community/versioning.md b/docs_old/community/versioning.md
deleted file mode 100644
index 023b92537..000000000
--- a/docs_old/community/versioning.md
+++ /dev/null
@@ -1,26 +0,0 @@
----
-title: Versioning Guide
----
-
-# Versioning Guide
-
-
-The Outlines project follows a structured versioning scheme designed to provide clarity and minimize risk for downstream dependents.
-
-Each part of the version number (`major.minor.patch`) conveys information about the nature and impact of the changes included in the release.
-
-- **Major Releases** includes compatibility-breaking changes to core interfaces, such as `LogitsProcessor`s and `Guides`.
-- **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatability with existing use of core interfaces.
-- **Patch Releases** address bug fixes and incorporate low-risk changes to improve stability and performance.
-
-## Releases
-
-Releases along with release notes can be found on the [Outlines Releases GitHub Page](https://github.com/dottxt-ai/outlines/releases).
-
-## Version Pinning Recommendations
-
-Here are our recommendations for managing dependencies on the Outlines package:
-
-**Small, Risk-Tolerant Projects:** Pin to a specific major version.
-
-**Large, Conservative Projects:** Pin to a specific minor version.
diff --git a/docs_old/content/docs/guides/migration_guide.md b/docs_old/content/docs/guides/migration_guide.md
deleted file mode 100644
index f931641f7..000000000
--- a/docs_old/content/docs/guides/migration_guide.md
+++ /dev/null
@@ -1,244 +0,0 @@
-# Outlines 1.0 migration guide
-
-Outlines 1.0 introduces some breaking changes that affect the way you use the library. You are likely concerned by all of the following sections, so please read this document carefully until the end.
-
-This guide will help you migrate your code to the new version.
-
-All previous functionality will be supported until Outlines version 1.5.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
-
-## Removed or modified features
-- [Generate functions](#generate-functions)
-- [Models](#models)
-- [Samplers](#samplers)
-- [Functions](#functions)
-- [Text generation return types](#text-generation-return-types)
-- [Inference arguments](#inference-arguments)
-
-### Generate functions
-
-The whole `generate` module has been removed. That includes the functions `generate.cfg`, `generate.choice`, `generate.format`,`generate.fsm`, `generate.json`, `generate.regex` and `generate.text`.
-
-You should replace these functions by the `Generator` object along with the right output type as an argument (on top of the model). The output type can either be a python type or be an object from the `outlines.types` module. You can find more information about the output types in the [output types](TODO: link to the documentatoin on output types) guide.
-
-Associated v1 output types for each deprecated function:
-- `generate.cfg` -> `outlines.types.CFG`
-- `generate.choice` -> `typing.Literal` or `typing.Union`
-- `generate.format` -> native python types (`str`, `int` etc.)
-- `generate.fsm` -> `outlines.types.FSM`
-- `generate.json` -> `pydantic.BaseModel`, `typing.TypedDict`, `dataclasses.dataclass`, `genson.schema.SchemaBuilder` or `outlines.types.JsonSchema`
-- `generate.regex` -> `outlines.types.Regex`
-- `generate.text` -> no output type (`None`)
-
-For instance, instead of:
-
-```python
-from outlines import generate
-
-model = ...
-generator = generate.choice(model, ["foo", "bar"])
-```
-
-You should now use:
-
-```python
-from typing import Literal
-from outlines import Generator
-
-model = ...
-generator = Generator(model, Literal["foo", "bar"])
-```
-
-### Models
-
-The model classes found in the `outlines.models` module are maintained but there are a few important changes to be aware of.
-
-The functions used to created a model have been replaced by equivalent functions named with a `from_` prefix. The function `outlines.models.transformers` has been replaced by `outlines.from_transformers` for instance. On top of this change of name, the arguments have been modified. You should refer to the [models documentation](TODO: link to the documentation on models) for more details, but the overall idea is that you now need to provide a model/client instance from the inference library the Outlines model is wrapping.
-
-For instance, instead of:
-
-```python
-from outlines import models
-
-model = models.llamacpp(
-    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-)
-```
-
-You should now do:
-
-```python
-from llama_cpp import Llama
-from outlines import from_llamacpp
-
-llamacpp_model = Llama.from_pretrained(
-    repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-    filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-)
-model = from_llamacpp(llamacpp_model)
-```
-
-The `load_lora` methods that are present on the `VLLM` and `LlamaCpp` models have been removed. You should now handle lora loading through the `Llama` instance in the case of the `LlamaCpp` model or provide it as a keyword argument when calling the model in the case of the `VLLM` model.
-
-For instance, instead of:
-
-```python
-from outlines import from_vllm
-from vllm import LLM
-
-model = from_vllm(
-    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
-)
-model.load_lora("path/to/lora/file")
-
-response = model("foo")
-```
-
-You should now do:
-
-```python
-from outlines import from_vllm
-from vllm import LLM
-from vllm.lora.request import LoRARequest
-
-model = from_vllm(
-    LLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct")
-)
-lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
-
-response = model("foo", lora_request=lora_request)
-```
-
-The `ExLlamav2` model has been removed without replacement. This inference library is not fully compatible with Outlines, so we decided to remove it. You can still use it until final deprecation, but we recommend you to migrate to a different inference library right now.
-
-### Samplers
-
-The `outlines.samplers` module has been removed without replacement. You should now use the arguments of the inference library model to control the sampling. Depending on the model you use, this could be done at initialization or when calling the model to generate text (so when calling the outlines model or a generator).
-
-For instance, instead of:
-
-```python
-from outlines import generate
-
-model = <transformers_model>
-
-generator = generate.text(model, samplers.beam_search(2))
-response = generator("foo")
-```
-
-You should now do:
-
-```python
-from outlines import Generator
-
-model = <transformers_model>
-
-generator = Generator(model)
-response = generator("foo", num_beams=2)
-```
-
-### Functions
-
-The `outlines.function` module has been removed. It is replaced by the `outlines.applications` module. An `Application` serves a similar purpose as a `Function`: it encapsulates a prompt template and an output type. A difference is that can `Application` is not instantiated with a model name. Instead, you should provide a model instance along with the prompt when calling it.
-
-For instance, instead of:
-
-```python
-from outlines import Function
-
-prompt_template = ...
-output_type = ...
-
-fn = Function(
-    prompt_template,
-    output_type,
-    "hf-internal-testing/tiny-random-GPTJForCausalLM",
-)
-
-result = fn("foo")
-```
-
-You should now do:
-
-```python
-from outlines import Application
-
-prompt_template = ...
-output_type = ...
-
-application = Application(
-    prompt_template,
-    output_type,
-)
-
-model = ...
-
-result = application(model, "foo")
-```
-
-### Text generation return types
-
-In the previous version of Outlines, the return type of the generators depended on the output type provided. For instance, if you passed a Pydantic model to the `generate.json` function, the return type was a Pydantic model instance. In the v1, the return type of a generator is always a `str`, the raw text generated by the model. You are responsible for parsing the text into the desired format.
-
-For instance, instead of:
-
-```python
-from pydantic import BaseModel
-from outlines import generate
-
-class Foo(BaseModel):
-    bar: str
-
-model = ...
-
-generator = generate.json(model, Foo)
-
-result = generator("foo")
-print(result.bar)
-```
-
-You should now do:
-
-```python
-from pydantic import BaseModel
-from outlines import Generator
-
-class Foo(BaseModel):
-    bar: str
-
-model = ...
-
-generator = Generator(model, Foo)
-
-result = generator("foo")
-result = Foo.model_validate_json(result) # parse the text into the Pydantic model instance
-print(result.bar)
-```
-
-### Inference arguments
-
-In the previous version of Outlines, some of the inference arguments were standardized across the models and were provided as positional arguments to the generator or through the sampling params dictionary. Additionally, various default values were added by outlines to the inference library models. This is no longer the case. You should refer to the documentation of the inference library you use to find the right arguments for your use case and pass them as keyword arguments to the outlines generator when calling it.
-
-For instance, instead of:
-
-```python
-from outlines import generate
-
-model = <transformers_model>
-
-generator = generate.text(model)
-
-result = generator("foo", 256, ".", 10) # 256 tokens, stop at "." and seed 10
-```
-
-You should now do:
-
-```python
-from outlines import Generator
-
-model = <transformers_model>
-
-generator = Generator(model)
-
-result = generator("foo", max_new_tokens=256, stop_strings=".", seed=10)
-```
diff --git a/docs_old/cookbook/atomic_caption.md b/docs_old/cookbook/atomic_caption.md
deleted file mode 100644
index 9443f87a7..000000000
--- a/docs_old/cookbook/atomic_caption.md
+++ /dev/null
@@ -1,193 +0,0 @@
-# Vision-Language Models with Outlines
-This guide demonstrates how to use Outlines with vision-language models, leveraging the outlines `TransformersMultiModal` model. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
-
-We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
-
-## Setup
-First, we need to install the necessary dependencies. In addition to Outlines, we'll need to install the transformers library and any specific requirements for the vision-language model we'll be using.
-
-```bash
-pip install outlines transformers torch
-```
-
-### Initializing the Model
-We'll use the `outlines.from_transformers` function to initialize our vision-language model. For this function to return a vision multi-modal model we need to pass in a transformers model and a transformers processor that can handle both text and image inputs. Today we'll be using the Pixtral model with the AutoProcessor.
-
-```python
-import outlines
-import torch
-from transformers import (
-    AutoProcessor,
-    LlavaForConditionalGeneration
-)
-
-model_name="mistral-community/pixtral-12b" # original magnet model is able to be loaded without issue
-model_class=LlavaForConditionalGeneration
-processor_class=AutoProcessor
-
-def get_vision_model(model_name: str, model_class, processor_class):
-    model_kwargs = {
-        "torch_dtype": torch.bfloat16,
-        "attn_implementation": "flash_attention_2",
-        "device_map": "auto",
-    }
-    processor_kwargs = {
-        "device": "cuda",
-    }
-
-    model = outlines.from_transformers(
-        model_class.from_pretrained(model_name, **model_kwargs),
-        processor_class.from_pretrained(model_name, **processor_kwargs),
-    )
-    return model
-model = get_vision_model(model_name, model_class, processor_class)
-```
-
-### Defining the Schema
-Next, we'll define a schema for the output we expect from our vision multi-modal model. This schema will help structure the model's responses. We use the `outlines.Generator` object to create a generator for our schema that will then be called with our prompt and images.
-
-```python
-from enum import Enum
-from pydantic import BaseModel, Field, confloat, constr
-from pydantic.types import StringConstraints, PositiveFloat
-from typing import List
-from typing_extensions import Annotated
-
-class TagType(Enum):
-    ENTITY = "Entity"
-    RELATIONSHIP = "Relationship"
-    STYLE = "Style"
-    ATTRIBUTE = "Attribute"
-    COMPOSITION = "Composition"
-    CONTEXTUAL = "Contextual"
-    TECHNICAL = "Technical"
-    SEMANTIC = "Semantic"
-
-class ImageTag(BaseModel):
-    tag: Annotated[
-        constr(min_length=1, max_length=30),
-        Field(
-            description=(
-                "Descriptive keyword or phrase representing the tag."
-            )
-        )
-    ]
-    category: TagType
-    confidence: Annotated[
-        confloat(le=1.0),
-        Field(
-            description=(
-                "Confidence score for the tag, between 0 (exclusive) and 1 (inclusive)."
-            )
-        )
-    ]
-
-class ImageData(BaseModel):
-    tags_list: List[ImageTag] = Field(..., min_items=8, max_items=20)
-    short_caption: Annotated[str, StringConstraints(min_length=10, max_length=150)]
-    dense_caption: Annotated[str, StringConstraints(min_length=100, max_length=2048)]
-
-image_data_generator = outlines.Generator(model, ImageData)
-```
-
-This schema defines the structure for image tags, including categories like Entity, Relationship, Style, etc., as well as short and dense captions.
-
-### Preparing the Prompt
-
-We'll create a prompt that instructs the model on how to analyze the image and generate the structured output:
-
-```python
-pixtral_instruction = """
-<s>[INST]
-<Task>You are a structured image analysis agent. Generate comprehensive tag list, caption, and dense caption for an image classification system.</Task>
-<TagCategories requirement="You should generate a minimum of 1 tag for each category." confidence="Confidence score for the tag, between 0 (exclusive) and 1 (inclusive).">
-- Entity : The content of the image, including the objects, people, and other elements.
-- Relationship : The relationships between the entities in the image.
-- Style : The style of the image, including the color, lighting, and other stylistic elements.
-- Attribute : The most important attributes of the entities and relationships in the image.
-- Composition : The composition of the image, including the arrangement of elements.
-- Contextual : The contextual elements of the image, including the background, foreground, and other elements.
-- Technical : The technical elements of the image, including the camera angle, lighting, and other technical details.
-- Semantic : The semantic elements of the image, including the meaning of the image, the symbols, and other semantic details.
-<Examples note="These show the expected format as an abstraction.">
-{
-  "tags_list": [
-    {
-      "tag": "subject 1",
-      "category": "Entity",
-      "confidence": 0.98
-    },
-    {
-      "tag": "subject 2",
-      "category": "Entity",
-      "confidence": 0.95
-    },
-    {
-      "tag": "subject 1 runs from subject 2",
-      "category": "Relationship",
-      "confidence": 0.90
-    },
-   }
-</Examples>
-</TagCategories>
-<ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
-<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG]<image>[/INST]
-""".strip()
-```
-
-This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required. It's essential to include the <image> tag in the prompt at the location where the image will be inserted.
-
-### Generating Structured Output
-Now we can use our model to generate structured output based on an input image:
-
-```python
-from io import BytesIO
-from urllib.request import urlopen
-from PIL import Image
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-image_url="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
-image= img_from_url(image_url)
-result = image_data_generator({
-    "text": pixtral_instruction,
-    "images": image
-})
-print(result)
-```
-
-This code loads an image from a URL, passes it to our vision multi-modal model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
-
-```json
-{'tags_list': [{'tag': 'astronaut',
-   'category': <TagType.ENTITY: 'Entity'>,
-   'confidence': 0.99},
-  {'tag': 'moon', 'category': <TagType.ENTITY: 'Entity'>, 'confidence': 0.98},
-  {'tag': 'space suit',
-   'category': <TagType.ATTRIBUTE: 'Attribute'>,
-   'confidence': 0.97},
-  {'tag': 'lunar module',
-   'category': <TagType.ENTITY: 'Entity'>,
-   'confidence': 0.95},
-  {'tag': 'shadow of astronaut',
-   'category': <TagType.COMPOSITION: 'Composition'>,
-   'confidence': 0.95},
-  {'tag': 'footprints in moon dust',
-   'category': <TagType.CONTEXTUAL: 'Contextual'>,
-   'confidence': 0.93},
-  {'tag': 'low angle shot',
-   'category': <TagType.TECHNICAL: 'Technical'>,
-   'confidence': 0.92},
-  {'tag': 'human first steps on the moon',
-   'category': <TagType.SEMANTIC: 'Semantic'>,
-   'confidence': 0.95}],
- 'short_caption': 'First man on the Moon',
- 'dense_caption': "The figure clad in a pristine white space suit, emblazoned with the American flag, stands powerfully on the moon's desolate and rocky surface. The lunar module, a workhorse of space engineering, looms in the background, its metallic legs sinking slightly into the dust where footprints and tracks from the mission's journey are clearly visible. The photograph captures the astronaut from a low angle, emphasizing his imposing presence against the desolate lunar backdrop. The stark contrast between the blacks and whiteslicks of lost light and shadow adds dramatic depth to this seminal moment in human achievement."}
-```
-
-## Conclusion
-The `TransformersMultiModal` model in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
-
-By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
diff --git a/docs_old/cookbook/chain_of_density.md b/docs_old/cookbook/chain_of_density.md
deleted file mode 100644
index 8a822fc2e..000000000
--- a/docs_old/cookbook/chain_of_density.md
+++ /dev/null
@@ -1,103 +0,0 @@
-# Summarize documents using Chain of Density prompting
-
-A good summary should be informative, concise and clear. While large language models are generally good at summarizing documents, their summaries tend to be long and contain redundant information; their information density tends to be on the lower end. This is where [chain of Density](https://arxiv.org/abs/2309.04269), a new prompting technique, comes in. In this example we will show how one can implement chain of density with a few lines of code using Outlines, leveraging both Outline's prompt templating and its structured generation capabilities.
-
-The article we will try to summarize is the first three paragraphs of the [Alan Turing page on Wikipedia](https://en.wikipedia.org/wiki/Alan_Turing):
-
-```python
-article = """
-Alan Mathison Turing OBE FRS (/ˈtjʊərɪŋ/; 23 June 1912 – 7 June 1954) was an English mathematician, computer scientist, logician, cryptanalyst, philosopher and theoretical biologist.[5] Turing was highly influential in the development of theoretical computer science, providing a formalisation of the concepts of algorithm and computation with the Turing machine, which can be considered a model of a general-purpose computer.[6][7][8] He is widely considered to be the father of theoretical computer science and artificial intelligence.[9]
-
-Born in Maida Vale, London, Turing was raised in southern England. He graduated at King's College, Cambridge, with a degree in mathematics. Whilst he was a fellow at Cambridge, he published a proof demonstrating that some purely mathematical yes–no questions can never be answered by computation. He defined a Turing machine and proved that the halting problem for Turing machines is undecidable. In 1938, he obtained his PhD from the Department of Mathematics at Princeton University. During the Second World War, Turing worked for the Government Code and Cypher School at Bletchley Park, Britain's codebreaking centre that produced Ultra intelligence. For a time he led Hut 8, the section that was responsible for German naval cryptanalysis. Here, he devised a number of techniques for speeding the breaking of German ciphers, including improvements to the pre-war Polish bomba method, an electromechanical machine that could find settings for the Enigma machine. Turing played a crucial role in cracking intercepted coded messages that enabled the Allies to defeat the Axis powers in many crucial engagements, including the Battle of the Atlantic.[10][11]
-
-After the war, Turing worked at the National Physical Laboratory, where he designed the Automatic Computing Engine, one of the first designs for a stored-program computer. In 1948, Turing joined Max Newman's Computing Machine Laboratory at the Victoria University of Manchester, where he helped develop the Manchester computers[12] and became interested in mathematical biology. He wrote a paper on the chemical basis of morphogenesis[1] and predicted oscillating chemical reactions such as the Belousov–Zhabotinsky reaction, first observed in the 1960s. Despite these accomplishments, Turing was never fully recognised in Britain during his lifetime because much of his work was covered by the Official Secrets Act.[13]
-"""
-```
-
-## How Chain Of Density works
-
-Chain Of Density starts with asking the model to generate a first long and non-specific summary. Then it asks the model to generate 4 extra summaries by proceeding in the following way:
-
-1. Identify 1-3 entities missing in the previous summary;
-2. Add all entities marked as missing in the previous step, while not dropping entities;
-3. Make the summary more concise;
-
-The prompt also asks the model to return a list of JSON objects that contain the missing entities and the new summary. This is where structured generation will come in handy :) The paper provides the prompt and an example:
-
-![Figure 2 in the paper](./images/chain_of_density.png)
-
-We can now implement the prompt provided in the paper. We stored the prompt template in a text file, and we can load it using the `Template` class:
-
-```python
-from outlines import Template
-
-chain_of_density = Template.from_file("prompt_templates/chain_of_density.txt")
-```
-
-??? Note
-
-    Note that we modified the prompt slightly so it returns a JSON object that contains the summaries, instead of a list of summaries.
-
-
-## Outlines implementation
-
-We will use Outline's JSON-structured generation to ensure that the model's output is consistent with the format specified in the prompt. We start with defining the JSON objects that the model is asked to return using Pydantic. One JSON object that contains a list of `Summary` objects that contain the missing entities and new summary:
-
-```python
-from pydantic import BaseModel, conlist
-
-class Summary(BaseModel):
-    missing_entities: str
-    denser_summary: str
-
-class Summaries(BaseModel):
-    summaries: conlist(Summary, max_length=5, min_length=5)
-```
-
-We now generate the prompt by passing the article we want to summarize to the prompt template previously loaded. We load a quantized version of Mistral-7B using the AutoAWQ library, and then use the `Summaries` schema to generate the summaries with structured generation:
-
-```python
-import outlines
-import transformers
-
-MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
-
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
-    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
-)
-prompt = chain_of_density(article=article)
-result = model(prompt, Summaries, max_new_tokens=2000)
-```
-
-We can now check the results:
-
-```python
-print(result)
-# {'summaries': [
-#     {
-#       'missing_entities': 'English mathematician, cryptanalyst, philosopher',
-#       'denser_summary': 'Alan Mathison Turing was an English mathematician, cryptanalyst, philosopher.'
-#     },
-#     {
-#       'missing_entities': '',
-#       'denser_summary': "Alan Mathison Turing was an English mathematician who was a crucial figure in WW2's Bletchley Park codebreaking centre and designed one of the first computers."
-#     },
-#     {
-#       'missing_entities': 'cryptanalyst, studied, biology, father',
-#       'denser_summary': 'Alan Mathison Turing was an English cryptanalyst, studied theoretical computer science, and contributed to mathematical biology.'
-#     },
-#     {
-#       'missing_entities': 'biology, morphogenesis, chemical',
-#       'denser_summary': 'Alan Mathison Turing was an English cryptanalyst, studied theoretical computer science, and predicted chemical reactions in morphogenesis.
-#     '},
-#     {
-#       'missing_entities': '',
-#       'denser_summary': 'Alan Mathison Turing was an English cryptanalyst, developed computer science, and made strides in mathematical biology research.'
-#       }
-# ]}
-```
-
-Not bad, considering we used a smallish model to generate the summary! Chain of Density seems to be a very effective prompting technique to generate dense summaries, even with small quantized models. Its implementation in Outlines is also very short.
-
-Note that this is the first article I tried and it worked out of the box. Try it out on other articles, and please share the results on Twitter, or by opening [a new discussion](https://github.com/dottxt-ai/outlines/discussions/categories/show-and-tell) on the Outlines repository!
diff --git a/docs_old/cookbook/chain_of_thought.md b/docs_old/cookbook/chain_of_thought.md
deleted file mode 100644
index 42d6c8550..000000000
--- a/docs_old/cookbook/chain_of_thought.md
+++ /dev/null
@@ -1,127 +0,0 @@
-# Chain of thought
-
-
-Chain of thought is a prompting technique introduced in the paper ["Chain-of-Thought Prompting Elicits Reasoning in Large Language Models"](https://arxiv.org/abs/2201.11903) where throught prompting the authors generate a series of intermediate reasoning steps which improves the ability of LLMs to perform complex reasoning.
-
-In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to apply chain of thought through structured output.
-
-We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
-
-```bash
-pip install llama-cpp-python
-```
-
-To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
-
-```python
-import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
-```
-
-??? note "(Optional) Store the model weights in a custom folder"
-
-    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
-
-    ```bash
-    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
-    ```
-
-    We initialize the model:
-
-    ```python
-    from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
-    ```
-
-## Chain of thought
-
-We first define our Pydantic class for a reasoning step:
-
-```python
-from pydantic import BaseModel, Field
-
-class Reasoning_Step(BaseModel):
-    reasoning_step: str = Field(..., description="Reasoning step")
-```
-
-We then define the Pydantic class for reasoning which will consist on a list of reasoning steps and a conclusion, and we get its JSON schema:
-
-```python
-from typing import List
-
-class Reasoning(BaseModel):
-    reasoning: List[Reasoning_Step] = Field(..., description="List of reasoning steps")
-    conclusion: str = Field(..., description="Conclusion")
-
-json_schema = Reasoning.model_json_schema()
-```
-
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
-
-```python
-from outlines import Template
-
-generate_hermes_prompt = Template.from_string(
-    """
-    <|im_start|>system
-    You are a world class AI model who answers questions in JSON
-    Here's the json schema you must adhere to:
-    <schema>
-    {{ json_schema }}
-    </schema>
-    <|im_end|>
-    <|im_start|>user
-    {{ user_prompt }}
-    <|im_end|>
-    <|im_start|>assistant
-    <schema>
-    """
-)
-```
-
-For a given user prompt:
-
-```python
-user_prompt = "9.11 and 9.9 -- which is bigger?"
-```
-
-We can use `outlines.Generator` with the Pydantic class we previously defined, and call the generator with the Hermes prompt:
-
-```python
-generator = outlines.Generator(model, regex_str)
-prompt = generate_hermes_prompt(json_schema=json_schema, user_prompt=user_prompt)
-response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
-```
-
-We obtain a series of intermediate reasoning steps as well as the conclusion:
-
-```python
-import json
-
-json_response = json.loads(response)
-
-print(json_response["reasoning"])
-print(json_response["conclusion"])
-# [{'reasoning_step': 'Both 9.11 and 9.9 are decimal numbers.'},
-#  {'reasoning_step': 'When comparing decimal numbers, we look at the numbers after the decimal point.'},
-#  {'reasoning_step': 'In this case, 9.11 has the number 1 after the decimal point, while 9.9 has the number 9.'},
-#  {'reasoning_step': 'Since 1 is greater than 9, 9.11 is greater than 9.9.'}]
-# '9.11 is bigger.'
-```
-
-We notice that the 4th reasoning step is wrong ``Since 1 is greater than 9, 9.11 is greater than 9.9.'', so we should probably give the model some examples for this particular task.
-
-This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_old/cookbook/classification.md b/docs_old/cookbook/classification.md
deleted file mode 100644
index 18a933b47..000000000
--- a/docs_old/cookbook/classification.md
+++ /dev/null
@@ -1,85 +0,0 @@
-# Classification
-
-Classification is a classic problem in NLP and finds many applications: spam detection, sentiment analysis, triaging of incoming requests, etc. We will use the example of a company that wants to sort support requests between those that require immediate attention (`URGENT`), those that can wait a little (`STANDARD`). You could easily extend the example by adding new labels.
-
-
-This tutorial shows how one can implement multi-label classification using Outlines.
-
-As always, we start with initializing the model. Since we are GPU poor we will be using a quantized version of Mistal-7B-v0.1:
-
-```python
-import outlines
-import transformers
-
-MODEL_NAME = "TheBloke/Mistral-7B-OpenOrca-AWQ"
-
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
-    transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
-)
-```
-
-We will use a prompt template stored in a text file:
-
-```python
-from outlines import Template
-
-customer_support = Template.from_file("prompt_templates/classification.txt")
-```
-
-## Choosing between multiple choices
-
-Outlines provides a convenient way to do multi-label classification, passing a Literal type hint to the `outlines.Generator` object:
-
-```python
-from typing import Literal
-import outlines
-
-generator = outlines.Generator(model, Literal["URGENT", "STANDARD"])
-
-```
-Outlines supports batched requests, so we will pass two requests to the model:
-
-```python
-requests = [
-    "My hair is one fire! Please help me!!!",
-    "Just wanted to say hi"
-]
-
-prompts = [customer_support(request=request) for request in requests]
-```
-
-We can now ask the model to classify the requests:
-
-```python
-labels = generator(prompts)
-print(labels)
-# ['URGENT', 'STANDARD']
-```
-
-## Using JSON-structured generation
-
-Another (convoluted) way to do multi-label classification is to JSON-structured generation in Outlines. We first need to define our Pydantic schema that contains the labels:
-
-```python
-from enum import Enum
-from pydantic import BaseModel
-
-
-class Label(str, Enum):
-    urgent = "URGENT"
-    standard = "STANDARD"
-
-
-class Classification(BaseModel):
-    label: Label
-```
-
-We can then create a generator with the Pydantic model we just defined and call it:
-
-```python
-generator = outlines.Generator(model, Classification)
-labels = generator(prompts)
-print(labels)
-# ['{"label":"URGENT"}', '{ "label": "STANDARD" }']
-```
diff --git a/docs_old/cookbook/dating_profiles.md b/docs_old/cookbook/dating_profiles.md
deleted file mode 100644
index 5852e6df9..000000000
--- a/docs_old/cookbook/dating_profiles.md
+++ /dev/null
@@ -1,232 +0,0 @@
-# Generate a synthetic dating profile from a description
-
-In this example we will see how we can use Outlines to generate synthetic data for a dating application. This example was originally contributed by [Vibhor Kumar](https://github.com/veezbo).
-
-```python
-import json
-from dataclasses import dataclass
-from enum import Enum
-
-import torch
-import transformers
-from pydantic import BaseModel, conlist, constr
-
-import outlines
-```
-
-## Defining the profile with Pydantic
-
-Here a dating profile will consist in a biography, a job, a list of interests and two question-answer pairs. The questions are written in advance by the team, and the users are asked to provide an answer:
-
-```python
-class QuestionChoice(str, Enum):
-    A = "The key to my heart is"
-    B = "The first item on my bucket list is"
-    C = "Perks of dating me"
-    D = "Message me if you also love"
-    E = "People would describe me as"
-    F = "I can beat you in a game of"
-
-@dataclass
-class QuestionAnswer:
-    question: QuestionChoice
-    answer: str
-```
-
-Users need to provide a short biography, with a minimum of 10 and a maximum of 300 characters. The application also limits job descriptions to 50 characters. In addition to the question-answer pairs, the user is required to provide a list of between 1 and 5 interests:
-
-```python
-class DatingProfile(BaseModel):
-    bio: constr(str, min_length=10, max_length=300)
-    job: constr(str, max_lengt=50)
-    interests: conlist(str, min_length=1, max_length=5)  # type: ignore
-    qna1: QuestionAnswer
-    qna2: QuestionAnswer
-```
-
-## Prompt template and examples
-
-We will ask the model to generate profiles from a high-level description:
-
-```python
-@dataclass
-class Example:
-    description: str
-    profile: DatingProfile
-```
-
-We will use Outlines' prompt templating abilities to generate the prompt for us. This help clearly separate the general prompting logic from what is specific to an example.
-
-```python
-from outlines import Template
-
-dating_profile_prompt = Template.from_string(
-    """
-    You are a world-renowned matchmaker who understands the modern dating
-    market. Your job is to generate dating app profiles for male clients
-    interested in women based on a provided description. The profiles should be
-    authentic, show off their strengths, and maximize their likelihood of
-    getting matches on dating apps.  Here are some examples of past clients that
-    you have successfully created profiles for:
-
-    {% for example in examples %}
-    Description:
-    {{ example.description }}
-    Profile:
-    {{ example.profile }}
-    {% endfor %}
-
-    Here is the new client who you need to create a profile for:
-    Description: {{ description }}
-    Profile:
-    """
-)
-```
-
-We will provide the model with several few-shot examples:
-
-```python
-samples: list[Example] = [
-    Example(
-        description="I'm an author and former professional soccer player living in Seattle who publishes popular fiction books. A typical day for me starts by hanging out with my cat, drinking a coffee, and reading as much as I can in a few hours. Then, I'll prepare a quick smoothie before starting to write for a few hours, take a break with soccer or running a few miles, and finally meet friends for dinner at a new, hip restaurant in the evening. Sometimes we go axe-throwing afterwards, or play poker, or watch a comedy show, or visit a dive bar. On my vacations, I travel extensively to countries South America, Europe, and Asia, with the goal of visiting them all!",
-        profile=DatingProfile(
-            bio="Adventurer, dreamer, author, and soccer enthusiast. Life’s too short to waste time so I make the most of each day by exploring new places and playing with my friends on the pitch. What’s your favorite way to get out and have fun?",
-            job="Famous Soccer Player -> Famous Author",
-            interests=["Soccer", "Travel", "Friends", "Books", "Fluffy Animals"],
-            qna1=QuestionAnswer(
-                question=QuestionChoice.B, answer="swim in all seven oceans!"
-            ),
-            qna2=QuestionAnswer(
-                question=QuestionChoice.E,
-                answer="fun-loving, adventurous, and a little bit crazy",
-            ),
-        ),
-    ),
-    Example(
-        description="I run my company and build houses for a living. I'm a big fan of the outdoors and love to go hiking, camping, and fishing. I don't like video games, but do like to watch movies. My love language is home-cooked food, and I'm looking for someone who isn't afraid to get their hands dirty.",
-        profile=DatingProfile(
-            bio="If you're looking for a Montana man who loves to get outdoors and hunt, and who's in-tune with his masculinity then I'm your guy!",
-            job="House Construction Manager / Entrepreneur",
-            interests=["Hunting", "Hiking", "The outdoors", "Home-cooked food"],
-            qna1=QuestionAnswer(question=QuestionChoice.A, answer="food made at home"),
-            qna2=QuestionAnswer(
-                question=QuestionChoice.C,
-                answer="having a man in your life who can fix anything",
-            ),
-        ),
-    ),
-    Example(
-        description="I run my own Youtube channel with 10M subscribers. I love working with kids, and my audience skews pretty young too. In my free time, I play Fortnite and Roblox. I'm looking for someone who is also a gamer and likes to have fun. I'm learning Japanese in my free time as well as how to cook.",
-        profile=DatingProfile(
-            bio="Easy on the eyes (find me on Youtube!) and great with kids. What more do you need?",
-            job="Youtuber 10M+ subscribers",
-            interests=["Kids", "Gaming", "Japanese"],
-            qna1=QuestionAnswer(question=QuestionChoice.D, answer="anime and gaming!"),
-            qna2=QuestionAnswer(question=QuestionChoice.F, answer="Fortnite, gg ez"),
-        ),
-    ),
-]
-```
-
-## Load the model
-
-We will use Mosaic's MPT-7B model (requires 13GB of GPU memory) which can fit on a single GPU with a reasonable context window. We initialize it with Outlines:
-
-```python
-MODEL_NAME = "mosaicml/mpt-7b-8k-instruct"
-
-config = transformers.AutoConfig.from_pretrained(
-    MODEL_NAME, trust_remote_code=True
-)
-config.init_device = "meta"
-model_kwargs = {
-    "config": config,
-    "trust_remote_code": True,
-    "torch_dtype": torch.bfloat16,
-    "device_map": "cuda",
-}
-tf_model = transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME, **model_kwargs)
-tf_tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
-model = outlines.from_transformers(tf_model, tokenizer=tf_tokenizer)
-```
-
-## JSON-structured generation of profiles
-
-We will now generate a dating profile from a textual description of oneself:
-
-``` python
-new_description = """I'm a laid-back lawyer who spends a lot of his free-time
-gaming. I work in a corporate office, but ended up here after the start-up  I
-cofounded got acquired, so still play ping pong with my cool coworkers every
-day.  I have a bar at home where I make cocktails, which is great for
-entertaining  friends. I secretly like to wear suits and get a new one tailored
-every few  months. I also like weddings because I get to wear those suits, and
-it's  a good excuse for a date. I watch the latest series because I'm paying,
-with my hard-earned money, for every streaming service."""
-
-prompt = dating_profile_prompt(description=new_description, examples=samples)
-profile = model(prompt, DatingProfile)
-parsed_profile = DatingProfile.model_validate_json(json.loads(profile))
-```
-
-## Results
-
-Here are a couple of results:
-
-```json
-{
-    "bio": """I'm an ambitious lawyer with a casual and fashionable style. I love
-    games and sports, but my true passion is preparing refreshing cocktails at
-    home and dressing to the nines at weddings. I'm currently looking for a woman
-    to show a good time to and get a kiss on the opulent suit I just had made.
-    Send resume to this inbox.""",
-    "job": "Lawyer",
-    "interests":
-    [
-        "Stylish guys",
-        "Gaming",
-        "Ping pong",
-        "Cocktails",
-        "Weddings"
-    ],
-    "qna1":
-    {
-        "question": "The first item on my bucket list is",
-        "answer": "be married and have a family."
-    },
-    "qna2":
-    {
-        "question": "People would describe me as",
-        "answer": "charming, stylish, and funny."
-    }
-}
-```
-
-```json
-{
-    "bio": """I’m a sexy lawyer with time on my hands. I love to game and
-    play ping pong, but the real reason you should swipe to the right
-    is because I look great in a suit. Who doesn’t love a man in a
-    suit? Just saying. Send me a message if you think it’s time to take
-    your dating life to the next level.""",
-    "job": "Lawyer",
-    "interests":
-    [
-        "Gaming",
-        "Ping Pong",
-        "Tailored Suits",
-        "Weddings",
-        "Streaming Services"
-    ],
-    "qna1":
-    {
-        "question": "The first item on my bucket list is",
-        "answer": "simulate space but stay alive for as long as possible"
-    },
-    "qna2":
-    {
-        "question": "People would describe me as",
-        "answer": "easy-going, a little nerdy but with a mature essence"
-    }
-}
-```
diff --git a/docs_old/cookbook/deploy-using-bentoml.md b/docs_old/cookbook/deploy-using-bentoml.md
deleted file mode 100644
index 5d000cb52..000000000
--- a/docs_old/cookbook/deploy-using-bentoml.md
+++ /dev/null
@@ -1,232 +0,0 @@
-# Run Outlines using BentoML
-
-[BentoML](https://github.com/bentoml/BentoML) is an open-source model serving library for building performant and scalable AI applications with Python. It comes with tools that you need for serving optimization, model packaging, and production deployment.
-
-In this guide, we will show you how to use BentoML to run programs written with Outlines on GPU locally and in [BentoCloud](https://www.bentoml.com/), an AI Inference Platform for enterprise AI teams. The example source code in this guide is also available in the [examples/bentoml/](https://github.com/dottxt-ai/outlines/blob/main/examples/bentoml/) directory.
-
-## Import a model
-
-First we need to download an LLM (Mistral-7B-v0.1 in this example and you can use any other LLM) and import the model into BentoML's [Model Store](https://docs.bentoml.com/en/latest/guides/model-store.html). Let's install BentoML and other dependencies from PyPi (preferably in a virtual environment):
-
-```bash
-pip install -r requirements.txt
-```
-
-Then save the code snippet below as `import_model.py` and run `python import_model.py`.
-
-**Note**: You need to accept related conditions on [Hugging Face](https://huggingface.co/mistralai/Mistral-7B-v0.1) first to gain access to Mistral-7B-v0.1.
-
-```python
-import bentoml
-
-MODEL_ID = "mistralai/Mistral-7B-v0.1"
-BENTO_MODEL_TAG = MODEL_ID.lower().replace("/", "--")
-
-def import_model(model_id, bento_model_tag):
-
-    import torch
-    from transformers import AutoModelForCausalLM, AutoTokenizer
-
-    tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
-    model = AutoModelForCausalLM.from_pretrained(
-        MODEL_ID,
-        torch_dtype=torch.float16,
-        low_cpu_mem_usage=True,
-    )
-
-    with bentoml.models.create(bento_model_tag) as bento_model_ref:
-        tokenizer.save_pretrained(bento_model_ref.path)
-        model.save_pretrained(bento_model_ref.path)
-
-
-if __name__ == "__main__":
-    import_model(MODEL_ID, BENTO_MODEL_TAG)
-```
-
-You can verify the download is successful by running:
-
-```bash
-$ bentoml models list
-
-Tag                                          Module  Size        Creation Time
-mistralai--mistral-7b-v0.1:m7lmf5ac2cmubnnz          13.49 GiB   2024-04-25 06:52:39
-```
-
-## Define a BentoML Service
-
-As the model is ready, we can define a [BentoML Service](https://docs.bentoml.com/en/latest/guides/services.html) to wrap the capabilities of the model.
-
-We will run the JSON-structured generation example [in the README](https://github.com/dottxt-ai/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
-
-```python
-DEFAULT_SCHEMA = """{
-    "title": "Character",
-    "type": "object",
-    "properties": {
-        "name": {
-            "title": "Name",
-            "maxLength": 10,
-            "type": "string"
-        },
-        "age": {
-            "title": "Age",
-            "type": "integer"
-        },
-        "armor": {"$ref": "#/definitions/Armor"},
-        "weapon": {"$ref": "#/definitions/Weapon"},
-        "strength": {
-            "title": "Strength",
-            "type": "integer"
-        }
-    },
-    "required": ["name", "age", "armor", "weapon", "strength"],
-    "definitions": {
-        "Armor": {
-            "title": "Armor",
-            "description": "An enumeration.",
-            "enum": ["leather", "chainmail", "plate"],
-            "type": "string"
-        },
-        "Weapon": {
-            "title": "Weapon",
-            "description": "An enumeration.",
-            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
-            "type": "string"
-        }
-    }
-}"""
-```
-
-First, we need to define a BentoML service by decorating an ordinary class (`Outlines` here) with `@bentoml.service` decorator. We pass to this decorator some configuration and GPU on which we want this service to run in BentoCloud (here an L4 with 24GB memory):
-
-```python
-import typing as t
-import bentoml
-
-from import_model import BENTO_MODEL_TAG
-
-@bentoml.service(
-    traffic={
-        "timeout": 300,
-    },
-    resources={
-        "gpu": 1,
-        "gpu_type": "nvidia-l4",
-    },
-)
-class Outlines:
-
-    bento_model_ref = bentoml.models.get(BENTO_MODEL_TAG)
-
-    def __init__(self) -> None:
-        import outlines
-        import torch
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-
-        # Load tokenizer and model from the BentoML model reference path
-        hf_tokenizer = AutoTokenizer.from_pretrained(self.bento_model_ref.path)
-        hf_model = AutoModelForCausalLM.from_pretrained(
-            self.bento_model_ref.path,
-            torch_dtype=torch.float16,
-            low_cpu_mem_usage=True,
-            device_map="cuda"
-        )
-
-        # Then use the loaded model with Outlines
-        self.model = outlines.from_transformers(hf_model, hf_tokenizer)
-
-    ...
-```
-
-We then need to define an HTTP endpoint using `@bentoml.api` to decorate the method `generate` of `Outlines` class:
-
-```python
-    ...
-
-    @bentoml.api
-    async def generate(
-        self,
-        prompt: str = "Give me a character description.",
-        json_schema: t.Optional[str] = DEFAULT_SCHEMA,
-    ) -> t.Dict[str, t.Any]:
-        import json
-        import outlines
-        from outlines.types import JsonSchema
-
-        generator = outlines.Generator(self.model, JsonSchema(json_schema))
-        character = generator(prompt)
-
-        return json.loads(character)
-```
-
-Here `@bentoml.api` decorator defines `generate` as an HTTP endpoint that accepts a JSON request body with two fields: `prompt` and `json_schema` (optional, which allows HTTP clients to provide their own JSON schema). The type hints in the function signature will be used to validate incoming JSON requests. You can define as many HTTP endpoints as you want by using `@bentoml.api` to decorate other methods of `Outlines` class.
-
-Now you can save the above code to `service.py` (or use [this implementation](https://github.com/dottxt-ai/outlines/blob/main/examples/bentoml/)), and run the code using the BentoML CLI.
-
-## Run locally for testing and debugging
-
-Then you can run a server locally by:
-
-```bash
-bentoml serve .
-```
-
-The server is now active at <http://localhost:3000>. You can interact with it using the Swagger UI or in other different ways:
-
-<details>
-
-<summary>CURL</summary>
-
-```bash
-curl -X 'POST' \
-  'http://localhost:3000/generate' \
-  -H 'accept: application/json' \
-  -H 'Content-Type: application/json' \
-  -d '{
-  "prompt": "Give me a character description."
-}'
-```
-
-</details>
-
-<details>
-
-<summary>Python client</summary>
-
-```python
-import bentoml
-
-with bentoml.SyncHTTPClient("http://localhost:3000") as client:
-    response = client.generate(
-        prompt="Give me a character description"
-    )
-    print(response)
-```
-
-</details>
-
-Expected output:
-
-```bash
-{
-  "name": "Aura",
-  "age": 15,
-  "armor": "plate",
-  "weapon": "sword",
-  "strength": 20
-}
-```
-
-## Deploy to BentoCloud
-
-After the Service is ready, you can deploy it to [BentoCloud](https://docs.bentoml.com/en/latest/bentocloud/get-started.html) for better management and scalability. [Sign up](https://cloud.bentoml.com/signup) if you haven't got a BentoCloud account.
-
-Make sure you have [logged in to BentoCloud](https://docs.bentoml.com/en/latest/bentocloud/how-tos/manage-access-token.html), then run the following command to deploy it.
-
-```bash
-bentoml deploy .
-```
-
-Once the application is up and running on BentoCloud, you can access it via the exposed URL.
-
-**Note**: For custom deployment in your own infrastructure, use [BentoML to generate an OCI-compliant image](https://docs.bentoml.com/en/latest/guides/containerization.html).
diff --git a/docs_old/cookbook/deploy-using-cerebrium.md b/docs_old/cookbook/deploy-using-cerebrium.md
deleted file mode 100644
index bf010d583..000000000
--- a/docs_old/cookbook/deploy-using-cerebrium.md
+++ /dev/null
@@ -1,126 +0,0 @@
-# Run Outlines using Cerebrium
-
-[Cerebrium](https://www.cerebrium.ai/) is a serverless AI infrastructure platform that makes it easier for companies to build and deploy AI based applications. They offer Serverless GPU's with low cold start times with over 12 varieties of GPU chips that auto scale and you only pay for the compute you use.
-
-In this guide we will show you how you can use Cerebrium to run programs written with Outlines on GPUs in the cloud.
-
-# Setup Cerebrium
-
-First, we install Cerebrium and login to get authenticated.
-
-```bash
-pip install cerebrium
-cerebrium login
-```
-
-Then let us create our first project
-
-```bash
-cerebrium init outlines-project
-```
-
-## Setup Environment and Hardware
-
-You set up your environment and hardware in the cerebrium.toml file that was created using the init function above.
-
-```toml
-[cerebrium.deployment]
-docker_base_image_url = "nvidia/cuda:12.1.1-runtime-ubuntu22.04"
-
-[cerebrium.hardware]
-cpu = 2
-memory = 14.0
-gpu = "AMPERE A10"
-gpu_count = 1
-provider = "aws"
-region = "us-east-1"
-
-[cerebrium.dependencies.pip]
-outline = "==1.0.0"
-transformers = "==4.38.2"
-datasets = "==2.18.0"
-accelerate = "==0.27.2"
-```
-
-## Setup inference
-
-Running code in Cerebrium is like writing normal python with no special syntax. In a `main.py` file specify the following:
-
-```python
-import outlines
-import transformers
-from outlines.types import JsonSchema
-
-
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-schema = """{
-    "title": "Character",
-    "type": "object",
-    "properties": {
-        "name": {
-            "title": "Name",
-            "maxLength": 10,
-            "type": "string"
-        },
-        "age": {
-            "title": "Age",
-            "type": "integer"
-        },
-        "armor": {"$ref": "#/definitions/Armor"},
-        "weapon": {"$ref": "#/definitions/Weapon"},
-        "strength": {
-            "title": "Strength",
-            "type": "integer"
-        }
-    },
-    "required": ["name", "age", "armor", "weapon", "strength"],
-    "definitions": {
-        "Armor": {
-            "title": "Armor",
-            "description": "An enumeration.",
-            "enum": ["leather", "chainmail", "plate"],
-            "type": "string"
-        },
-        "Weapon": {
-            "title": "Weapon",
-            "description": "An enumeration.",
-            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
-            "type": "string"
-        }
-    }
-}"""
-
-generator = outlines.Generator(model, JsonSchema(schema))
-```
-
-On first deploy, it will download the model and store it on disk therefore for subsequent calls it will load the model from disk.
-
-Every function in Cerebrium is callable through an API endpoint. Code at the top most layer (ie: not in a function) is instantiated only when the container is spun up the first time so for subsequent calls, it will simply run the code defined in the function you call.
-
-To deploy an API that creates a new character when called with a prompt you can add the following code to `main.py`:
-
-```python
-def generate(
-    prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
-):
-
-    character = generator(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
-    )
-
-    return character
-```
-
-
-## Run on the cloud
-
-```bash
-cerebrium deploy
-```
-
-You will see your application deploy, install pip packages and download the model. Once completed it will output a CURL request you can use to call your endpoint. Just remember to end
-the url with the function you would like to call - in this case /generate. You should see your response returned!
diff --git a/docs_old/cookbook/deploy-using-modal.md b/docs_old/cookbook/deploy-using-modal.md
deleted file mode 100644
index 23a846a6d..000000000
--- a/docs_old/cookbook/deploy-using-modal.md
+++ /dev/null
@@ -1,206 +0,0 @@
-# Run Outlines using Modal
-
-[Modal](https://modal.com/) is a serverless platform that allows you to easily run code on the cloud, including GPUs. It can come very handy for those of us who don't have a monster GPU at home and want to be able to quickly and easily provision, configure and orchestrate cloud infrastructure.
-
-In this guide we will show you how you can use Modal to run programs written with Outlines on GPU in the cloud.
-
-## Requirements
-
-We recommend installing `modal` and `outlines` in a virtual environment. You can create one with:
-
-```bash
-python -m venv venv
-source venv/bin/activate
-```
-
-Then install the required packages:
-
-```bash
-pip install modal outlines
-```
-
-## Build the image
-
-First we need to define our container image. If you need to access a gated model, you will need to provide an [access token](https://huggingface.co/settings/tokens). See the `.env` call below for how to provide a HuggingFace token.
-
-Setting a token is best done by setting an environment variable `HF_TOKEN` with your token. If you do not wish to do this, we provide a commented-out line in the code to set the token directly in the code.
-
-```python
-from modal import Image, App, gpu
-import os
-
-# This creates a modal App object. Here we set the name to "outlines-app".
-# There are other optional parameters like modal secrets, schedules, etc.
-# See the documentation here: https://modal.com/docs/reference/modal.App
-app = App(name="outlines-app")
-
-# Specify a language model to use.
-# Another good model to use is "NousResearch/Hermes-2-Pro-Mistral-7B"
-language_model = "mistral-community/Mistral-7B-v0.2"
-
-# Please set an environment variable HF_TOKEN with your Hugging Face API token.
-# The code below (the .env({...}) part) will copy the token from your local
-# environment to the container.
-# More info on Image here: https://modal.com/docs/reference/modal.Image
-outlines_image = Image.debian_slim(python_version="3.11").pip_install(
-    "outlines",
-    "transformers",
-    "datasets",
-    "accelerate",
-    "sentencepiece",
-).env({
-    # This will pull in your HF_TOKEN environment variable if you have one.
-    'HF_TOKEN':os.environ['HF_TOKEN']
-
-    # To set the token directly in the code, uncomment the line below and replace
-    # 'YOUR_TOKEN' with the HuggingFace access token.
-    # 'HF_TOKEN':'YOUR_TOKEN'
-})
-```
-
-## Setting the container up
-
-When running longer Modal apps, it's recommended to download your language model when the container starts, rather than when the function is called. This will cache the model for future runs.
-
-```python
-# This function imports the model from Hugging Face. The modal container
-# will call this function when it starts up. This is useful for
-# downloading models, setting up environment variables, etc.
-def import_model():
-    import outlines
-    import transformers
-
-    outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(language_model),
-        transformers.AutoTokenizer.from_pretrained(language_model)
-    )
-
-# This line tells the container to run the import_model function when it starts.
-outlines_image = outlines_image.run_function(import_model)
-```
-
-## Define a schema
-
-We will run the JSON-structured generation example [in the README](https://github.com/dottxt-ai/outlines?tab=readme-ov-file#efficient-json-generation-following-a-json-schema), with the following schema:
-
-```python
-# Specify a schema for the character description. In this case,
-# we want to generate a character with a name, age, armor, weapon, and strength.
-schema = """{
-    "title": "Character",
-    "type": "object",
-    "properties": {
-        "name": {
-            "title": "Name",
-            "maxLength": 10,
-            "type": "string"
-        },
-        "age": {
-            "title": "Age",
-            "type": "integer"
-        },
-        "armor": {"$ref": "#/definitions/Armor"},
-        "weapon": {"$ref": "#/definitions/Weapon"},
-        "strength": {
-            "title": "Strength",
-            "type": "integer"
-        }
-    },
-    "required": ["name", "age", "armor", "weapon", "strength"],
-    "definitions": {
-        "Armor": {
-            "title": "Armor",
-            "description": "An enumeration.",
-            "enum": ["leather", "chainmail", "plate"],
-            "type": "string"
-        },
-        "Weapon": {
-            "title": "Weapon",
-            "description": "An enumeration.",
-            "enum": ["sword", "axe", "mace", "spear", "bow", "crossbow"],
-            "type": "string"
-        }
-    }
-}"""
-```
-
-To make the inference work on Modal we need to wrap the corresponding function in a `@app.function` decorator. We pass to this decorator the image and GPU on which we want this function to run.
-
-Let's choose an A100 with 80GB memory. Valid GPUs can be found [here](https://modal.com/docs/reference/modal.gpu).
-
-```python
-# Define a function that uses the image we chose, and specify the GPU
-# and memory we want to use.
-@app.function(image=outlines_image, gpu=gpu.A100(size='80GB'))
-def generate(
-    prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
-):
-    # Remember, this function is being executed in the container,
-    # so we need to import the necessary libraries here. You should
-    # do this with any other libraries you might need.
-    import outlines
-    import transformers
-    from outlines.types import JsonSchema
-
-    # Load the model into memory. The import_model function above
-    # should have already downloaded the model, so this call
-    # only loads the model into GPU memory.
-    outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(language_model, device_map="cuda"),
-        transformers.AutoTokenizer.from_pretrained(language_model)
-    )
-
-    # Generate a character description based on the prompt.
-    # We use the .json generation method -- we provide the
-    # - model: the model we loaded above
-    # - schema: the JSON schema we defined above
-    generator = outlines.Generator(model, JsonSchema(schema))
-
-    # Make sure you wrap your prompt in instruction tags ([INST] and [/INST])
-    # to indicate that the prompt is an instruction. Instruction tags can vary
-    # by models, so make sure to check the model's documentation.
-    character = generator(
-        f"<s>[INST]Give me a character description. Describe {prompt}.[/INST]"
-    )
-
-    # Print out the generated character.
-    print(character)
-```
-
-We then need to define a `local_entrypoint` to call our function `generate` remotely.
-
-```python
-@app.local_entrypoint()
-def main(
-    prompt: str = "Amiri, a 53 year old warrior woman with a sword and leather armor.",
-):
-    # We use the "generate" function defined above -- note too that we are calling
-    # .remote() on the function. This tells modal to run the function in our cloud
-    # machine. If you want to run the function locally, you can call .local() instead,
-    # though this will require additional setup.
-    generate.remote(prompt)
-```
-
-Here `@app.local_entrypoint()` decorator defines `main` as the function to start from locally when using the Modal CLI. You can save above code to `example.py` (or use [this implementation](https://github.com/dottxt-ai/outlines/blob/main/examples/modal_example.py)). Let's now see how to run the code on the cloud using the Modal CLI.
-
-## Run on the cloud
-
-First install the Modal client from PyPi, if you have not already:
-
-```bash
-pip install modal
-```
-
-You then need to obtain a token from Modal. Run the following command:
-
-```bash
-modal setup
-```
-
-Once that is set you can run inference on the cloud using:
-
-```bash
-modal run example.py
-```
-
-You should see the Modal app initialize, and soon after see the result of the `print` function in your terminal. That's it!
diff --git a/docs_old/cookbook/earnings-reports.md b/docs_old/cookbook/earnings-reports.md
deleted file mode 100644
index cca1e9513..000000000
--- a/docs_old/cookbook/earnings-reports.md
+++ /dev/null
@@ -1,276 +0,0 @@
-# Extracting financial data from earnings reports
-
-A common task in finance is to extract financial data from earnings reports. Earnings reports are infamously poorly formatted, as the SEC does not have requirements for producing machine-readable documents.
-
-Earnings reports are often provided as HTML documents, which can be difficult to parse. Investors often use complicated parsing systems or manual review to extract data. Entire companies are built around automating this task.
-
-This cookbook is a proof of concept about how we can use LLMs to extract financial data directly into CSV. Comma-separated values are well-structured and can be defined by a regular expression, which Outlines can use to guide the LLM's output.
-
-The example is a smaller subset of a full demo found [here](https://github.com/dottxt-ai/demos/tree/main/earnings-reports). The demo contains the full set of pre-processing steps needed to convert raw HTML into a structured CSV file, and tests the results across three company's 10k reports.
-
-## Setup
-
-Install outlines and required dependencies:
-
-```bash
-# Later versions of torch can have difficulty with certain CUDA drivers.
-# We recommend using 2.4.0 for now, but you may wish to experiment with
-# other versions.
-pip install outlines pandas transformers torch==2.4.0 accelerate
-```
-
-## Load the model
-
-Choose your language model. We'll use Phi-3 mini, which is small enough to run on reasonably small machines.
-
-```python
-import outlines
-import torch
-import transformers
-
-model_name = 'microsoft/Phi-3-mini-4k-instruct'
-tf_model = transformers.AutoModelForCausalLM.from_pretrained(
-    model_name, device_map="cuda", torch_dtype=torch.bfloat16
-)
-tf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
-model = outlines.from_transformers(tf_model, tf_tokenizer)
-```
-
-## Set up the data
-
-For brevity, we've attached the markdown version of Nvidia's 10k report. The [full demonstration](https://github.com/dottxt-ai/demos/tree/main/earnings-reports) processes the raw HTML version of the report to these markdown tables. Pages are filtered by whether they seem to contain income statements, and then compacted into the string you see below.
-
-```python
-income_statement = """
-Table of ContentsNVIDIA Corporation and SubsidiariesConsolidated Statements of Income(In millions, except per share data)
-
-|  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |
-| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
-|  | | | Year Ended | | | | | | | | | | | | | | |
-|  | | | Jan 28, 2024 | | |  | | | Jan 29, 2023 | | |  | | | Jan 30, 2022 | | |
-| Revenue | | | $ | 60,922 |  |  | | | $ | 26,974 |  |  | | | $ | 26,914 |  |
-| Cost of revenue | | | 16,621 | |  |  | | | 11,618 | |  |  | | | 9,439 | |  |
-| Gross profit | | | 44,301 | |  |  | | | 15,356 | |  |  | | | 17,475 | |  |
-| Operating expenses | | |  | | |  | | |  | | |  | | |  | | |
-| Research and development | | | 8,675 | |  |  | | | 7,339 | |  |  | | | 5,268 | |  |
-| Sales, general and administrative | | | 2,654 | |  |  | | | 2,440 | |  |  | | | 2,166 | |  |
-| Acquisition termination cost | | | — | |  |  | | | 1,353 | |  |  | | | — | |  |
-| Total operating expenses | | | 11,329 | |  |  | | | 11,132 | |  |  | | | 7,434 | |  |
-| Operating income | | | 32,972 | |  |  | | | 4,224 | |  |  | | | 10,041 | |  |
-| Interest income | | | 866 | |  |  | | | 267 | |  |  | | | 29 | |  |
-| Interest expense | | | (257) | |  |  | | | (262) | |  |  | | | (236) | |  |
-| Other, net | | | 237 | |  |  | | | (48) | |  |  | | | 107 | |  |
-| Other income (expense), net | | | 846 | |  |  | | | (43) | |  |  | | | (100) | |  |
-| Income before income tax | | | 33,818 | |  |  | | | 4,181 | |  |  | | | 9,941 | |  |
-| Income tax expense (benefit) | | | 4,058 | |  |  | | | (187) | |  |  | | | 189 | |  |
-| Net income | | | $ | 29,760 |  |  | | | $ | 4,368 |  |  | | | $ | 9,752 |  |
-|  | | |  | | |  | | |  | | |  | | |  | | |
-| Net income per share: | | |  | | |  | | |  | | |  | | |  | | |
-| Basic | | | $ | 12\.05 |  |  | | | $ | 1\.76 |  |  | | | $ | 3\.91 |  |
-| Diluted | | | $ | 11\.93 |  |  | | | $ | 1\.74 |  |  | | | $ | 3\.85 |  |
-|  | | |  | | |  | | |  | | |  | | |  | | |
-| Weighted average shares used in per share computation: | | |  | | |  | | |  | | |  | | |  | | |
-| Basic | | | 2,469 | |  |  | | | 2,487 | |  |  | | | 2,496 | |  |
-| Diluted | | | 2,494 | |  |  | | | 2,507 | |  |  | | | 2,535 | |  |
-"""
-```
-
-The markdown tables extracted from the earnings reports can vary widely in row names, column counts, data types, etc. The advantage of LLMs here is that we can define the data we want in terms of the data types, and the LLM will output the data in the desired format.
-
-For comparison, here is how the income statement looks in the original HTML:
-
-![Nvidia income statement](./images/nvidia-income.png)
-
-## Define the data we want
-
-Outlines is often used for JSON output, but it can also be used for CSV. We know the columns we want to extract, and we know the data types of the columns. Year for example is always a four-digit number, revenue is a number with commas, and so on.
-
-We can define a regex pattern for each column type:
-
-```python
-# Define the column type regex patterns
-column_types = {
-    # Year is always a four-digit number
-    "year": r"\d{4}",
-
-    # Revenue, operating income, and net income are always numbers with commas.
-    # This regex permits integers that may begin with a minus sign, and may have
-    # commas separating the thousands, millions, etc.
-    "integer_comma": r"((-?\d+),?\d+|(-?\d+))",
-    # Number is currently not used, but it represents a number with up to two decimal places.
-    "number": r"(-?\d+(?:\.\d{1,2})?)",
-}
-```
-
-Next, let's choose the columns we want to extract. We want
-
-- Year, always a four-digit number
-- Revenue, a number with commas
-- Operating income, a number with commas
-- Net income, a number with commas
-
-```python
-# Define the columns to extract, and their data types.
-columns_to_extract = {
-    "year": "year",
-    "revenue": "integer_comma",
-    "operating_income": "integer_comma",
-    "net_income": "integer_comma",
-}
-```
-
-You can modify `column_type_regex` to match the data types of the columns you want to extract.  Adding a new financial metric to extract is as simple as adding a new key/value pair to `columns_to_extract`:
-
-```python
-columns_to_extract["diluted_earnings_per_share"] = "number"
-```
-
-Additional columns are not well tested for accuracy, so use with caution.
-
-## Create the regex describing the data we want
-
-
-```python
-# Create the header line. This is the requested column names
-# separated by commas, i.e. "year,revenue,..."
-header = ",".join(columns_to_extract.keys())
-
-# Create the data capture patterns. These are the regex patterns
-# that will be used to capture the data in each column
-data_patterns = [column_types[dtype] for dtype in columns_to_extract.values()]
-data_line = ",".join(data_patterns)
-
-# Our final regex pattern.
-max_rows = 3 # We expect 3 rows of data, firms usually report 3 years of income statements
-csv_regex = f"{header}(\n{data_line}){{,{max_rows}}}\n\n"
-
-print(csv_regex)
-```
-
-which gives us
-
-```
-year,revenue,operating_income,net_income,basic_earnings_per_share(
-\d{4},((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),((-?\d+),?\d+|(-?\d+)),(-?\d+(?:\.\d{1,2})?)){,3}
-```
-
-Pretty hairy, right? Thankfully, we have a simple function to construct this regex for you. The regex defines a header line, followed by a data line that repeats for each row of data we want to extract. Passing the regex to `outlines.Generator` will produce a function that will __always__ produce a CSV string that is consistent with the regex.
-
-## Prompting the model
-
-Outlines does not add system or instruction tokens by default, so we need to use `transformers.AutoTokenizer` to add them for whatever model we're using.
-
-```python
-from transformers import AutoTokenizer
-
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-
-def add_instruction(prompt):
-    return tokenizer.apply_chat_template([{"role": "user", "content": prompt}], tokenize=False, add_generation_prompt=True)
-
-print(add_instruction("Howdy"))
-```
-```
-<|user|>
-Howdy<|end|>
-<|assistant|>
-```
-
-Our prompt roughly describes the task we want the model to perform, and a few pieces of information it may need to know about income statements.
-
-```python
-def extract_financial_data_prompt(columns_to_extract, income_statement):
-    user_prompt = f"""
-    Extract annual financial data from this set of pages. Pages
-    are from a 10k filing and were chosen because they may contain
-    a comprehensive income statement. Note that selected pages may
-    be incorrectly extracted, so you should verify that you are extracting
-    from the comprehensive income statement and not some other financial
-    statement.
-
-    Create a row for each year available in the income statement with the
-    following columns: {', '.join(columns_to_extract.keys())}. Firms typically report the
-    most recent 3 years of data, but this can vary.
-
-    Each column has types: {', '.join(columns_to_extract.values())}.
-
-    # Relevant pages:
-
-    {income_statement}
-
-    # Key instructions:
-
-    1. Look ONLY at the "Consolidated Statements of Income" table
-    2. For operating income, look for "Income from operations" or "Operating income"
-    3. For net income, use the TOTAL net income figure, not amounts allocated to specific share classes
-    4. Use NULL for missing values
-    5. Operating income must be less than revenue
-    6. Net income must be less than operating income
-    7. Ignore segment breakdowns, quarterly data, or per-share amounts
-
-    # Output format:
-
-    - CSV format with headers: {','.join(columns_to_extract.keys())}
-    - Use NULL for missing values
-    - If no data are found, do not create a row.
-    - Enter two newline characters to terminate the CSV when no more data are found.
-
-    # Definitions:
-    - Revenue: Total sales of goods and services. Usually this is at the top of the
-    income statement.
-    - Operating income: Revenue minus operating expenses for the entire company. This is revenue
-    minus costs. Operating income is also called operating profit, EBIT, or income from
-    operations.
-    - Net income: Operating income minus taxes. This is the bottom line of the
-    income statement.
-    """
-
-    return add_instruction(user_prompt)
-```
-
-## Running the model
-
-Now that we have our prompt and regular expression, we can run the model.
-
-Construct our regex extractor function.
-
-```python
-from outlines.types import Regex
-
-csv_extractor = outlines.Generator(model, Regex(csv_regex))
-```
-
-Provide the prompt to the model and run it:
-
-```python
-csv_data = csv_extractor(
-    extract_financial_data_prompt(columns_to_extract, income_statement),
-    max_new_tokens=1024,
-)
-
-print(csv_data)
-```
-```
-year,revenue,operating_income,net_income
-2024,60922,32972,29760
-2023,26974,4224,4368
-2022,26914,10041,9752
-```
-
-Voila! We've extracted the financial data from the income statement, and it's correct upon inspection.
-
-You can even load this into a `pandas` DataFrame for further analysis:
-
-```python
-import pandas as pd
-from io import StringIO
-
-df = pd.read_csv(StringIO(csv_data))
-print(df)
-```
-```
-   year  revenue  operating_income  net_income
-0  2024    60922             32972       29760
-1  2023    26974              4224        4368
-2  2022    26914             10041        9752
-```
diff --git a/docs_old/cookbook/extract_event_details.md b/docs_old/cookbook/extract_event_details.md
deleted file mode 100644
index 063433ad2..000000000
--- a/docs_old/cookbook/extract_event_details.md
+++ /dev/null
@@ -1,34 +0,0 @@
-This recipe demonstrates how to use the `outlines` library to extract structured event details from a text message.
-We will extract the title, location, and start date and time from messages like the following:
-
-```plaintext
-Hello Kitty, my grandmother will be here, I think it's better to postpone
-our appointment to review math lessons to next Monday at 2pm at the same
-place, 3 avenue des tanneurs, one hour will be enough see you 😘
-```
-
-Let see how to extract the event details from the message with the MLX
-library dedicated to Apple Silicon processor (M series).
-
-```python
---8<-- "docs/cookbook/extract_event_details.py"
-```
-
-The output will be:
-
-```plaintext
-Today: Saturday 16 November 2024 and it's 10:55
-```
-
-and the extracted event information will be:
-
-```json
-{
-  "title":"Math Review",
-  "location":"3 avenue des tanneurs",
-  "start":"2024-11-22T14:00:00Z"
-}
-```
-
-
-To find out more about this use case, we recommend the project developped by [Joseph Rudoler](https://x.com/JRudoler) the [ICS Generator](https://github.com/jrudoler/ics-generator)
diff --git a/docs_old/cookbook/extraction.md b/docs_old/cookbook/extraction.md
deleted file mode 100644
index cfb39ca5c..000000000
--- a/docs_old/cookbook/extraction.md
+++ /dev/null
@@ -1,94 +0,0 @@
-# Named entity extraction
-
-Named Entity Extraction is a fundamental problem in NLP. It involves identifying and categorizing named entities within a document: people, organization, dates, places, etc. It is usually the first step in a more complex NLP worklow. Here we will use the example of a pizza restaurant that receives orders via their website and need to identify the number and types of pizzas that are being ordered.
-
-Getting LLMs to output the extracted entities in a structured format can be challenging. In this tutorial we will see how we can use Outlines' JSON-structured generation to extract entities from a document and return them in a valid JSON data structure 100% of the time.
-
-As always, we start with initializing the model. We will be using a quantized version of Mistal-7B-v0.1 (we're GPU poor):
-
-```python
-import transformers
-import outlines
-
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name, device_map="cuda"),
-    transformers.AutoTokenizer.from_pretrained(model_name),
-)
-```
-
-And we will be using the following prompt template:
-
-```python
-from outlines import Template
-
-take_order = Template.from_string(
-    """You are the owner of a pizza parlor. Customers \
-    send you orders from which you need to extract:
-
-    1. The pizza that is ordered
-    2. The number of pizzas
-
-    # EXAMPLE
-
-    ORDER: I would like one Margherita pizza
-    RESULT: {"pizza": "Margherita", "number": 1}
-
-    # OUTPUT INSTRUCTIONS
-
-    Answer in valid JSON. Here are the different objects relevant for the output:
-
-    Order:
-        pizza (str): name of the pizza
-        number (int): number of pizzas
-
-    Return a valid JSON of type "Order"
-
-    # OUTPUT
-
-    ORDER: {{ order }}
-    RESULT: """
-)
-```
-
-We now define our data model using Pydantic:
-
-```python
-from enum import Enum
-from pydantic import BaseModel
-
-class Pizza(str, Enum):
-    margherita = "Margherita"
-    pepperonni = "Pepperoni"
-    calzone = "Calzone"
-
-class Order(BaseModel):
-    pizza: Pizza
-    number: int
-```
-
-We can now define our generator and call it on several incoming orders:
-
-```python
-orders = [
-    "Hi! I would like to order two pepperonni pizzas and would like them in 30mins.",
-    "Is it possible to get 12 margheritas?"
-]
-prompts = [take_order(order=order) for order in orders]
-
-generator = outlines.Generator(model, Order)
-
-results = generator(prompts)
-print(results)
-# ['{"pizza": "Pepperoni", "number": 2}',
-# '{"pizza": "Margherita", "number": 12}']
-```
-
-There are several ways you could improve this example:
-
-- Clients may order several types of pizzas.
-- Clients may order drinks as well.
-- If the pizza place has a delivery service we need to extract the client's address and phone number
-- Clients may specify the time for which they want the pizza. We could then check against a queuing system and reply to them with the estimated delivery time.
-
-How would you change the Pydantic model to account for these use cases?
diff --git a/docs_old/cookbook/images/chain_of_density.png b/docs_old/cookbook/images/chain_of_density.png
deleted file mode 100644
index 61e01f40459e26f082fc90ea2964dcc4bd2d70c6..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 515603
zcmb@uV{~L)7d2X8$LVwj9ox2T+qTuQZFQ3F*k;GJla6iMw(fbpJH8+H{e9OsBQ?gU
zQ>UtS?Y-7qbIzRz1vzm<I4n322!tppA)*8VL2ZLT5Iiv8z#09CHwfSX!bwO{1qKFY
zbz6Q51R?-QiU_K>ubgGN=%B7{4o%Y0tmppD`RyMZOdAg7YPj3b;TmNW<$2!h>ZNKU
zIv^tWRV4mvL0en8saH0?St>6#Hcj*J`?~o?)w{8o>yAv86W<|ruwuR#7|#DhhMj2I
zfKf<ru(KPZ!~evGzVO)v_}2dh$Pzq2fZ%Iu+YgTaiO;H@bBq7Ei9;*CPJYk-#IdhK
z4D^@(Zxp}wuwwo1r@Q#F`2X__#ar9p{by~bEbjK^>hpG#po0i$X=$mc^Ybd!Mg`>6
zi;L@7J`Dc9-}z1LLs!=^&O`d-6#Lh4VHD423is_}a*y@lNAnYjiRWDSwwJ5T@kKe2
zM_RhJTy`4{7M@`K^wVc1ukp!X19fX#^k;d+W-T)*Ekr8bA)D1}lFPbo-d`fJhvLU9
z;o*9O2<P`1%s#!BbzKTr1y$7!lX)?w7PQp$Hm6gGWJBQWe*=@@nlocTy=X^ELnml<
zIGI{wbN#mI>O9Fls%P3ITr~8frSry}<{*?jfAZOzJ5t7a#2URq>i8sYp?(2NuR+Kh
zU%r!f+4;@BP2rQ#I}yJUX(f7s&6bMvTks2oj29<8#b;f2SLYbK!(7efMV)2#%galf
z+ohQJe{pztI65*yNkvswR(5iHjLZ4cWcU~1x~@RRW<Xrr<VYep7$^#ZzSH?w$qhps
zT!G@ZX*v6!Kk7tv(&%HlTbylPQFq)dSZwLKnyTwFLS*p;taI~dXlTKel7$K-DpXA4
zIkRp1kqVY9^$7KLRKm$hs=9TqiEfgXnr;|-U9gCrzDEd-+1a6?V(96zBwtljRa;*X
z`U8b>ad<tDcxRz*%_1nL6iVlABB{s{O9|+UD=SG!N$1y#14Xs9t*`N&K06)IQh7ZN
z^71<>BWr)UC7*tB)1a5<*1&=U5|Y%El+G@8-X1M23Dp<ps+Vm<p`d7KWx)fc)=3k1
z5XkD<+XKb&M@B~<&vIo+Tq_?RKNpzSR-|$jN<kn}QK)YbbL#^rQEiWS=KFg@voD^l
ze(n=5nc~Gvr6yw5t1Pl_h{rTst%Rw9ABdu+b<0O6U?3T`+eh6Zx}X8G`T7NLkV%0=
zS#4R@c{2hVL}yA~E|X=inpurzSLIkH84*$M>}*|54oG)q?Z41bRE$qeEiEXRTZig@
zmK6~fPaH7;`TzMdPk(u)ACQr;7RpN9=zb|`X1~^X$>bd#Z&+nVql96_LW53C#B?xF
z?lw|tNlTldB_1h(!_nPiTYq{KPd7u<W^Uuku9x)5G?~>E2!(~{;_}5y8_I(dD^t$0
zP-SsJQiRO?o^5M-7USw0k)o8IoSLNMNN3mh6=9G}wAUv&_IqYdZm#1@^fNHF=E&z5
zahJ|+j{Uut`{|MYs7$n!g?%>8j`J3)%I)3n58jI>_2J;6G->37MTB*I#wd~{ch_a3
zy`;+wMoN$UaJdob3BQK_{8@W*bEAfS&=-bK-`JR7C;uaj&Esvhaj_t`=DKJ1u>4bJ
z4K4ZlY`{8xn*S+TS~?LG4XuJ^?EKv3WmT)ZWD5=@JW77;r~w<ab;5IZ6|lapoAF6Q
zT4*|{*Z%Rq+ge|Nqic~<b;!5kvg6L4nZ--ulH(uO<A#6r1B<{VI4{r0kNb367&<i-
z%S{ot$AC$=%iHVvL|4m`%u1##hrr(2meu^=4lMS)?Wp-WV&t^x3=+hFjO~3kwegoD
z7?RWT#>T;d1GF`E9TWu3;p*nNQe0MsW#>|%)2N^(R<vH}-HGe-`vD=76x933&sREH
z>XrwxnHn*per;-U<JMI{nNvK0-oyvS0~E+3sTW@`B|rPK_V{T!lz30_M%PU#H)K0M
zTMi85o+9hZcka80d-O@Wy}eB=9&5ss3i6-M7e~fslhN1DmMfmMs2?00yt(-a4<B_w
zM?7J_b2ynj9Nh8#>3^M^f`GNfFg3#WG9(WIC62Vajy_5>HL|ry<y%@<thpUM#WlN+
z{UnRJ=RFyEEWytf>P5O(C#0t*w?Wob3<>W`<4<N`V<nu$I>{_8{v4k=zEij7&#7}|
zd!NAJrywVPgl|KYF*a6KbS&-e^6ekkd^*eu=-dcp^+UpDs&1o?2wtv!m^?T@H__6T
zR?d(mA~IH3-th}8IyiWZdz)G&h36IJ7ea|={DPEiV`tezQ&m~+_3}0!rGEL~ZtseH
zP8DCPNU36otS4l2lV41)qVhYL?(QyyOEDNWuD<>*(iHv?pDT8Z&*#G`X$X~87BND9
zByEMuun-DFQe436^C1-~vQqCm|2x9?Gut{%zti{fg6M6?ux^=Nzl+;z8!PeBNkvEL
zCGJv~|C62kW5wK<HdV0hAj!-S+3##JSlbQn==h{Ux5YX5R$+wi`5qCm?GZmhUS^Pn
zJW-oC-xxU7#oI2H8Wl4xe8iveJwq*7UVx$O>df98IroZFTs3!MPK+lGZEAy!8gKn`
zq`fwFi_<xQq+xQtyl#Hm^(ArOU&ScO(_&LX$!Ub!Ug7=%cf}fezqVqLboB7sD$cks
z^F@caPs%e6FrO4R%bW1zr;mfnNW?c$)|C$Ay3Oj`&R0TUSdj7gGT80)+>&96R}Jcx
zff%lD_r`#$>dO8zib05l2qK{2A)@7-o#}0FgPSEy>QJF_30kD{4S*&iVlta;Tt_HU
zfRJV@_hP}4hQAerz=_XJEp=@AoqJkRW$)kiua2k4IYfXMX@InF)eWD+?|JYNU+_ff
zMN=zhGR`r3W5?!ZfAv$Q4|jWdHd%Ije7OFv+~mmjHPH$jwC_!}$;e#-LA?eH-bmTr
zkAz2$sGT)6$G|GrZhy>;)dI(4HEJ%E#5YM3%U|!Hw=)5Y5YC^K&uF<?yk=fk_4k9Y
zVdH+&HgQYd;h&g2tO5#iFX~?mb~d)k%~oc@($OGOg5ODhE+065o}7`wE}$TgIh8;1
z{rBY<IPR{lyGVVxY}j5$KPQ-RFtP>g&=!4mHw7dumscP0r<ebXqDStwpzESXp;&-G
zlg#=qkIQfDmZl1FBH2z?E_Aj@?Nz+#;}4UbtF(uvITieklRxgrwUm>Hq#XT%{W7bX
zU;=}Jx_sZ^Kn_k$eNbtihhJeJ2iRIy#GlxW4ebu#26Eb6Pwm*zi=_Xa&8NH7Io)iL
z$|1df>a-esWMGz2>F}rR%>k<mJosEh>c#JP_r44K54%Kl=%76lYHi)|p@DsrtEXbt
z)AK8z_e!bmhj@-ssUoA}x^5&=w^7Xo6l7jq;$U0EF!;jUn)lVPq<cg~#o5@gvKpN>
zORon-`&;jRIwE4sH~LCPd}8vjk9V3^sp|zI{+%6Q1rg?5ij#aRmca>|@O`^W6+2xy
zQh@~TnckQHj@&NW><zIL*Nw35Uj=(}bhO=Fsbdux|H@7-=mz{!*V^Bt7e4}$OUlb>
zuQD2pmW8;}eUF5=G96#mOi_ZUGY4C<3-dd0Kbylp)BUkO*AnLx@pvAjC&$FhVIg}9
zt}m#QXr1@paiIJ@zP;?Ho5(Y8lUL>CDeWv)Mm=RGia_u|=ieAj;;ler;`A^7qwmAd
zbY<M)fxy2F#4gP&jTXUH2w%BsT9t5ZE!mkBpC2B#*mHRAks`q>PaSK`V3aSapCPRP
z4XTr&v;J|umwfjlqX=jI)o3+ec0o7Vh0B)yj*sp8;WscRxdu96thwG_S9g@&SS<b9
z+4Fht<(wozg11!XvlqTCLAuB3Hkh}6T4rq(?`Qw!WPg?t4<2+#m^A3(a*v$F^9hun
ziuMeN-zUa%)(>Nr`MQ}QF7UHVwL(IDyF1@+;o***LlGiOOy;cnuX2sWT7cvNvqr*A
ze%W{_9N@RCp*@~S$}uG-Vsh68n)otWo!s+pKMDi_GUSK-#%Hv=RiOeZ8mf+N=y)b;
zhrGRTg#O-6!i*c3AfciD>B%W2YFS1`MQH0vHDQog8<)1`!$5_OuJY^SGxab_%OCJh
zm-nP9Ed9Gw*;6$uYir5I?zTpIwy8xZ(B3`<-?i8C0dG^H-|A!TY6%MXrJSJQH1L~G
z>US>8Ehv_<cyMxlPvnEZeU;B<is!dLYH7H?V*Fd#bAVXhz`{yJRer(ijgEq*tf?tV
z6hAgDb1;#1!DLe{z>mR}j*EhM%SyyV%UhmP9WGWZjS``?qP3ug4uN|<6oX645G$$W
zdGIbi=}&-jc5Wt<PI6U+&h$Kd-311U86PhZ6xa3XyLb=Vo(D=mN0Xg^IX<sf>e_;0
za(w)9opDt6cUU;2eOCsUVBJ-AN-7w{xpl>FiAqzC7$63--xSj8K;4=X{udR}ay%{U
z<yFtCTea!K<NLzHv_7}?7I&FHeyVQt&~3n!Z@jbN10y9N1z<-AHf=q<jpn#%ip6bW
z-O%##@(8-~K##ihp=NLpFa`9xUL77kzC#BEyrJQtVB(=1Gs##W1u7DUe)?=vDNQ%l
z|H?Nq$_ztHnls%Kms5LqJ0?D|adH|RC}-bh-`}3a5GQ0$EJ;~fwDpbR^?F#UK9C0q
z2n+A_71MY6#L>}F8rB6=R75u>9UxW3u)WhTh8fo_Q(9tPf`fO7zgyVZ<ud;OD!*qd
z*WZagu{h_2#n}*bI|~aQ#{+Pt4#V@vqM`%l!>WP;QwuAr3XPi~j>l;r2}Wdp&SF$q
zTU&c!9@(!q7j;##wm$ElHc3ly1kLA*w?_YT`CD$^?D486Vj^Rz+7Q^@?!rGBa%;Z2
zx@-WKnwshKxM2(ufJMb7<z&UZP1Ddy`DDCog8uS;>G*l=L{`w=%!Rz(V*6c*mHsZn
z=VIpGoL-OEZ-4Kg3mzto%M!;7S%r#~=IwRWzxynM=O=Zl#P5XU<hsV%ADl+bx79u8
zzJZOlT2bLoP$2^VsPlbJJ8g9MoMwJx{M!-Z+dyV(W7X<%2*y=TK=1c)hcXxaJi2(v
z{rDd8;dL?NvNV1&w=4}T$l*JfL&ITpsC``l1rZY#ruF<Vwu~Prwg$hG)N|mKXPFo#
z9rT^*LT){jB>D$Q+ua9h;1$%=R<;}B$IZGt@So@i7-)&3#?7!XAF!BbmYpVKFfp_I
zJCC&cTB@}dN=tA0<=5SgIO@)sIV^jZ2!k@Fy72fsvJ%peKu;*i{&el<uScwVM2H{_
z2EWFT7R*mHv%2NYPI13iyn#yL!h-sGmdZ*RXim;(7yNlUn(B!jv$`PRuADk!vDR?r
z`cbpG_Z1nlml7=6+J`qK957H$&5^wmpSu?9_))xavKPyVjDfENM1!KRoKO99xNcAr
zR{4EAQKOS<LuEs;F&*EyyXLwUF*z-_j^8lZ5NjU-bd<5Yp!p2{!I#$}9ysck6%lK*
z+LTr)u2|mJPeWGovSks&h~OogM*X7C=cV#gmt_^`aiLioSN%A|*z$r_-y^TskRe43
ze2@Tf!?o{8A_U41)d)@|ZzbF9T^zhusLkPKS;WE}&u)YLzx@8PJu^>uu&>ecp2%}9
zI;KrFzp}o6zFeb4_=#pgg7y{;^Wll5jm<i>Kvir4Lp@$oW(47B3oG0s$U^Xa#G^)7
zrLrR6HN{48u!9%OW5v--ID|5M+E~>^@e4qLC|16R96r6Hu3kQ4=LEB_g)sWJ%VbqY
zAwv~GQCU^ME@GeKb*TvdGH&|_dq#oHZbw78O`)%-s5sm|U#^bN*YBEUV`KAhxf!R~
zY|s<9u%HI434DBf5GY2h&FL^UJ6qqxgd{{LK0bbWZtm{kL9fm%raLhvBBI6Za^rNd
z%G%262SR9M<iDjFV+>j?mY;hPA4Qih_ot)({vC{^GP|BFWpcam`M#ydkWHpCA3mI~
zLV^>LknC;s1OXo-GBUDUy^7EMDn}%Ex}u^2xE35lCZE-AF`c)wvjd=gNNDJTy}hpy
zBF4tX^z>Dv(n-TG*BP1Y_IR0cmz!OmS-f5#kP6j3aARLz-|6Y;;rr>CnUx0X#Yi-8
zrgNs$G5KshUe7xU8k&l{yucBYveMERU?3(kICqEQhB9<rT_0j`xj`T}IJnNvPNIn;
z;KK(51RP=Jj`I6`&@gz+-t3PK#bAF~$!}?S-XBc{t^%I0v9d~tk7s0Ixx2g5sWu46
z&Gk>q95<uast=bT0|rsQ<!6W=Ktu%#harK1$mBAbO-8;2h<m@kxM*l-xVyW%xUgo*
zsVgehG&SL~+pPg_B8or@0m3=6ou5;gW|<|--Oh3{b3S-}G9{^+4h|0OJN|VrmcjOK
z9y3!_S#`a^XBPd^XV^yB$?2(S_}W<f_EKC-jEtGvy!`QFrM0!uH2QArEJ$2#;O%K7
z`&)XC!CJf1=La#rM!StE(=2oC40EkFJ`d-^q8@;S-oBKTS0N+DeRZc(>b*3w)P!nP
zrHAMMV6$$U2L>CgrJbb*_SdJ6-D2uPc`a>8nrQS8p=GD7rN@#V?uOFvErlU(Ps^i2
zlX|f~T^=@S@sc_yP$R3qqgvc_7%dY`ncf}SHuqT*2lXJpAc!SR9Y-I2*e_P;AytXl
z*{*kdGMs!5*P99M-vf>>aoLxv`ys~+b(%I$EQjN%ifVc_w#o9!--oEf#PUgt3(h4a
z(>Z-^4(^v6{nJ)fjEg74MQ(I7$f8sg6*UzVG2fm?>yoP;&+m5P?6h^2sn2exVB`%4
zbC{o+@}@0PLbTdZZ0}rL==545CSslNNg00c29ZzJ)pEAb0^?6gF$Dtg`5=iG(~ij^
z-rb8-ynXbMJEuvIyqqm>KqP#U$vb>vp*){+7u(-nD-XZSk8a7EdmG(+2oD)b!hi(w
zXZ^@X-en|Of4)x?xlohGx_vKR(n`aU#_%jFEz95=dSB!Byb%f(%m4fLV9x;K`{?yD
z0@C|TAr>MuDd~W1`r^QcjEHcP?<?N8S>1G*x~3$ok5;(IPVK;Ld082|{U-csI>V3X
zBy^^(03}P!=ZEz#H#hLxA+oaLn6k^5smttko7!8PUe7zv7W`pi#rZ>N92@6z%cBzg
zN5EPoCZ9}A`!Uo%kS}j$r`Hoo%l(eGE_UgnC1$oZDI!<%=Z~}J!&1wqgNqAdZSdL(
z);lrXi})J&q-J>%eQ1=5I{DV_*ykW)>ZkN}#zzRiQppn<9#yoqu3X@1qpoC(0iZSw
zO*-Fb2QDfyo)oCnj9BAg-CN8UfAdDb0DhREMCD{Lyrc?6QA;b$%gqKw<MYPY#A2(#
z5)pKkY#YQYgMPXkwYbVw^*S&sBOw#uQbG!U*yDRZ*B2&(b@|aaE>T@u*4kW9pA1H>
zyACv)M+Rwz^bliNbw7Vjo7c+Q6F<=hAVvdy@R-mCNdjU_%oW$E8ZE!?fsh2n<86Af
zzQeAopn}9w$gH?wqQ;nZogr<b&*#21Wah-Q1%SV{x|Eb=pQY1(CxmJeU6pGqYkS@8
zdC2SrkC^;Cox-SkyjY{^fw`tw=JW^=)xn0FCjLTEgT@b!74^$<QLj(e%W-${$+YuZ
z(X{v2E6aHydB_N+rey)^x>V~UP2Z+lFKeE}=-pW>&Y}43Kkb!JlhSVb+xgK7_OQ&y
zhSfr3>KCrpdm@!x0SXHmDhnFV_<OUGjqXF`%TtQ`ewzWKI`mHMXM&hjx)`b`;JBak
z^s;`t9Q{*HR9-vyAx|wEH7!q8w*oHv@xFUAA*jum)?3(oy4w5D$mAaSP1%}B$(rec
zgkQ&q_cL;Jvv#KTstaxJ-(hMMqd1K%<q(ZY^$+mT2gb|U1b<djTC}U)z`;D~>!!c`
zn=5<GFL%Sl;FlAt8OW-YtX=0=;2^2=(T}(44<EcZb&kw|VSAb}dm8WG7jAqPCmj)s
zua=sw4=@@QEGvJg)sMWh$>wsP0UEZO<8-yj{`B+|9uCgi+uO>@%4{qJ1%0s6;>g)k
zS~{vWry#YowDk61Tv|pZ7(-1}b-l~)^W**1>2Sj5>57m>7?^oL`C$4H9ZxJSm%)K@
zJpv9|>+o`$DUc-jX1P$IIUE5nw_P7EqvPY_0|Wbu^ok-P{kMmcU?2!+1ox}09-vTQ
zV{zbmp-276%<KUAA@wS~4zEYK9Fn_(@${smq?wr+0GFz$s{>^Om$v2TR9Q#oY<nB!
zd*5wMWkto^{eQ9E>GN7+JhVmpaJGCq5QQOe`P3ir9o!xrL_kk(zdIO%3MZ<m`P>@<
zf29uw1~!~(q|xSDc_iI$RyUc!SqFS6bc!9Y7%^#C+4VZ}xVSiZ1%+Sb<<w^7wY9Zg
zj~74zkx3zNXv)pamGBsh#U1^$S5cXJd%ku0>E3HRzqp8jj^62cf3i@i4O|i#DU@#{
zDJ>a;#rQTrJt3fYbx6NnI4z-fDl4_kIQ*Exg;LimK>kJBn)-42?`l(ppZnno$D<<n
zW1j)T!xK&jBo-Z5Ku`Ab*)Mz^58umeW|f?H!uTzfv`W-}K+Q>A4%|(30k<4RD{lz_
zQZ0J`AHY0Yf9CF@!WBZap(6#20K^(c5n?bQ35)k;?(P^}wP-SdHv3;i<=xmbL#0)K
z7Qf{EVi+JMt6B)GRTU2MrpAjUuOoAzZ|#3rs+@E@JO(x(#1#Xe2QT-rKTNm|={JL+
zuqY`hNkv6HC^Y0<<u)dK!de$a{f>SG@OoL~*Og2_DJy6M!?A$;OTVwGH*-UB=^%g{
z@c2af(_&&|p+ZED&IHK4>idEKu?Og`OiZU2ohG6a#UJ>+9N^I{e%{`6xZ(H1urTZ`
zEC2}-+V2Cr-|Xdg1kyv}k<CvlE-Vdk0=g+!->C49@!ZV9f&!cuC+SfWCd*${d9Zt&
z!$vhgK}EvjrHvI<06MjUb1z14M1-h;R<F<N$~K)S2=}|!jXrbM1{dDdn^%U^u`qdk
zoi-0~Syj=}$_e9I4f_p8W8vX{4%+kctS)Unu{q#~V;y0WN;5NlhiHR>5|d-0<<Zv{
z`DpI>pnt=^c#7m4s+)9Mhi`$Ns=5aQL_oy!aure64vvTn++o%ArNG=QLcieN`Xl&$
z?Q*)N%Z>|gQrk4z1v^3$%ovxL%O-{)Tx7&Oe)!+booLlYJ3KrK#WR$sy4}X{og&GR
zsnf*_VzXnfewf+*o~t763u4(H&xa{==7b!5X!eU}Yx+aJvJNjwC#w?wwH&iF|1!dS
zHtUH~(h`B7842sOPdg7zd-ZH{D68E)45wyrS!@|L>miKn>bxoej%ibVGTv-NcowxZ
zpXoes5Rp{=dbVsnQ+46X=_{;8jvy@kb0?`u@g7~5Ml%*ymD|?J$b!;I-l)+w+J|Io
z6P-IY)zq)>8P}BsEQP?z)e&ShNV~6>7Dn-k>8#HKI2t*}g%!wNW|=V^wU?*c6KAdP
zOQ^LF=oS(pmaLkpqTk#gFU(DM?vk@;>yswVbiX1ZB4V@IEVa6v0*gpd8Oq7&wlXk<
z*&OK;?-x35h!8OMFnQb^G&Fw1OW-gX1&93?L-m%k|0!7+)dz}(hK51Vh*B~#Bwt}h
zlW5@)5NIeUhR4TweczfxL!m)i*SkaFu~&0~$H&Lv5qEXwlUUf;{N7Ivw`As1M?liZ
z%F3cv{~H}0E%JKH?<+w@;3)(?w86m6{usyqF*Q32XjdC?_yAkAvrxH<Km`d2+3x+U
zv~_TFPf0_A!(uVDSfw8!k_V)XpFTx$nO64plYTx#L_`@G8Ae7%1%-v?lUcYwexxM-
z7jAb)ima@xpzet8<R8yrLr$8Sn#(f@#2W)c@G)p@rlaTuZJt+Kg1WkR##YqS)K9LD
zM?IEck#Z;fMl~OAxAFiMPf9{3CeBYJmnS15L&v}Xve<fG9|Wk|fC0$k@#*PN%M5g}
z-@$u(1gjURS_AWmU30jYn0_u*8**}T0%9DwT*lMevy+sRl#oy_0LH%{;e@Zf0|OJ;
zS?93J&&#YWB3~dK-;OIQWd+SuSyGZy9`xq{z!raQ$ye9oKGPoo#iMaS4IuqS|BZCQ
z!vu4m3-;s^=jQPJQJr@q*wy>uwQ3ypKt)AiVs>bAx3a#p1PL-=O0C-Ty;=#Lf0L1o
zmd82JI?`=(VJBzrdWR<;=x_$=`?K`o{+`L<SjxiUT77NVzsDYC6<w;T5*Zme>?4wZ
zJrni6n_@9YAkglbIBgogy0UUdYwN+kd+u!%oG&*3o8g6=O{do#aa>rM8<1t-B0C<z
zj4{nyp;uFYoDJ5}*r>4_?hEDN!Rf`FK53ZYD(NezsObUqQj+oV`i<-4F9Cg*`4D>Y
zwlEANcu-HX`#rkVPrs+9&CVveTp!rz=~@2HxUw?qP(;ilJ~I-~*8D2SIUD33Cz$;O
zdC&K$*6C)-(1eq7`Ms?X5pz$3C>{g?<fq~ZJt392gqa)e&bWmN9XR~qe=|6~zI1vW
zu*gTS!h(W$%Ty#ZZ^8yJwF;`DrcLZ|oOFZXx>{0JD|@gOFy+-(m$bAdqI%YuIR+OD
zA+@y$7MO+k_4U7|f3Jc-%Da3-YI3rZ{^sMdvQLYGfz<FkQ)ZVw76pc?Xn**V>Bim9
zi~g?{;0LP_hhpRp>|}?BorbbTxexuJ&&GNNjDwMIK#NDfWdZqo)1Xlsi_5Og#fRoN
zVQy(?s4de9YzZ3(nApO64?6cUH8lZ5v8mXa%b;RR+W@wD7FM7ZYsipA4U|=uhmC(M
z+dNEkghj6M#F<DGJUf}or85cY$<a}hKVHJCPb(O?UGO&6)rj%wTseMERCe>U&0w$_
zd!glympdEE`ASPW-o{Q>->76Zt{-|=2*Ao*BPV$OyB&1*iL`yjkO3op-<O>le0Gxl
z)LV6Xy$|ej_O$dgf~ttiAo7t?oM(XY-4~M2BD9Nn(AjVhNlf@h9Fq!vvNLg$!Zm74
z^?4K6o43v<uK28^FYh)UNg9jfN<(%!dq}Ttzd@L&4F+Q2TKAgFEq=6*Or@7p6B``K
z{d|qX5pD<YJMdC{XSWwvOS*qjQg0sEUfB0Ln*4+wH^DVKXPD6Wo;iroz!c1brc^`s
z!JtopuPXXm6uFE%sC~G9>gF>p+hnV?x;o52MfPylLT!P)`yM8br%dKb_Xt^359qvE
z3==UjWUn{}Y%@u;S(}SvsCqG*)0E8IU;d2l@9#aFKVUI^EffCi4EFviB{h1v=GAUC
z@Qpz6=Plfn<t|^x07{mDI60_0T(1uXvR6XbeDn7Z&|oKG{lZ`=r(c+VH|W2H##*Ft
z3_0Hc7@OLq3V`lF_T|S@qhSf19v9iX)A}pMx#G~@GCjn)&!TtOzWL(u*tRBMh^-c1
zKE#_oEpw<6GVvmeC)2ttJpcmzQcmW69>}O)kpT%Pxts)q8;k2S%f?&%9N)eyp3-F1
zoX0kvjAL?^ERovyydPd&i`!~q?pb=*SyOd>@Uh$a>LX`1NiVtyU|78OQ~6cVVx3Q=
zcvV0n%ZBe7b8pDWS^<HsWIcHLyiZV-FQ!_3klWbt06|flSYG`1&$J=oJLog4GbsGl
z;y^Y(54MoAoTlYPig?t%gATW&<*xXZqi0u4Pw~Hln=9;d+G238^R4{X&@lsy46y%@
zIq}^prRjBbMBnNfmn8p@6uu_fMV~T5Q_epo9L)=+e=kb_18Ls3gb~a4#Pw9=R95<&
zpdO-lK3=S!uQq{!n9V16yft*z;o?73RaKMe^|LZF&ySBi-kuyGARtiERa8^}zJZ0E
zeROn`-~Gzq#rcA}+2_?2kV{LIi@`y3dhN~Di*$Fg;<R?YA8)ByS%1sR0p_ivz5V3$
z6rb0#9&kbQQ<>bK%;WR>LMDv5##yldl*QHc^>T+7JJ*!^b}tkqWhns5!9Z_sZ(k+B
zL&wdC7sW+IO_ys;TUuJ+(P=nb&r-6oD0c7i@`%{k*_l#l=;;A0@3jPOIuwgrR8&-F
zHva#Dci<&gp>lF^0Q3j>KRvkzY)Insd0#y~egQ2lE#-iJ|F<&`HEoF=<+3}72olKV
z^ZBYO@dJ>2yIfAuK?X-txgQ@ND=Qg`3i|q+larI{>+iG0@(m3QK=IsYu*PU7Oj}CV
z4-^z&!2(<kRr>s6`h0wRGGxl?>g=SXqG&g_z_pn?9<`~dnwFNwKs|E3ME|3*S|D%l
z&PLyrL#@=-7!}Nlg%zU*2ZhAIdNF&fnAV$@kUC3luhsNzxvvDJ?~qn2!j<;hw?xCb
zPoT@ag1H@JrT+Sf<fNQ9{g<x9+{?h>=-|Gb!8iKEs>}iUXvmj7;77L_whC3l-z<;U
z6)2WF8GGVB+0};}Y7!FC&sH%2noq%SfIx@2L5~+pkuua+R5-l#=JTHP_1L%9v*s%K
z@yW@;@?SgrVStj@_)V%{Me)q$CHFtklf4QwZJ{;1m-9=(`8it}Bd5o-u&|K5Qd?71
zHRgwL@*j0bsIa`6Oooqn04D=11)VkzeyWQLCY#g7w|e6^N!bK2(Rdt2Zwi_mEp5%S
z%?8_bGkaUWXhe$$1!a@KDT|2s^nC4eC6X~q9xpMhR!smX<EwT|HzVJNHd<dnn2b5o
z7KDb|4)lTJW?uyP;CI)>rP+dLPR{%ZT1S7YD?m=BrG$TTX7G-S43MRLI9p<Jz5<!7
z%$sV^L%fHKavD*`h{40hrLeCGkMXXqT=86SdOQ=cU_}RNDrmfMvtn^&;z41015?Jx
z+k%yKPB8|0B)=G_qqeuLio}^A13ts_6oI@LKst*P`+00{?Q?v&J9|hc#5E7{@0QPw
zN5)vw{r%?;?kQV>1PPKutvDj$nplj2vMNBeo!idA*G!cL(s(j7bpMvzRmd)KDNn{g
zMa$Rp1$Z<d11fys0_eQG7<c%P(uElX`$yKE<F6i2JYuS3gShUh{JwCa)cTY}*JIbh
z5=V@n?r6(I71iI5pDy63m4E+6$V+@H8G-9x7=)V#v{%lS=dQeIfYG%X?w%Sh!SLG!
zyd<~_ymefr?okngk!%cGPSbNXpO}M7yCo$Q*?gX$i0=IpCK)ZQ6|YIbyG{A*E@!p;
z?bL=dYiLlnyZA%h+3JPvZmF<tXfIy@pg6s{+Uj<h7B>RzZ4#Lu?#n4E`cG;yMq&_|
z8tLJj)7??$Ns(AtC<yepk^eOYfUe2caQV|JiPl>``iK!ibsayk9fLU;yfI=#xzR~E
zaDSwacP4e?E-Wa|U8kRvlNCdQGz^<W07`zrIv44Ho=JaEku%^IAnd;tZJg!~A0MWa
zVEbjgPF$6t5us<qwp5!BvxTtrMFM3&0WcCdET7+vb<To7#E81kW6ss>tp5DDs_M)l
zegr6A72Jvotn8T2TyK1wTVj&tRUhkyLrA~E<VTC~q&Zu^-KGwA`T6j}TC?z?Hjt-A
zLx9%gTI1;ERObZIBXeSv9tq=KrT*6FUrg2Ug9Jp?OcxF4Rq{)7=X{1=A@NaqI*{2#
zgh^ZAoEdBObymYLeTuvHG4uGR+<F@r<%^G~PmigOsa$0BZe7o@o-A!_D$2_tK>dS*
zN)*5I#p67^yqZ=?b=q8QZEa~mTR<;SEfB=uabp4`TYxDBd_M^Z-&8IH8@|!g)Axgd
z1qcYvNrB!%MMX%(7z8qkyNv?OT6fKMc6P51=Y4E0MKXdcY3=PiAL9V2oSvQz1_Eev
zRJi@Ukwon#JKM(>cMp$vz$?`auoh3c!2t3$w6aRe%#49FoSxZP$wT_V0E!R^O2Q14
zDzsbbW>jkPey%-op!M+Zh>3~G=?EhvC;y_Thk$?(7#N7^o=*p;_V)Id0NViuVvy6$
z={T%xXkY~vI>^A*HinU7d}2b(q!|hxy`;=YoDUTh^{;-H9Klx^8OUhNK0pAcq^8dO
z`vXv|AaF<xET<Jo&Q3)NuYYJaVoxhWmH`TB61`ZqI#E1_DNz%>*5(upH>FJ0*2l%Z
za97yp(?+}VeJ`aui_I43!qnE1Q2O5l`Yq(-O$uBS0?zsQR)CKq0`&?hAte;w{<D)u
z`!0YlDlpdsCxHB$x8vs$?iIZafyMslHHSQ>swXBE%Fnb<6GwUEOk;v2Oz_p<URGQB
zJ5mn?Eh&^WCfXlWhe&D;bO9wPtfG*W;->Ak<hYBqW`~m_FqP*7RGTSAPJLZj7nTA_
z$ol0i0uvL{J|-P<=m60~<Pkb1l7zpp6g}XHYpueANDc`rd>x3}B}pLA_uY-^(Qh_7
z3v`}bDPQ8qj7(Kk(UvHYUDr>nPEN2N@OYrbb+TCS(syx;VF+kxecrD409lU++^IKW
z!&IUGs3U!!RFsr*I72ZkC@7dg(YM|9YW$;}62f0x!vvK-hesXJ8<m>f@Mq6bxSg=x
zl802Xnpci7!Fi0XP+@i94rr&Qj)u!${=AB`v?2&YVNE2U^leWU>wBtNvRpUqhmDx@
zS`N%VYuST9-~pg7(-`lLUn8s8T*I;rvv6DD@L<_XfZhV;H4xU@x}PgFhfYx3niK@W
zsVAc3L}SzwW4xGct3<ljd>dMez0d(Hc;z50p6(yQun?}!X%@Yu6hjYqL3*)nE%+nO
zc#j=bq??BG#;NB*97-c;AEP0-<t3#Yb+P4-<0ssRC>CGMQsn28Fs#}UuwTcTGGgFZ
zV0h8I#%i{r<)<|6E(_NlPn88JxGH6o0`JL4F$+W9TM6B2Yolu$!5?!)*V8Oy()`fb
z#h%25=2N~hT=<Bkmpi?v)a>(Pmmt1zUmgjUQ<ZBqb>>*8TCL2a9PtaWaX1*GA}wHp
zUcaKTqPtm4>Vl#Dta??$(Omtv>tVRs*ZNcUVp3FpbHY88kstvzbIM}Mf&dlHbRZIq
z+x5&s-CsgdQZO)jX=Me@Ig98k46yXXiV_o=JIhOO5v?v1m6ersbrJvb!3g(N5#W0L
zXJmwgE|!*-wzlFBjsOta0YCR01sWQ7JHhs8o$I+4&`rBm$!lmV0Z0W56dN1+7Rot%
zUsJu+&o3&6Z`uu-U?>*QF9xuPnW-rl2pv$`0Oo+;+uX3|U7y3rTq&U3fx0(3ywD&+
z2F_-~n6V%?r6Bh_W@lzj=B2wIc_=AiEz@o=a_WE2LI|ml_yU3yhhc9pz*~aH11sJ=
znyqW=k@drn#70~=S;te$i`yFLPu*e36!6!`2!NVSL1#v8!-joZS^TcWjSRJ`$!r0d
z3=$&whIr?qm`D_Xb7w7r%40o-;XWC?D7OtPK$)jFKeUu9#v{@fa3~rTb>&K1>B1ge
zm<DCQ2V&?%Woc>2QWhB0flHN2{Nu;fkFPV9&U1s1a&UTDYJ5HyM(W9iu54;R2lINz
zc#@}r3X^%{?J1Lf2Y?+DE<(nZf_jAUUI=b3O>NB|kBlcr6QatHumUeHo=QYj2!17;
zjD`%U4muFu1rOZK$OcPBza`vADk@e?4J!%Z@fiNVI?A73Z&P_JP$;!#_kaK`*|-`j
zJL&m?z0n5<r0_v;{|*}oXScn9<nWCV=oHLG@$umWc#APOGcq0i<j}#vyRi!2By^Es
zO-c3S-9(tvxqttpY~fcuGyxpJpnubOzypDGoVpAy<h%1tR(<AJy`iEV+*je*cE6yo
z;N^}@#HU3uh|=-ssGJH*YiP2@NgQ=y;$uL3X(5dql_Egk$l5M)F#F%wKf@Z09TMgp
z`YQiC*Y=7lxBP;7@0AQuVsV7Jh7Vj93so0H?_z`jiw6KRm}BZh9m5XdJZnn}I{o^q
zY_g0D5`fQqW2AR)<y0S#tK4>k#jzYV_P5Nbs46Z|UQwFxYpzd2*g*6M{PPVaSg=(N
zJTvZu@;)2B1#EF{f9?hwj^Q+=#8JO>5^!Y3=sY8z%vhJO#21fnY1sQ0Tn1_U3SU#h
zER@ZER=TUiOVCA?xOpqKEM3N!Zqe|^)qBg_lqt%Wr6dX{0dPT-DT6~1?Z=gH7k(Sw
zT0t1l)r2m-j5B020rEV7E;%(V{Mh(8+%LP9v@|TL_<CtE=tJ1K`FUQagJ_O)0$)au
zKcMWwy{3&cAtLpx`-<daaySsI6pB}h7RjUrYQ{*AtSl}0>qGPQ=LiNtu9)Ur2N`;y
zf1DZ{7&zRRyn+dUKo(L(tkw%K<rXR`9DN8W@D}Fg<UqS`z`(@JJYvG+eIFSW6^|nc
z7{iC+@a>y^lo*f`{g9QGhEI5tR#m+R$hfb~vEACRppw#3fO!*npub`fh0GUpb2(pO
z#l_8hzCBD&OAC~nE>HF~(k5EpNw{G2{B;#YE4#D4O^@Jflt2suNwjkxZn}93%`9F{
zJ6=cZ@BgVpg-h%)aMg^Kmq`!WgvLfnq>L9d;1F&ri-hy5#p6h$0SROr<*vh6(FFOY
zmG=z%EvZO5Zc*u%J2^R#mKvRo*S55zoO8}c-Za1(JpF>_Qd7&&)X*rjDMt0@g^^;<
zSOG=}FhynjM}FIz6Zc$2DsXbW+1%BgS=*T<qz+7P0WRQl{$yil1b7c+5?tCJK!4k%
z_AB{&2a_b$ZEvJk5ul=!v;jR=9DNbmmd}bNjhd})W(pQ!V$XLuY-93~I35#Pp<HJ}
z?U->h_$WmxQql(7+ZEUK3P8AqTY>n9K01bLt>$ISj!*MVTH@dshOC$#u0PJrlQjt-
z_@V){5dgo>@}OW^2x_~QLpa+FN8T)_x(GGh=jjU)(o?-~TFupheJ%Sp9N9(7?*PN;
zNDp82kDg43HBH|dIwj2@2=boo1Z~62Uc%1SmN4<ppWh_PE3p{!|7xpb`K9OL5~K=!
zVGsm`DhXYV#*PzTSGC+kaU(}Y2<SIE=~b;{J;iW`IHe;Bj%*lE->m~{xbvn#pe*yw
zP}Z~zuD^YO8NVw6nAjjYb&*eJc)+j_PgFt-iek?vmhY=kN~V7|sW2_CiJgt4+>#ql
zL~&HvlV%(S3>KiVIacyR;!bOUCwJ~nY3Q)Z>ZLgqN2a^C|K-`F!{|t~CyfVZNt2sL
z5sev;pV1FS4INM?W1UX3l)=gr{lrjX1Na4`Y3!?x-^{M_LgavPMeOYcRTA!d%+TM4
z&)H=Ii{yGmu>a;G1qG=0bqrjvl2+}n4kb|-(pVySh6WhX3@0%um;za8WF%a_g6Z(r
zzfq)CnnA%4q5>*t@bK_Zk|x}utZAvt=5RV-4JM3i=wEjbJNN)`4gIhj3Iv0E7#zdz
zbTD>*x(M|*NQ%EiBZz>U+}zms_V!R7(9SY45cHD(GJ;q#83~_nt<GG23UB9!s?XaK
zN{tC7K$lmSkNZ(Gsu%hB^|TO(qIOSDJNAd80GzoE+v?E&iLo&o;0&OZIRW-jo$u2*
zYTN#c3~z2khB45yh<4~%N4f6_@ztipF!J&e`P1}z9nWi@m3vjcs&}s0tR%g8zXt{<
zaOZTt3(HU}H-<3vE_`ArDUTYLN5mew)i24)%7W@GSrn*s(yBaYjiME$q)xW<E`7pc
z^O+Fg{zxId-_YKTHUS{+aQx;g^K9LdbKW`tY|Ugco$uy2P1xSFN%E#qlYi&N%E8^-
z5cO6#V?|%59sOZMHi=r&hZtifAf_mgE6#o>!^(;l`gv~2!Zc;+YDDp|IK41)w_MlJ
zWy>vV1^BZa!O4|NuFOTcLk7J2P{LR_mcExQcH{HTO%>_Jh~?zQR8%08O~bSgH&-bq
z_db?2UWHYog>zJuRnWK*1_5UMRHYqFRRew17X29u24kK|p#<8-#-EcgbS+OXZZ0o{
z<|&^8Jv!$TnvC%A7F;?i05v20>RPWKHu2!!K}ia3)-H}?iikvQoKxXOACm5v7e9DE
z<C*!M^)DxHBx^fks#Xa+OU&>9A;n~-)*H+s#8$%0xdK*KdOJA2=qAngD*1t3L+?u1
zP{B&OrBqANFApi(tlZzbD%0KKy>5dJRzsP7*y&_5ltuAnWRLrHgYI4JTgY>Hh@5ZQ
z^^N7AG&U`D6&N<Umh?uHtjs1lf`z!}c5lc|g}$C%EJxYV$qC$@0~n}Uqhf4Kru+0R
z_#b1nDil%9F9{NBYq`}3xng2LCvInSB7wM3d<<wh2oUI=|7GAbsF@fHq@<)IQdLn_
z_OGfD0cd?%{#;*Qmx$d0@`89A9vJBC{9L_K3u+|@AgD+e=jYJ~k8;e}0aDEHw4m6D
z<~yUB#8IQGt1CFW$b?W1FKU2*`|Cp`up_8YI{E7}AE1O#Q=bCLu#}V(pv$f|I~unM
z*TFc5x6L*Vc7+G?&oVJFot#*?*nUthbGB4dONAXxsdXY}b%V!b{fz~35Hs_4D36wk
zGaeQ*hs9~p6IO(^zi1lRhUkr_&Ql!)!i3OZCJ*4NO~Cs!U~4F*2&_)O=2lj^?7xkS
zon-(zD75d(*rwvxC_T^58=44_-aYeoQ7-l;X0zXbsj}(UlZYgxrOQ1cz$;DjH>c99
zi?#i0GC!Pw<+{r_y^(@mZt<YoAx-Ud-qx@ITX67}Z)?u3MHmYr*YlTzG{WdQI55kI
zj3lA01oq-gmR(j8LI8Yz@JQ5^=vU?ps!4N^U#W!BQz0Rj++4Cmu+0bOzOImbLe)L_
zzjG1?Skvf4lNa--xXF<bS81?1)m6J~jcVM{BQN7_sjH~?A;AUp<%<h`8z1<?`o&JI
ztp$=vgM%#KBB)0`r$e*lld8>CXW|&5Ms7cPks^Bk4K9CeB~EE0K7$+xPzGekr{^x%
z7gVZdBwk0gP?4X&=7^rtsfYy(oITPGHtXl1dvJB@JoYbW-eR(BTk?tB(j;fhP}~##
z4_Wv>mY88c3=)eqRiT()V`GU9Magak(@;Odb)}S4;ey6lV^51sDX&<_G2`ZcsbkdO
zi{pajpW^yh%vv)QzfOvqtD5`{J4XIF<H(?}0C-T-g<RjTbutV*OMiIZBSe>ncl~gC
zH(X(+`hFUbs*>4CVeqG-eheoHGm|S&feR)2KSmVS5*HVjaXo*EgA$Mmzdb0CladlT
zx@Kl%Ol7gP66)}X;0GB0D=ID~=CzfyWEmy|L_1qssMi#6v7cuv8JRW2MKtK}EqooG
z_un#);Go7aE{=|xf%ajV8gS;`>i2;`Kocymm$(DIf{u=k_~X%J`Y2{^B`qy2Rn_?m
z)8NaZY8N6Rj$GPMBwRRp3|n#G;#m}z3j57YaBhEKN*Bvzaye{u1Gwk?{T<LI{ZX4+
zT9#K=Uk78b^*g<LOwjF<H>ZX$!U6gdJiuUf@*5(N#^#iUvi4AR^~wI}Bf#+;6p<a6
zkpX|ZmLBpHC?@qCq5j)UF={TK1FmtDY_FR;{q}){G!Z(rwA9a!rKvv2jFm%>b*|F{
z?Bhtx-D7Iy;&oYx-rJs4QB@RlQjcg@)whF6Pft5rYjoUgMFZSTrIwh$0qje1SX%(h
zz2OpRvn|ro5tVy;dt|Y0UT@L6iul*Oo;&VMB`F-vA(V+EIJ7^&*fK5D)f*dkFu8^U
z6Uw@ayQZz1(#j%JOKVzEBTV(3(9ld}cH&+<!TNMu?s)<t`!Na<l--TsmR1{flMo(R
z3M6$o9w_L^-CkFo<?QJWQn~<I25X}3ub_ZT1J?$OLb!N(cd?3;R*u;01ZmlBTRPA$
zJ*Wu?uU`F~J%$gO<zsty2d4nU^ww5{zKYqUxq8c0yxk2kV7SbR2GI}903$Q5k1E>U
z{(${ycU4(hT3LLqMvZ~!)aC^AyaiQx*+^w&idJn-I5`~w*%utRm`Nk3@edHN-|yo3
zHfmnkM;HRj4Ivjz${+AJpzc8-s~bp3+u4fdH-dpuWWWmF|16HLjhQSq%3?469UdM7
z<T{z%a$$fkWSUV;aG^LzpS2tKt(TaplL_752kbJKw?LKD@ryCja-!}~&Fa5>Q)z&6
z-4BA80?gmq*0LU%aly7FB0JY#uNl&7eWu79BWcNLznCkOku4gil@PfF(es>?1#PWF
zeNRJ8jI>zoj-l%ajp))2j%6#Hcxw)q8<iUI&U!^3ISoFux;@P7`g{ow5}y3{jPCNb
zySrYxc{-|HDyhZSkL*$vCJbJLw#~HUlZ~VpVO)`0ng;Yg!fR(>Ne#RmU^Mvnidvvc
zU=e`2_eONc`<Iq9K>q*zKrl%?1_ViD4grdF7$}M={lyIpXxCH#)=xx9;6t0XkRbVX
z798SVB=#*yI}M)&=AzoLFNsETCW7fU3=s)%g~GyufoSwQb%J}~qd9s2f>uHRli<y+
zcKIwPhX4eMm&jl+1pg#7F*St}dyoahZ17*Ob+v*DDk^tZ|Aj$~hNGh+P%PR2#_*36
zK%o6+)d_(3vcjv}d_oezi$J4ElCcPT<uWGJ<=s~S=;v7-?zoT4*VlE)U*a;j91nWw
z@fdlTvE_&u+!madgeb}B^>+?oKu_<2G}|Q0EL@rV>ZoKnrLyd#^XOqYTh(hkb}}5z
z*$A)B0AT4YtymFN>P-ba>>ZioZ!OHo_)F)Xb8e@D^GEu!<KsY`>Z-PncNQtl<TEAW
z!o@?8ES`-&8wG(1zRaURsrVap)}M})R2D9Ohs(Vort7RJg{_m5vGE2rhiy$=m9t;k
zU2kCEzJ3Sgl;nWKxzQ;uy5+r)ZE0l~$P9r^70{5-`E?yJo(<_<e?!K2Weq!W{2+|j
zg0_b9HJ3SQ_>{4pdcVOyQjc~D45yP$m6hl!=Vz>W%e;lRzX>gw4wmQ!<=A0OJt5Cd
zCKQvRqT}^^8nl02Q4%wA6EHlj6`r25T~bife*vYWWA%;Cy6UT5TNeU*+lq^vg0i^o
zn%?gNXuP!X1Rum{*g6K{II;=~985`|G)uc1*<ex?lI)=Ef8Yi<@8~GaGF3~@Fi%_h
z@aont2i2(b{zi9->!-9G8dnO)3k($YTjgIlnWDwc;i@m?Bqy)R=SFp**^n@~!q6s|
zBqUCA_4xCiwP03lP&x4KUJK$b3>81JENmK7C4pIsXh5Gd&Q^$!H3C004m_^3GRf(0
zu!yP*|FT0U`nz^+uX7GobxFb;rBKo6LP}{)`6Ny(Q_Nga#FEAC+QBCaui~kg62WnU
zp5yvqX(J!7Mggj4c9f(1Yf~;w{+ZQLyU$e7;p_!Ql@vdJB=%uLs8X^{<_r8QI%vjR
zO+kTy`FoS$GVCli%S{tZDk}j2DAk`8YfnQ$M#n@_OAD)({psjjjPv*_Kt@Z@QquHG
z{`|?blg47R#Ng``CpWBjbdNHt8io(x5|RoL;5m^0fDxe>3l<g@0IblQ$c;9ZT`$c%
zJ-NQcnE@JHTH0E_E6^{7bEgAxS~p{t+a-q@M(<cM1K{PZe(4YTzg_@x@?t<Z?C`jS
z*8GN!PBsy0(tJ%H0TRm;xWO8SF~>_UOn19Tz70id_xgm*=yk=?uT6*zw7l*EM~oVX
z+>(8uoT_d5{`TRxzTK|J&;R`!BX#%Bl=h5_{EkeHw)gjfBGZhnpAdRXmf-BA)uRkq
zD}ZCeTGvN-t9T0E@fuKD8X8=9I))}Teh&^775uJ$#OLSr3b5^nmmrBNO&pVHLFw9Z
zXzW6IL(Eg6*cp|9erXN*@)Ppu2%q2BXm>i6E~8MdwwsbUQd$ag%ys3FIJXKcb<Oi*
z2Js0}d)P!HZ+=ce9RKHGu8t@_GdBkDe8aDiX-58O%|pYW0ZX7qRmb7GX6FgzDeYL>
z14_n+1}5f**Y`r8ydCu>j8vu|VW*EHg?uYtUcM7f*oXC|fqHSS$1choFd`+nZdMxl
z*t3ksz7>u7(oUz6l5xS&6`Pj_`1>wWH1vE4t@6VZbd+3oA9$S?in0rVO^P58VD%A{
zQy}IbgeAT}MMdxX**@w}BX+{X1h^5q!0ET;c{X9s&fNpKbUzRAcofvs)seFsY7@TE
z==LzuclkyU3A520r)qjUha6Q1jG}g@o9BF=-ZApKnR}u&`3YCB(w*1%&E9=No{;B!
z8d;F=szRU~KYA#S!(K;N*jO9)1?v_gLL`mR_FD!IN+amSg@tv_(2F8K&xWBof0nAP
z@Q<|<L)V?@+L`hF95hQ^LL!oU_MVd=2hVHa6Lu!))57X~by8q2r4bvA{a=X+R2<qY
z0RkE3^x%n5)KJ#x>Rp`8NMe1n&Sy28RNdrRA#1^8?qi3|qk=Nj=xrH=4|CX!K-bdz
zxk`jJm5tsQjY~^T>D=-5lH|w$WqCVGee<rvDn~u@F2D3!!qzi!UtzEqN`h@H$;j24
zAre^(3HW*>te-I3vJ+q2)pDBd?Td;U9=s&)G7G+7?EW?UNcfYad)fRCdIl`r0{6#?
zO>$8YS;6F5m!BV_a72Kd<yVWfHGroKpoaWE|1$ytq@$ywfE_tFC?*0IFwXUL6e|E;
zRz_xSZUr;-`<FRX(WOJj%Z-lV{{M%)w~VSP>e@iJqJ*TRG^lh5(y63$cPZUScS%W?
zba#WabVzr1cX#JqzVG*aKktwG`#R2Wz!;o!_Fj9%ob!o!d+}U@iYxtFI@4!&_om-=
z2|}SJQD5G#|6peQ_YcS{-f#o~6zWw!HyPUbU|@Bo@s-8lY!zZWAYH*U5Cq`9fY6Kt
z1+wKBH0yjM5XX%uF<;4l>3UIRfc~Z}p7;+n<NgFsh)0hpGwiMNZ;7$-@rLc83y^CU
z<`q%X(KNgFT@AWfII`M&#9(`5E+l*rwE|B_pj)PbN_6*aHAO*`a_)6uACoYGfJiPy
z(&B0Lk?!@sef56nMOuXbEGpU?<v;a50|OoiE-xnLJT>o7pK9$9_oW$!ds+u`h&fG7
zM&SmvZB}{*3QS7k>AiazzTb3uy`x}rKG6E(da&r-wLIvwRUii&#bNqbuDI8TLi}{M
z-?0}lGQIyBD3h^dN^pZ)Zx}{lZkZ<8m^<6;zrHZ;WR}_s5zE%9wZe~&iHNWw=8^kB
z=EL}-T%R$%KQfYOWN#>e5vybz=DGKj#i^8-Xkkfyi1e4)-A2?w#Ivo@mqy!0Ti@oJ
z2%^OrT_0F)I~n8p_i7GVn5{#$gnVj6C5NuT?Pz`)&aa5wqSuIk4dFI5HkL}|*0R0X
z+r$6L;HQ06y=@PR`rYTK8_;F*%>NPs4H~%r^I+O@?U57j7k(A5Mf4f~(F^ZWT~m51
zAe@~u0IvcOmuD9isOE>Jk>S67j67Jm6Ap23lOoY;-We{n>ef2lOt4gGH@RKS)}qE>
zO%`vGYk6r;v}Nct$65}nTAS|q(r}m4(!SRD8Srq!vA<VXLwrfdU2Qudd|hE_IEES%
z&KVLaB|Ys=!_fToCP5NJ)c-#4T}DjFEIj@?pej8#%sDdbVL`LVmH}Fk&rM$d<E~38
z*}stVTMX6g@W-TINos0JD4hJ)HDCT%)Y+fW&!gM<zio(nbLI6e+2j741Ehiw2<&0_
ze4CuE$`(ecsi@mtC%||M2sKY3!Rk&D8xcAqU)Bk=jHf%J7YiV~_>oRJmBdRt+IQVk
z(ih<0uaXldHLNt;c7JOUH*>Y@?t9O~o)DT`T}6QDdgVC2q(6|9s3M)e^kyHK7~bAo
zo`h=J*v9-%ihe_|QP6#_9;5OLi?5$)bm=dcKbEwb*_j?}h9^s4U&U-pY)M~TojYE1
z9gFA=Sa3G<uR1^covAXWy&YuW-QF)Oa(JBKeht;NIq@};pE-3Fx=@8(`4BTTSzKX?
z+0)aIRk(Dse#^yGJHHTKCA?m0X;|K2;Gt-a`?&pRw6<4$gZ$irG09-h)JpHOcOGCs
zrM=AzmhxZ-5Z)+WFBpH&4r?)uInGJGeHf0b<ef6}<F!>k{WjpYe2{v5Uey~z&CV86
zre=viL>QtuTI<=mYUwLnhKzu)a%3Q%ukao|=-01tP?i`Di~Ga~AEEnG@a=4f2!WQT
z4?!~Y9_F80$cd^_XM6kg-5m@hAt{+Fg%$q20s{|wHwfUi?`1Yo;NfMIl^yKu0pNsP
zjWNDJ#mDCX0@ZGP?WTrg36l@q`hfV(#>9kPhKJ_|Yh~1Se0iY0ehZSqZ{<1@)f-X9
zs}=Qz6*E8xCSvXQj5nlnX=w>;W1^x|bdtYmXiRw54i_K8Q4<FSX_nmuobm0X*P6w-
zi&#t5aZ|&C?K2Y8)qsrM709IfO$6UsttIevgpp7pi6h!5Vqu+s2OE#6naSwnXy8M9
zctm`0K}JStZjhSf(jPaE!y;p#`fNlxe+Idr5cuAGg*=OB4_mB3R2rgHzvAKGXg)l#
zgHT6>vkx>ltQK?LNUjr;69bx+I)En!YBV=DzFw{A+a>^%_Mlfiz6+vw-k(Mp2&ATq
zgzC#=FP?v!+pL;%Ap8Wo4p#9d{k_d@)(QANbv3m`w`(7<#8<CQ0eRlL=0Jl>0I(zv
zi+0$K=0yw|?Qj1e5i{w9V~DCH41NU>u%C}#rnv*4+fNsF0)|HaPa_LnhhcVVPVBS$
z{e@aAjFcXy>94rxz!he-@vbGj6$a>UfVdqTrdFcyAnatJc~PH&g14MkIEE3yG2FU4
zQZgterjyOF|Kc|xAow(=4R^~ZRDJ|bACpBY)+@b|wxaxUy#*Iq_oixbiBfO4{J7?(
z%tjrrYMyy@@9)(R19gi?E$2tv{^4WG<L-M(Wkdf*embJmM#llZ5=1CGCWhDgE6zL|
z2K;>p>=SQD4h9j1m{=gt&~P~fL{%~JGCB<s_W@;7R7_;3_?KYs&Rrmul7(UOI*R_v
zg=fIUnctA40mMbb;>uw}J8uQvHa=Y=Tz14U85e!pn`F!sPLo~fS)U{;WJMw5{v@Ot
z&!5iBQUjt9x;QrV^45l@X2zBgcaXw6Je%KC_-^C~{6?M<bQgqMi3W6q!yK<p<oVTS
zuWe#WJVJn@%C@ux8Qcp$1p&FfQK6Lk)*TtZYbE<$TUVDblyVRmIK5S6C8^>1_LgU;
zXF$8YC((n&jVHObBImF8{z1`7i~SW5FCZ8J!jI%of6j<e6vdxrr=c4agdmZR2IKJO
z>f>VRw3<~`)8SrVz7%7M$;;aSZTLOM$IjX?Y$a<aj@pBIR1{*)?zVYaMve;?Otwn@
zm>k%mUp?L7MR%&59X5;lOtp491tGw8fR%-9YXe73^*nnt4+Nk|qf9SnN4|vF;)d4w
zx})6H>21Jr;hRF!K#H)tJDahJ4!995dqc@A50lAo-TRY-cfw$X{V7Hn7@z@X50<&n
z%#7OK$U?ynO#8lpeNHch-cGuv^c17$W$OvU2e>em48^5JCm~)l?G0dEHD?Emp1yZF
z`kaU=NFWe1C9_*(iK;biSrYuKKa;sRBL|&FD0vh+)WjQ?@f5e(rnj#)sTpz^j}kqj
zHN2E#bxb^4G1Py1<j1`<5#dLl5>r8xW9752E_(G=To+q?NSUN3cgkQvq9Mq6LxPED
z&zJ>q;kzu~kglu;4l{N-KDlDH)Cc>%k|DQWn{dRhZA0>Zdax=X@Ygb4J|`!<=JdxJ
zV!D@2EsFL^yl2vqWW|Mpu9eU)-mkr^+w}TY-_Ig(bHy>G)9K5KA4Tq5;a%GIKB^4Q
zdY8`Dnx~4ik=&*$Z6HOdV&sA%A-M58N_SXuUW0S>#D`J1<ObPyh)Js)a)~@qVaC7x
zTWKh80k=bWDVVXY5%km*Hl>O3ZuJjeGOS1Grs|-ENd*L<X`ngN?piuZ1}l!wN^2FA
zYjojK5Z-0d^Fk0VRs4XIDPXRmKwn`#3=Zkm&8%Cwb#{O00XzEo?zkcDqn*NFORj0@
zinCbc_$RLP>NIs@6O&hc!|h=p?ErQepk57-MxYzAS?_**&gAFkSE62n67W>+aHPeo
z81;LV^m(AK?@N=XI6$axw~0__`O0kY%DeUW^t8_Dj9hK@`1rfvMR;Un-S?H}%_pxD
z@I-;V)%|i6_@u$XL138!^hx3|w6!){D6>DZM~r|<;Og{L=ffL7|9QB*z^KMh$mFoy
zmT`CI_4@S3b|><B;kclH`n_4Wx3~AUnA72W?No^-43xxSV*o_6F);)pBa@Spx`4IG
z=S~3qb4NN^73z1Fu}L2u9v&Z8prHx%QgL^G1QiagX7}RaVogQG?2?i};vLwNmDzfS
zQt+?5oatWK(eXNg#T0@{N=g8+0ro&vQyy_E<DjI|BNj--X4NP{_m|49(8%xKBL%#7
zGg=yaJb1Z}xkr$dohzOEeT%4z+aItAU)qc_y^YweGL{55c)^I^)6byC*ERMVa(dGr
zco|g%Nx|kl{Dv65xY*oXapCpf^5*1$reyUi!qq=1a+C^>kGV-n`#ZXDb$#@O0zk)4
zJX<3m0E6G+ar`M4I4KZv+Qae;P`E$F*?|fV>i$fqn>nY8JolC(_$Gl!KTrnxef;x7
z_Zo~4kS5h!890h+xMsjl<tEJ5C8!a(5!#Ii%F4+lZ7^=MMr|cwZQJ&z)7`zYgJL2h
z6L|I8fa$=BZ&ZE%2m%BUwcOHNb-7-~)Skhfzt^m%9}PDvcN1nsq_uylbd9#%_m_Ep
zm-PE$H^IUZF_P382+Aat+Fq+hmNgDtlcBo}KurmE<@^_Bt*zT{l^l=*O-=G@>U)|~
zXFldADJZ2|5GqcvY<FP@X-i*?kHBbP2zfP0NqE^aA5>lt67KKf<3-7jUK1mj(gX>e
zZM~ZB1sS7&hkE7Q%=Gj*X?T~u_uzI)|Ia6~>LTrl@$q24dz|j>*MA+HPgKbJdb<g@
zDh)6E%x}!y)}~Vt%J|B>IC7-pf7_-TiUjEy0A}VB3WEL4#KgqoY_m<*qQKjxIiTH2
zWp5zH*wFLo`v6Wz_A=XVvIyu{7LW7s!hW#f!tbU-b5ixSfkfFTC?h{V{s<SF_K<{D
zi~XqgV(#893dBIIZd@OFH-X?(sbnsYlZYu^B5R%L{b<QSVrUY3PzhC0CGhR>@fAs<
z*HoRU{jL4p<Fx9=#z3&Eoh&=ci#4`K2B5(u<~l-+2WaHsg|XYV8F{iuP)o<xxaXqZ
ze|eR;2@{8`R(>*}wEHHcErMVB9I8g0cjVe}|At|hmKP)Ez#8-Mk$J%^GAF5X;Jvn9
z%;5UXTS+EdwvwSY%&@PbMeC#P(c-JeKWH1&57XG98VX<!`3XKhxxDUET;EN59(o}t
zCV5-^yD{!L*ZJ>yz8%=tr+JYSJ*l`(dYv?f-!*NW%xs-#Ay25~qHK$zaoJws1Ka`U
zuP=d9^FwWF2tNvnOj`Q41c?7^uvvEs{dmQjnHyZ*VuI2M!Rh~8Y?1Y&Cp8Z>uKLBf
zgqh+T(&-`+Ge_GH?}#%HimCoOtg|)VnD=6Sd(BU|IL-_kN-L-`7)&Gs-0XXeb1W!a
zHTEH&M4_Dn7U$Fv2W5QaIxDtD;pgIqoTM%YSqg5G$4Fl0Gd#L&e2>A6e$y!irjn?-
z)>*1m?0d)NPr1wq0CRpHWM9wYgar1Cv*Y8mu^sROB0M}GNSRt#40d+|8PG?_^UHf6
z=Eq*+^uY`he4F0(!n^g8hU2wjB1*C^{I)Hpj+P$q05CJVAL4rg@~bd>w$Lx{0UfH*
z<B3T10O;odcD%OK09*usXcEJh!+O2@kJTE$K2o@y&<O~*jYr>;k!1t<RH=3|C~3ez
z`}_O6Zns%^l(un`Jv|8V$7g3SP?1{IE6q0Pr6xBUo3g{>Zo%*0KLoU|6<UFs79p2?
zS_Rlt$8sdkuC6RiOm>%=n}H0>%j+|cPdPfWqf1TC&)+Z9JLKf3G};ZPeNIS704ghh
zJ%b1v<mKRN8R+RBPd9>o{AdF*)Cd{D>gs7UJNX<5D_h&(kdTY3tEsWEL{{?zAg=QI
zln1O&!ooy!zJQ<{&_U-`ZE?$;ymKCOOfse3IcZb801I_ac}Oy4nf?P`A*k*yQ9S+-
zad(f<584Ez$sekD85ypZ?b?|d$rK?DHQRR{`wNB|7^+jyag;E+^AX_=B|K$Xn$A&U
z?1ZB7)M9;!=~R^i4oy(w75&>U7+F_~F33EjJ$EZ6!Jf}FHSL$S?mb-L$E*!E2jAOR
zAJr~#_?ozcyqxQ5>}t|f7Yz8)F3R`@=C+QQuM%umMVh^n!nT|X)tMK7>n=s~C?GDM
z*`W`6d34qlIxwiIE(L$Iy6(7h-(xW->A&%~8P%r8gnBT?=+qWRzSP%&^JaKzVQ)Wx
z6-D$;adY6tX)!6a`Ft|+h+F{f0>^T)q%?PK2W~46`{-$PZEI=w6j1b5N-OLw=k&f!
zwpnVIGcjfQp>i}>`ZqLaM5)?xy3K9c*%&-zJ>0xm^Z{}HKCZlj`P}p+uN&s8j_K~N
zBMmb+L&<1IS`=Q_AjUkTKVZ3E#JSquvvER4+#J-XGV^j0Hx!ip?A|sVA%ePZ!jfnm
z#Mf9oi=aS>2@H~|JStip2oDbI%^y~0vScXl94rYr^PXZmL(o0gKQH-p=p0<TJ@jTH
zsx_<v!$*UGxN`lqUGk-V=#!Sz;c-)P6h*MCf)S}pMtpwHX~$U%-EGxDENZe;O<DE)
zAA{Pi6By^ZrJE~Q|G{cQDFU;TJ-B2kV(zq~T0T@Z;+_<#O!LLxy}cR6s5d<ro&E6_
zQ9o2>yE_@yGt<Gv*gx>Pc0;gh>bu9DO^52N#l}o%(HTGPF9-G#mIi3)`a$Y2ulJJ~
zI1qSyn0;g#SC+Z9t$2_tB(|wQ$qhZkkhVFSX=1)p&|!-AzdA}{cDId6Z)(T6zQ(=f
z@mQyrZy%kGj*0GK?JJegQ!wzub{ErQZP^Ygj?t*B4-CtHZ{qfDAl#kzv7}IHtoLL2
z-*7~4sX)IbKLd`>l$gC8q`A4QL6XyjbG&cd1_V#!%Tw^QB%{;|F(r2=zmy)7;D=x#
zCE_TClGgW`+YFB@zmo7JuMmz$;rh6o_hZ~pIU7mfcE4p!Kms<ejBjJ-z6n=CO)u-~
z?^Y-M`=ap8XL$VLN@IqrnPivE1;5p<=>(;vwmQE{54C!@?=XP@3hSCrS~#JgGU+3J
zE2&gDwtfEfyowwS4w~ekYv$Z<s4kk^v@l)K7Dym{FuC`}3PtRwH*%|nl6eM9_{W1k
zWgg9rAn?L|&dSOXq{CC2==?=CfK14F4%Y4L>@0+ai(3nTR{3112q+{fY78(v6SwS1
zJ1VNHvGqqo!onytz0?L|&IVT3)<p5})Ya!OD5U<>)sa(B013|d@yg%#pEA?aS2i|S
z3<n7i{Lk@Ut?ljtM~aW{-o1k;si<ndue1RX`r%@ux}01_R#qhtWG*j%?TOd#@9!5B
z4F%eu8q1}}Y%(ANii(WH<D*!FEAA;GkB>8Eo(04nkH;HBSnt*iYoLIy16L{@ORul5
zUpBK3LPuCwSXNZviV@YrZMTalYACO*{j@!j2^>;yudmM!7r4mDJIBX=-H$JD<06AH
z89=%a0u_~&wHq8yhSLS$P>C9VhRDo}HgQ0!(b-NyLIPjphsV21+jqVF*F%|q>g%H-
zYTVIKa;fP_8L2oz<P|&8#;`EgUH;TnUSpv_8M<pX#LbMr!|CT1*)vE=Q88f}bw?=%
zTe4|5_E*eJ6EY<wWrj>XfSsi@2?d*B*iLVb{=T)bKFi8_1DrPqi|KkUXdgq(zgZZF
z3Evdw6=5D-n(3O)tgn#L@yIjF0g63A&+ltOENra1Cgi-lb!O)8GiEV4b<WZ?7RC0~
zz$_9?x@V-d{Tl8fh>DWdB<14|vCN*H-rXIO8qQCpUu3O%@S8n*#iRfT8$~tI+CIHl
zo%w^m&uJXkV7wqcA^`!dq`RBBbFW!Z0_u~hx-t$<YFu)ZvWkLWwu~w*cH#h_$nNbL
zEv;@3t{zQe$HZ**GX#%Lb#HZbQIO~6oH=rn%LjJoQBzXB!QIX;EEI(wWBy+Iw@ae9
z<YqY98SH%IA;b^nxkVwX7?B1;PvddjVDn^f3$x^>#5B+|pdcqRG}t)C!w2`})s&*{
zu*E=#_S)+c8qsS)j8qwUdGU0atn8U&vm!GtwHim2f&%q7%s%|t<ukNyn#oFBYymXX
z^tdeF>b94_{_sRE%Gj>2XAs~=lQU|hjjw%oPyTMH6N8kVitIwu$WhV`Idv#D#sD85
zw~n2I)dL;}{)H?t#sf%pB6GnBZ7tg+ATN($T_GtdNujzIM;nueFkQD)9i(RXhuv<b
zX3rsACN3%R^3csup9~n=%r2J5i<tIRbN`*1n(`OvT*tv=W<8Dnr}qRr0ajLKBbE%G
z+0daa3bI`C?5xdAEFhl6H8!RZ|0r@di6MOI*x;EKejeNpJV7`^$luYdts60<eAm>l
zvI@u(N`dzki;jw(m>M&_Gi+z0p`;8B5dR`RWVURbojpS?>{m@#Os4!kUFHog;T<n+
zc}bI&qLAvjzVLnZ*siaJwl?u=tf<(fN)#Gu>Yv)rTS>q5O}U!TTWuxe)+LxY3F#l#
z4_5}$aweB#X3x~ess|BE2x@9RARe)SA})XW@Mv3qb|Fmdv%n>LkU^0ds{FfxpMm)W
zB{?#AX4t@hYiHL4e|yK>*zEO)q0#4WuS0HkK91)HJt!}@H%|s@W0QFhSX=Gqt@L3J
zm^VpSnTxva<Q|8F6l62q1tlM~y?R%`w@Dch9r>F0N`wUSTJPp&3s#Z!`*`-m{KWzR
z%K;Upn1!O2Nr%4)$-q~(9!A>P{lb1Wp{fpRx$*_OB$sV_!{Ikcu%V7INr#++bfL+#
zXs_<ApWoYf7yOxcp8hF5dt9K#1&7XsQ*pFIPDA~a_Ce8Blop!KQ@dc>FDA`Ifzn@o
zt7<?!$dbSev0Z;WJd)<-ExmyhF<<!#=_XLeQ<l#<2U!@IImX>GH_=9n$q40+R_QU{
zzBw;fV6@Z?uZdCKOFT$bif_eW78{!xMXIhYim3|Bu9HCY=9^@j<QqD6k9}(LT1unI
z{n)RUR;JQGo6fu=d8rP`e4}v~vwG|HQ?75!y6|jH4xL|G%D-Lmm1!V)U*ud<2D*4z
zrrOxCDN79}CByq#Zk9r7f8q3=4_W$em~Kz&9*|u;@@1!&){l;cw$9=uawz|PDLg$p
z6D3+gA><7A>{EUX{{pBEfsT*>1;{H*jEv5KIaLQzIrOh$>h&oNs2)v3<ki%~@FZ2#
z)C%(5*mnRit)rtufg0pP7e~ZVARGi<AQ(_60Y|yf$lG+8?X4{wG&C7CHEJ3ffG2S1
zw8JB0GHh&Y0PhK4a*=O5cgGnJULrS?!cz#e@pAbf*aWFGAg`I<54%1+x=BhRLtw{D
zh<7`|c>o1ZS=xL$0zyJ{f_o5G?gOy^kjrV%D+>$ffJ4dc?$Uhpb$%@{V+Q(!&W;WQ
zUw(7*0}vns(IqfZ0VM(SXUt4Yw{mQCd3kx=-QDf&?R9mWz??cK2gsb*TwLxpm&}@d
zy4H~KKPoaS{R?0OXZW7J@o@`#qupASEY(r;(zSgZKGJ&H6Y+QP<zM+jRUugG4w89x
z!_8)zn~^K)jm~6U;*UF&-F-{Vb`gu4pOM4aPu_&sDEmLu*Hk+MpxxG(qG#6HZMa5h
z7l*T#_DxAApS&*xJl{jDE3>;d%<!b+_J}z2?=ukrqMQSe=>tuI^W+=2CiTW|h)$B}
zGKUM-;zFu7N9L#a-|oKqJQfB9iVwPIzJB2Pt$pN?#zW-GkIQ&2k2)tJo$}O7fLI$b
z)ckZj_Q_ZP?y3tKa3DpilxC3ic^sQ9bh+^8XnAxKTyXE}?Q?tZo~;*zPk%}kM3zt^
zc*K2}3J618o?NJHe9r92i0AXD&Yn~NGWmIr^ZlZ&Z0$`33}hMp+%ZESX&CzT4J_RZ
zqGteJsX5}gb(y<kbH3fj5`}*SF)@wGQ-nUa%n1<R&spD+oTqxmhUkH+wXQHS#)Hdk
zlJeS=t$;9mdNta{dsaq9Kyi_pd|DC|?05m|ANXj#Bpb;n>^<xu>Lk=L$cJ(;<FTH!
zwGlbWG;$7<Mo+qQD>G)3LpmoUL*i2fg{HH2Ew0kY8YM=w_D~%6H;J=fo$#bSk32b2
zcr?BFN%7GS`y7|6BJgg8??K6eot^#uASB%bCtb#I_x$BZGtAx9qaSzJ((6tmwA&&4
zkDxFc)%EmnP`PkS3M05yudJx}WVfT{LhOCCb30HEjC997TLVj||Mm!2uSzKBy=Hs|
zM)Jwy>qk7Frk7eA#j*-AalCdvwwK1X8cj@0=jvs^Fe^75eOTPq-U!kDWNXIlG@!?1
zI21SAv~jZ{KtbZu^o})gp{KK;vZx_q$gm=A(;#O}7`5hpQJ}9Tu=T5%N`>oKjOO~1
zGOxP;oJZj7eJ&POEVt_^KQra@Y;;g!utQK&OvaCfZAWVA>LD~g_=lPg*>6?_Fh=vP
zgVgrOLx~AbdHI<a=R?%59?R%kzy-aGkB`?ABsDbLx}-<EO##a3#*%q7xBW+E{U0$o
z!gtsRU3UCnDAE4&pNYqHAJYH*Ds%Vu4f6m08yKDc?<0}vF){kwi)7uc{lEnk<_ACE
zn!4Z+f$=tZ>HS$*Q8u2%t+(Z|5u9@_I9sVo3L4f|tx<PjDKW9il@6-^c`A=JVB@K`
zcTi)ttWGL2!&goRfvF2nbXx6ri%2gtoTW87V1Q4&HQ~|Fw6~Z!=EKSn6(72GI*bGc
z<d^jD{~qD+6TPBhG)#R>!W;Z*kdJQfVG`knJi`MI{z<mU5@yL?q}qY%kt4+QF_Z_6
z@sszzuO4p#?90Xy4a>DQ?UAXS(<=3YM$>;ER#WeG6p`7e=@Pq2jE+qd9scX>t+MQA
zEiJ8*MnvC#Z#{3$C@(LF{7YKe=(b27&nF@(S~gMD^j272NojHM?kD&gxBBkw=}Y=H
zzTBcBiSya2#IZy3#K+uQr~k~xs(KE@n66HW=TPcUn6$+1=6uzJVpB$`{JYepMXrDE
zPiI{TxUeoiF#7Lpz^8ktfKaSd^JUkf@c+%|+r_3w5vJe~lK+|E{P0zL|1ZBam%f?W
z5&6i({CiK25F<7`DROeRO#t>HO6LQVR?_siV{=5TOfBNV#iMBs{R|C}7L~-2Jo>7m
zr*ABkDQ9zby5#FubJcpsdF#xHl5GAgE$uh^2X;YbO<+H-otgs4sH<;1bcS`A-fz}?
z`rnHjR=%lcOdu+uHz1Ut=19V?IjYfjdMDs*mFfRvwQO{zYM)#bOP}Pj8*Heq!>lWe
z6dfHsGBQa`PxWBMnDEW}u2i!&%c~FQ-p?8wB{4IB!iLZOJ5I6`sOp>`!ee(BF*A?*
zBk!U@zzLawmi8@%;UjarSoU*IG#{|lwH__vU}NnWnIi8l<=${<)jKIGE7#v`y|dQe
z+T?LrL57&ACk7@uvJ2n&iXpfWpS0bD$WUS?`%X1k0<YF~<s|NJ!OV$tykxArj33zB
zyNAD?`Y?(G2AZ+%U8io*Yu5{H_V)Ej;_<0TDJf}b4Gv7`uN`rD40{Rv?+D0Xc(4gq
z7%iFsy*g*e<?K{5NE9bF;4Y7N%)UQi>oPkW@-^8!S>Z1^Y?g{cEmuN`&jptHi|Y{~
zTGcXMPF5{fXB;>#@+hbC_8ma|bFX%ATL4EGJj6{jbT-PuWOX}^V0)XpHa|T4+kT8S
zUFP~sIx3-g4OrLH_Ddx8zd}MGV@x1aXZ>U6U}0vqes&`Ods$IYb+gvO2z@`;WY+uI
z+`qq1&X|CpD^gHE86;inwDtMx_im6~vAI73jVP44pY~UF-v*8VWHmUIGjTvi^W?rF
zx3CYDr}}?y6Ak^UMu*uThkRBo;fk0CpH~m`LWxy(AdcK=F4peRVRCr1GMZV{rW+Ul
z^S~)c`B@7nU=`5~<SXoY@u)v*xiVi~fRlz=nrw3OF}fn{h`HS#ZwJ&z)NSo5iTMmb
z^~k_Vw5KZ30qivIaYPk`oOZLyP?8TIAItma4BGXI>dF)e_1^}VknO6Fn~#7X!J-dk
z)JOsPA|+|cnhXnx4&+?z&hSuiMDwSg@6KpqW|BIe6Oxyb@R>|6tA5P6XVz~6HrvkU
z_$W}&<^Ud>b9-y1$71ahpnbYNHu_5+@;T2gWbc&}<AZCfzFM|Id-bfPDlrP=bxHI0
zH{wblu1w{%`r1DHIV`m?@P^xceEv71JPilIX+mW^X?eK~-X9IlkV(Cpy4%~N23Hmq
z#aF&>o4Eh)7;;Z)j{Z2^=uZR9|8WbGSKL?}Q)JeZ{oItRUL)XAbM@Vhikg1Y%1O^z
z*Z@033dXZCxCp#P%dMFjb8=XLZLm}>mz~w{<T#pnS9-<4<AQo+6>}=LM}9#@gW;W}
z#X=Gp&{*mb%`}@{hkKzC@wwd0T7v0gszi^i-7-5mpI%YyScu&NI7(0@S`LYBk_WG*
zh<aa<x2U7u`Q1;-3KVkA*@RV_o9CZ_VccBHS13Fp2K$xoJ?aaH1kLpt4thcSbaSo-
zSh`>%f&SbrTN$f5nQJa|($(h<BoO1r4)1C)pW@$SuwlUSCQ@Klvw!767?5dlr*-(h
zUCb@9c_K8&fnjQOHL~)&KYaLjv2p&_FU-C^8YrzH104c#&rYcXDwEb{5TM4`9qsfQ
zl!Yd9ziUcv&-w*kBk%me1*_lciXdmo<=>$Yqauc#o}ngyr3{S9eNO+dbKq_7E&g#k
zx-E7Awf78W=l0=uT|9!VX>R^fzh{becUf67zLmvs&dxsfN-HZf-knG36ovrUM#5XB
z9Ciqy5un48Br#hkFw`8z@?MfY_oTw;9L5a>`xQ(JSa;1UMy965&!OCc!286Mh?!2T
zFyeWEx4h8rl`Fzpr<WP38bs0pl8RDF)w^f9ZEGh`OZ)V1x>~P)$IE}l;rz<Yg$oZu
z7x+*dw681ZzVq(iCwS$1O646O0}lZ=ENs+aSD26T7*KbAKdfhP|9-Luf%CnDobBIA
zZ$=0SpL=;Exi9QcgNW%jF$6%i0Brc6pQos<Q!y)n!AQu1$($J_InL+nxU!xw5CH#D
zyDq`}ayma3(%_H0rat7+#J_QsQpCzFsEZiFu~c~hoox*!eY8o4$?5#JEE>+}jBR-<
z&Am!3=H$T6C#NR-h=+f(S`nJU{TG-lJ~&Nx_VT;l#-GLZ_V$MDFL5}2TC$|1{@)w)
zX_~C5$s{FA)~M&GwYPaI?|iW>b=n8-!w>GL*SKj?*)Elo(55@>i}~jtR1A%7sm-Ua
zB_&L4T%yP=7h25I#*9I=S<(PlG0Ib7s^a2OM^nFAGBU|iPe^>;G85(BjVX@!5gFNT
zWLwHxSW4r1uZR@pu1DPL)%(x(7DWHbci|o(NXVRO>J6x2n3gzW8ItM_HtdRYgF1}y
zNU+I$8>f6nFdgAKKzeRfJHKiE?_B<$hx5)AJe;ePg&WsLzX7gEi}`B%wd(#NGne{*
zN;QZC6MjY4CmX@*)2D2?8CoUv#UH^r<735K(6qzh`RPb83k|lo3h!HLSm^x1?(f@-
zMSl@7QATE#i4K?dl5e=I0kK8WFVjNSK<RS+B@J*F1BL6iDb0RETMVkTE7@0xDHaPw
z?P}ZMp;Bx|+$192!Gec}laiH?zx>4(?oeIbt#MRWl$JO;ILoMgZ%@d4UsQu`0MEXz
z|8B+GH34<`fj}WsC0f>luctd2a(S4&xf>lh9>T7Z3?Fk96|CLko4Fs)SU>Yg5>e35
zR2X$|o*3c$@7?>jCs~?_=o+ON8l@5A67H?xe9D<tQI|+&9xSdu`%5uIh6IaAaOw`e
zD_5GJzh}U7{fzCgDQZCHVD5qo$}1>%-jerPppL_#XMkb)rYl`s*|N^wCV}gYf=eAA
zu3ifB)x6vB8z@JTXluUX8JL_2kr{0#-u2Ixj|0Yl12kwnu#i+wuvn9sQ$1OnSmCi;
z-=))|EsR76I`B9=T=JZ>rZoBVtzltegLOV?R7pM{%2Z(kdA5H2nG>EO39NbnB?ZP4
z$jLz%a<x06>H7T;Y@($!Y(O<zt=F2wf?wml<0gVkTyDMND(fYJ1YG<G>Z}(wm&30Z
zv~JG##`bPIK5}clOB`^%ZMz%jT0@1vDR(n>8@+y!{}HEweu<=%c%k;K8aWM@!{|9j
zhDY|BU~oumiU<Hs5AD7A1k8V@5?Jiod-%G{qh`h+8_dYi6_QVBI^INl!BSNPp`j7T
zscEHX%S^j=!1zn&80z6~@15S<SDqt5%V#$cj8>A^G&F^f=Xhms>J}{?_wQ7&I>AW=
zHumV~WEco|<pO#$)UI4qQnG^h0-yEh`_aaxk!9c3nFj#+v#;<^1{Q&dkgRNaQ%O@u
z{f(7l=H^ET5{&<5#H?ndBKY+?Ji;*aNy;Lni-UoR3N}Retj^V2>~2_VXE%gPP7djn
zZ$=)8wrtNc&*9%cVd>czse~#Iveti*uh|=V1V@|A)E@a>;ny4v+T!uhzTKVao67Qe
z=5+)deA<-*LIdDDK}0An?lEbUSAKei^cho3{I5~~Y-O8Q&FlY}Z|NT0Y@J4{=;&+c
z=pDIPhI&p(4?zv?Yd9o}O#%ekBu0^pv3+X+N@nzIrOW3WrQ{fHt`Fd)u94EW212}t
zeP4$ln;u7La}lO}HPr-vUlF%z99LdQ#Up8O>nssx^lVk}bhY0%C?fz^QcA`kA)z26
zp<tz=<EOsCPlcy@{mS>z>*yY4iG_#6{TVME;c<|%#h2-N6mnq+Dk`mt;E&ej<sc?@
z<t5ey)sX)lW5ax#b7v#>42~DG$^fZatY14xrUBR(wj?S%3aA1-*j)xwT$Pk))?QmS
z2;6Ot^&Nme*pecI^~BM^HMFg)B3h_*J<5p<EI$nzN+kKe*|P0_duf&Yh{Nr%5fznK
zT9_%6E`#1)_l_(#YvA54z_*qvM2zm5-FnR93-HHYStEjE6g1F@aOo1XfaBr%wK0@e
zT#N{hVtl0Y_J2>@CxH*bV}YBEbw88lxn(copOPBxe>B`<3-6cY=eu2=+PTD{{O<+a
zYWuh)|KCpNGfNE04*$0^%o1PzC*zi(0p_mwcz+TSc1O+ZgAk7$8$izm>S!IKw53H~
zb6n86Dl1DuSXfO!vzV?AC-xU`8jhL*Vypt{-leh+8(7gB8(5O8>KYoLi5qCFmz74c
zv?#L2mtR)KNW*o-rPKz@EP+;qipsAf6ns9Y$^)l(TTV@&b!@z9D%KaK>mBUbIKi{4
zWQmtBn49YD?gv&2;33)GE;(nd{v(e$G2;z<(!NSva^Ou(jgN+bI)!2RZX3n=dZ}p&
zF|pP`BGObTp=58NtD6bPn(|`1seE`(Qt_%BuDnnFL{Udt2sAVt+4>vS<xD0f1$>9y
zfZ{nhIT1D0+Y9jbSK4J3-(=0rlM4#U8<OV3XjW=k?f)}9eiZEPuK%js^Z-0G=(pWg
zlX1nSh)|ci@x^HQDfoFv^G`XU!)Fq-7cUxVsOU_s4)@p|#|#Ge9=B>)Ee`Irm2(CX
zq#qmTwV#@V8iChsrRrHD{SsR0w$-BPY|%~Jo5RFtWd*53mZ#X8zUC=dz%T*tXtp1_
zWS|RAPUbS7yw5IVdmLahSwDmMu^1hz>^h+aI7`nyAAxdIJ4H^GzC<F6$Aw?<l1y-5
zZ?9*X12@r;3t`JsIF>1!bRr8d3f(K5yw!jFce$}{8v&;D&1ql3R|hE<-Hp^!VZH)v
z<DfwCb_&+-wd=wZn6H}GTmagA4cByba)Cpvy)he75ALYh>aNKLE4F0z1GuDl57Wh=
zz+N%FCz7QaFE9n3rW5H^Yu&-ef@d3v)7Rho+Rc7zbHif-fsUF!h1K-&hrVf(a(rsD
zGmrVCYuK3<-+A{wmn;9d=)-nTP^sEgcfc6<ma8+=+jO#f*6C2#f0G#PmRDyjMspx{
zZqAO=dHJs`!qOGy`bLeJnn3QztQ8KhkW(&RDu1}TIOM5gOWhluvG2=aF*g&0f#fMd
z2$}MZ)^Zt+cpXoNetC@lZk5s`=;w^P*A>}+!gb=9p-Y?`Zqg<j9~+w~Hx4NZJ`baJ
zvXz)GxKO`5yl+gjiLEZhLc=UK-aD+wMNy55XN|uwN1Y2xX<%QmdIS-h_Jun0?VW<6
zLbaLN87eb=Pj|mot;LxiOCK!i_MB^p3?)Fhmo`E`=4m&0X07?)4~sIED{-yA$vo2O
zOOo$SugzWBl>S+>xcF+<W3;xJ-bJ%+aN`8>6wZFE{-&sX1?sbT(yW$qQAr8px5}cT
zCsV3p{wRk_F?P|vFu^FiBkQ=<^0nGs{i7chS)J2~y09U1=6-XNE2<T1=_-(?zxqeE
z(WO>ZHSS>;-5vjb3J$?Y7as8($WUaY{L=Nyy!?V-0^T<eP?-qz$>je6f&A}oBe3X7
zLX&&`0+V}5F*QyQ2~e_maBo^WeLms`Ci|+i1~^@V=4P@Q0I5UA(~?&&%q-T|)hC@b
zld{kfHI7dysV<L{`}*C;Nyyi)E&;T6f&{xd`CRV_K*_fJEhZM!GTIi*eFlIkY;3G3
z^xnQ6R)%lgcgJ8>X%d>BTmf41_~a-Oa?rgTg20l9!qoU1Ts)P8M1FjD-?roK?eg-|
zL3_ePZn*3meFa73DD-kw$$GV{Y(oHVy?V*7sd*4O8M9hJO-&>4_pOsgN_2L-R})dZ
zij0hkj6zXPK}KfrOK1mIHJTu_sU$Sd=CdomoPiQ66!!){-u?sj(pOtES1BnXUopnL
zuV!X?l)OL%KtfF(DwfVUI~RB*PjaUyqav6&0vZpV+s=*eiBMwt`HEo0{sOgs_UXC6
z+hIVT0!+a8_=NGvN#7>a*yw0Gduub6OJImAo+1Gp?u^HW@B%<->DV!Zbu#AVOJC5g
zYuBUz)NA;7XkPh_lmYpb6NphS?USd?u;aC~WJOUT3Idr5XK;IZ2PGxMKw~CTtrh&6
z`{SQ^tC4bm6fWYTP8}__tD?dGR$i=kw2hXRl#1@%*m(PN|GB^idTR2+)=WddUyh)n
z8-RsO#cf~(Mx)R{Wh*qq<9;6tdQX-+F~tUXnv0l@xXb2aLXN~lstO9D`I%Vx2pMN@
zV#CL~*Hl&2e%2X*1R4OZ4W}nCo^jJJR8?`QVE^`xHqhmkXI0YDa*L5HCnm<@;kr-u
zNi?|c?(GEfJ%J%;Mhyc+jY%c!49=)nZE)OWq?hDm=RYGZMHGPH0c|*m!f^0YyT1+>
z)6tzGz5?Czs4xTn+rv{pK)_XpDXOQEnQfqFAS@sxQ_fhk-WFRZ!P@#vXBFK@p&(Ph
zkgP*`C3kl>!6E${E%sP`kieWS1~AG7vSRjWuW2y^(VmM7<Q(!&037hAe_-OohB#+b
zWb?;I%ge#Bhow)(qnW%teZ8Ng!5TC%R*WAlwxOU1+@T{AI={MJYG4y$bm71Q2-r}6
ztG$!L#MorouKaa$MMDM?Ew>yfDvd1CQ*l(4G%b$KD_DyZ|G93@WoURtLc>MzFNNrr
z<odbG4FSXmq`sBs61wkT0DUB5M}wY$nKlEIDn)Ud-m|as7?-hI?SCO8ES5Lq=3&qT
zMpbWE9b5cwMbjszrr=)q!uzyY3~CC-@=-+t5)^n{cSJ|TE6VmTvNA)j0JI?{Dr#qQ
z_?BoCoTh=&Ukp9~8|UdDk6!o}Kf<q80FWYwB%B*MxytDLA0Rm<A*NM_F}aS@H6Oo}
zo$o2F0tX=2Zz>EV(npMl44-D}bW3aV15!$)U}Eu8!v#xD<{R#}ryZ$%XFE;=+}>Tq
zH#E4C2W=Qx(oo#3=>D8f3JF2_1n+gkA1@~d5`e&r_lgq}NBnI-v!7fy?V-32%r)2q
zU>dJ|vRamhHSbUo?LV-x?Sbw}DbZ1dMOj&;xfr1C8J+yspv?JC-LO$5IElwAk>n6M
zTn-}UNM@**$iO&Pmq>}IivyE*{%0VH2@)6Ufz{XNbK8u>uSzN^3Ml)gBB<d0`J0F-
zjT;Xkz^Ab=zya4mqbFwKY$q8~63V*sT0^x50U>DF%E`~~7TbVO5YFekUry<^Xpf+F
z<GE)@RrdDUJ6_U!uRa!5rtsLPcU=5p-RSiupmzu8)&|Tp6%~IyJpX}@ho@1;8R$%m
z=-bwv7Z+J>Jr_D@%gOVRMqO0)1nxq*{l5@K%kU=;KQCU0+x_x1MB6i=U^h4)Bas#(
zHN4Vkj`)4Thu<DYlBUYgCpF>CAKAoeB7iu`?Q)#2E@Le=yNj$2@VGpMf=?vj5fP8(
z3u03iQjy~5;LVpc2Y|lEOdZXDW?OyW_;rZ&v}&Sj>S=uk5u-D-zmMh^5wxIwC-4xo
zCH1@od=PsQHN<YS@^)r0MBK2diuu_xXsw!pQyLlIHt%A#aGp=+AfeH=d!bGZN@AtH
zs{U^EHGzB`RohjiPt}@t;C|yI3<LzcgXCiu=2)u}!gclr-G;PpAWtyxniTjas-Wit
z3UUxhX)JBrNB4gCk)r>i1P}yT-^ftf9`62jnIk`iJk~BBK#Lk6(2FGVM?P#e=+FVR
zlkJ&4o0M3JV8U9*M+WG4OH~?7O3;)JX@5$c1kLQ)E%#{d)5xVi(r%5#2Kp=C(eAC%
zncrmF3oau?prk~r%^k*{etdJka2E!;xJ9<6o!vag3#rp<otOO!2lENA3r2zKvCf+4
z323%*PK9Cc#I1Omo>yF`UVC`eo%hB%5VR`}C7hNdIzlDlzykd#kEc|@p&EWR`gcNq
zYEp?4BO5djpiXB_d{`Zk94SIhbij`NRieW5G@V}oaubMzlL;$2T3p!z^I^XzkEGdQ
z>*)Q56JBoXvuYl_x`_1F4jT$VeAmZO!@0u;U^EE9UL3vC3a&g4{?|+SS4^2|QQfUG
zj9X(SoP7Z$jk2OLol2dstn9bZ>k$f`!6;w0QtgRmO9n`#&g30coYCY%q84>;<SDUL
zE%(Ud;f#k*gX7(9y3sRWaFgtU8WHmZd@=ibyt9f2K-<yKq+=tqbr??>$|TUXdWcW*
zx##^4{t?(Iv-wvLID{NZocK+pQm=0JM(wPbAy7IU157xzqEtXG)ULN?bSkg2z1U1O
z<h~)E!UH&Fxll8(_yny_wd?FnZ+{^GBH@dF1wSVoyr#9Qi$-VLT$%hnCCk+lN3I&P
z>1r1bb8|vKp8b?gtJM_BpKY;FX|Pw~_FkvT;|hDLBoyF3V3uh+H%I7va0Gh^tnS9>
zzgjqqC*X$G1}R7X`t{t?oC1G@{tMW<%D&&ISr_hry7%LHgWHwYd2kV5*g;KM`7+{*
zb;l-mXJxU#gqW0CpzQ-?ANypp!3Vo|T4@8a_c^q8)5=ncLN|*c(u4v^MROYR^7YO)
zZD6jRIEFySXO*JstDxJTR=JLV9bgQc+wR4L|B_Qtv3*#@J^89(1M+2hTH5DOX&>7c
zk-p|Z0yak0_cpjlx6ykwLXDq2)SIn8X*&RnCP2N>ipF2QLx&Mi?rHlu31*tbg>}}-
z0qFg_1z30D3_QJrrfDxH5*>DV$&vrjxGwf@`as(Z=##c|*V(z1wLgv`KXlCLqJc@y
zyXQ1B*JXJboDZ-Riu$j*`c|jiTGY|K8>g)13m2>;Dxh}Jw^6RrA7UC)e`Wj9^Yk>Z
zgJvohoMiPF$x_<PmY37Af1$LTYWs^@y4SLd>qPVvH1)ucj2<PKEcJn6Nc&k!_lh@w
zXmBA%4J{`jApsa1i1L$agANa5w$xw&)_c~o*T%S0)6;)eu97Wbe)JqjsU^4>9od_T
zBU;}yoFIZwx6%bQB^<>F@TNkEXMwa9kj^&gv^15pP@vFD`yZv?{bR-IAk^hg|30%D
zLVEOW5f(Nv(GE29P;1a>1dmw6We%K15oFptO+uL?k&$vZr5e&XpeD%MaPf^ksAs<7
z$w<k55Fiv9Nd%xhqq}B<n51II**Rm+%Re?xC*VQ!?6tz&S*SDyHw<?7s{OVN@vY6T
z%DySCCq7iBKzaH$H!{1nuMGwP)Y*u5^&ki&sb{ezAfs-yeatEdJT0zA4NGFDtBbd-
zte$T+EDS!!p&?C{nyWVT06M4i#<?D2Ix4z5{Z?zvWc#O1A!yoW*@$0<N!9UTH+ww=
zf*2Z>lFr%rtf$j{ji1N;c>yy0@aEMW&`KE&YCf928@)q@x_i6NzuP$7pKpYJewZv(
zo4u4DvyTh=PbOpz@L1YYLeBr?0z?t2Pe#4z8rxcQx_JqGwJ@>n7!w#L7Mllm3$i)Z
zV)d1~)uySLHIUtgrBp9!-h-0>oDgT5_xHCmn&2Qf?lyV?{VaokUE=k+laWe01E2H)
zpwUu>+umM3L%&w3`+kTe9Y&Kvv`c)38A|-|J}|dyA6F6v4y?`FrD|nsb}(A*kITZ%
zygV3@a)a(jC$6A@`^=`fix2-w_8A3uU@U;F+_D@FpfE8l8ake{8yaF`*1I2FBQP%2
zO<~tXPJ<v@DsglM%W6`uIBC8i(Sax~(azENDPgFh9)JQveGEuJKVdyV5-!m7_)F&l
zD2BTN-WX^i?cFs5N=MBK(CG)zTJzwJ^*2b_l*23=U2O}c8EXDq3!WmH$b2^%uD=2m
z^2XFfCvdH(%CX1YaUjxz{#+gJfb8&q%fZ?CX!fsq?D>T)DiKd~R3aQys&&`d-Wm3W
zvwLf1LBxR`q@SmV@Fpy!y`UGsyG$L<IeXWK{ll!2Ou%Sk(0o8sBuy^?mu@ywnc!pK
zx&%v}n<cuKmLD(zMsUslk~vj~^?LcF;;&0hsR|P+1lT^xs-O-BAdL69Imo!<M>7w5
zG%@{!dFy6NHsR;!)>hUHt2<@?_|NJX@GlJMH89~{h~o^_x(vv3kNC+@YSbIMfab(t
zPuAjXs?r}z=r7s_Wyj}GvM?M3h@#Y-mknJJvVZ+ORX&WIoy*c7+CVCeuB7>OuZrzh
zbeOWipohbq5;m7om<BL_Xo3tAe(1@my(2$JhVs0@5gB4vR0JfSl@mM$M}o%NSU5=I
zk7KmF;!CKGb5@y~2k>~I<ssd>Qb%V3u0Qx{7zF6{t$PI-Kia^#cC>U}Ud`-*kLG%?
zsd;kibU98c$4AV2Bg3b=iZ11t6dMz9$?dkkxnTnks+vPyOjML=6}uhggoyMe8Or&l
zEzj)QnOjBisbUTORYXv^BP2Wk90EcQg33L;3ipdqJDS+-y`ASRfMBBK+-zlK?RvX>
z`+%>NfVX>xLPSx#Yq8{_ASbswKe4h>EUrno-@Q<z2#}b~?**{ub{>v*<vD)Nl9I*i
z8=^@5^M63a5=7bfTqe^wRej2I>YuV@pQUA1ZGfleR$pByV`pp&If8Y0BH)7EpNs%$
zDO-epU)(UQDKs_t_j$Hk;PW^U)GKHEj<z3@2Q>+yuZs*Hza+o5yy)CQT=exz=5R6s
zBqr$l;29Hi?8uFt#!>FE)hAM;L>nI){>!>8GrCA?gwYCG9`EqF=Q8?4JT%tTE-(bZ
zNq}9Q=f86S@d!XA8{J&_2(<pQEU5=Z5u$R?K!sep`upJsOg-oeCZWOq{mUE3QLFBj
zmE?~DL(LW{m}}NrvY)?g5&MH=_}TJeX0Gv0{Kp(g*R%V@WYZJ?_Jb<6#j^FY<@VIA
zw;4LZ$i(-nD<OS2QX|`ybX->vs_5<QoiLtx0;~I}xg~1Vzr6!qLtjK0!GZx`>m?5y
z=d4PE*cUYwkPN!uqnYWLU7wyDIN7+D6c@kiQgUrpqSyScw<LuZjfaIHf@it-=OL}A
zVZJ!=46M4MV%jesPD~?=XlvmjJ(YFj%C|a9GewX{qr>~ltC_$aX>}Zy6Hqg4Hd(zi
z$@J!2(*+f<k<8}YJb`8ROl8hQ(WN|jXyfV*qg$^JcA>4Bny{pK@z##+`n#|f!G1`)
zaM1hb&mqwIhJ~eyRSDY<_7>P<!G$JejTix$fUEUY0f=EiE)!PaAxa5FixG1FW`G*b
zA7!ev8ZWV4`SNJgM;Ia2n%i6bHZ9BHalb6EJR;iS2{ke`R#LG3!cN6Pw+5>70QBDk
zAT*pqBQm@ntA_0o(%gU89(j_09JVz)gAIXj3RF62=xC-UNA;NEL4}jfK_^!RS_bcC
z4bnH)w|ZEOuWsmzc;EuzHRvA<JW~cFxIh_w!&QPvG=<dE#6-v(O%Q<WqN$jGMnmqs
zPYDRm)RfG`!tcphn<|TdgNWPo@*|LXoHS_9yZjM*WA1Z&hPrxoQ=MFdkADO1d?C4B
zqshMS$b5M`K$omYsL9#>Ku7u(`@#-1=N<VtqJb`DvXg843R>UXT=KXcUo07G(>7#!
zW7C5)ik$-x)hzM0Nl8<XoC_@^tFdG0G%`iXjS}&=&GdAdocSqcl*X-#veT39^i;%x
zQ{Vo+Ddo0oEW9R+7-Dy@B0<Qi=Wq~}?1M)_!3C(j|FE>)c?C|Xc@Z7T!NOt;>kUR%
z2p*TSqoSgnv1)BV1S#}WiZMj^r6=f`98SzPAR!6LRy|=o=;30Ni>3lT`~Pl(YgPrc
z#!lh&i1?ddkhlJOn)a#10s7$93+fW_5r1K6sHj9MR9$&frE>eeh=Z5166*j$Lfo&&
z#MmJnMt~X7l63$+pRJ=FbG%TbEj4xb-p+1TRu2B|UVcHrix>XZwcMv9pm$XaZCqw%
zF*w$?xvq$#qur*dcr*#g<zGSL3W{^=yqM8qSRrIt`T2{kbRh2D2-+%}0r?0l9IXGZ
zn3!jtz<B2>Y7Ph#|20q`lGc>0%KMY``?rdgR4DhiNSoZk%!y7{W?uJ4=!4@|?NL}0
zuxFH=xIEl@gGoeDkec-{2v))722S|xO(z2c-PT>bN3EG-D`lXjj~Et`Qpn7Dg9OPa
z%5q$HRg;GaXUzoxB@YOV+D%Dz8aKu}NV&LhAP^^&E;Qg1ptHJmd*UowLSM&^NkO~_
zYNqv$8#$Cty|Ey5QZ99S_H{q+n^@qhiX7=$_q~*LMM?!~*9I$4B9b8Bie#pH{jEY3
zOk@6Uh{cZGyn}J9J+q^2^Zn=N91QefUHJt{86#gRO#$}<L?PfaNUh4q8}AdwMkaMC
zW}vj9qVa}7P>{e9-vpygN!7j7V3RsN^!q0zO0Jl+vifH&UKYB{%qpy?o^v<+O6#5H
z2+Ev2x{?p4&lfc7A`PvTadQ2oQ(|2ku)pLd)QG<K&+u5%vRFudmR6{{<(Y;i$3L<V
zfFwE5(Rv7y+ixrXYM~M6urK}bws%z3<6Q`rq;mid{)o*Iyd#sxxKDSgr?YE{kCnc4
zu_Vd|Mnw@Qdug$0ILW2Pm%A5BFpMvS2Fn3#2S_j-sDnT!<hJ1Qy`X@GnkED|Es$*K
z?rN87!`xjQ9*$x`1caP${tXN?G%%jAi@eYW=XDt%9>SHU(5iEm9}mn@1avnL4eP9-
zcQ5q(Ba(y<ZQtJdm1X)ds52|OI$;N8m*xs^eqjZ=3ZtGMwRb3lsgHg9D7pW<6aeF1
zoz62d<Hx_3+}PaD?S7DZY9|a3fkO<x8M1acn#jBZjal2?W&lhd03<2RuE@~)sKG?g
z6K8p`f@{b6-z;ufMgxhpevo^#jNj*nyQH&*{k6SjfTY5De`aOh*EJy;4N$LB6(^A6
z2i;k$GDV;kKocBt(xRuMfrG3M>H!^~V|%K83q;83joVGQ|Am;w$us=U-|u}cY=kj*
zWTv9xAp5}1{DB=40&NNmhT=v?I>k=a7_{$*xUI`GBHtyvHQA7Cx(2rXmsSp@O{-Hm
zqI8s$FCem)PUWil`q`Uws%k2Qzu}>_w#=NTavA0by<g#J4dv+2mn9kqSjY{-)^VzR
z^hj_$=?%dk+Isd0sL%jw=3o74Z09EHSS(Pl2Vj8f0W~jm{5t?`X+;1Q5L#DTBBK6j
zw0SMYe+R4y%ToS7M15scR$H|8Rs<wPMG=rtLg@x+kZ$P~Y3c4XXb=Hu0VxRyY3Y`3
zq(izJY2=%pd++$hI6pWJcw_Iq)|&IFiBIsP?b@ZgpwFrMPUAvGHee@|m)9k$-@%J2
zcy5MLOe}eOxu2e)*WuipsXWF*AShif{#kjeW5)d8Ae+mD-HYL2otjXxd0es*@(2lO
zfqJ>i7h0lOLx_;FG6_H*n2D|Pwz(j`6ceM0?eYr`!p>GP56StOnA{dDjX=V}qg#CQ
zD@9+jZz8m`-1GNP5ZGx&t3_UZ*Q7^&hv*8-#>}fXX?1o=QyR2>G6<geR#Y?q&5OGA
z)a`ov){+WuD;7-5U_|KDkczU0Ofoi%`p;0y*o)yTe}J|AGIw<ki+)Ow*vF0hGB>6u
zl6d7Tm#T!knp&NcyN>SX*NWJ#j-|}s9U5e7kH3ysG2kFzNOeRb(ZU!I0`rd4{co9>
z7zq7QY~egLRhFmgM`A8_*U4Ku0lD8&aS=x%i1-WcT7aM$051emZXp>WpF=}S-`03i
z1&9zWOH!ZWNAdDi)<*HbJWA|;r1yMTRch*+=A6j>xVPe2Qr=7?Ten=E-$}wRb$$~<
zhIw*<`Iw6Gm@ER+MIbHu8Tx?e;hp=0_ezXcULwp~8_b65Eg)IM!O<Diwwkt?Ow6O=
z7H@vlU#Y`b@kS<75h^ruiw!{Q#PYo*r-`|8&8Q$9cC75!7NEg{lKxHSWBvN~&r@A5
zKSde<%vj#`8jgLhKSYQSoTY1TxheOla^Eksd5Tfp@P(s<-}QnXh;wBTFWA=f%B(|H
z6kiG_8(p2+lAc~iKAMzx96NekY%WbUxLlqeV&B9yE{cO~r?IV1NAV}2eM5-0g+>OX
z1&)^+x*EZG6PjG${9Gv*%E!})oL_e$`>+%-I&h+t|2#5%7%#?npZ0$jRGnyjNf1}(
zM<I2I$&Jk*kgNu_ue$6|Ja2v+sXS=?lKaUcUPI^zMo2vT5cI3Dnemb3{vyHF0M#=*
zSw$twqBnb^0{W3D{KP-V#EF&<PWec~JB*~QF`lQP!ERchRrQ;0CZrKTw%d-)JrqrQ
z;#i%BzgJvqEZPg25%BiWtz)F1V{@9<P;%XF`+2Iv82<d)o@hxL(_zN7dZC6V=(at<
ztx#2B-oy+FlW|>IaM>Kz<ly*Kx6Ka-o6UyW#s0xCw1U7D>~MLbhbIpX4zF$xMzg=r
zDlsUMQAZy*il?XkS5_WbUNO+|ay_GDXXWV~WqQ??{0^dQ<m`#LiaciLU;;Mb5VbrJ
z7Y{3s^B@1OB>bZF4L&I>CN0SdLc>oN44V`gV`WIgAoe|8mZoF2QtGvTmxZM_X-9<{
zhiddt+pWMtu>VOWXA2+t)oG%hkK7Kg|1p!5($YG!zGN{gGitrIv9>*3sVSjt(VdYZ
zbtq@)p{I^#KP900ob(o=K}IV+bVkGDZdF)il&{c8kBN$?sA@^7Nf&5xsQxmcGJ4K^
zQSO~tYb7xtdPn<ow}ye}bq!86qqM-8zUzJrE&)wV_GqXt6G9~CcwdFeM9XR*tJ=CG
zS6f{@>kE0`(!OLs8E1rRaN?+f87CGb6NSPwW%{}d0Qui<#U(t53jMyQhnv;8@HsZK
zl4i~Oy20pzdS=t5qg8Nzt4Yxho$chU`uGsBvvyrAHMQlIRQ?VcvsC@ta}fg|jo|!S
zcQELu45mLyu?grZTBGJ=DmdS2qu*Ts*t#^d8&o!Wd@iiytJArEQnq!C^i)(6O(J<b
zrzq$jn?spUf5FGW#dH;m+ErhwmzFLJpgJ~B_~;1^86m>F6H5pjtC3Mg89PTy`KkW)
zs9Mt(mz}#&O3<4CP!hFLD4L^PH=c#2mXdM_n2dn;(V!r+hN5tC*Agz_aJbRuN8DbC
zQ?eR&h79il+(+S+DaB{p{>9r0JEv4bFkT58fhjr^SCK!11aX(sg+ap6e+-a58Nb_{
z)2Y9e#0ESZNzY2#nx;x4n{)6j$cJ=bXoVi&<uT5yQy2)eVkiw+5;~ZuEh!a(64$vn
zesNgQKkCqniiigp4i-}Ab$O0WdayCC8aZlTT7l~O<2@zJ;sE{Umlw&ko=L-6P56;m
z<b15s<aoQMo_n(@RN#a=tn%ci<28j3`Yo2}!s=2Q0UGc+;!*tpgh7V|5q;>sfG8dj
zF%K?$Ch)|*4_qh^1%sf9Q^7pVXXO-_dpjBpItc)$ZFH(V!D^j`BQD;Se7;dEME-!^
zVAW<ry*K{!E&@|)uQjwql+1$>?$@x&5L=1f`B@aO<vY!gPWbc*9h>!G;8x<O+(Rx*
zq`26ga))3oy8qX}OEooW^3|jhn!n}usqT21c|cX?kLHdDiCZ!-G7>Ud>g))o=TeQ3
z4YGav{yosK7t@REQ35c!G_9{7B^4SSc*E#V(aE*@7`PZ9)y&_<Uz<}^9vaC#w_{M6
zvzYwB$IbMWB-F{N6vWt)zT!1ugKZHL6IH^4tG~TmDaRWtt_dST-Z|D<g5gQ_WX~cq
zK9bKx!UnTAx&}SnA0p=cLYdJy!^8I~(=ftu>_qgn^J&G2iUKv~PTt%r8+3%FXfI<O
zyq@PLr|K#y!SYMz-z5=AS?NbP{Zxjm@qkODm5I95VGL#mXBfC`9mL35E^;+h_*2%Y
zSV37FX|=L<pbo4+D_u&t9tYXFY@$Vi*NQyIVxwDE4&Z9SSkn}~W8Co2h|m7bIU)QY
z?L8oW!lehIAa1*}-b;x>uUYSdhM7A@;Zq+Cp>CsraHst}tn7jUG(_)8DBR&CRKrm}
z!~Yt#EkhY`fckc-#*-(>+SCFCvBQuu%HM5&pm+7f{|v38mHgakYk3+#cTgUlNd<Z4
zrN{{;x8Jn;a8wS_S)iNiFQ1$aI3py*KmZq*8nIsr92y>G-g=dnpGo80KttAices!n
zE7RC`(q$*sq~waPTtQLzFH|(~`O`C0A9WP1c(p9d%xkc4o?PMu;GFd{1x=YkIoNJT
zak{2ubsGf%RNk~i1Atqpt4=jy*S9SRCg$eX5gU{lRW*B%m$YZ23gVXk&nm0XF1ANO
zKFY?FIXNxPUb~z^zqLwypYYFNWkrh5$FgWQpxF$Rf`KoCVCMJvgoNudEDwP51I(Y6
zuCRY_xV?+X^=PpAT~LI!i|NOcrNMz2PUdHe?HxmnsrpV%f4rurA^YHNN}MNHrB23B
z7vQ0xnTteq2o1yj`GLs!p!Z$oH%`a$jn#z>g-j)%WX{_N9DeOTo}8khBD+7C$SK%}
z5zahK6Eo}Ok?HY1!Pnbq_n-s4M0rO?%bhsFh#ur#>BPHc2VI5X%=|xRCAQu3$QOKG
zf7d%4Y!J!5_8Y#F$E1;bGyzA}D+i~*hUxl#gAyM`sQdoLLsqA|?jaz7jR$mERyw}o
zbu{ROfQCYKJ1wFJxb6!uf^rrUQ0RZzJ%pW{!ZIt(j2!9R)IiY3z2bH{k?-N<Ub%Pb
zsdq(=_D)`jPsRS(3Ln#K)%kZ91$D;Ql?}3XBi3qer)$1@hT=85-UlaZU7gY&O4%pR
z%Tf{&Xx9i~R;B&SMklD2ffF4b9<FmEhFF}9SF;deFM~Do@%VQ2G0DhqKkSj<6alCl
zghwfyZWk4PuVeCOpnYvOl-2d(=We}pKYqybqyBx?Tgbk#o+{pG7*O762DW|S^m^l`
zAVaXW16agm1JQWxQ7^rPuZ{nHUqk_&yWQ6EAt$-(TyYt2_&H{4|MatnZ;wgOiCJXZ
zczsO%_~OVkH=_Ux6&Q+IieO^nuO!$;Av|otXECGD=Y_8BG`IqxZ+-1lNWwFjX=-FO
zUO8GG1t}m~Eq3WuJkmsQKbxBH5zm>LOSz}m;h{wvMt5)Db6Bn!oj-Muv{wePdcE(L
zB@0fixOjL>+c_TVze72zgzLx*_NK^aVh#GQKEHY4u7F1(^2|(s<k8&08Mn#s3j`FE
z?pq_%P)dRW0F3_Nd{kCcBjUEiLw-~}OFF1|c<=7<!ROY`xwJA4Lx2e>mJnemoaiNR
zm`%MV)4PJF1DH9MRYzzUfipcn+D1j11F%DbgLu7W=-^v(bI5_gsHW!m)7MKXIApB9
z-?yCtWPE~ND<dcK1jRMtl`;fp>|sceuGCzA`yA>{_C#H6bqUO?%(o7eHL4pv=)RDB
zih#xbV*)35PU>nsymO#yFdhk|Nvzo7_Lv(4_D&@_8jRNvk!4ls{BU#Y<>^Th&jR7S
zyR#sB2iHSj>N`sM!n8a-@9sc|;_T6@`q>Je<;S#iys?zbF|VfLZWnr90FLrRA*I=%
z#N6B*NbFdq<R@<V6i?)%E|w^TJGQ+fVCED0-E^@76Ri5FZnKHSJqQFOD4gl;07>d8
z7&$;k)G_2LE?&OU`fWy_4p7f7I$Ez;H1pTlQEw(qF4?(hdA8vKK*<K=eZ%h;#B^A!
zX&-L@!sEWT4%1|oPjX7u*ok7>(iL-h<v*ky49;uCESo@60Lb${dAqQFL~*1N6b?ax
zeLX3>%1z!lWIW7s4Mot7QKT6_;k&rB1jf<&yG?(UHbwy=2GJYJzdi-AeWsdm2bXZH
z5Wu=g=dHr*#XQZTv8Havr9o8Uip_}{ukf}a6b92TD2VTB+z&Pa#%n_Qinoy)^&SeG
zKg`FkceG|H<-9{Xgry5in}VLg=(3@qEWrvRV2CVs7zg(YmDJTe76=-QSiz$U4A09`
z=qRrvti{CuvNmTex+Hm3dz&&pu;lEjjokpM0Shw(&Zv+H+3CuwI3q`CDLEA1u2E}#
zpS5NKmSf8+xu)COlhrGzg>-NQDB7-Cj977S4iWqM^Ki_^>=YE~YeQKT#sk&2NMp-l
zvdKUocQ$HJq@he5YPO?vA@UCw`+<V=|0;6$ruj+~@Iy$Srtgt!O(}(UR{0eDxtb@k
z;I|v9-+C@BXfgmdyn*KTJx-R8(9BQ-g#yLveiV3rc;g-r5)#33C>qNf9Aq6_eYN`K
zFL-Ps`@zu~r=*0N#Jfa*&tY2$Ot5y9EpBX=)o**8Lr3^KE_Qa<p%M}k%b!W+MrT&O
zt<ix@BG6TqvR8+Qm@>$P!FcAD4E+01Ica%Dmao6$=Lk!$zOu>6CZGLPR02LwO*b)p
z#iFpXC9p?JCPvIh4uXCfemytM0De8SKzzjB+5S^RgtLQ_qN*a}Q(8t^S{6>usiR4v
z^AYbM!=gCdhddB@0ZxPC?IljvSO4-hE$FbY*oHR$31y$dYvrU`!)vAG$<B1uW`V)E
z820V;goM0JnCu}9<_*94US~V|8wgL*Ojmw&X{e*OH|*wC3KiSLqx4jJOjKg1z{(fz
zW15*AtuJA54F7`S!h=^^mxKY7Ym@Svs;a+W3<I2foFn(U2>K`p@o*0R0M4D8`;CpY
z<eP=X7LOLtK;cHuVxpqA)<(j_(~XUnV2;<Z^G?xLMHwZrUwlpwK%ZH;v=nUB<pnbx
z1K#l1x>rp^hWjh4w?T$2X{@PHcD??VJ$&Hx1Q_tTJwKfJJsT?NItv2yNQp}LSyOSK
z@>_ava7Iz~yBR;$%nYN9%rG3BP;5b!=ipW`q-d_WxGv{Z21=q1TT+Nxm`EThnEnjq
zdl>egEuIPZL`#cH5RxXB!V{6PJZJBwKNmqyt^88HcdTDXPd&i0zjIJWTc+fj$V%1O
zSyh$098OuZ3_(n6EaX~zyuAbk2>bvm5!AEb!=sC3Bu_<u^WtMvl&R(G-!nsC?z(vb
z;^pg6SB5*)%?Z+V^$M7eD2MySAbbfwC++FcjIQp6milI`%VwC<R@&p>w_k4F3?v`q
zm6UJ~crSE)ZHEkn<y&Au^|tr%<(`|TXVa8NRx&r6!!kaJRq7RmgEG{G?<Hepsd>l7
zpVHBIJLOJjXa~-<K6{88(i#~PLuz6pliFUSK~EcP^7>1__JxPL24pggy_oI&DJ|la
z9UAIL@UC&rw&E-=p4Cpr%kXY^TS-Zg=m*1xB$lY2!nZ<t=IlYwGQ@oTMua}(?oCo_
ziop4n^@%ELlkkAVgOM<ch+{oa<j$?jK9&j`%!*8VjNrAj@J)RV65Xph`<c9*9I?$<
zU4BNcH@&})4{Zi*G&POXM3M2kygYhEovGdtX;m>oBq2d-(a0Fy^qwHw$ZJb;#V_qR
zE~R(f8-#(ob9Ny8(D*7h*Id)`C~kn_*5*v`rX#Qf_=_|u=IF0p4W0d>jC{{0*4x;t
zrRGfT(?KncMu~yJQ>!d1YdQJzkLMd9RWY*@W{*2kO)X=-ttJOQ`2M7tdpKLCIXQwp
z^hf9edGbKObRZq%(Y2cn3kRn(W`v@4c2caG5;Dp%A|#3lb#(<msli&Gz*AkB`|uh+
z*J6X+KesaUfjH$IkaokIoJFII0Nx3!J<Yqj&*GZrx)b@WUTtF{{%2MA!4G4*jG5!^
zh^FO-Exjmv1fNJU1?iVbsnW9Ul}CQRGUay+H0O<08|}}|%lKh2`LhrcZ|~XEYXvTD
z&GBv0m<-DC)(*SLuYcNS-b-&ZEZ9QpD|s<0p9$QRWgY{FTMw{;-QC>-=jVYV<k3oq
zj5JF7@gp+wZCNIFu&IrOp`WXag*I`#XmLduX<F`Mg$M^{d;FUxs;Xr=j4tyhEFR@H
z&w1ox?ti82lqMlL%w5;w4B1+qFEzWQHKNN(OXKTq{Ll<q!BGR|xcDJz_?qXHh4X!r
zOG|6=)Dar=F>m==e~qvGp0RrFhg&L*>r#5&vR6}^@5CGS)1)Xm8vl4_#=zWCG}8$n
zX)c1S@^a_M_wSR!s0RDY&8(s*qewO_K{Wcu>0i>~>zcv^`Tx@buoK-6nVv847b1U3
z+s->N`B>nH>PzN$WAB**+T(w7_<juxI=VRA=Qssl2~E!0e>YC*-;FB?zxl`Ufvkz<
zDLSYu{2WXn_!1yLft2uT<WU_c%%{$@`K5WEpE5Ekq^`a;0~t~(_SRM&#9u&OB$!Q~
za^4SV{i>7$on#Z<D-o4Xkmcl5%AO@p8qOZ{qN2L`!^g-^dEaa{zqr8nom?-u&)rq{
z%_h6hhCDdr+f$nL*JiZs#|6pT1r#rc@7=?HfbIVQf_==Q-v9kgOZQD%OFL+OUPMLJ
zv6P+0z8F3?YtZ64-{+a})+PI%9$DE?Y#0d5RKm=~!QS6&Q1mjH)QD9|UBcs6a0-Tj
zff?aA^-uq1lC3!lNx44Jh?;8+T|-_7!P@gze0+QchKBWLqzJuMU3uO`b43kzpfI)p
z-n$1aSLASkU-EhF+KkSVJ=btu+*2m#-{GK(?LHTz_)OG7j>ZyrRUZ)@rLLl5u5S(Z
zOhs3j%VhwQMyg{)VaG?clNnxW{=9!cI9$fHZF4)B%@7mq-MH89u?;R8t~}j0kA-kT
zL%c`H1aQUcy~4L0h?aMMDWI9||9j>C{Sg2Cc)|5TTis`^f><ds^6pPJ5ENf)0J$JK
zDm#44aFjPfyFR<wsYNtwbP~|sGYUjJeiDHF#OKD}$tbt()d)yY|C3nJgs*KOBi8y7
zwH>sm$pR}1fRsz>5g^6(`-uq&S6fwA=N~Dfk|ds$mzVp@A8h*EjamK*6T&)j2-}Ay
zkM)Nox$u?qv#Osk&P~@(e=QV|&>jLWhL=9ezUe8MUu}ML4IkBCt2<HSmm*k1TWi2~
z8@z10nq>G(<_BBOOm9^|!K1=2yJU=70LsHciZrsn6%rkp5D}%mYwFMdw=I2&5Sw~`
z`~TkoP1)U}$cPABccPf&`237)NI)6J@WNF?hwF23Q&JMI@FkDa#&&bleB_byfAYP)
zxW`;y%og<;T{>UGVoyqXxIs+hw|^Up?XCk`RNB1KqVHKsV{bn3>rz}r!k!;R>6=#h
zMn*=v&vZ=G3SNjCni_KW^g6NziBjHi*%CaPjUE`&(k3EW<-h5<KUxwQo3%M!M?=N1
zAowUfo(H-Cg=4)DtG{%t37H|wQ5|>o@udxw4}6}BFR*v(|JrZO>ec-_8#@r{=o0af
zyw3G>{XpVirTcqDS@OY<y&@rwnEZCuDO=%e)$E@q%^KLgR>Y$YKRyc+_QZb4$*D6v
z|4MMX@UK2$SaEUW)t@fPNKrL;();i99a-vV(~H=-%8Hq*9Ry#*y=|L2vYb^>*5-Pe
zPWa!GDSSbOj`q%ugk{R9W!I<r@xKT9QM>ZfM6SHQrk!W<gmUmXOFI_sEz*d!X5xg&
z85o?@@^XQWg<hw4{hakeAoL;IQq(7rHt;OZqUwbR@<?D~V`5-qYMyruLL~HM6vxFR
zi|Mios$1vm<8%mNvLoTxlW<G_TWz;Hwg!K9aB@)}*)b(Xhmf-zwboB3*tA{-2w!`n
za3a`kdNLabS4>sy*CAc{_zx>4wBJ!h4o;+;4ik$Ww^A;`SQCV?L-RIke2hDJZ@G|1
zNkQUEjq}hIv0yxlQyNT(%ML}Gt8RA2{co-Q--@}y@XQK@1VP=yYHy>fUU&yG0XviZ
z_jX`Xx)z?CHu*QzTm*<G0|UzjeTO?y%4C8#TWk6>wC(44=b5E>^D#aMLVxA<rab2B
zPG#3HQ%f+>+BR!vvr}K=LeqnNh&#V3U|>o&0LO`o)xo1ENu+;#+-Y<8lx%>5)VddJ
z9j{8p>t}!Pdaa8Jgok4)sVXM(8y3gsU?lc<-@8k2#`m^u<r|KoZi&5=xSG7G)7D_?
z!L{GLwE?is{<$3?os4&IT2m<!d{o{{zwNw4DE^U43<-j~k62DcRR65s6O@VUfC{WJ
zbq;xDlzny3REdfCshrdN6^t&GQ&Lm9aO%PI(5AZ|;&s@-h}^`)02>Y}@-}WP)GX?A
z(nE|J&z<FGY6FB(=n^HRq_@ZR>CPA)-bQSlNClh^+o9zmBwP(6U7DGg@Sf`|t2#qN
z;PRJ%wR8D(K%<bj$u`3kf<I?J6{iTzTEztl_My`gg$E(6uN0E%l~i;v5k?Asx!ofr
zm5;E;|F~k#&d$J&Hn29o7`hnW$FKYv6XV;ERe8FjSOuK)yr0AHFMnINsZXWd@Xy`b
z)<_(?{^oZ9msxe2sr=jm;rip9iNV%a0yjO2^|z!^-K14yKlc+4W$YRLba!*^UxTY%
zeX-&P)BIu!XD*a%^7LwT#5C8sTW7VjCy|DxZlm6A&^wGbrlJJd>wNZ?OJEjHex8&6
z?f90fj+WY+L92X*EIHD`+%(yZRae4$2eivL2)0XU;xlVmNsSYzLKc30_^M<EuP&so
zLlxKpr#d%|3)B{x+-_jyUEN;0hM@Bj<6~b(lKI0WrMiUbBU;-;ZJ6%FgE0gkY5&cS
z*Pu=|T|Sg@=b1>-)$L1`X!%Ly$+Z_wxWr;o8ju&-9nhKUaqFFciW*A8+e$XMP;ACC
z-4tGngvuN{DYMf`8TD!pcf3mQ8AwV?Kklq?-XIMo=e0R<WV=i8X(b-ZJKt(+^>`V_
z!k>5Sdc%jF4NZDyVf{T=Z!k3Nz(lUAv%}+p9ly4GZS)u11H-x%-s-A46;(LyPfN0}
zglrFbQR;rDKgg~y3>aV7J1Gs_1a8LB+4)+-n`!N}B;8wNf|pJ`WbKfB=<5DMK7Lh>
zI%@T>_e+*O#*X_R4#!dsQ@IovmfZY`IVC_sthEJa$|KTfAEMtkE$Qu^O$Fqd<7eM0
zOL+XG`n-WfAZ2Hsz1t(yRKwNC`tP}}wilcg3Q2r*kWaFw+c}{s6Q#^*J>&J=tvO+^
z7?aoni-qs+{;x!B&k;w*{#&HHJqvpo8I-Y%2{y2jSN87Si6Y~34}N<UcI0?q>r{8e
zENK<EewcF?-r29Ker36yEG@VAC^j?OJ$}4WxKY1pA~;e`a!e*wE6u-Z0v^IL_F*@^
zms%!AaLKUI9DLQflo$VP6yRmTyE({Of<hMH*uNIIMX_?#+e}i%!GZi<D<MEx-R|h@
zwEUyV+uaz`M?h|mi4`s@EhVAA{ji{}pm#l5hO#3A#(#mx@{_>ebn9f7a`Iz!U1g6Q
z{r)c==fZ~}Bjra>O@kNSiXqJBYs=T=m@|KIH%u{+%MD@34bQFmnj}rkBkJ^fcM-Pm
z*|$HxF8lwahCR392H~k7DrX|_r<4~%Q{6YhoBy1f$iKXn(g)IG{3m#S-_XKLY{0L4
zis`hpCe1h2myC={*{>dza|ilhLFpRYNy{2<^d;*&z0074+l){XVy`vm)#T+zrdy?S
zbS}s&%yhj(9Av%~N2W0k4qcIdpS?8J@iY)u>(wZXFrgL?A?P);$(pG-N03Dtd-dLw
z-J|{P&Rz>^vrFUgc`x1y1PKzgcd%pS+Q4zLHq&2To@YlSnip)X8Rcm^$tVJFRF$0^
z?4G|_<mFjaA?o?%_JjH9<t&yQ&Y$LU=hZdySalWaBh3wt&gRA$G-Pqn=(0EEaQ`sy
zgV#Z7F^G&fLhkooL!dlP3>g}hS{TjK8JwCj&^NJvb<7E=hKNRqeFwB&kP`Y*{O2=X
zZXG@^l}K4}Ia>!in5&dF>!!yb2O<q?)_810<d?rrt!rmX`)^gp&u@}@<D&vzv<Q=i
zkN#@KyL)$G&sBveOgg=2{GrMuh*tA8D;K>Kr7U%tJ_Rkvla7a+H)*|XAlS_T7^~|Z
z$!+98TJAh_1$LLT9=sdd+oYyX+JKHpmuSP?e0g;6_Et2Y17(6eDNgHSe_vj2G8lLQ
zdU*0ENj?F+w~{w}?F=PI1WN-X@gWCe@CBLs+9rmh{Lrs~D*L|~1s~Lm4lO{2agg;v
zX>k$pm8Cw_eP(80gm=DdW$ituaA0b)L045>ohY52iWfbo6xlQZWj7iB;8w$Seo|7B
zh<|pTb{V!<eMwpQRJj?*R-<8n0)#PDRz3rdAWlE=7N9pFLk|>O4p$a~X5e;7vi_k8
zJ#ysrb`1v(U&*`*`m^YRC(wl;4UN6apLg_OPy~ctn3=6zR+q~s@{vN~2mZ9q=aIj~
z-M`$Py=$~3GO72N6rUT**C_|0q^|U$xn_K*FGb)jJF12yXSBGo%10LmJEF8)dV2S~
z(RThxt^uCa+loc5eA~lCmq+Dl$d~*~x1}a2I;;{f{uMC(1if`z=>Cboe2SpQ*-TQR
zCjHSttRf7GjO`^=|CojNf{YZt2I%NAGbu4{6k#mXz~BrV1kwkam(s`rSu3Yrc5WIl
zXR{{+MZ8H2iF$uSQ9~3xiBTL61<^uJl1^s<71dLwt5>FmWs6yE?Q2P}SwPVQus9lC
zf4lOcJn6Klfu)9)T9Nf1dqr>i&HnA^uKUJIi*5CC32YAQ5njBCZU3(Q)|xXslaIzB
zbOG_r3fO>6wYus7d^^v&-zRhY6#l$l%ukYz%PPve2?=E~oi-=Sj0aCxP+WdPt-9Qs
zkPsh5aK4-$7VQTQEk4IkWA8*yE5_Oz=<2E}pW!H~0yqJ;Nrs@$#@t$y+4N_V8(Vc%
z9IViYm>6hfwQA>AzJ4`<esUqed6bL&1JRm^p)oE}Cvn{OnUjY1NqtF3k^55QIGZ~M
zChB!qAX?St;~*Hz$jtbmQ}F&hWW{`a)3yM26#Gz+v6g3oti!+Y73oYxbJq|F5-7Sy
zx8jvg-~F<!zNfPzk=J2jh9nXFHEYCE@43T)aki}c;EGXI`O9uIvUV5UPJ=$BFKh61
z_ckf=%GrftAEY}dMnJxqkvPH>$U}x`aYF!UvnV55l1UOLa#Y~qvPvAa$i9W@3&JVV
ziE0ugxjvM&f3({3=zVeV;PLG%^19$prJ*$fjb}c%K~gyuUgcPYRDZA41bdZ?n0mZ=
zvCCEGv+_5+Oki^1>K&RFtiHAQ@l`#2yR~}VBpWdk=YsDOfQdS~uW)%k#k=`9UGLb!
zpB{clGeDaWy9g(z<+0B1JA(2E=R!-yBJq1HbWZO_8GLTTwmDE_<LyXXV!uDgRzfGB
zk}Bx1j|o5eRe0$=M^CC7F|FR!Q*h~votdgJoYhL%@@d7sA|4nRC>nth-^#;BbeT|-
zyXn#W%Qk}&$Q0JxD7ieU4e+rE`hyeJq#YxC5e0jmvZm_vPWr{M7<@N~_&yIWKnZWt
zt}tsG(M7)~nEreD^{O#FYhoftFgq{!R&<)YB3`mztq#1Fo9l46^_pIz-aet{_TaC{
zdObdeeY|*49{YKEdrp>#A>HKXm0M)CTO<5P|IjZxC(^%}BnODr@#^|ZtH9i|7iDll
zA3fw_q@%z=mdI$Y{T{U8LS8lfD0~=GN4)yLdGczmxvf#$|BUOR_mLARlY)zwn4e7j
zw2;dV41#emk*!_0`q;H_9cGPAK|^3xip#C3sZA`yzz`N0S!||E4WrpexZNO_cJvol
zGEe#Jv{MT#<t-H=1Y&&A1N>72e8l=q`yS)%DtxZgP#^smrA|&kaWHhDSw0ferzc)k
zUZ(fM-m2z!_3QHe*^*d4RaG*cmS1kJCeGVb^&9_&Yk7@a!qE^J9d(>J!ty{CehFNA
z{QNp%LzOzD2psN-<vSEol6+YE70g_y2-r}yiMzTtb?|>iOJ{a;DMP)%%UhkzPK0R0
z#cy)E>ZsTmHhzxhBegkxW0;hryQ4Iij5d|PYBwcTzwr=v<zx*jI{Fu?Un}$;Y*}(3
zgbxXld7-BCDe@D&tKK1adU|^c#zrx^w`XL|)mdy_w$?z#Q`ZZQUxLFiZPC#Yr(0)9
zp?RNMu~wiXN%f0|hM{*6!|>YgwAnv#qkE@2Az-DUP-$H!=xe5kWk}xPHyOTccmmn|
z1-;G0#EFBf#b_r+GkI&s_w$CLsx$x3Qx|(PFKKC|Gnej_n9BY%D1xWCtGCAoN~<Vz
zE-2tG4#%;Ei|FrLER@%}KHMHR^Z{ooAE}_Is&R{LKvP<zNLwEwq^(^Ij3;FO@kS>^
zC8k6r*1y`81tfVO4e)>Dx3097sFAFOYero%<Ak|vBHstfM|QibIl%v9^}W0?RmtNr
zQrf;c>JzhXr?ELTJ_I}AIc|zk$@or)wDn2S0qMfp)o+&YE>sB1qOZ58kPE}{srN*S
zkZ==Q0t*|zjs!>%f|RI{_X6Pv!ZE9`^S*HUu%ehbD(eL-1o3qlq<?zoF6pWHa%qQr
z^X9@Vbr{ha8j{NN7%hA2q$nK_!)DV)@A_FdT8V5|Xr;N>`%7T!J!f2kh?J+sOT}U^
zzp|Ce_`l%Sv(7T@c+E;drq9)zE2#HPsNsLkswR++kukT=_#Pj+y-$7fQdN!tkxlr`
zKqDpKcBIt3aq{GmKER6(rgFEuZgUIegGj!r+5<KdDDmsP57e}@f`^{SU*5R>#<y<)
zMabPm-f=UrC|!b@k{T?E5+1~`KuE*&l+}Iy{G5h495C}q(3^<U?QKO`+I8Fz99*1R
zUZl543G=_DFR^!f$Gu~imy~1}XzE_v-D86W^*!pNulj~oZ*C|&hm@b>e|Mlg=e<jy
zrFROihZeL@d}8wQYLM{!WK)bX{~hW1GQ(iv`5k}mF*7rBh(-M@u!D|tN(X;9;)D)@
z+YPnOl&LSC`&W0%Q=X%qsE;dpnL<3Z2ZyJ05XWz2k>&kd9wBE^fL30?SQRIrIsTb-
zXi$-`oYge6006iik?=HbZufR}RJ#n>q*aXmnLPeWzjtf?+i&|y4_s2RwPnk{5*G7U
zn4ceR^`4T76QEorYA~BtRg6kfdS2MfNlC`pzSWM4KhfR2tgtwg8!O)+9~={d&a}Qj
zlf`>;wa{BbQ`G?;vwY%FuYyOaDoW0)C-G2(X5%PcnhRWEWPCLy{lotx!NAI_&IUSy
z>@6oa4_rEOUguxgjTIf+nwrKId>EPDh|MYe@T=Y$6#~L)D^BuAe$#Ptj|*K{Aj}?f
z`(#?RSGO%^>cV^0Ow;e&8LJLDo3z?~_qO@^yXQZ{2>;k{U0&UXTKkwpV69|3*1IfC
zHWa7tlr-i}CGM4Qdv8Y{(^C~Sr5f)gdcVhQbGF~dc1DQ^AB@_}f^_?jt;64oh~GYw
z(JgwB11g@`m>3@1U}tEWSDJHcXcxMLy#GoDi#LHt<d0KXMETXq^g(9#w_J-8&$`^;
z;GCe;($LcCjllAwqwVcv|DRudzUcYq;#~6uAM)^h((ty0y=&k+zJ-8$g~hP*21s5a
z0G9%ymZJ3y^jVbBg*KL@s4{OKyp)j<`7HD?HjA&$g#_6(FhE{`o#Vn7-<&s~xzf8!
z?~`;wg7&?qpD<=#0)>!K&-`5}6oM5XH-V7P&Gs(R5+sT+bFdHzi62nyy6PYxj=GMR
zxVT85>jIDQt`WBN^f~cD-P6mM-Ma#M2;wJ3m3Ey7PJg+u)a!44AN|5YkO$bgcMv!l
z!jBwNes56;Q>NY2$##OH8xHE`hnp|``SNlLg4~ku9^MzQwSsLtrBA~D6$AnyzA9|}
zV~;RJ)zncuic+2W4PUl}|Jfh@H2{DNHEi;(Cq}+*A*}{Dhfd>`xK&g#a(&WrJNH%r
zYC#b)b3=s~4343-q;-*BEv>B)H<=jN68lz4PA^8voz2W>-p0ih&5`jOY0F0|4%kpD
zJg9K!LQ5&DU%ZF!Ul-WkLx-odd_RSQcq9)&F#VpB2L^?N+{!E2<%VJ0K;O9Xy=>T;
zMUaqYXkPBxRl?Pg>bv24_LCZcU%k{Mx|^og_E;gMHjdkBe>K8-&n9Y0SKQHw<z>B;
zX!ye|D<93eo0r~utUOOaO5HDo7FN8PnVDM0h(LMyMe2>Clx!GIh)m}XK2YC_h`-Ig
zX8<=4kAzszVIrdkK9zPEef^i!@sOJt^u7-!{5j~<nDnp_dwg!Y>%!7X_Q+Q9%A6uy
z%wNZyuX@@Z9`@$qFKjjHwaE~?n+OJ8jtbGB*!rdc-pw0G!z<sm1;rfp!{_P*C@h57
zXw3@;larGxy$J$_>!Ob%Za${JK48@vAgy?<2NBAj+$=F^p1AjKr)XT*yKy}=wTPRI
zcXVW8U(8m>Nlw>NhmW=S;WIMD@}1W<>4jMbBzuWy;>yZwtan6}lgX*6DJTF2Zdf_E
zw9!FBu<v7eO_|{NFkyzsuoc@sz%~&vbhwzL{8z&OcM}<<-eBZ>>ndH|Lf_6zM&?^r
zpgu4QFc6qJlar$YurHo)+fUJ_?s)H^Tq<+I0TUXW)dPhKKwm`T-+OOb2@9!WM=7q`
z8fY+FXU-c<65=hJ+wzjK))p4(YDy0Q5wlFY_N6#iD|^3z+6Vucv$K=2@rsm`l$K^3
z6(#lP$S(-LNK7QcB2@W%o(A}{l^hNzbX+>jZoD014k&!R1Cjf1Fv4?%&YP$(rep4|
zrl1fGbdZ9Af~?N<fUYuDE=U{zJAKXD54b$;I4s)8>h`uU0UGoSH?vNw7ugwOEsXW4
z85mFzSQ*g9>oATEHeu(}AaI?am~4E-3^*M}*&iCSHP5<NEw<>T0+=`6Lhi(R0uMGC
zTxCN;<Gs<2=H?zv5wF*O&z~F3(cMVt>+7L=N>4)(7{txXO-uKOR>E{;8J2`vjESQ=
z__Km3<2UEFWg|ZdDQMEGSx#xwhKZBoJ>0YuT_S&o`-Fz7wxI5f=wtX}K%JlRQ!%`2
zHMr&Q{abE<R9`31mX*~X@XpJ{%v%RjhW)d@W|+K}5KrN8G0sQ~d4J>fZEn`Mt#tz<
zla)I`qUgL9>>rBk_K+7^Y9C&fl(WW0e!AhiIDD8L#{~vjC27iF6yN-?%#XN$BVYt|
z0Kx{cMz}uYXuYpk*;pOgjUvOM#SE0po#gV1BHl#%jaU4b&CM<7Xzzs49Y8#)-8WhN
zERs36kBLcx&wZET-Pn)!%Q0&B*3TtVVBRZ4yu&%@)7cmc9$@HNtE%GO2sZ-!M{rp{
zHdRf+-`VA(r=moScN4-<{e)F+XBrzVeNtE1N4;((2lHvQhqGfw%AtI6k%BCF0w=J%
zg!YGL08vLz+d=oV<rOo0falW+_Y;#M6CN?j+m==^Fnnw8?rk+-v1X+BR7Xu486Ck`
zm~Clh)z#j4`)+fHbuPeF7n2!K?%oMvpBj)?Oyy#wCk+sZLVefhS@+1&(o$MhJ}M^i
z(AMSDDDK%qd053(y8J<ubL`#1ar)9)FcZSw@jFbOQ|M`rjE!<|A$hc@i%re8E%&&#
zL>V}mAL;yqwPH`<gY@>O6Q>1RM~cD`?wuGnI5GpBzye>>ZN=Zmji#5h0x3`M8gd+B
zuf1BEKgH+PL_ThI-3@8{u924=_l&w^(zu0L5i8<W^e<?+JOySaIxn;>jb4+GRJ!df
z%-_9+gbBpU7MGP<np@K_(y*~`ikgz$pv|XXx~i>9{*W6C+SjLy3|&1QOn5ixyXC-R
zgO*GY*bTZPF`D=CJKeoKqrV28KOW9qPZ&O=7f7)&(+|ew4WzvjrBtMmR}y8Y`Y*ib
zD=27LrW9RO|AD@^nn&d$v#+^BaOsWjMN?v=r6+|2cS@kTaVAE6KBG~9fvSu&3>EBL
zVL>we10(EQ93Tgc`Si)r#jfcUbKo@h#3cUx*!%aR!G1b6u7Sa3R5UKtVFo-31j+!7
z`QDR*g6+wX+;xCK)W0Od3q?mnp!q@(z8@@oNF*%!q}FCI2uRwDucpz_>Z|LodB-%d
zJnHf;M4YYREH4HnK}&`T;H2w(AIU6AmmhNx3=S$m&5t1AZQM8;wZfaLU+TSLpYLpf
z*2Z?*FcNv8p)xo)VPa`S&cJZ*-epxuiHwYLvsv6FV$8wG0f}u?StfX`t%@En9YHos
zaQhaE27S^D;pJJF5JFl&@$}QoIq@T`CIc3rN^jGhtn8^9+7R52MuhC7az6T{9x#`O
zqZqI}K!mR2B6yu0AN5y|63{&ik8nVVWk<Swa$=Uf5vUKfA_QJsh4zQ^t)Tcmc!`a)
zVBB~lA`v??cY9#G^0X-y@Ho66ti&Kn0;CrG`<egOE_Iumvo>Z{D9GZ?^86&}jmelt
z^xAwrb2XX~1BD_f5&%6Vvf6##Rji-(%hl4*A%Z%OxyUDIY}EHmE!(MS=dRUFAq3H|
ztbga%`(D*Ya{l0&Z_nK4%9BLq+uS7@2VH}J8$d8qU1)I9>edqfPvtSiZ$8uO?A<eF
zaqnP502*dmSAza!(T6MJy<j2o6Ajhd5|p)uPyEq+U!Erv4s=I-BrvSyOG@&mX;%in
zYe=B_F2~^{v8TW-(ZGix&1o}2OIQ|kX>3#of_<d089766PXUHYK>-V)rog?T^KKl(
z`uIS6Yg#w@t(4(muZL%E?ggHqo@J!JW8`sOUF;vltg{XDJHb^bxb#-iRR(+X((0>k
zxoLUd(*GQeJEVqUgtXqwZ?yX{Ln^-T*G(-DYiao&pV!eq4j#`W){8^8hySMqxIwm(
z^O{b&i;lYQ%xrTt0``}3&KPv7^3$BBvskX7Je0c~Y>WEnZ#cdvbtIj?!PtoeDD7Xt
z#$_@br$`zSQz({+g4_>S+B<De#tLcu($PwA^T$KR+{j&5G;+K{Df?4Tfx(~G5tw@8
zP$yx&K?H<F^PFwFS;#I5=oa(*ZBW*;eH#Qn!h{>p32tz0JzRdZvAOu#%(~olPUjX^
zY)VZ2?Hp~vQ_rBSD~8em?T@ybXVkY7c671*pSC`p6?u2!u?OxiZo}t1yU2GSE{3SX
zZ(-nsdy`!8Jbs|a(FBb>EI4R(0dgt$Gd>`k6%sVZ{g81RCLAYQ!6S{vNXJ|88+w$x
zZEhSK737b8Kx|gzKt_h~=8R8htAUiXG!RH%fBMJvCJRX<!3O;&(S<dB`q{$|Z3{Fs
zw4V?7({fnwku-yuskNbB$A`B3KTL_wriX`%4R`cGApxqZ-N2Wv=Y|2!>91+t2tOhy
z&=CnTtQ8j*ZKZwMF6iOt^1E|q!iW743)H-Ub9-trKHp2d@E&}MiMZTIxOYs3_UK&%
zH``Zo!k)6kzfFnJVH@glDXiC!mR5`vMqMj&>q^`8?EEk)1QaFK<CQ70)PAVb5CyQp
zcu9)?P?|al28{fpihggEc^$Z8qroYwptoybWi?r8+#Da`{q!eQ5G9L>)y>#0-3mJi
z>%SoGxsE_Z<3f-na_)T5Y9R~pU+^EsV?`A(OtErVPbTWUc&ZR#$Q-}Iq&r>i#y_Z3
z={zLJ%jMIN=5~t=l|n8iaCHnT>jNlU5+<quQPZBBFgDI0l%)=D)2p@bn2J}v_NAZ>
zhICso<XgG&fvF>zCt$kH?9z2-0I{9#&iwjyS+$Gc%6o`xZ_AEQ6Le3WGD{?u`H=7e
zbP))U$1OQSdLloPZ;yFAK+=*0e=h9(2@B&rB2^<!J6oM`LY{g3m4@RCxe?$`ar!50
zwl@`JrbwCKqy1A%)oaMH&1u9xipI%v9=wm>0ZYJw-7~f?8TN8ZD=Ufkf4gnw9Qu(Z
z@gE7fmA=~#iM)^AlHmqW;N8^4{s{}f!^IlScmGy2yk&Vg`;!PtI7|k*fa4JLmUZ6d
z-9OUB1kKCBp4xq%rjj5&G3$v%4CK}8FBNu^$)8GPKY5>BL<8FKQSuHSX-ca>zIK@<
z(2x+geY1Q^ATI6`tuW=h?GHw@=AWJO-9hi5Ku6sPao!RP7Ld5KG&OYkJA51YZTk$~
zUP@Z3*`Vn98zFK$VBWT8%TPmF=k0!6N_cA0j}%x5JdxKE-H3q9QFUW>KrM3CG=bt;
zBOC{VJC)b=Z9?qjiLgl-=`H}Rd3aVu?e-k9KE!#+wOrMGcbx{`6hA%ved0SuN{5Pu
zhC{h}6`U2NkI$ea0j^Cp`FEImPC?yVtZ_)eY+bl{^5=vPKfle3ME8wM1LJ6KU#vn4
zsavP(naZnQNH2N3ujZnzzd=G;z0=SVuj<-OyR1!IxE?=&hR(w=WmWi6;hN?x&^1w}
z69OA?mq5%C*L2*RtaP&8dt>GuuR63xr=E2?<1zr%WKvQUX>9t@jhe_NtwJmPrYkou
zPy8L@Loe8JCPx`ZxY*(8aNYP>mfeOoI9P)F@DIPkJj@Vkf=G~A${LZ+m>t(~w<h0Q
z?Q=NG$ZUM54FJ&%i^GJGzJZ$jOs|*32a*YXQPri**4gCud{3gJ3|#rPFuexf7fo^&
zpa4s-;&Eip8(rnF;Xkx;*)Oi56v?L|7u~i%!spDxa}-uu`bmiVna}BQ%5slZrR6|_
zoC$OM2G{meS{f0HySM?9l}^oL0~G%)1sM36pT=i(aJFX%b@U<$kN3}s!I~eXp<Qzy
z_8nzh6Q{<1IrScYug(9ncHj+M_QKE5;NT#UMX)=kr7ZVd_l4?dgLCVPk{RgE7#Q+l
zqp5aW{*he3iF~egyk}+H+>{QkSmbvFR^2iy^*?ax&Ec184V_J!`;~WBxuV8vVauYS
z>42dAX|LUzRMs=!P1ZB)5WIUNF1uqirDP&9lxXkVh}k~=bky~FXL~5qMcZ$~8jpYP
z=m9c6OT7h%XFFwP1|?Gpf9kh%R4}saY@_le3z~|XyHi))E-hy#(-41mhKFd--=}SF
zH5=P&_IS@!T2{vNvZx@Z5*67N#CEqm=XxB`ZsGt-=<}yYMmjn;GJ5su%1xFceW>R9
zXA3ny=|NaJ2on+C<~b<#Ag=qo$v|`-h;uN@YDx_>0DK`@;h3@@VjLZi$c7tG?ZqV|
zrPv6UvnFjlk#;Zk&87s%9IiD$)v51p?R^btmy~%zok>ld39>d878dHL(@Zw8)&N?x
zcR62#Xc!q3WTo@-a$(f4sJQ;B=Sd}YQ%gG@0@!(X2VMW*3-ky!!SFs5Uxm-KsmxVy
z`2nj3hM|Wc-{?+I$HVsOmDt_Pbj8N=$B6Cs5-83Mnd87;cmsV#_@(_v_TX~~O(e0I
zhMw`>6FgXf@TXAs8XILq(CLUiP<WoavQ$4hIOr=N><`_=jhzAGB{KQs%4arvov@LB
znmBd!Sm9T>&8Gh~U#wv16Ap}~^R)T2NTt}@&HlU-P}H$kl;ZBPrkR43b9<<UZ(8Kt
z1;M{~l2cT%rrX1#zaju}rlm1<c&0~7=9U4ZpQxexVMD?1241a%b(BxmriM#=O?&z;
z+F^+h&}*V1uScV!u1E*a(q!*Qy=AhW0C+T!-w<<Ohm9GSWFq23i0^|FAGSou7|P7d
zc+kE5#zV1z070tQF2r};_z-gM<wu)R?GCI#rRJxAJ#f%SOH1kfaKt$a;H*RX$HuU|
zM2&}Cq;4XT6_R3_G#vbFEHq{LNwkk1HNIj+f9(fM7wBdevQn#$$dYPbox&_sIt0Nm
zuxW#a5F3rnckb33p$8m;VUM4zxNIE(r;^pD2be-2<ww1Hhs1XG;9|U)!Nhp~HHAzp
z_`$%9RF$SNNisxfh;61$Mp%vb`TfSL^teCYgh<55@m_4SQ{H(@Y)sXPF_Ug#zTwsP
z^5u6URlpJejcT7L)e9{4LWCH-_c7~+EHLK7pn0!o{|owu+;t);RtPfYfKUJByN0-y
z*08HSo(}=%dETk<_+#V&TPFqIGjhIXNKzO$MNU1v4VBrTLEGEEr=*^flMfjPH~v`y
zUHaER7VS=Q+a5xfNW)#AzTIZxbrI?x7({%b&Gi$@BpxX^<gfBsIGf3iI3^Q-S!$D<
zg+8}0grI2`*0-~VbXEvas11;3j-S@%D)6}b$8xWU8=g{vOLr=I!Jb4y(ud!ze-$fW
zHZ9m8FLa?ECcwg|VKZyeKz8e|9w1Ql_%s+IWAzp?Iz~66C$8x!TR>h|aOe9LqsaH7
zU#`l|D}3lK)6dPV803(XtuqpkAvE$&0{y-6YvtMINh<$b<28zDUD4S#gQ8Vx3LZw%
z%$ndJ6og6PmYv-mfF6*(0<%W+g^!<;orgQRxcz07cBPi^ay@&SQkL0TBvP&La3(~j
zB=f83@^byLqf3zM{RgBEnT^x9^54Bw)jg+py}Z12ev0ZBMIGvOG$y#($N=7A<&^^R
zDB04>kig(t$D<P)0#D#~Ow<^wj<7I1#wSMVpmt<>i}Ds_XF@~M$YhDU-1%bI4?1$d
z)oVW7{JS+zBaq^AafnKrk8)=H_hs9Yv)Lb>u2=HN&G0~Li!16rnjAt+z=fDjrxC=4
zNnufT*Wo7jU>Tk1$K{JE)vkV>{-G^Z(!F{+LK5QglRV*Ye_T}*1h&)NK&IF({uPfC
zC(-ueRxEg8D4z6T27NCo3Tib$haF;``>JlCKel1Xen$|yDI`dj$K^t*f#I<tAM2Tu
zoK@0yh~=JzVL1LbX6#%@pt8A6iXtSXc{=Gr18P9HCmvjdpFYL-w0$4s;rU|IT3C=1
z;+8~CM{&726GpcJ!&&mHuf5(;)KH0x`($ovc<(mR1JXOpKE2qa;p#eCxQKl7!;j19
zl<}eNQm*58`1^b`R6Q;2yT<L!AY!%1{?=OaL1IuJD09&Q5Hs32*zwr8?mz&ND1V+8
zhLVyJBoF`}B1V;I^3(r$&&cYi4G~cSh$@@jZZqf=f3MD3(N6m+fQ`AfrY{%2N^<w_
z@Xv|o!-;=`*a#A}wN|?0?ws$h{h>33X)>F}lwo2w$nM74u1vH-NsS@|ItRyRo@bk=
zCDk5(H`n@7rd*szXZptu=Z)F_J?+)UD5%#P3oX=Y-2ZC4JsHR$OX-_+`27HWedDK7
zA;t5f&9N4f5&+{*Nw^$l1OK_km7k#U|1kSf*tNNhi-CpiJ8Clnpe;LVKNe0E0ny^l
z=HdhVhn(LF5rmnI19vUTF@2!k70y6%zfV<FXo#e=l+QUMDuO2kcg0U2+%;-fUQb~~
zZ1)|30@><oi=Xv!>s&F}c&ncjwMV}ld|;<>XB!@rc8$HDT&x6>|5(&{-`j0KzixVw
z?lD$!VWQ`D!;$aI858J}#y=MlxVk#VH4?K7izJ!<sG_XA=KOa=w#S;c-PTthMXC7e
z5aq*=C*A~}!sO4qwlS631eQJORNIFbPhiTxLKfo-Gm_W&1B1h3mBuKE&2ZIENhfb1
zn&WQ$6?XszknrQ@sR0x1l*h>fV^=u%X`{<+R$`u9tGHl1E7q_~O--Bii)vS0%3TUk
z>?1z5n*8bE;fIb+j$WK5G?nu$L{WWq_i%1Y3~O14%;nC_8!luw08K3W7Sme49`<MN
z7nPEN`V+oquiBO&IB;a*jw;3KU%7uQ|Bm2%*OGnT=VvE8kBtEJ<KeBwO?gJa^BCk<
z>{_tDgA9B6>OjE(ozWtrdSfS6;a|ujAMNYD4{~WtqGI(kKfHgvn^cE^w}}bb1}~o=
z4e3uFNztr911Lgv4vUF}b+~^2JfEyo6gRX!SK@#hDAjm?(DT-!mT2%{mJ{q(KvR<d
zeyOhRCYA|oPg$4gcMVH1vCN`~YeeTvb5K+yhA6ju8)HPHprA92dq<+wlE7vdoRkmU
zgxxbjvmE6%T%^Hf)P|0>T_J___8FQzd_Ma7=f*~kic&d!RjP=ue0|9M?&RNFs=IZs
zk+wH)6q3O?V+Y1Lz+oYExZkA|m|(#62caFh@iy+6;y;b<_uYTOY^EeCaPIe1-x|r+
zaz1KloVqv$0MM0hMk~c?*I19wk17-Tco6YgTV$o(9c{6D2^qe?G0U2`NSRf>soOqP
zmVixWJz>J4V!N!ooZsun0N!}TGh9VYHGJO|v{_Qpa%5i&p%mtZu5gfrg|%OT<j2k~
z#$5dFUBE2sq+C~)=N*?@9)T70l_?2vtZ!&<UB-wb<g-Re9Fa~`Ea=+jhDB$-y>8r=
z-24xJH8-&Q;jn=Q+V<&boPUYyg(okdH5w_<`f0?xaA)R!;BXU@71lUeNE|c{fkI(0
zS+~NSFli?GQ{)ll;%V#$8y7M@(xtUJgN`Ptg*J9h4y>V-o@B=1)vxUxscL$4?sqYw
zQ{MAQ5LmA;`An;iFkJcG4S-05;z#^!ZPQLSV%x;LVd4YnEr-&{+MXvg9Mg@xFnndU
ziWouSm>ey2y$s>D0>hfj?{<?At42m1YQBhw7;(=VIR^46tadH7ZJp~0GTx*4`P~Fe
zacu4xlL4LP6e3=$L-<~2n|>9e?I}J9yrQaNs;c9_SgilyZF>C-;*3an-4_cJUA)T7
zU^LfJEeMyA_=>ss**t!v*WQR&EwQ@QAoM(;bKRUanzOrcdn|Pt7<D$f70?elbE>Vf
z_Po%$g_PT9%}zJF-#e$?SEWacj8<UmMG9YWVMeK}w6y)uk=N6203OTrFeH-}z-Ie=
zGM;J9Y&GQ6;VbBOq{KAMvL`-*%#0ol3jwI#i^StcU7g*hENyVC%~yZA_7mcmj9Mf8
z(j}^*qkV*0?q3~|{G-p~o>_Olbt?g0>J$Dw*tKn@afGx3yc?J7SNf6*@-uIf5ch3v
zld`~+sc&>_kQp_TnX&fTf^@37HnR*zN>Oab=RMzz$t<cv7Sxbyx3^M@_oW%bxkkE_
zp%hXUsVyq%7<=K<DZ2CVd#s)k;bgT#N$Ds2JH^k&v{HDSefw5yrk#Fwln~6kOqRLd
zqfDH}V-y%17TpR9>`7)uNXXrhi{ei`CiUJ?l8>dvCbd@X;d|4dks%qzMk|Z=Tz_zr
zjn~7#*v@t#K}LbS>UVT|Cl{Y<u7m0^#Pma1g)(tsg4h<_dFq(<G49T!9q6+@m12Wf
z*Ni5ohxf=!Z{Xbn#OMx25(Hwi5Z-=664^TZiwj1k0aMInb`b~LT3MO0Q)wAdVuDN;
zo$a%!O~NRa2i?r*ydY8pygBc4&6~n{TUR~=!0xTY5sxr?0x!q8Lq%8&6C-L$S_J}*
zl5ma#g@pj0n``r$GD_-hjRdp^P<;GDj-#)FfG3mubRxUWRP$7;!u<cy^p;UoZr|JR
zjVK%t5Jf;h8YHEpTT+lt>6Q)!B%~CP5|9?@E&-8{4w3GbmhSF;r@#OEoZ%S959jQ?
z?^tWidCe>0KHPjYa8jJwOWk+(em^*cearR-GIyeo$!l3{G_vF9_{SMCTZZN#702{<
zG>uaD`{xu$Md<0y(c@WVWH=7u^n-p)SP$7=`xnjBOqF`!<M@mF4+qZIAQUW=cacva
z2mOaW;qUZqbj6HcGgj&Un%zNCC|F%CcF2@9eAQZElD0MNZEE!Lhu7gAIiWl?%Pc7Z
zsvoL-6)@^v`&tE@6lLnEz9W0sSlZB9f44!Z;DbX!{$Mo!I+zJ&J^E;_NG4`%;ciZ+
z!u3_;Ek8jp7tFApaP_d#%gOgXeA0n>xcQ#%3FtGphWtBJ7l5)vO6ru#K7^cHN^J)!
zi|g6$No=!ju6$@8P*E{<&B$k>+?Sp$gEuD+L*_lr89f&@0M8T#mI#~m*~0qIZEylq
zfgz-DSnsv-A*PJX-U7xnbXMNO=#R<5?zlUYr29@QWDXDWKz@OD&Nn+7D-H6zk6B?f
zaWyu(H()sefdGs`i9>LZSN8W%8v3UR#_~Q9uRbDPvNz=9<kqWr9nxq>r8CZ+E}cep
z`woVMrLDDn_6-ClcoUjRtgKI;IMYgxC7Xc{{=Jk<r^X?4XD&n(S6qto0rOHMj@&aj
zG@sna)>I<)k<R^N5|<g;Y|2bjL7Y3t_}4Y`H};orZs~K!+uB;jXezzXQ>}KLZFQW2
zi<oZ>wX?H)oPR<=4fDOU_!ckdaXyFl0Q(&XjX>R=so6(_#JGW+VBh2P5Xaoc)X>uI
zYhrwIL=3PZEF^aJ%@s!nhQN3Ms7EkhSxp%PjTw*qi(0$kfu)#{h=JF%l9EPTR#xkv
z+6F!joJkT?uRv!0>w83|89ww6fSd(Pxw%ytoNb~-%Zita&HQ=njS2kUwuyBAd`B5e
zzO-H*jBTXhPa87!k%VLo#u9rheAstSV_`V#haxH=ZJ;3~rJzs_6T$AK4I$Tr&iPs_
zjGVmO=!gil&)F2ez6dtzXv<<EEn%dd{oC#;uZi>2>nUWO<<K+1LvwJe@$^?@d(gG>
z*C+3*HkGnRK;GWtYxJC+(^J0{X<d)G>IJn@?o_SZ918D^zjn%%P4ORqe!$t++5Q+o
z@JxK+mCaaiQojAQ_uXWJBFI=paZA{qRC-*|-~{70&95dVwlM>li|7UO$Yj^=p+Db&
zhyGaTw(aRBtOHPiAq`3fgpY~ZSjg^PgYXIhsbN+CBD<fZ<+ZHLBWmQ@D%zn<#tiqh
zBtopj)q<XPXitoTvIg{8uIz#cZs2dlr9|F$6ri3Y9y`P#fX9C0o#9~zEf8!}AGyzf
zgIHj5zV#Y#gMfQ)8QC)HP*M3LTT-IgOI^yhbrz(!;RM-V9_W`61F*+T@LW-I0$$Ow
zt#SElcz|kd9y8F*w~Nm_82<z>owB>bi*It?KR;0m>Gr)|U9ccLqly^sNAc{8sY_CL
zqaaN6383zw11uVfE4p2P4Woy-0K0w`jhY(V_~LRsCUCwL_!*j`L(?!SSYiw7ohiF+
z10D~kCgGGG80_cf7}{Pl#`=6Zk)eix5eRHF1b2tNBCm>@nu!F_p2!{MDnop(q*qpr
zoSoT)6}MAde*8)39n?7!j6S*z=b-m4D;#8`PG|XFDk|E{pFMyn_2`KWInT=|P&_E|
zSTEO6=Qtnq6gH&b8eM^mT1ZIfndsHds~tF#74Hxd5ydUfhF_tCh4<+}>JFM;pMVH9
zmBZl~DcI$u7evQiU-d_tKgm)U8XdfwR5M~fT?rhuTR{(lyYK4vCDxwkyirIh8Dx{;
zIPJOge-|cRw{U$UwK$h(b^HD?g9oI%wz3wT;6Avo@?wA_d{$C|Q@a!~1avc9x>8_#
z3~6Y85}V{NePlL#|H1wD_iZJ@qaw;Zx?2F=YgeUTjng{5@m|S3q1v@9KRCCrFeWY<
z?;)|U%LwOVZ0C2PO8UyDomJg<HMJne2YT)HSm$n_4CEGOXD{y;i*Loi$O{vnC(n8f
zSVen#w^2X`@Nza{m_;Epr28?nzl!A3QYND#BQOK9c~iuIJ$GfhxhY0@hPcKil)v6w
z0{VEn>Ozv>+biY`T5YA<!Ll(|KaA#hS=I3uKm5)B`xz&ufULmpu~Y`l9i6Iul7s9s
z`#>Fl$scS1tNWUtm;MWD#nVuKpiPn}UO077lv_|zT-v9k37mkd%ND&-b6YDb{qhcT
z5b{-xYXcyeKVsjGA6RQ_ZL6iG!Y|rjq6{oDaPxl~viCVUJTO=h`k1@gpHv4qU`P*B
zX9N^~sJ>!-Uv41&O?p*UO_N%xkl79-xz9CmjiyIW$OYVw6Mt@@DdmzrYC!`8*;6@i
z=N4IV@|P$M+r1CpQ7-qW`YcG-xwjIyJn#8;PJys8C%jOs<T{<M7t&z(u}fk<Q(Sq?
zZvl$5;>B<v`Jr#9JTHrzjsE7U&t&EC_qtS|QOkPfXj^JELzmn(tZc1wzQ;vE=l&MF
zzW})`ny5g<(5<^EoLv91sT?d)EUd89)s@|e_XOckqzZ$mREJ7ks04U00;8k3rcDHc
zpLbRY+8HW6Gg5vfjV{DSpHW_i64rDJUeHxNpyjPN@gssbIGx?zy@qUX`NAGKUIuF5
zRsR%V1$Xn&j|J7kA9EClMbHX@VRizmGHstY74}W3wVXl}N_C1;dsWr_0d*4GF;1SJ
z|FvgTW!W`<l?OOPU0nmZF>}^Sg5=MfJ_d9wvDI$p!rSagV9>IuHXe>=ZqVF<ksK2h
zgO28R!4TdZlXy<p6_=Yq-@D|{?=L7U!bUIQY6mbSX#AeKk3#XFS7}!Q<4Y}~I52Ct
zMi2AW_3^o$U(rMhV38yyj*n2~0A)GQBO)KUqWTnu5>$!CP~7uVL9~cg8GVk5uyG@v
z%gV@x9ypvXC>tuZ*^h32AG3k+(=8r}GO@F;h=>Ramd`Xaz?rBYr#kon7zE4X-x$Vy
z*)u^JRz#KzG!#n88aEKEB_<AHQ**k5%(K8jmeOLf)uoF-pO*-hx@uV|TW5QBZ#&cH
zn25Z5`@YfPr>{fGxfziU(h|1|N|6*<3yr{Lqydggk>Xq5%Q=&I-nuI}JlOO381*NX
zRqKRWmz#<!#xb@q=n++u;?TZWD_+23>h4JpIO0b~fzmdy@-I5FChpIy$LU>~Ef78a
zPQ!7vE)gR_fcmwVxnUF~n(y82o%}xs>sA02f~sa;v#{XGdA-=Ty1(U;TDVhcXEBBF
za$1iPm3}rUNbV7(nHOWoN1QwNB&Z@}aoq!Vn7EdAU4ck~e#;-@<_!iJnGE_MSt)A3
z#(fpI=;tUou%+HAEDE1pNO=o+rprjme&ZMzn6H&Cdhr9ihz@$v>z4a)WVN+ro#J4+
zrCaF~?%j|saEv<ZMIj?*Jo!g?UE{HM+1UWLfq}G)xA5;i>0xNmu2l&}bXj`)d$_3>
zVUVRAbRxjY!XV&qpr-DOO%^7tdYHd;-`)L%1?xMkZxRBZrPf8ew}0;C3}bq3S@zLW
zlP*BL1RQD$QYR{9U8l)@!*JSm3lr)NxxTK5H4_&Q5^2i@$!1<wlqt%{JbU80G9gx4
zuKM)%+?}TuV?s#XQ17@lF}8d+P(ncSB8LS9i<YWRc7B0hqoMZ16`QZpj8|DLpoQ2W
zA$)wNaxZxTX<(DpP*u$?$P8>Y29DO#-)qZ}9n`Y%f<_y3V(7n}|0${85xdu)X@B`1
z6QhAkiRCyr9Z&^bQw~$c6@95yx=MeLI86|BN8{{nvW|0LF!izD153-?m=W%mE_aCU
zk|kfaXWGCWm<#b9F$(y88sPMHRignNaT=x9ol*8H_MSd`hiehww~;7E1qdub@UTKA
zW1Ff{5ByMX#-oddO?8cp!?2H!I8^4OKD?Sg^6clS_shF}bAf$jfhpOe`Clnc)wOSH
za=osqPS_1p6j7K%qzazsa1atVpo#(q*sJMqcw{(OHdum^li^wl+@-s|@E2!gu?SKT
zYX$swiwYD9bdh)n2$=h+AnwRm!ea%}-(aTgq4GhLV&$iXhp+*C`DOU8|HlQ0Mo0RY
zc#)NjjlJ^?*|^?iZ<V=*lZ-vFg{68zp+7Ts14~<%9Q12E4l6Z`oaLut&&qQuN>@sd
z1b8qHtk4mOVOf~4dnT!ysdvZj5gvZm(N%dQ82dGjIE-9?pY__aYu$B$z3yAh4FqHp
z-V8l=34$fFJ{qx7)hnwWcq;G0yrOqdMD1uWNHMG3GhD;5axx(2Pe9I#{^RNgpHS$J
z_K)v=)zKE2W%v)48WdLEKc1640Io<HSuC!Ez&{W%k{Bp^3FjsibzX74xDa8_dm!ns
z60A4?+htYtCquS4<^ESQytwCK2WYsJo7dsSz>lh@4xQ_!42Zx!ki<Q&2^bkE7s3A)
z3n<0*xscE|Yo1F#?->2n;d}RvGIpeuWr&bT?*|{QBq%2&3;^aMWRy|PExiz~)12x1
z!@79z{K0<)RPk$HwT&ug{;bha_MV4KM#Uuxd^cQoWP2c|EQJE`-ioD`pf$B5A~y?#
zbNbZ4@XGguq<{O3w7R60u5QhbyhpS`j={ZOsp)#sZfR@k+_v3zf5H4t)Tc*H=$n2k
zJu2ac`*RtYv?62CBSRz5cgXANPF}a*gFFz%1Q<N`p0)j2TYYM_>skXZElA%Sns^Y2
zH__zYsh;M9zJVN2;{)CKSJ8*iQsGtB((TyD$V=5JVXv+np#KA<$f=#rD1lH`+MDIt
zt+8h8iLm|cHF3T%#ZKTgL8i18ul-9YkP$vLg7oMq^e_mx4Nv;z^0&EDlhzC$I@k5}
zRf;C_+43YgD$r-e_9lvQC)MKP^-JDMDwPbIH04+YwHC-$U`Pg_BM4(~NCj-IjA?4-
z&5B<<llDF2<2#I<)ikon33Iw)%D1+gtRsi0%cy<|MWePCIzzmkDT&_RVM?^&CKL#e
z!`qs_D3#sZr?XCqins!XQJ+iu9R9j+zu@JUkY`>2!f!tuv|r=tH-QRSk!(7+>0hzz
zCw{@p3!IC8-WNMpi9k>`@5M+<zTos8`D6*6wk4RD!3I(-H*fjp=F#7-9X`;(BwkEP
z(sz7jO3nPxeTdKd4Vn>~p}jRwj=w%2iICvD)uVe+07V$!mP@@!N#W7eFy}9NNqfJ8
zVjR!*R_=CCPPRWSL!f`yMktL+q@r{slyYtDC2}v1SLe}z6$!}H?Xs$X+O7{FFFw<s
zQvSG$joEm{J#sd6lH<$FuP$>B`Tl13$7-ZV=(z!JhHq$$hN+XyuWD0Ja>FP~f)PZ>
zHxQZ}u>-`@4yD~Ru*fVYJUkkF@@#A^rNSe@B0eX#K)eT6=_3r1#BQF-EWTyfd~y}I
zVh%_+z$?EbHby69WuM^^VZFSQADgG6J+Z@^5EYg2nHvngX2u%{MzDub+b!5U*D^0}
z#N7$T``zacrT}f<ThpxO$-6b`Mzfi+j=lS7BM<~++ALxIt+TLb>z-H}&7@HhaLAB_
zb<Y0x7FF`Y=g9Ms^$>e&_3PWXr41mz^RxRjOY-qV`meNP)mc$`S;3U3qRZ6U;?F{T
z-dC=yiPLV<3nx4Q*Jh>V<#F-E+v7`)Q`?i}-Rl6~z@AR@*OSCLcr=_cc&QgikP_4A
zS0cV;nQrRNk2wO*$Xh{(h<aJ1O#hm$SC`m{fdkATicqFtR8E>khsO*~#krwamC{vw
z@x&xjH9^fr#lWBE;`#{tc7-lWl`%-c;phY5a+^uNXoF#jNM`iMmX{!Ue+K{EKSv7f
zj#&q!;dh(T`)3)>ec@X2Dzaw>l_J+aygC~eu^ig7veRLP2x{}Wd(*FY6Z#16&J0Ml
zahYgKar-yTzOEG0va&Bh%GP@~k0z(20)m5~LB4m*EhC4;%6TOw+&^6-4<oso?_Yq!
z8Te-lPpcUrXZteANpYS>bHgoJFYoXyAA+IiwHqbE86?r=ak;b^I7oOF{qrce;=}Kh
zzkNQfje=yXl<XL1YWY{>Z_SDNq(%;R@#rwN;2#{nm@2!esoJTt@(K9&)Gcl86s1z$
zC<6Metq9S3_abV@Sr&{eZ`)M&cI+=5GqA<Q3Apr3b_pFPT%#awb*zy9&*Wz`_W~ru
zUPvq?o`)&Rtsp|zd(MBr$TcIAMjHKOd%V2AyE(jbKGne`syOtQ9*lN4gw0S9*>5M|
z@9$0N+PS#6SX=8Ds}Lh85?J?h;?HKOlY8%A0G;ujg8-7En>|uqWa^V<^FC%|reKGf
zeH5Z%@1xu$jRg`jxY7aWUtcuKRp$H;s-mNQtf#N+mw_mq)}s^<kC{t_d1htY*>yrh
zX>dU$oSB(<CPk2P?=4vTsA1lsML|I@7kkbU-U27*{d+37n_$SFy7q46?#JoI58X=#
zw{w1$l;1=S{zB+&&X?o&I#{*6J>;J60n>|$GW{KIo*hQoa73hTWwnMhh$^CowMsYt
zMnU`sg?3T{B;J3lB1ULvvQb25ew#DMU>uxKpWE14OPlmJ9LZ$6P=w2ZmJGaf&OQ%z
zqafJM>Vjl5!^3_fU2}NXjw~kB;y(hOsXE?Px!=*vl#_3ddzUiq2+aZ!-HHGC_J~}n
z^1o}^K2$P15=Dx?ffB!3yEY(ydD*L&|D5&)-Ws85Zb7C(w#)BkIzEAM0f9fwv@}n&
zm}!WDtJo>|@SR_p9&BUH>sVM?eIMIsTj3?7drOBt0Wmt|Z;1I*Y7@DT-!+P&^X4PV
z3GlJz$7nUjFUASy&3khI_-Emx-E4wk1iE~ebeOk<DKk=9Ue>Es@p|YtQJM20r09Bf
z=K{6ElHy81IO00<h0&cA_S~Frg}DXPxI0-bQeTO!e*EBgsNf>IhE5gk(`*cREC)O7
zw;TWX(Rn!jJ!3?@iN(kS+<*Wm2~spE!D8v}TC#X!OUDfaGbl*;rrqSZ1WQ<UMvl+D
zC;f>HI0PNzcLX6v3G0!g622fxTVDG7Tz{{IjmKQU+dJovZeEyOoVs&flk}~UKG(24
z0Na(=NLyQLMLC7c{uL3g7nI&8K(5)3`L?(F8d&5t*DG2ovczYX=LEEwi_(e-U_P}g
z!|<M~@S0Y?4(Bc3hniIH8<0!`1nKs6Cfq3GK9xk92mr`T4{?r#Dcx4{LU;xBwUmuo
zM-D-soEvVSv@pBGOhq*}x8ZO)4uf*%(!VvYSKkYrhuMr3_l^$;3eXWEOWj|&ElxM>
z<r}dMon>fLo?EZaMU#M{<{NW&n#nw>fge2+lUrpqvlj32!q{^imn|?i>OP5rx;OmH
zRpSj#-ivn&3gCOV%%Xh@nfUpQ#BfYYTia^y1N$+V@Ko_K^3F$^XwR4b#-v^X09YVM
zGJpI+fdaQLFVhg^LdUUl`Dr7oZxb*oYNzV1Plngoum~LI2P}~k75Zti!e=*sk4~Vr
zvm*^`_F<YuO~T7dy;`Lzk}?dG43dxmO(Aws;Am9*C$^EYX!=7X?-(p^x@@nM&+YnT
z*X;~%&*p~VJ)Jq)dXT2x701WLB}C0Ec+WCKxorVFP1EG}Un(lf%1h^lPU&uCciA|q
z&V#8&qt7dPPU6kxPM&SN2+$IL)z%g^HC<b^AX!;O8iHtVmv8(z?|heT|BGr5GYItU
zbRp!v(|^5xOP4^PTcf!$qcd2IB7c|H4>@?9X6#!ba5pnM9oy3?4b>+&(z9*C1!ld^
zdryA9bez<CXfC%I`119u(y$;m@7a#-lZ^Xs2;Y7QTZS|z{Dnox**M;@{9DtUuN+6o
zOaNkNF)Xxw|Myxqzr+)vHo(<Iua`qJUCle*N=u_x>!t`EY6#&+|LXnbd>+8^Ub5=G
zYntt!W81DK_W91qgY9;{)!aD>vl}zBXqfMk3%?}e_1IgMw|+)r?9eu0{;rWl;phR@
zBQ7D3_1+;LF-ZpS8wh>C;(_;7LoLeQ*0$PXi`P#sN${18gqjUp4@i%(9+gyN$U#g7
zZpjX*B+dEeu#3=S$FBoDr%8Kh1#qWOYBObL?}waf)n%3<tzy^t-nU94_&@T^thr~u
z^P1D0H}9Lv6N{Yv$=!zme`t{B%%49Jsh{ECs|p0$!kL5pEwC>AASnTC_wRfOyCyf3
z7G1A+B1WOmf{ht2RpZ;#uf^A3_yzuw^K+3&b*@~N7b_58A#xjQx@MD5TwF-a2#=QN
zZr}&A;AK(~Pou@N0?<)D<24yebITA`k96F>{cEO~o`L=ZOM_yg8sov%IX|c1y5m+2
zZT{k-DPH6t#65qGxoCj-!dsnu5PIM09LdXL{P}YbXM3g~hayaE7#hMUy@$X`zE?dw
zGKBbG%uD%6N`QMd#FW7A{*`tQ%pUrwFVThXMm)E+v~_lI&QfYSm<dJk35|J#oSeGW
zy6<!I6d7*Kse^n3Xj-s!HC@)AmzVCQqXM@sLUoXk#ej~MA^b~1R$&$gIBdU*XZJoj
z7%KhL+WS68R@T~TZQQc-e|P9x^z&u=>AK_COMe6Z7isK(Sh2&WzL&W9batF|zZ;U$
z;BJM}{|Ji}z$DhIHr@?2@R}$)PVudkcav0?G#egjpTDI(ok+T&kLC{_#QTH^iV0r6
z@$R@?RY|kM!di65rYgZhK(o@XxDb@N-+jJ-b|fw#@z3uDo}{qX*{`_Un(kgy^#V%h
zq6!)>TNLrOh1|LAcdbx86Ne7M9$=wiN9@%0)&>1YtNLmM-#OWwzp!W)AFs2mFCyY#
zybWFl;04??(>yfF6Y=af{6E0M6On&-JT-ba^|!RNbg3)uNmLP9o%`Sbzgp|WHR2&$
zU;2$ofUc+_Lk(A#ZovR0td)t8rTZ#DM_Yq^CGWrBVF*kH2}Zw5pj_LG7|h0iytICu
z{Yt9M;8Q?A+o$A5S5pNi!SwXZ1kC&66WVc}=VsOc&jNxcCbVXET|ry)__3vl<-42Z
z;6#{1?>zLM2v~7~g9=&L`r3w%L^rq$ZrT={3hGNsXD1c*4EQ;=d^#XMFbnQne8u|^
zbjt7g1m1%(X?&t@g_MeBU}IHvsXOUolM#I0yQI*sc|qU?0VuQ!Pi&*9_Sg7AgWCHA
z`Wz3sVn#G>x$(gP4ntv5TY309{9$_jgfaS;aWU9a_q#L4wwjcT!NEsA&^j`%qNu2&
zsbf7J6<DZ%^R8(6nQVX}vB*oG7`*>nT?nPoB4sFWrNeee@c&BF;L3%BC`lJrRsH@g
zQNY~YH=gSQ{T+^fIt6QXdV!^BPEl%MkTz%p(!W89XjqdGcyrQyB4J<}q_zZKhT7Um
z$TnUn#UV>x-dZ~7?0Nj2|2Gzz>QnuZk?}MGGh?_#=x<4|M8VsTAlUiG&#8(xl_^4c
zdU~x229Oh^|I?}+cP0?l8Y!vGyWUt(x_mF0F17P3k9ir^_$=Vuy(+wYq5S9#V>E}^
z9MI-sDVO!6&B4dA2FU4E2|YauX0(!UELt+iI6lDxl^aX!ghIM9LzFU#SAGnu#6Xx;
z9%E1Fn+FJSt@v|Q&siXeKMIxH%;*>393F<rbNCO`#Oc(`+$}@v%S_Cd2IH2Tkkt!!
z8vejD6EWz8FNJmu)h9EKA*?lBsf}!l1)}gaS2j4gIdH$(TU+L(i)Dhg^6Ar_=@H^e
z$Ue~ob3^eEC<_0(36aup{qVey?<K`B!FUN=7w(Ckme)<=i)}S-9%R0a@G-#Ol#*&Q
zE(XmmT46df6SYZB(P;C_NXcg;BP{W}UYx#-_bJ1fyX5-j%5#fTGyRAVa8x|?NQw&#
zugqnCCkn<H#%x44Z@lWFzec<bsKPC*j6d8Et&=2-m#4YcB^T0t_Q8_kz^N=eE4MUi
zA!#$dMo@QR!cvW~xp72WU2<%A?5=MkR9Pknlmg&l{*f{2ACc^p%PSZXgNjQ|ZKzMQ
zx2NTvf0Hu9eFJ+JWk#S)vaqml8*StacM?)&l9_S?M-`X_;)*U-+5KQ8b*|x+(Z2$A
zegIN|t-KsTuqefsxCeGx^&MS*0tTlbqRI|ndu1(JO3FeCR(~6D<)`{j^%XRA{tcSI
zxk~2gyYAVx_BNY|nivTxE6cnIZJmgAQ+0e=LOOITv=vfP;x#Mqwfn+v`S2_5fB^r{
zuo^7EK8ddi+kHbyB|l3D@bF>3wae}gkVva9Iu6Rb4I4V*3+BDtL!)KoKPZ9&5HxrZ
z%5AkfO^dM(oIvHu{@_z_YF%cW<-g*>*V5SLyVck)ujcVMxvbKt4Dc1M8Yl1LhdC?{
z$B@hSlBZ+o7IL`%o{oG!Ma!0PWIGjHGH5y_ZZx)b>S;)Qlu9>Yd%${U6c`$i{e`_K
z0n5_#?mdVa&r~^Za)y^MP$G>!dc57RqLLTl-Qf}+66EKma(9hJKBM?7DdABo2y)=@
zam%!{A;>`brdMg&r>4QK%vhV=)ZX6%9T3<C21q!oM;i$z(cB9ZZ8&$W0)O!V3sR81
zHFclHCh9D>-|__&>ie36f@r_V?N-IRcL&()?=V>A9=##P3En4PD%)SizURv&4zz}e
z!RTi+0~SvBW|>eR^?nt;g{}K<9Vxu&6Uch}mB$)R75K2G*}o1ibgXa7U|@h4)WkBC
zg0<DO7$WI8hlZ1Gq5V50w<h|#H(QL^>!!ClW4okt8-e<(a8%yb19^hLHR7;y@=2j3
zUdubH;=f7tZ_jS)guOSPo~d|j11#WR;c%&&!BP~`vpqmHwR=hRm{ulz5H_L2hZ`^e
zPww6Rw+=Z$xp)f(uHxXhi|t$nwVuL5p0lfTbBP;wqY?54-0{9(_}cyHmR}|=xFW-s
zeJf;vxwnZ~C3qM&vi+Hw3TZa}{B5sq6ZA4b@4&;j`H+n65o1t`I%tbci)ns0!?+;)
zb~I$x$jH9X`kJ3d5Bc=44qmbEf`}HhknRgR$vlORBE30-*=RR?DP%i)TT2XV&x?4b
zJ3E;;I9~T{qckK)GSycr_~TNFefaP0-cM=<hTXH%OLCE(`uCdlzitvcKniE_)pZNs
zq4}=hinF?NcNB&Ruw;_ZZ;=Z)H+DDg{;g}2WV))&@IEPZpDzb@n$U;BC&m^3`^3}r
zIAIq{aI08!Y(*=iJDhoRuvwiJ@ng2B268a~v@PG9wzsixSiIP+2V!GjLZyRx3fkL+
zhOIhdo|(DcUT26roT+pcUzuq03JSY6h07Ko`c$X<?Y~>MPJ^?<Ouc)iDv$rBplemX
zF7}Uq-n%R>(1EUcN<bwM@b4E~m@qtM@+Ldr=Vxi`_{GWiC`q*8SUvV4ndZ_YE8*Lx
z<gn$LssFop4qQ)sk86T8Vq$VZFS5O(#1daUU5|T^`<44m?CSoN2RU}&v#8iee6_ph
z>qG)pzt6Zj5ySQlY&UVG!iew~O!wF3mobT81boTO&QApG!0;hs0=Ar-uCel~G4Lw^
z1L7;{l*j+~!~Q!Z{PAY-9<abx{+zV5EFs=A$|y}Y<A|^GPtM(Mot$5!Ms)pqCR7VW
zzk6C##PtsC6RfG1x=v)CiwiAmR)ael^zV1Qrs?Zd+LC{Tomqg^9U6Wh=<M&^KF25r
zslC?M_>o~Ais4r`qh*WPo-`W0N_N~2U#WfnzUD*43r|lIfsl`N>6y7tDgJ^=UJ2h6
zUTRuNKd@FqSvxsuo)|1EE|$?(80l<M|C_F?q6Yq7kC}>S?gl=ycs#F4;M=ICtZ}fv
z3xFxe#cek^ar{-aHBb;<W({fS+sHH>&6D~ZR%#|{Thkn9^bZ-Bq?OKC=2bOXvYA~p
zvsnk-0VMLP45W|Nexar^k?sEe|DFWZLNW%9|LvV)?49GwYLdOt)YL2V45oiKL!!+9
zn1#=;w;mw}d9|=EuBWh1XBN&qId9&$>MN=y-S}q7SyfT-!pgnn>~ZzYxgA*EE=9?)
z;W21Fw>7~@Z|5c=b#iK>{qy}Znms6zlf5@bE@6H-uB~G;QF`Row+`X!s6L-chc@%g
zfdTLJwd`4xtkBs6*%Q@yfS_Qe_eoKUq<lk+_Oi||<3mNXpO%tC&bzwM*y>KdPM*H&
z)(|of*DO3)E|LR>MHyuj1hOUoA1`^?d=F3u<j?n)e2AKwhGKll1B6};GcOxKpb09}
zzApKD|MpVP7Xe35WB?BUJcZZGHy&23d`+#}>Ywqti-(svC_qE=+kVOu!p5iSPs?w^
zE&1Pz{_o!W`^%M)M}1felw6=C*cpxobe(j%exkX3y5kqK&^+FFUUs^W^(-c)*22=_
z6Y##laK{&;!1bCgg+oCxt}G{b*m?FK96BM00tc$8LA_OOtp9rZ6T(Wtq@thb1Yo%0
zb{lz4&;v9kVTXl<_0JAG=B>W<TM6vW10e!;&@6yAN{!=xzQbOUp6+<`i2+>N+uo^Y
z>!TpSgk4u(;ym{@XbJN3)AERsqYaUoH_Z{@)E{kutlZQp_tdr~G_>dLd=?`Eqi(%x
z-uAA%{HK37bG$)?NB2ulz0KR%VH?xsmP;3AsGOGDg$4H+&CVuihq;_dMhn02yww&{
zu~N{#n(<us``;b;Tpt4Ki-milxnEzYbP=85aXvl$THr)<ba;93kCSTFT^F#`@H~m4
zWuSc5as1z@BN+M7W@e@L@-mN%PKbZcEh~FWTT2(0_>98Ut>S1RcDg_Oa}=W(m8bR7
zFYTtyEp+#GT<sNk9&qHy+~*jB{SdvVuqKVMC`Adt5c=V<tKa)Le!mwui4#MG0e1z<
zXwO4ytua0>=R+y9vz*Tz27NRF*3VZdhaOk~KBFKL84+<-P1bdc(VmF)(KoRy<HK*w
zdeQVj=>6JE?}Az}_2Kutsv}1`#al!euh`YzFMEINt#kv+12lJj0Q$VUiO_4GUQYlJ
z#{NZ4T7U!tW0zcnIJp<t+Qf)Q&7?9VBjwB-ESbP|%SyL(@ZyKkD}3B;<}6f(!4i16
z2$)Jig7&w9g4}O8pgu}7p5fu*LKSs2?V39++1UKHE#v{owIaV}S~~^=H=A1FkO$P2
zW`h)|;1{@oXi1oMJ@ZBkOZXH*%TdoQ;NvTG?PSOOV`4^tJK$PXO+k*$nBQeu4BLON
zu`@;;Z~&AIaeNz7`d8jx@1J@QhlE-Q9W^)ilObX=b<jzS@wus*v%V}rXY4G<%q>j&
z1{TxSCT2z^X8rAkh)=q2NS`$I|1AJ#i0W)xseorb)tp;opIKbYAk2`Nni?pUdE#CP
zY~|yrb%0iYg?gFrh^|K$=0ue)^S_c8(`Doogj`nn&o&ke&lUIkLD#+|<S^9rdxiB2
zCb+gO5)!2L!Z=q}<rmPoc}fZA!s5ojGx%S=(QDN+e?@gJE)(gH6YIBj&xS@uw}l+;
zdp^4NpQk71Vzl~u`MEh`gEI@z3>}m4GP2U0KPH!d`}TYqRa!E4NMs@XsXREcT)i@o
zft?E^0siN1$NT+#76;3$24M%sn>v&y=RM~P>yFOyk8%9>_jf^<2yYP(mY{J<iS;-=
zvwfEZe$MOli|KB=d-5+Nh9F7&>IONs)j-^QJ$AG7;N;|w?Z2<2-^SN>N;_WdVXLJt
z>#+EI{FqgX9t!IXt6zGbf$6+qHgcsTJoW1OOV}e`nhP_Bm5$8YR~+QsmCr|+81oAg
z1g=>3`agV>4hjrPlM5%WyReQ{=h9?NxJVZ!hTRk*b74l|#cyZ9`Nr|p?`wyr>({nk
z^r1_mLLw$eWTa+F2K|gZQ|v_K!G}p68mhba#pY~UE7#8!B;}$o5x}&hPR^n1)X-2T
zAzEps%i`qviTHfUp+~UDmqMcKf`XopPpqPH%E-xqfFfcToIW-&4_M|lXWuNIks^Qh
zT@hTo*gvCBeH)CgeVDm9l%$Hp<elX=8)gUz@lX*NnZSOOs}&K5V0izD?SEwt)xwMz
zo$o8~mjd(xX;{d5*YGnUY1gvzs8&v+cl#W2tsw)O&F$BV19NNZiDK1dbA|W=_3ZX-
zzYVVdhR7EAha&voiZ|2fZMWGgvi%yJijrY*Zg>XZmyrQpF9-XXxQm|^&c{XoS7>^2
zYT8d{x;R}t@#KWMFFchM1}2Xb;z#Wxq1QHD39hR8xat_|;cNgLrE?xu?0)~lpMd3A
znwv;ws+?VlK$LLDp#bP445qb&ufuAs(CXdJA2HuRlyW+EpAZ6?0vcrW!+u?U$4_@I
zNZ;bn&}3dsT^<Mps4yU+nOA&CbK4(V&zJtBWM#d+Udl|>q$U#Xni!p!7nfRUNp`ue
zB|MxloSg)LNJwnN#c0LzEB8M(J^gsgHG+EA_sP3mRCQI}T<wKmTgP~-TPOFPTYF($
zH^aH}vSmodI4v!o#|FT~QlswIT*EV`%?wdwGhhA-yoqoU7XYygp=o}b;J`7Ug7|o_
z6D9uuNLDX#gDKw%`8yoXc900JuGr4Q1=U`XsXV@OOB(^Lv)V**P?;bp{j0EQwZ9T)
znk}~B)~Q`zri0=S9ARvF?Ts273<PlqAKtrq{PX!guV?^jn6-<q&=vpxg6l%>;i?az
z(*NTE@W0_d@}!Dyq6`lBZ<0MA(wvoD>l?HdVMMFsq$FSj7qTB~A4<gaUV^5jg|3Az
z19p{~i(hoT1~+^neG9;t!VqFC2$R3;QO-w#nO_bf!sy2#N&<!mgM>vu(L7;Cff6uU
zQw@r+s%m~Cwg(6}B?D@69r9HIpiGF&kt6u$OVQd{upO9}XASv1<KyG%QPCcPQ)h*>
z5ZcoEo3?Mo2|^n0dI_8UXhXWKt?lo46Pj4rSeWLpp3p_MW8%04>pxebj*}ZHY0w0o
z6m+#<5c>Xk=T$=d=cw`Km#~K1+$!dFT@kQYIsFn>P?VjXnR>Nw`W(%7c6Ey*>H{J2
zo$ZNMxrH4pfs>Q5h=DywrG$_pu<rpIC!)3WG+#JMfPKn4K9Q4~2R0r+v49M2l{QnS
zl7A^Wq?<@eP6a2N16>;rEg2X}W%T6Lab1AcmNtJ62F$y>35f;K5G%3g0&k+)=ujoU
z0NkL@KewOJ%&pw@3JSvXEw9PXXX0m}tWSCNK?2-ZxV9Z@If;d+7|Va;ePyNs2U&}M
zaU{Y7!L)Y+X{dHP(^KUo<}|U=3k!u=D2Y6?KDXyF)(4}=Jd=g#KxSSVNXN(QTcMJL
zqX61Ah$8Trt^2_-@c#S(DM@m?W(pi`(9o_#gY6I?&Q^IHm+wj6zZVBxFH8zJN*woY
zyNjhk+jx_1?sa9aufD&e1d6Dsny!K!3L^JFK;^F`@WnwI4Fw~Znbd`~BQVbdM*PAM
zCu6j((hKIn<)?r5Py1hCRJz>~B<uav^?P{isRQRwG4F<^;D1otvCe!b{_^eaRo%{c
z$-mB?Cey3q>=V{-dn185-uag}y%%a=(Wa33*`Nk<cHgz>(xbGDvrivzKVF)eysD29
zH2eZ%P*PyxfZ-eI9{0Vl)(~>izHM0|xtD5P=W`Jpkh5Z8Z(?Z|)NFi9he1gA>lL-4
z&fD|+Q6pe80~#>hGd@yPk9|N*$8+uYKev@g@m^4a6w1pfL`UJ0FU?~hocuzJkATwh
zUX1em-2FJ1?$`V-^f>LgtKg?=B84rTKLiu7?YEnevY7|L%2xk-MfFs5SQ!{*SC^=8
zFMz`ff=bbov-?Gbb(9?ZI2fL2Sma0ZJ)wDP<B-`}{sw@?A7R<?FoHK6gm-8m>VSn?
z$3SzUa}Zl738qx9c4c6@XuFKnRQeK~m|u|jF5$#O_}WwK^~*QlV5_Y3nH=vgYXz6Z
zcd_Kh07;FDAfN=yHuo6l9|6^`+l63Gdivm0v$B$5K=)D&4mQ}ae=jSCY++zys;H<r
zIKO#t0;$kU(^3+6fghk*KYEg_fHpt8{=KE;5oYw<;#OJ#g#3nt$iw4Qz$w2Q&aR>o
zkuG&2<V7&l{8BNIQ$a%FCSt*P);XFAs}PQQ;Uqgx^KSg;VSDPd?;r{7N2xsZ12S-7
z56&}zXsPu2;L!Y>Z}&1)F8$q(NmA-$WwoVX02kQ|F&`2pWto+^f6;Tw3WTUQs<O(H
zWK%1u`BvB0uy$GY-D*C6cey3m3~W;X<!{YHfSouDK^z@}GcrEFZT#CR1j%OHEGUSu
zN0$VxPmPdXxSVZKv~d$v+TQIT&LMT_lT$Bek1c5<Wf>=3LElC=q;`%R@E)Kg_bS5r
z>FQZsasY%bd;??r{>9Sc$Yx@$Uq$7G!kmIXs;hMjNpx9wU}2_w|I)jhpPP3R>-ci5
zw{@oN_}s$b1v4lltc{Dm7zvZXt^oM#EW(CBT+OD1Js=+7R0QPueas-(?*xdj0D>A#
zo%IkFX_$#^noUWT5>CE;#B2=w^TDi7_Li3Tu5&Yztv$&?o|{3<<}CUpY+U+fHb#aH
zZDsMtb;C5QcL<R`rM7oE?1@SL{&b!vX1x<T@Z<IUz<u>tz30zK5U3-#IJpA@g9z|r
zIkcDotHzA`G(szps|I%QNE;LVr;z6fJl5mjnSJk`Dl9Lr>53&syQ|JoqlkNRD{gp%
zc(=gVg%mSx^3Ra2=FI2FczYXDHW0-8ImQ4l(>Km<cpXkm$Nj@rHu7U0C-A#nfX&AA
z2dRsx7K;CC%sfclrL*IM=OM&J^V`IDuXKs!RL9YqufB=c;m<&IG<4d@^T&o8(*WCR
zpPQWz8yR@QA!4?O_|zoW7?5|TXFp}tuaim!efdv&Egn%4*g0XC0rfO=I+j{);wFRh
z=1;9PSqOe_cmxx&&O8)yQ;3p{j*RrYU|w-k+8Mr0r_6Cto%2WU90zRBZ@E>DSDnSX
zkR=GX{U_Nk;zi)JHfi?i6+1T9usN&#RNe7P%i<GMIZKfEw3o!Tgb|I6N~`K?x$SLa
zmNCVqoZ=_<8k8D)T59njg6_nwS3^b+DFSuh!tctIzqkO8fwh9PA}YDaTeqtr${X)K
zAM*X1+$P!ezybi58d$jP`>{Ml3HV*w=%Tv1SbWACHs9WhPp9$Don%A_94B{u6(+L9
z^f-LPK<hbESO74e+o>Ov!OO2J4?;uXSVnxH#oX)n=;^L@SbT((mX@c<H+cwt4$CSl
z@Pt7$3<F`m(9_#^8<yl&RlPA)dCR}476&6N^f`S|jyoQ~yCB1l^}&^nmlq91zc`6K
z+w(&3qt)U6fyw4bt@OVh9Mn2URE#>aPHo_He`*TF%*Y5hbEInx(y{9C9#<Ds)%-ZS
zSOn9fu%^|R*ho@7<1dDTaPli8R0cE}($LaQfA|QE2Aql5I&V;Xz!+6oW)-|XdlCgr
zjZObUM(w7nUoE46>H;5MT~(c|6y*i;5v1gSiWnLO7)SVrB@;cdssNR%%Ik4oI@XG*
zDP1!AisYXSxvh(PeZFXjBC2orpwZ;Y-S_PnWMMkBMmn$^{#Zb*w;?aRzy$aEp;7tk
z&w;P`A(H@XPx`N2HU;hzkwUm!@8WB~xUL2>UICeg_{0K81U+fda3;V@uV_%CqM>o$
z27>ggx$Sy;+9PoZ(S7_C^#LeP<gZGWplWWW6Y|)jGOVzlJkJR2QGi7Y6d2!$STHI9
z-N>)S_2-=(Wt_wDU$yO~-j)K0+vM+tD(WUV3`5N7sw!~O*1uVM_bw$y&+wn28q?vH
zvoka;Eg}H-w`Wzajiv>D781(Ugjbg*S3CtqI<Va0m}3gdHLP_eYNdN0g_4wc9g40+
z>arZeiv_dccVYIa)jKCze@hzvuEJ1mX<kcTxzeM1=?(D|h_dpsj+PSr>o+Fn))!g+
zTspT+G-y&<lt&~+9#2e@B1#(i4Br(2_NwDr!UTOl(O8!u2l%f5<qG0x>b-l*pX%Mg
zkYUhq*=WN14|;6adq?f2>R)eFgqANscI{&-Dw8+Y?hpJyLC^`5)tQR@zZdOIZ8}bu
zN0-(-+cLoF2o4l>Cw@Ro<~Fxe1S+YtA7}8>!Q7)wI`U|C5MZO0Fbq-QuRxBFkZ5UZ
z?|^WN`vlRGwbv)(AVMzwHw5N86nAmq0i)|mr+0PwVcwKHMStN#?0Q0J&CEdR2n%bs
zBxRjYF_qP3s*;tDD^Y@~)`K*!C?CEeJ-%eVfmB^hx60y_pWC0B`Z4|qV@6R}R%NEy
z)j*lNd^f-}(dxs%@Zf)CksMMcDX(`4>tEmhqsto(&_n^4r>Q=wp{4<otaiK>H4SxE
z{mOSz=_D)bF#LxA#}l|@7}6t3D%y31tHkgfw6v!jylQfCr5UBVP!Y-~r&eUxG*>-N
zOvD1asE&>`5*M<ECuY^v>lCMU%IvR-#DAes@YJW|WZXwI9_s{naDuxj)2N-<j~6g7
z?(TlfvnBYV++quMQvcWKVTpBplaue~9)suQNTKfgF3ZkPPoSMjY896k&hC{>{L3dJ
zYljy%rQH=Iejdh{$QNRgqr?7$TGxNviWNss@j<{d(Civ&{S|u+>Ymcz!n!3Bxn1Q5
z<mZyszTRE15%(3J?&xec{{F{b5w-t-eDiQ=vsy8340a;es#5wR#2)Ere*qFIzV%O2
z&mMIR6`Js8VM%Dh9(!}}=)5(S4wG*jRUg+lZ2~B|{1rd=r`?NMlV1A}`R?1x-3P~q
zVxbcOZGUKQFF%~Fk8d|ltaQLf`1u8XZcUCVr0el|F3iQHCv0tj-}A^Zzl8XV)D54s
z_a4A`Xq!PpKxXGbdT(oK?&P8|WNLz^R#DFFfM!Q$GgXBFjQya>1b|{iq&=4$Y3Q87
zM^0vNvT?9*z-uvq?ae|h31M$C|EGry@goH_u1hOma}W6}ip%hgguM@=KYylYW;xwl
zVIWI(+3$$co8SiK!^ZjM1F%=sO4&|{O(NA91qXGfUw^`p?<U)g5Gm=LA5-J%Wc{JO
z_fop|W=F^WAZ&@Y`u7Gd*EP=bU+cE;QI@h=>{vn<@uQzMQpT5G-$3@4<)nlC{g1bk
zxI8ftn4YmS9!RRK1bg8%16?Cvv{!+2>drrC;DN7Sb=o1NGgtKT9s8~6ats7`R0CWo
zY!8v&b&6`5s(c6QJrkO}fU@BeM0>Qg-9)(K_b0+t5r?G{d&O;oWD#?B9RX}J3~)h9
z;CG1=qnxRKwF}q#_KgjQA?rLYj}Y^_c=*wg(NAUiVAcQJ5}MmT#m65;D8AhC>^Kv2
z`?d0AC-xcJqFbLea`HKx=>EgB!mP8Ctzm0I1t?X7=vYN|hDneHP?O^B3%}Ywi)Od?
zg%MPRpuUnro06Rt%urxIFPy45KGG|#N!VB@%4DTvgUSi(c)#-1Dg-Sswx!3HxhxYL
z_!Xy1=?XgDM@l(|`?GB0N7Cfs;D9io-ThrG50CTHruZPtz5e1|i(+j5R#Z`4v4Kkx
zLjC#*#njLF)>e^kqdn`J0hMKB%=f4jx<Niu=X}WQCs+6C{HY#zhCLK<d^g=+Y`FDl
z`HtOWy<50w!%XFMM%oW|`r2Lfha2D0#S8$3t(v)Z69g}*UX;JFr11tqwM<{IK^WFM
z<Gpe-zKQRZ>x|!^G)p2MjNfp`MQ$T-dMToU=|LcMJ(@^;HU%qB6gSZKsp8w)yX(Eq
zg+rvIpDyxie?`8#Hg4qf5$0<$Z~Y_;e-?o0`EFzFB3FDR3-|11I3x>$J6Y?8`UF8E
zEF|S4HdcH$I~RM$NmO*KK@M<PZSEjzYa~!G1JS|WK)_{dff8{gyWe$In||~hXfyy0
z5U;JH3D<eT+Y5G+jak|68UaZJ55T92cId)&w#KGVo*XZ(vmv6>Avr6@=86wJ2={cg
zWWuNW?_L313MDl7Ktj>}vnXkz#)}qep7AFP|9Ko{<>z;{c3v(H3RIWd0^t}hVSQj&
z2f$MksWWups>6*&Y>-ximqFe3EbqguzjcY}XBEo`1*c#{qp6+t&%LGSCyHBo<WB~Y
zhliVA-cfroGW7eIyn1D38U8JFXBU_H;GkQG(*mA3OL^HKcaeef>P9iV*^F(Sq{3cV
z+b9sM@JnB~&i(4<zgJ%((g@g$a4$7`lP~t1q!oa`7YB#C`_bQ)Mo$6Yh=nEci+qm}
z<B2_AZ^Y9j+y0f?M5sW{=dy<z+bJpOTpw0gSU3}!bi>Dy?v7Bcx98TkTyT4hEt(RG
zO^!;)%F6TD`rQ~G+Y4Q8W|p@EOMTJ%3yOa&I){{={Nk7|2@h-XsqHon)ARj^A4#<^
zXC=ZvGgk94sP}y33d@t0ra-rfZOQgC6uy8(likNcz*ipHNC~Rl{>{!_@<|(Odvt`^
z?sNNgNs1x2KOWm8&q{k@h_xo@jxo5ws~GBdKx8gpaB+LNXW@0FqJGUyWPt|uT)_%^
zI)p9m8%zWd(UE0c#5MUdGxcLG7{kDUC?Cfg+Po&#B!CFs!F=h+3$>!YaAk?fy8DnM
z??VD4G)Yi8+r?()$JRhx`p?3I@R$feSD;L8KRng38T;<OnMaGT{hC3?zHPyci9lfn
zC(^%r>wbOK&14QezaJBivQ%dZXsYX7oz^D<eG%LP#*E5Lzs|)i5s!%0`O9*b*7@pD
zh$#qGfPfO@Z%gK2n|IZlVU?jzb<YV`^rb)6{V8{Q3_lQy>|_H}1Iw&46u3Jsg$Axy
zJ4^qGB!iJf^)WfB==0~+UPj?nU)Ktl%)x-{W0SG3_^(@-CB#;Yh{OI7D!{v)%b`;h
z9>jce&@+?>FT-F<xbE%Y*5Q8qXc1@M=;QS#QkQrqSG-?)D^DY{dSfFZa_lYE*1!b$
zZ8!}5hbG@Q8GGJ%L=mEIM(YQ*x@hUk(dz&WM#bF*bc*dkPGRo4K-AFC$mvZv1W`yv
z0X`8K@9VuTzx;s#6;A%zvNBtrBepo^Yj#yv!MX<0!M;BqLNSn6+j#J?4)rg`pxgjo
zO`x4N7#5elA&$V+uQi&8<>%7Vvq4-*?~xIZWu|?DwC7vBwoW!Re|2vt(fxz-v?`w}
zXnd3-u(YyTJ-z{v+1*S$TMoQa*Uj<F|JQ6?#0G{2*SjslK2XJ{825+nyP0TvoaDZF
zcsmuh2Ely}5LCW-<D`N%a`(3(V(8=}g66w>^!ohpZ*dIpgT|}g9a4kylBT}0y5<aS
z@)s%=77absiOEUW?tc7$DijD}nUxGkR|3D~hvYlj%F_D42sq7%0>-sHv&E)4Y}8zR
z9_Jg)>u|im0SqM-l9HZz%Ev>F02id?(?bZRoz#6uO=YrjfCI1T7c2yJ@Y!u+1W{~A
zca%?JBM4Ob5+9esdmF~PFljsp?Ma;sANUbeLRfeM0Y|LY-|@iI_GR|fvcBi}OHUJ3
zvDd`NC(Ir2Mg;QA;MC%5Bq4((Iedw7)yS9dD+{8@mJTWw;arpNEp(5vZ;1NT-ysNA
zSJtR<ycERokBo={%@DbW8EV6q@Fr!KF>M{%N7;|Ev#wV;N@0Zp{sKr5WbF7h-ozsR
zSY!FG!qeBB5v93jy-o~;A&O5_hN<!^OULKWEwyeXPE*p-TH0E|Wde?07aU4*pbH~d
zSm$sm9FqwJhlXf=juj7UP)Z{~t1tkh<ViMIpgU|8Dkvt^KP;U<Nr9bpeBc7E;uVGr
zc#SXOtFK%`0XWK<*uhHQ(bX~Q<QQ=CMtoDMixZW|#!p!4HxV*B7I4Lr)M|$@Cv8D?
zm4>kjBl2nI_^CoB45wh~9jPiJm&ArP_9-?y9~7EM1@D-M+f~GLKMDe5tdW?Pr+&>a
z4+Aepq;FE7Rs!QG5JVnewLt8#f<{)cCJjm)9x?mx=HHWgq`*q1$@W@_uRh4E`0?%2
zeMGnR)z7M`mOpKVhkMPa%X~M0sD!aY{$_Q&AfWAq_4EDGJLD7Nmf#cs^dPxMOtq~H
zT~d<D;DEnILhQJ=*WZSU++ONRQDK1O7$951`XD7u1XlSBzvf#3><~I?|2wD%9eTC2
zOQBNR!n?t-*RS^{EIIGzzVkqtGd9*rb`%gGGRShr_JffWvqo&yS4{2XBre;%w%Mk>
zlfc$_)?@*FT@{U)z8A{sk69i`rT2q%P@UOu&LYjdaM?_lM%&LgHuSV~v~*mIPKn$Z
z$1;5aA-41_zXzvi`M5~2BcK)n9LGB39`%KXAnYvY>Phb8$yi8JjS%pYf+K4yD!v%@
zh%h$vHceGlRB#Hj-;%f2(bh&mB+b?=tkaT)hLqBSR8o)qO~#`zs4qNaWjkl+Xs8oM
zz+c=+^g#T{uU^97X4tJ@7Bw_7Jnz5@Q^5CsAt<=vCD5EPZwH_A3Dhm*r;9y*&eZW$
zVm|w0RbFN8ZuYm{TuUPOK(x&TV47i}nHbqIGBEUyjOeTBGCpD)9~og`pbRA=2kT)s
zzA*@c=;d>xj|&h<xfmaibd-@u^UvRZ*F@KkDvZGV6EQS1`uL+3s6bT?TUZCd;3w}q
zKd^f7D-E{G1UsBKxviCrf2xLDhZX(c=K<Seb%7FRimr(tKJelMcb|Nz-2Q12L+}z7
z|Fvjh+4gZig4Cvq_Z2W}5Mrh;?zv9t>rJx~@|<{hK_RFac9i*bJJy^CR`&7Y-J92|
z6M02jOs?9vsG8PmTU#vGwI#NHAs1zyhsK#V!7kNT>i>8x6HoyHd+%z8eP7Ga!R~4Q
zCX>1UJ#J<ON*qXwK&yR@tt^S_#sbQw|FkknA{!s>9LM%9LkQ@RnFIYhA5C@TTzd;#
z8!KB|bVRm@8#_=!MPseKEv4TAQBv0OGk%96DyEyzZ2GEl5=DQvjB9xCiEwm`I`|?%
zZQWd}_^q%YGCby<+V9V}gwiG9N$ZRIC@G0j=sxfQ-dMbg@#@t=-P6H4QBsdj90O!K
zA)JV0lm+y?ca@7r?WcOXt~}N|##s5ybB>R7wtS9`W)3O8ruDfKhl!64{M@~0R2Xwq
zb=Px#PizD|jKweNFOQFtf^odLE%-thWjFD4PJ4g76$n8?Y`Ap{AFL3Q5P$7UkXI2C
zqX>Tz<!EVbg^0SDsT;rk8uUYD@|=d6+H)l&{nQTDZ(mJuQ6_c}?BY27`GvV9Kg(FQ
zLfkFvEXT^vMUR$s*_D+gZ@HDm@VSz6d_gE^ha*V>F4OLuk6|S$`0K|%y~Xoa1nmy;
zSyQP3x;TKoVZB}oR$`%{A{jJqqm;gTCiRuSaVkvOPVdU9^V6;GsOU$538}pp^_GzP
zM{Qo~ant+vcoD#ES@o(9g()rP#}}c3M|?0pIyt?e8mxV@PL?%F%}B{It!b;O^5dAD
z=+~tfA#DW<pi{&DwR)BQ9E39^y`*RrMzGvmJl@G+Xa$M4xBLPE0t_z}9C+V)2|Xl@
z6{QM?c<i%}X$RT8(GkxwJgdu->OH7ggc(?bF%af%c^BIWUWaGIYsJFczIYD_9v7;*
z7b!?d5xZYLMQt$Txao4!xAA~$YGLTk+dB^}t+OE<_V`ldOpXTkbMu_TNu95hsncYm
zNj7Xfj2%H1m-@)p9LzHVDeb)SLST`mP~Jya^s1*uUjij51g)CYfGg=pKc3L^MjE<9
zz&su1CQw{fZfJdDQ6mkQTqlBowi@Ft;-`Ik@4};^e5U&?mCm$?qh;?NJP47RDSzt?
ztRcqew2YiS)_XU~x`E491Rxl&C*RmiIUFt}eh;9}SX-Pkg>b#B?YFpB&o4fQ{oi5%
zV%v2bl(+C|0@(K@?BB+bUER5yP*d|h4Yp>XDLupQ?JeMhXK<^Q^bNE~IN_4dP(4mn
zN0wt7QvvD^vonB(dYyY)PG%j$|EYdPj{@3RWV<QY?F0k`0S`1dI2hF@+K;HCiI@t=
zrGSv{s0YMkRn-@1ns&X#LK}2hJ;@(sqGUlsOr)`;Wv)D*)*gG-N<0dt$gJUelC59w
z40Q3dA9=*JTj{yC5eOW;A(~7-FQRKLt}FMcdF|eisc2$gg6zzwtFp3{w7EH5`f}EN
z{Ma{+)x&CPIcaLZ3S8cglOy<QneK?;A?i)Faw7LAvUeE_mg<!MtT^50I6VwLzigHe
z@%SLT$WY;HMoaOzv=8_@!eKv68#hp;6(S^_!G!FF%n-_hzhh@Cktmpc8;i5rN-scJ
z$<jvSe)r(fxnp5|NfUGw;7&ZcvrA84*e8yHl>AJvo|zUMT)pC^W<f(8?;0j?v0h&b
z=T&~HybH>a^Mv4m5ry-!%HPcvoK=Q+KbQg=&tgO=AfeW#{TJW=$JAQ@Wwl5B!W%&;
zNeKl6M5Lv=OF9)rKuTItLOPTV3F$_VlI{ip6={%?mhKRcj=MbX_ul(G<D5BX#$)Vf
z@Bd!wmq5kep{hu`(OmEHH_o@I_X}CUo_#s+n7}gu^!pmK)0lvOO;r=Y&xRfYmWYl9
zhpk5Ozjap^@5Jodzq(D&rrx~qn6$5}?_O|ZB@d5VqqaV(YX~IE`clqPPPpDcxz6e6
zVSOGJsu~imD1L`Qn#K_D&T9{_9Br>-e!km6^$v`wpn8R*UwE<3`aSaK>1lRZTHfAD
zzF~f&Wu;VqO?)vz+{^g3B#$0itfXOCePk~X!p+N$%WbjMZzc{OZd^O}1j|P?pf-~M
zlnul&v#0%SsOjFD2mXC3-=k37fhN^!!W4qNVwR+O$Zg6D3<C9h%Bn9KJJ_y{x1W~T
z28!d}ZAAfqR1J-S?EHFj(OUw?Yb9emLuWomH$QlYQeEG`yixpiy3PB??=vlJccG&4
zdKozKo<3saU$lm&!;*RbiR>YK;RHsxhd7(!%@rlAq$ekCr)~|TM8qqL%jnHm!KzAi
zXTLi?_&azt7F@A3>;M0r5j48!k@3#eO=D^fgB$<->&psP;mdH5maTE`MR?87`$nal
zqGZd-9lqN5ASaFd+XM)9uDCv1%SwMo#9<JP?}lv*aG1NlyUj6G{G@4uBbf?5i+S{*
zA<r~Pk>^?o1eku1m41Mi{*Uf2%`V_e6?Dc#K;?{3gECMH8ixRO-<)vAw}`pv6PF{1
z`1#A$^9%3`{30eI(R}taOFS6G&b}1Y$65WgxjYx+gRS@r*~l9m5k&N6WzWGF3FyBw
zl`sm#R?{;wz8%H975yYp=@~4AF&y^>_^Aoyk%7U%VCH&-u>aKp_>mvC=+EEq0Vkjv
z0jU+x&)0o+od@9yif$DrqY%lqZZ6n@Nfg=LbGn`0Z);_xUGaMIUA866Pz=W9b%~e|
zvjKm9AmU!R1^iaIs58$ifOKAx0&J5mP|9HOT3A~A=<bBnKVv@t4n5H$Qu!V7le<TT
z24!#{w*zVc%w6B?I#}ePh_6n{K_8%*tUGIba!d*@k+S)!XYJ?D?xtyCmv)Bi3HooW
zyzDjz;apU51M4#Kk3#WnZNz*pj!Q;+O$52&R*&G{x!bX14^N<&i5H2LeQ6Df56q#g
z@!h5lYDc)Qxbp4*&<}XAiQR3)(Wy+`1B>vc8!2)3%09g1|Grm!U0H$RQ?ru^^2nwv
zRR0=&!C!4YQ{D$wL)9W-DgmkQM>0<;B9(E;JW&z3(WChh9e@g8UN&D!S0X8^WWVkM
z7DlKVbO1KQarM_)wcczi)gx_ftfhBNH*v4~sA{Fh7lWx{zwhl4CwLZa-GUGunLFZ;
z-2+BXQTIPd;%=GIt-ng0{^h`^sZU6hm2<Z*{DH65Tn*<Yu2#L<Ev%mse4z^2_!!5V
zDZ+~x61=_?8L+Ovwg5A{;hg$F0j|2ni_`AQ`w{8aEi67qVD0Kuc@P{x8Z_N3uRC00
zL{7OIN~(bp!ji(<Gl&R=nm#1sBMXN>><m3S+DKwl4F1kCl+(W)-J$ZEz1wg^Us6<b
zckA1nU1UZHbw6s=E-WdqtG_LgbTbw?_IS;~@#NP1x4n)|PIpCIF9-f`Ny%$Vw@X@E
zi`W-t7)r=!la2RGI&=J^9NyaJgOD!ZXG4k9eHaBtSEj&J1G9(2llx$j1)mz+jE|ia
z<h#wSUgbPyeD}&Sue{<5+u{AJTR(IAj}wCJA=S>S)Oac2EF$f#fg|B9*#FcycA|W9
z`d9tlmYnzkbTw6VWFT!^__2A$uPh6!J-973+`P#Zz>F+e^5SK;UnycNnExlMg`v`S
zcfZ&(e~s$8H|vXP5@fk2rQYSTa7l3)io3U=!?8nEUtH(zrqw^whnchH3TS2@$#g6&
zk?y$&V>=A#irF5weg}aE?0f}91?WEV3$oB%I(06o@d`bgDfg2JT3GT8#NFAtjGHf8
z?FURwV!bbZ4Gs#t8Quj22H%co3_j`pp`%X2tIoLvGnf*yKNf34z=lFLs}46Fvbba%
z&zj=!XT8W(addUfoS2CCN&fIZADc}=4xAGMLW086DF6k6MS5d{yb}EuNm*6(F4gi~
z63UaI!bcD>1}_s35LK9+qalq{E+xv!SNyKuueGK_FL-oJ<nGbmde;?0lc&mGjxFG&
z4@5tveEz#q3}TZ<pwpged}rYh2TV>?Wpx0$*rQ~?WFg5K))?=kDm4v_%IH=MuQzze
zKe=kkhfAQu0wc7z1yg0R!259TF9l`#Q6M=EyX|2mG)^Vd`lXvb%yf1Hg&(wjW+-p~
zg{xO)+C7e~sUPS!HZlT=kp5I2ViVV}WX|N?<qWQJ0Sd<>dga?9wxg9%3%ODy>}UX_
zCPOIzzt`pu`JrX09m3aAMz<y^0nZIykWo(M-afqONUqs|UH?Pe2O<g--&U*XTDPHo
zcM#t0?OGTpTi=cDgrJ;zl!4Ka_<%)TEp5P&%*({ol|@5`+Fg4M@mG5GUw!?qqr<)E
zM^;4-L58X(F$<kj`iE54eOTX+BEVQdq&&)(&a9S9A6zl2e$Yvw2CB|9?q_8AmYIH|
z`Sj?*^FXnW@YwuC%BC++y?!#rCM%`m6t=20AkW!y$K9L+bgI;DkyT<4AF4RTJTIAI
z6<qCiCTclQfJ`y<>`P^LVhtRExU9r@{UDm(bf9!0-WL_%@dVguYI_@CD!2!u?w{YN
zzC%iXcK$Ati?n#!^H@nsT|!?qH>1>L=`_7RU6^|zHyDQP)_b;a(!H*47#Yz3Y+Svc
zO^aPfWZ@r`4mrB8alk1JdZA}c{)sLAg<(pNbD&y53R)`o?w@KrOH8_r`Mnaw&e<tb
z?T7iOAk%NL_HXsEilFlb0^uUz2$P$vsVv`heeqA8*JusAQ8QLZ$IYOoqr4x$sP|O2
zN)3A<U#mJ5269kUe6EM&p<<mnw1^JKnq$?5`au3^3gibHgQUCli-xJmA?uBRk;x;9
zUyYt6KqUj;m)qdpzHajFEG;oMI_86op^cz)oh6e$cnUs1>Q~WZ<E{7UAjMN81*g3b
zx#Bex&^+zJQ@}0g*^@WN<}20zqLzt<=B1H|O<qZ|rW&{9CFci>PJoY;&87Kj$l-A_
zfpKW2ITRPym02Nsd5IKSPJmrSe1P2t0XkY!u$-pNRFlIMCW0nXC-UXKX<o_7#ufp=
zoeG<MT&%>+^-aiD=Ua8^1wMyE;4jP%j}^3?`82iU%W|s2yBAP=LMtwk2ZE+sC_%mj
z@EkmIaDumNigB9!<ga9FQ8gN11wwgzko6v;)5WMRKSkeAZ(T;-;zDjAi1clrS9omq
z<a3l^f_h#hL2&HrkAFrte7j1+zvV#3y{NV4qT#HrDk`eWb~-#@l<K5@Q(2S!m=Q#W
zbp6jk*bMuT3dJQc!dtA>?ABf_)uAHh`G*iQy5vQDhHX)H$-DZRvNDVH&2MrjujY@{
z83@g~14e`nDW12u=7JyawZ$F8N~=WyzjJ9}?QqX&wVj@A{#Nw3RUbQ5ze(sa?ziG*
zY2~LTP`(NiNn(vlRc5A#g_b{XJ}HS}D4-+U0#XHhrMbOYQ-iH={$Pa$O`J^*PEYc#
zt-bPQPEnIrzbC|Z1s3M4T((^o4`{UymQLdKiusN0@%#RtoqGXNmwfoA|L+WEf8|H@
z`TYdxKY7b-F*@kDWZU+9>1mY35c|sJ)#lnZO~P*&MqV_yFbT3YS7b#?etfQV*b{E5
zoZbd_YekG}MQ+)+6_bhF_-}OY={p2>o;=a9wX~jFn!`fMe?A6X&edKrpmFeBY4+X4
z!@^y%uggL`2R9EOpRX)8)Hr?vXLb9|I!uH=*Bs$A0Q~LUNhgrD)<qBR8_fWnOmWwW
z{yn*f`6;Hql3s&7N8pe(i2GU%MvDNV(tCTVtgGy_)+QiVPl&)YAIF-K_30A^v}wFK
zTdo2rK?i+~UEMU`P7snAjHLmlcvfA0^lTOspWvugFu7LV!0v$|L}!BDG+810c~Www
zMN3M`X5uY$sWzG*>3J~Lk`TovB$Hut@`IYDv%RI+lT`Pqt`s#s(W1$9^wyVUD=HLj
zXRM&S3YVj0U_8W71xkLZ-i`NMSIO`-W|TM#pj8!Rk`<!1cznuEJhzNc50u5IR8d9l
zN_jTDC7(U=fsF;~?j%|*prZx7hAl8a_L=4wC@8_Qf<0=Lt3+|O)LjW#+la>qyOMGk
zhReAo-yO2mYhPfp0*Mw^g~Q0O23;7Y%Q*|ikWj2sX@#rZ+1_N(YCsMh0E+7O&m*xh
z5#B;w!;BIgwda|+1zQ_il*fBGt_C<(PykkLpyDazn_O@o=}G((?VlLj1i{@5%#2=D
z{-Bt6SMcn~)5~MVWIXD7tfnumf&}g0f?v~lsu8JQZx?P5Zwp`eI~q5%usdrJj;e5M
z<6YD<Gd|GO_!l{db}cI7V@?-(b3#HML=?yOvpl{Q7aIlK;${VD$uEjNKs;a|6mUI}
z+m2EA2s&hLCSFl4W^{n}o+mc&3sA37fkT#nu=7`GhiS^D^@Fe9)Xs;`$A_Ro1N7Z7
zu|#lzCI@~I5mAK9w}GL!SXcN&kTyk*Rwb32l}ji$ry7^>4kHuORFl^|7MjnW3R9$M
z>-ongpKIMr@?Kfl$j{770Qu|BiJaU55${88KGKe6lZxi+!O|1e7S<4J7A42c!HIva
zucNr%TBK;WlL^vx1{#_)%3||fG@o?3_gRDH@1hJr&zzc?3irB88{g&=eNqHU_XaD}
z^k;uBZ>}!Qg{(7&1_ygNu_@#5AiGfGV7CEZQQq?CGjirou>E5b8AG3Idk>j5hv`=k
z1cTrM@GnQf>HabB;yuw*AwfW4+Lta&K|%m&Z}3usQqPRy&#){%iHv{bUzvv9l;i9e
zPQei-NusX#R3dW+(B2i5yI7v&!I`AsENv2*9%whQ%DWob0P8h)Ji!P|@B)L3>w0S}
z2q4G=Jf16B7QD}Pef;Wl(s|JLRR*kJx}wf3VAe=fN)vQ_aT?fj8gN74js^0o+$%Pq
zUfbF_HLds%xBviofu<lLUK*4#voNHSB+-If<KtUOghiKMR!SEK@tVHa2bMoMDY;iE
z>^#u*?8L1}RZd1C9!4$THWiiRvmP_R6onmM;;Sd|0J`h{JqiA@&qjtpJgm=lBr)=i
z08nl|)>JQ2;xQ;wgcs6;&^2fZ6|40WfG@VVcxG@0>}MlC#BB$%F1RfyqB|-0=&0dX
zFU|Wp{MF7+KjM2S7_TmpNkQJIqobpndN<(2C1c~!uje{1MBO(%Un-D>J|fP}v{N5`
zNR1!cz7U7|u*n?mcKCyp?c7RKTK)wOlZUYSb{gEvf27Vp=@J|i1Z~}E3fXy|nVcQ!
z)oDu#DtyB#V;+@o)*zr1Dp&~*YzhTL362UNE}Ca3J$0k`U#PZA-{|>tS@Q1B-d;#p
z7)6)n8Qm?c|1oA=34*pnPpsw2do$JPIq~Gt?ieHPbda2Lkn(*Jl8}&)Y2C3=OK0~j
z6%BEGi%(L5XcqWu+W`)^DPC8<>{<1PQAhvI%nGHih#s|aak-a_SD5y1@-TbV5MALn
zEzHzdYd1I!p3goUR5h4C05r0_zde7p#QXNY531Spa;g%^#<vBb*?43_AH!eL^SmcN
zyK3u8+7=B>o2;sCP)P8VQTOQs?Bz?B1u_3mi$?@eW#EDZqZc$lrnR{6r1>1}+ff=M
z-446Ls@V~9UD|$sdAV_piOm)5h1Ys+&i+52OGS?-^%6J#eJ&jzGXSeXkGpdwW|V<~
z5D?6PQo7d9xz6>3^YLRU(PePe_9aYs=dXF+^l#mZ8hris-dx!qc|c2{4hy=+6W-oH
zWhckE@!#jWx4&zJS?RA$Q73MWf4`X*ngr)~`#0bOmz}ZqxPVt(QN0VbcwK$HM&${-
zLxuFJWJe|CKL2Q0d$-M{h1p+Ig+<vG7B+3zTVIc_D0*ak-~RXcGmAVu{qpZ~92_Y}
z8`HXSbsOzkt7%eIZ>4wCjuMrz$tS*Ios)*_FUMnVM`phN_jCW>KaMX9ytx0n)VFV6
z;NMLifc8m4Q{gY=g{YrE!^=dcry=bU^4Bl@Cyd<F4KbAx_T_QQ`hA^aaK@!4usmSo
zFicr0p{kyhc`i-luOQ`}Oct@N+qBei^2}rx8^f?-d?;6mkdW|v8g<{2k@2A{F$i^J
zWL~9YpOkEueXggY6)ykRGE0jSM^$M*-Zm?8<BHTJ9zr5@_}mstXPjPLhC*-Nr%$Fo
z@IKs?3;#JQDf3JPv~_AAZnf^6eb|i$_HX$tCuIuT&X!N7U)CTk3+tlcbAQ-cw0c>~
zl^}dAkRG|ZL(qOxt+loPgq|2R_+W@CMhwF=tMz}bgb&2PJ2P6OU$N3NbJ-Nug(jQz
z1x9GHrJdICCItk12eOt4SN&LvHwi+Q;XlaUQ=agUD{Hwf#gaSY95{Lp$PE4bxxTz+
zX=~QaoOD()pEy+T`SUB@KvaaZk78?oJuxmSHZ{5C)p+KPB?V)9m0~V>-c48>H@9<i
zw305;YX*&1ylTBRL<6B<qwanH&X!RYXkJy*gmhGI2X@h3@x#_>$-AB5<iyB~LqJ5-
z^P}s8m^TTkeU2;=S8V|@Vr4b8Fy;KBn3#OTA<AL0FhQJd+tt&jtk|zzVc)4rdff46
zA1@?CnZMF@#d}(2zbs?5>wQDccO$RV_aN&Nq~^C8*?ZK6%D{CcFlTGjcySH18$6NO
zF2+h8?(PlKg?GdhU0CDEZxhZ;E7zo%|E;t3SIE8|*mND?islHBB64YIIXrm!SYwsz
zQ^egAFvkRvtzNQ_<Gde*5g59chK?b__39~a5=41}#2+b5Hi5wf3=42vLn;=94*4az
zn@fv){|`y}DzgrdGnzNcb9mwN;iVt+Z1z*dw2=Pa`yK_l#zD$>b$&E>KQMy)%p$*Y
z&d%E=7an|38@64)bP`u<Q_k-lTi4}%4wwV65Namg*6|TfNaep$A%EY!kB*dj?tH8~
zDD1Ck_E>^$oxN$^B-^Ff1*6~g*|gq$KbK~ED0Gqqs(CTxJJ0aaPdzMUo`pw;HpB&-
z_LmPfz7NTEa(Z9I-eh;mKzUX0lN4O3M~QbJs0jk`va|DNcFHLAY%^cZx0q+W_S}k%
zirwF-9Hi!-N*A8KNT+3>g|;})vZm>yaq}Cv@!g)g$8})bg<lzt9=vpZx3lZW7dt86
zervehL~q{2Mfcgy4d1om-rBkg!Y(~-?>^FAy{Yu0pqRM05KM1WWFe==v|_x)SjpSx
z$EpjStr#*eGRVn2;C{UL$n;HdWQIWtZ&I>mAIg`+P)%QVS4=|>-2e9gfRS~&$%_E_
z@c41+iz!HJyP9gkf&Ez8<qV?F;n@Jw54^zlH?3SBi9^U$8%<^xGn2o+zjcl_o-wAj
zy4GFdTM?Hew{E2ho$ZX-(8rJQQ`n5ZJp1x{cXJ94?vCa0@emB`tRlcUTMcai<^Yg8
zSBb2oIOm;Le>a<wYkK(Qv)jUXHAcWidfmc*{Niy=N$x}X$+1t5)t*xj5Kqp`h`N-X
zrz8%6(*-4?F8cvB!;5Fn-1ffyDVb^l%@|JLdT#=6qBu2sqSDyd%v{tU@|%h%Bqib*
zi|&AJuXc-1WOeo7c=eFBwzlUC8l)~LIy`~w7qEMdL**i*Am6;9t0~)U`kANX4RPo<
zEDYR!Himh70fi8$P19a{0)iVpQ;iY%7FvQ6WWb0c_~hvt+RqG4IRmaVJHUPP%kL9B
zjI=a^f0ir}H44sOZu9(=u`0RXhJrS%-*yHsxhJ}QhSlt`JuKM8yNN<!(>kTaUb@~0
zb5(D9{E_OJYlTZeZRGBc)6?&3Pv}R6hXZ7!!RFNu%@NeI?BwLh@Ki!mCm5-7wN;1J
zG`EiM^{;*(9)@di<vwU~PftwE%K6M`8(t(HL0r=Okpl*i&-F4I8pETeAxZ1>PHrb%
zzP-yV`Y)*nNfO~FXRy0mQs}6wN2M9EU}NCsKF;?=@9t0Fbsa!99k=zo;OzY8Nqqi9
z2M=+u%2kvLf^`54RSCBvfWnn=m+7DbUiCehfYmRxi;}8mFCLSmC!~Cc!$m*^&{s@}
zP}4K}+GpAGrwkwT!28otQ1|Imlh1!$?pRnjW_TNzD_Tw#-gz=PAq0OhNWt(=#O%lD
zMt9;Ffu0(!+hHv>Qx6$f-s~R;?5)^14q3d)yH``?_h~V6o^yfY{@=fjpDka(q_+_L
zLCcEwCwH0)fT)9In4LRAP#8Kfd={{^z8)SH+}+pPzF>Dh;7)iAr|eN%L(K5`yO)4h
z5(cl%ZAG-V7Ty_ZJ|jOns|>?)y}rM4DN=@rzv&A+=t+UQWk4)gS`uCdtsiYWjo;Q)
zm@zSWX=m3|RtEUV#%XGN=zJHq46+2-UA{<vKiM~qjw}g$xknLE9DjItssP34n6m@)
z*Xy$NZYl^P%7=24=AF<G5Vyj}K*aR)xt#gLV(_V(;6)l)bmzeyPF~^XI|6N`{qMV$
z?16rW2n)|uV}QA|!c;}-QMhE=eJz%_bLu*PpZxb%93SVsZJk>vg6X~kL6UpD(&Gry
z6MBp}M>L-1Xse1x8@sy)mU!ElOA{fDhX=&)CMG0&!EVdDj(s;aG6wPw4jij;RjZck
z-a-_V+wm~~lWR}ta7uqVo4&F6*>D7ZbR`n(wJ>O@EJ1s$<L++mq5%$yU};V+4o(hA
zWm^1WK9}vizX`vjS?wM@l4j6YUlEf2_7(j`Al`>)Yjdmli(_8cp%@dpmzI~8?sI(C
zPB=J3yGIxd7YM#5P?9Y7m&Hcnudv-ns_D>908a#zCO|R>Ym!**Y!;FMQKLg=$_|R5
z33+*czrL;))NWLUOL@;Q3Ub&aB_-kN8_B%?4L!RSnD2vF*-G$9NYX@|K7^0r)#XKY
z8l#~6gl>ta#Mzru!l2b>^6wg|4Em8!n1ZK7(mfy!6)3^dFRysfMMEUa*k3M0|Kuv?
ze^LAodHxf(+SmJ|tF!Bp{na8nEE<~5R3xDSo?V{|18%G`-0)r7rz>sD1ov8UGMP@o
zoqaNvlDs`^j2o=ZX{pxfW_<pVI1yn`nL>jf1UCrm^O15b=LP*M2E1hd&6ziEMt+Ei
z{6KL7;c@zlP}8so+kY(;Jp<w;lh2+vIW;cP^J>{dm{MQu`FTg0F9<}2gXX|OYC|d7
zfxDF|YC?Fm4Nt(daHAFrufM`mip+sYnf>ELuH@e0q5?cz_sMBxbQo?Xq|ThLiEGP@
z7GKB;P=p86zXD4`w<aY>rMdWcuKB!cd&!4q3t?7SPIul@j!()sES;wEz5OWp?p+zC
zyl108qsK+bOe5Qa2l-jqcMf?^mzX2W@NAhGnIP>|%;OY@JOOHkLhx0Mjx13TbpZ^~
zmj)XK{J|!L1W`ab8X`vv`G48ir+4SzZNl+7m<a1zdq=uTE-I49Vs2sqRy)>nCoTqV
z$wyl1&z^l}qyE}HdDYou`R?62s6^Qhzi_3(26VyS-|zgIq@Me!sJEkzf&wQO^U>T8
zse_eu<a;_ACOQ@x?H4a%u0B^~mUsQF0#OclL2M{z5Cbc2ta0-^?g^-j-4i9bIWh)9
zNKxB&h4kS<P6w(KxSHx(O>c&9Z$#(w8`~Mk)ta+a+@-p$M7N$<)P#zdDy8>S45Fd=
zL#yHF=m<RKfHaYW5~nAe&;PN$ENKH{g$|VdnX${<BqyUFSzDfSzZx}3vskcnIlB&a
z2q<0U<>BGs<>gA~X`BCGgno4G-vI+btoC(N>h}X4&}P>*AJ*F~GhnH?Ew4_DOzX}L
z;8PsNshHW-P3`S<Ed(Bchg0eN&k*VL>z%;U+g_6rX}Y+ZJ|$7`O!W8^T~p&3iuioP
zQT|#Y!VZ0q5S>tcco?Xv@A0GizwSF&Kf1?upb??P#UoJF66hwlFuK!@$nkSEKkd=6
zp|V8N-X(iB{e}xVb`2#B4H=nrWrJ1hT+Jif3;)7IdB%KLbKt?#Yetd7l&AeEZp97v
z$B)<Fr@rRFH{bgbRb0ged_ZibO37hX-0EeRC=!N}lE$z+gN;?wnT;tHs`qdrUH%nU
zF6>R*`Ww#so`U1%bZIJ4*J!FYm@&G$@#ag%;d=>K-<YX)(^FGp7v&fq>S!s<<Yc`U
z+-VGYt<oU#F;cv<yC-aqwVz$?u3;Nxsh`$+Z>ZY+(n;r&5N4L@Iy>kdvq43oJ3|y*
zRt^r*e31N4l~&#1VvHJ}lRn!!I1~id%vw#Zd2e6(Mg;SocxuZ%{{-`G)5zrbNK-+9
zJIpP^l|3Tp5eu*RQ9Y+&h8at1#>KOxDR~bs=b@F7px}SfiTF6pTKAK!ndWh?ZI;X2
zH+!2)dB4!q38M0Txj@*;OfrS8o|xM@13hzAaSj>9;pkG;SZIectYoO~+F*Qw2}n_~
zTg-z6@d1%+4Fqiqr(mMB)phepw#|eD2P-JbpIrBAw0q<3vJz7M;U$E%B?t1=H+Y`=
zf;y_Ql88(MH_&1G=ikU(xu?V?P`A+HgU>eWBkNoebscT>=J%vtq2#0_^v=#3{xUb&
z_4cInnsZCU5N;mcd#Q;=mcCZpaJ(m9{+b+}`IuQaJUX+zwjeg*)>bCvenQ>{$lgKz
zDD2VGP*1SJo0PL0r#NN<`yEuSVco=BxuA^SWo6<;LBN{{R5^;cmW7-xc`vi`&Oi_Y
zxLfD!z?8(ABHDG;2&u{^=>BvMT1twGvWB3mC#Kf#(@KQRK529LXSa*EX(A@4MZ0z&
zKQt*t!C^<v^YS(_%*SmrRh^q(_XwOZb=f1m?Pd%0u|G`8^J*O@i&3W?PgkSb!TUNb
z?16h#XSsa9h<fLW4ZvX82#UC*wETiB;5#dMMz}Z~26CRPTmeM=>@meV&@`X>M$hJ^
z3!c(njk<)~qh}tH<ydt*g|i{@PLMS1=?Qv6<6)!sKHKZpWtCxt&`PVVS_*__#?T!{
zK@oQTsgrTfZ}+eqZx~&~=|bo%ypq7<1J4Ddxav0=up*!^E*h`&RmsIuzfFNp$a?Rx
zM!gXHN(<FZAW999y5a0J*mQACDNN?d>GG?%m68}~Vezg14=nJ32a@P?`7d*aESj(=
zJzvScP=1>!SGnw;;gr7|F7^`5B5dGEsdrnTR8#!J|G!#*=+4DiXx>rxso5whSP`4`
z(+kMs)()E8U?x>T@$pt;2>o`=2Of-y@l9ym6A-O)KmCOQL4y0+QjWtElWQn5#j<#t
zYeU*P?m}IS#|N!AR5hecv*+OGv7K_2ri)G&sde4@xmG}`I%KZvOeK09F?MAv$*W}N
z!u{*4-{K~K95-WLL!kZ_vDG9td4SY9{q^vvv4kHjD%?m$AwL@OV!BB=Os2`jqoa4N
zv;5mW3iZSL%y8=KDu3&oyWA=uh4eT`ncSYcxPd5iGr?IMM#(d^w$>ErT%0EO=F8M@
zUUl)MrS5OLi^ehK+h!+=GZ6X`6({*2U3hP8WPFm3ddH|$V*@i{ThHC?DgXususkzp
zbR8*E`*x-XTQnvQMO()A!8e2>t@Bm-kvi$ScZ+O#^F5n*usC{5=|y%f!ctwhiFZy8
zO=myVG^Z=QMp8FOS0=|fZ{NK7GT`-%`u88$Cuug94;D`rCJ}rx%BN3tu@T5<f_r;K
zg|hz5+QASyAFL^Gn!;n?LRjtWc;l>o`h0@|@q-!B*%?(_+bsvmE2G4mXB%F34!bBR
zT3Q-M$lj((xx1~EK!*C|ZPVmnDFY)e3X*IwYRoBckFl<(=`I2->5#Pmlkc%Cp_VEI
z>F=$2t3eMW!PYF4Oqx&ejm=E*7d0@&I>+aN$o_emqXfPG-tW%+8HLs5IU{a2w8w2r
zil=;R*Tkisfa{ZON6hPTVn`DT3jc<bd}p}#fP(9lU-FO<f=?ag7K)BQ#%s9h_BEe`
zU-UR&xnQ4IT;6{2Slb3DneFqh?cZ9(Rr&bXs2+Y8B$k&`@;cmomeA7BW^kf{^?;Mk
zhPur=ygrEu(a;>T<4-|XP0L~i8E+B4F+1CcineL>jwQXAvl*P20KgpdSiv2$!{M)>
z+v0Z8ZH70MLWzxyd2)hIN<^~115$lR1cOq^S<ltT<fEwWCCFK^d<D9bS8h4sam#6`
zI@>d~+^#=euwuywDT8sP-}2U!=+-)1%v<Kt57DV^uCLE~>j8`cZzeS1na;Vi9ZT6c
zJv~X#K}4hpf+A56o(XuoxH|Bvs;b27?a+{GD6_*;_6J`mu^pn{watS~4VI(2is;8r
zbkgIZhzSp&A*E~RZzA9cp<7!MLV}b6{3t$fRQuF0b2iosyW{`fT0c0TTxWj3bDvfi
zvd~7>)}|-NSgrJ0KIhdAcXv=h-^M~xQWDSbQ+**Z*RAtUA3tK=`0~MjzmGx;0|}7&
z8`ZbF{&d#+ps=XPsI$$rn2pk{(J?Y62IGCueD?Mw8T!%IRFmfrY4>{w;KufzHekPf
z_QpD!>vj|0U9m>nrJvXT70W5ne*gN%$bCoci4G~j>dEm84-yotD9y)1wF@2>NUEt7
z5Z!b1q<io{>&4>$itXb_I6-qhH@-TVZ{TBns9*2XYkGEy{!uxCf&#=tn=5^9KqoTS
z`Va->EgX>5m6g}N^fpAm`2W#Us90b0?0j(R9+q#(bp7lKS__1tWtHSc@6Q=M>qB=v
zkr0lb9IZZth<=f~MY%_|Qw^?(a<l<0?`^ED7iP&+Ii~bQJ>gQ6l_l1!DI1u-u}S;*
zHZ3>Q7j8_4Cs|NX>C7gEP!k6ITfrUA*8PLTRly;DBRaSo!?TBth=7EWZ}hygX`=&g
z3l(w1K#s4Y4Xqu#HtL0IT<5U$5_pu760`7PIE{Iq6iuh3xzq}S)<DTCo#bxv?sV3{
z^#2{v!%)iX*CZvt@Q3C5^l2=Lk7TyfV}@1h+K*+O-D6^$g3lyhguzTG`rV7_hiJJ~
zln5|1Tz7n%8Gf4U0@wkNu#2$QPc&xawx!|e_x6*MtjQP67WZdzVFYDy;vt4NHw7+V
z9epu6@^~h<V%!RHet!AE(E+`1oG66Qs~+8Jnsxo7kI$FP>kbfqYkw=_r+K4M-E3OH
zwd`pE0s_%D<|Uxm{E&GqBno0gP!I@_6uY8k)At@$+&gnbV99`|8BUH%Vf(I;X&ou$
zzu@Wu%d7szOX?gNA2!ACuhjS$dd-)jw%4x1q%NldYlELNBf@T&@&Gq*Xlz90yFV3*
z?mTS2;G&Y3mnU0YJGe8$R&Lxn#YK;spU(gxBkBx9h`F`pP74<WE|Q~?7W5xY4R&e)
z3Ka=Mbt9uE>aQ6QNKtV;IqmpP58mRM>T0mcp(1n9gnjf;a#^4+&3a2a)!?u(h}!VS
z#lzf%=&|o;b@k;p<hPf_;y39>xyp$@mhIYRL9-@pd=u$O;@E2F;AIpb{+G21$2p3@
z=f7x(oP4i%Cd~sP5<)>QbE?N}D1w~aJUl{;cCX(!<#OH5lDT>7*sUwKC3bhC=Es;#
z`llETBtg`P@U1vCVQ@5^wnTrEPo^fFDp?7?G)nJHh#or3;sRkN@zarE)0Jg`x>~U}
zcGrADy4Ghz2^iL;LC7}!T7xOJ&j3nIQ4xEL50UtMJa<qL#{4yp)1PWjQ2q^g|DyZ(
zR8?)SMn`#>Vsr9~u@LgHBH`D@EuZ|P0X_HaMDwYpERrGdYp3s<hEpyn&DCzUk67`n
z6Z3uXr5b7zfbVG64ELm&XXqS#6%`fzWo6t?PRB<05F`Ze0X~MglApgnBCaJ*pryx8
zfg7?!VQcbMQZYv<?Tb)IfA%?duDk;53UK|Mot*ZI?@t(;7|PH@(LSJ63X{d#{oq-4
zpEW+z(eyjFB?54ljDwRcW(%=x%a>1!!=6GlIO@00Pr|yq=k*)x8KdNehX&(f@X-;h
zY_W^4bBIA3zoLYM1c9NU15o&@M{5>*mf!21$lk<p1cD2aqEL_vDmNknkd{CQuDp8d
z#>hVPtmcOggq6jt#>Nn4_Gj;JL{JC;De1;`&WhLZ@igjnB;dxW6|WEtr)kfRxgT7(
zpj3Zgqc3vR5%(9sgAcR0m>)WkftkV9mi*?B!CODR3*|G8-U^bYc`GiC_u6sjqiit5
zG=dltH?V@63k4yA9R$$zCa;r182D|o7P^?2P<+fd6+Mn9-O|j#@6%*ts;&=*0z7Dm
zN|d%HD9Fwbr<ZX|Gakz*ChC75BXM{LGQ}`lt|8DJH$&Dr)RBH9Sit_gxv6QP!3|kG
z;iRKux`zA<kNM)jPXX6-f(W1bBk}dO;*oK2>oq(FOLww$eXgor0@{lV{QNmTzwGZ7
z6%};@fnu?Gd=f%wW~l?3<CaiA+nT}Oy9oP7v#R}?+o?nd0|S#x0Ufu`k7Nj!UEl0%
zk4P8zMo-s4PdAHln`3JCJik3Xw}zs&bJenYAzTFup>0;|{LwWR{Ptz9x$ZO+$kI)4
z1<ZOVpZO>mycAi}ZRT#&|H4MScv^I3g)1wi96OpWB|(KDQiclK9WtPtn^k~^0N~I#
zO?0FCzAtwDcBY5|REgYCq9fd$kqV5rNi*iu(pAnbdo8YqoXdIvpeT0rd$rc@<PE6M
z7-78XjQ)<(I4<hw=y=V?Kvneqy?ghNjG?t3dqFXFQ_mgrY*JAmrRMeDc`!C4pNnom
zIz_%k;x?=5Njh~p_UY3H=lXO0{%COYerz)n1kPYiXSf*;!~o#Ia&r1RcsCV7Ji)aF
zCEDOWO5*hn^xCDvxrP#9yuH0HGp^%WZ1`g>Ct7l8C_Vs&k>ODu>2Gwvt-(d>uS~p#
zkdZ-x*XH`=YW{sFcGT3gKGRQLh<jfk@IvLebWCV6Jv7D#+0K@GooZN$dVX9h*4+F)
zd&I<P2?NP)jt~%l;F-4DbrFt|`#NVwOM{up7A1Y2&fF%XFL7@*Q9w+$i>=qsJP_8K
zi3vA7eRGvF?t7u3e`8C;nq7*tSV)qxTre9&<9CzuJ{LE6UUim!JY8RBqQb@+Tzmn6
zL{D{nt%hDSbG1D0zfTxJ??IM^@+CPn^8L+J0;;zAMy!`k(R~Ja1^Hkjok<>eLfC7M
zd<osker&Ot1A(evkGvi`J9^EE-*a^0ekAR;u}%()$47p1#N5&rjL^e@5hoBJB1HsS
zi8J0O#!t6jii8ZFaSV;0_zU1X%KK>X_Q>k=fB^*ooM_&+qYw5nV8ec|>FhMqer?%(
zHZVAji7Zn3lM)g6r3EgC-Tc3_!fpI?Zw;bbx1dQJ@}}T@JR5&|-a)hp&3ly8t#^Gp
zhwHWy3JT<4%X_8M(b5{5XUHDA1<1t+`>$lRu0N%gQg}>=3~M}16RV$IVW?y|#~cls
z$qMX15vM}MK{8$^zY;N=Ew)alGO@6EX=wW5scv*~{LU76WMb^X@!sF?BJzQ%)$J3<
z;88-c$CpwH3YnERtF4|J2^p=M88J)^5BHEufy>@k>W@Y6<-_)F(+apCkfE^ygQR`!
z$J~f3((cGV0u&c-MYFL0QQ3UKmOj1maQ9U1xUOw%c(loJA0nc++0uW=De5=B2_!Yy
zoi?}ljK|0Q<42Ete~*oYbt;e1`icUl-N5HC=dyv9Po<@APv4ir^lxeMB#iPS#OqHH
zUD`$sJPZ$SjJWO@Te=V!tM)h(^gHpp8VF2zza1Vm<5A$|7UyW|jH&5p))>rKvCWF(
zrSl;f`NgB|>7ufjn3x10*$#b_{WJ1X%}xH8XK@J~6BBEhl*rU%MNiau{`cSzLz4Ux
z-J-0D{lO=)yn*|BC#WV5zR9ZI)o*lMxXZwJ@AYE;H2A^0y4mCuR6MuV+}GBv7ajgo
zHA`PQ*Bez_>3zvs>l+Z?*~fSS$5~(Oz>}8|d|8?2RgE_oH56J_utgNixklWM4tqU$
z|0G7XpwdJ!u&_vHISrcY&eG0}F)%Via`!l-cJl7bU$`87lzMiE9>WnLR%^>Hc!9GM
zClUU(?Si<v#q&yD)A?x0fgXH5@7ib5>g-}`<RxFg^y_QJ-P3jQ#k8V|hX-Eu7~rJr
z^>R7GVjI7TkfQlus*?U8Hr0~bRKvyi=INjq!)AMj2e1Ss`l8qf@0(B|bb|$PFYn5<
z5>czs+0V=;Ee;d8$tv?Z7IUEyA4u7z!O^J6qFd^Eu?{b5|L(DUUe9`Ex6_wpm(L_&
zqGF2uKIwOQ`tT8E+4EfewthE1ga1C~$pkQO3bf%P0884}z0lSigTP7Ew1zNI8BZ$x
zv+ttPhNa~{@e%Qn$5{)6N0h>^A;R)<jrgx)6$-@9RhH)YeKSQz^Z&yO;Hx|>AymC`
za#o;ZG-ke<{dfy6z0n8d*fGQ8v4!jPiNtR&CA>Y{Jg+u!P>7Q;BEl?eEKqzBYQLh&
zs-mL>{#%BYo2BRhT!<kGR7&Tx_|xV(ux}cCP7ST^)zwuio-E=P=IAsxKBOpQeRO-k
z!b_gY!t`KcflAtpz4pa6fWqLwm)iaV`Hl!aAqlXmkB2zX6C##{SUISES>b92M0Fa&
zUoqJSANl>T=kx&s_q$$GFllH$8^OT9xc#2;-ouAb*>YV(9!rohCVGW6z1fcVn~8_A
zc;zpM5u_!|FUvo%)?h(0q;Bx<nJmSwrZ=&TdI;$Nl(n&Kge36npP{BAUwRyGLL*Z)
z0L<_vwrvqd*!g&5N^;~dTNWNxCC})B;0{RJy;yWCo!cyI5i#sbo^D_<2ip$NJ2NW}
zsHs6FnZQEQRorD-5k-G}w=uE4VPk#$V?1Fd+Am-aSV%%gM~ErNP(=7JYQ*BvRx~Mf
z-fTMCyNZ0_f|2cHm!qXVrw_vK+IsjQ`wdYGPbj%ZEzU{FMTlH`*A6*W%GSE`4MY$^
z$TROi1PRhlo&*Pl5R=~G;-M7r_%#6Yo`ghsn}I7?5mcQc$Upz(>P@WoK9o-LcKj=>
z`eH5zH{-i`$VBXi>SKMEmH=<|#?Eg}<3)^3SDrl`gT9I{kwJPPhG+;>Yy!rNM8@ma
z?hJI!(A`_jK5P^sO4WXG-f(z`iyOGGK>D?TCTNE1USPWfFKMOJ^RKxqw7{{TN2Fn5
z1tW+?(mGuAR24{&6Bp#dkL-R7S(DlY_jv%IU=U@X<`3&$;^pP(_(~_)Z}mM8IK}VZ
zb09>{n92f*(ygpA)$-t;fS8H-3%-}X)Ig_aN7;4vhKLFUWW>rPqH``Cc2BPIGrb=L
zkIKCBckQITSt<cOOpHh<%<cN{T~WiYxI8yPEU+Cgqp6FPLEW>6H%`vjSXhvZK!Hq&
zc)}Ic)I2(Au}R@~uXFyMk`Wm+LS8fLeSS_Z-O<`-O`a}V%lLq?Rp*@M&ktix$l1pA
z3|;Q=qL)AT6uzn+_DL_~ZH|9w&7k8A-#&vz%$q2P2D49p)?&157y|TD^Wgz1LKpqb
zfZf&8Uq%{I@q0F>AY5ZcJ#Rc?zz+7D!9gH5$SyVK(go^m#>-l2j+WTw#WrPqHRh5j
z_g8iFcl7(x1v#Y+cbD&;o&IGe$7qw3Y^b(P;<R`NAag;%S<kluj}JHV3-Z(W8}Tmg
z-ilxSz1lW;usQe8@eO3?$*QF7luUXhu=m-6-#SgycIGB^)y$CuY+o+ISuHoCF5bb{
z!T%!N@}OkN!ATO{E4ocC^%;4`FcY=t$N*~;rr+0`s>8SQYiD&&kcJ7yn(W7U(E7OK
zkLeKww?czIGm209(lv1ij{R!!24oH!r>+iGbt!CO<FEOrzyRmGC1$@dW8}=%0!i$g
ziH*)H&wv0?d)~Mf(S-Pr2*bRq=4+-(S<TIt$m{x2&Sf!4`CoQ7SZPswdJR#1f`WEi
zn}quG>Z?6>3?b^Z>+>rP4t0(APqXo%i9@+WL?r1Aju%iFpux2DqgCWqht{6hqUEeD
zkRyVCL;ynlsGILccXw>8=BrnEIOMMo1O_<Rl_6f%`DpKclpKhuVfqB<`MOW=!go=(
z_2HiruY_##3k!Q(m_vhp+^!U_ciehoCx?b)tn3X&lVZ3uf>73Sh5jGoZ1j?w>BRH&
z!&CA{e*VBM(_9k)z5>jOKooOU*e^fSen3mBp`&SQ^XgMkJ_-UPE#7-@Lk6dtw4}&U
zo4?8<j*t1f!CL0FaJaXpkaqh}aqKt3*Vi>aJSf)EE_0t+yqud4{Ae>}EBL<+wl@=F
zM@GQO4Ur=!4*gUYlrw`8i9Fr{+MT~9lq6=Y@HVH*tjFG-MWmmh`&U<258XNF-O$X+
z{yoNWEh%Xo<OL8aNW!Y$Z(2q=J}Kg0|JU;t?E^+bn^%oC`zmRKrlu=24`5Hc@_QYJ
zumk;Aqh8jZ^5Wv!yEVt}<=+mO<2@DFIA^*f6%@)#Yo!5X5?8I69h&ofn^t@=HBLF>
zg&kJ}(F+z<uPABh8?1*WC?A~nnL@IfX7SZzzvcK#cj%U}u#neCaf!RIzhjY)xxgZI
zKhuRO+usk3Ub!?RC`_yF5|2YY3PRIV$IO1*k(HFB(Cy8(tPTpnr~z>Ev=+o2Y}y8-
z@sS6t4@PVTndp4wKR$lg6PBxzhJjG1hzUY_y1MqWGNZ}wC}|J_0tZEQQT|7MX=pb%
zh;85tme$y08~a%-#_Rcm*=6pWiGS1DQjs~jQ@g=F|8t1Fw`cyMCrfEn84W&$(L<#n
zt_l!dL3-6WwSb83R>X-v^QetU#ds*mFIdk5W&cHyS-q&S|M9z3^<C=9L#-lyX(-|T
z2+8hIO{qZ`VpI&!a09JnKu0QSTEmb!GWyepk`D)n8(nX!#@8!@kirkCAxupFcp7zy
zdcB-P)sm`4d6%zt?H~(kzLbF50xJ{)0>+td#?4mR5SMho-QVc3RCd;H1U9u)qV+Ay
zmB96l_mlz@cdHD>>m_^_f%bv%CE(8b{vjK4Ns37eMMV3?y1AL@(#@|&T?68(K%ziL
z@6@A$AJ<^7`qKUl1*x{yi%;Bdo$h86<o(!`6|IuXR$a59Tf7mP(&UKqKwM%|!0sjz
z$C}gvU+s*A&wsbMLm<w`N~{B=nZ@$%ya_Oa70YZ3_J+e4?GCx&H`SJ!13rEFN^wRk
zO;Z*v$wBcs!qf-Sq5-JgW1UU$XSVaHNaNpL{rB`n;9$DD|6aXA7{xzis-X4A#76Hj
z-@!YTzubGj{;9>979i}2<8`U2M1rZ89m`HXUYi~rJ_PrU``YpyVsh1w1270cugwu|
zGW7T{voJ}l?V@8eYsR_FbPzIQ4{4S(0YB8MRZAGr-dmlzy+Y-2-jfZm5*=+ef0G9e
z0s;iYX2#WJ<?_~-IU3)lsFyLC9ugc@);ZEL9YD$U>DFQ)0KNv`jL2T^I~G=epuOa{
zw58vx^K<($(8%M`^CM<B3}^C<Sgn&u=Q~%4dvmWbcT6YQIu<aY!Wy(>6ujsNWU(ck
zi#}wb5fr(fEVGg6<fNP~h<PU%AEnh0VB<6AqpN`viRM|E3fBuJon+OSQr+oks=Ylb
z0sg0B!<VmuU0(>gH%3dYZOwk80+<qb7v)5YS)l-}M_Ldp?0-dH`_@Rzd9-Kk33x4F
zM3bkbQ4l)f5fl`Zyq^j4T2EQ(Wm!h)Q4a8DoLUC$NyHv0HRvi14vHXo^psjq73dI^
zQ<T3u(bE3a<z|#P1LH&Y-Ni>9FZY~ls@=~_m2s-98*tCD)K7lT%~SvC?ib(^s?i?;
zt-HBT4Ycm%s?X%luqX6@{*LOZCkO<je{|RfhQ^ybD-N!1us+mduc^z9ts&BC{@M&+
zb5NTB0=-gi9Nf-%A<ej4`}uR1#@OpuAGi@}2DzDP*XflOJU$96M2D})gfsD_2*I*Y
zwGFuV8Sb_Hm<!luyW2J^!^g|h-6f}b+bq~PGCKC;r7r{5>Ktl=?0br5&-5(JGeG3I
z3$JY9TOWG4Hw*(#L1lh->+VplN|J~XSx6orm$OZtT!}0`Zy>Z`rpeP`2ZQLWt^yKi
z%FD&<-ZC@ZvoX&Y+xe1|OeQZEl=bw9EyRC9x?7HNB2oHo_%~Cw6frZP1{r$vp2?if
zr7}ki{()#47vUM3xZbaSp8eR-*3ekIEb3krs%uKMw)VUL)t#Z#Uz9Iy@1E{N;`pht
zU7duQg+J<`i_?7eL5|i{K;WQUf&?GPt@DP(;jlsqVM^PKT#$_llrvl86$lDV;bGw3
zC&b)e`L#?+0<Ngf3`Z5Ly7hLKE87}*dAmTheK(T1cSM0Cn3)NUwkbWyqN);bn2(H)
z@ODt#2r`s{^04nZi7krqQk|lwCi|yPi?+7@$sc%zUHs?3COEB|CL{fQba)sG8(TGv
zTkq-8(5W#ziqnpQSO<uYb@CWAV*w6F^LBT)j~HIJ+r`^|7Fd8(U)xw)BqhP&8Gws<
znn}24gO1I$002BKIF{HGUcSr(IjGAS2#KXon|>XC8t2NznIa{(&`5u^0dl@ED)QXt
z5fQ2`9R}>&+{1L?RNVz`Hc$-B#5A#UcY>Vz-tHx%9339lu<Vn!rI7>E<nb^l*HUax
z32*^t5J0wyvlTmZdi=5JRv`<x4}+!Or4$w>tj^|4!qfik`-8YR&B&mya7wDxCc^*k
zJE;i=!;;6b95sHJOaw4f{6AIUzyKK3O~i61c!4D*HsIz~-3|!)!0`n?Wz+d}Uwd0m
zLiZ9)#~mptUm$EkLol+0rK`DeK&9r9*P45)ESaah=PSnhiVySnd@CJrc*Vw!fR1RP
zum}1N&$cYL3-3TeF=&bqA3&~d1xqgy0+el|E3h-SA0YWLDKQM<O~0-Zv~xMw58f`;
zsxv-j_wCty!4yl!!eY#68Z)Qnd3krvt`-_UU<%DI&L@XBRB&B6TD?<Jy}o^Ye;L*!
z;ybrwvwxaSC?$$GkR=T*_oclRzD}oH`dIVHgXqqj?^)lzN}GJjhP(N4deSOKB}7%h
zlrsTNtA!#KcU3wa-a<+atJRKkN$9V5@Ze+Nr=YLDTHKT2Df0VjOp@I6W5#CrpwnFC
z76F0R<#Bj~K6O#SKWSsCN`G(n82Oi{pne0AG&DpWGPJ0%BS$pEdd2p97hj}_xsVg<
zJbNZVLR4AKepk-&O@-+}GYWyH%>w|*M<&GeYmSCL!F&5p=QJ#yeJUZ5k<|TfDr`nL
zncmaUA2EiR!#dMMkAL}Gwe;uENVuvJSZu13O?=@E6S~Eip8Zv(-QDF%#Z$NFG=L6O
zBsF-UV?wDaT+t;k=b3<j){5ED{j7O~?k<p*KI~C;arT4ks^t9zn0g>_R$oImYw*6a
z?Q9r#h5i7J9L<Hbd;*UVnZkik&%p1#_lj_`qh0Sjd{g1)hxRl!HbLoMTT=d?)e_H^
zxvbl?gliK^T%C+Cu8qSj{k8O?DV<TI;QKo~F9Ew)mB2&b5s-c)mPohKo5;v`FD+i5
zUrxmoW;}Sm&H5Nfa3)IAK#+`}knk=ZR_g-h{&EisD<edYOPsVlp#7RIR)3c`Sxrq%
znvgm%@%H=MyfJ+AW1}PRBch6ZE7kOv^ldW#>-otVTlDgc&p6&0XvaY!Le)Jr6}^2Y
zYEN|`BpfRAW*}bHU|c(<K~7FhO;#uR+e_fJq-Spp@KFf*LO_sBMlQ5nV}~lb)j+Et
z6D!>__2;rgo&T!^(3<`^F>&S$<{l39!Ouw(IXS9NYvHQ+E*k3#Hz&uOUoKC}LR{3(
zpY&#rsTR>Y!K?Z5RAU1sfHzM6KKEsn%>vKBa{R<}0un+xyYZNyeI|8WPUV@*>mW%*
zOPxn1#8<5ZghVls__JkB?J$-vn6Z9j85hon2Ila%8(M0mg@u`_dY15Q+3E=#(XlVi
z$LchBladf(Vei7LOTWy<a%UDKgWNnLi8oEX3<)?{ySni*2hE%AUjUB*i8o|q$!KVD
z!e<Qrl8?q?5pe~D4=mIJ1HWK62EB@gw%(wSVj?6)1xqtAu*AiEAjGEzig=IXv0j~B
z*VHfrr(khmqkU&voixqsxe%PcB3gJY0rg2CF`sCQ3JNkxZ`Wl;%1LY8NrbWrKAsv1
z5hoO6Vq(&@PZ=V)G<8CyX-I&2W{rdy6S@!wetjy=VPc|5NsgbLEO;cV^F%}XQ&vSq
z%}+0aYj5#yEo*Ad`gp7$8P&!2Xv1ce?=doRxX#zY8I<1OA@WcpwAU2ghNuWE>m#xT
zoQaTHEc^&SP`JJX7=9sLOwcq8AQN+%0xjWR$4W3wdKq%CF$yApt3WjQhoG;Je_ekW
zK!l2FL=7>!XfAc<E<#tBZem)Xn)n|R4Z)4?wg%I2x>WnvaM8`6No^-?2|Gix5cIcA
z=&I17sP8PhxZwLo8FC*|_#M{t-D5e}n}1|m)_^Le#^yzZRoTT(-1RK)Ywz0nuACNH
zx0u^cE8-644ln3IW<5CTohmkH%R+hXQuqXPa{A&JfZKT3gR4qeR*{R3ZPshM@FDrL
zta4Wvnb5=2SQ#He6XmX>=ixoJbjj>{T*)Qg151|9h(YOjMoV8hh?t?U*+qE!n<8Hz
zDJzH|X*}i*205~1GGZ3`N`$w?+-@LorgXUxRA@dx;C8&{LHGHq<Gk|cp~m6)p!I=C
z&-ffAf`%6OH(UxNY)wX!GBj@oCPtefE1>!imnO>Pd|*cYRv~~ZTH?QT71mpg%<kxy
z+_SYXbX#9dU)>q%5UqP#INitv4GyZBXXp9%&DZ)~mI0VI<GH%)nbl;!);7W%A9)>i
z&8n)cujAdIu{$c(2M30L!yhu+8k5j1u=k6(eisHkc^W$ac6OV=gCdRI@&1W_z#vj%
z_2NDaOOF9N-(^{DX&xG)LOCw>&FvacGwy86_lt_MIbNSmm89~cMC1;d!~2N7PjA`2
z4y>wA3h5be@AP!Q#z}4Wr$;?)t^QNx=9fV(*p(-ek|K6=Q4u&WVIxEA4u-E2ky{Cy
zE(W)s!8s)Gg7v*jgUb?02sNn3K10M=e8d%~3XtDYdY6D+j9XQM-yqw8-o<nQ=0V8N
z+24GU?)wT2`Uk|E-FUuWzc%Q9^T|c_ata4&RML=?d?)9E{OIYMuHTPVn5y&2uP(O9
zdrKKC=jiC#^y=g#YoJk!qxBPqQ20v^qPnX|JJb8yAI|eECUDu;&UI#a)9gsI#&0Z<
z?<#i^{O5vVXLs??XYKd)fmg^@WA>`U^2ZAy4%FVGsJOK3-2G1V>!=GY*W<n)v`D^^
z84a2Xn0=LTBI>0Q76Au(Z7n}iS>+6A3ZTCeXeo7>*n9+A?tOw_P{#VFrE~(lZD^ad
z84p5O+XT?M3Exzu7zejOxzkE|wAJ$IDSC-cmDQr16jGY_s;Ho7>AcIRjk^KFvhb1u
z$=tPnYD!3sK=Xk?=n*q7RI+@m71Ixfd+(#4cmbacWXVgVd&q&X*XCEf5U;4b!sC}|
zsSgoQ*V~G`23mYferSbK&};Oo`N2AI=t*UIWJ06)w}4;Av<ScJHs^=Wm6a5M?U0kK
z0Y0#ZX>_wbouugG8plBoX6%Zc+lx`{IDAF=Pqm+knze>F>AVRJ2}u!g4U(gci+P%{
zIu1jsIvp4BgKy;_SiFJJ2{GmHi_`a~oN<49S%G|c^L<$C4VA~z_CsoM?J6q*WOeP8
zYFU~ZRQTF<Cc)PG<7t@co7=pB4zO&&(fzva1)M+3iWTAED#jZpWH6#Bz3-UvKA*Dx
z*#p0O*?3ABNe-f>@uJ+M!wdP*%_g07#0tc_AKLZ)bWsqiudqB9@EmO`-|M=M{^mZq
z3vcKn5_~|ms`ulwKjlE&wn{Q50b<&pQe0Btxq0xYTb>J?F7&hv<}ZblBINR_V6N+@
zoZdUKKLK}M+STG%t=qmH$U2c)$G@YQ#N=-_MZ&yxwsY9D%jWi<Lzz@oR(74$UsJ+h
zV>sygngRjkNl?EJ^}GFP!b)cBu;A5{>AcVxvpe4vjghn6`kYNTKkq<VM8@YDm(^_v
z{xK9ms+a!*<L@Sbup0`amX^>jY$Td*YDG4hF8$<Tj+<)vOUj0Uy`VE*ZLtL-ZD|$j
zcW-QmTUoJ+q-3jCYFJER-eFBb)!<BDPhZ=Ry)fr3(yy`~r?IeLEcf}cV5S1V9}xX$
zKIMB@;FN+46?F??(-?TE5M}=SHB50E<PvkXS@CRIU=XSQKUBR3IF|9-K7JDkMVTRc
zW)>oQkL(!|%HGP}BS{jnH`yy>uTaU}J3AwL@9lT>{l4$}KmHw#qa!`f<93hF=eo}G
zyw0<t!mBz>%t*Zq+Vkp?MG$Br4@Q1+)3UTSohiz?X6Ba;Ky!6)>lUmJ{(mtZar5+p
z@6tn5AdU~WV!;ZPz*$32Z$g0b*4#90U`Iab)n^qLa?wbzeG^R&LZ6C{Ri%ip5>_+7
zF8%7Q#TCkHtLd69X$teVZYc`p?K_rVZ*R{wF;45aA$%!DD|3^Jmn&3J_H-nzvXuK8
zGQ%qd({%6!(gx6mfzu*5*Fc`cbzh^vLpeFQuB19iDV@CB>bH!VP$i^h|7PVu1YO)A
zF<K!@jhG`Y_{~JPovezitCyF5tljr)rvgdIw-^zvodc5#zrrsb3tsLA6r8UWq?J$e
zs51gcyFU31VBm}yAe=v3qz$gr7v}}-n7^b>`}Per%Hj(sd#lra<lbq5OitDF=&}M2
zE33rbM5_6R8F2riT@TFo=!?D78y*n_i<e%V8>7#xl#~>TD0ojnrlxe3PF%dXm1UPf
zZA@<v)BvALNkbI7>$Cp8dJLz&tJj_p^pfdwY6Q5^1}blhZouFJ??W2w8w9aYA$M(W
zAoTPmr?ulvfw#P3BgUr`iEmTcOjSev!|(*JZ`kI`|K3}kA#M()m$r9~p>NcNVZ5!F
zIT73|h1$D!uasQxP~W}6yli({l(FGHI0(6iBlGgBr>M$ZEJ86%l@^_Zy?ZzX`Boqr
zW+|{je5*V+mNOAws`!rs_n!bOJyrd#N~Gf+kL{Ne&=TU`cpFj58HE|t>Z09oa&KUA
z5(~+_M@$jspCY&38T*@-15ZP0UEoth1RA3MSy5T)2*u>nsiFF}4@b#DiqSFA@*wRY
zU2d0^cT&rRm5rI@7A9uc!XN=NLE!3aT3O5NPM<nfNu(S)0tM0J0QhiB+U*3~ADIuM
z93TBX<hlJj9)<}n1zH*CguHg_8<ICD6Gu@*Aw4Xh(bgDRxp8#y)wG;~DP1ADUsp=J
z0r^snhq?HO>&f+-B<U}>y4`>GvHhbtTvsOnS1umk?dW9DvYS*&Qvp+`9oO@w39Xy^
zP5`RW<TbSL-lqb#BTE5>e?aPACh}KF+|u;1M%WUqI&kJg^DW7N);JUHbJd^;0nLfy
z^SKoVp9|N&tp788@t?mw`ml*uX6(4(!<@=y#a^U^IwSD9Fv($UVS$b?Y8?qjH>vX9
z#S40Sd7qXI9nl~ytf(MFU<i(=@ONL%4jb7BV8^WbHEE^!roJam$KTCVFKS5Av$8ZH
zu)WvM$aCcI8j733c6xAl^Q=Q_jGcuX+68{$0t|B1t%2P-R7@~72Gt;!O?E{40*ppB
zNbLMA{oB?g`}7$pf+m~!<Jzm6DcnC+kJG$Bw_kTi`8TWWZ&(vHacgXG1@{o#=U?<H
z9G57`Ms21KhRbX%(0!|axqvM1WDW<F4sTzL5do*%f6)(nfAWed7mr`Z$P;x<eKl>~
z;fie-34r*Wx!VB)rC7+HSC7q;f-+NeVkkm*_lJ6oN<;#}((VL*3~h8gc$z8!c?vk>
z0xU3V`Z))%tixZgW1y3o_VyD|W&TG_Tya~F=&X>8fo@Up<a+s~C91?vE2}k_XCAe?
zn4(;&$gwkHY__eu#r)#j0r2R<JzOGUVm%Nj9#x2@seqIl3VxOt_H)~ZySv!<c#)5|
zhfWJkE)9N6aD2_ma?0+G^85Dm;)Vjl;9#o=ElXgtAtp9<ZwlYWOr+@O*l@Cv8Ea`3
z{)SuFMWc+gsk5bfj9)`hO<sRk)gc)sP6+tEm8F<xW#{j*jJv$v15IdUs(_WRNtib+
zjE=!73>Hjg3W?_xUNlB?@^T%lYzq61*Pp51%)kYCTw&n`013Nt#peWl_zpd+WIFMw
zSHI4O&)<4-6EGjF1jkCEzT66BgPFP18~>t4BPr<^MPxTm|CWKMeLi|IA~L$9B0m3t
zw)eTV1I*Hr$NnCf4iAqngzq#Yv&Adu*P36#h=5895$GZiz_0;rcu2v8<$jKXAes(C
z!y}}uBC`74x8({uQV&CwkOy3kAi;NTc=?o+iDe`Epa7a?>6^Tc&ee@>7x`OvlR&d^
z`mSa6-F4bPGMtFWimF6(1co`0&(OjHE~&4NgGfqA0nn`xMc98k4~#ZgSkmvPs6JIw
zeTqjFu2M=?r^5xoC(<EzO}X8Tfnrs%g-{*(*iLdg6Oz<p_J{YI2f!rngM25XW$_b@
zae!R_gsZ893>1$Qq^HlLyhusjAXwX}bFmg<m+}O0(|DBw3q_buY7C#no8dQZWyD>H
z7jCn)({SbX9ADoRvOjKplm0&U_w;Ks4bqTq(*maZ`FYHp?HyFAGdU(I7Tjx?Pn+JY
z27h=D!4qpM{$!~=s0b3np+m(+|7g$@3;BGn$;Vn}>vS$0#ywzk53cK3y)<*j2M}jB
zI3C79iTd-z0R#pBOi$vTT;!^IFd`pGgw&qN$w}#8x_Lvt3%|aah6aL&agv0H_=)<a
zXaq2sQ8$>KN2(9Nu)zNPNSK@hj3$hH{A}U!7M~6YDuvI|X)~EmG9gtsC^;LIJC_H~
z#x>>rJ@plvP)DKWy}Y8LzJId!Z3UhXoKSW^A+F-?LRU3ae|~y?mjG|}&$61n>Vcl3
zHyI3H!YkTNj&cSk2j&3D-aJ^W(*@xteD@nF`b;YzKNvEx5E1LQ{#@;u^Wj_tP_};S
z=WFlfU0p5?n~H4?-0*>YRGaBLE@*}d>Tu&s*<RQa6iP5dd~nbR?v3{IJresJ<78%9
zxEM0TSOL3wikG}?Z9Pt}9G}Q}U#->NzzhQ{eua_eczkels<Wy$3}$!9@9D{PviS7t
zRDV0Pf8{79U*p3EAw+OsOp2|e;K$(d%CVh;>2&JnrFTTcj6u!oBLN&3Y<vZ;jaTrd
zSq>WGd_eiehfJhZ=lhakpJ$zpk#yMCloy0|2t5uK>)FUgUF+R(5Gb37i>D1ab;QLH
zB+G;ld=YRrW-yUrAum;w&x6E8AZ;=14P}D>46qp}+MM1}Q$zXF&b0*P=S+ihT7K;!
zloG8tn?f$T+mv;<K%>P(Yn@`e8}Etx(6RjZajl@&Wklnr-ei4n*-Pc~zWuNH2R5O@
z%nZZ;0DeR=hRCG8hTp*_75iFjr;z9QdQD^?sDscD$LsUq{_6kKwbed*r7(*#ZZr3|
z6Cg>j<#=o1?BQZ7m1l;CK6(9qEjSSQ>fUtN^(6@yz+%0cg^(OBuvF*0C*Ty`J^OWK
ze>I_&8fzDZ;$R$$J$A5vY;0(Vl}T2*m4;|+f~~ZQg{4@#(zJ6_fg_-qUOBb#d5!gK
z+C|+x{c89W@SOhhs=7EppuJ6&8vpp`iI0y%o!8dG)b}|6n1PTH^e_IGI(q6J8;d!k
zw#?ColPklu!=3uV9tZ*>s7vg6P*PJ0dvE?utH-r9KtZqq+|Mo}6UO=V{ps*9Z^^ow
zI*S`}LW>=sNGftT)})+rAEsBm)Fh9n@ADnNC?F)jTsU3QL7Y{23cpX5%dK^eIZ6V4
zi=U7Uv&gRVfEbDSGOnSj<8<DqAGJve#*O8D$Gy6x?_3BtJ$61C!2DNu<Wksq&DqKY
z**}j<xA{%lsLGZJ-apGf>Kitdz>u-bXb$hfU)*iuMnpTv{!m4R$1p&QId2eLH=6k7
zXF5P=o((}~8--q!3JiO&5(CE07RP&rZMaWNe2#y@cy`Kua#>RZOv@W_T5unt-qG#-
zJ=IxUU2c7JsAw?xL0|UL&#@Bt2jtb#c!0se<%8QDEcui5=XCVH+=@%<Ph40weGoG(
z7eR`~m)##y2S|gP?OkU%p}URMTe|$a*Cle}>i*HMSugMnY1-Ai<{Mb$14Dk`UhK8z
z_iu>e>CCFn$>pPIBMe6Pm?(_5-KKVvl``7e@0H9D-&i|Q090bKuSd4CvK1`r)x@_O
z+Rix0YRQ0Eq^5@V>9WR5U$tcuB_uwkr>7qr+%gZrLd4SU=<rd(Ap)CQS6f%+AbbNG
zTT4?jqma00(sT1%PDAqTh*Uv=)w0%~BMTDjm?AQ6*_fu;?FQ5JNtZZLD#yVU7@&#f
za_MJ|nuzN*O%*&EO%BAt+VzLUvD9xY!fx7Yf4e<+r1A>W<WyV5xY0Vf65u`2&zWED
z{Y^mkeU#`j*O(bhTs(PxZbgmt9W>TQJJ{f)n#Oe(n6+>BjrKyl&;)@|x`seIE>SqJ
z^06lP^z`zxc47hV){NUWLHpRi06!c6!<za#0ys4sjo_<|BZyPwIQy-r$W|C%UD4)Z
zNl%5zpwRvn0`<8S&xXpq^arv^l{Hy}gcEbHaPBfZkW>>-2W8NNgCcAbUApCRy7Je;
z7&V@19koC4JXjP`PUS^KMFo|p>=1<Wpv#Xkv#@k@v}rsY-M)|3Ge4bSro_Gpk3G<O
zAVBbx)7i-_u+y%8S(l!UiW3NBjJbskx-TtbFZRCV-j2Dg?O4Ut4bl_z&Jhr~!JDA7
z6laV$-n;<*G<;7LyzV%hR3YF{T<t??X?k_s@7D95!WB*A1`UVX-c2GxTrly~V&^B9
z5%5li9vGMsZQva$dDO4t#$fn@<wb7X+lLwN3JUTBy>=p`*Fa@*POye~uFJy2jB^7^
zUa^<%&rX}Bn5I;o3z0UBA5_iDbBpb!WoF%REy0{1#mB`7Y3u4hzXy(Cim8^0M7Ni<
zj$zUUCVOFY8kH_0KCeIO$viqbItmphgYdt0XpI@@SwrnOlB3WP8oZ#FcVITBgeLuG
zCo#vLnl%`Azm66V9iw#v6H_sjvwd>tM*X`*FV-74!4S^@PNyEbT{Y>Y&tzq{7s4;l
zP}2;kG11-x9wkN%Ig(hcM0XPH(wLXp3%j0OhrxAFJT3kn|2+2*d3A9pfr`@TG&mkx
zELiHW8%03)v{?INDF)|%oE+Rv(81T^pHs;P)Z6dTKz%m2#L>x;zwcBL%c18hozgUm
z79b61+RT*VX#dFb2QEZ%<roNfvT`I%LC%gshzxEY>hQ?)W!H|ThP5Cf=(fqqO2<kk
zIemWhTVQRQDPn1M^jRs{ShWY$rt}RQjC-5#C$NtERrPgDcE3-5C~mxVaXB(BE83z|
zjH#!yMrE8lx;R*xosHk$cRbqHI8qwb(CkeN=&^r1IlA=*0&(Apr)e^FzLdD$@$8Rx
zFqn(&x}MWxn}Y9Xz3YV%9bO0tU%3^rGV-G2ck#6M`_K>=RxauZrG1AZ{)3jQ%s!>j
zoh<c}Bu!-a3&_;?NbwMj&~<a*5&KamV=AVRjI0DpsrGuXmThh>Q+MK{4b?woxp9wC
z{T}!nI+jmZ*yv3`+MxOX=X*)RFx%;CJ}VLy2bTSB3P!iMxjFmJ4le@5n68%Z_;+_J
zz`{Wx=x%6e{n+yOj}`AGY?DkXOlwa$Bod_-{t^qiUDj4-Dg;Sy^cR9v_SAg>xu;5-
zWyE$Ex0XGY#C!wdSE?U2DJM0^s(xnG<pz&L{kATQFjSF1XtK2>9lfqyL!@!YjKPPi
zTa@ay?eu}NoYJGd(L_F{^nG;`HMwzE1Rz5hSb)}#_r~v}h9CWaQD2M3{f)eWLZ_AS
zZV;iZ^~9LZfG_dx?EDgLsyN@hqO6*g-~CNlI0Se>bj%Ew-F1K7wMf%S_}`JhNE4RX
zJ>g^JJXUAQ2(qv=<o52mam}mVIZ;(lZ{B6iThGg@m_oReMS=UffKEc#-BT9C^Co7n
zyzJr5q3GPN`JL&8D*_@C@HPg|Sjat*jtwgR%>ecw|HQ$a9GQcf;hAF-I#=V39Q@#&
zak9|7_GSG!XZ&RIpx28l;oVdMOSSEsJ_ZIZMh;3sCL4-3k_`11+bbJqVUUri&KO!z
zemtNGOJlEo1Jp^Gs(Jp`5BGKsSK9_o2bZ?B>_nJYSh`b$;}M?@18oiarOT%ATK7l%
zUVY!a+r&5^@wR<$RQ5)<LI`6hxzk*i6lZ+7b-y6@eyP3v))@v5HzybxzPm-jR7Y-a
z?mE=TyOpQW9Iwv!PDVO5>Zt;$2$oRy$1m$pTR|*BZeCvKUH-P__OvM&#EDlZf`qPm
z=dy-|v^wQj)uzUq+a8x;ogFOwudf<}lJ9L~#Xd+#*gV64F_@rcLv{T5`}nPnzr-}D
zZln2NT@a9hmsMdQe+b1x$<_Oix4U_Qk!@!B<QJY6aqC0cKx!BKPfE|R&A?*u8$1TA
zL}BEQzD%7YHJhGc2tHBn#rq&B{qA}s<40)`dYB!c6n-~8_AJe2arzs!(v!~hslnE%
zVG`u-$LzYxFo(Te81Ju%Us>46=x37dWno~~r$M1pVrY3cFO4hX^ypzc@76If5mBV<
zucGM`35;7*c#YM&N%gM+TCebI9OkOS2ZBzACvZW*0vBN3zUGMXI$qGPos7V`+Q(^S
zk1EU~6ksZ;zV>W?YgC^9$vc@AL64Jn`MHlB#(!N|B=?4`nJ2_`F_1(`F$B}seO$12
z15nK7IRi?Gc2Y|C@To$g`zP3FAlKCqWc=7FEFMG~X!}hz?-qTjZV0E&{A|n5&aQ+n
z|A1?VfB!FE9@Zl|yeC$J5q8m&RHMeR?S}MKtH0gj5nfpxU*?OLZf)9mD?*wQ?F2;`
z-L?KHD}VtpyhUExnPp^4%Lpj=V28`@Uo*A93(5qZK$pxObSY1(R!PFOlXh&##0liG
zr7oc##IYQ?t*DTFPW7X378{K%-`3pRnkz@6D|3o|!j2|9xBJeIS^Gwl@FBzMmtZsL
zgT@|PXWCR5ResZf3%@phbr|9g(S_??oxTq$QS}x}^AS~1ef{LAb7*O)Jp{X&yGEs?
z6qR*(*jSO0-c^0LLlZ7b6aI^l!LMua+WYm%^Ur3*=G>N=PKw%+L)|P(At5)xEQ+Z3
zH<(NJ7X;$Qb_#f|?=O9AYdUjoGmr<vU3R?+sFWcRmyjqLE%0qU?&bNJU$QY?UU@m|
zzu7NRQZiL8W33eA_VK=yoaH+-x|`DK&)gTCX1EBj==tbvAiL3+GfxXQHEHtvAkl?g
z|AEJN=4*s1T=x9h8*4Rz8K1jx;qf6@V<jeDGW;iV(QENKeAYOD-HU+pr3luhic)=Z
zBmU+vv*lQu9~(Fkxm29m=)UB#5*1W(2NA^bWF`s*kPfUv;NIMBTdu7=1=<hgdx`pz
ze@X+a7*>|6)8DLS1gdFO+M=SZs<^p%AdwR0xA5P;<>SgmzK747T}y3vz{}M?IDAm&
z#(S*39MWkfOBW$Yqf_B+<*bwvo+c*tvV{9fgXgiQn5UtdjSV>hwrDC$#un=$6se7y
zRpju<WHYpolCFA#M--N)^X1Ui#<N&DI><ya`yeTvw^E)MtUk(~0v>hEwcRA~V|sP2
zDujg4Qk;X|G8jy$FhT6byWBX}aBbE*Sm7_)Y_Z~_k~$2_z@iH#&l-h=GWkjA>Boad
z?Q~e3jc6sFjkdk?hgB5pC)o5$8Y(M`w0dnrZnj)DwPqO3x!CG(7&s1ca%LG&Myjdx
zpDx8V2%-2&XPjc7<rVbL<43%a_^FZt{R7y}H@7erjaBQh64ll5=PVF^!^Xi3{y1LZ
z=H=#=1$WqUaVUD{>TpwO(hZpi<U9>a3!Rq2o33|SNI!MfcUGc6qBJ$qdsDqUoFcoo
zTl4mG#%exhGqA^MsjJu7jS?ZQr{@QLw2aVv2njtmGQ>}*cXe3r+jQ27t8j9<L=kg3
zH~@d;<w#9EFQI|{=hf>j^M}`HCA{Pv_H>UZ8oY@;w_4~IBE8S1n_xT7&(Hrorj3p!
zjO}>}ve0Yh7G85*^jE&0m0}ioo+hUX`W!5b6~L@K2^mS7g}*G_vv2kiZc<tCc~9V4
zuG2!Db%=Xob*!FcpdYT!zY~&|=e#j@|Mm?Y+bMGL#6O#>U-I&1JvK&Q0WW_3IPa|>
zPJ@_}6Xc2sB7n5eSuiup4b*~cbad#K<7<0Rah&h-ogYl#wuY@out;P5uNEM(IGUY}
zMOi@!-M2;hI&<ITy7HY0_~j#$B5;t9o6>TYEpVq15->q#1!yQ4!r#}sA1^Etewq1Z
zH9Hy4m0gMN!%W9LSGD00B$*zWaS;(s=6YUSxeA(e?AStdU*Y=NKxs)RYjA`vKu!(k
z;P<g4Y4qu4Q%>&kW4gCblLesK;bvy=?ryR9@BvGx*2#<V-tOE%9d=IW@3CjspiJ0*
z$5w}YZytNyJl<_sQeB(v+~?bH{nVXO;%DXhWN}f^;$ZW?#Du?VM|G99%QMBi<rslY
z@r}k{Oe}7pBc&}_UK#TgN*&}=bYx<3B>1_&=D^0Lki}0*0nwSW9UxLxcF~fO3c$d<
z2G6d?{I1T4swy2`dqb1YH@BlZ%{$xh*e#EBLRAJn!yhEmA>kp*%(&(;-}dpj`jRJC
zX{n3tm@oTBWm%=p<H28lH&<DG(BPCvb&i6UynA&GjFe#rdtE|=npLb485s$j#A6J!
zTcp2nSImK(NNsQhx8SAC)voy+Q#NgVJ@vJ}v)Q6)lODIX&;KTp#r9fST5|JnrW?(p
zqf)~FBh(_VRrOeTJQly1fM!j2Iz@^;`UmyPzkj36b5TVN<wybjjx@gVuFvF8IDVVS
z*lVDrsRdJQgSkUc=I6#$3wRTQ(K`z>GjWl&fx%2)ommyw8P*4aoUO&~%!bWR6o$aH
zk0x;T_hSpE-yS+%QvNkKi2h?=BS?}++T{}hlQ5cVRYiI=$43_eWA}Q3wYEZ&`Hx0}
zF9QGUxI&JxSrw{3q~RF29Z}QKvo*EUmTirRfK~Co_3TZ`RKu0|d;%nOksyZi_!sQS
zl@E9?Y|f^YQ=1GLZ^aa6tjw((w$2I}G4^H`<Wo5;6J=&*&s5t#)spdO@3_?trv4J*
zIZC8USIbW<bWW=KX`gZ!Fonw=UU{FNbb@0Frk)_g{5b7&+~Nu3^6A;v8DxNtV@G{S
zy7?Cf%G^C_loJK@b#<+J{WA0^Y|mV5EWE%)SMk%;Eu2^Z_n%$qa|2(Jo?{41IK&q1
zoSbhoUW`<^;IhXG_^c!ZH+W|R;({F!=Yu{B->xtBY-|d_9AIor`*n>VUWtS%Dm`BN
z{MjgkMl&j<QpSGL&@w2_%F}uFbUGvP=zW@h?V00P{Z(5*73DN~U{?YUCelWSXTUl#
zFtpw1MNfU-WNcM&>y1REolfuHR|fSb2$B;<zHV%B*_ww9D-yYdujlIO8aN7bx&6Fb
ze(1f&?%MP5K212t(T-g`3Uw<aWpu61%iRK`nn9D$6M6`bFr<X$Gc%8RDYGy!{UaqH
zn`i17EPBOfDI|iHc>ev4=STIW;S3@HZ^3m)J1+4=k6(%=Tq(KUar9~j5?4V>!qTX4
z?b-!MMtnanapfo_3BJA=FG(c-3)x=n4*b})t}o<?vp$W5G-%oB*S}sGPQ-0`cXNHb
zd2r~L>-kWPrRDy}uG{9q7V^O1vNUPZbIcLd@5Sfh*vLf6<Nd$cPAsQK8*DM}+v)I-
z$(pO&{94n+p&-aE2jQWv3%Tpr<^1tOS_oG8TYitv>gd=cPLihP)oSkg^^#`^30MSj
z6121o>^s{H<SCo%o_@&2yxZYV!ZelY*Bc-1NcfnPjAbx%x$bSw4+YJCdKjBiXO*fy
z75ioKg{7gO)oCW?;37P)9b4bo$h1=VmpqQdzE`U@nJU^DNxU{O`RnW94J+FoQ85+{
zro`z8Jo(_svomuPA5RK7(hAA-%f~ChrPHT2BCK9sh%4p2x#bsCPV<&vN4kQSk(O4X
zt5w{*XQmsM*pt-A%3DkYqHP&oJECB(*<B}~vKQl^Y7*_+%V6x8u07iRoZ<EQNuH1B
zr|fqv0qQaq+^c;*uEi9MyRNK;=;`vdSx5ka02@a`O3_S~I&8v==O()oAG?;I!2<ie
z#{%QD8~GXa8{iB@A>_?(v+K$?inBh`+Na_(TUue7&EOrY)R|Iif1#v~@38u-1XzhX
zA0)XJ*%jrUVBZQ}AI-^0LWNtbpdh2(XzL{|Z>+yZRYi}AzADLrkALfYSi<9s%yL#P
zmWlTM%kwNNP`i#+hFqK-&h5B@;G*8~pk+B+B9nB0$Bp2a03qs0erU6M+_wDt5m}YI
zGMs3~wbj3WXL+25zLHgKr5+u4T&^Zp#bV%zLrUc3g*(TyhMu;zS3lM*jB>if#I?0G
zhMHRIl(#nhukTA-gLM&CR%V`Z8x2{6$zG3^g}0eekn@0w44|hz_IN@Qes8*NYPu&o
ztptpyvV4*n`ihm64L$JfU=S{d_Dpa}FD&Kp8PoPy;t^$d=yk>W0*4f~cP=@>SLa*$
zK<mU_o1Y$Xn}Q@VA*O%mS5$Gv6UAQ7z0Hhw?;s&{$((yN%_s*pOZ?-7=1CoOsPXm>
z91k3woNXq4_w_xt{Q1#^Qk>_DPUyb-_3Q2yB=IO>y=IE3+!p8NoI$J>y@g-PztC_r
z_WkHt{1xtnhDt5`kdL3oul+RJP58sWr}$)D9nGn!oX%M<0fdT`h8a|1{X?0s@6T@9
z&$TtE;+p~?QxOLoSMY}Q7te1(qlNGYz)<Aacn*iYyc0LE5;=;mg#%I5@wy^IBnp)f
z?b?sRPHs}0j;;W3ihFsdpZ49Q`(BZcqT<H-O5{>^3IkDa)h7S;>PFy>ku+?-B3Pc+
z5n%A2Juq{JZhVfA{B(V1W?^Qq|G{0w#+xK(Wn099I3+tl>ZWWF`2LHHO+#bBSMJo1
z)#rag7W5$m!&iT*F1>0xWvcEvnow|vJQ~7z(kZ~Zd(Ga(>2jYhsp)UILGI*C#%1Ha
zG|@QDVI{2V_i%_2pxBVlwb_ev(PW6PVMj{~V@H4g`{ud0YFA`=u#5~*VIoK^(ic-U
zfdBH|v51)X2+=GMUZzP%=-@E8|B;m^5yqwZ%~m0oWPQ>T96^F;N*+JttI*Zr$m2Kt
zvA)9eiR83r=1#T}sSId{Lv8|{2-c&&twUoNx)Ox>?$fhzH2&%F_XV{3(`zAy$aYV^
z7hP>HLH<X|9nv(LnPmnUG2}LE#`r^Um1fd2sKaIFTSL(hMi~L{4dr8cy_WqbZB`Y&
zAm)nX-sc;o4@)qSKbwmmCp+MBP4N7()#V;{WiojCH4yjN;=%g#LD7{F-P?`wBp6;Q
zP>O*B@z@GoF)_R`YaG0s-CZo?eTecs9%HI=b5|;E(bIz!KF_w|(v{qzfn6*O;#JUP
zOP=s2IIq&&5kv4>Le@72(G%zkgW4sDI%NMuM@I=xJ{K3Kh3!^R@gvIjp<mjnswfED
zwRsv+$nBu<BeYl%g->_llS2rDBmq*f=E(uEM!TVE-Uk4pMBNAo38B~ro_B%6JZo(=
z#43i?<KfDU@vo+Ck{Sx)YWBvCtejTj5iy^l03hsAP*k~8ZNdm66}~HEPH;a$S{+7b
z{Px)m=O@BRjDHU_58gV_aXi>HZ+(*}`bpTZMACj`RS^+AoA`IRNBh55(X%n|P$2F{
ztJe|e=m2&2nLhWQ-;vZ@*|9%Knjw;{yIXC5^Y=ZRuvU4nH}M5DF_VS*_6<Y0*7pfl
zSQ|9u6`s=|uqC315-Yplg)>=(GCy$agZPkwsPe8lTvV0#<(J{IlE-DYnbq<T-=B9r
zLS#&PKREjz_4p&11JSF*Fp`0SRJ0C{sp^GH_q9Q52qZ^zv?dzrtb^;&(l!(FS%s-0
zS`$4$N{zS3?xocfiQ5Z4zTfcpw3wFtU6PSt5acxdG4QB}U1?f>)4@#9W|WqdpMe5_
z3u4aAjC?VuaD-ZDIm)kL+HOtD8z=bY4NvCVH-gaK$g(clSIf%yGq76~C_$vp+6!4?
zHpN?DHv@YxU+~k~I)0ZO$3$MM{O5XvD32t>#HhIiEn#_Q?+Jyf(5JT8?tNr^a(At!
zd{c%Yg&>yUg9PMr&{zWta+=Ha)yE+7(&4o;6B~<+5G4sY210LQ(j7@h_OB+}urBQX
zB+1?j#RGFk$5ovz#^Xh%>|2*-SfucL|2W4Gvy)freI<ns7RvRQ&mIgWY@7(-xuR2F
zVLS<vyHNYwbw+O~?JY*V8-Rj9ZbbfbdQhl|JNFN?u&59qzJuScKA%%c=cT^k9!MG*
zfExZ*ye#yb<Qq1P$3M9nM-Ddz=RE{!ob(W1;cZJxb6AloUXj>tx-lc5R$W2GY`(3b
zsw&Mf<}<3?2ta(T0qX1;{;YkCyp~-@TNlrxhfq_Rw6ofOns3q?4BN7l)^!Wl_ZPq1
z{U9HH=chB(ZR9iTCTt^ox!=K1XNAkDyh>A3^ThA410D!p?|NNfFAXi=w57@f8Z+t(
z#XWY+$MdjnAlsd(TI&Ni;dflIkgt6A$^{uBpAc}OBeHZf@k9Lw;sP1bhfd|`e!hoC
z1Xxlf-MZSOB+C%AQ>M*&!cRKnN3@vD_6;Y1rhoN7cOEa|<IMvBoo8**n-T?&$`=Dg
zFqovc-~>nUT)W@=%di9}gH#TnITzbFY$iP%BG&&HI#eSnIy&^a#2;{n^_MEA>I~2t
zPWaOOLvMWZ?2qml9-#PR@3t%*p$_|qfBKs@aPa33LdDZ=n}`}EiBh#aJ^Q2D*IpKD
z&Yka{TxczbrLmIw_S5+R<NPixkVP}ugPH(vK2vakw<J_uzmD+hRiuT(ENZ4eyIh@~
zuDrUgeB>ZX<ax24*y^XejF{}j@N0ENGquEo_^IGcvsTm?P_qJ?>_T9s1$Aoa8!e6J
zwf0ks`KzP?x>iDZ<U6UU_aYV=>fId$7mG?u2m1Z%&TXf$gWp5mVf`7^_coO);2hPK
zge`8BGVt+_Ka;^(|K`2st?*o2?paG=-6heYDP!2W1hqv2y<$*%9$D2oSX{&Y!OYx2
zZcQC8iH1De+Y-=O3_uAP0c2-)w>XjxuZ=^*vZJGr>`DY<q#w+%{%unpeIp~Ni1c1<
zZbRA~(2YJqY52B(aY21P?#?pYKf=y_Kat1c911Uf?VIrmaAHV4GDzbOF3Zo8si=}$
zNq3?VLX?)26h&B-)mqDlD%<<+Q|uLi<)caa&bO<pq{j1J^Bp8kk19lz^gh{kBHe3E
zBU5jGMseO%1l^D*)%F#G-9FDKCb%^i8=Jap8oxgH3AM{(qiMC3Ih_zC3}1$pSdkM?
zvA_aLG8;BfKouPUcJ?nUb$ByCc4_!D=(APeEsDob06Aly6zQi|7Rl=tR2F$pf3fiL
z@TuC}Omw<4ksDl;FU}AdAWof;hq<z~N=Bx7EfWh}ZSO<*Jn68rUERS2s&9Jw&oTVK
z#sVUmk`0`XK_OxCvIC=o^<EpOAU~;ekK|_f3?w81vE@{LW)Bj?n=*!mmZxgn(@eFW
zC8P>@=g!-xNYLiG;aZH{k)*;y@ohdy50=#Rp{gLrFL3F#g?#x_U4Bh@`Lb{6YW#0V
z8@g&Yu)lde!!<H7p(l;*n6<C6)3-CVocv+mW?YIF(z5vKuA<_pd#z`K=+m5ekArM&
zR^e}Ufdyqoe+G#Q=egvv#NEIDwZE{Py5_sb=)=qPsK$A(=nF?0yw|t{MrUCYs(xNH
z^cpYk9QyfoBMg2#kxOO0or@B)spm<P&yxV91*i8qTnee?Uo#D7uTCOP?AgOR7*ETp
zGr1oL4Yu`tZ~w8m6S)MEKZ)xHGo>#dh4o&UY{!W_3X??;(!z|GxEs(T0w4)Cs?Y#o
zguQ`1_S|-A3SL^PlUJysH>K5&j^OeVLjeuQKbz3r$W$N{O)ivZtU8Znl%!lXMJx?D
z!;f?za2#k6nFPj@fjyg~f&PA5b7R&xt_(mYiE@UP_iInxx#<FE!c~|oi&L)H$<;6n
z|8o=nD3|XAd}Di2@0E=nIe4Vs-bPF4?r9dW%=Y79dRh7#GU5REZAdGbnmTenSjIw*
zD1>bdIZr%)Q0B0+K5I~FSV9F<VPhZn1|C+x3Az+@Y7@$5nJRXsYo|z@feKs=Sct64
zFL}~bnA+yXw?_2RRJ4I}-mFUQ96f)>9@Vga#a0(0chn4u17Hw~ryDSkJiG<n8N5P%
zT{Wf!m`}i^6xlmI#|V4I&;Ky~OVHqwa+2_sN0qsGL4Y9gXYYh2e7Py7Z}J<e`N&_s
z-xH`{YWv(<Ox4S)X-~^8t*PbyavDGY@^pAa!6EC<-JOrFL!6@sBoM(M)EVxzG@ToN
z{ZpjY7zYaKa1lZ8)Aa00VAJ$!yggt$-13*`(=jIhZp^ud{Q;14FRyD6$~=MomHDFV
z;!x-WwMUWO_9`e!5%NU-^jye5{n$EB&hKow^Wh+Gg~TUoZ{iD(?e1>XDBr!VTEW7E
z?~xF@aF8tBsjG+A#oOtTs{K&G@S%N$a9KJ6EHYBEze`ns*>Vn<mHa-UM1MjT0Ra&}
zEL)G&81U@xR@q`FD84loqZ|kjPZOp37$Ps(z$g21X?^+gmoFN6>SwzLfT`I-@ESle
z)Na550aPCJ<bE3?I}Nib1ihMZ=ts|>($!L@(!$^{;5akh$7tQms+kQ>6yy|p+vH^B
zFtNi>ku=d8i22JCfYg9&fKR-7`RAP83N^K1U34ZN5jD|tGSPe#!5tM^$#>*){<fDm
zDki$|Y#TEnXD(Fm3nYXyFqF|ckb*<S$S^hn_q7DBOH+(q=Tp1<!h*=7q#I}+0%VBm
z{&AitRrM03?*Kg>^y;!D&42CJ$Vfpvg4K;5n;DH6u`f!8N4xXqRSi5z$|?1z@he4k
zTbszMm5z|(UklJnC;G6`!Aj5cB4j3{4v@a|bo-V3kTvg#lcsPrH$e*0KjMA}QvE!|
z(JJ9n-+LYG7=yakGC|U7b1Nemx?0O%Rb?;r@%@MHu8uoc0X|8!Uzfi3AHOx$AO=wM
zd5Jclw(P4|d8D;&1fH$RPpqHMO-ix!>Q|@d<}S}J4ivj6BcI+kez0}4_EfU<Hg>$n
zDIh-b1Ykx$<h&uK!h}AHJ!O;4do0?V8`TeNogh}XpW~6cg1M~y<9osee_YnsnQz5E
zYLBObiGhqej%U*nJb_Jr?Qze&4dFHd=m7ci!tNRZjUeFkqJOxC>`I?nhxeDzwkYb>
zzg!Cw+V{b?miY&o-MzWE^`mvdYX5y2yjog0?g<qh+$ePgWsZy)lq``-S{D^QGr$Kb
zq}HKjee*i}Bpqqy&FTKD59aQAUxqieD$z%_-D7td+Ww)0bGXa$tn!sD&>rCPJcHpQ
z%@PQNv%CbM<BlWOqle7Q>=#bI(2%dI)@}eZpP(1%lrRC82r$|YN{JIxh-Z6N`r<J5
zR?gK^e;UBOwai+f*9Is}jVUS%dnt*}=cKuy3K&e5n69g*y(}#EyOa3akH))g>To04
zjms;;^~f0MvS5DX#MQ&e7~k*&3UUkcTw=o5Loz{Wd3B?mH)Bie`a<q=tN{v}qd8_(
zMf%X64lc4j)zH}AIlOsEFz!dsk9@E0Z&FU)`{z0wS|11yTl{?B^JH?N$>$6qWKIB5
z#kH3K?5e`#X#W*hWVN$qulcn7)mbj!uUnP%KgM$h%V`>y{PyxV+O`a}$xJI6X*Ae^
zRYWOqk6vN=EEeSh7M8#f1<tLo2%<O6WF@P{;A?x|T)xIPj`O9pr?|bs^PgY))!)<6
zgtbzNn7&oSI82Z6(6t$@upN$|XW^YZ@VU3WfF9NgBb13FMSUX!k{(=sU7SxQ!TJ$F
z9(*>~f~5*f?>uhZML{GmaDRXEj{a78=LQ%oJ*%<ZTXAUc=H2AfVO6|!FvgW#Zl%7*
z_BT5)F*qf~*2HAk_vQT1a(*XquYlX(fSZ8`I(h)`99!eHuOY<N1D9OjywyTD)y_D#
z(fEfoqN1K*H&G5c;|!J&8|g#GWY+F10bQS9TfO?Ed+QPEgRP435*cdsf>Xj?I5Oz~
zIhz_y@o}-3O-Bs7*YX4N=|OO;zy$kvaAZ?g9zJZdiQGmgoTuA|yPsCy{C#0-I=8ua
zk2savbbu8BZ>oB?Ra^v`6nO=u3m_auXXnwy(<Eu)`D`L0Gra21-K*<Bu>d9@xTFv7
zApANBRV1-`F`Au3KpEG&?+GtBREsiE-F^=WWJ?#Vm720_nDhU9u?Q*F0IYzKv=q%9
zKheAVt}jIU$|@3_A%8?_&vbgvCkb#5@4X&CEz+&%BTT8SW28+?vKs0=JdcmvHq$lf
z@GQpCYZ^vv-$_Ys@wqxqWLE&J@d1Vx6nmdnx(_bjk+^khKZ^|Q-}!3K4*`<wM8x~&
z-&fI^QKN;=ck6i`sTDnMbikM5Kt+5b%{Y<;J%YfJ$CG?{`GpJ(9*17N+vwH9a8YaH
z2jvwN*$UQ-%mwjz?Gfl@z53ty&HClA0$6X~&`Tn34I@E8uALou9qx7qhNXH0z8!m}
z3%P+mMX1c-2j_%Rd@X-)h9cD%;$+4g?Y=hM^gV~8oxol~f;PG6!<^^Q>a3BerJA;e
zpu;7n5p9#$wKH@fj{rw*=+dAc{b|xTaYUJZHZa_uBKVS<SzSuXZTq_cNJdz0rCcyC
zH+=Z8+b4Db1f1;@41bA<x}>r%k13{<{23YE6F4Hr3HumcX$fQu^1^EBWcLK$ewr8%
z65`ajf0x;DpTHP}|Mj{oh%S%^W-U=|xxXQ)r~&tKvVGv*g>mO)!uH7<P$aMw9v+EZ
zt=v8ZW$ER`S*WoMc`U`hbuI3SdT4k!MX2_oOm~N4XW^v4Uq&f?9u|@Evp2`Q5wQk$
z$;mi~g2~rylJBkekr}&5nLxNsXeANWzKYtjl;p&m;sXB$n+_Q|Jiw^IKOaa=J#uno
z7h=4iC<7G@6%A`c<KH!r=m=0O1kq6mi3y+Met{B%mP>&8et-hQ*u>c5-%d;m`6_fD
zK7@9H$f7@Ys@7So|9#e1zNF>2V>o<f&HkU0T|F&W2p(SK8W~CGVP#tYHIHz3KhSCA
zGVG_~;Y^&&m7l*tb_1+va?#}Q4LWXA<n(ZLdM6YNV`g8u0RYYP(Gbi#!#K7N?X}YT
zT|F<?swm@FE30zqvI2BKI|;lvTxMOX0_C(>n(*qxDn_h)()0TosC`X2*dKK7kj2Ku
zCmRiq4UDxzBH%Mszk$NF*uQf@T*O-<Hw8WBz6M81$Rt22HnjGDn4f;_H7tWmn}rSz
zQJ$wTJM>T8lT!eA!WTPpcxWN^N4n+F@bDwxI{=9X*4{=jB$SmCg^G)XQHU#8mM*Vg
z5Ej#@=muNadnDwH1Xut9=q247pQ=SiI7>@GS@nfq=PntEw|qXz=d!GTz&d&P69t9)
ztd}&Y`V>WyG!Oac3$w>Mr9XTq#Vggfpk^!uqB?5m-_PQVUtc%3Gn%ULpWB&LXF~SZ
z7Zn+<R`9K5&EKlf?XJxAk8{B|KSW%>v<V9f3jkt-Y?UzE5ur9<^M$`3l#BCt5Y&A0
z)=iLVe!RI4YLbeHkEf^az*L_nF(Xs5tgHl@e#@zMVbm`TgEBov*|EC&|CQsAgrT9S
zd3LzT?h{F6!G_tM=cq{t<@iX>w9h#!UJyWOd4*XdB;-I2{$1I8b0}<QZS0A4%g&ga
zgpAbH!>!HI1HeJy0O_yVtmHO@u(Fgdx9~uj0m2_Rx9gkpifNy?O6>tm>+kw*Tsn>J
zGq-ng&3mQExU|e+FEf%5f!!}arsakd4Hv!<rCl!$dl@>an6;;&j$0OV=ze}Fbp>1y
z>(>loU%nbQ2*GHbkiU$XErse9q}+f){<|x?2)bK<$_HQSP-h-~{&D1X?5CkNb9?7<
zi8*S5p!9SjC~wWVpN+Cp40K2Q1v_pEJR&hI%@e2=V9RD;=xpkaFUki^0+?q#*V8ky
zGn=1VfG3NBEY;jN*UrK=CN9EsYD`T}k&T&WdUEL3@3ub<u2E?XDR8Hz$?h-Oyt?K;
zyEtcVWfL815SNr-`qK8DGv>n139w76mX5F__2Zkt{Wsa*jKXKOb9~<SwGF6WLc+iL
zLbVpTRruF)YqNRi*(4-FHdp7+5MVE1F63DPDJ)>1<_l=fV%x>Xd^;^vhl0d*h6Ks8
zva``fzIS)8)vFiGqPAa92l|1O0(>K_zX*UD=gB<eg)g);Gv?*z-dvvv6s2o$BU7Lf
z?WGXzkduR&i!gd7sk#9@l>_)pCT4bsmx1$Ex_8aLfB&$Fu|)fP&9}kH>A($LYJl?T
zzrl+{I2`~j?jP)bs~0rhf`)^hAqS?t)&H$DHoz2TS}V#bDD`&q509)))L{g60%t9r
zp?0>v3FE0c6P6K~1-Y)(b#PcQF`C+O@v!Hiikd!E0)HGcc?W`RUGUY-%FN0#Q)419
zlkpgH%S$g^ZOT!B0#oeTDY({v5O4DNy{w&-zH6GAs?uVguO9Wib6NgJv}THw+JEQ&
zF92tNvS$7uE|d#)0(#P0HdG<7z+QR1NvU@~+zkoNxR^sP!;e&tEgH|w%DeNbcgrp~
z@9hsy5s}XsKN^(&fgjlcI*{c_g*=fI)A+YIi~M-OO-MwzHB<4|j28{oYyJ;)1(93B
zWFTHDo<RpUVesGpyPlWmw-$YHzGVETQ2hr5V)E-fI@bKtYl{y5^2BJ7es^Sh+~9Uq
z(vap?l)pN^G4|7h(ElVAfdl^cga7-+JrUdxY4V5rqp4F^Hy1hlj;34w3(EY@=boIr
zAiwp$FaFPax_UAGf1d!~^cL3spNjc^U%9vU<G&~9|2`}(759Jfp8q>!&#$5X`-cA>
z&f=246ZijjaR0t3lG-RHBSCX{RlGa~Y7>wKUP-nC7TT~BIx+UFN>)oPy-=I;>gP%Z
zu@WWBOr11(HqM+WOid+u^^!9>A2VwbM#JgS6Li>z*A&FCoyTr%>UKdXxzRx{Qb$Ps
zMmp;>MPPIJ#psUwVpm+L<=7jaa?SjFX27<pDoP@<s~-!%3yKOdLU8sSoyOS~{-oHb
zGlWJ9EkDL@34Q+9CHq{^(y?}{W_~#pHDF+l6@-vNE|cBOdGpaqshr*yt64rqyqn{-
z|9D161v)$cSa{7}G;T5f80+`GDq?FeyL#+!{;51SwHYQ{L5+FtFyZ@uC#hKUa+bwH
zW3|fyzx5+&_Loe$+c$u(PU2sh=<9pm!)_qhJKbXnz0KA0K`_R@d|WOA1;=R;Jp^;M
zVv#(w?gL`UL_NFb*{{V{&y33+@P;U7UcFL>Eu|%tw9>0LZ6pUx{H3|Ju@P;C)ah>X
zp3syY4y6F3bHKo^!3q<o2%F9-Q)^z)({ZOybqUqE=n48%mW2ChvwH8GUaagjfGB9H
zz|v|>qKSqc?+Hd?xoNYZE?L1$SnA^5u_)DfblsUg_~n;hZSR2KZY0NkcelXO@-j#c
zj=|Y>iYie}PrdBL%Ci$KzVQjAga@W;Jr{+_&_uwx1zQWzN^H-)iBbn2O(!y!@Z?CD
ztoxVvE{rCSQ*J^?2s)_&0vf*;W=73Rd;KV3y>?Z%K~Fk)jymQangP6NMj^cjZe;Vg
z!=K!slG(KS3Ej>VNgB*s!Cwjswp_FLYS14&a$>_f+h5&WoZAO!l1N_&K%1792Bu^s
z+MXaN0wGLhwh3s>oEGfqNjHwSJPaC6Uab|0fFQFd@4}Dg{4VH5|6-%zQlPgpkdl#<
zj&1b*hk^U=RlNKVYI>Qpg5MefAt$H*5*l+lKcid>4KG{qZcs(GuUhWITKkCeE+hvb
zeZP!%Tm>Bo!(~BgphT^1zlWU}lt4;~sAw738CvIl`H~bLP3u!(H{4zd$T{4C8X79V
zhrDh4+6E`sLEKOdb{_1dp1Ar~^vJ73U33I!W;9eN`Kzkso&=e`GzG<&iL!--C9H8#
zQRdk#(k6t_0|#aK`LNV*MaxOmgTOF+#9x%k+`<z14r)455(6i0c<40p(}N?ufE_=7
zLxYI2=sGz)qKIvswa?<y%_;vAD=buBFE}}s1kE{f#l=BQL(1l+O=4)W;Mu~76~0ju
z%pfboKUf?XPPAuNRZ;m|kpETt3&>*Hg&zU<h<aPwx>D0jK20Sq)_}>wr6<(cg#wL$
zFd9L^j+~~HGSt}Ex*+?DTLfY1H>cyw%L~fOOM{r|PU8~n;1BJ_rI5lS%L-XZ-97Dt
zBO_|CY)=J1DX*U14^cM>aSBo5G~r;=WK$>r$rQ9>1$l*$HkwbJi$tlS>d~>{C#f=o
z)iqVAZ)4t+#y=hW-KL@!qObjY0Ep+A5+D&k^)jB+OcQ=L3mnYzesgms0Ct~OqUt6E
zqO|&Qc7PX2HC01@%j%*cU=Am*-ZY*cnVtPkn>D6h07y|#H|=+^Hi-76zXV%12S#it
zmm?DyJ+=^#J}6|UvM}cx2dLw~y$2F6g(P7Eo;yz7_>s-VrO0O~8Q5Eaa9v@O=MWo<
zQ(&mg*MlwxQl7DNL)7uYf7g`J`-!&ov`0Eu%hG{-jR9{Wj$1`b#O?9}^`Mf|BmT(d
z%JK&+#kmWws-Gsa?O|ipJWtWEE_JZibxbBr-dtUm&F~`GwdRp|x^c!<@{pGc#oVZ}
zy}cVZ>|;FV>&dhX#)&N~Hbq@B7AdJ_mQg!@NmOKchq@%Mzz&Gr7$Q?Z^xTON(VT(=
zQnE+foN{uFxqdKd7(kO&6v`9ZS&&}@t-Ow&E{o=ms-Wz5@5&-2OrPS^T1n$a&U}f8
zloeG~Er|?p1O2y@N%XoV$jhK2p*Q)viw%ZW|J@WW!?x%7Nf|K&kjEZ7&%DKE%VO<7
z($Eit;O!g9Y~38GA}9#F(GS-a`3~sBOvBX;1a$iXRbNK?`-y^^KW1k`vqRH~GvnX`
zH}~YXy%!_<?Z~I3ga*f9>6L{1M^kl9Bz%tGfl_rH_Z$p3f*b9x6cxL&8B7`F9m2Jr
zYdoh>WV{oEab#P?&D+hwgcsaQ&@mcH3R@^d+(SXJva+&W{jH*LLo|2^vN@)_ZvR=?
z>{!}^0BN6z23}SgQUtac&f`%7S-OEdBMg*3W1YObTY?B;LQfwBLjP*Zi;=i$G>uD8
z*<3%|vH}-2xAWy$32$`Y_qlQSX6!x5d|+?m0dNCTV*UrY?|O}Yt$4CBO+42&S-ZON
zBGf}u1YM?%6lB~bR8@Dn0y>pb8n64#MU;Z&Ev>}0lh|s0fg=3|Suwx#;+%=Ht5=!i
zh%X==%WEqUPw1e;9r($5pPqcoUm=;hNV#}(oPV4vTj{QV<>P-!e4(v13<Piq-AZ$P
zIm_D6&=}N!Qn2DoNO*=UPnAu)eND@u-_YQ32J6c8#P0Zqc6lHL4YAUKP@gevr(GY&
zrRi?&X|bHx^aoyEPOcmV0>POrHYQp?U_4dOV`U|H6e8h_U0r8zn4@crI^D~2Q-ocD
zRG83wL11t-P!&t6_lS|#2Sl>VMBgU${_UrpliYds3%e)tpq6d$K5M`WaelSMbIYkB
zV!$55gRC(T(01=}+an^PXxTHFjzX)Y3reBaE)I)(C${WLx-JwqFoO}__(8^7<U^_A
z*;LeSC#T6Xg&v5griSEw_)P!<h2A8%10IaAMl0-ih3PV{NEIej-*&D>S02Sg!6L0N
zTUf|}5_W3b_;d$%5a<XEji;Dss196%;HFE*kcIgEJFH&FnsbL+yh1a_0~dAkRK)gZ
z6QHvL7@i<v*=Aq}ZkqidNdsvqIL5so?W+#XIs4QfdF6?TZ{d8myMy!ujHA8BGhyN(
z52_SJMTNYMftK1mzbW1R&pZz~Cu`4MxJ$<*0e!={s!c0#!6tAdAJXGJ(^v+vd%u1|
zduP0~Soe4g9MTJ)SD_;eAIbPW8&LA=(Q`DuI!Tzxlq%HbZ!|k9!YldJ{(}P9ob<W7
zzfOqs<qk=lRmwLno)%(&{0$MM!Nc8}0UOi^Efw%tZ(E3Qb^u|spv&BLK&STxGiT^k
zgYuK0>j(fUqW3*;DHIiZ(R?SoiJjN^xNQcim6BPtSr0*p*hHIw0*~R7dFflAgRLlk
zP+XKZ_SJ+qwo{keC#Etc8@|%~E&+v^s&noi?xzs4FvYfsVuc*~^0Knmj$g^Q<tdTJ
z-luOg9v|#GhEhyI*akVd^@Z)U1U^-OIQSAq*5p}$6uv%L@lPZ%7#3I@y#~MY|5m|Q
zK6~^02}AogJp>Z{Jt{#Y6ZRoUqSG2_V_|w$YB@hP9|%KF&Kp<c$fu=%H*bDI3DIo0
z6ZfPHh$|xu7f@<$RNr^F`;j3{MV29Q#ql&3ldIDD`NrSbL+ERI4H_<<4p(iSo=ts&
zd4=n@jvF4_BG0i)O$o)RG-BwPdsPEE0kbfl*Xjd<gL-9#J0P|N1OfT;J8f#uOj*GD
zlm>wf2LNA_0`Or?2du6q`4QoYxl&RM)9o8K_YW#Jgqd{FBb&twmbJu=Oa_ams~rwn
zl0mXM>2{lH0u#!Tsc-g5%F0!xU?Sj6-pAYcStaz%b;Pr{oy*Tc@ulkPm#Po{5&AA#
zOPj)a;}?#Ug!sH}i>L)XF(62fP$UkLocbmNB0ZQz9BdyruiYYn|AKatnEgx-wAfeK
zTj&OjE<dOT&Y^~BEYz)ey|(#2R~va;?sAr&g<ZMHk03QC=!@Ceu#_Y{bg^)gqVM1!
z2x4XHs@&pRSY7{WKK*6ZD2H9=aUCf1`nH)|w9FVHD-1VG(^MYY?r%%&nNgF&k?sIt
zU?5b}Xg?K01f1EZ?WukppVPo&J7pIOLs@OQ;O5PlDi@$85m<K9a|-qrnyrvHE-U5`
zx#T+6N2$G4fG5v(^}_%=&qa<gb!595d5AVEO!J1fbq`}!IG@b9fjUHrKVfjm6W!A<
zK5qy7n2fP!U)9K3r(m30h$TiQ!gEragKNj8S5XpBJP3aBJ$kh~9;e0_9=c=uFd+R_
zN|M$m3AA0-Gv%i1QG7^Wl``%JRFjc6e?ikDA>q~Yu^mc)3g<lZ6pEtQ3)>3WFU58+
z1%4gj=H^bnrjro&rS|FOvtEO5Z0v04zDpc4a^3t50WsMvZ}_-k3QyP`-j{k3bhv#u
z47iJ<CD^(a&v=XO4J=?9L4tyiO^+d;6lQ?<P)XR<NyR5H92D|dSv5m(NvZdAm_PKo
zn$Q9vk(=N&VdI3>xvg^i<B`@0^}`4uj8NOL>V&3Q3IR7I9<R+Wd2-yOc!@p9tn4X#
zM0W@Q%Q%#u(Lb!@$zbe;ftt;to}5r}^ZrT@dr^dKExZ^ls+%k43nnLD?>T?1lAk64
z$P?sh92CUr)q9_fjaLoitSr3X1=tcqaX>vpyq6T2;euP@o9js}?1zJc@%1m70|sR(
zD#{l*bjX)|21mx3`5s;9Y68ByGTA;O;(6|Q_3}lqI)yE}0|@<#?T(#98^P0Ts^&<A
zj!5|)?|9cQ*UN}uh>sOD*gE~ob*H~%_rUyryUMPsqjT5lnY{*YpnQQ@*K2T{27!IC
zzTnEbIWFQU+*n;rfK&_W1MTT8c8619+tEc;Rkf$5du_7)XLs{`E&&*#YyUUs6u;2a
zEr1KS7A3myHDB9lefxZ(8*j;Ry>^YNACCpfw0ySP1-ejeSd<udbbP$_1B{p9)2icV
z06)Z@La`hrOXp5-d{YN2GutdTKli*|5m=y*kD=tl_q-f_^@&}ErJldCprcVOi6^@A
z#_L<Xa6Kl!`>RoZ88KO--D~fq=6)@dgTpD(*3AN%@cjH;{$HI<g_TA6&!(U-`X)Hp
zS0YLUb@pV#k;*B7iXArEoAWo~()eXxzP;hN&HYdZ3kwU`=3}*oehaE6KphcL3BAJk
zA0JI1vO?M58wrmqa_X?KO%1a4wt&LKDl=D&j}0lIc|mu=dRaJ%w_mwfRDH+gxF+0y
zBYBmrNXl49Pqoa)x%5Q7H)=-Shjw+<(wXNeOl`{o)xmXeaG;?05&2wH)DW3<vQ2#r
z`qkAnve1;rqeX9|b#$H>Kw4e#XDR;F-o~Y?y9n68)<11|ETf!cQ^<>n87z7!Or3V8
zw8zQ@M7^WX{dW2MwpMy2dsmLPZ<LGcQP78=D|;6U643UHZv^C9;D^gb{Xe3<GAgTe
zYkLtQN-7~C4blQiNOuSdQi3!R(jC$WiiC7Ys-(2Ev`BXeN_QjO`OSU4v(NkdVmOAx
zT6fHO%`2c+!pC&e(F3zShlaEWAE}zj@-7W<s5*h>QBEo)<?c_~w1{P89SIdIZF(tS
z8olajFn$OJYp{Ym=tS7hxV@^~TNv-bYDM*x`={mA1s1_7c)R%z0NM=&Apc!lGJVk>
zKw%yD0e?93V&Jrd%nY^AvE0*BRh|6R94JAvys+eIPPzMk%kYKwL|8!AVv2~*N$|8a
zpGNkSa5WC;udDDTk5;7=4GsNiQ%@2?r`D+v)1F|Cyb^FA#Bp{u#~^gR=Eldwa#`B=
zbo?E_h~)1-4&7Chc>46IuC6t>(GIEtoVOyzc3WXyFtFt+@aAf&&rDd)x@Xa3S;bmL
zdcncrMK;xai<KD^^52DGZ7^~021f0^n%a{Is5AarTgP#v%4V|Y?bKYLErztv-rutP
zeAnJXfQN@iKq0H?1zSaM@B@fSlB56n5ZpE{b?QI%-9dnOCD%x!KF<GVbYr6WOZ^ub
z#!9glL5CUw=N)?*L)jnrc(@9SY*$8m$lbmpNQxrWu>#d<R7p+{a~qG6d=Uy=MwgMf
zWeY8r5)1Fuw<Trb>RcA2J4smm$JOnOd;AQc@|mb&uZ_DwyTtf`qRM4>(9e%{I*%{X
zJp8U1HyLt%_Qh~Dm&K}AgKOUIQTvc=C?t4=EcZo;#4p5f1FlDQyoprP76Y5*c0}M?
zpT~mvG8z>9MV0Y2tf$c(=d`>|>aj8j{StQ-$!}!=oiSGCg4K(GI3)T~C=ihzkGyf(
z1W^@n+v~@Ear5ifBLXPT9&p`8kQFC)P_|Tzq-eha)6f4lDt#c(Cj*Ayn}+>cAuTXJ
zmTXD~t_~pJ?g!g&jzKiQUv}kZ#{K=<?Gzq(l?f<bi6i~bu9K#;_xbhAvaom#6MVsI
zK>GXpA>O@m<<mrwJT4f+;Dm+(*#`z5!qC`FztMH5ZvnBdUcGmC*t?!}O)N)MfX2Um
zJcfP+n4l?zB@P{aR6_%v9AHQ=sEC0;pN)}zgB@$&im5f2kpE5Ea$f@Fr-<6CqxpYm
zu@f+grLs|f;U?{5_4D(a3$S?&QRuuKrSi;OQgXhjO$pU6)kgyqj0>MCuGSuzu@=`|
z4=*>*yM4~g4T9cK`M={eZK)JVS-pB})6rN3g+70J=iVDSeh&`TXH~Z${&FgVfmls?
zeJf#4irJbgdP{FBE@?EM{$m*)kyeIdRjugcyAqXkOCmzTjIlB*N!rLV5-ET_ejAv5
z=D1LO{5a=Ju_R5}RooO%`~8gi|9#+`CFh@u2tILq6fmC{i&cnY`}lXk;K=f-9_B_s
zpp$+m-Y6c5z5nzB&hYRsz<3BXuSigf>Cx)&<kX0ava<c(&K3X=d3YbhfbY$@8N520
zhQsT=;XNW))@kVli-5P!F8S3b0Nf2)BDp)f+$;w{S3O<J;a9QV5GJ%X)`Ir#ri-)7
zuirz`s~}Z7J#|m*(f|1#kMxn*O#H`34kHc{k#f}uoB9i|A>S6$`K%7@c@Uwq;|H~X
zO5!4#%=KeJ;>J{cPIM+eP=TeC_<jeQvVPfG64a;ws?yk))}~GGO5dy`Mu2P!_a`gK
zB5_~axS|ZHJX<St<l0RWU`8WiHiu7c1;0f;UkdlpZg=bfy>EYO{A?$>h=P)mRQTl|
z>IJ&245{&Nru<`FOIv@LzQ`TxzVzRq5P1EV&L6OIe)r~Q``PcRMiYrbBC+wP9^f-q
zJkr@*cWu<6RLT){%z%{z-f$r7b)7stA067%YY6;RA}RwdM_5H%xjG~AEN#I3U|)O{
z*tx)BGLpd-_Zic2)vI_t6|TNxBtZ7%TPO_`CFw&I7GCbL0G_VMGj%PQETub2^qEjk
zR11t1V3l$5^>DCRG|br!WqEVb3XJp&kBaHzA#zf4Fgc~8HR`v%>et1BS1It3AJXUG
zER!qB{`~oy`w1=JOt^X>|ASAjI`PH^!1MO$_~FJxiT7HDy((iU8bX}pwep8~;kaqZ
z4)`=l4~;q}*W-~H&>PnrQO)kWhOC+KVz-OT(za)*d+~)Gt?25{pSzR3#zRzw9*TJ5
z#;baS+voz_pY~!8@L=Sd-KZ9!SUf;x78Yewrr)2+;7|#<E~$Zg0~|e!mwYG&bNeS&
zxXXWQ!U@{KY8@_(^}JEQ?YX^EhU+Pe)x#-8OFhP0kDHC1d8%`B*EZ2)#|jpW<o<wy
zv+yhx?kkB%vjLE(d~<7JS?^7m>7w-^P3mE0yFXgI9C4ne*MOy0Yw+CfN#I@njnVp!
z!JpVbD|ZI#Q$fqIjOGQ9Z76n`XDi+510gAhb}hc<4Xx%v3_z@^NTeM6KH#4kT34j^
z2%-b^>fL;f#KmL<^y3o8s?*#0;2SdEyFc(uUkOC6S6htzIa}qzRc>uQ%Og0@=EE&t
zxh3r6QYQHQ)%u9;bqFGg@AA0dJ#KH*efO@#sd~*#zX3#LDHMMruTrwIs&7BNaZOaj
zVa`h4LI5InrJ>DjZk#e*FT5CJDLkI6C@rt|K6#?3_og)cqAj8yh-)t7#QQS=8+G9G
zsaJ2`ncX*=E{k1fe}X-h>IqZ{<eT~AO>Ik7aKu-CQE71_qy?O9B^zUt5<gOjyh{}-
zY54a2ZDFqa+C=7dI|+0-L<&o2zCfgmm63LJ-EI#8lU1+xSC8BcG#If;17rfC(<g!3
z@v9c}(Ec@*xF-kk@Q*OhIh%>tNUn9+y3fTwJUlu(`F?A=VSTdH1I*04^1d}vWarn;
z0iM%_w-p)~#`EiUNYK*=42?0bnfBSbanOjou&Mwic#dN|)5Y4mXmnH*H`UYB3rRNO
z2CPe``9psmR^s$PEwV&1*7*A$1syF+)DM7a+n`<^?LP_U8q{Z!x-V8k{@fLG!#yFW
zbEEP!+d70pNP!-b@fr-f?jzsbFTzp^4nkR9!O#<)xM;vz*e7T|U<U(MT<bn7rRV)6
zH`fBNx1jy(f|P$`7d+f5ix`-;@x1Sq<t<Ku+Pkbo5frLaR=lS6LfsJv5LAV8t08>Q
z*Uio@rHKItFcZ)dMa3l#JT$1FZ1nh(QX>})Y{~C@QhRZdPG<Pz`bwbaz5fn(mTZ^K
zD3~$5&Y<zQ^pY+4M|j1^*C3sikP`&H>jcpWNgDlgMEzp=O1iqCU5gMrBZr0q^#$;h
zZ%1ls>d;Qg9!}xt8`1esp~-4}OkhzV8XA1K*v15V(!%;C7)ONcKR`!qgnSOB+pf($
z;?gf(JbkLpo0(sGyFd+Se|e?p==3*&m=np?bnYvWeSy>ofcCINK7EQzV7IWztxvU)
zG@`rtI9)0|XHeaQ{Bt2BBk+QRs)ZqwDZHX0E)7+>!s3k>(`2NqhHO&D%i>XXuX;@M
zpKxegK5{Ok8}M{4D<{0vxfb7Lsgd+KGsn=>^rxvK;09w}1$+;F*+%{}E+Hqy7SN5A
zxTmoGfV;n_ATK`nO={}>^hlb?doO}>bLC$tV1ZLX9w8Kvr|tq^cvqq#{v`rBp-xs7
zbG6*yU<c=?1-ZH4r!662VrBfbATvSZS-Sx@u($xvUcN*4n4=OZJh3l;w0)V$4-O6n
zKYDg871guya)45y;jGRwVIqJU0mu2EWH&hYkiMXYg(c4d+)(oW^nX>WCaWv|vwd=&
zHWw8Zi1SA5Tg95xy+{z7MEs1ASN}C=^#VdSC>&ydSOrzR@lA$?&k-~PK#HRoAJv3_
z^UmB{a4B5*2XSqn`onKJm>*{K5eM}DYf4m87LLzP^k80W`86*y!xXTFYuAF+)-p<;
z+UUJ;+OCMxBo3MEZ|@)cC6+70Eb--QNv@W*Hkfjkj^6bWo~#Cjvz+ngB5gfo=`YU~
zLP7x(s;{Sv?_y_RiBu=8WaeczCC7~qktm>pppg86f`f5hx0|ZWHwlBo!k({ZXBA|2
z%<l};FR_DsIDw3Ocx0GPQVw~RK6C?JZ1U&gjwV|L@Vg`>jcRC&08%L{<9@;RTv_=>
zB6&GC&oL@WRPcz=X!9R%d#}l@&YBdiy}bV)7R@Q&Io}nQkjD;&n-{7tAe|5^5aq)(
z!SnCpzi#&O!K+^V`?}L9a17CD_nfO!^h~1;4(yY0qWY{v<Gqv6gBc=h*^=R;OToye
zoFvUB(G9XKt*)HD0~}uNxMas&r#*r`E_qk$W|rnt4XZUT|1QNen9BN#pMM%0P*eZ?
zAw{TW(rfg8h!*ORjZPHK;V!$k=hokbTR?|x9!ogn<l^-tecjgqL&w>9O>zGdyprXa
zVvq-ok7+v7KMd~!6Uo*-n_V9YAi|i;Cwg4I(G=e+qKTBtwOI3ZVZEdTp{j?=)@iBs
zoB!UgZw&pfyGV(nUELD=&ph1V^4eH=Ixug%T`BuMa;vNsS-V_Pefj*&tgH6|hy|iG
z|Bl73T)B83)Ov10W#}OL!05Nrf55nU--@!_ulbN|buwywLJs8JI``R_>4KVK`=HPd
zS-h1U|CZS2k^onSk%?q@@Y(Kua=jMW<Rd2wOgcf=5-?R-ZI62bP9tj;7H}izs-85R
zN&1k(#R?l;=6P@B0}$5UME@DqT*b}z5Pn;V8wTgojXx(o^euNyJAb8vv&Tbw9fn-y
z%_Bajpp^j9LA@Ao67G?kA56bj)Zlr*E!Wq68f1N}S60sL>{NF3h1b&Df`n`ZO^hHq
zs9~W?f*KzrvTy`HV=@D!_S|6bm8cUp^7*q^f~dzog_Dnq<iGFh@1Dl|0L-+w_YHph
z%LVXqcjkM*y<*@ETu;X9HM~k7aQ!6eETyfTN4L1YGU9yrp#z2x<h;f+q5~Y1n}y(M
z7Z(jAW4DJ{PgdlVq4S~Sx9S%iX!i+=y$Y`@#SVOFiMBoZ@{Xp<3yalF;!JY!XZTXj
zo^5OvTI$x+)CjKLCX`lQ#a-Rh(bP0}RZoS+O7%1Nb=|~z%JOfArEdPMAX4G2+G5XH
zNXYT7r$Bjd1=W$7daQTKSBRK#vs)hw&cxy^(B>Lx-L*}kWg}iUvmGybrlh^zUZ3KV
zgza;K_c?KcqA8B^Ft7@(A+WV2>KdJJzrG)M{wHtPDT0=7b8ys1(#J9FyeCHEd3YHn
z)4!i9Y44G+R>S<BToWvjmU~v?&IU{e*7<Bqgl_->vE}gUV70S5Jm;m_^A+;koturm
zBYSvtvYv&}bg)Jxc!Vcx$FU?BdNKc}){Em@BLmgk#_-W24{tvGt#6Sr_^vJvnq{84
zur#Ix#F6$+xt`qwGUA;qeSVBR4sFY6-<i)5+xowb&p!5T8Hx0z@IO^h^~$(@s<55R
z%>|UU=n}HUBgDV4tm46oloTn-)W|p}{Qm!Ev7x1J%`CZXT)*D?RmV?$!eB84EHvh_
z@`{y>jVJPtlSIeg-@lg_6}PkhP82cdG}d6eu1%ch2yYVcb=?xB?IeR_FXm|7NRKTM
z&rb8}Fk%a-s(SQTL7uyk6To>LZH=pxotd5a4eJkWBu=HHMd@jxo*x!fMn}dV{IS8~
zs@Qh+r`tm`|0Wa!e9#6}q8?_s?QU-a%L`O~G(`6lNf=q_aNzIm@S-5<se*=eOfw-V
ziOi7e&!6Mwh;6yv%)-t4g5(O83Z<>Bt=Gf#;fArNx|VXE*SeD)yq>{_V7dNHth%ti
zQvur(E|TtL$QLg4YF>SrFse@};`T<Tg!^G}1vef3_NV#VfcIC|)Kry;`h@i@f0yce
zeE(A1kpA|-z|`StmoGqxh%a2&AIb<3>W61tP@?izLYpUheVi+O-w^Q$Jh@9m(NvXD
zlqc*xp(%v&DJA-BP?TgfQHr+r(_eqJZ5+(m?@yWAyAnvl!J{OR$BQMF79al0>N^>N
z{arhVS_`_|&Z(~#f2=&kNBy>o3A)9ncL*d$23kAsA~RHkT3XNHfQ!4Iyf@Q<^}?LF
zGH4#WjVG@wMcDoJQwvV&dTqk+E^8_7bHBzYe^q;aML8Gz*+5S7lh9|^eFPBJ_P|iQ
zPBrh{0#Uc95d6@9jRj!-Uw!6WP<f>h|MrpB(tb`Yy~=CjUu}K%a{N>rTj-3}b*)3e
zBNV<Tm<3E6!6Ps-e*ur!!q8MjL&J4zU=Qv<@Pzc$=e8aB>k&Sz7Ms#0IP>if5E_Bk
zhJw5`D|<}P#@gOB!pB{|k0(QT9qwW4?!$Jztq4<l%?(%rbm?i1dpDiRc_NcJo|vBX
z)Zm>R78c})G<KwR_kD^^+M*)tG$=zw%D8`(aSwynpyb+<qg}Zt^5W+xZjb4Un4`hy
zT2MeB^$Y{zYwbxBQ^VpavQ97B`uG_tMu^dcM1&p$cj?~Tb)byad5!sY+53&dgTWs5
z4}JDG+7%sL9Qua_)>aIzy_crX&Z0Y5?%ogc+_Twds#89#R#zW(r4m|vG_h)7DG*Ao
zu(aIO-MNEz)2O17ZV@s7z0QAWsQ7JybhmAOOVs_4fsYEIzQ^*wdd>Kj;TPyr?s&g}
znn~^g)cxdPq|)iiL_`axf58(;QqWQX6%V+hTpPpWy?bVcrdCGgQxk)`8zu+|s*sS7
z>b)8kmUd?smy_%79CQXj%1%j26v|c{9F-Qdvr>bX+9{_%wzA%cjEty>dh8^|x6zY@
zy05PN#eLpZ`$EJ%=y4&z-7TTlX@yx2TE(tLCS}_zBBKb8cAc_ssIcbcemcW&(2+f>
zYmHWYm|C~bM`i?8xH~^N0_kRt?TyAic6-|n_y_`W^uFo&?_XW==>dxWo_kC_tfGtS
z^o*?O`57c6cT*E$jchGjnD1nVJWbeugbWcB6bBV~GTCJnZSCKaJ*H(9u^MV%1Xytx
zfV;Y}#!~bZsFBH=i{eXHWa}Rl@6>qD(g(?5<A$uPtUyLqen}LnYb0-hE?3j|Lr(g0
zYp3qh;jRhYlqxky1xh`>Dpc|F`^EpTR!~q71@ZK(U4PCV3Y({_?0n^ou6*#WTuSw6
zD{u=T%Z8w7z-vpl!fX>p6fZ?u>dv<9d%MlaN7$BiN-E1oo~Z(Y`%EoD^xG@JBN|N1
z2wPo|CR53J6dG>vwU!~-?PE&y#IJ$dYIv3N?+-Bt(4$p{k`fam9k!S^|G8hzMD8)|
z$G)%WT3f`ba-H2=6KWbS{rv598`U~{_Y*CW53ve&_QMLlfUHeHb@EUD=Ph=a<0Ne?
z-6cKPbpGYBl@ooZcm0O7frZdgpEbD=xEo36!`zutw2YkjJvGUTML1(uos<;?qi#C^
zIrMMRZ(uZ<nEH&?IEJK#5j&Vq_p*1&2SYZk1xu%om5#Ngb)y&1Q`C8!osoai12Q;>
z^9T^6=^MD;b(^cgAVydDMQlPWz=rVbYHPvXI4(BU`}AkS=Woa79g=dtF-`@v`3ifx
zFCX++?pE|z&PPYZ-%n&!yfZ(@<3BJqHa0Rj8M+VWJC7;{9$p_<a~O%VKx`SsSf!+{
zIw|7Y@)Gawq@j_?5;=a<$7J`785`Akt_j&!<~AiJO%~>lJL0{?=bTR8ObVyEf(It9
zk@Zq{Vg=|VN=kzBH#qRoQ4sX@1qe`+mk$)(!^zLjt#{eU$p1iyAn%)sbc;!TlYi;_
z)oho$oB_H~M9flK8wPhlQi&UfKH)(*#j`<89uG_Fg^sWWVQV-$QN8iKj({hE0RJER
zTz<jPuC6ZH1|`Z-_>-`(ag`o-3AxxEtth+e1U!GCUT4;5^=`@ojwZmGs${>+EiZ#>
zncejeB1%%8rg6v1WYRIMH3Y%Na5OmLy)Fp%aKxXt8Ior%nu!O-WPr7eHCU2I%7HuJ
zw-Qc(v`AHKY%H8k|Ge4n&X2!r#7N)A-571<@`6l&M6Mpk^{jHp5(De7z?bj3`+W8~
zNwG2J$?<8a#p<1wNbuhD*PyX-k@WxhGf{dWAQ<VO<oX+;T&3DPZ(`#nEj;j^%*~Xm
zN(h?mEg-)gts1Q|_-NPe)m33ZZf9`gms>E82dm?X3U0|x#u|RQH{J*WqFMsRN2vce
zcoX-0VphCEKy<6h{ZdNZIkdj8V(5_|8!?~bDPdc2#nYai>qi?A(UJFV1%Y@24e2-Y
z15a|u?SuU>=TuzEi2Jfw_=-s`a7cLatiDwnLBJ3zPlJv4?}6MRL*z)I*5KGFR{GOW
zLU_$4L|<zNV7Bw0Pv>-x(&)ewyOQt|k}fOT@Yfe3Cg+cO+lJxD<ia?8bzf%@0)hgA
z!-^|Q?-%8-y`mcU_3J~j&!a?CbiO;;bAu%h?-ZgUP;`NvF=j-E&cdy<f#f#9CrNcA
zCn6tOUTWOe-5wM*`*sRGH@%972;!w{P0qT96>rPL($2$I34!#eNW1Ya1Yd&_98@M0
zmX?|6nQt_&Q^(zcm<+F;nUj-~tk>TQw2oxuwJ^Z@ysUeq6O4<fFP|+E5y9{i`NxoX
zqTW3$JUqU`is}gq8-~YucMIHX1|+@Rok5u1h51b3nygTktgf2rR=WUcFFW~^1lzQ=
zj2R>zaNplRp6RN(UM;mK{I+onN`)|ay7Vcpz#!dPgTwh(^|iGb#k|5S6(5!C5d_4<
z6IX31BSQo3<QF^h%dZhJcT?i!kGbgu39%Q~xQ67w3z6tAe=S@Yw0X6j9p~P<syQ)q
zjJ@h=&ry(k_8!RT08fNp!)o9{&gjvXlT#3)^n}vr^LwV}-=V<B$|>aaT3fK~u?E)U
zJpn=YKAm6vf7+SLxcO+JPw=P&hen5E6H|fSwzew5#lY_06R?XF(xg{cQCZ%75=BLJ
z>Au;$3Wc1y|9iLI%c%H>n6#Y2Q|sTTs0;qGO2fZK>cUT~URgtwL&^n%lH^lq1y$>)
z-&W1k(wdrP)$!om>W|X&`AhN67*Je%;T^|^+n8JCX1_9}OHbmo$^YH`?!ySAT9UJt
z78Vxzw<$u*2KYLtn@ER0CIICiA?1l_OOzWVjI<xxIg(QCoU+r(h#T621ztkXm#14D
zREA6Y%pB*aXp%os5HMxnpqhlo31M@TPt>q6fRpJqx53c7M$gP2>k+sDOYy5K6<j+?
z$yJzOn$P7&gL%uAGdblh$%heuUz^~|==HM&+!A5%!IxcKt@h}#u&_Zv`q6VlrYHxv
zK1HMulYcrsPT)X%KV4BSPtMK%bz9$Tn(4<x#1w5(r-y1RSmcwJEz?Tp%U*<>vP^Pn
zc7ynST1vW<hDrG0nfi>XIuV004J0Ch7K<QC<&>-k1u=cHC*;0+KE6VN{1cK94Qd9x
zx(<V~>JL<R$SwsQv7sXUV`HW-_bG%a5HTfL%dXzKpvPcg0aA+*YLtLml?uG%w9y@v
zn<5Tnl{-$ZIM_Gh<Ktr^<NT<3G5l{}-gxyO0jiJ5?|c2-&HZiT|6Cv2vqrCI)6-B8
zPz*QL<1HFzLgakJDCgY*pVM=q=9{;noqog&@MFvBX~7wS2x@mFCRPK&yM?<jEobh=
zfSf`Q#{agzPsr@LnVsgwvAqjq$D<*&ZfAxNwc}9oAn>m+c2i19icSojq?CdrJI~)d
z1x$yDIltUsTI7dEC%x1|TCsZ}LoHtk<sB8ed=Id0hA@EJHYPep-{aeJ`vVxKDp7MM
z@1)y@0|I1vZ+sf<rQf$;z>s<>O&Z>0Kew@Phw#ApsJ(|{!~FdyE;2)d|Iy$X#ZyNG
z1<=t^k!z@;qvK<dg;C&OH(6n>WuM$8Ao_Ac<*yYRPTL1m$i#q(C_OEs-Tw5L7}F01
z#WW<S2q#Ga{q3C^Wwmou$x16tlSf*jB3z3-<41Qtg~53M;awi{gR6b^LC~mq@!iKp
z7q_;4A|>ws1oO>%1i>e|Wn*q=SNj9JHPqDusLV<HzHA46t-ncz_y$yW#!q;L)ybGX
z4?7}sm7-sGf-7!^g;spxSNz8?AlBU6X7)o`=xC|+FSXcF{4$XDW#}Z|&&tX@pUhGj
zdJsVc5{8~0_GG&;eC+G*>euo<qaa^%a<`)bp=uJpjERjjl?1zsf>6?((@Av3vlYBg
z%~H}355lHXALh|OW|4bDB`oRvkfb1UaWR)g9&ODGe;C1TQqn(85wu^Kho&kc=C|@z
z)mQ3UsX}|gZcfuJtbF5w0y?^P^)cSsX><eyW8NVort5Qtd)m|mw>bbo)^a<oc$18V
z8Sb$_c{=z|0sq%g)w~(euzIYZfF^m1RD>vmjV0#ZOiO=b$OIB)coxPOfH?9M=}UjX
z1RJ@#xbK@BJ*;nH4M<79wJzw!*>^^ez_jO2rTIm;UWn^~ks#$(S8}6c>&3K`@R4Ax
z?eg;X=C{oY@T;D&yH95#h?o^Glry8FHySxr2m+|FF{mBY^{2&XZ$Y=rMvN6;$cBN2
zz~rPb&-PD+TLF^L-PDAJ{Fry2=EVomyC<68*}?vC!ZfUz$?CJf9@82DLE$5csG*UD
z$R$%5S64pDE4)tcK)?TfcE=x{UwM&{h7l9r)+&9;wTyOv-c}5jx~Rwx&WWF9f7jM1
zUFI?G=AF^_H5$-h1}i<8mD=;#+mvjZL9t|I$rGrl&d6kd7OPL^yM>A2SgWyv6RXwX
z@6T1!p}5_;Z72u4JTaV2S*-!rkE{fNi)x{DFTMO=)Sp~`{0+4!d1>kI%rx8IpsGnO
zHPMl9cx=wk=SCffBOb_xkBuXp`&)USgL!o%E-M{p9}KD7;V07*Dte0#{36BrkGyI>
z{O<8X5t@@eFWK)dIHg~Tvh5XrYqAkzcs2hp2)G;RhTzl1*lMgMMZSbpjZL>H!HF%2
zFVt*8hRyrSb=;FeS5{Wm^X1KD6|wE3rhVF<c8PNvTqFVHWJ#y$tuMK7LZ2qzpCeG!
z^61<1e)Hu{-`y`NOh%xpT3Etk=y7QvC9!if;?%S%lFRPZy@83#$H(R1^i8|Opcl}T
z%zc)R81!_kU`=G4Uof2g{p|phT6cwebSa-S{6#@_88Q4#oJ&p>t?V@EUmbof68yVf
zM}p1E(Ll}082?bY%I#WRI;FulNhqoF{*OZN>&(u`tmM@CG*Ne|3wB7Wq%Vf+a;v|;
zT}YdQ0XyzdQ++1H<6wn<02ra7_<=s$*@a%RP?GHD`M11`s6%NImJB$vGDnczuafxa
zO?FCb+fxVS#Sf;doSRl1H3lD+YUcO0dC6Qy`$4k8bIHt`+F`}_kc*AA=vhQG4vx=7
zy8mgVZ_~WR8X0{{i+4RU;lH04ng^Z~p{3DYD|x^9Y(&K_%QzpV^0r*;9th&2H{v%-
z=W#imy*>B)S&w1{K#GQ*a!X602uA}F(z4{)A(=1q`t@r#$l^03@X?9lV*~{LFx<-o
z*ad<Y1dBd1@!m%{e{@Z{1wk03-4<#Y*%?+1UU{8e8m+XI(m_J0ckt8Ko8*|tvxs;=
zK0o)(o@*b7Keg7y@>f>ZZQ3EeMIC+pY6|zqHwQdrCR8XRIWGM5Pp@l#B(>G61V{w}
zG!e1p^~zgC(}RUr7$OZmvug!wjdgD^)~o94-_TN1)3UuaOO#n%Ei1<nZNG*sIg^9~
zZWCq}LKwjP2%+U*Owxab>1`~D>p|wYhjsG$WNt1RpOIQbPj7Qp13x3veJ=JUW(2dE
zyit*!=?`nLx&NA;2`v|zLr{0$bnAw_OILR<GaE~BX>nO)>FEgDoc1o(O`?TG{35>F
zK}?Ewx>%V|k#D0`VNX%;@IZ~!a>zn~o<rE|veE0$XsL1E-r0VxFdYT#B$oJwo1SV0
z5))Aj9fx-2PywfbI7_D8_JL?aaLG$gi3bQZWD~^4^NDbg?o7=wL{__}@8MQF0!>>F
zJGNqm1q#Xo2ac6d-V4T&?q<&WG$*49K8SDr2W}v*{kChceg-C-e}3ucY9Ez{Of47Z
zRtFGB$-oI$gZ1W#)noI~i#C%ZDpCDMuC2y=ccf&8dQ9hcgr@?XxVY}7^->yl>+t4Q
z<tXo;g?qoDK@c2gb);Sd@0mV@A(a7%_wGdB@Mv*Hnb4yR^@b;&yMT<SL~(tZ3)opZ
z%O05Qf#O@-nfw15f5xT{q))zJg4_Su^>;Ll54u=c(EOU1)H+v`s}bbB3?{_9$rxcf
zm2pG0`RI)_r~x2*>%h4TQc@GwAYic_<sS7T&wZvzfQ8@|2a!IM>VZ-_F){eSm*VOF
zovFecHRrUzhZ|B*kQaWrZN7IJTk=5V!C;h+&Ez+avwNFp@Zu6UZ8o3+x>s5U*#Y{p
zAs+H%w{96`2r5H{wISJVk`yg3FK=B~uwJD_PDzxXU+u(UkEObLOn|F=PHv&HhN^Nz
z{LqM+ezl#VHo2<eb25H%JhN#O@qUG?nZ=OhoIIA0px}I+>Y2571QReV6)X+%rbr14
zwCe-6YfR+3!{%{v!juGe;>WLnK?5Iar(TSGcWD_3walPzxppn5;`tH|PH4NE_EYt$
zSGtf3XTpHyrAzO=X`l<`?5pJk8=ab;zy8<_m%O8=-C>MjL{n1Rf3E}u|EP$bpPw2*
z`arKDE}Vf)n`u>5(#iKtjm)zOa%_xDrZ_y6t_DnuK?-ZDQ}fSNpPym3KT~KDvgBPE
z0=*Lip8`4pKqv6M__e}OWRI36uP0JF&(<c;{*-{=IpEGZxj4+|9AIG|d|aTVV`%9}
zX_lm>#>&Y5DmJrv0;uCPq!t_V@Gv=}Ad`koO1sKQ+{j235FO<6*Y8rw;=R4CDl%FQ
z^J9w}HoR$M5phwTug`+ZGf875lvQFhRMllDX~kKtnLk%iNtVAQ=Itl}@|`ol#pZ-t
zF5cPBxMtIG((A|tRyYmlK<GVqYQkWXhMbv|vGVh0GoXvzo^xW5KB2pQd2aJ#1Ms?s
z*-E6s@=BV4yu(d@PyCahqnHY!`0TB&9-14R7L`_;RXrBWu%}Ie4Vg+|g}GH#05j{T
z>h_4~YpCktyhs5UPT@}Pz%vaZqFZ`^28DKIM+Th|Ce3Ute9ldyXQmOEwA^)-WoMJ}
za5~DZOW}!?h>VXO7#VD5UD=s}q+oa$^fZb9lFdArLJ4VlV<^lCb|qSgYf!Sf3uu*k
z)d~pKEG!zk^X|>KA4YtWW5+qTp?K&;AVl-QwM-PskF8lMxC0))Sd>$E^S_c{8Wjj4
zoF0c8U(^n8Fqd9ynwjSkxV=8bkXc<@Yc;e-M-@3!piKCrB<3q*fZxCH%ZZ1`1T51i
z*n5a&ETNi03>qg$(`8q4a+zXd;#gVQp<x2^`d?+;t*u#a7t~}u#yf0!Z{OYrvOld2
znQ4{`0pkZ)_**LmJKNi4mbsRexw6u2-$pk+_Srje4I~U@X=#6C$j$<3U8qMgmJp>9
z@l3<f3a~m%i#63)hWV&7@@kSKZq3Zo*{isAb;vCf5yr*Dty@^SSJzlr<^qEq(=W1+
z;I`_=?!)j;^XNHhR2+D$j?qvj&S!H_zcOcj90Zny^xZJajfjl|nxEJ28B;<~%mJj3
zLuoTO7{%D9lm9gzuJ*1zj{H2^ndRl|8#n9!Mjw&UfES#9<C7yutf+*L)<2HhnzC)5
zM<n5<x^@DE=u(`i2}8@*N~&*~A3lCvffGU)l(|PiaP}Yw$tJ0;r%K-ngm`5srC3>7
zE5A_pTAMIw7GvgOR@YN>i3^X9_NU+oD%;w=VI2MBz^;^&2hi1`uV16^v*AI71Y@?f
zGVbo|+!#RT&#B2|V36<Xg8l;qp%Q5g9+HlYC7hONb8>NkDR=E$;s;XXN0&m{kTic#
zCcm75d|b={ch?vH7Fc%%=7s{+oyOAFPOvm=sFgh+(i*_Rjbt5W=8mYrLBQp&xs5LT
z+J_eAzHWT1#y0^$UEO3E@)hXl@u-XpN$&0KsWPD^r*P2GscWjKsOj8n+CKZJ#L_7b
zW0XXa&>P{e9giASH!0NtUW2|uTKesB-%jKFcAKL9jF1<crqCkE8Eh1#udf`FklUIH
zwR!~CgvGT2d6;ghc2S3&os}vxo8T^SWYk#t$}N0q4zQ%sR@Hr?mQh_z{9+vf;-Usi
zCTjhbmf}`b8Ec-{<>wR#I?UaTN-W5#D9ouEZJKQP1%vx8Us$Qog}-n!!Sauc-7^V3
zlY)Ld6dUI!-IUM~&RauG7<mBGTU#|nj=sXxiCWj#r$f5zH77QfN)8-lkoDoW1p!y2
zd(4Wey6O_i-piLpy4F2wYh%#Sk_4n1iLf%a_k0v-`SrSc{p$+oW@riWs5z42Qs1F_
z!-KkUC<3?bI}lW=3QqvAF}HL4w+7FFSW87Etv0RXQyecHtLzgs<J|X2$@%-xfOmGX
zf>HwuO93(lP!Pb-prm#gAzriSgeXO629V6a561M%R&u5iwXrf%9fG>=mQggALenJ9
zuu*ZObEVr%c#9EJ`<&kGE1o-vgAvNYFKpM}H=wNsAzFR-$@$FAL#e%_s5mZprB{Q^
zAjp{=scTb1dE-hY0t@UW)(wEFBjvx$Y{R}GghZV<Vi`r7`rHzgUp07~*bv^N-{7T0
zOue@oCDO)_m1WY|A+KcbN<=^)Y`^r<+BzlLU-x-$d~25Jg9LrwCo-s<N9i>++Vb*g
zrKNj24;#GWo28__X>P24b-mIQBm?f`{9<FK1Yc50RI2|*S{UvKlWwRX31!;ne?QoG
zA0W$c{kP;zCQgsp;MysYWF{sbevEXdq$KSKg56DuByBAcNIxbq|045&?(C|&r>ntz
zZ%PhpH}A*%&K)Hxw^gbS3;c7q#34`wE6G2$w6dbb-#)U{s|Eu<&E}*?%4Q!z2;bh@
zUEPbfwJZB`MT>qdN1TJwE~mfo<)qs}_{v92^m8d$^5Q8c=l|6tObv<CjkFYtM`E%H
zLq<`1b?^J$zB7XwOyRlpUGt!4-b9M?8?B+`CX&w@@`wFV#3M?hFR*QuAmp}Lb{ED(
zbi#pGOg8uD;hlqz#bxNtM5=c^0W7(e$`m;NLoB8-a?{06vhB{M()T~)E4vGP7^xV=
z(tvch4`otxkKy*Nt`7KQyZq_K(`Qh`b#PD{uP~SKo)=pt*A?V{Py?IXV%0liV;}$X
z(|$^nK_hzMA3)W$s*jE?&X7rTI<`i~j`_cz=W|U>)_)g|7EU}G=Eddl6C6z3i~8&R
z4R`37|KOoC+5O3z=Iz;Q2!vpbb5FV}TIdpHL=v_@!usEjH}lspjQ@;#O`ee@X}vZ3
z1)-0xnAurvkM*KF{?|%0;A3M*W<H~5SOKHx71u8UiH0R+m-%tTt?AhuvRV;7_;6T2
z#j9KC)<%q0pY<;n0N(2~_!%iBtrg}K+(5E`W43?7I$hWDv-{7$luKKxB%0ph^744i
zUbxqLMF_9(uFRp5*4OZe3;(TTL!xnqkzBzM5$n$W^NItlUi(vcxa;&M?a1kIR|<oQ
z%B#hBvJXK7t>9t|&GpK1(pag{Ui9)mHrr@=!_PEyAGt0;(?`?s`}^m#{71Hj`=Txz
zXXE!~^{-<$5vlWBSx`KModPHYV|G?}8TG%b<w}}xVX?wmFYim%gXYsehDTD=Gua+)
zLOynMLfbi4r8Sn%+2-cw7j;rZ+m`<{PBz#i2c#cxqzDsJ9{<`}odm_t&fXqQdgRmf
z=WyX2vI;qF_~z_<+?vu>Rn^v*7S(At=6eNPK7GD5pGW7Oo~;2GbnGoYiQq@%u4IoF
z66ijAC+1^}K#Vq^h=YxiSNPknjU>@1Q>h?gsj022$vk<|H-7A6_gvKAr@Ib1o`Y~A
z|FNn_&hLJ<R;rl-qD)Ne8+^Rn;4>yVdS*ZpHlkl{W!NA=YR;@~lG%{q1%+nf=NSWy
z@$vBx1^HPT<Ay9uOp;K_gOGc(N4`A)@?E(yTEVk>B1?M}y4Rmw`*)3Z$|-BA)ta4D
z1Kcc`FX~Cr17k$6l*%hCsBu05Hu7prhs`)|!L@%tFiO<MsgXnpnmv3>A#_yYGoQNt
z)tAsZd>45q$AciD_m~)%FDDBbnpYuPU0FjN*W;i%E6b%K{5QA>1aP5~1i(1=G!7s0
zcmTs4(%Ayl<gih>ly8xmKc=Ll%)q4#U#7wLlZ30tYx+o!5>;GDBk9v8+aJmyftX!#
z3b+?f<T9VZ+&AE>2b>`wW)}OQpzt{A<m613h%f_2qr3Zw8nfRth3XqF2r0wc4BuCg
z+G*xL()|!GMP@}ve>+muul<Jmh*p}u7ku~R`}b(s_N`R`K=k(TA0m9?8y^*epB;k3
zU<z>!@D<k0%V`b!UM(<IN`#b%*Fve`;TmdEBnY)!CX7XrPtj5JR3z>qKiaw!vO^B6
zLuCj6TH6#{tAEHQmkHzzny$BP^JmsUwh9<hzE5C|IX8ciO9XxfLg;>tOm%1BsSP1v
zkL8|Rl8|3bzMH4PFDYr5sNVZF%>dgoJH)Fd9zo8}Juh7D?Y5&quwpyx%0E{=Wd{dN
z$FJSagbCppycc7-f4}ba;*xKAbR?w#Bi^H8kX-=Y6DsL4XHYr{U-9cZT(6a1TBfpc
z0?tRn01EQ@F}-~?2|Vu&8RJ{d&SmY!-p0SYKC#iul)QbUs~Om;tEkv;5%*zl`zHcH
zAMj%@QaLD5zoezD&BZ3A=<1>sruZUNwlufP-F__FbAo!lEQVh93E^3fLi)60@J6w>
z<jfPoS3l3zj>NaHaKN)%YQ3@#>;zKo2M9&^kjL_rOXAsFgr~i0d<1Ek|5qOG>R<Q!
z{6%)*Zhg@Z37RnSV<37NH1m|Ad%6wo=0>XEg{!QpT<yKet+M>|BL^}AxD>5k!!^J+
zS!}0k6_v$t!3|g4YkS3<UTh=1M&q5CUwq*t)c^N!-=7`xdk<Ai_Rmng86H-1aNLvp
zNe3hGK{LNMeOs2fc{t(=mtd{{p<O5d?_%S=U*Fs$&W2J?UcL)KTy1q~JufPPW><E?
zi~#JsuTwuuj*p7`IcYnE1?^euSwx!-;8TnFUy%O>-`s`kXeg^14Z)=lCZO26frx#<
z+?Xy`*OdEOl_~v1ZJ|BL5lATCu_~+vCV~j=hA4DpWYS-tN)}Fi@2a~@vAhGt5(y9p
z=y#dgJ2nM9cb(O^5nnm_&9Qpr$Knv8qbM)GJCV5#kcy&RiL0w~YNfIBG;yK6(7{Yc
zh+(sFY;-KYx0$PlhsP_B@igJw#jSK6BJ-js=v*UvK?**(RMe%fj7&savs|6JU#X^Z
zaf@46>t2BEufRW3_IonvfV~21P2m=`#&dPBZ>-3wzy)Z7vD9^$N>^7GK_HFKaclNx
z+U<<>q<E!1t#6(%hK#yB_vmh~(b&W(hRuC`&^>g;cZ1&l_ATPh?oNJ}Q&j(fo5*XT
zeXEl(c0kw=1n!0S8}lHuGqN^c;_tZjzKNNYIkmxaa=7KO-gB465*}=s^@EUXkZ#F_
z4|<$W^qF-t@enN<W%s=gXlQ*|tj(MM;2{@q3}Paqz*iil?aZ|6RQ)MCI}uSVTw~@&
z4`A9i>9PGCGMOHEjKdvnM_=nY|Fema1D-bqIVy%fkhODS-<d<kLRxLIzCrtN(h-7`
zzIyHU!~vkKMH3UxT44l%_jPplQL71WtKb-iF86z1Mki-i{u-s_m?1O2?C?#-+iqIp
zV<Ih=(|Wqa#`P(oCJ@PxoL{DIuyUY%gn|IHo*ll`;Q6-~U|{11K(m@Haqa;lUtdt@
z!%P(>O5LC4(U4=utw><tBfbzx9fy;*{YWUG+t%i2cpv+HleJ#MV0ZJ^tgPqqDs#IB
z;~eLH5z4mkk3YN76aGL3k+9q5QM$)L4L}(`M#fM{Q*6cBL&xzDM2pM<GDpE&Zxh%a
z*U;PX(K)oH9tw~vl#22eGBdNUuWc$SKCR?_n85Et3R|my93bEbLsVo_*t6A*&dC1H
zMNNf#;1B~ti{ww*sz>RM<@-NP)QH&3#nhALKWEeZ(bwlX)6R_eCg$!YlTJaFx|Wth
zQ=BLy3?g8MPz~A6KWFVwjZ6GX8QY?<MtNO$iJ_$)jEVvcq&Ijarn#;d6a(!AW+M`r
z&&bG<p65?a3^evgb~c;sx0bU)Ed>YSB2lad?p;Z?n~RIahno{WVVZ%yv$L}~=1PC2
zkP7m{axBK|-jDiv14h9P27NB~O1+NWq{9%lC%7eU>0bL=QIA`8%Z&sXy~X}DqTa!D
z(H)Ma?v)kgC)WaKtA!}oLt)&`&BH?w)I3sA0dvgQxC@HW=|5J117k8LK9eI2e&7>a
zuNVBBIExVFO(gNn|Ie_yw%8y6GbI3Z+P7c~2DkZ_KbLs=ve8NUpL4!c?1kef>7PP3
zNx^TA%gn*dK3(Fq^~Fo@=FO1s6ahSh={`FeXrGVgue<KyyMfW@)YzEI>*?M)!OABo
z#h=Ry3TkSS#HhbAF?Z|MT7XA9L|s}Hu+vW0lIO4U*-krJX-?=eBH+#D5Ww{o^5GYT
zb~1tpfQBq&ILLj&-rZVPI2y)Z?Ohz2LO#!7%gg@Gd15NKY;@dP_Z|gV*WjQ#)Ec4g
z)-78DRtO9;&#vXXsCjuKsV5Z{L{d*hY4Jq4b$ZQ_)%|R0=;<vL_D&RjgSP<b%6h7>
zeLn{^`oME+@LF&ou`GSP=#G)@+?Y%XKA^Hr+*nJ;fkQ2(9GU}D17C{ryw3YKq1T0z
zga3v3e;Bu~-5x{WGzqqho?;dtLL`pg9ajAyu;Jpp(o4>d@C@ua>$1N?N8YDfQl0tk
z3qW9t6Fcxd(&ki}85uYih9_F`pmzYqF&KUx;N3(+@?R7mFN`W|P5GdM>%ZIAyd3Iu
z(BL~lT@JQ+C<u7d{HZsY1Pffi5*RXJfm{_@#}`ds9y8-^@~DttV`G#W{@sOSBU2Nf
z#+St*OFe2T5zBL^kTVRo=0u&Thnr&J>|cK8RzEb4ilihGAe3R=82Cw-4s+9^&51dF
zuheKTRgaZ6JGNxU;lbfi|HQA5sI3L9$!Lk8x`EuIk`VxfqZzAWVl}E^KA)E-`@!Vg
zziFu_wXh)b&DZz)&gJjS%C-afbYwvJWo&6ZE9_`!shyPZ_PA+09yXTIJb~?mw;f0}
zAd%|YySEMdc$A%E*8<*N4|tnI5<0chc3Qu<_v(zBhp%Y#JB8opdSVMBYqmDQFx~G~
zx?ukh7`R<Weh21;bGM^L@sq;iH6(PbK*|Dqw93NdgtgTGCgM8N`?YQ--Ce#0S3j;4
zhVNd?s=_K?vtaqnk-DZ0Kw3$C;o&dRy@YNT0P)uED1mgMxs}m8H*d|%y!&HKMoyJU
ze(2&M`#O+6ug0D|ICL2a!*zjNu!8(tVxn919`h3N7BUhl@_b1>ZEXs72vRDIc|Z>2
zyton5Q3x{^sE$VRH9Z$wQ-I)U?0qTtyRWj0Ga?H-9Lhj)gN@+<LwD~wl5j4~azkwp
zffZIPwLFxC5d>glcuY#tCOjc6QPGhd-MzE(^M5xs);2evt3-sK5RMPUe1R(_37BPY
z-IaTG)Iw4h8UjHNDPWJ(DJNWe;WPFL5H;BCQ>DaQBn1FX|K>;mm^|1W&UCR%On}m=
zLZ07<=zto*j#M;EL6ZsvW%p87J3LhYI1-S*^>?z$WPHillF#}&vN2i3z$>!40e%3!
zV7<FFRXl3B^b=T6FpeTfz8i4?sOd}7m)|{mOpzb2fs`|{1bNH_IRymSA(K_!A`I+k
z1Kf*CcdBdnUzzyWyL<u);m=|=Z4kCWAB>N)lo2%VT*iGx`Hx{*PJZD9-<nxsVj_P-
zoiJm;dFvJKe^0Axi}591N8(|dfQUFS2_(**3Fy{3ure*g<^~T|Jhi+?o#1yo_XTDi
zxUx)znLQc%_+;T~qTPo8XaB&AyK(=66%1d16AYwEH5QYYRNNJwiU#j9aTp)YF9Yq!
z5^HGdrf<_r4R-BEJ{3y#k5}KZ^8M#%QW0l~m7*mhCxdb;jH@61u-KRw8V#7jBEFDT
zOUpuji0>!piG|_M#m93K+l@hN=}v~q?PwLKG_x@H@;(8}ttbJjy1F_hmJm#c5V0p}
zZ#{rUc7lOUcaJ8BImb`!rizWB>FwLZkLh<YRUnF-fv5Z+{lFP3IQTI*fj{}x+9`Cr
zDCtN*_kxre8$(r1<-v!6)pd*ab?#`0uOUfs%<A8<M{@_-Y#4NMnZJgIeL7e<$&3)h
z_YZ#FKX{5hBBrY92!0ji8d~>CCr3IZ#Px6yyQ>{XsTWYaLkWY9{{Dsvv!7b$dga!z
zx;p+RI$m^VXJ<%WYY0E-4xq;L`@U{wKw@KS>snSB85QG8y)sxn?$2x%VM>iDk%l|W
z#|^%v#WY`wbAse(T}nrhf8w4YnQe(!#`&qWkt)HQZB$$&4|x9F=6EdqxVY+_IQ{J$
zHI`R`11(bFsX`hW8h8*tGYXe)I?p4v!KR*TIrD3CgoLNLb?N?HOn{04_prR890dXS
zjfi&`mHcrA)lV_~Wp!i1m+d1OlSc3e_n5;vjBlziPwMJ|`T=mU_C!<!EMw6TAdYvp
zcg}A7plTWWxsmC^x!b&Whw$k~Zua{$r2oW)&p&i!j+9b)u3~0mxV?>k>XvsQ6ldDX
zMP&q?>b0C0;Pikgw<>|8F*!7t^!*$mEl?xMxa6n#C4c=YC-<5jfqRdAf7p_@`O=01
z@_qwF?(+yo_n2a-T&K8(-Sn=7VR!d|1h>{q$aT1A&=4k;&gDDAG(?o52h!O6QBkT@
zWlag1fcq>EDXVC0F7fYKJ7GKA@H1@oeK#<cn4}>#FHkI#F3Y7g9wD3BW4blwYCKc{
zrS0Pz_R`nf!t93}r^Vo|lS13PL$bDE7Jhuc>H`%-SVSeJ-ax)_@^~Lzs=j)aO&tps
z{gE-rrZ0b$M!#Sqj+N&Jzm%6nZ-v+p<X*J7jYURo+S`_wUT6}2O_y2c`;(KCSA!AU
z_~!?o*k|8(%KZ<-@n!~0o~q8|*oFJU$OxLOl5UlKo83MWtAZ-4<Io->o)Q!TRhcF~
zzEaDfzk$v}V2n)eMrmE0?E>mT-fKrkEC`8@kN4Pmbk_zV>wyjp3=}Xr&My-D(VtV8
zW8&;G*wS3aGt8k^V`6OfOk?_)&KN;6VZdz6FYMkOgV!L5^7gI=8{TiGPbKBo$jLCU
z_Zr{Gf3cb_QB;>^URD72UJ`i@tH8-AOI4(proRS};BH*iql~^6wncGEivCtY_u*Qi
z7W3m9+N^tReV&l72toiBcGj)6wabBlTR}oF^cop?zR<_J^jjT2?MAhLAUGfIrv!Va
zp)MfU?irI6>py@dUX(04H86GSF0uV`V{`@YJ%^VQHI7(qM+M(A)FR}?=fcG}-+u4H
ziG&O>j%2>Q%}}h+(5K!gG~uZS$qv_3)p_5ptRmR1uCD$f)#M0F!u<@t*W_XC7f;{H
z4*5G#{pS9?QwF|<>E1Vn-_uH}OLD9AIRHKaVSAW^A`)N#m2thd5*n>kdO5ex0kctA
zh4+!?Wjdxx4j_RjBWiNb)L-P-Vlp;|q7FPQt33J|N6pI}^Fs`c7KYh(yKX&-djRh;
z#V0>L0>7`LXZ>;~RDjToTbzhN+M3;2@i7!+E4ntlStgRg>2y&LAt9B5j(jaBkabd@
zSZz3|QdBcE42sEkTb`MtFBS-l&+?=rbH88besfbDt1FDR@4S6y-qWM6f6wb_geC=B
zFGj%UBHM8Bv0E=WwfGPCX>8w``5{Q}7?+lIBr|)_SL|@l(w7}VXkL@xvJaywp!qCM
zq!r~^e=ce=ul>sB%Z=Tr@32Q&$3cvHJuEp<^i^xZ{B;9NaDuLOJ2KQ4jMSHxR_~b}
zy7hmUWwPO#+40E<Sn(Q?NY#3i-=-wJP3dNQ^=T<f#v)g~c;_+Nl;8Q0@Z7;TNc8bE
zQS9$YY;;s*e`qn_{D-Gn{J`h*=M}(+O<_xzFf4nVkPu{OX#ZHpGb%h@*23LX&s(`q
z9rKcRWhs0R1f&B4_Q%JgAP1qDnw#CEWfuSRz)SCy#+^txCo(>lKXsLyB!HKlA6mlE
zT%3WfE2}U+Q#|Bal+xG5(M=VXuBxHIhc2FMyj5GP--H1*1n{@0guF%eM?m}x&0Rp#
z%X04Fc=p5^_t7-ieKcCRM3$AFNzzydBDsy169?t`)t};Uu^;Q}7DamXAK$-U2J&J%
zLvLe@7`1^%SPo}p0Z)JI3lTkh^rt%sc;XQ;8L#z;h|rsK(X6wh26x{!6M}ObZ&FHf
z;-&6%C^3fruUDy{YS^TC9Uy)^q$Ox=jE~Ss0c(DEy)tlUl*4aeth4RRXU9E~;Um8n
z3-SBCWDz_+Y5`qvrQYS(s(v>B5}~DERb7(#;bupWJ2cnDYc)i3pOBcStE1y<9*cg%
zff&qNei=Veb==#I`utF9{P?VSaq-DU_~6)BE8jCgc5Dp9SBjRVRGc;vI8}QZ+T;&`
zf(25ZkKHcfEIe#;&P8$sZ)j%kVB93bY3%pKQ}B|Uk0f{njZ0BhZmmIB9~2C{HW%+k
z(ar`N*Nh*GpSAGHi2wUJ=_1W4BEW9Ui)SALxP){ltG%Q1BU!lea}{6eub_@a0PSz&
zWV?S)KJ%HKCfhq&@bQq7Hw0$2mNp=1v)}%_b9{AI_*<Usy<7iL!PH)8k|RIP!ar5*
zi}X&7w073!R^MZ%>*^hQ35Dh#kypI8gI;g7whXl$;+aslWSShhJh$I@%v0`6g8dwJ
z?OP*RH1A_<ApT!G;A-T&8y~T;VH|$`cxhR`*h##hnfv0i3ezx;bD(?sq%$gZ)AZAQ
ze-t#I)38vb<o<CUyf@!qki9_vLR0PKE2G$`xJKg&xObAs$cYJw95=OqnE|4-fGCyV
z9@FK;J20J-xh$(JKOJ+Hh=O@znkxj-VJ-JaMN#1#HtLJ{)`DTcUE;2Q&tTi8Q8XPK
zh>3=DvgX3AeDBfW2MIkz)6H&k{p=&z5xM016=s)gh!|*bbR>3R0GPbN*?s8p;svmd
zAb6;>c%$d}!*5^xdw3DC2>G0^`4Vg#_nsyr|IpZ|iW|e5l~T{(z<ruLb3DPfU6UTD
zAO~z-8Tt$rrh76*5lpo{H2nIzPq43R4IN97vp!t-7~E(8=*O)inN69yz~TF3E4SM5
z4;B)goU#!K<A$6s-r?-sA5ajrU@>WFh6lyb|FofY-)GOvapkN$(+bM3MM+@#H+CnB
zNIzDl#IBzJ1ty%-hZQ3Tf`U@A_xk#iPxP|zt*^5C_O8Kq7zFNz3ILZcMOp;}`LUJa
zgFp0|PiQW<5#m#yiPo)ij<}Qjg{bhM7AYK=RazeisW{#U@<)7OdjW^n-pD$_fze}p
zY)pXu7?L9RIGZ~Mt!mSCHJdHN76+EUXB1aFm!iFuunp9UU5C1<Dl}nMbX0G^2WshW
zql>>7)%TO2lX`nqzoGc(ZbeDf=JNQzOIciewwLp-*~h<s$6YA?)OigJzH<bH78<g$
zwk7}sJZTf6f2frw8zw)Xa{=Bg3SxN6_7G4b;|l0Ro!a6K@N9I!=681I`}h%MyoQ>?
z7<854*lusyW&!HdldW4Gtf|NnK*ZopziKm8G&EM@-bD*(WNGg@EMJ|G0F0`xfav!2
zBN#auMJd?Z3MHa6JOrkd(t#b?KcYdS;*XF&iy((oA}=6N9(jGi>wd7?vX;vtBm@kx
zwEytv6;o<FFMrKO%N~MICozevF8IS-DlY@Q&~En*8fZ9aA_G@RM*%&CWh&&d<_WWa
z@cTot>#lt%LQXr~1GVQ8Mw0%}e-q2Ff|~T(ACKU(Fkzs!LbYM7Rk{0aTH~*8y1(vA
zhc*1kHh_2Pt%tOfm2Y{2MH4ZUC@U{Eg<o4d(z@z=N09JB9~eV7LlnrX5<SW)DsEs6
zY);nh^WRUE_(HO>?yymCNFm~8cf_kslv%->_0L79eZ=p25y39zCFu_n>`;YluyvA>
zlH&fQR$(<AqT_!0(`Rw;n){emS{f~^HUTX%QGsDhVAp_Fwe97%pFbsRr|OFGbM?UI
zE|OpwroL5yWbqN$O`HkcP{N<P*1QIu)Wolq&Z>DBKmoyAc}v7j;@uTK)bpt+pnFie
z2?+vpf}<SN^bc502OdtV${yard|UYcxOxkys=BUicmoO|DUwoxf&vnPfPjQF5(3gG
zB_Q1;p>(Gph;)M>f^<kpw{&+(O4mPq-|u<8Z~SMRaSw;yoU=J+?X~8-<`r9(j@;yb
z8AV}|(Ip#Nr2C9A`BO$}R)@J2-c8IaONm=n)65Bi;0fm6CL+_>Rpq+CLKQA(1CAa@
zKN_vJL|xz^7Ivk}O2b3`Kjd+8p6%VBzrBGQEVSG#`0($}B{+*dxx;+g)Vdec0fy4l
z{4R1L+7918b09)Ga_Apk1E7NZgz0o8>1VJy2Egd9>p;_)lcST9=Q+jIPJ3|5thL?k
z^o=+<XhcAF^myXa<$FD?Uohh6(Wxo(IA900#W2@$ep@*VvJ&(AsRgLVt1tc*Uzl_K
z?(Z)?=%vXGIa82VDoD@V1-+x++!3{b>Lcx)%}l`~O*x9`;MN8FjqG#`r{b|qATB`C
zll#u`2`SozrfTn-H|}H>DI}EpCx5Cx=>;gg(sWv%Tmo3}*Snx}OvTLmbeE%*c$N0E
zZSDMgWjTyYdNyhROhXLI{N%@v7nwCR14BmV=K^bsz{TyB`K49w%*a5eEG<%-TY0^G
z?mD_lVRcoN&C^YQG^Ra6mX_$^Nkau}PDv8%-6!1CT|P#g9~so7{%63Qa*j=^sW~xH
z<R4@6FB}*s1Mbpb(&Cp81iUXZGcy%O7J}9o@1XB#_@ToM`aqntO+(+=-CN%<UHDdU
z-bgE&iiSA9J4z^x@%iYTpHxU3C<)2m%Ug!`<w(xB^i3S)Sm54m$;*4_;N)}^?I;8g
z6_8#MSzXw$k1lYK2>k3OHZ5iCeTxf^P_Ra}YfJ!u58}_uA5Ylr@A2q{*gQ>qUx&-e
zY6YHy@&u8R(o&!i1bhYl=T{7bbaW^1T3Rvmpmhv|J++$|K?DSJ)J=0YSzY+CLK;@r
zOcZis!0!4(Bb5P}hlc=}DKIS3dJ0)|2&ClXvT`b}^;1X%3?}neJa+GinvY7&^MN57
z82Y}D?zG$A|42ym^!5#;Z&-=s7I9udVf)%g|E^wE8oAzX*4fEU!!o<`PfE%$z)k?4
z0#|Xz855Ya^$Y2e__r9nS^4YmM?qNukooge6H#!106H3=d<>>*?jMa3*55+N9PBU!
zxUo4o*3b1G$0zH`$Yi8^QwtDZSfYb{3BYDhgacO?)E4!N%h3U%z$5}k?dOcX!ncG7
z{yuC2Hj*^>@dIBb0H%Z;l17Hh5X%7nE}l+wSggAKFESUlhJQj$k@SZw3}OQjM`M`Z
z;e!&v9C+WlaEs|v-S6RHnn5N;T64A+MP-F7Kn&8B&78rFiz2i#F&Cz-M+Mu<v=23p
z;xHs+mDSY(hR4Ta!6{_AAH=yp!9R1D>g#Kh(gAA+8d_1X7oIaW=xHW~wGtB>d;_b!
z<At%|^3MKlVBkS`&KC$!YeAC(j$rcKOX$@e$VI^7it%xF7ctNfSDERz(P6z1Z<VHn
zku&ELE?GxAGBdME@C24CQH&z6{XRhXNagwSTYKg~fkDuyh3RW(D_3TeCb~H?I&gjj
zJC#L8PUR&~=o*6OF%c6}exxdkFYR4M+K^A9PzoMlJskY~yDL&fMEfHPqaxgqQa-J%
ztxU{xC!k|Jd8P3pszsmXy>rD~(eyPFvwwrVkb=6G)PjP^?~_=9pW7<_hYNr`+q|gM
zGlBPBjwD_2+EdB{ZCuPMHEnkh*Ed|*@}g=1adA&(^A(@#sK|*|mE<!qJs`n+agcuF
zrlPmbvsp#pL{;s#QJ>OF)*$gS#zCEIt`yHf^ny#dA;nVmZnp;d$$QNI=`%V!gMplF
zvnN2!zmpvw*zaU@!a&)rdp#|^m}<pO^o<zVb5Me*eY#^;7n**K69PUXmG<v@C+V87
z+(X$FA0-F|voPB3jFLSfb_1#jggog~dKhQ;3ZLLMK^j2ur?F{RTS<{r4R~JbCjVhm
zh{x)!n)oG;tg#x8rg|z|3ctQl2c2kdxi>rEH#PI0`)>H;cN{I*V{SDA^#eGJmoK+b
z5iltOG5oUug4A<i$Qo0Y{gN26D2eqS{@;f^zWwmN{x0<j5SE=A#+t>{`&dN_rJnQN
zy|MR+tGuLSV<P<U;nh3va~!dle%~~QV3+h0olgNf5FkhK|A(1dALHp0mQ|Z>kRc}Y
zoQ$jjBSQRS#hWqiiZ3QsHT3^6EvV=5w7DFg!2@N{9a-y)Oi~0mvt(=3$7$g4`{cJW
zrjKsg$8`B?@W(Tvy!>2~oP4ll+S@r7x`y$H7<g6Lg?W?QtY?xmtPk#qriYdi3i<&9
zGoWoIF8a9MD<<RG_vowTKsx-LJL`GQ=QF8NKnT6$!arlXyT1=UEBeLtEA@}MyRn~`
z-1%?k$hvfNp8Pjl7KPZ%*nnqr2r7|rjrb#`--AQMw}|hk7ejPH9CH%)-&25l%}C^e
zZg^frm_F4=evx_}!3N$aMcPltimZ3f#{l@Lcbr=+kN<`)@<Q=^t=0_2Tw|q{OS>Y~
zfui$8Uz!DS@*cK#NFFeO^VU#B@$mn+KtPdW^JVtwfa1Bz=!80}_S2Q-Iewr7d1$i&
zXmX)MF8pL1CP}%3?~|=2n@vh%{qF3Mdp7_N8no>PU#(9^Jpdp`-j>vqz8Lq7mVCQE
zgGWK7sy)Frp!k0en4T2gJ9i&2aTxm@y;iL?W(E!Cw(3*QVS6BN?0o)i-v0~`RO%`^
zIM8>aPkZOaWhc1JPxk%CZS2T^1ir{`)?Y)H4?n11d?_G!w^!=+4^B?>ik++YkgF4j
zYrMQ%SOK<)RSQV*Rld$7jD0RTRrPAoFSpPLZo=6S{&h~ow*xxkaO!DqwbD?kOeYvA
zd7Nf32<&@{Tzybe+6`n_c;HfAewDP9KQ*-()Uh5Ji~3)P*l9VtU=@(Mz@9KTVQ~Jz
zD-D@Y;(F~a`~u>hXziKTZy1e$!rATU(!f#&1^GGoyRBoA(7&nWU05&x<d`2BMblv)
z`{#z8Ss9pFiHY3+Qp(p3-`|`#*Y05D!OSB8$d=V1P8BXkw@A$FC~%Ff%#5D8Dsfly
z?R8gI_!wqPP7U$!-;)*>@0e`<R+qXvHy8(p9pVM#=dHrqK0YSF%e|;bOl+*bj|6Rb
z*WoU0Q~=_I_c)ouHQZQ~*5(kK6$fVsehEJVaTJ7#-+gSLk&T;^0ROJPpJ%uJ7Q2U*
ztc8X7JAydO-%6@>@oZtDv>xr<o^vF>K`n$tM1Oqz_|666-q-MQ|GOXe4li2&Ibs~-
z5(fEKV`(e>eWlfiKloZb;WO+(BivkmhhbQ-G1@se0*^#hMWq(|#&i03;ai`$jJ~I(
z)nuPFpZ*d@e!^FFOAHeaFFgB96l53lCH&9kWi`Z!2f_gVtkn+BZ125KrSeJ{#Jn7w
zRD(l9{BC|ENJ^&fdqal70r^)EaacFqmzH>>#MF!H>gF0heon-+zhf*xzYlbCbl*B*
z#e2+vi;2h-R8RN-<!V(w4d$PC$u$ojK3X0)t5F{e{S}hdpEEEJg?NFp$X-y=&)cvD
zpnnR%!2!P|YWScg2R_6wF?8_p@xb1_zlQ}A^9p05H2G^a*%g8=TPq7oG>o+83@kKY
zYG8Q%!2U<en*Oawg~3{Wpj|OC<`qs=`A>q*c5d}G5M^G-c4A|OZ*FcvC-?F*?@6Yv
zyv4RiYHHzIa}(>0t$WB%B|PqCZ90L+JdYVZ4)Wo>XX}}Ej7f4dRz3P$-k-?720~Q+
zdrUx)MuZfFCxJQzv+bvy>&xs(aiWZmSX{Qq12$z?cJOadZQUkXG~`ozG~@NU#~jR3
zn4j(8u*ZX-4cOZN^Fg&>W}0*BhO;u!5>RqwKJ&&Whi|WkR=44UPUp+7v8lQvHC4@5
zFC{kM`v!e|fX^#_tz5#lE@Deq^(lV8#5016xnh~I#orvpfi;1ck8iZ1;)QpkOo%uZ
z7S?s&nt92m9+RV4%gaED#vuO?7VDQ-AfwVl0!{*K*XnohI(pid1l9rBIL7c*9{EjM
zt)wh=lIe+W+x|LcOO0Hb0yi!;l8At~ou9t6AfyyliH^>|hf0nyb4Llc)-M<@>w1{+
zdecry;ETnSVxZu>2|Ua1Y@8gAuOT&VSG!HEquX`(j7<5UmTRJ&mPus8bs@jkl{6Q+
z7!BS_c5e<<`GNP8yt$nAkac8iXpNe1_=uQ~yc2FKQf&HV19SVX2O09f`>#po-#Z)r
z;t$-PHmj!oQ$uzrI=Tj@i2U>bTkmOpAuT26Xh$C@S%=a9{0~Y>%4Gf9KLtri{iHjX
zaB%F!HSqt-KAef2o9N%!-PL*OuB5I&^YlqtYU=K3=AfiAAwgh;-T^xZ{ggQFaJ*eE
zW%%&}G=9{q;TE5VJ<cgeh%WTDWVXRrgT~?0pRO^E8se_qcy2RSNQQGUB0@*w<yC`D
z4|as7{H`*#3JNOoRJgA}vj9X63Gv$C;FsfNE??}n%jYq}49cz4v-1i_4_MxmzMn3%
zAdlQTJUlEcWa0V@<=pSSzGf4Pr6)t9zu}LICRX`bd0h-XdKIauf&N0T*)$?G?{Zy3
zOe_p6HL5>!^@c0K@yS)7K0MGkOPiNecvyK=>@=*WV6=_qnf;m+Y-Xfyi4i`3E<^~f
zFVzl1+7myjHAj9~7JrP2h(TZ5vh>qizr1|P#uv<+Qn=yd=_VK+-n%sE*6Z@JhiBWp
z*DdT16C(>IF3u4JzwLrmeTp@a5uKoJvc+f5&bBU=hm2#80^l2u3H$*cv}#Z8$L#M&
zc6DA#be0(N-1oZG<Zr1{&&#=B0HZmp{Mg+_uq0tk=99a9%jJYuQcdj`?2ZTs=8>OE
zkkg{kc3)B2jH!dtYOdUB<&A%XV?*lP5a$o$FzUXMq33fl^jJgz$wj@~<#hi*kGJOL
zTkHCc?4%FZVq;Zwq&2}7W_+mg54svS!R<C;;N5YCVSi)VysFiQKj@<rW94W+s>lZE
z&t&e5{Qh0De}sc(@zwWX42(jt<e-g<m!vH#A5<84G@NYFHPNa4>J|SZ+VSZg$4)=a
zhkZaVpC2>$;c&pJi$O-VN-0?Jd1O{F_{PWem5maM!STuS52JX2OmUA$*PdZd0iiFM
z&+If7r#>N`bY+=nM<XSAA>y{g;hUw!Ivm#(_C}kDtQo*-zPf*Enk_aE-`7uCS;@~D
z_!FCx`QvMGvCDB^gUa%o^IWT$@0KBFIVpWRF{L42KM%XQRHRw6;&Im|xhg3|938|5
z6vEc`l5$E_O^M)~MHp`6`S^5qm6`Ya6W9k1z?T6FF5Ev;mfSzleP4cN2hX2kJ`Ku0
z#YyQXLds(QA}L?I2;v=20Q1+c`T0)v->BD~h0uz|bC@OM2jBkO%~R@Bo#mHMQ?vPK
zh5@<z;K7FH+7s{o1FomPg>;Xga;eGS<Q#esO6@kkqXUwH&B>ZV7J3BGlp^gPA>~Fd
zm14>q556BB;-aYc4GzU`aJx4o#;58<!e?pxl45g^s6S-w)uP+8-+;L6ihusMW;Ck7
z^8DU2c31dtEcnyA<6al5fleENZ{at%Q^45S(`m3unU2l~G9Wfa3arVOLQqM%kIsz7
zgfqUVSK2p}GTc5bNOzjso)QnmPgS(AD1h?Vd9J&gIW)9;tB?hZN6ve<C#xDJyB4*Z
z&%)Jt&KxamDD=Y{jpdS`8LVe~s=xyeTaK)kI&8-1H;uoK!p|YID;pxq`))>hmQ}BN
zLQd@7h_?Q^eZ$89!;uD0x3RHtMUH}yfFLL&2+~IP(p)@-W3N!BBt#6bJp4#-|Iec*
zd|~2`Ng@R$@_OVl-JA;&A;nZR?e=_ZBwi+AVPk57A3Ofp`$&aigsrEO|H-+#oitOR
zWa4A@y{!ug$VEU&fg?AZLxS~e1_<}q;;PbS69IE!CCjV}stuE#$MzRROocl2V7dZ{
zK-Hf+0UWy?aQ_}N#4&JnTPc{TgSl=%Gi?58R%;Zv*%hUm=X9Qk$@c0~7em(fLO<Cy
z4OGHZ6obObhMOMry0#o{E@>jx<ZI2;5U4>d;8NyXr>iEaWc!wv^%*B~dPeEwRQFQD
zf-y47Wh-jwf%0AcnBePI@Jinf4g_^RFul)W9Udk&&4DX81&>WxX#pxifj*(Gsh(kE
z`1WlM__~E3?dIoO1%$o?H@dZwFfeXUs&)3Rt95d4fce~BBNkRL7$YDEeB1QD{HO|9
zoT$-Do$Q~zJ2wZg?K#by%XQy#g<4ngsNKi;S!FErEPA7T4Ts+5{b<11!b^U}!QQvg
z%cy{~&>6u~9Nm@D-nFPI2l9_1BZ-OA3e@XA3QqPqb)<Wdp!DjZ2|=>Xg4^YrtY5jN
zWE8k7vhpT+3W3|drKCM8)AQ30Hdv*vXJ0Ok<zp`VdIvn(zpJ|dOB&8U-`;%Y`uf4m
z#AS|M2rh2xdcc2gzu^xPnt8#zz`OoyNN&~iHD7dJA2hG3ypGcL+kG%j|B@G^D6F^#
zX+6if@oRibjdY&BUF~f5llg0)JW<h5x%}I*Vfu<*tJcH6bBU3GadcvgMdOmmb;&~h
zA%cC!q=`DbEY>gG`{K7c7D%@dAq_?S@tD9~!7<1`f5wN6#Sy(1l$Xt_gC-=b7!;eu
zyZ#9EX5X)lcWbO05tZtSwi+1P5z4pxn^OBl3y)_G4{fRWJ5I(r%fW>6&iY$86@^ou
zxt@LyrF99Ra#<Ufjl0zpp@H#_UrI?zfR6Zz(Y@R{$$!X1bCv)gI1fiQCBI9OaT6*M
zKgPr3a@}B3BGp1SJ<&fd=Hq@yeVvjtfb1CINtWhsg94>Vc?~0u)SQy^6zW$QXu&t7
zIpg<ipEV*W)g2MN(d4A8?9e?i>%@^I>+$#R^?DC{8<0h;j^gF$ppUE7;@l)kRdlBh
z)2w@A@T>I^Far$1ef_fMqV>bZLs$l<zXN;jQQRho9y90i6@EUKlLp<L8$Xo!m2*eh
zQVgaLeH7Kh=hZbnK>OsJ|IRe{s=5>!zkT}_R(e)y;kdMR=bW_(EiK;>#g-vknt@=5
zEmdJrK4o&m=GJ=GoHc}dLH^gPK|%D%`kqVf#0W83f(`Nn!X=yX$*2E$2>Mu!^pldW
zVPPtOx4i$QlClCl^a*|a&+%|WcS!l!*%?(-#K#+^EywZ>kd(Bv`!p<DPrCF{zH%ZU
zU}g`@$WQ<H@e3A08h5tSUZTUru28&7`L|~-OhkBE>?!YP`Hvr-l7yuWZEf;+P2U2i
z4f-_16ouh1P}h<v_*(m=X=(Y;M22_Oo))Yy39hC^?6&!De&U?MSTp(M8dm4B@_u$D
zI7k}UJ3vKZevuB~R;1LFL&K6SA0L_Ib2P-BS35u$9nrS<dg>=WO&_<J_{`xz;NAS{
zx=F1wWt&@4{nmXtIu&MJpG!iL+nnsbNhp+UUsUxPTG5GJr=bZ-$YXyw?uoFFpm{+y
z@X<{WVg9Xmz|-a~FNJh%I8ZaQjI4PkPfx{5t22Il4ee|}>ZZ~q<Sd;k9s1S`*~v_q
zLU0biO3dYYfZnIb)H;L`hnABWO?z>fBy4|gwVe{Iry===1dm;mw!clNzOeAQRPMwL
z-*$a6^7S}~lgGTNU1ePu`cX_+v4hlUq~=3oIO;_P#K=P62<>$o#L#f5K&{H_1<~c-
zE-UOR)nkQX>Nhf<-~DOK{Z!!t$AKwXynE~b=nBKBMu~{XlOtka<b1L8MVAz!-6yk_
ziMc*^FWG8NjDv-Vm(YFM=u%MBx$unj^vDqhhuS<|49%>oOU){?Y_ei%X>QJGY(#|U
zoyb}Av<kfCJP-F|i(<C;vDK=qd;Xk-KgK5C;5qG&CUIX*gRqKnUq<eFAqAbw{Dhk?
z#Ry0#{)Ibna(+DaS!~N8Z8+OJz3<O$?a&uF7FHNW>?>^J<UqxvPl^P{=g#ZvGj;^k
z;J7CH_P01w6P4S~C0bPDj^ljWlK4_LTEB4w$*#Z}{T5)7<U7(i@_xfp)6_^WHsTY-
zr9f6*72ee&E!^_bnzZN!a3b(;T3r`jr*UzqY@8=Bj26yzdh@R9R5ek*XU;Z1Z^N1r
z=fCG1-u#5>_%+nx%pV5^DqN@g@E<Ft=Rbo-R$fv?#Y9*;+CE}o5J@h=yhe*GE)tiP
zI%@7^8ySKDT$&k+A14>r!HW-_O9Ia4=7}q-5l*)t+BY0k=f<TxBuG|jS&A<w>vpCu
z%tvZo!)|g$U%&Xfxc9@SCRnY^x-hen;d6IvZ0t>(aKiKR`4f+H2_6ucW&*<dQo?Ov
zT$x|2Vd+Sx;dZPjElTp@MLQb6HAb<5gK)Wpb?@u9-frBv)_jkK#$)2@oyq7~cCYTp
z(1?{r`KbiqY@Cp~$#I5@Q243e5vfM6-D4L2&IVVi<DT_PJ6$y0XJ9j2lNJR22SL(j
zVR^=wL3*Rf9<LuCD{ch^Nn-^EyGR+nSwa=k)O0gZYhG38fd(pFD-n2251%TazW=#o
z#ihl9;A7Tu^AH4F(+`*7<oxl<n8lic*V!v{DCy5<$U_{q7^iXeYy~o@PA#S*lx!Pw
z|H%Bz6Gyw6@w-t%-iV)Gj<1256HXehu&r%{7B;egg@M4n%~ho70E3UNO$zKLbi6x1
zgJXkjG&M8?wWs|hM45G;S$TyT*-7Vr|E@1@;w;D*z#2X}!m~j~fa#O;Dt<uljcokm
znRc5tggk%k6@3jZHjs_NFC~UF8K-}?sAeNOkD2_37WV0Acrr6HWv!o)qQh}2B_%~q
zpM5AHVRsqj`C9T*dedxwAq!}vr^?QdJwrkQ;`oHfwZ&DozTvl9<}*ydx^ySKdzZ@c
zrNb2i2IFJYVyBsN8npcxY;ZefE-Gc`J{|io`=7`8)#8N16L45kA$|icWF|?nw{N97
zIi_i<k5!7bWcJ8~-TUId&vjQxS(Sq{Bx*o$u@$xod0*fDR%PVpvFgOT8rR#UuY0h`
z&9FFLEqvv}yFtuQp8|nXnZ|}RL*Pz~j+8nH?dWfi<I0`bUjXeN(22*xgQ4yEf^H|v
zgZx;vq+IqpaMSktk(f!_>Adq{y`lL^^8pP*1Cuc`jswfEzot*ZxW*Ed2sX2<!)4VV
zcsxs?fp7_KpG!_R&y&vafka_?YF@`XO>6{_IT<-S<%yhwFRMswlXEm~1_uE#B=R8W
z#>c#zySi#h*fVyxa`iJ18~)bHibTmPSwu{XEdDvjviED&Q(kQTP+PM7_p3vJoh<v$
zwp;>l)Af>9UPFWZ;ZqswG#E<`C@MnR)y5Z<mF4Q-fb&)u4te9yw7aYl_Z@Yk;fM$%
z++(4I%9tTR@@MVFuU~#-sN6RL1A~AU2u}d9-A<ywe!cZBC60d+O>{K|nV8GTR!L^P
zd=Kin1r%Qm+q8YJ_4{i(i_T0pvEDw#jXPUyw`0XCUydD)EAl?I6aQ3G^T=0NoL^N;
zEVtJ?@hJsv?gJ;l`ZA~}Ln9*ay&Ie_8Fx<dg&@O@@6uDN%C)369qsf(Up>8=mg27^
zxDA4A_is8M3}x#YBW-P6mlg(|;TpGZ6EQzx1lNY$dKagL$@u68A!59D?#N~o7q87l
z8$&H<M`~VfI>p$)o4~4UCSnksC4R&Oxz?jXFS+k=R40GG(1N?)<&GZ3e;$eTaaKoU
z6vSlAbPSJsdsP*M8@rlU=f~2|3uTh)ocZ+<{mWv1O{6$I1fHFsCsD<_-FFtuIAo@4
zV6v|}yDxT{iI?-Yr)Q-I&FQj^=HdN<UNJc+{LJ4qR&g@L5E8+&%<|k~Wt;r8OVUjG
z^Y^HM+2?3DIDh?c$$VlX^B6mpvwp$zNy3dPE{>D^CR1s%4apU~(<)wwojk7!_gX2U
z^uxBL5_mE@mM<^}l-+Xj*Y}w2@lV<W46<R|^i4}ExIcSRkXXYaJ_2o=ej2E54?&>s
zUbqT^?bcATuc?K3898i5^G{d%b!%M7_023`iOXtfnfD1z@LiIikE*4rKGm|6!1yPQ
z(N%37AJ9@3BST<*?d#*S`3L<WBcr)SeO0=5VL?Fv*JX+#L`$I{_MxT1P;JAt!<!)x
zVPa<awo6xDw<Qi+4@!Ivi#=zM5K3VNW(F^FJ%J3zH$wF!Rkfd=n{BZ9H$eRfKPmoB
zCrwcL=f3LgUoLQJMMmH|kuOnLUsXk=2`5W+ed2|5s^=owLld1ByMwA9>_~RbhZ5Ca
z$%f+4(cOPX?;=>`uI%$dtULl`yof2T1`~V9lm~{3@+Z=g0bpW~ZblpV>9eZ5lGb<q
z_IJP?<kz+koXJID)k#+kdhr4RT?vU6LqoB%<nFxO<T-ZfuTifqw}J|efG88E^anbs
z(#=WHbhu@AdWc$S4xc0hfSdFCBwQ<}mo}zjkVm$(XoxYcQCVFHS@rcw&rul*%gTDX
zn$^^$`X_%oC-lPc1wMIT7xDVR!RFG$d$a;=3Eti6yjxxo5q-}Db!{FOH+Z8be2%{x
zu2l1Zh=gQ8JKaM`OKY9OCNVK_6IO~u7^(PXe?Np&*c=44d}kkkGhYN9C#NtM_~s(M
zBjdxkh`}>0kne4xtKBQA(oOqX%m4I=KG06y<}vQse&0}G6?C*#O|EPbF!s&P%KKn@
zUzq6m2vBa3yG-{kR(q7Q^>E=g%L|GsP8BGU1C6Rcr&ExRPep0zZ-kslry@SLxH9+g
z-Z#rP`i7QoL&U>bl{Y^B#D{Bd*}vFCA2mOAFf!Aes2AUO?ghrd+u`UK|9Pyuk_4kq
z4-bzEmnX9)`5+^oPR2&RcPKLdE!7*kadQhJn}r^<%V;3QGr>#2Hz>yP;1(t*rqXCj
zjAiyI_`es{c|bG0icgHN^^M5NNkO>4)sw05D&Cpg=tGKU!^x&LGUgvZ9rUbzjYY0$
zfSnyY{Cv2co0yq`fhl2JEQ%NGuBnB02r2xCnKO-+uF=}09g({6X_Aqx9Vrx(`c+m}
zIlEM7XeKz>_hn|@|H|AgD;x4*Ml3}sN^w9u+}4q5&crY!8S)7B+kU>bCtKeDU^Q*%
z%j@~uQR3L-yn2Zbf?EWv7w_}7J$rax1}Gy6&js!mxaG(A`2lyw0ixFYn@&E(^Z%N1
zfTqQqYh!5QHI!SrHOW!&AJf5Dccyp5((548j2o<jXs284`_=%>U@#>!nAFaG8J|*=
zZ8zWL3$xK_&GuXvViw->!y6hIZ13m+^NO4zD{vP};<5PN^tu|PTD}sZ;kIOQ^pUWO
zj(k<Ov@iwLhk}B-k%cMzxyO(lT#|{qkI3}}`zC(_MfGN`TXS=B)T@E(ae;YxwlFg;
z;s|FT3@Qv$X{(l;HmwVQhCp03ZU&p$iWLf@U;hY<uWlI_c;~Dhvxvtf6*|gWlgh@>
zGcko~CEmQtK-b;hFzv8-15VtIytk!JQkv2n_1|t(rV!ssg+U-Xx_|1jwuXj)L&tIZ
zLhYiVk<%u>l8O$^T>w|df446H9PrJ#E;+7s6{=UDGo|uB0{jI$Ql6*k)rJ}OboKm6
zaq3AFXqo8r7QKI2XD4`NBWUsG4}kC^meUqn#jRul=50PUNyD+rkBFs}V*J)=??3v`
zG$-2P0rQ!pjTJtVlx_%-+dfd^&&c$IK%Rs6Cvpl3bK%mL9WQP^gp@V)7cXw^pTGOJ
zdqRzVwS%dIs+(4uevgUIYCUFmFiQqn+rG1$Q9eAX^i6mT3VlKAwDLH(tHF2resjyc
zbp4iQQ6nRxh<Zla_YpBsV6QVeIS>`6Yhs>5y~h#-A<KCWef{d=Vxm$Mb9Z6Io05r(
zpF)0*s*?Ou*Y$VO()n@zX24aIHp|b5@c|c-v(<{1*A3X?0gYy6wg^W(brd%+&tYYK
zKpq$IcY;;g;wy8nc14iP`$RvUln+MiCpQnZ@5gn@cD7OQHhauhJVr)Al9iVBWD%dA
zhGA%6(!Qt5G-)fn&2S*eyVXUa%oCHhw&Yum;k<KEst7vJc<AnD0u%$u%=~qb#grkG
z?Ua?}dCjW=FYx<!cDpxn<oq3#-*S$D3;O}*2(nhv(iWVv!a$~KVL{1Lc|w&g5oSXT
z>u!3wpN5(F-vZP(Bqbmo8*7)m`n1GhGije3*hGQe>MmL3afWh78vCu_!~bvrK!YTK
zP0edo6+Y2+3zESk<dVP>2HBa&GlEGf9FH(Od6V@EzGK7(N?slw^^i_k$#p6{Y@CDJ
z(TW@E<ewg?$V_5@N46}^4_HRz>l@H`U{Up566Yu}Zr>>BZCtii<gmTAdU@koT1-Yo
zV|@c11a_blnT<jN6`t+#e~X03W1ss;KPvKrqpK3$CKut0hmhX7b7y?T4w)@~?E(O}
zU~<0HLM?pok94QX#kyYli{GxaS?k(DYZY+#^?qOXpV3x3|CLR&F&LO%nyX)m$mcd*
zf|&)sOWjMS9)!)Kz#oL9lP&e<=DAl)fVii6W4L#9n3}((y%jjZnp|mjh1uJB5-tMz
z+$DqOtqzZ?stRY^e-W*ENr}jg8T|g{S=Uz=yHYg0CHRg&n^p6PQ<cyX;eR`}nVwv`
zxIUYoh_%Uf8O@=id);2|m!5qbNOi!m`<BdQ)cC>3%%kj)JBLR>tBGnHpr#cc&@ooB
zIfe%<O7o=^Y)m}~pFF*KM(P(z$>G{lU>crO_pSIzcE5@Ho$8Bic@QR5XEglhadPYY
z<%UUfOS6x!Z`}F!LNJ9k?i}FG%VT^_qxaG+(S3Gy2k@Lp&%ZG%eCR9GLbA$qk7yZZ
zT|CbWz~tq%<Y;f~X+4A>^G#}NgIzn!2zzxZ##h3=T~UGtLDS9Or((31_}YIDnk(|F
zLEZ$g?ty1AvzRB3`_@1&G0KW(OxfY#ZDW>mbB0UJ#9Hfv{G^wiZ8Dv0=VOXj<p78d
zP0S`Ux0DOr6;gJfEKL8;FVJu@adK$)#rl69m)_sT5(k3vl9{jn_XE8Wx%9Vx0%t6k
zBt}QZZe1~T5_c~N?vVe_C9P1We6QLv4)*f?Kac&Z2&<WmMF4)5|F)mg#H&|=Vd<X*
z=2jZW?YS)9a(4}Gc>}4%sSDaCmA047ak!eMZ^5?K?26&K|G(GRPtME_mdbV-FrgLH
z#Q*Qoyi$Ev9VV{vLQzh|@odYxCZg=n(gE`y)En}GC^j;7S+yD{JrIk-^*J=G3I6to
z<-HKd0r*_^C_L}}v$K9{X1W*FJ3A#t1A^Jh0&}>10Tvy!3j^X@o1yKiR~!<vm3jXQ
zccI>m3{d7gs2X4Y|6M~|?=0wUUiQYv#zZ(1fphfS?c2XXZ{11_8ZeRePzUEgOo&R}
zDq-VjV1oe+8X|XPb21jpjr=>?mPb)gBZY3?ir?fb`kr@3Y-xV60;+f#nvMzMBb1c1
zyth4#A6JL_(W+%aaeQ+dUnbQ-Ao5GA0*J2G94>mpuCB!RBe&9JqYw>T^N_oAbd?63
zymVqMhhdHicmA{mmQWL1aZU%={QEL5vtJjF8huo*2>YKhWyKR}G@wL}kMoJ<G=gKZ
z661lfv=4mVq30KuCJ{qj5e<&}u>}bNZZ=T!LCt1wcCwChYujK;0NeT?VRN06jKSoe
zH(%e7*`7~siJKS<S1aFK471SR=I?#ghx}Z3(A27>$ME;NqT>*qucQ1d(lpbnUTT{#
z(+G6S84vX!OWD6vgX7~|4#OW@f>kM>+IHxDo9boxpA9I{^D)uKAA*`b)L<p5#S+)K
zbokxs!h_3VVqtdk1TS%YXnJ}#SM1OG^lXYm{(3|Ivu<Em0VoWf=9lvtb*|ptCB{lJ
z=W)|-P0a1B4=L6j?@ExgaKrln{@f2g9G>ZX)&B~^xEK6YZg_-@2-JTKw4AZ)YZ&8z
znvMDkr3X*t71cdUkNEmB=MC~ZF0lu-Ir&zrt0`BTT@JTNDVAq9D?K`UJOr*<={(_e
z1X)fzl^H?o0dH4xHrpQFBPV7w)a4@W{)XjKz2V30L^@3;%HxS>SdPIqm1<7Ws|VRg
zpMln<?%-v8GEcAy?^CxlPpc2YCswlin$6YVnSQZW9jtN=??iS)vnw50q*LC~Ft)FK
zd}f8e_G%-j$hX~)W)9ebMVhjpeP{maDP1+gaSbyiGZR+E(ss}v|2=KTCyX60ijt?7
zE-H^^ciVE*{$ee9?o(AsLNNDtwliaPgOnmBR@S43v%WbGaS*^?1YE62;$<MFNn2Y+
zo06FctK@IEI+S2DGeZBnm^z_UgMZW0OHF2O3#VmgbX*voTGfCyvcJJ&eQ_7)i8Ivu
z)^L}vzpWbA5ah)=3r~E+%vA5O)}<@_7#q2q9@>NO4S+>p{&fuj`72*p0r!(W?<AaY
z)PXd`+>$J)w(88-54}xH%%6e5e)Y}~sb8dI1|Y#IEZRRO+%6YZbuOP%N84T=Vj|E`
zLUH(IgIilIM_1c+72nNBObm^Giu%hJO(*K4YG)j6CI-f5Rdxpw*~jH<23rGB1MK#)
z^wA(bR(TX}1&8%Yu(r(d?|7^%IVwTVo)DdK-qpYDb#@S@$pgc)<S&9=NnFEMOUWNI
zdA6@E4Xn6@UMLPUkM2kquZ(wB12}Q$SZ4P|wzWcw{bZY3xv_IPyn-7Yk*lWAcvG@t
zvsEJ1`vyg*Z_z7K<KDHJii3+J{$hXQgk!Cvn!euB%Rk2_r=_K3{5PF-<kcoOhR?_9
z{!spfpo*B-kDiCx5S<N%KSV~EtYViIyjUS;M<%=h_vZ`NULjsW=fSxcetMv6E&EY6
zS#cPrGdFX0AH3O}scj*wIr9B<jMVf!3G6euIn|;oZH4C^5PRS$t#B@1myZUUFEfl@
zwtI)m91)>H9bP`NhY_7~muU5?H`K{YkqYb68>|_StRk#cQ=GeYnVu~y9Aoi$Bq%uW
zS2LZ~VZE}F)@iS#@BeJpS9k=0O4^!5xrKWr*H4Q3L`65z#|;hN)H_p>-Xek>&x6u}
znK&4HuJ1=m($XsZ8+z*(j}p?{)S!Yc<T3VbqD-5L$$<e|F8{9u84p?L(UHGBaxDHE
zudP;ITeTKVuj{j`%Mb^p7SyEW2j`}Lm0J9K#mQM-T3QOcwA(kXp{EDS&|`8f7_+Ox
zK+Ih0LsS&%ca*h*o9~4nPZ`=EQCcY=Y+g;lz74@<?C}M;sp9t!EiBJZE!v<~fB?9M
zadVI8nI5s9#?5f))><p7OGU;JN4GOF72y+QGB7Zv2yp>s42}}p4>pu_P}oVejfjW<
zeuA#^91kvJ`1f#NVBldwvjM4!j>6dU>|l8@paZ&f*WDri3>rW>Iy&gT&$<ZXI;V<P
zto3d4zDK2{d%L$)2Y`<a&_BBOzMvvszJP0rNajt*V1r#PEDV>7%rq&&oU0ntb@}<y
z+OkAH@IlY@IX54ewdMl@N+iU@%b8GY4~&i$WSu%t2GP^9xAl~jlyGc`{BuEq=(Q0*
z7*zcLbMeyy;u$mvQd01>CXU?rgE2miS4uShYBfbwKI@oCQYq$;e_zdkb4X7FgPV)f
z_+8h!LvV{e2*BYpmX&p25A;EWsD;nnsfxas*My~Dk|47-Bmmna`JFoqgx~Xm*ga8D
zh1}dEjFf*w;ETNy`SO6HqbT6U$EG>(JA}_?PvcE86r3t5iXEUP>*`UjwmDm<9=U(N
zS>btLfIJ9`<2o}3`A~#pWC$~fu4r<j8(y===gjPOL{E>Akr8$37Xv-rbxIiY#Bhbu
z($WwQl!}kt#LK&S4fVNF3{Z<zl(iva<sl8tWw*n76AQEaA}b|@E$tgQDLJ%C5JRY<
zGd4Ds3ob0o%q{tZuM9-s8H9hGzqgYwl)jPtq5ORiHxS;u?r>YvOd6h-vA+zzgGSm{
zn3$NjFsEQWWe_p~opHbqL_>r8bAoaVk(42O(a4RqTLjQJU?aaOs^6OCpdl~ikX3#5
z&JIQpI*Wt~p9G&@qkSGD4wXH`+rZVz{%}e^CA_1)D66o;{@J(aCYq_?E#B(rD@$1k
z3o*CMsGPK*Pqu<)K21x6WrXD%6Al@r_kHPH<|w;xxqf9uni3LZeyhx>%+FLUrDM(v
z4WZ|>QFhUA`mAR+Mp=N%{{9i;eT^X<?I8t=`vyYkeRU<>DS8<rpN5&A`9lEwuVw4s
z%2&3Xrk~IdnE!j`rQ49fCN11VD>&4j(!*)nY>lKSx~YnSBF9fJk1?5Hv=^7YZ$Dj#
zKKg8ILqT($b2XLPB&J%fE}5BNX8WFZLmTU(!n{Jg;;|fWEKE$yT~1-t_kBY{$~p>k
zG%WNpgRN(x5O*S>ng2cS)Ig;Nth7vMf72XJY3d~;Xo4@isA!If{?DsnV~lpVQ`Zy5
z6DQxnc66V{YT$#3g++d0Zl+=`92+$^`pq%Aslf;WVi$aUeF1km<KRA)tCN=4SFf)T
zj4Ug2#BR#!);P|UeoY_k(sFf3bTt9R=d{iiRA$Mu3$5qP;#mJvy=Du7rQcph!qpM)
z9Uh)B{t514o<tl?$A9}=wT%-5qi#2*J8tqXvW;&23y8JeUGL^QqJ{$k$DK#FbiN!y
z7Nh--07~WqV*$juT%}+-;>+i7{plKyAN~`P_U>1DKTo7Dp^t{e=yt+;0|~+Y+w<jF
z$Qx%+s#R`!Vl}Y=%nF#>h}F!IJ@pUMKa42)g6_LFPx64X<Ur1M9eFk>4g{oxBcTW0
z#jaD^Fls^A23(HAm^c2#)G%F^L!#J+SRI&he;)qI&do`zC0f44kLu;(Q3t0}6Px+M
zc6GEO%!I!?+s#*<`khFtrpEIYJp@zHW!68j{x67dT@F^J^U3Vueh&|UT;<Pd`#w0!
z0O?_9Pb~VE8w!M?fpe&{vnXN0%BY&==4g$axcHra8at3?CB#Ms(mrPYJyW%-DvRR)
zv@WxW$|GwTE!3(dwo%K8tRHRj$I^==#B}u81cA+ha@1hvMW!#fv$I>m%sfRu4}+<j
z&X=z6UUZp80BH$FCGvj9gWMH6r!Bq@!&JhA9rM>q-05A=eztaPX4V%#aSf|=b8#$~
z0#`y&c)l_Io+N|aw%*V9{Wu=2!m5N080}n^zk~tb)C3pm&+332m@3s_Vr!Sbx&B^A
z?PTK@B*NZ1bDFHYx;FH^sJ9uAThyOL8|8a1O#iGKwTaLolNEMXzy4sb$c6EhTOwZf
zT6>?loeWQOy8l^Uni=%=<HIG}dU{`g#^s=@{#|A5)#}>Ejq?7v3@EHOh6_la^r3@-
zjE9Vnpw6{re&?uPs@)V}>o7V#?A}1Uy5CSgb37gT{Q9q$hKjgrcL$vbL)-GdG5pb<
zEPikcI)s;G{>YD;!1;#mHMHT;d;IB#`~Eo8HA_T<s6V^bqDgv_r>N%u*M$qwD^_En
zy3I!x5|sBA?z=TR;*k;V$tqVl=~wALzINN4+tb`!nwmfmwR#t0FeP3m%+cMP_+j*^
z={0h^G|~+O2gK{m0%8rpL-+Nx-J1@lF#6R{kl5ucDAA(h;x-J~CV$D@N>y~SnyGL<
zN-aBdq@e-t9qx$<oa5zn=L4?ezWzZOV(I5PJ!9>WSHnpc5>2$}o-^O*=>=|`D9&>F
zp-ISrt4eC?v9uK^g<x^PIEUg%A2IStX82n27=T(}3erE+KVE9#1Qy_~wk=U7uO5U;
zj1-}X;8O7~EzL2~(%M4xj`Yc^*4iIFU5~<}66AB(BJghlrsAG8`9DsF1jgA#wFgN=
zu6Jus&7U#Z5~EfD`v&s2ThE;5Jr`KU-RT)<9z&&vx$q1HSl*RxFX^HsJrf1K8@1L2
z!B#z5Zr2HT?e)0Sj~`B|ze^JPACx_9)|$RwnCB)fM_yvF-09o9uWN^b=B1Eu&(c*#
z3IC7Cy~Lpij{T*^<^)0{1;2gckIJ^~<`%<7V|kKnrTwIfr=`K=Sd26>n<4Ob>USIx
zG4V-oVd(?jkksne6Jn2S2`YMmfL-N4+;VOIoY9j68Fiwn_bo=9fo-B^=o`|W)h$}C
z;>&eS4<gm$@qOp{90qE}>P&E`DeiczpBpWnDEYkj!frDO%+`8|%nd$c8*O*As^uR{
zO6t&i1UOTpAz>n$Jc;!McA(UDT{`_TJj}I9B@>%Y!ngW2w0>JoM(#S|q)M{tuSHTg
zS#iWRyXH*}F3mMR2Ul45Jx=$!OWdH(5a1x$X>6KC-%;|Y5kKa7NT^+)UTFyU@^lmA
zF#!{XkvSjn0#|LhqqPy-89fpIN9t_kP2c!2!fJ7`U2c=|Hc#U=eX}DenyS4by}(^x
zdrLi9OXPm_nqA#L?288_=?8Qv9`$%#PxYXLzm>rwRL?xnk^W-pzJk}6n`E$)iL?Zt
zKD8Z9OD(LLE7t}44V1fd%mzc_%v{67(1VYaT$w!ZMnSBSrbLpzcsvx;!u*W|1=a17
zn<OeQ8kC{MRCsvd@)W`grpMNJzfVl|jJCZ#S9@vC-|rSumxPOe<Vd^PMipc{x-Xx>
zy?PSuZ<vvJH8=3{A6Rm7q^P2BQ6hHie>YkZHgr&7j^defot5<N&}V^>1o91XSU)t*
z?;7yKUPVr7-y0tHC=ueo#s5Vf??$&HJEs<Gfs&x22(744bF{o1Bq>5ksmeP1e#MOX
z{@ZT3Zq}Jh&DAjj5^-F2+k@Dx8YOu71jQ@u!}*p4xs%mIKrG!e$YHd6t6%Ecq7UOu
ziaTMBLjUHs9S0NR33tj1llUO}TEt)>mhy~u^~0nw5EzMw;*g(E;O5|a%>3Qh@TxAn
z_r<vvPPf)3i^WOxK@>Q0YZJzKA!1cwy4GR=rg;Ld#tqpKz>plOdYh#dTJ=dC4;z*L
zPHoQ?a{%WN&`yC02N#&qz^3Nrc3!4c<#><z-oLC%xPpmNM`cI3X|Nvb{|c_bEzNW?
zNAbryhaa!o8H=|xQwv`B&!Wmm_g+J!?Ov&qks+ieX6~RhCz7r<UNI4NSh)mt%5+7j
zZ4$3bjdv|D3`xr@jv7FNw<NusEa(!@G?&Qhlv`Fv1#wga2;s*Vzxwd$Qp=B#sU9#<
zq>d{>6M~!@u&JvgEt@O<=u!7-=9*vQ2xQRQKpt_C%XUZlz-sO3Xr~k$Hdw!JrIWA8
z=Cll-u%L2ncQVId*YBtIAF(fLi^6CVu=9WIB4lGZ*!~UQX<G$=4K%`My3gDXo1fE1
z|A5N{Dm!vFx(%`Zv#brKUYtyEh>*PGQ^j}Jz)$KOB_Lr9SMzJP{f{CfGvH?UXBhMI
z$A%MXQvLAY5RZfXOciG*aDuokr+kUsR?|<RT(B(c*?)6}`V(sSeg!J`f$4ro^P)JS
z32d5!5oJ!UM+19W_IB^VXHY`pA#zO=^7R~Cn=FU_##DdIQGO1!U``DT_6wJDSxiMo
z#XffITHVja4L`7#mc0rf#L0QcOBRyPy;Z2j3p|J$(e0ZhYd2H!^EaB;uw;nMVCw}J
zQ%HXIxng+>a_7k#;URnW)SM1`&J$D)b1)73w-&$u@|L_`Rb7b)?Nr|&Bsil<{aQQK
zCa<e1frXsEk(sKsNDuuX#`;)ZHy|X)!u;)}d()lkE|t)MEq=@T@#D`XwJI1Lt`h~5
z^d40ou_33ACuR4j{@pHyR!b`z-a#a$DKH^Q6mXLI#Q5&@(Qqvvmw^|!zY%jjfXD)f
zDx?qo0rWR5mJ#EqBe<wD14H9<lc9z*K~`t&ejTIC9pf$H8ugPWS0BqSKOGHD%W)@B
zDD7Fgv{||TDx3Wi)SM6#>OFkFH;9lXcr%PYMq}bnc;kx(3M9<q>4n7lM>6_X9+clE
zrAH_cHHzV!vjqXA2io?0>J@h#5_U7}7~;YoBwj~uMq(plp)PoKNjhZijuGE2+>IBq
z0w`xD-)Lw~3|mfvaZcIWlQj7L+u9^`Yi%ApqP>a0%pa=C$?54jgU+gS49E61$+MX%
z|GLNTI5vhK&A@G#z3c<*SwP>QaSY#_NzKao@D~5puE6r~8Wwb(fT$<)xdgNuT1>eh
z@$xk|H)Unyh6a>i0?O?x+|$Xr`3j+AmPLeaRs%eiG0w!CFCP(io8&fdMgQI95bhx8
zZWDF%>G4(}^4h@Sy~bTd-u;SAPcs<q6!nTC70ye)vMtT73EjnaW+X*!1c%f)9D#@F
zhpU!B7}f!AfknGEFbNym<7nODTS;^}{Cz9lA9Peba0_h1h;3bs1lGcJ#NZy<eHuk;
zw4d>Ug+R3dM=Ak3^DEX);@DW_f{9WzoUbNoZSCDNZnK|&_z#R{`jCB5k-_lg>H&zE
z1Z5hbXIr8e({(evd+7^h1-f7Fvc9}euuh(UOTmMQfV(buUkMou$Ksfgh7|q{>4IgA
z2uYc`n5wrd?YF|P)hHtqhNmhjI%}JY0pssMWc;}sat$!NDBW;>l$LGr+3QawEWvbz
z)DNeO+IoGGF2>UM!2(zr>)nD(dTrnsK%(i~kum)PCNs-7jaI|~|9}tgd!OJOCH^dX
zf--S=g#oXK<|T*jJw0}E^1#*O!t}{41fatTm!-rkzdKj=+C(A2p<tvC?+HpoWTZ~D
zD^H~4qem>2?%TIC#Gx{6U^CxIL4o-;%3U7L3#beUio?Z%s}dMB?m*`pMk?@k5X1aY
zH%ZSIOr&@h^RW#kaDw5YS#_?X{M?KmEPcWWPkI-(lbNcx#3%v#wiH&n-@hkASe%n@
zi<#E`c?DgB+?QumP2b8JQi(nefg}O&Z{W`Wh-LwJn{%mCV$KtKSGQ2=jfOWRx)|ty
z&VY6p5U1*Il89Qd;rpD276Z;jKu0&27XZ)$;xe9a{(v+|v(FI`5%7ICQTnI!?l~Ob
z*ieIy)*Ki(N477Haby8i_<DV?TD5CsUtBgnh=-B83=AkpT+!6LksZ+JqoN3&+`5B6
z?4N~tJv#HNTcm!JlO-SJwn5Bziy-Y&0%;})lSv4{!oi`aoR6$+M<PaiV&|+7e9Vv=
zH!%%L#(-g=LuH9X96pYA$`VcPTl>2@>g@UeE&dHqi@jy~OiL{>BgL9@^QJpckw)6|
z!K<p102$gpqu(tLKM?u9Oe9K^&rzDw*5p2{++LUv@*swz7~US;OT8{SJh={^@~ok=
z31{dMA(zhh?M6*Wk{qF=qhMrU)IT|ZicnJ8v&5MhnR@K6^W=SF3lwc++f;rH$BjDM
z#IKIVUmdhhHr>IWPe~@RJPt$gB-EBn09Yf~O&^3Cr8_(F^UO|5zM5}0EUXFxRChFt
zqkC%KHP{P-`g?ZF7P6r_7d^wj*w5Xc0D1;ONk#_d-mV@XI(w)C*E4h7#H<bn7nCf}
ze$TC85%(UIlszf`QHC8H!cvq2mogzkM4PS;uDj4GM@UIOR|vf77OZ`{H{J#MZgjlU
z;M}ml;?%%mJvki@L#0QaVDr=oE%RJtznLmC2|?O5bbl1DIDe+9;7OZb?vkR1^3UW$
zZ@kKdkwljKL-@j1SWnXDSdO)?MYphXaEd$Hsq?tNjAiSUgm?45y*;IIpfSYL;+&W5
z8j>ZqC+@8VsxQ#*eGqO^U`+`OK#<Sfb#@nx*=Qqi-y*m(@!O4~SP9<0rm+e_o%m$&
zaxV$(=a_;7kr<K8Qpn#7`;1&gca`C#e&g{^7UJya`oj9@rQ**z++)?E^>bh2BJ;xf
zi2S&Uhr{2ovE{lN)#;DoF=Zd|<V%yR<}eVFDY>bR8qa(CTM6$w8ANXqEFXUAe4!tt
zs3cX?D&Qh}eWp(Xzu$%$2p!DKU445V8(A|^QG<O}TCXKZ<2oIj9LUz5Q6R8Fp4xB9
zzy$A868a8KO6W7#C)P5>+;^eN{}o1NVa5H^(o5Fl7zpKCudsCLFj0}2+?oc%j@>~|
z(N+)AjWxgN1a_OmDa1?~YcS$ln*hHfxF#S%V0#;r(foHTSX&%>;^gcU*7=j50Z~zL
zZ0Zif?+EUJYk%`78wVc=q7)M&aQ2x`FL5PfS4WHG2uv}-Ru3A0Q}<gkpzBhSc?Sat
zVD+K=^oU?a)HD;ctF41x5UHrXI<Ticn0L6_D?ox1WzYL1*yiikN0z#LtiqXT)Ykz=
z*H33hBI2T#;v$;nQj)4|*HA^k=S6zIykYyh-QN8CZHdaKsnpeBxK#`IXYSQy=LVOW
zvu^fJ;K<s;3Ua-!PYTExanbSDs_eBI%$uPhVGI2D6nAu#w9>Nj2FAwXC(?VXj~*i3
zHmr>}gaZTPek|kT6C^CU24+TqC@R{oG!@iwAVe%4^7v|_vYZA`W?=}DvG_!jT$;t-
zZfoGjGSx_T(KNcJq)Ztb2P{yl58wcEJcDx+2m0lq-=lxSKl1&XV)oQi62vQMj-=#G
z>aO5ky?x5R%%Q8nw$rVB1A#o7QR(OR5Q0E#gm+xSgoTa{wx+h%818t#<u@GL-`DL$
zO1f*JU>(Z0mE;L^%5i;0Ls~L3k#CNJj+YNgY;CWvn%uaipWL2WdgI!Mii)Q~2XWYu
z$1l7=-zFzRAKlW@4DcicZgpNB`k#KsgM_ZDcToDLyyqHprw2YojfZ#Es?vYB0B4s6
z6Guiy8B<f@a!G6|aw=J+r4$I9Y;%PR$1{v944i#eX%~XFPrO9o*JNVa)?owgn3SR-
zs`8JtG}QV^d;~wRxg#$jSk(%!oD#K1rA8s0Eh)y@jQ5y*9p3<pbuXwktBh($AV^f>
zMP6^ml7i~YH^)~;61Eo1p+iiu_9I}`1yrAkxDVoN0A`q}u#6QS;Unj#odcKLvqje{
zMJL^Mp1y>|tUl%fo9@()D`Qj1Jx2Y%4_7e4lUp9#hbJZ$79RQAZ4T2Kpjt(@i(+vl
zjf-i^LNdd33nrl#)DSX*+{M2E{`Y{-=PaHXHt(b++<E)uOUQ#>or@})iMlVE0&duC
zy0Of<Yh;8LG!QN%vXv%k`b3><ts`$8@?MPS#8BUZ)g&X165(oMfxr-Pa%NQ-?&ZM1
z{gdr9t<{aSyKc#skab{aU=$e>75gdit%WHu(QVd8kDhgmN$qG#I?Endj<Dr4l<-ij
zl5PEVTUXG0g?71quwcf(!hnVhQK3EjEgWHWXO*lpTV`*o{sTJFCc20J@%fz5_MaGw
z+{LZhTd-Tmump<}D-bFp2t(fQu+-0|w>6n-E!~fY(_*9Kh||9e;UYB)$q~+X=HGJ2
zxF74bo428O`6L)NrV5&|{owl2uBRwTh{&h3GX*lBdesC_Am1uxw#HGJNtIo>3N;cw
zr!lj=GwR&1&jSa`9#1~bA3ypy9^nUqn+Il67<yp88(Vuj<;I#y)5xr^uTQh6*xB!X
z6Vtbdkb{GTyU0CX;jqAqzTs(|SDOBR<v*+~aF6sMn5-=H<KShn{Ov3RA5O%yh#>+5
znyy1@^6Q8p0SITfq4kE9yUKlQ!<1vok_AYo;epZtfzm08zWzMehw+;hhAA`5b1Pc?
zFpDWGEo;<m?cGeQ8Q8lyOFsGLMDFAP(UTp1-UPp2S$@3?oE2IuSJ{E%Rj$YR0>ELf
zh`Y)Fx)lm$|Js+xe$Rf*ks<CK#|`%v(q?8RYKz%(aIDZIV_|rmp8B1#O8jrlwa;MA
zm7SJ%D43T1JXskk<Nj>bS-!#>l|N^{kb!%r*US6?>b~Fp<y-T~m^24Cy`EZ{M>KsG
zpBOO{Sv|OzGbGgN4C}xiid9}+N=oQi&`sOb<2Ddp8ir`=D8%nR?ra_nSXn<PF~wxT
zn;jI{ofZzZK%UlEVR*_w5E#N{lJaK|gIfo~FC)VUo|85Eceh05Kkh$}gjO^lXu$EZ
zE=dq5o7YhYQVztG9dMAz-FgIhWz4+jJ}%bkl9T)JNP`j5J&UGFzq}yQseT(D7B)N$
zq@>{k$=P1w>s8CPZl&nDC2bHdt5ao-8!0)^C(qe8xO8G8N|5I%d+!=ViB2EMJZ+*)
ztEru=Z+&cNr0f8uR(@3p5dWzmyrWD<hP>%b?5JOQA*Vttu!ePhif3w?d&bS`D<R(M
zAqlNF_DxK@@Q_NErGgr*@TT8X2h_HMUbMIH^^!psBrs`n1Ic;FM9JrH3Bbgn!wuIe
z!SypOfo)P$#_ba6r}s(g=cCS4|39kUI;^Vg3maX4fFhxQAgzL+bayF;prCYz64G7L
zU=Y$EAl=>Fq0-$*gLF$v-{JXv-+k`g56}6-qkC`GnrqH6-tor#K(Ve4!-?UO9qK58
zL@fV<eDG)2DaYTpB8&bS1`k@BZt*wijM`lPH?Lm{vNf<?3-z?DGTmAAEWFDjj7BQ)
ze*2}Yz-GjOO5g}0ihPKzc=_JxYi{JtWWz_asJD?{Q|?T4PKsqp{Ie8j2%uyY3Kd$!
z4O~?;0gh3U_fS0xzH&$Zjas6g=+{qw)6i#4=D`~(D!%t8;p&o$>y&T7lp`_H{Fv#|
z;jHqb$(JZ150$6p5>0RR#w!~>Jqpao%92&n(vnvfHa`Ca`&RO_#)X2z&C8ypz;qos
z2C+@d8eTrEn>S<clI>w{1jX@^-NomIsyaKoO+{sFTlh~VTTWg;_eYJWpU{{H)1+s3
zU}N9#_x%EwmD~pJcpU%U(e+lc&;5>{x9?}I(1~hYY}9ld@4r6#Kg^qxD-`dy21`mJ
z8ecp<w6<n|h#qMI9_;mNs1wu*ikh8EHf-20rCwq-oZzUhAG8gB_VKJvjmg1BXlT+)
z#?5ej!b5yrt&W7o#1(S;dh;8vSq7Ox$uVQ$8JCw=R8;g;)Wr-MWex&rr0Q($2Jk6t
zsv&LYSNW9@@Q`QaW)l12b8^0<eY%E7zVM~K0i%>>#++<yY|K%y38B%U(OVoHE-*HJ
zy{)YvuL!>9W;>v67)zIJw{y;fX`i#dBn|W?1J~U7DoZ%gJOTQu!jrm&>eZsl5iLEV
zrCVs=fkzUphP*K|Vx1Uyru3}3qJr{L@cp5`F?d^}h-$mvcP@!aFMR{7Oo`f>svtXd
z_WqTz%zHOFbgH$T4`~61v8h_Pg;yC0nn_4sUToCRmIFxrD4~yAlNH7W+O@`Q@n^wu
z+lZ&5?rjreMk2Bi-;%HgpPz<;4TqOoa#|VVZ?|q)%(I>i<=J}o@6#hdpLl1mf^|bK
zWa!#gEbq73##AUsdAFrNMkR4OKPBJCTSO^{=v(W@2u)_1STN4@l(CYK#J(AJ?qN(%
z;Nq4^CFejeOGS`4&0eDR)x#=H3t353_|dz<o)5jbb^hT2jaVUGs}a{_xv3d2jvpwR
zC`t=Xbc`grF{74z3fh#CxoUsI%8Kp(T5`Sg)-^p7vui>)Xh;p2cedQ?NWa+-g9lz!
zzf3I71NtK>{QZQyKed*4|0ZOkjk==~fgmGWT>Qk0n-&3cR_K9ER5{_l?mpL1cMKiT
z<lqZWiNwD=e^CFL1+-wiSrcfYN-l@8<=<27CHPEnSKeqolzYY=8gO_x)U<3R8*9A1
z>&blA&gir)&uq9ovEFrq!}dE>)cG0XWFzsgU1;A9E`NPaj<;j33+vkwXAFXeSf3v-
zG2L(#V?a2$xpyn_uq2+pq^9E$VGtN==@@A3JU_{;Y%(0DgJh3`A<>M)RU^|t2YV7W
z*GG++G>ZprKX_k*xMY#bd+gblmoKPi+Mv<^@1K!>zVr6hx~w(tEAAOxUpMHae|$eV
zC#9J|o7?AO?;HmYn|DGY{@=F~JSgXM4g(AS>1NKalg6m?YHlC@O1TPFP*V~DVKC@G
z!dt*HS(De12uU*Vu$S37WfFj=C1&@}s4=uvB)9K`$He@XW{%t<#MhRUrJ<oIGipc#
z=sL4@V`Re-8ZyIP4O$xUwav{>mrUxAyYraP##(Cp@+TeP9m0vpNtJ`8EBn1q#n6ST
z6in;dv9mUTX_553TNav-toA1#@-UgycQxKfs=m_py7f6sMHbXBo|=p|=ens>s>ol_
zrQSqTrAM<%iBXzHTk$E>cgI!QX$HUg>9J_nSG)Cn2h0`Vk!yk;>6y81CsWp_URHmb
z$HDL4lEo4HXG``v0d$11p&V(T_=7;Qg<a5725KC?cd1@kTyAtGgQz~I-+w4ke#3xV
z<~aI2h)s2P%(%F?^KuG|jm*Nr!_)l+=enql2Cz7osP`eW1Vp6x__|E_`A?GPfy#rL
z)9v1m&@1r@@OS_x{kLw)<290tJl!X*Pk#JU+NCKX?LA$Xp8Z8ot=<vG|EIuzn7#zm
z)$p$&s<kF0(N7^IzQJnAyF3DY&<<(3c>bCSy$&2)K6ut!o=YE45>S#*{5iLmlTuSw
z#zg`gwvMJb+~c0V`BQrqjM`2U=6n5c`y<`KtA9?Q`ub_;i)Tw4NkIArwStWU>FCT%
z!q*sP@nk6TP(Ay^ccw9eC<W{r^6DNGCS|3SNh^p^oIE^9_#o)faRu)akBy<eTwTkL
z=2;yr1sU)i9K`bSL!mKfYbJ8JtixmIo<1mj9fa&rT}sM|i`h)rhlj`<=x<?c(6elm
zl^x(<9~v6!U~i2D+V)0+s3BIL$|uQpy`7wEl4X29#6EfQcK1l=-?V^grg|`O6B@0T
zQZE$8hJm!*2<Ng5^UtW}l}Tx53EBzs(6Gzfh%ik?(Klk0HwB4>S&wAIKk@%CUd`qC
z4&twWBkn-f-5VX&yWpJ@8jXek5ZM0m=ioD2bVWb^@P*G^W4Wd#|JLaJ`!n97i5Id`
z9yfN1DJ?&Wg{FkO%EWB+zeRAz-jQNFe8IAq<37IcpyHZWXGOo>I)ESg^DWA@cy7FW
z8%rrLesV0KBr5;<e#2?+4A&*{YGSi9(QRdGk*-F50tp%UR?DXA;ynEv-|IR8T{xa{
zF-JJh$sn&@_M7cIHpVheU02VYYi@;^CxtUaK6w>dCh11t5E>dA+vwJJ--&M=^C$B$
zUFk=RUq&WbQC}m?C+(ZSaN#uH1zTba3<5xIZ_*5`un%HC0jCqs8g>arm82Q`WUYtX
z6J+OSx}ZKVii6Yx7*zwmmbf?e4k)@f7%-9kc}JGnYw778gm(zy!K5NKslTocb(QUn
z>*;Tqa~~S;aEKnt<i(>RCAw{Y{GA}cA2`NON|L*Ss5VZxSEuWrzwXdy52Uq$Bok-M
z3vk6KVP^-K9Y_(PE-cF{Dq84hZ{EC7s=6uc4;ED%w9HSmUyb&Q_Pza+q?g$XerIL7
z|DGG@oTv37?U#cYXdVm<H(%$=l5ieL3bAtHQ^SgWN{}wI!Pm99^Y84d$r!y_t_RD{
zKkavK)8BdA5B;-{$H~$5Pox%wyZ4nCSSiX6vVZ#YDI}9}l_XyGH$J$Gil_R)*Z@E@
z{Oi{er53=JzVf)=&u>9+NaWqf@F--5sH=bbajaJCMn_|AZZ0CycqvHw=lssz{r|YM
zANMUky+}XB65I{#nNNGSc;MkCNgF<___g_z-Qw!;BA9wKoWPWsn}_2hXAG$1B_;bn
z>c4Rf4A-W3S&qwdBc)!xMDeVu`WBaG_D<Qt$sm1Fa^?OJ+RGPRJ<`|%r!*{$MFkB5
ztrPfe_gDVm(^}$oPS5UyN`{4f-CRuqht`qNkzg{8n!LJ_s<iHQS&x5u@UM|d;IanD
z-h`@}?4z$B7XrIk&|pDH3oh08_zKbi*r4our0RbE{(T{>c%Y94bZ^#uc)(0->u8VS
z34I0#aJ;yc+sV3s697auP`OZ;%P8G35Ob@CzYS<gT--VQ1)n2}m2qL$8*oERUzS2I
z0$wVS6{8?VvY0GWR+B<OK>DJ}cx2V~y}kcbOcyK4pHBcsDI`k3?NZ)Y?U}ty{@1j$
zPoKn{o+&v!iwH~k5S?vp=876BNxCFRUTxtQ6qqUd$Rbrj@`a?T`Zp+umY0a@Z+hs+
zDRCLsjg1C2%0D7R^<ven5_@K#wCU6Ad(77IFkK82PaVhz?5HJVZ)7q;N^NbRpa;p`
zyVF-{wI*tbI_MzI0cI2&MqOPD#+;#R!UPp&s?_(HgC-;#UL_<1DYKY=4ogAx>3RH!
z@e(9A6FmknxB<%Mvu6(_yiZ{{gVX{<9+IyI1_xu}Vv;j*1UsF018Gw-Q{$uK!=vMK
z(;kUf$EirERW}p_O(SF|zmt!$b5a)9e;=$(Lp29OL57zc3gL~eqd{N%@NW1s7G^!T
zbzMFz40M41@or5(Cj%2PJ9{=2U{spv-(#dnOHWRShp=(g%}q|iXtFqAkLvo54g1H3
z7j`qrDS2tx1<La3z;~o!p<ChkX$bRU&V{J(h;mk*?EGB#jPCYsb~biF@M{5&2a`lV
zi^1TgcG6J!72wSE@%cd&k%8stAoh^Cy1U2xTkl;^WX5TXPGn5XT~yD8grJfz-M4S0
zz}f?{xFlERff4hRnNMMGkVH%RxzCgK$)~zDq>!;=L!OsLV-!wn^646Ca;|^Jk_{*o
zS9*>P$%Vd%-uo7P*T7&!Nl_Wi+kCQ)`yrE<rgXPU9wz-m7&~DX0RS;v77WM*P;QK}
z(EvD#fy#3lj5_;A2f=I%!}I$Q8XviZiMoWMG<mdi0{h#kF!WM6GZ>zRg@rBDNhmxb
zWcDE=x(#R(ABd`I$Ur}w7{ci?%1QQe|8m$j<W|dn2p=7T!0WHHzj^dbp1dty23MrW
zT^Sm4#N&a?5G`1uIwn=s)=tioZ@csQkXu!4%v&&}mx+eTJ05dNJshQxN!{^zeBURn
zo+zl+<@jpD$If6G=x^ikhLpwS5;tOv>?EwEqHAMsc3v-_5#l(zOl`*^)uNAB-&m7M
zmp>>w+8$T(m&QQcgz5vj$+alr=fyA{l!7JlvGC(PHS8}|d4-vD_*_Y)-q`CZN?PGj
zI-qo*n&Boy=+^?ES^+EEqL6!nl(V#)zJ0_YlBCnM4tx8DUt9=w7L5v}MN_>&>6I-#
znyyfl)8ge<92y)&7ysRrXul*WDN9DOKWFJCImKrG%xjD++H8E~Hl;d>xv931eiHM?
z3lW)$={SyLW~rtlBNYUGNui1n&Y5#w7*F(`^l$c1j6Qhy%2`JFaiz46xRU476P^Ln
zfKsn;W+^(a(DRBtH_NtdI%f)3mFN()*aeeSa--$fxw+DBXq;l-<JxWK(qrKIkZlgW
zessI_Ya`ewWIdr3?l&JR=||@!4;pH6Qzq5RG!+<d!A6EhOLe<+_|^mh4$T7Q@7z8V
zg!IwO#{{mi$4^0lJZ~<xvHo}GHvvIdm{xpzBvXvfhgpTkUqj{)B7X`3vZ%=Dt);aF
zyYEv;2w|8Y^U+c)7HPcbJ*aDT(LZ330#@fB+Ndqp2YS|vf)5g7{r&x8LklCj3`Tai
zK9iM>$8Di~GtG3${i`(iQaZ1Le=aHYucCS-sT?!o9#tSA@!^T#v9pDaj;5wv{0;Dc
zw)c)s{A%y-8_Uh9IaTZKI_MU`p)Fv3b$ECPN+$52(UZU(8Xd*=f4DXI`F`b8;eA4^
zoBl7Hk5@?v=M0UE{%pI2!IE)yNf5LxhFUAE7rxNY*ifnxJH3$|nwuNphl%0lZ!gmO
zNxH%9>Gz-w0ks(6#0`U_or&~RzvA0SUEph)Ti1nZ93&D6bf^XB;5(n3IXihoe(dWS
z5@Mn*EvizyM+p8&Q&Ur%aJ;1DxxQ(f`R;WoAI-_dv_HO296eqQ0Z1Uem-9{>8XMza
z<IYfG2R-CJ_Yb=)il39Ho<6bu$;nWbS$F%Z0BEEVTmbg@Nv^6P?~l&Y8xm%cF5Nge
z)h#CQ!2%;R;-mZV=4dsadsP*)R*m>`9tJoq6PMj6Y3`!^do9d<XlR)EsS!0oe$T{+
z*JCMxeegy2-0t5HWCzg+en)xbXmcgk95jvDzf;YT8KjgOciQ#<-w#Z=|GM61$NE*h
z(p2wr8yM~}tPYF8L^umq*NOav-u(URb>W<^y0;%3iJls|WD440gQ*6af{t6PS47Ra
zQjm`m`}0{OsSh~y+by+z3l&@Gi8)Grb{1S@bTR=Pwn>#<W8+YdGRC`C&&dUC_J9Yn
zdR#ckopeom{kv;C>whnGUrlC~+@vcGd(JByaU2}~$U^XYB4`cj9~yGmJK&QNOVx-9
z5>aOkxMG3B5PU!y6$efd#p%NzW@g}#O<jqkCZQnF(iOT}Eai+xzR76jdHLI55Z$&Q
z(vjdjpbIO{#Cf!8;3q3OPOAu&^z)Bi|6T(&IVP%|s@QjD$zcbXbuj0HE(Ta~cd)<%
z`kSypUHWIIjhC%WM#31Ur@4{Le{=m(SR1IGdRa}a&VM{Tyd3!8`m(I7{mj99+++1G
zl!^NC**-p#i@T}tD7lwuUfsU-biUb|+6Q}e@-&5&wSnhb5XOHuEOlwlikB$XR6lDL
z{zrpvrqATS$6Ajv!n;>v*U(QdjBc}Dekv&Fg|~t~oTa3Deg+w27e<dSW8ulWpgY5E
z^$5E%qsPMj_X=o40oxrz<twAd|GT2_-^-hC960a(_hNcZ8y=z5UoWJot~2Zrj8Xip
zEmy2@_kCPw@;otKc!xeCH>S@|5;Bs@&x0EG!GxN#f6TOZmz-IJF>-Zn9W<?%flSAR
z$w7U@ex#2G&CQwe3n>&7x}b3I#yIWNmtPHHz++|SZ1SkQc4Wr=oIZSQW6i|K=zMkb
zl(;{R>@?rV$f&)$cWECTlBygkM%RaW&c)_zoOJ!fReVnKF|Lg~3G-}AHM6+F3Rxs2
z6?dzD7npwaXno_EtCy%k_$Ql@i`*w@d0t+^#o1<VOr8L6_1}|dEN|DRkJxMZ`RVWw
z)hk4LbU0bQ@%ZpT=w$-iI_8HpqyJqCb@bGn<wlAy3EFTF3!{4CM!ZNi>RQ?Z+5vb2
zCFk^s#l=Kc8=a^hO#b?{x?R4nBbXQ2Ldk9tRuMTm<ipc&T%MNpj#0a~cW8R6dA`1f
z?{dt|M_0QU@&WP>XPB{t6&Q)h!f>+E8jtRD<rpp%?M<qBCy7^FJqk+Tbpg5%RL22S
zLg%J!0kMMqEVrg=Y?Pz#eCt}mq2!ZuYD%5-ND?2&Ps^$jS5O%09DHiO5cin@63bta
zPvmy{2PAvz(e;5@rQ`+91?}P;LT2@bpr7+89s-{#$BUh&k8^{5E`1TlId!hl-%I@6
zqO3IbHEM`c5_}SwX$BL^-Y?zZ1mTROVQ!7}&&c>40)+}Q&FOTKlcHELd7FtrC9-}?
z4vrxWkGK=}z<yNlB{pCmUzx<g!G=JtsQj7LjV-6era}?*?#Av^->iDfpo8@R`-_~1
zBI=Q&?C$HEHT;)>U6~_`vc%Uv*}DoOcge_Z-?;YHO=u81<}TUoVG{>jO1_g-DZ<ly
z+Lm9yxWc0n;VAOhx#Pd$v9lrc{_i5ZEpV>_$hot)maMAlV49e?Z1DE#?7Q9nS-|;E
zX~Xx|uQ}Jux}B^(|2}Wy6Bvz(T$~+fmYY4%ellOCb`#~zz`5E_A8c7$5MeBAZ^h>1
z<T`F$Ts+qiznLum2;y1YKTffUZ3no$e9b#__D4U_+GDMGHaytUcam9K^CR8)f*ohz
zBSI60-dG`-2che$uCAh4itF%B$eWWGnV#47g_POS+=laC5~lfCE}f2uuU%`=VcEM1
z-PYpg=l|v6Qsc4n1!7}+=dI2THo0vl#G^GYpFL};gpU8uN=xp>Ust7RsNJT@w#KS>
zc2{-u^ir!iGID6;)XUv-BhpWLnKl|_Q+~(V(>d})O;7R+kaP>3;dwL?8L{mX7eZ}&
z?pn((@$L5{R}0>p4%4~L{AH@|$LFt5n=Yp{KB8T`j<5eD|AZ%-ii*q3YU~w7f?#0K
zm!OLQ;@!Y%=Y`(tmnsV>RF_6i8n81`zc>AR8QtL}?~fk2dP{-x;pu?;v9MgSY@J&a
z7$!a>9A$9qSla8(1D8@meWSVU<&*=Bfw3V|G9pbdOKU+LC%3TBZundsJ!(kQ1+W_(
zPh&=8dST(p(!4j;*RCZ%Tdtwb^p1KfOR>;1mzNipl$OgSjNBtK)Yrecu>oFzTW0x6
zH`6%2L7j?w<dEopKca)3y=46_>P=e?*zaC%CT9_v|4qu4``oo;u(B*gBTp@chUYjo
zsv)=Aa?U%UQ7rx2aNRt2^1a+szoKs%&I~)ki2kU)=`Bvq8#IP!bV%0iQTESD82WB6
zbI=jR;ejd~!JA*Ala*5&Rqxk6@`)%!KG!DS<WDy!8Yk!WF*P^HHZf?*&CShz8y)Lm
z_3HaPY4dN0%|b=l*;)B0y=}m*W~_#efI|TqBGoeDD?VsMMud&%nDRivtgK>6w$$GK
z?qze$ygzGhj=~T-6jqwrT5<r@!ZVTM%r=L>$ia4R4+jTlO6zJdT2)c=MQ+;0`Z~l{
zCMC2&R^s(f7^lt{7=gu2e45PuEd!&Y@(aNVn(`Ux&n_1|_pnHDeJU$T5TsfYdYCp{
z>+Wu8#qdMcZ%R+!)<d!qQFL5Pf?C^(MhxT~@+H-?M>A40__XDgR+QKkRic+aBK!iu
zp-53lNk2_!zXbnBH60dq&VEDNDy+{>?TPclX8P}Ov#|{(%adSd6ltdc<8EieJPZFC
z#k8rG76rA~$AfM&%EI2+F^!R><g&aE-}7wj9caQk>@XBG{LL;84_%Qg;CSq#8>#-;
zJ<40SiHMFI-<p1m#6|hU+MqrTmF#4o_Y+Ua4^(3EkBC+A{5fAQI6ZupQ|IeKKtM)(
z!ssn|KCC*o*}`CL`5V^A^WeK-u_{)_-xu<0kI*sxl9G{(mZ&~_xCAXK?08uM*$x;8
zLS7Km*Vo>iEL-MQCm7>;-o;>q?>9Wr8P+quNK80Bp*lIixqO`iPBmD)ikf>Z1QDs>
z)zR6d;U7(&fp1Gv(YNn;?3}~LR<=rqM2e3Ht548JM#@_yykV}GPZEE6yw<U7g)+B&
z(RTPiZY5|w%ZCiU$jI*ePwKdc_%2;r*Lop)dy$zSZ*Om~)c3|H_AgBx^ezC&kd(Sa
zq^#=B>YGdx0u|*~;V&TLL>X=j^lT48`T2nO_B<!o6g*H2*rO{meR8l7QA1XUd0(>m
z99W*deg@W-aJM599_JCG|2^U4pyMelV5a84MCyd5MY40t^5T@G=t3ohwJBd%@(#~~
zhswbBJ7q75Rdo%F3j{$88=|HpLxVr987FM*0PzJKT`b^v`{#IHS-1j`)nR$j)jl|q
z>|9mLePtp4Xu@)V0Jf6ZR*zHs!i9Y>NPL!p0SV3{u}hnaz(sJW{_^erasfi3BlcTR
zq4YK}F%_C>xQ7-d`ydo-@*XR&ar1qHM_{05@5n~Qt?Ulgvv&>_Zf^C3!DB1-Ag0UP
z6huDz#BL-e3>CsxNa6TaT&RtPI5)xdB26b-S{yOZUqT*DX~s7$r(V?mf6f8=2q<Kx
zwaDGe?#&Vt)3reXUt*)~`n1G9T+hUrcA{6eSjGHTOlWlArM<2FrTJLo>jq5`I=095
zOEo<|PMyhdd30ArN3@piltDEc5i!)?NVNQ}w!*^7t|<J40YOI3{ae>Jl$mELHh7A~
zQ4v!W#_ge8GO$itDx54!4eJ#(wBH!CUWhR;n%SF^TkNgJT(aTw6JMXqjWy8%aAU(`
zyhFEipmDpP7YuCVFWQGPjL@3sn6-kb6Sjoji=TLHtSpoRsdBWLi#N_y*J;z)OWgO8
zVnO&^b2gRqTHtU_fJ2l^QXtBT03`i|y0i{EB39fBgC)Q0IW98Dw@%2AJl|<6zO&N-
z%CmQ;uazDViil{q^PYSnCqz6kr#3y;Fl#R+O0v~AbH^HiJ!mwG>mfq5)arsq9Yii-
zJ3oB?fF)RkM{oQ0habAgpZ&v2w;c!Qf$tH${>J^>=}S&wIBBzP2g{_RNU+=IOD=Bq
zsWOLZlM$W8rg;K3mfy%ddVZq(;{x@XB#+sIPP7u6y~w-?&+apDrxns$zlO|NRWpp9
zhwZ*SeQRi1>T)LW^5%XO&(6`N2tFx!k>was+0E*@o>(sP?)GGz@V6bEhZ(f3IehJe
zSS@AaH!s}k9nGiz83noau-IV`Y++&lnJza#iy_diu>nc1517$Nz3dZzM*$7jnWH7T
z=RrVTd2Gela7n)!#g9p|@uDx1=j&qv(0x`PzX2ZC`qIs_!)+8qKHKhe^D{Z~&cVP1
z5c)w@pjBbCJ%{3kB1(fV6DO=fV^ZZjpN1Pzv(gj_YNu2iMYnkE)_RC5tkj;aZC|t*
zuBfTu2dfAPg(w9>mUv;v(YDJPc6W2utah)}iVe}JG*`4Og{w?(!iaG(Q+|B!l+Awd
zmvMM2xVXS)&erUU(nbMZE*Z!AuKD2c0*Yv|X9veJT2lf=EVtQA%hE3Lrfu0CevPc`
z4X5s0ungqm$Y`dcsedgkH*H|5$K~7{a_Q#$phxAYJbQlF6GK*b;90b)oeo<m5Xo0E
z&)x3&WeP?~<O;rDqzQ+^%FWq9qt>SQ^s?Id=V;3BL8*o3!12|fF=?8&^7lv(vXd1#
zG=z>km&Xk3yq;WPp#Qz7F-ME}RmwMIkjwLKWznsQXajz+qKcAho-g?I#@7QMrU%AY
zzjmyI8l~b_l-2pS+pBG*D=h2JQtqRjVPKx7GD(t2J_xP2s=4~UAPM0zgEdyCtn$AF
z{ah<nL>KN!H7&rt&ThCoOq=^s^2K<OLDV2x0knAV;7Nq0RcuW}B}CU-_mn_bY*O;O
z&8)Y!hOf{)$triK_?@xp!PEU5XxY&4e|f>{ytlc;z+*raaCmyO-e5A0s^m6WGD0@n
zN{8y1?Y9dd!NV>)s7);lJg?%beb;|1?C!slA>(poMc^>hI&*J4p1Z*h$Xg9~VW)}m
z*=^b>+U4z4YcwoJ0iz$Jb52#YU&K?CHY!``Xp&1gxwuw(<M_?2JdltV>5~o@M7Iyg
zw+6??CX1ZajI1Cxs-J#^Mf<a0kjvrfge#RQ>`~x@vA!M?HfSS4#j^DB90*hz4x7*V
z2odSK0=lCI9Ws#)fByI}0590n%gg;wqMO%_Pynyh>ic@hPnTO$$<wXnU0j@;!+5kR
z1+&7=dwJpy&=JsMxE~He>51y8$*j}xZYb9A2J*(xRPXJj9A1I!qo8<x$CGs?ec$JB
zv9Za(wVU&hr8cha+{s^=TYIyN51ISD`Tpqrf~wv6X8L|qApmN2{yw<tl~yy|jkD1g
zi6)%>|8>c`J6^q6rvndM9W~=9Q10D-{E)5?xsZsffBz7i+)P(+UsKIozj>pxhoS2D
zk5qViyr7Gms*VUO)SH>Osh=4?Jxx|EcQ;kzGynZ76f&pdxt+-4ziXDe`#((<WteA=
z-h}xVXxKzK;>6lsrxF}*i^xkEcE<>>?M^eawo3FRG)lgb!bGwaoezfjsx=$HgE1A;
zUFL<9!sx?WSSooJpIhERRO6rJrJ;Hkhn^0GU&Sk)IdPW&i1=VDYOMm6>4Hc3ec=!B
zD~c~wj^^vl#USK^{kXj9?ZVm(+tf&Wt54qs@Jcfq@VM|!saIMHcaG6KEB&5N{gSl%
z)y4LDCcwXD_$n<XrQcEi0msSl;=;VLIsH3>A;XnJ+jV<jxd{pt8~nb%kcvf0ft1+p
zhi>PApQz})UlW|94A8q-$zBOh@)V8?4@=8*iP}=i;IIAKaS6I;c<w)yH?r0mWWD@j
z|Ch=%zoW17`~Ahx*=GEX(;fj3E)OSi*-p?5Q}H`uKacRP#8DEOo9GcRjbw=t(!)YD
z&lmWydWt|D7LXU$K_DUGVPkm4V#;6?0ZR<sQ(LG8JVjvd$Sx>ALA*r&p2`S%iLOqU
z#*&hdZ<;gEuOUtUY;`%FJ-8j)1Mz0BaF5ud&DL@P?J#VIYVNSzc6@im$@iEKoI6_E
zsbTztSe<nCq%D@+(0k7KO(4~BXB)YP?-!Xrqo~j|QEA$>Pv_VgM0&)&y>-$xHq7&E
zR(l%d9{nl$>9LzGmQR<hwT+#vy@aiZfli<c0Si9YHT1v9=!;IJ85C!3uiK8yMIOt3
z%tAp}7G?nBrTA6a^L-|=Ew|Qgi57txou+RzF!y+Kijd?E8gk8*!0TnE45j1dv5+K3
ziPa3YT_eW97CHj`Wxe}GLPW&U_yXo@%9+f<LdWGFRFam@9vJ@`SMSm<3Dj^<Q;+G3
z6>^)&Gv*$RT;EWyvD%*AApZB=PY6v#XSMeYUstm77_fc=&TIxp4yabhH_cGqH1~Un
z$dKS64KoQjzaO70oNmGX3`z9Y5z6PnXEE7xZQR5Ci04Xi;Va5Z^ZB}Htk1vyuI-~H
z%U-BG4@zWaq$d&9f3tY8!Q2}JoRI;xz?v@e+9QLN#lxMUeOSSI&w({r_^bN(9%oRZ
z#)}tSQH<XR2&@fZeyzq=4eR*xj>J3VAWdeGBA3|OSpTAi%O6Xo2%MKS1}nm-305bY
z#qF45lb_IaYMl3<O_3cw8`(X{9kradsv=Wnsdb+Jlvz6oyS2rv%crB1HvQS}dYa6Z
ztz~}j8{^%;LXdY#hnHe)1~w<v1dThB<<DrNc0}yRWs0Dv2J)*T7@eN9MPpVnJ8n$P
z%ecb0BB!L4m_-K~j_v8{@gGxpIW-b$V`GxiFNAE)u|GpnvQggvTBCgSS2Um3fM1|C
zb&Xl>2));RYjHHUoE|+yJ2J8hNu!wsjO4h;gA`rxa)bML^0zlaq?ZEPoxckcOldEK
zJx<ichrwijLR`4YeY^Qz0P;el7OK}Rq(HM#K}m~XhWlV|TrVbuCfy=kN>eKiSibO9
zC`}y$l;&SM^fer}7WUr9juqBfpKwC>ywypKeC#>bUF_uuGDlzW-<FLQ<3-NdgNC;d
z_?He_LXT7);cm_xwN)>`-DYFaNvwa5K8#Cyhx7D1HJ!C*?fcXAf_yT4_VLTvVCtT(
zv*xcnJNV$e^_t?seDJT_>ct1&ASb8!J$nU$1%>Ax0=Ao;+1VX{FD43k42TQx3Q$Z`
zqFqSf{hhcIXfjG{ZTTv((JZOZaqmw<X&WAlO+m|gWv$;eyTgPrH1OuXS6L)Rv3Qa$
z_j#f?jAMI8Y1!AE#Kb<v=9b!=fmV)ks_xSRbU4FNMXimaPg7R0-$CHazk}w-_Zf%v
zwBk-J=*yvZBK&d$Y>pmj8QK7nC+V{EMiKCb?*%FIJ$mD~**JOUU_yWxY=yo+nX8=E
z*qN?kb!`_#1^?=WbV6$3`_VnicS%UG*Js;$Mw||O?|0Vxo}~lR1r_2<_j;Fav9Zah
zZ~~8x`G0MLCmaG16}TRTJWURfq&a*r04EjpyOG5ksfkndUn1o-S`ViLz=hvdpDy-j
zTcY5l*a}+HRvjjiBsO@X{Q6U$&uZ5yi?Me(lnmuQpphe-k}Zhr4+ei`m-}IU+ql*8
zd~__`>{SN8ldC!N!+MwY-gRr$y<1R59<gMOkE=7_@bgz#3fRdu2F;sKe!@TzwU@zx
zAdPF5uN~p%hfW9g9w$E$;(WZ>nw!hP*^g&Qpx&iR-G|EcYeF?TIvUk8G#Mr5fWZG)
zYC)pXl9=0JQ5KrvXzA8>woY|5jynV>Nw51X%y8jajQ{L;2m$6Tx(a3GF2|cWvXA;7
z9CcOk==!B8_&BWFdhzl9Djg?(>vpk}Te-}rdda?Rt_D=7IPxuy5}R*IBg&D>(8)1!
zXFjj&g?qIBi<huVOd<RlqDg598!`a&lze8gh1*)l-1J!*zwqpIPfPD;OwT;r3ryrI
zk%UgF#tJSC{$n6AS7(V!2w@`n@`N+Y-NVhi)C`$*A2Q6=y)Il*3AW74p@9q(?qc|4
zmwI~YEYcTENjT?YhSE=Z320NAe0E}?BXqjjl-nOM4^TX5Q)wYfz}clHHI5NnWmesz
zWA36gm3|{eSZ#jYBQ|`F=cuBO^+^*SL83!svxgxdGMI!-A4>ehYX=wPSkE-vN@7(k
zfBOW+*O>0?Bv4vl9+`<vr~=lbAD)t%n@XP$_V(kM<IZ5QW#M-FL!8fwDrkdvffQWB
zuyX%~pe(=3u(v-4V$Y#rzDD`tWuQNp)i=#sKe?pxQ5M$u!z_FoXr<rGEG)R)_Nsv4
zgttDXkt_?6SzxnntjbQ7-*{WhXa4gkU?w`P^%0UZf86><M`7cCC7VPgAI9wv@Jj`-
z0JmCJcD3+6Vr-IMc!;j<!Z7h643waQ-&&g~bulz$=;C4^D$uTwJt=q+d?T?>#Aa>`
z7DgZ?+clDiyJRH7HT!M(42FPR)C*3v>&s9U;@Z30{gX4@0Q$6-72HKU3+<;d6rb|*
z54(*Cem+HoQV|uAq|5a-WMe@|@<^nFi!MBq09%!2tuEW(lggVnom^KsJE+fK*l_>J
zB_gYt^UBu8&Re(mE`s&th5Qf)OqIa!y9Rc)lBaH`jIf*!H#=8u+Sz5rPgE}Ly$9L|
zU~~`YNs-n7auY*SAxDX!QK=V#(^@EVGe7B$9JpV+W)P(G*!s~SFxKl+c@yHOp4?=W
z$)TW5Y(W)GYP6J;6@?#Ic@eb<jlNJ*P1o36$2ur2FDB)2=0%ztT+tG7XtVT^g!SH9
zoG#9cka@&alyM&K@4`r)kvq1J7!hqAGD22I(^G46U&0%qDt6NkP>}FvB?1Q*Qcy@g
z=<0AK<D|)!MF{JzqNtokO0?Bo^Zbr>2Ixc$uRw}6#o3<yKG}(e*9rgI|80jvsFbe_
zl?4QTRWxLIFv|PUz}5xi9Sn?9?os8=t%nZ{YL6}wVpWo`DaZB6Th)okkOMn=xj|*6
zeVfI+w8BiEln@wB+$CM#NB{7Ge`w%)g5*8lB^y&yhQD*4uOaN!ao0mazLi|{TVw2}
zJ|PJ6Y0*XVEHA&KE~%35FJXvy+8lj$mrNg2C(;<<x*+)T{``Dg<idS%ael!Nmr5ix
zGglx<r|^cJge$m+h7(%9K$_Gu2Oc!Vn4lvqQ}`pfMzpJ>lWLy>kI9|bHYS>%bD%W+
z?6-HH>sGrQQh%49pPQ;@rmx!ZrXj%PC9NssPQoPBVCl~N?^K!mxKtUP?M|63yE^IY
zB&XUs70#*aM|D3^GWJcf%^$hAK!*0huDU*ZunJvx2kb77G<UOZh-6bn>!^~yGh#Kf
zx&p@r;b1o|_io>)!da9ix=6yrk6+cWF*uHvOwG;n;ZPbd>mT?o%u|RUMiBwYsdNm%
zfb*MixApfU)%37KmYWi!sNZuYva#7$>Z7M;j_A}kIlxt<Pq+1P2<_9=ZuHw3$luQ$
z+u`K&1yVt%=v~30on3{M#J1KzQntU-XEa(`@vt>NVUi>;G=3XoXa49ZK}`!eGak|{
zEcoP6>kHGttRZ?4Cz?RS^OdFMVB3Nf!fNt2$?}`tuhCYM#iA!L76zZtuCenH!!}yZ
zN7)|R!Vd5EbXNZ;(%*s9fVkq&`ybUFnso-z0$cFd(&vR@<|!6B0<bCEUP&`)=ltw!
zzSWN@IET?S{K8-M3erAESh9bU>nbWU3dsHN)q*{mTo^k)m^v|=^d3|&*(^CJptXSB
zWL3vSKtqXwd|f48p!|x)B~+TW(RFG0P~-S(0#~QEO;IJ8U{W72fGc-q@DSxlA~DQE
zGCs4A690+GiKJQez#lx;R9(iL2rHY#X5x!V79?xoru*52YPUac*}J5#^QQOQoM&gI
z!Nj!0#zv)J5*u+B)?R4Cwj85XbOI0esSIaj_a|?ylBo9XiJ3Gv9?TS%mb0<;_qBJD
zz~|&lFz=~#1DUXYVG4U3MYygx;%&qu!jULm3SpPxK@#l0W~|9Z9%c^N(hl>3<z$YO
zS?V$=sykWCODb)N>!<s^Z*_I^%=iyow@?c6&VQ&p30HnLi}wExP}OPVjnM;lXGaGR
zTw)@8yNl8K{yEVlGacIS{E4E7j=UHFTa(~|c*^GzJa6Z45GtY}n!#2}@sYxZ`o$@y
zC%(dSlJxWV1Bx<Av$9HBrSBHF$#E;)_SLCg2iIc;r;M-~XvudFuMg4~Bc7^^a`}?g
zS~x~`E>8SaNTT_yEf-iWt>e3v3N&ky47PHl`aj{dvLN@ABvCx2%_2QQ6lZv;j6%2&
zPtE(72sy4}@~oe|xyn>;<5k7p#m^L9KDn0%oc3+bL1xjPfQ*2Ru@Cgx=~_GKS27-_
zjNU(&`1pP?;K?3gl9zQFQw9IL{pWqOW`o{7(2i`E9MI4{Du4a*;xdW%?>+i;IbQPZ
zDZg2z2$gpk8OR$;L$bSbv0vj_->}Zo(Z<F`WALQYJbm@HaQ5Lph@dCIefiwr>)7aM
zYfe=fA!25pS81V{?u^xG|Jj6xgX=`x&zt7p$WEtt@%5V7uFTrR>Craj3FGoetfRL#
z#*foe%4z<`A0OguPHr4jcR{}dN_XFu-vGdH)6sAsbGuaZO4$@GUdp|Jp}S8_QF-pY
zZN^`seyn1l7#Uz8yAj%{PmVhV)_C8iNRT3%tz~%rg@XbpIdrdcay`HJHLqb;yox%~
zd<PYQYUN9LBe;;k!T0&|HSH?b3)o_fqP#I~BTy%RY(|{gJ3s~-2m5GeCZ%S2@TnHz
z%IF_CBqBpp%m_Nbwt#b^N=X{-fg#^SoXeb~!5$HR<HI@YX6O;);>kzzRP`<1@f$UZ
z>&KxYP}Yy9I4wxv1v}l2ijE*S=DwT@1x9p?7{-ygAQG&=R3{aZ8}#=jKBce&KRtg1
zAP^!l(M?Oc?|O7jJ5aJX7+7vS_I*4t0c!7#_D(0_#BVVPp!kSPh;}>r`-JYf3%-;X
zeT0R2<qx@M)ft7EnPz{2JCojh{5k*hBN}yrgbxNLy^!}Hdx^u9gXC+K<NgK;m{0jD
z=ZWzz(OM8W`@JMJ`1|AK`A`2l9VeMQf1@zMCmpG@Yn+FsN1Pxj>D@cp3-17>jB`_0
zKaY!#PYjwvP&`enmI`;vG0($ULi$8RF1cBC=>9jlw-iOcz|1Q<kDuGE3Yqh)9b*Ae
zt_I&r!t0+po1$rHTmQM=BpK1imc$3*;QA0?ZOHL<AK>kJqcnB>3>3Ww9Nv_O+!7m7
zKm%#;q2`#|Ao=S5vUAa<x;XT$kAwLm!WWQp?fN%NjAP|roCsIGdQ@UC53;yKob*}X
zpvMH}1e)y^56rk%=EF0qPd-Dvr&VjSyOw%}a*wC6s0C+oAcF#_D{%F>bdx7UN8-&U
zc(~6nmp;gzW-2+F<1itTZ)g|@u^PTnDIDP=&FpN5{>UQvD`6Cb;Rn$)r40%ComaZ<
zc6pM0-3`DJ8N##Hh;4?GR>H8Ej<(X2V1k=*5k5Kzzju!8L)(Jf`!^V+o#0$O@Vx<~
z3`s?4e2;AghMx^9E|NZkZe79~1M4%Y%rFJTCJZ0GyzKmX!_ymQ9S>=H6%S_<+Ub$m
zpkVXNC<X!?kj*!i0!5ZAa`me%xjazjKyJ-!9~lDqhmhX_L$UI5sjl{?6l93idqHGQ
z<mSQ$>C=~8^Z|i^!0hVj=^5lG%{5h(d4a#WV``#Ho)|)SC`g209Gn=T=)XytEmt_p
zi_^3th~>mZEhl=oI{)H@)PhyDeD(o+Tz`#zH=t0!kq-TQ=|+~AWQO@S>6UR|e9-zc
z911Sue%~99tg<C(o_hT>pjxhP*Lz80V&Q0lYowX8urJrHp%gsO_m(y18tQ!jZAR+-
zlf6(#s}|wpQq@occmx&s!qxSLRp5}E(`ZpVTXB7DJ|p>wzAY~=Ib6D#?=#FC0+nTP
z8SD8c=g+3HOQQybIron5H9;k|chesb4Y+{iWo5vq1x?Ip>3N+H$~`9Po6oeWt<T(}
zv_DpEPx%~i3)3b#4K5KAp+ON?m|Lo##)tFRNLCC+Cm<=HAsh{+;4<6D;36Ak#PqB?
z&$x3-=jT`(0RsIXswW_SrPdmi4PrYYUqp#-B0zkCieYJS8O@VjmQ7880O@Z{fhD=*
z#YvpK7Oi4AWnESoD{OH7y?)XMXPYIeYrvGPf0h_owZkBdH&NR>|Djv`G&woh>j!JH
zz_&}4ZEDJ39mw(l=;rAMw69<AhsomOHZLw*A<JmbM*dOEdq07RNfc4%cafURDvt<B
z==~D=72*=21HC^u?u5_%RLFa0C8|O}xVrOuQ&my%10g5YNn6x=bqQ{S)3+tZQiHRM
zUT0(V;f#KO(?}4FJXfPU(YFEBq?|t<9N#E#F#f$-c|;%OSsL9+HR!&g;cWu)gC4b*
zteQ}U{A1cy2`KzGDmRB8KsFZhKV_5zH2>LdlR6bUD9FM0o`t>9ixL=p*=-E7)tN`V
z+Dg^aE>JytX=L>Vvfo^bP++L}9FNL%ZM_Hm<ye^|vG+T-RBxQ8%RM@0O=9oPyK~mT
z4sKvg8V$`2qm=%)aanF@YH(#{X}B0r{_9#8wBjh_u&*md`IqX|a^EkUZAh#&N{{j(
z!xCmUJ~uivU9kCl4(3=e*MyxK_HcZKL1d1A<qlW3pL!i@4O{A``ULaN#VWaNvEzm-
zp&ZV4f411X^*{Qnsg(Pv`+r<X(YyZsDVhIV<-w<I6-@M4%;Dj&?X8j?iLn}g^XCtO
zLe%32=7Y7leyAIwwW&L&h%=*m?$vxrOl0C2;*MVx!oaNBsCKkM+{j%X8C2P9DR*~`
z!DM1uI+|kB9E35^c_gIdN%`saU|)m4m`=rGG+8e*bVJR}O$gb%+`1Jl?EpbcJkEa`
zKDLo&HnWo;W>@x&7*apSCiR;8pFr@2w3Lj9&2KX`r)O^Us~tl|WTst5^KWIa))uyI
zAi)XZj~B57DJ@NO1<m((x!M+X|8@|xZjO7rzle;7`3c~0FrB6%!XGZT#k3PW`@;>l
z6-rwxUUIB}6iOzh65xqFjAw((iS<mNX_n{y<8;HyUgV9wk*_N=5wpfnwpeX*8cb@Z
ztW&*{<HU6~%%R$d7bx?0IynVq$fP%Vqw{`2R+U|Ez~i^~z1APX5Yo-rUc?ccCEJZ|
z+Lj_$A~LMrxZ6}GjI{Lp;PNK6cD46l8|RBhXlxW7a{G1afcojF(#v7#+==oV*48%N
zER3XV4l$5Z<^D1p^D%&qqKl>9Oe?o{*Z;bD&A|);W^Qi6i@K&>1D#Z37RluDckV<W
zR;#Po`%wiLr)Ry1LT<ov!^BPeRp-`M9t-qLHLh~TFmH)qi}(S*nS^a#a_0Mw4K7#&
z%F4?nY2^14qY8I<_ef%U9y2qAcP)9LG(ZamR0j=*Z8Rk9$-7cMqma$%RMVtJmgv04
z_*{AFjp^DnPsjdYAW4m{Yq^u*PVoZ+Z<)fS)^Y7%dD<mgkstQ%|I*ajg2)}V2S?I#
zx$Sl)9#eCGonfj+VuBXGH$#iB({18A6wn==I=fWteQA(5`@?ET=X3rc<srjZEw8{c
z!N6rJC$M6X^N5VfbdmMc(Y09zlXbx<&25n{Azh0fF^^8HjvHMgOH<JhLDxM+MxwTQ
zKSyhaZk9*<_S=I2*#vx=+cRj-?X$i%)Vtieag8t|&0N*CZM|=8thNm}O$!8JV`fv{
zZ3}wZ%|N7c*cwLp2kL+e&8viZ!o(2ZWf(&nr_jakwEP<ef6(<)2?n_m-fqz?1*WWk
zMl2}Cvw_M?5<L!NN~gbz#w)n<_O9-V2RE^vO}0&<A+3Rwm)mZ?!4URtH@%<YDb4Tb
z>};t?z3)hWtzjiWCd;P4d!V0ka0!X|OBWFPEy^$jCUFms4nEa3?##SlJQ=Pq=(x=n
z5*E4k#xg2C+;QOuO@V-bIoNUKQV$WWp4X{&@5;-;@Pw#V007eEAAy0eze*mpX_#?@
z-QOa?p=z<e=?pwW>?M(f<NEqWikg1+{x28c4=;O6TCx}W?`#rja*D10Kn2%^d)wxM
z&Zbl%Ztktae)Tnu)ixIncHJ)pDf`andIJg^+nXBxf%R9cDNly)g}#8TkM!;;o?^%9
zF&C#44UfvoRNX%t?rMGk>*}dzB7NN4?BxBGV?eY7v^~5t4GUfTS=;*BdWz~B4-Rmk
z_mK*x<yz?n1I*%IaXWM`tdg#V{q!C}kCI;!@l1wP#91^kp~HCXg=wV+pjg~63Iu9!
zhiMCbobw0N^9&i%V;NGSe|%aUoM>rE9+Ph|Xld+fCK481qh3H;$@U9v(C#8&_0ZVc
zyQx!e!P#>i2ltP0%Sx5vgC3Sg2gm!?_6|=q#2?7PR$~>?SRZe0mE-{=S^pLXYlo>K
zXOd9KJcH8mvUU{r29V0!@ozz}7+&7AwN)rqWhC;Cr=WnU1NFkKN>CfL*`boErLJ$C
z(0c$>hYbZDxJS^~^%s?dCB*8yG1KQXKF_jF^-0C+1e(^%o2A=Y0-*W;oQFf8t#dF1
z{z(l#KMWc`0SBj$P_4asg%xvYP0Y{viE(v6R6buj;rOP+#l`rKiB_7yo}06^W%hH@
z`?pmU!0blG#&;sV0^q>Q$KM^xmH&L~)w+h5#>-p<pFdEJ!zd9L^h_noK>CHT$>`#q
zyM%pYLPkM+L11~%!&6%}EAq`j9sew@W_y^m?yTN;I8{P<q!oO4%Jl1eL9i~$jmH5k
zQ@-g9R&KfP;(cbcESfKV6~!MDwC-oBJ_z4juN$B4FW07DH=CU4`t@_XiEUUg>yye+
z%u#`-)}qO3S!P#-Htw{Xu}!t>@#W76^+o&K$gJZE*{3wOj3W+uDSrHX`axPI00IjD
zjKjoR_5V3fzh+mh2{`BR{R@SMxS02!pj2>=7PBmvs4*6+ni;>fFfhI!)zzX~ob2@k
zFfUlExKRv5{`rhdd5lc@?9r->+-%%O`@1D&tayL+=l9mZv$R@JyHQw(J6~%iu3Pq9
z2JNVFq-#S`ST7_;^uEH)>Vur*wT%sj>z4kU*P*%w=?m=qsq)#B+Xc7|6<l29Bx-#4
z*>-!M21NQ!q@1XB-wXP;yz1-5wIut3y2O10+@OYO0!i40n?w`SlJDQb5^yvLUJiMi
z`~(P4cgi|9ryvJwFAUw*8;8WR_xJ95@LvG-NzObIg82L_$Fme07-eQ+=<8t_9vjoC
z)xC{HHx)3-%EZ7@SyFaaFsTrxNbAfnqwvjI0fco;p_L(V71T8Wd1l3`c>q5d7$@iF
ze@+;M<pm(wI@tK({kE>sB=X;y-zJ?~fxd~~9-HpgpXKEYY9Cx1Ya7?w$E^e33rsVQ
zWL~JMYCcUj@~OlTGtv3f(4ZaLKj$3!h?*g`Z2_=eYfuDre%*q!*og^skRr$otJh@9
z%c^$g<va-Il9367D-U=bz^I@OKwy53$HTyI(@V5HFK2Xk*o=U*%wh!Jp!+6$#P9j-
zPC$`4zPUCuO$?%<zoBPlq6_Dm^!}WLJ=D^Y4)4D;;hjyqHuI5=t2|GsiwN?Flm<z>
zE`=G0`Z8t9(&g6{d>?UY6vMscIb)UN=d0osI6ZlP=T6k<XbEtl#tJkUZw-SN2Q@V{
zd=dC6@SSpUVy5a`sl|T!El32B@DbvK!MO)b8z5*50+4{$4aey3s?ePE{4ZFfl));(
zr9CusEVIAbi`*tzQ}K>-fw`8K10*`^n_yT%Sy)h$fnl$QgA>?GF)@GRQ(WB#Fonl-
zsfC1e&TRZ2^Hvm-l@E=>Amb49ZRf7_Q7q9r3rk){@a)2{7}eDc=3a_@Huly)zs@o!
zcCxo;h&-RlG#mrjH*?fjflL+<(;5?bKB`58MrbKNDP`q}KOFiBpN1l8&g8G60`$$G
zw4ChhJJ+uXYbUh7**BFfD+j7^=!<liKf+)_!KY$Fg+(m68SUJ46`R(-hZf(jF;8cH
zX<qwP?xIx{6ii}r_V7RnQZ~Ae$X82eWJ{Th(GIt__qTQZY;X6JNQL}KC8e>yC-0fG
zIEkMLfFsiltb>((^tI@5K!0HJp~0P(0-W8swQUS6^!V?Dqy9u!f*KDZAiLWBx@9gc
zw~P#vf8UDZ+TEZJ55W9a?O~{7jpLTSo_Tg*VcQ#^Ko3g;D*fM}-$0YHudPdrKAbi^
z&96ll4pX35`@DE|QL1p}0KERo25oREgZ=T+eMSu9U}^U~r9}Y@{Aj`?U><GFtLgAc
z`(Zo@;cBJXfS?^!bzMy%4lShxGY~K$TZpb-%Z+4O9Vl0L@|G>C%am%S71S%=zZZN^
z?#Rob{8gwT>zOG2U>6PsyP~0i<O)*)dLn<060k>j1+~5k<K_7J#{8n<4IJD>&=rmr
zsgmPHOK%S+Get{7`!H|*69Be%%CDIGN1%p;pgRVRyA|Ic0w+CNuFoP~iTc#-(0X++
z{iU6qo$ZUngCRkEob=A-VrpoCKzPnF>5wZ@Z?4)7l87anlbplYhxlvj1~&r(10N(v
zFc364uJ>C2fC^SvPrsax7U_Q{e~wwpMOJCeJEcr?cmo7BI26NI;ejwemP@=d_+!d!
zZ1k{8*-}2xz}E`+bxysnfiMW9=Ykryliw%U7^NN9{yoWhXBDa5cnvNvy)4a-Vf&{>
z2w&aU5UR6R>p@BJ_TE?7<}wso_y7C8=93S4#paY{@5S@m0uDSGp$P)E>#F6Qj{~l#
zU<ymy_f14^0xUiyE{xPJ2Ov3sWIqysi|vV;`Eco(D-#%y(@MwYLU*t-;Y7Gl-Qp{r
z0*#%jSugOqKq&J4CmmSEa##!(@o`)|Er`Ct1mEqDaFrsD<I&f{Rv?eK!u$W}Qi}T5
zjUx&PyhQJE*gT>P&Xj+=!lK0_7pwPgg2qTAOB;UT>=F^Fa<DN3AW{zhKud$`Ma!@<
zocaFJM5ALK_~x;mZKlS?el5TI#CW#GJ^q^N8X4W<@+JLoDq<k{(AY>aZE%?}E+zMk
zZIzPJNM~T8>MZ01Lr8c|`P#3B{M3&^Si6Tes-4hni3|SfrxHcO^UJe1`^#Co&Uyd$
zOR4sT(W3w9I#T0C?YL2}1E^lY6uu&o=*8nNyIk($6#{gHNo9G0tWCrZ39s;ERTXU@
zv@Go_XSJrVTjyV8dLzVBl7>>qsQ~2Y!uLCRM^BF~+Gy~zfzBav`t&Nfj6H=L`r0eJ
z{K0{1e@`G0A&%e0cw*67i@yRGgu%40A2TxFz`4Q6!;_y=aNk3+d#2ph=Z-&_*k~fK
ze@H6hq({%4dINx|27^{0?(EEmU--OAy8CjWE#Mok1I^wmkHgdH!ZY{6Yy*_0<wFLs
zk7Q%4Z8Gs%3JNz7e*R&ys8_H{)Vq^4zS+ix?GVU|4%>VP0{pwiJdNA<gtA?_@q)J2
z+ug>6+Ktn7dWEOCzVAuh{<KStFo9SMqJV;ZotbZ4rwbx`9MSPAvDS6__Z@`y=!_C#
zD-d3{O;1hQ*T4X*f6IXI3h<FnR1(AkzABNevOZKg33lF6G=WbpT&Vp=jZ2A@`|0s6
zDu%`f-k!#NhTQ^TeJ`3gE+SLyx<!<;<Zrh&?7kG1IIyFz!h9do2I(Ij)?__?jpH)+
z=Z1mNb`sL4r7_l4=UVS{mZ0pw@=vgH#qeIRLHxl*iUcLv0@VkQ{{;1H*a@rF>ABwK
z_DM~CLjJ4qIL|oz`YiL%e(*EqC@9pRoq`&g<nsGa4i}hcq19GWirC&4qGkY)&Rq4+
zFF7%#{fBz@pi&`i#xIFn4ko(xPZ{<kw}HCW4hz#0Gno49aBU!6q128ORu_u<48P(F
z;F7R6X}@#W4idtg_;l@XXEHnu|9u^_Rv%A~h3${zfkg_i&8q}nT9(E?m#LCKz=z*F
z2$&6*dt4f(x#bK|($T!jAc;ZdTDQ&yiVNRK#~YdIr7J3fp$i<$TWieW9S&O;ZqVGK
zogIF8PZ|%7uP~1fN%H&^Rk$<dA?R_$Ngwfa*oYq!kS^KI$10BiSALqDB>xzOjJ_@u
zjkbMW4{t6@|35T+cR1Jm_y0=@x6qK08M61@lugK98CfBFXDcf^AuC(<$lh<2WN#uF
zA$#x5@ASET-|OlR*L@dnuW_E|oX1KRVVkq!3D9)~%Gxh(x<(;ka&cD_$POY_TFtb!
z#+|=(e)aA8&RNJw2c)L5nFfBV$$Us%K8vRQXQz$+$V?;R%EbO29X7!I%L<>=va(D^
zI+Eg38%_IPe}4A`&Bx2?yujEC!1S|``X}Y=M($Yl+~#xRFiLveBJFkCeg$e-I^JyZ
z|4EB9sBZtKIa;mtVTQSz5kqF}#P_wt)e~?o|8x*^KJ4J=CgOGer<Qg2`}bG9yVByP
zU6(886}}76glPZni&XdU*zC@+GWj=91%kukbn{aer>fd&E1^pzH)T{x5!ih^1M>JH
zrZ5Rz<*LVe(PY0<rYgfd;P3O~sWJHJw~W>=J;<rdP^&`w@_oExykpx(O#!xTF#t{h
zCY;`mnsd6k7c9dRkA|oQ)VVI9WbdfnqRB##HvHA8Y-WT^$W2I$>ObkgnC@geNe+|c
zFH3ja2Pi$cGntOPXMx)cM8sDor{omTva{cTOEYc*R$-A(Us0SfK`J<bTKf`EgyZ9q
z59Md$U1an;>45&Y+;3_m3dCWNO6u&io7cZhi#E(T2yk<&Q`+|fdaw6MsL>NBM!<Hf
z_ehta$>rlKz7em`?0KQQlok+&O#&>|-erEXp2MlLXOsN@uJffhG#^!0o)~uj+nE1*
zeYRw~H!RP+n{f6I6uP@G57(q0-bTxC4%wzNoy^U(_~R%*WYM0JYNezFIrgmsfBdIK
z8xJ;rm1Jhlu2hcG(DymiH3jC3ggi|<+#Ga0I#dFhR&V2x?9((!tI-xLr)ZOMfqz;i
ztmA9QqYp)%4GjozKF|>Y9wDS-K?&31AFy!7hIYZm#CQXghuBD!f>78-mfj8Ii;>Ht
z3lpcWo-xa+--qu=VtUlo1AdKsY5sswTpvVIBRd9{q>!(;qCT$W-_gjC94&!n&Ld;_
zZY!GbuaRVcPLG)ue&}^zO;fUl3mto-h4&5)QS*yu;{2G2)?g~h;Rfkmb^Jo~hXq+R
z_3pdar0DFc%POkVNq%qSFfcsDyE>u&U`$|UVeLsy>Mp(kRD4pB?>~M7Gt}-Z$(dJl
zDPZ7X_I9rBmIZe&IC{7WX@tFgeR)(`)R0?)GBmTyD5EP<?LMyYkzq^7PFftZA%VaV
zYzt#s^Fgx@U;OYuy&%lr@<vPk;cckRa2rO1Ln?G^P><gFsxwXiJOv86sD6vedT6=}
zJ!X<~%IA5#1|(D4I|h)Trzf(5yJBOX)V+}Hql3ev11&o%HQCmM(Ppo7Gcz+4ef<eD
z0ujTh$slH+C|>yZ2nE)z%+%5&HZ#penJPBF4}OOZ6TLyd2)K6c>9%EK_kz!Og6!ZE
z<Gn}>F6?7s(z?ei7zlPe^$TZ3#pJWu_?V;%>Go1~twCy3@o>HC(d^sI0QjDcZWT*A
ztA^f;-<_QEi{|?cs&*w?gQ5Q6hqT(B9n}8pjQB@?MPuR9ri;C>xTfQc4_E-CyS50h
zGBPrz2wA^|doX&F-nS}o^=*_-HPoomRAC#fwaN##v5N{{4%n070mryJW|nBoKrEzn
zVR4ZF<HqS7@ih?^LgALr0z3j4Nz&CAdbG{C9~PG851N@zDMm&rekdjbBa3>|FQ6*%
z$`CjN_&Zgc7HgnhUFu29YDnlOGJCOfdgjTv@#_i$k6e`Cgylu&8$od<odeLpom~40
z?34JUrWV1Jnhj;8Zk5;@5v{XPNj$I>it==+JQ)4&`nQN;GI7eMPTX9>{P=7M?IzAH
zt_3o|Mv27QnqyekC0JRxobIXEv6N-pxN%%S;jy;a8f(EGMJ(h1jK-1#g@<<s?`x*p
zg9uaanJ$HZH?`2>!fx9h=E2rQ$pjjw&taC3*cFf&7Z+!~_|+ReB;mHgK)y?hpHFZu
z`vW_XaiT=+KG@TNzs40VGbNxUOP{Rc#Hf+m9#;*XKJ6=eev_!N@oND2J!{zR-aC4m
z*MCGVGEpQuc2M+z$0!Y`;OSz19{by0_m<8_KF%Le2<*;{rc^T%9<7XSc-4*?eF_pH
z0f4jK6jmZUo^v=!ig?H~zETiq!F*SiQ?!Rt?O1rCZR=l_80P(Ln+m3xCbx~Vi^Jv+
z)xq+on+A|jdELLpm<m*e0$w{^^Bop3s{K<_`U7sfL%z1!P70p=o2m5Cls+4aQ^pJh
zZrs<o`S^AZcU3N*X&GittK~kPL=e~oK)Y~V#PFwvmUbL|SJx0(NGsz#>;*>ig(~2M
z_mQ_k!0o2kt6dL?q8D7=9f+UNR;t&nY1Gm7N@e2PvoQ_Q0YDY|dI}f|?t`J2o$krM
z*6*tJHr8>nUQJTL4%Ccv30P~x9$6-mSj77DZ~u{AVhV7ifFZcXo^TfZkeAgNSo?*~
zB;K5Bvl(OveL5k5j$h32^c%=q-=?#EGiq_&I$y9ZOA|dNWt$x+9}beHy>f{EH8wCj
zYEb(!^oKil97L4CVG45aDKFy8k&%2L&A9=W0yWyR2kCijdk@tLN~8LYO68P3=#Vp+
zPZiRBx6k<+?RjZ3HXSg=mR40!bxe1Ii@0aO+7FZlz)l!|L<q4NVdGQ%%+DyF#?PtG
zelpjl3;La$gxAr#m-!YjqC)jmbo9AQH;L$0fjlwB#-v{7o!zCOQ)5{w#WXrPd_nig
z>b>0BWmrRIn?v0saNGNm!7nfI&P-$J*w%Djt~pqq0EzY2Bcqnm(kMkci9KP@?|)#E
zAtNOU)c?+z0g_etC@Oh)8{Qgp2Ulcz$0U0@yC7zc@WMATP!F$Z@;I9AZ!dCg-|p5P
zT?t;GOw(StEt6>}-sE~8d8a#8oKAPe<0_%*xH6)96Vz=MS3D<+lorGhOt9B0*KxWW
z;zbzBGU^-aKML(Wleek|^a_=>Y2QML?*I7Wj3rqdA5rn1Bu#kAd@l@=1@v27kF0mW
zmV}XEz#VvR@ccpK1vSIKt>ni?Y~-*R&p&tX4KATIU)_xo_dc&O(CL8Q5k!Q*xp|hJ
z$yI+2Z=Q@yuz&~!0-A{F%4Zfyia$%XI1s>JnOK;t@JY_Mf0_do4M@>lA6NOeM8+?>
z^$<%g?%JA~pCCu}wi#cE^-bnN)#cc9^`xt|PB%DxP`1$1Nepd!?@S~fRRSG)1|rV3
zxETl~a%T)|w0UGk{hrFU&N{)aSkyq&nmHWygP?if(ycT|xovhGBnhxH1!R@s-p0EO
z<Zc1ys<)Ny$?s#Saw#<f^^~MkbmZrU51yV4z*!oN2cj65zRM3iDEQe1o?rzfyu4ui
z$ETpIR90F#IyQC_7w?hRf|G|H^b)QmF`1A3Tb^>;egi&c+&+$~Pe67+YHYTsUQ<CC
z5iBt+W2#}V%2nZH8Sesn8%_QsSeju3tSqctCQK-3Yg=>$azYVW?QH~kv~p^5)r&Uj
z5q}Nr(Y~J~2ysuL9%1)*M@P@d_0S!WLjuHX>2FMD@<ZJB;gvK9@FLz<R3s)iBJYw?
z!&}jGW7okuq!`XpK2}lnOqExjb&T3ds_4l{$$;6Se2)yNKUDp9)6U8GL+>I`L4YT_
zx~ekRJ2X0|TJ6ikoVT#J0PY8fc=}=;2UZ|lLoQoZTTHLtLlt4?ZzQh85Ca+(2WS#C
zdy79PK`>AJs14<<bRpUrYTWkT$@id4a4)UPiBqHlCIbWOV-<sZ;n^L^r7zIEB((E#
z@n{rwj@u~H2E5LZFyIG_xZ-_gULO9v<(ZxnXGuv(6DOv0_Nvbud%F(S75Jb_9B~0H
zIczE8<8h2ApKA7|YA}KNV#%S7a4~n<kirObe4C|(?tyJ47G?`PJUn#`0@=PnpDU|t
zDyyUX()UA&P0Am(eBQQlVL^aCHCpjY?qU5u1uG^3l7i%%A0eDA8FDOle(OA|)+qxS
zFeuMqn<eq)((D>tY^T%i?#`<>At_IkSebcQCDHzokX;@Y=fE1p*-_wQSr-UdL37vJ
ztH>*#Y(GP<R#gq~ifZ>T(+U~e*wEz1M-PB!u%=R=kj<?@gaAAFTYA=Q@Ze8+#86dR
z8LyDD6$}5SBa`ZJ7v2NHCwF7j;mHEBqmnIaN-OLGK{q<_`dW1RXoU_`VGw&iMnmPx
z82k7*%3GOUefL>swZW@>uk2n`>zAGgLF$07oGlt$i&s#k9B<qa6AOP_8``>%^+SUK
z9&Os(9C_Ni^s<<+X`wUk>+5WEhBe8E^Q5+Cg(BUVdAs4SbV;`?&hv#ecdi1QNMi6?
z7%et@C8A6n(-ivjoxC}0ej(w<`Yyidd;xy@wMVo8I}f?^!WAiCV*&o_zWXVyv)fG#
z4bW{v1PkHa{MA$qJ*~6^BhX|4N}f#9O7DYlV_$`^jR<Sq))rLF-X^>7pTr`plFfQy
z%L-*g?jaKk+hg~uz149tI|aSqn9Q$B$=ATP5}cE?WU|$&CEr|&TQLHu%Ysu<ND6c6
z*PgCEUS8gvgF`8DRDK=MAI7kQKp!l_-`4j-rkRMCBozU|3b|O5OI3%NET&he)IZxn
zXvkOB<Fuj5t1YfC`x`z2$P2I&8NlCx<ly8X!WNwec>l;bXt2Kj<$3)GP>Icj*<3QT
z;u01x1Rv(EKrtu+zooKJPY|O}4fL#v$|WH90G!z8Kc;(OBm0VcA*6qnrp}ci6;h2I
z+$~6eLo&f+pAMp|!ekP4Q&V)w0Dk)!{<m$m5^qrTqz2BGnNhhP+VoH_@OP6XMP8j(
z<sbX4r@HDB{5oq-|Lwv*luAu*l3`-JL}Fc<Y7H(&`QwqVvmP+fF;Gw84o&rLB+gJ`
zmTJe{EZS_CnVCKc7Y`2Qw&)Y&c3%|i-}6u)At7lLen=<gIA0@reyoFx*V#U`(iIH&
zI$3k}JRx<^u6!qydy%!ehDTjPB|kr3K~5o)n;20sR26g|?DF*~(TIyL(0J*hFX8Nb
zbK+_(3)W!u`qMvby#~sWZ8njL&-w7a5hGwCk=Lh}pIxG<{>&mH50aS#lr6AV?eCJO
ze0Kbu#>Nc#-es2`@b6T3kp~a9t)$Ep9~<8}Ju=l;|0u4NNKA?rJj%=SATN=IzW*Jn
zC-r;nv+`dbgSvk<2nox*4X;%%02<!8k4h4(ZUHU)b_RhT|Jmy>-{hukV<#buNieE&
z{7pe+z$g)jb%(kkpnX0*Ui+8cmEV(8bv><emr;zH+_6CLO>cG}DDmF;i;Adnv9|tw
z%G$_P=`8<{I;B@t<DIytLqY-A^1u#?la1;=Vu-@?IuXJ_4!0)Wn0!(n;;LC+mrEU{
z{mDqjwzJ&Hoh^$;^JhnH$nde@<qC~BBrBKcG&;%xpeB?6?p*z$0ir*O+JDe*ia8Wl
zSL0p?e;YPE%N2)JuZQ$P*23EUfy<;mnd&rxYzsQxKG`@kZ17$3MUc1Qq4jR~P1{O2
zpC0BZlVBnb@2-se3LmYmMsBCSYdZVFNkktqAL54oWj@2g{hYElRrDg-qw+~t*Ky_d
zqMx<|Ht(9=fLMnwQwd;IbVL%pIVGY08WL=bN+_N2+D^RM<rs4A_!(&pEaJ*{V`Fwx
zsX|uzu>O7F#`e$gc&gs<@$r$ol*W1tz8Ip?ZkBb5bN_GD?jUen!C#V!`7UyM`?$X4
zDj1a<_$Ygbkku#;1mG`!eM6UR1||if-}2Yg#`23<dCis>F#$g0ln9bV#pV3n+#ogh
zxQ+&~Z?HoVpk~4#pr(`gRq3Pdr%ewji&msH8}guAYi6gX_j<ME2JA<Kfq})a;U&6|
z6vt1{8k{5^3I6u|X`xIeVa;aHNN|2Wx_)2%1Ix!}+S)B{!%<mtY8n~Ig<S=vXI>-N
zAVHsOGBt8llHHbWL21!PtuLhTq5ex~)vfX-#|udj^r@I>4nZE##Y9m~2gvBjX!Yf8
zc%`H~Q{jb?R+-`#Gb5F%s&2LAiIAuPxw|$!DiQJX4sMQyxC0&qW;CSPaSIYi|3>q;
z9DzGvv-ezd`@D~L8Kl1M`CO(XKU-wijclDw6KR?{i($tKY1J)vbM}0}hkyrU0XA_C
zT>ggMMo3qHTpN$=<$@1CwUF@|tBw3w$$I_G*?BRlvgs^Q*pcru;(uUIi*B90mM+pV
ze`R^h5&bu=3M|AC2dB@%j<0Uvg*0AqZ12P_Wy|11Mn18zw7KU0db1R^ljsOtIPLNq
zD&jSl)5(Rpz#LF4h`-vut|Ztf@QYHojufD~T6Ll_eQkX;4CK^P2m+Hi9YaB>zD*{S
z-+r^7nL)_mJL!MIyEMFE;ZIIRZ%4R>g({8v{mn@UH`P#V0!zR#hNoBl4{%!9J_@p#
z*|pSv3Nj+`98Gw4+i#BrDe*Wy?t~Px@ziwgsKX1qt>084rL6qy;_E)&<K*WNM0CHw
zVaxdWOT8bIdoK)7o9#wPACpn8V`x<otEOXRW&NT3=@d8^rl8*zOuzQVpXC3z0J~_9
zBn^7ElDCe6#tImiSrdiv5jfF}P<XKrg>!|iW?;bWK3JHXtcN&!uG|r8pbTl{OzN?1
zyd~Z4CSsBEDaXxz=Fhq%`?(7X;WLLGEYFW8$A(d%PwtU%_eHq@t@G=w0~wyh7>Z^~
z*qoumY`3&z4HMS^-M?^E@9VW2rm4^v05y70`O~#8iK=M^o*q++27>h-%jFq4`3~y?
zcBYc(h)%7Q#S`nb4+llf{f5su5)pr21>_IRJ^*tew~5cmg6?@xPW7kH9@F6np^K)z
zD|%sPc-bXh3supu(V?NF#Ak5V6=LXG`ailYo)hdptF&m(%?!XaGxs}n9>&v<(3eot
zm$<vS5aM2X?Mr+9xF}}yPRr?mi>r4O<+8G8aQ}*P0)>d@H3UL#3>p~Wx_`~iTgJgy
z?V<{pge1eaidCTZ^CLf{K>nJU7kj2wZ-2RC&%=fGh~wwV=x_HW_L)QuHy2}7M{qts
z7??U{>K$j(iH1&<rL)4iRy&sKfKzgroSVDYu`F;o>Z_!rWNYR^_I^uRN~Rs^uv7gc
z{2ewwa4k5`o(>G2@gs&rWEx9ZOUSz4<`jCsaoo11f-=v61h};<7;QX649#5h)to{U
z93i>`zEwkHyJut_<f8mRvtuE&K2MB+N*G7|_ZzG;QDV}PrudP{(GgLS_7ccDBb9si
z$?5Z(p6RO{t}2Ze)ovY+zI`IMsiq9>(|&%0;h;J+>nN(F6qAAMTr}_qkFXVAV<r8G
zd})<^nnCYrGhS;vW2u8!+F21y$5Ttay-g*7Rehi|-tV+~Ircmw?<V4LLz<{;noQJb
z9B%8Ce)T<{1PXCiph3DXf-N>xG<%xp<2-J+4dF1o=kw{zgZ69ic+{4iPNP=Wo{#NR
zSdNu#Z4{wq;@lsro6S;YNf-w8eH8Q9JJk#)?2$>S*%bOmQg>}lc4rSb+#H9l%D0KT
z+|>O!v0KlXJ{(-22Q(D-VPK^_*mWJxYl7vWa|y)h0=}r-moBk23=Bs!{~tec&u&r7
zI|!uXWBT|+<+@yY9%6t)*ZU7KiE*d9O2kOSF4@FXn!pz{|Foz>_&}Bu+hJu#@I^-y
z)D4;7G4_*;h2PrX3bpF>{D^OV>+EfQ<7KplM%+bds^1@X>8ILDaO?h>K6!8#`O{gM
zkS2*~B;*d+J*b*elx+L%V{{=gYfa5KRSQ<53dFWy{%YCaSoVqYu<TK(QOjQ%T3XkG
zAE`NU6z|0u7#Sh|nU{~3Tzp||?(b)hi$d3;1V%;%<O>9Us&ek6Nn$knP&i!u1oQ2t
zqX9>zcnqX=Q{?U2u$8efrjDz{OrPJnEQerDJ(vpsNp7yWVj9PyZO8l?kxw1jX6c12
zc-??Q&gu;OHmf>gfxjd%Zb5=!C*ulkWF+#%q<khw_L~2g+*4%S>sSf~?F#9=>1G@u
z4>Kbl$@xo~Y0<;7txwhf;qEQ<VEJ8089k*J<Q1jnk~ya8NV?Ygxo|Z(56u_Sz99QI
zp>+BiM?~)n`-z&o`ueuB<x9AiTC8U@G~8!%@hJ#cK_D4f6@|rFyIm*h9gvUJK>1HY
zl1<+~w7efHdw2tLk}l@$5&E0LLN7U{JSZq<g=JM$_mEeVQ+{WKTW#F!gSnma&B%kl
zt|5+pFGA!fqWK4=4jK=F!G|AKh@+z#TQdz+HiO=mzuCboLttOzm;165rP2Yeb%Oo5
zb5KD!6#|JyflZ#)mKjH1w!&tr_aNxxh(aVPA}XM-k$3$f1gsaP>tN-sz{DqZC&^!!
zUz%*Xc0R1Ewk}ZzQgBaES?@Y6Is9xhq<Yt`mOC3pyH&Q8U2G&|z{|Sn6}g_iT5ut!
z&u(C1ofPr~-}7K_Utswc1oj@<%wQp9(}D5Y8voo%`{0?gP^MU|;AX;aTdr0wfNfS@
zG$rC+=96(<^<6@v&#1Pt79`4(O*_NyzTgb^_~TQLygozfJMszrsouNWHSiYK1^xx<
z;cJy98Bdhw>2u)s&&|BAIf*lJSXnKJNt5j9MQy_<R=p!QZA>K**<{}O@+-L^%7g#g
z@cd)}je<g0yyhsoVTkXH&Z=gmxChE<<(ny{JB-Q4-@yTf1*!9%Vi)4`h4Z(zAANJW
z4m^Y*Q>~BR+6^;kXe044e}&(p0k{ZJ6yrA#I7WeGhd8uf4W7MafJDScS1Ijyn{!SA
zn9->(ZLQG}TwLC-Vt2o9`92bk7joO*esGmyFc}w>CXy1Ha&#6^V9b84|BTE!PtAQx
zJlI!fZ~zy_*LQ9p4bE_ot~TB(;2}ZQr%_*>J02j3LRMrA*iXV+7BRyfDYGp1y|lKD
z2J3N`!|wuRF`%cPu>A+<z?F30Twg){EY7{{bDi5ZL}1dHDnH_glo7A8_mcb5tV$CM
zWGu1+G3N7#YS)8Xl$iM2$If|C9|b)a7zN{73UX`gjxMu_pS#=ZV|neo`ZlhbxkYm=
zKoZHw_<ZcT>;=)g)<nvb<CUypob?Grbr$W5zqNfDml?nsK<>K#J8k=^sgLaG!eFsG
zIWWF|+_`5+c!qzHt{Qonat!AIJ-_{7M1g(U4IBq|TpT=)Be6Mc?bY8pi79D<jvYTH
zhQuE*?9b@Q-}zUYU$@SZr0Ds&vifo{cmwYua-XciwUwkyH4XBb0mPz<y?=+`CSq5r
z3}QC5_jd^w>EPMwC^s@t($vx-1j8LO*tJ6|ek|4s`D@Ss&1k4ST6&qMmT<}G4o{G&
z3@#=2OM!oxs04)ZataEGDM|1(7)}=!709&HKDfhAH8tyZwp;WEQyH(u1*3Jm1Nk=*
zW$Y;>bkEx8G*0$>w>*ve4^f#Jz(Tgfq+r^->8#%9>l2BK+a!a#<@0QmpNGw;(rWay
z1iFXTV4Qv85hJ-N1bFS+^LJcas{VbvQnI-@lZFZMa&*rIA^4a+Ji>bLg32MpTfAyd
zwV$a=`oNt)wr5RwTu?C6CMu<I{P@=6a?Ek-()GW`GI}J^PXPWO(e;i-|ND6F_eki=
zLZ%<V81Un+`%~cJ3y|Tteo87B%xwPM9zi1JIBCffA7*F@-bgAEiu9O`KY!MY?Bjst
z?JNz_lgKBpt83+-(hYyR)>f~(^%WT#zX;?xPU?T`>SpdRQ!}y-Po8Lanhs5xqN3vG
zsbje(^0*G}Z5%A?7qZCP(5IHx_O<0T>)8d6=ox{7*%x(4`^mhJNg?>O0p6O9WST%o
zibQ!(aCn?2LSuZB8y^O&qGe-6m(vaQL~hB)k9{|bS3vU+KRiAX7x#X8a;UImR90HI
zJg2H>aQcHpW}vK;F>&x|OzW(Te&ZW)G1UW5-|u~CWDmILRgB=kQStwe3lj_NfCViY
z6WI_Wfr}~x$ydV}Xm-21P7vwZA&%818F|9oin?j}!Yi@;QHfY`07jkbf74VDV*7xK
z%6D@lX2(<}a*w2-EbnG&w&!Y&Bjl@lHcS95<%a*Qn89;CvOkMMV_o0C?>oO8K^{qh
z&lW|%BrK776eDBG32Z^N3I&urIyiaE`Y(q&|I%ftvK&3BM@$RuO-%pL`6eFT5!q<6
zV7-C_h3(-yWf!+9%2iGkNEa2_66O)Bu0D17bwA$9);#JXS_w-}Tzp&x_a}LGh(+=d
zCBHYEH!7=c^fZLhIFQ5vK7cbfKCN`nOs-;JU;CTwYmMP!rWHES{b{SI={K9GCe&lf
z%m2s9s`OPx>q>3519XpF-yLC(5in_hmVvso-yb+4SrB~X4@LlzM%Fix{4#sQf54p*
z`7$+HkX3=*`s&c_0`fl9hPi@>g4*W?7rdwDEqfv;ID4jgfAzMNmsQiYgrD4!Coreg
zwrFoXbGMl}rxyy7pj%s8SzfkuWsH=`gYLAd6DIh0!k&}$B)L3p@M?fApsl>BSYJjD
z`(l;q2umYUurl}eeswaRM^j*(Kti{4R2E!%F0LUYBmY5k>D-w_kID!s*AZyvK~Dz4
zxSms6E@XtC2k{TAs{@1V_gU14oV<Ldb7q|8VEGOUVNqFp8^TYdt9yUHE<Ezdy~`)$
zr($rdqkevXE0Fs`28OG7`_paCu<rR^I?qgqAxm?Xvx6K_!Q$}j6A+$&h@1(XXP!7N
zTMIW12Q|jVhEMdSiKR>LldfLzxn|WT{pWf8;{BiZR}IB_&d%L6Q<%IXqYr6m!Iw?d
zY15Csd`DjW1slx49gG_P_83y<!e2QI&0R~fzRjdL@23ZrsdKmvz5064%UZ%hcoP9T
z6oxuf#%E$57oQNjheOuT&2(Uf+&cBlQM0OCWGB!!($ji9@~mMVTC*i~Bch0!&5w}}
zuldKt-BX)<v{fNOaSM4D&NdVHkm{0%E3c!o69ng2=q$kX;1J09od0*|7rrRPXzG@x
zdlGd~5hU`J4||cZLO|u?rDwSfda^E|b#=r5rR7W3E)`zv>SH4+&z_B$zP229xz9};
zoLM?qz4i9bDHwpk%?|5&`PDg-f&Sl#BR$Py&2L8}!qoD4e6mMMjvT7&DId#uYx^=I
zemm+`bYNfNwR`O7NDI3@TOClFJNaWLwoi!3z2@S|GJ1=Seu>0vzt+!l#*UPcRZ`W{
z5_6tjQtdV40{eVbaRqt#tY|hKtU)v3U}NZKP@a2iPyU#q5^5*--DzUvOl70%R@}u!
zs9)wRy{p<+bmi4Nn=at-#dpE9vUIV080vSD`L=x?c|KP+XpnVT)DZEqt&}FRj6LdJ
zjHL^SAHLC$++#yLWLdQSNycv@1I_Bn;B<TZzn2-4Nl&w*b3Ns)?xWmwN2`QxlzeBn
zx4eRqJmhk8bkw+iH#88Zy?Hhz8*f2;4m1A2@KalZ{@R!B`hYj(RGl~Vb6PPLl+j8A
zdPN)#u76Ha^BlY<j%BnOG_wwu)64rTjeRl!J1k69XTjayPTqG}y`8ml9u4VAYPI*!
zUP@~<{dWOUGAK#DdgYWC{3l(+s#|{<-bF;XW@(-I_*;SwwR#7SQ0i9j9{$v;lM1or
z_o}!alN6_Sa3xSLAc2$``8>Y9_(6+GcNECiXpi>x@aOmJ*-p{ILZ3)c*uMSt&E)v}
z$wg$g@89<)b9T*F2j?s2W41h4v-SCxw9vp_g?fg;#Pr>dj4eT1&(Pf5?!1V3mwUD)
zWsg?m&OL<?oc7*siyL`4X2Q<K-tA}9r*}cOw)tM2uz9(=v(s?rg8G5<G0_2dNZ{xB
z(9+VvR>981tj_Ap@Sfn+iOVbfB@gj3slAZ(zVlGO6{d%@-(&3Bwlf3p->YF`hpS3q
zAP)8VRj_<Q^A9<g>n%%Bk7YDN{#d+{D*yY&>6UjT37%(ad1FdZ&RbOoT0B`YaUMq=
ztw{ecaxo!TV1HNzyZ%Jtps&v<U_0{P5%s%hf*{2g<vQNO=qdk5s9=0P$$0FvGWp+)
zll}$_Q6~5%hni|?;J9S@(k?V2D)!xNSzkhiD=f6=7QeHKKka$j%V~=rJ^<e8EHFuP
z=00(-w&t}m;a>N3@7a5)hgmO)?=MqM90hGae&HLLY&3s=t3S@P^Z4q0MySmE2Y%<<
zj-uonZOmg!heA@#!r?>5oHX_`KDY1O&xSZ1<HRwBenV_0Km7fI)=fHe#+MG}3#AjQ
zmWUsPl?|WE{n6BtuO{WQPeirFmKUjSdmcp6IGp*}8S;!<?&C2cIgZ0A5mf_Lz7n*D
zbNd3w7uVDMSqFi+XmeUgLTtXZ?q~7m7xF(P4N<^Z?cfM2Mqn6gZK4xL`DM!n;hon`
z9NJ$SEBM|cPULC%S*I9g7i>?Ci@M7-etUVYB*jldC|`&F7(>!uN433(m3#P=G&Y^Q
zb8EumtD=0xHTm3aBLl9vGaDY_U|^(+P(-t3hHG9kaAI$~qnXHV|4rMI0KZU1iuw+k
z|7d;XAI-O+;b>nTxsSy^A==f>Dw5&H^ZRnea&~W%lB{8Br!2}94%9-XuH~nj^yXq`
zexVQ=?;if?>K;%eZg=PZud0sa7uX*A(aE=7>CLw^K7gf%>PKZ`)zF6I!iMCR3NK~!
zCAemYUtcGBhbhJW)wK!F&4(Yz!-V+RJpBJ6{-E(IEM$<C`tk5n&Yi$jDTrRb<{zA#
z93ON)00=vML48S2seA(Wxq!7lZo?OBEYIloIC^ofAp8j_NiyQENJr(RzM1Du7~cD$
zk|x^}4eUvG-DS8}jri}iEHv6wY*^zWovB$Is^$M7{#k1+UVo`(d@*7>XPs`0{XVRQ
z0zQH>Vksxq-}j@T{hI2+&T*-Mm9;4@EGB|1dw{q4;|D$0G7j>&yym)pN1_0dF?=6R
z76(cpIsm}*^>J<xE)d<z&gSB(oWJ}2Mi5?OV}48;g+9gBTs-N9M&Z)C$NB74r<P1i
z9dLZ^hA}~b<kgIayraOJdmmL=rOeHF!N-dT-`TN(K1TjOdB%b8P;)W4hc^zAE8b^S
zRNQ^}`+bFz2=L5v3m>o_v7MU&bn^Pq?V)H|hz!lhD}wd#!1##1js^}QLQ%`_`Y2W6
zldCJYs;ZWgyR))Vzc)S!m`}^xx{{DBfODwJRlMa4t=YSRO0(ea2o91lWC@+}n`eVl
ztckPv&d+tw9yu*-{Nx<E&O-x&YKT}2HfB)n?HW>az~4=iQ{HG;Nx`bWO@r=(yuv?C
zYo8}bh@(Cmo7t7yV%cLSRuf)%Wio(z#j-Dno0$Ci<?lg%@jtf~c)OgxjSselc;Zm5
zzCJGmFeLca!K1?)JRbyXqPG7Oms!2E6P)+@bu@I4K9h58{Gbv8q^VI$Jy5&GkEI?~
zocUngM<c&gx-5V^=Sphn)m;~JQ9Mk5ei9nY<vl`Cz1SPR=8&bO5@=wo+CI<5NDDL5
zjd|mN*mg4PH)|L6Y%>^v-xF2dJ#TkDX#62U63DPX26GlGx{rZw=FILgKR>HzZ0qcE
zV!CmX<s)k0rPw!3A0aI}<`Gu(TZl)-ymY2xMQ`O`O9-d<EdH%!sdpoNFB{s|;8Rb_
zq>vpZrlN@nPb?NSb<NqOX1^<{-8B|zKYlI37px_=Z-zQ%LZDeHjp7^nb@{le`gCB$
z#kb4(dwD=(dT^7(oP5i5OyLx$Xm}K2X`{V?owu@bau{G~PR;YR{(rBU*PoqNCAJ$C
zTr)|PhemJT{mKzv?452{`!X17@m@{MdX2<n9)IKXU?DI6b9Ffi1i{Z8<eofX#>glt
zseZoMbLh4TiyI^(RRJAB=T2Qy+^>B(=FFa)`q(H;6r`BOe^$GfU|&2_;kAv_ew;i9
z?IS28q6x9|uSnvsXv#7iewU)31hoZYc!D)ebU>O_ei5Tv`ipO0`(z-C2u9hRt^@2{
zCtCA`0qc+@SZBt}=a~}!oK|k*P&9hlK3!_$<<(eKr2#A4kK!NpUL__MDrmY!X83%q
zNq6-&=~_Mb4G}$hzbAf{>c6S*hx!HrDTwI#RKyQeou9WBd_wi4b*+lWw*M@#OU}r0
zBq}0rBcjybC8P_u-TUH>(n0r+a-;O{sE2@00&Bj4f&$1C_oSS<g@Ex;*xQKu75A2^
zrmRvF6+M6dBO~{LN43)tDfXpl+}55FYOAB9?tscADK4Jfih&6Ro>t^b@qKl;tXtE5
z8kz>c5`tt~@YS}l%)mr$F81lSm>)sn8h530a>_I6?D=Ar9+_w|48-!BBZ&pc7+hJN
zF{VJ5eT9eS4=WQmXO4OP@*)xUr!vbyG*8rfbuq9rEPO;k!8P7omjnL7t)}mU-{Wth
zLIH5)ic~ggG1q=#o${s$co%IcpB>ckX?%4)+y?VBasqOqly`wIB_7<4i;t&^BE5jw
zA8@MrdI7qJpYA2ZvFZw72KussyV12r5^7y==ZCo~dzF-cWV&@MuJ$eg&=zbg91Q=V
z3J5&+qI|9PW^=p(?^9GwWjUS?Ww}SY81v1_pEZX;doQwBY!D8skJ_)KG^8{l0IMLP
zPZ-PGvYYV{a2={MRAz}L5!?ToO>ewKlc@MY$u_mFPUnd=^Ho>_v)O9Sj(w|`h&P$O
z_6&G8r6qoxkX1K59{74rp8QNNp?LBEGjnWwEHfJ&--zi0KXoPBsa@Wb@82121howf
zsdYSC{cdISH!v#B<cM78appVL9;Kx#F-f_-OEKpO*hq`Xv_&O}+~iQymL>hq#)qG@
zc4OA5zT+t5t6A303TH=Z*<kltF;4VvW=T>6l<X~dJ}s_GXs|-6IXv{J^yB8|zCS*5
zfHHorrMwAH0Ruxv%grkWI+-M1pNj1)xcANd`j2{VB2?@%GVMRQ<(|M+R&BtU3QRS*
zg!%>^*<XOMa3o|nX;DT(h@N>7pQZJImCZ{rSOMARMnVs{g?Sy=Hs;}r7jmuG7JbDl
zq4EU@WoHD030(Ihj_#;poHoz|dtiyL?j>dpBG0ex*kcC)N3!C1pCXm%i^cs;Xc8D^
z)FJ8#-j?uKQMGT=sHNMf%1iyvGN!k-Rz-~08Ol^!FE&Ld-H!vqvwaL7-@yxM<-oab
za+xUZb+#Gs&pKC5p`a8_O*!BzACL&*cvjFrk;TKpLCs&~S2P5-&L*ai^xqRgK#dU_
zU|{1t2l!p}nI$zn7mwez55`SMTyfW4dYhlv#{84?{%S~&Eb`@zAUE0rMp|52yrTT#
zYyKU}<Atj)k?|_K_{!MktT*18uD$OyNdX;Vf`JE_-$63Usy}7>w7_zawunzhH!(<7
zinq2ZE5Zs4EX4gzWH$5839R`|Q|<3Hr4?TM{dE&^+qnKt`fM+6G&*o%FMLU}73XN*
ziKSXzS?NvU&f{;y5XFUyeYwdE)d)CGcI(TtrXapw;x+kcuO5Mmn>BxTe9CWhcqIbt
zf$1T8<o;hneKo?ve(Z(2+sD>KKE#SIc8xk0x0T4&&y+~E4%Am$oK__GQl@Bk4q@Bp
zaT4?`vf7XN05=kZ_-8S*R3E_^3Y^pq6D9tKkO$Jx(Lr*+Ln>ilvyDVdQl`s&#oxx>
ztxyQt71!yD-exTOW2(WT=T@~fcJ?uDN>xXvI6HfR^Rpk3y*D|~RZt&*Ae@nZJ*2%q
zS<It1tj&At7OpEV{4cuLUqeG-t*<}T7nH}KrSyh|Mbh65y*oG8Af%_~!}C@OWY#>q
zd_$|*oiD)h;w8jQp)QxJD|p6+N9|tPLE?0LT%1m`_no_H1aOuWqe_(xd?dYuZ4V<D
z{d&?3jAkG-R4Fo2Lo=Sa=q1j+D`s6ig2c8J7L<Ugp0l$fgbhU<)<FIxG%}I#3lUba
z`5MW9>)*{;$7728IW$yQlpXX!41R92jA!Yy$_NR%@X%1WIN+LkXFJft`2{><FSvAn
z5-!TM68}?I>$31g%9wH_R#a5ju|;{W;1bb$5r#{D8d_Eie$%AyM?}xb85US~{J>pX
zgu+NiLn3cD1c&1DPCo%d7^<K?sdW}&5L1$f(H^O*b4RzYS=+wu(UbtoELeKsU}Tpx
zr?#0u7{I=}D+VHc!$aFbmN3OYMI!>8nA`I>z39UsvDtM5(hCX;nJx@c8^03)Mqe*5
zQ8`}x<{LF%xM-jCcVOnJ$c2YhypPQ$k8NXNVk02JPZ6-&@A`BzCg3Th60X-XQlV6c
z8;dX=husQ)u|d>5>Tfhza65t|21f^n1~&tOV_&1#?n=LpjkmJ4*49$1Dz70E_Pjpx
z2qeJIUzm&xO=v-o09_0r>#@)5r0zG{+h$yyKYui2WrZnZ-`XRtv+H%Di<QqC=H(q#
zioj_U4hu2nx0^xtcb*Q(p1+HHu^!n5`ETZ$lhbESP0i9Wdlp7!Z8eR_21ncqh0nj%
zy9!^Ti%y|kh(|xAK^NsGlI7JT&@^Aj&3(ZXS>pmfvPG%1u+W%G7u*X)C_WgeNIkqu
zK=9Yqwdx)Lfr@p8F**WwUY{abA$xcQ$I6OPn;tVs*}4Ln`hj607YtL7jtNZ|VNDt~
zITe)`cx*)}sl9K;CQwy*$Ir`ITvB46H<bG{X<)4D+S4yU63a4y=MyGSa&nu|>rPJS
z{>8<Lz*JyMc&4NAZPrQF15da1#r>gUIYs3sO5KS9ecj(CA96Fx8a;JyUlbqokB)rW
z)zfui+a)Dxurc}c1SGk?`L7BJi*b-pc}5ouXcks2kKpQOsW5=9OG&eBqARPRK}JI2
z+aQXPDW{?VfS0b$uUeYY5DxmnKjmhS1Rp=|bc3;ynu4CR`KU_tYwyqndo@*ER(j_7
zb+hO|4Q3`Lf_r}rQpx`=EQm?{fc6VJ2*kCYvP*(I0kD!)lCCdFUH$u<o`%(6W&nW5
zGPpX@n$=21Ps|eTy$SFUr$~^6%jXgt7Z(>E8&F(~6Dmcvyh%qV18j7AJ1a-WikUh$
z+AFGIa;08Yf_P16WmjJo!rKiYaPW1Pf`a9roy@^yrQTk`7j$w)_y56K;j!qkoi^5a
zTsn&8=F;x`7Syl((**gnJ}i#__$ENs_hSOvLli;6drJ#T2Zu*xesy{Nm)d7(^OA35
zOuC;|N=d1tuC6X16<5u*ps+Ol{d?GoQ=R7g9~YojQ(aS4Q`OC_!jx0@Mv$}(pV#@g
z>Tmv%4bCffAR2b<6ti-?(1?$Zk*kOABe|4VP*?yNbS_Sx;U8*gsli&+w0vB;jn?E|
zR~;wK7>iK^wIj9Z90v}F0ZXGklq$<^tyvy%a&lGG@Su~G0?9#&+SafoSti@+WqxC$
zAjBJqlG~Cpce^oFf6l)i-QJ%h=(;czv>Z|mF|3-JpIVCWvH6+VScg@#Ne5=`>$XEb
zonKG@&pqU>gD?JA*T<Rlsll;{tjx?itsGvzQBSTCwKWY)Gs@Oh*THuR9Ee<8t3)vB
z0n+{7ZY49XsCBFQ=T9w+SSxemmsIceZickhHwwZUMN2~qQZaRtQc4W!zqfUwAunI9
zIf*aehe~;P2pW*Hn$t>;kg!D^I#pD_D+4eO6wBt4-Vu3IE0>0((3tp7f)vq|12fWE
zTDe(Sg%yQ*@K9fs_IA|E=<D~uOJ7q{nyNr5^2Ii1={8}Ct(9F@t*~G0n>TY^t0emj
z1EYiQ<724qhQif^SsnICbm8Xa#=(<(DSCRaK&g(e-@fMckd5lUdPX#V7&nc&fBWIu
zZU6ZSLEX)aK7EwTpvpuwI>CvtIn$u8{%mQ(`LVpbG0bDod4Pu)j0$4N{@}<q@wkzU
zR&GYdzr&XAY_I2iLHX?H&Ai6rLCDiYyGVsfWW{sZ&2QW>E%+jiK1kv}-^05e&QL;e
zm$%e(H6ukXz!Z&q{qzpjabS*yFUu*`(_5U_w8tk!j)pYg!#*c8e4Io=ibpZ-2dA^i
z+}$iweB{CL3A67zDvIcI0Xr8%Ut@dy!jEL@bP?pAN^`@wOcn|5eabb@P;zprpb5s)
zQqy8$E_#K3m?pFKc-m&@RoVSvI$FBCteng(aVhG~27t>jMZHwj)h+CQ*8&UTS2(}-
zijAfwK|d9`xw)B>*aEb79Rx4a+}&N-Sj&i4IUg{CGA6@vH;{8Dl(8e4D5MqlfEe<{
zA=6_zNOpG#pJ!xLKb3sI=}+v^WS^i6Fy*+C^i_3vU+g<Xgfzb*4`a#wc7J*ErHM_4
zvGa6bGduH^+*6Jb(?3=6^Ai&V9lC6sf{}i1O1RaT6ahd)t90HI^NyIil_EdiY#4${
zh>y9?Yr+sJ=Y2S*1Ps?PxqgUjL2=`~egs^Rdo8%qZ*T7yJyKRaJehBlaC^>?RJ)0J
zb<an@XG$zBz-!iWyLN}?@96A2xm8L6dz!dpzM}%Hy!|N2*xdhoE=O*Y?8+4r35m6Y
zLCgu5u`n<*VoWwF-;;YO`DsWdB83mQbiaR8w-nkXlun^m-2Hx3$us=={6GwqaDo12
z@k-3}>tvGnf<CwqL!xt7*e?2{J_wZq3!F((lI@H{$1(%h{&hlrLEKAalYYc+KSYyH
zj*`l+_~}IlD}0iXY6E0^vT6Pn;wdzN!1UyO^do4HCZwRC&_=IzxA^GJ?8yeK?%9sX
ze%Cmb<OZKYEk3e0km&5Uu{|w?m#*_G-UgrVdOqkDAKbj?S&8lcDDro4ZrRd)&BtX1
zOpyo0oL?x<e0FPtvl|R$h0N!4wzAJ+e)W7GxBR<Jg}{+3;`njwyQ5*8a=8!~s5@9|
zvpHT3HB6GLslA6vYsRIk`uH*Ysc_a)G*CL|JJ?%YeF=LJfFTq5($)tBRUSV!R8UjX
z)59i^zMa^)Pa;^{Qlzdcy<o3SU;1n+hUCk5$DhrEy}NUFp=AWk?7z6t?Oh<-jMux8
zM|TT<_F4nU#eY(BsMM4}@0Aa@MN56jhH83jGMW0#4hm0eX`V2IN~vk6a14w={6pyT
zGMMGqj70oeCl;!!58%KPqsQ}2oi4s<t;LeeX{#n}P7p7%C)P02JS@XT&&*6i!%D>#
zT$)VDXJu>UdogtX8d`dS(Z2~Tz^i62&o)nyOxW>b!qK2lmqHl$M$2EMB#~Tj+JAVC
z3!j8d;XSO%-?oGnu5La3L<L`7C7es#P?*gzH!IPvH6LwY1<%11iMfQW*@Bh(;Jvii
z@})2Qtk2dD7bC!NXzBdW024-~@haP=rs}b_;?^;*W#Fl}BEE7lq~}V4gOiq#iqrfX
zG4g0u5PpRh<>Nr9w6?P%pIs?8-D@zG#6WpZmALmOs<uKvhBv!Gi?jG635oMV%^9$M
zVAH9f`#Vb7pBWJBi33CxG^C#@05Wz{1|0sGwfT|~`ku#?B@)vve=RA`mW6>6C-uD?
zoT*-T6e+OUfH4-Al#78|_^7jz2#3yj>4PiU<ELd)>q0AG#yXGw3|CFB+zmpVJvYk$
zoNo`5NpIHtQiM3!TW+`c1<OidOg4+XQn^}UpT=Tb-L3xJDuRK+#l3aGWK&q!a0oLh
zVrgqNRbku<gfQfJUyl`4nZ%$+w#gj&rlBO~##DU&{7yxMxIv3gzw`1R5V%^fPoG0{
z`THyhS0!6zyJ>J7eTT1LWZnIGldU7^Q%==7d<MX$O@+~$?&woIRaG@U2!=T3jk4+B
zW^CW*mKpu-4^HXOPa7%r#KN0CKs7y4Jw<!{^{~e0CU!WR(GmWn-)xgE%&i&I^5Bm7
zYN;?ORP`S(6%MqXOWUq<+M*lPffORxDA$Ezi2Zf?cpsEDpc=a_TJw*pHAnx=th|6n
zXoijb-{Zh#iG4fdEP-zA!JSZjys?QPll>03X}6XuG}R^N|Hk+<2u;jr9`--hl<xf2
zBX!U{&JdVf7#5Ti+NG_Vg(Af_ZQT=Vx0^(zCw_z$C`s7sw0K3`ceXklepJhkTfZ#t
z5&YfQIL%yWI)C#mEA0TbrY|S!ebEr8HrW{#muQaxl?0@HV`F6D2_b@DJ8xLn(tei-
zxIPVQuhG%LYY?&V+{BE5iVD^YTF<m+W=2C5=}1XeVIn~Jo}o&b$q~EB2L2cO+kkjd
zmEA@?#SH02^Uuhm`UigFBHkx#RO;ja<^9s~y$<V-TH{$u@J-LLFw;#;!m1s_dToDc
z`O<2^LIQ@YC*2*FK0h?}*;BuA@beOyWW(M82u+}caU@Mij-f_(QvT25z^x~L7OKI+
zSb2Es%+SiRHp4*B^SELF2P3xcxr(RILzeUT8YRDA)Y-dtad#8m_1qd;MWOYc*qK5r
zVZIQ*_=2uOCcwA}5O9O<X<#98_r2N=pzyivmBNhDY&%0k@9p~;9}WqsEMcryFmtD7
zUY4B~-T~mK1&#W(!^&i*k-<&l#3YilG_i>(E%FU9gTGdn9zQ%H$(DF`Q&Mu3N%)~k
zq<aYgn+(X%R})VN|DjGFD7luHJGyU?iy#*(2Yq#hKn%Xx?hO7RL-y#o16HuH8P|Vs
zZy-eZ8T>KVJh%vXZIz@ef&Dl{r9i_)Zn^Om+%qsLY@dfyl8KJ)#aJ2p`+Cn)Q1!o)
z&q>kl0~1&fj~17f8h&PYDCB<^&7tMqgX1WkSAPFd$2>^<{l0tQ?te4(T=TnoV%Ty9
zcmI?6fE-bZ^mX*yQNy~``tCVyFj^T+BPj*~K|}z9$e#KB5==J?2_73cT>lci!|~OH
zp!v=p5wFXxpQ&daBGd1KoGGU3O@5b!<fKVaKjyGzOz2+d*4cffqtwgKv_iT4-8zRJ
zx;0UgabSP4t(5=&_h8YXi>c7hmGyrNMyP6lC$sNJ`tkrsKw7hT`6Q&Oy0!&60s23+
zb~=@i^0HvfudOx8k)$Ys=Lb1IB&o`oD!`9PAxR0mr>3EAVs9Or_>p9P{|1DwF$^!e
z{H?gNySwo*Y525J2z-V$v^2B-u7Xwvo_J2j|Js}St*j4pYhH<oQxHSFXT%_Op%k<+
z%a|D%Qkz+eJ}>KLWGs4A_y^0?b#(jfvf+o(?${4;@kB%k)3>lz>vTj6>u$erUpQQ|
zF32k8xj2&USf{38Js=<ZP4h}bjFpBNOe7J)o>alBqx-<tcth*f+%V*%6d8r_Fy<{h
zxZdseaav!E^Oy3lQH3h!2@8(H!&9^6Fm|66=-fi7Q6i!LyHf21`GqH=n#W6DI9PaD
zX_nN*f{nR`hWeD<!mzPh8q}JD6tdge=ry%8p?wDSM474RWnO1-e`y|nVh3q~e8J`Z
zH%^>06^hQQvB$kyZyxTTFI+s6Q|kY_^NR3AuBG92i;TG@6wK6$_i8}DGcrg>mjoI(
zu#verSCuOT!NI4~1l-&n+-cqVqi%>6FT*`Bys*@HO3TB;msPa}R1-EklXJl-9RmX&
z4-1T2jPmj{k{4|4e7w9w2_HJ;o*X6qC-)Ba<6&J)H#`JqJ<5CqSMKYADrs`0u6;zA
z`ce!CNyoqR;_eZ6#?CIcbc&bA2E5u~a_I^z=kqB6N2g;tp7WFLTCm2OuCcP!Ic^`M
z2?keUyJ-|VHPee8hNt(zri;KFLY7tos=78YuZ#NaxP2VUZ@anlS;ywajguSBFLKSz
zN)20Z5uosbd7DTG-iO;nEiM~;PBWJO*ly0;u@L7HNhX$q%2a)9*f%)M$3`XpwoK8=
z>IMP~kzy1Ie62s_T4dQMAdep0w$m4L-ZK94YfszLxtWJ#>XFgYqG&+D-X|n<U&P%)
z*o{IqCu*pUPrAWHYE}}Hba`b{(4n|JE1xI1xkSlHp#jJVF1ZO?QLs3Mz-h`m1nL_4
ztLv6v{M=CQSA6^Zk9^@)g(udeuOD%6a&>gG>kHTuLOT<;CC3nccdDR(3Ee+8S17zK
zSFNCz)lc`Wwu0Kw%+_wHk1T-A-`l=xXvP&4iT_rxbZMymRA2F)8XztrB3U&7;1DXO
zb5U^#m1br-_VuhN5{BCk-<hk~`=h#jI{L)Q#j{I$PC%^v7KoGAfo)o*W%%y}Mg144
zY{8LaauqrD6+f>Z6GI#m!4CRnQhvBG!S(9PO#DxP8^BRfW<pPVX(~$LQ?F4)LsJ%H
z+Ty>JA`GXUlxdxH{MIezrpu`G!AYa`Dy8%Y^Vg1}-}C88Qkwq$@{c<-3vsoYfB6>C
z{IeTUdFtnx@1ffaTnwK3R4lcpFp*rXGGG1&a7wsLHsHIGn$joQV+%yv>uo71zDLkn
zfRWY>(M^2=gV?xaG4JK)v-ib)S)aL6z@XmXx2EPk6WDl*4ECxGZnISxs@cL&_upQB
zqh!SVHJ;wP&%*3Qx<FI=uN`eMY!egwTNJ`UA9@ud%Z@$w_<i0+2(CC&u>PLVNf-10
z7oQ2AKZ)V7twv#c*wJrATn~pp7Ex<^YHroy*}6!628;L$d!rLSQ;+i^j-;M}qdPFj
zDulJzV6*{0kVD`)>r*m*&#6`OmIlqsPv_2jB!npt%>K5g7Hk3Ak9G^q{UV4L?;Kvv
z0~qq`*|QMb`)1ar2CACPHobH?rE{xu`K5)}2=v7WhYj#6UmXhw0Gu@SAr(TxjOuNu
z0=_n_Q92Cwr~75eKJI^^l#T=IcHt@mAR$Dj3-NCq&AhnaYdacprHifiJjsHQ?+5L}
z_;^i7=Ui|^^H;V2oc8C4>-P?`g}-V0g5n<E#_KEdz$loAAOYJ@8HUQORLg5`(nv^2
z(}Z0?^azV!N`W~rH_tZdKQ^4M^Z}WJ8Ju9jwq8t3UWZ}g-jfzE6Y;a1Rn^p-P5HgG
z;Q`L$`MHz}1`VjRSjBNYSlQTo9s1qm<#!5KCqM{+-Op{HX)ruFbnV}Mk%t5q*~P~G
zW#i`7mj|#KP<;brEc6@Pn1%R^ha2Jnk7`)Evb@@41yv4vi^JGYTx6S$<-G60|3_|a
z;c7pA2MD@(8b0^TZD_2bB<Vk#b85@0!cp4xgZ{*@{>O~x>~17@)EG58gehBq%Mji#
z%gs~qER~KWWpV%|>(h6yCaa*u$hcqzFNsFSHqR3}8(5;$Zi$|Is;fE{bcFuv&&BUt
zd#z%iURHE*<(HgaUCp{|h4uPTozwL1@d5wth0D0}g_)e3Dx3bVrleQ;JC!y(Ti?V#
zXlo>skf!a3u~4z~nUx1__5a}r7F%ZWHO_k(yX|#6<I;I`*UeYNL3ADzOca}y1D<$#
zq3HV4-yR|kjcdA3imqaIrO%^x0kjgnj1Km|Ip_avdAKq8ebFH<{+?eG*f`aKI13gx
z1~xoz4X+<vHRF){`8hU04F{Ev9oC$n%e&YBTL29qmH_2Z41{I%_QUS3uJcda;Htj7
zYI$`_Q2Ej#^r@nt!}s4y-@h7JX6tfTEQ0(8=(WdLS=6Xbaf@s~2vZv^X|HC$Y3ZvU
zyV23w`)W+2W_^6dzzWVmqwC+Ww)t2-mbG=$a|i_5SE2hytCm|DFQI78<Dt{IUUVIa
zev>AO?Rj{>Sf_87AQ~i@JqV!G@xkezrLs5aEvHX<6VDj0j)J-yizlnEDtmqdpdIvN
zZ2Jl@jm`ksr=GgUNy@buC5?C)L7;`tywT!y>*D%4$*`Z%h<ONG1_zH=_n6J<&N@(_
z;m^#Z#xv~}?~O$^W=@<&x4q3V-CE?&e*KfFRB5Q_A>AGqhZkchS|u|R-NASNjd9ao
zUN|!`SzpflwuG%islDZQjtB`VZS65?b~(&OU$Bq@>U_p$9d$_qAmLMU<z`hVNJyA`
z9K&*XklF|EYKgUFT+9b2ch^NsEbn7tP+<XN3?fd@X7lilPJJXFTr%=DaT-6vLXP9K
z|B+^q@%`$!m`%ZojDq?F5H@)UYbq&?j{ZXH93LAUwOCrD{gnG}18gH<f{g$TIAi3x
zs8feeLQ^7{p4Q!~^9}7mt&^s0P2unf?Y1FFuLm=-2d=6Nk&;Iv2bShFqNn?To#wa$
zPSx6#tS{M={~t}?0nX(cw*4Ua+mgr#8D+0RwnCIWGO|i!kL*1XvdJcU6SDUxo9vaH
z?7jE;uD<tue}|)^;&}YV{oMC;o!2=Ey8B~BxL-X!>fW<_UlznkVN7S)#J=;+$<|ty
z=_ZF%eq(t6{p+~N*qG@s(=4Mw$(POS-N`=;<T&;QqgOw38@($>Ul4w49j7e&z;<A)
znS-T*>3$X$B>+K#YqM&tSk5R3-FRItGiu{p*}KZ3TM>OcWE8y$pf>zmWm%`R$1zho
zh;M8cr!Qz+HAB$c8yLV6nwvgWd;G!tUAa&vemKnwIqt-!<DR!xcSD7Ra<fL2{?o_b
z^jdnE^>M^e@!&yYuEH0(IRlPo%CF~Z+hJM5FRv`UOe=!3A0{q=%6?QgDE+!Sx;lS#
zDXOV0;y*)-EVJE%0*p<JUxmf2Z?6r%v;CEQ@v)}nCbU@&bGINB&f1!*Xr}HFH&0o~
zV@UxOwK4H_dak=TKxGxsB#%6+DX16Xqt|}@x~px1a9S8@O<*cq+<K+-8k(mr{bCr!
zn|)(d{DL8F;{0C6JvFtVFfQ(enO0hCY;3xlWOQiFnQu<)6SBm&IYp+Hrp@fZ&5ORU
zQCb+=*_W*wG&w_0@bYzAEj|d0Zjp-75z^DOY?_;&2oMnwKg`V7bBu+lR;Fl=kLO!;
zSG|`XUNCxzT76&llPL}kmIdWH5!QI;kHU9!-^Uql=ai-k#mW4Hno<bDbVDJeZ*pQ~
z)jmgm&r$G7Km^%oC`*C*4%+lUe??z_7rUu1<xkN;`7JBk1K`QZt&Hq^)|CVhv%-LX
zo{%szGvg-yMjw7Yu(Lj3zMwa_15-6`jH*K}%7ErM@zypqu)^w+s(mui-QB&*JFs%l
zk@Ne^_c@=&_?3m+pwhi(H-O;|UmtI0y25-DzG{B))BdXr*BkLL^1tPUgvbI<rKE%l
zjEZ9jwma0+Av#(|v2{C~O>$TAH5IO%p`-wz{|yA5At-SHg93#A?Htwm+S>WJI)J2y
zzlpTic;SA`s*sbB`PJH)l-$NFM4O*m4Wo^(Ki)hSf3D1+Cw8-XKcX2}q!Q2DU%q^a
zN4hXWUi=?ZaCB^>iJ3mc{esmez}y}|0p?~Vz#K;N5dsYv_$a0QCFF`**f0v!XMWku
z6d?T)Mjfm@0+gySiGX_d;ubi8-HsRneq?Ja1KN2l*3rAlOUm2$h?H)5btXNl5cVYx
z!~*B6Jb{R@wT&H4ri!|fE|@(4W#9QPa8I-E7a`QFLV+(cQ9ScY(PG+x2RuBiKoVpD
zHy8+gG&DMYwftB97Nlg}ox=lzwq=!z`5yKd!E*9aGPiEuF0I@oC%!5!Z1^$KnM8&0
zke&VC*I4kq@1T^XXHD$XOCTi&FPr~_hsekiAcGFj_J+a`Z4uKSCjZ`#_O?fNWoX6R
zG|5eWxbO)*5_oug+pl|QSYdtpFP-|I#;>}!*Xf?Mh=itntOK|{?X;9#r<!ppSMp!=
z=uT`<Zxwc-YuMczf_aB5xL5c2)dG;}!5G^C#vDpYqW;)Gu`3x^aPQJBw=flmdP#E&
z<huUi0Q+H_@ys-wQ^&Mx7VG~Q$5vA`wlS2yem7%=Jb|k01?vqVH?Ie1g(uG4u}3@Q
z`)6Qb<0o(BBuEq9F=qv{CRh1;Bcp@AI=aRivYRJGc!W7YFoXo*99U`z4<%qHn{!^8
zlK+Tn!Om~y@xRqd)Uv=lkuO+=QTO#CR-788X-<9H!QfW5PzXhUkfLk>x@T}m{^p38
zm~`{CW4TGqYR1Q<w39VD<TqLILmQoMUUk9<i&OoQ-kP082~J>pn$2BdN=elTRus?n
z<d2olH?)loT~rLrRK?hkjS#`eQtGF#lrj@XEL?0v-EHop_|4w7Q`bBzmAVnwO-6Du
z*_~E-gMzY|gYr%3uk0BLyIZr$aw`s*RPnJr9xFZJ=^eh&b(R^tOrMWAZcqL)I@+O%
z&`Nmz^ERI#-p%i=t9{;R_9#saE9pf6Vh&Yif)TdGapS1S*r284FGrvDIqBCgr?-O{
zO!E~8)r<@kHMB<EkMC)I9YYb-Fg^7I_diYPPDjGP_O`~;{i*M_rOll2+D;$1AEWMJ
zV<ltE+}3hBsS2&S>tmUYiRaXsjP)1)XG_K~?&J|P>yK?)TU+WvH!DjxZd(`rF)T{*
ztt%<vkOpU8=d))L8jRdR;7Ba!Vk`Ol`5B0Yz-oQrZ7i@|G&O4@yr=z}U96sH$;v`u
zdwytJ-h;m9PB~F~&#eOerk~^a*M|iE(Q)~|%*3VQcU&GlG)hizxM0_=_lcHtbgpx~
zH!7*ze0^kqI1(NFI{$@E=ugK?tF1Qs#`N8$n=MtiR11gMbXFn|1RVvFWF7z@^^Bc?
zk$HJ`k3mICUTqTK4~UR)1gZy~(;=}RbT3cXLJciVZ8rW`%$*4XU^}_x4N6UEY3YGg
z9JZKRmr37LN;H4EzrAjLuyjSvSa^huyL&Hz`Lt!xi*s|nucsEY2md(=;sck2T;%=2
z8F%-GClnye9wR_tltedREYHZus#~Y0t?42_8G4NWpB}b;4Z$<uo$ZvyAYqW<5}$;L
zbd&TYbD9%+_+>;M_=V!9EM&}4rQ>f-;~{`>F^^j0<$P0prD$Pfwzpevd(c!!Y5v2`
z?oe9I5?o}!RUc9yo@rg|G~DaRi;YVxecx<P2DrriIB|Y384xopnck+pcz19rJvLw+
zqn`b+Pv`3sH%cTUGZQ#jfFg^4Z)B<$CGx2btN!>8=R|?@x1srmeR5}7O&mBvb}&7M
z!Gp4*Dwt=Cm)LmdGGmjHu72*gTei=)-8^gi@~>F-Bh|vmFNymv<W#fD3lha?5(P`f
z;Lt=YW_CkEWMcuyOm8w|h}$j=QzPl6fG~pOJuL8B7f&DBb-aMz#i`RLEO&T_o8t_m
zrcD(SJ!=y{1waL~v=lvSC}(bLtleO5qxb=H?G~c`#9_2<*2C1?bh5;z@$1)(2*1j3
zFK&h>3thO4bIpP<8jFwDRK2Rs%&8LBcu}00d3Z1`?P701_GgJEgG`1PJ9x6=`!eLs
z{Yw{h2W*2prdM>?ax3SSm~Oq3U*3l|$D>b$&`B((?GWRqTGaz{cXqjc>uPDaw$C)9
zO=}--od_n(^#3k1SiXuIdI(&L-6$ai5fhV~<YPXjDZ{+_#D5nyW@Lm&hMeDyY%wlZ
zf4P$*{1iMe=YRxe*sBY}N^u&m=joyMgj|Pe@YFnGip20Kg&-;$7E?@nDaWIK3UZCV
z(68`2?2>^+N%R*;<A4FOfU_0Jnh%;}otRF2rUHXLS8hSUbj$CZZ^v&Ne0FtvuY)*W
zeELi!{`Be0NbByzri-?#>;DIyw$n4q6MMEs@~^Mi?t2{<eicDSK<EvB?N6?W$3j+C
z*AP)CM-uvXB(U8@?5&Px5))96=I`3)p=HXs`O%lVTL%4C)0*REy<3-(12QcC^mV{f
z!f;muIwC3p8%Te`N^9vuUmR-;&pXm2X}3uaAA9^B9IUzScma<5z2`gUYEL2Q+}OlK
z*!3sIKmFD-)a{;xOL9PD_;Btxww)8Wvh(34tNNHL5k#=-L@G8sHVhm}ew#_<`h_i}
zfcWKm<ojZ?g}HG7#yd;F#1)ZgTS6pkhYoQ(Snbs6CN+>Q`8u*0r!|_}vX6vKhn=qs
z-F}6+_)#KvJ4*<|b6U|)(UXUNO?9b$la2f(i6L1x`Qt~LY*{*Jo-X4kr#(KGD&NS}
zqNhmR=nKzEL)UU|%*IGqPCrg0Q;Sw%?S2$FblmN{Q;8bL@qb(Zj!MJieNTCLq)E)h
z)2|o7Ys2l`K8>+8ZJAk^D0|WN4$LBmF2b24!-a@kxN2zd5Gh9*L`(kIMTnlmLiO$~
z$<KK~fefLjp3KITT@y@lIQ$BgX*qhvA=xuwpZ6GuW6%+4ogzjHR&DdzPZbppc8{$P
z^{e;VPMWR9cF_$<ayHV;Q7=tROx&;j%G$}Y#&ok5<qVU=%FD^TJZy9$WqdGIO$GEH
zpgn}t`vG-x%ba|s(YeIpkc;2Dc)5vl?vxqx2C82(DL}dlnD1cl4|fo8Pjb(qH4q{>
zarSGE3FMoorb;)BJ(GcGY38c;&sKr}Dl-Pwd}@OKo?h|75D)Q$*u8(|?A10Ro>DZ4
zyTTNX*I8Au|3vX&r(b@&ncDZjLHv#^@Lfnh4{Fiir-1iUgz$1^1|8LOJ@LVxQNVV;
zC=FhqeMG2fnIGO~lu&*C;Lle1I!Pb(Vhbg}m(E93U}m^5w;9&tvh2p6aYt;{?LVi(
ziqNdKA*7My$5r%F7VQ2o;rWRzBi$$SqPu@B{g&hPQiFB^^n!o@ARSWD>6Wh-o?{>|
z)kpK>O!u9)ssczUBg^2)@9{e4Gmif2tdcK!K;h}=_7$a0Ie#rDSJ7QfDdIFbIwS1(
z=DI(u9TYt|tJ-Dfz;_z^rhWqfy_#YtC_pi{{s9yqV*;!VwaG7ZAV)dGDY+o$n@)q{
z)%-cHg*Kn$edQxc{vUTbw$5Zjl8?Y>IF8?Vsy4>U#ojD!U=_5w;B|l3cXP_+)ARHq
z&H9b$?i6iJUXaOXpNAZ&rx0Yx+`UccXbD39h8f%WBR4oy7Jd{ghi)ecmZ-@}@Bz!g
z{56N+d(hUlVlUb9Wt8NdwXE(#7jQ<O{GBB9VbHx-%F63*OQ;C-u^(6_ZVJc42Z({c
zfsf2aQWGpp`-LtpMqQr0M#nf|)qEV<sed+&Et;U9I%RmjNS<=s<?tsYe*Uus`_gtI
zf(1_+tZA@uoi>&>I>vkmP-~Q9WtKp6gHQ`|o(|NnL(mf+H!jkvQ}tR!>o1?af>B5M
zc_o{_zhdFk&UvKno44=N3Npv1roMcM$3yP=*5gUc58Xz3PjuLr*iPNrhC#@DY@q`j
zUOvYKqIBit)MM&nf@j$eLGgm`vU>t%G{_&=`6=fElSLXJ&g#a$MJ6exru6Ki6t$75
zv8sv+Y7=yRFqj1XVaXmA(myr?%s(NQE37#2y;^(WgVuS_Ap?u$^yG-T?Irael$3&c
zZFx0;jn?a8htao+i?dWBL(RVk9<}sOQ%|!p{Kpia_m;$dInT7cX05Q;-oc?VN6ODr
zZ(v&JUZaRGM@gG!e*V_6gKlWwyw&{BaqpB(pHnh+_^mlTfBoy!`Mtis`=_8Mrwf;I
zxtg%uIzi#`w_Q9l&v>e+saa{?Ow8Dnw3$FT)vz$Xctvx5=>KRL<&U=9%+WMPtW1=2
zHoh*ujlPxsESN1yt63o6`ZXRrdCc-l)>22<ZhtU$rI;4NYu=AapM-jj8*yAU?i^`v
zmAqOnnys6R6aVwiY;pDmO}K*Mh{kToHA7d|%Z?sOzjkJ3=2VZB7`8LM-sOFuh;mu4
zP_yH!(b0i)LQ6#}Pc-vrdlFbb=qJv_<77RoteaG_?IC1&=>;Raf;b4sX7)T(lvL=x
zNFk!iDQN+p@4a}_LjNGJ^)DNbD&CPcSeK`C{D*XOxDE_Y$vlXF&ol>11)+}U$Pv}(
z5cT_%w7zZKLUpNmW{ZnxIlDlF@NYbg5q2t~mfyLPNK^S&%@s%e*rxd%UYb7K@1Iu;
z2Ajjp*(WX1-*b1q{aTpTYnAW=QRpdvjJ8*SU5wSaw8@b?F}HSaP>8Y?6>9v(K{|Nb
z<@;Pdk03(y(+v42|9%+Vv$TjEMKp2*p$||>@oKD%F8$UMGpmi^*@-tF$?%fQ_Hd5J
zd!lm3<}+Pdpk>kEXRIq{;kRu+5UZ`8OuY8xy~%ZiquU89N{U;FGl*jy9pSa!?jRHS
z@wO~Kx)i>O3P}(Z8xu(SOhak9`b9gbWWPAPtiA1V^nAgI%WI0ps9JM}09mT;q5gv@
zXXb~v>0(M*E>52tt`n0IfYp-4Om(Au^Abd3g2PEbs~$vgjdMn{g@V1_^R`>53N3S1
zk975UAM=!O57(lfp@};75yXFd2A*S2--ds!=@$#k^dVK2AVpw_gm;QE5J4D<m`jB=
zzm2%0<|E!BQF%3TM4>6=SDeR0#s}ba%n7-GdRUBh!fva@D=-^xz_{%17bRI)9wcQs
zZusC=*3x^~3g4ni3?CWFes--<&-UUE2;9l>F}kr|&L}j}_2wgxKW~32BRhLm2BtZn
zx3xWyql7F!gKFWZD56PS;SPy3H6w=y%06=Gscr3T^j+-zKN`A&10PnJX=y6G&JxY+
z@<l~{I@?0d()<3ObAnFMU}GUxV}}mEY!@LQy;YbZ`bte%LfQM71f*8g6p9=Xx$LYk
zA_sWbzzlVB@w4CC<d=Pl;)5FAL8r2Z@#5oiynFg4dJDi^{3rl+WmQ@_&y<*u6(iHk
z3}LWf<m4KLKb>MC9%j=BCx;52-;_9_hw}}kVe(47()pAI0*crk7S*LGDTWVZtKowc
zD{Fsft3J<To(9*$tVh}>bVOJZ8XBt+-lCaGki8Lo&t*&KGw0*}*+$dAUzSo&z4E?&
z%ShQtWpg)!ipjBxFCifwB2`c>&k)cc<Uwj)r30Dledi4rQ*!IKu5gS?sY_LwlE|~@
za4<t$EqY}a5gn#DD)Q-g4V;VL8e1$G&7vb~WMmmijPZYWl7u?+2mhelN+iV~!hUD`
z*0)gM(d~EfA3vf-fo)ud(rSd^?mwUR8~);hLnFJpdk9o;keNwByyx!L;FO@Ou9B8s
zu=vI>@L3tsB&y5H%?e7<so|EY+-kn{#eDWeT0uOFO~N_|k_J0Hyp>rD1ROA#^ZCQj
zlvP!q9l@k?)LHS*E&?>(&qRt2n?iN4aF<CnI=&m}>3{xv!>vvz2<OoM#?K~qH-wt)
zes22X9|f#IWK;kDa%1HGJidiPe9zyY>pZbr`I+l|12y$q8fqpE^gtgg*gptpnvye1
z(|B=Mjt<|`bjQcVL-mZ!?n9aer#3cgr-sXLlyAQ?s_hHwgqIZqR&B>k15l*8yE$*o
zR45X&obG$K{%_9mg(fihzm#BUZ(1hrpNODP&Yo$qrpc?HZ;p(VBcj<u{mKk@e#)v!
z-*R%q;f`+b7RU9!fofgszMXaDV{cCv&93}I{OyiXZdqz-@%~b*yi%Vf^<Pa7$@2d+
zk(6(l6v!5_p!nzwar>V2(SfEpls737pVns6KZX5x7TEh-q#;+F36aVmL~@pfgQ~et
zO?p`=kiagpSsVTFf%isfB9qCl+Ve4x>53;TSaX-RRg7);kPb3WeOrY;Q52LUt*SPr
z!I8N38JiJO^fxH3EUubwx_fg;oZ4hVc=KSi{OBtbb2-rooPvrrWkeK}5Q$xSd_R@t
zbgy=Oh>8V4Zr@4IOu~Nt8!vZm1dBec@6HLn=ofl8{yF87a&t>4+uQEj-33|0IOhtV
zmlQQM7#yl7{R=Yy^Ae*HpmQh6K`taWx1I0Jv*dhiA#BB5qOv~&dC|DTUd<1|<(ZW@
zkkppHQk#XRL}YOF@xg-3{RlAOx(7KD<^sowwT`ctqL{euvQmU>ute@ZmC(Q}SBNOM
zZ;HIM4~KNIukmaG4Jj!E2;f(UXx6S*h8TbIa$(!rsFz9oRzJGBrrN{CvVg9<C{x}n
zLwzuw$5(L&CgO;S1^~N&G4}p)W+eCz70p$#c!Zl9Ntw#p`B=CcKk@q1j+>ho#j_@!
z4EYyngiYDqh*r$~x;;z=O!dW?%#F;%DcQKnCB@xg1Eo*`Z9@mfZN@P$wgGpFqj%jY
zLy8HzjsSh=nQKok4;r}A?;FZ%HyW=`pL05Fw<Df@3Ht|~|LCLEazG(QM+`C>r|nK%
z!7^LIf?ej0fh{Ji(6EHiJG*c+2b7aTpc5EIV9Z(PP;if^N^q2^F@^u<12bO4JU?FM
z!5pMlF`Lw@FcKYbA$)c;wO>RT%uGytLsiMhjwq+mz_VqKj>r9K!x*y1&^>ihe*p2d
zeLkH8C*S>i-m+9zVOXd7g>bUjW1nX?5&mFb)S{}Xkb5@5<1WwuyWs1=7(GIH2;R8R
z<hBSYaR5V7iX6ZNI`n*U@aHWopTtgGx%;q)9^7qT+*_DZhse>T#-o&tumGIrHqwD+
zYT!gzc220KY|!~QbfBCdCX6U1j81@3;}=*18OkSDE&o&Dv3(V^@d681TkKd)1Jtgd
zi~apZ<8)FUeb11J(U55j1+6q35vo3v%b&)N(J5mb=DLE<@u|``qlw?UIs1TaOVF9(
zyoe{LRX>F4(M4Z@0J^oT|12?&=DXqVKWdMufaZVFl?I=;U3qy}GOI5|mOSP4fyOfv
z=e9ADLq6NGa$sZC+VrI+%W6KXGnA6XUUpC5-3x8~IrW&MNiF=xAFFD^KHcX>C)=WL
zJd1qe*K%#{H5#^^Bug^+r2XY`Fs!AT4x>UVc?qiq|J(h?{J6-;#W87QzaiP?rKZ;S
z<WGEC%En<5LZMYaHYx&Mi@3TeI8XJvP^}CwghwH7KK1+&`k^VpOGb7ezib^{zii3A
zTT5@SJn~iZ?el5y&QPLeBH%6Nw~0m#$3nCe*X8X&j&6Qm;uiAc=O{a#Z4ga0n&urW
z+b&av*nK0;37Y7ag^N>UCPFJV*|0q6OZ3uH{AvHX7P1MtrtR8tUFx;?U-*9X^O4sJ
zNJ{!8%zL&($IEJHKUwwR6NQN>Mm0jMOm?LNLLh`1cKB~GMedgdy?J&CUm!CR5*Sk4
z*!7e9uZ{e>RKr!6B>bBu*r)qK{C{!Z^ouz4sPAxh`_q9*Pv3ee0MSv#z46!iEMXbu
z0UtK8jOf|!pKkat$D0KUP4ecJsHZ=a@!03V*aTMqq<?CVf1|`U6KP5rLZQxZ`SX*H
z>VYeSV|Z+<W$%}`Ejecl7$MXI7n?QfJ81%r3*yetRgJ9r>ok;MGdRO@rV<yUCUCxi
zaZkXBhlh<*2<SP^J~z{~ZOI_`Jgz-vWF(g~HH`6Ds#J8l?BFU}cO!ba=A)8SBlFHE
z8_7GQ+cC`67H1S#=0EFQDXD@rfh_9tOjVy_C9%rGU5sM0_Un@}fqe=4rNE7L`J~o|
zS!HErGv(&#UzHWGXZt>jdOr7k7A5q1{_03q(rmhzm%Z`Lug^K*%{L7iI+D&5X~OMi
zNuG@F&<ruvoJwa7vY%`<&4qRQ|02Rh=xA=$bA%mYM+Lo?)Yk6lX$7;%#l?JHxb>hL
z?GK}dlK4dq=|jBD!NK9k5zmLVOwkf_gtz>~^*wHR{HdpjLPe7MB#A{&U&<F4x^_)?
zZ+!v_FohH)KU{x7hu!#8ucU?~$Xqf>Y_G7CAO1iS{yWvo>U>SVk8zK+5q^z!oi+3~
zF3Y3AGIthp6mJ{!`melpoW-FmD$b3QOCMN$Y%7nwP0D3DkrCRxF_x{?M>!p#G^pF)
zY_q=QozJ_)@k-@E5IN>GdV}e=h>GZ?B5XsdiPFvfBy(nWZO@GF*O>t_LAYW22{^D{
z+A+s~Jhi>`VYXpIm>WK1TgmZq@+cT^UT@xd!?7|aB*tG|(s)N4RR`cYF+Ne!M_J+?
zY^jV<&r%<)%$Dx%4Gik{#c{5#cb_gXFf!RM>70fF>}jgT!p6uLF?|ANiN59C6W67t
z-Nd_zAFRKZ7H>t4{%H&buP;D5{+$N1b@_RDTR}<Jl{Ace=KY?Zn~+&5p__L_(8~{T
zb!l(!%-;0#wx4;-1x7h{?vR3<grSNNKGT%%pQY#~Xl0*Pi)8rI(ICU4qYfpQY*Kpr
zi_AoHA6Yuup5OfDF()nZn}GnaSe{-EJSC!`&la{c-XPPUyLDC_dcBez^Yx+Ml&>rD
zUA0s{jFfBSu9rH)HA`e3ro^2no|sIPI*0g`_XG$sW@Cv4hY#QJ86k3m{ASl`q&iK*
z&Jq|M-Lqn}QPEn;sR~BPNBJ|gj#u(Ltw!O|35_t;jdhE=|0O%Ah*ES33a|FEyh0yc
zVQa?o*(7^&>pqI{3me9WF5CA4NYmkz0HV-bP_Ij{|8yRolCRo*Xf$$oIWdlfKQSw{
z(0>2O?SV4=69!RD>HVWiNs!jCWYzh+oo$U4r6wYdO0n`BE(>aFdjR)3o?W<btf|Je
zv@~!-TUFrWQ5zekA$^@H^`CW-cgdbL@}`tqMFSLLdPLl`gqSEzKYvBSRbnE6l>+f8
zro3Anz1B8%sGcD4PgMY8vzOgnNB0B+ekz6CFnea5r_Qp3eZ#-gIB^xP?DQTYIz6j<
z1KfF$rqUC-1G6AoeSi)ZVnFJEhdgNWh3J8&&}tA!g9zSGU5WttcS^7bhZm%V<QeIb
zAfH?HA`SDJYZPJz){hT@zC{z`+2ScpvgSneUAznkkiuBl9dhJ~1A$jBSfgAFS0ByD
zCqI#XG4N~nF0v)$ii5P$XI}aRh1`7{(7p_AR|7;uChKcqX8rlRmeSo_j6_|HyUj%n
zsc&b0yjy0|Dc{z6WM;Jx!rF)?sv$WcR49k{59;>{!d@vfoo)4f_CR_r`9GjDynX*(
z24W{(e?jkcwSd?))BsI!<*AvzztT@%!KC8A{9{JQiR;x+)eXeTygSzS=dN5*wSt2B
z!s^yX!g^42(d}QH7(;pwP59@SqUk60Yu_$NmWO%RqcS3GA)#u$*9~yGt0S?$8qc_p
z6zNye5dgIm*RuHobX!x?=;lHb3LnNcM7cP+)R4qPHgvcLPhk#3`Fv^bTpAc3Xo!$A
z+{Nq!bFiattgH(H-y@@=&^)8F9w<wkYRge_^?&|EgNWiu8uvX4x0%`<TBf}jEPgt#
zSW{z&*{N3?EA!wE@tth}TNYtPQe>vC`_y#oDzXfRiv4(nvzeqX0TVH}<FQ|>Rs6nr
zkGpY<*R|kwbK{TWN5nZf3{*46BylngF59s0Z4j?)uP$-v>)Xl-w1aPlJoO)uFg0j<
zL6$&G;OctoOQ(LZ6x|gA!`88gaCkV%i>+-D*E%;sixv!v<jqtChG(hY#qMR{{eaN}
z%Z<ar`X5p^xoX|(U{>+6g>cs}{$Mv2jMhB<Xpc8AASZ8{&@SpMr_O9`kAo+5i4TL@
z=U>nbK6E+Gq;8I0+0Hu$kER-a-LJ#ZG7qqQeul;XSu^sb$hT(|(k#)g9NtiTiPK=-
z_!t_hTYrUrLS(&kVDrl3i-p$}RL;v2^uB6J`i@RjtgkD4hmN9jRg})S(bzn;0Q~n7
zJ!!hPVI!9Th+UwLZbvl(znQv=)>jIwz?Z58lLxCuxBZj9*7n`MRuxTEn3zqEk533q
zdU9Ln{YiRzv98B)(@7B=jCm$&k0&YqT<h-g4Kaj^#`3@2s0r45-*-LoV}^%zpB@np
zB@J=aS(s(36Pg0Xx5G|{|L`&?s#ibcOaJ~O#n-;Rx+ea~4-SfV3n#g7lL6W1hM@ae
zi$^Sul<*8cB^oPDZ(YNR)$ewHa}R-S;al<>7X9URkA?Sp)qAmBMYM}<j~iBIZ^{-p
zCTZ_a=#B|wJ+WU1#yoevpkMyxkN3_#Yg^!Wsbsl0<;xNm8hyQ5xn-*oC-TcW&iZV`
z=Q2)mcDNybnO9%6(e|wX-g%t>4$>0cqGpvRqY_j@T468loh5uc@Fl^?A>oGX&{j`(
zjx!1p^%eSg2V;gPDZ4T2ang%rt|FgfEJ{|lng>}7JdLT}tE$VgKUg_57w*+gQ5#Ji
zUNl8H@YbxIzp~ETA`vW6EVmuAM)5qnVs`$z$$K;!88L2{toHtc_>@ttVl<15Elp^<
zfa3ujN@ytMqqK;91AQYPr4Lp*03Yl4$X38*t_#CSQ^Z#;JD5tyY^<UAc2-Yuc?Cex
zh^KKxTj!XBvt_*TIkj=A07&gYUkxE#FNT)y#6&w)@wE#&+5bIo;(ql?`(8oc9V0{F
zvGzSG$?ApZ6SY5^b4Lpr7Y$<VcSwWSDn@V3y@jnN@w}Uh%m32`SB{RRX?F@Tdmfq;
zbeFev@Fxo=;~l^D>L&uIq6%t_Lg%4NizFd;c_o<4anr->|DI!qw!1i02?xptBynl#
z%2%cqf9Tn31!bj>i*2Lxyquo(MAeH&9~@ne@_8p-$A@1!ceD5W=;h&No&PfsA)idI
z@=6Vi&gz^J_Sd^+-L;=cKJ_w5DKdO#;N@P79}*N6=jSVl>q<!Ov2?LmxovH_z6PR;
zjoRNgPMlGiy6h6H_xDf`AzxPYk^q?p<F7S)68ZQ_bRC_akN%*2s652=vYUBWMFaE#
ziKkumAFRP{jM2EST|Q3seuM_gQeGzZVltP(&sdFpkIQ%Ky@tE~uSU2nXBX|U54Jf*
ztp`sA_T4oq%-d(6`{&)Z2JE9sQEWigTIpa#R?}W4VL#A2(R^ZG28n&_PUrKCu_#CD
zWSz+f+o`*=hw6?HE|S(B1j<64nl)Z<=e0nA5uHxD_Rxb6;5KOOpC=<=DHAk$XM)?X
zO&^31CmE3&YL?)9oR-<?ez?R6C=~8Z&*6ih0)eAv2~Jxt%mn#OAs80a#80&^w?te8
zT{rYWsSE!9TycUQrlgP+8S6}WwxCG|DzlS!=L7SGm^TCiJE6pxu5xOhii=ou)fIn!
zpi}vw8(`q1HhL8m6-Fj5@XZp4{Ts^m%4oTX%l+3**|G*)0Yt*AOl$>3mXJnOvaRQR
z1R>)6tB;8y1_W5d?VS&Phm&fsL(ljnDSV{=cjV<-aCYr_N9x{I%?(r(+&^iV`LAAS
zFqBuQUsSe#DkWhPai&PPT*F0D64d>gS2kCp!w>f=gEiT88(SCmZsLDl86A<=l#Y<0
z85tSDIB_;MH6|k^1Ij6NF%tkb(9yNa&-;ePeaq~Wlap0TQECq*orz<l4gc)1IpML}
zWw&rR$}32O{3ZMXPYTBzMemqO`02LsXn~gASnEwN+E&i)Ii-L2aCUNtm51%aa)X#n
zmONx0L1Idc)ACiU|0A%sgr6d{J<FN%MuEh<-m%H~7iJe7yUM?0g60X8JM_QA)M6w3
zL_`3ldYnzs1F(%1_x0-;WEc7P*jmnZ)Z**GB5C=ME?WJ_?W~X>C@AQ8$VOyeAGoWS
ztn;5wti5xs6a46nS~ycxYB3)z8T!siutA-Ljg67HC`Yi$5Z*v<C=2Y(C!e?<SaH9K
zmQgVq-uaOa_r`~doQ}^kkae9d8$1p>>}Ejgqing)WE8pi>4{;H0pB)&0+1kJYBHR1
zrRaS&TYXI6L>e1^_uhD8Gw_<E==gs-kP?mankZcV{W1t;6;##r0$1Yg%8W#|31SDU
zcWUfg|1#cggOI<N)D9`x?lA4b8>n+@r}J1?Ds<#*RCn$yf)lTN!eh1!O*P8U<*hZ>
zgBH@rIZJqk-6TEhL;39~9Sz*&c21}01@-%TC+{x1o`$>m;gbTCS*r6sojf4de_(Af
z5C%|>^j4^Cl;<-R9v_5J{8{I=9F1I#(-5dMi{-O>HoKGbP%b&fj^c0&X5A(I4d25o
zZnp(L=6Vc+`!=`-M@ELj>@=h!+|E{XUJ)}h^jq$Dy2Fdst^YtS)E;t=+jeA7H|^%3
z!-tYax9!w#i5^!CYd}Tkn*RxGc3^Nr<U82Zw|t<DVfi|gtriOJ?43KWd?Z?}pBw-9
zy2>VZ)G>^Os4y)Khu2%qEJdoR<mHyAUevtl`ZVB8a}IT+aG$FeKNl!^cGLY<jZ7t;
zf0D#q1N$f@flHgJ>L-Por|Yq|6x03iL9G2!woeqzR6nQj{b^cpx~~N8hM}cl!q+dp
zQ>Q$Z>IFF+cLlgiK-r+^eGZ}6#CIymgxf+XEA%=)K1HMP#re#@!t|VQ1+n_uzcoOc
z2Cu8g(#Y|gO-;4DAb2EeRuwOhj6={QB|7W*dsUuz0ojVSJcC9V3K457{ZHv`F=fkf
ziT>KJ8yvTMy2mC1BTC2DF3%1i#X}6@af~&bj5VrVTTiqnc>sj2NG8%G&o4+AO{#T_
zf|%wc;SU3KanN4T{A`8-J8<%s5cHJp=WM)OX~$WI%#yfN5KA3IY`*>dkBEov2zy<2
zln>u4ufeIQj|;m8w&i05T7*LH#fO{Rq27tpVCVaa=gezA+;0%Le29llDwtvkLPi-y
z%k{Q$RIjF<R^#)fLwMKb_D@zSu3iM;;D+4$y}G*9rDX8@`N8v;rq=E;37T7R;_<vL
zYZn^Iw9bCMH!&6$7S&W=9dA#XuB2S8L?s5f30wiMg_Vu*aDP{;(z>(h%FP!yL`6kE
zBcr?__1+^ICXOaZp4*(Qc7haYZL<Q4X*X&Cp@&6pAX%z4(_qhsj<eyB3X`2tEnO}Z
zEn#4?Vg1!;*Iy-xo7&RKm+RPPG&6NQPd|G6^CkYhID1fVTbfRbt#6|1ulGU`VQv{%
zXhWQG^e=Va>Nx)h??fpU3*$NU_yoBSN^iktuqI9Y0l~S=J){zKV|W*yHvh*37~zG3
z+PSJa<roV9(f1G9z#T-C8uAxvYXQrSUf)_XGt+Nc+yh9vHer8CLs`Eu4=7~wLqtT7
z3J8+Ul-q#YzSc?1&r-%uw_sR#eiP4c{T<o+ACFB59`A|QA#Srn_4qVJ#E8)?Yju^Z
z#)eSz?r)bGf}EkB9-gi^R+;djAo4Bg1PsD8NpfZ`h81XmT|l6$Y4g`Pr)51;u`!M8
z2)O@z8s;U5;`<|qC1c;wk&w^nQmO|-`z5bRM^kb!*3Nmc$X4kld_7;mC3s|;<;S{3
zHFW*c9{<j3#FEGjP%g-8ieDqD&s`^@5BT!+>isLje1%9q_2>NGHN}or^lMa@%Jt(k
z^Wu4T$Vv5mlhQ&;v(odkBEn0&;O|j+)wLDa35A?ZY4}l61xuqVOeTE}3Dtv!oN=Nz
zJ#V_FC#6}0c{LWM;USNA$Z?_$ym2nd@5WP36o!!5Kg<0duh~QW==p^BCFu#dvKJ=O
z#^gKqAk~o~HAKy@_p$a?y;b45?Rdls6X&HKf#dT7yv;g%S}&<Lc{!hsaTJveiq%1K
zw!OS*tY`Kgf_sPy>GoR|az?jCc^|9CCOFm9f}1TITBiRZ+2O<5Fe`X3m-uIW@qQzO
zjpu)ZV39Q(r4*%HaAzT6^6`8KiRyRBRs?=K08^HT%RuQMC+F#9s~!YbRTNkB_bbd>
zJMpLzuSnEOu`KPQSMWrG5sOJ4i?q$722t2sL5)2ulE&-ToT<ccWM2W_f$Wdl?`R+0
zo}W`h11d)0bX9tr-XSi2^PH8i&eqvk4J!p+GP_k-rFQuN$-=j%_aofu1i|Ql(4T^I
zfH^0hD(cy$p6PU4;Lph+*{4rYUhKz`k=E4mgF!V{+*Cn6RYdM!|8J>O1!q6>zBu7r
zv1RUaHXd=pTcQ6Vo<UGOJ2agM7~0e9)Nl;M&h8a|u!X@};S}8`Sg_RTC0bJo-LbH+
zz)C|Vt1G2LI3i6l{p2&u2jfv?#k8jfr+#())Phbkwa~&S#So@cIYoy0IiT(ZD}UnN
zxKtQ{r59{&7-eS+KRuwRFAxH{UT)gog*Sfs2Js0Ac4g(@C=$+PtE^m_R9Ki081dyx
zTJ(d-zR3ttY6#-WD^Z`9#3vVW{jZSrSO@GSApDF1Z+3XfOi%yG+l`K-)B^cfD6{ai
z%_L|PrLo&523=vQDaEM4KolJrZJLu0?xfJf!|2T@Cm;SfIsTfNkuf?-%>=%+X7uk{
zad#`X*7Rhv2bh@i!8RA-Ex@=8LJj-bBwaLNra?gj&DRqxw$S#WfdR@#djGwaW@Y^f
z87a7Sc4$J{%-A+CqP2|%QnvTg<6Y~qp&{crs}{ZD(CqVzapeI9e+imY1yBI_x48Mx
zgsZcx>b)}<+e*6N+$xkqZ2<a=hDJPVJa2PhV7Vx@3K)jid@wOKR#a2Au`iK)F`l6L
z#0(dg#1O-d`xQ8zX`8b)h~{+Zu_cK}d%fiB9|NaE$@_xI;w;ftRf6!0-Grurn(fEN
z15#O7@qngRj7FQ}<w0699bs{5u_E|+I}Uro`P5Hx8zI2FVPIeY220RUz}Z@&83C>>
zo<+QIGI>Ro3W}<foVNWLyXpLobD!|rLUV4*FO=cZl=Ie`kD?HiX<!P3P&{!(ML?|r
zNrx?O`Y{f$seU)g`*mX?;C}I&Y)zlgA3T_iOsi}8^GjkK1$Ahm?yjv%9$<o9E;Pt8
zccKcfgh9ycC^$1VDW)b_Tb(Jv6m@dy*pn>OQ>MLCEB|wG51<fHaKzs{**~15AdcWO
z`TYl=Q<%w4O-=RpXTa)%Zp7fNg}#yJXi$EHHzX8-Nt7sqN{Z5QU(!USgF3dKSf&DG
z=&7@0#rWl?Hh_TOj(-bEgASxJ4-Ji=fLrAD@_c4?`t;Nw7=1#TkQ)0X`sxo@E%t}|
zZS%W_P;~mYj82#=m0^Z>$nWZ?=~^1UWjin4-42ia%B1X%hleE^9D`qII#!5JDY&xp
z1i8!Sp67~geX6bzzdgBYsI7eoRnohpw)@juEphM8VM3AA(x%&BZ)$yD^p4h>4?Cvq
zVHQrnhJf4N$vW?L62C*Sd{IAqEZwTt?%Pj;{EO$jv^AOX^R9o_42^fb?eP2zJgk(;
zML9fsle}-7wCM-7c4v4LUgF@fLnNe?c<EYSbLOO5?JmG7gobdH=%g)s5XIjj9XLP9
zwGlX?5SQKCvzq>He;T-<c8*7hkBzxTN%Z}C=Y#DF1o0AebUhAWc~|F3+>_AxKELnM
zfE3S)Iht)}|7c_8(H9Un*sk;2tbN*5eh+Zc#Du!}bk+M`Eu^RD68c=HZLJZHmRWn9
zPXnH2UvpWufzd#zj(&txOKCg4R-=>Q_1tZJF5|-zi6<1lG^%ib(T$6@IURjSbAFL&
z3Z8H(=uBif3oEG^9aXMJD@||Tg`4ogC>$qF$Zb%!_9z+ViBMNw^cSKBs)1c9Sl6I4
z9c~@k3@E>E?CC^9;?hx+lmq|6UmqQ@yoek(qKy-;feTe=pNGx<1B*MGU7kzGyFh_N
zQ@6Zw%Nu}yZAFd6rN347tzS~#J?QK-HL^6UHJ%O-M)RZ99qRv+qSy@XTbw#fQ5Mrx
z6+X=xb*=>}Oo>8nX%LYc8HfEZlRo4#*~%M30fzx=M3j_sQRarG*kwzqcH?#RC3JEu
z5BB?J)j)`9Hr{{{+MX<2bNuwzkhTl{E&opt{TkRxctEVTeIznGG6FyVM9T3SU*z<M
zrR;Rf0SRKj<#!m$?X2FX^^*saU6>&Cay=xQXz8=JxV)SQu9Z5$`T&JzbGtk^vw6zU
z^u+$lDfe>aKDM|{hak+>30%OLVw@huGBMeBXgu(cFg7mfOMF5=%ANJ?qP|&&gghU2
zJ-Sb~?B^&DqX=ePcBC4lu4-OM&ouN1SuK1rlWr6i0?LVe5~s(-%3JMAw+AwblfzR}
zv`?&Neleo?1&du-Xz~^q^sidZEK&%4s7Os!#2Sny4vBtnK0HuwL&65Bi<RZY6Aez6
zr-5n%`B8`YF%8|F(=I!AL1xbF%28w5ZTt!<RR4L9^NJMI)U*vWbVmo9w#j`x<#|Yg
z5VOY`n3$<Rbe79?ctF5Q)Ns$pNny;_r{TBi0maMeJwrDa4|G~<Dc&mV1);0C@f@%x
zINgW8do{Q5@lEu^aaPa%Q7j?MQR%8*U-k<MiheoQaka@jX6AUX5EoD~_(NzChz<6a
zOWMePtD|<mI#PRBd!Wni=yexMXB@uJF5DjTDQ(s3oWM;GsFQx<$_po5k?xcG%Lrw%
z$HH)w<<@0JZt7^IVL!xSjF4A)O#aRb_y=AvRJSqW1YPsj!u8$PE>_cBuIm-fu5UnY
z=TvJxs3e<x#o@ezh6+AE(PQds2;;b3$$I<djVoFmnhR=EV!q+x-2`{)@^ZU+T0byR
z^s8jMTB*2TF`KO39v&YZ`2F2JF2+yOFWXf_C>{`sip?p>bVg_mR`=p){f@5|=NCil
zQ%R!1V>3Wnir;>Z1^Hx)eZw<Js0OWRJWvfN;&NZBf#s<fQorSOB~G*3vWooyvvbm6
zbM|6v>!d|rMFpm-;LRcYdDqm;tj6Ygp!Ta4(%~HEwtST?a7>l@)f_+^$mV@E(75>&
zvmS7M&`;F>kzS;7v)BDlwL*=B_;`cFE+t)N`#)Zi_U<tYOqhHtIG$~|;10l`2L6Kv
z-{%|2_$n!T|EIa#CxR%J?@gYiDu67Pm|ZvFkcJ$g*f2AG?9<#XCrw948QN$)-3BO6
zS{j_u;i8UbEGT^49UHnzxhm8IWh8eZ%pvEKVJZJx-Cqq?r8k~b6Cux}pel!AoeGnN
zI1nB62ZqMbjc0Sfs?*<covVMQtKaynmv*}5EUDfK{YBY>2iefJxODW`9|_Xt=Tv1#
z9ye8_Im7fomMJQaftgQW975+r!F@1u5T293?{OOK4-kj}rv)LViHV=rvxXT;unPhq
zBm!2mHL%UoSki)T5-#o`{QQedrBnsmzX#~}yoUNFw*p%s`0J_70mJ~lBF2uZuf;(B
z3R~T1ZEE_%MH1qQ5a==p{u4f0pd-X=^F71i*QO@2II$7n>i<4YHmXWI>9CD~_={UF
zoEY<xNXEUKTIS$=1a`s}mMJKRgv3p6IjTooFJFX1Y9+L(7B1D6bk_S=IJgJMn;}bP
zNYA<PZ^}~bgpTyH_Fo@VcS~8GA5Ht=2FsSDSm60dXAkjO%qS|o9-SKeh+`rbVoy(R
z@PcvGh+f1aX{2EcaxU^%vczfltWG|;qT%M9?Uid9p=1m7xrSf9wvYK^RrUJy26y9i
zT#Rnhob}>eqZIQ+FrV|UvUBPnwt3^Z4KGu%yGW}R9f9f@o`xB&9kAOfMXxp0<TN$n
z$;no*|3%7L!&rgW(uPvV=_Ue%4H9-;$*rNB7XWDr98<iM?f$dp$W67V?0Cu3{_EFW
z4h{rB^6xpkI6)p!KpR@9yiO<I^M({Q*r>kpc{mLoI04~q%H<D(&$Fj5Y#`PC%HhPF
zDeyYSZdMU}&*MwcV0v+R@#~T|#AmK&Q<<T5I>LV-I<ay~cn7qg4t9HoN9v8vtMP{(
zU379}WaO!=^~By!(U1cP7o$Fx52TN7(<1#NBVH9aAK`w7LHVq!maSd6K`?wY!VbI5
zO#I2TFJxt3J?I_QiIfr(>lqopUuPIP8b^Igu?{D+UANAxXJKXFQC3i&^XVGw7Yrqe
z($b~H#V@ju!Qa0>5z?cUy$0rT@VYAopsxmGP&49E(LLD^gjo@pAoVjz&C<HM5?H#M
zwMUhQa&8O#Kg4bxh<SYJpyR1U6S;<9D_n94CkT5+a}xr<=QL1NR8y`+X!2TA707QO
z<0vb8yeQ!BN(juruHMh0;}eKzM8S_mYXi<F04328zL3z!1v@vE@D9|Leyh~;V1PN!
z(-!<b+>b7}aywht<z+i`+2>ha3l<o&n1O|EX2*>@sqw(T`UVzM&r-vBRQuS&NbaR~
zym2*4bxb<dTUmKg@;zR6srRnP`;+#XV=kbd)lX@^Qi&WVBZILmSnQlR*MOD+7K?Aq
zHwPeRRTSz(2PtC>R(;CA|MeEiUc1BXbn5j9);*Dmt$0rQv%jm&AUgr%k>6zwOU>d&
z^aD-WrOrp=$*`-gt>u+OX9)F0TE?&9nBg9x+hoHin)LF;gNX9S<DI#EPJ&4xA*u>$
zvwob#SHejk;Xgg~0eMG60W?ka^M7C1!c2#O)$&jhNLpHK+crw7NPquty6z{Rcdsb~
z@zgYDQ;xQohu*!pXf$XWHZo@|C}Lms@I<%CntZL<+C~pjJ)h&X0p`bd(I3Oce}NBC
z!I(&koMGT~Z&(K%MBbljeC6O@$Txco*d3QpZ$O^ODaQJ$3Hi!;rbu7Shw~ZO#$4;B
z#)@<%MrO~^NMC212C10wX@z`5<<{$$(&%zHVCbITLqww+WhFJGsR>ldv6If672Me)
z{R#Y<S65AVd2ImMyn3n#bczQa!F7)aDDaLu{2%#+B%Pf^0>Ta~LLh|J=I>9%DMu2K
zsuj*IeQ-*azVk_^&pjM5agnf~@j3~$@ggC<lP%JRvq~I$$dw@4)Xl{PDFu&~**(Hf
zH-4nID?bRTD6hcT*>7;$zP~pw`j>^aP44vUbP+CAr(0(AWWj!NfVz>F-t}el$L__5
z#+}iK%e9`2)qQoDto+jEm0M!AW!w%Me4m|p->j&|(dRq@N5#-|b<NjLHkAqM91qD_
zzqyChU7$ke#Ew4(sC}F7b4z*N3Sa)g5z{#<sZFKO@XsyLJZRA>G?$KX{-1sxb?06(
zu`;qIG&-Vt%Q@JtaM@aAL`9vlHx@<@<wZv~_MrQV|1GOMLjPzPR@?wJVDWoznccNZ
zPa>jKXeVRi<YM^^VI$?Eu#j6fvI1fF`(N4@x&C2YjVu4%xVWA*6Ta&qNfQ+rrS_JG
z47MYD^893)%s+idxai7IzGSkFkh`rI2M0%k(7Gzy>40pQAC32JMCFdi54BeQ@-izO
z{l@3lc>>O#D1SHh0Y>T($14`qnlWU|Afja;(2W;Bz4MRh2?~Bhv?P5R?J^-g!Ps`<
zfN?&2j)Rf=9s>9v^3R=ZVQ$*vo9?N3J32S=YHoGQdT%GmsWiLe2nJQ$-6VjMugfTI
zTU+1TG#@b3_;VA<5a~PEKZ+DGJI2`zX(@a>PR8J*34ya<?n5Q${86brQ@BsXpjaP;
z1?|koQmYcx?F!>BOio|(PZSg-fZMJ9T18F{dx@J)jeV(RibrMP)tKdUgiuMAP+yqO
zCqy(^Cv`7E1JzT3p~mXeSv)yO=%aUUr2z}L%>$w1==kutM@j6jhI%oBY^e+-uG~`l
zHki2p-JB~XOk?OlZ;~)3!lkPZjwKp5C1DQnb;|YhiVVip`d>a_sjTgK3m9a35K2h2
zdAiR*oQ#^o`AqYEOkpZtPY<(^nc3d4={*Mf$>Hwostgrv+4A~&e2AaNyL@-al#wIZ
zG%qx*J;6-uA@f$TIPFVXQDytk?-N7J7lpkc;KRi*%6RD%T@?C|Uk22N@b)@AH$54A
z=%X^bv=EmNmlCTE#N1MEp$WOBc|V*Dzh_6#`cd&a5B}~~a332?GK8T`$IYIBRB7jg
zpySFxD_QA2xHY27twbBG;y<DyFkrF8KaiL0&dDph74S(_b&0PUP(a|va+j+*P&%j?
zno0*&{Ny(RPf`01FUB#*(a>DMZ1fXT>&Z1(!!fe3ou665!up@bRpqIBb>;Dfh{<~s
z!avKbE~4PW#pnF@+lD7F3?*^ywTvoAl5ry6U1r}(p0fd{FMbr5w??)Q^I`j`%(1A4
zyh99jHXb80;~x4)*jiCNXWsr8$`(gMln$QQ>4%ClO?&M8&b5gV9?MIJkq6{}#_K=C
z`=+?~!K#L)rijr=)zC8Ask6($#=vyv4>@?hvnuSzW}3I<o99q3>bixCJSUT+$1Hl2
zcy9T%M|>8~Kt49lBXH<SL*oH9!_o0cs|)=;t$J<3(g=bVBj9{OfUmY3Q#4~_Typ6u
zNW!9liHwYJLthKhnwa&<Ev~ykq8`a$3tu+68`nOz8qv<o$OPvf63g|L#jn}zaOqya
zBv6k*Qu7)J+Tc_QlmzE_9)1a$a5xpTO04W(zUKGttvdfR0NfNep~PE?-v5}>YLc3!
zogX2PZ)<Qwt!Nfi^w+#$X|^mD&W;4*F@axb@_6NCs9k<^o#W|5Piy$g%&Ge8ocZD!
z=l#FA&FrPRsqODte0wHiSmU!Za|foT2ANsZqhDeDTg;)Pq~&^@7sPmePqW#z%XQ@b
z*$v|BC0R_D9c_8vbraGiXU3+(nHu~#UcbHb7rhiDw5i`RhD#+K9{<@nRqwj=3)!B&
z!uT@1xwf4cqkAF<e1HQiT>I~(=!iLM&-XBDxIn5@nfs6`TMWryK|VAPde(UXB}!7q
zO|l6NDFZWA2JOJa!#O;3bagN4{}TxMMaeq17*L`?EZD?4gHW(E4(@8jc~S6cU+FJM
zK*Kr6?$f5rCP-CKQBrzuU;z_EC7SS+Rg<w1Uh3J*>2v(;iJA|<UxlWp{#h*`AlSJ_
zKMl<YSNk4ycQeD;MMfU8vb7eetNt%6Nt7iV%+WWe_COgP%1HFmNJ_!a{Z+eM;$)-E
zegOTxZe!2t=m<Nj&^{FSlVsOCJn{Ej@W8?8?e0gg>}`Fj?Z(WM7xt6`oMnDig$KqI
zpm=5%V?yZGQyX9&Q<5iF88oj!k7r_10_8>YH}$bj7*cDzZ@Az%jZAIO@k~BG6RCLG
zep<_MmCaD}E&64q`X`>7|NHQzXwcPp2u63i0JMQ|1Wf*Ti>PV$(Xop8+-#EzLsZ`w
zHHvyrF-4tmEcwfre-fv0cB(Qi;a(#rqiPd0r_%WI#?bVw#l+MgVRBf|Fuz?HcaMxP
zPE}72I}DUi5GcrK(dyMKXQC=g3r^0R-F95!*a_0ahi?eZO!sK$s%2*tvI*Zu7iy>o
zA;VkVp4U)OnwW{C#sH)j&%T1*U@<gVaDUG#I9Oi%hZb;<CWjIWQgt+5-P>P?lTrL4
z^T6S7|9u`qrPULl3({s%ehN=*C{6d4q6?Cj3Ag(sL*wLB)kdF1-KtBPw8a+L=%T4I
zHvGQcbwq&W2^WoP*Dv&G)JqWj)cb`_)&_@3wQPBz5MfnU6bgf+Ah%jW<Hd_5EIQi!
z#Qn&E;5&)W&#4c&<7AEwj}(**qN1Zj+@kwC4L5H4*>8&2Y7sy3QBYGp+_ip4@EHgZ
z`bVgr;{1yXf<XY=)x#_r7q#%t^qqmZaqs*trErM(tm!*iG?IrH!n!+jJUYa~ppnST
ztGQ`>lbP#oZ-_mo{Jess!nTUWuadn0$U$3igoH%ais$ER(40+3H_g}APeE<xD%^NE
zg@Pm$%x%sEw9hLKCM_?L*uRei<m&}Zxazm*es@|g(f@FE-g3r4gS+eBxha+`?C(ZG
zo~shDH$JJZ0Ss@AEvK7_TieSL*#kOtt^j0R=1iz-fR(|5#SkTbA~D5qlw9_-$4Ttd
zY&1N59YHoZJZQC0y*OysB@7Emuzdb*&v)wQEi5c$X9&fxf<>elR@`75V{~eIL1t><
zgLRej(X(@B&Xt3yI@GxMAm5t!^2}4%iM$j#^td+73J4>H^qu+#rXjvS@Us&WxjHe}
zHya;)_)YXvRCgO-4sXBEYtiZPe$hS!U3vT4O((q<IdyUn8B0KO{wasSL&TqkrRetm
zuIskYMXCk6vqGNuJ}<M!S3lhNWqQ1KWmoPI3x4;QoqPB2dd1gO3N{27fq-#tK*5bu
zn2p?<sdVf3yfRp_M>S0xBwy<IgM}-uPNvY3r0sZFa>D5I4JwC+()-)lQ~sb<``?E<
zi-_IUCf=!PG0xq@`DHS5?R%Rb9Fz~CtOtSIFVUg<fGM`iebspE@c`xew)W4vT!K?y
zr?nF+Hh-U|lYiX34I*pK7i2j9_c#Rvgp3bx#Yic-H<$_y<e`)z(z1Og0|h7KU|l7x
zR>024dY!hRa-BTxaq<ZWv|ds~?`V(q!_^cXn_L`tL}TL3d&CX~<WWuE_`iz=CC4ZK
zeC~Z9%+Wv&-N)vlo%_uey$=sF)j0^jpP(^iMY0lgen`buPJv?nn*5p9_iCT!l%fBr
z=l<uFx+`6cq0$qP;t)spU(H=(zoiEDSgZaQ{zTEUz?325gX@WAPa5ykjyD%8s&C-o
z>ayeL%D42=+v|THyG{-NkE*|bs%rb*$Kj1ADvE)Gf`D{`w6r21EseC&-JJ#@B2v=b
z-5}D6bc1wvcQ^lu_xt;d_hmTlP%n7)*=Oyw=6vQ8aoC?ENFpNez_6Co`KoU-D+^mH
zHI>VVN+?65^WG`b?(2~HQ?sg<=^2G!>s_;m1tm}4P9xB9+`5&Ck`fb5;`0uEHvog6
zE&TCW%=IZ+@PZot$oaRr`Z2uMuW=X&BJ%Pdmt3?jui)~_C3HhSsk^6#r&xa^fp?_*
z_y~EP*R}4wZ&LJ><LpO?$#9|e;b3!NWz9qcfJyeZc25%JR3cRYU_D9vS{p~!HIZij
zX0vhR;z5Vt+EDe$F91W`Y-lfU$~m@JOa8NG|GUYZ+o-z+PF{93)))6<H#tT=#OB()
zkK>3DZf#}!>UjLMaprQE@Qf1+JFKbK6xIOS5MJl>0|@MHA`1#ji}H-1_HHe6J$R?9
zz{ZtYnTUzNWPviv&e~E|MiCGz8E?Pd?tW$#CNT-AmA~BIU!>LZ_(5NKeXVw83s6J5
zAN^_!0L#ingiA@k&&w-)C26V-r704y&y$p&E+~bqgGtSrjDM`AIRSlZw?6bjej!4j
zl!kADfkr_mCU@~9%rDT;4!;PJ$lLa}u0kbREKY^_2yfB$?yfnY4x`mdniKDT{53#_
zowI~(C}j@Yew}xA>u@^@p086@hVH)R-*t71dr|gxwPO=`)J%SO<&@09`8TEpLyU?W
z*J=UhO)4&-X%QRybOpAz(aMk4nO?)h9j3?xIMMq&Md?3ny8xR!bFEzGa{&4z4BCoI
zijnx@?Ck8yMAtPqs$j$j`Ko+O<U2y_!;7m1owzp&HFN+wm$-v*M_>+le<cL3KPl?V
z7c0mvX&*W8sQvaph?9F5rfA}1K_JD_;bQy;Zs`J2lg@o!Jq~C?ayOO>&)6+rarkei
z+D{W>Q^R+9PEZ=9rp^=o{t%%&l<@lnlrZ-;JrMT|ICkT8pOf7GBT#N^cxhDw8>C<=
zg|V4lVq*69@Ms(;NdgnxAd{e;mgDyB9$lNnFr&QqEfzL~{=fP+3GA+-qR!}h{h8Z#
zaG?p=hRxuAPYTO+RSX@N5JeNL4Vhr@th@G&eL8(jFoe&RU*lUe(<5YkurZpQ-ZZ&$
z1kQUJu(xn|iIBle)O<O)puL@*^_nNB&NK^jsxy3K^11?w8V|rj&^C4dGvB)hBVU>x
z|35B3qyKVSAdK7oQuu3DSQQA8ykl0dGaZZ1OY^m(*LZ4NJ9s}uS62{oYce(m2Eqmv
z6sj+UKszWZ1M#}Am7;`Kp5H+8ZcHN8myGJ-lW1=0dOP2H6DaS$QMwF^2~_)|A|Ez0
zioH~q45^ou*NMOKXl+*w$m^7TMCz(a8!KxCmSgBwJ|M`KFJC`Zl$LXH<KTom8?xv8
z=^O=Oz?v_H!Hsz;<z0hK@#W=u4c`}G$L_YfMx#5FSayGkccCpjvFR;T&0uGOfQUNw
zS@B5M^Su1@<X0A5;LHu$CwZE^z5T~*Ci&Te(0^SQ+86S|R!|6qM{^^VxT$B?t-tmU
zdyZ;%&#0`@M845d?R^>4?tM@ueK_YcZIADW<5f@=y@^m^wM@CXgCVKBqM~wS6yZ>5
z(w}iKqJG<7>;ofa7e4p1WnJR2^2D{5*Z0i%Yd~jp4G}`q?)cjLiS$H8HR1xIBumqx
zYVroUUYw!4wc1^T#*aCoB`cKSF(KQprswYKZPW<REbPsO0Kd#DuB;vJu!DGkLP}O%
z^_ZpLBz*}VLY}7|#ZnB63NPdF=ro=|yY{@lwu~hWY%y%~mb3tOX#>g6cFQu+82bPA
zrMT}KYp)*QhYjYv?#!QjGV6(g07tgJ*GC}10`@+r3Sc&V7yt24n$luHus-;BLc^F%
z*Ioo%Cwp62S=mafGxo4Xe5|2?mL8Xb>OZffqTh8by}G*^&PnbxI;%_oaZukC-@F-(
zuoEN0{NZ){k(_p7ZWh#F4D9_dD+bb&UTOn$^f0wxfB-fG&yQVCM_z2$8hPe+Tp66C
z4)T?kuU^}xqVvUF4CJ06UEFlDpE7|!Sxuf5aPl?BFC?v^wIZ5_QwY>}CB8Yf3JPpA
z`Mj*xmcJ{k9Jzo_b^0jyI_#(5?F6b|h$vuGJ*#*_S8aRnyPCj#|276dm0v37#fUFG
z8mj?En6jdL<<Dokp9s`AdU~+*0YWK0pB@O-z7r6#rMm;c(~G|p_1C9x%Cv8*dp7ny
ze&Ur@YyDp|7cL4+mBeqs6wLcz)pB`N$fhNS`!4cUcGrM_UQ(I!>Xmmu>%?cjusv|Q
zjeuVpojU6QAg*gGP7}fzSO^_`n*r-^Dm^zjW-}y}jjBDibcjAJyRmz~>9h^%(7VVJ
zcD_K!K;V2s7x?tyZW3iD4K1yyT7}E-s%MqJOTfsG=Om##-~1}9H>S?;alzH*DiXz!
zjfO;iter`-#DzxE$f(-&U@D~O6~WpX@D(rjPPNysDIrJPX4gg;8%U`BTCu_&2cyNh
z*OBmuhxqu9+9KW4SXR(4Hzi>zIHN-Gm$fOB{<n%=Zz{0kQOMOdSKm604i1}$koe`o
z$4^Ynmi?-<2v>#vN`tT$Nc4e!1qm@IlgchV`M(39h;NjNt#Vt*O{)+Gf-j7A%C8@z
z{wl#QMfp|VOo)^Zs&vKyjnx8yO`SFJBc0l!wOXZBRj~R4-+FkknwUvK#8#G90%tzO
z<k*6_9CH`)+)8Zt{paxsJFz+~rnt_qdJf=o%1~d7(w+<L8}r58z4RkO-aB>IH;9qp
z++CfxPC$y~lbzExl~@T3TSEa^^yBD=ketNqw)nsJc=+|0HxW?$O`#_Fz-tKU&INop
z>xzRB{csBjJHwW;i<Y*WHC#O4>_7#`NT-^pbk&rq0!@t#7d>A=$#UIyR&-A-6O%*W
z|8>S}Gf;+g&a4e?JGTYvL(&2WL<qk11P>lQ3T{+YjReb5u+3}sKgIN|N2E1mGMe79
zva`b|$w%<xpQ-NL?QL|M=aQxba>^0!DUs*a;;kVorL}(q!s#e{U^H_R+7sKi<d|Rx
zKH1cK*ND?~eLGar*lwM(Y0B!4v*&2pL4L;4f>_wM4<EKVsc13|j|Kj(R)va%@+k8S
zC7XgG?0t)I5gC^HQVD+z0m}lPvhVel0nKDcKyg{$XCacIhC%(Cev*>a8tc$@wf?1N
ztl5N{3zDy9<cSICyx}Mi{baB@O9epGAvk^DUi#yX2(*V+%uRFNNRYW{Ym)iwnk$rI
zwDHfn%B-Q21dGaMeD7+2+W_Ng+c)1O`mQy%$D_Q;r4>O-=LHO&=9)4#L?eldcV0yb
z1g+K@VgSxTdGaiMrRqA9Wc#+nT{02?lufIiE4cH_|FrwY3|yA<?qINEo~UmyY7CvY
zcvF-4=S>=zadv1{!gWbJ<lY|LLe(J5jeC}!b0-4?=~@MEhJTo)n6A7e3Hef|#o8n5
zBltu8mFmiFw<`ehLwVyoZg`ER`ltfrF7Y*};HO;G(N4RFv-pZ4_|vEISg>xURYXBU
zYj`cFEmyQD7yJIwrQ*LYO<hf3Oxn*^l55&P`Q&;`S6xuh+}Mh03N{{~f`OsI<2#fg
zVtaeGDDm<6HxQWI!CM2K<nmwIuTn|Glz0djF56gIeSR%gX0?Wcq!)&RI{f5O%;bJP
zTW_;YkVL-XooAzIN!sXN#N?IfFnvDX#!-n}O`NW|i8_M}yqmhf`PreU{zg7KHPvo)
zR#{m;CFbe1jEn*m{`5l7-g~!xU{CUyP5OJw_lxfraX#hxz9@U)JNP;}>2MY^nRYO4
z)6z50hlr&_M&gP5R1y!yKt$4%NkW<m6{%@yQIQ16&yk%A0DJ6gmqdY}X>vX1gSso|
z)@W-IL6cdiQGMn8hqDt2JpN^67tCecgdB88HGl3{$i>^i{PBuQD=p1{>F5=N_1ZPG
zvnr5kl3YVF<hQq!0e1O<H|XDqJ3Ei-5|Ako_*+^|8&c11=2P;*=owFtBq1?VSV?tf
z#kcm?|9u@2%{UvQ!6X%?kSF6WA!KG47-J82Sf%8Gd9}rCz>}bWG!Z1gd#VRr8oS_E
zU=b){v$Bp_yw$Mg42R$JU?9|0*YAFQ<~;q&%eEB-grpX4FZfVT4j_2u+WT?dmuUMN
zU{FAj<qF&xRrcI932tzk6?KjUVP7KJe&dvdkx?dmQ&<T)@>4AqdVTQnkE9XT%tI|h
z0oRC}gpPjc#wkm>RCLvs>LK+W?nZNPEcS=a(^1HU)U&d71-oy?I5WXx1_e9<VG#B<
z7k6N~d?)MkjScX42elXJiRXm-uAtlVE~sMx)`g@)6a*NwJCt`@Yyu(N3R02h+ap!p
z$VUVDeOM$#w#%T)kQF^;kHtsO#KZ&!=6D=N4?P7EWU0Hgg~`ZL;#cM8=CI*4Z*#?n
zrNCoyWmwHwpxkz>+L0Gef=z+KM=(1xD6LDB43jaM3*244@j>!JOOFTmdR(>LBgOOq
z*<F$sNF=YI_@7kz&0AN1mzlh>1WvFhOzk6>GN65wmw8d4Q_W8FVCPwAz1BB;-c^|W
z5v?vSZ@8}hl2TV+W7MF62ZtXoLE1xivf+OpD%r$qmN#UWpaW4<7yvzBl)ZZNV<?Ok
zP=7^EoI^Sk#_i1|Y`jx;O8Y<aYs@8qfLns(8c1?5s|EKlLHvoz_J|sXdho`^X^RE0
z1x$UcCO2;)Pg%R}ZVaJ9SLFsy((yHnm#Nswh9Mz+HrG`2Ra&PkM!1`Rht1B)ni#Y|
zufn7|9q#7p48s*wi1y@k6Sf=dsJgC1r);skNqEQG8g%Ed*R*740citF?b8(?*XYXo
z)7nJx)&VdyofY;-PtVRa77wTQhW01#)b=+3`DHxg@Q2HZIs%0kbJho}WEjRZ{XU<|
z8faC_iDPx^4Z2qJlV^NtCJnT#+}YFPpTgN_alpxa9v3I)%(ckG*-Qf^sM4?#3Iphp
zASx)YDSm*1^}Nhh=|n1~#>RHJ&hpi_xKLBjm_|M372aO}0{4|)Jqk5*^Vr$S7Yn^U
zGaZBzlpz9IR?Ekds@+U*Mi&)7t#FvPaCEG1pyvkDLz#!e0Ez+I7x-~s|4d!I@_WfT
zY*xGn%qaS5&X|DR<5-j`Mjmvj#J;e09`PVlDRSusW0W0q^aH|Yz^=Ob_zvTJaTt82
z$WHD!)YqpVUA?_u6RW;_5qDIpmZxcNWyJEWzw+VU-f_#G>b`omKhdzVx_S{%E-th;
za{nZ%3?uJrMsCrK-T3^3{07x$n<0Y1F;9B%F|wznNlgDt^6}wkA#FsQajdMh|DdtE
z$^|Wv${}xKqpZ{Qj`e(7M1=LdeckD`le(#wXtTIj5FZiLICDS6Rg?<zXrhcG-F+jU
zQ7=(JUQslbLSRvol9V<rsYJ+n;0m$FJ%Jl%mLlux>%_#w7|lR9X*;&Z6}Wf@c}D7G
zX4WtJCxmLchOvE_iwa*ymvxO|zgXlKc=3Yo*M=<bBlz?<<5cVDIwrg&C?p@x9T|to
zob;?!9F(peKJ49(<a$>p-8tdJefcWHyAj|TD#Fe@UTs+NDQ3-{(Qvs;>PCXsQ*&)*
z=LVhDA9`WAYUb<ZPdwb1&^>_(-!oeB{Kn-Kx<CHf9gpUCe0r_GnZpS0+de@`N{Sx+
z8x?iAzlS~W|3S1G9^)(C^W^e{5<1z&oLCn%P0!|I+SNU4TEXZbssjB+!UTx6^>yZo
zOHKa5+xYKa#(Dhe6Z{NY$;ZU$iK=A`xK59I#{);WUy~qGj*vCqT+HbB?D~ZA$QyCF
zB+WQNAJJ9s1pk`Igcx1bP83(2<diZsvtL!$U;oi`$)e(l;REf^^fOAzgZZ^z^Xp#x
zA7gS#zS;R2yOp{*HrwEtP9MsI<sFBxzkQHJp=S_3G4y9{bBhI|nJ3<y5d(t_ad>$A
zjLBzg)b`blvxFuxD@#$JJZV`gUH$K)dJ)MgTx4)ghZdvAQGYNWW54lWrAACKbZ%LP
zv0ldwC+QKlRk}18Oy?rWgcKBXd-Go$e0o`fm&C-v#+UCbuB+R3HYp9fm$y%s%cwmi
z*x8tqSr$#(C6csfE`pgN6<R9?Ooz8#hQa=`T-sxNngirax6|IEFD+uC=pn)cY$jb}
z&cADDgtf;Tbe>Pdx}Brk+~8v+b03g0!ww1Q)T}(;{+|6f%P+$cbwgE6_6h<jnAf6`
zjSn#d7F8kX(9Vpn!QgT|cu1>D;w3GuPC>8-oaA&kzCPar^3x`&Xvo=I2g9l3$b*_^
zi9a0+Rl5A$tWYg2y82!_twX)WaiJo@_~+=*cq?%5P8s6NS{RI&zkESoXOQB@ODeNn
z7|-E8z;iT4j~LBzUib^cY{xHx$GRAgSBE<5(%C!(L<fIX5g)}6tZ7so9d}Fz$`rsa
zj7sytU<>3SPktw(@lAon8WAc?T?;GvHaxrwq5sy(C%y0$R)abRK22#3Az@*shVNU1
z|9nTsD(;Th5lM75T?Kk}$(ZWE14oVOZ==?eRF)<a-FoBMhUnYdfOLm-JihimWQDy*
zuIF>nl@DN0Kuqe50p4RLcL?q}9j1q9wE7u`Qa8c*{8Q!+&&dJGdw67c+_#K6_c4Gd
z0fr5@Uj(b1Y#wpibPe6&gHLwF{EDXlr~9@vEJ0jXmzt4rp&fmzXPm;qIi{w=3JMAv
z!}ia3AiqXBqnD*|ALdil_MK*^$GgeWP2=JNg7tMM&l~Ub2UC7kR8pAQc2IITb2WE~
zRs~8p8+4OriuCnbTYtJ5!KD;ZZ%~lXuWcrI&36d6k>~Ofqo95S5g~+oE_?J8S}9<s
zn5pxCK>9an+UHcx;Ss<TVx-dUSAp-(Q}gaG$Er*woA0Zu%2A)II5iaZ;gc_Ag*rCY
zQM^$J=Nuja!#A@L+sV0=CZ`rACZj65IZTRRN1cf~AQ#3$7k<4P6y0Z@QmS`EvWBB#
zKH>hR+UWsPm#|_eNdB-uApq&_1Qi!uTeXSP$Hug+S^$EztNzqF_^}7E%z}(4zhLzC
z03uE3X;q(@r$%Q(O!okUHQYr&gu~ut4iSZXCY#TZ!E$R&3AHQ8p#ADo<6|-ex#$0o
zBWIo7%V8g%HQqll5q7`se)SV=t(ld)pg_=`WIWj0+q>#=UUi)qCSHvxRQHZs=jL$1
z7_5i!wiM~VJ;}X{U#oo*B&T~De|I#bn>wnzv*|{1Rern|$;-^%-l^GHJ@O4LAI_v{
z2|87dHqoMv_$u)&?t*~-g4?##^GAu{#Bh9{?FBb1^l|GdFpCuv6F)oMxM!^~E?9pr
zOvmyW5)YF}S*56-(o%{)QgVNM{c7`a8NeP@8JPrPPHb%KhOdvE`VRE`-Tverjj1g2
zdP}9Dz6YwBZ@oBfheAN`PGqS=TXID6syt$lkW%J2woO~jaeNz(&AhwCNBh+??&14F
zfR%AmFjP-WcK>OR2o2HQTkVe!(JZu(x<%iP@~2^+i0cR7SAEPI{-ui2piFCuN*^RV
z^NOi(F!ecjtgMWKRDSz91Z2{<&bYNIlOG3#UaLbkC+lbMIA2-k?dp}zi^|PR<Z(Qu
zs)@2M-y^`p+qc=X+Bj}~mzJ6{o{x<vW)HEBI1g3UXD!DP4CaLmRMuDVzJiQEso1BH
zB5KtJv5}Et7YW+eZ{xS-%`s~7TvU{-txiRT*M0L57ZH^_u&@nHR+5+BUFtJkZA!rZ
zvHF-@Z#(Dx2@$u&oRgH@x6TrLL2r@y2{0({*w?TFI~5ppl+PWkjys+%PKhh#ZuA&;
z704|>y`ZCGApP1EywvT?H1G*<C!8lgTgvId!owh9&L?KPE3n8fL5CR@qQqVu?FD?d
z_A#}8tD)wbG}URhc;vgskh(cThF!Jn3|R2&@?O-7-wDmDkj`me8-AZc*bDUZWS9^>
zbTt0cRIAqsPSbL79X(xdFJ1z!OXwzb!*3)mL{mnl%x=amL8N!QB{3%=Bl8P-pkIBE
zW&X-04Kvr1%@@3VC#)DN+~97z+FRb5(Yw(XMMWZD^N0WK_F*^s8of^fD{K{obp5cd
zSMFq3Slc-q5HJ1y^Au^C)|vR}T3Y1260Cn8j6g~0!EfU2)1PYNsiw}KPkUr#VP6YQ
zZ6_C|B;a)1@7VK&1xikyLJ~;D%oX~R{VA2hcy<Q3I$&dm@DSz|7faB2)z&7Wu_r(Z
zU1sKS>!n_QvMc|vbOYrHc#nKe0ITRvt!MMzD6q|{!E@+YUr%mJV(g(!*$LN(`rKfU
zMLVe<`4tpsSzJoMVwn1Tv%vOTsk6DcIW$bES8hauw|unf06_>nJoXm*m!#eXBx9#G
zc#$@~Zf$GHjjG8{4`-YNyL`8PVje>9hVJvbi|J&y;<DC|^S##wtZNS=2BL+Ij;`(X
zdWY-VFRzk=u_|qi#gKhdyc@<N56^qY=B>@N(g9&-lAbIpbQN~(8=)X&C@z+_E`4};
zOfQ<t$jiIT^s6_AF*7q`lx4~=;X%s5Q4zlsQs&6Lxn4EtrN(9=?cVIfdO+j>5L9|*
zdfQfH;+R0NR2qcIB8TgP8&1sN+wMtn9|2!RR3t>~=(ir*6A{#az=*z0%pz+mpHC`$
zHHC$;%<`|uaJ>24j)$Ug>FNF0DH=HPVo9sNhXw_yRcjx-*l^SUOMgFqcie}&n{F2A
zU`45*kRVG{QJRB$!!sxrGKj&q1^bI2d0glu_6=b9*laKYH5P&g-23RoTKW1Esp6%Z
zkoeqOF^b@WwLuXcInkBr2ji|;0g6ual71KmL|o&tN|X^XTu{f+OJXG6t=F>D!aONL
z+Fy6|_f$Gx#FfU}km-9VF8)$OLuy+BOvpaHxWFFszjuuL;D#rJC!`7=*JQX{cUCo|
z&R7N)Fe=8pO^x#MGS#n_rNVAl;$MSddRF|PA{nO0=;+G73w*h!k%7&IlM3r&Cj;vK
zUccX(DAxoDQnai8B#DS|+G@LR%huP>D|SddOKjE@@DM@br88=WRi?Ff82z<e6e%JE
zD6}3SPjGx3G;ecGSvdxFGr_YRmMk(HxVir4kh)0Kz;JlD7@xx_9F(@XuHHWPS2cD*
zbrxnt7>Lb?aeVO~S{Z#uJFrF-JFcYC66q)&y0vrb{82i|(yzSbzA9@+q58xl79U1h
z+N^T>y(bO%pX;VInJ43vTGM3RC!M|FjF_GGk9?Xl752pA9F5cQnjhn=DioB?HY>dW
zy4&|JYoso7Hr?rmy5uQdvnKL9iP@E>N4^7f?$N8!akPROQi{UD<&~8rQ)Aa*5>4{_
zWm@-g%@@Xd$ooQBZF~BhjHxSdy02x(8~NwPAo0X;Rn$7$a18Q}M%P3IcXedaTXqSE
z%PEUe)AI(yV5`6{{HF&jATR^%#N9l)!XozoQr(gGze`Yb5C@9D%i@==$8tIQ4Bq2n
zWaO`j?@FyA4~%=E|C5UwQ>c$+ScD8mj%rp<IZKud2Nm(`iQkfhAA%$KPQRkyO1MO-
zZ}3arZVq(Mo+{r)xT(;-NuAms9xf58bKj@WSr8}?8U00r@?1i#vgqZj>2R?W!qK9T
z@b+E$f~}j8ct$Fbk%F|lHnz~Ot>``EcRFl(_bD><;=X|U9lkhCWreeomN=ew#X57$
zmm4i{?=F>&)j=-dOP&+9r|vA-uXZ95r-KfBL}72meNB|mo&KBnT@0_9y_q_P^Y224
zCo1XcA)98~Cos*`<ikMzG)iR5?K}t`goFu#>=AJ2#CrZ19wX4d@LL|dcfLFCAUTw;
zr>v_x=c^PrOkbRig^rlKDQfE-3^?pE(=(D7by}v|@LvfD5OH<3&C1NZ<G7nlY@hIV
zQM_uQj=XYZq+$ei)(pcX!DFSDXq~Y&5@J$A)rOLjXWp>jD{L(_%gktz8&^M`S2_RW
z;`HFf&~I+Ky*jY6Z0kN9<m&y6aW9sXl2%nwgD>7Z&%=J>)63fBJXt$CGbr`vZD|KH
z3}N@xXFQyppL3@z@+!yM5-aI9Psz-@Y`T!22T#bDd@wfu$i(*a6h@C$_O(`O&rGy{
z;-k#@!d1(P_XxEA1s1tsA3wSX3C;a-t4h?GCi3DtSJZ==3yeMKe!skN)0xY>pVjw=
z9#g^A-t8mD6Fy5gT4`unB+y7ngJ;|+0|XKrBx4YyMeGagECzc~>EYn(7F#t5X8*am
zzVj3m1>6-Cz5Ux0u3&`&YvIw4{_DVn_cV(c11o{dEMrhT1Vo13w_tSGDaRf#ABWll
z1Mb3Ppqy<Q%$i3AdMzwH@jTAibuc+8^L|O!%$d_@hSq~pTSNbD7>7&&6hIh68ejf(
z9(b=`szY$A8BGfY-iwz{$|@?nJv>;+RtB4iI0%moc`IyFWvT725V`!ga!Xt|)qNti
zE@QiJySmExhNr*<_Xp6LW;5C;qu+3O-li868noPBuP!W{T3&u^t0N(-7g`vUm{eto
zLrKZZg##M9VZ-z+UMEbcY(`@pgST3}TwLtc>%~C27;PR?R(i$FGp(SkTqfz<YEKjt
z6eRa}&9vrm^3LSDHyB~yPKKHBx-)n6JF6D?<S~SflaRBWIo8o~t47r(CRQALG(K{f
z{{Bhb8zM*<ox!3D%Ctv##hID8WzHKf42K%8X&SH^QQG6i;HZ?8oZ{w@PPmH%sU>Fp
zt!*C)6iaHTqp0-xLX1gI-}Zuz;zwX=Kr=k+H(gJ&6Q^g&Yaq@M9*;V$ruJ6%ZMu&K
zYogELfDd{{ifNvx^=zmNM)5-R<f6^gb=JEU*5%F@#0AynyjZhyPPQAM2!TAbSg#sD
zhmvJ8;I29yF5WEA%MVQW04cI{tsH{qKt46=xPwJQ&P5UH@N+_Iyg^V(D)dc|a&3Jx
zA7jnrORd5W)v@J~%li5^;Z7(WFu>>~BqSt{RJ(_T*@y6iPL-K^ZS4eRNwlrYtLur`
z$={>G+z;HIZizw5Xp)d3!@=*bL_5?nmRc1O;^_!^w0~ata}Jf+7L{>v!*d1e(Cxq$
zAOJ5|;qNC>oR}>fE3NrT`A&%UR3DY}?^vu49$NidHeK$|gBuLarJ@G%nwpxpTuvqa
z_2uVw?r*;e3kw_V22{6?G5u>67yUO^1{#GlYp3Sm&()c!UX63-A4}T)+tl5k@8rGO
ziL>9WQ$9Vzhh1JthH5stvQ`lszU_tny_WROA4}F6brmN)-qtM=z$Bu!jy`5bk2Zu|
z;uXXm=cwm$On5+SPm!gb<<FJreH~drGLcPqba$~>Z-WS$-_5ns3)`F%99U+X<sdem
zmNW}hAGv>=*E0R~E80Ie0fp7|E7)IU0qB{P)ooC7PMRg8T9p6f0vrX{ER&YgG7{WI
z4r<5n!~T3cb)QhzQKo|iH?CrTdWjTT-}b-UaZSz5Ue}IFyLwC-xY{B(Z6`V;{TS1r
zlkL>f0OrHH7v~UwTa%dy@f_aOJ^<svsaNfksdUg`Z?0kXmh;5UtbEORs{1So3dx#G
z`Wp#fzR;Yfot^)8$0xEgv{_-e<`(w)M4fM}NHwe>Y+u2BSo6@@YR74ph#gu1^D1%u
zyGosIRgo1renJ!+3q?Csmc|naGk%Qv7afu|Tx+d134Pwo|BnmMlZT(ps#iC;5+=2`
z-d*zHi1Hi3(C^eSrPsXfM|xY^witAc#_Q9BXWrOYIA@t&PdvbID`j?mrWE_~&fhcI
zX8SazMgZq}OG_(|J<c|+V)}s11xyL2=c1(68K8-qx}OK@6J!u0kr6mpSPeC<uXT!^
znkQvve-DgE$x1roVlGcil$ABPZcIy<7y1mxS8&_PN}a2$fBWd}OClu&g$GOU&q&=N
zh>*}zJ5?2ppf6_iyA7r)L6E3hZodUtO^g~<@6wXNWJ#=~sHFd26^C)txn<_8Xq^${
zMb?&WZPTn)ELLcrrlxh$G82O$ns2&qiBVyRwV0dhzBB2C0-Wm^t3(gjnH85p95F*)
zx&zL6VsUn~NU@>$+^Rw2!Va)^kSElf<i5d@(LuTe;O45MtGqEzDcsl~-VzubY-??a
zbyJAgKNev%7@=OV3-!v`GdB9!Nuq({cKm&OyoChvuY{Ap>l2b!Zo#(l<A<V#RIV$x
zZ{zDYah2rEL=z*eU1Fax*;gmf`j^*7M&Kz8Sisl@1zOYcXG5^GvQNGu@t&bxpJM2J
z1kwE+$PFVaqoX|?qbR5D=U}KPmllD~b17UD;e!}qYrA4OC8gH3X+S}@D@GRa9^u`4
z>CnPwn=xA@*4d<0pktCVf{hdA!Ry$taBG?Rm9l7ewf1!(p=(O|jl;t#AD(Z{DtpXk
zOFw}&2!Pzr-~8MU_+)}|#{^n`=b|D!818{%?{%r6jP^H+aWx-Z?#}t^r)p2W^n)D?
zB3ZzIuEKurZF7^LaAWmFz*Phei#57-M;?Z)XCz5ltu3u6&&B2b45qUoB80&nSiwuv
ztMSR{znVK7Pj?+(YTO02y4-`LmX;O(!)zzscy!srlfN3x6N7`-u73Z&vZgK4<}rq!
zu&rIbFI*nMBki#c0oXT1G=CVZZwu`xGVW-U>72J2J^(IFQf@A2qZJv{^9{$Wv|Jyd
zdlovLeSt%_P)CP7TVuGtt=OzND~x`%r#a+VXmBvEv&+TmqhQFuDRJwQB`Gkb<lv8E
zNqTjAX?EUa_g4qN;rExI)PbMnvYXA9d$rFH&hp22`D(?TekA@0qy7BgIlx$)oHo%c
z)-8RvDB+>MSYl&#{Ii@gR$*r(VNOLuqsHSp*w_4m3hNF&^CWO?Y#@G;N@u$tBMU03
zcg&9nOon4Pe`wYyh))NB?AB2-2F5+ntK6}iZqF4%g_6PQs<_uMAtl4i+*sS(IBCnU
zM^0Ql#yCtWvV;lcQmw-n$?q$o6cAS7u$(W+Ev`#mPrsrhCdyQ!I->lSGUp9*+>12T
z@F=jg77^c>(f0v_qb-it^!U3a5J3PB)A(U=2kGtYr3>dB8%ju;-DV;M_UmkFN<y_+
z-*l%~>>68rZDo7fOmnDdJ7hrK-54>O==KtU+qgIC^Qo|q4zRL7*O3ZaJIbC6`Ib-&
zTKMfmfupJF4m&J}l<?a~fNnuY?6bkDEN>C2$2{>Hzj;+_Dq1A6Ld*Amu#<W_Nj>H$
zG&|mHp23A#)4Np>YU+##P*Fs&Rh?U5Va@s&_~uA*JFDVH?JL@L0@^k<+9nUNt%Dz8
zg5==1<+C>77Z(PE%oa4*UJ_hIX8NKOf!|`dCRA!day$5)f8SmlX|;YHhg!jLusAHL
znU^wKvNZTxMO3uA>v^05@9ZpJHG5_z4P1?@lbw2jj@#IF2`gK!#<M9Mb6WEHXy$iq
zP-A}PC?cc{Ytv+P$B7VRW@p~5auC~gdg6gcg&LOpAJ8tz^Q|G(l^c?ZIAoZBCUfRw
zZ*Nz8>xK&g5$Mt~GQ_05Gs7hlg`U>Vdy73)MJJ$jCA`YNJXU0UJ<9Gr<HTTbscb;l
zL5=msOIc>bptK7*!oO)>-FL~N+ocrb<)5I{+_R5hw&wE=o^f81hpq>VA|NFWw>?e)
zMA}-T@(f`djm9naI^9qX2+1f4ihxPAINnFfaK7f1Q(<krBS^w@nM9l%61(1zQ?W6A
z*DNE;djIg;@Ij;5VEKx+c-?eYe%fjSM&ct*6KNUC<}W2tD3kp$&Tf4Aqv|p1*S)#8
zY>G{c*$wJE04+ztIZYF7xmy8Kpa}sZev8#!PANMEKjA97sbrXcoLwhIEaofyU(i-T
z+5NInYM`oq_GS_hziYMa0w3aV57Tb%bZdpf9MEDYreswVe;A!9A`Zu=^<@vP!zx~m
zj1+1epde2R5J~>~>HruU>bqOlkOXNmV83X0fTKk?G_vcaD+A_33atw#W9SbD^A3_I
zr;uN7pO)L&X%@SfsyKPk`KDwiL%8(Jw9W+uO=|kjy7&-6?!B2JUmu@?t*V;{VAt^2
zIzeWVk6`2so9#*Lr-P3I$orVc9}ef)@!tMGu$8eCmBg^G9Sqc_Y$rqWx`&3bC?}&_
zVACi3;rZkOk;QrMRL45QC0=bfb~f2{@;l)VI7-<S=0!1Jp{InmrT8ksysfu*SD_Wx
zY^0<n)@qR2@YU;BYg@ZPprVH1=hl4ty1#Q5L3DLf*4245UJ1)J%?@b<bC^%;Y<53D
zL^_+`+Q?_SgOB5c^48S_YGW0Z20UI>E_avgzyQc+w~ekN{=DjdkjIX8(rxTi)qdUB
zyl=l@?bl>Le)$qKF7Rt@L>W+o?F)^lEtrQ)<f@fkLtJvl2MgU&Vq?t!(NNtp)Ye*E
zUVY{?{WW@&2Z@)-fYj_uuLRWUoA(5WLcWLI^sJpIHR1eMDI6;q$bS`<lIn+O+Gq1R
z3vTw2Aj%^};JSe61axwkS;T3vUO|3Z4B;XgLc)8?-9jqDcu4Na#1bt10|r7^`ZiYf
zXzxKZY+c49Wd}g3B))MLaqJlcP(uLV9RykfX3s0geMU`6@-*c6>Bs?uAu}`gz!sgH
z94=H}5B4nA5e3Na&Nnm<39rMs!Qb(5=%?xHv19wyw-|`2grDlN5h?;XSBB=)oU~Xz
zp9jUXLCygvBl$O&v~YA6%<HAXhc}|CeAz2pVMTzFBRUP|GbEQK1XuWaN3j6VFHO}m
zfq$XZ*S{)y=&QlY$8aX6q0!yn_B|#@&!qQQQvP;elay#HBh{&yX=ilB>ilwV!6&vD
zso)?jF(&?Tl_yvYmEXuHT0uvJAfH}9_CuK?t;OypF}<xV6jNZ9vAw<H6QKz3q7%I>
z^mBR=F%~=hW6pO4;Ehse7}oU*OfhQQub-b=!2iLHyazn}3;H&!UqRaBa9K#K1nKjm
zh{xc;D#&rV(TmZAyif6j-d&-TfyjI0dq~)Qf|CcGB2-wyl};}uLR#7x-*znlV~5VH
zt2dv~^OcK{x<++qC<Ouu)y;)(fJrR4qNA%L|2>4Yw{LkU8+yRuzWR5=bmo_jFT4GG
z?@U_f12!xl!E2sSpP-$o?U4<a7Cp0=y=tkZF1It5@*OyGG&Hm$rA1Gwu|6FZy)(O!
zu#qTDb>vwext#gG<31V>I|&XuPS<UR>A_C^`16Mto&v=CAC-$-h`~<x5?rBs6w#7N
z1EUg5A0uO9l;>op`bz<IWOlF|pnPV<>l^Q<h=>u(o(_x*O-RU{U!J{({Xk#e_`%jT
z1_E)QrR-hT-vjbCZPiReCP9IUEZUTpN77s(57+O4T^%;+`!{4)5Idt`{SZpMb}Cxh
zLY-ukC)4%D-JNdC7dU7!QdiS%2z+=C%8B<_!OgfXy4??ymHRVDomq+s|B!0vT3W%l
z*xvkweit_Otm?AC15Gqu7eNnp0EM1!c4CqxGKP{K-S!jawX?wWZ*|^1RJM~cv$ge`
z{r-jy(?umQ#cVV2XENC{*6i1LW)NfuTFj^oKW;%MF_rT2@=VxLfl1))wD0fpx~aWA
zD>s*zqI&CsdRdN<=^+xIk&s&__e7TA_)x5$5YtB}A}%!TC{6p(ZIA^R+S*7!AaF}3
zFKOVL(@s5FKQ|OIe^uFg-CTfXHmk!Ap<msjcl_Ht<|&r&_3<(_j_;C9d1R)TGp+AH
zV7H(t=MX19fk05e@x{PQ#lzcNo&upZ4vvo3m=<pm_gV;ic*;#lq|JKSJP5d#3D+Bp
zSG|sYYAS`w0zGO#aEN4Q6AcXm3Ig4p@h8P^cL+B4Z)E6GQosmAk$#7eCqFJIXgcB+
zpP-hJN;zcNvdGPpME%HnQyA~8<Ud+@>J&=d-qwQlemW?K4ZrjfEXMY&uj{%yjSB7V
zLR74tRUsQ|FQev`ub=xZM1o?bD1#PIAvMBMv*PJjx(l<J{mL`KLdr`DI;??a<^wN+
z#)spveFVkCq%189^_d^u3d@1e0Uc(?^E7hX<=;@N5<Hr5p6S<Lf`<_vusWs~;o#t~
zteWV3#n9uN-SWLh5;kxgyd+OscC=G+Jv&$(yyw6_7$2h1mG+J<!|;&`9v)C-vj)rw
z_#EOh3hr~WRmG|<%+CANYXQJQt=UR}<6EP4H~dfY)UB&=k|)QUHO~I|tE+0r=1$Ga
z%0!IrBGrZX$yfLD#_wq|N3;=+aouN%%-e5xJvlp@J}^Wv+MmkHO&)Ui<IL%N^)a8O
z#o=Luxu-Wk6CNZoV?x#U1>)}e)IW-TY)o4;pL^rK&Q40QR4+fu_W|;l$H{8zi><Fx
z_DH2RBl5H2nLt&X$IIGSa3%I5Vu@B##m74REFvM<YLJp9p41JX7_4z+<?u8Ym9E(=
z*sf+KsyUdLqgl=5Pwa3ie&pV|wavlC^wAt*11UFAkhUn^_VSoH;7ET$&CSa$%BArX
zD04ll^(Y`sKuHdlpr#0Jek=0Lkd+4Ex$Qcd^$M3w0UKF-YpQFehqq7m7$M}n$-!i_
zHm!3P0gxuDpP{t01O<U!^0y8ZO%t4sFl-DGAydW?x58!c9~gcfCzjG<ATPQ|`D}LU
za{_7z7B<SS-nzw}2vR_Fkb4ipCcy(NJNZ%qx_`F*CBLMk1dCZ&5?)pK_TDA`Ci~#_
zX1jh^os_7kq-KSdC5)We)YjAV!`7$fu?QY8Vxh+gRj&3``RWiS<k2wZIczPrf4`}{
zXv2l<Zv53g_<<2rq}8oN_}0SUO)0}Vc=BD2EG<tb8Frk~#-*EP{X0FSf*(JiRk#rJ
z@On6*V`L&o5(pm{S#WK2=Fj9=N%^*3A8oiLK7x2W`5Qm{vR0b&vXaL7+je`)!~w~z
zbLSf{`^m{IA};Eh#lhGArIm$={VD?Y(&75#-;t(xke_2(O0+$@<@S($u!v|22QVO$
ziQ-6yQ5&w9CSj7mhqueIf<?bdJ0H=A!ERpUPL$XdB905syz;?15W&?dw7UHd$OYWP
zY7>JrG4=kV23%-J7VACr5i~N^1KtYOBW&?x7w~-TY+`0LfN2?+J;IyoP8?Qov$Pa#
zPdgHGseUp!{z>K!+L-e_66Hj4PV$oE<lU{Jg)1MPtTwYy0?i20`<_$X!@>;{LF)xB
zJ~nRHy&2um+|j4*`^DX)X#PYjJiMgV{ii8)#nge)@@i_b8X8;5o&f*O&gs0my*FBF
z{Q%67?U$A#tix9Hs2CVBa?j7v{;=ZhEqgu)6925tfAJpJ5!ZQLk~82_=rTiz196F>
z`Cw>dC<?OlC<17}5$uku5H!4KGkhej<P+i~K*Vclx9G66+Fe+<J(Jb})#|%#k8U{L
zKD@Y#GI^Sk6*Z9KU~{2|5$4QDrL!>p60tbSgqXt7vf%4LUpJ+{0Jmb;Gmv|`n=(+O
zyLJ%4fTV`#F&_2<%53{`IeCd_m}vqN9byxI+SHq$tbc07>1gk0-Yc_9N-UJcbO)jG
z<w$Twqng9aT&2RLxn(ctyN+EZo$uaew>IJcU346wIzJf&%KJ+<*z<)QaP9D$wal9+
zG@eppyvSNOyyB)up+9g=dcorv)WgW}=bOXP{I978BkVa~@B~h!K;Sc<a7|CMou4;l
z&sHHh3UUx$Vc&4rqbP^8sltVFX)2oY{Wq9Cm6fNEqOv|zjUtj<`>|Gtw!c8GN021Z
zR!ek+hQGvYES2Eae8*;ulLi%5YO3xkpG9(P?IrW1(n-!<V?3&9akSD!HOkIE)&ng2
z<>eJ!_i6#t&r>gR7>qxMnJMrtcB?E^rKI*jhg_#@h7o44Cq$%4G!Ye+dkiDbSiW+9
zK?=*V%As4K_s!>JH#mb7BJ^Iz@i-p?pjb(*aSJdiuwJ=reZm7j8Op0zEY7IfUzm4K
z17qW{;HNJ8K+)OhAZBdc`v2{M>}2-#2LS<S^n^^n2b*nJ)7iCV&r<~yiC?=T5k^D}
zKSBjbf}3kM=xbzgL1XOh8qjdkPhTA=MI7wOf9~C%heKYkgU(Nn+Ix@Ra^*Mb$uO`f
zz*(56d^xL0hxz8DH%-^o&UK|DpMEwJbkFTH*P<uABIxMbC~jfP&A;ZhI*vw-LbvF!
zn9Gv<JUt^Ui}vpPw?msrmLKf<<4q5BHdiOvI=3z`YH;v9M@52k4e)Yr@>#pB;Al!-
zl=*LI(u*;lr+jc*><8^X6!sDk&%6ddq;I)Rlm4xt(%96OvN%MoQNDM4_rVuIm_R;E
z`q!x*N1b&u;b~!n0r%l`OjYk<t#q)JwO#&Nms`7+=^i5oz^@`CdGR82I2Y@EE&t)U
zL@WN^x}z&>_C`~(+SPCEvbeaIA0LjJtgAfn=&JF7Wk{PReK@Y+%%7Q`ehA~=>7b3g
zbkBbfryDj@AdHL)T?L;46JGquJ5KsnD0FEG-sUX3R`XO?PPKKS*wtQSF__=9jP?@$
zNvNr7X2FF-KpwMQ3rP41IF&3tV86US%f$E4S)Os$<o^QcJ})F!qTv1efq4>4pAL;m
ztG>>f4@XSGmk{z9wUyN^Hw77(Y~i}NL{<FPr{6F;yv}MNjuX9d`EzH~hgV+c>azKg
zj~xP7`IpSOK~!9vKAKxj>RaFF+RMv==2#VsGE4>weGgsfI2idE&T~R^M>aG(nJ!;>
z@%0OeK<jPv+sgF#n)12|A+ru(aIDT<!Dc`6@7<p>6j<KmFu#u32!T)-Wmgxt{?$n6
z)z1Tt#KV$_yV4ISAN_j^sSP^J?z@;@yDv$8?)Rjm+^m&QF2DH(+ui^B%K$uvJg-*c
z%exMLouK#rziYS)#x>SEZbR!*{|c~0&8z?2VIq%CTpis7HhcPimm<yb3pi>pY}RjX
z=`dZV`Tze%I%Q>Kpo`}Cf8Q+k=l{MrL7Lj}NC-al%hx3C?S1;+-BmtxLAm~a=*RKN
ze}Mb(jE<Fs=zpI`&7FTNxCDMFQRN6}s((x6@}epJ@5cGROpxBb<r`Ij^Y5b5McLSe
zNf{~iVL+mkCiI6<8SuTL0EG6|PF`%xGqc&EfcLRu+yj~-lag`z#3-EtSIjhc59WqQ
zPk!*<x+$$7b6yqW4KHv;2x?|-%9=9c5C7hEBvuI%Zf7)mHym|bF=z;*h7f~#$%n=n
z*MrOv`%fg=)67NB{L1!#t_2_##J5yC?ZYaEk^$I$H3bWo*E}IihXcgQH>weA|8IKn
z_vq-MWaoMe7*>bFB**o<Fxd5#y;2%DI;?S8fOkh<Q`ZyH;NaN6n+Sm1*RH&PaS-wr
z^mr*5C<u_Yul&A{EGh+VXnY<nLDUl*--|pq`Ai^Frg}NOcrg{ci5t{>B*SEhJi+wg
z%L;MCe*g&XG!993W1;~oe4iwQa8?}nsB@QziZcC)^}xjryQiNLE^)o$4)kr0BBI0c
zNXfC^K79!AEc&*Uf<oWB*R?CcE->PN_Btr2XL0aw-t8mY{PQTtdJqJk<gZ0R{~q^q
z9DbstrQv1_<j!NhBQl=pt>A?r&N3M$h}j$KusWJ{1ezk3<2(tHty+>?o>^YH0Cw`-
zIXl0qc@0_dKvo}F@dk79VOMF`!RRA3hSJc^&Zbj+7r!rGwttVQLO>U0@KDa4Dv+2Z
zwjyDut?)VrO0eRhlGMM@j%Hj0Fu{R~{L*1A1$P*5{sPom{FXJQ53AAPWZ+C{pG6tk
zvD0-s(}bRM7%Vpkd%-hijhlN9-4iA<ly^2{0>VIj(%aJ$&gnXxeIy<nB=@<(z#!f9
zP1UZXPMgVwZu{9)c&ld|;9!?HEepiH|D6pa{1*e9+JE|ydj$BelVpUny1_vtQA?{P
zQjVE<4h~U5UNp38-shSaO|!sZM0q><T^ZHun?gWXSac+p<yX5~?byZ~ueVg5jyOAt
ztFHm7+N+tRo5kU@`|x4hX0BPH^pmRy0N}sk?}ahAYENVK_Q?~hav{Gya<aha;jf+}
z{BXYE%KE+{PL;C_q<i90Dqzn6eg0b%vvKd?l7Ac@r7&_oGb<~u#^G%L%$vep<MLy~
z0ZtNkY5e~den+Q&5^-n(^flI%sdpAff<;zG|L(L#Yuy`=`}WQ57I?pX?=5c2HbETD
zr~4*9xcbQ15oALaZA6AoAThFebb6<M4UxJqP`Gvl1pyMz=*`<#AeMlXek8IvA9x(e
zbZsM>4m|d_Fny(@2u48&vU5&1uIlONK>*JY%Nn=cXGw>OjmfR@8G9I|j*BZ@^k0Xq
zd6AJ4%ZK?HXXt#x{}2E{H@C-<%3%4H|ArX~T`&f1nV#+C6~EHch#w%b7Y57^B);71
zWR4oXkS?=?lmXdSc_|%#H*q%}1iL3@*3q!DK{8;?RRk^g1sLJ!Fw0bXm8%#G{sm+V
zaX8|#>O0OcG<*bmGkq+7>%}l9F6GHY*9F|UE{?ecT-P5G_?IUPoR=p@maLXr=6$?s
zpjQSe;g@fI8$Hbm6`mBx4IvUlT2}Td0(P}{e!?le_PAOuU%mv9Y^6j;XV1#xP>$r6
zoF{0fLBouI+p^DnCjM1?%S0Hi17wO9Qf@dB3m^oHH7No;;`)i9Fgt#?sEzjNo$;92
z*)qAd@EzCF-J?d0Bw6TRjE+u$v@o&#=5i-Pm_CcQ(qLiF!W{JSIt9|HPhGyw15E<9
zokwyMizU5*K|##s+Ol$gfb68K4Dj!8^O~zxP*RMkBu1E6%9Zypw-3=(y`=9W2w>_h
zEn@Io!(6F>ywccMkIQnjDLdOl{}ng#Bv1so_i(oh-VD%bq@t_ecK4ec&sqYto8XyM
zp`TAdYAP23h7%OQ-OXdy5RxQq-Q*NMKNZyXA3yRSFrx>B3DCRv@n=Yhv3x#$By!x}
zr~-@5m2>-Yho5)wy7U`kUvijTYm4MF9GpafEImIT-!zs*c>Lq@rLEr4@{R$WcP8>T
zyj8c-=*lbq9f3E_$ldsag8}!=+*3`HdOT9|`gMZz;S;jewxd>P=zyxNwzG4vI-%dc
zD~eybDytx)C@X)7lyQw)>)jgw2A#wH1_p9K=uzqXFC-8nD@^Jk3kmyq4)7vB9Ul;T
z*=cT7QInFc4b>b>xsDhcAHg6G*m1aslg-ZJ>N+|CWh?R=t3qea$<#C_LXJ|*jLGrT
z0D0zm&P>T2Us;(5t;0uVAR~YR(2br5*(c(YA_)YJ)Z*|^PGO<ZNG@vWm35R~K7Vn0
z<rVZK2s-2cg1giuGJY`0Bsu~%(vFTo3bwY^{Cw~^U+7-?bI_*w6}<R=G<mjbK@}Cw
zA|fI$3%|wiutdZh>v9D*r{~Q1@=4w&=~}XZA)m6A>)sRS_qf7ZPnM}Epc$KE&rgqg
zP8<wgTuK`Zl$5Fp3iQggFe&e<xn#fm|8N8F&i`cHeEMkDk`(Q<&J!k&6NTDDmaA;G
zwZLEGmucE4vKkA84he>AU}AmNkFPM$OQ&1_)JAYt*iv-_PQ*{Av5C~GKKoMtQ|DEF
z8y=q+F2~x%A9|@ZOR!NHtk+^A`K~*7XfflX^uqk+edYAzy1J_UdEWy=&DP6Leb~(>
zqJpD1%=?QKoys-a17d<JbQ}4|$vwdsjLly6!WAq`8IC@4)fL4BziiCYtC)7)8QKYh
zxeKM<+fzm7yM)?l$&wMz8Y=yo?O6bLF&(Z>OZoN<P9(_@YxfGj5xa#sY`Soks+8Og
z^Q!LiKa?)HV^6-7nyak6)jt^%hYiWEPU$5+tSWx_&u{3?jX3KvCh+1>kOUF3(}po{
z336M>fm0;GY4h60FD*36p%M>9SD>{5BS^Q<pVqA^XrH$J>VB!8UMj+c6o}ivnKg4V
zQ4@G4!*nok<^u5ei{JjBdW?$Fgt6j4mU9?15&&tyf~KJHlHa(428CVg-1gDwL{E+%
zWaX|}uQdSTYHVyg5{?HwkkqRRp$DS}z+$p7ISnXu+tcTEJ9d${r{x{>|F{4qS1jA!
zc1eE4IN4q8-~2hk=z_9bs@<7TynB0kaIDYgVf&dkAl7Q7hTgN{hKAG-^r11vEl2fz
zL*s;x$*`|HN0lS{n3?Jx{)OG2P<7`Y=3PC#hYAXkT4}J)v+;HPupKyWllbNZP&D}t
z%3JuJGRGrRiLZcNO9P>`uMfPMP~S_ggtj)fuy1NeQcsPYK(Jf6YMq-f8K!mFC?h}m
zO^;d#C4wWWNOOY_qwGE<xBkw4RUMO$1U$c%#VeCj^AgA7I1E6(^%<#j-8ff|f+E6Q
z6O5+TFAkTWGAOp4p#f&4FJCnXH1wDwxY_eEE(CDIg+a-~H>IxO_Msx{Z)stQzD||T
zm!-jkvkl{qb>4Ne(Cfi6_R^sU6akGy&2sh0-8gCKfQfvkrMv1>IKEZ$^uRKHYBU&6
zu^7jH)nBP~&Pphe-%{U8^WMLHs>>Yosn7sKQV7dQzX^{F)lhBC%gsm#XodkPVlbP>
zD7p%Uh4x?&ugQCIizA22eQ9~?FD{suq}=fiE;^Vj_*hGl_nK&Jd4jODZ|Ha5&=l^$
zeBQ4)s|5zN*vL>~+SZ`3;9Jg^=j*Q1TQ4gVzK29s7!jqU>VmeRd&veZMg!uTwzk{%
z@a?4@kH1_cmEmWOZ~@v`YKFcZ4kc4W``osNK*BrYGrjc^uc<AE%c3%VOhB9t2Y^vt
z=W}N2h^n_82N#2c`_H0-V|M!TwG9U4fGGZallIl;A|0KFx6uj^UG2(?fbVTmeqVfc
zBDr3W;ao%GB@1re-SBkt5FqZBsnr}#*W4JnKsPMyI$b^xP&*w-p2gBK_4b;y;;&}0
zvMM}PaU<i{KmQakcxb<)Cuu5C4>Y?eorTkN<!cej(cwR@28sZ*v|?cHykqy)Or!cc
z)=mX3Y|dAZezQ@+-OX<EoGJ6#puoV>nm`yTBxy~<h{%1n_IvMc_@DzAhR&~lr6@6*
zo1V|D<#_jx0(5K*sc3Xkxq%-v9*i3Qj(&7&zU_wVsY!jLB#vqOoz-y?&Jx(l=%n^r
z$anG%kG*+sDoF=Kaa&(j*k{+)N1Yd8Q$Z{!H-7{F5qYZJatK8c*&`z(qtLwZLkccM
zbxGUq(+CK40IIET-q9F3Fb+95G&MBLKs*Gr%szfQknFG4Mff85R!bjYw#P?C@j9Yx
zvL|E(Xv^eo^=W?%l%rQANA{DGQ&(12a#YISKU}c{aToW87lOnvnwF#ZM*Qt>4@<w<
zaJN|AT>kGf4%3m~nBbBAHk-Xd&${(sXjSw()bF@rYA7fJ`G(i7SGsxe7$$<<!3&9r
z+DBJDL}s>tV}7~)nqV{_@qDfw3x*C?!&+Io2BMzoc>vCE%)Plkc5(?^N8(ZXufA(;
zqI*-gJhzgbPdOEP=ImQP9kCM%o?ENqKhD8B7I^#ufd;t)SkrcOnsRb~dp-y*8~k-!
zZ_5yDcg-F9N%WY@W%E}{X3O6N7eTFVs4g4*@PutnuqUFF&dff%;c=?FcS+ph&>vlV
zr~T><Jr$iwq1M<*osQOq=U0i2Weio@VpZUbS3<?)QA(7~x;#}iVpFUNkvEKLKZLE0
zfO_Clpb`?IAg*KzhJ+d=HXk&^)f1$D8jXdnj#dIXY`6M$EZp9C<C`DyeDxNwD2No{
z+?^<Y03F-A0-fXCRW7T+CwP8nh@8ITK!3m4NWM|y_{(mlt;on_@k;%uhmWp2S5VmV
zY#mrf;{|z8Xkh62P|i;J{ZA<=$;f*pRp<5Tkn49Z!*0HZd@r}Pyz~zbxy<$opQ%!N
zq0!--JJ4Y295h6Yma_{4m9AfbbbJfnz?d0SFW`Xfc1ouOYZ7icfNx#WN+XqK*ANfh
zoBY<l8~anwfESTzY*abV&@rup`fC|uBwPXwi3ULIkUgT*Kv$3AIue?z@;dGBlFf1V
z8IkugaJOU2CU<zI+2T3}%=@hJ_4(~_Um-XA9c%$Ke8sJ5%@VABawjZA0SYvF<!9b1
z3#vi&L3+(De67^yYbPK<ATN-sBPqz0@)ezVH8Na$Vy3OSyquhxQseZ1{vJ9N)|XVT
z@>zLi?!C5aL|SzpQap#gQqSNsr>@p#HGzy(Z()}C(QS}DLqQ4wV92D-LiU>qW$~9|
za4S;RoTo9zWvITRR{Xd}>R(H$aimre7+#m!Z2)Auna1040b$qQ1J#H!Etx^%=}O1g
z_1_8atP1J!G9KVRoXVNo8896`I<2QA7xFwRO8cjn8$2w9+L4$U^f7$|EQRk{bCm+S
z5sUVX{y(o?BX8mEVnlEQ)NsZ8_+H)tIPaSuPDDJj^N5Z%fd)EoMxR8Zdfv)W`2VBn
zDuAkNgZ06}KmiG9q+0}}OG1$D?(XhJMN+z3I#jw-=?+0U1f;tg?*8}AonhvW8S$Ji
z_T7E<3I7KpRN)qf>nv{ARI6Yb3GB}^&!*tAgRYD;-g*}}A7If8g0gHAL!_UX%NAyD
zk}6_GfzQIP73;OqAk+4Z{Z143Fxo(<5K!gX3Ua=&0VCN)vz#7OAMl;-d;U6OY7Ka{
zq2trP?wFMMv88uB65NBK0snF2c=+To#qR0{GAh{0Kfd=7+9a6S*vQcc@J^24)4(|j
zqaaq$4P(a<AJaPY81Rtlqa}j_8y;U*PYVI!>gulbd?SCp4{^qO$AiB6d^)ec>d5Fv
zDj)(STp?|-`-|kW&w{1Z<%ui?!C$_-d4F@=u!BneI|2O@KYk77{=sn$op+F!<>T1j
zzoyJ*^7Y&qZgKxoGBDiNX)^@^gziu)9Ih~yzoD&6jcu<GTf+PO&juWn0o%Hmf7I9A
z9UU3zY*&{4+W6kSgWl;)mP`v7;^kl9srHE7bf+@8*hxwC5A<8a#$Y~q{P@KW)lU8H
zt@&D91U%xfe|&I{4_@N9|D_rmY~^Q!CT=GH&lR%#fWNwb?uoH921JCYz6ztf>}_sG
zp2So*6}Eq@^rsZ&=f8AQ<*joD%oG)ILcoEC5&yE_3QR4)#(uDC5#b9*A5??q%?5~{
zN0U!#8^T-1lwTzl&x?n1@Y!9oln2H~cZQ1_=V|RX3gpO(hT|}sz9n9?g!RuVqZVik
zPBuqAQIKKa9NBjL^yaN|x>%<iUqbdq{HJ`&U$?VA0QGG|Vz-#2e(`ig`S*r~N`Y^E
z-?xr6oRyTkOt!CJ_-K<30cZjig7A~xFFa9a!JjBN{Co7haUP^$2Apwk1BU~+8<b>Z
zEFj9$O2tMFOX2Z3SHETt4hwn_(hlh)vt0FeeM{vw36saqw7GI}3k3~dMt1j(TmF_+
zF(@_E*Ixpiigfn3g2IGEJPD>mQ1SwWL+k@g4@^OXxLD=x*UGJ97r_1k&IJLl8#cmE
zaG@o~0<hdEbqzPEWK}h#TEpdwwFHj;7zFUJ^E?e-n=POIx1W`dxpBczFG$RP7eN%x
z-s#aEs(}RWEVxZA8?Mr;0_xp+^#=<J0~Ik_@7mSf1Ikt<iN@XeI`O^{&{gP{g66l;
z@(=7ouVK{=&D(#vA55$l9{pMbg!ufXZ*@Q(4GfIxt*aVlAlsM$7v<w%rLV^?P;Wu(
zZ&*5)j3OU1mor3m<m4Eo&T8%4=k-iaILcxSf@p*NQjnM7Ab_PXu{A0hp3b0Fx*(<a
z#os^aV*^3y1V*ShD+{Z&fqqOvo3y09ga~J;D)aT_(OE=7gLm&Mu>L=v3Kf-UVPze)
zvhwad@{>?0vYD<VexIE<LX`xuc>G=qT|wn-1u-!dZR96WO8BtTkBbwLx9<TaRZtHx
zpZ5x2hQO1{mBKU8n_Jmxr;8dl8>!+o8MvhG>J7!&5X&2Z@6deuQ*KdF|2a1oo8Ebn
z6zDy~gCCdf+qi6V9|%=aWE!jEZ6IJf@H`r14eM<MWe1PDpGcBKLke9tIWuCd+FZ<Y
z9(^(&v+!R{<%g!kg3^Lgl6?yFcfiyig7;SUc8<nuoo>w1{c4L79E((_C`*ZX-Ksop
zA(+Sk+AEND`1`z_tpn3suamq>=zVMFdZQ-`gm^DQaXGU~@?C#L?KbZrUiL-}kIIIB
z0ZE~&$%MMjf%U(Ym45A#t>Gn;)R|PR)NPnnp^G5x_I^2k0E5EKVk~UbmPDS8{2r}Y
zaoKE3e9}e(3MHUCmE;Y^f7iS{aEbx9$jpyC`+=}YW>9Zp3yJ0hV}@6$nm{sDFW1Aj
zqYa5pO=z2I@UqeNZVHmKkapey6&{?fcn9~uF1=^T-pQ!~8X<P;+5T@o?t@tO3tR%{
zCqIA((vycQ4y$T9s@^*YRwboHPH)_Y-;jfZH^wHPB48t~=EjMj*D3!`E|>H999VLZ
z!=;DVKmP5z{@1v@V9>!oA#VWsJ(L1tX58eeT{oNy9E9ncq{ITIeM|@^fO{bIMO}hu
z-*R^umVw;Pru-#3#fHuszm@ReDD(Ez63d2%!=IsEI<=9!S06MHLHMsN%c|mRtc{=J
z65_}2#X%DD9b5ngIOlo+vURa{*jZc<2VNw0LcC51^q~O8shGf6f=N2?uB&gc4=(t6
zd7$*p051^K2yy=ts<x4{Az()jf|6p*=Ji#d&bme=HMjG`L%z0o9DDmi<9?k3fK0(C
z0SCYG*+b6~i(9?SpI!MAptOcGWfy0MCJ#mqrbLF?@y)Ka&%I+!ml5D6hDvoAhDziQ
zxA=Qajd0WlNlK_AXadMVdx0t-i2`6Yx+4_WiqqwLFW|D7_W8}gO9G)2B>-rw{k;JS
z(|@Ad3QvZ8<;{%>*_imd2qmNCaQ=d2z8%WMK53|qPy7TCe<5cZCZd=(i=zO445XYg
zSr{-|{jLG96Xf0S9$bz50ToeNwdwM9y};Nu1DNWZP(lWK14W`5!J^b*QvI2eq2p~6
zuhsWyUKFne>%-12dp$Xj<JCiW`dJk8#R+Yc9jQikWnjOoq^igRpfOyHvPm-sZ2EY=
zE69n5TUlQ20IDaaRUB+gYzoTK@HCzm`v;Y<*=Yf+*Ng=5USe>f!b-;3xhN#st7ZBW
zUS!}@2JL>IUkJw$9r-mSPVW!QnC8N_;Oq_yPD#r0!aA;h!x&<qTrvf7LY>B)`5A_r
zU0Ww-YI=Hc*VByscWMdMOsq_Id;xY<v1z(tUNf~&t7476>OcGjwt^_?>(@~WVW28X
zok>CR?Oi^fe{}Wq`qT@E1VB3fkFXh6pXI}-rlve-oeEYxWM~gJx;u1VUmWic5ggRi
zWM}^}%F0!4T?T+$zHb-x{Kkt~@i8Kyl1ZDx#@I=!-<QMyYH1nAWJkKU&TXN<-3y3F
z_kJ%~RbH9KW;PR#OTguvz-4ABCT5l{kOdqMkN_k|cCa>U05hhwkOCG9v-R#MguLf4
zyg=|`kqNbwvhy=AJL(m?aaVJxc!t<>ZHN>?cG}*;Cd61}wSh=fZqnJw;epBF|464y
zc8{D66~H=;nwp-=T{q7SKN=`57VK8xK^{*Kc8_+UKg3<z1QB+!FllA^8)7!&9`JSt
zCHdH>Z8^tek9aVQRJ3$7wY3}+6y?Xq>UVw~z9@$A0~lrryavn&oQmJcY#s~nkwNoa
zkazb#9Cj%4v$2v=rD4HKYzco;!hLQpNzInDZTyX@_}*oJh(yGLC-Y3IXJ#5;))^$z
z_A-Bgg#eNnaLXlf<mtW@!t34M(Oh0;34`fN1PalF3qYaJlYz0d+1VyI2$`GLtS2Ut
z=*b}c9C}L()H1w~c3L{dK}Px`P7lef0pEviZk)Td+F-F&W2Tilqopm{K2zsl1A*xP
z0l~2q_bO$i>(84e5HCOiFfsy+7k^u4`~gVFE>Ug#%36wuKe*fQIin{P`zpE+FR*!e
zL?uD&4IDz4WI!~5^c6%@px}E7AWC%o6Q(gT#^#qC<lshSHC>@Dpxsnn-#auMZF7Gn
zozJKtd%E)r&f>xf0Y?05*mzHz`8`l}f_zreNDjIX18ZaT!M~oF@i-{{r<%pUXO2};
zf)S0C`N~9v16!g4TKV!-Mj%cCrF>)Kisj+*|1F5HSFC`|^#Flol#}X#s&!buyl5(u
zA&>Nun`##*8PEU~($a!budSn*kf)^hpS}KhLLD>bB&+5tCL&Q0^(oLPz}RnpcV+F*
z*qC}kbx@E*L{wDSYa1ay4y3=<L#<^82RmB4fd$$i;7pA=;bZE~ZB*jTfek-U-l5>a
zV*#cf$<?Os+Iq(!s00~78%qn~K}F4$?(Rm<6Vq8HDPYh5qfm{75`zpRz~Y~X_<#k*
z#AYtVBmPO*9Sj@B#}0S{!Wm!Dh)ML8lv!TR<^t`>ZSWW*XFI&MhT7f&0s<M%PZ+>E
zqH)zdCp-7^XwXYEFyD&Nkxht7g49iZC^`Pk(=V|PT^rQ-(kD<=l;faZBSm`*9M3JQ
zAIK-W3}g(1&rir1cv)_?JwR0gC?lMoKoEfEyMUSw(z(6fOIqiQKS%sCG=jh8I=rzs
z?3?-doOWKI!`hBD68aW4#(z7(ckdMGh1-QonPeq}-^sxbCOVp|FiqsE^{5@5>l>z$
z-AIm?r|ylryS;}Cr~_e>7K7gtdCQG|n0TEXeHO{G(rivDk82?#1^+1!ouckmBO}n!
z7kbBMysqb7@+r#6$syj0i@|ZX`({M!8Z)I`|LO3bXOFJWSot9S<!*UPMRhf!MuV?N
z)(_35gkQQ1PoB}y(-!6D|07VUTk4<IWlQGO+cLc*)KH}#1JwR;NbqpMIFdkD?Slf0
zPu&N{ho`6yr^+3km@lrxjdh;d>>uobnFlI*dbGlbNa*X_;k7FK@kfsyC-tXry=s0<
zOZkfSf{u1*_gDj=3JmR)rxwY{8h!B^dip&3UfueoV;X8js^-rS0E*b{2Q8qdbE=hI
zrOoV;dj40dP}tej_4!c4ArjiG<pzKF*MzOhzn?@xUc6kXO}gtzF%@6FZ)s{-^|wD+
z9UEsi^d7|J^u%%5p+fi$40CLHJVW#)vRZAeMTTJaB{G{e#Kg2AmhT1zhtbi#!p;2V
zb>w^ZSpdchjB5g{<fOQQ{Fn36IYrg#VBy`jI#>i*yu>F=Cl+@;b(5d4QT$v$yrg1J
zZ`U_7&u7r6@C^ufV?4Uo(LqV?Xx%v`h1g?Y%6TGzd2w_Y;Tln%(9d=WpKkO>(uo!8
zs4|JkD)>5#0flnJetX5gL*&ka%X$5k2V0UIY>`-TelP1PJ6oH&wvEzY+zQ}_76wOM
zR)zo@;{5*SE6*z!AQM7G1hk!h&8Ydzi?H|%^kv65eowz;H29@i4v?Bpn@DSKFB$dA
z*Um?QyPUO+lcun#9RE@E*E$O~GtYp}=Y(HBi{$VA$2)gmvD+c^^d>M=L|C<Jxnjg^
z++r327uqVn(f?*H1MXnW3w^UvVb!(b(9Pocs5*y_H+)|Rtig^-e{u1)lIq6d@4|vw
zmFtIGf4on>9zME8BfK2v4#Q>j-1rwn{+x-6jqf$bQFtH~EoqroZwEX*GgfaDX_X`u
z+s?cAxG(QE(<Dav)l3`}L_=S?FS@aC`{3a{PqlZINo6g>F-TGxa<&+W={oA_Wt_$L
z9k5m&6h(`VbYp<Yi=v<8ijA3>Ri3<r_m!zl-s1jQ7S%w8MG_>DsjDA834SxG<l*%&
z<i1zrK!S840&!qph8U9G{#1zFSp3x&5ljyemG{61^)Z5`U>zZ=rBy$BgE)~jTqKD;
z55Pka6BpB{=>y~3KNDTH3p^|p!Ear#u(7ANo$nzk2$N!>dbChR>D1_z;e3X@;jn;|
z6YJo#_0YO1lUd8es|Pf0oN87Kp)S8NhD-MqX`dq2QkM`(zR%c+{Iz9)(JJ9cU@rPk
z#5%ti8G-p&O+=lRu^H_mp#HRRcW$>UAvdFl%Vt3Gi2qsmU0)?^i5kY7I^ESVDVhRY
zoB2knzxxBCCr=(fJi48`$vP5s^YXfjfH122YV06M24$Xz@<~a?gOtaNWMuuFozrvD
zaGmboqc-ULBH7zDwQl=fTDb#2kAXh3KmZGPF(P&oN?s#i<OvE62KS=Xl@$vMr<!kH
zh*${_iZPfO@hSMc9|bcIz6lAY#>ywrtHo?Kn$t^0DgN}YyfQi!AX(x+l0<-aCfcW?
zFZ77FPe`PmvbKF3rST-@eJ`b?rmjxQKxR>NE)Lg3j!bf8FqMXubY`_z-t1_BQe?N4
z%$6es+PI-hTSp#Fil+$2%E}>`ina|5^`wnutk_67CXx)|4LrPz@=d2v93oUiT<;5)
zM_fXU#|uj)jM_~|zQjh$h=`P=+#fj~+|lGwh=`(6QY5Xkw3<Z(5i!z%f|Maa!Qd^I
z+0;#Vz0bS;j55T?eC*e7SnEQSrFyvjBPuDJ&BMOkt^4eFp~?yh0e>E#qK5CkZnb&T
zz0&(*R7pwFA9o<#q661xEDMLmjDPb>6OrQv3gYiR2jcbU_-6Bm2ft)}9wIz12pS!S
zuTmPGb<&ZL{EBQ1IKOD*Tl9>R&Kp^CM0_2_oeAj?F-ZDA1b@_(RmsS5+ux6U3v*6Z
zXMB8+pZAdQGqE)frk!rqFJe|uYzsO$Nxh#4=*mwR8)HL!W#JkPa=5}o`}$r&V&w(q
z_0{*V62ea;?ocw#aMflKdqqQEH+jrYUv?j|fA}TjSeXZ2kXAjyn7Qm=*rD=u;hB@w
z^B}NM!o_%zo)-C;bL1FsRnM7($e`|mXUJjQJ@4^RhFOz2r8E?l;)56OUHETom}VR>
z#3gCaZ+Dr}Fi;c}=N(eM{x?)QqIzp|Vx1*$u#P&6g^3n;^_;~c<8xTmN=<3jN?+30
z>Ug%pKXw6q0Qqu_3^Qc-&1J7ixc<uw@Lw~JkGc7|bZKl$jB~?_|1`Qgdw2~2CF>y{
zAD<zEv9mPZ&9A3XG#H($D_C1o9yOQlOA?MV@oP0mu@@LC{b_2N2L?kTpY~+7t=rqF
zI@+m#`EyF8CqPX4i=RyXol$JSYAX~}yi)^@e=fSq2kn4?Me69+rB9WN*|#6G%XsOW
zm}RMzTFRNa-g_*C>^nBbr;l-?CYi=j^tEK|Ham2s=d*$%*Y%E<4cx|S59bDVLf64f
zCPZHvr`5Fa=JY8M8!P*?YO$QW;@f@nTeX?P1Gk&Axi-5je$R^o?w;}S5hh$V$03nJ
zm$!K=;u3T>-$VU<JMSXc1cW0ouaQn!MNaP|N@D*Pq?<YrM0OFt?U0?2K#lln_COFE
z=syc~=r8TN@+1u|T!@x`>?00Lx_3EALxO?0cmMwDM=(+ksUQcm)}}Z=MtSV@;<<mO
zt7qCdA%BC-MobV2<r9SPyLTn^FrZvyx5l8o01tdiQ`4BJm<G=qLpQhO^~3nm^0mtl
z4or@HKAy$=L$#P0@wab%4_Q9k))I0y{|ywz_88Go$2T|6H!J+Me{k@Obtw6@^&6eM
z!58LZJJ0Qrp9P50=^=Dm=W}9>%O@dAkCc#4*(cT?{p*IQMW*{(DKdKUdzx6@$H6%p
zg|x+U_bBsnkw5$0|LoVF%xiF)MmUSBG;HbDZMevNs_yoleC>_aF)@2_4MhE&oSXpo
z%I|$4sG*=GAdprbiZ=YfB-+lbiD=L<15E_0MZXMzu);jN`t#y~f+sP3erc*b-sVm#
zpKxxjNp7~0E8me_v-^koRl=68h5w90d^_=<{^@*x#+5pJn`cmSyB2hrihN+tJd!6J
zXKp1Se$!1GyqklZk*la6Cz6r$o=J8_s<UgT0LM=!!)n8hXCdsvq$xu8q|c0lM_NSP
ze6fb9t4>QvtK9yF6fAZL%Gas;N(yQ}1>*#?rt;eT)3P<ym7O?Enc14*A)b#HeYPc#
zk@Y@oCwwiJ1%c?39(Xb1d%*`2Raf_WiC7iIFao`lSzLnn$k@h7Yt(1RPgat;yL;N&
zsch$cHew=>_(RPs!xS26dD47&XxvS|v)g<Eegpk0N^>>`Jk+4J+U1}l42>3BzHfA~
ztcCea4-*2`HfHLbmd|^~ne65>^_xW6J&zB(-FM?OeBQ@+e0xle<E}O{sBb&|#xd^r
zP&=?)-T~OQu=b6P8u=*1e=op0fphjKFzXu8)=Iu*Cx?4ZVGQbZ=o=hN2GL$%Y&X%j
z2yaB7ArRzk?L4!^4X#*waIOi;iwny4G$PNP+UUeC*XjLS9tltT_y+^8!zUn|s-@W^
zD<~{f^s+BAvxT&!MU{j4&Zl3R^;0dD9Th8aD;h+ouXBi5%xiMmpR1V_{4P6y)ovRV
zxt*I|TG$D8vJ;%+JUrQSFQorhU_S4IiIDVvs#R_L9|eiNV*m3?>X)`t=MhT_7!rM}
z{YiR{eW!iSTmw=iWlW-v@o2<?F8n*-SAL}dT%FHjF)%K!BkV_VghR}TSp3ESbqx`4
z4C7xV@tce}l185Ka<6t+4FXF5k-+B^RTa1L2cKxIOri4`_!^N?Ff9F?O9hANZz}Zg
z0xD~$H23HS70++x4eCHsUGw9I)y@{KkrCFQ_StUvd|F(j+X1*Xp<!UnmYJn~Xd;U4
zK0Bl7X+0#_u>UO|g?zd*<?z$2RlQdX<KbS~{{M3UI!Y(MLn;vj&!^e(dCp#2J4+KO
z+id<&-r?WsGit!VM4K>d?3n2C?e`BHl!HvebBkMK<mqZl0)A&jE34E^*3-W4VGmDm
z6{r~1-pFVwR(#E$80ke;iwsyZFIKM|-QRoU*k9E@Hd%WvK%%EfD=R4IS-<_@sZ%8&
z(J>W$H`Mr_x>F3;ydXmWv3Gpvb93;ND~torsp1VDUUC}jn9y*Tb-mKqJvu(RmB0J^
z+5X`14)5N;RobUfr>!+~=z#FYevlp>Ha|i+IoSCti({VOY<6ufhrSRo6wd186!m=3
zg@@SA#}Dzj%=X6edl848rJ*TL-=2d^RWi|(wCrCg-PdC?*B>Uo6iB@2UtL{gq<vLu
zcd0F)7r<%n7c{(q1?oJ*9y(fLSAL<m7o2K|UMJ4h>^#nvv*dJS*9;`*fb+hOmnLu`
zu+|?P8uH`#$FsvLDJgv)!n@(7bwdYwe&!0!`|!Vokuncg2cBa;w=}gZQ!R^V*XKXj
zZ%>hh$iN@U4C9I_7G`G0o@*F4OwS49Bjct@oMVS&udAv~e*PWml+pRh6cZlqGP{9g
zAR11NwpXfyvE$0uvO;?Mu&rJAssEb#_eoXFgS&`qa*CC~#Lt#*=;9b4Tm(Hcho1Ho
z)U1V!!SXp%Zjoo~X>3M|G%r*JK22M)XXlvs`+QMUjF>KQ{)5JwIyP4IHRAQ|Ks8G$
zkARBC$mM#--Dli?6k`q^2cun5GBWN~^=}5?qtek8Ly~D=dEdT)=#NrOO6r%YfT?Z{
zeEQP2J3#}%Qd{TWl<4B%EG(U^C@L(R8sBBT5POL>kji6!+T@}xf}NP|bFJRHZy>U0
zTO*^R4>wk%*s|7h|D&7n$B%acy-pkyx4xDd>r&sy-I&}zld@Kx&B%Wl-nxJzxv~)}
zf$sJ+R(yG3VSx`rknRIGE<pkj%t3ge(4|CzIgA5Uc&5~-j?cU{)psDaA>L+0R9mVC
zv@WR9W2gT5S@tYCkh~mMAwFR+Yj~}A{()A5{Zjik-vR~7!b0jhzBLIr;;ik4%!!i9
z>B*TwPyO4xr5--Itnut_S?Ca%=%Bw-h+)YVL~`cEflyVaSF823leWGAC(cqD`ufw;
zsHm7TIc04*2z*1KEH5Z1xc$Rw^xKuLcX>8EOXTuT<Gcn@+9anNOI~sl>ZR-Gc}iY=
zGD1|lyd#(LLF%vG&S_%%o~kNmhbP~1bB&9+$NwnCLWpWcbh3hodT@Yn=I}a{A(A_0
z(o#!h`j8syf8h>^c*dowX6uO_1gYB1Pv<IUmDR-1A!gP*0>fKKh%A#<Rx!rp7GpLb
zJX}Um_}<r&BhkdNGE&6oIQwfH6ab^BwR~u{rQ=^HD$64fZv^i9F8(G)_|BRBOq@MC
zKUaCBQQ_U+6E~(D?tz-l=d(<_Hy~hve_C(L5rOuid|poi7^jBJWBaj~%NWrL!|X2*
zcRmS@Ff&7|7)~AG-dg`x7Ew|@%b_>3wig}=r_7?~{+xS6S((1J_d@m~1N0Xhu_9#X
zL8g@_sJWS!UXUt`e7e11|N1~e(!wGcBF&KkQc)-|-eKt$3z4IlNV8~F=z_xB2Oo3M
zetjxL1<7a9)cwxyL~%)zV2KO`ldt8w(cZJWcf(tEC+Z6FYlL-VknxXO`8@Ss<2dvi
zZ<S5Y%*@yyR#6?b`hI5qY@n{Ca{Iz60-9rYW94^BoixI9A(^)|N6P>HR9g3oAD)*h
z$dByu?R7|kdz3uP56f6epe|>|$G=T-0EM%$p?7h9zMx+v1qFY})Y2WcWCj|VFj8~i
zhJgz~It07DNcSyezI|S@Pz1TOa3|aKu})=B44n##3L@j8;6?qAE+sAG=+qHL)NNco
z!dyo1j05_bNt@~%$1=)4f7naZ=pTRnGE-A?ci+c=yNAhcA3aj8^$SX#RAQh$S=_5M
z8!IC<d39$<4pbBvG7~5&0)9Dm$mcVQl1N!uS=-jSKN^-$^54Loo9Kv-hnn1Sdjm_e
z?Jx68(8DK4h{asJ@=4r8(_feMB7V1zP4-5+F*Y{v*sN!<2=RA44f!)W$Nvm-Ic-H0
z&s#kCCN-$j#0jTe6e+tDE&VI!`UUh`1{x~%`ECDz>8Ge)F1J0Ti>9A~FZLYj<s?Za
zM6bb+T9VZmT($^23Ypp(<?+!wh{Qgh{glIz)#5R$sU7#@u{6Re>YBR7AG6#3$UrJO
z^a&vvo)|(_8G?FhGMKXB`mgQ5bSmWb<b@FncLMw34d+bNalqP3Yb%DQ$K4nmT&a#~
zjrs*2Q?{H=J-X_$Uou*hIu4gczq)Ye!tU)uCFEwEb&SgVNPyTXug1W}u9$?Sn4Zbu
z{qh$BEJ8-0r3cM(8pj4TJ-Oq_Vr=<~nWUs7@RhUDnw39n&J29Pn4+Mh)jIpq%B22-
zAc@Un>EOVeoV19$5LF{AmXi3Xm^ht){PwOrhboLK^AkpAxZOl!|8`yH0c$d!XuCGg
z>KDqlf`T6(2;C&T`p<;@y_kLR&+C(EF&^^i?PsXOm#?(m6!ULnXTK~i%xuu>9gwhQ
zcb>E&GS3eu+IHU&7LeQTajMGG)U38v=QsL&_MzY#oMIYmKlpvt3qSwK1>h>gBo&>;
zH?U-{f+m_Wl(-%WBm5HfA36p+eP{^K>!yZpg@njVlo+YNV6*LCIm3#pca*>IhW1Cj
zmb|R&Pw#;3(&*L8@_?V6of70|zE8;lQ1MyJIWogm!}NrnvZ)nwE-%*28GG2SA*;kk
zoBEE)s_=RonEtGfz*1&I^-4?U^G4h5+uMuq$H%vtD>-SKlsLHz4<qJ&ytLiCdR)`c
za4-^YI;v*RZPqdNL#@>IX6rg2bzIdZXLv0y6*Z$Eydvh`0mTX&<QmRf$H#~ByCrvP
ziF@BmXgPBx#>CY9a?XRtEFe{5HL3qUYYp4|EjzXn_yv&Jbautb!qnNt8rKU8W*07k
ziqh#%tqaap^E@S0JCGhO1(5ywL_3a@uqtD1ZzM$E!zTp=v<7T9yOXt&lnSEXRQP|7
zjg6V*V${>~<w>Ot&CLxAcCNy;H&Uux=9(rOFu`U>skc`P*t7=({)?G{G#4SrAbA%(
zPcp(8xvZ@!f9nR$f%UECD=ny0&y>GLWnTUAsxQ(!RW&g<?dm*jSvR20WA^_$gj&a?
zUgmY5%+S!gx=b_7NQIE!8TO;<%FeG4Tl3{!`Au`i^}NYdQ{Sj*xVC13iymF?(Y)fz
zC@H~Nf2*2jXNA@ryt0BVO%*xs=R%?vKV#kGhJO#yPLTuus%>8%l*awnEbz-0OoZnZ
zW^6)CR76ZvO47i!JM>D!WKsad)`qAY9z6Lp?~}mlbAIp#zc$>}zKj{pln1`4&jO&;
znDUw6Z2k%ulU>at_M%`ge`6dj)yXs%UQ<-k`sB^V#%kxbBuDn;o!{rOrd1x#tM8!w
z8cRjaE$<tS6Ek<uEl50<4iLgkoc~t%O^zqf&yNEkiNl`MH%N@};`ZQp`eRFGh>Epj
zVr7Xz(Ihtv`md;8MjZ2+<h5ob6QvE9yh0!tbviw(*~_u;HlU@7lK=fmyY@rEmm%A>
zb_OHAAan$e_r@s$)8zHQUBdyKJvQ$=Ki-hDV~xf5HPyaKPFl7b2}gR;n9#86jipO@
z1)*)<)-b1UDu7?q)MzhZ8Z`EYxk)HfkmnwWBD}r%{!Q<CuCz#@4u+>UK|t{^*-$|@
zMkmrwxKs{@kNJc>)9NSefBpUgqVu6k`IKL}eHvxv=SiXIe4^mD%357fHbGB9BJeyi
zzaS$w^Si3f{NEkBTPTQ|e>5a19sGhKYZ4~rH77h=`5L?iH@o&(OvQEeEy-y}KCJm!
z*UWpIUu<`FQU9HwMMk_?KR85vFV-6v6vDXrw^ih@2-1MEvw`*L`ObqpnktMeGLV>*
zuNgHQ(D{0gTXaNU3ZKzWu06K3>k7^g_y!U2I;wy_kz!s%Kga1u#%4(1wmFepb94eR
z2lEo~-VXFztN9Zl*6i9%&oHshgN%=JJH6hRYr)e)aMEqajA(0X?`)@n2^dyfkgq<G
z*;xJRjYS;|Vy4E6k(KvQL;qmoSR=QOS=7Izd6KNo_m^QN6LRS%h3V;G8DZhil}nd<
zf&WI2&it~Entw6)`hWh|vX~uH0YSKIY-~tx;U$a}@4rg3GB<9Tonzu9`&L+ox7{Lc
z8-Yy1<aw=4mMVbxBR9GO<DuWsULdp(04=4VdLa-jn<x{_i11D+0)FE^!8s77bEy9{
z;vDHE*`|lMx_`(f>l?=Y{@%Rdwa@P;@_$Dt*1FxFnD|%|JlEuLOTjfZc+2ouPiULn
zYr_!Xx=;I(a!12)Ykbw)2BEv+sIVjTW#cs;2@Mm?a*q?Lw)R}r$qA5ju|rr}CIFX9
zYdnif@a$dU%ZwCxN3ycwp&brg!hC=i$|Vjoyposdp9uJ!yu;4L8Fx+idwMqU@FHgg
zGKs%PgYaVJi<2MJWQ7ry^RqKbX`CmB_~Kd<uff{eJ+bw0=0|o+24q5xAr%$m2w~4K
z!B<o`0{BS*^9QbIctjuA(uO%&4Vn8H#$IU|UGI;?gQ_Gh4ja0Tw;6i}&Xy`~*sJ?A
zEIEN<Oj}fhf<q~VBSDIurQh_R;fWa3DUV;<%bRa2&OfO0b5ihfvd%95Ryumu+cRMF
zI+5CylN7v^U;jH`J?+RfmTW-|Pa8<6K3a_3gfk89KJ{M}K}`b0Z<ex%fv*!Evqz7h
z9Hd4^Yisx{TumbW!UF{pE@-AzoPML;QeIJUd_sE3%8Rk@b;fh7QQsq|saa8(-EY}3
zx)<05+a098OL>gt#tG7?JV@J$r!YT{ec$DTA<omYA+YnIkb;7Pvcj{ByoAz@4iOl)
zo!F*1FS7x0PfTylEGzuc$jd_KvETFThMY1a_Lw)I-4F{a-qB6iCL%7zG*5nWV?$$l
z%F|Ql;;iYPH*ZM$#q(=c7P{niV^PmAFoMH;jtNekl9Y55td03K6(u#~Of;;F%_U7u
z0+NaqHUBzmL#*H6H5Sy?Hf1nvwNmwk$8_sgzNwa_{?~{cZpXK{eyGA;R8up11QA`>
zUuAP8({#%#Q3v19&;&JHwX8rEU&ylK@6U&$on4I3vLW{Vo%eDYkGQ$H6)0{rOtM_h
zHwZPcFlygV7))oeyGlTMjkJHbA4<SyIl^}7-_Zd-<2IlAxfx_<7VHMUHKyWSvMy{^
zw+)j?f`?{j7}4<WDfi2w7YGPzz2rGe=5<g%^Hu(GxYFmjvKsDb(vZ%>p3d*lm%ji&
zze%a8E6+{CCM#Yl5g~qhbaL_c{4)j;HyEmUczD7+J=YyOroW!TxVkZD93K`ZVl2Z|
zWmBCxzGO0Sn#?Wxcli-jXXfH_2+=U3=>`o#RzqxKbtQSKba7#!nS~AN)9|-#H^7hH
z__vl&Y>G+(TRKu^^SX8Nwyyh#@!4HvMa4Tm17y;fzIj?}3ut(FfGbV>0P~=w|0b);
z(PNdOTI6Q8+``P5WR;)#rREl!5dp)!;~zBGHF|ryuc+zLxs3E7G;D)gr<r1h*O8$A
z<S1ZGTui)ZvLy7EZben)Y~yaD`!e&tToMMF;fc}FiBaJ^mXhyrcYV*VID6yiAshKo
z5DkeQv4y$C)8l}94+4Ihz6to$3I1fvE+;5Rk8>)2SV1J-kXTeeF5g4QRRirh*}>z-
zaKmzP$YJU6tiDRC*@?aQD+s_y*^71C_VhfjJgaxzhu4mvu2D5HuJ)&$uXQ^pHo+=N
zVLpD1%3lnlgKy6}7_a5B`^nY_;P}rh^9u{DZSnwU9$?$83y@XT9v&LX|4}fIIG)SR
z_fBMT(mLUdH;Q=JYnzikmEya&DE^4g%x;Su-$I=M8AaGGUu*}s6sOnP<}{|N!@#Mf
zMaMw#XAsE&t?|QKa*){5(W3C|8vCHVNZ)%^(^^_umy3NaIcZt4i1gdHIvnN(%%dA2
zdHW6ULPyd~S||T<=PNsb-Hb!qx#U^r`!~X<_~s=p)r7_bTj(4FI{-|XFxG%5ZF%Re
z4HPh02hQcT-d3TkWATfKbKcH{pE~oRZ`ki6PAy*4{YziGmzaKgmAKR?Q#*516JGTZ
zww@wix?q1`#a`5CH|@|G7bay1x#ff!Z2t}26S%9b(9>$Y(gNo}A{&4M+`q-nd=r$H
zU(6o)a2Q|QNk!gHMGm{f&xd^__3lr$v`QW2X2$&(hwbE;blAUF0ubZac{t3Kyj|=?
zY_H}`ucJ?db;!!fJ7QhEbXJ-r@{Aap5Pv?sA7Q;_l<K_L_iiBQk|a+7@w(HpF7z00
zmF2~wdo2ssC&Hw5$2~dd=~;6DUIeTyu7J=Y;P${&pHZ0Uw?%*bYi`yVbVWiLCSbDB
zI;BJ{fP~=Knj4!M!#<~YQ2QD78=O0n9|Rw{vlzn%jus^(3{RW!nZPB4QIDs@g-TsA
zjo0pIZy#_>9??(brI8sU8I60@WK}$ko<sc4pD!xM^+x<LTdGrHxAuU#rKqA%TBhl|
zv#2z%7y8tK0W3kU?tX+H#NVTRAsmV!%1u@%(+9^Mp<HL}<pc6j&%qB0Tqo*XmpkNE
z%OPD59q<lO?pJp)ygMquK`IfzbJBX8X939^-hDmG5*NyeuheG?1Swnncgm(cp`0X*
zM0H;Qf+(+`w9*h4T@0vIqI8<BRpkL>1M}ZjBf(m6x{=|;LU8DEPifOq=#D3n`_nHv
zumCYq>i@=WV^uYI*Y``sXl=A8_bMCgKm5gW?4Jq1v%9tm*dTyZcYL7^w(^M>8yihq
zyC8i3yR(*)&B1^yt#7Vp=bBhgmbx6hkB;tqw#DmuRd>mYuD1YxWZRNmv(e3DY3$jK
zi5B4xEoUDq>q$#hkF$w&c=^)Ln(p}moGT_C_>G#5lhCo|pYszrqy32Y*b><KYQe}v
zcU$bpa#0+NrBL08(~?z6Canjz)J9Hrs<@qd{tKy5VN4AQy*&AON#h)n?&F*8gFp%o
zcfEE0?N_QmCG+rluY2Jq6QNcABuif7BNXGIb;N6PzkL9Gd3cC3yYW8@Ld&iP3$Ph{
z*Vy)`w(L>pSYRDJX$~-4e$#36YTTyfJTi|?lcS!Zq^7z^+^luyqa~5>*SOdi<4oSK
zn9^^vzPAkW>mE8IBLEklt(=WxIdu-7Ft*TAly?WR5GLc`V5f<J)t|vJ8=E}+SA$?M
zlTlQhSMG9J1?OjA7j=cXf|4AUvnlb4urZgLpzM2Bz|evA4*eLJMfIdR5VQ9miOrWg
zgcCVVEV;(~+6vs@vNOEBX&pcgOB1GeFCMr%Uru-_PeqwmHn~+0O$E!`V^L!_&dDXw
z6CK+|?~N2XM8tdH&C0m0NUQbC`{`Nw&E67M7Yug&Jw{ogNpOogo$Fgi0*FSWsLq?+
z;$xzmd)l3fCcnU#-sFw{MWhAXcTKzpy|3BMj;D3W_h98wZX$ITfjTIfLXJY6|4v1;
zcj<!SRr-4avQ}u<dAv^+C!hT1*(-y?Pf1O_b;>z*&43^3UsrgnE^*|ucZqseV~12x
z$z>uF?euwBI1t40zZK3jPNKUrFfeqlg%Fw}bW0~b{L0Wu%FZm?IMFKBuwIq$iSfCP
zJG)xO?>pa6S-55EGR^;m&%X=VRx&!7*LOeWH_C>P;3h~?KJL*ipBQzfU9sylWKPyd
ziu~Et(!UX-CMY=5F(s}h*OV$Ai}pga_Br=lSpN>uaP%=R6*MPMKx9011`^mfS?KUf
z&Fwy15*7<xx(p^+Uog;6kRP>*h9v~{6o#28n^02{!G0WzU-TL)NZ}$2rS|IRWUNs@
zTs-RSOWjNUnHLcR{VGA9^+8ojI$cthndw)<kt^LI`0;M8fV`Z%a3nnfj!g&}##~-T
zf;JogpWj^T-`7yMej5r1e3+VYch5WV9voOMGwpmyN!gQ_h^lw;-OQ2eu7<Fkpn{gw
z^;B4phwOU}X6veo3L-9NE)phOhBSENICdsWM}H^vI`9j760y7cH5aS6ubk$u)x0Ra
zBdeMYisBnI!tZx-$SIi1nuPKs^soZ(Y){Sptu95!#FUg)sh8X1a(!%F0z4CDD7ctx
zm|%W4@D?jSH&ann^CO{*)aLsxu(gI-HCwFxmcQH38<XOfad{&VKezi&lE3Xse%Ue5
zGDe&dC)~AX1zy%{!>oX>Ak*mhe~`7B+>M7cQS`5ufA`0f<JJ@BiO<gt4iC4ds@%F;
z+)x|RGsC%8zr=-Gyh;P88x}Ai9swu=4-YRdQTnkr@!ZUeyqXfPMV*=3JBnQCl+=gE
z*S;=V^;Id~QN~1j=x#RXgTxB-i1~TF)*@4r{~KLrcBM|~rCBR0yMDIam&EQJ->p3V
z$eBURCnh15go5ur)sBV0{AUEpvk?L0Y!p|Xw6`y*X=uN`iTiXW`rit-ht{V$)OeK*
zKW5(#!?U3QVY~BB?m;vO^^jhzEw9hVZbN$}LtV(E7}pwv)F4*&>WJ<uRc1>W7k2Dq
z3i_E}Bpcgq(N8Uft7yr|*Ea0SKyqAIC*mt8C#SqM=&18vb`K8E!orP%9$EpU;4=A6
zEllRg`SB<9l8V9;@K330DgUR-Jnd?j&%#enk6TpV!^3y&)%e=^t(KLP&7cNk-9x_8
z^F@8rXj}wPDstMRJ|CKZ$Hbjv23pd-?pO<bo}<YuZkVgya7l>^3d)!1(7cxX62RBs
z!OPz4D`3*q(_>p!v@@qy-#=F8(G^}5*RgcDLy^NTEbO;zXf$DY0w9&E>-)@1QaBRS
zIR?LkR~u-SsaJZQN8L9~SX^DLw`-^1rUIsxMgBLo>WU%LH<;LM_b*Q^tLOE4=X*u_
z567?ulsjy9$FQ*LHxv4LTOmP`$W@c#d0<P6Ag}|T7uTm*SXod!=UVw@8m<R5`G&<E
z3fz37TWv{xO+oEDCAA<H%iz92pOd!8!i}VZVDZO^I)I^PMeiNdbkLk*6i+93<2_|{
z-M)lBP*M3KRFIO6jz+fyisD?oQx4=)mzK)O3ldm(5j}rohDWR8E4VJVU-wp7+5eCk
zysqevR94DhEs`XiCaW|5wJ7&v=Tc`l;XBeBEEWB)#jCrQB{kVG{SVOfNYI~E!kYgm
zG@U(I5*;1A<Bd6-J4+p5J>C5VuM>L2ALG=n<fBvGG-<0Y07`Rmdb~kIJbGXFj^6H$
z72BAFxKGT&BUhS7kA(5s3F8MiT{sW+2Asaj#U%_TvwMh0Yx5BWjLWLD8?Ke|MMi}a
z@*cqb1CAU-+`xSvk)(pHj=x)!`;4toX+le?nLpFt{WC3R_{}O1<-NCmKugnOfRStv
za59YR_+g-j)9VUt_2&eBkv3N=r*552ePCKh=aR?hJ!T)#JZyCAlN#@W%*?W!%22_T
zU56oAW%)I!N@~Nh>2?-ipuz52ND3>w^U{6Z%E~G>KK`!nvIO=(G7AVf{Q5VzY>|F;
zEwtXbY6uCI*mq3{3@mwX()>D=M{n_r3D%1M*S!;qEO$`T07ek~%5JCQi&h{4ju%z^
zU6B28_tH5tCp|m6Ej2m*Nr?E_xof0T>sy`pZo{7L9#gYI#`X4INX8-NFY}pAirwDc
z2EE@KJgewqb4%qa%l(h{9ZZ)0EiWPR;C`|0XWn<{Gfp@-aJ>3Q!1DCyIE|-4>mo~1
zQxfGJN<VAzz~EpwN<I7RfX`qzJbD}Ev?)te&jO}Pokr?d>A+gW^*y<hlafAv0(*OV
z8&8oAZE^nhqJrh+Wx1I6T~v(FjGKs^$G%PIn9S!_W_dk@8^OWifFe=6yFY+R_gS*Z
zeI4aO{_yZn%t?e32cE_Af8KIp1q~_m7w-DiIBT1Y2VSb0np3~0#8*~Tl;!^#VqV9A
z_i}SgxFv84V0qZr-Jdmr3m<GkT}(>pJaV;vaZw&7x%+Nvz|HjDqJ}jN8N=t+1-O+!
zpBE)X|D&j+uCC5rq!T(9A0EE@pN(l{rO|RMIjNk8`eQF;#LxAwcSTOVAbA;Ce*f`Z
znW6J&Y{0pda`TU>kETtfeNMYR`NWOT0AeWp{_M}aYRfmr$4?U^e8jpDQaBHJUyYi*
zxd7;?o*;~6^aefeb|Y(gYO0R7`F5?Fh2IBGeWomb{u&3X)45iFe44d+U}Lp4=c1}D
z-O)B-I_9VVo0%VM)<V5b1C4j?d>BGse{9+J>hN&xK7L5#5|4C7L~#oLGO)0q*Jfb&
z;o)MlbY%<e%vjpoPQh;P(@o{9R5yCrk57KHl+%w}jg9@r(bKQ6DnCw;YaL8!{>Ynk
zdHe=1)Zs?><zFU7uYp9i55B9?^Aqowyn~XTdJw3y|6V_Wakx|3fe}&SO<`ZZIc+zD
z4BWx6q5<PmQy&xN#1wiLVbW6mdimMA`Y>Qvb59Qrm%=Bxiy*X;0=Mp!mAFfAm4B;s
zSJbBs=i~GgM`+#xF&Xfq4iEP$kJBPQKsJmW9~}*ewLXs{@E~DfU9D*|JuxAiH)62#
z+}eI<K6BP=RQUSz#^^P!CEMi*FXMotF12w)|Bl<Yv|qaV1JNm8N7h0^rBbYd#pGop
zmLJRfKNmo?VZgY)E~)3C0a@>|M6PsNAdZGN7WF(q4^v+vk{5K}u=NCwErD?E<;4xV
z>)$}ti$Jc7F%%trMHK!51Iz56_$DBX)D$#Kc2{cv;xnc>NlQ!h_A~lgr#=ktG&sR)
zqaL0bt3>D2_;9F=9kL7?yX|^y$_^0_@_8TZRsHd|qL#alY<oRP<VlEDF=w8W`EkZ^
zq#GtMbMs9E;t{lsm1$%r<**h-VDUYCMg58apGALdo-yD_O*YR@O7@4OEncf83w{+b
z*++wsKd6N<g-(x?zf@ovU#{_!Q`ArKfgGV`LgxljPmwOWTZI;R{@hQ=3l!Fmr?cbG
zjOX=40Ry&bNnBSRqGiMd1=_=Vuua<OtgVQTi1?Zo!A40NXO!4CILMio9nqaaNd$RX
zwTmY#$<JN5wc?d`4mljlE#S?xE0SEqh|10(_hD5)c^0r#X2C#1GRONMpxzsjDCg9P
z-7{T(F4(4AjJ%`rxEm>?;^7C~MO;MwW$e8>8P(O4&mR`jit`ff5~cL^wzl@laWFF4
z_)Jis`F;Lehe9|r?4`T@8?B+CA#n5)SWGAgYNCXokvtDdacbcmjEVUA(q8Ri{p*m-
z<N3kr%>(U}!EFT$TMo_JkxVDz0*+tsz5iLp#h@XdNgQ@xU1<uIdyKe|Moui*ie^MG
zX-9cyUNWS#K1obZKNL`PbdDGJ&iz`6>W1e+fN^8gC}QhCLHLXj0}CqVnahyO_dE6B
z>AZy)E`Mz~)Nt4NmWDWgU1vQ&!0;*GsBux}Q-JjL{d=@!c0cQlsp<E35HN3$1>jql
z6}A?C?Id~lhiXO0uXO><?dhFWlW+BpgF=FwiX&b%CSYiZqoI9$_%>jfiG;=|uayK=
z6%Kh}Z0yy-?cNlB_hGiU`u*gD*e4`Bw=S3M<?DOb$LoPi2pAa6P4ktNm6ZVeP*fBT
z3@StJ(df8u$yR@PATOz?s3VhCS;=r9@ufG%0-*Y=gpM!`-`U%z1h6{<H0w9<@{P^h
z_JqS*F*zhDu60MQT7^fz0`a?dUoNq6m3-s8=KGflUqt)v@pN3AGUQqsn*LO~Qy;Hn
z)3$2L$2a%x$x2{Ii1$_qHD*9-D4w1m`Ey8<ZEQ$StzYwxdS)-6?v5>T4P@$X-+p|?
z{`UqiH6Q?)vnBHUJQnj`ZL$_P7{Xr~cfU2C_&nCuH#$DL-T565S7ie|%-#JPQY<wz
z@a3tz`R0&JN+f!_TFB8c(S~Fdq}~~JG;SSXW5t5t@}>A4_#}@EM31Qs72heb9w1tN
z8kY}hH$0#0meHOcl2ubmjh#gol-JVwbljODO?gF9HwD~ZgV^sm1=<?ZqEYX$jNVba
zJBK#N<(J+w<P=p)Hc|s|sb0{f|0pUH!J;1Hv|4Ox>^FAVrg!8bqvIA8lOu{%P*cnK
zP92VY-iQ^Ol%{8xA8z&IyO!05=h#4>3k-aC7!jCHO%T94?V6B*_#Bn2@^|U-U@fRT
z#ef>9B^^%R9`uQfo@+z1_PQ7Svm{t|a1c~fRYANY3j2=gdsKEd^E`PLuAmAJVQ+zq
zj1Srx8Xk7t;bmpP=#;H!tRm(jKy4*u<7H(Ym66yjF?X6ct8?1^*rLBR-O=4hK}DC9
zQKUXUgo8(9chpu3=iaCEPg2m(!kSTs<f1XFHV1TOfP3~O8L(Zu<xyGb>Df6rguv@*
z9+;a;sb)sgXSWsqpijV7z)Q#=UJ>&XPTTEEwjVnEyn9w?@E<Q9>>~^04Ey=rgXZIG
z<l|NJYvxg%Dpvt3tJyj!pm}p292{^EPW2i!FW4;dUq>FIx>nn!O?qE$L!moztibML
zovu86P1=3?U;Sw#rzK+u>Re)Ah(kb%DCI%Oy4|ggs_NQSH`$zwHdu#4>-iMF@b=3m
ziJY%ryWD@|ltn>A>OfylScuHOgClqZ!tuEeA_orv5C~f8gt+AEGr4DoM`AR<11PRm
zhlYvGq0UUoitqq!vp=i8dTDtX1U`Qt6g?y?B%Q@VU0Wa{4W!t)T*1t1{q-N(_0OC3
z=1comZiLeLvic`2Pw;w;0A@x(ahD;Dj7u5}*`muD0`|1ER->*_Ee|)l4GjA|h31YD
z*4libBGdI28kU|dz$*Ywy_{uCT<*ogXubcCi?B^gvRhvl5QY<zWdVB{e%(JKBe*6P
zD6%aJUu$#PhxGz{esW)v)RmR}rrw~s(U4bm9uUjB3q&{x@xWs3+O4S{+&PVT;u*4C
zPgeUMfaZ|G+uNI5-cwK>_~k;x^S!Gw*zqyg;2NR+nu?npb@=RfOoYY^7p}*uG{HFm
z;kT@ueO{_+emP~ec?VdMeIj)!QGW91x~o7PD^`hGOC91oXu#gB-hTfG2rV=;-LPQR
znzgn+b%PPKyh7FeHd01Q>l%3SmTdU#Z|>h~)TyxD`E^|{)Jj&t$_l>{TweC;>fHaD
z+t5<eGpF+h{L+O?gQ+_AnAXj()hvU6Q;`4Mp2_h|TPDS1x8*X_XSm!oGCj*P3VA`m
z-N49k^a;q&DQY99ggg&m$ICC+!bikkHs5=R%g>)4yL_NxGq?N#^N#O%!{nVicObIZ
z65HcuCNTxi;<c~~&^>@4bYcBzZvR+AA6yTX5A$F-u)4f3X>4&#`5Ncl#OzUxPn%tM
zM1-z`!)CD23cHPmeMT$7_v$0*C1KTRgD2weNc>tq{&%!cQ+E+%efh!>fdC*U($IDd
zCbKx$n2fsHFu*&eSL&$e<n%OBIEC4c;H~Hq5Vm*b%V$S6?cCnwgvI5?#QdvHugwR4
zbZSg{exc~J#yua`Q+EM9CXt&S!_pnTy*VQjA(Chfbq(l);#Cy&!vgN#e`lfms$AV@
zcxsB4n%efX>2$u}{MHEVp<m&Kx*BeGv(6~Igts(d1QUmO@BVuunF0lD?=qr>Pk$cq
zQ4jXP;UKX7)_(zJRkzf|FP}dGk+qk<=M5p6qLR{&@82&R)_cco(7SKv?<GHdN09>r
zZvxSY&+l?lhjBR`e2PqnnUc<_s;)+EL<N@K@9_~U3mbmB6Mgp8tF#`IQh<^4CO;y<
zHWHPE`6b#OEQjZxM?bp9^TD_Hdyz=nQ+hT!7P|T;`=VZ6t$BlTK-`*A$0H{lJ}%_W
z%aR4iokmQIM(hz+Xj)X<uUX5Oh!}YV1@@VV_wUp)Aguy+bQO)73L5cgkp+3*q%vfS
z>gRv__;H<<s-ZIvpilGaS0B~xg7xifMcUA$gp?lrQi#rlPgSn=J*D9)X^IOwBsMNm
zNk!+Df>K&RNK0LOF)YtT_I*M`1llviXNj*Cc<$nSO7lZCh6^*WzdOZ2`Ey^7nD5c!
zhbXM0@Rd=*&lYF96?=@!0l_KRJ{_xKBRBW*5}F8j&#PqxLX4v?-|5(`vyQO070qYH
z?<rwX;F{6E@{_X`eEunnma%}&0R!y4J54p&75O#yD&>ro*<!^*LN7adI}I$~R20^&
z++<PpOZ6A5&oJ9A?VW9<G<-cJhPD(A2G|V<%~V*jac>!rKSKuUQ@7y_-T3TMt^Df2
z!5$1i&mW$3<mUbxbtIQpQv=^LDNs5H8&-U1DgrM}Pr(DH)B3=cxPJenDfK?HkL|`j
zg9GmhCWQp~$HvF2s}+bZ&dl=*o{P1~q-WKRyQ#Ug|9<yG_Vdy2*a?6^X1MQ%ig@5-
zl2TLW7v(;-Jn5fK^Qx0FGFn|(e(~is6F5{A*qCy-?K~IHl*}6bIPU}8I2N4DPTe6O
zJLwZJS#56vfE2j*x!>5K2h0xsD{{*aXRIkAg55(br+Gleb@DCa7aFTl&cB$7Cq?7j
zJNK#Sio1FlxANa&!+J8SpdfKT*U!(JJo+5QPaKpESUr~9s{j=O9)2<qZGwUydbsm?
z&v)BitvkGVh75+7KyKcjkDB3>9~?;K6_S1XbGu`k?@!tjfL-b%UhB5X3VA)UD+P)+
z@wua8@M6^OiQ;*b<AA`R;4>Z%^yep$9&(}m{Qv>&er#1&*8m;xRKX}*A0S0R{ST1Q
z*>g?leoc0V$ftW7J8mMa-?+M}ijp$NnQ1IL*XT`kMc8Xx|GPEro7NH*7WU>3qPyv7
zt|JptxMp1QA$1{EoY&!oSzI?j57^nj9bHs<>oeC_T3~iDD%QlP<M11imjej#01Gag
zqP8>}K&6eyKJEF~;24381>~u{!+nS@GgxoDd3bhqMjy|(VvqExTT;CL!i2@;N<p_A
z${;3Z;*I~U=ios5_h_Iov8;s@xcuLN4dnqvAGT$JB)}^Y6B9?f5TNxVIoV{lJmxak
zFdR_*G3T4iY$UF#s+vBI<ol<?N)^%m@&*;?Q4IXVw?XAEZYy5iVyME(!&!Q6+y)eC
z!%KA*y0-a)EX{no>h)9lGi#AJ{WmXf&}Zu0e~k~OkC3hV5ZuA+386|0NRe*$mRcJ;
zh7ZME;xZo0np{(YcWpvKN=mTzF_}z;JNDvdXFBUk|8*ZE`KZN^73s~@b|9?kwdrYC
zjTQehPr>`&8h3kFw^WUuEhO+(2Kh03p-*S&AAuWXIlF4Vu?FOo%LXIchnG5SZTkjs
z^eK7S*(C$Um+5iKKKJ|k`>9e0$oM-U+zxsU=@YeAt{a!;9iucrlC;f@CTN>SLPbUW
z@b=-8+oQbu=cszgKb&Cac7tOIwx4iNvzz;M|92UD-V9w4zm<|Ag!TGu;5-#@LLfPG
z%|}jdL1I<<$W;xRpdByKvolwly42fQ(wuA~P20wz++6#%;Tsv1L7Prc3c^-OgS}XE
zxb0SLV%yo$+?>y4<9}$n%BZT^Zo5&i2$hypl#ng~Nok~!Zl$D=lm<mWKuV=c8l<~J
zLb|)Vk?y?f{l>U|3>@Ql&OT49HP?&`rE42)%=zHi@RrMr$UHOp2pB@`q#LfGGzU_0
z{NWKo7wA{B)49=;$b#kjceA-$b~j2<&f<>Rg|zJKu~Eyp+>;va0oJ!(-+wh^iSm1B
zgx9^*FY_IbaytYrBJ{lp;djhS#}qT!NA{`R!?3Q&1wraIwhy`4xN%rpm*=U6Uw^L?
z-4>^2^>&>)G`AW-MfKrznG-PF`;XIm?M`_Sb8_(VENJB^&O)L?De+Y?eT^PSbn3od
zd={FC$XmeqE|1ai_zJDsotT6m0gjIT$$r>9(>=Vsu&t_&NmY~;wtAYqR_`0Lj{~<6
zE~@JXK+%I-9{oD^jh<%UtATdaYO12gyVc}D9iAg^ZCa_$o22b;%-8hm#6+e4Y*3M+
zM`_{sQovV#m^=_2)?p+a)43!Dkng8jOGR5-iB)kw%S>Bl=EzQP#F~j{{Eh=sg0TuG
z4}ovISDO8wlBxMDwbp|N_O=^=Ok2jgb-TuWd_n!)yC<(-?_Ry1J5_A$qBAx&PZ?O7
zWa9a7V4z_A{4BxOHzoJCo<&Kaz;i1glC_s`u%$=kY8EARJiqm$!hFalHDEbU&Gf2D
z7d#P+hPWqS=pjeQdG=9Y<0G_&Fh0MCD13e$laFqHr_)0~;oT-}%@RA6jN(E17uG`D
zM=03dp6A{>U#@Q&8x~Wkq!ssSTopeS5Y$zT5li>|;%;tgX_@XYwf%D2nIZ4j*MR49
zA0P8@Li_{k?da_iIcCCdso}#g@-JRBiJf~A`i!z!RMef|oE~9*^bJHbTT!L02JY|y
z-Hi@5aGXxjN+*uvdF<^n{)Bj}GeBho^kVLYCSwrbYReXknZtm%?)?hPlw$3;JxM(X
zNSxdL4R@s9=i0vs95}cK+S;##8nQjey*Q$;MR6l8Xs%o;`fsx!pb`vMRl4H&h^hDB
z5;aUxG-do!&}>64bMp4?wpW|pB~9MNu?4#{6<gh}kXMhtd~qu-R>LMeF98U&&A_3J
zuE&5CbX(Lel`v%ir7y&a)rb-lj+tw1KU)VSl2j`v$cQg4xVAo8JK@K40U5jwSLbSO
zU}ZTRUxjIn>bCsjV;2`XW?qWIiqsLbQE0a+1kk+??RPuZci$$gPM>La0~w#u(8lU=
ziv|v8UHOd1Ow{zx)Kqk$4*6E3?6UKY`MTRa2Di4J>1gdGi=Q2(EDi!b$=JDdQ?Y?;
zcg@z$Rth~i!hW_cjqN6XiQZTeUI#wrtJ1AUWamxDPv+cBVq`LQ?l)cT)MfT@HP6iq
zH4sp?9G1DU^hViH#E&j$YuzImeU7}%CqyMl>=@Jir$y+T<+Sy0G-mujEqW?E@IMuF
z1@Le-s+gl-$YL*deG${Uayz4i1#{m>bkJU#adkHrW<-FTPIvYK%*mJ@Ti$vqwIJ=$
zyO7TTRM}~pc)h$y54q{Dtr_rWmIYWQONvX4-mmaGUC~mo-a~%kx{MO+0_(W{`vVU~
zHr{!*2)V^P$vrFB(ro{`E&gl!$ln&IiBcL0AKG8F4!}IC$aZgieH}Kevhq)6e{OL5
z>}@&T{Xx*hz|bHhguc=<oj+XIW1?lKeCP*jh>{AZES4JAKb`}!H(j|utNlSuSU>rf
zXFdCXAM|EFPESqPAs;fZx|;WWY%o_G?=c^bs*ddNX)EMPU?Rrm=9Sjx(^a#~L~)GN
zbov%0HpAaCGwOl8O5Lr~*}*h8G|U=f(ter^Sn9^sF=+ZHCNiP)f<rx9v*`1=T~2Q8
zALsobC<&4LiJAn7XL~y#u4Hz-AuCG+l`~E|lvF`&2C&c>bqn8u2%LYz;kfoGl*8|3
zN!$rYmd_g!+Vpauxu>Ux!l-O(GT8sL{Vj8G3A=gr$Ty;QHxTHn#N&D>y01j>%@l6N
z2~34vc6Cn|6c=M$Zs&t(sX}AlH4q_}#=E4c_J@b|37BS12p=^Y6~$q3ODAwYLTV|g
zW@@(Mpo%pr5|1RbC{WK!O`J~O*%oTBbehb|{=K}g5XWj=4ZES$c;0ncNJy`p-=*q!
z?%^lFM3|(N#tbv)a1kIH6XT3JwGT<oU{E+#YyTAM8r|$Z@y3g^AJpvZ?A^MRmB8xa
ztb3&^*Yej=4m|!+ZM(r-q^pa;jKjQnVYhbrBg*XP>j=1eklBr6?o~Ig_ueF2&a79O
z*a+|5uG=B^0|aYp!jRRR=1oHk*|C>UgNuUDG+VcCa6DH`+>e&bFIxEy$=cQ?&ym=w
z)n$iE4n(WY6$F331F$6}zV>YRV&ABb+K<q$qXI^HM|T$_|2+wm<&0rD?7f;o<iR>@
z>Qbj^CF989H?W8+hv-;5$^e`p{_l(BA24Qz`9PL($9tLW)rT-uvz$D-Ai)z(X9Kx9
zXRhXdsJ`9$uYpWZ=d51K44d@=44$*WVey`suxzqhN>b8}tU#YxeM^-|;%k7Rf#FQ;
z1|C5Iv=bOUtCj`oR-^Zq_eS&XM&wOF%S!d&PO=yw@B*~8vhQad%tvI^hExA6Q$p$o
zsrMxuu@S9E8yml)1_!aKQs27{hKucwMW3J{ciKJbo*Vo6WN57JjjSq32#)%EbQsr*
z{?nmzquy(<g`l>QQZ;pV_H~C{XOArR^!pGcp5#k|L8mgSG9u$Z5t4Vce>eAk$YuCv
zfL(Q0?ZM&rp-AxLnd?2jYL%J2!7opRs)bg6nq3@b%}nkS-TUJ7CES#$+Spk#MEn7g
zfw7(_v2xoam`*cG)X_V_BqkxoRVz|8vdbuIvY&ePuAtN<{fzlny6ZIB><h=ODFvHG
zl*d^d6YDuLT4S7UuWRC{G|vXo;b~WC*ZucE?^^Fx-6tzh+l2egYQjA5+l{E1BCDB0
z|MTd2g96~AjazD&8D!i?_<Jtm`8^B^24iK@T_#;x-?9QTY%Sl6u9A+5ar*GLi-Pst
zPze&geuEiH9u^jbExnW5s8hSHqobo>TJQh27(_{yCF%)a3f0p0`jw~6@v<RB=$-#C
zobGfcK>NzIdHI+-L;<J@*eKb#pYY|2x5I?RUt8P;qoRa5*Zqs74Y0K*Ba2sOlLp=u
z9yad9$y&#f*%n=Rlx~&1s-~3fK;_W+eV!lB4UV%20J#R583>&{>yJz9?G0G3f4*KQ
z8&gBSd-p~?FhhVihs++H0JBRI;hHzLVh8(hU_;m1)GSOSrv0|LD#kg3>o+&QaBgLV
zXU(pD?TeIyW&G#r*!{js<5wU_fpXu*e3J9--CfYrspigDnwfdHUaKT9{s8)6=2{fc
z#eQDVfFzE`W$zBFZ^g`^GXaTzZWM?%Kx}z%-G{Rh9TU^sBy!VB=;N3LBO}<CHINg(
zFMk`_JYIQr3*mCTz+k$6A90+jfs7UuaN(Xmv*?ivRbc1VRuOXshAIgCZLOM87TJAo
z%FgR06}^|;2&GTT&uIGR)AGeS`Q?5A5Gl9$9J7B{-a=H=EE*Pefe#Oypg)Y@hJJ-2
zXvt{_O%Bwdvv$2u2xdNVls=aizKoFDvaP6pH!=(`&S82++=O}MIsFqytvcM3cIut5
zo;x_n0vZa}Pd3LY4jtw2%qm<9dTt)qNANSb^1iJ59KCM(DbLm5d=BHH<p~W<ePiRk
zR2d;gAExWlQ;I$Gx#SV8o?LB<-58y@`LP8F{q!T;oPTz`{|$Y$vG#U%@5$3lF~J7q
ztBojIVK}(M7f0sc?~b=`)Z;jRi2f*OP$_JRG2VUS_@rxJd}?N?Oh`+*xJ>Y36~(uv
z>Ysaya0&a+P2EfCq;ep8Y)%>ttQey5>L?XwRc*rHGe2zN@s)?TsPexANuJakkQTuj
zgj1eq{r9M~Zrup5A78(I4H%W2n%d5P_~1YkN<HSW3t;{ryn$$-y&cykr}dv5OTf^A
zYw~Dw@^?<^+{*RQN!vd1>hUZcEktcd#r0?bYC1B|$9SpWu$4qyxsf!1dF}ZN93HTE
z##L2ExDqPV)RZs~m~#CLFtgP81L`rLz*<K<T_gO_R{W8dh>(Qa@<0MyW<b=w<-}L(
z))P>*2sfz_TRPlw$Xl2j!ymh2hYQ+svaCkeKI{OZ1<3s-2*pxM4#84$Tj;F2^D#vC
zDyZ9!FYV#XL^xNDHwFeiU=RX$&*A(gG~DK#xqJ?ey6|_HZql5_l}FOBW+R?sotg0B
zHvJDE(01H>S&5fSS3!M9HdU_$NFa!}mWob6OW$)=T8B~jn+xxJXq=Zo6mVJl6abPS
z7l%!d^nMSYn_V5$*O8C~jq=Nc+FsdIGKXSL!X#os66Y0naWf#yD=Vo~=+ua&54@L-
z=^vhUKaJ*TFFxtNwt&v_Tk#^GvbFAcF)<!5@sl$vqPp~UcXyLshyflVYKZ6c2NmJS
zho?=;!zLR<$S<uddKUpd7kSZlRR_u9SgPQXk1$}`+eyAj@_Mkc_;Li`y~~s@>0Dv?
zO{bmzQD(S*?(`4wqs4O@-Y9%QNc<@2ec~)X)sKo5PsO(%{%bh@eKddg5WjNe@IMuZ
z8$QhwMxdwK4=h~4f-wEv{#5$ggP8k)Vl#6eXt)j<LFGdohXpvOe!5o3bN@eYH?4Vz
z4)|T$o2+?CEg{|FBfr}dsx4xKo#VyWKZlD4yH22GgOLSnLQ*bgY^iBZ=M+D9@&>FG
z|M9_do^pBr7xkVAQy_2qGgWv0Z?u;Rj7-!|pD?PNC7Z9rd%@a0;xWI^GfsdX!NtY>
z&g7Mg@LF&oUywLtj*~c8#2Ime?mV`m+00GFJ8ImH?lSD+nLDDE@?u8PBQ@+w(*B_`
zc$(f>eQ#){te|-J$5cU<z-w#Qr#a8FX(*DZ&`B#2&dM_~4Wjo9kR4eMQ@rx1KYF}z
z+fK&c8XFJNJ1$39)!k4Vw4}=2D8>!H&vR{8koo@QI$5dKzA$}3l-u)HtMK}Rybr~p
z-XCEzub~la%LyO5Y#XMQnNd=APtV3G89q))%X0B{uftE8ea|OCv{x1;*+qw}r0clE
zEWfxErZ;aKBA$(N?(7juoVXgbW6WZfO<bo6r<mTvUff24n!=m<?ep({%W5YUHLcm!
z?UA!Zb^5pe1`QEd1A)4Fd1-0aY1x-P24Fj$V-kNkiE)#%)BI$&*RL}hQK)|smethM
z0EezQ3yi^Zy9he}pj12KeEsel{`2Ys*N5!vYF9X#K(rz5?)TE;)Hy*fgO)Cc#1aDm
z@moePoNQ?SWwnA82IzNFvXBS3*eVL@FZ53GQL4gu-FDaAM#Y_+)TII_ftCJ}Z_V}c
zf~jH(oW^0X(6c10TrW>$+U6p!lKW_Bvl{qCx?jXZMTc8o9@V4<5GX+;uG<Off0_yx
zCxIW|zvs^sG>^72KV~+#4!v%fRu&hnaQnwMCEYuPN%6~*l;*46$BzUwk)zt9dS!Zv
zNg1A>AKyy)s`j>50*j9sLve;&yd{+IZ@KgK@q#)d!T#Siw{@qF+2S=b-6eswwTy^b
zbWEai8@-3UU2!$<lqSz8;8PLdZs;uE(zZIYW5EKJyR@{dkiF+JmYNv7MK;c5@1VlZ
zuW#qc%_DN|_^2>&_5b#V-6MnGu4qiz_S)#QjIP{@zQ);=a7&-PrkNCz+(&Kge_%@J
zbU5no@86VS!0`ioQ*3uyUqi;pdwqLq)_c25_hy&YavOf#^fc3auVg5adl&Uaf+&i@
zG5Nt}pZ%X|n5i+Y?)c40sp6#@>6w>XckP9hmOuP|E<ndhFi&ZzRmvNRzHym}M|K)D
zjiid+b4zPM>=nEada-A>Jy{TUar1d5xlrSlql=S#=Fip0a*vpeo`jL)KXdtCNzLzl
zq)5K&nA%VMBPQiCt>^HjC$VbxY?#!#2g=;XWIs%dHH#dEbeTtW4hu&seoU{QC0#hA
z=a)OeuI#cby29K!jv{rSe$zI~dI$0YVBuwFn((`pW4!lbvaLvM7r53}R;wgpj$~`W
znkBY%ctC{Di0!woO)@z{9MBm5(xYP9-bHg`PWU#xw!%#F@@QKT^7yP<nPiLe1XITS
zvUFXXqow1*4ZK8KWaBAAC9j!hT<MKRT6YJC?Ujj%99GB+DgNpe&|>@rQ<@E+-nFkr
zjE`vq6Y)D**K3>xKu}|D@0|*?>*o?3biYUG!vqov0LTnz{^svT$CQ`P%!V0!A;hPq
zsZPItT)@ovf|Knz4Gjjj4J*s@u_AF!wAaciDv7pJg`%~$&ML<>qEXChV=3F)+qXqq
zpar+5i}=8`xuwm0M9j^xWzAcbq9k+Jv2rlyb$)EO9UT_OW7=>qct0WJ$o@DtE|!#Z
zQfoXTEAua(f3VXI4a-_s;wSfYUpaDt)E^QlVI3C|B1yO@Ewyii@@CRpPa@ND(<*H$
z1ju}g3YlZLUH$~-e}!CU+Q^v<+;&Ku+!hfT8r%VIYAfSk)Kz%I#~CTbonw6{EWg=_
z>=%Z$N315pGilk;(jZ3@Uaiz=>Ymek%D#g8QXTEQlT{Q<x1{lWf<5ZHCLZp;mmBCC
zj^=V`8k_Ctjg%RH08o@!?RFQt=<mK@t%HDpb{tpnMAN8*!=;#-NSr*V!&~`v_)uOw
z&X_9M{vbWMa5Y_n`!lvn4~<^u7bN*^F4h|r#8cL~Ud>ZhdOKKx*M3$6(|5_#%xwW9
z!<7_610de}`gq^02gCHKD*LU`Y5SAW%=V7Ca8j3#TYO6D%FT26s`T0!(?1iE^L)#u
z<N0iC_jPKVR#xBY<|UpT?sj&zyGNYWS`dpnaYli^Fzg)gRC@Mhs?u(i)4+T(dHc(U
zrVtdq%8!CVCs$byXlaf{6&iJFo%dfd)rlW7PH}$>(e{Yhy8@F{Q`6sB;ir(ZU`7}4
zoXyR0@?HYJQ|!EGc$>lFsNE<wOnc6TqNbGW%~%x&4hg=lLGafuH-Vbi&cQ~){qRUp
z!zVY$q5eCX8oi396x)R*yI$8k*<S)YTORulz<;@eWMpM^MQ(h&T>b|hbttc9rsI*2
z2-T3#C`w%)_-tgnt~N|(f8dBg9<$QoBWKP|H(og!u;*1ekIr7*JUhCDkl<H2H<s>H
z_4!YQ!s;^&aHTP@iIDxLJ<cUM85r*O<A$`!3o<0Anp>SAd)GSNglfgl$Mb3Szj6`;
z>ltY#p2t(VHpe}QjVFRk!P?R_m6eVC>y_&p2@M|zH@PN4k^GfrL1eVgUYO|zyg8M4
z>vfBrjwID?4S!dwr<LhBr33OEpB(9H%lvM?@eP%rr1_H^f1a(xa4XV<nxN)of{l~v
zahECXPz@6nO8t|bsVNL@xl(dYT65M%a=-D0_Czn(te9q*N73EnR$1)CHSYI~Tam?E
zGi6asgqXale<(t+n?0;ZH5%OgyOo36{5twv_A&ON(vp~Aiyn$?Y0pC}Z(-{GWw96A
zKKLe~FiLzuhYz;=f;qD%;AfOKZM;3q!OL3d-$Kk=?;?NA|DV^CO7@g1HLGcVxq`8$
z)U$r=iQ$FallJ!CJfRAi<opi*CQ*VL8TJ$8Qs3}=G+)_Veeeh`Sz2IxVoU<XQhZWT
zajEsbGR7@7N6+Kse=b`hV)n)>6u9^F$#~)1so@I?v{Ac{?YAhV#sv<8EUCgg1O)F}
ztQBi_^xkA1ylKTm)E0SK`Wv#%R<OV+(mHq^9gGwmH{oOk6aEt7GObwB8s5qTZkNR3
z?AV?>%qr5mh}jMNwYBwwBfD)LN9nyRPW7>u;B<dkQv)hK2xlrUwkF{*jdXoY-zm+|
zDXm-Winq%b#>8_JM8uKdG9vpZYD}N1OZsw0fi`T<%<Spb^li_j0N1WA>CP_cWk$7o
zNIS{B-;ULRW)VD)yD>iO^k*KA6}wcJCj#rAR`)Gk_s|2$D<={xYFvM@V8S<VRPf0j
z@#(GB=ktVGO{YQ_Q$C(MX#%(P<^X7nC{w3(_~JCmzH7@l#{c{g^WNMbf6$8mmRH|U
zpGLXC&|yrOI2Lh=rzKcSzoPXGoL`JWha<ppD&Jq^B1leNRW$})34$Y`*dGnWl+nw|
zG7BQPW55yxtHFF{369UR#=1LoBCcO4u`zBfj1T+$`6b{Hs->b5luNw2K~X2Zgm-Ad
z;l@Xb&IYMr*?{E3A7HkgxB-D(qTX};e={p9&(!Co#$MvbRjs?R78ez?cXcpDEbM5E
zjTsoND5<Cj;^Z@PGSJdeVf$g_m`h2vfB&xAQ7b&3ul7ED*qnook8fXJgS)cTs7UGM
zCRp5!ly*ZPtJ8dkfBW)FtQX2bae;WqvpOegGfYr|k_7`-AeKbax*~P&2QGeqy<NVl
zqGDHjyX+;mtbvt!WX!PA2(MeX+VPO9!f^iH1!bs2aVfji_Hejn=aP-oH;h9#eb-%(
zUvFqYk%-8vDNXiXjs0zZ1Ei?xB-Xjp0Nqns;7!BMe?zlad?0;bEvLfszC{JTG&VX-
z4)pv;+VyqQgnm}73GTkN1BfF{a&LN9;7|%vV3qAR0(VgDk)5lB^zl5M8O#HF28Pb{
z^)F1{YaiwHo3Jn5tIf|)<Clj)EO(UjRcyq}bPfZ4fQ;jn(BE0L$e1^+i_t{2)oAqi
z0VQQ6?lF7QqrLS_qdWQ<m97kS_Ql!sooq$6;A)-C(4|}BlwO`w%Y8HV`ddNhr+Z%D
zD^FV+@rDk)h@(3QD8;R*Dk`#!rVnnqO)<ue7ez|j#c96rcTm`|kY|5`VXKIPgOglX
zSh&LUJqD93JFOj-6(ox5<zCp|bmJ4$h9H+fDb4g~qN{y^*FuSR+EVU$S(@sRI4ypO
z&&5J;+iHvrGOSE2Rn(hZ!Xuv%*MWkc5QF2`HXcG7;lx%@uzs^=23KTEmk77-eR2W&
z{4&bWudI4OlBuxKC?5Jf-xg?f6=o&0;~_S)w}0WnL%cZnud7Pw)F(_b6%W&=&i`I=
zj-pxpTWsBG`s;QDBH~9XyN(oz`GH0H%6QwS0TJyXJTU(4YE~B&U0N#=IgJvRks0dy
zBTIuz$?KxfxN_4oJPm^Ce~fwl5`)}G*uN6p6I?unYMqexG{#245!3h%GLXO~7ZRS{
z`@P;%j1s2f;RuL2*ehZ6!u7*c^zQpQD)CI@Mxq-JL5O4mzsnOj`=}_Qm}tU?urL9S
z-924m+e);V$KKVSlj<7!2Jd<^b_cOo2m6U6d7qXkhJ<;K?Kednlk;ZJjZiG1rF8#7
z686zWC?3X%3HvPVS$^y@%}C7*!d-}x#slU+>}2(Q@_|96myWzT&#+&<EWLqflQJCm
z%RhtXRXQ9_Evx%A$PeevddMwIqN%8;G32eK(P#Et%*8#SWn%p0wARn0QCZOO;p5rH
zDDzv`Nu;H?qat)t`*{rdSjcvr?hG;7cs%#$#PB{<fXe=t6{ZI~!IO`wVXc*52n+<P
z(cePAV*I&WFY-zf_DMOu^M;NcJ@PB#hShW$&9X7<AXs@WYo@Ea11o*E)y7i%gB}@s
z1^syoHcmx_wy@fPK7DLUhKy=x-as_ufBNR;mO^Qmpc@nUv)Xn$G0-fWb3Go}uh|t+
z%zWXzaoNHSI?5A@FZ5K3aKRt%9}A6ELi9OU-HC`$u!Nqlk;_h}R^LWo$qs6%;2QxL
z+)PJ9FJxw?_$r~xzPRMp(eV}IpTT1!Bs|p5;Fj`(w$6#4$H-e#y)UaNvZRv+dp^^W
z?w9yA-ah%#{n0`^BOLT;?Nx)vfaL?rTwM=4PHdw;Gmid`OW*z8T7rM8I&~&gf($a4
z_o7JLInTfiLgIQO!FGYW&(f~9LqFn8UhY;tJ3U{uhT6bPb{W35B5fGeBK3{>>I~6E
zHCE&?-S3u!1d5>Lao?D)wkU{g>Ds#<zSZq_;WvY8_>1P}d}MY%Srn6qPhb#R@L_oU
zSyXZF&SOONe3GbNKuC`ms4m}l<vLyDz|FXZ7lrgnPGKQ5sO=@6RuoL-_OM0X6fkOQ
z1}CB;(%1u^^^bm^Sm@1338$f;NX-gcaC*stXJ}F|+BejclKJt${$R*`VOfypg5Nk4
zOJ5=|pa?Da<_C5Wk4vxygOByeDJko2A)jV71?G1gkq`N0qDfiF$!LhZy%2megOI8L
z-3+7mm!wOd@0X*4^8dzaI<CB0$oYfv=Js{#oK%|^?9I1-PZyIrH_z^yh2H%ZSVHMq
zhd_xB#6_e&I&k9QhJu#X4Ft;XUxuH%jGo}3-tr2fWek%XU}M|(B6KS*$cT+K%G}&w
zhhxbNDtEY+GmHe3gcfVSROweUR!I)*8L4MlmYJ^}11i7zm0HvC=jk_bsvo^gshNpX
z8!~9{V>SBlHYGvjv;h-TUKHH!4F4+ZoC+2EWn=f!V&>bPJx*vtLoe*ObH><-=ew!~
z6;~Lsh$riyv?$)9yU(LZmy+kNA^b_Tu<o<&yfMtORBqG%UK#=H<*Yl3#eJRR*)Hig
zB*=5GZmdDJ&>7Qt+S~2vl7TlK2~~Xse3Mm5%6ra`DK%_{RHmk87Pv3<XUA%B60E?U
zhIlOSK2G#H8yddHVKEcv4jCg`<F)VV^R4gcs8@fmG&16d={k#5?M>EPi79&uQL`uj
z@6j0_4}*MO&IE#VgPjl0rp_2XA+zJdCI>rzpfLohBH)U)?`0ne7zXJ@!%^roMnP_H
zhx6m|&l*8-{duyR^$<u@poZZmA}Q4q%U>bvDl`v$0OBc`3UM>3X=%$A?dL2{eOx6w
z)QYT$B~3GDTq8<lgBNhV1!7PKl9!kd-8vx6u8EN%P9KK<YwTshWkik`@JEUZl6}CZ
zIg?r0U~>~1W$y-BEMJJ)V&)|q86LA9jchBFr$@cvzTq<j@aXF{t4;H4pZkyUof}oq
zFd}Y55)y-y?m6Y58%vqppUJ2dChx~?5Ag8Bdbe^YDGi@PNF)i*{AEk=&_<bL>YJ9~
z90(JM2oHtsWngG{FN)7%^5@G`Lqo?K?=~~uL6-sDh{?hr5mNeZQdLtE1G#yA7{y}f
zYM}Hjn)ZF)=a&O3Cy6z2K0f2nsz^LbQru9P_7d-qD2U(Eo~S6z%*lbwt-@mF1lD(*
zi&6Z>M0-bjy?w)lh0Ik>JbS$bOySO88Bo8KGh#Vu^qbk#^n@;)QHv!1sP6|VvN||z
zVejcNcv-%s&9mz`G+o2TUFmWyLWlCIteey#n;{VM@CO(iob}(iJU=QeWd2{x*8bmT
z#;OwL8G-UJ1NYwgfVdnsgoUMY1uO-){j<n{AQe-+NsFz{=L0Ke&sCV<?GcaW-H+)S
z>>pfZ`|xw~>`TE@up9%vYg?DCt+i}VtB~ZQsVax!(&G9BYY1u*Y|rRpv6f(UTj^dM
zoqoYh)!qBb<9`o>UgSW=*CCg5qKaQinux`2);D-9H8u5hTb}0Fgmer$N8PX1_Wo=|
z7tVX$lLa*!-|;qZ;_H}uN2cm*JK%V2*2+!T-gFae0SpB3%q<;`3>8M+F6ImB6-4%|
z)Ty60|0*_1$AYTpug@7+wX}*oK<v!7D_sk=FGt0Ze$iT64vPuzzYCq2I3P#x66|T~
z+8CWaZ0=Cq;P@v$nNLyRcxZD?<jxR~I={{4G9HWewRvNt#TJrG=U|RO)YbF;=yNk2
zzpM1XaQ|e1lT6(ceM81}j@-NDc<!7}*<u=vZvPt#KR>nAa9;j-H~GhR-KYQ5VK+V$
zzU|)K!$g+__QG*I`g`S>w!My8C7m<FfA{)+eED;;`Yx)FkxJ=&l7sQecw!t~pqK>>
z2MEjd&%#d8(1yeD7g~ZMp%kF-h+x>W&Myme!L`ye=Yl==X#YZ^q2{vV33h<34Z3-b
z``eXEVyKe=>;7AkQ{<}ktS&VR`iiCH`OUFpv%QUvn0TB-H~x(nrW>Dj#VXOFc>2^j
z_OTfg`?t)^oeOg%=<>L)PeLw@yBqI6^Bk*b{}7fU-Muq@($hNP<x~vyvDI5j6k+Vu
zl*)P^gVqf#3>5?j{J^{OTObsuZHyG6gp`VwckD{bOhO_=Xwf{O$=*ZNDpJTgBIR`e
z2Q7#|<F9cBUdG_L(oKXl%v9aOEx`Ol(czxH7=~a$f?WJvew+Et_?zMevLx~?FD||(
z{z@eJCvKvDc|w_*CwY6-(dYDGdTC(1nLHzn_^EUZAKH^#nZL~LQt}jNhy_De{8Uvd
zBLyDvdSIA4;~4Ys(170<jvos#6rA7|m!nPh-S)$l-Ma>?>J*`l2dn<wr^agP>d|fD
z_I>hJe+f2+dNp|*$b0(-33=-w2BR%@_5#N^dC4Hz^fY(TO+)?cFt`kPTX^|V+^^P1
zhe*l(bfu3x!O4l`rEX^xE$U3rEwaB`R@D4_k*>V7FmOEH^_pzC{XK~4TMg6cc*WD~
z+fQ|0-l3v@8ud?EpN}H+k|~&p?a3{41eFA(*NMz4OrjZi{QWC}?Ate-S3JLAjia|)
zrtsF}oxf}E<Hw4mB+-F)b#IkdOGSiiyn__G!Qp-q-g@HTmh{@|41-}AKfWmGt%0l-
zZ@LkvA9uka2w5FTkN9rFA;Q-nQ5*J57ri$}>h`_Um}T$8S~89C4{F)wn;kx6-;aFa
z&?N7#_`D*))Yx2{z`uMRDiN|3f3R_u`fZFrPwl5GVXX~2;@N_OAOU~ocO=bvIcr2D
zi6+lpf_I^tpg1M=YeVHBIPM*XULf-FO2vjP@qP<piC$q1!y$c->$)E8TQxCRv(r=U
zyMuslQlA9t<Jl^IPyAG9x5Z9s)}M%6St5b<Gdc)CP@{J^`?(8{t+2}bm|?Wm;0HKQ
z8}AORghi;TsYjCXlxqrW=d7Mvo$r-9QrgYrcAaC936DRxJUgg#VG?6$_S8D=irr?y
z_1CF$Dr@rKpEpjEB~euB)XsNAL$oR!_V-Q~%ARi3IG#$<;}a7<+M4{hD%wv6YhtZ4
z6dIVRa=gUH0kAnikmh>2j)_2D8K%x7m4YqB7VspQ3M%f2hsQzWlWrU*dn?thcH;-y
z#YwU>yg05ePX^z?zCZaQ0>8p>?-LZS^+v<z&s!CG?a?Bah=CO*lzMREW7MjJ*cX~p
z3$52CQ8WF@Zq?3M<h=D(YqX~nOYYtKnw9>7>Ao}NG#)j&2|8jvXN@h=nK<|EqF$5+
zX2Bg2DBtk5_Q{3YipZ2lzU{UkatE`j$TUiKG_K0faOub2z(wVe&V9{W2mzDs{k!~g
z<HOJSpD4hmUESE4_MqW^68fq>fyX)@pa#|4DPSVHOQ}xXt@!}xopY%yu%xqwt^DL9
zn%*+|9$6gk*$LQOVEhLlo-$~I?SaETmpVum8y#=gTc@9a))f!$@6hZI*&Izir!x%X
z#xW{N{ofZIO9|qd+<qd3djE`&2U1O5w+}=~y1r#D^q}gBnc8GUoIg!C66ljeK0rhw
z5-H$9^y74DWSfEK$k@!}85QM7k^Q||yD-`GC_-Y6VtZ3uMhg<2(vNY>7CP_VJw!G!
zC)pv&a!_7o4vtc|&L*{D0ct$a_WDXW9KO7_prELjoS%P3{kF&y)|kiYNr$SHK@a2u
z-QOH>0o-r3LwGj4M3FjhD&LSAF)&(^j^gs^$s^@5ua+j2hDSxXDwnXq>FD_}FMdi~
zF7{Er<0oOd3Fu>>T*>pD5_|*|XAlX;18a5wp)AG<I8O!>4e3VUZF5^w5hBZ7(X`}6
z4+5GQ_m5w?ZnE!ng~x|3t?=Bg_zQ8E&_Mlof3ovCYg@pv{9{X6oC_AM!)Z{<Tzs9Q
z)4s#A-%}x<+{$ikCBLCTU+&T>HEmotd5<tVH5kuSuo-pl%SpP&h5guoNkQ5TIw@N|
z?SB(<(1)z26rg!~qeo=l@C@nIjOZq-A$8wg%Ho7Ct7-JZ%Cr|;y04%0H?xUze>5g=
zpzd!P?M)URNlc~&A`(mlLuPThY}jEl_Gc);c@1mfzut!3Da}gE2tOdx=CGk{80+Y6
zk8ZKG=~RV&)m_Dvq|tXS4(23$dads$FCG`qZd25~Ou)K+z)UFZ>%bTQCXbvi!Y#zI
z9IIh1qP(jA?rn~bllkpM)@)*O>RqIIeSOdW9{vyY&dbvI|8hT(ezJ7=R%KqDcMl1W
zkogp)aW0p=Wo=1y^@=o~vw83Rls9W#OdBP*h`ejpMFRzHn~?j*cY5Y`5FixNRNkH^
zSY;m%7Qs8sO_N}w+|Ihva2BINBBP>d21;S>Jvb8vUA6dRRbIod;72cU5vU0ADx=#j
zPzcO1KXo6S{M|d-VUf|*C0J@+P!4EHG799Y>z~2Vp5T6B6;#+A!69fL`c(ugMq5KA
zJVN(ZYNo>Efx8^b1F!m}@cNM{hj$7E6;ZD>G^miT&3fv+_|mqTJI(<-*bU$NS)(3M
z6hID@&tELUy3b6k`!x^xjR%dqA6Q!iwnGkqb$Q=YpSR3}p%MUR1x-#Q%AG~2zg)G8
z_<3>GK12)!PO{H^4$c>1r|zM}m6g9~$pXfzYjv!8$O!VZJPw|rkY0x<<06qUMAW<<
ziHW&B*x7UYXeBqAWw8;&i48(Mc4kqg^Jc2Obh)ZupaUfO$MPU)H0%kv<)malZ<>j7
z)Pcx|X66JXLET#RNMfB6IWjx^bGCqsYOSL@rq7wRF!%GeXqJjnK+F7bxfPnHtIns9
zus?0hRyqWA_PzNlA!Df}hSO9v8fK+8JR8<2Xg>*2_fo#`;}$K6)ZGh@=~u7wQlcfr
z<9HPCXKSONOJ#@%kvvJo`+SeeuPtlA@aKD!S9F*>#%G;Yo$RQdJ=GT)3Jj0rx^vDx
z(d5E-!1%j}b1fWC60z~TZn?Q>{%4^*d7#M<z0C0J2y#0iL;&q%&DXUm*1tFuWXtzR
zKv&Ie;q#Y!efcY`-P+zdyiT)|VQ%+#rjkEv5vP}6dS3=KYEGx8=J<;4_w;@<10pi{
z{Y_bE_mKk!m=LMRKEuvbx;MV`Irgvrx1<;By^VSWFJ8D7C*56Nx8Yy;G`V<<<8B)r
zEda2J$c*d7PfL?`R0DL^;!oI88ffPZ#ZaUJ1AqO&&df}l&d`-L^lY9}QX({_V|*wD
z*J&1bfAHGLUmLw7LJN_2_EoKzne$R}OU3=UYNGttVc?Zsyv&0JipDWgEE$F^E!4c?
z0#7ND<46`a2e5yHprI66%Et}@z(CHQ_JkkMtn!Z=ydCL}^rigNu+LBKvW~TOw^2Uy
zlc)vXr-Z7VC37vMPPcQ$p+D0{Q2$~Vq@)yMm_QEJ5FRex(_m+KPYZ?`CCbXYDcq<C
zfRExBk39dpji;n?JBd7+(~qV<zI{L8@kVfPLO*aCI%NU{R+a#Q4m*F@GuDQ}_ah=<
zrZP<1RY*_sean}X*EDaxDA9(ERGerVZ)_5h;`nxV3MIC-2+XeZLsIuq=`gGB;<kV*
z(TjJt&L+nCp0IDCrz*qFx;P8;s)3iGpx{}+!B!XsEz;n!kP#V|YeV<&ct>NL2!$St
z8f4f@>q<R?`8jwoBl7C%$_fgJT<FDrMZ}taC5(+#7~4YBc_0Sne1d!EC`?_nmoQqd
zEk2k|{m#!HZ2}jOR*03^M_;OQym;^o`}O5+Mm|a$Dd}`tbp(Spz!mKW6(V{M110%B
zoQT9}l?(cbywU1}U{Z8cd#6)xM4mcZc!ol6feZ?;(PS3_+Ax4Ky56!6jL7^MZ2H=!
z6xw_i8056d^U>#1t&WOAkQ#*($u&kOYlZ_P#;}OrL==t}ddcPD`H%NL+(vjb%>euj
zyKxVzVL^F9P5r^$FWiYc{b9Y<$-Ym6(sapt;3Lpna>!z(ce}rnLPrmH^Jbc-xX|Tv
zJy4Be&PK{WqY1)tTaAbRwzJal+<uY$>apXQ{och(SIM4N?x?_4hwrSHWMgwxUQG=k
zkhd4g<|L$>9+VOI7mvP*B+=@;SV+h9@<3C=5EJ$8%L}G<9a6i6V28pG7WIbUYS)A`
z6^^$}PiuR32UBmad~o$gawG&jU&Eb&yHPV5_yB}aG10ElQZq#5JspjtPEn%iSPTp@
z@5OzA|HbghrMiY!sIdVY0%4FE^3qw6*iFR4wt^D6Ea>(Da7G?e9_pKzM3sc0c?quA
zNFCiG1u$1sABXfJR)Yup{tOIPAS$oD@6T1!+FN9GsZTqKFDgD)mZfZQa4cYJid%Kp
zF=x}<uDGPqe(S?#)7b6cdEs2&%&aU5-Bvf}LvxOjG>4t3Dm(Bmb%lv%hpKRwj7+!A
z9jCp6vEsIc5XJkfeqUOHem$jzt{9TJs_aZ5=NT$VU)ejx0eckBqk2|;d=)vijKf<T
zYAI@o`<f9Dkq32EL1d?Ji)_vHj>*8dr?Wf$#;0bb=k)ZI1f-hk>P}k?cqCY=YDz!D
z<4sKUQfB3sm)Bi+3@E!&|DOwRF%8TE>Ti!Af82rupDvt6%5Acmba#Hq?9(exd`tGb
z#6+u58uGYa*ujULR*e|a^Fw7|Y&E8#p#9t4uSFa7hQ<Jw7`;|QW6JL<D+>p103`-@
zPN1S0d;!!gDoo#XEHEc0TC%9I8#rWh`gXU|>xLh1R6|+xVCO{R(<@boo$jB^EU5`k
z6LaWZ@e-GNSz>9wC(VpF9<7QQcV<l;vksits}j6d#^O0_OmA$Ft*R)!TZQAqbD<-!
ztrI}dX>YO7`K}1&2KUKcu8jg`s75eh|K50+Es7`9)#TjH-26{=A;a8(LnT5;NPmGW
zn4Hgbe9;<C0LhM4RK$Syd|CURpGXt(w_|`5>68v+=#;_v@tuV*yCTX@`ff+Ecv--c
zL{2Ml4fW^^W3xoZcMt}qY|ia_Yp=0o*YtNR&il`6>_mB$+%Wu43?sct*P5NWuuCs4
z#Xv&XqUpAUhcZneNAY80F2elR=%oL47N@|JI^|-QL$rU3`=gT|n=ReJI-3MIo-Olc
zA}=6l&EM2tky?P$Fa|-$0wVaH-#-p;BVVC3|5sQVBbYw<yV^z&?U8tYqS->h$!_V>
zk5pul_%A&aSCq%XD){@PJ`if16&v73y%=^7nccsHpKv&6JUUxD@*WJqs?4RwMi>5j
zz_I1Tv#?eC^>l!QD%RCDSF&{@W5cMM!#gdLj#!zFuTqwg$?9mz+pVjbLtyHzWXu>Y
zp6xLS*Zk(Hr)vrDLJ&`1mojwV3|HEYwD$|ish&5j&894lxbqj?KHA86tZpELeM}SI
zXcWgq`t43A{0nRnooZL_Wy(eFed~)NNM6HNM62K(=^MV`d2rB~mH2Mm)D+aog{O*g
zsx2isww+$$Ai%$fe0aF{Zs;thEpM7$gIPR^%Ngb(IXSjU(v#4+L%1Qf%_S8#o@A<L
zv)=i^y3_h(^S`S5o$>q?#l?0bwT%>^tI&QGeqxIGZNTn-&#Mor$jIBl999!-j=zp;
zR&PVDJQT;P12<gbK3jUvykH+B2xy|k@G1HsY%haJLCIERgZ!#a!e$7Ck8ogTA8o3G
zcdM~TybPwnC}te?YlCE@neJ#%41N_C_W7$mR*^v%$6fB0wKozinbNM&1gB4m`-$Dv
zyfS}XX<FLf!KT)x(bldK%z$RsOOB<33Mq(+_6<(B<q2>F?Q*_<K*b0EPlz6<Op8Ly
zt|5DoY_vr5NKJNY<>5kFR1{$XZ{e{fi(N7Q3LXi+`}kO2Vp3sYO$@rG9~YZYsPXzH
z8uH;>gIV$J@3VyZyfJP!kF7Ts%r>?GYS>7|-2PIPIzEd|TWuXKHn(9zBvdTNtHY;?
z8kM|H`e~Z5{piu164Mo)x1Qhq{7T+eR(!ALrfV6UK~dx<&EdqlWc_&qjOMw8bg=*D
zbFqto6j4FJzN4wz-|N<iT0Kq)?s=n9pdcAa{=ifTwv6jLjE(Iv78gzLi!Ql;A=<lE
zJU$bO<}LK+EpuRaxMboT)EQHT2ZM=hiE0<xjkNnt2AAuG%V|{i$y-(kaGxg=Og#=T
z@)S{(y-`>F>6yxMOM05mwTnR9rH*{25|{%~$s#*vSw-pO-f9$~nZk5zWF|`f)fh0X
z9L)PdB_T+UKf?APOyZd_*((i($7M7BmipYTHceox4v)7$0NcOhof+aO-5TomoM;@o
z;AEmH)y}uwCtu<j#3mrGQr|p#V-jpRlk{;aAPDpV_Cj+t=LJP+GIn<P;;sYVD%^FY
zL>~^ey-?C7rKkg=*$XXNI@~ZxL5C?_(<iu><|j80<XxOA{u_0C;LcQ!KE|8Jzv%e%
z=j}dvz%{S<{uJZ-#*7HXJAgs~4C1nx&yLUS4nL&3d5l`)6j5ohQroqm@;HneRmc9t
z*-P7XR7IR*V6&oTkKqy@pTMZF*48L{2Ak$$z1{R*doywAW;Ma(tfdl6TK@rdn|bYO
zmzB9tP?7oC!z!;gpJ#*PHg}0-$@><uYeO@Nt*Mdz!AK6LNpSrF(9#WVcGw+%zmZfz
zmmezs<sNytQx~upx&fM4z5s6eXkPG2?4uuHCV^uE)H!NZFXyK(zq%X)S3cKb1146*
z>@P~Qbsv>;-pRCJB|K=u5xc*d)fttzyP=^Gt;vx+S#jK(>*kB%DVazgD<BmPk<5qt
zQ6+{gIw{T4k7p-Jty2a}${MyQh6KK3C6+xt2&eVE^iDPr%1kNoa;w*yX>95F6)X&>
zI6yUaTfe^kLh)1|pkmanxN>N!DR0@Oi#TiMbGBJfY2vLjEX$2otv3YBml5(e4=-B`
zvpA&VxHzc62cr7(_K~*C(>W*XSJ=N=Ri>)Iu^byQl1qp^yV3_-HFV+qaGuV3ncWK;
z9Um+kz~J3a7S|eE<RWs$5#;jBv7L*E>HhJ_mlY_d#5J;nZyyTZ8re``9LqOcnOYrf
z%^-aCr33%8UiG<BS}Z1ub5qF2O!Du#HG=1jX!}MQn$ci(UC&^~X*FXDxX#!tlMt7c
z%D(K+81A9ojKO+_4S@SUo?WB@hBLQ&@*L;yW0!aLysPy)qH{i-cWYz3{K)b}GEtg{
z^PZcd{L+U~D~%d`Ir-i>$U3A51vG%$;H@rFem$F5{oBgXzz%ppSOKCWn8?A=W`Wtv
zlMuxu6ogH2;V>(Fi^tJ?|A-IymDHeCV3(+;I*%~F>)cVzo1x;BlazG-$McaYxP|A#
zNmI3VM}w7?`4NtGvf8EbKtPhgnB4U!9x`h&@5e=CQK`bTKvvfO6y@#rOHS$%+qkrd
z9jW75u0=O{LWIk-xbo!*1%)y0tX|y@ue@Mw+$!u#b|a$rf3_kqxoC)HWp)l8x@^tM
znT0nV$f{-UX-EqreiRxK!;?2Rq@3VRyotw(I>kM7{&;<|Xj+_VDvEUP$wB|QnOgk;
z%P!aShLl}O+ARVUajvZr4=9^kW@0g=?(2@Qymd|qew95KKLbZ)aklGP`ynzrons4l
z0Rpb<<?jynV5B6E%6H3)EZ-4`PFoodWR4fU0wILYEj;pIZ|L(W8zg;Tk@3A~`S{vk
zbZJ&<B%c3VjP9|vJ)j{?ea0DCG6}p_gB9pJ4!2P}E2Dq&w(vQ6aypk;S=~R~goxpp
zYMHtGu^QgVFx+pjUX7L(Yn@^WT@U}lHZn9YvXzqT!)14BSS8{Y*mOwQPC^i4!}mo^
zR`C~qbj||vPlj*6T6)VIdD}X8fvn}GrxbAukhxfF;yV)frM}5Yi0pTMeR4B4j)az$
zb}6E@p@HJS_Eb-oPn#5Nf8|>y&TlhQ;K3cAS+vi4+xGD*rg5AkJ4Wp3)Dl7#qEKbQ
ziYzKzS)3Y@V`El(Cw)LXf@Lk7h5^JPz{6Np4@hH1%+V)(d1d8W7a?Vvo!C4ANEeH!
zPk2_?&R$v`>=^ldgLR!dQ@H;|Vl_2Ho|nwyY}MV6E864cRqPQK_hP=f^rAHC!&HI5
z(c8hxospYwttQIKR+>Y4jS{RRP0<4amKuqR)M!PUDgA%YE{C(PpNr9kO$P+k5Rghe
zo)yO=FVt)M41j2&W?>vG=2V5u5?8Re$Fqj%d`MduuYb9g6Xu=8Qu#;w?n??voytr3
zh5{g@H-*$@N&%Af41ZN{UJ0=^RN}F<9r)I1{^^x`6T@lkyQo%ygWzs5r~`oCN62eK
zQP>Q;u(UB%_azd#`*scQ-ISk0HVQC8&X_<*Ql6jTIJI0akYKOCLQs(|__iRDRq=+O
z$mh+dI`JP}T-KYd5SyG(MR)@Npw(e*@QOM>=|4>aK8{<KVA9uvIS)&e<N*`tdJk^z
zjcO}3HR~}$3Pz6Ehl~gJL{){Y4*xB{*zR;!Kf04gSm9v=<@>~XJ=OqR(-Tz{<u(_2
z{Q@O-7rWA=r1=KDE!~GP_!sv?C=_HZKY3^Nk#~eP8nsN<JP~jNR*l~F^eeGFm>4>2
zk7pM8rzhq@hkCR<t)Z;m{I(a&2{xyQF%^Za#?&-#CGcx$bsDln!5sO0aPE$MM4KLK
znT3|Nf`Yn&db3_;Ys`d*udY{~`vVA0lO*?rcTMOqGxp<eF!nv~XrgsB`lJ4_<QGFP
zR?*c)x=F`Nt?B9LR8<uzvHfnI9fngH8yJ81u<c;^+C&CZS0->jP|p%;SOq_#hm?+v
zuH`MW9k)!{;q2z(ha*S%cke<a=!fN1#h+BhGy`)rU7BQK^<Z&x^~3V$Rgzw=#IrX`
z`E85axWv#g&18CMzBYOE$@6{hTI;{Bb=xEYGcRS9O5Sy`t{r1x8LTeNZS^H<x(hzR
z3RwEU)s9;msJO8aAsZ2<{DF&})_Z<=p4Vc}QRG$LUA1EFx1sdOaObw}DXS{-Ilq0c
z<NhR2%&;gjSqvlGKI;aFK(Y7{=W<ps9;)!o8oCuZTztID-!>Vzk_MxOr>JcPd&~B8
zu-!ubVas%P-3t_(sx<+l3e39FaLLqG)K{So#IzzPgMQ#N7)?k@DK4n>=C-%fCQ3-8
z_l3;hBh2i*_s5HJQi_W!&u)Lox3cQt*C7jR{}z?G3sfE8&VhrBIfqnjExu8j`29_K
z#^pUTG>vGYJ#BNV_qKaG(@A=k+HnsKsValMC<?yz!Pw*hSjs1`7oJheQ7#Ze-ef9;
zHHHtVqQEGPG5%!(!*fIC)r_v18s0l1pII%wvpt3SF>-s5E<VVjDMZ(uhk$$M#+eRH
z9whd@9k$ad9kdXX9ePh2pCmfa^e$#(zE6LZpOjczMf2}c#G74Q(#lfv4#E0|SO*xB
z!-m>A_9$MjMp8^#XSf$DTwKZyPpgGEqEclqjeh;1D$2v;)oo{5+LIchAz8ddm}sXH
z?rM07?r#A2ynRpP5qEGzp6EczL*u>4A9`Tz3;byyO+E@}Zm(MS*V~<1qdDPo$>rMW
zndo)i5p6TMI;Ss&DJ&o(^Nlg?zmKM@@Q!)aE>medPmXCYc!DGMjcYqOr}Jp*yPF+!
zw68_<_}okvnR-$K_4yLEfW!fb;lM3_+_KaDN=j<~<j)icVC}fWnhmhXMhOhOh<x{>
z*xdg9+-;tOHQ(^AD8yQECFo!8Z(#2&UcXXKQ^?#<WyqP|g|AdeX*GMB;mw-ecu<#_
zQ}sIZyc|6+YoJx5(5DbK6b0Sc-adwJmj<&HRh(F|d@CB7H^?0n@V2d9ym+y=&)Pfk
z<8pJlL~(?Xl(_bDqaJ4t=Bv0R0C##OA92w;Ttv?4i{Do?s$^sD<#FN7s7?~12=85G
zS090~6**6NRe7W)-2)(x4{P=jd55JsC1Wb*5jt0LW0Kq2n%FMvMfDfjkYIs-$wRYk
z*db#@RoOMS4i48>rle|5zoLHL!#NlDA`oFMgL!T3+K~U}@-AvMU>{INCIp$>NA&!j
zEsldnHkFBq>EkZbW@FmC+xHjFU0f1pXDnhh09@yHibe$NLndlX0^@RCAa}fzCMSFL
zrX4En*S<EGI&98Tg-T-UQr&#D+1TRd;zE)8rP0W0ruddg6QAQr)1T?musE28Mbc_=
zW^Yck?2Dr(6SCYVb1MHR${v*)G<+2oGo@8)O&&9R>&r}e#^@x5APp`b02YHgPg+ST
zBht?kYbN@f$%&84>&ghCy@0x@rXDTeyjMKtlaqzhK8cg`0^tkh&d_&Jl|pBev@KPr
zqfp&@;3oo59M+@J2#IH=W+qshGv^-*1iFSb9JHkrZ<bgT@x(y8rlG0M#lwvtp|8}6
z-^$7zGwvv!^qKNIopp}3B2MwtKVHM3>jxT@DO0;gDx-pg*ionm9FlG|ynuX}+lW*$
zJ3DiOf=7>7W47aI-LD-*jlVmIH~V@N8gJLgycVTj?D1>)bv6B;c+SOP+n=AQ`g*4z
zQj^u;fFLXP$<6bwsF$Kg7bP)eJmnP_#DmxEhs<NmfR<vcZg^J~=GdA&6~3gFOAsFT
zIkds~d|V?M;7{aNRx-o+dKRI*y*<1c3G{5I0;lh94NZdhiE!?s!f8&nrR|aArm?fK
zTk&gE&Is1^4p;0&D4}*})KXRk^MPGS-$iiC+^9J|S+IBy`vtX2fXK7;we`o0*i+xh
z6Q$Gsv!vkjpxFy<>yNlpQe1)+pLFnAgyMW-ng}sg`VAaQKauY>lnL3<v@Nf6Rh>SI
z4c{Tb0#}0}HYw-|p-+G{Dv1r_-^RK1k`FbgFJ25nb}jK#chXI~fVQ7CL1x)wyE8Ku
zw0G{{2Mi7M4fppe)$Xv<D$A)7myqEHz*%j)ml>f@lK{+m_?cV3`RI9|P%R_ogg<1)
zZn&*AiQ%LuWSaga(a?AQV8Z@8+VtsSYw*29R(XYY88%#)EAN%NgYQ1U4^V#0h5(Y^
zH!x&UTnqGAkaPq(6!=s|XLBy`<UY54GkJ#)LrhajQCw5%(dS#AL@3tQjXTC;+msXv
zWkT-0sbT#UaCp^*!^lOS`+S0uHQv0NW5{roM!2`No~o!%6I0VmEG4Q)rerl5(~9LQ
z26|8wui(?0z73(xCZ@j&58dC~RbnfB9w!0Uqs&n5b7RvC&TkQe#sb4X>YW7?1StMd
zW88YW(PGqm7rBG_846GTd%`NofPR7o=;rY)l={_&!s~cUEYG9FlCd@?h}}`SNMzxe
zgg(ba1CQ>FjkVvIHjC->>|Sl*bv|@Yy<|CiM+{%gv3`?HZ|h<6kB~sy@BDqJc^iYc
z%w!^0&g9YU;J6!}Mgpe8nkS1MNzN}WV2OOtP59$h^O;P=y;7egiUh0b$;UJpm*k~q
zVF)JwN?gZF?rSDhmX`@@G-BD6X*F%EIH0CFnhBc86CcQ|nP^{fM_$zfnQL=OQx(X3
zNqBG)Z@&D{H1ob2W`{svkF6ps%vcNNy#q&qjk(2r2sUzRj9$_!;G&~@-%hV<cl<Z!
z>+1z^G~G%n^n-I8E@IcQS8^*2YfrT-{eUHP8U9LrCz-q%ge%aZJ%7&7IG3;Ve>8n%
zR8?Kq_67t*Km?>yKt!ZPx+El|L!_moL8Kc&rAtHvB%~XI25F?byQRDPoA>jL*KvM0
z&d{?@?6uaM*SzAEX-BwS>6=|8Lf(8WD2qaIaE>VX{8Be3Z`Ybxo4;9Gx$^T9XH^IT
z`U8{5N{iB}4%%^nPDycUD!3~?nYTUIRa-_$G4>eI=(t>eveD4|$l!@h|J$p@d*B0n
zP%(SZcS(?Q(6^A4$7}7ev(pBOgWKB^D%vj!bH00@O*a1!H29p#Nh)uJSp}B}mOyR$
zVKZ{2(QzT+r=<R3&zyTL|1>1BvVJL28P@WC)@>pZhoPaNQtKK2P*$Ram7vGqJrU3G
z`bLxt2$1H_x&#5)tn}V@?oG1MwUssM&x2O+1?xWaE`0WkC*#c!PmLSZrP)mtEtBDr
zW)#KzI*pXql2kw?I<N_tS|4%Nuo1L?FwNPJ#kx5{K=as>`1h_Fv-U~Ehx2t)%P&wu
zNE}xusSzfVJwOaDJYUDe#6WOdMy&KP*oLG0N^W?ioZVsZn}qI~cA4#?wLZn0^>H_$
z+dkXg8>NZoDwg;zX|S0Q5Wm$2cPx^-it(H@$kFz=U+5pO&fVIiXd-BvaLEz5LIDLs
zLg}V<MQ4PLYHDWg*@l~H!*2$3DGjxy8_6O#2p?s*Bm<qjz}t(?`}d3c*XA7b>;D#1
zy<pRN*|cP7Y0e<z^R9Rc#rx$w&ygB`ieZ-R{-0jJrY^h;;4Q-C^Q3*VheLxfG5^!A
zvOjOV5~D664`?|(!vw5u5l+J7p0badqxD3{>(0aH&wF%BygPE2uFN7Z=XT56b90|i
zlwK;!n$#`4@$_azi4Y|(4rzS7v`nH~{gw)W&+sc|J(m5*X?q;YzaEU;Z6_*}j$XLA
zNm)-C0|ybWre6bj!+7#5DXF`8AfV!YMsDu6k_6}0ZztS+KTbgkhiR)98x^Oa^C+<L
z8p(S9#9LBbUg@}ySE5^%;ger#YS6&Hhpv2&1R*0s59K-W=g(hqb8<zL<lWCUntPt8
zV_$JtTA|C(&)SisqqCYn#n$fi^C8AR-kPa&h?S)4O5{iPdie=6NG4v`iy-XTWlr(6
z$5&nlImqgPzzTJFN(_HcC^}u`TliEdvz~O-Ij#@03TWr}ihyn0(?WM|k;gb6#TiF-
zPLbXcweyPKnYXF=ceB)tIW@%nltM{)75<eRXTh}7oSV5Uy;z^4lm?wT!tD1~2$HX^
z7AMwzJ5cGHpZ^WOoa~&puGN<6US!}H4z&1RzjF19$tiDVWwz%O7FrA^tSb4-F#OW$
z%w<mUjOy%?zr8uZY2I{Wk?Vyl2dN`Lkh;FUN(6gaU$!GT6wUH()Zp9>W(U*}4;=Jv
zAe1~h48-4i^<db>a5p&SX6V^^-7Ai}x5)Z5n69EcL)T_*(!CQ)8#wXww%K&yX@t#D
z>Vt}=W|j$`i4b*ByWz;-j!NrqQLN8SL=jmGRU>z^)o;bfLD)pJ3<DqK?TD5R!78XQ
z`B+MGgR|0+?3^6ArceF9ei;<eCO)HAYn{aS9Tp#-`;vu?y(rs^zuLXR$GO|=Zcm=R
zfGGFGk1-a3urvN5vA9=tH$@-vc#6=yy!P>LHkz(<*RMw_L9JiK=KXB9S&NQ}GE<P2
z7ZjD`XJ$JcE|x?K!Quz9FrUAg$gut2?@N5KuB@XIVjVIH?w?slcribI1-GOO=+wy5
z%4GCISK#DS!W^()_HDdW%xwc#z<qA-$-y!3dm@kTH=m^Sr2~bP!%?{w=o>FIb~6>p
zDUO-0mis|`Aut{fTS3uO_5S(|Tm(oYKG;Nb1hB0}NYYsh9^a@fEu|Lp2P6;sdTu2b
zDhu4-Z)~TSH=Gyo@VWovv|AlGYU=5wEk0DS_is)9d2JEUzDL?yb_^WPFGb_Z+Al=B
z+D|a?d{m7Rq!N{1YQC7>sf2ILYQtR(im~hB(Tf2s_#sSLfW==ZZF0dBRCh_Wk0!46
zTQV&_EWl#~NFq8*+*a`lSr0$Q#9T+>)cMF+Ywrv|Hx0oB!Tj;W#VP+=2l41}^24g#
zYm4k(lx3^FUgOiEz4ZNu7FISZ8Z6r02g_HF>F(Ki_g|wQAnXu!-ka>nFgASsYtYBr
zNOiA|(Co_O)aQ0HcpZw$xLrtrByl<<=zJpMq-=IPOvfl_)NBg}YMB!{h==;^QzglI
zf=B!6Cp&e7LFL;^_T0NADd@pb3CDt7ZDs%e4Z*@BE}ot7g^R<l1gF5gw~E|8x8eK?
zjptZU`f{P^i~8nDbBASjX@;$#7Mr)B!S4Ns>rK_-+1eX4*S@uS?};yzV6tDBN65TA
zdMJtu>bASV<U?;sf?$fe!tQN!`7M@x|1&1STlB4Y;d<rY8jaLs+DX`iGryofzCp<W
zFC&H1bv+>Xlj8#P(jLM0qU27*MLvkt8(nD!hRxibLi_S7uvMkePQZM$W2S&`B<Q};
zt+|$OFRz_}&d-;)*2MT*R8c|oinS&(qH#Wu&y;@Sf-~iNyAhYa5|!mz9F!su3CvwL
zVFf%ppQ`FwO!^k3N7v+c-*%kEFfj%RWp?RD?(^Fnm*Tcg8Dd{C0P;yfb^fONpbhYo
zD|gF7T7nQD-Lam#nfRmtJU&ynYgY&9n{;VgN(tPyuWD)b2`*<Eu9Iyf|J&*Q`?0SD
zAqNayN*7IdvsUkpsXgGqK<w|GA!Yg4tWL)DW5(C}&#oLFBnEFd0*kGz2$Q$E#N&GI
zKc2Ea-)Y$zpd-?FG%N%m5tP?>D@RvMKw7ADSp^JZ$0M5P)(4my9SR*h`Oq6thVdc%
zl>2_$XPdpUiG{NIvM%eN>eWvST`X+kUTV(>k7fU8tI$YZXkl7n-%Cad6Qi)?k9>B5
zqM2Wks(pCL*fDkSrIPic>Vw_Q9Cd9itq&Tc+0DITwH~qI|6X7FFxV*j*u4{b1I$1`
zPSU3D>FIszdakf0RdE#|&Kz^J##=ot<>fV%hUt8@nffF=p4{8iG$y5Ri)4V``|P5*
zV+7^ZX8Zj@o5Wrz7L3R4C$A7bw|}U$r7!4CfAUdZYwuW{!iOpG={;^XsILu7GyM4#
z2G*hFE(UIRjCz74?2n3Myf2UJV)gK0&P};tvn~1Ta&moBcLGefYo2>um#w29ugr`y
za&qwDMk;zR)RE@l{%klP0!|ObYvF^fA%A%~6fx_`g-`u6v$)43_3A42+rz6+ulO`~
z<$@0GY%e-4#y{1l40%_da=bBppPNOOY`F<0Zt@DDO$4&fo(->;J^ZzfyZn;SxN*2(
z14)4hQsAeLIdQ*3^HpZ51}3}R$dwPOcvPT{_7Tmz`Frtuz_V&L5)c?286Jkd%oFHn
zu0pnA4k&6FsuuaWfyM4+)Iz~6Pw<~<vk(ISt1{X35uZCl^<i21-GJt|eAf5#>Xt{~
z^j6Ked-|#q6yrbyUtfE#V9zb2566Kt0|B44^B<$Z;b9Os!ZeDA2;+kG<YF8MX3QVx
zwmAsy%cKn`w@!TQfyRj(!?<v6|Is66GMs0PbxdD~UtbY-NzPdpw}aPwv(fN?9l_v`
zhQ{tomVbB|&X)r{D+bRq`u*0#ze{E-4A9Mb%TMM1sFY!jvFv+7!N~L_xWqy`48xc$
zpN-&ov%taj$8<4<*}dmGFRK|9B(5rxJA@Fvvo%Kxn@gu}?27?rlfTt|Uj=DD@>OQ(
zTv&2o(w$|_`#&y#HO8)>;R5!l)RSgMI=#d5m2M<OZtjK~w_7<Z_oe+?YxiB*A7yHf
zF7hPlsHtjbW{9g=e2qm_2SGr!WelU=8~BR9#Arh6)Xq#~#IZB<yfB_QD`{BAqpWmu
zYMcjzr^K;eVnAHKWDLKT3=7*MR>F{G+!pcQOmLL?smJW=NyujO1842$OHk|qLnsXn
zfd5WeD2*s3eekFT@sL!bdBDHIUUR!$tUN0ME#uh57zc^1>_h4ZE558Y(3N2wiYvoK
z;He-<Bi^ql@5K#eWgo_g+cI00?1`~mZ&(*RTx2(Fc*0%dFs`bi(zd)m7KZ-*87oLb
z^}&5+=N;5`@oVG?3b?1tAx-n3WB>UwwJ>ws{M$eHRo|s3O6VzEu|;UI-VGWJ>w2Gk
zh=_@7lo`R+-Rg0?d&ASst@(1a+>R2LSn&0tGE#n*U_5(Y?Yr_zRqbaCcgXc_sL09y
z=9Jt7EHIdM-#~rW_<HEkcZdyo^yr%Pa+HMwQOzkX&c_L99<q{cK0GU$RdV2d@#P=1
z?*K~hRA;msMo{vs00q^Zva8dA(EoO_Id=&(?F0Q>S4)&(022bO#5+EXtaoHxP5Z3&
zeVT=--TGkVklmd5+2JaY{qa@uqB*c|fTjeBj1S)96WTi3U-f{e96>be@>EYDOzs}I
zA)G8^9r&P0r)B=t;`14XzRBl9+V=qq01aR#QJeO$o-<$k-fqbLO>`XwZ?E~||I_i1
z3UJKp7faHa=5j3Z$HTi4&B*BJ`Kf+ON=;&Fx;wM9^hHwIz1dxO75G4GHi_DN@uFx#
z)LCbCVfEjU*NlH#Enjzp3-j7aRV%oOrr2#iUS_8faCYB8?0|@hylNirg0F$?Fkz-8
zZ}}240`-NR<hHZpF@|qxcer3#A)4<_YqnDY+L9f(1K2LaFRjS69{zxMi`1^dnxyW0
zi)aBCYfNOF7xUvx*wKF6;e`~}co-liOWPATxKQhaCkeR#dFJ3=Wunytv~}crw1>)i
zYQN!V0LH8GB<~~a$Jgi%fUKjW6e^ZG%+v=~9Y@i7N@JeGRF-^Yomr9N!uH^->m9Xl
zrIY1$rTuy#xd=@_qj0$g8gj$QN$~iF)U#%=NXEe!_+ypTjQx+U8-0n0noLhqGoaB3
zr2je_U|uc{H;9hQR!Eo!oaSI-9~*GqaQeGf>v%q9-NQdT)e{J!o1@jlmw)(PZKgfI
z`fVmp&F@HCIbX#z3%X^W8UD*Z9z+h$OE*RdlOvg8<l0R7XHmS&<jEqAb^%1Q881!H
z)Uz?kp2;xwFil@_&`7F(N`WPVQg@L}kHLXMm*(lMDA(puwNRKR$fDQh{c~f^Wm5f|
z(3V1=KzF`$*7o&5!+D2+!wiZqhK9}41pW{`FQR3q`NOe?a}Lk6@iGXQr1b$Q=o@ZS
zJm!ColfRWda~Of;s`Kd<lF71AXG?J&cZlKZTke!F?~=b-&Jv+{=U|T%Ie2_n$+IKm
zzCM~aaK-tk{l*e9n?H7@S{u#CV%*$3`sIg*{f6Abx`7MHNgiTE-cEQ`tfwx$VUZ`R
zDqEGW;;~v3SzierZ@~L-r$I+o5w;gk{itIxBu$$))@n{$^VEtewn=ZCJ&j0Av;K$z
zyWv5Q{#!`dYRYx&w{ht;jtz?^LYkIDw#Y9&2rVg(HapZbI}M4v(pT4`6x{1B4s?5%
zL6hy;-ly*YZUpGQip#<@-r<<OruJ57^TQ~01@*BWaR)bi^Mg0ETu3In%R>lOjpyy*
zg{jlX-YtLmY&%`^KxD2$of+5!dIF_7)o-Z7p0Tq1yG_aYR6@dQ*RKMc3G=e^O@4kA
z<SBaU|C|$c3^+S+bzC649^l+9+{yH!ENle8h^g|!=2EYf6z_dAAMDo`ESwMD^we*B
z(?!6Z-2cp@Ctf5+3%w4KBR9r;7;01QATkX7fP81klLC|veZ=T5V>_@J87_Q@gS_+b
zEBrVe&*{|g6<?x(K5f`;Z0wKK1B^uH%YRe|gm1YWzV+euOA>iLN#V#s{}>yAY5>M~
zlNo9rkQ2y|bomkMhLc(N=1j2-uXa=HS=aH_=J)dV!sj?qR8tWfc*Ui^BBz3@NgPg2
zEYw{=ybY!n-+LY-Z=P^(3|2g6dgT`ulabuR0#Gcnqf;p$GIyG5>Rm>kN|00wP20jV
z+yovsO80dbF9MLFEQSVxa8Q#egoYtv%rVr0Zfp$PIOujZ>yz$%eiZ`a+FeI!MmJcF
zq^6i7I`cOQ1dkuox}zaD7+x1EbMyhY;$^S93=Y=1s$kX*pBK0gn7*LGAn8mXb7OzK
znYo!HTop%7flkE_>Xc3e1(=%oecj6HApo8{sB}^PlK=r$v$1KdYzpKkF3&eUpy@cf
zxc{Hfbn1^6+B>mGxeTaVtL_poN6Az=Zy_@QH~tXfI~^=)eNOoTR3KSnkHd{=Yg3@2
zfcE|D=m-iewxc;h7+-bW%2H<*cKg=={TMUzBb2(yL6fU*7U8##cAP{E2!Akfvj}lM
zB=fJAr{pETcLT0e&pUcvJE@e}T;LNyzXp6l6fp+kZQ(bI!--WgsT(#0mtM<X<4G@d
zj{zpDyzbpKep(f2e(zf?>D$?zAT{Zl_Y2T&q2%%YZ4$=t1%&yuJ28CC>0<D-W5hOv
zlpcHqL>w|ZB#&!9h}w)zOcxZym@W{S16iWCFM9n03G~5%i0xf?fh0ac9M%&xJ|E5|
zyVcy8PlJGCftv23t)cAK@7=A-uJM84@n5wrMWR~1GauB<*l$qrpN81a`v`LQo*qhi
zqe6u5Q)$Jk=Q5qwOlL>ClLX~vz)*lY9vpzQu3prwg+1uejjRQtzGZJzv{Le)>{uDj
zGKQ7bHc*&eIon(XKs#E(J$U8((3=znUM$3#BZDBiKz#oc;36^^v@vp~C>u-Hr5ukz
zN-2aywlM(qeO2=tym3NoVHm2A#6gabIto+{%}iiVRbJ1k%qG#+y#g@pct1|>Ws}}d
zXHefUBCqUU77N3mIZKfo8yT^ib^`{@ypi<eA0MD$f$sgXs<ga<Pt@&fWAekJt`|Iu
z+M@-6F2Dmg-X8zeIrD)2k=a;oSYY>b$#!qpk2hA&_pgn&^BPmf$IFt1u_SI!O;H8E
zd*?@xI$DrDIt72QoYQD&nNodG<(|8^EFWB5z2w8h$2d5l?m6|EjEyDveLSAGudhTo
zR_oB-^wS>8V8mv6EQe9tV`bHRj*proxqNw+^NEHj$pcJ}s`{0o-2oEMOaBcFylVT?
zgAFn@KM8TAZxw<1iGN&&mWQke>HX=?U2app+85av*d4Gb0*=zz(NsVmTR1O!GMV7`
zsfJkZ(t6TLOi9M+$0r@nBHlLHSO1Bu9+2^%Y?-OCT0qauwW?`qif%K3>p}AG)W1Z0
zl|Be!DTfRCdV6gPdj@Zb!|b>(dO0CJVYxe=$A0qbGW(@_zYaqVLF}`nPbq<Db<<BL
z*usRIEKlwsy-|a71c6#k{74;G<Z6W>k^0eXXp@q|OlgcAT^`!n3V+Auq^ZMdr0Dy2
z0=8-r2fLi3<Z1U2pD(+g#M3A3E12n7oY)U{Lpn8)of0m${Nv(JnOwD4O4ACvOFMjO
zUdyisO;8Yg`aPtU(&u;3kYmG^@13Z%XsR8iO){kEb8{D-ZhfO$(E|}3+X{|MA@9Sg
zt|?--_h&zzczpq-WBGVSR2ID5CtYvokeBTV#~Q`C)&)l$wh6Ck(uJ2$M0y-EXK%{h
zpu^kwgD*XNsQ0t&wG(}nh1AY7b>sCJyDCzjDe>-urE*LPn-?F5r1^xYdS*G4aSh1b
zEN5mnh3ed9Xy~!TW>IPmy2iEss+o~0+RhcT`BZ(6nYx9npBENuQrtp}ot}N~r0|Kx
z7^ke95IyeFpD36cK*htIUWz3wX2_-(#%b9Nn>dn`^s={q%jf(iR)MBt_Sg`P3seNi
zZws#AemIm+Hf5&B$@>^6&7uxhmP!H|&!6Xw{E^;&t;E|qkdvQVbFxY9n(jXZKfX2Q
zXfcH8!rW$K+NZcO^L0<S0*NA3>89QO%4!IS$d#X(g{{Sy$1I<Jo6rxB^@`{KVDt-b
z3waV=pwuY1^o}`7(?z|Z#KVq2d=qs>z`gcHz<WAoehA^-Kk_1qiD|jLCE&7mCw!sm
zm<Ie<5CrvIN<mKj(pz5JiG)~#6+WZ#f!4Xby(y0itrF`wcux%X@9%6U@XcB6O-r;S
z#K)J}X#e(XyJ$%L1RtID`hng5?so=8=C`|@eO`jR%Zp1kV@?oBgN_7BqJ}TT7J?;&
zRd>7yR*QX+M+9usD6_yYHDlAwP-X%p(%WRt`>ga<3#ET#>X)z5%*wWedi#}Y-RGL%
z;zsiXE*ruZwdMO(7B+e%-a6t&{eMdy6&eU-q-D5W?S<hPCim)q5e8)oEIQH)5$n<3
zr)=T3;qbZ;mug{@^a|3eaIDD>b$spI{RdMCYR($(i9-J+q8KQ44QfohEwoOjB_1d%
zq+XpKdA+ptu5Zu4ZjzKDtYP#xKS2{yl^bt-Fbyf0I=qE1BBO1#7x)Btysx98Sqlo_
zsp(2XCpESGtHuXadI<>aVx#nb;=__sZYy(hd6|`~=&b3Q5#epHG2WahF+DYk$p1DO
zrgKb-*#t7Cmfa9dJ2LW>uCHA1cbzgC!1GajBMfG-XOODz{%qd_fUH*8|7)*65HH1a
zy?;nY>5mtfkm7Alh*+Ja7q#VB@ot+qGFA@|0)J|bPhVZvkNozO3A_wGN2pKq?#bBm
zhY!z$j#o)P3LhTH$|^|X*-h85OJ!}({Bc@sr~t7NMd{X-oZp5)c<p^@<CLrg5r^E&
zkJ?TOn`v3>;*Io(nB;rsSI5Kqv39QIJ&<YsT%kkAeUFlEXt+R|%W+4HC#0Y#4;SgF
zHu&Hbd;|OnGo13ir1Bk83DlF|t{9vCY&GWjATQ!&EsChW(4gV#XM9^>IVP@v`vf_;
zLtGyliXdPXT=6?={NN|}Yt?TzZi%|Lzji=;Wf;IemdEHsnJez#3IE|**n}sVcB0Yp
z@S%!<k(olmLBmLL%Yp(W@7Z*knKz0Z0)%+0T~C+$Qq7fTXAgS3roHz4xLsfQI6O9n
z`;MnTC#N7sMP2uo;riC?zP0t$&g91nCvnf0mWU9JA3ifV(SO_8+JTGWG41yC+qiRg
z@v8BXh{LTgM#?**^s5xISi2vnxBg{5(`94l7MdU4JfF1tov;u%N7zz%3kwUxg}OCc
zTDq!?v`JFbTR;<ueo)?g-}91yu-3Y-8qbgh$llmU+BfM?QjxoeoLonE(nUuaN0b<;
z=<twRI!J-#WV_1WF7V7Km2M&laRZD&!GWK@#28e}ec)-yw#``i1P0qAq!$H8P0h_m
zTceIg^CfU4uyKb>8)>B^X=7FG;2Os{U17aMuuGcpO(9cbw4N+o-aeS1xyQ&@`D%;B
zLCu)D+v8YdV~%jU)K-GF^Ba!^teCsEmY&N`!fjT0C}(SF&FAsgM1yDvJhWSS^i!Yy
zbex-vAFm8hs-E~nNHkUDu-}}=3$$+YhK8-*nd3)gD86`e$)Hp9!w9L2U;BjaP`b@u
z3?JIttDszW>r)H``ib>)h0|riEFQj!L9dd<!?SSK_3?<z+oZO<t}7`hRYx>SqFS}?
zS|I)V<V*Uv-M%;^Ej!!X()}-Pk4N+U45rTSUK2&S$NeKiahy(cL6QPqSMAFyrN1k`
zzng{&t~j>!?gVy>{JHa?i6%mqjrYJfePl|Sg&6xrxOBjr#Dh?$jOTJDON&znl>sk~
z4Sb95LZiSB{sZ#Fj~2a75;2u}8~A&?-0kflFEoE4&`_a^HMYp@?|o0H<WPzQuGgOl
zL%M3_++12y-hH2gY`m77yqrSq(l?u#R7{;4tI-q*@x9`CnYq;_^}4uDC7GFwG_@n1
zbNKf5KksAzizg*pTwD_MJ~OJX)&dg)Nr`^9^gk2kW^nU|=5HkEAMaz|LXU0e6|<YG
z9t*zpcIZh7Ub5@{$X?^Q^Q6a#V`yG<P?Vve%5I<9a$gv6jGW%Dx~cL0>h$7C(H)zz
zlHy+D(x`Vy>F-Y#cWqY+|7}+FYN9+vez>2Co4iYHq&&x7$yIQ%)?eRTV83@y#Ai$X
zUT*K9j!W-*+DG(%T<a-?saP39HrGi`vLuVIF7FT%r>b5%z2L8slGkmB#5acZNZa{U
zz`m2Ih0WgxLBIl@eq)o(;5bz^+w&I2Qw5E&Y1<w6vAE;k1ZWmrprOEcrH!6y0~A86
zQ`d(RW!oY^WQYu8Kzc_8o^lCd1<59-E$r=K-!LeNOCU=Y_KsAH9qd;xDzXc_pp5?f
zZP>;T2u6sQOki$is<q+k{+4lZdHK&`v>aObb!FSgtAWZG@XAFyKlY+I`<Y%0y<2m8
zgS^;#IyyQ;8BB$*oT-p_9qs8GwV1xJDFp}i;v6`piAdLa9UET8n=;U?Iw|XDGq(p5
zzh(rcDlrjJvwd}L?wQ&m2|oTf90RQ;zmf`_U7O5meR(^GyXXmmV2}fkd71`6((bZN
zJ9<XO_%BIX>bkk9`6qz?Q?_hwVuS<-dnzb)b+@T$syQX|6E*(O(-{f)D1CB<2^$*m
z)x%#k-=E5=)Ejr>;4}t5m!lE#p0+^`R+c1=@0q&XRTj=OjIWoh5hmw~%4g3=(2K=J
zjFXgbAKom`O^Q}YKu0pUp6Rz8>+pxhJ&V?S$&TSG@cJ?}WeA=#OOPjwB<4~}rKR(E
zeS%N-?(0ZoE&WB^(2bs5cVpWJqOW#GDOi@P%)@T@MmOG8bmd=vx)!Zhs?#b-aMs?j
zbnByGn???A^j4;N2AXg1*SD>oH5(Y_yyd)#8J_gC=v}H)4%tMpE_PYQIXy^YZw+Z>
z<4&Hq8sh5sppp>+uTPrATNyR9l0ZZ&PY`d*C;R5TOlnJ`V`){JBV~)%Z!}0lOEx=E
z<o&i7bATXhVSJ{{Rzlhnr*0qZUg&iPG`x{kz?4$?@Zq1s{@Q-oht2gn_f#o$jOKne
zz4d;acj}g@gqv2HnxU#E-lgCIHu>gn>M<##)^EqLTSGf(pIaW3|Ghx&`XR;wkyKSw
zOlxPiY)|W_&vy)JEy+HJ4-XBK^ExSxwzU4uWK2(KGT|C2)U~!Y2cxWwpRP`}-cQSh
zun|a;0hSwZV!+}Brq&d6At-eXolOm$>7}J?7XvO`O|QYqiUMkcRRV&)&oIv1E5T$l
z)g?=!Z*Xw3{Z(LK?~0?WF}?p~m7_Kpj{AtV_ewl=PNR&RT)tMtdr1bsy7epO4)=zv
z&J86{{rt47kJ7Ph(atB3x?EFi-tAG6yBCIY)mjKRY@;F?LqmAgh3#!epR1((c<_>|
z#pYQ(Y5Quytf7hz=u<&5OhwtbIXSSsrHs`zdVQ?O;Y3W9|G7=OkrKvoG&v)mrr;c|
zaqJVKY)PwWdTsKW<Swy=MQ$4Ir<RS?+l?RVe%ymWSG~#b&`?KPd$s2wIr4=)&Mv#i
z%Gx~g>u2(le|iw!G{cQ`(9mKqpHLYVpZja-&O~4<yT;HH9xs+Y^i9~!|0*Retkb+V
ziRvBHR@Z%dbn?{dk%PBTb{62#{vKsPz2=l`+ZP$@*;!F-iwgnH4&{>~FNk&7j@O)u
zVBR}EF7WUnm=D}QxPpr9mLc5|g4|q=>j@8tD%;5|M6Ev$+@x%&ZTh?4BfavK->rIr
zDGMKojE=qD2|7-gJ)R1F_x}AyG_hKd7j)e5Wf49y=C0>-j<<>jbcB3vGhfAiJ;`S0
zG^C-2lR!#pYh=W|&5xfiT-q}n{BhV>m|tkBJ@vaQE)FvYQ)~w{E9}SdH^U_9?8|kF
zmvzf5a{&9CeRl~tg|JT>&lA{Sj$VlGe}C+4FRSJ7yd&Fw093=>i9+fx)$=nm#Tm<T
zxB<#6)T_AB_~U)k;@9%`)qdji5?<<mo|AM_asQw}Jj267!o%>+Nly)m*)G;8!}3E9
zRu?e0CVhTgx5^4rZ5m^`*o{^Svduk0p+6`oo&53obZ)^+i)DC>5L5VHw~)`O08+h!
ztE{P3kXz_>HBW@lVhJGGxw#eI>uh1qAf|^VMsDdrR7~Om%?Yr$&d&;}klyLkPZ0(w
zHx*oy&K;dfwi(8bHpe^X-ozG;GqJE#`y3PF$8*rNwuHX@RYp<mFcbCnZ)O|e6d*U1
zo?$5yIE3_Qw?4exo6LZoR>W;<e|C2txC^J~)!$WfvLZHpj^>A^{U4l7yxg9s;D;=~
zAKAd!i%&?T+5`d`K@Fp*+m^&VmwX)}(+n85B??SD>-`!>N;UciWIBzUKe3!8Y?P>`
zL_3TvAO`VUekbq`6_acC?g=%RrvE9)a05y1Z0YEa*A+j0?5yEVNJ&YtB@64T>4ey3
zpS{##`NnJ2(S4a=x+H;#Tm5FO^~Ayf{hKq`&z&9p6mmT{X#Op2!97Q4#+D@FJqu+c
zM1Aye?w+20@W(i!=-n24qW3`BgXBSg-g)@o5O1gpg_Fs3IC=+2?7+(4NWqHNW+ov)
z^Od3P+qr|NUA2F->%MSGtF<ncJc?y%z0OEbU?5N_<9|EP65Ao{vbQs-rN(Y>jOuF|
zBAmYRX`N^Lb9BsDv8`#(NK%sVX7GJVHnXn3R3^9Fy7r_T%#>=~7kChs_-$L>CTDma
z!pC4|8L#phXg}0JZ%Tvwh_lfvwdF$S>EaW}Xr{}IH!~A|?gl~J5!_ZDM-3z@qT%!(
zt#b4JEiR(eeZkFJTr$-xn31Es<7}%}>wcOl(b9v6f$__$iK}yG+rol8L8mP)u*b$K
zW`bEHm&3TEi1fnzBB-{|&BgC$Wo}RXTn#F-Q+>+}Ub;EqPzZTkaUrCeTyIxX-GGz@
zD<=sUTS1K;lO*|Ko@9FBMo|G(<>jMY6UXc86ekj-2xAz{@AEpy?&vcPaHbbWfiRdc
z$eAto>>|E*Z*4j;VfE7|s9yaGTNdD50X<E-*3?D&*}w8Wwyf~2WPUrWPns86)-%=N
z^(Of*tE9BG*ZSkqh%_`-KBs2naM~7kJn5-=G-8wZ#H;^mO6RYFv)Ruv#=jr)zOd@T
zE7|H_L4Alhainw+kAu*0Ya3W$dlzfRuRSSoLVOKJfBJi*)--_(aJX0O_~g4V$i!vp
zMqwX?@YTOI9!*V>yahUjDOn+G_oPSh^r|e|#=``EIqYZ3lX~p6PHRy=M?YE4?0c_7
zGv%{fPY6XH+<ScGW)n@T`@N_ZhPJ@H^4uAXgV@Laez2+b`#R2Bn7(dX1FT(LJlms6
z3<S`cjM=cF*I=FA-5r}8Y#I?|di;n2VXmFxIXOY^fB#{*qG#pBsW`_+9P(a`1TZ)6
z{>CHwOnVKwWulm&Cr@l-|66Ws_f|JPSP>8yL0^jE>+|y8RxPS2dPOZwn<yT{PuG`T
z+EYdZJ+D6Ac^D_J&EfJQ>mFym$7=QMky=lXX`}jnjAuT}$@qR7;deVle4Tqf&PckO
znxvy33f-u*4<GJsH0OOCeg6f_TE4tMR<NA-+&`yXub1^0oC*7^W@Xcg8lCjo+onNV
z04{Af7=-7}_ho4BKy(7saSm_y(H*!i1a{vf_f6jsGgRSZZg_8DgSqhK3o%;+P_<a^
zu}!zXdPSO<pN=4@sdQGFY)=j0dqz?!DohXF^Js-;{(J4{k#kkB=6&OXIJJe{F}$DS
z?{kl{6mH$d^0x~RR^Nn$&~2-x@`^1sAv&Sw6$f-#h?tTzHqqQR`_kY3kA_momfBY0
zx5QEEL?1ByVO0A@m&sTduN-v)`PhGU5s5^q$Mi>=$28xKDj=hhWks?r3owwob8Rd8
zBnyRTNJ0ySZ+5>qLQB|+^h(zg#9))98ScJBXjxf^)TpmH4z;L$*Tf9mLUb?e`nsYI
zS_*N;=3@y*>#T3Kd$i1QnNcXk++B-mNQ*DOg}`T}&xuh@)F0B;>s5Z)V>zpN_oLyn
z{xbmPWHoZa<8Py;ic4+fnWH0gck!`uT^|M1&;LX>dlk3jr_`*HH)Er}Z~Hqaw2v~H
zjQaKO_cDI+#VK3q`RRe5w@pk<C8eYw^zzoNkdtTmQE#I|i@A|El@!f4=jV-N(9z;L
zS6;X(ynNZ(+*S|OdCYp*r#1!$wdkK|iBa?E2qfp=ly1@-alT#64O3?xYHx7dFegD-
z9y%IGgjbpuRM+^7AQoxhb%<IQy(|)V|0%EG!yL~{QZ!%Av~S)WjQsR)3#-=_Tt)w_
zax(vnCap-tZ-2Xns!LUMAYk-1?I#)Jl}SoKFg-nMOjKN0E{RljYX3j=n{A?~h=09U
znJ+Bqrwg#=!EIFgY-Z7%TEwgVdct-`tm<AV(=w-hP-`}awxDXwRCwB28OP(*I$PQX
zaR~*YhrDfa&+E$${TNwvHT3ouD|&vakpR;y1Tv2u?*-VRf?MvL@4h%^DOV8id>d(v
zr2A<2i>u!=oRN@KXKjXEXG%{45=$5V5dm44Gfv+0z7$U5bby!kZzIm1R!$EnTYlFl
zJy(-pU3-qxj<x2oIj++6ri0=CxBvnVc$2&4kUFSOrbWB~n7BZlZtZ9r$D`n?xA}ZC
z?1hF#nEDVm_h7GP{8}!!@$y%=)&A{+e*!BVY~txsQ{8m+J&jZ$H`d!Qs$|Ka00>kK
z%iJnyJFeA})qb~^g2ThW67O8lYgK7(gr6L~kPjc4Wrlg#=JFzDtl<1Ae_y0aO@i3E
zPf6d3_AZ+FN=lZZ^vhw7oSghoUT(B%O_i6WDaR+(e(z!?ea)FjSS(As0`D=AyKbAu
z;<(V<!6HxltB{vxmz`%<?z%J#`}7^@v}{^U2uO#qztL}xdAD~h3m<%fL&Kh@V_{<k
zy#F;uX;K0#PImpDHTl(kUH@*>;Kqr%WWK9m1ItVba)$d11ktMSo~}8?9|~MH&Hd3V
z^!Rhl_+oiYh>MnqR4?=#;vBh+%AvTFRx&tPhvX2gFs5b5Pz@!1({h3fkfAA;)*&u#
zjK+|iga|3w3Vrx9m8X!z2D#Ndyk*t5Zti(pX69|OpvHEDga>tWcIQB7F##1x>`sg%
zgIr)pS#dG=*};QO|F`hnr6qtfJK7&<v3yxh4Dc+uiSEXKJ3#HFZvW`u-6PKaF&k!4
z3lslLf0>&R9grANwv~b%=3DoQRly0lga19OxJ;+!wS$zY20UcUPnv$F`IPiHV@aV`
ztc<=NuN0tcDchKg_ywmgl<Hgjt~doW54v^TH}_?p>fg=@a=s;vSp{ZXl8Vya;iq?w
z+ANxOMo*f`w#C5~i(?cD_eyfX@v#>6$c}}ekPuu70rRtO&4Kd#9&_VE*bgF-E9zc5
zXlrYaS@Q%*TC$fgAk#HhFvcG4SBJZ!g`;z5Y<I0|0<|GFR#VbDoNQBwz@FRq8NauJ
z+|%FtUsV)&g|;&?%uwrw5?sLE1At^eHBiL7B+@>_wWX6$i^?5LdK^GZNosy|!}ejq
zyUQ~X&-sGPTq8qUf&VN|wVejmr>ChBl#>25&Aq84kAN2!cAXFFY={5XX_0=3B#1n%
z-}d>gfS1fS0jZv&gX(%}6y($vCMgko<kYcX`|DTA=RzZrz&)*?s+yYm%Q8RD0s6C*
zQCFngbIRepzn<o}$wgAN^Fa7oJUNCx*(pCG|C6a!<!h_A%pVPH3l#;AsM%N_v9mt%
zp*m1hlBV$D8gJh1{?iiJvRi%LJJwed5hG<z@<kcNx6Oq0Woc%4dT7Ve*%>D5H(nhB
zB}||qv2C~Q^LTIOl=XazvuU|kRbQ%qOr1)7zVp+i%!OShGN5JQ-@hf7udn+`0X?Ru
z|7pt!m#AM@zTk0kc3iPD4qN;KwGQP+yf`5TZ%j^gboA3tBX)4d!{bZ-B(3Z4*)_2V
zfvnW(w&lh(J`_@PZDarb5u24gOmd+Xh_DvCvqxmvTS~pWKH@lqS?TUyYU>%LIvCO`
zaUkk*Jx<7Y6FH(1z=zau!MfMQA;yaOBJ;adAKoPu9via*IMMeL>tW7n*|z9y9B-fF
ze^RI@5*72it~1%h#aoWiWZwS-BeP^)_(6){Daj2Y)BBP^bY8)?KQwcoB7q?x1zI)b
zmkyOCye;0ab(pd1dxf5C@EN(vX;<b36I7~8pK7YS_$Fv|<}yZqM}=@_eAVS39d!a#
z;=wxw6I_}M(c^B<`wtoB^7$LsA28K^uao?Hr;$UQ!d$TZ`oiqex$Ka*?qo)B(80k{
zTEDf=<P=5bhU$6Iog;67U+nYQCDqOE>IGf-?8YBOBz-9hX~(C->f&_}rF2?z+L}nI
zc0qpBS$8WDtrhj7{4Kh}77?;H(-<s%pTUs9TYc(32E*fTg9bmh!-`xQf<?U8!nBUP
zC-OE%>eW2aal{&q(@1N&ux^CbUEji8Ou{uXzs5Rn-9W1w#?8}9l&<mKXS@GISb#?a
zP<`kKTgk0rOWqp>9N9vpbn}*X@~3!ed?G4*qCfVVw`#r!`>?zzpF9$+eWX~W82D_x
zw;m1S=zP7mgSR&9V^V46UnO-(mpZ|;@0_R@Nc4m7A#>@4+?<CJc$vdA^i4JRU6Au-
zX=nPvV0+sgMYPzi%1?LV2?ha|IOK$_D7<l4!lb`5!R1?d8F8eheD7*raGZ$w#A~$h
zQrgrsoy4vD5mRfy^p8vSn(gzVaZRZS`PSsbaa@wYK&se+txk+uyvv&i@O23@dnA-B
zt=4N8Gr&6j*2%3#4$78_$?bki9p6e@X#n51rS(zX)f*Vi!x;d8JMqcCsVSY<6B3MS
zJN!NOFK#Gf1B1$sSRdixM*N`RS#*%s-8ugOj{Zlu3@KTc%7s4T#T|C_MGoBH-^<J_
zr`eq>CTg0Vu?8joq2Hnhif`k;J07zy_RGZd6wr}yO7q8G#WR%Q1V~>wh#wy}W4{Cd
z`=I^Y+f!WC9CV|~KVCUd>#~t>`tgN7CdGl_nE(>oGB7+iTH}_O{{BWl<Co~@CHB|Z
zG$&_HM&a+|zJKHUm6e;D`{(E#7^_w1<J6*Rd?xB>2f$EE7SO=YH2LYpl$HKpzphkj
z{C+a1!A}_;TI}d5neh2Fe3bf9G=1T!<gcG8@xduMZ=?YA30j4&t%NZlvg68k>vZ8u
zA)0>mtslR*_xEdPk|XaPJ>;U@VKs^sm2U9{53t?VNBmEO*BnC2ibQjjP-M}>+}skL
z(D4L>Rfc5;-u!6jsm;JDyzle<@Xg`BM4N=+^2*TAFeMyMLW0Bl7|04_jm{fE&z2xq
zWmD9)uv^SMg#HG;Ov}>`QWweSQq5ikEJ#XB{Z04tX}854m+eORP#g%T^8#R1OhJmQ
z@dr%z(&nROByWAlPt6ZylA#Vio~)F%vC&iBIkz{~Rs2?AVqq&sNId^&%=trX5vsGb
zmdoCxDO>FN{K}iN3!(#EJw2y}mzn^M_bI91$AcaIytymi_Ef+RMU6t`B$O0LwlN%4
z2AgkXD;v`tR(T>~B5Z8zHks!!F`AA~sNp*WUD?^olC}b<7HM<0NW^0A%=~bVgH0o#
zF)Aho%ahPA$)Q3>guc%3_uK>vI~&{I5ux1N!kpaPf6jdaz3#ya<-a~Pz)CbRG2x>G
zH8yHLDG4zFU36RG(Mwz>4LFY{-d*-8$;x8EnTtY+8`}YDwdtR3iRI;da7#qqS@ytL
zdRuY9|7FTHTtnpr%V}oM%Sl`EQ-5<SHLK?bYA>E?KbJs5Ru>nN>1r>sA9%vd1hC-F
zB+mN+JoW6fE7IXLU_;($kA8Zc#2J5NZeb%eQJ(pak?R8v0^I~s_xkt1<6kh5UD|fx
zL}urzyI*4Xc1vpG95_KGwZDgLefAO+QN90+nA0chFeC^};$WCXv9?a<_ARMNP5Xtu
zsw$81S`(St11{1~@C5V$Y`CjqWMl{*|N9f(GMLqYQ#J^<Lh~sRa2+P3yf-dE7}Rw8
zFr~1NnL*N&<11Keo3nq8ic<h9mHYRbnwy}(g6SiesC~q6hXmb6BeviB%pXK-42=x@
z<;c&-7;5jwnL5bKGz|<4OrZ}$R@T?dEpp+)haQ(0h;E{E>IKwJXkrq^1CdWJKfHer
zkpgcmEAQ%k62C`(o(N7n+4w3t$9Gmj+t06dBGQf7nrGe&Q*z2p*pDSRi8I{Ow0VH6
zq`f-jUNDt&n!x2Ja9z!NH56UEw}w5c%M&Mar3zdQcU1x8Rhl`~R}%TUo2`SkZVpoR
zYC-W|nH?2mk-92BJmkmZ+c^{eT@kG-;f=>To2~WhEIUT~ZqjLrwViv{%A~fm8CH3d
zwE<3FH4^-}b~0Fqk%J7!p(YK}_Zhj`RsbfmEOm|PI?HQ$DlyBwAsZ)O_XxU2-QAJh
zmTNq`O;e4A<Z9<C*dAg(#Iaf`?yq#-2;0zvT9Hc&H0!F$FDC~l<-aR)KN&2)PC;df
zZtLt~PkAn9t6hbP7#R`1D4aQaxEJy&JTs%@344*4tcoERmLqre?s=XkNhH&WR#gbx
zd+Nb`;|-;fqOAXRvs-)nqfu*~IC({d4qv-HmS-Q`Vcf-07kTv>)+Iy>&KTn-XI!#M
z4UY_Uw{|Kj^f;1(3>eB<1=)_Tf>xG32_nxh+~H1@kqb<-z8m|ww3O>E(b7{Gz>W-E
z5fCQFCjp=kyl_(JCC!|dJCd)PpOqHIMnbf>6lt!i?~M`>P)1VB{*bk!we!oDRC<SM
z;rE%9+L#ITG92j@70uqGujry$UvWTNZDKn9MJ05HinBhz_g)AwI&1>~(qV@LPiWuq
z>OIuVWtIc#i^Kjc6yJ`ey)Z1q7gVXviu(Hc3GMoCwFQa^Pd9Ge`%Z)M3XoXIw;5)y
zp=zKcI8#BS{SCjuzqQnUW~(nU{<@4c${b~_vJ*n`U_HT1R3VP<OS92`&?cO^^p7Xn
z8*r$~9QOCZa1JYU*|eJ`%tZK~6y)S+Y00WcY2`?>W`C!qNdEfi%TG9hL&e07s?mQ(
zg1McNrXc-eL0)#60{@3OWWqU(J`cPT)8^ik$HnnWD_uIZMJ%TD-npbFNv6CV+p@4J
z>Y3`bCL{Tyy4t93??3IXR{<+ys1ucNyM0?bE~?_W4aT3E%dUWoiq#~l1ejnR@C!g|
zAyVU!czsI`an7xvr4!$;x*YcUYeFK8DWcFg_ohKaO2qx_6Ag~n#P_wpCfim~!+YOB
zW(d>)8Ug3Z!Wu*HVgmBO#Z91Bq@`!~wzN+gj`p?@wsh~kWUedWpSI1KXe7?h&DF28
zO#4`AN%JTK+ULJv&T+LSPvBj@tvJ7(Snk;MyC|j^ij6EqYTgcGGOoBjss58P7EjD#
zoP9E76Z}8Y{OjooyGQcOQIhj5cPWxVpx|jNaarTVQkHLUx+QTk7BT|Sc8G43Th^bC
z4kcn$YLvGztz1x~Xzz^95KqAYVXG^kko8;uhlb#YlRyJKzBeLSU}msTg}Ek*oG;>m
z82&$LA@8HTh8+53_8H2}`IC7Kv9kFpxLb=$xiBV<>}+lS+JD=&3G0{>@mW27xQ8vu
zKrV^+?(f+CmaO9#1X-2yxu7BHyH@tMnhQ~lb2Yf9b<1i<O3te0-9#8^cP9(HJ)ce!
z#*BEv!?QVAc~xYKxMt}Eb}W=GpSPJ1_}~<aV&mVuA^92l@<hJs<GMbz?}qrW6@mfA
zHL+5x@R!CpsO;$mY;WR*`2}`d@eEJ!6pWRwi^=QKg*5NEoyFGSPdDplUk)hhHD2xC
z>CF_Pls=1~+S&4YdD0>SM1cg67cI9pYgV1E+Z>Wb^}>CqZr+g3ukt^rB?Phr+!af^
zt`@6WeBpUakAT%uQVNM40z6kuaUq~+GU38PdKqw#epN?K&OSiS0cf_g$Q>CR?26|k
zbCc4Dj3p!{$j#3c@wy6*&wcBpz`FO1%efdj9+@6NJWiavf;=d=?hxUVS=eFLw)erg
z`j2OM#^Xd?Nm_8l<|YD|MAP2~+FD<Q-Oo~E>SqluCTp(0iqQf!4R2%S7~Tdj!xpBV
z9hQ{l=jTH9FCqT-hwUAZbO(18wUBqs_D&F%Z6SVg8knl#;vS5j|5&-Jr)8F<%Jg6F
z=6&>bq=sJYK}1Ifd-?HBTxERla>@SPTVq3}iwkD#c+`TzrVXPhv&R->T8i`mz(|wD
z3c_y;z(B8^sX3p0|FaO72uexkU}$T==av2|mwcC9CcUsceQ<wE0{Kx!k)N9yYWOu|
zAaEa8K~r9mw0FKyZL079K;I=F3gCeArl>%}^p;xFa$VWI@Y{`mT0@QyGo3ilMZs%6
zKlm2`m&j0KDaRL178^DLvTFO#QU|Kv-qq8jUJvz5LV18mrgMGRIEXRwm?D>~1V*I@
z$Iz>WxoowFfu!~4tNkAs@T8LJ5MrGJnb3^f68F;U)OT_`BZPKnb*PVBkgujfhmHr$
z6K!}f@#JnAityr(R*yOB-0XJ!!-*Q4!kmeE$BDJ?chb-s&Mk5}N$O0I#B|Xu!|{RY
zGgL$@yv&Ya8taZ%8eQgwEf*DnlYVg1S1PvI7H=YpiEgIVa(vQ<HCPSCml_kRvX#6E
zE&R4+wacGqlVp!gEx4K$GBUcBm{aA#<Bud!?W0bs(V|+GF<A6W|18>Eoqzhr<)9Uf
zBFDoQP7^A08z~7DK+FxxC<FO05PuDf+bSi^=)btL^}1X1Vp4(~Ty=>!GlI^H?I~T{
z1#0puJfGK={Ck_3Gqmc|>F!k-&LJB~Xyx=o0}`$QF9JT;e{-o{H+b7E-bY48`W9>y
z-4lF(7uOmZc6|%Uu^`KCjBud<!`6~lt01x>J0??OXljm>kkCuCy&=rS^*gp;Iz2a)
z+iKi|T!GZdRhY`=stnxxK<;j9Yx&GsqvBhwZ8!y}A}NJz4PX12ct`cINe39h00k4h
z7FtZz7pmpgCvGrntVv7=pwFArb9^6(H@C8N$HoAnWMaR57LWb>#=*&(@S>)?7zgPk
zI?wY>$zFL0X0j^X^tyVKmj<<7r_5q6USwueEbab|H8M(l_VEninC?1MTa2YhI_AKk
zC&1#h6E(NCuJpRB4kTBoYct$1k7eiFg`Pi3CjG*%0SJ7|?Q$})X7lBJFqaT@+Ip+!
zC1Grw1{e{<qxB_Pc}|c(=LA9r(G-K^g@CIsob5Hk#BgFl><SoLY>tdD(=zexIdtZ)
zMiciWqv<IXph5~jJ`PficbT1@J-feF@-1H;P$T6X4_-|(6a(B+s@VmSm3?^c&zNgd
zvNOAwo+cC@uGh^15Ed91%t03FTUU1X7JA%`wOe5C1YfIBYHHpWc5wkY_&W`~2(*6X
zmSYXRb+hejmM6}22S>Znk#{s-sFNDn@!Ezyx77ao;Sn#8RQ9lgKYdaISRY0rnC}?C
z+>>|a4<5ymP(x|0f{mWuQ^TmpI~r12L1MI_R12O1>sR&0X5hz9L6Q7YUnM6a$M#Ag
zKA~lyV4^zj=g-x@O9R5z><YLgz&2qZu<%u&LH&qc<A#}tCE!%&TBZWuhp@-0yO)H7
zqU}>a6!Js4uiqZjelS=z|D@Lc@iw`#mPQzk2JGl*TF-X;-YxL$nlXfXGkaRPV|+B6
zH!=#m{rU6f$e1tBCK%C#6ZhJ4C#x^uky3Wgdm68)SrYv#?pE#{u$2Ix85!VJJk^&~
zc}G(m_H8ZSZe(y6cyDbT?0nUB((r<+U!E(<KJFcn;>km)>lG_1g|xCgY-}ieAO<#~
zB~6(L!@V+4diz#z@E%-RDfxVOh!@QE9n=;&?T)Ml)iwAi6(Im1!XU25Cvs;0d_0~!
zNLDY`<l4`eT92wAK7Je$d8#kOuSe`=PK0BkqS%VE)zu@jl$ikMMR{5e4;MRqt|vSj
zgUe@S>4Hkt-)p{q|JgU#bTj7{9);1wcmJ`2pWHm$JJshuCJOZJE#gxBgenzpAoOU$
zjzLcu?y#_S8Z8hKMytYZ`QMsjKjQP))rPNE#XjfTHyrKKpc|;76nggnB<mW&!n%X%
zTajbG$BXX^8M|QE6<{dwaRPVR<A;JPvmkFo{VN&Sz5%XZ|Iz$1GjkiZ3HYuY%!2Ya
zCJ#DUcr~-0mR^FibkxR`E%v-!^iBF7J!9jAJYU*$S!_i1p#|W?7uH`B;x42p=3F(k
zxOxYq{w^OnIz<jU%4uW7$U#H$)Vi`FM+ItMg7_6nA;zrqoE)b8XdX_U{M^Ei$pK)u
z@#9DD!ZJG<4#YHYJ?0F)p!C;w8XCBT9m3i<F--87kly0qL+;`+N}i8~D}M1Hm<F>P
z4{mTcadJA8@sZQjmDkiE#A4(8FPxcK&6r+&%~0Zc?PuFV3uZaFi+-^C)Odl#oCqc9
z3ui@1b#7*S(dVdgQ3lG{S$SdP4U?TGB)MB<^!)b~G){|D6H2YrD3?*D?cK9fbATQO
z(^XjOwM|W$>MzD_|5#lz_kLgfO};0>?3&=FN{(YCgJ;#RG%RfX;}U~K&IB0-?f3-6
z1feTng|mC1MV8pv8TgpM{@iM^*Uh@lu3Is;xuC=ty=Y=;rkF5FoE6!7)MZsPx#PY!
z86eKs(b3VJ{Fo3qn>ZNzW8)7jVHxq>_I_)aslt7L^qO-E=&8zNZ*obbZ$8j?^N+Vs
zuf}Gi9(GCCdOTdLk9M&==hl5zbC*j(9(Y~)N4}tCrs{v<Ol=_w3`gM!UMd0azoI*=
z?&g-(<*%k$fO3JbXmJF{)199D5pmt%{vdAeNLp&N>SrtU3ri9;kuB&X=l09eQGdA%
z;4f_oIs%ATSs}o##eumw9?N<}8!c0V@-!ek8M@gpiiKaE_7sLc-5Qzg?RV>Vw}I;Z
zj4}GPVgk3_5jRdFAX1T{jOvA7hF=^Ej4IPiDGH8H2)iDH1}u2xcSdnR3>m4j$#wT{
zlA=H7cc`x(1~al$d;V+zb__y$?_P~V#~J;DZm^hu-;~v0!+teB>rP36r8J+JkM+~p
z;4-&++k2v|{U#ej0jR#E>6y7-xiG$_WeSJMQxq&*<nXntPE#U-_XklKMJC8A&3dvL
zkyp&dj@@&lT*nivl!Kd#7QgbQZRiMF6V40Bas_T8j;rG+gvmB|GiB=w_YjWAe%!dd
zc10TwBrwuqk6VjVE}^uBNlP+`@sS)ciZ7bK3xh_6C-dnyz1a1oxBNvluZF4=I%W!K
z^48Dj2H5jg2$4uE2DkW)-=VSp^~Pa*<iK{y3pc})q!uN0(+y-Wxo_9KY>vG8`5{2b
z8Hp2%w-G9Ma9@1%)$cviF9zxsp~2bxtji%X6&BkmHj<&P_vijqk952b99;bv9X_%7
zfFO`BJUPUMuplO@4HY?)y{{UOJ>Guv`IDdVCu@4YxXGxy>AkB`;!TB~RI+T@S03li
zD$J|z0_h$VjF>z{hkB-bcI#USS*R1_Y56TjY1YGy8#p#*lWYJN#`dNmCa1o4j}6=g
z-b=N{KE{CC-{&F2s~d1vzLbe{q5#>{O?b{A>!S=q@9HN4f_Ws;KzX#LktFuZ_hcyO
zpwf$=9(DTHKxCq<B(Lr`NgA7hy?j%#qoLDeJ6s{Q$OE#6$PtG7O<9VR>mpXn2#^L@
z_TJRGohKR8Q5?sqfx&e3mhe9vadgD$xr3Wl+`w`Xq|&acG15{{!iMzV_Vyq5g(Kh}
zponQmt#<DIKbo!rsH(1OUr<3sP>@neq>+|x=|;Liy1N@ex=R`a1f;u5K|rKKI;2aa
zL-JqmH}gA=I#=M{6Z`D7o?3Uj8Vx~xz^#Mz?5}yWJm^mukQ32#+6#+|8XCESy~Bho
zeweBLdqRwafN|5H&jqrKaIX0}XtGsW%~l`Z#Kxk<Wev-onHhmUTs{Q7Vas2B5vAhB
zl-~T>zSw`Xm<pKOI&o6w1mih5)Y9093Ic;>=gw`O$HZCuzJU_t7@L!(*F||h#>3N!
zU#AVYgA671sVVa#0@ezM-j{xx`{FGPHn&cd))GJeFz?lhHq0~w5sWXW+o1}+d2b7#
zPrZ%$?(%rI-+eT+k*zpZP4Uqf8A|w<mm?we!xXtB5j^sho~CXJa?zgdtadNMUsZup
zVTW3-MEfmAc7p9~*a@bw?9Z5MP9ac4`MslwZ0qIktvt)jmF@;mw=R=lM>tqoSh>9$
zs)O0;WUtChD8;~Fg556#GNs%x3e?lRKG(X^(!A`#Y+W_Iti+7VzryMsChUm&56~Vw
zK!adRBt%KStoYPyq{H)XMl2+}r~EG4$nOR5r`q>M1COA+Lm-|tt)J!9m5EaR4}FK|
zxF45xMhjCv1btzAmc*v{-x72DAouoT0jH}+DKb?g>_%M%kA2Fj92^J;(k?W$UffaS
zdtVlXo55giZO-sgXlHU#Q%o#X=4Eybcgal7OL#b9HNM5)T?XMMG5-O-$06@~!3v^x
z)m9_CAU`3dyeLOzx0<SO-CH;z|8nWS_M_toa-vkL$Bl@j`E;Ke67A2Qe_vCQaJe&F
zDJ8@_VVk=Bcgz(YyZjaOy``o7pri?EgRr>?pmud#Nigr@r*$s9{@wu%zK{^w#=s$L
z+OV5{X}F8|>(HU<v5nrz30nW)KzxpL+u>sXKAEBPNF9P@0J^}>bWjF5zY3^y*z&(&
z_b&5X=(3-KV7%|26XH!_=unf2G)!Y8;q@3aVusTSAR(iAlh{|J^R|sO$D}fG5z%1O
z_EcU+{!oz??`!{z6?LfN<Hb)i+Pzu}8;uik145|j8OwhkI+y<}vS3qLL)BFikIEc>
zKTpD9_B9nJyhqj^)vxPl9u@tMfbW(CN&pQbqwCS()>jTv5ZDjMJet?eNGhF9*Ly2b
zAkE+8IB<KC7LA#uLnw~_Y`Y!(-;1w7Q6cwt+cpy6RkED62p9WR(DoPo1Tpu(&%q2%
zoBjsoL2z@Fl$Y1nvATkYZ{Ay>0B`a>>Y0_ajFkKN_shQn(s*Ta5g`};VW8Fkea$_D
zp|H2(>-fv)OeDe8jby0D*TVVwWg`Fbc}>jBSa2A{6jr{!^v3}o6zJCb@I2@Co75B(
zjIF*1xaZ7TMH_=~`XdZGeLLaCF$rUd$*Ipbx6t6~Px41o4jl_-7OXeQbFd7H`?SFZ
z%xdhPR$ty8=Z_5cTV0hM(QQ)BWf;W~n9(@C>pRu^m%n^jGdnqmmzdB7rF&i>fA=r<
z`TUHL$^-nM`4%rq9x6s^&JR*~u;l}mMNrBYh4E?R%ijNM0XkL7)t$CdO^uCD+&J*y
zQ&1GMKip&o?7zM&{Tln<KVfwd&?BiO#S(|B9sSZ3YmU(7v~>YRA?umwmBgH>oU@~I
zlf!&*;<sdeV<C6<dyA8!1nmW|NQG6t_egm1)LhZ&{rOLs12&POB8xW9k~eoQ!D~Xn
z%4@LEM#lGS{P55LYR*-A_aof$<vy*{=LDJjb+<v2*MHJ^K{)^@n8}b{5a;=IbN$DJ
zghL{gmz!pm9y&ZE|F|v!<^4W<I4@d5bKMgN>)z|G?Smn0aI50rJ=LnR8+0D>H2dMo
zMhu>m@^pa%5Cg6z9E0fpPcV8)y;&Zjucv7SW`md@nC5Jlf{<8rwr(%zLuHy=j}+h9
zNmyM?g4$%-Dhia3ZJksOM-N?WXTd7&`FGSA5tH!b?8t}7{-Omk7Qru&%QLwc>_b=n
z<*l09>+wm^$$PB&OP-lYPqE^Kgaj~%c;AXkzXp#|=xe;%TGlchUgw?D>$te8N<(eL
zr`kmOi>pM)M&MA}=t7&zYCn$um;>kqfS$oo_;+CLJ2##@$wAX4HQ(!57r%igvp0Xy
zRQHV&d_>yYCW6m)IlXUSSTV7_Efu|}yqs+Nl6!qo)B3M!Pt>SPG0L3XU~~=MMeor^
zj}iOHuMKOCe!RvjSD;Q{&LxCl;e^_hH7*;MOjEbhO#{s{m#%*-q7vic2w5ux5L3sM
z%F1sA^S$8UN^H#pb)A9afq^Dv+MpVQN5ObaL>;^LF)vrGVVaG4b%e~z-0N2)aOlnT
z46U9dxc|7sgtV;&OA0|}GDC2CnHH0fu<z-kCx}3acfZUYYM+b8juot^WPi;uf=Adt
zDX_nPP`$Zl9KJ4_#j+%C+Coq<ZH4@PpWr0do)C0%BFIQYC^tS<R_q%+N^hg9z8Sz_
z$US2H%^BZI9^czuJdQjH%rYHr@mP%;FLBzy-{WGvJ4t<Oc8vlKl3zr**OG7W-Z@Td
zr%O{PdMcJ((nOn{)7$-WIL_El9xyh<?=B#n5f8Z|M|vZ2lR`)tcs@&uix}+E*8OKQ
zW^`>I-R*Wt-m+6I>BIHrOiU-uhBf7z{*^ibXH>m|W}(Qozzgr*PIo$n;$Zp?)TBP+
z>S^WeA@7sfZzk2uk1nTsKMa@{-bLODeiQsXjORNpAyyM;SmBP8C!X$GvO{!oUq9(N
zI87hC#gM@b_tO{14@6-JmAEPlMH`TV=Dvqpq#P=Bfp^>=I@cU{(lUI|t=OTczp<+0
z(q<T3asCyeqjG60t`!cmx3X+|vNT8j+h|<3JW#VVZtEPb<?Z3-ZS!0N@h<6}mH(UZ
z@o^_t*FDKUj9%B4GyhBg(EUc5y=;)<T|yW5;S0z?MCzE3n7%?y1qy6r#5;9SQ5?N0
zGKQO1zZ*X02UWzpRdr!#(ypyy#lyiQ#kzm+K+^9{qL8_Tjo3PUYMK^k7oTr#NsVlw
zqen{86k4%sK?^J`oFSEsjhmQ|r~;)0ZD3qN99XKrL#n2t@_2JnO;TYgu>5w#)H+op
ztNnY`^dH!OYHC=)ng=F2a`mToJMR~o|MwH|k;0|Ve;QOxTe)a=OjaF+rO)v~e#65+
zukEX~sdEu$wqHMg7D3;Ii0Crt>7sIZfyVR*tZyO{d@ERSB4!;jl*Kt!MU9PIoaD;V
z!t-<E$=}}n;qyZN=Lylru?oy4`C`yyG3vF-$w<-AGGP9O0?7pSwlw^$SFdP17Q8oa
zFG>??utO#KLCY^G4LT2!;qvmbC{=Ia&iA<mcR6iQ5D|hu2Kvgsl)QX$-+b%Y$zqs1
zX*}SGh1gPn+lZO47wvcCEQ5gw#Ea`6Cy@Sktk7RZXYvl2R3p7$Y=6P`+S4zx_I)Np
z+5mJxkFcu8$KNU`sOBhAgE$Lz6lz+L)fHY|c79BE#tx>Ek{qm1LvqTz&!#2{Gzl@o
z`*EX(q;=PO^^Q}LFWym57z1_EjKig0Jxxl|(e+)#xt1+WOLIuO;U^xA2f^VRUAemG
z#K5>5Aqg|iMN%`MRTLd8^em-`tOuqVzQgR{qr3SXMpX-KXew$dl~qh|MtNN>gka_f
z8wkC?tQBZ0=(#)g;rnO*j%6SA3E<#u7rX^0$*e^?D<HZOLpM@J_7Ui6l^;@&GN*JI
zDA8cEr$w4MAgF2Rv-k`T3mw(&;t!ZdudlDenBU}vyKQLJh%aAC9^PUbga6Cy6HLCT
z+S`}xt{3J<iAmC%;=hFyRjjnMR9m<CY7%yJ9oe<~NxqvrS=giV*Vo@)TN{1Ovzd4I
zdzAJ7Kf2cW1P(m%(J_$L#g-Ol#N4`!!x6FmE$LAoDHDZgXlPzB&<M}dP|&xwwJkY1
zr5A?A=J?OmrnxTM^^Hl2Upv@OpYwvt%1BK|%OJY8WbbZ&pr=8nYWb>DV81lxNo`BZ
z=+S9_<Vp`1Ji*3l<j4tTqMBM5i>2JEcR}SP`=2I=ny#$0R9Z@|nw0|w50A%j$zps1
zAC%`o-QCM^RmTw}<&rc)%G(7Yj9V!hxr9V9^t;>JsgLzB^Sgt(gHVid5`lbq!b==3
zEd|6_DV+=?EN(d_fi)Bq)`<xpRn#Wm-4#-lqAenP6mW5YLSC78&K5YZ{^-wSun4(s
zgGczt&k%VTEGthm$QBg>N#A9^6I)MMSjh8dn{eYmQqEip@-Z(D6#y}(l~tb<sUM*n
zSt`<cKqQ}ruI|W0zSsq&7z)u0c4tM-a@>`F&S{FiBPVF8oNL(V2V~b)jWRth+iVJ`
zPR>Zl$ytS8sC-*;cp5i)yA~8CNuE>N7#nh5)~*78pSGtOY5X4r^7?PZ_tAKCzI|ni
zIKEu|&h%$88QZCB=h!V&_qtp0R#OwX?SkzIH%~%GJ?=l9LMODB?<HOy*J$g#iqA5=
z*p}Xsdz~&q{)C4uLz?`F@5djT?kG)N@lJb|4>E&u<uIB23;ObGyw%9&<h|oG()bh>
z2b3oo#`@pkgR#$dBLA=)iCIrTi}Wt?(%I>dUT3-NTODuUpnSYC(~?MlK2#Cy9nn`W
zKpGCdP0k)FQ%}}Q)4c7akw7%1!oyfV6Lm+Q_%7@6P@G<Sgr1h4kAC7o)0u>bh|!4G
znQwlm?`6lu*#Q^#M_V0xK&wy<B{Pzilh({UM#6njuB}UeDb+nkL3};zBKByS0g37L
zY~{`3_Xg%%F)prty{2_})75rV@a(~efqD1+&)L}u01FL9hC?3xh8E@x%?tfXfMgMV
zmk~7JRoGzovw-fxc=POPF(#Mv$$jE#N~hxcS6T77TcV6NFJcG>%&e)h4x(?Ibh&%{
zCtsEMwwBW7dl^i<m%Lc<ILC?0ZQS$lPF?y;>@ylJaF1MkY_a<Sgpt17jIEnpww>Uo
zn^0Y!=O(xoh|r74{f9o<4zJ&0JD;GZ!9*Za!}-xU127BBuj}}7i{FX?tuLR&vXCGB
zE6iC7-t;COMsR_!zrkfm*~v<q<W(QXHxi#n82JGf*$0L?mzr7=<KrMrZ86+i63}VT
zz0wOv`21a7!(~U5(PM-@|C!uHQZ@c$`k;<8?`1XKNZSg@?N-EDP5~G<8wOD97(EVr
z?-M(kTy4=^oelSm94{|M)g1Sy9DUSGP6IUOqT?4vlR&HBmoG&gB&S=)fyF8+?FL5m
zcu>Dz{tbYI#=<hn=!JZ~ozm_H+Sci#Bn_=p;H$zXZ6K)icB|uL`S!VT;@YhfE?mP~
z%kz~Y*4tc}8>*kb&^|15Z^OrY|G)PY@B@8#n@XUrQ&>^aHn<J1wzRJ{4hFtVEZz(F
zCZZbeITF5z!IJX%^NX_r$B}T|;~c!!&Hs)v{PuxIU>^MU|6V#D^6Ojr*(-UwrS0T@
z7$(Id&N2BQBZPHwl7qFr=E^JKmbHUlY=*0;bh;sz>OGgX4NN)O@JyRywtwIfWfQJe
za=Yt3+jv%ykvqHd>GvYeQqJ1WAHw<Nr)@4c?@ofO;fu0f-ZIS3&CQLz7{0i)MUC&O
za&)S8bP|?(bQ5@v_0$xMtxhNTnRmPGTblBNx-R04H5zSR6)D1JA!+|AkQ~1lavdxW
zpP|+3L{l(XUv357AT{--s!&f^yQBOeeBC!<iJQqUrNy?^yK~Nl*49M1Jmpchq)aLO
zr@wSHRc{I$kJUbM&bymTHCRKEkJgXieP23T1&R?ND&x|Y8)*;6p;5hu83{z!R)+!~
z4Ty519E3E_QrEH=JT}hJ2_y4^X6}?11K#4>MCbp4>eb5S@n6WIIQOmbeMLhvLVES-
z{k4rYKmCgi9k6uhw(Vyz-jAe|+3L_qFFyLw%)xlVl-aitb;;klb9r1=yJmajpPyU0
zb$M)aNyRd%*e1UR{;8WIoD;22K$u(}GaR0MxcS`<k?yT~?xL`PZd8IjI;Qs)skavV
zwIv#r=dN7<Hm<IL^qreaada#0@X%M0P*5}M?|<KIyb!K?r6%|k^QK1|?^e6ptmA1<
zq`jl__I`Tjl1wJo?cnvTpTeV6owUx~k?ft7f-lTZO>yv&Zty#A{htHyv-jQJ5UdQo
zY9CHJgCH_SOiV1l7-7X8Jg~m+<~!Ll9c1kZc<1TtD-ou*>$giHo*2ZBO{=-mGf(1=
z`%B=bbvX~@U*EysW4Gvv`)}X?d~eUn&Gr3T^xt+|R_x*xJDp(alPBcJzh0LW67qtt
zlE&@7*73IbvpO9>B$n%o!SM8B=|Vzg?63I8AYGoP95_7K`CwQC<GeDM$o6;lj>FDP
z-;l?u(rHu}^dbY*ZsK@D(T3+ag$m?<+Q7faw);kd<}p!Y|D1vX3+T5vTQmP9nY8+c
z%sng?9lK3JsY7;Hfd3v5IJBp!?GroFV=T8SyjZ*M{?*Vy!)if*!!}&u`xGxmB;8eX
z-nr6wGnLq*&i|x*qm%6cJPW=AphwR8yeUH&ev5jW*wQk$<UPy7sY@E)>m>PxisSYx
zhwrMK;YpRKe%8xQezBW#S{jO9M-0`$2Vna_@BIDm&d}nI=P@B7G&CNURI&pDz!RSi
zKR#}MLSOo&p6Q?Oz8*-HeIi0;>da+_GGwf0Yt~1`@D4&HeVf$h{luQ;kC^!i5dROU
z%ddS^WML5)_$5+y0ESZ!4fftaKnvc=!?CktQD0Z*(9j|~d$zg;4Ddqx9pv{fC`74_
zi+s<P1@0LK3cQO5>M~g2jF!r)^Zj8`?aoUaD719oMu-70J!3M@QO{$;m8W2B<0528
z6_O;!I;0KrQ7jyvP!%)RXsp535fW;-u7?sDJd)7v!!5m=wZaK2Z(-&qXvb;1hK`mw
zbrm(c-!XxS&1ZGk(!yeAo8a?jka)E<C!3pFmlhSpChf$PZXyVK!1WhgA&>ZKP<eju
z->bVpNP-|J((w)esU1kaZ>`NWxZ-;i|C1NsZc94I;G)@iHOzlnatR38%~`hN)MkJr
z{y2%xRaFis@r97aw~%=+y5|)wLzjJKRh6QrUNv=Rh^%E6#|L=e5%%D4lMQnb$|=}C
zcrU1|pt{=3ChC2sOG#N<Dmw<J-B|%N@>(DkWowL1<dVW(s|!Av!+FaiB^Ba2<zUx1
zw;advz@9nV1t;9Y{L%^frld}!M=OpQkdX;j0?q`S4<d_DsXX@jnuH{^9}?2_zG#S(
z6LfGgDJmXg%s)WDsj#%%0UPbbnVMf|X#^lxT+>-G0fsCod%^-a{-5Zw!?|(8p>X(N
z<PiK7U+mM{9|*q{Hw|}D%U8m>a;hSqFWEBs)*Ltqu}wAH@4?AYQdxJU0=xsU&&HLf
zbOAF`m6cUjILnH1#=kAn!`Wb<0VN?V?N1^?;@q3w6?>J0MDF7-D39UIk&)u#6L*)G
zwy>|bimn?3iTtv_G@{K=UsI9}ZDTXv^A!Y>@8Bb!46=~WFgg64p9Ymfx-AptIiOB4
zHpXGqCk@OT_ys|gJu4?W{|$JQm+gy?m(gr?-#9|8|DsIW-o^ZA<-jfEN>m9wzUyIh
z$g)GD6-L3z_)i1x`?q%yRxY!zSW@QNbRY}}TpXKD-0i+~qEY(F5A(7sD~pOW-@M7B
zq}%)xRwQ9ZwQ9!+0Vgt)(xMs<_{<ks^Neu_2O$=z`HscS&25(J;`YMnf^-;pm{Ed+
zB6;Zj&<JX+m!6cdDX}S4XAu%3pK7E%_xNmg1{X;R%3c*loXu`rU!R=sEF5ufIG?&_
z9B~eP`%N##2<hF`r3H7s=2^aPX|`toyAc?p@7_I!Texg>?pJ7+LDI)W(2Uw22}R$(
ziyY8PM^C?Z&^&A?z@kyYQZpDMxgT$4YYX|PtLuRib_?txV==DI+n971X<hp5{b&ri
ziG7Mjn{?daW0Hu<DfpYAm$A}z?v!|d;8)$R#4*~>cgZQvn|%*y%@WzHYvZo4;*h-;
z1s(l5mn;4qc#SC+jMq$bt5E97J&-6Lh^Tlh_RoPceJ-d416N1mt%4u**R;CG5!G?+
z4x`t`G@~z#V)yhiwX%9pqM9mbZE=9@l@lh8A5g|oq+F7m`|4uCgV@JB-=y>KF(ggv
zRK-<Z-7(O$nB9IpLy+(RvRl3x{I1}9SypXruTB{F;)MwIRU5+D+IfvW(Tz?wmUieY
zNg{pmT2T(4LrhtDb)@=}%@Qnz&L9hCC#Ub-6r-2+3Shl-3O1jd_*GzRWMOWBjg4J<
zt?-}~I5Ujku{oKQ8^I+I564r<{U(bKwm?VmGtrcWmFn?bznU4p-|RY;9E+KOv@kDE
zNlr`QO{Ns4nYFd0*&myS$|53eA^9jM0g{rkwpJF~)ygpO$amF_pM~ocRXx}u2>2y*
zYIAamYHOX3zr<d7d2OXkgK7`Kq-<Gn`nI%V8A}=HZJ}|goR=?IM&l4~6%}Xb(Etj9
zi5TSM>YA#Vi5Y$9;%J1Z9>243R*l)le?3o2U3nURDSQzY9tsEZ3CrQli{J1%q|a6)
z4|ax-KAL!&6qjITlTSfeaOkjPkn@teGYEvT-#;sBN~6`Y(H0lQer8P+0y*)z4Q9yQ
zj{7=9SfR;!5DM#uVDg%JOG^4}J9P%)P8RGaMDIDPtX*_>fB25QP?ASZ0fEEnR;mVy
z7M5BB4@E{x2Us<wr8gcgsLi)r?td#So)KnkB;NDA3t{Pn*=2c!**PU2_$#Nm+~?fi
z<h>P(8b<Ov|FCMK;q@2}lFjX{i3>xh2w`DyK<5&$@6Vs7wH4|J8w!nb)G8^ds46Od
z8Zir%ENnS~0O{>We7TocptE#0S7lEd7#iY`vK7(N($W@UDQ09IB>Cl$)ML`#H%Y?h
z9s2Fv9D74UsjI#24ho`JmD7Of-Rs55)>gwUgUj4Z&kZ`ei;>chikMFg4eD|>{I;Lc
zOCDXE-tqhVp#d&SUEOKJ2lARiq0YFhtoAPrBBC0Sq8gQrjsG0AUwV*JOAwXu)il+?
zuQDxN?it|^fq@nl+oLQjzS%$T;C~fBM<!-uyu{;74G@oNM?-t;eG+m;U3)e>@*|qr
zR&VY$F0;|1l;Gj#`}bRXPF0M$y?Yjtr|u#eoX?g@n;#?n{`%LNvU>&TUDJU7L|xq&
ze<mu%B_#c<9GvfJu_vd<-SzwUaY9%;Dp=gdpmG}K;>{%!{UxBmMnXVu&%f%;QrqX0
zun@9DtTmAvuVSP7jdbtnvy|!0*l*N)(eyi*?MgaI&5K9s+Co@}-xX8lO0x6p4WF7m
zc!%(AI=9r8$_i=s8x)!QYUZ}J=?7U;5v(;Bpd+4yejiqgZorffS&p1N8`3*8t)5m?
zj7#$Jev$^62Fu=Hl^%up6>y3ht4&W+=C_@-8WInusuo7Q9$t8Z4ABv<d?3qmY+>PM
zb4wb5U&J<b_5p9^FB}H<)=n@h^4)udJF1Kv9BC0)GyQrUwZ=X(J&vEtFfL>ZUY`>3
z`ZpBN#78!l_PUbr9z1S|<mZ4;7cKU(XUtysh(CO*M5m+<ePWX_!*=5xs$&E5|5<j!
z)$ZJ}oxkHZWFH!zKaKeM@{((?kI>hhu&%zI1dQ5i#r2zeoSh>On)QaN7pxm+M3(?e
zJh*$cjU(MCX*2Kbv{co8@SzgyE8?Vh4F48ej=p+yv(}@RouEesM`wzl>@nd-1U7Dc
zL@zUYCsUU^uw>^IO|nqCSX@_@De>aZbY+)YhdRQ{+LC9}`AYb^c#ZgT9+Pp5v&azr
zl5*QA-zlmKM?#>dr)c6M>TPs`<q8zC3{Hu8r7X=cpW;+jvF@2B4w&?d&anRF;m%-g
zcXNu5j<&7pvU;r)uQSVTZ&+7bo2EG^7MYd$(f<1eYZEr<ll^riN^9M{Q68RC8cLhK
z)7A?zXB?bWVt#idL=*Q6zwTudbM@lNZu1#`!C<azIk?N)GX8c3ql-rKi}er-c$d|b
zAS1qDhlL-FKEqWS$^BigzrmmjsR&322yJp^c=ZGw{Ap$9i!JsWNHh70g$XKsNnVT@
zlk#&7N57X7{rvjcifDh%3EqC4RlBlXc(7@tuBoLpX*-RX74ryDT|&oR?uz7>I>Wuw
zh%rZAp0yylBfs~qQoUeM*Xxu8A4gZ*^M}h><e|g$=xcQ*+d8<Np>!QH83f?2q^vML
zDZVn9t-ZB%W$XIC3bg9*VY6tTsHjLL-~GFOiDhZv%tZW7U;O3hqlWam|De-Mia~eX
zzwUdFd=^9o@glUjU4<<7UjRb)WzKbqiiQgw%J}`sBNxrQ><<(R!O33z2Ul7u#8!Q$
z1(u4XWttzNJ$lk-mo!y?mMP#HYi&)26^s7RDgE5nsHiB(F)O+P*#kSRtdE)QUw1ei
zVtd6Y%bhj!ji;wC;d(}8@CMu1L2d^~q4S-Ko9+3oQNz2qcU(^(+hNm_+3L&0e{-SY
znhp1|;VJPd1GnAhvg}g@esoW{PAj23Ut0Zcl(d9{M&ohsks3vwu?VvMSktdD`hJ98
zuR#W<_Y&ksnMreLi+cuNqG&ubhbbZq&Wn}hUm89Lel6E*TP^sR@-cSJb4kirg$?FI
zwRgL(I~`T*f0@5*MJX)SJ>r??;<TM>^x6$-oc^4l#c^`O&d=KzIZeF3R@eLug?(d3
ze^>u>tWln5WG~wwwNScU^T;Z@p0m}#JLcp=uRhXOR-CaNT%YAD1H{`o<>h?`V|Dwx
zh<Av76`4D9j9~|%LV7W{F--!m5r|O>mLOfW5ZZ7Wa|V?s36FnzR<j#PLtTJ)dmp*j
zDWfIgjI?9TW1GJIj0ao^2)_?xWs_RF<~6~6T}~C;R>X{M3D<;({sbz%>c#--#GdIP
zjv$X!4qRlv8o$+%P*mJUm3I&rc1D91tRrd5L>(ErjooS#NMG{`3o@KwvFVL3c`NAB
zVQ9%qP0F~(nbr(h%p12UMcMa(xb|UO8|mUS;pxYUjDi)$;b|C~y`03!-+NTvaCNDp
zL)AR}nuO1NI&vi4n(z9e>ou9PbAF0!o99ZEe~Z-~qk?iwv+v4JC$#Fu>e7kkvIeKq
zh8ArSNgrHjbS6eeWpeE~(?5Px{VS=iC+~VuLj#h^SLuYRoLn*)1|6~;MwO5M9wf8$
zPWhA#oP?Wx!G8EK%Y@xMFuwB}A8}Op`jH8<l>FrFviZt^VPy9+-9|fyjw$9=JV`~N
ztBKkpm%0fLGH1pNKIWe8uuHRnBIDPlGDoL#&F=Ht8n5Zm#Uk$_=y01mdf%vuSripM
zpe=21+gN?ZZM6T83Kbq};*a-M286U_j@N6CUVUU>AuRRncP#JwnK#uoSKCyk>C$oe
zX|N*u)P?=M+Mr7_+XXS1_~zV$1~UU*pU{BTPGlaN284hB?dPoW=%OMP+qlxS$-j1s
zOj7c$2SUx&RMB?T$vmDWzkfNd^gQpcs@iXNOD!gKHrsP4i;EqsAYBON_O_J{3BRY!
zWyxnaL(Mmzn|aybGjFH<r#BZ>jox*fq_Ek1mv-oyr$NpFf+W1h9fz29Nu$u#b&hxj
zS~XHxrl*D`d+Z`^jGpFN;>kPK<H=A~x3*@8S-x?_Y<bv=c&oR#e<5q;>oR4;VjdVD
z(2%_O2<`Pa-JE9hYeMg{?mYXT+@5H*2H&HfQ_Q&@zBp+Er~EahaR=u|1yN<{+wV**
z!LsX5+-PmLRWhIBuEv<Yys@|Ht;S8y?|wWip_4&@aXQK1Q(6lVCo4QR?62vmA~;&S
zZa&})5lhG@k2@}62;cFBlZ|?ItoHT`0c2^vNKE}0dk;Z~K}F9eU9!XCm7e`AKzwL+
zmIv|T_qxsl`^s+t<37j4Ko5=e@HXGGan$0_PbWq6r@h|!8K?vaB|{ZVo^F?^VO1}~
zm}ERdCg1BQ707VAZvU{_*+N^N&e?xsQFLajn_|Iko>_X5s?yiM_}!oqwgR2>jAQZb
zGQH~MAK&I1Y;CVlnEY<<Rd89E7v6+<UcLEJL|r-k2N)^#G#wqK@P0yUoTM+*ub>IO
z=iWF^Y_p@I=96W&)WRjUf{yg=N<9ExYv!ASsmsG;gv6+b0xw2}WK*62E0*|PUY~Vd
zgUT}!eq3B)V{2pkzD3)Vm5%)%^85uuxyq)4dUDrsZTgW%Z0%!Z#UHyU$TR*n4e&2{
zl9Ijl+*z+xyO!%d>R0`LEdae<6ZM7TTw}NzxFeSu-2a41OK2{w+8xc@E8bck3h*yS
znL|cU#&0=pYg=l~j&_~G59RZGHkUQ=<jL>F-6cwL>iY<WS4n?Q?PmYBz2^|pSFk_c
zI3L*2Nx;UTeN|j`Zvn**2#kT*Qy6DN!qvd@yQ(}c=SosfK;+#?)S5Idd?(mC_i&?(
zEd2e2XJ&ckI7;&rsr5!*$J$5wcjRPT@<Q8P*K9jgKbFz^j(DGY(70)SqWRCM-KDM~
z<saeI%YABdnqD$gSEakuy@lp;;@Nv-au*k&A@}w8eV0F+y$|ZnC-TI|PNQ`?Y+v?O
z688*&)ri<X{`bnsnR(jC|GzPYUkYh+7j9+SJHWB*^KW~%B3ttYZ)kJdOx!2J{do1B
zhZFt2i@ri-_O~y>o~~XY=Raf(Y}g+l_<b%SY-XmcPSawT5Xm;oIEX8TOCSs7S)aQO
z?A7?U{0aZWF$cUf1_DBZ{xq1uq9WrCu>Z5F@m-6E--8wo0FZupA|6)zj;Sy5S-^Js
zDO%4uzqoKrU&(Q=zq+<E9dNs5KkhL0P1vi9-!P-65C5}arM_~t{432z)qP_k63RG4
z#k)*;3t@hKeymt(MydcCRn6HM7+4CJiM=@u93C|XJG=El9S=wOSLa_<@#2p$h(F)_
zF|Jrl^jml>be$ZN&|$>V`gr$ae4?a`-0r+8+JE7gdq=!tmgWbrcr*IKQd2eM_nv-9
zgai{_7ZNaYOxL;3zvm++#Msy$QG(745fhhMH*`b|jHT<(%Z?O7bvl>5>|Dr7M8q6Y
zT>m^t18eSvJPDuURmF#f<g<g<o9VY`Z&ehK{o<+|5C9DeH#g)gD`JIyh?+y#9sTXp
z7Y;7bth|dLhMvR*sSq1!d`2K~e$A2t48r*wLR9-d8+ZQ_w0$=IGvnKkrO9P&y$c<c
zwigxxvMZ!Q&b)ZnrEp4|G$SH#WZ2l=h^i>Yh)08l2)rpsh(2?rVKNOkJMXHe)A&xw
zDGFu=Q|;H!mqNu>_{Dtr?WRx>F(FD|NrZUR$r4R|Df#-mypx1-J;r;KSPvc^I^^Zy
z3{$-5Ub2Hz@}CDbWx!-_&b0wX3B$HQ)$+Keww{E~C%-Kl6Ol|e!9Qfks@WGC&|I#|
z$!2``-IOr^v+v5%?J1Va<pLQOS4uISoNOv5>6&!YZ6wboCQ>x2Uz#g1oqdH~tHFc#
z-s4ugdpLIrG3Oe*7u_3I4k^&K9(4oxH+;X*jp8>zj=3k|a}6GYoz~t`7Bgknx?f1j
z@6FNDGCbqSzvLbxY4JF+S(#Chi$dFc)SYsalxohpau5XlEwq;*axg>JR%qdlpf9I3
zI;OK|(N*m@0c^Nkog?@^HWw>e8chF8?8TS(G#|*V3p?VBxw;JfL7Ba~)~<8yLiTfV
zd!(geIYgyNPX9=E%6t&xr*HFq*I`x>5vKU&#hd=Zp^R%IeSM{ouEGL;lSMV`gtHs3
zZKtbyh%Yv!8aE#3B$<1=b5zV|i?WKWhhba?)R)iB`1ZodSVh?~`<p1_W6>Gh@RD6n
z5pl(e8~i$RQzO_GXS^;7$xM=FlEd4di?9KTtZ0hs<U~{xGg9XzmX10*Jgd}}RlkRD
z*f}~C@mOYR2v?G)C#u^QnbSagHEf@r)hg7&e-*KT?w*3@MR(dS()C~jO?Y9@TZlrD
zNUx06njyfK2tsIuU;M5Ow$N$wiRXB$E@1~tA@XlCNZyg&weEz~7AG~k|JhyRlP3BD
z0s~kXcm2B1V3<wT%x{6fPH}nQjP870GRzP_{DyAHjnHO7rq4O8-Zh+&>XrAO&$cpF
zp!5Rk$FfLnrP)~!e?7ru**2vQm+NW2U)mXn%pM|A15m+?V+M0d*1W%qlhdYT#P2ki
z@k^oo=LbgnV}d%@R6lZEye7n+)@pNf+T(h~sEfXtzk^NukB0X!9v)gYv)_NcN&=s0
zPB(xXoVzRkF09Qs-0$j|nOpRyG7p&l@^I?g@HQKY6FrhYZ4xV2-?rOZ8aqn5QqXhk
zBzT_oHoon#PX~dXwu}=DTY7q=5t3e)Y9p6-pOuYGT2iV+2Y8<*dNZZX7OR7Fjn}8Z
ztWj8wNp?ea)hvWwB1qr7=`+tz3!f3LO-RHx9n|=6Gc!AY`{}>$2QRQZ8=FRz1O3cI
zI=742$Q9R@)l)~)EKWxxjicvWLct;zwEAxQKOQ()yAAJfs%Zy?pMNJ;+b|x+18^V0
z(1{NZzZF*FxjSb-nFFmMG@1fFxOWkajt^l!S^Ib;PEAd#yU}E@m~FV3kD>p$EUGZ{
zKyjlMU`YH&>ow6E{_b^n<#&HqsMFAd%4)e9o2>1}_SWBF09<TcWgjrMsj8)eqZfdW
zAib-(2{|bnv|R`y?WF@tcu%4wAmu>wUomFJCdJd!TrVAd7pIIKC4n@Wc%`wJ4cqTF
z!&}E%t|T^^^*_(9aIiva+_H(du|t6`3J*0ga@IwAjsKrl;NHE)2omad8+cH(Dh37y
z!qEkpnXj+mu#TBo(Efx@P)bs&ft$xZII`#CiKSZkRE04l<Y42%(@s8%z02iBu}lUR
zL1&+Q%g?+0Y0@5{ZX{q@6^i+gB&RStyNtWjh4kXlP?n`|-hHpd6#!cn?-2**9t&G9
zMc=8+g1!iJ62+6&mDRN_&MveJG_Tp1kr9~sq_4+9A>n2&o)*u8EVze)(l_M%OHN6M
zkx|>kJ$Ow0b75DKm4z+GU$`R`7>G1;d^+}*w5ft$n>l%6L%vgB9Cdb(2@44{*g8?f
zN^Bktykmm2%Jpf@H2D<raZz`eHFO7&F&Og%Pp{?cAOHj+kmIy&ys;`NEy+DK)b)-j
zZKBlY=^_87^ADDuz>%yK%bo46E|DbeDMO}Y(BsvRpJHHz8kuk(T;*9t$}$ZOgE_iQ
z>n=)sqiaEPNW~N4Xdb84yWCzuJ|R|?wg8f_4%)FItaSI58w>`DIrhYM&h#tn`pvu%
z_tB#N)JfzRCIY|(XIY)DuHp0XdIqWlE+;u_`nPwQa%ukL0TrCuHa0$<lg`hv)Em5S
z6*xe5^5z@j-Rm)j^d+-^5s$sj-0f{14(SjjxCUxUhEEW5>873&d)o!coJTbAPm;3B
z3!1HZg%BYZ$6TZ|c?}K9o+0@`O`7iRYZ;f(U0vUsPHxJ-!H9^2%(CQ=keZr=l&$aI
zSCz*pORcTJ=ZyI3K5gL4hQq87n>V=g77Kudkqa_#iFf4kr^u%yB&TcYC_(&h1Sg|s
z^EB?d(yGyCAa*F_@LjLh=a2iXM97!y-B1D=-Lf-AIE9s-=Ee*^hU;&=yGzK9N<(5j
zy-7<;%OWAGqN^(>5%Y7&E=>BTV%z8Nko%0z6+KIKmTHWTA6)O#a(@3cO+(6@ms1rc
zl}D#nbw9XMMO7t>-?dOcg+0UQ3PhN&tG8lyjzkjqgJ1GuKZAGWjF<X6b#ZHXSrPG8
zNhRtE8<1pzvRppjK{~`&fJO-FP}<RTYUe8PCI{8&P!Q+;{;`*6h>~M%Z$Iaa&h#S2
zp1TyqPLyig-03u_u4QFQmRXn7YX_!>Jn(>l&R?z|O*S*c=&2p2m8vSk%w_8%j3qnH
zx@wabwEDjsdgf81KD0$UsWa8%%GyyrjqU^wPdd+*><cWPY2>7$DlA08$B)3T52mj`
z>veV>dk4xG2<DyRk-l>s3ymGJ-<N1+NgBN{={^+8FQD9HH8^u8-lgj8FsUA4)o3*+
z46Dist1^(Wx=M>V1Fuz!BCi0lnw*rJ>Wt%J6iFzfz31?-paAcolD5$3JM<HqcuT1L
zUG<lsh!w6yIpY7>|Ff5o8VeIsx7y+vakQFp4BF+Bf6Ux<g@w;byPdeb$1Z+BJV$E4
zucQ)nt>X+U9}Z|Mx~V|R4pl)W@8c$i8zT)-ivNmRM@~X$_-1Bi0H5z(IS61Ihp=oA
z*NJUK0%$Za!0O|}j~h-wMYYmqKG4lJc&*TLiFte?rhR;p{gxUEOb{^db#ydh9pgjf
z0bRfBNCA|t%8-fJT(LhrF~*Efhv-k|ANX!Y82tLw+RV-8&C+EEHQeV7U@xRDFZq1F
z<v0glo3o4sQWgjgL3;PgL0kpL7!z~JxOn2MJ-u$?b$2uyz9}d7xwRQ&n7XY}g%{?(
zE|70_xL)s@w1o=nV@8^qvUbjw7t!mYduE$vPUE^2d%0>(?sOzKJMWbh-hT55=_sfg
z^31_F>bA|UpBr$yi`eLo>Qh=1lR{U+RK+5F;K-T&xhUa#5w<=5$O>g9Ij?z0^7?gu
zPx;ZJYOV5F;j!|VYBReLT85}^%*kEEM|8)S)ue_vtl@5}tO732ltBZ;mp!uz-!?j?
z-9b&hji4wo?zC?P)l-CZb>}gjs*_w_v!BrDp(e&0zJI_|iEjmyE9&-JClzhzi0Y`B
zb}<45w7I-`q1Hgy^{>XP1#s2BCVrGeh=Ndlc@F`^P1VOPX{EM@&o6x&yO5Hz`O$G!
zgzd*O`h(s6aE$kGRj<rjD32&UKse!Mgiy#SC-1u}P?Kav7i*r_F4uED|NJI3N$&^2
z7alKl)C>%S#A$X@(?kd4VFzp(M~;n--*0e#e2wuCLt|-m-?9m(^^Ug^(x<GT+MVs4
zk$Y;Mp5n$T&RMwJ;K^K{QfE#nnL7UfTCFmp!{{d`TaF$kbpYlfUtK{uo&R__)+<m`
z7B>DHY!%=4#4i;%oa(LKbTw8vnG5u;c;G=6q^9}{cDT^+P$WTfjg-H3ySmFO73!tm
z0tR&(MyD%Ca@MDgNtH*(Wx!`XTWhTZ8f*iH?Xj{~&%OEP0hIh(N|Yn#ao3d)2OF2(
zQM<xu<Vrqk1t#m*EV%o1Kv3Ej<)7;(4EiK5VWVwWg3T#OP5GyUG^xrzc;w74b#h;p
zRhaeW{JXmHsyvXCSV#<q6}BlZG5J_`bqx<VAVr!4e&+;_?=PHH-n_BcpJlSqwN$0w
z$#{YUjc-bNYPa<8{<xf`*mFt4k1t}?aMN1%h81*(tgCyU!Z6MZIdLp~D51e)FndNQ
z(rak&^bb+{!uuWya?-kcmCo9+>BLJdbFtCU{o<6(+&uqD^i+O9+*ZkKOKWA}mi1=v
zgOzSNy`~Tea#oHp;Bc~~%hM(w`AAE@Lh)&^F9Ep>$fXn%v7z6tbT<4t;90)d@)Y~?
z`Z`X;YeF3jl>s7lSeP(>;k_yWZg^<t60*(03I+JJHlN7;WoG#!Os=-w22bW_*wnsJ
z9D9p6Sa2);#Q%!8nS4P(6ZMY(H(^+zpe3gWJR;FO?T@2Vc|0jI<n|~jr(gT*pNM03
zzF@;a{N7kuA&WPsr<cJkZmD^sv5;ZFOwc8Q^#`JzglkZsqNv^*9)ZDbSuo27o^Tyr
zVUdXxaWXYomX7CBG+9a;UUC^qF-aLV%)dM?SRabMX#aE{K}%a4QJ`+GCt+!B4hmiJ
zz@jh~3zd<q$`=cLWnyxwC>nf%<e2fj$bMj(K)B&t{T-NP@{^NYv6*QQxaU-&sG=ZN
zAe%t;^A!~%3C^7^R_5oX2CQX6csUcWr--$Zk(2*eGdp}3hIrS$Q5g{IjqrmQL_rU+
zx_=3hZiK`LP|qmHj0er7)d^V6Je7Yj=X*}aMxR?-ivGZn)H)XgltztDaoW-{p2MRE
zvVh?}8z_(wT0DM+Ycl7K?Y5&U2Q`4Fr==yqS>W6{(dYb!2!DTN=$m9j;YwGi)A6%B
zTYE_jEan_0#Ty-My~TGdzn%wv%GOfQ=pr)T#1NKIzIzc#Jo{EuV@vpZ4-iuT=D6|R
z0MS8+frfyhPa%u{j-ODHAU9TLVOgQIrS0B)$9*t4J$yMpr=ekv*c>Cp3zd+O?Sn93
zV<p(d`#Ld4TYd~MA~N5Yu?`NKrITLhynTBYfybcN?nZ%V=%j%;UysCh3JYmXjZ7G|
zTHi~M1ARYKTe<vq!Uq@B5F*IgEjfq(`v`b=S>rOfpgKyGx$<!2uaV9`r6+6uQxG~a
zs;H!eiood89-b0v&B~0OCy6qt3XQEvAdj3lyjt6d{L2G)!n4n1Gs3*OM}+&Gy}wqH
zdtXuCML^HZTGu5aN`w7=-9!`2%W;YY%nr^fI0(p@KtkYFM^`5oRVhvdOFS=K6J8bP
zCBEffCQA`PMzG-+6Jn@oY57jyipk2-XABCa85#M3`g5?dvI#<Dn~Z*qjwiEq8f;_7
z>(wcXzQojP!}v#-Bxq^=W&&qwQ8mkQL`GGWo1QG*RhcaE?jsa3s#h6c2ZWe(ncQEK
z9Eh(SdBttJJ#@R-nr0aa3ab8D>0QIM4Ejv?3Yf7_Tqr*RWbEdB{eQJAHxK9_?juw`
zDqS8WmRlCPj@K=ojqF%ReY)Sc$EYSJtjkjz5JK@oX$>J5LcVHVgUWZIn1+Fn@(&9M
zjrom-xAL^qs&ueyXg5CBJcok}=^0|`%btJD$F^}J#Mc+Ga!CKa<pdyJ<g%k7#sgJM
zE!Fvccv!gD;t~fCFxobFd}q-@x3o#5_}5hl+ph$g1SBPj-$8g4N}~F8j_@xqQ)OEQ
zhSKcuqGs2;a!%UEg2BzB+7A%{qTIh&YA9;{8X;qh1xK`$cK3IV^mK-iiD_rO5d2-Q
zSgl@A|5H$v<i`05R~!U8HWOkXp|Izii(tqQMx5glg?AZvX6t}{V)6m9!_dD=pSwOJ
zEtK!Mu9^BAZsL4>-o&O9_NJ=H1$I&N$Yrk3_Me5oW6sS=gXC#iOp=-)r7u|=RBez|
zrO|GUNSDGC+$O*NfDmZ^^dhl-&!Zgqy#$Q<)kb>};e)S`Kei>UT9UzGI#IMHjlF6N
zmz2x<)BqTC6~Ry91n=hkzIo|10Pkiq*W`IB(>GmhQZZeEmQ>z;A4bE0R31Z);b^_z
zVX;O`S>zl7ieVuD_X?ZX%a=kewbIgiHCzsxKOVg5E5tR42W-(xa){SsnQq4X<iU8$
z{G;p>k}KpZ#MjnYk^GWN)@Z1F+wg+0es91`hV@dSnTw|nK0FnFtVn|lp@*PUm<uOE
zLPE0n*DfxumUU{nnss=x5~X5JBZ*_&RgCms-~m_<qda#Py?;x{Sn(pJ0CnoP{OV6J
zD4`sDrYWPW{9%5FpTGRmTQN<G&t-#o#T?)^Q)>SV3Jf-Owu7UiBaAedClo3nQjD%C
z+-6G+#oOPIwTu;D-C#Yg4^-OE{-$N1ynw>{2(SG5hL8{*_b==?OIk{f^?EhMG{YqR
zi~Ro}?vu)4s`|Usb19bOKm4k3s78Q0FZWk^vGNa{IfeTvGRtg^2%TUtXyiMcYxG(?
z;?v_k#-+L(NC^V@tr(ncqn}@1aCvw#3$~!u@fu(Y<KvWI*}vH?gXhSOcH2Ss4;?Gt
zfq5~;Kb>cZ$cUP&D5BS4YEiB7h9s?sR)R%CUHnIg3XOKfa$>DJwM)x^E)r2FNWEyT
zIJwgC7qeqsBy;`toP=k)p0BGMO6UqwQ&PR6WPe+*+CkBEz^TolIc(OF?T7G{u@Vv&
z7l*158A;%m6Bf{sk0#NzEG>3$87l}3Ki%?D<L|y~qD3B7Qp>5^wU-ZXQrDKjo@QP0
zFP`uJD|MpOuM%w^i#wF&sz1P*no;V!T7YjXde7(-|H%45Bael)-pt=cqo3<`mg8ud
zsP`_@xglS=-n}*a!w1fC^*34Q#(ajUpb)Tmaabuq84n;I&OaK;m$cjzY}ejy<Cpti
zjtZhN2#QXg%1+z^eMd=&x~i^*sxHK5J%5gSK+E~Lwlvapslyq4e(`g=y$(0F7DxG7
ztBcCc<5+Yi$0n{(9jvEM8<%~LG=E!{<#VO}2;sdjQ9E7HA1Ct;Ktb3>f9j3u(pQ*g
z#(%V(>Qi_3`3F*c5)fu>4Aq8S;OCa*zoerhfB7;!F{9ai)`p+OeWgi-Rw@j&#mh}c
zxWrY*+?;<6a#!_6S=v6E5+Cv_=qfb+;fX)G{!nn%d$y#|{j__gzlIKCZx-_KXN9XP
zrV)PM4pkqI{rX6_fQ#_k1-V2ii+@w<!=0V0wPOaaVN|y^Z>(hlv8UZZZP7mld&6%C
zqk(4#>husNsFoE~pBB;Y*o{CnG!j2~MuJH~hk=HZd4`!opUjJqs7{Ex|7c71lSJZe
z!~Sm5le0?=d0MTa*j;?qvqnadgohpk(7KvB^A)jPxO4|>9!>%NSh0M+>|e4i!GlhE
z0;wG)r?$v$waypXCJ$9)m6UU53GPEM-tcsW=S_Z;$z~5L9wk}<<MkEw<#c$<q(UBZ
zJ<H$4ArHd6CuQ^_^>$-fUAFeD*1JRq!I$w3?5aqgnwo+J4EA_7=WkbJwX_H()9%FC
zpEGi6f~)4uH&XZw(-~r~)8~oPKq0eDaRXo)i2%A#s?t-chz$OLh|7?pfbyB4ocw%M
z<>OcNMMXKv3L1enVICqX&ayf>*MF~due}5^20x8GyV5}po7Z-@x(^L`8(!JDZo^9o
z3is8o3<Nt{@9Y1nOiqByS-(>K_}dd)tj}+$CDK2*vVo>mJ(GfrhKmjO_!K@KuXm<a
zUWy?0Lje0i#J3b)<kfL0D5fg6c^Lu{rJp~yi7k7;8O`s0u-}X3!`}uNVUx9;Myrjq
zK+$Mpm+c;E^Eg>ivE|!l(m`E`kcT?U7p=e)c?s~si%`9_mCrX+qJY<D!jWj0yf#z+
zeFwzOPFdA``62;PQivz_-9+y}B_5||N8Zab)>t2i_>}YcE92Mb`_B>Q*iY;ABdbFy
zn4X(JybU!aNeUh9OYOJn<I=d0oT3<q1<a^h)_!e@IkSWgJxBQy1TY$R3zh#SDN4l5
z3{QnTM-B8^d<)$x(oLqgmAS^fI3BP_@%pR^nz;O(b=~tVHAQ9e^F>yFq`Gd4pNr%)
zuAmqa70Fosr6!Y^ep~4gi*j~;(YUE*|KoFi5^K9(Dv6;Y2A-b5aXH9XND?wqq(`wd
z*bn2oe*XMvzPD+o&G{ISYF^WbmefT>Uad~Yo|ci4(PzX0DAXN6C8eVh_i)1%tn8i4
zmdZ*=;F)4#UIcfBM#^JjVUYvsbqy$h3}6jCWT9z3$VV$=X3@jMIVa~DZ<StKU#Z`L
z>A~aU=^1I01)5Hbn`?fb@(M9=i4#Yi`y8O&nF{{F#s0d~&FyM0`^yVG(6(iy3Aprc
zLdjUPZIRhw3B$TU1q1smLQuipT-Y%8!w39hQ4S|vk=sv+ieoHr^45$#v-w!XMWAs9
zyk|`##2WxGcvCxzix8-rIC8A}PAFs?9KIDn9t-u?5-($AWFxY8A3tpIZqKi?0Tl72
zEW`<dC>WF;hk4=fXRY7tBJ0lM?YxrDeuRG^vESm6VsUY?>-291a_UXb4$F{_C<s7!
zACHVad5QxRgZ%?Ze`Q($5|ZcUP#ZXL2S|_~-}LST$#w6}%GMTDy+rVO9U3H`K7sbi
zcHVnxVd4E?|Bq4;h!>vW&-<JYl9wU}-}~QEXEdLe1`1)G$7MeFZ}3xVkN9$BJk|4O
zksPZ<Ou2=naOw!Lx`{qa+RmY4Jg81F*MIHU1llyf&rkwFgPzWCeQJEGq7xsVR-ViE
z9ctv@&PF%3VS#-YF9VjOTnOl)renat#QIZU7a3MHnp6`37LEW<K0sufaT(Tb>U{-B
zUwj2z<Lz5@I*Gg-m`Y$=tzNRw`rYMDaOayeL2}9<)eiq0g9i`nKhq4YCR%R=-V;VW
zv{joY)l3DUqUOTFf*fT$ltAz!dyV4YP0A%77doK}fkXztRA_L&zGHcEo0tH@l|g~1
zDZ`dGqS@q9M~E;pHy`-fi}P7AR`x5naY3!a-$Ql7A|8OUbj4L2Wd8WgO5JQv-ls44
z_i_qod2^qqa2r%lXGAC9+$&UuC~)YE-@MrdY90w(hH0H9XlM^G@4F!*f#2Qn@DQ|+
zJ{Gul1R=W-@{LDXHVqmO@9IvDhtDv95gub<k&Wm%>*<5pE*TjVowz}`!QzQxXV3J6
zMKYv5vA#CTdbKRpw>eoe5!G%^DTI5&bo<Nf86V<@CH1EfbL^KCx$MRHrBM-N=uh|3
z62BTARAE2c3%$%*Lz$OoZ}vJjFbq6EVYxsuca1CNvE~u^D0T<&G7fMNEZX>lKy3Lm
zk2_ym`^6ozxZ7y%uQ{-Cp}UtxsQMuLn&_}2%))`!6P_L!a?q<~nJpQoM|_G<&D6;%
zemOqvXIA5?WPpsoCjGq67e0XO=PF9_i6kfEUSuH6FLTi4_W!B@gwRrG&g&^dP?E@X
z4`(X)vrz)F^K%HbvVNe;$lP?iW3nnKAlsomvMT*|ejY>g@B>h(fYrNu>jZxY)Jm0q
z@DKxb17K)7>ggsw*c)Q}IhgugcZY`0!T#M@U2GF~9oolJA?{;TBBvmD@(GaWK^>dw
z*?vk~ey17ea=D=&Ib&NOl_VVFM?pdff3w;OX+$L{A2|t6E&mD-3lce@sM*yFjRpmZ
zPY<SIVqIi|Ec5%mzG)PRZiACEBi73-1B1y%Iu2I+6XuK87~#&kf7RbAUk%ib)&tQA
zZd8+as%|QB!%Fqy^MFX9|JMR+wOZdisR=gwj|2U2NuD8B4&7!^*_)EnjfT6xD?3(8
zf+v>Z$v#viod%U!b$j;^MpBgI$9G}{lX+&tJTYqkoaxTb50=*jJw|^C;cvUQ&9doC
zs;=i1W2TH%Q-4OQ!n(UDSxLzRx;YtL3Au!*_(`8d_cU`e%1RHxzzj01`#mkE?jBim
zugXF12LXG_Ga0*2EsAKtv7nu$*BT&a`lGk^Q#svsGq@$eLCeLi`U4pV{F9{PM`c$3
zqK8*DL$0u^4W^eN0+50{@ArY5nkfjbBkK1Sa?y~!PYhL6VPoPhlf@%_N)Zc}m#M2y
zYd!nKkv5RYHQxbndbq~X(oi;x2)e^G{r!H*1JDsQw6#GZ77{|Jqhr^~6yQ}?Cnvge
zTP<bnneY^H1Q<eVm^cRde{OAVhJ;1cCUbM~)7(QWKaG<#8U+Onyz`13rxro+V@z!&
zSw8uk1jeoW_=v?8&mjl~jle<3;B%OO+BDvj;gkN@@bC{$Eed)HcdGrky?=e6)O?Th
z=r`m0E|c)ko9`~SLqdQPoTRx1j4MO7bvMCt?RIxHiMOCF&1!Z-M+{3%z1b7=JrR2(
z{!umNJM=n?_o$HxF~o&r9;H0K9C3}3q0|~mb~*bNzw3!HSeKa@pLPpkQs{O!<lU>M
z{t#VaO#b>cJw2$Xp!nkzb<T(;E9&EpB|8~h%h|y@H6K6zbF{`@Gj=@uKdQbmD$2GC
zcMu6hLIgxWL_|<273prIyBh(K?rs%n5$TfdZbm|oZjf&2?yj?azjfAG$Hfm>z%w&X
z-1pwszT$SDA1(p}w$wqS_sCC8CJIoOjKX2_gR?{vN4z<)TEHzNp($Ag8wbQFK_KL>
z$9jm%h)d$M0(B*=dH0L1lVbEY>Vd0N^V7H<^Q{j+bCK@xme=jz4<0=PLF>Q-&P=^>
z+$h74F*}8)nJPVvf8>e>@GCj%=V5$SHky5%+O{1F{@#__qMjj~tLrokz1yDiIOKk$
z`4FUDIQ>&lk-#Q<f8jKDrFS=lvJr?~zzT-Jex3C>k6*X6q_pshtY`P*Vj>^w%|S^8
zxFo(mfz!E%sl@u?KFz03AK7Xv?hkke1j}%_&fSg>*DD_Tq9e$*vIw%1wfS<~Gn5<T
zp&XsZjfA&e^?yuLM2K~^1xSbukiK$jbZUZG$~o@2Bvaf$Mxgd`#aE!XhA;D@ux{UI
zxD(hz`wGl^4(NFt_Q(Re1x5sR!MbcU29UJtEa$gW-y#<OxZ);l?Q?P1)|}?DCJQ&G
z+nOD~V-=mX^1ZzU2KAN`&J8xPYVsXX5KtZb$aL0TcSJXse?a~lVT2Mukz_f{#4%T^
z%gh%P_CpJQ-ppXB*z9ACPmCXm;*31xk%YJhzh(dt{|#=x*5rJwQvJtn@#aAg2a#lF
zQA8z#JGPyF4EXjJVki;17CJdAlSC60=MB+vE|uHdJ`5O}FoXMhPXU2o$9H*jRs1vM
zwqJ9BmCe_%FcIH?E>VxW1awd6m9*tfPfWgcOLo4!M{_b$75??R4NzYI^EVx_L_uc~
zx4el^jg@geSvZB<Ss|g1U?mNlVO1q9+lw|2>sS5%F`Ig@he*r}O?q@+r}DHQ8t}Rc
zSr?#(5B(CLy<L-%L>8P=@^_te`=sq%ZCK44b=vW-2hS12PUk#(h``Uk46jK+RUoKk
zU*g|i-Y@|a5ly3o#>I6Yr@*)k&}@f_(w`v-94?KODJ3O!b-Tk;em(*|tc3}F9<v8y
z5{mrd0LH#n4-%^V{K3Q{=xW3LA_zw6We0R8!`d|RzmY)U1&e@}(J&^_yyxosM@gq1
z`>@_R6Z$#n%-q}@;$T*A+#}8BeND|WZE*Zp5cM*XQ5(hU@W6zMGVD~s_>nWa8aM)n
z$8E1|fJYeJ)2C{(s$A-xu)1`#A@~IJZ5q0fN&0_*;NAUZe4-CG;PxKOe`+Ui9`moc
zA;$!K{>bzfBYxyFR~>^Bm0N&fJ6vg$*_O?7g4@Dx<)-y>Aa%sfKw%I<b%0mKBP29e
zGz#phFIg+Q2H3LWu(vn*_<(UY+&mzqXrm-<TxU%Ny#=5rAQ*$Y>b+Y}g1P&_$L+gG
zOu&YNV7{4}V_-w9WL%?hk@M=s&%to#GO@KQ(@o3xEWVu#hZM3uTp6E<(K9^H&(2qn
zSDfYeGdR?*ET;_0*YoXNZ2!ov;UXL7((IUi6U*f6)OMzEAlw8`1Vnb2KRQBqm7m$6
zs$E9ZNQ8LsY@XOelokx;AGpRGvgg`I-L<-@q!+Q8qmZ;*TjBiE>T~SrBBcMQ@=O-T
zHEn4Bpi3*`pINjzu#RGl!zU8$dcd_lo0-bE=QHA0C2?Df?A#3<b(TXP4UsmS7Bf|<
zOsG8eqcS$B!INhGv0Xo_*yQJ5`s3X8WQ>+hueZn$RUw)v2&~UthyL8QjQhJ8vbgqo
zL*drTC2O|i>N9*U8oVWuMe>gR#$_@BN+lj%uZW85HxDnfC3MFV67FJ15m2+;NtkJT
zf9v_3CcUDwfhT9-N4@{1HAI0C@!v97_1RwI2JvMMayvYlT&n14ocu0C#lphu`jE3Z
z?sj1F-+KFKo8zl{gio7EReB>-6ccK#n!%vaDVEgt*Q|Aw#U`lr4@7W0FW#)M-XcMK
z3Gs)f>2xK;Q^S>$JhZYzYcz0l|4nJTsa7N^;!#Y{FWtTioz<G^tSYJ#^Od<v&dyF+
zSY5t+*}0oESi8LLdes3uC6N$f>CC}i?Wz*v)<~?(Q1-$xN#mC;V@{_tRken`xj>wY
z5xQMvzYW82&<vqOxubY_(d_Jo`6f=Hdr7FEKH~E5nA<zlB9Gff(=N!)5C8Ukw8oZ~
z@;%P>47;ZBGX$o5Him=i%jlrSM*Zb-3s8t*{0y<2`dI<$F-0LZTAFQX9hw1v@?5uO
zUa1N`WZU7P?0{l@V2sCp18+PEBpvgjM@%tYAn*p(My9b5s9TZVQQ{JvY3Wp@HhX;u
zBG8wG)cQ>hBfClygp)DB-F}PPj=<3R>f1GZa3J=)e$WMrwZwhLGfS68cgr1HiPj$w
z0R;5T=YqXHx9y&-ncKIa0~5y^j7rMN?VX+OL03VNboU9PoE~zQ8zVJ@4I`?eRo9{p
zdUyR1e8w-QKM9{Cuq&%-HukoPQsLt`J@mWcxBMs#whX*abgDgUJJD4#o3T#X7C_L!
z<GxFRaM(N^M;Zjqy?3V!>+9#Asou6q+FrGG0%8Z;0_B#OuFi((@ERRac5i5#p75%v
zN!NOeRTdXNr{fyn-8pJJwT0%+*GCYC>}aggbiI}|N<&@!^z>AdrOGzK>k1PTdLbwA
zV^zb>SDL`;qNSrM1D>8fA)^-dWU>AHJn{;bMfe9S`PEA$;OxYb_-^qy1$=np*^~F5
z@fA~Wg=_r?jM^%O2f^?k>_~g#>;DXibJwq_H{{Q2pQAD)r4O#Qwct?}{k<2SZ!C!<
zb;?dD5to)`S+aFgbt147;KK2{WW0TXJ$sG!Tj<ldBKOJiOS3FxICCZnG^k$1{i9!t
ztmvh%?UEbI_dXZk_36mn)%)EK%cosY(m4>80I9sVxL~UJC}NmK^;U=yXQUbzWLO#f
z>U~3>a!c<;h&z)XEUoyL5}<fDV=dEbzxQK7U6v-_<#0Hg3t!&cd<dX#2DJ!n6g@fu
zKg{W>ta7Ay@R*znYB#(en6O!IOd39W*XzJqPm2Nti>RA4NCoq4)x0MQ_jbdMx9pBN
z95{g2B4RrB)8bIAKKJtbt*NRO;QeKKv_mC`Pedfdx=ckWC#@_Zo}w`6FG<<n-XZoS
z5Cst-z&_{Rur1y&(RU0;kxj|#R-A3&BQIBe2M|#06A?zi7raOU*?mmR9Y{U>VmH>8
zEPE4dIYA!<iDUBrxUBykDjw%wdZqSd|Co_N>gu-<Kz|hS48cHn@bCbH10EtEJrhP_
z=hRp?F$&+u3Lxk5Aa`27!uLDfv$1(qy!-wWSqAs-^FE7fQb)$~87e$Xh^CJlK~mh0
zr)+G5H>Q~ml90RXY7YvkQhvq<ooAH8mhW<K@YG-OZ;jMVFMy6x{tf^gdPR1Grbu;z
zA+-CPJM>d+C~-lm`$gN8DN3clj3|#YQ5viM$@owwh|E%XTX~_ekDI6ouk&~aJxsXw
z_b}ihkZ0#As43A?QU<*dNAU4^dA1Hq@zC^gyYGHzta=4C>7!lt@aT9Fd?Fqp(<0cz
zU@KO+3!}l$?e9{sG)9y<VKAX{vzFbuMMhJz>z4<^Na+Gg<stsxzc44==kX8``&#4D
zjE`h5{TJW9F|oL{>b|BRNftc(x8m2~LA3-W<z>xFNA;gSl9OMt@T{H%a|-zQw5>n&
z=IbM6|5=vT2<oFd=mxKb4$N0D7M{mQQlC3*9wXhj-_TKJr3wnGS;x%#tP&Yrwo7O!
zjaijl4XszInjPIhj?K4_AfStWyiGv|&M)+|t@FE}Q&{Q+)p5c(JHs3#LIJ4>9|pUw
zD=WUOy%~}$kztV>)(Z_VCg|0!Gb$bh8D_qfjqKMbR0Q<x^P)JKwI2J!Q=-Io-e=`H
zZ%uf=%QiDoA6PvA0}!Xvcwlyj-`<;;V*L@$geX$Y`<O-mjsAg&yUuM`bIGY_XboO|
zpz;@cLzY@d!AMI%;bjz@)y8U|1TzPiwmz7Jgfc-5@v;nIk0AbP8J1T8K>_?BE&Wdx
za>@Og9>@@z8nQ<;%|ZD0LfW=44^PP|m4S*ejdxv_T1T5$c>4m#?S?B@?w~WNd#0sV
zXTy&Xs@pXs6$K15th`LL#=SU|*cuLd9OcnG-FZb>R2zaHf$9oRE3_3=q3}_Fp(rfm
zzK8o$DqWV0s7rVb!@&}1HFmv;#D#hj*7PWLh7GrmC~$L{Md9NS1%e2ARpAgs%<G|l
z=%3RI`dY<~rLjKV4x9WN-S1*+mX}!P<W9d+(R2gZelK{E>qtnXnK<+jr$^qex7(<h
zJz9Inh2AAAp^u-Ix?{=*Osuz^o11AKyK?+93f*C6xQjRtSy7c_wx6jo8o6q+-t~tt
zW_VLrx3KGvS4?0<(hM;I*yU}zj{ZD8JgmSsoMUz?bxNLUT^aqIMj(vlj9Y2Hg2ma{
zal_Xes`|u(Lztz<#l*t=@Ue#Ux~6hjMMcYRYA%;En~5d+*53Hc!3THHOC1e3z{C+A
z$J|Ix9uFddq-e0JFPm~lRUt$1{nZt3;qf;psqA~&RomR<gP|OO4l`X;1h4mun&b05
z#o%V$3F6ITCQe+7wZ<HT*VUTX9p<k?yk=!l0(e#N_(fL<$RMu+WnXO9R{OQVsaR`6
z*ONMp)#DkW&o5*bkAtRariySgjV_#c?Ddq;a>&k*-MaI+7nN^ZI(_qK_RGwCm${}1
zop%HVl+fdnX_)mpSL`D{D4;Z`>RnvMpdg>%n%_BW(0#e5SbaHPSvX{ij-ZUpV#wpr
zGS9hLe^U31@kZpVVc8I$$Rh*+8levy?cVBYoE3yH0EpvX=4tS3q`kad{v8ZHv$;O|
z-4$x1EL`T|Aln?2%%ms(kTj#(*BxRLx<gzV(`=+AQCq!HPJ2s~<VAtKX-kCMYZrIw
zi}{A==;)uW#|6A{_Zb~8uL>smbpJ4fO0hVEw6**mIo5v(Gb#XNwf^kf@N2$-J}AY4
zdn@FI7wB2ST~U-PuuPGfjE@)jKrQR~=3!dFnqy0=QQZ@H{8On}D>m4hr~`+fVb2X~
zqo=|NTtt9xowk%PLskEmn?6-l#DU%5^|T&3Nf99c>o{!Cuy%NR(ct^9kmJJuwNB~r
z9GWnnS5i`8bxe!-{&vDmU}-T7f<87~{wCNBRJVxA-oMBqf+!qu`4kY~OrX{J3Q@Lq
zwkwYOVPR%|;*+sV$3|um{1N&B#oNIm>r1R4iW8T3>tJUGS;0FFtTB@K_Xq*UNgZ6B
zq{D;9$C{1Kauar6;kI|LORkzpr732gKBWTUW=$~J>liBP7ah7U!4vr`JA;OS+U0sF
zBEpo#_^RQt94@zX`q#H<8Aj@?^PF(n=!n+O+XtRhrS5{<;RZigOK5CD1qpp5;8`Uy
zBI&Yl&?@t14AcRw|G1h}uhphL>r|p&s{!BafyrCyK&{h!)VtrAz4;(4d_yJQnr(}+
z;hgj`BJvwxKik8Zyf4^+^d|qk5CR_yEmw#{6@@;AhC)705JO*0ZLp&{_AKvhJ-y`n
zxRyuw$+fBTXPcSVz7K<cHGI)iPd&6dOIjG~9PtZvm=3GT)BIq`dTy!mH*>dk$t@|=
z<?UW)_Eopb<>60{d=yWVUzSZr<a^V|!Kt$Kz3XF%6VvANMe)ns-P6|o>kXdPr+JPO
z$?4hFs~fvX^u|eH*NTFG_j0lqI@)M@DUw_p?;5+FCJ{4T_pn`z5QRUgHE-4Kg{EpN
zZkX~ch_!La+!cU_>C!24KrDDUH9RvHxHGsH-2V#q#VtzQ-wk+hWqUF&vBJ1;*lCb@
z{nX5$EPcdqV1-`f_B=v&w=-g-ev*e}ceX{Ypngoe-fM7~VQOm+ZNdFLt9<JEKaM&o
z0$@Md7db|~v3G%%QW*e=SO?Jy=cqe7dV(@ek0vWjJ(V32b4wLOlvx-#`%(|FxwR62
zvFiEp2T(2oKjqn6JX5fUaizO+-wu6Vw0!Gs_ZyNZe%+VRlY%zet*D?=XH{XYeUuC?
zFG#he?F3)l*D-CJ_PqW5{b6w`efd<i#{HvzK3Y7k@$NP{9`%B13)t1%$`yZ+;k=EZ
znAGU0%j4yhju(pgD7vs>!YJU9<~?NT#vEPFJwlwdE7(;ugib{E=(_1(5d}6+h);>p
zQgL<!S~P95N<AqaGqaenZ#q^xRyLhJJuwecrBe=7yU1FQq~1Tj8z`%ygUhQ+eSOQ=
z7L}Oq8kpCHh8|N6?Vbd+8(%#RvW}&O3e(S6*AGYb&&ts+(0<cUx3`NQx6UbEckcYZ
ztGz3uxOt!QI{R~g49sxf0w?jxN$^tbVodvrrlqT%3Fo@J`Q{n10dG(PbL54>_C>-s
zM+w*Ut){Y|JRp{Vi{da-#^o>c)1iv9Ce-;}k8IDJkI)K=RA!VjO&o{p|G>|7WFil<
zPGk4Y_pFb&dU(Mg3aLGiGUBFNDGGDBB<{a9Vc+)*mw9>(L$IKRA7m;8QNbG1TYkA+
z<W+SaIn45J!EXLW(|BbmA9?M-4|+3{3r`fH@TbooA5}`cg703zFTCSuKXK)sMY$l<
zeFe~#(1LaK2`}aYlWcQ~^=gk2_^WAf?-!$z4-LE%CPon^<|*e+{Fs`B=P9$6go45T
z*-)dvrMw3jAMaz)2=`a#)1MtD9E8ah5XL{_=lD~#Zu{(cA~i3Yw{3t|qA6eD<XoTj
z*@a(7(N0KlMmf70=6lDX6`ct5K2C@QaNg9$)kXZ1_usz#DM^gtVpIN(F2eK9Pvr@W
zosnc#n-k;@KTjlqd&6(1iy4MS#FfP8VzKiGF8RZ8<fUa44q_HaxSmMN>nOdvmpmTU
zY$YN=dH$DvI~Z)SC~zn^Gy;7mR6tr}&GNq9|GtG2b7!{x<24!QpxQ>gme;fwq(snV
z-p1UyzRJD2_Wu5TK~W*RUMt{Q0g5c)MEz^2py5Z}Zmbu6FMF)YU|s}2L-DGriUvdu
zB_{A%f?{?Tm8=`_@01^4q6hZ(P4`J!UR|nS|Mv#2Z(MPQh=4`{7>AmbfdJD;5Os@&
ztb5#~44bVp$$uZE(~fe0v6j<O&?5~W7J(qXV*~;|cokj*<ep%-A}ze%y>9>IvGvHg
zz98fX<eL`WEKHfCp^wfhbFXZR-X!U>oiRFPp4WS)2X0<}*Ft~rygeP~i0bvK<%kKK
zPcs}1b@2<!DE)tD;CfrPB!)S^9KJTY96%(mwql7<I$t?+xsLh#&OP@0cOJb<$2!b#
znsgD@SSQD9BoGq>2L_#}kS7oa)M`ex`L&NME8(Vl{;aJtCQj;XBK-sv)BTZN<$JYC
zK5TntgH+z8<Y?kh6%~a`9<sp8PNHT{B61|I_D_<NN{QX1o8DeP>fe$%bb|iX6xqz}
zf=AlmNpO=Z@im1!$G<CNIWHkq^RX}ozLseoA)GXb1Fm?w-A^k6S^oFV9Cr%GQ=iio
zevSO5s4SnAmq*TNqzKeI<HtKEr?$p1#KBNp0~xzVrv?<l<MjDnh-)wufIh2Dy(q;e
z-nVuN7}nql3B!qxA3avMP05HjEqmy~{qJEqT`Zsc_9ArzUd}tiG}b(89zX&iCrG$7
z`a=%8)7Rc?Tf`4;A=_VE8qVxvIbbK_NdB9KDwPd}Jc`Q7#)}IUnn-aWRW_=PCYXmt
z4rV|`(ur1GPq+2v+9HJ106o){F0^`#>aF!Iz?6IF7LGGJJMDqcsqwGn1TOUg%`B(T
z(&^+t7$|Q9zZ{0|K&tFL((Q_0!tw%t6(l!6tcAiSCff(9uUY;N5QY9c+vIW;i(V=i
z&7m}cfDjZHmjz%ulbK=#sp_Bz0?|lWSs7#?x$bdce%hWYUIb9{ff_sFrM!wUY(d(s
z(kKn$nayA_2^lfeBJ<#8-gBnKZw=xAWS>VtR@Rr}WbS`P`A);zFIXMfVdzlM1Jj5r
zvZzsGIf!Yp_rpbc%=awWGpv2+SR_<}d600Zk!rMBv`jMQPG*LoHr7wAQ&iqb#E*=e
zpD-DE=*sPkx;(1dzZyi->zg6NbPsP9_ZF7p%*qS;rxD-x7FTZRAo?C<n;M&9BP6M9
zCkhIUYJq6JX{6C9CXqazo?&=`F)b4M8GEQCiB&^exuPWJ3;@nsZ=ctWfadK|wW(i9
zN(#i}h<~`Jzm_}jgJ3XEDnmTcH~Q~mjI>t*Yzy8R!1^U3B0L8;Iw5?d6LP!;uCyi!
z&?aqEb!ixBpxtzV-;UJ*`mzkMu@&G%E#<jo=(uo=G?8PzoIJ*=k}!N7U@E;twTy~X
zg5irh=al8N^Vdn{29dS3u8PDu)B5{?&FZ!K{vdZi2zdl~T+n@q<BSkcd(Ug!^^BV+
zkPsqVoz}?{e(^O?)6%dWxZgyGNxTJp(73@d%7W{3pYHeC1Z-ixA7v19yaz@}dGjeB
z>d<*6kVU@-rnGK9qoig*cW|U}(8s_GcsQzsNkn;@GT<RN`_dJewbpJwx__i5r$X*d
zet68Ye|gfnHR;c^+@lq5+Z%72UY4g>cjLjGAv9YfRkc|E=5L~*sZYpxn{E6o)sd91
z&mL41nVaWPhdiuqvaAnF{`G!4nTw}hL*@RpyI-aZd0}d5t*fQ1u;HP{^~3(&M41G(
zt&8;a$Vj7FS36$krN-71Dg+-5jo;WHO?Gbf(D0B}_1zcls-Io4<z;_xhq~5!9KL>N
zzPa${05aH*R<?#yT`S!&Lf<}(=P~4p>O|V|3iLEp+0Btrh7ng8#|imBW%<zKyjTK&
z3Z_gco1)SsD`z>Gq!}zvDjiayr3|d)p5Hv%I}?6W?{APlDl#^e+~{n-*Ama;+5Bs|
zrKxs(cXl6c6!^`-Oi9dI(mFVHP0HWW`;3e_Sp0GXh?h21_x*+34wo)#lsduW{@yov
z9H8!udlZ}>ZNV-HmO+J@^;2E?4<3-5&?PR{Ut1X%++zXG>0_{EZT0CzL2$5-%fzrL
z%cx}5WGs#6bIPL$aD-G4*`;{%QFz`Kr#hQY$>elB6Co0z4O^j4>D9hP@P>$pIVx-L
z4~HrH+x-z1u%q@dVtyeQkrOMIq|{uXoSDsT>iDgq!=jDeGP^H*eKdQb&8Svb?oH`q
zWPnzFVVZHS1wE(iQ{;>K;y#h?HHlkt1X9^jxy#JaS;T5Dy9^nNKW+SuC8&!oOwHO%
zTiij~gMS)@&YVhXs(-NS)J$ompX<oN_cz6=L4?v|tRJdIW^4IHhb^ba#@eTitzDDQ
zUi%Xixzffp{krqUPPn=0{8E*kp2w6Wfy32gDkr^`YjQCZsk;@_|HN-|+EuqZV-qgX
z1ZyGMKlSkX`YO1fBmS8a8A&C595cOxeSoul3}aXAiVyQmqH^b#6&lfT`a@z=16GCn
zt`+BXZkJbmy9&Fv+7XZrG?^XVV)DL>T4ZQ&=mD#Xkdbl^>tRTf>(M102X}<{F3F#y
zZX;8Z4`kf4NeOZF?-tKYD+Mae)dy=eqAJwc3V&UqI#^zQ|LV4*x|3TWscNcBH#|5r
z{j-Ue>u>wQtr^F|H?QBYxis*yGaKY@hn(X+IX<`Cp3bL5@Og`H;xDu=?z$;kauB&$
zOy^~%*81s5)l@ksE17rS%1ka;n#?{djLi>TTw4qTg*rI1D2?>Zju26EGjSeoB16qs
z&d2kMOe-4`c>==;`S|&NHPz;382LFo)MV&&U76L=VEr>Z^pM;6NZV;08QVWC7)2Lv
zyRpEY#9SwkY*jzwK)D;oB1)9oTq+|2I+-znULNjSH@Of9e!N2d%j3<l$k;^Zm6PY$
zhWSNV3-DUFN%%GH9v{JrWm7@l2DS8V+!V-W(gV1EdiJ);_)DmmI~^>T^i4mEUKGFq
z3xR?{OEEF==9(7_^x@ijh3_q-{x27RoTI9uz~eGf3kaoGoU8`dhS2#e)Z6aY?uf!+
z#NKwYmCVQl|N1(}>+^$rODl>Y798@+r6w*m%*`{TNJ6#rZP})qYAY<YTlLGyuy`EL
zw@~xk+J1QCMltf(UIp~7!BX9gal=9G89}n_7mBvg{wHbNGKnu^TjwRH@DUAz#?~Vi
zpYCFQbl9J&bx+MM?mz41bUfKypPQ=~Fn+9TVQEUwy)2g6&oznsYvqLS`sW}NoD2JZ
zoy&a1ea4x1ZtJfCr#;>YK&RSw`iULqh<5WKxOB`$hvm}_SC^)KR?jse&UvAw$$<(2
z%J<Vv=&WyEf+#AKe8&@S_haW`Ffj))qOVrtJ_hFm=kcSn0-bgv+iB0BBo=2b1fQVb
zWnbMCqjsHIxrc;{$dPv~Cn3T6{2JlA|8Bd9NNgjUj<ycCMRiCd#Q42gOvlr}Seab`
z9IXLVLqSYpR=aL0tB%(Csj^3iv*RxSF5otNLe6e&v{dFl-|EykB4Tg9S+1cclal0I
z<@gl;i@8k-`Ig17GSPA46S&Kp-m(0nag-`3%C1ZxTTl9*9Ni?t<Z?3mJ1Xby;?Sy|
zj5s8j<kjsRPWxUa*ecUp`$21=Gpg%&W6CyDr;lUiA^MpM_%_`@e2-9xjv>Z=FuKMn
z+AzP|6)mr%g@t&@{EDZ{@;J`%QDnh59^wAqhUo;6&(}=;xYv+B=9664vmFV3o_y-o
z!wvz3(q?~Ig$cQB2TGzMS$Cq{Epf`(mbqTqeeaBzwou`w9a9=jk5fxtGUhz58!um7
z#PJK<LKPGZu9|QO_!*)xGk?Z6fu)7<ImD)FWqMP+J|I$jLtajycz-cb)g6gt;cF54
z4Lzayo5$cnnP8HJAbl@8FN4bqT5WZ$&)v={%e1xVpF>KxmS1oEvP%~FqE!<|YT?k5
z-J756+tgxLGnwkyhDjZIiX8g40B)Y9vyhPC9KT^Ha17mYcRh9@A@!pv3%Q~?8w)6}
zgZT%{J(C@q9&YepaR#*Bt2iylFEl1+!w-t6a$Kh*3j8mJ^iO<bnEn0{=o-;G<EY5D
zxy{7-=I()p7JMVmamd$)hm<4~P!V|KyG~Ujd}hZJ&FSHh87--;{Rl69R4#=<rme})
z_eRCqKV3+V&l2p;`~UpeSQ}Vv^B>eZZoC&R51k(hLb&<q+xF~fbNxGyI94{#Wrk)P
z*1@KXh2{0{`CSVw&5)a?AYF_mCcX8kF`#43!`*f0&z~vG>cN4=E8?Vsg;m>Z!tCto
zvHt0!j>E5z(m_p~u3wn?mP)^ud@HJ88Zuzs_I8ChT@bk8!n(rry0mHQ6Xb*PxD!>H
ztMH+QgHCPzuViY!Pp_ug2P4CGi<5Qdv&dqbr}%H(1Z>_ZN=iv>4v|*cmz5V0_X1m0
z_-VR8@-s-Vhp)m}CY{iZoVD^ehu@A3{^*e0-6gNA+|kK+hsSkQL0}Jwn^d$H)>PXQ
zU;iy80qBozbSLo{88r#PLzNuaxt6-6>@6)KFFHCzeJ)<6CRS%d7kn#2RRXKjbuP4E
z8dFMATuRd7Fh-w+B_l1vb2STYT?K9JUyIhoV}yiJ`is68F4WYS(^#SJiq#WaYdtQv
z=D&K+_|L@0Bs48qftO!#0SK1Uqj;#Pgt<Zrva@R)m=qyQL0aV|B8ZsdMDN|7&XH}^
zyvRP}++svTt!oFfY|1kP#YKNO<my4f$=9OuR1v4QDz86D!e4WGtX;I8q>Y$U*UL54
zGH6%Fe2eV=)5T%G^*TblxlxS!i_<`+v7@2kQ-oJVWCl_r>Ks>N)6g01)AQ_ANu$(_
zwlkL+Ev*FHhnz|Z8Y=3_)?0tef1z?Mj`kN(3nXX#D5`>~A2o}`O4i#OzZ^rV*4Niz
zZuE!sWnh3rRDmiZvz1#iIsw9q0y9F~o%lgQeC+gOm$>*_iQX=G6%|y37#<bN)?wzy
zcZP=PUt0U$K3^qSTSFj-at(hC4E@^_EOoW($OV#tvmsT)6BnnZOtj?q$gRQ*&QhvD
zVyFp|1<uAuW!X<k`X>@S1aM*?vjwJ8>Mr6RIF7<^AXfekKuq%L?^P^V;UNg}vEH#C
z_G!<4+me+$+|fLhI0m+%?z=A@KNnV0d;WKKaN&;~wmrIozZivPjRQn8aBx4NeX(@f
zR$*4v-AzYI+X-iQ#PZWqxNXVx-r<{6jWj?UK~!V}$RWT-YB;?#ydZ=nAZ0EzMdZ%I
zkL8C}jtVKXJpLZ&h#itY+fA9jcSPmrlqDq1CjDe?`C=TPeXU0&_Zu7<zL=8dqrelR
z?NOJ0hPaC+#~yr|P#&MQ$V5*6h1PR^!dz-_Z=h2<6W{dPGLPD4t+lj8Ohi>2Df!7<
zADN%YRc!^mquQVG7dCSf9ig@lGmX1X%Nk6EGuM6_zA^<f!GMq|&R6LSm%qJ!EG-(M
z)H-s}JJ)SBE5js;a(u@o4)gA#)%z#!&E5;~e^*2g6Ivi-VW}*h9sYtP)hRJzujN6*
z>UKQgK05SNM#WZW(ow-6Cu+fby>3pb`F!yCea<(=pujo(bn!=x24r8W=6VWA^-h;-
zo35vax18J{q=P)Pw6u79uF`WWFI%<xCclvs$?}4(DCON%J7Jc%ClejUyFYuRah}jf
zwuZdaozKb1A>$l%u@DaaMG?Me_)^Xw>jf<>eLSba3;KYRB_nEhy2JW=iMo@9#@Zi0
ze`@}_?^o~q^c&_<1R|qFBy!1!1_wj`5YiUY&aDX0fV8JWK5*-nYmxQlUZ;vB3gR>N
zgDRR?FS4)ziOy&5N?<XxD)fF9aI`kpvj{f7;&*SpVq#_pv(9)1nHntbyc8kn<R0m<
zdIhZ%HumipP3``(9{w|a9HJ%afmH<J<u(<<E8E}NP>Ili1PvjpuB@m3?qrBuw(Jnq
zheBTAyBO6CgdNq^bHApDSoJ!r07`@x+zXo>Qc}UKAtbMU=sen>{Upi4%G50Vx60l+
z>~mxFLkE}*DJg2|n`K$P`{9?e^0LrdjlJ5gMqf|wRfcM@5dJAuhA*W%7SlZgQ(COm
zlKzn)xgY%=Rjk?ipBn24O!f7%Bs1gUV#;umGifjB{2Ej6Hv|op%1cV{ar^}32+Wyg
z%2e|Fi^2F<Ce&h7MOYXO%XTny*Z<M{JyMeB%k!hU`&aKrzj~v6YL+)`tSkc$e;D5v
z6{Z_5D!x$CkVs7_IioFyo@Hn2GL!%E{Ij2y#9N8$o#(s!zYQ3oXgojANb9Ip+N^%}
z2R{;da&G5Xm^Ti6c*ci<0B4=vWvovE1j5x^JP3raELnqIlxiN4Bn=n&BVoBNJ6e2y
zy6*1beDw^W%+zN-fU21DjAHr(JpI-xw;3hXmT(T^bf}06K6R77GTWK5OtSeFn(Z&A
zWFP!TTuEKoLf`DfK+FFrzOn?IZV-;&ym6BsrJ?*edBhb7v;>^&qvhQ=cQHTpzhZoM
z4$yIi?6c;l`G>SGa`A4yVtJjFQ{eAJ#_c$)2#iU{{Di%gH#pe9mTi<SkIs<%Hn&!`
zJ^27X9)Vugi)%MtC`9%`bu&J6{K!g_u0|lNOgIY_d(EBX2?QJFqN?feo_y0=^3}4B
zPYuu~ISsz^LNF{1fykO0K0O>?Tdz|dS3(VapmlHugHLvbJ4bQo(%o^wbQInN!t48Z
zS<q)QZsCMm!v;cMoDv7Snr_hu+HNJ0IqjIV?|+caFG#5f?Gzxxsm$rbPcmLF#+mQX
zRw>FQW8Od~b`7}O&QVDyiti&>vc0iYj)rI;vi^ua=a=Ke{KCLBG?>%H>>IZV+P~fq
zlh^RSjCi$t&Xk(ZpSOowj}F1fD}9i#_z6BS)Zdk5Medc`*?Boozc&2wH!#$ImfF~8
zF;BIm!Sh-#Plz%SfGC5zg6b0QT+Z<~WQP__E5+aGYv>xs!UDZ)ZTb1m2C&O!22&?G
z>^0J)0J`YaJfW?-6WVH-fEM8PfB^%sb7cUn_xjc}p)j~<L%KbsC>4(mMB(OVX3nqZ
zE+eNb=O!OOtUcV92t=`}K>EcY25M+)VDq<%WI2XVKXGg-^G_u@yAMwco`g~8T*gsF
zelfId1M+LJ(1js07YZY`ML-IIKWsbXgh}__h5Ryp@yS(F+sCx}gUP<7*8b5^Dy<m$
zjbVd<PtEDEu_P?Yk;{x)GsT9rVPX7QTHFrs&QNqH_&ppb^^4h<xz{h!>;Eo^M|Fov
zUYPEQ{r;w6Im7$qq_b!stgpy(%7P4s|EqE)zGSfuG5>G;U8uV(4TIjRe!bhJ(r9W`
z4+RAJ!TLk|qgl@JuV1WRciswZMMb0qs@-;&yyFM|{Ycrr^zaH&U9~hhU5~cd4I86Z
z_ai&bh}_?909$d|-EM1PcDQ*_ceWr<#bUZ{Q2XVLL@#<X0^z%>`+KLpsImm+8to;$
ze`dWCfpKy$wH1{ZLr-vfKE;qz^pxu=sxt#d;Pg*S`@HI&n>>KL;QpjX2kUJ5#W!j0
z2@{7ld1lv!_@rR^_%M1fjC(iDrDa%o0p--sjfc};h^|&GU%vd3kNG93PW|O9bC##G
zuwi8Z16E|dNx~<gZ}Hz8vgvj2K`h8U+-A>v_RQ$yxY!zam<uxeL0>SU53oC{w}0q$
z{wOojq5r@jg19ZK$aqqk^(x!`f%_fTzbNDgzNfNonnWGi-IjDhh#W~}`Scv<6(GA^
zOXJdXFYY#QTH-%+-%DCldS)1*Cz@sUC$H#wT9@Tv(9aeK9#?nIXt5rwza4jbOn@ap
z5_hH6pnJcY)_}F(Cg~x8{gPvBc}=Ih;q5m*f)rnQ1CvVnp&#~}!n#K0*V$DTSC&tg
znd2Tg)Ru<(9-uL{bWb>fEIPfkwHE!*lk5^qCBq_^^d_NRqOSNgJTiC}Q`klX3RG|&
zVIt09)UHKC;O#hRkkYO$o$~*lcfXoY7+?a)QrfO82nm1sVLyLP-&*l3gSv&xA~zUe
zV$1g)BOs-6n~V6B-Af-zz&gLx-v_k-VHD2JMnY_2Ohf{>N;jFAb#!-QAl?}*cE$4e
zd$k(4@`}1nbZrcWV+~B%Bax6cO4rp5ATa`gV_x`c@oP-P;7lp&k|YWun$z^dTN7tV
zGfffYo}lb|5i53*#UV!(17A5P!=i_m@b&d9sHa}GRbslE>ip8p!XbUVGvy}}ttQTs
z?<&#f@fQHxSdumQ`v?rqxPg+}8@Af~ddNW~gcq1<aFUub-AfJ1i%(l9LrHW8epb4m
z+p&kcp+nVCX_J1KSX6LK_O42|^Fx(CQSaU@uB@(!8=IROdOOp$wRSY#!2-KN3u^mb
zlX4XISG&O82SfN(yrYU?`Q<!_9Pf!?E|+1%#l}8diC*5rfH1|U;cuZ89qkJ_a=@0P
zHX<bq5PCyGnEP4Sg>rTmfiO3x#}9*a?g!+%R9t`Gwhh^aI8@k>9f(n-Bm~=wtGAqo
z-rDA+{;r~Uv_64V+W$p}GKSDJz(jR=y~I+pq_SLE{B4t@H=LfEH&AHiMfVp=eq-K3
z)>~~!J`wc&(zdcqKJv|{^&aA+^F?e-1e4{kBB9};5fK(*c2<{`=J$}}$<c2x3Qezm
zaP(7E?V_9u?}2UP_l}&{C8NmHT6&_SiC;bF{CFmB)-mqNRGb&kKcnmEWc*HKWVCp&
zIh9(KcE|O1SGOeF#&)6V7XmUI4Bvh}xKD4B;joUTdA{$t<fpz&^St-`=52&m5KV33
zUd()HL@e6-^YdJO1r<u#cI~HkzaS8I_ad%a3?7oul74p4iKHJ<?xUQ=1Usw&5<`)r
zLc&jCC<wa&J2uXfEHl~{VYd;gOC?r64_7qC9H&)0T;vH#_s#j9=Wh6nw9n+r{VVr^
zbYe%cxy^aAp|+kEah0GF3*EAKLm~x*L3paWxccWZeRot_nFep}33sq~36$u7A|N18
z!qH6aKE8vFpqBgehRJ<~nJ-Rc^u}4Rj-AoP(%@l&{YPsMSwV6^MV~Q!g|V2lRAHep
zLp0O;UOAr0<x2GZ@Rs25qH&tDF)UlA@COPs788VhBpC}6ai=u>y-U4`pJ`@~Z<1YX
zZGioAYY<7D<MmwxpPT+b!<Bon;Jsd*x<nqsi`WQ>RQ*IK0%~gP)Qjxw<67!omkCii
z1Mb>b%|wM?ToGxc64f(vI1%J@Vb|5Bx2{a@etMQ*OVg?+h+Z&WMEK8RP)=5M0jamR
zS%g8|^7+YU?CC0d)?uor%<cfwxz#m{%GL+kW`>LsKzBDZL~Ls;m0yrOB&&dQUq%U?
z`EY+wRz-u$lc=<~_z4TiPO$k-#QI|tWaaSi(EX_IhX9WZmG1{tOE}LQWxb$#5yNZj
zBDo=PlV7{ueH*Dabh#tOF8EbUh%PqIyN=wOufeC^XmLR=)1-qoM=~flLyCOn&!0Tt
z3e0t2o1sV&bJ|Sh7|IyvsKV$h?zP@N&h@rYF^savwwSZBt79Z7%Zu1w74J<lG;V!L
z!sU7Gc~tgsL2+3+1OP{R?YfG*3r>tlkvpVaOIFR4yWjl{3a0LuC+xEl?G2G8`yKx)
z3wjuztxc7>CM8Z~_G-PrFGfKeVt27x^5*TS5gfizc44Kb?;_hYM?tiRpbI49Df|gv
z8O<;n_WeS$uEN-)xlGg*<FV*sN?H27?7$?vWqFZM)<khzPmnJzqN6&zX=7EPNBgI`
z%FOJ0%UmIquCJyMu*w0*8weQk{#AGhzI)XOg~|Jg?`1)1gEX&5NdXIRIkak(4!HD^
zuvK=L)=lz&x;91jxx0%Cw5rJJY9gYjva+JZ(Zgx?FFM6`zrG~<X7p+egvo(k*_o@Z
z_pfuz7<<l(7XYxq2Lw@6M^F32Praf1KG)CaXQ`rhrb-5NJb0&@{P1qwBs4TcH0&h%
z0#OFmR{{BecoX1EquFNqO}l4hXK#&VW&+@7-Z{S>xsuB&c)vI9IgG<8`%|1tw?2uA
zYYp+V(Cj(=9Al#HbfMsJA!Bm*^M#o<@^?#ZNkv&Shsk6MJ#r!Gn{S%{75;&p?n@G=
zIx)Ok=Q2k_woOMB6!_<b)z(qmII&5YU--6Le*f_WsjY%&n3}YD?Y)t<iWt{6y}Pp+
z+BA_np&yex{k?u$Z&tfcb;Tt;ltGKrphpz<(!$9rS6{{S`FoOF;>XWyo-ir;W{5nM
z-~3~aK&;xUkv$d-%R9;D;2+mxM8!q$S=(mE<YaS5Nl_4ed^f`^JMiSMfAB?1^tbzK
zisgh*j#M!)SZ2TEGCWRxp{<+<Pv{zHbghdWD~r$E=3~4zOJCg&A2@86x<V9duU~!}
zi&QzpfaDCyFjW?&H~vB$J>8~8<v-o_rw1ClS~~tNt~;t&PNK8Lhb4RH+q}S;{URhF
zAb(qj0;oG5-kU)%A;={K{viwHDLy`at}4vT<)CN!>3*FsOATlm7)%06Y@SE2F!<UB
zC+Yj)Twz|$k`<dHxsdLdCko<*_bjAr0UT+e75|(-oUc#L#E6K?yKl1VYz@o2Sk=tX
zFf1XQX6jjochlmo^Tn68_r<E3Dl-%Wn2;>}1P@b~Qf?z~MC9|?)Rg8^{CC&M^H0<B
zMa!(5prn<86dA2^1p!c4+JtURow)rv?Hob7k$myJmy3sYshL*L-Cbo_8vj;_>o$s6
zcxcMXew3x+_Xt-9M+I3~HkZFaZ-z|-^aG$_qY*vd3&>foZ2WlRv60Y|yKg@UKLuMn
z!~`PsJ+Qmi$NbQZjFdydYZE@#xG)7qM20;d0uocymu7(Q4zV9%XEJUo9~HO<$=Iy_
z-Ox4Gkdv$CoJAxt_Ud_e&jG!vEsL1Cv>1%<1s=E7y7;j_nzM4M^O})10Y?@UT*sOe
zN*{{V2Woc0!Zj)S?t-W+V*sv!>%J(bhP}OrJL<g&+dVWN<6a%`bE&T82ya{P`tjrx
zZ4ckT!0qJwkF>P+CKee#_&Pfof1lBWi2+m>ZTrATO-xMkZeTNDBxaMQoEoN4J<>P4
zbu&F>&CP4$dmv#jZ#BGP8^_M3Y<=PV^}^~t`1Xrc&Ny-L@VpWx@y1NesMrEzvVb8W
zo@@iyV~%GSZxh-Oh{Xt{FG4oQ{8YK=%#}XL-gWuqPqpNVb80dw3dj`{)m<WGsm^wu
z%Qf6TSY1rcO?BT?j($`1Rz##ih|=1Aiz-qq1T6{mS!{It3nB?Hq!RD!a0PUys9Xf^
zx8=;^Q(7Uq$lt$P)HKxW*8iq}55<mk*#Uv7zI@J&;HAm-;(|oyxizU`)%fW6fTlSr
ze5zOMgrTjaWAS+#nm4><Rc2%%KZrU!Tz=rh4Fv}SXzLKB6xu5Ts}1e5cctuFnh8(!
z+rz?*OKtb^=(__0I%DJ1*vLY9@4Kz5ftA+mE}A0;QIH4)-p|XO_URrGQ4IwOf{YB~
zz0(^6!r+#XnkKY8HWe(b!bRl9@RV85gO{7TCi@xfUH{*I4kT}Xl2lQAS6&2LG(-l$
zyXJd#^BGA}G<4n0T-qW=6%~<7U%f9!PPfQ7-g)^K&y59+q@k2rvvGy?N?bKIMkj_T
zDJc!+W+JO&-oE=Ys8PAyp>7MCo`(A9zw$MacX9XhRj*Dg@dF3@O-U;J%j1U(bPTIs
zq_3W4DSklCmu=!nP(5E|=bcqOX7o;ZFXO1vh>Nz{-c9?p<%T2Np`$2_5>Ru531Vj+
zDRz8H&Omh~wo^%c^-M^df$7@FVbp0|E+qDd4WTctEkxM-21}JDyl2Y^S(!NTvQzH`
zj$iKlFa}MMN_tk_#@<*i7G+}FeXPG7Grzz8%$gsz^eE5$rm#6*QdB6%h=!jZsc~HQ
zf+uI?LHNwd^%bA&#VeAqHFI`F4SpuNWe;}b>xwrF#ng))+CqU}^JPX7mft8OWJHdA
zO;C-)SGY(CQK>vUOf-0#dAM408GK&yiL9`wSmm|F(Egf&>0l^LV!P>!_z$dlUse&>
zT;3QVydfS-lVk0FMpqnZNpE;x@Xi^yr5qeC{J3%lenb7l&E?hA2c)DEVbjozN9xEP
zAeT=>Q!{gImwx{;%w-{n`hMwViPzRENFA0aI^~DE%_yR#w%VBe#P#@5Sa8$<#@_{9
z#|v4$ZSQH`nW5|3IXN#vx2C@;cr)T@RuhI?*xBWH-X=6WTPSYw$lGjc%kui6Y<ZoI
zwJaHC8t((O=JvN4)tBeH-;}f>LyA9KHtkmYZTehgj&!j&db1XsvXXwVzAVoeRn+ml
zT^x&}`^Z}D%V`e*Ghy!+&8*a{ha7w2NFF%?Y-t|HSDP+Rk>a5RoP&CM^2=zW6VG2!
z^2A!LsT+yWS4t;jR7{&U6HMG1P47C}AS@;=A+_N?K+@YM)}IbB(0*=qF1&|8)J1#F
z4i<miKn7n>n+xaXPmP3@nOzcgo6@OX>zhKeCq_B&BRjPGNPEJbJ%%sx_C2Q^Rcq_j
z3-&=RFiO&S2w2Sk3OvwALKocgpEI6FKjegC+{JLI^mQnh{lIW`tG{*v%B%N3ye~{H
z*ridf(TKX^R>v}pZN$$WS^W6rMLUn%%6A_<^&e_-o9NEIN|ImPZr?UF{jWKz(ubMd
z4)I5Zf(lCE4%q3lR#i`X_M_N(ulMlp%4km6kF6DS*8O~!skd*}s3Uh_$eD?s9fvHO
zm~@|t%BOp<bS%=zG-#>1xqn`U1;@|lBkkrz4M@peRr?23HLdVb)ITg1`g`9gtD@j)
z8R=v9H81{W+V=xM?&>wZZYwCuRaesMkT2kV6rx;PHMz8S4n@*Un2%A^<GfEjHRh9h
zvu5HK62wnYkAIV&GF9KBlHi)213zR97t*p;-@dt+JF^WF!+S^{JVTH6-ycG=KC=5n
z4-*hT`2J$zYde)N+W7ybZvPIxw|=5V8jPP%=TY7pVM|jG&sp5<NUzK`?mBuH_-lj_
zol))OxueVuZFF3WW<Bvms3WJzWEP>s{fn*kOC$sat{v>6jxgLz;_=`UJ!f^Ab+q4f
zJ7`cDz=aMP{yaA)M@(YJ=>iT0gQoEzDt!ML@dhs2@wEK7;R}gxz407(U7uyhJh<Lo
z1&iJc<H>Jmr7Da%lLdI8tp;zi>u<VBzIpTKn%m?t-WNB!^5lEi|NC5Ro`Ff+>kFk$
ziK)H`I<x!?p`w&wMU5-Qy)!@0Upg-R;kES@r4E<6`V{1H=W|ZR()IGSjBAyn7MJJn
z^p8~VoO|=4ygKG^-B4Cm_T)p&YuK*z&4j+cV*T;#?B(m4oSGV~?kD^nFok1$ekrcG
z3{<PVfDX^MFmEE~4)LhyFb*T}-Q8YYhR2n5*g}tm#l!je^QQ@7W|koj0C-L;*jn9>
zs?UQ`{^I3p7)oEd4hT@-u|u(m%1!)|_}^uPt5e?`I`LjnN$LAaKnRY5+^gblFvs+2
z{_u;znoSeOuXAx%u`TUr{;ofyJ>0y(LrWXgx{Qi|Q}Rh5xs3j^cb&GmR;}!_?@zpn
z!;&L2Tp&KrjQpt}YHiM%x{?y%y(dirB{pNS*|6V3(*?4dkS=N9m(+QiSek-s8F-4A
znb9%5wuKhiK}knOMMV?|W{euALGABKDk8xlEkPC#aNw$wno0<R1L=i;q{@u2hK7WS
zMqtz8(&Bynr)*?Bd^$jP6HyV>kWry2{80qfB^jOv2mhA~psu5)*525qNErt99{RrG
zfY2)&yaYD;!~=?hL%Rbzd-#86N4-Jx57}qxv1?N`HAY6uMn;CH2=!6QjUyDqj?>SC
zV8TxVu&=z=D=ikd;r$~sS4&$?I9)a`pKj;hq@tj#y|Qm)uz#rE#KLItK2duHrT$Ar
zNNcYD9xk4)s-X^D16mRzp&%+Zq5&q-BqUcQ%Fu*?52L9Bmexc1n_&n9G+;$?Ma`NT
zDk{p_{T7%v$Sp061u4TyY;3AE6BWq`_|-?rFkCglWWPs*)6(XbM|>*5uPysoB4Sir
zUNlB;cI3!Rfdz8O@`^P1?2-T_Ib~`*Oi0baabS5CVDxTL_pJ|{Sns+5|KhN^#E=w9
zYy^c;R1JwBzqhis+^>18@%&d&Jqb>3<QWAmEk-PUsN>3v8GKjEQZJ|0-kT~Z4?(_P
zdn7Pc0Dajxxrm><2<g3h!`tm1nY4<=z9SaOf43Dh`RNx`!A7IAD5z(L{r$sx<Sc12
zGPsC_){kkkO!pVx6c?9Z9!w-g={&>3)U4p#yv230G}X$^ltF=sZ+G&sEEYe3+~#m2
z`fVwgwbTs_eP-_Do<ve-W=xE!E-n$8P4~N8!R2RCvHUnbp|<aIu{Bkb2Xd*))|&_~
z-`Y(eGr)8jM#O@s`^ZvGV90Q?>cCL6`NQ0v6Sc_a=5#|1hZ9u9zvUUG-fsFOpQ=qz
zORX1c=<)p*m=f^)p`Ql*>gFvrOhlio3UFyg{&YbmsH43#hwI#0Q``H~R9HQ%7cdLU
z4kRBwG*6bTGI!JkN|iTXmHndG$`pnXHMKxkR(SYjjTGlMg+Ua=6w-cd3CY4tA1(su
zFfqMWdGf!Ir5{;YR=V5LgoF`)2L_~Zj-aQ|{AqKK9}j?e!}$*Ym4O_NQzLF8{->6!
z??Rd}3-c=gGDNi!_|xSb&8ILCUn$ytHq)XY&RjVBj?a}w{=6oQo77B@iLb-kWg|fV
zWw<~KQ+pX{JC;%{E^bF2x()0v7m>(#wca>p^Np;zB>Zn(l2Sm7>Kr)<Xx2yY6_RgF
zxQzfigZ6npX|(5<CJ>)|LeU9{2(dm<j=4fH`totNWN_v7*zw;*(mYLITOU_Qy&<tb
z{dRi88xK+R{*<F@i)GSO_PVZ?_|8y3M@~8|s~Gb2D;6%|Dr8yS&-72V#;qBsqCN)t
z^9Y2kk)&Q$PTzY)Z@%}&1`%NqzWIDW+$wX-%*YV&;7_(S&cwid0_l%Oj)^(-eZ#{X
zBnvtU8WW?V3zK8Jzq7IAUIBsaUQ2LU$xA^w3VtdW<wZmc(T-U1<NY3O)1AmDcjEH6
zTr<=!7NQL6(kQvT{$XR~?$lHb$k2chU^ZE!!SOq#nhEZ*03k{ZQ(8pBP+W2`h};pd
zMgUnwkOB|Dv=-?^dS+Anos)zKme?3#W5Z>((TS(&Y3Mf}HZw?(uznag=?5e{ns%&W
z${Sb|ye!thH1SS(Ca84kDDnS29ZN243G0s?Vb8m3vwW4>D#JPOE^d1qpvpvcPc+0!
z*aLwN1dWQkqUJzm6b6D%P0c?!M_pds$jGSC$aaDFJ*>lr_Vz?{5Xz!q{4UFLQ=lmQ
z(}N{T$}lm>AZ(Sqsi2gV8YKy1$%HRs?}>Vo2zS}eRCm}B4UpnmN1G$$1K98h2D!`G
zGqa{htwU@{x;p_kf{n}#tKIS~!2<Df&3Hbou~tF?zzc6lA?{`7kQPnA+jI7D*V<g^
zI}a1*PuncK1i}wD%>R*&F1J$iup~L#s0Jq06lkQAbd<DE5yY(H8VZ87#l^R5qafT+
zy>9qEqR6fN0?pQ#POX?6F}pT!dQja7gL`yDJR8AFqlKNORvCCQyJEbX<{=F1BgL1*
zf}Ydarbj%~X~sslYcnH5)q|75v?S5Nx8IBpVbNe5t>3%cXs(2=zdf$j>NGw^!7QI1
zx{mL)Wgg`+^5U`AbxjB_ZSKkAjh$fE%Zl_*@RUn*zHl2UoOIk7oj6Xa%eA`wN#EGC
z+~zVee$owHka?j)t&*WfwP{ge31-rp%A@XxhF=X@1JwIQQWE0QQ#G_OzJ)2D341_F
z>3tYKO;(@AG#90mxP<kNN4IgtRh*V{uG?)It(3<l1N68yREjjK-ad-}Uk_fp4FZJj
z4nk(eo>sfonGp{Y@8;^-8kgs44xr*d4UsSoi6JHhDWRHLEY`<HNdCSpoI*Y|b>V1H
z=dq4+>=%NSuU5xK%W3R4<ALD2AJyaST2beG9M(NYf%$iBz<GDip1t}M)&o|?B<rHm
zw^F@bos62bdh3@HDs^1zc1sr#Ov{B`N!;C&U5XWMnb#iI32_qQof8T65VsLiYgG+c
z!@^};$zqcqPDOD^H1fS6Q($_UP`$^iDiz#g7#_`ARSEAKcdO6(I+q0>k>{i*qnN+J
zYm-UI?m1E3q5EIuh}U+l;-RHY17Nzrlcp~$_rGOj7I%n?M-mwU@}O!L-Mnv{gvFBv
zfvy%GG*vYzGjq?*J`=99<YvFBaa{kPZ$?EF_`v*^{!>hw>p6oiwr}Aod@71JN397p
z)}sn%^z0XipQJ$!?}ZPgnWL7vGK}HdfPUxIIiIgy4|%+f+r!O0!?B(7uN9RHNg1@*
zM74s8-Bo_v<S)X9>zi#718+q+atZh%O-pR-qwNpW>B3;IzTcBDH`yL0N%?6CtG64j
zwQjfz{^gzAqqS~V8Vt@2&z)A^-rDnM7@t)GuE7x6s#_ND2WXoZcSB#zOT)N)B<n3C
z%0!C+bhNhi0bT+f-9?Sv8R&Y~4czIn&k{J<#kW`2D<9o-;|Ew~*#Ap2tQ-$t^&t>9
zZ~R)rp1`{`0Ng!MR}Gvfa9Gg_PR`7|cU<netZPY2?>X-6dgdr6pE5BqE@piH9;qk=
z{^n+0@nRSt1?;qXE&OH>>1=dvF=JqAFqDdnBN2=cqXIMhi-tZYuDU>pOlaR}C2ZkK
zSt3x#XLPIr2UNYj2{iwQsjrTzYU|$KC>W@KN(mAQN`tg?w{*93gLH==f~2(4-61WF
zA}QUCh@^xdo!{i%`~Jo^27g?xaM<VUz1EuZnNQ^B7B0Q|9269EaX8%+q=~m2Quxry
zN*aDpn`i@;hUG*mrG}y=yUi#Io(TohM-K*WCT6bXu|U+<r`~9wv0ZC<$?{*{aKYil
zX8fuh{7+F3Eqc6@AhabpB_&#FdLu}bqU@j6;+kvxo9c||3@`2bRi8sUFF$|%U`i(I
zl4p~kPUmu}8~aq)S3ILF)k@_ay1(G#h|DzpOq^dp4#K^2T|~ZA?eU!|$qa?7XH`HU
zBhSfZCQQiGi)4huPaeCcqYL+4tGBe4gxSj`-a%BMBmY+Yj|I9!mn(EtvCU?tLR4Ag
zNY8K=D&?#Qx$#65v>miduc62H{2vdTP7H}+902jCAh%FIH{Z^+=(U}4bc}0#2`az5
zf<oCdc5Uv9l0r@0(6ke6Qzi5W{`bf4QYpLe{;ck*M4|XTR#zvQDNv-0_L<S#bEu3`
z93&(l?6isqftX=aO*#ab_(KVY@%dQS1=X|iTXAFguWtI@@E3f-NmM6RR&c{i(+mhJ
z=Umk+Y;4BMODKO97@8px+h6bjrwg3VMP58GO`{A#Ed9Tu&I!o<B1t){mQzUoy}H&x
zPiUOLH9i<a2v~l#>x|@Ef(2Wp?!5B2EEESwcgEx=zvtli4+?zQ2E4y`*wicPm(-Lx
zv?|?0+h2uZZD|g4v@$h1AQpBo(1yA1IStf*)FS44#3&-8N)~WOUP<W#qs>P(+NV+C
ziWFfnd=>nZ*iz&y)Ho;9tdBe9H8Ga1o*XbTvL!yVB2Hm|G=$lfw;eikwA4D)uFEx1
z0dQf?d!m#o<IRRT4G)gFHyzs@l{Hnpe~ABiP4)$bBAV<)AAzRo#D5!H+R#EpMdXzz
z@0pqA_QmtMb_VUiN<omhz1!&iz9VLq<_anbkxm6^X}#vgH$@&B8yM&-d{1un<8zgr
zcvvh)nO=Ql3!_wm&^E)W%_({8K)&!{vH&knHx2dp&=e-T^26J}@}^mSRWn|gz|RaP
z?K^s_uZyp;m^V6j&gxj$%dt?uga{+RJ`sNu^-+yK(WWpQ<_zGV@sd#&JO<msx6ez{
z`~|^-kH`DyxDIoBgZ`|o9P@>-frQ!7Ych(adZoE$ZjaOM`2~}Z+k2j4MTxAQ_neBc
zpex3|S1K$ewDJFy2P3Do<uZ4=+otP%BU&^=2{P&%phW?G90~%SgR?@{n>_x~vzPG8
zs<zANmVC-plTkH9S0}~R@4cl+5h^4f{o<)XTx0EPSn}Y*1~;*}@p2SfH|S+S@#3kX
z@MA_?au=9Qp|en{KXq2-DKCX;%u%(u&@3ub%xkvsg2j6GM)!o6kdWs=Rp?G%6F%XC
z<6=#C`cRP8Cq^}{{=;likn9L_=<(})vpl78T-Z!ED_rB%;uC7>Y9vM6fTT!i!%#@0
zcz=P5ondlP4F#F4KN+uf#+1I}h#>ETwQ4NWlqQ?})o9P7-?M;rVc)SIk;N8)s$GM@
zo9o7QoE)WLgm{Q6zAGk4dZ92y6l9UZ(NKGQZG9FZ<4f44pm!moX)oVM#yB7te)Rf@
zKH^jB7Eq>Z?^+LxywxBg7=tJDbawZm&F#NWlk524eJ4UZ@wqiN!e7gDflC?G=y_p#
zif>~{w6<otHCt6WLb_)~cxS0$<jaIaf~kWG$>IK4&g-+2-2D8Pe>Ljt@<J=0ApE+r
zCcoQ8#`T)P3~qhegF5E%)SsP+g~s2i=RU-+m}tAY9M*FHkGpKJQFl^)rd%+jnxrHL
zFZEo3*k@&E7h#r+_<#pmY;^1f#&awXh(LSkn-0XuDJQg-j4J5ClA;(SWqRMOF9(oa
zAYSno+4(u}MvVIbQ}f^M?W^}tU60ox?{h1T_QMdcBTt+rQ~$=5Y)x!h7@NO%6g5<C
zhak@%Vw`b_qK14J#{+(65|3?Q`?B|GH2(JCD%@xor<9#z01yEV0<$$nXWd7<J+!0m
z6!ET{RcSZ5<kzQvZR^@GUg43Ma?L7H(9sb{QxGgp?ChlJ9~=a!Pq_*T@=brc)X!ZW
zo0Xkez1)@yNiEA?(lb;7f<7GXTJ&2T@5&R{3*&h_ci>w4wCl~y*9q<REY)58E3;bD
zGQNB$xy-HE660=kIW{$Rg055#h7YGLMs_ziJ$~{x7nf6DRDI$AvK-g`dK59&vF4iR
z{?8H9Exl!5Oz_Fc{51^5&qFyOB?TqkXZtM>8ZuU-v$QrXA@L4!vkOYb2r}loLZ^Uw
z>l?lqCr+-`MtS8mm@G9-wvW`@NB9+mJbQLpgO~e?iwgd#H{3)>@4egN4IHmFaThVd
zNh)?Z9X?vNEg{;ls<za+ipZ&?E9NO#wq7IVj{gu%faSnqy49=L?tAFb_^xzdR2$`9
zO|rn4^;;s0N&8(NK8SX4;~DO4b6bwx({$Q&{JqBH`13hDr36<6urQs$sy%(tV)0EF
zugB>w=#W(VQVR2np)>yM!Sk!mWEY@%=xSsiqQn1{S5euSb6!ytv^%iYR-xBKMdak=
zAK!g9z7AuJj$Yt%2n&lIZA@EQSoyav+zOWZ)NPa#hQ_F*d?~p2@ok>JRBh+n;55x3
zCq1R%mBXhxyG<7-nZ}wtTYX6lGc%5=EArTHw7BBnQ_D+}A-`L_K2AgPAt;FHOckx?
zHUGKy%ha5+IuW0L9%|UMuDwc5PA)AiCCcNpKQBA=O?UF|>XP(m<fpt(>i6QQ_bSC#
z{mPdKn%u-T*5<w~3+U)UnevWiO4R;>ZN2;;nm|K~Pi&xhdoQWDBu8<sxzOfiq`0sZ
zUD4zuXPWr8m$_9HpBSBQGktuKr2zE4yxgVir!ga}j*ffO)LlCPa`GxmJD-7c@<q<*
zlDyevC5>&$0+NR8>?Md427o4`Xr$Q?#u#V>cVJ?|@v!jvH6d{RkYqJ5P0h*8RfD-h
zx4xtB6sUd7Z(oaR2kVXY5TfS~H}}c|Yl2n4eZbV*{AEN)X_pK4EI3sGx3DU!D2oaU
z9pvL#mzB!!=xDXnK?Zmr7i7##cjp)8Zw899P{XF{#o8PAF^}=1Ff}DL?z?mJN*JTy
zp^Ff`J15(i>*z?$`>3p_DAl~+=uq*oTeO=`Tu8g8pT8-yB5`C;Fo#AmC!0S5W<ta4
zEFf&qcb4=em2}{Y*!{SDH-faxoSmhgEUNQPw`d{By?2y6{p9{V#s>BPAVax;dL@nj
z>HGIrLf^J5WLB|FDCe}jG&RdtRZ<2Do|#g9JV6FnKnp(|Fs*PBdTEmP>-V>ge$WcF
zx}B_P%PA{^eJK0^HYRecKj4}+9^FYE`4jK@m(0N{h>~(NE6Xs<b&QnnR1Ly+wugP(
z>%xK&4tfQQ_>+_Ip`p@)iPQL#0HK+ibMuyDR@Beb!_T$qzpJ=@;}?5+e*VtN<{b{4
z!z*86H0S3$*)EiUOk+}FZQOQ^Z=&y?0Yb+3^5p>GXtw6%|2hc1@1;NNs|C08Qb63)
zK0Eq-+2qVL6BmmLOesbUhcHkDcF@tMaCQzoCVU!+CPu11O?)6j5GZmWNku=!h>l@{
zT(49kQem*Ma9=r7)6GnJ^*Vic+z=ZOH}Y}|BV)hs_ge>+?Q}I|jRQpK<o=Qh4&Q4_
zCZo8h0)|9-lUZ~ZDyl#RS1=G0dq11OaFjn6HC))geSV=p>no0P3uz9qRVd9`)FR&c
z?tVODTDh_55B~J!#B$hYgDh0bI?RNGSyj)7Q^I=$zv7*$DJ=l)o}rTbWbZrIIQ?n4
z9QnO2?G32Cy1A(}ALlf#qZ;Mq6<JjO;T(U2dc`b2*muR;?DYS`<6rCIg<sqGTk&b)
zUgwra-6h=r-kT7{aT)U7)!6@4-|y#;+Y*PDBLJa#it~SeduJ=X;7*e1f585E20tHA
z0pO@Q#}2#{#BBb5(7gL$h7}7uPc{sg{tUiF=VB{hnb+}2h>>ij_W5s9i?e1d2oi5_
zciLK<w>CCEWlqm8U_9R6A^mZ&vbalOV6Hz6G?cB_pIZY(1qHE77qe4d=>CJ%ao6`T
zFJueV&<J{pg6+4GXCQs{Jj<{Xh*E0te=%ToxM5j%vb(b~TNF&NVDn4=LUnbfylx-!
z?8Vt~GYS!Y$hH50*Igx3N&xYP5V1~Ac3nBiFUa6=8h&+jVU#1+;l;&XP;f!z9-`Y_
zCSng{4ja{#%_V3k$BSPbMWVr|_uTpd7%Fdd_(1p40VHWCQBOm1p-}_m>z+jBTv7Rj
zGWW@@+H>cn%Xd*~+^R2jy>5=Kb?)tpr+a@^*bL>oo01YrPk-|HH756NSS-iV#V@nw
zUW8EHhw1zIosSfuBs^~4`oqTfus&I&T<kx(dJ;;+nhua))1yIEAYDU)aXdQG^IMgJ
zWBi;+?>t9d`A8M_-v4GNeamC583Ut(S`8kyf(l(btc{TMeZLLfZ+$u(wS*JFq~N1*
zR`v=YJP;2+xY=aP0<yWa&u2TG`=dOrcn<x<ResJhaBG=$iG6kT;wsRlKEFC@>pQr5
zL{CR&+0-m#oZY5N!0+t?bq?@{7<Sdk63$OaOn~JRrpm_5>h$|3x9zcqSyi3x+sc^M
z_X20c!rbR;BVT6kdzEruWSE-e&cE_LIe7TLdgsDVmw@?meR|ioaXDSt>f**U=f2Is
z=r(}g4*Lbpj(1@O1J!$6iND9kh6KShT1}SakpxWOj-wzN@dQzA;Z;>-FDomi-Lc-`
zbfE;afU2vc$;9T$>-Vt{u4kh}v@Hj472?9_a^(BOoPpmLAl8zLhwPTcU}vY^;GsN7
zZNM)+t|S1++6IgH)LkG&bGEa?Lific%SdT&XzS`EAN(NS!c*%gf<kbxW_6fBS$dc-
zRRzMMJVvFKO`d2vtwzK1sGL%$j=tme7(G)$K^T6{*1MhP<Y)B|syN!&$)ye)?{A_Y
z&z-k5fH_`q1c;Y5Cx?oXatf#WQNct+RtO=_*U>gfG0pEYJvdy8yNtY4Yi6{5X|<LE
z(`Nx;2Jh0ZW3K1$XMnH=6#)VEO#?9}+f`J=WpAm)FR=ZFFM`n0bw3Ry33VjIau9aP
znS{5u-k}Fr0+^D3CgoXOWMzy9z^lY!d<k3@5c{kE43dMdBxiqu)>a3sfPgP4xf?lH
z_JcWY{2|7r;E>0<qB=HKRK!SmCFvFI%t^ia5L21m(=fsZ2W#3QMkBja3nXGZ!688~
zuY-Da_C^h}Ubi?InID)FN~zlDO;i%CTe>-HIe}CX@Vn>Z8r7dT2&*^S-oA3S-y>!d
z7Ez~SZ~7zJkpyLW;dF<b_fmxf1iZo48CUF&jZ<y>L{&uv>cO(ScM)=wwI`%M0kAKe
zE|={~CZ0i%%F4Z*KHb=)nI9<0&_8C$gQQ3u-s-2@qi4ta=|eamQ+*mj<ex4sKJK?g
ze{Lg?p^zg@yu9*~B7)SmWOmwnyta~fgQu{xG(s@#HX+md%cp}lGA1cld*{^_r};Oz
zn~!haYzR-EBT0vSjg?hYsr&C@@G>F(%Hks4Ltd|Ib-!Q%`*<|~iAPYs3rxG7KP*m9
z{|ZE_!9nH#Od^M$yJ+Fo+DQ*LKtUI<7z1ZzB8yiA(28Kvu<0k(nYW7g0Jy-;%E}iK
z(srDwK=oUNz5&OL57JQN<@MTz-#M|VJdU7TI{My0!N^zN?NCt(ciGBJuT8&_J0!&I
z<orI;+%@R&^Rs-_Xe>6m(ZVDSg1qMcwDrT5<81d}y+=+tf_D$@k51hK4GoofQ=6%x
z0*mS2)%M0R@=9_pJ1i>H8=1Blv~)cv2zc@i5`sl-o!u%(U(a~lo$Wr+l;m={F6lD|
zqaow9wv|9s2ME*TMVq-!jTfO*`=+T47xaI?l-=3sde3!mD_BEy3D6K;56NFUE+ABg
zk)_L@KeoH&crEqU?8Xye>rfdR<_&a#QQy}~h{$s=cQ`Mqje}!Sg1iI7XPJFla7PWq
zFvrG*W@^nY8#m??_^<{)-n)T{H2VdFL)%yS6&@*UY-}F)ZS)f;s@BU5&;7r%1K}AG
z(#GY<Y8@Q~{*iuVbeP`U(d{$|XkP#qhx5>65S=u++x;Kr1YQzZDYl<6T(>WN*2|bq
z7dT(#cE^i}Ha0Yt@>36az~RyLjnV6MfX4Xx%oH2-!0PqzYv0+(er(rOS^v0&y8W&b
zK?q{pc743fHH?3}Hh6pOQe-+=KeMGMSeE!&>^2dLb%iqsB`q;bzK%!0pr{z2M&zZL
zHCw7~3NY1t!xBMYHs;l|VL;KhwBrJPjbi&-9It*EUeczi?qgirjOVt06qGLfc>Cp#
zO&~zs&tWjl%`4KbR8kc?5$zyCnkSW|tcvKIk4Q+dZW5}Tte9FPaocZe?ojrd>iC^}
zpET_-+&N^N#eAHMAb`#*<PF2{3<4jetXL?6+8W$`4-{Fr89RxpU_fHnl6XR2BHQ+K
z)^7&gCe0@Ej{rkkuv^7tgaX}qvi4Yqcr$PHBLKz-(tRb1KLg{z_|~7&ArDdn8Q0LH
zrW#-*>`8ZJ20@)nQr(-DdBEmDCgZX)PK6**6Zi*fMX}jS)G7Wjt-hW*BDw&y5z1Gp
z{QU!f9fBfL_8~?hhmE@DRYj<C-4C&t6{Bv&*{rOtS{`U?sw<z(D}KeK%?6F2Dy=q_
z131tbjeH)Mjq4>&V05O6YcH=Vd+zjV>q|xslyGn$b9+&~9KDi~zZP$;t)0llHCyXx
zbF$<pYp{^a<Got$G7CHR4d2@j#{2sDx7Pg&wCHVM2|bQjS#Pxf76|_{qo4SILOjF(
z7r+lMta-W@=VwQ3+Lw~NPJndY9{#}9bS{}ASOu^+G_=W8gX_M1Ixi%?v#>LBxRSq`
zFJp!M7xD3QuQFO*BI2-p`$bN>PCqYlFf^y{Cn%CypWJ(w87#F7aJ``w%l)>nL~gzA
zWFNmfFP3~cnD<Y%)K<!I1+YLe1@D-zjUtKHy9S8qfWK4pXvxhPKRVVd)ZNI)$Z$M7
zG&6eQmOAG|tc$hBf{XlkX)`o2z`HgPY>?9%O33>7Rp02%p{4DkE1=_o-@jx>QBo(f
znbpU~uQ>;hMv6myvwyIj6qP{YwDoARYABh>b{z<~>|1|KO4Xq4oh@~p;hak%e{B1$
zUjQ~GGG>#MxqX^3AvN03=$i&k*B=fK4nKD#f3|CV2xANx45xtojcxS?Mo3-gS)xtk
zVrE+VDEt^^WWT4rx6IJ}X=X&4M?)=4S=I)kOjucm4>o7JjO}$Ay$-HwD5MXClc2kt
z?F;eu-vwYuTa(>-?*8T6_w6;~>4IwD2!N#X^QFwr&aw=DG<FH<%fTEk7$M$`7=RC*
zlT#z3K0rg`gLbXc_wM0w=b5gZ9m($4h~$sj>e|fuD=)>aa{-^kB!!qhiRf}z_;WF#
zh;Yn#qT+vCfU!dLQkSVseuRJew=*dI;O7`&@ZQtYgCJalzb_dVm`UQ&eGkW*`rfs?
zE?+ts9v%gPyQ5=U7_pw2_3_H~P*DC_ZLJ&P1LOC)1Pge|i8MzE2_MWh+Vj2Ys|tva
z2STNVDm~+~em(~lHF<g4rVFRn{8$fgu7cG*fZy~PG}QV+p#7A}^4vS-!3fvfC|@z~
zh#VFL9_VsUGJWNM%N^qWi2-6}PMe8O8e(GZTjwoG=&jCYl8)CFevTyYz1lj*dxZ6;
zb~64#$>#D?>Z|^`S^ety#l)fYfBRq)@zZko<2h}_CqG|OOyf*}@D3=}--Z5FXPO%y
ze+I{oSya2;;uh@c;*#{#^k~R*y{Gef?%fTIBF%=|0WBLlm|s2EYinvE#WS1z+R5)p
zn5iEDW8FX$BhDs+k)4^2nyR(xx<G*Jy?3aE577VtPUiJS6L=3&A~#Gzrc~W6kSe|~
z9)hw}?QCed<ju_I)Z$_l-MJrw6Y{#UV<tUMF#>KOQ0wL9F(QbhL+x1VIRr4a%*v2(
z6L`Fu+XR&?*__LnrkawzzF}a?eB#56_OORehldaD++i(M15gSDvfNTo5&S(@RCFLL
zHhNW0n#j8u?*ptf<#@P$BxUL6xjI<{6?SLSUC<k-BR5oo2bs%&v>m5f*`xk-x|@ZE
znTEXtL4MTD=+;hHzhnaokInB`90p?Hf$uyYoUNlFKOhVcj?RPCMcZXBU`M6VP#<q`
zbz53?clT}+vyHb+?zCI7L_G~@^rE5NZw2>_=GPeuqTzCKskW~5@?>uTX{7pT%Yyuv
z?CA&C8DnMaWGz(go$$`4&}l*S7MkFfzn@hnYK}lGkfA`OMURPK?d@_@y(tcOQ>0OK
z1LAm}5dz;EBYdjiz-PT5R1>;&ILALnG?C_rg`BPF`jxfy-A$qT*zC8yS2p&h!n@nr
zpz;e(xtCSUv$T|rS27D@JvwUoq-w(~ABN^GU~EFS5FNdel=RRr;KOsLU#&FkAhSS_
z>tTuYgmx2yUuh}WNg_x=2*2{@<~!lY1l4D@>*5ANekA1s)3+O@hKEBovS4<v#mS*B
zbwl8N*IGPG6cGeA%ZI!U@3Y^!sk<6AH8z05{wH@g*vh&;O7XrhCNrL_b0gceM+<xI
z!1DYFD-rUE;?IwTlWHh~o@;GhlseDpNt0V4&2viPl#j67wv;o2(QxFk1hCz=k{5@E
z%4a9~ik<g$N`vpXZ=q$1Z#5j~+<#d<N8-nU@gz0Z<)CNlD0yqblC7M%oUqrE`7egw
zm(QMCj}lr5c<gfLirgI?J;Gm~p)_0h*b4|m^i4=1@1qjN%Io|B>MR^APcw2DmU@zB
zTLy?d$=+f(HJ$vh1(q^$m{_NzrKO;y5Fed*d@St~svO4<Jn($a0eChi7(^5|9@B6g
z^qz?cU9W4pb3@R%(Fo68r13N^<~BkuaRm{$F=08{X2k`~Dnt%6G;oK6-WFygKt7eO
zJtq4bj`{>u01hMUJ3qOcx4S@C)qk#Pa|Nb#H`6|e#MB4ec01zyah{Qb{IK6wXyTRW
z^>ov&{TWa^@F8MW(S%N1QpQTh%Th~F+|)&1LDI!=w=VSn!k#6A>Zj{Zp>PLnW!hnH
z3k5@AQEe)|w4}M4oq2qA_Hlc|>nlCwCZ}WzgEQB|!57KIOz*7*=QsNwp<D+r&5#wC
zt&uIcyO5pc(vy6(ee|^P*9C>B5X(KVh0`ryA*7K4^R$7&>be6GXQz{Aw_D)rTow$I
zR=jqHPbG+g1^|1)B>efV>TC|Jnq%2lG)q*tRlIaUB^FoHX+4>xk%~t^-Sy~@kqmmB
zaZ6lPnXtZP_KV?Ns{3x7oVOpa-aS*oK;9nLTC!OzOwRR789fo1s8Vp8{p{&I)i$Eq
zn{va8WPeTC$29vUa_Jr>9wbNE-P^g?)rrDyq#RQlzcwtf03jdb<RG+`OhbGk(gb9z
ziP8HJeqp8Cll>zD=?ZyJ`wGMuGCi=wmC}}eLIcY5M}y2vPo6x9!FJX0)55`VmEj>e
z^6+vqUc$p%?d$FpAiN^Lle9e|Xp??Ki^rg=`S)b7QHDoQkSv+o{oS>lK)u`&iuhIj
zNJDdt+QwuI1PfJOQIRG-=n2+v@K2)@AkX1HSRr}taj(JB(oz^R5qW8vdQdw7rZ?vC
zjn+eX-@Cq}_+UqZ2w`PgE&AXQgsbOKFxzKlW#Qj;6A~gr7t#J(_gDV6Lyg_Tt32Af
z3W9=P<#ZWHpON#B{T>*?Q|%?eyc2%gi)7Zrt7ltrnXTl42juSLET>;GOlm5R^ff5x
zo@Aw`Cmzm!tjVk>EG*R0Ri${u_>nZQ{?x-EA%a|8ckWNC#Z6_fRZe1&(UcN`KcOt5
z&Z2({jjW$d7b-&sbSvr5-H`6=J{6VpUToRJJdEgwSQQn8UE-euO}Z)B-^H+?bfzQ)
z&v|+(Ds@#=3T$+VNK|yR8JQlk8_?Q?&8#D-+1dM@<*I(u$4y=*GOA)z)5E{Qd!z>{
zK+2zzEju_Y7YBK>!%D=wUgt&)hNh`;@v)DN7osBL_m%HWkp2q5C@IV>%+D9)Lc1c6
zFcwYvM%~+caf)|VP8SnxOqeLJF?@wUM7Wh@OULxMELHEL;ZO=!q_A#mP*4}u{dOR=
zNlJ=C=r|gxe={BxJ-zD*eaB3Zx1mE}Ox9(XZ+v-nPQg{=mRIsSw}95jUtP8?=sT?!
zB#MsDAAV#wi8_=}qIO)T+#;&OaBwRBZwbtuHCjewW*WMOt@pQh7?8=UEBjZWkr=%{
z!gnL(a9r{Y332%Q?McFDQRYDI_3-4bj*hjhL6h#lv>ra-6e<i2naCuOb{Re*@BI1m
z-T*TPQt|N<GaZk?bo2z9ingUyVJ!VA2P4N`H7j9)`CeZ=!O9|R1q))BbDynDJOuA0
z^fcO?f1|*}y&Fl6XzkTK!D88K0S~@l4U!o-P1QJEx!m-0OW^r<4TZg15ET(=&(6;F
z-mWC|-{`>D`IrkZ{Ur9+8|XI)IjrLwk}AQ>WWUE^Yuepm<wN*~zMy0?5s|xZun_Tk
zZ+?wQ&<#vhlHRxyZ#12$kW0BE;<fT5l3Ze^Wn;YXbGUBWMM-|Kxh?u5iN;%Q7n?0Z
zt64N&NB$zF?n71DCcP(4<<%SWrhJZr;~ckQ(}lSkRc*N4j=v7fRwXoq_LX|VMse4O
z=@=i~ilvWGGE0HC!WY{s)PnBOBiqv-#c?)QoYp5(5>!8#p3@Ep4wK*fl~XJq%$>a<
zVIg@13-xv6NyX<^;mM61hwqP%2XUH0x@jFm-jGi?yhb`O@ovs=d#A)p-VOa#YrA|t
z-@paU$F<M9H_>kj30=#~9H+q8s&T*A9_2d#u4$oG)t_LlAkx64J6UphUrVA1{ROkq
zI~XJ)zwR3z9GQ@Ksj$8g)cac?lBMuwPf%Vy{g2$=odF)w<NVZP#s?t=*PR>g+?Xu=
z&2O!9?pwn=d88XJTkCpqPEw${KI!b@Bb5RNb*y(FI*QL$3}HIMVIDIc4NXDgA?*b{
zfw?ZngP#5`e%P+{R^A&huK4X@hW}O%zQ&i!$|~@g2kTlZZZdLWXLR;_zwm7}jlW0?
zR&%aYtFdP8k}|i<#k|l;Us>s74h?<NH{hvvcYtz8z(a8G(<9<{5>$>JyvQHyU%$^e
zc9#qQG{cFB5qP8rUt>124PM*PAr%hFj;zSankmd~BFW8I;;XXTlH9|Q=30w56Nb6e
zdA3;X?)jGcX7G~3pzt>`>Sq=zA)bM-7WzXSDG3P+OH1q<YXcnw5)wDV3TZD|JYr=f
zrTQE;XPD`E8L6MbCX-Q7NV_P6E4wY+H>j&^Kn8XX_g?IFWM}1_&*$9)Q@e6|JKctx
zE_+XZgnIvop934C0dn%E2?Xp#<uVxxfCkK6Lq+j9Ty`T9^1N>#bm5cA=FLnY!Ssk~
z;_EbO79|9eHPL^tKOY=PoUVWK%`gg6s$P*2+knh-_oV(X(6=}YH(oV8BST$B{<U53
zjRayvP2z_s@4x9w1d7U=ndjo*;-Vl4t&vd+KqWRdTGrXCI~yC_gX3X{xopLuwl>~e
z)OhqvOm>UUjniuI`PjK<>o-$y!Y(Vbe-U!KKTc3xLO!fT;)UefD%?M7#x@0o-|E~i
zY!j}QRj`aKF4nZmCW+()hTgdYgi&6H$sE4Oq^N?T5f<+7c18a3)5*^-Qi3Re;Ay!0
zOWx=48b6)%9@vhs)s4J(a8si5>h<0G4J?dIA+j$-r@-hPz#jly*I?aV%xA0RzfLX=
zsn~_a4DAYvzaS$M3ml%K##Ve(yo@#8O}q}AL^JoEie9@(B6C(la1)rq=~k@DIGmwx
z8^7y*3hZ0<BIjkCIBbdYmCOQlmhN7izYe-T`2^F;3G*WY0&{M(qx-ze{a2exBVdfX
zkEcpYNnQv<cIT4K-UeXZIo+T9`2K@nq@lDG9l<1oCy!gQ#W(Gzw7=|!*vdw?Ql2_4
zEpB@jW}2L`vbBsKEdM;*`1^OLXsBSH_8b*!wD0}lef-C!wIZZj!1_!I)Kr|Dxp|<n
zq=pAG821~BPyMNHk@v=#=jzvQnbGjgkAry;kta=bXN$gDti-Z<=ZOzg&#WQCM5;T_
zxbH)E^h;n{)}*C<nHhP)N+%Re`KBcziP7;Z^ee{)zsbn`Df47C+(eriWq>CJ>!QKk
zkwRiH-y&|AK}SruH8VS>M5}6Lx|IfsEs<MGiGMh*BaR-8b47Jhcs)~<Tc7wzKXB{a
zhM>ok&+K$OE%xt!7dHg>3+k!4L@P411KfGlC+#({+r1^K4#Jh+=|`lpDNTY+)Ogr9
znA;D<u~37bsH-XYGP!xsWf>Vk{HDDZ^!9(PoyFXhac~AO*P&$y4SUz*xQr4i^Vfj{
zYfo>F25)&58yzjn+iCT;(p>-hq@?>PA3QdelFqoQ>oJ4j>Dbx97tfN@Y1|H|BAy4A
z%!$^o>*p;_bYJY}2P?ChdO=>--P`+s%lre7@Xdg0GYvKL%6XCy_e_Sq{>(8%0d(=Y
z?&tRmzICZx#>3t@^1W#ws@3RL-by#=yEQ`SZLJ;`9U1!}da%5%Wo2Fn*r1wP3g(uE
zpR7m9c&>+KXPKB<m=Ub}PO5XkK@1HuF*hc>*4A6!_rIc7Ef#YQfRFd8Phnw*U(y}q
zLjr-6lvJ7p5z&1(xmMiw+J0?k*#AYRrl#?ZXQLvopwf$S`Ht5oD=E|9c%YXchKqv(
z{$F8#_R(+LCg$ia$gWZWaezd`825<9Et{*pq8NYe99I)!BRD`!62BJDM$(~4_}k&j
z!DP4DeW<-|p$ETp!Kf%L1Nu_&2RmXuPe#4&K<Jl&+%q~_TM!zvm}cNj@vG5w`5bCj
ze1~Jr3;qfUh9;lB<?x3~IK+2ZorXURmN9v1TM^5qeg2H<SXLJK82SACz}k`=_JVJ3
zv`?@48p|a8`QFETW<A5q#`5a-*6HsZRx771iq_UAGWg72vq}`9<A&EaH@6rt7b|j1
zuZ<7Nd&darndYC=XqEmtzkW2r%9=tryj|yb@Mo6D?|Mz!w<os=p8K>$dh9=Pe{hsn
zS6@j3A3Oxs@l3n_(Y$D$pY=QPXe_RZh3k=&Ra8?mOwIYdx~ZYAoJGnI9_z8DZ=GLQ
zh>9pEDB!|4Szvf<EF+OsFzpZ6@0MrFQd800KcfOhTs@qMTT*zNJT4)mfe#d)!mtGh
zmy6$zPv`te-*idV+sKz;Cs>(Vj!$e{wgvDtDLOPVlEm$5=Z?`}d!l2eKJ|c7LnvqH
z$%l~l{Wk`PIXGl2we($GTj9fW=du9a`=>u0WbK}CBtv@m9GbrzrBZR?H1sweynH-l
zpnjKEm$;kt%Vm2~S6kXK)q)Rbgi$$td|vwsrxREGW@$R?uitt2(!@YnL8EG~q``E<
z>biSE&`wat%6KVRfP|#vdChM~nq@@Z_={0f@4!T7qQC#sa`ONfqggwRMhujWjIc83
zoWKi#EEi5P2w9_XJIC}Q(a~kVXL;qLJVyWeOB%qQWUalrrlzK;1Ez3Z7TK6m(lD4j
z!w*m9UH*#(8wWuCuy^>2v;rj@{sX?YuK3BDouhn#@mnnJ5<2=0=xfo?{E_CxuGTh+
z^c;(8egZiI@f6c$q4nJC+>h6UMsFfWnjAjvT_Yp2`Ijbtd1+5a5`5sbSYFOAT6o_l
zyIPIXt6it|?Fl*Yb361NV}kjOyzKnPG&E7~KNMzW=H{n^ums9yc`dDb1h^9=HYL0=
zJW`TU&g;B^S14aRjS|gx7l~?bXarv8{as?%=!;{zOFDlVa3A1IC38{J@h~$|jbz!B
zSYuR8)VTg;-{8L*2Wh%jgal8X@4ebwP|{XxFc|Y1X123S{Pt`3Yq~y~tF@Z2lC)Lm
z!+`r(_<~s9b~_iU&nl};4y-*BYnwdrJhp>K%!&L3Bcc}SZ+Xo<7E^%CRt=V}Jd=aB
zZ`)YCErnXpcW$9WBKwT?3CG@gady7lpALx%VV0CW?-Tl5I9ljE-3bZT<{B<;(EVtQ
zi6UHCOV3Zg&EMJEDKy5(MNtg_VZ4v>Qgg=*e)w4Lvpq_?dp$f6B;TkATY|E@Iu7An
zH0F&;v^&lUs)|XRw(Hp&b*<mjg%7WwnEU$I<iTDBC?zvfe_U*AP^i@V*jQ>xN<kVO
zs2b?w2=R9tAsL3DiIjr9Z)j$uE&qG#cQ5cyl#rnM+1L8AAM!FZ3TtcLeEYSYs8xO5
zJF{+8R&LL3c~sEYNbEUxbr$U^qUieo`yMsbBYE99ES&qQs_ou;`;gtPsv6h$;^{)4
z_H~89K@r&dmvam9X6jE!7h5bLJa}^Dg9~i)+1YI-3|fr`fAWBD67}VU_zU9sjja51
zS{ky;yAHP<?z*Dl6!+ul9rwxc!=1{eccp1lK3+Roy_;uZA~g3=e4?@IGAop#yT$9f
z3)kU!f|u1GSOXjgH8Lb5jIJctQ1Y2Xq%hf6gda*53F+^hJ!2!Ic54&b$|5zOgs!i@
zGDAF2YFk<9xU(}2Yb$#T`}i;*a}`X$I}Iy3&~!93aIx<#P3NfU<|Kp6h0AFJ?~bxT
zxjM{gw{J@HgvZ6|<Kf0+W|_P)Gy;J=-0{U4yNCfsSTJdN^1xrLsCZWAb~aj>;0!tx
z*bnse7cP^33!u6P$Y>}oW=rYA3=X`4v|BrAclQA2q`b^NS1CU~sEk_`BTMAEPhz5$
z(UUB~>a8@sE8;KzxJQy-?hIyn`)gLs^9wnheTlc8zRk@1bvV${`mkRd>+Zl!-R+Q#
zc72IS^^e{cfA#;ZwD~ZC%FF+joAgcEul)M>H&Stsv+k?zL_G(yQZ)93tXJmQm(^8s
zlJN$s!w-3MgK-|$ZBk~E7@hCNtL<(rF3vBF&K)g_&~Vm0#@}CnDssAk-a3e0m-M;)
z>gZS(=dSu~hsxuFjaf5ozstfN|LqylsB9eN9U*hnfPL~IRC&z8%em3p=h_kG^fhpo
zGGWkJb>YB&>*?*lDJ^tHLvg$9-@W$Sk&~6A2lEGb4(T(+=+8w#L6U?+IpO?AO;2y?
zub<begsLrb``663TfF4Rb3GCaS^n!KnmR7rL!O=%EKRp^rnlP7%nfA~uyfY!>7rot
zzZ`GE^B0`&kaTvU`}gmBl6kw#d3mX76p9Mb7rx5ljQ{NQpDsR>kYwmFedHB806fUa
zX=rlHi+W^vg?E=P+4r>l`orn+2Ai$qo`b?8&v#O~IVE=cbl!#WoVK)+AK1*af`clb
zs{yD1zu;IYBc;ZlCa2;423Roj_6{UbotCm3Qn=?Kob_|6Yof>xs*aO~GZbWh4@&xP
zUz>s70nf$WwPK9ONPYN+u>vvniuvJ|(sGM8dgIlCSqiKp7ep5W-~N3Ijd^gTX$n+;
zRouI`;ik3a0OrMuwbGgwlNP3_VmaRWOJfCN=NvVzJqM=w1(4dHySR>yu0=(~Hu+Mj
zG@;3HtwRTPy1zUC(f9L*cSf%iHpV}5`j$H{#CxP-Yuh=5S2D?=-nqQq8uOcNz*H%$
zuzM_@Fi8f8Tp6slEN|YNho9(|T)PlG<!ppvFh<g9GIjou4gP+LNQS&}j4czvKOgF|
z!mCH7fMkKmKX8FzC5&4-S!#Lwb-*W=pr)b8qo#p;aaCp2j3i4oDM7*X!otKH(h`8&
zH8m5W6N6^;pRI^qJ9gmYVWi3m>$CBC?8u5kAE&9J5)0019(-adP$hwVCG7MWM_z0R
zz;=^}G?X;gaA$J(^=rxA(1LFk;=;w#<Y2N=QdnrdQLdn1X=b+S>nkWv1MQ7OWQdfw
zxomTKZEbdbe(ljU<k^GUMn)o_ZhZ1Y<eDD}azxT~9f`#Dnl457GZdsl?&K+%LpG?V
zkdS+ED##L{%T#fav6aX(%>mzw<-ZWS<KS5FysXUi2o6UPBsH_6yyBVc3rWb}yyCMH
zgwCt$V|}&wVE}1hyR5EPy+%=deo*@3lFUpNE(+C(`$;l#7w;s?)M<gq%$t$|<!JKh
zmlscm_+_gz3kzGllse(pAa%aNCAeKtPF|r9SXZ`FaMcE6SJUJ^M8<~2Uqhg;nXIh^
zUZR6fJ$Lm{W&hBKak&~?(UYu;Tp!qDe7UO&>(fOXWJN_aK->yYN-M!WfQeF4R}o09
zU;F)gKt_<V5Li(8`M6dU%kld7`!>Q@sIsr8w(b5~N8WV|;R6mE`c+PggK;Hi5Q*T5
zz5}S|@w$I{?iYSlJbOihNb(-(P?H{AUwuZ5s!?bJj>nSy2*=KKRCH#?$DezxJw*Bk
zhY#()*8;sPKi|Bd7?(!H%7p~?b(eah%j?h>BL&I)#$p)<%V_{K(#P9B544Dd;}hTz
z6%=P{XsT8lwrXHcl$KU$mFwGrOG=WAqPz-tn3=rL`Wnaj>MY*~?)M$Xy_u8n>3aK%
zg{uICiS3`cZVs|GX*#0-X+ggw9tTo!95}T#6+K(mLM5p3e#*)#D{HGO-$B4lQ(0L$
zg4D{YXRjc3$h2aF2KT-&nZS*lQx@IbyUx|;-p=dwPgbyDC33M9NlW*cT9Sv{-W|1?
ze8h{#6!m4i(X}W)2+-?VWxQZ1)l-?20}-1uZVPjDR}lk)g_XOru17}zSpXc3;uG^7
z&ykUxm2Gfqrhp4s=W^htB)^Tk2)%=M!4%MGXr5cJXHW7Ji$#PT|MoEeFyKG@G$#{e
znS3t0yZwwL+qN`t0^eHd4G9dIw_5}Z8IMkLgxh785qj>cbyY@-zt~t<Si<K9R(FXK
z#;$-F?owA}g6>Jv{f5ixf1oAc+qWl_G&IVpswGnu8E{G>ZqFx6v>L%#SX)p~FpR9u
z{0!Hj-FoKJ@unZe$}z!%H4@kr4-f|AeRT9Vgss4GkOBLk()kT{OwV2&!5lRfZzoK#
z6G+9=w(0@)Gi>h}?<a0-eF553P)Hcuu6yY3bmNL3Fbf3%9ob$%97RqhAU1a#N9efr
zg@s!Ktuc4~&56IZJX>!H_&(onTG3mK_HUy-deq1*DugA-fdNRc(Mj(aRDi<~4H=@h
zIp*z&8@?~TTB6fHK{VM5lexWi^Y=6nQdS-zsJz!QZKF&sdqymr#g`^$Y^0`ruQQNO
zhjBHEPs0Ql2|Rg{VCdzFK{D3g56NIAJs-LkyKwO0pl837(4Lpjq#G<yVRxM;X+D+3
z_k>vy@&PKb&T>4P8CL`W*B?bi#iO72J?m&vC_mtmQan^OB+Zwr?s!#iWn6bb_b-F4
z5*~=&6CI%uH85k^ow(JS@^iK&2Py~zr~~&aP!K4vZ^fQ^{oVO^)KqNB$bS_9rh$CB
z_0(r$V2u@AjH4q3s<d9G>x-b*Kxv+`=+DP#3d3iY?skQZ%3sj#=Wfm3-X2sVS|=xi
zFjYzY1AAjz$Gp&2vh-^h59rW8Bh3s_D#X*}XHbvNdDFkfCVs$Ud)&u48NVz`WZmf9
zEp7vmRRi~pf!PaSn=H`;A@PH=Hoi}l)SVsT7mb$llc7_(jr-SqMw;l7`Apj2+tv#G
zrQvsJcjYbEEwB32@JZur*jQR7K0h--KHDC>_`C4n_VM~h#Wv|=|F_j44_1o`Gu3o-
z9zA+g>b19%<c=}py^D=NGGM98mxm5}nnO9Nn6dEWsNX{PMK#=xqqkWILP>8xat}lJ
zYipeh*sSuD)-Fd-{e+IJBd+B+3>sCOw2#pdC+ib~HfS{uNYH#F7Q}SxJ<fJ26OjI%
zSrutz5$p^O*JD91uSZBCw`b$gArVZ13|NL^xa!dVD=aYQzZ<ZSzbBVSZ`X(<*KN3C
znw+w?7CL20M*pa@E&tXexcP$sAJ7?g%zmU<xoL`tX%KoROSS^&^r8c@(#wK64PhFw
z2`(})JqQR(7Cfn&nEJJ}2xg*Cc3tInF|06P=fcjQqNd+=AOW>00q)%=PapRST&A(K
zjht`!wx(6F{ErLJt6lHAzp`phvxA+k|5Gjo+;c!FOG-vJ<H0%De#LLDdfZg8JWRTx
z4Cn*NzV*QIoL}Hw8J3_jg9dKRJfH6ABn#dHqB_e(H?pgqW@c1S5=3>rkJZlzeBmkN
z7g8|{C3gIE(?)(wu>h_ElJrO9FqQKYdx7O$UlfzYFP9RN<fn0$=suspn#(rHs=nUa
zpKi#ji-GcO1JOXfWYRZTSw=%oL-*=D`RIAK2XXi+=4uPj5N?m1hYi!-D=rpK0-x%F
zN%+d|)sTO{j;_ugF`>}Qp^7La`6axtvKss=x*~%j-d7mIX%x6Afa~Tu-i=_qs&yO`
zMDRRSSI0|0%|$7A>IY0OO&9hX#E5Eeq3-O!zYS=Eg2~vGf4ldSM;@*r*N`<r0^wC=
zR1cJ>POJu)Dr`_BEK&+Q5QTzVGkW3YwEIB4`n%4&2Orge1)hEeNJcpx?Z@ZV!>_fc
z;L&2x-mZIf#3>P3R>s7Q=C6y_eV6EC+!*%mM5+A+OfdNR;<Jm3moDf{FVgPo3*zWn
z0#%EJV8gK-|9xS|Z3G-EC3f*iaPPJnmw#)A54JqaiZdnz{^dpTv+Q{xg7{B?loArQ
zdH$fdV=vNFS5yS(>qXYNLEh6y@^br~E*fesHCYY}IHar2+5yFS_OW|ujD?N|Od7lr
z7nZidJH*FEMm$$ytH>mw4(#>Kf~|G7{@fWi*ztF-9{`+ClRSSY0}kgO?t9X*aN8}`
z_JEsP5?EvE=?&PX{FO~^K!wtzD`kl;nZ;|jF-h}=?`MyV(VAOvH`C-)L`7x7hUy(_
zT&1JRowoRxAhASt=&6~78S1sBIqyS5QDc?sTqG@@yzlywDn-Bg`tsk>?{=ymBf+&K
zK)G+GVIgX?v6LSD1lSp7r<7EkKu+}NJpwEm4Iv9~b#=dUuOa6BhkElFW9pd6%HSiC
zg1BWoTprK!{E7;0c0Lm6BeS%yK0e3&pLykYP-pN-r&JtQx0Ww69IZ7toou#H^yw0f
zSJF{);`U@w(=c53nH#mf7xnUxP#_1&0aaraCoVoF*VC2V?(R7i8&rfxd*e~^W_q=F
zq$T;JOvQ9IRA$)Rf>5_PIy&k{K8TA^{&RFHt1~n@Ji7gaJYG;<;X(_ifED1(Dd)|p
z#{0K-9aZTyzmd>9ciORZbwx{P6^)7kuCt+}q-0=mhH0g3Hfq$Q<?6Oi^Lw9WtfzkS
z!$Y<sL9LJ?1sM0Gqhg~|3mYi7^^AAg<8ee<TOkseijIbcp8m}6H^^xxZmIU_d`>6r
zO9BtaW99GhIV4}7t@DMqMxIs|z6t!d+48g?(&&$q%-Y`?yN*y|2Pw+7r@oilZm+IV
z2#0B&wDcY1RgK+M?US_D5Tv=c%^Y@5gcKXSg)9SnDko?qkQ?Z(B_#zzt<%^<Y{uW@
zpp(9=bH_?F+7_q~K$_lN)V-u)`>naF0n1KMe+RJmbG33oQBeW}NGUrzx87oI3tt~s
z)9f{ieY9xL%Epp9Yufi^aq+UMY6)OVngmX5=hu{0Z(&fNX4m2(rYr;+GMdB7`YMX%
zmZmVCoSZ<ta25b45Tr98SW?*KaV}Jq6=Q4<5EYgbe>SjzidbJg4*j!#&pT-R^P5?c
zu7|%yn|i20KTCV!NG#d?F)<SNew>eWcwnySBdUkLmV7Z1PKKgAy{!bMpoD|4AXE1b
zao6`+CTEE^uX4|5qog*b&5um^#7}C(lHWpVH8i@D86P+rD810qaC@52dd(;GVBXxE
zdcR6#WuYa&t+uWfLHq^P8-A+7cmbiLqH|v$>9C$7;_9zVN!c=kW5-Dc0~%AcHJ9}J
z{|x~I0@?<YRL7o=t8*Q!W))_3ynI$XS5I?(IGTJeo5<T_`o(gYw?U1z6ILjD@yXdi
z)-Cp}@uUxlOzun3Iy|*TMvG8xfITL_*60XRxgD&%GOLC7jukax+1I3jp_&(05JAcp
zFC@1)Y8kj6OA7j$nWi>vlz9FuaCkYU!cmir#=3u6Q&U>{=-G3JtuF?EhHlP$zl#hj
zDd)&OfaD=nZ6fcsX^|d2_oGQ^cg+Ij@7~&tYnZ^j)tkz8Y?@Nj9330`y=Hn*bJC3G
zYzcg87&7c{Oh-l~Fwl^lEpPh$T+k(~KCd+R%V|L7*VRQ$CUgezQh^K?zFU<|Rp(z8
z?J~#LAxmCU<3o{ljl;%UE2wYuaqAl$x8}Qyg4Z_+Ki)P5UW56MiTl@Kh>O0R$XOOg
z?jN4gu*>1T`qPL{(SU5Oyd7{05UtYd)*W3^h(2&w>FeeY=NjaIfyGz%XS$>xc-{3N
z6N4n-LaWSlwi)G2E%F|KdZANJ?`&E>-imX!n)?*jPWJkRq{LOkfhg{bv&jOF8F|0l
zb$0wcae|p*c@dBP$jVB_I{9IM{q}Ce=Qrd=BE-S1$5uViuKz`YT1L0Vavq6mjEuZz
ziF<NTNjN|S3$)&%T@+fyr+&l_2{xy5@Ccsc;-!3s+w{8lOj7E5w{$3s{{lm05uabt
zKzMzc!r?_qN2VwyRbH0A2GUyO(oJc(`0TeTU3k#D1v?HUYOA5DBm|fk9<bW3a=}u4
z;7|)rbiEA$llB3h$kio-xa+ZQzdf5`TyeO=8uNeE@P0<>b1p7E_x(bi0GVVC2977v
zyMt|$b|-C>XFPA=HcFMM!=g0J2b!q-FRJM1*t9rHwqO{V;e_QWQOgQ6W@bhoQXT}r
zOi?N_CNWZDuD^MK3VKWVuNYT7wb_CJeH?i#8xT@a5mxmyg#ip}EqIUpzQ+v7DO#BV
zwXD1N?0rclD|fQs-<z}uDb}#6l1f)sm$>99S=KFyNI1-MJy3LgB#oIq+C5a44>o0q
zk4WQ1qcmH6U`qEV4BUv)hE`$t1=!(FF5JB$Ko*ijAw|Pn_A&kwk>tq73^ZKT>17HU
zl9B=fyj)xXSm^4Jk+itClD4B#BIA}vhlio<&p@S@45AyN93I&gQFj4d3$`mdR59ta
z15lpPN_5IP(IVpF;`jtBuv?L5WxmtR)pWr85RZ?Gjca0POhvWR@Be<GBYm6eF{(hI
zO2?p_YSePq$jG+h;9$aXs@^}CL?~)Ra>NLCU1;J%F-2`U3ihV-bU^_Dor+W3P)Sgo
zne?yvw?09;&~lKFRfTlO{IYy%o&^%;!f0$wSussH-IVCW$<wfyg${G`Ty_>VFmD<g
zt+H)@3|+W}hNJ`YOZ-t8c!KN37y>zb7_bs+9DbMIQ*uNN6(!u@fkU~roa@^E6*~U-
z(kARE|BNga#exT(;qRI)1jwHsDWd$h8F?36wXnG*;nf~%a*4Nnx-jVqs<$CZ%_}N!
z-~Dw|REH@g^_^3P;J>~~1{>X<jAp7pth)2QXfRWA)hblNdUxfMTiP6*3N8V&b=`N8
zS?2lPQ-&-@M`()1bRaq2T(9~Ubv)w(6JIcs_B?9y{{41jLu)5X==VRM8~(nLDNBAn
zQ^~?&#ar^u9<Kc#$en-wsM@lm^!Mk^tL9vooBg#iy1}m-=@AxLr@wP#?GYf?y?GnM
z%F^=Gnu4`|EDL!;^`vq^go=jd!OAMaFCt<DL^z7@h}g(AbF*_cQgJQ54%?7Bcc3Kp
zu9jK30gWX%cFG^G;f~YBRw~DR3e07@e+EbOxQ6o`^qW?urmlPNm_QqkfTWP<0ntgN
zxwWyrmbPLNc!_uNj5ETd7gEfF6y%)gX{3V7c7kv}erGHsyCImH+I;4;7$7Fvgc)qw
zD5jwBQ|i{g_oamwGv&hB?n6L$O5d4Teu%>qjfh7ReyIP)cge{z!ATPnHEHt^WcAmh
z$SqGSwww#P4@^=^su#nRiDwrYm#;1Z&%uGf^KaM;5rd?Vv?G5LFA!%Jad4Vw)sI#U
zhLx*rG^WZ^0zTMXLJlQ5Z2cCy&cEM;9lGxCQuOJ7G~##E){YCh=D8ZH<EU@+u*3FZ
zhkeZyd(qw>Voe?SiJx@JIEyG}$1vAKNA;b!73)<LFPp$k&qhE0>*FmVhWz_hQ=gY|
zQt&byd$?v!YnZS2356#=ADX~1)3-EiHhcz62mr(5aZzy#{ke0mSS$qeB21^gPCH!m
z(ru*N2>rK6al{h9x}V)6z?XOn)i-zLGjZ6ymDk!ytLgBFQW0Gp;&V9&=)4ggcnw4h
zxCF#Ek&7b{Izhn}cg#P>=gV1gO)`Crn8APwG3u?0!!Z@6a`RjoFJbcdz7kCBO|%ue
za)N>agWd%OJ>cm64ns(jYMlEffI}r|`+MX67_2U=xZQ4gKEg5*(`0sZ=gyt<4c>;@
zh5sJB@dM$Efrmuv@zU0$Q-AZe@G@Kve;q+ZiSlwaJ!SEEBFNrZysOjFY<;9ioK@r2
z*5b?YiUYHk@mGF_{UL^c_{69L@1sdzma#o0fTbtj^46iUf>dH>;yXJg2|FDPuUB^<
zG_X*aL3yJR&t|(`m2?7BohXVB3Go+p&W;n_hr(MrGk}#H5OGF)jN!5A4N3jcKS1_&
zy7uI~XohQT9o$W3(r2g$Y-fNtWaOtS>FC(*wI!`!_KAa68Z_)M^hsg14@xd}a+28$
z(+2k?RWXb18#WHGD+eTmJ2SB|67YUgl$U=*2C2yTx-!;7i3Dh_o4;-vKtTW#TWEEH
zgJrSj%+dXSZmtd1sX7yc5)cqOE*{ejgyg?{D;n1*1ND=+xj7KkBR@p@?*Cpj4}Zue
zEUnDJ*Yszyhs%^tR=^8-fA38WpX^_^9X7<W<hFP`9&cVtA8HTT-JERYreHtNma8*<
zb{7>GhwP4nYHSI6f42XggeRw;F%1s3H1HJB(ybky0>`YuV`Hw~(KFEvKGVS5aeF-a
z`y{cg8?!yVEi=j*K}qjFcq|?l!MaA&_J5xO1PREX-T5G^BWp}!gZpt}!<+28+b-xC
zZ(Sb{K2TCpI-aS;!@WBaHi_OIN^EUop{=IoFuklLZ+o+rpHDjB4aRej4eH64uDAdg
zKQF`s_zv2f1xS>^{Wtxc_CxiLAXuTK+V-6bNWtSG8u&UroK-TEd{s=u{gQam9?`Jo
ztl6CdMA^2GKgOlxTf8nWBE3r<b{m(!#oVhXUDMFjy;$k|o5P@!aRwdCF0eoy$LQhi
ziHNj{$tKa!(+k+Bt83gQS?km<g}jj7#DgLLt!<}%E#e*OG(^NEQ#BsIEa_<|VOB{F
zSnfGr%nyy7Z}p!`03%MgeQnj0YtMm}hJnxZxPE1c<i}u6lYPose7rqgNrlOo0kr#D
zkNhkPYXJu2b~#<22p9!b^%p-aaQFq)*Ou?>+-J`wnDZfUz1=<E<6<Ig34ds5Yug{!
zmzd&5zi!<8^eB1FoH{FOvE8#bD4zugL%?rEZrr%-wLM>jiahmWG|sGsA*39o$Ij_U
zlnH}~^Ey;xi6^^2WD<Ck1`nyld6CU)(Tua&m;053&VV-e6XO_V0kd&^*85E#UWcur
z7{*^FcIG_D1+%}ndwd@i0pNJDM79Ks8R?*-`Ruy<1(Rv=h6Ud>zvArl;fc!Te<A6{
zM{CQN@#pe(*A8A)uru}Zd98VPdbR)$5T=^ILC6QNl81)N`QV(4gQF&3fBnZc_P4md
z-}}l7Yc&!%-M3G(3Ur9AnX_M*Qa2?eB^R?z#>PbKKl~+$^8O`Lz@6=9%#1QpebY62
z;jjs`iRw?U=pjD_$IQmD!hr@yug%ieHTd``z1ODf5VV3XA5PX~l$OpnM?^v9?}@Xu
z{Ck_Or~DZMZ6Uj}pZD3pUwPJ#B4R+gG$AM`zxjQCW;9EmK=JM4IX;6+WOclOj*V9j
zE*2bzuRGbq0l!Uk`$zHy03u>xDPBA4i?Mtv@Vxv#uHG^#%dTA)z7PpfKuKw&q>=7W
zLApewLqIyD8$?7(KtMpcQ$o5V6lsv|QV{9x-luE5Yp?x%V>tZ5^9VOr%z4fugX|6<
zh+WzZx6!S2>usOAnsJLW%T)^3Q15Kp@bK|*{+_e*`9|VshRJn+I0BC@<0IoM9k+g0
z?7#P~ug<(L3OEq_+-J<#-lLE`UKSdEtOoV6rXO^!_Tq{`0x9k|`QuR6iXmFiyXzRf
zG~0M&zEZO1aj=YzfZK2FojRzaF>q4%>)l;DIUF(S4j;Ee>Wu|h&G@&b#T_pl009Qe
zN%{r5ck(|;Hq5=AIOGdV9b2=dH)i<!r*(DpPOmrN8eB{n@v6r_2N9GVARt?tJQ^cn
z*RX}NOP@i*OsTtGBE?O_SQ>{A^Lo2d%1FNozOyw&D%D*)Qaqwf!4CB&f47(dLPP@K
z=y8XV#QX2pt&SJ^#?l^Z=2juIdS%;7fyyao`VE5Ky8;}2env_OkX+~XAraofW%?Mo
zWLQ2KRd!`{i>Eg!$hgbP;FxA-Wn*D$N#kKPR2Kt&Z&;STf-VW@Q;-p5UU+*r8#GyJ
z+1SjKMw?Yu@@lXdygf`zxND<(HU+wT9IB)IqI^+j6OPmFjN%**x2;qdmH>~GRqhk1
zaK#AG4JvhRv1C&W1JMX{9aO(;VZQ08R?$?SdC~RqCOzG!cBSN)B?tb(6YPzxiJ2EG
z8SY1vVj!Ne8{O};xs=L$nm@%?R9f-iNYHZ!;4lFHlSCRVNM}H+8HhvabUK|4r%Syv
zVGg~0?nyi2m$l}X)@*EHkWz=*Toq;o!A}1Kl2nuRwD<Y9I(E3VCImr1$~iptm?}6t
z?B=Z-Py@@y$!G0zNPp9pkCzu*nBQBw$KVacOZI@|KzK79K!h}ap9|6Lunc=~DpF&<
z=`{-dUW5Ev6jYH_jqD@tU)gV}x_~gS@~iD&NP{iWs~ye{q!nTip{c8?VrOBw|A3C?
z0bNB_MMbYL$=Sf|u(wfBkJ`Jx{r78qDen`M^zFZjv43^QA|1VbO`aETXq4}PgrQ}I
zmxmQEOh5)?nO~~y9alVOj}rb3L<ivP2}4g|y|AIUcj)T&i1DK#7s!@k7GA-PGpM)E
zZES2bFu0FE_+0B+bwp~TZt0&mVB1le0n|efg038!dKho6Uwh~vHwlB79%~PDM&3l|
zv$MP8LjEb<u<rm|<9Rq=S(nh2aZ;YJMJ43cU%pgQGWYc+0{6{v-E?d$za2(v(B<<C
zkEArXCC8@3e~co%LkR2uOXCrIN<eUFXfja@7pnAbwavv<e#vu#AbsRRO!Qv!jE1Z#
zCpXK~&<rp8y?<CL($Uf5usSj`GqMPMl~i#GRlKC#HWEz``~Oz@3F>Pa<rl3{29sY$
zS!3Z96N7Dw)Bc8%Xx+i)#6_5IT4Z`-e17Mz-wU6uERkR1`OwFT?Y<4s-{0;E_N*OM
zCJvZ?1Ody&)^y^xG=3hQaFP}8lhxiayE&9*15ezU8m~4`&Nf5k$U_mILoPw+{#ajW
z9J7UO$92mT0|9aTr(37M)8So@j7qRIGI`*YcfX+a?$(<*xZoIw(s9yU?FMc|c~#Xv
zK-@{ldq$jl&kc7+Mo#wOiIvNysMo?S$s20PElOc`J{o3pgi_4?#S2phKru&^wDr^+
zA$0P^W08ItDq%8~8w`6VpZSeeCmIs<oSnn^YUSx<H-()tTuUg;o5HJdk`*=jUM#RM
zv!MLi=1c|2_{>*FDg^V<T>`xKRmv%iUtVm4<V<x+H05Re5T_2hjRkRM#dCF`ZRNK}
z$HlpX4)@&+pdP%hqPm<?AdSjnA_1(iIcAX28Pk5B!LtVUOXX%rTk@UjP%0j}*_?U5
zk6gSa^hxHO>%T!Vu`@($jyY1QW|r(B6w)LFnyH+&cP(uS4}*QWx>@?)xL6ZeEgK#M
z$ZMqMOMn_{s@k!;*gFv1;+Stbes+qw?6M-{0v@ZUt!i)x1WCPJ3GAMV3yDSx_#yK0
z7W+AdFI8mjB3>e(uvXa`O%B|-AqyF%6U=XJUjwL$gm>DiJ6!ZPRb#{$N6KXR731oa
z@-7T0ueQRy10ysbOD6AGgdb|v)-(nJGH93pee`RR`iSXrvy#P@gdyIFG?d`>T@ja$
zt`2eq=F_(iOIl)$WFN#Q>)rfse0RNlF#T>A1a7L7JTyPkGnA1rKPD!QiHr~9I|gZX
z8l!)tUt>YY;5%N$h;X1EpebU$yY`poMKyiXzUw9h9_h~U#La<yzlca=`J}d@B8iqo
zM@K^*g!oa<um};%H;im-()U6V614de!^-kN*SB|auHH@D>9}}T0Ufyik2{eBQ3$mU
zpPVCnb3BEdJZD`&!N;iHNM2D<J#2<Q(r40ttKLih4GGjXmq}N${I)Z5=j~m4xj1M|
zxSp3?;u%`g_x2vQV;#R~5*}57v^Xs%g{KY$m(nXqiKO&=GA$;MfAvu+nKh?6@$=_}
z(7%NcQYhq7j*2|?7RBpW3{8AO4hSn890{<bbmY~ZKE)G7N0oqc5Tne$eACr+sR}(E
zow&rrl9F8fyXXHHl9!8FDwqiG2ru0SbAfm1jes)Yg$01!29B?m<y9F4piSi~7wm9c
z<l;^=C42ttc}`Al=F%aN)Q#(2;jGeLsv5;}LcXgXfW^<M=;?O+)hm1?u8&_JNWs>-
z|6akV0rSmm?{qj6g7PIQz;Qr6ys!{3o#=0QcIMQWk1FKu7*OBF-t{;+-RS?eIp(d5
zf!>6RdFzg%5U9q+hE<4XN+ANlxk&l<LpeG5rAr>L(FAvztXf(Hypzhf=(P&^i!!j{
z^oH80qEe*?bJA5f{w~QMVuBDZPU$D@8L~EzPz0Kdd-*F2AaKndFt3}`hL!00y`JMU
z`EM8qdNR(xS{xv^r07F~(}I(loNOja_R#ANZ{EQ6-Cd_oILYo%`v6Nd_Kwrq)_WXU
z*};AZ<(|~`;~#%O1CLvM2}8*8rukfd|LeazBqYl~&Bt^nwDB|T5;!?kOcoLz+NBro
z+C&c+{x_9(^#q(<AF^&|yS>KU2@fYY=uGa+Ux8tic!Q}V@3z6uQ0TD>d?MgdURhl!
zR$<l-kfFHJxnIgS69JY-8e!BUJPAg+^ocLvPLU;WvItS*X?S64>Gi7bQ12}Kdcao>
zmJ$Ki-Tt{sG(6<Tx%O<C^P?BKkLFw2`v$sUR|#ccT^Dzn-ug=ceqU*fseJr)cAZ+v
zuqJawYp?Ii!Q2$qMK(5dwwmSVdcQmew3WUqqiMbU9xE(P5}K}(XY<G5a%t-0;Y%Ne
zt9WKvC&E>jKMmISDo3%LR#f~uOwI74zRyPu{>?Q{Q7IpptP=c~=lqPh(p{hN-h_(_
zVI+75Tpw=D%$&~EUgb&i*NdJ-09!Y>s~w~yhqbzvSbYks-M)>~#lF4jB4Nys>D6Fm
zXl+B$$a|)R`{3KQx{NwK^@yTV8@k%@Ch)BWd8v2!Mw8pgDV&f1(M=q?5?#)#RO7dk
z6t>G%wwxIGm{SblgB_!T(_Wi5QbjyLF*rH%?dR<1>GrI+atf>i2k3P#QX1W2QVoJ|
z>g~^b9OM#?GrUTUd%^cVI6?+|fL_xC&#Pd%&VsIjzZA2P^>@zFAN+$n$us{m8ECJn
zyl1%2f0bA)nfT$8k*(#q80)o$XV0G!Eb?5?VFhI5r^iAM=kZE6h-Ur`vR0=iz(CPq
z4&*t24-iRH%{QXFrsLV&HpwMO-@7%chp2p5V%{Z?#C4&f{5&~Rmqj2b_;apE?Kv|u
z>_>3k=vY|%^B1<WrO7biZeot*xsr(EyJ2NzPvHz_b>&)eF*j3p_G>-Zf68SOAD<U;
zccEePb^{OBu(bFtKOfYrn{YvNJOUxk;LiyXwDb1%mbJ*Q%0VYx8vM(6|1CXJeqmM+
zQNYOPsAQ%*Ui8zf>@4v1LDDVI-%m(osAaimx#g*Ee~Kp!#^Z~CJ$idN|Mc|qZI%hF
zd_0E_0UT9SL<>ta-<X8{UN>xq(2(HAep2r2)m6n+qJ*|hZ{E-V@sPkC5|vS0MdW07
z=X3X%ExW;2h$y{k{y<aeDuw`x8f=OV?>O*e9eJ9FjDj2ph(JRNnx?!=#%EXwsN>qw
z_9icxo`HGO30Q2{f2)g=xOw?Nd?G6&2LZeUGo^r3JC$l1+nPSY3NU|uCs_&N=80z7
zijswd`IXfkQ`nG^ELToQ-3abvX7180*45GyM8>DI;e}2;YKTpV1Z}gSNm>L0Iedj5
z+NQJ3B-Xk^=o#n}<L`#*E0Oba@v*B8k4J!9L=pR-r_lN%_6*EJ!N$_3`VFceA_pJe
z?cvP6r;G$I>09|8SC`B+CW@OqnVS{0DP-ffnWMK%W|ijS6wEFtAk~X|*XAop$Ta6L
zVY;4(7xjqnrT+ht$-mG0uYNQ+UPS^D4_bcb3f>t#JtfKs<B3*XU1P?NAgH<Gv55jZ
zbGly&r~}7qaw?VZ@S+<6-84VNJC#>ca2qiqDi-romn#^x5RXn_)4MsdUR3X*RqLf{
ztKS$L>GzR<OgrKySajFdhBFfz*pU6zfua&MTmmYX_^a#vHv+AH;QWsk0Mm{GzD2tK
zjCc5^6n;1inBf%DD|~6G@9Gkg!@vrdpGR1vRkyvA9r|}=Z3W^uJlt#O=@>$(MxB-}
zX`$NyMkyZN;kF3M;@r;{zXrGwMle0Hu_cNcmjy?D1787@ELiZ6B7hlsCwuHWucnuL
zrtOP)JhiLKj(pw0B^5V)3!fSyCcAd0Mz=BKydSrm3v%x}RVI4~96GP=%#V;EGD@;H
zBi`#*8bQF#Lk;=RrmAA;V4pQI(&Ez@(2L~wCszj4uAGi^k33UX`LGCjq|89vmDye}
zR$6s_uZ8mnBlAE444pCRz)mISmi#M5^?cRG2jY)yDe`seQ=x<9g<(`=-KOYF=-^Am
z15~7G>Ik4kqMl?ybHl6FCO|Sc2<Au}9LR^`VWTEj8sbxaNZCL}O4+%zm#akTSww-q
z4A3SVv)kU^EdcA*UZDk|SpCZMxsrt8cEMsOM3dKlO0=E(O7yach+>-)P?RbgT_8=L
z$*au_%@8swmREqU0P2vq0TH3FQLIL}fM<iH%ko?nRG3caEg-cX{$V~;HwQ+ITl-Fe
zWo7*yX@81*KTIKJ7{X*wKLIIOJa})9ao&f?Lp9W;_X>w8LmQyZg}Rq4lTaXRJLf6j
zyLk&dVmG;?4SX(FMrSb0pQ>A-fwOq7t;+$FIzYYwjPjmB@K)I5(7>qVn3SsOL!HkS
ziAuS!LF!HMD<q02!6zQ?fYcZLWly;%@X-4>`X;Oznq`*cmV51-i&q!gai4JBBhcTe
zzkoiUK7+4!=mWz;g5~A=j=d)yK^cV3TCwNlH0T6g4M>N`%bgvpj5H)Bt0_yvzQ(t8
zdp+<`?m$-d`2Ws{fsg-xE(-T{<<7_$@B@b0RHg6)((0(Fyv;#XVI-iZZw0!1a<Z<V
zV6}Sd4LOB199{PS!M>U%V{sw01QAXX_LtJC{k!9HZ3Xhr6wV{@b5$#7XqaPjsl}VK
zBfu;+>)#IM9B|b|K8k8rb6cR2$s7QZ6`mkgKFHg>$1L<*R`nj2Q1REVkEEnJuk;=~
z#DowJ_y;e6f*BrlzqMXcoKt6Y6}dcKW9z~f(Ao;ZNhhIXT9&=GE4|Qo0^=thYs=;^
zYD7JKs;(YW$-@g8lc*>)&}R6*dF*+#I@@GG1uTD<j%$V&5isH4%GiWfiLJ5r4vDLI
z!X~g#1#b<4D!_(Kkpq_>uG{yFc^l!QbVitJtG?thG6_Ol%Ogn8yoGrSV)Xz7^dKgf
zpRSLCrzK4VW_FO_pdw*mk|xg-_&s$X<1OvOgDcl>@B~8^(CTWzrtJlMtbV6^fR_Qf
z0MsKe7TX=>fcL|S5ku%Mt(^2y?g9ML!SRvbOiM{g$-r{Q^KxDjLS6tdm^;SEs0Kjd
zeqRjP{K{<W^q-#>?Mv+X56|WTrJ!Fr!mqe$RL}qMcQ~}Y?NSDcuBU5f|JBCKXHN&q
zQ?QGH<m6xaKhv?dCPkCd!-tm#QxfR@_r95+2fXFw!z-D+4l*Xsx_WTsLHyVqHgpbj
z<%eQSg1r13RLCcVcOO3z$f=6BeE=a<`nx)9P=ShQz@zK`%}}!a$Ii){C5}0BD@!m!
zlZ(2y)w=dU1MBfyI@08}n1!zgxR0zdy1P0)#>TYXXZXy>;$Bm4X<-SSFJSN(E(Q5T
z#tswcjTO}we+@+U$HFXi{Pu@mF*|tv8a&8u<CCSv83Y-z*j@O;lN5;KvIU&dcBf-%
zd0f9V0U^OWeiZ2aQ6%1U8kTC>p<Y9wW@ZS7@U8>W;m@9c>Nx;^Yi_>--?kd%A+)*L
zq9(<-VS#KUB0LOVExW92cYpsDXh-2h2-OMRCzEq+ldwbOWaV^rJA#>@%+Uem^~2n5
zpW}I8C$R#uj&9@L<Q@Xd2IT{AaD+u1=2B6AK98YbVPWa}Pv63Jf}mQDwav%-&$rKB
z35D@l2L-B+v4g5d#PO2r)Tpb|-eC_*rHqU@tV`{~kuUKNVoYFVQQtiB5E7~p5QamY
z6`yr?PZ!E-b|O=7hQgx*;|{FR{k&LMI7KBTz6T7=-+p?0(D7X>CV~?cfWH}Pl2Yxl
z8qaVj|4g<Fh|$tofj4?1ULCj~aVTvDM<W{((**!kv-t&#p6)gA@pmEX1V8((ou#47
zr||IDO#h;sI&BwQeUB%miMh>)1js|&^VW>RuHU(nyj_$VlmD3!#sjalHU@_5pvz{V
zaV^)<2Wy({{}Jacf#W$WJWP73nB7|{8vV!r<-W6(Qv^$T3$gAVUO^|1M;)%);jW^)
zdqBm<k6X3IYy1~5ik`mA1`+H<ebubzwkCmV^JU8uH{8v!F^f%(ZeX;o?Ek;at32cR
z`*qb+INiIHZ&+H`?iZ1Hp3Sxt`EE*0fdGt8$@&%6f4JDyLkgkr=jCXxVIVks<z&7`
zE8aCZb@z};#lT>$W#C0n?ipw($k%Ng9QPi#&^8q^!U_OB8&IyoE0#RNx<19l&+9_g
z-)C%irL`C5<f8Xw%TF%7AfM6Y|5_NR?t~8*j*nIRsjdWiB#)Uz@1TC;iv6yl?di2Q
zr|$xl4MaS4&l4|GA}%mtZGz_{_ow!2Q;io_?vOy5(!H{-{hx!8I1LQbUrN>&t**j}
zcem~9O{tBpP(XO9D4Y4^3U_zc!6qm2e;gX(zkM105xnEBkJ4Y~)XDwtbo9`B>~2_*
z^B6|87uEExsJmNV`_|#+2pAh0;Y)S~qDvc#JEAN(9cstZ5rsOnPABtd2BkQSW)V|U
z+TMN7uj<|0Yfe%`N4+8{J`&U=Q%>*3M8<9whQM8dZ{X})(d5wyvd`|Av$~f`P+D?P
z5N?`%uqV0!0cN4eASPh&A23>4bTrs2e*R&}XLK1AB%|&8ZFcfL@w^RsEEARZ;7%J7
zH<q9&IcVY=<2Xq}Tb~++h1zN9s?zyF!}{hSMgNLZ()<6A>Q=q3VFPd9NqgL)sJGgB
z!NlYeT1`h!KU$=H*e3pJ|7|?zhkXy{4_rSJM&y2RoO;)HF&s8^v?Y3+(5d?}E8CoB
z($(sqNl8<3I6X5rVGH!lJ9mX#{!U8)%O$bWQ`adYKV)uj_1R>mvU}d(ug?q@z`apX
zQ538ubLk5MoTY3o#g{gv>3;g5$%9XyItgVqELJTxZ`yzmaYNjv!|>?$Yzmu;Tp4^S
z`o`xmAor%4X*}&r+q-X(_t|FVOI@WaRFnWZHJ0K4VCW*JQz2x>So773wi}D%WJIbw
zJzqCK`)`Q#%N0IFue7nLQDH#=n6L23@t*L=+ZJRL4#P1}f*KRk0~$a2m7?fwMA3<L
z`Ax#)qo$_!rt^JB@X*M>wbzig$fsPCpKfYuN%L#Z|Iv5LwN=S8@B@J3=la9z0dLj+
zc{~850SSl`X)nNtp!e_JmiZV0&_ZW+34SQO#Y+nm5ZsS|PIj4=hZ*8JpNG>Xf;|qO
zOgV~fvgC7SZ)6M#xCd8>iezNvmzX*LXJ7jBSE~K|qOjG3tA1kQOZNBxr#%c@85bbZ
z0p}V7Yye1h!$g|>Exi5y4cN&~JgJCOpK6c(-BCc#O!Q}yjY9FeS0|(mhu+p!dWx?k
z+G$Vo;X{k0czHZ|?)B*D-UY1Sy6oVTeZVYK<*uZwh|LS<tT)ua0*;LG>8VjFKYXl6
zh=hn+65k`h`yp<7pCQ`9)QEsARnx#A;_!<WD~x4%p9J1eYkw|?$(Ay5z!IX3$ffR!
zkynsIK|nV3<~z*|hH@+7J;*--F{rA|OOq@Jlm)LaZKn1U7S<j`-eL8dH$xLSR<j^^
zT1!hE>UDZ&W4jl#a$csXrEzP2iigcDsN>UMSq>lg&JXQZe=c@p2$$#8NIESA3Go5K
z&cL32k{@>K092-+^tfuD)d-ZDsIW7b`ZmZU;~F$G<7Z^(YUJvgX@r4p^g3_cPllF0
z2=iJ7%?#TYl<vc$_U4P?A8&S!#kP*7_O=B~wi0U1isL>1CRzPtxyLD6mp(C}X<=cJ
zljRdx@$x0~1wTqp%d2^IYIt8~4FVRGY>FC^u_+@;3JQu#mR^|2(q%vxS!H%WKrKnI
z<h9p*21D21pqH8iOIjOQo7<WuB*tB_iQ?DaB!Nuw7+Zq>LX_Bq@8inz^EW{}-~$2z
z>q52z;Z^6dO4D*dJWy9Xo8T~dX)z0$m9DO>JY?@?$d@w0;8v)B6s>oDQ8rG67yY26
z`S&Y5Vb{UIL38~tMK6AQd3b{=x`-9R)D#{llJKglJ3EiO@3A(feGD`VxFtz|ZmOej
zdV)h%D3P1V%BjU#m>JDHe##&VZDQjSntR9RxHn{zCMVmT#stwt5glZaVZ}a@mA!!4
zpwX1-4-0m007^d+h<=PC_~*la{<{9Av~=eU+Q#m=@8%iw<~=#>s8X~_Vh980t*EC{
zvgpu?jRN_+9H^lyT*v;BGIW3cZap+fXgAl&9nZSXj&Gc~skcETzSkOFsNYx<+K?O$
z=IAn``#M&h!ii9D2KfpgXD)YGfhHwLzwX<d`P-8O@$XO7TCTsgH-17cToKrt&&$(4
zS7os=R^h}WZ!cKU4rhoTU%o*h*3O#=fdA}wZXo^D$`PI6IFQW;#3A(Qh1`<jDam=t
zv}tR$E<6#`I)7503rBJ3)~}CcmBB!E-0z3q9x*a95cxc?!F}Tq2F1oh)KezLe*Qvz
zeI1=MdNvAVUwCzP=argDnp)mRD13$-S>=2Kknm`Ao{g&8uwfGtF2awToJ_c*3T{gJ
z?Y-s6+M%$|ORH=DN@)^DOTaf4r|1wVn*}uuiSY?4Oa!^I3(%w9wm2$3#Mp$3hE{i$
zJv929=ux`)&s+CBg~^^hhvMmt>EKz>x*Gu*@k0!*rPbJ>L2XzrOQYBa2fAv;nK3^@
z8-2PZ`1)ICZR>`9d_z4|bRc&HmkPrTCT6Z2Va5GJ_xCdJs5)E@4+yanXQJjPkczvG
z4x4cpba!k#wu`7P-C|*GLIwJ$1;QqOp9$SYlqhHHUf2p5jZ>3`)j9pvrI&gFNjJ#y
zipzGRbAVj->Wa%<LF+HUrS~i;u3!5-T#5?SMSn)^+rue;KQJ{>F@J2B^D^}B<kUp?
zU7=U21k&Jn5DE)tEdm>2ZukZ<=35G<3JqZnW{#4MUJe8aEX{m}OOcgcoHO-s3aW9b
zZ+mrje8(Dv@e-=Q5;ljcT_gIp8G|L?wEV_dpIzm#xE9(<DO7U>Y~h>7w6{g;d@A>#
zPI9eAS2~cB!rA%w+Cv#lNh>K%yU_|1?=u-PK3G7PH+BX&Uibbk)T^zqic{KK>S1JN
zSM0O?_37@d0Jx)2e96c~0O?rk8`ee(q@N}PxpN`xbGD(5iHVEPPE+ss^OX1)R0J|c
ztBduro_0JVbzVqVq~jv7XCKND5PDBSLO>dxvhIE{xKuJ15+c99?-a)oUo!Pj8d?!Y
z+d$IDwXc|EeN&gm_j!bEGN2UL*g|lTpdA+ar>-^ke?#RN8G+<maPF8YrW)CrzQH7b
zc+|!3o$SfNaClfz!uLE_ey5dl%UZvBkb}CWINZw8`fVKn@};UuV9@~<u0P}7X236z
zyl$<J@fK-rM)4-6XfwCWv$C)lpPJ~;tN5dwI>44^2;qnNPoIWFx*}hkeosSkBNQ*P
zA0`w1OpHHYDd$18vNC>YZSw0rVKSV$x5ER;B4x6MEY_Y2he~E}Y(Q}J)@!-!pYgxx
z7OthsKv0|FAVc!W=LejFAYTC1HZdiVkODvOa&sip>b`JdQ*<%8L&v?q|NaYeOPnsp
zNA|IA1gV&0lkwC-v%yj33r0o-a&edStaw%;vTW&0RTK$z^_K@B^j7>EAbv%A&C3#3
z3`*SYpBG$|l~+NZp#}4(w_Es#^7qZ_HVig4FFCzK77Ee22X@bLoLSn#NnIu`K+-r<
z_oDY(X7O)TPDDjttJeLZ3k2%OYZ*E&p5+>&4WdSg4<D22?2bnFY&VhH<fIR2qTt$>
zIgjC%;hcJSki<{4E9>13ZGG`uRf8kNq249B(+H(`Ls$%fd<`!37^;d#P;(bjwj16M
zNRKD$E)_78bt5QHyV_E#r`|0tdI_O^z}U0UHPQU*OV^gV8bjEXzIvs_nHaJR*QQBK
z)VG$s%0_!_<)ZoQlJ*~9IZ^sj&+U(n>KbPvMSBE&q|9f2YvRz>WZ?;Yu`Oo2P#>BG
z7sJDSCNC7<p~Hr2e%r-;3uh@~plUe0e4Js6u>pWOP<~Lwt#SDfWFfEx0d<2saGW$5
z<0dv<5JM!6ijzx|4LG;vjb~erHfD^E4i#;3ScpQY#N1bS?9aP3P?kI1+Gl!Gik?+g
z-wu{MrEu1;8)v_bsV1IlGxznd?2HZh#O-M5(6uCrp^F$!f4T!Jz@ss^p+7n#d<Q}f
zScwv4=w|C&e7@HVC>tE3Kn~Nld9-rWH{X*?*`iV|1r<rRoR0<;%s5b*$EKg@lt0Xy
zFYYr2MY8A&4kO_$sD49z%}eHrygs^USEib#<Vmx0`!j{&?TNcQ)^0IMW0K8Zl9DQH
zRWMbVXjxD&l+3`cReNtk2c0saY0EuKP3QhX47gD0TsLkZ38bqq!U53;ZAZGcqVqyW
z<Ra)u${2_2+;^kW>P?}$B<H*Hgb{x?@4H>|S(>FqR9Z|F6X%zy2(XaEu@VpbJI<%)
zgoqm%zU#`mz4%oc+{xZO|1M?_ALOmo<{sHm2zcO?mfFn!(c-7DDFT{31DypM^&(vU
zdEwIV=(_-E10m1z>1-=&KC?*z>1&Z+JQi>$BbHV+9JdZDQ7$~B8@%t+-heD6NbFs%
zwg<a@;L+Asrvr3{&dzQY@K+@X+U&kR=~g)k+7_aI`-nmKJ;g{?^*5};sxzq-|F$RZ
zZZ7o#AI0mys@>pK+VUxK@RpLJwQjkH0PB2D01I;alwq|CZLIkN|DgMag{GT$k$Gbe
z&zh2N{_p-Sc-e659R5znz~Gr~neSNa7|OjHvP|QHuEd`89+VMEx%USBdW~GGs#|}*
zYN}$*&&mqAaT*(=POWjJKT{8HbjzW1+Kn5CQ(1Z2g8iL=(&eSZpk=6?11MwUftFFJ
z0F4mQ9f=b(DamVo3N)R+{JF}?`s!<ODaVx~Vd~1mM6j@&c+PCF;eY+3NaxWDRdYWy
zQ6#F_55*!<UJxkQ*yzD(!hdG~Cf%)Be%7x|SKE=@MS%uy06hNJkZZ#xcMT&gE6?28
zI*6P|`)ziB#d%;vM-%PI&=RPyNX}mSV9(UKuro7nOf;~ZYCU~Anq59~h8qkY>VBW@
z>RfJIR1{6XBNRjm9o3#EUQ3WAj%{#w7?4f5<4@<^vQ0RcF8U%R(MA8_(>JV(Mn=^A
zbzI3T?u<8bM{1G-rCzo1hmZhZ5tB~7=R_R(Cz)2z-|T8PHGVvL2nun?Cs{UAwYMPl
z6#g*)f_4$gEuW#!vX47@x>9((xhNw-m1X4mI-*2TF<x)#2?xI8ulevrFbOwUR&U*=
zg?7$mYgABdht@O$QpNl+_R$b47DtBYfTXM+M)uDXk{Tg+YRPQrFA1sBkKeCC|Mrmz
znKy&N5qy=74tqWmMbvZ^>Dw@4zQZNKL`=9QezvV91l||A<UCQsj<%TVJS@pqY4E&4
zIiLf}5F(K~G-0h&U3@6jnmtoLnP~;UY5n+A=U;R3w)b9Vo1(c@d?!O@kutDF<Nhgs
ztvv+Mmp;+tD-#X!PfYo$_PRGF6(`2<G2i|?tb&J0)2@XW)4Hh0Xkq?J>08$T!w<aD
z@TxXnVdDSY9vsd+ch<<mWazZZ99k-v!pltsHl`YQMRqr!JWh+1FIfo!#9PpDe-ytD
zdjD-<r>X7Q{>p@R%pT4D(aN4i|2tYVG@%?m8f+*bTi7e@q>E_BK$v5&6k;{$h<LcT
zjDl-=or;QBmFmc4PWDgMs`p)#lv-)-#9oiZfJVCPr#3c)6_pkJk)K?5%xM&Jhps=R
zi<B6zLRAe-8?r$9Gj)jHFYe&pBE$;dvb`5AW3A72SY8~?a_8zryN^(qsbN}R{6ky)
zhc8}mB4Bk+5wsV8sK~uIp3R6sGK}z)qzAUcLHGRm3biiTM28M~S`Rz$c%>=g*3N`p
z%^#PhiupQ|^H27P-&K>e#C-*!_^NL-S&6O!JEF@_g_7S<#F?Hh^CZOPpdXHsD}nKY
z(!~|{Cx$Zr-;>xSkTT9m8wiTN>aSmS{|NF=$g%K{Dr(<+*S9c~$Mca#`BWfPIddSv
zgb=xY&ey}rQS$mxI>enEZVvRW&9=bQ`?TXvVbZigUQo{BkmhRwu6POR;k_(JmS%xo
zDwc(~=oqy8;$MM6P<Y(7T3l1ZQ<A3A_88Oc9Og%d-QBSwZFI!hc^W1<+HiDHzB7I-
zY`^rhW~=;6D{RK^9~+W+pR`Cxwn3v$)zuA$^C9Xx!MD(Ad&>Wx`JiaNn#X^_`ln)_
zC%@<x)c?TTZ`L*}3wb|G93{6y=|QA)MO#erF#Y$ZK%95GAzR*_S7v4(s$v=96_^=W
z*;z_}AcclJ#Y>p6AOH+I`o{K9bi;}o1NhE2#eQA;$Ku0_ZGA?md(#3cEknAt8cX~W
zy3M|5T0^YwX`)<qzeNQV)jL$CVrg{dCJjggF7d?8M$>vJUJ$NRQjqcfk|DYilGqfx
za(SC_56>RQXgq_YIK5h*`pz97GvI)4H|2dBfh42R5|=im8v);8;uiNjNqCZ7th4bI
z2G(~)o|Dnf9YuWpt~s~#UgV5@_;x`oSYos1aRVvl%;OGw(!!5!@%^`U6&X=t=D4Qo
z;O|n91<@nYN=X*?ZUk$NKMnmLLziwx^LS1YFZc(HtD7^W77*=KcJsNA6d~bDf*W~-
z8Q=Bgtbs5f$v^p`NiLa~(O_3l&{Y=t2A27!+I;2`bL-RR+a1*yN46}?%okCI_jwfo
z1hKSQdvFu_LvFo;e@l<ZNI+T4m+!ex^R6VD3uA9qeJ$Jd9I2Hbn{@XE0XT>a7!L1H
zp12&oXOfc(feEfMa&8OhW?>!~c~+WAO1g^W<N-cBtFb*dFNWxY^K5a3lo9f#^+CpV
zD!;1O>EI{-&pA`;_~0Z0w76>n7ahU=ttRRkV?l3e5Sh#U?3DIIPG>x5DX#&$G@G8l
zQJ<;M-x<$~@H*790~7=e?KIt6*i<R;76KwN!kD>SMf>ZB2&K5=>BpL5$v;m0lO|+{
zn^z;`)nw41BK@0W*9p?%o@D<_2IfLre<|*fm26rkS-b)|0)$ny<9(HMZ@CLDsGs@R
zm9=BMb4;~5f1aP&N9h(NV-sx-5Bl-jh&?mq`@eROeh5Dq%RI~GE_WK3UhdamRDJua
ze;e%hpyGzYt)*6gAHMGYE7clEF9DKo0(2SNNg0m~pns|&H~f315sd0-Bsp~nt1Al_
zuwbJgfv8vk@7otTO?JP#m($_hVAd20bJ-ZQps-;>evOXB#l+nVXcB7$M#_^a)$LF7
zPasSf?uBSykM5MDVTGR*j+0qhXWnc6RNR+@)A~WA%=Ony_!L>ILZ*gIvmoyJx4`3U
z&H(tp?^ICfz7vlhC^Vss`dWn%($l|vzlVaP_HCtQ23R9EiNReW?6WTOV^d4(lk>*(
z4aJU3<M}$5O%q$oYe-1Qt#<oN+Dv2Qm;59@a{yPGFBPJqrcS|?=bM-vnH<gk?S+$x
zlY*=TNq_<;H?L&`6?Q7egZ?c{D9=->qu>qHWDO4wOXO?4qqUxOV*#34pLUrcdHF9A
z5);ypDVPUa@V7PfsK*5T!#yDNdtnNnB(RUzFhkd-ZORA##06$h=)>x`r5mwXUGX5=
zz5!++U>$H%ze-#c&up`@l-%|E7_7Y{0k^*{f!XrCjp+*P{BT&J4aEKD;HZO<=fp9m
zyGEZTQc|-0XZ*@rtoFh0>w^;ugU*~2_PS2XEC7k|g(`*nUITrlf%4M7@##_zTz0H{
zXL0<(-2<enHM%&tL>CKJ%M#4qa{YrTuI$x~m1>v=2R6FYwDM&$NmDmIx|i12lx&M9
zZeGp(+Hw2zqGP-1?;aW~=6lrduXJN4w^X~0%C*+79I>1r@O@-*i{l}5ZQ|rsRTXdY
zr|wz(+Bqq++2Y8B8qfCM!&p}pq-|I1Z=n)XjfFz~l&ZDlY{10C6rK+8uM4Wro&qzx
zv$$5>-sg=1sd>Pcxh;(z;21yoAd@vXxEsC1jsm%ejqNez2vp9=NE&^aGe!hFcbG%m
zu5Ij*JOp-K^ws{Kca_~5tU7508F&<AuCp6L2*4;a7`lhna{^w=)M#iP_OdINw8d;<
zx~EHL$3H;1vxF<+F4Cw%=af(>Lw$R8`|Fy^=8p#N$ym$((E?13LF1b{zpJ8Rm~r5t
z^@Mx#WS3F*J9R{YArc8B>u=&mSPjYrg>&N#r5F03Vti|+bY-W3EG9<XvVb{m#x5P3
zqtD<3mvVUP8$|?=uP`Qo3Gb5}d%WbTiIwBVyWIWvdg!{keH7#68v_lA)_j2q46wtV
z1op_#DegjQqQlZ`i&{I*tJ;GxyP1{QO72Z*2xF>S9&Hu7$9ZC<?crJWXNk(%ByDTu
z_w(oa#YOoi%PZZ}5HYe{cecG-i+^NmP*k#{=hBp&Q}AzUPCkpsF*896__|Z$RaH~*
zF}*o&!2#SyHxXC{k3>uEmBY-?VLff<ChBpzI}%+_z>F5rB1;!#CYFNt;rjIe8WL^9
z?RbB2(eXat*7#V;bc$<zqxarqWD7cXyn^_Zp#o}+ER$iY!`M=1ZL~C{TWz^uCG3SF
zVpJn~KsdsI$V{1>y`*q>upbjULXz_<B0+PfTEN+Nhxu23BL!g9*G2r~vdP7rQG^$%
z?h1KzNHpahNg1j7JRI2NO&mC2XnW#nXTZi$0{xRB&X-)fsi(0i?obD*){Yw+69=O7
zfUm)BJIv5oV*OJcx6fOc->ff*nIQPh;}++NsSm8ZKRY*uK6^e>AV(zI+t@yF(+Ud{
zBUN_#{D3JTX&*Y_P^I0}7T^0<A-?wZjPpo~V|%PKWP;qg6AW|plTABCnl58gu+5M7
zdsc~WAA&BDHd03Th~gnZ&suYTX{PjagL^=$=@~`<1jW7xgB+y4mzie7eUxV9g4eJ4
zw5POas$`A60-wo3;8CNGOR5AlkRr7ywc}p}4SfR}gLydlzS<v%l#QD;mb_#Xf<TGG
zwIM4wP!D3DzG-QDPkOi?kg7>zTg2~4wl`}(y&M<u6~bI2`ciqk#SrVM3Zgrl1YGVz
zb?N|$I0{roo3ZAd>sKE2?_gWSawPCW3@_7f<K6snTcp3z@xDgf1=Qr-hKfi_f-{`t
zps>&;x5ZOPETEjJ*HTw6%F1$ESq)>LGOWe#^}XXpUhH4>7p7uy>~xuD-6wtFg6NM?
ztYWTAhwW>}N0kUx6xyCD;(ye529nEg4~m3TW;PbU2<j*<?~V~2@VU(W#t&a;x1JlW
z>dw7|8Ss5_QV4<J6}FJiyf7VWE13g+cpyTVD#^fW|M$Y6JzO-9^F=*;^vsd^cd0rq
z#=D2YRfUNUY@MP#k0^yG$%uWSapO)4diC)u;xR|yQbE6;Ea0PXR$k%Vm;NJysU2s7
zgePswt}%I`!d-;2@2Ipo;bRFs0&KQCT@;RxxYwepY7P0csGDKa;r(xaU;Th6mmp}A
z?)Wp;!~IC#$1f@)qe6I_3L~PgWNGxVv03ZV%^F?yip;ILbssXcR|rC)xa(}V-C1r+
z`wm;I#%xObh-gqtyUKhMcE?G_iy!%W$S=iTdvgjJtVHD#<cP{M6CSNcw)kA9eZH7p
zV!jm%PDl`@$G!S`EBw@_O@$LREz><&vbO@oafm|oR8MAxf5@Ev@|W%_nb}U_8hEjH
zs_y%MEmG!Va&$ZKcG#_31f;GykQ}jp+AFMfUwOR7Maqhw-JqNvV1U(t=&PS;A0TsY
zF(!t&tIMO#w{zTRwsmXTHA{#1EYvLgBH+)f$ALK5H*Z??W#5b~vJkUKAF{ZG7EmX|
zo*TdF_wHSt3Zu2z?N|eZGD1T`ZP8&(M?t%pxFh!XBAwus0KJq{dftm5e^?tqWnlG*
zHC+XYIv5GcDsVdb-lV5VWUIz2nBq@LDQiGq6MwMzP^Dw8ao~|ceCc*>jnGSadYVWT
zwHAU(V~YT#+>x=pewv<WPks*l9n3ha>S^-p*Y5pV^aojJ+OMIF>oIbwCJ!{N!}ld4
zC25#hn1DifM(~UojYX3y9>NZTMhSf#xh8<LC?eBPPOYy<?t@^rTj-g<3y5Z8z@}cD
zsgJ>zTU8qnElUSF70C7ayUNY>Mux5+k1<q|h9P>yMC<+Qb{1x)UH5Z_Sc%*LfME+(
z6Bt;GP4Hx;q+oI2890UdZ;(Z@vNN46%=Zt%SWkcz-<xk^#@r<*N5_>9Q)u)8fd12`
zr<1Ys%A5oQ+t<C=a;uU@Bx&vt5)iDe)|?-*JqDc}q@pFZoFuoglevlLM}9LOoa#Vu
z0@XA$7Do2Q$(4koRZJc*K9f_#R%?7UNe-ivkMdS$O<GF(-md$^Pqfw4Yksc{xW-n@
z+UuK8=f?c*Zq$36%ujInhgZKVu~?DUu=(L5wtskRwU9pdEqI7n{}jzge8p+<do8zF
z-z!U8wQ>L(p~L-mh5Vk{&Ezo<cGOVL%>8{l8C-MxR&na6H|2K)e8)L=B}l8N1AD3A
zh2;_;zCcRi)+Jqky_Xw)xI8M=7B(g#(y%E?&TTn9OObdrEPp}v;bWAxUW?am&yn$N
zE(5vZ_jVJa#D%?Wi`BRaBEf{9yiWiL!XQdYTCwh2v(>+2@XLJEcJ?G!AR$m$KS(k|
z7Qc<xekY~cL|s~R#yHSt{C=bDUh&(APHEhA%LrPwcUm`Nm-s`<ptDroaR@jElc}eP
zhK>R_(WZhNdauuiuI?FL%m&k*q3^GGPfabs`+;mj0qMlLrHAQrKs$z~T>ZYi;1Wn!
zYAm@Tm39R8N$R|Ln{Y46YYpYiOVc_92CL~mTF9@X^%THTm1pX`AXPQYWAo2|Z{4d0
z08YO!&TUTpWezE*-@j*qw7va3>1D*<lkbabjXr+nVc7m-Gl65aZb~jr60?Z!_1}?p
z9C{6YM&(hmSs*0*84vTlO8u(@9v|>S70sLpJ$Tz`5_VTeMs07dQZcdjH@UdG7{F7C
zJAkA9b$I2Pc&6t=1p=t1o&nvj^L|n!Vk>1(wV-b^yQ`CP>)#P~lp&)xSKpV&sI(>{
zhR%o2B3N=TsUOhjyUsL`GCqz!+wzMRE9hn?=9Q8F$q{$|dshDQ*~U`VGS6hU^z5uy
z#QC{D{2DoVS)aB!r4HGkh6xQ#N?ZyWGrUDh+-q-Ke`KKqlK$+ekU2gwpoj8H2a0XT
zAWUh}CnLADu<NxX4HF})?GAwf^>KbR)Giij70R&MMlKdEIOy=;HcGrUF;95iYIGHV
z^h}?6Z1%ZbRhgEuq5ppJ>pbgPvBb6`4`)RC=i|Tzg;pc`vqWYD&cdl1{qR1V-JlS>
zBn36%6~n{5+_b_%9-no_Qpxv%4Mw*^BrWE++zv!&X}z6!d7XfJ2=6pv>R*9Y;<-U?
z!6a=RGI{0$asS*cb#?cX0O(v-){tImN4tVG00gALl18SMg;{w)IAxmFz^_JS&?GCJ
zogXBsFs_?e&HVJMfwD#5N1A*~d?@ZZOkBjq=i5Vh>F3W@H3tXqL$JX71zR6hz$0vD
zXRi1kM&-Z^t12}AbJQ9#GN#SMDKLCa%M+mG)=`ZdcomZV`XwGB*WGn9G$j8SP}zc|
z_O+Cvy7Fq7jVVWvX(6Rm+xb2J&6{foBad2CtPI^87nc?*PMBP28PEMnW7qo#fC1w*
zxPUP?0x~xAx8*~efu_D{dU~;~N#2}N35}S+hsc=L#Zj_V*N>Et3#kt0To_OZu;?M~
zsO7i0l?BQ2+Kc$UjxN*Eao}aYXmLjLU0pVWJi7P!b3XJT!mp#9keFU(gAdT>hRzck
zz2wMr6rwf!?%!i}E06s8`&A4c>tPGew$9=pMn-eK3YWuwEX=JO9b+Vryu3UD%mDEF
z!XIX4aRtQzc_=Dlk>C5v?@)UHw}cKM`kT{H{3Xe^yW?80<l$efTX2P{ik0N$qZt=>
zzWp998;7((=W8ei&!OiDB!B1#S3I4!F@RVuN-3aY3q1^O`Mp{*y}z$ecWq{RI4c<)
zj=LH*Y`q*7D<DiGtKRTl-tbFug6=BQ=Qz==kUhc8bNnp{ddsW#75H*vpJZz5Bp54J
zX6$+nPiaPK_866_;`~DgW@_+)>F)EF0H3Tb)%s)nwx=O>w3bG@`tzqxRnp8}E<IP*
z_?%zBXCcY<Mky1Z$rln$v!(WmQ`aKeyF0q_OG=;^vRQ~7`SO6f@&Pxfx3H~O{QgC8
zEE2LYun}N2<z(+2djS@&9oGA{yQri%KYhTwfVm66U0#Kp%9wx(n+NfQ)92SBWlSty
zFnE(A4i0-0<VvA@EsLYM*=9klM%KnPJlaE99BeGo-(F#3`<OE{y@%5V=mThi&0kuC
zpJ<k9u7e6<n8PAD;UG0;?lbh1Clp|v5<yR3Fg>1yuyl+(O(ZBHrdEd97f0bf46A5Z
z4c_Mw(N*)8p~I5O*xkh<#|d1IZ8e2qGS4FIQbjfL#mg5sh_fy`DLW43cm&;n_npvI
zpmN;Em87NXzh-c8G)WjJYn*ngw=bHSF6QMII1LX;g~tNn!_^4m-2r3>!3t&+d^ZL!
zULm%afE)*I2Ru*L{Q7^k$(<T&v@g=kxwG8dJ5<4gtAP<-hjsnC-^J;tPj^MV&d-76
z0hy?MrZi7ZfE6d(vv+yK>QA44r_|JI`QuOvTcimd(}6rGN<?*KX_Su^UtlzNsM;57
z3R$y-84-yYv0xb<Uf$;>53-ci#~zfIJ9e5xRaVlMQiRtPo3RH-x0+j73BD(}b8z?%
zfrl?`&NUE{)~gH~1Cr)bQt(_|^&u(2zt|_Je1L~9+@nrL9~Jo!35*noL*aUZ{FH$z
zAta<LBWkdTgJ8R!JdFJ6x)>DHFRrCNv#@z1#O}Hia$xnGl|;x~!ro%#9h>K5(!j{5
zy>ilt8O7Bl0VpIyUe3FvO81#~qIdpdD9?ETUAj``B%Q$#60R;L4wh6DNLc<<vyjUH
zsYnPCDFajt7v318APf1dUL(EMh;s`C0apg-;a<L^M{2y6E^X|n{z$?b!=QM#-eN|M
z=X?Fy-f_;D=#Q9anIB*gF43$*0$W7gcFzwfQU$q_${ZZ#8k7+b)wCpP^zoBiRR;Y`
zwuX*IOaua8AlehKc&L9CcigembQ?bGT_dM>g4$fyD-5b^J6%z;moF}V?d(mtdLh&Y
zdt(-ku?X___q_9wU*EzK^5K{hEm>Q{_Yf>qNLpj#CHT5OJ+B57u(*iq^kC%%qT@;d
z^ut8Gar7D6R#}+?@`=~GD}hk{3weehA&~@<UX%4NhRGt#9s%dn;l|`(ZM03|mX)n(
z)W<01+nDoWbP4gKzCe*tl)T__-4G7hz^yEMAnf8+@*Jv0ASLGvx1NZIxA2L8B59NT
z1{RNd<Hp9`pOeoc-Nf*Q(UfKCXq<)SMxZoT@AotWhQC--h7o_Z-jO>};y5;?)?wzk
zu+AEUBTzS?L&$^u$@}2a$aQP4)ANJoJ7YlX>^nj)GaOhRI|c*h#!Xz3L5N9?!*6-~
z0o3DFW<hhLeE%gr4!w+>T?1ItWnGkfkDU@JEMm>7#YJkamM6k=W`oe6My|wQWBq|C
z9oKtB;loVLu$j8aY7&h*i2hfs8kd&6pV<5>MmvY@PD`_yxnk88&zQ?Zudg&HRc}uF
z?(fqhU$3_B=#>MdF12^H;021Us#g=*JgRnArXij96zB(AECg6_B@;=pDI@@wj_xy!
zjPRa?C&^grPOX7n1V*UaeVQto6>yFOJ&%)Iwg1n{868knA%<g(nT(gsda3QM-jt}#
zT4S`3)`|8yV?IjnEW)Vx6d>((+^xrnFZCO-zPhwH9^SeSk-dC0u=C6uIoWY9!R`Q3
zB7vmIK7&E%HF`=m^+;|t&R<&D=r51!XvLu<+D_vYCfYYZ?KbABbNBNhUmV|z-RSef
z$LI%hO;#-N3L&A&<KE3Lx0`(uNJ$x&0QBoyi~F#i>}$9jAF9z$!yha;UFp1#cjs|G
zet2DqRB{n`cLyR;ba5%gn|$_g%6MoZ1)O)^g4#j`-vmqikPe}al=-Pw2ZwDc2OOyL
zbTm`ct1{R3v}}?JlB5~qj?>3@mxjSbKrmXQ|L)f1WlOI95#_b!+_p={pFR1E0#Mg@
zALQ(jX81^tK~)YobWc!~AwG^Mv|;HNr?pu|NXRbaC}14ewT!Wfa$D^kUI6}qd@=la
zYeg+o!`iS+v%{vKuE1k5i5?4UpL$LjxlMTSbMa3tXXZPwpXfI&Vk6`KE=5bbl5v~=
zhBF>!z|db-z;Mo7>_i>ToxDYH16u%I!d}7kQIwx6TSE#iK>Xq{<N0G@=h1me2epbA
z9fRGRBj!bzTt0}NrfTuW(Tx6eWBQlcQz6a6KROn61@}wqA=;owSj^nu4GsK<nGi@(
z>0+mcqsBGAFJ?#V?KoTCe!uddaAMbOOi7Nuua^j~tpwF`ZZ+eGZnFUoq=4EfY+pYY
zP_mN~3|L{Pg&5djc}ASpEPXHksESZ>M9Q$&&w1?cqn$Rwj=DKrhvOa!qxD6gUa4}H
z(|snqU~qr$?cqL|-2o`|%+)UDqV#j|3t+ENnwjD*02Tt&aC6-8lTxlM9UeK(k<f!b
z;CSQCQTmROrd(x<ku2?Mhe$kk*=%fSkK9E^1m08BLd2Q~WyE_)D5Q1Y@fs_S1E3Nm
zLfn1b5N0~$*T@<*s1l@%I3GWJvERXmb>27cYp1w%bqy#&U(4(Z;tvKYH8w}z(7H$0
zr#-VoRXt&M)($0u8|r9qbt11}O!@M^xWg;w+riiMlCMNL6Gm0wy*qm?uSYR9vr#cu
z4(fQPgT||6MM)my_6<oSC@B11q|ALi$boSI(8l@b8$JSBpjw~qDT$_qhAEBI>3YF5
zyRFx0UvQzK_%_+mZ`cdbgf8aejX?wgv&~WC6(b$%HP2t68YrW&MDXINn>>IIrhkG_
zc|VnvPZ(sEoJ{n>{1Hl-_b@_;{3IvZ!YE6vcSMi}{M@y#x)#4P5nFS2)3w_JZ+{)p
zv66SPTwg{vI}X(l$uOGz=Y0Q(UF}_81$)lApj)9|1I=yZSI(F>&<MB37WS4&_{6Ux
zyy_!6#?o4!66MVPLd+TrVxQwUd`^yk0*-I>smF>kWy9UTky}kH^@?HMMiG0cXZX6X
z&;E}IPys=fjQ!DQOz=WG*wa8~TictPEZ|)VM*i%*jQdBn%v^VsQ<}b;)9!D3;-)^4
zYF7e5#8;Ecju(%S*X>JolcgnjHN%U)Hebi#<|w7y#v}4$_>;#77Xey5T?Q8=fZ+y(
zgU>?vx(J}FQ!Z4jY%FtSFDl~#6JrA6eN~pr<9z-#Li!$$uXxyA?6i2=)f#-Ms=&PE
z?zptsV!&<j7PRPk0*gk*FDL%E>3Tf^K@HgDScsLvNd#Hh#ZZI^;m!;a;yOE52F_CA
zHi!b-rj*q9DKQQg0fefPDIfjRV{GQ9)0%aJswtPDwF#sl0-N<AKE|Ki_$F2e0l~6%
z0;!V__Jsc4YdnZhP8IhQ0{ueelQEunCATvX{Rd{~gDGyO-*M>&lwj<ryGDWZi>!_&
zp9NE?y}Px#bv9mK-Z0btN7)DE+p?VU76u50$VLU4yu6l%OZ{VF=I&5oARzveO#BLa
zb|$&D)+ZD=woQ~CflKS>I*Ag%$BDe&a!R(Bo2b%cY`u4pbN@%Fjc$tqKJWLUPC%Au
zqnyIlqX;`ah;`)Y@b=o&oq{Fj=RK+_3cdDDV{kWhb5$5;kEl9>VuPAZ%0z5EQJd2n
zrd;jjs2eUdbf4C3ZQB!tdYfD+ao~cI!Uc}TqP8;+1JDuJ4#uqyW#dr)<}g_EEmc+H
zBG1&-Qw1FTx{I2;W)I*u0C`bQD~rs;>00%B(1V2BY70<f2Qntq>nxQq488%ZMoQ@)
zIFcOfU4@G>ix&5m;nwCkJrBes5ddI$g%;3u@ADp@(xxz6MZCUF1G`Qs>Kz<u-t@X&
zOc`DnczkESdR%68`!q_tTK{tb{T;}wyTX<e8MS%as)V$ZU)6uTLbmr?KLa4aovY|`
z*}n#8n;ZG;#%eF0v!!SQ)iZSskRJVM8*yB*X~4;;4ekMI=j~ecS4dV`g=cnsMXf(S
zWu{(NV6hBR-x-cF!b3T3L-(D-)n*Znv9ugrl)Q#W|3sqxk)`D)5F^$u`rL!HlSXBm
z3pHwTLYqSL%T1)igo`=b`E#tS!EFLkXytbL)65R4&%$-xXRR=2Ylw<CLy}aZ>ri{Q
zo|M$ec&FoCzgf~KH8IrY4-YTyMDm*sKe!?m8w|bfch9azK{1?sJUgTCV|)z$?P+SC
zc@L90xVE2igP&R4M)?If<&QCN1u9&H{Ul`0Kkf1ziXvb4Z`K_B3XTal*$dRKp62+q
zQv=&~`j(-@E6ry;^)RW6c>Z!<n?xgg`SK{yRUk<lsd2m-kZ)?%7p-sba5(A;hz8)o
zJ21SPXh7QQb4oD|O__5=rL93zjq?KOet2Gp|DzTv5qHiP(~L{>d!_yCxg5Hky)2qA
zJ?>$ofX>S2@*9k!{o~K#S)WV1`qcex20rEO;8x&bt|6%$uN<40@Iv`Z44l`uDOqFJ
z=KQ83{)gJV)WQs)DWbY4y~^ETpwVU|z#$nnFi^%tfD0wrV;B3Q_y-h_{p+f-s`f*t
zF_PErHKi@j6XO1mOY<X+?(FL5TA7^qx!7;pBP}GrdJSO^3||{<r9?o0+~jod#N2Hd
z0>XnNuOXijNICR)h>?FF{_={vMZ#lr5JRRBS_p+z*&c9_u#=g)<>pR(>-lkavB|TV
z3X|d@+yfL;NLoe)tG65@Ucfup@IqeRRCpf-e_-+hGiaDQ@zw(uI?{7R7jk~nKM$Z$
zct5}Zikg8Q=Q~t?`gFS1**m&t*nSO#TKIm+*RN$daZ2AxD61XglH&94M9O$=o+U_7
z8zypTvhry&=+!&qM@7xA5{81do#Yl;#VX^1MQ#K|;likuWkGb%k77L`sVD8GW=n8v
z2Fsir%4_gt(Vn^DP>3M)i;o{YRm8?`CI1*ifAen-oFmj_G=D(nxr!?0@f*=8exNhv
zU)5xN9Ohf&6#GpZ+FJF1KY3jM2XV2R`}AqQvp<a%c`GFu{9_)3d$-o-F#hG1S2=D)
zf{x!}DKa$x%&`cg{K+Bz-KU0($a|>vn*G@}`}41}A!d~&Apz{|AOMn_JsZkVa1wRo
zq!AkJ=%=P;yoCUVJ+7&vt(DNUi)*>(t$Mu&4ryTGgZ-QOxBvO~|HIQ;235U&f5RIT
zQ4~-@kQNa*q%@M!NOy;TD2;TNNJ~kJNK1D&(%nddba(f&{5|u(XMAynIl^Xp?d!YN
zC&?if6;w79SmE3A`*%G_m)ag39si4!2DxW?z&bgolm9?k9>Vcth>@cpDYv$kH#;9L
z+i^3Qt_;Qw(+_XvbP$DR`?Vm4X2C}bd?b3y#~Y!jz_j|J^dUD3?O5NRUt$$~$v3i0
z!)4{HVpFIf!%eq|jfjs)h{))8N2kdeDn&`J$$EO;JqAlJbX{B!ABXV^+S&E`vh~65
zW9FfpLYCI=rfZs+^OlxYmv2?0ec8m+80-FB7w*loFKH0-^yiD;W5MRsyU5AGQ7_S`
zncTNBFEJBS6hZWc?Dc8Dvpy4-?mNHfb@aa+kDO>mQDWF5C+iJ~eWdd9v#hLmRW&p)
z5E5Myp*yon`;N}l#KQVT<letO*wm{k+YJn!U_L_f^tyAAM2_Z!9dPv@f{c?DPJ9Sy
zkc@N!&pZeT_2%Wr>v0iA3|ziP-)I{=`ulpzDoUzsSH>+ZszjbVf7%=GQ?8HxY@Csr
zm4}C>tz|qD0C{X|&^D?lm<`^KM;!Ba#8ERg6jr8<3{4-kz9?X~>6jmZ2R(8l@F|bb
zx$>=b@CoteJ_sT_?KMV6UV&!VSBs7b)IwP~S!@}quM*aACGkL3Y+GD(a)_CqdBuPg
z0GG~^X2p<Gc=WF}n$_0K;`fq@p#Phint;{$jZsl#kJMKPpa$jwoTsGxE&)WZ?szvN
z8uhlQOX)2(-nW<F6^PckapZ&>m^>Mk`ep0>{bSGc$ldcCcPR-cIOuPUg-}P+ZKx1Q
zY*V(r^}8_J3GGKVlc#_<E;>3239+#QH$!SzCzpoCVta74TRYW`&;qEF2JrGSE7Ova
zI=Z^1+q;Q#^pA98A}FE5{l%SYAh!5j)?>b1fF*M$uMQ`o6B$AikPtc!q#r)8y|M)%
z9XmVo8)e9p8`kpPUGvh03MPV$?Dx!7Fq*@4UR-QbV^>7=_yMHH-mfrIGc-yn$jCXQ
zBqv}S9j|nJ8kCop$2MUF5D!Qk{>4P>RqQb?7(vnIN9jb@?(Wr7ZSi~PYdJZo!LC1U
zmdV2X|4nlHG*EZdkWldzpIt5d_nS>|>#Krhr_L+baZko>NX1rV%H5ogEA3OYgG%{>
z0G9_u`QTyTjix1)AJHuq^?x+1x)(QdA_Aa({_;hHOcn17l(NE03_v8!$>Y)U)6g&@
z5bOTnND3xqo93ssy`9Hi%NS1E55X|_#HppH_XP2-2j$|<_NbVROS5Pm_Bv{L0L&0~
zE>E8X2)iRa2XXW}W_8%wOwWfziv^<t)g?2=%`nhB$rH}sd!~F-$Yp(W#ElG8i@__L
z$%ykRmJD}ssRK;3V@$M{5ycAsY5t!UKpYgr-$Bm{wbAjrLG6p%+!GY;QtM`xe!u*I
z$y1<t8kD7khUF`lG@QX*W||MCjG67p;Dv9@Qf^vw#sfob^ZSUTjGSQ#B{>0j-OzCr
zlE8}19Sge>4w}3>Og3`ta=hhL8X@mZi_F#iQG`73o<0>~i8q#C+&%RY{{5KaE@DU@
zEt6iWX8ynHXVU#0$evxhYYc8iN<o$psq0Cp(#diX65Zsf2AN7|3VaN7bfQOrW*txe
zvX~#e6tLk5&(tnnGrGk=3}lWRoBjxY5*?!olQl>TA4z>B4Vve)JUN0qO^{_5TzOOD
zbouLfL+%+Rz0wHJ1Bz`3$`53>1vj~}GOop2V~8+!7vp8EQISjH=DPaA!-4?gu<gL0
z=#IfS153*zo#RSv4UG^H5+lXR$Hq%~CO93oPc*hF!>Au@R02#BruDG+S?)HfZ|%O*
zQSL9_fb@`ZjO5Z+eVJA1#<yv|9d`eu%NFeH&Qif%br)flN1wqhZASl2-9=7qL_s&$
zaPeb`8`;YH4V?CYz0C2-GdRY`B>aUQ09a*^mzT1Qydq3VB^x#&w-{28K)+qw=R|-@
z$zR`y3K%6hYGLDKJ|corlnK}t7UfZWCSE7ghrA^(iG@7uS0#RJGFYQG>9i!V4>4I!
zAin2=jMP`cVUx9|lWSpvA7|tmlBtwT*4kPJ)~@}w@q{`$ItW;-W1C~2zy#8n$X75`
z{Lrr%)+BSlIhOte?o1XRqCUF;df^!n`odt?oVtc{n_mkBLxm3AQ`w(uh+t)@R0XEf
zXY1`cm0LP$O60h*Ldd}ej}Qa9nW+3BOaznesY@zT(I;a2r@mhSqgP?j#3wm_n#VaY
zH6aD7kKOL`6)Q2y2sqCAseyA|*@1!fEU0<6<X{X6A?3ITyv$q4_~gkluuGkn;a><w
z1Hj|pHs@#%(Uv~`sQhR7Y^-TeuV<Cf_*Nvu_qW`@@bw+$XWcP$dzfggO>fAcSAhW>
zX?=rWl+Z!lp7ZF?6;*d?jVZO|3ZAU_=wL?f72VK5frWTP%=4~%gq%b^@Jo|QXBsct
zC(h}g#PQ&jf)b8nQ2Lx<l4|#BH7)LAw*^yv6GwQq@(M3n8YfRbB^{^CrRyD3tF_!E
zIgHLdsoV}m?=AXGEulni;VAGdb@l10b!>H7A=9CL(S1aRd6e%Xu*L8k?n6D@Td9l4
zJ*EkMYF5xSxb>U%`k_gCHqSc)Wp~Dut{R#j>%#Bl>1SV}fe$KShb>+y3M_Ezn3x;y
zIRBA$_QrIE^lkeHTuUWGmFSgtkFU6&>nfGxOKAy><||rzUbtYR-3@MK@D=U8T6{BM
zgcVaV9~B^sh=rBRdR5?!Y3-U}(Ya?D@u4aYiRJrDP%r+B*L`JmHQxC575^<+{hNVb
zCbClndy@1<zo+Bu*`pubYwl_yvZ59YH`>g<19U*2<O(JsPa{8d&+rs#&FtjR)r3$#
z*bhTH{{uVS^_4d^|F7v9H)&sI9P$F6rd$7GaSo42O(40Xv@7cx=l@wO7NR@~NVJQd
zLwVNDtWbNFQrDn&4cPE7s(v&Hf1jk|5-`9>zeFP6Zx4~AxQ77N827+~vQr_Q1m|ES
z@RA;+iee|bBil(V9|zIsE&?CL3q+rgHsL&T)hyN$MxH;~7t(Do6s~iNxi~j%YG1k*
z(^|LX!s@v|3G=7$dPDvAW?YrygAn`tXxf;_?-{VXlQ_Gy)CS###U6o%+fk|K@=GY6
z&d!i^AEdOtATNp)Cll9@eYlLhIm^VifcHzDT7G@S`=3POp^qSWxPQmR$s2asaeb<Q
z@6CV^Ux<g}4cY1iy<#1JY^bwYe1GqLJis<brbjAe3rN=Pd+pKAG<j5Ce@HLV>$TSW
z&s|zs83hrEgImYL;3h5=0{ZovqjOll06GEpiSIpP5YHy|&(+wJ2#E1+F-XVBw72>H
zm!AnLHU?COJB@ATF0+6EguQ4)^8Ef?BBI}L`7}GP2E$Xh_+$HujQt<H0!Lw1Y<lxd
z{jBmp%b^pGK<_K_D_&j#5s@}ecek)c5=2idmodalZz$aOq4=~y;ht%D)R}s%<FdzW
zp1DX-!#!?7(^M6r|IU$r|JuH+|CKID3i5bZ^vvl)n!VfmZ=V<Vc|6eA{Leib^_;ZH
z<FszJt$xnotYw$~BbWp8jMA=e)lIEC2HnHNg?VRgc|#}}_^kSoV!#u>cjlUgAx;?%
zq;xWqk&L@Qa=K4ka9MhE^apvXx6PcD$MYXyPk8eQ9|KkLY0qo7tH*&&d*}a<x$+Vu
zgcRMaP?H#9ZK(6<WKq`)$@}-q2v0RSBq9uY)70Z+;Fy*s$#B@DUhi?1ind$cJYM1%
zTEg*8_B?sxc5{J__#!ypo+gVJn3$EZH!i&wAdH2V`1pWQA3TvD(W%;X>TQ}XtxW-?
z#8m1F9*kA+_jq%U!Cl`nPk5nYAOxCn7=i|8FM&@P#QxQ<@s}>Xth1vBc+hHMF1Tyn
zC`;q6OZ&I-%h`STaETnI2YxkzHHXroa#l8`E=IbA!*ehJj%5$_{UJaoG^9aA5R#|v
zo4xV3ETPpuQ9S$X%hihRH2Qs^K&+X~l@}i{BF^Ppr7xlGjYL+%7Q4<`-3j)MbXYl}
zB<O)+Gp#cE>ql`LE2ScTN%0P1iLDP0A|ezoPq#;FR59nq1O_TmCD`A`?l3lY4dntS
z4-QX!yhi<EIkPNKx59l46`pSviPXr$TiV<lv7hjow1jQPj{ywM!NgAYAAOD<``Y?~
zrinoL9~DQQ`qErvdUdE)hJ?fjabX<(hoQ{^lF@(cjsNb`CgAZ^+PLjC-l%B;5rG22
zabK8y<dxo&9S^X(;457M75<q1+X?+0?$N3UbOiifz9a|Fi+<m?;{5ToJtf$hx7u!7
z-?Cl5=!t-LzSyDc2zH<_d8FRmOBr@D!#(N&lX!&%2x<ZR2sp{UpjB<o88WwnbJI!2
z_`6;Ax{`&-jUAWU%#u=8I3n6baR@i;BEIoICwxOf4vTnw0qCCBtAJ7gARfdQ+N$jk
z8Wb6h3JEPVQOF4026t117l=2>R~dd{Q#?gG#K9H&{KhhQ%IVpG%->u>oi8yEDU<Xn
zpK~>+vC7uP{gIAYmV>V!#YusR-pEE#4TyN8zf}ddr`a@!%FV;^Yu~YB&Dn<9IcZTv
zf!DeD`2w#aJ=1Dau+C>(r6oPq@!nZ}tg~NM08E<dE8$BVG?wDxq8iwFtAqMiRox4|
z5myXOPb-E021%dG(NQ|+bnf4k%4#fIW0|=^DJixQArJTUy#NY<m2HHj=($%9N#nR2
zDFSFICrD>>mG1Uz@Nd{~)>hGB2p0#jG9J#;2@Oqs%?_9=v_jb{Uqd-E`ti+u12J^n
z>Q5Nq!f-x;ooQc?fKA0rr$Itp{efRI1i%48?@w^m<irB66CD9b#~lTj^W*cpyNmoM
z8-rzF5#=G7Ydt2?ctZa8K5$EfBMp|ycEOj@1x_Wu5*%z|q;oVi)pK>ra#Mfi3zI)^
z!ggBUKxZ8Q>}Wo}KLRv0;BQwMdv!`=D?r>*t=!!fN-RS0ByqX{1%V`xEH3#&Fd3dU
zxV(p%o9a@}=UT@1;PQXu+cX-KF%Xf_tNxkqQO*y%<e5U4cSx;Y`xw3r|2<yoZWy|w
zjr7-avH~`x=Xh2;Bn^%?X6^f_t@;llI;)}ciSoxR0gQ8FxmqS4Oy~f4xJE9}Y6#|Z
z0ZD*4;TR-p&3~#4;|O1!nI7C1a;eBjbKjos{3UkE078S~o|jUUHEdD|JW$}iPC*`^
zsOd7=jP3IU^ihBlIi64UzI?a}z1Z%O{o@`jxNe0reF^s;1Q9Wrg~bI}*-7_?FMb>_
z^YUzm{yoQ$z2E<$<E>_UiHY0iKf&>qCE~v)+o-CdlEAg<d+!&b&~gkCFJxhz(&Wek
z$=MFbla+rxP6aXnKC&9I@e5XGXDw-qRQ;*r{j|{xk0mK=3Cu_Ko;xDnfXoOWYIuf`
z#Poyn;-MQBA_;It^Y}jinQujQd#OKE6%U0wHz6ynqtdYz8dOTlre|abpAkT;`p&e&
zU8jVww`u66F)s`+r>{QS#N;Pq_z08F*0`A;hPWwGadUBfZCl*mWZxBJ<Sg_c1@JvG
zDoU7qA6!m_*F$-Fz&eW0d9{9EwAS$>o-6#|gdUNVp7q*R&ZE(5YBauh`87PC^R+m$
zQuFM(mJzyS2+x<af5u)B+S6^UZ#Zhb{%AB%R9b4*KHJ>E7$cZ$rd8t#&oQ!aV89dV
zY?aHY<bfqw?fC>7lXkh|&%i+I_ewp<1{w0Hg|9fY8zv_P-jGs!EA(FrQq6yOw<!lt
zX7;GsZ*U`%0SL45_pSGyW&I;z9B1i}!#h?FEuCkCTug(5AITDPzp^{g9|?d=<H+v3
z)5`6-cRNxjT;^+kkw6%FN~lv)ZVr7xn&l=2VDJDm7e23sA53a*3pah)x}TvP8Rl?A
zMMo)SMMCo-Ch;^WYGv(bWRQl&)?O_i=2z72#KnCFum!X@iJWe~PwfUkt%epDC?EL@
zs$-JoYMtFdq<TxcF=PD>y!2ce1P!-88ot6J?lfJksHDa0^4pqXm&18Uu9STg%w;T!
zi#@qNwp}o$G&Jws^@A$5t*NTkp(d%ePZ_1TS8D+!CMMZncK~QSe$VKixt0(i^$Gv_
zDyl*E%`1}72yqHnS3zwVfZg_c*NPSH+ud1@$7mwdKYGFH>VH=dZn|r4K{s7g=C~g4
z#DLnsN2FCyuUT!PYltYc9q8@jfudmuQPwZQ@61K^430uUv`|qH+wtbS$-usUm*UlH
z@YMrNNXqX9;PGTksU-+Oz)UVj_Eooo2gA?hs)u7m+tpU;M?$c&)mwE73#g!VkJpO@
z2Ymam(FpuRveugj@ic(8=8kF|I+jP^8YEBA=PWA&gSmt$_LsVVz?h9!y!k(b4u@*5
zkbxhC&?V)_C~)wj`qUP^Vks&wHtIzOw5L!v@h2F3IV`588gvUPn06b|+%#+7*~j!H
zxaX0WLx08i)XVfVi;1F!h=ksVRS=BSf^!V1@vp!&aM_WO->Odste_w&#@!2YAJ5+c
z8kwT5c#xrl0D{Jw9c62_o@q(?{q*Hs#M!rK3<RvSihOR*zxhK?2l{<9>9bg5gwes?
z(efrZZ@3c5t+oT@U-<H!9(yd4er=vVKnol&$rdIT{dKvZ{N`N;Ayz<I{;!LkQ?|r@
zF?kKqP@<<KxQBlSZ7?!+HY;o)0xGWnFKBA~{d=?b_>UoUDs6mXHl*|Y^SXO2gIb4s
z!i(fyZxD!82JKWi>OU$gK<1z#2u&X%98dXvFE7333Kk^~50g4h^!Zl&EwEStJ;JVI
z8OKNPKiV9GClCt|zyi7__q{W92v323qTnbItGC9GhYKa%-?JqQ%r$u3S6up6AH?@Q
zLGytOfC$R3iYhABw)of6V;+m;yf7A^KMcK#$asrQ&?JAmLH-ovrSD%2%-(I$`R!3z
zel)zw=<!x^Y-j{HXE=)giyGNmf5?w`W4qLz7MV)bhum1)3hDG)zjD)*<kV^nSCJq%
zfpsL(zZDWK9ArTq0(mtFmmzC*15WFB0HJ-5bwG|gqMPi(!MX4id@d&33T;3z{LH%T
zz2Co@07T^o@*fB<Z{Xm<ypN)dD@lR)`|88eFrGW;F)7HO$TnWqVWbQjM-UUgp#%_1
zTdRM4-5%%{1qv*bdkw%i!~A0Wk*QC;6w3yMfZw9@Z~-2FvQD@2i=&$rK|zv%5)uvR
zA7LOQ8ZY=A&@<UY<k$St<H8lXMU~D7|Iq!jv2HIeVAr~lEaTy<EG>iL4a7a7hyG!O
z{@{u8^A$VZ9$j=%wuAiQ@7dXA3g6z4eZ}WGi}Dz_Nu~0`*gEA)PTmz%+yN0K=>324
zte#LFnE!L7y_<0{HBA8#7hJUnPa+iPt&-mcWAsGpY<^jFnnD)KFX?Kl%N@`N)b8>(
z-H%~UuUArv<X$t1iHtU%DC!>>&&;m=(Cs^lme(RJ7R#!ao9I!vyGcfbk&&BwwAE@_
zq~?0O@LyO!Rs_}EyMBhNOT9l<bMx||BI3o8<=C&+md1%d+91Y)q2&D2fiyEab8XFt
zUj3DrC<kA!@j`q@g##%MR22yo8{sf7uaq4k9%JKq3?KeX^ymOe2;`5pTQ1<~oHk$S
zv6T>)s<v+~$*mP6dtB>?3r?FYyJl(86aTwS;Jv{l<#ubm{?n~`L&xsdy8^96N&d!p
zxIB|t8Y(*T)(2WvR{sU=k~}3bG%z>TAIUD7P}5vr-@w6*v6!xIAMf~UZbrmfU3iW1
z_-k-59-Z32Pszl~`uyzt^W{j^kx^fC68r<oq3IrB5%sK)Ze>-)%h;=|!ptwN@6S<u
zzxasJf9Jp4XVIm@fF=G<rva);{+#o+uPO+8VMcDFqDX#QTr(9%yHgR8SpOHCte>0L
zCMPvYQ;z}s0u6(HQOjf(5j`InL^<jWc=ACDLWARkkdU_d{*}7XJ?`InrQp|sj)tn+
zz4ejKkUAGTOtc4#|IwRn)34cnZ(lA!640AEB{?~Ms8#(bA57ZcpEGK|YX1W@VLydH
z_-nUq#^Ku0sCol|`v>bmpOF(li8fxn3ZpBqvzaA8-rfi#_q$zxl!+(8CJfE8DN)<X
zjIA_QBvLJy+u1%mt193hdVuN)-u;Vu3^+XZh`H~v2U^}~tk>2Y8BxOHDyReUEjVh~
zHX{J$Zlrf_=4OWzh+=Zb*VeZq5+Z6$&QpLQWhRI%{`T#1nFW3pfFN6|q_v-Rq!h7V
zQ25(H1`7ab5Vv*D&)0IQ6fQA3S}HsQB{%F@%zsB-S3n5GBb18+!-sHQqxJ3qFAFPj
z!A$LDZ9I2q`&7AAZ|X-z6y)at!dg{kPkaO?O3e!@`WM5S{qUb{yx7kJcopC(vpJiN
z>tCv3br^6-fsqS=cnStb5JQP{kIfxahqf<Pm}?ylF@oLe5y61Ndedu=uh&{%mqW2~
ziW3W@KzJ$azBlDHdZSBA8)T5bny=N604vV=RGBa6sYXVvRC8x8Kyd~-mm#ClmNXm$
zK2PCo%f0~#kNXasMJU|+6P)9A?G0=I)bQT8xx5fyL#Wl;3h2pOz1fQgq#7P;Y4QD6
z6T`(ehy3a8alFl)9Rq)FiyTJv<jr+mq?>vMqz2GGd&W|ysEyCi+v^!PXloYC=vSnS
z&%}y5a6~+MG<<$tw9koWx1_UMr{h7oyST`*Wz*MJ2AvYyWwUG!gMq2-{iSGTv%l;l
zg>(39F4Ee9ubL=(Ih{_(XD0tuN{(4f51N!|<dy60?&N8f|9wcqZZ>0g(uB+Im!puS
z{e##6^2ppyy`-n?8yt^6v@ZC615f%Oj4=`xGj4JkZ9CINGgU#VMMr(aQ^y;#FAqkl
zp!``peF*MIo#1-IzeX~YF>`Kvjw6Kv&?Xqn<Fs?pTNC?xb?u1|(bo0~o!V2U11<6R
zq{bEL$LrcHTgJ0GKlLJ-;1HUbc7Nw_?>Q!{Ko|cNA9Oe%;70o;>hNiv<I&PY&w4~9
zkGqToAKJ{tRqN)tq>6Tf9f#S79M=pN17DWGWjUIU0>8g#O*zDKtCf+Htu@c~*)L8m
zD}JQs#B$<V>Al?hZ*i_RC`!Um>Yf|Cy6}v?<f&>~lhEeH+4@j)i_sLDj1vP&Ux&@1
z!OlUw$t@J@CQ<NDe;+WyZV%R|G^3Bw=W*@uEDT@)2Tox7qT&$qJ<PQ!_R8hT&XN#l
z+hQ(sx-9>#<~Dh^!fd!YJuYh<=Jk*SPE|D}I#p_0jhV;*0|qgX)Kh_iorF3b-VmZ&
z8T?3Rd|ybJ>R?pEJx=~(gr&%Q`*aHZk*MfW<FA`rcLP`ui(cLJ>WzOl&$4+RD*%UR
zq*##peaUM|N=hg-d6>wIWWEIo^BQ&mjXd_(Z1iBU_ifc`a5m0q{D6>?mseNNSXo{V
zSx`AZ!Cf{B39V2qp+qWYr_)Wm2hxxts|V^6RnQ6pRkS<PS-ij{f4_^u9wSac%10R)
zUF5C7ri;Ub?e?o=Vmc5lxuF0v+e7_X4Rqu+&W*j&6A+Lu63mCmoP#{#zqYoYBo;#V
zP6)Z3@kv4*9BiT_(#<eb^9ErmR?Sjc*gp&VOJ-_Hi%3%V51ohiE^ZS9sxgFGni~du
zZcUMZ?oK78`;gRdhKv*;fN27LCZ1{S8Lc1PT@$2>*rO(Puo<lM{uE_mzvl^BL9Fm`
z_DbBL!ZgG81cc8()NJ1Pl@u{q_TZbiA&63-K#_-!KTUcQEUubr_{@EH%BriYc;yzj
zMwY5VcWE;-sTo&G&9PgOG&vsZRg{*3?NeGxf{lZNkk@q}bKZN9A*D1|iz54%re@st
z@9%B}xqZhc$H!mAIRWDKGF@AUyC<Fv-OfKKD8es3M@~&$;j`#DW-byZ5d%dfZn$_)
zSEsau4BIb!#*o+m<b83&;+|Y0Dk{H_?I~oyyE;32ZLF-aTa6SS!rw@fevDvvLCeki
zqI;^_;O58P)86k0&nI!y<#v)$61a(BKUC$$JNl{gPi$LWC;a$ELKs7arL0Je^EAS%
zeNiOEBs{ZPHIxBDW#JFGJX^4YA=5KX#QTncI@1hHQtWREL&7O0ir54}oe-W#E?*{?
zyoEJ6>)ZOC3I80%{+EVz)ea|z$Kjptr3pE#?W~|e=7GtCAVW-h+smaDybHFY*<$yX
z3+7hVwn>AT@D3O*F9VZaSWz-)#Og)=UC-c1*{E!CVPObv1~y9qD4oNKP~@G~cCp(<
zMpW$b2`{$3A5fR_6{f)r$c7xZCner-aar*}C2!OMI;jqfI(qt^|K0yB;Kj}kt&ja3
zlo8_3%|Acl#z{*{AI=>4n-pE9U-NhWO!TJynfe5r<e-wpN!%c18r=Bi7ZIKC2`R;V
z@HFTp_~7>+bn@^uxQXw<ZQI*p!WpZr{o`x!SIFr*e2A7KJUXjG0=EugfVuEll>U!+
zT?j-nv9RECK1(W;Zw_;npcwiw&4ahP+Y!Kbc7?|7H}yO%c$sQO&gpdq9Q;xPZYnWU
z8yhGp9QtQ7s;lY$!Old*k=VWH<W#-*sS;|#0)$8JB73#I$0sCw5obJH{c>iqI$XE;
z>T}A%5c9{6?N5a60oy1mTd^Wd2ny-)-+3b@(hAll_7*=K?)-yWQY`QSvf;BZJ*_o)
zx9l!L&rsRQDl5ZJGpBI{K4tnvUwY@QIH#fK>}Y3@?Vsslxy%B0ZDV0+Vd00aoEiiV
z{<7;X|1Cfso*C{TWXULtjjgd6O`pcBZl)y&gMN48N)Y)6=M5jrOByZh9|;LWAh??K
z_}br1`>C)Z;k9`RQZeS;GL3}T$R>T}o?fXU!{0St8M;E2+?s?9zI;6Oe+^ty#Fb{s
zZ2UwHKj`OR2a6sb*op;Fg!WgI7o%aI-SzaH)p2!4(UtRRo>!sQZp6D!-PT4tRl&x_
zBoNiv*%=xvNlkCSVM$Q<&Z075+rf434#7DZEiEMIrzuRN_^ZE6U$!;PVMrzxr$}ii
zkd=R%Rao3L)Z}`6;FGl;_sEo%)f-2`&_u)6yUsj@Uo`KjWZIbJ^pk*WI21vhg@v!g
zQs)NRy9a;Y2x*Z=Sj^<vP%+1)jjr>xy)2CV`g4YBao^EHmsIlMLxaE~**gX2HqC<-
zM`uhg9&H-IKM&lKuI?gM#_F&AgvcIq**yXc<+541oZ9;C*y9Pz=;$I5a``te_(?4p
zZCFZFym}4)_vX3_z!w|n8}h7e3t6Tmg~^NDrv3<%d-cvsq$_wd_y(=^3x0)<&sJAg
zwx~P$HG%V;H*eMB;H|6-Pn5mWzsjiladin!Q1sIkI!d7|CHPPHCgz`VQWmhXlqO5x
zzF@?3Ku}ZDsdV7de`mM1vC|vKi)VX~i50--Dg_wELz4;pLrdDVKYtJd6pe`z>Cj{E
zaTY75r7$^=@%!ra{7yat;@wBn#oaMEvd3-iUq9MZCQMe$RN9|A80${YXw~w}IJtVk
zo82^R){*2+izk-EQ7KH`-PLDjZ9PI&f{MV#{vP4&X5lViW$Xk`bCD-3L6TjZTJ_#_
z34ZaWuKLC3K>=Tc(-o**W=u{_+U^boydh(KV59^GQ$fYT*qF)iD)!qDGaeolYt}8!
zY<t`DV+Mrn`pB$&f5m3J^-!{qAC>G+*Hb>k{dm?p>!UXW1ddJsJ-(k!!R2NzpaqM*
z)5*}*dIW~p&!(>!Qqs~pBDzF8*2u+3NC+x1zv1>1dbR3{=Jqcdx;>Z{Iw3qx>@*kU
z3h5Wcxn9@K5a*z0?RYo3c=uHLKjUYpAwLiZM%mLvKl_T!A=NyU-D@{RYY8rE;$l6f
zf}i|+ZYP3w>{;PHdfM|~m)8jAqQun55B%xDarhXBF~*-?dF*!Dwl5I_$=dPi!&Jwl
z8mU*`SuG;{YNk?_O&h+i=H-0!m?~CGjF_J8+3Gt!yyF-8lEnr2|Fi&?G<6W?AOwGf
z)nVTHD5I!o@R~5@SC9W20VZjq(n}6TbbR*c=4HE`5eC<TGp8Wcjj3gVH>V5<l#n9t
zxO8`9O6p#M6y=NMCm0~Nq`WS7HWpRE2+XK7IIM;+?Irb<g)PUEv)_Gn_w+)_bWyzC
z3>dM3CP_4PCWMQ_<m%|^vmi~>!QP)J|5n;?o-g)h93Fi~7H+=CjaAsFq`YoVoDPT(
z<CBH?KZ1xz*sC%tkER2>KQijnxm8du7>xK}YkJK2RX@QMpp4+1)(YL48%);pNYlsX
zaCRKeOm1%eg)i4Unomzsa0dkN!<4~g<;7Y>_8?+Ucnt}%?E3pOw#`9e98^*X3E@%A
z`pgUouetQMB8QmEvjW~8MGa$T^mcYL^0=AZc9!7H)!jN?jMtlT?EK(y*}rIwh<Dg>
zI%GIe8L8q$4IvNK9P1+nxi7m}&wE5AuAK!WUK@G%v@h<?)v|uPVSc~p>^NtS70&9u
zVRrrY3A>3}Nn7xTUfsZ(A6!g5e|*bKuA_I^3zgrM;7XD^pOAW`(`j@1NhCuw!xLyl
z-|&u)5&N2OG4nT)N!B7Xk*g*;pdG#jH5J!O6OxFvnbhtuFE=sS`|XSS4*7|?RrZIM
z<D4U2p}V5eXo(-m@ZZaIKP4Pv*c>q+E2f(&+pehmvIN~=FuIpJp9@le|BjEsvFl6W
zyk5jdsdz|Py_i~=tFTa$rodVS{ptG~vp6jfuW@LT&1faOz})%X@cy{{O!zuPq+29j
zrcQ3*K;*Cms}mH`zohQl*cNKme^|84iTSX)0*dE(sU@##zZVZ2n`$S9FEG&wnQY%h
zAR164$LO?Q`L(aU*nBO;EHrtyH<l}TE{rKN5;thI%xuuCBQxMaLWqaNV}4}bfU0lS
z;|K>qj=2MP!z>i9Zc<ktpV83K^;n7}uzd2?O6IhiZ<}+GS=ca}D>H~E?O*L*SV_2y
zFW(kzX!vR+7wSBcN^+bj#(TpIp=d#Y64%$RX69-jP=omNU&DE+j0qPL$Jy;37*If1
z_vxJAqRy^4`%fNn!(|8NITzcJMd7E}M@N?%$MF~Tw7MuHesCu&?xU2mvj@EWFV=gl
zLJmUROy}mKBv-Kgzl2%^tTKw|xyN>onf)N<Gd^f3T#78+<k;nNyXpF*LXf9&)Txbb
zei+6<7A5$b$Z&!vUm|@7|BAIjkBOeHXIH0SpO5VEgV-^xJn|d1Z%<5VY01MYO{eZ%
zOZ%jH^ZIqW9Bf6!_+8G|ac8$KaO~>$>r|S>BF*;o`L^S|BUEAt4b|S5tZH7`Hvq+E
z4ArO7itekHmry<4PH<qIra-Mo>3h8Pv};v0$=~zhx@G=PBsGqT!m(hzRnbwG#2F5H
zUz;!6$F|64ij+x`tOTQ-gEzf3n158F0h>a@FbViizp!Xv!}gi_Ydh6}QlmeP_GdVz
zwB5|?@ozH>y14UHXt73P1IfGuFv5r!c3X2eV0qyGbmkZ{aI(H0C(a$ME4OB3DP{&R
zX4puUP)B;s;G=uQLCWW1MZfO^n!eMkw_bPcrMi?dc=19^>~e$v^&P~u+tKHxqX#`3
z=)jARVvbYV<IU@;NE$T}8TB9<-j!CqGhmwVv@Aq4{rmS=xV-Ng2$6y10^{VKEvrVJ
z3Qd&WsVOS~s|Rdsl&s%xT!r-`bmhitT<}b3C7(tlE9BtYntr760~CZH^P`UYJTykA
zFzSjoE*dOi^=SuYw_+WOM6t$Vz7BlmwKdD85v)WbVkgQlp+<p2C3XqWz&X2_oJ!j-
zIqD0^FDe->c!a9m*^)V<I@Qjpx``yGq`b#RpKJ;%Y5;=KCGmH@qJzAOcCMDnJTcx4
z`Tw2JhQ}nlPF6PxkwGN}4YM<rb8dGL@c6i$h{|riz3(nArl_S+ecAZrs3sR<A7k*R
zHMD?6?zU1kKE`#sgw2;^C*S<D#?Dlg)5+&$IGBh%K6Z<xO5+pza*$*uuowiQ2c2u5
z1Q%%J=dhV9P{o*mmHI+Q7*VCAqtf?`%UORW=q%%TOgGXyg|lhK)KzF=ck?616#Hqi
zwm4H|P=hUYf<erdB}jc`uYxu-=KuW%5wQwgNNmZwXDQTEMdz6J1Uv#)oY#`v^N{1w
ziTGqFvPCKRFI1T{IFXSGhDySkJY=XEso0sTr0VTf>lsIF;BrnK2C9-2ibe??orFoI
zkE^mZ@)YD%xp--!iuJY+;NkTGZy9FKfjraPVLPsw3Jte?lcQh%1MNm@#KxdbHl+o!
z4DWgG%<niKU*zp&EfS)h1+-%E;;L=uj?hGDq^j5I6SpZG`C*KiFr9ML6-6|-p9&;;
zuQLi|T)K#J71d`)rO45gX_&VxL|xR2&-JgG5;H_}JGml#DD-{K-{Ai{3AMuFd}Ri;
zOcX*%$bwpXdM`uRCZbF^Qd2BloGVdKQAozi4BMt#W3|cKZb5<jiiSDY_=0VCozI0~
zEPH<l#YUogV&?MEs8Jb=KMA;e=ZA_K34+JCowZ<1g7P4kiZ740oG)jE;<RoYf+cxx
zc>n%YrAW>Cf9K=gE2k@b8<ydChaFOWJFm{9{#bUa@g#Tt8inIC(%+b^cG>w)+gquU
zJ#>$|&S{vLOBS&sHmx&qf^6U!02@0kHwR!Sm0u{pO^JY!b$5^r=bv{Qeh0C+>Z!)z
z>jJjtEVQUL-CLt|EUt|Ar3DoN6X+i<79X4(7-8+XMI+_OoMz|;rg}SIWyuh*3iuSJ
zmDcuB)iV$TPnNiRv=yGCXlzJ!xC6DRv602*t9(9MPz=EIoM~csnD`Kxn3P@r_w*-3
zVToZD8$nC@M{~n}^X;u_-WQ;9hT(*mi(H1iJ9}}?en5fh*@kp;aiiYLmkvvFjCeW`
z170wSC^iC7=VA+a7>IY4mV}9NLIj~iu|<<N|Ji*@_jw&9C@?+L)+P!EP|p|h-q$n*
zh2idr9}f(~G}K}NX1e!s1s7*>vJfUb3d)SOS|>0t;kB$B(i@ddMn<xJ^$vbRcC<bk
zEJf*I(yAI38#y#ShZ)q2H(nc16W^^%nA*Tz&Xo`%oEy<X6V-Lplgpmi6!}fO1**eG
z<tI&_KWFk0sId6E>~(Ww>mXzS)>08zE0Usnb?e+uFM2Ok=;I-gYy}U0Xt1)13$@HF
zf7FDf#F||8r=*k&+z_Zx)A#_Pq7sdGldj-O=uC+5N-yaY-W&sdO0wo0dvncA*%ZtF
zTGN;(px+Pmc$avxNL%SBPeV0B=FvmK*GG&;n!GpCYqsx<^-fAm|No2Se~&(yu+B7a
zy=igT{F@1>LbjgqRo>p3+@Ey5x;#@|<Qi?u(C!=w&;zW8x^OPdD)ps;vAvAWQC<zA
z&~<+rI&Og$tghm!sp5Pw=A5qZ`omIKv{x^*S*`z3;jM9)C^V!(aE?zsm+)}lz~k1O
zI=5*;BD?rAv-P9avNw)pwLq9;vdm@bUXieZk`{SPH+p|QY^a%QQ}KWDD%}tB10Li#
zoH9V>uwcw1p4jivOeW)hMV^#^8yYaHmF*g&cYB_E{*wt3sTf;hw4Bs(WSN}q7d#M^
z^{w`*Nm&w)f+9UO?)xmb$C&<Wo#{o{uW)g38pz9&+hh=%fy}kps?&k*xDNpj_|nwH
z$6EtEFwdi(!UxKKVi;gx$j~1xeT#q<@^z3~-Ou$nOj6#ddgJYcOH6aMp0K}3={fFq
zJae~d1JUL;3?BltiX6}3cCiH+MP8P(tpMNMOq8twZv)|n_^Le9O-))E!v~I6zppQe
zUb_=GIhCSY7z(Bu+f52wyKJK7TYoBy=e99<YdldXaGkGJ6W}U^LCkGuv|H|T_+bSP
zryTRapLJW{IFLM{NBT=J0Zl~bv{nL43xdh_3E7U8;)0l~u4L8yJ~>?hdkb&XY~AfR
zbD?@q{HZoq872nWSEs|-J#UbF7pT|ob?gtvT7rN|Z@BFSXq90j)&iY|e<!9U#LER!
zWM{S>DYQK%I?7>t+N&LVy4H<U`(nfW#;|7!tif0HPBWn7A|l?JEVn4L9ldwwJbu7P
zkt!Gk@iampT`run&hc1-|LIVopuY7+ihtZLUkkK|!|Pr<q;(76y$szMFAk9+k1Bc1
zTkRVi89h^J*WY+OhsDC~=W?Sk;yWstYB8Ox@YUsXD=_vvo%vZJ;~so~d|v^0$Eo#G
z?*$G>G4!ok>ilE_CA$EtgGZ0tmtrW40K<SCEna)A6Y1Z%u;%5_yF%xO1@m}JmTi8)
zaK~9LjlF`N2B8{uA=Kj<$7L@35fU93k{8xr03PZ8RGTzU6D2S^hsh?-`HJIrVR`PY
zQO^Ina^aIhyuaIOr;w!t0OrM$-6|d(867Vrn$$rfT->Gu>%PO70>BU|wRB*0k8g%j
zrfR4s_m<&cONQT7(_tqkHy0TZCE>Ps;`V|W$ZkQp@EvZPxHzb4x=ZdQOkrJB`?Y85
zlysp|izKF-mW~Fpz+LOfXu0ut^Hs7A4<6ba9N2Ls2$R>jFUlCdzJ&7?`m;XvmHo(e
z4d#)PT7ApZX%Rq>C*|x4S++0{#5O<{Pwzj8SsH>YhSu8VkV!D?+t5Miet_uvVmY<*
z=j<dXco_=Rqoeu^F?fd;+B)t?@RL_R1xlJ+MeXoABO}nN$Z0tkU$SXuDPcXhi%inL
zd@wY`0+W#3CK_7MduAs%v~V2MrS64`*FJS4JYPY)5s=^Sj;(-M1K{M+QeaXs3^cqg
z=S6zrr)ymaDI*v&$E$Wf{S%LOEFHu!SxQSuK|=0LbGYT4hZ2qQ=Rh=u(0)9n_aBz;
z%eQBbLRia0mD9a7_J@IiEX+1A>?<k^Gm6c);t|SHU>%^Lu@Usb7EAuL03<4`G{;Np
z4~wkW^qNplHtlW{lte$2C!lZMq-LmXW@ifzR_w&Z#|;^c#B%In62127l71`A_|ZmM
zT3WZQ7W(9^v;?sNguD8B$BbA<^{wn>NJJ&f@ip>J)`Re222GH0SZav_CQDgG<!$;9
zyqVeA+2Bl*k^xc_;uDx|Ujztq6?*KJHJCR?BI_5Wm43ZmTek_%)R21)#>`NsLz1@!
zuj3<r#s!zkC}h!bFg9dX0=4h@8;YQF>FB9(7X3mK!5fH_gCE-8-_!T|v*wF6+iLjY
zJ_;_~J@4ja^J_O<W2Y*wKl6KzG_A4S%Mff*?|h9Bpa=k)p`kMTQL!fpg~0y=74RiS
zhsPZ%TU)q6<&PlD<I35eKc{?I!l}|iLI4`{HP<Xo3Im}K$4bRP`j#`My}@QTxvZ=V
zpny8m(0}|0r%jb$T)e7h+7rat=fIvVqU1s>1fT0yS?TV=l9KRFTo+7|=7n6)hw-}X
z6Ifvadt2vp`g^ZZ^*-$`kK<RB_;1Ar&|apj91w=j<8cf=P;#$ND@{>f(jseXkr_gD
z>9JN<@IL*6ewg9yC@v0cUWRX?{125!3zC#*#d>Su{@m~1C#tKLHuEvsT>Y(oX(_a`
zx?CL@Lj19p_!~MNK|Moz8HSBR4G6A<42NKVMkGIqV#MIZgei$R<)bz=5RLs4q&Zei
zOuX{o$!YK;{knOi*tq!cOA`(G`elB5!{bs7bK_^4O7!e3EY}&`L+iH2pTAQ1^bg;C
zM)Y9Jw~hRXd)!Z)E+50uWsQVriIf{h3p*s)u8%<zoJ4Z|k)-{3B)5;hL@B%0BSPB=
z-I%FwJdpJaqu!9+CBlF9?0T#EO0pxTry%S>?^h}G63q>voZ%Jam50lMk_JH+*x_8r
zF0{!WN(9Ptc><)%W-`jkY(Dx@WLFPD!$`9fh;<_YV2^1>CB`SbMJ`$7CmvgGc;Lc4
z1u7Pc$@-e&VwyDRf$z&f&PRkQvKM>1cUtW(Os+tvh4HiM({f*(7=KNOU2EV=kX2Yi
z^)r7B`SF?V8*p`9oUXsNIFRa{BQRdOrN3FOl%c>Xaxt1pOiw-k&nf{H%fP_zi$us?
zjrv2kk>!RvO}M)Ti+>6<6lg!KFro)2O{R~%L-;#cUnVk`F4R41|0ZkT;f{Y@S|vT~
zkwR9ca$S3-Z1CZQY2aw+s_))Zd|tP1UgD^2SH8>~ro>srySCKx%Z1RZKlZIdDu=fT
zSRO<`4Omwp+Z<d!i7a{T{}&THq2HpS3}>Rqz<d4S!!L{z-Tfq48EJrguPn}%strIU
z@sUwWo!P@4RO2`KJRT3srmA+KgB8{aVm8;k{`~HSyCn%Hqcm3gBdn;yyO#P--qSp@
z%Rom^ukL%`MzMx60u$4UPvU+T@Lz&pUkRoUuhu|R*Qb&-!b*U<IB-Qnfn^*so9F|j
z+YxK~jDrq)M)gSxTl;qy`ebg65@Gnrck_Ldpc@G*|2D6q*+xqtOj%c{xR;uHmhE;I
zq=@+D<$aWoA=`}9H1<~3S8lt%ol$yaTR8gjJ<tdb6UB_$?ScI>Df^yP^7Lw}dSnqi
z+&PCsJ~rm&u@9qjvs4GJQU)^Fq%P*ly{}*HZHYYkg^vsoLetMF9CvcFeo<1|Ik}#N
zAte(F)e<`1Y}sEWydsZ?Vfb;#h!gp|gDcu~@~fG%$6nbz6_Qa2rUd<K%qU#N3-%+0
zIT!DIdpq>&nuD$7+1}ZfzV_<h4FjqT41dfg;h_$8#=S*#di5ec?o%6K5f+x=z0SqW
z6x8tU<#`)C%xB{clW9d|49Y>R^Cfs8ge>~gvGX3P5yq@ECmy^T`~04Y9~q%aZE1d+
zmjCByuK!CShW^%x{jGg(G#~w?@hpj`Y(O@nB+=Isg1-izY>nSX9LD6{Fk^HwBGDXs
zxT>Gs=EHNAb;@Cz-)-kxT%0kmfht>5&R3>k-MHAl`fg|rkBUbLX=%QA@O3A8o)%fb
z5pr@ufd(MwD^4&IRb1Cs6lSLBSY2&4_4IsE<6zjnWg38z%o<yqp!AAe-rC!>tnepJ
z@-1`-2x%GIutn#Yft$xwuvj$Jcx~Izpc&Z{q+mV@h9)NZCMJTZe_ZX>ZL7C_@C2)D
z8d-kS7rf({O(Xv-UWhXMhMArVIi?+M$i~Heq+nP2p59|dC<ZTnt%c?@MA6uPrikB_
zvY-&4WMcWz;DZSoLj83WLGYiMQ55-g_@+=}9f$_vgM-s!0lGy*k{mV2xDbuKW%lOs
zE`W2=<Oo9r8>|Mwgy(v`7tC=yd`tmSlpN;MnZO;=g>${0G?+PfiDM$d%zrH8M1<gs
zoqV$YWxC51@rl(A;i)hGK&;D-)$X0bGe!yhu2wTm?0EYSwrB~O;0=+tk8~*uS&Goh
zEq$~eMG_cdF}B9O+iEw8?U0tidw=iMHPd#dnEloM5(^?sIox~82pRw8o=l-KfigK`
z+*?D=xYrvQ9`D=Z3FDm9s_|5C_usoB_6*KmWj)&Rr4dmse>K+BG2J|Lt=t*^ZDWWX
zL*N2+D-L&J{j-#Zvyh8o4URj3kNAH2B|eFBp3G(Tl5~1XcAWv?G@ciEF^#pt`8wA?
znVnH@(jOwa_Toh~r4LbpO4Kwv^M77uFf+!*Og85QF%^mOC^#C!cNOL-N-!w&DME>c
zl2aG$3mLM4`&LGIL)i~SMCfRS2YZmT>GK#|6i-M(n=2g543trAC`dwe)wI6-3}2pU
zWZ`8YM8H#7t*3t`Dn=JFGQPC@+R+64Dlrf5G;@g!37>ntQ~S@AUV^E@*eWL!iKwp8
ziNW!4wba4)%S%X}vUg-z@qlOeFG|8I30b*$_Rqa4Dx9ad4bImxsiOyH$hMuoXP*xD
zo3MUgiez>785o-8A#23+Q64oZi<_vr%QM?OTvwRM$;~m>IXK(Bh#CGy%7_(_G-y<2
z!thjb0V@K>+9xhfz2p~COZB;a=I92Toot$LZ*T28vY2kW_{oLLY`luea;$p?0~}Af
z7^%O};^g;*IP%Q9nf=g8{$yj-w%~^kH&bs{r*h7Sdp%!|tUX#&7Y)0gsj*(4J~MDT
znfdt^v!z8<PUvN|Dml_pGPpr6rQ=6NI|Orv+pMtmI9(N04D!F6W*-HsWSE=J4fVA=
zo$<u7<I*m4qb0<Sz9|WqUNY0mfAX8UrKqe*zE^@;wOjGb*ysX|hugdMQEiDs+K2kD
z_bGU?B5b%7t%HXpK8V<t6;DoeNhZtl(L3u&us^WddDA%dRW-f#z{%<>ZfkL{o7_x7
zR3(;<)VijN;*{v4hDo7ZWqx{R`R90(@#82HC@G&gu`I0Y0t(9duRpiD+mdoT9IJ7{
zG?!MBRu3*x;iq+0PsgeH&dpI_oV{!IGEBLa#;G}lNpqr1d?lTR70s-5*ociO-Lm_o
znG1H!yGxb1JdZLpsExj1Uda>95wZCqk<7{c+SLLtH1T^h5(3`5N6rb6Ta>IC0(wmG
z&GgiC3fW53pV54edbSR%%ZrMvA`l*E=^r}DvF>*SuI5<1qEIl_a>WW3HOKmU`|?^J
zwrubt{`>#hZXhI69b)~^(uaCS)#;v6QKo;72fA&q{VD^h1_bp33$wc>yN2e*#BOUc
zQv_pweD8bD{G<NUEl6#ET+$+$>lrCEHn`Q$RitFgo|^3bti^DPpGXlUVa%!l=cOUv
zd+D~_vYqkw@;<uFz{p~vuS!mDN7wnr%jS8<8^y-~+43FP&LfNM$x1X0Z2xWS=QREf
zIbMvr3M@2a%8(&?h@k(@rc=F90cERX#J%6qr_h1|$MUplPvt25no&HBBie%r*zY4$
za%Vthxwsj4&IN*an7bjRd6>O&w5_bWOk;ZicN0_s<R4^jySWGwWGY;4!@K1<!j>sX
zLK8VmHDEsZhkwum2l8US=oQCQo!bi<qxPL^5Q^mL)|Cp0ze4nA7uFU_*q(1O0Z9y$
zH9&W9i63&p!w0A16rE4k&|uL05ircBIj|Q0iBY!%^mdNFr}rS!K}9v_pT6wplah=(
z@EJ(pb9tVUZwhB!m<DXAlNB~Li=ysxCQ4ObW)^HwkGQ470<KlWlpcTjlB3SC!l%ug
z+PP$6c-%PxiMStjzHz^Pf|!kJJ*naQL7(-S%!<XKil4Kqc_@Kuxq@@%cgP18{!f3d
zGP+BX?s#U_f4qQ>d(Xc%+BgI+5k1bu_jx<e`&(HFbdllROWYpnFBxm)qbh8HOY?Ae
z`y*>yEtTzNn*%q-OA@j5i+Y*eZhrf!>4peJZdu7w?|^W`o3;r3cAnFZXUL8Dp{Of-
z>bZBw7T>bO%S@2eR6Hn=*BX~>PMza08PBV|MErff@%L2Br>{%3!Q8P%I&~Ns@kESs
z>~j3?y=$bT<B7g)qN!T?OEQ(W(WO5eUOiXH%>1#&zan12%avfUvWNl`TM`kq@Z#Fd
zDtC>ou;Fn7<a(}RAjSvHxN2-U<DA_Eg48u93x+pJ=P_SkycAuqatxM{d?A18V}W@{
zsHbLIFjwue0L>~L9X+DfY@l(1#5RPd5~XIX>rDF@0XaW&91)W#WP*4@%hQv|q6C?Y
zAg5~`nuz@~WWuL5`+;__zhL{FIyzT&;-V=LVf{^T!AuT|0i2}%@-Gu2SGy=v(eE*0
zyw*)t-<k#?JXakg=<!bF8b$-qceR8pKr9(3^cI)b^GDYyW#YZt7h~vR>zpo0$~4ds
z4_ACoH)rnTJRjnSyOQ6;Ib6A}mYy1%MnXWJGVtv7)>k3kQqCVvQVdP+Cd%Agxw(~_
zsrld(K(X$0r8(XL>RnIZYNZo-Fs}J*x0YZfK3x4WS8GGsdqgrL7(@TVv@$nyP1EIs
zv~}%jr6-oh@r^FR@p|RjH;=o<*FH=F^0JYFMR{GW$h<A`hS#6cX<wgQiy__>#lj6P
zhx!*Wwz=^*B+?=E)~6HmL-M+7Wq1BtBm+Rki(^bHYUi?s*KLztV0Xj|{^>F1Bko*n
zabY1>>`Ri85-**)!|2C(?~W&2*iO|i;@;FxpN{y)V`nStBP#i+vx7{j?u^PAE2Ub>
z*2g{*X?Zfj75sW)>wT!V8?Os;<B{LtPn^25y5_cR(vwL4ZbVYU$AV1vk4R*}iG%z4
zj3JSzh~`?<g(MlS?;Pp7HM-B#(s7U)OWUxI7;%+wPXY*{dt`+?*gvv4aV}N+6t((%
z&9V7Ox3f9;f@?K|KA0^{;C99e6^YFJ3f0oI07ploVe>_p2_b@uv%HRF4Nak`o;GOP
z-tbL&3L4_q-LEy@oV3aCQg9ig5@#2?l2SIXuzqF^Wycp0@Z-Do7E;nO{bx6LP`1YL
zT?Iy_?Ww9~!5twpsQELM-m}u2L>2%i0-<%0KifXsOwY%3Zs4k~qvI8k>MgD$q?Yx^
zNF{IT3a0Ig;gON|V}IuAHz`5i#%HiY-V&Bc%8!R=(eM16o)e*9a`9(;+{mgWw=ZGx
zPU9r{6hMP;wej08Jksyk0)iGA5s^F@Gu2AX&`$(5Hx^1iUW!UeK17I%OQ;vVXDbeS
zW2yHDvM1qu(*2kBAVyEmqTKje!eg$VmYu7d={z-rlIgWw>-XUCyC0d?2>&pgp0m;0
zcu=3%{~vV(bSUUn61Hy@iP7&7k)R^ZWU1^(EFibjuw5sxXz<6gh}HfF?ME1(8PA(~
zTl{}o0B$SCP+g&24M>W7on{;%%4jePcF`q;k>Iq*QjG1%k&D<%MbJc^14V~%Qobrf
z#8c=3_1lBHHJC}5zqwCI>zuFqx`(ejXEK>y_$sLc$Q8C)mSgzc@m$k%0ARcAEB?XJ
z>@ma-f<a8H&i>(`>-faP56;@vzdL!ld=R;W_zQszkm|1Wih~G|l46;<n3B0!!KtMK
zk{H!~k$bEM|FQxQ&91@rl;scf3NRcdP!R66iC+p;*GRe0EL!$ZJWGsY7~X4X<8ZFZ
z#UylpUb*8bk-Fhxj<4A?gI67l_p^fflZ=k?Z+}f}B+oBYNC??B52d}MQGe9MYf+@e
z1JzAAt6ec(^R7n-&+FtVd4Xu>wimj_bU)Y3w|!3deUwEoAO6fN+sF+hLhOBHz3aIw
z@DQs>$A=?MX;VNcxMG<!l3;l%Je2!&nWBlZsjlTcM3zJ~<%bB2zrv??Kngn3>*0^a
z%yYpgsWaDA)1Nkk-K>A&FuR}hm*u$u7Kh<XJUuH`o{_{x@D1Jwmvqapp;3u&;}d;}
zxfK}+))(;>?x!w8m18?k^D|pAL>Nz6NsH2p9Os?dtGA=FmV*)tM4#0&xZx$(w0Dgk
zG_Pf3IFnTWa3ao`pKn=2#eN|~^x3B*u<Ms0R~(m=(<|<V;>J(Qs7Kqc9afYOd;2F`
zR*wAv%`Uze8-5Ji{5a=Kd`z6wd$91<R9J#^1026?<s829Fds*LvtOP>&yreM-#MY*
zrYe{J8q1q!S+Mi3xQRMj^j#1^UuycpRol&>CkQ|oyckf<E|ENCGBm3<IN1HT`OI3j
zO;>-(l2BznhDJG^jNvDHj)(!1vQrYCi8{CDz2|oje+P;bFGQdmJd}vN!}l0fzu=CT
zjFM^)NDrKE$efRhQLqBGAS29IGy9Lx(*ia4R-!#W?GL-+PRuo_GJcVATVWuSXhv7I
zT|n6t(ra3v%L9RWz>$AGN<;jeUqM6oh5W?1PXjNb;+=(KyS+Jj%u5^<@W!cSDI{=J
z@BV)@T?16!Z@hoQ(pqiHURuk`u4UV{?ONWJwQSqAwrrcrwtb)fz2~0uo-<ZIJorA}
zkM7o2Aa-Ya;^+c?-^d_;E>$bi(g;wGS59^gfZ%UTHw4<Z@mFrcDW8VAMmlf3?doum
zGjbhpB>hb*KA%?P!o1X$R)2!Jvf7zX3DFvAZ#rK=(QQVx6-n1K;fHkbZR}6N%6x<@
zkG2pcIrR1ni?i3*$rjKW#2GCN;jTc%30+q@#aN;n+Q^#gFI|1xC3qBQ0H-RP^4^`>
z4&C9fgTo0y<`!mGg&$2(>fi@LtH1F~?qA|NTEJ^QmdH?q*wH@Din?$^j)GQ(qAmiM
z_jK9GdnV_hi?rKA-`O@OQwmSv%6>^q)gpfUS@sNS0rJZ!u%0T8Zo1l8;SB}z#eIi@
zAh2^o*i$%8P8yrEI*We4t+?VTgK5+~!-imT>qcoNb+pbv_i*cw2Aoc9B+OOYW+St;
zFX<E&;_k9lyICY_(PH)ce_>0LnY&#)%RY#{Y>OD!J&$l&d5((t`WUnLplP<;kIF%X
zi-VKJQEl|s%4vV}SyX<afq>v>ER7vdZoKg@FDGaAKxBni16!d|$j>%yEdhn^CiQ?9
z=r0gFNdzp$j)cQXLN7PDR4Kv;)U|0NK>x3$NxRAAXCGFr&CJHZ{U*wdi)Wi1CV-VB
zB*ZO=KH_`bUx7h^fwbK$%{6ZMR{Xk$dM<%JLakX31+v_FnOR$k50Oc0ON@#`TdZ2S
z3}gd{M;tf;qBlb9E{5L}NwK<p%P83;v3&Dcqshh*bi(_yU~GlcLcvU^mZ&w_{$8iE
zhX-h3pbW@*KY$I`+N6c=Pv%>1?z`|hY}G52in482i$aB-;pJ^bFL8uO?1yZr8(!K`
zdw$$+kJmewfNi8qtD)&6(rT@CXZW5PuxwxD;`U}$NCQxuTpY&DY5{W7X8-sSxqF|W
zZ)(W|*i-&Pv)PKty3l|FFhvom+2m5AMsmO4eD4DV+BWiCa7`^uAbcEP0AIek@Vp=*
z<IJm;%zy{AX{x$K^h+eC&Un&?J45#apjtj{TPg%!b;emje63_5zEYoId__t)Dm|z<
zTY7!lYYTN7FW)2l0(jdmY2>x4T@a03Tc=MOk$6q|my3|k0})tU-w2S~sMSc(#Ak+F
z?=O^Dn!)EukS9>%J}^Td=w_#y-{QU@Pl4Ndx7vPdz2^nq^${FEN2a=@N#eZkX&9FX
zISXxvFh`s?SBh#90REFH&~G$O#O)0S0nebaO5Z;m|NDzUtkP%FfA0v2qY$?s`Fx8g
z6Jpbl{Eio8CtMmOCBmbp$JpM@sw>qzTXu!?lBP>ghyWUEb>=;n!L7GmUz|#Uyf`{U
z&V0s0u%St#alB|Yiqdqjn?-&Lm*Lrr3`nxo8>gubHP7m9nqqE&&VxtSnStrxzLHBg
z`y=G6;1SyQXHgsY(3aK9pjjV_xlOK)Tcdw8HkqP~3{6i69)kMk{bHDvJvlV#6?SR|
zJaa36x$dp$$9Al@J5r0n5bV8~h7VH};M1bHn*!iTa-V<-D>~8QXkajA!Ws^h+1RX9
z9<L~dPLjSSd22F>?G(cDQ^NIYT|>@7E)pXo^QU5v6N+X&IEOA`m?G842kcJL!~Bsu
zg$k-~iWH4}i=&Fj4h_*74ubZ5(P<m%!G~b~o7rt>SUTC^OMUT=mmr!|Q2`8nz?1)y
zPME<q%$arqKo1Du(y{;KH~qTDPn!bT7IZr!9#X-EkSK~V;dE|zh*R_RusIsTuD`|~
zpJXQh`T$lxM9<i~NcSBCP<jQ^e}ZrOkO06U(9AbyJ^)CK7}@0b-{&!{#?BqP8pq34
z?qx*SA0jzPjP_v_kqs>+L-&&MfGaMM*SXgVm}U$yn8;?UD?V4uWR1!IAs*Ln*U!z(
zzr2UWjG5?=l?4yebZ6Sd`X>&u<$pf64cViEMh^Id1u8|_X=$JWlaLtvO^u^L1UlKz
zEjM#LJ%|t*T3W)jCV<<Y!L~{iRabtkC@tMMe<KrdV|K!{H>>#hzr;f?59e2}G5%UK
z;Au(;aWca?+Bw+RIhd>#o&r>>Lxog^31Hp=J6ZDKmNIvYG>ELMLR7XLfP4Y>$ev>_
z*lCU}3?=fi4dr4^b)43j)=ZH8(*D_@acoHH2mujFa5;v7c#*scwdHVv9VBr9fy&a?
zG-KmH0kXHfYVFfGK_NT_&$cJqi3X)-ki*blnjdF<*|1$@Y~W&8w&VGL0ZClqULF{>
z7hH;i{}|!bP=!&ARGUa+#xS(7Y%ZOWhsBgrS<=@+G(YFwqES+=qy%~*ytLlIi~iF&
z#3+CFu0L3<4X$>hKtLxPvW#mdo*ixGluYISrJwF=a6+<r6byvKf6ru88>051cKq+O
zx7QLzKYCOU1!(#>P+wY`m*$uDa&)%757fiMW#W8{&?sTeq+w;qYU1RAf<Q%9uz3CE
zG&%EK-CLP1@z^Ep8*2@6pQ?IxMgJj@S8Y$)u{G^PmoS%O3A~3`ig!=6Ii0QMM2eIs
z(FXyC`>c%(w-bcKKmgd^&PuZbh%o^4`^N7#V{r{OVkkX<d9mg?jFrV=^%&pz#$SL`
zcXd2NAby?s()Lq>72X#JK`|0y5vhKGLFs5BmBeXYs!>*6j`?{1y~Fbv^9>aS5_J@&
zL-)8Jkmd-?_X7rW4rk>;DiC=v1~c0XBOvuO>Yfr7jOu`qQDlw-_0qmasEIe5Fa!qc
z8{fV$9bi=l^eh2}LQj3?6$fU3mI!v1m8RRT|LIwdKu92v^CT*-%fb@|S}tVc>=fUw
zbzcntk!DOsvr%aGMs)S!dOX3=ac4n<ZneR)s-qyv&8wC*TbU*lyVlk~4Lkmct_w7$
zm(|)H;`#%n5>=jYsEONCg%Q{11Zt3c-Hyl5M(8XGdL)n6C*>I*S<T^YJhr3n>sg0?
zipp5j;-d4m5<0`j@awYM?mUr~95~c~J#r0W?SbztDd~mJlmEpSAc&vv*r7uZ+&p(>
zj`O`9ehZke@HD%W=VdY=H`mR8>t(ZAL4?$FQINjf?LbB^C(H>mpD0CiDM}q#07q*p
zE$Da!EJ}j;2G!w1*MXva@PWMlcgh7=A%LC!j?cf5N$v;_Sh8Wm)`D|&P;n3BPv#Vq
z@3&8&{@>c#*+let-uACo+cB!C@+Q5Wx-3D0`V-JAl!B>PT|~K1D?mq6PS=y_t2LYk
z_2-um45aSn5;5>ZHv0aN!E@A9YhA3d;qpWiR~-~s!HrZ;*a3TnVz4BT`AIdiC36W@
zC`g?O3<Sf07&4zUpFNl(=pDAcq4`h39rDj+J>-R%*ekypv!wengN`rRNugwpd0GlO
z!uNy>O{|cK;ex=g)L?egAB(YEZ%mKzr-j*TFj{xlhU2@p#UT*A(U=}bJq8t(kM!3q
zD|gC%u|pUjG!10RP)5H++W(Q{hy9;#;uG%op{)>L@;$8G=(#<4pGC0bRe%d8@@mOK
zV_F0CXL%U8NIMAjN1ngGn9|7l8CVg6g|bC*s%omBAYl@lEq|{~qzF2^r%the!vtV1
zBncrMPuqYot5MHd=R4v>#)RDL3^<i)eLBs!UQg5RUeN5NNp9RI5I{^R%lQHKUl}}s
z1|YKbd>^`?0MrpLO+3s85@Bf(-S1}j$Hx{%0RlyqVs2#dig>c)s^fns#5HeChML(j
z4D*P^HznNHBT33lym%=mh*V<NkMgJ5!~C8b&)g$RevaERRh!QoeSEI&Jdv5{;0R9P
zY~=T0*3}x(mLo}iv?$Zajux9?ykjPp?sv|PulW78F0<>|-nsL=;)Bb@vwBkJSoy|0
z4M>7yl+IIz@%``@ftDWO_JT;-hrOckp#}-&UA&ljI3}!zA{V5Rt%~V0_5ceKH%jE9
zCfzzS`>a$4G&q)>z*$GM13Ude4v3z$|A>Jl2(N7C13DdNz01!HC%h0}(4#j@O~j|O
z9g1k{XOi!bhXa(+5+M=@5EDT`NJ&Yd2_t{iG;4(29I17G;aou~$$G(i#~V!d{VOmu
zpQ>L0&DgqP;{Qk-AA^Gmldn1w2064KnMvbvB!rOwiQGDWgO7rWOo$|vhFxzk(!1nJ
zwA%jo``ag=5K4c~s*rKAi`CgtoMZ3m>TGY{>_*<4Izs#&0n}*^Oly;qoxeo3?cDmE
z(E_`of`sBU!UxbK{^D~DkVS{TtI?qOaDXF9Y^WD5VknTKuC5*=G}BKx`c+LTK#8iP
z$5Q%pl-#})+)97Y@cKpEAnqmuNtBsoSFD6FG!LDo6Q94qu+caNz7ByoFQ7Wm&<?wE
zL%Hw~MpC>#ql#PXTpsT3&IRS}Arxcqbgg|(SSdbZFf2rE_x+h0wBPUa^rZCkq6`kd
zZ)dmHAXQTM*3y~tNZC&T1)sP4#o#rPQ$>kZqyp)A8I-1g6W533=^70@ehf(siEJwd
z)-S6aFaAvl3E!8n0LrsuLw&Xar4}L45;EGEo2Dk7zrf!|BX!X~n^1u{3SW#E#pQ@=
zYiq|XEaD?=0TYP}^b3}jvRDtk6(~Iawd-9T9-h8JB?!$O8l4p>o3Lx|)zo_XOIFA=
z0>y<fDYd{9J}CIzPaA71tL|;YuT8*J#-*MnNoo{JN$&?N<l)p7h?A!FnQ`BN+y985
zMQ8h(wV62VgNcWRCMoI2+W~FZRL;Ux=~>AK%p_}Z!@ZT?b#*R>0jd=>b=AaQGBi8`
z(EXpV0DfDz(p73CaRY#q=f>@)kS9>ko}d32BBxHCJyo{N$YJtc=+CA)&1V$&JmVvc
zjRK{jGdJ$g0RRdDfnv^tUfY8)-&_?3woZpvMq0{Tz4H)I2Z1bWbysjqJ!jVf{H!4r
z+dJ$81c)=S_veqnvnug`!Q5deP6k#7VB-&N%<r2TE#@I#rvbwNnZbXk=nLRHO2Iu3
zxd*D?1q-PVmr^^{`};Q`hJSvqG_B#k=VVFM)0jH)7w!Yz6g|ZoTm&kAfdb8{g=Gza
zeglSJSOIkoyp|4=ItHeDhdKse9(Af`ON}XM2?*67TSgs;>et5szCxjw0{x)}9}BP?
zpV-g=O&8b;;kmfH;*b_`t6p;5qZEpo@tt8J`z-4g9kD9l>MZ0%9mXr534%*_{qa>S
znPIsJm8~Ajp8@p>cq`%Z7>I#b#AW{>OP)|n<usW<t)yB`0!U&?3h;zxq9wf)wDQyv
zxahm<?xCJd<FYTLs|vhs|G}#X7bQ;SFW?vN3cjYD`9N&yOgYmuNvZk6gcx6%05crK
zzj<d#q+kjJ71I#RAYKnA<h`mBgjdv|LWDkNs94TWYtnNjJ5?hx{plG-G{h*Aq~BkX
z#rY5h`IcR*PQo62+R*~#EA<J+S0LxE8l$7{ge%o4yUj3#RNIxQ2>Eg)Lwe7i9iJ-s
z7lpaA#+}hW7>K2gEoX*(9tb;G<Fc7Ps67$IJ)HUd6o@OqH}^h?lHlmcqtMZVrixDc
ze5QdgN(|UJCaOXEYLr3waz+~iqIO^`Bf4RZbBar92fF{h88g^5Gp(-I)Ztd!blVJs
zSfL@$w?P_Z&=As`G3a3QMrXDt%N44##k?3X9^A{Fl{Ak?iJ|#NSKMmyihH*grFXe-
z0wK{rk9kAHYh!KP<huJAi=wZH32RTZ^#@;v9go}9(eV8c-Dt$2Ds5ejF;M-~&w9fK
zir1}f31s-I0s2(bPe%JE5*?Q0A-u(M{K-tdcrkK$D82nq7S$RX<V(WFrWM?%?8R51
zaaBg<N2CtapB5`qWue|GYU&1YTL!0_BWf9@e!TuS31~PeqGl4aRc(8e1il(2pFfB_
z)ToumqlOXoBq$5HEX|G8?4Crzt9GpB>V847>o0^4?9$Iu?WclU!zNW7s<`1<8E66>
z?!&JEl(nP4!;P{&TP--2q)Ei@g8$$v7E^(!VD@!MVAKzvw`&%8Q6JQOu4Zsq;_uQX
z`qszwe}b+qZ!mjy#KBa_(%nRt<j`Q(E?v2#3M{uy&Wtx^r|gR4OKrLy8crFLQ5&uS
zY3MU(d=07EZe84gD)MBqsA~m2@)n<6gDHZH?AW4A1(>K_i>~cOR+vM+EzO+Xh>&&!
zsE!f&T+*>Ja9kfh&iBdcf-)zA$NQhCm8}ovi?wb+!>IxA`l&R$g?bh;G|XWc!fImY
zI)4N0p_u}tM7-WN(L!V*x*Wg>p$cpx*-HT^M_f*Jul!wo=CzB2+JY1jcs)CI>d4;#
z{*e4mbE4c3Xv^E*wdL5ZXtx+dX_QgY(52%O>Syu-Qb>obwynO6iaPoC%aflkc~epF
zfL*#!EL;PeJ_fcfo&!lL)>_{}{v)Gvcqst(Vhle$wJE)39h5a!PDN___hS`1xlfS!
z%EX~rtDsO>Mmc7Hk+a*1gqlbXF5m;v1EBP3HRHa6fFt4<N`~j|OOA`RMoLNw&=5|U
z(Ec9ShUI<%qjSf^1jrB-M$+LRHM{gf10(Rh00}9O!IdP5ij`2VfXd+5{Va>kRH#c&
z66|z5DGqT1e3Tn)79Jw!b21co@bx9Lw(d_?Z*IgQZL-k%A`BcGX+D$E@;HtJSexZ+
zZywiuU-)5B^-Jv27v;fsnl1)g5xPCI?@B(<zJhp?3DM(;O;gbIhk<yw+BAFJM&<lm
zsr>s){&D|)NInRr0+iPvCj!WQcXoEN4-WFE=iXBNYWwbFJ;&NVJUzXzV88a)5G+^K
zHFY=~Hn@*V$+0)Ye?v)4aH0kZXN--H#Mdv!$w8;9LpIVzn{%>IsZN8**aX%f<?`}c
zy~<4|PE3d)zkiFdHW}M&1yHO}P+WwdC#I*nM|$p0e)e^oas<&4a5Y{I`<HmXYkR!i
zCypD8NaGHDGG2shNJisL8p2Xj*8u*@01%a@ihFl7X9@T|FahDhL`c9d3p8qh<423j
zCg_Zvh9AZQr74r*RQhwEVTB7gCfm-|{$}L|vg`6u89;<2<N#e1P@A4ER8@J+;g;+u
zOYYWi0N;8Ns84^^T<<sBK`je*6B7CBUhB1Ud5Wl@A*>N_C+in?^)x%@Q&_(@&CCLS
z+Z&1?qo#z}a8+7TJEX1D+6aR@s|;JgB&nN&>lW}FKzb@*Gy&f7&iV;pybKvlMhc}9
zPIpA+`QK+oZIXX1DhZu>P?mJ!9N(`ss-6H?%I_7@B%$wayB2px{%E}s79al=OqgNP
zXrf_dI6a-601cjjH$(UnE=X5tKmEs0H%;UW6*#nM=MJDTi8v~976M7#Dza?%DNxz{
zH@%9cK@-P`{9$V$6s{WZYSgR91jvd(xV%BI(dKL#P@+TZ%*Jkh<)Q;=F2NqqCO4eS
zhjL>9@&{X~x`nkhd2zW312))>zb%JA-m_ZF1FhXWetBBi$gxKx3L%Hzj&=bGG3dB$
zUfcnm|8naO<--PSlqj)0g;^Ti`9DKI%v+rp25c+1J??QKz#V3`OzUWWeRT8u_}HrI
z%0wC$NOMjdro+ckh)wc8Gd39v4kH8$0rf87YBLZ}yy)<7#)unnu2_IxXHeS`5Da=>
ze<ZDrlKNazTdUS!E^45F1PKE+0(ZY(a%D?KN)lZ`UK}GgCLqj#ge{7g0YaN{)$$c~
zw<|J<3|@CFKsPhqwi1PZOldeJ;<4Wui@5NC5`z9av3g@Vw^86Hk#+ar=93h$Om62h
znSTB!Y7_8q4MkDWkV>*`S9shXI)7DV9P<O-|K`USK#V8UK17G*{IkZ!LrT4K{0LST
z^SZ?(N34Qe<5JD0MlZ3ioO^Fkkc`I5K-UuSd*J<-KZ+I_B47l8Zlsvsk9IlTj_2QT
zX-o!tpLB^%t8xM>UaW_@`o@6B7igoAtwexG_+zvJAl>ZDjdRfvBN7NbtEcrRth{Wy
zvm+hHe(wJ#eveG`th}FM5tk@ab&>b@X_o-z(?$@)w~5;mtKG>K!k+4uouk+f8C!ju
zQ4w2$0VT!PC5y91E@Xt(C*J?w$X%mZcd%6@!QyVhi_TQXNB-%%IklWctnCo5&t2XH
zlU#<IS~KY>9AtdYqSj##be~i$@c}Cd^8R3XwP%@w@t*m+P{EWwI<dSun+hy#tmPIs
z%jnBG+aZag8GW{ki8@@U$xcx3Ohu6)mcoTWhJK?THE2|b5F3;o44lHq$v%UEx0nUj
z$wBA#&o>A-q->%o%Gp5ltgK)mMdnGsLtspI)-J+=8P`d~e^f75m#JHXZ5k;CsH~xa
zg|;r=0Go&ilpDZO&oc(PuDzv7s3#n0Sh!%Nhr4}AQc{PxaTK73f0x0TrFaVkNOA!g
z0y-b>@BIPp2wEj$y6rmcCZq;dz~K@C0Rj%7;zvnPl~aPvc>4U0o>uFN;99_HIh_m;
zP3}>nlu^KG9PVs9J#8NXLa%%mJat@gEnj*oE+lzHck~=Z#yTURySsWbp*95Kd%Im_
zpW4lO!(acIwdjWvWN!nwo($XDE)(D7;GSsWK!0;g$!~zud0}k(so0-4Y&p<cOS}q@
zAf45&kM`=S2{N<jCARMv{Dtfr>_*CledO_(ve$oOkux}(m*yn>q#VfAAorI4I5p7p
z9J(<kYoEuxH3Pt8AFAMo#xo921s3o0o{U(7`N*eYfg}wXQu)X4SFpaO!u0D?%MHx+
zc3-QgKP03{c5vtHqi-d>fpCt$S!y6#5AK;Ew@AzPiz^T*S-|r7#b4+0Xq10To4KE(
zQi36o*G^Jr?(3uCV&iGRs1Qw(5u6_m)4oK^&OJIjeDxA8=?1N%i{rOy2F!2tzV3le
zl`m&v0Z4geQ%yKpZUu`DmTAHC74nf%q)5(p_hyjB_?hf;po>+3Qn3MFFU&U}6#loQ
zKpQhK5nt>jhUe(W)NL6-HZnUKAoZC#ZV(`}z;Xx7jLHH)O9KFj8VT_ExV3XNc5DQs
zG+^|R8CBsoSfW%6v^zyC$?a{;b?<Cgye=BO1$zM-4agyLpQp{33r1gvJX8r(lyB<h
zj-lzanz@t|ev4GWX!f0)RB6h8z(Z@ngX!tc%;j%xX=&;vxAF9`VqfM*>{|<CT07(A
z<xN0)JmNL&@F;WT!SK~vE?obTxy6U$!zQ|UzVo~e(_*{g^~KY!C;E+oLMce9U}a+i
zmiTFQt3+rCa_Y**_Y)qS^YYlDOu%hg)t1NIvpxIAQ@u3Evv)*##06)fil*k_B#!43
z7&hS5F`Ue6{39`OFIpMT@$v$Z$GUDL<1)o}EQ7;`BeGAk-YosBC~UYiP7(`viiM}M
z=TudVU}&j`d|xZ4ZMoqJ{N+Xd9sxv|gOYk;fC&o<_@}2N;Z~?MG}XnmSS)m2m&DWR
zG-aiT%BF37?VmPUNR|dEa#Vv74HfC35#h39{Tt_Q4%Lj%?Rw{f@3ON=Zf$iOu(=LE
z-~~7Dn*JbuB!#Q?s4fZu5{Asd$fK<7s!w$_cCBe2k<bpkZi}|&P<{1R%?@|ff&x;i
z-GOJkPWZSu^m=Qp2E)^=v!E{|QEty0`#?{cWqdH0_cp!5^AcF4L9qE;VzazoPdJjp
zmu?ikpc9Q@fuiWx2{7SrQj-*cGy{|X%Brf7i$JSL#9Mg?aHCVc?|^y*BHC?NE5DEV
zk%HXt`F_35+9o^|`V4OU<)=K$&EfL(|K|c!r1F`Y+l1UzpLX`qXI2+K44(_4C~^k=
zcT$<0Q~-HyfRYNHLI#KRMK~8Kg4<Kn&z2S4c46|BCg-l%aLa~Mj(O>QH}3TCy!(*4
z6yRAD8;4G~()4iBwOe~7^&jO&y;{#@ey)%K&2*NHIn$;K9gVBe^|4BEm(TqXTq_er
zstS#|-RyPoF2yN`FVuhOt)InIqPorjp;O)DWE@gSq}0@x8|T*RqO!PrCVTIf)Grg}
z*qFcc5An2nu(0w{a+4INESIY{ZGnuRemI8%vOuWk>*igvBdb#2PG>TzJG$+Vmq#{N
zrgJlUsmS)C0(_FCq^w4><LEP%p?vd7Sle&iXvhb^TZV&qkWQ-G+skUJpN$8{S%((a
z5s)Xx_qAJpxW6!wy&F57$KCBE6=$?>4DwCV(o#p~1=yLTH+L*n-O2%iNAaJ6lL}Lr
zj4*p<G=g5wyQ4=6gk+n4drtTePu|w$`bUUJnp5NqH71{vqbTfbZ15&B+?6Q)JD?*l
zuw?OBSnL%U1-@UvZ1e!@{qp1_<M3*EbG^%Pei|@oU&^lo*(Vxp`gjmEb!3hI>-D}3
zl^|39IH#peMO{J&oHA+2n@s!ie~wagz`}yI%Un$#)>wxB2b{H@hnb;$r-#~wP)*II
z_?>)^fH#uZP+D0yKe-(CS8rIOvbCtl{ckD`-fa1||E?&(NE`$&^iZ>bVhvbCNE7ni
ztfRSJ>021<*Mm4|-i%X+qkq-Hj^K<04S9629vB|x*P#H}6@0BeyIVMK-8^5wps}!6
z0NIl|b>5crhs(W_o~zL_fR_axw04sn(!kKWng%>Rm+6bT6r@mzTr%}rTkR0Zy{y1j
zK|h$L*J5mGwml2_p_NT*GdEaBW+LS+2)hCiOiD^lujRCVvvf0>J0_>)G?{Hxz}wiH
z>tDJh0!s}pUY$C33(qYA0@6ymyEgshSLf)TyP~#h^nNV_X?S>en;)Y=7_<Bp5`pyy
zNYkk)p96mcv_>4{?>0Jrf#eKpI8EQsnD9ON!}PPVh>DA{a$Rbvl(;$0h5~RqI^ngu
zy#HusZlj{Di1r?xik7^nAm<I_F)hN89p5Ui!TJ^oQdz|arkoaA>d?lU8lSc1K5!G?
z<i4N;#>7RP<V_7ms<iwFdXmWLQ8lz(;=#49FxXAML4w?99)H>h=Alao{IcXtiJ);<
zw=JD-DrmXc`octE8F{+%-8mmOTx>*t_qc8`pE7GgSAngF(OKs;3Gy#x*XN;x8j`DO
z3jOC0d8{-UAJWMApv6{aL}utKP>a1<6`ePf`R<jlhNoc-dH1&>kz;FL%d5&jk-$Fb
z6gC=8S3wayy^!`u@xaaKvjJHdk!E`3gvTzT2L%nDSrz>8Asggt6wB2JD*MNSk6P%2
zO4!AuSxG8sO@BQe#XSn*L_#A2GsMeA<r+&RyV0hTEbg8CzXW!ql)FfJd$ZiAI4i#U
zDb$g%^;t3jr-Zk<FJH(=)Inl#Ry#t9fdMk#;Xw@nIX@nUbbU<as(|#&SmQ8AO(h)$
z@;<(kV>mrs`GddL)X?}%+GVRq@t<d`J4i?LADVlH5p2|=Th!RCcVzzD3ulL_)9xMl
zbfFZBkw9maBP|_v1qxrEi{+f`?C8E-OUvT145ptVIio4pOUI9q|Gk~TPMR-d_z(~v
zGK(qm(#L^|l=QT9KLq^ZU;n%pesf=Yg-n;gH9uPX*pi!<_n9-y`}t%u|JXws9@oR&
z2g<hzP;Gr9>OVt%i3$;M9(=@0$SlYq$Vhy;E8>zA(9)FLl(p)#w6ui&V@l7*8~Mt@
z`28E-O83?@o-1Lgc2jQ3MFb(o6-YrP*8c#){R)fBmr^$`J<2o69U7O-CPwZa=_l8u
zyg4)@h&}iw!ez3(lC}{cpBY8;d!pcfM0gLLBhp7Gp(zgN?^@q1I8VNOg_6BP3-I8g
zVQn9JaHBbLqT`ALA#F0=`NZYC(xxO%k__|=i47Eg%tv-<!9eH=CB-58E@(-L8?6l$
z*+tKq=)rq%klG;GR~#8eW&PH)7wc>r$AQtQsJb`goZjT@jpx;}t2$9eZli11#!YvX
zd}-HYHk$dnvhYpa;ld3i(l=KHl=L(haN%?KHA@adxD~XAZ4E7+&T;vYS%u1wKe=BU
zP!{c&kFNR)VnEge0u^%Tm+)0iv*#o44rg8YjSVSfjP2#tCzcLdRYm3U@^YS9+rS_M
z!-K8!^bAk#Rkzvd!{p>-fld-5V`Kc@o7P)j+(S)zJus(%Da{-;8=vP5NL$kj3C6&T
zTW)={+wW`w+!B$4l76+-{dsdF|5Copwaz9cg5C)3!|yZK7bCzaC3g#5{zFvr!}n=K
zz!I%-%$TxIc&kfls4&F@)-teginWxL=#1!%V{B~$Z0?4TTOv>Hx1@xS(ePuW&ZbBt
z0MZ-#@W#C0Bh`^{gWh<9jww(St+%QtRIIF4J9d=S37fg)e*2WjVEr+=<>0^b{Kz9X
z84dXT%{s6s=4f({Q=XZP|5=&20#u%CC)~f|;?%#I_C~s#b`XIYOHo0C5F4M>dJxt%
zvSFSdk2(|><xCW!<HkiN#c9i{TU#0ZJ3f3pI1WsjqA2-74P=AZcn8!}gB|X-AArgM
z4g(!CKR+*&3E*=}b3T=6EE{IYDVl0D!Ot>%pu*zs@5rY{W@ddlx_qO3b8+3$^VY<$
zAX!rb^6#Q<X^&>(i3aW+&^U@>vVo@Y^%5#Ui#zmRt})plo2R{PVC;h&|4WGN1QG1x
z?Yhu!Jlok7!$wR1uU#W4CC}98{N5AlPn)PLRAzZOFDVlq#-*)qUX!|xH^0Vr5zU^5
zH}e`vq&q*l%&V)vhRRCGNRA1WJoAO`>|k<+yX<sgYz6rh&rNp+Ee$@OEx<!yq5lre
zou=KK3dJ8cHQm2^7alh=(|=N1+#Dd0^|W8G))x67?g+ea8`;qyPR_OD>ZPtkZIvfZ
z%L`FrWG<UU#P7d3Sa2pwgP<qi96$@7L=utTl$Rt%rqk$2a1A?cJuDS|h=<!rnmta+
z`{)VY#E$x0iIq#*X+kDQ3nh*)nY^pYX!UAQHLI?v^1NMVVqBr3qC-Rstz>3<K0(5K
zRI;|(n%B_G%{MhNG69YRDBlG@FYrg|@_Wsa-)BJ=ArgJ%%7qhXcb?u77MziNhBb#%
zjEt5GR=gRqS-;&qy)C|fP8cdHE-o)B!$5$YVW}%T)~Z;}mHRR%fuuym(vc~r7T4C)
zl-HV#ltLu;#{2H52AJz9qUH4OseOPmOl&(2an}5&`m<|&UM&d_=8TO@Nb#!bEuj92
z+VV|DWVCv^TIp1j6{RPpJ^(IIB-`n>k@23BhPZiBT+b(`=l5LHL~D?(OJ#Ls9@_z2
z=6f|{#<Ie~uyDz5E4@RzL%HXMEuh{yE{#m*41CZrbK3&ab9q*fp7jJHXk?r#j2#v-
zNHvTe-?s}oRp?FmRZ%g?@qdj%_*=g80#)&Nyrhy;Yn?w1<ZDlN7&2UAs%4Cj*Zsjt
zGDc`!9rKG;*cqskbMJ{V619}H3<}peK-JPa$jyFbDJYl&hS$(`fT)nBX7V#N2QI{Z
z2UV>6sl%@{EG(@q&jO3s5PXm0Qrf7Tmale=tZ68M&oA@*|4E$Bp??S*mC`2ABntvx
z6tYQRsSfg`wlB%Y$Mql}Z1voYp$d@(F_<Bz`7IJPGS~%%c4c{i11v8?lGrEUEoJ?(
zjI)PlGgrT+J*quAME@y9d{>4;w6B}b<D8R%4fVfNKN>o^cgTnU@B#%FP|h}2R$?#U
z(&%!_N+@*leI+A1qV^ZS0VU8CB^p2&|DD5f`}tDLt7{@*V4^~L10ED;Fl(NlZ7d85
zb3f5jksv`#V8ZEBc@G}R<)oyXdZ)!LS#1JW?>WUM$SBOir=s@oHI#pl@QPdFf9yYk
zo%>J_W{kWXz48-DkaioMmNk)0*^~qVlnvq<o0WkIL0jP0!e1zDzC0?L;;`AO5xOSQ
z^<;76!ZAm@wy@m#N$Z{5F1(9Ht7gqR+OvJ@#e#AZa)*cF2YccNPD<$i$O<=NYI}=<
zDBohqX$qh{-<v&+3^B0PSU4OhTaka+AkC0Iv8OGg%sO{jR4n+1CVWj!9frw3X&|eh
zQ^=$Yr8ks$jgcym!TlN?P#xLPhVHAEGd9}!U_h9$o5_I7<5W1PcZERx#*rfcq#@R?
z%#AN%tKk2&F?k&JRGgHnUwYM3k*)_xkiT(0Ty|)MwXgPX)68Z(GGA}cMSpQ>ssCj%
zzC4DNT@Hu&>-SkMOYacM!pVo6tEVc8QKNJ&XV55|GdA9)Eh~F=S~M8cK*n|ZyB+jc
z;yOSHWR1?(7s864IrXSx%JmGa#)<#^fWa{@S{H;THH-nigjEaIh~fh@i09IHDrp=(
z{$1vs@%k{~?@*5xZYG8ra&u!rw6nO&Z~oc%9S;|)i;Hu5D!Qzx{Y0VBNnu9}j4iN3
z1&7Njs4A-0S*`CybuHRdG49&ClNqFJBI43C+H7Y41ODwx3ZS9Lh)$N5QzPrtE7feX
zvY_v5vtDQfaqwM3Y@c|E7lx<Z7Prp9(2x(Vx7K*vc`kT*>XbfI#W}d;J8@{wj}LEG
z0^zD)8q~4Oct{a-0mobbiGYwMu!p7MyAFs+rm-JusHy^lnuKzox~6`5!u+t8kx{{A
zHvaj_3Gz){{Cq14#`T1cuA(rNTc0LP<o5jS?}V8BbNd~Kd@WwpZ9ICsHy`eO7l`PD
zw66y)$4s$CkNGx((sIzXKSE>{x$4d|NF)uD8>QQ8R!wG&h!8kK+!P>{s#VJ9ai{iL
zZw)g`?f)G`f_EhnE24-voralmt6?pO!jE}j0?8y>WPPJfnK%`s<o<yJGS6uSlQ8BL
zb(7!NeszAQscK*w#~gcuojI~#!<Y2!C3aJBpvWQPNl*9B`ez+9**`h&DKoq;=eKG~
znwv~gAHImDq=^b@1D9q8%gb=1D=>9p@+Hq<TfuaQ6~+)AATi_pKpyJ-Fk)EHVQ4P6
zfV@Jf30JnZe~PVxAG|f7ibD^|nERbKg$&BGNu}?33LwX5)io>LJn^GHb18V;U!s|3
zmx^Qwm_={Z<N?8FdIAOrRt81~XelXUqT&VgUl!M-D}Y@ONE<uCQ7n*6@*S9RTDUU#
zEY<|qrKd^V&0(`v8}aQEfRsL*-CY)>qD3n%r@Lo{JL026nrP!mIv$A_(X$#y3pz>;
znc8+-T!cKg)O%mtP*YM0M#zA`F0d4JcwW;{lE=vuRA|;YUHq~SYG4^ZrT<JdXi(Qw
zpM-rdBqq+#BLp~-MLF<*sHdgaMK~_&x?Sx*rrU~wj6zLC1?9^^ognumHmET1;UU0K
z=8;afrDyZJNsBI6D4EMKIW_F61y!f&Ts#En0>nBxU{Aq>hIg%xrm}lpo^%0~;`kW?
zTw1%q{txpQPFB=xgAKr@2`l0m4S$7wQo@AC#cRg544rZT{~10q<9erDRCakRgY)7=
zFmQw-l$_W0##V8OI;wATqEHFaKQa^<EQ!rzF<+N*+Z&Ga;XU?`)@pScd$2T1?sEQ6
zS7}frhJJr`@2T%tbV5uw+U>G=`1-{Ygh(1}e&nj+riuSmiH+1!E+f$TbFz#5G#~(O
zZ@@BwoS4BRHlS{N{!{dCV$6=FA{<S<%loH}kwEZIS*5w3w5qWCoHoJ^fkt}_-?eky
zUWfMuuuTryI7KMV?w>$oPDTO;tOKBI9EPhXF9#TU^Z*%QMr0}%NPDQJdKFKTAAZ3s
zQDe?Z<#UE@0aoQwLcb=&F>}#=je*99f%VAl+YPj6PRf6kWZg*y2Avp7lao7D8|Sl^
z%L84Y>7;~Y{JZFU17l@DCyUi!&gE`=W@b(@DAd&_^fH(UFlU{4yu$i6y)JIryvE8r
z_%p`O`himkeKOyW{a8JC55VW_Ew!2VTX0cs_WUidh!&*0vo7Ow(vzz$AlYCQq^$@W
z8+&_a6&EvJunU49Pcsw@5nx0MYS{s_UJ;SL2z*{La#|So`)3V&4ttmVZ1-L52oR7f
zyXp<7@x9;Q+~gD$CG~9*Cp`M~25>dlnAzBjjt+G3YC?UJQ21y+h%i5{CmtjzDURrx
z(rbg!4#-c5iDqTL+1Yw;&4D&^smWh}*wkoyypRB$BBRgJ+`PwxgEu2`7S)rG2qG6N
z=Y;RRNveV7Aw(JoB=%%v-$8O{R&&0fL}`0FlsHxu8QGOn-D15l1oF?)sNzTD%hV^*
zo(HxEzE^JGF$+G;oLtV)UDM(yYh1kE2f3%3`bYB@f$H{|RqNk2QBWg*F-!Bq1L)7(
zDtw3GR~80KVBH!amMLS}jjP)cDuDr)QxNS#c;BWfjdrEMKJ?Wt=&Do>0XtXKd)MWf
zvJ&o6%{rw!S#(7lVp>urwhFz(wzS|VsFw2b!a~@{zM;XJe{WzQ0gMlXitT7LT8aXj
zEqzeFK@{bTI?aP_<^`L`jDmNVh{Zw$=>4!4Iwi**W%al1i({y#a<tkuwPsBdBv(s1
zJ)=6?kA9L{{4#e<aFZ~VeD0pBDN?e2-|Qw=gJrTakon*R4F0U><t|o6S#(wMRY}`^
z^mDQ(tT~HYWm@`i%9}inX}QxGRF8Q7`<U?Rr%1Lojx%P9u%On8tqY#9?xNgI%x=Zd
zq>j~BREWz1(fiRXI>{yH?xNK>LO9Z&P-;eK`q+3l*tes0a9yXCs!q-!5_w2V`Cdyg
zq9_fgRI!td6Qu3V+#nPV;7CMV4_sQsmwqdq2?SiN6>LdZXc#LOkI6)3JS2YOGiD$g
zX?Y8gfF|<tFVFnxUsjc$)}`NL+U@v(&3+Rw`#<X6-wdo=3kc-uHv`81iZLiYn4$bI
zw2OvGFpzwUWuAIXtD}QhD#bj-K9Tptq09IlA08EjrQAX3a5NTk6S&@|Ph<Hd^)uX-
zx&OMabf7cXAy#eM96Owsx~aL7oBa^fw{oBLk1qSI4FU(6qQd;fN>R8VqAA)D$B}qt
ztFyFK=CF_YIYu~N3rO!uhsGm>`M>eoi3{#qZ00>oK5q_wbc1fE(Z)qQ|7X|u=Ai~_
zZ!Cc5*N;|wOb1QuVk!wuwimUE)uuW}PEpi4%A(-lV8Z1k2I04^n_!D}iTf1wEs@Rg
zp;g_wdc1SZ5TeIAjy;i_R{8y&WSwuWVjn@ipg7y<#z+)3mCN#B`KBAk=>aZ_+8>?(
zB@mCnN-XkWJMCMJJXf10X!H^Ep`+pP)tiR)_&|Gy<OIt~Jy_dlzVBNNG#%_8mJydE
zV#_ytFsV5#Uull?M?&j8Y#;*RoNl}AOYb!gwIc$n4r?4P;>%t}J<0yP1vW>g3Fp$%
zBiN4gtxo*nUL`YwCUe9u2!4Rt>mQ#$=Afw%WYY~j?SpmH%nAByX7G=wW6y{gml?1v
z<#g0suArU1mR?+5&f$EMJ=7%0xtI(Rr0kb#u5Zgcyq^#uVsU49Xc#WqUIii6z`EDk
z{8?QhpDoX14v@ux5&x|SDJBf&$B)1o<m2aQz+>ysYz`SL<_K+PBJS-a2Z0>O4Go~W
z4jw4-883Xau{k^_6d2rPrm8E}X3})y7bA(PL{P!Ni0hL{<Fj8>mKF85kG|(>u->B@
z#e&NXq9Z2eUG_c$``-?&`p@N@UkRyX9}$H)ed(6v-e}hnBnd5oUN<=L^@rPoFNA~P
z-1vU>=EtmUF9D?0?RxfjmBl`P9z-4yHS?T$Nj^IlmKO8cKXew=U^zS2#?vOZm_GJe
zx1but{qzQFDzu+3dubWooZBOuR+~E(<aah^bAEff4*TlYXm41Ei?+7B{rPfFiANF<
zVdUKOG#NfChNQ(v2zk;1M_@!Mw6EXi03MOleT(s`!-H>T#bm(62%109;E@;sl?2H@
zOos%#2Ko&8M2Fc9bT@7p@ha1{dEO^(u<v3cKg1&PoMya+%4ukL9~=qe#&KE?{sIZy
z@w2lGIRVbLo6khBP_m7L9&=o$p`yK-c)~*^3=QIL8&o-Yd>I$7NN)Jvp{FN<`4$~d
z5R!;H%05wXA^qz5uCapUK;El$TR}DZ?i>0?3~UF<cUauA*AU-#`6VwY9$#NA?A%)c
zc^h=e<cnS-d{g)^0vqe%H09O(U!0@i4+&9MmMOvxzruwEh1r{#E)N)8fU{pz3@zl&
z8!>mH5Db~}(&An+w+<1JLo;);x+Q$D8+9*Mk68W_RlO=Q-;dw1CZ6K7cV63;09hC_
zGfFHh)8iRAqR8PvM_(DR4NSwm)zlL~>tklx#Wd)gh2;BmP*(q9h;JXgIy?NErNh;9
zuyrwDJMFQ!aT8+pSBiRuhs{pRqSFx+t3eZX*smacfCJcmZoYA=`>VXh0emSD*LsUJ
zItE&wQHXv^j-d!jXh_6n-RbJ<m*1S|O|DDD14ZWH;i(fji!L))@K-6k#sZP9ICv=E
zSl<{JWIsG#Z!FE2b}T`W<`^31=2cFD+&u)>dR1FKWIO{lgZbYH6&@}^AweNHHhkD?
zbdHARHUXu`?9$R8<Inwl|G*$N4$8N=-ve9WhbjOc;?2m2jedC9G({h)!$T%8urxJ9
zga9DxgwS^0*_j5>9x!E-$+~-lg@qvBWJM*Fg%NqOOE3=(-iVOI?ESBm0VIg2Cjh1X
z!)*A01ww+V5*)05xo(Y?Li9EH*W9TEODyf3k(0krcRp`F1~-~NQ{l!9?q0?yPJ(=-
z=#RMQHv)?MUS5JOHiVF`Nd|9(g+WlFTyGd|KpZDG$)a~93vPiC<NeLdrP(em36%;+
z#_7;mLJ4@*!=qU^1t~qhL`Ar2Pf-htsNmh&<W&`gufx<>K}NFuvYA;z>0xnrv&GcG
zK{Bx6vO0CT>d(5qeg`<{d$NR)eS6ndKfC<$nkXGG<duFaD$C1A$$)LO-^_yx&uSLr
zU^mz2i_lL!?E*G>|83_%7AhYq*vASY2%oa%Hu#7@wM-F-KRz`hqVNg8fy;H*w;8fG
zuSClfYis6JMOk&{&b^OO7J_Z2?NtiQulGi;s=xCk{YNj;%h-6T$v(G|IfS2AXEtSg
zM{6xY_1f#=g9oRD4yK#xki^NUSS$zRg~Ob&)9SnHlN+x)`P%wgKxol_hz;)GByH=d
zu<r{AivSIt-r4&E+vO&wo4cNxK9J2pJ0hKvy*X|}3#qNG|L%R2&f~ZU0GX;KptC*U
z^>n!D)_=`>Pbn^LOy2V>V;wc)J-xF54zwVZjNQQw_2b>Y%lm#F)Mhp6oo%e;=EuD&
z`vQ8Xzf^-E7Y@~yzYG%OC}VR1Z&c={!ED-U9P&SzsR1GK=DcSA!X-_CejyC`3OI8S
z@wrbJ2DEuTv)QWgSOpP;i@ZZZarF4Fz~-*Cg$KhX2~?<{7=-*OVM|~2hJw8RfKg+2
zGE9vnL#+BE?yw&6r&{<D`P~DIZ*A=s5Jyvl6Devd9_~4DlQKQ7b`|W(aDpr<F-dVM
z%6es*$g$n?yIUI_WCWx<x&HrJp?}PMpx4Uglam)grV-I%$JDK4W@%V$M7)TtK)17h
zYHDlh@)OtGC>AD$q?j1K6R%&lR!8Wp%1IQOsf*PL*ihB}?4~3or#iPSZjypvt~Asm
znX!IGJTrV8g3zX2&ju&3MeNe)@EtW$fBF(j+__hfv+Z=m>@s^9b9>Qq2|x!=PngP|
z6sZItHa$Bjg7etY+WOri?)CWbz2fa0Yy6HXU8<CXOiexh)-KU6*j(`Xatt<P2ssje
z$I_5f;$UU``r;cOkI9&Hb#}82LY6txOaS;wR^Gzz?%u;t;00i^qVjT9R%S5?89`(v
zHzOxw<(Z_IKu5<*At50Uci}O;^+Jb>iwl|t3&65MNGa?`YwA)>LJC`HI+NpK4#i0!
zTubwL@M8&H!YB|no@o|+(k3{ry&{zQ{XQSycK`${EDU0gxj(xxhU~Rvg?!b}Y;xm1
zc`1_w@nuf9Ba@w1wZHd8wY}Pgs(u1u327i|)9o1|(x>B>i{s`v$?L~kMTJ<O+x5#)
zT0GWrP50UBgTOIf6-<*`8?7*~QjN<t6gHv{oTH;-YB$;6Zpn+j{*klKPs|Tk<ll{1
z**z#dtP3@bXiDc$QBpE6wD#U#v_=mbc%E6J#Zws)lcek)YJ1%?HrKEBxt%541PbRq
z?a>T77bO)XtL^$lbxlGJcTC5)kce>2X1me4P&z9Y9!UYuf#MUDI5&J=3bt3!qX2{z
zR+XKM)+azmhx~hnrw1+al!P&1A>qRVv%1>a_Q$h(*{#dDAAOqxVpZO}X}niKGA8q{
zwtCha2uq#^f~tB7iK6gC?3{OhDn`W;#-qo(pK>3@19_kD{QB0h4#w0A`F3|>4!Y=J
z!$k;qozBJzE*qfOoc831h%#0|pw@kJ#<cJF)WqQ6fS}3iq(XFbbm`TV)RrFjk>SX2
zUwecuk&xsIzanRokc5O>fR?}5cif+;rD5q2m<2@~RG%?pUNfi6z&zz6(;JBXjcsJ2
zpskHY_#v+(CoVzN_v+Kuw$U5tzpjSd)cr&IX6BWLhlU1dUrBb#9tC&NOy7ovuuLlf
z-=b#XH@_xyWmQEqs!LlLDf4Lm5T?l}Dkyz}u7zZl-n~?X{XZ8VcYJzUOGiF<C7Umk
zNKvY`K0YlzYG!g$faC)U3n!OeLW1h2V?xO9=;-N*Ni%EfkrC=|=VP(x7@S2uj!jyn
z_S@S=&UGvtAh2|zsxwqy8N=9Fd0T~!?LUfME8f3zHvr3krF(h<+bB0rjO^ni(lM+#
z<p9*P6COuuVWQiWS$<+-BRdlty{?8tv=|v?LM#<kzazUePx1P=IkVu;7ig&1*m!_u
zfGP^f{|Ct@65Kq+$B>X9Y`lGNSikEws{ftc=UU~kFQAM;UXVivx?vpB(%^EiSpbJX
z@Xux_kW#ftN<`<a=~gu5v2e1ME1IgPDy};;b$W@2YNEaNeTl5Av&#J&@!%Uv2K+FI
zXe~bgmxbUc;v}gDTpZF-bQt{d+BPTL15nh5fS#c^rTNO?(I<eDgSw22q$;<RJ8wk-
znrn3Xf`lXoI5x85%NQ%yW&iM0R8$1NX->9qcNg{&{Fz?vqM{N}Scm{JnrW!w6JoX<
znzk|GKxXj_3yYNt0aPR#6AST2L~!+>`2NnR0n&zeNZ&fwG&et-)>kyLPi+pY>^sIQ
zDXBPf<`v{9D-NA}Ch-uZ7892cf6+gu(fY<iPzGFbS2f=hmlPFJQP<Vhe=f*ZZQx4N
zZrQsyud&s$O9WRTZ;wcY^&>dSF9?AYL|w&?fCutH|LfPcG|nu_%Emn!F4|IBKfX>;
zkkZcR$`&-`ku!3ZPsu&dA-J_hMYSXRj7re`gs)JRkseVSYb0Uu9i&fZ|40q1b<*-A
z@fY}z{uQ)2H8nLSXtD}Q($ZoA(ft8mfrNx4(G2AP&W)S*Gi#Qj6Rmjsukr}Ygds4j
z_IN*rgQHWz(6(JNTr#$bvU1<s9)5U;HSIfL_ojp7MV;RH1D@`lq*ER4gPkAwJ3AgW
z6&_)-3p}8@F_FP}>=?Pv_>*!=6q&{u=?wx|jYfyfJ2*wX_5_Lko5WXHsIRnIjh<b{
z3zAolTOUOAITp*$8XTBkJI-gH*BKHn!D0yZWv{!T|HhItOI+Gxs<*Sqi{PiIpun5c
z(!#>0UFtY3@Dvvw-y$ik1^VE=5JvQ7c0`2x4~*rW@%`ls%;DGgXLiajB{jV`sikJ~
z>Jys0xCl(-Kn<yRSaA_uQeg|2P1y=xzMv1eg1)YjjyxqjH7NzfBbM<dMO1YB=sGiK
ziIAes@?Jr$(;qYgRZ^U?GSY>En%>_I=midUc?NqR@d0}2{>+YmW?UC4I^tl!P(Lkr
z6nxltbzx>^&CfGc=-={*9niP(B)N&6nV5VE<SH>S-kq&p080x`fZ9VvR%0&u5<Z40
zBdC};wdHhl<mHrxqu*mitUauCXZ$mJd6QM}B7ApvZT)yA`<Fh}<ptGX>x`I-)fVhh
z&h7R)s=Ilrdnt%7L*U_rT<`=+Jj<^42iE{55)?4OHMg+fb-v1s^IwoHc<}1(B|mX-
zV#PD_Jc+#=E4a}%+TAHzs#TO$z6Xwh+S-8klmgV5^t#-(6Hi1a_vak~GB!FQCN8S>
z3^J)M($YhXb`OPisii(B?II!|g4R&*IYSoQI3kk(S<C>)l5HzM@aljq{z9$K>D765
zaXmLFr>LmBZ>;c&`CpXB{Sezw!muv0?d<E<g+AF8-Hyjbo0-W4ru-LSj@Pq=-LtBx
zew*Z_MQ|{JUH_xZbwg-SbC}N8O#UjHC;f`FwT)E{u2)=sec<`9gOi?MzS@uvv-4t6
zI05SU>etwR6HG+~)bURTR`u%-$B!?5#K2MGh~b6_at9v(YWB?Wx!2S4mz=q>7V8up
z!N7S!m*ts#=nmh1HRY;(5xDKHyF0pR3qC{-?8&b<-YG4$Z0QJIcRs$Fw}cyZe0dEk
zfkd6Ar}X@P*ZIh#Ufx#e580qow4=UCpSgK*;1p%Z?!LYQoQXA`E0UN2z}*73zkcE7
zLm6sbxc0hCY05Csp_NqkC;);ZS4VpV_+L~5Zkun<%*EV<RPje<*XEcV*wya*eZb*y
z)vt?!W_aX<0)GT--ks@{=t#yh)&3v_^x(V~oWX%{Y`I4LnPx;jfa50i?|Pl7UtjH+
zfk$*hP1RGK_U+jE-_2cUKfTADYY+hU0fDp6gy*{*YC@kN!~2IbOZ(ksYF|B&4DBzx
zCJ)zWcQ0aSxCy4tKF|D7GmuBnByw%?bxu678gBr!!s`iIzI)P^VSZLZTRLy-Wp6Ai
zGUq-6D(BbZ*(+h<i#*W_>yst|E@Y34bvp1Bw@?YGj)(i+2VaYj{tG)YZR9?)g^dos
zjSdt<O(QWY5*4L@@(WpMYqkQIq@#-pQl;T)fK53xLM;c*4=M_0=WoAM6vqC3PsCJB
z`=I0K=%{C>X0G0pOExsviJW5yl<t6@GH$!xe*)&AzHUyTQgKhFU)K5u5P7$#feUt@
zszwuYoFw~maWHK`jU?vpRavGSnb);x@5#G1##dO|+Sx6pD=UgeAS)(OWHmQz@w`d(
zAr0iZZJ)b=3LTtE+T?Xmkx>9@$~ub)DpU}34mf?w!?2DU{Ug+OxHu!*4&lhxkNfr8
zS*@)k3>++9$WlQU7!d&uf{17jkj_0TyFFDJLvEcwO<iygW^O=5RnN>OIo2N%5)7_H
z@W=}6$B=KI-jfyN0)O<)b9U-beu)qy$oTj?_6MAtlCq@4R3r#!5d%P=q_h~9$%6gj
z%lhR#h`<BKFQbnGvy0G|j={m<6~=Q(*m#u-*Zjo#$edpDa$KC@z!%3HN51NurS1r@
z*<#(L-?3aZW`LBa4s&yh?jDL}k5&-B(tDQW5>2CGXXa)}J-2N2DO^PO`Ii{kqFNHn
zHz%uBA3X$!k@0S}6*lr-pMf~!8z_A<eBMI<oSvfyAW|t4&2N(co$$>ZKS;(E7mqKi
zd-GEU-ghTHe!LPvsbT%Y1WuBAGOopr35-YBzm&vO1$p_Ij*Rle!hTWuj)#ZuV6PiS
zE~}3xz{c+_FHuyAZKz_#I+3ZXI|APmH1#oz$(a-l_2DxHi;MsKI}<?>6_gfcNA)>s
z)+$AX5A=TXKQw)1RF!Mj?t?*#h=59$Af3{sAfU9Av<OH^cOyzkgLH?4AYIaelypl=
zNOw19y5H~Y!H+$L3s&7R=T#Gt#^PiOFDg`owQzcUzi%}U(j3>#3~Z&>ygaHw?xu$U
zD%i7Aw8z}sI2Gla?y+*<vV6=zOUDx`N`cMsSW!`ReJ6VPw-a|w0mLOEAZIeg*EW~s
z<5LI!V*b;{rAq5OP|krQ1cWiryDp;xzQ#{Wwg$>V5q@rAX6nGFv}0qh4mgtO{RS8I
z9;RkxML$h@6`w#)|4To+O3MOdn2>CW`aw_Rw??ewqRn@Z8o{)JH|googjr2y8n7YQ
zc+1jDUi{k-t$0%)n+5)U+o?7w*ic66sm%+GY#iK_*u;qVg4{I7Hz)uBI^^_%H1@>{
zOmdO4-IH_0=8a;E2gvR&Rx@>al~zIc52UN@uIv!#S|TFw;4wDeCn+69e&2^8ZuX8M
zh?YUW0aJ3-b2CU4%h7uW(BTEq=dM`q1i+uoxdS=`&}_hY78#*r!WI8rJ>jl3Knw<U
zOkl+=uUNLmZFxxZCCd;&mkGjVSLNp^q@d%GmsL@h&^T>2N{Nn+rmRqn_0$#9&zOlx
z1)NJy?49d_w*uV4?66UKmzEX@QoD|}vKg5lJ|Mdz+c*z$E(~3gyD^QkTxsto`}*$V
zl4xjaQ&aYol~=6O+#e?ZETvx8dv$FMUQSZEPnq3BPnhRxe^H8}lKbMck<gw|r@!ed
zyN+Bwp}bb#I5JK}#Q<YCBpVu%pPWgfAVhqLf5Nsw^YNU}to^mwyCRTdKv}zDYWCv!
zbC3O9!_qZ;f`xuYIF*!TWcF)njiSTY)+!&uJp^inF?)O1nu~8~@n5vjDXFN~zm{?m
z6Z;VV^%ERDQt_qfB@FEJ&oo}VRFmxN9EFVSUtL45B=bI?GZyAxV`9u(#dAF#0?sWp
zwd2(c4o0A{g~>{GhM_^u6)R@^WhXBY{eKm?BHd*IWe+|l=i{B-liiJ67>N7PS8`-{
z_$U_rbdnM;z~24<zCKS+`pb=?(&Da8)}P8!t{=aCD*04Gb?UPDU!mp6guif_Zc&+5
z{>8phKcQJsmYMkM;`GEFGcYPX8ZNHQQ(siPfZyZ4z~L%+d*g}RIF!?%tPLWvm}#E{
zS%CD@Yt~%Q9GUTK<mcwXqqJDm66qIh&T;1?y~P@mYn*NgJ^ekJTTri8nZfzSnM}ky
zFr`Yr(>DMo1Ep~<<7jr?{QV&yP=~XTw*$oY&yK&g4M5ud)LGJX?}_Ob0F#IY*7;fV
z_K%Yx{3raRBJQ)GXTrw>8me^69uAC{p5*17z;n3CVc$(~f2*^(V0pQsQ|A`$TOBV-
z!Mo;i05t_RDx~#;<zK(4N>QNV9jK=<I&xialRT-+FeNuIbvrcB^x;k6{SeTm%j>S3
zJ~bL3N<(HPJT^XF?RwTa&vGx!rlB?wYG!3=={1YCX4sc=b6q1NBcMwSYf>~cJg|^7
zg0mc|KKp}%@P2EL%O~aU9PDK`(w$3*h=_T%ey4=RGIc~4zgtMOt*h4WMD;7CM)}j4
z<kw%k^AB!!J;YPoCKo=_;&sROYlg`Uo8Ok`5dQ#a_(8l5@?yGA&40-sO6ZHZOWT&F
zeK4OCnuhIsdMfdpaIH@<2D*-=P>X)!Aq|Hqcr751yIq)qH2yWKytA|4F|id{Nsc56
zZyNlFK)e8<q2}I(zyDb!{{Zk>%Ng68FJF9QsGtKG$QdflPCPhU11s(4W8mQ&$%UV&
zkMEwX=0>(UrkIOd{Q>@WsrQ7Li-|*s6AO`%4eA@{85L>oU*Sch;AoGIHp$7;;q%0N
z?s|;ovo7ES;vZ1hk%eZb6}U|QeYCsna&Xw(!C4?q_2j8mZGLUjYte>17Re3+JVemx
z9Q*87lUtI%_#8-+6_}*YesCiSI&coP3F8GEpPpPF64{z`%uvkbBKr2V8ZgV?!OJAj
zw|D+6(5ie4FDSH4x5imlOuL+?`{&XgNT27(?^7`_G|h7DIMrTvt*|%5@||4`u8G|~
zU)eKHw+Gb+5glFUUz_4MG(<u3x~P)U9(;iONr#&)`pV}=<h3wy4QiTSt-opu>OJO1
zgdo%PWki1UsV{u_hZ~cEHdXFNW9`3sri$zx?d^lBcuUjLX3ty9h(`;>WwkU^>mBAp
z<|1CIKUn$obrm%0$^0&TKiiSvrSaei5Qq0AN_=@4LNaN_ZcKsDWM^H~+dGPhQLjSy
z`%dhju(T~FBz9nyJN%6ZRI^OWW(9rQO8R$X8!QO;DfppnM;wA$ZclfAP}4!30eS$)
z*mf421}_}hOdTyPEdwrkk9O@LJo;U-^?d;wNKLfnO{~;y{8kf68%$12a5s=}0oY*=
zx1+DTksV8WknBxzpTu`OymyKGcj>pb9DeAw9$CDzeooKu=eQUQrDw32_{oH!h1L=N
z9@t(2^fJp5i@GKb$m`aHyCaRXgr=Q1?(dJNT1ZQGLec4wz>k5PU;j`zSta1Ho&!KY
z+5PX%M`;sNW=zpH{07UMT6{qe3bW<*HpTeHbof|js`e-t_;C=oh8w3I7&p8C1AQ@}
zG39yAMMrlPe0F^{FtjG|o}NE21}&|0RK2W2Kk+eJ{=U*|TShJ32>`2ZzNK+k7_L$2
z8n|bmW`C<(2jdZ-YTMEkrj2}V_?G~5=7xrux>Zvs2tX0R!6dLFMcj6u+~EuFe}KT?
zns?*%mBr!Ibc{+Gsy)?I9>c>o!J_36T<fqm9b^(25upS^b3{~4GOx`XyH{S=yk7$a
z&ZwJy9#b1m_v`k*YZb(Q-F1$k_KyM4;T||DFm4D+KZU8nwJFIxEu3wS@Ly=SELun|
zJ;`e2B_Y$a(`|4f%4Mo^Ih!3C3v?4s@oL*Uc7tOjJ5(S|&tvIw9+*IdTzu3d9d#9R
zBXj44zGOK{^~>U8!|#Ck{I0OxV0A8wh?uL_o(XsdQ_n%=G)*ir3<yq&NG+f3>B5f+
zRcH=QE4dD5um#CP97pyGdJ0}~c!uR#XMgx${;fJ9ipapU_D#pRu6EYHKHPm72e#?+
z*_kEk?R&r|Ld2^!zvDG1_A-TpCi_G%tgYXR{&6~Xau@qYGaq|wTL~8>bpECCMi3C}
zqx_M6Eu}0o3ICKRX!E))`rwglwtlvmsi_)^_VQ7`{(LpT<MMYgrRLHl&+oRnxKwRS
z^^a|ln&%FWlrS<d>caZck%7@THxZ!DkJ5;dcUqMLS9@j)Uw#u$5dI5@S5yddopms(
zsv|`N?Oa1bwx+&AXqoxkF?&)0D{G<C#o-6v>aW>t^GB@Y05w3a+5xT21H3nao-4l5
zS*PJ2T;+HxJ8q%+d=&9Kkr2~9=-xvsOqn8oYk9PIL1(c0+RV`7YVjF@OK|6j{ifUN
z+tGjKYt-P)*B{tGdNjaH*WUjb8k0W!U@pfiKxRAcml=~%?fUWKGwh0Tmy7kwPCCc4
zl@+ae4?3PgFVV*9qM!zvA}&C1$?p8UJmwCU3JM88^U`!A2g9Mf-TjNzv5TcpOL_)|
zM1I%euV0A}7#N5+dF(m}*{=WcQ34McKR1~*74hh|$)dM>2zF_uSX>H6$Lx};KHxwx
zg(4a?0K+n1KHxn^*+|GYk|Rmx@?iQW+s86&9?A^vs<u1)2}(BPHMEM$Ls0y}HDPl^
z%0DtNU}HRkGZRV%c=aOJV^TX|dR%n>8&vR3jxOCJth1^qHNIp*s-PKV)x63*Thrin
zJvs~qUPkn2WO$SuK>g;~&6(Q$xt$I3gb~0obZYF${}oi%{p6vpETy}B=DvSl$mSJz
z0^oo`QmZoef39G05gm**Yl7EDyWUC1?agimow>Fnc*9`Ov472a&TSHl$=Gn@&)>Ds
z5T`R>D1eV*SM2|N{8Qj)YQexzN2uZI^yC6#UgG2g&3$+G*#)G<ODjIv3_kl%nvIh&
ztR3nfEFTmgtpGivlg(&{b+#x)mGhBasN;7MfrXV>T9%$0Uc@oI=@Jjv?a-014ugFo
zI`enX32BplYn6~&9!0mi!ESxGHTaqr<LFjqrr|laeR|uVt33H+SX8+1rp<fpQL(;W
z7_FqDHo87tzcUpmxfyCnOoR;$;D<~G`NVn{zFX~aP;$@0aZ{&3lqwvDl+VY`7RI`0
zy;9~d-*p1>SPz;5&*R7B0`?(o^DyB@GFu<C(g`DlyC>TNCRaTpsw}4J#wjCI8Tu(g
zwE+eo^z(qjy-I8I0<_wH`szu^bwpOapkF6yhbRyDy+gOZS%9NvMPK}!x{g3O2Etek
z?QiDE4=gN?diFTedlbp0J<n?aLOW`!1hG6c_27uS1`VCtJGb;tB@u0rHE2f`OM54k
zn?kGX#7eY$RU`uF2!N|O!$2-zAagQ{{%K+$pjy>Hytv2d_=8r2f57nU8CEbO&-Ic`
z);cZT{}<ut-#M=r2l$OqhxZuk<YvwzMs6-Hm)e6HUcJJ)VCyM={8+fw#l~vu5XN#&
z56u>forR(O1=*eA;+>eUX<4c&+EeYb%Bm`L?pL{G`H7wL1cZcUra8M#wIJiSiI0cD
z7|Un<`bQPPFiQ*#CuN!4^he|p*}<(_pz5u+hXCD6ds*3oB8#q@<IIwMdEe6Nf1K5w
zfkh7G&q$plGxIk$Vb7T#89)bf(&5yI)Nk0VD?t4QDuI4&V@xPn#PqB?4!hUn$@q#)
z$I{8n^jDhrV1)u1xo%hc{rgz|3iQTDL;W8-$c0E+L%KS;8=RN!AfIE@NQ9h2p3;hZ
z_nh~cNCFGp!|co{*L{4md;rxiE?vmvw<tw;_j{P>nFX9%3H+L;>RqNM_I$pNHNsp#
z{Bmk`zQ3m%XdDAC3zMfk57sX~2jHOmNZS&2+mtoCyxX_5wr})@%|TgQdA-M7&YkMV
zlql$;pwj(qWJcGedrInSak_vJnq2KUTD1tpF!+j3h7DK$YXK}My@c6`umJ>zZsy|;
z1X5;L-|?j+#H8pdh-bXama~CcS!_Mu(9jU(DgoaJtqF{uLO?jyny-xueV@|HL6p~v
z`_Cb~8RUx^@kDk+vJ|*~tavfwWGEvW$~PqKkd$8C2A4RRz{k!_zoD4VGc!e@5TN5}
zNcESY1<zo#C`DH?drRk%G%Ftw$l-@eo*8hMtS{dIqB7PHbJv_QlD{p4GD7MGjCD`l
zUhQ_xqLr9$3tlzMWG<`CR-^}(7$Pr!TN?yS%FEMHvfo7Z?aJKFp&e^7&+h`e>IZmd
zBxH128|dxYb2TmEDNU=U3}}b%?p9zNVF40j&Kw;3`v$TM0C-{qA=w=hBXbPIg@4@G
zIX~mu^9)53E>p&sz`qJ&+63mTM}xzlB=-*rA?2GS6EX%>Z#BOL!j{^g1B2GgnTcUU
zFvBh(O}Ny|+e`WR%M7DszuLt0p^o_g?J^EkQ-j9{qKx@PiCbL!J`L3!1W>8%TsX|C
zt6lii<Ma+45Vn53eJI?441qS?a3b5MoRP5emy~=m+h&h@CBW1=6&}Djrd{;>gDu*`
zo>Bu#s1d&k@Du!C?(V`J!OP_~cp7y_ksY=Jv`HUiWFZS})$kw6XT*Vm#C0qin=Jl?
zO;13<!S!f?vE<Nnt$OC85RuU2<-oA!6Uy0O5}YInvw4ozkk<GrPm!t5pY4#o{*i0P
z`1q&xT{*MDy4O$|J&L~ZM$-2f#3z7+{%l;|8LVK_4Q^2NhAg5XkLc;wc32FAKDN!G
zh$?@ioP1<Vcnu+`FAZ%5`Ps`sj#{^Kw24=L7YRVLA@#hZ3oY)ctDl;h(i3epN$F-*
z3SL?ky`o?OWN2t4Iz}yB{1HCpVz5ueW=ezG;N)cbjPC6*9~4nB{rORUV?;qAEHLQP
zO~hU$3bKH^@=3s))VD2#fQ^Uay(kDwHe43(N4)@p2DE9wonoW9@BV}~$6^^*6L_xO
zPrtzb4gD2WINXOZ%*llmtURDa{2n+8A8#;;oQBI=Ac%mUh5P5)dp7gkkK6vsWQ@3f
zmBi}0gDsk>`Z1o<1I!rBn!I%n7v{w7mrX1_&^EUO?6}WLfLNN#zY8@UgdbS{5>xAs
zmQA4F{?|YK`@mFn!XeQ1>$6vPzcMPc8}KLLfN>^mFg1bOxps-pvR8f@e4DI{g>aL7
zHhBK(z~EtK*4N$5zZsuC{Q|5)5Da0ntE+2#Bq2He=q{|W_gAPHzlAq%SWY|pXIam%
zq9ANG$GE5;z^1^Jn!=JE>bm~XcwOO<!EBdVP97qvsQBF2>F3`yN5b9|A(sUOvXsKK
zg6N27s5F^DfaNe`g@CFzc5OI;AW8*V?J36b)O$bbBzu4TV!VaAU*oZ3$HOA{8^f-<
zYbGm9DiZ@I12XOu6&2$m6JQKBJSz6h8`!@b4AD{)MX&JgC>F!Y#K*ju;GKATeY5Vj
zNI7Xzh-Ax89vfcZ*dFDUGe+|UJ)j3d&&%r%+k4T}+*}<61!V<=kD2wbuVYw~Ly4pl
z=_{ZjfiJ^i-@oxs(W(YFa>LI9zTtm5`{0e~Lsx5)xX(0LNd0kT9hK4`2ZCth;^PVH
zO)l$1&yCLY+H$R_vIYiA#|r7Mf<zm*EA<9h=Y!V4y+|IUy|aTwlS@rqy||#@^6%PF
zEOie0hhl?S5-1Du*Ap`rM2aGSAdtJy^FsV%cw)rmpPf=zMItdE@a}-fZPT1QZf=`t
z7mr;Fql;YSUw{R`#OkGc_`bf+&}TRd)0^2`Cjb={xIO72Gc)<_4~W!2hGb@8A^cCY
z@i|(pN~azLF~Bov-S21*90r&3#oq+L$WClMIy+hMLE!t6-nGJ#-~R9=bAtK1oU4hU
z0JtWcPj23xoGi7q=?7&v3wEf#>4FsuPX3uXs0*e`syB8%AV>mz8<>Phfwxvw1(JUQ
zI#jz~(>*TU&Bls~;h%?sK<phAr3R0AJhbI_aY!Dtsmi2{OF(+>4#D6swXyPQ{!G>@
zY@+u0Js@;IdV&+z!K}^2BTT6-t@IB@mzCWZPB>DaN1f2<j$>DO@e+y{Iiwxr#`pI@
zU&qlRj`AlXhINFeon`<m+WTnB%&`J@F|uoPzN>4f;vnEo198|*FAWD$i8ZRA``iYT
z%Xg@#_J3v$!Anz7RmqrlsUF8#f?+Ey3=9mGQ?8#QCp1I9{@1n~KY$x3>y<9ed@`)T
zn+VLZ{x%GI9eDGyGVBNH6l=<uJ{%51MIzHG7zh^w!S}*QsZ28SrYHVEub)eap5ZAs
z-a|y84@de|xE9vE-i`0nMShG4d!1e*MMH&byN;KJ;~H5^m2IvD3&<zC3Oa8-p$(V7
zCN8m_`b9S1lV1L}ik(`(eM?nY6{S(2w|7fWfE(3zoQLmnk_?Z`?>Zu$h(RvN@sMZh
z{6{(^9gnAf(k%q`W0{88S%WqfGG#w<FiyVttJBY66=iTr^sy-JOZ@+@1qj$6yCNWd
zofpHpnw>q^_j}ppWD%^cM}S7W9kQCmsz2&uDcza+9TF8aj5Hd2N*kvRmjV!Ma7bB2
z``d2|9Xk4sYhDh;$;p90Ng|?}N@B68C|<w5GW<&fm@cyCaVW~E$}BG@=tq+&=5{B+
zu%;z!!#s$Kt-3K=W3#(k-FrRop`JP4BR0U$#mhn}-kNIo)+Xi$6r)6^hU$8S>WWHw
ze|JW_?5-P(8*(VzJ@OwY*%gzNYx(&QQk@$3Y4+On!~j{jgLeu;4hh)rfG?8*58X|e
z{M)d6q60y_Fb@y;yL_avv+b(Ory|w@2ph)&ZPA9M?)c>JI8tbyY&P)_kfH$@Ipa|z
zYtEd|wsgI*z}@dE?6pq)$B(=m_Ie!8P!I?$vnBefiJBDROMv62&(Q)!OFvCO)?@nU
z8whTIIM6jqdpF%s6133}(19#_TRzsUMxn)`YQ?{az|@3|y3-4^!Xt8iRjTD0qF_7v
z=K+Vbv|?wx4D|Fuc0&)jg~PqtsK>v=ESs($4G#7X4$D24?d@jt4+*}G__0$52<@zZ
zW&uPWaBQJ?(bLi<@_r@fA}j97pzOHRAKh>#_2Aj_$6#imKfpqdjxcb?z}}pZM?<)V
zhx=R=WR&sicF}Km!?B*A?Z<3X-Snvif|OrCyaYHV{=pAsV1e?F++Y_m-lfDtr^0pt
z7Kz&LgRd!b0<$19BC4kt;9JcQ?+vXIQ@5<^Oo(<`KJXW@8zh90^u8ER)A*Pv+0}W8
z!JPR@9~An1p@%WN<jAk9*wnHzeA;d$MU`8K?lo_pnAy5}W5<7w_ei=^-?zrEnnGhP
zP~>>#Coc@dLB39LJDi$ip(2jIPRtDU)2La1`Dsf$?A9zHy0r*8>R+9mNrJYD5f6)<
zn;Dqs@n1Ps&d-7Z0yL*JBm!l22V0(U(Fd8O`W8Buv-&JhOUj)&Mn^?a-yVkS5F+AL
zaz5+mo>WUfdj6>|38WXdgq+H<DjdKzER>a!pu4@<Y(yxXo|Q!yCvDAZ{_=+=Pr>lu
z@bs1)wqNtsWc3p{)x|C!#t5e92ybtzB?UF5@?s80LAUgV<juIde)7rI!BEu++ZC<F
zWx&|GZ(j>g5qkO&F#ZiGN9opGC@5>a>}Q!BG1QEWf2A3(TVp@ddDK}EY<9HyyUbxw
z4!Qr<vJ%w9Qz4Qzd%H%xu`Z`ogAq~0R=lx)*sAYJavC;O`M1YfkQD<U^}vKaTb~7=
z!*Hvon<@Gj{4z-6Z-p7iv2eYe0xg3DwCB(g_NtfRAsnS85Qwm_V4%ocJVuHqBJ6<z
z74pWfhfSw?VDX*u>m~dk(;y)Z*3l<t6Ql^}=VsiG-=)rYu9J&oLpO5nzVQh}5}yL4
z?{z2G&-u@PD%S+mlty~)JeL}BRitG*L0ehxvC@@)(&?JG^`8u5c8QX@`}7|Dz{S(P
z2X@C9jSjZ9SV(H2v$9`0RgfHco3Tj6_b)!Z|2(7bMM1MHk<tJ2r?{)q#xl_?R)2hE
zad`9m4mt!`sz2Kb^b)#*R%gBg$iHML3DE<BgwKhm{hdb_lC13P{Qz*=d)}K1`Z@8?
zggupDefQzsQrB`(+sdP<lyp8TCqcxGI6nSIsu&-MaM48TKrOYtla`A8t*@WBsj&S%
zIAozB%h)c`vKs)aonp0&wA|%?#Ku&aw7Okbq&PC5%=4K2n%C~`FNmUUaK6HrH||3-
zQc}8$Z2O3R_WaE7lw0dnAtCw=^SbyEy@}xwa~GlKscS7Q0+qr<^t2MF0Kj38)8Pfg
zPGj;9Ob)zMQtC_{d?tQBrx(h(8qd)Yat5ODDjaN+NQ7L(fks0m{f5y6x8dIboNQEO
zaZDd<q`TXEY`Ne4`s`tyB)c`tKEE|nVZxlC&Eve-G(CB+1cx-J5{^&M;DDsKMekRf
zfxzt2!5n2drNQ7%qsPO;MQg<cc?DVC=YJ4UiFT`q4txRIrrgw`M;!bt9Q-3k(tl8v
z?)IL}GzFba*E{{hrD8(AAvhuf;-q%sJZPC^XafU2vEAn~PPso3U-JDy-k!)%xq{zQ
zZewQVS9I$t<(0aVBeVSjsNX*J&x0(Uj!Bn7FmP`CCi%zP)3ryBxP|-m*4vpI3PA%I
z$L+ckx;U%(Por>h9M||qVS$IoV={+K$@O$jMM0|S%lF*Kc^1q#L8pbEa`#?yhHm?d
zZ{5sa^n6WrEG(IE4FNZO2fYb|l34SvdcVu%;u?ZAj*^<IZrp18Wa|gn=j6W&2CPW}
z;ZaG%ckhv3KIW(3wcC210nKt+F^nONC^nZno9`1l=Sb<5{05!o2{u0h@>F4WE2ZNX
z26f(Zzdr?XUeV2P+Yu3bY+8qFA4Ifpmn+`6!^H)j;@cZDCl?9DM?MM?#$54Mq}(>V
zAXkPv@=10e%o}BM|Mjz+G3MN@=gQ;UnUX*_h5SAV5y>e*XtV2y8wNH`q(Lz{<T^L*
zYzG9YS%7T4?%@1$WP+tJ8-k$UXzHV^0Q=dmo{nJuPtJ7W;BDZ#1e^hgZU_Sd87myL
zAAPc}LhDL#0y)fxkA*t$Ie+#EaSc%54fs>hE{+#W736JJ7pxJjjR#M?|CFBNpN=;$
z^-#*uQO47g<i5JH!Ib-``;4VWOw8!^Lckecuk)sDTgqHxc%-CU;<uc99M8K+aY<$)
zDdqjm!z+tPXaD1-Lz4yXl5Wke;*;$1KV;6xsj}l`DFn7~Cx9N73Ru_Sf8eJ*F7d^e
zpgp^Zi?i8*l1~u>CHi!|a{FYEu^A}c$@O>`<+XRrQUR&RYqtBO(QmCyzC&KZve2Q5
zvA9+HTJ<SvgDVO_)G&+O;WA4%^S7RmLm2n5f7^+KG{H-z#&0ulyFGSP&+<hlL&419
z=mt+%0{7V2YIhB^Q04XqbiG=cgC>h_TOkgxgo7i3E53l+>?)ZQt^}DZrX&{n7k)dG
zGU^ZR(fR2fl7I??fr&xbd2wziV7|W;K(Utt9M@8Wp<Sx7nUaU5J$OZBxE$VpU0q#H
zl+E6ZCPr44Q7c_mS^3qEZ!xU6Kk2fpoF{85Z!Y62l93M$0muE}#bfC3fGPv}cVuXU
zl+Rj1Mg4(Kyqcma9;Oe(zrjbPcSrGVb~gc|@lXFdd$+w>-V6&9EIjQZ?;qBC{$hZ?
zrW(#+m=^(o&b8M<+Y;hJypDc-{=o@~HI=OVAW;g@I}b{2_p-ZB95yTSe3Qj)7ptkL
zsK1P8owf2#Zdlkk2~9@5TKcM`^>Wqq^2M8%U+Xs7zXUF&RXLSkNEC{B#m_dkJ93Xi
zBipnLgnrF<0Usg6jFHD8kSco)KpH54+f|KYU!QIs=eoAw2Hc<u$9-_uabsi(x|C2J
zr>js!_+3zf2lik(lRC^F3N2JqeN3yycE$q{u3OnwgwmknrjXNG&xOeFrCYsM+kAPY
zqj<|t6vReCG(Kcuhzmi-!MKT#U0qz$(`DBXnF}A%Z5nw`+$|ey&8r1=#b<^D3`UYh
z+4vs{GQJ(OL_fLcZSoU(gM#pVKj~&>7N<(fD4mDII?BRkw)@yGpZ&duud1qYy>ULf
z*ZnQ)=|QDfx`&LM?AFG{Lp%%s^X?T)`UhRQo#MDogk$pyaSN}=D(63X8ocb?UyN}d
z8#mx%Q+m4esKvwVw+MpigdY>egpVn}FQ9k~ai=OuSjae&ECvR~0&~kf^sw1O&ud%O
zWC-pt4eynCnonH7qg#>PQqqWj&?k7h8`p%FPdXDkE`u+)FF_ro$@A^mm4bocPY6Q9
zAs3-$rQ>mKs+tn;C@QY_W8(}E;!AAQ-BJLC;Fm$PJ^ewla!Yu5b0zI1J?d$7ANgkX
z*Xp;eiI(nfnb%iVhDV3f6m!itHu#)R5QXD{s`Qe0DT-L4P%ru0W;tVdU(Len8XFT;
zQ~qizVsO|>Ny(rfX41CHo}XsKkAgYjw_*xsdb)aiDE&s_<N1-kk%7+?c>Px0OM8BU
zq3uk?85w)cnNFJpa(2}QP6jeIFRG(_;y&#z%~T8zmm{gUX>yd2itLD2+ut8ypI%Q1
z_sd@M@@>tk?REDF5Y4cV?j)DMeohoj&$uT@qpT`1J~;L4$&<^LH7gSndJ}rz;&7C3
zsr<Y=$&}ZCrvwMf;$x2Gh}e!u!dz=O`gq+pOMrOkUBEay+JxF-bA5Gnebw9JLeAzz
zyuzb(;z1`*<HM?B1851mx8Oq>e5lUUeMbV-053PKc*Xn)Ww;cs_x8YW!fi64Cv6#&
z#<7cEQd$O@@Da=z0&26G)FoC}BICttj?3gSFF84Cs%rwopXw_O`Y`e|QD^8$Juj``
zy6%;^z_q~Oy0HjTYY3D1Z58c)Ko3I+d7e+LO;)~`RrVQQ>HdhB=m$&kg6$(*%xYJc
zQ=;1+Kt2UgxuYqSzt|1I=l1mz9!FA?{;wBOQvTD+*L63>HR28Mp5q4vOVKQwmWE+1
zr#wS1GVesad6SRoeNKA5|783l{F0!cvgHs%0+sZ6P;btiYNrT4p16-gWNHei)6~~5
zk3Qc+kq@K58&-~ze({TI2*MUd_B%+Ew5hQV9&-%^21=udCVcsk%Bf(*gebgJ?(2Vh
zpe7a@mvmN;V-V}Ov~!{|{CJtiulY{ggVI;=TtwYX`SOa-A}CgjOYtHHtE;QwX|)nb
zXiStH@3q0~Q;_0T93IMRldpJHx>ceI(X{huhR|!y^tF>)nPpzI%v(1HDjcfyzcq0Q
zC8owy2T;kbyc|^eFuVnv@Z?i9(ecSqpCf#2Sp4r$OLTb>k>KQIQs78%GHvrtjuS|G
zh`7zwd=p^mlI=5X+qGO$I$Jju_8=q-)GmF~0lnGbrU((qE>Vl*t2#G!!a#t$XAUMP
z5z)9%8=Jj*%$jJ(+9=cD=<ta$`y+#0Z$3`<UR^$?{e2)$C4$Sm8y1cC<_wpo&ffP~
z_LJfsGf{C6B5v{|sbL+Ig-x5sB4xyekiZYef2!w{hp}+E3twyZ$x3P1<auRabfJ>!
z%b-Ac7-0Tjt$&G>mXZ2fcAmDvE1;kGZ{9qKjvfiDZ)-OgfcZv!3@oP~Cp?dD_{ac$
zaJn&?;#)P!!<i)^-dc<4uB55SN=X@>7+;Xr9WUhm$(YaXbw+@UB4PIxtuXuF%Vs)=
z{B*0Qpyr$y`fUtpblZG+m}b(P*HA#51<`CrH>lpzGk&P;yaV4~Ve_<|gG0~R+t%%x
zw`+&m=`%DDKNI7Gkv}V&Zi8Ew@YXM`lbMy4{Ry4nJgne$B)>83iu+38)zA4IrY;%P
zeZR)C%5n_1({tYxPgOZGzs~Z~Vhye<Vxr(`gZ@dZ7)vJIvkP>w>q9MTs?063R<*3d
z9ByAF!5Q#YKE>wf$sfpVgadn!v!uGlwZeQ19RW2YXSVF&ZKv8dQ*W;eCH-9B!h8N^
zW$rujIyL)4L=3Z5#Ynp-EB#~bN4OfB$0)-1Q<6{aSiV92bai~d=ZF)f0G@c#^B4ex
z?vd6P3#c69J+0#df_#tNe+)d1Ct3w%BE<ebql^3V*ZIssR6}9}rq2op)Bvq4z|Vh@
zQkqrF#`#>fv({x#PWc^#@lZ$1rZ`?)c2;y#5x*B{>8V2tuenjub9vd8B!2Bj*U)i|
zJtNZiHZv-yjO&vyXCI3`Xe8)!AMl)??@bf#@0XO??&dRf5ke%)ew~JfCb^I^VN95T
z={x0_zS!hAbA!~-tUfa}rm_8bR>_c*fDl<@V`C^P$OKw%wXABCzw6H;_u#YrcL}f7
zq~q0#7TN8%ecm)b_+$ykQ@QnCFBB_M9p!IUn(7^ne?EH1NIgHdfD89Ke1hi8vSk$f
zGSsw?O!APC2h{!)x}5kG&CT5_9i{j9oGtIMXlAD~5gBRewWkeC4)HQmuk9X8Sf;;8
zh}X0$SBJu-VSW0lQ$9gHr6w^d2@P4^(>toCh64zowxvU61d_~{8L723)f${yT<B3h
zbT4WN4#qgjdrKuQ!zZkRAa3_RU(C0#aXLRB+F@hMNiE6&;Z0K)4KuY_?`6gWcOiND
zwi0cK=kc*R@JJ=Y2}B9;utEU0^&Rwe;8q=l#88rbQ0sX2E`_oU=~lWtK#Iz^;FQUd
z+iiVQoU!rl+O-d1mx1C`-!v+gKW_O96IaB?-~W;YQFr4}(!PfUO|RJ=KQ4bSpVId8
zA<^=RvdSo!kj8>bqNBo55qZZLuf|k&KRK=M)@bwzsH;gRe@;64K9>bKwmd}s2w7m;
zxNDT!bj;4z&zMi?aSoKgTc{}TfCy7#y+2BY^NF?RVv_DCh`R_aZFaijW(6~Gb#;yW
zsm=|r=n8JcKSWlv#=&AH{Y$Q}<4D=W(KVRo@S|OUm7P9HiiSm7kQ^yfkH1?h_#wGL
ztIRE_IQ)st3x4P=Ub26b=<XBkof@X+613^xwFma|IX%pVUpVLEL}?sp{Wa1GaYKeb
zA|5j_2Oobl!1OsE3sSx!tJjjG#>afNE2Pj@t{z|Cb{irYX!*I*igVQ$zVIzS=Cpd|
z&-%x&AwW*Xxp3@jq2;kR6+J_{?zWK2=3F@xKs;RB0lOAB^#|maKe3Mw{z_<FLm@{j
zPN%k3JsMuCC~5k&%|0~Y(k4}vx7^wb8x6s!fQS#{F@?r^kYpw7eE8H(@FmJMFjT_T
zd&-&AzZ7-!B=^RhaQ2<&^y$p)=7pL0S~@yC{XZPq3NGNVE2t=_&#8Ln-<jdK8nxze
zUALfJT@5G~zgdNg6faiXT1;0?E5odU0+eBwrRfg<(}U(YW5fJjPg1$TaBI9?zRIV6
zw}S5b-lagny%zse=>a!d*T8gr|5&zZC6#CN$A-pAN~L?oXY89S?J#(5IU+(hvJUO9
z-V-=59x?nvYj7Iw`@Jx?z~&H?2WC;=;%V;cH-}DX<At`}!Y+CmL2)8;NP8&n-4M5u
z(LLazxh*HHG$pFb!_Ph2{5$;v(D8?H>&%!2jRwC?Wez2Z-z~&;g%Vy5Z3aPw*0vB4
z?tL(Fbj(_5CrO4LSdoSd7s1nzg-V`8p+2RJ3?ZK;-enMEr&mD-R4ThSfm3YZ`THNQ
zLpEnFD7H=1ogaejhoeu@cMbHik1kspZ8+jemi<aI>Ov}_p))_<v9aj_`<$@rB|oP?
zj1*12b~z+K1%P3h?GNV|@o<w-3C{~}VUI+=pfY?UCoflv&vVYis(r$n3WG|o)6tC9
zIaF*=#1)*~H=Tz)VvHowG;i6fD87+>_RI(L6aPM-fpH1iaB&nYT<C`2)9cUPvfAFY
zyZ7y>UhymVEh9-+!-vn1S6r6dWPZJ@PDVL}HCEG1pO4f&MtvCzzo==N`Du5wuRy=>
zo_X=Lk?rd(R?BpKSJyEvhL%22a&uiaqpakR!@NF!C>5-eEp(DMvsxt(ujV)Iw60=T
zL7$~4j%5y8--wVWX*^T!fH#Dh!sic6Lu?$8UtPfJ`Z|%hr2vG?fx)cI462;BwA4-0
z;vkLd-j3tyaAm3nh4Wa2^M|RNa5&;zU0j5wv>`~sr-_Fxu7sRGTkFcseKm8Z9p}w!
z+IQF9TfN>2QPZ=0Gaq^EcDbJ$31@hK_e*fqB?&k$k4&Mx<92O{@BxW6CB0$lTA!pk
zuOrkO-QBL8pQENK*Rm@rLe?rbUxd*VrMNPnG){D8<mG{Ho-i%%+q0O=ALKW6lrmU-
z9+CNJmhMg@PP$>E&aW;|@Oypv5+HsArMFV_>0os5e4De|$f$lt(|01%uhsz_=H?qO
zUj_x6#e|a1`_@-}W9lwQ6xiY7%`Ew<p)LD&;+d_&u^v)X!PVe)cs%$g?CWZLxay(>
zOUNVhTZl|Dn;bj{&bO?rxA!E`<Zs_o><R|~6fhJEV>b{bI#;Y4*9<=H4vqv9t(rnN
z>cXF&8(C*LSuc5}pWucW9o_2vy>wFu-Oip29LEOuSaIT1qyqap))H!de>uk{$~<v7
z`$h?ZQ3LO}!8C1~J*<V;l7_9z?_bJcgn%bT#0}IB1VZ1hyNZ9F0hw%~fc<fQ<e&I)
zeyZ?CpW1?~Y|0wkS-s?9@XM$BWOm6qXRV+YT#1MfxvIhl{GnB4bN55bAB-F3(OJC)
zmY>6K0pVT_%Y*s|i&^%7v+eezJ8t{Z-e~@-$=^eX;ZYeQS5~*)2Nd@QG&lJ8N;){$
zE43ShbNU@svBN1Mwe<K$r_v+*PpxHfgS4RPXmBH)9vEMpT@~yZH0OOX!fmOQk)0PA
z9X{4SmYYhkwek3EPz@H524uWQ(MAGKE*o?U&{hW9!-z*+xO0`do|rh!vQ*(+vXWiN
zD}!_*k*z>-);kf5OisQreOI;=Lw@TF<a2));CS+kJ&2X~;%@gd^(8&TKxNc~xmf=#
zdDcpeEk>Qr50iAVWMUVDLcV-KLGEL}?aV6AkkEiK5Vnv@6(3MmSam^t(yEz6s$nrl
zaqGd@xor2lLdbc>CPIQfg{&EtF7?K-8+ym2r(Z*W`EakPXut^!3Xoseae~bYDYh-g
zjb_KXSgnu*z{K>9*N;W3!XeeL?F(ny`8m1oGx0j?>ITz`=!(OTwo9JGR+)I-Y1_^9
z70yK#6og-UxA+PKmn?!Vc#timi#jz^p>O#V#t-CcRJ6NJ#$pUEUTf^X?g|-VZoRrs
z)ZUtye2!A_A41WeOoHID#xSRprtX#<KiRcbxp)i&4-fDC<>=O_NVUfq8FE8tUtret
zQe~MKi4`y#-~ED>)FyMy>$<CFZtk~dC>xNw*h?Z(<kWI{d&(5G*tPAp{lTsDYuAx*
zDYu=4>Dv>I5Tkz`380~D`wV~-s+Z^$F__|lgS&g5>TJ!FUjNu+8&d$7wdJc)t!L`b
z%Zf763RD4GowY%EJyyAAU#@<A)XjHs)hF`y^$*FY4L1;5rTGlBqpIKpm>BFiqoS%q
z8E@-PblBFt25L*di2j~3tFb0_b-PXsb_!p#<#veDn3Vl#6zClF2;J>f{7*Zn4KRS@
z?e(7at|!6y1z9djrzh-pvm70dQbRPWjW6?0$dAfI+~0Z1Xw8vc`CaD`c3r4+yebZr
zxO8+Vn<;wxy?6atIx}48nO}^wt4Hp=l1%slQ^;9x(eK;w2TKsBA0>*o(;T;~wLR(W
zPf=!Q=^PZI=FBe2af<C`ln9EwdXSlUzn2Ur3Z*QjyWaY!gH4ST4UqO&KEqdmu2<Od
zAH-lhP=H6F`pyPhDf)%y{8*qG0AbVftx$34_(^U~+BOdTUS97I)2-2Z^B?wfhwx6h
z{*ec<S)e}J(d9wjg&#zjoP{R39i9*KaJr8aXziB6z(K#!&tn@Kq(_#z5=5r#S2QqE
zvD)?O&Z)ccjNBPY_>^#qZhPGYhP};mlkX%-4(Gg~U{SGrD+hfVv7Rv?MUZ(#M4VFg
zZl`v0)56)AuSSvjnOfC6IULfg;Esh?2}s|3n@g8^E#5QQn6R+l^#?VQ^1gqmbl|(9
ze3B?^|HeeUB~g9w*DnTaY`eG;j^KZWWNYAYEsTxR2Oq-49QcXs!QD!*{gnl+A0CE)
z`497kH!D^EgcWU~2~O8qcVo<8O-%`5E%xA}e@v*{xZyPy!6ENB`q!+%^4&W;%o{{R
zrqF>Q+shrL(ng`Ii97=Ie2^1vJUgkrRzEY6ee*2RcO<0*+1<O%Te~${^66K4I-Vrj
z1cpvB2((SjELmCTHA}2EhZ7tT3R-S!-`>o-_4j<MJS2^b{tPaOn2%Ylt{V!cG&ud>
zk?m#{I3iiDIibt>_DxMW?Cfopyu17LNcywo9ZsAWGiWWMS;4_|>%hJ(^Z-E=N5007
zpFR-2`<w;?(??51B{DfqHZ4q>gcNBM)s2y%O`EsfosN!Do2);3eH22TXJ*#<i(yhb
zX;O=1iAke>YKpT-%5$&vadxkE!PkQR;kUWDkEk2eZ+wuuvY7e>uk$BA5&N5;=XE)`
z{!2nv?n5IZGvL|(yN0ewoX68&{EErpwXqC5!vDmxxq;`;#dqn^r#e0m*$#H@-YZVI
z5cuU7@4<lqjVh~Gii6ivP<d|qLtvVsqK;#ArojuBx0dp(aWbFeWpcme73SpISlcNm
zhrv-n$@f_QW4MQ=rlg9!?8_JN>FG`Pd5n-Mw~Gn|3Q<)$fJ}2&;$S))V3yxDHUJUE
z{kspH-h;K33ftcIvp4F@OiY-a)E2nL|J{!;dq=St!}qlM9d}JxEfup+Amy{8eJjdH
zsU_B##@B4XA^ZoN<>eVAW+!A-kHZ^r7SwTh>tM8GbRP6b`Gxbd;nd^D@SJd6lUejo
z2VsMR&AnG!g5|X!Sz$pvI~NrmGCn?`sX3vre+)+~DDtFjmCyu;1<4xDj;IMP-rLa7
z&_MXE0{_vhXh#1o)->h4H~2xc?0f`>sp%RZ4I~@p=K0KkF#=TT&!sC=SLY;n`MfW%
zU~{uIi<T<<KH(jD)4kA$*u?R{34zw`u|`o=L5|b8s03qys2B7s^kNc>uSlY$XlQA^
zP?`Nc^=$<mF0mQbW0fc!ZUG@qO$yu~HV!s2z4p&hxdpi!8-|>QYF&PO|6dDW2u<_+
z-Lu-_^f&Vd_V(EOb}*;R<B6WW{@tjjJ65(50h@3`!bX!(j~lN(g`ZEqR=d`<viluI
zR2OAsnHw%0aL2a5rylb;+PR8<?3E_Ra)h7!F=0&Y(L&RkVz?AcR@C2Y(Xqd|B!5a3
zzKCNY*EGM!Su*4CL>%uLisjpP6fI%r<0@|QPxQhcz0M3SZ@Gsye>KzE)b;Vpm;UMg
z{*e(2(|1+Y8=r1-#{SnLT(Ko3X>X7w`J%csTbG?Mz#1(bTMr*0)U4N}9^FcyUzAsT
zsgaeRxhyyyB13z1kQZ>GL<5bTo54R^goCYoC;Qk&H`~+quQXvhyP>Cm)jD~7;r;(s
z^~}7;(%Q$CH-tUO-UHp2(zs{*0nnEO{rgup?fS;im937<gOui`u7$%XG~F8fe`hH4
zEec^BnVwPJ|9jy7@GWXjO0sT|rmqK@^$vIbdsI+tZ)*CZtMb6@22nX34>Plx`ed9^
z5wn4YNPtOLV$Ad`o}Lg%%)|S8pw35wfx#$Udez75B&hTBB?H3y+ui1SVMRqjPNqx<
zeO`9nMK4d2$a6y#dv)h8>E5;6(tBnxe;FI+_wM2;A&t`x3o$p%N#cP|2~QG+D2van
zYYqNTOAeG^I`b|1ssFxw2$pBA%CgnfB5`_~%kIxn;BQvRm;rLcT_~wgJG}vgJ0U4C
z{(Z9^EK?DC8ABCW2mpXw66jcr^=E6|;169xpwm}ip<%yjKA*7X`oB-3f#e!l;a4V{
z`{z7XE2|q2gkC35`JF$MY?1!trVpI-kJ8UYTBk20<Ks2Y3z9|^)Gn4ZbE~r)0K<aW
z9G8_W7YBLnpQm-WU$8v5d$_LD_LJ@l3IC!%>(t5UhUV2Nsiww66IFU^72n{Br=~(S
zxMSu9I3C{g3H9JyTC6jFXA%9D4b|6_frY-as|SWspEpv2b+1>air?9LbVSzdD%x9R
zbf}(@j6}%mOK$Gh^nJ(S3(JqqvlM#%IJZs%V_x*On{KUVM$lg$4g@Fc84j6b!>xUL
z))b-3a<g+Eaf!AhVMpgVr(SX%%j0pw6W&paso)UV-^Cke*7}peu2ydyEZoI1{PG4z
z8O+<)wxN$mpMi6v#^aK>HLg%j)3t46qvjF*ZeFT<bpZUi)8qyY<3{K0Ws|<WnB}AV
z6{V3vokO)xifS|46-BJ;^n@S23_g^+zdCPKHJhlZp_&|@4CVS+v-Y!=F8~D5Ifq`f
zO&oaQ7Ki-bDm;-!|M>Ko&y}}fho=bfD9hIR>Pr<>vP*ikg0nhCh7&(~F5P=@RAALh
z0_!-+>>K$ZsQ@SNN%j@98kXl>_x%NRY1_dDY01c>ga$ZcVf)kHt*db3c~$umB<U{m
zsw^=)k7l-$&Tfk}XP~Ya<32qeO#3PF40QMq&4XqRCWDhMlgO(A&2y}km6fUec{%O`
z$Pe`<o6eFxv@i3^i%bf(mr(8Lt?{@TH^kpdh8cAGhb0Z^A#bO71|c?edrl>G>x5Ca
zhW*mQmqc1nNUK{$5~$i+lmD#!1k!}ZCnB};SKEa%A}~J=(5UN`#-&_G8>h~^rgN?+
zp&xW>FLOc`N0QDU#Pkhm_l)-|$^YBU=MSx3<<kV8mP_vxrdDThb23UQDn9Z1a<DLl
z#S*&ocZH=S?;fVl_Y=#=mjgc6($d~<N-<*fH)$)&xT&e>zI>)`VuG5n)!Fa|Qr)a`
zlM8T^R2NOG&|}*6tn6Q^D58QS3gWCLfkeQF_!$_JhETAw8bK(K9F*^AcN77>eTCgB
zs_Qr;mDjMGrKqk#dWS&zi$ZItJ-Jh^uW#$;GHgD+tSBFKctSlz_HfbdI1ZHWPY(9?
zC0c)ih7GzN$uq^A1q7Egp>J%=#L`IVxPTW9eKGCsZf7pSz?)t?ip4@gLLnI&0K)@Y
z_(~(5CFF291WIkN_AX-r#3a8|7ug@EriZXu1Zlpp(NL8u0P%iJV1`uaUO^6xFuVGv
zANQA52#HgMrlvOL7QlYsTfjjWNPy`Bl@}CBKxP1Vl=Bf9O>hjmZ{+{Hr^4`9sI;_H
zT3hxZ6@x=i0w-bMEoXwahI-s~a_roGkaTmf-ogm5S}<%~{~vGKT&h0dJYt~_4wj#@
z0x>cXP6fC4oY7|`xO`!$elaw7{+wOd1=Ltm<y&k%GA8pCw^!R@e~UwU=+EaG8m2>1
zjQuFQOciw*Q0|VB;Iwr+?_1&7Vh2Mq5nwUTJXhBa4ro()<>mf~t3Z`5mthVGa{Y$@
z+k<Y`#AHR0ni)6>+dV9Fr}G629bmoO-q@I#B>>JK+ud!MK4;9e`kTFg>Y<dpyk+}Z
z(PFeJYSQMyYsS$lK?qN^cXSjG;Qt&Q)%K7Uf{vWw4NV^HVqxBm99Z}p7jjtf;l*_S
z*n>L{lxF&*Vm{khm&Fh#&&#;I#7yYs=6nzy9(DI_!t&!0QgbMK;Xm9c2i)#%=Gk!e
zZ|3IzxN1o>E{>wH^=+Ph8oZ9N{!3ZqcZ$}z+|q;nKUBGr!D1}g)>=A7f+}W{?>8aZ
zqAP1Ym5}qhVAI0z*)uvx0R|tA#3m4r(q_Lc8hdwLXiAY=_4tw)T#dk^6A>jyjJ<kt
zRn)k&r|Sb~&HVDwsH9k;&-YwNYKa&CT!Sps()5i1tX;D0lHFgm=aoc6tDxbMPp(^D
z-E^=iB3|C!h1CKfC7tsR9xUG?5%GZx<cX%t&vwEomU?gBy)|Vb$`~IX*Y%YE%Xu+b
zLF}yIC<`-V@2`%|UmZ$eF=D;mu<ini6`}etG5KDaaXs$bWBCz1Gb|BfOEa--X)poB
zR{exip6dN;Gku|=1#+L}S?F<rt>#(k2MPy@UGx0uY>LdV4&V`9TsWb@<oMQyG)E<4
z)#FSxiX+$UN5B;WA1Wzn5@`5w4l__)GF4SZft7YV+Bg21$HzB5kM=4g)5yTkU}?8X
z@TT*3b}|;cpRCwyboiKJIz9jd#wHNmo+Kl@^DLtu+}5d{#<zm~85j%MhDOVYhnJqk
z#^@@j>hyb(DyX_xrbuTfqU2oH^q`1Rc?>xTQ&Uq+%(4K>oo_R}0a@5+p$?2Mf@o$9
zWvx|5*4JxU<x=QTQBg9%!3YH-UE$ZS<D$B1AQ4Y1rhoWYh=X5_0+-?ks3sgT@=;PN
z3WOmv=+=;NOi)lzu)S`C_<PZ?=;WxVNGxPAYwIs3&z8CKvoAa7VjOY!=W~ydAZuF)
z=DMHW1<)Drk{}3pPYkPm>|ID{h7dBu!v&wFy}dn5mB4XB(yks(PYl8v0EErJAhRJq
z@an<;ngghA7Fj7->CRHn2e)UWW^3kz&O(Wn*DA`(D=R7v_7A6PH$0bGZ}Mg3i&7i|
z4WBj}E`ANguNnS{h-f)Uu;jLEYOi*+s>TQp&s2?Vf^jtA0|Li`#UyB<p3|Ga>4aD<
zO*adS_I9>Q!5FpwM%(jhv1s7g*%dwGk55jXewfD%T@GM=P>KUl1TgeW!G=`J_RWQR
z_yrmbQO&bqQg|R&ncq1{*Pap*ay=x6*bf_fGWWF}_cCihfP7>Ma`H9G^~|KD^Js!c
zhNs+5*T2~3MfW7L6CefCwUJj9eJvf7?0n@lHE0M2M`^=M5fQ?o^<7az6~HBaO8T1{
z@tkJk@dPv1M4@&y-cbTX)1)jZ=RVX5;Cq6G1!jT}9<)@eNTX1&P&z|x0|BAp5l)vJ
zWjUE;85y>j<t;|`TzKMh{gFoG!ke~^<7FPhW%PFuigVfhTOwDN@%(|;`~{u%xd0HR
zr&fBYN_IEl5$?8hC!@s-7P`wa&rdXYN5{~dmrE?5K)r^-I?#Wd+%V_J3TI`w6rQ&B
z6vXB%$&-?~SWHb0jE{|#+YeTcR8#=5=jPN13^rgp@V-vBt!IwXHI=CtP*3$t-*G8O
z4o*)O8=stluLRkyA+|bdy$QtqF;;uKl&eb>_W1AN;X1v}LkOH8Fl;F(*n})<@JB8r
z0P_gt=h3<d!^4(XKJ$_s9UO;@WxEdI$_vgCn?-ld=Bp<Zew{bHiu@nV%%c^(dItv;
z6v%J{DSwWFo(NRE)`ts}_+4{gItEaHKxLT6@@((sHN;<DW~{*o4G|Y_J+EDJOjD3r
zRrC7P-tAg*d6weh;<{qj0sTI7upU$OEzR90r3be>@7_XYSQ1P1_71n(jl)CUqC-$>
z<<@yHD=ypeJ|NWd**)X}Rc(;C(h-zN!^}*5_9KIC`40#RyW@p#BAX&^ZcRrpJi_4D
zi_xJG-71^Cg;{rKJCN<|ol<JL9@vRQ!Q=h?^v=Exn+ufTgzHL>!}w?2Cr_EdhlbMn
z%UExBi*+xnNV>Mj=5OMdFmZ}Mg<40%dInzHb5>Q7@+n433m50jw94I!i_zu{>%g$e
zZ_a>anXVW01}giq9VI%ecFkiTorW3fZ+evlb&0jPO=*!pY-7KZ)5S;~B;NrMAnIv)
zALgDl&bzv}LO+RfQI8ec{BMV!h5FXw#6*R+-bIfom?9ErHE#<|V7o6PX|&{RXhuw&
zAVm{E>~Se2w66|>2j!E$!;JESTg&R(heY~s-hGZra)mLypg5uybMOBF*b<<|*dBJ-
zU%uD_SO$Y^@G;|>XOruk%t7Z01TpE{Us#>}2hB6$8R6T%>>EJfVs(|c(*vrPkR*P|
zh_F-zxaDOlxK#mx^Rd|HU+*q#AO%fS1eXRVeZGF$UC}>Ntlm7kO-?T2{{AUtXD7RK
zCxcyCc~6pHP#GCoNTS;=lX!$g#>;<FFExwl@5`>@s*BQ1>3VFjr7;D042&maA{688
zvsRvCwyyz%`VzQ^;`Q-{HR%6Xbq*h@5_WV1s?Ab4Cky`QtkO;QUF^A#%<k1mS7^79
zSp?<^d;v>xWD#Lu$Lj&WGt&!96GUl!pB)_=zKK8|ASs_ry0qeJ$Mxg~>p!{=ylGj!
zF3(L)QlAX8i;}Mxqt++%-Im5u%FNKS<xTLMjEheW_++=U$_#MV(bma>&^EtX9MA3L
zm7(YJ&D8oDniI$$ApIQ+bdNv3!9g}}3c@TTjQy+TMvv{-!U*r$(hT>fFMr7kc+T$}
z+sKRo&JHj+l<fT^|FAu<D+07-V^dOY|4-l~xIfWzr6p6@(9<AP82MjIduPuQHL&Gt
zuK$Pr{r4mGdozR<Cl(ufZH}-~V7(}Mv_hCXc%2{$t?jmjFjlR;p<Xn^zWtJLhRKEg
zVU55&d|>gX`-4KUVE_B}Ora;4lbWo`_tHG9F|B`GeeF6HvTM@?kAxrS^U_Mw_2)Z(
zF=ol^W7axe>|7)aX3If}gH+3&;W3&ZOw5d=bE?C1VKYp>7cWZ%wgOm@D)!wMe)7>w
z@oTQ{u8KdnKCpX^@H(yAL1qS2aLFR|t`~FkOoXnh>ksjd?tHW5=6;L@5VPkgIwFYU
zqNu6HYT67t`j4ce9N9D)0>rwxg#`+NCXfIxDzIKsgo*OK7n)b?xp%6chlu$UI&z5;
z8qU(Or2`!2D_+NuH6s+&=(kWAAYLtylV2XwZDF>B?qO#HU^0rRfqsL*PA&W?kfOW)
zJ_EhhzvVu^n3!<oFwMjH_E=_hbCaHhg>;VPCAN#WRBF1nAfi=g#X&`*fS*grC@>07
zL=#N-$MLW8pDAF*;C5JI(Iv0`O;#W`%90!pDe02;B0)!CG2_nGY=o!LN2q6}e{g&$
zi%{SNNF~aVUVK=Sg)vmE<((g{S(KJ>fkaTenDSDmV5T}DMwhL_bsbKFb4%Tk!C_(B
zSyNeaQUt&adVDBjCk`}^7JK7n;iKIVa}5tC0@Ctft@YT^Y0Ka%CT>mv9-amS(eVMp
zP)f%%bl<WG9BHL_(wTPukEg2)3u@c88xS!Nl$I1JQ9`;?I+aGc1O!BoZb3l0LApgs
zx=TbrK)R$Qq&uYHP40c~o$vhcoKJMe+H1`@=7?vkxFo3yo~KveCCY=xw2h^C*O50f
zckbTuwNPhrKHe9Ul~q>ZPk&y0de{Q+)r_^q0sr<UjB0NBeus(t{=a`%I?SAjWF&C_
zL6wv~1;;W4DVdGc)r(*Bq&=JN{>P}z>F!O}F{WXxhc?g#?kwyKc6PvOv2S#;wLXqS
zNB)d5r&d8$@3;9=t}|<$e*zv*6wy7Q!$!c5CoGBEr`511Md%p*+Kl5epnzag2(j6*
z0845+XcU4j;cKpaca9n_b}Q&(-j@PYukVhnz#+k79lIR)>woxFU3K(h&n`LFbwt;{
zNyO7nLhPom=u*D}=Jl(xi-}S*l-~7pbxhVY6q2pH+I6O;5aDmQ&xM$h`1lqAnmTSg
zBM&3IdxQr>`*&N@p42q0H>%%pEYDg|p`HI&C=ZV_K<*OU!{6m~`;dOo;yOw(*|@&B
zinsfifm84%qFaw?VQI<A$$A5;-qmcT$agEypXvjAF#7_P|DhEuYyPe^Cqs(aIZ)E_
zY9+u5+64zpnoy9x_^PF>1DMyR4$opb^IkJ69xjD#2~@ksOJoDb=Tp?7!$*>UbuYZX
ze{}^LnqD6p0+u=zCM10Gf)ZQ@fIOmv+1Bsl3QyFqZv|Mp(7+*m!&B!I?~O#$w_!lg
zOEHtoZlG<VdJse6fjrkyzB&w~iM!#kO{n)4YE|s(yy%b@@<hz6x13;C2BSaoZ-rib
z>Q$H9$+daAeKZurv7fKw_+`uuz=rN<=Kw6~8@{4IX|aQNO%g1BS8^vu_<E2(J?Db`
zI#n*_Vp4w~B-P)u2n^^3O_}>wBC+4-?Ev?pJJ9;}t)Mqk_tg{@plQo#rbKx!0(6il
z=`77bp%sgOeZ7&`S0dQw&)X@Rxm|Z3@amVJe}QX3@si0-(`m<A7pWuXcZ*C&F@J4|
zclL8@#YYP$7od^ON)>%(_cMR{yv`HrRy8nTriLcr-54%cTA5iImO;gIa7ZEewXo0#
zOoQwX9;KZ|nNmV_5Y+0}XJ3hkmt&y$l4wuI-8vGQyM4{kT6e$6{<yOS7qI6s=VQIe
z)=R{YEaZQ;;=h01YS6SAFe+SlBHatn0!W6GQ{TNYt!sol;1Iw?GUjnm^pkS<bd;2U
zf5SbEi_FbM^#cbW1wxRX`sC0o&kv5YxeQ6k^6Cr>o3kV3j&`V^G<v2Cvu7|c2lrH7
zA|%}6+3&<<D9Y%lKNo1!{`q6g7voo+MxOBTClRtuycXJONGHEdOr&)$sP*Aq(oJ2X
zMxCcn2K{Io_ZE8yEkKT`G1@oGRB0kizwB)DH?J@e;8uUhM)V^OHEura|NRrFg<$sr
z#3(T<Zu>bW)c5)(=ASbP;3(?DKqCku)94mr$HT%#f|fj~7+Zm2(60X!^?h1hU0qq(
z?TSk{`&fv6!C$cKM7p`h22SGDHr=+@eFS@2@R4yo-jgj~s=GHRbWyDOhfRPEja*IC
zNj1Q=u#c5$SNXx87mEMR1B}~-$3zrD16pQ3%WV@sV$%~gH5~;<1I?Gj6dvfnXyCt?
zpZic;Mz*q`Pshl}G5!W`|MsjNB_EyVKW~!^Pq6X>YYr0ZE6y5FfPt6&l6V#M{R4WM
zh9Yo#LEhTfYSh?8)WmlH@IxH(cwEqZ^%{pNKA9CG6`H`+vPZno-<h812nqGOTDrXA
zj2F9N!M43uPh=aEd?7OZ{g;@y=1qi`mlt-bcY8CaUZWb0q*z=}x5dOw?8QwuBBYWQ
z1h9|-#%`c-j#)amy!zJ{BQK?}v9eJ?ViMA<zb+91Xc3D|z_ZlIl!&RR!OV<;4+N|2
z&b3h9!_tHjS!x_b|L=f!Bs=jfN#E7mNMs~CJWkh{*>(|TBD<5Tf+{Wb_XlIxLb!&9
z4Gf+VsT2hlkY-UF<B2_N-h#nLqa@S;x2N$tIy*tOXoQQ@2Lz&f0ieEXD`!sMCT#Q&
z8f>wLEc85ug@t5@Zeu9ob#uNQ)r(Yl<Z>iiS<<tp;A^h6S=RHP{N&hVel7DbjZz#P
zZ1I)I%($K5BHm7mcm-t&D69F2Jv^^m%OQz#+t{!~9<b7qBcR@ZC2lx(MP$8dlkT5j
zFky4ivG+VM%IJMa{IZ0|a;D6y1;s1&GcTf|HVl~%<bs>7lZX5Wj2$S;3YM#{A)h$y
zACmh=MHxMO(scA8Jw+`dpxf?h`Mv^Q0pzM7XT93@G#LJk`zpuvlC>MMI6GTMB$R$l
zM?sX7*to9-z+FSXZ}>FB@Ov4C#xPSv_4;z51}co8zyL{!x4_##u6z`!sq^wIpu8Z1
z4LvLSc;{IY56|!BxbFihk}e;#95<nErKDvlfwVqEtmD%M(OJK%WL|0Gr;l)^F4FBa
zQe<vW6Tv!*AFFx6vcrnDdZ77wZ56f{=lzATkxEXGB~YGmy->WvB4t?eyK$5SbJ{m$
zUwV2TT$N*FM0WB(%!MHS?Py;4U1qpORM3|Jpa6@J;<cHHc`ll;y}Zuz?XLM8X8zmG
zAw!*CtUON-ax4fHp=5ya;u9+?ct56M-tNBjkBt9fdxs4;E3Rh<POvAVp&pEBt_Bs0
zU9)@WrL3MEe`h_h`?)x84Z=US5g^@XFd|-p^I8+*evNnu|2m`&k1Nv--n|$0y1=_7
zMDEhxHf{@)Rg^GqpqgPIw|x0c=8nL%!ZTi$IJMWbYz$xxsym1i>6<?LVhzkxY^Vh9
zO#U934(*g)D`HKSK=XRf$^GN-jyRgvX3AmPjjf62k+iQ{4A`Ub_ItBxv{RlxSpQps
zp(`0FK@~ym;yR}Bt5|KNFAg(B7(vqiu!+2junN5vA@yw@1td`JD#`mV1YOLj*}(Cu
z2xTSmDQ5hA?<;y7C;NrlQ5AHqp*;$qb*Kig4P?IlEjRrOQh)9kgx~dG(O~f7{)Ud}
z(+YX^@17vqjhl3lETmX;tbN}MpTbix!SF6O%6O3ZWvqYj9!cuaMDu)Mp%rlZ(%w>Y
zGhm(65uJ3uz9xz?p!ug*gC5)#Mr$$t-G~=+DZ-A4{vafCAQ|9<iv|z15V-gfpN(fD
z`YCE%?qp?q{t%Xb-#$-oEMv`Fmt7$*UD9iJ51%|n`_cMZgjFIJsh|xlln!H&pNL7^
zn+z&KD+y{J6wZX8#%Mpd3#Aj*y%@i)zvSGu5Muru^EwrCyQSr8f(_9-G!t9ST*c{T
z?0G?%=Usgte;-lH-W~*c^Q~LEaC(A6!&uE)RSOhH+@zr$(c%}uXhjbgyO+1_p&~Fz
zK+Imh@k8kaOi&qbDL*?KA4vh@!1`N+9dclH;)T=crtv-fcscU{BcphZ?8wk?LqMQv
zc-#{v`X^7WHV85+)0X+Yi1*W}WV09}(GnYcj%H;H;LQXJUsBQg1z&l2WuQa)w6oBH
zj7;DhX?hEw)|HpFibFCS9QKGXZIpxM>Ga{`-#_7_&6GuOfSA?20Aa=XoD&Uz#t4MZ
z<C4XtJzyrpCo?tsf{g$){j)Mcj+W9f#iRdx*J%NsH4+{X1LF*#G~J389)Q+1mMEGC
zZ(R$ZHzzEbl<WP~=x+a(VNaxF!2j_oox=h4{KMw(D;B83-J55p?{D8cd;aXn>xqQ%
z+K*o}03Q6Z>}0&`Zy9~Ii4MdjpRT_wCMMT4SFNUtSHJZSLca+A0`a-%bgDL{WuNx+
z@yGw4k;PF_iMP9(M@L<%iYZt0^e`de;SWSF&+4kG6ixaDyZli0D%0=fie0Cd6y(tP
zq5l;0_z6O$ufQ=`&0#vqlo|d`qSdw98g35}_?Q^pdo9^If7upIf7{28AAgA{BSn3(
z3LK(-!W((4F(-|k8C_{()UUr>=zPD>(eZJ(lano`H=ltE6|*$8l7$vYUhCT-RU*tq
z+47f=CgTMswpqWuhFbyismbRppF1nXe)Mb<G?er)%y$bW%H_*x1lU~X_MH{UC!=E`
z-cvS3U;GPKmhd~t3B~L&7K6m1y4p~RaSM~32;rMjVN|V0VoZ!Lwhk|$cYr?rwUyOG
z!#oBTVY_Ls-nqxok3UiR+nSG_-_v{W$F}Ags>stc%@dao-2}IMmya$K((k-b#_|tZ
z_koi_e35PEW(S~pX!A85eR%A4%=_TVv}|(t-HX}Tt1$z6K;ZuURXsu1Go2UGY6`l>
zc9u#qCtLoZdV-GL<9beZpQN^U#I|pPRwvZ;f>j0R0t4~$DvwR3tMX-Ky84)m<Mtar
zct)r0%=7C)KhRDCge}DDkQ>|h?IHKFxVC7dZ+r(Zqi$TQU;1!=*$ZWgyIL~MlYeDJ
zs5HBxu%Y2upor<y<>dekU>yVVAM5d18V;5o;dVb23sG?FX9;<E9T9{6tef-j1qr&H
z$vwDRtdYRYKtS&~6SgfNu-!iU=F0@AupTI>-ytjFV&OUHkJ5*s?Lklu;9U4?oK&~P
zUZ@AY%+1NEs{EQFk^QrSC4w6UtOPYbuzn18G<P&}@^J}qH9c4`(NMAEPo_Ku9%75-
zR}nKfBEId#rCV|?ogVCct0eRLKQ2I662I)js4OD8`}fVW(#}ZGey2}6I@*sn%=-(u
zGR4br4vnHCRL{;&QNk=<ua&+#!4uHNB0LvPW+jc;&oRE;fz;{U=ywu)+Wp<@@bu}p
znlc#z6bG;rbl5tYjT68aU@N_<<204EmRv7xz4vNzo%c@2>jdxv=vjwSCg$cTLiQkd
z<A;8fmI)3~o5RMIt5L8H*-jNzirUriALj6Sp$V%??^G{5GeH_xP>+Kc&P5^L+Qyak
zg=B70;@x^N9R<@Ip?8IpLXX3Nl2?a8?;{5D)r~1n$PI3r9(mO|;9K@t(f5Kz?FX<c
zK?LY@#BIhs48VHr_;}xzKMp>L&?9ZsgtyoU4s1kr?gC_2sxGuEffLN+UhwV5u+b_!
z!W$Xp!Xv;pD$#hM^jHNQ`2yKcUkYgNZu=;GaLY;-WO)W@NIzf9^j;cRTW>X-zCYhM
zsWOQq&nWZr#EmY@ZJc2udp{2JZ{O}&uqqgx^&#HwEw$U31o$!C6FR^TjlXESz|`Th
zJc5wN1!Y`frpAs`7EU;xAMK?yO@9Cs4cVSEAyPsYbR)If{na{9YFkc7r6hw{lnfrH
zIIbKV_+UfK%8rVFN7Kx-;!lVaA!YRzeuMXDgxNbgo05vL`&C)QsVYGDkpry!<S?mW
z70U*;CWtwLLoiX#zNUFXxG5=<qnRYpZ8ls|EcKUa@AD{d<9<&~Q9YsR9vC2ci0xvb
zHPOLFSTt3ipILY{k?w2MOTQeGjEsti_X*|`)%8dFcH^0Q2PMATK?uMZPpF<{D@_yB
zKf6Ud<X6kffJn(HtZi&SM#9fTi4UOuttVw;q32+Kc<Y2}2<$x}aezNU>SJJt<Sk;m
zX3CV#2sHxa?gj1VoE%`7Lovu_+8+qTt{Cgyu01z#T!s-YfQ|m%t!X$JS3~~A2F4Gw
z@8!;bt0SH|;*Oe@d-1h8Yg?zpr!Zvdkp+9OMHxE_5$5CB#AOWxBzGcIzn3JPkY&eF
z(Gxx}a-&st_(E=Oh6-|izbxuPh}vM@VniSn5D|`*IY!?Kl<h?R0D`~QsAGTh1EdXP
z=IB@b(((!uJ);oB%-ph4(^hggI*KPO`u<B!!i~*n6%+<YTFK7&=|kqGKk#i}kqZdO
zt?}Ub@*cH<P!G4g-vRmrCc*gD0n%6Vm-NpYfP56eqiQ3)0xE{yXQKY^m{NPKx0cN;
zEIi4`$3~`6kbwcrEKX2tuWqj7=35IZNKblrs^YN*^e$n$rT*J^co%rsvzV+Y#3o=+
zWt}nqL6kDA*OTnu^%L~@+q%z`F#V$O=AEU!ltyMG24&Wh?g4od9o_kH%)oE(w-S5!
zLPWQ*tnA5SP7!J%*nELPCg`}5;8(l4aC83O)|d!oR7Cq+6G6%UH@m)fZ#bEt+dNHZ
z{@K-4__9e{TAK3GiV`a*c|m{xspp^a)~oO7>`1J0qQUZKre_3ac8T$*+v;yJoJkVc
z$atKuoJR+gg7<R5Fa_w;UZ@TAkGQC-KwbL!{{0m|%>5mUpq+W_i$i=5pP1a-#Jp#*
z7;4<?ETXRpSQ)rzs6HU~JC!3A`bFHGz+$qLf=0+R1E2E7@_KY^>}zv#{Fn{pe9tv?
z#;M}JOEcK1PXFdqZOj3q1tMNpUZ-oe?hBiaGa8HVdyT-;pY&pmTA!@<?HYNQ?CYP;
z<|DGTyslkpOSSLWGdwcVom7Q~5Ok*scoekfkVXKNx3qPy>b!JR=ue5~HzVLo{iIzE
z;^P8^&fzn<w3%;oytF9zMq5Q&>N(<Vt1XT<Y>A;*-}FiKoXSa02Serxn~hG`%K)ea
zqFAQke3NFe_RGKsy={d+5xrWo&Azz65XH&eL8ksxz(Ozt-9eS9C*b<DkAjhn@31*D
z2kfLD-M@?UnW;dbH&p&hylb?hg9R9COvwUF=K<HI(G1GQpu6<!8bA8(TxK+f<MT{6
zbA1t3HQ0T;5H>j~&$=lzO9t}f7a^5r<v~&pAQ=+s<)Qropx+j1*Ge4)OvA=m_F2Bh
zZes+rRn{}N@48Fwpri6*Rhgf6g7T`-{<Nq~!q3_IHlhzWjz6dPO;0F%|2c4GxLJV&
z1oj&M!Xb=m#CjU*ylZ_cP$rzaP^adHdHK}p-yp^0hH}dTS^*s^yhF#q6IBfcwIlPI
z$gF;Nw>h6pCB#}EJ^tCm=!AhtQGiV_FYx~{!cu<elq>1TbNok0kuhodOyB&KceDNg
zYiVr3xQ43EJRoqy<STFv`HGfQm4QEdLEhJ74$t@xAFjQ==HWT^mOCjf=k#FW5GH@{
z{Kb;ZtjsU|^bf)(47w(Z)z$D1TYOG#r`R($1<yNPoAM3(SIX0qfe*kTY%U3Ttes_?
zZ-OWG!utx6>Uo0HKn*&`S_;ml5h{v_td!f*Z;kCH>#fzRDk}2xb7P}p?=0mG4G*iS
zgclX&r`9>)`!IDZzI~&^a~lDu3M%B77(Dy(3RntTLT+#~<?KT!`XFBbf*r6VXsACA
zjSN>+!nl&xU9Bb<&a~y$uHx?*U^o=dkK)suW%>%PnDnGQ{?qRZn^#MmY*F{PNx!BQ
zFHgjfz~~KFqEPH;H##JW#F3htl$c$tTX1cC)o#-C&yI<Rd6pjj;xUa??BbC0Gz?Y~
zn~W<`zA8wlVMRy*PJ$rj^b$%k5L4Cb?X5RpHuS4Nd&n+hPF$?j&rh-^mCfi=JW4KJ
zptrZ1j!pwi_hUp$*tz886!6h8FYs(`c7VwN8nWPb{FnjN`7h~M)0&RW%Gho7bRk=L
zphHP{x}5y>5YX8GorAzl7Z&S|*8*f#cn+PzK=kO;e6F#;;as__sOU%-8w=`qOdkbn
zbMu4viW~mzA?2kX0|N+K@u;GtL2?w(VICA$6c=Y&Sy`DZ<o*GqlnMPqzaO#jInEAL
z$bGhywU`JA$}wkO{k0MWBz_>7f#(khO+Zr(OSFMs;LqUexpd$tfbCo;#S#d7NY^C<
zTy`x#L<l0wtHsYtYhZ(@bhP_D5sK-tb7Ir?j)n#u(Amz;)i1w~el9z4sQv{+bTcId
zyldQ>r!HCEzI^%eHX+s%H@iQB`U`us|D|u8aRFb_&*RD7(OK|%mFFuf$z^>D9yhv6
z&q(j@581&W2#B7zMwy;}K7*}V(f8@)wbxrG&KCCeg<r8OKn?`n`d3Y`Yc)^qqbH?2
z+5qLiPTi<gh2v}mZK6`Hiu|2}dwwUN(1BA}{*HZYtY1woOKNXiY;5UMHW5vyBe%wa
z;Na|<n}KgWHtTCAlPyQGKMu)zz-($|4y6LBPxGTE^y3qu;8#e>nWzwF=V&uoZ&0D_
zdE^JPSdbPFliX*};kW)YGYl4kFj*5tz*jG|(lXi*bZsvQKDRCB2Y1beK4k^5i)7n_
zhj{q-lSyPiZYEDHD#~vFMI;c;`p96=2A0f|C01qL{83)*b#<^kR9w4>lkjF^swPAs
zC+edZFt}s72<F~$rwjKDjDOBe|Jpsk%E8G>Ztw+uotyrK|NC<h-O%n?qWh~_TJcm|
zDB$PB|GjT~AW8Te2t0WC#z8!yn9T06b7uW@x*FtngguEgx19GVMUgA|3oEM0I1DOL
z)BAoIf5`Twhm2y|8jz#HUBLr{q3{=?#m{Xx)Hm@>vLj*_QQr%Pc9d7LgMS?ytQEdv
zeT`2Mee*YYSy>R~3g*#w_U!=s&FVMPU>F(e*Oirp<`G^A+y~=K;ko6y>3d<|S^*Q}
z`o%L}3Ol0r!cVz4Mn;G4S=o_W^@baQm$Neu4-{okE%1|H-lL3mXa0aZxR(C>M#ka^
z1HWKub6R2|G0E}@=D2vaQd_7^R79*{8IPRinC4(kfnI~$kTNOLsIcBBJc6_K`W1dA
zMID{tfl0hpiP_z1xHC0^&q_*|bQpY0ak9W!dPm-1JNA(38N~Xgea<s9PHX?W3$jP}
zT|edjV<1=zB3Zx|a!VrW>q#So`-WegADz39VwqYRdtqvovk#MPSOMcm_`y>S&X-j#
z@uy4LH@coYJ~*(0Qb6p&3(Pv8?MvxkXW?95ULP2GKH5KW^$Iyzh3AY}zlYO5xq4xp
zlXyPO`o$U$`?kgP70B44>ET`T@x&w{<*d-M!hbQ}{S+TuT*}^J2t1>n0VTV1=&lMt
z;Bj1ayS}Y9bLkDh=TvC!@1F}(<0MN=To#)%ma^YRRl!8UV`49MPVI5q!@vg+ZvLIJ
z5wKYD%QA%@qz$eN0?bU#<tLGu5c+gzv+C3H7ixgr!`oM;bqd(kg;fNuefPGVKo73x
zF}^;#3UD|OtP2@`0d^N6!NJ8Lkv+F}Y*(A*n_I#uiyPEh%gtl<C4cAGO`6yWEKv?g
z)Po*Mzy_Y1_ZVW=06udFp6C^_jEpoyyzqPstSwq*-<lX_aTRLi=C>#*sXS#PqYO0J
zqlf!T45A8=$?pFGGES;OhUW`daI!s&fG4Z*k{EDYSi+`8A-v)*17}xeMYENh_#1)r
z4g<3*iILrr7QiqYi#kZP%~(}q`U?d>7}i(dF|_JGCB=%{+p&3dmH*&d6Y%^&bnGj)
zCWGN<>G={3*l4-f?l;3^kDZHy+~6b1x7i)AZ2=;7P{B(mfWAb;0w92zj`+8RmRJ6+
zaL?zIy$xZ%14NFdk7K?s?LbI)eD03pnmShb>w3rtf>vUaOej)lBZ2D%j<l%qGxv-*
zx_G(BsF;|D2nUC9bzqA9vjJ6L-$6ugTKvilctf0t-Gn!Qi*;ac51$L<<>XnyVc%-_
zL9`Fn(C8owtf9WqjR_M2naFSmwBTg#dv*Xt@bOw`A%lEKdQkh)ZLsk%b1E;XyqW}q
z<^}vj0;mm4%-fdsR8?a<f0w6Z=TcHpS-p83YzU3upcx<i^<;_6PPTZ0SG2K<pC9F6
zA>4f9Y|D1+9xoFTH9jSRe+o_j`$&swD1FPTcR&J!3Q4oUkfn5FcE`z9?lTyM<OQ-K
zkn4o*%p=RDY?7UBYZBREOe2mCwj!|fcs?`4$k>xx6@C32*gl^uIdSk03ftwN#qJb7
z*V^)EycijlzCl%GRT~>}#`b6wOaux7j#!2yK9h{mr!j9lL3yw8;@bldM^k=h3;I@F
zTkGfRduM6&c3h#my2@_vkk}Qe{UWTPC^-MM`Anl@-_(>+V6^uKyW6ONfyqQ4p~ssF
z@GX>bl)CwdQO416EA+7C+sOP2F-518%eo@$v)>l4x+mEzMhfCBHuZf%n#jI?7riGE
zCizqNVH5@b9kTOTAt9lyvv_4s$Fjzd;-8-Yl+L(3S7S;(Eh4?2ZCXE3yXKYhwP$#b
z&chMIjj$uv+HsT5_N<Q_PkgQlIeyOl?#wG-=SNXwpY2*_&HM?~cvvtwC^yh41Qj^h
z;*E_pwrA&8?X|MAsr3Z0j!4`e7B^ofu{>jd*BO(Efly1J+t}Egzp-fyX%_J=^RMzx
z+*>^GXwo9!yim6R<H{Un-ONixvMJM7hG#8Wzbh3i|MLeF?*xUP-kwdm#GV&wJm~Fa
zR#8)W%5}c;a3!@(ilJjhCheN9*aykBgm~gi?nzm~9ur)4_{la;^vl_mFMpohL(cg8
zN#XyXCT1G=+~NEUPe%_0gPfVUYtifivYS{;CAEO8KEwD^>eJu9ES^1P0=LheJ>^Jw
zz)VNQ^zY8-r^U}vZC`SWzeily|DJf+(J2MWzx6GO;<8Fu!R$wg>~`>n-r@8hBML6d
zWW>aXxwID5l4X1PXZ+^~mu2xzl}~s7fA6Kat;nm{SV}hF^1$|jS;Q!KJY^%%qY6(5
z>H2!XWw_(bGb8)5#IAWpcS^&myJS3}sjmM&k0BM}FfUGLZr?LV@8`xcg%?b?{(Iv8
zo)G8^-kwoOu>J4-{r4tQQMJPS{_{(?G;}4LIm;aKOHFuErKn>s<vH<jaL%T1?gT6>
z==+E=yJN!(cRfAnmJ5C^Jm`r6q>3%u-PTqyiTR(~Lf!z*mj^qdqMqsbsa#Ts%eY7=
zDyHV{TnnHOpExF$qE@mLWnIfD^Vv|RZ1doQ_V}h)WJPv-n%B3j$oEn-f{#5yTmHlb
zr(PDRCQOHGPO849w)#SM>^b=@T#icUY(M(o-PeC|x}l!)n3?G%JX}rBWt0pfBWCp&
z@~X?&8(5U(HCiS5ZTB7jRuC+dFRpzI<A~rm$sr+8As825bx^jCU8E>2`?{K|P#_zz
z6lbqyfAsjBrn<Pyv)a<MjE2%~Ax0t+PreH}9%4a3T?5MI>FJC2&CN56T}>Eo-Y~Je
zh-elORkWVjiuDbFY;CFVu&}DK{6w*bmrVEZ7EDy5V}KT;uI|{M6PbDwMnXtXYp0Z~
zQf@g&jhd7~do&nrY*=_hbZ35RC~I9>a~eedSy+g^OGtB98uV>fYup1H81D6?6T;_B
zK~YOggDZCXO|grgE>R_Y#7Yw^d08djF<%MO|7Mr9Ul4@)DdhM|{6M@u*AlZj1_lS@
zDCLsBVJhJN8#7GTqs=}vdcMS+Fg@J}Qk4-!xSt(fqe~eyl0+?qq#AIo+4uCy1oeWm
zjr3Sel)q$4zZ(M}|3#>Xq^pJCbh^Oyw+FL7;)4qKVbpNR5xNM5U4>Q~yharlOrd@q
z#<auPSsbtS!(B{{vZscUDq21CyGIr8Ymk@4V)ULL+Ok(P*jL@!vQAp2>)Z?3Aw60d
zJMi|xjm5a23|&-~l!nLtG_KEb<9<O<uZ*sJluH2~?Pamz+-sa5-asz7Gl?G=Kl6UC
zrVtCY;AcjEkfLG=)9gA~$y?puddr>~D65!MWA@6BomG-e`2LNveN4p7j&Qcb>3Y;o
z+Vaf+g@KqpmZ-!JyVRpO$8A<gIRa}r+zH#A%MB}%ql(*PcO;`G5-z{;V4TB|sp=ao
zZxV9MeJr6l`o60D_Jq*q?f?TD_Ay51E~63N{NI=Hna&fRD)?U&#`*CjW*+PPD0;_k
zTRLEv@hs|Igjn9!+{}Kn%g{ciPzRDTb)80!rH;Kc&zv?-3Zn}46-*a%P4u<E_tQs%
z?O50R#Z!14|2oQYy_DljR8iACCahY!e5O(9bYX`*+RgXR)VSf25!Fe-eyn(e(7#pb
ziwcv10?Cp&-mSkgcwS*)ht7YJcLVcUpPYHtgrC<{a{tzZg>v2Jg_2rHr{`YN*oPN}
zNtKI#-?e8N;bM^qoz|Hc%pYDYMj6?&H}t=cn3U#3!NtAa1jLL@TCp=XX``G^&hQbU
z{A}m7RZXPY_PhsPy=-WknfFCLog)v=MYE2(t0Wzc4>~EqTGP1cqFva`)U!*YYTNlY
zA_n)6?VVlTajndtIm3*VfP09(fly4d{@I)Rmo`u1f7Q%(rNn{-Dd}}tQ&xx1q&MEy
z)ap;Jd+L2^Z)FZOPSNG`Xy)QcpdxT}_+*K<c#=b!05r{+>JV<s5s@*qU`afrR(fjh
zyZf@7L`Ug5#k2?ayvm+%7yOL7xum(=_%aW%!5f3cSY6~UH#48`c`sRpW0Z^AJSXb6
z44Zd58fVAr`H9o(k+V^qnxDRRvRSVe(C3BLAEjo-jwonUNSS^`=q}AzCDq%vWCZsA
z;@=m{9Zd<N>rfya-eB|of|-X`w6L|Qz<FVF#w$55V3ZJ-L$XE}A;k+3B2FsnTW|U=
z@2yYO(DPOOkqx~*>nT;FE2A{`W4590AI15lf9COqMx`0KPmf0NP}SGQ2p=@X><iiQ
zm6;n5V~4@Fwj20(rb6o@^?*>V=@H3x)1F`{90tw$9|MceZ(JyP3sW~@1wb;`U$Aq6
zJwVr#p*H0oXMW|^>0dvp>vHIJ<num$+-6|LM)B}Kq!ipI`hFYpjFsosaL{ZNeyINa
zYlYmcW%}dE$qDXs78n%60n79IrMZQ9@oHx`vz$Wr-Fx>Ec&5&C8vn>$XM6Z~)5s>b
zG}3!%h(0NawC4Cq{PjAoleX<y{&83<%`v(-6s!b`v_cHq%F^cYR>`H256^=H++bHa
zP2%5gijK}|=%!EzBOy)%oS&DsE{mFm)Ur^cF*aWAm9(b#K;qDQ5%2Qotjy$?{bldI
z`QqPKa3Ln<tNJdiJ7JA>8=)aFzSW-|MRhP7ZL@GN`ba%|^5`dAa*2xNC%MJMp(+Sp
zRUwvF(w-c2?^M6w<KwMMX~3k#3FrxRfZeH2{JKws`u40<8y`-5LAqeuSbuDG9UdNz
z2sKWQYxVHRsAMa3pBcN>lc1Y^E(r8&y8hx7m6iA9pFUT8A>$VHn3>K3q%XD*A^F-&
zOb&Q?SpICu>^718*X+)99CV$<|59h(OiD^B@@_%-*4--N9{gCQFt_k_SRiyT5V_MD
z%=^;aR9ziN<~w3Oj$_W&N`soD-9MK%Hv<9`^YX11W*1r6I5_)A!n_m2#o9q2q>wzP
zqNElP`A<zs>TRw6leM#o+!`A8hqYBzVL?G%wxzl0jZFF9fJyb~16N_;#@U%0WSm2f
zo=G}k$9t~&*2zQ_#4`!`K~&`au8GHuAk9mv{Cx4Ypd_i@J95DGfaP@??wv+;g;#^8
zIW$FCPBco3V);Q|bUGVKm~1A-KNPZY2-NIMBF~2|R-9MXI7RCs)4tnIwa(Rh-bfpf
z-4sZ5uW&9NmhT=@9S)#z)0Q_>Q6+sZ!5%%3m?IFBJ@w6))u*c^YE`*|y)?<|*W(70
zNH-j$Pja*F#oI&Oek!C|(@8N}>%;B#3Wx8H?MCwl!t<ty_nTt8Yf64C$#eU3IK4b$
z&F4vMOVj$9*fW&m*D{AZsNxVx=Nyw7f16bOZ{?8ilpYN(Q*vabk6>^ps*m1`IWR9U
zQK*Lm$;h8yk@IU@aFTyvk31Osc~6x^+CEoKf1LiM^Orh977gT)+38(X{J4>D%I}PB
zt5fJmAI|V6-Zpf^_T?`8beYzk@ln?E$j~E=I0KFlkxu>R&KeSfeEbAkJZ-}g{kzPY
z@`40$)Q!?qIMSUmw|cNDy}eP9zR^(%g8Nhq3{WDSvOX%WY!uN_HVP~gxN`k|979Y>
zFxL0mGL7qk@Nuts$3#kL7+R3$%Fc5?Jz_+OWcn2+j0;2Y2W?f~iC$vF`8(d_v;Wb4
z-NU6))RD!dFSN6A)95oIENAa#a4H%7&O5rXFmm`2`X>fcTeIEGlZ!SIbd;44%{AOA
zlV6Hpf|zVuw|$Y*qr`vxFy{5I?4&j7lkmt3?TvU_)Y2ISo-F>1g1YZwBR@qi)&kZ{
z%-`BuZy56^EAtp3%e3}F^d86Q=eUHSgVL?jB4gLXT4C}SnKs^d!ZXgcdP<BYB!K&8
zK`*;WMN{e~{($_dhZ>GbpV|+frIbv87O7WewRMNHFZfD~R*uk}3(-aX8W>Jy<lk|7
zK2y45ag<-F!0aswqBf49VaOe5a6Vt%|K$jk50Jzrhcv9uxOtOP!lRXx)KCx!+<kFz
z@w&RY5L%+i+itH-3}`{Va<uZqxF{>@i@vE9<)cT>RdsO55~b3We6OMFB(0ZDFI97o
z|6RZbqOs!PTY1;Sm_)sgd2RYnM+jb>xXc3__%LIxF9FTJm6$~Sli8Q{Yf|!EkB*-1
zFA^!TtEj#R4DqC*p?M$spw~wf0C$hd25@5mPT(;*g-qw;VvThM=CJqgMSjQuPYpxs
z38TC{a-^b^5FXvh+X)%9l_g2PdV4|Tc|YzGf9dy#yg(>!YrN~Vt@SR~;|H&}Q;kJM
z6xjo2b~NI?xyqyg)B|hbKSA+jr(>E_={rWPpwM&nJ!5G97ez|OjoUZJp^ND1aIm-c
zmuyY91ru9p>K`C%e9<dsQ%Z=3_c1^b7|ej=V^yjp5F44<xx9WU(Vv5(_5yZmu+w*|
zuQ&M{V-sL~msy{;<7a;lF~abRR$5j%QY!0D@s|kkxw*Ln-d1T{1wV-#(^n>@ON(WZ
z1&qJ!?AloG2L~5evLn#X$gvN3#wD+<uHVHWr=g)YFy!A+eap?uj8BMLQB`s^ld;~q
z5fo^0FwF%5Bfc?9rT+H0#tRf|F4Kn*`@HNKP$0+ow@-D~WR^V<WVsL5AP?5$z?wuq
zI{qb3XVTDX^UekwQsx#G$}18`3GQp@C<op8F<ZR(rswtJP9A4;NZ?z3NJp0w<GaFb
zC65z8ihJo7GuGTABd4<eckOBB%@f$z%-P42Ps1cQV<KZB9EvM&?%jK1Wf3MTx4FE0
z4FSxJhv&Y50p-I0<5z~3uTTpjj^%%STGUllj!IPjS({OqoeqY^L21u<lfQob{MyQ-
zr@afVBQ8AJTUUvmUYbO94^(smySKLPr-*uckMSJmTC<@DOTRD(jGFrPN8j=+k=deo
zeupWv5*|M0b+)GBkHaK0aS*eb!H)no5thqz#dbLtq@xJYz~Hlh?3c#Ii=(V)(0bc-
z*<q{mCf~y$=V<NFeWp`bRkgL*LPs^4QIG|BWLGO4l%}|GV(Ci#_TWdKtDJOne7s6R
zTw+pJ`hQ%2%y}FPjGF}!=0+x8vsqrf4G;g6q^7Cb@;Fc-v#YBE%xjb%MUNW1Sk??i
z4p{PF&1h?DlO-op&i0!bC&Uwh7g;+UEiitiWj;>4)Pe{Wlg2I^$9F)Z1+6z6Yeml<
z67=kCu5M!83IJSlac(X{DHl##RC50KIb-IB5BHD8p0Hc)w}r8>aV#$UdBkuM&2Fg)
z<Q|JRyt&KD($AjF%Q0gM>Q3(m7tBT4R?&x^qoLh<Ve=?4{r%TAq6){c8nEsC))w~Q
z@#B40AFi?;i{-U7z54G>6@~5vv9H{X=A)KG{QteFs;ES~7KW-KSa$Afj)D4xn$;QZ
zHzCFOuuUBEI&Nxo<Pj|g)swo$l7OdD5AFtNZoMng<ql}p2bVkbqKWab;liT)!O4M>
zb(3x0Cj7E`e&hiSqfz^G%hjkmIXrq-xh)5#;}3BH7z;P-cpI_`47Wn%foKCA(121T
zGx?N=(bm7IZ*a2nRTW$sm_rN)FqQ80?|3|kW4)i+`iMT?>1Y8xM20F-3U&}ZTlSUv
zTa_DvfXPSW{cTwzCKvSZxL7dwVJhJAZ*{!84vGorCMaiLpkR|H3ps0e;_cLza5Idr
zYJE!lK}U;w<44=)SI~e$D5QWV{`amfDkfS29CG}(m;Jynxguiz`PQ!tQyI7tL&J>O
zh?20>rds!dq{zAFZx1itzc*=I<Kf{3@xHFV#B{B#LqR2NR0qHPrmdM+a8g0}hia#{
z8be*9=~Z8opw7JR1xwTWdxi+=PXAz>nE>tM$KKot(3>pbyZxD;mzR=JeIEYg5kqOx
zRZsdX{dwG}6|gDP)Ri|@H$Cp*)jAy5!l#0Sr$ij=+H0uFTfKIn^%YH5GH2NdfE+bE
z0vxD<v`Y;cePv=|BQe8*|B_U!?JNvoBE?CTBU6Kk5yvUQ)VsUe!EUmu_90v2g=juL
zfwydvb*2>cuDd$dT5aq3HtXwN^Nm$k;ixKysnx(T^gt0&T5k`&OEVB8MyVGYtdWyq
z<>Y=HnH>N3@R40V{}z510V?=7G`Cr~LSArSsiBc_S7r|LXu(TP(~<rDpPolQG+E=p
z#R><f=S#lDwq64msp~a3t1BDHeLJ}Qs22oyT~G29NT>NkdB}bM5%IeX{f(b6()MhL
znWt&%-0ez}U)=b93SDF1-p5thRl&2PeX)__B;StL51*|QC5v!*E=U>m$<#f0gE#3u
z=YXMkb3<Cdje9SeA+`Kk(NTzZ-Fu2un%)4>^_ICWs@Q^psst+0f8(%@wzr4sf;9JV
z+XF2Ql&^=HCa*oTNBdS$&&SQ65U%Xi+F#FbJD^aSG!64P(GYam8l-okI+E^}KK;Tk
zr=t;b^DW1vgObkbot)c9_;7RshjY;q(dZ##jZW5^haq-lf8K0<SJ9D9zUN|88LKYh
z`oheFdd}?o<BWcB<iYcsOw0{R#M8twt0>=M>a}!<H<>3X^q+q6uFNpfG`;_A{h>WY
zFE0i%I5c%vGHNQ)@!=XjQjS%bo>eo>c29CpWVzgBbvL(cNN%J$ap)cWVOv{(hpA;~
zs4a4NAL6G>nY3~7@WP{FZr%)YAtQF)8lhoQ=xMyT$miOEAZA*coWF~n528BKXMxsy
zuR+@tJ6TG)3zlMttW<351y6RQN9VWo;%t@c8wo9A%!NrkCVDjeQrC!WQvNXzNa7x?
zXxh3b<b`+U^h9#m17D?HRbbdr=vw88>h`-`==_9^j-rKW!j~5gaNQ((`=xeeG@d}p
zU0`@*EAEnFayBBIt}KVP@owVQKDFKWO)^5;qSR8}@-e5MZtHG6738+o*tz~Fa%^ML
zGYe+kT`f5ny@N4*nz>aA`B`+W%!_`fM6!w+g<EI{pdzorNE5DW4Si5`krIGEsCQ)7
zWx;_YsQJ3vjo@7AFYCD;+dDU76tY=X((xmXg9dqQl7<C1qA2((Fe_g^{xte;IJK+B
zrDyT?NkBfzAEnKZ$x82p=~^#r>|D&p$sJM!vI@y0J~u=-Jy1L==7Ie@F#QUB(o<Xe
zWlwW6<r7BV_IX9}F-NBvto!%r1*U93%Jyl_5*tw?SiH!O1wO>P*tqsqHpyf%OTzBG
zltqYG1HwBR+E*w^Zr#55^-#|qJ~9Z8gQ|CHt<16g)u1#k_QphGvuAdU@7Q#5qNar*
zHdoy7JJYcVjW-kJkX=-Rcs1+)+ZeZnu`u`+GKR2#byw2bhA%5j@s3YV81!oYoU|%W
zv+<icJ4?x)2c~?QKC-!J6g;Wr{le?1mkPZ7j4Y~C$Ia=+7vh>PG^K4WzZOdV+Y%ei
z;T{vJc5j6q>Ep+r{}|UmdC0z7sw<+~Gu%OrKzz4$@`g_G2KtduBFMpi>hQV_HTSqc
zj0j}Hi4(>#CArQIm<#BTnVWZHVmO??t@-?bnS|)A75ol3><0z3Qw6K(uUODvVUrK8
zv1RU5b4?Cb6KRFn?<cL+XBzAwz7lFhXvjvHlQ~()=j6%O3&<xM=jP0E;4mJoypt0~
zCLD-;;=<b|as7&2y0&U*ZMuinq5vtlhi$FH5cwu0yRs78@b=_}%V(t6d}@^n!(qkD
z0*+?5-e1i<w~)5i7AL1C1)3GVI;I!58<T?;ds7x?Yh7A#5Xd%hw>vzkA7i_Qj=uOS
z;o@l19zvn-<*7jk=HY>6y?#&YpyawGLmWfWJEuzp%<C;(?Q2_WY)0cM(cQyi%`>Bt
z+Hxu&hn^WVoqwrElHY-GXt{Twxv;ukY_suL?Y-1vLb^I53`Z(+=js~fRK7~r2?+oa
zaOVTwqo~s*-f_p53#L7S%iGiPIQfmjtXNeut!sAgSb%6b#M5o->e8ZP@E}v_=+Ph8
zcySQ9r$hGkmF<QlliJ#5zcs?lGPZbzO=|1b;?$GKESGvyL~;f}H@h`eyRbO-<#S%W
ziO)>(8sP_jY{qkUcjzD7*E+{cf)<@?6<IjqZ$Ow24k&qGL7BWUbld&e$lO)`_T(qb
z4t8yA;oY=jZB9c3NAL~6Hthr%=$`nzWbv5PKw!%O8JBx^p2_3RK(?GZjO*GQ$?V4a
zYjb8*_VWqyYie8Ll?Dm^YX$d(o+nn7B#;ul?VcV{64Nl7A0LkErD&ghi#(N(|GU$$
zrknBxu7YIrM3>wC+TW$c#ai3VBY^IKK$26-J5}Rw(X|H@rj92|ih$WM>poptexFQ5
zK{Ga{F<s}}@#}|%oSfaodc5M+3?3mN8bXJm<*7%(P6e}s)6<jR6B;PrKw0^3bJEhp
zpwi)B_gxMai9Gx7@;4QbexBLT4YdqgPEBtr+g|4}E~naJ7eV(sn{qU-74om}t+8{l
zjEoF$*IT!Dngl%`hVl<SE)Y%Js<yG&IU5_~*QO(0#m2;{<>gzSEY*GO#HzhjWeX1m
z4egQ#|L^64r$o=gPtSJ}$Bxh!76}6b1LbTgo1Fjt8XxXLfw<PE(H|OntTxDc(X259
zYUrpRY3RO)N!baF85*28<lp`m)b8ly?0hRokTT(PywDYV&pq;Efsl6AocsiGb%9d9
zvax~YWy<V>qJI!b14UN+qM+|z3RCv*xESYu+yACW2tjT#D{rJ+*7!9P4sTGT@p`VV
zu6m?K^>(%R-1r<p$w-s?^)oSag{PUAn_m_BDSKhd*;Vzp?qjpQj}AEJ-1hW0FN?&{
z(+H`BuF)+-*i~|=BGIu#W8G&W4t6-JscJu&?Bp6Jt9|**!4)viy@q-(tR22iuv+?0
z>D97#HoHN0Xw9I*9w+=jx6y+HGJBP@MN7-07xunQa4~_)&~CVIjrlVpw)`)lp*;Z6
z@?+PB$W?6$xm-s;w3mL51xlH)t*L6;iysN9U9dr?Hr*1hdeqs4`VEBFg^b->Gqq<&
z({~Q(6_DpI)d0<U@1J-&=A*?-F++(H09uB-RB6@a!BIYgzkV$D9PFJsyg~6=DiVE;
z!@I30+k30pgOceiyFA+1^HO*&Q3Zk<g0i0H{;O1Droc$zE_K{rbOPT7<Y}{TubDTB
zuT1x^lM<FL-zQbquVcW+<!O7lScXwmT~nir?ys(*h6aQR46n%2-D35yl*y*|t%gEw
z3=&IsU$EQ+@ag9T8SHx=7u&Hg(vBB3V2zif-TM(Q%T8?7`!D1@%IMt^+rJXRTg%!s
ztHHxACx1qK2PKg|_8KaR)<rO(uKr{O_&6`gmqH5&_rM{<<%nmGT=B1%GnUrfslB)!
zlQ@4q$JxMy%@&)I2FPvu-04qui#e`$^VPNlpMP}U3Wz_hjT@+9F{fM8z2#HeZzdn0
z*dv|q1C-Bsyr_;W7PFO*co!VrfU=euMybdJd_}8YKP{dh2y4jMpLu0xkFidC^p(B2
zxmUh&%8vQVxu^&a%9z3>vALi$@>45%d__;WVHYhwgH9QZ_uL{jRBt<G`_6etw-2Ml
zfL~3J5^h6FvDeL`?c%UW41|;$3%ODF5!0@Lqq?zGq{FYrc=B<*kZqBRg*=AWa-P$z
zWKpBOjJ$b7KX#EFsa8IEkUd)n#aNO!{n+g&!qCL?JO(@yjW29;CJJ~+@IAGS{i2oq
zqM8l2`}oyLp60)PjUs7Ja(yD1#!^M4+sk@9RvC1Jnz`I{wJ+xE_{8(e#?%AHg+i`s
zE1nwkRU?%!vfuV=oti*&sXg&?+<IqlbYN~}HTawYmtqH}a9ht`0ueS+ekb$tI!HY6
z-ZJj1KESwG!owP&w*0!Bn;_&cBGZsv5L72$<!~~ZRKl~X`ik`WsK`36sp~H-i<9=A
z$c9%FXP9bt&3=TD(I9Vo{yF7%b}K~bVNwW6-N!uIzKevnDJ#ed(uEoY20rWg9t}?t
z`6Jr1<XmAZE|TLJ68hu!R!r<vBp;Vsd}Ny8P_VkUw&!+J>_~VC_kj8{F~;v}=mS_2
z@h-*Y#jP`MO==W%hXX|csRt|*$07>F3oQu4E2}noJcfeAUDkPrzg5SC{$K?u7BKM~
z)bEmBdnR2*b&DWpO(u9Ep+3i_?<Tw_e#E$Hq2eeej;X8Jpx8Qb+MP&^)o`A0Tb+lP
zUi#)8dk#k~Q|Ol6-OulYbOTu^Dj4(G_e{rip7ph+zt8&VV0&gD=+b{GCiVfy_=hFc
zDlcVhrxtZUSDn|{31YV7nOivV^{+Q#&m|SegzQDMWLug?ZZVqJ?%smNd3OG9NQfsc
z@jWU^CcpAfl(dG-mUhQ;)|ms$>qc$tjtUA2I>{vtoUzg!=2{M$hcnglR*QAOn%aH)
z?kZWF1gWm$MHh6Qonx68F&WR9i)`xK?1i%Ys@7tlO9Y0Z^<<@E%gB@~HhFKylWztk
z+@w7-qo6@>{F@+uij~amxl)_n__bz5O7J*Rc_xM16>MQ6l?WO3w2FTfm8|TwYk}!}
zM_|SDZ8xy&09Q+Ku@An!zQ9^gR@JF<nHnK;#R3}Z%A9e|*U!A7_`r4`xOZ>bb7AOx
zW{%p4HQ)r=wQrdaJOcOHcUvGOa!RGytFqLF7{g%gJ{wzGEKJO6z8{LF%J=q<pTH9f
zj1qW#@N(ua-Yxs%)jRnkZ=fTHjEfRyx(RgkAgm>3-G&PN_|(fbGAhT+z{rfq3^&W~
ziB@&kOotOrn*#@aezYp+Q611y9%waf^p${Cd37B@FnQ5HEiTqZmkl9;6ub_!O=(i9
z;x`eu*k4OaOB0o+eL%iF=d>ieckk7X*~~RwiNWFfYx7k$BW|fZUzy-F;LI7${Q70g
z8WE}_r>$frR_<Lt*=kWW^Quu&mCAUD;HVJXczX?yPfG%52<duZySasc+nM3vMdp(c
zOBfLD9qNR#W=8OIPaQ1h?u9Cwj@QajXs)$8wi;q#h1zR56wZ{MZU>YBvJVdt)oU-#
znn`0(GFz$6)uq39tlntm#$T1uq0WvD<XeXG^m13umj`;9IzBD4XKIWz%8!EF&d#_t
zA+9~QJ_+;XDX=A3;(x(_!~GoX9YJBS!+y2h0Vg7tJolru!})MFO4Xu@-xdE?o~VY@
zb3OGX9f#>fveOaE-yRDKHIW0m5&q4$n&g-1b4-lW)vob7VIsGeT!L=gPUdzshZXKY
zQaSs!T2(N}kFOXXNg}o6-1=?3#5!g1FY;a|u(}{<mF{l9lXI>uKDXVY_@Mdnigl|(
z{pD!B0{|-c{*m`EGYpIj2=@;4_q^1SrFumD-MZypj0NZ6zk|$sAEC4^`EEU5;M@h;
zVtA4vzs`V-fbIqFd41Z`Zg`wAETwVb?=dGu`!!k!?ZkvMq!c49t@ila>L?J19Ux-*
zF(7b#(lINqR?Q3O*^4@04I=&ITQ3EERI$g}ru@VqaQD;+PnlgiCWorOHX6V@bR?8q
zYU0r&YE&d?!wL}wlVW*Qo;@9`32qzqZ}U5rrtFuQz<U&OolEm6JV=bTV;`?`_2c)X
ziXCoy&C12aB_u>KPUG>I$%JN1QwvD8&@A0o1nFT|Yc?E-R*S9(2)>&QB5ZD2fAIaV
zZ0Dle>@R{!*;IT3x$8e?cKA-01Soe;iGmaIN|OrMfvuVU1%ea?1tyUPk2z70>8#Ie
z@p6D4l9O*z#Ey*4Y^@s33{NE=FqyKyBr(B{0W=-;?Cf(CEvi?3nfvx0hN&VsDpKw<
zFnmA%0W$`o@Q)J#+<anq_`Jn1S0$mQFhASpG~LJ);O)?E3mTdqY-P5{x1k?WkXMaB
zA&&BZ>k=(S=I@_{y979v=2jEM_KJT7<J{cbAd2?P_8tur)2lQta-Zf0|LUu=Fp&Ya
z88ojiAElgBEZ=HS53m@=+rg%T+{MKQU49D@2R@Cs^Fl7yo;wyI$I6O#?;dOrqNcJ!
zLotISLYo6)na}_#53sEf6QiWn<q*<PdbC5et*a30<AZjv@C`vh?(UcxnTIy(w}lN#
zEEx-lIVxJ~USRNO-WNybu~UhSjuhNwG-l;tZUPp)4RnN0SzAT(I?YJ<-_`O738g}8
zX<+F?1QX{5fPOS62oNP|?GDsj+UgE0)S$nD-Lvg;ETC;5uCX?~gFssV==vu)TG<YF
zWA5<@#0&N?pog%=Ybz`Jx6en6TiRKoBGmQLkllNoLzD+K33cptWCQyo-^XP2J!@BA
zhUZOI+gV5G2=x$c)C!)oZorqkYLa6V3cf%%rvmKh+1pIX{yC4AA>Wjz?jV8=vEgRa
z`~<0g$Q-&>45dxS)^?v#Du^IvB$gNm7O6j8kywjxRLi?CT+O8;Zl^89n)h-#cRe_{
zeZ14$R+Byrj`$l{;0%(m!8DJZw!we$HsXXyjxa2A7vM9(+Bzj7M3!j|L$%Wai+8Ex
zZ=MgmuL{xBlphmVn3iV6Rrvs_?u^F-$PxF^n+1$XhU%`6`RiWQ2|p-gLPsb)C7evy
zk*A-6;SK;mx%KWgim=;8V(RL9W}Ag?Uj-((4zMw4>!roDIQOs5`o+hI?=?VmZOn#?
z5D`&Bn4uac5=@c9NJ$ymKI`Fxd4{hWeH~5M*^$QQ*r-=ticMBhTPs`{LD<EIBWCXN
z?$O;i6r%dVkNHmK<E8idRE8uDzhT(q8#J%nB%2J!BU-A<KfBwJUs(||s`9YXZ*3Q!
zb48}bi0O^Tqly@U^ITsG==G+velZwKEepQL$eTm-OM0|HWc!n5nWYUMnR^|jA+MX=
zB<HQJ9jS<#OMDP{`bqAUm?|c+uY4UI0RP3lXp&@bXBy3DPRZsjmp1(mnk?FkJSstg
zsRMPFmCJW$s#-a%w-*_7pJE_@uwtchk4}vGmuql9q^q-cadCK>msdg{YqCPl;0h3E
z@NQ+7l_ho<L#87G0Tta7DtbymZmHkJukewyY#&%%c5*|%hajKgsvb~QQMuN*eHpIz
z{2Bs33kU?Xm=C{;;$rS2!nz6ss*H>@5bE0=GThnIZOTf7JXh9oJe|&L;u^T~IHbdv
z!2NOm{rrQu{^al2sY-BIVTEL|5y6TEpa2f^syF)>2qh&70&aNn^1Y-bLt80rIawxl
z4)$oms`iC2bjIeYY9msGGN-Gh!Ss}t=%QeCZ6#V=&AsUEGc+A!8$5?ix9oVQgsr85
zc4wmq{jS#rE%km@$|b4Xq`dnPfV1bD7=rH-=c+xrWnSDfK>qywi;e&U1i{l_Y)d`R
zQ5JXE=a*V?*Xp(T%K9qIq|B2l)qq5G^kX5Z9(Ab1kwVbM(f*%a1L+Fx&hGAmc$Bn~
z8cqL<I|!KnK&bcu$?N1pOl+~#{uMi)d-%oX?7T>pY5g<5$M`WEh8Y=?latpT?j(kh
zs7?(Hcr(13nBeQc+l{MrGkbAS0$v2b-9kY^<j~G{%#3%p-^$8L2L=V%+u1#AqvH^W
zGO=bw<>ujmL+Ie(Fw^AGOPf_N{6Fz{m7Z5S;sjyN810bG0KWq@wNWKHPF}hfW{Eg>
zcVPhX;k$oBZwFY5hfW_+j;A#1gAS*RuxNOg3l#x_J+B?!aU?uB3cJOcM)dNVzLicx
zd_X^t9)<ASU7`{~9ax)?EYZ{hCJgE6v0)AgdEdA=f8hrigijxFlOCIskA(w4UQJZg
zO<(_QW)rhpij^H`ZIuUDx1G+ep$WVjl>V6h%x6+qF>ZJm#uMahmv0}$MjSU*fF(rd
zlRWAltr8>ePqs^b_q6R^6#frcZy8qQ8f^_P1p(<0ln|sFln@C)=|;Li8blhT!Jv^Y
zX^;l#Zc!R(r5ow)hHtvhIq#p3YhQb_7rJ=X6Zbvm7-NpX>?=m$u#X-={oZC-H00~>
z8mNQrq5o&@5q@EY`5JaB8~xeubC2SM*EV*{z@d<d5bHXJ{nPWDx?cy!3v+W|NR*PE
z{D8v><U2Y2`eLY6=EL{|ajW%)rBR=HuO#~47SbcVFc&>6ex7kN`yyu$wwr4=-nW+*
zon6dh?yI5-3dJ8ASjca%!qi#*+!fR#e~_};BrSY^1fpXl3-cI!e4;k#m}uOctqpn{
z?ZU6&N)I;sUFBXcNwh^u(I9N?OKRU8SZlcvc~U>@&E2+^&tNQU_Ws_|cRp0f59X&J
zO-vSAP!88Z#=+q^kA@)IkwGziZy{-6q1xKoK?~`d?~k-{LNJ*z%D;xS0oTCnNyN)P
zQHuU|d6{Hgo17ex25cFJZ))n!YRfc)B=z+5N>qIGtDE>RJqbADE7Q$gX{4}@TZlke
zl=s)yUI+%(g4b&~lF|Xn{=HhDzD9sF)RXWmKsFnJ=r36d{}O=4e1HaB??mZ|t%FQK
zMKp~TVafXn6=dX7|5R#nzv?DQDG8TnBV!DsGC+jN(h<CO-QP4}3PDHMaLz_ZX5tAx
z2ZW;b_8f8FCvqL{o9||K{rd$66<%}PdJR}oW1}ayssm?&gb<=g?T>Wj%yCa^TKYTp
z+KwUjZ`1XML9SbspwCZJ3`vQrFF0sKRfrts@K9%YnK9AZHt#RSh}PYF=rSg<E>)!9
zt#rAsX)$oQyR_KWLD`cL+cH8OpW{zX8^(YD9bhv)QW)`@+aJVR4NEwMiG0-C&>Zes
z1bnt2x%7Az)b{$50L7N%c&bKRq(<Cr1WIIG*z~q}AXaViIeGHlb1s~U8r0rC>)xp4
zC>0_VnnA(-$CAV^eY9Ox>0^zem-<H%BK{;1s6OoQUA&F3DWv4Q{hL(rj~yeHh-1pf
z_frUyUH?dLn_?ey52Z`|v#XbPjxPhmNYZ07a{9pT=fKv1nURs}si#_mNJd0(FotWS
zF#EH}AaGRKZ9o^=$s2y9RBVv;q=0wDv_)4RDK&y~@v#PCuGNl;kX`_#BXE&Sj+I)(
zPoZ0MgdtPl<E>xFJCB|4xDgC*ZvGMrRF~ywX}(hYys~0A<GFh3)9&!fbxb|huGZ>~
z3ufqDTpjWzry;lBA=7`!5wBe(mRJ5Fy|-4v!DMy_v@J!MA7~Xbd4-yX3?doZ`%gF!
zX^$R_O09pKFFnT)i3{18FSh3b)Cguc0r2GEu6`tRKt=`wuUi{7mNuw}!?+hT91k%0
zl5ZhYn6S2FA{OxvM-+2aq6sF_>$0ors<4-UqB0*AGke^$f*@>!E)cu_1M2$^Eicvn
zz-^Rq@B}65E~55~;+p&;%BNchG9nCpS*%3oO^4J4V@o3g%je^RGj(6f+@K|sycH&#
zUM9)}F|5qRmx_C{$<_mYSbjXZi1#b71l*2j&0i21%$cZ*_(4dW2j%-f&|k4v3<WmC
zm?%Swurx3bBK{g@0kA8ZF(~}wYeHwLZkDZ`-Pa_c$si#k+whVpo2WSdg80C_-;ALE
z;sx3$osbEzv9KJT&NyKUbZiU+upb2uAelhCF%uzT_2;6`pE~e+sOWRq7cl?(U`x_V
z?UH(#!oo?7*|sLV<GJkB<AIN1Eos!TE6iyh{F<JaGlDxxa0;Hi?I)EFxWQp)c@(!B
zS-uhD#O#(!r=Wq;+dp}gS`<i~;R~igGBatI3*Ln1k!*0kwB*0EC6uLec>Q{};MQj~
zEftcj0)N=rp?N%UbO6INhl=WVc4a;6Ec?)}-xrooi{U!h@N+=m4lX`ex=XG_rF`>j
zQsvL@TZny%JvROwPkHjMbT_ZK7bg$eS5EBi5%lzo8n6~xZ5Q?NEmqJG%V{(jH_e_~
zQ3T8QM?QIg_`V?6&m>ES#ES+KF3?sS$aEX0CF<@6|DP6sJWVk#m(xOp1WKJTF#gKR
zXYlor`15;&vkCVuR(Vy`G0erYi&5bQ4weQ2XboT_oj3vHM-HyBkqRNh1c3K+IydQT
zOD{i;o0}U?(b3VqV-Nf3T(!Z4I<W?k)N>JmHZyMuOcdt4`v@188s2BS0qlT!KHg9L
zaXaE!#SsY^&F=p0fQ69rc6WE!-<RKTDcFmHqoxbyc9R-@BsKiY#FaaGlEDZS3iL>t
zUlJk1jH+6n8tINCISuXuOcbKvtXH|PBe}Y|0Ppd&5YIA&j0SfA`IJy(PtMkME{uFE
zCNibesDJ)DJv(G&Ks)7gqwyLRaM%XysYb+YE^EL-2dH5$7B<Lv?{`aeI1>N9_tX8A
zDM?75F$b8Dykg(mMIg%hol2@_s`g0nPzI5EZs<*@Ktp{hoR@zcR=#wyJGXbR96RD#
zZU%|>nWcHIBbWW!g1=d34^e-0bb!a?^Ji1W6RV%+W->Ex0fr`HG#vg(2tP3CNYEKM
zjJLs~0i0Rna|qCo58sD#GrI+d;CoqCw*UOu=-N2D9wQSh(`|2$k<y6XWV;mWTgzA%
zAxbyA>R3_rQif%Iu&qi$hMO)bu*uL|B{(=xOFMzSL5h{-9FplxxnAD*2Z^m=#oyVY
z__ZW^NYmk-Tkn<&V6%mswPgQWFZM3xTL>*fk~_h(UQYLif^Q+1lDh*JGoP#>-rt&A
zoWsfsKr0R!B>KakaT$x$VPNZR?=@*wK+9Px)#dQ@jr&GfLix`9=b_(1CofIaYes_y
zxz74)6m5Cxx%-WNMT10^SQh2Rl{l7N<8b==&iw{Z=JC*N1$+xyYB_lAhn0+Y@lgU3
zfh{*vIq~|eSn7h1#+7y=FxgqRs8pcplJK4<OX%m&`$YI2*Q?kR>^rf^7eW-vSXg~F
zQ*9j`s0fyv4?-p)=m^T;)K(Sx)7H}+N#4hDo=nUEM?4=iP%vNPt|1UEEY@6=pD~p+
z7u8cfFdL4{Ff}x@y4v`R%3z{6uxz|FQ8>yDw3MGD%iAbWuS@4VzwOjo2^*1|$7?5+
zgI{D@#GNka24ACFL`D+{$ThaT-QUKJaG6ljLWj`KIwZBeor7tyO^u?QB!Sr^#A^GR
z7n%3YIm6W;=f~^KM;1O^J|X%)^(&nqwxE%4w<PnM$Jg|xkUz$V7k=NkE1e)>p#I&)
z4BYqkU#3Gr^{u0QB0as=R@s3$Q1r)q$qB{te?Oo0R}iK)gj=Y5pV}1ly24f5^iC2P
z8PqGr9JSzwp*lS@<d+Y^igQp?$jLoKJoeY1T<6hF&wC=SEn~!mVO>Egk&YGa{<^}c
z!=Uuz?c0NPfjw%4&i1J>;n5u8a$?#t8C69e<a3N3f8jr&(64vexy~@7oFBBvn7k8W
zpOm2%#TdO&z4Pw}gXQYQBGd=yf}b5_oUjlyv^g1}K>?vg@wqj%&nC3ALBKw=dWckC
z{Z;TuvT<wAPXZ!5hpS`HdvDOYM#g@_sty9LaO>dYkfrp-NzIfS?{%l8x$g)2wzC)>
zeBOD*iK7ir_-k8Y=;@I!-bI0<-IMB4w?+nxu*%saraMTXg+x8q2^Ob+r_bxcOVsfa
zlyZ3d=kEOpC<svCJY$LdgjAgW6%%|iLRF8{spIc(aljl3<Q&(=&??zk@UHKUrbqPa
z`8^WqH9rvQM#v)HF0Cv~4hy?qSP`yeui9H&<1E(dGs;z69lyglV*(?5aB8~OA$&ue
zFPV*+)%RY5sv_Y}a${mlx+8@kCUeqHYATWV3*R=tn&W*cI|Gg39=iAa{&7$j!5rJ)
zC1VNNS@j{cQN;Ui-ykU)<8Dv@8tAFgGt^rJOKKt`27gaMPggpYKE6EjGbbHy@SoAp
zdjXT}3Gcsa!J`bEZhhQNM~JkdB6I}Uo(;@wX+J#;$u1HS(a^{(YO?F;yJ90Rhg`|F
zmX`TVbBv$VR_1DhFUrxen_Qd+B#j1Nq^^jM%$acvxPXiqmI$}~%_SbjHE*&4SwxzL
z<(E6YK43u>oKY1kZY+BlXZcVb8yl%{PYZ0`Z(ZM@@Wv~Ol^QGx1<W*2V&Rb_hyJ<>
z5FLu+vQ)AsiI801*pWbYB>K4x8Yw}>7}<0%*rC!JoBI6#pq~>BP(lOg<175!`>g!~
zbq<V2ySmiC^gHA!fn}yR#VP(>fVqFZek|SHO`53X;_4Evtd01xJ&%r1w?TQuaL<`1
zN72i8FnW7^^F_|E#Ns_25nVYANn6`++{9fJ1mppZSk3q*^#N#v{2q#PFwaHcl9Aj6
z%P10kUu}|GFxH3Sws>)YYCYA2gXqK>uiu;s?PH^hBE~1Wi-@8lS=c+xU!Tyu2%D+B
zk(5^YG1ETed}jE7?#=*FQ9(Va)jHtbw-BX-rgZIy@IyMRc>Ts+k7|cmyx6k!S7BFq
zv*CCH^_vZcpPd0`gU4&PZT!2lzp@PlVt3ybP`P^zf#MnI3PWgSof8S`2smt*kuhCa
z4x{rCv$4?!f({i`knd1a$<bEHakss)39)M`>Oc=Nm%Au+2wdVs1OgTaxSJ3NtO-04
z=Oyv)eueIx#@*1$0cOtKp8F(TmMVe<a*T5y0m2HW1d#^1Ixr<94@))A6)L+lSMe8X
zg^3H$rGZ?^@GF&(KXsbii6hc`uzo;#59K|MJ9P!e5W82o^Xh_g?>JKQhrj0)W!vb8
zR|IOMsuh3R?Fg&9Maj9C>I(!un(j#Ro-Ps8Ud`V^2r#?eL9j7lPq}Q%8{S9QF4o#<
z=xB+cP?^6%tVQ;f#ZzLOWIa^HjZV$tzREnJJk7vEr9h=XL_5Z0m<|yOfhtMYgUsUr
zCf(dm5z&E1)3N#^{NV)QmC-!x>kMAAk^Gu{(Mzw0o-p!YY)6>W098wSk_|_Bnvls(
zyF=qtO~8kklj%Bp=5i~|q@*(O_mv4g)2`OF72hX$vw@QwibR4cu3{$@)DUlR8+{>2
z-*dqhkg(Fds^1iNFv3APzly;sgD!ZlvZ6|nLFF(%lpM6l!A{G>((6Y`=TB+29xUmy
zxQZK5a!ZqO&9?Qo+*f@_p@nz%C(@_XsMIcFy}rh~2riT@8SVnSN9>Ps*uGZHdZ7Bg
z)O}dJ*E6GxuQY%^JZ4cAC`nxyIQuy=9Ca0yB9$Zl*E?Z}SFR5bHTirkC#aMtpI#EP
ztxOw3_snM^vT%Gh-(%^%@%o+!59C?K@3{T_{W)ZCu6TVAfr#pAa5*}LNz0K!y_EQv
z<mjX(<Cdg&J#ygKfhPfR3dlDxJV#1m@mLo1Fi|iN_t0rsINmKgo;315rO(Q&=Nl&_
zS+$;af~*v56A>CTW+6;({odZs5JQ;vEwi(uhZLV&jwXVmY^J)RqVCx^;1Zaa*le1Q
zp`7j4tcaAP3VCeX+9rxrq@P=0lT>X?c!}2e-aIJk;fEIkBfry>@+8|{NK5YzPFo&f
zg@mOv-u?&cP(}RyJvAis7?<>>y{*d{#4p5}JI|+epo}Y;t{{m2>477O(cafyR#kKh
z%uI#e&GzW%DktjizSOt5Y_BW`lIdA-b?=gtVLj9w-o2tuOAANKxQjCRv=GZ<Us&1b
z*JlgmQ~{hikE_tHA1_gEVnF3=#;(WCS};~+u>~bnzkbt)DHMgLoEAH)j?PX{=0ZTx
zVg<JlQCvG+1Bg|d9X9+LE`I=zW>E6af;>ZOVEG-KT7cCqJv`virar9&eYp4e*avu$
zAbkS%;jdq*Ai9nQpE%i^Sb;0)Nw4Sny^fR_J(vZ4KHY!FYVGo@wEYk4^SatRo71Ih
zX9c*hwidBr^Xv$`@Ir15+^Hqo+1aqMu!@R}oNDwW;K8{+LBD8sx)WkN(wPKG2T>}@
z$<w(KNkl-POajDwM&2n$MMXV3J&k2cQdU!}?DpooY#?^9HUBG}DhA-Yv|Ru0eiAwY
zgt=b%$Ug1ubWAaP^4c56(=Tp+s4-9BiRuCiKiY^+$o~b{Uz?{0X8UxW^(R927F#u&
z^brz(-m>(nvuQI>vG{n@r+I;hy~*=P2NL@Z(s0@2*&x6=WoBmZ8Oy0!L5A#SdTXtX
z3qGzN6ZRbBVK~?!Wb1XQYcDwB$;b&Id#B2*Bb$fO26R?H5>{_crAEX|4^J{PF_pa?
z6d<|-Su$$cv$zCtT<(4=0yf1dDSDo|51|Gi3OpzWQHp<T`<M5q3ILympJ|JkU?HO8
z)cN?w*tEISRW%1eqkze*SGp>)NuheI0W_w$#@8oPqocref+KIEnTwYE)2+?T$jIoD
z>PkqeBTtV5e-uEo8X5&)CevK>SDX~pS1RxkWiQCpDO*78gQ^4xmAoH@dl$zlJlW6t
zsLGpSyb?>EZJydqK>h;saX23P3%|#EpfFhtqsaOCO+#0?+WMd<*vwBj3T^kGwm%6>
z6wWVD9osE34jv%0vGwKY9h5v^2^38+&kd__Z;p|{5c0qTL1Su^1YH!E>op2Yhcj>c
z9yEX(spr=6w0iN+*0xqqSxD0C<2c>b90?sZWfy<S2wI#o7ym;+P)uQAIXWDm^YZda
z>@1?lQG;&<EbA;H5TBJmX!-QV?n_6#7Swvi;ajgk4F^+M{8is(n^p#Z(b*Ps85;Zj
ztEZ=Y#^w&(R6#&)=+wP)AG8;ex*G@thaP!KA?e8H>$S9S3>o!8jF14k#D!<_+LlGt
zql;$64fju>h=JuYFaYhM`1lqAluLbmMNm;dVfOL%C7~*JuDZdO%-Fc7f4e@!_-JWO
z*3{I5H>d0^R=`IE>4-bK1{*7@rmAkix6BfcuG?x`v(eHJcGNb=#`M%+qX-GeDmZ9R
zatpl;uFTSu*L(T>s^cXk9tXz|Al$%`1BF}%Dj78W9NzM5P^iN{D@*qcA3p#s!+L^N
zdw}~zMYSf-<sUz=^>E$gt&wt16azhSk=E$c(Cwc|O)KBT9$=1*Yiv)Id0i#M03$hD
zd%^(8+>IH*(pjv!&+ZckKWP@aw`80}A>_{Ou&rcb(smr$7aA%LwqC8g*0T-P)62;c
z@=5NytD4G?Ci-@|WZ7RnH2n2LU0pqg*fh@kan7953df`NFz~|yTm&Ego@k1Mc)Q-P
zRlMAeGRcfQdWDg0lF+-@%8z$YKGm){+8O+no;0&wY_066V>;zMi``)LGoYZlO?Ige
zq@UE^q&2DTf9IEY1~+0Z_iax{OX8?nu8Sv2ptzP<&ARaRQHT`3E3Zo-Azd}zgnq0q
zvX>iOJGIuBp3ki@tMl>R<Bz4~IHpOv9?vBb6!QJ2E+!5lU~DM<B<?ti;trysV<!3C
z>g)Eqh*b^cttY14e;TyIg7q_>HnbVoVs!L^`ZQ0y^D>Sc=O(D^=B-unUq8n~oQ;LD
zJ&aG?Kk(Nh35&xB=ZV-hiYinqxWY-OJpPFNrt6aJeA7qlAfq3yoEA4VTHS<4yC%K9
zD*c)H?M<$|i2D1e)E{XT=c>2Ahf0xV<S$U=8^+kI&Xv?mG@WZE47Uea+<HHVw6Rd6
zgm^*p<W2Qq$%Ahz52{~OYtKe$I$vr~zHEVdu9#P*^IlnXTt|jkF}Z*n9mAv7hgY2&
zFbD@DL$fP?;g0L;@AtplyNV8diXB{U?rgtLkOLx-*wtU4grhtFd-0sx--jdPS_tP6
z;eQ`qYknft$T1X&BlmV|qI&lb@oM!Rq9)0mz}5M><QWn#(m7@3Sr$yDPQ|#H>n=P+
zHA!3}Q8s#mA8MRbr)sP4@>QSII9oPd%Vvbg(B|&a6HDp>!&~oP&?4SnIXIIM<<O@`
z3GU!{;g3A{R-@}8RLVxXJ6wKe(bPM>qN`JAPUN+y>gL9N$r{Oa2g^xwo=km`Gv_ys
z{HQVR7Y}s`cRuO#e}0Unpz%=B%Ei54*Gt%nBSQQw`P(B-jUt9zM8;SjRN^4?@a~_c
zC-rWJv1EXbU}UqiRQaWW;w#CTs%*c?ELWAcKPGzl<5<3MLRL4EF9|`a<JIwVzT}yS
z2{k1(Y($8hoaAs>es*?x9x;MVxAD*L<c7J~R<*s9vcm?`O|Mw0bZc8180P_LbQjss
zPKx-$i`BI)RyM|rVPcRNz;A@dHtI9|gVnXnOp_^07X$+NwjDT>fZ0O)_L_LkFmze^
zO<t&)ktat-_v4}|`|<Q|t$7N#o!fI2JHymaqBFUyL#m;`;BIG^1L$lYj-b5Jtacp@
z%EC}PV}X$pm((cd>-~`9P<g~NyJJI3_xj{Y-w!GBo}L-~>sUe+PH%vpcHcb(I>uBs
zkx%l`Bkxosq)%{YtUvu(E03@m8Uz^L?jAS#$FU~xjK7Tpksi!)j_9H`abcWug7xJw
zxq#qs&rj#G1sMFNrtcRM??C##8K3fp+etzp8R$CzVHcg+LB*y7sR%An`RPT=#x|U7
z<JV((!A6zbX0i?rCFSQDI&SNmOJER7%8YTfw}>w2(cpAKN#%%m#P?KDPi@4$Ebs)g
zKV_x~rY!J8yF0rDT@DX_t!=~TxS~Y`G^8m_Xr_-J>^STf&o=Q)|LlcyoT;o0XnZkN
zdhH$TBt4?QpHMvg6->T{JH)mdlso*6T_d-KleV_DHR-VX3_WA_IZ)j2(dAeRSIr>S
zFg)!M3}A3R@DEF!u6Y;Jl+t(gJx7c}21<ZD6(%=a<lE`>X67%dsw!htRc?bU(hO_?
zuD+XrCYG6*8G8P&0gF%pZSd^qam~6z^-05Vc2*Gn9#v#7J>pW*%8Sco_mtOH*v`pL
zqtwwrMEPJwgnlN|Yh;;~k(J_s+BF%mBMLPwtB9;D7kx3V?<7s6q&cU<<zX8HPFVJI
zEz`AwZNUcxy%G=7nCRFWUba){wg9WBt>=U0Jhj=w=+C~G6=Y<9Ie4wrNsHJ)0|<dj
z9=u(yPl{_Xpq>QuJ3<n0m<y&(vGvJcji_Pn{f9bmEb7kx<~9^7?jg+IzJUnVi|m6m
zn}r`SiJ+Fiey0XAJ6znA-*n3FYLaeEIsctY*#gOxjna_%naxbKL-|@}AWZfEC=7%q
zD5;Ml?>L~K+=W4%S$D{gKof<BK~q@<!*c9iT(lws^&Pp{ri%f;$^x)w1F=+h4>J^`
zFx+(q!T#wdWY~eq7693Y{GJAr;@*b|5N46)++6YPYhmLj@eIg6gUl^%p6ajkO4_6E
ze<uJ>2iq#3=qKcSrKaF|J3NiFEoA-n%}iC7cY8%1c&lL2mDgl`%i5akTlqn4G}yEt
zqil-+2>9P3fI&y9|MCC4M8-~6=AO3Q%@gyuYxhJEp;7xb{o7|1E8S0#UoU?%%0u`0
z#7^pm{_1;BCx<C2yC$CL>Bu@JfFxVn%~_ouM$e!e2KqT=TSQ(Yi-wkbx<2KUuCgrO
z&p&YzBqQo(o&IrAjUhF;=5EXUK<^T#SVK$&GF3#>0qQoUY5$iIp~rKZuM_v!>N6{J
zl#MX=dwPB#drnU^2y1V2UF!RIUXs5uMyc>^U$m{SvU<@W1hN;VA3@j0wBeD<#^q9=
zErz^4XQv|wlpNELGp!YaS%nEvk*mz;iVwFS=Ja!8;UAnB^VT-8(X+zW$yVlXKJ%8C
zX{)RAXyF#=m9|a{vCy(c$<je$+Jx4P(D61OOMVs>;`D7<&yOJ)XXl_tK_4G|9rK-b
z+S-OMY?XWUuSB0nMMQ5GP2YLiEF^wnPk8GV`^oUh$?nw3WA}Ha@|${n8^8FIZE1uP
z<Y&uaNNp~>;!qYO4})1z{`I8fPPy6L3EWorGGgOkI5;OB{=R=_Z4ygikqQ^h!Blo{
zdt%9COj|H@w%8d*NS{LLIG%0W`ad%-(^+Tx@vppI&+B4h)W<Znn_Q00XMWOWw%ene
z>@0Rf$Al<IDyb=wkBt`R2R(GTAP0k^@Gk43J}{A7);IFdo#HotldGwnK<xwGuzh4g
zwkIt{5G<IMs_!9eZ>`r9n{A4Om6yxVpYSx#y7iTau2oBtg#dYKuPM&Msj!3Tn=89V
z;_4C?>$QtvpF}vy?H^#S^*z*;`;4`R{ptMVcy$U-LJwD11oa5~!oTNK+Z2=gp`26v
z$N&SrtTirvZDt8n{!7i0$|}qn+x@-a6-bf=XRCg;<OB|P>&@JB$X#|;NR?0!85-hz
zTgbXmBq;R=mu&N`GEk89u3k~}xq6sPi}5$p{uT=Qrp-kQ<*He#e3s6exl2L%4=vWf
z0cTD6$5G_^5T)rA{rKm$JSF0LYtt!-@hamj*;mKnWNbxj#BV2>8~q4`J72XtdONm=
zv8Nu0yl`*QZbFuoKrDmG8r;u5aEYA{rX|J^6nQg+-VYQ*k~_zIyvC~8MM_yQVI{@y
z<X>G&At@S(uG^*6hEMso=#PO&j$14aJwcbA8+jPYY(wh#7e-qCVS}K1Qr$X(Dn6nt
z7BTPEvFo4nkalk@Ci9zD&2H6>f9q?ALBV{wrOx~}ov#bhSCx$OP(N7<$hwmx5ew7c
zK2+WE`+^njk%Nck)A7mX4$X`#qf8W)^3fM>cJG%oxFVN3|6VKZ%#6sQ(T&XMY9>9~
zG~G|GcBROaRN9`fevVl;A7{!d-<Bt}&Jk&2jzy6DZF5hg{^PbFF1pxs<8v&mL(D@b
zhyj6r6Aul6$JIGU2L&ZBox@0l)UysS!5b?xP#s9VsCB&~kNpn3z$nuVn3zH?6*N=m
zswW*d&l7Cg<a(Bb`;El@(CWObeArVr6P(3b!2QF}mfr~#KjQcBcZrbnd`b1y%%!?X
zhw_x9(A#oNGtP`dq|!Z0g&Qqtk7-sMLKnFN5*X{5oFlh1Tp4TO18^Dt&Y4um|Ev1&
zvjCl3M0SKZHH%-3g%@dkE5Z1t##1<z8yuB;>AI_~0yGzd@)?WYwrk*knvl6Tw(N_e
z3~G05GnHi24u)&0)d`S{f)`frxv`7rt7DmY7=5lJpD)@Pk?*}#%`yHiDXb}uwTFZ<
zv-fup=?&YXJyTf&mQO!c4F-DUHe`xTpUIQIp6Jlo>*=hi-O47k52p1G#4W3=N`lAu
z+U=Iy%iY~wG;HgkN2P-VIg6F&UYwwKakP6!!sVeuumV8d_Z{Q;DPqsdx?-K`C5ulV
ze$RetXqbWl2GX(g?5g9<r8Ga0V=l(^F9uf==dVaWu=BHb@?>GlBB}$6lq1dL=<52Z
z&8!3Hc-EBL;aR*%j8n52J3qDPgh7q-qwT?Ob=v+KEPY+QsZY%m?d-tXoS&t)S4)rn
zJ5xbEHaV}RnqKMKc`{AScrcJ%IQooov@<=Td(j)%w(z^+vqkxR)za~}e)6^r-&=_9
zHb%4Qi;)JWlTV9&5ZIHBpgSaA&+?9q>tLh#Ag~Z~3d7ZLHmQ0|bmD%BPfGhqwSN+V
z=kROXUYo5BV!a6giiPk9g5781FR$E1-e)^?b$6<$mg@U8KIFY-=jAn(ep8qJfxM3d
z#fvd@=F{Gzw7^Ai;rA#A1a2!l0hUkS{^U%$g_sM2gyMwANb>W2xfibo`inmAc2Gq~
zP9(fxEk?aWt#p28*xj8l8b?3b+ZLbo?X9u-?m=Z$_J=FCxY!sl6RdT1w3<x0^nzSu
zh;K~sZMZyT)O&<`j<Y^mtBfi|@@RJh^Mpb}SJ%qQVz4DkSNr4+0&z|7Nc_{a-UShS
zOT+$irU}!1a-{K8NQ4IWx2ozy-Tk?^U0dga+$ib{ww)r~hWd9F8`xi4kj^Wd?t7Vu
zQk53?yQ|}zRBd;rnjG#@T7jtq;?%}s4P=6F6QqjJ5HWUgPS)Lr|3e_=PLQ9Fz<<;7
zU<Noag!Sn5e=ol{_)d`k-@S-}EFdd64PN~JUcO|)&A0{MCHTnFS>(Rt`1dyvh~0u0
zpW)0Q%<(_G=u}dBv!VCj%kO82TFv3R_jCr`91Qm~aPa<n`TxC#sIqQlVf}xvA-v4~
z=Z#4u(!;sAABiLV_p%pd0AP*?L<j-8*c+Cms%`ir@O7H}@_&9GS&X!S<naIdD9$MV
z^HJ>9^B)KNJ0Jw&QtN-;_a$8f7~R6j`|ssB9@PK&qfd}Ci}Hs?lK*=Z(aih5zYjM!
z;-f%$q5Hqjia>Zd{O_Bjf6avKmjB<IP=ep2HbLZnhl+Twzug}+Xl`k7xV<(Em+fS1
zLSuYVtJ;a(i%`N{+fwF0Ew6jvMR8cF4TX@t8NENnRt99uuS^!syTMGBXM@1Rx=)48
zN<i3;ffqX$>RVQYx|7uxe>``$Co~n*wf75(LZ5C&CKentdpvCr;D<rQYH3wq<eU{K
zOlaHG@<K#u_{Js=k2hvA>f=5%Sr5;eTN;^|8L3vc{%Y+km&qC7)1_<oD^&W^Z@;tp
z=8rVRUwwaBy1at_cR70^v`=9i(zmJ6bJjZmZrP^7uq->2b2n-~Z1lQ|65pc%i;8rS
z^Tn+ToYdu|`)S2E_VA9Px1X#ztizIleEZf+$Jjucx&j4XG7&lP70u}#w%lr#lmF8K
z?2M&1rhm9T3s(1~!V}Kb9%bXLbxmC$@UG(ABHQ{_$kN0pE7#+4y$}<oexD<djQFaw
zQhqKV{Zp<Ay721JB+lMlshW}k|N4wZEMMWS-ri>WE4n!Cke)v(npYQxDk`re&y(y|
zW~i;5xP?2rfBdRV{(1DBpD-o{jna%B72Wr8{ixp5vAw?<P(71^DHrez1<D3|eQC2y
zb_7L#*N0WbE+-L3?M!+9e(@YWn=)Of>*-%j%jDhEubG)31fO4;_u_r@n3nr0QGHq`
znUIiBui0}pIv<LWrbaiN8jr0UzJcnvo$p?a^q1V)N>o3?VbJX{4<{bFZ*P6K-28Oq
z!AkUD#o^~V->CKwfp_aV^459;%m#T37e}WOnA=)&TixAEKH)DOy;4)tYIa+z7!URl
zHo$PN_t+Cqzn=p$2Y7*^9PvpBv4gEV=Pph#f&~j<BtAcB0XAD-OTS3s{=3^B9@|ZI
zX3+PIT)b(Xc@7JZthBtv{2=$IVujJa{duJ~TQ{H3?g+=)Zy>Kt1WB@3Z?OG$%fD~0
zbR?4t+t#~sJNokMov}8BWR#DO9~SwO^ra)*0A#)pI#$-_UWc|$PO3UJ&K>PD&+8@H
zTE*L1vDataJC?a?8jSpp>9Kd{aM2(mJgOsXro@er&Dp_9$YQFKEt!8ozuwhVL+cB7
z1}A9O<TC~>D&(!%1ocu}5~b56>n4Ke!|7<nat8e{S1rLVQ--KFk=JU!rZqF#nSx4Z
zf<KwBEKEE~7;SB1O^?q(MMRgCa;-eoruyeam2#enPdH5co(}#*J>K!1-Ja@8r=?~I
zXdNL&BO)ZIzu650Lm1ZLe5?c2DfeCDn|T6W*R!LhI`}~)X*EwN6DPr<{KJVGm5m|C
z^XNb$Y%*R&Wo2f@%X_Ic@MlW&iIQ8+PtVF!qMO<2FMs_R_%paqF&5UbDC5=Ylw#4Z
zSCikaw70)G^`l5yW*9)n`|SGFt`{>Q+Rmuct^<)1Z2I-Q4%b6;+usVa2B#+vB#Pw|
z8&~0q6Orp4uxOTE|BVl}8gd+d63=05FLNW6VI?Ns2CMsjm*$$~cKWVJ>ump*=wh5T
zEXy^D<mVm@?<k(ty&WV|jg<!LGvR=iaJ_oNLCD1D2qVuf&WQ^D(qMIR>vBvyJ=`CQ
zc&w|}SCqz2LOfN#J))<Od|RlbtZe_dvDu$KTr!h&L*07PGc7G`pk>B%Z0^?jn{QJM
zmk#%l3=J1|1*6)-q$OoR_>}P6v+UcqUt9G!?E3ZNFN7X$TKVA+zsJQ#7Xxq5h-jSE
z4SE=Wpi+bz580yyW+3$Z8m$g<Yp2$cB;+FlTGKzB&*O8e#uipC&EYLN*dO2t!wn1%
zLv&0`UcO=f7e30zt20HO{&>PRv;N?R|9xVJEbLlRTAp}bXm_prK5H#9KoK3EsF(Ya
zx|I|KZT}iitDkwagn#gKe|NHMTMKUB!2y^1W_68)M<tX<T)$NTw(iqM54b+fW3+lG
z<VjBXcV<vkX<>1z)Z?hVgl7j&n3DZREPF~pcDHnYX99<TqwHqRlI2rMA96u#><0Vx
zjslFPZmCpdl*7Gx+cT|c|1*?VdA}omqSRK|NYLQU_)DY+I@i`^qcqdpLzloN{at7o
z*WH~D1C)~Ozn}5*p5m<MbSpuP)a<gvNA8xEQOJ8xEOij~?Z*Ws1;_iJOEV3g%^$K&
z9(-mb=z!<L)YP=u>m(g?9O%A9TazIa&t;{9+*Ym^ne}aCxK&5KL|nsw{M4ImFq$Ft
ze3mn8ZX|wQZL<Pa!Lz;+<j?J5I~KWju!%WJip!&(3o|_7&Dws94wk@<ue?t}i)io}
zshF03kITHpMy34S(eM1aS&xO#gv{)7P~datCuxIfZWm7M3=D2pRZ?bwFCJkM4|MdD
zmq$mdh^qWrJBaJ)?#2^tbUz<3ugGUQp1b7<No8j}qbbcUzOzGlpLzx;uk!A7FIn<{
z?0g?17)~N+uoFkv%DH$FxxD_)-D<=W)^>i#z`zha#S3%4C8cGZ-Cg6gE3Xvoq>ywY
zV`5mD*aW@vuBpLEIqvptiBlYKq^}vwpvDgrK{*BqSdLOI+%;g;z{tk-5)B|UWo3~N
zNgDYF{9mi8h$!#bQm^*b@1o>$@NmXIHm&*C;2~+QF&&T@7AL_|%!ZZ@h9ZUhC!?M+
zvwBU?iV$Opzj14ao35va8CN)&-{}FjN6E(qusOCM*1W)Z*m1!gFC?R>iAU+UGe2B6
zD+bBf84{SK<)uYSdnwOx8bM=?ABZaq0HfFLv$1Jl6=uL3-+Z*`r%MUGuQ2&jmltt;
z*2eg}=ZiTa9rAY1baejs{gDALd@#6v?9qwfh;KNVQzXMle&TdHh4^G)DN)<``kIC#
zzAVEJ$708r9TyEATf3zRIf3JFQ5x*c*~!Ur&Oq6qt7zNH->w2=JNju^>E}CKzw6fJ
zL){P9rY3FDDk>bcozefTA9D(dii#~O7&OgH%#7Six~xHi8k%1i8JS>tcXfZay}jkS
z{DaU3H{u2K4v`TN<4qecSi`E#F8uBqiEj#d+@|E3@=p7d+?(%|Xz6<Sg!(d$HL3jV
zkvNH5_fUiG^n;Mls_gP8Pfd&Xz|#3+Im@cevHQp8x~OsQfChTVX@}xPo+c%Aw6)Zu
zq<FUmFFHP^cix_K`^_7z_V@#JOsqpwg;bqt9SQB8jB@R{LU-`l-x!~EKPvM@@(=r|
zz%7&&)LWF7w>H*73%G?OO?PjPexm`Yu%zP4momzWXTgmw)zRQr^^FVy?a1U@?vt<0
zZ;$j17O9Hc{w{E!A3cb(UuM_jPSY>ODu}~=z!%$7_>|-KY^llCu`a5xext?CW{uIo
zgq<=xdrMpy*I&OHjvmVcfZ*aG7}l>}YmLEHU0N0HgOoqEd4d-adR%qNw6bzPHahz3
z>|6%}^NlnCJcEhx4+=HP0XdK*<F|*t%kru9Pj*-5E&GK-U)NxDtP_eFi>>|ZYa<tr
zrpm0d*du23qDeA5!r``QtCbO_MY?|3{!XBj*!n^v--KXdbu*LH4sd464-5`!RlUjh
zjA^v?aXk0#7yDmDx_qpfwR-9j?vw3*ppb%=L5VRsS$m|-D=1aaEFvgqzbd3aOjz;f
za5h~isfCSYqy02d`YYJB#RbE>|F*_2Pa_rUH3ut5?jP-Oa6T#07Bqg{e7r@bRyeDr
zt<C@DG@MXonN_Qai1JWSKuKR=a&YGO*BLWf`hF1Ia}?)|i6zA-de@ML+!g9LskOms
z(dsJlsT^Kc`LsYC3Zq%Pd2{;ipuzp7Yk_d1L10RqRTuYus)mNjFC|62di%X9-HX{K
z911?;7tS{Q8y~9igl~VCgZCeaB~tjP?-3;P0!{UIZ8kh+<8=<R@dt70?D{n>*T3L2
zau3jI632HlRpm3u$cQwSl`-+s9wj~`7eG%r8x=+JoEA5umm}dRGeV6cwB(6VSLr*q
zT&Th-XqStpB?`8gX#xuMuepu;miQ|RO<rLJ{+#qYjiI4KD^j)T8YiCJ<LByTIe0%&
zqzoBgm2o}0!?u+?HLbJJWU@MUjdXFORc|5j^5uM|!KXg5$sHkkLnpc3oYU6(>BXv=
zEFN-uts*RsSJ`5uli_Fz`PSSeMHk?ah*VZG`=$N$5qq=676)H)s7OWCNG$fdr<2YU
zH7-l-`(h5@{}EE0e3J$+n>(v^V`)(Dc$LAQ&v^t-ZEp4(b3n{xF(oM`<_qDhswyhd
zbZOJxhbjs36#Q>Arpd(7(vO(q0$RV0W5f=x9Xb^d^*nxXy1cWsvM~+Zo*aLE{unHt
z<29yh6}-WTE<KgVzQQ9aH_e~_^(}n-<INYif!<5@1+)%||8+p$oAaU)dw9GLuO)-C
zgAP}>ACm1WJ;D`mJ^nG8!_0Os$OkQ5*^HeoC)3o_WEPZ6k@sO0Zp^Kr5o#`}sWI+I
z`P8FkBZ`)lWwy|Cyt-ZKKIz8fYnWCXmmJlzc;J5Oh~;^9KEF7}^z>pr$}B>Wg4O|x
ziHT{l+Cit;WChk|>I|W$3F7Gu4ODbb*?4KU673@-qvP~l)~B*y0q0y{A70~fs$9Ng
z?VGn%FP?qsXnXUW*$gbgcLr;z^|DmuVtjEIFs`vv?jn2W1mFrA?qwD5h+{td9a%6?
z0)=o~(F_9%lc4)Ny0RLP!rir*Ld&rY=M!?lJj3!-Wd8=7u(!uFq1>cZlj9nwx<bu%
zy=61cJsIwC2uo@u^hL)Ml5*VJy%U)M<?ekpeNw&}ia=3V2LLhk6>9@`1+g|-5U$4{
z1GYe6w0Qf|HW3kFYW=;mv`58hkHT=*|NNw<IaFOO;3304*}m!sluE64s~OcUuW3QX
z0urR$-QVGU@$&K#J#EeuTHR#yj8fk3PZA1*gmE4L0qCAb6P1mCCH9R-ztpqdD7soA
zjOj{MR!FV?T~Ipp=YH{wM#)<_Q?+-{Vhkjs#u8=s*1teY-%OhNGFEeW`Ef_#t(}FY
zm)A?kg@Ucm?mY@+bYfg;RAT?Ut&!m()O{S~Wu#UgW$8e>(?(CrQ!3+m>EXO16d9YM
zsixNCIeH6d>4#QwKH-uKOsuRzkYfVDbiT}^cJz3{a0_Rd8PB+%W@c7wuI_On-~Qoh
z069>yZ<Ag;0=L8OjfMQrHkP+8c7A4{qX!ouBa@J<@=VbZ6Z@q%%@na(`d{Eo)tYQ&
zSN0Hp_Wt|~JiUJPutWO8n;LSeh_R0Tlh$~Q(^a6!tc`PRjMKt!rDe}Q6UevG(#+4q
z#>Q~JiYct54V2|`-+QB}dDiH!fRBG8*a%nUNvWOYY?~_T;Ss;<v7w{vTd$Y%DOVCW
z5BbH?m2idI+Gy9}H6e?(`l7av$3S>!1x{Bt)4OT6Z!En|72GG}0&04C5wR(D+e$AK
zKqwUXilOdgt}D@b9<Dl6E42DM&f7vMyf$kq(G?#*{;>S;GIrU!f9hLqu475LW{L69
zuo=6csOXYsBcV`EW?O4}OiY}&gEk-ESdm_HWo88nKmEyEL{ubBaJHY8rluxbKMT$A
z`9-tqWCWI=F?Xze&xQp7L3`*cl~?Bf3%gtn{{8anjzji5?#qk>p-^#Y>F%Ck&+&Y$
zd+29d7rR*)rka{x+S+N+e5rhJQ%-zv7Y<G%CH8{Vt-tA(GOV7T1$2Ia=JJkz!@b}(
zi#1}c1QNiMyeXRVnW~~Pl2eMF_PONe6lCNSIGFFmnb||MnZQc~B@KRXq)^U?r?~}r
z`YVQbV#<4g(5@07q0VApxgS7`s|*GwJ?dn<w;Y*3TK_Ls8Qa!%QkQPTyOWlvJtOfi
z2@kGBWrQ-)=5{bX;Qo-vDK4-wHh;uh3N|@T^@qv}`1leEN!EtObFXdqjbG0ih@i-(
zme&OJ=B=oVT2OJE8)<6Rj^YT@&pz{JeBHY*_VU)sj|@Yo$go_`wD7eV8CbrptR6b*
zHJ<U`k~TH1DymXz<mMsEh%oBz?(UrHvP4TyN}@AKEzHdg7jRkmyPTNtfXe64(J9kh
z35-mBEXisTs}dE=yGhavWtLLQ%6$$L{g_0?$@;w9^?vTHIHcR7g*qM==e_~CUslo$
z{`KFUq@7Ppx{U4Z>@1Yu#Tq;Y<xEEi$?}a;e%b>{5u__9QPN+*Hpo)T=Ft*ogOmO3
z+uxRyj+LCuY>dPK)vjvXuznk^_|AG9JlNI&@-{5h0C&fmBGykAC%jJEO4FV}!7|Y#
z+S5{siXXb7<@!5+G{X&F%1id}Vc`zyGMBWnNJdGCMx<NA;AMv+i5x^(U2PQpSUjK*
zc;jKJ)_8d2DBaUt(1c>m!C9`!)XkNkA}3e)is4tB$M)%Eo&FanIjc&`9z{rUaD2|^
zdr1)?d5;V~$ZUuk-G39}Lq2$CNl7tw_AoMZ#>K?wH@lim=hHe{>jVURzUz6k3~6?9
zwOCs6(m~QW!`HrkUUy_1$}w`VxA;8-1URzGzAAlw<nZ`kRr<YILN4Y{wK~qMwXTWr
z133?w10$|v{yOR5HEo`6?ah&rqWY;trFM5u&!lPcS8!SYI8h|Xj*lm)b|^`F{LYw%
zJq$o)Q<Ie-Jg8u@l^A$zW;e!KAiyvuGmF{>`S!g3Zmxq9hp26Hp9OxfV#YyYf-nnF
zaAM;`o3p{=4=->Z5|BP5a6dXP+_<mcWak_ldgg=s!;;bfCSxlOcf5I)_gqGiY;Y(z
z($1#MUo~2~0;s#&X)dsFtDxvQ@7C_kGh5h`!Vh^V*A8MVdHTsFfjkQ{%IR|`C}`$=
zlVmFj?Cyq<1oqV+m9!JBxBSB=*7=a%k@0C7a?I=8YKk@L&n$<J$R6rHdW3&X{z#k#
z@6J(Sn+OdasXXsxZ#;EG0@vy{>l7`ji1#9~w?9BpFUxU+#BG4qzg@T?<Q0@yZ!Tt_
zxCh0^M74uxAuBbf<zeeuHk~+dCX3=&M_bF|yKmF7Qw#j&DIc$}-J$Wm=6|cGqx2CM
zpW3d`jw(V!hv$OZSJ*;2Y9ZD2y~1@8)0%>m#khY!r`yO^I$V;gqy<~bQh6IG(a4fM
z$+}d&Vkd7(`~WID5eXlm&B92@#Ez9wtOEimn!}?Xlqxqy*V)`X-AqhGY}$!&`g1Fj
zR;zPfPS_$WU2?+=A6_urL$BT((>{@1yi>Kmk3lD!Z1vW7?(gdrp<I2N=56m@t<&Hp
z-VG)7EANM*pPs~*^$_)0{2#bXXK}56@Z7VJU9Wa~u#x>*v&mFh9mko;cj!BUT*g7!
z!otl%cxXIWJ?@*%XsL+%?GNB`0C)M7n<HH0>5E-Seo|)DSafoU14g!&5)z!$oESH+
z@}k$**}Dwes0RC=-@krA;7=+gDKoq?+whs-$%3m$q)rZ4kkNX2@-5gGxOxcs$}+Mr
zg$64eZcg2e+t>hWGz-D26|GvB!nCv4@mX1u?0gcR;fHlCwrIJqK;Trt#aUj~oTSk`
zOMfiUM=k^iA-iyZbQbt$#Ky)_`jBl}6(#Yr$>l6UBTbq3lTw)hEgcqPR}bB?pqB7=
zjwPOuxY7wr0arqFNadY3pJTFk<)oROvd#ZG>tP<X+Wxyd)Hz5?BLy?u_<^y@rgnRQ
zXq1Af3XgCzGc$38Y4E#VDCErNj0eH0*Ue-^jdWQM5@Vv=SC~Ym9Fz628j^36hllp=
zsw=7dwXHq%8T~!7e|Cw&6k+h#v4Y!Xs!m%&<uh(sb>%f-k63(SeCOi_cLTdt*zd6D
zRE@Na=WV|xFdQ~iigajO+}qIAS#$rx&7Ehl23_*$TVqp`o+l=I4^J~<zLb?au>6wU
zTak&$GPHhkDd$tM^x^Ie05=y-alf;33ZibDe3E^F%9QLr1^Vyhb@ujDpShuutZr~T
z7;*_3?(gpW1<_6bcc4+im8~uC&aA3>O3OXnGDCF3#40yiV>&*|$?Uz%(ne1Mxn)u_
zmR1%i(MkJG$A(v<ol{d>t9>E@Q(ZG7`P=FRr)II9@{+P()((Cgv*ZBgf}f+mKk(Ck
zm%g}MFK_zxSbbypQdZXctW?NXmde61zq*0^-?{}<$k+Hqa}`xhXam8|CMvqY>q3u(
z$l1~1uKU4Hr+5C?#E-<XP^QrEgT)%%PN9%&y+KZhP5Foae&!7ZFzoW9uqCs@rLjBk
zZC;*kzmj>6+ImKpP1~VFPi%HE5gs!0@v^RY8ZX<vcL&^ms3M+~=(GF05;nAdt8@Fj
z>j74r_7x@7o34+L-#2}@znK&s{h2=92dFL{x6_u!N$csRFtptIIQp;-3|C@5Tt2V!
zA0t@IVpNKzYN0o2O#nS8$lOTu4FXQ(j+279K3i|yt;);0{P%7~!B|pjW`Ch=^3B?D
zqr}zuVG_T?8%gsod4YoiEm!y1bP3`H28@h5drI;MI>0-CpHzlC>_m_Xo$~O>)6k#u
zNqluXKVmEe$cr1JGCrhiw-?sq2}6a<U+29tK^RQ+)zIzeP)e@n)U<^qLeg<<J>;xB
z$Mc#^YJf)^hk^wc9m!@|4DD!R`e5rq+;0YXrp{quH{}WSRfdruG}E$KN+4Pd#qI9x
zWoKg2<FXOoJs4n1o*EdfRfM|bSzJkNULZJYG(87IN>xYce6y{`F2fH%7m!HD9m|J^
zAEKeDs*nCkJW~H2qU~14cU(2%ADSYJMoX6im9-XtsI|ejWHsaWjQmhhrtAOdDc21g
zp2chOQeUFZc<#3L3@2;r8W|cx>PPhMVq(k8$i%=49LmhE-(W_S78kwEaPKJ5nd|o-
zzM2XVG)>Kfyu7mA!A6&}x{};6@7MakhBA*@-}Mz1(%%Wleb7~ru$iPoE%Isp8mS;Z
zKl&bFbNv3n^YmAzytIFQ{jS-buz!>PE%f08wGZ$lDw<3+YXc?KckcM+^y%txIipn9
z@d-7jIrncTB&-u;{}NyIr|59UNev~QZ}#dPIz6SaWDbt5IIJ6`;^a50xj~4Dxmxf&
ze@<QQN?ZdFtU60Fz>{5_o&TsI<J@MeQK1o;$hbIVWfNLDo@-YDhwJXlJ?ernGHic-
zuV1_H*7vxH#`8uy6!rjwoYGX@5%QS$2QO1CJcw-|BcKm2d(0DK)V)A(At=B^&AIV6
z9z%xc*Rt)ivpc>&uTb|Pv~0N6^$dOAo}eS-S3mnJC8Y+pEu!kWiT3W>IQNy5P42Sx
z+N$zcPs_>ldmJrK@3%ShCpG^v%me1M#H6Fi&z+Pa^_j~H3ZK@8I!RfXnQ&4d(|GVn
z=_+bQJJf055{f5W$EWq@4|mO4{RstF$LT?((zXEja0)?S8*IrTAu?m7E02V8ZP@j7
zA8uj96(6`XSnJl)(9o)@sd6v2UX$HEzmfZ2TJm}){P5^VB8+@GS{EP$^>e4XpwKE)
zFRu^AnmW(|(Ln1;goeBpKTbo%D8NKQze9r`2pu56#<;?O&J0;pU~eS@(zS>mdHj&i
zZ-v{o<`49UAZNi97NezIS|WrM9*xr4=_qRMK3I)nA3v4+{PgWOqo#X+Bn?qqEcX+y
zmsEJuIukO(Yr^S|>7L3b$j4<5%Sg+>HI$+;wD$TyV=kN>4sEkN$*;uS^&@_c__x+?
zBEMX^g7?$uW6L*hKz$Kz)SW;=nwOvN5u~8g?21ZG5+0R!axIb>kuE4BcD7bNu$a@o
zNKeEOA03k%>>!(&Wi~RhPnBV~TXWY3_rrT1VMSv{+2Pb}h*6T)^OAluAtY^o08yv5
zAufSS>+1>Y2?_N5daG-1_$Mau@A?1%l4HgdxlO~dZe(JTRT(3uD7AJBt$}pUu2}QM
zT%aTk1i%drsq`~5=#<-$%HLMfQzYK!kd={928L(ZXlmQj-_li}%lBt85QMOy11Lir
zk7bhOxwjtz8Iv;Z^nm%ShNWUZ<PWhS<p_+9h)+(A3Rbv(3-1SYhO2;{D&d7@Co^=b
zsQih*Y5;q&aFlp@yX7K)0kZtyFKatfw0c}d>%aOE5nuTAtFegzY{zlY(SQZcO;WEp
zF!}$y8}>O{N~*S&55xDP5j_~?x<t11Y=~utA7G%v{V-ZyUmaAozl|h8!x*L4;1Pmr
z@8D#b$B-)EuqS1NQPk{0Y-(U_Xr6Z8apmRrkVh64b&+2UH%Hsrn$3)JiwiPX<#KYe
z%)mn$s$VWXRw@p0Ky2H)LCVF>%G}@i>xV>KWP+Y*8@BsZsBro_TRRSp0R>t4@$#Eo
zntBOn^2JT7Vrt_dB-KDsQAb!QqnaW{dJQ0k$EzusqKb+$?cGZ7UhKt^GX1S`IoVn2
zYH9$@czq7Bob7A*_4^r{pxs8_l}j3OpePFiOYGFodmOY(bXEC5;JvXU=mIwu>|2n_
z>9{>c$<MQ6^p<Rf^{>g7*lfShY`O^1^LwcQ&DovC`#Gng0>46DJhHEy^1J_~|5rD*
z=TUo0U;7Jvc4}tJj_psgf4zZF2<wV&D;VWwS#yvESw><(d1?Cf;#@c|PBa8H0m!$7
z8K^k?M(18ILBM^J+ttjEU%zKY_%?g%!{dsBgJeLa1uaA9@w*{MKL4L`=Se{dZ{Jy4
zOG#O{nX0KAN&pN@K>ew6bu8H(Te`@?!eYE*%!OHolUB?(Xuo?*VNW<hQeQzrUqL-V
z9?mWkD|3e`@NEHY3j^vlk~D@L*lb|eDiN9*(JJ0$tR$>t4?T}?z>7z}RYpa{y+zS{
z8$7H4c`$r?%JTOQuph04rLoz3UV)-i*=EL`4uX|j+_IlBv(^rirY?W{Fr0GymGewc
zfA?BMRMXWr(bC<DwKol{;!u$nsv9nF8Sw-8>i9@Wi8IZNt!5ow?H)vRnsZI;QX-M?
zZ3!DVzWqv5BuazN|7`MVGeG8lixB_0`FT9DsmzAR40?GR$4>Au>=(_DnyztlTdbO%
z#r*PykcP@`GuHqo71_*W1;~W^3KZiNRwo6cJ6d`U1d5E&mT~giytbPbMlIaC6!h%u
z&R_wI@-8>8@VPs$B);14E9JGVjq#Dj{}EPr2Q9B)!nc752@_Wm^0J8nf^d;892j#+
z>F)&s3l0XOz$j<}gJo>|ly=gQnFO#&Q?<T>@VJ|M_ElT|rv)(0NbAK5L}rz%{D1O}
zPygg31AN=hU4EE402v72leXR@&oSAv^If}=?Ckd8m2^PS5*lS)IK#XC0@Po?V|~TX
z{QT1mHh$oLf{w=jIWOM0{{$UcSn~A;!E>}TH&>tV(U5T{cr1h&PJ7XOg=5PMwNL&>
z@pAc6tM0W`Bc$tSn`qEPbvcdVT=#|th%&`!IcG4UAq!jT;^7C1X1$s?=i(xak-o~K
zjqd{LdT1!oU-(QZ&^V(vtkks{%lzqOAdpj$AywOW=5!FhsCR|+-@&~1me>#y7%*4r
zCr4giTd(NaV@ct0X#aiJYUr>TWPTmxs2?!>DvXgRMd_j-jb!&JbOz~=>p5ty6zNR7
zx*_O}n_HMQzK<<S!m|(drp*fjw$&Tgxp`$Ip7#O39sU2P`U|KkyR~~9zCl4i8fm0M
zT3T8G>24$yknV0JrAxX*y1To(yFsM88@}Z^?-~E!x5pkspZ&mg?>pAIt~uv5XTg&!
zMWi@*`#)7_d*f4Mj|$0OF3<ZKs*A1GzMHEznFD^nfy*<Gmx~EFzHl{$mxRxNf7G6!
zs6kGx@%zBklrs2t!S?wi#E_=~5^lk&{Ye8bj0*~Aj4~HSXs{_M`y3q2=9LcN@7YXD
z-v3YkQ*x^^@>PVqP<j4#);vs{5;R9tga~A>Er#r9Pz4ZCHL5)6g>t>pQrR98YeD5R
zTFP~;T~~KAvBl6W^`&OCh!-vOA+a(Kw5kDD7R)8N=)uhlE}A+D++P+a{Z|Y>f8{Z~
zUIlQ!$5WIzWukylnSqK!;M4$ap;iMoJ@3_BjaD6wc=Ye8<M;Wq=6sXl<zt((=GE`t
zfLXK872M9x;&k~ud`Grk`UznsoRgnl6h8!T6Ip6HReMTj@87CORaUVB{0+2kC(c#D
zu_XgT16F&UQ$WBP`ZVyhf~v*)iT4$fhEmgsILm)8tfJfm76O&npvd$`F0++SiAkq$
zu*poyr0Dfd_ETfix3A*R@gEe*!vSmumJdqIori$lzdw~F>Iw@(jJ9O$6O~PG1y#+{
z72l-r)aMuE0cbbC*B{ZZJ=;l9N~xoB_U+#;`sAJHtl#^~Wj?QBP6f#6Xu09YGIM_c
z01WG%PdUm<StiEN^vf$u0)KQ63ZcEYq~0$HBAn&YnpRJ4LaVp+jP$>n4z8c*G<n9w
z>3(~Ojs1#&QCe8HiG_{Bv9tmRQNGt4{+Tys&l^rzRQ?$Q_HFQ^GsXmyk{}Cp6Rdx9
z;Lo)mUApb~Sy_g|vZ%JwUq2dV>7tSJRTLomtc@=tT1AUL;kkmeD!vALWkvZn^d-)>
zeBdyM>)1Nlz8#&&%PaB#FPu7?m>d5pC<twKTeY_>9T*tNk$(fKH9)ea3ZNAg6?HMM
zVGVEClJI*hH>U`Hv6Zy3k?45w&U*Wu#Pbioqm!h}wY;<I6N{=0bv~s3_5Ku}aCv;9
zsH%8+x*vkx-!}xfD<K4+Pc0@sylw(uS(K{*UI;4F0rL69qE>RS)Bw8BNNT{J{6XnZ
zuo|Na85(Y^R2Jv=+dA0+v~p#!4B6j)hAAa;=X`KK8R!70HdIv1>0-RmLDaDfQQ9A4
zD>KtC%9@&YKsW>%w!S81Wi7B9@i<!N<rX4);?Qxk_4i9^^El)d=egXoKMy1<X^9$o
z#N@cHCo8j@2m@DqJ<t3AE}(I2E_5q!{@7}0;D-;tHZ(NA#EkBs$goMTj`a^Jw{;g%
zyt35J%KZsI`^nHKAOy?JDbQ9@BE09T40>>)2MN&!3*1Q0y}{o1vYGfb5O9yae*yvX
zoT@m}v73AXPoGvT`X26^nmwDruGJwbR}5mS74sIjfYnyJ7uXDxk-D0c0^&YfnVOnn
z2mviEFqPn;Ru$6>7L5kgHW1>7$)#QGjuizBx!{~C>V%n?8?<hyYJQ3bj985a?~B-=
zQy@0Dc83uX*9n`O>HrpU0I_Ux!DZSB53bPvB&zD-@)0eq7`!hzD>W0@Slh7w(AugM
z=3rHmlB}gC;32S+d`38vI=Tn}JlN_LD*r}NpB5<O{&R1&re!54!Gp$Urs&hDOsZdA
zbT{yK_;Ub6r*oe)Su?Y)4rgb(mQZkA9e3>i920Bc<Ff`<+6~OF3B#}20!hS8bO6v1
zV|;%L)(i)F+;v-c;oHBFO@Q1tGc)sOaNu{;Gn$|=1ncJI(LCrNkA<Ci#Tw|h!suz~
zNMS!u%x?gln-GFc?V?LXcKthE$z=}9^S6Aw{_jcm7s|$Jl4!n*S5}qrfmGSw-a;LW
zvc!U}Q{`|5m=d~jTZ@4{QQYqo6(@)GRPB#8>s$_OmGp%{y@>=aCeTRndY#JCaK+8f
z&jVrV(NbC27(TEmZ76RJumFuS6W@4+6pa#AM&kT@`|mGWz`+Ly&y!R3XBc#I3o|L_
z(nKhU=W3QZnkdN8=eg(81nT8fMd4f6y)G*y)<rQTC>a@tCd4j*LI0X93JM1(avwch
zWfR`kTgG*Vou{$s?`GO$^iqz`DhLDGo{{?r=xsSzCzTBI3U#Xt=GwVM;a!Is<~E0t
z>x~aXfRZhO_9x&HWbk+~<My_V3GjE=63PKn37(*%s!bLrU5%nTGXe>=QoHFBCI`v;
zeUwpIe)!AZk1u@pj7@lE=YV9y2&n2Ek{m6L>~97YTwJ)xBWoSj=>XbeQ}^4u9h@)e
zkWlpRZ@~_6kTgA6CradQGP?pNvdiHxK8qc^1;Af`mfO*w?HYS=2Vn@m7)oH-D3Mry
zV{nS_^r0Y@v2`Pnz!VfTAd*>4RdK%2@}rV%%H$4a&N7ps!PlZsEmPC;<FmaGS!O|o
z$3eRU8x{q~LD%2)&C)F~;T5}^o1n4!RHFqhbt5wm`zKU5IG6kJ_BuA|5PVtkNPv5R
zY~SF$4+uACyER@jU&$kTf5z~8*q8jRpzblS{M1k==2XH#kAsg-0A@Nll9>bBBEVhO
zra(s=UZb~@7`os{;fu!_!ZCo$YG`-@`pVjMv%mApCL2*V;AS37<TP1NGz54>K^!wX
z_zvTf>U>_)|5QxZ&K66h<vvmkV;uqK$0yx1m#205*Zv8DZu5mESX4z<yK8KvB`e>x
zlbZN=!QtNBF{SVYjio9>D$SgjhK9+&ZOru>z{5B=$QWX4lA=#ePAa#*OMH?8NdJ;V
z#(y`n`<IE_VbuH-N<oq;YWjdvbakl>5bJvN>J`E6UJ9QZ#u54h%i(<%<Qq{~C=EDz
zfJlP1mV1JRgxiz)?KZK{8wz^2p;NyZ%jLYHA_qGMS($;^s-w3;+W_x5e65Z6+*_UD
zaBpj1{Uo;d6WE6>7wg?NXFUPf?{g?%Ug>sfJE*NWlh$Ot(U!qnP&OFPb8@|N25=%_
z1Q4P_v|RJhUBs(pc~P`*b$4~ru}+GIY{HCo)$$*8L2kkPyt>25PVkF(dux*!;|z0D
zqmv*9=Qjc0n^WTg?#h(5FoXbgFO+Wmck}~A_e0mFHRatM+J2I*e68)lSHNl5U#)db
z1n@f8dp!P8c_w3ExK*rGgKt%+-k_l<OBtCe;5Js`Bo`ZGdweGE#D%M+qpVf#uqD6~
zI$hD=wB&kz;Q-?2sb$gv#18FuSIO}WCn_4QI!){Zax%$$g>;)YRn^u{BSuTXbJ=t0
z?qXM=hAv&6vLhn3+8xj>(&CcKRSpl}Lh%#&rzQCPMZjXY8TX$!tqSaYO5N|dXBPkv
zd+ey=2}2A=@d^p_@6%Ruf#~_Uxp>_6qmfuf=t5}3a0Pkg0PUzbT)YZr3KXgByx7Ad
z@0$as!WxaFB!gzelWk^aY+@@sRe*%2wk4^+Zs`<Xi3a38-S$@{A*9ELr(GV-;yK@W
zT38f;r5td$pgTtwf|;>$xUi;)?z>h+tiOW-JS~@8-Wv-+ejUJHv<#J5&A=cROxt^&
z4t{i>L_#`MG8*1Gg9E)e0L!z+GQmc3cX0x?==pMeX24Q+@3tN?hH^$mCN!Bo#MTq8
zfO>0WfEgrGOFky^*zEs^i2JPRbSQ?v?{WF3zfK=!#s?@S{DdNbC^x7oiH)@)_f`Le
zR1VRB1>ojMeO+%|!xP@|GxB4Ry}I3%k&!_W`XQSklknEHZF~xl%7DqGq5{$x=QAcE
zvY#>0AR6lS>V%NT6_Xftq7+Q?T^yhA)Ee!9At^xKfg~`JgTNOppgrJ8skWGbi4-r)
z%(Xt=re(v@w_{t+9LWO+T%gyGgQbz6VS7B#yQqd4B_-jn)Z6I-EeH^<05!pT1jDa}
zB|6o50B)%r-cMfqx7^P=T>}&~${$R@^VuwlYZ4Z0$E*9wj~p&5i~$@&mxt5rar~d^
zcXZH2m$UWhH42Bi1L1#e())!({ERhKwJqFNySv7O_yp1fg1bO5Q7~f#h!G%%at@fW
zY8iZdLq`X6#_th8k}CKs0D-$UK3gp-ym<v$;PA9RfF(-fbXg3dI$LhMI~L$W61*Vu
zHMc07B?>DE$9=2{8mca5w_+C+&Y7sT(0TxHDIus4G4bI$M&HZ%w{0KHye1QQ|J*ks
zA<ss{U=Sdi#oBy%dFipyNk<2B+*BO=WGwh(%F4=7F)0mJTiG;iWF+Lj_hi4`L`tby
zIoCE+*_B2|{qAQmG^uQ1OZp-5`y1g0hiB}%VPQj;dt)e1faNx{Fj=+^@$XruqV_0m
z_>K|S2@0~Up<5|13O2v*2L{O47pIbw8&8A!fwAv%K{O8%VyZqZA9SSf5S<G5wFj}b
zwMYQ}*`P=Zwmld6tnADX*7E8V4uvmBY1Fc;0#We`PHWC-pai;cZ<e3!m1goC&0Q0w
z#A*g%7(x2%u<6?uGx<otK_=(1*&!z6F5KD0VK_>v@GG1mu-~PrlL!dAJG(uC<dV6}
zb`!CK+jW3o2IFFT$F8q|(kGsh5apjVG;SDJp%8(8vYfR%%C{N7bbRDUBW-OeW5$(<
ziJ#c8GOkF-PXxeS!l$#YbZly7cJh;`PqZn&zzh&$LcP=Tr$Fg`or@s^N>O^LGtfSy
z%g4v1g?FIy3A8l+_I`Bu`N3J_ygo(>d~5<@{MQdBSj`rLLo#n(1pz3gzMcSR;uJo}
zsfnqIW(N)71a=~%55RnvK^!IZEibow5R(J4lb)d|m)oBwHi!}nj;|x0i!{@`LTXv$
zgZ-djgD50U`U<qXL`U?D^jHYKT~t}la(p<nmrCv`*b4x<_D2?u$s(ts@}uZNAd3{&
zQdf2d!qj>h>$)kYL;l^B93y*a1~8F5m9enw(?eO}86<cA_1V1_S+qpNTfF~X)^~|5
ze3GW0{1fR4q-bh7dgPq;M|HHdiz+J%8-HWA(4o8Ivy-{>_5|;UpQf(TIBLcJOlGwo
z{7(+tgWC%o0XBzm^~Ug@uVQ2~Omg%7q;1E~9{<=D<=tFAkf4sXC;~3@0oL>&9k`a*
zfnz!gNK!T<fB%UBTNqI5fwkk}9N4p>dcF)FAz?G=j$TE_HOh$&a5`En59|AB@k@cv
z&G@*r8N^E6*zIyX*)m}UgE_Tkm*7`1bTT<oVTV9c5ng69)PMuXc_4{$CrXY=1ar_q
zKJqA_7r^H}ECDDIPhDLyzl|Q)jmO9BL1|%VHk{n@#nimk?rc6<>~ee8m}|OH{CcTM
z6d|(DU~R@Y80dGA{6fDjd_WjVZ86^+guABM9)2c)6}6CE7NKAk6%{3fK>Oy+>dVaR
zejYDcXk`sQK|@$nrhl;iaHgRLtd2lNx!6bz)Cyxhye}?}w1cgZ@>{G3-DO$?%arLp
z${UG9iJxQPClY{>7InIQF!^xZYM8(p#e3GazIq-a4S;NhDEHzq!ICa-s)AwdMpH}E
zMzg(otN_P7V>5E!%v?qAl>ukC2fElTCY^^{9Sk6#qp8{9|7%?f^blK&I#WiYx!^Vh
zo%*EHtEdU`Zf)$~E?q0WlbH%ST3mA9PF=QJP$j$Uugb!G^eZe(4)F?`zzF;Z2!WLz
z`VzQhz-N*(_t;m~P+<m^G{gUDJ9-~NivM3g<@$=lOFO6X9QmJyJai-5qT$8)Nd?$M
zB&4kLp6>4ftk<ciz{Fhg9)XsgmfV+;TkItgQaFf{IDbAqKKah=Piu4zSs!iQ9CzKg
z{SJ_5-z)b3(5c*PWet4EP!irh+kq7b=|MqAr~!mTr>EdlA6>QB9!zgu3<wM9=9R01
zo>H<ObQL8ljdl;4uL)#KkHSJd9^6MfJ=*V6W3uY}>hk1D+MG*6o>m{9oHZD50$W1m
z4YvLQaQAV4jAM0mOi5-XK~viU=nlYAE-|qK0Fa}#bq~h32tZO3c-%>X8CYq*whwf+
zGv<ra@A(BKDcaYp!RUcOtAd(R)|fF04Uvf`mW)dkl@%=`eL-HH%P9jTIqT+~T5zzG
zY;v>ppK~}BmFc?m7v5<B-S_T4hPuH-Oy}%^6{*hempKx*^8>dlimIw)%srh8(%B5C
zlvT;mK1CtO>%z{I|4t|u>+bs0UQOboW4ieHabHD%Tq>~m1$B8C0jI%3YzlW>9gtU?
zRwZ3i=;ER}u~V4`ppdY>kJ@br8ftcXIswvFAh?SJ-$$aKf`Veg+B+4#J?`1<HJnz9
z-xiJI&iP2B@fVw>j8@yT)0O6zM~4R{mX-u~_G+`09g(rwY?)brAV~DQUeyH*t++us
zky3UM?|)Id$9F>{V$rdwbA3Y`Ok{yQTP^Mn!y_}i8+#%-u4UZGTu1F<C?vq*PJ(jW
zmxTgw@F#!7uzJGcdF|+iUr19&cakemC5589V&1n`NQnX6fNw;7;x3q61i=u52Y6tp
z<HVVM+LCy&CA=c~nUMG!Onwhy9?%cM3loq{b}b#8k9kmHA|8>%MZ_nmm44}K1W+zz
zFc<gy0`YtU$`)d_{F^T<z<<T~*I45GC*O!{5J+MSRQ9I%4BXU8r>0G8{=|YPJ9A3|
z?{#s>GBb)CcDHq~^gC4Ra--5=z;XdEoJiIYVZQz<9t=^t)mU}%XDZT?M<S4<<kRhq
zCuBHV5&+%=^9ObewsTTj#Mif(CW?xkB?|_usK--Jqi$P|`XtO~FWHnBWkPAOM9-Yv
zg|K4vXQrno^Uls34t7bJ%d+c5m4t<L!X&cf*(Y11rGJfOTB9L8hfNXHf13$#>yj&Y
z5;B9&mXhR{8ykQIXGH+RM!r{ZQNJatB`7h|n4NJ>P21x!oFgKgW`=3dj@o=GfVi^K
z;E{n(AXtbxxa&{Q*A1e!J}2vu0Ob9G!a_<3ue%ImA%k-<=8<7rY3lFq@zw-O18u&O
zD5$EA&rIy>>@|U73FNJ7Hthe(RnFR$@^Z|7!(98Pa~bcrj>Erb0%;g%`r2OVM~YLz
zMyl&*1fqlGPeS_kekQXFxS5GihLNzjPmMGC=ynRuS+FUsZv<}wpLlX#pu_@8q94T<
zbJUvbY<MFoE)EtjuY=)L?1krdcfzkeux?_SOX17TF?n~f6NnzA^>Qpcz)HK>(e8bV
zPVum8O;QBS%QKSMqs^%bAdQOEdUsq<ru-`Dwd#N2@n3U5dbepONsvN)aX$CbcND1D
z?=DV5<)lRs0=sf^bJe3rV>PuI6QC-ag~OYQ$9k6w2B9~Ac)3ZGsl5knlbw-llf#9U
zgG{J817N)cbuzeEZ`Z4}U#K$c{vh!m(MJApbDg)Q17>Emm~hE{rbbqoJ*@$-q?3jc
zyyk{Vl)$TR+IoWpoBpdb*I=Re!9i&j@Sbj<>8!C_iTf!c(k0T<naXeTt!Ev;|99c3
z%U<V|-yN}{1=0k(Zd!{O%{lM-FEoL<hE|jN8i0XwMzlo{uyJCJTYyO$!R02%u>r6W
z*!jsLw*jpS6sKb&U{aGRE6YI&RI#Qme7>W@jC+&nbO`CcfBzXWi=*&e789|UE#Dhj
zKm!m0?g$Z3@@FZ&`EW)22Sx<2*lNXahf8JPnvzh+vl0|peIz8<wbE`;RDAM_4nP;;
zy(B_HaZ^%I=v!0uItVNU-Or03V5)6$dJ^2xk|r8A>H0bIyu6%z&w`oJUtwd<Siq67
z8;u23o)3<fvo6nwv;!$ZW~Ov`2XHq{Uf^D?cT62w_X?7W=kUT8GH%;90Xhz_QsKy@
zy!*P_$>IX%r)0$R9szW)jJ%h_Nf$FwHx)rrl3eCX*ckZw`i5nuJNH&$mIs@!F(O%b
zuZR<Q?RG)<Etm+^`OpT2n{A{iw-0@Q!TEBdEqn=k<<s#Ayln<MDHGYrVpl=Hm1W77
z3g#Ej|24!L<2OJmBmh*ZhK4bJJA*`igUN$3@euTSm%G%I(5WgZ8hUy<X1GP^$K?6_
zHN1~qT^Rvr#N6icK=d_+wROnv?mbYV)j-SK!{^`}F3wO~T)gac{X7*<T}eq(_3>){
zui?hVJNAU6I1+0L<ELR6utQ2pf{9HuBWA6(I|j`ro5MlcnvxV?HURjQE28TtKeZ5f
z1qGWW>=nXlz_lalixCOdMoVS^P0;EVGa89tDD(E_Yqu@wG*;8u)#Ki6yExe6C`I5)
zf{hJ^Eja^x(kch&=?ctD;b1@8NFWORP#d6UE)hWh5{$;_DR4PET4{Q1b$5->=7E+^
zm))ux*t1g0CUvY?bDPT=8X7je4h{<hT6BE@&nWi+j-Y`J$ztm3k@0C|tN+a^3SzW%
zbWpr|S4-Ovmiq2BFv1w8MtpwmlTHPMtIIEjNn*H4E{>AGk&5t3*olhS)10+$is8z~
z4lk}sv~-Dk^Qzpxo+nI8d}0U9yEr6;FM0$Qpo7gTb-&BYEs_2{{UXtxQ>}9xs<kj?
zrxnm0{h3w%>5ozPq1YYQ2~@-|n3<K;!aM@zQ2U0gVyFo&DlP!cqc%dK+vx701^XBN
z@rm^R<<mS)z6uI5t6*nUADWwi@s$!n{r?U&w-CT!#7k!<5(M8nV9F`6dlAW0a2z9?
zXS`@vMgHtQQ4gy6is%Q>t8~kx&Yfl_UHzQ(H=*@hLR3?hf9GH@0tRTn1R8*;k^t`=
zm1z0ZG{k7geO;}G;M1X)TD3~CHk)M<HB-Q(kf58r;CxHqi#PQ29ffSd^1lz?mf>)Y
z<z~+aXUT&&bg*j@;a_0DvbSK}mmu^9oiY~hwEQugT>$+W=;-F_DgINxdy!KJKY+l?
zcsve{G6}+vj2k(8PJ?gYY6TU;>d}m9PF5a^V`fWZqt++a&5dvG8z)(ymKQGEo4Xq?
z1#_~~c-$g8)~X$x-YaqNgl%$RU2ZOUXPNzP#M+A;GqToVVi(M!B$c0~5&s)LKi;~|
z_`<GYe)*$JiA(_X-^Uo;OeWgDG+!D0$-kgijP>s&&#L%v8_-ySJu_;m|KBT6Yl*6Y
z2L9zH|MDX`ModCXMn=)xcEf*X&F}qzB@X3+&x`W}4hu!iD};Z)usTw1e$fp&g|&nI
zWa=*C9~MhS|6Z>Yh`c@`TA=H1nPOG>_X6~UyQ@7@`}x0P;{SOCGIsm_|1Uv|jIx?4
z%f03Q7B2sO=swkx+~~NnZyl5+f?YJ#4rwq@iy)F;VR3eTadvibH^7TC<lh)zg26lY
ziX=QTfpA?VO-&`G;3p+;NB?^abS!XGRcWEt1Pxj4J^R16{w8`a65=nEUZ~q-zyCME
zAub3E;j4bgk_WQ&1M@<23x*4O2apUA*r}Qn(zkvhHPA09H4r2P9s@uvE-4BAE?I7O
z3N(6Pd~*O;NBrmU4d=G5ss^Zp_f(luIlK}Oh)p1<sI7Ch6l8veIdXo@?tX=|fm<b%
zr8udqagnWpi}0Mg?gsQF;0c8EJNOz233H3PyVU~!$D(}`93JTcBN1RLP@)Or2J@Sm
zS47aUm$aDX1eh|66Eo6&T~vL5m>@kpH3kSxp<e7bKudx4?b<gJfK^~H0V_J{^RNH*
z1JJ7O@dOKj?<i2Ffx{`_b&?N=MF<!t<ctA`=xYBE$V-nsY!jXkV86F2Dga4_)8Qlx
znX-}+R$4p}7U2~O=b^qiUs4C~Q~19_=_xl9IEIekoAuHtCmUeNhBXdPFwheT_=5ku
zX3h?%DLRbzn(v7SLcdy>h2mCY1&Zw8Cq#~w0iBPpm`KO<e=QQA_w=T+A@F3K5wXja
zj&r_wcKK$e^S9{W(lG(d4UqaB{M{5HkDRY(X8cHriE@4cm@D}6i-Io|D~%86x4T1o
zH6)~7|Mx|IZ2&On1H#VEE`uh!5vUj}JP@F7PjBd7R!FRifzq?uLCg~o5F`J!hgdtg
z#!whz)@cTs?Ci?I!lJwXE;rD%*VH{8icoKcT~E5too>BrMo~B{bi4wzDO@i1+kiiN
ztdKDJgb@SiI7eF>$3(%ql3L$PfMN!O`#kk5YXpb^XRYgfOGmD98z(0ip+!V8V{4Sd
z!)YPSPrt-%5Wl07ivPfJa=P-ZZE!1Z=sWcElAg_$7ee}eTy-YE#}^&R+p02Y`#&wf
z)^Z&i?-kcp>cm)J5C<3v{oms$gn8T;(i<6O1pZNTbvH`d7_VOaJc32EDN9-41oSJ&
z6v(UI(Ddp0!ZE4LKB??zwes_4)FB~ya#Sq$lxRAbX8ym&{mO9`0m@5Mr^l@caJ5-l
z9mVg~ga0QDHVd1DuL~UXv8yN7dDVtNpFe+QVPj!uV==e1JV8T!iGXBww4H34A^-n9
znw_ZEgt<sajy6_?F(1z|SYQ2oV%^LZElDA^RByn<Htr@TCvUXd6J=J*8Nl^qot#N9
zwfp~mUB20@c`_XSgN(|}K3dPPb}i4;)Q1lrZl-}TCJ}EXE;ecRcensHHcoanPVgZR
z5m2wtfMJg5|MLz42IH`z>O-sx)qRfn1wP*1rw6U;8r2r%9v9~@bfePf7#M4>HMyMY
z_&{Ouqc(n9ISKsG{~o6vEZm;#2n(DG)05L5yF29^vcP*c^W4_6g#(Ep858G}|NT>2
zdwaR{a&ny=1I6zzX)U%zMP~7)!T-;PXnskpl+mX5)<}*7m|HP1(I)M>f<x}kC{9E~
z6MmS+Kd|2tp?0^|=G`%rq|e<m&d0;qwi9Ef;&7m>>HEJA<L%AbC2eb~MIPyV@rN1M
z+bU@@gR@-H&MtGY6?Rp`Fr@JPt77Y#wQB4pGtmLlqpPTn?AQDuaK!%iNRXt~0DA<C
zXR}#L8c<;<X>)*g=zAXAb+}l+wmdWd+;o*J23k-UaCip>XZzNc!7fBg+b>te{l7~M
z{HfGoFzoQ<cjg<f!@)qdEzh*wdGKwdQGb)Sw>;WQ3d<Pgqug+TYi%e5MJ5zEN{KD?
zT@xJ>W7@xiLXE#t-g=piYA`1~t{!={k|92BT$vkx-RNDzPLLw<r!>_fos{{`1&oAs
zf?#8yzXM@aa=-mhJVMD-FueC+gRya>CBIXp7$RQul`oFRj&FK32pc-7*jqF{&zaTI
zD3B=Zt;wltp(<tl86M-55H)0ki>Zu1J-K2|WkEK?c;P|?9KJa@Im}FQC8Q)g%}Jhq
z$wTp6J(%G=at;+9xRsUFS%wBDC#M92yB<vyuW+!{RHxq1_R`bSN;`xq2&6@gpZ*!Q
zRWbZ{ZZ@}~cwAkZUoWa5@j+rO;{dqo3#ThG49Rf6o6VVDezIx~0-{K8#W&Pgd7l5h
zQC35TgWHfw;O#lRPH)<txLg3CpCQQhV5%2B;G=-G^YqdZ+Z0aveGY9&ZcvyMUN$N{
zEv<Sye#NS_uHt;xUo(8hx!&>#_dBa!V{h@0w`vZbbv&E!Wk>0>kBrg&%d}Gdt38aI
z33ki1YRZ-`jW#=EW4`F_@x$kRbMr4>$PHLOep38GPB28c@*d;!2MV8-5MmO37loD4
zTvBGPIszF5vA(cTA6IfV8mSD4T)AZoSL>?N@^BOOogocgA++J>s#~GZ?ChLrmZL6~
zrH97C#^_m6tenOMTe0mAj2|wT$Bc`!mICvnj<q*WB|`n%^fC<%3^)F=;UORFB)am<
zt25w0?$3nCGk}Ez@7pj@yNz3IsI>C!11d-2_wQ|^KemC(EwN?sT*aWsN8gq<2~Zsu
zYqjvoD0#StW>-|uGqK=nbnBIZD>acIAABn;J}G7U#)6!Y>Y-{~%?w&_q0tgPcy@I4
znV5XOozk>s6&IJZIKoreI$>~@kg#DysX`>)BSdc>7_j(ie!Z5>)eJ-4-YyDXA8lx)
zD!cp;AcuOmbmB)K;~--ps{;eYClP=c1R(S`+S-I^O`gLc+Fx&y$}#%N#m)(UvwS*w
zS=B7+$RD}t;ar}))#i9&8Fgk)wFL!k-RwhGq9|i!h(5x?zyva3wXWRu)O2q*(hy9i
z6y}oBQ~2>Aw)mF*z=KMw%j3&Mli%t}Wx#w3tuDwn+Y{=@$gU3q_gfkwFX2L9#@*JZ
zp+r;RM?0{X`<-pZC*kqCoy@O?l7!0syR#mH-|>0e+{e(L$!Xx{iCtNRJOyzJ>?D1G
z^!!HrxeOmyvXNxrh8Kh`pS2<``TP)7n$osdt1qp#I^Q^&=^QArNG(**IH5Aa$HzxR
z4urn-pjq~d%S}rCE@^oJ_L~E-46bq+_<n8ii`j2+xN?_c<XSPP3tBno28uEnXHRss
zo4vIE+OTZQa}J>Z;LTkscXCa*$7_(8LFm!m@j{Dz<@Wr<XRr-1ML?(L*!AHRG^?(W
zQ%`)P^hw8NbsG)1e)gs@>olx1*xcNm@R*oe);SJX?i0w$j^CttjZP%6M0RjlsW+J3
zd_l$RnW<(2kp?%?u`!X4USr0Kjqc^;XWhjt9Gtc13v+U)b$tc^kPP1(O1MZDomiOp
zm#>FCU{0-&vGy<Y46`Rrr>D`IJd^ixS#1|cNp-zzIzf33c`PNYKYhD>;V;_VX|-T}
z=V8i@VBhU*l_|Gx#kPHZB{abb!Whxg#D4_KXk{Rxh9V<{;x&BT3^9rG<dwShZU_<0
zSaMkf4g;6EGsbW62GM%ZIM_L{8}G8fB77q#q^)wZI2Zlr%Wox%UN2nC!1dMU6wa4h
z%Vl=r1HGwW_j}ZQ)yU}i*pD61mCEfPBB}VNKePcmW(tjj_qAYSNqsgaEtR&a=qDxZ
zsjhVyC#P_r2E}cD6Fw@IirYABLEj1u##Wa`riDMJmNV=BQ6M#rJ%zJ|?iulb7v4oO
zxy;`-M;^1J5<o5&7%f70n-ou9lC@HFP{0-U*0IV3M**vX1-LOBuhajGE!7h|5uKcv
zAjH8}(pHk9#0*8YugWnadxp%M_O1ERlP3iglAzrIPw1^twcl#(4b~sb=;){kV|!OH
z7iGr9480vqqNSnlScK{EmJ*i)X@!5c*0q}rFmQ2c%0Z?bKNxc>%S%WKHfw9-#{84!
zDS}(sgpx<#F^9c|zWl2hUL!k1{?L&Bi~i>&F8v(+<+H(9A_vcyl!@xn@{q7&`g*#Q
z;6=i_q^w%X>HI=H9wI7^Olpoy@_~N$AWo#Sge`(>6v2LvPtz)&sXvE&zY-oW;xuxL
zxpjzah`D}MZDP4fZTsxKj9igeU^qj~IV|^H68_cREdfK;peijh$i8U$sS}`;vjtZI
z(@HiHm=+WMgMuiz&glLY$LS}$vhSM{+U*~MwK1Q?(iNOcEv8>7i&c-~pR4&<CNR#~
z!*>Xc8`O<|`HJAq<3p_Q`n57Xr!>hJR!W_yA$J9BH>L_!(o|Fd(=Q%fOVsg@?=k6P
zRVR=hmBr_bw<N*%uU%ptRQUSb+Mw>{9bN(R(3UPx6?lHe5F|MYrwO;VE`7Q_^ZT~?
z7OZK0XJS2^3=Fk@#>-bn-;l$!IFGq!K0`)EzJ-6z_EMwDoyG1PL<daG&2e}YtEf20
zC`-fQ(wNLe`3(-wElpryV5?~>0lN|&tFR6AtrGuk>Zf0$smlVY<YSAI#`UrAc1oIV
zVj}ih9v&ds@6$W0u%DUN*D=6>q2<BTLQijWyxo4UyXt$}T5+lX5NEmH{Dp!3ZsOuJ
zGfs~y$Hm3ioH$P`FVljL=<;9;gxV#@!6EoWL_Z_Y(~LNR<KkXG{{Fqop5KXx_Bmf}
z-d?|mQ&R?`lk*IvqKyHf#+BcGjppz2pl}$5d9f~kHKfuJcw&o@{K)QCz;#&PWNXm2
z28>ka=Xp2P6Vbv$oL%msX<xlQ*r^zl0pTYsS;fsT+*(T7DrYHf2<195D)=Tf*_*N3
zv%u)I*8Ieq`2;TB$|5f-JyA`bU6o&qfRd9e@7RI`Cm58Mqp9hB=ezN>o!!}+8k^Kh
z{7r4?7=<E#Fq3Wuuj-47Ev=q@U0%up=q!VOs8c-6h1uT0A>d>8<n&ZUd3hq<j@*Q@
z*dQu0;?d5IHv39_Z6zNQSwcXP*<2JH2rs~5@qCz%iN&GjsJdv;ZNg5Ak4n;gr=LhY
zF389z9jP9`M7)g5<I%-RNd7(IXB0@wmCaHdpv#X4$Q&Nr`SC(bOUU+%EkUR(Gt*}Q
zL5IHvS&>-4jn-GRyRy34#y09}Lc-gu_fE_!k&%&A1+gEyr~$(g7!<@%q49Ql$)~do
z6(TXU=*lu&!(j0-U<KKj_jRh}tXQ`&n6FmlWClt#^iqCCF_E^eEHAe$Dr+kd5@Rnc
zEm6?Xec@ja5|`H_hs(}11p7R2mNKM*+nd7~oGlS298O^~GyZYk%@rY(zP)3dltcgn
z`Ly*8F#|aVRdgQ+Fgx2+Cl^yuRRP(h{R4gZd31HPO~4D8gN<c!CWRY=O<i3bBm#hH
z0n!`0=k6WB2^F<FJO6xGVf>u~cWc{9)zb5`>8G!DQuL?+!c4LGLE|ANiCEk-<}WAw
z&WZ?=5<zK`nvk-$v)*FZ@d?5sA~?N7F*mgYZf_iH90Y_x=?ytOJ<PdPMbE;G5?Ev=
z7AFyQrV=#j`b_YbmiQemP<{aZ&ops2e-M>~07F}CrM17`)HG92M(P=Nem3aSxm7?Y
zjl6Ep0&{MDy?>1OVwFBy<XaP6UV-^x-70%n7_W_cxrM9O7A7VS^BiCYNMHba!X}DY
zwPpE3y9>&jXz!59$)N(=KheMml;6_S@?Qh=NKZ+b1%0baIN{rygG$}*HkFLLg5{*S
z9gVF%8dfc3<bXn|=8)gq><Hv}2t2kq))Y&p&1%jBjBZ!XS#sK3NSNODdrUqyXoP#R
z@I55;;X~q<(t`MR72k7Z4dpmcear}qT@s)2O`V8Y8%bIQWj-`;BW7ms+{ty`cWR|g
z|3rjn&QSv+e7SN<>S8{}O(iDHC1)^av&ou(5Fa<xRA3_>gO0m*cxZEWUZ$bvaeGaM
z48MN3F_0X6-9GnPBc4UZ&OMZxnu(T&nivX_?9Vsbh2G15^C}cQ>!u-nL+L#XQUyc)
zjGxludcuDL!hZ@HFIY&%VJg1A6Sri^O6%qznt8g;;wDbf5_uxvF5tzv$X8a5M2y?s
zzS%Slo6h9vdo`=2UUnpi9hF^|^9~z=r4x#6)OZQe-Ai`8&H}&ZTXf2TIrHM%^=)QB
zUJWgK|7!23>4u>`M<U91-X*EVI)vnsU_OJ|%$`IWj*DqDo0cG3go0Cg!6C03vH632
z8ANi*wft0qznHobO)9er?Ce6GKktwO+KS|L#-$D#e%fXPkE1|n<*#k<h`1v6q6uU`
z@-z5&9B`BMy-V*HinD8m1s2MjLEd-?kx!fYb`@PFolt_>ccF~s5yyAg6;IuTQqX&&
zQ7-sHSM9%TBmDzmMqUx?#%b$#4V>}4r*V{cffuwd8&!j>5on6gUVLj?YjYtUSSO4!
zLNI}8eJ~GD>6cnEPX>7zGI9qYThz}wd=!&0!|l>umdY3t5ynM4hd}oVbZSs*U&f1a
zyO)%do$<T|TR4lgXWu{uk8x%Eb*7r!0pj!f)m75*tJ~W<7`?zD_f4%FE&&+7k5s@u
z9IWXSpf{718|t;K^?DZ2p`tQ#C}>;X(&CjEE2{0`6MUCMB1fcC*-8MhmfMD<+~W7N
zgZ6M`a&rH0qrsx$;BsgW2CyvG$0rpOl^BdD7a+4sjfq%|>E7Jb{H<&j0YOrM*)E7P
zAA8ntbM1Tutl|2)@-j)~awf{u&Gq#wz$Fh{IY2fWD1;zV$~WJP*~!)va7z{qjEtfq
zqs~vzF5TC!+Q-6td{U&-XZc<GCCE4a;&gUWDFy`m3ne#Npw@(HUDqw;#h=sDvyk(s
ziilg24~4aQ>wh3JnpFO2u4-v)X>zc;iv!lYfk9NH!0qjQBm|^A5}?Fj_477nw?XU!
z#@Gf(F3!#d3tu&}MJbr5PX})3MO6P3<Xi&Zo-JnW5qf90+v{}Z<U%r9GA3FwS!$*3
z;{C%NZtL@a{M}C|95Ktyo^n<#9Q1L-^EFtIQ2TFi=kF2_fLs?b{{RzvS84Y@138x>
zu)uQGav`;&WAZa_tt>vY1i$1m9V_@`O#}hASYRwZCHR(<mTY=#Zew??$#(7!#S2G}
z7mCK*QXbTwy@>;M???;+u-*<1S2cJzkY8Y8;3z97Dyu3xU(e=&R5i)h@DLjF3g~3P
znk|x1=5_v4G-rV!S&9<<50;Wpdj`y}!p8FEW(=sdt`_~@t<1vnFU=k&PE}=w+yj2h
z_&?|R(N8p^Ph5(Vle5MBz{dHhc#1w2YF#&Ce(@q8QS;LStgwH881$<=e>09GoQ=_a
zOY96+4cDsW*Iv!RJ8lV9#i*Dmj*_JnFgW{V*gRNnT1`ShfS#)*uP&=DYf4@>AaBg~
zW9Pl&4gqRFG&=6t37Y4{{5nWJPD>^<yV~9XMJ5c4rz~dQ0`nVIp2KMpT#VObRE!jS
ztkn$-Jqv=y*47A+F?(WN9)?d_t;11zoZPn`r4is+qZ^TwQ<Cw(=*U?;ek$1@CWmt2
z0QMV@G@aYg0ajJCBW)v7oOO8xjfeI*3j<$HR#r>`;*Hx=7MehjoQWO*??Gwg1OgdH
z%jTpwG>9EUyxtXlYIgP}2+hs?(Ok?4**iE1+Sqz(D#Jqj{7ssUugxvY)RYv83#_ai
z&u$-P*l@mfL!+a!m4;g%Sp9w%W{xf_?Cl&_nS6Ey@Ege8;-@D(`I=u;BsMr}+<=q^
zR{slNCjJ5*K8vzzftN>{zcIlH-A}U2?Oa(_m4|_xhn1SR%yb~zVdZ^B!hNIcQQN?P
z-NBXJO8N{H5>o#PIprJcr|Fm!O|IXtLs!q>eOTBy?Hyff?G{AY&4~MJ(rYu3p?DHx
z!%QleUqwwB6$Qb5esIt@Yx(8DJOTFo*J?jn$ocMt*JT~jLhu{*AiqqB%gyI4hQ*LI
z%>)R`h9FcQxNnQq-jyh04sJo`AB`Lp>W_@L-OC3HOlmL*6^}HvY4n~j<9(iuph5uD
z`M3J@v9m&dtFS|K6PGe*_?wytr;BXeZe}}f5CeTW1m6^v=01lObp$RQNkW@(|7`i+
zgUA+wmR4O)zGjOGMu?$G;~UZGvXqyKjJ9(`95|!U4>{?b{e9SPf|kn}ig{q2vA^o3
z61!VSPR<17B8vbAy=l{Y;;=BsvQ)~7W_ZsU%s1%_5Npy*5z36i2m9^CT$uR4J_7gq
z7yeI3SM3%im(w-y{x9S@lRu*J6?s*j$qy~fc{Qr({ik!_LiHJ1NnETf3F$rE(lqzb
z?EqWJDW=P5O}0q!iBpx!G~ej0Qac>fs(%7SjQ=dtHd}9v-1M1mfLZ+zMgfUi2RRYv
zV|>MZ{w;Y3F(tq>-*L<9bbP(^9n@>(wO9AoaH{xzs((NfuW}%d3+7j2LwVAo1!Z<z
z6E~T86`G=MKLQ-oo=>ie>mqI^p?WY}GMivVW)e03^QLfhYBp-1fiBj$gldgjNX;cU
zQUqqzX8G*e|L;Z+r$*8`8)sLsn)ZRhAFNN<-^~f|h_&)dOI@8lbP}wTANI?AlBJH0
zgrQ+u^zlnL8S`n#G0A7Uk{Ic|v~}4$x(SLFF*P>t9~_2*K74@5#Pv|8wynWwWW2m^
z-+AoE2<QOW2r;oLP%FP12;N%85<^>TaJlizNU4qQ9dbhh17gr~KatVxb5I4i+8@>5
z(f7dF74;e*jZauEmXVN0k35}rED^?~@;7#XEBRuR|F#y(77#Kp?<EzJK`<l)V$)+$
z%s^ur`MoJK^Bn}1{7W=sord#*0lNS~HBZ7K^RLz3$@?Wnay+w3lk#d7mL{Qck}wbs
z4iR*Iab;&?^YOV$3ST`56ak}dWTa;7%ntFn*^jK8VNbvu0KJ6*;=jPkloS*wld==0
zb!k9NLgKM(X=p|^9GwLL8cr1;__J00rSJEZWJGifx2?6DuI|$mq0>^6`x(;mCym4f
zhSEo5eStoD+z}9F2olT$>%G?f)*lksvg=%KNlkIX;9fu`I~v5pVKB=j068QiNSa_P
zhau%-jg`(O;;y|r@kuIXCMnS=^NkS~DSmw2(DU%@8GxFsJx*p1y7Ym*VCR6Vy7hNd
zIaD&kG(z&ZY!=|D;yDkLv?d9%C0Zl8^~>|qmZk=8?fj>EE=BmpWJv@oowjfw^|wx(
z<4p<61}(|lWe(BcX3{N3pEEEtL4`m)w;SrlPH+&%Mw6Vz?|#3E11(lKiTFPIt7B|-
zgm7_HYxm#;A)tg45D|Py1(9DT-7zc6N#0wyc(9Nx>J{;F!{hu;q+f!Kn9+>LZb4x#
zSl)1n2m;Sv7*P31KCEx@L!}KK7gZA-O}sVWTnO6Vi`=WyIn*9olz~`?hZI2~lEuH2
zk>hN3=r$xE)D#^4t`1FMnXa+NOS2rYr0EivZ#7+@6@&rUGt<cHeM?<_TRkTki<IC6
z!mmn{IbVsETj=RJ32o|6PS6s?o1&D)Q-}`X?&^br5S*^gyQwIjLB!!=Owa%9aK<tc
zPCvR{g1`~E8KKL$X2S;&*-l^43}7G!195TTLIxXuPfkvl60HvOoYq@Y51`c@ejYa-
zRvu4lW8;-Wi!e0HIzFdT?F#SUXy0vBH%IfEwQDlyE$6t{pn8iVmF*+|q~PJ<oh(-V
zo-Vds_&xjn-umo3uQ30+)0L-+gM}@ycvKS8O67gWy{MmLote+Pws$*5+B?vf{Iqf2
z$JXNEWA~3YK%AebVoJXIZ2L^*w=|9j?B@ahs<R@a1%6io{MTGlo4@V{T#l3&v5@?N
zI@X}~s-grpwuWB}(DAVZ-@ZMb+Zx9l|DlN+H5ox5kHN~$`W)h))S(NNG$12q6B(qE
ziMWv|_!*bQX_(EP9j>eQ)$X#+SBjUNi&5>9JM-DvtW5CVt*$AWlra7ofyv0ZsI%&f
z=aA?&#rjW9p$sr64Fj>pr3MF2^uFg}V5O#DyOSzk3S}-bU#JWdC(V6BdwO!}=m;hP
z&?0DG{mp93tAk6JZSNS;)>PBgoFQeTI>Cp3o8y$i@^0AMP^0iI3FMmR0fwa-sA>#L
zL*shcbRAJ4^QqZVc}pgm@9du6+f>$&p=T`Gv`i9^dQLg($<oi<)*yKZog&aV<`xzT
zm3{W@c-I(ICF;=oAKH2|l|>+nF|gg+O(Wd(O_LYrEBDH*MyZE{HVkR15Z`ytVW1s0
zI`^O-R+9uV8;4ZO;%N$hW5U4^Ta3q1#Qu<_WP`_$Ai6_BOj?faQ_v2hm)9;N3rhnz
zGdpCD`_~0(4cdWtN*vdsT~a5wl4Ipf!cZIxc??*H$k{Pq(Fke{GXkF|aofe#?w;No
zMlb6W6B#~(kxx5CH>)(m`5s<`?`3z%Y5qoVWF8g4E?r+9{=>t=dR*XQ5q;dewT-Q<
zjV)dO05CeZLPM5+^D|`790IZO!u43pXmY~(cAlRj4+rtRy<s8ijoT%Fq=TtdS)#cB
zcVgqCZ3bG#7vC-~!<@uy0I#8}tV&PU>oEw=>epWRHz19PGAom8&+5_UgP0o+YA_tc
zlj3<lx_p%_{dnJTvYR;X-@Vzp{SKBG4booXhg!V}59&2Ji@VmWv%BGEEay<p(V+LA
zL_w#n^fONQs~#})kaN#I9vvOsNgGzm>-cfXvEs^!$^U#T3Uv3LylzO53DoU>{Os5&
zY{~UOEj$hqf#qKFw<b%l*HKc3SJT*j1Z5%=k($8BOyh&xh0B~6aGqe2xC>CnAM_Cs
z5u^~ny*0Fza*DIdK{Y@V_#84|<>sY9gU~<thKH-k8Hu`ok#toJ^KN8NT;1J}5~l=H
z(88jh%0Y6L*B7*>&cE!P%EZK$C@51WpVpURZJuwpG>Pr*5&{4L(fNK~<16%wl(~SE
z84TRUq(obN{qn;;>B3A67&d?vysR9v&!E4xm6YurJ1%a{_vJg(n09aPEJJJ>&-<yx
zNqH&ZgPAm0nPaT%{2+38A4H^3dBvYS>tb%^I-YCj2@KS9-p<w_r`-OKXh4p7gB1PP
zv8DtEJ1$mD6-6mPf-<c6@e$Pj8WFJ2UsQ5TgqBo(zg)Z{0>0PKbGU1t4)O^4myk)q
z+(J#ESpKhEQll@KXb`pNZP|&$vj-onyqWeJJ9HIsH-A+<YiF;goRY_PXq6{{VKTY-
z#U_>}MY)CDU&?}Uho5}|&=y-&4#;up>7*em%WbG{h<xz#eGawm_(=&1NBbw9b~_Yw
z=T~}}b&sOsBOZs`H=muI?CcBzf<<OzFg7}-`-=_Ks`qxfM4S6Wk0-jlW)$3Yz3-1&
zmtS8UylydE0;K3827Mf(K&=z+>9dg2C2|HzXHe=6;4;^`9Fd)Bzl7T`Ywf<gLM<&6
zAwR`P3bwQF47<OPunMzzbCj?Cr|Hl_EbzGmvW=WG8qVN0FAMH>EC}Ae$1NsvJeLsG
z>9p*`a%TSL%%@SlQ6#9y$HMR6<1TEz*w;h)-8U{8+n@~V;vTOkG-T+=&k$eFM^mQ5
zPLOWv=GpD?ch^4JyX!o5cJKDG*!V->G1a|>9%lr9W-rV9g#3~0P3y|s)P)NH3XV?~
ztUoHUjw-S9)`<7FOmP~FcM&$Phy!r9Ufmgdf0%@U(Dl}hVn-GyB@SV~1s&$Ul*m}-
z!lY#RPwZ{v*(rKm<U^!jk7JHa{dD~|i&q5L8ev)oMtvtn{<PV!e&-2okz`QrL6PSD
zR|`wc?|1<}DC6E)NNau05&xeS;CSFNT^7>2d`>)l)dH_}HBGLgt&`<>_E%JQx}kRo
zLYHSD2#3#8BdH=;z8G(@2o3+>KGW7+mrz?L7C`jl?@8YRY8^lKQ=_IvL<W37uQ<Ca
zFe=|P){y^3l2BXY(vi5_+&p@0jC;iewMvj5%s1QIM135T?M<G3am2Isj*~PkCBVNI
zdaI-voRmb*!Cz6-^sKHNj{$yJBJJq4bI)+w(m))r9~PGiLrMCk9HhZ)uphc8FKmFM
zINzC+f*2|*CpFTpSEd;Cr-E^ibH12>@i|y;ggsIMU$R4h7M(8q6f_tP3Y5zksj$Fz
zfk2S?mZlaF>6n6IYpg)-c3?`3XQ<sp6~x$j!jOR)06c+4VJL{I!O5JMh)9SBelIPY
z*U|Z^R=E3;#6Aa+|K7=R;T>3iVkDb@QGHWKMRxXPv=~}r&4?RZXwKOEQZ`73g@A90
zG(KEHAB*XI{4}_n8Q$SQ{WMnT6fc}ARy9C`bF$7f^ey67YaqoPO;mqMN~nyA#*T$e
zATS#X6(y`H&B5)yZ^T5nJ~8Sd2g@^w*9|`he{{4+R0R6o{w{F?Hn@PkMW!#Se}{)|
zZ*QSrm5tuqGCBYqb9Rn*S$k{@(?9$r*Z#0&HR0=ZD|mGP64JAZic;Q_;%yIUfl7kh
z-0<thqwR?)pno!bXWjI10O$&UIRKyc3c?b%xbOUa=2lG+UfaUqpl6M4^UOf@_5(X_
z=7?Td#z;4uUCSHZHKXGLczrW9rlP#O3pAv=ZVnN^Pf;>3Fyw?0>+wYXc7Kvi%f2}u
z(>78Yp^+UPT=9JMQ$j>VgO-tk&<5nNbS7!5tN)0LGdkxtI&DT=87y^MP$(C+n^<?b
za(%<k=d=>uE3O~nXe+|IipLcvkCs(aTY(HAr#!jmxhiZEXyGrr{=y9J;2mso)1N-V
zYdtPK-EK;TgT&uPhJOI(;5lrJq@ioq)dd;^&FJd>pov4%j0*@LgaB^1wU-l3l}*9C
zLE>Mrp}zt5O-GU><wMVx(w`WTUxblpzGu?&=<E}A?mbs{1L--3I7Oj1kV5g|c>X1q
zDZfy#*2?Efen_E0Z;l_k!hLoKr#L53k8TGa7y>}+T17{%WuA&U;g87mggh0fT$jX3
z@F}-GC%d{vCrC=j-;F&#3@<?3+afn%z`f@lSCf63Pzn$FMhe1Ov{>&9N6rt9=KPI?
z^*`xfM^BMqA_Se<8e+!)Y0i&eRR9(=b+Azg^HVBPL#=}5xeBlU`aE_H4|lh>v$C?D
zZZA7TqK)f;IgJzn$74`se(cJ^*Wf<X@dSV`D+lYU%gmOx?z>XicVXWXgE0T04nVL_
z^r`tdBzu2tw$N})^dqFbZTK-YGDd~tP^a0vuuz9-q_(ck^X}o3y1`(Bq(>zq`Uvi3
z5swhz$j3u|AVNuhJR~U=L%z!)f9WUL<ho9b6TATeJP!9QElv5_)GNFO=+#CR8@;im
zaPhZyLPA1X@zR3=k{SvM?gsNRNRVEL3qKU~fH9DT8neAi#yv2PP+@nr#OJQl=*R`e
zrxhcBUX`7|0$?$SwBFaIWO0vDsSjGol%L5yK<cIyq-3-lWV9E5$DWC6yh4KO4v7`K
zD}raS{UMz_X8gnZ+x0bR;!@9{-w6f=PQCMC&tYn6%}1)hJ`;e_0Imu7)bYlETb;;q
z*xYXS?bNJ1GZn`(jXv&qK!CwN^JzmyMFCU0wo&5iMt5OhGU75L$`NrMr^l!-QH>f5
z<zy_D(rDWCpl@eB@;$wB-3)BeX`b&cm$S#RuM2YyXHMYn+{|YR-)}aAZltfD9Ui8x
z2V61gmcIpJtkDxTFHSg|2D623&_2ZejM>`6u64bwb#*2r+%=JV+Lq&c!OM#poHe%N
z<KxrIA9mkN;f;Gn=fxj-q3?d-%q|W1fRm29Zht>ulzl_Iy0s4AI&5s?$0NPAHO>;<
zbz?6nRngl4Dd7(;Xp{)XTJDM>%H}36T>xT(X`OwK>(k~QI54;}&N7JzB|pdXY18XY
z+#z|K@Zrg3O0TZ5jOW}}!a;!aeI`8-Xukg~-$p!J;S5EAO67-dBQIV<wJ9gx4zfod
zkFoHOyUDIJT<v13N7+$6RM{W_PA#I;??ZYds7{ND_*uBIp@`h#G?de+;^OAd-W{0W
z^l@vL!27Hn-~#IW;Q=6^00l&4ZM@+Va3rhp%XwLTawYGk9mEH%Iz;xEUv=HJhllIV
zOGoZN5%OtTOG?673c5ItJ4bAQ0EH*b=4}3Tzz7-C{AD_;Zc220NqvkT`)mXT0&<_;
zVQB?;C~S9MFx?5^CP2FMi{he2?DEe(y|;~$4^XS9r$FPxb%cbbcrPKeodZ-3d>ZtS
zHRI_%(JoOo$`LZZ!C^g=(C<syI?OykVdpXZ5t_yFG>GR!hF)$I0#M}8t{=g#O>c3;
zwQ}0ZOj(8j8hn)+#{N3xgPf(fuziQzm(O3pT~?x{G|mbl9zjZplDhIKRp6|k<D1C$
zqbc04AwCaSJY;x3p*qin!}|?FZS5t<dSv;eGki<bU(CKT=V-BBK|v>0HU}+|76hBp
z^#}DTKyTl^yS|&!1k@h~XKqgb!b`;b{5LFkAp<j(aj`0ZS5s4+xv#^2cI-s4wG8$r
zAn^j8*L7JO0xWQVH1*e8M0*5_=AWqY#QAy*0>*a?yA?@?a?e=suDA5>UWBKrDyv$3
zYPkh{e6I2?Gj4H6lgb2J1sd_)$$i!ymU}WMR}Cop1>JT&Y0j(?g64fNF!sk&WQ{I*
zsA$H-j+}0zfYS*Y<=@eF{5^A1Xa7H*z5^P||Bd^Wj8bGvMkvYNWG7MfmJnrU?@dx>
zc6Rob2-$nf-g|G^d-Gnu|9jroaq86JdG6;P-|uyOuFrsb-pp+CE$PQNQ?>QAZse1v
zCkl0FmY~*ucs4akQbG1JE<vp5;7XQN|4ni&AiJk85yDD@e7bo{{5u!`^Ma?l-UJFV
z-nJ?jbh0-%#6un2T7dIfmAY_^rAH2eF0I&bNug&V$3R}?k_jZ}XB=^!FdAatj&Po9
z58RKkfE*0c`>W*BwQ)clNb=C3R=bP(+qpSxC3qWXUGJyMb)`oX!Qm5Pb)k7eX4ffQ
zS5NMrxAwG9jP{R5YVSmz;vm5JKt6#(!|%~=xnyEV&&dX67cQY|#X&ychk*kH%0nia
z^Um^_P*>?h!PC`!k%P&|VO71FA!2v+jkQe_#NNIHm-Ox_7a1uvjapOJz{-wa(cM59
zm=~^(<cXQzT{dM*>*3_)qA8NZ<rp1(6Kad`{C((ufbaK&n7fGgS%t2~A%l!EnU4iu
zHcD`-_Sxn}_oW2DKebkZ3Q#H&aS<*_t@Ow?4d5-Ei@kn+wbsZVB%A-P(n;p5?fz2j
zHyC+F2Aagbt4OHi<$Lz>t)%tJa3Hed6x1Qya_PUy@~SCj>b(z+*pJyqA*9C<%qp6z
z@8;c}3{0ms`S?V*K;N3UCn8n)vYtUgdyLR&i7SEB689OfLcWZljb!JMt%&6^=9;u_
z^uFA3f4d_Yq3skCb{An(D=d$+Z%rQ^NFT*Vcy~6AhIxdIFFXbL_pqjGtap9={kAv$
zQHMV*F;Qhjt#qe=-Y__^Mp~vvv%(o4$;zgu`5V_V=Y#KqO~G$-enm6$Y4+%-n~L3O
zChylmD=y1+PW0|1!SrcF8jqp%G!g2jR-5g-OGl3R-@L6($5Bob=65FTgw{SfsXcy@
zU^}sQ8xb36VRA1bLb5pN5or3U(Y#^ycW)q`frI}KpW>JmEek_%#V7hIEI}G^ylV5^
z^WEqoK9(+((+I4IfG?wH!T&Ze5I)^1%Od2rg4^cTMPWcp@gMuyhPzZnJ?jZ$KiAu*
z-)rmI*L>b!Gk7#)Gdl)!5A;vz{s9Suy!^|8INAO>&A1~uxt2^vZ&!=k9iTQSf7X_`
z7n~WsLw0x>h@Qg7|A)I)ZR6w85{^wm%i!b5_TT*J9(q2*2KkD$(G@q>B>qi`<@M+J
zk0~w>eC0%$E!(Om{>nV9k1~JM4r%EXAqKK0S8qJmEXIDDs~E_>Q?;MorR=6MfiVQn
zXdcBi-2{`%TaAfm59W8}D5xYRKh-mX$VpZl=60^Mpp0PNvcL;dJi|b*to+}8Ch86-
zSf#UL{+}sY8m6J~Y4~EBXi!!Zy@NnhLfLhjv5Su(P#xm6rhuFU;jVhu%J;ewJ$NCk
z-tVCE$FE8pDia{h4=I12TcF%mQTcxJ<P>6W)yzg^ljS2>v=%r1t-(yVv8!5VYa=-R
z)mzDihCw=7A>)OkOMywv{{ApTB-JA&IIySH+t9)ll~Ta@HUb?Pl2B{?;znVOm0df7
zBMd}XSe_KY1Rs(|2Mub1J$8Vj7eqPj^WS{)M?`q*kFUG0xpTGdCq^sEzxiXn&y372
z&%dqy@U+l4bu(vi@BqFE*GuLt%a$Wmu{d<o<`*V4!@}mV<+Sr=D%4&-ugNUUOfFCd
z<|+`m5X;57EkO)9A@?JCI?lph)w+6mdN4kGbo16F*4w>P1~r3rOKj{TLH<&t&Oe=d
zE#_G=ZXa%+70^5{T&u|AwR}_WAtVo0$831@SC(0{s_zFh1BN$U<xDZA_tA8x>F={=
zT&Pa&;A^ERbd8VhbvoBHmzwD=<gr#UAQk29O^filZQSb@zJERGNbOnI-E!2z!^$P7
zq%2lelU$IW?=e9;;kQ`O@NLZJ*$ViHCGb^TLo#xv2OkSFt)kjYTXS={#Yjm3%!zz3
z(0zg_e4AS&PS)2y;_sX^q=YS*l*-c$E=dv9?B1=i8c7hTTkrm{%hp3g`1mUfCM%rI
zawn=ZG&TQJrAWfPaLb`9Np5d_-ZNGBp%c1yn;tUyX8|Gd>Klr&I=zrXSY@Q36UfS|
zu^Jm^0t0~f_)9=^dxWB5{6b+e4D*($B8MMrGD?z)RV8_IqifpcmsZb#*EXk{*~O1T
z_p}|d#YcEJA#Z1OJ+Q*#U-yst)(!w$icPy{i3Hw#3O@eZ$d6ba%qGB9;T3SgB_Rv4
zJt5cN1`=L_@+%H$)*^#0fB)E<fll*-gTtGX9;P}vqE>$YQMD<smeu2V3$GtM3VCsg
zzDQgJ9xCWLpMr64a2#giMwHB&c^ZW24%!bZ+xD)eCZNna`6dbq51P~k7y{<dOGYl!
zzN+#9;?GGF!E+RMDJ&%<>zBzZSs7wNTy73FvdpZ}iZ2_HRmZA$mM3l;_BTFRsOy+&
zeSTP3N3|)~craI~j&tix1IngW57s+MUx9LUx2aoGFMx?j<`^V6Q0f;ZM*Hq-)$r%6
z@v<1|0h!0&Lb#Yve*rIn&!^&zy6=Obk(w2cT{h?QGjIbo`S@UT)EAc1kM!-GrKNLo
z?;mA$#zsh8E9`S~ayPd;IXpTnC@3&C&Z1g-h5Dl5{2T-LP&D-{7dKw`czZ8v^oUO$
z^7HXG(&)y;YQU%gIHhNLk7AjvNA_<-;f(Y|#Q$o~cjaF|K{nzFygt#~8SvgLq(rTo
zi*X6fQL$GmCvYmcQ}WtMGFMgIn^W1$Vk%DY8V5@=y-N9q(l-H7`!>6rjl9C|+#^H@
z%j)wC-A^kdo$l|yb&NbaQgm*9l~S2$@e(hK=eV?t$Yr;1&vf`V+NZM`^d3BIDti=-
zVukmg>ve7_e@WxkpJnBJZhyr^&k%B7oRa1@wQXPA&@x^N3z1M^k5hvVE0Fh;=O1(p
z;c4EdP{8l%Xt#Ogf`jf=9&Q}mtaB_o)7{Z7W!Ot&@<H?`wZtk7`iF`R0<(~&2^cz3
zJQliPdhu#}uQ1OakYam*3E5vvfC)!)lPMD7YCZ62*0B>a06J{_1__j&a(nM!Hh9Tl
zzg$chX&>1WUaUBGE!NF~?iCst8rW15pOOClV_LAZnUv(~{GS%ReF@AgPf!tH3jlip
z-I*gj{n4eDXz1i)-O$j;{*j@eASqKbGY;-S32!o3H^|B3mubD*E;0adX<I${7t3eQ
zt;&JlSa905cY8X+(BSy!2rg=ObVhcnAo($Ki-ek*gQK$P3tbK{umaE_Z2u%2dK@%#
zA9##HUu^)M5-nkOeDK9vBAl03=wQXn#KdFa!7Fl#l34wWoD_ol<Z22kx>@<DIk~G#
zpa=UI$pRoC!R_W&VUmF6ytg$Zh_{c=rB?%xm0{oq^fjDdf9MDcN(xC@OOArN0Ro1_
z*!6D%PHAMM8qmo)+F#MclwWIVtA@s(^Z2zyGQYb=9@jfC5THqpkdTnw!Mu(7=_OWf
z=-8VQjU!uI+Zrny(W@VrP$9$O2`T2HpKlDS7PYu*rOIbp?Y;))$xQ+GGvKv*-atS>
z00&~&9qpZFFAHRDBU3nuk)e?{KJEGEq<udK>INICtejjU(=54%-^Bl{t*zY+_?B&9
zw5Vgq!@&id2N7pGJzGr|iOd+k<SbuS1SkbCo7aeqd-yy<7`^{isQw-F#@5!Kse*wq
z3DnGN&o~(56h%$P=F#Bj+}th$mNhssBRP3W#af>n9vu8GD9HXvziuRFNE9F-e13K|
zHag15r5v;NLt|pRks4@-<K^Y$1{%?9lwma}RC?@ASCeCHWO+O_o{f!-K=%#GX9XTH
z96F@))KjthHJO94ZC;MRRd01+4xb>0;SnPTio1}WO#Q@sLuqrT^APAy#99xJj{N-m
zzfIjOZ*3)lRwblq4rqI!eY<ZM?N7tL-nejuO%zs&mAC4Wz5y^FSXw5eZ~IHVCAxiI
z`;9VQVl;@)n#H~Xh#p+E4>t=fZRV|8x4PLm#5Z2Q?jXP?e12glEN*Y^Zmj;*yTZRB
z=Qg6Jw**51xsiT(RYhN4|4Y`X2;blL!q{#j6%{TiALb{<C;V=v{K)0Bxom~bIwvn%
zMM0G;ZV1~`HuD_)3veJiUP+BJMr+LU5|H9H_q0~J{-s8;a&ulwzBV!0eu(EM>A#mD
zex@s|`^UjnytT!_*1iDG4;)E4*F1IktID%O!%_tH@hY9pkh;CcSTVZ^LS8OML3tF@
zZF|L1E71<CywaYavM|0s*6MMxB=zSFXj)iM0-7nNYfYN_Un*tYOO@t(@j?$69@UpO
zGRH%<iIr1NYa5w6=0q0%27WYo$9Q2|V@5#WnAsQY|Da{e+ttaKjxJnQ!f<6i;=zBM
zb7=3gOVB{N@pJz^X;*201P75|{Cx?Wt(Y*S)TAU!E34@md#fe(2g*sOQS~h?&hu?G
zacV4>N#P^%S6WFp3~)#ucgLQc-n@abd-k6k=QJd{4<uV~)nI&bxo7=9WPHz>U5;}4
zc?MbP1w-G+ko#$qdEQWNHup_5eY|KYfz7n7WHjqxB{R(S>$}zh>`&})0xADsqUqme
z;~qS=^QiA=6ddvfv6gG5>p*n<c7bk(_tq8#<@Ic~`Av{s-3BsjcLHB~3wNJu4Bd4m
zOp>mwm%qgL%~A4fA0Hga{wMQE3=S=;2`#%)*6`fB{z6~7M@dASF@BridvHTbi+_zD
z3?RQXem@1@SadHSB<2GxK$+*p?RJmPch^<puTMG8-tNrQH*}s14Pa!5qieq-xc&dX
z+Ie*7FCp{Tr!k)2rm)uIAq|6JPTUQ&C2g$#y%-MO4%w10+JAyEEMce9o0G*hCk>0K
zbaXRR<_SXZ=(H6U4>2YOFB09l75ovTY5L~W%(Q34GS<g?>5ADML@avCkMH}m=fr+V
zQ$}-&@e{iz79$3TElhLn9NLAFy`K5xyIQuqOr)gSD3>vR<{&~FE?9kO#XDIs)KM#b
z>V#&ozlRm7|N7Q$aJnD3a|X*v6XD{P#2&7>Ziasd^DmqqoqVw{yW8Sm<XdPk)W6ML
zol&aYTUS@t;A9i>ZG(;uD%w8H$r+>$UNNS8x`vG2Zgi>x1L4lq;2Ki7V2=I}h>l?1
zyhz_S#32{<N>c&cCq@&(QY?fM**Ul`_EO}pC_ya75H$v+(e=T4It~X-N|GzimoG25
zOG~V5te7;aKAmR%PMW<aFKQL;9hj)~&`HYAPpsE5GS2akPTf7iqNih242il#C+)(2
zvU25#0{%#TMx2L%fe>b(U0~Z6Wx@#&Oi;PveMkYh*!5~1aD#!Ty4+JN)}MVPfq5&~
zU+UHJ!osBisf>uck+VTiS=p$ESn7RJ>Iq?c>e<}|6HWk}6$*f^&eb6;s`>471<t_D
z*1PW7@^2P5mS3hT{?+p@&+cpGjBU^Pc-|^o!I9~K8c55*Fw(a+1hdU2S$$<Lr(E1z
z6M3ifq{BD;q+T&I{TwZ@Yn(B+vp)o$E?e)wsSp=wrdpmN-aPRCwBHTd81kTdc`_E^
z`3-OmwJs-9`pnJE@!MmAyyn#MI+<$k(kDpDjEIiDO~y`1hSrvE>Z$-A$yG16RpcV2
zS7glTdnjEEiv`)n#t6%G$@b3Q@d|Gimb>7saEpg{gfgnzIP219x=P`S2uSsY*bW|7
zo040VIKgla;<!J5U+#nkb;1#&61VE`1BVR^A8&}w7Rv5(<P`?p&cv~b>X5YEV<8MY
z42=D2mPfTotGfeHPy$?Mi-*U92N-cya^7BIzmdD3OPF{j)Z^LQi?y<LW}6F{EGVC3
zGc6|c9uSaT?G<AntLsL=`F*2#>hCKJUh2F$Nf$PkgO|p~#)jPDx(4<LT{eK`w-Tm7
zoyC$Upo1)&+wQ~GXXkqKafv<db)ON1u-o53qkG(AdDGRp`%iE>+y5RPVV!BnJYk~g
zXcc~=ZLwS{oeDb6TQ~<|L>c3e9vm}L3wwK(GiM9)TV7AbC-^hQiD2crHUEK{C+|np
z_j`#SU&z;Bot&O*O_v=&RT!7qqdWnsN}+}ZQ6{Eez@66A`~h;O2GP%*GVr%Kh0sv3
z9WSG*wnfIFX@FnaJTpD>;So*)Vz8N`zr8s%-FKe=4@fqDKR2%Z>-^2h^Il&c4Y>Zk
z|6(gZ<@Bex^c>?WuHP4x4k8Zi{E6{#ErF_uasF3i8>&T}h99qE$XT8~7Nn++3=g*%
zI8)gkQ`H!`M?m^+rnjjDxY{3dOp|zy8fj9xU01D?G~X_-Q%+7yNa*PobMnBVdET(G
z0MC3Q^@j=<=MzV#<<62`Wwn>XwU+jBlCM!e0ahip^0w0YDJ7+aw#RkH2!$B+`}5^P
zL|#Sf0U6O#X4(q9J^>_YU=7BtQ121k{u&qO0-KKZY9N%Tz?)0eT{y=;*H3#y%=DWG
zmq?Tw?9J*ze8K8`E)<tbWhp&inMPqqvV(y-WnM4sF>z<HDD!jKcb-8<wy?t>BFXa8
zCuqQ;>H`8QTOX)zcXoHF(Ez$CEh(ny0MP)wq<@>Ws_Z+-w;Av|YzlKL-R1=B`U%C4
z2Wqp6iaMGbD^pVk21i9Sr8%kjWL4CyBFq1t9MUu8J1*dhqz*t><lZ#l#^$<>mCHcy
zpmg1WOyJ#nJRu1Q?+kfZ@M$i~zAVDxa9LSeZE0^|ZZ=h*k7_MpYj0m|?=dhw-qO-y
zVEjHMpFdn($LZzJgPFRy&^eeB6qghe6A%IEpSnWBM8ieiNbIFPm*=NMUE{obejDnk
z?2<ef5SCVE`b7D-)N3*<-#hqP<Z@kC2Y6PhE}mjP$LXo#NWMB19o;9*)}#>MgIrTK
z+|ZX;SI?Qbao;Bo{^>M}GgVUL`1<vx#@mVLxPFczxQx)AKE1X3n0vof=)MqojI{ga
zRYpnPJE<4(Ah+0{B@K-H5sMoa!pSK{M2xS+>QM6M`y+&tllv6?%PUFWE_M#5vw1ry
zlL-(KbVM<NQ~fI+oZEhF?Ol`<6sk%JZ$EG(#s>zDvHh5;;^VJm?EE%kqI~-+9(i0y
zYyy{PDhO7Xe<v`Vs7VOpVYK?3ZPd*sMLCE#=@WX@heUpVQ~lu<G4aaDVfw%tfDcGr
z_pb-|VT!3A@JVdX+Lr=~IR`^-(4@?ROpxS`chJ#+lMUmon*vA`6TP3m44tEf^!%*j
z?9Ls8U{UYaFUAmdN<O~6wba<=e}qy1y@V{~ZBf9$7La8a>>VCqb^c|loBz#UYA&Wp
zz|UU^q+Vd+?=Ug~QUN{W8^)6I@})rSUPhbnx>-wRS-=w4DFmH_IK`%m-Dq%lcn}yZ
zq4|h^S5t)r*Tx_Y5}ar*2ZkCei+;LH{acC|D~H>DQd6_UBr?Fy9}#)=M$c7|9!SRw
zjsYaXNAIpo&9y7(o-z~R;#+L*bdC+#bhH$|VLaL`QbT?N{p=3<z~JQkCh?=Xo>J##
zJj8Kp1Jo_p$ecofKA{To+kA+<X=T+f+(z*uA9B-2oQQx3wN89O<;NpD>uSY;q7U<U
z@NhMWy<FDNX#d-eh?r<pW|r3a{_Yhpk;)`yzAMuOU@Suvi^2U7we(t#*=7?5xGwMt
zDE05x?vIf$+5>f}kX;e!1ql=pu9EzlFU{{|zTJTf8<P%&KDy6KBBF@dUB~K}fnUE2
zsH({kz_(sBP)oJb?>JVam-~f!G8f)*5@D08HaVwF`_q{_Irxq7nfe-Bxloz5eRYt>
zKv{@nSs2~M4#>*>WoT%qp(E%L8;Nf12Vu%aj>?kG8DM<spML%D;fwbQ1s*y^*L6wo
z5$3<{r#-s<N)x~R5;WUz%TIV{PP&}v0lN^ET&h*FJO7Uh@OjT87?9xNnI5;qt!z3a
zG%T!+WH)|*<(Rc9!Ob&coQ&YR@Z8+;4Rjg~{$QUtDUnZ!&lst~|3bEOj)f6`Gew}q
z<D{;rsj)hbJY0xURaYrGud^bJ&&y4-oLw~rOW(;s;3==y?Lt8bI8nlM&F{A8+23t(
zYTXOFCnrtcVVQ^=E?FG3Wh9yy8A*&yI2>Q2!r*t&a%&vc0;Yf-EG474nzePk!1d(j
z;zmBfZ%}%a`_}7#-4x`rv$3-`2&l=vs}Bt~ln^(93ukvX@Od(B0C5+;-2r0r<y#iV
zqkiocV-VB|U)sm`_y7|y{ld)3$npDTEaAC>#ra0b;lWX@-8K~igTl+#Y;4cEI-bn7
z;`@YB*;;>eR)6^zA&B@sJUoP22aKcBPYxhk8ygC5s`rH2=x$(lUcxxXpL>DJ!__{}
zM?u&A_;C|-IBuw)^R>=~+C+|vm|QP5hW6F+V3Sh}!-E4!5V{K9G^eH{>DIcS;p2WT
zs~Q{8MoNgE7@6R-9vpzkYiX@){4+Mq@-gBBV}|^xDon?!gzh_}fhFg4eue&$GIlDk
zp%e44P}U}aiXV6C2?u{^N?}EHYPe`}+bb#dRy%-h;m|<*06Wz!M^_gKLtD-HA_)h=
zMSWw=@q`yoKt2Jh;64_(A_nyr-i|{XhUyFk5&XyG5ANSrP!8dDooD3de_*3zVKL)v
z=EBccd0_hoOoxCy6ihK$WF6Br+Z$5!)6mxRa}weKsw&VJ^7Ct`+EYAWSch=9IE}YF
z+q}KPtU7v0aS?Hp_sEVW<TCG-W-mt$Z@L7-K)9{_RXpGR10EhVeU-qpFVPh_lKLuL
zH8h!Ydj6rF=ITybuigxn$dn{jc>$(D!Fz^E6aeKN2xt$m?iAPV`^4t3BPTa+{x{4n
zbx2FohX7Y9_VTKff8a_0gUPY3j_JAkD8Ue;2m{<$M~wfvQ8{U2GZU7dS)6SBh95aW
zB?lxUw3Ab?Zb{&;MDzYqYwvvUSN)4{Vr|&>ux`K<H5Iv413|JlvMsQKg$b2oWZ1pn
zAQ>QG*0^mRue~KT+a!lm4Gk&t)-=p(H_7AKtRSbP-k@F+Ioa$#1d+G;{ioRWN8nfh
z-Mb493n~&QK4G(}_d++kEc%tIxzKs^k2}i$cKkSMA#KsmAJfLhO1!zL(&dzIvX;%h
z;<2BXm)Mz<7Ko#NRbqoQSZ1UjIA1;QDo)V9R5^a&agQ`-Bzo{5QBy-rE*+br76NAy
z^3s!sXM`<xPJ)3~26Z$CiREF-R918u^q8OqZ&{86Kn5WpCH41{x{1(7v=16V(__r}
zR>=k{UbZyBtbWjlcXSNo0-p6o;MQ?)B%#WzUHD2<`cFZ&({^$=aQVc~Dgpy#47)FD
z;CATqa^b6}_@O1ZfUgAn0E+-A5PLw#2e21H9$v#XK~MklnDUuSvOJg<NVT`ajpSfD
z_S<fN55dF^V9d{g<_t(-nSDlHUN7LcDJG_x<}A@Hq}YpWP?7+|uJu3=P~FsMrsB@7
z5aZ%<^RWGKa1=LW1rbmiX4lvdK611*c>{SRtPm3Qk~4Pk`$6HJB;u%{`BqB_0|CC;
zi+5EE;fkQwKuJpaU6xWvim2wlX?T4D^l~RnLjyi>0aD^`*_o-vx*E^VBKDp6I4UY#
z;1m5{Fs`zq0-gfcKdg8+a+ai83LpomYl6qp(kKf`5~srz*ihPE6&Dw83%cf{7k9~P
zXJ%#|Pk#N0wfr_-(2d|f@B`MY;oi~Qhb*w(1!F!2Vjda-nnB3?`nSH$!`}Prrv=G5
zDyJB2Yt!Q!tniSc=&TIi{nUtpJbCgL)*Yb8evG_7uemU{L`*<h@Sl_cq~AiW^LNl$
zz0oU9(JWBC8o_5IfQ~(3+P!p}e|DE~uaMpEZ#I6FG~iM&ZC_b@Br~Awunz*LJ8rv8
z=G6Z-uZfjnM%G^F>8IssUzu%GLi^orsAHPtu-&4sob>!u0|qwal#^lih~uD^m%sWV
znf@^=dvS4zI4bhzPkJRfWj(k5J;NuH&mxb_-(4+r-a>ZnVqWPP*PzsM^10P({>;SL
z1|GlDhT&-@%R8{I0oBV1VD!K|fBo7Y@^c<P%B<(_*$nGvw-{NO&g&^JFN0o!y~r9R
zDXUCcO34k*gBr`#=Wmo3fQ#M;^=zMx*GL$hnXWZFzZcUfFj9>X?BgRLOEG2PaU8p^
z<N^$L;(Llp8m*Jh-@NMWyDIs*rj4g5F4labt{J@ls!sy;InH5rVSPPBJWI*ODOTcg
zaueCuepdwi^5LO<#PWJB9A>rZekUKuCMz2+BT_+XV3lx(goJ#{TcR08L&S0$dG9Wo
zcRe=Z?I5p^xbE1bVj}5zh<U+p`Hih%1qt#>_>z|w5mAsAdrrExwPj9G)x8~zYC)o`
zM`mbQ<5b)HcGAISM(H@g&0(^d+i+Q#HsEjy4}kgd3MZh0Wd@ti)H)eAhoO;^e>d#W
zMno{aHhd_6oX4$qwT_p2#tOKhU(yE0E6(yIeVY*xeHFX+R({#Ly|Bz^2|3I+#H1?4
z)BY(5?GrMTBiplg(K9jYTX>^<zU%kbAzypPWw*0`5~!tUXg(++pb55<i4RL@*w4h7
zq?m!27i6bP<c`HOnl<kawp0m$a*u`jO@_Nuq_EhBCgdG$+*itb$6F3&JgXmvR(}Bg
zM%+VziVy|ZE%e`KV^iS7>Fi?RB<0hC;|F8HiQfij$Vv72C)w5GBSD|W`|mBzJ5ReW
zHKO|N+>YO@U&w?H$rLvnFXX;67yt9cfnB12%k*S_f8`D&+eI#COmOE5JM8Vi+wzjh
zis-yDd_HFh?T+G1mFv+?w+aGSVs5_0taW+5>m|x;Q>a_;ww42#aK^Toh@3<~>q2rK
zn1MchqphH>um9fu->9dQVK5F?uf<1mzd*T-OJ$%Pz}n5rID9zQwUURUtfG7~?skCG
z^ohqke0;Yp=j*M&0REDTb_1I&L0jBG=<Rpg)3MyD6PdLID3NIlZv=?|e@_x316qeX
z?K1n_Khe=fl5fRT?Cv5>PcoE82I(0X$^~nV8@kkW9B&}tsr@t?<icKpiGsE@4sOE-
zPtD904JoFlUS66lj1RP8AJH8Ffg1Oy1S|w=o#zJEbTMdUNe0HxHbtyo9WEU9J1iA`
zwa~6Ot=%J8VZ}qH(QiV*%O7L|Z+=i5kZB;37#975h!9v23$5QOEYIkWtel*rnp(4c
z=lWZTmD9)eE(fPoK)R%3cxvq=FjKmHyfulpb9OorW$w)P7MY!29vhr2G*}7)N3xk(
zw`On)U0N@NH1sU;k0!=sf@N5U*vpqJiTv0|mdcES{brJU$z{nL*meYOX%3AHSlQcI
zn3>`K`2+JX9+t_3XEXJ4mZxYhVOK(a14jB7YZ<9q^jiqs5q>1|J8mVwq;!jt*O?SK
z+ME<U@n=aB*>%`neymqkYPZ0mwP2Wu-Mv&V0|6k&Xtgf>_?H7sl_9PK7B|0}pC=ah
z;k1vInqWn>R=UmH2cZuc@?A~2`v>>f$oU$8Ysl=B&+N253@&nV+L4|4p}5UjGsolc
z#+lZsCPz91z$E~mS>Ca(G_0jMSZI<NTM*wP`4Oj$iS$%!tBedj3$n3vvK%<<V|s~p
z8cc-?9|ZZ-?$OA?(38M>uyaJHw2(~kPRM8w0Bv~M7OxiH;oP^i`vYl6u$sx=en=bc
z_s3m{RtEVL=RnCu(&=?>|3jc~yD#3v3q5D5;v7m_XcD6G;r`vJiSItnle7Bwb4knU
zbxd*k*qj@GPa+?C!*rx*y7x=hv-RTZqFtM{{|M8TZDaTt9y*xRRH9cVov}ivoqjJG
z$R<RcpGzrCoo;cxe``)9jKfIO(cb_wT6&(^I(Y7fxK?;LdxfoaR(#)cTTHt_c-_rU
zM?par2~0w4_YrTLA`}EHZl9xijPO5tlyRf3j4c)UB=Ydv?^4uHi{Cj{WAUWztO<yT
z@d*jpxY-ee6mZI*1#++o;}nrOJd8uq^o%XGLEsk^AhZ$$p_-o*8a=j^HaE$g;KZ?Z
zzb8OmNlB?k#~>n#k`+3X0u&ZW`4cbDyne78Q9iRfYZ!8%{H(^TRc1Rkxy?L!(8)*?
zC+OIV2$Gv2tXynikzxOm<ZojI^kEC8upWHYF-Q?IdKlHARq?SAjNbkIrQFZ$Hlu4<
zpG!(h0#DeBiteiIUxXC(=L_tOX>=qE56>`&Qj^z0)u<`+*pmt(Y7ysx5^+3e4E_E5
zV2pNMp?}=?q(@9ljQhG`SmgS_%xE9uRQI^#3pZBejYcdAC`B+3<>p48F$gxp^W*HF
zCOS(>eI97}rJ4!d>priD%Fj6d=;HKK&#4Ir&UBqYUtA<aL|ib?kdPOi6=x3>aB&fN
zB^5w#sUP7|1Y;n&;&*3?7I<U=>+@%BLho@yM8d^T;o&WEaqLzd)*Slw7k3Xszqkqs
z3sK!ipr+ECsus|$YFXHY|Kbw-=`gp_l~7<ezydaD`4`heT%4k%?h+4YQFOXHjU5Hv
zDe1D_9+XKP0FR{g2c%ZKWc_D>6uVE@35~LFh=_@??qEMi%J_Vteb2?=qU7Ugihp~K
zn)?+U*Bno!#|LYA<PH}Z^^+$7^*Z8#C)H9Zkp{Z@RF9wDLa<XeP#!Wo36?dyx8sO}
zgoY-tnsdS_iuteF#Xzjs3I<wnm}7+*sOf1RvXXF#4qif`h|~>_e}47op5IOW2R|u3
z@P6B{WqMAY<}!ov3FTwX+t`0oBVebAd+UoF+ivaVBv6F%vgm4!FF!6~4)F7m5h32B
ztNB{x%CfSmAX!phfpInp5?^m=ckKLcP9|CL<}cWBdPizz$Ok2#$dYilwTl@WKYjL0
zeUX8bIPvCxwO8KOySAg7{|yZJG!ADyjd-}Y2$j7xDR7)F5fP`C=|3x@R;09VQ1q<q
z53m;^h}=V@E_W>)qU!w%@otv{Ry;xuoF>okT`_OF?f#Q_ZCvhs_VDCv^aMk#K0jc$
zITg?vq;9hFrjs`HkfMT>@q?Iet@wzDvY6hup2X_MFSdNBHE8tjeH0p3x?_P=!^FVg
z<+?kM7ltI2mp2bht&i;!lQCmde5CYa`<NyO6Xfq9JJv1)gcqvXau~?`d^k!{MaA&Y
zR7c}EA)k}vM{$-LV3eTO1LhUC5ZKT-`KDUhHD9Xxi6%pTaYN+&`+}d9BT_S^;D+98
zV7Rh4GCV}1^~Cz==GDPQeBiq3FYU_0>gt7v8GiZ*bghMz>-)l?`QeC$3A}?n8NXXW
zEgCNM;!LY|F+(JC>@hJ7G3F`DLn<ow<)-JQD=d@(?&HU=WA9OU;FDpGw6uS2T_`RN
zHOuWpqx=D3s>ft}eZ%6z14H877jB;QIydTwk2B@7k%zvY{X{vEjs-Z^iB8gz60>Y>
zqWi>GI+C+SQ#teHnsd|KNUW#0l0-z(`+9G9j-Ikbic#BZUvovW#PlrfLi)zN9V*`0
z7dypq#~=haoBBw4I|3bP$~$k?VT~`Hd@Rl~rOo<voh^#rS|_R~z{T=ST4``1eE-PZ
z2u0=p#ne?R*I<7e4H2s_&8o!EHlhhH-Qs(rtjTATG+u}DamwtwGv>@Cb6VOXmLEcA
zC)pBkuEkY@IymGY%MWUaWAi(~2IYxELX-$hSC84jc;yCiXZb<mGlhwh4>Zz}ulXGh
z?!?#k4h+%SW3~A=eo(ljQvmj60eT|ZBmapv&lbx&G?~@3<)V5P?1ZTT0v~Od8uu$I
zBA=Rdc*(HG2@<2JF+*~H{%STTa3Q`r4H$9Q5+<SwJvg)-Z`<@>%wloCrwzD;Ji=3W
zC_XcX`bIq((tbLBEH^D#quDv1G?oAMmD{r}9DhsdcIS=X+11J=ZbgMtX^oti+I!ZR
zRpzE<vA%Vs6&Vyt1U9Ua4L<od%Iub(#*;ihPOWrG)1o4eY6<oq1adS{EOeHh)j+j4
z7Ve{&;FOVOkd|g3V1G!Gl4)~#;kV`dljXDNB<;5cQwxVUS_RN><>jIL>Kht96C7N3
z5IpT8q;(~VBhzBwVtVoA5p@|+aBJgxEASATQ;=$TZv~v#zjX#hp*BXYAJ|c@UCzY<
zj~JLXJG}aF=60*sN&k7heLHbP!Dn^UZzr_V8oX8|Ba|3iuOmds<bt~p@Pgs%XLP;#
z7s)o0seP+Pn0SPQFF)cco4Yv)zKFPTT4tw@z(($?j{#8fx$tE$XWAOkw~DDo(Nt$?
zSs5DK*Pl~E+%t)3gD7nve<Qj~j5#Xcf-E#KNhFhq5Z`AD6O}~EjET?7Q~cEplxss_
zdhK$nk=7CIw~~@-5<fFaw&cjqS>_D~`;fXXR@jS2K|gu6FAtS&;qFnQ_%xDFR=Uk6
z7BCa-QDOu%Gi$SZ_||OoDXsLWJ*N^q!(t@*nUf;mzF~yz!(sU;_Mu~PWt5zW$%KD8
z*2u4Ss!AF>7Jp-Fe#l5l78Yced-b21-F=q%aB)e`Wc?pnT-zcxB9h2iIbBklwBSB_
z@x9p~QP9a~aOXYM`%nql1c`O<0{8r6SoDnGw#qx%JQr5dr%zv#5p^u>LP*MMS?r3M
z>W+@~Bjze!-?seP+zb7oey}fWT2ziQOmLtSu(Bp&V0m&(js|~XA>;(~XMg7`dl||!
z^FJi*8M3Q`b-GU@bit^fboK<G3TX4xsw}Z+XQ8_taI(0t?s0j9TH5MRoBvNC_pY@|
zCN&6E9&<2AynEN()VAX$Z=_(AonF2MX&Wdy<CEz((C!ljc$!njOpLcpJ?7-$s&P6s
zH8Ifu%&%j_0+v?Isj8!m(a8Xkc%@_(w>c3d-;8D1>m+|!i(-q;_>cncq;xxr2h2SJ
zPnG`-%d9W=#Ds>&U3X-VCsGE{n)Y_-tb_(U@TwPZ)+t?Z^ocI}GKCR@WiOuC6{o=$
z|3Tj94roAk!@}^OHPOXzSYP{iTp&JbdW5=?DL}D3JY;j`%tuEjUHW<c1&-3^?(>NM
zX54=bif-K$J(fL|v3#=u*Pr0FUFRz9XEHPOXHM^N6VuYusuBd@jvRAX)l-(9<dOR-
zs;*E$>gvE2g6bGtBlqZDQ*%=Vfrxh2W!>!Wh7>$Vw$V`QL8<FRW4IYR)=i&rn{Q-0
zmSu38?z2g94a;$04GsbXAovDa3>XAMJq&8_1eaRzdU<CQr&B+F-=jCd7bZkqiGmhK
z2Q_M~Jb7Q&lpi0w0h%P3cvaf0w^00hoA$6y=Ss)fJJL@q<x3gwS4ly8%O2t|O1Jd%
zOSgM|P~W8Gq?j6-scS02Ok|6<8^tVT;Qsvl2f>z@?WCXYZA)V8t@#mH(u|)zT?{@Y
zFAZqcnGN8gW|`<8uW)J{S4g~H{I|Uw*xMn0<0+Di*J3sa{5=!$Qc^xw{V>&4+6O+J
zeCL`FH3NgLg&{s6epEsXIerw~<Hvrc%7DV`ItutSmb-3!>8I2+;Nj*%MdG^Nv&THb
z%Ra{(!z#8r6so!6CF^cHO!)Tap!sL1>x}!v<>^eD+{*eRovmDB2)6rY@$?Jksl!8B
zTIa)y*U6v7+Z=Rs^{*DE4`GwJWbzHU<}_+|yXZ}T?hAI&dY5OhUu?=Xv7^?HE*>M6
zyqbT)8s~nGnR3Sk(gv<RaqfBWF$o8`+7>Px+khek3W{E?TN2pvH5M~Ayg$5KGD3D<
zAtKWZ)vmuc@Vq3_`{W1j=p}w`os&<Jgl@}8U^)Q8Y;b1xo12>;N)u%kc3q*w#5nEf
zut?Y(y~ot+N9OiVpY&+GD>as|k@^XJ_oVs&gwFM99jBAP980LV3G0fFj$k1f9SYjR
z!~fhPk)36&hk@d1N-EFQg_9V}+5E-omaO{><@CHk3LtdzE10iTyYFLH{J8^svk7}j
z0sFHX=&Z{bEgeK%Ud&SXV)`nFg<KEREJpo1iE8|Q6+b2l=;~k*%Ql2*kNxV&SY2>3
zB9b|fY$CpRErj_pTZHH~B<Mil7oh}qhqnBen#mMJN!i!eb{*zXhCS)&>CjHZ%yXh3
zqo&-5W!YpmcjH1TM`~j4`te7-FO;~1_(Z!;A={FXo%C$S?ZL}Yx8pgQ@Md@k2`f%{
zn)NCcis0rCE{?ULCcK*F$2PZ#&=`zu(V@xH1U1QFKMo<<IG=SBE3)If78e<M*xpn7
zCzmW9#jE~dqAE{M6d7}rWNJLGi&N_~O?A@&3W)f9Y~O1#mj+6^FI-a~Sz58a`BP=F
zIsV`N5CpfnJh;~cL9bQrsBYZ$py-3Lyw>XUTg2BGogjZca>^3{q5PyQUT&&e6`l7S
z1t@ZOnWIK7p9HsZkfmTCFy=yg*$kCkbD$`7>hECy;GkvJWFFaXHW|F|6%%DCe!b0T
zPk}(y_;Q?V@KS+LLcE!?^bb(~eZ*b=*)u)(4;3|~36Ein49*v}brL%gRkocxo3VGQ
zK8T14C+GSQy-?QoIpe(R#ut`sj$T6T(^CNAn4`{5vRfnLF+xxY5MMt(6qs*4@E?7=
z`diRPb!+>DNvX-<ca*@C1eSM2Z%k>!o#xtqZMtAbeUH3lfAV5~dAB#)dt&0oC!Z*s
zkBO~v+LPtmn4|?x9-Kt$o@LP_2my$qc*%qkXLSEKUDE1IhjH<23v5X3<A*@JJ3OqA
znRSog2I7Tr7mCX^l7C&5IKO_GO#`CS+DGCSdfd8;?<uZO-|J5r7XRxpER}7Y%#8GN
zyN*5DRp^L~lCG5%n^5BM6TM5_LyvUv4Ww>8ohAIQUeTDE#q><gio7}4tKKOsLirR`
za|IfslDw<g-6C+;6#XZarm?%T_r7QhTqru5P56%}p6>=L{M2zMzRB&pY;T-PD@Tfh
zlx{y$%I182#tLyOw)W;d+R$!;v>jDKZhPG0o3y!FNYsm~UxJAV)uP+<@0p`Cw3Gs>
ze%?$SxW*UG4?;(Vb2Zsb&c0toiu#5mlB1!Wy6`aD$=zfSYfBV;`SP?oFks+HI!BpG
z)SXdKK;ou7CtZ*4#6EG5(#p{AP%tUmH;8)FC@Sz09rIZH-q;j_jc=GTwr#<G#4r7$
z+<w6xc~fYa=ZpE2Tga;a8=9K=<V*9T)j*HUoOI}hlIIk{4Vhvq9qn9PS|LqSM&=vh
ztkf@UO@@U?$o`09HwpFIl_%6s?=RegtBO(wLtc<RLW#RD=%m25%g)}9eEMLxF;<|i
zs@k#?lR41&wbpbJ8v(E3ttm2e@EePc1qa4%?F##=el8_L7VyT6=Qnvv8LkR(Z`*>l
zqC09TS^%ZN3@EU1?ubl4$Z0vHV`SUcHzbH+k}0}d#?Bg^8@S-Q@dogb)-L;r5tz32
zVU5NFj4U&&b9A=cMtUpDV8KsGX=n}cy}uvTrCnjrouY#5tXHVqxQ`b;0Bq_*+dnli
z9vD;%G}R^(w8~zx97HvDF%>BVFdyH%FW8VEMxCy&Clk#+GSp@Ax1k&u_=~552hu>c
z)*@$Oj;eIs%+NXyPL?Ae<9-09N`FRSTX<s5vCNcMZ8X#uzVf)KIf^CbfA>Qww-+OO
zw^qd*tJU&i%xb`_n#s&)n`PL-8*CpNSn8m53;w5(7V&l8z%L*mJ<Wi(>y6_@=gTx6
zvAd{TnHmn1GCzbUeHvq8NEEXF=>PQCbT6xmd(Eo!(#cFNzV;udZnF28&fUZ3utC*=
z-vLsT@eG2ohlhAf$4hv48QWX_9s9OD7|Ev)FoJpb;-6<y2HhGIWO(MRVrn7d7Xcx!
z9ud&$HbzcytN?4rs;aE)jq8W@Y#X(tk%#0as?2Dg;0o}{#``d$IZg`6@S|OlSKLLT
z6-TFUmjtW@F5_608|Cffv%#tL=AS&-YcW|AJ=!%Wh(e#$s-d$%iTBy;uCB!X>4nOA
ziF91i&Ge7r`n)1CcKi7zW2W8xLl2(xXdaw(@jrg~byI^|B%ic2v~g4Hr5H?jXr*J$
zdz;PHQk)RWi_T^ky5Kf$9@rL7v~3nG=fA147R>oT5sZa2n*Pd=i&viM<+Zumd&)ul
z_)$8^oj#2kz31CYI;>*B;DKAVZ@GN2R78|p=6G0=o7>soR8*t>NxP^zTEOaJHS`FK
z+&Gq5;)EQsivlXs(!SIQeV9d5Or$vIBe)%Q|FMM|JnPc*9k!To!;E;+ACWuuyX*aY
zCsQT1y-IIXelf5G1hqC{f^2LPDIHfi*mGTdHtPwEgvYV$KO&`=iJ=@1fT(*#ijzVt
z2Zdl_f7fR<buQ20aC#NDQBN;TX@KL2bj&*3C4NOw6&C68bTDB3{A$i^aj~2MLLHwu
zZb+M$#LeBie8!RSfi=1r%K)|RmrM7+GBXzsm$xvfw2W2kr^Y#eYs41}9bPsnKapT0
zTHou$HxvCQ>h!E!Yl7R_sY+$NR63!&t+7KwR@J^IXl9m0Z1glXyD#jk%{L!X(Dup1
zRoagKNN3TmRZvh6Kavbs?qO#q3yV`@wL1#?+A#vnwi=Vh7i!918bvTU5v1ochaZQd
zxO;Lh_gA`Sm`2aL@JDy|{~s5itQ}6%EhRp33x&VhHSQRfe@cshDlPgiXKOngBs|K>
zA;-IiiQI;Nq5k}^fCr5-#It}1bv!}}Wws^W2Nt(mornvMQolUqH}NZo?}R1J7?5dR
z#-qS%9|^h1nxYeTRaV`WZ@2E=-Lzn<xgFIZXn*zrs0<!h)M9ON{u$mda1*?X`4#gN
z`^2?oC13>3=W*#n4;omrw%<7&loDLuvp~JkqgCM<$@5FQ+Oht}RpjbVfYpD$$qk(n
zMHd%u4f7QvaOGrs#>LO0t?<qr9$>!s-8tPXOh3-QR;JDd_6vqo^l6VC_Y!YFVFne9
zqk#KAz1I2p{t9n#j)i1;`gTxCTI(+01^7$d68l{JE8^iJLvdAEa0i;Z=XM|z8=Ig$
z(`zI!sPie7KCuVq_O1M1zXk?|N=l2ebMgWM<>VEW$prhsy;7<@B}EsCo=iR&m^mlQ
zX9Gf*B=QJ6Y37wo?PPwHMUTQAUELFiMiBKqJ;SiX!PeK`|7~mx1Q)su%+}^s-KZvW
zFpLf)y!~o&e4L2rG`lzuLQ`j1KRO0Kd>WxL1X5;3A}K}DhRIG6854Czdm~3Ztrw~B
zflCP25SY|Hef?U`uWn@@{84V-4j3Z?^W8;Emx-;Oq!+C{fF3(<ogsv}l-a+22K)La
z_XBytYjZPK!BGLBd&(3LBm;7KrFgRDCNW{xJNm37*fc&JfB%O4-F!~<9ymD0^N!!6
z)ZBI+5<G~m*J)~b5}~I0p6v}JSHjBByRrrDoLf%CP#C#iL@Us}D~`>%9a|q6sj^5+
zlzcE`RQfsP1AT-6uh=#~7?61sCv4aP-I;SO&;ud?i3P%BPPpz%7-yNqhbE)6U+g`l
zt;s6W`xN`{q##v8V*+?dkD8iW`MzNWe4`Ms{FLnJu45S;mh!ng+saYzk1yum?iT-^
z6(8^a%wkQ9KYk-KSJR%kM~MG~<q6O5`$vdn`DE@70Raj_UNMiJ`UQZ=o?vgkEGQT=
zwKsKDRCt}8<?9<5@zuU`wOQM%aSp&GxC}rJ)HDC6sjL|r7MYyQHagyap(HKvRkFCa
zL`?Q$XyONB+luOVF6)ceGCk5#J)}t)eC``r!O6I;LNiuBUN{g+<NSL-3hM|{mo!0T
z?7bXJ{s&U`jn{^`_k4nrO^IA`8+|=d2Lby7VE2D1NRRQu2ZV&<0@KTX{9uggxO+D^
zCnpb96Gb&jvczac8m7mXs~SKoBqWzIWF5q~OU5=CG)RcbODm)xEo}=;e^zbNlXOh<
z@cu&o)lo+>3gIE+&t=hXW?NbLyo8+7s;hpGh5x;Sc?Xz1sp6yfeqVBbQ~~tb^5Sl9
z@g9~x_&Szp<|TV3M0{1(W>1kf{xe$El=F|23_BAriiT;`USek|Cc&b-WQE>}ipG`w
zBI~^{rGWyH+?oX0;BN+y3c<{lT~QG!@-Qohj)tQVrj?OBVXyzTw<YN+po$O<7L8e!
z-|~A3lN2CLG)xlMwZH@d5Em`8U!JH#4b(!1M$=htElrO>n*e?6=vxg;^u?2=#bNHk
z^zx^LCBA-AM%ziFZ`CowB1=k2nwy>g8w4s}?7o4aZ3Apg#kZ>R%G9gfGmlt#rIVx@
zKK%B=!08<t7Cmg9v!<a=W%!xX)g|q{y7uLzc!0leQYfW@f_nV-{%7uQqvPP1vA825
z)W8-DdOpBOAkNS;hy1Ps47GcaVz}4k36-)TjDi=+Nowg-Yy3Zd-}y?t`bbl%p&orp
zF%1idU|>5m_RQ%fb^TylZ<&TM+<8rlgOgRFH$ruEji;xG!y1=bnx7C75W+^UN@Va^
z1~394xw#Ry5HTDXyii_)1K+K4R`uMh)iVEfdIg2P(4&&6A`An)rAL`*KXd4w1ySOY
zQoyHa4JHrVF3yCD^a+R3*tW%z{{gvK*Pm~>pBp|vcZmMt;7oDkvZ1RoIAo3j(G;*D
zxXW0*U$kxn@ZyJUA<H}Z`*_r1uN)w(s+&Uipn23$^ouJcs5zngTuas6sl`qX688<M
zC8P-Eu{*+p<Ys7+4@wz@zrk|KFtV+<_D4zLC?Fu<*Gv^he$wh-0jkI2xS`<3&$y<h
z24FignA!RTKuURl8jQZw5}4gMYvB!j0V$@ONlra|t&8s9^XkzLL8sMH@^fI-pKB07
zAx%xz`iyL}C{W~N;1<Cb%(bsjtH4^c#qU}eT=3`%CN8y@h_O0ad9~ZvywfF9Td<hI
za|)kUDAX7=3&1}N{4O9i;2X)&X4Hnq9XGA`iZgGxfq-{LAK0S98pV>pqTKu6tF+aA
z_wa-oAL>;YXx2EL*L3@uh<>?PQ7>_vKI3Be@~q_R1^Jukn8`{j<CPj3;7eUH1>A#?
zn+(wCCD3X&n@g8wFhQ4KXX|0=`oY|tYNy^P%QL<F@Aj>49yh>9SBrA`4(uA)zaUBl
zQi|HT{`>Us77tGSC|Ij$Xqepg+sm2wkFl)x04z+@Z;jMm>HIN1qlkIxAugqKS^uj8
z(zThGpRQY|!zbU%Pj->byM6*ies#Qz_MO3VtQ(%J`u$)2>rejo!pM~jB_A+5EZ#l)
z3F%xjpD}U!EkM7%;B$mY$SHb-0+D`Onw2`n%$>xv|My`2cPfONYFmQPso;PX`eQ(~
z3!Fx{(r}~$A+ORnT5&L~@!{<qXaMy;o!l)V&Ii@#A^d>*?eVt7%&lo)uDMdkiHd$0
zD*XCc0C$s|JUF2YLQZAY)s9P}ujyjcpHaS`Ik2<K;Jp-6y%U`1{Ksx7K3+(_U+&6n
zed`4=Uwn;o^L!6S^F^zlum5XV8HtV;cYTrw7N-2<r2T>dc{tca$KtQB%F3Kd{``T-
zQl-m{_(Tl$^2=-=2YXqH;8uKMVh}EP#`r0)oU19T*1Ar`{8oomv2i@&v&`dkeSm0E
z^^Koq0}14Sw<M#a8=7+`r*guX_J5Zi(y$Adx~8O;8y{;*U6W^Vj$caT@se%D886oa
z!}FW9%KIlbJ6Xu&Z1JjQfl@@z)>P3t_E+)JPU6$tH#k@Ne56Q7WMqZaX=lll&z^7x
z+f%W&Yc{<3`KWFRjx7y5fO_2tAmw%EDqd3Zc8|cj|4#2&Vl{>wm<AeW;js6Pq;inh
zqAUIOtJiTqA)%*YzS17(`~F?!<m4>ahU8Rf&1i2A3x;@*{6TUjBO+ZLhrFvbWT&iY
z5ymt}rP1+KZp5$5Ewou3476=IdA0<Ul~iJ&ne&jkl#^FdOH7EF8Wsot@MI2>#C0Rd
z_IB9mABVoWCMWO8j6Q+E-{0_-V=amOi{|y=;S!h&U82r!3R}|)a%jND-$><rhc>)k
z=b%P7;y*7hgMv}Q2o<Q-rtCO3$-f_T0=l5Cu9BLQa#0?X@hk8G3bB^i=N_d&8A(|u
zbKtAuU=|S`eyOl>2Rjg$8-EH}*F!V!BT)VQtW?w04G$_&bx0S15>d}~qrl6hV*5vU
z)b_$U<i~n^52A%CPk_CmRkC$gS31z=YE<!(>UT6FrM3s)@x*qVu0q${UQaklWmH}|
zt)<L()j;w}x}kxCtvqbNK4WjMJywkh3$5K;FQF0q$`?!d4ay?`ol7@`9(5i(S`>!<
zbYloB69)HZ8uzHKUqZPH<E{t-!u-jyj+Om$4Kk`r^q9$nh;X0AIf}^|5ETvU{tQbI
z?bS%8-Zq>A0s*o!(R~R7k3<W%rQ^JcjFMXY3yQWydke~1<8bnGjSvstm-_}xu{<3U
zt=ngt7}XoSGas%}1uAR~<u-d3b_J@n@5(D(V(6vHXRXDIk=$9H+hhQ=)$Li6j&H(x
zqXsY^up%$}b;Ch64gZ8$27~+9MYHbRy9d`X3Ou}=?6l0Ge?MG?6!opz!eDMnP#eG-
zJimkTpye0xs}RqO=%iI^Ii%OEm!dQ`>uU=Des!+EqoJ%x&#H}=l=*Yt=AV%}LAw4D
zkQ!i4h05yI{s?x#Kkn^mw}8cfi5<x6{7=-Q<HT!j*W*pxDgTKYxF(7yiq4-k4@?}B
zb^q<y{%@f=b-ACEVZAwT%%u`>MI_ed5j)0_K}Xj{F6iWW=r!H{<)V_;ZSJEXpS{p^
z|ImcdlDD=hh`YgsMA+${=P;lPK<)b+3FI~ycXoD3XUN6)4vV$Y0>K-)NT?ps5wIgU
zt;Na7&JGU?ixsk)f8|nX;T^YRS?>8fney|CHCyj5WuwYZS_`mg6o{yV68}RS3Gk)|
z2O(8Q*l?H=iv!=5Ah9R-g5X?;F0H&=<>gD`v+sVxk;_K-ab50|1ds<0<V9uW@0onh
z#^1!CikN9#!WaN`P}#62+IoL072LXYX}k_iBEw%jLX3<|E4DYl9R<*3W~RqOQ8yUr
zneC2Uv$K=Ub<M8oPVdLdCi0D0?KSPLMrH#$#KE)A>J}rdl(wA2Y-{NEu%j)o^<cn)
z4ZIRZ3rtv`?p^Lm@a%p~TPo~+R`t8==83D6%>al&sCS*II2c-6dxU8id|o{y6L!P>
z5Ah&ki3X~?^?3rv|4f#!+TRRL_FJ##aU^`2saz<Xwjqg7Qx<}k%By)Ut*qC%8ahDB
zpAu)i@2ak;Sz)?+0q#wuD4r0R$`IMf!@(uawa5LCj*eGaRMd6qpT65S_KWr&VMg}Y
zo|Pw=88ma5yfVKFS#4adQZh>1dxC^Jb*tSjfIbCO3&?bWns(X^GcVsF<okCLj*m@g
zehWLin%g7uyN>nAq1!^PY5hhYQc`4t7RXxpXDxZLqg>7=2@&2h3k(XMpE-F}_Eu`j
zAv0y}T*0mUPDkwMUSn17yW%nApi+i<MU_#Vj-LLr`7=&|f4W9uQc^#`tf6Xb08L3~
zrW6Hu`y3h1KnFYRvN+!NoQ;F)EJJX-@|Y5uqu#>eh*KaSd2G{?b^9dshV0PMN95!Q
z8BOP|IaiDu;fMb{@VI89=ZlmzG%_}3Ob~YItwTXVV&c!+M`oHFv%bds-8}CG;uh|a
zeFss`3zgTI9Kgc@1~1!vX+f=cfQ{~w^=9vZk*4FqH5J5obj^|6$5!Wc$Vf>M-3|k)
zr|TWR-^B$4#P}{(^IRyw+7(85Nm7s>SM81;5)%RbOr0`;hO&vdgLVxZ#Q6p4rhlpo
zmJ~hjWFc_t_4ki2FD>{qz9;UA`+7q_!rlKa<{enp=PY@%x}aPJbkGO8^KX~KvFljx
zsPx0xuC8zJqfAs()0rhFkGtPhf%8W1Ub;!V*J1@%hk~|jN88@cp0No{$KQ6iNB<8|
zUm2BE`n`P+K|w+#q+38zQt1xqP>@Etk?v4rR2n6uOHx9*OQfYiKuWs1q~FbNt^ayw
z`DvCj%JZCa-+N!V%ifmK+ts=6aY4w+%*1qC;a!#P7JyF>qg-f9x-I0`)7wePZP5hB
zBLTqzihs+ZXX8QWNBOAFpQpaN2EHF<W{R%yYB%Fh9VpVv2~}wDT^ADvC5dk1q3HAH
zU(cwnyf#b6|7gdTW+_7{<(l3eJax-#Q%y$2K(2v%E@fLqN7HYyO%GOrA9FMF+6xOy
zUX6S5kOgQ|oI_F0Jrc{B%s%!;mI~(K+8Qs1Dj-M_=|i0UF}L19cr})$uw5l3rTWFe
zp}aleUw1AJGt%!@Kk*r4?nmB60GJ1QaAf2b>ld#5-hg(rz*e}He?mgOV+n3z3TXd}
z3~FgWw!%I4IqK_Ah)(!##l=itzVDYbwuFYX*`KJ7H@dG0-pa~K`-!^C=$YE8cXTjF
zz)A28m*OTuhO->k9{e+k8=`pv{hkj%(=X_-amCUU6?+B;)#lqm9LbjZV<o*_-5Wq2
zTmBgHZ~6<rq2VT#%)EldPX^4u^z?l0R?D}*!XUt%P5Le5!SZF_8<OBgp`kH?#WOqu
zAM}S0w~_u;tZhwSaFy8*>gX%E&ep}Pq#+0hKt!lkU{eifR-qKMXyay}ZK0mg^jR6G
zHd?)bec1dq8pbhD3$f}yzI8!&_qGGZv#n%dZ(+aQd`locWY)X8nDtNUyV+UVW?%Eh
zm~mw(#ze2)Vwm*%2IDe}wMBY@<;L5yusdGHNWq41eQYr=D?^0IVI`ILyQZAk7CU{u
z+T|t(M=vkGZkmOZtm^7$IWVb&_ksLUjG2!N2gtPzXdxgBj*kA!K+VSH?xU&k{5B?r
za(74k8aGNkt*845pkL#EUfyUrYl=``E`seWoWe5YQ%v#IPVUsilb*yQY=O1+;VIQZ
zey^TN&WIX3p=DlP$-VP}r<^&K?Ixo6r0>1*WEuB}Z2>E-Nb6;V@l|GNJB$M;O-|dI
zmD?gQ(b2`+PvGH&h-!YH4gZ#$S#M#=Kpm2&qP8Cz>=bOo0iT_{5MWRwlFuA`EB<5T
z`3vzMv=1NB$w8RJ-ky|{l&@AfVEn-T<>c3?Y05CO3FlTcq_LxoiKX2Pe8<e#an^`C
z_<K-R?pb$77qlyUui!Q>9)q(89w-V+`|QlL-L@;Amy!oVGL<H#rci<ce+*1{%z1g>
z&T-nq1g95zPVS$Haaaircx{UcbDfARXXNbs&WAJ7^PoyXs6M#aq}a-84e{@uPxE}P
zp>Y%O;2Zx`vX?$LI6%88g9Gxt{+*oEh#gS%9u%hL=H%-)xN?QJC-8b0Ia-~bbU%U1
z-ceFczAq?L8AvGRZ&sk>H*>WUdo|!RdP`01lNj~vGW!hG{o5@Y`EHa?dks0$aLe3O
z*49Z;AcvD?b`_X8kW!wVG+u3vmZ$6OGa~Mk_&;V}!={pQb#_P7?wOqI*tj+bl9KA3
zg&2xb--|ChB@6lOHl5@q7gj{hMu6WDzT>DEO&i_HpuTyuz;~!kZ-ge0_q3Ijw*uc;
zTVYMZZ9o`|jS;-C=N{Pl5J;!|yuzN|p5@(t7MJHcyIh)X>>$%%mM?`G?NFU#-A3m>
zT|cl8#^>59vQGm#Lqe4oURT|`@kWxyx0IXlxs|$;yse&(tsca}OS+NwWLj6wpx?wM
zOU=suc0tW}U@ZIZA|dwh&>0TEHyPRU_dmgN1jE(^>f0YbM5+bra?|sN$HwsS@9XQS
zpDpCR&kJ(Y7gGfXL`liq&t2oE|NGBCwBTG`{yZvLztQ8EgYJBMUO+AglbfF$jS@7x
zYET~@7;Yf>Jfu4yq28Kb<y&uW55s+oAWTe*XO957XP-=pj>fmPTK{c3J^ts9nYnSN
znG=WrZv?bZOgFsMwTVyA0rioF5=C%p{1Y#J=T!^GJxq9pAZMbhi#hWvU9fytiPN4C
z9O8`3U@iSq;|Y}gcutQ>L?JPiz!UqatR#>K(Qa&NYwPGGdqCp3NR<omG3&+DDD<63
z@snJd2bd`6{@6lohv6Ayb$DbX!~}S5Ao`XM-p1|hGb)ac*MW~y>{W4An;~q}!L8{>
zAR|_=+XhqHuil<)HH&2>awro~PC4r>gliuTX>cAD<QB}oehS{hjYV-mwjxL{%?P-G
zuoe{1kv!|aw)B8%98dv(<(Z)=8vFM74#xrHY5}g{$(I%%)V0^TpRJsPf0qb0j{mwB
zlLfuvWN8X@D|Y@Cb4_hmR#YUn&%Z)c#=E+&W?P>!Ke=jYSJBdngSX;WKA9~mHfUY9
z*&r-$es^$C=x`e@f`p1qG_0q5Vc!bAHF{szX((MrUI;*ufE*JeovxZ1jSwv}^`mH6
zI!8V-?h>#tEG-d~MRo$uBS^u{EKpgN|Kp-l@BT-8yx6L<CJqOO;%9hm{{hg{`ctpg
z9dnf#tiB-Hndn$(%OV7$Y4m4Jb)9AJV<;|ha&_=qn`LEv`0yb$R2{Bv-S=#8_gPOa
z4JnHXjh3?VG*i$ua(Jd?Ko-$LBXaNdKNtkjKrkmIc`PLZ6iW|WB7o!lhwce-0P0!Q
zTgP@l9SJ7&!ilP>p@A756nXanGjmZ#XZM49)YKvk0qwm%8R&6K%hd){uJG|HEzcT)
z&BNfBdHpmI!m}P`eg)17hl2YtBc0%m@yg;p!_7YvB52|5fPz<+7jG11JX7k1|M`P|
zSfB-~+r6TC^hirv1Io0%?f>gdeS->^IIqpPR5q71R;lg7&y>OrcQC?h+~(CNd**zp
zu8hdBkq^PadeyHTm#1lgK>r`KB0_;j+j7S|OioVr;QA(UHZpm-a9JP^PfKoL?llN<
zo}tU<mtLJ|dOB&EPDAsgC~>plnY{Y^QgkU8^hzHTZZu2}|CCjHk(pJ&Ccw})*73a|
zjf0g|URT%VSI#V`c#=z*kwgKHmohq^<uX`x>UXzLV=X^390`jBDJ!7+LrV@0$x}ec
zyc*mK(l~A-InU#R6Br^zL1Q0PmpGYzQo6*d<+a>A|EfYmfp9%CSOA%<^VU!|ehn`P
z3`!R>e$GTZb!GW9fIuodKLGg}Y;jO%4YalA#+tChIgud6Ed<CU*@XR5#U?m*t}mw&
zD=klH!gx&KIv@Cyx5vDya&u1-&_1w8!5a?#i~$wO^PNsEP-Q7k)dMGEJdXzvjf$b~
zlZ8BP2Q(8AuTcv440N}<0p1}SvN7p>%y5j3qax&UX(ej-dLjFzmq0>W8@;vs{;P_A
zwh5ykeC^ZyPQxO&w$ngf|Lu7`%P%;j9<b;zL(*G41t4}{ykOU@KA~p3xoRQ&pCr<B
z@aVH=V-dh<op|k*mU(>%LbNb{?|gKkl=u-DhK`|})-RYMwBY#E=|PdDrEO1FXL)6*
z$L28(!fX4I%k6jB@WHR1iQ~h}dxg3+D)7<377Rqm?dNZD&~AvnaDh^sj&hx2c&R~H
zU071$`SZ07C>)wj9$kN79lwUF%FAur>P&qvw}J5*tF4ytHo`<UCaTeLc?Ijd;io;*
zI0P{)@)%$mSOKy!&F%g)^tB&v20gq3xHb^;@W_m^C4GvFNd^j{?xFRTu(Lyh7dG-d
z3Lk}PPkjT{NXSKg_UxHs)ueq)+uLfc2fva|hNsRNO0AC%VFR~EeG5z|#Z*1hBWjSL
z90U3!YWM-rZ*6G}UGb8ziX|8^fTIeHZY{i!nM!V)EVre#;^p%|E5x<8w6@f(b-4mk
z7N}Uf_V8!&eNc6$`a;#$`Iw0!ywPL%m5S_70CgVTS^h834^(s|m8HTCCX@&Oj=<Hm
zk6v9220D4I-yy<6AffHSRxdnSXE;eAK_9)gF03Y&4uOgX1qQ5kTNC93i0h^ZA!!QH
zUHrF7?kAHN2oSUq9#W>qc<ExGrD*Qnyhrg~r@~Af{+*iRpW)81&t&bO*xDNNda0@N
zt-UEa;tM9SzuYUyaCH~|hasmHOGVC#+nNPilF)BydgZwCFS1kzRT`w#K6#(#AL}rK
zsu?`7;NzhnBiz+=<_2pF{DQZe)OTVe!^YtyOKdwAC-=)z$7g05z5lV+W*E(ZnkJ?1
z*oPECUvM|bc)VZ2%oX{083hSm@97X%QaXZHTYZ_u($-eM*WLb@NOV?}0~%s#{%GUc
z833T$%eDn@B6v9x`D+?u@wBzS968Q|e*^HJl&5u-@1Lc{<?ccdw;8+cBWmh_44Zoo
zADV8lxh*a(RBZ8oiTVVe&;ejbSoC<7$2S*6Y3}Py)mu297TPA5DW?2d-iIo!d;d6`
zSAXzn$CH$B4ocD#X9RpQQ1J$%3gGNFj+uvr9iAq>@AknA<ToG41AEroCr$~}3tmo-
zyw4d)1tVYJW68{<sP%+|7Mg;a{BrP9b)9?$@z#E9Au9_L1W^I&>W6ptHo_3TQCjvG
z5TvcWLxC0X{~a)-h;F^RoU%-9MWOH3+z5zy*bO)vk8>rpJfvXB0g)0g+o-p49@1y4
zF%D1j6f6?FKzY+lLqoUhHA?r8E`o4yymhM6fsEv2eVtT^Y+AtXSI(2|8z2g}ck;*%
z4vcGm0f&~F+8h{p=x7`|fA9e*K}Sc2Q1S85K6<FA^=#l|+C>m&85EtcnZY46lbtaz
z`23cCULFZ#5nkryha9vWegk&p8QrHQ_O$%;!TC<1M~^zObzIF&m+9#;;ZXfM?qp9A
z+F-LQNVmzJ;JUCx@J>ehYJYd>ath$nyASYcy)SQ`A|o>;fHl(+n_2s;tq7YCGfH;!
zwZM(CHB(o&-o)`r@-cNiEXnlBWLF)nkY(F7*72B*<vs$u0f5yddzfEGMV-(QpknZu
z%B#v0lx0W}NXFX(;|RTvo~_tp*#K%!h!6Yr+qU@e6AlheXGgMNvp>XId^Wr?3JM3E
zL(3M*f)tu++Di15u~E@naSCW2eBjVP3_+#%ccNNZP8%KhLPVObY~uP`103Ayqr;PL
zv-06^Ss4VsMq0NOEh`r(0&6_R%^S$e_yqX*k20SkqGI|uKEF!=Ht|@}Qq@93fYn9%
z#nMtWZ84)Uk=#`+#Q*1G9*Uqz{{KkUZrwy5TERkyU>O<NYXvb7Jn1$#w~#79ir}6t
zJsDI4F$iN?eA+vE<g=_?j6sf%87<e{+oP$fQF8j5dk|az8q)k6e22~R<}NOW5IZzf
zn}^_XesQ4tKP>=)#}Lvy|Jt2QNm-K_pbBa89e2S4%bvEb8;ED0=aBjVd*DxT9vyWZ
zO$}9I(v+M29AxRr=8wyGcwo2$(FZ(Z>UgwBM0E7P&`@U&GgK6>O_Bb9RZNI>eYC)f
z<yT1Pe<TVdBGQZN_pCS?zgav+3@z=@kt}*?=uB!hgqn50r5rqfo4s23Nc!iqfBW(S
zd{s-xGU0H>rwE^^KGeiVpGzowS?|9mN{lriQOfzXODy5jr=e>aORw7D0rEwG5P?-u
zJXT!MK$BFefs4cn%wD9|<-?uOJZxxZ7Q%E;8QOSWQOxBiXC%i2sTE~3qshO2^MWC%
zZzDLYtzQ)PP(c395G>+BT~;e=pFqGeknGAEQ@s0eMv$7R(7{-my6AHn`YGTC&ggpm
zRt>tX63!@pAp{4`a%uEaV=`$Rp8l(>+QnlIk2!Z!sn*20^RsZ>eas*PIh<=mC8Ovw
zEl#8lL-kM7E2IH1T`ukS6K?TQQFHT&IJz=+R|$g+H@#@nK+K-sZF{@w;c#1S$JC{M
z{|DcnIshD`T$s7@jgJES;qOB8*-4MQ;A<3jGvmP&DDi*ko!6dp){{O64aHK$m6KBf
z&IrvPVyv!RlH#XS0Olg{guNUTRtRzfmv&f~fiJAZs2Ene0CemQc1CgK&v2<jf&J2g
zt2GgK^S_U%{+(g-kTD$;RPy$%w~qvEQZGcZ0ZcT%>#nX&i$D<1z1wlI1|5cv7Z*(i
zjGqJazLWX}NP#%5>Lb_R^^+_?ks(L$79WD~Zl&s7A8nMM{@FFt{M2}8aS^mZAJ8#r
zB0t=SBXtoLTD>lG_-|!>Y4Jg?R$14N;jzhICT-6dxN%bP0DjGwD=96$=?^i?Fc6Su
z*}}#G>Q|Vd;Ol^+4ul7-gMrb}BjBz?JLI%HCo*Bm{8kVhog|+zpi+;DI7<CAeC+x=
z2Jj_Vs>I2Aj;Yw_T~{Y}r_R7JZ*om#y*nqmHUJp*RS4Hv-MKf2+tJlES><gu+K>T9
z1B|B;DI|a=hwDy)1lM+=j*W&19%Cg}Xp=#K2f05H<r;>CU;pZQD|eKOcTp(c4(#D)
zW85I`fj-yg(a{&@GcxV;fMZX(`PNuotN~QW5vOo{P`$-VH>6Y`yi<=!>m!F7LG<=m
z<Uk1vQJ}w5K=(1C{{Z#<#RYj}s}ZWd?5O#)2!X85V^qn{&+r32cN&5GvKjxjy}fw>
z6+1(OqU6_o?J+weI9zR}d?K0+|GWgKwCh;!zak74CMHhP!J9xx{Cj7zFv@FNGXDgX
zxDNYv-Rs4es;V#1fy4<tTMfDD?w>^e^{QjIN*o_%o3$DY4CGwH-##YKLx5ca$UyAX
zJ;cJ=vY!}>?&Du3iyMI5c{TQf;}Tp6J~IvBp~?Uxcpm(;g{lomj$EV1K#=fbG>`uB
z$3$2kWxo!rLV$s#M3rD+Y7oGaDE^&>n9khb0B$7i%w+Fx82Rf}#?39ZX0?-)68L&*
zt$zRAKla4CQx8Iwwtau-V~1U;D6LSId=ZW8CDJ<_(b95H{L56oQy%mCg}Q;=4hZA3
z!?uo(b^&tf?COQ4Icr->__9AXrc<4ffYM0h*f;ZQj-j2^d5?m!-#tN`_BE!^Eh*~9
zz2g6I(cQ<;nlwBcL}`5fA6DCgHs#-yOy3?T8B_(e8i8)00T=7hBj#KFHUIqw%9Z){
zFq*r!``2a9%Ku#14L^TDR&yf?@qhbP(D^X+H!o*!4>|(=GcN(_H}x3kFqu03`%aO)
zmgzHHeRUoB#)Gv)sV9uqmr5a1JjQ|m!jhBbYCkPh6P~KpK)k^R@@VAzTm<HZii$`S
zDwa9*ZT@%)W##4%%_b!gPspJ2gjOe<gwJiRfB!+z3;-qXWCGvGOf@Wf@2y<IhLUpV
zBM72tLAKaqa()t!(B^VfRTK)-FlorE*B2M_33F7Zmv|0$%vdiU&lV;;BK9;nFEo%S
zs1ccae3|kKX$5jTNeJScciETN?}81i70}j7yVFG(Au}L)1n!PK#*Cf^<EfO?_>W1t
zhu7l*D>H+#rjERU<}D;D8a|D_WlDjv<$Zir<h%PNSvwooEwqyvg1d7@M%}T$3>r*)
zjg8fz)ij#)7{>89V0TKIAn<|DZSQsXarMy=4np;LLW_krQZ`j3$A0G4vuCmjpse#&
z`17a3LK%Mv32z5|z-<lRu*7^Uq|9|w^onbyB~IbmVq{bd5`Dyf4x4tiM<E1fYifGm
zzeri$d!`{<&GOE2@44+fyNTl&=@qVDeS=xwm$}XP2gKy<W~H$IEC|1&+u8AKo!<pE
ziqmX(2L%5qMsu~Dm!`8|nfUUBF;v>#nwI$y(|PWVa~hZoIS19i^WlLeb!@LUU9;Yl
z;PwVd&CGpchU3g1&}s5XmJX!UlY^Lgs_veELS5spQtDyKvJH;=g{9xlV71)QtHH$#
zytZ8aYJQq1Qbo`41q%UZ+CaNE5OFUE9$t@00)A^9%eZb>gm<R0Ui;lMl|n--tn8Nd
zPk<MOt(OThyrevaKfKDxTwPy<&8ezgwLXU=I+|bkgD(*KU$e}uEb~CHl7{z3TgMJC
z?7pr?Um{%((VihH8XAz#c~&paTaNgb3_f96uieKEBNj3ON6T{(R%U_Yr70PhT6Qry
zMFh#&Kez7ruHHdL*^)IV1^dCgXn#?+Tb_p>QU(JDOr+7Vnjmp(RP#)lXE+IuJ}gKc
zxdVqb%vpb{Q4yHIvU?4Pd5p#$xP<vJgEmtbcRoEY{W)X{IgZodYd742wje|pm%jBX
z=-~s$Q8a(+2g<p)6bHXrYHQUyM%Pr8GzgIuEKoAvy?1X_*f)=n8AVj)R$$BGoC9!5
z;6p*z<2mGkK3tIaeY|-irEfh|1$^R!!J+c9;u^jN0(tF2rJ1lPy?&YhwvLPw?SdpI
z2vdjma&Mx@8klx)`B&9=2{}5DnptN~jwkZBki*1+>K_uq{bnbr3sVvM5KKfXWB($L
zTwyx|oRl0ts3J>xTjmk-qi8^Qt{DF-DH%NO+|m4Z6?ZczC`eb(2Vcqn58kA15L8y$
z>N9{E<>W-H%Ry|7jQUC|FYo%#ja(NFZ<i2>ggS{V1u|TuqteJqvyD=V*b+qZNY!zF
zV2{~@*eF)L`rT31TC|iiZXR?Uo%jrqh0Ys@0-^QDDXYAuJ>a9N&!59hO+X!a+g%Vn
z07Oz6FI4TUX}#P}KnV|8ArR+m3pju3TegN;gxowR+BEO02NwcZMhj}gX@px5|AJ7E
z2>^^2{!#wx5HSp)h@rJDH9ga*)Y<KKAEAc|co?`tq4OCe+i#0J)V<NG>!;d}ZEpnv
zg?JaC4*{$vLuL%Msl9(EanhSDc71M%&u=w;nO(m{^B4DBMtchhbtllc1Had=o{^b}
zjX+4f3hQHH1q8xq3i1jI)`JSDu6@VgtqaLd1j!hT`C^mr#87XIy^L#bW1!FciVaxS
z+Gnj>l*rl|0MjHXLVhTSiK9S~)7G)+EKD-HDiX-f&JkGDU%R?&3#w9Ls!D5PA^@<H
zK-Bd_$BU72J#1;v&nk(D`R2T7D9`k;@&5GXOHRorU`5}8B$Oh2d2Or5q!{?Vk>TIZ
z<ewdG8MZDuj@nOGzdm6%T_nKVb0l1%hlB0f<=Hyx@N^+;Vd<nK^*fPbah0ypam$pT
zep#hk-JaJiL|OtH02eIm0|NX7(u8<mPr<M4D0iAD^$mwyBsgfr+6)UUB_p))mHj!~
z2cKJw)Gd>`8Q%(IS_4B}QCtGSxUCOODJ5uDoZCR;WjE>ZA+cK;rxFf5IHo?U5B|?>
z<hvh}FSa-%FHic){?@fW*b;15<m9o1Qw_dnVjfZi2$e*D+qMW{Y!oc9Kt2<)oV6C&
zul<NB3jx^_P&5l@e7=dO$kPxYZG?R<dstavzl3ubqi}F+IAklb<rS)^gt0n^HzDFE
z^`0-brKKgpSd`DiTx4OtE&2(y2}0aKQjSkS%u3H>Wn$rfpg84%Qj1HL!0q!UyvWbM
zF8fK1r?{=29_YH5C>I{IE$<H~QcDrUc5jl22=L$R#tsd=E;Z@z!@&$3D$gwj%hO91
z-qVk2+q(^ek>Jo({gVVFgOqgpbj@*&dW~?1ehwBOWCDy)@qhR+12K^d=|8BVuV0!x
zv-Geg9S60wN<e-8<RAnd!!I>Gy*miNxWJ@!^tGOBx!}fnt=L}`79fVN?B-@2afv6N
z$;yJ!4KwgleDe78fU3S~=cdFMJ^YwCB%4o-2M^Hv$|+FVLP$^$<z!Fzipwf0Hpje7
zR@V&VyP5ib7scP)Ypkeol5kPH{vqZ5vG!m3*$Tqeuc@K7uzv#16wn02QG?+Mwt&>r
zzhchlEkkDn!L7dFGS?y1(t4(9pqw3>AReZqgx4<O=Hq_}h~1ctl8yKM&#mH$cUC=#
zLVE9kcnHLAum;z?rHT3v1I?$jg7d+H2lIYHxSk$Ti|GXGt{Xz)cJ`D{MSBJ(!7~K=
zDTrm?Hyaum;gvY;^D{<eZ|NhC&4yo#<-%*F9N%ELx)$n#F^Y!3Y^|fM4T?{&42?}_
z!CQxd>@9Afy^lOX{+{gd#7*j}JU{2)zJc6~sk@@PVsPcYeeYi6o$T-p(C#TG&%%i9
zy)k*Z4#47>4kmbCg8=Yr<&pt_<&i83^MXjnTT^SaZs0~PbgOM<*}ui0G_6Fou<O;T
zt7<7y{UvV~xP2s8W`AvX0)A3LT?-rvJ)|16xT!^AH22MwoNC+;kKUR#-o5S4n;=2;
z7k6ViY<jjg@oAOucBEIhF0j{V2sksq$1a)x1s30Z`qJ79DZ=%H5W}5Rl$FxPukU{D
zws?i5G+YDt&KU4}z1%L*k@{b!U~)(zYp=(}951tF|0qc!;yMg0-q}+fB0-D4wYbfr
zc0vdOF(Iq!3@H8;F>SpCjrz5Re`6H|`t6L#rTO{70`;?%hv=2DFoH3$!{HanEBji*
ztzXO2MKMKF3#U6v>bUlAvWc8CQ8d(^yKgRHB0v`pWh;5kU(LcHKY%m)rr|9vYN@$`
z6uat72w}(<0qfIfQaAf4rv{fkkE!^@E8}ez)by1a#Jd90homJXRc=cC++y31RxRfE
z{=K#ql|bsI7{26Pv6_U;TI1ZF?e@TnBc9g=J_N?e2DMJ+TZz@jzmlrWdxdU=w>-=S
z2S@!%X8`F`l^KiPLR8|XBmSocF0XSGuY6(u0#W*T*=!Ta?=`DDz~k_;de9L0;Aw_%
zcf>y?%Trvf%Wwh(?3usYPN&Xr(NAk~j;AVZUu$i<t`{yB)F$T);ciS#U#J-YJ#YDg
zC|rWUjQ!;s?od&xe1p<Y3iK#Vb0<#lP3-iq#~nLez36@b!g=2X840Sp(n?=Y5NPt7
z-mtXjZWfvtwRaG4nq7;#qp6zRm&6lV%d}Q%^z-LKo2lyXv*XR$50%?A$ltcbAvCmB
z_|j$x(Hea5{+(hYf>kbBiZ48l!xU?e$Y-|<&8!67j<^w(=MiobRYyQ6vXM@zsQNK?
z7lvILndjy**}q)(p8F-WT-e|1ih34KDny2O#u2d^2yO#h=k!1zBR#T(j0}BZFN3e%
zD{e!?5tEMwhOjZ5?>GE0jAiJvvu@R6kUwd~*yOc7HCY&Q{M+$pn&=_pYIxe4>5+Yg
z8UuTa)a;>3<?XS5KgaKAApQWXmC9727!KcVPp#ap*!-gmZvUgqS&2<nR{q9A*Zx%w
z2aIb=ECJ=YXqrk}IPvVoq~8Uh+x?GKoi2y1%uj&0<h7Y{dA$u~Fs0RT|6Q_at#F(E
zcsZTE7O#Eab{N!(bL3hl1+uS|ZTis{kQ}s;#iPLzXQ1+&EH)x4CUUjJ@zK>#<~SGq
zwutxMSni#6B|XG{sQpOiOF|qGaq6_2_HZ0=USss)<hTJxAl<|4%`vBlvnMQqoGgNB
zYHB~WJB4^&m~Oe6_}~%Zi;riah)PPj&sC`^C!aV-gf#ljjwt?qP4QF#1coL%#mkys
zMBL0=*$yR~$!eM(;a?Md+`|G7fBG~}o@P>{MRdZ}<-N9Vlrh2fl_<7w8h9K1M$=_b
z4K_H-Lr@nyDw}`Y{<S6D0u`hv${$q)3o@L_KA-M*><M5rzyzs?MBIHK@ZH?&c*bWy
zoB{s|?bTEt4L`3i9k;H&Qp^)Z7jZSr!7@<IzSPp*n5u5|*-iuIB2|s&1p)qAZ9=W;
z3#n=YgFEco%f|FQ^iT&$uP*u0y(cd-$6~(b1vn{RuGLX&+F03Cr8B=GzV0079kSS8
zQdoB4Nsr0m5AKPHj^rI4RaaFLH^6~_=voo0t&LSkD4<NAmsU)~K%m>8br5YP@y5AV
z)A+<hqsRWza>~E<0kQ1b&B<eG{;jibKbrno-ra$klTkOyA9qpbN55ud7<Y&H3>5DQ
zv386`M%?w=J@8eQ-~GIEh%4xRz<s(|XEO_7CXJ=y5Gw+P<Mw%xJu%S~Q4w)5{jdZf
zYJvB1)|U6~-{<ZhTAX^{W5z(23APW+J9iFCH-<uWc&<krkd8pN7lb+c{PLnMx?A`T
zMhY1AjpouX&JGtZ=#;X#AT$pqFi;2eD1>?h1Ff84JO?+&izouGIFgLf@vyM4vJe)4
ze(KD$C;mX_TZaj!rzUX00Jgyp4KL1z+jrFW?|?AdrHrTco&X2mh{vWG=z*A*!{5tD
z4iJZ+{En%s<|QE^HHKtdg)GQs;eP~TnUMz|s}KH|R^$g;hE(fAz=%33bH3-$30)I_
zWsT7f6MDizf*#(zO}J#X2qk(_xa_d+OO;6OEvxOrjv#CDsV+Z1xsZ2M0W7ZE=ChcV
zk2<q)?RQU@5<kW3KUd4kEi^O9nHioWONn54^yrnDRbs5cSpQf~R^D5qb(MWgm?gPK
zMj#jzNv$Z{l5kFZSLN>atwD=#xT1m=)kn|QOqbl*)5=O(!Weo1QnUUearDc{xgM5O
zG}R+xK8Ytiq-G0K^Fol%twF)cFv0_KyKO$6?WIl?+m*L=@N46Dj0W{9Y9}r{42Rt-
z14nIXRfmrdGh6Zrpc(#3+gsSAeI}e2^^_|6ce<FFuD^tUQmTiLl@<|kw{pw8xf~Bt
zB>`v)3i<&4qg_n@3O}YVZyM~pbiA5HoJgrduntu6P`-|o#SkiDyU?NvO+VA-3WX$$
znSp_{94dP|v!I%W;^L3-22?aG`PtvF5ujcnv#m~hs(*}#{4*HEu@79z%emb*<;q=L
zAORTkNfLVHXLa4Tql!&`ey6t9Wc8vXW@V)6pXs_k-gp-+P|A{6>!x5scs2289CK_u
ze8A_bFD#xvVtZ5oeGzIAG7d?1g5|8%*I;I@YA<=251&I`o)w6t^1pVt^kfWZs||61
z2<GNZ#v1-VYgSB?@$-0}HEZpXxSZC49`2tE#3(5mbADDI+I^~P&!s-S2|6taXI&qO
zfa0__HSu!bK7w$Upik*=cbkviVTPdQjxIVsHu4}C|G&PIv^q7v{|9RKpvMHZjqiM9
z5K6|~m9{(nYxqk)!N=T^JUMakL`gdSka#TQRZ+@5-gZWMIwvjjt?+hv*$<>EtFQOj
zU~7QPqLMExK0--b?V^%CJ`%pK24qN|Nv2zlroM+QzBhA8{zLF?migT4r^9tn{$X_g
zBOL=nMs`MM`@HI@2cH(Pa}g|-ni?<mh5i3bOaKWi;^XM^5gv`f634BavqyZ@9^-tF
zIUQ2BP&EOP$CoTI#gWZ2v<j|l(bCqy+Q=Bonz<^0cQLFY>q`!!T1d^QCmt=T3(t!}
zBX}T8P0T9GE7#Z8Qa`9NE(*Fuu&^*aQ|tz4U}LJAj`~qx%k06H>03J|`p?AV2_V*E
zV0~z6IZ%6MYD9)}E6{_NcX;{ek(D=RTz*<se3Jad0V6r9;)2rHpWhHoH##_&m!EIZ
z=HofI3abdzT}TC5TU+;abS?QHi)#D1Cus1r=<5DT4=1*F;^F0$YHu4_#bQ--DNG_S
zHjI<^@L0s_BlT`zO7&gNQ?c@-4`>#}sp}95{YICB`XT*B<&#pUTs}Szy8E4j=}$Zj
zeHhHybfuJH5r4-}zrzTTehBw><~N>T3;%%2A;fwc9o-<Ja{*Xit>g8Hp|^&H-n8V-
zr@U^s3v#E-0pk{rJuyu(27V7#`5fvhckg+?GiYVKuAEr?B{9Ciq&e|KiL0@(w-MVU
zW7o^mKu!Go%=%1Sj5i)`Xtx^Mf0uuJ?}D$pdU^G5p&2`hb*1kwbShH6+Lpnltby_G
zy?&okO5<RA-DsTgt%gQ3tPn~$b*KG~U9PIj>PGSz!;BLyWBrKEukbevCtK?O-`|k@
z-mBRm8KK9h*lqUMfeO=&UT1z3R`*cnap!ND9$hH!c&^pLM+K3_Dbw{!;3yfIJ8wJx
zk#x?yYyCuBOl)9b*Z7hS6&t}SuC%q^uJYV3oL_mA-Dd!ohMQLL*X%&JT=)UV-`jIk
zrw?1V6!WF<doCF7r%M!BWxLN0kY?FA5nMTv+H!k8XMGH?Uj@F|U6-K$Ea+Mdm;Vk-
zd_t^u(A(JG*W?A&PDtCu{5FWV*FpvYd(aPp1FMmG^btC;F*XPBiZ@S!B6p9CEMIMH
zRe;B}AscNKCwywcQ~TW&1**{0d$%G<N$Dp;+Td$BJL14wkrN4dv&%eGRCF?Q>xrj!
zP)Zl2HE0Xrj(1M`oS1(Aes8{+L@gZD`S)p%PCd6HZ2GV@neZm+d?TQ4n@%%p4aave
zZ9cwv(=4T?_V$U%9{Y!ycuadioDh4%l?X8*)*tr%5I1b^{4ylvV?-)dRwfZqqp;tS
zHiJHZ88AuVaQf}eK7Q;=0?A4blqj3AxQYCTU?4pX>yh}^z5eF^|I^%^vLVxA`WN};
z#DHTrR`abmH#gg~xe0+pJ6=IY!w2@#xL6?~=#XF9U`tC8j_Zl{XQ;&RfJoRECD$^J
zmhAn@1CDh|t%gXzbKv;sN#K%@Iw1{Rg(N5@aO?AXBqoQsDjW1P26P&c^Fxv%ul?&;
zuG%3#DRH%+t8OX@)cR-hGRMwb-V~ILvALGSjlp){j!3R`PdamhS*4CR%2%DApFb>F
z5cla=h2>Q25jbveZe#Is4yQb|!*`os|64ivZU}DW5Q^2Yq1Zjm{^8*ewO(2%)az8F
z9xkO(Wn93vPO2SeWpCVR&dBbP!xZ8A@3f~jIWu#SD_e1)$MP^3FE-*}Z-f-jIAGUU
z`lyU$1}!CM7w;dl4L4$FlH+$W$uZg^AlJ?0sZAKtm~^S(b6joT`WrDmZ@!E1E=gw;
z@AkW<Q9tczQ7qxkZgxm9UKy$#m6?Fv20==%6d{N2*=XndX|s`mQmuGxp$qNOPD|$m
z)UVROa<OlBewRG<OCo=VYm>cs2E@5sz7a#dJ4`YCtLBYQpHqW0bwHF1eoKFjN+cn|
zH?m0W4i8<5D}n^PHBbR~%4rh@zfjW#F|67kCt`RJPk=PmhyK@TLVdB(CatcgPKp;x
zMYV&ga}9$$v($1?wDpjBND+ReM5<`!+Dprx(kt>bG3Uch&a4E;3IFl&GSP4KUbKS+
z^@5wn@umBM@%FNxTFt|+?~sVZcpSu9lYRZw9c)n@uUbFDh=>ToWA$CYF|L}Zj3q9v
zx}@<4DNcAdgpjJ5e84@G6y!WEY5MI9ubN$XCw(}%oFGG4I+W;io=xP@!=Cmhsr$G%
z3=GUdO*IBZepw<{nyC&6ohFxCyzxh2IKKGv3_F!$VB?U|S{;ltCY^MJ>hOay79v0P
z7BX~@IETre2ipmSG~(TO^WAnx`vyh$^jD5c-roiK6?TUu%22<74)*?oGUqFf0$W>U
z?{jpV@0T`@5?Zz_Q5wm6w^`1M{9%Cuw<ZO?sJ7hNk5yV5-U-*0v1GfDD+-+V`<c3y
zE${!}Qd(UNwB>$oQ%vIS-7kJpq|^RH{e>PL5d@5waK^P7L3A3s{R{|ycky;wMIJw;
zezaIRz<FT+fBL0rz}K%|e+CZ8AWdc+>{jj<mN1G%^nN+yq!2MPaj?`fRQXWyiKZxY
z>2CwMh<3={9wr@ntiUm4k#3FQQOF~qDRH8tb3Ht%XU{&Cw&QM1dOe<#P*CoEyZeOG
zbZ(5NSAO@OhzJfoIy!Rq_PYYoQGgf>t~B9{Yx9$SW;qV4C!FW-gduLmkfpOb^x!`}
zEZ6Uqwa$kc`eM|7gkNpsz(YN-U3dOge$-B;{U>VE(xw+8`b_mz=)M|q8w1<#<g_QE
zxq?$o4Y0Yj)S>JDpHJ6KyG8x!1{jQOcx#m*tvE~ZQ*^YL6x{Jowmk_6Vu3t_xft%C
zYGL-calR+p?lCdZK+(dviKfJbkI%o~0Z|?v9%or!P5-{GdUH{brAW+9&x8rbWjnjY
zCMV1PX#xI@I5moL#+6xI+6X2WO*Zv;Ul^d(n?C;1`N(pmM*1mEF(r3IEh-4%T()0z
zOuW8#Zx%#|j*TxsF)Fl=&B(}d|2{q$$1nIvU!8y12J|iEw2qO9jQ|hNg~&$0VUqDc
z0VDjw-XU(;s=eu}>Ghv+!I1F)-3ss?pd!2p7kh^Z%f9TEtOXFS&XuLKaQb9m6^l{?
zUe8KfTU#0$8e?hxFf!pNHwSecud`V^^HQgr_%$oGUU}26jY0$tw~~Bk_B+pT2OtFl
z(rovae*OKAk?j$yh~Ei-UzwSK)w-I?V2J{#A>Ojg6#nu)m(V(^qV+<^uNO-KRfwcu
zv2F7V(tEJzhsHV#2kORU>)w6E?`xlibpZ~*M1HsG;@t7xHqTCR2JSfftbt^A&`o|w
zA6h-F;de4q8c@-ch#nn%B6xHZ8Wd*wIF4TWKls<6`~J&&2DVD5LSqOcZ(TkFgR`LL
z+-LcoY!yaX4uw!`uzjBLZum5PGmMRvcSjwR^8RqN>s9TzD(=D)4YW6GnlRMSM2;2N
zr@%YnQ$D^e?4I~3hQoI7AC#9qN!vF0nyJYWSCoC`vRzR1@9V@d<;ch=EJLBht51!-
zpjI+I<hdxWwAc4d`~E#5DfI9l7q2?#9(lsw`E`Srx=b`I+*a@AbM7Va1AcrXK-~LW
zf6#&(ddO~h)xrg$-oQh%kbEISK(H9g@_KH{9e@iAjNR(eEEMF8sD+IczsJ#io+42l
zdE?N1&Sw}7i2KQ@qO8T!KaFnepW}w;cb)k8lV&f@TPMCB_mmO3yjaxYq9-LK*>OsZ
z&+7`%iG`T8Y1DRow8leREDC-T6<zVc3<lx0|InSxl~ECctEQa~jP29`|KSinfBD?;
zau0|XN#wQDF=XRAmT<0*=HK+!wYmg63<?<_JaLei;9=_sv%rvXY)rI>=bj)39~z<%
zf|aF2^8Wep;<vvM-;Xo?!+jX}VbpHQy!a(vTUcU$LAvmv2ROwlA%sS(`RdFW6z!f5
zc3+(lYW^>JJ8I&X0*pEzKL&-7a5x`)P<l^`^ZxoY{<pgSUz;>%oQ?66ecSLySdV9Z
zw|)Tp{Q5umJ#zguz<)5jqjVMN6kGU!AeQBe!2XN=i7!<*2}~a@o8Ea56^;D;87EH=
z9W77z^yD8J9UY{uf8%G@$r<DM0clafe!C@AUlpE9y8p^~1!F;~q|4^!re@*HM`bxx
zz>tLv{?W}z9P|hz!*IK?GW)qL+^l%qOEAp2e-+j(iDlE0mjH5^JtiXkoxdQQRx>Nr
zv<`Q1$@tadw_aqz18}MPvd2G3I3_I2&+}b4WMPBE*<YT-AoJ9dle<r_xUeL2<eMBB
z2U!s;zk0G1DIiU*yHAREWmT8ot*)?ex}E*+5_w;On_#wboG9IWps}D|iUKbdf*s-7
z%Tgk1kN#Tq^`MPh*-BqHq35@Xrp{mP05>thTTgv2i|Ue3?tlC_KE&XZ5&)_WedUz)
zhxDkoSWdt6#0z7XmfKrf&n!B;6csgWCmaR;jkI*@9I{+vhUO^;@~X+2^=H<W<b1!x
z)WiiJGyZ(_eM(beTgLlC!_|%@@DgQF3-25<Hflo*yl^RCTdc6tbvm2QfaDAhK4SR%
zBX?#ph(y8~?Ftzg8DAT2hk?lEpp1&hj~wz=^lPn%%~G5HeFLp{g0AM(Rznd{9jKQB
zWh7e}==be@b+coIe-L*v#CZ)-<zSp}mwXNrz&7+W1os*uNcNqw`usW?L7J|K<<{hY
zcE%e~@TjitYZ}etp#V%^w-~RO99|E#GGKPBuk#tTnVkOp8wUZ1RX%YP{)mKxxOfs$
z^hTdbn`moeV>~1(Duz{W6>=#CvMb{v(|is46HM7iEz1ZIIES?xJdNQXk-y^v2{@!~
z6jhrCz+E7Bi3yx`R#wTFUg+y@HASlP@VIzaK*M5o#!s2iz$%yHSMa(fHOtD%jaM&I
z-4|vNbnN-h7A8hOI`3d$fO&IkuzbQmNOkP2gNGEN7akIku8oe2kc;@tPG<J1(BfnA
z3DT@dHGP9Nk^zq0!^1<M!gE%Vvt!%S@PNWW%|C7XNuNLUz2JM!O@gFa=Zl5$Zt=Ho
zc`pk5Tlm}$4tgeVtFl|>cXcYZ3vzPu3JP$2&$bRum}fHhyl}~CJ@?t-6t+x_@QLu}
z>g=3uC54gLw){>?+otI-ab)AUYD-c=fiIvYC$@+`Rv{)TM#N(Y0L=8v++MTNQQJSW
ztA`5G(%KSwo7FSjhdQq5F+~O6aKk@fdS93UmU#c;OEqoJ_30LW&1M8L;CSsPoIvb&
zKQbpHlX!>8eK*L3fFK%VbmjSuQ{b%3E)0u_&J6VXYrR~aFyIXcfahl0CPeM;m2l#w
z2nGj*E-!oXk%{_TL~6j*3&)q0FUoeW14ifUJ=ObjMgaFQkJ?~*;$?j+c*sWYdE9Iq
zduS)?Qu`)e^Df1`MTZ3wPqUa0-(Nf1oX3!RK^M@mj8yB-YMa;Hzq*JGr;5t*y|<g^
zb%&(AjY{@wShuJB9uJSgO)2Def*z!N6A`l)ho3q*T7uk0z+aM{od3u#$z$d4*L*()
zVX*1YjNEg{e;a|{WR#Q7%mN=hV!eZdF+T;RWcqd9Q~muPF5z`N)S0xlTFbMxGPw0V
zx$)rYy8x{Cp-iCYJ$E5AU!D$toe=;-G8+LZ9wDd2_@43V=a-ovnXo8iW247JMn-tx
zhh2@+%HeKIWfY{ztNMy{VOly)PEM|8e%bEdY{E&5wRAO-W$guH6J~6v{&XJzhfoxr
za)O@!vNZQ=85zXuboZ0MWzcsA^EQ|P$@v`ESZPa1Ed}`bf6%6PbzR-Nq6Whr^&=_C
zmrvP*Ek|ltLF9E&z|zh9?TAwNg#RDwpJm`J5^7~6jOP(%c{Gpe^~!DgnF~GZ5_kmx
zB~t9}Ac0RbzwXC;|HsrqCFQ;j%<vC?8rDS>6<J{`yAwXS=x~$gnU$xsyX5fiq_cQS
zZbV7eq6(hul^if%VCzifdYl}&@`4l%nk3wW>Qx2;nV$&NCy$Iix%=^*&{_B%I}D=Z
z)$*fY=z%I8pf;cYSCsB9BfgxwBtFSYgo;g=gO0ndE?rGp`t%zC$;v9I8VT?Ti(g(j
zKVLxJw>(pj!w&zDw>{`|T^Hr7Upqdu9#>bF9}*H3Te<BRi!B@I=(w*@W#dgu1kq0r
z=dgG(=2gB<ctCc}^i(iema3z(i>me|gaigO+gRA>mfP)v*#jmAFv%G-c$PZ$Wa(N7
zBmEYv@`-va(^6lEh+lq5fQOD%f3@qI6h|?JurGTw0RbmxY2~)CKLsG2Jj}Yh!be9u
zTs*w%rsw_Z<28x`5j$_+-hgKW9zghQuOU+x&?yKraSnjl2!gs;>Dg4pbvIt9bimvU
zD;$Io)HJ52Mtl#otA$U-=JdEbbIEq5T!g-m>+-pcu93TBRKnRMfX4hum(=`reV4{7
z1AQ;B?nhQl%y%5G?E7`hEXoXd^xXWB<??O8)U<H2Y67<CbArFS|5(@czV|y5`7yvZ
zGtUowyo(R~J%-LPJ_LQ}=#b6$djlARu3x=;o`!C!4j9+EHfWJ;-A6I(yUd`hR)aJl
z_YFr<PDc<JAH7~?6~ClCL6gGW)(2R2rr}If*5cp$m8u3G&oi{T!km1PCuTQKdm*XE
zWwxtl)Ly(jlp;Af5+bQ@O&n@fdh+4CFRdhfsihYxEDye!?_)Lsr0eD1Vt}=1uHXe6
zeZ_WT7B9t&#CAO0W95wZqpbH5>D}0f$)cj_V<$i43)mLCv*TJHW^ygFo>^H#;BavU
zqe)kTL_1*X+fFm5t2j4q+<4GO1_g5SSx=$F02(5QNmle4T~b<RZ6@jp?Ng>fATpkJ
zcnqGLcpd_e1MiRPm0PGyF^6nG1OZ?xpETOu%HYz!iqA=0odR+pl9l!88c!u@?I3|h
zJbe5_PKS*TuW<f5tpSw%jTrSdjd0F4MHJ*faeBnG`o9{IfM)k4($uuVE+O)S9tsO%
zA@jB<NEC)+G9!JyLCbc`rUH6WzwTx{=moT!Iyvf-w#v)I#81!@NRbkutE)y&&E~qb
zKuz=JyT@5F9GuR>Irk*sM7UY+QwbhcGMxAHL9m>vGe((_^i2MeG*C8&$V4(627Y?s
z*mwgqZRoHL@#cq@VQZ=`CokO_zyReUbINoh*s4fKQ=+0iWo6}g!yj?uO=?cgU9$Ux
zN0%*So$~?W;&(8EZ%yEgmAt}+@(e~we%I5J;8+;q4}I=+%)Yj<%@=aDUtd}IB4|x5
z5_|o<gp_?9qA`coC;h<DVU=6{pFVN&Ce#nu##Ustb(!7?mY0Jp+t$`rP<2RT55=-L
z$6p5+u;^E_LDw3$;M?_%9j({yf=CV%Gx^bL^Wt%?Wj0pkqT6s99|oU9L0qNf=k|m+
z65NW?$f&zObXvhZeR^;mn?e{|?mAKe<N46_8#`)hV)}%a_v6Qp`t=?UkjW|DFY!bi
zalarWZnEL{HLSVtT3Qg#e?n8zi17hxmTLyFwCuG1O$;(7xbzeVndc<RmT9+LzQum-
zq{Dfik(nFjy0WrDG;1N^GWc(0?43(Oh2P3x-@PyiSO|A_?>zNBXGvM^>P>6#x)RfR
zwj1$YAX4;u>1Ly0Za3^ZLax<5hhxz03g8a0g0#31RDE5>zmRVnQ+$dbu-QnQxu$1A
zHi-Q)_H2GeLG`H5Hvl3bcqKz@uzlMUyqOs-RQMAe^E((<v0{LU^>!29$B;xA*;u{Y
zmxG`d500C3<9R5Atl32!99BW~pZlc7#~0T#afDS-8prG54dh77fBZP_t!=Hr@LHE!
zJPTVkYzz#&x>y)q+bfTWko%w15yPdjXR_3&Hv*o~(IoU-F5yP3=+-NQY}~?*ii{ET
zIC6v+UBycP5p{`Ahy&4FVP?6@+<^yi+&J9$upX{>`MZb&A6jeh^6-SG@IIji9VG2g
zhv-#TPwjP;p@xqTIm2n0hrGsA-Gw3S2#{1cu70#V4Rq4g@N#Rh`tk{WNqxb^!V6rG
zd3kt@13EF|_wQ#{N#DXp<8#yi{RT|^zub85Tk3Ie(EJ@TWV|hR$=NG-f5S<8$sQ1O
zMZ94%vE+h$w;4GdKJfc8cH>3>@ISV;`M6tCP{ZQ7B@`v!6D%(aH{Eu#@pi?LYqpm9
zz-PDe_200m{H=TiJHv_Rd}T%ESJVF0gIe4=q^Vz2Sw*F(ojn!Ija`FtD{rO{D%OW^
zt(@%+E$cif>||!NvCBH#9GVz0MSCOq+L<US#o*m?wLVRx`_WGn%)k=kzUC+PGeQTR
z6`d|4`w!kg?G8Z7dT<v0&j48K!A{3b7OLdP$;rc0elK!B26v-k%lu8tZ2#x((JN5E
ztc{MI!=DRl&GdcqLOEyDgw|w!%S0EEJL|qaA0>al%46R$t)u+~6?r3j6Cr2C){G^9
zNexUxy=hA$BO}2<LEfcC=PH-dq}T{dVt|&InwsK}3-uToTNiNnQoX%#gQ9jSpM}*^
z`z}@p1?u}2TW7ry;P?nB#;>oDTNai%5DPZAidAIZu=22|>T8rF4V3%1_)#sF4m_x_
z;r3zm1lk7G|4lLs1y6(K?xjRvApp{N;gF&Fuf>xfvI=s<PduxDS%k__a<cg3NK+#l
z?(cq$uW$lXUqG1`gb<trHv!t9ZcM8~YTcd|Ffc5{CX9g?&Ea7N1)(5XvG+f#ogVVY
z6t%+RhNR}Dvrtp(8_YB*2fX$lQe6GVK?}ogpt!1r#;-02sIayMHYsfh#Xo#T{;J9F
zT1kfvH7*BlQ1cs}<A(nJ^f}=cVv+f&ENb}kttVBPPj#2Ozjb##w6Mf-%X{oFnWHIY
zFCpGy(10nuMM}Q8LMr@)gcJxGC>67{vN5qT;<ozNAAW;QnTYE?0u(mNU$415xNkj5
z&Htg@oc8+nuGx^PJ>puwEBmYu@t>|&P`MRHLh5MC4R#5uJSO00VK9URzi8&&jVAC&
z{jVhhPN@czA~xNIW<$<M6r2|yoJ9?kHD5>(BUlG6D5>QI<jDQ|pCKH5n5e+#Wlg{3
z>W|(`ZrLH2AtJMYF1jHK%9F%6b%@=o3KC0G0U}al$Gq?~zX9A05wbr^_PW4Wfji*v
ztkW`>@D~t_QuIm-X}|k^kZszGXS+$8xq=8!(}w#x1E6L0!_Bw3A62zH&>Txv{u8S}
zWyE)Mlml^#)hpF|NLjfX|0R{Re9CkdLPbM!{`a@3v#g6Et0VfZ=hE^fSlY^BGzb%t
z;*(!lzOpm3BHN`lYte8smhizdZh<q@xs0b#%bc5M_$7;mqDxP(|LC(vmegspVB_sK
zy(1k$$%R?Nf-{qm0tpS*@!N1?cWzcoUecI@obl}s8r}p&n){ZlWeZWxwQnwJ*R$Ef
z-%a6DzhbGy&wy(l4uwyj)KUB)pQZ`1mVZS}jXdrjf-tS^2A+tt?{(>WK9bMX<g|Sx
zS{5*jA%!<09^w|@QvLZg>T8rUw>V8WDAu_+JhzS=VWw5~(T92PDL>XV!Pm~T`*6eN
zRJ{X&l!kvUl%>xZAr|)b+g5D<`=$}mm$I_TG6K|(GQXBh{+PtN_0%pBq?8pg((nO8
zYYMR9rqBuHE_@{YU-!|GD}O*_WE?A%C};->@b8MEzRfMBz7l6*VuIdCaLhDNAXI(#
zP=)kpQ7;aD<wNwZgwNFX<4BYHtKXO%?t<=;|4%$h(R99k@t2f~f$#p(zCZ^-fvnv<
zlA+VX&45eo1CC&21tkTg`F3gbm(T0G577@M##H>Sk~b0mEuZd&t5I$M`Z{whk<_d_
zuu9w~O?hhiTU;G4<MQH^ABX%`?H5&LO;Xbh&f!s?{Ubds4?gE9U64aqSQO;tq5DIj
zyh(<bb&w2}+h4X|>9wfnBq&Go@kB-96gatfpw;@q>XVqvd$Aig;4>@h$f-)y1t*dl
ze?AKTZ-)*lL`5Yc_U?nCZ$<Xj+c^bAS7xpM(#}V^*`&i?15p%96{CM^{yntscw7#@
z`Xw6hMvU&RtL@BzWejD=%sp1@!=Rw9v44{E0ps02SAc(md?*qtD+3?D)4`wbM4M2N
zC#1~Vm7LO}e;DZ@Vw)3EEXX!C`=b=#+CI;1k$8`l_s#k?titWd0YgbKR9CODUCNyp
zaIq_{(@p`D`%~jIvvJa$V99M4XhTT)A_wwG4ehgY7;zCSepIAG&pvVLSB7wP;nval
zK}}V#3W#9uw}=|_HBT$a%6?Dq<X=hJj=wo){)f1J74x<V<ll>Y>T%ic^?4sg8@nCe
zAW^8@@EYbgrY<?SQ!BSTmE5$wVITLyR7OY0{L(yX-bh==mA-E6ERBU`ncFvSY=6Zm
zN3V=_Cxu%vl!5~v`7t9yKuGX_kYL@ydTo6Tfuz8bUmo3XoTzGRh|xs)%ZQrziGA{O
z?p*P?D0T2#xf<q%dh=Ue0#OkF)m*BJ`xc#+28mJK9Ughk%)~4)BL$~k318xuv3G-m
zj`+lHl*|z%*?+k{rl?T+^fu?mj{0=Fr+z1Gr>t8=Z)Z4(ux|AA1pk=vtw^i-^nE<Z
zEdrDo0rg2p;0*+FR2(O&G?X<39<t5@Q3lFfe$Ty$Mp;?`aR`ohuAtl4u=HwM1Ttc8
z_`~pcRTQQ0mTmEhqRKXqEvVCoZ*1bQ{PD8+T@_Ug&oz(z@r()ZzZVo1PSrS3y0KD^
zx=#?he*IRl#dIM~tW6}_bS1$NF(%tmt!`Sl=Nyoj9|BYoct$PUzh+8pLk<EwF6aRt
zHw<Enrl4-%tUL8Ce}K~YvIW<qLq{@<?w)ZA3o3v|#TIC9FKsnnshm7|P&Dm4RY&`n
z@m+5!-tDPsoYcsZ(qnP+dr7;i{ScVCU}!HxA6eBRU{$z<y-s<I(I+mYPfV{&gpB|`
zA*<L8p-SbH+~2)qWZBAMesj>0udl6hm6V)}Q);DPq<H-pTy;kN@|WJd>iIy;z+7t8
zYE<^rdhi`bN0stbVwOukF+xSn<~uta$wB+RyaLLn-)#ywdk-FX)KwT3(m3<r<JFW!
z^=y%8L-Ymt#%5J>2ftpra2#XW3&Y9vbd{U_%PlRjFt+}JC0_i;4=t7@8QiYIW%1GZ
zs&(EIitBeS)0bNphJoHB!LnEDg*B(A$uVg2$`X2thyR+VCK;oNi2V5QIZyshlYi5#
zG+90mmbxTrq>3k51I_NJNPoKWc7xBa^Bel*<>d*7oF_||wH_^+YR@49rgU}2es)qT
zB?fl;jXxEUJ!W8bOltI|1ws&TGd`~D8~z_7n}v@Hpm8Ez^Yzl)$iH_BO{3H>F}+ZG
zo|9Kd%;h-NF-ysg8u{rH3v3Am-)=z#iP(1PRVq3pO5xqVZ)0n#Q(;k91Chlt090dQ
z-YzSzgbav`xTd7DaENh&K|xCL-GGca=E;0w1uV0`E}3wi%*AEkb9(f>F|K=(>hIqK
zH$t`&$+ggln853AN2m3Z?KJ=VE7sXaRq*(sd-8)*8P)&))%F%#RdsC}=mG?!6$zzF
zP$}sWkVfflP(m7{JCu?Z5h>}Gl5Vi*ZUiZ5B$RH>&GVk~{e#0`INXE1SIj-voOfIS
z%m)Awst3N7$$Ww7|6xpX1}#Jgzs`MM)7_LPOiu8{;={e93&A5>6SZN0gHkoraNNcf
zP5rVs-E*6GaS2HWTh?=h{*JIHxq;z9s%N$ZMfnH<%@&bf8AkfMScr>@OIqCDLhlDr
zxxu=Mc7_g5e*SRCl9ZJ6K10>E;Z34qzTX97I6CSg_!6{#(U3HDLjA8#yO}|jPBl}n
zfIm9z2)PoHadD}n@)TYW#Jm|Q379S*X?Jy55oPZGQ%l1aG#Pn_X?gz%XQF)Ga7sMe
zk@^F=R}UR)zputbK}*x&-xJKJk2}5>#$SKk$q&C|)g`k+=KfI2(%b+cvd;jtfL428
zU@geV$aS~;d0qP2)5y@!u0N`GRzEc9Q#lB@>v?(kiqTRjA^vT-eV=Jy1P6FF&v@Sx
z5d-|Kwg$^l`Y}gw%X7uh@AFVo0V_KQUm$<e)EEcFeN3c7`Diti0D(dT7i>N|J<)uF
z*icEJ<{tPuQ<9YhoU8MJvf=0RG0-Ct;fnIR7V_O=Z&3i*m~B!pQd)sfJw}C^$z|S(
z$Fb}2^NfbDN#(m{W=jZ&c7PK$AZiCVh3#agOTIHB{$oQ!gM$z_>0U-+S5;P~@9P&A
z7k_aw|4C2{ViN0spM5K3@@sY8Tk}o(LNG3s_wqL$k(BVkFF^pTai_kc(966!)!x&g
zY^0)+kx^DqazRZWOz1}+Tv<8NQ*i5UYwH2kBj1DHRnUu+lnf*$Ch|9b3@wM>#mZ_D
zw!0^zdc3vZJ7L)3Bf6@v8f)Hs??4nPQrvrJtrik^z}tHplPKI#E$Yebca5ySF83B2
z9o;uXS$E755l_vVF7%YM{sU<S@^a@*$m_;mqvUQeZl<JZqPn`e*v%RozW4m9sPV*M
zRMvxTb@&kiDCAqu$kFdL`h-^)XaS_ppJyy!9zy2t<EL?3_)Tsu8PXUXQC(G5zBbF=
zlAd2*FRiCWiGZyzR2SekDJl8NkPYKp3iDkj!{P(NJ;uuNPl6x^)8u5^N}9(efdK{^
z<7%@Xe>JwX<qBg)MW(F>zJ4mQ=gDicF#?SYn03G}_wdj`9`DaKA6iT#6o59Bak(}&
zw&OiboUUiTpV{MplZ7bDeb&N~5{KzVr*D;Y(4|BMrh2X+FqpcADjvHf=GNp%h<B41
zF#Zu5VJ;!|76|d)c{Le<<sPu_5d=&R3}*)t<a*O!4|UGF3nlq!3Msl6j7e>jFJSWt
zDnpHFr$&vQ-R;{s{xm0Fj*he5Mia(%W~=UsO11|#g(Ln+le+jJ!-InyPBTMmigJzx
z$56meBC}$kdm{T;$cTy!5-(R!^^;ckRItQc&Pf%5=?~8u&~>1L2_eZ9TfTe#rRO0h
zhCd{oZwLnKM!aW9Uwtt#kBN(^F5ifaH#+L!^OvM^thk>g@KWbW6}&g@kJ|uUO44M#
zk_qpwv9$?c;pnrB>9UN$*APFA-)re4WwKrn`AlttgZQGng6eNC_awl8+7(j+#EEEI
z-M20u063+0$_?5X-VcHU1No7WkrB{g)=<~XF?W`JJ32IQkByD>J}Wu@7f=5xGj;h#
z&UGOZQ+U_>8&VDb6JRg+YCnte<yixYk?0-&%kK1uo{jAHmD!8jfFyi4L4tsOHp~a*
z4ke_cm${>}SK^jH<!y$Lr1->?fZWx56coe^wb}L57w!ijHB9}6zDGxiZ{N}ciHmjr
z>(g%>o0xc?B9X%3)>-sllm!pC&3-WW9OPx^Lcj*_JTWm8R4OGpmo`1HL85O{sBxl*
zfA`^Q{}L23!>{ACz&tIl92gx`Reu{D7P{h+dCbDfT5Qo%Q1l7KfB6O3$<)}`jQpRo
z$c=zV1%*|68JUnO4I?p>S0j}X{>jY-p8p>f;LBWf^+BG9FFjMw92bieSLE`-;ZQsK
zGiW!!(gi#vkgb%K#s(Y3;XyKq%3nZk%#;A%`uzOs$hD9_ScuJ6l=T`uFsj@$6+=P6
z4+#h>FLzRL)_9+aVfH#}nfTGZ2a3V>g_ge?zrnIp7(j)+#)nm_!e1r{byk&&s2BO-
z<$AmOGCVy*ujY?pD>QmvQgAIvIJ3NxRaa9Xymbp5`C5ucCW}>vW@2b4rf5f1u1EVv
z8ny^m{QXWYBP74Lc%>;Q?a?mtJr-;qQWdBwGBEIesi&@49lVtW(T~<wdvLgKZT-Q*
z+7lEMh@q{6`ehAAD=8_7tK2cSR|n9)U1Qa}Qs^E~kGXO=?h62wTynQZD9v`Wzk_L^
zHj`Ihzg$+sj=bbjQt4PM%L;+cZln9Lu7*_TB>^#vleqpbSASxwYhapdmRC|;g9ZVq
z2VCQG6%Hi*o=X3@;fobjr-Eb5=+tKCO_JpA^AFkrWM-iMdEfHRJzNU1-XqfqWiNn>
zo}2`OS1yYGISo1D{%^+x4x-|)CQ)jxuYiM#eTT#M1S&{+Q<ICPW^c`I?Tc9D1Az;A
z+;?-9*bSSPd5Os0P<bAMi8nVlFTh(Y&eZb;+-|OJ7&%6YQL8}9=X^D^9~Mh4&I%Pj
zXD3={p&x%UT~URN$mbjlB|SG+&E@AtKlQ2Y6>e+l+~NjsKvtBfL_7}OMH@Z7g32&G
zH2u7Jz*;i<(H@pn0%_0_S)BLj*5@5__K|8o+I);<^wXv~MZxdM6D%T*YaS?%e3siA
zw3iRgWngvtw(xKJwKobF&i15nEu1ZpZ=K?2Q8sQFX{p~>U2d}WuUZ0&5W8s~e!n#<
zga&+oUdUHhtPvpJDK}>ueL^^cGgFFLSlgn=@PTcJkB5cR;=OsWRATxBN@U+DL;fq3
z!vmVya{c30<u&9UQP;a5#e2gm;(3W6;Q)+QwHK_Fm8->hg()c#ONxc{UxzYl3d){8
z)eRobj6<HK=Q(ZuF4!V~GQNk7)Y56kDm<Gg{t=ZU!Bu_tD3thWOAJ+Xb#FqFf2Pia
zfb!|f*=iJl8oxCKEd`C%T)&pXm_g{PfI^O&jls6V(GS?Bu9_0{8UR(9@H<*;yjwWX
z2kWU_^7zgz13ESiKs`m0ZGAE4-MSd_;Th3@+H_v84zza4d@BB=qRB+32!cQ|U=n#e
zz9lgApYhgjl0Y|3WMt;FPSn04noML5FwTH|3235Bx`pbW^*x<Q#P_GbNa*lNVsc`M
zLDj}Y!7h1qJ+)A6137u*r?8z5cx*uLnD|`9Mf0cQIru;jpd+{F0{{WVuHRf{GwP|%
z4S*;pULenUI_B98F#s*b&#DVrc-$SXk5?P6`w~e=>so!Ai0Su6)90<t5WyxmKqImY
z>Kz*#Q&-W<ep?&`SN(lnxpKL#B5!GVDI`cnF^Pj3=`h`CbLq3_9MxKb_`xiI{N%N;
z%Mib5l75Y)7f=7Q@DjT{S(%zhYjdSE@ACs6fbn!Nyqn6c0ssdr6>JW!O`aD&K<Wt$
zFo5&IsF;R{fF#S-6z}!x*O4!6Ref=B8c->0!Sf<OVDcvG$AiNF^5t7yMRlDQi`&mP
zeNv1u`r}1&B^Mn%eYUxSK<)W==O0)&T3Y2pO%%7f3d*QxnA*a=Cm;E4_nJR>B4^aH
zbUEk>GDd(pLFOQauBvjGa@|}ZY4W<9hb$a<`KL>V(pCZRtiKK^C*4+$pjCtN%)ct<
zmLI+uC;)+erC3W);5DsM?aX|dQX2MgK=?pdz1cvpx@M(B!FQ6HM;hDu+i)_|qe3yY
zDWVS@IkY#FWOD($Y4dOd&>rD&zD_K>-*}&$$y(H6uE#PKBHiSsXXqd;aRkttjGx;z
ztUOR02bc?55o^#jbf^haKk|IQOdly|4FxwqkwE*MJ5=&ly~F0F7eRpsI?jhWP&I3_
zTpr0@RJX&q_4lZCcO}2zftz2+zTPymN5oB832A(%iRM<9gNRN5I19hrBTAH<gabqh
z5jf_1gliTx&F};!-6robswXQ>y5jZQm%o%in-;*)N_X+VS?LwWnXR^QE|}5Z733#E
z|EHn;)@Y&eT4L&Wk#4p1?EF>zyYA8*t`8EFUsT70EF`zC*j7PJ;rw3n53V$^oNT#~
zhYlMZUa5vvF8jlI@^hkW1e_evR012e@dc<zpHcHGoAT?Xbm|%^JB>b9&mbU{ruX-B
zKaZftu^DiW|L8yv;JGD8^m<WJ2tvIFp^16J#vGqc*Btb<p-6k9(JmZ%s|hjRVV5B7
zPjg!6L)+`QD!+#)-4-H5pdVy}+2XTI{<nuOy%1*)Gz5BV@9B~M=^nSsx-AWr+OcyM
z<O_f^uTH$X=cv1gi(EtfYB@Fn9F~;4OYtn7gk*(8{;}?<!;!k~fK8iV{IXrYJ0ne>
zaImly$wVi+r`-NT`TTELND*-p;1wB!qRFk0L*F%RC~`pq$bjln9akjykQ%RVX6B+d
znNOItt@Im|6Ttvf3JnFLt%k#7T@p|(WfW7lZ%@CCH>!nB?=oA>L>h!bN8x#gUMK5*
zgv}SPFdK%J0CU6-0&16^cQz<Ml}rXjOs=~tImt3<{S&knq2tTrlBOeMlV93>ZlGZx
z(aTIjh&80D3Y?2;s2R=88JVAS_2r4`w!mi?IN?vj!W|550v{U)$patp&|gbEOzfes
z@|BVTRyj!2FB|ZIG|Sr-BvevWOKn~l8v(XH*jdUlyaRki<3|=ZFKazLd7Ib%{{B$4
z0B=&rGI@y2D!X#MCTv_`8-ym2KOzMcV)O(SmIs!$)@mx6l3WwOov1#*pTUOR6)MfI
zU%$?63+|{oEd+tXqL06zm;?HZhKy~$vLAvtLz2;BnmN&w{tE4)iKFa;g9F3dws(ZT
zf$szqYY1;G@Qt*z(&8j+?d?fPNC1QKW_jL*SBN<T8#$?;bbplEX$<(N`Ipbq$A3lx
z!pQ4*y%#WTErP-eLqkJJ2_L)WU$UGgsdX&sTloU#<W)19)yK++9+1xh=7pHp*WW{&
zH=onm#Y18DS0gH-{5E#UpheAS_U*e=F~%5}dtfmNDv%Jdm1T$obV4dXN=PJ0O90V|
zTsN~jvyG*tg6z`_03vz{c0Ij(6Z1<xi*^MSf<~f)H&el&3W<yHy{lZwM9m4tt0;n7
z^x!^z2$(l&s_IHJ#)5E53@-4x(4|`SgJc9!J%ma==LaUmD>23bSpQf>bc7NRcAu-(
z(vni|Gcot}uEoYCAFfHNi|gaw4O4v5onKHqJUXVTq6x5qp03=k?oJwd76d@J*x_M8
zK|x-JoXZ9gK;ecB7z)jSf!DyyPB~v!QccYXsD6>sr|O0FvtG_?lg0F>(;f!2^?BKq
zjfv93L&{mo_aIRL%DH1X@L^S1NSKBA2<kCNxbf#4_@SvrR-9A2$A#rNI;x%^|H4wI
zHBZW70~HZWiFx}TOneyW>1~Zo|Bl{wQ*+|74<h1xqM#5*{vo{6xb>>XEmU-J0Mc@W
z`JX<#dp}2M3pG$jAsO)}FmN*#^bq$0wN;W%Oj=i0K_gL5-9RllDK!)g+~^leXbQni
zspR<5KNRCR6c3yO{8Wx|mxp$+?^xbnlzf|#%yjSZysHm<Yv2GC^&uGKVL|9b6JG<o
z%cpO$78mRMti!{LL`5Z%K=YYtl`*1=PwGE19qJzSb7nKJRRCaMb|5q=Vfq!0+ea$N
ziqyJ=0JGK=UqeGGk^Hr1=hJbhg~!Sq09OebGdv@s?%SH&vLE6!u47>(B_{n%a)JII
zSB2yIIAWk<DI2=Zm=DL-YpY92iixoaGNyi?*Vawg&HZg<X-g^OfgjuHeeicorm#>%
z4j}B{f|dCi*lYY`S}4dq>d-;-0aL83euF9S^c}4*s6qhoaki=t@ufhWI+)u4D@=N1
z%hJZ?L1F&lCYX4>9K{fJf`Bh3L!@*F7+Z~J-`<rwqN0@EV=v7r4z;_rH}c`bJEB{}
zp-~Dqh=>5tPg?pE03smm4-S#63V1Ued(^^rMQ#0fLB5==`Ld&v6s2^WpFIV-x_slU
z^DhZw)yT<`ara%IlmrQRz6UyR51@KH(ugV;@U`jZ;64bQWD%@iYI0n#0t;y0dn3|!
zR{+D9Hz~Tar_aG@w%>ZI^~+pZDL(b>+qd69TLBPQlm@hie)tKI7-$QppsIc=ub+#T
zbB>9NqfH6<X#ucz5eR@%KP-g<lBKw$`2sX)XJaw0FKnayw7}o^d10B#Vwxr3Yf&0!
zmX=Cb3M*n!=I<@E($9Z8ZhjBmm?&=!nflXD$0cOGk{aOt=%S|#jfKSef-@d6wDIYU
zngwzj?uopMXpVv5CWn9OEhX<wZr}D6XN)=6bo>rXliS|L#-=Zy_49rk=R|q_?Ae*s
z$jF$=z4!eX{y$d5e9fl7W!ehBJ#*Z<S|o%ET8rH$LlyNz<Y~1d9zNRc*}1`!lN#O4
zE#kO_doC_Ly`IGJ?-de6?01TSdV4GH;-vVhq4n0hYRx*v{r%e*+Of^9yE8I(?r}V9
z4g|vfa{@P_K?DsgKnZH3u673fcnYsyz51s|!%Tnvw=XXB1M0%1d@{H1=i**1ld3%$
zXl#zfRS2`9w{8CzLRZ9#O%h2SY4te%w)`T=ZhR<@Dx);$IOco{_Le63$k2)L+)MDA
zs&QFJ0-Q535%Gcpe}!S&><6l?L-dM4Zz^CRI1lOrtp!RcnXA^fZyN!7b?Ydj$zoav
z1Ea0Q7#KVXz1$+h=FhHvyu+GP`)clrwtYD=<2c_5>7Vg7S3Yr&efnmXS_f-kYG80|
zWKBg*pWt<|%Q`-6+}5Ou|AI4s|8yO)uI`nn=NM$T5%w>Jjs_uQS+~6Z@c({1sd}Q9
znIR}L5^+15i&1M_8+iEe2JH6-_4ZQt&*6Q|i9t^0(3BMRSRTJ3Zk^G(Q;4XjsfGBm
zf8-1Aw`muRgz5s+7oomewRY*|^Tt<iZ|m6z7ISNJ1Ew#9-7&NYR5%<UKX}N;b$<LB
z9&_--E6FUX_3iCSa~%ROF%VZt*%Nwu#rN|-l+yXcdu8Xbm>9GJYtBFJe$Y~<67ki1
z`t<iH{iI>DFB3hN7qI4=E+c(Y`JKx0Yn_*d(CY<tMNf<e^Eq9Hv`f6?q|8P$Ow7S=
z?N^QMnZ*iZ%D2qfKuF7Gx^`IxC-~Jryl1>grSuH<YC+~7EHEm-qu}R8DhcL-))n@j
ze60A}Si12uZ&7{<yUkB#;5ZC8FfUya+=$uG!0h+rHPpzL_fJeVZ7$NzC6ymPmX&=-
z%aK!7U<pGk%(~pZxI`@3-ZnL<eW^uE6(xB5X>NffCyx@Jyz@*q_7Y!H##YaOtLb<+
zGpuVpq4vj`8l&%}4tV2S0pergrU%79QUg*T-#yv(Ewm!wT>~K+FLKg5Es)*tTI0dq
zPP5bIZlR7PE31Nn0xrg`(O4HdXvm?Iz6O-?$9PsrJ)l+AY)o{&Z~~NYKzX#=_X~Qa
zPKF^!^U_G_wuM;e@78%P5sKXXy6{h2{54=1V^qq{7}PFHEo^PWqax$u<Nx^9KzGtm
z$)^qvY^?zSXF?ZK>k<%dqlLQt;lo&NF5^kd2Fplm+6%u+7D)9W0vBrm{gukf_f<JT
zfO!!yK?4PA9*DID?)+i}q&M6|Ip<K?)~}&SGD`B&PnmlGx>RTjLSJpTy#Xu%7Afha
zg-t&FU!hO{-$mAn&Lh7*@4P$Oo~XDHm)MV}T6}f|$s!ME1xv2@XGgYw32=e$3@mIS
z?8I^Dh4pe&M6<yzQb~Tx@{S+IT2<m#sSF}ioi|69STAw1a`Ki=66TX6Lag!6ZDQWM
zxq-a#KD^cpnAgD_9_V`%#fJvgY6I@s+t`sSlGkZ6JI${Nuwj5yGkP$+QGXJWY9JxO
z%60{dG5L4)lL#9fkPU<;zX@UOCTf0BZDDKeMNDLDX8F%j#l&Pdxn?^Uw|dEl(1nLM
z+Q3d@ln2(~(~S5tTXn5*V4P&QO~Q>%pGJ=XqR^n~y>&W@ze{1uU6;X@omtp~;MPzP
z4vs+*U7YY<eD8B`5ORtA^S~G1=n66*o|kI-<Z4)Yq<7g@ZC`l)=}MGn?}I|`mZi=^
zPh3>07LS1L)tgaQ{Lr>Ej8MtRm91*mg=&BfKr;w!fzQ9wv>E3z9YB_}Y*duV14Bby
z{Y(J6>48+FvtSK8Z{tIUiwZD2>yekxJLk)ougFs>DYUQkxxD$3|BBtBxG^guZPen^
z(%49-^o9J+DyT>($*G|c`@D&QjVdK2R)&EPme}RKR6w)!M4xf%h6CDkn?B3TG+VP&
z+vW-^m*9P|g9j=o!a}@RK6D~K^jmz4wDg>IJ>yF?YfDOZQ5p==>{?Pz=8T7`HtO}U
zP(sleEBLhQmERB{fBO!EZ@U<0ZY9L%Mui)eJg{JSf9sus5fK`4I82=q*1+Fwv5w^U
z%Ap!OO_JmJF>10BjQSX}V!@(k&HBDul3-FL19y#fadI{ve#)x%w>Z550?j0L&0A^?
zzD>>LRrRH>J?7Nyk=od058eyTN!#?WeU8&}YOMQoEV#(GZ{Hfpq2dh9OVXQnrp5B)
zp1uL$XJ05BxO=+|G6YC`Tg45Y{&InNV1wxog)D$6gHPpS18_BQ|FiYoZCeyHc0h)j
z{biEnV0*;jUqInPH5~*ovSj$#3|1vUdJ0fWXB(~7fV7q;DXb~@w;m@~%Ntd+@e4Ru
z=1Yr?6BHU6ORc-+SKy(?y3)&`T=7t+{AqLWJMItiSWhjWp~Uo0pASHmRu9*S!+}a)
zhiYoN^s&LcI6Xh%G0on#c^MeZW{(~|Y7+l8)1SzZQ=fZKl1Ga4BnUbzk8J&l!rgOz
zA4detM&pIpflu{^F)9dj%|kYB!m3X|CEDB3!$=qkS>JMhBJKVAC<sB*l`X<M(6?pC
zp_vExD2LtNOQhE6GRE)OLr^aW3zd+9-GXwS|5^$M!$#Bi@Cvl>P+y}0C!{A)V9D#l
zcP<?k(0FWD@_|;pht5w+wYF%o8Ou4nW>2*6R*_o!)0kT;@6?8X%2QM1CiW*3LMJH6
zq<u&&$SC23*v8LA8UT!kbN}E80~T|h{KkmYp-%{8frHz?n##OD_ctpXBcON|_x8Q6
zvby)n{sX;3D#pC*QoSZzol<?iZ5yktJyN}v7daWtY-ybN_n3tnoq`n28$HOY59b_!
z%d;f6482KBb@f!RgSg9+(rFCHp!y>H%<O!Bd;#w>8&@{s^HeUAu4h8Y{bSyn%S|1t
z69ARUQ=sZ^NcldmU+YaTa^I{+wjESeP##OaIGDzazNI#V04~VAFXFy*@QEykVGC9Y
zhl{Zj*XEdk5^u7w?amQ&SY1Eay*MyXDh2UWin*d<7LwkSoOGIg;qBkYKpa1gPCWvK
zpNCU!)~)Byys=nb{B4l{(&9jKiXHj5F|F){;*TGaN)$68cPxMad;<`bG6&mGTiLc0
zgV$(ad<3KJRAK1jnGB5`UlAJEvd3*+hDU_mv9dR}H{WG@G-TF9zT=!<QUG6mv&&(|
zW%oGbrhuij(8Wg}DF7P+JG)$9y;z<l+{6mKUuaO1iKXpq^G&*D8l1thWCRbY9v%|l
zI_hkX4NP=IpScNB<6KwvOe9m;#r}0jWoK>8$IE$-xey?vQ2Z7@Jwk8n>gX98JwD*!
zH(y}y#nGN=vzaw-$ck?KdGOaKS==q1&f|IAYFI61qG144L|1XY8i@woZy`~`F04*z
ztsm1*b}2Rl>hnQzT0PYNkN#xi=5*@j{OMw#Y4!eWf@Um7Yy15wK}zB$*s_vZKMWM>
z=Z`GB74(Y3@(J-@8W49SbS7Sp4a{WzUH->WG)@h4@nPeaT%_hrnx>LTJ#DHapYVG%
zEDwSfCV73L`yPYm``B1msE;-Ad4wRx23$RR>Wrv^J39i7A}-0UK%O(1F5Ui(UIDYx
zpk_m|fi^5$t|_hMHs?fjR@TB+Ks(^5kb%L$GS4PGVA#Suf_~#nbNB>cB_>-ErJZ!P
zKcvR+IBewW3^QqHjn~?36)FGq+njcJOk_!p{58L&?HK=oP2i?pnIe%S$L9r`KJkr`
z{t98?sR|(#v9ixN!r_E*Z>3d<w6z~sPn=x1yDUADB63kqOS|0Y*=Ri7=n<md-5^iW
z<sJL*H0PakmVm$n{Oaf?6pD6?O?GvDWfN_A$Z<cX@yymK@9gzAEVY$Kp^@9;-R+%U
zFV1qT*L5x-dj`-SJ{1-iYgdY$dwTm0j!}BuidFr@PsB~u+npth{K5%GkQ`MHc2K}>
zeDn)tZZLLY&51NOJNFcyd~s<W`m<-$a$5I^u#tzHi~rp9VNk{xzwXvzWXqeF(7$UR
z937-huEjVp9UdjGxzH{z*QUA6<!DjShDU&Ytz=p`Ypo4U>EleL3tw_CnSl38+O*4+
z^~h+G-yOL?JPgcUtn*`m!lJZvBOt*<$8FRL)RE86ccX+3zePmgQc`a4+WkC!l6Kp9
zU0+hYGB)7+<DJ+Mh1TL?mX8!0M`xJ;HY+c$ncjnB4o*1g=Fr#Ao*=S9K+Y@|qwRb&
zfmJ@*H#P~Q#lCH3tQscEa5AzDd>-8z8?nA5bzN|>&FwjvE%w{0vYz^MUTV8FLy%VS
z?EY~x@=T?@r}Jw`Q4tHreJSOgi{yS77{Jiw7G_Q7tu%DTEHEij;|ZE|ESxy+V)&ci
z(NTMGv2CcGZGLJbJay^0?Rzw$g994^;%}t>G2uPo%MH_xR%Sc{gRn4kp#utr<E=gO
z#8~s5=rG?$?vK+zw$-qXMDUNFT(;_cq_Kw`K%YRR8koks$+rIy2m6@(Vg^ud2C?<8
zeBCFamSe$=Wy?T_$9m|cRF2o(lRNwqQx{d<8G^RaH%`d*_MG`jbhbo2#=6f?KbeT-
zYc}|-P$88zgAnYrtUUfwEOaR7<&y6i&qYgmmvGJg+)F9(AqgUB+&TIuBffNIk6ie5
zrzz^=AbD5!6<)>w4wtdlA<?NnelXRJI4LSBVESW}>|@|<belee!K5V&eS#DICD#1?
z)1`@+?-+ZV=m`5Rj<NB=WgYogcV4&5g8kB%_woqVgE*_trACT&GgZkoUdPUBgVRED
zN1ff{;_cDrO^y0TG>_i(zHr-`n9s_AmHElo{?n`>ckPB@%e~Q?$;g~+&2&i{n{TAF
z2y^GJ1NZ(gR~(n0^|=%#$QxGgb-SO--m^5qc*6V5xs!~@jg-smisF?uLzKyl8psD3
z@wBS-Mr!Po%u|OK&JyUc#I!i=YRMW9|NHIG3SX4MciwWUonjl(4WViJ7{!#@7`J;N
zaIfBNZEw~HkojtW_Zbj7@0C}mclqbJ$3HgK_Gvv}Zga_laS_O0Z5<A3J@N{_lK9q!
z3%v|vFp)Q5z99UwrQ0&oQ?Ovktt-4zZ^9@~fGg5qaq$QLK6Z}dep+}P#>vTvr;sc4
zbzG4@OFs=6KH#oSsbwhRl8+cYh%;<(V7|-KlwUjXw!ORCONWi<;`au(sj_I)2hE1+
z;<S~fhun*({!J;pBcZ-pB|)5t5tr4k$oSewk#eVH5}a#s<8ll`W%MCy>+5@~%PEO*
zlq5IzX&!~y7;^#|N~_~=^B0s*8T&t8t~%J*c#Ag~t!p}PH$Ivspv3dMK~kVvpgJ^!
z{+duOTrj)oiSl2aAn|$81Ggdm9Q}LsFOMgOe=G)<uu%&C(s(eM)n29;b+CPHHYYYu
zo)KuPZUrl6eYsh$z9J9J+ZGuk$?`ovVIQAa^Vvrql3S`4v=1<qea>GHZ5uE4P+O^w
zIzC2!_0|0Qx2%r0Lo6f}pqyzE;u?T?;FEf2JolNSz_R}bVW=!avzI6ZqRD#tedNcL
z()aM#^ypdb#z*}feJ-D{mfYN2&vT2}4CJBH44um`)9bI#Zj+No9Z-5g=hITEbwcHP
z7Iiyt!D5PI+oMtZ-;eAKm6t1>jM2p$9i*H()_s(xS8hq=ingX93OhO2Dae0|N6q<|
zE#9cbLkL(GqAmX1>URY5MG=;BNCq3@w^?_^+(&|Yg+BffEYaQ(1;Gc8VR*o5HYc~?
zyB`u=eoM>PpW4De7=C)V#^F;gipVlQV1tbM-_AbPf^-a@Z1y|<bD;$JTxACA=!mFD
zugQVYrM)^F<51II7Dj{-PjducE*_nKmhH#ilpQrI(-z_@Mfc}e$OEe#VFlT5p#dty
z4`p@}dwS&gN{#NosvR|r`M|V1iS&(?K=eTiTLPDXoLt4S6m8(A7dED~cB85-Pq-xM
zI7Z88)a5#f!~C{_%XErzBgw(K8qKHa@0$?1vf6dm=0&Di_;yXpRM6rJ$0eR@i|-r_
zb`{_8cy4WS{&(<Jfg~!TKvq}&bAx$Yd^<#{{_OY0MeW0OSp*?lcp@pLZe7@CNVOR7
z`}Zv*{Z;*At@wnAqrA{3qn(5N0uv}mq;y49Szl-G!frT{?iO8w(ynCK;xXkHu^sUD
zUZVU=PQ{8)v789Qy<=zP9m{=VFovRYaIo{BLgevM<njhlXjU{z#B*@_!`Ds!6xiMW
zoO{00g++!a*ZAp7`Fbo}!$bCpr%aso`ZVVMNI5#}bu%F}8nQ;)QXu&f-lsqH9NyDw
zZarG=vAWrC9unbLSz|piP!%6t6S*(r$gQiJbn0p&c*}8ZxmXA*w3C*!G+4iJUC8F`
zZ2mk*=OKT8w{}}rTp~!FK0Z?Bpfq|)M*dH>@P?P3HDHbx^})x3H91Yh&9XDi+LV}c
zXh>3%77ejxmUH47h8=IqCwRok)zNye(wxtJ2o)VbJe{-SZIk8Mm3t?>Q!0Z>9p^^p
ze7v2s)i#3>449iXMbyGkak4oYq!#w(*5=kJF^MT3QpGTYc6Yp|2k$d(f!mFs)5ZF}
z9$cE4hydB*x^m6nQR$)G%`nNee$Fdm8a)FJI8sVr@9o%JdYE61kJ;t*8{mpK8M4GY
z>h3rhAAI%@C0(JA<Z282*lVoOVX7Lsrz8DAwSb$4hl7U$0|8e5x}UxnZCWG0?{DX&
z8FT7yrVnAi?kpG_8p<BD071L`JLc!JVd$^X<qQw6*$fu_1AXe-xWTEZc2jV^K11z#
z$KoP!#@KEoJqhi0%L8_!$_@6or##+&5!=2N&j1o)oDaqNm`JmPpITnd?^MQbEnAlR
z07ihc71#bjhTd-OVo-6Ui@Ry1Bpo{9=wLw5q!2uE{{b5#v#?O5$-@L>pkBjC^`DJx
zG00en=(-Fl;X84sMJwhJY>XG0h39r_7Arkf%n_v6BR+;i$++V&1?}w&`?kI@@RA0f
zCibsC_ilKWGI2QMn3-{t6>q1<Aj*b@CCU$AeG>Gn)hdhEFY)IqxP~C<D2T^;F4roO
zei4)S)qRR`2G}lhTfJY+`(_EE!`_5Ny$PFf`#Ul3Gj$a!wUM|;2fQ!JpW$Y+iaX@w
z<FcKa-1OBpwVDL;xe#Qa;<SBOSdvtzv{}&}b8g_7MMqp|+HbPkb#EyY(CA5#hhv<m
zNV#-u#$dVPu3H#gvU}36V8)WTapw17nkbYDM_2|MjlWo@9{I%`7IE*UG;QqL7VnF^
zds2`qx~ynFSk&3dJRYkC=!`V<G^|A4&S-9(PpGk-6azIY{>{Qc;)6@pe0+94J?_zb
zPOc%XOIcrA6P%cO=O#GrGV-iq)=P23oI!-v@^@vOGg+~Wn>|eC)A^V%z8W)D)?Nd>
z@sx&_OYp#OI4GI?PNm}a_Fh@8%C?mwwu~aEa};(y{<N=OA{MWhN=CNv`OUq3Mhr09
z*x=ibl`h<RiDyW)<%NGOP(~!;O$3I26s4gN&Ek}fuI}xfTI6w5Xno#hURz1YE-$a9
zZc@@Zs)Brk=sry&)?*%@zHjXu4w2;b)fO|7_lVwY@%|9fV&3e%A;qxWX6oIiO@b@v
za}e<a0x5?>-=pc%dFm9`Na?&7Z?^DOhP(h%Hmf(?JPDk>8-wX{m3t9wkz{Y*8#z9m
ztnK8|c(6Ly+Ypsy>8|d)pNAGJo6Xk3&%|(d`<GX~f^|(7u^|IttiqHV^-JrSz#mFc
z^HVKHmdJWC!?{sN(@ux-`c!<56eaTVbB&Aj%zNB@sjZy#PjaQMqf(+=J%c0fBdbP1
zr#hNK*kSH32fB#d8L)AR@8T@K%~-x6KC9YI0+I=2HP3hPKcR``^U|3F@e53n-`Rm0
zMdgO2@eXV@g?)n{-uq%@5h$DQT37makuQ^zgdzvZR;H%>E<?46YD`2VJ0K2lLi1>S
zOw{f4V5h9CTwcNNgOT>C{`gJA@yRRA52MG}t@eyvEUaGW$odo0ll_A_Y7&nXLv4t@
ztKxF%NBd=Xcr71R4{pbeYp?5CRPc)@S~w0n;Ol!K`O4~_@03yxb$ow^u^^0RIAi0<
z=l8FchA}EFy??A>;P6^N-y8RTH;@rqOvCN{ySf86<QXeqgBNb_>NYKhke`Ir$Y;k{
zc2poKQqL7i0>=;&I!I{Q+A3a(2uERHkP_{LqoYD>kIXsLW#=WU7N~zxU*t%>=cU0G
z+X*}v9i2PKTSY-W@ynR*B_B`MWDYa$pM_cwT=_p$Fm-lz4!#((A9+K&wDb)vBpMwf
zP1xx|b@V*?)61nv1?0$QBK^Y0lYxPe`VnEi;-0s3Tgd9_d^3d0y?SoLT%4OWwi}y<
z=Fe@4iXpR_yqrvmaA1z_zJuFleo2;>RouJa7>@p$%0MXgpP3owo!HK^4b)v6>?i$M
zc-9}RKmUYIT<}xUf^{?{itOWv$I7}Yv?digckhmp-3hb!l`j@;H-nCTb{-OKJJFb>
zGc2va*QJ(r6C)r){nOm6KzM&{@4WND`ZF6Fe#?_a2^2Ad#z(N@+t1X~$96WjZE{;r
zUdvFH%v~*c5eEa*P*ZGFyHTol`3r~N)p2iIVl;U@A$fDA+V|k{L$H)2+&(2QL^0ZW
zIwIX2sx=QhFV4`fUOmoXK5wxpVy$p;Oy;rN)4QDcyS=_Q$e%+@j4Fl|pvZUe%g$l7
zJs`1>-M{l$1O<nC|3M`6CPPFsyXFU%1H0KbSY(0uL!~uJEi)qOGOv`2p$%&{>w|!c
zOpjrMV<%|WLk;p8Lhvn%f$+&mh5XQDP{g&EcS%$?i4{^0JN$WGG4+dthI-wgB6eJR
zfPfn6(su&FDYwn-4X$V7GZO5E%kely1ecu)j-%+Xz&xVmIBqKz0>1b&&P1xM7EMY~
z4~shIC29A?@78>F?LXSAEi9b(+=UYz!e1z`tS4Zm2>{V|tfIk9fvn7mlLwRk5W_!|
z@5SA}v258!9J7YwHpwQ5mADUhE{{tIeDd+-Fp(~^rFgEA?Qmg=Bo|CX6@Bd+_H$x%
zY=+l`clb(P{Cac$4W>Wp%`_HpcK!RD-DufuW6W*i#~lENNwuN|i5G}$Xbg;vMR_Ih
zThIKKkO_bQu-W1mEL)pjy=smP@R{Oe6`&|yZRao-XIm2KpDH)8KUEawvz^Wnp@#{Z
z;)T!1m`8;th>nOD2%#Y!%D_lckbkPv;6NVUi|QZI!@|k(#ynmmT&oC-OYXBW(-U{R
zY4^H?74m)XUFRZfy0WtJzBj1u$H*d&4MaOX4+CH6Cd3#3K6)%Chl8a4XanlD%}le5
zgapV>(lHBZ6{!2VGLbRM;XrU+u-5<nX-<6m6jnu3(<9_r|0{Y3i5!$}`5h+z`r6=T
zHOW;z`s2rS1XK~|GO1dH6}$m>?kRK2L;`W&<3E_sVuCk6nI#cSQ7j2+so4exJIIiJ
z`E}gePDOi?M!MreEgXstp_CX&VL}`QV7lg!mxqQEkpq4m`xlScmF^LK4Pn<LCugRL
zXC@4H+Y&zF=2#mTr14+p#in!dT}}6GlMY4?I`hFqMQG@`k|pJagyE{`GBZ21G}CUl
zIBCVL?}xvybS@{YGhkrMla`c<jruS;lC6SvXr!m5u71|!6J2i%mIQYhbBd-aBf06J
z^!2XORG_ZE)BIm}SZCLp{M9%4GV7VQaRno#lNUvkdS6hBxY7x&91A7qD(skWCZ6I`
z-94Y~IRE>jzzGuOU>T7|UKiz_X2d^9wiQ#68!sOnImqijwL1B+wma#z{^Z+{MHC|k
zN8RjP%tBsY!Nor@(ML8iyRH1=F27o=cR@BODvjS@cj(3vB-+jHUu_I1DPb2_&Eqx8
zghbkK@w*HUJ9Y-OG$e0No)i>>mesdn$6L?ok8=pzWqrhX;;X%wQ^+@BKjU|F8v#;x
z8vdizpBV<A;`!y@9ANoS{G+N509@a3FI(f0EH1U+Lyo)q+5-h9VtEL3Dh~c^4mTT+
zBe%KJm1e@@Aq0*};Rmq3{uU&)o%VeiZ#$Qe0m$(o=)}f8hRTv(L}_XX+s~(Y!>9WP
zKO21r0=}*e)tTT$kB&wCx%Tee+e$gm@w(;@3HgD$J!4kCdd&H;a&>Fs3U}s3`gbKX
zua6hSRHa)%zDyW_5&UP|${{%w#lt5An{nFjhwj#<d2;NSaqBcrUOJmM5crC5aSi_*
zW-Olh-0Ih5Q{L7#`Bd0vKO}-eBp)X6k$lINPe>=3HFNiIMV^(VCH$5cmaPI@74%J{
zdH1)6oi_64k;k4<+|Dld=SU;(h(<PU4gCbTnDZfYuU%Cg*AuXYed=9PMLPXP>s!n(
z?bl$l78b1ZqPXR^_05b^krk<QwB|KleY}&@X{z{_yvm_l(Z#95i7DE`%8E+JVZE!W
zMnY&?$ejp*3V|Y}=)=Vt-)O4#nUj^7O`*f{_9I?r7XdP7tB&o3lHiX=zg}=72n0f)
zo`xvb#p%HIXV}~$>avAeP%j<axD>t|GauMFzCDpK4X||c9)s%${y}xfVY7%uZ!)jl
zc<$&l=DeCYE8>5$|Iy8U3PCPRyu-uM`Ru0uQ3_!pP`UoV_R+dy0p&|XUHC{EbKKKT
zx;qT35*$u=dzg5ux7>GneiK{ED=28P8T^Hc-Se9STE*vH5ExT;X>;6(0Hq~1y*jpg
zM;nQ6<xHDb*4=O*6IAGZa*;(^s_pdK`HdS>zY{52ECag3;zTEXyCT``Rij5f@=dU3
z53Yeq`XLj)Gs9bkxa&g)U43JM&n}s`BU0x0-c{z|M0-y(S?4j2OAHTjwaF)JO?NsS
z>^<q$DmMCTx-=H)CnYVcs>Fznz>3E2a)273*y6O{y)*<P>m7ps%ZcXMW^QrvcyUo9
zXt?&xz4D2o5UPWkB99-N-E2wZ?CQ;sXbgo^0F72w2@EqvdI>Kag)nD?cuhHE`4CX!
zC2{*uA*<`FJt;h`#U8hg_I#bLqaeNCzteMbS1zN|Rxy?RW4@St>{zxc#ADp?1-N7!
z4|ovi>nb^o`uuh~Jv~JqlSudbj6R4~@5Yy0RM~NP2m>XWwq?Ii$0MKQW@zg-N80dS
z&Vg~QQWj>fyKzqiD5As*-aeh7eEY61oxIs{4k`~rT%1rcGGM~oU&UQo5--y(%E^(!
zSiKugfiI(%jD|djdm!3E)AF|eMHy^pf7aGC*&4GextWFHdyNGI1RfUtGcvcd)RB@3
zN^MTnX9Q6iZVq1Hk`|l=bJ50+*-kG1%$VMh+WqA8_!{9WS;n#F)_<&5TMqH+N>k0>
zzgR1GUMC+l;V>k?%}{<|h`kFJLrJ;oHp8yfwHt))Og62*GTlj`@j%1od7~_&o$2qc
z=MG1inT<Q72N>0YHk1YWU8JNOYx10|@5yv7XY+BZ+>?5ud!%Sc<TxwJmmKR-Q%kDi
z%y{&OpX3aU?BUkQzj*QDq_0flK^)Y17*q=i%5El-gfzAkZ4iUr*YL1{f?w`iW~i}z
zYACrLlo?TE6ZIjtDTNhTTolnN_%ihEW1|?&y-uH08S?>!h=_|R9SH`KmE~oQL|l3a
z302!J!=_J}zVC+IAC*W^I_0P)CdPqXjO7a}I-;-#<NaDoON3>CS-l?SL=%V9kzidn
z#F;)A5s5qdQ~h9FL;3~pm2pb-nf*AGuZn|;N?C4haAHAd{c{`@?GAR5_wb}AB_^s<
z{m#y><!9oH&dM<}G08DBI?7#;H(L)c#0Is8rKP0<*Xqg1k8rZ%<DeJ0jDmubXPL%v
zs+~&?olBY>?5J^J?F;Wj_0jL<{1m7IfDNw4uguKfhra*Z`c#%qL!ai*p}9NpQ{+6o
ze^YM8Lo`y6k8_cDCtRB)_f@m6Lm7g=?WJd{XDHvkv(tfo(PxN@v9RerM4ex0W>)&3
zMS<oXgvMk^WS;f);xYH=x%A}JVsZCkPtStHGd@>7Bn`i7{Ld?w^dT?ryY3}avCw*}
z7%9R}ss*4|uHpAVo{=6e2p7*ITQgQH{c3v3c<mREVgnCoRf6w{BLh6K3An*IW|S^N
z$0_ws*ITNis_UCI)H|Na$uZAn4idcRj%tG7?(Mj>7MtDh`ho#B)U7n${P-OZ5{-$G
zc0Tfa_r?-8H#bR7qNa<VZ{Psm!K3Zg2`nxV7wrDaZxndqRr>ng;+n2oplTIFH`vGe
zN8i9VLy#`;hWP5Pk8~79d)Dy-eqCJ<a>br#$%@nCg7SCTmyr?E@eeKu>0<fSW%(w}
zR>#yvrhV4~m2F+;WamD)_=VqncmGzK@oNm)zyYP}mow*KutttuR&9v6FXk3EK%F%m
zDg`$;Co7Xgy*@QJQ;oi&37pAN2{_q(p{P^liFWn6IkS$73Fqlkc?EgM^Lg0uXajGY
zY(0xwF0>x*D)tx#SwM4qjB8IXhA&A*F>H9f)_;^WdDoAb?pzNX0K{v9iKb#-j0sZ#
z{x#aYD%h%Hj{|OsU(ZOpb>MPBGGR0Jmlh3;VAEH)>Fy^53d)Ux?Xw2AXuGSsI@l%W
zdELVpa6|k$+(9&^6ZdRXS=k`;xN0t&q=n0<Gk<;YZF04rdS-5|;6igf!wTS@fJrL)
zv-lWZ!i9=o_wh)`IvykA$#1c5&lTWeY~r%Nd(wuByElGW<-9K0(#6F{7ic94IX(|F
z$(=jDiQ@kI+^S4~02wy4)(k}q0-`q7X7iX7)x<+J?Ruj&rpLO91+IohBm}M|Nt6`g
zaAnHtR~LCh+=y`5;$Bu>p^(o`?OfLB$>Cb(i?V$D6>|{w%2V9aYjDJhZqE@U2^?5)
zD*01uaUuUH>+%0iDF-v5H6k)HSBZ+$+xs#d#pf^Id?|QeT5TUhqMhz%4wo&{?@j-i
zn;G+qU>{YXxw=2vz1F!{tKw%J3kp=C%Ohpy5}5nrBR}MOwFVnx#(ji%=ly%HpUVd0
zQ4Lf%$^75By1!5;uzp)2lmAQZadIR!`#gZ3g7wq)V~2@JZQGxptzd5gAgy})>l-KU
zEVw73Lq*P$uk?R|ijt63;Ig;Be<qD5O6mR*@-Uamssj9#)7*><jv4#KuhV)-yrA61
zay_H;b?=K}{p$3nx;s~6c6G_>YVh$KXBTJC<aHGMn|^uQh)eAjoSn_ZVtn~1!{9r{
zzWMyyp6llWxO@BKCHh6hy%~Zx{`+j8)io#Xs-$XO^P{(~n4-6)K6~n%y4t?Xp0X;>
zH>tiT&GBP1cxGiRe|2vAuVZbVKfl#MzV_c7QjUCBfRFf4q$t||%+!Gb#F11hX{kOz
z-OD92UzAa?Q2^6A!EwK15DYx4D2PI*-BYM?v=(t){49)#nh!Jh>KZco0|k<{!RM>r
zZPoreCIQ6zj$7ZBybiw;KfMbFp`-jY6-6yH1dI3j;~U~5w>kD|0Ooc2JVqV!^?%nf
zu=0hX_b@~;uf}E&WLI;~GJnfy*#+u1?m&HWR7|DFrF+>oOM1k`G5zYc<5rzdz8!PP
zoGN*x9==}J{_hw!PE28DN8P!zBaKVRZ}ajJO0PnP-?x6@^n?14V7`5^iiFlBM>90%
zt}a}i8Fv4(s<UX0TQ6jDRAK%2-=rIcC%Hi;Z`fd6pkc)6fO^*pmLUkV+_r;SnHW?k
z$F;`uGYYo{UfJ~~gZ$~$=SJyZy}wN6vi-L+`hRzn_m!fTptWsssIC6;v)DNx%n!fQ
z!NqtT-IHdESEC${hmK6+qp8cW(s(BWng5QX3eg~BTd;s!Ee`(;t&+b$5_N10ZE3vH
zI6QhmbSfj?+spo8LD@}BZMfwRNAUjs-pStjSf05Xy#y7M`U9;RuzcwOOAncz8l`M8
zO`N7OCF<3~IG`kxq3pfR2E+8#(up6w71x)*8#OKi$$kMg5n5gmjx9@|g60a=TRksJ
zDAuc9UHsW>Yt4s;hxh5b*mWQ31o>qMCZtc4mUhEOn7g}!)Z6zrEk4iAJPha$9C14i
z`aYN`c<;XpPm^*eo7*^+TbBY5c}kN76%|!RMus|ObTdzoIP^Fv$<zLFbKK`*W)c$8
zVkj}V4{p3Ojg||Flv52gw#8Xl3NHNDGqzQ|#fJh5oE}`hb`ag88x`}R7thEa_n$Wo
z4ZNC5a96&r@{GG9|IkL)Zob~M@KpsA`{38U!LKFurCUh4Y5_Emdp14-q6Z?jqK%L6
zukow7Da_?(_EWVP)(gQU3S@kXWUFdEAE5}xrxkCMt8`#gV@a;VeU!NM0Jd?kGVc%6
z|Ky?a-vr5LY>T4M?4=cYC=-An7Tif?Wie%CDo}A4VFqa(+B73|mU|BfiHJ!_iE+Tq
zFFXto^MG+PQw>$8i~FShDapH}nl~>h5q?7Ba-<}*;-&G{A+l5Td3Zxy^gXnQ85!;x
z)jI~PKfX7|XB7yYFwX^&|97c4&xSL`=bfg%@+my^-S`tIk_^a!&PBre$4qgu*@Oj^
z;<~gkPg@B)y+UM+AKsS72xv8iN1~@!Vc(g9KH$d5dbVo8<@=xa3)V0&PDqugvjjid
zZ;H8_?k!A|muvC#e4_u~yfc1bHI&s*QxNK~@AegT<m(yKv*Gc_^z91IAKn&y9tC?S
z48ITiDp-79{l{hbMQ%4qpUJ>KKjEvt(5idQ8x>xCKiKf397U3exe)y~fY*I)EL*@E
z&RM+cU2+Al{rBcTk?T&}|NHaRaYw#io$<djz(4yi)K^FR@12suJ%j&SHLqT(5d8oA
cv8GEjp8M_3HQ9D=A@K7=N>Q>z{MqaO2M~=0ApigX

diff --git a/docs_old/cookbook/images/coding_structure_diagram.png b/docs_old/cookbook/images/coding_structure_diagram.png
deleted file mode 100644
index c30e04a5b07a9bd962f4e91d1cfccf22d22dc702..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 73586
zcmZsD1za3W@-~*B!8J&57Pka~69}?+Ah-otB(S&!cbDL90TSF{afjf+Z3zw`I0UzE
zc<=sq@7;a*&BA17db+#1rmO0yrwCI~l6i*l90LIX;Tc%=tttWnaue{OM0*N+Q%8v*
zgn)pkWhp7C0+y7dR&lU3v$TdFAb>tQ$F|BNzrpQYPPOpY0a=*aE27ZI!So;`1bDr)
z8GRK8)mN`q9XOIPX+fP29fK2cd(g<tdVjuf4C68lK8fegcCD2Z^Imfe$Mi26og?ZS
z#XAb2<!rc>bMqFImzEdhlKlBVYNYUC$`q0%`*nN3|8QRWObF!gGjd2@=z0=yYVo(a
zDZlfvp!waI>w8-cL$jA+-%KK#sdC<iO{P;Nw<rp4OuXyyh>H+?Avb#Mx$uYWIe1Ne
z4bd!BU`=MtqF#}+-hQCGvVwf$?9QF3hWbJ!=EHO5o?JDi&whwR@(-71g<@gP?5sO`
z1;Z;>#v@ReDfmu$YV0~dBcQ5a(mD>u`%FXmlBd34#rwtrvJF8Fgn%)2b_Yc!wLWtu
zxYJ1X`@~O`dZAXL7t_86UNk?G{C<%BfRU`VGQM3$LlZ{WK2x@eFdmf&9`Xc4tpU*j
zn<*+HumI<12#6t;2q?fABJg<*e1H{+3r0W%e&Yk5w>e1vcNMuQ2l-#;l#e&QQIiCN
zf!}H-4iJcqqlK+gJUlN3c+`yLdo3p|MFl|<TPTN-sjV@D!wqWp*aSh?O%OPRLY$1K
z-JsSsj)HC?w13?p2%JBD%}GoB*DX#~BD7kHD%6s;4iIWS4lWKZT2TyYYHDEzQ!_!;
zx6=P?4*XAq*22ljPLPw+)zy{5m6yZT!JLy@KtO<#i-(hkhaI?s-O=5~$;ge}#*ywH
z5Bb06yoETLI9S>_S=!oAKR(yU*w)!egqHTPqksMT$2cKwmjCX_#_^xi0w&1$c!iUj
zgNyTD&jy+bKYlByV(A94)_!XV1;hjNA<8enFZ|d2|8wQvJ^rnw*1uZ{yn6ZfmVdkQ
z-!0W0Ar6wZP@qdE(SNVZKO6u3;y)V-b3V@eZ<6?jnE(0~kh3U;Fz3J4Ocdi;PuUd$
zf;a;B?VI;*h<ll+F!J8Bh7jXP^j<;}K|}(iLf1&krwSD`xsu<&)b!%dMG|lqvat~*
zl)o{~iU0cWssGP!U+&&Q@=D9j%k@l7x~*t4O;$^WvixQDERQ;fi_6y1^vAIb9yNt~
zrO3A^Z^i%n5De@{pq#*Qmi+Gvz^!bE%uP*r|DQJDzJyEBO&Px<(L#Oy>mbx$LZ+og
z`mcjhN<7>~MqkNU`hU9#FF!!U!~L(rFPfHGs^JUAS3$b}eikj4vj)n;^=$ia-K>F?
zfzO@u>3+kqnTWzBFj5TJ4^2CTysTxlaaa4cK2}?4R^>?bDvJ1jf4<)Wx&rr=P?6K+
zw5B1ne$l2`uD8|@<!#Og8e8oTiLHWU1D;lo3!9pxlE(j-NUQ>N{O*a^{c+63^I938
z0Tg{D_xOh-P95}S+U`Fl5sLRc|5r4n5T8qFeV8wr69Qs_e%OD=$_d|3A^ad894}Zj
z_B#T&B(t(x(SANt3;K^4?<-Ts=PlkZy~p|Ag=h*A-zpvArbb4^eksq{>SUe#*cV|W
z0(kSM$6>2~o!kiCe@y%qeX8IWHkI()Uh`qwRfF8Ca9^|yr4>K~@p<V&>gs>*TIDU0
zgFeTa`n&Z1?f4@jusUW=f-3*BoH?N=`M-#di@z!Uhvfd!h*t!$dAhuUlz;E#AF2ZM
zrRo&nrar~^A3}W83coaT+3_;6Ol-ox&*ZOj5G|mfQY|bW<Nrgjk2)zn0t{m&QVA89
z;D6~T;b%a5X$^?#!2dgVGhi52(qECK{^vajmjyH!uOfVb{y$ddagr*$z%X8Yne_YL
z8B5UvathUyg8qm8{WbV+slYJO#&^X3cgD2%DNUhKs{bkX937;1Mn#_P;Agzc-Z#*l
zAL^R=(|>knD{9N108%=j26_@IuldCM*Z6)&p-_43aXa4ZH>8soxOKMNo(yeey8h%A
zyX?dn)^i+zY^9iBPd3;Z@>=|$uIYSAo<e4pk{j{;TS=zU%H72>zpibobBs-v^NjWu
zN&BT#3^Yn-^_G9xvxbb@IM5w<?z7#)wwv-)mjz43hx@ycwFpidv1DBvfts~&R=#4g
z;s#{&NHX3&J^NlFvvB4@E}{L}c&}*J5B!e%afC~H_QgE^oVf+1ACBl{2Zhs*H(PP?
z9L!cIg9j0uZCVT7*E?TzH!mLU4<CkYn)fvp{+j7;bU2{fTd|VZ+!xC}K5Ly7t><a#
z{`!7V{6FV190r-qs(OhX)Bi9dy+cwKIZM!XnTN>Vct6}MrzabH8o*9V;QN>x>1*sX
z6l)RiPvxk6;fMOd^T+K`h*%^`Be?ZyBVKvfc~(EUw%Ph-Jx0uiQ7J>KUZ0$W0JnKC
za-pgx-L}n=Z{Cz(m*#WfM&Gr|q$0nz)RO|K2}+cYUf3CqqXH1$QPG>dYUsj`cwnKk
zglif9sW?!X72t_{i?;2pZ|uu7>gUbUrzGu~g;xU!hVkXujP%*7^uX14xXsoX3w2Yx
zW5OmH+7P0EE}a-X=V^7blZ}D1_qisR3e+*%2xFHj#*1A%OKlvpn>OvpQ$Bx*)UTeY
zay!c-jsK_Hm(mwKEB6~rVpU;o(lJ!fEU$Cz(_dd=o*TPmS~qSc*;MWEIwiBR<~$2S
z>EXarl-YB;Tn!lsCbvlugasB{$cSBkPW3vU9|6<>0%N+w4n|gOzdx&h!XYxg@0dQ}
ze%}@7;JpWS*`0S6fX0oR=(BDDuU-=GIPw22FF8`S_jM7Opp%8R$HRF;dur?XLW#8y
zxq<s8k>g%fe_BP`_297Z>F_v@^(~-A_!pKkf{0b}+moXnVz)<<=|?Q@wRR=}p;|_Z
zUR%_xM+>Fxe9dkvP50bZ654<12)FOo4oW_`IcT=k7J&Z3LUI>Dozb-w3E0fuzMfK3
zhMxcQu&n`1fy>kUr{@ykHpr{RAOCN3Fa1!d+;a8^O*0i$9henP(zUvZtHW#G0K0o+
zM|tzXN8Z#Bp&3ITuad}No*}l?@`>x0lL{u@7tPD5ORmdH6wXuc%z}tCD#=pipVU$#
z;USso^a+KCfn9*Tq^3p{{LhX+6o_#uL3zo!v7s|k)Gv@8?l0O$0F~9!ATvT5UXGwm
zm%9G5V>0@BJ(5jI9k7!*(vO=F%Gd6HZofHbD1buMRGZ~%Ck$f9@WpCh4RN-=gjoc4
z)z7);<#?R^CL-ov?~maJ_6Zi~SE!oet3-V!L>yvEy`oe{LIytp?=Y*Li6B`0T;kz9
z_aUcaoJL$l`~AYDwS{h4x{KDb&%=$OWbpgKKqHR#dOak1c5k|Gzt)bO(Fl*$@%Mdu
z?gE1;rEulG7x5b9+2&RdyZ-fM!k!J6?zx<n&*9J35e43n_bhxi@YmhcgTD@qn|CIy
zYevk7wN3GAfc<B&F-86f9j2b`maX?%7}MVI>6lpi2#111&ByXL`5)73!!ET7HH-$?
z%30_8j#WQoEl3X;0*3w<^HJ>g;tgrmZ$>6m8ASnype_+gQK8Vg5dP({F|h}@gTcq5
z0=Cn+3lS^+6APhjhh=`(pGQ|62_Kb3ukvYx&Ks7!tq%1M2)4MgY?^l9Mg_`kV(?S?
z=j&AK0$l5nyw9&&CgfPpE~68$!`Wai)29?@*+%!4c7|p)e@Z1hWa-gDE`EB9Q3-wM
zK&bgMA)l;&S}W*YIKdIP{bcMJQ{@_kFfX`6>gIJnuz!(xJ#Acne#I=+@YzYd;h!7K
z%ajg4!PQ)Xzh-+ctVnB@#DdQox3eOs+9s8Rq#9oM+#WM#Fb27MZKs$-Q*DlkT<DP-
zo{#z5Pg1yVeBJ{5l>Mq~B7<0!nZgVQIU(14L$mEn*U=3Z#U-DsA;V8onG7}ToV`8D
zKkfS|A>e@9K@DB#D@k3uH)n>Ir^bK)Zc-n`?KJp=SgWJWH<^!~?Z?5<LcHhjGYVLD
zitW`f@0e2wUwuNdTEaCl_UjHiFJlK_l@=P)Om0-HeQp+)7m}CeyzA}TyY>K&r%aA}
zgCR{IJ5`k-$t|kM=+BVheOzxa8uE*((siHP^%o<6k(rKl>|+^;KMf)=n4=r^%h$fU
zKC20<fuT+oDy)67t9Vc6UY~gyyxg!~zc{MrzM1F}%dl^-p~GXbz5%>?)sD|6m$mll
zSVJbHTNAgaa27YCw|RGY6K=xYli~v}m*oPyIa`kWGncQfcPo^O`@EUJrG00mFLCQG
z*CLYP=NI$lFS=`GJo#e;wn)2cC>aD9g*hLNV|?a1yXNktp@3rrsxH<<X$nN3Yt=G!
z6NuZ8@g?z;mb(-4{!vMq^ZEnLD#Ua~+kE6Ozqp7U{m9`31}EXoVY`odc%UKbOwz}o
z<V=;yoWPS2{$)~w6_v<%6sgcd1pNA-^%_|>y#kX;x7gl<<PA3ex_GfdgRZoOzUNkR
z-NhoDsJ(MRcae!<b-cnHMfX}>T9t0AimLQQ5tmac-y$~5dNMTdODkIl<F!JPHvmi2
zFhMgTQ9G@nOZL=mBG854VF5%Y>glFEjWrmEEDQ;9Sonc9x_b8s7;GveNxLtHOCxUi
zc8lLd{8u<@`8SIl>$i)RhI8HrK3h+jl;6a$HSbk3Fp${BQM$}^ezK)cylCJ>_<p)`
zGQ`Obn8~6$%JYln%}RY6j|KC~Osv^%6O%E`_TlV@TXV-AJekY%ydh5}vLo2(sL(zF
ziJWEJ<roQ}?$c7V#IT?rm(G>hH{qv8`Zh_j<4|JDOc@lxSOg9>OU@F?xy>$T^jTIm
zdYH$W#()JcU3uIA4X)1bUji72Go(CDaXtC6{;I!}Z3(RwE4<H2=C*>N<U%f`WIU#_
zQc^542=G5yKKEB~dE~5<*o$_MttT%P2YUS#t&2Fd?D&4Vtq+I$3lX!5q?!W8|9$r#
zL9j$)-Hc9&sgxYcY~=n0+);Vn@*Xhs?>8lMG~V<*i`v>t!~66Gwhzu8rtzErGZkPv
zye!$T#Y?_=$&kz=OrCkP)Ze6A_5ty}m($y5y}CrCo^10<DHCu{)HA~m?|+w6*swQi
zv!4O7u-$U}z-%7t+gbJzUe23Mt*R1kgA&dMc;%*az7g5^btrcF*T8FS@{kchrm-SL
zqCbU;6S5|c6`k*hj>{ib@*CQ)H1j1M)|y8jdlY<=^Stw82$D?cLRrJlh4mkRH`f2t
zP=wFBA<pc{(a^OV2d9u#MQeSljKDG^S>G(N@bOg>%U>NDen(bAJRxd)8al+_^4U#{
zP0&-&C>$r+bwbwVfFkO0gJH^bej{sj0XI8zk<P16zJr8KX}Ft0jWE=N5gQ+eFu?*A
zMatbJ2In}HXdRa(mpWwrEfdzaMu;){bch03WNPfxcH{eJqbm4Jn0#a-`<Fm|o#!J0
zWp{sUf8@NGaTpZAO&^ThiNh8?>iD*3RkKJHtoOqk9va5!^7A?1{CE`&{ds1Kb#b%f
z!nIEj(AQHK^ZXUd^FhC$Y!$?())E8j0v(61FJ5<@jz4$@Z=cRq4fiGJUB7fY`ao`_
zfc><s;npVb%(!S7Ffj^sV@AK1&C;R6o~I+`sKXzg8{muvMZC;q7{O(38p$bj3>6Q-
zU)hvU6-H`_H8NU@LEj2{K~Eg^!YR&@o?)EXKzS;!6y^TUSFuFqCW;7Szr6Fw?E9&v
zN=V(g(8X8?)u(R?;|%o#yXLK3@)R@FqM0ib3)Y<@pIrhai1pzFl9-H}cE`Pn7hCDJ
z^@EL$HD-pqbM{%%7zfe!5un*1kIop&jawLdFh>L^$L*SH%>0*PN5m@)BcB}KPj+P6
z2G53gW!L-7)rdQ%6Z|r`QgO$cr6Ckd0-6f4fzuY2FSoUBjtM0`V^&<LDT_eZ3qb}-
zDh75Y&KGTqqvos&qwbA3=&K)cdeEybU@RfvZrE&Qb$H#yQ8y0f%}>X%gX$p;$WM?k
zjuC#i$-6P$g!<RJyvxiIqAchK!`xPE0ZPK#T^Z<DW`KDtgkUgdVQ3n9ITuVOo`0cO
zf3{N7>y^ID*d^rI8L<xd-8-KbJC?Jzs|lBcEsHtue(@l!%A+rB_uB*h`XLYgm22_~
z)~EQFCT}S=>t@2X1+uh0-n((XorfeWMiFi=f0~}NM#3`2y$RISQm8aN3Xy-oNu>qS
zWfo?l0*qZXsiW~&I&;rv6AAdCr6rKk>(4kOJejrI7tuZ}(DpJleCgihp1!2g1MGSy
z{<*(koP@Cb=t<ID%bdwb6#9&&upjKf6G9@t!`=X&FhK}<Yt};=>|em*G6H%uG&&I`
z471afqinlXdyOEqD!*8^GYI>n#b}kw%9B9Vw^X4jp*0&+es!Ft%-97)P*>=RcURml
zCRwUUy9aFIB(vXHY-j$vNK`?B-!rjQ{4DBQzK{Tpd-wE4&(0?#3Cv}79n*p{(h^yE
zx20lkojG{OYE?f{D#gIy0n+!|21arm%hf|2X!F@6b)+HNBh?&I{Fn@hK%gFq9}3}u
zylmCsAF$Y5AY~AzMXNSG>372EAvu-_UYI*WV_ii}ea(As$Gc!0!M-QUcQ6`v&L~c-
zLWpl%Bcb_nmKTnFwFqnorPF2V1H+Z2wZGxfPqJ(2oFX<E#1se4Vh&$2kn9R3rLy^}
zGbXe@+#QC=m9xsLij?|bjb8dMRvj|?GH|uT94pAGI_T?b>;KZfD$VAWYDgn-&T52h
zeMJu^wt=_3OsnzXXoxKNq%h!o1H^=mH#e)So*<6g+PwW*KXapR8^KV_*&4U_Bd>mc
za%>bMBE{!$^s&ibXIc`kuZh(h^mTucxZ{0i+{eUV(>Q)f6W}!LzxSg1P?7W;q2^<o
zn77w7oBViLaMIvo{4O4W&gDu)S9yEoKb`cNR@yXdep4pUU2HNiIRV&n$IQf*Oyd<x
zBa%yxEqlV-ab^Ye&b}i>1o4Am2Q9+}p2xi+2@&}x?l#+-PcNovWl57|((Ujf-Af2_
zTcd1{2Mfg1RTr)*Q?C3IMF+lgaQ5@qkASAzba`@#mF(Q<KkI;|8RalyD$JL%<L9U2
z3$dDb4hZD=kr6GsJo|I=7G+1l(Bma#69`Nu<wfG<d!{OSvoZ^z{o|zr{hO2P$FOav
zc#XdSEu5o$XSIWpc#kc0#OMArd(jTd#}1wUP)^nI6J0dU((*hpIjNDyPg*H{+q`0B
zl%`PeJ;;w3^<7#*_Cpp<s@oseV)a_??(RM813tM}<U=^*U+T54Wj_Y0`}GLLozwfW
zGH_05QK>AXAki_l5FV>~Eqh(OdS$}JDN<!FQyLZZQ@XT_)9njat5O4oy{C4d2b03j
zUd+QvqwZ=7n<3oj)rg8;*BlAuSm^Cth@uS@WZ6|$QuY-#`7hHPMWFq3-;-6wq3Vm4
znyoaru)Vx|#}_Ay156n7KdFo&S+R^SO2!JchfFiDpKyk<b=^g&DwMXw*Iqo;WE?WL
zj<#%>NTX48N`Irhjq8Wafa{o8_Oe1ZBoAcT@SRw|ZbkeiglEXJzej5Ago2=&E6y|U
zg`3`wcW$rj@aU$g7D@knV0i(z0~?<J29$TfuJo$qM~}fu9_myhx0OVxpt9QWQH<C<
zOs13)m9PLxM82GZq)otT)QJ(4>?J|-kw}6oqvfaz)28YVSq(;_3!jnSwk0S^!wxW8
zd&;kN9mhmmKOSq;wPqc1+2@B)x}==BggJ9i!0jQ73MnhZy4Pei9erEc${$VL6y0q}
z#k@V2ke7{GGnTVs_=`l`t&{W8QKhD7ko4uK%@RKwzaLKb%V*X)yzGXAT35vYp=)iG
zmd<xRkBcSOMHuC>9EFHDIi@1Q?o(n9W6rQdjISS7Ax|;6wE5C$s>bikGWdmM6jDD*
z4{Oau=arhE5B;u>tmCjX;;>`#&SPm985`ZNe|5@R_^ZqB5NB*w&c6YRhEWNP6O753
zqxVIixqj+Id!V&Z9a;hDh(`Rh)f3e?G}~3E-rqyG^s#iYO28gHY%ngCrEIKh7!gpp
z{XvUt;ALY$)OJ!kCUjwPgy$Y{EiXOehkvNa?~~T7x{b?CZhV<~FL~S?)S93Ayl)7@
zW)RCI@YIh-UA~5JNg)a<L1VBYQM6USC0w6AgC6l+Hw9+_tT3)`ksAa*Nb<Z6hb<$t
z&{TtvOi#~vuXz|08KfG<Gw!@MTH#92W!GgdXj?cbau?Z=(~2v>E+cY~_tVum?y?2q
z^pmQz<$W!ORsi#9S6>>~Y*BS`?EEJ87#!!^p?%1ub{|jDwXtMx0g0{RFa(e($af@A
zo2Ka*hcPTveEDb}hA?Rr8^}_twCbN45}~v67Fvm$PRy93-SDhw;tyk=LDs;6&e(^r
zY!Yg|(80TTxlR@->2158A6iVRK}~hFt>YP1mTEI53kCa3nNF(j2Ov~X2MVstIL^S!
zd<fV15T8ftwg)KN=iVL0Y%gLKB-3eY{yA@PR8tvB@`z(npv49nH}BWiB#CZ4jcR;(
z8V8ov7{qHQchnurWiHM`M@g9pDeX6kV~Gp69F!(Vo0w4Rc?vbw_?g64ox>E<7_!!-
zt4O?7Ti&bp-O%Um>=(s1%d?Hvn8rX!sS5zG;?qJgL1sG3Y9G(og&Hm#@BeVA_TUXt
z0u3)Kl-e|d+4@A8nPf_Bs}#Oh9UAY=<=LcHe|x6Z{11E~O&+0vK`*Te&NCtf>fz2s
z`*PG*xPUff)e1xT?M{PHv&q+-oICrbY&(X*GK}FzQQiHV{aIy15J}JIcqCMTflw}`
zQxI-A@`;VAVSJWR{zeu~jDaB<{*EqxJcE4POw(Mb_O6fImMuLxq!TX9+IS|6O0hg=
zzSl^l#7`(R3dM{sv43Wg<gVF9DO&KWzFS?D{1J!E2}2>)wdVc{DH`Eh9OkmM1t}m`
zKSs1#JlU0{GZA2FK8F}*x`;?}eh&F6_+TKXj!9qSz<XDQ-q;>@1(G$)nI`GJW%X{3
zJ${=YL(k)&W%w_LnO#zm*Xw=vcQ@&FpHlm&Jcc(4_Edd9Sk=9HY26M{dgk}bM_H?$
z<{9?Vl7%UN4<=pxF{GX#jsOE(^qhTK7`50^-r6@*m_j=+<@jiR_FzWunq7*2Xj+r(
z5%U4uA`lGJ6lvYBcVY(--H4SO1_0Q3i0MLmbC30s%OVs)83vh+hG>iwTO?)X^gUaJ
z7nu^5<qs=nydF2hu5cPeM1Gn;bahtYqhDW^^z5uUqSe&w9X$XJz;)7er?~?;aJxwd
z0i>VOrr1*Cyh{l`C!e|cqaZj6b-(@Lew^;nG{-j}0lv+6^hRwJ<pR~ILh;Ke^}YFc
zJ_NNx$DFjG&X=OA2vd}<@XwYAzZ2Ayn1$3yv-LD`p78;ZTR=oU<6j221P#~Gb>p*)
zbQ$R_%w*sL3oY0RJ(etfZ*Er4QBkUd1RMZFUQtxf7;7xpN}$8lE}My(X=bCzl=X<3
z;{ccolexJg(n}9}6V0G;JX94)?hT?hjA;E6w3))PwO2F7SB2G?@N99U{^Q)LHXg=t
zj^!ixkd|)K#9St8b7PE*L|+R4$df#o>kC#DufPZ26=rL?v4s+*yH|83xXbMn-0g2p
zF6sT2SxQP(HHc=ojNJyT4ZyGvld9aR_2&ZuE^WdMM6LVYg|@SYNAthxPX}7-&SrFP
z()Sz|5M)}%{Ho*C=j;IRGdXdEUwA(?ifIUls5D&Qov1JL>dTfM<<Y=ec>b3>;?Wg=
z(afk{eHFVaO8r<|G&nkn$r}D<;hW3+n}t|bzX*<|?~y?G#E5YNzzx>hy;6vdGpGg*
zcI=d|?Q(LhK~>58gR#B5kr0Mx*7LQuJSDZsJx3YZvnHJ33LZqKjBdBb{Ud_Ar6vFl
z(xfo`MWN3C3rF}!oEM0H;!Rr_>v2ZgZ#Owd@MI|PL^NCXjEP;gB&9!NwOi3=%zIrd
zK|Q@{xu4C2ZnC;8Q`mJN^2OiGRhfJ--+Z2n3t;K>X4zinVM#{o09>S5A9EA&`fWnb
zT!hfYC4d{LyXa++yFCpKgX`US^4r;({jT@9+coSPa~pc~<<9c~b{O~D&be_++R<g(
zvd4UUk;+l>0m!!LezLh%aBDlhZ2<&(H{x=YYM#ZsxWzYhLt%GjdPhGd{p*>-w7TZ}
zA$=A{?h)Ji4Jk&WprBDx-N{Nx<yW1@ZSrxaJVYyP`1$Fa;S|86YSqRsRVp`LyKbeF
zBpGT9T@k=m{4wPc^Hz8cs|XA}b<xw=yvSCFI=NYX__;h~;vT_X52~4X3>wTXk5+?8
zw9AE#^)rxJ^?nlH18P3bGUKgER9Xj*0o5t^bc)CfUjo^V<uCi%8Y*dF52m8;4ehQZ
z2l^MHCIkRvP=}S9qt;uuGS37OsU7m$NX82U&7Fv*4XZ}+NMHgykkWbyyZB%kHy`dG
zA|KXrSl(uc>NTJ;TDuK~#C@PZ1$$x)&iTIpF~j?lWV_@yWJbS92&JD&h~8hWPl#w~
zz!1sp%C?&5nWrT~uwUKM(EsU_7!e6zXdt$v-_}`B8{w9+5juVD=yiL@8_u@uN+^Bs
zhNo&7@K(m{aQasPB%BY_1-QW#i&N`0S=-_<;XI{(eoqg1o68n&Vvbq0ah8-f71Njo
zZ1auDKMj%SydygdQ;JPuZi?jTmtL3nrdR2L2T#FW!#76PTPS#uIW=l#KtL5e-1%Le
z4k!LoASw>0e#@`YmXpD=_tQU28fPVGIwKz83eK=65sj$bmYAf|(nekQ1dJ0qdbDMS
zWSmC1aa#?8nryfGY^M?dd7t?>>E+l5Gah)pSGB;8x8>>Z6UOTalwNn{90mKQ+qe41
zMdl6FYIAGA@`J%TU5;x!6cJHn$&|}AD&w1Q?sKE`v?`@rQ8~W(P?Nn&_J~&58)(2g
zh4BKc@B?HV;SZ#;m_eP`_aDjNXuHLbOup0R7ZK{`yXzhV-mbTIHe`^Lqkt5^gWbz;
zhq5{>US$eC+@6jAwg~>ESy;Kmei4FT@;#R3+yK(lrt;ehJMc_hXInd1$bnS4!|NAL
zSI_TNQTmtkGcj6MdA3&b<jmX>k$La0O{h4wnz~D)E5tZBr4p3I?rNu!<#1YQ>)Qzy
zr-rRgfarTMnd{))vU-t5`7nvBzLpFr16y(6fnzQ25Q4_Ln=dX9x{jA`LAAN^tTHiF
zA`3dp&tG{-)DydC7U_Il?A%W<*Nh|Tdbjb(<juyDZ*d=%ta^BT&1p1RO1qpmr(3=m
zQ$li{vMSYeT~J$!R~xwWC=v)I2Z`)&r5SjgO;5YzmY-;~f9Utnuk)`HX4fF{#xlz}
zGD|Eit)5BNu(>;%ZOB6LoMEX=UiKihfvpCIO}~tDTmw5UyxXX^tsR$sS`l6906>cK
z%Z-)70Y<_P#;S*o)Eedm(63paiqDF_ZQWD?h-Qj;w$}{He&TQ2<U4jXr;;n09`+En
z2-8KAp1{vbd0`J-hKkGKGcigZm21iek`28r?HD9UrH1I_DXZkCx<RXNP${Rt$x>cB
zxJZ$K1l`@33LxPMvd;7N03))pZ5iKWYq&&=;O}qMk?B=6rS%IRXHLYeo!Jo8_ee)O
z<th3L(QPJPc>Q@-B+FKqo^q%Y_*OydP$fFf^lPR|HRnfDOq?;bouC-4ruMsi+pR~C
z#%A2;eBvm>s;n-luHJY64!~|*O9f{-^Nkplrl%)lI+<{9QTx@t>cLl0BYu?CD=Bum
zB3Txdblj9-DJyY(X<ncW7~f>Yn}Da|8E;pqy5u(i_Y+)}{>lBICoZ7CosB;0^*VNz
zpO(w2a{I*fL2*9TMep*%y{jY}uN*kZ?Yp#HbWob<aUs^>i`j<!G3nY&$FqZ_=6>c9
zC<!WWCKdNXRj2N`NkVsMFK_Hl2q@(1o4t#e6Yio50o-P#s_haV8y>$?%0=&+eMPqT
zM%D<XHST^2lT%?_j5x+sSLT`Wqsx@9hFZ#Q3I@Aso)@${jjE7Hr@pN;;~pm<EG%<!
za}s0zl_`WRdQ$a*gHRf-8tR2!v#0O28e|6KGoxmf!lOX+HhE^(4nV}xkVhu?RVGJ!
zHt-Np><I0^R2cUZmnyHvnL3g>@9iv|n^cZ(U{{1by9U<p#=G-H=;EzlRntlk%?=gP
zc7|iLJ7CWT1r=12K9<@+h%6+hQP5_o{qP4#vS{4uk8}qQ322A{C8uuuj}Mv;bq-nZ
zcXP1TMn{M6hN|p>i-kU~B2;JdWb?{ByOT89);CVlHXE>WI;REE^t|ZCCs9biyA}d`
zk?E<~PxGtHosJYF_D}gOFb^rT6Y}?3N59fv(1arb{=gdtpV;0^dV{l{Oz>qrl0d6B
z862};(YjuBsrfe+H~t*n9`=2`zaFz2<o})%d3ILX!DZWG${cTT)HQRNS7Myl(9_<d
z)i^*V8ePr%*vmB9%DBEp*l^bi8|J8a1Lk27My(X(J*wQR9$HuqlEZMU+h$h1k^ua`
zseK^fQ>167-^u=TmqYa>q*_}}orQjX?@bQbY+%Ss<Tq#+c5L#dG1BUd5?HvzSbCRa
z*2R$giOg!ea1KX%p4MPat%fUaxuy_=@IGMMr;0K$6+hV)dvL$#P*MIDWdY+&4y1f&
z`kbnPB2MJB(M$hv;xM$@ID{lNCd)~B7|*nKlOWj&jY!K|lyF&u2KuM8CN`D)aJvAS
zn&Q7Pgwnt{5bb2E9s0p?POZr}IJzuq^bl6LA`ES1rg}<Mem)RS0WX#j=Dmt9IrHk?
zuDKJHk|Z}#LuX+0bCQdv$c?P&+h7SocRiZW^$x*ck2h?@E1zYkst%Si2xfEgwMgI=
z8LwJwm)rPwE0>i@vtdhX_(AEfIo`ou*(q&3lXsAU7=J1i=z*+8_t?YW#zFKyLpofH
zo6ItuXC`*OPYvD!sO<No)rFP3h-n@|O|fI#d`X|H&Ezt9+!@8++bvomyY?i(LtkP<
zKOPcYM`My+U~pqvo^=f??z|C(#_f`hBlfU2?(lzqt&|;>{NZ;HkcemBbFs|y$nx-Y
z%zb~@ayr`BD3qv__XO{|AHuQ>JR!{aIlqlw%<}>1h(+rO*L)eHvVW6H#IaEp<i?+^
zJDFf0+2i&o#Bm5Sg4|&`dA_}Y^n?nTelgy0c5K>6?>tWl6WB{>!Q0XUG~@naP*m-2
zF~TR`Re3`_q`A+6w2;NJ$cA3aQAq5eai{%3g|L7D1_Ox)l}{`}5Oh}GwJzibALCEa
zqYh+`_p5|&3M+ic2NzaDUi|!m#ro){K})}8S1nXzqu_vohto+g?tr}R-AHZ-l};VL
zET23+jQ%Y1>gM(ebmwU3Fu+itj&yJ0uc~YfW#daa0ag9UyzdPR#0@~jsYjQFJ=hbS
zll4ZA^Wj+#AsWYjcN%gG7r4$i%Ags4axiiRTW-Dvm=^zHz9gd6M(y%wbgQ^OD?=t=
zRH)S6yLABJw+T^(F;pV($h-%EOhDFcc>`6e^ha8BlRRv}XJ*>%Ku#4Z-FpPDT}Fpg
z9*rwAnfjb2x)3V*d?H60y9+b|wI~S`U>BmjdFK>PwcUM!Z>BF<6v3QYPFc(T-eE~p
z6ztOT2Z5U*`7Yyuzx^tTrc+iS<zgr^o6i{ijJ=~pJRY(B`PC+;k0rZG`5^_lfz^V-
zT5ZY)s7p1l>W#e<OnW}(HQ8q>4n^6W#7r6F7p9jCVTKov2x-3Y0tZqMb;m{g>wT^O
z*03jJQ=k8~>bz)#2IEqly-_43q?4EeZ#pq~XSmON<ym9!19>7ad6QHLWeq#Kn^yGP
z^)o^8O0vCCMA0VpH@#kD@)ZONfe|)T5<-;{LNr$798DSIvpYuJ!HEfUrJw#xqGOE$
zNn=$(k8AL9#_cMO4~Y2yRB`X=3?sY2AP%RWU;jl?-U(BhBxceya4&e{E+7AxNG)Xg
zs!S&x31S{D724&P6y=&_B+$G;#!jHg;#9Q2*?V_sC;!z)>05+VqEsl^pY7*wb0N5a
z>sNosOc-!Xu*j_KQooa7p`+?sAm6!O&;_8vIMl?*4sRm|6VA%IOK-Z~KU1>vO9~Yq
zN9W=!d1!IoXUGWLQKtWw643szKnhZEO8u64Idl}WgDXqke!gIp?s3rgwZ|X!j_q?>
z&Pwi)KqEjD`ZO#MGQ`f&JSrH%Frta}!t<z8wNK@@49%Z)xv=rxvN=~oY35-`i;x-?
z9{X42J)`X8Qc0wmTQQC^`qyE3zXF}^yz@BN7GSJ6FGS%v8G-`9Yb)nmIiktq&dk7D
zY#U@$>FX$fNe_Lys*JB!s0{E$$V=qOj9r#)s}D#C)!>m~554Tk?ucHH=?05t6FAca
zqIJeDopmMcO?mGySQBjbrDtRwJvA<j-N|d4V5m(DP24@&aSg|zLyWOUS!lKGbSU@I
zc*$XwaFaib{UB%b(}>fPo<jMzIloXInLCn&hbp9%rnri<|9(xP=vC09lCh1YFds5b
zwvok+sFlWJ)FADL(&4X?+{RNb59)>ye+TAEYQ3~Sl?sbiv{BIZSs+~VJH>6^2TxA^
z5G^fGO0gLrfmbyE3OvDB*mS&IAq-#)a`uhoxQ=@G*-Zc)hDqr#kSAPqE6}|wE)dpY
z$HB4n>dzB-PY}I>LzhS_y_UYo$*$o;Q*COXP9jP~WyD3_^Ig`#`zpy`EM|S%L0;Qz
z0$1VPHl*G2)13shBr-cRBJHQVFjPxkn@p|FxE?W70}OmT3sfX5(jfGuqY0xA(fL03
z|75%Xv77~=UlF;w{j0M)+z*XXDHe}hDh@~1zU!d85C$&BTn5~Rar(~iVEm{OHtt;B
zk(cq>N6kH|RgF~k(^5s9L8O|-h5}nbWd<~KwG}^_6~?MQ-JM>^C9`^Tp)-+|hHBki
zI!Z8_8In0I4p5I6f)o}5d4hVM%QvJaWIE7?%Ef7ESS+(d?>br2LLbr?r^l*Hv!Vy!
zj$_ao$#uqIuZ~xzyI3*DiSG}dEbCIU$un9pQEVr8!`m`1<-RTUF%esA$0;#F7jSIX
zN`akMUx*ao*WC064}mZb1TVL-hb;9`!r%J(RR9jp81}a-<N)FI<EA%wNq^fQ=+Foy
z#u?D@h-D4GGe$aWRZko5yd5|q;Jn)O>cyFTEtWOQNy71^<&sz`K}k>Jowr?uB$-Wq
z;7lXnlk!dONc3g}sLa{FI_E|B^8NG9>qzr+Wq<NNK%&>;1GcT<@%c~zJFgaAjsMNI
z6ipO(4_2v{Ru^9l>f)dxj@Tb$t)S3&r*m;gl+a%~e7AlpkP%O)U;3bxI{@Dc?8Lu6
zqVgppORy(bJ`uO;zuG<;+dtn`LkXAjy~#KvB^&jksYR+lKbQYE2b9_tDac+QTd403
zKfRFy@yi5GLVtj1M#uJXc2dva3+k;c9mf}?6den|7B^_TCR|b(U@);sNt~tB-Tmrw
z|M9o1;Ar-J$n{F_89g}VG!Cl|)U8D+$v`q|LL!s)9(A$%v@o#@VpwqPrFkw#JZ2X?
zR=R(@d&aAKqp@4b0q+kB!q$f|_un}K9i#(D!x*kuCL@PgI?sLMWc0Ww9tZ=mMeZo=
zI4H%n@KgMMu0&T}`c##u*R__Eulw6~9hY?X=KxcTU9Hq^--)B>YlcHO4W1ma{jrJG
zxhKD1f{?LR)!^CT^0E%Rk|i#mg*lpU=&Fn7;MRIuFOiVS)pPhs%B8js3yo$Hl*@4-
z3GPw~dyI^9>onfs%dPsi06Y|3P2JbHH|bkHb&GTSFmsWpvcj=Om0-3LJq^(Zu=Qw0
zVtxvtpPGDkj`WIQ8^5VniG7BP<gL$IFMe7bwA~y~1z_~3bc<vl&rh#rcZ)SFRw(0v
z1Q8t&xe`-f_`e}GyN~O_5C6$jr~H3>)*KiLM%gd_C{pRdGyAYayD|ptDQ)x2=>V1_
z4gNEXst*7<Wf60CkCF6}&DWb^8(@;Ev-q>}9-D9nbv2x9eO3q>o=^%cQMBSG#${`S
zA@sw&z-*g(t*cK_*_KTIND1I<SN2Z)3PeOHoI+1DI=?Be26NYzXGwcqbN8<bFe+!k
z=@{oVO}nd0Z#1rs*ZaTnJp7qdhSoz22MVGEwlm^!Sb=cuvo70>>2ulyJ7EHgS%ArB
z`mhq(HKf;ekktg>83aW7%V~Sr&3l0h)JRVlI&!G0cD%EY(*|ozR!>`U^i^v9rIQPx
z;;TgZT6q*)?JOExfmh0@nk25TncDniJvn9qxRALZ_PY82M{LhXi?t~JQpx9Doz0V*
zfuwh4%w<oPQBSD&E(0JrmRVKSXiX~nV9>RT7Q*Z&w_};-V(#989d2Gf)if>j<+%S7
z$vaxfQLUgJXeUZCIVc!_4)`2J)}ijsYsyvY^{>_mf+)}F9Wa6foHKU?F(3tQ20nJH
z0MJl7A8XQ|BW_zgz{sGawK!lC1>3M>V`}`ouD;TO1sl7KT!N?5m(Yr4?$pKBGHYh=
zC^Y74nV-MXy(P7FwJ9_(rm_Hgp?tbHfXW6-?}=m|j44*A8}640q`6n3fBxV*fS!d7
z=dCIV)g$wIged=Uh}WSkNQ@V#X_4Q(?hn8bookvDBr$+n4G2aeg&4a4WS7K|x(K4E
z^^6?C+zr~cI2l#OuRmbbAU9C5lD$!Lw$xN&$!5Iu7)+NARyi^Dg9VI4XZTqk$Uew_
z?aGriQ)3#wkT2z~2B6rKN07@Ubst83y0&+boC&%fUfspS$Pk0z-Fy#UVzEE&0o*&U
z3;lua-L1i^?BT=a-^C?qk6ye@pYF;d^LGes<b$6W=?d%oeMk6E@LqNRgsuO;o5sp5
zB{GDkElwfb*uU!zNOdH+umhRAf=vz1AN49Kru0#C;wOyBokEcwP`LJKQR>d1Y*PQl
zNU~LVqjY#V&fB4X`6kW;AZwoDN^?-YO)>HLXJ^vh8T9gkd*KpL6Ud!DMZ%te{1{-r
zTG%vh61&RAFuP_Le^+92of^3<G?ik+<Rvy`3u((|a6cJR*AZRuMV7<hPXe<|19aQ+
zI!`;LYUOpEJ(Sts0OM$=Y~gNnEyN+ns)1muet)r{LcjNJmiZb6y_rPB6AHnk8E8|1
z+cCfa0t=hDCP{W1c};sO*%0glLB(B#&)rbp*{Y2aEz@@Sh)CS4pEC`?CeVT`Ah|yy
zeB498XTpDhtzpB2g?|Zdgc<79YY}dxkT;7Ra$iu|3~H6bTpO7h4ev;I>~cCY)0w~O
zMe+<*E@NvJs%LXY*pw_xuy1cRkLs=UQ#eKn{K^T~I~{oTtri)1?MF9?jj8vx5Y4|U
z2mD$9uO4t$4uOJ;t)|^_))A)glcru`-Go9j8dlzeY)~N@XTWW5VDj5S7kh^lKh)=i
z&2|{;VN6*TmzYmxg%*5sYBy-K3f@^iGV;i+!c`sTj6$SJq(8Tw46@2kJ~3kWljutL
zez^{BmuY4#`)<#?zAaZWWWn3yDzFpSprH+4JagTy0-o%Nr*?17G)D1OlneDy&29L!
z{BV2`C?Ukz^oy3<gYqDv%w!tEg(8LcMmw6}#Q4R*5G?9OBJyQ<KWv;13p=)Tf#fNM
ztlidu%wW@rw?NXnV(=$og5|QG4IBM~d>)Ipr(?@|YQH_H5cPb+&bDIwSS5%SR@1o3
zR9-)!(9o~m|GOc?L1M4@bzNRl)gxCmiZ?4eV`AtT6nh7I{Y3$08FpLE_9gO2q7Gn=
zUd#pMIRR*!?q)%M^g|h!4<~2*BcsyVpN;Q}lYmv8?fl6NaS3ZZv>;KV+NKvI%^g9x
zSh;7M%UDj*nWuon!qhfY4frE7a%>YymA{o96GVx9W%J6T8-sSDrQaWk@Hkr;%VEpg
z`BuMbj*|5$J+G^#z!2H;UIgUq+#EOF9K2Df-3CsLLyqP>WBvMg9!e9ME$>Gl)T4Gh
z^92qfQ*PZdv>!q?`_Ww8b#aMl#+_|SubKXM_k1k%eOv0{yV1LpZL=)h?cZPoevcM|
z(EbW%jIyj(d^S;B2x2IHym+&BfcW?i@{6_qWr<4A6pr;H_1$CjeG&K-#iW1iZxyf%
zU~CQ6p2S~Czk6#xFEvh`#hzr!Vohfs|H$Nbo|LMRu#9GwTeWvDZwJi#V>wQ>I!9#(
z(u^G7Ax>_%hpXz-Pp^4uIfn_EeWWRsXv@=Lf%iNyF4kXTC`Yy*L>sWOW9`?clwx%X
zC@l!<`QSKWPar+8Z9pS$9fvd~V5hm&?AI)S%mCB)y2Qk_1<88zbj^A-LnF)ahh~Bi
zN)X(gVD6dvJ>7f^dCD;k<cQIX;bQ2$jyC{<$)k6&Bu?L3qqJTBX1&3p#m*FaTK|3J
z*)H43BV!EP0m8f(&V!{CM1>98y2&Levq&CgMw-T7;ZN{$5PQRL;gDkWW#N;$0g%`{
zL;0>4(EpV)ko4r|tn6ybnCMOEGv<v<yb5+hoF9%%<ahNB`lB-ihw|J(xN`n#gl)_5
zZ9RY3%d)zH*gC2dI(FDApqzsg;ZPx%@PwoR)rq&C_^PQm-2ef#bW)EXJtCbI569uK
z`mkpsSas-Cui_H=!I>0MwaQKEw}^#NB@2>pj^k3;kO?A0oRbkQ<dzw`S3cn@w|6OR
z+g&U_<Jf*j&$Hg2895pE8;wgAoPr{(ZD!#7RDNaPMS%N;SqEJx$xDMY4Fh*bx1)mU
z3&cVnOvz7ZWwe|D0J!wlSLK^sFltw6HL0JT*bx;iKl)QdG4R*L>~nX(b-R%LIB1Pz
zG%;Dbu3Lpe7n<JLlbP)uJ5+$@Kg&Fz$^9!(g7$qE9(vKFBU3(s;h}{T|4#^%$nO@7
z>x}5&52%U$R1UzJMOcYVlDWxeT@>3@**izs#(a@3$b8&HR<4!yM8Xw8plCM=-1FFw
zZAmR3kzLSoVBn=Yp-LYb9t~VVl@f957F*JRW!m=wBMJQ==A)q&>vY*2U@dy-Iy}O~
z0#MoM8{}u+W+lj=uP>og@VAUsV<1<w@Lu`%8;=v|x3@;=nSF@U7Jsbu=$Tw`{$^M8
zR(PC}Jz_kn9^gy@6=kuPyEs0V0nk^+!yJ~HR7)$U=nWn{V+fftOT}M!(QubX<&jd?
zz~$Th&COJ6JU&2{Rj0wTp$wEg{EBiA$mEgOOneFe9)p|Q0`P!mx4s^2E7<;ERxbm%
z8c&e4aQh?4)N<cLo53s3!?RK8;Uwj}2r*395+KD_^Cz|ds9)Kn7ljN-Y?y!g6mcR!
zuc_OX(a1J#52(^GGcSmv#bppj;^yI^>nSrw^N5*Qx9t&>q-6{ItijfVk0(njZp?JT
zzy;M9c97S59X1}f<*8KMFusVMjMn|uZq{Ma^40Zi@q-5vfH+hVp9A29SNl78K!A!g
zzB1Uks4eob)P$^7bo5*7jsU1U&lSS9y)LjSmVFEwWXWJ6+)_^_70Lv04CSum4Ic{f
z=B@Oer7zzyvgMs&03*~u$X_a>6W`-PPqDMb`g&lgU$SQ2=S9XZ&PH4HZs_vefNrWy
z_zME>qb}y`xR{A78z%8c0Zv(kY$5{u7^7A8xSqDfN6zo(w7D55O(7<0$Hx+&j`<s|
zOQbnAake~1BR#yp8Av)l#i#z6m`bfePtNR?%dKbz(#~$A*FR-YnAkWxO(wH$1<Fot
z70)vV$k0FS{pup(pDtKF0}_5+ZOUt}zVA_LP27>azpd6$0!rifQ=`snQxLSX2Y7}<
zjlM|ppZ=yDsFIGRu;hi@obt}<h=CQpH;u$(iE%oVo$bI5XvFbFx4!ok2n=ULFOmiD
zv^Wt;sdPv@V&+|nIu(8h<#whzO!io(ZSmhD_m+l3ynh#cWn^g^2ERzR%nfGzg$v98
z3ngX$i?+l`k&5xmOeNOOkGXk6E232mi9KVzD}&|j_t$d47nE4=sR6~+q5GRd+Z2H}
zs$xLx+8OWrXS5&CS48Ec)x_>R$}IF;-ZMAW;qM)+tJt~1?Pb#Ogz5dJIWl5|*(Qgb
z<E(-vUvfQ5&gVukON9`3Pw~^|82=8}8m2O590Cozdjx_qr4U@_RBYO=w(QjA1&?7p
zo569tUIK-(L(r>-c>2!wpy4f=fe@fF%JX&cUzI{YeGJ@xV2yO}*Z7xb2gg#MxgzK%
z*G;SN2@CG_7qHj9xPu`q87i`&1ZL*XhAi~|qWy4Dx6kNUomspdBwoIHNz?43P~CIZ
zPq-s&UKCnaqF*1>?i|=@+0=Jio>fRP7*p`^u@29#3s1{&<kiTxcV7k>Tg;uPE~Yk<
zbjvrP2g~j4&J}0m{&k;8ox$tqxGZn<I40j+IRIco3P8Hxt?{6l)HjOiKg;MOjEk~|
zCTkg^cYCH<O6<Kpy%VGfc&!#=O0<<gAC^tvAdGvV1txt@114V%0wN%C<Fr(hlriBJ
zN0v-6cnm8Decq1OT*-9p02Pb$gV>H5ajvqr8&cLAH*^{xHLqRNuC>?5H#Jr_$f9Si
zMKkOk#UIZJ>tAQV(d_yJBtOW~`Kc|vttD8#*N-}Wgv0F34jQ-XH<R@oz~ucqv-<ly
zHo5SaVM_TtQJ$H&?wsdRNM;8=^DDf?EQD;ng2ME86A0ZVTnGnW`aJ}q6XZ$3l7U*Q
z?rN{VLIwUMC&*&SlJBd;A0Wd}#9jy=IBv~8)!zTma6+#o&^ajSLm_W>Grb}##As-b
z0Olf8|5$z6YjXFgDC97ZpbfLFf#MYD^ln=iVCj^dajiA>irbf!UL6Fh-Cu2=EfShF
zf7Nr_clYLV7(M-zC8S{V>msoTxkgm^()*-zTi~<QrGZ_Az<>h>b5nGZ=wB=#N%TD+
z7(SQDAi*{)Z#~z!!B0DTvLMDc1~I6YT<v{1P{zkra(0cL;kFhA(Z~W2iEeChF0D95
z5WX9BpuO2W_i*EYm)o!05NRUWIBW+|!%8PMAc(jx?18M^b={;Vg9!stxNv?Q_rI_u
zK<hiNc?L-?T5<s#)qL7UYZ>jX;!(pib?9TMms*f|b>Q9Ywiy2d@IxLdY0otO1F!%l
zrv?cwZkQlEpfIcDkv>p*vb)cs3-p3HV&Jh4X>g<ncn)I65xcDvK@I<;pmP7=U_rS(
z^*XZXPn_}lFg4S`L@mCnIe|H8ec0AAu-%89O&rij2B)Yg3H1KiU*s=hX{OwqWB$k`
z4CK_XpiCr4S=?YK(*weV_Jl7cdZ_VF!i<DZCI<W)2tak8uIgesWE;p!YA1CjTL#Aw
z>N^DI;4-4iaV;OF^uUvVg`U)U*BoX3KodftN{qAjqT|%G`x$?&(d)#g<Fb9*tnrJ`
zhRjC5N#941U2u%({Ln@)jS=vQV*7y1_4b_%B?MdS)~<Q*gY1NB=6cr~<TfUEW!}_T
z$B><!E)Qf#?SwcE22Vh$CkdG;8tO0w|HG*-GBy{9zSBEPf_|P#;w?I6qDvc^#iJab
zz@&equoj{_9w|5H6Jwaq!@_jfYex8V=VR7@>ZRr1X3o=BsL!68ZX06@MODYrE4_Pz
zD6i%qkd@3&7*LR;GYEIz&V;Dn@0p9SE^f8|NZ9|wm$MQq$AO`$FtGU~@-4Kd@`khB
zz_~yQF2Kz1cC^xyJfXV!oWyOuSE3iwbt0moB#p#&jr9VkP?);3ERW!93CuH%gL8vJ
zm$r3G_UVJgplQNvO-ip1^5d1NIgWa~nV1_BLiqQY@D_mbZ5v*{ctt+&yA4u8d_u*9
zUu<0~lozD^1=I4)%-z%o;V+X~=&ykMvs|-^=wlhL5U9t=-*S6pKKxbNZL!Y({Vf2Y
zux3Ns9TXyB&vT;i!UO5fmA2DOAP(9$y83g$aMqa!j+l&MZUO@eyGex=Z_s!OQ_f<-
zf&o=O*UvChJ2UU|MH-UyUFOpg^h*C=(FV#L=vb9an>>xGR4g|HXV<}5!Om48W7|q>
zTS5^+H&@%)w&~wpKNr=ppc@dadVNPQPm7b!RCQwgJnj7#nGFG{<<F@<Q2qpRe+UiU
z{$3q@HcV1#pL>(kZU~Ta<}AKM4YDYvxdf7UvT0~B7vm+!Y@G-LjIB=;eZoI-^?(#s
z5k@lWiwK6kJQ9q9+LpIWz0ZR;mjoyCR-|7vxisI^dFT1&Q3TJ#%AFZK*<hd?dZz(a
z``#s}(p2SA$n0~^4N&x~od%z`N45{^<!JqW2d_bc_Uoh+QvA>;+@=ZBdF8dqKtAgg
zJ;VZ_N48R%gVR!XwV7jiU^cQTpi+|(YR1q@&|nUCmNfdnOp*WL2ZPz>mXL1QG(Kl6
zP++%&ofWvPq{Gx=yj#|wv;b}3O?e-I`y1~(*&`D%MY@Sfb8@C7Tj;TLWqxPJHtom@
zlG6oFN>wN2LT{JRZh>wz?STO36hNf}GY~lWZnZ8k=*1|s=1x+GOQ?Xt6n?{bOr)fE
zxdY|147L7qdR#{W#$XmgwqPxG<8TaO9^=LD`BBcutOlf*vR<&9T&-qAY~!ZNlS|I2
zmL6kB&nph_>;BDT!=EPna?}UmZ7L+EA$<wd@WLuAZqgpt0Jqi@2)<M}OifIBQcR|9
z2=>so5K!_u01cD9Ha3dA*O-W9-j0E{<e@rc^L8r28QrGkXr1wS4ZX=yGk{YPEWMi?
z4cdJ<OH@Lfl=*`}L^vtm0v3<YFR!bTCh`Ax`^u;+w=G;iKtj3{Bt$^ELqJIZ>5wi7
zK^mnyL_k17y1To(QyPJfE~QhYL*TBLz4h#U&b@!{7!Jla@a20~&NbJZ&wL){#>4~m
z#1+kc?F_a8Y<sNxk4I7yZRR&v*s{KYBp2o0L&j4W$`Oax=Yt>*CC3fA*1FR0WsfcN
zWP~A`q5JF2Kb{8q-qAB1kUqsU6h1;aF+%2b`^tNN!4nj7*PzaRl(v5?Va<rDBGb0U
zvjKvIC-UYoU(s`*gtBIW^BSv)qSNKwC;Bs)Qyh$ioQ(o*<W9P&xrVC}27OxU^0jiO
zw|#7zmo(AGemLHlG=36?y?0hB_|<-Bl;yh7wBK3K@rUM;$I<Pb+P%&(n!2ctPEY1H
zjKeMlH4&XyJ*jsr41FrT{#3FPrM!Wm0eHI5=<P!EUlbSbW=#tS=GATP8aOEkQ4K!8
zt|&kCp)4looc7y$*Jy|B*^B;WL#n%fbBO=@+_Kn*`r|1>tOCS_{g!iBi3@Fgr0^%{
z7j|Vsek=iOMzqG8qW3!7F>7`eob<c$mnR4N?*_s%PaUBVVnwDa0IC4RJF6i0Qy?67
zfHZ0K?TI1V9tcyErKh3zv5f5au&HXdlvoJvly>Z(x1`ki?@G4A0fRlK7a>{4Udxrk
zW?1~%Kv$=nF=RQ}P~m}{b5>HBdB3#%sZ0i0@gjskBeSm{m3|z}gKze-_Sy;4Z!GD|
z=$BC>!`4Yew7$JD+`DI1%}d3yb9vS<Q98Ro{ajAmjy^8y%+%DS_8o@%v=p7D7r6Cl
z`#OT4NWMsEW0B*IQr%VUa)<fn;}Oey+hn(1rm^aE|NMGHL33Xqt-bGL85GE27D#I6
zkasFH#J)xQCN(^_e8nYO*g6Yevv|4ZxDb8a^7$a3Q#I{Wm--3EF#q>PC7eiC<XPee
z!7h8)^thVKb=IdIJye9Aawx*#=|TQ#Ce&BG@*bnh7x?^Gv)M8;X}f<#_>v^xcuL;z
zk(c{+<Hs8xC(9J1=}CrhhQbkcDmZ;Q(h*n0ZI5vZaRzRH1tn!p<@r%6+|O^_xh*Vr
zN>2mq6-+wVKYoii;+AZGtgDM#hI!tOu4eK{<YxyT2{>k=i7F+?oI6XpnU}K@CX&l<
zQIeEqUx7kymRTPaw(J$9u>G=5v9g%$QiIqqRz~oP=r)$Z2K{Jr#&TK{4a8*a{vWO%
z4IB@@NJ!isRR~mQcBhW9)xV@fe;oLMg^hv#LXkqWjX2nswJ^b^636$x(G1a^J^lgX
zJKyMMs%avfdX#ANPu$Y1aQE<KOW&Qpv&Rt|j$yqb3}V#oFqC}0D#)OI+_E?>iJQmT
zhUdedHX7&~RT`ksn}vzIa@PdCC=r6|j`zPeG!U6}1I3Mp44baF1wNK{yt)`$@91V?
zj8elUb9aN?_BypcbU0~Fg5%HN76t9~tHsa&LDr??4DHuAboCmvc{kKAcG%wP-O{?}
zT6B#g??F9NfUt$pEI&Z(WqCE9mBq7l`QM#Z5XOkpBlsBn&0jL}vl&SrftogsRr$7h
z`iGPw-(}HB1IEELs6yt;D-{o3Mz*rMD?hU41Zn!&rL;GgKVEPaXPUkvz|DdY>E7*d
zetdb!n8E5N)6VXstk;G??J(HRUT7$ZKF@J!!xoCem?9~n!)C7V_|!MZ-U;iic5Y|)
zkdF%cO57^%^Gi1LOi#?1sF_7S>T(lYO*PY$lJ~5^%BQ5Zs<W)#_xl>ZXL+&ralbs>
zMZMi&u*lX(Gn#$;K$L~>RN=1IXG#E?)zWYhjb3hNfAx#{-is;u!pkyOiYLu++hDxH
zKzW+%%hDf)0T|SWx<5`H?V`^|-(ziCeG>Hi%b<v)5tOV*^+9<r_}nho*<6cGU8J4v
zS>^`!Ng$O5y}M(7tS@qq9^r$vk*<~ceSL>r9mn~qf$$!5O&J5p-1bb>vR9AQ-70D?
zY~&*Q6}D9D46Op#(5;<xBVkkh94}f{9+YC0VELe0VNO@tD#znR5sfK~;@|6I_kTm%
zoWOM9mNgI+tl=~-aE)ig_mODO#<HT;#$Q!>Nb6Y;^UwaY)w$~s!?R=Ga+z(PFlFZZ
zv_FcR(##+Cyl)KX!sNg=R2(!w?Yeki;c61RmdJ*>oMmW4Z`olGd|~m}#5d}3!E2!`
z+m*VOFT+W(D*V4Z!WZGuJ;ImTNsWWo?+tlHv#8s4c)#^*X(_y`lj3*d>7}Bh=+$-X
z_FsE<+kYaz`!dgz8?CUu8QuOAVSMM<ZuU%s`rXw9OL=3Y-(E)Jp+B3O>6}OwowA*z
z?z^ndHB<)oz9CJhxs|f}_xd}Xq3XOg)Ob}lTwrQDmBQGk;PkOYZf#RRgsAULi<U$k
zis5^SVB%GW$)jF9ya3PkWBq``=}G%=dQ2K&4ccaN#tRYihqDU5D(J6Q;D7kO#+;~A
z@ZRr#r4h1v|G7>7fYQs1z8{S}NYnD^&&nU+a(qf;`jyO@6l6QOrq0;z;YPQrBX0jq
z)!o}Pw{^b(Pf9gLIhlaA3xZ*&;4Cy`kuD$%S57sp67*JnP!VjrzB|&UMOA<sz4?rn
zAh4|WpsvII;lv`*H?w<p-n3uh(8SwboW3Gw-e?}ZAJlMsa?#Thowd=OYmxwEcI+O*
zuPGc7=;&TKZ}u-ErK;NL`-x{iJqNYhvV`bTMeE~l7x6yO)i?8-4Twd%(ym>dx)*Iq
zr=<&<X1o3*(5c1=lKw{is8P>vH&~Vq-?gq|Yeg$rJ4r)uo74+6igm78)Nv@Yx|1=Z
zY&+b}Eo)alk$E-lpqe1iiaLnJ@_?hMJY?Kf(B{cxjjkBa{Woa~lrqcW{356XI>!<4
zF~U)--0X?zn7$oqc4u<zcl{jY9#cGXPeXSfKp<(j-ND^(6vp9$U#pele3zBQw&|V9
zs9LFxyOm~RCM&jDF+obO#~!%rOwvSrTX@@^8gNZQR!_q*_##EX-R0E}nuhw4&wgnP
zPD!0lXy$l3<m;IIpH(tL*a{sX3Yit*m;fHLtoJO!mA;6?;cM6^1QJ7i{Xr3xpPbaP
z!DY+Ckukun{gk1w#-s4f#+A6=#2cvxTjr_>I<pM!9zVN?jEduBWFu70kG9oR$H`0S
zXEj#+ri9dz1MQ*)C7Ff}se{9&o@(qm8w{75ZJVf`z2SQ>6x46J!Q6K|g79J6_VKC6
zetzt%O>AS}xj~@S4q2v!Oe{Bjq!bxu8KRm#Y1D+F6mw6i$@Mp)mRvihu@j!{x@Csk
z5dQ>&K;fPd1UbT^H(t0{Wb3@ozR?TrCGd4j{g7hK+EAs;gaVV4Bi@K^(n;I81q3?l
zxa%_=k22aoH{Y-jT}{WfjP^>9yRQ)1UQh){%9L)Mzl*}yhI{AAzh~LpnVqEX`EYNO
z)x2&oM5@2`y<a#izW;4)gkN|LJ9))>0xs2NoJm4ytB=B;8zUyl1pLq#`y@B2Kc*9F
z(!c$JQIQz&p4LvwPt6Ilx(k`z@Z_<Pp@>qx!U8~-y?OCs@*WjRioC3)(8{Ypmp#2S
zj4o%S(`e7WB^B`n<R_Y?#~1HZh~6F$UtAd6OTN7PTCN@Q+t*^ISfVL7X$b6BW1!C7
z=+=F&Io&Pylm}I?od^d++3U-Q6xdVz4jNQDXeIL~OfjM<7iD9el_lTXLSf{`NNkQ@
z`>PqSZ9P}1?Lc{fy;%O$;c|4r)5pgK0?zv@1F$zqB6;1E_sK|#FB<2YM-0}hYtA_w
zEF~-b(^U#O_kL<_o8>H@HlGFf{MbXu*|V9B?(oppKd2z(&?Cf*PaEk(Ruk=tt2dO=
zaFVMbBU$M^xL6vBIP9(Pa`x|7#~K#<7PoEhEzf6NHJflW_tkmG3+06@Z(F^5dIxLH
zJHf{RDw8$G>@B>uhW^vhXC7)K;t#|obx>~-wL(NVAL^{R+C`KwM%5XML1`Y!gl;si
zuwnU(SwR$c&)z-lBZIxlEFUv%iZ6qp*vB&D@wX337__-}y5F-bz0#DEN+&euU@i$Z
zvCT`PL$9sG_1S0$=!xvlq&Gkf-YR4DUw=Tx6UmBaTic?t&DMo#u3J!SS@3w}tuFq&
zy0~@A9;;#bti+U1?A*w%ZHi=1_ul*)G&(hj9ggT;KEh}`{^|uw+A=^P(r}Me&nxLs
z;#ovesuxF({oRPIW1fqOxM5Pp;^y>zYs9hUGauYL*r!%QB$kzm(@zmw-n3Ggr?aP&
zXveq{YZ5}iH|7UIey|ajHY_CuYGbgf4y()ElU%IuTPlVH8VX%&iO;7E4WpY_+axup
z?Jb19V63*@!eVqiL(Re84-KGuS|(F+#TKlKIbUVSvMp)(p=;Uih?+FP$d$3VEH{vO
zxOAj`2u_5Hu#HhhQ!#t`mfM@d+g0rXhDkE}&8Ze9U+#p;OJNg&#KFB&)E2BK8SRTz
zrA*F+87m7QbfDUsC4jslf<%H9$J)#pYCEYZahhl{ZtGtcE-LL{5@#gE=cz|Rboxej
zCD#8F{=8;Ma|qr|uc)x=Il5>Zt!biT(`R>xnvIWZXxI4;M?@`gqe`EC4UzyHS^fAy
z;10`ppv<g)88x9;-a$k%7vtXMvc#}_B@hkb8u(W1RQ;xh?ootVWX*FNt<Mmm8NbQj
zTJG~ds@sKm7`%NOVED^0VH0gDg{R+U=e+yddHJ|XCVSC`Ai47Tc4|V`&4c|S+bf4?
zNoFE??_wDXZ0f{@$tVoL6ql{M@_`}Ab=1%z$IJJ4avPt{h<MJY^d6}0ztNp?8BTI=
z!%5~ZodSi_qw)7-rw9^=rc2%<`j&Z1x5!v%5YFEvu>7m16xiL!=Ngt;UoK0ZRz9dM
zBCy;Zx4k+W_N-<z*1vv^Uk=E`4yR;BGvAOP8AMN#-JU4JrJvPT88C{Yq2fc$VR&%y
z`Z1BECTY}h7qW@L7;&A89!VdBXSSl(>Tw|m61)T@Q`)N4(HOv%*-d-hR&6#5H-gzL
zxP&|~=Eq;#|JcRuzZzIcrrDZbR=o-fr7B=JvIgO<yd2@URTdgkDXg(uzN*d!Toq1%
zo$X8@S{kR0XR}Bc#I!bj7juB^GD?t1GZ}+!-}aH^l8<o2)Y82%bO?$Q@F9P&dWSwd
zef#;w%|#wKZGam_3Qy`C`K@ior1$b{NRJZH-Wl#So#@)mnC6`MKVp9Ax?7bUziFll
z7!wkbI8+~up?*rt4Br78#z;)T4x;uc-~Ec5z1r%_Jc#fy9G9M6Jvc%H{>?1<d}~Sl
zRz7=-MGKR<@0(dn-#nR7#+k?zPE#c(^=ZBry@}3iaYnQEQ0U52)!zr=&b@s-Y#`|h
ziqv)!%Zm04O{P2B1zip|vblq(1n=DO6`{;iFLYJkdn9-!S=(EjqY+4VQ_)5W)$Sd~
z$}uv+rNlaRfbiOVLdz|_;kF+dtsLCk!AeaS4EwgPe=M~bElN)=PFa7LC3015a0kqY
zMOuYhF9NMpYL(UhPU_D1v_hn09BS#gUxxx_ze7<#1+sw2z`+Q~;n<l6y?hYalf?Yp
zvO<HSd;AU-_?S}E>#Ih-om8;rk6~f_+`CzyqrOHgi*|+YQhhlCY17EZz65=bSgzCy
zd*R|~^l234yqtfA)}`Vm5LVDG`;pj^(qzIFsGAI~iV^vyhO_3rh2R5*YIzA)U5pu>
zMo^7hRKE45$2YdOO<2D01jQ`HM{HjO=E0V>Jho&Mlk~wYGbAaq44{stEr=HW_@-EI
zA_d@4%gdcDKhR}K0)ubcUFdX21gdh<|3bE1d*a93<96bd+}UpW`CESl6@EC83de10
zCpWU~L!z2B1>)5srSWylUU<WkK|roy9_z|0i4q#%jPdhV0$^#0z?7aZF6_^9s{Fl<
zW+H?v;zOmL0`!wDJ$*U^7`ex5lZlGgmipnR|2yyWj|@+kGi$$}8NHS5{HD=MI$9O=
zSeucnnf=|jc>DL{UASG3qAhOMh!|`uECKE(T_1i5s~$jqBofM)MjHr5pnz$4X4R7&
z<3+AA+jju0cV`MMC~2Ny1O;<hq7a1KXFIE;azwB^baV*BhKH6WP}c!Sv_wKnoBB4X
zodS(C$*N*u9I`Kw>>P60zKu?<prirIj--)R9qGw2^a2YCuZxuJ_NLEXNW7@AYwtZS
z=I%wd=EuSq(s10x=umQA>W}z1a7=mG$5WYUv+P}(vE!+Y>CU_b#f&G9sj^ECX@9g3
zAP@g2{qdAqA<;1T!PQS%F*mG^_o}b<#6B6)v$B&!yiHn>`+D-RUV#;hcLlp;CYtFx
zD4dv>*?E*s45$Z5O*#`E%7eY@?~Ucb;FgCUD5AmG{FTBYUb^C?r-eqK)yF`H1rtI*
zfwOgK^$Z}3*g!~6w@deAbF4>CML%+0O>f@2@k6(+_`HWe!0o~O4TGwOiUH&i{=QVE
zQV*`5SjEK{Wj~rR41MZCJFp|s_DnIKYJk9HzxR<ukiJHihbJ0_8aAQK{3=8Jt9xQ)
z-rj;R?4G4)4LV+eK8b;-=_HP0NkQkYLzFU6xIxrIHWn5Asn5Tf4gX3i+y0=m%}`x-
zjPr1is}N>}Ur-TmN&Eg4@5y@ksuMf5KXbiBN4d0&;ljK0zKT{O0pk&RLisp5-AI*}
zUsmH^;WU!hu6MK{&#>8TC|txSV<lq{fszzaC#ob4Oa7CGx}0fI$y0CHabLf^C10?(
zsJ5~`L~LI8#gLv0`QhElLMo@2rM~a`_usfzEJ&2quh!qsNVohec0xIVg!p{CGnfA$
z$5_#Z6qix9R{Zt7G{>G=c}5G`U>J&3Vzcu0)j-*-MMZG1Pf>QtULx%S(=sF4>#x6>
z_og!C3k%dq8H_41LVLD``UR(D4+wSX<1X-cMjUX%Z+)eMBRN+H{(&&$n5rUGY|fC3
zq+eJ(OXMK{FeqQMwI>uu6}l6DB;|?!dL^UnZ`N!MAt|f+)NC&mvLW6=j#r<<%;Kki
zls_RCMd!FB8x_8;GFwV1=vW)$#_-f9sti$6KZSP)OZw}y)^LSfpw)Br?cY5!MvQ&I
zDeL4+g^3x25l9e?r;DxlrYz)mZ=Qem$A&;8`kemTBBL$;N+RXsx4PO1yw(<)u^W<C
zH8?fTi5jW^NgFi{pJaiLJ1s;e*HBfPYLboG_(rr-2Or%b>r`+qq)bR9<9!EGjaj6$
z?o7npQL94Y(t>P{bAMmVE&^X>$*CSa1goY_nQX}&qpEQG`4o;Z7->Pz?lOmi%nUjU
z%jZ&kI6oqb^CX(>!io^tR$sBFOn&T7US~Pj$}t|Ny+3+?QP)yC3ChN6-l&DzpZkuE
z)Bb$Qlu!D;?K0)!t_C=TGA3Dz=GCj~dfvLqzBqRd`5F@HFG3w;dukChuV!i)7*%4&
zlovYnll3_E?3MG{IqpPCmeuuXx`3leMhYG8{ldcwZ06-}yUD&9g{@zIUY+AF%FeS+
zFJf!aIIQ+QN@1K8<8H1DnCZtHK?a^&G!BKcDB}S+i+Yr<h314wQf(Sy?Ot;MHOp8h
zWwn$->>Ci_IsQRHWr~FW>ha51wym3vqV1`lc^mABX15CC<Vm~{YyuXjDCqSmf`)lh
zXa2tSvo;E3Y;;sl+6gyuGlFu?453fFF`W{fc+2mS#2)({H-hK6#c_NaiOnBcrPeC)
z3+ulb_Gte8rjrSKHFCRR$+4_MvbRT@bocxW)Wm2O@Lo!MO>23TN3+He)IO8%5%GQ#
z;L=LVzI762AleGz9_W}Vq%Cj*MB}8xlCO86rh!!NtKbHK@Ao_7Cm+7iTYBdjef&d*
z!m4JpRrl$-#k!%^*_`AzKN6ndl0ser?WYitqotn*a>-WRde#PC`Z>Ng<|9DH+QIS$
zV?H8X>lP;|B?8i;u<Q-O#!a(i$S|!)rOwq*PHw5h{RV3OWv_02a>jtH_}c!&^dETZ
zpAW<`fP~fuzQE-_2z#ht8gf9_aKqAPwHNqq0tA1fKmRs@K7c^il2uQI<v;ZYzqf@p
zpqNbP*Ln?kcK`lo!U~Z4+AaBi*JR)PrwK0n+TG*zdmvBl88K8ic4K18P@aEbnmAH0
z|I^B6sm2IB<erUR<!>q;=H-Nn`{zAw=U)H)*1sr!p_{BeKzKMzA*ne2=WqWa-c6lF
zePyGQY!FCSpA>P<f*|{l_3gK32VW$};{e0U)K%FP$b-zZur!~xyB)#QvX0)w+1KoX
z{2MpKeOcuS^5I!VO#dE+M|gmd*?}GINEVS;atRRjaay-rRy$u`o=697pZ;7Pe3as5
zFt-i}8qGi|mxDvU9q|B^B664W=4<`Nv)#0={+$qwm?Pj})qxjH!uKvjEG-WKM7Q*1
zwB?39AyuBV`QMvPlFycWunKK1z<XX5DM8fK|NLh|87BE9YTHJxQbkq^r*-QbP$0wu
zJN^aY=`|kDRYX2{8L5739|+N+CXEst{hOYf5imy=GuGE@@x{TFl{Sl9fCzj1mkbH5
ztm=r1-Sg@``hr@qf53pDqfi1H>i#o%7%I+bNGresn<P80-7`<?k_r-${$)|qmvVLZ
zr~=g1JD8uqtpBkbKq>^RW>FT48poi*E+}{H44?*uRmV|kx@-Wkr#WkI)`sPw=W?`V
zlpSYGJl5c7%>fdB(E`)m_192}`QBA_GT#*eV1eDdp)^3oARe!l8UHt7_H#;XPjpNv
z25}y^bXYr1Er3l9ief!|1t`IAe3Z00ce+3#IN<>qVOG>I1ZKY~OEMAwb;k~4Ir8KU
z1M2{K`10|DS=hgJ6n+wzDmAuxv6heXFaca)-mYVi-sfRyIt<tMm{#otaBdTb2$=)K
z#p8gl-E%bvk`I0@5J@I|0i!(cYGjdF`Vhmvc8-o1#K^7kY<#*M#|fB?K>*Tv4Y4iz
z7)T#48>11bGZvt;G?r@EeG!vw@wnXKn^kzFXhZ#9%fIUZ7}m#=B1~uG@M}3u+r<q7
zP%V3u)>@urCH6_+xB3wg^f+=rRR|Fpb4W9my}#@rX7~>*_pd!k4z}>@x~9E_bU+S-
zQ7SZ${>T()^$-ckp_e*_Q<|wNQUfqxNuyal9L}r%QziP>YK)fv2IW9~Tju)V56D^u
zG3|W@L`AvaHg`3K&0TxnFpd=5`5_ABK{SR!J;Abt8^)b~tY*F#Xoxjp_)m~cREHth
zG+-t;Q7_m9tRGWhV^<pzg6su_<$)Ry<VK32=oM2H2Ja_&f`6MET25$5mopQc6B8D{
zonv_@5yI)X;G82%c?5ES34rDG$)rR0=H1C<i1zvp$UYVV7GZ~(WNrEMzJKm4@OdU2
zw5W#3HCvn~sukibwrV<3HLs%P-v=DeVcTcG^|q>*v+0imY?_02h&eyg*8-|smEv^;
zqyL(sBw;Yq43n!vwNuW=@0`ucrWU|lPmuC{e>@8nk(?|;IB$#6He~K0?&0GC=6`G(
z&_C}X>c6)6`T`v6@D9-m$Q{pXT55<-4@!eFR@DT?KDugkYp&LCA|2SjdG6<L)|k%o
zaV-fwU=Ah>T!EcRc$Dna&WfR+%MO?^5^h*Nryy9@1c7rOUCuy(%YS_d<wzG~V1IUD
zHC%z@<kXpkqD4Rj^&YqU1BQKwTLjQ$8dfa0Y{%&~xvXk5ZT$~Rxisj1!itEL8%<f{
zfZ}-v`IW-+*)X`VtEG@%AS|;YkOCGARVXq0s}IuvS0@859kXF!@r?i4FO(xzi0lBY
zz*bLN|LZ};0DBO$GcZ&e0KJX+>Pe5_RhgpkoWVRUWHte*X>HXU#Er2QAurHd<Xua~
zv5GY@Bg;MA378HYSLS}?Q~uX5f4up6LY?dTpgOB&uei4H{9QSe($+%y5G!+K9IKDq
z@*$e!ncla~faCZlAaBXteGHNOLo7i@7PbRy{u+++_LjygvOM$S1Hk&%O_cois(J<p
z*yHf$A%0DIDVVApfJ}HUTw##%jgy7=S_50k1`sMO8i44ku)m@LaTsv#^i^u1uY3=A
zkQ^=ZdHBh}N62mtfGe39;L5>(L<=U=4S$t_$j8Fa>+Qm702*>@Osxjh8=R4FsNZy*
zz9L1eyp3300d=$+ij4iK#-X=tbOVyEA!^)~2UW>iw*^nVH2`fchV9$`;L_{;q2md_
z4S1krAG5MTqe9=Q>yNhEuHz6SDc<AqJPxvNjOCvxRZ$_#n+D*>8UWnQ;~5q;#jT++
zv!5PHX!;&^5xuOG+J$)0S*h+kD?HM1cx7M-3dKb^fKBJx(ar}EfmC*H0R0-*2E!MH
z?jVG^e|ww(pnE>mfzh}8AlqcW>GXTDypJisZkPvj1sGPPP-JH>&dFUyok-W?%;;j>
z`nB^A?>eWt8(^Jo3n#cs9ZRMDLOv;vL_f;+rT=ZK9(A*_(e|~Y$(KA3#g^#o9wP;9
z71bw$C1K!YDBMAK5@g-@v)hj|n{4|H8HU-)8EV13Cr)q329-(w>}pFPiU_|C{L>H?
z>sq1SN#6;;C<_t9=UzW73J7^18F|{F!-CH)n34Et)e#W0S9o@Y;Ii~R@}%tp)+=TZ
zmy~QV@Cfoi8UcvMVQC4>A@_23Fozqv8q+0!&tdJpteB|=)yxTi7h+m{Hkf-RS0e}J
z46IT<RqH`)TNB{huJ3!uK=i0EGm0J5s&elfrX@IZ$YlopG0kX@)!DAW_x&LFvXnjj
zS9li@(p|@y#&x};VNfV&qLyKiahQHLg3cx0G@I^H7BCSx#K{{d&I@SUPj{zi(M;U!
z-vZ8r;`s&02)Qy=q*%UmKZ=?kVn^En+{M<Km!n;g3g(4neo2#r4F0(9eI27F#8P7s
zowz~yu+>^&#ZX_B_ml5VG?hx9XNBZGOzOHaO&VjPK{5d=nmP1;yhU1WZ@Ok%f`qvb
zH0e&~eJsT_jg1h{CHQgDmJEg0sy~rLwj1T!alpxBfLB2~#0?+I9VC>1eV5xJ_Q5eH
z4HRFJJtdkYnw<bnLjuu*>Joz4(@Q-e)&iM$&R%Vgv(+!~^2wf1kqT+!fM!bH-xqr?
zluAKwKe~E0&f1wz&jVxzYM<cQ2JTBcESWKum(^Dd#scc0>25}jEWEMDMWyX%#$5zz
z%KNnQ_bErx5I?}0fS=Rn(jznhA<d%7wNxb&EQETWg6&~J=2O~BTh~MCn0VBUdIIV@
z|2|z(D^NLAzQ~G74$%K>p>3yaEos~jRCyO**!5I>Gy<;7j%vAGDTst!ifKB}aLo4b
ziIO>3cS+Au6QXqJ>XPgp6#y-s8@m|B;@wVLFFD^{5zniy=oN4&hDyd|7J2WfOC^X`
zIc~*^vTr7e%`6nCv2441jEbJxfTA^pAX3QvMfKB6UN!J*D?@~M=o^%5v>aPD2mb!v
z5llq1l8tO(s$RUfz%$WJ`{|&b5K?mvqKKxuF($Y|1TcCT+*}Y#nT4`zaw(n40>A1q
z8EY40xnRnscPnq|2X`Vx=f8C`QmgSa{`RDVs1YoSHdMC~TT(h{UcGZbW9$0?d}XD+
zL>ci{*P%`Zk%j`(LyrQFAo_UM`}TNv(zrx1NpbRm92HtLx<i+2t2nms{9@HFTRc2C
zs%{PkF1!uo{^H2Fcx$Xbav(Zlspk0efU9V$5csQ;xErsG)TFE?7bdo9|89!1tX^M!
z=m`8@9rmRz?WIYGeR@2VQRwcJ3jsfk2V&WbE@t$7BIg`xrOYz^{yTgKX#7nZz3PVF
z!sfw<X&-pEwiUX{@0x5KG2jTP=~w}2nP;h5D0YG64AeDRd+JBO!*~0@3Fdj3oG%Nl
z2Q<;td1zLB>!QX1^7=1Fj308u_QLbppw@*Y2{l0TQ}NO;>W^><Ep#nR41nej_~SSx
zJW&56<~{!C=v2N3?Tpr&l)OQ_)*;S_O-F^#&RvN7zMy%_LB9BllEQOpXeq&ar!#<;
z_@HO2{IRnCluPkGeyE@fz2stE+N|vNU_innlrKrEZn{D&lxhME2D0`mmESxLI&xqz
zViw`Hnn61qacOS`ga5nWk>=l<fmjwikbzoo0}Z65^;U&if$n0MtMPj&!b6TntC%to
zw0lxj5sH)?vQ{Qsy6h7FtSUk<UKI*Te{tpMAq=X5CmenZW)}Yr-U&lD%g??F95>$)
zg&3c;Bh_oUsxndcx2a5Z1yAfIn-?YDkh0NxOZDvm7c0rdEh|aMh5={vrhPZp<3{l0
zU{K~3<UM<1QiNsiB)@4#%~pj5=l0zFt#^Mns}cv`;VtdT#+9J~H9_{KGI9i7LpCAw
zAV+ch(1YKu%@Po^*&Z7y|LYH_!(zaSQR9jH(>(_8MoT37q?-ja<0aGtjWT!bXS4Lb
zAF-nauOfBPhcQ8~`s^cA1U>O&qX6^ITxx=wdr?&4-zlaC3PsSWl(xEK|9e%UK>-18
zmXP=xaS^cxVJcw3Xhkw)JN$k36>`LjHQ09?oktmWz4=;jUG>8vw_*mE<nnJE1FSw!
z@!!Bmn&=>K+tsQ)zxon!gCtd99AJX(J2w3-kOIDL1)B7QXDWar6oMdlgHn$;P)JN`
z%c5ul@#}M39*(^h5J&oRjy+2y0}4XL{>dSbv+vDJRSFKx1b}PddNG`}V%(l7^7m??
zK;Xp-6XSWEnlA?}uzKHCpo?L`4C2<Ed+94=paQ`d<XU7IGyZduQ;wh@ly7jq%in@_
zmp3zo(7Sv9ccq?r@wzuEoTMuw_|JD^(6HM3Y^xpp^$SuNy?&+|&K1r^a7s%zMN-$i
z7eW8?EE6Tg<SL=}r0LBDj6nH#pLcZ5ZcNtrix?3XZ3_^E+RNO_`)$Z65G=*Pm?Y@1
za;g13CjQ)vD3XB0%-ESp$?p~MS2PCN`UyB7R%1*gGx%SZAg~Oc0~5iM6neul_~#F9
zK2+X=R;EMSxc2YmabtFFR<|A4O8>7{jXe4LuQq59K$JhM?tH6Y5Xjd|%g(F56iYQi
zM8>>tA}4V63exSVF2vybZ6{WQg~rfVg%$0gxtF_fbs${r-ch~?7;!rU{+&{vIuNcH
zLpS~1NBl%D?HF_q@9`%%8>$GvCseC((77b^j>dHYkT)$Ck0kBC$7vtP74cvN=7YEf
zg(JeE+kxz!{mu9GAs)^b3_PcX_Ph{cpu7-UeOHUPSsIERwKrXMY<l4$;HF>;Vl00?
z?O+KypPSSQEndKFXkH>D@pt935Iop8OiN;l5#T#M#yxSix#wqDJ>6{qd^)4qDoESK
z0RaRUO7Q1rer`VhYh%2_4coR6X89O@V_N?#3Fw0Y2^jmTw_}R``t8kWjszLxR3h=g
z|G7Ki<OTynJ5#$!`@a(g-j%L3f}Uag$DjFozj~pJLrcnX&hC#T`Lhg!grQz-_$vOV
z@cpk*0zb_jf@W}A@k#y9I!sLkrkL=s{f}+_%L%#pSp^oN76}jO|E0$YU`e67=#Ml0
zZ{ufxWh)9!2>IV0(?TW<sfpDr`hOR_C1}P(x2(kf_831BjMF>ayIj)$+jwr!4D;pe
zj6WUy^8rl?jMM-3=@NRb7y9&=HAcrlqgLa0cH#Gk!h_z!V;2Pq;h=(u3(*<~vav#(
zjzEA*A15IfQi1q0%^!sH!7wrnWGRR~qagAA+TkxfJ?JNu^{0Q}Lw#iHxXz5Vy%yq3
zU@QI$8U0yEfCdUo`jiLVRKvWu{^#@Hb2G3XC$5B=xZdq}LSJ<ZgSjmDvvvrtZ1Vo4
zP6q6pSdG|{y(UhsU+YS5gyma-q;jC?wb)m#h4C!vpLS`37)(LKno>bV(la<CN1*Lc
zEnU9jFdC)u@}KYi&xY279vjV+oG70K9&80jomAyg^MPu`fAk(WWFXeFu9-_ssetDV
z%F1U4z<t?aD-EKm%;vNEcNEVK7YuBY9TzO7n97TUqt6Ni)dm2LR6ldaVbwBKF6bZM
z>;#sX3oFpj+m!;%09;|8%4ds-6m}u($?t!kqrif@5g;=m^`^^79HnmnhhGh2Ls7#W
zzKVY|R0tlLqgF8myk`_F2(%4=6N_7Jx(VEE5ArpS{@I}qLbKSwsB_0iVGwv*ASRYW
zWOQ*LAia<X#7d?ymFfQTdeDvyY%r+&D<b(M^TJ;0bBy02=bPOxAiCRLZ4evLz<hDz
z7X4}Q1`722H}-tIc@aGjd5&w~gDJm=gUC<n*|12sWuoT?R+Rs##+Nz|jD_V1L+){q
z5E}gJ>7>`m&V@#5Rs(dUmH*slsfM77yRp)FAs>Vi;rjMKdB1t)PWPEf!9T}g9VfQ_
zpgxn#p5GfVv-U$t7~}T$sK?g8H_DQaKD+qtc|EHd+kAPfHM>yAR<FOFSPs#gYxOf<
zfa=TP(%6arxCM|$_z-&QhyJOgBqwlR9E@c%93Zi7)K_1#2BCXUAJa^CYZP1wI~F1N
zE`{?`3!Vo-HelM#thg_AUnkiQ05^i#sKz+(UlN3=q$rh(vmX2kDkLN#OCSKr3N~wO
z9nh&87?T&YY`!|*9(B%6a?s=eIktFTLe0+uK+DUx``&9#0BNYW+=dmX<!RTj=;4S@
zH-Ru=VYzgBJ}4b&nk~D@Uhv}RQ9xJZ{yc7#Tky%VcgE#FMf!1sk;j303ebET1ef2}
zYbl-{7p{*C{yHlbftz^o6rG76tQER@lezs6J~&+tO@J6PtFe)yoyMkLr_2v<&st>5
zrwAPPw0I4&yozu#CFV2<Q*rPIuG%cEQB7EkR4}f0^~2VKb5N;LHN0MJ?j}qkQ|<mF
zibgj66_CK%XyWX+!=YY6?m-Uec5y&TRiNyQs8a1OW6YTh)_ga|#Jzl~AAmtQDsmRe
z7w{Oa`RM5T66}2D3M|ePKu~hnUe3t?Q}qo|p)WhZiVw=UI>B2GMzunDk?v4(Yk{*B
zybCC`M(iOf@$s(yu#e+dsK?>Mij{C_%`T#nM>i^$cIaTs&f8?orl!<VzXTG1#}e^t
z?DS1QgiSMsQQLz03^%Mifkhs6<G2peEQ#2)AEd)A<8+bV>IT8zN}vN)0(d@;YwQ7L
z0|jfW9$Joe_~2Vf?RjS_I3V38D9AL(P||2nR0AB*TvBhEHPumd&<T=$wW=VY8l0s#
zO4FzCIdCi>)RonpdKljL={`oyoK4W6c^PnnNUloHZ^GHBW6*M}WS9}Zd=Rn?;2dBz
z9DHmu=*QX}m1eQTv#3<$nQ6{ba+tls&A&yda2KJLSvpWsFNAlE^%oBEAqcL<LY083
zUz>h*lUS;cK$H~+JSL^-v6J#e;GE2bMt>`A^8+8M8^A2C0cydDvANVG=H}CH#(7!M
z>|lAiCvk2603QGJACLd|p^yo-=K$R`5XGBm0m3gZKP<~Cn5uEp(L}|qk=;p-svA(w
zZ=-L90RQbhq=<I_)>IrIHlJO&tMn!TPN0J>vAq9Kwy!%8J~j|CTS5tVdgo8eIU)+p
zKy=wpAssB<o*Q{Vup4V6yJCl)-`tFMA=OKM-70R2gU$fF-OzQwm29n(Q$1yX&Od2M
z;*gb*@Ufg9C^yVX&^3^{=fQX(Ge?UP3uG7vfkcV%=TB_L19#28Hf^K_4S?XC^)Vm-
zsQ~qaF@0vh4Fkb?UobKbbxM|VL;mC*QxM>34TMO`1*2=}r30mxuFD)B{ZNy+Chw6)
zzohrYHhPIGA{i|bo$31{V#l}qSNZn~Ig#>TKPrM~FNnn`%|8?ZGKv1^a^?5Rp5G5H
zlFm|GTWr6d>^ZJ2OgcYlp0BfgJ(TD?RP#M)I5Aj{r+(5e{DBu7LKptX;x*T#_Slul
z68)gQAYj?Pm@o2*5vB~vlV3ogP%wc<m=_5Oe@T(*d<mFxkvTv<vI6AUrf3v6Utl^>
zHVKuIZ1#(O#TJ6|PDiI*(g7t6gFs}}{MLf^t=sXCbx(fkO6)EX20f9Bj=dp%iPlx=
z7GLun*hen2@<tHkVYe5)Q(~Bf)DT3(!W~4&B;`!KHCaL)l)w`lnNK4Bl+)}>y6<HR
zELGSiay@GMwu=In^ZqM-zJhp;e0g^u4Qe|Wa)~WO;fHYEkgFI!2{aH7z-EsFNf4M!
zENfdL5bHFaHi`-UdQb5cB6cnf=#*Lg3^m^*ByXqfvYEwgva!(Aup6P3E@~T#fUc}m
z>I&R~go+_PhN3X5IeejFkny<D-rSNdZZ=nZo*@y+4bf(O(T?>{nuHsS%A?S|_4yD0
zZA`r4v_}Cki8a_APY!@hhjbqE#R>UBtXxzu?`26tTUZ0E=N$~2`rY(%jk$hu6}3JB
zd2cA(Ro!<E5~w>q)gYxcln641#u>rff$OM>W}w|@-X?QxDoFPBi7T~lmj@s!w0KTa
z1BXR-$9HpRwL8pHIsgu_Ycc>8uQLwQ>sS`^KY$s4&~r`o+PBXSLphW9?Af7I=0&YV
z2|(zTeF-+Q2`fOLTT&L}lf7*N@Jb1wJd-KMX6t${gn!Qj5+)j<P>ajw)@VBfas>Kh
zE*ohh%Wfe3m+g6d#l3U|L^VZf0`~hr9d%!IB-0f_JemSl>Cwp{#BLAtP4#}s2XcU^
z4<twxuQ6b;+?2QI{D5;TEzh%%#P1j)#xXL=&~@;Ec}iuIL7&qYHdk#rM)TExR%pCZ
z4Q)+(v>aLQ4CF2cf#uy(jrny1(gxgU?Vvo)l)3^mb$Xvgqr7v-9_*jjGsfL@gff>K
zVDPWs&vV6O$1@htNWfRevxSoKb5Dt19w{3(0N>&Ec4caIaE5rWlsZuk&s!GH#6mjr
zCrF1pxoRc4;xE2zy1uMx1o8)dmKb=$9O>g=8{{D%)<?M*NjvG*PXh*&f<X}de|3SL
zFVCE}kEjg*FO6@p92y*U&2k#h8l=#iz}UXE1-|fYu3WSI#;<-P@ervUP)aEovm<BD
z`|fI^uqDucbOZ!MQ5|53@zABrbfM9?bcjhNfqM{=DXyN-+8!5cyJW7lbP|K}?jstR
ztJkTwV|$S;!v<yWc>x>J?xGKYs(RYKd@{!~IPWQx5j$S)82WgmOJJ`Cq2g8_?W6gJ
zpM%%z>6P>Fcj%|a+=9^FfIx_DZsEg-(DM!YFgGx8xbX`NKbX{Ps_OWf&wB~^o&vp1
zVW>*_g-KWe2L8^XO+QmxHF2W&rHdpOj3flJdlmi64Y9noYq5}db^|UC47E+;p_dR9
zQtp@F6{7%&H2jtP(2#gX`ct}Y45uel`6ZF@=o+?+`GhSX9!h{(3Py4EEJej9FH;je
zm^<I2LCUN*6+RtkEQwR8L|g(8RG$g<L#~r~)10K8(D_r(Yg#>+p5>Np!8K3C>tXI-
z5}s=K6@Uv;Gt`TLl2Resdy^Z@z{U!{$un)RGt5hniBU#A`lOb!1xe%<$(`k|&3NSR
zUYEtw_jhBcW_i?tL$o}O*ESo9KMw#qeD~D}AT3i*$(Bus2e$$NLk`$|;GWE{jg=Wn
zK!E~e=W}$IyFbC}W@sYvb<m{i<;)E7%yZby@ilZ9jyR6ZH#q28&Z+b7)fGEe7!N`Y
z4xYpF3(sXRBQ+^i=95D%o3hL;95WIvu4ZKUYs&zD6JuFj+#K#wdN-<VROErW1_8qP
zb%{rVi(a<g2>|hx9sE%3>a`6=p};ov=JWInU=FZDrPF?e1QJL0U6->+WR74DRs35)
zve?4<7?OF6h4?v;dAlATfMC_MaPY52E}L>>-hS%5Y^6buC?E^_JyJF%;y_hz{d_my
zIhmwRLpG6XI5Yg+<vkALyGUss^}uDceDM<L5xAaISbqCHIdnd220=TxftTeCfW<~T
z62xCPE#T;DQp}Woc~n?6>B2%!fERtBmZg0>{8RAyER^&){c5+TXR1I`Jf8(P{%bCs
z2nx-^*n``Nqw()DsP~V{j6H$`LAV*JhyneDgnG4baKrf>xg@@-ZV1uBz9sz&9O<xA
z8HU4=Z-7L+rYR&Y^5vu)7FXmUw*IfvpANid<T{QReL0Xm9z?D+y(%-j_*5IJH6MUz
z!Ef0o2#*T7prqFIttQBiF1{g(Wgve__2s)Qvm_Jw;g1Z*?+bN1Ljr7sBiC3@-IPYZ
zw`gA7oK-exBX)I(=WI`_lY=(`ip+?VrA5O?#?a$L%cKcb))au|S9<DH77&bsG0`;A
z-<8I+sqz*VQB{J=g_}UCBfcZe<J}A22OM@IKt09Y5PJ2c<Gmf-Jg3N(XpHH;xhNoZ
zr-jsAX5!bChlN!la&h8SwDKs_=w{2ujiXJcUs&c)C|AjNgcW#ArChwIYDA;A*C}W6
z%wIOHZ>~O28*D*7dSO{9P2aL6=O8KDV^4ahM?o(+Px2f_mwK%Q0aq_uGADFBwnMh?
zzZT*MrOp8>w%Uxq_N3V?fHSEVNM0Bvz;*w?r?tUEfZbgQ?1e3vYP`V8l1AfG&Zq*k
zOtC(ge4kD92)X1{6_&#hH0I&xO;nk3%qY3q%cQ}kPbp%8xw;n<>kO`MhsWPGOY+!j
zJdnG{_Zp+-I)l9iK>+&#Gk2BX405}IJ7}Y@Yq>GTc>pCqg*R09bH6yQ>Uj#ZbbUf~
z&NwHn^YM*^XugpA<^qT|o0XX?sD>n|F}c-t3&@^**2LOSb}dL%l&m^Z)3J~0t@_m_
zY@?0B@A*{kF+kqDyZx}f8DOK*nOe+rk6hDcw5gFcUBbe*(w>EM8So(5;Ewo|0B<DT
z4})YzQ>SHCOzvJe9s&wQn7`2FlVhVvdi)WLl5>o$ZpBMZNf?Mb3l|p`qkpJ;;<%^X
z^q!{5$_a41nYm@_4QGjB@^;?2%i7vhRnqrTR!4<DM@Z8<=sc*}YMxKO!Jt|31&qzG
zI~FL)Qlz04AQ%r86p&LRb<Ee>SIAazv8puFMQJpjne7=)`m1xLZb=uBSX0Bo&CT$6
zoI)m`L2*!MQaW*zwZA%ls>E7AWg9ij_1baUPIf#~tEz5B^&oBwUb+qs-m$Jk=qz+G
zRrnUaP*gYiV~|?i5#<p#WWKgfLqt)|fa9u%Il*k7yq^ECjcfzr0&L)t=^Rp5sx^m*
z?_>&>Y;e*WoFBC4rcRirlz)|H=8=D8c6o6k&r4JiKX`sQO~|Mow#6!@RRV9sd0kXb
zU1_~2aQGOF4Cbi1p6SfNMWG`Ax+36g_Ht5Qp&#TAwXs*WnxymJs*$gQ6jet^xen%m
z@_NZ;k#=|m`AYo}v9h##w~iD{ZPpzCjI!_*28*ve^Iq1+vlT=kp~53#(^i2-XAy>O
z{^a!iu+|6L$RQ?;RwhM&k*#deH^d7wT@1P&Ey#P|z4mN6uke#?$RZZw;!x!<J|Z}}
z&@i)Gw4stM%FOONhx~?JJ!xVMt;=Wbx1(fl`Isi16EJM_QD?FDQ?q!I2P^D^Z$H{y
zXi$rC22RPIrH1Z{q!vN!ad3R=OYCDZq`pImne#H;ds>1LsU?&#ynUTQ)^0)}WIy}a
zc*n~oCS@5MdB2cs8u$aQoo$gdP&Tr)s-8&2v-^VfGJjU#s5A&EzHdJW;dp2b@`coV
zN=*>I*-4(|TIZhEn7gFeRdsCc)4^WH3rIX%um57pl^YS?f9_+niL6@mWRA9NI+6*m
z!ealR=IBD{!iVQ<Hqw<BcDDV!h`ch{ZdCFf0oR%EGjaYh90V~!g~>Zra9=9TiFuZQ
zfDGe4PA?f4%~MbKc;@RCBxyQCPIq7ANMIvLMH1k$r-ET@gcP4IuY=Za*T;KuG3HHM
z*K?YTQ1DxCLurTwaHl60E{;cI_(3NzIxZ0(G&%SsKL5^EPXT-~ey%kT%xp6zuBDGr
zPS9X>HcRnEC!*QOzs;p6lQ;B|YqO5(GUdqJ$+p+W9h=tAiwrxzAz}>%>0bxFMcID_
z2ahNMJOvCR{VY6%3NZ|UERxyWiI27A`<i#*-_2b+Dw8W;lE0B`)2)`Qs)Ya@r3qH&
z<G)_|@R|zMKyp~fb@AQLA8tqU+3(K4kn8wBY03hLOh9D*{doRU5NW_dhV7e^zu1i!
zKJN-5^507~YdYsIAe6jiN;jCmGt6l^O2rxKrkZW8E8uzcE3rb3tWu)IGmaa>@s`!9
z+Z%^xsm~GO@wBnV$6>-g8fJx9xWG^-Sb&>@0ea74PoyWr7^+-YdB`~dKQ|Q*u|cDs
zMM8IBRQDGbhno?#!pPr$DaLT+m&JegMPlBn*jB2i=(Bv0t2XFUG^3pb<(ejh)Mkr7
zdfSCgb6-MGj%x{4s8OLr_7aOqJGFqOslFFeI-Xt@YahU6Fd*^asm;jtYTNB!yaoU{
zE?-mMUC_<>nQL>7i#0C3reQo^g`p<)M?OvnOh(Jwz24ai&*2*ef)Vb>EdkhuF1X}~
zU!OR!zQ{&z)5poB1gk>rOV`k`GMGu43=*rOk_RfyaKcvKxURspdtj-}$0rKh8(vPp
zC(Go>CJ?T?1r$d9$aP!9OYdYY3)ed9vMo+vZd1&rc9vYS#Y|6up>CteLslL|o{sJ^
z+T!7ngPkbDo8_-eHiq8kWrrC%u+k5f6j*X%tJz2~w+M!XtVzgO(G*=)g|0mxdUTeZ
z4=){MM><n&X|7pqVaiF)g6d6+zM=oIUI=8^<c)z)jUqF>ktm!u8V<!sQ5d?$#@Gl-
zX}sKX+2{@{ICS?X`wS&;!h6>j+cZ0X{+W}M&+ha(%9DlEb*BvVJBM?Gw(SMb%buK)
z<%%$?-pR46^1X7fnl8Dgc+hI01cGeuuG-F@j?6(KOa-V3SSsF0YhB9HSTV}dyN?8)
z4^|n!@J^71!YHKhGfk7*P+Y73Yso<j*UK_#NOG*b1Gz?*Yqxm1D*jW$(qo-7E*Owv
zk9w8$Y2%Wr!Z3ul+hnQOZ4_)k!kywTMVD&_d$*8g0Ai)gwO`2F8xdWRP+`oi3VUpB
z7@WNpkwH?^*g~lYh$i*g@I+-JIjNf3TnuJ2m3h0;FE)PlvS>nw_7-x%>*3@!-;wW)
zYl@e^=eK+@vgC8k1a8+)Q4i*5Jd?MY*20ysSIz8MYL^f|oMZ9sQILvIQG5%yWHgBD
zt?(GiMb)${eD}$xsm<q76s!fV>f$1!oyPSrSNMLb_bs2{VZLofsOWT-rp0omlI>^<
zRv~Muvb__JXl^(d0ih;2rsUDDQZ<Q-wAwN!=!Rt?)~f9@Dz==6&or=ZjkjpLV0#LR
z0=bY6_YWpS0|ADyKLoev$SjKTf~k6{Jvk^|K3}6=`s#CcRn$wPG3|D#{3s>X7C<jP
zyPbdLsv??bt%NrW;Cj=}!BH_oFDE4y(6W0#SS}&1Va6)Txz!8)j4b*Rhc>{g9(Ut=
zswRWo)_Cv6cdDBM#ShNgicQ%tIR`lt@DrLj{A#vg6J!b`r&I_QF1~BGKzvzSojJpg
zL6rsfa56Km?iZa4erp}Y&}!Z?EvpJ#MyB4-?sB!{$qcv!@T+B2D=x}i=5mSCco!cl
zV&}m}$^pggrYcEsGgw+msomV$o;awevwIPxg|2dcL{D34MqsSv(Y``!G31g=d5l>V
zjh*EkbHVJDEpx;NR0ltWBg~iDfm~q%M7MTIt8*3usLr&C0i65^$|)2LJzUO#c#Ic1
zFSg5dR4bg3zTiEQt86gqw7Psx&DC^l-qumAPl}+3J2Lm}+%uC9{d`tP1X$U?2SQ`3
z2D#mj<uNSM;Y~Ry4Gz`@4c%z)o4DbF#R~-D&<OMOleJ+|6Vb|sxtfAV0{K;dndU?Z
zxz1xGn%Npg69VMi3V}1pg;^5+nCnu7DRuFV1HhZ0&Yt(|$IOdMklIHo;=K~g3w_Hi
zIp|J+pFeF;-?lzeW&S#W8!9$S^DfB7`QoMJF__!O=6-(oes?U+)K#`OlV+tjuXKWr
z?j<aL>$`G3ol9lXl}9j;gGzkceUGazeb~MT=ElnlC#UAIn?o}M0B03dwJzTu=@=p%
zd+g>fy3MyZR3)_xS%O$zwX;pRH-ENS(UHP4;ajNQk5O}}12ltjOPU0KXeJJMms5HP
z9U^W0n!jTvc^D*#Dovg0C2@(OlJIPCPAM<u>M@!PY(R<Ut0IPtKL&~M`N{VbuZ)~)
zuti+k>uVdo*gfWWJybau$rWCjO*m@Oa`n9VEV&<`+RP5rQgnhp77Aq5(QvI#U~E8e
zdWl0FfZ?(Xi(#%k49yab8@JF{FZz^pxdyv(h+`}WC^oaRmn){Tr@1(r&CSb<4`Y}b
zr|s@zNOGrHRkL-BlD1vkf|msM$6um^b{ahjXDZUgizI`ag~V$jnan!%Qy#<2EJz0&
z#shDxD_m4$&jGG5QYGkpA!chg!Nb%RwtN_@gTX#V>g2l2KrS9n7|uH!y3S<IeG!O_
z@6B@%s=|#TUZKZU2>Z0FC#YC!Q;(`{6s5AyEOr0gT%$buyL+^)N?z$H2GmFPFYq=*
zs9=xtJ<B|7Bwap?*4uAuI~0`x($J#_d8M%hi}({`C6A{H3SL@)nG`Mwmeq7!X^)W+
zsBuOX?^dM(StX(YNR8$K7PHcN9}ylgU))Qm#*Ep1_t80sNlgWFPrr(o;R!h<%)eCF
zjf*$9KAyEM!?22ovwsmLPazrx@WQT$Q25XWwXkk2{usU(ITn-y&3_d1Rjcdkd<$+i
zHi#&84C|U}_B3XgP~zpD`zF;%Q9r{n?p|Tt^P}Z=TBP-2L4@MAO3c83i_5a8Fr=^6
z2JfGO9X7?Q6S00Y;5Ug>w-r&~ZN+Z+E`E@0iV!qZONdGS&{4|MU(y#xVKC>)svs+B
zExo8Ra}TBddZck=%TW&6WN6evsamwXvFjP$I(D|B)t0kUNdapZu+OEo>qAxUZsXT-
zgQzw04q7A|-Fxpv&<dNqr0N{<PJFawNg~YNB!r1ppaECyv%1h_JM8JJovE^W-vP=5
zopvLtEMZK;eGqnAU`TQm$|vkDnIqP}Z4Qql_1-`Q!PM{Zky+1(#gt!i=O2BP9!<=F
zYK_joF}E9Y1=Ph}fys)VOyyA&JkdtpruZCuu9D@3q8!o+-HjbU>31J*8cJdSnMeVD
z$lAihNwV(IRR@+}r68IImxoVt2bZD9{e){XtN9pRmtK<)+d+=MI{W~17^N$AJxx!s
zO~D_N)WDVoic5hzdwB-3Iejc(mCrG_J+jXtRn1k8Sl3L}f8#vCIL*-Lm3|^|UcEP#
z9MSFvx0Mk0jRLxM&YnsH2~WO)Z!t724^8lYf)L)~696nltUpsM=RrVm0ykGO@hWDs
zPSK-lx;2wh-%8kvMQ}<E3|YH?OIdfd{99(4>G6#%n&&T2HfKh)d5J8Kg~8mUGo%VM
z0H|ohr6L`d`Mb592~j=Ru@yi>q#Za){QP)0pWR#!MQD>+o`lTjrf*2aVdpvkH;?`r
zV}z3*fNTA}jE(0$lEujEUFBjO1E>qb10>z&lQQOZ>%%G5-5>86CfMPv5bmRe5djww
zTh<%IFXBQ-`DbWaAviyTmvpC{tFv8K=X|YNVP3^b0u+=Pd(Z>ctdX)rQtOenR8KZ-
zk6jQN-+UFBEqT;OOx$(=)oMMI(*@CVm3u2&=dYA93WV#DUAE{*R?O^-kdp)Vt+|0O
z1lHc8)bAfatZHEn2$8OV{Zu(0&KX~Jek}epFR#kjV`M2DMAQRa-qMjHpmYj(3GpBk
z<;99AR~Q}T1@0?zK%q2Z#sZ`MXvo9*#B9!16187kB(BC)2PykCWrLYFRna22&<qoV
zx~Bq<6#UfXkQGD+BS6Hm&iyoR^CUePe<L3jeNGkuZBsWS6uSe0k3X+m49-EQY5Ef}
zM;;wxPhNgq`8#mZ=z^+>Y}2vqRa$YzkUR><!;LT1^2LRmXpf-7S}4^8X*^Vo$(1vu
zvHmLe+R743raupsE`~l<sxO-1|APuVL-7(!kQ<U=<26X)?-C^^AB(e@N+HQ|rRL!Q
zh?j_#^t~<{@iZA?9SwIIt{?cqR`|szMnceXc#yj32hvXl{KVDGCtOmt$`NGs=6W*0
zWV>sjp=KI%U=E+X1fz65nQKBlPLv!@lNhQwiud}dyDHNvC}I}feHD+MNdbnZKJLM1
z*;`!_C=^~oQD}@+<4VWgAPr`1A@K^p_(LBuq@8)ezRBEr@#$tzfQ35v5_Hx)mMzQC
z&sZTtwVrYbg6AKRQ-Q!=#*Uj8TMo14Po1y=j|s#&F+&<dXe=lZhk)zF@8Dysbqc6!
zA3DpKH5UqW@9Dcyp-ZNLfgl7MqpViOmnG9ohHF?C|Ke?a?>>}&nW?e1uD9D%I*lj^
zyN_zaup(kZ47D@>as0KKj`uji20+AC6lF`<ZG6Pa*pq@ZJmdK6&AKm2AN(Ko-a0DE
zHGCVDkwyd+3_ut~qy!Zh5J4D=6d1Zg1%#oLlny&kX=#+sp*u}#EEpO=rBg}ibH5;N
zxBHwwzO~L-XRYu1{j(Q)oO$2pectD}pF6JWx{>#M7#qi$q((8SaXHOL=>CcWb06NT
z2Z%}ev{~nDljWX7MzXpi6(o{i997qPvYZ`2QAcG1&u+!NixLm=Dh((<$wce>lbUhp
z&7dB0egY7diWY`bErSDPbhVVGw@ZwffOfzv4|8UQ*|qpfqngtPAfHYQX>gN2aRM2F
zzOVCmM>zBbjMU9D4~N9gL3=Pb1_JMAdLwUcK~eYdVdEmx4I1a6EI&E6??~11RKUqy
z^it1EvLMAAz^7e8_buthPoT{Cx!Mm0=})A;_^|`l=TBNT#}B8Iu6=vseDfOwhq5E*
z45Yhx)sxPp88A_jk$(kq&Po{7%>LvB?(!U6mdU=|Xig=r@8DfWLcWrCPGHN?>S5+X
zbPE=bPWyHFT@9px@!|?lBZM-@<VV6@?+DdU^}|VbBQ+gxmN*q&KR*JGBn<2JsbTgR
zW^GOuG0RU0xOV14v0#1b2b$8g34VRr^5Mlv4U1p7Ycmu08%oeH!Tr11z?C153%0d>
zI2DpG5P;!7ry+YgOINQ2UMCamwfO6r3b?}*rAm>a0yP0b(l;q%a8+GU7&tH;oo1>y
z`}P!AQX2d8J~xL5WC`$+8)!d+c%4PeS(f)vWZiM5p?BfJZ|6uvQ};r|W(LAVgU+Y*
zR=A&GNyKWe+-Qz+0}<;?KmI43ya-exAZj~bQQpO)>!&ggHR5ZF?rM2Cz{BP8cnaoQ
zCRHbxf{gsqA(+Vc2bw9*S0REE<)58>sIZ8fO&U~47E0s(IdRw;u<281$%y)dJG~JQ
zBi2!rx>2v6GWHPX5zh+@EG>gWFt-gU35E-YpK>o0XmA_N+!HRO%7_dA&gs)8(Ej!q
zUbp{E{oNPhN6W@?3-^+jsur^Zllfk!g#|cW+OLWCx6NpV5@PujZ{IsUbx6f=w%Qdx
z8I&`ENrB~bPSpgccEWBy_ro*CpW!djvQTx5zW{H|2Hu*5aqRQpcNC?nX&+aB@v`q`
z8jp751z1ine7U1>WDNR6wl-c+K$~F={i)uaT&!imB_(nP30D~*T=jYIbewk?wy<*;
z4@D_$Ww?Y`Cx6}C#+{jDOvVdZMJG;Qzi8hm9fk<H!i6{%uAJKr%PIu6xaRft?mgj^
z*yn*>E3?|D?i7P<^}aA>G|mY%vMaqs6RP?l2}?{a?{|g>sct7{y!#kp9*-=|8ikQc
zk>@*y1ffbMs&r|!%g^wV5aP3d?cgU9tIuJVl;r~W$9ovBD(|j-WPAMDu^9=ri(BYd
znQkQ{zom7CV_=wv>5kC)R3W#u4gSV6drxx@KdL)zNwe?f4xXo2k%Q5G(cAYSc46bm
zO!>2can^459L|USeZIDZ+JWOcu+`wXoTM?@m^`>6AM=OW28EGi){t}lwi;RF{aADi
zJoJxp+h7)Cr0c`o81pT7EZV(VoWJj|KAu<VV6kIb{nh^ZJOQR&iR>lRPXD;Zmdz!B
z)KvfVITPHJ*F?vEbMza}%BTXb|GJTw!e5_1IRiJP*p|Ne7{5mPl@}KM`%#(Hzdl!k
zX-)FEi1{@$8`t*1O2S_I=Gx8k_vf^5Q>J4XGXFd~^4HyDcz#I&YvNy@%Yj1{iMmbN
znDt-R-h%qUg#O?^=oAkn%zsT7FC0ed)!Bbf?*Chpo)UZM6f!NowKR|;N86h{$Nn>b
zkGtYz#N1kM?)+z!tPidac?2r<G6>XukV9Ept-P^r9~rqCt%!BovHmL_8xMdrfT4f0
zPvqf7X&(NNcSWTMb~K-amGU0nDLTi|mdxc_apW5}>*K|8bW4)8?}sRu^FxX%rvr3<
z4F`@_#L=qd+!Is>w8<H%$!APb!AT#0lX!be&gMOkcjDI1BZksMX9FIJWd*4SZLY*x
z8K0zvc8eT=zusgY{EE~29uNAj&qZNX{m<9`|CpKCy!NLy|7^qOT*u7nnewiFc~MS_
zv1&O|=G|TTF?FkMuIar6fGjqG+Z5SaVmW9|j0nfH=x|PY)WS>u$_e{9*6O_c=7oI8
zeW8_e`?mcjMJ^gj*^RV5ewX>IUI=PL4mA{e`WzH$Dgl0{lii3yQEOCU9akeXD=}B@
zZW<LP4HJ;m6sC3NR*td&6@P&|sI?A8c~`j2UVz+;!{@3G63%d-=Hp3xgVuWl^h^=j
zq_Fp@hXbe?4MZ_Wzz+G3)i9xWCpmpV7^<MGV|=*ot1s#K&&Xg(1@!W3hYfZgHL^x9
zY&W_POpa}LX@4+-!~q|y2gUQI9h{Z7gV{FMHcuoLAR-U_x&?~5K2er|ugb5l`~0fO
zo<8vDni%_JzTC%AlGnqO=(D>#y8ru`F8G+tR|S)oP%MQU<Rrgz6sBxRv6kky)p?`g
zo~`(#VGTvj_#L*1Cw=Je+1yK~VFmxj`psDH{v&JvZ0`r;iiMl@0xM3QCeX%JEAZBC
z-`jKOMc7Lpt7L*1HwRp+fEI6C0e3sELt2n^SKpg{1yyBo1sH(7o5Q1jCf%@_q6~Ub
zi6K11)O?dLbkR`N8MZCHI)F%evpA)Cee)<l7x?&fBd{<8G|$XDwzDs(a=^oH3u=uS
z{h?osZ1ct`CcY3`YiYpH^=qm!qF7KAU~C1@hjoA3HC3jRV7Qam{VKy?8ea?WLu95)
zaeSM1ii<-5XoSX|ZgCWAPOVW6mi@KyCFS3b_LG1$xMApL;B{jM7OfA*f%WS)q@uHK
ziSFiL;VAd(+Pjvk>PTMC1J5htvHkZ66?_&gSR;7aR@_VC#9>qj60YxS4aoa##Bz7u
zb%T*dlKM>{&XC&)$0X{@Z-a&tK&jHh$kO}8o=l&MLxS(ud0baE2acXXsnWqebFT59
zCEIZqIg*bE?z)8(b=C)WpPMlRnc?fLPos9Eh{DM6WnZI?Zkk2LgSZJ~TKb-6L)OE7
zH#m6^F8^K+^*edAkxA99l8vHAslo~F|3Wiv+wTQ&j$C0sycL<g7wX%{`7S^z>4hLq
z@{P@<*vF4cL$2P#|9TZFy9#hdhin`DeRaOwJkH3~UsoAKokVfL)dk3e?);vYuROR6
z<kyd?=!5t9UI#zr&8VBI-sY3XC7^C?hbLEbTR5>7rO6M|S`ZeqOLp_gg?8~E3nDUH
zOdqz_D$LsgYv9(+2gm(DEl`$`moTsKzo0qD$O{uAXSkjA_t4L2qg#=im$n869@vB&
ziLX1^$jru3quP<5pD(61y|)Eb0PCqm;T7fP#B4!TVw2#VXA#~Ap{Rq^U}Azh*yuJF
zObVlK2eM#jOFl$VqK?C2_8-0#-?sT{Yzc)L1zfzpQizR%jC|zmAN~S13q|V;>eQ0L
zP##<d)<G&Q9MHw@8o?;AU3;}737p>&bD|eq4f_tBTD<AC%iEJqBW*9FfwKQbAZ*$?
z^oXw7yO$2uzowGMry7tQHaEKO?c(m}V(Ob4YtP#v9z+3i^`rtDU3FvIb7=T-sp2gJ
z_{6(AI4=YXR>V+pKmi`GIifj*iQhLyu2EJEFnTLJaK1}jCXL3Kq998VtO^hR@F&|}
z{!QF>)HWH->+1HuSL()I;)#b9yzhs&$mWas<pz8KOGRg=@Mv?6el0OGWS?_N-TBJ}
z<3Tp?7p1x#n=`(#sNx>Ne3-o&{P5RRZta1ql=F%o{9~fw2LI1P2Y#3sm+a<B+*onG
zLa+!zcf_9k%kO0}f(iJZ)c4!|Z5U#l3@kaJuMQS}UF8KB!QG#O**15{uP6AARrz(5
z>u{A9k%63lP1in{u7<nspZ#?eTErIre>drnGQAS+{>ZL!At@Xh%S`Oc=0;JQ9?Ax=
zR>b!GCr$GOSP_G)y6ZaBFNwgR-&b%+m7SbV+2jVj=rOCz0<G84d$^%oA3*IWy)vWT
zl&a1xY}Jx6f*U-t<E~FNR1&Qbthp8-Td9L@A4KM$fDWmynfF%vny)S|6xcQHhCmEt
zFQO$Pk!l{=HGl?y4R_Bq_3z-6VeAW?N^>|uzPcO9;1FvAICGGdaf$OPfF&6F-VESX
zKPc?DKAafV*qo})-|u+hG2hO+=i-(C@u;;!Jo!G>ksLaUvPHkZ?L@4=ZePG?HRn=r
z0QBHEIjC4<UNTjTp=Q!V@AS1*egCBKulY|?K=1TKJmY<KNUvxp@yDeJQgT1Q9;X0o
zhsk#Fz9%=&!kxLloU()c05#V`YeYOLEu;jH?avKN>u+Pn$c?;LoE$)L=xtUP(_UXw
zUpewp!0IXshP6XA^8z4ntun5_-t9Yl%<gkJiv&cQ32#nnA#<h5EVmj-0`D7P#GEfE
z4yN<fp>2$o5DV>3`ape%t~o<N>9gPbU%P|)l^hsunLQ7l|Fu7w1z|On?a<#?iof^H
z|9qYLe|ctbRdY&hsNdEKy~`JJVUb|k+m%L}Q!kISzMs%)=&1sCF$oH<;cTLgks!L#
za0D_5UWLtSid@#<N}pMhYQ5Z7=6&Yk!I{I+x;O93t3qaBD;eHV7<*~grk#OnBN{OI
z*6p8G5!o0SUv<%8Xe2{BC*2>cp4(pVrqHm=nSLrl+S}{<0`z6_o8^wWdL~<{=@#VE
zj<&6(#6PNcEXpWcf;<utM62>Tk{MYibKbJol&paK#I4#uN!S_}5L4?nnEh)p;apL)
z!3K+vk7|T^?C9r~#Zr?R)HDFfC_86C0Z-?hSq<rkm9chSWO)WlpLI~Iw%}5-M9el2
zYBDW^os}{6sXv@}<+Wsyb$t!Eq))2A#IN4&`@QURd<Td??F9nvhwGqDdZWK%x6EU9
z8zwgXOek@iLn$oPChY#wG9tQyfLIVy1DVT2#P(N+%l<ltd}()|05-B;o;zfHh~=)+
ze)Ge-C#u#~d%ppGtUBef-ElR*7B>Ra*34TOjMODIKS|i-zPVtxLsxd<w)~psltW~*
z{)U!`A%5yC7`#Sy6#fguG>St(Y1ks2RLFOl0|z)?8YD1ea<-qXglWR+QslDIW@nu5
z@7>&iP$m`mGzFOPE1<V>QojvpRNK6}Sn6l*jhcI@!+8<dCCicK)Bz~z3B`I$-3vc{
z`F7^GAyf}9EX|E8g$Wv70BH3xu}Lu9H9vxyKlR=WB*ocES0Kc)HS+jzei})mf{0eT
zHE=p~G0E?e8b=}}Y`acOLz<7uEL_U7^cn*VoaVNHvX5F!RH4u<f)tln$`wv89geiV
zbkcFigesU;sQ&U7z+e<nB2}lY4^l-97-+x`h<=jJU!^s&I6w4}^NYipKP3;Vu=zv7
z0?1BBLea2g?Wq}H%$~rT^hMV}ws*q^xQe1;ziz@fR0&^`!4%S7P)~%c^3k6ukV&qY
z%C~H41o161BM_Z2F`Rk<x?{FW1<n&);tTN~=7YFZL;*5a6gv*fYyi6G%C#wz_4$x(
zZG^71q1i^A1r{G)M0rfr(ToQ5^r;O%zL$?Ds6i!Ouq+^~$8~Uj9!Or7JZ=QWl?_5r
zk5h5@UbD8Fl7ZbClH}QaB;5*xpWX-#Q}oq`K$+~6+z!C{_{+t;`ZejoBucj@*#I%W
z&29&k!<xcV?sJDe&}&}eLM_k%Al&<`{crbnCwU8WbB7}?*~GB)bSkJi8BRl@vD1cG
z4uG5QoEJcgM}*dB1+9I09P+ewNGm0@9mPGCXRhF&)s!0S0Y(CM9%yzQCNDV|YIT1F
zDo+uPP3--8=6jAM@f8g$ISoT9MKt1e9er2g?9^Nn2$EPnlAMzB)@*42J(CMpw;Ar~
zS%#%0gfx~o=BWX1Zwzyl>W9@(_sRK@`PZLadOW-FkQ|JwMy;)B$5-zQ+P~>(;D{(M
z2Q!SZatH$V>gB|^W@!^*JiTFAlaG?l=_=hL<<A3C41;fR2S_(1S&Ar|sI^YEXBHiV
z>?I4pNf?`;9Rocg_D%ym;uFnXsaOfa`hrIMnal{&{qGZ@_Su!y<VIA=NEje8NPhR}
ze8m%|`CWju0(=RPpJM^C>^YrO{PsE@&QiF-4qckNcvJdde%5O^)^8YSKA8Lj+Nq((
z5`**!R5UDx`tl46%R5TDMav;(zX+i%-8VVG-y6*jOE9YN)5Bu+XmND*<%2Q();d?Z
zBN8Ct+TX;kXwhG?&rHM{Y32<FMr<1l(g+0`daCx!+p(5R{bP&MU|x352+)g8?Y?hl
zj$x1jfn7gN1m`ukOybs(C*bGCBlcKD7q)tp9HMxR6!S7qx2~pXY8?ZTNGoMK?{}w#
zbV8;(W1QX`&91}boh7#bM$jw};k`N?Ti94byBka@z>i8P`VR?jTJqj16bl*;`3>q5
z%Mo%#YQDi%eq~S-X(tPrY5)5A0rC7ELsR7wU&tL`l9R@I%7&97OEVe)P`TsCU~Vtf
zSC*^1*6hag3W*Ud*&1U)qZfTCqt(>tuck>AvryO5XD5zhu4@YpLqsoz4~&XxCg{G^
zR>P*(WCx4v5AI<S&)D5#S6B`+-?{RW3d!t!?~yfmT^DJv>Gt@h?H=kZvctVh>5wGB
zG<{nvDC}tAqR{bbx>96RaEG>wsXBEygYtsC<;UmeZI35w@5#0t-0#wL+>AX<z7qH$
zRvt!x*G6koY31SlsiXNYHXR~A<(5)u><f8GwRLIYEoW(3U+FIt*>tv;6MARCMa*wt
zIlhPx=;&EQvM0MSO{^4O1@{(lOfjISnGC=g?@ciM?TjyLW1Bu&%ot+$wZ>y*gq&|^
z5RvkWJ1G=@IK9R{FbJhjQL7@gkU)G%p<UO>{I!B(s8lNF;zH>_$;DVp!<_oD)<*mm
z0!XnX@SS<#=3RA@0q>dzUB#2SUd;cPZe^0izsMkarJ+vnBH~|U^|cNxVybu<UQodi
z7PdtxZ2Ifve}ef6pHY|e%;nW$ucj6_wD_N<z3E3Zu-e4+5rn#gk18;Pdapx)<!a>M
z%1MWTMl^#q-M)<PyR~#KQcN0ty>z@GgEzunzT}b2I3)$0L{d+}0@UwS+RkmgU(%QL
zBaln+p?=GiaLQpp4%pUccW@5!#4a;zE-Cb7G73z@W6fc(L`6hhOtf;Ec|(USuwg{^
zstIH(;sw*mo@tyd|0u)S-ipw0Ed*Q6BtVjmqJ0^bNsaxKAO87EwVZ~r*6filzU*5}
z1DB`kP2G$97ul9TW_{L;MEzbc6ObdeA1l)+TZvWaJ3<HFMO%sB3(7)SEXBKD`rT~2
z4q>=BoTmd9-u@2wu;VDCqhUKZ`~FE$stzp(&1AcILQP1p27f1{e{Mc(o^K)brVTg)
zd|0$rscmRd3{X3?S|uO7Z!qf${v=-O&Uu;p!Jy<~YMa@dY&66mu}bM$nJ?~NY~?Ku
zmd!eyC8TN+!S<?j>mo6vRrkvNqK-;DqY&n7D@NT7NP|SsAu|b^Rnh4g?LCWZNE4KZ
zq2OMwe`+jWP+rq!q^VOt2dcTfx}0vCIB?5J{>H)7!(vA)o6W<f_CKuW@KNAA_dZs-
z$1M8$$or*HZ`_cORVk-*CQN6I{;cj?q#hN=az&i^3tExc?^p(!WnRzaalHN<A)qP~
zWg@mGtK|+wm)9`~5E?RZ50O<Dg_?HUPp}QI#RjN1-K~BU-g%7j8c6OZ`}HcuQ>7P*
zi3?QBe5&E|&teM&_TcTC1Me4<k0zLO_$y~b3_=sySCsy%Ifx9#E>WexLf&)Xnye#%
z#`>8~2+VN5gtH?@r?}&uBRY)bE_Be^GJbz?Lq!h|g1Rng5h9Li-!NK`FL<?TP#J+w
zah)DmJ?yO8BwWH`fzfus^hMa#6PNTab3+({Jxu{1N}RF~f|yu8@bR=Mgyf`{5(pEm
zr_ID%vQV*a{8?Muo&hVN<|O8t6vD2GDo;BVC1M$9vpN>>rj)*EHr=VSdCm<^78VjZ
z-2+iXwyEr_^7fX99RsJe5lRti=BjsvmYPthEi-Tq7Ov<b5;3f}?d+0m(7bGW;Wm)!
z69cWyyn>JAg%MOb3_C<V<tTQjFzm9I7it6Koy{uXR*#KDX566z$7frvgD2qz^JhQG
zM1{Fa69x=-&y(Lj0wIv8N$&5g)1h9Jq47(*Ph*5ug96(Kmdq-sS&Tid@((QoCo1`f
z8ZOCIuWDg{_1k=b(&x&;dD?`@;&0m#!P2r+KN;z(%?c!q5uLhJMzhZl6<eE(u#*>e
zTqgDLZPCum8c00_KKj@f>D8Z?g|GHj_dRc$$5)QER^GM^Hc!O&k}fRx(FvyHg!m1D
ziUD6{0p5KDfbMr~03nW<x&%=nlPnZWUSH#addr3<f<A2muAi`%bQ>VePnh-SleX~p
zYS`d))OLLX_a+jY)=r*k$GXY9^nhr?DOrZO$ajkkww(}Qca}3h<w^z!+*8$f8H)53
z4?0zIC94@=CQ%a0yM32yV%J@x)oT29_E4rcIJKL*AVfag&2;+WQcJ4(+bk_(JrYkk
zjlKP+-isqHW#Hv;dk23vVqjYqT5Q_RWnIe(?%{n%K}n1q<yCm6YnrxwsctK)#kne=
zjhF|}MZ!^jl1ypoiKLJzfQAu9eu%A@6tc1oa%Mw9e;=`FM6I*Xh6o(ZuFm_OB<dHm
zGr+;OLioR(mbYF+NbF(37Trbg5O0luO7$?}@D(>Zt-C^_dZT4lA(~pp0gQ2JXgKwK
zhBZPv*WZ=VvxDNP*B18D%e(nCDV4q@NY##a6wuk=WB4RawOx(3Qi0NcziUEA727E0
zBt#+{F;^|R`G;bBZ(^eWKg^b#eqQ(20U0+!-fX8y7V4!Cu2Ghn)OazTwpHuVnP72J
zv(##iCTC`cjo97LHQi^Sk+Ae}yJ9UX0S{`#*oV*hrcvXTw2%Ytm?<~FO2^9E{>ZTT
z=A+U+N+C2Ho9%%0H3_GHdGtr99M=YTHhrEgDCQ%_+~+_SUa+j~aC25kuCg}fhi$?T
zt(2!SJ=6;!M?MjVLvOB>ExyGxDfjYZEL$EcvNYq{vM){u5Fpud2H@*qjFD%5=h~??
zl6+(yIy=x<Egm3EJOce9i+me^><1h9uylWlbU63V6=|8S1Lv{PkAm)Dk~3{&X(~!3
zT>T6>7pz%8g$UEO&5!_#wnc?2*Dw`LKQE_#r37pETG^%GJ-`2X-ISm+A&}5xm-;KE
zcsCM0?#kNRe}2>8Y%F~_NEt_l-eK5uR(@UV>tS5Z4OIby;txg-AZ>bV#PF3jKY}hx
zx)6Rz3bImpFKGFnLQU!=0)R|{rN85-`0HQB%-4e)De&xuN>Sbip4X*z4Tt;7yhYz$
z(uaC+81{V2kTQr0Zp7Z;bN?tkxZ~(J2=$tP@<NOYZ`IO-#5A;CY_Nt_BIXDV0lpL|
z4iF|H0w6r;GGV>dup$=3gg(EmKXe1IEfEkZp@Wnfa-pT3_HL%_>t!IEz3-WBs$xHe
zeL{zduOPiTx<?)wBYi$h)_A-hU-39&4Gs{+u9RA=OI;ur*^`)=fT=w{A<9sK7BNcD
z<f;(?v1;FbEID=e<5p_VXMiER!aM!x(5S>jaeN67=`S27AT&xl2+ge=zWIwK9Z?jQ
zzU#J7Z}~E3eL@vDakWcZs$OM@YL70Pc}F@ntgf7fkiH@EJtQSQqnJA##XpB}xg}PC
zmxuJho)T<BfRIeZ3#p1>f=Ff=(HIbxUR%y{1$vvm-^DN_2o`o4(QttO@1$w#UOZC{
z^*Vx#nVKF1gLDCAr>9{Ev~Ibkw~_oXqb@QH=jO<H_d@kA3Rb~u!%ssoACRzsv9A#L
zpUX=<Yoe^S#TWh07=v8S04RN6fK6T)3(#+ef&2Vf#}IXAe;eI(d<Uc&NW51;#WZri
zb6ZV}2lx(YyL|Ixw)o<F?^~C$xg}*BN%z4YMw&;)Z3D+X3}Hf_xnW?_T<+>145jlX
zf^n5etW{RYo0d6Bew|;jAb80*t0VsN&tf4xQkf=j!WGIuOeys&-3ak^ZlM`12f`#d
zh~PydNFWdqJz;RU%vuq%o}`|v%qzR5F5Qb1LMe>^hn&5lfaKt3M6Mx79AX4UF?e;U
z8^J<LEq^~-<^{?@#}TTlD=e_FoT{H+>5!l|9PIN`0C*)!W;M*faVm_YarbA)6WtJ{
zA^iy;`W!L)J~T%fovd23?Tf34fzIlzOZ|wM4Ucy2sLOziR80~<vqb<(s*JPs{EqN<
zt2X(&ZKB;T5Sc8vw3V2uF0>jl9cfWoi#i``B_Uo`Sw14`4hS!*I}xHbddE&^fovoZ
zUcg?DfulSrtAO$umwSij%S_Pw3QTA6NXvH2dxRD>-n)CjDDP6(Q65RUb;Fjk`|A99
z`-1Uz!eHE{Aw3YGZRGm;9s9E!0Y~GnR1PCz>F=G6KHKrYi6a*d=Lq1h^CsRhPsHsx
z;8`^++cOPu!?98-Ty_fCftBC}PoFnadm(GF09SfAp)C^Wr)Ou`D<2|ks1~aq{kT5G
zovE93xDup?3v=gB)-B=RP<B)CmQ30f`3n;Vy;NLNCkaz(;ig&BA6Ghnf_!;A#%s~i
zUCyb7)+%wJ{v|&cMlJM;apb4a#6z-89O($1k$n{DQBCVcxoz*0tJY%pSig^uvXT0t
zc%F6SaTn6&8<8izxzfc-VUI-9!`)>?`ew!)Ank&rhCBO_Opz$ji7cw>8HxxRrvm*P
ziWL1L+AM{HDOG`#h}EQsm@D8Y4}fO57=ZO2c<YpTK$sW-G>`WdOs~9`zj2sD3S$70
zd@60qHA!AbZy_Wr#O)C}g2;{U@klsamx}si62B&pvJIqtlElB><s4(7NT%C;<QP4x
zX#ONZmdKouX#k9B_Gr-BM~im~i&HMfcM-ysXYH3lO>`DnQLtk-Y-i)3NesX9f<e3M
zx|P05@bR{LKKUKs+Yz9_EdJ%Dqj4PK#dVPFpFK5J6l~8rx(Z!+Y>xqS%pu)Rn<92`
zRmEv7j-DAeYINS0pe8y`mUd$AiK}t0K>Z)DtXt#h)mRYuV5+l-FcDWY(^i}&gO_~_
zx9X3?oF4{-3QTE{`jnDGMUT2v8u`alcm3x25~dOP-m8;w<1;=nkSqMiC}*6w+h1sR
zCKC%uOuTlMAxV0oE;gNJ0;&nIjD29AN|PXITa!hL#yCCli_)-!<9d6gMSc}N)KG0p
zH{c3yUM_wgJP6-p8G}~wXVZH-xNPDhbcrgFQO)mXp(~agwLNq8qWtMpKP*Y_C?Hu^
zWoa6a-u5q+H0&V78ODf9YquAbXJ9i3%*0=(3;1+ObNkYfR;y4e-Iw*fQwsCpowwKB
zJKvu)LlTlO%=c$%g4ws+p(yE4?g;_`@8=&~2(e6P{=0KG47#O<*k7rn9gqS_&<)M-
z&r}v2SWWhyr_&~smOIS$>oZfN?W0#b$?d(?XESpuL(8l^$zWpMaa%!D(^g06TGB$X
zyI}NVJ}Nn|w@SWg+a0^PK5`fmE7G|G+?-Ul&X)KACt7RzZK`-$ATWX7E+X>DU_mNv
zE+rg3E7%PU=53G9XnONtut9Si(q82@FUBgb&YX7wMWjPx$y1hT`@@)yrMW)AUneaO
z!MkHX*<$)t9WF|zKAcERZb{ELDMqU2<B+xsQS|i!@$bwAXwk6W)CnSk4ktdB!}4Z7
zg4PlT2`U93>vTEZyiPE6pXPx;Nu=cqn!oj-;k&wmAGYz<gz(OzVXX%BVGY;XAejM&
zlp6}Knc%+NlY5wSD!9d6yDVKxXOUff&mm8-vIH9gZMGui8`7l-OKhHNqpPQT`WKBf
zl?aM>p$zjT1(?W`*|5>)36riAW2Iu@LT2|Xq*bRchkc@$T8z;oX6P9YOEw?TAuCZ_
zbuD%ZuQYe4N}e$D5+@y6fDCnRe_(XEVRhv2kM`|k<O=+_S>G%?9qE<qia_v8THRYn
zs&E+4tXy=c^A5ZYP?QKnlHSvFt1+hTn)gqplmxnZq(4|^)l0KD=|0+Hh6>wrk3tbC
z?r<S6)Cvh{qjUzz%KJNAWR0mOe&W4Gsq9axgymNy(O&iR41OHJy5L(HLCt({_`y4_
zsx^XsSY0}0vVEOrK<Rv<2XQVUB-eVQoE;}b4q{XitYelVr?@K8Frl$C$CjaunXud5
z3_b0Fxhak(OPC^156@mPET}?>Vw9YF*7TC4N=J4kwwMZ*96#T+<VZLf8p(CKGnMWs
z{*05|*9n#7JCmi-Q~WzXEMfZ{)U|n>!HTV6(NUY^YiI4{+doeZe}1oNopNL(esr-1
zzio$Wy8AA`t3OAkX##S?n%_p60Y+&J{zJ_oI7yTZ6AT2clzVpe=yabSkh;adwv5@|
zrf`D?2wRgsVr@%pv~<lv;s|$Ih8wE6Rsx)cL3GLK0S!SwzcAlZvAZ70HVBT?B`g+|
z_Id9sb#fy$U8I+JmK;;*TcuRk;AIjqYEXbc$*=0ne&Lm^?~e$Hk7C=kzK%YlSgxF%
zKj<~ZQEjyXt;+hnt-~-jv5KoMgb8{?k6;O_k0GQ|-k2#<wfs<jA&8@oPiJoWnk$c-
zhDnx>aK>1NvZu?eVPupDb(msWKatDKy=qX5&~nxXs-ev%GAOsqa%^vI^I-;WT+26&
z^&y|GK4;%8E@>=@`j+7E<7}(1>6NUDoXOQV&r7p!b+^j>RI={!sv|&f=RO74Vj0PD
z(6||M$1JGE2E*o*N&%Wn8rQ6q!j0%PqYrm$j+>m{)B?CAXnD?pi#{%K;Df@5%2<p-
z2lHp?w2yOJ@ssgUdsdD%LvYHO?9qIrflK?r_~)<@OsADrn*B94#OwF1-$}jnXwJ%n
z^st@2ekisC?_&7L<bt53NowD5*_rS+`&H);cB`c9%VTdY>+4Z0c1krOHMb{fY*iX|
zICViKOpSOWUGHMIeu*BwTQOz#$;jaGmCuluVKki_7^c^{P6xi1gTMLL&u@=FY`Bji
z?c@i|aqBecU@!XT-kvkTYB#VJ!KcDz`r|lG324T!4C6Q@_=l824Ah6S->;zcF)tM$
z7F21=2dNR7N%e|r#<UTlbzIWkMOjY?Zh>Qg{EyvZhtqq=J2`&xH?7d+7#6l>Fupys
zo<Z{++4;fq=9*7ibcNrku;xYnBe%u}?UJcJk4U7G#TszvtW6t)t>|gBhMfp2?>fPC
zacOZdsYWPIzNBliWa;RB?(TW-()YQ%oSiWCFBEB|!}e(=qT%8T3?uT~(WX=f<ZNXU
zR%YI=#<))$cYEOisrExxn9EgMtk-xm2R^D#Z6%?Dw(aina>I`X4pa%4>49YYjhA(A
zNk^#ot#omDR?=8Or?`pxVdEHwr5_uRF>DfQxYA09$Z0L1B#^M0g%3%>wlXLeRQNSI
z`kL#+*hUZ3JMm`RBP?YUYFEjW7TV-3NUE!^dA;X=blnB2)l>SSW=MR)Q;F4V9U6F<
zewunF{&39*Rr+Z<5&K+@SJ4nX6yD3G43Se_b*_uGyz3R#<&o!=nSu{#qWEZQTu~}r
zy}F90G^|e;=uW%kv~p_(UCr^jq%e1AI8Twqg1=EcHW_aog0E;0ZfdnA=-OAN)(7A{
zr&h~1@-FB-WV2$V<}-rBWg6WHj?y1Xovb=hN(=#*A@#K5*sjWp1eT}j-6g?YTj#RJ
zzve67xKc`%=eQP_rXSH$z^6PYtGgqG5|9*ZS-AtrBo5J)9nBmr^c4h-**mJ$>F9T-
zcu8Au7k%3SQ4lay>ouOpN@}0ff1dm>buouWo#Zp}Kt?IjK%HZaA%Ct_U9oCku<=u*
zeDSU-IYNhM88*2~E9A8LffCi&+4E^$Z^O4;NbH3K1Yz6#Wb-eA*IN}_y!4bYS7n}6
zI_6cFCkn<slKq7X*X;$@f<eo&nvaO?WlumdgS``m`R_PlU-cZe*7fu;RA5hN%4|(H
z55kEZr63`NoCGTSPo80Pj-<xUs>f6B>--zX2$uoA`}_coq!i{*Le6qe$9oJQb&)11
zUSW@A<eJgjB2v1!R<3PRgl^H)`Zil}MB+dotLN2Q%v#h*^lIP3B&B0IjR;WIc;tC9
z9jmr-FGOV4XH+bH1w*f@qh^zvbVAK1%Us!?1z7gssA20Sz#@d=yfAA<*(RqutR30>
z1BbkS@KMhP^}Ow{Gw-weCKEH)dP`splCT%kZ#cCZ36Ja4Pf<{q{X+GrRL9O6ak7Kc
zLTkM;+2EMzA2_+S1&n*IMMS_s%}*=0-iUx8mTV&wQ-bk%av81Ic)=2?*@+5w&7xy#
zF~rP4X5^6}#_xCfzC^Hi*OlGZigq4rJ8r!T@nbt(2Lh#$iWDO@-7;Wtv9#~;(-iZM
zc+u#n847!#+f$`u9#$!`pc@h)1$6LPc!B<;%3eZ=vtcHIBfJ08jIN5^N;#r#|DMQ^
z`Z44*frUcN;k%`vR2afR{-zQgH|rH-jTuyqag%EjPSrZ8Ic0Un$;fLhs?XED#60nc
zvAwJ7<8>gD5BaiG60ile=$R<8+|ZemMqz>a@`+^(xbX~15W<<CfSQ(}Xid6#!1uC)
z!|a^(!ynEv(7%%>uAk~R1=VQnG2>6xj!D%Io^5ALk%5%V6;~4CSI483<O`nTlQQlR
znLji_Z3Ch54ZxXAXRkDW$6$%?gSvwOg^t@N*H%LpilsEsCJyZBpx0LR_9TO}s#=&R
zJ7w{HCW+#ks~@BQKt`j!m6vDc&MWDuTF!B-E}kXx8!&@N6NFW9K+Fn5V)*h^TiV=$
z3PrccKtEH@eR#vRa!1>;K5fjT7JbpQY7m;_W93zjG*cP6_Z@;4%R#hPe~9)@XKh_5
zwDjF{a?67NfxMtaXb7Sjo86fmqfioi3K0#DNSw<)D0{y<ThO)m(V~UbI@|Z7#+lV7
zgCb8b!_EB;{eAG#k@OCC-076A@@zMrjLB2o?kQG*mHbX?&&*$YDP&z!a^Z+0OL=$V
z+rwV>MW|X7kr7ZP-EVntGyGIDz2M`vOM8`sk)A8f?>~7o5$eV*h0f{}x0A#^6(32}
z3k)e6LtuHUNcyPX&JTsAAL@-!NPF{<;53t`PM_pg>nLB7=oJ|B=vneTFSYS+(9y+o
zb(Y}WE}cc&s791ej<|y1u*c-q%c@)k46NHyvzBiz%80`D;xO=e6o($?LRTP)$l+q!
z7-OZ?394^3GQ7J*1cEMTNQ?$nn$8CCS=zgMw*MUbT1eW}$NA{Q!M#z6WFGbcKR_-M
z|Gt|(_{*FVf!3nC%a`Sznl!D?OO)Yw5D~*KyU^wO%j)}(&+K1(wM&J(;4va*y(q8D
zqH>maKg7s$mUQ#6RT%?<f@n;T5Ay3UC=i_1e_GNAii;1TD9i1a!lpC@hBbD4DfPY8
zeK|#~=G3|GRuzi6%8zbsL483nw>#>3ni;C~Xh<*A1WG>%fZX?rxzuW|RFk^Azrs_z
zl=jL;ci`9C1nsur2o<Rg#NP=DT3j1R{Q~KMvs)!7jA>snQ7@@z8%pUVbEl3Nr+Zb@
z)r}L5=4B^zXr4;c$h#;#Y?)mj?9te_;JM<dCNL_WRmr5G+}h0Rs=r@$<(g(*x}Mcq
zDVLoDqM>4<X>U}27L(?t7H!+hPq9{aI<v>4VgJ!qa7(62?5Ea`LB^wGYZ8K_ihJHi
zzaCw>GSaj5vvX)kwdWT8JacFv(1j(Q4uD((n`B?hd>rdT`K4xOjyU_#)$^WOFDY$5
zx#Y6H1Hb2+a~Grb_Nu7W+@-FH#leB2OG`gTg^z1~UR_<ymB@;4v@&4ob{VcI34EtA
zuUguwW`zk;IOQ2HAqa_7MnT-*B)4Wev2=foijfs*Mw@fXqP&vM%4*Q+6;-{KAwLe!
znE+K|dQ638UTd~=FkhbEW&aZy3B3xwS(M(_A>gsjyE0Dr7_9DD*S<e!_2Um|B;TVI
zf5__V)P-0cd>oBufG0E%O#%*|Gm~XSlH$nQc+Y&oV)uZsam=H54-vnDta(JcY=-NZ
z`CAs7lOdN&r&9L?#1!M#L~T0ut|=bv_FBB46iZTZfed9H!|{B(X9V464ceXJ3$f|0
zrP3yMrJD8>^%#8ai2-N_7L7--bhmPlRoQeFnnV1h%5R>~5P$6))UGa0N-$+xQr@$!
zv=gpiz>OaD<ai$~(P}|{c|U*Tw_w(f^u?O`;hHpY0SO@*J3du;ui5W8HTiAyncP%w
zY^nB#Q|rnC0IGPOo@k;XU2>Rog1hywd62f$TWxG+^~r9H^L1t)IhH#jE^|)?O;prB
zc`trwFcaPGFBrY7Yco0fN}@#TgmtV(mM(Ao9xd~hjr=tlff04a8eoKAQ1n^mDR-?t
zU7*9OCS{q7Bqb_@_a(G*E~iF{c38ysDP0HiZN`*Ro77-7)At9e^!T2u0&&gvl{f{V
z&{y?lU#eB$8wpi4lb&a1@33@c1JJ9U8|!boFeVQ0;9u}LEIEb9=lgFiee8`M=}4Go
zd7>CEnBE!wMi@!c`2|{46FMTqsaW$M3&{m!JOLVEDuf@nk;}gOz~>EmAGEX?2I)+J
z+(kAkH+iww3Eu2mskY|}U-~(!16j%`1fukvB_;NbjHt#Z6ZwQ8x~s(J-u^rY$rLX8
zn(>2N6QmOhnrI(^^27Ew!;j*RpF)z(R2H8eG6UK7ZVV(*X$bdZbhk&UrQAH<i_@jO
zO3Wp{?65HoJ~=GeI0*P%@pjZpYq#ktK28kgA7XLf3qj!7C9R>tzl(m1JaImh(FXyV
zElIxDNu0g)p4wmVLQ??JM;>5`{HLz4K2xvG0vhpu>FYHqU#~^;aQ!pWlqe+qiEvH2
z?Lqmr-h@FER2dYQKoxT*fo~Auv##GtBNf}4WvJ8;Dlq<jd0{$T!?y{5#wbHCs*PDk
z^0$04CIDTiDJhQVHlctNBn|_$3r&Yqan)b1W+TZPcM=dM>EBK|gl@j?^_xI|#yF%!
zOdL4(uPTJ^2%|}J|48w>sLfHI)4;CCfySoIG_(tPS{KC4>d(@Zq{Nzsw4Xr;BF&kR
zI!8g$!%-N@wY}B|Wb1{@o>c(YA!3RaMd`*>mBXOxm;G4(=}D)w1<0j6EQ|)c7yytS
zLG1f4#8_8S%_yXtFCtmyP<>nRJV2F%8+)&ike*Wmkfjy{s(PVW%OXO}X}dDfRWg8R
zLQ5lwL|y2)Km@&2Yv}Cu&4jh6cgh&*YK@43Y316gJ0cyA2-gbReU{9F)N#9KXMxqA
zbKN3F-(hVFk~%~hSQY{}xWp3Vbu0n=+75-#3m|OJD82fF!P^1oDu1I3eX2q82Sk0K
zynE=jrQ=xJ$1Y&qm_sI%K}svG4Qxm63N(+kUO)3IcIOi-mnN)0JCwi73mOGppmt>D
ziG_~Y1Av5+Z+q_@f7u8ErzGH_t>PV^?M8c?dJUpmX|XsneEIR9C<`>rHq;2$5Tz)F
z(DUjG(GC;Pq#wv{QWi$~1$UAvOPBTYEiOJLOv@=F=~>XbHEcxO0N~OEE??`IFL_$F
z4Jr=jK_lCTW$>#M%<U`@us$43g*N&{_aS}96NLTB`9g@<@8$1pK|1^$q41;CFwX=t
z^FJxbki{rM^UDZOW&uyHn=cPu4mbH7MD8CjdNfk4FF;s+CwU2YFoh+6Nb!3Q??=(y
zK)O~FmyWi9LD3hgKk0(fv%Vs0#>{RgZZ(0-*nCdOSZ;$gE(5La>!MQ%RryA!I55uz
zEPyj;lRY7YbR+S9_nM28#y1J_uHBhwvlwj$vy3}i>rUI|wu6j^wN<31BJhTNrw~kh
zYi#^4OgxsEHf1;iFKoCa#k@F8Ktug2A}%3nre%{ASSSv+_m|dugjXPyw<RK)yTz7H
z0KN2@ZoB*#AbD{7M&U+X5qw!J@8ecu?{0P#3<1B}6TtvbbO}Prxly3yx@LRuTSVey
zt2)JX0BLz<BPqr!VSrt=>lD#fiUFmfSjOz624jhI*J;QWT!vs;Tt7(i(yO~@bXW);
zCoI1Z0qAVWVrbD;5T=@E7Csym3u+H}fa{}teYK!t1C7-O`<#n=(C62F#*L#8j2gXQ
ze_V$K#eG(reAVjbPDlam+6s<zXCAigHYoxi<mgQd=`&DL7)~vlaY#k&w{`)h+Eb*u
za-Fc6kGSmLS0O!|yJB{O+56}ic!UE8I#+99#rH<Ynm%rP8o?$aE=jysH57M*!GDu)
zyRhP<DLT8N$zei~g>(bHVZ(GRNrARF_WD;Eq`E^9nNl|edvt$!6{yGi^*honM5I5?
zRur=BD$emtw|Z9DjKJlW+8(ls3F??b5cW5*O!LTh@Rg~q5GpfRuJelXUX=t`#tjla
zZIqk7-Vrd7l<D~h7L2Ok5z7ToQFufP3lLCfHV$8RDW9l`0`(!$pBLJCz3rMWv?YSG
z^b&blYhU5G<<8KazO=_@bSp`en|n{SRYe3(vI3vC4o4T*9f6e7D@f%Fpon374@^8O
z2fPG|P~(U+a;-jHTAdf%+M1q#Uzfu*BBPAwBS_2s!ejV?iX`=0LEa}CK1`v@CJLUH
zf|`7(*eztYutTS_6*<pyAxy!RjQpfHNVtC^qhjkN<TkJj>zaK!uHilw_7-z5E2pW4
zf4Sq*ngeO9)V!S~pHF;&bJu8F>L<`SU|$S)X9d7w8ktxv(6V~i1i4WRDnCMnzZ+mm
zbFu{uh}g+^SQ$vpcYMrk3h{;7XPhG}oW#5VNSMQ!3;*#MR~)6v^b-E!C}~UH-)Eyy
zmNM+A0YJmW{w!!4H(}$iO%a9{Tme(wVe$iDt?4;RhxtWb<7`q3w}^DFLW=UjLQaHB
zS%PIalOw$JmTAmnXAxn4;vxt%Z{HrT=d9h7<x>fQ%Ffkk9`D3!qW}(2s|ZDO*B?2;
zo5$jl=xXRH__TZbTBjZ+!Hb^IU<y2l!j;S2?iC2(&^t1F8G)inCE|ByMD+HdwF_$W
zt#i)5kV^fG(1J;_HH&?E{k=D~?QZsvfLQm|3R>EF>C96r<%_eHT0wT+Ggn+BDLzF!
zlE;*b7&`NnAH+%dHhZ6{*FOB{;qN0i18c7|3gr$BPMng{)J5OK#h-`g9W;+6B~>6;
z=`&goW#*})77AXn98pkug287~zWS7&v^?CNcjZN8|8)chp9hHeTkIW^VWac0^qs}H
z0jJTH0?$XJNgef*mgJV0>6W|gg1kd5`Lb?talKFByzfUms>c$tzaRp@g2!biENB~g
z>m5h9pBiV?zw+GGMDN-C9WosC7AFsciPQ!SbZL%Hf+pMe-PbJ4ZCn&)S(MtY__Ua8
zufhn*+C)cBi*C`^ZX}*h7d<_kBq;&~5rIJ{b`;<S5TzG5>#{k`iY%Mn%+-;64ALxH
zQ{KVxfxqfOeB70(U(|aFaVPW;bVW;K{BJJ^sfyDvHM5Cos&RloUnk3~Cdow3OU*w?
znlCSD=cnaN+gF#B?l^$gxLDRE?$CW7*7{A#lEtpcI}x_W_`<YvjF|7<NrW`-XXUCs
zDaFETqbsc*jTr5$Ic=MwB;!S>>#t<k5XTSPVY?*4lCL<NWM<#_9m1s7yI_`L1jK(c
zoSvcX?0445UWJ1}3||=)rQtsGo|sk{E0~(3jk#xQdU-ZglRgV?NyWfECYslf_Qgql
z#T9Ds^%v#iF-s2PNs&f8)@x_7SfxC!pCJ}aY4v)74id$kk8G87>TDDoZgdMzmkR3B
zE;6{Ec$OMD&=|gpO;zbFB77}{%**>2&drl0gf_)ELuN>`>2l)d3inQ|CMOmDUCo~N
zkV<GQuFOwM+<(jj8|Q?iOlRmciJ1|De6yesKQSt^b+xr>)hq@&dg_n1fk-}@!CxbT
z<qIO1DjFRF2X?n$$!<)mZ<OEG{+u|hk>@Q)IXTAXrpp(+TLF)O#(uwu1ui?v_5g<i
z{ryBMN8l2*o!z!i=+!b5SDh0Ep#t793TLKRICQZjdSy{<M-(?ozkc5bcr0B2q}EoO
zZsdVr7u=*t8PHU^N_wR;4CV8D>5H9d^0|?CS*1(wLGi{U59#nK!?7MnkR17N%@)-x
zQ3|8YV_KtzaYF+)=!KSg&dg7A+k6<9TLa-`zi-@j`eXBaFClZr?DiCj$e*{zF)v=M
ztv4WsghXNF!Uo1P!?Vh#6qbTK_EMBU!{aO8dk-u)d<IIBB_emoa^(>)Za0k8A?l3s
zsro`0ZDJ5J+pR=IIuOzHSElNdk_OGm?&q2P%_67k<fn^`poL5g9z&*u&D1j{Da=jw
zNdIg!{(#o)m;=LEcsggPaWZdIWnq^nHgxV<L4)m@Bf-n7t~2$GJG`O`Mm7MwF&*53
z0MM;G2Euqt*h(Ts(JZGI;2h1zOk%b#&eM`gT_E!$2?FWQ%(i#lae`GL38LI$k9F0`
z$AlXekP<p_?7w)!Vs_M;dgZ1+0xaD}XQ6NpoIUP~OVEb))|{|H+g&kuJnT7ga7Qz5
z1~_BnngA|#P_1<N!L!3J(lRw|o+VLyS`dV`!3$TS4&FgvH^BRP3b~sOo-NH-MUX#N
z&T$&nN^fZ=*;he_GIIn{$Cr**$8Zy5Qd&ySBp47eI}64D<zxHcnPqqj@K5wB&jhDO
zxu(=DBM>nh`K)})?~qnm`vHjLt<O64HLZwMOXE9F{j^;gJ!(p_39%{MAta86>9~F(
z$)Mmv+DZ!pQvBit6qtq49tB9kY@RI0Ur!K1<)f=(>xh_*OCldLDnYvY|NiTqjt)03
zi~WvXd|%L6thPf<vcsJ1R!t7A_icpAn24}-ZwbBq1z~|Wv{tS<b~kVQ{x)j(AL<=(
z!}ip)>How@>wkTPFp#y!O1AD_(GKX(LU42zr}Nj)84+cvpGFc2kN*8gj0fQ=elLEL
zw>P3-kP-rh@O+1={qLc#N5{U5a1}--t=|T8V|HF?Ll{}v`Y7<XG28h3CIn=+;}ACV
zpJPW?!lC`6!ucO|{>Fo&&%ss5Z)yHH<1ooTm{T}zJ#Wp_7ygX6Cl_u|(m%&Z*85=3
zQy<aEj(g(@9KYm0NqMv2sfI(1Q7Y|qq5<_QH4|X<<Qx&ZC4<%lD&YjUQ9%u6i?gHJ
zkQ@OQA{(U`dW6^db|3OqnN-mWb}b0JXAV5WgX{qr>$Cfn50}S>$+~i;<gdwQ{Hc%7
z6G4frcrR=jO5acc{m?HdY&(&csAQ5DqE&ArI|t5tIXObKET6k_g}nX%<2`H<)WrHv
zxIDUtCDDC2A2;Gnu?36PLV;MqeN9*1Bd8E`HWW1?0SlD6Q~8!diV$qZd6#?VxGNxD
zjRc@vIMU{#UXU#L0d|-o0%K+FHa}t00<q60jwnuKy}MFdeCJ(3^mbhPUSz@Iy@v&m
zRU-)4tDLpEe<8o^Fdh{^;}}nQ*7eV~0}J3QFXO$Wh93$4kylYgWLLgwJ_GHbe^x*I
z^*6r!*An~R_v_lx5?&AxlU%<>{Y&5Gl?*Il(Xg~-iVDE~5P`>MWedozsXe!M{(HuL
z3Dulb#h#u3yxv$nBC4)$Tf%j1>*s%Ot{vb_`PbgMYkzzKu+6M}SrDuut)*lL0Io|i
z7pvdnpZmuzaIcY+iN_CNaVLrr70Dkr>1~nr9YC8i1$nwXOGcP7k<0JS1=#%g&f|g+
zH@f*zdrshuE<~j%=Jd+qERFMJX8~TpKP(^#bg%T4v?U%mC^7ey9-(%QCWl<>KTClT
z1Lh&cFn~HZxe(pT8~~kf5m)D*uc+=!$?KbBN<^uKPSKjO^erYXBLXGR?U4xGTTI+$
zhAtwUs<q3TG8#bBxxH=QCB#`t71o6qvSjjZGxM`!)aPVjrcbs$ALFXys&%?`xXo!N
zaFLLE0Xq^Yo&et6zv~{N0EHB30{Hh(AR#ne^HJ{|Jg1q+r)caoy;>jYV#-+jb?-$O
zHdIS*3{&GF7BJ3AFlh&`Ng=8lz}s^>RDb5cI*4WFqyJ|Nuj{=jdV51(!|Zfd=Ww^$
z)PFx@A}Buy6kLDxNIM1H8VvxZpx$4FlnpajkEm)2E>N=-eK?&Z!R2SjT?4qaf2J`Q
zl5$0CBRfu*=lB`8;K|FB{eC0N5%UZ>^1imIuQ{)lU!p3RV50f^E#MU3EvVsTdvy|N
z?YmQVj{PA?g;?S^4diVHKB~T^u)nWV{g0{Jc$1)`M(wEhjLr9QFn_I6vU2w5q3eIb
zeSI$>w0iv$sD&_zS|{@VjxPzucif5diHnAxf$Zu3{v5#2yA?%I3o-(=PKW;c^V`VJ
zH8`JK(YRqCd-gv+KZD#|PXx7a{#32gx&Qt=0e*e}!}&zp0PcSNzduLrt{{Y3cm;Pi
z{qN6jB0p#1d?ITAcbEOo&wpt^R<B-Tu-|{+Kks%5k#>_kidxcAcrwcKpI7rF9j=~b
zzoU)t2?nR|pQ|GxP!6gOs--~0_N3l_R)#53Cd<?zFC@&MTX*dG`^BPJ;l*3$AI!EX
zATN0o@t@rn4u`rqAG7ax8a#U6e;&OSIe&L0Qd_G-e~5VQe|FTbWnlx$rhw*r?=+sl
zzMBjEpMO6IcXvFw<L5)P*pGvZV~5?o1KjE#GlQpXCY0RAsO~66`uFWpk-Yd|8WK<8
zK;P{*$OrJvQ>2y9;0j&EH)t4!A9xoX=}-s<YwpsceZp@<@XfW*=3H#ByzO5?6nu&>
z=ohVdz~nwX3?$BE6Y;QHh^!(Am9j{`H<;pWmI4sUQ4KAbL75F4z=n%MMh+n6tf0HS
zRshj{=ryJE@(zO-l462XwtS^7v*{YJ#ydUSW=G5bIEs5BSqBQg|2z%jRn*MK-!|vB
z3DiKA&v^qG<$w6+>$m$$8SdAW_|CviKPbq_y$Ol_Wd(X+8T`w%uC;!SuGYuQdb1fP
ziig~%j7Y)BaVl`R?L6>&o9mMgLyzm999P*Vr14xsvi^qJ?VC*Wx9fvTw$Uqaos(18
zH8NQib7lWhjQhKa!O!mRO57)>bqTHG7na5<THIsZC8{b~76p)J^Ti<!+gFiK)PFuH
zcxT+$xY0w!e;NUJ0%}Achew&y{~y0~{jaOAPL%xs#Pa9o|8wd8BPRr#gUk1mfq;;Q
zBCBM81?njx(*Rn%lj|u$OIAe#{PQ;tAqj*;Wp<kJQXA{PPu_>UTk=}YaTIleq6uhq
znFF|*ykr<Pc!GMFD;tthNa5qi#`>a0<Y{_U4$x!KQ79mB9(SJ`qq*|Uf*B>XTPNQl
z;IWk~Y2%x6ps6HwoVyVG01-Ot&O!mAY~NLCG};&Vcw07hAhws{X*}ysb-AsK)XCV7
z&)W$vNd)5_jTGF0DEcH#YNo*I`Vf91q**y-kE$ie^+~z>8bun60{5P6RtDXN0(RQF
zmKQZQs3L*bb;Uh~8(fDq6r$1vk;c6-AKtAyXeZXsdvYDLdzicr>siv@aRq2{3IDb#
z8@>S}9~f9%v@#0>-ha3YXlz`=l}(%WYtNyPBmUXRXp!GK02_PJ2OIYO-#_-jrXca6
z)5*(XZ|tlNzR}~rI1^n#BmCEukPFwA!NYwfqUIm%<Wgum33=(s%2f9sUtbe}u<fbN
z!S_3-feiEu9D(%jOCELx0ztN;6$<%}i#rpRZmB`R7dn(AFJ-ea{CRmHEWDPj%DH|D
z%+eI4iHc#ONM(g`sOcYS6MS8+2*yzQn(oVWuWgQ`h6|5v4|wc7ap;eU*as64Nygny
zI2$(nk?6y~q4WoLDUK1Il8)?D_Z`PmrCtg$nyG4B(-0#@KsGZm^2WD6JZK+mEeQUc
z_3|T?-)H9U(6a5R>=w(+)XhuEY=GF&Q=R*-=Y{B<{;g|c4l_8LMFhL{T&}m{pB65)
z=b91M$ps)V;m96(`Cqr71Vd|HE#Jb{P0P&p)3#fWq8BQo!V=D7wLW>+U4O0yv?099
z;XuK@ZyK3;q@Ss>+H4qnGmy>3j&Cy->|>z1^BcHVbnXk(Swb2@huw914&2|}pKiVQ
zN_;1(W@j(qZ1y|tN~?6N3n=C+czN@EPGRr9^!k1{hzliMnjc$7*TTyVJd_Q63d^{|
zdWGc61U}dXo93Onx09lS$_|R&-@$>nJ>O++sat?2HD(!>J&fI{#>}hssu?y(hEC24
z&{~kt9ckG5YxR78yxzQxU%k}PvPVN9UF;|C_c{7)%us7bJedO=<Y7<h4FhHs^MI0f
ziSR}W=?}Fx&v@kgsb9;l*N$v{rAEa!%gvF-aZO6L3v_2QU_NwDomonOGr!P=Rn+>s
zCM$^c*s_S(T5Ax1(eg9RR%3JFBgmo#ZYaU{A;Qo3!lCaMThcTef{z-1wuX{=th-!y
zC(!zE`eL&qUTZ7!sS=FtzXY&+J!p77kb*g-XNUMshiZk!A*(wF*3sqEn=>UyuPv+#
zWCttp&sM)#w`N60pOu8veJJY`sr{MI#s=Lx4fWj!fNZ{{Ec5z-i);8`Rux4*;_I^p
zmsxR9R?~_H$k7X)YcCPxUGC*Rm$%IPS|SKZc(j!uh9um)TwX41?&>;}Ud+}0`j8~3
zCCh(ysBumvkEE^Sg?nL-fNjv>JK&iGsewW>ugK)lVyo6H8z2DTKYOovB0Y-Bl1nyD
zaCAQk%CWQkw|axcBz?CY0WC$_L`$3(F9be$dnTcb=v}38&kd?6?_rpp&S%tg_Z!~J
z#69yvs+Q`mPf_Fg74Kl@U_huoBV5eR$an5_<i?tnJ7)sWly$ZXQUErce%L9axCL)(
z3&ZwonnUW|E&Y&8{H_$8LXD#S3GQ2~@gQe^fn`fZIPvN-<Zcp|brw!xVVU}`eq#Z$
zkg6Xa?Z!S&-H9|RJ4lu_;GvdFYr)j^DN)f=%o-Td#KPyWDBkaR%!q|0gX*pl5c>L=
zhC9lnv&3x{Z(KCbvG((hs^|ccYCLfE_LI2!)+-O{ku6!nbRAe|M1UlClGKOA>X2SX
zGV(`QMJ#&L(@NM4rRkIBm>EN0$)1giyvX5)Kv8*qGyzH8qh|_gqaAF}hG$M;3OLe@
zQ1cmDI;nmdS^mBz)G6HBqea@VX&r|?)M)9YZ^4097I5r{@Q^miObY_TK|4*y0XOcg
z9qNv}#9**-Omc@Mo7JRfS-P`1q>D;mjVD17U-YybunCF5yy`>Uv(q&3w#e7FkV7v@
zZfNg-)K)WRp#vNWw=|=94Vtyfh$=;h$oHJyB^lT<n+Z4+Km9z&kefb-J5a{Vg7^Kg
z(2bO)>1*K`!l!uwR0I5Tg%C}<B&ci+0FJb%efQRnTlheaj~d;}>U;fkmL1<v(0yc+
zdjJP=0BM;B+WNEdnXCHrW(J>2ozgAavY!z~&YLk}tBqlcuAbzuOjUxGIz*&O$TWoi
zti*Uf_f#QhB_aJ^oremtgp47xZQfTK;H*OeYvT7|+sL|@1nw@1yyV^&{wK{5bFc3+
z3nMw<F(u2c5y<uD$Tlg^Gr<g_W3V~bkAC+kra&;2kt7|uY7YC_1HP8l3se36w~f|7
zKgqeGn}Tll-W8^!MqR5{JfL%8+p`N1ihapujHs;_&cl-ReL~CFci8z7#>0K+!?U?H
z-4$>B=jZOd`uy}wW-v{uX<eX9QzSz3#Y)1mx-N9U+ep_R<bmW7wI3?J4^GMFUxrIS
zeqBUXb`&)#UGZ3Iv9=sGuI(=O^lF{ccEDE2m3)80k7f+WdjD5@Ul|tV_Vz0XiiEfk
zL;(?06hT0dkPuKxQo2JVq*1y{#Xu?P?q=v_0F#!G0cit<76k^7koR7=_xbPl{dlf(
zo$Gu!`-7L8y_tDtp7pGC-@p48GKXI8y%wBCjz`56z)y%5ZToXvxJXv<X!w}JOWbFv
zSH2CUDU)pg!Z9d5eR`39cDS-2?vhC$Qq3)>;48FQ4jt4uV70^K`pLxAVQv=JkOalW
zND9<3pC&ay<mp5c@}B!Fi?Pok@D@+$4Re%VCUNe?!%8csclZq%TU*i?!0TN%ZZC(t
zPCn9l5mLV1cCSOq?;`t!BBc>1%id1JzUSgE)H@L~w!38|%J&$mCVyz14=g-nE3cE6
z=LhL%Iea_K2kLs6;S#9Xz_43r@NUpTX4~ZfOOh9uSOL5EB*9ZQ#d^mSZEhRG8g;EP
zDztOwb|6NuN}ehpkEd4NAFC-k1mA`ECS&Fj2?J>sYlklv*wmG3e$-$?kA2w7eQmjp
zeTRP8=<%ZbH$KFR`2~*rlH+>uk6|M<a{odKkE4wfjEde;Icm<sR@y)6x4Yw7)D7PH
zul=YFsl14Mn+;kWWanrBtAi+CNVBV;8G?r_V~*r&6~*t2-B%!cH+(CbIsoOwFOXKl
zv8=<%mbPr83(iZ}n^S`(IA50xBY1-R{dObFzYeiK6aXSao&`Wtu7`+ra?&!DVO8G5
zHiK^7ygc4B6vBJXfQj?|db1#Wlpkn*tJVz^ymQ%GfxcIS%%z0ZKAB=2rhEg5QF1Yo
z)*T&<?+WDjyx69x{46ohpz^Z1Z0^jB^`_qh9+5{@a5@PBAEco;M;;l+<G~|J5H8kU
zMLfQrk%*)ZQM_bq#hK-If2}W$3|i>)hxKQK2)BpgGL^|5%HRI-`jKv+0oSiA<`MTd
zq;#=V5c6m|-Y_I=IYbSGrO#PQMH<qE>-NA7a-?#6=|0oEAg@3%cX}Q9bFFY&Hrjv%
z$ny!Ws!Fwo@?eRz-&6^npwu6zvg#fel+}9u6tvo(p1;#OZnWYA>pYBsh;Y*Y)WqRQ
zHZR&i{X4)e>E7TtY?T960xtCFNZAe2OH*Tv!---l(NGECbvZg|4@fT}#pSuikMF(%
zQ4(c!vabtn-lkyd?vJ?9zI_W(^&DY*+NH!UX|(h)Cv@lQHYBtUnS1eO4)GlgsG!KD
zp8{Mlgb8JE2py=kf3u%jw8wdAd`vNp_njC&cycB1k72)Kaeb(AsFF;9UD_PH0nKB`
zQT9jgkGDEgfd5y{TUduA2g!1>42D@P%3|5kna7}4B=iopz896jtxoOF^0ONGs&y1%
zty0MQQ5EypLRMgPRJZhwwAqIR#FGlmV;zRWyn0Cpq`YN?FLvO*jWp@zC-_z*4d4oa
zF@|MH6>twW8l?9&r*xCH)9Q}q=4FDY>+iFzMUM0{*BEYKzC`Fb87N?wd{%ocKyMXd
z+j7#6?8qeEo0+egfC6Nb4ySPu1f}Wg9xUr>X>Fo&7Jsa*Hmd>QuiG%6iMbU>mupa4
zCIlnCE-)XpQ)BYU%YuDvnY7<)z8z(5Pk*axsU0(dgeQT|YDaQB9+$R%3l+DY9<4R`
z&^s~;Ttqx5xnB4VD;uk*6!0`5c)54kRBw*GZpPsasHwjV=x?c9!>!O1<aU8~Vn<a~
zCGV|JKyMnQtA`&~i^75zm>F<F1A8FaJoTcq%kVdt58(Kr2~}uFEhI23_^gL=ADhX6
z0!JX%t=mSo46!`BreA3C^&315ai)AG4r`=5<)9&D-@9{$g{s-xkUczH+o~sg^G#$W
zpK;hT?yFJks;?6vMe=&SfD0r{-$1^mQr=A`^-jbh*58)7nl6XaeC1F5ojht{U-}+~
zx4`_;%Z#^Ycbt4`@ZD0$kZbp^;?<>zi4V0ajhJQ)Y#qh`NM%*knl_d#9ONpn-_yoX
zGhtt>;UIs01@5xh7Gv6NXCSWWj;A;4&g;}S>yIFQp9Q%6-1mw|e_*iKL%Ih1xWrlL
z5~@sTx%BuHMED!FAd@h}eLB+emT|ISuaMdncBlw<#xPGU>*nnuihU>-yIo@>Lza47
z%br3`f}Rr(<_T!WvT>bq*K~GLuRK<tO-hny*=4sAa-p;z#^a~O8Nz>{_^0fd6)zBV
z3{&U|DZTt*Q||{BXZ4LWzt<u;G~0`DnMQE3LvDv*N>bSC!Vvu2jKU8(Z3B}grn!#f
zMo~35VTFjtSv!U^f{4!-u;0_Tb3%J`sVw?XZ;T<8hwJqjbX+V8bGth)-LEE>=g(B_
zV796$Wkyv5GK(uom?<3??dQFQv8<`^yl&Y0=n8>(o}!9(c<yw3_D#^^l>kqSY=m>Z
zDzp*DBx<guxn^a^)Kg+d$F}Gw-8}~OCgI6>5T-tCq+=egsp=3NXwCM$?sFi;LTJ?~
z5!XVW9Z;w8qqKVL(r;C)L0Ew>d#uX{JJ~=ZK`aV{+FgzBy{~tuvhT9D`hY5e9LlAy
z+??gyVKe6`rGE)j#0TOh4qUDwB=6thE|^&&hiH!@H#MLgRn4OuQa1jCshi{2BGq(S
z;G=&5J#O1yPngLInTO42Ny|`(xS2M&3d-bYH6^WQ;4PeR7|XZxGWQ6nG|CRba~yZg
z5axTUy7iGsNU%ZYmA_z*X3w;zolAJ<h-u+?nO2cr9MBnuKJ4OLhr7XDjP`^cQ{KYv
zcT4wG<lS(}5Yz)}T0$hF^hR87GSk||gYU;@6pY{Q;&N3Wvr@KCKar11SRBsTZ+*4W
z^OZrbfedM*aL90uYwlS5T_w3D|DaHtn{_fQY-JBs?D_krrJN)lXwhcQU(tnbY%7Dg
zJN01IxKpZAnMD^vf>oSZoY9Hzz*De?(yjK8`R93Zd+0_jizDVKifM*aU~KT2c#hyK
zed2>DX`H9wwbXdY>1T|-GtBc^OSpSVq-o^8M}L@6(S`6I)Lv4JIM4RTaRhAOhI5&L
zypT@eh%LVG!#N-|NX`=puxUMVB}BATPZuIbatm+VJa`9MN8cX8%05jq2%As-STru~
zy>U-uP{l8&;X1}mq_UT^F5V%*@@Lg`?7MN6!Fp=K^LhziV9Soj)TEpvT1p%>BdN`=
zD*L6jeLl5GyG|!t^z>et$O9k9qqjZPjv%f3X1!v3Ixi>Lx;)!LWyOkv_`+X5)?{o^
z%Wf(~?G#d38qt+o!=k;;pRx)Z)t<4~cOxG&*58pke<SdD@pxJprE<34T<~-vBz8Pp
zFj^ny=qTkRuP}{rZ_&P!2fj~dNAx3E_3V1@M^Sb#dV7Y(EWS{^C3U{XbB8dPQMIv^
zmWZ=3Z0M=IsErw&;Bn;1A|7XnadA|aH=Gd}B<0_+uZHsY@GYKIa#dK1WT+uXRo33*
zD0s^iADTlD!`WnBd>qdy3l!^Z3igpaPKNh|w|l#&oV)Aw&51*BFBz|G>yCKsmtpIh
z9Vki|Q3|OZU-u9!RLuxlH<|QpFuxhO`d**rny>Ti*|qSI_+;`+QT_5n*l3>Rwa9v(
z$?1XlG@&m03pt}}Sv+6$sN*6$p*HMi+?pke$E1cc%3p-HCL&oz2}Rnal*2|V$;i&}
zXlL$YsW;5nA-OA00<%wt(GrmskEiq8her0NnE8@TY%)IXJqLr)?BX??0Cen6Dy=Zx
z$ld;Ob+19w=(8a0`gFz0Ul)Fk$EZ7==i*xR+g<k?q8$A>)@MmBwxV~(Q0k`Hsh`8)
ztzJXt?xSk8=ygdJck1lv=!m|lRY6utyDX}`f`fRhzb43{xV<KSQbT?-y)$}lsLLuY
zq=zpth$W1eh{_FjMAV!`t2e%`lz6Oxd3Z#98&W<$uFengq#3fGapZl5|4^6ytCafM
zYM58-H@gP^VJ-sAjs-!h-b~L^DOpclr$ux!OmZPnk03L@q}kx{x_4_;OiRSTU(|83
zHFo78cBFo5e64=EMhOG4vyRtx8U>X)+@W~0OTUcAZ|>Ntij!;uDltFd^erL&!sX}q
zHbwBUIZ!`0^g?j7PcSO`G}(rR#2Zh;R1uG5v--m}*B@i+j1>8Gn+ks7Yvx?cFX6&w
z%)o#kNAPJV{@Ap$_euS0sSKBCc;X`0P83+mozXVSE^Gw@sA4Y;$N7`(3;4^jMx9?m
zn4-WnN3PJ(`f*6wEK_NO>(o6mqxQ#em0~@_zmJjD>TxS}oWv8Bq?!|y3E*h=l=xyq
zyj16<r8#h}5CkV%e!U9A4xzX4BFFJ-9|cGv*m|-Q=)awOJzAj6Qx1l@4_$YGQ}mXp
zY@Ukm?uSIHfnwuRPJRsbg_DkRJN+unP_A<0T|Iot9H-jDAG$DTE%GwA42MSZs|kF|
z*DM%zsW-Jdx`c>R=a(Zqi%gMVF{*1QDltM;diUhYfvWhFd8L}EwH$paItOSs4H@Lt
zo}N9t(wMUKlmhxVd%9>Y{zno|r<}tLo9^yC{!Q@s|BC8pUbniL>FI=H1P|<s<uDwk
z)JSb3qg9kRlvMFMhf}`Q`SqdmvB)Qq-RoXqI^|6KjnusTrR^ZSw0JC4KX~9U)A5yc
zC%`rY66!{jTa>hZT1YlREE+JWy<gf5wm;d1j5?`gaUk-ZSwC9)%`imnnpm%NKmqUj
z`Y$b>*E~H~iv;JMl}URAbK|Y9Ry>R5tA#cZ%QLb??pqA*N3AREdW+b5e|ZFmkV1ep
z*CmtBt|fLO0vWc`4EOu09VVJ_DdcDT{hLUR*DVyUeAyaa{XRCro$(Wlc1{!t<?)RF
z<SpH7ysU8~;08?(Ln&gPWy-2!Q^}K8k@>fJy{nymIpC`dJ5uLU)GlSS`pn6>NN)j5
zl;pDkrX44WxP)&v@5^v;iQzG}_cP{actckdH6JLvZgAb6?g{wweynm+zs5WFqIPqx
zV9~Rg0bd~Wd)&pEfAteiu=ks_ok`b6&0f*lIAa@Mn~f40uPz7(<6Zr<=mZ%4cMO(L
z_3IB`Em=L<!>1Ok!{+U1#Ow<9jFV*OfA#OjFi#=kj@m)HjLy}_Npwpm`j@==pB&v;
z?0a|C@FJ&Xe&j@f;tpU8({N+O#IO4n8UYBm7<Qc?KO0=oXz4k4kYJjWca+%os7iN6
z(2u#V;chbpG76I#k+Be3X5Utoo$2X97ki;b0~P6`%zsfC4;=Ywc8o2W0UR@Cp`EY`
z4O{`FK?u*vM-Pc5dFPD!@?sq}=i9D7I1oo~<jz`_*aN0UrmIH{_rgg)D6+K*zYsSy
z1%Qz7vW`pQ)%)o_g%k(tEC-=>t2Yae=27c?N3=vryfwM|yB1jE3lri?PNP=}JEivR
zV>FSIx~{=f0}tXWss}X(mAX{Ux$hjtSH0CaG9sG<58@6nrJ-_@2YW~^>*LM7ykGao
zU9oM(H+m%)0wfqn^gtegda3szClgH9%n4VmlefX&(Y6}5J)kp{9?V^a4f)9{Wb<SZ
z0tY$)z)IvwOLWt@*U&jPo_VR1rthRnzbY&abD6cZ#JajUA%eg4Qx*q@3ixJ6ye6>P
z)pMGf#f!isTJjNgY=S1p;#DB2#p{QMs?O{4SROAA^wsX?y!c>dv$oP7$Bo6Paxz3S
z70fi`t?SWWU~i*iX;s49RawhcWzjpT&v^~S;9k!YAOJp}LP7h>CbxAWn02$EOrzV6
z7|HTZ2Ike?hf#qbm<RnZX3JAmBX$13pQzsE8R;URDXrTG*ms=ydi=x?um>-8Kbj4|
z&sf_R`0{BN8Pqxo)TlSpah7!EEp8?mU_C39=F(F!5M3$L@%b&qX{TXrUt@K(LTKQI
zs%@^i7U@+!cMGxgT`yR_pzS}LL*l`Z%yq`0BzdH9DuYF*Ex+Pyc%dm>(m~SpITxsB
z>Rg^jk#G)pJy<q?7I<-~qM%sRJmLF|umt=2eoVLGKD)d!XnIgL(-LOxmq>3NAaNWX
zSEitHohNoZIBlwvOk&<Os_7?uN^zqNFioMoQ*@ONFvuPQbyC}(o-`|c9Hc&N!`en6
z^J7{>nZ9bO`5Jk5R$oq>K4qTrp=jn#2>vL&^P$|)ke9Q?M<*@t=OySP`1mF0m+e9E
z95~v!eZ0q-MD1L2&s1cu#fK1_ItqsxDlCS^r|6_!?)b64q14(#&^?}`{V>Q+u!17W
z6+4PxQLJcUPMj92N-jKGkJ?WqKd`y_6)8SmPFUVd#nXV&r@pY7F<D(v0U?~3==n_I
zJ0j8_NmS%kWYY2AAhiIr?v^ZbN^G!db>LNA2^%*S7x7-=T(N<f!NRDhxtL+O#fznR
zCDJ{>xQ|mJEWyY@{G^iHhJfv-+rZ1ayW+KxXWtLS_-hnVl<YK{5N*l6GSi#U1;ktP
za^5D`c3lsD*c(2{{u#!oxD(5Z@j;eQy`mQ27t4vZML(=Rgi~;&vNa)XQbtx^#kk;J
zeIEriRleC3?<RQ}7o-+J%qs-KTh8v|q9}u8Ny+pj%H4ZC-Z1!bC*pF+N>3|j7f^1a
zB%;zDP%|2<bGb5B@7DcX=A%4targigyphrUe7_CZN{Pvv1*Sy#D2_M)bb1h%77q%8
zax6A3O@edp(>4$T<-N?~BsgqaISE7E`EHAMUGm&#M>oqj4A7<Hfsrh3#+4}qt(h#e
zX5z_FM+RnNO+zjtdrhp@x`5t|DWH@Kc8jh&H;i~v@QYRS3m=LXZIw=iB@NmLOsTFL
zKwWlZu1-W0GaX=VBPiUYx3Y;`ncAnWqw1{>3HcA3hnSF7D32qE;+$gaZA{Hq^7;dK
z5U!X+mRLjEY60*t9`vhHdr>Pak|boVZ+z0}Mv;Eu@U2tA&bgo`)2v}soQKU=8*X)@
z@X0vUxYN?lVv$DW!vrou>flM|FzA9Eo|j#MClCjG&E`VmCb27*ERhy3><ckT{g>^1
z-ryc<F;J+R4Nor`^=6yA3HoBoUIcWF33A#MyRKXv8L}CQ5MQlgBfH3?U*PUqHgr5H
zywv^YR}08eFIS+`1eARN@Xc&XRZN%J+k+2e!1Y7sP|fXiP$ZSUG2q{nC8A+-CkAWl
zdR2=lzzpJH$BacAI?N?Wi&^`R6Oj<3c&82Z;~@ufwY9~Ogh#j0yH>a6UfK>Ce7x}0
z;9ISQ%Eh&H1r7FAGg@d#UY|}4r%LhN%o{2-eQec}8jz=7&E0`{I5pW9Vn7Nvfo*x`
zK!SiBFTy-e=8wX0-~?_=8;u_{AQ0&_)PaoSPK=+TPXcr;BMxDNc?fXq>M5*k;MB%c
zbkL$00Nvp#C>0Wxr#cb<m;6-lp8QEL9jC>9lCioi-SM{%yeVo3Vf-XXYWD)D#WtEp
z!Q|coya+8|E;8i$37ndc63z*ysXg`SP1dhPIE~9YTrK&edgOcNPw_c44aD3YEWUeg
z+H-LnJMx?A<1_M2@x7y(e$LUCz8LZ9@*r48sffpmClyHOmCt?oo-0WQG?$ro*zPOz
zXqY2%!Zu>2tNkBOE*>I}<Tez3#Wi#>+JE4sFAQuJmi0@G+`4^$iz*|I{GScu6|P}>
zrFA=_uv*wxdky@|QS6J9QwOoXeXFSMjFGuMa2Z(jTY@_Zpd;t*d$t&ZJMEDdjuy(O
z0FQz!3!Y*2TQ=^DSw<sh2J_opKktj$X1A8p1NJ=no>43LS0pH~z{_#W$Sj|w65u3P
zv)90Xzk9UeLaXLb&HW4z05~m_NEFE}asdkoOXY~|CY|+#@8j%zYJb%2!QH#oFb>0m
z9Xm=*DfO!cBBzCVldsRNb|M2TuJ>qy+nIORO_Ecq&~jDsItv-Gp7wIuW9DZXm0s)L
z@t`|<7FRL_G9Ny0uCEy3*uZT;uf*iR<939s0K&M4_8&lIui>ffvgdXEnxeT4t#giA
z<@b=%8{9^9#T(~ir;^U*1Mr~bPg~`E1Z}x|6qXxUPfN{w?PB|w@h34+t00u@p!xIx
zDy^<csOM>uC1ADLmee~<D>E+G50z#8;CT5(R*M=F#15>{F?I(TFsAQB&^Jop4IHK1
zwS}#2n?oWLW86(i0GPjr-1bumYhM+@OwTC>i~jX3B<LcFSZOQ+`HqN`@u3{PLKG|~
zfO2If4SBGu5lVVlJIuZtU+_@ARVnyb?x)()RxLqn{NZ0qfMS)NNUjX2%-&~th^Omm
zXk}y!O*jZ1Y+O$<+$oaMEqzudQ@|X5rSJkNgPkgPY1R9#K_9{7p>UcoNE!f@Y)hl;
zl4g3Hj<g(hRCH%;aYPnRc5c{rNo*`k5z5T~>-L8i3Gc)Pn5>q(YE!(4Nb5OP7tT!x
zgkPImH<HVaJfa3p5rPi(b*_sH>u#6BeafiN26Q;!v>%()(94kh_bJgxiH8`zoM)6n
zE=veGLu0YAH_hW5P8-k$9Lw89bGtcQ0lf#){0GQ&vfEdXs`@ILxED-cI}a3<R*8et
zrCnJ8jG$v-BcHH$oG+@52fvL^3sd9k-eA)p^CkzUkdf|EWLjjB-&_IeQddR-->t-}
zz2ob@uRynV#jG_f_C#i7(p*-h8ty^T%^kW+HgPmjAj3&OfYx~EZ&^Z$h#SVo6H2gf
zf4<(e!Ccihjc&lkz{60adfr3!lw>FmWJM=PTvTjxbvBXEZ8GeYrtvwjtyQs*7&wMl
z8NruJ#WH*goK)t^Z3emP4`XP=tO@%Kbx<#`h{e^d?X^y1#H%w0zFbs=a3-=S378Yw
zQm1)RYAOrQo@IQSH?5`n-jpuJV{|JX-|~)=O+K=9ErNJ#COui6ep0$Aday-LP3n4K
zb?&q__5(%{wtvjg!rJ#9MNK2Jz79?{yI>t^0f{7$=Ii8}{hSNYzI_6FX-9$%$%Zm)
z9q&bN=Iy&ye_@JW@2kSotcvsi-@AkjuxhFFO8^1%Rmtp~{>^eXK<pY}H&|90Lb*+0
z1$(hsUI#1F0@+EQ+Aqox648KGrp+A@`fIjOjo%yUscYzphqO%I!Kp|6TY0R_44BbU
z)0Ud8?hcB%zg7>_?gc-GdZyB`Tfdk9UiQ21YLYqptCyPy-3r({c8mo(45zQbShp1~
zk8whpS@W8K=JSNQb&x@=1=(zDK&5Tl?J`L<Di^%(mHl!oztk?^{?+QIRz>o|&ym}s
zA-fX{g%}B|)2r*}J@`#-p&_F+3{%v<wWA2x04CCY$@!qEy0;ph4+7bDEMu$WVPLpW
zCdt5?@SG<}ul?@l<^b0TU}~3Fw{6@7gURqQlvupScWxm+2KA}UG%3IqP%J3+1unD(
z(9|3)-eI6w^t99Tqx$J8J!Ga|X>AUGzd1S}>c^lK>O?sQ)6DgYHA<oWfrf_O$3dU1
zq>}a5UsD}1x4K{Lcc&#SMkBbS2gThrG=PXT;m7kW$=m_&Uz~_L6Dr!rO-I{op|<B(
zf!u-xXp#~6sPD|ZepT5ak1TTQyP$;MKRR=!AV)*I-QM#Jm7rNdH51myJ`^(L+eZc3
zR_6V-e?E?b(40x1pV8ud!U^)nv)*cejqMBDln)|5hl_&`V@B_{ocwMdtxs^aa<A2A
zm;*sB*N<PVj{BI;UQDGp%t7wzLUe;Zgk1Qpwqzfm_k$1TKMzv5n8=97w-l!e2&-3e
z$684}Rw&7ykVQrD5fmbJDOqzEB#Z%nUH$C8e<93>Y93IJA}a4FXIbMqL+T8Mj4iPC
zNm_h_4Az%%Vr-@DRz|jsuA74iktOeOnjLKKIur|kwT%1oHnnstZGEjYej{kOQ|VwH
zgS%CjURaLVxNP#WK>ysR=L17F8&1n^Owi83N?J~>&U3SUquy-ueMK~lRv`_)-_N++
zS_cvYWB)ld2edbJLy6^a?i;{dXqi|8xq}<Ks<xGp7N<vivn>=Hm1%@NLkZdmLRI?;
zJELG4&a8Ax`xMSLdE2SAn9dRR#n8j|fVU9cW6}f-{KG2KjQ&IHpEQ@U$}ZDq-qp^a
z(clx*|44!+)cc5E(X%sbpX-5OOH{~18Ck8X1h@cm_1Dei@j~4zS3FlOulQ`PUObeY
z40F($rH<K0pn2fQiE;Ab8n<?ax=eo(`-oq=Z{#CGo??HJ`{Me6IaP<YpC-N&G2fIu
zE_V~hmx}=4P!30x$X!9tT>E6VR>nQ9%MDhE(7&mr0jXq`hg01)QfY2Y9b=jc-bUg(
z-D-ilgS@>}obJk*Zz&kC(0+)=*Uvo*BFf2U&r@zlr!1?Df;Sv6UO6_#G&;K0;>cNh
zC~L{3>~c<b{29dixQ=LQ?Q!YJaOcU1NGs_$G@f)G!X@HXK8WxPR(U)o;cQbZ3wHbd
z9hk&rt1N&_eJXi1)&+nXO(8g>tuXTBHXv=kBNG*NpkWe9A&VjH)%{DpVBti!<2ZT0
zp-x($^@}n{dQKPT|0c{2hKhzXX0T%B+KOO+qqKH(M>v$?mjCVQAAhV%KFbD7`spv5
zasqe(_5T?nbA4BH0GwR86xOc}6cM+;kW9v?o$E-jZR}$ad82R70<k(%kP;Y7sJLL%
z;BF>I!{*Xw(555(DmR*5Ey#k6nghHGs=7uE<kkae=4)t*jpV;-hW#pw;?xcyS<BxI
z=4qTc9JjazR!6*gfmtvBl~JHm-DYdPWZ{=YZH;(_w1~`aGKcaFTW=CAZB~NQ!lOMC
z?R4qjJ9?JvB@t&XT#n-GZHs7-v|Je}&+9D2paS%}{iK#$sbf4Nl)Ry>RZx4mWwr<}
zPa3;{po7LEx6<0s!A_a5ewb4$_2gn^a|P5);XC;C4^ZmSc+4Y*h%0ik$r}?%D9L_6
za6lc*R`0By6I~0~_X<O;Ost7^6>pa29`jR)B>8Yh<JfwD29PkM`6{CwG7nH<)AI;6
z;L5dvq`K$8GsLuz*O@ZhlzHP*srUq9JgFCP1u<4|<{6&363*IdFCM>t!IPIxk;7e-
z(67GO;n46iNa@K`)eDyqFm_l0i>FV?I7BZ%gdPK!&$wch*$%386B31-7WDa@^+)O>
z)ORY+hWR>QmyljmuaC@~2glvZ&_#K75+08=9>rq8JEHUBEpfY-55f1>4G;F7<%Gfe
z&b}yu>7v3Jm`#5YK3@k#RPnwyu^;EVdeH+R4#d07pM@&t;UvcTdtQq&uv(sM3$Nlx
z?JF_bdCP&Kq;is6^{Qg@s%%djI;~x(W3rmzFT@jo8DP=;u?Eo1$PRJ@!pzQJ`vj%0
zDGUnnJ{*q5lFwV7pD>M)QA^EDdkbMtp$*@$E6w|<k1sWhkIohQ4-snGXMpgz{{6#Y
z5uif)U|zXL3rb&8DPD%SGGuRVTT&2A0U)STzt*l#uGt+R8@*V)*)sa%JcS`F?w(sj
zbm&fzYx<{kJZkTL1;BthY2@}SfC=T-B4B&1#76(TNc-06lYJqJ_EN9P%uJ;?zk{aq
z%3Z??S{Hv;D&b^T{0${$$20T^eAG-=d)Os39SWZbdPMnrE%xJE*TFXF1@mbkcN4o7
zXlE0rrV{6hsYJkQ$4edhq0bkWK0ZL*9+CwBbywDCrRqxW+I$!0cbW6z^M%!75zp@^
z>t-dGzc}WrA5qZdtz9NnNwo865=;M<cPu-vDRQ?7ZM?puC^OL$@TaLoBo?6ZEE7GE
zeo)VN9ST0d_`3z2?V9--m&w03NP0P0W1@FeZ{>TQ=bx_9$Ws%tV_xSc@D*sg11ZXt
ztb+IKbfx9f(_Bdvj#KjLKu8@qJciex$=ii{vk+!ZT8_@UxE_ui)5)ny*ghDkPHBf<
zN8Oj8ke1Qf)U=%ot2wlJ;T-wZhSOq-=0NbU0M^Oc_UsLX)uqUYR4eC4+<j8UeB>|p
z+o8svSB~ZP)Q+NBY7u-Ee`*riRBL5R-{9!EUf~i_uDK!?C)1UB9@4q?!%6BvGc!Kt
z)pmgbn{8)b811~wJx^<B%zRMR-=y=Ekaj;Unw3WHlZ<)E*5<pnX#)&UO-yOJwwRZ8
z2YpxasFr!%uDpkQz<+drZr@krOg}ONzL^Bu%AD=cKaL3-5VSma_x!^-bo5+6vAus?
ztp7)E;788>2{+s(zfTZM$TKRFHJ{O5_}Wdb&IxyDUw|k@*mI6<vxa{Xj03rQpqLqX
z2!ZG_18Ekl&r(w~=a3|Pn@b{ftD1EE?~7%5K+^Ea{s4bMHtKIrumWWE{Aw2;Qf{L4
zdIyfilVEx6lLwrUQh_f*dvCUnyeEUP`K3(M-yezvqWD}9(b6xXT6=BJ>_{aR%I0I)
zVlDsVs2Y>8_W;oXuHYdOt%*)YMDP9EfBx78MOa4*yy@SUw)eOH4Gy;zB^(^+jsE!~
zfIH>CzbHsNfW#Fv=iGPvzr7i9ViMc`zkct}E2P1Fpu){&Uj83HC#my3@b4>y%e?`g
zfHD80oaz7gIhpe>%6~5A@5_294_|^t&dc}z<L5_amQSEB{$H;^hOoL^<?}Dl;{Dgp
z|F4td|BrvM?8ov4df*n2z4)Utpn~=qxyA6=WwMfOT)zz|D7yKonK#2PhI;(Dp%O=c
zmHxneAomLK36>H#ENv0kZC*9)4pO5C&Jw`<CclSa|FFzeAX2$rkvxrNPl){F7`ryc
zU{zfdDBOHx2%Vvl{c_~bP0iGTyHVM=a1w^P7kbc`sLZU*yD^~}%*1?2g<Th7C#sjl
z{(Sk~XW+1%NF+y)t5E>2d3!7<uj4PMXUD=wz{B{f80a-5B<Q2TNyBDyeJLdnTdwfu
zH~TnMCW~vYD!!L17LxssZQ0)KzL^G_i29Qb%-@gHKexIM(6|KOe`THjSubHXpiHOz
zT4wX_1p(W@D#!w}C;mvE|FvE-Rp9eIs#*9~pbJ_`Fi`<gz2NkdfBmoM^)CQdNQsrZ
z{-0;}?>D)L0x~zdgiinG(L#TdAcuSM|1bUj^>003_n-{V!{OfX=b!h%KXTH_QkWaJ
G1OE$<ObZkM

diff --git a/docs_old/cookbook/images/knowledge-graph-extraction.png b/docs_old/cookbook/images/knowledge-graph-extraction.png
deleted file mode 100644
index a24b1c61a9c371e4fdad95dd5ec9fc31cd71c9cd..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 33127
zcmbTebzD{3w?4dRC8Z@*x)CI06Dr-^p$G^fC6ZDW9Ri!~1}W*3kOm3q?ndd(cdm2p
z{oU_-|9RsNKj-M)d#)K{j5)^hJYxl@D$C$uQ({99geNB}sSZJ?s^E|FEllt)U&-rj
z!GBO3)n%SSMSav8;4kQ}o+v$mppr10%NH2n@3-w`^&BCHj1m5WBB#!<3qd}Ya*|Ip
zU5)?F;1~{TU5j5=*^)b5F|FdsE0Jgn??;gCj*#5p&W($W{(7bPuGjv;)_Akgs?I#O
z&bM-_J04@48$;n!DJu7^+4B!kX_YMTx1yO`4RZOL54YCDcUs~W9<|^;eE8^>hgk92
z5|2}#UwDs4pIiU?zB7+LtL(dXz_g%Y1IIIO_<wSl8NiAl$oVNMJm2U4zyA7Dn+uvJ
zBIhzp?TO(YielHvt*=k?x;}eV>wd(hrKve^v^BQT9zykQYm{?udbKxU=x^1WvdgR+
zbH$s92>mKsC6^f|ayK_OwL0g&xt<rt!+9FadxnrtwhuZsM4g_Q`B^y`AvfoFLPJVQ
zs^;LpbG6$-@NRor__6nOJ*)6r66MnEnl;+Z>6%AQe<!0nJU#Q?pKQOoxjNA(HvZAs
z(IF#73^nb2l|b>C38fK{_QR)1r<VyG_?;ocDlVSFWmL=1xJmo?-MYlp`BBF8)y0qx
z3Yz8aeDiRXoynON<kS8HY){2vc*fPm$<PNhEEZ;FKYUsQOZNPqwjtZZ;aw8LsY;tR
zB(+L!#p@Ay#gRQEO&A-`bsQqAL*xBekrA<mCt^iMTbn5x64qPVka)jG@?1g!!oa}z
zJ&^WnqojlO(Ul?uy}$nf)rYd->e!(B*RRZ(I+vkj(f5*c8ZiPkld3zjKiUrCQc_Z4
z)$4W|E^-xOxJ<wO4Hen`+W#)t#QRE15f6Mjy5Z*fJ&%p+`Jok0i0fkO?emSCl;uY@
zORbzjLb1}pcb~j?@#1j)6G3+|umk+tB+xBAJ-x%TwG<6^_s2<(o#bgNwrAh(kZPEI
z!o(vQdYR}7&P~Z{L^M&@QJUc+qBk}shB=$s`*Kg_Ol!;i_>bJnj?gTJrFJz-%UlFD
zq)JT00?|y>xo~?P{+2;TM>BDuR<+H|nJBv8QSeGCdhfMOCi~0exl-;2aW~u<iU6Ci
z&(6OmwN6M#c(@#?GPt|YvZ4n0u*s%E?|gHr)z#H44_12xWa3>P+YaZd@x5J^<sn+K
z%}@yanW2f;ue2WHnyhim>9rE4jf;Qm#3m&z9mpyVLBO;h?=AS<8+^a>Lq^2;&sUjW
zdlmA;Op14gbP^F8gv@G)kQ$pQ0SB8X{tTJ0?2VzEAq^E##xTf-_Q40#f$u@&+=J&w
zTdWTrJSpfh3)t~b;5847&BnfcJ6z_2`Kwn!8@-$cmHr|(9D9=*Wfr~Z!j7w|9icQ}
z_5(QJ>6EQod~F>aL$yv@%Y=@IK@w6@sYqs(Xqi+2gDTss<z+J-Ui04R8b?+pCd@5G
zl=WL+>+lvtdasYG8UqC1D<Gr<9aj~3t};ig;}b)Q;})$3GBX+SH9QiL*9!#RcU)Ld
zeJbIpLJdkcy@GR@1Y)NWw0Z3=eS3UEkNOexVLtS4Cd=g%!}e)Bqm#YPt@O(+!*y3t
zpz_<dNqn%_IXE&&*|q%%2^A1Zf_Afr5?KZz0&am?=fC>-L+=rbCmrI~Y8)Kc(BwyS
z>?X9U{T}Yab}G98lbdrLw@gYClIz7F?m$k3CMVaRgsNG01;cl?8!Is;**YekQC=5Q
zbKUgN{04A5*Vw8#uLR@wfB9?e7d|4?%@<n(4hI#uA2_@=O_VJ(tcei2K7HwyDW9X9
z)av#gxbre0v}*=T0D*2|U1rurG+AYrf>8f*^7j2ol+Ht~H4E3*c4zx5rrb{|UbOhC
zy?IkmEr16h$iWs0x*V)3@nDp@AH8<V#5U16SQEdYAor^iZWzBj8Z#ju%g^gg5jRn6
zo^_tG*}nz35`n|A{WGK@p7jNXe02Qn7h>jQeANEtwIE_v{#iOEF0N2o?@I?abO?a~
zkKdZ1T%UZ0+ZXKg=w7*MikRmz;Vwf**gf*0!f2O2dvuBQUgXjl5CRMSc1yxk!|qR<
z#okf>D#Hb*WQ$+155jzyr@p`Vq{}S-LkgF<&3d$eFhLN_=faxnh5)kliD8-#Y98q#
zF1GD8D7)`Ah%Xl7p?o$?xhK5l*lz<t)AyuPp=k*@xd2_x%d&p4NEv?+M>(Y#z6x1U
zg-P%EFOB5utPrN}Ep-HG^?7=DfZeng1iKkO-j^iOR&>4Amjr$z9Y)KkUA6r?!w-R~
z?+ijl$?HoHAeRZ#!M8M?K1B_o<d>&Kfm}YaB7B%y*>xOCGnJE1@$3FhRnfZ*5{iv^
z^qoypIV8+iHrt>>(*$5v@tp_#;woFCg^~Q}V5~t9tAY$BokC-ZVnyEz7}dL~(LRV5
zapAf;NV!4sL-tr;R)&A(Zmz4UU3Eo`TJi%9e4A!!og!r<`uh@v=>_Pti%t77)VjL5
z-kr?2^w2{C$cJCB=AU}3k=m`Kf$1#n{rU5TLFM^Rn-5dyRe1EakEWfg@lYg&o}psQ
z(}A0SlwRrV>}>zmZ<FhN@)F^PKK#=H)ggBsrWOQRZEi#HWZAor;e4cuSejfU)9*rq
zs%1i=A)fH?@KVg@2J-TO6Y*-j34%Gp!<z1k0nG4L@&`+x2vOKEi8w4uYV5b>2Ohj$
zl1)x+s4lcmSg35N=5$C{ti*tn(14c_Gcqv^mRk*(g>?4!tJD+ve*WA^4;{7G_)g(h
zJ)pNWm~Mxfj^HIR)H5{ic%6vo(NjKr4tZYgbyVw8Q-0YQSo)m<?yF9E$({OxP~CF~
zGQ@%h-pz%Ixoi*DGr|ICPDPwH8NwjGX<)&YKR@3-1YX!p&!Ej278)82wpjp}IDY$B
zR&LJUj~_p_(z@t2Vo?W-5if}zynXu?oOlS8;2<0_tf6^Da3s|JJC#ctr1Z3I-JP8h
z`{>Hea+<EL)f2KLD#T0M2xtx7-23$cDj@1_*ori)29S%lmeU9ktE}y@t+ZRGy9S^&
zjC-8!E~Mi_eIG#dKyh`lG@job%u-12bCEZ<uvq-)V1K*eJA#8F7#jyi>c=xbd{e5}
z2G5JY$B+;$5sQ!Xq7})kX-oCbYKNp8a4iQt_F<<Mb=M5`F|K(lHGFZjC3A87x8;z>
zC-ONB{B_dwk*Tsb6Qvkk#MRVQ!nW_Fr<q3bW6*qFf*@)6c`e0z_RY)Zv&k4iW=}O1
zDm!cj{HK_z)F68V*d4sXQ3Jc*AXb|dzJB@qO+Ttq$W*F3nyGEZv^Gjpi9H^P1lphJ
z8ja~QGrm`S)FqB$K6S1jP+5Ycp;VinlK=HEiyoR!J4mIT=R`yhxi5#ytb^bVRyzk@
z7Q8+p3#_6IRAElr({T;wo9z30dxTDqCAH!L%6NcGLU5M`xI0)`S>=!@(|A=%wi&@w
z992e;JuaA>v+GibU~46Zbp4Bmqwc6KZ8nipxr9|lwk^<rYwS<7TxR1bt5$!)Qt3kw
z?Bd`+tA}Q9GOY{}11CND$sg~lbRS_ul;Fab5S^}f(<mr&ygYSKTf^zBvYX@ThLWHE
zisi{>6g)rNpeLKH|64gFNdx`y1Agy*+H&vuxqLF5s9BE{k(xs)%{I|fTuYU|Y$*R7
zkw)z5P$MubEQUXw-+ELvj8@bPuj!F}!b+uet4;ir0K|t2T<mqTh6`fGVRM*OZLN?8
zq%;iRvFmZKl6eZXXUkrx@KAoqLqHcV(6Lj=xEn7^S=|<uKI#<eGlH)LPCcwLZ?{R9
z5`xAFfbHX%c<f;m8voEA^`G@PP}&^Mi!4BKWwLFVjt;2~aWue!E}ml(H2E&H_zeMz
zl5~SzqKi4jpVx;BvDZpm4(@$5T7frr|IXCuAW>4i9)eTD*s*@4m^{9}(qn#Gf@&Dd
z{PFQy^zQ^9h~qR~?gvev#??P(GLgkkWt`5gu>o;!t&#vb`UH#{)y~Ob;U|XL@xejM
z>!nn^>B^drN45v!s~D-*z;b-#U%W`Gu9&i1Y7begw1Wy4&4sg-5}2&DTvjO}*`V)u
z^r{dWyLQ1>#`O95=L{;TZqw~Gg;aSL5H#*f$O5r)bBEEouRqYE2bm$?6ymN#zN_QH
z_wNt&=+9RZ>?{XTpOj)kqtDT?CDe;U9)Yj~#-N%Wt?310phCKSV2n?|!G5<8h;vv=
z*3{IH+_`h#8k%(3oqvLji;Lt>FK#&RE>MAdgh6b(d->PSJCZS-RU`Yp8|uu0K)=o0
z`i}Qa9S4dJ|J@I$Ga#M13(|U((TC^g=jOuLQ7I`ZAajr6hg4f`k*phgc+_)f*-U&~
zzI_-tC3b>B?=%NNj^LL@b*_8q6PV4_2l$75`=JQq=Vc&&1ap^$KQw6c`oTd6@<jA5
z;&s{+lx-a8O)*qx<g3JoB1Q&=PNLr+3pKkfF*oe_WJU%`rOe)ys1m!}edZ3*T&Wt6
z5+8yE`tw3};6k7;S(Lr|p2seV=+Ur7IxeII4!QOSB)45ezkkI&GQSPQ)~e)QU_+YC
zV9tYGk<586JF}(Inudmo@D0Qd2|2Ffo}Hg}Q~nMhX3o_qHU<-g=P55I7oPPk3+F2W
zC3Z<H&3I5^<`pRSNY;lzd1D^H!p0^a4M`}?2o8H<LLjt1E_c~6#>U4dTXFt7`Fo`&
zMv4M@{d%rJ)CdGJ<jJ3jXDxmN<mS*1c=P&X?s+mOXkE;r_}FZUrEiNs?tbXl@0d9`
zm7@KVJ@)1FP$BDKPste<=n(kxy1ceFSriiK&s%&&LPSK@4LzqhA>3vUzQ8|u07;-?
zJU#%~eUJJG_`a&3lZhJmS5XkXYa^1AxfCRUMMg40xgoX1xfgeQk{7rT10i;eSJCGu
zJKWtiljUjq`)}1SeQH5{_FF~#ilJL3j5Z7Fo`)(D<q(@*HI*tZF0NFP4-S5)KIf_@
z6iz0n1MCdogF2U;XvXw#dYNt_XxOi|Klg&j$M^*{!MlagLWALNTI{u6SGBn>cSN9z
z+>`dYgFcas$qMV?dSS6opS;qJK5_k-Yn($O&Mq$K-B7<1pHOUz1YZ!Hu>L!H`}rmm
zJv#S4B^_qBp+9NcBAa3usWP&u66D}MGYhcn{Q9629XeJ$p>aw<9iTTv!>Apov@sYe
z#mG@f$=;r+3*v=_P1Mu~kJf)bL%QjGDlO$z!yKRwmiVwrBjWrEiI@R(gKRQH=?`yY
zyJz49nU68JM|wrxt$zhZN{i5qhEg94d&<sKL5G;PNi1i<TEUZHUqpxS(zc&%HejTJ
ztV`m+b?SBo3L0i75ybbomLLy|Gz&_--k$#c7{;I1Mo>uR|9p8!US573z9oNA-Fk`A
z%cQO+Uy~4{HjE}V{3pcFZ6MmTJNBNuSZ}=A3G3dw({Bh|+Yc9aP8i!ggD&p=Q@JRK
zxb0_h5FTyM7(}OPM@^Ksz9P#qiawHN5f6W|*B_R(&FRU)^7eMta!hyexc2n~8|8(z
zz}};vs=kbM7Xlqmlxg2u?mRGBGMJ^(#dh6}S-?Nmk6yt#m;AIxo|tv7%+xLO0UPU5
zl&XF#@)^b}>|CC?+tXqH@w=rjH^IR;&{La0)}t9owQk)=X}o(ER!<gs%2-~Qa}+8m
zq+;w_db);~NNfcC45%tRN~(`0dPPuhj#`Z;E-tReyYmuq2ZaY;VRg0pA2b}x(ZaVX
zbet1h>@tfiPAXXid{2YK_$*5~#)7!N6f5S?Nu{N8^6-7H09(>B7Be)(wkd|0$_x%*
z$m?_gOOLJ=)XiqM3)pjlLwZ^(WVGb1CTjV#Xi@|JW7Z-r+ld^6P8)+#(M{ieYRXUF
z?~yiGVxh2R{D|RXqJF{yi2^hr-ue2{4Gg%J{2mM_eCdENACS=BH7!T{#MhXPe)G0T
zlGf7ye}?S^We+NdDGJD`DmxXIX$L=+PoyzJEZ12rk}pkJEZi74kE1*A<aIJtebJ)%
zpuh7Ct)77n_m&zk|FNJlE|Vg!v7LdmPaGfk+fo?S>D)AaP0KA$(S2s>)tNgR9cRRo
z2paa3pBaM=QGqRv0vrY$no*r|DLOw*b~!(VBpR9ep)0q8*?cf#W^$+XbOh#cJ|aee
z)lO3FUW!HepHzfy&Re$GWLcb7zUr#|7Q+P;;vTswogDhr4?YF<tPE|Y)j#}1WD+lx
z_F-n_#7bI>;PYTX3=wO_K&V~ji%qetZEMIKq<Vb$_4Vf=MK)8FrRdRYk0M#OZ>gQ)
z5SmfCyuPO#?ycFoB>Y<L(`4sgV+)Fi=xC#lESoB<nyUQ<P|pFem3DvSu&4(GX;oO}
zmmM=TZD%${4s0TYM+Nc1(X|(TgAu<TN2KIqAce|ADrjoIma={|jb>KW_FXuNqj`!<
zQ3|l@OD_?<Vm(xu@O`HQZi@felYa3V4h2boHpOs!s4<oIvL=mk=g!8^$R9gi`10^s
zHb;SSuB+T9!N{bl^wSYseO|`sK^*zq<x)@atkW_#lItc8`8SQRg9?ExPUd`4=zglQ
z_UKZ@idh9G^xu+o<CS`(eTPav2@o~+Oq4Se5cIP7XBK$V<d`j;P!P-E;NI42{-FCz
zzD=I>B1Dz)@TqT6B#Tpu`UyYu1H|cgr=!<LT;+uZFZreOLo?Mbozp)YFssSpxc#z}
zKyM2UH?I^;9!#vre6nR9c=e96jmt<O@223Ub|nf6#D}Flq{0I7J5Tgatxp!x4b73V
z*{YZ|<z}|^Q5&76@*V=n=%j$m;ldIUbtlhqisA=BtY=B~dXFxHyd{Ss`VSMH=3O9b
zi&54=i91jb7>KP$_j=@86vvYa`R-YwSF%iWmJ6EJQ(33#!|83BNs-kSa-YbqKWW}p
z;be2P8kuJFMf!JMgyFD!b9W(hOuDlA`ncq=y55o<%S`=OivY>D52W5jV-@b=1*u=z
zW|1^<X;ocZ{^ujN#)^B0ekZxivMM}(%N4SygQScIyPF{QENEV?+}@|)ZuL+E=E@YY
zRRA?%5H>zt4@&uKG}iqj43fUv(N*G7*N+VO`MdBn_PhDiunEeeWxu^U$a+%B@}@t#
z<+0H>qnVsxPdo{3kV{z1R69J2_P1@ka@Ir5Xh~RN2|=}5Kz`Ma4D%=dmQy(WO74?-
z_qbuj29Fv#-hR~;`NO$A2c(z$gREsC#(o;h>l3n7HDATnXgvPeAFs9uB^ACUvyrNn
zC=>tZd72waM)KQ*X4_`-CaCM&V~Q{Nnyy8!IC`pVEP<n`@w~`*u<2X(60%Awnkuh;
z{-Jr#{$emM8++MWxDwCRviK;)e;G$p9kvgCl$}bNIpS^U3!;jO$AjtampUwk>S3kT
zrgEQZ4Y{1gD`cW{6oAW{8*uOMm+t&t6gUe*k<fvcyFgsZ>Bs6|PWRhKR==J%baC_G
z^kwE&G~agvVJ$wo6n3tH+UtOo4JWZ8l!rfY-2Jo(;=fq)735Dx$~$1WE&S3AG%EKY
z8(u3xj5;Y$j&mm9lk<59e;_N&*Jvqk=_qk{E4o1WY(NBD`+C(1o$+tG$E$<oS>w|?
z^2z%Og1D993umX>_`y$yCxbl9!rQ#xbac;uVJ)|Rd2|{VR<-HmN=(H(M!x+P;|EAh
z;zvr&6+m=#{u|G+VfccL(RSV`l8$w{+cdmlkCIp}NQ*pDa3ccK_?qvbV)TTsq&z2b
zMPEz(Sj4((?ZakfSKZqE5q+Hik}9sM*?@%w?_U|nN2TeH{v$-;urnI)G7!iquzdQ-
z>TF@7Kt4Y3Pb@8t97ZZoo(`t+Y|b%$isn$tJE$$_{*-Six5|L6Hs?0Z4GuBj#dZoE
zi@T%fb<s~r{nBYakGihfzUNuq|8Bu}e$USpTzNe;c|Q7hrV0IuI(#(d_Pc|{uPLn6
zS*ik7_b@{RZ;a#xrPYSnD|7a{2@2EYh-1F?E2ypFSiLnpRUhGU(UNCn<n?ba>dYQY
z*#C+#TQBr_og+Of{H-_Qs@XU))dk#ZWv670^u}c}b8M_GMjU&Ck%TXvXkK-XcX7Aw
z?wH5X=sk5XNU$X7=8iYjNyy=@VSm~sZbsJTov473wW4WktlU;VH@Ab^anD$mVrJ2@
zOY^9KOAE8lSgHi2ln$xh&YrfsBp$tb#Tq_%Y^G7U5MKG>b}!D3vdz6%Db^WNR_A!p
zv2d%PT9v#DN*`Mmh!WHR#rDK&evOTdh;J&SH6?)$M(z~1eKaE7w|s_V>59y62-{eo
z$zvez;WmiMX~k0&^;U~k-?Fn8Ql`i~zvfAa?&8+f$jt29Exews5WP%~8lLp(Uo>MJ
zoa$~YJI+0@8x`BuXdn9o9{5iV#m33Wk8yhZcd94j@j(jFimP-1QIRq(swwjGG+ovo
z@qOVV^HXCtR-SChUw;<J(^}1`b-FnoWl)iJ@14+-+Lf(gu?0N}L<Umr3%|Y*E@DtQ
z&=wury`@`_nNrBP!lqGbj8}3>u}ZKr-zGoGcy>5x*sV^6WpL@y<F{~>&rdjX8=lYS
z|C*1I^=y3&Yo|%0ml5qqKv|dk=gTP@Q95<`<w)z8vf=Z_s$>7&!gE1JRzmGY+hg5$
zjUzKgnShF8r!tw(5gyuNGl79+dh$QhEg3UY6n*NiqP|k#AV2?!n3So1ovSMUHZp3Q
z;(>f>%HFFBwhxgnkrjyy`U;otd@myQ>h15L4P;nc-h0Lj4P*jZW1z-yO|quCI=0qn
zXoFJ&!hg5=i(>z%u=YCNajoy&@gF^q@e&(K;>Wi)&)j-K+r%^MDL2yr9ioE_LLcxJ
z<^k&$JB=}nkR|Sa<USUO7S9o1u%HkfO-%-cRZm<A6%b{zopxr8qW!UPwkpQ3py7qX
zjiFMr+5Z}B^6F%cTS1apB^mj@1{=-S;nt&j?6CNm14ZIx?nOjx7|Qx>a0T(*yLWFB
zBs?rPheIlh$JUUg(Dt+Gn7b#O+aN9c4AP8Zu*;>Ahk#826~_N9vEG_%Y~&!M5q>NA
zzf*L;?3e|xaC58t?+@miKY`T(Hs(Kb_i-G8cWEGx0j|Q;UGIxGT7>fb9(U?Qh75t)
z@&11L|I`dz=;-L3L~6>)p`b(-9rl!&$$<`Uf$Hq3sw$`!Jonnj|Fawcff5Il<57(1
zfKuuqg2tcCxD0z@ON0TiVEre6Aak|ywclNxJE^UOe5rHVV(JF9o3X@DwZksBAmzZ(
zWUTh?QBWV^6yAIH0nO9i{(AwTxw(1QzonMF$<uvwcRyOFmdCX9mL46bt75?RdgfjP
z*K(t*lY+8Y&=m)Jsk;kQcs!ie5ahIEjuZTTo<`IiN$auu1lR>TsDy0on|3D3a##+i
zngJpV7p)zhZLtW4dti7#&b_nvGhPBt0In~=gQaJ5hCOdd^uyJ9FqNQkw11_|6q5p^
zO7&k=XA8e-s>sQ-Bf&rA>YU!c<*KTNv$Mb#U*EveurPuoac^T#S>l7rQUnk}VDR5E
z_rR)jGq&kB8_-f+VDw?4&ZXE&Dfj<HEsfuZa(AbFNzfN;X&^5YyVUSPF}b=QfEu~J
zVE$GBI4YGyCMI$yK&jc=RGVUG5J1zC9=q)`z~$5EcR^Yo#-mx6Kdk}8!?2~IL10_B
z_NbcIe`lD7tp7Nf1=aKOO|fQA06O{zgL5?)%2A171gs?ksME8zi8mWiAuX^awlE4Y
zTw$%FCiecA1MEA%`3^%bP)}~RdwvEWL@ZpB3s{eG=+WuJFbK+eLM?ah#Vts)MQS<K
znw;A>Q#t8z7p0no1yjgfo~NVG;j5GF84@xwCTmHUeFbQHHzjlz;Aihb=I*ZcC9zve
z0>Cbc5%N?APzn{OWeZ?@VYcLDM=tC-(#V7a1r!vNQq0tQSOm10O-&L2U+VAA0f-qG
zDHk3oZ0>$~db(N2dSTUU1Q<!J%5$3-jPwSK<kOt?%x@hqY|;F|Pd{AihKhM!9gXq1
z?mh?GqyP@;;k;mAPKp-*>9mUsSrwr3RNI4TasZkPgY$*djg4tV4+ox=`iksJh;!`J
zU;Hm5Yf&&b?$Ij%zWj^GO%2;5T|g1cKDA(C5?SBBe+BF#tF<JRpl#->WH;$isN7G7
zCPly|zJL%AnSqy=mu7g-8-VTjU^hC{U+(t+bo6`+B$++}c~e4Fo(wAgf_FdjiyL0+
zCnd^c!ni3geM;M@>2``LH`@zHRUFvE_eBt61Psy=$*2Ad<%9vGv0<qrETQnA+;--%
zJHUX!b2edy?EeGa_yz0vGeBB#gu7-;044~GDPY^vz8sJ@^%xq4AxA1-T#Ba$YyALJ
zl?oyYNZLR_P*lCJ?U9C&K{Ovv04u2mB?#IDTBn?j+O}VO+OwGhKxxd1tg<qhxpC>I
z;xD!T!nm6RV7Dvg1*5C0h4kq6+CSwXb~UG+L%N8UDnpiUr@jEVm?{C_STjyfUDa%7
zYWYFr{Oo?X{$LYIo(FN9?|Bmh$oy})9j&bcAeMkY@CQIrskE+a#ojm9rnR3!Xhq{I
z{|87ef7yYrwraHjXjRk0+o^=`TnR3v0Y^#+CLsM8L!ep1-hiJ;C3uP#94O4MtE!}>
zrIo+iP}j?!L|p=09KaG!T5U9c7~#OHh6aS-gBV&06IV|0JdR^=`7<8gaJjFrIa-*I
zi&h=^=vCi9Mwpm9@Dq|Fz<beeVA<b+k#jay)-)Mfx6dGeF>f|EHzSjhIJ=>31_p-D
z$A2d1yOTs*vH|uP=;7f3Aa6=64Ohr{Q84xMXLK-C%rBy?O*8fheDG8`BAB5LFx3Gl
zLpmfKFl;@PzX3nIQS5z{Ux32LX&W@n6jELIHUR_bXeObRphw(X9+;B9r#H@g$jhrf
z3K?=UGD;Q1+VQz;8=eE4o)wTSEP!mGl>5<u-_zYqG?oliQ<?<jG_>GvMWF}*Gw_i<
zKWV&CYVpO(xV-dGvl}Z)l1+DAP=a(8zVHaU)xKQ`%str}3j+y|jLV#7RW2Yq7ARpd
zL@*by!9GCMaDVd>^B558fd-u@h99OVtAt<fx7vhHF@WW)f#u9Y0J+TqOt$L5s+Q3=
z;8#q_iLxM~+;M_lEeN#RgiKK{33S50fZ!=XSQKI&LRs0_-IM`%e$0xo_p?u-#1B(s
zRXn!V($knVoTxsI04TFhWlnbeC}2CyV$Gn<x;By$F8Lt?CdS%qe)&()Ry~32#b8$I
z_}8QUSkCge$jE043k$J*XnqTotu}#E(S$IC1I<%{9ZJ0a_;LKK+Zv5tK4rk7LAS#T
zVLq~|2et~E;mo$y)2mRE6JVKzy8cX*^-8CoK8^y=xI>+=#4~$9@?`>cYU*wkV8J*G
z&@krB5(I40t%kC@^hX}6dBl*AO`*TvrHj+ZR^;K-YC4T)R-8iN!vizp+69ckK&D*e
zIsh#9Z&iAJI4_${{u0ABU-#F(Sx90C9I_A2E)`9&vaR8p$UQ1Lf&0F-_Ec)jD$~Bz
zoDbYkE%iYZv`+&F_itd)Hj~F+dgC7X&3awf=%M8vPdoO*_G4c7`aPNP3+f;<R6bGo
zvF%$YyZ#^D(G~#}0KmX>;E`SQsjob8W&mn!)t@Z7FFVrSg!miuR}Y#-%Le3S@Eqjk
zcL}mRK>5wG-19tCI|oVA3gG~|E!s5Ma2IhEsqG^WwWk3g?ZdR+?|mJqU2Mz=?hW8A
zQ*Y~!C$0Byt>dM<hrb?N9b#{Q0-gLr`c^uwDaZ3$lL0SLrp=2l(yMdkgb`C(6iyM(
zb3VX}_P^1u0@<W#NRC>DbQj?Jl|eWeXg$T-2Vs@yBA#N8Y6I5O02AK7Wl5UyQfzE<
zIbaI4>NbeK+c<#Cq1pvZ*jNLHQLQHJ131o9e~l%VZxwT&Y*|D0bHcv>W>w8;p8!E8
zUsw@<Qn$XW2bB6dKw-IMzCPL)ob!nMFIs&`EeC|lxI(ts9F=?eWfpiKvw&3yePCAe
z7N>Z~6GTuzK)8%-{qKEQ4DRGy<$|WN-I^uG&NcNU#9qy+W9ahjCPx5#+CQtBaZafA
z_W;_SVZcwhAtjT@)moF>ll}TBQbff3q&;kF`qtw6g&|jbaBjAL_B!s7XOO`$TjpE+
z_4(FfZ;T)t=1KdSe?Mu;BYv?Q58E?pG@rF*epQGTC?^_H85N=jz-_x6&Q>3I#$H_i
zlwLact4tHkpV=P?(cC7smG_yb087=@-ah#C!82GBAql(zfxDB?G4v$bWmPFrDBRe6
zgAx1It*<hvtkekCui&bcyXSpbf>6)W$WdOb-@A5(=FeaaMyiR|*_T<@CA<=i#w<YX
z41gK<OLn&jIk;~XR$2B$bMDJl#F%)+8G9aSodX7l9Y|qvV4}Af5VTfpfuyGNrV90p
zw~B2;@<i7b^Byl@+(e(+)`?&5g)OhGF@vvbgN+;Tsjo+rC=Uac;C`6qlC{n1uQ+-#
z8}yEapq8;>6Ye6VvO{2=K~Bk-c$9#&_8EnL^cAM3=B|$g<&~AAWxl}(Y_Pv#idYh$
zWSPP~iUKjfvz;F&$#?fe-X9kr^f}qs0<YJ+ul;DfPG10NMmaVxB}@R_2QMxySplqd
z-(NU+c1iF8@H##8stIrgm^nPjN0T7pQlyI;Ip&56c>>ri^#$w%LyGnJHw5CB_rtW8
zI89H}513OtBAz~dx~%qmVqyYLLIZ++j^z7N!Nih9Z#-i*EGc3ML-jF+Cm#TQs;VfS
z=W?a9#41J`M=5Aa{f~HJBnEiSAYh)kq~i;eewP5vS&-F0PO>|!s}KcTaq1&b6p@f4
z55BBb+0I1B^j266Ni(PczMtfx%<3qfwIMqINOM2}FAD@b8_Z{<j<6WPkJjBEgjU%f
z?ne<aC~(e=vi=Kj_^n6{H@BKxm3Y1Yz=PywW=akh?wy?EbkM$6oc>KtLLv!dWicju
z98ywJi*_p_YsWitD*sBb&Q4GL8~OS9wT6Gx*S8|qLC(Xq38mTrgL~)8?Z3K}99vvm
zEX#H`tabNL>%Y`u)MYDxP|IrmgcUS(h=qGs0$e(&Qz7bm>w`e?qaYJXO_*Wh;kOtj
ze$CSOnr25&(D{!pPe>QNU2tUE;-abAs&%^10y@l7r9MiewvT9y<}_eB{$17@AY);X
zt;-N`^0p4}M3;d!h`(cXE_OU^BcS>Ucbx{e-QwgoAQkf4u3o}FQ|vee+MbQevx7zb
ztewUiV+P^+?<^{6$wCf4SKlR0!noow(Q=z~%g>*Zbd~`pg-cqk6phaRBok;p4dB8W
z2nPgWcvV$Z^>LnFnFXvc44)qMP3ajj!osJ%Gh|Y49^i1^PYcv9Irub>f$(dx(nbLm
zFIL#j#OqhQW+b9Wrd+ZPxqG9S{Cc_dFG!W%9c^lzgZe11pg?h&u*16pkbiBCf18Kw
z>)Y^Ddhuu;Ybq*GdS`PcX2QAvA<cb91qFq+ITYv|6etoh+J*X3IOJR;=E>=cgxS9H
zPT8*V!8skFw4(Ar(?XZ$z?`^z2=0HFuFIxy>n=&#;@h)s>!JJ(@{^z+m?#o+uxZ>f
z!R;F|2;O0FTt4RVAbhNn;t9kwi?wXY>Pr<luE-!6hc`2OW2y+s){Iw($F0`9dq;){
z8&{yA46>4JxHYXd>wV*SE`GCi<Ik(2G3vZ046UPqTp&rv<Mg|gT_tebDpTwN5PH+M
zvEkf7JThnA75i(>vg_-j^z^5&CNhbqYjN2>=g{?av3XCl>9??)OIPgB9LVkQcQ{hj
z6shXzv3jin?w@Y+Qhz^}3|fcfTuJkylcNQoGI?b2>&`{ZldC=JcAW51@yi7QSh5ko
z3$6d7Is9&TVKvv_xrq8YxADjOlR%N<yCx=W<dI`eD32jaa(8hbd^0n%*a+qZ+gn|c
zSH5^u7PT+KxntM(%OHtHc$b=Pt{QK2qWvxUlQ|Wh6Br+V5F`*#Z_I{L;VNK2h}La!
zLnOKVrH3E&TnXm!k_b-7)cCH{c6|ko$v8fA{t2+`AxkUR!Ck~EGY{Ne3wix@6_;o)
zwtD-vlm*S_8jch~N1Bliob;ALN7k10>;%vuoN!XM2U+Yd#s!agEIP%y;;~o?8RG<w
zeO|0cyUYEpE65tvrvaw+TZo8>0Z0L|63P8#=#o-eQnHisH&_UuKo8MDl26%|?|#tU
z+%CGw`{)s@2T8{bDFgaQAWlKDyi{U=JfDVP4dj9#3AVww(O<9?qvIA1&LacjKw>s%
zfE?Unlvxd~9zX?RN<(jt>%|G&Lr<?IUKJoN;Nr#mIjPd;hjWN>2n!Y{51X;>#YW`#
z_&^LQ%1th_qWhin>Azf-U%*cSnPuve=pmPR>di5-Ay?=e<P%Uxvon3sr#yA&>L)16
z+Uo(g2*~4q<O<5I#eYBFCN>LU0TLL%o^OE@vAjjV^76c|V%VV09`(x;Psp<gq+Gt>
zb~GqZYyIAoVXe!h8y^)V_29t+K=4m~>BLTbu-R+lW`^Dsd9Y`Vu$ko3YiU~sv=n?a
zO-xlvSk_Nb)(OOtQ7v2j?s-|tC&RJ|1{JQ>;v<lyNXGRygJcgBIOT9=BAuHPI!rF`
z$@)3GA~@Z{eEv5lyMsLZpl#VhO?W5D1q6v9-xerU6l8h>(2;*fwZqa^1V2i{R=COg
z_7nwTA|gpJ8mza(%q8z{Jv16pQ(4N}@kfXL!UOsNLhv5mJuyS1+ww+t6uWLiYQSzZ
z7ceQiy`T8hLK4q>aS6?#0WwKFh~Smmo=`+icFS0b=)tc?rM^L3DCM}|Bxr-};?^os
zfaOHT#K@pw;ggtKQo=H4utZ&?8z+k~ap0IQDrD0vqXiMPZ{blyG8+fo>e3jZLb!gn
z!s8MXw6d~-cMz^(Kjr28>jDGA7JZ4V3bE~I18hL0<eCdo8x0^GY~wuem?5^K{#7;j
zYzQ`vNmyS(0EC#B;fNBRm+kl6-qL$;NrzIic6)@vkwXm=`U^t*d>U-QT^?5t><Pvv
zF(js7+1rP=j*hG>EV4+qzs$3>U+mJbPF}eAErUcvE)22n4;*I=1{<){i{FYmZE7;8
z*!stC`3D9@@E0iJr^0fS-G_H4cB${Oq7QHZ@rr9KK<wfQuRbq#d17E<Mu1C;<p4yz
zZAgL&*p%ztBh5Pk_|D$HfB!E14DTR1@IYGhEgU!6f-#I#C3wL5i1XGiVjLWtXZy=2
zr_^6QfA;51sRQ>b0Fhi_EgdZbY^4u9*W`xRHfLPs0Lrp!27<)yo$-}9N4;5Lx6|Dh
zK*r|-rvD1Ibw5=TpI_GE58U>tJ&KrAk~yq(iLoVc;H(tIcDnir@JW)YpJm|e20*4K
zNr7kqd7&k&pZoRS)&{vU9U4OfyA!Wc;(+_=fGD=+1_Tg_<W~nX$Q1AE8u(!#r<g|>
z9&!QE@+;#mL;auD0=+UbJm7)J_+X*7_}1c9GC)KR!?j2^1@hU+FIBE6P(K{J1Y;5t
z-{@ahjemb3n~b=t(g?0OP2^c^rBZ?bh_kM=aGI+u>M7`X{!R~7CCiu?+n^3|<pRc0
zT|&&P{1xE-EC+?V^~c1;C>Y5>V;>j!(_C#>v{hB_3w*(Z`ai-CIqnnL0a6|$tHxJ^
z6_nBfRF_@xXPf)byfC~S*{8M44X%mYbgw|>MM6!@(LFx_hqL9*@bqB*GzdZL&f${)
zWQh+MRXi4|9SEg&JwV$0(17?gz(s)(C&RMK1odKKKA@g1`HW(TmeTVo#Z!CqI}?cy
z^H1P(REt_2Zd+sxBT1m6Ai!Y$iO8_^(*?}`-UVbtgsFV;^+n?iRaYeDo-nLrmx|lz
zTsmri{;M?DWTiQ(eS-JD#2I&{pF|elb}CE6uIw19-`yU5Baa=`-2$*%c;kG>yv_ms
zBo&wAOh->IHDSEM%*%3pRx<n}ivUB4D-G#UFIt<Br;!uLX`~B~<k(TiFWrtc$bD&&
zqm9An^UwVmr+<siY9M{sOM7!=Qb3_T?z3h3pE8VFb|KqAC(ZAHQWb<!9v-#m#GLsV
zIX)}5$19}N*}Tic5T0*PC(86Kmkxj@#zk|Fm%PM+b5$f&CT{{+C2p2hsl0-ttQ)_o
zb`wozP7VN_mn7q21m=6wjT}47S+f-P(sBz+9}CCHS4u`y6`)G|g^x|zRrEdp7H~Ky
z$qn9oC-bx%Fw?`V;<Na5*P&?Q>|m|!V_wmMQUgI}dL6rBW=XXd!89(o-38MdzTUMK
z-#{ytC<5k0g=_FF^G_xt@_{w94TKCn3G+%8;y@d&M;|ZdS<CTZv^Sn#0U*n^Gi;L|
z*~=|ORimfJ<M;wiJ_J=XU4a|cp1A%v(!uW<guc5Xhg-K&wmzWcM^M<JB}&$O46CXu
z=GInZ_12fLD#e^v2JxW!_R%n&bjqR34<?dI=ND1P3uzK5o7Q#WYPcK5)@SVI%=>2N
zCV;mXXLz6@rOD9&wQcNK3Bc`xaJrB36H85Qr=^N7{B*E40h#i&gt~eJ2chWo>5{2u
zYIZx015*QE&{MR!prDDfg#!r|C3i2<ENYIdZ(Y*teOUtxXC4Uz`7W9(gS^)LHUa_a
zeO)=js>*GE=3e-vioI#&(Liu?J5CBm&`i~##v>kU-cii!(p7DBg@Xv(EH!z^wVdC)
zEnAZvG7Z^bz(7QgsPTx{6AhBdTQXkV`Jx;rlP|EQzM*txh_m8{d&AY5@>EB+JX{<@
z%lqD*u7xF!!1VISiO%nO*7muSdo!%Lp(%LJ$&InFq4+*fG9lpvd+a4~SijhXqE{QC
zJaV~9Yfc9dz1q1{QH{)KRXQf^KcOMcXwAxG*PGCje?5rGSejC`@!k!KTU9-1rLja)
zEqghumE}WwiuI-7P()NZVA?7T<#oth4-S4rb-L^$d8zt~S!7kyuW;weOfpOFvua(2
zZm>GYzq3Euk)3_ey=Y{<Z&KY@Ud}hPv%R&&9`;y8PG0iJnqPQgKwmd5;^?*iq7B0K
z_+ersl5UT~Ab9Od$~~c%g%QT8>By)kRz{4CY-dJVuix5}Dqqz1ouY+e^d|DU7auP5
z^CK*ZqdG#umVD>RvrCPFqV7DqD0x#yxbMPy{pa&X>@#kN5@@?5h<p0_1Tc<4JY=~6
zae=*Y@s=}_YFplZZ&+l^l5f=`1#|hnw<8Jp51s_{R$NJIB8jYyrK!W!kI0Pb5RA4y
z4|w7eY>)6F_7WJFUw!cYRk}~zvQy1BaQG;|BPtfRUpC$A>PV-pr>EzzZfCC1#H&sC
zHIu58{3fnRStg&8Qn{?GVt>Xh#<PM;%;><#DA^@Pyt!hFZ(5lQDjJN{FNq2e9i6kw
zjZcNI@v8)>Y)*XJ`UT2oD*RT)AW9%jlpub8Jn>5D$yFP<2~vipEmSb=W|^e~MXUe7
z7-`Pv#`{gRlq_$~pDxPGfL@K4f>&&8>g(?CS<Q`oBJWpBHiydDX=O&1h07n=S>1%h
z)`7^+JR{;U^)Hc!3i}GS+W@8fv2hX#)aEitEgT=>t91jC2#Z=Xbtxuz7^Df?*C=Qg
z<VJ^*6SKZKzsN77(KWnEoQ&WRA<hjn!jUYctw!`aQ(dS;Q_mp7e1x9iU|8M-?x(cA
zIg&3QM8>Iz@*_1>!mCYf^MXk;ZJARrbaFVN{;rX{j_I(SWcEf%d|<@S{1lnr@~;a6
zf*%N9WGpC+uXcXk+sV6d;cz#JqP|O8IeR$M3n&Z4#sfR=3xllm^gd)+^DiP6Y)w;w
zamu`<%$oBGhWw0A=JMx;BaUw>D#}*7bl(&qj6c!b*8K9|j6A^I=x@*0<saY9la~uQ
z`kZICa3<kk8b1Iun`YqRZ1O=-syZH8!4345GLG014S0&FNooURPrnbef*#t2WGvH{
zzG-&ZMAd&Y`J6MHNIc*C{mYh}@X~Z-`YRQ>e0CBVxon{lNp&HNkv);~V`?A-<vpfm
zQmKA$o#sc&TY+Y*`UsTh$Ryi+L?Er{*6J6s<qMw&e|c&RZC993xv|{@LVsT2N;Te+
zkA3s9zpbxfnL8*{JvmNMq54V*H|@jU{LqQTt>SYLTu!>Z=N~xG#z8NE1hIPYNro|2
z=SZOVBL;y|#b=&fGGx||9&!YJ)5vWhIf)B-pRWITfI+J?s)VU`_sv%8&ccsax(;1$
zdhYAT&#tG;Nz9yOqN8NyiS`sDy-76!IBM(G*DdwlrW6|K&obfsVN(0T@vY*PuEa(T
ziMn!@`a8}P4E^pK5oyL18vkN8FKVfo!U~L$7Un?@y}rRR#^oo1)*q>0v|Z9QDFq<Z
zAOPPGlaexa7wO3Ro#Z-GZmlXE=}V@aNMMIcX}ix0O;8$83i0^<G>;4A$UF@z{9*o&
z?v2m>ngQGD0YMyV&c~Ma5NC1?cRh-ziGy+C4EfXP1b3xefQZ!1#jD;UpAwJPVT~cX
z6-E~{Tp5nNZlBGeteCnq+UIE#w({V}dIur{7Y1K=YpZ0*ECH2>bAgqN*BE>D$B@#_
z>wHGNUK5UrUB;V-dKPh6m2;X`OlrRjhY+-k4GH@66W?GljZC^#7p{SoGGmlUwA5=k
zn;-S9lX6Gi!&}8~ai+<Bw&bfT+tLHX?~ZxDhFlhwhNYg_y?rYRNgQ5w)^(K<-@~zR
zw!A?hPmhGkDU52kUQ?31${t4!x$s}+H)P;&=9&KxuYIp}x<jTN2HAw=@ReC-ceB1+
z;!PKqR<G{0)(y-XFPAc;C7m)f)BqaykC8V0KXDR^(sQvqx?Y*mkzy5PwUA07Bg<py
z&^%~FzPJ>P4^U<n`U=nmL45#{NWHyV$yx7-v!v2oe_lPsQvBWBQz5DTOnpfo*FQCu
zr}WpDqD@y(!=0LkQAC2cnO`)XzCNSn%>A<QbC0i^{d)~rEt6vG!zbl<?KX_e%reNC
z**LzE2-lNcAvV0Aob)n1<nG0tedkdL6>IJh{`V8_>Q%{55^-T8RW(8MUCi!XyC`)e
z)c0mD;%hv4R`e}NXy81_5nQuY&PQ#(v)C6|m4fNC)SU;3y6+M{BdnO!TYDxZYX1BA
z$SSUq$z_j?Da`t=iUDLGdw-r68!u5Z{}`!0KgcI<Li40yGpIkQIloX3mGq{h(0~M@
zh6}-<y;2OowMFy0#5Npci9O3NR-u#6N^Cg}+m`gh7;)n%8jH0I)C!?%vk|DD7^!&o
zAwckqK=Ue@YLVsoxu<u33(1W~V29EU;^I%)KaBGm{r#VuS{53AS{ad?tKJz_3)4;K
zMhat=4a?$6WO(?+p|EC2o&4F9!Kt+Ekt~Z34V<j#>g&hr&iMgR`XGH6g05SM)xLi7
zoWYoi`B4t79oskSh}U}VO^QJhfga88#FH0#xuO;Y#~R*!i1(J@>nD*1f<V)EKV~_C
z=8T?}h&4DdSZ*`lzb{cv*lx25I-m&6T{PY9e~OOdN<Lsrc|7{^EwvrxUBj8|T)r>q
z*<2aZ`K7iRjON9`*{0$q<c#jA{2=EsR1_sWsGGY|j*cJqVbSL;e9X^kR1z4mRU5r5
zi%${bitEueI*7)978=GhBx$_C8-MZ?2(DtS2@wHgn4Q>5f;~+o<!{0YduDs`=4V*2
zdFy|bXxMQ4Szm>bR)uZR5FIZ)(FUET=9Q~Db^^PUb^52FVQF`d?r%0Tl8^|5sqfod
z5(z2*HS!(onY&UvOCvXTn($3U-$M2Q01H>M8ZC$$m;QL;4XX7s(pM!fMR1R#zx(=~
z@L4A`Q);mY7vmSW+e#Efyc)B;`?ij3X-9HF{wM2MmV0hz2r1QVotukqSUU}ky0$(&
zB4^C?lo_9UXilwX?)COMQ1Mj80o6Y{K#`FqR`uY18~Yyp*ERO6lDj0(Tw(FY`nN_~
z*IqoE-hF=V3!T!^`?{XnasktH-rv%$+9-P>CU#zx{80IQ5pFT{UDwvo=xA5_GKWib
zG&@tLA}H9*{BOE!n@PFj&SnbU00eKhB|ze{=V<})ANDx%%%VYai>aAEGO|fLo1IZ`
znOy6%SD{`Hqm^YtK#-`YZ9!<LKW_}Pg{+`aPv3hu`Ifuyz<}z|+`(AmwHB=nN6^(@
z%;AO-+utG&M*10B+^VK=``>y?wzRmsEr>{bA!|4|%Et53gLFSgn+SU_N_WJO<-We&
z<?C{yMIA?t*9V!s2DjaKp2+np>kt{^s(DzyVX~3kjG;VnEVy{V*^KrbZh|D=KjN<s
z=!6=qhM3%j<!m)sIzj?MHwbe%`dHN23K*oMNioPW(7gz|QI|TE^*@$A=e@!6RXHvr
zMlMwzCl_kOMLho0=q)P1Rhjl?%#^Wp5pQar@bp46D>u4vM5F%Z`Xaj#?%vq9q`x~U
zt=4WD6ppZC+)qMe{gT&iws|W0xGStFOOHR<z`&I1^ulcA)caJAEF=B$A%k{wSLALA
zhj|IL6yJOFRyN_q_j<hLXPK65WBX?o1a^!b6$?Aa8`RV^*a15M)UEWS-f3%;YubFS
zJ)?31(K4l==KDY|o-iFfezpl$PQ(5)Q5mmZm+;i-+z-0<C5tvZoG+^ai5Qs~wFB^R
z3cQ96aWv^xi?jxD$`8o8Y`#nd?@fCijp*#NALr{#eI0Vcvyw55w=LcCIQ>`SH9_`L
zDkmap-77Kkf^8uPH$clKXHz}2)Y;Z`<m6G})OwVjwv1G-;p3vr;s7<ek5@{!lUy3_
zFuLcxoaBP6+i!UIxFGvdKtm)0bc?wgkmZL>BUx#zy|Fwk<zBjqinxVFb=sg2$_lQa
zJ^)Dt+{m0)U%y#&e^&AoG^i7s`<#!1u5Ju0tRUWERj<9pHYQ^qQi-|m!6<+PJr`|X
z-T+L1)IPd1xW8g8MS`aWf%b!~8yet)_sz%sDtUc<^RFVbnGe8qQ_B8M;!0<f0bYO|
z2u)-u#B@$gB><+AaT72T*&|sff0`4p!GRhfLEKf&%E}r}dxoWk1&zUAO5_jtz?ZE+
zlLu(KY_9%`j)4)tTb1`OOi>M*2O(w!S=q0QF5^OX5lOXpKR|a0&|ek#0t}M5yQe1u
z@F%iJgL2CU*#~gv>?VZg0Fe4xfJe3^CH?6Pryp+g_6Dfc7%U$l&f`8z5^|u6V7BD5
zeFn>tq_Zp1pr#7A?q!I!7XLe5@<}>`5_GDClz3h^V-tX8o(BI!Zl<-MZj>gcfh8!B
zK~!`GiP%e+24hQ6q?S~3QtUJ=S-r3Y8UTIZ)rpd#o*jVreJlor?k3D#`X>K_yz<dX
zXuy~!@op!v${YpvAtwt<>g<pApq-FDQOM!r<<*r`T-CdULL=88a>yPJZcIffzA{rx
z#w4H(UcOBoIyKz%0Zl4yO8;LF+8RD!)FUMJs2A!hwJJ`<C6sEo0hSt9gMFS2p*7v$
z`Tl(F=E_8^LzBm63&74z&BN?rVN`<nUBp6oRfRx31SC2Cg3gI};BG|QyEgN<iX_b5
zX_^?e2_|M)Y6ybn4Z)hI{mF%vo!st*;ol_}LAQ?vvWx_2;hMjI7*Rl?G}QpR5c+!l
z)Yv%3uNcgV$@Zn1np&wv!}VPtG~l8G0*;=Fl!miaDoQ@fuPkUnL!hl(DopG=m*pUP
zB!=508)hB;=`RJa64pe$TN&sX6iNGv_>hkPP;_?P4jVgr2@{DA4!r0)M~6o5-j#Ep
zRQsNrz6I+Rx(BaY;>R)Zj)u0ny6NN}(k4<b@M43_!7Ln#XYQam2;8#(@e`s2jm^{&
zqUV{jtoNF<wtt)$Q}UR8E+7Q0y+C>>1c(qIxKOzSZiD16xI$Ozc&o&s_z)}tJb*G_
zY=#2H-vPa+U(nX#@)QMVJS6%uQb0c$=@CdiH9!w7K<-`wSyO5_K<)q;0^CT+cWM?u
zWM+8ek!V81U;>m201_oksM%-sJ_beHfcMRbcMz|H#(1N*I3AUNA`)Kd0XVQ!Lh#N6
z#V+Exz0b?H<0oH+i867qyqXrMdNTpK!{&9t$xuEKmEwKbj;jYG3Y$)y4@0o#sR4Rv
znF=Vm&DkGyEP~df!JvQ6`h1yS06A<Exksr6gnK~x<U<cX;X_13gjql<KeKFdlhN+G
zeE?Ykjo809s67ctOYxe}sy5NFt;IjoHIH&ZeSo(2!R!PzbYXFe%j3OCm3mAPpSt;q
zR97XIB_(_hmsuda1h6|LtX0BGy_OVr+c#rSnRl*MQce_78~ruL4ti^095{(%%<Jdt
z{`n&3?P)n6Fala9m@E1Dv=|D7QT|UbHM+RC*g7NcG*Ro64@?55<Jr@vk`zrdyQ@Fj
zLkflWRFsrRS0?T6%{Mf8qC#(A<(<n?J02Vu0E49lAZ{zTVHD6CoyDV>$qh?5PHwl+
zAeTefF|$=txU4Hwmf-f!eor4*!M9BXn(lz8@}$V9&cZ1_2<RRIfBZdtQO|e@6OwrC
z<2HeTfvucN(Do)!5-#t6-nzs;DIzX?<xC`YqUCM}7JDr0?8F_zuAnh4UmLU;z6SE{
z6PWPb-`|hm-=cM$M+ukRn{l@LaTF~?cIS=+=;2Jqo$mu~t~lp$5W;e>PUI~#pSc0v
z%@8AGq2T5w3^U22kOc72by^c$>w4u@TA&BaijBPs_Xd04T+D3{c%ObM7`f;De4dI4
zN@aq(<i-ZJ5U`{mUi5t>fT){j9Hy29&b!%*JL!*M;X)-W3)}(dS}F*ZDx;Lyz>KrE
zvbGy<yxd5djv6kuMTVS|V5P4N)hbXA=YTzd)r5dxoq;7IfVS+?U)qaTZ8yUvpa00d
z8*-;Tla-xaf!3z}=g+5`;B5rJ$|!*XSoFw9KJ;A&jf+3SCU}d90H?*Rd0D`^-{~Q+
zUp(9@ZtTto)lgNH0AwF%aaV=arSJP{`##NqM!zXv);so;aC3cPID!YP>p2xX;;RO^
zH`CSOaF??+6xwDJG!<J0V#9ZMkCMw~8351t4DN{Jht}qS@~M{sgs5>?DcVnAN5tK5
zN_TaA(ij6iU94=k>HGbEKK*S+RG%^qunKd$CjRPPhbcz}-~{lyIc6M}6IwqJh^-G>
zt3$6_Y{I5eGycyLGq#_<268+2Haj%AIWpP5<zwwI1&ghI6DB&j-18v=U!8Fr&Utfv
zI0r5yJv2+oX-pb^>_A}XQ(q-MYzu=P>0)s=812PNwQ?|c*Xra7{QDo19vc@7a64EP
zUcqPaX`dJ9vMJ;F?+F+4*b%|+ZSlTbx(9Gj<7%6?R5mRCJ!R|l1oWMh4)Q;!)sJj5
z@Mg9ML72CJ4U%%7>36ymjziiu@O92GsG1pWkFV*Sgo~vA{zkK%EP?3Z*8-GkLEC)1
znn!TM_CKHf@r*p(GP#jg5j)ESf=5i-vT5hPAHWR*7?srb$Mtdu5M$uif^<>(yjrO2
zv8itOKND`SBLbR;OptfMA1(^~vZ)b+KO9x$`D*SXT?hK>j2b-RiVSO>uB@yaf|pPX
z0R}uTJ^f>mNh25PKEN_t9pO|0$Pz?m+}4tYL61qvRjZN9jFhpl@o<rm2;7kCv^gw?
z82d*76BrT_#sAM6Zs2Bfr$6Irv?#lVAkEKw@XU7r!1T+6ThSbPPwndWd<%rbnA`Ed
zD`%d^fYyIlJ`D5gnwpxDZEbCEPckgUUk9-u`~DgWHOJ2X!|RI57Ooos=A3^t`E|<n
z|Eli2<Ej4t|Nmn~9D5%lNhqU%gX~q2y`rd;G7mz@9+_F`pvc}bO2$c&j0(x7agsf=
z2@${hQ?J+S{rY_0pUdy}{rA^DUFtl~^E}66-tV{D_5M=lkRCGW$&(IkB8tcR>fgY?
z&s(?$T6w)aE>Yg^ViARbtwDVrqBKK9xv<D^U!5{Q#Jr$sXeq0KWT2e>Yro%3JD12P
zX)|CEYXrISakJSQZ-OW)ao+di<5gka0nz2$r`kIwfuYlCp#$l@d>D;dEPhW3;!w~O
z1C>bTzrRjL(vq^WvhLs=KaR&q@G4FYHK|QQ){RwdR=#Nvti8v_KB_HjPsc5FjAh>8
z&3X+^9**+K%q<YiAIEp7B!#WbHLGoc_Lv0}DIomG1%1kg)srQ6hC|pLOApsMem>y^
zoR@28Wp9u{=>mx-QOfQ0?(YRV2>+LxK1ds-L@^2TZ2kP|EOPSXgIf!OI^I8jFp?CC
zBbH@keRRbIETqRJ{Qw(fkR)!|6t~H9o5-I&&GCBY*AtErCD35|g1W(aV;2Oj)^bBI
zEQ2OVspZC(w<I$ECo0MK5fDpW{r>s2+*AyNh!A;)D|-o>Q;mp-h;q~Dg-_3GT}fTZ
zGWIhGr?#}VW`Wc2ad%J8Y;C|Q0#j1^iolc_iQLMFnCbNlFG4mUmD6J%8y}A-3`Vx7
zetmLv2Oh8QgEa&<NY*TTSE#6|8OvA$tXwX^yqQ>e<Sjj8I7aHvjUy}ZfA~rh6(Q?w
zBUc>ip5nuvZDwifHKj}6kqeaAo|p~^3L@Q1NZznEOcNi{%hTMRez+^<e}b};hWFWU
zT?e}-W+|LeFP0zoaFhV+T7EVm9z`F5{C=-sa;<C`{-Jy6n+FXpq;uRk?+fT9YaNaq
z^>JJ}6Hkbe)3VSMxJ$c0Ry)gGr>Co1y4GL;V1tK@5TkSGN6vU(q^{-eRZ5}98GL6l
zBqyL3<g5>R$KLw!i9hz?r~@6ufewUb%2LfR?o&&P`BdeULeOkT28C1)yprj{{IvMf
zkcM~7#)V@pQ&Zz97qdLUBgj5*Wbw<9#Ep-MAYlt1zFt?guV_h%&j*e7`l1>=5I0=b
z@#y5fd$vJ&y{uRcy&Ee7jMxvZ#G$r>Rj~e<-+q3VkV$_gL!P6onu(9k+0C1Ar|Es;
z(jJ(po~!N{-Z^t(cV(5wzo8;@p&zt@Z@i8Vb<JR~8$qsCOaZ&}RZib(SY6f1WCBS$
zmO*oLG!*qN!a4dhv@*QX?q-LUtLrl#>0Uum(Q}DKiw^I~_+KaXTV3^=+vG-h#)mec
z+v-TF_;FJprKMj6YLhV7>MQy_%3gcVU;}f{@}tYKh<+*2eq;cD>4HO<Y&wmi5Nf{o
zxNCORpKro?lime58a5!TGd(p5stNpu59c<|xv=>~Z&p^_qT3)Y%96SWg=2K+UDaTo
zc=7CsS=kW;s?qlmYtzaih{c*IGG7e2<mNY5lFF=T=JOGi%rp|Zr9$jHd`KS$ladF5
zVzPIy0`S2>R1JyZJ|A(>4*K+yinmcZB(9CG_;I#0Ou|?<u;JT4x>!o?yo<}RP~ta+
zbY?|t;^e&!{&a1WnKE%vG#v9KfV@?Q2%gJFO8$P)bmNngVfPvpQ0W4SW{H!r9U|!$
zP&eR6$8by<ni`KmT0-~p8y#G!;m#2h9SxHomgSdUjJweeoq=-&5{4heDNFJ1++}2D
zHp(7dw8X$x^+5-zUgPA)FR56c(xpNsE*e+06b7t3JyXHqa_@Cqu3oz0b)H;kuv(Cx
z^P%*VLx@U!lMg#&)4c;im5H5-P3Vq5@*-+bq4pNLEe4wviiwFObf8Lx-0Yp!Zs5pJ
zKjA2PI0|ki<YT&0SrnhrV6hrl0Jze5Jpww8r#q~@n5?RIBF&Hv$NUN*3=hJCe-Ew<
zJoJ$ikGs&YkA{SVEOTaf_>^uwe)1JnO^w5#`VZhx@Z?oV$4TKONjq$B*QcKNWiuCp
z!3m)3!!a45gkirpW!cZbl4<AifqQjzI1(;h@1*ignoxo|)`Wga3GYL4beBVyROS;5
zlFqQ!%mO_D0D};{J07nNuk$aSk{WFNa}>OP!e<Ec3>aj{o)?tD%uVP58n_NyAW+e{
z|M+yA3wH7543AIVpaXi1096omH3g2DC>=#;M>N&szk&!PYw#{$5|;f018%d>TM($<
zxjQA{ze%$<R8n4!`AS*MD1bYKsG$QD58wh`F<_H<^A@?M{wHjyvFJarrH6IL?FVaR
zL0;a?tS1kkEH-|e^3?-4NuDTW+4lfVu?2Kk-hodrU$?flsxsX58m($TSW2UNqH#7`
z<`GIswh_%971D7OjLwpX1V~_cWrZ0ncj%BhFG&I4A3zkR>(>!!$}eymF+qeb$zve{
zX8;c%{(JuRYd}{4t)cTBpMTUn#Y}?@UCa*4ow4_faCaah%>4ZPMjMysi6*oQWfOWp
zm9&rQ@wX1hO81BJg7O+Pff!^J!AoLG0OejM^v<a^iP!kk;V^9o%gI^v*E93u0#5Sd
z2y{#Zlp~uqYm46c`V791o;x2##vM68!KLIUKRz}F@8}<IOZ@|Qd`A!92$b-hp_kV^
zANGbTsa(Hy?EqDfk;^e!dV1`ZdP|Q2q3!K+<Vb#;3?Gy5sU<E9*edc~Uf=jaVbw(2
zxG;21UbN#08dK&pbRl6gt82J1?4BF}+l02=ACCDLLf%SO+j>;O_||l{puyxRUeH~*
z4hs>;FCE}wW7CAzn>7ewm}jYKuz1kmAtKLOun9e>LTaKX&)up$jxpOWT{_T>#7_3P
zgg^_6ydQPctTfyUmL9h7orjb+jlwZQbPU+XWV6bB%w?-ZgV(Qs@|Y@U=q7Q3Fvsbd
zVqB(5J5S+CMkE;tbvqDP6b50dLC0Xd!RsXVzN!3++vFkvr_yD>axY-ZINzT*zk~ps
z%5B;J;1t|}h~c@Edpw5%_Z5KJumT1F110w<2G|sZl9G{W@FcU1P!a^1(C-?g^eaCB
zK#&mp#unz~Pdq%tR<{m?(=gX0RTk+bfY~En3VA$3^YB|_)OfU@py29zf!IBJW$Hp!
z%g2vwHmKVTYAZfw=vkmpBC))&(NTe!Aq)7H2m_Xjen9zieL7dP%2QaOL5~FV(R?Vv
zo_dR*lqAtj=tY)rOmrA|Yb18iQ_ZpLaudeix}D9%j41sgn#=X;xHN5P2!SMx+%<d2
zh4BT2RkqFBD9%5)C(cK*gjZItydrj2zKBPKO~6`D3&INQ9xmc=UUfKcutSD1Zm?FQ
zzfHz(n^3S4k^!nS;LYeA19?VI$%^l<RYsbrZ-B$gqWc&KKSSr5edSn4{&JYo_8S&*
zpcVc%`sp!^E15jch-Tw~!1(_U{*%nU5lUMUIP~)Vj)1VCJD-*T@tO5eo9Mjff|Kt8
z@Yq!g55kR(v=Jl0&@mo3bp;G+iX4F_JU7RabX4N7TNf+s6NJe;k06XE&T1#z=+Jo>
z9HqiBKf8QM*N}h+vi*vlLO6(fM!7$`gwNq%_406`Z9;b*Y(j5M`R#WGQB~dr88EIW
zr~52N2$R{qfssZ9RQ}B4fD-J2!AygbYad(!t!uWQy8>zRmJw(0feUp$FGNXfgKKo$
zF1$-~j1D7^ScDM_lxlP`{5Hqg4a@jjX}&K}WTGq@fE-oDh}g$EkHIDG3&*Twe<hDh
zDt-;@L*cE>%|{d(D6!$HcBjX#DerEf(QqoSK!O3*(u&OF{CQzqm*f|MYc|1`S^3Gv
z&tHG6Dm>C}hp5(q;mBZu!70la<&yKyR1?F4f`b{!RaOw@6&7561$=Oyn5O72(>U@Y
z|7J$zFbZhC=L)zr2gZVih6V`Mq&U}QssvNEaP?BUlCAD{ZIu2Ntz0-jcnE>cpaWBs
z)gHJN&6zq>r5X|SN|pFj$4o%~Z%JW)IHK8uyb*Zr7Qi{n!oeZzpnl1BEx|lJAejNH
zyug4h?&+iIMfCjU2a`|O?1xj*A&*28i96U(JF380)re1Zu>we>Iot2|uNE#vjUYdB
zzvK^pTM!HQaVTRIV!>HBXlZGw+_-X$6wbq1?K@8fmqs$dfJGG@!1d<bV=RCzLbB6p
z#3@_l1<{4#;gDrHkmozp!x0}~E!xve)r(kdA`mVRRZxaEaUVf~qi~A~PLDbkM!2HO
zPHn8J8YIdH2A$K1s8yYB?7hC^aB@`dU4ntmijgOu-v+<J@710xVy-s(FK|7AtTP-C
z1zON*S;rBtRhrj~h(Z6UCN%0R1GdQ4k9=m&9-4)i7@1JVJks^oS>c|6_T`!#nRhw^
zq1x`nF0w`->IOcxVv>3|_Wi(sS${Z0y;Kl!zm%^k5mf;Fj8!LAIX5Tgy<wl&d#WJR
z1VOLBS;usp6W+~Vp9|?DDN4btf&ySMp`<*%1C43#Q_M(cX#V2Q`ES~=9>^lX)LG-R
z35sN89Yo5GjL*w}j@=U*7GuLPsI~lcLgr`20@!t)Ni6zVMvD5n$D3M(mXfiv%8Qju
zQHF>-V6<Q!2Vdvq=$hDqt(ZaNX{8U2h8|`vC@vNS;~k@oRU>e_)FId_H!nam<5=gJ
zSAzyja+&YdJG9PWeMGb<YVu8k4CP_8&g#L}O#@GFYg!g(dm>Ov5Ao3gSr#!h?|F;Q
zNWkm-kv~bueDQJ-!Q{?>&s(?}Ms$eWLGW;9Vtiaf>G$dbVZOh8%>8mr=z!232Vy6Y
z$`}j<qKt^t4b0z)y}p{?k}Bxnj)qscu#!p2&^0b3h9ja?uq^Y<C}z)VLkB16bn%MD
zBquk33!4q-zL7{cVdhig{(8k3h0^)j6<ufpTvE=ZE!u)QaCuy(V_k#oW+aHO{a=ev
z{Q`Iu73rslE!uUSx4scuc-R=BBx86-DSM1bJ^mYEm1_czKRUmf@J2F08Ce1_bs?eE
zzX@0W`q%?)3?r6KH6PGNlWT4-0K-q;A5tg1h&O%W>jSV8s+&JSE=bQAF6cV>4Cwsw
z4tRov<{%ZRBm?p$qPAGp#Y&+4xX>yzqXpUzAsswkFbkpsZGSuJ6Or|)OpMh?=>6+V
z1qg<csVhSz7%quR^;=l4ueJmH2KQ+R*Typ^saYiWoDs7qpuj@)*ZDoZcXPq}eAlH-
zJl1`V6yOjN+Bqq{o~Xwf|HoIEsg2K`rM9ebKa0O3pf7qipE-#}c$9C&*0JNdq@7!#
zJXK7+Z}=9|XkD}KIrv4c(rtChW+kP>8f=QX$mE!cSb=CRvAs=rhJ?@lr)H_Vt@d%7
z5A=fR*V_xW@Sbu$CcOQ}6`IgeI*0W4AOu-3%C|A~y0~ykopk>hD(}NfmD_p$f$d=W
zh=`3INrGAH!n8fMt?;t9e7koo{iger+SN<p`BAEyQZB-KEQFHI)5HSFL#ICDmt#XD
z9j*zlEj^p+ImvD~@-D2fw_J(+uwj7>X{MI=mtV2DcV<t}irkfc$kdRi$injCuHo|R
zj`IIP-cp&Y;KlYz-d9G#N9F2!mh&JkQZ{CZtz2S3M%It>`_~iSMQ=BVO!UcEq{Qoj
z<Zk)C7WGl~o<rZ6<kB8#E;)b5E{Id?t+KhDysF8?N?0{ABN5c(BK7m5Hn9?84y1W}
z`dd=3@42HDJXl&jNkuEo2VPZhADJv`swc)P7W2c+;zcZf&W$B>v2?wN@7`h2RSeyc
z|Cy`aI<D*S{Gnl@c;=5HXXA18gF4Af+S6gNJ;QAlGj|g3OIpHW=UOv^PEsVFrAmqK
z9}VTSI@8nh3Y^!q90i_;BW_u^H+L$Pp=b6mueRHmnIi$8f9N$w7RPewjQ=QmNe)|E
z*~{)ia>0vh-s(B{_Ywiw$bM|~*G;}2u9COWkwMpNRGwCEf2~c3V}FoPtG#?r8lS3V
zHgb3?*@}MY;p5`-vnGJnD4Ns|Zn-heE^n!kSg!ngeSnod;h><e0b#_!vvgssQfttE
z>`H2n7g4{biHc<*4R<$Wai(KHNk4I>@V9d0#Z(R1h12nNi$$G_Es`$$HZQ$)zL(Kc
zv>cQPbG_}6H-FV4vzcJ9{aJqhQf!uqw-)+^kxa9%rLy@G0lUlIo8*|<?FPFyiSn%%
zc&xrvJ1zd~o4QFa_Nz0arTn5q9ZTJb)JT?)$f~B!`u^poAMF%y1$GAmkvZjB#1sV3
z9|#uoIEmT3QLO7>5}E4@EWfT_6PcBG!^p^zjpAeGy!MP@9;?MYOmj-M@YIrLQO!+&
zU@7V5QUt$JFY&9j9lBSDuQK*~DyM($OC34KW0hG~H63Q^-WP6t)q(DNXtO&O`a{o;
zvxqwsyK8XSn9YTgaQ%JWNK8UlqJi<X$myE;`o5(KnT(L)&yAhcm!k6tD7CJ|_F_L_
z`>r5kzI4<F+dr3%fY?PDzrm`xoLqMt8?O4HpC>B6QRo_@q{^ShCdrIFLf=#LUE;og
z9_1j7@U2@{4Ic-xJa;F#W5j84zAfX+mj2oybbxHhJG4}+UWWKZf-)q`n8v}_+=E2!
zT8A7J$9jy|az~hGRqUwO1?P>Twm8<MN;(aR!h^MU6S?uM!GSB;!a4T`Mk}c9WrdJM
z7Os}dqxAJGVlavFih(sgFKWVWPj#v!)^BPRYy^mE)hYhG=uDEPjF6$)d+da8+z2SG
z`~#(-@k4>bpC}ERRyb}1g?8B!pN=h2_x=m1rhEHfSb2M5a71R{n&mxx(f8Gn6FLSG
zKL%2RG(TpAU``ZdJo{ZDnPfQlP(wxG2|MlMHBGHPF6?FYJ*1OfegeVZg;2uCho6q<
z`~&-(rYBFGiaB_yP0KfdT|YT<A!J!=Fl!+?_s;Gf{%6)>h(e6{(uwV@`uP*$99J6;
zC9hvs*m@e>*VMLRWP1fv0CT)`aVI{$mylE)myMA!I2a>l6iwPclNY^#wc#!K^7z8S
zKx?X<6t4F`(#gg1Fm;?Pkv$Gd;yWxeg4q7mFA=9xXVW+mM5f)MXFnJ#P;%`!g=`l&
z0VM}dp%q>|{M~cw=>2<4_}(9)f=30+BT_C@rfK!9ub3R{ALI1Xa=M8>-`9aHB{NyI
zi()E}!|hNRTX#n`tz`HS^f)gm%{y3RCJ^pF@2%zgV)3#i^W)*8jCtwiujq_Fmo}UD
zUiiKBK5V_>T5f;ZditnV1MLes6Xl9>=~2MHo95vYxXrrFG#@sHPKou`8HE^{#+4RJ
z-4h56i?TBvRjs!eCu|fMM8wORuX$WpdYYx)ayG&wi|oTddt`gUrLvGEspY>F%Ap%c
z$rZ{Gi$=dMzP1(2%$T>G`B=e)6}_`f)%$Ne+zP&DxH?PND|IIOva)FNYmPIO-&c&z
zzBu~U@XXke*7>T3<vSj=_JU(W4Eqtb+s-3hjpi1Xxpy66c@^xGT^Wz;ZWYsByyA+f
zmU_X$$Ea?QQu#_cLSPSgc1O;sDcV}juipC11j%=~0oR+MBPGvmJD=N)DToEq@<j?C
zbrRFqXSRDd7n5z9+5Y>;J(2NEhUCh?EgNz{XI+rl?TpQhYgi5Lu+#}Blp97lbg&wY
z;P2pVt9*@QZE9IGv<)4L5BptX)lf*g5^lt16u++xz1yv8cG8x*?7fw~s$kMwgg}Dv
zB93J<_S~U`+;`KSZ!Ij2Nj+AjU6QcY<j!SdF<6$|^kO~ST|D%R#aX(1${e1PeOv_o
z9S9~;MS4<C-iVRkzh*5<y7taJxA;S=P&y`5?{=tOkgS~ptD3&#7S@_#@Jm?a-7Nf>
z7eRLj@h)1t6>qIWmXuQjE=&mM^Y<J~fQ^EUieZf5wQUUT>)QsnASP!0-?c@jd0vw}
zbV)x^ha(PDsA60NPMeGFQjT67nvvg>5-R5iw5XN0px*T@zgervEqsEni}S%{UzS#e
zup(<RC2O>Qg4BfnleyG=%Xk7wUq`mtD4sBMrYBtYbKrTwi!t$iQDSS{f$>Lfi52*j
z4jX)>Vo^%keOo_%r|R^~?t@%W{n#OjCe9tMj+{Qim6p-d8ApyB0kRsBm?cM-TZ>jY
zeFdHV6W;3ToGL{?X*Mjj#G1)me?ucITb*N1F<4W|qm-nnJ<%p*e_g%j(DBx5!IvbP
zOG~Zz71s5RSIk`grtN+`eK}V!?71*b^T4K8%d@9nE#jP&B}>}nb*bO#-3}Ty8cU|#
zQu3XNL*QD;`911H8MAW~<+fw9a{}6>@tU8p*&=sz4s8yElhv^t_YM@rI4Yf^FqO01
z%#HW@;V`Z3Yy9MHKI<2*-_HoAHlk%(Q(i`9d7CwpBjCT^juf1V!iFE8ctj@;l!ojc
zIB_W6<H?6y6N!=&+tFPWE<csODn0k0(n%I;`*E2&cYf)^ctj=AO?*f_fJy{xHW1X<
zP;*^7M=O}aiDi^l>_)#i9>ey5ir?RZuKWh8*h3zjm7V)4%^uedhGE|sMn=WmzZz6f
z>f<m}5fZ9@)%{^&t<mjr=B0+(<P-r+<<I;n|HBW4ds5F94zGVyp)?WtGf;*RhJ_g=
zI~FsWP1A8*Myc2E`Y#g1(rbgVg>kfcWt>c<IaK9!Z}8a7{`JGGl`T(f`>x%qzr1mI
z+-5s&oOJwAOU2#%(Ziv=FLGnbUdss{&20)!6e+*a`PY~ZmM)lcA&rzIXd72A94YO7
zGcBpi@2=6<0=xc2LpC0X!i@MIGLdltXL2_=S#}%F&&=~_X`gVqCXnzMU-Ro}F~i!l
z_0cv1J3iInlh+<Ooym)OD{kws;m-bCQr(n;ZKs(2pF25N-Li`@f7w{<aX{?7@Ieic
ztX|HY<qXy}IUTF9Z#rz|%$CfiCGpF2O=hf0F<JA8*;{P%>F+CwCbe>JpwH%udNY;0
zMQih>D|50pUTshD-cBE#u!y_8NV-D9Yi=hkhT+t0E7qwTpES&}^HIE8JG(({+I)|a
zvgko7Xm4$|^{W;dhp%7IA7dR*aLu^A$HPn0Wp>SeE@6(Fb?~-t<H@=I+p;uXi$1Tj
zX!4z5Z~9oKMP)|fuHZdfW&qud){!Q1IV)35o1gNDVZE}}3o*#x$v=wP`Zl3`fA?H=
z^Yz*GvF`YYXLzUkllZh&Bg^kY8e+4uH|KlWeU8_Uw9##F_qV49U1O|Yx$|H_-CXjB
zNKaFxZ{O_E%A5@L=Y=<?m?D2?Z)cq@%+VVC=3ZTyQ{{6b(~^EbM=Izy%;MeJd9$}H
z_I1}LiHc{6T58%=<t>vp3UzwvHdk(xRh5q^TRFA#C4KUeD~VaO7r<@A@9N((d8ef4
z_^z98J|76CrW!x_zl$^CJ*C6@{VY{9UQkzf+8@}g!s{&`nv4|DW3zL3b~+v7Z7YR4
zzr6AKoP=A+{QFZbhS4weM`C;Zx~_iq=aOa}Xjrz`zi-hlTJN?;*G#3VMFfT{IQv6h
zZlPpCxq|EOjFe`)v=V;XJ0_RV?Gh;=o>_d`S~%x<*m_sLRS~_@XLq9G?;~^FZBn7|
zKBvd+{i}Q17pc#)IW5^c*Lhmr@>82X!g(gxOK!-pvvvQG2O2wRb(Zo=Go%CEGvDl!
z;pYwtUyl-{2@{ch{woh1>mrqu7)_gX*4f6$PFXb9m+Lg+49y#bN1qFBOE>%fxH6d+
zxOy|`$vYu^u?wcLvsN!}=_J+*wmMij5BsKU3Td`g8q(dE_^cGAa3E5Z``PP(7uJnm
z?~N5K6uSP^PENTtCS_TfJ4)~O*vLJvW%tAK$&YjU+r0!+9=~KV5VVnq%@Dll`(i4|
zTKH+Du>IX(wFyacNts%?7{i&D`@WCUT5M>RKDgvR`B&#Np6%~@H8*2&^j=iJ%43;x
zh2`g)Q?kfJ-{#1L8R5Oo2F=;wEn~0z=(g@jAY<84y}K!?L6+c`dS!$5e6zuKquD*S
zt9)wWh~_<hZDHJfue;LGY_;*-gjpv~W83R;A=~0P2KRLO3r~cdFg5;uePlMf(%=dR
znz_Z>D04+vq&zss`#+D-MvS-G9{IWAF+aO-$Rc`gvDn?7b_$pO_{88Phq@q_devkW
z$Isth32$$n?;G*ttQbqJQS3`J8YFqa8+n_yIjZ_oW04A=N4Yi?$N-C2Bv6<C)6S#p
ztW$F|8_ROOWF$Jygly1Td?uP0`t80M?o3((8-ti$&Y2r}xt8Km+N)aPpLe1aJX9RM
zFNJdCeRQT5&&+sw2OsY|>io)Fi?i&me!(#J+bHKRCdFJ@Ph1=0;XV31$!9D}At|lL
zTz&h{S=jN+ed}Gus~*Pr%|jFw_I4ioO@2+pf2qD9`#>dPmN2Aq-yu=oQ=vcR$m@$=
zUzFt#vDUW3Dr#M);^VOrd4arsPDM{(vs_blN3Oj&x<9bO(>unTDPb%B0GFg-+D(Zz
zvNip*^fS+SW3omzTPkGgHHBo&&P}u_NUM4~8~WoTF5xzO#ZKR7r4hFA6V@n+^SGsj
z+r0@pM$?=^#rFwD<`JSNZWEL;d}YMWI}vHk<v8>V=r@eg--ocgmJKdnzh?CFCrhNi
zNaQc~rOanXBhk;C`9nOXhN4YJHPJWq<YOU#01@L(C?>=k@=MrDs0mrN@}%6@pR|l0
zV^=dy@n06ukJ#*;k<L7#p|vd8r`a@kU#sDD@ekFDiP;>3<g^`2rVVBT<x6=tFPb}4
zoqbx(w47Cm)@j4n_6?i7F1u7=XJk;ohX5TPZY7)AO5?UC3Gv4o2Ue?16nqlGb6d1b
zRUWPFG^OD9bWKL`ejOd$v!_)q(BbML=9LBKpKb?9<ul<K`WBd03dh&Qjk+h3o%MD7
z7_}U=YT^@8i;7b*r5jv+yHh!%TE7hY+-Q#r{pxGha(`OJ^e8JvAynq1!CEcJPsjHT
z&ls1oiY-QB-wew%F9vq4ZYTSvScJqsmwfumdtmA5tB9JvXE9?tyTWPpO6TuZydf5@
z^&id`-z~Vk!$yBe@R+ED`O_1+=0{1-C;H?ZfjGg@<D+Pj{vVq;8@e$jhO?=7yol^k
z0a}u-jRo}Cjvq3DvcdkEHSx74YG!UP)0kPGbeyXyy_pS5YKzNDw>fq~TaLF{H-2YW
zw~u0)7xItk8{0qA;M=<K<>sZH{Uz_B<9}3!jx`)SexLq(0E6J6bBez>(?_4&l_+`{
zYs`{=^j^CnA#OxLOr_=JH9<;_c-iOlI@?s?Cgq~(7akX@_lIjNGcsf!YTGc3XS%ZF
zp=O<{n|6ejFL5yTxVHTU;|44>r-12M#bG@H_)Z6(V;3JA<QNm<It-rwsunZt?&xQu
zK`RM}Sg`IED-<cpczG3c|C#Dp5HjcVHA{rd8Nzk*s!o|vFt1<W!70rTmZD6>wHCn|
zVp=z8TMVO^7>l*uZ+Lf{&p31qV4=kn#b1cRMlm;pj=H)D7&}u9M*9Dy@V1*;D0Cwx
zMNOt{J?x+2#~HIsiWHs=rB`~;*AoBC-*|2OZiReqPO11!Y4eC<zb{;f1p}X>Z#w45
z!|2;doJ{c=j9M8R^^IrKG8al)47GZnWn`T>n`X&+(&wG1e!PjmC)w-SMpe9QN=8MG
z9-7~Is5$9|Kksk}>-~P<O``U@gjK1DrH+@E!kJzfW1eZjHot^(_uXQcSH&Otr^)&p
z8#}vEHO|6ZojhY2iYJ8n_&zDspN1q=s)DyEz7j+$^-D<1Nq*<x&9_1=XWE2l>J#=b
zpcVgo2JtAs-k2nYBgMraw)a+lUC@KCXFn3#+Y6$oliF~LUz2B{|Az^B?kEx#)cX^7
zFa}PcO+>*{eW909q%xnO(3VY0ZLA%|2xY+NU%`(WkiY@jVyxu>$h4#X3PnIHG$wzI
zkeL@jUIZ$`a&>noX!gJV{`lyq8jqYi!+5&iud_r=Y83k2zvG1{Zf#IPKn#QhvU=&j
z<J_97{s;7#3rQqYOm-6CO3owlA^_UK4?(u~e-C0xA>qw~5N`(Vqb-#OpCLq0A&20?
z{i<m%f*QP!9zyUSX@aQrA13L)BB+OdHk`EoS|yf{lmsGrjm$MEVIZfmH_U?QQP>+Q
z;F+605_Vu6p;R877Q-EDk5YaHa1HpmmX`a1sHb5y3ot;NA|d3o!4V{c%*w)YNLwBB
zX(6J2V@E6YLr8mRl@0qakEsB(V69ADP&)^TUHmDM@IOMxy`YNw4`~#w`VVQ8s)Im+
z<huVNjY_S_K7m|!z}v2?#LLOQXg2SH4WNz7z|=JH4_1_RUC@$T1&+|%J7VkT=_n3<
zZFMN<2hhj@SOFWX3zk6Y!&c_iTeObO@Np}xK?12gkl4;u3y~>*+xU5P)XMaq(h`;b
zgypfM%|fM+VafU+<PNf2aQ9b^)>+dA5yMuQ$T{IbHBLdY;AYm%|0a?CCtJ`$ZV&2{
za&@HY3%u1Y1zV9d^wUW5`@PDImAs9{cPXqB)gY}4^b)GO2<!pX3!Fe5$em~XfR;7c
zk<M=O3SM)8667t~2OT&-E2XUtptbw(+@<!abRcOBeSfc5Ofpa47X(o1l=S}!DE-gr
zVw9I6BnM3zg*>5b7Tn<k!I>lY0OW2C>(Roy^kcwEK30APg2PrO0}%1UFDBPG1IpD3
z8bUAHRjtv;^9OXCUZ5kHH}aaH5_`SRBQyqmsD5}&CmlMXJtQhMdIa=+RN)Z=_6^}^
ziISj<K)HO>OD~NBp5q*nvXTq5{x5oI)@|q=pjR^P@*xF+sh$Bq7Bn^iv<Ee<4g)On
z)I1GQhy$rOq7O;?dby@(`((Y0PY84(Vy`w5d4}^$UB<is@^Qvt<m0+<;?zj<J*3X#
zUO}6`l9M|K_3Te_vLS?~_u19c=47FeE2494d@Qb+u|eLEdI~aQ{iT*ifhC$VPe#PX
z$@ks;&k>N}_`e~W{*zW@7toK=b2m1|7c;(E>%UM=o*^*<l5WzL2TpL$CJb2n?sh5`
zsQ1C3u)Ww=9;){sLn$%+ASehOxPY5d90&tP20Whr>FeZ?OVs{QgvKV?cDywK&O|>W
z-Qi23@)X2te-5!Hj|5?Amm0moAqT>9%GJhYx@70N%}|7a=Ae&(yzS8@^jCK&(&mk(
zj>ob_{o$Z|6xT+5$-~VO_c4&q(*Gw-`wS6E1DLm%(FxSdX?G?#&B;1>trR@JRM?ek
zx`Q+f1TPPy3P7&(cQ!cEl?+LxLxLP6Yd~~%GOTvRZBYgNMAEyEV@(xj7XCT&q%7H#
zb3!k%?A6a?NQ|SAJbTnv`D$NV4mcG}Q*_Pr&HWt+8ch0v-9z^F*?;3u2g?7!p>o08
z0l76upQkE61C7c5MG>Y<SI8T;F<@s-R~vb1K6?gW;jA}~pPs7pqvSEwHdh(0%~a7v
zLnpHcAsG1A=w<v*J}OV%zxk-apcq1wL!VBsL+1c{-Ut_EoyLPDMJ1oD!vXmtUuPdY
zc%bpf`PHNY7w+Ypw86f2%0v_l<WFPgPiJso<MvRgT>-}q(L?R2(&JPoa9^o{VvwAF
z5aEhalyc#R6l^Sj<}_Y_A{m)25Z!tKnP(96AEbOJ;#<$+WoA~}E1Bg&2SN<@a10t`
zl_6ncTL9|{75fYbVLK>b1}c7P&~m6D%66MHD><)KVMPr&7~4FdFNMBgN!yQo2xYb2
zb45^sdaND+r&6lCb>dX%++@JkKj6nn5}j!!c)e=0V49uAzTQO7eSUtr_f82r6m*1X
zGmCC01`%jH51CICx70#9xRi_Z-^vgXc??q9@Yj#$9qKPYoPrdNGla;yI!}EnvR*(W
zO*K%nsUVh5$%<Ul1(IlgOI1SuUFD3IoBJH_THt1M0h@PDfbV~BRWBl95sDGbQDBo|
z2>eJbC$5Ol&{qL_a8=}X*i@cQ!#fjFCc4mGBr~A_^{#zS9(oQcuJyhK3y}TiazWGU
z6xEA}ZBaf7!B?_11m}S!W}eoyG0<~Ay*(DNs%aCF3h@sfjGLj7DXUdHQpTkB?`et9
zBcZ@WE$}idd+?^byu69~M8K0JDJdA>%R@cFi14%tka~Wn212G?koY4696Y2eXTYk&
z{3edQlgx$HYA7an2yy`GuK=DrlDG+#4KMV*c?B1J3ND)NqPmqH$Ot!v$Pj9Ift#Ao
za*I%NNd~I49tf(Bk5k>eFeV*wV5*J*n^6ezERcVmTAllv1gcck_7t;Mj-akWNLWaL
z%mOkQ8ib}woZzTdL9nuKKyRsRq(XENKcs()>mU4Tv>Jk64V@wRANbWE5L2Se)QAXP
z6}{NRtk((EcWt7Z>>Fz%j|TC}a8-*S#)v4woGXUB0|aoTL4;wTta619wChVeO!@<2
z<w)H3FA!_-=aCoC+CDR2q4uD$fcojvJ*O6G1sqJ$g*W^4FCN$*-Ru<pHU1a#p=v3d
z+YlUAZw|-Oz_IOV3SWOP3wB_`LP8?GYLajTeTBk@Yn3#_w}g^FX?T9wpoQb@54;%%
z{=F`wRAPz!EB3mtB=;r)6+EB41f6f(QsULO1LBuAEJ4*-&j>O?N+#ibzkq6M+nsY3
zsVscwuCT1^BjD<Q1EUI$X8{dlycAP4_lo67!lDrfGBS?Cx9C887>*GwJK=ayGS;=x
z)R{<i8C}ij!7zrDf>hDpg5xsbIF4B5z^kBBJqpGBm{X*<_}nHw+`mL|?46U^H&eo#
z3x?DeRs-H3OH?>FEK6P1_$HmU7ZX9bZ=aE&ArlM<I=mzbqo=L|KkF<A)9;@ylHM4b
z#yCFD&<n@tvp1pF14ZP8hC@G`vP32|UjDA3|Dv|4#C_dg%}K8ETa_5_1(biB$LT0^
zE1{_A^Gs8}?adDV#7h*fhTWV$^Ae)LUV{tZLIk{|;|F0N(U-eD@7TWqvwaBw!dq_y
z^dn@`Qoy?&CODza!zdb#0ro1MCKY(=oJxMR=BI7H)l|lVGA*6Nq8W6tmA{B?7lx!s
zkeLatzy&an$@Nf{s$Gs;35ekM?ry_P+Rpt5(^RYjTTOyIjH!Jip*NIC`_w^?UjiEL
zcg1cM4g+-QN2Km{2&NmY$xgU^oaD{mXTVm_yEcNh&t4UaBkEgQTYr$-uLN@N4DM66
z&MDH!qY=Dn;$%XHRJvQhThQGe2HkBic`MrSYn9G4=r5yVf=}V(V=GVDY!H`aDP0aV
zq31wB43#U1<o*p05t*?r5J~gF>|>xvLuCC&YOJUb1o}u%82&m!$y?jGWn7dwr5(>l
z)3I`M7hk)6z4b=m%OSUh$+ycsOR{5b)I*^AOQUdA19I4lLU^wtKo&3RdIR94g0YE-
z)*GNlbgQ3iSl(Wg9d+B+$-;mog4(9)K;X@*F!Y*Ofi#ROHJT6L!Yay+q}$0W4$gCr
z7hjCHQFh|tSo)DrY5u@v6lkptp{>|et}Iz%II?al`bC#~<8P!_#QMabh}0lwRU>wS
zONMt0WLpMECEq8ht{`P1*!EIyNR1Ele{uLY8Pq`?a0<FT4AH)};Q4)Qym$anY`y+H
z=ZrA!V~uAKW255dPIxzkAnIEd{xk|mVZ@?f0L>QJ5%QL>L6zd&6{TG@K$<=U&&dvG
zH9<`srp`sd3+1veffWK5ttaz#A3cxr=W-22%uGXBLdozgzeR`kGT8tjxByXTG7VIC
z%8Rw0_xsB(=}}WtBOaLd*?mC7JPpqx^S|h-_xY*1VQd}ivImWJE*ON7N`D0~A1w<9
zOsWxWWtz~J|1uG<G`@!xj#P8je>L301QGlg#Csb%(*`>0X)ylegA>J^$NmOl4(#F7
zQ&_{xlV(_kV?1D~qYWpSfyFW)zEnXil?R@emh1^wHVFg`gK0)YQquVOfZ}2$HB`lh
zwkys^4y!r@x(b@<99V)IUb}W^ZKh9~mjbll(IC&(2MCAUQz-9uocK5~Q4{tM%eFc@
zJ3FhNuErwT9bsYNa73Q-dc+)tB&2e9-Np#b7w~Mt-tB!^FX%)ON3nIXHOLZ|g;57^
z2XM-Bamv%Q{#heerdC(oQ__z`?^#aeJ&&QeB*Q8HrmXsY8PkjS?zbg|;M&0^*p#XX
zq^#`iB9H9|SJ)3(%LOT^kRdC37}Y@5FZY(@(0|M_nMl2*vQKXt*Fl5Erq-w6waSn#
zz3_5D(KWgW*$yF9%6#&!ZUeH+9jcwVw!c~b?ym=aCS%Yj=5c{Q?TXuZ%Q=9#<p8J}
z46)=egGB_;q0q=qh=Q7h3Q6EM_9N`n#!pCr0VpN7+4su<u~|VHp7^E*3i8#=mcA^2
zj)fEiWQ;>fGJ*X2g{S!L2hcu;qkQJRNFjCPe{ZjJ7()T|dB+#-(@fy>GM69V1>Y7@
z#xjpNl#b^}9($A$Gm0B@o2vi|`GiQ=k5{xe5|K?MDAO4q6RG!o1DT=0?#}OKcM?4w
zn0(FrSWy?z@C_i1eF~8{Jbq;nG~l`y-CNW~YePSPhPWwC)?d#3&<^=~wQX3@usJ;0
z088-jE*Kjxo!htA{1`Pf2<z5+;Kft3B)F*1rmyJB(T}fm%F#cA@+jf_w|Tv=l-hJR
z<t01RRspikqV`kXy+dtU9^cXU3kDOW4hqy9NU-#vq6rFxs@El`((U~XE<X74<M03e
f|NlS5)pwP6V{e-~85G1Iw{})TSG`pAa?t++)dcs<

diff --git a/docs_old/cookbook/images/nvidia-income.png b/docs_old/cookbook/images/nvidia-income.png
deleted file mode 100644
index 9727bc272fd7457b07f162c81dff8e987278bb48..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 185563
zcmeFZc{rBs+ctVhQdCl=NTiuah6b4`5|YAg3W*Rx=6Ogd$<TnJP?@Jp$&fKqk(42o
z%w#HK=4tI$&-1*$-}`>=_pP<9Z>{y$vb8-Ab-V8SI?wYsj{Vs8{kU$QJtMb{j-8Gm
zh;{O(WK;-(2H#T0u3m-z3H2)6#s5~?9+y{JjW4&=mwfT_W;<C;J5?)VJI9MQMudr_
zm4y+%t)Y#Pk)`crE4vZuVoAJ+pM22?8>5SMrdF2RYNi%Ogqpn(x6l!8WkWk|AweM#
z?gIzJ4hV~h9^^iz#C=vqEikT1fgrdEd6{EsPEUTeI_qdqmZV0TZ%vl%;`+*1PQx0k
z#^kueQTUUBZF0sQe&+AI%{p|poLM?s4G*5#e49Qz_I~06Hi5j6#Qy%=ZGOI9bbGi*
zmljKdCKh$R-phQ}JTW$UVA}RWUSFi`@S>ZsQ(azPzl@Bu%s*e0`|q+W3C4fE7CqRA
z9Xx-3CB6CofB1iF4d%Eq46D}n+lk?i|CS@;T8H|SFV6iQFgvZDrni{#IQo*YagxLA
z%*=v7^CKZ6zx<PzE?o-yu-CHS+bfIYdaL%L#n&&VgV;7<b9mXQy|955zW)Ado+*QN
zVkw##&)?dl6^#4wQWJe2B2=T2HPQlq{`|>PS5sT-la-al>pJ}X$oC@m^X>=P)>gcV
z?T88pSZCIh<fEbIOmo3=QFN<>i(0=}Mbm7I;zI$`LCdBjit|my?&=V}M<S*hwX@A=
z<mBWuH8o4h%BU^Ay(~@9%DOu7qn56DQR8lSxL<m@V2Vy2?kZM6L0+Ex*N;W+u7}0M
z-hBVQ#iekoo}Ql5Z12Mi>+VXERvFyGbA<|mo00L{W~Ijm*Ci$<#)eKzPEty<T&IV(
ziHYUEwR!D0QrY_2k%~w?pK~)U%_^_u_h{#d@`{R!HKF`YeG!fw-QCym$+N#t64Fji
zxj!bbz*mRA=h-x8U3#~1W@e_NL0PH-e|h)*eebFuuI|sN=MUTTtd^YX-!5|bBO$dk
zW9aj3RwMml*{b(Xj%1xB($D8uo>EehH8bO7;n)7{uRUG->x~cn=;UOD9GClz>$D6!
zJ9p9?IdbG}b@iJ7jzb%mnSH3JS1l;NPhU$<uUc1@oK`faW-luzmz=MotIL<@qwP|t
zr`z5CDZ02u)YsQH;qqEyBO_zTC*0WBCpoT=#GfCZooE_!Y<JhVw7FaBq*d{vLs-iv
zEZxuVd7F+p4wuI%Jr=b6ak}w^E(fuJmGxGnglpH#Shvv&24Z}CJl(8`mzWi1x3I9F
zVp>lZYK(1j8Y!F}of$JEr1kZ=L-%QUPYyL34%El?*T>R(EtD9w7r9&i{E}hQbb-a$
z+4=3ecZt7Umft#H%i`YJ2@4CetzCsbnp#>CN)t<&ZJx{SlYZvK9*d(xX~kW0lS7AH
zr}q8$5J8WZ5+zd0v&ui)i<h+yW~<>|B^ZAE_@PXBn3|fJl&pWy*2sBb=77vfqEQrw
zk#~|Er&dj96>;0wxB98a;vC`CTl28{eVA};sOPWjW;HKk94TIIb?*K986F?BSaEk*
zZlR;}dd+th6B84=q0j4D^X<#{(h9tTB<EJ+oS#V3D@1bMym|GSHG6H{rI&t2&CECv
zCGAU7hbYa1P09BjKRzEPhO>QR&8A%mb*k}aD%uKN4><hPT)XA)2E$7K(igg~8{cPt
z|E{x7GhJF$&c#LSNA2TVq5S8c<IJbOd?}NvojqE~Vb#&|?m>9w?c28p?R&3L({?S)
zPF!xy7ya#0sLhayQ~hzkPEc6*#L1JZ;?6|SSy@?WYHPn~XxOA0clwyRy85#sonjA8
zyy3CytJdio8PRZXaLmlj8FZE3_W3Nwr+CzBp*a`(?awZ1R#II}w~k$;WVEvk7hXr%
zwQt{Av558aQ@^MQtTh+DpB!wWK0D6t#p5k%{`skS^NYc3b5>T?--Au`j}KW1G*;cc
zyMdx{<;uS2(b0@-YzZrlEzVDC)X8WC$9fX!9*Zvat^12+R%pD?DaDzkBSTwDYa2KB
z)3;hFto!w3u-*ECZ2IQrcS=g6%gf7yk80ZYf80iRorymoBeQbFiWMEby)+jtTyUG|
zy!Gg)qkzlr{%f6Ox6F%jUcGu%QFOBX*$LXF_~(`H%ga|XY!xp_I2YZ~*-2HX$3Xtl
zWRr$ZwXv~r*NEq`>$8y(FDh!f07ZFoBGj=!L7O`L`LshroKnL19D#lO{MI-VD$RsW
zjwPp<{ou;$rG%oQVq>xf!-HK)Z*W`mArdaX_pDmSej{Ae{LPy;E7q@Hf4QyT2)U2H
zfB)v0KP4mc`{6F~=*8F%eO^OQao0TD&swOrpNPSM2)Ef^ypYLj@^iBJg2q70D~pDn
zit_S=f@VEerP$p&b|l!jy3X^ojE~psXZ_mN({m{!D@(QaOq8UqK}6*$hOPUE@(&-*
zY?XA|!*R$`t?x=}e)j3^Y}1Ba_GcF7CU+h>l=CI?g8O91lf%}>5wU981_sYX%o<NM
zQAy+Gnlke8@(_9_LzER2Rh@CZ9~jq$C)S}L^YGMX8YzZ+j#XsM$jN#5%A&<DARs`2
zOWY|v&H{&FldG$1h?TMNuDH0k!K^!X?mUrjeWmjY>H59`f><TTQg`xeY3Z>SdWF0^
zJW=jbVY#ngCyUe5(~pgg-d)cr=Ii@*>Q@V;&6jbz8i$Rotr7O2B9N0<vu<6}Z`*~L
zF<z@5@52m`4Y8-;w&91YUgXJD>@{DOk?~j<lM6Xy{GQF)+L{q7iKypo{q{|}tGm1D
z2>!0Crzd7y>GWxtwB|cV?ggW7=zS_*@4&I;o6h3r84TMyc$8_whNpuC*4;OKeZ^gV
zpNgHGn^VATHBDR+ue{(k=99F0|9)jL$KjOmD;5^9Cj;0Ax2USBYUNl)@Jcz4U!r_I
zT*%Yi6sN=y5_$LT(|SG^s}m<qR8&_h2n$pi)~8=QQOkDo=FKE=qz=>inA>x0qNepS
z_4V~?U!y#iv;q*yw{!FIZkOtx?aR6HRYgchsM^g6nQ4jc5A5N+Y5non>ye@d0}b)X
zb@%SwOVY8XD(UaP9I`akVr4$=zx-u!wl9h@*RYb1l|f#NU!xtWW^xS$EV1LqW2>UQ
zJhP>Dyty@(mlkV8^te1`p5=d-A1S^NXX`&X&>$DVwPnk`y?giGfAC<t)WYa>&GYB!
ziBCh5yn>guoc=g8sgkBA@ioKn)&kzmW`4?o@zW|%H+T1)M~-O8$dG&b;>8Qzk+=7@
z^YGlicW;}erR54~JF$p`Zm#9KazR_=PoBK#wlG$e<c3J1=%>7NK<T`FCVmq^Uf6kN
z^qH^k`zWak5r-_l9~NC-InBq-EiHVqf`*Wq|0x_YJTkH$iGYdmUPHXH`3k*1XPb>S
zXy3klmwtWAzW?yy?uP=p^((jKSheeQx${{q`_Jp+q;yvXJV;5|C$FgZKtMM?q2S$v
z-O08$*Kx$vUvTQAkyla*`T6sTR_@hjyB?{B5*_}Vh1HxR&g_mj>KOVp%cKU`xCaTk
zB63=Us>IuMGZ)wXu&}W4f%vE-<hEo*W8*kq-z|r)O2;ZcIed=hld6h}nqW(FvxbB#
zJ~^L|o-7R!^`mWY(q#=D-Tks#3^n(*%k2;r&iX=hC`&EI)L-}ZZgl$jHB0XF>C-E{
zPKOD}hKziQzD0McvZ5mKHzip;vCj<mw^Nc9`NC85Pq}6;JKX=uvQ_Ne+qWB(m6fTu
zk#p6APy5jmKYDxR@#-e7*{@%}^4ym?pUhp-)!Q2zws-m^RS8*gY%X5Bxbw4?{_J=!
zuV+;tC+}ivPTNLq&s|!_I#PCw6ifsrnT&N+#D*Snm>i_k{*1!bnT`jTB-ch9)sUg1
z)tnk>zmTBU{j~By>9+2UcfSW3WMy`Ui|fQr4!6EO>%O$K#JgN_joQep3TwDy*RGQW
z1_nD09#oILkbCu5L_@Z@Mu^@$Lci2@m2@EI(YPa3PU`AGTP6H{E<ZhXU8L{y<AX}Q
z?n|?MK9!xOb0*(uaf6eE8yLUl*~VA94#YlG%a4wZR*>>sD)=$y;o;FsM;IF!$zQ&l
zV61LZDz$0Trm=~M2hI~e-XRl<_DNU}qu=vv!)<X=DSCExcF!_4i`gX@v$PbtPVbPA
z&{dT~5c&CQoTDm<&&!#bn({I3WM(BqA`A1a+9hj`02r|nT|a)PC!`|j*!=E4r!;%x
z#*K|8T$>mf7b({3qyZg;GGD!_jF0l5BXX^~R{@KhH^3pf)zP6(NaGAT{Q5>i%pjfL
zyNh0K_4@TUZ_D>CHfM~*Z8B=z(S^{YB64$cp91aiF5+dh#3du6i%7k(pA8FMJz7|%
zS*{sG@g2;yXnBQ53|~N+Ozz&a`wWv<`{&Q<s&RXONk6-J4CnPVj>RfIBo^A|t`Z;f
zZ2OMetHp4B`0!!%+O;z3>MRE>zTJ?2u<PQR8*9RK-2j*ruei!@+_cI1*EdnZD<Xn(
zVR5mh$(V4$28c${f8Mgv3m72SX)%7_!fS_LB9Aqg>|0}EcI?=_J9$YW&p}68TKe`c
z*ZV#`G(adfQN|2$c$`p9I|l~VaC=&OKtby)V~~<T&|9PY*fi;G0WJ`+{I2=HctKpR
zz==_M$qBH01)=^zXALba?Mw6K^#F`pr+$8=R#k0KNuz{yPl`+*6cpU>`SWL^{J^of
zNL=hhU8F=)@kC!8HDTFa6o93pBGA21b9>f)R8Q4@hN9KqUMvOJe1n!@tJ<IyFR(r3
z;#3L;dd=U(mqt5EXS(j}Psb+j)5+a%;?{b~>zgdSzjE3ucSpIcU@Qd$Cb4Z&t7ic9
zjgVfz?!hl*Wzmh}eRWY)2x6j#R>yBh(aGsBDgnhncX)VsaAv%>#>n)A%}ASs{m-5r
zwLR@lRcE_1X@I>I3o9QyKBShcuEfcDg?-fZ<@Rmcs3csc%7L)e%{chFOk36x01DxC
zl~iBqz4MeCiZgYZole|BmU*Zxrv)5Mk6c!U1BQ2=%&};>IXltU{Vm%(*JDv^mvW@e
zW!nJ1J*x*d3wvx877`)|oaHY7VQucyx?3b%n6$LC4%_u_K$*R<Q!%vU>(|X32d_xS
zpNa4*R5uE2Zus7k>9(7mEysDJt*|2WwQb*ev&MvL3{ndlknl7MoDS4CG%#{;1p&{x
zEiX>bbX;c`of}HqE-t>1=y-A$4^NW9jAsyFLP$e>tl~uj18<~Vx6xAC^o)$XudV>K
zegtqJN`j=8d<zQ1uYAp<kzPvSv$7gdFmJAXd}yp@e=+g4-Tr05<}1p)j}|q;4XXk>
z5xYA>2R=T#F4mT5bjSE$*Ml~fA%3^<ce~AgL|6;<Am{j0@7}d5sr$J%`%apRcaocD
zY>(ajIeobb_?SW%^?!VJ>dcvwdU}!J+0!p0U)3rPO%F$*9S}RCq{JT_6&$SgVoqk|
z*x1-9dHGK<%_&-&wx9HuwzGSw!1pC!^GxyP7qY+DbsKXzT*tMQ<2KWpJ{C2<xm_->
z1A9G#vbuqtJ;|71*Up{W_UuV0m(=*-BvD()U*nkRWi7bWRdMI;f{`EH;>5%R_1)$9
zgXW+2?Z8XL#MrY;>P|*Ux;I*lVGj@34>k&2i1PLAPxGQ8Gm6dl#6%g2Jw3P7N6NRK
zkDi^lwc_n^*->%vEkHBGxCyO#=q&~=gSU6Lz5VcE6+r@RQM1N9V^7dIr05m0OM4Io
zDD;P&MxXv@$H$<|j*gCQ92)wbmqmBEE!UcQq<GomURIUl;%`;A`CmCRGu=UqOiVYh
zH7`vYPWF<h?xVR6nqKuPR204p&WgRO16c&u4+AT%mR=gFqCsL8>75uK7s+~1{Vfp9
zXi_)YgVNGcDl`P~%2DbA6}6dY_xZCuG&D5+G>mscLvMCh1(}cY8U9ol7g^+^2N+_`
zJ*}vC;>wkjxVGIR=JKHex^zd69&N-<Y}>X?HuS(&Ti|PZ9l_T^!osSBesn!`_g~xe
zD%@@co-6lUn&46j9~=5nnMNz+HnZX3{tG9OImbSqx4QT>)9A3{@RqpKVb90!xQ1{f
zd>@-D9I?1Y0^07~p2|RP^fbD%D-T(<$pE(IJN)u9ZHU{(%S(r=s8fxhnV(E-kp)f<
z5hMWG)u?QLyg%o8DN4*fsXP<eOw}{k<M$^yY{$Uwk@n)_%|!?sho4`raUONx#%1VP
z_DQ2DP<vYS=@TO$q-^3O8r>7AvNQFH`=7bbPhD<#mHML{n5pUKlcr>ihirFok7YPn
zO=3pNGgVR_fhPB9zg*q!KFwJ)(?xGL(#DAkt{@(ThZ~@T3m!Od47)Ri%Tc|JQX)@)
zxYH<s;;&igA_`W6=&5r^?l*712QR(7X*V~|44{ihnga?5D<P;~LI8165rFbH(R6IY
z@10UqEJJVF=wE?GOT?<J;B>RSgM)e%P1@!50_RN+b}Or!hwMZbBJDSpZ^|m5Q}U}N
zNAZc$8Uo=;sX}0sXPY;3^B~vUHQp6$omZ{WwnULZQfV4P?jos=_wP@?e{|qPYLo^G
zdOn`5QXXrGFX<PlT_y*8&<|b2vGu)uyEN0N=3}zPex=zJG{+1G@_eFszVq^>$XYJR
zg46EwYt|f7QQ4T8m36>#$&Gn~oxMHf+Sf$Y_&1+Fvq(xxhD+<7ITQF}(xSWh*BuqD
z?n$}t=ZWwUJ55bCGzC;dMn=Zy)YMJAG!5Z!e7ZZxvt?+HaeTv@Z{cJ#E%ICVgCbe7
zS;}2JNq&<fo<~3W0LN4NGV7-<aAqp1mPjwMp?oz0B|8_BFBBI&5qHi!ZBIg^SUFi)
z=9Z~LZB8dbTJr79GkWRBv}s^sqG52<;>wjBLP9FGdVjZGS=*D*b^Ulk_5mULE#=M`
zRvvwfEz%%-eSCc4_$yxz`1$^9$x%+CB*%^O@7pI6-;h|>*4DO>l~v=11Dn)t^gH#r
z)`l4`Uk0OSDO|mp9@pmGsz216mbRIekB^UWuLo-fQbX<H%5g`lgG=HjAGRpH_G>w&
zE;C=C6e*tf#Z~v-5_2;x<FBl5WDDalP`|TQ;?sbgfu*HDXlSU=<&WGUTeoi2GdM3R
z!?S(6Y+CcfIbE}_nXxf(aY++^II-_?GBY;<Q<%0D<Qe}=(|wJiFfcH{TU%8n7wI~j
z#}icypqgL~;z3tjTwE>k`t|FLNfgDD)Ktnh^Rz<sL5Y;NPR!Z#SB=`v-@9MTS9Wt<
z!^`H13R(A&Eu5Ua<+nE_{RG|=ZLdRP?f3ZcV+9b2d@nf;GBSR;z-dA<S(LLR65Rdu
zK2yOeou`?4%c7#9KDTbs=GzT;GxMpFq={KX%Key_7&1Pkmny0E7mOayFrr<@9uuhO
zx{mE_)R<FmHS_y$QF<yWs;f3O-g$XPP`^K6QQk)!WkqKA+(_Ru6716RmhpY01P3r^
zdS<5g?b~YxKd0Vd+<vm7F;TT2Jwsyi{tu{x2MjCMHYS{-W?^9=`2#et*NtjIO{+m)
zP$5a4ynK02L1E#uh5?W;KFCblK(2rsl&cLFt#NDp6fI}i`Nrq+$EQ@Mm6Xu*T?2^M
z2LbU&(p^mZ<z*^cTic$Vo&?eN2=<T9W$D54kVMgC;Dy`GNDrmxbx2aqZutA=RoAI-
z%&ywo%Yt%a+_Y)pb?R;PdrKi2g)WonuG5xwzrHa8W2I+hoj_-R7hDHIPqLwTY#My&
z<I`il0Ra`*gU*o=25=q-5%RrAoYngJP|@rUjxFL&XS0e7(1vYbWV}`xz)=RO6`!N5
z#i}^qg2$|^u<9@<negS6#Gyln7QkPSP`I#A0<^;?d~~<zR!6fO0QT|mH+UT*JG&n~
zd>59$Hz??qUYZ=h$|&G*a`V)(@7Ub-VT)^^H|aV%I}wG{=*g{@m)x=Vv?v>gU4ADR
zSN;N#k?t~SW;akz12~jxKg6P<s(P;2!yVDPakp~hzP}Go#F?0lEW_Mb!?{gnEwA|3
zb4jk{UZzxdH@l(%N9-#croKylLw+UD84~}qAG+B|BK)r}3frF}`InSIx%j7b`@i$U
zIUf`aYSdKxL4MsQTDSu}utp!s{M@veO{{NVpt^-S2<S+A+kVfFAOlAiChE%C+O|44
zIP|yVjAU6NG8ie=e?EbP57L0l$&=476nkK$y|c4LkyUO2#9jp5^5vBU2S}kvd*Njj
z5<91-r%N^E#>d3ye@@Z5++HMsK)h&axgNaP{{8z4u!&!gcJB3ofnS5pGS{ZJS~aJ$
z?yx{{PGh6m4IiJ2z11Og`Fj<>L+W|V8X<`>fPV8rP=TS+yfWL8W0mZcRjNNjeO>{}
zM!Eg<kNop#IVm^y@b95!2E@YtV8_lkG&;FgsnGS8-`QW}yN=^v$>+~gn_rFxHK(3G
zgcP)bz$se(XujMhwOmTe<wkIOsz1B^E^B#NY`;#f*N2Fs-AJh6zn?tW@*U_0B!je?
z8q-OCwrePSD!qNZy%$lGa-Ao3iP`mg<=YRb=dSatw4U95<x4u1u*nA*X=&o-dd|6&
z$J7KF2!E7-2r>INU<j6krmy2%L~<^rtpj$WbDA05hGG`$?TJH2vftzx#Lel0!n$f_
zx01NhUc`Cg=33{_+Qkz<*QNS`CHPZaJ;+cvh19n6aVVq$%uJfMCqPD7TJe$}5_vaR
z0+QK$|9*UV`3rDQ)R$t<Vlk`f;Z}TX<JgzvvU}c;GAsrZNecwJf7`)@eP7%*Q+ya7
zkC*KENjyY6#bzJ2A7mmoxOnlG6`Rnd)hq%!zPQbjsw!Gyv^z+u3nX~B*!r#FzBo^B
zps^r-39_xd-28%*Wxt+3u)We(aJrN1`u{=du0DVMJcxQAAuhmapKp1fbxDr=%$YNS
zhYsC({FnoD=?#1mo_YJlIeSno;7CQhLPD6YT)EOQG;{|5g(bXH{mK2$M1j*74=5~J
zBpO_G<*HRYyLPP-i+J&6j%;A?k)EH45F_*M!-uSB`>umoNfrWFy?G|S3_-6Jjsphr
z$yg_F%MtIT#rd+@+I6^ttB#IUcekCm3iaW5v-@ZV6|&SVFx=8<&C$TFnniBTw=MWz
zED}!}2FN7l<wZC-JC`;$vw|+$hQ|I5sC)gH(N3p@nGD_*kQo#?tUrBf(9FI{BW%`S
zo5xrM>tW~d9V9gjPCYk2-=N%=(dl<TBcL5g^n>&~fwzOq6z;%`%vJ?8rt{aYbpW|)
zX?mPESXDT!qP=L#o9t=VZ-J<`wqSW_e)RY68?muF*$-av^;{gfkmLp=V<*OmQ%y_Y
zBuL?VV3_#o@A>xhz+$qfo=sV8PmfWP3$2l+$6ppv`!F&x5ZuT!RG<X)RCa{Sxny<z
zg0V^teB_w8l=%4F`HmwmB$p?mmT!Ps2m4&(R=7gB?GB&Di;gCZ3qH}&+f!0fHf-47
zMM_<7ZKIz*{}?UDK>3V=^X%C(K)g!TNd^LIx-#MnEn5Aa-ri1}r<=Gm3O$NPkmvGz
zJK^={5!*|P78cZ-tJc<~fSh+9JWx$20swpl&HzV^)H+a@1WoIAC~=42*i=j7KMCIv
zAiIc_UMbqyNndPGryih_(Nj@T5gE9{x}OCE5am{EY^>NSZg3)@fQuB5|Bg8pfYdbj
z1kI>sua>6fsQ`A-a?r)GJ~wZgeR~N~01UUrEEEB-%}J84b93K<4W!Hh>n0bV#s`OQ
zWo6xsgUwr4UM>U81$exget*H!da$33N6wu)X8>lfG6CWWiJu7IwRfQZWo3FmjdgW)
zKC51_fq{X6<h^g)Sjot!_1Yn|8cIbd%IR&8>cj?c=|}7bgYzu`V)1$Ogo|hln*oPI
z4k97l=VD?^)JTiuEyi-G)ZFoPa&k&|MYUn~?%m`iaaKFS0&8${3C<pCS6z>bAa?B7
zAwy~dpyr=94^c=C?MIlT<M5NiM~~7I2Si0f`XXJ=xPKn7%S3BMImpkiw3E{aK7_x#
z08I_(-4Qf*9zA-*ls^RxP)4RFt4?y0qoZT(#7Jl7Hcm0S4UG42GuF1Y4KHh(nkWcA
z_K`1NzNn0-{ta^*R&|eL{v$6~0XqITs&8nx9jLM%NXQ@!^w`ttmX;P)=lzxbo3{%H
z2;j3e*`B>B)!p4KI;j*PMrqLNc2z5Mu|EO{(r11*PuIvunupY`Gf(#5&cba;HhAcB
zs=>vJ=lYjGDk!UJ01sB%Iyi*xQVLIKkHLR0NA5<Q=iU5AvhMxe|7kmOLBoI@NSi{N
z2u~1v_C*gKJdlx-Q?#^v0RmIu?Ag0>*B$Y~!6#sk&6m9Nuv<9R@4&Q)TnogRxa0d4
zT`i@AS5WXQXnPr%9Y>FzKN+}XFHsKq|HXx30UjQRg1p?%8wPr~mS=8}`T|}ka*dQz
zlei6?sl5=cssjL{-Y8}Y4i2xtN}b)bQgA)Nva$}`a1RJ59v+*~j$@QpT3N>785UpC
zpH;ge4JjxqD5!LI_V(^X9}9jQdYj+FhYyLn=<1TjQ2>%60SFWIKayGHT65+~!sPv<
zXVpV?Vgn!p0AbGqyi>Zegbm79+qaP~a;`Je@gGTrPO?ucQwDH=Y*u0)hlSk^3JM}M
zdwld5<S$;U3P0wg7m%@&T|n32=%bho{@GK<kv8Ug3nNd)?2p>@bIZ%i-+%m=57Of~
zXCQ@hlN_`kAlZp@A8;Jj31MYsj;#ls#|Nyq3EVs2+s^MzZtiox&rPnRrBzf=@B_qs
z3o7bqB_*xd-1{mx$$jJV=bm(}@Iq=r%8^%B4~>b9l?V1zYXin_Is(;$G5}WPTwS2G
zbRKe^a+6f|{I8)QJ_rP%zkl1AlIL%;R=!=LukV%bx#wb=6V9Dznt()~`1|Qo?sxCs
zV=Y0@CA8P=RgDW4^FRR)4*H9n@e2yl@O;F+_NGnLGKhx63G~4U?-b9RQ5SYXEhBed
z27EiB-}FP+L5uL}c>pp5KTs`iDn5(N&}V+i5`gXfCr?rmL80k{j3DnnJIMNekct~Q
zf5W|TT7r|4lQIXAPjqsAc9t1Dw~2}3>C^sDKNHMZ_N-P94+}FjmSAD+*M4;+Od(3t
zJT&eF;v%8I+v^ncQnhd35k!02T}KLSXo#%L%pD>kYQm2*JE9;o&CJfeD{!9ZL1_%1
zfN&sLV|}0~2f}=F@bu{DS?9_u18A{{^}A0$;^vvnTBfT9JCc-Wg5I2W1WX+hBXIh=
zzTMlmpFqs5^dgDP*oRDn$dY8LX2x-dcNFC0`{1$VO`#1X4PHeE{w*c^!Wq+uUEWG1
zO>+C^$9hEdTn2WANM3roA!G^o-u#8I>(rGs>(__ow7Dokj^g_+)lGDuxKZZ(nEBN5
zb8>Qi%%)J1L#q{)m020@!Says5^Rd;i`GI{M=EYePqBd{^y%#uD8aV%UMg7i#5Z_;
z!CSPGcxv1CAzU0pGik^G$a(wrZArx)*3O}!<h>KvCUZ%SG;@c(51a&vOaKILLSjQ-
zzY%Hba$^GHx0j~Z`=UIz9y@l-$Z``w=4J~go38RTAg*7C>z)KBfsaf@0DGHRl&?S&
z38)pks}{Ww2^WEJ(oO2N%E-zlijD&jQ1y(&pp#8LSidK5Oxp-JuQ??k6p1h}h)XJv
z?0*pWys60ZKHu`4#=2DIbxI&9!*Ws<BmW>Bz>kE~paM_2xrrlGuMRfueIjO0hkk1(
zV)8mdKSZ-VYYVF(l$r24*j~&UHg=15FERb}Fn4POfgWvA7de<UHuq5WSNgA?KU17N
z$T~ed97tl$((APFmhjq74<IJsNFaE;afl7!5+Kb6U@mU~5Ix1(?~G)S&WgJT!0@bj
zNR%v;RVGpHtGJhsPPm!@G`_E`ZR&gOv5DyTl3_TQ3t=l<47o3E_DaEeL)rOeD9iR&
z7Ph>+{OS0_(ZSwcAG|yDThEgSCy~0hpd>{&=pv*%w-Q$&J(3WI<hsV^u)DuK|B^>h
z{B}PI3){APH|_LDyPS<p=C`q&eISVA6Nlun8?vGN9UUD{XRj>W%CnY63fl<0GP5*4
zLeffKzG#+LR+3W8HDD}Ksh@85ELQoZ7LHV5uLB&SGQW~Q?Q-f4IFxSLns0L<EoLZ6
z!j%Q9$DgTlEIlS>ilymuW&$Odg#?vIN>qg9_uOlcY%VS>%;q}$Vn-neL<QL{B$Qe1
z64{%}mX(v!?Di@>{cIIdAD>WR$E--vFks77)G(n7ISFI?hmW3p0wNCmngLJ@i54Mi
zW!UzeA?-L(ZsThf78cUctN;Koov#{MjMy8LxwMNdr%YckSkuJlciqh9y~DFkLTL#m
zmq=p&)dYg?FV7E^BNYf5-rdmOYg5wU9%4D~04&RP_$oJKq8Q_Yo0mIv1-DmgqI!!7
z@M@~72L=Xe7JEnln2s$@w;PW4yo0QC1I$NeW+sS!t`10Gb0@xCGB-~SB)Qez-8*+O
zLqcz{qTH=%Z|6crIR-Q&vl0xUK=Wh(^?E+_6tABB*H!w-ETv)QddJLM{(nZsXTRD}
zXQRV=XUv}DWew6^Batc;zQ{l`z#|~A9!Rk4{!T?fWRX9dCi(-?OBP!`v79hsZ6vh*
zo5ZZR<bc*RmL$(<%pjxAH!A8v?1jg`l)MIHL!Yps&AE@$Ufg}ou9h#F2Ak}Iv)&CB
z{sieZYISU<dg~eSLK->v#}fp5CyqZczMSAOo?SDJhTs4MX1GDT;3Fb;{L5>{{V2h4
z)l3EBS}Knbn<BzjtLNwT-(CJ@eC#eaKmR&#ki5z=qhvl6>w~XoaHjT9yA0BVKI9WP
zJ0_D%OJ`-@)z(s~KnJN-6)#Rbj?-~*$v1w*NWd~C&B;-`b)EF4biP_rb1^Zw1#rx>
zV@FJ*1IzSal69EL5rL4{m?2b-?V_UD$J3fwt8(jT0G97UiS+dJ^yfUvf;{Pa=Z<RH
z7p|iYrzgb5A0Wd;2ck(&&?^=oX(4-^dn4~ho?*EMzsww2;Wgdn$^h1DbYcRe{ZHYM
z3-WcR%*@PgPs2Fj`xiHYrF5H6%2sbfRU;w)x#*MIcz9Nre@-b4-mCiYrKze;zTFz>
zrF<7Ow_=1o2r|I~@0Qa#@Fo{=PB$=;O*$#cmHMGfmNhoMJpQYt$o(i%(<ihvq&I34
zaUinv$MEoaf=y80D|`{%#6dS$`k;rL)YfL-b^6iMInrvhh9_7Km;5jEGd0Wv`_$UL
ze%0b+U8xDyl5yk4=L133A8XJC;IH<egdG4so_$jjUcPWs02&FBeMcezDX1bGfvz)p
zFigr*5&?CK6tuvAU@47;CqPgneC})cUohJD=TIUhPFKZdl0s-IgX7dst$n)rOg!G8
zC;Yjjwn7L5^Rh=HeNw*Cz36J^3R2_5zY()T&`QO4fHPAiXbA=`No0mquxgnpYAdD#
z$2mbECwi7w!L~(*EtGa}I0!puYUC!IOJqN~>1ZKvC#N~<fQ@Oo0EGfc!y8CdltF1Y
z!6KcP{fFQkJb`LLa34I#0+Jbuv+N5ElAWoTS)5|UMkaj&jkXhd*T)Ypl$VgK9)ExB
zsBAk7ok<qVv=I2>a9eM0Z;{t7>?7Lwuqr6&;#I27kPlcG@8L!U8WSHV<YZ@upgzgN
zg+mDkw2ae*-R1#IpowlsN=h7sg@ro>1kQx)9}y@a%^i(t4Flz*Gmf9%2z(^31b~x_
zOdyw(02(#kLbBNsWGioIPz9NhP_L`2YlJZ5;Ta!(Jz`dQF0DD4Q>Lt}3}z~s-XmgS
zNv(!TN=ioHr;<`3eAK6C2`lN+7`5ZzhD-s~h0Hy{;`-^O?VmhCg2g{PH&`&$g#FO&
zJmR`lz85kx4t&Xou6_;E(2qRcQw;0Zs}G=+`79oUa{{hEUXp=ROf5vTjVAYE*-f9y
zFHR2)yIajr+tTVu^$OjAj0~A1<`Qsgt*yFdpHmVYI?#rh(RSS@n~&&|0V2(N5~$y{
zzncf_`21wT_2-I(OA?iRq)hkV5Y(pQb>3qm<*3wANuHIj3DP3=8{KutQqS;L-PaBj
zXEgtm0g_B#(1Xn&p?wGJbw{kbWD>jOu&J|usLcfLwX3LMD)A$`^O=_RrJI00Zx9Q$
z1stdd9!s+(3qVIc2T!vdvfKjF;g-e)cV-~1xA60p>PsX2jsd%E=jBz%N;Yv0gFc17
zdL(Mj0(80tT{#)hfHWt;T^~U6wgv28!wZl(9XNce*RAu($vFfD1Kh$#Yy^nIPMAs)
z(CPTJfEAJW*;dGbd+`BV3FGF+3`04<;HRKDl5|Sfrhv%O2O<LLO;Smu#svPo8;%J!
zVdJ%W&b`c}nQ-@Rxl{{Mfc-t($^kD2bm?PYY?@Z_L-YBPWx_!6lW2!Y%!3d=4vstn
z*{@)6swHV1^p!7pw!6u!Mv^ozP+tC?djP!H0i){R;}-nyU|!k4!9h}3Z&2yLw4)y-
z84Tp@Pp=)V+$L%dky@;AYc?)#2tt>X763KAx3+@(Swt=n61DiowEN5x(q>7rGQiHI
zKzeAbOLE#h?jTXlC8i<6T*KX#)YoqWWKccY4ps0D`QhiLM;HL7Vay<n1!$f7o71>F
zW_wxv*$zB!oYU3Ox%z)1Sj?*;Iq+*`uEgn-gO~N9Alu!?kLgL8cjCj*!?r*CjgNSl
zv=;xbIQPpwCfiA04D)~tuk|%%^x1!X&8c>-g{=M8*K=NfvW@!x>WAm;rSQf=U;(E$
z+WYW=<j{}c7E~d%`R&`cQ(S;pDRY3UR0NvWKU!4?zd%I4r~J7(H;OQ@KvNZzS*eCs
zNYHB8My1{~bual5Hk37g2!xQIKmp`2C!PYMvq`!gCcvS}EnajTY)mxyD)Zc?<)!H+
zxInKW(U`O{&22jW@-icSzWeZ@#<K~uMC(w8i4vGjyb+w$SGJO37Rz7PzJhCzWHQk)
zQ73il_Pgaa&(@nhKHEe^nbDk+HXA-Q2qPNyFnGMYfBuwb3<?oR_kqd1MBMe8j9Oj*
z7d}d5{NaffS|30)Nto#fY0?#}^jJMi39}V*HFNFrcfiZ`EAPY60mZ(O01PKPm%3b8
z=IG`joCMN*YiX(1=|VRMc__L3^CvEgzli4_S(Ycm#61?+c*av6k2Nl_&S!wcQb>$P
za9;lQaxbBe%##bm8r!G{u&z=0$7R2B4ybU9G=-l{TMFHA;J~>;7V)bc(wH&m0JgjZ
zwvbeXV47SQy0FaKklmbiA#HFDj-9LHJ@R&gjcX1bJP0ML0$$NkP&gP?Fd&1<(UBC2
zVAZ}lGkQ@;w<48Fb8BoWl@oXq(qN4q4t3QF`3i2Pbf7>4nQ8US8#fa24Gaw#IXMFW
zw!eJ)_C{ausps+%@b-SYfM~fO1C$#iBN(6ep^AE;KQL@7a6YB1oOZU@AAo&5Fq?O&
zz83--ZR#yptUA$IC%p$XLTRA}(;*!F;mHwdR5eUnG^U={utj&g5fH@-9o`$*s}ZoI
z=PCo(k;g<MB$l5beI?(r97psp#;}oJ`=u8t)L@8$v;lAW3Bs!XTXyo?ci7m3Qkk0}
z!r;ne`?GyJHTY99onisu!nzIif+@d_Aj6-e=8>|=C3yY%=*_j8y~){%YKhqbE1(y$
z$Bd0NN-V90o9ZG~)9ea_YTWsMFvwHl07@WU=|KMYV4cC53W8nbnYV@)n_OQIF4xdC
z7ZenHL>WBZo}D=rozksxAIvPNH<7GvcD5|W0N@=#IQmJ%?_2y*=!(oAp*=?sXc7^p
zBJDMA-<~1`OZ0eT6V5Jj*^5Nt=&O8QoFAPJi--t-ZR*u~cO2j62CI=2DdDVoPREGV
z1fc!%U1=z`yd&U&U_}iLjjM<H%cGlb2SsxeAosZe;u0KWv3m1WT%KyZq)As_kSPgu
z^S@H0pp+8x;jKy;$4182Ba)-*H*MSq?h~MtHx=)abj5mrhv#qNImF7;G||)qNJsS7
zW+{(?>c>ZH<G>92!EW^92@Z`SY)E2#DB~<xM=I`JyLQ1@xNFt*V<99PinAZw`-){R
zXqu+ppFj60g^MtvjiTblH=ka5LY_+c`CRr6*hQ2w81-npu;1_C3{sl~JMSAGA1_w*
z=FM?oy^*4MOP>~@i*Hue3Y%QKxD5xC1rD5$;a|V_A;|OakPNhQ;FBj$%FD`5fGobX
zu__*tK={y?mKH@ML{a;7Tg5ePKOnU=No4&+ODB#)%(4NyK$&{i*9zd`US!>UGUJ!R
zsZ$#u=hkM83=Z;&MVy)353^^;A&(!d{Ms?jKj@YMIFG8^1|%fxX<~rHVfyjuwMQr`
z1gr&_Bo#IpE#9s21Y(~xV72|@M~{5<(pYiZwIk}tvU&5?9VJvCwa*$+`QEy9izK?+
zZzH}E#jlrg<HS;;IJa!+LC}RyMM`<<^%kQZ#!dyH6H)j()4Z#xA)V<|+=wxs@jG+`
zJMBAY>9Oc8?gt0&Ku<4ZRLvC9jD)C?R_s3Spror9y6+w`>1~WN4Ukg--QC+s)(}FB
z(pAGhoM*c*0ZHc^^LrZ`oBJS$(4&O_SHk40(&xWbBLD3aMw8xR{dU4Lj=pdQ?18*i
zfVYEBK$s-xpU}~HqHqD+40|-1;ZU&`va>MeG>XCw40o`qj10J&=(&UVU=&IGjJ7Ur
z#bJ04LgDF568ljXm6lkH*bVnGA(>A&FeIt&@}GaDmcU<WXbMgps;S0wGEy+F6)M)?
z6KEC*ze!WGslR^|Ge#cd&zu1g$5fV?cYs{r7DmQ<uytm<dc{J(FZ|yz$#Vfwjxg4P
z_Od~0Jz~@2%YOYiTac@oTCjqY>r@KK03wXxjb-}E0NqC`!@%dsQO7fqm}&|^(h&eD
zqtS;k31ynE0y??S+DTt#FWhwMKRkjRb^L^{d=a#dAj6n!E<mA@D1MDGJ1tSuz`*#+
z>N`oXyI|eu(m&%WkdOE9@PNP4=&cIi&`79-%XNh`GUSUj!+;>?W)J+aY{{e3YU0Y0
zT|B0zO>FRbf-@3Q2j4VOF%>t-x=hYhUb_LQ(R`frq_jZ`SDV9-8D$QVl-k_LX2NYK
zHAig@EF68n#nQFgq3ON*@Ijz4@;|e?^w~{t;s$P&NIC)n`xv@MMb3t<wQPdYy0$=)
z;RPU~=m}CBiHX8zps_&!H>Fg8WH9ySSY90Qq_l#Y3ok%xLArOqd)G8y@YH?wiY=%Y
z17RY)0a4F@mJ@!*ufqWjiM&z<+$<||2_83I3v?yn<D@sundaJ_?pNPhS{^zdj33L^
zNYQ-ey#D_E`#bpflD~M6T>pb+G~XeP;42v!aUY>*6hH^xqo!SdxtjI8xCjp0boR?X
z8`9n1uh}yK4-|NkXRees4__uOqQJH<Fsa7Lq1T6=Rc#jHH;b7q5+22aWB?W+0X)AS
zb`8<Ti=nk7F^8*H_e{@XXTr0`{|8nP@Yg78aC~~Y*==fkTxIC?=00qQB4>M{YtDCz
z7rqG4#BOxg=HpwoY$1p1fNHo=-q5+<NJ~qLdk;{QHmif~!2U{smUO9@5t19(1BgpV
zkanGwD_3&w*|QpFk9oKZyj#;4S|+$`VkNNy|A|N>U@x1|e|LYU6D%m!3p3V)*PT0%
zTX;mw(_!U1%}LH<Jm`Z&4HIR&rQ$V|19M5K>F>6Kx#H$EA#wYf4-8k~5RG9ZEvatJ
znl*znXkRWvqKPX4qQEG{!$f|Bj%K<(mCQ<r)m1Rke1a{EbUR~ONE(^pjzkVfhVVMn
z)(TX&RUONb5^(pBScmKiVC%woU+ia^+V}5Q5)eeV4;^APHvYD}S5HkU(pZO%^pzet
zvYF6_#q6@u2T0f8ffXK1i7YzZ&&m>BK}}8K8(h5;Eax#)B4f@E-eLh{C)^zz%*3|#
z=EgM4Qpr^OaknJgCyQUV-%pqGY9G&{cQ)Kx75Ee-AiNGb`d}{lcG<Xx@k{a6|7Ldn
z>0iU*Wb?$2Bl`b}Z@A$2>n$9#0#n1S=da{J(z)zyNxf=q2^Mawqm-JEZhoOlPs0;1
zuM_GeC)=W;wpdwr(&);4mB=Bs3nMT?VD9hAJ$vTt?J3F0=3y-x+1Px68%X*D_Jdr{
zWl4wO@3)cjx*jM?UW<z2LU&8Yz!0R8W@u;_9Y`Np2r7Y;zBMtJ6xO1qLaw4;QZ#8p
z9KSR7%Q@UbeD7Z-ZMD++Od*dm{+<`$21;1Sr58kk*?C~S&R9@W*Jj%5e&O~IMiT<>
z(5VN90AvFI`yib{Lo`Gl?{A0?qWhMDPysMYJ=<D&?KS|z@$rY0y}#?~>XdEupi$B-
zQOzEmw)~nYt6Dj81S6}nt>0EKZT0M~V_svx8vo~jLHzLhRYfwT2}||HPbEw{O0I#r
zE5k@>IwnllZWfZEF;M=S_tcr({D*n{`2N2#uVdez&y9(6tH42{&dE`yCU6%f=)^^*
zdeSjHYV{wE>8*%v^)BbiqnaE+)WBFFTz_>X<xVV0S}P8DQb27}stz1D5MBT_H1Vl4
zc$_=#u^Ibh{wI#wW2d^C_ck;%+=C0j-``(l$ntYan3#u+x+;bwz#-E`WWgvyE))hQ
z%e{-hRISaiY*pzcC%poL;N-xLP0k_}gL}M=9jy%HJm2|>X$D=LNYwPNmiRypQ8Trk
zBbC4QCTHrz$9OP(e_W09=M7zYW`3S*yOEltQA1n`7obDz3_LFni%UWvN&1k^F(5Ly
z#nsa~sg0tn&u{MTf*#^MU}|a#%TLqG?sxO5n->Lf5Qf-6g5V1n_9}?LjY9{PINigS
zg}re3KWaph2v8kO=nIS>`=5t|2eIPOqn>5uyZ-~hdF}c?1ZQ)u>$Ct0*n0Hw$p7Zv
zJJ95r@e4y^2k#dIj;i#-hZk=(q@tL%Y)J_$d-G-k7Z(HRMk3KZObEvAWhm$dAXAC)
z<^X`tT}5+)eCP{ER^aeqO&V#~GH!q+gfang8~y+`Oa+jU0s-9@o-&dMG~=%te^v1X
z;~eS6wd_4Vesp1^R&(8a&p%!O2Fcl;O(E#Vq>y$6T2a@ideWBF)&G;lc*Ew+iQ{uq
zzv#J_C;Po8>A|~E?v8?$fiv<b7@Urtp5Kk;lvOKNmcSZ)?b<a=G2|<FVhW6IN$l|9
zVCTZx-A9fdP41>=6(HVv79Br+T;$k+Zf<RPUUMD)=KWi+{v?eJO?tDiaR-RUkoutK
zkq^Fq|IRa%4;L4S+CAtuZrn&mLq+c)2arg|Z!V4W)?4;h0}A|U=Fo!lFxtMnG`%^e
zzb+9Fcm^C7Wt0R?f0>IOhBLCk6_58&n_-9FgCvv)(mDz)gmllLjUv|+$neY8uNT3b
zGbc%*=Ah-JzFREt815u^N5{U~){#9seuun1RS&H+*?R&1lr%On0lVLTTZy!t!B9nD
z^gx7c(kaV8A)Q@aE4Zbcz~9s23!|LO(jI5~0}i$AaXcT{7_ZEQm|qDC&FI+JIjV<1
zuw76oD7~QE5(6=(!T{&*0<gvY2`8Y;^-uu8<-)F+^KJAZOi7pPW7c>rcr`#LCg(Mv
zbcru7OpwG>axx#l#!A3&x4D6M*%usITX8&YVax|7+8fp;(n3bwKMD)@*Yy~70x(t2
zT|(EJtl<oxN=lb+T3b1BljYtt^lEuvIYUj?@|o+9<L;B}7Pw6`AZJsxMalUaP|qs8
zbl$Ga1GmO(>$6Oh!Ps>|5pg1?u;*aef<KHNL(G^WI4@BJLQ@vhT?S+%V)oh_@2AFO
z)AG0zmaTtq-)(;rZRUk-iarpw$Vp{#6rK)!3Jx^c++%}hFulUa%uIP!d&D0s9X&$l
zBv@vBE2|*%zvO;F4zGeY-x@;K@z*=4_e|IqBc5*Dx)lv*f=0FiTHLw#Gn~M@9WZeX
zQreUyXIDbLCy3LS2DGh(cwbgiLodCwz6d3pxLa&E3LJ6(W2#QBt|#IfN?}ze`9173
zI)4{Y+az%d=Xy|yq+yVpv%rH20G`N)3#8*-YuD!fcjTdNob?JXcu@>sn!#A}Dv%Us
zQPQAo(}EOj2&P}Zz7ptR6o%U}jQx=I%LflOAvo_qo$1EJlfZ=Noe5AY1~6feLsP&{
z;OPC4G|8zn^zCGV6@-nAuDKmPU2=2><2>GYU;{$%BiK!lxoqU@Iws*);K%`-1UpAm
zRI@!qExibu2vh|h;2$Ya!BI?F`GBg_dg+XhsmE|-8P^_%K6BCBoF3S^G3^2u;)G4Y
zMTE2@!XZe;2+T8&1TU^QXw~MA>DOr4J1lS%%E20?9o;Z9ot3*e1(B`hAT}HQY5^E@
zj{el?(>D-&m=5Ud>$?TnL2_u<xg#Xn!tIyftVY_adK@ARDZ;M+=5AqKmj#+<J%6@3
zx+zUBmy9*n2A<x&CgV^>qWTS<1YJg9hX7s*pPGf)gKs<gbCMk7eG3gY@s(fM(B!Lm
zn>o|g!&BzNevO;oMQ%7raRrHQJ=Etm+*eeW7iD{R#O)+GvyTa&jkg?&^1fwk7)aE~
z6T(@0^X1DU3<;)QEI+<_+l`MAzX>LMlX(||csLfsMvYBMMPIJXq|?yQSpRsx83yym
zt<P4E$>Yfsy%7AQ4fa#ym)2~%v0*u42jR2fxbHr)W3(4q05btI)!kag+d1c$d?elH
zo~EVMA&N4OZrLWOMZ%-kuc`HojVr-}W**_e6?jbUu6(-T`09HVek}aez+hD5V-+GJ
zBfoCB=fn|kvJ5M9J1{U#C(q_9XW)jY6o{~t2(gP$rI`eD4nN+$5ue;<RT%YI+zGBj
z{>OLAX%~X-KYm<&@4Ebo|NPGzz5aUTpO^RO-~V&}fB!lUt+0~@v$;sPBPILq3z9E8
zqOJa-g-?RK-GAQcZRrlGJ^#Gv>TNRXr#{>yOpFf6{_Cx`$^1CQApLJY+X%}4?|ife
ze-`Vn>#-em*vB5Q3Tp%!BCB8tTt@biO00sB^Wt&em;buuayb)FSY|?rz|QdF#q5dR
zLx+ivJZ(#)pRXGRT2e0*>jfOxE`<rNubi11y0&j$^B*6XD*~EJ1EP;W^IU@27?t2z
z5T)?@tR+ChJqH;uB6@P$jhVA-mmG<ImPW>yoE}2EuUF_&0l)*bYC8`9st!aEDG?H2
z#PCFp`xe%fm6Zni`d%laoz1_3wh+dIx?N5G(2xn`@^N3~8-JZla*fB%@v5h=ZQ6C3
z;NG{-A2x@?`N|-!I*_DZ<$1sE*BxQqyqTN|#$ED)I{`%n@pm0O!%qg!b|IrSfKUAS
z>lZACv6yxyN6$#C69T#Py4Yz)Opjh^D+t@TQvtr>)2B|Ix*7PcvXXSC$Id?pKIJws
z0a+1KQ~bbD6-W&5d?2HQc9HFkHR#}P-`LSJk&_sBLQDP{1>Fx}&-c84d?*OU^ozLS
zv*SMCh<C&NaiULp^}Wd$TJq47`{*0=8GW{9mhn0e-%c(r^#C{oKK5F#>({CAXqIU0
zCFlT1PZv1Zb)L(t5n*443fl*1;fbVsI9_mqU&^NE6rev1N!6lLg5q@@Y&srHQGqqz
zgH<E2AcONLp>Lq^w42uR&GzXQdqjZF8e5pP1<eHSCpqRKVp@L#Z6h7azSYuKuL=Tp
zkOVs+AtCW0QbG}j$P^eaDJpvC$UMm!*{%Yy4Sv}cAj(BOYQ;90Rr3Ej>|FfR)YO<-
z@;e#5O;VDiEe>Fo2XFT2vIy?e0L@2e)M4xA;cvr*jOc*3yTDXpYiPVe_%t%%RpLIN
zbr7u<p%B7b0lJx-q7XKIPi)+D9$H?5-_Cb1;06K+z!QL_VCUd?yq1$Z4=^QrhD#y+
zD^5<3o^cq!QDES9+k&A#?gP#_S38eqe}{{yvF8GY+E&xkccWn9$v3JJ!-ocJKH$+g
zWiaQj@Va-85d&QXZnO7cLV^+=@ywR{##<M5*>k}^*Z*_WZ_Mn61{MSs^z$nuymYr&
zn^X(Pe6U)s#wo|7@v~i@y1Kd`%E8WfANG(8=+YqL%5j?nM2%>-*>Pe4F%`_Gv7Y4R
zGd-6*2#1k2B~&N^aIg#TwWm5{C^aMPLh)8SM`JB#AD|OLN!hQCPc2~s8ho=%TUg|v
z>a%%4uw&=q;v(=8Jg10_Ks&1%MEw-OPfjv;!2$IN{_%*qujpqsi(KA7pe9}dsxe!{
z&7$xUXEQI!TOt^q8L}K-a1>gp`FFl6m$O38fhz1XKRrTyTRyp`>?{=ka-%a96^Z<^
zrA*g@rujg1dt$69_54Gilq~qf)KfH%9Y4MTjohH?^XJc10^-pGY*tt@Vul=?8X5Sm
zPuvY!)7>wx1^3Q^HhvbeKuua5HhIb0NtnF5H46>6Z|@-M@k<Q<N}V(PTIbI<z?Tf{
zxvJz)yIb^&)lYhA%XPI+=snnI42PA7#kXK^5fR;ZOi9D61o`wNW55!z;{D2z;@ZU^
zd=t_htXkSBwREUX$;`fBHq!q|)^Ew}*>z0E3%G>-6`xzSIWy$;Omy$Wkek{C*o33^
zC?rWKM@iCbWF_PkmdNfCPkjR9p8><(#;#}<71c`G&YJN?@)=Uj%EdeKBgspf4j<cA
zKEL~4$u;W63K}{*T!y{`Yt^J@E{|g&gD!Dyo}`VMG1`iDLj1A_yMSIng5~m~xTYM)
z2(F!lPGjDHjY;!NvLsmD+lGjQ%hi8w4$Y*t(SsifxChUFRkg8qWcx#aA6mOr{1M0s
zZtHF9G2eh@m`%z<-0O37p}cL2&y5>`$RxjiZ{&)HIFTi?1_K8ta1bMg>(H+V>3WVI
zS%8`QG-l7?H23E?6j)MHqShW=bHx19b04w&9cbRZ6gUgxeRPW4!Z3k$Nz8Ivx#z!?
zlPG`8k)6#pqi`R%W`w~rc+T_btU$k6J37*S)DFu*N3WJ}_8J^jDk+)=;00Jk4tnm@
z%@+Z_eS3fA4AvBJQNQ@Lz}Yr#hzoLuSc)UgeoxIqlII7R+7D}lwt%Fho_lJ|5k5*1
z6FJ37&Rvo-5oTrzu?isxy02e}iHMPf(oy+eU_f#+^mb^P!N`o~G7MyYPfn8912y$@
zq<9!+pupnMl)y?4Vc;oJ73$Q)c(3A32Cfgdb+R5KRabrdco(Jk{rmTsvu}ms)Dl&;
zh}rofuJ7|{2mm7|Xryf|Dk>rwY5r;a_GGp|-&%dJ<)<GBUgMGV)4y~{20x+oTkE+n
zb}Rqmsqs0V4I_~!XCLj__b>3V;dsBy{}#&mcPIgP^XLzf`xmrX+eWEi5BQ-=;r};;
z6BxHgA+%}de?dWQS_~yA?^XT{acvWl`+xJZ{R6wock_BKEnF4}XOiOb1K6$y`-*wI
z10o_j5I68tWFgfs8G&~Ffa*pf1MDvw8ynpJCtz^Vgc$1@Ktm2rBJYt6SVqPHcydiJ
z-HvW6eh&vA1<0vHwZx6ERe)+cn{Q`Yw7lestXCf}gapb4zM;JpwCPk-(4=9<z(AN;
zOT=Mo1BY9y`j;Z~P-M{osl2>=&OaakG0lLk&=~#{t&MfSY>Al5grfu3IG3(}>|2gi
z5E$>I3@~XSpxiLP5eN|y(8UCOWa8)4nrvX}5~TG+-PZ?UbwW+J13*@fF&KdhMNf8C
z%^m2+tt9=L#<SGEa<UT`HhYSem5E=Q36=vqKq3fJG=SFN2jlnPsXM0;Hha%KUkUaJ
z+H-x8yDKcyb5Vw{K*yo`V&YR@i{C>kOVleAL$%q>%R4Al^}HtgnfX4hEwX=|l`(lV
zJE~sT0dKh0_qe#YBv#Vyf$U`V)z^1*tU-mJ_I$hB2ry2Y;y%CUdq{&kNy70D!4xa*
zI(3>K-B&X3jZ|$uWZ&SYPoJuF`(d<@my#4i@xbg@nB<2baO8?A5*Gmf-VT4XTRH<$
z?l~0BI4uu^%JSYwP?2$3IS4B(2ut=A<j{Jwq#`xNUyJPW4gz$9kV>l7XP_H&A+yR=
za~VhzhDm)T&CQ2ipHRiP)^n^UU+qXDcq#>0Ewv@VT<t~4Ofssit!4J~Ts~c815hEd
z9I=8)POfU1>5f42%@-MNzLNf)t$%j~Bdta5^TFUJ_o9|^sjU!%CP^y4glzPuJ$P=4
z8c644Xo-99K&HVd%pme%MPNofYnX<o&f)U9JCV1(V!qA*&n1w2FX24i6H$d}>v%Lw
zC#jggAR7UyR(F=&3LzgWbRdx9(3w<Ecmf8epdj%=%Bcq75sbW<)KWbA<FT3{#}S*D
z6??FZ^F15A9EZQ3BW+(-T3#vfg^h39LD(;zdGL+?DJ+J%N3M2wWnTWoQvtS<iV4`I
z7a~ZQUqyTS1wb~sr#-cgHMC&)C$U*9p2Ik30(7Orzl^Gv)Pk|U1oC-A<V4Z^-O5{O
z&cX%|GCF!y_MBbu!KImQhU{VJoqJ%#JckvIMN539RWC^`@qN@x&|BcIrYRhV(b63R
zW^O{T+eu?<VPV7cQ&6VefJ*i{q8bs?G>r#ND7H#Ejq3LT%2osKH(5Z9t+f+}%4#No
z?)5Brb$A67_Npb^&3A|UCCpE@+5AMarlic*<+Rs-Wl)~;6_u57kI9FrMW|pdH&*)E
z3O@)^DhP1$-l~h2my#(k@D-^JVYt*U6QfzNz;1pp>li@Ff$38nVd7yfjx%Q_foCB!
ze0wPVH4BeBlRd`-+QQ>p4&-zBWVi9iYgUzy8S^V`V1%$g4@YlfCaR}jY;5d`Js=b)
zIT)`}=J%W(H?p4xTh{ms^Wvu&n~^P;-QC^MY^a5S_}cxDIBf|j-gFl+@;xsabzF1r
z`By3AxECs)GE%~;Oow0EJD~w4U&*uSWjgiqS^OS9azA4MF#T`}Rh~*y8R3Bk3TkSN
z=es}k9#twzMRiheb1TsH0O%bo!cz>42fw2Z9ZNODWc1(^sF<C|u!BX=b;+X&Qpl`J
z0EPwrDDm5w1{HLFc6ZfUwx_}Sv{d$LWyaXd{Wi_Mg7^vlxiT12Sf@FMs;b|;+grP`
z19udHW|9v=wkjrzf{{M?NbMNQ8cVSx5hy3TX36A`N(JV)@bVzoHJ+2y+Yvlg@4O)f
zJM-r_?J+fEeqsEeAZ8k6&uuJz9n+SKU2foQkCES$?WSG$s9^$+SIZs-wz`~sr$@pI
z$l+0<Ch~w?K<^17ufJ<(%*CF!_sDsB!ra^(onamxp@fWCQd4sWAod*42Z0*iIE}%f
z+h~_JgOt=Oa$DmCdIx|T60<;S-m0O-#4?B~grJa6{Wn%FuC$^va|OTxLN32A<B@~N
zHP3LGnD(6Yf{JX5hZqu)%L}%2tO6+$V?*pUu<bQkpv@XIM`Xi;`3|OXc$U^9&tQ%^
z5~%Aw4AYoecFwoo4O>7Ad|TyJRl<1e8`62&gta>z`ck0fE)bx=iDat;C<&A0N)Ue?
z5Ua>>ceB};iG`kbn=ph{jyjl;p3W^M7XI<$Sv(YgG~A+{Qk=H6pFM+zuROI(KKpEi
zi_5Z<k|(AQ8k^s>gb1Jwu0Xg^35tqRa&QhR6FyzW_hhb$m`=fT2>ct#y;zHAW<|7G
z_C*Ih@H{CsL;u^ix4|Os1^0-7KDQ~I7cyh#;Glwz&KC6js7t{}ft$8$d5>P21dhPy
zt6q*Ug5x)-kC7MmTzWDy;{<~gc|VAzZGwV#!Rf}$hJi8fMYotZ>^eCZGM9hp-F*TI
zdnAx8fuWC7N)gT!Zl3nkliiv4cw(HzP{-g}UMUSuB3U=)bCF$u?884vK`nE&@83Cr
z0z)SGXaz3-K2lx;c&LD@jabdV(5!HChv%MfF?-%*_0%q;E3(PP0V6sv2q@_ri>H<l
zAOKJ+vp3S?c>xA!H3&e5Tgb#K&c>bIjI1P!PT`2>k~ms)vgCun#Zwu{M+kuSt!+H=
zQ8d(Zc~OJ^f!%xg`%+G{G&EFqff%!wb^uSq+o8QEF|bYQ9f;vIYuA>)dv_ZSoacKK
z;sW{O;X7l8&4E@Qqmr0ew!sw9q$}wW$7nn@e~N>ZwF(&JW^u7pU6d3FC(PcsppIaL
z#aru|dXLs3*7EuHYDMhW>oBuC^_&mKQIrMg$9^8|edCRj=>6Y?@Uh~V2WFX)+buMj
z&9~>_EM_1rE-g$blX?$&yN`GP-3b2*VBM40Vn6|%0;hY(q>lleiQDo?<b&Xh9<;kc
zN&l~1`8BdDk1zmRX$Vf8OhJn|PeJ37GZp#mBX-X|U2dET5kEE!@17_N+2o3;X7t-w
z^t0F8QuqhK`>KUKdUO?@J3MvmYrcK*T*v>x+INRz-S>Z=+IyEMX;BnHcBqt$WXo1Y
zRw%RVb_iu;h3pk7tIWzME7>7rZz6j>uTNdqeLuhFpXWH9<N4=0?)&PB^Zb54<NbcE
zcSn1B_;YdNuCK6`{Os<Ip3KHA;+HuBsEPkG5tI;x%j6DK*zuIi*5l3tTwHrj3f}Uk
z-_`Q5MS3bquOG*nmM>*+lCl3imFgvi(;N13DJf+r-047)S>+`sv$^eJIS8K#f(GeT
z1&i*dLr@267tkO!$cI<%$Hi+@kqNf{WK+98yG|@hQ3}L$GIl#~50MiuxLs^S!0N^g
z8M_<6%zS}e0pi-YX)o^wG<dcAUbc;T53uZh!_vI$%GIkMaYH&UwOF?e)f;-@X!#9~
zH2A#DmtOGRfrJ2?|0PJ@0O8krk%9M|Xzd3EN>L;oRZBlkNi(XIYbpi&ASbm*jiU*d
zY(hef2Q8OGbId?a21-ABY+oX~1{=Z-2nJM(2TJ3oTqdH_&<aEW=0!l(YG<5Q@%7h~
z3cTw6l!VNEYRCRI3u3+i>2WXXp~57%8m=_J!l1$Kt1>b(GuKW*A?&A`=f|ul;SgMt
zbua?ZKfxk!OZE2ku_Gh~FDPbs^O`jvC03zcm49!p2epX^7A>kHVQ4GhwqGSIER4$c
zmC(CubxqphLVE|Lis|d*_bj7Bga@7iq=2hlVlRWgQXJBlXfIq(;zD+uCvnrNJqnQu
zDc7X#-n-X;v@#$Xmz(<WNNPATQ(N&smuIH6W;n#m&dvfD@<v}yIe|cxG-`D@={ht5
z$Hlcj7ZiL0(&V_bV6*?GSe~cnW(El#%A#7+gAA=beK`5sP^H1hdLLz_{`gb#<LIgT
z>ys<+7WwR_Phit4F2z2pxrQYQy{9z0ZewMoY(mbwpPiXG$bu_if3j^ST+u}XL|+PR
zdcsOxZbCu=|Miygaw+3AOHM9bCVxuM*nb<M`(ld7b%EA`0U?$>Q;hVDYg`1!y>u<V
z7y*I-o^TD<It6bP%hbtW;`9pWN^lp@id}rE@Je~B*xqXmh_%4|RYa1}CMT61CG{~!
zyun|HOZM)LkQlN-i=9Zq7E`Y`c%hW1auF#e2SDO1HhmH9E7IK}#Zx5OqeGqp0ZB-@
z2xGit`E2J9&!JYP)414h{anEdAVwTqsV{<q0|5W+WL6O4+Nbg0(W7dpQ*a%uI*pQb
zbZjgf#fwouS$gvm%{|5M-@PNt2&fkB4SmFhAWQZxbm6W5Ll%2F7QPpL{#=bOKx+Np
zb}8XQjPK!L!G6V|$N^{rSVlSIRVPlIxB^>sfGKAtP_dmb{2l-`1e6oJekm@<G_TtM
z?hJ?iKHA8<cYm*k(8tgc5*iYY5hb?V$TnN&I(0FeYzp4wraTB#8s!qOs!*GK3E@Y_
z2OH$2nDJDfK`Ry$s30cvYN&kQo?`07TNWN^Hd`gDI;WrdpO#`po8bVFsNuv!0H;3e
zs@;UG#wNgWv$waGJM7AB`eV)EO(`|i)glp<FDngtLC2^@qpZM%Z?!4?1^l%>fIZD2
zP)-~wZZeZIi-i}P1F|+;oyEZdJkZ(U3yLnVwx&ICj1dQ0oDc92<plCGUtqCpMSsC6
z3W4#{K4qd9NAA}4y+B=hK;<!Q+Y3ucfA#|!ldMejJ+9$^e3Mu?a2$W}W1~mM7l~pC
zgis1uT(io;3Rw6nR<3-5V`7Kf^~gF$V&`RCLZ;st8JR8k!=tX@teN9UFrwZ>yJO{e
z5r;1G<)kw7wY3~Te^zqX<W<+AjRUi}8X^y9m{Pz&GD_*iu-yAs3lQjYu<dz(GKcN|
zJpb(v!o+hP^(B~$N{KBDp`Y1iI{S*Zv-}@On$zX=G6cQ?7D^?9>L2G@NcW9bor3%=
z7cViESFBiS6@du`zOpqCV{z07T=Y}+N`8(aRCZzlYmHD;Pv051Bqp*@;2KXgs&&O#
z10Lf8q{1}ND5~kq))2Z9`p?|?F-f@8<G6mXrA-d>^u#b$;1rNXR$T4XXCs*l{j<Tg
zc6OYwrXVjnK20#LoC+}fIR`o*3+b(qIu|8zl@i5^C$ehW5W;Yv_V&z4T6g|O4R}Q;
z4P>VnScIqySo*_C(_~TNkD;1F9YBiZx?D716INrj*DTiIqLO%i9Ke=T>MnkuT6MGe
z=j7(397l!#4FdGIAT8|)7&CE%A)zT5GI#(UAlUqI#2XDs4hh7AIfMX<vKO+nV<C~;
z5tRn1O6R1Fvc%Vajf59f5Rya6dD+()89gX_NU(L>^IEQ58c7>9bY^6Vw_N$lSqW{0
zVqK2j8ajkeShKH@UkWKOOEnQhrU49WRO^DAO(97=0Gb3O8MuUm?8A4w|L`G7o#&Iz
zBu55F#OiH!hplmI)@q)?M~^|-OANft!jAK|T&dYEz0oe@wV!_7=q3tDi&QwISs~b4
z%TB=ck6-Mi_XhD_>mh6a+);wlIS!E|aK6ohm^Vr;1&ALzGSHfr$M30tvZu1TS~6`)
z6PP1Rrx&@}IuL6s^lAxZHI(83xCUcxU!;#9htR3pW7r&*pfV@idTp&QHhu~W2>Wlg
zq(+W>FDeps7(s;#<&=n)Rw#%hLk5cX5&5^StgUglT>%4JL1lJYTRRGWhUfeQBpXF0
zQfw=v9JNziB&lCmc>hiRn^oEjaVk*s!F|1In_WIXA75KXhj+L-=sYIn*ZaYhTm!Lv
z)b%V*TLL3uiT2N511Hfh5tFI82YJc;;p;#RgIV8-7L8aIz;JIktQk}6eV*?;%2$8X
zjud>L{RrYCF9V9GSeG<x(S)4uUKy>Cx`qfbLZeYH;rQtOeO*J(4=@MC{=NjKaR?qs
zb*${6kW;tACq~Z6ZNqQyhob8(lwu&uXps4XZn*;xm64hObXb6YA>|E>L{LCGP9?)4
zWJaAqoPABl5OVs_3P?XY1*`H))ExW-d%BA5TmCrx@&CY2x6lIMlHa*fY$67=ehKz^
z%@`68Wu|CoX_@g}N$*<)^=T9e*8W@+PR!)C$Kr}J#U&hEr=P5rajBUH{kdfa9{cq`
zcE+W?`sC2?qpsVq2?41cX?_djNoGKLUk|=r%E?_|fKVeG18^4i+``gQ0l{=T@$Y~l
zZm8XO6zqdx(L02nrTxGos}>CLY(Uky(_PfNn-BHobJa;KT8U=|&(}Dh7mxrqxC6BT
zaRl~TL!cjpiV?kb3AL8}v+kK;e3DEvkT`}#{sAN7P#gg|(2Lgtu^6?N&-~l~r`PAc
zk&)10KD#EX(e8;rc=%JnFx7IP{w9GOsu@Z3qRPtta`-)#r2veEP0D{p)IwN((z6NL
z@c}!M$AZJ_<uDsK1Z@+O#2vVeiS|9CsiA=hoEyPZeEW06WOp#YB+xunNUfF4^%Zfh
zk#@m5k1NZ7X7o@cxe^k<>Ev0^%toL;5@7*;8aIp=E5YM|#iAeU;LD=M{ikf#WqiK*
z*#@dze^|8k?N^PAN|0?!)E9VRsILeJjzf%w`}%GiE{cz$ml2x>qH9WgnC|7~3bASB
z>XPFSWE84eIV7`e*tpTf%}vtcyTY}j{F{t9+1Uv-g6KRy67Ub%HNbo<1Y##f3|25Y
z@95c<jKrjPV-&>3_1%9#;HNxhnm{Lyf;E7!<SG=@{gLwladYx@2%-UaaRGHcG4N?=
z(E~Mr1gil5l7?OTzG~f%A1|P5Ayhd|;8K7Bq}Ik3jYgStG4SMC{P#P~AHa2}vP<V$
zZZ*yw3*0A=U!BH_0e)~0KnqF2b8>P@M4EU>|71d?0aDHmL4I6-a8tBoXimV7uLb0!
zD!A1uQ2-z}0yWr88ll;Vy#q%<ofY~`C1RQk?w|Ju%Sy1by>xV_xj@x8qhUo8RD^mz
zEw=|t5>T564h?orPOzgdQ1-R=^n8YdmZ*?%{v#+oPO$%N2_lI<<0+F?mGrK_&%vTR
zShsUGyY>nskh~OocF<@Bbs7Xq;=qDVBRG&i@c2HcvWR**^1bRz*1M_xZEma+JwA46
zqQy)X4*o+c(4+!R4CFQ`fw1E|&c75$MJzd>TfA20{}GDiw1;l%M%fWx)``LyDiDS&
zk(tQ<XetVh=mR)HxfJKNZ_Vf%l91Na6ol$QVociJ64(g*gBLRe)e5r?xH<nrZU`|w
z+C-;~-G-;F7`#REv1cI!+Bn@i;X&21esu8ur5)$PK=6UyiOHJwy2Ycr=Jh|I5&02(
z{{JrAKdblP>i-LqqPum=KbPMXT6e}Qku7)2F4D1?T()1sku`WHRcrLXml8dJ7L{ec
zX56Sp;*I|P$M?64JUi9u;^~{JabVkj?$19D&HRMBm+!(_?`OyVK{Nk+jsP9QtIzbT
z{x=35|N8%z+=M>vtEnjhP)}v$Z#k>)w`|$#x2cCgW@U+NJ*$tk9$iPRrE9bMY;R`{
z17#1tfR3JFK)+|A`$jk#v`;TBs)K5AcyRd7$A=yTHc!$(ywxjtjSF&ds>#fU?Xm}c
z1yCHJto*rHja*BYB#lg&Y-RSbriRhNp3vWJk5*&<GL;C~Pd~+FAd4*#WIuCM@$F$$
zp@&h^=Z6A-0}+2jt1eN}Q9uBJy`C2LR{bfN#&Dx(55M5UE`M})a<VfcB?9OUL@JMQ
zX7A`i{UWK)!^6zW`;^U0v&4s03jgN4`HXpyw8So3<h4ayaBu$|$YlugEyb5vso;$g
zOZV3R5d{T>^a9`Q_kZM+g5a2bCAx9&x(SoqyA$Mg1<w=reDnJC`9zKg6{X4bSmm%~
zknnW@tfN^hgWt6}#>R~Nw%qL7Olr{^gg5LJgDgVnJ@Bolfc0XFzaepmo>2je<HM}7
z?U<w_B3?t=cBfQzN&59Rq#PkncmCZ?JYDx~g>KcI3CX`E$4Z|2n=d;ri~kjF=*{SJ
zTy`0rQ><##x<gL$?B&8+dd-tBm7>WoM3Tz+b<{<ST0%w&$zRK8IID6t2lH7x3=DO6
z=o`fVfoRz;{qh&=%khh8`33Z+1xN0R0^Qb7n4d3PpV*4`Y=Hy;GV;VdG4W5aIggNX
z=)Q4QanDasBWCXw^;IGigMOznkjsJ*phOf7?NoyXFAV`3+%;0@{z$k7E~~*GX(SC-
z-zJtmIO2lGw{#OVz{CcvVavgSv^`B`LycV&CABB<Q4TKf>TeCCMpR8VI$yYMC+6N<
zMX`n*i@`~z%!)}bhI^P?&t@cPG$2_LjrmJhx^Z$5Q4CIL8eBAooMy#YmY{#3<uNJ%
z%SNg|l3s(bQ_v&sz<$)5()G|x7|jTyGy^Ayd+BF?zpKXLy!w0l0IWKkgC!wITt)d&
z7y9KSBp;9YFI@t4h@PGVTml5t@y`;#sn#KPnv$rYJ`L+gtJ+1rLbzBA8p2mX6ak&W
zjz{<cIX0W2GX;p|&#AvDXKq}!v^CEuRCMDW1cV8qWfLJaz3mXKnuGmO&1~LBo_jx5
zRdF;|jQ{=}0R#MYIy$L@<UFF2QwM%D)RZp!p(O;h5JCO-vv18oYjtu!b>=fY6pJ?&
zM)Kgfx|CK%loN~%+1WWz_B>09=@aHNpFV*C5fQ6ZNg4t~3WJZ@+u2>z$Yo3{!6|~L
zreUFNHCE?4(<<aJ4dcmcHsz;EENZ~&?~a6ttWeJubJ@ngYdg3bWUJ9_jp16u8GJw*
zL8BdLAl{{$NFp7fA<v8+hd_kmy;YE;M!;6{+|$)nyRjyY2*-e=f;KI>`(S-vWyEJS
z;enE_Qt;)TNXsD-<xP7}#a7R}Nqzk-&fd1gZrLv|d4`KR(@kSt&=<->BsI^b{@;Zs
z3t}hC;MqYarEiYaFu{Y@H`x8q;p1HpaW}@lY=w439Kz{QjLON%Ux-Ec0;wm$Bf@_P
zT0RDJy^LTEk0yuKu32+sDn!udsO&Q5JFQL?Al!(rM$mRrVSo?Gi{(LF9muBBX;rgj
zGHJ*T0BH29K|l13!{<%-om96b2Wcj9=e+>(Qs-625J>}Ndy%I;$UFP$1Zsfr@&mUQ
z=lRm;=)f#>Lkg+zO2GvU+};wu&w!U(MIP1g@4wmG9vkWyjsxPAN*1@|S1r+7Wg+Eg
zxoRz}nf5z_DcWK=#|>#VYQRA3qSs3ws6Z^tw6vG3eX-}{Pt>pTBU7!0^+~~*&8@9u
zfoU5W7MH7K0OduM*^q0p$GAQ*0%elX3}4>-C;Wu~P8(MAn$lC8#tx$^@gL~`$xfY}
zHMW|=O}m@jo|j078pgWJ+#M1;H>0)H>3*d`Vs-nTlU)V4sKJW@6prhWy#t@k;wdOF
zw1MQ!<2H|(pC`)-UmEON+sKF06RNw+P@4)hrkO}~F)Jd?CfRlv_pz}bE2-OPIQ6T5
zx}Bb~voeAO;DT?7FFQAP_q5A)BRbbVk&yQ3*RQsPpC^o_gEVAi-!p93pbbFg&|MUL
z*N#%{rl!6C-S=Y`4+LupKy7kBZM!pl&0c4R*+~fb0wMO=C+zqtZ3N3I)vVs5z0ma;
zxfOg_Bu6xJ-q}vXt1`x^b!ES|kWWO7Go!4c+Rkp5o;d`TsBMmF`BAOhQ!``TQCM>1
zCE&O*&f$Uj45uhUF2Cac_u=SfaHT$y7fk1oW)$^>V``{+HU1b{I`4xXi>(fR{kczy
z{#Hmrf~fM^peunqjJ0-kS*Dg_agAX$N6n!#22F4E)=!;1dxKs`-C_5NZ`IX09nn%N
zTHb{T$%vjmk7&O0fcyFui{4vXS{5`k1fX-yvpRaC>RVacAyET|P>2tnOpjgwJ+6XA
zC{sgr2c+@4M~hx~LS8_;_|C_YL*0j0t;VrUmrKtpb8KccneI{W_Ht78LV<2r)`0^Q
zg+wKmLLqPk&?=RcA^7;so7>4Z9R@*vL3%|r)2hY7&H!d))7ANqTyguCtpQ1$BP$KX
zXSKv4)do85CqJ`nS_9|TuUpp+=ohr)dGfoU@rDwj2$d-eE@BxW*bNB7Edk>ka;@S$
z;0mCQ4<RZyQP-$W9MM?!-i{h)2jBo-ZOmF=f$jccFgHtW*z`dY+@UoITA}DNjG9+-
z(QXX>@q_X%a8NU<G>4xBC-foUJ=n1ZwScxTcW@mg;c{Y?tu)SdUh8>Y2dlxh8B`!t
z)NCAx@&NvFX~cnAD0ama;HrO7YeTD1GMXz<HT^Izum9EI;p4Ltm-%h_J^@0xVfyW~
zRHT);1rWTST|>b}3lM0Yla}6`QNOVdoj}dpm?U9{u(wpk2a?}GP?Ym@3%_;m2ehj-
zQ?xWRR4z;YnW?5yT@t>^Db_5<mgcuREMLBA;n#AIKPBUD2DZ8VDN(1#A>{xo{17%J
zwEvQ6WsoZw4z-Vk9GaURYR%Y;&jFP5_jsQ?6l-#)QZ?S2KY&Q24chts<r>Xwc`Jw>
zrq8iEr~Y@OFOxiDVJ_Mn3hYLPR!tS&M^Y^6S=a4i_6fIFusrD;cq&0XYF$uv_HG!U
zO5P^-WZ1U4q@zh){xi5J1?Xz+G)Q!4P`}$f>F?0Le!VPnVn8M6xXKH4ONAY)R;^O*
z`MU9Ny487EgT>555$)t^T(TRE9Q7I&j!V<PqKa2a5QSv}jM2r6khLA3e}l*DE)Vq#
z(d>^~3rq|6+_cqb`*o?8J0SJNDmmFhZ)sTl5Y^q)XAg4HgjND@(uPwM^(IGQZg2fm
zntk!GV&;#8>TQ9S=2+w+*8phFhsNPpfWXyH_Z7VJ`cU|Bzz^FUq?r?YIWHG)`ZC`1
zPw>^ZXT}&VS@k0(YuN{ErknKKPuA@`X5QH=3pe{}9KHoOwlS=VRI2kQ%@0$s`U@*7
zpE<ixBMUNGsTqzV^>u4rg5v2H4itH`;Q-i~<*PSyz?wn+HrNhw=|D8YZrFYf5X!Qb
z+|dA7+0hR@e4hnptQY{iKd~P{)zhDkAF_1$i-XwPLKlnyBKy(Dt(q9%w5vOg#q863
zwQCL5R6d5!i7a1^-cS^EvFk3Tvu6NI4%AB<ADjh}zFg+8O2xBI*PNDpJ9hXC>oaZ~
zu(jyi{&MpEd(<mt?hBe?Yt*>SqHnZ)LWaatVi)z&>@f;3YiPD#)qr(^vC5yvhE;>#
z=348(>f58kgUwlfz@Q;}JqyH~QnS?(@_xM^-{0(Dk-LBJ!Fuo3uO66|bCYTR&%}x7
zdTXkSTEeBR2)Zffht%;4ZdRBVHxp7ItxSoYo*MvZUr%W53Uh0k?qLt~|KwK{{AyFp
zsXABZ=I*Z7fU%AP1J%gqvC3kd>Q7b94r%2`oHW+fK8qUCweY$nK<>{r3v=GWPL64p
z+a7)R>EQHp$#IDeBR>*Nq(9{oMLQX>1KJ(`Gj<Lp&SK0JGHeJx@47-+XwM$#kqXc`
z-$tS`GC(ZNEjmCI+_Jkp+A&bD{{y<$m8boZG^mg*UsO@C8B${)G=(U@Qu|F2M*`W^
zULnghe%vNCL*m}&7=Z)uQO;`G7}qOa)7ko`feU>e2Kz*5*%Ms5;X(=^NL-vuTca@=
z#B#7k-1epJSr8xCG6Zf$ot$#3cpCuzhW?Ixyx3n;Q$63>@}Y^^Hfoz>fk|Hq0~}5c
zi(1+2M`nYJ`1s$V(c(7w@dc<-SejXNkvj##W{zh55FrPN&zFL+bFSc4-%U#!t>%bh
z86(l?@f|W5f?E3>0MlL@$#t0L)m{O=tS(MVihb=ncA$chh|aPH=P|++{bFz|N^*R-
zcKzJll>yuinVEu2XUMt+oH7l)!f}@4yeN{Ol-r}_FtKyBg9DRRu6>}DpUn*<kh|iB
z*=jzOq?B9(yeBTFQ!|=##jW=HYjGSH2&NqIv+o^w^!oKf2-vW$u8sS%A6HA?AT~Jk
zJ@%>y&Zl!dtQtq<BE`5b1aQLRaKiG-79<oxk@+gnjH($W=uPCGZ=?36;lJ*$G$Icm
zb+{#G{ja(MgMBlF`S&ELsKCt{xSN)njtEk{YoBVBZ$_+xqp0UTsSzOuP#vN8Iw^pL
zEu)WQ-OTB0by@;(y9((Rj(gdcp>R(Oa;cA&Lij?ghNO3QHZqs?>$Uh(AA0Z9GM#W9
zSf89F3l}%{s{P#F9SfX${ynUbH17$?nECJzK3r@Py*YM7eu`X60_rFwYPutw+WOIq
zEQ%xKpWV%EbtD8;?GTu@Xq?)&Od1PNb%+yHk%Qqz9!XW5QIq^ibYvgB3f*SlvHAkm
z9_+#S)(;eX%DPvY?ocrCZ<IA!T%0rGw;Dvhc{D*OS?STfn4Q%)_yYuucR}TY;!?@@
zLFrVK@*C~>w3~;Jpaq8k*Asdk53pcR^_Rd0WZ0BMhb>a^{JXgq>cihAt+~*M`IW1s
z8(4jCadBU{&5m!VCFO#YW9m&SqC$de@IEv~Z^7%I9N;;5QWwn;O5o3hg%8Aqmo7rK
z1QbRW5Fgwdqu_r|HjMx&<-@aV9q*5t7z}*@-y@*EvdOT#;#%D1L{#&;Tk6_Z|BsNh
z5r=^O*B2l2^Y4Wok-i3guz{}o!VqvHEK4`Q;_X8AyoVg;WsRM#4cD5*L)A$1m;m|S
z;q3i+Km=<14itSCfuFhVVEROq3&RUMg7od*S`TAgOm{PXKpA1L6tN18n`K`kSK-F}
z?Gsu#aEtj27avo3d-&;hk>uNMyKpSr#27m`bqQesxzPB0_a3h8FUITFU{j%z`jt4r
zc)Llfmg7*$n?JDBV`*<t9hg%RDTY%vUeOJ&HK5gT&K-9iK?%Oj&6P^{@ME+pf$CtA
zjQl<Ykb*1#;I4qH>7OS*@{Fp@ADKzfrx?9YCA?YTXOhx1A*I?-|MBCUw1K+HRPpWL
zmA(7zYzDu7%y5{~)-x$jYR-y#Uh;|7=u|~$b%r!Ou^wPS`Ext)uUa*-z%C=1Yd?bu
z#(DI^vxp08-`n{i=$k58Dkv73#mx4YZ|>noFNO*Qt8nVeWm%gns;4m*{!W`22uT=f
z%#dau(X2OZCIMBSJ@z>a=p*69A3jYEvtjig{8r1hdAs6UvcRB%A0b#inN?H0-B~Ed
zE+(^8zjPNl_?z+%H9cC4CPHe31kGR=j&9ButcAZle`@)v6W7KKP_xFO1Gj!OEjx?p
z1!(TbykwR3zBpE8pBK0d#$Q2xIO{Y;2W5#owMIFJ%j#%p{pQBBdHnZ=>8<0)V2u}X
z!eEY9zvo)h26KKdNB^TY=EGz!e1&ok-*%)tC~fjcIolN-<Dj;vTmcLe$nc|XX=&l*
z<$cX=IBoX!Hmg^W!ya1N+$n|v9c}GY$BM7Yf8nmrY&h&U(4Kt!(`iW$xf#fZh}^Bj
zT&48t)=kI{<cAD(b&S`)+_b<lV|#GA$*^qH@8pFmqKr@Q_x2JQ7Y_SVw|)`yRtC2Q
ztQq2{h`Qux9=-><D~g<xz_lTr$R+twpQuV^Y4ub^^=m(-4~#r732>=Aq&bU$%6~05
zQQ~^Sit#ZG=P`WdTQge&h6ET)c}l(b;5gW17&$RBZ~iobiF^A7<e`9vh_kObf%<g(
z5D(19u$2&ctx&>OZV*`f=jP^MoW0+|Z(9sA1YH}zMmw#2B;4-6fzINf%Nhdvv&fwc
zoj!77{&Zq_W}$~EoF6MHJktR&D678JAJL3gN(?mO14__GgaH$QTmaJ~f4c7h0wD*{
z2Vnhx6OO0AL3cXl-0)!So*!F}B77{b)u{(7y`TR6e)co;+xL{BUf$sT<gH%e@6Fm<
zpMr3bi!s%Je&%U-!kheowP&Q~W};?WY8Zt~L7{>RyFM`rUhEEZ4)P%16Zseb{t-2o
z^x%N$w_*qpywUde=}7c5`KOvonJ;Ct2aIHS{8MesLwrVBPx*N*j`xvaj)qdEta6vJ
z_$e?-1!=c}=A#@16^G-X0v#n>zfrSJ<0uA}$$(wk;9w2a9f%GRr5Ze7gr1p#gQ$)<
z3-1&6$z8MpZ7Ej9nZ<datSaB0Uk|<c_e>MDbjGFO2lv0dGL$DY=CuGYjvz`3IyPKv
zXG_I!EYdG!=Vbd2wS9cmkWxL^6`-l+1lR$mcTr1=kVxnr0HnwYv+A#l!tcN%2EkPf
zG?Ga*^=OoQKudp3u{ZFcZztS6*vL~x!WoGfg%b@DqRUR#@-*##BM$Y?&3$a_S{mY8
z;HWRiJgKC*>b0+(S#&TwY22)=JDB9|1L%wjm|u>|*K9jIz4XVkf2Ii`k>CRi5-6Ux
z)6xdEw*<+GPc60N*czragDcxkLnDsG=fkFjIR}0hYnQYfwyZrZHow^&UMyg+RjMXY
z0P_4Virbv}R|^mjKyr!k!lNfbP|V6lnd_k*icV_=;ejtXia_NQOg0~G%?s7`O}P7D
zU0ZMO7Z3|7IYz#9xsVL1t7b&Y6}m!f!z}+<{^x`g=dY2xv~(>%&?vog>i^xw6)*%k
zEzUdwiV*+1i~;fSh21P_=(;1~$4^&^!&eIy=abIl^~6o9pgAdX$T@vArMvu2f$%uw
zUjd+j-Vk$`Ei~YzqaB|K;#PN;0<sVt!8eIbgz*J5y<y);u>=o%k(4VRmqC`Awr?9o
zynd8EbODO?L(f^X7VF|%W+(gdvBs41t}aK5RW<rVH*wh{)VlXyn+~ZW;YXf@IIjRE
z#zN3<=P}&D)8nW0{q1c|T$oM+cWKbW5Gw;jaRuO9(<aMr3HCVtt~a6E=IsV>zvlLU
zx1~XA2tE#zQdS7>)n#Bq^gc|r#>~z*CJRg|T#A~aKOI%#+wiMiD{k?y;1njQ25v+=
za?37vCU9q$S+$DrAW&Nd8U;jlUYVffYLUFicOK56=1l9tl1AFlL^b*2yXP3O@t^W4
z=0`bG5_iUFW#2l3ZY-{&^PH}3WTqhea<ak;_d`Nj)E;e#>FO!=uD+fJ($$AmgZ#Ea
zNIa2O;1ne*H`gk%>(-)ZM(M+^Utb8?aTo4&E{3wjJ%IBX$@Q!=`~K2Y3*Gth*{RRi
zkWlBzMjFTxfu}1D5}+%$ZrjE9_wxR1KTk}Yzz(oyx=U&km=?XrO&-W?_z9&>UrWSl
z+|d_bJhh%M0@?l5Cr*B*?Ge3m!OnMs>vy~a%CT3-{9Hmvq{Gf*iex_h>Gp!?iGkYu
zy1M6};$>oQTuoBXg`~`au!C3<daa8K{Zf*C|17eYBdn|*reMplq)|N>)B%&U9xkY=
z@>2EixIQ^a5UUOJ>~|p!)U$a<PX~PiE`UQT*X{UFShy0fJ91e+;w&+!3}0znm!$Y8
zSkPhChoHzPlL*iW=<S`u95+A%#Q(W2AbzP_gWI}nH%#weI+^5mkMp<r0q%@=xRDQ)
z3o1d#uY-uv({aHT>*+f}CAI1jV(Mh`GzGEHNDOUb!%wN&c%?!>^SG5ofRr+!t4@N=
z;C2}f0YGZ}$1b(10V8~ZWB;ftNqGSB1a^|czF-Zmv*Pr6L0a_sX-d93`!HI99bx-O
zP4hs${>w_q98YO{+L{nWly+f?rO#2`e*-kVZYOIQb|SBB=Ni&vPklQ8+I9W>Qi>@$
z&NVqsT8U-=m@WJMXf&$9;X@T$4#_O9_26z&^({Qy7&m!#5SJT%x;82U)gggHyN)Y=
zc=twUbZpXbL|z|vgD1)_Z#FF<m~)5+ju(Pi!MvG=BDuK;IdVnI{@SXETctpBbw`m0
z$>`+e<CEodzi(=4Z2WKoUHoEFX^t&m7|X$Lcc8xhg|Q;FZA&A?ARjw`Nr;PyeP}9m
z1K<cNv*+BeHlqdz?NgwHotR3|{~9}#eYVh*qU;~UX<zWz(lhP!;ymh&8$YesK0c8M
z+{JdxUS=(s#Qnmkz8lJ;<4Z$4i7s@WbOJ?pXMUrL-GDN+Gv&?4IBOGfVf6o?q5x<`
znZ1sLF|%h}@xpnw+Or{r@5ZnFp8JH53+xI!gIJI;<1OrYH5b(yGBtqI*R9ERRE|n1
zjKCkXM?wTI!lXneU}Lm#;}?VDf7;jfYIsKSGt-P!a%{Pf=i%A79AdJ&C5rLivMwiE
zZWlV}01{BU^3wyDYVH8aCvv1CJUsAI&Fv_Dg5v9hL*vQxk?$Y#S8x6TuM|F`ZW$m0
z+e3GJ$6xcA%v$D3*Kziv2QTv%3dY;m2AL-kOwxj1?4GEW?!Tw!_36R7IE$B2Q8>4c
zkB;jCV}61hk&JyKQKf!qw#>Q4L|TqdKy-uHd|PQ~T*AT*tX#h<a`dtmyZ!VHsOpzm
z+Kbn+u3VvO@NTA^K{8dePy_TDj!BEbURgAS9!N>R%IJxaji~9Q$AU{jJc*5W`RCOx
zf|@`ODtt@*pskBM=d#UElTEzw-8-U4-bBymH7rw6qpQDc%eF_$*e1q_-RK7>9(V;o
z4=^cao;1<Wp!u~hwR_K=ys39W*Mge?nCM){)Pr51tc%C?<+X9mU>^$fw0r|~$(JF-
zzYDSxa<p!1nJu*L5ApMUFMPNIpi6Eb1yY;1nFjs7A_HUNQuiR9YL?&P4F84~s3fEW
z?U`J;9(jz<n!_Gv7__{i(a5rrw9+e;mX&K~?wvGAyzf2zYp<aBeQd%TePdmNo4W43
zL2&7!`g<>V!&i#?h4v!LvcvSo2{*sv#ltcj49B)RH{`DShbTX2(U@-H_8>~n`q6=w
zc4Ro9=p`Yz*KCf`Z%`8uaHQpO1y!{`Z#PA!RXwx%#=J$!Apx<&q6M#14r!mf>%TLh
zK-2fTQN_3KcA_SCWok4HjXht=rHZa(F{)`fb62c$X5nLwFbTfU9-ni}86M~o>ah#?
z{lcOm$2@O}W+aQLohAKx!(Hjx_>n9cdLMX?u1r3^&09DZ<D=bFmuTsvlT^FJ-1naV
zhY^``=ZKm?EcHa#7Wzy`Iq$|u<&V_9@H%UMF-a?{2<1s%bBy%c{3>C~|C*vv;y;#F
zwq>MvYaSH)M_}-W?iBx<Yw<6cANoIYA4b>O>0A8oK@bO8g#Y_lu%8N!T>Mwn{^zIK
zfBKKcm*BkrFMqt41}N#-UlS@B&o^i4@mJMgm=0|gZkulgCkg*v_#JF0WNr^|3qtQf
z?6|lf_gH3Q&RG@Iwau;7-~gi0h(Y9YNET+F4Ja#OV8S5Kv!330VCMdCN<i!>Ugv~~
zYe*BmCmIn3wdJZbDf<`jhFpf=_dap~4j=B+`W#j8ZR)hq$Y<;wGJpkA@ZuMIoT&H-
zq6?-U;}sW9rYYLfqJ)!1vtEq;4b&n6>DwV{hse5GJx0zIS1*wJ-6suX!iRvq!r0@6
zcmA`t_c10tqM7DH$=5@|MA}iyNP(F}KlHJzYy%)1a=4R(h#y7c``3Q3>BO5xVr?jT
z$~J`VVs}Gs4tVGU<jf#7IA5`#h9{}NBay=>AE9N=2U@(F&-^fo+gnFI|3&lKHL4gH
zL1kPxd-juB2ebf^aGVl^0`yn}8Ujos4TgE0!)|g=f$ld>Z?KUk%vw$%!xkV&B;3V_
zk06vV8oO9G2XJJ75TRoR&3YCW+bHlp$ehQAns|zJOTZEU{aTX;t!Y2o{b>Vm(@(H*
z3EPdMYRT$H)9%@QVnYkNQHTTO>#IxLgR31$9InLa48oKEU4&o;c^=rrBBTlVZeew`
zH^3T#;Q|dHnN4tiU@9G|?XM?m1@KW(UHf*X<0OYpnT(`{vM~Z@CAK<L%0wK5jPf?1
z6PeTKeB@w)!Kn4PIduRyry!#WhYnzbXKn1c%pgvE4~(RdcP;_WhWF^(j{#8#Bnx>9
z>!dTY^6QuU0Nx&^>O+B4f@y`pH907xSH>O?d6?}DZ#21nzN~@5Qwdj_93H4FYBU9B
ze#kw&-jGC6QBl#7(o#vdDbHPY1fA|HXlLDpELsp>t+r+0keJ;?BgpWP%R(@f=^;P=
zIh$6}u|1gYuuK6jdt4ji=6o<NXITvjcdIT);=UEah9^%I4<a@$b60x@f-w#ewuB3M
zBQDb)yenhLt-{C3@Gy7vhWGTq>JllnfcMWeVZB69xG?nUUgI&Ljd)&)&iySpS(-C=
zQV8*LDG3(HfO~;6_A7Q66I5g5lTx@4Qk^r@XqSQZhZ!cn%OTMDVPoM(S@kPu+y=t9
z<0Te^pRORZCmgX3ed4y1=oA<)zCpPLr4C6L!Mf=H;(<bn4`U--z(J82dW*9IY%t6~
zWKk?&R*e#=MWwlM5<v$shD+F!zrp5=U-$ZZRII{vvk$iP)v!u98k4EGs9}D>CYfJV
zC3krkofZCo*SvEL{9R0SqZPtS(8a7mNi7M9{CGM9Aw)$_ii>0N5(uC#M>x>9Hpc(N
zYUK%0Py^KXs;R1ufp*};+e0dUi|wWMj?2(6z!M+}Hn)o5-$#n^nv|!Vc9l&{dqA(S
zMZX_v@4yMi2qFYKt?C#X;5gU|I{{RpPdHLL!4$P-&6*F8=^>Q%Fb&~Suq$c!EsxT=
zi(XEuG=LhF=GT(lB#Bl*afj_sa=>$Jk>Y$7R1A55XbLV_?Wy)YOsTE}r`%hcKn<QS
z8Gt{V>=T%^(bmQQIobhDq?zWk>FHa*cOazvfOr@v5ZCQEdOB3d;hoJ$h^8P^&A`}0
zK@vPe6m9dC4(z6zfPpE*Y$F;$CK{1Z1C<8jrr$dw)9FCzTViytGrDzXuQ3wU;q3(G
zsf$c8HeldOsKX=;I9pr#7Hrb#*;yP2yWn2@5+W!7o6&yEQKROB0RqnkLRPUcUQ}^s
zBPve-p4VuAzys{T5v&Tpu4s<Ma6-($ATlwjfsP4<1(w`a2%FJ5knl0+u5c<rh?Gp-
z3fVS(BwU-u;nu`6ByK8cqV@;%I#;BEym=hrajrP<j-eQh#0Bx8ls*kxi6S>Km7cyS
z%7^v#YJ`gxLMZM1N(aN8Af54sUhrHK<R7b$7UrFFeE)uDgt2fanrsbyNTi-GPv5b0
zlXM9ip%NMXjJO2KFR;Okjgx5e2MQlAJ%_3a>Km_4{WKIu6coU5NoJ6-ATHkxCq%!4
zV{Q`q1rp73_Q5)a#^~}MgH~y9WXpd-yCJ#57-2m)2uUCgimy0R9b2yzSNPW<BZg=N
z;^ZC>!`SXDSYJ_HoJE0Yr3YtX7|u(`OW3g$AzM6yAAs>|w5al{9bPVjGTs?}#H%_w
z`A|IvAhRET1NYiSa<gI&K80Nxl`^FBh|#_aQXBuL#dcqu$!A<2eI&R80t!r~!1>^*
zJHW-sxm<JcYBkhR2t@jV!Iw8~+@KKWAFikya1o>OQ+ape00=2AtRc)6m-iDhhV}w4
z5;rh|or(LCCkEBg8-I=Y$pFN|FC)WMk$0&3>*q(J9Rhm70Z+_16O7ywKqwKc+GN-u
znMr`JL0pW$9jnxmubOW#{D;T*Mes>{6&Z*rfVmOx4w0M!bjBgi4rR51`;E%Q4@1Ml
zmvEVbi-Dl^4z4&^B!Lj~Mf@cB#W=MgPZjlLR$Q!ozxEkz-^4yK4>;%$528czF<$UK
z!%}m@vG@jLF1fOxoF$P5{2|Y#m7z1g0onWE9>dznb5bTF-|*j&Njex@8aq(zN6V&}
zO^{TiK5^oZhpa#pqjf1j0x6KV94U6#Xrhy8XLiq1_zZ$^Wt;;~T)q2o4n4jmRX+wL
z)$F;P<Rb_AcAX6Vk;#DosILD0cumCnP-nrP(ls=ch@jW_W0?7PX=1PQ!*ueI*U4a`
z*%~(k9+7kAfO{4-b9Iw7=0)p4W-Z}!Gz<*hL^={9&ceAE3pGuD3EpZf1%fH3zCtjO
zqEzu<OETZ0U^<GszzTYkb3?=${T697=q%+Wn0D^k6;08@&`6RJLrub<R>mUTO+y7m
z&d!%0$hQTXVI^!g;f!SY+o>&x8E*Q3M=41a4^FoPzO#*cd1Vq+QbfxU)yy*`kN>>Y
z_6b1FW2f;pBygr*V!`Mm+$<7S(ACiF*@L*fV>@YR0<`H!zBZ)b3czKI@PUep&YwH?
zWv}xRdt7i68;b>Hu?f}P@Mn|>%~J#Q^tb|^jz}(Dh2~nxSrw3JxCzQ88RxOzzwM04
zsaS$B7J8T)fn!6GDXIp74X+o=e_4?;z1a<*sS%>zY%Q(`IW4L;G&MmrBRPR<rgvx#
z;k{j*oskq=!DTfe=9ymmocnqq&!)gbN|QZ;jZ1-bN-dnV8kg=$O%I!T4j!H!h=`CL
zrOJjuvX9_~%a}ZUG1+?eS+&q-&t${HK7nmygjpI4&kejafE96eq{m0eL^znNqQ({n
zv&m(MyW#s~KFyM??mh$6IW{&Go==V~h=5r6m`$YOOl_A&UfLI2?wTqPNZ;|xdmDXT
zwdo&`&tCX)SCf%6uaw{1MUBzX({llUBDqNsr9sAFnhE9l^kY-~M7BT}2ttz8g}{9e
zayX;!#>wK{yadA*V3jePh;Ly5=vz3#anUwswOo82c0NESZ-R#c;m063ete?>fTl?B
z6#s3*F?w?kj2XDJQM?hG4Ds|Kp_Zr(Ec0)sMa;@^;NKGe9DJ8vbhAM|SgTc-dKh@p
zkSHAq#WmFA2ZiP`A`XWK2H!afzH#FI#PLpitGG~iLcuQITx4@*n{(e6#Hj5r9*#AO
zkm^L4a=bDfb_OJ~<fFm2(!=qHL98CQyydY*qAvz~!r?C?5%p%m(D;HXl6?A2?OBGL
zHleBqeaKMRfmt=nzirGjL7fVjq02UB$X5%1y1|(vk=%mw>xjw-gzT#i58s`;Hnv%l
zQG*JlZIxw<p~*jo*bRdvk8&V$;q=WcEG$GFO94h8r6`CJ`tujDCF212r?JY5>)xQu
zD0$8z37asuaLLrikGG7-;X!+G;sSt8H6JEza{YlZjle-qtY+w-ED(MUFlZt9>lynO
z%nGp|a2m1m@WkMpg5hEf2Hp_SBZ5Ah{igIJH0ftC_d))ihZe+$xpTt;&`T784edkC
zfdvjPybN_UlyK0vzk|sU{mT)E(U1V<j|#evpCc#HeZ;lcj<5d+Gb&<=?RvXNCD29`
z=xrs;vrwy`mSM!8E$F2%5)~!cQ8Y0mAqyw=%3A^N)~#EIi9*6sJy#3_sy<a#vNv>0
zf~<x#gE7ktr_dcgodMjnL<NaS(UAZg_X^qj<5d$8M)P-DlN@You~Ff*CON7j2M>kT
zrT^tQH(vlq2*-h>^&29RaLz~?^T!3azi;Jsz#uno-1ru>T-_K<6bzK!IHgId#oKG;
zM!igHMw27!?e_BtFN|+@ZejkeieIoFQJUz)4oO+Ld1_(Jpkdhrw2}d0E+hqb9lEdy
z+|%Z4hdchr0)rA{&AN3}1LE;?Cdf(Gf&I(X-CfFcy$B98QXWN1hkgZ$0{JC|#z!M#
z#*1+CgWuVa-opgZEy|fpyIBVo?L%unSYi_I$`9xr@FOMIx$z582+ASH=VzDh^Izs~
zzhKiIcJ=h|C_;h<Tc1+;brmcbH#w41guSy1y&t)N@FQW~`0~?2Oj4lgp7q;{cqXI;
zig!>&DgYj?G90QXScnk$!_uRrT+QUTIK%h>DYnL>QTnt06kD@MV~7Htf|;D2xXF9B
z-Bv}dzg*KJL|hzyk&4%{cMHzb>ZV<&wt$==i?IG!D7(%oTy-z;)8Vh{7u~=1*|Vwv
z!KZtNZj|6MJ=cWG?F{Bx1&TJiNQ*${29g#c#7oKc_)!!l$9fM>Pko&Dl+2y4T2G>L
zzg7`Kbn_%W2(E3+F7-m+*YDM&ivA0z?UwD^BQ7t(g$=n-tor-6NrSk@pbFgy8JEVs
zqY=ehep(Vswz|4H0ck;uFcpC-34-owp!>%$tk7RewI4KX`l=YvhFe*D%Mp*Ll*ztz
zbJeE3u>)%%%<MlIoS$d-IWLUP+#bcu;SA%H(c}co&moEiwG0n{I`HgE;bJBXR=BMb
z_|U_Th8Z*N$`~0g^-A;$$B?Hd=u~g=w~40_$y#QJQhQ=CDGO>Dv>8R{AHaF`mLQAo
zD-P-jrH8K&4~1s0^AWuxrc0bnUPOOE>D@Tg45K=_o^Whj`=x~`ulemX2f;W3D>)Ka
zQSoic`2dD!t*1at5opkaQt$R?`vcZHtcz*{3<gUFjY*E?4D65%sKcO{@ofp*fI@<>
z3YmuKSdwekua_JnLJ<nIE{HKCu)>GZOA<h_zB#bwnPL+mEJQx+>`Dz4sNf*X6hFyv
z5+NE0NB)M@I{tpQ;eJ6%^)rNHkcJ$$gXQ4&X8@GZ^4tZ8!V0s#D{zF@prS}_35=_o
zPosoE3J`R-UkeZ-pDZweOXrT?@0O3HE99^P$>R!lxszr=3G|;|isE8tuUwEA>z+So
z#t})^7YE9PC=<X=k=U1dhDqXZ090D@O2r!`L4S!9B>zA-5yb(84^GZMEozP}jyxA6
zuOQ0h3N)>@rG>^vx@%aY0XOO0$V$u4t$4vFf(FI1ixB3>DxqoI^#^);#!bvV_*|Ji
zjeN;yFrNSpq7Xa?r}}EmuT2>sHc%Ci$xp-v1~`da!SHPXJ=%ed8l7N<HXF6-13)E}
zK*oRSRXl?^0K7GHri$dV0(H<vS&J@nHI{Pj!lar89Wj;TW#bSyfgw<bD<V)b|A}XW
zue<l*6(obvkw_R$L}F^#<B{pD89maD79*4I8-)O5KlQZF1wF<q3&Tr|b0xR)uebGb
z@YlV0`?gBTaK*zsf-B%nmZ40eK`c4gH)^C#p@+K(Um5(E_krKx{+7sdJ0e7jZ>x_6
zAJ_Y9RI7lwD}ZRL;A{c5mHFa99S*u!#F4>THb33r2`vD=w6|_^qt$BADhYp@M?x*(
zdW2H!CBC_TU?9oQzy^i>j&zmZ3ZW@^hK^e%*ZwvRFL!9QpcW$@b2M;r8Snb-Mm6j9
zEf%n7<*FmS@z`nD`C|cmC7_Zzid+`F$|#B=IDadQ+Ra%YjTS`_Io)aCLv~^+70H;S
zD1x?Z9;*VjSc1+bqX3Z^uMyK9LVN1gYjZ0rMUr$1#cQT*PEO9wmt@!p4F5HFvvSv+
zjGDrH<cKrW8C44!m{Pm|E33nLc&H@+@<`FS5mM?04<E(@T8hQHi^Y47LsfWg4zer-
zsuGOtq=Baqdxy~#lK;rJ85x&R5yS%$V}>)5BtGqARad4m!lTaoj#h4EEFLUl{DrJU
z$l5aBVoPh_2QXny4R~ATom;nt;qAp>7@w7-57Ku|Nd0^8=m4WxCpKD%5+KI`g-Q+D
z97Xt=eI7h`KwvRQPmHjtUHV6wGLn~(KizKPhPgoqLuJMop%Pp=Xg?(#F#oR~%gXj9
zE*divlC5Gjiqp`{T^K;storHE7C3&o6O%JiQY_*-Qd3*o5TLi6uKMYebJ$b|G*dOs
zRLd36XOUk_Iv7B|N@xoEv2PuDsAcVUkbMmQUObL%pMU_Bx1j+N3rhm_IQd#sM^L@*
zc!~4b-#`s6UraA}0U4!_IV`jA7nUH#0&nILE+kgjd1>K=NCW}Af?d%Uzl&KlH8v3O
zuz<oOo5W+bi&JtB5mt#7X3H^nQ&IOS0tIHa;re0p>R&Cu@Q-*LFh;nan6VWqg|JMF
zfVag1dME)vr2=^#Zj%IHT=6*Na=rr`OaS|)h_EL;GqYsCWIkD0S>*oZw;uG@ypF&U
z0zZ;o3oWfY-l5c+p@*r51h+d&Fr{HgS{KH5bLfA41*}B@8FC6ZMhQgd-5dkq+&US;
z6?1pvQ7NH?XGRVKc`th9fr?nyiYA!;&i3`&ikET{3)pjHJvV&+eiU(dWYteJndQad
z@iCtsYR=;McU^>T0``K<Fq<26;QmL{<B6a(7w~`11(LWl2VgLS?pWb2mt6GZ-#|t&
z8y>TkNJLDMD;ZTn&Pex~t4|?j#EqPfY3waF^7l_XmUG34_btOrt9i$!CB6R!pk|8z
z1k}e<#05r=IZG>0jRhJcSsf5s1=u74H+|-N*`<O1CK{7EFZ8vK-~F9`y3N1;u&?>Q
z@E8BXCE4x3b7oByhgJ5Etag#7b$XjS7kAR%9am592xg$r3p&mx#s9SqBRKp{=Uni#
za&gYF0?Co<JFndK{_8jU=k!nK=I&GSEBg=4gHQAS!kzrVwEq0#&)*J>MLU~mH6*dm
z#zy_E-l5~ipTJ6fRo~dnY^ZT0>-K-Y_`!R+x0ani2Z^pmKF`r%tf%zC^Rj@OrM_+)
z`}PrkENY}{mfk|*l77dpLLJsw7U2CdOynPuLm^Ss7d_BGA*;V|D9H4jnCE}r>S!C{
z`Y4mu*54Z%$8hsx6iM$E-Y1}@7C$>3IlCK|_i_A<5?|)F=f|k|EIMBT?IL|hY;5e|
z4kQDeeg9VPrnUJ&eUH63J06k#$nMW=8+lU)9>%_JxeoD7j6y=F=X4BYws>sXv}1NK
zMVI`Uygj5U#}z+A1h->o`LbzLU@wAS9EIyqik9cvQm%uGwnMyv_^3Z}S2e;Nh)V|-
z9HJCYTvADqOBLISCa4_u(Z;M3?Ga==V9BjpSM-fbfmwPzxmgiBmpEnS_ceH<P`aEe
zYC1esVy+xdPcv<vP5gXZHS#g5<M(eU4!I3~eE&HaMeY6M)ORpDGf+rj_n$|w_))E}
zWr$TfC+t|($EmNpjkd(=n+*7XvUj5$>>OU3Tn()-hus;4n#_BNof<Q<8~2R21B#)+
z5qw7EksPqS4JHwPA#eI{R1Mz}jW7Gi#kmUaq$x&n^ag$mh(B&MP}7FKdk2fs6V!NA
zuX#?r80{!}W7H)vNzdzyTahI0o;0g_0=z*M_m=HADqjEHs5UO>2n**1vEPGewV#em
zcH2P9`|YK*i)S@<d~f65+=J2FTE^$X@VoWbBnIhT#Aj-;&BmRQYudC6p(*GdFR=3P
zB+dWcj4DYNCp!tL-9#tQ-%c-NFNT)6w=q%8YvQ)8ZNFe1Mw39JLBcgiHb-YgXlS+m
z*Vnfy4^9p@h)m}?fV#^~3vSP!8f@S!@}S8_QAZqt?CEHM3vg_7rsIx#W7r}3_Wirv
zKOPpk?b@1aW9HRFTazq(XW2@xACu$7+tC2Hml}V5VmrNsvgAy`Nxpr^TU~?ZK>jvo
zC=HG{y3>y0=ZiksD~_WBO<=VfpY8Zjs>Ma()8ETy0YqU+cR&)90<?q>Q)drE=UoPp
zP*_z}^=m;hhiPzdNdCzjz=T_i^S1*V8UJp{7z3;aLZ|@`zJbDJhYp<WeLtes1B>qh
zn-?ckgvCXgAouxsWq_lI&A-*KJxZlI8vTbaW>{1WDSpxgvPocK!pfN~+H2TD`5oKa
zRrvG6Y5YtPgym6@T#SXeiHP7lVJ$9$)41cR%$W)<?1%B)s4n&l3PI%zN5Oq&Po`~y
zxv7+4ctU$#Wtj~oV0L`i9Vs8Eh$59o+cQn*1gyes(~J=8MvF&<YToLz)y<pY6YO&n
z17z~63V@m#N0W64aq`h8P~&5%h;~eU_m+qZMZNS|XU-a9Jju`&ARyA*=^#eMp92X&
z5{L*cqTZm37tJAr`2<)*67`Mu;6me<3Qy7E!KOSdVf?m$Q+lpR7w7g7j!RSfm&ntr
z8iUJLZ^Vp@he(Qhu9;)Y0ArR9v!bhF!sTj=`BZOxCle*<d(bJ@9<=X0Hsjh)4aF|N
zmHN`bX#`T7oYw6+`f`sFUV{UG(7;|^bf6Jx^+}mYCGVSm#KNW!{$Tw?J6uO=G_w8U
zg@ih9+I(g*R&w3Jda<5WV;aO%tL<0B>wWrK_Z+W3<#shA^X2$+-1_?J8tUMO9K(YW
za0BDOBO#NS)_r#q+|~<fO5SbBvhn<Q=MHf?$GlRxX>~KqD}x7tEg-I_Nnp(e9E<4t
zHlFSRgOQS^2RzxOn{vG}n-5!J<^J9#vp4v+_*Q0uY=5qe144?V(LkR#-rpPqSSp*#
zv81zj9x;RnajjwKlFJ8vo!peB^Yg<U+d0FNw6rrbR5$wC&uh#8vY<?J9~EO0LK%i3
zYAz@=;Jb)DmH~O|P^J~|XYb+?=Gvic^r1aBZxZVpz0<4<uH~rR2~wAMv!f!!B*6KK
z0q_J7TqC(i1!nlEfQ>2SSGT6)kN8;A&nfIf#AU+#?p;Md1LOSW@A<7AId(Z&Vr*o)
z7t~6d=EvMLJ!M6tg7|%fFTL20#g{R8&^}ns73bqsQ`4`|x^Li*o?LP@%e3hM*;ems
z+^Q3o)5<CLaGa72k{VQh(Z`Pm8d&pGp)K3r>3(jnb3gzMtcX|fl_Al0L4_;>6qJI+
zcy7UVf&dUYF_*x6JuL-3&wJ7*S}O2_;g~LeZe{>bbPs7?=G`<jZS-5n2w3Dg;lqmp
zt|yje+-obZ;F1BY`qJE&dBfh5Yk=y04HlGyIRW=-*OhbUT$>}xwHWw#UTIa}jM$CQ
z;C$xYouFtoz=PEYPh{@GTyd!IQmMsBu9a6NU#So|CCOff%#X6~)UDHs^z_Jzy9j`5
zG+y}`PPI2T^H;4}y~-RTWYL04z=o@mX3K@o@bkms!jnD1p!~*U=G`}6b6y0K-rg_q
z<?c%2=j68<z%$O>p+*1iT4@Z+1BVvvw6vIN1e4jVp>M6$i;D{J>J2Y-2p8ONxah&<
z;EiKBvmq~i+P}2r<BsPnULk$n%<NCEH@|!t1&oaQnW3%p$v{bE5sRx(m>ci{gn9>z
z`p|q~2y#d~KPt_#>9yN&*un*eSNY)Hy?dt9IuKw$%#f+W97VE2A+1wpc=n>@z0PV0
zCZoC8{t~RF$H<KFqH`(uJXFp%wE>2?^8`}8ym1e)I^UbrNWF2O&2EaM#ggi-e*qmw
zZ&S=>v<A8u#B19Wj_Eas1?KVkjf?hZl-J%(FDf|>Q{BStO#zhowEI972@C;Y>7$KB
zZkuO20ptd|^$p+gi)O>l+$K$Ob=BTsXU-76J3LqE^YfelGY-NkiD#Ah9X(3*8xbTj
zy$?U82467Zd=cDm#!ToF7CI+#ozQ0ZYA=&%Dk7$cR?mI43;G(s)S!^<t9f&%`rf?&
zKI2`^!}){g{%n6cP>59pcnt?cUA9ww5y#->Jp>`94xCS9s5_+WDDw*;SEC|+T@V;6
zG#RzEE@SqA2c6)$(|7K!jKkqJ#`aHKZ0I{Q4dzgST8}jRt~+<ny4MUB@^0W~c|w!;
zJ@2f3Gw$Ir(S~^i;^%XI1zFfd*e>GS#U0(<t3F_XELsT^#7G5EV2Fyu$v?k1H<Ep8
zbQwbMh_x)XG6E&i>yHzn7cM{sT^#%D(CyiACT~XR#U*z7Q<R^e|ByQTJW(yZcUK*g
zQZ_GmsOjA2g8~k-Cn-2#*UV0;k%f$8CgAw?ZC!=m*>$!}oZ7LZac}u@Bsf2z<1Bi)
z0J9c7E~Qwt<bAjd<#MJsX1~yG1A{_Fdb&hkcXM_9gf~6XvJzviqhQM#?``;<lW6no
zz=a%(Ic^H-SZNEe!mFZo;NG?^&I9CY!OY=>(J6-md28S@27fpPd+Fq0%SPvh1T8KC
zXEeW7F<F3&-9PWjousm|A1|;0M?q8c*A8gJZ4)+YI9q!;U%$26jV)JfTrzlaIqRuY
z>d6~_nLn8sfO{B%O$1_JoZCb<ULqwWWjHrcPnjRDmX0&>{9B}*ZH3Erjmm3hA&0sC
zjXe@yeTu#L5Ox3=Q0dD9`r-j6Z?>2Es;%Ft{b_oX*pM``YzA_b@vN60ZctnIjcy5D
z+fqDTdv%PI==C4(p24j|*jyB8%Uk)azJnpeT~W*MQ?9h~<DJ2zq?%W<%aPS&J=7vq
zT5nrn<arxgVV~%E_UqF(;7=utGM~-lah5<HeRPhQ)A+trCv?AT0UfuS&u9n0%~Z42
z5Uovn$wV`tD&4J`>qgh|A{bJ5X`T~W*@9YU`|1G^C#;xj-`FD}A_6#Wa{%EAH{E8D
z0CzY1QHOBPN&tXetMcwCx>($-yOEdFnysfWnB{XokRqkEVEg=OnbA20p`{mY$>nBY
zL?4l}^LpD-^Ut)@;`axWS+eFc?<PMN7KYfDLVS5C`sE}V2POa+K(XK=AKOGpxD**|
zhH_(HbC!H@QxL!(**J4O=G%7iaIoC#o)R<irAF6EGL)gAmte}7ZQ1(}YE}`@WiCNM
zYT*U@v(-QLO<0gAQ>fS?Lri<ko*jh7YAJDa=~}Ou+^G_uFf6qnxfYL5&Tt|;c@ef`
z0+mg$iVq7swywP_k{j`ODH?z{i8LDPsz{!q%Y?D*B8w)`bDX%|D^r(EEBl-lw$wJf
zOZ99^lAXZ1dd08iCp3^PQT+@Q6advi^a~>Z3g^(9Ke3${ke@4MquIL~!)A%7?3Ab5
za?P(g@1g3|FpEp3<&KmIJB!S5H++`YU`EE@9d19wtv99%E}<7Hr89FZs_&jUw6z`A
z%>Mo|ERTYtfCM-eC(R345TE!)E=`7lG@S59wUYKiK_26P>9YIMLql+S_V)L$`%tKx
z8R=B}`hd;Ik4nL%g<jc9=S3N5NI3q~DWp<TN-qX@J;|_DOwwR(gmNA0HU+0N_bV-F
zF|jq+d!Iq0VTJDIv3|vB@dE=8pccJ|s|u{hJuD*%$q-(JTp7rX<*gQrr814LUOj^|
z0v)2QNW}Kwjs&fF4h9R4#<3&GN&m#yADq}gGmFsRm_4%sNS3nXa|F^EI~LzTvaU(q
z19UE*%W8Z^w=<IIC5Kz{M{qh(GI<Yp5r1v--f^@3$#R}2G@K8?RMIH)_`lJyT0+d&
z-rO|fYIx8#(Ft8dl?vs;OC*4Z532z(1o85jba5!|kIF}Hs7uZjp5rooVGygEP?tyy
zZmTCpcw*h#t1TdMKsEdjp&@5-mIzPwTGo7Ej!}xK5VB_!yXz<V>!mrOaY2ed7vSfw
ziZ)SWyySm+x+P}>MyLZK&Vr{-VHW*QFtw*{aV;)c0Z7v+DhtSbbYMG=QC1cP0rM0O
zRVI6Bb0=GsR`@V&$DGcMS=J-nRb8d8VDu(8N72ST20tQQTE1O@h@$&z(C3-FrEdJh
zC%f72;@%sT$}qj&lFFoLVn53~;~#5MSxI!gR+{GqF6S(I30yGQHr~2P@bBr>o+m0W
z4maMUZn@Wkf%>GreHGd|RFHP8s!!GsVJfU)lArD<So_d(_s57wn<v_R<*-gWX48y+
z+)liA?*Phqnp3)K0DXew*hGOe0j!PJUapnBC3<IsN6j$zTn8#(@OzK#YS&lDpt$ZC
zupWBs;c;y)-6XE@NNK`8<;JjKKzpn9$1T3XO^!hC0qsJk`bI8e83~4<$C8(zy*k0p
ze)Xp_m2GHiW5poKTBu^9=dN~H0W9xEmj$Q5p6bhP?`Y39HT@d3%XH#AOa=9^Bm%m~
zO-A~Z`pmTLJ|W`QLhe?_T%uZ{UpLQ)S^(lRm2`VPK!J>Cq`pEGb@URL9aL_pA`<63
zZxOA&QAeoLf^X~~x@Y784+#B`(>6ACGoBp$kz?oAl-w<E-yC<jXIh1$=bw>mVvQ*O
zWF*+ZTSGh%Ikl!Wi~1Iyo-D9Tj8Na2rB%+d@8?M9(M15^q_Rh#S@;<a!H`tdlV)Fg
z3f+#(Oy$6Jgcd%NezvXat}DqRG;gFg3Cd<xx;UV>Woc^=E_`^DPl<<nm%xrgCYqvd
zVPfO~RDNd0yvu9BG?~sTiSB_7E(b`#yo1$b<}@?~%pCE{oc-JkbNj$I;0s*lDUq4C
zfq#~cLQfy6FapC=MeeT~q;NxmUrUb-@B+%&UB_SIG!kcJ4i~&;K5zyt!vkC#uOUGM
z(`CV9D*mcFWGje*6Cb|c?7$Q(;V?I99Yhp?4%t4TOl<O9n}xSz-eLzI_dSRGh?X3H
zXRV}bc7wZ7{PmZAB$3lZiG+zs%{w%*EU%i~;zc7QmbxKj@?J?n!F_NMm2dSV8`BN*
zKkz)as{5*Ipnd&Cyv0)22Py#8B6zlzdZk+Bmw$SfJUB<ob^Qy@0Maz0XiH2wg)v~~
z*HbTJdnC7~<a=iq%df#c`Fm4;VZwHudKh(3K6!Vh!*oHi@nClHh>K$)DJ6Tkn2Tkl
zC?-v*2cT1XvX4{LjCUAS!8^O2O{_^tz;{<Vzq)iueYWoinF<F;Yob4Z$Bay5M!wKe
z-f(grnbnDIJFqZ0myJq2%XVT(q{_?xspjMQp^cmMuZpc<hXVn!OpV+2_NP!WQ5j4(
zPL%-f#DT1>d;L`q=j$6`XJIaSjwmJa6mkv}Li?hRXG)PlB}zf<`sp<$ue@;(7~r}7
zVZQd$i(ecK#M<o&x(QIk%Af1m*a-CkvTv<HtESnT3gAUlGO$Hh=CHPRY{Of}If`0d
z{N|=SVAB@pD6bC<-mtL9M~EgG@gp>LZJ$=6PY=hjiGY@a#a~syZ!HtE-f09#=;4Na
zPr$m^O%5~QDP@{C!8o@0n9>zPL-9nmypun#S4%%+prq&)e~K;-LIe=q`tAzy`DNhr
zH194GSGPa41a}hFBuP0SwAM|7qbQ9pDmGW&y*HYu5d@s8GEI$#d<kF_Yf*4vkUQLo
zmcKh^>f-WohL0vGKZ`C8e8zcUQ^+ZYk~<4?s4LeKsvcRQsCVyPhwvDq^W<Zs<#FmG
z7GS!kWT(<`Ef0_f;X7}8%eT3)O}|xQX|(Ajts>ZbJ-i&~v>*uJRJ7W#oA)X98VBxR
zjAU~MCs+x~jCq%Y<J=Gr{?5xeXRdRAWeKn9`XGv-K|{lEocZ_PDMke`tWkWGsluty
z<7)7jNJ}_bAfpoVK`%hsG+u(RvtlgR-CVXOQP@*7lWt_{Pnqs|c%MY1W*<{3sHyR3
zbzBNwJ~yQU+bmQbR#wxqvtx?+53eidg*dTlMcca5p0b^OisFI@8>dR2_gTI6KV|4;
z>fcblhlfHU5OCX67m|1)zo$FslX2>i<mTqzz=#mp5jmw=pQuKSI|WnaDw2W))`oDq
zYaKcm`>T-pl&8BM$~`joFTBuoEAOq>rg>1@v?J#S-Q2sIzZoeF7M)jmUtis@ab0-0
zA2k52L;6YTKi)e80dM^cffONXB!AQ<UiJjN_SU3{`&6k<7+NcTHRT)QQ>`fOH$(DY
z79gl*a}l$Dy32wkoB6F`I|kfPACZn5H{2jnWH8ZocJuCENo)yUD9}FYfojg|$)n?2
zcKUmFx%&U2?5)G9+_tFkje&uRji}g&fS?G7fP^9v(xD(FQi6n1B5i>|7${PL(ygR4
zih@YDfP^634N~7&oOACv_xt_vd%p8L*XL4uzx#dHnrqH6#~9Nl+2y`_GWUG5ia&UX
z7cgdE_td2sx8GzK1(!+Xee&M6B|EuxO1~>Jh(%DLI>ztANM|wFuj;HTW}g1><HoI9
zVegR_DVKd(i`!TnrPjycsTM+7#%<Cs_J4$S)#CxjODB#}tv+@suL&nMk#WtZTM_Cm
zv1T(ic7rhxx@@HXUTs!R+47>ACl-OCzdpnJ(U!ySul+g;L4hZlq#)K8YrPk)d$2=u
zp#Wn*&O>Ea^thu4Y<PB9eSkj3akKJJHdwB$9Y1$z3oVFaIHJ(}^g+9+ei*oH2W2BO
zD+^3by9Cu19^kEHBoi`@+!!BD)Ln!%3Rt&S@4h0AtRfE#)j=wc$&CkvVpQ%Rz==NI
zs$JV(4_^<Knz}x#ZCP2Qz==`Ic-3lHMLcr<Gn&h0m7CVCGZ=IG<&Ku4Hg*N3?py(S
zj05onoK<1BBiiCP5ABp^^8&ihz>VJb&kp(Q+3l8=Hd7Q*>P!V?t>!UnU>%QK=mKH&
zhc15in^Y)<HaiD82Ri<FG!Y*bS&8vWhiw|Tnk-V))6o5dwD17~2Ec*hLCW^A&QeCS
zwol(48->BdsPxpJEvxKx`W%iwq{&HSeQ-!3Mc>KH-ug=)Z)g+UVc$s;%MWOk+UFS4
zqtgM+q+@wgA5P2ZRVzr-2^`V4mz+RO`jL3+7@Yickgt*vinzCE#eECP6$bd>zsFl9
z(*QxYE<-k1Q|o>W2Fu7V8wMp>xR*}Qc5Fd98qR<^dU`LfI>18jt=zBA#eqDYFvBBW
z<ET#}!3Alu8y8nBh#h|?-*7X&r(gaAtJ1u;al^-F5CD|NFA=JmZ(01^@EMXAIF=Kl
z+*gZW+vdO{x|*zhPw?i<Pn;NC`q~|_Ec#O1al2CgsU|P34-UykjBUR?6&3#b720yZ
zepe%NA#5%FtQq<611CpRR~%iN$h##WlKEox^S2_-e`nrpIh|-wO1(bCQE^;szn~?u
z4=mXSrS7|OhQd~Bzu#3Ck2)zJHSYU@*8cwv60~t8%86?pg6PTNY#;AwoO0k1dB}CM
z<c-F5x7+Uj+x}X0zVqMnZqJ51{?Gba%0{Q}XC>~>{ilO<_TiN5{|s{*S@XY!r~e;m
z)Bo>}uTXzPKjJUR8u?s*0cp5l4;Nprlcfm4N0C`EthK%U9>#kjttE`hC?#fS7(XLI
zai&Z3umGpVF=YXDk%kX6fQYwWs;b5WQl$QQu{k<JF9nw14coz82Wce@fDC#<C<Q@M
zI0**P(i~g#u#oN&B@6%reWk?88f1+iMgBr?$NtGgtMp%2cj|>3UpZ8#m>+l&1r%W<
z<JZX$COYCfi4FLbOwmqCB7=J9Ych$2(^<skb<IO)6b%J&|LeAfNV}}3f(s3C@zut-
z8<fLB@}so5;GyDt!}K>ol}>Y*O)*7vH1iE*4HkYliUc^80uJ=R*$CDpJbD?Uj=^0H
z1PWMe-vvmETNDmX7kmtRze0Pp#{ImAuI?e2Zs$S34jDyn3fcJ}`}`Ul^Z?HY$^vX!
zh$*8kd4e1RScw7>BAF|Ui1xyN`3<zwK?ngZNc7vN1V3YZ_^MT_5cBFmKzuj$2#gSv
zMBt2}8oYp$6y0ngKx1%8b3sgDR*n}nf)Lhj3W(Cz=4MT^Q=Aa_(}Bo>N$ml+Q=vT*
z@{*f2Z$68Z1C*CfpuxlJq8xMP(L4;NYXr{}l{D7<RS*x!SUSMNFhwVBtgsTPec}X%
zR$@0G?o7@0?LS<ACBsLM?nzP#ApH;@A79D9@es{_9w)8KcM2kwvh(*h0vX^L*j>-e
z$;nw(jZbf>2f*$kRz?(npeUShM!>3j!vh@^aH1$UuU);mJX+^(-pDrKebR4EtFmdN
zE5Y4Fg_n<y1s6!j9Z%*Vi`q(PD9T_h)jZsEuo{d(F9pc}Atz13UdDelxQE*|Z+?em
z@(%7AaTIQJdo%=qdn<z$Tm91Y!GlP^@lj~<9T8SWW0fug?5ULzXpKgqc^`#48I3^&
z$<Xk1bazLP$)os|)d&$jfeDMW{T%Xu3v8O1DpoK&ChrRSqm>b8`Q}VC>@?RvEggnO
zWmFz?N0R|r9g#~JVeJ_@IzSm0Pf-Q|mm9HviHEasV-TPZC7d8N^RrV{AHZ{xcLqlA
zY!uT%zz4kq#f1?DYe@!IM3f26O2L)lNT`>jxVRk5_{cXw$c{##8Abkwh<gOeD*_;g
zS?{<U|2LM07QO2Mk<CQabJ5udTQ{l=a9T8_1Je^&SrtMaE>nVXXW}Y!9!lsvdYuoJ
z$1=Dm$!zt15$oIFobZ-hE>ch{y@nNv<g!tRPP7p{79pFQ6%m>E4mZMZP(KKPWBfbP
zh$xB`>NO2jg*CZ^!gEM+96WTygKa^m;qW?#w8X_9YW9N(y)}rEtW&Vz<z|uiCkm1i
z-4!BKM!4CM%%OPUZ9y288iQ$)eTfdU)+-}0u<wvjJ9_o_0T_#{eT2EA1mTtvCa`H{
zVywuLyzC`b2MFm@QDC?{>U{$WOq6`EOFD+oefqzauY}7ws%%6x3bu`;awg^+Q}DT@
zlVX93*_MGwH#{*R%KQ}(it+=~wJItqU^wE86Owo)HRWN16hPY>=zlbA)Nwz*To^4g
z2z+Xyb5IB3GR1EFM6!jyK98kIe5A;Wn@`^~s)>-l2hBv^OK_b|Rm?yLw*@R>1iw(a
z#xx6b>ogRtj*7(7<brxdWQlO&=>otUf<`WJZa~~-y6sk-DTo0Bl-B6@?8@tTNKeVx
zjQ=2XVT@{A@(^y{8vhcpoyfv|0~xaYfE8zeDh|sluhn{o!oSzcA(8D=6{S<~E_fPc
zWAf9laE~*V-%)v=lXt8^7;V*iPJIQ2@r8$@l16X}q<5jMC4Ryb=_yfRp{sX@x2_Qj
z4t*@?$y3a?IcHWa%APdb@&rK_b1xqw9(AV}g1ZI$U}bHb#%!%lXh%9))MVf@B&#Fi
z<G-=(j-r=PKgRugA`%mv5==fqF`Y1xIaqT2B{<Gx@Cwr5LO-5k$oj*8qbfyTl=fDs
z1vAp6`tQ+6*aIW`3!kpI0-E%lkQaOs73{rFl{{HkS)<SCkqI299sX<;_71^b+lvMT
zE*5Z3ogJiiuaD@Xhe;;<Lczv)v+W$_ct@x)d;~rK#r=NnaIct<?IO1^dvD#-@idXj
z0U625@kIdPWw3sJq?+x;-VaC9vvBY_mMl<K(hAZQXwdYDzfh9AZ1wu}U+e0;Q4YQb
zWfB6(4Z`?4VwFjJ`V{#U)c*H*Y<AYv)I`e<y!{PWL}3?JGa8`#mRqEJr)3--ROhl<
zlOF9h8I|C7t3%g-ET6O{h;*(%2L{Ay%bq<RPo6w!o~)j{HkX=I*N0<G!W1UKlF9;x
zS?p-85_f(a1}G-2$wNSv>7yI7KVBB`9!aaX6~DrT;EHk#1q9pgEa)15sq}G_q4(yB
z0tC)Cz^45^4Tzg${c|eseZJ0Tc{yfx;5W!PPJju4SB-IeVy))l_yjlLaDyYrQ0Me;
za;zjZErlFN&S3EzzU@}m_3^?Z>uwVq*qa#_+0{P=^+is{fnfLmM8ccEEkUV%n<~u=
ziOjjEUWp<rD5*B1)Fi`%$&_FiKUU!I*DVt`s1;a!9O3|chbWfN03#U-h6^p?b46YH
zZ&K1G@*H`fGhzj**a8G8dTyhw5R=GmO>|mx#HlQbPd?UJl2=;FE?<mYEic*gMod)n
z1IipiIuM{nqWuI44hUVy3<TR*P?cL?<3=YWl#`kQe5n|V{04v_&||EKQF<^y^z=zX
z*ObpiqQ{z#&ZgpM*0LX2Nzxd+k!>U30F2C(jOyhWPbX#_z@+iFYp{G^V}d;bph5&u
z_IW>*k8d5xR4BwMIQ>`2WzzgdtpPwTyw|^(3&<UkevF&8d1umKZK!onq(NK!e;-zn
z3LKvkvBJm%0i;+yycUO{RywSfVvJ_wFgSEKp}GLE?Htf@1Q}T+*Ym-@a0U@u6EvzY
zTD1T|0iw%B(Rc`dMgjE6i2?=~jTqW3WN<c)s!qf^q8S7n1arNS+LM3{aH7)?nk_10
zymA*_6b3vbX@$QD`aA1(opi%;8O5-SZrik#D_6?LYYG4$5kpS6Ffjq*1qHwladIa8
z1q3TM%#tEo3uJDE<kb(~*6Cp<^ux9Y?To&QjYu!JNC3nhCwV7$cRg*2vwsLf3~#*!
z*%MB=9?WP!K1bFPILhD##$G6;2UYnpW-`gc#22m%(bsQbJtNc|1|M-rd+@)Zbd{*y
zd*o!1_|z=ulD_0LkRjqS1cN9^Z}wOWyN)B{hJQPvY?XJ0IsTnLNOF`Au3kn)_ALI$
z82T91yLhMln68=JrZX#e)|wV?V?Rn4_w(tg<BZZ$XDq_g#C#ufmM9OhZ73F~+}*LK
zrT8Ze-NU<UM`z~uPc3{@uN5sxm>L^tm3J)dvM=p&vl~ns8lxvtH6u){VPR&LFblxb
zI^yKyM8c{b((j)?e=1>u3E$1O`!Zzc(aH!Adb3WTjje<w`w|;H;r)9#1SBIMj4Zso
zs$jjIe~Fo;K7M}DNOM_3y~xaAq|5S<(Gp@R$>y4}9n`4E-D@xh7_u!0?!^-+k*bq0
zvgi$g06=qQdin^CzL#KUfK<XtRC|#~Sy_dVZL=1&2R5~v&rRPX7brO0x;#r~77$hk
z1_(jvQiHhYr2w1W-@kRaA*NS`6jBd^Q{cF!3|Xl%Cbr;;Y7719XDW^&84Nf%06t3;
z{$CX4Z9oPuS>MnQ1-3vW%HL@CMRk4L<Kp$NLqt|PL@8>)ckv47O`X{I7-%JVsfY_3
zMH0^+xKS7==|dqK%pL!(R1pOid=^qTC|~<j(SgUk(8lq|*Mp&`SHR`Lm|0J}T~=_z
zKLpE%PArB|6UXEM<Uz(M16TyALRN5-e`el(fdxbmO^Djy(Ct`P^LO`;gs#Qon<;bN
zM=*OD1a@7ts)#xRvE~ZgBY<ZR<Q9b4q4dZ9MCP`=-CCT~LH|t=?{EBk<HY}bJ;br*
z-;Nq1d+e_3Py{`m*r|!q{5R=eJ))VInCzgVCqX0p&o}os>ummaP247MnCk!a$3yDq
zs)3NXj{fwt#wA8EwCFsu0vNEV?PJZOE=JzP7ZMT@W)b?B2izd^-#5?PwV7gn4=1Ec
zR)B_!6Z_1VV%BADGE*LU-)1YT2OHyw7#t&wdo?bJkh?cT<|WU>T%h~z&FX5AZ4bA}
zigaaGF!^-kcriK<CcwL)CL2Z7+|cL;p6?9|1P!o%K(p=pcq-n7p-E^_-LEBHCO~RT
z#_!~S^ls{rV_RVT+hL{D(8#Z*hu!t6EZy?0dVNAenXq4NEbRKV(r2LDCK~xSjFL}V
z>gfX?Z-aL#9c*ePjt!x+iGc_wxvy96%Sq(KXK23^z5!>ZP3%0~qA6Bq*@@2Q8pzrY
z9^XeX!?kr5iacuj+EX35#h}bzg2%~vsAjk`dv7z>i(BavQ(j2>-e%eMgfKEOZE=`l
zhl1@_M&S#xEcvbbF<Wbbp+;f#2w#|t&t32=ab!ByDHF-gmzI7cG>3yU_~B==da&{V
zXs<V?N*NHR&!Lc@N9#SSa&C_}F>{)ZFx5bh@GDcGJrQyfV|s#zmkl=cH>J{|FGWET
ze|zWNKvQ1WPwVFWnfQI{W<?l$Ux~?yCzuMi0uax;u9SVv(HPS&!E!#AUbMNbog6~!
zk#WIpD8VZub;MACT!RJh;GR9|Io9q_>LmhrB{Gq8>%OoOA)E4j5T1De>X7z5`L(pv
z1HG{kRM4<VHA*=zN)k?rmjbbnLQI1%=~-+uB+4j1T;LRD-Z832pjiD452r6dw!X3z
ze;TGfGiqB@Qc|3tT>5z>O*a_iNF9lzd)M3yuU+T5zS=lOqrQtN^K+a;rQXT&TXQ}P
zx1V|qC3_|M;2@K8HR)lw-nZ<_2Lp$m$NFg1fddPj4Otq6g@uH-LPPW9md3XI&AzC;
zNR&enI1(34^sEnWUFm+h{}3bN2Y5e7dNT)0FD5zI!9Ty~bwx0|N7P=GeF2WqK`g5@
zQrtj*Fsx#d_5DKT451iJI<_nCb6yqOK@N^hs8kW!SIfR+LN6Safa{#R!2)=NF2sp@
zdh_t>M_<H@N&0(|TS&;E&ws{0;cMiBa$&}*0|}=4_uJ1^h!Jq773O>x6M!&CoJtW$
z6R%I1ngp%50%oVZfaMV~h@pP8BR6{gL(KAugrdvuq*;Dt<!0yMY==nY#KXklMzrV^
zENSK@Hn>dJJ(B2=s87+z*v<2W%r96@y^f@ZO?H#r>S6#@b<ro4+9C9Y@q)Ke)%vtc
zIgvxIsb|tDizS3O0B>swve8=AFt4k~BC03G@@l?5u$Pfhijli>X8Ed7a3=yXai5Q>
zzjsb|bll3U{u)0e`#jig%F<76@2MNkwXv7wKgz-NtWZQkRyKXbDhqU)vN8cnVNQ=f
z%S+bwi$RnbfB)e~m1ihMp208VMP2w+(Z_VG`KbL3e*{sL&W!eO6+uaT8uu&&^9oLN
z6}ifOee4G1BW$U%26p<rzR~+zLeQ|;|E{U&N4lfp_x?=4kM~N7=Ma#jYfVwx7-84r
z$3(kla4^ZZ;;xp{yfRY@O6(7qumea2$;v&n7vw{3Xpi#qE5;frV<(1On#7KsMgUt&
z$8sM5OetP70EwZFuCCX>A1df>?)TP2Ju54Gbl?F^72~PbYOs7(k+Ck-05FUfCLX;#
zuSm%2ALzIAxVOLVoGmDRxj}c9lZF94^9^u7FP8XXUiv+(PNT7=^xIqP`1#2^C@g+9
zwWQpvD`yZNM}R{oAXg1A=P~`^jL2w_v-k$d9H!-75M*M)<$dYezaK=v`zW^%A@93p
z@S9?YR!hVt1DU|u=yv3q)9=J+QObp8?Lp0O9x4|ZL6JCDcuhQStRNO3gsOm7UWm+I
zK@QrH1#H>1XtMAL4O09A0tk}}DDO|0cvPJ*{rT}z*f()vSul2QbC9gcjykiS-)d=p
z4jq>5gg2*@58pO0ZA@&D)I$nS-T!j>7(a9X;AV#!T#Jw?3MW>!m4KKD!mwp-c?4E6
zdd;Bqo^QNNN+|o<WY#<v+PXChR1MeX5WX6c@OhV)r)mkA4AbVq{oGN@;VZ&9G=8cc
zQ_6F8U%z~zWY%U4<WKaGUYBUu8{0F#b904Tm8Yyt9G4H(n6=IRGI_S$c_{bEmvgN(
z>yIWMvzu62*BIG?#>Z8C)6!>o<c@4Weyv^BoYg=x-}hF8z=qZVeB*nMQvjE-FN~~$
z`0@~H?R%~po(Y96&O&|HU6I9)_RReUuI=c0>~)&^G`H9B3d)RgrVU?K?X{<58rA<E
z8Y=5i48PWx!iJ{Dkx;CTRa=;(T7T9j9+kFAgki=(G@g+qp)v3k{VV3c%B>h9-9F7@
zbPg10Sg8n@uB0(X!#O+P)H4UWsm*DIFJ!7`!IkF%M~DQQ_wonH^?_~^emhZrH#&{z
zoO24q?W&CrNBfGfUeObHURLh|l<qjJjeWUHrK(e)GM>KHw?1BQ4zpnnLU@URMHnqo
zj54if+s?<Rx&W3g+lOXK{=x$BdzZbOlnZxUyNU`a|KoQsdq#iu=P?Ya(tP`tQ1&6k
ze9B~|xnW_H23`VZnM1vWGP>B|kTO3z`N6tmCwPP*iuu@m_2`G8unI-}N3FG`hjeqT
z=I5Z0Wg&QMHn$PfvCC`Y)e5j>$S`751CCZPB_&kTV{M|Sd<y6U>_(Lr7k++p48Rkt
zoJcH_ldOQ{2t`V<v;<gAaR7!Wp@KoaQ#`X^hXEbu6%;lDatsFv*nZ?I9cjbuzqe|v
zkSXv+XTNH(V#+OyiUK3xAxQ8VD0cASHZw4kcs`-4N|D}`@4i<St<Ck>dw`XGIL?Fg
zRFKTh1@#ux=Rsqw<s=iFo+2LG1P(%yeTtv|dKDZ)KcTM*@07BHT^+C~k(aFa)}rON
zeb=rhp`oFZ^@msy0Q3Q`+gH9@iS_g0<UGnt<ibWy{6|7z#j^+Hl8Tzzu7v0|p?w@2
z&xYS02X6As+1N6+1SW*-`JS@3RKW-!j#x0tn04!KZx5blcvXu~w^F#`DrU!}nzm4j
zr$#4)`6JoWRh1oCkl00$5*H^PmOyBqvtF8PNPWq6$)g7L-WfQ93@GSm7*?&zx|H(>
z=Yr@mXQxEKv{-(doKch-RpHG_aD%!>>kT04Q1!!HhI611Lh#!TucdZBvuvs>_!^x6
zV&xD(GmkT$fIozt0Pu%Y*fuIVnnN#;(YXOMNqSGV6M-$&{45#YK>-qZuKD!sX-xR0
z7t+3a(#*R|FfU$)0aHMWVf0I+Iuu~v1ZfK08eJDtVfb}>Z8$BN9S<0suVSd#!-L3+
zfIq=+Iz(D4)tXH$_+8Q&77A>RQ1s?Q)tLOr*Ug{T?iCj0aFZ7^`Ywi54)*rs%&^ce
z>zqxqph}vULkE%YL)a9TIS?-vF5_}%tLt?}%biJ+5MUXZ+<bk~*rd7i(<h}1$}Po*
zn3y<W`i%;zK2wk8t3vi~(jN>l%h5w^$Tzc|_ZH$A)9Isb7(dQoTf8G+jCkU1-m+x|
ze9`snJO|;_Ni-qkkQCI`ZWu=jSOG_6r0ZqG4`Xbss<s2w{@ih=?8(;`(YvK(xfEuf
zX6{BWaCsM{`nM3`twHG<M&KP3FLp(j62Frl(78P<<u4A5fr5L|0U#nqBhdwsfgCW)
z%HiLR3#d=L6+daa5(6SS^Wd<gKYar`%@xefGj;b|T~}o@Yd8h?`(Fd!+ND5ZDbEl8
z)nYHwc+zC`y43`{jMwC9&L{wMM)ot>fM35H8VkJat4a6Z^ywXZx~oY|k2DD`u!7?g
zdd1xpwz1KbJGpM`1E5c?o=KxhdJtd&5qU?%swE2<7kRN2fiFEGJE3!NF4Y0h5Ao(i
zjF1Tb8n7Qg#b-qA9lTdp;DnNzn{1RC6N1stJ^0M^>g8k!ELU8)XTaEm3dE;uO&uEP
z`M5A6jw1{FTvWLeeYGR1-s}ypg7H+zxHphAGcWe#!jPA8&58fsuc=<1k@?{iamp|o
zv^>1HS2Yrfk~dMQSPw2)*hXAVX35H)Q^>GamgM;c_tYD};{vSu9LG1|Q=T(vids+c
z_3^Vf#^*~!`ztbxYNJF;E+;1~HiK}2_!0UcC#ShM?>4k);1i<uSmM(EW!^#zflGqv
zt7wBz1mlC5sb`ss?vmp;apEkqsCVm?fS@Wsw9|mXA~mzpHLL2-9jt77A$#O`r@m(N
zMYe#}y}~F|?_6pUPdtHqh#>v*Q+NR4GBU)eo}!**t8B(l;(j0*%A;jBIBQPc3SE)M
zC4HA0pbVUeO%k}DG^8aOph(^TJN>sOr^74Q4&5h-n!kUHH0StFZWz2Ey&(Doc;UsB
zK$gk^w=hv*VPTK&(q%ZE$l$BQI1Q^xP@;Q&L~g-jp#h8sH!|6bVqk69gnhrENH55+
zG?WQAU5Ls9^MQk`iM3AhCn1WMm>48lx6wRftMcb8x%a8bi~cpV8l^z!N5;nTfH4el
z!nAHVbX_IQ0(RFH3Yi+l`!#8oVzF5?yf)NQOc_5~7(mz?BYn%#QWwrPd}Va><TZzT
z>1&IN-8LZsfPxR~`PZoG(HzmK8*FMYs82$-<ulHWX>dYrvdIY>_;iYmg8hX2Rr|;A
zuivEselM70i09(6<;!04UVl~UsXju0<1L3BX4%jg3z4V_<7GLR8XRoBdLQZp>%oT5
zc?R!Vywf$g9Km5Dn;4G(z6zex>RmhIF30X9xyMA5G9Heo!F&xSf5%@=d%U=(B+CSS
zJiueWfUiN3aJA#{wxi(m9(1K8H3BT?*0I~+H>3jiw&7jw@>!OL?Xrz`+|oj8_GAU6
z<u0+UWS6GM1%r=AJaB?GFj&6J_7%k4=Pr8xvzjUf$96d@m)B+P>k3#pMm2HLv6$4X
zm=YtiZ%|z_jVyf{$uKlzxlP`8f%PIaGswSTuV75p55)_T+}VETKX5s8am$a6Q|nt<
zTh|({jee%7S>#1)dQ~#Dr<ja8F@fz&tXctV4I)!z6cnB}^T9R9R%ggC{psPb?@j=o
zg;NdFT?{0hkFfflsH9$ORc<W=%8hL0TGcv%TU2-M#K71UOCFJz$Lwb-fGpw=-*BwP
zUwum?N?xp5hSQ+UO5f)9_C0*X8VZttamejLX2bIGngSG=WQjN;fO6#y7i|Isls%F(
z<M8VZQM(@Z!pkCT<}@&|C!2lNZLnvQmOGVEcaK0h3ynm72nrjhz*l}wkC2h$V3xjO
zS0z<04%RS0kv^~9(r^EUrf>v`!z^1t!-~5AqAFl^4-c>s=*GPU|GdbaC!$E=Qjw(>
zw{q=!bY8Hyt#+;=_GKWXRbWsI_4?i7z?hTD_}X#|q3$BCWsjJ%s?Fblo{vuY&6)4p
zh&y`?yT<G7o9=*Mtf>iqP%^a*Sj;KA!QDc|eg(kOUJE*;Po@eZrubl}C4txBoQ%bM
z)%faX1a;NXDp}+d1{r>tHaYeciA*SW!i<uQ_&6Srk(H@NUr{{0hU#4_%*goVw$QD3
zb$t*?G0gXaZjS<R`Q)PY|0D-ycLD+!>J(R+QznNR#ZUtwOjXh>d|J%;^m=T5%J$QG
z_v$;3AjBempP6YE_3#S;<g!Y0<?-y^7OlqQNeRg`XcgaG9B81gRjs<)i!7MS@aGPs
z-GDe|g3R?m7$zf#F=A$YUtM@cHcq9%1&p7%QeJ#Aes*}o@X9`-TtvEle`??f0B*1{
zPCL%M-cEn69?>UJIi+(hr0!5Ni(kp_!2E*c^m|_<nK5H!VM*l!(#NZk9tJ)c5m`wx
zH-z^ZnOsY=oqc9*%4r+@PQp`m>>_*cobOm%&#QOOD$*nzI7=UfdBDy9_s_s(rgBin
zI(`)a&E{^w$MZf6&d}+u&ob~VRmmku?fP9@SB8VmR7a_98LABBJX#PsuOjdtF2G^E
zquCc@)%x5uY6Oowge8OdmBj_iI@}W$Kx<_BSNz1@Oxu%Yufn<>oJWhF96o?1f1@Mf
z<1u@**-VnTGi+|b*jGZtD~8ZdXBUA9eaVqONVu@S4vK-JSyk8vr%q1CE5GyI$q|bh
z35)3hN({2N!C@wi9GHrro4U*yyu}ONY{6&d|D2$na#F%uR!Wzl$4VxU0^+(d^hTh5
z14R(GI4;luRpnNcHg5F%7mr=k`IhxZ+CeCj4G^kdb%g2*I2&Q<<tIfduA!l=Q;jo}
z=;-N@K-HI^td9-`DFG#!p{6TBo4|*7{^U$ALow`o@_HD9&?jasLBT)W!AFby`SjKT
z@PT6HLj0PK?g(0H;l;wQnvHYJ@<4eZ5*WSjuk^5s`OdsUIh<oX<D08>_Uv8F_hU#Z
z>RAsm4tt+0e<{!F@+JeZup-^+I!Oh=0F6}6IJqS3ER_lG`S7+QOz(w}l*jnByP=T=
zZ<jxm<-BLvFvn{c7O<(&tNtreJBuVwa>i_=pBebUL2Q48g`G7weg?P$`ov$hBaPlY
zw3$HnsIGXGo`8l5cwbUtL|C$I4U3lo?}moo(mfp`hdM$dr_K1_qK4jeDB_G+zo-qc
zzMkqdj6!g4?=oowNXS?<2jq<uON5{}hbCLYBO$nx7ozD*T<}Tik2Q>goH(nE&dpdz
z>X&nC&fTK7)q3LL5oY+}@EoB3;y0XP{JIw}yZnr9sh-Mfe4b^f@MRatK@mJlz~-NB
zSzHKDEQVpJP0jpC0UzltCc3)!;b$~HSmnyioLC>D0s4Ntm$|2IqI$_b-x|fk3gt*B
zI+3E$*Y0PNUYvqxyPJ(j%YZRedqhtiR+s3}R72|^Yx`ALk=YH5Lfb%@?yF?X(hHXf
zEY+lE=JB}D<FyH;R)E(?9~ZVHg^F2ctLcxA&p<)E#~ChL^-k|Xz(rSHW*%;Up6WDP
zTTv?$jy+qQWe%H9*!6KOZ+63i7g7T#{8p6vM6_Go4tINU8$NuN8J<G+({cuxqGV@2
z?3Wb3?K68ipv$+qNb9uI8$({-{b~qxVW~$k=MKb!_TFCK!E~Z<axS$gie$j7D-gY|
zxpqkO*-NHpWY~pSZ5`q%vGTGt|K9%tVGN=xJ!HN;?3W~|7F(7ZAxJ8Xj*vd1Pp7AB
zVf@Hw#|1bh!(`86GWBb|*GR8onJwGGF2L!;1K($MeyWFE$RO8Y08JjQ$*C@!%EoSP
z#mA>txitOtm<~|o=Ex(}C9*73I9*Tn2v{!Lh=@&1WTdt{*~!;y<=)xcL^FJ4=2|1v
zF&vIRPi2`zcrE&dtoQ`1<{=7sWEa0sdGXzbG*l`LbQAry8Y!^=>(6fJ^BAFLLZ9?o
zvILW<&@I35r<xh+xY$HvyurHN$DObDNPT~%$j-d?V^ZRorp%l73>Cw#Vgkne`;Pni
z{az?r<(cNOdmHXtOH_c701h@%wL8s?xRxf+SB3X=@QnZb6s<pk4cdV<trEC-1<29!
z*S-BN7HwyflN@pI^)o|(KISw{Wy5{=tP%40qZ$e4m<9kvsAt+dBaNMk$a0#GY4Y$M
zA+2WS5?<)=g8+AUF>72NN_?f^tN$zBLYg#x+FA+=3RYp$Sr7b-<f%}(diC8&VR^Qi
z5MxYk?Hq?8r@#@EAfec5F`8MvsEWJ2g6U;5k=@!h%1TONzM2zp?aN!!XsVu5`W|zO
zwOX=I%Z`ayH2+K^*OSokUGLQIC(~x-T!MwhDLrVC$prF+Iuz2jV!Q}G_$qvqGiE=B
z0HY?@*Poyatl0D5$<|*$b^*I;4$aJBd>A}?wyLSAnH+qMIW=;*H7|inRTkPI8$w)i
ziH?qL{(ZM?-auRX6R`AduyL$cbGakmFl;{<$z<gQ@hv`z?T9-fa6S%wmF%kzn$Lw8
z(woPZ7Ea~0NDw%ec!vY0I`!U32wO75;QO?0vQb?aayl}|IET-w8;LpT$U=@Bdf`>;
z;9%VP%?Q`8=0JmoJ}R-Rl-zO9;6GZ=^P%X4C%F_qe>8}AkbZ`VO$~Ax{UEr}Pl3}P
zK^8!Vf(h+QgVt7SDk=zB{mMknt{$kN3Ec`TstX&J=#QY@=RiwQj-^QT-C^lBZcf4H
zNfv~isE}LYP6V_#j}>9ZXmUN7=Pf7(#AzLvFz2;aVdCYj_1znY$%`G(lF*d~8sM2!
zM``fqrAE?|{xZa0MAJy7T{l?u4xc*tO7sbxVg}%Y7_E2*i<hvY-129#ub}nJy=wrO
z8ThP`T0OucxUuk)7o<U6yUCc8-6CZH#Q*Ld^#)zo^Wp;3&D6d9U->m$nxEYQ>O3lX
zMn1lspf$ev@|6tE!PVy*Xk_%3{D9YL2|zyHo1=%|^7Av>>IonnfX8HnTY|;Tbth&r
zhDZrOdJNS)RrbT5R{g}F1xBIyFJSe7-EZlU!H}`UqAdNT<wWLEs;H_}v6d`<Omfm@
zSrCMXMpPS9<`d7a=d&7g1KE;jJy7z&A^#0d+yH~O*mzquZTj?0J~*jt9jGu@!Qn<$
z^%__nsN8E)@2(`1>t?40^04iJs)nY{!;^BDEKxgIF2e>dlyIaI#Nsqe1u0AZ&(C*T
zoMzp@mTGTpCEW3rxlfxGS?CmZGG0qR6~?rGKNao<3X+cjl{%5me!H}2cjec?tt79`
z8w*;YyeR;{ArY@)4HK{q1h;j<a3oU6GTMf|ap>8Z(P<u2q7ZRfm|=o==>YI9i5E=Q
z3#j8U3lBIe@%|#MTXcLZ;FAuLPbpMmDsXs7*GYUb)q+KFttM*QKtoFCvm7BL7YE>M
zBmL|d&}&b`r@qWnGw6+9YYbbDTGK#JGA{sBv=3m$0xl)wm1H*Us$O(I2zK2RZ817O
zeCs)-mH#-hBEcBV0C_--u-qQ-OO31wlkd4ltCD#Q^`qN9US4|S;UJcDZ`mFeu=@)V
zfp5}&gnL9JSm@Xhi2-p=tc{Vlb}gb*%^#15z<5ZCfMAqHt11Qf1KJ73Q4Y-Jyk9D=
z3j`4Y)qc+V{H^czxfl8Fs!28+kWR;_X1o$vMZ0;|mKS=Ia?o)DGORg}yn+Yt73Pib
zP7F88aU36PnsqNdr4M~o6#8SQu24X3(m^-J3!A&%qI0-HY%e*T@Q@2&U-Ozn$K?~1
zFv1Z;yM(DypJ70Yw&q*-L=nms(7pui`9H?o?E^3?M|NH8$M4;<X%l=|6v);(e;)GQ
zWw}5Yl5}qDFdyV1_K<k^z)%Bz^`HIw@Fvq!2o#1^=TWuvt%%UrZ$2CrVVcn*C452)
z<6E?78NQeA7kTLE8Gb*U$ssS#5Ig-<K`G%zz#Amc3`{sc@O$BO*xbAane>e9kK@2(
zE!0y$6$VdE7wkK*mr!ajF15(pghFlw3SxZH*t|b)@%(}v1{7dkXntd(Uk|5#$n4P-
ze_r@FU<>v)C%WP++>E`AW*Yp+!muWZng^9)_sbEqyjp7#mCvI;Ha!v^NS-{op<p23
z=PO{Cb~nRvKSbeX=F}rCWB6@Y$tH3aYBroEDKmf)eP~eT{`AEY4J0@j$O@)`_^s(J
zKjX)P0enYQ)4foPB;5G4?4()47cgnyH5vr)4<1Y+2IilUv3@X}x6OBFz}u=#V;K|k
zJ^k3rIG|et+yV|S&>L096apfA4Qp>ytXI{Pw&FRF$6cY5Uye49W!(jbc_WwK>DB`~
z6|-mA^GvLIDp!Ijo%`O-N+xIT!iMuK(fSv7$EVGF=gzDiDVqcePwY>?9jSlysd3v<
z5;dg}3>U?{ThF!b>R);AON%xfm2iy*n?{-*Oo#8{)L+LFr)8l3{i)I7NneT6Ei34k
z%aL4}M6P!r3Y+&<Gn0w$q!;@Mi|ujXoh=VmkI2HAK~2d^J%>4j=I<dbXEQ~KX<$2(
z)Kk;cHZ7bhKB(A7Z-2KLKRf3~;wn0{MIjtZf6*5Eoc$kFz^`Uqx-0)nx1c?!^Kbga
z{sh3<OG*4+|6~?4U#7eA*L5yk?3WBCw*L&IA-`hdDM9(q2O-Ji)aw=hk3SyPiU69o
zgP#7Gm)AHxe&vX)kPyOl!MAU#wX|AttTN2{K9Xs*_>~W!@9mDw!MOrE227`%-W6Nd
zcoFUo__EoGd+7@C&m0W_C{JOO5-_u1U`HtHVd{Jr$1_-uL9qNl=o`|wB>!;AB{h`?
zQI83;-gqy!;KakuSA>`3?pTcAQ^n$Z5p0!6jBR(Hwvl6sMc2DJUTZPdbcxGHpW!&q
zM~Hw>dtfYI8(jGDhV}2Y@15x&XV<ap>19Cjz*0V<6tR&IY7uVvX2|1(Pqm8^WJ*4!
zV$5H~_Bz~$Xcg_~<2FuTXQ3U2A5DaOFf|VBJXn|FtRKPBHV3_Xw>Ll3j`+B!b&XPb
zpVmwe)*ia}cktl=;6N($26nomz{+>m$<NJ_L1G{V6FLG6T(QN&@M7pr)VJmzhB){g
zeun#6f#O<fOn*u<_5+N2q3p#Bcw&}=6NQNM?)6TE`D9kb=D60zsEQB+b}~d5m^tDc
zLYXuQU06wyNyPmP$uzXIp10K1_Akr>r2v1hWaQ7^dM;i~C(st8e7PQh2G{Y?tr+p&
zA6#wX36H-`?m<J;fwgf9*44RMA`Aqlj!Y-PHDWx{bAY4?aPSAf^%ZPTBHVh_@iW`O
z4;3JOV?B7f_h1JKy^%IVtwczVbWYpMj8+)b+2LNE^SG%K-(enJW~l~#q&CP=yg$@N
zve&F#dlg6_^kNklP=;qkI*|y)I*2I5?E{~2r&1^GDH)cYGG<^U-~l}<jC7C{%7Eei
zAt^~0l?_f7KfEEKfb3Kf8gYE+?j8!Vzd!#i5A>NwaExL!WiDW*HMF!{$lnR)SWORO
z9v<R@7_i3gpE7cWT>*SYui)Y5W8m{A($58V28Uh;QbZC>#AX*8ylixyKY#9(n3#Ct
z<jIQ&X_abk06yDJ_D51`0Hf|UD`;#~qhjIU7_o|CO&A7~Yj^Av*;DYHcm6Y+N4r=%
zuu$cBL8c~NOAD5s20q8a>&OI=twD!n{9gx85LMGir*fRSDwWX^mkgPSGSGV#G*7_e
zV=gi$qd>3*i7djhvcQc#7%HtARN6^~aN?13mbq+=0rjHrMU22_XBM6qYH4Xb@nu#U
zw{)!>wqW^fL1`J8i|F}Wa`ru;_xOw+UOoy%))6qE^b8Fn;idE3a3LnH0JK$R9Ei%;
z<LIGBOQQ}^z~WPK1}D<*4C+o-Viiusa_G=GCIR%zZ`;@$5r#AWfg*F@Vo4oX2O?Cx
zRkA`Q$8+{i_2MXmiLp5|QnGD(_r^;@gK~#OF31`0r;tLJG(F4as71xZEYP(8KfKCh
z5;Wp3KmlG0*rhb@`UKr9dh`zwG{6M_#kRGlrx0lY#;JU(ox}}(0<&>whQ9(r2lssA
zjve2S$c&E;{`jyQyho5?zz#+Y^#E}Mr>cklfc;}E>&k2gJAZ$FD3IX%>&kt-_C{+i
z74C0$U3{R-0yrIH$~5LP!Tx~oRd7!^kb7l?HDOfGCkO(-Z$;R0b#+|@zR*^-%U+n2
zxCxT<b{(W2cM(~zDc)jpVJTO}s__hrq9u-LPzEraa6^88Fr*Zx?t?5lyLRNn?@U)0
z7u2z2Z2An=KT7>|U{}Jty%N|E+_$4%EP}~~(YULjcLkisjM#=c;WxfXT~kr+pat4G
z__1F<e_jSt0}Qr+IMb!s#tIy_+;86=0`exD7lam9s5q^yt%0B3gfAS%`k+&qS5mSb
z@*40x^WYj{`CWpa`bw{Pyhg@(Kz=vP%|F9U1H@o5b17h=YS;4PIR4<&`3Aq-_WL(v
z2rHpjvBcm`<Xp>cfJKbk_#?$+7dlHd);TbDGp0vhtr+uOp}v72+A};nJY}x`%@iO8
zMU-?D+{kdnvT7a5U}GvR_@h7yUJvpT!#n()JrscrTq1&Y<mj7r0`Rb8-wwaf#xb#A
zl;gk|MgY^U%Xna@j`?%wW>yd?JJczO#<|W^NM5lfBBbrm3Z!^=cu?0|LgNZ7CyX>e
zdzj;z#>he%{0Q)O9ewdR`RL{Nnd<7E!!D<sa`=;UM$ZG)CvhN#+6Zl>(_fwj^xecB
zMQgkgiX?>iq#h*0U@@<A#`D;vBsStGCn*n87}D!{VQOY33%!VQzeWxoDV$Lr;2}d4
z9bl*_GE{@+BqMGo=8b7%3rd4+#{>6-F4W{O!7xxQAXKQFGq`NREk`}RFx^f|reS!K
z;hZRT1Cr3SR{&pG#{bkA;93a&qRgT=9AJvphA<R^X+c#c5DUD9Z<HdWf!IpoAl(xu
zd=xDDu$;j8abcGio>>UE@m@#d?}87FqpSeH4sig)7>X+h%@D-ow$4L(6OQbjFfc$*
zDn|(d^aKVTO-XyV!sN$f0dxd~;PdP8W@2bd5T6_XDDbDc19Ihi@a%Dz1(%rV>*u33
zHpJin=vwCx9$)+V5-lvD`Vra(Y<|owrMM>k3}y@whjUzIFzy_Mye}5<i`TEk(SRBr
zA3p@N4D^i(Y%FBc2~C!#xJAk`!aSfHCp}D*_{)$Gr<k>rqOu^#2~MCY>IZP`{*2Ix
zuyg`1eGm@-ijFaG;Po;*Tn1;6+jWDMjB6tg6&ngEgc#slC<44<>jPnGOtvMo4SDGP
zl;DT-7~y5okn|SB>#axRmgK>}#5Zo=PIeIU#h4vL_>o%byR@l*rFX+(3&eLYkWRqq
z2w5-~F3HWy?Tp9`gA;XIu%}?9;BF1Hn9=r3fEN*{VWJQ~VSvD)BS#)0CF@ynnA?iD
zAOTM!&evepY&_ndNsDS6QSdlCiy^_9`Tat+GAkY<x9&kKOW6O!;0+L-<*Q>F>08OL
z9V!>6Ns_~jEF!FMGIR}Lb-sT6T9-Zo{#GE)%vY?9*W7$rRki0d=k<{Fl>CB%Ep&9Q
zFJ35yo$<!+CSv|7%7F7&Qzv3y*ers+52l0HU;?g@YKLXOwe$s-`}gIrM2?ObblFzB
zIl$x7Buj6v{V)~K9)cs+#|7IsNxeS4KTpp^JqNZ#g_9@I)PeRw9#aAahHB72*ob+|
z_wMCo1b!I_NaDBml=eR*PxU=N|2*-a#!8LCHA~g;IxS@WL1nItDt-KwVLIIsM4gnR
z!3?9%2()g9IdKglu~n@QOsP-6a^29S39^fo5u`Kz+W|+@=){CFGOOb@IWHB6CGuK1
zh)4N);Fd0j19Pn$^eeStFnJQEGD3D^s|T~y%Tz|thBR;?{9F3X9)^$);j{zBMCeYx
zMto!z-o|!WDcmPbK{8f7O`#@ELyb-Sqa4#bVDs;vKc7uYOX~?SNR5tu+XYZso$5Q1
ziTJ|zfq{WC7F{?)n1u-eKfL~Cul`tl&obk0*+|C~U<!T&$Mi(b;=C-=A|l4NAPaTy
z_>&AH%MukNQRvA~MPYOB^&tExI1}wbk|CNSYvquL*RtM%2!gK<rr5GnQ8+#ul5l=7
z3zKoDen2w>muB{4+eaoYU@t3UK8G?0dnZfk>s8}vhEeTo3xbn#HI9L}#NzU?xcaUn
z4QBxX0p(t`X&lbmf?xs~{9*_kneR0<%Jud2hVd&!gqC9XH^Lia6<)bwg{}_|yuhi9
zkkLh%wPcg=I#d{_pfWQv6N!M^_Z#rXMQ4nZQiOt{LU0I!qasmOMghv~HEBnTA`4(v
z$_j+Eo@GZXConNVIm^}xJ-Qe&T|Cx`OT!5_v>LAktsqIeeyl`4RR?N59KLBM`VaR&
zVnCSy(Wrm$jZK+Gj-v+Aq~eSvuXV4sW?h<>Jlz$M!5Z4Vf=bS0(q2Q3Q%=<7<fs#v
z6y|JhZG91rM<x7y3(;PdRF3U`CI^ZDP@u<k5FJ9KYN$_OOhZR(^a184YF>yS%@H!S
z4W4Nr2IgU#5T{oY)+D5!1|Lg+Z^06M3!K_ZK~KZ(_m~7wZcl0qe7%K%Vg1}9<rBKF
zHO!=dg6c&p-p|yUIO3FhhIzAI^1yEF?a)fMWCouEJ6|Nc6u(!|>51G??U+@aI$<6W
zax_wOoX{)R2_5zow<pdjzCO!|Z6j(XGMR*ejMoMu(NW+)Ib?xsAIzNff||SnLl`+=
zEA5h%nMpyGMol4I3JStsg`qXJSTWM*X9IjY6NYdW)Nb(bO?i14$`Vd{&H;#S>)Mil
zb>~KpJ=3(ks9{9G=?OYwtjy(xcCSqUBhbPs!-ux4oS&Qffxh*t#37@&a9Y7>;rP(D
zKF4OxcRdLoP>}?T<1#J!YT1ZRCu_U;pZcW|CSL|df|#iTuYMF<yhU@q^%rp}7J-!V
z`ndKo@a%(-Sob$&&><Ve8i_86H>AQ@coy09C*X~+Rai?*E?}~Y*%j9UADl^z?f~Z9
zJly<vO2XXy1bL7|iGfy2gF$Bu0j81P;sg`Y<u!b`zfMP8eK(GIdg~X}#(kvO7Zq-k
zgL3sDaL2zTR1i)&1b6H*kIx3DZn$?!AKNoRIeI%FPhIR9D9XHnr_xvd8J2#*o@;gn
zi3r9yE?=B&StQy+*bEUrE+Wo@4ulLR$9DBdO-;29KxjGv_zgk+^<Z_MkF-H-Pdf_s
z8v!?qD<O+ri{M8LxUt7KZr=QvR%nj+@{+Mj7-rCOQoAG$niQISdF+oPmv2Gc@e8C^
z3UW|=6@A+~Ptif_2{pXy8+lLu>AcRnXPQmFly9|>JP=SHS+ZFjSg6Gn?uVJW1o&s2
zpe5ZmKVLXVmww)QN+8jsp}ziCXJ<Z;B@&tQ!@}rYKb4fQHI4VLP{$@6_f-iOkfEoC
zFCPwwugJ;4?^8%M)j&ZC-^t7~B`KWgZ#28c-aJD!O*#S4QRQI~03V~)bQfE_*fKw5
z^;ecC-hzx0HRb_`4viaA+|h0_NGnBl>U`-~B4Cbq2h|1yeIR}_Aok4w3QNv2N1WcY
zW5+{K8DwT~Hc!tkf$O-<$;Za9NlOrTRI=^UEa$h;V-^l0II5W7a2!)9$iV2>3G>Q{
zx~F{{i3X<9_k3?P2PkA(fbSBH#E!tks4wK8%IC>mk67ZHi7ezG3yWm<FL8Wxcydx7
zXecK7VMLD)k7ivAK~hG?$KliP0`IyES;flp9=i)c@-u|01b>`|S_ilsW?#8tJ`n~P
zLff_th$)h|*Ib61k)V`X^7{r!LHEg-HDh#H_9emxGA@F`BPgM&Ci&rKaF~%oo_E{v
zcD>h*vt#)jS!aHoc&zy!E&vyfo&eJT9G7KK{+&3=IiR91oO0L8WzP$@-ZjTD8xEtf
z&Y?-{gRO;K(?Df`i9?DetyOXe67}^?C28VKe`F6%T**=8&~4#t)Yoq;IeOy{pP88^
z)9WTo+k{N61b4U|a}NP=uGCzqg~cuj5<tp#i6L9trZPKyKFjW0)X8MH50q>~IZWgn
zkYmds(zj0h^}x$Ry{=+M5;i`z?R&^UP}(IV$3gKz!3YLh0Y)T4J@tv<*3UU90JtoR
z3c(;CsUHwr2wW;Lt%P_$U(#~Gfyp**dv5hxQXi2hj0KA{_-#s^>_r4XqQi-ci}Ud)
z4AbtOxMpTnh9xK=Df#)!IU(Im;L<*O_Dr*kZKNduiVm>o384W`K`Bb9C)T*&sa^76
zGo)YRw(hEzFK>o*4~!0eBpP^v9(4<PVPIN}n2_d!{`*ZhH>1Nuq87LYM^)ZCBjN6j
z2~Kl95u7%{%g=9MH@X2p7vO_$Umk5q1CmPE5J*^%N!<aP9;NuYW+BA-)|ys7L$iE1
z3HQ|UBBvGG;-SmPRB%WD+g11T{WXvfCt^SYrczkGLay8OvA<4W$fsh(uS&l&e5g~^
z^me2V`IuBa9avCA$^HN|RhaXlwjhq+WUv}YvRBN^0!>pV`)&NXvVffsV1^7-g##)K
ziO5NV#P;RMvc`pe(iL6YhK~dy8EIAlu<|*5+dTc#V>j5rt`N`J_5wyfD0pnk%IPbT
zBH%W$Tr;bf3yyk-f_x%@*IaTQq2Y)W1f}&=yzc<tlPThdX!zs(ew4PmPu(EaaK`bt
z9V*b;m15gK?Nf8R^pYgjDh-qaci{bjSjahT3p2`77y{IaNC$DUltCjd3`OL6;AwTX
zzm6?5;Vu&1-FnB7=UYHmpxJr^?#`+%v5MSTNnBICIeu<b)6|#+=m*@X$352jRhaVA
zFh-BCZ6sx~8igv$Ky`bN`|j)d7bGuO=#)G}e+<Vbkj;(g!eGp=Psd44mQ~`ZqUsH*
zzGzty8vq>39o1;4&cH8R5{tm2mnz;#iVG%p;<%N_I;&`$F~33!)Jg6FEBmIgH$F)3
zm)`)@EK3iG>bHyZE`xUZjhW~trQ{FHG9g(ZLxf?RqSd(hczEC@n~bieii%CZM;vE5
z=+IWoLpFsy;caa%Y%?1ufQQtUgJzNss5uv7RL)@HH)=r6Bh#zmUA|GF?0Dat09=Nc
zjxcQKKp8-eBYJc()L!XN`{AenuuO*LJ$&>CBJ;Im%<F8|&=74deh<vbeW=ecwzqAb
zcq(94k|$skoXgyJ{Y9Z>8|YI@CzO%<1E1h^oI6DZ1fx^So4$v;tPt#~>c&Zca{z$E
zZ@F9<Y1z1G(-|cEm|1%TJTo)yCB_1{R*|44A==CGu)mv$^|y|Q&LK3!_h{c%!D40I
zauFmwgbM=K=xcrv+mFam%J*7aCnvkO`T5u=3Ypl42m<O~HT{ttn(iwvC0&yz>U*|u
zc%7gi7LH%(4WtM1BR<+=m=f2Cjaa{vR%+~1xf^fTO&m~`8Qj#=)XU<V_l4HLy9pD5
zik=-5OZaJOyN>s;52kw20zD#tD$a`QRe^Aj0HBp@CdJ7wM1LF4l^#UO)|&epBk%EI
zWi}KtC0y;!CP^N6*A~OAXNYiK@uD^L=Y?=mCxgp~lJ~5ckACK}>Oj(7euF0LKy8rA
zn_DF*33hB`FmC8p6e~J-oJ+`xiBV=r*&9*-YLSReT6ZWV5GxH{&6q*+2SQWQ@-NG)
zqm*f6Wo40D<|7(gjyhpn4uv>W+tgI)LmCFlImOwt3hf}oHgvs0i-))n>+0&JJGD4W
zF;Jl1IDmxN%v@k(!H&>q)l$e{Q;4Z_c{TNrS7FtbQ~Uf<_I%|Q$AO>)An*}tU2H$G
zx^}q9JjOYPM623x<t+9vaTueZ$=N*qcGKcA&}G8WuEYaz1_6YlXn*j!M+UlIp4xHG
zEhf|N*%QY(IfwdQ_5Bqp2==O@FC8nGxXn<xa(X=sr*`4Qxr(?zfDdFGZ9P4Yz#+u4
zAs7tdMS%L}I5#3TpmG~^pm9rf0NU%7c$QuYgt37{oOC?#m#{_U1VCU#zk<YkbT0@Z
zhSjoZ=T1rN0T^5yAkzr3r?8Z%;?oTs$1B7LFb_b3h(CyNJ<(<XRU>ncOHXaU(hQb(
zaTu9MBGL${sT|lTlF1eL0*_V%fCO=Bf{hF@&qprU{d+HjCOCkAbNi!=LQZ?;>eT~S
zzobzKUbq~3@4EwNL_8$)2|pTnGS2pJ^b|1cY9n}0z&+6V%>&&Pxa0}rAIFhg7r}q*
z^#EV?;O#OgFy+W}bG!rWINZrYhC@C=XEGaiRq6Tp!;`3=Kao5J*q0#2alqfC9%?Ou
zgu=D~ZhU-21KHPsg%F=K<hV;`92y@<NxyLuj;ep$2n-3X0dGsRu+!j)Oom^;FC+x~
z1fN%C-M69eIYJSd8(vPpQ54N39{!h+$b5rO2r-@|fTPU9vxdgGrDJ9&gFGEfZU#?|
z2m#RR5$Dmu;!kFaJ}mpD7(M#eK&|oOD>VhKqauGwQchrL81snUwE<c`gg1iF+9&nx
zxK92$<Q7C5OOi06+=iVhKtmA5q>(2j3!TM{_BSQVgF4gJvG9neO%94V5AaKIKT6hV
zRT76?0EQGqwMe<Q%Yc3~>@oKJ5L%@J3qBasmrm$)Pjt7nEsdf8zI8+`C1x_U7#DR4
z+&XP-ZDe^h?-}TZ9-x8q{KbnxeC&JF>m_-B%Fz-U)(H;-_;y}yTt5ZCH#ho?dE`(R
zl0!&n=<VY-AUtBpkn9PMtO`@u;Wd%0(a9w3>lORxUGrMT=8ptOL9ZtgADI=KkQCDB
zGQ7m=|LKj6uf`^_8bykLNgLrThz7*%fLN#C><a~K7xPBYAqrb68c32dC=YONx@~92
zDvTMEy<Pwt_jC*m#iLRP268}V#IIeqPKoUM=ypW2mq@&cHyasm562lfG^Z3{+E-pk
zhcRN5nk2(2<AGTK`${-EUdG18CZwcXB;#;#94MizlXL!qG~Q}d$Z<|(J^hJue>)bi
zf~SXJV4~c&M|HF2$3{lvosm#fHOA@JgGmgG5k>`={Cn`=#CLe8C?dmDLg`2_ncKH-
zM*@bA0H4aLvzXqb32Q0V>-+Rvh+g;YvKS9vJuIZEFfeS-!N6ZFEj^tDU`z*COVsOj
z1fd7ug$ebeBO`@Lj=c4^V4Cpop&ags0vh2gC`+PH<n#jnCR4chx{k54%TV>ylCgwq
z<!}(zNC;VWKV7B-nED^crZhpl1QN{ehE<5dVH9rc&j5R5ipuCn$b%<BuA0nLOEzoK
zWDs&xkC#DCPKJ3Q6O)>UtzJW!w2jS2O@?&>Lz2uj$=5IGUB4bphSniz4E9nWW3}`w
zX#fC*FZ053gcX%IkLUiSWZ>*v4nP#veMLDqi?7vHoQ{@$b6Pb%6O54v92vx3%P6p(
z^utI>;5`ND4a<=u;^cB>=QR)xCW!7TdXRLroO@exoIIwq<}y$f9WA<EmzO8vr!_Wb
zY+&FWJBmlE4Cl0jmpsvZXfe7+p>0E89#vJ<NHk$0fv_7y8;>|*pj_GLpMf7Af=@b~
zA6aK$&~>zkz<;HDNYile)!-78QO+-oflE$S#(P!IWG%4oV4k2g^ElX?M&b|lgOtI#
z5=<<4ffgYXP8+qRLw|L&Z31!RJ!x%iH7-kH`6Bccp9|wU9b-+s&StWj9e6X6{gi_u
z8s?>dxxKt<4%K9=A{j(NOIs(gC^R|9AJ;kWxqQ0sTC|Lv?1OLJ=T&*=>u;9yiyx`~
z{>?+_;kob42mKC3>xN!1?|XjJeKb+5l6_}j%ii%UhH$IRp|g&Hd%AAM4^55D+qbk>
z@pU-}9uyMVi$%DSvNPr-;V|5v>%wiPhH)>M>}}<G_Te`q)6piyC(dTO9+`1lT()!h
zQ?DLCaOi36t#i#j3jES$W|R@F+IzreKIi9C!={N~$UqH`8&w~{c|Oi`PI`$WAzm_q
z!LKnI8LxId@p-Bxs;#}hy1JT$%aLzCxBJqo_b!u)s)ms9*T8@~%1EMgKfQ+*??K8-
zLfz)z_|UuI+0vzwe|`%+*zEitx&Ypz`YIRozJJjil6Npt{xQs09r@Mq$Dw^a^z=ee
zM&#4}{W?b6*_rk)Yed=p_jiv_iU0FE;%w*F2><(UI3y{X|M^<siTtvs|GljHPcOUu
zU)SYDIdsbQ_v)?F_K>AM%tXRy_BOXGFXd9ad67(@Aj4XF!FO?*u~M%WdAGBvoxM{Q
zKX}|oYWH84_%w=`hN)au5x!GafE4&#^8(fmT<4D#`|`NDr1SZ{_PzQ(fB`1R2b5AQ
z!e!)I$1XS^^2!9w;i<gW8`Fn_0TuP1Kjl?5qDWDRQ4Vgs<X+DGi+XJ}48kTg$rtF*
z)<S1*wbR$HcM95-Z;6LimRH|Lwh9Q~T=PXq$+f85;H40mJIdzA6P)HlPiRnV*I@(*
zj4mi7WW0&P$+N%T7*BXRf=78}WmK)o_e!+myZibUljdJqYdTu*=B@`A_w>!14gCBW
zy&L|r-h3)qSW@yJ)tjz1UW=BOH%+nzDb8Bgis1cFae^tm{0}v?%|bstI#PFlAZUZL
zvt#J@mzS~^E{K4t0f*R-&3)HFj+<IUeRDRjIg3yUxJ+u-f&RJzJsx}dA_w~$!?{%I
z?vL+gX4;2dW%`}-xL}8l(7U#AH@Cg=1IK<%wm1dj;3nV|@Zp>N(>GzNsB}w3s08;g
zC_4DwgD2ayk(bsmcszyaXb|$lkCrnha56>n`|rG0dSS+PqJIr9FYj(X`;X}&e=V;N
z1)O%992ZU_`+Vl(gOETQL$f*;i6Co|gQKJ1P9CHCNl9E#{!x|q@^sZg?ep%sH+$>Y
zjJTBKUDQLA+#laC<#_EhD|_6}KbR%xrpjq>wKeDrX@}OmcHwY&3l0O|=HbbaPaz7=
zAKkA$JjWtssYutlmo(WZ&PcbHQBq9n6#{<ES%M>+kHR3D-%SB;weZx|!ra_ZJWHs^
zlkZrv=$SKAZ7jog{(dOFF=#+POt+Q}dhq55hJYQv(JBlW^U}VVRTR`1$GN!985yzQ
z2P;zKmEfK*FDWT`V+SWSa?(xR5(Vue$8VJHqvTeFgh+ms`Q}8=NQovxAIXGNbTf2*
zy%UAm*H^3tLd#T0C2W7pzkC}Vyqv+rL?zYC6F5kFW4;ey23X@gcolZoGAUUb#`DOE
zazW(Bp25~xPeBEHe_5a>1hzB-L}q`sPjcsSG}$-u-|BZpB6U3K5^b-M(Y32aj~}tK
z+l>BRb7`=V111o=T-deL_}SIp3kNk<=2K4FO&$f>PP5#>AtEXwH$+?H6dzw3dVb3~
zWL)l(z4xlbZ_~DI4>Ser&Wfq+`eSR`T_3y(gzXDC$F+KTdU5hEr<7xo&RZMIi@L&d
zZDTL}&a=v*E8V4}j;GvL@vs*#XiPQ918`d&<mdoPkh4chk$9=7s-Dr1UO%yJJ;$ed
zdLJJjBFe<N(GH5I)nH)wG1>dzKL4#1lOqE^>_FVW?&d&fCD9SvjC?xZojbw%lz!B1
z4e(Lub$8oceQCe5A{q>reSNzzp$GlMOTWexDt9=e4}#TncGdOk*9?u$;!W9#OE^Hp
z3<(KY?G*q$6(STkPkV9egdu(ndK<Ulh0A*Kc3}BrW)|f~b|;Sh%3B47Bwc6DJhc@!
z_qFSWZ|@~@gCQU<C$~@TBscdq@EM8J*wV@}C^mHE1sWO~8=HNstTYr@ORvylWYje<
zAPL&GJyXw@g630Sebh^2%vfmZC3*QzquW@I=hY}*)3CNa*_`b_!^p^p6|<a@kdW}y
z+q-~~La$>pHOMffI;#eFMkT|V4a~*+wTzDQ?3Z)S+><Pj{pPgIol4~^hIG37`|n0M
zE`0g&&V%PYzQ9`g19-ncX;dQSCQTYi5{|xj-tEsTz7-W69RDpL<#U>TYN+X6*=PqP
zd)@m#Vu6C+%F0}yJfYdiZR+L(`ok4|{@3%9yC`_g^zR-z6`vmfC=f1?2{>NR!~kH#
zW_)<R_2AD(NKg;_DHD1N^eL$+ratj)XskO%2*K!K4;+%2$2^OGoVE%YY=AV2143)e
zhio-H3yNg3Ta~u<YJP{EC~8+w#Kk3-QIJp)ELjzeexSBCbJHB1=^4Rb?mx`3iY^}M
zg#z<|a&|W6O1R*{p(ncwWemj`QY|UVn|Cl|TD|FKP*|7?8eLasQ%o?iTBJKt5nL6#
zzN9#1y3%M6m^R3Grz|b4R<_}3pKi`{paC3sWO(LB(rr$_ozDUTJwan}v%ZmwLgxl1
z1w)9w7pSq==cbIO&!4k!SQo}mWZNw4Chjm0Q~v7jcN-tKgv0ICUN7dSFJFEQiqCnB
zr}-8!4!1iGP7v}rXO@SAhwGp;A`2J6Fz%(1YFFr?4Z&guxwy90GVsBccNOvg?|Ahy
zC@@K`al!kT2x{kjU_zMLi_7F^&WzzMKLS(==sSPjZfc{&n0{Kj?!|G{ndy;(#(cN(
z7Z;|4Nb~;UKw~@FUZ-x(o$4$JAhF~M01`^J{d9L->M;~jWU<HfjiJzL&Cy@~E?y&V
zX7~nx!Lyeo*Mb*CNZokO0kgtOeo$w>IcezYA4YcGnr6vtZf@S1VYDrkeWG2AB5Z5R
z2U47SqCx3C9v(VMdCYAK#7{aj(L!rA<m8-F&6-=!iLb_ZJa?4RSI}7#nsi<r`QQkZ
z$^#rhA58i#Qk=nFrZ_{q@#$8FWPg3uE}VZI&-dqgZ`S}_O$6PR8xT&KZmdFCPqyJn
z;~x|z0|P)3v}nP6D=hrv;=YRm9h*3hK>^#KOJ#2;&V0(JD!!qA6%-o!06BQDOn}p_
z$0*Xap<h^lpnE}z;_K&k5OW(q5tNoHM_r>ZH8MG%25s&}UtiycnKtr45;g|Ac8w3!
z(GFU4?mK;88Rg@O3e@;3Hann_CWkdhz@RG-;~<48Lmi!a0LYe6Ncqobub`;-tFssQ
zfggczTKU%RK2AaHdLxi?72Sb@RC9j6`H?~biE{4L*H@o;cM9%TPj5`ID=8^aDAIo1
zB0nM(Ns@W-&!5lr_4E{Cb4dE~r>w`1padPkI79CpcJf)E2z*BDOY0n>>^pYM^NoOQ
z>+}a5^<~3D(>7e!n|B{P@3!%y0lF1;&L2LEJ>0t4IW)lg`F$6c16X;aX#{oQ$B*av
zPvz$3QXHc%L9G~g!G<o@2|GMaRLr8?nGi;el(xpHXS-#rS#zx+Mm2JFYlo`pVEa9|
zA|dXro@0T6=6<%*+?}Uem^Q-s65VeCYJn4(mkYyc_I=K*{P`7|+@+;&V&+7q?aXEX
zeD94~_E4~4FLpZol|6rRdX(^db6rsw7Qn}i^o0{1%W_K^?2s7ZA!~ndUFG%i<;X4j
z(_p4}{K|9Le=U!i0Cy3lEiYL=7(=Dw{O;X>&ayz-t~Fcs9XvR;#l!7s1JY;ZZbV|D
zj&p@0a1a(6g`%OMfz*dMnW4A8EYbB1uLU*on)ka-t6ZTzNZ2<mAC}`F`stSTjE-Fx
z+9nXkw*Kfs=gtEHm-VP0lx>ebS5UB79Ap3hML`*Bnr7v?;r(FwTgl0ro#;a>%+aH_
zxP2$>JA9sqOYwU^wt4FBzfM_M`Hodf$|x&eOT1=wjM7mgU*|7g+-}?0(AHk374bMC
zD)KE@MIdH=1xoieEG(Zf8L-s;r*o{OVAWw{qgH-`oPxq<M7sNN!EijGv<~?3bmX&O
z1yHpZYgx`6d&1z9voK!~uX1T<ZqEL9SC`27^VI3-=_IU<j*jjdN#@oO4LIevn{vFl
zP$HtHdHQC1YpN}`jO&9n$u<mzW@U_vA3sV<-GKjBru{@z@rs}$W#5r#?SnP?&GrJC
zYTURx6&p5uL=Oc+UCKcwic9Xu%DDB+*O%qSUKqcvM{X9~ZMc*r&isPbRHZ}5gq4|j
z&D^x~!w7Uvo<4haCnQ{!7?;kO^{XIjbOvBa5mpmh1+s+;TIspB3IUw9d1gCQL>tJG
z6B*FW=HB&s+J<Y;3COft+=JAKaOK_zIBftWQ*zn!z{5C(kzpub4#&ButEU&8R66>9
zXnXT$F59+k_(VuFt2C-ahGb5u5J{oIlzB?#sFcX8A|z5MijaAhP|8%M$ULQzDMJ(@
zA=A4Zb=|}BtZ#kmTkBol^T)I9`@Syxe&=}%`@ZelaEdl#iFMnn^+koL3%-6AFCT{0
zI10w?cc@njcDgSZIQ<Mx1=Z0M%toQ$z7Ec_PrEG6P2FB5Y1|J=;llKrQ&Q`rWEn3H
zykANFW6ITW+n1K7oA8FI$fmS7Vwi(ff{2YBcc@*7x6;3~j^Dk1Un5a-xILelg5~>4
zHs&8XnCV{+3)?U`@-$C;XkcJ5f3i;QZlE{x{*2Vwp%!}TOK<=EjmbFI>cqJ71Pa<E
z_wHpwrBZ}>o5CjJCy|lTUlw&<cbli7M0vXMEQSW8Pn=i=6n(9>*{Rod_mVp8k_WQx
zhJ<*bj>e9b-l{xh-39=GhGJQ|R0!w?RKjm?N?#soS^7OSwLVdG=fKym!RuF(*Hlrv
z$F^kgVL3T<i*O)kx+fWBc;8fgd^od!-%y;RAprA1Ht{oOB9oIz!SeHaYkCO=SJt<%
zueL%>MdW#2y1Wher9fAiitdl>R)nY@Ol@4ihR80J8jf7dLtJt-^^ZhE3V-UaDaF?O
z=LPu^B?e$pEQ>=p_yZq2s7STKc#3t+nu{>v-+_5c_T|diuseY~X5R|FMcc;CzTP%2
zc2zAHXa_wu;JNmv^g!-_CNPe2PK!F#Y=YUU+oYuLDiu4$eQ9|HSzX}N&{nyX2n5Ns
zo<XBCP?u?(*z2o3xQ55l?)2S;RaPB2*H){`t-gYjkQlJT)y)fB?G2*lMeTgBb}7X?
zm@j)FCB4u&=!z6<GPjF}tVTn&>Zqoob2)kTDy%x~DiJ=>y=|p;qc-lbIgG0}6C6IA
zxZU9RPz9Hje~tY*0vwEQvp9L^N7c0lzw$Kc^g4-Ie0ZiX#Iaxv_8_?wH8Y*})jZzM
zSe|KI)qZ5citg5!wk7~FqtCO`?=y_(K<`#yk{)Y>nCvk3zPWsKu}RMgd|^AcrtRrf
z5jJBSn}=TJSSt$T+VR0unwh#Y6V|CvTfv=W^EKTk<os$QIp34%`dgIqSiqDl0uR<t
z=+jH5b>tzku#|i`b}zE1zX!XIPgmm@azQ8xQyb1S0fj!+;F-WF1D#umiOIyUSIXH<
zN*k>elswxwIXxTAY8Kya#jn)-n%7$pkqkJ{V|R@l);SIh<-xOD!5_@%9|w152FGxJ
z)t!y#{^2knWk|ti&>d)W7jW3U>sMXF%8GI;bbY>ZwQMR4!8lYnSdaV5AGKSue0h1a
zCv$zi{Bm;pq?-BBP)`e7-SUz80PMgE=pYk|fO<?q?n&j7>9{GnfF&#r@OH=fC*3At
zHyDZ5scHQUrs<XBk#QPnG!F(D+z8safB!~|59k8QWs81Kyw7*at+VQhJ5FNkqpmXs
zqM@hEo1P0%Tf1sDM@hg32b{ed;I&9=X{|?DL{K7eo%o)KqJMD#R4%%?HO}QGc)pKc
zb=M`>VxqL?M_r7Cs?uX2I_xNVYSWCwvQLxdvlG<|pXX*c(+WXuaX6sw(^#mH9c@z~
zByJP*97DTFsRLd~YOg*{<-Yxs$OAYpIPK%lAs3yWH}jr@XR!@O9#?5z?`cn8Nj_&u
z_74&g5O$a06unXTdG15=OJ$xP7==aQzK6_|{y@mCL%49TwlqBD|3W+!=}*v3+S_#W
zNy^h16f8gBJhcLiT#r`^56<Chbj8I~(L~JEAG7<`3TWRzgN`W6leE3LHf`7NO`-sm
zw0=}LuD2QQuM|^J&+fKo7OTg`#}n@_+ZHU%RK!c52^7IkKv^JA7ZjX3e}b~iHb3xj
z9d;$j6Qd#_qrVdpQij59(cLv5MUryqQhPfvl4V;0(9v1pQVn$y;a|0OEtlwL)M*=_
z{roXi&k+@(k5iI+(gC2Hp{1qo8qQZ;PtzCH2t3)C{9x+7SL};@`%H$*kt~5`e(&Yc
zttu)ii&|hrx)vw0EC!&ST#Kjm<udzOZ>8gTk~$uu&x;EWZ=A)c4h$;&u1&4ft*1|g
zV`5`}^i^;4ka`LN;sYTUi798v$&u}xD5_YGOFx~qN^b3j0_<n3Qs!IvBpJEKYdWh^
zs{Mq{%}a2Z1>!C~*O9Z7s&m64DME8*Xl{6_TT?=={IuHndbdjoM_&_4){PZoeB=cJ
zHF%+cF*3CkU7JlCuYo}tj2f3=;$2r-=A{$&B>5J^u^jyTILlKyMO_I7BV<>}`b<!~
z9IY8i1SMKgs;)mkstz$1INGCvV>$b3AKrxGI2yebNO#v!HSJkMe{ljVady;Y?{n<<
z@E3@9ahLux5{;!P=q8Enp8@&P<I81>6AvaZUs!{r8D@8Zp7LN59He_)zp>?5xBD`!
z<|4ZxqAJ9w<y|OY?1F>CWWmFOIo1YTNhJV{UmWB5WLiURV2rv}u44~CXV6*y<|YYH
z2dlU8c0D77w?wS5QCJANd=O<}7ox?DzW8(DW5pVmhtGM^z`d+%s&j|DVfgdQn;5Fe
z9MRrT8zEIb?=rI$b(`oL%Zo=Q25Bg?$4lX>KVWWU3`9w$7xi`L_wP#~_q_bQfxajF
zQDk)VH9!v;2)*Fu;W0AFl#*1B{oES&QokxB$}C+U{o`RA)|nq_m#@~-twR~IcmICb
zjv@+NuVUf|r>W_UZ>@H9my|-dgq@+8*-Z>Y41@9yGJv~<z571jY8}LcQRqZ~y%m3~
zVy0tQb{G~-3I_!`2J4?NGA!G0<B7a~B(MQCQS%kRC@%;Jufl}&HKNkJeC~0nfBoS-
zmID6=Ml8+JLG|R(;r9JNRJMt5peo%Y4Yx)Z+|_Q@xDSyJ>R^_ftT&@8IqC+itQwsV
zSk&=3zpkaeD=*K=!r~A>P6+(UZQ0qIf`j=dLwEiGU}Ly&Ja{aGiOE#m3YY;`$3Ha7
zlqqH5%BFD)PNOjGBo_koXBT0`di2<_V#I=?wQ!$HP}bE#qTIK(m+M~Ie<!XwzeYL&
zATl?%uqeLxO%nshLmC?Q`O@(#6RcV@(C}@Na(%>?j$Rf+{nyrg@3iYT-@Ehp_A2*|
zuTUL5M=BOZU6tVnlff!7M~WQKr;DgUHSGsT_T<3e;Pwl1j*HR^mY1>NUUrXMx=|2y
zjg1awZ)EfgWM;2JCxu0MomA>Km$8?@9@6vInrB9IA-lA$jrZ4VLs6xzt<5HJjt#a8
zo^|(j{&?Z1r2#9-6^?_fxWrao8D!n9_sScm*&$h3VtG!)z+qzN^#^~UY7+WTOd}t#
zi99bhiOc3bOwttFew$4{e@<k=VBLXV#|s#-xw+*Tn5{h!bGqUqGVdyY&kolMj$eJp
zcW`kQH#V*YdR+#BePn$6O*^aBGpLpgOo!A}Wib~Oa`}DPW-P>@)Hj%K8wbaARU;#!
z&PNH?C1}dRrS!lZ*G?xUVZ=5C3kKccLzP;^W8e*4!(@p>7d!yc#byiyvt$zr*vFFj
zO-wQc@;Htp9|w$jFHX_VUIH4O4AQM2|Ca`Regzxr4ZIOslZlCml^45cQ83&Z9q-Sy
zuiXMd${~4q#?`vIg2;WaiN<tsJ<K{^<4b@jMzYx#{hdin$BLeNzrw=cW=zsz%%0%V
zxCHzIR_#{E#h>vT&(3rMg+s%^gpJ3m7E#-8>;<o%=Xt01o|ZY<=8%#SXV?+-s`&S@
zW@2)jbMCVrW+azqCFCSXrS#`q%qNgW%C~bi!iuO@RYi>~)FapaXF{@4MN@MzIxMA(
ztmI+nJv)5<YfQ0{%&DK}y->GjyK#qCI*a;y*8cg$D!1y{{_|@vXc1K3$FMb<`e(95
z{(8*XCqn-WP!fw-r)CG2d8~+ujlcD3Y5U6<u7r2_%yn~$mR?EUp3S_;``7O-U=Juy
z9$v9HvFZM)Jo{tEPqhUxwim{5MgH^1{(i9-8FQBZ%i+sE59sgb{2#ykzh*=+4{*gs
z0oU*R;lQ@f_H1IW?WIdXNZ&!LE{j9)%bfG?pG?ST!m)P)w;gJ4BU4ibVEWFpQ_2a>
zs2V_=B?9QtY3GW{O3=2~g8d-t>RNDW8Wq74ur1J;UJD9h1zw2@L*+|fpYJBVX*=CQ
z*UbQ&7y`=QpuQ`7sjGGPDg&exo~YPlV2}w?fG04#GXKrDP*5-hoA`BsuuI%io0^+R
zQ3-acF{G2Y7Sq&HzO=?CB`w(!pwU-VTzu$@<9g$PhFE5Y16$w(;MsEPxBgrzF1Tyu
zJy9^j+g^S;xFp$bB<6J3fvdtmaaSVz1N=KR`@LeM)*#0R&{8%CvYIz9`}0D0it(`o
zkv?Qw1%CTx<1#-Yax7+BeN~E#U$8~NU|0&`1-Q%%6ych}R(KHxjito5Q}Xe0>L>;`
zm_uEF&jIRSLr?|@QUxttXJ4NerXFD2xhK1U?}pSW`p329cecqAunM;}A=VHtRWPlY
z(cK32aEWQy*z5+AaS@13z!cr=UyON=vEXPeuB}~;Du@jX7FJePBKN@9jRU6}6zLw2
zm}!7<u>~dV^|Z9zkmfD1u&{u*-VhBKL|UTr82*p=hR=pU?uUv;y*tlsewq=n4Ia4j
zD<*6{JYxdQkNaYH!bo#Rp3BT0jL^Akm82GN6b7NSW{7+(fSGaNAfloTaq4!g=>cZS
zMj;SwUT<Qy5-+zD(A3V1Zy*z<;Mtr*t<}u>M~1$bM-l2=x}rzn;aaF~=!$?)ZByF~
zBVF`2LMZe?K!w9u@xb5zMC>oI%LAuQYid^T3+&o;79g7`^#G|Dxz6B<>7FZ%jf)ee
zqoZSR2me{5jLqpsH~qb!ptH`BkHMi>(sU4PV+MEY_B{0%hhx>^nDCqiK{^Kqbze`<
z8!7_;v|8Su)d*EqwTn17aBDPkAtq*UC-eVLPqC~;sCQ5|o0ynftimm$Rp>g$M)i%1
z2&lY@xiT|5Eg~W=Zua~4?_CEE>cv=NfmEVedVuFoR|IZ4M}~gW<X|^Ymk02j7ZDRP
z0f3%@^&ZWxyC!;fQ6U>HP`pg0SK!yMP<N0ksog&?z^yWe8sJ)P?qtB?IVlqel2IcU
z7Cj|KY#4@_T;ce8bTrEY*#U%92689*03FL0sN>#Xcmx5^ji@N2*k7?Q8N}^F9Age_
z3DBv**a{Dnk_OEw_mL0-o;E^?3?jFn;D*ECI*R!DQ?SxB8wCa2Z=?7!&`MS(d~9u9
zM`fUdQ7AobnFd&YJF0Tr8%cL<O;+}<C#xGwzOJ4gI{Y3FI{?s(TOHEzcaaOn07O{^
ztL{ZFKyD2D0Z-<iSWJO(wF%$-{qu`Yj`GW{H)h*^Wfxw01{l4os=R;e>pen&hl)M}
zF0`1|&wUjiq@BIB5}_*Kdo2pVi$4Q8@~Km&R%q`e0cOsvzp~$A_O}i5blX<%Yu9K$
z<=HWVFE=aXLG1f)++jZho$VT4GYbnQMa8O-53PXkUwkRg1rdj(Q!zTO0xTcJ)MRrd
zDLXqmst9|Yh61#+ir32Ae6g6A7{q7W!6@kN{u!K+lVC3EpJYC|tx}{8=^QAulS=G3
z+lL^yg<-Hj=4|ey#5Pok9{3UvHTuCc%J<oKyrqjcE&#*acRcBX?--LXln!%#444o5
z4#_9Zv+kJg0#B)HaL^Y_aWd4w${~VU_<!T8QWVTJ7;%b{88h0HrIA`UguH<Cfb7Fy
zX=!;UDCma2|1$WMjbhp{;+t+RsELJpR6&89vfu&9!c$H91$GfmM_7VP?1f?!sUW0<
z1byc+FmiZg%GjBmJ!=TNO>@k?(eiBDxswqqAFR4d<?s+QDC`rS0?$SU?-EEW!GQz)
zJCLe{mxWtFMpKgwXYu{T2ViFFa3G>+iX3zV_(Gd32>uEi8yopz5P&`qvs(K!D~sO0
zAKQ#rc!JZ$hyos3|Kh4DW(@OCo+OO2clk!F!z5(X;Q%DDItj}lP{e-*1yAkwZ)i$i
zgO}^*=t#aiZY`n>ghrGpXlAsF4j%-1nNi4vfRijDPbfH84%{+@P0#jj-%g984l`O$
zTv*SZJ*xxF5eKB2#Inw+_p89CCist^9}Bt$0>H!kCUbW6Nm7TX_+oXD`6PlwFpxQd
zU&tGD#~Ei2tTkD191#GFdY+l^xzNp^dk=Q}6wZnJ+&jLQDWb8%JwUi4+`PQxwC}aK
zcmMuPRM)afN`8*&EB)bsh>wk_34g;C(oti-3mFzsX`BXfYG3jtumTDDprxfxTd#F-
z^9PzxH6LJr!W5_f3oK|xQwWMs|Gt5dR~w*@P#l(4urH@8!h$McGsfG#4F#aEoSYmN
zKYum&FIwj2+#0p$&DK-DFj)`*#W}~MYV972h8<knwiW;U`4dBc4H~C|gs8c>x%E^F
zTw8NUegyoV!456Mrq#suZo>ut{3+pvQkuARw7^8tLPvmWaDxsem;~#*7aVK?eRXH0
znW-u8AQAHGaS62HqR_(R3!T<Rd`uR$qG0ImO({*FXQ2fJlGxmvbTZG-<JqnKI)z)O
zjVLlGQwxi?<l|qyENbkRiAf-XJAw!wd_0fy<^kSYR(7@tK7=s<EpiE+tGKg`g#sU7
z)m{Va1Y=4QI{#28;h9-j&f>G1;+{96Fl_x^d9XkF(S5vPQ<%D=z}7^Ml{<^I=~v#9
ziAQ>=XREiP3%M|Yf`T3-CGqM&^ot-#@$rcX6LMf|XyrI~Fr3B^9NsfHjo`}nP%qyx
z7#*VT9Q=S;sE+_VH0JTh@f(Hkg>pda&q0tMN_gSZg|066p)$+?RTK9j)T<b``-8i8
zS3QKO=SS#1#wRDuHcMPKB*A;*l?N9{?Kdq(=0mamG*hXSel(ho09%TC<>H5s@42rE
z`CRrdvB*t9J_Tl?g^Zru1Ay{r452GD!--A%5fuL0ABhW6tTC6Ej$HY#<Zo>59Z3Ts
z7ht_QJP!e@8H5nu06z8FXq-A#gw?g4>ihb2S}*hN2M!5|Fd9Sbf>jnTL8)uOI3O-M
z`ZY$|RcyxJ(5Z0x(`f3-btK>4rS%$X`<AzNXUN;i%F{D!?m<)veWe10dLfv`5SClC
zRZ?#A@6a=^r!ZsQXY<0#ivjCq4fjJ##z}59&dT>V=IL8-swkST=<?;F&^%Wo5+pPv
zWUb<{V=FbG-rL9Eo^AaxMCS_9wHe$YH+hO<%3}<m80rVORONz%c9n(N?=Tu!SMdGY
zH#TbbO0^)vyLavwS`mYro<a!R>54$upsx%Gq96q@@wvXQA2<lerDXHiXvxUXiaXf+
zD8A6x(4dL4^u;2RtB3aq_z98n1;7v%xG@gZUE<>RLA&%$Oi0K;{e7|baK0p1bjCOh
zzd`OsYtq=*7z$4Oy}&>tXl@Ph^MQfI`SlEpjLboW(4f%)j~puJ-*~`5kZ$Xp3unJ6
z%6x()EOfC?RW3pkB1E0TGN(1c5#2__F|h6Mtl=T`f$MQ85gL#!uI_;m!yb5c#Kp&#
z!_)V{ty_wr$BIyuW#U(hprKp4YSs2<1wa^Bs38QpkQfOkM2~y~%%iHIVdmxK#X{kK
zF#5APjvuG6K%udY6LNBL8ooJ;aykP=NJ!ps5N3v&lh@64p1462Es&jPaQo=$)=%aE
zthi9cdzVV7aGL0FTd2mzc#jOnMLB366IYpVQrXX*L6Q7GnmK<vysM0;_e7%uG6hwL
zys8RxDp~-)?!OO(qkC>aG*c03dwg@0Se$9%d-j-8&_0f^882`>zy-jowg<-?jUhhw
zCt+DxSuO#AT5!lq^$T73M<p(Fx))`x<9t94t|Ldv>^Hlata8UJb=8NH=)>1jA)%pf
z*^JL&eX&sEqoXT$b(#h<wlhg9*C+-?_xv0iI|CB983hNH@<o;2wk{y?R#GTa@|a3T
ze@|HLzUTstM%nwq)|<8c%<o>6ye~`@=ijyKxe1B`?EPmJs-gWdO?v@cl~ICtf+3F|
z8w|hT!V9ZNA2Q>jZdKLR_Pd^q*ki5jV**ZC@t5A-(1?gFB%9U81n8LN0@8cgl}Yn0
zmIIb>3z(@4?lXb+nk2=<a0FnW4WYIWw)^fT?chuSpJ5!B!l?gqFpM1W4Q$$tLK{a9
zljx#EZ|n;$6l!(jw=;b?sat$s^YhI2oWUg))#hy_Lwo0lb80CD+CT8kFm(^u$QZL1
z;}*)GOq8z{&mj|S9ti9`I2pKi?PAr01Y5w?87DX0sL7^RcVSb|4bjFF%uO-vE#nuk
z5Hv=IgaH!hNGyIr7}0eBY^5k`>+dSY+#7n*y>oK*pn|?pIp^IYkQNh~h1WzmuOFhF
zvNw&o%fJTRu8Zfdd(admjzi280D{pjDDyo%Juh6od>Btg@l($B)_C)lG<-T)MaAUT
z7g2F6_B+nB`0>ups=X^PhPZI$N*Nx19c)I(RSm%|iia<iJ{xXA$pFKJ>p=0g^71Z0
zeL!GL3HxCdPR>}B+&_3(A~&8b`IBgOOVL4gq8G@J+`5&*;*rJ8cM$=v9<^^ve@r`X
z+JA)w1$j9Jg^`hw6gp=xgbi`B*EKbvRvD|^yF*4n43|5x@PS3r5(t*KNA=)*w*jAE
z(c@Zq_);KzQtNi$-7qe+c78%EoA#cH`HoJR8O1XZ0KloDGo;XIa#_*+A7jX@puw3F
zrIk!x<OdK-NQsTx2CIgE>%nGzU3e@I6w~9tOg7oRQGseX6B-&yZ56b0ya)mk<qr5)
zcxW2M6Y(m7wFbgmneb_D;G!e&BqUQvD;$Y0(|9(kgI_xZ0t&1uN}5xF>Y>~Y+Hl?1
z*SDxp@%M_KPy7p?cOOe^U=6={*S|2D{WGtVK2-!t0}!l5r+%?qjn;jPePaaeM{G+o
z>Ix4xjV}jiiU{@zn)+5eRJ<)R;zjq#fQt(7E3s_@8E4hLW%#1H=+A+szQ&*!+^ZX;
zio&ss!}vPNp38Q2e5`*OrTLFI=WdeUh$oBxqa7j$A+D;Z+G2oM4<zh(5sK!IIV*0!
z=yxu-aive5T#lmh(^F`*S3@iH8ZLWD!-k7_-oTPG1L{B6Z_%P5TBBXdMz?h;e*^}$
zWZXxR7~$i%1)aGRzy6pN>9ySjI(m6|QMh5iNnsU?ud2eiy^WiDF_z4A<U$`EX~vbt
z=&oY66r}p=*y}(+(a{@pJx^Z3Nm&Wut5UJG^Xfsf+`JlL_?+EP77$nxaLUe%wC$;k
z$r1(U&C2-o^`UpeThYB$juHrNOivnwXr+dy<Tx=Wpem}X*J$4l3)7C0z?&fG4~jf<
zOH0z$)YQ~~bkM8av1VuG1QMO5hKE!)xVX3wtPkBE8I_<}2ZJ>LJ_*|~EyIHg_LT)F
z1jHCPh8)g*gg5$O%+!TClq6T-)zX8~%x<gnw?+Y*X9+?+uA@b1J#!M+DtWn2&oV*v
zxu}wH(IQywv*T*)2|=~A1}y=wpo1AW-_ZWWH_gRjfCLuMDK)jj_>NFYQ50Akcr6U%
zWkPu{>(0TMeH^rFlx9oucU7Ehg3I5(XIq#ZUyj*GJ>;_<40Kr0<ya@h{p0mGi7A5(
zhS*Xouw=cxy%R!0RANkV{Fb9Iib_q@>zl*m&mZ|I+}zxexIr)ulr=Xu_c39Yas~Qh
zY6y^*4Qb>UB~sqM^z(rn;J6CXEW|m&!ore)woq7S?fg+xIi*-sn4RB(`%E{08kx4i
z=-~i4Jr70IrirExAGFQeq15HuY*4?^8H><o2&tPzm>534XzbA|r+4!fs4z>>NjwkH
zCTJN7254(9f}<vS{<Cf*!R9bKOiW#VYeu<TfEVx0!@(D|thVQ1!ndm=KMy_u0RU9N
zO*F>*7;2qz^prP#DF+Uz{_NVeXU|%!F_}Vrx2aJY5M51NGuC*b_C}pY*i~#BHjn`}
zI7Cy!dw~nqL(Vm9H}>TiMun#aA~^rtU8HA2EkPCR_)cw=ltgUTu<m+sad9+No#4*&
z!9Tbkg|S|wE>3A;;Z3<?m!i>G3Tpv4*sP|WcbS0T<xaL7isfr?i~?=u0(9{JzKo&b
z)OEpH!!?7m$TIQZDV&T<kgEWVB;6e}++KiPi1QV8HqncsE%G8Xj8ub)^sku4XhQut
zIRbxiItrR2FOYTVrW!I{zkQo1(@UsBomBxd=xER*_HHnz4Dt+4|6EUuYQYTzVViX1
zWKMJH^l6fVbP={qxR|kvy{&XfNboDyxeXW=01@T;;Sd}YBnTB(<zvAyMMWj05^Qgy
zbLVtTU%;~<5?vh_zol5!<PD>4SyhE@n0D%J#%}#1L4b>-J~-otSg<H#hZo`#T2?MT
z7R+t6Y6Tu9njFmxX~V1f`dh5v21Hi`tg?*1wZ7g6?Kdr5h9f+rDO!<u;a?^Un}CHg
zxF1Jpt)pLmOd8#2E!z1|bT*`w(p<M;1FvY*;rbm|YO8i_+jc1O70?R0B8=op_;sND
zqbmY#$op&rBQ8O_*ef8NqpZ271fM&Vgp`DGSg3F?xZ4eX(u#D3SA3A@kE8W@q!%q?
zBW>S0;~#C_BErI&019{yFnM$Ldu9rV2#-4)Fog=7pQKFp{BZaNq~epXa(VzjNFR|w
zQ!}$Hpq+wP!8Wu(Pt#7;OQ4Xpal-#=!4Nab;|6ZgP@oBoEiGwko6rg?L02{4YT}oP
zTd3}wBUy?9Cs6bmHu&Auhb+-d){BKF_Y+E>=Z?1|@6Ak4Gkp^f6AMPbY6kgB07yt9
zitYLyhsk>sPm3xc#R+;@i^z~my(p_mmq@C4l<Z3S`Vo)|MQny_>Slzj%0XYQ?Yo9z
zqoS0%Y{91}(|D9nD-GiN$6SFc>#G071vt+>M9hsF8qVM)L0fkl0;q=wYF^yb4%&?E
z&>lg-si=M!U?4}GWlFf}s+RBA5;#%Izpx7t!-E@Xj!k5`=MEy%T5diX)KrxK*fp_5
z-h*M<1`0Rnh)L_J@aTIpY_4bw0biWMhWQ39%Qt)tB?NcyaB~Y{qvhc4>ihQX9Rw2Z
ziDfYrm!4irk|_vDt839#<AoE2?XaoJ!Qlb2o#Y-22*9eU0{^ZJd+5beIVLXT#Kong
zyj#0Vv*^mV_r*j}kG4(~Slk0i^gSrfLWQUlv=ishWADNaL45sn8bd(pU6mohhN~u~
zU<??Fj`ZH$yJvvBzeD)9(62eBqVCps+^M@@6+zFy@DzhB+x}O7D(@=L-5#JaB+rmg
z^Tu$Di{Q=0rKVPMzClZcp3eyTMM+op(Q(N6zkhmhZ~FId;@;2Tjy^m|rnUbg`Esrm
zxRu06Tl^8Z^^;z6*1{S=dr6gz;^u*e2hBq?7^u-#>tvZOo!l0SmaR9!jo^o~;_$<1
z+!gobKI{7RZ!wNb-p?Y#YXT#Rcg4j~d2bR{g4oI&6ilH*TUB0NeS7`JjUt$=qPi|8
zyih#M-D<y3CY(k0q;%wn7n}_rTUE2dVFIGpQdGjBD6vR(^9-tGHe)|1<b*Lui%U#=
zi?d!%ff6pmG=VJ(DK=0{kZ=0+>(@+lA%A*XTREmUNX;@JFazo8gAjg_{AIMOWmt4K
z22Y$kX~w*EmmvU(Gax|t)%O3{XP7F4ptBC<<15}V_(P=ZJ=*oqygUhjQf46GnE0t5
zrxnBm1_+6NVsf$w9zO9U#vr8)o`67MtgZE;F6q~gpD{Blr!wGdvZR@~GBdb8l*|`y
z?MLzduD(9->)@a;n3YKWDkBX%xP*%;eZr`-ewl#a1!4XamzGvTdU(_Z1P5YfVoJQc
z$xrAC;L;XcHgQj$ydnP>PbdpkmiSkULBWW9bn(xdxu5Sk7lwNK7o45ap-nk_7EESC
zSZz0NX4tl!4}$~zm9yYerVpR8T(rg5ApX;~7_a@vZU8z+a)5S0IGXn+Q5jPa0i14x
z@L7U+Xc8wwo+PPVwsFGS>ov&4U7ek$+M`x%zl$q|RDeK|t^;ScyD95^6j+T1O0(DP
z?VG#1Md=yBX=GeiVjw^qU2$`-fXOnJgH1aKvNRMb<&=m{f+orwO+?9!cw+T-qZQkq
zf}&&ydjayvgQHO|;GL@3vxzZ4>NPEg>H8sd^)=WGN3h$;><SG<C(y0E5dGs|fp+)7
z;Ieyv$EEn#Vl3pxJmG1#1uw7+cLls4eK5ZP@6QM#M$KpDE3xpB_a7$D@z0Mn3I1=9
z*}vwXF$_4an>W*LOHg{b^6xlv&&kURcf%riKYp9bSY+sUq%0sQ?4Lhgaq@7-NJYBu
zu%1jof@(X>zdsZHjO!!zJq%k-+WuR-w(0*Dm6}xXzolpbzJFqY{(j3@HK~I%55Pf%
znKzJ%jT5(TM<qqm8gp<wJKwPdtTtjK?RMp;Ixi|+z)}u#zr}Ku{t<}*3=%tX`cn7S
zKffy`?&`w)!L|IlFpbncq$*3dbg)&vb+;aA62QG>4tAUFrU%=_sQxLZXj5}kdi+h?
zR9koLN_jnlv#|(60i*VUgQ8-hk+F$jPfL*5r;N!Ev^5^J0^_~HM%Rk=0G4NyJjVtl
z91+|gkjDM`b-H1$9p0|jj;2lk_H7d1;{}xvir6LRcMgh-9-K5bG2T(=k4^fz;WaJ9
zL^u@GEnlGaC_X=7`OKny!-l~@BA7!(PMli;e@!Af!MJ3!i%4!PlakCor>utv0|cYg
zPu6_4QmBm1Knrr}w2Dm6k=uuTzss*&2;EQi1sfwHj5C%2UN>*d;TIJZy%QYjGd1Nv
zcm_xVE`=A4pi7zG<MrCXFV<|l)hj89zJw8wJ4NBh-g{-%K}`NXkCW?|lG5Q^$k^xS
zoWRqLyr0V7(Ac=racl?I@kibWVD}J+Arek&t6QL&pv9qc`}zG=Yr1|H5Siz;)57pS
z61?*1x$xv$wl@RHPgX!{$HBn?v6Bp}rqEM(HctI;2n)Kg5B2FZ`i8B1`(KvifxUYJ
z4ED(Bi5W)f+ow#DMQ{^x4dBsG2^{}){Ake+<MIo?-mQPM-))PjscFa6zN6N}Ht?0(
zf_{wkVHbviN!Ju5zb~zK9zJA75H;C#C>cy}1@0=zh>|6vA9A>;#vhlxa(XQooJ(H*
zA;?pxId<DW)=1YacBi2unVpBOT&Zhmka2cCKpb(<a^H|YEUm9!(Dc3PB|F5>K1As?
zF}Q}!xVLOPYxMncU#^23vYj1F_QR1_yky+4*2w{;0rVj7(V7tYKCs3fbTyipwwzqf
z<IF9!LHF*_p{fha)#>SH#E_rJ2=R`kwhI3}3y!KoF<k!k&I$<N8T^gQZL4sZ67PRP
z;m5ToH=fSqOO`8o(Dykm`^kJEH~>_FbsIOLBtPIfMymJN;qj6mJ)}p1Y4HW#U4Ape
zfnX>vYiS`?X*MM#k+4}dDT}}i1m_}0?;FNX-aZ~Up>RhTet0efy!zsg22KwNJP5(o
z_RW}Gl@-j*@Yxj{{eDq+$0A^A@FkVUoAuwa-wQ}&(93Y{+M>zN0l~KeXE*$Bd18Rd
z>Z_nlxo6#54Hv3w#~+GcFYlqDFmtQ_T)71I>2sI<WtjLVv>Tw2+@+mUf-&Gqs2U*V
zupOE`on8hvy3WBqFEnfSpSxV0G=3evABNROfI?(kY~<2>c8+Pynl+!?Mk7`FZyFrg
zuA(ND$~=lkh|w-xtx1(2JW->*uv1_(xkJ28C|S(wcCDyLm^U@`a`f^da7vs~gr672
zf5+9=R3o+<fCFS<JJk#T27O~#Dsxa)76QJ=KoBPWCE!D)#6T~}4*cQG;dSdESu=tl
z>C<mA(kH|<WKfzXB+DwQqOH180vqi`gY}Kgo;PnYQjx|tmqBs2(r&`i$9N>1g~<b_
z2>(NI_&hDicl>q3KWER+iJx^@^2<SBV%~fbZ3*rTI%|IZ0nqBC7v`ToyhA~Q0l6>H
zyqV}<zuQgdNB?g^JKcl`+{MP&x`KOzuR-9eonuaeD*mkVy`R;!weqMVB{X%&#~s9x
zVr1ig^-}+KPENVVbIK$z$+Kgy*}UZ$Pm`K@eMu8*va<_~^V<eEG%nQ+3nSqiWdR#;
zLrN|#)u<mj@%LVX^Gd^_WyWQOP2vW7@q+3nhSt3VW=%WQFsB400!SaN&vFV&>E^K~
zM+nL7qxkvCPiD&EZwVPr^Z=%>MJa7r&;UYo)@f^O(9?hOquavLg>0)AEg}LDQp8sb
zfhfCzq&GojMs;FP)Z230G8Z2Nq4H}A7v}ea<9w>?J2@!&T~p5CpisuF=P*diKTe$J
z4QXExJNC-3OJ+Cpr%6dkz$lNHnsUjUc*LFMj}cfWDk*S1+@;5sQ}{b9K*PRu<ffN|
zIIVfXF4f;_RNq~?cg<P`7neEPfUjo1dGy>LAGmrAiFf8!R&UzcUc7rsGT?Eyz@L5#
zX^&B>F^rnb<^d=IYw$WemXYz`0ruLt56@Q?@M&~^e<;3KvoIogCy+-hBf~RiXf?#e
z6A}}<pznEKV7(C(O+uQ(Oc7gX)hfz8vPc>m)(FVqr`u#6o2HisWzurorSfwOOCgK)
zM+;#beTx|o6`_++MKyXaGb&rxq462jD%b?~VnOSMn369sC9R9hF{eVENu-0jrX?7<
zb?xnc4t-pgnQL=VTnXE5pz)<RSjJrdixI-tm0mb61C**hQ#D?oTv<jX`+(d>2PdZ<
zSzF=0>c`7L?a=k}yMYdOxZ~x`2M@(x+=O54vYdR!C=8##huU#@;^zhEjZ(%YN`tHJ
zuf!Kaw@gF*e-Y~W$0~+M$JywSBfo9mzERCrY_O654d@NtS|(oVz`#H!&cXL+EC8}*
z6uxKVP)^(bx8$N0<$ZAF6XIc_49}fo>8zT(+eyc$z`wvFz7Cc9DK!<|OOMkF{MUa*
zR~BU_XMCl+q$Eg79^7OitwW4)B#`J&V|}Kkdu@N3`+NlJik3oqSDEs#$ra`t74P01
zHaxWgvr@iS;R!e!Km9oyB^0bICi%<QY|2UicBYq<XggV1f#^NdO0R9}HiWW-qLP1o
z_iQ2LaY6zG^B#910XRJ-u;z50nR}prRvUQWn)CQB1^7dPiiEqyigf~R_c0KD1J1*u
zd2r1R046DZc5Jo7%+!S{fUth3lZ^1D9`p<^gBB|$F79=8^BOEpIpk7FpE-wodz`3V
zGmrK`2*%Xe@kJE}U)m8lrk@k@`z5zPNM4RP*K>oNlp);U%+H>=h9Yh^jro6z@a$+@
zXMUyD39aR9g3fO1odI+WCF5qboxQ!p3kK=N6a{(Rc5(5^g>UWlnyjs@MLz@-?#Erz
zJjLqN%H8r_TDinmlOrthjOedbif4|U%iMPQno#B$wG{4Ov<XvhHaH1n8XZ!L@m?}M
zziqz&->v8~OcAx?Q#(GtxHMVw?cH;yU!O099yD79cV6Y0Q+e=_UcP1aX|;Gn+cVL*
z=MG)8GjHBBTP#7N{;fSXAfxUSLa5|kjP=s6AypUi8E8h5Oei|rzO%>?BHCEYba4!v
z#jH;h<HVxo<mT3|zrW`~=F%hLA_3LVV0DD!Kw6)R$9<o`!mmv}kdcAxmd0Re)s%1u
zXDaO7I&p2mt)OsXX7}GZm`s9*eH+@qUtXSFU%!T$#(3{W&D@Ws0s;&0>3@I73$xHu
zR#vvWpe!r0eY7J;`R3R42<`<ujwtc2+b)d{#O-&Fj=j7wWlL}Uz9=u-Dpc6cGHh%4
zjKf1GO5g3EYl5WjLL33Q;4^1)p&G<AhtS3eyR^931mnjw0|#badp2KDF|gO_a-8K>
zbsy<?$>^?Ed~JP_fne|L=PpC1!>kjpU!Xv<61DlnCZ`O2FCB?$$6-_duKG4kds}$n
zPXuJ&UJAs61f34D8Xq4`ynRzX>~KJ;b}oh|nl9~NMImWGm9P`DvU+}{>PW~>nV=3*
z4I8ja3|ezWh25<p;vncJh?;Op&C_qsOT2aaRX|NdgoSNT5U<uMcb(kxbQd5<BJvvq
zMm)RbMu7@#NM6d(^|zl(nq&rU`TSX1_7aU_%#Q2rDaLjQIp056%;}9AAuLfvn>`P=
z`H`;Io>zZ;3zZ0;vhJ||dkE+^N6Rsfy?=kIx}N03KQ*u4-WRioN58-6?Y+z6W82+4
zXtn9qHmE)^Jx-lwWQc2v<GF<~uKfaLOHVNvgf-_|gUBKTH(_ex4*nJ~nm4PAr2Wx*
z?BMPC{dA4HMFy=$JX-Yv*pmXf3;BTfx^LQNJMby(n8&E4<x|n`G3ROnYp!uv?iUiW
z)H?lQi&W)e>FUd^7{;+Qw>C8fi)ys4n3mxQ{5-TU&s-E5G)n^p*qH{aEe8O`v~IA}
zP0<tzpO~MYIge(j0tcVnJjWwk0*5L^yK`f$29lxzSGsS|bjD-=AYmNRiP&P#HCBZ1
zZ){G|aO*iRDPF&3ZjWOud)Ow&yj3ed!mJQ>JM(j&rpGBn;p{oyK1;vE5Kib|cQ%~k
z8?C&#uDUuLbOBL2u|n?rx}PJP|MfIQQBfR%>+SQ^>k-<-jZ3E=eutMw5Ow1Mz1j`?
zfGs2i*lwuR%4PS7lf*RZp?CmVZLD_%lyVhm>9r&ULo){snYOR?l-B;VXsl~(t*v|Q
z;~On#^m<X;iO_2x>(Nx*FTCQhDlmH~EMy@b@2h!u-!!A_rQ<`VtsJdAHwT_nY-o77
zMRebcuw#CV&Pn<GTD@4}G)1#K98I4^Z|L=uO$-T4HvUiyNi!-b<MBwkf#;gZb0`ny
z6w2wYbC(VJo1=d1v-|S3E`?SVCSN$0<v|+~I`&m|(x$!XY|#iW$JLo`RsV9)@T0kD
zu7&*9S$=Zt=Il;;`{RCK-q+L^M8_|KY#41-(-O<Lbw{g@x=-l+`$^@+zB}!4pW~Rs
zv-7p2IDnZhARwz|byiPITZBx;&SHpBdN)=vGbg8}ZV1wYOxHIIP(P1OOq60U<c^?&
zjM<Swi-zYya7$}=0jPhz3e#1nOPqQJ3PE>uygX8aCIJ?CJ6G>KUU^EJD^ScDg$0=J
z!ge2?DTS6z<xEN+Nh^OhjK_sO!TqQFRB4Rg`^F~4-0&c~#VP@#ir+Ki!h;;OzT)Yf
zoZld3br&mw7~0}W>R^-pfH)LtLgnt>xDas0Oj#dsjo#zn6pVixZ**^GtrP$92rR_(
zbO96?=RQC6B<@SrwFBN|lO{p71*=qPAEi4Yb5j1Ryg$pbM0!ER;KswPt3Hvgn~g2z
zsz8UlY5vIHwuplJrwB$On4XkjiDc8dhm}sZr0CDpcw{460Vi$^b{|nPz%Y}<`Vb3#
zOpp|Ucc)$|Z1(Wr_LaF6!f)_#9<7Xd{i}VT>vq|@xglzP>h#=dEk*E|u&3i--Sk|{
zIw{yzVf6u$d7XQ{eY?i+r}ML)=l}DtBw!A99Y9#vn-TS}v7e?_EX^4f>Ai4N*W0@z
z|H{)36I05x?)V(2|3ZN{5jO~sZ3AuEff6F)!5g-o!`2^S+FZeyK|D{CdR(Z2c1baf
zyD(4ViRE!A!(&OP)XmJ7B$y6bWANIzXo1WdfF*r?f5fx`*$cojF@y~Viy*bD-Z%Qy
zMZ1i5DiXbP6>nAJ8_vB1aTqv%JRSo))s*8979O#;xJ|^2JM^yNsUA#aae~Lb<(MiE
z9&%at=K&k|<DTg5>heH6N`x|5dLpU2B-ME%AtWSVrq{$!sD$+gr`{VFQ$@oi4>6xd
z_?bUs&`4XEbTk2FUH0nT&Ir<BG~#Ngs5p9UkB!p!W(DV~CpJmd5C|KQ@+}h~VfK}3
zlkUA$FsIYXv(Ep1@J{`ZkjM=8(+I$^#6)8^*koK{ZNovOl6I-X54ACxq*=UlX)LNO
z?T@I;_Oh)cc}6%jcATw^*kyOnHAqgHikq1-MC{|24p(EiWD(Pq<KX};K0JSk5Qd6(
zT3MHuwd;<SvkywW9<}|cGa7dsKOCONQ?COGzU`NaV(h+sh%s!LM~@wQ#&ZemB?>ZY
zv9(A>Ey|}m4<00hNFg!FgN`9#&MD0DY=o=Ah27h>b>04M7KGU`*hU0HK(lXUt@G4D
zDA(rmyJxMAn?W@}Y~Q=Lt9LGL%aXVHIO6Jxcq-lMHwjFiYd``XnxlO%GP3aW#86A<
zhr)M`9e<2+O76dDX$iS8<$(Gi^GfxGRBP0OQ}NDMQjuP-J&e+0s5O=Ja_g7O%N;pl
zJ66I3cbKN9!10y0y1D+I9Ew+;1bGhOzqc^zX#{&$)H>hc$2EW+cVVd8I{&==i4*dw
z3*JHS!a22*;x>q)<<m%Qn3f+uXn{z$6y0oJoIx}KbS^tjKI*)zP+)gEH=j}1@nC@>
zd+GfrL%AKag0PfKT_1|C`Xyj`97k0g#0i$uXmR#l?OwDHSHeT#yDNG6u4!XqEx~~C
zN%MTr+j~9%0p*xlRyely$Scmnleg9@a%;i&55Z*Q1zflwO_lNEfvaHM_77wlKa#i#
zAbz!JZRquzC@l7X8u^@kji>y>J@EV<n=NIAJhBA4dZ0P^ewcYvL&M##l9E4%n{NT!
zhTZ2qnP0z%SFxwtoE&f=ue-gbRUyO_@+mkpc@B=<^*Q`xUiG%Sv?*2RJ{(h5u2@l2
z!nFFr`9?1|DJDEwvD1$|z^XA0SeD9Oll5OG2{9CP{nz^Z{rQfQaff|YZ8~Us1KYFV
zse1CH{-PaEupB$lD}yjdx&+pYtRC>g&f)OID`^xm$^L?AUtQUo>te|ocr|<n4je!Q
zwBPB<h$&pCV8ZP4S!(u9qDxiEq~4!S+F8rACQNRly*+LKjDsVcJ(uh#DbE5I6x-z}
z;(tbm+g3Y-cAKIH#MwED5h2(ePqM0U0jMUba8i9Wp>IHGT`%r>U`~bBm85Qg@(n&s
z5Y(Z{2(c%I_6V19`2CYdF=SVL=*bxzv)D)9vR@<+&J@}T5>$p7RYRVoz_PQ*19wt>
z^NUprdZ0ze;nI@GMGgp!JFj}OWvnVp`8r8TO;b})#eRGD+(O3BeU8reQIMu3BygM7
zMx@TqjY6A!j%7(V2;ey0Lxs+QArD=I`S;-(6tgepM+!x4FK+AmVT^W^_+lbnTUFx+
zJt5Y7|Cmjpzfu&kyMF`axd51<mFu{9Yr(<ku|7^j8o+RbMD{pN3@?54@n9B+vvAx`
zoS#D_j){|b`%diBcr5RcD-ECD$XCz*Zq_hZ>DozoB%jW_(k=mO`lC&AGq`k&F<?R8
zGw_Z0t%ldkt&uZ=NZpTJ+#dw0OAGTZ2nwo2Vtq;HHrZ!`n9Btc4|1zMC1~1U=Un=s
zi~&3^z?pZtzQ*5+SHi&S#_{6f8^{`Ye!-|<Zg_ve!n_kexN@**$Y7u?%Pb!1di2n%
zajA95-h6I5!HY__Du^;fH?O&NMG>}8=zg<h80$2pY6Ms#<`sCT^kmZ8&{W)C-#+9&
zL>ai6jqNBXwD3YbQ&AVi!X_xq7vBNj*+Y2uSVD#acede9L`}SYkMmw)Ico25XEIQ0
z<mI1{!IW)Nn%=_&xx<1OMB)r0Esg7yX6Kq_=S06(E;`}gpG<E`Q$#{V6BT)6uyzRO
zym9;QiP=3Cs{nqRf@|;~G*mfP|KJf8_7iQky_a(~Lca4#vVIP&g2(+tduOwEM|1T-
z0Ng=w3-6f7p2?YX*A3Wy5SWBSI-xWrSjmE`E5DSYC9dK)z@DD_s$v?K(sl~z*k1Ef
z9Dl45CDO|j$oU#^X9WhLN9(+nLNGw4Vs-3pXJPX?l70FRjBdl~zo4wpF1Ij9sN2az
z4f3Sd)V|vxv|f|D+qx-1KRnB<c5JJJ_UTkdxtqVf<I6BJ&QBIz?vG{u+bBqe<21Sa
z>t~E=rQiB(^L%(D^@x24f6?hV^$8&_fe*!P1Z!(YRPB4i&DZX)W?|=NjC7D9I6P*O
zMQU$O)7W(6$sIkq7=CvaYd`)omvMFkvepaErDcunws}P&CcKw59zX20;{0)CR>*sK
zThA|9uQ183Tp&YFF4E=sPL~Ec4-rH9?-fUJ@*M{Asxj{50{XYV4r?pU_hUmTMF?}Y
ztI5karvYTe!Q2OXCMi3O8xB^1x#t^Gv_!(0d0l+YE*W*J!1f&tl?>0H`L)t+K@ydP
zT8f$vnd&1n<77Lr*#2iq6?jbe5`KV-;{Nylw5{a6jP3sji4;Ce``;mvXN}m;E%{eE
zB{|Bv8}F_BpZ~%tiuIpAfPcK5`5qPduRn<YuJu1NBP(8GwfyI=A70BvBmLv8^TOOD
zMr$iqtr|;wx&7b&Sl%FT=3iWZK$ByX;63#uH710=0I&X8V~rX}2u{+VErt6WoY@Py
ziB^I0Rj)yi-a=4o-2-Q2`6s@=UtdxR$cZt?%fwI^fUB{E#WCzGHe`2_^*Tx1zd^h~
zmgGH@{gom7Rd}aYi8Fs5n1Y6gEJG``7PlB|OcgPmN7(o1O-a!o^dy)BcUK%rHPCSH
zh?t#t1ZqA~B5QuH&)<T)iMqx{r|*wpbnAA5gSDGePFWMNe|CdSMCO8)SsC3Q_}M1`
zo`?Mg^`#3@VcU?y0_xCRDL2<V`;qy##s?gIqq!eHW7JZDvn!cdq9cz%BRJATj`a(+
zwuEH_(_XQAPE$spxGe|S1O3|1e?80+vmu<DU;(a1QWOa1gm-3O@EElQu1aZ`MiU)5
zNH~L%xG#zUyW)#tpn4bsRroUQ1o9t=p>~RPwnFk8u&WH{9ALkzNZ8vb*a7wL<l*tD
ztE*f1WWV&s^j-{zF+_$n$qh(_mGMr%9Vo?`zKXjE=))6KXqI!iFScl;XcmLbbGg6<
z6B{5gBtB|%bTq&=MTU4<1D+zOE#Mu1=JO5A1OWf!aREf|g({4KLV-X!fTO#S)(8vM
zy>^4da{P(32QA1`chKUVytLJ~LJhvOYE&<b0Y*NRL2kz?y$@P0u3t*Zs78ge^EFUb
z$-E8ZxY^y;_v?^>{GSWkdm#_t9glkhDZD;nNE1U<3KY&phd?pM2YWdZqy>-MHQ5Jv
zc#d7Pvr{MjwiBDrqAFeziTGg6<nO4N`S8gVFw2%KQuYy$!@<oejWzn>^fc}YR&aGV
z!M~_PUP#l}ySyzkaK{D8#0NFu)rIRTw$s7#Ern8miY@yLFI+?vl250yO8-D?R25cW
zGF6FQ>e&a7)N*LKF+iU;cX10=qM2U|x58VfXYpK2=Vzx8=(B~a+y2^<u$`I4u(CN1
z%w1x7M#;;Ui|!jkC3Nx9r8kJe1pQ|TE4MU6&JD9NWB6B}fccKbv<8*OTbMv2IYKR$
z&-XQMNRnWRu@_A8RHF|wks%hkuV)_|fl{DCjRW%^@)y{4pFsOVQp<37Q{Y5Kr>CF9
z&<V)|*zdsc=UOAy>FuMigaY6x3aSIkW(ihQ-0;5Rfd@BmQBFxo{p8zdlrW<|gyj$R
zb}7_x5Hp2er9X+ZGJM*K$5K)tt$(QuzJ;4Nc75s;Vti3%BJqbGd#V@%J?scFNyApb
zcR>=P2qcb##Q-5TD#k?&3~^Vhcb<$O3L~7&w_y;4StYjA8gPFw4AggW{c|iGZn48l
zr@|kHVxchuP1I<93LzOR#NVPo<WY(7){a1jjk5@@@=akLUxtPhk;8#>jccbr;{9)j
zToVO?Sznj8z|k{jq9t5%v8uH{CY?@IzzTLi#|pNq<yBMN0A$0Ecpv~U<EU*+YKj;C
zU;vzmT@34TbFdbgZE+MLVUD&4)dCF>yW^GjC8rmFN}-n=4&Ehb7t&A&;iQ>%V7%iy
zzw>IdfL~P9HVUGhuKBqc63+ak>V5#}abaH$K))1CI8H2MlI03M5KcB>G{2}TkDwi<
zA*@r_?1yh~u0o;!6umuIBW5@Th<!OqZS#CnkX=ze#F3&ISq$IrVQ>KYOeFD@4;7}z
zKpA$JgjWflzD*LBy%E3{&AkK~8Q?`Xa7-k=d5Uay1a%0*ObMA&k8zk2hksCs!Banr
zehC27On8kt77>mlPMnlP!h@2zv%kL_dOqmxi#s|DQahlgA%PP3aO>fng86+fCLQpF
zBS{kzag8i*!~_Kir7&o>UKlu#!f0Da3JX>T>cW7Q7c=;37DNHYxQzT+XdtgcryLAA
zfO5>ymDnalfF#NLg1uHWPEy2r_~DGB6S%amB~kSzkOF|_r2$g&M{&W!VB8QDH}atF
zqv&}xvG1=-@PZ4n(C9mHA6ew+6*y}pFDpX5Ph;Nplyh|K<|r^A%!uBG9Cy+!2j`41
z-7y})r9ra-0WjF&`)t2p1|cxM{HU`KxU>z`<BSBT{W&r5HdJKioqer6V6A_JG6mya
z@NT&-Ej#SZf;2+}{;2lnkU8F+ohP?Yc<%Jl=s_Bq5HPYIcWf8F2uS!!b5Z?IexEg7
zp2e-FO6(JG8lkev_BzDp8x3HB$9)XEtOOJAa!-oTqr1`;f~h)UTVi%_aam>)-wRwb
z28pEOUCYWEL-}G&+rsEI^*3olD8dr6_$`IFPx@0};>B_MMq@o3p$URsHzsjE)HkK2
zrE(-;G@9Gj7Lm>L6{D=3xbb-1k@~kKI|aq^l1P#Wk!%Bn<{3-NB=j_iCOBP|L_V`>
zJ#h><v{lF#jl*rqi)UbY6?Z->b`2-e?~>aAa&Q7h@=Z)k97BA$oHA=KcG8ka3irRx
zvq7uR1~4*9wUv@86m1lRDN7<%i3iiT$q5IkU5>z?1EGM0oXo^49xI0rjq_==0`a=}
zj_O)@mlN-z9*Bhni~>Z8+Q`xy^Tj!63m`leIvnyeXu0tYk*)PL8vd#VsV_~dH_)F#
z9h^Yq#fU>cCDjoEphgQvNBrKiKj`b6<%eKJt{t+z^$FbaYb=6goOw~&sG=2$1tOxi
z4@sirY2hi;pTd@mx%3#-61O{$@#ugyb#g&?qX$K|`tr0EP8{kvd||96mm<Up5tlH+
zfJna@w@DLn3%^7GnZr#=1o_m22CKZhyxrk2FVo=9Sfy6kP{nNCAa3LLS8KTFlv~uk
zt7AJbxNWjDLxc_jv6jNk*?!;V^xp5^Qzz`<{sf7w0~$f7D=5pmJ$RFrbCzcTA7JXM
zm3f995D7ewJ>Y)<3;Q*QOiyga$$FLKcY?Y?;(o1kAOo>zi19hC1vHlif9874Sks$N
zpojaTCO}<v>{0}Hh;)$=W`8QbA13vUjga?0V0WGJFnZ&qjA&AX(p|$?2oh~$sA4Qt
zt~%^nT3u8`!5$C0h-8sM3kiO+5AM5DnA?fz`J1X&;;a|3`Lx$@au^aPE0ULl%bVCk
zK_ot)x+$mPH9~++;5NdL<V|_GJilAV&=0I9rAK>Ll8d&H*K>*$#XQN~h2ygszK*4C
zJOYaeHVnon@`9qWYjJrt1%3)svt)lrUe#rO^)#l8gKqP)*)x`XK{%;Hz+8;e89{SD
zII8N7e5#m_A(&|7eK#Mw4M*EL(Dh;2A_IOUtkHO2ix!D*jcHe5PVEQ>N58_oNkKas
z+1vQ}<OMeb4p4Ty6Un0DLr@}RSE`;2ZNWXpC><6VAE(=xi*OGBmkgsH*0`$l+x^jl
zuQ2cjLzM)fqSSd~-{C?$aB=yP(;S@$<|iefRW#LxV^#Lx>(3&y4(TJutmAPfTt<71
zYHSotO%iQ_1;FrXHU)K|1q`s_Th7-<(?cd<3Z^~aySVj{TT5PQ7P=n9sfWru3*aKD
zvTU3>hCscVhBMQ#&k8@Z9N^CgQtug*mL!<#vD>^>yQ!gJ5xRahxMksTAz>!K+b;mF
zMSb82_!g88)kc4pbE~LX@6Y|mTMm2Q1g|A4>r&o?C`Ubs-wn3W0gh6PL;$n>iSkVd
z^Ii-d`C0+}cX)J6ix;D7zP$k^G2wtc^V^HhN*}2S?%RtQnXsCG-Rqhr(c=Bp6*5cd
z=&oP>=-i&ulpZR6UVrUw-3K=L_HXa~?SbZJ5W2J&LRx_6vp*Y8TE0MP6P6`OMu9{s
ziMbk-U}O%{6-@U%6tzakMHV6np>r4jQ5y9Lg+thaUZ=BnaWFo`%C&3DaD6w-1?qbH
zonrOGLJ`J-C8}nES~-_HKORz6X6>s9zjgceZHZe7-Y0u;XhDSU3$Qs0qcdm#!{Tbt
zO91#V0Y+|Fa~hi##FguKWT;c-&R(RLsCjwtX(>Z1tB0oAgD-Q`kql_39MFn=_5`~D
z=Pprx0D!5$HUX`l;5=}(8NBY#pcK%@eiye|xW|d5)lELYnuk86F}1`-ocU28+CPQW
zGBW4-hu!F%Lf8`|H#WAdue<vij_A|A8yFcG=jSF3!kM|?yi1$Frzo*^FBS9h`Rn8f
zPY3iwuZoalhxJv5wS^~J*Lrf}8HdM1KZ-o)a3Tlep3A?u!fOS)3XHTMiNM#$z{W#9
zcytz)*cPM`@}Y|D10jxp%ZMyDL{_roh^oNoPoZPWiWU}E3v5HzA(3lkyWldAgTvyV
zP{Ud1lL1>`qS(h$c*V5glPFXYC$p?^QlRJs`Y8(_+I``%@GNh2ka@LQ=R4xK1}-8_
z!bgaPss4o_n3(*nztGJXRGt*@>IbGeU@KtMto;!$kxm*%aoBHs%|-K4$B!esLQqJk
z3%eBr-~7yi73cX7`GW_R2{jGyxoW^_ml+m1=hEb>?W@rbo1Gc)bl6XNIlK@Y2$yhp
zUcT<O0f#P1bro=Gh#&~#101y^ac#St?|_o=FCl~kio{&KU@6H;K{T@ojDB#+pwD`C
zdJbhfnXLi+6mBZu8ax0EuxO(`uGC{l>3sk@B8u>La4K+t-FSSzZ?yS@hDM=O42mck
zGE_-*JBk+bI&w>4Jx~g}5P0LL035;@{|4vIvk%Y7AfzSr4TQVcZ6=lrqW$~q{MX*z
zChKKrZC#DpSY$%%FW&BUoy%+jEVG7lI?$`&OIg+}??lpaX<gkNz|!V(h6^Z-TxY(E
zcbZ6w0cs_dxnxy9k7S+=CbZ-HbU!=q>6D}BhlS^!m)tjMh&hJFKp107Y=Q9)c_Wt&
zS~8>p-Q^jV5Gc_h#UIj0i}B%bVOsX)HpPNF0LbwYq&KN!D>;sgqN)ILIJ)iSjJ7cS
zl!zK07KJYYke#12P`K64-~l*Kw%ZetC>Ulcz!ZT9dqX;by5i?htAOlxXmyYyW)Inb
zG|q;v;DBQ5V=ZK0$0FrDg&`dQVz2T0%#jy_9jGq<7RRA4Dn`I)@qK5wPeoD~;E`(D
zz;)Ual{QT7?kvCHMWk$~wy)!6Q$<!Xe1LH0GvU;Ot|<g}U&5PrbF;JNb&=$28X@95
zXqVPqkSRz&J0_zz1W#R}F%1E$DeGd-5mej7U@Tp<x9`VYLQ?naKtK>be*)i9_}_2o
z4HhgCGtc1J61^XwRY(Fz0=wJX#5&{$D1}R?ry8Kbjyn;q01z?}@wq7UOA#)QIMEA0
zWFTKk?0ZQ20NVyaIP{A_&}X54sszNL1#AT^H~OYYawF3cP#i#xkf2~EJ|zpgq^?;$
z!0;rF*v}J=2M{eVy3pwDIrW|_3wWs~>g1-i47Mcpz6-sFTa!lM(o-AcsN)<%T-PxK
zHxlnKRBf1BG+^8o(B@+O-|&KE8z#YPq}w1q3;P{rjfJaPg~}$LB)CERs*HX!2;gb*
zN1WiSxJ*d*Pl`bDnjok>hDOHi*@t#myU=pu9BOH+#qiFg_!_+e)CVN?9JUIFdGudB
z7e@2i3;l|=9nQQ2AkBOb^pG+Z4R;zE8f~PTK3Ii3BXxV+lL}}+S<(N-0*y&Or~y%`
z8fAjp*-{H<2&**S`7!cwQMjp9A;v9D>EWK^JO%}?(08Ta+U3I^R{>?M;TtIzVsQ=l
zB0C3-d^%b&9Wn+C%R`5$OrbgFwfKNQ^JH3X7+N&&N<lJC%MDIbBX59@kFx9hv{@{0
zGGBDKXN-)ZoF|$!d{M-XVZPNG!TzUQ3(P*~Fe*I_6QN)*X$XAc3ui5|D`GL{sKnE=
z8(xi80^8-Jkhr)8j^H3%)}SLR)latxBkS~QM`0lZ&gp0gxdCYrT0#&Zr4Pui)*{+-
z*3R2*xP}w5J6`Grpl<OE4Q*qT+`g@$75$vJKCz<Fo`Ka29~fO4_|>t1I0}F5n-luV
zqfmL^#9uh^FBTBOv5)!(vn2rN{}Z<PU;p}3L8ku(S_3_gQ7~Kk7ue~&OX+{%uD{un
zx*qI2DXG@@A9Pjbti{^CzR;EszyC|j_1}MY@wWf^TqTUpu#r=C@uQ5Wi68b2!Fgt6
zV(~VtIT_e<iP^cS3?sns+qQ4-R<)N4e)LGGeA(s<0X1)gnmZ2;d)GE0#FYEo`IGVf
zo6&pCH?=Fn^FcYv`RZV6vb?kNth|#R6J`1o6_l}sQ87xr&o6F+@s45IVJK6%etqj9
z0Va6kkMlPE{npNF;pwG#^89)2moLFtV9!1P1ZBOXq%(3USt}kteueC5wA)CfLtY4T
z>Yg&cEw^Jm<?`G<WLqb<5SokCkmWp?wSyT_|5Tli)bC$a<0U`rTe+~=ZDCH_)z$T%
zbi8DEvwl7cZK_(T3O(kR+-ezHF=j#~qG`4t4lITkM__F3bzsk)AU>TCRA{!qKt%0C
zY|o#?NK$I;t|Ht8{+lIMtXnr}bi#eA7Rx(cwcH;S{*Ur&^k=Jn3D3rIFR^M(i8|I(
z#%gz=sO&?S!Wur|Z{_SDD;x^DLJlmD?-TS>H(a&0=Fv}KyS-u0Sv1`i5zQMmUcYAk
zty{)pcACfBO#X@2Qmx!8sd?>Ct^q1V=*jU8wT0cz<^(KekN&fE4re{i$HTrf_~n@Y
z+&EIsmjN>VoFRx4?Z{DcZoGh+FI!wSkL&57b2koNMDzD9U3Ws66##G5?V;9vkY0Ux
z7rNRVW!na^b7a_%fFl`VT22hhF*3z5<Ky&j4>a|stmEAl=1Cerx=nipes@?ak8<=P
zpaYDZpq2p=<ZPGZ*MH`XuA(Mf3d+Sbz~0r;)R<mgqi14f<^cE{>b$nZO&tu@U_SL|
zh2|t}k9@cJ>6`FmKzW5bV|8T^zv7qgA4y&m0e*XWWOIkxn}3s5EGkO`vcNceG*$uq
zM<Qmmt6*P=K=*B;yI3Gc?EF4*8+<`>BjNA^)o!qS|Gw@lap(7fkUlJ4wybYR!mep-
zOZ&?#%VyrOb==b9rqwVrGCg^8@lJpCa>Pht#6qn7ityboSTuQ~e(m}8O{r8N7^stx
z@u?M<1;wflwU8K*A-njK56b{Kkt&a0Cwu!o9v^HDTZlp9PCh;{x#eI}-fAu-Ly#Sy
zqVnrQ=I()I%$LTM_U~Lpy*nGal6utD^9Bc&$U*ACojdM6KF8v+CzF7Kep|vCi8AkB
zQx}6*xV{^!t8T$zz7g@l+0O3<>QQu#h2LKXOx^M-sOQ<1EniRpk_y1Kw~DR~#uqK=
zI^{3M``_Ld=-5BG6R`l0dWRk`Spt)heC@R9`L4;C5!^{%!#}Rmy2Um6YUhp{=>@+P
z`0Xw0+!3EN+Hc^7N;oQneg73dubLW^(igstnTS0AmSwib`cuVN;X)ihKTDU#UmHUq
zHkTAJt=nV5I`r}LMrt@!2in=!fluTCm{u_EXYVe=d~~C;`U_$7t98D|(HheQ?6FQb
zmARb&MLit8(RI@g&l4JO;qeFp22wz=cHmP36!y@_wXxg}&3iQNw@(Fjw*XqIM4geN
zPJ|yghC(PoKVM>~C0*kqG}_znBuc=`(0gg0Jb46D2NQ++W!SSE@&SXhG+ZKd?kLkZ
z*d)#_4q3bQ(LtvWIp%dLSx1(h8~+k@)@K!v98c6g33kjM+Q{gM!m)NU){}nb*RXo-
z$j97(oGn{ed34!^KE2>kf4Zan)vLXi`utj~()9B;C^9q8{~kYysWNu;gBDpGWmYtk
zeo$0QkN0u}++J#+p7yR{B?q!vS}cc%_!Ri-U~`g@MkIxQ<4vZeI_xWwmt{x5@#}?c
z-wvFED2$9Oj68IbZaO)lAX_ixdca>j`EGfcm-E$em2?|DkYYf1zN>xy_cIG5g$X(~
zC5T!R;XXbMJ_!w(Md6zQY<dh&*aU<0EjV@FJ{Dky@b3#2pT!4~RUq|q_&8wv%$o3x
zGcuk8oy1+-oV0H9W*oKjVCr_^?>NtV&)v()BcrdQW7(1r$n|^qil=7Ftq&gH)=OgD
z)6x>et-lHH6daoV@R`5YkEEMqW4y7sc|~rAYqEM8_Y!OnQ;fB>dr=6&-H+yiM05WP
zq(?N6`AyWg`1t#q0!95lq<wch)_?!+g`%OfGt(eR%ZMncL}f&XB2>sq60%3B5E@1z
zGn>pZvPEQMWRnqPW@Yd5e1FG%?%(<6JRaxV|J~n;>-v1&<Mn#37bv`S$Tf>M>Of=%
zT(}VAjC{}iqKZ<hBqc&R2Da6P=`Sdi5JLoFqjt>_l@jJ3dqQOfY^%*XCI_j?+O@?V
zDB}!PE>jRDT;%aasJ@dudohQqy0G;PcG<Rtv66ZAj4zvRRr!AgxeR@VV%!26j=DI_
zVAO{C>z=`06((Ntj1nMhr+V<PO4@-A`)M}y%=`5QC8S?j$3lHLuh4L2cJ^gJKn3oN
z4=q(dWpyiqI5`BAc)sYu4=U7QTZrx3jLLtWeTJpaVN_)Uuu5to^51b*g7;5UFqt_@
zbR}Iyt^uN^Pn?|Bhk}%O``OJ;(0o6(pv>Hm5MZ@^ifIQ%`IgU&Xyoj>n!@2k%YfSH
z3GU_5#$bm{tgIEqvTuN(+s^va(&m?QQ?Re1&~!oBqBvidrzu9(B7hw1!@}w>E>0Ua
zBDd~61hQw&h$p9{oVm0`HT8BW1VG}5u&dkXiea!bm9J0FEG?3$AE;&BJ=__O-Q9`t
z+gl5tFf<B_?n=eQ4tTI)*0dMmiwnzCG?xA^P9fWD&l3i&6DI-yeN!D?Ho+>R!)&mE
z99b41$1BjP(_NWq)CA2D$FA+JP&u8MYDk29fLIF?uAfE%L2(nWmfpaCR8xBn4tiEc
zlm)eQ^#p&djkmFQy6s>YPJk<!+N&WlshwMSYFwi|+r&xvpc)ynfYFgqbkgH{k}*k;
zNr^9(+y$Lx_PhGQ&CgG3FiHHK(QP7tMc5`7;dg83!*7&1bsNq7vu2I4cz$9ixvhHC
zGm)H2-Hj&Q-`Ym>Z|<f*7+}yG=FGs$$M=m{_U*nh7BS-}!CMo0J5JiVj8ym;96GQL
zmR)2+2qWVUH`Xp#(wOe3Xv;#LMh4D<%{(9X9i2{|%DeLxc|}ojHI0oW`*kwsE7^_%
z(23+OIPM5wK?}r@eY(_;(a1VSXaTjfwTsZyyvCw=!*EgC3Y!y+!Xw;0s&Uz1&@hYn
zbni)5?nYnp*g?al(I5S}#XfN=RQpsvM5`j(=;E5D=9oRyWGFPK3e4#NHP?T;1fz3N
zRV4v*gA0@YuGz>%YDe}5Js!FLi11w!5eQ~8J+$Y5fOl=w#Y+jcR-nG?LsS5I)H}8U
z$YeaBw?U`eM+;z^Dnq&8@G#M5{V05rY&d;~La&fU2eUd9P?OXvFq)nLG-#yS^e`qN
zM*jjDLqt^YxbXreAfv+rb|)GBs|A2HAs>~zUiXiJcg^qXGjR@F%9LR`Gxht&6ijuk
zt#RW6Mz?Qcgm5WKRXIqZW9EN2&~m>`&B{`3#|Q*-|3aYjWR5#gy)3ai$j9B1wV(Mx
z2PzbyXC&kuHIJ8-z5Rey!@@L*7gk|wZ_F2`lFzX=hwZqpf9R`>re;q?;M)9rENo0>
z+Z7tBDqftEogJ+nw=K_|O&TrG7eD@}mMJ)*XCS50^BPb&j5h4r8-W{bEvW{Sy8s#Y
zJ^j?dqa00j&hbuLRxO6ap^S^CrF2BH|5g5Wji(v*UZeyB2JmH94XLGH$-Bb_;bi#5
z*nrEp@jhX#ToPNZ0PIn>@d3K-C_u9lgB@bB@`qE=^t^{c4hnbfH+&hLmqE#s09^nC
zV_$_jv@@!98@3>4opQ0VlBr!62~1Gjl1YSM7OCp|3?2^+r6zP>Qz_05Ci)$_LH&zI
z$x^Gezm0{&m?<S*azZhcpqZCozr0y95=hWEh+?9iPfj*@G6m^`@c80lDQE_Nxy!yd
zjDhDoL2ijn;U`?(;kQ5D%P;SpAVh<1e{`@z51uj9Oz2cehQ0Wc^|Jh5(<Q(90|ZMl
zEqP`k)<^HjozH$rSFTFH_<oEhCsl^|{FLEz8`21Xp$V9plHp01l&rF-WuzqzvJwi7
zBaj8sEs3p~0)MKHgPvL?o;m~&o$3(&;`ZJr>#)6a4W+uSgO=?4b>zxdx^*64Mc?09
zv_|<idML^%|8m?Be*Umq@v8cBHGG^@w+2=e6G2wXM1O#Yl_z2r!D<3%IdlYWtH3-3
z1b`p^TByl?`vfLMS^BZ;gpv)|s}`84dS^g<0BX?tQ-|)_r8)$lsJlJVL-{J0H%f;2
z;49TgSRoubPR3_q-)Rq~8BL6IEa&`t?#O}Spsk+UIAadW?b-SDY-H9EvW%U)nMD1K
z051^C4);h(XO$RXJOLkp@_TX;Q><IkN}e$}fzlRR&+_U{AdWez!(Q!c$I=={V!XKe
zdwxkQQLlV`Oy5&A;p*i{3_<V0DZC(Zhn4IZ#NEL##{j+^jo5rknx=)ZUUZ|eRNiOb
zL8HVrAQZIg8Bc-sg`q<}uQ4wc;1+tsQ2($!D2qtW!$8PxM`#8q5bDUHUyMfQMxvwH
z+1b%PeFZ#YelpI(-F@Gcq@6-{zIS378{1?UOGr}e&rXThyH>AgYM+oa><N{_4D$;y
zJE9XI^3QB|q4(A4NNO#&QK~^RU)z0x3sZoG$zU+WEb=&Gu7pwp`LByn^MyD;KIwWv
z<J7D<1Roj1aK$O@eLjyprl+>Bd{-9OJS~9T99&%8KzYC-m=CL+IhO@>1Nt@-SlhNp
zT&f~-dAY~@>55Xl$3rhg%ZCbxTh>qW)hl!&o&(|dugSvEF5MIm5E!Etwf?9<Lml(y
z*@VTvrBK0KUwuyxszmY(+yjCI&ww0S9LOQVF1BT=w)R#85;hkfS2cK7PNsR-Ee~eI
zv{vJ!ioa&Wj1ylMN2qP@ug9sW=9zZYw$9b``!;B%VKShvq80xOi+c#fvYhTa4J`3$
zy1gh|nFDFM+#RNX`I|G1-iRoqre+<7{o@ueJd~BE16NMR3bnB?4@|zf*Wr5Dx2Ji2
zK*>ZMzT>{0UnGE<>4f$wag4Zc%LEY`0;SP+S9cBp0dIWCanKRwmpgFKP!T?*AwkXA
zG`8k_VD9V_RIABI^qHF&dUJ~-ZEud6l2V<`!HJ2zgk_;O`+4k3bd&dIL0nnqa^@y$
z<9)4{ghNa-qi_9KzYK}<;)WL@8NbkXz^wEdNY7uOVubrnGZ6ouid>Zvkx>3@rN*Hp
zDsT_w*^L}54L*%UgIOb2vTm)T1gU3+XC>-CV-+dK*CujP0Ov&!6Ubq6WuKe4kRGp6
z6_=nHXvPhl^s<{v{_=DO1})!$<Cn~;LW2lelySCJ4*YtKsn}n*Oi^z?{NEJ)Kg&7$
zK}U@25Ka1!a<jZ4-e65g-M9II<AA!&`dfUsq1K!$lZt0v%3EH4YFdfs%D_!92ld8d
zIyNAuu#)<F4m$a+y-$D!{AESxbzdD&_wSjHxLRhx_dLz0Rz`8;cYi4o%|C3*kvX_$
z`pFaP#auo^SzD_*UE{K&FN473*T!g_10U)ay0tpXd?aCM8FX`Bb*?C|!eT*V4Rp58
zz|)NW`S^RGtj>yPxXgZkA&Ppioch#`F~NCg8V9~r30nP$7Vd8Mv64~)%4<6$t|Tre
z_q<_v_``h*E~KZUzuSP;8I3nrJUZ8|&VuwE>!N2dP^Hu>T_gy7cW!p+Z6hnh;nB=Z
zbD^4Lv-4FwW1C$mh%N+Y{&CcvucIak^Yc4%stm1xdon_|++F;<7mWh06ld)5ZYX2W
zfWg&2zwe)2!Y(ICZGdoK`uVXEGbb*lP^MphIDXOnsv}X65D4s32w(N9%-6x_xWAd$
zE904O2Xg6Au6##ewZtVeS}Mf&k>F@U<-|xVPniIL)pm;?emsWujwXc1UGMsJ%u4gA
zwE2m;X>!i)xlf8j!;6>K@8awN_RX7~biH+IXkd336UNO~;ywbj$pp?a0yhRkG7RG4
zg=S7YRtiD&$#OyA%~Q!id7d<*VrSlO8ZB^H71-KwD>`>?3+vc=Bg6w-!W*`KbHXn<
zZ-K-oU@TDisG%WlGa?!WY9VYlpyWRUrUg__I5)ef!4^bam00x10QN2e`o?(tmV0I6
zF$;)NQiuVNg(YrDTq#Y4i9~Qknlj&q-32GSv8i674#uZ|&4x{p2z}nEr4M^e{5^!y
z_KJW0cPFz*j=A~y@$TwS7!#8aywyf20d50kuC~l0(0{LwdxDwx^6|TlfMo*sM`^}e
zW!`MQ8VorWNEH1)`Gw^HAmGilhWwAp1PXd^D@F~_FHvD=`C7}5r@G@111<xjDc)$~
z{;Iy5m;P9{;v3Yp5Blp~Q4*<Y$Kw2!d6@@f+F69IWESM%>yViefS@8xv)=Bb6h&1e
z2^=|+kGLC1OzsZ(DBT)e{<zvaK6C@7>IMb1A75NGA;>|QVfrR_bsvX0r2_0INDPdh
z=@Gb~05Cx6^O~(FDk|jX(Gaifv=9^P(q!UHhwDOW%i(95WgouhEiHI`nwt7;qqA}C
z%<@`|@ubEDAtY$JPll-<r$k1nx?btVz4!L~k5Bl?XAk}rOWa!(h3Xlz%Dn}5><}<)
zpeJrm_z@yK7U!oTgjJ}xKXMAhUEP}WK8zp_Sr*TAq<szQWzU}vxm^Cr`>@KPuU2I)
zNMkc-Oz^N7Y|wvb0q-ZSnYiBH;#2>;BHQ)cF={AiU))jyDOu>eE)30O#d7!a#z<e3
zC%sWG_w=_$G2mjXZn@tzW3r@?%Zi=tpOG2atTOK8@L0%wiH2yuc5>?$NlR#&F3!mW
zSI(IaG9o*oy}dorC~B%vQ!(4p1g>X7n5CvIzVd9%CD=@J>vxuzn8w6stTSEZe=-e)
z5c6178JLVSPF((la&DPQb4!co2KFi-hDTTWNrH^T?xd)4*W`}>q3*;-OQoIqo?+sI
zR_=zj_M=)Jo<c6f$OGeAv+cm^AA^hPrCU0y<TujMQTG0C3kFJP*ej~DR1W)tM?u52
zrn%J=0}Zz=O>W1UCSqVFgFKDJIRvZwpe^dI3!zfaTYQB#a~EN{ujp3DN6YUzdcG;?
z(w<__tXXMB{S#k>24X*x-uikoAHnY2yg#3Xrob7Ae;b$Q+4E%mAK`8wgj$<DJ+pud
z0FI<vlLj|;h(%htZOgV?3INjk9?cM%9Ta41D4;{aEfO_AH1k=NZAo1h%*0QO3pu&V
zL8CNuAMVjM#HzEvv(VQbMhB4b2eMffK3=B+axpNF<uB+jaHj&N`7x6}9Plbbs<1KX
zHWScdt%XOc2l%Fp3(=v`B1IVLe88&bQFRc>7VuDP;Ag0o?Ae5dou>a?B<mFI?}G0U
za&o4xX(0m#p(81UjVy2Bk6Yy;k4o&>xQ2LOdF;rHBXRw8u~)rtp8td^k;nJ&`x}L{
zBoX_!u&J5-%X(X?Lx9~_=CjayK_vO@XTm1Uy{LvEGTMsG*|6p=J}xf;H<0Z{)H+(<
z{5<NBJBarDEClYO^NSY#LWYw3LP8*ZlX$MYfKLl@`Wql!Qb#{wO)*2qd3IsM!F%7O
zuAFVLTdq0*14WhGgR=63-P{&v3d~#S+1XL9h{GL*<L+7IqN(at-s_#GDF1h#jAC0O
zNXw9roeu?XlNc1Z(n-K%XY6G-Lyp|&DiWv~p!x9wCu?=mr7(uOf*&q`!u~llB#Ry9
z%)r(KDGx?8JR_orWUX^ps$bsu@sCYUbHUte$ifm35P-ype3o(Ra6!|C1F<3^DcWN<
zix=m3usVNM1qVa6!iOQ|=6(90ikm(fZU<YUT{1OvKrPdoG)y6<sB%Z@-SEQV)_UN~
zKVm+h1bnjue+Hte!jQ%h&JQAwf;dj_on}?_E^#t+2mmuCSJm*>dkDav1Um9+^1X-#
zDJ7J13xZaBqB%MJjfuex1D9Yi3PYkIMpFBIH0{AKq;t?9&P+CHrY_ysrx;0p1FB7<
z0j0D%ZDrp$r7@O<v|H{w_-TcN(&>X6RXl=^Jm}(-8lC<FqLbg5APPoCvRL&qHlNqT
zYi6C){-^vYxtw4547Av$jT=ux!Pandr+>_K$m(S*mVrRWoPPGeiI#NUJ-~|WS_NTX
zOG-(33wZ&?Tpn(dn8g^X2SQwh4p>C+nn!|7OJv))CzlnWsYGBg)SAY7vuIW7i9b_D
z=8-M<EVXTUFk=nWNkM8h_2|(N<gDUbKs`r~hX9_F-?nHKnuhq{>9lFuQ!rzo#>*IB
zT*;FwgDb5Vb9j+bfHA3c_r6Z1VuB<FYbBa?K&~81l@A_dJkMnnkw)M=gADc37E&7U
z!?gsCJ_KduVcofdhz3TIKlQ7FTEhXtP-815-;}Uk9(AhhFTxE=nc3Hv81&Y!#IN+G
z8HeDOE`);Givj_?0gH}1^U(oCi@)XLwxPt@0e$m~`S)Y-O#6aafP>He^3{a>=f&{T
z!mje#95o>l2j^9}rZ(6O*$yV1`BVtwd9^n<WQGq;ECT-4AiIKEB@sm01r-!aK&vAi
z1KH54*M4+#xplg!V&{Jt$S#_hNK;)H6$-Jl%mvze_u@3i0)x6`cY#=_PF-IOS&9R_
z!JzWKZQaIPhK^=zYU)*Iu)NQ{OAn3n<}sQr{k_nE--;Go>Im;TVLS5z@Jo1>_1#;Z
z|1ovgk~jf?gPYmetFZe{np>DLBYY3F!a|T|wZ^J-P;cW@u0kt1V_NV~t}RdiMPWY#
zODNrRi@v-J&OF|QXv*)$Aj3O@%G6c*YRt6hJ=|@#=H|Xlwi>5Kayy*=m)<zz!VrGu
z^V5nzq34r>4H&!hTds}mAUq7af&TX_kI`{TdC%0fZ43^@AnVZ~iw?4)%rEy>oxXRE
zPc+k_7X|zQxvQ(yT<Q=GOk$kK!G3Of827Nlz@ZqdS&Ta7hiFI});sngb56Sdwyy;H
zEo+VB7QW5Mq^w`wwlUl(hid9!ThrS8(S5oF3CU+aZ&OQ}-lq1pX=_&WM%mvhZ9LBW
z=5(!j-IP>xWdn<Zd`1U*3X8A0|J%1g(^j^dbSv?@z8AFdr+;DnZxfA+_RAIj<JTrx
z2(GyDzuDncV?Qo1ocb@)_T&xE|7L)@Ht${i|N6&8Y6v5H3#BJH+Ta4WxF3W{5Oh%K
zYu-ITd#!bQH+CuRUY-9uADv;1o&c`TK5-R^zdzkeDQ^Wr#^EsaI%=p~@%1Y=<R&OQ
zB)zzZD-+`*pp2#m?g`26MF<YAAVU)V)w<8PZy7u!y?zEG7--ruwac^`r2z5K;ixPP
z5Po077_kL}fMi$=3G8!PBm^w^X@Gn&eg~(LBz_vI*lv^v6fA87ia*Ca1L*M4<HtF*
zgPVdQfvZ+y2MI8*!^)NuxQ{XxT_hU>HPBUat7i%T*aw&;xqZBZ-GO7!Vgy3>_AwtO
z-MQ1S?|>2{1`1^Uyg(ZO1KmXkD~Z(r0B4xp{FHW?cWGw^z@H=tUEmKNqcRA*UuyTa
zN;23VBIYF67I9ob>uZEhPD4E5kTwx33w|+n1r1>mqg7M)<4Khi7SbTh2?*FH=qE=J
zmydV_6a(uWQO5AY;k=?M3-|(r7|@s69B$-3$)J{Y`v#CMcZfnEmf&JchahM@>#^%*
z9eB1eSCV%_TViQtRg>r7uzUAzh$?BwNrh4a-~#2kmR2EfLnIpOA`S*;Z*Sl=-x&eS
zNALtw&dGBhY%5N*HBjilKTaZoasH7YUyP;nhc$-k1(fV(9RN&Z`GQDzi91l~<C86l
zyXzq$SBsDf)}V@&4?Po$2NesQH*7S@`|<79$OCd!#ETI)cE92^FUA2&8LOQ-<w#wJ
z;-I6EGhDcM_-`592KG3>gZoeUK>v))t--XC;^J5+1@@i7^OsJ?<l3G!^6(Z^^Kc0W
zDE)#Kg=zpd?1dl;Pz7t`@q{ntxNQv10dBS*=#>wr^tPNSR?WM2lTy;sl!3h}Ix>rx
zMYk`(D8cl>F2eh%t7{FLDJskyJA|w5$FmxM+&jkNbeET)#qhP2(E#3}iK*ulmSv^h
zZxA<^Aq>0xa<@bsI&rgBm=R(I8WR&Ksc1hrs;r>%*3r@m^A|MoH)wpRz_;R<&p`y>
zgS$dU+>~o~oM@h4>OoMrFX4s(H4(xN$s4l%0zbepSeFDJGwGtKUl4<A{Di~Og1Cyf
zoNC75zoil3$ZXgE(G~=3qz%p-V)l3fj&_YBpn}0PUud=PZfe2x=G+58M5d`zu%x6X
z0>}x7Pe@2H*aX8HB32bZ(EZVJY`2DnhV{^_xfC}P=?|f&?y#SauVFM|Nx1qxvi&~#
z4!lNMl65pWMY4FI2|xh!@+*~HCy928ENr4W)6wzK`;GpN<mID-`rOc<oK7xDEYxij
zsOdSXn(?WyWT#KJ#&A#+YXDU<fjmN0gxVFg3^QX<xU}yx6j1zl@TelpmmQbkRnZe*
zMgt~pX_;~7_RJuFYGFvY@UBRh53DXw$E~KP7dTdhcTd_MAd#F%F@}aoG2xmpdb9{3
zNwyG7o+HHvmS895MPXf|r{T_=LjCZzMi1#6ztH-_iAJNQriRA`tK3o5672sUh(H42
z0qLD~#p{>Qvtq8-eRY|8|5>eX$c9JF0gU%%lH5GsNvI(+jo`Go0SxrDKXYZ58HU18
zNN3b1401pU@0sY#i_Ye*d0H37_DT)q!U`>DlfHlh!(68zjaGYgq_I%k5B&}`>TO{>
z2s;7}*k*7aEcNQP_IOF2dTMrUAd9IA!v(Va?+`5_G=7{ld|-D8%Y=2g7k4Iig0aEN
zj=2~;*kMqCd*fMLLhX81E6aRP@&WbAnW6NKW5`^<X|e~?18{}&X@5C|Z8_{4-GC=U
zWUk;Hv_V25=raiK3tOHoXSFTlgRpmqKz<bpT_f@N!qDf*t+<zX8nLt|%8v5$@4?I{
zEPC)Jj9{qIEXvlTzVY*;N-X|h0Y)8X{(FRelEMtDhkk&JUctsY2LcXn=@!llnAIMF
zu&j-~v<H7{-HUhgf15;$qG0OVj=2iY!OU}taN7ZhAgC3LI!Ez@Kb4oe;fUMI=m75s
zG#&^T*p70+a{J}&!O7qVZ=&kO9pXL=VazQY(CG2qVVmT*Qt3!V1F|$6ryY@HlNJ~8
zkh};>pr)oK^Hp(r)E+NJQUUhEpGHlG(ocfB>q-1Rwg~vWbyI%0kB}CPYsdBJQ$A~3
zU`$wvp-{iUOVq8dB4omTdvuH|8fv;V@=zd^J{^JUh`J65D`prUQ~s)x)e$$-{(bw-
z8+V}0qVGA+$b0h<bsehMb(MG|(kr9E<n*Wc98^=*7htZ>wldwy!t&-_W=T|h7@|!G
zM&6!_;)kU!8M?f(mX?<5K~DhP#lnTc4YYF-+SIoDTGM~u<1TVwMcpwqO~KLs$c-}(
zO+bCFk$2pGEhDL>YH_SSUQOkyswzn}iMr+USRVB68hJd*Bbj1?g39=HG2lUBKoxLD
zcw=jL*9cR56bA5Ot&LdaAstWzq^JarLb7TMZ3@pRJeR|hhh%KG0nK{V$La_Z%z1(B
zmr!f3<e14qLVoEQd0aQ=edco4^;5lmii3>K5rI=FUGVTzK^1&IrK_#2jZppVX}BL*
zyp!RaK}0eUdq9O#KVORnLdnrEWA>np*n;&ism73q>9L2PcC2ZIqWrB&36vanVUt0D
z{1CXH#v)n*LJ2~hxH^HJWb?1uRTR|&Q95|pRlt65#TP(jyUE?CJzE$mO1P-JNQD8W
z^K7q=2(}L%Km-mv1T*|nSJJdChMJ4vMQ2hwmi0ZzyKL?3HoMnkV&po`tbh|%o|nQT
zlfkiLuM(#joONtyhe<s{TcTfH-yq>q=Y1vUEVs|~3LNQVyeoi)3sWz;ElTfFuO2J@
zP>4%h%WG?*2Ue$gN*79z+exy=3u32yuV1<HI-%9{rnEG!kgXxOh)ZM7-R$C8rr-dX
z9Ec9Autb0@=<hue=6}1lyA;QXD+3ZMVrE1kgb^-UZ^!UG>lT+mvFl^19rny#xKja%
z)$K&~%}3gj)h|`a96(3BmV(?y=OuJ&BBd#CEmr~%#9>@_aE_`Pq}aWlic&}*qLt&B
z@k1JpSGKonBpx~wz74e)UTz8aLyyXWxu2z*1(-eBsduU3#Qg=d3Wg3o*>^fSSLZ8v
z^Qau!IqMpS*G74P=dS(PaI?NEbYnC-al&CP;)#L`tU{*S<;bH?8F<NPv`K)sMFcmS
zFK7!cM&CM);(s)Szex@mj>6>BR8GXKAI_cGEm4BtEIw#@5MY5w`1sS*nB>@t>2)um
zPF2Jjse6~X;Z#>?I^0yy?w0}XnUsvzw;8|#a>v<(7vY4v1RJEJwDiK^*{FIvNq8PV
zp%0?p<Os6%Tb$d5n275I4@t5S=B~j)7?M5$p5uGA)n-_a!alx0ML|;y-Tt&fWB7UM
zBU=$=o^ri_8sxqt>a5M+H9UxLmjdvWr+y&htB4qWx%K2ggzCqb3u+~(7^0d**8J5b
zj3vO-Ne?;~j+t^?q?@m?{lS9<DzjYJ-W~*0mic$%`qOI}7=)U(Z`o3WD-(VI@d0HG
zmt!qI(ot+zw=VtpC9Q6AP%+ZIMhozC6`lkdMxsOk(87`N`V2@c$C||!Q?8>$S6saB
znVWMWL=A>XMh*%lA;AGFka>o{>uCOqVdd+i-p&570^9kY=7Lm#a}48yaN8B*tHVWo
zROG|7EDjnZ+?>RAoL!vmSU2VD?7Vs7#+L@YHk(^6mw+H&;PzksR|{Zdw)6x)ey<H<
z>fclwI}{k~-?00UVcU$#_hG4+<<U$o5V6>q$w=qVvBN=>xhuC=`MM}{=M?Z_7eBYG
z9b?C#2!!G@DlwwLGLrK>#D-xllFJpCvEc~Y`v!RB@ft`<4t6-fd$E+619|}*Q-|O!
z(9V+(w^NcYUF%S1kro5^e$C)7W~h#^Bq{C*!!~d_p`59$8_^)g8XU=2GR?S2KLMpJ
zltN5qlW;!k;KZ~Y|0;wSXEF{rXAy-FK@eg3bS-`JOBaMCi6@vH&KjA4QhIU<syP>Q
z5`V_4#1zT3UJjM!dh`@r;HP3lt|p;w0cNxZlXGy4<v4Tl@p+=A$M}j*IBOv9p;qY4
z{imH%Ch+S52vmrpi1jy&7ot&O%67zGRZP?n&0Y8-O^mhpjIJ&&*RdQIaHIo+tA;w)
zi4WUj#VK@69c`aos)C%I>2SHRidri*vPgy1kY+4w43`kCMrwJZ7N(GiW4H~`17N#*
z@4x&Mj!ZIB0evQ9?#IwN)bsG*%@NxIv7>Bj*|M7#>3LnSQ0K>G)8#|sQigfpX1rK*
z95626Kyw1JU4xmFQm+nNfC3*Pab0vg*Rm~NQqj=LBTC1~H2iZ%jweM%2Ys0irah36
zYpA;w0V<Qb05fQ?`gdZO0)#X#;g7Y)L<uJ$!f@b#Tnfd1-2xjIK4VnIdbAoSm>h%c
z=Um|)$KK^9Kr1$E6VO@+=cV{&>4?(>n$Ks?pEIDgl`;|C^d6fUO%ax|9f9p)lwvlK
z?iSS-`JDIy{hkij7jik$zPSFqsP&t1PEe6x1suP;%ut=alcr=MJqdj)MlPG7^gIir
zH@XC3=LUg?@@)gIL;5l@Q0Tht1KM2*()i_Qm$d*wOFE11Rl%ntQ<hM=-N17qlWq}I
zO?=Hb+L$oD1>$+@5~3IZ8Vw`q#9viDJTt_2VBJbIIJnl#97gJ5yb|^(VI2@a8n6+R
zz%KwG0ENSE3&TH<ta`c4VOolMlnqzP9^!eyLq`jm1hk#7M{p2ugZqbmAOt0i<nn_c
zOlX2|vx_w4SizDiV^U8c7KSVw?hqf?K*&=22Wh~>f+Ln%p58wb6aqG3iT0wCB`hm}
zvM?4BhX6O^a43U|;dwrM^60<iy8dUJItsQjFe;(um#yJ#u_}3N=}Y{qm@JFyNEcNf
zd{KLuHPJh?efJpyMC1+=>8|EpqQFGE13WS+y0UBB9EZXdb`+;!TR8o2_z><;(Kg89
zLf9tjfk&P7ER2T_A0|0<p>h9ao9y;b#c)sJszJj8EP6O60KB?WMphO-nKyVba89F=
zyQoBJljH8wL{yBP<!`r46oWe3zc7&g6}&f;f@`@h23YpGu_Nrs9ZeN(qC#{8Ucb%v
z74=>uKLP)P&Ga!KTf4s4e?g`5-GCDCnQ0L!*@Jl~F`2J_4g!-Pu2K90(&eKV&Gi~z
zHOBpafKx}j7}Tl${S>51%G&dPpihzPnbd#ZwJwkJ1^gM*SMIX(@(T+editSLR#(((
zis#$Ix`5P?Ku$_9GDi}gZu|RhuNvD?I(>XB*)xr&h-c*X#tl|q{xwM8b6fJK2@1w2
zzp9}6uLzkUE}2dL4W-$B;`o35V@pE>EM#PTycDZ_VaG|^naaw|EG*@|MlGGce-=R*
zy;xbf=j0vO%YT~Oj~w#V1T8xs8W#pJ^O@33oY7aqXi*rC3}ycZn{#2n=rsQb*oYx8
z`*ZBL*Kw6Yu}TRQAkv{wxa3(vw{qnTV_oOK_=6aDC)Biz8yCHq1PIM9ua5GpT)U;S
zT+rwkw1Qa~th*Si^ThICs6)d_uqQotg{Q#Ev~&&2QL3$*60;n2<PAHgn0#hsif*}N
z0rkpg#VkSm<!sy8O%!w{kQ_Wk@}6dzK3eFR<2i@M#&Cp(;ETX8Q{@>U&C6RO=Rmw;
ze}0wSz>h^2w)p_p{{2sxYNHfmZc3!+e0{1l{9WL7(+hI-dE}94pS1KLuie&O<<pJC
zxle3Nhc5*Ro8BUHBMykZ)+o7wYj^q1Q_J@~a(2dtm0qXP$e_M8@SdjagTVM9=Y-;O
zd*X9{%rt#rksymI+Ys<zxN8&_In&WY1B7~t!6yoc8L^GL_&L>j@~n+KrT`HF(OY5l
zD=7XAi=Q!xqu3f9ZhTI=js;-ZMkuqnE}VKffMCOCFw@b{t`9$3>bbi3;PiO+Xq6aC
zo5fJe(fum1MSzS-eR#Hu+DyJp#`xDe-$TIqp`Dc2ZU~1V<m_;6uIYHj;#;P&=j8ov
z@Q_6A1qU43S+mw)F^9_6HDP1XYQ8A`IaQxb+5CW-*$p<z4oe#n++p(Tj}Bs0Z@f5$
zVGoCh{@l2k4=#ps;9W7Ac_*_hhTbZF^$%EHv{1k8>dD#Pm?ws~h@ID4)9P|&R_H06
zS-zPa5mgnG<91Wx1ddrU(F-5iS=wQ#6peWD`}ccv6Rr^x46Ypoa8FL%K>J=JV*_t;
zH|~r4xE0qcS%voj?{t37YSrO;vrfov{vL*)2^zI8gvNRH;AzUYwgtMltg;ujRvPgQ
zIR6nmO004We7D@^sA*~^76qnGkR+|$h~2Z|M`;-jn);o_dr9LCXx7OPYLe&Mmvlq&
zIXhIJJj(AvNTmxF!)AKKYnM1}+Y-<iUhUEANT(?;y5MEjDi?6<G$vxq48$dl`UwuT
zT??4`jk;1k;+_sPax>kZR-!y|F9v4o_H^rhLiwVfJeRk$I196_`TTLSzWlU!HPRJn
zBy!G^iv(TyN0dPDY$~KEaku5(l6^x;BGaDVS5XDe4<Ayn6@Uj=qM1)8`f}Vhm;xGG
z8$-qEriR+eF;o@l$XYn`EBuc8qAg|i9<P-Xt3k%1YWkh8&|3rf&X3y#x5edd)Y%L6
zOW6tOhQaMQgf@kMIaDj^V&*;k!{X(EMq=y#-I}MFE(++w_pL;7dU|g~^Lv0=l_BfN
zA!Z^?>8F~xw06y!x3~TH+?j<v@j5R=l=<{G#cW23!4C;Df!Zk9Fa=N_{=&7LP5e=E
z<BJv@Cb(-D<i~%$;kQWOSoY)&6=LoBYAa*BMeJ4YRdTd1L(@bXqDrdKSf%(i`2cU+
zcd($q2N2SSrG?5+keR=u5%6Hj5$g)ZoFUw6KZkQ{JQp^{81z14Bi+dqn+V721w|>6
zkcNAfiT7T3&Fe0xqhm~UbD#RU<Lg9S9BRpTsCuTHKHLTp?nb{Sr?JDXF0UgNm8__6
z``g2u3v#JfuGuhAe%NfVU9|LGLs3ou(^2aWhi2&#pNvjnv=!D^+MQ<V?18ID(UOFC
zcwLJz-XS#m-8dkv_f(*S)(6^xoMgGX=S>VeS9G2mcs|~%kN8($nA)JyuPyv#sQ`{C
zvUdByt{K<OC@fw8e}Y${9PHlYGemw)(8Yet{qp4tgZg8+PXGbPWK4A6l456{3U#Z7
z(E{rrL5eWJJM;~3?2A89MQ)?>xw|FYs2KzAoSpwD&N41CRILdZ8FA#dKl}y<x`CC|
z9d3PSyYk~|PPUZxXfE&OIcw0A3P~-Ap*T{EeZGEqfgjquJt$2jwVteJDaOX-#rWAK
zi-W@-K_%q1PXQ-|YZKZ^oAwWLU7O!y2o%;m8fE{2TmXH{Y@AHN3QrZe<huI$T)e!z
z3YG>;6r(Umjnxqc(FQQHDD6F#o_Vk*UQWNG?l}F;o|9_kB{Sw81IdtieQYD3H&9>n
zH?JAiUB`zZLA=&=`*xhLq&VU*cXu1CQG&(g_idxqYv}9Ph%Etg`ugkNbFKW-%Vm>x
z8xAd3A0B1K0CDcmxDiY&O!w$MW7YoZ`8Y+V?8wcTyHHI}-pt9#Nq(P7Qnn@rn!N&q
z?~TfK<hp9;E(lW)D@FbY$wXbKnsG(A!k)6Ea&C{8#YyW4mmf&wcx!2rqFqTG2;l65
z{HvQ$K7+|Qc3k+Su;Sh|+xI_$NgUSqI=9v|bDsNYX$P)mzs4~nkI=J}Tcu~(Pgu=}
z49n)@?V^pPsl0gcXQhuDEtDS)0b<5?E75OTPY*}{i7gG_-i$v+zC;R6?6V_Z2RGRi
zzWg3sgPO@CDLpDJE307P!5hIlMU@S&@NsGGqUuk2pGtt8Yct&pkUl)|3fpt_y>_TU
zeJqh%c??p?e^lW7^XscL8se~X9&gEgh(A{<^>p8Hv}~sdAU7TI>!DjN#kCHdJRV{p
z)`ycdxGG+lJ<5D}&NYBtdMSGR=rwgRtr<-!?O4p1<k;lYCTw3EDqJd+0mp<z6!J7g
zfnUrOPs$8fk-F(@%g-M_PJdNRz4GzqN-(2c0K|YPc%hN);y2%qOSrhayu335T<%>Q
z`r8g%)0;R>$qzOOR>umeDBH@j4`wA{)1N!fP5oASC;#9e-zUlN6hH((@+ENFcns$O
zvjWT&d8xZtz_@02^DjvWi80`wH87k}KW9an+4?O!8rKdNfBN+Bsn7oYpj~5DS=N*L
z$#sgJ+PZdkcpWMRj9J1M+vYy|p(!+cRLeOTCQ>_3%4J(3_L4^n@xM!U^X(Z4rJNue
ztAPj*UpTbT;m$7~RSiDfdiWv`W%TB^fRa((UHRw<MG9rKZMx$<5<jm!-CD~Q=Kc0k
z;gb_~Q|?qWbX{$U>}abxk*J5uYMu2j8;9|A9TP*+NLKYjd`qovMF?IY>3q^vQ;{N)
ze5m1oopopZ-ACwJehdwH=z8ttMka=HKx_J)nxa}SPfn>M>#xkpOQSUXGP1Lg#+o@I
zU{Jh8QVmH<M`A2r!!*?tLWL$VA}#AUh_Cw3hk%L`et)n>mX?9((|4<JVxelKp}re;
z*$uQ=-_6>RdM&_lr1hp}5p>6yt&pmEl=Pi*RyMWx#dxFSI~KdMl>=YbLy3+;HS->t
zuv6<)R1l{i0Zo8k?_hcUX{QWac^~f7KLuXTF__CS^<>XSu`oQT>O=<zv;d!xv0zq$
zIgGFI);vXR?xPDIAL!jMy>}M&YM3RSh?oziAgi3@4Om`bY=faGHQl>`W+x6(qz+V%
z#U)xICIH^T+b!9CDVnwg2yJ4VYPm+r)Pi{BgDAN!zVG=u<|pV8TKME8E`KtB9xaA>
zGRk9!5RV{THC8#U*xiqxya(`C+AtPN_S&8I5PI4&MWE6-SkAIYj;A?~V&uhNh?zn~
z7}f6V<6Xxe^Y=}s?<E;_10bH+V1q}pR<ZoB+u?nevX8Z8nTH_^<#%(|cF@E*S_rEk
zULaBc))|_He2!8-0kiHw89x!9gtp8dBO_J7W-=^BcHlJ827@#*F;~>P2M7tueP3i4
zzBoC2ykk&<eeQVemkFiS1O_F0`wT+kR??;*j-_XIj@hGrsxNb@1+9zKH~W&qSCTr_
ze2}$B`=WC~wWXt~s;ePymD{}W!!Z-nSy8uijVIvNAPa+X6|NVJ?8|{(H=7cb!>w8l
zg&~IeNXt=?TmUpF@S!60cC{n!8I<w)amjr;W!_ZP2LtWl#7F6W4yGzI9VAn}g1afj
zs&+4Ol#>x|xB>T5xO5{{Bd_@OR#Z47ZWNpbA~_=OD5x0yIRw-*KV;|a_UnAoMk*~>
zgosaznx6Us&)D4LK!3f7|8oWqr?<dwnhiC`ZdU^|4}J7&5uG(|+Ya{t0&s`z0JoXx
zX!R~M+2+Z2hKDgM=+O;v=YhSy>ly$yxriOQRxLYsHT&%9`?otPA^Y|l=nj*vR)(_&
zyhpNrw<Iww*JoI%BksBcPhw^k<m+m~9GGZSBS_JWjA$c-zn&ir((AYjM<7%7kMsRi
zdLXP>Puz7J>#!%F215OAV*c6*v_G<MguHmQBy$GyKxURNESODa0c`M2lz;FoLW<Yw
z8wYB+Maju(J%O{5lAio3hX^U?c335WHS7Mc_IRXhMUW_qqxx}+KgbFpjaYAEyC=wl
zMzfIy4;%vhX~2v;#43YzaV!cLHOQpr7^dGLsYVB(9t8PkQ1bOK^>WUo6ZdA0z$hlr
zvUr@kW!@apo{Pc4VLV~D3BG4qc_D2Xcl&yn(fqos_fGs%TDv`a7Kwx7c59{&E}rh5
zo+prVL(`~(Xlv4QK>+g-pSd$NH1G3M&SMkU`QLrN5E`TF=LN_s5?WQ39hSqaQw`kN
z_@98uVhl9XaPeNXdNtKfQ|dzN#a9Qrzx*it3n~xIE4D<yJGj3LfPJhdZ5dG}py!RA
z&bcV9Htdv<=~CsJCK?$Y&$&00d-I3E{^E|{Z<dyq8_|KjAu^H7=&%XO!Ya`W2X@Er
z&z_Y&wQ!df{EiF@?ayjgu2gfiSAPeQMLk+&sPpbQOHVV%I{?s8(<M3gjX>Qtvk7fq
zvCvrkCS7k#vNW%;xv8o8ot0^h5jYOs=x%5ri1WtIdA)VAhp1?Cw$55KR5Xd6=b@2A
zWRcY2S-Q@i9!bQ5A$2T#*w^({R#p~jOQ=agq#KStR}mY!wspE|n*ppf+klXSU?Bj{
z8EDIcToG!ObGWOW9hQV^Y+g38Jo8?1{}v@z^-ZT-Uh2o<SD^*lz70qPjD-1=EIV)G
z8#hjSS>_xI3<%h#oP7)s!1t%pxe*_BsDh$wn76@HG-A|*e52;;jj$mP+$JVDp~+z6
zIi-D5W>Q155ciIVP5-jXv5a+v?@hEi^Jb+rd{vFNw0-|hzsaTLz71FV`^mkNgpDaG
zDYR%mk#@Vi0tC%ipja^M;nmS5T0}%x#%|w<8W^F^8HY54CM*gbu5ey5(6<T6XowE=
zU7)#ZViJn4N_GQSp@^l&u2q#@881(yCCaviPtH|$%ye{kz8yM|)KkL#aiUQZBF%^3
zn>TIPaD9Uzf=E=hkIh=+e5a(D7<I|&x0){Ouu(-$V!fL*pN87uJyos=i^ZR_dA`w%
z?g?kR$mkH1_DHYWbN>EexEX2H(+^^*mF3u<h&!z)JSciy{eIW47Aqf;u6*?7f(bNP
zUjhudjx{v!!U2l4{t`gHQ;-Q7szzx&eX#`&0kf3zFL#r?24HYrz!BP}u9+BXYw9Af
z7n~OCo|m(2qH`P|6%Uu+-R9-A`O2a!PqGlJ@gM&EKoQAo6}VgoyM~MI&yZoEo+b{m
zlVLgXfq?-ti|HLCu@`5r@ponS$+TMg&ieb$@8Cimx3in=j|x6^|2HFg10oY@WIj6K
z`4JY|%o}Ns*KU={*fFNO)v3-VOmD*a!;s*xb$etGX%uEG^`f@xH^=%EoP-e;#rrX{
z(Ve3m)g91_yx=swm9yQssyot`M6dD+@$WfT8G!DF?tBK9)T?(_A1pt~SUo2AyX&MW
zi?jxrsR<PJ$$F-@jdxPVYn>lIHX-F_S~9OcV@(Q1*h0lajshG*HP}!&i!J?+es`Jo
zYzPjnlcLrA>Q6}E72BQa?lkrY{V7?Tw$2t_+Vs>RFkvxQ=NlilCwA;F_rXp6-&f?v
zXK0_fxTkc#mZx=CbOak;w%*uHCvD$X$qa0+>vUxgX_-nFMJ=B4s^*(WtkrPSp-z5f
zVtgqzr_wX_+&$0uA10G<Glp3NpIlx1S@YNa+@kio+5J_~mj@f%cZm9<xkFNqU+2%5
zmZH11E!w|9VRppsDAjiZu*`bQas#LcLF1MpAeL9NEqSp5qD~h+u@$o76E^+zE!7!u
z5xinAKW<Ge4AAsdU3TH<{jcY}{iuzdy<R-@Cqx_~UJU&Ekpf9PO1b;vOxQ1<XKyx-
z_Qb1!aij&SV=rp)41d+BDzarGN3q7OwJ#M~s-_vba}GGn{^2~go}KoHh>CDNw6dg9
z0HGJY>`3)NuZ9}rB#N0k8{+JC?C#1{%f8siz5BrkbJuw2Q1(WLiS{?qP>O?m{DJz6
zcF*wo;+&8e<s6aGSc_EPY(P;FS^-nl%@gfk-eZpKL&>lUqgnS1UZRhUR`^v0(c0a<
z>Rrg(3Hn(kINzo06|*t@x0ylB`flh@@A?N=$Q<X6N;AB3@8;$?&QRHg&%rLUMomRI
zKaeS$(;i^T?Q`YcfUJY0xHJ7OR4kxKI=j1_P_;^SD4i`*^H7oYmG_X|kUN+LJJJQP
z4Slsy3^=36^mX>ecQ?vGi~tx0Ip1eqda-}eKG=9m$G?l37;s7ADI(`z;<6eTYRxp~
zL06JdY-ev=8%ZWc38_|kkL2~D446sOkVUNkF6k2{1}Mj?y}$Q)4ft;4))||PfgkO~
ztQ$pC0+7Qk4j+=Y6~sP)J(df-J(q}x>h!nc5D2~Vc(Z0-E?NU#gmRpCBM=Hd2lfhC
z3mcg>7L4RA7DsxpOg^E<D7U_C`vuazje_=aT(-X9ipz@TD}iVmbj*L(x#A@f5B;md
znhDYai>C8}(p@6y0?*#O<wh0z!Kn2W*qrwq2_<O_C}oH)7ix;Mg6Jf09YBa6^wCNY
z5y@G1fnxs%XyOADU?{K&VMXs?bmzqkBD%hyo&gnjaV%i%1#abcj1W|j9C%4D5%e7A
zQ(c8R9)Ys}+|mQm=m(lhYf%Ot1!8=ZmX;PV>pOL6`*zBqrP-*O6cYF)85RnVKSnw4
zrb#aerb1PC#IlKW&TuqYnr;9@d#wGZ?v^AY7IH7)Kt^5WQRcJ1Cf-nGYN*5b$oJ`%
z`Z(+5FNXTdR1Jr2|2f(E6jT-ofP_ZgwmEVg=vE{o^APe1buI9VipOG0Ay>w_tV;j@
zhScb}PchnI1z_%Kd&WN}8$GLl3vM_*tvNG{8j(S>0VrW+GQDF|Gj$E-nIo!nRkr_I
z&d{v~QC-lGU=nLK8h|g-^4E8w72dt?$O}H5@_UnITTzdk#WndF4O3R$`LSjw4;@lR
zf4XkY+;Iu_BnVU{4>1zSH8B5akoSFWMX1&LTaYg%LqsX%ec*!j*%b({>%>;u!maRV
zIFVDi^RjWvW=N3$4YGO(%jM3PVNE{Acf0$$iPvepBI3c21g0`Q4SPCOVlIuQ8%w64
zmeJRcjK{B-1@unU-ul{<t-o2w_<P93w#*ZDe;z^EHa;_^r93-7XugSA<TfLD>wPbF
zt=J)a2BQ*)ca&hWfcs?l|JY17)r6srRLHs*XmL+WgUij)HaxQG^@I5MgNm?;#3)8?
zU}H1dU<#z47$zW2!ndk<qgWtthw+p)wP=QC7?eJg1J{!Buj~EB%H~Gk(ArH*pJq-H
zDH2FD+wOzg^AJ(^gnmawmGgS4#2vrsnfQ|K45|Kwxquq_N~E4p=M_-zGTx&y{GEvU
zAVTQQGw^8#=5^lR(pmxFijxfsSwRYO{Lp6XXm4-dvQOUe5?CUHQWigBDFLBM=w$k`
z=kD&>UkB=~ThVxw;&E~#B;d}<FpFko>&f4)n4MNowQh-JZ;0$QgWe{;r533oRU982
z#z8001r}cqP8NC!jJg}Uoh2}9)F|RZP|Vn0<!#p@^2&JtCd0(Q?yx+$omf>!CUQqQ
zVnJF6PPpapr7n;1DsS%1^9-yZ*ieKG?XQhAK|{zaaQh4NAfQSW_G@^|e7CZ+q$FC4
z<b(%ZIQFj=zyoTmdkaOYGtGv|2+a54?k+S!ML3X7JJG#g&Zly^bQKX6{OFzQ5;d3O
zD-WJz!(9(yC1!m+>oa0lw@#;_!Y@+{CwG3_g5&)+X;Co^xS;$5toQ(?RRkS%gXmQr
z0f6_1TO@!fOx_%V${^iA4C6|Ae}6PZc)?wKSX_LiKlYq6W7F1%VoRPq*^-D#`MmJ)
z<L}K|1x3t%e?bcU6<m54;Krra@Qy_6PJI_%I_9tgw-sigor5lL;#t+3!|W8Vy%)om
zmY2?f8|L1T7<Ea(pzhrTc$`Vn-JMhejP2wC5kZJt*dB`^J@K8Hm8V^|qXHeo7<yb(
zQHr@p;9o)4Ju)H(j)YP47X9b|%{QE@OwjF{8#I#C1@K9(q^&`yE<$9F)R%D84+?4Z
z0AaE`g0MMzTv6C~B(=xD!2uk95g`0c0lIRe^5R}(ngryBQhBUfPORVh^(3%e^h5ln
z#q_Yg$pX?fA49*?oi&o%-PmE@?RGe~v*Wl0VGXeHUxrEMn20L8#qCeTb{Wcv7YEVx
zRRq3q!4N8H8fZngn-4bBb6%jMXUs>8KvTvI8oA&Tp#Y<@ZO0zscuGO%1>tSgc1_z*
z75g*BTO5!Wp_*<0*)SbYeuuD3E{=#vP#7#M%!Fcc`S2NL8{?Aaa(4?-f1dwsdLEF_
ztbN&*++e<69_8D23lA)hI3Uov3-G|`#yy^_@d}YLOhFt+E-ONytxzX!KK51unMb4M
zRYBab?`O=+aTvo(n*^f}g0tdUHQ*3m(W-t1jgEP(ET^-e#fyN2(~!cD!`sqzQ=Juq
zC<Qp}Mjo}a1LvNzX)R`&6F3D#^`<|aREQG!i9vgaG^ouLrwcQ+F&k+Q1ORt$u8uoF
zJ+BgMg+c)3k&60<a-YLh3_L1)Hx?{G-VzTPL@gnxUq);<m4`PqHz#A=M_%;%1ZJY?
zqZS=>CZC~O`>{t7`owo`L*G()o?X01B$O96a}l3U+V5t4#PGst#+&(Yu=Spa$&d5b
zch=h45=BFt{R|a~B0zDK1n5FkT1wg2=;(Fes>C-rvYnM<S@#1T0*p7fqU<I-&+Xm2
zccwB}E;aUYe|tO&zH05z`_xn&pFA-w$>+BwqAy|f%SYe6LA#DsFNv=)%YX-0)^#8>
zL`QHmP6cc3A-t%t49QH6g+g|kc?B0J8DR&ZR8ywTS|_@7pDHTc(OL7(W{phr=vPY(
zc-*|F(s<F;yP>ve{J=ll^jPj|e}6t`I7KXmEAVeO1xD^a$|h|48bD_k%wbUGca#fD
zw5Z@Z5iuWj??}-h$Nr>}2E(pYASq5*;K`Pyo-0(0<Ybv6*!K9A1%bOh^mOYP#xd0P
zgm=Yq|2nshrG0i5t@QcaY%3z|uxU?u<DlAzQLcm?YBueqKdD3#iGe3@W~wIslMvvB
zt20?^{yu_p&w~c+=>JIgdB7kIf`SO{((L)N&6@ld<<$L@JXTQ%BI@Z9UgmZ<%&&ti
zIQMqzD<GvDjJho8A-4=(z!n7Vu;VUCoX@u$XLS*ZvJ1l-v~c<fHbFGYqPu5!>45>)
zU+ntXmM)mF@ecJna&u}V)+kD2?htT4l7@~4CF$w#$uTwcs93yuIo4obLn;r1ChVC~
zBDi3Q92nja*abgK--oDaUlCWA9oSq}Ys%h9$+1eYWnzNB2<PibA}N1=)ibNc(!$}g
z`E*LvzHlu0LM6-RfRFchEx&BIu=H2>FCkyX$=@K|)jt9l8G|W`cLLAf|0Vne7y$On
zHD^YD_SbG<-(#c_zS8*H{nIUp=UEiqSG6A>{8s{CX~XQ%Bd&7VMQCo*{=fd3|DP9s
zsSjkDOaF%j$M)0z{Efj!@MZq_?W86D|Cbr)BQ4qv&=cO~a-cUm%+265uBK=g02aV4
zp64~5?E(KKROr!H-}g8tJpIr2FFH4p!%3N+9CQZc3E|bqp!X*i=+;t@wgB494^gN{
zC7G+f%y<3Hoo6`ow(eJ&slI$+92Gb)^gic=C*)`0$3O{*0^$(>f*Zj0v5Sb;6?a50
zu6o)ri2R};!GRHoh~e;)i5`|n)u)8R!omnEfgijRiXogUmr}GZkUCFKPw!y~TqB$4
z=;&ayA>qGA?m%Ni2x&AIH_&)sMioBvS`_CNDTDh?+F{uMf?ek!SqMN#(TsG4L~sO^
z8xFnWHBCbL&*4doLSntA^ac@_Sdduu0$d`#+(jJJICiKYwvxFd+%nbI+35%@6`kKL
zWN70iKT)dXHC_?^v)NKJoUT*12V^`=ClIbEXt`lT`c2dL<po?qc+R&Fcg&u-Utbk<
z8AJ>({x)D0sbzxihqO7rq=XJGM8xOlL+*<<1rzVwz_1T|2F~FJWGh9#b}vEv!Q(@@
zsI$DZ2&Cx8;NY8|!Z3o6<hedfUVuJF>(isvC=8JDua9#13ofbC=g+&e^8zjpy;1mt
zjJko+?`%0chH2yM>qJL_X<rn0_n?!+C?OgItM1;nZ{IH=DO*o8S%vVRDB6C9wi|9z
zdQdn}2<zRJY7G;#QUDz`j9DCAnXYn*ut(x`UB?q6vN$vWHIR+2XFn+e36`g*CnULk
znc+8Z(jqOEkxYny78@e|jhi=n^^IK+<>$5%Z0DH_BJr}P4qi?9X&3k>tPYIu%4i7l
za=iYurW!^OsxwCEAaMd;F5i3sn(f!r-p`*`{j++D&c*-7rqjBEnK_o2H!0?j<41Ta
zpvC#i3<L^w06GinF}!pYTo+{p1*datXZdPO6V%c#QpEx+rJW}hP0A!}3Ij$c5WO%6
zPyq{50p;5odEh&O$Cl9iUB<uOkH@<IzyTTK4!o=_gy8YfTP*6ugN7*xrHG;<B=jmN
z&*mY#Lv(q+cU)W?5zF%-O%_rQKIZwoLl9;6lzscAfL$qPhj3;rLn-bEbSM#&dPIxo
z452o?jODr?z4QP?XRfbIK&<K9x)q7Yxq9$!X01?RT}BnRCFB+6i*)|`IMW$gXzn6=
z3`Q2RH4#w+xIe$4VH-9NjOTy3SCExWa#l!X4(=V0Sr;KrfmqTelh8I?6EL7`L>e&B
zRQqb$m!d%r-QI}Lh3$tO+>0`+)@KOr15}w5GT8|PF@YY@rq|}$**FQ^qd=t=i3Oou
z8Nft-(62HHXiC>r`fVAqLz)uOwdtX7eGF8E(Qe)ZLK#5+@Iw^gnnk7%NpAowlHi6}
z6HseEA*yA0)R_47f3XxoFF-=R#omYUhDbjU{ZIt*0ZPQs_F0@p4<y5WU@l4sNH(nd
zYFtWfOL(BXm}-OYHYxOom}4G^e~;OhJf$)sjKMUcvMpuU<9hY6TS)-}mXSDjF{=m3
zAhFwhJq_h}j3Tg3(ETtuqD=^^(PI?46L4o42Mjqb@M6EEx}swrFF%~uchegI@Rn$J
z5$7CLAj*9?b_;Ah=wJ4h&i!@c*j5nrD5Mts2s3WKkC%hTPnSJcv2GOIA$+2R01=_5
z`JiEox|aCvq3`a(hI9>5R0~SBN>o=@r$V!s***oxb{xkKm#_+P7$LJMf;^b3S2^Qu
zkIlj-9}0?)#7U!d*}{eidtXtP69$K*G7C(}wrD}m3u28G?CHfzIYm(<RT`^jugvR^
z^c=|DN!&glZ;q%NM0o&goP;JoxHphyh&5(wOMwWJ-}?5wAV2RDEF*36Wdh8DRMB6j
zz|3CxRTZpxP&4-$m@gTe;0ihwuP^z%c@qaA#9$Y@nEgAFE{->BfU<y?*alkD6H}6s
z<Z)+R!eO3%=su*IYve&RD#0M2f?i7rnk^;lXR>($kC~j=9%onB0MRd2AP@0lry@=6
z7t9xwqi8C80LwznWnd1u7TH!S_rd>|V~FPy!H2YLJXuA^6%-L*<-`o{OWAAQF!I~c
z{7C;fsc%C;*Qi9kf(N^p2Jlc>{1R!ZScDbr2cWcl8Bd)DZ_Wo8@=+|U$wCIPeaM`;
zjBN3hxL|Mf539rA6pP-Ox(=ez%lJQ;cJpZpQZe8j_MalA7D$$2u)X8(oA#aBx=;QT
zYU}eBl|V4|owBmBih<Z6&Ny#T8HSLn+!0E^r@^y&6L4;zCd~QI__!ii(pUgWYvhS%
z@y`w-tX?CJJ#X2ac5k4LTgS*K9!G{xdV0G5NxMvg=9CC(B^**nzzshn$k`aA+)@U@
z*UVT9J5ULq5GJ@hZw0A%%{&dbete)#J_?U~DwM8A;awmeM{)`^PK|v37>0=#;V>iB
z<K39If&q!99L1{QaM1XxhzjN#>K%o@?_g;Ypjz@l_z<S5eH*`ox(?L7QIaw4nhVb+
z&cCC(j{AxaqkWK2wI6;T;|@eNYzYB`@<N;Kb5uO6o#ch9f4nC#J2Mj=O?Tbab@HBd
z>(o9`%Hvm(VV6*u3S?F_Ha6;2_0M2OL{U;L$8Ltrr*^#YnB=ha(%pM-lClu8`qYq^
z^_Z8C6gRkU%bXRj6r?a8{0dcBW24I9DX4*!90xG)MPcAFb!kO~%rAsDK7R6K;PWu%
z^7^j$#(eB<2(=QH&E!o`Z|^8h4x>m)R6<=p$t<C#7fo!RD4LYf{#*Yd6C29v>gqyl
zwLYxQ$MrMF83d$O33~J`Au!~ax5i;C#Jl+TdLWDwWma<&F=f7ogUJGf^r+5ANN|NE
zR_7XdKt@6{kKvWSyE6JmA#AGpUe^IGst_v3jr~8KtpwROj?x2Eyss!S`f4Oyo$u(;
z%NRHjr*Ru+2d;^OYEzPpL9bfMkXGPse*~KO53#T}i8w4}w@+b*?mLxg+@WEdHML`6
zpkA6gq7Glh8;dknBPSEi)e9**{(i0gxB}w9r4F1tZf#}7$xNmmY(-)^nNyAFBz0)x
zy-u7%w0kTNPw>Kl{(L%5iIb3Af*{U)?qTW>7Jj|WX+TD+Pem-vyDvb;7z2k$EEt;@
zoJ6q@lkJB;(i|lO2P!)CmK*NC#4bt2Sol+v28LR~IPl_;5|2w&Xil^q16HYG(Ym?5
zf(C_tWi(D|rnZrvKf@_oSy|tuFUkh{eekBpY<6=*LS$ozVK{U^6l6I!7~vPP;cNyB
z*#8;h@B#>~3xzZVce^AakRSvp1Q*wNjfnFhf(7e!%2;!u)DFJj=s4+8PfqhzSs9rx
zIH7SgX6|~6Yi<uzg|HECU}i1@YGdw@a_g%U%2Q#$#VK4WLon#07P9VznU$Xcz>%V1
z9s!loK8(eJyHaW-0E7@>V1TWu+n#Ai)MPjX=?27s*7GP3R%Q<v#6@^O!oLoQ71m?W
zuVRD@lF{M@>w9q-kqH`(_@8zgjh1JvGUmi#X3-WbF)~2XIvJ%wH|j}@Zk_i?Q0F<A
z>N7(UZsg@R;25IQFh6a-wD9pxf$vP)Aeix|=pXP!iB$n-%nhi>aElz_y|6j&Z=HGj
z8<aZa28I@Y2E^=DXwFgYNgrvitJ{LthrdmP2&C4=%efGAF03Yru+m6KhZP9f%&Slb
zML^p_1~VdW1%`t584=0nJHDd)ZOSm&1N8DUVw;GUgoQ;K1_b;EyY7_FB}0=p+yS&(
z9q#p>+NU6ng45;cQ!=23cHVxjpNoP<*7E?(b-X@WH*PjjYj4bqgWz&kke@}tgt!Xj
zh{&Oq@84M+A&#IV-dV!UNrqUUNvL*{4L&JK;Go4toW2yWm!JA(@%=KOHX?Rv+)Xt`
z7G`F3Znc$i*Ke2J8aipTh%wy^*-}SxPHtDT#d6WcVO+4Nn#lF?!-ordji}GrRANzk
z0s~x8E@fc<yl5Ce>PPgryqN+vxPK>>AorI<axDc1nQiwR$6jG0iQ~>1=xf&9BtCQW
zQRWT~wT5!iMw|q6#l3q43zyuH!`Z5kGYcHMg};+2yc!{^#SCH#j+iv`wCW&fFUOJ%
z(O;l_>q75CvIf8!_T?@xbq@`F#mW5<`O5_MM(qN{<|7y}IIM&E`t(}TjjM2y_G6Xe
zX?#XyhF6I0!|lx*7L<uXbF+NHxW|g&jv=Ny93ULnp=6FWC<zR@EJmK1XMnW4tv_Bj
z3C4j>hc<9$`&RwuyN|+k2W8L{nI#38Ao1I;CeY}+tGJt2Quz7bpDo`9za5S~T{xW}
zr6YC|2oNf905YMA!8L<O#4Cvb3}Y49YxQ?iy{=<ewN##~dgTho@rG~TzV+D~o!N^S
zY{oD|^?MX^eQ0YErd!ASC@ah6p-BsehC<i$JE!FVHjJ>asKoO6mglf6bW-=;a3dZ!
zmeM!QRtl~wy$=B>YTpC;SyAPSX6Wpo*K39vWp{NORRf-}qe$)&wOOiaS3f6_nqx;<
zULWWb?)brDCGAa(m1nnXJiFmY!sCOqX&NkR&u$#GKJ9S-0h{BM=;@ZXubV=eo=OSL
zFN*hM)AR4#`Ebm(;pRb6n_#;j;onTRf|@t31NT9+A_(+)@$%(qDXBH-F6M}il|ph5
z1zxcgmoHyFcKo=?K+9hcf>;(2-z{I1m6gC@!He*dzP`Q?AY{8I7Z(>lIzYNl0sr*?
zoheoxymL+pi6n&{-;NHJn%Dfg>n*-FH#0dbjM0$_8o-Pxzw-VrY=PH;M$C9ZPp~FJ
z8Q!5EA_FpUhK#{W89JTuvoRgsYAElC`wJ^2{*|GE1IOhEv~a5!7)prtsnKA^b7zDg
zU~VEiHZ8YG?hTZ_jOfS_CXsIx6dM!ciYuPDgwD&z{A`JF?X8&pBhEo77r$^}HRbI6
zyIap4QOf<fMU(0@b&uMUI$w_SP4xh3>o}w(v~S&dWi9h!)s}0Ds;YBu60R=T>+9*c
zct=G=iQ_-S9WH)oi3Jh+bBB_$&U{(=iK(a21Gx&<fIxxTMcV!oNQF)#!<+{^>(!Xv
zwmb)REL9?bL!;}4n}le<E;F+br07f4ld4`VxX%znbOzN}(~9dRunQq{W=}G*F~HT_
zg#9|A!NbLcvNQ7o{phc?_A6Je!~~|d3MS_yC8=jTuW<{jN}M}K2j)=%7Yw5MB<}O=
z%}YSeDSSHXVf#TNv_<s}4=3l!eBYR7Xkr~PMdV9GR@vL7^Rb>V58)bkj1n55vNZVR
zkC5yMvWCi0Z7%kJ?MZf!Gl*p0M6AJEw{C?P7mtA%E+)scG(prO2ET>Ja~1^F+H!Mq
z7v$%YT%DPju`qciEqbsL-<7{UOs!g>cv?CW4nNBiJK$HkrZcuW@`Xmm-P5<NBpOZK
zo<H9%40ZJ$N@q&tLqm<M`?A{pT%*5M=Kc)0Z&H@23AOflIl113W*)bP52;ZSU;#hC
zJ?rYNtF7%Kdt3Qv!R96S;dblb;5rKvKE~Ru4;k2DBcWuj$vq6Bp5lnka24k=#XFPD
zT0>(8?RwT#`uh6h@d9ZeIU@iuuk;`LtCzCh2E61nt{nhQoM{2Pw4KYk444WMiW2-a
z5@S9N_(~p$L(0gx@OL306k%cEo8@I;@6yT@FWSqZOUVaRgQVoGnCEf|GmGcpsidMH
zf@T$Bh#=Ou;-l_6w<kvG@#Dwjd%$;rn(d>`CeR;rlwUiv73JkA7_<wK;5xj!7q4Cw
z<0c&&sOP~#pgHwbR5#Z*crVo|3Ko(0jT^h*S~+NTV*ma%n9Y!nT}DNLROE%1S0SL=
z43F(A$n0T+8KZA?xxL}fK?FpEgoi(ZUb-8X`xzY+eWzhq_|es+Fgyh>^!|y+2fR<f
zMNw1I+e0xZ5jHRv6gKxt<sQ6ycL3pXpPHIx;*3wibxffURT0a!2hb{0G>7f1R^=94
zgM#9G^Z#P+&7-;O+xF2*BtwcsBAH5M)}Yd0j!L48naNOOEDbVdh>T^7h$JC%qQRIk
zp$L^i6bVU#%=`Gb@8^Ag@7`;#wb$PJkNwAPt>;;Hciq0%_qwjnIUL7v9w(qQfM7o8
zucERZG;WoVIe??cwq{M;Z$=;RQDzvI?>^zLGu8DXD|0$Hsy&wGsjpldrHMYBO8q7#
zym4bCjDq*BKbo<B+)5musGp661h{|b4X^n4&|t9L2NkpSu=L_u(pj;hhYus;<Ab1?
zfIOKo8s#HDq6c1gn~GS;i0?h+;c!`i0%tF4W^!U;2+j)Qn`PeUWCbK7G;9M1GDCU0
zhnsc-%F@!*^af_3Gc3*c?qiw9lX`D&41j(D5#9T9%A%8ZQ@I2n2*MtWgqz(VBSW8X
zT#3now`4;MJd+)Ge>UF6(P~8HDj+G@@~oK&2r3_{7FFP>ejNoz5mk+h5?crR+uOBa
z*;S3+V5rbpSC>=xXwq`@*r*<9GP+&AsQzhFr=c$s`-{z83P9vAakmWaXk-IBIy^cS
z78VKGEZk@k>@b&S+f%j=DRvqDJsHu4EW!WgO>TI0Wo<7;p=^%og-iDA=6Q(1@d=gn
z^$e)w6@j`jF*8e>SNY`@3#lmr99fC`Q-jY+$;=E!F-4pL0c1qCwqHR!CX5Dg#eg$0
zmdyoWXx9>L1<|AG<`^{`Su1e_2A92KV_~ROs{s@tz(+(yL1?gvlxh&dWj5`gzm@>(
zB5d7mU;HH@1SXIr^)FsjBO5z|2?wE9QB{>{bjITvps7@R4}>s3ScYOxtTRp)ID$18
zm#o7j(rAxbO6Fz`i~2_1UYO}w;FpwKiFK*JM&GYNC{qSf3Op2>d42x=y)X-%F(&23
zQDp?+jDCXscK}!djLmstq@<#`In15rxUOz3DwkKlc*qPfI8?dciuV>%qA5E!Hxsxd
zATeer3?cSYvbGjR%qRN}<8Iz`*&{dn?VBYkb0`L6frPyRM1b80pKz(fh;hH?>{aO2
zm^`?QFo1)Ot#5EJ5WVse3Kk)uIFXisUdf6!(4+i9LRS!xEm2<MrNdj_|E}cnR6%QV
ztZ(Qr?sE_xCk=I?$d(2J^Vq4F0yyewNC;m3r2sZ>=j4O|Z|2{yfez6LU-?XFN0V^*
z3cMGj6|;W}JeGz+bWs$_K$na<d<BVb##$IL+qk`7L{5$mtGS@P6+WMhl#iv<XZ&aS
zkq7)cI-bJ0(_8Z3`pN?<VTGh<XLG&h3h;Y!O?Z0!IQbcv5kqVVM3ath?{eh=9qPno
z_yyA&SG&0%Q<nidlhlnUa{zW~TKf7|@pt!n&T?Vu#0T}gcjG=Ak%{gMx9pqhiMt|c
z%y26(=!1pJG-kl^+3rhjPCy21{ehRSkda+j3aS%fA)#v6KXyU8(g2K2<n;k?mxaZ}
z)*>Pz>W2;mfBp8Y5Eobochi(hWQ3UN!ql-}*rp8|q^PEkA5GFmanai-F>FU4`U|)q
z4uYuTdU`fnw{BI3+3@f0J;`tc@c6x6x)vJ=s@GC(-#WyLQSuPDh<rLW26mwfd5im`
z(u{(V5yA_uHms&bwV;0d_zeRr1O;;!+|G-Ni?#wQRz$e?L^=PX1vtA`PA>U@L16#@
zt!=<+gg0%fBdSwCWBMrCO2=a5PoEZnwSBohga~m6#{EC1veCcT14a^)l>DAG*^Byi
zIW^NCE>~h1{Z;gl!v_npOH@heN<(8~0qW-h^zYa1rl;#pS>^_nRdHF888|#f@>h<4
z;KNtLfL=U`I`Yl&2iZ96Vi>2yAK>o9r>6_!fLJ5<S^>CL$4j$8rk0C*PudEEnHRh(
z|0!yQk%;UOkh0;dsIH|I+3?~;Av)MRbR&6DQL7`T7v^UqUESQQVY{q;;6MQ0i4Fk0
zoak_wn3(*AnLaAzoOpO>=mO-I2{=r4(6D2Yh}_90vR999f8xH26UrH5s3W<&w|XHN
zl}zKFi9sd?Gls2JR#w7WwlowK73rgTH9*Pm=EJUu!#GElR7nfV66F5n6mFkl9ki@v
z@Ax_&7}mYXRr3l74_`P9mY{RqVbXWD3-#Iy{OGu_uobcF+;W~zyNrVP^;*Fwr1T?w
zi(!ubq3tVQ6h6+$2`ks_(j41$$h}$nSE<&m3n9^QlWPC6foX^4;Cet-yY6K7k4RR5
zJ0u>4a6uERfZP=T0sZ(NRs|hD5_qiNGjj*^FQ+&_d;oZ!*VLpS0`A~t65wzH@aVvM
zNzqd{M4GVB3*r;UAJF{gW8@N$Vl2`7NJ&X$eAQ&QA5m4MXnXo}j9sXUqUUxLM}9YM
zumiQXla1nAa)d^BF{~~A-@uv!$G!hUWunKM^4D4tK^VAy{z`-?*T(-`75O(){vMBx
z$RBHaU8dLe-#^4}_VwS!Mx74MP-??}|6BZMM*08xNvw*V=28E*|M=<-C^T$OohAcA
zxFPrQ@>ti8jY(~nszCU(lU=9i$%G%CMvraYRrRf`bNlvcBDg3mTdC;ZQis;VPPXXK
zZg%!7ZZR<#5AH6_$}y&)U>NSplx+ItA{=RaW736@FS3Nx$puGzx2~D^tl9U|{EU@O
z%Fh)eJu9D#eMr%}w*naU0hA1R@Qzu%dNsqFJ<gx#wExZ?^EzAfP~{y0{Vu?muQJS)
zP!)t?(!pZ3#2EtXA3OKyPRtP48y+u#IJ!hGsAz=&qqv>tCEB%hUX^IwIrhH(<;$7?
z<e-O#*Kn=nR?OAlEpqH!0c~^*>JYzSZl1N#Z(DT!UdBKJH!Lyq%gt$#RhO<je;zC8
zF2~j#^*obnT7E!IPcLJ9imUId!Rcqgzs#u;%s%f8T&a3x`gcl7PFMg7cZJ>Tg{Sb&
zbKjqsdp$FU`DLOsEebUyGfh@Z4H_B8&TV<Hl11ewr?|3-SwpVCwr%X>mcl7KyC;uk
z)hfC3>yxDEfq}!C(h`REB=S?Y(Wv1CWX<H^;|s69%Ag)E%R!M!0q~y%+&*nTX0Huc
z&U<3?2RpgRm$ROnAFF6-$*mfE@atza9ul?%1%l9#*DzTR2rO>5CB1`Dd06Z3Z>5iM
z(kSd5d7Ofc$ZCcKTem7Mm=@Wx&FbYSLq;v-Fnb{*GxJm5m1P9$0e*4i3YHInALa;)
zpWSa#?it!}Lzs-ca+wJ3c7eAUrVi*)tt_$3ZkGZ0-8V3F71VBZbMp%DtKNuI5IQo*
znVOEgD`m%6ga6~lvQ)*37tCwp<1OB!tz}!myO-e%FKWl@*H>X~2fT*2KQm{=qg5>m
z6Ep9yd(*l(?#G);9HZ5s`~omCRo~FyqxAiwZF>eA$&q;eeFOK86E)P?;Ay9GS;chx
zTTr2{BdxmY;LAI7y^^P2Rr!}-5*stpw=**I+xjjSqW0}Y4o4~Y9JEqtzmI}~;5h^!
z@7opf66g(*gbtX2-X$5FblS3)F1=Z`lbmHWo%)uRx*=&xO-x?Gej&SiaHFuW1+-dZ
zX{tuub9n=UwUm@yhgb=?dt~Jj_X`!7w#t2jsanDqn1Q*Rc;P7e0gK<6I;{_mFWYqV
z7QXiqy^b6YclZ#P1Hjw`?m~r6Vk#748|w@V3<L|K#=6Tw$BK)|^_I=eyn3vYb?p7F
zO9Kpmxx&%Mo7Xkp>i!|16o|Hy7Ci;AsI_iNWTeKEwyN0gL10IF?2EQEeRWpbRJa^A
z>4u`ZGi&O~$GW@um_T(;cJv!@*e`{V)qwpP8X&UI-#;h%E^w1ZoHtUXeNP$hikf#E
zMy~TuFDjajY2QnBA9^J@gmZ|~_TZ~j{rT%rQGN?^-@V8>7g@Jq?Y2JcApDn<cW`i{
z-Zzscmvlyc_+R4`5EW%cnR5Si*J5xi+qr-NQw+*sft_z%fAq}LeSLkTluc{6Q98g!
zsU13G@g>`U^2U>PuETQ2<rF5lJM`Xb96F?k_U**Y1ZMQ5w^3$O-Znu&pTnYOJ<)^&
zj=nEWLA8y+ySthO4kqiryi5+i!^S4K&GXP9v0<A}`)7*Q{Hk%;(5#nx!rySGqoCPK
z?#)}aFh+-R?Op|lpi5KfAjibdNOXc9J!aqpMEU0~Z@jO@=Lx_BAOXZP%qh6Q<$U|s
zVb|`i3*-D)50=%drEL@{3@>q6I*$E(-;|VofTd2c316RaCySZZ!{b+8VKq7)<>skh
zU)il+CI(p_Nl-klL<?@2Ylq7Qgd^qQ_3&{*6N<3h`^9)AkIZs_Gwhi=VepK2MBZem
zRb9{Ql-^O(m}>rfA!$z_-i_8*9S7zv{gz6#ZL~ayu|hs0V<R*BlXNWS&TT*4#adls
zbUbl>{rbH8VtT4+?wcfj6q$<Fn@90Yj`7QzGAd`p#fNO|?%2gj09jwA7bs0$1;lS>
z%(kTb`T6tbD-;>F8CgOS5@T6E((c{E*862=PCN=g*hl34p*q<77kpRgj)NkX_T~QQ
zPpKO0ZytTG|8;Pg>$YvkJ5|rSxS$cP0$!=|3DDB>l6=coSe}cYS>0MeI&U=iahlKI
zP(m^eYSAD7!o&ntN?IlWQ;)Vbr-ESQeznL$SeQOXLc+E?gKhhc?X9;xJxYF!UfU)u
z9UXQ@Ou@&8<bZmuJI`YdtEu&(UU>ZeouZ!Vvd6NrM5BQ%MkWs?{iOr&vnt*-GT~ph
zzHV1O?CJ`|Co_o9f#Kh-=ln0eS{((H4Yx|l?$r|Sac=KNjvtWhh!-M)`3=KzrQ6^k
zAzN7r@2Z-d6nXMaPYA72pW}s;cz)>hUBy8r8SSABUzCuo#8YnHCL=n|b%%th!h(YF
zu@8Gzb8-eJI$BsPS-W;EDtP^qm(qJ(kE3l7T`f8F3_~rerlzKF%o5ZL@B^ezy-srx
zw`$MGNDg|Rq13vD$K$S~@P-W)1jtLgRa4sr0JpxeQBYQPI_n4P5(-`Wy{@ti8Y<J{
z&>>P3)F!O=Me&+%b9EgU80ZD+6CFHb_{MgL=40Di!GNDFk!rc5ocLF)m-XW)E4vyV
z9*E&*Oe{3NA#L|g9v)&*OK6y<Q$Jt5__!W<+&A>$4yYun-WUS|?*^{&`sZMrjEAhR
z@0C6{Zr)K1tH`v|!)S!dIji7LM5XKSp)3@rhy90Q7C@*(8T$ph0Kj~#Y+)11b6oI*
zv)MB6F5}#CCPP0)-aTYq@$8ZSuKVNhbp8vYCw}osQ4mJ>>U8FmoM;@{<a4U+UbX9U
z2CCw=!WCA|_025bkE*H3<7(k&L;>UU>1m9P``Kw)-|(`gOWJZA@;zD#a9YRQ&y@h-
zi0^nxlS{?$n-cfI6~uCKoAkGU)v>@tkY<F;exE)&)Wwzs!7WL!>FEIm+iYCE?b7{q
zX=_0N<qe3eAFAQ;-#@D%<mkg3AChLuqbpZh8R|)=9Nv734eBp+uY$OElF@Tb%KTve
z3DV$J{r4X#$5th-<9+qiVB1JnDZ62@gMV7oYT*kP%~5?D&0$0VEh;orG-zC^hP^4E
z^{W6p`2c!^Xxrfau0|Jh{fL@(0G(RAaO`>Pj{<<ansMc3G$^H<$3_ggCP%gu>HnO3
zDp?V0{(}AK)7$tOfX1r5-iTmf91VVc-tfC<cUoCk;KNa4$=`_s{9=38DOjY>bFbwQ
z!at%F1Tv<ie_>ZuRki<cm)rNSr%(4L?mw4lG<ls=yJ^CV>GP-U><}03RSlvwln2rR
z!4TV)8`x+H;K}6K<sf8dY9X20d!d&)vE7lpN?I6j=NkG(WK%}Fsf_#g&7rs;<rdlP
zi`e|>&5KwpT&)CleLFilWVhFxPJ^n3=Q7{P0bLbZZILX@^$$VUiut(^EXLh?=@$dc
zbyW5Bxn?hW5Jz>)SZ6oMV6o#;0N;1-HgR5$ja4-~d^qQfV&*ZGX{mDC?UPW*JTKlT
zxoO$$%Xdu51G$Z9HKmm2(T|sK^1NW-a{$42JxQ((L?|p6e1Ps#e+$&cAdiF%b<4&^
z>NYYIX;CigWy@pJE|9Nl7hMMx4GnqjSz7o7G#DBf-U+|Kj)KOAU4?PI!aBbFmn-hK
z-79^JY(tEG&;UC3RW03q>fP#!*idHZwnEr9Y48I~ICkvV-`S~c=@;DZ9BfM`#ZK&5
z8t<~^W@2QHA1Ki7c@P?m60Pm8vlOZWU>Dur=ulgeV({||VJFBiHZLFk0edH|g4KiU
zi<7$j;6W$`f|sw8qENC0!Z8Ol6dfdJ(08uCA$)Xx_`5Z92_=}M0Sz0nzqz^F8E2Dl
zv6#fA0FU^Eh92?H*;LYrB?-GO4=|*qm7T2&yCo|iF3tkPWAP-6jjf>K_MTb=pHU~6
zvZVZ8hsG^n#prv>KHsI72MNr~6zOyW^e^MkeYH$N!pH-yC&`gMI7Y|L?iCYT9wI&R
z?peb7k|k7t`=zSt54@Ox@FySLzxgG^)8iv=w?SQ_kTVoydJOoz#;yJ0+ndTz73rh0
zUDBD|*YRd1X%bZ$J>0V^osJ&OKJhR(YY*foaq;n$@86T@np#wWuVAq{Hgy;NH|;jp
z(4_}pmgc0DRb*`Jrs5<F_TW`=hD0ZYYq8DqRGt^r`}F65A>>vO^-<3!fUpPR&EBwa
zV@(Y6JN3hdub}ve?R}3#3VUNe;Ufv35M3EoGFQjEOdu9PiD5x~>9ZQi(zLRVAII%K
z!;SO?E5g&KgOPH^3Rwg`G0=CeUP#nlyLbIcP$8ArX;U?O@u!s6$Gth(fL<9)o)k9?
zwtz}z#YSe}gNl0U1xB~=&3B#uBbK!P;>4R%4Sw3nD$`Du$B$!mST91<g{e>et<!(d
z2z2w`3F&cW$|Eb1>TeWjy^{HapfnIR3aAyB4c9K)2+a>4k&&Io`GwUJ8K|qlA<7O8
z5<vGh6&wh_9f5HnK@%ghv+c?~6&G@0wz}r<54i{wF;ZwP@OO9e;wNEW^x7dS$8aEF
zeSq@1!voTV;eir>ro0bDI7gWr)!Tk=CCc-X0aH)U{RM@EKz#`c`SjA%Dip6q9ss`x
z0kH5sk7kX8Q3ai(B+Om-&(2M|u)O)mzJA5VjT?*XdQ67$pL)!2hN-x|y_1?M;_(cv
znRA1jPtu=1l((x*%aXqZZe+scRq+`^s4E$2z#kBCJDzgiNc=X%(al(AQ|f4%V(B~o
zt3cmt)ku^Z_IvpaXQC<6Kf7I+^Y^{RTCHOj=dYePaUwW7|A);h+ZWqZM!mkz6x_2t
znS`w;_1{0ftEI<*-K9Hn-c*OE$a-HqW?XFN<q90*rwxq^?C6=&$C!pgc;wat399)l
zs+ZSQw+x;%T=4bA`%9?C=XG_K9{D?35fr4ItLTdC->$M-nCE#?U6pt8>oJsGckl-M
z{odmx%DZ#wx5wXc8hqyG6RrhfP>8@dm=3Lgv{SD<{7l(UZPlH0bcEcE9%7eO(@mz=
z7aF4@u3xVLF#Y0bSbDlJ)$6fED3{@WHX2=kI~R_|`{MN_SqIM#53UCDfM&1^JF`UO
zxJBom;R&UYC1Mgvl5rv;YNxGEp5#Yr2BIZ+=U&>~yQLrUsb1tkifR)+^4ugBrnguG
zRe?}>J0+!(S915P$@q&`Z{Bd6QW}4G!c*(=<k-ih6qaJE+kPlRjP+`Mwo3<!h$ne4
zIOs0(;K2(d_!&_WWX_lQ@~X~Q!y8ij_VEg&q_lQ+E)1W;OIDK7#!}N{yk>K$laor$
zCI^kD=2~(WbkB}HQJMUOV!~GQz-zJm=i>&TS|fKJFYuj@N^arh0S=G%jtmake|XHH
z+tD%9eV<+I{X_Okin=mk*XbAs3p^J-l?A<NZ}%%G*otdgQQLLCT+;W?t#!PI&&;?l
zH_+A73-t3-zI$U}@gcv#yBnS`ZRm2oJs8mQg~>7Aop&<QKvaxh*z0g50;fFaj4IM1
zV*7Kze9nKG)&itugN&M-a7EDT&b_?~DU;rS9`kJ5=m9axcfU;No>5Q?K78<?PeFl;
zzf3Wk%^4@D73+6-gWR_mE;`li90*McmfQR~=T_TfP)IfuM7lzJKxkqB$FaMoSN>gK
zVLTpu#D7zoh<4FRxN))J#xm$X<6f__jpy^gfD}BLH;af^07ln5empC(Zjti6F}FHf
ze3PW)!kfpw196I5&MkfN6dW%TuVm6*(H7^Fzri`X!F_BkU%EO(Bs)WAeqO)CNdPl3
z*HB#C!4Zs*JfwJO1z+>1`tiinRKe-eQggsU=;Cza4Gq^Ie*iD6?Xk)}qgT+TJ8<cf
z+q+2dLnGE}Mny!qIPI*g{GB&;7>eZ#>=SSK&P?6h;TU!}R#Ipn*$a7?2H?WqscB!3
zz+fTc_Svk!VHV|8JxGua%5&|L&zwCpIlBO~;=L5uv}y23&zj1gWP#mg1(WVaWo)d3
znhg#P?)F$z&`hyCn3$}4kg)Sdc_>fh&FjAH7T=AzEbCvssLqO&6c86W{$))$kzW8W
z`!PG^ih*8+=+A?L#KIZxqGYGd#?aJn3?J*5Wd>cH&v9JrNk^VVmrQ{D3x9iQaQP->
z%QQSiQA=IGEHam6&##ZBdD5y4Z`~?lTecxlUfp69pZ(1JkGFKdN2}%PtgPyhdQ-J*
z*ZiWiU+r(|{~=}5ybknJU&l+hGDInZJgOZ0DgD&;K}GJF%b=2387n4Oo9;t;Ab$SC
zYfh8Z8#F}<byZK%UE6p1^eN|WwK~y29N?NI55g=`ORE(mlgOG!Ki7;~#onygVF$D0
zV6Z(+iPBdLm2$*pQ$HBAy?!n7OJBt0+4Tm!+xnl1H}B3cS{+zabT+!@I1K`|*q<Mj
z7JPh9FZ@`7d0_??78S6_U`F~LxmZQlXeBw`EqLC&ruW4h0SOWlyU7XKa1sJ-P$wpz
zM`P2$17ihoD`Wx1c8ipxcKg@Vm6xgyzb)y-_p}9meN)CR0F_GYI}NNtC5z$;ZJMuc
zpd?L72teTwi0LMKf-|9C;(~pd8fM%+(2g%`%c3*Q800*|f9#l>n+(N&uj`Au-|Bl%
zI6yYnj~{52<2t{?K*d8^RoxW$x4eTxPQAvPCBDjTKi+!Y?XxzF-s<fW(U_^u2M-l0
zxka_k;>lARqzveGvBYWi%flU_-vae%^Y*Uh;2_u=kSdbu6|F*~{rvpk>UV}J6MM=)
zO@8XHDL(i8<30oe=*jXLKe4OtEOVJE=$deh@A&(*ZBtrWS_Qb~4O6jkaaXaDJkk}f
zgH%-;J7Rm+lk<{*{esc4_9ASXK1)LB9TE}}fffY_;MFJ_YipOIzKtf|6<%#0{j^^0
z5*^i@&Uqx*GaQ^c>xmxCUQiQvh2=!C`jObuR-t0JefO?E5%8W`=j-Tw_^>){1qc@9
zFp^F{U?GWTzP@m%EiFURbbf_LsTc}9vk;w2!Fv~>uyguF7o^b$)XW2vOTj^BU|t!R
zoXmH=8FZ8LOow>Nqz_2;E4WSwiG;S(3hvyw=5Vmv9^MmCC!>I?09qr9vLMD#+?<3Y
zL&5@8@nEvM<MpbF*dyN$l}UnmP)2na6&+nV>ERc(A@0a8a7T`(>k2X?IeV@i=mlYf
zRL;8R**?_CnAr#n3)3jw&qv5GO7oGev9n_4A(a^?V7cwD-f$Sa_pF?qBDFL|U?)9j
zC903!py()gaib5nKqsPDhp^~_3DiDd22{nzkLxjafTFEm<|`{Pia-Jg!!YFKpT6le
z7<!~d<N$4`d^za3B|<vdnJ|ToGLmen1-5hHBNmJ@aDKzh>BCroqQM|uQ1ztkK|J}J
z{tB=BTfk|KJ&`fureS~=$AA3r0X~P8kCaHJp-NIu-;{Sr<Sl`O%O^jQ4U>MN9?#HJ
zIv-yR_OKfxkV~lSYh``+@7xPSO6db$*JTpb&4K3M7ybIp9cM(`w)b@+8m0gma~f+?
z9XP@kUPlxUk2nwO;7rP;f<($yTczn_nw|Y1YkSgbsM42E^9GMJ(#XEU6ot8IeZAWD
z@dCAj2L&)IVefl==8kd0sfmB`9KqhvX|T@-p)rN91rtIkfUdiD+g-Ur6496k@3@u7
z82Tl6$9Ur!&Bsi=%rR~{)SfRy2H>Hdzk?Rl$ivIa3pc6q!%G(OTB1k=%(0>9)CX?p
zZl6QgL{_v@rf%sFDbxT2Zw0PT2NzYy!pUuW7|nralxIIR;Ub=^?0K|9Wm~AmwfH3T
z-6l=Ra>+9dEiI~8$K0Baf|G_ySHH!C8Xy~C9zQyN|Mt~}GLE%4%hK4}P71)fI@JMg
zNM?W_eWnkL+6cTdR?=7zm)7|3%RT|ox!IDk$0@cy(@<71|2D5&BLj3A`>RznG*;MW
zKah6s5JN_)DziCC_;NruJQ(V+8<3`zd$rn(ijzIV4M_<T@)l`RF+^8c_*|L5tHZ;?
zafU^_g}}%$LBa%cp0kDdl~*H$Qou&<dE_XB1=A=*$Ho7I8tG_iY56Q<>R@T;@=Je)
zd{ALkR<yDL6R1oN<pCl5_sy+qR7Kj*QU~x65jSF0d5w*Kopbid$`XZ;Hb8C}i|@58
z(U0X`;SUUp$y`S3t5=m!yyTs2ST$x??)@gq9Hzn>d84DESXZwL0Srh!>d)AFy)P$L
zV3-_=ysENi?#Pf!P<KQo#|A<KNzf|j<>leoMp0mcGRJ>IZ~*ehf1Z~9c1$o6(bAx0
z!|>+vXY1CLTmqgFU_3b%CwoqW078&t^dhfz<$LTQVyWbfhf7!tL=;U-lH9kozpSN)
zFS(|UP7O#pvN`QCg4DHO@nItlXJp|Dh%o>aD!aKAw_3~i3RFW}r?3)YgD@@{u-h;_
ztVZL4L|JMaR6giI<1P32Y$fy$`BpDD_0Nq4SAsCn(qcpPRo%?H$IQ&EVk4tyV^<d!
z=#47$8|_)9d=3r{xJ@)vqU<>)s_gw)$3>FsCYaJKzZUwEk_evfW7pB^1rtIjLWo^U
z_S}pkX4y0|9PTyWa)uW?fEA+QySm?PIjUA$2tnb|%LLJMjzoA}b<T$;`7p-x3-mXi
zDzGAF8Tu3WW=KIlm~#J=U)0GnXULM)b>U&PnA7xYMDd!31>z}|$eyU=ZqAga_|U-e
ziUp=Ti8GU?blE1MfR7S)d9FkKjH@Tc7B~WpFVBi-NJXehb=?)ijwS`XyP#vVf>aNA
zssac?)q1NnA4Nb)r4eCr0s;>B&p2VspN(^U7!e?0+>cycP!J=hxJd|D>Yxw5?mwQ@
zW|T)HNZfmlsZ3NCtI(H3vaunl`;s*}FOyjCAacggfe-O;@$cHjg>Aqjdtp%tEQ~6P
zzP8nj?VWu<^cT6g;V48{F$nzY*I8TJ(^ZX)OdzNf;ong9?f`BjnH?Q@|6~nKmR|LT
z54;}6MtS>VA7B2D7NDwg4Kn90ZZ;nnosm}q=d%V`(|KXu!w5@TAU-~TNT`vxiw313
zU9`Se3qFBFADj;8#Dj65qPYW6CPALU#>(mq3e@6V`+Xuf!*`C6Fs!|mfPdCfdGc0!
z;1Vva>e=55_U}sAaF4El%Ys1es>-!B%+I8u`mZVVTBtcBC$|o`{(E41!NE*7x1R~V
z_;V188On^xd7LlDh$+m`v*g(gYS1lPMg%?F+}woYavS~l(&NgBXlIOr^Wwpf!61x}
zl=l1n*0X<dzAod{oqUTa7qZ<RC;{I2N_ZgomA#x!np{eC(9x%!OOPQW)7VYu2QqtT
zc-R8c?zoXr<KZv-riD(g1PKWW_{<+1{kq_d?nl?jtw2}ckS0WOe0|H8-w8V{!e_(`
ze38_z86SRtndr|NMg*in`}H{!sz`t6KrGVYcF;0TaUVHy1R_{V+y&yCL0l1{5#ksj
zT^BiTi0GwXV#~%UaYB2{6>oLKv9R9@H2m8)(BWahr9i#({igCqsw#yojajK4U05PD
z^m9Jkx9*Uk^G}|;>32QCts9OyVEj;@kDKVKQVtVZU*n-W9Pi(q(66YirFZ``_Ta0N
zCl{Jye}K|Zu7?7|-od+Kv0{n7H1kLN#6$ZaG!UqQa?`5?(v0B48c@v1NhZw%1%*I@
z=QrQl#|*>)U740G{ThLCrpkaqoxXb~Xd@z$jq#F5VS(g2*s+3n(QE6QmHP7X7BgNW
z>`H!+u0%eF;l2y$nVC#Y56wp=2dAV6Kp#F1BB;-cN2dMkE<rFwWV=-Q%bVsMBX7y&
z$?ZKvtvdFSQ_^HP1_@E#3kV6_Nw2=|3<`;$g4vAA(|8vZ1`~%dbIyzDNX<5L)??;`
zs>CWv2rD_^*J9Mx=S0M4s2LHJFzvSKjIEN|Zp|eT5vS9kTl`>d{ujj4feV*{Kd?=n
zcg7MWOhPRK-;njI_Fu31i0^Qf+HqlWe&0I)C2c8A*sGdnH|elpzq@eJ8cHDTY!N;G
z;@AA=|KzxEXZ~C3z<NyL*bZLmpFc}GD{|TMNT58!v17+YFR%D*JRy{!-m^4g;X0p^
zj>fw=hyPMd^n}&*zd6g`r#-B+dGW8=H(r^RoMOYJ|Bs*jKR>%%3b&1f;eY<l1oIub
z|Mk-TFaPlbnJ=w^W2K$!azzm_35gXrP5@{ZWgvOLl(7M)D7~6LO>y(Ta*_I@1E@p^
zubPu1L9kkMg$k&wH8eCZnQ;0@;AC54eSHOl7DOJlKba_C35tW{sOafgrZ$H|XAa*o
z5|cwhmY+R$uJ7yDAiTHH-!7r^@BMMXYMu+YUgCKyiNl#f>?GBu?dvpzALC#Y)sc@;
z?-E7i?fdr`!4o435@9H+TUY<@M|Cdz{K7)Cd+6pM-lZs@b+53kf{@A(Ozc?I`F#w@
z7A{CKph1xOb;o`>^pu5SXU1(P1R`Q%1M%JxWiv#f`M*K_ky!(XkmRBNB_SKrp=jKd
zhXFh<Dghy#Yk%`5B{kI-q!N*2L&a=)m+#x75?IMX1!QJn;Y)ZD@F}hB?aHX$*f=?L
zg{0@gh_DY@eu7*^L`M_JCHy6UEJnNB;#;?F_D!b=P<!-6l)RhUAH8oeBh^TOTzlNa
zFdIP!d=Q$2s(0_WK}2N^zt<AjTg#R8is5t@;yoKO2reJy?ssw47hjSzZH!9QLKM~t
z-Vp#-AD$wr3KdW*d6-oMP}K_!zAd`yBZ1S$@j(i31ad~#R^;%38wI!unr3)igyTt7
zzI@4oD)Y;*X^9gJM8HaDFfl?;?l_#6;B-Svwy5D$E;ly*ur;?7M$Z^KQ?)mM>=A3l
zh^7Ox+)j0vS;I=Hm4v_XTA0_=)2k!6(2q|lD98mdGm())py-n)d;9u)uxpDSON(Lh
zu7Ioq1F5CsQ?rgwT|cAS-2*^dVE68|IXOAx8)19}hL-+F|Bygc*3>Koh<v{|u6_-W
zMx!MFAeh|T+)Vck;yQ=m(1Yf)l?X1^uI+<2P*<VNN*oE|d5!GuOrAgS2FFsG6jX~K
zHkk>*fi@%+tvke%DyTC{AH(R4WGM31fK8CckOX0(-A1}-P;!&Ri?r4=7;u$TLg*J1
zR8*Ei-imuJk0x&(&ffMPpRRT6DO17xQ)vBsGo+;yygy*@M2C46WmFxJP#>d15<)8p
zZE0S9{+7bf7qOE4P!QjZfof+x3L&zEP+Ldm$MkPIBs|!Ytp)+e=H3iayR>r_VrcLM
zgroo%s))hmajal~@Ka#V9v;l^?8Nl^@u`Gyas3^qu%e5$0zy;+Y)QQT8*0%?wD{#;
zPap?cVeoyk-iL|``3^{3-5@k`d@TIW8`g6i)zooq?URTIR`K!iA=0&=;ND_@n9DNc
z7Pf5An4w$CZIJRh1em~v>cUPAzc8=)+4!~G+*>IGpRhe2u}g~H8(;Kjh9~X?Z3PO7
zzej$A@X00NmCVO?7}LjFMAVVEr6V|3-9J#Rbl^1t;PetV|2c0K)I<>du}>uyVy6(j
zH-h@N1dD7k5=-xmG0LsATu<PPNLzt6patq30~EhdWuikn9s2D&86u(ge(3N4my7vH
z(|v#d!!hHLK%|gxAZ3U~wGnhsO-%}eFEPmgQ&%ouev;~f1E1e8mLwzrEXYTT=UNA9
zd@c25U|@sG88kKY-f+6hOI`Up0g`$tY6OaG+#s?$0&5hZs2oAB4EE$2M48vz&j|O_
z;i=(2|NMF5*;!QZM2d>~Bs$)4n!EC+IlfbOKF;7)${rT6pHEj~P@KjLr*E~aC-g}W
z!q*@u`6uzbx&Q7S1t&;ajFR;H=EjsRlA}%OFeWDZ@Chiyr_P7|vwJ7P@yCyTn1!CK
zz?*~0MDR0?y*~&t(#P>XPUf*(j`zM6aJpYuRz^m}t5<9lv0uX|GAT|9F(5R4g@uKh
zbKnR8=g{HUDS$7yYq1*(<2rJTuaQ^>IY2jAwbl#ZkZf*1bS~`&aY78UQE6#m?-FHt
zU^C_r8PIkJCc7e(Ye6_h4oEilrYm4PpkID?dgC#)AB>>w&<@b!Q{Hg4Eyi^c9$|YE
z0$2u`8nB_HOvdP-H(p3h17K3-2*)T=7?12&$-v+p6T^vBw}RT`=-}&{nw0U*ohNS<
z#qSnif@4QRK~?`b?2!ogEd@xNh>l=O7IFvX8R4ib=_a!OzAOUp5;QSrZ~Vg&5)v$t
z-JtOhJI6HLNd#_i)PbDuR;yYwK9`2V2`ztj07qr#!w=QfN*UQWE|_Fefn*g*GvoVQ
z<vI_AsfR^HBv8W|h!xWZe<eDB_8?e{oIjPYnDQfT9)mm^!ks=)cmP!IcFCeWML%)>
zOFhzTSs7V!i7r&7k?z}T##P6C-FwiMMLxcM{W_`m5aNhC15+JnzCdssRjsY8zy#Nb
zuxGy%R#W3#y}B1?bhDTkU3eXPrf8gun1qj@sOU8Uxj-yzlzLYd1FUN(zw|9@Iv=02
zQ;(CP0S`i+f&(^7fE295K!M7%rUBH=5D|Gp7#$o1!lW0kIY9O+L=o2uUG)X1ieS{X
z0(qR0zjR3!%^@W}@D6KXX{hdUFVnO+Fs=-5?n~e<so*)^x7&biP!<5a-^f0gKMz^D
zC<e^r3DDhCRkeh2!ZAMda36+mFVFsEL+20?@pnPnpn=W|>6<9|z*R=d03LFNA0>tm
zHAi+3-fiPXot%@OCN&>0R-nQFNJ)eI%C-OUE)<n7JTP``WA5k(*EAOJZ?`ezG1OCD
z-uiegwx)qjQao^g9t~+l|BtfAIXy_fmHj`Q;9E_Z7L=i!LR!M2&)p~mNWp?Dh2<TH
zd3)AVcaTHC!V>di$U(`UVpcWh<tI$WT>+1zq^GwA_XD>}36gBG!7=4hO+T5KKr57z
zoxL2@71m9IG82Nr5|6*Nt?de;AxK7Hx0fMBMcZKtxCSGo0r=8UQBkdUP=LR0-@e`Z
zZD;c0Yt+6AB?i>h{79du^n9?n3nDVTHuJgHSnuV7gIa+%194y}HAXuRg?der5Z5h(
zc(y~3+mS25ID*Q7iyuo}m*JHHPZfZHX7XLq)9Lpoha<=_kEV;NUBCKXnU31&JC4A}
z5td(oh>`++0Mip>G!UQ);BR&<6dYLBvkO+1%^PjX@wB<gI4f4Pf_ACG#(Por%hrxS
z#4(B-4h&Th=Mf?rRE9zIHv0H$2&`rRUa{q>25@c!uw*0>BGtkzs710V+x;b&^b0u4
zudr7O&4#m(@4?^4bafB7xD<35y?*rc4_*;oMEU24jihQ!*^2dupfSjV%)sZ*DOp*}
zgs8_eZhvTWI5;HaIciC0eVA0R$*4|$W2Rk>GlEP!o`PkFG`3;UdoQJ@4{m3FMBj_S
z_c%-u*rAd&gr^-b8BIV0gbS`EqHVV<8*YUa$_YzKAAKX4GU$S0t}(<8o$GSy8di^l
zNXOi~*+@yjvWN^16eGR;!7xez7)Fo|+wxsYu_|;e)d4XS`7`Jj_NL#vSBE~vuVd+M
zSO#H6%nSsKJ)8pUakuSpSky6UMq`GFJ0oL-b$A8Lp~$nxT^vE%W`*>Qt3gdjr2RZt
zG)%4n1-unKJ$*=Be?B6%>nKRsLQe4$Qs7TzU{q2JAg37H2d(YwUQkjtHiIs+e-;~+
z`fx$aZ+{&clA@MV<Vt{<nE{)$LdUj)_oHM#aQde1?nIL45qw<hV7hDY;Kai?oJjZi
z+0xeCsQE1r1C~?FD^}#R{yuj6czi$ry<J;+XqO8wFK>Ewb`vqjBf1aNUX}R%O_*Eb
zqO7s$0|yCZ2+S0bd%;dyyuT@Ji5dV&T6A^)BjN*H_~?-hGLg~38Aw`xPE1S>56T=n
zM~Vb84}fG~0O$ly6?>gsa%uFhgQ%FC0XAy|3X@G_B_pFbr4OxIJci#EpYJ0_KEr~O
zjO6)JU6>v6Yes2cj&Zx*cO0{*!1EzQJ0$nVk*F&=psJMM#5cD-b*cgT2r6(Pr#Gbc
z908)RmMTU~61i0&th%P=Zq+E7kgWi<us(24WC1`aDT)*a@Q;Y|_w&m`bl#??r$_IN
zyGm?yV&8fKZa=Q2WsYHx5&S7fSZ=1a-4ab*-5BXV8@E{5+Sa4GGqA8o)qHjx>nQl-
zwj#;Gil7!J+Hd$H_G^$Q0)WV+4*Jivl(nOyX-)~Gd|<~)HvIx$?KhmX=e)*A>DSn+
zMDLBy2zUQ+G9(|Z%i?qm3rlG~!oCgW@&>}+<DH2|t3zG@JkM1@cLuKI^{U)EIiV3T
zD^|u&Hk+>{Ge*J_T}ye3w@Us4yQh6YRLMgNLtgas`}Z4g^SG#V@akA7W7yU=_Y3`1
z9Ug2Qj_F%q|H_Se+4_}K9(<niftlnJPdm)LWBO{FwS|RX(OZ>^#0h4^hLz^rxpSxz
z6L%j^Q<vgbkg8COpq{#o-s~!-OGv`i+*6d+jfVI}8r259_gXG4r)2l;4<GF8oo9Ov
zVv}_o$X)l@ztZPN3|Dz03OB)Z5a~=BcTKn4(*tWK*5(-H?GA(F$Nl$kbZaSO(AHqK
zNVPZfVG#T?o|mwIKxM!R-bN(Y8ccN7SvMy(fc-Imvywjk?nb`3NA4K$;(?k8cdj=n
z79%b8SmLk)8TS*`m8xTK_<lwP=<g1KU%JiDUi>mV-1zJddbV=KGnkhN{y=w#s1xk*
zN{t|-30nCciH0w~!RZ-{-J2njk`bBr%IWFtYD*%dYk>eFlsqupnR00(yEl+SYpM&5
z9bI^SDxP#tIG%aT#7OI&W0c>YEE&~m8Ilj&z-iA0#^rbsY0Mx}Ty<)ud|}=a==4b>
zQ&R8>4sz~<eMpL9ADuIr*WmVZ&dzKzp&!upt)*lj{~n8rjASE|LO`Eam7Yg|iw{L8
zesM1M-w>jb$LxeM0d1Kjjz~6AakrY(F=vcuo+R@~rl#fXY-cfJV~-q4CWH=~ka`@`
zas@E&x%xq%L=1n})YTPt>h$S??rx*}b97JGuJN1_?4mg#C2xkvzTCUDzUC8MvJa+#
zO~$ygPW&+TMuXLiqO*nqB;w6|?d$nGo0j#FZ*4+64434UAT4e?bLI>QKLFKgxh(Aw
zF_%-{U?SAzN<0`TC&m~QzS2|^NR~7QZEfwlsuy)OR!UTWv>gGJr*3R~3oQl1X!nt_
zaFn9XLJw%9)`!<2dUZiO`FxCZb_cAg*xr79_7vqxJSGfChqFagOzb3y(R<g2pBw8o
z7#XS>)*XF9F=O1;V!XF<7sgA7chKp5mEF_${_?PS8yJLD2IY<YV<Mp~t*x3TrH@tm
z-mN+RhplPSp(`&%jfYqEtT(2M(HA?R{ij&tk4k_VXQBiX?p!szsxM_~gvpH&XNAjX
zIZ@{-BToaxQ5m}hh~0a*N8z7tLR)+h8kFULG_wYsDyDe#+63mJp!z3EEin!*gnk~Q
ze4K{ib^T;>JdupRn8s<WdpBSTdNeHEGgHC#$57I#Acft=5k!H(X{egFJD}txH_$*9
zs=K)w;bz(Rf*noFhFua4s%<xkF(6182UW`n4^n=k9L>wi^9B@;il%Z7Ukd`pZQ}?f
zB`+OCGR><4xM1Xf;{-Js<L!|$=Bs~975GSjr)G?D40U%ehESt}b2H(3kcx|7U}RLl
zWa#|_zlcoiCk;U!fPF6;>eS~bp`HJZn>!ro4u1N_+Q!BbV~QD&hoObm>eDBrBw!-Y
zp(%T{AoxO_3WxuUIR<=o!5d}!jvb=io;o)&8?NE?=LS%81&-JwBO|wQ?Z-Aj_(Kue
z7+zzs?_zqxjdTD6Y|-1NaCzgJui_7qaAUh%6ecwC*ssf<a-_<1b!TrPRagldEPb+#
zZIQC!i_OiI#L77*EH*xN{?MC_4tmXLdXBWSNdIxzPeU#Uf!gc$GdBeV8oZ`{GQ4oX
zeojPC04m(P<xkNCT}PmeJwbd_h%^<lGL`+kq+P36SjkFsnGm(zKaNeet1}allEQGC
zF^0_r<ABd$KMaXb0@AlYhz;kb2Kn;Op-BbctK>1c=P_DxGQWTu3^xFI`&)2rqoLr-
zKw)A&|6FQnsub40qTXjlj%D9s6GHQMp`{q*fiKDf(8QR&CKduI(;H_T0aZ~~KUh6P
zd5t+%j(3{B4B>QO34~+(&!1YT>$eoIc=ks(8O#9c`3g8X4gJOTYKj6d^x5GXq870#
z==@&u7C9VWcNzNk$d`MN3a`K&0oaO**ET|Qm3w)Skz(9E3^rdb?VtL1+MbSo+=}Nj
zG~SozVQo?yI<qRm&q>YKxR|`Jd-gmz=5_isj}<a3g|W=j@xP;@7s8+I3D4R+vp~&1
zPJ33g#l7HrYHB@afp*$S(`)(9mm63{<a#*L8`Mg<-B@k4>Fwm*nldX!1CgSu1GDo|
zlk-N8qK9Q(ew(`R>w@}VV$1IDEzkINOnxvpOP!pjRk)dbqxc2$>rM#>R9<v8H#Um2
z)NVkxk~(W{Zy$oVS@Hh8;UAhAI|v^TX_sLidH{0f7J(|STyU1eMuO}aLl}uVGtq*c
zUshI@>E6&KcXz^nqw<Uq_4+fi7IoCNIkCmA2R4-#(=GAjSV4TC$1m>w+Cs*HVS-E%
z%9SF}7K9+ISaVx~;bbE@kX6bkHhhv&PvjWwMrbmFM5chlsApnw5)`(GT*815ib1GK
zPX7J8D6K{9hW8jt;c0shZxsSoIM$AO`r9YuD}6XSRWTJ~p57h|4AqMHNXk_B37PE1
zSQxfY-#3~;LnE+v?>cnz5gRuq?69M!F!*(+Y9G2Iew;I$ftG8r8&t8mn1m8UbwfA7
zkw<Cy6H_lu5pS>HbskxJgK7El=cqJV;j&mNXOE?qE1=wu&$-mp(DD(HRXst51>Y=6
z6vkALoj;{06uc49E$GPrqE(=>VAh#fsX4J+o+4~HtbJIyxUK^4+uW3PLn1!9OuFY1
zlu|%Qe(q~-!l7CQg$<cbhV|M8y+_z7Oz=T~nk2$dCod@%Vg29PElwr`W(JB(#>?SU
zkqRip<u|fmgxZGq=0l6mzUmIZMg}G(U!Xha*L^T(-H+1X!rb%@Al-BjCMe^)VCZhG
zgOE}@YT+t~hRjc%^h2h$>B8=_B>-nAvJ9FL$uAJD#ek@m0qcaj0UhxxK=6{bc+OC2
zhwY2LfO;$eQxKm*#U6CBOT5)>+?D{5S<5M5bQLSFmlF37*!f@}A&-E?kUkO`8&vwq
z(avmxE9jh3kw3{!hsGFjvPq2SUjTw?sU7>WDM;BDIp7Y8yu61GRp3l;``*1}6zKv1
zSCLx^WrZc!v!$G!UZJQj*75Rg%f7My3cA9W1cm79*L|VQR>V6gbLy_5KNGQ_f$5+(
z{$&-knrMdzt_nRW8|D%bB^s|^MO)Lnj<A~OSn5b{1fq>vWswd5uz{uF0(N^YPMid%
ziM@lRB_1;}B#fitfwP4f5cAeIZ)$+s{OsS@2qt(fh=V|=l2UQ*$dnG@UitV|0<R>u
z6SJEK5b{2+Qiz%)05dX1Wpy<z=GKVa5qWxuS3I~({j$rYMl+CCRx~s)LNfp;J6zH0
z^<dedXp#7^DGY#r0`VUw=d~YIlRr8(B03<hnW*3cpw%Wmuobb4rjXt!U<CkCufTN1
zs_Pz}3|PyupE`eOmOgyj`zX-+@+c;c6;TFo@$fJMf+$szM>XaLBe2aTPfJQrJa>~_
z(d@`SE(t*dQJ`26TXzSG7Anc(XRUsB=o?J+_(k928YzD8rCZKhyg~N#2UxP%usC}}
zitVe<MxgzX-zEXaDWXCi0!@VEZQmcUe(%I+`g4Hq92D$+0<%Y8;un1@<M#vPa@aSq
zID>>4F%d?P6BMBXl+I6NSg`c+gSrKWKv6LAD%3f_x(`l}<=LWpo_pQJ?;B;JT1M=E
z-MMrksOyo<OK@UgKy#plna6x=y^D;ANpD?1kX%aT!!>kyot};k2fpI&;2rRQ0?Bm@
zP!z+C9S_B@`DyefW*#{ZC=`JcYd<)lGYdz<Lwh!S!AYt5@WEI+OZNyIh3hqrq75lP
zzSQ0&5W}QljhpMC_T>@gTect~_z~C=p90J)6;^8&z<mYx?K4i9Lls^lSw1U>Nq-eQ
zEI-r&_`LX^%B%jP1>nF6I;J!IpFTZ@hq^Knr5}FCX024e4DI{Z$H&K6VS>=|gr|iF
z;TS;sYGLsi$`e5fGDf0Ll=#woOB5}3^r}v;mpEYLWGTmQwemw$&(P0LMb(sc0sV~P
z`-liu8hM!e?yfEafmOIy+L(AJVmJ9gh?xy-?d+KG6SdF^NxObgFHOGpx;nIA<^(#Y
zmCxvV4ht&Ne_J-;149jxLq4dV#reSHQB-NU1ZA4uw&EWL9IhRwS0nKG;hb5Z?}0X}
z5)sR4#^|Znf=Tf@NPP{91zfDjm<60jV12h7Us(|%*RyBOnwguYNbNH|(Ro{ewlp*v
zv-<>+M5L<#JA{{lO#PzrM6B36bGVyGt$>uRmNZ$1^oBFJV;l}sf2XGU5dFFetgeli
zg=K%@+U3MAW8dYC0;lne`el53z)ux8m_Rg^j$6f;f?**8Ghi62@F@UC7*?#PMKRwy
zFyM!IU$Kw77hlk9FlSoul=U}Qsr{D_r9=-eW#(O=v}HAJ2$|w}dg(V4MlBmx1QxJj
z)gDaH*RP-AAlNs6=r#^d?F;bR=0)S)JYnRwz;suVCdTFb`8u?b+xEV^=e00TjIt1J
z@?oAyCfrev!T1japKtdmanLAa9@2B}x1JbQwxEA}xs@hv4cdKrY!xI(D6p77ST=FX
z*k46rd@*|DUjokzo`^`c958gzghe(%-_Z9=6@>esO2_*fgpPxspPxe4?T3axQ};nI
z76+IC&APZ>7%wLqM+n-b#y=kERibDGkUMQZJbsApy&Yzap*Z!fds4l8@N(hVitY(f
zta<lNwr*MA^$7-nJOCn?=FxqdvYKhB3~dD1gR57E%7@6*5<p3iMj0cTS6x0MqL6rp
zuz5FcaG#d>8qDy+%7PgbqHpk_l=y}8?dlYE0sQle!@MSnm-^4|*go?KQ`j2*&#z#q
z|M8nKUp4qBC~IFmxM7u5puFGrx3Wns5*+kB?jtnGrXtHHCc45dAEwF<t2gE5{PROH
zs2@Lh=L~<m-8k)g*!zF^!9n#eru<CJW3ZL_pI<{WL&HM@2BwKLiyyN1L4MNz|8M@k
zb`K7MSbwX4L;lY%t@+;N^5jgDU#Ws2jQ<aRGa{Evt;1m5hE<FRBIzTVoB4<_+o~YK
ztZBJzChrCLbx6yMQ66ViZNsYVr`s6-sx=nff*TH?NiSOn>L)D#D(^mR2$L)_;|Azf
zZu{c-BbTjZuHCzhLcd9NihcRU1*)vxd5IGa=`N>U3TSF_uJk`&5C6YK$w!z94GkrP
z99S({09nnk(h&7H!wBH=*94|ol!=iyZ)T22mAyb27@3rmGa~h$*ZL&WcA7Xa1W->1
zf`tYFP9z&)x%-Ii#kzH7Cr=VD!6;F^6}aYF{H>E`U>#06W;9)-&%!rHOm8gcah#+=
zu3fvLEa-#MWWAgxwy<5g4O>x2?i4|8=K+28-;c}62MA7l=Z+s|hqJuzvu7RwGfeP_
zW5EX;z_bT^s)!~Qt{fWX;g56s$-+Eju#`J@XpN1HLDcu5GDD{Ef&C~vaaTg+)ee^{
zRPUNx7`bLc53GVKf^E_Y;6K>8Sk>9NQ?(%z(hEEc;>3ouJk*sh1R@S<))nYsu;A-+
zU~RV69eNVvP{X94o@;Gwt-vQ0hI~cL#&n=+w1^HyhKPU@-M%F#;kCtjP`;vCC(;jG
z1~#E`q6EM2Yba9jJra0DWhD(t{u|F!(LX##?EL(0pV$04+%Zr%q<_~ev?k>SKMbQg
zgg3Q3&_<ROh*aB}E%eX(DmZPj^d>bs-bV}nYbI27iTgdcvE07?)j7~1eh^~d9@C)@
zo`t<C$$c=sy9$c0AIc8O8`5wyd^|;=79`q#6uqUvTButV$zjCrYIvNYO&*_|41@|0
z#^=6JQ8yU3L3jZgM#0pS5BIX>Dw~iKR6khFOOM{Lz>O$05vudB7pmmR6HCJFpcO&Q
z)`uJl<U#cL57_^~6)FXs1hL9Ohfj8`6VV{~-snwWaBofeKnxTsp@4A$Ic^|_W{4<I
zzqBNWqqVTWmFwr3(JW_Sk+yE)!hgK)eW&t~BP<oMX@=-QS|REtuOBg81$|1yBBg%D
z4s{@MRiO0n#0DPQQ>Xff`V~(n>eOEFI4b0_?K&lh_a4BnI(#K!t_8xAbX829Xf%l*
z2v#Rl0aG=@un)9WWExqKq1281TQn!VD0nd})Q7c*wfM5;Ha3Bf&1`<E3Y)=7ZyMUx
zrx#rb!3TRg%4NR$KWXf<sFpk=1%ha%5TC$x5e6a!fF2gLZ0^hftU`L7KOUDaKP3||
zM8<?Zz~u~H1iycTRzcY#cpH~^V;^n3mJ#|Y4oVpwlh`H%M4v{UVg<u>!_NKFb4D8Z
zU3>PtXap4@N?ip|at(z0i$>_jHsj{_qv_8kHZi2&M$cCNdj{v-9Ok(|x>OVueHGW?
z0qI=`r5LbozH|ZuolKMk9$`TN3N2=ev{Skz=XdUWw0vfv13c-;d<?7U<VfHhMsa2Y
z3;d}kb@uHUktpiXs_BgFXbMSAPNw)#*>BXFpZaM-Xg=bmX7F~gYS${ZHz4Ym(mydV
zF-yE*3mO}J1de?mWYwYYX?Pf2)?zw22V#Xr9)7!uV#bfPb37c94NrMN27tmFFNXKO
z;cW9grdgO{AmmhmZ;gOFM?fjy<Xai&djUf^xw##<e3{@<R;Iw-rFQPDL&x+3s<^$O
zTA+0a5+F)+1m2Q|E$)IurJ-=2{gnkHqZ4l4`~-fntIT5$iL8UoylCQKBfJEwl3eDX
zfm>8#;l+daPE79}B*T!o)kot^VWC=K8oT@jR45XngkD7HOJ0)FP6^ak=&UGjG<EBq
z38Ut7a&^5S3ghop2=U=Sf$C%0zPPCeDK8Q5v)}FJmbQ&dKZ-v})EA7bw^s6e(bmR}
zUdF{rS67!B=xHH_Cj_&^xEpitizH7xYNGj~>cziD*Ph7Mmj(-<pY#3uMTc`|&!Qg>
zw5W-iXu1CHy$jyVVZ~za=ok*s9f5lE%RTp@0gW<$0?`)b4H6o<(XFH3aa`cj(p>l#
zT@J@}t=G@tta>wB{jsSbit^RdwjJ5=BjsR(X{Zdy(gOMq-niDYKI~YE;!^)&@Pw2%
z?zfG;M5rXBB48)zD1=R!YG<0MG+kYkqf8teHLB^@=>y@S2u=rsl2Y~e^@XSwcL!HP
z<3^m5z}q>&z8=nkw{eFt_i;lUHuwj=hm+yasmY%}Gs)N=o*;QrrLq)crZ*JjfL8dU
z8#anBNHY94z@|EG5r@SN`Jto>fD9JT?KT!0Fr7hEp(tR5O5lCBjvkhs5zkedJpUqW
z`+gRd&~*=Q2;VB(c;w^8O`Bds$8OdVY};R$IdiFrz#0gN@x?Q4)<)K8(w>Knjd?3#
z2i}H<oH8?uxU&n7pMdGa0u$>hU}9E{`=OPp%!os43Bw(x2CY*QGJ61q)qtEK)4^EQ
zv=*U;Os7NUmL<i$EvqE@_kV>JkHhyF<>4_(w3Ns`Xj+xc%x+grs3KcZ-mqp*XXLWF
z-aHGVW#bq|F;KEp1<vi(>{#KDw9yIQUk?wft2f=~lmPr$?Q&`a&*<WQYirP}xE)Ku
zmJ{3`bweGxIAZ#U+_H#yfu?qSGl&Z#`|*&_iF3G1r_I?DPP$Apu803AgOnF|5B1o?
zq{9I!7zGG<Kmn&n-OzY^@(-lGYfX|n-4{N`oOLV#N$LL>&=8XK8l;6Aqgyr^Z=DtY
zdR^=5vEHXpVFbalQwV|xByT1HoW%?O-s*y^07Z#b$}F%l${U^Nz*O9BGWLet2VZhl
z{6JuHeq!m}#EKEe`ce6j^#T6=pEGl-{wp}Y@(O&-?oH)c)Gz_gFo%aMlY0&zY6Z9@
z5J3fu>5V?%mD0L<hy&%r5nbUAcgFDhDZX9a2(Pq%3lgWk(QZLl>4Q-wDmFzJ_`<K9
zxWN)62&iM}ej@CK+W|g+SXf=;HpQ2up#Z$<WFVEV!|WX7lm6>|6XXlc(BYgBK#>Ha
z0!AkE{->m5T~j^WL^Od1KtQ>rol$_|uYSf<cpTU)6+FX^0ed<mB0puph99oc?*}>m
zZnc|0XGp5K9Lr7rzV3}7$6~3&(Fj-QkOKk(mzk7Z-twq35z}74$9#~6QgBHAcU?Vx
z#5T|QDI0?sY-&+J3I_Ovgs_dw>p_nB*>D8776D5kBW;B|qokySA9WbMmNGhhcqwe&
zqka<U9F61-_|=g`E2v~?iDw7~^l;4b@R{n9Z)Z%9pNWAfw!4I(iv2%!=qPOlGjb6M
zDuIfD;8(!${DB1nuqLNNRfTlc6f3>$6X+j|a#bQagSAt^Mute-R4`rwu+T&rpU!#q
z8?Kz1cMh-+Cs)WR$wa5a$ES-n)A;M)4fev6X9ZRSqeM6mBh{pabVC8!tOVr3#=)^M
zLHf|Y@At22Wg7V41VIn<qTt9Aw^2F@Y5lI|d3d1o_4e|ke(?u31Hba8KoK7yyXo0J
zv3|)7oL>x~fY?^VSR&?VQ>>t`KnSg3XJ1O;A5&U&|A9s(AQ>X|(L{xZdQ}3iiZVWj
zc(AUXMsXZ4^d>7-3O5`Y5-7$rcgf0%6ulG)Ll|eEK0*KH^7+3|IppT^-=F7}&x@Sy
z$lgMV5hNfyjY$21lR;{S55JiB&8B!2VVjYv!o&j+2qP>pqa26Lj4;HfP#%_7?I>N_
z29}0m1xoX_RuDQ`5o~Y}!~xwzDh!-ZDA88|0DYo%<Vf^Go;_~a6hD}O)zc9j&`2~i
z;?!(POic6xTnfcG8(z~jC_2_CGUR*AxM^X(C5n9^MDN1O;b|^EyF2b5SEX2?``qox
z`=7q0K#1*A@{SwI@dO?Rmg*06aUYNd;4%OCD=sOTBmdo>D^FedKkxnA|5v=>e{}fw
z2lVQ`uVVk1W%ut9LKAsJPjyqv^r!z<`^tYC(^RkfA7P;edbDxHmCc>0f<juy^wwZ0
zQsWB$&7HHevt}RCf(okn8>1IaE*S@gzxjR^bD+kpC|)Mg5NhR&ynnP3&Q%#&&t%cI
z1j=>hs64p2<JA*ANtoxWyzQI2mzJJRriY|jJ5})~h4$_kyZANsh8e3k10qudGWD_o
zkP1IdKeXGu{yU0L{elb9*$LZyg7T$Bq9j*>tzIX4RxRa$@wX0qc*56CRD?<2X6Cfu
zMP31Chwa4$ig-UM>i2L58GDJ5F!05jwj?~rh0OnD=%*|>G(7wX$!&L;1E5ob{nuXS
zBEP&aM!`vBkExJbNJ;20o%_~KOm<kYQoQp~j8G@YMtDfH=jyF&OV_BuOQZF+X*m|4
zvX!{rJ7$qOVhokom3@1O{Z<?o&mE1?cSB=e0&Z;57$^zH<QAM(n8a`HR6WJc9<{}o
z+YMvzxbo&-H@jy&H*XeMx<iv^&mN=4XAkD63r5N#_FG4bVkByPQc@Be5bG!P@g-W*
zkMKZZxwEHk@oi`nW&YAqhd_-9EtX$nZOdjRN3S535{e29Q9r2|v7HqSNZN?zwWo-=
z7SG4~E`XWyV>KwAvzxnwCYc+vmPn312fw~9@f_vG%28*t1<!SR&oY1st1k#g^?paM
zB?$m9L=sCq=RdDW+px(J$|onZvgMM70XXRO@1T|Cym|8`3?k#rz0jvL&mCDHme?sO
zKfe&$EVwfP&<`TxazlCMyRiBT)tu9gz3ULB#ac@?yW?Z(B;^)DQ1K%1?Lp5B*rLBS
zyrsyq+SH4<T>+^sD1g|z=u8_u<SI%~Xlr5NTX)Gt_HUgJs`K-t*2y_5=6HDSah}};
znEZBD)-?=okmb_|d#Wua&+^W_;3U>5x&5EU+m6a{=&MIP1C&T?e-OL~Pc4EvyP?01
zxN4GKmJso3GT-A_m(Wg6sv<RD<K1#Vt?-W_Q-6O?)wAvm39H36*nymem)`?@H?U2u
ze);6Na_RTGj1{;sx1;i<y(JCW^(0LNF|vv;83<}eBXO_6?S+U%xNn{X6CVt1!5l<`
zTsYXQprl0m%>B3GTbIvyh`Nb;otNM|VFZI^$raxyUWZ6GaS<l*6GdbtrS1b0mcz`<
zFh>4_b??CS`}?(?=!)a&_!0gInV>s3IJoTl!5h6pLs!KOA6f7If>SB|>s~UMvB)+l
zq(|DK#?k$3fAp{hv6*!($5H?clzen!G+A+IRC;O0%I~u?;wMU2k=~x;jYVKzCMdFX
z6)M>ath_Afe=t$9bkotqipW+?3=?eGV>BIXaQ6@h$LEk<ffQNg7ZC92^o3oxd9&3X
z%f5dMMgM0$@4FNZQE}N%!8&%aukw8WkHR)R(a9JwlMv$W+3@-<`jMHwAFi1wk9To$
z2E5SAAq(wD&05RC<1{vV@#*EMRY%U)LqV$AUFNzBEzDBRHCAH)r!#a4g~?C`5)!6W
zO?A&B$`skX=fq2ds}4n4vU%5%1Is6@8bsgMRd_ykWDe<C{6w*>528W;tUq%>@@VPQ
zTm(A|@p{e~A7Hx7I36LOMC2Cgw`4i6m_omAFz{-<8kwMwmF4!rf}&P|MO=*I5?VuG
zL<)#6+h)DrK2>3EO4_FCz{&MFJp9xtVh8tU<ij=4>eg>%zBI5CCw$idHRACI$r?{Z
z?8eh?-%cO(9q+t<+^Rv2g84ud$jwL?MtpeU*3bwA3G=>v&(`%E_q=v-dNpPCF)u~Y
z7!@%gQ<_ZkC?H!RL#>;62x{=wb<phn(~C)eh=(d<__t9sxojH*w~@DL0~sZWIy_HP
z!6hlVHz;UJD=uMgXpj%EeKbRvnl{h_kW=s{aW|Tn<36=2@$R@wsH6HuN5x}jX$fzy
zovF=8@CPUh3rB2FSyJB4{dF(q#;`YWL+#NxZ-hVl_r*talM;GXE`5Nsm;ouU>ZVOp
zMd4t}=V3qAzm_k#i}DMVthQ}~OKqxkC^y%Bd#<F~=k*WgQ2x!IC^zZoDAC^1e&3i!
zC}3qB%3rI|t?y}x^9=wgdGIQl@1D~|zijI5zRRp}Yj6C~^BSZ?irR)woAl$P(?+CR
z{+{^787JOi-c_)W1Rlp6Xv}6^v6=AWzj?OHS2W*(8bmka#@NMJ<JX57e6FpLxDEby
zt><m~kEu(k-bG{!&?6jF8diBLGLSbA-N4)01XGA&C<+kO+K!#jF`$E0!c$<95@`(L
z@bsK2IOI58bPCfw-rpU|8|DmUUH(e>u?`LN@{5uiiXgB#c2+_S5=*#zRzc^xHT(S7
zI}Y@6S2Wf)cck)JVs8khWGzEtV-=0j=|i|g<KZ!9|0w0gmkdPw98DW@a~i6|sn0iC
zY-c$i(}|RzAvDXBH$-X;(MP%D)}vg#0;G*Rk($FX6|Wi?fm+J%ad+R~`LO*0lJeir
zv8J~l+_@5jy&|tZ)sUVELGRteGYO;CDbSi#BEOOz1>SUAdrvYr^pw&<h~r1MY;bBH
zJ9+azM=1+dVsens;vEl8n8VnN(18WjseHYUES6G4*=<nl4(=TKMF_}lnm;Qtx+GPd
z;73Q1RTolb6;zarw8Fzo9j#*ZwKx}z)<vt+5J}1KNZH}o)T7@&B6~Rb)XDp+LiPX&
ziXT$>9kACf=iu!@QQBK~)U=y=`Jbu@Ud;6gafJHTii}*kKDV<gA6)dV>0e`^Kl*F?
z4y>>qJ=S*vrwWZwCGsk0!}VBF-|&a~>#A-H6pFSVw4vKSy(w9zi=#_`f9*kxFkI`>
zAA!aL5vUb&NoY3ehvht`{Qz>`vC7gVQ>KTaHVb66wYOGaIUZ@cetqxZtBCcG2FP~>
zzb&=5LwzCz9kvlF286n1ssD|=H;=}$f7^I38l*JOb7@YaG^vm@X+n}T4~izugG{BO
zG)J>U2}x2(6OuG&5QU^@QZy-<BJ9uUdEWj0-nG|W>s@Q_KlUH{S?hV$qi*+o-PiT~
zp5OC4&f_@FL!5eZ7a1rwZd?<x`laX|`Cn7(?rt$)D`c-$-X}Z@o2to)@7H%0M2@68
z2jV>XVymkdFp{9nsL?`m$WFZ6Wzng>`0-z=7F|}@5^lk>T%Uw~M<J-GDqxY;mc54^
z8Lm6{J(-IVwbUpSu6>T49=U~q(S|{AWbiJqkFm>&U)pf*M!slWw>oX_ZT<Bj44{>b
zqo&NsywDmeLRX6XYEFIByX{TL&k=pz)*THSf1%%}C^+G71|=vTxVrP2wQE*DBc&PH
zgtp?K7ER9f+1OP-?d3}urExG|d<^)Q;xO$-#?=WeuPjS%5?PwxZOx-$V^4055|Ca+
zSXVXRmXB6jI4%9<90FdAmtj><fb0*h#s4Utx;J$P)*PxrBRSZF+EK8jX8jgD*|Iaf
z!tD(0fhW_iK8R{+C_XRs3{Ms6J9iRX{QRsb6Kdn^b{CGJ30j$H8$Ay7OoT_$<3r7o
zMcr?do*KStY!o9q+bu|FBh6mujH`bA6VtCi>zRl9D)c06!IVEIoD|y|$z^DIx=uJg
z@H9hR;_DpLZogHCJYg$hRB0E?eUuV1ZMi{3_~XtU#w4uuOgY?hY8`>k(c`@Se2T){
zZ}s6`;sero)27oP6_j$NF+<|76T)Mh@7Dn=Z6RiTSAr(fMcGLodp@o172i{8GY)C=
zp4oLv*6&0oMV8a|!JF2!gliVGeV(K!q;<H}om#qT($uC+U!0p#i@x=}!iT?F{TCNN
zE|Cdms=IHpZpBBDvdHbv?@v;uLIb!@b}x0`U%OA^cQxzlXE5LIL`%dk6o~c9ezu1Q
zyJs}r8ppIt8mrA`3kTVv<c~}R;o#qjwZ_iiK!HFo6xmMem%^-8b}y&#n|rIY`S@ec
z*o=%zp)fPVmADvZ5OC~(KHf#P<29Vx^q0(gGwWd2IIk^a{oBhgkIha!zI5U-9d$b9
zHfes<Zd|`1#$4{Z+#e){Jxa2@HzhjlNK=K`H;mbd``A_<SmT<M(ylo3b+-s~!5!D#
zshAYz33Z@q^-UYZ@jAZFer>Ux&Ht8IrB>4(w>e?MI<!+rW7B4JdE@;3>-W7;-;2Jk
z153|-|Lw%Pjy3S2qxT}bb$9bSX2x$H#dIQ%ucDC9?GP^y8X7~Ys4lo19GGntxfrMN
zi*BR07rd%?yyp9-PiHUgwaRoVI|Xa;g<}cQ=bhVkx9}CDaeal})n%D=fX2oj9on30
zH}>$jG+@)yR)3?GHWzAU`JHND_P(Qwez^bnv~-6Zt8#AYk2ir+ZVsRumI$1y6CM_W
z2T9o4^Z^d2r_vp(*78tIu-SO+rq+qust$j&4DQ!yEyHGKT-zXQJ#zBV0E-ki9-Bd3
zz=W`GG9rUmnRi#6Hr|K}d&|p8Pg4q6mfmm68o#)7_73uuc{sBpkez*Y<=B9wc@LIL
zR4R!`B-cwIdGGSi$N*e`%q`Cu#TOtZU9)c09fUU*7Z)rRUDo~77A;$j0HA_)*F0t1
zqkg)+Ivqw;w9?g_X?1>SlB=22f39kE<`n;Y7hh*b-dLD^>vA({!_a>bGPQT?fF##B
z+Q)LO04zC(tT(M%GQ1lR9VPHqrg_#Tc*gFJbxVZPVeK|DO~_Ey>$KR)s($w@#&tzP
z$L2k!(DXvCSe?V-HWiV{+40R^oS)VZW)j6_Ta(Mm;nNlFhO29g!Q9~4*=5r)HrD`g
zs{Ye@_AYVB3h;5-0%Mlu<bun;j1PA^I9Hzp(lXq>|BpA(dr6XI<TSuT9D28^xf$L$
z#Yts4+NHrQWKr(d{!>!ohndi&*x%Y@V5;ptJc%E>5wUl*(Q0Z#9ZMR;v~PdixO^<r
z+Zs{i#3GJu`~|C+n##px8Le&~j;!E;TL+(rTKn54Z_BI8-_o83WWV&Sg1LjYZA*X>
z@vJY<FO4=$T!YW8teOewSC3w`Wn!2jo|iS(l5fYs6r%5u=HT&bIv@y`d2_fOhiqHU
zFaOqa4Lt;nmP?i}B4~kS@0s0>oj7pdbU8uUlz=J1Z;H68%CtQ}EE@ebD^$Abs2eE7
ze-pOvN?IC~@4Q_5RqV%)I~q<a?6iN3Nx<@%yV~@tjtI4?Qnh+D>sPaGY_>5s_cWZz
z&XW-ut15XKo-zP(1J)IS@xOlGxqtu0!!ZqJnr)xY8lHyM58lRz1@2WO$!zk=ONB)o
z_+ZSok8fNyS5Z+&|AdTaoi%Cxyr1`v^(S98q&`zo`c67IhqSAi{_<AB`ASWArC)xV
zdukdzyb7P#sqT+no2=sx#ZfVYE&FPLD!YH|0D6)<zbA=j|C(>0DO`u#VFRY$+Y9&z
zHRVOZg2J|n&gJlt%yksX5U}Ti<R{06G*zA<DHytQ&?cG!aRp0KFWS5_Oj8+>d7!ge
z%W0O4EN6IzzY1Uc4y@R+=(A5v=RrI6?!8~=oN+mH(W6YXvV4vJR$e6SHM6)V%HbZV
zqp@*yVsyMTcNuK-XG52b0ne5$ZVHQt+J>3#;|s@yebc}Ew~f=<XKD)C?ROCR6qgsy
zg#)kC!eGLnt4dkv{P$<ZuKMm@vFXf#xP_NyViAMBzc|Elv-F7uoOC23V-A)I9qWlB
zE74m2oW#>=g<H3_>)NLtM&E|_CwkXw-Fi0+THzI8D$a3Q|K9Gldsk&W(i-wwvuC?@
z2Fu^yW~WF<<ux-ckU3YSIzUh<*dp{WO+IWZxr+76&*EMTPR{HBng0#_z`cC&)c^!#
zq(JhKd)p2(Br(0UskDjkVo{o0wEev>&Gz<0?;)qD)5K!F#q?=0z;}zke7Uyjk2c*f
zc}FjD?ceVg7-;UlZI@+y?Z&lJjNSna5H=FzR3D*kyX|{BM+_p~Z&jCiF_&!te}BKD
zvq&3mQ{(p9RQGFgup8)sigIU9n~{t^NW4C-DLvz{zR|2@MIqVOZhe{cX?cBd;}fk~
zzsr7GIZMJ}Tg^eMKWR)nIY@;L>#p9nKi@?-t=Dg%EW?X7FGRpBcMfJ`pY=axw5nAr
z7SqZ((BYakJ*-ODx47W2ar|+;)?iE{S5Kb~gRVIz*xg{n64(zQ6{KDn@gx72lVj<7
z8nS~n{W$S$P8uC4A&S1dV%v~;MLYTNJ&A&}3nA-?i_-OwkE?!{`koqVG;!jC;ySre
zM|-B;ds>H1wjsHd^`?o+>_TsMr=JQW-4SSM&Q9{F6B;&OML}q6aPZ!4@!k5>u=6Qu
zeffTdTU^7O%Ip!W`7(a;$d3=M#1FsY!CDwR?CSOODOmO9?KgDW=x^MDr&c>GMB1Yo
z#g_t*&Y)m%GGEvuvb6BwyqwHv{dKvos<BQ6t^FagiKXc`8cXE+)N?wFxP5JV3um#u
z=^h89yBG$=@uh$Ld{@bTQxHM8mOxU)2)WmV3&wjcWw+F^si#^o&D!%QiS?Z~u_AH*
zS-x<K;V8cgGe?jSAUSc1$FWo0;O!%|hwRk3zt2u^-?L}Cm`nBS+1Ke#RY+@;ov<DT
zEPpAZ7NoeRecN^NS&dg$mU#$7c&ovIyO5Ue<XJxwdb^wvNUBCdx-_97a^-TweUNbz
zP-)e@9@-H<67(Sv9O;pM?b?lb98ro3RlN04G>6}Ik(IqK-qDa|aKfcbmiatr5`)&}
zb%Y<Q*05p2HD6w+AG{Sg*T$x}Y+!*mMz&%0$=#)JU_h(5*C3@f3AgIcLZ7oWVA1B2
z=OzqZEs3{fnD|YJ#u(gCQe82TFDofJJjCYSjj;{4i6p`6%dP9HCBGjEEd<W0&8zdx
zZ*`eq=zU{t*1bpIStp()Cq9gOvoPd#Lzm3%rhZk)Nvj3l+ue7;&pA4FiG~--i^Jx#
zm8;Xe5R$FVvJ?$)kwB~P?q|oJtaO3lsCV1jYV-d6^*Ol0)+e-q$uAsZItD`TWYKmS
z;?VJO)we{Bi>tDX<?)k!x{)(V7h1hxm-xc=xMxAz?O|W^W!t5!c+mw39z7L!dM*oz
zT*7$lr+Q^IXqYU0eNp9e`Lv|*p0mrn{sr9Qr<0VOV&uf8O`ELU3_Z03DHgv)IziX7
zRvT4R7qc{qpz3v-`+YMBbf2&K4|xzIKd{E5M^)$e>Q+1QU$DM=UXz9{|8`%Dj7B#5
z&mWHLyi}#~TKyh-y-1$2u7=Tu(jPM~guc5xZd**z(3=W0z*dzsz#&z)R=%Or2_sEY
zUYA+W-m{V`cs%dz?{V8y{^MOeJG}mv%vgCn|KsBRAN-I#JD!oM@G#BSH8Z--WW73Y
z=B4Hx(#UVdZ%hIp19`QsZhE0nNmUX0z&I0&1+7{*Bv9nj#5s5C_NBhq<U(tb0fuS5
z!g@2U-=I}14UDAaz@mL=iK<y>QrJo&h%%)E(1osO9nQ`zybH3ke#?q*EvIV)Bjgmh
z$@Ml}^z_=$E*lBOiMCI5k{UYgZ{eyoznX~?3%}#W1ZG1BT#S-;p^M09t>a(Aej9R?
zNGfEpi<(}U!}wI1b#syEaIQ`<p5fUKg}n<>T)qp(z~K*Hk(Kkc4}4E$MfUstf%clL
zOD>ESMBdxwt9oPGCN6%o51DCLwX$G`G!_!lv%I|2&i}4an#azahOp~QXwE{V5|<3%
z+8P)HtZd8ZjIA6v3@OzBQ}ygItpV?UpM3~A(mQn>daG%ApLz90M~}g0%}lB=)9nMQ
z%~jTvZc>&LDP+b7mU;D%>8@HP4FG<A5jtCc(ueC<ijf#Igs97B+vv*m_{W+$R0iS#
zl5XP*XhJ)ob_twzxwF6|9K7#<X6TKXtqZJB7xfLV#pyz|b!+b)>FE4Kxl{fl47q!4
zsDn5yv<2j)iJ(Sxe-X}<C<r+%7~vOkz6|n=V7s0)bFpaZ3epENIgBAzAbqvFOgw>H
zFl+t&gVSY<Dt<!snl@F#*-vO(!hJ${f6=+ZX~U*XAK)MGUZXe184do;O8A@qe9}wk
zIad=lj-pN4$2#6~{MQPmOiC^b!~W=1nXydc_$ZG&p0YR5cZ8bUTyd}=S&|co(&YmS
z)R1#ThMhir&})H63@LxW#&09wfJ=9WFDrZkVR3OqZUF&{4mnagsZ+wn(%cltoSSHf
zEf2cYJ6MWb<Qx@IA2qh(Lgu0<P!g*1Xt$Nq&OBN)@YrNO7aiLqBc&b{PZJp~DF4#n
z^@ED$2dhKXh38t2{4%XDoR3y9;LoydemFW{*)cH6k{@3$f5^${rns{jKc_d7hnk=H
z>s_pxFtyH^2WbMDLEqvG9?8_{n(enT9Vt+0$1v>B$9*a$1IljuSMqUW$odOPJDJV7
zZJkKtl)DtSaiEG-73fbsGj{h!h@)b#t3`j~3x8>x@-0V#;*JH@#L=Jnt&w5`o#rkF
z(#Kc5Eic0FV%I*+dQ*M`j6+=!%>m6~89S&*zjg<>-VQJ%2wK1j)mk*Q78$Aec}r6g
z(jD|1_zQO2#-#NI1};iFr_%2cOVi;-!tElr+Niy5IUtaQLZ+G7gf=V`O;ebuSO{^-
z$;4J_y@yT%4(8F?R_9cdZm@jz)b&tVOhJEqsmSykOFXUhA+<nKTer?45QS0$<t{{l
z1=b{8sxG*S*k-&Q)292+PPb{QiRpt_LIHC~BkmscyKqFsNt4JI%lZ3z*dDt@9V#hw
zjTbU_cg)0zqgWgiOui<WhyjEe&x5uwTvIWDZ0kP%UiKdD=j*$f1a&UaOP=Y4FezUN
zt(cv}dkHd(|56MO*m~S6u^g?BWgSO;&CBN?$kbNgiNvgr`xVIq8C3;|s!5<0<t^fW
z7b$PSvgO`}0$1{kJkt+p^@fcxms**G>*1PDuc_-qVs*tPQI#qu;rXxC9|#aMxrc)p
zHU5Hw@I{9|tK`B#QjFdAXy-8$F#lUxLr!f{ki)@ZL_wcM63w(-k&nkXTP)1~8^PNt
zWdA+27$Br544JJMQ~bvd9N+TRhLsyOY><&VU=nzvTw7Y<wyRKiJcc*{l~kxk+itLp
z;=i0Z{-h0+0ATs0wJ&E6`M=}$HY(A>q()pjsYS%Giuq}*Sc~j~a$aVSCU2MKC@jzz
zine}BKfswHPj9?pLB&i&N)P+A|Jjo<rfa^v{N;8JC3|yF*Msl0AHv<JQk<7&R8j_t
zA?mzsSCb;sS2b8a>ztW?O@(iuA8St*`WK_xKzh}~w~rjmNWL3A4u5tEB^+pSC_!5q
z6*jMVKxeE<q!$Bt)vci4I)VwlB1dw{1sat(9v3bwR~bF(XkTq@K?;zRoH02J7m#&~
zcxb`3@kL7*1A~>#jkV)^e-NuyEpzFuZh$><TH7zUDUhNTy(!v%xaLSzlXFmDrTNEt
z*uq--^r#t(NG1r>$j{Fg@+j>~xU~<yfM(R-&>f2AHX|cLAT=3yAf=Zm+Q#)6g;yD*
z6oL7+*KNB8ACK0s_VBG^s;zm{-ANK_DkjhkBw`K68$u#42($l>9V4raBZq?m01xM~
zp?ALPr_{?PQ3do2v2XQ5NP@vkEQi|fo+-NogeC93$0?S`Er#9_ZE!!}4yUIU!+l!I
zq-73d>z4~D&Y}I@huXD3eS!6_!opENPx327q@6xR75<=%;)}+?-wp#%aGIxuE!5Q1
z1S*y(@UkK3Hhp*(>-O^2#(uJoNlxY>O1)G_>rG&3jyh}gQH&w?>^-{;nvaB^i#81J
zc5#Hks&kx#w6o*bk<Yl=p}DgmUvdtMrw&}g*XliZ3*oVG03#a{OTG9nIKrx6u|mV#
z>QL1a#17fKoaghgw!3syh+b$we-B+(35eMp*RgE?gH1|2=@s5Se6U;Fp&Jy)3rJ26
z_cPhqHD`ER&H?0EUcu><X$nIl#=g73(|)LTT-wV3%}0iA)FyTEsTl4&5;8Qq;GTeN
z$cSYIlp<Uw;qG(pZGpzK`JPeyhwtCX!fU4`gxNVp=125tbzbsg*3G<cFT;=m4tD9S
z)qL5~r5pxo()ncNnl0bw*$JOo3|p!wVl_xGYSsD(*iEILrmcP(nau#($9gUBgDkA9
zI`^?9xXtYk!|*`n=g_pGg1X1X>VsJ|%eoeED513F<IRTstJG7vn+Gb2@t-9NG2w%5
zg*W{!+$flU6(5C;6;;_D9?zqBLr@(86DI~w?S0*e_Q0Wc8BJta8tE^FuO+&2aa_Vb
z5<6c`l^EEku-szUo1U#Tvs^I}dvJ9ktDrHvb{ja{DGuFB){(O(&DlppYbBIy>8h}4
z<5-V<nR&x*2|Q(|!j8HJ3@=zH>LjIj)qy<nOVJ3+fa5|H@0}NrXo|X2tN|&3yymw2
zXHKP-V@__W^hAFCTp<EgLi=XwNb=EI2sO<^XSD|82WGzX^HJ~UIfQn&q9bf<x(X;m
zV`9;!H?*3Ci!9_x@J4amy@-YvnVdSOjJ$gM_+97AWcC}KBksi<!kNfH&yGL31zs!$
zbJc}{j)nLUpNL~YhKH5$W5=qvXbc?i)Go>P7Ga&Tb=RF#*15tUK0PA{$Rb(iSs5Z3
z@C_xKZff&QZgYO7oTT8}u_Upa9`|{OVJqGlVkLDN0sf_mfU|z5(Hjb1pD!t4gu5LB
zd_yXE$$Dw?@7C4}7gk4tCUaIu3gSSzu(jNdl(hS@ZdO5N{?9JV9HQJ7&L*RBEG;eP
z;5T$BZHoCCY~nPOKAgQ8U#qC9)_jtbxl8UIz$)G>v21^l0E=NS>-gHLteux)Pmh;V
zOkx2w@yfE2wIX7p(O0Q>yCIq~kc{HYigGs|xJ<OBnHfX(7K66EkUdEYMx{g{%ac^0
z*mW0o``f+?XfAF=$F%CYM%^GsT`TFROSQ4O)=sdY&=EMFoz0Ox0!KEN#|^z`5>#t4
zP>_kT2#!)=Ov#`b0bf`P)^0|BL`Enc<Sg2iOxhI_e#)Usjg4p98Tm|UO9B;3!T>?o
z`){4WqeVzZ)*y%{!wo_&#risU((NIx__5%ar@7hCk+96cBO?NZae;LIQYiTngM5Y#
z>fCwMx)p)ijp<nNEBD}CU?5Oh$K(Ctbj7K6k15B@P46nEjZDJUhiP^%Tz>xL^r%~f
z4aU8Axw~UwgX!O@HU4IK`lZ>o$f-*M>-QM5F-m_;mkT%Bd33Fwo4IylWN%K}!TnY*
zd@%F`vYU^MnyULIc+i;?PdyGbZ1@Hz-#WJbz`|+;5jb1jNlI$4fB$}4-3tr7e)IQL
zmYnFTcXP0UAE9FK$HQb4E#WE0@@zki<5Yo^YjSR5|C!Hc-`e22{^qAFS&BxU^+j#X
zt?PhAQ=FIC>W>dpDs!WN^DP`T$sxoX4<pm?Bv?STl=y*<DO02z!4z^<?yWDHs)}*I
z_PyUpNf~<C2~=55v&fAiDZ=;Z>hU<4>_l=Y>cWig_yEPx5E4ghG$mi3oK9O$4#))B
z^3lP;_t=!umlkFQURa68-EuP$63~b|V^N5jn>DGM{=B5ki@Oe--fOQyWNxpgH!Ej?
zBg^tUHk@Re5aIVsbb1o_{PD5gd6&Pa48n{~x)7wRqK+P3dhEy%<(czjoyeM*qc4{{
zBW228?!0N1A%~u{oJe2n_dbliA=#+us8#FM-6LxCA%<neW-X({5~_`8@o~IfXjr>j
z2Z|-Ja)gfWX(234)UKP!ASY-ILj+6c9!<2_*JV;0#RVCYOiqPDmgDfcjqaKnl`{-?
z*M{t;_v7KM{aRCdNksQ(($nnh6#LY3Q-bjbeWWuXev(h9bNL@NwplE?to-Gv4ogZ)
zOTWJP@y`y6te|tHDEfbYR^d$4OY`fC9!v_WpNtfLkLR2HZ`r}hODFvffOKM&D#tiS
z!IEV-ikW8kL(83#1Z8S3V1R8wubu>P7w6nx=_t2d0Sc{sJ5dS<tObYV#nDBKpg1jK
z&XAJlS69>Dmf)v0hi0F|eJP<iq#)({##n!h_o}5xa~i!(*Bk=yed?k$dBMr}M=n{A
zuUp58(nB~sm_cynrhGAU{*z~2Tv|#M+rskl!rJED8%v7JJ{K*>#^x%a9i9+i`WLsV
zgmlc3&P~yhxsbH@?vP{1ble$H4$X<aGQc5>DevO`3Z3GiXU>`^9S~NNPZ?EasinB2
zstkVO*9zm<X4(sCIMLp<evrz3R&M?1Biuvl3Thl-GZ5rP?bjR$KI%fxs(bqMD>j>v
z<<p?9BLoDJdhtUL_0@xeUG}ED-YC(Ar^e#hOwJ|l4;LdM?!y(o2#Sl3j`XbTKWkSs
z<GV?2#bG096Tjw)%1dbnca2$N29}o1R>4nyf9|5^;D2qNyXi04zIzg~?(1&6okf#-
z!ViW>{Y(u1r(GTuaEoF^#bD>^(!sjAAQG0j1D-??o*i=CBX@no#1}S_SU*F%5l4fS
zga%1s8T51;fuoGJ5yR;vDc=qzdzc*_vVJ*{yE=PyG~Zu7gATkJy}J=y8e+7qsC;az
z9cbBq@m8vX5Bz0(+cAuvmWL&o3LPBL(Gdf7Mq`k`ViTW=&J&sF_TXs0E@F``By_m&
zHZ;sAsKpdq+(e-o<5hApX@35}gOkqCI#M;D?C#AxfPIit#=v><r<V@(yN5zWHd15;
z;j`DQxuWdj@;fb~`cgGO;pTUytRQvvyafw7LVUgi+&Or|&r@P0g3<pOu3gmti<DCz
zLy^)e$a3m6XHGnc#dIShtQk52JiLd~&Ar*jy^e>Bb~iE66x7<9&o^$)#wBfT*muAH
z1VJ|*z(&qWw~O|oARo74#YJTwlu+k0T9wT4*~f~Dy;{R8Q`#wGjfR_VV4dzEnHWW<
z#K_-lhp~SHY>t931pr9ayc}E{hjr+AWlVvOhlhu|JFJ&m19J(_BaJ@7zEL$GuCCd$
zk%ZkAAOxr8oumXr(s2MVS-e1ZBiCh3ylzmhUOT?IpG9!o!-r$>OI;gvUC)b0MPcdw
z@yYS2Y~HFa6jSa20Rferj@Q=@b}R7n#EI>9<gy6B2V<uv5%z4w{6kHV<|`jbiFK~^
zB1vtekhL9J<CEa_^2XX=aT&mnZDfc=EhUW6?%0<!hX@W|OMCK{kwJt&AM9pKxbHM*
z%rSyE19L_j3lI1*N%3F5e$`SwvV5#^IA@ax^xaHOxu7y3GVO3ENS>NR27Mi$nx}n{
z$ekfD9O$=bA}_5E&S}eOjdINK#cId4UhXG<RcSk%uPz=?Av#{h)p!NyjUNG59AKZu
zPa(P524MpEf;fmNp*N=uuuZBebSom7*Dc!gcW`v}%2*Gd><+P9pg6ojHAL}iuO=Z|
zAS>#;eZ*=G<L0cyYKq`mlDZQV$ZTuzslYMy**U341J{ssP^=FU5EHkK)0mmaxV?2h
zUhm-LQo_VV&k;1YzN0(V6zHCIYXu}ZKNZdBnKAW;ZulwViI2`~t?3H4a~=t$?uddN
zdW(E|DF3L_n=hf&0wysWC9L+%?nc$GGpSj^JJoJ%Qj+%N;1LV<2_OmbGK%fnkJs4V
zQ`*T#%0H$}3a5prj<3uG<Uh?lRQ_8y-DB5({xIG<t?@sYyJwL3xyKh&>nHVFJ2btQ
zW1Xa%-&E3jJJ!)&l~$)v5s#b7f8y`xS=GHA5Ej-p*yf)<DcM@=IJ#oW(wj!1K^;4k
zHjLU<S6CR8|CpA%w4A=ka$2kZz8cSMegFNMlBxXvKm3319t0T~Y2R${f3BYVl^50i
zV`=@5b@spf;ZRl;7ul!A{)X(rq?v+O_?iWQY9C(NxKu${yw%wM4ks3?J%3L^NQlG3
zq>>X0?herm-hnEeU*i{8<@rY+`snMMcb~Fws%vGoppl<g#RWd#7G6S>nEd!TqgMmX
zyE|YOZy(X6QzwcB6#_p6_JHhSk{FQl>c)q9*gkV=`<~>7efzk}kT&)GXD__f!CjBe
z?|`{bquQF^L-yGREKo!=T9I&A`Nex`UK1+siciMqx(c+p`q1o($;T*7t9XB>$@%i?
zvX;smpbYy*;sPW`>b$s}G0Dlvr#_J5w3Zp2SjT9DI_R>scd8ipRk%OiFuI=Ocsxvt
z-zsm(rGCcN)z7Y~5Mw-`**Te{ReofCj$^)Z$|__ONrf2KQ}SgmO|4o~vq=wk61;c7
z!@3)FN|QsT7tNnKa6~J@)8*iwX}DuH1~kyV*`i`2xnRP#?#~u|jDFSTz`=u}oPB&@
z)S-PqacI`Qx;}cB+gEn9-}$VAk5vNIUz4tnU39#2$n0pRdi5iR8%{{L9Z|(alv^NZ
zt3E%iO^+TS`e{HZ@%l=M>h>geZE4{$e12*&y8`^xAZ%Woz`~-`>h4t&MCkdZUYtH#
z0s^|AN0PxMBvUdZKyl$)4EHyiix65Q7Utc(Qf7?R+(Iz+rCz&u_3C!~7r112mYX_I
zuzj{fVk@3paMr59EF-C?hRCS;ljx3`NAw(Z@dBs~yY|SjW13tsXGLqetU(`J()}Py
zV?20h#0k=ALw~b(_uD#~Cned}#UzSFDe_(peKjc0Zb&u&$`_lo&W!3svwIsYSI)Y6
zjT&v`Y!GKllOS{5?>#~a#y?^4O6v^ed<-+GIHhGO<+<OF{8KD+GzJ$@wU6N2k$fLv
zL=|bE3c~-8OG&Sy#B2dBUkuVo`bB)vQJYn(QJb97c`|ztG*?zL9u*T8rPIP&d$swh
zTa(I8WN#WR+)!>QhyQ)<2!n+bZo&fA7!J^P@zhE(FT}&CHkg?tgyi_bDuj0-lON=i
zYTthNA|LoUqSpb)G=DWh@<(*JAr2UYZDu<{$B~8^_0wsR(4dqq=r_TLO<x_CC-L#2
zD~*W{NC9;JBrood*4Le%vD2-_{uvRorCMLmJMUYp`jNmCIME3HM_nLID5)eS&f*4;
z4OgKFF`VNLj>~2e+pg*T-XEO+hnsYYY1nmz*}&*n`aYc3i4Qk%f#T?czxMco4P*>F
z0GVS#h3$&PVb(BnI*BKSZEC_|A5B`v{PgO)MWY^;^?;vKx(Bez_?1E6=Tab$Gc`r6
zzhc!YL)Kk*dA)TctlV&B!sWau1>CVp2OsgZ2!{OLPxvbVGXo0RF!N;Yvs1$rmj<1T
z4gAfzrB29fg;aNk467tgu^_#^^h*bb<DE|EIkph@FK&o@LyA^EN)1t^T9m^C<JWyj
z;dMV(LdZtd_f2|B#{p`aCZXlPzI~^{Lri=P-!rm+FVqwopQsg63K_&|8f5<d$BzqV
zycv0TX;DHiG79ZA*(TF&Z3vTTLwTPQdMh1ubj+7HE2NF5&JbhA)84J>B6*wmP|>??
zL`&b6EK1o&mZ|l}qp?=!?q0?eo@b19G|*o6&GM*@?SV$UW>_j?2uEbAAZh!7f5Mp6
zhw%mq96J3q{7<m<Ybrrgf)5GxQEZQLp82{4CNKV#mv;bgz}EXT;wRElX=_+>CnO`}
zJr0uGtF!~_$c-%I@){{-us<L|c){s0##sWcgArk-XT{&&fKH--BPulNW@cnJn8IJm
z0C4+0#chT<6q_hxq~V4Mm@dOK00$3K*|jhqU4K6hd<S!kxtc#eZssnKPp6S7pAINc
zOG#se-+0A(i;W#g<{l9dA)5^wi{3Oi577m+0%RpI?rEHH<!a**pkEIFSF5^+!3j1~
z{=~*S(8;dp&tF9mRQ<^X{mg>S@TuBBCP%?xMTE+A)6g%BOkDsztZHy?PwU{z%eWM)
zbd6yd0ZtS1BRg;b)m%`gP30lURjxvpJ?)Ckegdoqj%C!m1(B%_F}tM;b%w@gUUWs-
zq{44$9Y}BwP;jcW!({eZd^DV6IwuXl5#`~55m-{d^?*e_536lLrg`4tR}(20FyDGu
zSO0H2M1AAJDU9>$D`x1y{-9x4Kt`GbtQr;`-kA8o)7azg4&Yo<={D`s8TOJ1cpJdz
z`292xJeW_R?Bfn~%bJD!4&r=A_|{2u*^g53p?;qRAH1Wo0OZxG0Lo%RdgHdQuZ6M2
z9+HiJ<Q6yBziZ>dkw(V`+VxV-!06S+GNd`;CSvIhRtJ&J;$c&CyWpAgmMqaCwck&q
zc)*F=k>H_faCV~ixsQB<AY~sr5^nC^$KFGx5{<ebBWb9-|NOZN9i#mKsiIxA<^*Nu
z^kSt~vJg)2XSI|dYDS2xlQB5&fs*1-ZdB<+McxYh*2Db)ZBDPE&o?;?&V_#m|JEk{
z`0pXbUB&o7*W5&qGBG+dwnwj?7scH^w7%lGFU7E`fpnfpCJqs@x~5iO8F&cLL~A4P
zdQasHWsDi3aP?ZX4$^0cr%+)Pj>T-R)8N6=9!4|$sgr(XBJH2qMP1qH&M7xic|gGT
zx<W3YuH<Ay-gI{d?48+VOrUyasrklgVfII$eF$iMWseA~f8iWU@FhZlx3(?9QLLeo
zL<cscY;Wr!ofn5&DYQ#Qc;=I;bW_>Ert36}0~UjbxTjAi2Dd`k`S4KB;});Dqt#qQ
zL;v}eY;@&;i%J#GE96Wym1!5}Y_`atJn}I1H=aBPAJgGtM#vH91(a>O3^3eB*~ekt
z&GuN`C_$;R*zxC;eE@G-w%cD<c{2Jk*Eg}<?Pt1+w>a1Z>~I$nq>??O>4+~!8$5{w
z;5uPp51bFMG7EI`)S2c<78=9WPcVt60M$CuZk}bS?&?pwlzlMVzJBz^iH!zZJY4h1
zk)5|!Rca5OwQy*ck_@?-ss?ta<mj=jV+5c0OYOg%J9Zos<?x5n{UJRm;vBq)dDhmQ
zh%&}MM}MFfM#=5Zy*EDAdEfi2EdRz`C+=q5e|&s&GTGIYB$i<}&RtIuww({6RSX=9
zHh?**8o<E{Jwv8vEdHV}U0=kvY}l{^C(e5a?OwC6FeZ#Afy0E%**oUm(97)Q21*jG
zlaRd7nKwH_M>qh=*eQ}TfT-eWaC8))h(rWB`3-K{m)^M5y0~fmP7gA%v<q1?!Qh8m
zTPKqwnr*&SOESt%Te8oUeXLy%_qi$t`<*joK6Xm{lI&r^xKQ@#mlu!FDtn{M!HP!N
zM_l{z`VnhZU1_dkv+uxJb;Ht<lEsl7MwXZ5ZY831v&Gcv0IT;A>v_#npUMn8>Bm70
zmFoF*tE6ZYC7<tViI!G8REab%9-dCCp#BxtEAbcwA&@jRKAN1MJ>(kUk<sIre105}
zTscKOYn1~qGm8As?6n<B5~C6er>hYWZh)+GaQ*sS`91TPdKx6fQ&T>(Nrkg%5}eFs
znB+lyYR&DkN%Y%hF_^4y1evGVutVF)n8CJWY!n!Zdkqj~^4`jPK=GV=O%X}3a+?hu
z9zX5Inr_OenK#!d8qsRU-p!+Gz-Q`%TJlOTTKXs>03@%(=e!EnTJ`C1kk_I5%c^aN
zH16@!-sV#}oj)N0M#jcHuUwT+*7ktzx3f_<zqLb0ygK~JLA?98aZ9|O6Bvc!L$5n%
zR>Kj^R^58MVWL?>ELB6JN{qFhV2z+44Gz&c{QT;&=^;L(R^oD8KZZM}qL4<pdw7f>
zj~17vkk#wgt=o=+m4Jyz@T(L@4AZg8Xoahd?Kx4(N>T=rY2!bY<d}eyJnc;~-33Br
zqOleisC@rzNAd*4BILv(at>%g>SRkmPag<qR<Yhwx;Hke$}LH5G4Ysb!Z*-s<?oR3
zRvqZfyrlCbt`lO_&CD_=OOou19w0=U^ma4mGjB<u<8w83lTClsc*s`l?UGtFV)l)w
z#^0j<d}_&&9CC%)0kzWiS6*KRw}^kX$<yN4yC+b_o;7d5{G;{CZ8$<}3TvsWC)$rz
zE%E8eLqrr5gSGhE8<lY}G5SSr!hyarBc7052gZ%Xr=#dK`w_mx@<R!&;zDOgc(10I
z*Z}Oe=X3-kjvqP-d-1A2OJeoI8*k36bK=%lmVx2|Avb4fu!--MiOTp7-~XiO=(O#W
zTe>3Y#Sq9xaYK*k>XS}OXNv4Pr0S%MPai(G>R0^jm;Ax@^{tH!$*PQ4&f-Ub&N1XW
zK`p+*4m6Z6x{ap8JeusrCG-z5TBeE6!p7Ufx2PQ*B8X+i>s$9v?KC^eK{qvyL8F|h
zaqLOYE4ys8lxIv#6X!^o%gK?slVlgOrb=fSz`_j2q*oRi*Y#VJyb5XmU7Ure-D1ei
zrNW}k=!^G<s5!{m556zYv3~GN`|IGb4{yfkzs+<x+dk$1l!E}vLZXB6che6+kMRvN
z>+zKgSUP-8DXDGDw9yR;pO##G1g$*N@WoT4<|0Pge;yB?Labmdtc-JS5pU*`kCR|u
z%h=fQ;DG~c;_V}TG%t9(_|v!-MXe5LO`Z6s;+R2T=eOm(CK!C-qpfntF;X5L$4W?f
zrhalsQbG9PD_1;rd7j*quNO3P1k5+-;!ptTrfYvmtAw5WuKZb`VgOgAxTNI0m{ejk
z2jUz?{};cziRv|(10YlNMEpf7NAN>wB7*Z-sP;Uz_YgPmZ?1F2xL+HN&1za=V?W1M
z?{NlmIDFnam2A<l^$#kN+xPL~nSpyhRhcnap93`VPtvP~nw>jW=VT=i>qrBS2XqG(
zJ=CI?k(&mRYMuAWWOyiubu7zC>M3zK<y^nECZ_?vaQ;P?$RdNx@cw7^1icyPn6=1g
zx{3n2Dg%$juZ{IC#9DC{5cVL$@kZ@w74&>S5^*I`zngcLIhLiWvwFV}HAxSvD>W!8
zjC&`DZxPS7>-4MNHhr~oztX06zmo@FoN4s7zh~5dB{cmNF<f?WPKYE<!Y^LD)7n4l
zy75>r@3)^nyETVkFDY)_uLPl*1mck#92VtQqJER9dQcm5K%Xpb_Dl}*`daa1U}jh6
zH5*g!+b`^&b>_u@j3<Z9rYF9RSifmQK$DLBIx+%m-G0-gO?xxC^h_m)H-Y_S^}XBa
zlRYb+obmBVO_Zv$ic}Ho=m46k<5I-`nlEZLg&e+*bGd_XCrNLn6h42Zu@+ar#GBpc
zS=s1aBt8Z8Tl0eoDX05yN1gki%F52By-zNl^<?xC<F+mo3X_iQxlY#|*hY<v`3`=*
ze&OHWwCUM`qxl|BB#f>BPo&!j?f$V$J3d_Zx>`=qhxDQeKI?O5j!JpXXfuI*VQf&A
z)uuODg>)$v%;qOnJdjb6-)cW!?QAx03{OEV5o1453kbUjV`bsviiBMJaUeY!ueZBR
z^5b8+;pndu4mBy@Qq|Aes&+ZnZr`WZGiS`uyB2v?ggafQoNHuMRc0N~6@{?bNKET^
z_(2n0Pffyuhipfo@&l2c8t^&eW~9~RgQf}RKkYS!H8`Qa!D8k(OLKQ~#`j~J&OAJv
z$YA-C@B2<J8F4>7f7p_e1s{jmr>3XEVl!Q|i6%JpgI;!Bn}1NHcY0~9Ty95iz6BC6
zk9LFN!h&9$rCzh9*{3+ZPE)EMv8J@Xx$sCL)nCy1BPRGwZPH&zcR<C!Zm%0`n=3y1
zRCEgTR=joL5;jIE;+#l?l^rP^-MlsokehJ-#fjO3j<r{9KegU8l?VOy{rfEh98FLK
zPS!0{UMYVqo3E^!G?|)XH0;Lr<PNC0lU{Y!v|2rS;hJHYpKxfRt<!4I3I)<^yu<wM
z#tCI(ceMZhb*4ajCcdgv>wTBM9wf9rP`IErlb$~sJG<0j_9VX-dk!~Gco^{W$B9jx
z{QIGmDth1AoOJ^@^flZj9cuqBbL?N`rH=^9tleYg^;Okib5B}5jkpC1SXDu>dExRh
zS_Q&=0(E(ZHN@M@_OZE18e>y5sy#&`+_Y`pd0V;NQx~WH<mPcym)fLKy7CWcQgvnA
zp&sgF)6!dz!CV(+Twn94_PJS+x-S;@r%k?{e5?vlf19hT>!o4jl=uz}<=jVB`gzIs
z()V$-NNK4PJ0y=BT*(TujB>bcQGzvP_@{m2dWT!>h8Pr5H<@Lmsz~G3`vykX4Zxy`
zahO!ORo*9b#&vDfa{d}#FK}J1#hjnxXYHk<>+-{48l_hJydJrQC(VMB$)l?Aw;t44
z;%^yT)OKIrqISd9ng^M8rIE}M+e1$<<S#clmV>?Dt&?PPE~Fw5h1aOo1(<?mpnI@5
zV2b^ZNyZwsV%<!?%Jr#V=i6Tmy1NveJb3m;?$tP?19PxVv$EaZN<!A8xpY$Fud9Y`
z3dX{z4N*0|RUoz7W`wAGapAOul+mx(AY|z9oZqY!N|hLQ0i1qWI_|4u>U|a;2{wRl
zN=>s!`~{#QxWYJhxe}hW%RYkPyNwjtWFh)Z$S|6=Z{PmNpp}});^e3#<bS}~Ot+SY
z1Q60N>EGfGP@Zuk;68s@zh%qdzdbaiPwwLUIc5#UFIFM<csy$}hJ*4XcALDqZe~<Q
z&h^nH@rePn+Y}clKepj3C$#vPTQ+7dyik&3BRjjXm61^UcEGH4gF1FJQrtPj+4jTj
zPI6N3;0I+N;N+L3`Wy4kknrrV3OOX@D<_~e#CLX82W3IhnX$$hAVDhan3-q%isCCS
zCT6%Yh9Y7C6{I%D2n><^G&5@`&9xV}ABct?qiTTD+1f9DX@!iK>=Z)Dr)mJ;vw(pq
ztAK9n*Uxzwc3P<id>}v!1>0C6nE)|xXR>HN{jd6@0rnv*hpyz~Qm+$FJy}O%P)psV
zZArxMrU+64ciwMG1&!qC3OmK4S@Y%(3DSDmZT^88EY81FO9jd>qa}V9!zz-cj?gF$
zz%?@_z3?YKs~Tu_=<s|>f+tA+Xk`o(sRb~Jqz&1eW@V$ER_4kBLbvCw?N=I)NEoK6
zX{d}rZIGYRFXa1ID$n=X*<ImrCgJ9CRL}8-RGbJ~I{#Cjbp~AwnstyBI%ZF+!A2n#
zk4T=fps`(|D(_}-{zxTP{Q>%&q_zKIvS-$X^iMC(ttjuEylSKoiXAED(Q0(;+I0+)
zuwKxl-hcfX93h?{f^5Lgd_bbL0Pi^M1rO@2T(hPT>313orUO%-_o6kLF}-g6{oH)P
zU6g$QR^x1J0;S<Cm=1tKBM|tsXy<76Gv$vLUu_IhH}?6hRQ8#4i?vzhzX}Sx*zjX%
zkyLdNK3+Hd@XsAPoSatQG((M6%mJ7rkBTy#dt;F<fEsoB)V#7^_uESO%NZm|-?Xbs
zj|wq}1fv5*(IqI*-m6pB&t#-GUEaO=<<E@u>BB3}N*q(UlPc{XW;KPtca$2O%`_T4
zg}<lIkt2a0L#KwGRQ7=u#<P+HR7+w;Th_E%8XD$#dd}rDWH8R-$KxTo`u?rcKfet#
zEXHr_)p^p%9w7Ugn(y&U$t@2ULC~q@BK|xAHBJLdcKH2dVElDtSTi0Tx4N0Ul-$Al
zz*QcTN;?jkr1i%uy@I;#*4wD8pmtI<P#*B~3qCc+TlT<b5>Wf!P)w$=5AHsBGJyk9
z)~Oh{6wK}7SMgz%(MJ+YGnCI^SFawTj2*kJX5$4QqC!DXX&1YFkyF3=|HTEUJvpyM
z3xyu<DbyJr;GPa}7FID_xqkhz6DJ;$9mTVWi;p*^S>KVoDv8WK<kWUDts!T?I0dF-
zO`0LE9n6Wk$}uVLT{LP3Spo|l9h$(X4S(7N@%_JJNGE?lvDKaqN8+o{IWb#Q4Pblj
zpO<DEsTSm@6A-Y62YQ%Iz5@sm++ZB21jVsd5^fzT?N}7XQ-@9N2T<P)hr2;ThIkNp
zI1a`!n!_6G%<7NEGOg_=N=sE;wqFT_M`+BKj(h%mGNcKt<E?5o>e@~?9L#9YnR$>(
za}4LBP!s6w_>bAIc?oy?KV?Ok)m*SZ(42X{^hLtM`*oDXKQ@wTbmF1~=*F~~f^kep
zyS+jqJwq)hEDd4@7)fW^T%I5gwpsmjl&*Ot(=aa!6aW&K@h+pr;2)d;V_L-=&kzSt
z9_HQqgD%K#$|fdEvgHn(IMIV3CN+*AsSZVN)@vRpS*>}XZ^BY6b7iOkU9_(93&aB!
z?_v1eGrQzM>NvwjMZKd5Mg~WtBmA2#cB9l3Xwm3uy2Tf;V?9B|9)aCmi5G{AHQ!%N
zAv1~tlj7PIFjF-T1&we@;33*(`M?vA+Z-MqE_9Kgo9&CT*$QUYS6K_of;vYwn|pR$
zrE}-b@o+PGjNI9%KBb5MU!M8c88c2;%&Pv6gTA(#@kj5zLuOx-i7{04F)YzUaGA#!
z>NJ<B>*O4_NtC@d2Dxh=L`OBzM{hS=_ce7J2b>S!9!Zak=!pkdIKNU^OFOFQ<+O6~
z@t_Xzg-ye3n7tPE3^CMp#*oTO+jn8U9+{dU3&<;ZM}Of7!>+o646a(uMi0)ss(cUr
z<LDIipjWqyj0`crL?*ulo~SQDz<u}bIJVtN9u}DR_F)h7CR02j!s*J!TfyvDspPnu
zsBd~ng2Bld^YSHa%(4HrqVv;=HLJLAXJn$dc!&@mq;8?1BSWaTqt9saJiegm-;N?$
zU?(YGAh6B_C%<^*iV@1kzI4<ew+y)){14Rv;_kF}tz7>71S6wqFPjej?{0VE;Vu73
zmiTu1iAu4~fWrBW|LFtsTb}-Dl>~?X@GAU&fz#l)U*$G<O-#zY>fuY0<G-D&WRoO1
zKl?sCLXdz)y)87*39m@-Zb^0&LkW^JbK&^$8c~ILTS9i%7XH-0grI|662Ba`d;2SS
z>VtRhx0rPrH!trD-19Nd9^S;MQEb<LTwfjgFtL8z&L(liD?bP`aZT#}@=~L!b4knt
zN7df2(f$s9U%su@e1nQ|NaSPJU;l7_w=17<&LYTf{mrw(J{1*?s?)S-BQ2{&!#3sL
z7SK5_9Gioy0|o?@|NdPtVg0b2XqjzuB=7koNoZGPJ>Jo(S1-xdx$f=c7Hhn^Y0AK*
zrOTE=VqQ_%Nf&*QSdZ|^6;Jk8nbLi#RBw}e^wS*Ls=sQL{aTAAeq6VD1Y+omy?XSR
zS@5Xqmq}|+ottv*ExYE0@#G+L0F^Cn0bK}Xpt*Gke$M$fm})O8kv2qW_Wgfgg!5UT
z12rzpY)P)f_$I?Q_67AEKi5GI|FbkBvN#8`H{bcAT^~XJ>Wu<t-}^#GT--n&W3Vfk
zaBRRf1oDO7GIo=IzQ6nMAgSU01M6X^HydhGkE}BBA|yMyoI^A_h!c9NL6vHLisFf>
zf1W!}|2)zAB8<k5kG;m(PP$q2Ek=ZLJe+Uo(QP>es(WoLdC`r2Uf;OyAWL^Y;zjrM
zt4Et*n>1;MURzN<R1DV%S@X-b*A|7fa>^;v{sF3oj8(ISI`5&dN3WNBs>C$qCyfB}
zP>b3`@DW1(gMlTZ<Jhmwv^Vsj|KGcZ<;B}?S2ca?vo~^6VfLHOnl|ZYYjcytG@Yso
zxK4V`-b2>j2@jsycg|ab5Xa7wj&0?O-vMy))?Ghpm01Jso6yvGs;A}$Ow1WvR`+a!
z`hS{L)8D;2xnx>H1?;oui4)!`xAyqyU2|@W8-Ni=3bFEMX_O<Br$5c6R6ZBYk~#9s
zXFHc|oBcUqH0PFAJ{>tftK@UMD#riO;o3DXZnJyo0#JFHqej9=%Jyx(U>os;<ZP#?
zVBDSmG~k{N2Rl|2>Jgh)XVfR27_=F$Jdr<0<N2N8Hqp5TR#Ef3rMZO~Y+h_h!4chM
zEs!M~96ohxxfU@p@tSFeM%`yLgY_ni0GHHdr;`U}&f%Jx4Rjhx#(3zHHo&AQQT*1<
zz4vJ$yyGEx-(1%%J@?Jcq8q6`!CG>IscE~XOF;be<e0Bv+)hXnA2#hx4z-*8*hjzZ
z&2_n>NCaySS{+|=d)=1XQ2<`4t3I5B??5@v&a>y_>Zm3W95wn|v&s0?bk{PBf(~@r
z-+Q)Rt6}zyGO-UYe&^QK^E-BqBj{$EA}(9{`pVdi3uP`~RrB-i*e1RGxxd4MN+34!
zik5kx3*EN4Z%PV`%K@iPFFmR)H2<eFZzlaMjk)(q_sQoxJs*b}D>yy-&|B7Mt=mYj
z!Tjd^R{XuIj-9)z=oI1`Yw4}^s;qL|lR5&0P@4zNym{5)XI%vdTs8OaA3>{M-`q6t
ziS<1e223C+B3tMz4IG$2k-zeDA^Q6Csp{D(E-2kCLZ`XbV#Xnl==Rrk354`#0)^kA
zsAL6HS9<*U9-))E_tG9_LR2+Bw-p9y*Ujp0*QVQ4ddxk}d0yRX@W!N$Ck8pSf>z^V
z?4Ok8#0&%`SDv=nuW1#dugJ$MZ%e|L<PLkv`72a~{BOw~G10v}-A|E(Wv51jTHM&Z
zyLRP7%t4Nyb`8&hAYBEN0p~XNjG@PDTbC4m4JdewsMl|HMY*`n5k0-~Ng4#v;B%kF
z`wZK(?!@3tO*R$&cyc8p)hDi%3vW>4!ek)+F1l7rj=RtKabj?wm4D5l!-l@a0^@H+
zMs`ZL*7D37{$2xl7i^VG@^ps`j!C_*_U7i=SV$yZ4&9!p9D%Y!)qk2RC9<nds#{ll
zRk1HE9kZy?x6i0TrVU$0&8v1OvLL4DYVqOTkQnA39=7S&qi>9SBd6)(cuylzrIGTo
zesgste)MvHCQ-=&T6;VC<-S%?B4fKU`uZI@UhewYO;J5cau*|2pbW=K<NW2GJZR-@
z_D9c8c{(GU)7`eVJ7v}>awBEpT<F1gPQH7iBHGq!N~5kO?UFypxKZ5a9(WHKSUsxW
z>jP1(I*i_Z2b+q0kDg3el>TVo_tb|jg+nS*XN%1NNm;kJGM&EN9dn<PzN#pcV18%5
z1IJx|{n{jZ!oCuMUNK`Fly-4Nr2%=<swygPUmbU^xLMb!!l_IKN)WH|zh<^Q=v<;6
zv-Zmz5ZS;s&D>u`Y;RGwrEa5&BIgP<ttCq|nhmOVu=Au*WV?jI&0mtwz5B7Lye7_(
zT{PP0)PRM-*iwDnbzipr#>)_&Jog%P6MnAlEJ4ce{Cr>>0d}R&_xyRcPVY!aW?D9L
zYBsn5X>GSpLl9$<3(xhlI87QNr|{4pKOT7X>vCgVj$il&*Mj5SylBz{M)ZJAAv38G
zbJs+j>OB?W{inw#Ym?5x-R?1=6m^c@l8EKN$ZkoO&D%oT5G|I3@`wvFSH3S^`TDwJ
zOFCksaMhTzY}xMina)06!o*h%=FVK2wN{bm+?BUazo%P9FK-Q}VmD3sGqyfNggG=M
z0w+iGY7zPTc<HiZ?XQ(&1RT2>m=jaNgf2lyGXCVf!xJaaBE0(v?-OrN4jGu|RM!9A
z-j;5(eySw&`rvSBq(MHs-#d{v*M|-0q2bZAr=h5oeunhQr6&LqGs?i<*z)GOo>W$z
z%MLYZB5;&w-K~Gn2(b`Fxj-%Z6QjePkRZx<Sy5V2*!V{DBIw?K)qlyeQ*%dfKpE_K
ze3j&4q(`F|{M75s2$*<w^wP@eF8lZDFCWLQsi8jn>5S`r7JhrF^Zj?>vrHO{(E~&2
z0%8kfLcjI#k--hoC)UWks3JmoQiHHXxw8;cb@k|WzILtH3j3mzk^#E7(k+i{KDIFg
z#iy73&v#@qR+Ro2(n`;v8O6ac*fl<9%jEPs2oH?4lRH!AOofZeiYG@KmX#JfZ_z7<
zwh-ngF$2A-=`t{`5Ag^*qS4wfDHGDiCN>pau}#XY!D0=7?k41#sc}Y|k|9(T=D!d1
znp~Ye*1coN!Q!SJvTTV(Xmv$QQf@~Nzr%@n3Rhs8^b4(}r3B4IJWqsw3_!H~=;ej}
zpUw4OcB!ji&M>RX6Z%Lpz9um7w68_Q^iE)eoyHyRuJ2I;RG?b@%?Zb?qnX5fVd=pB
z{bF6w@Sr%xxvFbU<0`5qecQE5wztLL+%|KT)&o%w7%hEWgGEUf%{KWTT+@bAF#B~-
zL&~Fz%ggMcwx6+$hEDPpPG3-+vSh?cGR^E+(CS6hfQDo3&=ZmFUtIU3h3HpgCM8+9
zX{7yHBj#%I#fqbIqg*{aYB6McuPLbofIKW8m!O~=w-W=bRlvMQy6vmCly<LUN#V%o
z^0MbwW?OA9kNSRiwo{DzyI4~?kXLNd8sKU#R)Y%@j%`Eo`L{6LuB(q>LWm4dqk>4D
zvOVX~8q}#eb}1T|sTD7K*uXr{vhRe_Clv)#|3<CvjgI4IM*CQvVcg^9{lssQ<6CIw
zeJp}Smgf&LN^dFtW<HTfabcgeU;0K@rV@4w$Qqb=${d;-L*VQSzf{W{*f(TS&cV@_
zU%73yDKm7)EAETp(4E|-u_<0{h3BVrwG7;MsjM($dNa)Hsere!>KT_XE%;W%=MWhU
z6!ks1mP%tSO=iETcy(`oYTn33J<PKsqfHWS9qpR;b)Z(ntq^FEIl5CTv#0nQMpY@B
z<1nC<0WCeszYoMB<g9IGZ6vn@sLm6Qd472|d$+XY*#f9Ff}UP2{Qhe0_hbEWk7z5_
zJBsDS742`V{f<3i4Ti)tCfH}Y_nm)#-k(d>0`1fDu03da{9=(m9b)n&iP(|BOPb(Q
zee`S@22Gy6(RA|d?aE19=VTXUo(9>pjBtufiP*@RF}-A@F5SfYd)m^eKAIABtiKA-
zhH1pCVX+hUUAh7h*LQWm#EapVV-JN)H$5|Ezs~m}_Vf1LEg}a-gUY6m`|TU3Rd<O+
z+rnSJVlE-iF+aUMuUB&Z*6@gk)SBfgp>DMX+9j$|uf%f3NAFVRuy8^YpL<|lrv19s
z;3MMp>z9iav=DV^Zi~r+O={FI5Z^i|$KdwQ&&-Wp(({RiLO*pgLsULJ@%CVRtx?p`
zuBoidGBphYLf7Q;htEFh{JG&jMEat!Oz<W)hq1;*YB>Mio(aEgd?CX5vU<B(oNu-`
z_1v~05q4VX2p(&`E!daI53w1QxOH)SY!x!Q7wKU!57U|}^@>vdX46ZtytJkpX#w<z
z-gjs(f<JB9u|u6Su{!;P1q0{!Hi)+p7QiSbtiHNeJ^iYkTXeMZv7$2Dyakd6ML+e~
z8&L7rHNdXPASGkht8rs)d#6Xtowv9q99h{+%Cl=@8+sQW?=ZTK7<Uq7?jj~zJFlAb
zwa91_snv99e?QYZmTrFEQ_jw=6+cyF`mPHAxJ!9ict#{jEaj*DgFo`d*2`)iANV_n
z>ZGk&omRKWie&IP(iTnU^1pIog3au&cw!|ZbPyv>2&;1J_wkJFcG1E5{s{|*E+G*%
z#g}O-4;h^<T(P-;W^$4MBKLXc{UeGL|Hbr)uq`Fcb+7UCgx+qRi<qlXg}Q|7p0Lxo
z|ARw4TPp3qUZQLK0|1Vs2&kOAe17&`OnM{Nmq)Gl=I(Iiw*|sE+sj@mGNvgX?l}Uv
zwXjyC+M~ZedC|MO92iv;R<Xa=NfNLb%a>beEo{R$S>lIH@;R^JYtNqlz!7a@60H|i
z@<q#a`}rjzHNXA!>97)c2*4yRvtzcn@7#IkQPbXv5}I)yhYKbUt19m^HGCY7civFE
z+^&1cswXFhwd5EXfe@_E>XaU?s<E7-#g4hNloE92E*n>Wo|p0Zx(w~WLN4@-(A*&g
ze)Vo0HMfdO>eq=fI7S^G-Ob#)8O&yz?O5Ptk#=nT+NbL?l6~SAHB_p)d>k0;YjbE!
z%6W|$ur9rCl{!@nOTGj(+M(?4+pnM9U@V=2(N=@HJWiPdC3yv==+dPIDYi<#GjQqc
zko1T|SY(#b(^R%sX8%#+^(UO%RRG?CrG{<I@{yxQe|YDrCWj*IRAO8`iwjo%19R`W
z*N}MJZ0LqQHPHo9sVOdYQGcpW2FW_ud7_bt$v$nnWbSs~tN<TH>Axgd_2-SrBj^*&
zU9`kw=FRn5r%&MDym@#PwL3iY=H187J|w$qEz^XE7u$dSEPCpE2$KBsfQc~=URpN4
zG-%a}u5|w`?aHq%d)1D@uUXlLy~ep@DV1Z4#bxPLxLs&?*C%u&f5PmD0o}~q_W+}j
zDlQqOTl2HPVwLK3nz?`;%zgU$y4A9=K4^1i+_Y~z;ObRmf^}(>PIf4oMc>f1dL5%k
z%!g1SblMPsO^?o!SB;nkaQjNmbFaTrG0~OYB1~Yo@5c6}`|2Tgnu|xH<u#lA4Njpj
z>@2oIw0vm}XH^L`DsdQ~avM&oj19n>(d^UfM>$_=D%T2A4BILq?G{3j;u+n!vJQ0k
zUHjElTUV7TYKyZD53C3VYWP)LyTyRJ)F*xCm#kqk#S~A-0A7+2Z=fWGB#nB-L;uO5
zqQAoFq_xq2Tuo-QYP8jxVHZX3rE6^WfJw)<LkgewB0I5|=f{8)Q94GJ&V>J?n4I6R
zQZV_k3EYfPyW8x>Hu1FKpFCv1df@sOZJ&1*_aE-BxWK5~YEzybICq5DJ9w$8nh^FW
z&pIsqZbeN~jf851@11S^@-97dzU%4ZbO$}XDXWz6-xKUkenWKe!?~i$PhA^O`;8pS
znk7GGD$k~=7c8Sc!Nlt1#hK_nYTz>soM?L8tdq|2q^giIWhhk>@D19e<BtuUmb+#~
z)6AiDa|_S-^?v?#xGNLKGzk0N-QMZ!EG|9m<XT;k=p&D>|Kb<7&m{*V4={5m;<RX_
zz@zu|3g7@0=uxbyIkHdlql_vNUKHDU9^uJqSp<zLnV?X-y>=k&SMVs|WS*ZDUX#-x
zZtwhi&0c>UaM~eHc?<UQwm4;WbMmDX1@VnVgn*_*lxDPEs=Qrm!aSmtM|4(dvP*?x
z#gM{wz+vzTjmU1rUn%SGe6Tptcm^b#7|vwjIdhlqX^AafT9UtLzV8owlU)#m1m;U}
z=w5Px4j1j$z6hEwBT6hW%LRYmp0@hat@Z8e|C2_%q7sbp80cM(2|}M6{ign$v}jiU
z59YcZ&x}Mmz8`Yp_~W{b#W9-`Uj}gPKRrrLI;%{r)1&r|>NRknO!s*nFhMjt(q~_s
zZ-eOD*<@ID^tgUi`*!bMlh%&??~hNtr>iTA9mb?j2{M<uRdB5-(4}6#3=`@K$%YJ6
zfZy67uCYs7j%#kU<Il?@rn#OiqmQl~H<VU^xvEQxwn<Jj3?&!c)pAG^^5{_@CNeOn
zZ##vZxi!?s$HK|WcpHh}9AII~l22v97Qv?1R?J<Ko7iD;W_0)Z`->L!d;NTU=l<s@
z&CB7Gcsr0Xib!C`fli4*^GADIA20fm=GxY&SOp}V9#6ISLt1Wl5xs@pdk->wUq~!4
zmO8AuUx&O)W>*{^uv}cT>oJ6!6I*P*pr@GKsGWIpeceBKpGKnY4#T1|%aTZcWbo?F
zhZqe)1nfYH+0{I-UP;|%F6c9zwg&%$RL|aG)qjcKl&>kavuwSxSBK>mWhTOxJ27-^
z_shYC3obIK;BCs2<45|gUbt{-r<qH&SG2eR>*gO7Ur7V@Q(XalePi9;6{c)yWmHkk
zRju}&%byb_fB$8iKe$1C^O`jbLhYhQlB)H)she{EKssSz9fQhc;dee8EVh~cTvb=+
ze4m5t8VDO}mL+wp{OD{Z!~+n7kGX$%*RX$j6>>W0^=qD1rB6ZsgOzbc{v=y}_kV)x
z|3j9z%fsa#0p&eFVeOkWjfy|#Om;1ro9>@5-2PRs+lAwN+;+{l`r&j>Rl~|V(srW1
znf1EH?nwpp<FAhMan0X8^rchI>f8S<imdOi{`o&;jLrV9KloMH<wnK&=@wROj-EP4
zrCaw2!>cyFZrP!`rbVMAcLUuU@2<9_#}32ses@PL4O?|Oa7XR_ox_?o>Xc%mY8u_S
zg=uJ~b#a!8f7k8~PR_;Y4O3?>Xd1Tl*6f7X-!m31{QP+EkVS<jCoTLW-^{fBe@f^6
z7eDOZ5BYp2${jy45NT#eiH%f=ec*q4Z3c7&|8a-6Bio_kU7I)A%&cCUHhX9~2}~_+
z@B~95`|ptHM$c%vibSq3ZBe!59m;<1s3DdBG?!!1#?u93n0W)elI#vpAnu>i)ho-N
z9B6)SXa$6pfZs>OM@yV{7=DEcsJq2^cS@r6J$p_Fp26uf+cJWE&RVYZ!R(p7$-;%5
zX?y>d&k1-ou;2heB_KOBGat<knDsImU90~Ty&*$dl151vfYzS4&5Elu%mJgti_cl4
zZF$`?)FY?mkV~7zDibj-eggBcK2uysO-}<hv7Bzxp3hub<+rpDj7E>1PkbV?h&#Ug
zpXOHM1Hf2PKjE|`YB1O+;(H?MbdJF|Muz^3QES_5X70~-u`9o(^o-m{njj<2Bml?|
zV0a|rC;jVtM5dI1P!=}5TYaBgVK+?nW^X!ox6g2{aFoe_<h0otkwfV3%Hb?}XI>o3
z8Bv{KNKI3G)nC<at}W-0?jIPTDJ4b7wS<T)h!r5C@IEQZm{hh6>MaBnA%iI=t1CVg
zw?<i5G{FHK?v&ZS$pOyS<P?XoEpsEqr2`UysEdJhj5cf-Bn1I6s(q$CeAzh+L=}?E
z6UzOlylB|#XvSP1q{xk?DUN*+=M5R>orp{)4>7yQiN2T7GSur<XP~o&ij{MdCY8NV
zDxv)%(BZ*6{fwpF8IUdEnsi#m9fFaesaf;&NuVK4ALGsXK0Y)q)s-~PeXSwYS@`wq
zM>=8aqiPTZo)VHzaVXpu+B1)^klS@`b!C5^?^4wT^4$n6XYzGb17QSsl_SR&YK$lV
z5Y3&B_`!qUK`Yg)2v_mN=IMdi+~<ah8GN;ej~|;SfvNdZ1*;lx$MQk~c(tklw4#%C
zX(vv!`d1V&7?!F<2E9{y)Xn?dJ5`9!#0<E@S%}B0dj<Gq9h{_BU(w$1b52M;quf@m
zU*A+&#g#dH6}ARGMy`7XgAJ&(V`&_wBi(8I4dW62#RV8PaNtz7O&4VWgcai(eT(jW
zEqrxYsf%|M_d(VyZJ0#p|E=_$T8S*u%}osf%K0f#WPV~rf$gnn;CF$}(70FyuB~*_
zGfh0??Y$dzhfwCZOgD>@RHiZu9t05ZIqj`(?(6HW5-O0IWxom!&d}x<5lLUu*$%#5
zv$#KET1P2$hn2TRuZyN&oZysw@t<-{HyZhzzOKX;SRZX*^G;2I2VUu~H)!R1qo@L#
zL&FYrpVB?^5<r?5kI(_xzCL5)XszD3qrmhrq7!NKaa)0SQA*9mQUpi$&RrRlUB5o8
znW}twtDH{8o(Awwm1Zf<&y$igzT8yiFpihSJQpq!EnVXqgc8K<Dhl2iCP}-m51bW|
z8dY@zJ#|v96yrw^9}ee(BvOgZGv9FdoH=i`mfh!z&$2D6JQd9IF&&6r;S&u^b_WB{
zK(}a?WBU#rs?au$rKHV2bR8%IA(zbX^d1WBNvwZ4gYmubx5&T`Moa}nP0aDHeQkod
zPGm)S=>~o-_%zI+xSpc%r3*_E^FKFCCS3wbC+~F6=(Lv`qqy+)Up^i^e*83PK0QXY
zUNf#K@RcL)Z~9x@cYB@IH5r99{z#wpnT4i_`F=a*ZZFK-wM#?Fl;)>LRwXIl0_%+y
z;@WsOjKpDy+0Cr9Q2+Z`zrr+Wm)wtt_nidIlE;^FVWtPm+JPt*e|SCZ4nlmSYn+T;
ztwy>7*fga}DC}S33~UHap}l6t3|t%Q4)+$7@timxw&EWS5mAa>+I?VrIWN4ik{~`)
z4G9~$MvWS`g-|=Wi)`cijhB%&Vr<bv&?2T(yXsw@8(Up^gwP`s`W34v2oZ=-y;iMW
zy*5v$?lGy^jiV|4_5OSmYv!>>gAC6kP}oSiggo#Rwb7qLlO{}%j0!W6+F0@t7-C`c
z35OA#U3f_jBvf@}W7M_N6sJmgowM2f_~jFLzQcX6C>gR}rkvwEsEdd}28#6V5q#8_
zhSM;Z8I4)Xqa`VZdQYD)t5h?9J0=A(IjG2T;m&LZDq50kmmT@*Mr30ZgWa?Q{ow3K
z<p5+c`(a->@~Hg8GL?V`r(xvMrRNVs!vw+*=PfmsK(RQ8>||>RFPK<orE-sx>=bV!
z6N2QJc^e!WWk|S8q*dyA*8AUWy#kwjC>nx?LniE--7le6S!;~<Ydspk`Ge}zeo>9Z
z10oY1JTOT*IkV&@Jm(Hix0WyC{`MvEixCavK%sX>`HV%q1*cbh>bJVN$;=b+!{e2|
z(iZvHRx^?*F$THcx-0RuM*ql+4jrh^EGUQ{;#-_Z^JD&`y*sbLViAixPE5-vhrv{F
zc|GeYXvMWSPx>r-rE?y3skEgl%8T@|jI0l1&R&1W#=HfnuKg(FaM=_w1itlW!cY7k
z<(>Ik&-wrFUu7*BA7h5nAeqc4B!k8tvSrB{nvvyQXbi>@()wXa2%$ts6xl+nr4^%6
zQQ1O>Q6y5{rKr@oJu=^O&H3T{0p~h%UGsyvG`-)i*K>J1?%U((K9?6xE^nS{KIW>E
zyM|4Y(ed3?douQbJf%({nCJe6W~&)mfec=blopbvWE)|lmLrBK74xCet^V@|?U1xt
zAly56%Ek_~UOTB@I&8@Q(!`cVlkt(jMrGW#g{$ijv^uJsSMvX4#rOOru4lowmMmbI
z;z1Fy65XT7LVNV+0W>ph(d41Eby3(TMW!qHOM&&%f<=YPrVw2*hyO=XE?d_;Ms+WK
zaUmz&&3OT$`{oho4qEdgew5RNMfcMBGaxrW1oO`*%OnElNciZ6ugOIk19K$x2cC|u
z1rI%xXC)<Dxk#^&*C}ewn%c7Do#58gIo&u^1!u#=dVjIx>Pw#?@w{lK_|lu{5|-Ya
z*ejhCn=tbyMML;Y#S;l!VwGwIxUHU^9zLdS+|)7%z9^d_4{Z+d1b(;Ux_H!0b>99d
zTJqC)ya`oRjdQ4S-*ht+xxdT>;r@C-No4CXA~7AVwzX}|X_*fq+nieQ<Fid|reL#?
z0!b2@c&{234^<@(k0rB^Og?H&+qD+aF3O&TxUnUL5S5Nj{qM{Y#;AMZ+h<B})fEU_
zMWGjKqGTimINvPi8Vk0$w0KKP4Ee%8@*Fnhj~!iru6YV$RP7fpenAq{l7ghf5r?T`
zMp1aiZuIDg)NdrnfsFZS6mTU|NPR3heeOI1TLpTMm5P8?WEjVM8}H)#&6Mb<DCODq
zGjH;G=HsL`<R);c7X@dVlsjGS`ul%Fq%SU#Nk16tq=Y!$X}nY_vnv5c*r*0Li9m?b
zXq&;tm*mLowr$)mV&%t-o5D%tKYUUjdfUMl*OmrFnGaF~^5&Eq&AXHMX@U{wgCA!!
z^#WJYP-?2n+Q{4k9LT4aPrAh1w+K~kvNdcVabkV}0bU$_dT70R#9Z$G)8zgWJQyFS
zQJ2`^Sw+9sH5*@XZCzad-jHOa!)3Q8*Hp9KI-|#)!hJ8u5NU>9HjLLcH|5G;YyQc`
zid*pGnalG6u#%pC`Qb*;e$R7z7G?XQ>C{ElogICU9HD6g&RG;iFHLFOS|^IrZ{EZm
z((?Yfc{@4KisC&ADrCUMV?xg4$~ti1z|+1#=CR8k#b`l1&DcS6jZ{8-=2~8j61+TI
zo}KJ0MLt%d6Y0oGtM)xp#&&W06Rail(c}hKxqg{8jKtlO1)BTJ=uJn^lvJ^B;gwCO
ze(EylQ_z<FL4od4Or+eIV;!V7r6cQn9Kbu!_%nR*(`R9IP(s*JV`6u0J<CUZPOgR}
zpM2rMg}z_uG}OE~Ckin{tlQum-xMyjt3UNRS-3KU*MvuTvb&cC^a8Et(A>3uzgEh}
z#`8<llj^#uawsn^*%zQgZ^kVm92JHYI3j9$>E*C`)hePQI}gb@klhf(_-aeb8U0`y
zTJxH(QD(N*u77fIWfLCNNOBTb-7O^tqC?wZOYxZqVMN}o>^YNK%SnLQ=m=7F8%^+k
zW>8-F=`}$WfTRCGVA$2r`Mw+U6@jnX6l-HpH8WiLy5_;5pZ7e0v<3e9kxX&c*rBp|
zKaLKGDVfiVOs1JOj^Cg={0bTwLUXpsAJv>TDv2D5iJ75kz&atjqCKOZh57$&S>T83
z9p)MQ^iw+^^7}4zv<!;pg$vaa4w--38k^@|K;Eze`ZA&Q`qsqtQ{SLOUy0xqeaJ}X
z{7PD!a#U1?OtBCjkjbD9J!gBO6%9dRn2~i)tV<MQ0y#pth0JUK{}kY{-;x}~^SxBe
zHq8|I5iX}I`T6<9Jq`w)Sh03(+~nXzHVfSQH3?ee=lm3C#b4t#^IoOoSXHDQ6&>xg
zSEI||wOP;Nz8q=8uc6>FWBMc-u&xZqIqX)o|LH3D$Eli~xNWa+a*21xweEcyw?@te
zz30*O)yWMS3Nr54?ncs>;b9obeIdE6L{N$8O!Dk_8&bb0*7L&rW<3e&wH+~&y633T
z`H0dbsf8C>_$jUgrOk7urZy{Iv+XQ5^c4iabDF)v0|OJRf}uY_&iv)IrP?&N2Zwuf
zMz@<Lnc6}Q@AoPI)McWwq;GQA$Y^aY&yUEorzX3*`F=NW+KONLRX5pmuc^$QW{2V+
zX@d6g^ab!$r<`LGIw%~NEy|y|lu42X;-DtwYfQYbzzOw(fOa$%SBa+(s*b){97$ZR
zXYY2<)7wH1x&S&x_7)F7)KZ*z(tL{!3({39WPWw(QloR@Y3G}I*45XR4RR*DVK>E>
zbZZpgnN*zq2`Ec648&3Wh|9dGC@z0=<Pz4jXy5zS6Emy%JNU1)6v(p~w^<M-Vt(GU
zCp7HX@#9-*@+3ZD+O*v^L#cr~kRhP;)mJvRB}7iKhZ?eVfMjp#srd0}_11d{yyR27
zOxCqY!@P#k9$b>MhPCV5xp!kL_UZm`UbTh+vvG26P_KlEshy*rxU+2>V1jusFGyN8
zC4c940}YizUR(c<t|ym}YB4Iw4c>2VkHf*iy_LThvK@YwV$hKoOqg#Uh)lvt#=~FK
zQLJv*dC1@Ydv@Z&6)Q|p5Q_(#1N@TxqD3zs+Qr)Tg4Z*3dpe^Z8^y4oK0n;L%Q07q
zNyQJ;dg;9i#<c7o?vcdZY5X~oYF8`*jiG5P>6h#gs7=EIC{Y|R1@xtXv>7*O%-3Y-
zyR1m)ZM>_Y;a2Iatn6$?5q27x?Ecw*&Ke3VN(OVuCYA#m#7M$UW2*|{Asu)CvcTBA
zP1e_>p20U_!2AXlVRJ?38o_1=)!EJ|O)$fA?uZmtxOi--iUg$qifN%xu*l^s^Nw+C
zOY;HTDRm-!{U{(S$yAbTMDFZsf}NZ>(^LBVhu0(^SSGnkF^d^X@@wVqz{=)eZWo3F
zca-W#@5NuFHy)iaO_HA0z9h4~kKJZ0zFj3{(bUfM&ZMq=AL07=gOJ>aIc^O$=iIc)
zqdUw!Hth<f53-z(g?p?;AXIekL+U92DFh{3p^+R^xqRFOlr5kLqmWwPxAUCm{>iAO
zs&pGUYn`bkMsxToB3~n>JPm7uh>1yl7TgRO_0=B^yKcljXY3@>TA91Vr;wp6EZr1*
z-BhVkDo9D5IAQmfO10nxx7yEZ-|F&nRmTsEw2xsqk)KpD$$V4#62pz0S>1*Vsheb8
z6jA<Y(LZy}KkI*{0R7pceh8nfR({GK<iVBb4$y!!eV5ferjB4Aw5AsQ_%=Fw{%<ul
ziJC9A&sY96m&N7%;C~^pUH*;48vYxJHS~=)Xlzdsug@^PcVM|@TkR;PA+@~|TWi-m
zYEm_CgZ?)!mk$2<{}{3I@82_h{DNCQ{O|C%qZhOn|9e=w&i}iANHh{ClFphoKk4=H
z9245ZCj1fd!TgE8p$ZBV5JODC%8Rm4_6G4!&oQfcctyAlOpTMEa0WDQqSHlDx9#o&
z@c@9cYI@Xvst3}pTyX{HI(LE$p#rMJ;`S9b0PZfC*~K8w5wIHQ_0-1?jB7BU5SN09
zo2p>Vbi0!@IY@@vBAZzsTKcoI`g_2mnv9a2KrtgnUNt#=xYDIvb#DIca1CP9wK(_J
zGdfOmu532L{&5)Vbh%I6@7&Cps#{%qy7Qi=-M{Oy#m8>ASP!&+q#u#Q#4o8+o}Al;
z)WIBc0tKLp_M(|HYdNmD373dOBBDV~czrx!8W2os*!Vq9*PplN$Irc8KzFatFLzCG
zYx49jx5~ImM@6a>1)8RH<6c5h(&1#J11uU?*sBA}z4u<dTrjU_U*&-3AIDi%F4~v$
zy7_<Fer|UNRvP_{a40LW5pV=M7G>`<YXVX^g6@wxLsG^PkL>+A(T!fT^7sMe*}b?I
z*q+iTaO~!Mu3YQ<OHDt;-oZgbAsp_L_b=+Qsk5tV^Y8AoGupA!(sD&}S+?JLF7h7D
z<>|E1Ly-|&9G?p@LoC5@V`H~r!yxI#IOJ2VUe)AiFMt1DUec;ntJaj%Pyqc$SGOhX
zA|NkOK<zm&(Tw)Jm?*s<hcgKu&0sRGg0z4b&?x2pQV*aFdS`O*;KBKhj^dJbg@t2~
zUy{uwa0@bSI?%|GfBZAT8Qz@<DQok>Zvnjuh(L|-TWm*=e@So##W+6m9Zf7tQat78
z7`=EbSCEqYgX<tC-9>|O{(K)q8b_hrBq)y;5I!@Tlo#+K0VjwZYDO9}!ET<Rq1_~g
z9e(Y%fDUcJ+plra!18CrzC$Q4IXR6LKfezIFx7ha=cb(3B9;Z&l)x_L!=Lcy1@(Z-
z6m4g<?K%J&B);TW<i%SDnx)<5A;<)9sY|3tB;{6Sf@1;X)A;S*fAT^v)74oBRSbIe
zoOtOLg&$8`GRk9XoRA7go2OWCng^`T>vKGr)=Z!2(o78Sg16wR%jKE4R#(x5N@}mj
zQK<0)td2#86XKVj19^|{azuoUB%0BF_bE<#Ll=$EvyGz5WL<uS+}jh=-)%EZviK5T
zrUAGo0}jg8P1jXfFk@Oa13JqTkPYHKil4-rH<F1_-odrF{^Xge3&)FhO`bKDs^?UL
z+`0lH*%r|Unz`4lhf;41QtF8onINH!@Os5>w|t~NMl*ouqXi3%*OB5;U(rQ|7xIct
zhvX$x#m6(?i1{5_tT@rF0piW3MG3CGp*LuG`$dNb#BI-xj*Cm>s@AA`Qr%ofcghk~
zCXH%al1LyLjbD6>FH73S{Wtpb<LDP3DdKbRj6r5z&B)LeHk__xP04`DYRiJoo~gbC
zsf}~(Mvao7H$032+p!=d_Bx|*U9$e|ojQ7ZbV|nt)vD>Qn4A=}Dmy+tUIj9vNIu6A
z7!`@W1D(UeW6E0``>L+>(Bo!quSe%)?lary<D+fi<CDIG<<UV=-x^#tZp;|(D~B`F
z(<QfNE0q`1GiHMjbcaTg2x$!iO5fYY$qTP*i&;~o2|NTNI&OcCtgVL*9pbrI(F`V~
zVH`{V&zlBIY)_pZMa{i`s7qWz#jU}b7DT0;o>@<ivvAq6UThY1?$|CjXcpM~&<U&u
zQ<}Xg8FX56m2vcN4kdR__T|N`WHtrL<r|<*{;QJf)Z36%{VKDqsd?3KNEu-9T{%74
zDSfFTj=iPq`jY<zkZ+9DZZF+d2FXJ>9Z7ehavv%l-naP|tZBVXSx*E46++caFVVvj
zr})ukr1Q#5z+|I{qG8Gtj!_PDD_W&#?%YP27LwWp(iye>{X&V?;xZY3X;o1n2f5Wl
z4t#CJ5lDr3H0dlrBQvw9$QZhEfGk|Tyf=l_H*I$jScjGP=RZPjty@jldt^+EFd#Q>
z-VFGk;=b9Sy*}yGb-fg^P>OStQhpB`YAgPJ#wbrv(&fl~u-JF#OBj>u*RG910^A*!
zo_HxGMw$6s^BUvk0;1U4{+y{MZl(kQabI>bO#M*J2i;1(GN++fb^-Gy%5Nc+TEd$j
zG6^gAmB77Qwr>w$iEaf-#!hS{7;P6K#FW2qFD}Xs?#0-9KOF34Ozpwfohh8b`p<F@
zRVHJ+ke{YS_3;_UBBCuBj|)Sx*|_Vw(**5i$!Rx+B-zaMGHVV-v?VYw5WxNQz0-kN
zLH#GdS9K+=O#Flh1bXF=?(L<e(6a26*Ti>{$IFc?{$SRC5j>HlC6(`WWL#v$F<it#
z6061T%-bPTYqvzj#16zqtc+s@lz2+j62tLg63)Eu`}d#OWAN_}`ZQWPziylB(#65S
zVGG!e;wP?go`b}%O+zf>2(2OIS<7PyW-k^k_~59@Kaq{|)c!A@wzh6%tT(T|aip(L
zmc%-EHU?!Fvd{yx+I9B{4Grb;iby)5m*(c?c0Dt50(FHk2!KB#SxL@Q{3JE)bg6fJ
z@Oo$G8x=eIa4=+72POcqPm-lChGR~!bUx7c#>Q53w6JcS2HRNLxhCsB+Nsvx<NKN;
zB8uYmHq&e#Cc0cec{)>egi=Upu$Q%xLBmqSv1R2UA%pin=@0I#|0qy>*)DZQTT4s(
zA&oVP!T%6%;xp$t%i}f)3Co6tY32##OY~1W`#Vj01B(8I(u-IVr;?#Cf#B9ww6W@C
zl6-Rv10stmG{C+3-CVI}&YCqVqLSuS;<>*Yymq3A8^N!d7J&W6hK5x4vpUb{-{zB{
zA;j9(w08*QtZD|AC3GWc4s}dS$;8Q%+f(!~arvU!qOpUFIqT)!eVp1gWDOm*rUi>5
zEX=X9TE4Jg;l^D$b~IM(Ss{joYW(`da_p9Pv!-dVBfS)oG>xomrDM!-wrJmWi_9JS
znniACDE&7Cp$~mC{mIgR?L@yRYRCuUyceG~uTkyzzzx!4?&TQHfE$t@kwMWm^U~Su
zc{^X}t)AF+7s-HAlynfSn2cc?D#r!&b2L=2_SlUtyB~2t@dGFL(J{4Z-OW#up>xL=
z8hWlHWe_;V3$Ir{8cyS`iHwT6Z8d?f%%foT4LNB?1F(gkYkIXJ^tG6U?^>N>tb^jm
z>5xH}8S^NDkTt%OmF<G<8bQM-Y9a#aaSyjroZ;JAt$D}YLY&^r)5o298ybR=Yg*`y
zrIImH!ub7zN)MQ+ylI`RC_Z=}pmg6TY9b$h+fM9^D%BSjmy*Efrl0Z*vL81*nE|Jh
zIAeUkkSG6f^TEpKJ~c6s%d#>#5S(hBX71qUkiYEV1kV5yjLG=Ty%inev9O_ocX;rL
zY>V1__p^Z#2E{zcNmuQnc2#m7ywA^3LSEL24bW!8^5thCTBpBpt+-sUy>x6%>i4SJ
zT`!Yd-G6_8y-a-w+V|J-0P>8i>M-^6y^Ix`BGdZCnZAy<<KqGx_W70A>=_DGk)Vf&
zt_1@|y)29(GB+>O3DE8S4p*~l_@Y8<akNNBY3o5fu6WW*Dk&6kL=F6xO>SspH2U<J
zJ$u@*-lb3m0(j%Ktdx2Xz+p3@eP#XSn0Kf|q^5lQ`0*d5kNQhylJW<ssLZHrN*6Ph
ziY&HdKTtV<{U#16erC=<MK0f^CwIrk53YLmWSa=11`Zr34S@{N11K9irz%vG`G^m@
znwYf0d44-X8U#>`(14e(Uu!8SCgPt~iv6g9X#5{0PKv#dP?~e3i87zkO;IR^WG)WW
zfH%QrmQ7*9q-%k((@+4tk4}C<2QJE!*xT$L#F1-DJ1Qw8@p7+@9W$mm5|_}KYoWm;
z$&DCNr;^_3kzxjf&T55^SdxTeGbUp@Do}k>xK9xZk-lBrF}4mjv$Fh)d|;TJpGW$Z
zmeFo0!M&%tK*M*Uj`cyTq3@Oyv{r(4(SnFFCq@m%*Q{#vWM((D>nSq=C=PM)h`+hn
zsd+>@Xad69mR6OQTQD>$O6TJFR8RPVkr6iC2yHbPN<JQFX-2GZ@ukh)-oG7c6_Rh=
zuOLsb>7XELYw8r}tIcHkDnDj^+T2?Ic(l&6J89}S*4ee-f*KZjE&3yx!x3(MNS*3?
zz}c*-_NTb*Ue|MKhtRG`VwtGi6`7_iF>%5_<P0Cf7~7*zN1yI5D=ynItXZdI_G+|L
zFZUzS03>~``URNRxM0vz%i*^f`kY&H;pMt0es=w<L+cs?B#%(@h!H+2At@xSUySK%
zRhI4uOUq{0=w_#C%$Yq~NE2k(z=?l>puA_@{tYS9$iCh`vxV)9n5ebl$BO0N=?W={
zn~)5$`!#o(#`bGtOwOEA-mkpMbq+vjT|vX=E}r2d;_H@F8OAU)nWaG;y};3N3v95a
zr}~f3hk0pSRN*@~4Gjzo6zZl=dzZnvL8!<W3qWNJPqjhSkek=8P<_PQ&I~XKpMIYj
zS6S0p6IP{jSa6#^HKuf)5}deJq`TZaQ@KaEC~&%Ja>q+quN1o-sHkW&D5ao{S$3QQ
z*}%*}mH4%Q7+GGNPNTu#D&KXR<ELsA%a$K*s2dypJ8KRRtzX$n-=T|aZ94*YN?tLe
zBMzyZ-IB9^ic6+$kP%VTf@8Rs`Ve^ocT#=Ke6Q&ekN$g;YLC;YDfLNlR3Y~E_R=|g
z0TsM#f_WCK!D9!Pj>vxE=PoIuGJ*%?6?||D&M_}4*pj#H_RSkiY}x{qD%5wb_wBb%
zC(54Y9L%fG>11B9@NM-hzX9!=k4lT=mX{ZY)n&77gd}nEm;<uk=y$^+=(O=-jdTLv
zn>N!+uDAo*Ea07FI+5}%L+jXH#ZAN26_;ywrB);#asKQ+>_BytYUdHR@m*A@Ee#Eg
zb(5G7Tx_%-pp348c{U$G0(LPkLI^M6?I^s&*G3y+tUo)Dv0VVV+0nybFqA17jIl~v
zLS}7nQU8XS!%Zcyi;Mm_wq@}@^ZK>=$uX?eaj4702@^71vuW>wRt2esSyw!JUzgcN
zi0v~Kqn(}aHT0ZhAR{>-s%DoFjK*=K)3IMm%9`F7oP<B2R}lty!yB8^_nOB_z#LsA
z6EjS>-=^SVau}0|F-_;xH9wW(*&|7-n+M<Od*?!KZRU5;Aq6Y!KIlFwNaXbW`!B7m
z1LkcwQm)YDn?ae2n^s=4y?vPanWOL3L5J&fZ^vELZKA!(THocJ_tO@@d7|Q@1`?^S
z2+b4;Z^LL6ifX3$g>GyXj>YC4X+Z<t#D7(@LzQ)Le+O0aa8>^7**c%=p#nwp!I?du
zZy=p8)(tcQYZfil|K)<&=SpYU@7TLGi8G0R`-`STbh4xg$*~DZU`A4*>*BD)E$(5v
z^2)-xCLeA#^!Sm2b#^+Px-sr~zKAM~?Ij`*C7Fz<1&om-E+r&uhn2N;cEy6%bAr^Y
zya6W?hH*v6^lOAi?}6B_JbXCEWH^P}n*y3zC8Q#5o?CFQrHdB_E*?m<xsty6M%c>{
zqo$udPjOr1JaLi*G4Ra+xg@Kw{NAj2jsb2D_uF;s;##gDkS=RwAt4d76E_Hx7}E1(
zZ=Il&mv9#!UM^h}+Wz>}8uwPW3!XfF+y^co>)^n4<dI0qrjR+*F-_*one&CPKeGPL
z=MQt(VY;iWlS@|50zv_mfveg2_U-TD2xKgB3TK#@AyCa+q2xT>sb*2cB71u-psW+h
zx$ft~v}%UGp_I367_jzn*0#uk!)D7|pX}&7bSqkPf?`D4Z#yBWmCAeXk>nq05;ETB
za|Qu@FbkQfiXGi+UjHD<B&h><;gX?osHm*tic?*_^)_EQF{Qk*L-wBF2|)KObmy><
zbSZYaQ^>24%^DRI<+o?gCJv-fj_35Oth@2UA##q}oVSXu>qS=YiqD@)Ynv%3AjVCY
zFp4sLP3K|Al|m|>9e<elnyz4w(_Dt8(>!Jjn`vqskIo1D;W=8F4EL!)Mt0eEdiLKi
zbHd|(nOcXB&Dl@{2=J`zSTs0gZjX3GM<X4_^2vq06)BaNJ888x!uuEZ-07khRP?ZB
zQ2?yh(~x1RGekIx0*2Xa;{HFjXGL*!ZPe!{>@>PW(km>Wdjf`D_YIimK~B63exUi&
zo{?;(t<nKP@IG)rUzD<N_Yp(dUV%+KPYR84uKMxKn>S_po0u(tuF})fEg096)ac)3
zOqy?J)5c6S2^yX(8mXB2WpDr4KaW0Mc1-oB4lHS!psn)lS6N*+dYMbW3v)$eKtlCX
zcge763c5P43AqI`h8ydw_<LWrOYXlezdB|YF$r%tU`xolIB%c%d<xLKVN9;^Z_)kc
z@y8rocRpo7@`?*V7M3E6ortfvlCQ%P#%2Hj9xkV5V7bhDFqz{I=c>WLfqLwFs~Xn!
zd&ET{gtsg8!7HZ$DdOdP##smqeZgjK0}%mczuo=OA^R|Gp{63xM(pNXnEtjMJ8tFr
z6Kfd+iQs7XJVf~xF=)l(i~kif9)&qbVo|H?V<9$Yf8ByMp{T%cBlcYizy$*p!8ASJ
zjd@RHca<7Pqxf)uN(VTc!}JBcluYD8Rhf(UH1Je~Gq|O&hj5bQF*!vf&|ZF15*f_&
z3Ui;9NdM0T0DiWUe9M#qdc>%}yy(2U`|9hNv%zKw<?{@RULr1t)vM=SocAEQH2v1C
z`=>(DLkbv1hZxZMePf~?HN+7zd<L1`$eC4L@oLE#4ph(ZykX|%9YBolSrwqm8&+37
zs>sgt%<8<wSW%hmmFMi(s=P9|X?@b7pujO?Gt9ejC#N!UaqXV2yeq{*Ywd3sF`6JU
zNwf)yeDN}RnduTi)Xx24rz+%%6doBUKd|Htov`kmxWG=ywIANd{O&_88abuWHAxK{
zt3qF;g<b8~!uS2DohQpSFi%mjUc63{60`NnEG@&fHEv}&c?DP-@cRQ#D$5YQxI`DA
z(vyH$02CSjEukvHp}M|{&K@?3DemqcpSO67!XND7D(bpOr#CilA75%$mRQ{xia=w?
zp=C@X-cAp3g~7uFI#>vBj`3Ge1<2#;gh+q9y0*QEjSeWaXc8po6w`}XJZDGQYw-{y
zypi|?<cl)yc@)}?w)`&v*Z@+#((N-}!62|%cw%Tj8Na9=a1FLubea&`WBKc%F{3p#
zsqKV&W7_^FafsPu>^yyoE+aY!!T*C<W;nhTbRC@GPQkVJ6MLeVB)szoCyJ;WkjnZZ
zemF|eNOFs2dFAufWa91X>!8ij0ODOqguXNlOYff<!5tKdC=4>yuuq@!Da*7-apYWE
zNqoEH?k%=S&+Kb%o(9hM9<@UVDjtz>0llB+taG^UpQN|TbSZCLCcFT9Y?ZnWGJBl0
zbs8P1q!*EnF=J_2_nTRn7=DgWCy+8T3MxbAT^VH}QgpAVA7(%xqHq6$c3DQ;$r*?0
zSiS55B^k=(9c+lJnVGF9^)_OGMkyZ}<BZ-FK|%^`ACFihk%t_-o$s~AquARE%_+s3
z(^I=chZ`66iti|cegY8lrN5R^3fxY7Ut-nb{Ai+3#b_5j{s6Mknj>p7UB7rzGjx(*
z4P=TcI?Id$Ir*_mH**udU?Esg$K{ZpBfSvH0J%*8I5z>;>N(YM{!7yDY^KJ7q=>IT
zmK`aVg3`-uJfrj}ek&^+YH6!opP-3r1#Cy{(S(bw{KJRqAP<oU?L@6lKO;;P5)e=N
zSCbI~N9k3#s1s)OcVCCFP-3S<$pADA_beSGIPh4OI`W;<Z~q&Z_u0<RJ#589W8al2
zx?cUO3P%@w=)1^Q2~N6RM}38C`Om&f_xbp!yV6w~a99d>HjXRlIz<O}w4}~~Hwfa>
z*QM(B@z&OPpB~dW7D3nrvXr->pn<%wg1B<v3kw!=NX<tGWKl<6xX=NJgC_^bV@Qqk
z(}_bUCM7u!4ES4EyOXxDr?_>-0|!ta^eAP4hpG2V_&OV_Cq{KaSP3LmbGT>nTVqSy
z>pn&bj!1jkRb=-BqDc}OhOgq77qruoSl9^Ppmn31lwlO%zi-%ZS@A=%SA3Tt^<6X!
zaPH<>Sx=n!jUw|ljOcn6Xj<^bjX3#c-nv6pBznFdScCUZsnA)IOlX2dLw*;Jnt^Nr
ze-aGOD1q_7@`Y45U+e3eDBlkm@@x3^poE>nYQ~jS=U!U9Si=DDEANJ2J<8u@W#^_J
zK9ZJk<Hj!T5GzW-fcm~%Iw<n}A;(%u5KnG?+{Vt+Eely=<Ye^b%8`+F0C57R0~p^q
zDf#2Knb0f{Gq9VC?(iR?sv!e)+cDPO--U(Wqo_ctY(m?=n`e_yUe^z>PEbi&lVhlD
zR#m;7W3&R9<!zMgJ*FQ0Q@}m&?l6yr`c?1$9=)lgeOtk@A3S*ALt-LIDrWn!w!0~n
zJ6Z=NpnZ*~4D8gz(h@01FHXcGD2lh)MzrS`e@*lrq30vQ&m4KhhwrU?4?rFM)ugAi
zDgH4$-W&7)VU$|aFefu6br)Ye9gl)B-`)1Ghq`b}9nzuz+OsVL`S)GBopeP?5U@6v
zbK8oOAsmG$61Z3(Y}Ovi(ow=~NUHwpveX?1yzYf>cU0N6H?`}j*mF*RnfdlY854*&
z&YHgedTQ#Jj^@jIqC)sj+g*J65iD={kCgPaS6BYJXBKIO5o<<ySaZ@Aej2N*N-cX;
zo4E~QW9*<c#RiBSzGYFfHm5Ov^Ajq{LU18lhYwH6QmaBcY#;Ak6`Jra6P`j_(bCbm
zJ20CU)=Igw;lnAL%SY0Brmyf$dgG9#OT1iv4Fj6?oPpiR77hQ}c~nCWM#vO@+YwtM
z`p!tg;YPaUre+EdL3j)?H7j3c>3w5TwPD@gf^peK3@DkD@{jN`t<g|sw+FcfD%WOc
zxE`9csz=fC70^_k^De2wb_ECT5Keygi$>i-ZvHkGQ5htT<ng3Ynl#=DB&h>rnN$XP
z3^~nfkUK-NoH>7wLUKx)4=Zf}mKgw-q$t%F30eLY@v9x3RZNngE;qF}CYUQq9%pSp
z@6V-JetuuJOl&*Y^ugMi@$|Ttu5!%;Y`^!9-eXhahk*m)pKljD{=e!bRR0!H{QvWZ
bT95Dcu5OxQcl5HtKjTMFwmds>{>J|W@-@wt

diff --git a/docs_old/cookbook/images/simtom.png b/docs_old/cookbook/images/simtom.png
deleted file mode 100644
index 5f6c926d86b89493990cff0e14036545998aabdc..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 109657
zcmeF1gLh@mw(sNY*tU%w+w8bwbj*%z8y&M_TOD_7yF2RGw)uAd?z#7zci+G8GR7Kf
z?P6-KxvFN>_Y<kCD2)V<4-W<gh9oN^@fi#Z5)KRuyaxaSN;%r0Tm}O}6t)r<SC$nQ
zCslT`H@C7e0|S$ZOwokZQXR$4(M^mOvCNl)>x3i!3?T!TAG(fa6GACM3#5iZ^VU{#
zI4gkQa#5G4hcuQxWsK@%iK3-FG48GDkOlwpka98kFq!@F?z}H-F_qo+fDDFD!%MD0
zTmdK2RVPA7f-ggo^j(U^6%z3fl4%p#g;y^!0*`<I&JfS$Lw|h@Dx#`$pK|`$;=`Xw
zN1!VN0Sq0%5k{$RGW?Mg%mYi4k{K+3c894HGr|;CBab2sT&zPmtO2w2Qy4p@N9`&#
zFbCQZ0gV1DVJr^<*d0TBDUMS^UYHlj0JB(*2p|B4$hiCURaPKko~${Ti#)>YIVOh0
z<;QzT2C!&!%M{6IP;Xa~N`@3|&#~k^_vmqhN?3{SupdUF$;MHu!)A)rzQvhFa-<G_
zptCRL5I3TU?45LT^o&f3Zh5YbGF0&EL`t3EQ~>b+vw)&9)W8sGIN+v$BE?6lgfs*R
zs7Gd*kQa~*s0SZFcuL%aDxRdiD)K01X%teOFs6UHA*^4momdpNvO`xb|KQ7<);Kmf
z9$b#3LscbhrpKg|7Oz6CZ=mER6SHFf$&-mCXi1SeG``&u#FN8<h@^IC5}L6c?@{b)
z)n)3?wNzpQ&aV^sfMZG%Mt2?D<iLCVxTROe;6ui4r5w(Jhr!E(Lo7_I1$3{<WRd@*
zeC&gY1vSJlw2i5vUo}VEvUw0X=7A8gMi~k>R`1~it``sQE1;@|npqZs!7{ZKGEBeS
zn-(b310foKTvTTsAWI4hSn(q8+R84VfpQ4IH3b)P_{<he@p)HJ%tHZea8oJbQG|_y
zS$TaNHZ=Z&Z-Bl=kyLeEmzE>}AxA6=g<OQVBW+9?F1&*(m?j?WwVxkWgE@v9MML|e
z`v7+N0uh5P;gb~KVL}biQWlXK2CYAWOu~B=@X+sjDdT|@@EglwQ$P6T(N4+ZlO>G+
z^(v_g>bX4T1l9+p8RlWYGufD)>1#T|D$LiPJAth%YT;w8#CA|w4%~;~kuT!Y=kaEA
zJE>-&?@9L?=9m5R$}0(9MrgHj2}PP;Jd1t?|9pv$Vq6s7mqxQ5Qtr|g7ZiT6c#HQX
zwxRHA9>6w+t#e+YJB4t=v<7oIg<6PcG9EVduyhqnsI3vTzq<Bt%8%V7d56iZVoX9$
z(@7&vV~Og5@Jl!)Leff-P#p}aI)&iF@I!laX^uY;$T~mXK86W6eJSwp?Tk$F=SJ;Z
zxLEBF2D=mlva&kt^5ghH1o-ASypX(4Edw`8#jyfz!rMMO<_Gng;XwGvS@vX`1LOm#
z>Vj_^p^1B7p#lt1k&Q*jWuP$(Z<igiqkAz~i6+3Q@(>vzD@CD~0L=kJhTKai_&q8P
z@LAAlJy;HqtHG6}2pwR)n?#Q>qyV^la*hF1pm-IU^Z=F-sh<ohdgK5_xeQ-oC;*L)
zjVL2LnT#PG-jw}w@ToX5MQJ>O48IesXUHvSZGsYQSVcZx8Qj<4SP?v0jBg?svsaeb
ze0WS!2ZeGoO_rQL5vrw?^3!GrEMcD^&W6ECqK=IA(=mp6w6qXY!_170ZKyS(QcQkp
z5iLOF-f&hyo(u|YOJs%3JJPfU^<PqEMJo+iY)4<g`@rhQwQMe4N_&vJe-oj?#UO-R
zhvG6qB#UMZRTiVAv7sO&&kFP|kG_nRp*%*jjm&XiDM5FRn2(5#Jc;y<pou&iP&5{f
zlwhOeP4tlkQX-cJsEO%FuSoKV{`_=I?yIb#4penihf|SNMXQ)lr_)?6chcLd7-LVF
zHdXk+xtvKTzMoK<n3mX*m^qR*`ek<-lPQX6FfCDLG=Cz$s`!`6l=8XidRaw<PMKHv
zP&vGsMP-fpTBVo%UBa<6r!6NeXX*+MXZZ?x)7e*=-{)nW`Jc<Vet~_(ZwmMsq9ZP-
z;4AZ%r>N?#VNrSW%c5MfVzJUd-M?IPp>DzOtNaY@0_`GCo}+f5W>S?^;pp71<<zgL
zUjkL0at#L2C7hobq!<jW>lYNO5p>p8C~XPO3~h7nea@IxN}Kqta3>Ro3LLF&EOo6c
zEWhwc5a|*%E5<5jOxa~~Wea_={ldLwx<<bCY0cy(kym=oH)35DCHwMqzs^jr+*^g)
zjAx%W$!FJR&ASZ(M0Nn{CP6a(DX`V(ax~r!*CyS%b+xYdG7!yf*AUZ!W6&~q7O!yK
zh_Bx#%rQI#`H%2v4e;x!A)&p%q(`)4rDJzwxolbFh-vnyRg%7{;5^to;k<guu!8r5
zrMaIuM1#6H%|y<;>yf+{u`g3+=(>fS4Y4^h=Z4L?`TKI;0{N(M%^-0!Y$~Q)m|4jl
z{kZwK5CeN9ZRL?hlZJcs=mO%2%?aniLDj~erIn&3#yYFIqh`)Jvp?8>7FO0<3Rj-~
zC^T0#x7#T-S}Yx`)V6eQ2k^D>oARsi>A6V|2@*obo)G;B#|ZOA!a=Sj_8XU-j6701
zR6DsH(Vpg=^Q>_Wz8Oc;pqP_lByqGG;Kp%J6G*FOevtYahe*QfZaP1*Sc6=nxA<xC
zq_C_|OllJUH_vfOW+B0bZSWsk_wU}xFAxvr54FTO#Kv)aal}dt(=OBW)4GU6Qj8|R
zoerE{h8gSBMU~pPb>emC^_R|%Ysc%+yXB*K!4)?S|J#Mx>@Mss!i~u;B0raImu>|=
zqWArG!jG>X!{Djl9RarB)9LpUg!|0<<05Gyj+^;C4~+H;hzp~njI<nO9A!*e{5*86
z<c~%Xk}?`HwlW4~-iM7dNk$=c5p{b;Elbok%<W-oIOljm6dhDzOypz*<lBx#;zc|^
z=-|J>$LZj?3a{|#;I5%U8ZQAEOo#_oh7@B3(WPUezg5O~;|sI-dA+tl>GTbRWW!yc
zLt>;+ugm?AOQhzcv8IfZtC97~MNj@gUTiWE0c0Q2`^5PPtB9*4VJ>mLx%e&)iMZX%
zW|8N?k!r^-wuQ)m<Gl~65Y7`LR^eCrNQMV#c6?4^7yBCmO_=KB(X^q3n%S+R_A2Kp
zP6yaJy*|oDT6fR~<2<*sq`7;l{lZvPj?KsE^{GC%oHCh53Px%J@W9tj{``J5up_xc
z@QMEE=qWsOrBBajoN``PGHFFJwg6pa*z{twlI9_jV&D&DglR`ny$J04NXw_QqHY-y
ztY<6@Y`^qGfjyA~jfA=~*NXmMlB&5_;g}y>NV!AV3j%9C&sj%t=BySrIgt|+>B|D+
zP7ANvUmC<MF4Dy_lK2lE+L)a+Cw%!?%=G7dlAan4+~3ik<!mFpB3pMCAabDzaJCVk
z5&mF*(`mGuZTYN!G{~0r<N60nbXzpbJKyzUQ#Uf>sNP)jFO#D%vjMX{Gw!q(mW_Hb
zYsssqsT3846U|x8lp3#gukB-mJ-a=EMdbyVMVG2?dVgA{zOManHK6~&V5bJ7abC_=
zqoa1`Cusln8ws4?nCrU@<DW;XB*Xc$$^6WdO52L@7St9@JI|k-o+oV^xZS^P*jF@~
z&raO7?qi9r_;tPtSbufHztGc(Uif6~?e^`DtC{=E`4C^}&q?pw`{Xld?_%%V3%Lz<
zXa9lseaIL{SJ-M44Wa@f`fBOp(uD_3Q*IZI60T+vWZ~7--FcV8h?|5n>B%?^;(5X5
zTb{kg*BwZjD5cw6%<NOa6xWIdPve`;c@sqg`$%gmehkO`*X8(z(rMsyHqV9k`yuoN
zLO2T>Gnd|VtJ=o#y4Rum!v;!|XX9^!`;JwQ!IMv_FNgO#ns2o^Ke1iTPF?Ts<hTp$
z8#m;CdcG{Q>JqhhwzOaDu6eq)T%I3zH(X$xrwPsYD8J7<8r_lH`=1_N9`fAN9;%$|
zgbY9MwPmy9$mIAOeg8;)ZhXyHZuoAucwQ(-Ca9Sc=evF<cDwCtKbouKuhH4)?Ri-n
z(5>?_bnDuQ-i`0k`S@)s^UQv2o$F!vUf3t?<NW4j4ke7l&BxS7;DK&$zF}ImE557h
z?dwO=k?ugt3*-|dBzbH)#Rpc8p(HyyxJM@u*jIWm4iBNi81<}=U1*go?LELNO%)Ht
zSTrn{>*a@F4bJ=8Dum&MhR18rGyV=(L@`ACA@qz&Vc7Aewt`<xG8VZ2tWDJW_k>*V
zcLkL!UjK-f1MCsCwVvs-Zm{!5pJ09;%QeuBLc>f;)?7gWj2`q100R%V0)qlQfrBo5
z&;{CjB!+>(g6`;`OCk^AzjGns@*w~F8N3IS2PUd2E-MSVtC~2Onb|p8+PkFnq7{Ie
zTCh^na?w(d=QFXlWi~RkH#TGTuyy!b1x&z$5A<kj=3+$ZVQXXO%;zCU{!a-$(DUD9
z7IM;minv${l4~g_lZx9rnUQiavof=i3&E3;k_tGPn)7{@kos43&@Vx9OBWXhJ{A^t
zcXwuYc4m7g3l<<RFE0x#8w(p76Q~4}v!|Vlkq48VGsS;3^55-9m^qs`Svj~^+1ru+
zZP&=y-ql5rocwP`|MT}>|EHOU)xUePbN<(`Km%m?o5KQRW@Y)Gwn0?|{-*LNTX~q-
zXiHexg3JTzLx_WeN8q3G|4+`pd;E`@TK}%e&CUIvHUA^$f7euZHggiUw*__SBJ}S!
z^RLSPnfb4Z0xW+={vVe3FEjs>3bL~hya3DpyfYzqinCSF-c$rkRzg(G1N<}_HVaeo
zewc(;R!S5~4+;$mS`tkS)0xj%1JiloCgw0>KKp7S2d&{lbwXRD2Z}t15)BHRN@mIm
z?xxm$RM#srr?sJuy`v>}Ipe(M@#MTB?O44nodU=Pg$OR*gUb3JH$x(bJtVmQmH4+v
z5FkAsRifXI03s}a%n*S4?^}AjAxhx?odo)YDc%zjV(hT=|K@<oPyeqb{(t%ZFv$P6
z&tGHwFaHr~M%8Ninlh2iuW4Mb)g0C6^e@-tLWiKHw4uza#{b7@`v8nubs{h*#Ekvn
zC~0Qnsi7>TF0XrO24mK>oz&KgH41c{)MFgE`$~K57SlO%d7_Z{HKu2}9Uk?=Mm+)G
zY-VH0YEdm5ASYK2LroX<?_w>f)@z%?grsU^iAkx{ti8`Q-7$x&kfpPpLV&X@OJ&q1
z7Y{>Bcz=5tzdf3VK2z6T)SLXyDW~tT1^skXn)@0YKezTn0@PqU4#4MjR(h^L5}{nT
zrN-rGPSS17RkIwlxSTdE>R*qubQi5kQzt?q^N9R(zhD;rcy%7z(P!N#Qz`sBmFCpH
zK=ljG?)CBNhvjTR>+(g{i(R>TW$Lg|_uG9^P!PnjNt~eR{zQhP=g7w}i9aUM9~6*H
zXz?IoATg-rFMU*vjJ9Bc1T=aacZNpiisdMtl4X;qM=!cR&O?Ki8!Y5Hz3z55gHciw
zD5ReSsI0%duCkj?#Bcb$OfFWMAeO5%vM!L8?Lv^?o{~KjXx8dGjgW;uUfB;5xhI3V
z)>!xd@ZHGev|fNp0)+>>43wl$J%;Ybzhu8IR2|VIzddXiG%uSIpj=kgb{&(<DKIf!
z)?K5Akt|u{`f5xGKIpaWf42^S^;2zeE);asNFb}rpm6wO+cKIUO?*Vt=KHXwImLTG
z$a~(pQ{;KGf4%0uA)Co%TV&AfNB0d5i(V~$UP<`N{jvp#D8{rzE_}j8Rl`)G-P+I5
zT>p2v@iZ1CD%}+;7rjDcTqHqc?Az^VmN5nHHOjk_#k6SVpEAttr)(P}n8F6WcPS+N
zvM7q6*O&w}8InlFV4VLu@rgJ=DWlqetexYJ&*xG$k;z@FMg3U0P^~+D*0RBDG8~(N
z#h`)tCA@aouKjE`FBtiHL3wgq3j$JUrR<@zECBuWa3()k^9QRCYqK)SC#b25x_;#A
z8F7>+t{(S2c;A~zZoin9=abrE#caMG4qIZP83U`AOD1uiLXN5ccr0h=9z%Xl(sj=R
zBHpbKjMpQLY8_7`wY{+vv)jX2Hk&0@-kS;bPV=)N0{dGr3D%FHX9a$%;cyZ^T966a
za^tNiLWWS0$XD1XK%TKAfGYB5Is^g8b=Ekq_+t=T7qA1NQ}^+DAe;m2x*lUNo5>S>
z*I_fA>0>`vuhd}Do-<y&XulA`0~&{@Rj8FOJ<aim{K1Jc@_0FG@l62$HhZC-TDsp9
z6osBl%XOM6hKt?KRwTd6v)ZgQmbVBbOX*h2iyJzN=6z{*%Mt$0=UVYFv?(aA#iCi0
z)w0riD&C6F*TC`hmkg-!?b;n|KyPy4;i}U((;M)j&F?kx^JA$A+?n-UQKU`NZ;iP^
zX_TkCMFX>zHMgW@+drS4uJ_(bF)LV7$a~S`AjI%A)b~R{AA;U__lKY7^(c+t`zlHD
zHbIipXBiO+*Q4SzuPjCWGmfVN;SV|_0<K#OrbLRKb+41^>&t$W@mXmSruK`j_B`Hs
z*fg8~rTAZ_gTYvo*JZa-#5LhJiD9oktx{inoU6VIPVB$dXw?^VhwWd0T<~t6jX1&~
zLx<RQmM!>fyJexY+e*5lajV142DEMxkOX?4w+(BKwT3K2kH*v^ND(=+V-W9P$$!9-
z_>h)nIYn6{J_#L{6?S;feQ{sg8)mCYpv5EhxfB)VH5OE-g4%zFCHCmi)x~peUUSnG
zzE|emp>n0rGyHC09E3M)t9p{ZDox_6He(000<LdM<qf1iF6i2Onjr<jm?*v6pY5$!
z*WN9BQAiu(u1Q0EgGH%=Mnyzik<p8p68`Y9H`{v6c3-Ef?YNE_hC_fdA-;2t1>S+|
zm^}22U>x+3kS~65&$j8jTQI!|`5olO&b{h5Ow=_2i6St=E^PT3ZBChw4mf1oYdH0^
z@$sIfDj$B|+l3V90bo5R^n2WTUmp5AQ*R>uLF|;{ZV-6|s0B3M(M#N({aG{^x{frA
zYd1~zU`g@++(k?O?V7)1e9YvEA1@(}@BQt<f86_`^I#Zx2)>HKBPHZXXl(wT5}mwJ
zFuD4uQ7~%<(osQnI!STfiRoBiZytHh>M2ImBFizXC5z2`-fifxG{<X9R}OUmw-Aju
zXwUks`2LJ9-g%dr<y?6p<EQb|Eur&-`$XXROK%GXrm<b(jnyh7hO@bc|J!N9bS#nA
zVLouU4yTwVX?|sJRRYcV-4)2Kf|atv2nIaB*gtW^aGy{}2DaqY&?3RnDCH13o4=%c
zs~FokMYlx??CZf8pBG2*+G<Ta5x%&!!g1_RIMw^V?vMc83r(hyiJi|?fHX6Y_fMbJ
zgjc*klfm#$*OUG6la(gxi8ji?5`|0+H5T8F$L(knMUg08f6)zQ?)pJX3eUCe{B&=y
zCYcv5D8xXvAQ4f~KoXSm&WClRA?ELg*sp_GU|aK->+Bk=`oQ%z*(Lv*%xMfIfvtId
z>U9!+^Of$0SSyxK!?-)IA<?nr8^BTEu+IxH+9~T#0EuoRaVZnyrwgd`f(-+oE6Lf`
z(N;-)Zll4hFkp%8W|nTqm*n#c|BrU1gY+ldvw2584}M(D(S!6jtz04BDCUIJtEIM-
zp{&#&Wkia3S3AeRVL}XNj8c-1IN|rlf)mA=2=!Y1HIjLuYwOMx;^P8Ep$UpKJ*UWz
zNp6$6#D@LH8%6%3VZVKR3USh0R1eT+`^tWfnFbuT9L~y&z98<vxCoCWnYzY%*$O9l
zeHrpT;#hhAsj4xM^&mA-4SP4-;Xm~WkN;`_!(=LS?uP9H<VJ!h7(V_UG~Rc|SkC7h
zz#%G{yn(C}(+M~lU<-CZ#71PF&M;)DQJf0b*>1MGom9Pj_gizs!H#eEg)@A>5-6TH
z9~O9|EES00WssQAwvQfyId)t?o1UJ96}og*<lP)LTsdDz7|QxVc@LV4AFt$}B!`?^
zQb~t!N7IWAvVSk#_-=ApOl9HjMBb+X=RLm<WR0iUOREX<mioHF$ENFks8n@cU7Owr
zd{&%FNNRo+cO1mh^a$grfaHs8lXL(cH0;wTWxsZPynD0A#K*z@wyN^+ocdtc=aA~+
zq3L~co(dre?p=f%%xE*2D%vA~pvItid$Lw!Kge*IyUM$%D+}o1rRK5#M$8i9zH@K*
z%pW*YyMENEl(RoQ!O6|Dds=nAx+Qkmb3=%Xr!u{9o+YujU=?>>_8|x|^8ufcmxcc>
zAw>XEu)~+rRVoujh89t84~yFFcdy*3Hr3}*SsH_DHf+I<*Bc<UG_}hCETOXwdv4BX
zLgZ->0`ui?%c_&ygf@;b^R>!s>-K{9fU;lwbViFW-;iJk%#f^Y?u}$@=lk;surxc4
z$Z-JeT_CBD4Rz{`bidKv526@(t^kuvcp|h!wL~BxKA>PwQiQ&OU>`!1LI!|a@tAKQ
zpDzC^X786$XH2A$@SDpi=%wA04Czc2I?Ot*RlCzx3~Q?yv&u|uET&28;F43@Vk*r1
z)-*Vqrx0lE!Ij*D`eWXW6sRW!<`-f?q96Y71y9A|_w7Y6LeysMjXwE6VXFu&X<RHu
zM{b4bW{00z&*qXgnBc`gJ2VPpL)Foy{bmBse&z$NA8pPLRXmF!cJ&;d3$zWPA>ut>
z<@L`c*V3ZJ!{wzp$5?<E-}nY{>?T+trJHz?J?2*&2()2ZpS7Uv^oE*8u8JNg;o&8K
zy-ZtCv2RS=-?&fZY3lbX0-2tLfNBogD`CC^HPR=fgVdv8%QIabTPvd4mXy}iT?k&%
z4y){DbfN)4<%G$7e2GR1_lBqm@hs=h#0CWM-lV>K*t>>_?kKITaLbv$B6<4A#UI!r
z<~N@tgVJ|c!HvTF&)W`O>0RX41hec3nIY3JR-yJzda1xjOlWw$)4;2<_?|@4yRYi|
zv5u0IgwQ5DL!<_>hCRm&9pk-aqMpUefUC%tC@w6iEW-*p>jIK|f8!2X0zjEAk_ooT
zw}K(&iIMbkGHkGGqlsXdi$>%1m{@<-bDmyT0H0rd9ZsL?s8xT6rRY@9@pRkh_zNoR
z7f6TMUwkql8$eMk+z$s~hcVpTG~fsj)9=+N#`ucUPVf=T`IP!gu3ciZhnRsXk-q~1
zX%MW+chc2B2do19A?}Fg%~ZHqi|0Yyabx?{7*gy<73>2%cgOJ{Tmraq{8I{B3q99~
z$l>rjoXye4s1HtO7>}>Zt)I70s?PPg3|}k&yWkMV--or>9k(UOKA{Pqo?vQmk2avx
zkKnme=v{b}lCG=z?BX>9F|&5$O4z@GmW}~BCSM&ra6EA>??FDwg_P^}ZTo5Op;k#$
zm~}x9yx4LEG{Nf;3L&KP_!#ky{zjT5>}&R+dYDT0i}jq+omJ54%NV168+p|W2eDm+
z<~ND!gD*<yARMv@na$a=*bD5Ae3cl{TxG)3BzqK|a+deAoEChzJuhOE3Cr3ZSOkxQ
zML9}l9n_21RLXXh$TIFbphB_BYb~~O(WKCHpMrw}8~W--`)c@{bn*^P9fE?B2TK#-
zD!c7>B*sULuJLe#e#OczD$W?Q-#Z5f*{xsNdxL@3)g?bBZ>hL)7wt3BU(`{3gD|wg
z#Z&K$ME=1(EwJOy&U<~G;CGPij(umcPiAtho*y(YHL*v+=5uLDf`V>K@kPLBMAP^K
zw3Ra~BoAO=8K5llD^o#o_|EFDgL87cG`Iw=I*u#uAz)F!f*;&SIF|?`*4u*{B+{1-
z4NP^cGD#E0dmdg2^=CARml7g_!{ACH--&wOMkeH$i4&dj+M<Yxk3A_;P?zcXz?KSu
z!`lZlLzIv}DJ8ny1A0~}hjw$k<$av?g55yWr-BWUK%+sR%R>&MCR`d~CTm%2F`^L5
z`Kji_UdXSnLRw^8?HsOqUz`oQ9rr0Mji8QjLv6Z}V!xA9@>FIBn**6HV5fzmKI0lL
zP(v+Ju+|*4%ThzR9<1zOAp#Jp5Q0#9wiHmMP)SKGvM)a&%v}jJAj2DXh>?l<^1VMH
z<om}+qY5N2A0r_E^en0|Ay5L*vb<aEDQxpuE+eKIf=&H=VRjgNa0EyYeGjIxecSOY
zaR2a7lR%_6lU*kq`j|hUY(;Ug`;n50#@NP`16WYSpmzXg#f+H1%P77Q^v?@zNiR(R
zfZ$Z<Dgs>0ph`y}OC#pTUQv8r5)(Hs@lk*OF2th>ilUl6vCKAv_>{n|$-;GyL&L4n
z0aZ}Bg(Jz;WOm!CoR1dj5^f}08H-jaG3+$-iS*ZS4;2p{&%ERW5%kcyDAIhGP$*);
zMjQD?pNMDhK!t*|ahf0gpbc={EGj8I;DUYt3ism86g@CK-F162y*?N&{YnX^S@Ii=
z<sBJz2VLQG>p;lBBRb<e4dPp&JT()MW-|DuKn&M}a6W9tk~Yso$P_P@z@er?3_KQz
z=Lipy`1S&xhtW>Q&D70N7Jqzhgsc4TiAKqn{2Q<{Xo3(x%?*+wYK+K?u~;r@p7J`Y
z(G3{@CS)@>G(B_-wUtZ@Sgu-=Ko)NguvCU<e7IE9nkR*_8Y0`i6ZeqLE0T;iXB49=
z-$_u~e9!L)rQ64suZBVH3F_eiFRTvo4J3%8&3%Y}$Z*G{CK8(+ZO@ij24hFc_q13I
zv5NbobW3q8>M1YS@q!9Zumb6fGE4j{P9ih>a=sAyY@LksQ4Em+uOLxV^rpo|HWf_v
zN-@wX)g9p?`=({ogW-oxSeFVjj78M2`J#+k7+^$*G7Z~x`R$VBAOx!C7eP!p?Yil}
zSDy&jj8R^7W41Y2*f4@K%orXL4pj*Vhv?l7?I@g{YQgx*nxA9BTJrR*;D!)8TXNxz
ziqtwtz>=rnv(j$SeK8VKg>fLJeBt8hAJAbR7ZU_6A!(z$iV4mb#KbkK3Ce_x14Q7}
zZ}(Li_%{ls6_0eCA$n?oDrCz6)|dX*(gKjc#h~*|vB$E7RF5*_R5>2i7jDe-G=D9|
zSAO4e-GL-vW@)RS2Q*CE&Get}s{_cu$P=h>pcuF_)Ibr*g!G3D!2oti*i2>2+5nBa
zHNh(nieLHmK0||lm`E&HKp(~+B{ny%Udw&YjbaQ%44KU(G@2i7ixpri<0{gfLB2zi
zPZ7G>zN6MB=oR_G_fqiJMiIgd|CU!BBI~_KD~6GvWVF&csg-x5jcsPLbjnyv?bf}e
zPM1~OPdju`WKoz(_(vnOfF6J-cqJlkAOkL&q9d`U;t;ifbe)~yV9!{A?E)~yjZ}@5
zx&jsJbbjbL+gS{v=h9F_rol1-Q_p|yZAWUm6y-><M1YWw7v?5j^G#n?>1U8@5;8t-
zqhkp6uQ*+(LmEjtLKGzK4}*{Av^3*LZ)C@q64ntYC>q*a+FO(xo{;3&0ix5SPi>Fz
zG$!a`gDd%<v#?||C|q6EesR{kkW+-N19_e?%l^$mM5Dx+NTw7!W#W7RBNsFohkPSp
zJZc6^qE_TGGt!jn?P8-J!RcU*7ovYkcFGhhuC8IG3o}eQ1|({sRC}eqqH3tmky5g%
zh}i`eee9Dg<#hx)U<~$E-WDLZk7>oth^~`16j%jqB)g_MO9mM7j6|Pc_)0{VCbcLT
zejo-4hy_>>faCQ|B9~yT!%1P0f#G}wRFb_vqNYkP!f1idQ!!$qUa;+iT}!i0k6rZW
zs`&r73Tc9}T+wEUYEP^&1H@L3u9Px=-@1=i5|pz<dBtIliID?<)3a__K^Tnj^zuOp
zDz)Q!_e5Sn7jyQ3dn)lAQ3@;Wc!iGtq)+Y5yaNEtwMY~1gta$5c?r^lLQw*V<*Df>
z)?|uVcLglLCPm_4A_;<@@V5d8ip*JX(E3X$1PKTn`;%o#(aFpyi4CdY><CGY_d<>P
zwgQcvcBntQP3H){xu*M2<BlyxA_=oo^G!(#b_gIsEAr&C$h#fS%MnvE0@w*{Ww@5D
z4F364#*iUmm<Dz=6)>Y`XGDNxbspa$F%)q3tzxf|GzX>lIz<rTHdyNW(LS+iHk9Fh
zBft<iqlE+d)oUP-8$#B=F_R~)y8W?z*W>#zlv1u?Z6lx9<d0d_$KfaEAp6C-ABZqe
zAa?)BK}Gac3qA=9sL#Icj5Obs8o0rZuR{H3>MPA-G65qTAxpg9qxVgK@SY_Ojpwwk
z-{(R(M2vNcmd=A@Qwg`yp=p6LlXa{^Q&b5y7Z~b)EcR^>g80*y1dkRKc+xy&?pH-N
z>Wqi)hD@l|bb6Fz-TERa*mNAFGP42dU@dl!tR9?t>B{>(5B1bB_dOp#H}OWCAa|22
zkr~j+d>H^li0MHk3v+}P1^6WZF~sUceJeKey0viAQR)n-2^xIJiXSgn!Y4mh<#9y8
z7*4EluQ~*`;2P#w%mw93MVV{|dd#F-szVOINLBF+Pa{}y(!#;~iR{MxO!jUpk5A~_
z5PyHkUa0T0we(C74UK}R^V9F``3W|pnYG@rVRlutm9#!6CtMPT9QK5g2?7CwH#)&F
zP0ZM=o=psig-)nJi&b6*KyCpF=vPGpM#G@nVu&Pe1FbYC2~#0ozo1dh@Hke%!+GIe
zbS~b=Q^y4Ul)>A_MW9MTF~PxL>x11vtx4|n>Ak+|D5ywk^ixHF0GQC~@Pj<XMK8{9
zHMq#czExg9oCNQoM<N5I2AK0>hs`7}cB@LYrx>H#lCdLZQp=#uv}Z|~+nVZ8#FDSR
z2(uNAP(Zae(J>R49>E}gzB?F{iF6WepZvxSHAV{&!!P=|1lLw*U<sTB!l+n+XGF~^
zGMZe49G2(9=aDnr7z&k~SQNh;v5N+c4iZUx5?929atJ2fIeB1I=B9tg2r1w~><A1y
z2InU;T(z(XT?fqp8cbb4L9%()W^e&yjJ^xD{1m%eo?d{vM0`FX6zPILYe^Zal&%WQ
zkyhkfsKRd^SbI6V%Ws)cpz|^?c{|EP=2@0H+QUABfDqYXgBaNONc56+-4?r?dKw8V
zA;a_+yvOD2IN78o=&vi=9nn<G%q7ha3T6kNsqAQzus>59-NfU1(x%jjp$eshM@FJC
zkp)3x5SUX@TKeqY25xT6!eNqu(P4r|T<ZCdQhGc`kAtBg7Cqf%ItyI#WgJCMdj4zo
zaM{(D*N#LuIJk1}cl)@;o_{sPM%**A4Ldym2~^n@8{jE53B3<<&v6V29qo8&Ju*Bl
zv7u(8rahMM=z&J}*A9WECTlb=l608$QjTGD1Ff%q6sCEG`QLtx@{I6ASwmouzrO;|
zLU1L5OV3P)l{u1QJn2m>|J)H+!<rD7`wj`?q8#Uc`PIKGuTQNpWVSMCI)1e~nk3jS
zX3&e&3;kitT8IQ-F_hI=`9>N{70%^ENIjL07NTWPlH}9T9Dehso-Do+(9_UkO$uO@
zx|%xlmdG^>P=^bENPS3t5Z|EiJ;3XKyqt1S*dQ80TeqIOs`>E_PDXq}XM!B!3SkB&
z26<sCAV<F^662XHCwQp@NjM$8jh%zQEnZC!DCKW0)YT-|2)9poLgBfELiX&RdL!R^
z7k-UPP=-K^5z6!lzRvKKx;o5}nk7HIH$vdoH&=|Uh_&^U9?TP(H;r|LVc#(|7CeP+
zx~_5FdB-a(;>-dpyoJbUG*YKR2bKZ`*2`5!?DqV#aNh8lRmTaWp*vUx2o~r$k|Z=U
z2XWWWR46563xN34jSdb0+~ha}^Z}?iE+nh;AGAquNo?iOTw7it%TX5)G+zn?ceVyz
zB}hgJF*t|i%R&|l6n>ST8TaiN70S!Iqz{f%qL7`Du}pv+SjKg=>al}F%fKIjU>CeG
zHBu9s>A*m^AosM<ge%BxUyH1*?1OA%#R<J(&IHFOS!e~^F@zynAA{Kl;1elgk4zHw
zAdH4OwUZt-9tAFUc^ub~mJGiOA_d7#6x-x(ub60OjBjH?vvYRyb#)SvE34met4wBa
zNMC4u!aLB<D7N(D*R(Vq#I-Wb86@|R*w`<}SgwL^MSdaVLuxg%$j|VOIj_m$xxr>9
zu2ZVW5Xjc(S~u~La-p>|?|2ZrAz&Be7YspM(*m#3%F&bjQoAhcB<q!=-4fhe!!81w
zf^!crTe<G3j?6sBmqKh|J*6rr7oFGqgeSPTVewZIYss{6^-A;C=U-o}C~U<Nz%O7n
zr1*w#=Nxl{XoaOeMk{C2XmPrtJ6@6D@mJi2n1}E^fJ1m=MM{kc9foB)C{}2(k0rUu
zvM*0WyT5zV>shtzaD%_OX3dL1kOFz3g$2M?WjdE1C)ntW!^BJm`$fEnH6j{!w+J_U
zf(AUt;Pp|2XtNeMblh7J#|jaJc7MNq*Nh9k`|33XeG@V-(Uj4lauDo!)&dGyY`ma6
zG+ha0IQ*dI2M7#!rUxYS{2QH7lQ0jA-pL9Gjm!@B87t-cw(80C1tX_aVFq|4zg&&Q
z;2-!5xqi&`lIxYGiu3oIc;m91$*Flq@P+OLk9=-v<HXkTb8H8s<=-v0{0-O-UqrhL
z*3mkr_|fk45APiGr^L4ezUvPoU5Cy;#^n2dXQ?3Sy&}{9>2XjV_S(zV=L^S0m998V
z!@%z$sP9k5U16^?JoG%}Nk)dIyb7R{D&hi@)<O6_i?P>a{g2YUiw_awsim?^hKWNq
zB;06a*~hv;fxB55N1&waoMQdPSbe|kBtpqw)!C{~>Of|Rhxb<+uj3_;UlbF&3}f>w
zA<u@nS+WNme6T_}A%W|GJDBYTLxWH9jiFVQJwDv1a0lp}=vuYlRB0Uqw9^DZAdoAh
z)7$>4_W65OQVFo(5);$z1YcnOr-dKiz*1Fx%<Cz89*yUQK8QE8R8uLX2*?mtF>43P
z&niH+$Ax%`4uFGu^0u$G_-6?J7VAZ%0K29DwKKYX7qRc?I9U<cQeh&@WFoD4eWgdp
z+ikzxtgVj$dB17U@kk0Z@`}0sK?F$Ep|n&b8cBul;efw8Qm0U9;UIXeWJw!I2rgNV
zjM2kp$zxJL`zHoK4=wp7de}G7k1Xe@gug;kbDfZ;onesM_Wlh>Pr*8K4Kg$|{Ywvs
zaG{c-f_r<Q&S^csj!Ly@gpr|!iYUUtX(8AAJf`n9ng)pwNA9Z<BCKpsPFUdNvjT6o
z^Vffcj6fXGK-3tiws0_l43Gxst1n{Mz)ZDmh#^kEM7nXs1(>_NxBZ7uAZK;h6zPU+
z(;<IC)}e^(8KgE`Yj<B{1ZA5d1!ke4qoO*DllqVl6r$x!fj0CwP)jypUz+L@I}e*(
zFJ=p+yIhzqUIO#nL5C@ib%idemOK@EJDHO%G5@H~BiImH^mqN#;mV6OU-tfLZv0Ue
zmBb9gs3Uk}TLTzMx95@|^|X48Q=HTut@UX?b8nO;_m*F%z>|yq55RD?V=q<Lhho_*
zUG_POq2{ukFzqZ=_XN~7DG92_Mdw3D;x<UTyIkYlPdC{GonJDqNb-<8Tly}!0Cq#-
z1KhBjS2lZtCxhyaa<D2rDadBpoGs{RY2!a$x<CHNaO!YokUD_$Q140<&)Uv=9Tw6$
zqr!xLJcNKC@F2%q@%^sW|7jBQ>Zgb@NSW*(vJ3m220GsgiLaaF`ZF?!W5CG0=AxR@
zYri&%1GVsT@;S9epP6A1^HWoldXd!NS3LJKkVJNaF#mWC5|oGbw1(zH4abt{oQCi$
zMiveH&$w5;_EI!E%DY{@Hhf0<OPk0b1s2Ptjb&Km&|P%g#BO{c65VRuiIrn&C}x8q
zN2R?HgRn;X&S||*(Yr*x7r}2?%Eb%8o;sPZp7cCwkKmuCZoNt^k3}z+`CPplQR4k+
z=NO6Y>jo%>rqML;ZPN2PD#qJukuMJJDUgir0OR($wOsc9c<XFNrV9J7v714IGiIU)
zTn5-rfvkJDD3g-Xtj+I)5Xjc81YtzZIbmfdYF#~P0ktr&FqXA<vlmy~uop-|oR$(A
zZFaW8DfuOb!@4o%yP(wj@|BpCR<-`@lbU~BIK*_9%CZ<36A?t;)iB9LVO2`?f^Ao|
z5evLqV)I%N_l8A}R~K>M%SE@rcP?9c|BsiAPP4$KEZdbv?c5$eJP0~+WuZZrOh!C{
z;tXh`q+WZhVbRS(;jV<}7d;B47!MU=W`zkLov2Q`Ta$bSr&QPTP3h)Qvj5G*%_%_&
z5fq^N_5J0-Ti9UYu}u;9u)9spZNEpZ-Rvbf<!15z`sCNSl1L@z)Qqy=AWr%RfGQG*
zasfN`%}{IKX3=bKj|IYPJ+Vh|tXW0pT_~V!r9sK>WzLF;VWeyF=h$irg4tM~u<1(2
z)eo<`=f}re+sCZjy@CCg(lrIQr7UiH{go$1@gve6A?#Bl6_8qNge1o0CTNz_1k!o5
z!jOrMDO&Q8EJ#A{&;EEnfMM6BVNdD^DS0WKvtaBiCH}Vla7wV&?QSAmC~DxS(C==n
znEcw-U_O~huU_#O`u*+YAq^z7%S}`xkyBDK8=9++(64G}A(LN@P{fT4L}I-01fS4H
zXErF%;$mYnC&2HG$yJZ(%BUaMStknvl0g%^e_nIWE(XN7CR0N5H}|BiscOriKDxOo
zeS~F;c1WUmPHl5WhnU3*v-Q3@WA*c|mN!R!DUB?r!C+`s)gZ;1ib=7i{hY5mave>Y
zknH8OA#Ot??hOl-P^-=l#6XO|(PV8xW8iMt%;;z{+-NW+AITkz32q4gn3A+}Rhh-G
zh)v8jk4yDG_v*4q35?cL5hd7?HZsZ=c)jfZi&DtEJ>Qv~ueSa0xH86LH;?(Nh)RDS
z!?q*H=hp}&Q~nANxL-24p5)$m^wo<%Z+kg!w<xZGo*Hc4h#}dNwg4J*7$+pVi;1m$
zbuHS_u-&_){*pDLOvZ4hMV5&Tj)Bg_^seC@OFr}C9r?#XwOso40@1+gLx`~S$S!ZU
zOl8>F-s2i`qS%4m?ORpC51-%HglGAjx@aE9<)yo&xqjCODM}z=Uod-?8{d94wGO)u
zjsVdv{jX@;T}YpKLKbh>YpB|sI7j2}L&Ji2Le=z@!vW${=YYJ$4++U^hn#yE4jwW)
zT9106HnEu{`xOAiZZUN$%2dbEP<dA*oAg?0KKb2GyxnE2p)pLPi+rZT)46RcjGzca
z>i7wVf~NWwi7$ijyi?j!FG?SxQi?C@bORc#S5T$833PT}W)6>p$3}R&$AsN^+a>Ob
z5<TyZIoXUSJUl<<-`T9L)s~yBOn6`OI*U!hn?Dyj6{|(namb>wl99oQxGkebh~;@!
zm$2C$$+M6;5Q!S4U;u{XiB5&>;EG-B9P@sat1!^FXOE*nEmUdm&}e)gm$ITF>qAE$
z%OKhT2~4*klIj&|vc&$cjz)z=AeKkZd^(5ebX*zz`LLyH9lBJv!FQo;M8T2S34_Za
zfnXnM;|DY4uWH@Ji;Q6Y!RnTEdL(>~OJ*r1-R5M_t5u?wPv2aN`xKP6#|d=}9e7gP
zy{;>YrzG@tt7_jiQDjfC=Hd8f(Vd^H<HxTu9qENuOFw#2W#V(c$}BA3i0FInpN1q>
znk|ylzbPjhjQ2hr@f+6pyzjZQ?Zq8^Zarp+K0Ix^F*{T;G5zqpkC&Sk)c&eAEr97c
zCuKtoH}=}MH}@08M(qx?tPL>v-5vj?W4e{qP^^QX_QFNtFjQ$YmfWwQd#v3lxID=9
zm%_@P<9)4}b=v-nRsvcQbV0)8%jJ=R+k~OCH#b&OO=mTud$bT9BgJTjKEfK<aoFU@
zxEmWc>Ly+$2m^`3bm{?zHC?sj*wBLUx;su$69H)SotM{kJy-a%Ts5)4F)`@T^K+da
z>O~EW#0z(&RA5zp(mWtvEE2UYKlEd$My#%!y)ip#_j$Tr_o}m5X6tTXtl!k`@6{o$
zStg(S;nGsYWCjji1vdi%!BQQa*K^sF7izSbzMXKSndrVLtZ*Ie;$9uHZU|4jQ|)*q
zn!TD|vU;AUxW7)_^Peq^X(L5V_Z#@#b^5#x_LvSX-C?4Rc#>_E6~<m;Sk~3rIQ;@)
zT~8pSD5!KVk|rk}1n@jLTi&4qGhm5a;xA>TM#tdx!vig(2ae`TVU=Ds{!&J2l^O~#
z$P=f%QkNoyLtW?@Gazt-k|x_F5y!9~BoM?pXuAA)Fx|sYAQO;n7>N(fSSCMTlVj;j
zuHK$wE;Zjf7E(H6!k5~wR_V;CimXnc+PA}uRI16^?F?{H5Gy&8dJIbt1WhmV%*LPq
z;9BJqZ6W2azOyxz^)^U(w-ZZ^)+V?e3&b>g>btaOyTtAv^JY#erE<mn&0pSxNT|v!
zO!n45a7QPC<X{vQ^f<%<jt~f;s0j65)#C0w9>s%F6u6Su&>)&;vwB=LyD_cw`Xk%V
zgg>flvXSCBzW>R(1;(glf2~L~JIHt}Y|ola(~H{rPrQ+^+>0Ck$ig@LR)w94aX4ub
z#=GNR%&BMDC@;J@{mCwaBBPt62cb{}T6M-GEnrmSVVf7x&Ac*N$oQOJF$J(59@-9a
z7(kSO?CQyK*Gz>vL)+W^AG2^Gw-L8lx|HEQ>Xm@ht*3)rB|4=J2?;zSm3;JBtV=v%
zHq(flv8RjQQ@P%W?49Rprqjld6hxKE`H82$U05=jEh}_012}tWHPH+i13&$s&I7?^
zY?d>m7v{-y$T3LWAMfoQpq?_Y$tJD29NPmm5!yzWlF+*SDNuYK0t5tiW!*a~Y%ruR
z71V%H&^|f#ZF|_-T|R9XVhikmjtBR3PYad%L@g>h94aUhy%<#F+jRSaEnsZ@4dbwl
zpC*0?GzD5pQC#F0)A*uP3@+Kn`$M<f>cHhqqFvx9<NZ-;8l&%}HemfB@t#7v^GLt;
zOMA1yVK)dBdjiRfbCc)u_FtSCc)RwtuIC%ces~-)W(Zki#e5f%r=GxL`S#V03ATk#
zOmVfXy>Z$6p}>*4Fe=eJYLuD3CDjiS7C|Fa=&d_xd|LQ-we%tV<adr#9Zu_!Yb*wC
zWesYk1$3JLk*Qi)(w~E%;Kcxs*Qx0A$2fCWu<!Hj)y65u<8G4D^M*XO4XxM;>0}1G
z^vw2Rx$3Yqo+tyf>)}lA%OdnVdW#>3V4Q1pG+73@cc->)?J$FMt+X#jLNHRp3?{mc
z<6q%&qlk#}Ych8+5KgKojzdB$@QGC67lh0l)DbAC#Dcj7dBF$>vD3iR!9HCgg)d;P
zVjXo$kOUN-wVI(Zfg2byLK`9`iF`>8RW9iRKZ^V$=2b<#yN5r4g5h28#!Bmbx!Jq^
z8A4G7v)2E6s?asr#4fS->DJY6YY@bD7p>Zca#|$!f<9aJgkiR|EnK`&)aA5QgELb0
z)ICO6YPMYPNnzRlBzDf)#y;b$Pmaw2e~%jCFJfoGhi8II!bS!4O;wS%cz}#o=`Uag
zvdgQKT}ET^xA|KIJ&r+y`bTtok*P7|))M6kSiLwd84O1V1S^oLR}4Z_-`DyoMvY~m
zf-~)Yj>cgCP(XA>!}K?G^srEO)-Ph4@C^RwDZM<Ph_{96Q{p}6v!p+&1RxY4DIzfn
zdj1pDQDJKR*WdIkyG3lx);S7KlUg+hrgC%0`;nU<H`4xEFYO%bRKY>_$_RDgE&HN^
ze<rou+Zx8X1p*Nj+M{eMxRa?ln_VUz4Engep2cwXBBp}|=l58c#flE}7u$IqjaT+c
zzi&}m@U~vnNUwQ6q`CK1@!srC!RB5eq&+M(RogU~30nOw=$YI(S))~D`ZI?m4f>ix
zKq%%DvG3Dfn!ZdAH;Fm%he1J53<39)^?{Lo{$G5|{|4nD<E;756v+<zMBep&rv0<Y
z>x+@GBTbMN2MBv&@ygZIvj9P$S>)fm3!HrRBm6Df0KO6CF+4|)f;;OHf6&};eP(+4
zAH}xrds#_*E_>ev7Be}l`ZV>##)rYA{AU|QZjl6V<Ad>4^%y`b+mBW=LD0U^X;u<v
zbc+At)Z}U24eAF7HC+t1EvnX{=t*3fX!R65#(>iY=!E!Fenph$@55QhEaRDC4C@OG
zxK59tBiYz7q-~LSE40(&^dNRgV(aoWH`HjYT{TE2BrocT2~-#8`({}bZK^g12h~*Z
z`)(n~S%>IoHG1J$Yj!?uGJ(n3h6-WTZo;2AXKQ<=HZ^L``CUC4R(d^0S~(mQUrvm*
zITBlN=lk$u2m!em=LMP_yOEgVRSzUeh*GJKo;4_>s>Vk$>uPWcE&qw&J=W04p@7r2
z)gS&+pp`r5J^3<*i%w|W?`q~X@i<kms|F{DXhN_w+iY{1-NU2Ms-sTHI+R~Gs>V-H
z%c9lgeS26R_=8^Gi@*B{4Iqp_&<%a`yNvl4EdqP~TFa_R*wni=eft|K{MtG|MXlZ#
zOS{>;7$p+lRgmByERK*N8k>bl12bwsR<-0A1qVbiV=f!L1Q}bMlKMi8mF5=P?1}V;
z=LNVRqL~l|S%nxv<%2`Qc-V64I8mkRrfXLrN)Yx=`(E#kdbJ&9_L4{?T2jdGGLouo
zg-DdA7D5+z@_|T+&VYbsA~_LEGS&g>4bDZqohH{)wx}nE(rmXV`o7#k<aGTpu}hJ8
zh+q_x7#PG!LV00qq&gW&Qe{#xM0r6-B#(SapRG_F0mwLPG#KIrcC&gI@eF{&E+MnJ
z1eK_WNW@P6GsH@%M!2CxYd8P^!kT_g$B6{xl_XdB>q(3B^o6*=_9ED*ljffRXp6*U
z3>jV|)D*|Yt(Z`f6esT!L8$o13x;r-byufS&BwXl$J?b6%UBf^dvPawTh98k)^Pgn
zt92?MwxhGeIvmwlPAs_rbXFjV9T+fErOj>>cwHDLbc_W}i=wO?0yoJ&VOi3+0#~1O
zppaZMg~i;eoN+ytR!Qs4^+(m5C*Q5Ofx76E7P#w#x*SmqkBKasb<si~7D^bdIWdDN
z7u*9yGA6AUZaAu8-+;%&Gv^RB)!h5*&9v9yu-2S|L94-SoF5wu8%dsr(d#4Vpdsq|
z5;72GOLn-&7zYy82oYDa$#0=NAM<lTc4IM!Nm;Mfn7*038Sof2cE6AA!Ostw8DAP+
z8r75L(Z^+&%hP@bapXS2oPry-X)kglxa|EB{%xr-M@1;ACGvD*Q}%j#DXlDk`HG%x
z0eFt*QIBa~L=0bW$bvhlo6vd3EI@<dvkMSj2%*=h!0(1J;HiX?Xxzm{u7!!|D_3fu
zDy-~q$(Pl16{Fx&i973^APS#tY#vSK?)(CiU_`u}rOKNcI0~HKwYt*xr?+VqvSI34
z&8~o5=!YOUB$)W#oHSG`JSi-sJa`n=GEm6nQco~l&%CcPPk`u>?r;=nSTReQpTM^h
z!yc7btP^YAa~A*}D@!#ifnTc!<H`}odf$q?MV=(nKVljP^A4iSuN8VE6tyHnGUZ_q
zu>HAhSAHq^Y<`Q!FkNZNjqJjE)o$<E?WHUfnmE44oIrSQG%)$)+^ua|4RR(x1S+%9
zuh2c1@n}VxfdXlfUQQ$&h;w^Pv}4=ZisI|t>CU}<@Rsq#Za?Z9Y{eAq8~-F;1nTx_
z%#TT6E?M_oQaG$2Wn9dYa5se+tD2tLlUt1qS>A<I#{t>z1@eJmXH8>)tv2zUZvip<
zXozH)L80Y8qBnbC<#ebyRx&<tc*r)#M-9fH+jD5Vn*^3#klh&%sWn~OK1)p#Gq7i)
z*7G&z2cXvqg*>#M{~R|kzu{)gL%k@|F?x&S`B3hv1Iy^7#hw0fh9A-n2Tzf%d<=kx
zSfO0^)EiS46Dcwi>hUUvnpeOdT7TH^@0O93u$0{=#<hHoVG}!Aw#ZEZk@Nb@t(y?`
zy=se^oi!UOxHH&I<^{@g@oh12$&sHS5nw*CT#JuFDV`%aqtxW@TKE;&0tB>_GVF2x
zs4MUAQXpxeVitndZ6^>H|Mc@w+JB|dYG=IXbMT!Y@fQ#_=?5v=8B2hls+;2t*e4vv
zfTy5Z8Dy7qKu<R8kp8r)s;AbQ>j_Sn4qHHrX1;ZMWQ^&9-Dd_3g6>=G1%fjZH?bZM
zms0I?mq=}!GY?YRM*OO?S(Swh$yg3*nwEYB|0>h^Lfy6rNEx%Il_=6R=(yG8wjssI
zaSb9s;l`Ax&?slb$@VSyRo=P&#-h*bCd3QjNk&-)hh&a!6cC6+o(Nz8>L9$H<Lw4Z
zlnE49sDE64iJsfcr!yuGZi_8Y8=vh+hE%PFEXqOgA7;Qw;vb+E`xH5GhzNKDz>cd5
zFP6!wj>kM8&3{83+6^oQJM;MoRbnY=MES_tLs2Kmv??VR0z<jv_l5ueP<56;adhFf
z4ls}*xVtB~4DPPMA-GHM;O_34;O-LK-3b!h31pDq?iwVw`OZCc>;8nInyH@N-M!zn
z*0aRAe*P_`uxMiL?r=HSq@lZ92PR{F^~Rf`_o9WkCF(YsW2tj~Npg?tl)gh<hKyV;
zJht~98}~}v-(I-Kr2Tyu7?w(^i&0Wdhl%iD7FH^5ndVl0lG&^7@(R7&Z~3aR7SvOW
zh^d&ut*C`W<(+w#+nVG**8--4Op#dnCa80(EVF|~_0dZ12hB!)?L(fWHm`-F4&@+D
zv!Q|mMP;o-)fp`g&Lobr<<}SAq)V#;mnx48mIeFsAcB*%u8v_1{gI#t0%>Cz&O;<Z
zpCOA~<RLmR!&~{rc6=+p-OPrp%h0yaI&4k(57>%HeZEvIre@fZ#sS_LQuc*<jJ@ci
zBuRwCb6jaY_zp>o-@#ds3NTL6&k|o+Y30RfLu@Xsh?#xyNMM#fMp;DutFJShiMBhS
zSD4b9Lb;wA<p}<YttdutR3OKua#mJr-2=@*Nj+7gI$uxjGr`i+-)u0u0@jn_+d`vm
zkDfa64bNWofZDe2krxqrAFN|OE;i2~`8nU&H^uzj^8UUFjcPx0N@)b#am@<+V+S4A
z3ARwe&io$#Edgu|&%58Bcg(y<{KC7??K&^@ELiu+t8{QjP?<92mIV1nnU?r!A_Vz9
z4>m#OZg)O)M?FjvG$KQ)Is=f;I};NA{;<9mqYjb>+Y0AP&djTQSN)Jk8MR}qYUCgc
z43PwbNtDL47(Z>~8FYEQ(C%lB<&nf2`J^LHqaVC4B_ygYmaAmy{SXwG6B#MW^F{G(
z1%y?UHvd7CVf4q>nAR3~1LQCC;}kpzWU`CG)$A}AK9)H~*;=L!@g^oxb3<OZjhdnv
zyNh_8kiVSTo89abmtA~^D9g=KRLgA!WAah80z?EbI76N<vSOC#bviiM*k6>9Gpy3&
zIfg6iX7<=enGo&-8cWlaJ@+o%+M@;7^325vPTk)Ab1)(E^zYxRriI#P!L9P8w#H5H
z$?Y+@f4e+|4w2&<G*(n%i0K-rRw$@t8q;bW1tO@!_c-u@I2W0W)kP3tCu4ZE5U0(z
zW((xKdF(q)hp=4X%${+!KHPCuPJhIK8xmd_^5VfJLEoYv^v4E{u>TgB(=Jg3mL`TC
zng2%yS0#x=K_PLFqezs>^vam$Vfgg;kv-*`S+YZO<G-PJ2K#d%|7QFvC1@|x&MevQ
zh$)cd-Y`ox3YI;o9Px;9;rFOdOVw_q-65!s@tC%hT;M@_)F5+>f3s1gGH7PF`%(M2
z(vGg56_dIfYbdbU{|&I0)kBi`w4Lt;=g+b;(K>$c4^wiRH&!pm-&)odPEFoOGNj5X
z;Hi4H>5d4PJ`igk<+%4Vry{6VDvP@PF3vulZ%!UJto=oYG3I#$vlyBuR>yI;mNhz#
zL=h+&6{tjo$9gY<JdI5K1$8L82HCuw8=ncIL@mt}L6n0C+4{hGqR{|>6Ty>%2l@UW
zjPtV!RWNgjEr=5TGVd{Vv2Z=Df*>y)1Pp4dh4ZJ$<j3W+4lu`sGH5u=S|uT#^+TeO
zi0Oz36cLD`OM255epKWRJ->W!U(MLKWYj3spDPhF48{gjLp%w4OhOpa-pvNV08d4&
zq*nJO1MezsPj-wyXk(Oii<*$OK=^7zXeLwhFR=>dzd3b2c0;g1xSOVPEPq!+K=7a0
zABIrinK(_A(;Sw>MFKp<qTdDM#;ZtWR!#Y|W_b*^wlX>L=e<m|b^Bgl>3$V@=<96#
ziIOobE!85VGx_Vq;k)*CzGD%{M-6||K~v=ZsNuJwLX3H-jbixDji(UF`vhqWQ_vQb
zj&<V-7@;*}^Q9P9*Y#uUD48oNeAG93i|K;5Q_h0w$ZD37TAw@BOpYza<s~<e*_6u9
zRr4iltauzDrj?q3A%EvrDi*c-r|GSDy@;R86{UVj{W8gtBoF(!m(w85rNMC?FAYjl
zX7FblbF7W|oCZI4Xg?UZ^%>tPM`B%I9c(*3XC^8mGruUYWOR$>Zi7MJ`ey@euelWp
z&*QP!wpQ&>Ev>AMXV8|?Aj4I?Dp9UhM#(?F?3O(!*s(QC-_mgzlkV*EYp$+GF-6^O
zRsKuWC%0B>ODg;po33o)T6;XJR(0R+TE<<&G6gToO_rcbF_}CY`5$0J`5%8NFY62S
zMRfc7d&ybz;G1yyAp7p;d^&Fs>W`{BCfRqvq{zq05R%~x!yew!qP?{D)19>*F`{@V
za4YX+Xc9i}n7lJ*8?+6(MVHPNP4O`{2^jL~3(~_W>J1J3CEKX*yL7LA-HU|mbEEM6
z4f{D&83->WKhokEAqb9toQ+@u?h(ktJ)jZt`REexM6Tsacr7>`@OED$7W>~12Jzph
zlw`;y-o!f@dhX!#oWVznQ>;|=6P&6ZYEF2b%B3yxxW|4$x092~#DneJ66g^)Dz8!p
z86A`3tK$Fif4Rzi;dvt6Y7SB*AJreSh`1kaxIG#FIthClhP1>_MJahr_XWxD>9QL+
zgHKD*#yN<%cEV1z_$jQrXQpkAKQju4r3*8<H0;lyFM=89zTCXWgZ^TRbrEg|ge{_J
z;8}b}VL>cD6Nr;hM?qRg%KS+-3?_!pIb$(?@IzF-6l-F6Z@R25mbSO?A;pk`>+swK
zPx+T!)R1&w_`SpsiSAbSd8584#joi}dIv2>&YTyOIYBW@yB0e8cY=Lz1!Y%Y@ki+T
z$!s}y<_@z%O5haK<2<?H{kS3DCyh$I<JGu=LnBdVoV{5d=byRtMGkFvjE6kjL*Hgc
zlSA1&_GLW(7PpwKJ)z*$za1mvHP7mNxh<`qSd86{LQ+a5asQ=A&%2YS@)x{~VvWp~
zhC7fCcYzVgimZ##ubXJ!anig-B0U#cQy3cdAQ>EI{_zO06e$&}^Mmd0<{MkJ6VM8M
zBT~{}uju2APQ4RB@64vDs#J7{L*Rkt)!U;T?nIIp&J&uhcd8kvF$H?W*L*jYxoG^y
z5VF76sbrN1+v4IS5`yyJv{Q8?>Gu@)KGaB!V(5Ow^5~Yu=q~p^s4Bsh*|dqbpA2g!
zKR*x9q@z8TqKKm4uZT6qY{aL?h30$M)?I~(@PRj~!NF7afjDe3T0d(;vORaxUpR1?
zXrGY3rGmdeh`wXyd*l*5l&@dYE(r9wMC*rCr%H8EE9mpAXCF3!4Mtba)G9qbe2=Mv
zAES~bs8x^kQ%jE-9<4Q5pg;@xM%PNtn}wH(f-j1jGjPs6St@=3G0F;7rZjdPItg=3
zc1)D~K`ZPKJoqj&lwUoNeWCe7{jI9-7N@QB@(*(_!c9Jd2Z6mJ$&Wv@CSi41xX9l8
zUlxFMWm_x1BhzRQk-7$OZ-Agh^-2JDhm=M8Q<5V$EpYs|H+N9I0^yYLZ~kcBQY8~4
zXS;uIu6#O$DqWA#Q)<#!=Ae5MZ9Ss_zy$rsP0&zh(HfIQZuZJ_Sn1o6mp`NM3&&6s
zh1F#NXwLkJj(hE)%;ls$I?LsnqF!Hg*Or#vh5hlXrW-!~_WAa^tEw3P^EEaP5EG&v
zQb~0CB(EM!k~X!HTDuW%5u$2ZWVp8GgxI-G`A+|F`3*&yuvY0y32E9!Z9uQ<Xx^j;
z_7@r}nP`MawmyL@ycQB1^hSAVK^*jJ;+r5VN*K|(S%hq!!bACaVJ5f<Rb`Rst0X!9
z{;7C_7JKo$_h53K0hh2lxs7!5d;vQQMrujkfX%$jh7`IwgRr|u`0|5kWSaN24Avhl
z2czKG7HxY|B3zzk9OdT@JASy|KQG#7U4Y-g6vny{+L&@hvPOV;v=EHe25gZRi|rI?
zMSF?<DC%q0sWF5;7JfYit5o=17cB@~@{>yGMThsjFu!^GOS!u}-B`4QGzV?5q{)7L
ztwTmdQF){2q#Z&2x9EECR%6bR{b)dAm1TEbjwuYBG`?5=Pm^pIGN>(vmlc&tVC^+E
z<zn4|a~{!wu~{(sbdqI!$(*x~4gZ4Ect^h%86&z;af+it>9b6TcCC+JX|#%(Zqv&K
zC*vLXZRdcfoOJ6;z1w&<UCY1oG<nxY*kVfZkl#wAlqIX1ZZ3w+)D7oDKyoWJJl9Vl
z3{^=<WV-ph6ZPJ_j0JA#ec?4a?OtQA)ri>vPEPE6r`nI0xG0{)*vl$padI3G!UW-m
zJ2V}#HAu36ij|_3e|wgigUEcb$)1*FT=YA~{a7v(*w&I-8LVjSB%^e@3fi}t82C*<
zfEDsx{S6bV-G|Cg9y5|fqE{KHzE^uPZT~tRXMbX*-)`Spj9M|F#&r&TEJ4Sg&HEN)
z+LYi#!qh@f6SOpv8;66|BC1F8sd#rHvox6$CibxqnHm{^%$ENq>~VniX{W7+mSTkH
z)Krc1l5TJyz(-41r5cp1q$LR0oLk@JlnRU%pg~p|#eY`|Jp(Dom!S^(6|TQzQ!CM&
zcnl5gH=v`rr@k`>qJ1)3@GZqVh8!{nb%ZQ793pIi73nGlD9a>Gzx-9`@8Jgr?LKcJ
zJDUQVBuSzm-yP@>b%AdkH5%UVXVcNjovZC9rMXXvKaQrD9-)jRL{Dbf(W%sjPxpw4
zJ5fZN_I8BHG;3`J@#}OUN}RGJf%oNG^JBG1VIpf((`FUf+u4PmEEF$E(T|K)z<IHQ
z_QbaAu;3tAW|BW(Q%Gp5X{`}9cVQVPo0X<P6v1>i0)ayLJ2qQAG}5>Ji5gYGPl{5a
z2;AB_)uow2P&E8cIQWg-4*rd#Hxqk2(BU!x9n$p+2szo}I;79igejuHNW0HX!WiS!
z9@3Sf@NbK$B-~09Ch%J2p^3PAz&6j{xP<pzMc$3USadkb(!lek9_4r|DsUFsfU_VJ
z_B&&r&@A+Ew5?vAVd=BdRri-0!<o9-6MN%7hDXh_6L`r6p}l1`Zp?&$$Xwu9*YKa~
zDCr&SF;j1WLI`}NBqR`$1~Jt_(L#>$MzS#jXBah-!ssbXhm`OwSLlxY#XL^^&z2G<
zQVwuN66jR{lM#R=31p<`E%-!vD)Ks&?}v$DA^iQh`|3Mae$d>k#)z0&!Mh!XA|12u
zrE~gjpD-||FQp!uQ@yCFVE?m9U4np*bGa?0!~N>0;l?00QQ=dGRy{FG4V#oaO*8mT
zJr(H!d67V8d-W8~O6S#oLZ5f*ptT4t*R0G?VePz1rZ)>$FJ=T5%LwY_!9x1NJR>fZ
z0Ezb-i3dCCikQ|xMj=W80ySJ*+z&R+LI(bTW#?&AD%On^NQRN|2B#9@!4+uX|7QsM
z51G<CB${WppyPC6hi6t?zf16r$o6+*5}M(Z-F!o(izPeVdqDGYdwyQ(dD-FqFeV{m
z<Vap(69RJk4F#V6W&oe!#hcv|VN`-!JF5-QeEiA!a;uIJtE9J)ah>Yp#j6KXnX!6x
zsQz;&y`Dd0^s-|pj(fQ-pLfycY2vYZM0DC7#6!jg_06+Uz$hS5I?ly0MKrX_-t%-B
zz0PR&_xivoC7M-<R&R5J=>L82S0YD}5G=3#N3dAPhc_klpGp<|wo*4ZH?yxOqQDp|
z1!7+p%GJlDvHs{H4!Hp8N{;B@NhpdLxm((MH%{a#On!3HWTRjr+GM`BsKm7J@bLYL
z6X--$=w#X11QdGt-A|WtJss2(*#3_v+&vGdH^<$Nr-Pfsrs@y3zwMtDe`rYr<sbMM
zAB_Ui_?Zs~$0PzvQ1bbD^Tu_#-b^fqS-*V*uqYf3+yMNE@nLv0DMJ|VG;glvKp-`}
z>rU~3k(fUXb51w%eq}?}Fxbo?TR-bTM8c<UR6Ohc`=?T`qoAJGR0CT;@2futLC09(
zKP*lj59~R8hz{U`lMNDC10x<=$^iE5i_O8H_(#Cdb`rmw5aEghc^C*oxmt&;{aN+D
z2)0;~oqg5Ca()VL1i<+mTX22sJ1GWu83qlp(O4=Uff)VXQ9!!Iu%Z}|(9QfOKdd6P
zLkaP?yuQ=q=~p1dtLp9E5cr)G(7@JJzOBkQ#<uSlM0fV~qE<mB$o%f}(NbmjC`--*
ze-$TZ3k<jbiik%TkyOqq+cI;`ssM)N=8d{z{xQJukWHjkcv8%2BoP%cwR|;6&DIzS
zQwW}lzlzKjhyE`Q7G!3iK(L-G%1EYt_ra8N54t6~SHb?P-=nlCr;NjG68#r8n4IzT
zY7lv$>g_Ck*YtE&bu*2w6&~PyKE}Z-eEka`bP%+}##xyS@28obnj+ycMzn<vvz9i+
z@zU|_HZHIZr~N&8RSC^n^%0)yqE$}gSH+fp)eVy#`P*nK%A60cw!232n)B_t>BY2>
zkeTA?If#yj&POvDb_?v{>lsc)U{Lw2<co*@-2zA}G*xpJT6a^di$Y|st=UYz57)y?
zWw&&|l_BIG_c!~7lFRSO!+%X1(?_`JWKw~CH_d(B0<$&S%eQ?)ctv5^Zb{XDr3&Ue
z%`u^@CMp`M-b9fnk~dm>9FexWk|xKaKNbGeNSq%Um-}+Z)B;sii^qbPDB@oX`##>@
zcpCaYyZRr4Nc=08o<H8jy8)Juix)y3Tg7Xr^VA@eS~v4k>X@G4?<SkQTtPhI{PCs<
z!bU3$BRk_xt`z}vqiPoorzyj2ulVfLAhM;6&=};v&*1He3=qej7Q!!*3*aa>cD!Gd
zrYN-{;A<t=oW?tsLil&ra^vsfX{EOZalj~kXKP(u(U>~+dO_3O;;-TKf4h7u4qSgf
zmMSt?0N9Q-Ymo;n0inlZc0=YA*kRhoY;wm!6Zb_x6pF@@>imtih!Ww7t%dumEb=zg
z-SGnovql;@;Kx(!Qz!uZ7o-iJeLt{V0~m})zDU7IyId`O;=;2U<MjL!np~gyyobFz
zT-`5!SS>!$t_1JE7Stq(>O{mA*7Ai!K87OU;C|J$@8a}Xj{DE?_v<|vlDpv)#ZaR1
zLMZ}4Db2$gRvrV!?%RcAv$~}T<LipLQ^<jDU_H0&usPrko39m}nO~hjtD!TR{Vk;C
zLKuH0;phe`tc=a+$PkE@Nc4iUhM@k=F=)Pgu9WEHdK*VdF4TH<lTJvk-nYkelTqZY
z1b>*}DSS3&)v9zj9d#7XD2wt%t?=LoVSB~V1mC|Zs;|_FmE!_?3ze@#cCDguyu0`;
zklx;bTJej&<<Ef!XtM+sSrisIo+Ws)4kQI9V20NBqOGI{e^G(md_ZkeB?_IPUssGy
z-DX%kr47OPcM#QUJ^@J2al%%1m^@k7;>)7xL9(eQYeK~xO#l;Go`CmB(^4l=T?3)#
zAf643<B8<*hyckp|7GdkB9^$3AS{uS`BBZzJPZEk4@G_vq1~3X#zDrr#(bYX=wEyP
zJt)OvJ^0fJXj~bcKo0*Bkd?idx=)(`vk7Drat{Q<s~at6`gLiwmK*>~!l>XKbgM$(
zych17d;A%=7PhM1fT==2gXM2nwRqiWSz4j4OZq~xbFX77&KIG+c7H8Qt$!lRjQD=r
zU!{FKRG5d4D&8UXIo-$h!Ui(o;w)N;@M%-lN>s7G(cR-x|DyB`6n_5Pw*5`6ld~Vs
z`ZX0M9>A$a=KoY@!DA$(Jsy89zX4Ziqs1y0ZR95OvT1X`?oi99+pWrXvZ4{k=OSzc
zEfcj;PruqRA(3<pb^;Y6;Og#@z?zdo^3qsKSoiFmRo;5gAXd+GdArwO&$~uKZ9w83
zOOndV@d8ptUG&_;{E;ach^d2hklxn)UKdgTFk%4b0G#H(lFg7#NaLv9@DW$yoAF10
zhH*6Z!xa(b;gu4UBuNFX3rJ0SX}9EA<y*}_n&L23F!B#bVbb+t5<V5w`-IYss@l5Y
zb1{f_IV;D4)3np`@42Rh&*){5<fCKPJIurk-RM6}2G?jlwX#LCOGdnZiSC$jV;X*C
zAYO-HDW0eDx$JLLQN4On4WTG*&{Q;``(DFg1;l<>cH@2!1qp5(<PT;sQv1Nt@h|id
zGtjHIkgWC%x8NF{-g`vZ06B_NS0AeP9Zj;@N7OMBz{qC%yMg`aa_Z`456#nUhkP#U
zr?tm>xwg~A@a@dAaBt%eHp#mxqA?vYW-(J>hnNQEy~!Jll?cV~1dG?JZ`|b6%s?uA
z_<3ibi@Fu@ZVhleyMvqzWRz2J8aFK00oVxj>qh<M3Mzo_-tuNq`U53BT0UpGPb#zi
z;ThA)o3zg|V`6e>`DQeF^k$0{no~CS(}Z+i(eSAD0WooD?&A@oZ0qUUMHttId?6Wk
zQjMWZXDu+8RaSV~e1_SUUFcg9+!iX_<^5K?Trk35Q&2L!`sLTu>1+0%`As3B;g~n-
zN6LCWVXprixP#EmUD)~cqJgln1V&D8Tr=h=b`Xq-a1`)-UOy`u4~Ra{IJ$_03inqZ
z@*xOXtOG`Ns$LyRN}swEa4OE?H90YQWb8$>V+hz<0%gm$w+h{G|I4(K1*^wV*Ob{t
zi+HHMysrJa9(^lc((ZUBOPBCIL&IK1Pm>Aelll0M>`UP3gauW?Kf`Zv)Hm}DLc+z6
zx9f2xh5>;nU-F?QiUZC87QOK77w?*XCerMApgH$ERIGaQ_-wN{9d#-FOvn58$?jzd
zi)}v~LshGl{*GT$%1b#edBUuS>Fb1u+|*c^&VHGM<*%8j!U^-Q81h!wtgP%82<aB%
zrwQ-gS)TS2diY=&dmw%!Z0dnsDxCaer<R?C_NkQwio1fzoa)lLOMaYyFcH>F!T_7W
z%s~zYU4UcpJtY5N*VL%AT0ZC%c0&@UT*ogOqq#yukxz;G35B}Li)XqMQjgGWSE2iW
zKk6<hiIp_-bh3doY`VA8zK@k^4!4f^5U^U@+Jq@v7m0;9y$lP#{Hd}7AYE?s+M6Zm
zTWS84V3d|h8Y~)x<RM-A1N~S8G?G6<89i%FQHr8Y>$aLd<EmPsbtiXy{s1=b3k}1{
zg&14c!}$w48rqz+3iQL6%eOGl$n+Mk>wT4Zx?O%fD~wZ>Ups^p^M0=gELJX!I=vzs
zyUs&?qXeHB|62JM(-k23|MGZDk(%QjHD6}HiPxc7vXX^ogP9bmX6qvmVyrfR{aLmR
z7Z;u9ZDF<FEpa&A_{iHD=>>w$>Yjm<-hK>q=7ZWdB1MR7yO#X-@2OVH80qypyQTgV
zJYeY$)c4Uuwc2}~?Mv6P6fN1Pc%EcYB*=mDdIOe+Wdrz4$|V9|7B!5`jH*-%9BHl6
zcO=}7rltP+aY_uuWRnZW=hAw+pVIE-_*Ei5*`<wVuxq9iSZcM>TB&5(zy{G8ahiSE
z>=vQ5(TWdC?xVfK(5j=Ru2}SpJMlg@X1<rQ)$XLvlVdW}p5(cM{l4lz<li0ST~n^<
z@-K#q5Oy7zlzOV`GVaTIfry)RLstyxV?`yh7Fzeyp|{mqV_lZVcpp)f!h?=VIv%DL
zC}tf9v)A%c^oF@Kt7yJJ9VwGY+@X#{VDKxbZc6wcl*KD{>37%d%9zb}y2AH1E{F3g
zV$;3PT%P4M)(hnev;U6H%+Aq-KMtdw6kXY!e9o+D`KIq?hgCyIbR$5KkPNe4h7(Rt
zB}<BHT?T-`UNTn?g69FG69oww$5cBrV_M1kp1L7w)i=pFm24-b(}ZNlNE!lMAD$oA
z{~2sgW>}GY-}QFwXCn}jTU9S_`=Yh{Qr}p`QkXI_-xE~KQKc>F<#x{?w(<Aqwt@o#
zpY2ObqkT(o!}fyH8B~_{Vn5iIJS;D0>jdp_^g^0sJ1L&JH+kCt@OZ;apa|W{r7+Uz
z^y+>?)in?&$3r*DMu-VlIi(VzHClc90ZN2R*AS39!V?+6DG*{uy$>r&z(xTIxxo78
zka!qnvYc9@MYoRpGUM-N+f#VFu}H60p(C&_T8B(@@aB#Y%D`z2l`3}&wvG_MYw~_Z
zKwQeUT<dJpb6YmMU@r9~UBb=5&|}BIM#tB}!f4F0m#-{1RO<K?h+KI^1s(~2fqQbV
zWoaG4<c#O^YTso_Xj1c2)b5ljnbe@5@Weq{B`pp%$CItl!727jzOnK7w%3TAP8o1%
zq**%dvAfgKMQJ>SaPvBiO{5TxOPT4BPlBH$Y#(n6+#=Md%4j1ti(-Fp0><_E5Cz6@
zPBswNTad>ZH>YzW??xx<-w{f`f`DX>&y+|s%S-P+QRKI0GP-|RfHECF{tN}CJ=T~<
z7%R;yiq;F2b*x4|)A_Wph9A@DF;-F{)Ik%|z9y;@M9FcM4q7x35^pg`Gx=NwCIHFW
zllk|JX|^bsS;3ttWTIBiz8RuZ@6sf2GlywqQ1dOT&X?2bPYC>+z6H|izLO&Ft#TXS
z9P6x=k4kUY9mz%$po0HxN2Tw!{;}Cwd8vakJpet{kk4YQ^Ve8`27<1SXDo8y1Pn}p
zqp^95!uI8so$1GE1+=Xjusf2QCXI-A=X4Y#2Y5G+)s&dMs(j$J>|vgr(MBpL%Z~PN
z842|MSnN8pRxta6Uz&s3WE86`(D#$pIZ9s^MlmwpH%Os^jFlMqRi!gYmcvq{^cM$2
zsV@$pnHp)-3K{Q@ed3i%W{zYM%TO%e1L!~9b@OgsW;zV>%;{zBHIG6x7!q|-ihRV6
zRX^-;6X|DA&)*P1Q5|{3M~DGRUcAqsC&#%&2NEW~Xkp~AWU3zx<t-L0i<pn9LBghS
zcBS9wv;|BwoLR$t&N})OEVFZ805VwhcsVgjWUxc@{%qs<2A<|YI?YZjqX9#w`Mb`-
zyV-6Pvj;O^>dEi+y|$#U%mw4#@dv(ktL4W!S+L*b?;_#tiM#HE;oQgEC%%h2lcf`K
zAW_`;A<*tmq6{q?2fgJSu;cKNd~?Nzq-P9?s5VV{1lQ7GJ?6fr<Wi=JVU_3VLp9PG
z_$m#Kk3ndAuht}M*7j%AdFN)0tRr+R5h}KWGuc#D&tbo)HOccpyIe42xN_)nB+}Xj
z(3-lMGlgo;;I?jojKm*zCz@ZdM6^yXfLr%!5NkHQi5pU3`Sa%zkW-r4C|}h3-Pg?p
zo{&hSpeKSwxOu%j+<uHea084SIBgr~j4(q)=4>iNVB=@tf%LaFB}CV6E0){GtBO0(
zenlqRzUwBX)_(%XfOQ2{Fn&$YX|dC990F0}{&`28OX~$`_)_r72Yg(YP!e;<n`$pb
zj$W?gI$RTM%$;DPz#ZKxxB@JOQf~P`zQ7=XqU#+Pj5nVSJ_0O7l7AUD+(I{`bbqN~
zQQQGOm7BW=XQ=oPPY%qkTZl|Q+%Og8-5-Sjk%_qBK3B5VWb{Z{*^yuiBeD;R*Le;S
zTD`u}$_LAN^vR<?cWiA&D%3}R4x6U_tU=Noo182+Qu6{IuX<`etas#gTrlW-SFW)m
zS$2*`EnW(kDBEVU|F~g*o|@K&xQn=M0C3L?F*XP+HpA*5qi7%wHS>bC4Y3dijA=C2
zbKhGsD0!O{TyP&fqI#LKfuhi<^1(q!=ur*NZ$y;6pS;ZdDjx^nXPFCF>qb&z$(Ox%
zfUW&E00H`dmMPpn{<i83BrHq%ancW6i|UsItGooJ*yE%DgJnZH+VrA0qzw!{CJDyP
ztgmj{C9LsOnny@e0D&0YEl)}nv*`f~uY2oZV|<&^v3;-zC{e)e_!r8k5e7f9fN{=D
zfOYPI>JJmQMPhOectpIy251gjII&Yit%nXgXm;A~eCG++jBSS|m1(_bk@v<mLiTB?
zjdBp{#&p6Zv}nqOe<|jK4xsYO*UAL2z#<3L2yI=*UVX<*0T;+RPFV_*B8YDCc~GW>
z3U~U0$&OXTh%AZntr5kEUG0N*O8$1#>k@a!`Et=zkq_*ZbmvekTr%l9f|~7qi7@Z~
z;To@M%!g<!RF$^hA7d5U+3Lz05I2p-IrV=NS(?VL^N4|0L=ITi!%rb_Nf|>QA%%sA
zfWh(wjEbD2WT5E+XG5fS?>60{S|8D?KS0D=%|a06;JOd`;4A)cM8;|7JvHdgFIj9D
z?G5m2F7oVRpAqg2ZgQ(vXtp#TuP<2nxLcoq!p0fjf#2TyHDKV_QCN<7HdEjWfHmK9
zsfGNKI0a<Rn;t}zXdHMZRvVt65VxH3n%1R&VoWni4iU&s6v~W4-e+48atmX~!}N>A
zm_5l7v)k_AVPA=-<!rEoK0w9H)by%bf{kHF!Ax^#Tx(UQ;3=6St({lkK@+@IuHlt_
z0~U4%Q(og4!<n#N1CsnH>swvQcS8ew^o%Vm?>BRGJkMbRcoxrBTk$BSi)97Sz-r)R
z72^O5LUh<~7&rCR;f=IpeX#j#M2+KZ1Z|qK<-8P7;D*$Da~j~w-?8U`bg*?Rm&Wd%
zPd5AuZg(2)HbviUrH=towgo3s`hb_}y#qg_VGWpV`iX3DQ~Mxj2+7O-mq-n0Bk4<x
zj3=q@a6dau&-n5Mnk>h&?X9aF|Avald~x%J$Q`SoEfle;Gi(FdP|ii3qmTvgqbhaX
zYp4p}*!Pf~-ZAjjJUTj+>E<d=oM3G2ZY9y}Q-iR;rr|`<nDSp70<9hFv8ny=&2J2s
z1bhZM3(}?7<9|n0|L*MqN#UEg-$6duz+H`?a6>nk$ic{vgudI*F}XebC5YbVf;637
zMi8aXZRAQYI?P-g5xO(6omA^^k+>*ha(cW@>YJ2SSopcCaZxP_Oyg`tZ;q<dLaa(O
z8#%S#97^{@q6l(`cY^PfatI2@jgt}Qk~Iyq8%yDXZa*mr?#xkolsk}4CZ1k27TXj}
z%_l3KyQ#PnZe!~v+i8x;d%-fftN^wSXmhgjg5Gv$P`j&jhoC$?u6Og8e6VvP1@9dP
z7Anp^Nd#2`w<kS{1@fJJ-&%8fb1+17Qv<e8r0kO(^Q6Gl_ylt3L}AB$+?H%`2{-8<
z{u-f0%dx`S5hGD%!hJQXCT9`?&RtE<#Zbg2+tGQo22ZJ@^wia$MZfh??NU;@Y`4{)
zCuKP=YBNCarRP9PpvQwS1g#r0hIEV9PTbg}5-`b<AXTv<*{R!9XNo&1aP!PP{agt!
z>J`&TlWZJsr4a;t@fm`J_Ry;Hqtf+8&e18-FoJcD{)$<tQ-;hP3|<)m&BeDr3TZ{@
z4OY%Q8R1(ASzpSO!{zXQGy$7HVdh00jgN;*Y_kvMU48I)dK*=Wv>NEt!6BL0cfwJv
zWMCUwlYJuJq3rkz4fwP&pYJ)k^**ssL6^F5?Ru$O<}%8O=CTB94Y^bp*a|3=cME7K
zao->(I9gQ4sn%g)rOCojE%#}=Jk8Pg=2oTaAvd5gSswr~PU3|<h<{37N^Z(RZfAr9
zdpr<uobz)4F=cp|#u9;H%j=j+BA7I7F@f9t8LH|jexY3mV1s^YXi4jz1|e2pAqCDw
zPO{nD`gm@J{`)y^<!=QH(TXo0<E_}Yd%C<j9D;82dHT%1f2K=8=y{B0>DDsvdob&X
zBxdHcJna+r9i2&QpuH=IX1<Kpq1yl=Nc0-0Zy{dW(wQ+~q*bgxyhDyKWeb3ulLo2p
zwMvf^Oss$zjWwK(^V5>{uP9x#6<f0m-#-K|!P=7PgZ<7>k4@X4FkON>g1(5`Z>0I~
zsZ7;_4RM7x!ympjk$_VUn0t%A=w;Wc>G|F)<g|5f#qkTEO5n?ea3&-*zQ`NFi!-XW
z*3|vVdT~BWZv7`<aRK&;Oo(v8h>q3K*(_GtdpcZT)!udcEL7s^%3ENiZtNSRa*LsL
zaO+9H_9aPR;fN`nTbQ+n+vJ+Zv+9m!w#bpHZ8K=DXaaSDJ!&Q8*<jlsl9%p1$k@F8
zW050bNz?j&0@If-Acy#XK8p!NB%a_9DS{#N=x@xbp^?TEb_pY6I+a|uD<CDBCqeJY
z&KP(QlgBO8lWisxN1poqbbKvms!-muG>@uDAY)7iCbJkL^`PpTK8{3$AVjrXpAc~4
zsTazU0Q6c0&J+pmr<hvT<X=$Qi-c|2Hm;y5OFw+b7@EBEM?`MiQ;K(Qghb7`w79-{
z^&<ZXA5lk##g2~BzOTd{!j&6)1>s#OqmA36GhBY9xj#W$d>&XpoDewTis%OYxvBxA
z?#hou>cV@;{U-}AHkmTR_FXMrIX7I&wJd!$t$`HtffyQui*|s!o5BuY?pA^W|L_ql
zR=rnpXRuoam)324qc|U{Nfek9cDc9giKAbbEl~sp51m&S%N=9YPM56VGR`gWK`LJ0
zBxj-QNk!BD;`ad*=n?_vF^u&eycD?>BYS^sN)?^HpASXjtj2^?&Ig<GRC=2C(Zp~>
zYtQ8yQ~6(DedhBlm^JV^rE}WOf0WXv=Nsk-LLeLxkiU)}%!m9C6@_&tf}T#lGbIsG
zRvr4vHgBOnrSf(s?<WGwB7ew{sV}T+*-=XK2Lqy3Mbc5@ua4M`J+%QvvI5*%9IE+c
zLYR~6We0KnDLw)$MhMO%%;D|Y^CXNkP}0#4K%ZVSB<^P|>Jk|ve|gt6-Tn%eo&C92
zfzAAY@7_zrIYuQvQI9=oyNuLJ_aF1%i&4An{_Ndv3DZqBnmsgc^TP*>Q?_m#Kjj$i
zE~@FfUcs?>${r%E7OspbUZO!AleMu6dOtc$ru?CKGtB&->XpC*s-;G63#Sz#^ukb}
zLq-b`_Ei=khtlYLT4YaXVZ=h#O`Dm&?f=1L2W?FY$XmLnW?53lpSB@>JM}g0=53Uw
zny(%j5;!Tf3}ri#hCD&DY57SpboL|J!Z5^?ER^XX0=sDC&Vjd)T~D#~Bx$VRfR;GX
zn50@*%T3KY7#7xQ-~`1$$tAY4F_lj?x?Dp(o~eBI_9KK|5af``_&q@6Xo%P5M^+n>
zY)!tWj?-iZ(awQyo!zTQlS2`JE+RAe3W>I-93%wk*=sQpzHjWfSKbl}&ctBGeUQ;_
zcZrOV`n33WR0?>LQy<7K^M%w{TE{x!1&Xh>2gxcbYIPas6`}B{ck9XDTC14_U%63w
zgRybV?bjpK*8Utf2?)!wQBVk<7T)x~axFkGTz@IQq4dK7=#1TR4g40Or~AkV2_8kt
z)6cU0RxaGoFSM&bVqsQH9+am?j+=}X$cTdg#{kPLBGne%TYbX-?0Grgs|z1zn138u
z&9r4wiUJrGA1UfV|E=}UcrduXtT}M?;?Yr!#2h;)bQHLtbib7MgO^mVVyEoRE-N@p
zpRSJlsE<%4#fR=`#D7-XGXq+3Z5!0;sPt97{0NHBN>ZFJU|UACN%~K1E)IC}E>Zs0
zD$-h(D~UyAXZMO15R4yAb=Ln^Kz#)kjHrNCnyu)_x?i;7nuYE%v)9;Opl3BV|9^4D
z>rdavjl{$-1CHzeN6UT{P5(o0)V5I=ol0f?CTV#g1iqg|bdTe@;m<%IUz=g--Tx3a
zh`_|&$8|#UPF$n)Y+QyQd6{26XI(c?;Ha_@y~U{pI{lc89shekWiAW|MgEsHJey?T
zSKDu3500xd;a2OYUl0GCV#CcKqK#x_^v{lM2S+O5xQ2rU@b?rI4m!sFoy=_n*fEWo
z1j$gj;!Y%1t%^%r)h4kWsAX%@Ec%Ej$b`XSWJPja?t$-7woUfNsrK!Rf<d==F{9g-
z3@^>oe7Tx<xf<;QLfj1J^0Go8@HJ(^x>n2q9~C)JA*8ooU9RX~44K(}<xIWhAf<5I
zQk8}#pIuua@mf)|8m*==R?~Pn5NSpa>=*u^QC(iC&t;%7^-qU>>EMu`K$lf=RDG4p
zg1^&YL6u-*=qnkNr`}`7#Da*9=B9ihakcpiLpaC5=PjAZwdG9|Zi;*!QcwRzAn>aV
z^h3;7>+_#IT9a^qoM2$wbaRA!+bgvjhR#}{K)hVl>h^f&%Iv8?czf(XcqHQc`q-iR
z*oWPhJi5sYh}ViT0k1_T^w*!`%Hw-I`s`XUfc`6Hbn;I7;%dEkma8;E&5E*ahy8;F
zkMtrV@%svk&JXIpg&0)3WtneecB{2$xA58>KM$c3ybN!cY>j;8j*H2<4UWm;Ht=db
z`}*m*en^Tcfm^H}C|qC;4GmRQRwl<`*i~pLIMeyo07#|@z^3eDEFtG-Ajay@fm6->
z&-E0lEfq`nK`xa!`Ea3vhK{ajtifU`1(@OK(RKjs7T?Y4)xvtsV|oGj)DEx}OnPN4
zq_7w=1NvK&TA8vaI{sp{K7;?`@AYkfFp>b^_|4#pS%C<NZg=-@)_*v~QVT1J<aCUf
z=w;-%(Y!l-Yy@lvas&&N6~}u1kCpQ_q4=<w?67s5avp72fasM}v!*(#ZbMtA?kw?c
zk1w@w<y19)u4gYUq1KQ1JNd#^9F?5e^baAnp;#*0BfA#f^_R|lZ-e8+^R0N{(dOTF
z48QGrx}d<m(bLhC?LeZ(@3~W18Q8Z^)t&(*1gdRLW(Ug+rIkSC20JW+M&-F=#V%W-
zSHEYD|BELt5XjjEcI>wrQRlAwudMSMumHee$ax?W_Ui<~es2vR_*`GofWR{vo9m$(
z0gpo-xd)(XPG08zd$RBy!hG_w9~IGVXjdj&VoKO@=TTkZa}Qv$=7?Z)+=n`T-8(4)
z)>BdV8yhrwyQUs&MGukDDkhg8F$N1BOG+Ff;?vOZCOPR+%?)bB=gIAtqoJ=SP#&kG
z<JO5&)k3f55e<}dL9WZr>~W@7&#cGIf@ZX`KY&jFrDjaT0@{=@FQCT34;2FT2`Av!
zqpu)|*THzur&mEY43p_YBVg-b;rI+!y6W~@;U19l#C9jMGn<?mX1Evz&}kXKMqRHc
z^BsWAf(`Pjv{y-iO(J<!{#x#R0fF#EP^ydtaY9!}sk!pEUUPL@Q|H5&H=F&k-m-Cp
z+iYE4CP&$)jY|BZrt~pQ7M73dDNDe*jY6kAgC(qTl%C8`Zp&A*cKX^tjorGg`|-;l
zqa(4q%BbEw>ryOH3{~tZ&+PA5Dx@qV)66jbv1<57+2@rBV1)|Z_g!-O)4g}jH(VUL
z2XA*8qQ2Y!N#3lxx?x{)Adtgm)cb}P;1Y5X^{l+MjtIY8p!uh=0F@id!1k{g06%p9
zd(~Py@7BE6b}RwjfF!cODf}`eG{zx2&qt5?6}e+14*Cxcw7UV|5?0Plexw7n52JX7
zJ`?~<JUZd--^<xQO4VYwa^u*l&;Bj655-?=_$oFjwd+5Zj2m$1mur5fQ#i6?@a9-k
z06aeQg}X5fgjA{w%c`%>O-c~`qVO+mthjaZkHxr4GNC%zTs(#;N*pT%`HeM8>Rz!v
zgYi_+spH?4qf6zQMmf!NEsBq4Kcjuu>8_7SV_%G=0?xpglYgb{PmB>yIWd5L4-IMg
z(XmAA&<oQ&3?%R~3_<wELM1++aNsMOWS;;;@Wx|+e)C#okq$8ygM8<HX!tk_NP+ZK
zI!&1Pwmwf3X%|%S0<#@(melyCbBg>EHi#HHNk$p3G)!Iqt{zRjCK-+U@UaIV@pZqP
zu6D@%dwJ%J{znQll-M5uacQQOb-xb#0R!)oou+?-nAQBi|KopRnVoJ124|~VFjw8c
zDIWt0Xc$QNTxe60-Spg7UDt|VTV=4^;$kt0E#XeyB8>{}WjS~1O4QcYnGTP>7DQA6
z;XIW|CWD|bG_uo+LEE6Gd2XAOxf~Ff8jReE*SH0+nr`mc6Mn46%Q3eV?fW@QT>~_*
zWx%~zNg41V-klEXj$+Mz5%Q{K6S&)F9_o%Z>vCLT@rUM7(lY8XB0)jIhj*fv7gJru
zcV(J8G8WopYJSgzB{2;0K;NV1RmjUTAYpyw{iYv3p0x4alnrp{USL3CHcb-`lW`_J
zYAw~r{c$y|>7vKD8Tl5k<SI@KT<q%PGLhzsq{OXuLGaip=z(bWT}Bri1Q(AjmosTC
z+b1eDKM|R-jJ6kr33dR>lRw_V0@xg%%qNa@LHO@}p(+5lH{@+R$@RTWcOQNL;%iR@
z22VGJzz)q5KwW<Z=trtG27*0#c)NtrAhUw6{r~{UNBeHoJvVn;=<hz2Cs3BM0}$}h
zxjvatoD<o1{>cq{t)EfTA^_+Fd7Mx{Y5NQ~jMh-Bfdz!U5|n}-2pbrN#ta5iL4Gxc
zHutM-6ufx`T4G)VP7(W1vK+|z!wL9+n_kzW7~n&S=J?y#BO2V}6CbSO2PI7lXaP!h
z*8M73uJ)$xj@~8x91i`+gv}^36pI8?0=SY$xzoI7DT!ZcGwgrCl8i}1ON64x)_|yS
zHVkz3uuw_GBmmr(A1!D{1FuQjGx#8o?MCnrHkmE1!Q8*|=Zzdkdu@MAy`+=>XuSmD
zZ<d4@VWtGG<m=qIv(1IBr^R2Mh4%A;i>&5K4nE&l-TkW1_Tjck(k)b*Lwc(MpX9Eh
z@ouSlv^7@6jGCX@c~`y$UnFo2s69gYM$^$P8up&1H=b4%AZDn4OUgXk(d&(9i+CJ)
z(!f0!Sl1lMBlzC&TdhJ(@pP^0Seh7GB}zoFyU3{S_E?UM+b+3`p|;k>T>n-s^m11n
z9Y$#~d(?^|@F(6b@vdLI;O%(~_xSF}S1Q58s3C0vYyH~j0Ebb!QV#_|WUiQ9mY%20
z^fiW>PF0{&>KGn`Z&wKi7~QVM`NwH-ks5Cxb*<-l>eAD_8TlarLCJ5zh>c;z$c>Z-
z=mP@YIQ#DBhsy$g{x7%dDj>1E{pH^HQ8mRtLIhnxdD`gA<m^`mtAs+#kX;Pt(67X_
z9zySE!j8JTI1^E97|D!lR5|^_+lSmU{nJagwDjy{z6S3}KXZ%SE?xh-iLTPMqcroz
z^EENi>ADuFUh)CUt4(%7QfjGnn$Z%&B*gqGX>#y8ILJjK#wD;L-{sSp09<lBvHyTl
znGaxi$Vsbhep``(&k$@tLT|MS<N{6jsrD70<@C!q{3F03(C?s4G?Ie^O+SfivARJW
zrI$)n3Z;J5n~mu~VLg=b9%kP6UmH5~!8-xI$9IAIU{Ap8;F^3v&Xds%sWy5`t;PIm
zZIapm7OA^4Y|ldm<|up;m=^n21i8v#;*$Og?<wnY@gC{r$@o-}KuGR2wMqO{;A;4Y
zx3kpfh9Vr-s-;OTPz(h%?|IaB70-BwZ+u`u14*)q6|Eesb5qHxYnL!*T-#Px)+d@{
z*BKf4FvwQsTjDYzyEd2hxMNG_{_no%xaewRKb>m+k$hI5K5>%Da0J!sG)m==Dq&U(
z7Mb@4Y!vQ0gh#-U;4TJ>#NPi#-Y>!E(wMDf&C6!kVYAr@Skt2bUW%mQsA(TC(>W3H
zLyaIxSEG}YDZuc09w(goT`EbME-bp>AE5z_$|7H2d9Rus7Lh=fItLgxUf_b`kkITv
zTFkWq=sw4*eh#jDKt~fvaNN*)7M7Z*HzI$ob|<dXJ4kX&6!el2Zw7!n@ljALP)+gy
zkZT<mVS?h8^+Ub-(8MYv9uZ|rPPA;$vC%8{8So!(DMQW)fj6HcG^@G-WzrMr0P$KX
z6fs;02OzHu2c1+i3fo1efWfN-Mo6uy<o(5w&szj&AY^K*S-{QrGw#jrEXs|=l=~_u
z(0Rm(i8-qV*3yA!wwH~~yonV~D+nh~PSpe{IH_8bp$L5`)Cl{DB%GSv{>8AOgiX`$
z`DW?#`0cKlNl_54a2h{0!uXP?O-)T>`{OwLIO}*gu>agJckX0)yMySAc!e89#F#kV
z;n!_K+V`B~t;JloQGWo(zvJ028-0Vzz@VYloMhg8ja^Zrm8N12^jYlW^}--w3sw@0
z2m7GqXb%WNIqHak==o|DnqP(WB6!z3iCHWCyZ5^9U2HjRe(p6IaqVusVtw}v?yiaO
z5}R-Xs^>eC?V3QeOp|G9#$yAOLBa@=Sob8g%8D6qF)BV=H*GecG+8W>&U#mRP;2R`
zywti$=uFj01;ZcVSZPaZi<3KS37Jlq6Vm1js4v;)R=2z7z#n!Tlw@uLEmu`UTZflw
z?md+p@Bh6hK16mRKWuF*vbBnby_h7%^H{R_4}Lk=-Tj*V`LMt-bArlqxytOipS9jO
zweRN=jE%}Bcwge%w<4@BGI(I(2eCs0Y}MSw1jX8XfFrW3ztuiO7$Wob6;5&5&elZX
z8jnbqcwlhYzFD2?(2q#5S|d;-|3!vodX#P6m$nEtL{3iokPBP2BDAB1S(-xM5^9m1
zAU4|DKE?N^jn6WeZHXe=NZh)wj4{#1S|3b@MKTLtcG*4z-+GxO>b|75fp}oDUp>RU
zp1KimAYquE+xT_%{yHf?u~n`H*-wUsKb*k#+;?$sSpQSWQ1K^Y%A&;f4yomKNa*Y1
z|Ma^Qdl8nuOcn3x<7=hKZ23{rqY%(X7Ea<hQh|#ZPPWcD*;hOhDpZ73Jpd!%g5k8(
z7w&a4{x%#->r&N5Jo)0>x4~UJuNvdLN1-WUq1@7JS`28Pi_k{siNv6i3b{XF_m`k-
z>JLqdjB=@1u}RdJe23zFGd}HD^N-qPD=l37UnVD|%d<OLNcGF@+H^F#FzPD%pNDBg
z`<%U50rrz;>#^t*5vEKEvlX#&Y%tx{fnUQR^S<BXw#uL0)vF#0oF{!k^@LW+ijJD<
zIOZ<4VT#N0WOkEJF6>CDac<|7Ojo(>+F@y$6iqxG9M%|B@^T$Mx-#MhJ(K6@j{r{a
zO^evVO>%pZqnp`%XL+%E>;2hiUdKcBWzr1lch#YVKG(xd{ZKr+@YgCY%@mfz$LnZb
zHp~aljN4DSI?eiQLcwGY1Pp5nB-Qm1<WhRqJNJ#&qV86nifvLvPl7`<m~~;_;@Y1G
zi!Pj#rzGE3w&F%Ud8pv}s<arU$}>$R?>J|t>O3$+msPL7-bVlg+3`?%9cuSJFp}~D
zzUPA$`hU;{^l|^;`-Q){qw$Zf$2M}}@k-iwe9zB2`OZT67&-=OeJ!A|%8DuaZQd-k
zJlR$mL7Kh^jNu+uf$J7BQj{)^qq%8LzHmLx!_#RbcA;{%cme*C;E=Hv<e%Xp886eS
z-=;nM0p?`&N$WsfUH=3^RXoim%=O!oo_2Sr`m|~Su5HB9EG^a%hrDHW(oa`r(N@;{
z=VASQV8-F36Q9Q!c!D?+A!?aA-70<$bw=w)k)^@hBOGw&kb`@R>cY~9(dj_b-=3Ff
zS$u?QWyRq6Q(U(zIoLbLvd6tGy=mrTWIg&uF$X{eb~w#kdd8v<dexE}z=$;rj4dhY
z#HX`yC0DETwNIam7)csjPP-lEDRw(=5QQu<N4&jwYZNF6-@<UXNsPcqZ%MMexfmbL
z&<n78KgPYu_=GAKEoDhPvyIt0r56St$5ZC8`LMUaX%MO0cFK17=Sjs*_zIfUtSHW_
z+`{lQ;6^hh8LXHn$YGH0YHnjy_?W%7{~UUBC227rQC=oL)>%GYId^G7V8$#l7gLPe
zKZst%Wf!T1b6}HD-Vwim<BrcUGVb0g7=~pR4?Ki&8rJe0rA!suzV0erny46ci_1;<
zJOc1TH9A!srAfYttU*0r07KaN2wa5{mK(XVHy{OmA`wU`&=nO3mtos@^(7qXk9Fh`
z{b{{ct7GiWa-YHiR-LT?&3N659*TY4t(C_xO0t({9Bu9mCQ<;+=&%kdSF0=LRnUhI
zSq^*oS+CAJ?XX^~F{F^<)Ef<Wqtay>sd%_I-}$o~(E2aoCW~myW8EILgl2<r?gMa6
zjGuF=>}vi<WWL@ap!y%jJT3$__~iXYY0yb+5HnuL(Gp~Oz6(D|E?0&g^9)@%0dLBD
zsXC08S^hzIq6%e_(Fw?B>#B<D)yb91%FKpL<=VWQ-}%=B#;MMx9;I}4-4e2n1+po`
z*vj<D9=7u4cE0sljOJMLS|%^c?tMjVzuo)Uk0Q5FRcC%B(`j4pauY8y+D_EhD4xco
z?P4UO4N^mOP9(rW<&UcUKpXR&W)_5p^e^3~)5s|OM-5=d+-v{wxkLjD@a+JzFmD)1
zgRmn=et%7`8(^pMg{4KpauXZwy;v)TQQXCF?hKs&eeBkk7XTv)`2`#`POzT|>EH7|
z7-g`Y%_%BcS24%4{7&6r-Qb6XmCz-}#W)0i%Wc(G_WirzJB41SC2bRp*8SssC2sh{
zY#>@%Al|bn1)M)FE^I0yXs2VCQ>(|93-k>U1{)1h&?!v``_~v*g@^q8fZkRHh9ZO?
zLNPOnsC3qI^(Z9YB0T78e0m74T2#s;EcC!h$5_RqL`@^8dFiwwbJr~RUN;!EWx*Ls
zY!p4B9rwNL)0z^%aT`wR|A7?@>YJy$W>=z+*g%EFK-$5fUs$CPl&6b=)zeLrcJxQ2
zfhv%PATE%0OnnVObQk}J5gp0~C&mjSPV#;cmSr4r*8^~F?W!UcG;_)F#j55{8GsQV
z+I;xtF#8D@7TwP3N8dxc)>4EaQhUQU)N%4%#lLt5$S9-3H{q^P9F6=ucylyK?C3EZ
z$XruK(+g4F28dFGA(ZDsuBf7KT0U)F{<?=NPz=EjL4no_Trei4^!moLcbv}%LGwEq
zNVUG-^qPsQ{2)eaSnqJRH&mN<=bFsR1*4%K!HOZD)^2PqWqLfu)~b;Fp$=1ye-nbl
zE{1uOY%lKmFmP6Sgw|+i(4G^KL;VRU!eRtfX|`-V+4&olmqu$P)AUK=ssP)8dbTuu
zNI&taSepsyLGhJL$t0C7uc#Q!H!0f|bpw8os;Uf-?Fc7{Ktx~?!b+oQL?WgR6Q=LJ
zx)&OfgEY%Sh%G@4V;vc!*>1r<!y;pC?i~x@#RmF6_x+eu6a}Lgu!FD^R>A*=s&j6v
z;|tez>@;ee#<tPePSe<pZKttq+h}atww*L+Z0nu;&N+X)U!dK4X0m6k^<4LLGs;ZE
zuv#{a=5_Dk7<SHV*V!_UkO~!0NTdv4@_S?Gl8aA0<`){~>JE`PBZnqmx(BL(=uUUo
zP;};-LaQUI8T!Ekj~fFYEyAhCo{Or`O|Ova`_mzJm&WX0P4_-T4yB|iHVC9@P$YhQ
z80&5Lds7!IgL*q!^UkV002_-#1jVB9%7*ESg`xW=zc5KmDDXHSPW*b=0fK+zRbT;0
zX0Z@<nd!k~lTfbKBm(EuCPWa9oW(~+CtS-$uQ#$fP+M6bA9dPNTt9aaiX6cwgUTeU
zCf+{l&$XsW@unrz(z9jl^dK265Ga<mZKZQ+j2Tt7C71YE=t{W^1;O}?T(G^m3(0yw
zsWbC@d;@*xx(GVFba>$=%h3yK2QH|8Zuliatem&}TPzG2MvV!jqFPf*6YmZFX?%33
zKnx-UE%Y{4WyF=i086OucId`gJK=WK(|z5EPK+9XCzsZP>t*U4WY{8y3(IA619mR8
zT9E2?jf!hn)7G*6*Po9<F|hA^UC+lfO;=EAB-<!`s<;{&YV^2a!m|xV$^%+fqvhf4
zWzvfy!SG_LlO(us22&(lzu%=gvvTE2;6nQv*R@wm3uh_WP$VRYKqH}x6v(l`niy%I
zQkBWwSRYdyYB*hkII@U(_reFk#ganNi!g5Z(7?&<F}MNmr*!~CJ?ZhyQ8}bcBkjFq
zG@PEcuHdaz-`)hqA5#fr9c1g2Ze(hh=H0Yk0=s53W+}J-XyP$(=vj?3+<el}4m`U0
zP-UFk(dg0W;Ndkp^AP(Z;UW><#L=K?u)1Ge$-A}JACW-AM&d%z&agEoZF`;Ib79Z0
z;JM*34u0|FqxW}&=ZIh-)EmM0oZ1I)rC#Zx=;j*m{3E!C5YJ12Tr<x`^nDid0g=hN
zF%39au#nF$-1#nr<aKg2+v_B~e_s<l_G+Yf>!Pr5#;kx1gM<sGItu35(e9Dv!FOTN
zVm|CvE4!8z8z<ggWQTtKa8f}`Dbn7Uy~XjBCmD(?sQCCi{>uO;u51p9v7A11qb(8W
zk|K<jKUip+TJVSdT?6?-&gG@GIv@WDo`Z{Q$X)o|1&|VR>4mnyl_I4a&Um>BA@#{i
z5p40d_=9ObJhckXWK)+lK+J$w-T740ODy(Foxws%Fc(T1bk%zif?@b+E2D{6{zv<p
z7$;m4W$c#bqrA@!)FAg1k!PqXWnqTI?Od3G+fnkJ+=Z7~^j-}ytyY)JIo3<cxF*gA
zPp!@m{e!#L>tBZb8DVaQ78`mKly}IVH~s)k|H=qd-wu{J%+&R9hnSQx)&S=YCm^TD
zm%065bE|l~cpf@lTZE_Q)IhF3L(=Pdd(nRdxA!#MIb8B<T_q+e<(P#TB|#kv4$?JI
zh^zjNLG3_8u;dCn`i|Azk^e=3ph0iI0I|RO*Z9ut#}+Zol^QT2B8?R<&`Zu7L?OcN
zhd4+A3yA(*kh$m>5zQXrb{PohNda9@agbe>0J1<Y=AU&j4hr|b-mft^W@`Vr=T$^d
z!K0D8m5cRU3H%M4V(QhfUH9~I=lFc}czIjij5&A5hHRC5bcnTHRPFGE9(>2AR?OWl
zy0}OtqcQG}_o!oS#*KZb%0G`A`qHVWVI<<52sibw;gXW~a;_5~1hrAUipSB%qb)lS
z9ELA&jr>1vchnPDBwYkW*N1JB>}<}oUL}-G0)lC6&3D~~a_O*zrL_5cTz<}MuLLSf
zhKcnNja_t>(1cK*fC2RKl<d@a-WwS%JdKgoo>w^b@e@JBma1^)%TxVWfuF*75cJtF
z1QFk?r6|4R=ZR#hRbuER=EG6LxiVV5?LhiM!Uwpp!4*XxvRFamYTc;u+x=<<c6b@x
zLKz@$#T&QT4J(Skb5R{8)?b8-DfCfO4*i!ffB(;UYb~w)%~Ba@l~HR3Z%Lx#s9Y7j
zwEw!x&9KYEf;ql=iC=Hz)+SjrzH1xzLWdxZb>cjr_@*N5ChWECU1nAfU<qTKlxfz<
z4{!!m#a$0l)AAeVK%j_6M1$QYL}motT?(`9al*i>UpCc{04IqgPp31FY<c)U7l4|b
zTTe&OK-h_?RJ%JWS{}z-9MT{rzxs9FpYE`32&r+fyBzdj&V1imbyk0?NcQ6t<C#r^
z!tPSv4+fm?t0FD8saW8n{xbq6f*peEy30p(OYOdNy+}BoeholA%xD&A7eY<Vl;F&0
z=W^Kx{H%u!<P630&LB_)n#T;@YoOrD*EnSZMXsdJqJgoij)SMipAr2w81wHBO@Ggj
zplEFl2plv}F9L{|-tFId+{^e_cvx)uRw=lg!GUfhluDP}I~?G1Xh;>58j61ddDY%q
z@5YRFM^|b!58*!UwmT%UYY1lvCLMh|+~0Vh9ER~mc*b~q?7?O0KfbBD5Lkq%VljSn
z!vv$<fLq5=lIrRZ9cW`29mA@cyAHq{Jt9mBURympxD{s{KjV~#^K{~UOtK9xMgQRL
z8O^}_3+oYN^A%wP+IZ)y4*tUt_$cKE!p<|~&1q<3-(z2iKZ!+-%N1;6`V(e2bLRt$
z%luRw43kQ|n`t*rw`(vwY{7t^Z#bP0v`A*Toz4q>dHhS~kM~RqL7`@WnZXN44|!6Q
z`6#k)MDRjCL%3~Y^zXP+=od{Gb+vCQHEU)E-5G?9{OTM}=B(EKXx3W$OL+(}Pha=?
zeNPo%EIFA;P<AjZmRlniZYoEFC7c$*o=bq|A5wxP8Q+8JJEnNk?V3r;uMu1GC+sm4
z#wY)Q^^>2e&owTJSpr&5Az);%BgjJrZXRvE|M6N9K0bQDr(TOfEt=gvvt7m~OE<vt
zDMh+vXzw)j*Qy}SsIF-VN1Svt1H+@-hRibz0SssT5IkIc=EeFDV(7hn^z)x!_FST!
z?^Z$0a$LHA5r;M5FSc7W(HW%Xrdd^!ln<TTpXl0>=ypMP5VMkGb{3Tgq@(UHTEjz}
z@=-(K><VA*AZifx^?-NADzQKYN-q^yZFJcmv((3te7gW_co*#5mY|yJ;vtgh{F2M}
zqesvD{X#mJtt(`DFia0ybsotwS%(1~)zSA2_Dn>1D=69Si^P_QmVg#Uk1qyZH#dw9
zKG`fDD>dhe>^BO{y+enHo&MZb0&56N;zC{%FjeVX5nMxC#b@J^URgW%H%is~GtW3W
zn0G--m^yrB!$T~Xn_Uu4a~KDD2lo;pA7(T+wG@jutow|Q<7+xZe=o$%IhN!@a#KK~
z37Iew`ks#<Be&nSO|qO8ajGJ&ABYzuTWUscIEic@8(<<#DW+6xBT9-}y<QsU2VOGy
zD`S7MP=o#cOT|D#Bdf!wfu40mT4{u%RUcZ|y&7CghcTge@SYc~-xfbGO3?mL+))|l
z*{0n89ja>Wa8Hboj1Knou_Lkj$L0<r%g9UUaB*lt@Br28Z`+@_TGubk9=x3%DZ$$i
z6HeqV<Yn7V+0i2rBVpflTZv{N$0!id2)?$T)Av7$Z^%S3F3%$cHf8?S>wV(|{unT3
zbv-@AD0TCoNJjTGT)RwdI2RM(8+B1mrK>%%gI@L?_TvYO|MG?;cO70R0^JEQr!bc<
zf@ovxDa~8CRu2Kw5rw|XxDe*E&xh$&Su^$V96F#q$!jSW<HA|Jsce<dft9wPxveif
zwVvN*Hs99IFIdE(sM@WkET36-fRfeTD#$goc1#^iBj8TRT%3^p;`}?=A>kCva@vXT
zfxqePo#}EMrSzk~*<oTR31V!U<pjL4_zHRIdzEdNk_j!S?`y0r7|CLW+SiJ$nrqVE
z4BBh4FtI!}c%cc_W{AFNB=Sy!<!)SvS#u0|HPl~zpuKM#Mmq9lpxm4JU+je^aID^4
zw(r0zJicu|i14YZdKa)OdX=BdoIg2>woTocKKQx)l|+MQ_?6Uw3pNr4LXQe_)GSgh
z%k16*o!V{-1?PgSMTv~uz9v_DKe3wl9j02;iW)Y<DbP~*-d&<tCbgzkX_TG%ME_1u
zRRVjO$w<9MO}oxr7Q~nQ+3+_Kb^JtnH;cJ8BSabk*eT9n)GsyY+dU`D;qVs}9PI#I
zXK?*CC1>TnEgVb|n?h-bfbn`qvo-Q(Q0vfy1?CQY?op9zwbeb*ZX1HEcDnNHBsRk?
zq-A31XA4WgR1;BsM_Bqo3u{A{8mtwZJ(U3-g&IvYD;mo`Ce>Jl_+YXM5|qDKRHi^@
z<3%;^r8`h+FmQn(XuKFmIA~9z<wH1Q_Y*k)C-;%iR_}m<wzMxLaBgjvW`4-eoho17
zSu1kL2qg-gg%eE&pmcAPsGjUC=a#TDx(tuR^DoTs%oayl`D&!sn3lK8ic3&3C}iI&
zn8hZM40NqiWAn--BcP)S8Cl$=NmbL8LTR#LY^zlLN4F4mSDL<W_q2+UDk6#DuIJil
zV_NKgPG?z!2x;gPo&V{ME*i7$r8}{4S!t!S)wRMCQPDg~jb8s!%+>9X2tB8kuY|Tr
ztc;r#{YbpNV_TzXZ`C;K{6BK$e~FywuPCNCNdebFR1A-Tk&tH#6O`i2&lz-u()@At
zi(PTo4e?|h8}G7qr{zBYWM+C8rkbt#f7(MvLO^iI0H;HPV2&9KEZuEP#(1N-?4uy_
zcci(gOp#G`iExBB-x8x-kF#iMM$2MObwUR#e2g&*fCW-Uah3XEVp8S)KS5z&Ma3?$
zU5-g2oWl;oib5oa6mQlPok0JgpdG>4TqeTY$fKySXH9pM`6cPvN%7W@kvBd^^cWg-
z!MK|4^K#A8fdRvByA2G}!vCnX&u=H`CoqSM`c+0BMZ>TRQV=|u<vYGQtWAmQ9}SFK
zQX@GUrvHeePhLsk3+RS)JnzHCcVi3nQN(PvWn_t;5NQ!HF<En(v%j_T{Dc-ddgj&$
z3YIt#Ywh&wKl|TZ;x3Phvvu;t3mTu${v;TjuLQ*^3VtR;zu4_=Q^-|Y4m(dD=xU-K
z2|Gw>s;T{x<F=CXv^Mw51U&x{-?$s?;g#eXzOkCdd=9?vHWy+73aZQ1W;F^Vlg@+s
zbLGTFLIp|;3(p6blxH;%Rs3B~_ACn;MuE?ph6hGPN7~)Y2MB=ekl>1+@tv3TJ`asr
z{liPS_c_k}3`jkI=r3NMCJ=3Ed@^9Qs?+=o9H60!Md49}V=^RhpEnE30VNg`q^ySA
zrfSu6>1^}#!_Rjog+PwYevGc^6ZqB*DZm2D@V5m7+auTeV@ANJ_YaSZ72qwu2Fx4_
znf*Yq(TBPh0M#tIO(7!;GJJ@7sj;i8U{obJI`y4y)`p+17*$)&H3q}NJ4Lxpz#9hh
z&7`zlJH;0|j*P`Ev^9mm!RE#pq1U*uF|69-fI0aIq_*1IP|_)psKSrTfX~e;AL-6J
zB1@PN<&yXL9GnA(qj5ySL7(|CQo0Fz)9AmS{(Effkt+#OMNJ;Qk=v~FbehjWKOTU=
zjLg$oNVCC|Yqu?CHje)p{dn;HI84BZt^NGmcF>icuUA8pdBJQ|jY=9i<Yi$?!<r6V
zNk2?H^ZX9rsE`u5UIc%OHF~hWIB?lqV%OR<r4j6^tg$b$dh7kHF*GsyE>>7%M5@gf
zSwWNf;8W0K;=<>|My*|DGti1M#7U)jA8TmsR0xMW!t{&coAYAMZ*dT#zGWlD{kB5>
zvC%x>G+~9TDgSiUEa!Ay(ya{2`HFs)i;#t-&$abC{*y6VU%-5^1EBS|24>u2z%-n^
z3+O2_0t$*yi0Bv?q93d2)T#xFWKxNNY6W?tPM~~4sd3B8{b;db*Kk(7&Y0`L7YH|&
zF(cmP<=QM&W}X5UMmgom>11&vg~>oG3#g9#lzrnl)Vu(r$w18Fx^uub@>5%`^Klpc
zuvK8AnRR=x+ZviiwK4%1ipsY-9w`*kWn99ET}}U+$%hnk0tl^;lvlveZ2w5FUQ4?B
zwk^7P4;;9|RAQJaqpbT`30k}ZJuS^O#Ag{?4u(J!Bfg?>brfB3!xObatwu<9T%}5f
zZq2eTa_nL)hEi@XoXr@GT*lU5w$t;a#Nhnoz!R9{Ujv_xyW$$ql<x+pZWV)n;q1Q4
zJhWcYc1i5?r1>%|WKs8tZX{{$N~sE-)`H7rY>siVY-SCA0Az=xib@iI8RE~@<PRQU
zu~CBYzdz=GLWcIhStjh11TEwM;87<5(y{NqIg+Z{uftv{7VCT9ml)#9#5TzT&AEJ}
zq{FGhcQL0Plr1nuH}XH|+VI|D_DaPVSN@bVi*$aw$#1`2r1)M{l_2+RZR}_8Imyde
zLwNo8gyj8!>Q-kQx<aFUxuNq%WF>T}LTTGwBIcwj(GEoeY5DwX(RL}mi1g1Tn6HVL
zW3`Bo*E4=I*P+z7i+^f~`Ue@M%#Ir$7uFf}Q>>!m<{B%#Ayp*JQX}5%;@WSYgKJIg
zI_s^{DS`=_i}83I8O%1Qn=0EjX~!&7guqA;Zro-iDMmio1>o{j8Q60>{*yPlJe)|U
z==gmX_t~vN5BOB59{^VXbklU*a9@C4sz7=;{2VYD-=EcST2PKc5qI?gDoos0fXV2F
z$-VMW;31HYao(bkP1~yv1xN}PGvmn&sv$gwi0w@@$ujQ&bXCrYT{-TvJuL4{6xY>8
ztCLaq_k28vFIe*T(naT8Z)Y}%XL~P~JAI!eV2&%+olnMxIHDUh|A4Zn<gK@Jmuo=i
zE`p9478ZtggIk7oz9ZMe9tT7F^z;<P^m{jyf^<0vs0HGDAki8?w)Xg?`&%!gtc~A#
zofEVepZ7P6!5bhp7+FyLMF+G@6kF)zaVm2dH-pR+0n|myb>GnX=b?|De<RV?+oQQ5
zSQQvn9;9ZiZ?4<E)*hbO*y#YEx491hD*2ggz9wILkbILnXMS9NJam0H?eDDG15Cv5
zXUSH5W-W3=brCij9@wjMDh28a3T$!#;AG)G0S^mEn<a_=H8bGx=k{NFCf=euznj4-
zbgg~^bH;w5au}`R+4ivL-~DR*Hr19anO}=GlnESLxI_d|ScSbka8wJN)TTx~<}HUi
zT-2O0F*F~KxMp*$M$By7SB-}Cio$}G3AC<KwZC8GkYE-noaY=Vx{^E#4Ub@8(I;oV
zy*`ubRPHlEn1=|w2PN7+4W}1Yn^wS8nkgvMR!T{ZQ&=p~C1QGeh~1yG<~!Ofw2Z#B
zJ}Z>N`91JINo@j5bbc~zJZfZdfC@%=Y;g8((N4U#d@Yx!mf4!O1CO8D4Y@4|bEiru
zIxmR%kp4l=Tp81iwT$_9NEEDX$m7%y;9vAIeRQKfDV1V6N6cFS1l$M)4NENq-Bq9o
z=*;@#D%v}ty`87{D5%Ecqyiv~nzoWKbzFN)syv2J?EpXs9zc9(7;{Y*^;YvyWOe|i
z^BcHa_N7Y@7d<7V?fWU?KgPnV*-v5!BEb9gFM&NEu(wV^ln(H=V|nD_xdQZS+Yk^C
zje+o(aI4l>8>SE>vYIHn$ke@mf97pmthdw|gpCBVV}kBncz?JjnxpnmWdXm3d690_
z`VbQ3lo50cj4*H>M@L6;Al*0t``bUfpA|(c0NV8F2L)jS4%_ICe<8G<9zeACoh(Lj
zY*M08$T|LvV!Tf7dL>&%c+r}IxlFeJez~>4z#JgJs*AT?4>)RpLWfa6G;+v5b~|Zf
zJ2|?d;Z1Rr+EnQA{Q>uAqchsFX}@wd*XoC|*dAvx$*+IAh+}U<1s)%-Wf%MV9DE7q
z8{fz1g`upB*&dz8e_TIvXOkMOp_wabG#Xmk)g~Nz(-B2V5C3xxgM(R<`W-BMx1L?G
zP|C=XyGvU`B(DjPmHM8A2KL?Jbhf1nhn{!7f|A#7fvLbcBAf0xw{byq;$veO_IeNA
zMvA)u@3eF5%j|a<B1XGwR=3?14&uthpYRR1C@7}RLVn_)R@4VxSYx)mADHkx>L!`!
zUSno@Gdx{4licU{y<n@Bv><cEKU~i7*ExYT3eC-~zRw-VCz`dv?b`!G1%sWe?VP%u
z2KJL(1$teqcLbi4F#^IWUp)XN;3`rKq5;)CHa?yNuhP#Qd7Ss0+VoGZtCpWcut1<~
zU=o3aB&RP~%5~n9=``@$1_-iuMj(SQdecByBHB>sb!Yx%xu4r{_J|mC2tL9o)W7J~
z1CZf&4yJPYtmSXRitU^O+{QD5>r01Qddox*NDxqrAS^!^lVQqWD!2O`f))rfkQswY
zeFAAv0=O~2=zOaqgijVLHdt;p-7h48W^8Tk9i!!kXF;E>L%>qC_Z8x5USRdpRVAeW
z5j=$NJ`NkFY~6oT7$_lU9ECUcL7CdAA2hCW9Uq}6n442A&{}^??*70ffB(-t<aR+V
z;PjM-2)Tv{1`@;p-obzt=Vk8e%9H9Ejej@ZrjQ)hIpa#6g>vNn8i7}?O5EU)czGj*
z#}Dl&4M(gT!O=4|cD)h}4C=*D<RzA}d#;^YI)9=VLa-YZ2Kn4}|69v7>CN=F7DuxU
zQK*My<yNQDVhbe;6YWdUO1h1Vrt1gu7e2FN8En>^Y{#v<It>=7{R9J~#LW)&mQ98C
z<ov;Zg2`kJ6{RbOQ>IZc-T!)GYBnek9P<KxFs-VmSkS`2Ae`z11QB?<t$%O7;533x
z<x6o2q2pUI{tAaAB_fS4*>;iol*2s@+>|3MT~+x2@1AcN%d4#adiz~q@OFu?lwl@Z
zNQD%S?FKR-Bj11}%1In5KZ=~`L{FH<6_EZ>3y_j3l>?bMHz?%X4Iv3oLs_3#{s|@2
zcTAPSfgyJlrC!hUaQEc0OZ{0C^eIa=I5neqKg{wwyT*Xh$Lb9|oP^KoaqsMC?$oRC
z>wcM6{DDYbH+2^RlGFoF-`qXWJ714;0JA<y8J<!k5>g@CbTs{|^`JiaHIYSov9^{t
zwaGDV4s3GFx^PKD!r8bogX@u1T}KV2DMsR(Q*6=`qBKPpD8Ux`pY@Ck+M~I;w%5ZJ
z4%W->q-f-C#D_@y>IV#?$#&EN<r)oQVPWC(rJCh-is(011@>`w8a1QrWc%n2iX~?~
zvw44ETXihc%Cy?tb*Kut-kgYFxoG6a5E3B_ki#Io9=s-_V_A=^nWhW^n>b4c3DVeC
ztPEc&km4>d-U+2uC*z*R<_e`>HfPDpB5%(MgO`rw%xJ>l3=e956_n!#-m1_!7&T=!
zCW~izEl`1by}FfVxG&n3-qIyp49X^;)4<Pfv`n`UI-lennqV(r`5IG@=7f?tgE;W5
z8l}vOjA(*iVg}uBd3$s11fcB!iz_F30h0_nF)h*GJ3a#c^!kEYcp$QOG5l0GGp0cd
zK-~ZUBGZ`b)yva+NZ-#G7aqQn!TYg~fW0iPTQlSlY*k9viIw;t1}33a{jtN<ZvYd_
z@Btx-9DpL{K^6X1k=b(sM2Hz}>Zc;uRM7o8Lfd>$Zl}o3g|gahgOIW1U~tuHdq<$z
z(I|Qzu?M1kqLp{m&*%Rfz$4y(J&KcD+=m#2w~*rKW%l2}EY*tq2{wOSHQDD1Vji|8
zrY+a}h0<R>^iv92R#Cp5er6Y!^{u)ucr@FlwJi=-z0SGg^~Co>$^=mhzkY+g_HYuX
zzUtx+B_YUOx<i<uV`r1S3d}w6jvsEip_0~^qzzHi`R<%3K3sxce}7K1BwO!Y(ji~T
zykr9(siJ$w?yx%p>1?A;McWvDCiDn5+Q(sSAS;yK)%8(a73kQFvb@Ax3bsCa&<VX6
z3+ut3PScMLmqd$q=O2ZBb@FZ@&{?jMt4$U({&!7XH&Nh5FQA?S5CN;v0(EGpx&^uz
zE157H+2iBm6Y_iF5Ob3SIH9TiWDr5&-NslvT3pj!na-<bm9$_|e~IdMnhQTcH@UvZ
zbd;X?=3o)~HeX4BzY%3f?j;83-E_Vli0A>o4^mB9zt2RVe$~buL<_=t0DE@)oSV`>
zATVprFOubkf+?=YFFys}5ai?}@ohe%eyckk6=m!;hrrm(6Z)3rTD1Hno&^PY+7+LK
zx32k*w!;J)y&858f(z?tUnFd<pO3me=kzq{x{OR1ri`cdy$(94&D(D3m5}Db*DFt9
z<S64B!%97I1_*%w^Kwz1vRKpYZeM1X^L2@)(vHUDEZ@sx+Kp|(s<v{C4o3aw%5Tr3
zs^LS@UCtfEo6(N3CLF7kdRM2{kc}|IjI>wgVcuzv;a@SVS2IJMdjt+FkZ(*##M_K@
zyn57`O?A5@WGn3vm&jM6El|NU<Uyhv!|^cW2NT?qw-~rX^&odZc&bEl;s6X*;rGM&
z+4@U0UVe)HYY3u2ou4UdevVoZ-8A7oCzWCa3_mSHqL<Zdp_PhqIM`dBw;DmHULRfT
zb4Ov2MROlhSI#z(7Y(W}gPlr$6H`_7xf{`Uh2Q0k{S!>du}RlG5o$aM4q4qW$<kUv
zM59BZsAUwuxL&FqE5HfID9+`NNOM#pO%4Icy2VH^F`SF$Y7`BFRS(=06j_#jWUO}*
zRq*{!l}q(2qk>-UE-oIN-M0?Nm-@&-M!K04vhsMm!GEDX`-p+)*4674Rh8n{=E@}u
zY3vS-w3;<7YiFa%kj&TVOw0;2_+HP9D|XY3cMAc=iq>_gsJd{eSoDA%GbK&X<HRmr
zk2qB96vN+Q;-~nx{kTWEpQ)UF2;5Fk+8*t=Vcy<A2xTLRk1-Va$xiQDMv{>qZSyk%
zst@=6cHymBL?Qnj=jnGdfhF|T8}BY?j;?mi`BS$PmYQD$35siVNk4I82pI^DZ?KPL
z3TYKt<O0TM!9<q&7aNO5Pg;CkI4|TR<W;e6SsvIPRbPrl1|zOLR)D!fR<l@pkO@LC
zz}f7otI-nCvg`aaw1X%nd}3Z<f-ZuJ#O==uuO7$?Ax}SfPu=C~*%hln5x@r7sZ_z$
z|9KB4?#gwF9OUD}hJq>j73x#~BUE1Oa@{~Xtp1Ns0nfVz#~lei?*;Bk;SfG^k)eP2
zlD3rJe5*3WO6QHhll|ght@PoUe^Z`UrS($vvGbFDRN%_>RHw<;e4w3ku+#mV_QU;#
zgi5W0U4PaZC%bxxo?EOPvmNH|P{#HlWPmM}Cgmc%y(PV`7(Nsh4O)$m8pYGuvm3VP
zS8F|(6G|5~3SD})FIYC~$_7@PD3&Hp+0tV|_FK(tv&F(e88aNJ&kg&xr*M9;#vJ^N
z#w3u=3xAtFAvEqnFkDW;Q`a-*%JSGrou2P<zvSVr%7xV0YpjFO?lORlk~QBUz8{0D
zJjWD!rEpevAQs-GF&bjENQl62j@<RLEV+CDVSn`jA!oI5-ed0DP8--WlMqNQ?emW{
zI}WnO;D#UzN*UzqM?8xN*CdppcD7_EPKEKj3)R|Ae_jU0GT5_CBli}ESH)RH7s>P{
zM|}aAWhD)6Ws!){Y(h6`%6Uwb_<Ey<%39r>xDe&N1~amk3!|i)kZsc6KiMFtq|Jj6
z{v;8fCqvl>CBR0Ob$xFMQi4ZgsenUMJ})bimuOg2uDs%XO*3~YE!cs-JMybrxs!O3
zEWk;MkvmN<MADJ35KrVQDKe~UED0AjO1f;--<abJLf&73{8?r0CA57=rJ^Z;wwbJ}
zNwQ%=o2Xt^PO^M2Vs~vNV>nJ6*}@Rqd905)>)-fV7(=g9VkSgA9^j~eWQHfY&-+2t
zzM)ktJ36m|?UJpND(!-F8c{0l)8fY@PhR-k&!kYYxVVE)1$&dM5nQ^i07ZUV<^5JR
z%GSxaRRU#OcXFkb(!|d1cv@x>=*7P~IIYdkTYTY-nEANq_WZ2<&yT_Se-KW$?6-kJ
z>=suK9?>ABP*epvD<ch~jBh&cU~gdjMqf6_l<Au<M(bNHs6m1On02Z{qfy8_(&<6@
z?|FF0<D`w-9tVFw6z@&kv>wG<?&wKC9Rrdl|GkOlJBe|H{ZH)07&!98F7$=#wwK#b
zXN!o`m&#S=2ki}|4Hsq9<_qfewkIX*@tDEz04xr#u^6};dc-bmBgPG0ujFZ9&wX}&
z^u_Sp{jrOCu;L4%I~)J`Qccx<u=N5@aeEt&<I@t2@CjQEQHIjfR|TSy>)MNa=G$|e
z8>qja_^}7jk!|pQ7KRIrGQkMY*{R2F2wr&`0@~cdO;k7%Y_a?aupp@(%gbz$6?R58
z?k&K6@#yq;yjm+A{ahhOhto=TTr)`?k4BjyHKJ0yIFApu%VC(WP;pD{Y-PHhV%}HO
zDSzTwo-}+Z2g~XctVm`v*KjN0aqLY$3_O2(8knx)Zw36#eLB2~*$ZJyM;E_#58~!F
z+bD0>Z;{NE#M@H`Wm9mzn?0f{dwq;UkHcq2;b}j$MK$qAV`BRzG?Kc~mCC8vp``tF
zy~**di1M>0D!5BK)9tjaV14kov}gQwCtAv);B`G<aD3}K*okk$bh+h*Gzb>3gnok0
z8I(*w%Mm*@f6~y%_p7`yhP<7jr%|kRYS|r)`@!D|$tx-$*<xqkfi9LQob@09bVikc
zTi&kl*jszNRA>2-4tV^c7Rk?Z6aN~0B>*$UB69BUT~$^Ta*w!YiDLt$5xtylQ*+v;
z#{Ltb=GGjn*=<qB?{p!}y3DBk7Ua$U&+%xbG+blO(S@afnuq<!yVh~RGTHdOb$GSq
z^7r)X&3c}C5kXc+#pRUyBmFPa2)ipH0&7OKeOpoA5OnvCw+BzApIc9K7ac>$PLgJH
z1Rl?kwAzh$EG6f*^O8|KrUcm2UC}41^I+_ox|-^&GnO<SBV@@+_>)+j;dmoQ7LWAy
zr$vJ{nD9ursyVq73~7%5YPi`FalJM`2J6DIu0wd~Tm4!CuWl8})NUTkel1jXcfDBP
zbln$*x>78yGOo}6w1-wKJM;|7WTP9||9xC7NE2qd7VUL>akRbey?RWC(=DPp;nD2`
za9AS#z_{Y>0^c379q6rK^N_tvRm!pB{LD0=xIiMb9_xSeYcI^cc;^Tp!)+p^%wp<&
z8%h4w$=4#JuuYI&CeS>%5C9jpWR52g8==iA=RCbJebA&u3GcM2n{>2pEPYhfh9|4>
zU43Is<;aGyf?T;j^HA0Qd5msJhxFpu4K_T9wh(r<MbVw>)ThHR`0T~vZ8HnQB{qo>
z?DJe-x72vc>$U7Jkq6})7xYy@6Kjxw0a16|=h4@oW4J|mBU|!sJCp8;V5=yL-QB--
zem(lotolg3um|yHNxVD9At%D5y;kFGS)#23XRE7mcL0~8bRZKzyD|5g!WS$tHg2h&
z6I@pK$3RL>EGc940>(`eUaf}ym$7*_ksoB|Kn;v>oEmKdg2)CNZArqFpTBu2jf)NZ
zb2mk8$y_by+iz)tjr(c|4tEK~AW|9{#o9Y`h8^bPsqmh_xH{pG*j^<3QfthXoH!U&
zfx$QyTU)yx^o6#_o{(B1(P|$3aAbw!u-XKIwL-<D2-8BrDe16MbW|+vRct(<Ek24{
zit~@r7_4GHOSL4JwMwZA!Vi@ePpjz$yE>7d^xNiAv|u&L?aMhN3hN0jk#K|F@}z$7
zOw6&vG;V%xq9eU^Fas=NvVuLkP1aQ-JiV@6tT$-i9_)EDJ$+?nboXzdv#C5+EkV8M
z%fgpN&=h|qzbZSdzy8dDu3Vs)$*ORFh}p{a^qT~YFetJA!uXVP9`ggpNkQU8M#d6N
z&m#-W;jF_o{JCRLeg^md035|!<XuI4!?eTg%alZ>6JX%43p0j_(5t+xXR$Clpc7PS
ziy{)=eiSs9D>*N9?_0NI8r|kGUH?OhwsfAg|E*Evc+nh**LX+N`p+eqjasdK#1dOk
zzGR|OkMAE_Gl^=!4S(-Ts~~ui$#@#Yh4*s<?L6I&vC&Z{T%`>bCC$-PST!v7hjpPS
z`Ea2NosDmE$T~qEQSw4O)4;K|A~t+46dc=%=$8YuGp^pCsSD4SP&Qwp0g5iZKO%ba
z95m<<B=9cBS!(LMR;128KE@2Eu6k_iVLsENEK-&}O(mjewy!s`2tx881^UOiv{(+T
zS85P`U^QDAzQW)&rpVk_Ur~|kyJz5`)v8}T&qdeJt83?&_*mz$226!IwuA`9c6_V<
zx|qro%rf&5(n%kNPM$qWi2V-2Yq%4Pi_apfTcouIGpMA-l~aK1AH$*Ta-Hx<?f9LQ
zkd!PLq(OfY7j}E<&*gwQ3Z3mqS9hC9x9)b1g8KNa=)?k^*YqmH|31m;*_w=NrKZB%
zX{E79O+^YGR@P47zky{84@~!ZUw2>gypn`&6IIHOl&s76`4uvRCfr}_TU<jVT4i!N
zU>t8y&{u>?N-DI!=-+?xrSeM^9e$J_hLuOBRg|~)XQp?rz~YrOzF`ROP18NZXtN10
ztrAytu0)NLja2G28x#AWgzy7r>rF2RX>X?WIuAS_5~+SB9}ksTkF7)+%r2hS^Vrv&
zt%Q!#XUGhkhlVS%O(GL22h16tQRo$~m$OXU1MRG7&Lu}ouGQ!}kZ_2<`x`%2@YC&-
zy`5FrBbclS)X>ap5sh5=A9dvM;!wij3ad#;+nh!=X<^5?>DlmtLhU0>09=b2ENJBk
zp$MM_uBZ2%2mM#bQe#~#v<gPRH!#<}?4PrL`2y2XxGz!qBNqwEK-#d}dU~T<FgZPV
zx5eIIK%4n@1|i)dN-&$~ob{v(RRw-|l|KYw<*VIs^z&9ZmWHB(8udvzv8|RmaZqj9
zb37UMn7ZkY=TDX*TXHWBIlE~NXYt-AbXxr$4$E7Zr7F(UcqWFg>}g-{q@Q>A(TR8G
z&+|TR@soq@(n#jrsA%md0_0#T%dQ+@!-QjGV&&CkXKu~i5hAT92Dy}x==dsO#@*WG
zKM6$%6et!K7Csi~mx7zli3P*8Q}#BDojAXHjL15Q6ry!`+be;#QJ0k398HLDZW)ZW
zD^j7UiAM>(wD;WS?kf9(57!%vqh3Di$z*0W?>szy-IR>=m+lbtvNbjYa^`<+XD@qO
zE9cm<jNFqwYQ4G!lCUOx-WiO_Sx;Ns*Ef4dC~GKZyKoy{uI}5iW3+=2;^uzv2tk>b
z3&q$H!Q;z_%YYv0U4}J;<DEUc-}hYG*62Zh(~8t}=B=4SqF47iWU8#oM~yJkj=8q6
zpG#Rd#?>KTa~g~5Ib;9zfUL3174kdV<65!_4Na2<jlriHBDVwD-_}p^@M&M^_xYUo
zC@cZRHIeQm51OI?PRQXmc#z7@$6NOupFU7Kr0XegZt0&$0Vl|nn#eUEy3#t8Bqd5s
zeOixLPmRy5l*v=?PvzDTz1c?B#W)+l^Hw<$Axo@O4i^v`J~%dX)9E}KG&>uXk|tra
zUDT}5q~>KaTNY*&b-S<;^F_f|&s+KXUP+mOtl4q>%Jo+AW)gu{MuI#_*;0ADJAs>y
zm<OBKB<GdH)53PE!<BRi5miPr<kdLU`g@2hgZ6EG{qW4($EwY0G?%_mo9U+uM1T!u
zTg3r5w*PJnDVST-?*?M}2Nqe>0vBm#*A7q{jsgz;)R^#Ydvt_J&ku!#+r$Ay*?Cs8
zW0%+0nP>@Xu|3GFeefbgZ)euSO|k&64yaWYeJyw>wwrInAQ)m_Rf)9p+dnuTs&TaJ
z&$bpCC_8n{uSN1UHSc2FZ9Z;{nc$o4o&>b2o{$cP9`KEsT~qKq2gAVr;>u-63v`ie
z^wEKpBQp}iVt1Bv4^;u=Qa+15pY|m$1%>wgCbhKOc|6UA#<8zTPdYCDoT`?GpS8ek
zAfF!q&OnC~@iSECL^DqA;7UNuq2<W&cO(vO2@{eNEcL`rWx1Kqx{6hkKU(bFob@XY
z4mH~HDr9>+UoydYHxlz^&kRy3XdFwT1L?p#2Vg+ia@LJz_TW%z>gvAAtx$rCM><M-
zjb~g&rq{LkGDgtE%@u1-Gne$Q&UwY@Ly4oWC2`68+P5P1qUDXc!BJ!rrhwdtJ($N#
zwlF?d!P5u+uuRdoCSlx5$0QMTsX=~63pTOBf_kIg>+`07llYXuT!Ee1E2aCtPp8!7
zBX(}7p>xht;h+>m{NTobxF$Uo=9znfU6K>XvBHb1H+?&3TYhxzqud?c#Xcp>548m%
znZsy?mmb)RnOQ3ca4_rHRWCV9DgrqfD-H)^*XZdsw!Zs8j|5v@vrw=TPVxfV_?}<x
zSyJ!^*;jd=hog9FAOdvNue>J5Er}Sutq9VubT(L`hzjJ;KI42vZWjBcJDlH;%9=HN
zQKPH%mh$wnbHRhT189P(*&k1~@xMYLt1Np`x<>MJX6{V=DByd#-SPbYt}JA4*gg{S
zfE6<ewNmncDtUWO4NWLK%)Gz6FWJrCKQ?2pPOacS5kqd|&1neb&-6Dy%_cJf*-Qm*
zCTm-?nbEK0AM$1cr=&_;Lk;iW>*`Gx)JK!L7|ggdC1=Lh$m)h#;x?_TlulC{Rsd2X
z4#fAQs!QjRi((p`|K7EU4$<a20+k<>wRp;#UefX%%mK=v#MHIznoR8r0<`l$o2Q{G
zZ>afe?+tX<l976k{of}ojoKL})~x=80zv8Zudq<7!^j~)AW71GNp>&2%CsmNDrz~n
z;PdYs!%-5zyEFlRG@VS!QchkqK)IKa2_h|s3^e{bl6`BIRgUJ4(S57L=+Z>I{1X(S
z^zec^L@V2YLh^EsnmHi&k)VeNk?pR+RfYA)w{s{#tkKQi>0`)H)U`9wEe6h6R*0a9
za*M!(tX7TOa0z@fBKlLjXkh_OAt^+PaJ3Zb>BP}_D5|xN*aWU`k)EW)SW^AHyW|Mo
z8_L(i@Dlf*VomvE75nUQo5tc)nrs?$u)@kdC2JMUX%G1Z9UKs8m|%<-K;41k6)3<u
zIA>b^9J6M0vddAq!3izeSOaf8Y~BbP-QAmVK1KdmL6*hkI?+Z3bxz&|#(dlsfo+<Q
zbE_u8gg9Wb!V)<yIN{rokVoH;N>6RvW{5lFJp9ZwsRe700E$VT1VYUcn26iB0bY^~
zIdPAnK*|-$jc@cSZ-1_?>eD@kaKknJtw66h)la#VHt!=?XA|){y-d||);E`%pE2Xw
zc;KA0RR1<0>5j6@(EY+Bj)=#}=p01a8P1`tJpsB?Lod00R$!(Pzp_E8q(MkGdRLt?
z9;HkR1wk)uOu<GxKjD|kN@9Y@1ejaI1buk-?YM|bQszun=pBF*uT?^$TH$|wB3Fpz
z4mcW$wQ1Cg03Z4E3TMpYBFJPfXMcuHl?&B#=h)L3Bl(Ka>T$$q0btQ1iW64Nr;}c;
zAKpnO%^PK(pCPewd8ru*{97@lFx3~YpH{`R=!2+CVa}4#Vc_vZ@C_2%2FC6#ZCE>~
zT&|b86okkZhK4ZnP`(KmsGHfUq!*J>p3ladoNdQ?EL|%_N9yP>1%}+5{{VxH!y1MC
z><J((IgHXjWE~62)78W&om46fCRBFTU?P}N!f6<`>ex@xT<3yXH$rlO2n6&Oq`NBF
zvC`FL|9!uCJYPT^EX+w6Df~(}Db(#JzodD#*dCa~l87Q<W-=&>D`m7f8yBFHkAX3!
zDjjB|ZI<y<?~bt*538?--bs{^ItAk`iUu!ZmjXeqW{<HbDDSe7RjK2T(iBlJ9&vkw
z0s&~i1;7e4pFyq|$5=^gmhp4Jv!(HVN%*Wf0u|1oM{xxrg#l<&K-oLd5P^5Q=7pV`
z2e6^t&@&?^oF`kIXe%#5YsNd|q}TaL{NUB3(Kx+b-F2Eqky(eUBZtGN>!SrUD@Wyq
z)U7bZ_t@IQ$XHieKdcTX$XJ;fQs${l-6!t^%ZSQaa>U{Le#%;4Au@6=qrCx@KsOgr
zDrUc)G9L1PI!rr7-v#TDU;`9juJRGy27>A_R*frzR^j7Z)RxiXX0utGCi_6QKyuyW
z_px8KKtgyD8JNThJw&C>HhH#2WGT`vMN`MTJS_bEmwy^Dir+Ki+~2KjxZZ5XY_-bT
z^ynZF>hhrHcnL2O9&G{p0FYkA#mPxyfEVTLUYKGF5I+=pf4raf&u)j%15^>vJ4V5M
zTJ6s$sO!WDh%2<8GG<44<!sIt%96(hxw$JvEG()|kLHToUZL@91upFqr)0i=rZW);
z_}WMBE*Va5<OuYB2YvZ=T#mhCmHURr6L@Dhp2GCs6g-AVm^wU-QP0m|LN}fLvb)$F
zkgSv$=3)DPN>zY)zxKzUV&UK~^xk=#Pf|Vaz@2eCGx{*~hNDWtx-9&1Oe&CgMg=Pu
zDN-MKyFFd}gJ1=)Jg#HC-p>7~tY`+_r$@z=9_MZc$ow;xZ^eHul7D&GO`$%%Xk#rW
ze*oqsr<a>g%@UZfc|r!vWqJGcD9t%XbWZpCUc1_IGOarC-=XLqlR=tgz(Yd#2SqpY
zwqIl9xA*Ogc$+HodF8>H`Ng0AMlatRW#Y%68e+Yh)M$RmtP%cVpi2U0l_SdFEd0a+
zlhA=cKXW<XL>imLUO(E8PcPLn(-Wf$<BLbI<Y&7`1{E@S_A3C=Bm<LA&HrX=M+NZ#
zuNlqH*JBK0z<~S0ndH*w<{hxeaM6HAP#{nqp|I-G6S$u;9a;>trThd-m{!wq-3d%*
z%IaZ$LLg}chVbyv!UL$Nl<EGN`Y31OF_*@&So$f)myLK1<4l#G$`*B_Hm>==d!c5E
zol4NZUw(3&f#d#*V3~{oILetsCZo}>E>^q!p|gNxE5MHU2JoGbG#m6-GH7eq$?$Ls
ze$sW7-_(S@0Hu_eLxNkTNOs?1<w|(~H;&f!u`e|Kzwr)+_DerJr}>}u0OO%@58Fe?
z&;JXkTt)%~@7(gZnTU!l&>4pAc0QUNC@j>v{505Xu&1l5t1D<}ot}R>Q5WEBD&=36
zOks4%dN)Ea1MET02pAxOaxkiAaLmv@A>C2J?o}VMI8}M<ROWBhK}Ganr86?>zx9t-
zr>j+O(0*@RditJRP{4-^4RzJ9-aaNYpuAK?=tyAq#b|)Bw^c>Zfg%za&2I2E-~=S(
zB2YOTc1Gj#elPIzf06MrLKZ&eP*ab3-xdbW$4&MhE|Gxhk{p^}Z|TCL@&S(Fq7T|X
z3=paEQ51Um7K*t`o6Gq+7=jx=E6SA5snE`^KPfY!2ARhK<;}!Y*eit{@`8x=m+=Q%
zg`%Yf@g`q{KSkUHIFHkx{WP=U=J0QjG!2~rCBNP%X~LC!yAunAeA|kkRe<wcyq1g7
zSf{uB@G4bE8keahbOu08Z)-#Yetv!8Y)7;(+<ea{S08Jqyd|f4SgKqmRIk0^ul2mm
zkDjNcF$ksLeT#fVv_*jqw(jyQ&_XSuwD~aQ{~XimejCbO;zU{?XhNT<!74=U(|ond
zCGUd45j1c0zkbaBNMQj_;+Uk?UD-1YJ|m-=x%LNB^pzBQz<j}t$lUq9695m2V_pA<
zjN|<Z3J6{u9=<h!vu-q#L}D@ZX1a#3bNy5h>{Vy_^8FuSa^l4!tATaDC!|ZdE9@Ak
z+w^pP>KBUowdUr_UZE?XW25Gp8^e!qrxbigGrIBH8Rt-!xPy!S5N36rCF_}*%582X
ztpX&p3Ga6APB6iF<%`dcSz&5dt|f9-YM|!PVd=HY)$SRn1k!Tvz(vQwrWLD)rmLbr
zcf}i26R&2p<<GFUx6fb$lZIJD8zt_bcoq@Bb_D(q;m!s(s51b|+&;aY?6$hFVn6;X
zV5d8GJKkt(TD!sFz&P%&Md8EWF<Ny27E2*WzO&b}kEj}@<dNN_er)C&3%m=FpYIen
zuvo4=YRR<4WT%1s>)mmoCwu3bdW(1>;WnE|M%5*Q6fvxGLcn(RXVw!^Pj9(eFUoC@
zfS?HlEZSVkt4!viSdNSFxzsJHZ5W`W4qHV$gh`2ZT)JK|9qMugAPBLv*1dFJ`#kY^
z!K;!Yn~#y;G&UIsFcG|Yml?41h9Ay;eLN?>9{upPKiBGqBD92rjL-!y?SmBr@ugS$
zH5B#xF$=ZxBg?P%%UR?BzMuQbr)Xte%J$c9KwqOZ4!I9;lEb4<_j<0_iNA?*+s(8d
z5Rr}(K)~(iTdXL@H2}-_-Uj3Vf&?azQ<~_1RpPjX<1HT&KE3(CEmfx_8n!54`wl#f
z8A-k~k66rUJX<Du6c-mWH!j}Ki|4GUqZ-pw#V(Oo#amm=?WJj7lXA;Wy~@Sts$7L5
z%>s790PS!>EO9>@Jt<hNz#91C)^#wp_EU+iPo4}DOBzp*lpGfx@{sycAASgcL|doS
zM0;I!0p>NmEU%Mom2n6LrXN`*^Skn=V~T=S<oFlQQuVnm8SJ2@JbizM2yD+bKoO(*
z^(+BGz|U2n5sIJ%bgwUxRHaPy6*w4{@4uZX@FeJe<1AB-iv+X+?IlxNYia=<pYXjO
z%-7`P;fbw{Du*5^H<wu&`?W_3G}@lH;_qO`5;P76M@3-NgaW|EI59AsW_Mnp!)DP7
zuz7sGtHIiEwW}3zTbvJhj<%c~REd$wp9kA=emXP36$2}?1;|D|0FaX7p7Q#~xMS%H
z^F*k~dd}DgApi{vOe;^-K>U}H13-jz=_1H`W*dkgpw}Zfjgt6~W5yb6#hTciVwN<~
zx<fYfSSNd#)&ym}iJg!|5NqCOBY45^tX0(Hs_RLGU?c#iSuz>~N?!G(D-G<v$J2f?
zJtuB%K*tSHx%7@m3~Mj49~szK-4q`U&##0~+fbkWFF#Meh^j*j{8xN;hTLp@>lxQ&
zZ)`-gHSwxciFe09MN3oJZcLCy4~7dDZulJ0Qd`=s8AFp0D4kppvrhw`W<6A4@_Fr8
zX^4Eyax6|9xpc#6kJ94FYb;?-8^LdIB~PLlZC8wvM?US<ud{XwPitH6<eN*G8&=Dq
z!TxfW3<TdDP0=6X0knACDsc2%oUwE#pbSo?nPTcIw6@ygDckGL(a7N()Ws*~FbSQF
zre~lIx~vV0=*;SuYixpFPI2p-0=ub~;D7h@+DYf}Krhz$qy5FpcB^<la~Zx`q(rcK
zmQYVMdJ0!Z_42!sV8PXXUEn{U2%Es|yyY<C1@j6D`gl@A$Mo9ig*`5NDm69Lqg+VK
z{S#TC!LS2Y)2{?p5u>cl$Tb^hQQYVMMvK2N`uTYdT19r;0sNV@NvQr3#VUkeYIiM*
zQZ_-OXCZB4D3CBm0P;Duz-nWxUerJvR9!v&VC5)$a76Uji`C^UE)CO~D<0oB5yzWR
z7BT=4=K->?_^(q3m6-Lr-$1Rk=jq0qme`943@bQ<1|?ei(Ne7bxA%+NyX)g0C%e`S
z*ZH;57kDjt!$pBa8pgCRk9qC{PZCGdj&+gb!DMbwbHz6gy4Jpc)orRg4hr+@hIX-i
zv{!bt3oxZ+bOxd8cxa)BT&h(xcKuC@%~->jzw<CGEpJ)w6-%zutKlcm7dj>H>e<sM
z(n@d|Oe63S*o0K>it-@%8;B6%>tH?^?CgzE5Y!;T5D7+*>>qLwdfJ|PaJA2Qk6xN^
z*doG%ob|W%CnYySL*9rVvn-jSnXm`IlWo^{JlPX@<y8&>t;E=16NtVPMV)juvK}K&
zxS+Ysu8Qr^L&O}#9a=3XfwdNgzVC4z94HVTuf7*wP9#Wed@VEQUhdAwuzzn2Bb!db
z6x7FDe#VwPW;LkjqHSR6l=W=;Ra)%-C9PZ;RVxJvAR=E1`SC;NLW!zj06*;|1`wNR
z4~0Beo`G#Tzkeq&2hV7SYm{U@xwPpJir<yD3W|Q?7zwi=gpi+|P=p5sPEoI=SU-4Q
zH^5T+P--qFP<#_MH%x=SVB7mc5hN!KnIK61Ks!pXO7lH<%;7N1=0Z#t9r@gUnMZdw
z?{^J5bhXnpoq2^?v(~hn{NiErNawFn5Fqq)W7EtInt$7yxgba=3@kHzWEUS)3@nsX
zfB=YLM=+t>K_US&v!n0t=oq_dviH8@8*NTeiL?j2`ULj1bs2Zu#zJjoFr-vqR92D_
z3NjVy|LwZ4z?=fQ)8d@$7uOD@5AQwqs@il&c{N>ROZHYQFvk{fEacVHP-4<zW6REc
zu3j=O>LzyV7q;tAOXpB2puuQVE1bc===0@SAM(^H&T5P}j+VXVusLgHp)T|qQpy&?
zb$DAP+uDz{GyL7o>WE@!VM?qw{@ZPST7)I!AXW_<<C0p=jE+Xt?H7`pMjxaphBl@V
zjYRFkt(}e`15Pa)cqj-7<!4mq_+l&^IY{&8g>+})jF?^Ci7^$$#w>^lW2(q~$NTxl
zQ=8<4%zs+9T%8h(0$Vx-M4;(b!K%N!<G*jp$^!Ad;pqJLcnegMr&C)y$3=QGa<fkA
zi&2?zTRj9dCi)W{!{1SzRMTq!x({ZsP9noqQPQbH8VLod%&I%2Pk1dH14pz(VG`hq
zAv_pzyTKev<bOR4MA+b>qA4?`w`amN<UsU=66y{JfC5`B1vd1b-?0Q6Ow;-+>Tu-A
zFbNLB<!$|gKE@!Z84E_Vi&-_H$m$5CF{mB~rkQ0wI*CmpERAu%Mora-AIKH-obhC$
z-vT=5Xg`6HyoX)*%=6CzZYW~P{=A7P9psHa^o4zV22M-oY-oTP*_0gIYJ>Ijza&&f
z`4```Z!GI=)q8V21v_Q!{QvlR=l{I6FKV>$6DJKC+jbhOv28WBZML!1u#IioMq{f<
zW7~J<ob$c+e%>G6KVU!4+Iy`z*IHxFF-9hQjGTwaXCg#l!J=ZQZVZ(BMpxaAXI0%S
ze$8{A%@Q9OXh=2lXE1PKcYjCS`tq*KAsq*9C-*cAGsdxEUa#&AyP1NoR`zde>xcm=
zBESo(;4)kN%S{_c2;c`iqN93RW!t*yH&mlOx%-@cx;)KO*O`utJmyA2@m@1AA$Q!^
z_u75HCVY^-T^|CW_%X@<L7pAk(b{%nDmgNx*KN>ID=Ry!X)qslOp|^VGFeMsELqiA
zv2Cvy>>9DsfaKsCN7SyIGkph)-!-KBLIgKB(zQQ&4@h2#k3!)v##UZ{Z_uR?$wzsN
zZG?MZ3Y+}G@pV5Kd>iTHBp6fcC85)%N2QAR=+}i=cq5w8Smo>UbLDRf$*p06OF-?X
zz!iP%DMz2Ez@!xg&?06LVo2Sb(Z7GDU<4`Wao3Zo7LeSGpO;rBAz!M$cQ`aQHaB@;
z=b|+e`eIq?vt!5e_OoD#ULO(Fevi;RL(ouh?h4sbLq~Y2hK2Q}Eu=2hr8a#!C*b8H
z&B8~3IT`A&`7z$+OBvsPahOg#RXmP)ic?4|{_Dn4@*G@POiV5a1gbN7&Zykkt~`xZ
z?N67xikjtbRci!|={M!YOWOS=8OM)RYh?ezSl*PES95A!A=}LL*Ag9Hs5OD+AJb@3
zlaqf|MGiLU6kq75B1&>u*}NyFu;IB<yI)o*CHKbtTF{MNjA|8ZfS%_N_Odq>x@891
zsSzwCz4l*Q%Q7DzlF;8H$$7JE>EB4Ag!Hsqg=QzKZmgwa&|BH8B1+Za+sS~gY~)f;
z&HeG1<@v+GNQrK%FEAO`$++ya<vk=?gr`UT7ZRM1kB}2mMU7IH3QWs8R(DOp4c2qb
z(9QJHHYnyP3t}xUCo1VAhxC<D^b5FkYW5H$fQC3kppEK(guYpX0ACM2%jD`Tr_9!?
z2f%pw^+qQGgAj2xiCgD?S^xnRnLJWYt{%T_IY(RqlTY$?#RAr8%8h_vDKZi){|v<D
zJLuISo>O|Nmzb8O^Gz9_|J<FcU}4>5Lz9-77}C?yTiDiCPT1p+wHg=+Ws91t$ooh{
ze%ZMEZzNa&NH#RyQ@JPhDt`j4VjaZS9<KufDCda^O^U4`KWFDGBx|ccvn;qgge98^
zjfRn!9wcA@pEKUYxJ>uIsaHkW&-(PDP)`GU^dLPsu`)dF$|1~`o1t;R#ifZeq*E<x
z`<3g6zaAI^UxC`!Kktg_e>QFF-{B!36v)1Jsg=_P2OqkAma19|j&Gd%aQJlB&hV2r
zxykVTdva5Qr_q}?8xE{k7W=)wzF%ZH^nb6<EtoSo05Nx%Q-p&NmzhcVW(F=CRb^Eq
z6NNuyEi{VAKCLD%^8SY+Io2-!#ojDVz`*Jj6(DP4yZl6RkT_E@|562h_z4U=O`AiB
zs1_LK*`n3~Vg#Uqju_GSzo1nRf%L39B*DgBVjt6Owf=PcD85voo4t(jwHX%@cs5L4
zLR#7`_AxL?W6BW)@hRi*eFYzqnxAm~gU?|TPgS_|!W1ImI+x(J>{g6}c4yBl<v0Sy
z5dBDiKP`N%HR4+gewSXas(NRr{=@u??!xp~(}n+y$q9Kfz4w8u@r@Z#F=<Y55TpXh
z!g(yYh*_sI7-ZhhP6JLs#zyn32I$iNQ<WkCS<jjNK@$I|K(b_+*dV`vzr!dpQLd;r
znurH-K3Nr#H@00NimJ}pg$`Jvu@HPu!h%)hH2;0rW#Xw;mk{!wuK6n2ts*OWj+5(t
z75qQ*B|IUQ>upF84LwH2C9{;X6MPlQ0;t~w<Efct<|Y1p8VqHBK@g^-e!`N}N@^Z=
zpE_FbwI+$hYQh`VrjEDXT+z>3el(eaEbso$*KuT#oL66ZLyRT${~i{}X#idS?j2i<
zuIsQkoep!n_{2F{H+OZ|N<-V31zgk&s_C>c^ff&r>ew(aM`$}2ZeZZGDy+Vgw+)B^
zp*lbSR=V*X9KdEm4CwsZrs`U$Rb(+PM77Cw=<xFYyw%(u5Y4G7(kK|u08}b5n?l`>
zgo@f;At7mMa&`ys-Ada}0df7Gr_A*q|8YBJf;>bGu(g!+$LX1DYjQ7Ny&2cN963h4
zStO-xsX)=F_biu5XI7uD+}I2!OKsvWhv`_utUHdYx+{2}Pgx7La>joq6gb*?+C)m9
z`~F4zy=ykWPmC<*TTfh4!)^J!D?WeJWMUOcNl)*ul)nQbZRx#EFUg+?Q=w4@5ffNt
z(0dXyuPQ74Z)K1m*7quYo+p&^)qUl-tv(;UZI440)W~49i=)%4u~wG8#)}FCB;aQQ
zlR1U#vVS1GuaC^?x~zYqV2Tc9e?^c7PZZqg_?U5Ly7G+yiu{OVt-%U0vMUr(bfVo{
zwfnGyowJJzryGvmoa=k2kL`O(o>CE;el+}Ng|Q0d{A>`4g5EnON83?ad)Q9_UwNRy
z5kqvz#E%S@mU14XT}%t*3gJ=Opfl(CNaIRBhid-QwB=Etfn`py<`%ej{}cZh;EgeU
zQc@!MH9w7mRv<cc8XFJ}vb`2e%60S~{bu;u^)YD@;%h#h3>G-QvYyd>JKg0^P`|Ri
zwyV^jp*=-o8jt4kpywI+1G5^h(!_Q*pc4Jl4|;@?A3oA%ckeQMFZa^xJgZ<qH?|}(
z_kgDJ$Se!?&fA6#Z4Bo6Y-aUm<144o`|1aJgLMCa(*Mb<?}s1YR2$^qj5`O-_Yc+k
zzc2sqO#a^<frYIXn!Nlkzx|hWu)t+efGPa3nWL5R&&lea1^RCiUw+R}HdD22|Nnmg
zo)bj-0<33>&qed(|BH?P#s-kfQ)3lyuj&3h#r%KIO+mh|Tv((Mm*Iay;QxJ@B1B*X
z%>Ey*B8VpDk5)8#lV_3p_nV;D)ltpO;(wuojk>6yMpYUyFBySpF-_bWWt_M4F)8wo
zv5h4GC^0d4q-GaN3E+ZrXJlrQ<1g1IWn`571@pgA(h5#Qfcba40kdBW0P0wc?~CWb
z@7!;B`S}G7O>`<0b`vj39eVdcWwd&w!ou(eu`5DIXU#D{4Jg)Fjg$@@17jG_0?z}K
ztmBhR5M(U*hvyHCh2BH<GKC?<<%mN9V2%b0?^akmGoYe3tp19qvz=qn{9$!cBo3+M
zBw8#kSY{3k!JHq2{^`>wfWn*feMe<+*#1IIODml=!KhUyal7gye*p03tn2609078v
zWU=L|QrLWAS&GAG!7#lMVT#!RANkCus07i`!r>m1&Y?b!+}#Z$g1E}$XBC?i+s+r)
zCDU}99m2Q!Z!MxKOl)(eHINRZU3NPFxh{qdaJfI_O=e<8H9*pd@6=JlC`-DIbjDab
z`1VxTBP>x4Cu*syk?$mXIF?k*^>O&OC9ta+;gw+kb2-%oYP*@&1`uocCEM1M9e_ld
zC=>+EW$Lvv6lcqv$3+)zUbIlAInz~M+R?&u)~4J6xLCf|4gL|a-Pzq8UTt<#7Mr4`
zXH2r{Y!|nE`Fp0M(4@t};#lI$O-4-3phsFM`=&6$VaZZT2J)r`xQH8=bQ(mjPMZxr
zk!S(r(#b>9`O=)yKJTXE@5FR~O)Rx*XH@&e^`|6}f|}at-&mxRtBm>_r$VIo^n{Z#
zuZ|$Ag>rRx5ld<^@0AliX2`KER^~RkdSF9h1B0B3Bmf}f3QC|J+TeV=sAxKtO8jmr
z-gZ4LEoWf9(a|oG$|UqINzDc5<L&~Avr;sGqH5>!bp2bRvRs_)T8kt=U7O=G$ehUo
zKsd@5DnrrKPV4aNlZM>ui6d%!ZI{>ws4J7<5zXaqvIs-lfn3We`+C*p{6Ha6cVXk5
zAs&i|8(hsQg$I^JN~a9}i9zLOc_3dGi_;MCw;{lcBOkEe2((g{lT!^~xcEUQ&O#cS
zD=ZY(f>*9rD_2v?lAqLqiTX`JM`xQ31j+o3E|x3pTh+v}sjzUS&=M2WOHsjn+7aeR
z4PvRW<`xaLy8gS*;}WWQQXDP5)3WqlhHH&gVZdXvO)kd+Pz`ar$#xjPjs~-wDTu-Y
zh|7-47KL1IuMbyy(kl%%w7TkQWg5iq-XQ!wj}oS)rnwDk>T<ZmSla+FB+2b^xD$*u
zx%%Kd{HO;|{yoZDrBc)~cOufVCDcr#uTK~CL!Ghd1n4u#M(6323XIDW(JP&jgRe2;
zpt3nPyp;>&vl&3r5YPy*tE;QVCyq*uqpcQOHKFW3trjYhm>CROUv5`g&BlAM5<Ic2
z);^;n^l_AHcM6TdNyfDpf7KltnMg!vPstaCa56A0sE(<J=7EB%Qj`l2`qAOp(Z+UV
zl-A_l_G&*)$p#fV_5%wd;M?_ja%MvkU+Wo7=c$`}Wa5pR)$T}Ag*~`0K<6fUm&X3#
zUYA#@R_+){amt(R`&cZ^X*o;vZi%3)^}GR4?zP?UZC(NL1*yREd62dS3^D=TI|<7e
z;Ln${O~zoeirKbZh&A6XTkQh4U&4FqE%Wu=s#DDKO{HHF!1B_41&FOGEXL^1zr_h*
z?--jl!MT8LS=!O<b%1J1^sbqiuRYX@EI{{e?J%`Y-?*w2&Gi@S?w39wA71K&>;V+c
zboq7)+d-0b&0cq>6=&+YE=kta{g_EN2J8<2hjZ`H1E|^U0$f^qoE?BiHvCSM)$z1#
zIiLV4LGpQk;PvH^{+rgjVc*E}tTKN6$XJrKkE}kwuGDNvm(}*8ok={m(A{BMdj|hK
z`a$ZNT2|(>=h&}((hJaa`XT}yO|E!M^4nAATLNG&(|iU<Cd97V_ae%jN2p-E-d1+L
zwoKVo_5XlmeRtkyb$}=&l1=B31q2anA6}o%)^>GWmqfelfm&9D%wJ|>JUf84{s8DQ
zWD1VBZ(+Isg0rB>x#fB*U2E8;$Gt<KcIT#{L@|jwiHP$Wgv_t;3TRHKSDSaQvUWsL
zW3oV!S9S4G@y2E{`&{&Y$_zmu-UMFLb@J#Np)b`NK#|d1d9c~+fQ~lHbV;slav+p3
zXO$t_ZM2*l;gc5niM_y^_2zTCB>wJ3x&Tx4mF^xN5b`BNX^M(R4sdii(XIp+X5ufc
zD^rnLNNKq+-m3=I;o;tnRcn3Gf9JIF+K=j(<;mg&Z49}l-gwnxm?fIh(h+645vkn_
zIN(vif|{BRNk2Z(rPEaIAN|qH9+89y9j?lHQft=V&_{tIO4YABZ&L0$?!xj=PXc@&
z^5bFq=D<VIpu>*(Sa^X_pCbV2Hai-e8r&KLX93M^ZpdO0*GzIu_I=!&NRW!qLOD8n
zP_@vxN7xKl3c+A(O?g3EcL@5vz8&5G)vN;V71f>(u&*hU6_1EMLcH_Q%?rRj+34(i
ze!c5V0W|iLq~%x<5}p7Hn5J7_pqFE!J{(`h_wj(z<Gqs6)(co3W$?^w_ebDE1NB_D
zm*C$}FpxNUOc)^ej?9X9thbt(+B~;beipa6W13=viM^dSI{Cv=<<|9SIeLYIk<>3F
z(>+9gXrO0CPsISu`*^<F#a{A3?!Hs`e_-@gxzAg{Qk*{wJg)BmCy=;zs~SKJ;Q&kH
zaXb~RwWp~kPh^uFXbine>$;#;E@*n1<bNsyYM-1Fs)z`K(fCJ9Ukon5Izg%${zEW8
zTV7)}nvTI#H6-!=1p?D;VuYah)6?zA(P6&rhzUY51lSg!6q5kdiC1p$5$?w#nD?-E
zv4ahk0Sa_y>*TvE^e)q1-p0OKl(w@l4%TR3ldHvZ=tDjA5L6)kyQ|*$m^7BgItYsc
zLKVk%Nwk9ZK=t>qbx&U6xN1G?hzQ3xAw?RZYDSS*a2V85KrPdm8;{w*<-xD+jq|IB
zFTI)n(ei#*Xc4Q69_cOL&lcdNvzliprgMmhUKwIobB)g`ZyA~>2|LNFzy;cZXowVw
zgnN`+;?g@jvrg4m^?XxbLbR>nX7hK#&IseeXiZ1zd0q>;TLQ(N{UMO-BT=<{gQ3=c
z$3FI-{rpDok!`B;VcZVy>bogswlzo$9SrOz_z6vgB$KLn#PD6Az{@ampt*~t265^w
znmq6C6X7rZRBG^LkW7V4PUbCPQC)Ko6sU$!Vki`-#NB8(UXZwTt02Z%c^(P)UKqts
z2v|7U6~w%}K&Fds4i0GD4yVnaNUbw(GyK5VasPDz6HDyO6Fk=&5X1Ql1|jwcw7j6L
zdYL2lD7d;d<d_UBp9AUv+R?n15`)Ln$%o(3`LtY;b}jMA+XsjDtbX#8>>S{!>nef-
zu92a!@vx-7cbrxABFRy84Q5?{RT3q4Fm%$QgM+_>P!EoL@$23MvW(k2!L-p@z|=V@
znIePBuYB>VNi+Kk6B43#$4PJ_xevU7;Naj!?@mhGr5}u>2sZe$KW+FO{(T@pp2wvl
z;5CaB2<QVr0@5Cp@1!odt{i?kl}KcnXW+-r=nz04US-}rUNevjsEGI5Hv-Qhv`BNb
z3OA5yT_NDDg#^<n&;L1|pgX`FKNuI1S!WeXz%J_C=SIU&5={B~?Tl?OCY?cpNqG>t
z#p?xc<ERy>)q;(^5&nAB>Z&zR>>GJ8?LAw%EBFRn;3Pj6oJ|sv$9TWR8vQaSsK7R}
z_9{D`ars7OUCus?QbZ7dH-~d)z1H83wBJ!19GXEJBkLi;|GaBsQV_nQU7xI50Di&e
z_165@b4Uag_`&v5(?i!{PNpbngy|<!oJRaqvIzozjnA-$!NfPVm^zfmk)Lp^^_Xm#
z?3renW|%6n9{QI*IL!f0XpKPQExphT2-bYtu*krGFCoHT7^ea%&30M%$@>^=7gYz5
z^x!D4`Abv4)*w{)*(Y6RR9V+RI4y_eUY~Ub58={CQE=I6$}(a|h#Ah+F9y7Z=u0P|
zZS7unuBY@MSf-ga^O(l(_l_j(Glj`G-=>Hg+v!vc4SGZ2J~3r;NFysBElUNNFVJ15
zaCj9Wq-q5-+i$g2;?i^cnbBSuQfp*-0dyFyQ`Rc%Lkx9LrwmLD3l6TVqolj@VWgp&
zVv{er$6<-VPCjeTT&~4HMRG1WM45fNhusIZYF85$ex@y^jjfYhxdE7G>X=K9iYhXL
zn=VhV?wvmjV(IR-4nbZ(nNPR1826+@9;Q}~d%4FcNO0KcZ~ddekdIE8M3BMpQ{Z`z
z9gNPBZL-+{JO?iiX!1It#vZU9J1iu4Yg1$p2A{;O@ZpV^wMgUz)mbLduC=(Tz-t6p
zPnjj8LJqX6+D5~rp5&=YM?6Vp{Vv%^25wrN_Y^vhleA_dxh<Sc0&P2+Hs6N(SF+>n
zxstn4BupU<V0{HPKXbW{uI)4c6%La_M;7V3-cG}@AjczyPxD?ghEfKnQ)4xwsJ8eZ
z#3snCM~E2$XTCtN5ZG^;L0h4k!3JaYXv9Tmf3a{%DyV>Y2x@}%DCd>==+cA!>u0{q
z$RC_B3>hOk6LWp{Ehu%QF$`hoG56T8?b=42*w|a5u`asaSz#ti&_||fZ_|KphCxL^
zw>@+u=2I{oe;R^av4n<LL#`1+)1H_j%qLJk8sr#`Q3v@79@7L1@I>aB&8|gJy)jjd
z(WO!<C(9+0KH?K+PY@6v!P7Rdj4Of}zUe&;?}yx18-;~Ud<lekx%zU><p9^D)16~<
z(JfEF>#gmxeC~E`nz8>0;mPnD{@-9oopJ+nLvLc8itof9ybkZ&yO~JI6jCS${8<d!
zx<i-;d%9Tz({X!<!SR}4o`igvLc9NBZ*ld=0?WgV@ceIc0AMx(M!D*_iaJlEvu2)P
z$2kd}O!k!$%j+@1KQI=hFVta=>jD#8eR04qFIPaL7Nx?$uJ>c00=VH^jt(TyS&jn_
z?+#u4wN$H&K!d&K>4asuQ%uC)q31_zDY%)ULzves)?LOi9lp3uT;vaU;Cf8<Tf&Zt
zHr_ANL&7f>W999nX^tNj76d>eWcu|E54dWeQ+!<&kp<NkkJkO+d|+c>A*pgnKfN59
zO5dpgK$NRwxnx+U&N!GksNEg_NA+?t^f#C{v@{q9rEx-L9bSR}!DnQm$k8RX8DJ6f
z_wEoWoMT@u8>DitO7lVt_rPBT|D9!Z+6m}x0Sc`dE1;7uPP_rUP`vU|#z~Bh3qli*
zZPwlU#}^kYkH@{MUOHhAeqC>4fLoQn0q<{fuWfDwQgHCyCtju`4Yet)r`I@`C?BTH
z-zjJZJ|2nz&t-vI%k|Q`5zYD|YftM$@bipGAJ=ZopKjZC#exR|o0nhJdR7;;+=jxx
z83Y^ZI8++qIaCTAMTNkC(PRE59QfvDpP!B5GiWoK-Th%EdNVk7a~NJ`^Hbl4W0bx&
zK_TGC{{ozC9c@jT3NLyB`_)>7ei1+lr<yN!IYB2SCN?=-o@_N6X7JXIyU{^{up+|*
zp8Olmb(o^BV|+dT@-3Ra$y^ZN>2GiV8GZo-3xesJnDPa|V|Bs;p-ZUououh5yRj)G
zK$0YnlRF^l?}UPa5{~+j4Jkka>P^Sk*!trLZex;~mk%ibsrZm)2tEL*So!>SDAK-P
zq7FyVVIXQTRdAr&o)75;BLOXBkl3$z%NoMf6Lmlnjku<<nNBotsDhmY9NPlhF;@$j
z;c45h#cP4(LBXvSz}7n$ZVNinp_D_Q3I4*H;v=#J<iC1D>UTCm-=8sYp;SX<ZC&2A
z@xaPx$HIOMAh(H4+rzOH0>K{V1si-oWfV%MB28UiK%L}4=rF@E>@#P55%$70UNk2S
z=tM!#B{OAi<6l;paeb&s+>JHlp4_#P0apuAH}#=-2%){UgnuwxGG3}qXnCnSAun~o
zoyD*;G+*Q9Cl27xbr6@h2ijlPY(+uv_XlViN7#NF(VIM<p`LcxFW0|q#2}wB4KOWj
zL56m-+>dE6&7+)%pvAamVqrnz&}sICZ4RKV^JmAjX`m1Z$VF1mle+Tt2yYHPy)E$|
zcYO{ojS;x9T#7dtU326-g|g4Z5=Q<>MivWsPjvtb8=Ndil&2spEDIqF3Wfv^nDSy>
zG{AM!v9GML9&Y$CcXa3LMBvfh!IPPn;2eLqRM@_fnVII(%5ydH*8US=p{G{X2a{v4
zy%CPsgvjjWS*+MR+&wz1GN4c}TA>RxAPy@+EEra`j?PPY9`2L@d6qF!f0cUI{=(J8
zGsLQ_Q=@;2fiB&QgK&RBQG=ULgGXI!*0L_ZPJriLv9{Z`p&JB|6^}xE=a2#?en!iW
zJ2D$dAWIEmKeV}DX-V11bg5+XBmlgU^2{s=4=UD!f*0MX%c$aCR6~?<;L+oV!zRCX
zDDSX@ctlDQ;t;E!y7d0>$EsXtq|hNjrfZt4`cv(~IKo)yT{x~Gq9v6>rlErxe@}$6
zTn|&V9%$GoOeB;mM{)!oY=9iSMuQ64>O(iE?*BFZU`srviM$a8CRhc2D$+&u<wc_o
zwV_AoI-m*S5k@EEhS+*7rc@P|mg_|@#~D6u%X~xk8>*Pf{hRyV*QH{VQO_HK*@aS7
z%|-<kwqz&t60C!{@=ni>B~e||=!{$`2)v+v$jPtWdjvUT9}>1-!RsC1W-^SzREj`#
zR8XrkozJ_Be^C^O(P?u;P?r1@HYHq+$a2&`f91%>JTlHSZanz+3A0Foy_g=|q4r4w
zJITR)VNXT7phrxMP(FWq>T?nfrD@aeYt)d3*vox(5U5^UafkRy3)*k^!a>pYwQoew
zok0Vw&vV3@>khLYMsyhZO&@46E&ws}oo947_@K0Ha~}j<y>{!n@2V7y!LDb!?g`r1
zWa3as-z?EF9g#pDeNfnc$v63%P*x3FHnb|66e@V8Qq|IC6gLSbFZn&z>$-=drv1L|
z7M>l67|W;P{$zuiuDGg?vI&pp8vP;gZ$|dISJ}o)gO%Vu-VIw@^9ABkl~wU-S$3sd
zlLo_S4&IF~&%p`rHkW-3G+So|yP^@A+R$`4q2$(L1eq?zMgvg}!OpL2KCg!EZ@DPE
z08~&wO+*M%1EmOB5>>5%tU>k$tALu8cdkKfL3~~I<|gWhH9knT%d_fU>!`ALXp%5E
zZ<L{ROd~nqNdtjf;io;^u|~-O;gXAgILdMO%>rCj9!!0*Ss~M2;VAvOY9aW5+D-DS
z#oSDd=7JCG7eq_Q>Z;Mqr}<&Tvo*5b>!thkCU_GKEkD^X9IR9SSmYg;NtC+}RLH>w
zC&7LXR=DKgLMdh+@C2%<LCb>Xn}|IOcbF!DX8vX%lh~`FWf7(e0NpsQJRI;b4)}#I
z)o^@ljixi^v!|2DUTW_Mo|fD7svQGyfkf3n1;d*Q$x(X>KJT3+P$PZ8OPuPYe_?qs
z7B3n`<l~!*H{<qIML}?=FQh8wtB4LH9E{JOA7&F77QE0=nQq<zjP7XugG*)^W+R?T
z3bP7g7wc1u>$sGa3R>ygNx58SR=5ryCJooKe3TS4eobb{OCr}7UtYouFEwj>nqJ4Q
zx3eISQO`QHJ7)eDzZSz4CIysbf}TI1ouGd;y7|%|)B8#ggUt=lo}Vy(8lrL;hRGGu
z+RMkm8rnk*VQX*r+j+S$`I>*f;oa=eliw|O6YYU#u8!nwF(i4JLjyYnQ?lsmac+IZ
zx9W$2Cn?P(9;E#ssG*2TS6yU;NREO%O6TrcR+$GIf)a^x>K_*C`jI${1x2bK=rgn`
zo7;Rvy`~={iS0)Zz%zuqhYeQvE*M@lr(?LLb#*+i;=YMIZyzjN5i-W&O+?p5MSYFm
zJ#ipwzhG3^-!7oC2!iW5iFl=ZVw=xtu9M7;*f>medHKDOaB<KJo)zscuTZ^93H6Rp
zkzRl8MjlWG&#WFp10kTA2_9;na^3|UL4OT0=->XDCsF?eyKJw^?;Ko{p!tj^oLcjj
zM7?ELnMNLM$lU~Ry$QUbbVB)}M$QWjTA_lila%@3@QDiDn=bAL`Wa-vFt0onBIpp3
zXT{$PodkU60`I)+*R?mGg~~xd*(T>M{dq$VUVB9VU8A*$-~xs&CsbItlPK{RAhs=0
z@wr#7wvo8w=8xm2+C81wtbBvONK!2!LMQx{!BJ?k9^wNfI?YF~8lQkPu+LVARB65A
z*roiDWOMzPYMgUPcYh?fBLU(TgdC`fB8TNrpu_W;4mJwSeV+My4@Y`cKZTbr+uB>3
zAndF{({ut+Wp(AuAxD?XKE7w`6%VaYSR;yVvA%6Q5wfhfOHLv|QU|ODI}jZlzlSdz
zyc6~N5Py)Mh!`#;4$}&V)1W-yuxb+%Qwvgoh&Mw7-yO0DqB~(O3Uv(J1Dfs{#?*;G
z*i;#u{soL7aMKow6FswByYa^u3n1p9%(&ZC5#$H63|SU@nJALeD;#wmMkHYkJ+IGn
zEJvI&n_dT)lB`bF|8#Cz$<w8zpOq1#S46SK@3ZzvZDlKp_g9D!#gpb*=;1ROt>1#(
z^dbV6`w|JtB6aU^0*m=hTm?jFcwNz|`lgUzq(VHEG&bEMN+3_Ndn1ZW8zeZ7@MjY)
zIc;8fUe1YE2jr@Wt4O72v%oz}!;B}`&f(ZTkG>>(`6XnXXnJt|EZD;@sD^w<ZTgA1
z2sc0Rh!(HPHu-YwsK~Ej`#)A9EJO%{A=4{n?V%&XwS_XynmoyG>{E{5WYHR^=`gkh
z$|rtvBXFUEDJi{Jo@*>Yi-ZwO<~{VpJ5LM-H5KxVT;g~+ILi}8*E$RaF!m9BVD1UB
z_lC!S)ZY8_h)+-bgL-Ge!wsjh=L4cOztyIU*W!vN-aZr37b^vwjmWL|+_dORX&<3y
zo8Q=wRFFhfd!U{!J`_*XzN##(kbnl%ysAwSBE@JUcWjhE(8QNbp)P2euSbIJHB!(|
zD986UKe0}&YD$R`EW-GSt)Y7BNppzX0>cV@A-TW5fFcFeQJ#-|R|vpyq(CoX3SZ1~
z@fKu!&O^A((?7#jguPZF*X)s^6QPUX{qYQlvF~V=Ro?H)j>p;fYB%9C{-Re_jnF-`
z&gd=|O8EFq5JA3Kz5BZr_(+h_cbANfjJ-Cyb7*7;Z{mLXz|S@coV(VV%U_5$U#BRg
z&(pp9d%*pE3ciRqY&Me;fMs2+{wCDUZs%wNOY9}>;@jm{pu9`0Y49s!n`ke@J69XA
zKWnPIPIzpN0o~2^vO7jTHepu8?X7=o^H-mh<k4wO*d&}pg#$aohT}L<pmb@)pxSF*
zzq{S?ulR{q5Za?EpW5<A6%7*!&wHg;DN|ii`g2v`NQzvXlb6M^lHZqD7G@A6f&KzG
za$u|6_9uok$9&%A>$A&9=rB*Iz;qGuiL;NgnYe}6q48glCTFKnbtD$th{SOjO0V;e
ztxNfdHi_(siB{Z+AhG!6wmEwc0ugubqo$w?*+o%An&4EvVeFI%8c(`37&=^mUO@67
z?bi{mNuduS--9!J#3Uxwu)@6rI*Svo{4t30rp<1-Du=UR>`;ifY@Y(FhCCV~AAb3v
zc-KPIS&Y^#BwKCy5`EByFK*}e!Vh@taVQ;w+D6>#fyu%>Rn6gOLY}`oZ4(H1G#g#q
z`|gS|r1*H;{wtDx(BV73lRPZh-;QP9+6~R*4ba2&u&IUP!r;{y-(E}6AVH%Z*x{y8
z*q_8*eE*D-C6;rrdx8bS%JF%tQFmacBY+`Ect0M3-I?g`3X%V8tl&MQn9hjr2a!M%
z>o^PDN?saHfG=frYQ5H9jOPT5LD5vraon$lF<YLT#Kq#UxijhbL|%Jtj;=mquuvNX
zBg~H}oe1sLCn;$uqt;PRPn$e?B(G9psGg;~dEGo&wi6QjWw<1uN@WQ9(+ddrCFP7|
z@foP47q*Rlo;tHU!6jfrPi-KnnCl@8tOt<GDx(Iuk#pEDT}s~(Kg<Wt&1v~TBsD(0
zz?#U`_cxax)OK<dvrhR&*K=wT1TRzOuywhh7;~juEkfriEQ4l!mG?g0^?bDg?IqeA
zV%Oh59aZv)WrN>Du)HqAtN8)HF6-yAY_vZA$A2(jXjgD$yd%zQ9Cd-UFkwP}9%){P
zXa+_z@&}xf?P<R*?<HQdNK)_}TdPuzwfNbw4w;|UeV3ss@<RQmifKB@r4uqd!ubH#
z)oOJNu1qa*(u*l;p0^baaXrnKvdB}X$W@L9o~Xt3hU4W*wk+fF^@M_b#==!nA7w8n
z4A9|`kG#I-l5fHRG$<OEL8)+LOZV!OWjsn^pE8%FHfVlHrSk_jL8;V!+!qb?FfaoX
ztxZaepv>6liaz-DpZi2<VZzPx4MU@H-e4oxJ#>2D4fv8}T{6cX)?*F8eZ+5nTeVu*
z6Ukxxo=pt0Z%WovRSvh7KDwDW(m-;T5+1x1$eY)sLL?ctp!y=`HjU4q*Vg0^mnXO<
ztGtMcodg%|HiU@?hc{h*W+)m_a+~f$UEOXXD5GCI@5f{1IwU4*ct7*Ze%W=*V=U;F
z4Ru<T&^i=NtN#|aUB@mmRD6fq)__4magD_QuT)F!;+BY71MB@|$XuFRF*_(Rl#E(U
z$Dw^Yu~F47z9m#@I2@OR`nPLUOgv)gyyU7qq2B84!Wv_QcsWfadKT^`tt~lNzl-pj
zW_)_1h$H_7%?qU&&V5uX!<pOY=dN}zXmB}v@D{Xv-}<tUA+95y7BTp23DQmO&Bv|Q
zrw6;Uj%gTlOd_*L@~rc@b+^qS%b1Yg@gI@1@6t!1&bVwC2D!yOfFt(#lS-MpgPM}*
z&pN>)qR)(5?^Lie-@8_dr*J2~sxvfa^1mK%oR8*rES*x|U``5$uViV-pJzeL7>eS<
zwSY8F68doNXjV$LZ09fLSY{3F%m*xRE=_@|hLx^vYCiE1mmmgP8O23{1-HNc*et}B
zv1RnSv`KJ$p4^r(i6!w&p9Mrh^RL;#DFelPOv+(h5$feJ8ILQU>vKC8BBU@Sg;lIc
zD;D*+?DR)J``0h>E7O;D^OUmVZTf_fz5bCrELqQ+U!^=-J$K5XzSq3hzP3D;`DZKZ
z)b?T{0Q*Y5t6={6{BQ|y+EP`;hr{sU5Ks68|5|jnCv`M{XZ9Kh6WEvO7<Vi-RD9uI
zeyXruZ|u)&B*m&Is}Ls8fBR83f6SJzO3qRw-jGq=d}?MKWmV>CvPI$HeB%T@Yw>%X
z{c<=uhMbFE^)$^Tw{M_tM1Zr6Sb9#nTA!G|UFm5As#h-M&yyU>o<QY!mIqhF&U|hs
zEu;05#9wN3_^vTf|ExaOq0X?>E!5P~z~56nV!OX%3HTw}WU4dm(ZhK3DKWg-O55IP
zn%2$iqS`was)?@VKQxmw30JH#s2hJ7<hC{YQ~;7->I`?bzEaZ>EIVaCYj+c02Z#2^
zS9Z4A-i94|3x+m@mK=M3!H^MC==nNb-;v8HduV+XJ=vi5k{P(gBUg`pjt@Ywre7<s
zCaMI=<l_Yzt}(=yexwu0LTfe07QT%jHprxp64FyUrK2el7K$Ag_6Wbr$c**2^iJmI
z8nTzKI(jhYCDYa;R8-kwPFY162pmGm2B9a+5bF;0bj5Rp{}A{Va~-rEwR@N!N!PIQ
ztJ`8<(dP!Ed~%rSfiV&5q-J6E3Fd7-kX_0fv9@d1-$UlBzwljx*HCYheQkbKT`GRK
z$em@4i<?z8q~zE5jgdQ<k~>{%qiA>J&;?{Su=ixoi}d9K3@<q`xUFy%!;UI?YVR7_
ziy5C|_NZmn6a+FJfygF<mEWDzzF`OEXYisE<~__U^zPY$<|F02@KHTP=hA2v>YHtr
z_}yiX&LkF$?=$g30lM_1Y)n5U$^S5>Fe38|gDP7`?uO@|lN<}-l3Z@)3c_AY9%?Mo
z;SQhH6Y85I^1K`c4p5Vbxx7?=F22>t$Y5I%G2p^A4x8$#9Kf8dlkBQKk#YSjTw+k!
za_XNTzbstpRO|04q<R)fO8VqtvVb+dJi}U?UFArdE#Kfn=f`)k5TATJ8uX@j`qjg}
z`=x+z_ySu=S)JS|VngAlTkRN*eWdkLZ9^Plzsp07zOr~l*a>aITxq;^wYuxEHYx5e
zuMuh=XGN0shQN6IOQmNjF00v#sQF>eL>TQ${)Ba%go=q4z6J%~&ds}m_4Tz-2_cUx
zIyEShrQ4X@<lM@Xuop3?ev*W=Z^-CgAJ)-x!vkb{vGCF8Q{K)7|GH~A>PyC}EGgHC
z=+Umy;oe%utrWghU3kBrylR%=nh$24WEq}Nu@my`7xW~L(I`uAe3Ocll(qC|4%WOR
z@(Z?iJsnlMJq`u6M(-EVmsA>rUvus#9~X~IOy6LkI|2gjd%ZZ*F(31gQrdT$;p|py
zeJT43N9T1{<{+jg?ff!)t~$K%&OdKD<j@18Kq9eH`jzv;Xm6yxFK*T{N@Na0z?I~{
z_Wl6A2P+Zf2o3atZgErbrQTaawO*BYA!KlK+~(}ZCErmCw#I1g4=D_&t)hEsmJXs(
zP;)w&J|;O`e4*NlxZF@GQ9BC@hWMFw6BLx8!Wb1-gY)+frOWMn>hqY4@p(<oT<r$G
z%=xh_z7sG!l}MQYM&T^J&-Vk)K~$i^f=r5b_>cKz1v+{W+C)ikKAzSWZ#$GTjLAMO
zPf-Bmy&yYblS*GxsAl<9Y-EL5n!Wxnfrs&h(`By_L;QZQ#=}ZD<(LZ|q2DiK3>pHR
zkdu1FW+fE`1~jjml#8Ga#zx!6_AhgT!v|h5?16qdqGdY;f2DMPAcxmmnN=K1tP$5f
z;?Y(X#h*V8aIqAxFeMNYmcEL^w4w4j=9-AiL3^}4y3G7-=9`GiYvV3X$u$-l6H_Ia
zA0<rTr*0h7F<~jUx0MF1XYxK<V=Mh&z~BiV`z4yMD6)IgcW3glPF~~s!p&tlyDw5w
z5y#zZ^PnFU$V6olDYHJO?)v2{b}TFH7Z#~9W|L~A^NUZHq*6Ydu$yY?L=lhJp>)Y^
zZz_9yc9rgH`z1M8IMn=;K#Msco2=<j&hzQV{JX7?#b!@|+tHms)c9B(T8tN3Uh;1H
z;io@9><fC?=adxaNnyk{cb2GG9^p76T}3&Z{BBqf+i@bIppc3~{P9uKBzX3Vu02bI
ziIc;QzQs9#r<Ab$tMmQub(ncssxm6e+KywcO`#zPZuF8rw)S>o{10~%Zh4;a?Pcrr
zDkW^9Cc-MnX^Tu6N=nMc7b9G7#9;})q4-SpzyGB8bVnD`Tc**XR&sPVT6mN)TC6N*
zUdB;DAsbCUsg^oHl;Fh6ub7myTT(WAoYK=L#bq@uN%0ymXM+sw^ay=v_#`f6>e7aQ
z$2+xiQ?ZmpR(awfntY~dqDJ{r<QZjQCgIZ1H_J9;t~&+uU2eAG^s>`ta)jmv=@@|f
z!)H_kRVCE5tO*aaGw4J@5t-p3WM@SET+h~R5<fb6k5VW>JYH5pH235R+n%q!9Leyw
zxbQ!}SuD-$B;8C(3prO<k1m<A%pF(Ivsmfm^_fUn8kV8oR3^vw@E&BcseE-}+tjFC
zsyd&!4;^+Kty#&hGHulCPVqjen^9_U++U(dZLu{E-C`7q&Gv3k3#V~1c?eBnse1bm
zN5*r`K6I7jl%He$t-ukVR(uaRHY*LA<!j$}h;<agjes;Thng7+8Qu5{4bQ~u(-dNe
zL4?GP{Gi=#8|2aTY2Ml?^nfWY9CT!pWXd0G**k{%jTM4w==8BPQMJ${7?SSi$(JiU
zlm8rU_XIUnGmlsF=`rT`A!1G(>a=1|;rK2;8zhOL57Hz--5y@WqZ5+ywf0L(32$Po
zw36%rJ4Mf#<vdvGZ{PluMvV4s)(w@+CCI#D?g9<W<HxQ<C@mIV-!>Y^G?pBFzbMPw
zwB#dP%bogJs7t*IsW65c*c>X*RI(43X}{K$$eZzOUTGlHV%cJtpv>lR-SbinlUsBf
zDdA8kzSoWw>15(Tug?X7T&bs)N=+ml*L0b41i0b?={yyzDx!&)<RVV(3|no5VY#*J
zGBNCfG?}G*A`Q{5sh_Aw^YL1%x~r)AkB#D_VX-AaFrpJ8pK5q4cDO2TzYlxFqWQYV
zG$q$lk|Bw?K3<Wrx}vzo9j3lG`Yd|u3>xOkSULh-#&X3n7amHtYSt+!%9rGd<h!Ye
zF#7$EMV7-Gt>Wv8{Q7`wVye?g_Cb)@X#Jl~RP=D;(O=FdBSo>3a0}AirzB`QZJnms
zTsm5Mwx^=P@pe$lH0_P5F8v>IGs##ZLM869*2aB=s^OCBNg9atpp9=*BoipGv#u*t
zK+7DcH!D~m9NKDAGUFlvi&|@G+IV<XVhI`9_|^JWSpm=fytMw(=wPDpey@d*hMG0u
z3`HT6>ULS+>LMtJlxKO2ew4|?L72i$Y)+@xCDjHI3UR|-a7I%SKOL@hg?n!;pV_EX
z^>MI<A#{)`1_Fl5|EjH|K`#!bzNWHnfC}))p&?_ZR@=LWWdBZpjXWkayhCKyW{27r
zux_~~FME)WW!qm}>f#z(K#Ry3M+}3L!p~+t$2amcT=$3?k69^fqgsbbPz!B_E#%|Z
zhKz|ZheCP~Pfe{@Z2X1u=0?Fz!b7sySS&x)-z;4eraP9%937JO`>Vk<)JJs%MtJFz
zk$IoeozsU><HIhr-A~Guym@w;7pQ{eAF@+7;rr%HVr42Gru$fw`5P!CS-yHvQwT_c
zeSTCbWl&z5Qdffi8X2oora&QL$t-v-4qG(>+IT@7w86e}pvY2&#;24oiLYw#Ers3B
zYMuFX&s&w(YJ5E$%LmI3ujcI2mN#c2p{4SpZ(o3lR8*atzW8_3cgx`^w!)y+nD~*)
z@PSC-3sD!b=4#rlmX*d%tD~bGvuu8<bhZx}sgZir5TdzQ!A}r#oU@$5jraK4*}t9+
zRhlYeAK<)N9JVo_PmRyuO`boZl~**4z%?eyhTpBhjHs1Dg?jOI3`%IGH)^QOF(jP%
zct>~=<Kzhk?I`0qlP@hI^h)2)V^}0q#J?s&yHw6U`45he#z4^Tc4go65o@CD)(O?_
zlm)PAYmK9N=+(n{HHRcwl9h0-2aego;@tfW6A<6&McEc*B`2rpqhe-QcSi{ywVKP_
zRAy;Bx|(I=Q`}Z<K0e%OC(qnOlHcdi<X6<W!_#Hln%**lKDmg-=gAX<uv!wPRV!m^
zZW!J;i)kp{j>}E^t^r|SI`8f$igQOUHaNT(K#DjPs<UljVIlT;$4km3q*6zuZj1__
zVhu-Wj^z)^^R?tr=F-aML@tf@6Ib-FO*-++6d5k@f(mSg>>Xc<h=k<fcznW<_-hNt
z>#Cp^h|_KKSvT`dR&@1j^WcqjoGF<mhm8@LG&N3{cy!D^dHX7a&P6QD-Ny`3w}#u9
zF5II}9fXHj{Lf8xlNLb*qY2qBBDaHk!@n|1WxMReetDw8<wET%eJz=cPf|&zVU_+S
zb!Wa1D9L^5EPuve#2PB8X2ZGE<Yr_<Gt;LP)lEgqdQMEavFfAQcIT8!lt+1!!l*_-
zI_IGpy1>|f8n;~JNZO`t66Iusvu*6z@19nums752W%9bN>exYJ|5Je|Yw2-ggI!*I
zPcAX?Ql>b(T-D8n6C1nIzOKzSp1r{>_tz&W?aX``QQdsrqKM;*b*;k$W*srh4Cv&o
zws0R7Vxv619UDHR`(456f#+k@UB_<~hP6Uy;MTt<dmlY)MxQQa$N$dHrk*rTlUkca
z!8xzos_nN;<>&=rk*?xXt8J1Vt2G|%{!($34rd-Q)y%Ju&zn?nHI46caS-iR{<wTy
z4JW@lXd=tfGmwtEVXh?EbXZcl12F0>t+B8`#!w9d5HKCQt-JVFHW>vq*5SW$jf~GX
zyG^_4Ayo=vrOutCyydA&*C|7M4tFAOyf$@IBByl9#vp5?KIDmpm}1aWDn|OMB`XSs
zp5ZfS3#`z3l{Ai+S*eetcJy9zd$;T0v}!3gU`jK-e8YY4OZVx~Xc^$jagRJ?*5F^J
zh)>U;6ses|RO}V!puxaY%#Kf_JDBUxh>34ni&qEOyVUChBqj&q9;3TfOZrR;VINh?
znXz_TSiUGHVrG+&))ovsw&h=ryxy^;3X!$8WQQ?+=$PS|yIVeqnM`SlEM{?99ut8y
zaZkcZG1z9Rno7``Ul)LMY@-Z-bXsZ5xa9k5W!+psP|BOm>rumI-pW{(QOK{dO^XM&
z8jv7%gGE-2qs>yd&nG+nn22!3uP}g&kYzI5@X5Nh(zx}}?W>8sGoh*{3t0t)mXc9|
zaaXD@G|T+&2z96<FX`y=nK4;6`qxs;929;(!`YOb(nq`-$<w9y!!{>gBsNEXI*Z8)
zFUi`D@}!M}y`Hd?5zf%edH2;P$Imd15j0VOQ+(udkIr}4AxMlE*j`wBFHz%hzj-P@
zQKdkz#2dRgB%!CwfhOlSWm^3$7ItC^$(#e9xTUL$NFU|m5OL_9hn7>E`=^~CZXbc3
zO9b~ZAA<^EqQ5roGsmV;0Sch4M4>Uh*y^`7Ftg!KP62AdWIv*QZOfe^y<YY!g%G+)
zee$P~%bhSr^H3szo?;uf;({;jPk&lX4mX!r3ig<Wmn!dVR8fAB3vIKP{j8aWiTLGh
z<DB;V^<a&K%kKx=-9wGfq$L^~UqMNwB2&SQfK-;EUeb`QhyC4*<JU{r__$?}2AXFh
z$qC_yzl#w&$90-4RK(Jumh<`JP}##Gt#9<w{3|r|%r`aeqcHJl92#=+0?S=LS(KkF
zS|(jCJC~77)*ivR3O3xijJljxZh9Nc5V+T=VCxn67iYR^*bZ28Vh(VHFREYU<K^9a
zi5N##YIq8@YP=9xj#TKTuPc`hdUyW9rr_Qq!Nn++Az2aMNB>@SGnb!LdthsBJuZ%k
zM^+z?3n!|k7i?m{^J%fxi=$nk<xgo%79c}*ZWXP?uGeWYnNH7OmrmaGp^HGneBzGs
zRVuu|7j56+!AKbjgS$GUxjx8ZzifTDv*ak2!mc3bI-+>ZzLuT0Rv_@e13|z{%rbwQ
z`~&D1guGFEnDz^pG<xRya-SX<7PTL#1~92>>Q)xl7?J1+5m7_xeWiSEI9iLVa1f<R
zcP|=Y2W6SjO(#vx+qWM<lxX#4g1GN$e($BNw*h5AVZFC`rOfTg?E7aA<#9>UPAUi`
zPna^6+pb^g&@+TN=u$lgA_b$zc?sk#3_{6g8I<yjemZ)3FOe0pRYGFqkt`@h%gb6J
zS@#4yhfXwzq%9U=nX^JQUp~orIUa261|e1!yOAhI16b<<n*?7U>+e!$e>SQP=id#;
zk%6&zDa5r>F>&i_o<BZ+3%-YJcIiPT6A22TJ%rmCM`RpP9wkVT=V&J##-YPask|3P
z)Csly;hcXRg-zjtm{eo%<RvB~Avu(IfcVFFra+d$XWxW$f-^R4Bve?l=8ANw&MJb7
zkd|-f?>OuWMJev|r<qu-CF!q6zht{u13xc3|8dYykSu16G}}qYJR4RXeIvH3^ieD+
zp)pZ#kp~z%5x#cIRzfSkTRvl>2aRrjh&qU)r@RyBeb!R{2|rTqC*(<X9Zx*mL1e#T
z*P4Uvvn|4rpDo`m{4tk3I!~@vv)p^`ItXft$SZT=YAgW4uMBtNHh9lk+#{NcJI69r
zCYno)Vn9A=y;>x%>}$+w<GG&8#$R?%sSFpFhO<_esK_Tw0b}%F&0hHwFH@Z(YEb40
zg=_q)9>+G_!|%H6B5`Ip{@ELYtfgNB(lJO|0Nq@}t6KU-E<y(xe1H)>I6NGVnt{O<
z5B}jqamSHJI-5F+K8rc4zCwz*k=kW0;)rgc-g?7RGo^Fc`EC_e|2n-pP^HMlVYo)@
zQ<YXp>a38Z4HCs7OsgwoW5h<ASUASsk2SBhXDRhLL@|;0N2$*3_32W!rLL&&xb|>&
zw>prU*1Fg&8y(IE%B#}Q^GLDGE~Li7bM^<ybK}N{BV%K;EDarG*q;q@mgIdTaSV8<
zU7MUoW#=^l>9F++mm)^kdE~4)Yo}S4LjM-u8JS&RVyp7WOdNAaKP6I=($<<7TL_H*
zni)6V0m}-4>hKDGorjvNxtz{-vtYzl6Zk?g7oEIiqWQ7ZVIK0$k)Aqin1z8@yXKOW
zEcdgn*c?I<GX?~CJ;%E7=U+OVvohF}y=26cdZGK(1R)O4x2!Jb8rYxsq&_-YMLB}2
zz=ec`jLV(51zwLGF{C}klSvzj8OxENp`llHE1+TIX6a6=Ps4i-ZQ_L7?%zZg<B`FV
zBMx{8w#p49@78*Gm1tp#FCZmZ>MjJN!nn&24&<4P=EU>I|Hvjmt2;%-(2Unjp*AuO
zh6b)|UQ4Co@?0Hz2$D-Q@daQz=M1^xEn~;bk}YR4$&qj`<|PY^h<uu!J9*~30qd;v
zGrFICrDO4kd7jF#EkH+xb10Hf^x=^oj1;c7drI5tgtfIFup?>_(y@+&4i}vU8B2H*
zVpG2uySPLTh^t7&lw`q8POH@8+#}G>vNdR5XZm*(j8v=Aj7NiBoqC_WoQd1<(%A0Z
zANqG_NeT+xo#=GNH@t9}@o%`N>5velj~Mx!+jQHmlr@Y0Y%i{(A!$2wXkxfl_jr)2
z3U@gc6qFwqU|VQOiHeGvg2^bE0v2sEeb2;W>0I#l*0kT9r?nex*)`9w1T5g?g(W4!
zj#nBd`<^|{tZA`ULknx7!BzCiVFjEcWv%*I=R?=5UFPWhoV3>J)P<b5>@Ut)jSGdX
zK%s3;>zO#4Uimt)K1#JEX{}x%o+w!fj=ICm>84UGC#h$g9{fFTi>Ts~nw>&o1_*5C
z%tOR^@fcJB`sx((H;5>5{tN!);)`G0zYQP|(&zikI_0D}So_>QCzxdzKugdaWJF9!
zbCCu=u1h2OM<+<4QH(S#R}VY}`iCiu4>okW4#g4|obY+wN;DU|)Isq0Y>8x|1Tv~Q
zYfm&B9bcXQ>?K#_!;wL$6L%$dXU%&v$X#6G{{eR+&#9^uo%UHZ9QxBkBC2=;*`p9g
zzEKRHTIezMR|=Rmx><2JwZj^?kaET;4i0X$P^8%-nTt%bhhbx<>%3x?9C<${#v)uj
zM?Aenwu-bh)}7R1@$}m!+vvN4O!7hbbd%MM^N`Qqb1CJRpKD%lV&~*P77pOPYVkXs
z3feGkrG8d^YdO@JISG7Rn|8Vnb6m~C78BBHaciHYQ=&@7G1!(qL`;l)by4$_{4}~!
zY~L(%v-xa!=qA#E00|b191M|YkkzYz-W!WM73e<3VZBHM80N8^nL16}jEHf-`15u&
zpODr*KKyA(-EC*Bj5tPWR6cWFW-gX<`5b=@anPK;;%sR6)58vH<`^%yKidKBNMKzL
zGsT!5JGwC#>*|5&ltrJp;SvQa())i}0O_vByz}v%?M}<zAb7Q(q)NwY=cj@+I;WOJ
z8A{4Z=AU#*@3_Q(4hvH#g!0f^+Ffeclfk2^I_E#Z(z|i@r{KpRiHUDGL#`J?PR3ei
z)j~%b1^2IJjU6jkLH;mcP{e%1^!pxoUx)^TCxXWPu_Hf6$42;vC;dM(ods7^@7uOv
zX6WwjE+u|Q*C35FN;5P_mkiC2(w&kbf^<lCmw+JM-Ho*1v;Dv8{RFJV%<O&N*L9x9
z0lCDul@yMaAR83`N$li!^6P!J6<(^_TXOd|ycRDZK{H!4P<l!&Ji}f`e;j|-E0~=f
z4su`vjZCt6`F~efZ*YunVW^z*d6zxcb8mz;ci5peZb64h0ut^i*r{qY9(g6lZuYx(
z0e9_y4BBy(RTAC6jkhi`l5%BK8T$x(?Bjbtzo4o01jaXFkIjGm^7K-f?J)eme0kp}
zp7mG#e$aO!;|X?C;5tLZQlX$+NY;IP92%HA2h-aBjBnu!R3~3~1Mc7-%A4nzkLjO2
z*>k7<s~;M<4YRnddP{yZQ(2JlUyiPL|IMh$rQo7glPq)MxZS~SWz{>e7x8GbLcivB
zM43G15I0aUHhWjkl%uCaVU7dALO9^62W5V4D@5WAe`MMPl?5ehCe(g#`2B8mO%fXv
zGtrhjJt*E6QIqXY?hY#2ym^<`SmiI$Ro5>tut+8D{>HmJLtSNbq^P>&wen3mO#;hj
zX8pPUD21!@D1S1lGsm~l)}Kk7rc@9#iA&F^!R2`rmPo#OL<CGt12V}8ONvGRpx?^-
z43<>pELNn-7BrBlYy(ErjYDhH77-K~L-lGNPzee2DHZE2-C#M=sic?NfbH-~Di_6!
zWLseMtL(j3Gjt=1k|<@r*7$sL>m(i}=J{+grRVHN|EtHDjq9{8ZYz@BYq=GhsM&kf
zAOEH4b+g@F<973ovyzEw;zv~cT4^02rNn2Jl`o%Y2bn@XPw5{dx20mPEb)MSX;_MH
zq*(W@+WtO2UjE8^p5~Ukjga~yqb7dQCp*IA(Drax=GKcNGz64-m;gs+UX2m07JDci
zL|W@FAY=Btdf~X%u>$9k1Vk8_Qh^=+3}0iUwouzyt|=L6@>~&vwDgn2TX(7J6i7ST
z2f>}`(A{WZ0GKDRHTm7{SJv-CeK2uuOMf)kDBmC?@b5}IngwSrMhbEY7+d)V%CSQJ
z_Lioldyow_3&Q3_Mp9S%kKwUob7O%h-yhEQP24l}2_LqPr~0iP4@7Atw=BLGd0wsM
zD<0evtm2XW+PJEUbsq6ulKk(0bnoES8~F~3ss1l7P*PNbja=$9x59Rv$*7MKsg<HL
z&A9Q+SnNp98nu@pq;BW)ejRpY<M;16hg=>qOgZ?i*QeG~2j)t9UN%L|-lM8>@tQ0W
z<WE4oF8zNlP@wuXBGhq#(<yfWH}*Lw`X>h+`NyvWDq!i4XPu@%g;2F~2Ydo~g=S=a
zcfKLNj$|-!w$kcFRTMM)tv<>wzoGr<WCD>_GP!r@=)QAh%w;eQS9+6K&auoXm$fS*
z@=7M$o-i`t{l~*{or-7+;68vB%TQo6+Tgep<I4531^Rs0rNIxBF!kGg>MyWyX+)<@
zYl0rnZ#I-(^SQ)@#6;(k0TB_#-5~#gV*#=fiy2Iw(#p!zQ#<&AV7F<GwL)phv_EjN
z3t>fPXw?m)Rmahhi#R@O6L62H=DMV0QA13mJvCn@qE?;gp2;98tqpYzSq@qZ++P|u
z*=3`Oc%{mDRSU<{h|O}pm`$3_PxL}t>QpG_()?zxr87529<TiwHKvgTQ75POPe9(i
z#F)=tV=iN%^?R?tcJw*4^>oI&Of|I|MjHqO+O~o0V@fL5JIV}WZU1$VzoVwlj6y3O
z$$$10q>=z#xF8qFDInzjp}fydo{je%4}Cs{LSP$!fZ~T1>|zSMxn2-kb)_ECnBPab
zw%p(|(XUn-ID*C$;9sE}jfppJkh~$SmMJZ*abs1KimhmVjianBpb)0@rZ4rI7?Njf
zEUvK<eR!NOc(cy{+7%mgU662vA@Aww8EXQf3VF>PoSd=7ZZrH%dfwl$ndJotkNIa`
zT3>Yi&7WP9PdV>|U%q@fk&UZUUh!y_$M^UH<xuG9C^wDK7c1&oqAU%PZ_S}ZSY%H$
zIjDF|?X3*W7h4>@!8@2$zw_{f<{$s~nG)9_b>HH1WGF^9vYSMsW&DmusikBgBc{L&
zdZv_=`6cn4<LjFa&wk%`l~7U+b2R&}6oH}AxPALTbrwfS#%0zOXoqIxyN~|`QvQeH
z#J|NrC_lq?M$@VMu7FOSZe71`&^j{paA9RX!sPwEg3^#S(`0x30oc?cm$j-%VNRcZ
z$++MJv;J2OzXAh6UdJ(0j+bm-zt!^TYzmIOw_!3iTF~$iNC}Ol5~8#~<um7DfY=eE
z9NJ-iu$1sGXb53M6ckBB_A$g5$1w3|5*qjhxL2n3;Q@`WRG^jkA1kNKzPys{QNX;-
z4@q~#Lrw<%%hs!#S6^Q#XqIU0bQKT0e{j_$Ph#W^TF@p0t!O%T4kBLkARfYfILfE2
zb~v%N5j*NblExb2sJQNtT#5zr&h17;SU=1Xf3thwRx86}p@K<*a<25yF17)r7<UzP
z_*9eY)^-?~Ny62Hl+-q72CurrwT?iySX~v86{Uoa$Ivw#MGOtir*w{du}~Sx-iLYh
z=wWrQq`wt5UqJhSeWvm9wWDL0f#6Y*bUc~vN&4zx;^|>1OM1B{P}F=sQ*;Bko!K$U
zUGabXdgD9ZFZ1dBcLNmV;4cW)`qwb$(&zC2>i)QGDjwZ)(c=f*#EXNvhj)Jm?Y<T&
zG64aO?_!(RgIm%WBD^+0jbSuN;Xik1FzOJ=p`J(Aj}Q4d@59KsmtLu-S-(ls&O&d)
zai0_Q_e_278MY}%zC?#i#E5@+zvqud_Q3UPTC)7(Kk4-j(4BVFQj!)1IGh>(F}PKV
zVeBB>{DS-wjr$A7ClD#%XrZ^$hm5=o8Yu<~#lJJzqdi>58I1+ljOUEWr`?-UO3l06
z6BDxw68b`1((y}=zsps>PUd(?{+mFUkXE?OwGUq}BjNG$w}^*e4vbUxtOp#OFhR<5
z>4be`s{)qSok4+%|IM;<`Zdw2^u1^vgbd)1Ojb(FrLpa2=?UERTuA#Bh$nC!nq&Rp
z!;i0izG7evk#v9Rv!6sAq4sA0r<`Ct-j6kKnEP#;gd22xdw@!e*QG9KlL5YJh|u?#
zfZHF(W~;3lNA^{2xk*1wvTqhIgn!CV-<=K25u0nVn9>F&9&2M?Mi}mu1eJiRda$|8
zM6w%wzZEm<g@@XyWc_*>a1o7c-^&vB2_Rid3B9desI&Pr{L>%gxsD9u|De7;UJEoc
zKX=QC-J(3kb-4okSv0Px-9gEx-`d{)OwwtFPIj7x6%w7zv5ds}(K~=y`bBZJThs+%
zy_ZzhlC#(S@4Nq!L9IH2pe+&hRfe#K$T>NVU)nxg*sh^{^7pfeu6au{0Za*~bw7zx
zu*Yc_=AqZ;*ai$U!+7+m(`0H;yV3FSS5cr5&_rIWrnIfKkqZ$>IJiD}w#w9byk7vr
zakXuMSwQ&tZY$PBU4mSYmwEmH0FBy*zK7`NM=8+xdlV)?OZDiLzK1n}85e?Ejc<B|
z4(0W%mkY(QVhHn3Sld{;8cN74P)~Y%2PCz`mOAjIdL&1XDj>UImjoa2$0?jOMYDoK
zQA!I1A6tu9Rc`X1OOH<kMnJnQU-}@!hcnt|DwdV0>l_#r&0DL@v}mHpXqYNjR{38Q
zjmp1?UX6Yj-3}O3G||u9wyXBd+)O$;^h?{j%(_%_5zHxz=5Fu+v=b=D{{i6RtGvG%
z7}3MzP<q!Yiy)1z?a?nkKHDNbE*}D;eRPkz>L+w+h>m&&ZvkKLh*{Vod@zYcMOAq=
zjS~a-3`EjzKdt>4_)UU{@V#tZdm7hcCLmv`qZ>PL&#!G4ME&M#RQlB@M*A$eiWsVr
z>O-h3*h_F1(KkQd+GUo1bkj(;!IyzTO-?`&gP&dVdLl<`t}9Rxf1Pzi53%51F4$S)
z$7FOBS4h5I6I30wW#&S(iriSO1_B4y5hM}Jo_K#TLG7F%Mb2(0=kGfsQWNgh>4bHI
zn7%MoTWl!ER0m52iwBFM#%BuHw6rp@v%^uzJ8eJJB>I1kCfsH{+@e&f4hn9t9-}Pe
z=FQc52DI?6Fx5(GD2)Uxm<)>vx1Aa*0*g!d6Zo3~4TXpr9wb;89Lj%lUJB!9QV%Kr
z@(8zkvb?yt7xg|#S0}MwE`EejB=H$t3`i8|@C_|RH5`7Hebf4$=?asn@tt1#AN>)B
z_<z0Q<o=Hi?5gt+jE+B-qnEvO&p%1+mbLJ%QM3suu_ErmNb&;zKCV#}ZKFxuAtL=@
zpxYrSw2v3kwwRFQ^T4;3r?E4)ProF}w&!I8tTt}cJsi3yX;>mLzVTahsp_Y7%6I(7
zTOCZ5`EI7x#Oz67ALliC89Sl0l?LItsGeTlBXt}4vpDvt8_)e4DNXC3B|WI-MREE3
z);Ld%5<7+w+6Z>AaGF;cGpySN+bs?mVgaBAwo)9IIEmRc@Xt`5x`sAxzR+iYEzGg&
zqA-FbPqg!{e~3Je+?w_J>9>^E_BtP3$RxTpZb`1)Q|+@UlhTIXsO5KG*Y)rpsBv+Y
zO<!p{@l(qkJd)DU6_04w4%D+`oI!*iwX?G4rgGOLSPCSXSH>L~G!jnn?dJdqycwzt
z>AUwBVVlZcEgRr^0%Zxr_(}OAhIs$O<kYlJMf<5{DR#<i8Wjqn1t0apu|8AlW|Eb*
zUoC|&+W{pdMh<VZlD+q@**$^#4NSb(T^}>Vm2!sCFtQ>S=`_{zho?>NHEzuM9~b93
ztlKE?Bz`9>18@qNmFt4(VPRh9Rq;$oe;+39?iU(XR}&T(J6Y<$8GL|~fb<`LSl|L2
z=(72RfVCt<t_*<g+ps>s+PY+W^g^U%q{?UNYihuzib~Pa5nTuc6ox32Q(HWA;$%SF
zikmp#HCwJ9fv$;GD(V0>0+pR9vE_Ko*C@`*e691(nzZ#iC2Y3b?1$0|a1G%q4ksz3
zORzo(#=Z|5SY{rSXT@gB7T>0*PIC{uPQpxi92QVmfaQV?T_sze5A6>mL(Nl&iV1^1
zPdsk3mfy6kse!V3#kxftF3{<l+#U6{HB**NRQK-LK{tyFpSi_IFppw4o?Mf3OM(PM
z;`vtfQTp^(pH>>o8mSc$<oEGX_r)xT{O#0P3RKCIx!83SGb$0Vxz{c?&2P!fxrE$9
z)y))j!x)$(-e@H~4K|BRIL4nvJi-?Xb*o}OY;@gr+)7+&<#PwwSo1ak{QB7Hk!_nu
zWYl+@cLr>Hk%i4**Xl}-J=Z<2<#TN1w$Sd9Z$7~}OSywhxXVo&`Abf%W9V--O3GL~
zu@7ez*dCSZ^{JVI(}zd2OUu5e*@Y*I(l~5+raHy<j3V1V;prR!$aM>PWT2J7pP83O
z=d!saAuo-~#^;&~wB|2QoCBA*Kin?Z<e`*5cm92<y6~wE2)P1dcPXJO73T%Ch+j`@
zTMH50Q`lsiy$x*Gc?Li!nf!0zFsYtf0AWlIVDRG*Z}qz)ZWH>qtC22WjO9ugva07g
zxHT<3WE(z2%3YCl-i2iY1P|6ADwJM2J||<3NjZ7x7C=N0J9U|J<~R5og?%1i2-y#)
zl<l_zyq~ZBfu16_2;TwFQug=y@vEgRUq6rX>~fRPny0TLpa1A4X}73+*te;n#An6H
zW)<+1N*iZFR42H{pZI?;5rFbQ|8r-+WTQv`ag?B6pf4g)v~1ePp5m_a`tiMOa+J}}
z@G%=6w{YW%Btjh^?&bGz&_3H+pGsE%AoNvb_Puie<US-n_UBm&%kiHx(w>TIx}q>R
z`raTd)|8GOE^;(fzMFSFUdYU@PghAsGtyzLW|SR6a)&cRYb*MfbisOdI|@R=u+n<*
zcX2WEcMe~%%1eOZp=;v3Ljo*bcRxA5;=s!^8iv=}PBPXBC!ega<@uRcBjP1)GFGe_
z0N%=13f1adLAj<4P5zc(s$6@NQyHiGO6kHWRFxdy5HFs<IO+f9;)3$7=mLEQjsLnS
z-nY|m99dk^x$~)E4HG}0s`XTK>j0dm(wxYo@W6vLPKFHW7FZ5l57LNrk*2qxHy4*i
z@(Sc-6vkK<07+TkhY}(#LqNe{bd32_$jc%oFxSDJ;k%UNWQ&29Bbzw>fbinS{duJ1
zDG(S6xGpnwy~Bq}%g%QK94ee+WVsN&mwIhr&=c_LaH1z?SNZqg9)m08UAzeT6ad^)
zaqZaV@gT|r29<bL?1bM`8Nw$VAEk3mACt-_k8R4_B~wY3q|!*Tc7Q@?yrtYYqqo)5
z(dF(qidNClH)@R6_??}5YX)WBfAN3AUsdoP?_veNit=mPU_{^i9qUu`yO3QEbQ%&d
z{<PuqhhUZeVWKtMtxs$St%__g+w)Ou?X-|P)C+G1Q=NS-o^kyKS8oRrkA21t6~zpv
z!E?xQY!374^PGG)@T-9}_tUcRy;g;?(UoE<iiFMFX?p}aY^;pQuSlXv^A1j-Daddq
ztD3;Q8ES=Bd+JYBJ|x*8iw)#~M9`q@UBNcSACIvh80UN8#V^Fsu%%VAY%kJ4GZyh;
zjnDaDZ2q&(za8B+V#cMA6e}B>PlyzLX<YIQ{0{WDwS9sXV=$Ou8@%S?3oIuT-Rbii
zRLylfJ(D4>iFJOPaq7A}Yyf#PW}(DV5@io5U1u$S^+txROZ~E&@Z`tm?ylbU`rAk(
zzGAG|71B<wqp~7ta7x52twku};UR>}w-z0+pnj`=9YR7Bhopy)dj7s6$-U4=!^>ed
z?3Pq-%_l;XaeG`hL5Y}NdwM;OoyIb(@2t{cQLt@{9_}a)$4%|HOsygZqnv@xu+Qjv
zUiC<>g4W{!C{iE8%+m&5AaAF4W0%sF2ttFs`cR--Hb(uNSUBMmdiLvk&|4X4;ZTjR
zP2ga+`>$F0skb=iLsqruD=uB|LAA^tJ4*rK2<{onZC!i$mx7yEx8_5UHnYkXQ}5CO
z=k4Z2s670Tl7{vuj2%we9qC4l1*+0bkYlhTbf=WO@aad5y4tOm>qEOn`F0hEA)l;(
zUWZ}%Rhul2F<$@dmUHy_xcd*I!%)h5uBB<oHvE6%w|eOxt{*o3^X;ICIG%0K^Ge~1
zy<Nn7{L!CJNbj4w!jGDeja;-$xC7=8P_ycvHJhk2Y+Y;LY4WA(yg=eR6KLe$6}3V$
zv}tPiG4s%=t7E#BA7xfnYI{8l+zH;3%3GtnGja_JItt5(IT`JO1uf-b+?1A8@(*S<
zRdqbP7vKv3%;DCA@=I*V-M5{8{j~?4yP+R<f+EQg!Ra_{FPz{Yx{4VT#HQie!&|0~
zOnz+W|Gr2|Ja1Kaf9WYzEAsQ|Bh#5})sDVM2Iy2jiHiejdFiRMJ!kdhOQ5!}E*tvZ
z9>q%UeHGJ(qz!Nss;8R-fO*dw6I_OG#mZ+CP(x{5sikE!_^Z3<zx*3}4j4zX3@E7*
zqmzDDXg3n_G~5&e5sNnl6Bz~=#^Xdg1J`{2uM(&_i$!_U{r>%jk+l))SAdB|h(r8z
zh6Le7;*{954;OhXsTx|Qp;KvoQJI#{hcSpJs`iS^IXyc#c8aUthN9O0x}jnZX#xq+
zJ!j+|tJ#txczIm?j9i@|^>m=V>&-lxkhAQd4?LUSt=79F>+0os?@}=FLR7GZ9?ej^
zY^1oE2b<$><M+r}KGlSKrMhli&zU+Kc~YWl(QoQEDia51iWy?QRi#2YDDOS(y>v#Q
zhk)195JPs?^}HCqc*lHq#GI2al*QBb6m<xn8*WfKF`@ZKONQUy<j$X$8VNmbO3StU
z2oU!qkHqiGv1k<(H=7)^wcZo)-C*w(Uqc_dS*RK8so4?nxG5I&)Jh+FYx#(o66HN6
zc{Uxd>D)khn+@2PVM{hnJ0TG$RwMee{*q^aa!(w>h(edM;|K`9sF$O%T3^oEN_%@5
zS`5ZnH8R}tS<-tMH<UL@WP@qS3#nHboYmpN()ffXNa%>E`>CM&e$j{`DtnP&_O$;X
zUTtXjX)YhB*VQl~7%pjMpXFtw69#&M25`%Ja1~Q<igIB;kx?`LM*$nug*Eg*xTc`!
z?9#*<_(5#vGMZ4=?m8Go0Tw#sgnl2t3b`Acg4gzPU78NibTTC{-7)<>-)O<NTn?p|
z6X9lsNouNQPZN6tw`k&PP@ebjUbqBmI5n9Xoh7SF+O6ZEtItjyIP~7<%!i~4bry@4
zrd0U8RRP@WEoXP1wqKey$<hj6yd6IMDq|m|P^sCZS;I+08<qz~9~FO`)jD-RW~T?C
zo?%b0-e&u&LA%#8VQyGL8#2-y^85D}@qyI%_HSCnuiz5PbPs2&p%^Kqq>130yResj
z-EHaYPvpep_R#j+?hrJp{e(IkY}99Q?^#Zza^y6>$C_}64UOY<j=eQD4BG=M*>T3&
z{g7wMlORjw@y%}GZtl$eh``rEd|t2fY%z?{hU?NR$)$VjehNGYZ(+knxqQ3^9-Bae
ztCfX7!+0sgO8dl`cXm8z{l)0B^rObW6U`R2%IVhW+G&SyJervN(|B!zg!54zIG3$k
z%5>55EFucoQN5=KSDvf@f=tm9{L$sj2cl3X?m641Q)wlx{U!$F@#iJ+6Z9cHt%AHi
z+ic^Eu^{bgpR8b9*v23zx`@OoTv~~kXc4j&i*Ht-gFcTRg&5=a`1M9e5K128M&HK!
zO`xj@gA40*IY@6`n)KZwQ(Cuy5&HK&EVFMM`;a+M;>JJ^%lW+C5vh|iguXISvb82}
z*K&~Vyzfcy$xcZ(F=kv`S&fd@_*KfbZ49)%@3n%;-)$4OjXSKJ3LbZC=5O)^sIHVh
z_ch@nK?(6Y(T`zs<4Srhj`Uxy0zk>KGW0tKLY?<-_BhNQ<01&CB`3TZFcrF4C&W&h
z3iVGmC93!E&+XPQ+8EE7bb7FYK(aS1#9nO>^v&rl)?j5#$;$Q{gTm#IxdD7>2hc#`
z4gCO(xL=Gq`^Buq60M`J<M&)J+#_;=p#dU`*ZHJ>;r&&E`J5bCw?4N_!>8+ROq{Be
zFj8>2aC{H}dE;zgO84t?oB-(~ft-QPJA1XuFNj%}m)LtIo$Hbut7pBf);6)}xsefu
z#l1g&Fz5pQlo|9)c97l(hUh?hRDLW8Q5m{S#G4`&PberG2Y3%Vw8-x91+IZ-ei3^h
zb%T*9kC>sj=TuZbKF$!4?f)Z^4Y20)h>7K7C{*X-S8D79)M@{PjX|QWfvLhWI|bY>
z*Y~tpeq%hXw-mC=hcAed({%(Ad6cjih>_nXr$v2`mZu~}U8j0`=hmoS%*DB9d-IvW
z?y0D$RFBl*Nn_XXw|7*wm4zUwnND6Lzo31#wa{>KcNPxEfT2M5@aMOI3NFmJC{-X#
z?x;d3l_p2@Dl$wEt#Gi*QWU;J^<^UpLrNj@d8M8l`aX3oAIvW;1G|E+`_#gogtQh1
zf#wCJ^Paf4S#Z1YTa+Of9Nni5<Ym2M!Un1J|7fslyrC9jy*n!7ITDv2U^b7?a9Rg%
zDSXjT_KS+U9s6XHwB`421tM#B8<IgvtVZk!x{6e=i0d*RZwodLdWWEYf^GTd2N05V
zwGR6IBEE=O+ygQ>s~1BXNL$@Q8;Mt|jVS`a3sY-6Y@;nN@IgVpN~E*^nNr;6?8i}$
zhW#&H-gSCQMi__$1`W&T%5?Y(JP^peL}(+@5{!+)$dOSe(~^ip*C*U2R6aI>5YhQB
z9-KBbKTMLWqw=tLp++x6O=JUGB7)0mTTr2kMUJLHkdXdoTpLxW_D>978c69wuRtMr
z5S_vus2ImD(j&!*^Jhj#dgEBCx|>vLTkxCH-iSY_QQt*v3hvuh+?#&<B*~nh6;|PI
zh39WFsM<!*p%byA!?*5m`|!+=cSz(igOtiw(VNjxVfUK&pJbogVBO=+DjOk4YT|G;
z2I@1GyMVpASV9+oR7E%SCIz%4l#c<Y{dm&l5_H>(xtqN$Nq5GKC2DrC^VC}hI}-gu
z@N@z2*B%p$;d}dav2WJ%Nzm_4ov`Bb!t$HqkzHE=2AUy_t#a$e7rOXX#gU`ZEM2Sj
z$B9MEdb{g#>%9V9Xefxsv8%~NB=cOeJ$Hwt-wi;kH4<pG{*>|Bnq|QorX=~1-yxwe
zM5iNTtA51uld<|F0ms~&pDq73QGu2Z)EKst0Uf*R^+rDiG<xkAoYv-l7bXI%(`udN
z1qYIVA4X8Vb$L`*adq1w1{E^)=oY}ZRR<SBH3l)`rFSPhk59?wwdirs1)F>>RM&mv
zMXG<|ANjlFVwu)FGx{MBb-NvZ`I=9Z9bb&Jx&p8&<-5$yI8(qz>+m;2zkAxDKQHGZ
zHnX(ukM<@;wcgl$1pTC5Y<Dhf)5>NtBu?@eyK`2IhtnCAi+IAVLfnJ)LOcmHP|wvE
z6hV*R#R#`(5WLnRv1YxV#)tSL{n;n?CxomD^`EeysVT2NFrMyCkuXPP#`0E_55!c(
zQ_>=jv{x&H6Kw(+lc%>{l7kv;cMzsxki}^8<MHvQ)(Y5QNmYIDmlJK_#`a|X9{m*p
zO|ps5$7W&17t5_fOu7I5v&ivXRp}D%FExL8RQ+bS6w}2|GACTDXCpNa;>G(jL<r@a
z4+wSGgG5Qo=nLtqe7q8N(pl$UFCj3X=YCc#8Fw3RagZ4IFOWtHB3`#bdLMf35ScH-
zAEWmY<M9A%&T4q!D~VA#OWHhY3QKuG7iAMEci1dfmn3|Lm6#~FLunHI;?uf|;Y&|Q
zh}utu6d``O1(*sk3o$8o3HNv?2-$T7R<-io@voEq_Z@J5>#LKI4|$``#L*;6d&8yA
zzjlPf7`%xCa9z}Ea$9-!D;zP-=+b`orb65<XejM_ubka;8XJty{Qh5O4o<!AL;Z28
zi^3W&elZg4#T{rw>J%{dJ#ErI|4+bdEDh>WwLT(y1c|2Z^aIy?ZTujlyVp)+OQ=V%
zC%khjTfq-B5{A=u`EFbvZ||-UIJ&sx7)94PH{j;utp7O((H^dGvha<4evr%2v*IyA
ztWBf5?lXyZ9BgNq1$TVjccB$`<78=Y67uNL$BP!ZC)!@;n4`3{`XSPqHsXJUhpHlW
z<S>4#QtYSFktES7QXE=z98&g|>8!$cX8L;BJ6^$q-m{>m<%RDF+LUZaZ=UeZX|+*c
z6PH~HPg-~s4q>Xe!Dn-=UgP$6rlBtDHP=<cDrLE8?t}iumE`yR=m)xk9u}jyZU95v
zB>Yxs5`q)KIY{%1scNQj1R-iX3fns6kBpB_V`xumUAFSc4gJto<P~^hSs47&SX+IS
zZ)*`(8^6h+(G8}*F(DVndqUcilYkUB_-$f<ZNB9x)qNKD?lpGjgVCo>I~J}q&6g>%
zgm+(@Rhga-VmPli#wJ%^qb-UG7~gHxwK@e}=V#wB38_D&UfcjYlM)^s);?wbd&+NP
z{>qyX%O!9+UBXg9VM)nRi-dozM}j_NY=u4vBnD5)6vmQhlXr2_xE`gRgw>eIG~v_>
z{}M^0XX9=Yr>56*_wX&ludUxW+IJcSDw8{gGjoa_O{64n;ncQZTqHRro|mp{T|aRn
zFzTx(!wtTTc`F{?zWkM9++rZ__|2)TsKX;wXV3Of<LNZ<-=C(LK;EWo+pyGN#Lr#B
zR}N7ChWJ_FoPp$^@HeDnQy&OlW2TX|36neX_q^tR@*Z=)?DnDaV-(>q4iMZb=4YR&
z+m^eOxO&+ASKM=1a%{}(vLXhz=}ONl=8ewF`)HV(4@u8zy)RjL=^3WN{OX9&lhq@+
z<PQ5=)iNW_&ZjR02HM05o>>K*Mcn7(SHk#jap$LtrL}uWhE$Ea6e{*8`h)x}X&|*x
zbL1}@6^x_7*FN!FoSY<t#Ycqw1B%LST79c~YJA%HRz&&3@9V~AB=nm~ffathKRik#
z!C8sIabi6Xo_#<70IBV_VdOaKR6jLWE*2#~!d>RJ7Z(+SxomwNQ5yx?E1EIXmTin^
zV&63WU8!Xs)KPh6*%|6c7Mc{F4#Ed$+@L{pn(};#=~>xQDt}-&9#wPP?=)LR>=uc?
z4g0DvC{i(tTImGUwXY~pqUpVMRnYI)K5XH9KbMA`Q-)f6lj!8iQR#Ba6fMxVSMwG`
zE5-fiRWAW1xJP_IIVR;ze+kMaWv*-rG8PYtb)V!m<>D1_sC4#G_G6+|w2E?33A{5U
zO?O}Jj!?OJe{o&LpGC$3hbdMxj99vwy=+621VTO>t=vFX+a@fq^cpjnPW!Ce{!lJx
z)1&&S|99SqrbO+S^I7~LqG7YEyc~o7=s04r)Te5d)GcVbp<Je?g%&Zu6+EdFuUu-R
z^-!yP1u(FkNtx~6)*W(uJbBdprXQI1ANHHXrss4OohfurdXMysQw1g2nO6krO17jb
zt>zo0EtJDNYfLIP6{)uk6Ui$mpLFjk(sjam)aGUuI{A)0hHZCy?2W_J;Wt1u3pTZ%
z<T+k}UX6-|Rl#G_q#++eKFRnRTw~WahAaTxkiJ+CF$?-WG$hQ!N9d92`!lsN<gvxV
zCinOyk9`c9oLvwn=aHrFxDy3`NhfV|WXb3$kP8MDL%@%jn+(B_n3hC^uViqW+){g@
zLK%V7Z|n05sJY9t?n8Bf{PMZ5%eL#fbKv8r^?RdcH!DRfZ1Wn3i8pYxAC3?B%|z6=
z?))YOmhDh7lDu!Cttwww;-nx%>R7_@wm5ii%)xSq-~nZ*cX?%WXQqNK1KZp;%8UKg
zYDCCdl}Q*LkPhDR#K|LiP}OqoNd$X$2mb}Uqo3$Ehm+ZaporyojLajAO+J&)s_Q70
zB%Vi(F6p<oOV(9j>!F^)@(I~Z{I`Crg><>t>PqK>&Et(#EiD7^2p^BLua~8wUXiks
zkA}Lsg=B>bOxJLI?=TN_dkIYX?BxSu#OV691Fi3Ee<3+Uboa%QN$20X9b@m)y1>lc
z&xPjO4+rl?C)?#4-(7vBAw_;1a}M}b>DbUv^B2gqTmrL%#)d5GoOrikN-XY{0nCvv
z;+}-Frlf5>=D)Y-{?0e$<P5f-(V}H-B5}oo`9Px6%Opekda>8S%h8vDGpvIa{1Yin
z-9prF|C4<538Rb^ap-K^8rqaLmd)W_uG8YQ9|fxR`f)UVZPv3&&X!PMwf>?sqs;Y8
z;cc1+Bmm+*i&t9%idDV1Lcs%{%V;?`FE>r9N3OSil+CAU!?e7pTR7Kvdwjbt2v-mT
zqru~+bdcvw4PKyxwNSPVFf)IH&T|PEPx*eaYxzB7G}ERHCxjz=d%Fxb+X05+;08Jl
zlhD1!vBctZiPibBG8-cvf!YC~XQ!%{a8sxD>$i1;uD<mPiy!oYVv(<a6orp{875rP
z9&dfNo?}<zOThi8=|~EPc8uUhqBZyyPNtxJ+J)K*u%@v_LSuRWq@t7DZsGNVKntqC
z4GB4q98AdbJv4}oZ@m4y2g?nBO`}zS?*7rwslLZt(L>R>q>{s%3I{ZZ-HA*7-9N3n
zrEUH~pZ+15M>&eKr_5<17S&Wn%?NJN4O)t>!;acaFetQgjb4<wr+-VWQywSx&pn<a
zSrXznE>Q$FAFNojiG|F(Qm}5Xs@H1ENV&ns$HzhCfYq>s+oVnJ!^xtj;-Nzzn{5wY
zi2R;j<M-1kro!Us=K1U*3>?CwB&yp%YLRj3K<=a8kAy^<R&%EksL+9aB$l$-J0f30
zqOsD|M6%?c9O4+Z`~T?Sc>35_tXroRa>gki74y@*exwGz@cv%I_3j(mMxSIdRiLk_
zXc`y})@#Fl8O`z*CV&6>+hUDHM0e^cFa+D>J;&eAb|2@H`LeFno$IshO0R>Pu<8M1
z+$*$Z3D%Xy&-V+_2GcZ_oJ*-^mD_CBOS|>FAc-qG<6>w3E#w$1J_E1|u`L3JgL_Bs
zRbcCF>9)3X@r7sUz1qs8<|Mz~+!04r<xkiGKUmPm*kXUz(a%ts6U)l&uZ1OSBr!P*
z4@{C9w9y+u_mvlH$4aEH-S;Pm4OGNSeo|4IKK5uo##SsU-;gTw-*dyG*UW&>>C)Rn
zm1xTH@o~Q&=2y+A<{J0X5LW?bCJir?s1G)!T4x|zRLMTZK6@nhFPg;~gw-w^JoM>J
z$$Ilq<qC;NLOx73h(G?5y|Dn{vqmaO`+=qpS%`K&|0Q-nmo@!)JQgW8uXpBJz@sC$
zjzMum2^~*QF6b|Bvp#Y#&mK&_CKSJKktD6UPa|Ar?oD4p!p6=&3u_y22?@CBu^2=4
zc`b4a!+K7BcFBIeaML%aF%K&@X<vMHXp<b@c-WoDbB-&(=XFg^r^Kqb#$K<0+T8&&
zXH}6lYyjZ@32--1y#&b=C!UuJn&jWIY>EV>K1Z86JRLSpBO<Wk-wH?GM%+fO+pd2w
zGc0Dv&N<%pjHMRQ`nQu9F40U(eS?XxQfv$uD5I>PnB*pCGBVL*Z&kTsk>9AzA|0hY
zoHB%HYY{Bn=d+NFQbQNDxVBMC3%`Gr(IAFC9ha3RPc4SaS3&cAmMo-dk@H*#+^@1l
z+<#EN$$^t5BGYf9tY<lhi7^b`9HyGJHJsXa7h)an^Ro09SAzrSs3}DAFLqxYe28cb
z6l&~nNevY8suBNvkVje8dYN%Hl9@%psAYRFRfKAKKT{Q;t0sED><wH4QS6)RePApU
zfMXpa&N5-^_Fcy%iG_qsuwiFFdo>FSQ+M*eUEDC|b^DqB1p0wZwre;AIl&YIF;Xy(
zCVO^<vdM2i079{mjr7tS`JcJLeQ(^6`xXFxQmNoR!8XKUQtU)chnlZ#KRq>ls5$mK
zP%$NhDSvkd0PR$Ii@@T~`4=835X{I0hnCU(Sx2*$az({I5o2?Luo9y`p!x&r93c}3
zDi4H%UI}=0+2W$jRv5AYVK1ZP;`6x6+!|5g9UkiOT-y-ln$kDf!jq@Cqc=cUvtAsK
zGh9w<t7odZwi#BifWGq)!j=)doUjZ4C{ILbCS+76W`RwMqooGAB^QUEGptAh<Hev2
zBwE-7&g<`EzUsy?#&6nN(K*S%vz5laLmBIpbX8~;#A)hNb4e{puBX5RlFx$6FoxH1
znd+58h6qjDJ%E14>uv-aciVP1$GDE`Zd^!Sm_1?Bzjs8@uI(PWm~>659D&Lx#l`Ji
z4@9lXN?>0<<V=?+gh$a26fRG{zghM=rRk#%OP_etrJw1$JW<tZ<Q@Q)TWsr8)3PZ1
zV(Rvt`2+pUN7txkaU*oxfd&TIaB@pUclZ>=Xe}-DCO@FIDqqoxFmLY_3yTuJ?^hK!
zE%VnVTE*wgubKC7Jq~+UMz`a!PMwc8i?lmg_OQh}F1!>1yF++=ZvYnNvU}{g(_;u@
zP>=MpO{f*bnL83dkBGyFLEMMvBIsM|+%@3@@Zm4=FTBvZX)5w+RpanywHMuOEV?n(
z2S^t!L4iPCD0Y4Q`?L1kzX5H&6uW@x(94c}Lz_(?X<LwKOYwyuO;{~Y7URRN!3A9U
z-^!7|AR7~OK?Mlt+iGSR|C1pzazS9gdM(0Swh{*syKX`MMWOq4?*qCAA!YFlKL~)$
zcw7pI*2cRI`&3(3_a;iWykvH1zX}HeTQ$8!V_AJuZr59{7tQ3NN&EpD2OlcFMK}^!
z<m`WBIAH6tlll(R-`uD+lmQ7j4$hL|k4-Hd^M}A<X_zgKoBZWRAx5OsmXO3SF-VAD
z{+$aRIbBd-j`wcnSLK1vfI?VTP3+`90k!HbHcV_j<^+jxJxo_)Bf98aGgUJ`84<)s
zQwEE1U})moOE+L3Ugs!->6WlU;%l{)J5l6pAeH%_ZJMPA9Xk=~cuE09FLN<0S%+4J
z%e}B#kfh{CTYtNGhipMV)d}nR<i^e2q<j=7I-K}J$~Tt3`QVimzoE`%lU>7Ii$=7>
zwJsOUsGJhyN-Pw_7W?e<25X!<?wO>gg=O6q(ggoV2N6S%QBV&_RC?}*%)gHQrwwAg
z&M64Q)8o*7qU?5wrmWS^pbmmE`Yv_g5ZkXUDCJzVEcG`(`22b}efS;fgp+10*5$il
z=d`DGxjj@dj16)w((ag3o%h_kp)A|`h4HU0Xx}2q46((FC6k1v1p<x@s&x9!3w{k$
zay%y}w<kUbDFnS59VX_~vpCU=O$f(xpce&4vZP>1;wQ7GiIzt+RFqY9Eb9Fcjw3@Z
zW9w0|$FQB8ud#@QhPrG3AH9nU5?7{_OwVj=Pn7f%AS3wZ+HVzV6Zl$UdX6*{I&s(K
zOZrixmf&WEF7S-*oITW2+#Ua1_G1n5X3lOC$@hhbcgV;=^Yp8afbk<^1&~6qryGCB
zyg$#m*aEa7>3L7bHUV|Rux-5#zh*uc`V6UyKB7xtBleV_$E<yUkb>h^Bz1X7KBE`V
z#dCRX3Dyr_8Im$YJo8mDESmy)ih!e@`R<IUdI|<?CbU};bY|q`n8I@$2eH*OV1i#T
zQGe}uxlvj4_3;4X;PR>PMS^eqWAh)6o_<VO%6!@|+XcP06*%#+;D;Ta9!Kp1m5Rg|
z5fT{^_0&%p+TOXx0y~ncNo$jb;}l7hg(jioy2OV?ss##a(D^~dy=RC|o~F^oaZSH{
zznFsrb6fh4i8e7Kx--rl`j|L?uXq;Jd9-(LQ@&2|)hWe3pwOr$p65hgF3L65Yb38{
zqfH(gs}1ir*H5uSZaSqk@3)H11{yl%v$A@1PIZ0_J|X3L^(K;YXXS|+8hYj<(=+$8
zY9CXc6|k2-l3|`CBN8h*BVa(bJ>onzVkD0Qi<~f-_SD2Eu)ww?LWjC1zTbCt<ZLEH
zqsneOAIk&QZac)mknLSPBuW#sHj?=g%boP_&{q=M<o?cC+0Ef@SxrxVoeG?9s3bg3
zly{Vpj^9((MZeE{_fZb1mV+X5de!HM$HH~5V}qiagZ2qyQ4cxen5Vq^X3)DThZpOV
zY4*)&o#Ur23pNJT`R=jiPL?&zl8iP0qJjSc_dlzrMpu3~5qWV?&cAJAz$b-BuO6*G
zr0=`cSb6Co>_>pBL#K-p-|0RSD*d&(gkKWrT1EsZIAm{qBJjCRlM?QQ;HFoU?vq2h
zM-1uo2}2n4=u;9HM&fqh#?=i>T}@^R?99@r2z2P|dnK{cazQvc!AEX=cCrzqk*MTz
z@hP^@K)_YD!E6b&XbW!2gTK_E%<j_*XG3-lmH7KP`USt71}Q@*w|OP~gPB>jwL-@8
zx*U52NBS?c9|@~U6#UMQ;+c&HJULznDmj(L6ce}0YHiE+yP8p9bIT_!!M-WaM<ad9
zy{+rnA<J|243^Se8AMOs-Wd~jkg`EzXZB7_VxK;R(bW-D4lOwbo)=#){#!ME%oaMG
ze#tK3q)11L<nr8h6r*k=02Q^D<vkO4c{RRhLxF<0G!i>2GnQo}U;Zc>dD8VULJMy*
z$e&ah1B!&Q@N_Erc={*KZ^*YRtwY4@$GBVK4UKN44w*Wa&Q8#D@!J8nn`CT-8M%Ig
zuk%Ll8mEchL@vJ3nz8-Ns1c$4njOP7NzbktSKs~vd$w8Ez!=rXj4%J%F-#e?Np?;9
zb^cU4l_<9dvZhw0;$F*gZo}t4%O~zhP@oA{;!oRTw7EPCowbU!+K-`SW!75Oys#eg
zO$?T}kl|4b<j(?a+oWX^mXWtX7<2%o(SP}`Y792F-XuQKG9(2V1-bj99)w=isbEm#
z8-D>~8~uO!J4jrRg>d?O<lnyxOlk&1+6fUUXn_f!;3S(L85^gyx2<5|>dt|^9<~hr
zaRQA6WMezeWYs9WC`9TV6L%o6R<WCsg|S($*0~T(NGgHw6$!S%1i>9*-Vmz`VFQ`D
z+~|DQRqs6td$Cne=pSG?GuY*dKWpl9z;eGU)agA0-=_Y=dkNDC!e^!5%P=2w4}?;R
z@6e!qwv2>x5AX=p4ADS|ws4;yhv@N7-VPl%^;HxF_*G1HcFxT^q)f+4qB&1!lE}^N
zis-5d*Hc+^XwUN&;{gJ!$RXHlGv64`T~}=`Ka8{0`V98S%ceVm<qE$1r@s2<4~n<g
zAxS2$ZFu(d8M(*{`_Fyybi}HT_m#DB`MciGJG`)Sa{9X#46J6spyv4co5nXeQ9bKl
zU?c5y)N7e!nhlPBP}aO}GgbIlKO=Qx?V1J{LZzMKHZ4dvA1g1HF2inn^&1_Tex~t(
z`=c+uaX72zi;9{{)u7$qqiAU#QH^FDf|Vg7pbJc(x|dLHvXN}~>Al#`3*2{vO&74y
z?#u_ObFim)%9E-pOo8IOyZHf&AG3(oCUYXXVZ@nixr``dg6D5Pz*;6@n)pB}a_sBr
zAIph)wF9O}XxzRJ_RR^?N)nhE+cpG)SQxXFLBSGGOg$4wT3Kbw$inxvGcpr+K~0Lp
z@1L@$3V&98&nUuzFlGqh{-i#^_;)7|RSKis=OMKRI1M_C^(VQTT4h%{I{B`}$}u|%
zI!;$YQ=T4gwFVN`I<4U-R}ZNxkscw%PsR8nX8R&b!d<ZtQZB>IBspPN^><H^iK~vr
zmuQ%Rb(kqB`$XRt8?VnBP%UvwXrUdQk<{X%r1q>oV99jDNIi)x!)r5U;_FZSqRtf&
z&m#pvtrbTpTa<qYT%=dIw93teVSi+P1|HW8VmlzIOS4<zV8l?py24&#7$6YRH)Bq@
zW;|}P!oh?S{=qo^-YqQG)=l!x$ZUXNAnP9~F?OubL9eF?*|ojFDZtPoVu@?K1_4H+
zJi`LwDB8z(LKRbbAqwxF4Wn+)-q#*Jgd6C0IR3*7uELw*PXxaXyNlW*=lVmoV4+&D
zcKwuVGM({k;$sQSfAr2i2I3O50{Oj$|K63VVt;*(F=ZD&6~-Fj%M;x*=Av@-kCD?m
z(RAVazUVEcw-!uOcQVgorTHr13CaS;?L1JacIBUbSOgfkjrO-DsdBjQm_NOzD`E^(
zM_5|_{<&9SQ3w)k@Z2E=vxhVb2!G{3eYGb{%80t+!cI>m_}E*uw^sO4GxDIvuI2oP
z*KeYayX>o1Y5jqz(I)qC;th9+vc#d5o)@%Io2vIGN}WkZ@gn;@YnrDaG$^G4(@$-o
zZR_}NC4oiIb`ws{VD_~>7cvpUDl)_~)D^^$36wXEYX>%Gv*NFDSg>Wft?^}eFBRba
z_|z}je`{&4;=XS%MkRnSz*JRdxlINlt}*CAtgaD+NRjpNd#HHCBgn4<55R*5Wap5f
z*?2jyRidcof9ICcO?EY347z)de?NPWqrKXDoSOg2`jAgWGn>f!-w&~mynbVXb9WW7
zHY5eXF@`WliZYF<oY2jf%P}!>r_Gi1c1G470|2Lws14~ld!8>hJH0ATPrt>vV>lDy
zQ|gLZ$5NJMe;rO2>NxN+lq^`@j_8ODUq*-kPwoft6Qm_X`gw>co*MPZHE(X(Ifhj_
z8UHPJXfH=K@A%Y<TdaM8RV*&Qg%%2~<~{0SYCT76;c2XxWE<po(N%nLEVw-fk(z<=
zcl{Au_RWG>!%Esy-2m~mg4{KV5eAKr6Fm7Yw*=h`u>@o8wpnjn&(OnImt3-AG1^cZ
zmik(S)KKg&YJ$FFW{B@D-|b*tafl76Gg|n6DsM$L$tP$rK~WM{-=xqhuZ#>844wF@
z&&L<a?Hm{0HXS;}5fYb-xl9w3oNC;9HL)ST<DdQ3V2O$>irrD+?b=xOc%6QACU{?H
zcB*tJaaZ6&u>9S84b|Iz52W96rR_iIN2+Pa%F1dofS18)x-a!cZC_yc^sR?&$ouGv
z!<O>2`(_nG>OaeR1pzWY(Vi_1y3y!T0GZd4p5*c%yDTb`J%d?(7gdmpEH~2ZOD@}`
zhm0nQgzR}~ekopdl-l4Lx+;PwsyzmFB+u#{CF3ai?ub25cbC2tsT@dx;%D`9E1K;C
zMnI<I7RR~~jCMSd`V2DHf32s8G^=!YWkXUscl3{5-eyBOAXK_HbYu5D){wyhr@Mq1
z6KUd<ImKjEh1g(Kq)bP6G#};?bIY*Y=L@Si*r=qk=Nu8`Whvu@L14PJDJYMtOEvv*
zQolpa-ge0gBjs)z2@}mDTOh7->aUS1fZCi7A>V&@Dqmq$nHL~ub>5i&azQp}0>oUq
z#P$x-CAFx8ZDq5C?;owI@QSd*dN_r<B6)p~{&rmjgZkh&t`j*F`3`}Gbrp69tK|FU
zx9yn5nHgvqUKv<?am^S<ViVl1y!h~dqn%nSjBTsuT3V*o&U!<K(?|QmBCpRqq5q-0
z|4|?cO~&k~^Uu{^#XFj`p<Qe7EISZzXT8nx&VDk&3YTNMF!_vLz<r@=7`C9#y1eEc
zaemG)dXbBIc@x%ZTqT(ZyfNV9Pj;;vZ6Xzp7U@t#0N-D{*zW$i@VpoA?5TrUAfJRt
z)c!L~4GG?)Zf$iHQzZ#r02hzyQ!<m<lNQ^FTXx{%zxq7!TEEpj(Lt-69bx{hMk#{y
z{q9Q>CozmA?UT^Mvy7YsNBstDa_msV5Qm?3IC8IAC}P~KGwkVnWkI14+T|J;Dj26u
z@5YIDD(Mzi=WnP`{op|1ymUhi*D%9Ri#vmsr84pI*IdMG6}c0pSr8+$n&cWkzF>5z
zrmLd7dupKP>S~Z%nN}*bWP6Rog1pgd!elkXIUu>$8byEdA2ox=mYpjlx$qnzE!dDv
zoK=F;MX~&X-NY}Lz~XakGp0_QgECxTAKcUGIFZICk5E-Ms_Hj$|KabCo=>h*5@GL=
z)^FGwVT17fMcJEnlHQ1u17^4X@@gknm_e2u><3qC>RhI}HHz@er8mX4-S{eu1I@@H
zC_ZWIuwF&szLMTNr<7(FoYh0(M3SYNU=O{FN}M6=6v4B?xbLEinHr{o@5Ubo+@=9R
z)j!s8))(5FMDVKKijVh4g^e#kDWNbS1X>6(9fMScZ!}q&TU*<BB#Yik0qs(m)9fc=
zbIXfQzru4Fg;bkfEC0ueK-!Mi5gD3PRd5=a68|+=rsXmabTGvQneU6~NRCa?sEVEw
zZ@H1vg?5})PCcHu=Bv2pYf0EGVZF+-S5+EMs0<C1M?!fO{0@;p0W!^ja%`pl;#3+@
zBZi3Mm2VJmBI`4R5A~o0=M3IONiU}L%^wp#+l@Xgtw|OqY#irFq2z6;S%mPY)2QVx
zK2<EdvdTyQ5_8wOD?JDOyVzrIS>+3h!VL9iQ}T4IUzyo{^@3H+Vh2POlU1IdOuQ}v
z)ufj}HDM|HqF<0Y>cSOqEW6RmKhf-H=F*cuE<OA8-kPXKf{wm3Mxw1ajhFRvihGRx
z&LTyb1I6auc?pB_{T7OH_%C1?2S`_)#PdF6=OT#{jBl~eq%9A^UuIi5HV-Y!D=E8~
zYb*%rb#63XHDkSP;EDHaVq^H(e{pX7l^ni2-Q*2qv~PNsO|n1b<nWaL!OBge?r1vo
zHtUt?FGc^ON)j5pN%OVko2>Px6c?cveqHFlPhU-G)mdh7q7)z5yk3;P-N^5ppz;WA
zYdA1YzSs}BF1iV9`u{9|s$qR_d~#w-bu6$-jU&c2k~kY5fakd@<SgyHpyDFO1XVM2
zfCIY?Fh0=iQ&p4jHtMZqQY+FeF2@*JvI0lpy%y>-#{1xEQE*-@$cp(kwPWm5OioRQ
zrJ$MEGnBh5w1^sos35Ni4EYP|p<D&88jpm%U-{4*meoR5i68nkD#C9$&&^T&18p5<
zumq}9_TWiCt7p2`-w#*6yv67F4!zRqPvkaV8MI5u)}g&NpUavl{-{F;NAId2>>3=6
zl2@rFTPznSqJGIt#Av`#DfzbT^{>91Gmm(|XeKnwcfbyfO6or3m}A-eCLgJMxmhpG
zA=6#a^CC;iCux8p`?D+RFN4hcjnA5ULm_ByojcrXDjh^0UnxZ?34f6|ix@;y4k7+J
zeqw6<FBfg(m{0Eo^<`KkJaaee92dV$J%P&(!RG)&?>>yJ0cZReZZWS`0Q@3)YGTRe
zq!pmmzl~lr=bF0uf4|()LKUI(zgNqY39v9<b=BUSw%n*2dq!K$7#n+f8n>PrOd3=5
z10%&?OmxLFiP(fyU;efGwGG^ty{i05tivhp0hw(=XDbh#&er42#Wk)azClZJg`o>{
zp;8XwjFv)wNa%ms$Kt~YA+m@}Du`bO`NE=ZqtKsudeb;9%d^wSIVP}r8|ym=VqCh5
zB@6=pCF_XG#%Ku+Z|7t*SVtpy&-=!OnNsXSc@%aq`7%Cbf;8G9nnk&t@#>@nm`$<O
zV#oB;HlOseEfbC!t-8Q`_WqA;jDgV;Q7IlfYR7#q)~D}hhDGg7d37Rxt^S;`v$EG3
zpD0_&bfps9aa)kI*KMr5|2Qw#;<znXDQ;98gY$_3og0SzVm&3ih=5E6iJCscb?IjQ
z1-~MN{QoiamO*hfU9>PVz~F<s2TgEycL@;O2{veOcMT2+!QI{6B>{rFYe;a{p!p8Z
zeZRW*4^&On!0A5STh?BCZBDRB2)6qJ*e2+_2LgsjgXzX0_$KZl=nACzircUvxML@R
z3RUw$d@I#0-v(B6ZJ9dJLl!0$fpsmp{enl)q~HSS=y>jskg2n)buA_AH@R?dp?^k=
z@v*;o-<rV<ZpO|kawSokU4p$o{{0gd%Vg83xhdM@UBKq&iv80|gO!|yNQ^y)qSb)E
zsOvn30_jPPYbujEMiU6I9x_|svQYV~XD<r17h@lpW8+->``J7O<pxhJ!E?Lneg`&}
z(E@3{aHNW3e=Iq#%eVVkHkPq`DFqNoJ>buoggTt-A=kCZWZS$m->wdayA9Rl#N;$e
zHxuUxy1|rCNF`2?I9nhDc25wD&k$d0FeT8aDb^a{2f-j)x9Jl~euh)|ND<QTg<kDF
z5C0*WxA*1x^`0B+blg}|=Ex(1a9{IEh_0EF(`ExysnaSn>==!-2y;;?x3l|8K~Pmu
zScmA&XH&0aaU;%C8G(2|r%({Ahg7e|*YS)G{&T8r5yZ|Z3C7>~xY0lS@xMeq8Q9C$
zbR2T~#@@jTH~bOJA3`#@yu6&upwgV#yzWy>#_64ci=8VjHY~byOpVGwO(F~ca@Rn(
z#U&0qApsV=TQZ9;3)P1ZZhc3i%42r@m%eAcnHZZH%N9OZpVK#1D9=S*ZLFVeqfI7C
zVeI{}|Ku7@Ht%m75#yA~M8`RSf7s>HLxUDo91YzMq#vp<vKSC-kO5iHOghR)4NLxU
zJ!ugYXB=;E*jb0nrrb_DI)~A1luIi6mXFfo8_FXGTf2-dxHlfDg=KLz?whhN>sDVs
zEva(MYbTWP#1p^0Q#X9D#pSxzyu65ajz}9OM1YMKhv}s0M-DNGi#yXgPdVomC)AcQ
zJw&OJps^aN_{jOt4v6SlZPRkS8MYv+H~caXIu}9M$t;Z`24*oo`(DwNq?@Q1^b+KH
z_uGf>a$x_KXD{iY1+y!4!rb&vm0|R^gg}|M{Nk&-1EK;&e%4ZueFf0gwWw`xo9jv7
zvLSyA%fTZ+q*KeFlv@j>VB(Yhn1U*5AUEs=z~D1L8`Tu8*;kd-K$yomEczeU7wB|3
z30#-x%K3xolD|W~TXe(vR&=Asx*L7ti2Lo4%ax&EGs^fG$4A0;gN7^4jUxYfsv<DU
zs^ERF`^Sa5Kl|R`z?9J3K(<OYaisT=u|%O1gH6qPnT>A6=t9Xe^(YvXA0|x8E9RFV
zrSuF8316@~%FOgvPfNX77~i<gnP=GY(!L8KC%Nm`jdK0j6{Oa+|1;8K>SgBL1C<p`
z)#&}VONDwq87JwK7+HYzu@V)CX*(PoQUHaq?N>PzNmsEjB}I$moxc1ts#zVXRv=TR
zS_-I;*>d<DJ5zy}IM`?uIlt6$6WG+sTX@a7f&U_^iN&{Rc<d{qM5T#$Z_^C3+JWdD
z<tTIVHRp7igPuN49=BgCAGGDB(_GIbP4!D|&L-24c;b|&rj<Yr2lTq$KI9CfyPn-*
zN=)It;Rcx@Us*B}ikZNKMAo)Uz3{JE^-Vbc{G&w;gF};=FPQ<;2Xj_usUr(UKDP>^
z6Mt-JuD{p9lU+EjJT_orVf?TX(*7Wz_?*5XK+ATmqP)xAp4_02f{L6~$EOoHaX3_e
zs3LMelBVQo$h$>4&mqycdqLnGQikU=mnlSNtpahx4LT1rQDGw7fjY&%dlfecdc{q@
z$(;bL8V&rirF~fBc-uLNO6b!cav6mmj~nchQ=+SbVUR{j-&%Z{W3$luo#QDa*<6<X
z?BFY5V3C{bgzK0;#p}+xtD5PSWR$6b_bu#@oG16TcKPI2E(0;)dIo~mLZ~N|cteB%
zGT2D~5J<)W!M=GX6$dhh%EW=ds1nr?FroZlq@}t=-NvSH*j?GIA0!KPl&S;$7QO~t
zs0{w3O${>Ec*MlL18F*nnxi7#-*EAkAE3Hqk>0@K6M9#%B`UddDc5p0V4=#w+APse
z$Yoi-<xWQUhL-?~Ben(1G*{hDOMEvjppa2@vGaKNC&jo}^yMPn6<}sf1DYXA+9NlF
z%KBz^3$`y84a1@@QnPKjFY)y|&gr9l&V}U(?t5qXb(scLsR+Rg3&&)`>w|*3Zk>u7
za<Zc?rLDzD?G4L`yjWa_f*z+D#nQl>bFFD?`*8}vDabD@YE610Nf!F|JKT2T>si%7
zJ)j)q@gO=dBAYOW(d?bT;paUjYN42r6OG!z>huQo2vu__Lj5o<Rrn0rhx+1myaPTP
zL$3C@IgP{h4&Exgrt*Q~fdq4OQ?)P<thg9t-wuR9AyTOn)BSiV7y$z9{7uk7kE8+y
zaXNf6nelu-aY7SV>iJrzRdBZG(Re)Tyx+~Z20*N)@;q9_WCoe~26wg-bc~KN9PgL{
zH#<uWoiF0+I}bBN$25&4`!IMKze{cAD=p4>;07>pM9rl=LVE-IeIYN>a@{<@Xl>bY
z`7{iv+#q|OzZX1TSs!W6iK;KcyTW{zk??%}9@FK1{)2cp?28K<1RE-+h<6+v7$jy3
z8BUKe*?@fHj+k)=fn50&9S{f3dq$-R<eZn7GuZmP<bPg2<r+zF;J$f>O(kAHE&IAs
zB~b6yzz~eCu719BH2ySD6I?4?0yraz=N3Axj;3Fan7hRJpynPj{qny5vfNj!ua~WM
zq1MdvO&%0|)FwG6hu1*;CJU$$?|m6~0_t!TgNQEQstX+_Zg55{1B<@hkJYC#l$hj<
z=zi%-o7@?|wr1J`NfZiyqF$;VP$AoutSS=mdw^VWQR$^6*n`fWVR4xMy1KskquIby
zTDY;j=)j_{8nOjXChI>Iucd9aF+}@KZ?GV#z*)ishDZEeW;=0P=n|ZaKQOstL>_Zm
zIF4MnJd<w)9^Y;f5tF=<IxN%su<XQI;SdEbIFNFR4rD^t=(HpA01%MK$}wSKl}4lq
z49yuaJC=dPXTAVLlUaZV@NO|fP@Dt+nDaV2g&u$pYmSQ&V_yIb2W?Rqd(DQBzAiw#
zNKol$buClq4>{>f&;2e5UYE^Ar<WoAB@a7Sczy_<6Qis#$06$0*91f7RAL-ibxLYZ
zFJ#jcf^?X3nmDM}))bn^Kr1Zr_D4Yde?*>#k8g>^83Ms16DSlpVv`gXtSyDcA8@b0
z_5o_T$kTbbx?#wkvWVuT77QatWOs|9#eFq{DNlaNxIdO_1tkgst(fQpg1M13*H)dx
z`#QBmfDy=V`m5=aJkk>aGm*fyE2Mo|1n&MnfBqcDbJ-^+7vm#+2KM35qGATn!?8Q6
zH^2b_H=vghyyf%h_88}6PJfkeC6JTTxsR`3PV5opD#AG6`J0J0di!p=ZPEfh92~Y*
zuzVVaMU({t+r5&aVnI?8t{b}hat4pRy3!P&nKQRL4u~hu34=*%yJWs7LXPV%p5qMX
z{GT5fPr@u4w<2J}kvE4*Blf93AA*Vz$2cT#e+i%3!VTPxvi6h?bI5%`4wA)L31kEh
zhC@}sc($x9pj43<)R>$>*poorzEAL(A8>GRH2<}ur*1z1@5l9gOFV1{=qvvXnJpb!
zoVfrrN@$@ml#ZrGsm9u>Y>!PAwVQog-?18LU1Be#d-<N0)Vv5E=}DPJi?_0?TN;+t
zbWB~a$Gi<P%|r+v?%7a~J;N`A@~Vuwyq8It#*p&<z}3K8S?LJ<{{5Sw2XTkfroow$
z;O=<2Alo)o1i1+Vw)Xf(6KQs1bz=XspS4(_C@NZ&?Lgv;Q{D7&r$GrdDN&ffUpm)l
zm+@ZmEV?&a(ZqtL^78U=Cd+viwR}t4I8=F(rpbc{ur=Z^F-6QPT|T^7&QZ_rW=s+m
zD|HJ}-PsCm<pkv=G)0|%%lZ6xx0sg(w0^r}uno4BjWfJ<=sy5*M)@3bM+y)ra<AW5
zfBv1{u6}yUsSL(435UcYqd~?z#h@e^0)&{L2tX=A-O`e-?>AgO@+qrMtr$Nd^)l^9
z-vSAU?f|ITYRj{FoU9aWgs36TV(!_k4e6Yd^S|kR$`$e#1^!5Xj7oasg(<;8`qsNw
zDp>{OGxQ3OPR`eNXp?4FKMxDB29qlH)rv0#C4gWL+mEVg#wdcZPJ_}s!gEW8BI>9K
z25(PRk^Pw_u(A25J5T%b%(!KTM;Qbqh@>A6TqiW^8%!`4uksSt!>2AoMj0aqKG}*-
z<{_?ks`~qPe>`6mL<jdldQXs%J+rAy;z@WXe(|ApgC}XgB(S}P5t=_dKwp@DU6nL}
zrf~CxL|9wn1R%bq^}+Q^Z+>_nf{TScU#2m52xDL&s`$^j<Re?87L<hpm5>uLM-CCB
zR-_Lyq*z88ElAxcJN=%j1QNXcaVs>9L68PWeT*JK48lpS-&h|*s3`^@^k~!BTD=FY
z!IZ6${0gD?tq9vKk~CVZ;G=M4?bx`S(l^aR${gVf{%iump?R_kjR-p*?bM2Eo=*EC
zpe)y0SSuMuA*D?zcP0T?DoSu8Qc!BHT%5wT?6-N+ml;IU65^O@=3o=2gQ65{LieYo
zU$VqDe=l~1R+*|-#C@2NkxlS>OKGU|^^q~!(%FaNx$T<zhPMPscV+$tExH0?GUXJ`
z$A=+|e}QXq(0MiSVXTQ9o{N~REnU!qWgX}=10j;%X;mAPaUo8<d8gm3#6Ag3H0{KO
zP@2GY01ctqH<8{e9_YcaV*(GiC$^^Q*Pho{9Sx8-TXIqUIi5fz@J-;f#BYWetQa)X
zNUQPmqEvS%DJ5ku(S6k0YSpVodYD816=#qcf(+JJRs)!5T>OG64&h*F=-&O3KN9LE
zDXjdNLP7ghP@g?SmI-zLQF65Ga{-0Lx4r3H=&Ina31Sb~#+@N0p3=O;0b`DxJ#Zgr
zcsN0CaaxIuP?K%q7Mx?@bA(b%h{hJ`X8Toi+)grgQ1}D}-4q6nGZZBTlUM|Le6BmN
z2hrn(B3|6XZ@UBy3qt};9l*-u*<PKLY+IZK2!u(}877@h6fU7K7$2A*8#|ofS$haq
zI9pEWRNy4S%7Adv^iF6((EZLmpo;NTQ}`&r+ePfIY%lyw47^(!JxRCHr^9X;gK*~v
z(pt()6>vL5=F<bB^+z#(74`n7&lE%gBf%2VU6D;F%%XyxgJKV94T}`#<vV9HmzPX<
z>)=T&^4VWs5Q_{yn~1~4jmiX#8R=(vb$tj6oybhRrUirHg2k*T13^yT##E_1kR@<P
zkOd9BBc?pb_8eux1A%U13I<bH`MFv}hxn;D8bPZSDtysKds;T~n$F88j)*ouhULU2
zu#c-!r!Lqi;Mr}onras&7EZ~%F#*OH%yvjAl@XE<_=LG6K9`Pva()kp4lyAjSxQr~
zX;DH0ni3rUJ>OL<-5iEz1%N*XLB3H!;qY0FDhWy0OwJJkcj5aH-4n!@p1*stxkE*J
zkMpv*IZIGlb|_Ne*@8{BGSmL$*N(h<<BHli*?VZp>lnVj-wST{oifl%ZVG?7z2Y46
zoiAO5JsuM(9MVZo6iTFCZX;^*TpJHsK1)^e5{_s@c1;)!a<o6=3|T)xC*fNrMa4F;
zyHE<$ZIU)|s2lXu{Umc}wT$&uT;IMx*;geTjVQm79I_?OmN;Xo-_!=!pF~7>%plk?
z`mwX~V^dBERNB92l3&BEDCqqEDv9ZYFW*rud^Q<?r3+0ZXr!FtF)4sM5l<N62GgcH
zl8yPejIppu5kYw3>h57p5z<jpx2&nq%m9vOC<rXo8*usir_HerYvYbmun-X<FX_uj
znR!mQ<Al`|_BIjxfz?F2P-w_y6w?SR!$<q~%&31zFRyhRKrXtsK_<qbLfMKCE2C3<
zih}+Eo`@Vb3bOl+zWDdD7~(7V|M_P>tbP1MsR&&;os=O#{YcBl0AZ3Ru^pr-e85ZT
zvFD$2!}Njv{EBAP>TqiwDIk*_<E|S3BNB+#+7|&~tKKtI^t=~7lwd5S_R2xo#8q5|
z4)&}sbU#bElb1s=9cxxJ_#xY}9nuLWn8c<%1Cn)MV@WrenZhO_0*^zicTb3KCj%W3
ztI!E?5*7%1>j)#ywvJ$ewgGmL&5VcQbF=F|Af70!>3Pw+w_IAawl721zcb|lDWJb^
zbX#?<N*+jOQ8AWQ#C#TO3q#T{rx{fiDXXX`9zPJBPL7#RA{9+gIdel9SK<Fzm6-}3
zhm71bwjYKF{4jm-#JE5^+?)Q43{qe#H7UM$7vzP=X&-$~x@}QGHa@r7ZUuBO+Lo3)
zM&!b6iv`40js}+v`kLsKjP@eI>A|hd=0t9E`RMadB#dGCL3>o~SwbgiMBY#FZ3!v|
zoT}mbLAR02YKv&=$TRBD9;R+C)&mJ@!y&oIIr_pzVA?5nu9BPS8$BOThq3!uwCI?4
z2}sC{@_DR`cj$$?KAGxhSdHl&6T4Bf052n25Y3mWZ^_?2*eT)C-pSS8qv>j%p0|8-
zncapcA^}Ez>nB+0BmMiS^2~91C7h1mOGGWwfAfA|RwtU)w%AsAAUC}as=AF?1el&G
z|L!LP-ihu$ko)_J|LqI2G3I1mQvU9vyu|E#`ICO$L2x(lcv1##7G|a&b_|JYoBa5s
z&!i=+*7kFj0#rAK)rx&?GvHqrV9g6n_vQqCzZ*oCA@8VW^9a2E%89(IWD<JXe#An{
zz*lf$e05hE^K;$oQQcXm#WLfG=ZfHAGx_cSy8QwKh-T6_wh)dh_nF6xz1bP1MKpR_
z(wYmQ<J5=T&cV)Ehfb?AQd_tWl=`4mKXCWd@1SkOuQs?6v2N#M(7su~My=d#WV=td
zcUIj^;fmqFeh@hBKbmZG(|so5jzd_1NqAY)U+nWp#Ak4v{JRdaM*&|Acn}!v!?zOl
zv>b2eGp+uOxYQGiFbSM|GZq!2C!zuG?sz=Nx3z=vfst)tB45vPpL2wyG(j>?2I)aP
z6~0(zUtiShA&oPRUieB56)w2$`oaZ_7ctYo+)COrdv8Vk##Ll@rU3A{o)EZM?D0eo
z9?Pfi9jeM^o05WoU|Ew(K#I9~32(Lo+FIkWUpe}bzK46F!BGJ*s1FGTZ8qmsrE^3I
z7A|PxTL6Hp&tY;5S{DLJ4XUw3V<18rd-6OF{Jzee2;zQ3rASCVUO$BJ2!P;akx5tP
zhL@^o_Sh8REYGC_86l|%shp++vqXhyWPIw9g76J*AgMH|JVIAvoQDF06l^-)NV7zK
z1?W&~BcBuQ=&TmbmFIVUVK$bP-{?lQRz)e+hn0R^80zU;#4H{jmP74N@poubL@`ls
z&5Wri%1u_C?J;I%a!r4IjNeAQvN3PYutn`}=1HkziLgReVsEraXJ4@38i;;;Ao?L}
z8%}QDOA4{%ErkyNDK9xKg7T2eapkU`8f)`M1)S%uh6Z6Cc)PEKWml}L(H#3c>!@HF
zXuxJN{dbs81e6SYLaG@k#G{RWNQpzA(q%>}*wvJBDQ+Sil^#Q_P$vnqxrJP0g7Nbz
zI}gr+F8meu<k~F_CX+`z=KQd*(S{eU@==ig%i4%RwJw6?i?}bQHTjDk%r@$HmZByo
zRUW_8_G36FR4pjH_o+}_&_58^29Jx{{pH^I<hQxtbKFYxoj%nkm2F%0uB;?m9cw+8
zR_v6J?LzJnQ83_XE1r<OzupD>XuBjlkO`|6VC1=tFpp0CvL$e6lZT+v`*|!BEIZ{Y
z7AW?MHwixIp68V7a9!R{BkXU^Mo{N=ILTb=7G*40VQE>`oO)a;uhY$NJ^TXtu3&Xg
zCs6Q-uDbn@$wDj7!$6H(u<HCrV4L(Oo`Jw(B#C%VWVE=AX*Zb;eU!WDLCLp7JCFzE
z%~^dr?6{GtA_3m9PEw+e&BZL0Y_?r`>74o4ed3$=);NTIn3MSzUUQ31QY6q({B8cw
zVl-k@KHQM$uyfQPyosNbz9OdO5d^0NnRIpxy#`<j)NG5+>xNh>6gU*6$Xcvi`=>gr
zp!I;xEy=((C@>=HXc)`S)C~SvvH@-z!9KXo17F+`t~!-|7bedT=Ufb`dtD*A>V4X-
zq#rM0z#dmXaua`Dz=aRl`hE<s(-lQ+m5&;8a?JVKGYtmmVX1C60{;2^nX`P#`;LK~
zZ%0Uh_;fB2{2g=BzGc0*doEV4KO7gsbV+;wv#{WLa7L1={XMSoA;l3CkCFa0s(~>S
zJ%?+5K4;QvyL(GaAestb)Z%bl!>uuAT*s<3@bz?TdnQj<QAEKYrO@iOUE&{v`y6}{
zy}faw{<->+CTo4iW>y%iY>0pJkhW#<1#e!p6(1ZB%!1bypM|5(RM}Yn)(uYNy{O6X
z^FczeX8d8#zXrw%yE9%4U<_`^{Wapn4n=LO@DOmqREoFEyy%Kkt$5Xo#r+U?=F`cw
z-C>+EzgQW+VBvq20VuRujKOzg4-3WK=dm_EFX(^Lb&uK}4H}kXKq5KWmD^c1Alm6#
zNakH=@_3m^zCT=8hM)W+rMi6^<+Xm(`dN&ry0ReuQUJx`s=%=#%g9!Y5@NQ#U21F3
z6yY=VWm_bhfNWy19i44T$IL$5%8%j<iwbR$ip!iW4OTog;v6P8FltX572T0oW)n)l
z%J#<-Gnf)a6drQPxkb+93vs}50|$#{rkM7_OL&3H9<E|+2d!p2<9rrE9lO|=r5ZVu
zQ{N6&9vv2Re-b+Mc_=+QxPv0CXb2mP{e&V$b~ibO^M5qV6#uJ*W-6L2r&74fARZEE
zg&Jx#{^4f?ZHr2x=DFEt3Tz)LRV)V#(>=(2Kf@bJ9-d!XBosEja#YMAltcqA#yIXZ
z=QQCFDiO`IcQLY!B@{e<e&ZVw-e-?!#Iw9nyUGg1&_BD?XCrD52oT;pf;-3I`unRW
z*j$KITH+t=ap*0dTB&9*LSdKkfk1cJq`L3B?T{W@jR_wm<<{dV5S;R8CAt80DBI8*
zu5?!}UanNjb>FM;aXjn!E>S?nthxA~k}!WuU-XAD%MZ-l=ewtq);`<W$x+;GqCyT3
z!+|#3wb;v1o6<mGwGcR8MsAU$af$RCSdnPk#cItQ&4Qm0PXxBr6;jC9J5LB(UAp{G
zwSe6Ly-Ft33FOet+rWtBiN~BtGAaojx`R30@#!0eH_dEyV15^pB`It$<fpvk*D@dP
z){#!ewa~;8!sDcTXGWS*Rqu`V05u%c%1O=14yh9w`9<zn7nXKPQ;;9T@s{jPJ5XL(
zp_JnBii?|Q=0umm)b&gD<ivcYFEa~u;4hl7V#7ga!n?S2B%-%NL#CCO21qyWciIGa
z=*$eQvO8kv@|eH&hhVO*9CX4XeG>v3FX@m9^;XeBA@!&g(B)m<m24v!7drasB=Lzv
z4Shj;Mpobso`&^uKU=Bc4`u!P)9Z2F|FBr<bgazPHAKjbY-di{&DV%hyDlqZ-qjXF
zW@|qiO@Y|C<%$Lin~s!9gWN_3_nFizrhiVipt7_kKVqRv=`qgkw`1hR^;GL7|Juzw
zyrq-QYy&AveDIS-b9rY4F5;Eecnqn#Up%h%2X$9A&e2=hh)F7?UCKE#B;yS9AgZv0
zPv*_sF3}5(KOu(#*u^^WFk{ui94Ig{GNgZQ?!w#*+3uQ8myd?p4GSD|SDRTvj#@3}
zKs?%7o3pY<w{|zynXh}ha;V)>6|Z*e86$)g=MY4bpaxoW3Z0_M2GXf>$y!><T(TIv
zZKc@BNSc!)j?%N$FB%o79$V8dI4eO}0u|PwwITNJ+%16wsfmLNYlXe0$!|O>RY^9=
z28x7={7*kke)H4Mg3YXNGV9dulW_G{o0ABD(>y7&^}e7wug7^T)fnSl!Jd|=BFRIJ
zGA~AZ;QW&5$SDOJ+*sTeJPeZIu@-eFy^;syA1RjNR%0~9XP{R@hM(S>=OahQ0(Ibd
zQO>E660HS4E>4;RW^G^s1bfGk(?9bGll4;``8&cw0u41UP2nP54JL7_+_dxLDGV`&
zKzXd=t-ilIS6~i~)d%9S+4E)Tw%xIqPJ~a|sq|yv5;RX4u^Seo^b0eICNgfMr(78b
zp@%CnDCQ@)*~~tnJRKgk8wS%xhT83?3lea44_lc?cq|&1j-CS-^V-ImJe~Fk4^d@h
z*Vp~R8X9V{veCBtb@A?RtNP0%?FDI?nkdR%?Rs`ZZ;p(i;#sI$34)z>qw^WIo*xt;
z_SB0*S)NpWXZamy1(t7If>4Z4g%5WC3)1{ts~FDwg60mv#|Cl1`6zJgM`Xrcb7!E;
zH{iq0<uqx;$Mu9sJfCnfwRiJUBs%ePvq+u1NEA3y)_nOju)G;+JmmQgXDi4U{YlN_
zF<!N;)-OV=#-q;k5^m4zFNRoeL}a4V-oH@9KN2N0Cdu#_Y9vh`rKZSnlNZ%RdUP&T
zi9uj+oaL2+v^I<OTvZq-7r9BO|7qQMlzHR;bGtV|7bni&zNl0+fEKs&e%*8)G~*C(
z`2jzJ^4m<NynGTVjq!;653NIFxQ)Qxc{^{ef+4<#$!uw+OEjf~b2x-rE!bsy8XvlZ
zNzKS4DbKXBK+H2Pf8{gZyPLYV&gYgO8=JSwKRXJ4#z9+Bx!TQpy<BB3JwJsPYq}tD
zreKpuw9#_gT(S04CZl-aKz6CI#9wg7I{JrGqg9jBbCm<7;hw@=Tb22i27E#NaG9~D
z!G#0xN8Zk^0U%<pSQ5TYMy2@?zD2}PMSunNqqfFIBI!;ixwR_+CPD-Zl6!*pzMhty
zuj~3ZcpF*#<)_>A!J(bv2SLsesMVlDyU9mP0W<R>>R27(L>!=saE=1<4yT`@F>}bT
z404%cIoI1jo?!mNw`KC6gsx%W{nYW}Eli?(GjdRGFWAzu0E%myud*!!-uZ>pR7O_z
zJ*_vbO;QXSjBH`&uc2k_Z*3PN5!_uI2kV4OILbO8?I}#oxD#N37hd6fhzu5!t1Sj&
zl#xuT#lU5-Yc^ExAfFkk$h87`lA22_ffY1Vea)kU6~s2)G*oj3K&7fQ^BdOtRb4JK
zPu>xx@_g=&bh-=O##^wxSx#8n@zUB`k?*?e@`A7FSP+;pzTd)cQiD~vE1%fKmFL+g
zE;x`G1UOK}-o-!u#rEj)_SHc@6+hYrhBM%NyTV}S*j?b|DlfDpy^@lHPzEe8r5ShQ
zPNdU?PTnBSQtK<8acqhRkUcal8rc%j+qt-ve&BnwmRR}#$tMij3V{g-1wxemFK7sN
z*M6F%vHz;g|Aih06DwE=%0S5Red~?|)m7dzFG)5G(auz|udgn(YrJCgpQI@17-mRH
zm|gHe28gh1er(KxigJMew4>SL@WQsktl6oiu1t2z2KjVZ;AzfsqTZ=9`!t8hIQKV=
zBPqQ5;X@zFR3))_YMN|I=H;A#kgTrtm0C?n*?#rox%=<Gh4$jP{Gh)rY56*TDDa5o
zgqQQ!zV}VJj5jgf{Z}J<_0;0MZQ#$~6PRpAjc#A>_ufaTG%hNp+=c-%tPdXz<D&ds
z8Ct$%8+OJ|kY<VP#h-19MOU%bzs<mDI%X_9IVOPBSD+H_{Qx0c;|vbJi7GsheB+&3
zQdu@8`ClL)-usLNq3Y}7CLM|!(=DjjUyW^3VEgM9z%J~0L8(_u7oGg+r*_rW28QLY
zJ!ep1ko-1A^O|oRTHM*wcyU3a(?6$i=_~Hv)s$OC;8b|Hr0IA;6eu0)%Fd$zAp6o~
z$y9*%x(YGizpWv`VqxpmX-#31n$8wfMJQO)WRUnZQANB)CwXORje!M}7VMHaX~xq3
z9zKeW6JphzHs#;rL{+aEOP%@Uh=cpetyNLZ+KXk>IWnxJ>X|Kca_`>!m?j*a?!qeF
z+aoG=Cc(t50Lg-x(L{4-D)W4NblWfg_3bIVzP0~7MLWjpQ$#+0n^iw3>laQk)*F@R
zS5x?C>F0XJ@UYVf+X!99(i>R;K^jp7IYu;%!c6T{ruFpN`BNa$^0Q0+)19L$hR$42
zvETa(V9Etoy%tq&B&iJj`U?iXHWAsbAuIZLQDfemwA}@lq&^@Me#MyqpC1#*$I@bY
zbGxfCB`UgFVuN_4@^ZF!UGm-RC-*H`m8xFYz;Q{g^<gbvIxo9!*xbxwx}bxQhQz;@
z+k~RwKeOI5^6=}TT`d+8&eY%~G0KM8s14n0@nv6-^3LS=5Bj1d2F8>%^q~2pW&yY1
zuS1v~Ha;_{lw0=Tq$B~>9!U-JJ&^rHGq-cty?c4W`FEW65DxryxF)*x#Hx#z-rHG<
z=hQNXA}5UkpUJbu!X@pV-Sdfkwb7=(!(v66gg^8mcEM#m6wgrJd9vt2*S<F@YdjLz
zE!+`lYgeQoCR^ftW8>2jG@ey>LjmqE6(Za{ew^!>Gy~W59LAd37zxRhl&bFX_GC$Y
z-YLxJe}9VPMj8=<ZfRg(O&2#LaY@%+#`@;xx#vneP~4(zvs<kHl@^6ni{;J&a34}y
znq_hOTRNMH$!@>LmG<NFCQSda+#Gk@@=JQm$f8k|gXO|gDh$nbs6EfgxCl#K4i}%6
zHD2&ZsGo0uCgOAF?Aq?9;=-w<zJ5w>Sj`PjA-Lrq^Gtd=c&KPjSg_aoMUq1P6WJo-
zVB|#MK~;Z?$IO|E(JKa@JC8=UKS$XZX=Cpp`)nY}H>4wc5+dR!d#vT7myC?;aq8Zm
zn<OKcbqILYtVNK;#l`20iaSxA>X*4ggK5IJT?5DCDc`;M5zd9Yvy@?9eI`&s;Wzq%
z5FLL{AO^G4x17G`*la1K#Gx)AC}In1DrY0r0ScXo5{NSI_ypYCb%<OPMzO9L&cSR>
zLuqcMiu7GNzXCRX`>Mc*C&K?O7D<o_4)z)e-D5$|muG}*y~dQ<BS7j;H}h0D6!98r
zljE^F3?iQ4{DNPvWl9hs$83p6&QH*c&Y-3j_8D@;VP|0sQnqRiW>l~e{fsVnWi5=#
zB&*}2rlBqydyIZ~b|%Sgw>!tl9y3^1Y__)Lc{mXqlR&w32p%KEuW58dN2kM+R|*Ab
z8?zJT{CC)c152nVB_ss3xq`@&YfQwqW6k0Que1wi{ag@LJsoGK3K{GATpySH_8;x<
z8y%IKQEb~C-m{+w6>Fx>i6%FbvuG9^kdsM0r`3Imba&YwMtEmexok5rn>#8S?oxh#
zF2?i=%<ffr(0infPia)*$7Lo#ZRn}Ts<8WR8mR2NrD4k=OG&Yx(0)PoD&n4Fu7AF5
zs1`CN%J|>oQEn-KKPPO3&IfW{9y_gnD{vOqWga(9bKv#eDeIIJNRrHFF{y&PWTq4m
za;CD|b~&l!uhbr;`tWajC3-SoSfk+mXVl`avhD?SnRV_Et7`P~yQZ=7;^xF=a)x~L
zV`$gmVnO3m3h7O|uRF1!ulFZKets;F0eaAOS3W>+aw6M;no_@MT|xQj`9H%XL<AV7
z%*OVH!8htW^m>w(=4}6>`K{o-)GSj&DIoCve<tAczUu+hP?;6q^VTvogTiMp<vF(}
z(BGlJ2_gO^OOQPsy_>Z%s>OS#2N9Hu-`QvLG!*V&_3J!__|<b|ZT!2WwBX!vTeSFv
zYeOK^HeCdML&$|h`?QAo9|ZJjrh>)xs2qJq+pZq2m$Iuf(-k7wEfN3etCWYw{2ZUw
zh%>m}l7_c-%_+_s$d^13Bo)1x;Gw0wHXGXQ*<^3lh3#uU+oBY0ch)j~II=I`H_(On
z^_((1o4vRneY`Y+I3k1c`3hPm>k^YR7TUk3nnV6)9)km0C@Cd^akK>(dQ%eqD$PFH
zVI=A5RBf&Z4c4CZqTAiN{J74>^&X3-`jSSYfBh8TF|*M1gRAiquJ8{yiE?;<)aDhU
z1>W=*-)Hcjlv-y=maeNG{`x-ze(NdQ3l~`td=OV7nqOG>f_&f(P*4QHe<?V#Dy)z!
zss3|L|Gk8LA&``uBgcZ@64l>&vtT&v{H}kJi2VH<O5~7Hy)bdE?f|YIOH*$u*Pyrb
z`}ibM9>(mUPfyq6wyShg9kP7uKTbgPCJiRK0w-*#g3C5oE<cAEi|Q{Ac{p#4tA&&Z
zHhZER0=#Jq@AkbPb|aKEG%ZxpEh^r((Y@~c|M3&RuPJWeY)-+p<_NJR7i03$Gs^EJ
z6rMSKr;%m&BAI72@PaVSb4lXcd0lJsYvt-eO=u|50}<T>>G17x5YhE^3sQ!?^D=9i
z>*^-&lr2S7NmSI%&+YUo^H$yAX=@{6PccKHbPb7+ueN3F`5M+ZXg_{y{9peFY(y^t
zq}G1%7aY9H9hCtWe^@>L^?r7xvtCEZ=Px@F`;FV{D_m80B+d-aatN`+xC3f2*ei44
zT?%{3ibX}XF3{O(>J|Uo&~Uwb*?n)fo14SN*49Ghm}eu5MU=x7RRsbiMfJ!VUq3%a
zAt3|9lz{)c^4Aks9TEt|7I0)vfGX1~=+_ElxSM}#Dr+%r&bXq?6q$B!d}CH%yHJ_u
zGQl4~O+!;=HJR-S47U1UVNp~!jJ!8Hl+m#KQqO!A>O*hD#?8)aV`tacIpigbisMP-
zhwf-(V9E0q1=FT5HI*<bDoUL-AfO>LiUS=TT|-;DByRA55PsvR+p+qQJ+7vRM^&pO
z$mKf1-LX-2uv-=5tm-G<r_BH7wgB<$Dfly89=tYNa@H5toNR>}7WRzg&K_z>T0O}*
zeA%+PX{n<Yf&|PCN+rUXdb(c#h3S=zGX+pDlz>*_g1u}X74{uW(WlyV7J9T3nKOU)
zKu#V4$XFU_(FqdDe%IvPOy$>KWkGNU9o)LEjkU?@zn%~JXFpi0Zjv7hZp;4j{Pgxj
zkvGCqIDtT>hIOJ)<mV69pJUy)WMt#dyVEoCM6L`SU$R;vXS{*QWgk8;GS-nGfKq9M
zyKWCym>8K$W(Vn~07e)?Pic5@95@eklC-iBtH(}MK{0@*1q<6QUqm=q1-kPs-lO)t
zG_4^QO=S0d(Py3$mcm>ZoY;LF34+Xpl#ib8rW~Mpbt2JPm81FB1@pKgUY#;G=*2E2
zC29tCXjTokB_rkDJhLmZ`geHh)?Q??&6h<#2MYEJ)r9eO2b*0poN@pE813~8l0h6Z
z_1z+tTek=!8rJ{a<KS4cXI$gDRNmeB%ayG1UDX#PjkXaaK&nLP<s~OVlJ|kin7SS)
z3Fu-Q1)phG8!X#3Wgpv-kr9cd$-dR3gNkk@XXV!w`2U77o(O43>$XLdHO{1bZRN%;
zcX;PeXdiJqv*}=yD`L&q$XXC{sR-HIh)}42FQscqJhZuD5EKmU);E?rWBs|+P8Wq*
zktfOD_f(!A>*GEJuU$A3_~M7RgAx8B=g(ZV`BN?NaC~RAjx{A}kbY*-_G-V26{C#j
z*MRLlTZ_|@3`xfFF{-`Ya!d3GlOMyo)*z<x=hhx)f%Kkihju91j~_qk#C9b=v1}+u
zYn#7a%a^oeEa}e?nY7<Jo+mk2L6uQ8k>gswt3iM2PX0J>q$9a4sgL<^RS}3$v^Ue4
zHo!ndl@~cReM~*9{C}+*1Y`gPT#SGz`vmE7)-z7IHCY7f`cJg0lFl8GPwMrO!URdF
zI|D;92m%`9mNGID*OaMujZ$@)t>vwaaa$&9<Kjb5dvADHQz;wTEZ!odu&}V{Xw-U!
zx0?~&_Xm`QDRZ&e&pp$P(0h>wM_pL-_e{CE^z>?Xr(I=zyfFx&S)GcTas)9o`Rp`^
zUjj@ByV}nwPXr7UbNqVZ9rp2M>+0(4b_=Yg<Xi?dXA)q4{-&79Dkp){6<e~hHoHED
z#U&&auX)FpyYlxb0^U_pqO)JDMQt)MeOI|WZ(35_IzR=!>qZmHUT8}(0+T|EIIyzQ
zhL#e9z%^5uSj6QnC5nQKUQ(k?Q6v_}6gE|rQ@K<F3f-2Mc76D(qX6)#o|l`hpJdan
z2iz%$H>^aZrKA{yu803H{J+MzgL3L}p|NA^`;Q?n1VQkC7g3|HqDa-U{{GbG=+SYB
z{>09O2$W}_E?in!8M8Z{vG~y#ftI^m$w~K!bM-c=BKNK3O##+G#7r;j^L|XEe0kN{
zFD77$TX8}H#1gnS+yVb!4yOx{Zcf)T+{t<N5(wwIlETpyURhrXr&#Z;loZq~I4_AA
zL`9A7u}V>&K7G`K%@Q(YizQ``k-?ylTh;vvOFWD*Es1ztzbs>FYKmhV+#r)Q9rIOo
zCRaQJdV8IRi+4J@f-E{R{#HW->!^9Joab>6Z+73*584|DTZwfb-P2|M&XB{F%a-KV
z^0J_|HB0!NO)kJEtDBn5QXyK6#^#qdR3<|u)eU2Q2rgCm{r^XO|3z0UpI;5Qwc67%
z1`_gYwr;C-+I6pJo~#o1CW;po<2D>?Nbo85)Ts*9C;jVJN!|@l8s#*OFqXF8*goU>
z)kBV=wC^atOA8d&)oT3yx(}vMP*N()rM%L?=}4o&pPx%7i&4X;6SBNi+w;V{qMN#<
zNlbciV#(M|C2Oo`BQLFA&<@j|Cq^|m%E6WaczDU(^>OI~b*ICTg)#51+Rz)$H8suL
zDXU8_)W+JU^Z43)aNUk9Z#~3Tou-BEE}6<Jc?6ZMT$c1y4a^HTL{ZStG|a-|VM(h|
zb^gN6*DcR#XB0pWCj5LiGZR=^hSxqhNr%wtu3fZM*j&E<e;sLxtXOEMib1}{!W+uz
z6`#!#oJX2jUwqutoY|bG4EY#*pBQUaCT?3?QX&U0Fa9iGh6rf)7MS|rKFrdcz>U1M
zV^%O0(@pcd&mls`xl^<Gcu%ISN<`%AXG~FHGiItu6SDLs%7WI|SoDfn<P6IqpM09`
zkFlwF{Wcq#6;%1#{4Ni5YphNMkCK5XZE})X!*nF!-XM=(JTe>MTO*5OLAc93FBnPN
za!U%VgdPf68<eu&95>0gY>1-9I5&!>6xjIbG0!?JG-v<IH2#Z{95sO`Y5(`Pv^ml6
zXB;ML0Oc0Zn)Z1+d!VP1YZ;IIF@%wNET;he8{;l_zf3;2nPKYPSe9T4pjIXghzhA%
zvJE~qPuy7~m+cGrsSEI0iNj^=2F_p~?&qgT&YyNW`I_vSmYbWO$KE+@WL?1CZ1e19
zvHuOd&v$@Ap%tk&o(t~+Ax(HMNHW2YyfaCB0sBIVSTaF{*bgZoCNQRmW(3$EA+jKn
z(<hRQgL#)X9FL;(9e2gEf62`<;iRR#kn`UAgnoAHu59x=Sw`}d-2NLL)^+J@2swWP
ziX&ytEoei)C3XFUn38Q}*!5hy*{)<8=XH;uVPrWGW?o^W%&gB?`_l>KkpZQR0PePD
ze%+SbfqSJ5=j#DD;(V<y=AXB_rAicXaauw9YSzHtyo=XF;K4#xBm%}>M7tjVb(Mh`
zdPP8cevpZCC;_xT6d50f>?m%*sU!DdU`#i^>da&A(O<Zr^}ue$gDc7Ah<GlpXCh;2
z%4DJ~S2A~nM6$<7c{OmSvASTD&4&EAsh}pL@9Zmf-zOJqV+X>3<=qdyzMV4+$9$?%
z_dot{IjEA-(u15dN{Cld%<`QKkT7;2a+FdgEL+OyI1a7>fahH!gAKsP;IS8!GNS|V
zgh+ek-*dA_s$MkRJLxFvRaricp1g;G&U6SvJm^!otJZdQx!j`FxgZQrO1NMd>9~zh
zF%5G*3v)7BiXO3zQj{kC8_KwM6QXq*R><Iw4XKg@&%yngzom@@@v*q)?MC!9C562j
zRED1+5URZBh2#GK?tEd5?-w_sX0W+Pt7S2AP0s_orm4Dee;@Xa@xwv^+IK0@IzWI_
zwrqAg(jw-u)A?%aIosRepvXT>0_}wNBGeG4f6m~l6uePbo7`52%zkEzz7k9NVCTwb
zvFGxpG^>cq#ei*G*bMp=EWWTSILPyIe}R?G)*AfLjQX8@CbvlFZl*|~JP9*;z{=8_
z^ndf>Oe~oB1O&U&=ggX<-3t*sXBGKQ&m@f8%|R@J(Tiv6U5y`QL3kRcx#E7#W!=*$
zlKonx23P^@`A9{@#VdGYdc+}X1fVs9AnlyCc=Crw!AA3C8wFCnXy4}(>!}rbN~n-T
z!7M|8PLH5JR~j*Nb5J^Ip@O&T@9(OiPs3@HThidraL{yZpP%(=HGF2zu(34S@$dl>
z2oWd7-=2})#+IqHxjs{!zs}aqRYw*>a^;i~Bt@_)-rGF)4!hiN0xO=&ulEHR!J;`p
ziXQV!CS)cbI+>>)S8isx?<Ra?W^G-zGQMzo^8i&86I;hjAvc#@(-ZJ=<vpC_Cp*SE
zz7qF;=}snd?)tDtFf})5yr+A9cPD6ct8*F#!xByA&i2TL?0FuJ#no;MGr1n0T*+K=
z*Yh*)j2HJqtuY<~559gq)g#$zQJ{Q?nyMHH<|fynCrOV6mo?3>hI1#naJZRn_KgAj
z)40m~ia0;RR1SHM#hW7cyP|&QfCp2TCTxhel;%wDcRdRkE-xv}_AA~l#xFg0&E%Iw
z@hJo|{do6{q7A|7)X`;yg@y7`wCgFP2(a&^wx%CzigiokOEL|O)&0m;58OD&2q4}}
zuvE-tF3R+RUhOkklvD5EAh_X}hmz>r78=I37^l;AVHmB}_z(phF}b`XDd<z!8Wt>t
zC3RM=qw`Q%$6#v8r_qx40Tb#WL<yuS9uudwE+!=IcvA&6_NIc8q7SPlW~Hnt_N-N#
zx?=HN-O}7wPkuP19+;0`{0$AsEX;S=LN>&H%FB;D6&DxF7{Y(e%FHZ0KA+a{^Ajc$
zO8O}g82QEf4O203v8b+WoVvX-Ev<9w)cN_&j;vs<wNO$bfI!&tzp<vY62;KI$4W>f
zk`UL`MIbYsE&F_~iYNbbC243$2CssrfF@XjL@qA;vlN5z2vLSh9C^|3=UTeR2_~ir
z24Uf^s~BqbwqNQ0dYIR*@CXa$1nUZuP>|$3(V6IhaOiwpnV6Vbl~^qH@SUjJ5Tjzz
zw4U>fQ$>gn(r>F%co%w(#;ZuB%M(?IuoY9XXJ$#L94q(%z0r>AA1$_JkG;#8L*KJy
z`r1BV!}hg#D%fU+h-tc^+pl^BJ_S<?7!;UDDT6>X#HW5N62YjIo4)QwMqL$hvF1#V
z2nNny_}>TKOl>^aQ7}J<gm}}1PY?8vZ9C3z9=V_U2WasNKaI$^hgX-_PM#?eD2Z0j
z*J_e4Hfr?`=h8GX5SrRrE@|odjT?D2hH*S}*HK)5s$pG&&vVdUtzB%cQ;r$(R%(8b
zCbAWpgZX1z0f66m^9gQV2+Rz<*Tby2W+w^KV*8hWl3`o<;4H<rDN#|jv#1$92X6hs
zlzJOMzbjJ(s-%e;5Y5vLz8`Y^(+s!Lht|p5e%tBqdRgj_l=wmXPP8Y8M=&*EyuC^@
zBW?!XfC8?c|HD@@Y{(QMp|LI9GP$H(MP;{TbkHF8@q`%xh3t|4R8Lp8Bj#K-&vBEz
z-RwfD-HolSzfB+zPMpJm$nT#uT_?sOlNui1+LUtcksf@ujt-o{DpknH4i}SS^TnN6
zx-RV;P>z=pJ8vwBxM!Kn_~J9MU)UEXjo*HDXquVW8gWVEx&LSUdJw1~>ex&qug#2c
zI=?-*|94twwXxiX<w<OPYF*~KlL^bL`StO?guD?YNcOZVY9fojYygCS1<`9FRbLv#
zVnanaR}0KF@xiI@f<y*R8G#HNx}|uX6Wt;$1bt->niUb%;zJ?n|L(;fK3d>2Z<oq`
z4Ii3qrE=n8<BmG{6~gdlDSqVbQCV$_WR>&&#El?|HJL4=*l>sAxR>9-HL`r1JL8}K
z0u3l09OkAEZN=SZ5#vi*T8!X>M8BMp1+I5+61G!c>VpG`f)cjG78tJ8L{M6PQFyMg
z)^-v8pGRux1|wH@@QlaVk<*)Gh|nd9=(?wN>wS|JwIBPyYRJeim0}ZTfyiSdn8GPy
z?5fB52d1iK^gnV$P@LG-wokQ&WngsAgF=|>4W?aAE+w3VY~`2617@%e2mvD=#W@`K
zDA-C=7{Kr@9tNcmra$dr<W}%g&pc$^H=4@Ak};NQ?aCHq2IL_<@_&~BG05)KRay`Z
z(J6P^J+i*}4{5WaI-gqC7RwWhUSR}|*p%*Dx@S)gz$8J)-70Wi_gv^+o_$g|6ojc`
z7Ah?q#>ulo``=dsE2z$aqgkn*D{b7YTmU>uMOA6cP!<;0J^1y(1Vv%v;tfet$c7XY
z{_DvBnC@6XNdL~9(_JQh?uOb{$(_G}lOTaju?ZIA1T%;S`QZ}ZH!SH{Sz<xbVHXA*
zA``YuQJH43{wqOgV*R}`OqpGCDdBtfdKb#yGVWZ0K*1qn6Ju3axWKUs{pR6|LH&j=
z?pxCTvUn&xTtNE~2Q35X@{L7<)2g%>lzvLx5aI|Ev;an-rsRVujljdA>Knjs?C#Ef
zxh@e^b9d)CS#60Y_q(3vnPg*Sjeq$&Bsw!)AiKII<ASGe(^Fr65}AsEiaNjM;_RGG
z61?uCsi}G0kIFg)#Ln^nnj)CD2I!}1>vg!FATRz22GWtn7#L6(>STrl1lNY&y@NY4
z&?NSACjm$R16KBRhXpVKv>kK-?Aj=cC*}nBG6Az*<L7p_BN<?t(sj9yI1CJ#yu6gS
zMy$;?ppQ}jlpkgQZfi!PComs;rc$>d-2mQ91ZXIGI3K`7M!Us%FO&z0qYEST4QU=3
z897dNsD2(I_lfE6?=P(ih6821%Jy+^a2(%muA$F$?ZgR|wzsGA*sr)sOamf)XN?Z)
z?Kst|?Fktfim|b=$_~rz3x$E5SRf(NAkhtEaZ@I;ChXT8KMzYeXnO-Ugb+4@;|qM?
z_>=Z1NnGBwKqaWnMS$P=H+bD;QuwuFZW@?g{$BL?`YSeJ?OXq?URb?$SH_#OO=I-p
z($debLvY=)zkQPiGNnV`@pSG+vmp#xjYf~TmJRQ>rb<9Fl3PWFz)lS3@Z@B?goK2i
z;UW=Uu5jx~^J)@Lp76=ZW={?<p;ER&yQWzFD_vQ!ouTb}3UzIWqc-TpbFT>OEk(il
z@TDCyT^Tlrs`1kfwZhT-S0mp`Ed@nqqkejiA991tU|0NCSKPpPwUC8Wy7Ao$H@IJS
zR$IL38*S1GXP7#6fq*T;Onme!MjZVvlry}}>&kL|Y^yKi0I2%D=EhO+@!E?32VDXX
z+hu+w5ga}PjTLqWm%!>v-d1lx!j$;pKYz{zj?-IHae`oLWp8f*<y5vH@0-IS)(^_c
zvDh*u8%e3DazJ-&L0}+EzYe7@E-voW$hWtQR8&ET=)~q_R!1NUUb`hRV8kE?5Z{#-
zOxAT`h(4Xeqm%H&P_9)P_hchtkd_WY?U3M*Q!id2#^`9yje0MPag;&i*dM_3>ZIEr
z<{Z=UqAzl3X=xhj>iL5rmScn{n4hwBgCOB`18cTVe=ii+7+6@ecEtBuwe|HC;StgJ
zwuwh12_@Ez{qLo>W1ROF({pwE{!%(h$;gzilJGhd=H^oN6Nj)lR4xJyVM)0+M>Rb&
z6Ip_cTwF=7#AX0mDJOl|7T}_VqW<YnU}!*-Q|ZJA5?ym*409yh``6i%O7s;*K~5>}
zo^@TC9ZTFOB1}mKvTp+l4u*sNg}7Mm0&?&+TPa|$VHcC(k|?LN1WRD=?HI=_bB(<^
zYi3JWY7h@_ab@W&-zdKhIbb&K2AmAV%Q?~p0VNJ+`QeJ+-MZDs@agX{r3WK#YZ{Zx
z!sowRDjYs1GwW-RC5ap2lg}j*>AK}8J1LDlRUDN@aEx`PA08emaav7ioVj&nIJEzq
zi4d7S;l#21NCAh#*;Whysijvb#{RnVix$(ooOOrk<U@+AU4@@AbogpxF#YNS3r7y?
z^iaqJ+y1>g&ih?Q+@60_j*e_%u~C6a#;<FUFieV|mN&f)l>A1<{xP&@6F4*m+ciVw
zkDkDk#&<h&i$L0H!NknW4PX?~>t?XxO&)+AtUl)I>$7jXt?A4^tt0wM%=)l!4S+u7
z9G)1jw8c`F%J|dxFk#K>s+JA)yD8T5e$H(h*S&z|(#h{g6=P=ZO&32sGb4IkYW&Rw
z{;iZKg%!<D*lOrugFrz%RH}v_Z1sAqQy6#ToH7m-X(z=3h0KUgMNzH5$m4mvf)`O#
z<~VOO(UktY&WQZD>hnYBlBGG)<5H6dOiVfcJtPu@Gby+eMKwS6HZ%wrEc>0>)Awdx
zyIkZ<sp0{`L-}rrW?kegCPT+37X;eq)Ba(Pj0}-eEy>I4a|LA0-wtZE#6AFmonCE`
z9y7{TIeyK=(aea(W*0gCo@6y^N=sFrfo3%fN-$>cAbHR>z*{<7$+!jj^P=)aKt=p6
zyQO!>LK`3tgGRSlKejo6S{a%75H?sU1bcm)T>(It*Uf9S*}j}~#O9dIV+;NP*t)ru
zZ39$72(_4w_KsftuB@R)Q2^jwXKaRb{)+}6pibofue~dMgz|m&W{iw|kYuuqB}3Ni
zA;wN2LQ1KjNS2gQ_6){WLr8YnciFOK9TF;gWDiC5eT~w&r}}-*_nbH9A2{=7Ud(gP
zb3ga;+}G#2KA-D?+`neK^8SSG=$ls${Wdl?L*t!nJYDwZb3|sX<HkuuD%X)aIkJA!
zpqDvPQ=yL^^F;>>Gqz7mqySG`PC5_JA)l*wvM71No?-oNl}momm8hMt5;LC2m>?z|
z!-j^27X30LweRK7KsK8VSCG$H<Rfof7nvAdnP4<H+1{zIQi{m<{DyuOi2cl$n;i2i
zV625N&U&$QJi-=H;m`_#RY6$VBRJ7GGazXFWy1I|w_BY=(@6n{L_eY~SEsRwToPgx
zJnG*~L#J1M>m3V_g%#yfCt)O2{ok>6G?W~o;+#uUHAN%xDx;#J`X7A%&Ds4=XlPHs
z{hc)8_;Fo!OV>yjz9Sh7{#*@}--9#6<9zohk2`%bHjD+uct15o2i${Y{3L6Xl)Sef
z(6677&}ZQdAew^~AVpqzR++5Rk0K~h@uoi${2I3{>q9_A#KdG17<>#4Buj>$HcYQE
z6Zbu4lXFTSUm04jD}^YHz<<b3zA6Egu+Os<vU^Ju(lF<(0JZjv#CoUv1R~wEsnK8d
z%rW{d#YIuGWu9Mq)Z<>O_2(k+5;O$%dKGX0m2YUdPx?{Ai@$Hq7I<9VMgPmIL^=R|
zfTky(k~)p%$ofUOEM7Uvi(N%=Dv7N^QxGDV+p{RiHzXU3Pvzew-5>IlABjX{1Nxp|
zX}3R;`eAKGMg|u*C9Qqh5f5IOrlw(Tt3e*6xsd(Lz_9UJw}LO#gBI4lObLn)PKklc
z`DbV~L?AEyBT*EWpRFU+`_<d6O(Yhr(RIO-6m7PeIo1=@pc}Q(M*bC)4-s3br!jk1
z;Bffh`@<voTp2x(;AL<GT%2jf%gz2eT>Yh~s{nC&{X2e2o+@`V)cWqU)?#=`I*Onx
zn{ZUk9nZlsIv_IDWUO1R4XGU(N)A4Q&|Up<(~Ub4Hza|tbe_0db;XXQBp@8FhOpi;
ztQx1iB#$tlv6byUyUWQ-5Yc~#+8fFMN4oiV^Ic^aot<%n>k2sof+l^L`QWDqk_SU+
zI-jZI7)n+BvX1LN#l-RG*I*fMEh0Mtu<BbYwX3IasR7hXec)s$N<VW2DV>DzqEMrT
znMl=FI@hfS^@&a@rl0hS5P^E|M{qJa6b{aU^<&x_X2j^wa}kCfi2LzdxlE?=z#zpK
z3;|Yr*>+QanZ+!4K(qS1!J0IM)vm{Md*puETh2DSZvi;T8I)u;wz`<i(HG2gYn_s&
z_>!x)pm6_-W}*tThI6EaaTh4U_37BfO>5_?sT(2u^BdLKZJ8XOXH$Hy2YMwddz2wr
zJ8on48$5AKJd*ew=xYJ&F~WCBriXJCM*|uIOI1HGya{dSW&~>E$ENL9Iz0Bb=3{Dl
zTuN{y);&65$gB<Rp_+$p802%u4?%>;BA#At^i6MKNOC#DS#H%p4^}vo`?z=9o%xEh
zp!_lrQS%=^@hxzz9KksJdeon$Tn?T{YsIK?62V9K`x^TFMFCZ*a~M^mbZL62ff$N3
zGnDXoeWSu>aj8K@;Pv<FA5L7bIKw4Q{-TCR(qzCq70yv$T1_%W%STzN)kSi-$xCwE
zPUk={m22yHY-|n;(K#ePsIuU|3j^M^n1-VGN5m%g8&xzp*CgBKK6ozG5LMC{>?n+`
z?`Eu~IhChTL3}KEUbbb-(%5n?H=8#j`OB$hoefBx`^dGoMXDv`$D&-%`n3G$32tS3
zuwM3pPQPJIy^Eg6dK@TRD9g^dt$rR~x%R-<@_rq_@}}k!x;gZrS?4dz53^}RsQlyY
zE!IpG)|q>ba39QU;<P`~tAKN8y*MFF3$L~<D3VKQ@>zUHYT23VrL1y{dT`LXzth!=
zIBxmXXS~{wJ9zh#)1uUVh#ZM?lus;os3zFqlh29S(1$Mr>G-v446Y91iz{vfi#z+2
zzeuUC2+W2^L>S*|^16DH^x$7H`L(;Sma+EQ!5g+@=JdZq#YuQPoCb~IWB>QOkk{s_
zHdvdh*L~yb|HJn9yEE@Wp?m^K6rtDtVS|CDq*`VSf|2)sH7fGM=S8;h>W4ar1`(S_
znK3X`%@K`0W8PIQ@}FBC7$$mHS*4x36m;t*im>tn?5{aMlD`z%QP@a`XgWTew7WWS
z!ENk5HQUjnt|OPT_(33U4hpl!<%w^y*$e$BqEn0&2LnwCp^$+GGjd(zZDv-v+U`ZB
z<9Q7>e6$T84*^)=vdy7Z4v=)vziE#kYTD{b1zn>>d7n88M6suE8$kAN>M!AQfh~d@
zM4z;R$K|G3OAq8oWCx|Qam&fahodGR!BO2z1WLkwjO6f2)uS%XQaDYQrpbs4+%M;a
z=i`ViN01qz95_@*Ufq6ra&#SpqEkmO+EZEQWAo{o+8>LO5hphRpZ6lSGzi`DO-s-a
zdb#>s@&UTTNl`EQ^nhHw;3|0>^WYevp<1_M8~*O=UlE)@ajwJ0#<ugg{Q;8(CMwTk
z$iF#r7Dyv6a`@&LBxXx-EehWqtqfaJcX<`cioSU7UP;aK=g+emIKd9q09O3Vk>R#j
ziZV(jq>Qe44nT$l#Cm?6aMQu1ir_?@aL%-W0`mgPkm_oM^CuMIW$r}H{4BA0)9>ZI
zj_3ATI_)_fsbgACcCwSJ6G{6eW<dZ02L{<y$8-FuA;#vGE}*EgfBJbtSrwGhJsR^%
zcZBAXMNABi4RAk-ai%iDHloKJ!lf+KNMQ(4wk5EIPU;Co`SC%}1n6a`fS_PLEPF3m
zyR2v5^VmjU9Tdb?uic8*fNo#lH`aP!nW!CT`JI_xJC_}aA7+em>;d(DUv_x45;-&)
z5BEi;s=dFvoP>Mt4BvJIN?#LQ-H<Z1y2Jg&0mEOwaKE#=GM=x}fiJ1`{T6?_57?P~
zj$VNFU_1j*rRg|11w{FFF{q72!}Brcf_atiZtX<gBdj&-CA3v|IvL(SAkT6=H3IcL
ze<M1FWSXqE9e=(ZosYioF8J4vcM`U=veSjx`$V^Qs&``&>@^PA`{Swty}t&#zwfu<
z_Q#*+qi9r>{53T-W$&%9I#9sgG|k@F3qFrkK@7%7SwmBkaK209$c#5E9GBhYeAU8{
z3VJ#X64RIU-04{#yZAL-Y`eVENT$BWF)@qG5m1sTiv@vgpiG0<Yr$#Bbw0&>{L>mI
z{poUxioy?p?y?W$w<+qtS2_--zfb)Ea`w(eofl7w8L8d<>?JHFX0WUu9Ua}XHuXmA
z=N#*j;TQ-6$arqqZO`Dys{5<tWXB_&x`u|?c_ujvUtG4-VUI<g(tYT)YE$L84iD5Z
zcwMTPJbRO`ac;d(Yp&?qw-cws1O29Mo$AhM-B?~o$n!HmDts}<7JHqwKa}E7BMl(%
zs}MOTvnEW($jB&yg)_gPfb9m}0l|&fh7S_zD1T3QGQY;pcd_yVnBdKkV+Pnsl3#Ru
zydJiXiAP2Mte168xZT^wYzy$`U%)e~T7VKHEP|7)SOGZo;o+gZycMvS7e0YaksqGo
zy%8#9nIL6z?$xqdJO1@YU{>ja@a<6!4!v&WO=9VZ2jT1WdtaWM<A*qD;YNa3Pko>&
z#!N;`;Se^6^m^+JZe_g`l9278fhfHrzll#;Lf%tff$v9JH_zNC9`SYD+4Hde7*rS7
z_W6SvIKvg8ewmp91CGYV%xvuJ1-Q7QYFPX+HHYI^MS2jC^If}%*zhLvp(15>Y-oJ!
zih&`sJfG!+9C*uqL%+Cw2T6u|{k3k+o^xrzrb<H<)MA%QNE->t9#Vc&@+=);GELm;
zz>3NM_8zM#QQ=E{N=izZvF4ep%e-U4xd0?Cg!m5mlw{*6HoMk9J+0yfLgDHMFE#9F
zX*YyB#Y#B+^G-Cn8GQ@&UDGwE^~ZR>yJ$jW-h!lT^j-dqZS=3Zvw3;@+C;&WR7AdN
zT^#a52h!j|6XqK)yqX2Nw@lql(L|8|MLepSs)9oxDzvCQK|D2SwN5<eZoBGJEeySn
zrQMz2EbYEe8d;{6qq?dqr&3Lh$UfL}x|6$SGy$?TJdXGMvEDW>;rpHL=T23l8Io!U
z;olo<(Q8}-XeM)DiZaRY^NtWXN3t0klGbZk?HNU1c)LQ|3PE;?ZXN{tvi97UxPi#h
zfLTbiIBVOhQ^`W(JC1o$;()_|*=52_<x$#nip$A?WzX=1G8KVF0^Jn_IBH@IGJV3$
z`)jj=EpkA|$1r=$AW37&j28K$?7+}MTO1A?)caLeofp)OgHkW|o*94TF8Z4LK~@;$
ztjq}cKJy^$Wb%+|44-<3%V^b!&e~ruIk@9i%nB~$g}@}YAx%a1J{4Ypr36fXbxF|G
z?hz#s?Y_2#b;=cI@bjh(NAPNhGWK>>P%}XEpGQza7(4E%yZ}I)<|nCfs($yg$s;E8
zSTx=BrzDm)o4yxC0&R+^ySXSPOL8Q|_<UdO^r3~~p`POghkAo@T=g`GjMuulIYVF0
z8+ecbO&5m3Rx5-Ee+0P%+kp#e_n=5Hy{nW%A|iule{U;0F3KO?a~-|5It~71fEFWF
zZ<Y}t#+&5BT(bo-Mw!U{8p_jY;g_6tuHTL+`x*RxE#~Uo(iX*_Q@xkUp?hdj3*7Z<
zn7l=6-$gHSevLi^DRX`d*PTVp?T#6cT~VGhdinbG>zIc#ng|N0!_Kk3Zlz9J2Mtf{
zp7xuz?*8GyC=%2^&pS9lAJd&E_H1MQYx^mcBL?W%i;@Eg6%lInzRtpEH2KiI23s2?
z+G)GbGiYBui%08zO%<3FDW{2g>EhA}N_SY%^ew$#IfHXKh*v&%t1^<(8F@DDh(Pei
z<W~oZ%IFr^Add^`>i!0LJo;x;#<Zn*l%YnlBfz~PK$&dP<cX>r`DimK6y_Labm3Ef
zf379150j3aydw>|Rmx?>67JY3lwlsrS)#wcvUy-Kv9jY<DM~fd5{AIe-r6}WX|`;$
znf~fsTX*4hxvXE%*wglD_P18Xw?}Hk2LGB(iU=|(2r*VEXW6nd)Ghj1=PjG<!66N$
z^<1CHMIoc5vQgT+ejFw2(o%cQQ{Rp<;#NLKo=oWt;!J(ge_LdLHwqZPvsUBzFUS;k
zuL&N(I4-~DfS=LAebbQAqsZyO{L0J{)Z}}(_p5m!r}GjSls`Hg66DW>+S03X^PayE
zo8iTaUu%U2FFE?_%~-mPYk`x^?cvA17pTczI7-#VEMSvCmlOkAP*~v(OLa3aOh55@
z&U5zyeTG*T561abY6-d^>3zWoyQT%lZDbsa;aO{cp;IzOup{i2svI#su!Zl`w4r@+
zD-z#yuQMzRf64+v#Q6n|`W7ZNN<N)LI7yR*F&f@YkJBa_p!ewg$~;-KV&#o8p3M-Z
zC3zNF6XQ8cE!qRfST&Gc`Y~%^6VFu6c;FU;j+d~IY@5XA<ST~mG&Il`1QL;X-F0V9
zliB<19b+XsaossG+J8>P9XZ$<c4E-cG=>owBqIW^=7KVaV$Ee`_-x}$@*0eRd<(k3
z7Y5w#e005Il7`&J&$~M*nl{^VoGcG8kUn>6F|mLT$U|avP=xJ|cWBaNh#Bk*v)ACI
zQ$g)=;;Nn;SaUu+h7Kf$N*B9YnR}1;^I$gzV7?c?M~hw`t#z@=bS~CPjgmcT^FWYx
zbC>rrSEzQAG&dD9%a|RP&UQFoos61MZk|TneO@(^4`R~J>nIlzMbkHS``xvNnIWU$
zOT|F3nFaiUHNdC#Jaw{+GRBVv%KaE~-GG1UtrYZR$*^%1_6F^gJfCRwuK$yW$<{?`
zoIscMbgOhk+HXNbk0&f*YraFe+th;aT*UmMr)SOkjqt(^UF7h!!rf21g0f_T(inC*
z4WpBf%#E&~t#AG*B!y{#>6GH$x<u)BvYBhO{t1Pku<Vd&&-u}+;&VHDv#P)1$tlU+
ztApxaQ4zrYcNK^BaBqdKWe72<KzxH;7&BZ;3jGiUdhI@BQ`=|I5WGvt%XpBVLX|*9
zh$n3|`!j&QqJQx;GX=;8tBrs8M9A-F#}Una5gh*<)2U%K#aHV3=}KN=Ny90BjA`Ju
zBXjZRTw#5EeQ*f`XF}wZC0TW%$N`C=kJyww-4d(*vt!9d{6M%rR4(lqd$Y$lm}F(q
ztdpP}z$1R*7!(SHkH9hz#%MsYgeE@SSbx%KnEJ5Q8SvTzg0qLkHIz|RG(*V&0QV#Z
zyxQXX1BiOm;3xrmW8+ABE+{2uB00ZassjRlZ!*}ot$jFj)V0=S$(>BuaVKU`|GgQb
zCPi-;IRT&Nagfp$PQw*aAtMGaCs1(4GSBET)6unwJ`J~w2v7o>+F$8q4AJa2>Q_sp
zfr8g2kby}UU%3*0E>q6=bDiMsOc;5fxuFpZ=Nt?c0&%K_L{@845!cT~nP@Z$rM8ny
zkuJmgx3?Z}QWwWJttadf+2+?h-hK&?^x1bMlR0J?G(E}VNC%Y^$;Yy(KeVAV!=-B&
zNhzIvLd5#DFB<}5r1s=9Z~^(fI}Qhg-H%GEb-*V(YfudgBVmx-&BfA?{QNkC(A4xD
zvm|ziI#fNy-<r}#n75@EcLY`M3I`a_6<*kv`v)9iboeLVWYedmrNJ1QL9Mepz=Go%
z&nhRQO*w?nj6%)`1ElKaWeld(rk8Js;zO7?m^bYwCpG>Gn^1rtmU;hcYcI%$rn$bp
zE(3x=v`kFwcI1B0G2@>Vq!crqEDbew`-B2PzQ89_nMGQ<_gEtOh<|%9i$bulr~@!|
zsx$6%0M=fBys#~wV3coPHN;M#h&X|edp#8nQyrf{JHdp$n7{3EP?M{x5>WLVlvpwx
zD#nj^KX$976lO0L8^f~Vk9TaU{&4P{!ardC@6;QrlUZp{i<;lDy+7WZ-#?8}frDVo
zChJxb7+w^-nghDVH3z4qrA?okn+vzI7y+5Zk&qkV@K(KuWJ9jr?2pvK43$XG$buo>
z^e?2o2lE3tA7`!+jk3BjSH^1#LbSUaO-5K?br3%z!>b79$@AF5wLHgFx@E$C=^Lr4
z`LPSH<qLxryL*GB&5;_THHRTKgx?+b=WKCOPp$D|O_43n%kMmw#Nwx}kjbp0%C;d1
z7WG;Q)${p(uKe3>Sqsw<Hu5C3pR5l4^VomBl<7fll^xY=6#wTP{4r$QomLM&=kxE&
zL1WSz^2`3sG1F}RKkN4&Cto=`_^<!J$-frR|1n&Czo;8W=9`}z^6686kDivHW|2D9
G|Gxm3%Noc4

diff --git a/docs_old/cookbook/images/trader-joes-receipt.jpg b/docs_old/cookbook/images/trader-joes-receipt.jpg
deleted file mode 100644
index 6742d39462449d7264cbc2724668505f6ceef7f9..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 430638
zcmeFZcUTlj*Ed`}ff)vL1{@I-p$AX~6eSBtXa<m=!Vpvps}cnaL={k^NsF>52$&FI
zM6x-py1I%HOn_ictAc=;EFj2tdca-p`?{a!`rh~7_dZ-beY(2p)H%O8bt-f>Lyw1E
zA)d=D=UE8DFysV(Xz2Cir4G@-fe1M}>moKnNDgr^8e##L1qTm!Dm51lMU0LxI8(=X
zj43@A0zYAduK->+l7|Mo0hApD=M!)k0?&tYB^+aD|HuhWWk}`RhxAFChxv3mLKoq1
z3=LheDgYsIaCE>b+?-NZn)k=q(Rr>IH=JgmKg~!VYS1?{wlFZT&^N^mOf3v%SeO_g
zU6h+RtY<wav~Pre3cSYVVI3$vwMY29z@tnSV!%n?(8!RUIjr*=$ix`odw}NwPvOG>
z9Ti4+Mc@@j`0>DF+lS>~l&x7DsoolpPRknBM;CbdNV*~Lj1k@fc;-kR5%4l2ybJKG
ztbcTt9pQPD9wT`?AWvo%O{%|<u_@aA6QA`Hk0c{D8tCg|l7HHQCK<*b$%Aeqd>a>`
zafScXiydG|ZISvx(NASZcu)t?f8rTM!}czs>>5c|hCDJq<zfHC%l^c3e&Xdvc&c8b
zqG9}fL8dYA0!Rx1e)&iqV5OvDix45=qB6MhhvR1tF|sf;vd}m9k7;`PW)}M3QU88+
zgSMyS>>QDIeB=x#Bm>8f^pn81f)ZWhhtJe~O7DO3XWZAvZzb;MvpOKc2REH&zIrtt
zINe~HsTm+BEkXGqjq*1{gFl4k!l8hM@%(+nNH=Ltgk;foXm_d>>Oj~?OTq;<!oP=P
z8uU)8pgxGEkMvYZbD5v`u_L@x2imY4Q^-#n;ng*u01#XHH}rE%p(=+Ohn^uj#Fmj^
z$uQY07E4x^&5`Gil9!W{S5X|p<xd>1t~PPJsw%EEO&iyks;R0foS`#S-@w?|SY6xP
z%FNJmnvt;~B?Oa|m6ey1R~|J=*-)S=F#JEJp-QB{M(;5&Rs+!#FuDRZbPcJ4o=k{<
zR1h2?=%x&&42vzxk%J7U0db5@r!nYECId{z;;3%SP+%$w3~Xh_xcjg)!uf^?+xN3I
zMaM6X^|;rjW#k)?D9ceAr>ru5(qwI+j;^tZ=?pV-3p;y<S&mN5E}mZV<}dJGxX8~x
zAaF%ca7bj-+I8!rV`7t%Q+`d|yd^CwJ7>qvUAy<>79J=nE-5W5uQ+k?)al>PoIUr)
zm8;jT->9m-d8@wR{)2~)9zS{d`b}fg+vb+mckkOfI=i}izV`O@Q+82y{-9yQ{>?50
zu#3iE&>1YsE{wJw4!Qz^DKL;xv~_3sgpbiMOkndx+xH*8EURhc(KgmM;vPpy%lP%A
zcFHuVW&i&SOZ>mG>~F*VvFj<4r(@7~bOmICZUm3Exu0=uo=Q=cW7STZWo_kcj`MBy
z>F4Da4UURtognbD8@H!aeRH%qY++;K%2xMGi`3El-1z>1(v`R&B>w;lnsfC~Wt@(~
zm2D?E6`FqcYzK9<XDvOTZkIcC%Fdz-o9W-KxSd{a!+qi3Hc(TP_hQDfuk$ZFb-HZV
z>Gyt0{l$q@C$F;1nqSw%W50NNABmK=+nDN}#ti(RrqOU}#@0~(wJ$9yl;{m+@AAXD
zjK6&PWH#x1%;DcMZ%h?TI(|RQBCYFVO!L->6^AUg?z_J0neSP<RYrR(u%x7i`DfPE
zFODlTZwz8tFJHc(v!n<67$ceH8p00Mn>OdyDZBG#8Gc(b@zNrV`-fJ9tIh~>YBBQ)
zbFNzS+cJ;e_&er=^ha9UICE@V;NuEa|1+fzkBs+~7l#xrI-UN=bFpJsNsvr{>n69O
z#iq{|HiY``&^zk0qwtrSvL|?KgF@t*xy+D#zn?By>T-PXO9!oOvhzG|(vRq*E@{|Q
zJMrpL`|Z^(``0}?o0N3<YsF&OT}1P_jLov=uLjMEIKz(aS8qSE?P;@IQ767<(wxQ!
z(^)IWRAFwv<{ht8++W+c&}-?3X$99ZW+>R4Emk<z!usBS(Dsd7fxh2+9iuUwdEZT%
zo_F1P%Uxpg!O3$lyDxXIzM7%Umx^V(dKYw;uG9FvZZsz|&M<g%&9#;~uM-XeO}g3(
z9i@k_hY-n)yF!lNJK@}bgG-ss1sh$Zdcs*ZV&?LOfin}G`h7{h<M&4$#imcxqsPxJ
z>Sk=XpCFD`?;9waxaI5S`=o4W@q)mYG0e!ntU|u-bd0_(NZ<bbcsxVqn6YrrlQ9Z6
zbZKi|{5nqV)AbN`<p)i5$@R(i2EUd@Y3gKsx~i_<XesYLHs{6S8118D%XhNUziSVn
zqfY)C*YZD4QK^4+cJn|+Z*EsSnp`VaRyFW@{JSO0t4@Ba>z6OsA3f%~L-G(ha&+Ls
zQC9oOg0XF*+Kd0}uG!15h}(WP@T{}e4H;Tq#!-BeO3}lQ&3``VTAQMs;+Xfv(r(aQ
zCn0}k?z_1&*3<qpB42-&{IQJb9rUngTlNx{-%n;G>!uHUj@$cau%f^r{<!J6!e!UT
z+wo#viw#RAt7cU8d_7XNexlA<Mg(&*E4Me|)AXZJ+~ZG;Jr+MX5n{ONDJQv-^+-GS
zLEF}v{NH<&t`$9+u=wUPnTIK}K7L-)H}+{oMVHuyW17;6`aS$C`)i(9KVvo$^@}c^
zG|u-H&6(qs5?_kGn>=1QW5&bS4Q_s`rumEHW^wN<u>3T!W{REhy5jIWf74wScXwWj
zHvW2M+{Hyzs!4vgHp|#{)oK(TOI{XT@v!5z<6gJKdRfJ$3$`<|5C?t8SoC?OhhJ`?
z>+CDFCVyTm2(U_N&t5r3t1e}Bm{Fw}E1wm6xN&x{!;8}|_pDggzawiGd3|*6-tfL1
z=QliB#WFrCk$-_|CvRD)_oCwI^Xk2Cw=7WhyV=-XvD75pCZ&ZzPqj<>-gbnXKQ6uQ
zQp%djZo46rk>=U!{kmPVmU%Ad1|!7aX{Kmg60W}0QGVuryIPk_!tv`Riw>?h%)!c^
z)f)6f`MaR+UelRX84q)A+#Gz5A}m)nWPj9MX7vZdrsm+M+Q=ak7vKF(boA>b)4TZO
z>E|oP3?a-!g&m)qv5I3me$3-RotGXNokQrMgfUn@gjkK|9J`AZOOFjWOc?jBuYR@K
zxdEI0F3f)qE6lod2px9M+1(X#`c&hYSMmM#ed4CjMf($K7dj0DUkdErlhN1z;A%pB
zA>I?R+_uO2<ju(5-0(!ffL?DyT7FZ(e&ZBY`G9M1!qVsi9nZe{{wY^^Yur@@q;FaE
zeQnDb1+Ly7o94A{Vq+T(*b`@5>O68sy{g0f_%iI`i{Mn_p5+(vbLwtAPg=UA5{1ta
zE|{;%i@83{@vwnxecGXp=W)j*8N5%*t5#@(Ic8sr^4|nqIC1<-bswo(?>BY5gzhf;
z%d0^4yrku$&lLr)8M{D%7Mb+uKJ@5;_Z}0bDqL4CTm3Tdd0?d~Pk)x~{49OhX%Z}d
zt~iC3D4eyR<{?Zg_-~j9KNPQ2@<ndh=uPhn2BPOze^~fZ<CpJhml!L)+Sl*8Ql-8{
zZD9y2({OQT`MRT*2D|5++Ula(-#`BMa%C&-+adJmsNi$#O5cg)gZE~4a9=OGRXWpR
zb!aU=!SR=K{mrCc>Yf|-3a;dn`3aXW4}XPi1-8Vz+V<@4?GlFY@wbdVFRg8PE$`-^
z7)O@2n@i3&rf1ZXQy78YB)Q8brOrd&j+Mtp$IYGTU;6nleX_xz4c+-|!h_19jwH2(
zC%a1rGu50!uG~o8;8Q!ztsyPA<GlP3s;roQ#`Z>VKJjrvXN7|L%_qcwbBQVYQ=ive
z#cdMx@}hZmgMVEib(Uan^LMl^l~ZoJru_Ze5Ne%uc6PC?Lbb`uUK{nct&2asJ2!;7
zG=GV9+nV>dfPA`Eo?WK+Ky&r0<1^1h#2p_Dxm;GX!&h9JH)h+EduA@#g=;#(h^M6y
zzjI%f96GKk<n(TT;3&duzE_+$lX*^TbbB@H<-VTXXZ~2xEW*^6=CErGr;amp+%<Vp
z&0=M%-XUat+Nj`^^5s`EZn+xWr&qOj?tK;(QNvm@%c^45PGeLs2UcHNCpP236SVMw
z)`ef4#N`Zn#-?yL!^s`V>5XP(cklj@bv<XfYSzrrrz#zG&pDF9ZYr;BoOf-ditzjv
z6!D_%(qMf3;nWl_{ipn`Pw43$yVCu$%svOpZ>vLr?)mBy<5IjQ2FtK)ALgu;usFqs
zM?Ia`W@Wka`@E=mbC;~S7n(Y#A3w;b-+6grt-;AO8@-FJ-*>F^9sO7Fyi>`$pY8nc
z_|2iXy!C+ti!PL}iCyWl-mx`acS+TI$JfqHLbXd<wX?@<+!Xce%qY9_o4j7#8y$A_
z`Tb3E?2g%_2(}pv@X73&%~3wRb!&W8cd2KVjjk<iObe!~PhDvJXOP?&tB)>5Eqb#|
z*hA=|+iUTOb4?<fFS&MXf7b0K4RqW5`Hc*&PuiNs7Y3i_I{lIn-s=^RarA@tPW!hv
z`@9<;*9(`&491kpc9$6Rdt&MZZ>swGy`~;~k@sDrXsYAk$&IbS2_8+%jQvL)+v3&6
zo&Gv~2<;!=dH(RrWe|;9FLG-KG?mVM=!|cj^~bRSg}YmN=Vs6Gch;Y1hawDT-cOEn
za)@Qgs%NWxdL|RWeQRf@F!g$jm1}(MVe3T0qS!j7C-Y$E>Cx6ccb#reC{f*7<aflZ
z;$V@+7mXEqcz|yl+vc{3Wsh&P#@4*8sbybo=p;;;y6RhvQTht2Kkmo-C&_bK!tkb+
zql-!hw{}jEOP#bhtk#wEQ|VD}y1za6=!D}&JDCs8EuA$Xoc`=ETDT#4-`LUX64L40
zbgqnTQq9~j{=noNQ>u?lTWEUo(%Xr%A~Q<N23ET9*<8Ljs8vUP$(w^A1s$x5FaXEc
zbd8zMDlc1$op-sQcFb%DRn65t_OK=ANzKcJm0v#0-}tC!;Qir3?$m@+c|FU%|G9{l
z|ECS7Bfq8S_c3>DK7Nl<QGeD^WMCbgxk)4c-tTq1<9^REqWbpn&(L5pjSbqZr`qLT
zJR!f@ymswae167YR}fAA@b^b=T^|oV7(y=DHz)QDv??4O=v3c<jBc-rFzEIn(qoSf
zc6&1Dg;=l0C$)m&$;btJa7j0CM$`hElqH-4g?{a2&+snV31htD>|5oy_(8Adlf&$|
zW#!xA;*H!-R^@xsKL>Ai@UD#w-%8YWm&9y&(%E(Mz(&8Us&<*(X=gTUyvUFLYvED*
zwoLj}(>%Xlhfv+tm``^qe9g4Jk8Rz=H*b5ezbAB%{b;_zhC40S1|}bvJN~#wrJ9UQ
z?ZX4(b~aV?B~2yN`f6;pmmR^U;}Ho%DAtBqwd}B`(lwQYGf&6<c4L;St1EX=*XS{x
zcji{C54)<VmLuyI_;f=4=|xk*dOoQBMxH0>+eXcucq+(?+~l5evqne@nd5PfkWE7^
z#_wIws7X3S*Y)XXYr0v%J_U2?Gb6<l=FJ#o5cr%>{&e{+KI!H8-#@FoOK+HdscB-A
z&AaR&q-h_fdoJSIaXhc_h{5b=Bfl_6j@NWr8F@=DAmirSy+`M~DOId`Y%qipyw=s7
zP3DD!(+l31ytS&GyGlFPpi(1e&Mh;Q{`$c^8SUyd3Bex(1Mg6NR04s*pAVt_*3KrY
z>y`_3Y@!>kXRV(aZ-IWh8~-Jvv)3agJ^roaAnl>xS*LM~0{3$rGv3IiqoLsNgiG)C
z4~A54v$5^<OeXSK@0`;s?oN*J-Bgn6UEo%8WN;s=tvR#Oiq*34`H5|AVV7#QR{b82
zE!en!@Lr$Cs;~Ud8u^Rw4Iy2-ODATu%&)opAhlWeLj7y%>p8OmZr$>Im)4abn`mhe
z$Xznyd-BSpKcWW?WW{U#@+GC(dcrBZ$>$Qpb?*gVj0(oIWj){CQ^>eHA)<ZTpWy1&
zr{^1~e{P@VxaVqk?dip5=YLzk>q`oQakb67Ou=8qvZ^yCk91{SQe7YkR~omOtGVWC
zp4v0VVr%!!J}nGk(~515#?_C$+NHiY)v?}hri0bx4ndz$N5@&qu0|LyiL_g$v*hj+
zDWX6v|CV>3S+70IV)!$5kJrEMq+;H8vahEMb<+AWB{xQ^IA7Lgbu|r!ROPua9iBRy
zn_2SWuNupXu05_ew0pVHdF{H+hW6_kj$!Kz^`AA%TweF>f`7bCO(Nf$Hs<&*<oA8C
z^~Lvy&1I&235FHnN!#L$`aRBiUx;<jwR6s$MtdN)Gjkczy0*b<hkS(f8n=eizq}9G
zS~WiH`^W6JiOMZ1FB%$Nl1d&!D7e{e`qwQt-q=nYLRXF~Y5W)omY9B@H{WN^#d)1J
zF>1E&%|ARkv*=s|I$7}j+?kG2AGL*y+f7Hq;@&Si<v94zMD&8aHO!!Q-gQ^=<x{FQ
z<X1e-T*B>Z4&GGHYV*{;^w&Vj<T3ZQBIlmMSUt533SZ+Fty%Cb+2adzbfgw;HKyA*
z(6y2)Deh$LFY4S>uw@8+s55B1+iiR<`%$OyJ<WB``aG|te!JSi*H^8#w(c9$HC<u#
z%OxYXndd*Ly9C#Mz2s7T!m#M8_J_MOf<M~mX6A2dA1IeqEBbCCQaSdO)milHd-XtY
zAFDTbM*N8&g{AM}qhh&N{$yRb)S<U9BTyO(25(xs_pQ9tc2d}OcaJaYrRos6u=1c>
zJLN`&F}v=ZUGUP&__F?mri^8`4qh9hh?>|-jz5plQk<zCaxPr!W|)%U?rAb|h^PwP
zbec2Udd$X(h+gd}iBswv<tC?$_jYUR7AbG=d-Sv<>qYKZhvh|rUmm~9e^cnZ_qMv(
z&503KHdV$aEHAAPW&A31jc)4sehM4cv$2TVwUcG8CBOcz_nCy(OB5&UH1E$%z8hU(
zim$7#`H0Na2jds*SrjVRQ~jdg^^#ZNPf3lH&U@Bv;Z3{a$NIA3i^pKvGJXBM2M#Ja
zYK<<;oEw4P=Du8AQwPhd0+=KpIZQ0MR<dM`b<VF&4mfRRN|QKw`eCp9ht6v(ukQ-^
zD0e3RTutnQigL?_;yU|fFIIb=54QZ=qTQhO*A8*2)k(`|SDAOxTK*cV2R+tK>K<iU
zJHt&iReot@2WTceVqNMR2t3<dAK%~n*R9_Y-j=m2S4%opj&*ycMbD%6##?&4jlXDP
z(O&Bva-xUTQ9RWkZ&P66dzO25iSfOo%W@OsnuRKTn-X`ei=Ez4Hc(w%t38p_%FVGR
zGz_gv`Dk@*TUx{oHwRjdlcSgVKASt$t*l0(df=!>@D+n6!Lzv?!87u&)zNRdUOAu@
zALUtNWn(wTWV_$`SdQUf>N{Is3#_+cw#{h+xzuinOi#(@$d{Q}7RS|Hd{rW3Q_hZk
z5xsKC>)$L<Om|g#?yG4oeWS7qHa?1ftuf)M((ZS|6NbMOYB%@J&OQ}5$k<EHpAepB
z<ag;+$mEnITSmVXcpcX))6A$!+gB*K(dRq8<@0U7nH#uAHF7iW95#KTIWyr&jcIz8
z1vmVc);9_p%d*VR7;N@$YT{vd%cH{<F@l}Ei$0vP2vN_QeJh(bd(6mEVf&72W*znM
zP(Jh^W_B1?HKl6^aqCIl!++e=UA3e4vifuu=GL@n*{#EayG-xO+03tb`>Ok{kXJUh
zP~W1|&GoMoCx7v|spxYg!M~g2zullo;nyYGV!oG@>MLzCezEn?goH<9@#VAZXxQcE
zU9oEPY2o>!&`e{=D^J_zoNYnKMAUfu`@Ng1)P(_}4q{bzS7Y(gvZZG7uV46eh9O^>
z@ce>(v?;Ed-<I8-GVX<%m!DiM9b(Fd`JcJ{T0^M&?xh_j7s$(tZ$D>OnpIm}I&86N
zb+P)#86738owAMEU%1uveeYVI455gF$<=SS3<TS|#Q#}PepJ)$f!>tQt)*S1o0JDa
z%949L-5&S02GKV9o%dSh^zM>c6KA<&_w@<v_t(A~FJV4+Y&-MjzU<;&&(0zR8LJ^w
z``Y7>L1V#NVwB?V=DIeoye<ZAXnAksrr_Ax(pIj-|Ncp^{dDETEdv#r5iJt2_1R*7
z@TR)7&|O^v4z{A0+UIG0&f0-cw-5S!C5hHORp+^b@3rn-){pN_I9?ezP~oc(tzO;W
zv?L<j^^NW?$k*@O$?l_0Z7T-SevckPp+iVy(lSFn-*e+tn4jIdFy3)M<Kd4R4Zrqo
z+vs>D=nwqQQC}DC)>$6&ir&zWmL@BHeP2m2GkxK)=*AMO`23K?iQl8yf0=$+YOu{_
zX~;}-<?I$G<*e|*nLW+tYXrv^C&)Qzt;Iv99x;sHyln?Eczs2NF3-Jt1Nl7Pq)r|b
zI~!IQJUhNSp)+C@cUjd~y(7)G&7`Zrqr^DJzAuD)c+ciM#lEd!S7)woTG3E?=~4ae
ziSnZv1(B>Qb4&G>1-VX{JG8>gl|R`UY(4j^_JxGAuBZKGsG&~4C?kW*t<jcV#}`hW
znlkaCd&2pG&iX?EV-iJQix<o{W2hMs9OkHp`{o))VKe8ie!Bm@;d&)rQhNGwgO-xW
zcZ=dv+|R8aR~aha%etNK{34+@LP(gtw>Q$Ac*lNwmTgr#MB3^7du(3V+^SeI^YE>d
zZ1l%l*9`@s*S>3-M9kS;FxWt-Eu!;Gzk458>376NFh_2Ru-bKB#9U#@5NeUAJwB?{
z&1%fpv!y>*Oh&Mp=HheK`V_1!U9C5S^pxX-bu6saW{yYR*OI=K&y|KybUpdoGxa%j
zw5Is}yFxs!d0<<xoWp<%_fc*vU$vsFpW7y3JkP(hzgX>wjcsoHo^3_q$rDfZiEk{J
zY4JL;*AsT^8a$L6KYw&PJ-NDlU~ad`j$0~qVc{Qjl3QTTlDk(nopIfml(9U0UA#F-
z*d(mIG<Jed?XAi|^QsS?pD*{#;@0QZytprB>b%m>{#50+bnB~o8FC9@Axte_m&QbG
z^Vx{i|0#Ew%-}4dKWd?dvb8>&CP(^iQtZBc<BW~9?dT$&<FR+|w#9QouFFk2KkLBc
zHC2`=<{!GsTNY;+?<O^qZdQK!&e&wE>*_Ie-+BMrk@uHFCp~z!Q6rIE@~ak(jrp=P
zmC-pEa%(ZWXZ7~BKaW`%Y|9}|=<a*fvaR&l`w%~3Y}1Kt&Wq;WIdqjl+LV=#`r{9Y
zR^slbSIeq~P(|@Xok}soaPgtX&lf9Srl;Q<Tj8KvExSVYZQ|p1dG}w8lG&Y@J6ETq
z+xS=XFz%}G{w|%J9VgFQZ_V4}d&*#*cH;U!7R*u;i#R(q^V)($D&ePI$iKcJ&UEUF
zF{^6K{bLA)U;MRVF5|?KEi!cn(>H!^DXS{6S#<8c!h4;Bw`tX=v7^{f`>S(*+Nq09
z?pNrpMzhSS2QZVS{QcsYT?(5VVXw$lEi(SLQE%EY?ewS@LulihylLYGwi7&diRBJ@
zJ4{F#k`?2_ZNHYM5224qz4v;4t>Or=#Fwt;bK-P1RWJVHR0WfdPex6Rx9RPrvVm6q
zHYA^9rj2$~Cv@kXT>3Vn(Y^2Ji`tg7H9c0pl&$i|I?PxdB?U&OV#;TnRiC=SM&Y~h
zgQ=2TKDDbJ`e|CP)8}Vg+57gn`sa)rB}S8D>y4{zQrk-y*7hxjcJ~M7uG&4=$9<c-
zyT+id;7Z{Py_)Ov57({i9j$hKUi>vxxqpEq+RE~h+|y~1U!{(+J~FIX={5>~Ci5r#
z%A|rTJ2!Ujyvb6(F+9UjXxm$LSh*r)O^3H>w5p<2%%|B}kvq~gUu~JK7JJk(UiY)v
z?S&;S&BiYtuWdYSs~Y+F`IqeN+s~)d;`j3}E_vy-+39jd$nwrSzrK3qZyUDj-?*`q
zwKOli`^)^0M8{|qGD_~$rwJLXGoe?C^lRG3uKVS2r6A+0)Ay*l-K|H|oK?p@E&EvX
zs!RR*9~x(~W?=^^pSm@^>wl0Yauq5p{$!{uh*ZDxnV%lT`qSido&kH0*`$cl2FnN8
z^(jS$atoXCZ1$+kQApn=A>;f1N-=DH`}*8AldWlnS>8Hkm(>auoyjeZ3|up<aJLzz
zkdt_>#_4NMiD1RKEOU{r|JZ6XtH|_t-Sf)NuawmW1^nThUa(e$wu2G*SC{&ZAcw1-
zM6I5p^~+HL^Y@Y6MLXtc@31+qIXeH>T<gLMC8w>7yR?{&0qQR6J~m)gt2`2&U&p%;
zHkx$}3CcI_IgR~ubX>~J+u7fW27~;@6vQIyQ|)8E@(XR&7KcqZ^Wb>bf!i=mH6B7!
zhtP%&l~Q)?a<|uM%6B@d&U1XWJGOHxJ;U7E3`%x>I@R)0;+HNNLSxj+u1{{WpwGoF
zwd^ih{$huN0jI_a>(AY~FZ@|sDcgU|re)RcJu#_x<Gbr<-r1a(=~Zja9-Y;;ad+F+
zZ2t)F$j?m+3YI*(vwZcns3V<Mo!WaX?z&&m*Q*J$?p!c`+?y5F+OKC_eXpa!xp&;x
za9v*G(cXC}yPY}J2M+pmU-x=p?xo><`^(kHfti2AyUch#HSO1c3H}p)TfZbqb?uSH
z^xseHOxIU#VW@3*Q#d1iV$QXQgJ)m$G!6upEP0+WZJr=4{QMc4SHT^{w`b=pA7eT2
zypc~Tw(q+&)l6V}TaA0hCFkyBn+Hvw(ek|IjRkkU@0hwgSzM*J&FxR;r{%IcwD#RE
zxv#0I{Z@OT&7uOWZNc81&dy{*eZs?CX{;sn%AOG{Mq13FwH@)NreWKYR2F}YRt|lU
zp6hGkUC|hcb(<J(9Dnyd`*HD}RLu#IwI3GEews_ad*rpl{WUipRPXE0i&rRNEPZW5
zFL>t}q89RDQR0z0%deUl@rM`OdD3j#%jsxoH>%ILvPFa<kc~rWLjI>-Y%m0AS^0-n
zdL<eS*cTl1%5jo;d2YZ-{noB2FLt$uEZ?$$$$ve5TT2@sZPn83-PNWdnUlKTuL}}m
zy^=D9kb4oscg_%!`#yOW>&U&n`xm~YiPPie3U)U?p{D$uM8cT;K3=bPzI9#dar?7@
z8<yAPQuRFB(`V<LkPXNC-{dgdZ?FdB#&WJK{-STQ#ik~2x$O<ix~PFBYGp08KCO@o
z4mLWtnha0MSmm3<_aFSL!I*t?J^pQGm__V@xYrJf(erD=s+$^Ho6pYP-Sg}if=TYA
zXRm)V`1n`9!jZh`*J?JpUHR>Bzyz~Z)%vfG89XvCoxM`EW6Ifxc}d$GS;_}Te^6#V
zQ!$yar1E^E#@4E!uO=_Uw{Cl(86a4OCuKRpe(RI1-mJlF_es}Qg-y^lyD`fAvu29-
zxyp&{p9-3Ugo~`zR%S-AXqunF`J^jbuYL2hexefKo^=PKKeFV$onzlr;Px;xyVb@b
z`K809Q1d3g=U({{rvfUqWTy0~dMF<-hg(>vZQp~)cr^D~rs?Maoez-{^lkTDqn|m;
zvUwFm6T4b)Yjz*gJ%8e=@`{<eTgq=(>6Cm@GVx87ADi0fIqSpJ4Rr&h#kBE03rDMI
zX_~j>9xL4IQnFiBrdY+|#aB(W;Ic$)z_FwF!559+mR%GTZPzxdxb##fJvsEog+G4#
zYeg|PMX4vl&V|2Dy{3GkZH7~&dds`H-!^1?&bSJjZyrWYTpw4t4+|&v``xSTSO0!m
zJtU`l`=v!wWh#+pQd!r)Twmjw*puov$fh3c_rczQ_wr5Nx7v*NxkReQkOj-<`XAE@
zBj>Ffm8E~-tCsunNu~-)FKU8S){Q#YpLVSI;LT?r#O@OgCC7CbHKfJb?~jRadszO<
z@i91qbgI6)G?_a!Oy{v(Px-`1SJkM-Hc6L>oY9K%fmUqVGv185u!S;uqe*ZVwGFms
z+4&glW}AJnj!CW?bpoVm^T$YIgXyHL{zC5+E7zzm@$0{D<D2xhZ__)5f0w4sn~=zd
z+gF}#)J%J6R^z4GezRn$eTsL^OPB1VETgKiEB-Vad>Z6{?l2NG``ncAf?cCBnshZM
zUA=q9?D>IU)N=IQ{fDKYS8^@VP_c2mmcRDhgni|Ck+luwrO&ps|4z^h0{bKMEjAR~
zI{ejG?G%}DbMrPo&yG(`>$-}0H;{%&V=2GgRU#8(z%v@3=U3MC&O^x2mtnGsC+YLk
z!ag}W#-h$qYn|IC?@zRO7*%`P<Hk#8`n&n3KW}s^ORDa)*hD+|=3CRcqlTq7YG<i)
zCReLw*VdD_KW|F*;&$YlwVbvJahdddrhI*9tunpst_gV~+|}(o!S!&9ulCyOJ%m(B
zXJMZ_^wDS?`6G2x)c0*>9g!_$o;GNi`AqxygI*H%=bRu%licFTUG$+R@P3*6`k<(&
zFpKHa*F;Y9@elP4nC2I{dU~`^7`*GEKOI@xM2Eq9Fac4xZ@`M+HC7s5&R^2NgZ-^E
z7MRS{pBrWu5ESeb8xi0c>+0nfyVB3xU&F>4w~V%kULCeNAj$`iUcG8fq(!uqhE%u(
z@Ko}24P2TfYNeHiH@qfh7a9?O8&5NyrmqLDyR9?S02vYfffgS2vqmJqm6gVbs_WOU
zpSIp;T4=<I=?3QJ=F{~JryCmTK@Pphm^D#8(RynlHGfF34~X=O2o8%14qby&5`BC_
z*G5@sK>fpttqz+z_iw@f5rfsMDQ%^>BcmMFLE--mu}H6&uz=|v0g<6=Bm4p!)&;DI
z()>}}-|z3TVQV8+Ne%G#n;x(#U^Qfm1jG&gu8Mkl?(f``!7GAShe=BSvVZ3S{_jlx
zQF*Fru+YLj)Nd_SiL<?x1|`YDKh!VS-$Kfo8^E8Rk-wgaNuZ&gvA@2do^N14fZhy$
zvl#}y20jLU<_16HIIoF}@>$~-K*<5g(}F>WiK)5Y3}Yi7y+9LVb3J1peIq?{gFtgV
zKSN(rb3<Q0fBityALZR5g29b^R{g9GCC4A+_?a2_LKy=+Q!{;2J>x)gGmv9!sOJO1
zj04Tg4E+MkHE@4Fi&>!&t9_syf>-;j2$&uc7O+ACr|`0{6+2sL7*5k4xfHMRi3)_g
zRvL4H*Q|{mN%aa|9pD+|Lm_7{!${xU$jo$xsgbe1k<pBi9P<MrBEerMQPQ^;rOI2_
zMFjXng+_RVhOV;G`1|EX%Fl)&b_$M!+c6`a>lqL}!mSF%rG8`K<41X+l}4n`x&VKT
zkwm{BpEWB2pg%#I4rdGg4-osB1^W31`kLwK>j%uxGxjwy*7GqoG1v3)pW)|c5&+kf
z^26mLLj$AM`$PoTu7LIiO`)}=uVLasDGEA2@~saFpiG4KGW84$^z===M&8Rb9e%M+
zV>;+Ro$4~frcD2LE2Z~->RvyxL!YJCze9jE{$Khp0{=zezX<#nf&U`#Uj+W&L*So9
zM8F!D>8yt(!q6MqF=sowWnvGvS<Z7D;5B%7g<Z1BZ*?RCUhQAKCMv?iQG~-A**L2e
zzIma(=inlyk6&b%tynx~`0eTc^xN}r_zn98JxbPp%JILXjq-=Z0=!R-Lmt8}EFuc{
z<-nV+j|!vs-N5r<!sZYBA>gM*fP%n3rqX?f`HvL8LdsJg1t5NemxmqLMtxm_ulR}g
z{fYMridYSKj3JMf|LQd$pXNQx`&0ZzieJ5IE$AjC>b-dRfHm_x;j9IRDsn{5@cn@p
z;m8-QMZstlib8sDUIQtS$OG<+{)h7L5d@SMLn&V<6^#6#ggpvHtC5eiJfhkF@&NsV
zZDgR4p)?7T7r}=kFNcOYC&33W*$5404h?<VKQuI02%j)KhYyO@{8K(uf{<A^q-Xw<
zrd0q9n~c!!*ZxWK-G$JZRD_hDg!x4HNb8})FD(#0gX$<nXw*XZNU07!tXVva8{DJr
zqXUl+@&c^p+(l>)d=WKy1;}3VKiExuHT5@c|7)HR{G^{v@JDkMxuf`s3X0>D_{!==
z0(DhYbsN3ulZ;$!UFSO6I?eJ}vd-UQL8$jEr@-`}(Ac=-lw>!*?RzuG+;s`b@zT#G
zl*WxypP+6j5Lm{0IeEqZAJfnk_-sNB$qmpk_+sG4XA|S8uf`~%pcp2ML8mcg;M0R~
z|8@gz!<Pj^Bk%6>=sdW7d-X7myR#V{Ph2pr7S61|a5mSWoRiEzg*rhViIp5t3XTYb
zBq$TIGh^`I%Dc{F%%`fNk<{-`1Rt7Ugd9@f2o<x)BE;t*4h8tXQ5ugwn}L|nHE_%=
z+vK)sPPZPSB~o1l%0O*!1HNPt!`TClVIsld{;rTf=&O)+<&aoT?3lG6%bu3Z6p|IV
zu98j&yRaPYGN`PCm;qW7n24a?jZu&*?_QrVAMPO`9RErLAi<$A^f;t0pN7N)65{`w
z0j_vV!hNcOEeA@ok>}gFk$3ss!XgCn5fdL4i4mws2Qe5-$iiWB;1WneRY)T-4qVXy
zNMjcoABV4Yl2Hy)vu&OvW(Y-<#d>0nE#h$CvpN!^@o8C^lFbN_yky)eB<`0tPn-aH
zDF1In&_Hb-sGUZmks?xqLrWaV!THA>NG98zx!qQm1QC?p;dfK4ZWF`#Kv8K%reL@L
zm;u&+whSSh0C5UH>Qg6Tl*%z+FtI(_fn$V#>A_1D5IV~_Vli8n5&_w0DLh;cX<@P+
z0=Gp4o+YxjP=J7Y|3cD_rZ_Z#gBJ&s2GL$4SYu9992+J^5tN*?osht#f|F<S+_0Sv
zR<j8ytRA#XNERZJKMz_Bh+s=k(sl&^DI(wimi)38JV=O=j~Z}I*qECLpJXdeA$45D
z8kt}mpOk+5$l=qJaBx+xEgrbl`?#gMH|t1CQ-XdX8t96|G};yi1{=VH3!xM{@L1|X
z>Z=%&f!XqD3;C2Dw8|7~|A#R?e$P)8F;S6_0u=GU1u*Ka7{-VcCZw?`_1R(q5=aTc
zoIqcSJRcetLUy$vlp+?<_(hCkvU)mf=w3n{TYQlsRg%DLgD{2%T9nD0<g|mqT~x53
zL<5Qr3y?|~`5{0-j9~9@Nn`V9b%;S-5ujlvv;Yp)BC7du36IGT%leC{^4>IlU+2Q&
zOsiX99H=G%Jaxb)Qb1R!P#X2imZl*#)y(4I1|%gU{fI8fLn0o}i-DtJ+?FY)hl-Je
zOwn2-K{!VTD#Hi~f-;ELDHi?tq}#$5ho1C5>OIeafgwG@oeCxr1U2hGUa109{RE;h
z85~&=6^)prz^%Y!ud&_zizH&t!$C!jI;agSCcs1(jyYfiGzXuSNFzZU2{%Cx3J@4I
z7D{9Zf{?{TYS0Xr9B*D}VJ0n;M>^<nkWdSNNr1{oCP<^Hh4APvz{@qj8mi`hCD0ez
z7s&?U2-jnK18QI#4eHDUFmY*K59EaqE@n&3=BJS8vPrfv*P|bRlIjXYq<}~uH;p8u
zWgs?D2tW}kSi+fx<6<q0<3jKY!Cr4KM9V?q7)wAurx<J`VLAaZ0rTJ|FfoL05hn;-
zcXY=B+W*PLaZs9xkgkw{7;uJKd6E=?6L3Ft<^bq4CO9z(fH}F-_?JyGjJfV!Vx$8;
zEkQ^?KyE7Tv%$#)xJMeqG$+LinKXWBAxS|@c{!Lz2!%YvccGCqOwCOL-rHyDuwxOw
zD>lI?rc8wKfyAUDj>KRmj(C*3=F4o)KMOo#HUmIZ2l^LcKrX<9=>U51pbSGOfkyHG
z+#oABKM+$1A$w4Plu($^D(U=(mUULMT?9_^pd$e2BFF^Y6Nt)HEVRsb@kq?H=QDNS
z8(&(^aVI%KM1qxkBo<S~A;km&hxkjQ5%Y9x5v%YzN-pCNlqzD9cZ~XO7f8#U#}x5W
zbX<Rr`=uF&#*j8Db>|21Pznw;a?;QU?tIXg!U_m}`6R*|NJPUq2&c*xF=?pE<TqpP
zY##*g0#GeBOk&IyN>NZ@SQA+!M8!fR;YlRYMx|PZLQO80QNbWEA%j4)a$i0v=9AQ*
zL=}S&#1@g56Pv`EIDCF-mfFgjqm<&}0&zZ#1CINz#D-E(onjio*-{-q>)|n+8hxoa
zhQ>leXdu2glTurP81j+cokaKa{V>EziSl8OvCt|o_L3M^d%If3JX?0OjtdN(TrHi}
zE-DvOK)^3Go{PmuMWVYI<{*fMI3znog20ui$~Z`t!}g%M;Us7kiGYpJz4C68-HD`-
zDhaAHaS$?0JSYWjI7tH03%r9N6a*pO{t=JDz&A)j>PLx8hP#N6Pzi_vT#0|8KTSyj
zjRA8ls8reu#0Umyz*v&Z#93L1i8+uLj3*#GQPxo;mdlkOf(HoTI!FfwRtA7g3D^;&
zD<VM?lH6=bV#I!5FU0#IRXc$Iv3$_)AHt|BCIvSl833lJE(Ni9RQI5WN4YvgVF!d#
z5|FzSl_rG6LWB2l0<=a9J#Y=FI3|K=5!`ixVSpfFkfckLnJmCu*dl(Z1VEsXBQrJ#
zHF8sFW1Xm|4jL0;f}AIvRe@Y0Q<6Yz7Na7z7yv_zqgs<jna0ia&y<K^?gHlH;CPe{
zQX-_DGD#ve8e!B##9jh=NQvZ=5&(>kXlx1C1_Vmbjn5iLtY)_sf;vn>gQq}9AiBXl
z4?gmukwUwhY?vzHI;|vh6ek4ZIfo$FP!K?qhFv{@EhQ=i7fL5#Y!L!0FoLNe2D}CK
zIy?s&N<?fI9xbP#ylZV1k|2-|aZfrXP(?u}_=5>haT)NqR4p(ZJpLzwh{C1;9zbA^
zeEq;cklb_$N>Cz6cUwNfY&lFEoRtp|iAIC?;*p7v)q>q5jilULLgN)733wNS8iA>3
z?}7_|1N%G>ArOg@wHEQfM>zyo1XGL@0UKkJ5=EwD5-N5_vQ0bKE)P~TJoKf8Yv>Q4
z>`$1p$s)wWl}L~aH(+{9*#|#AkvBs~!<JJCv`m1_=yvLOvg=~uW{`tu1X@I+ZUI4g
zYA~np?}B@L2*w=DmH=TMR9X>Juv}D@0!qk2384;?gEc|==0ymDyO5QD9!4<0g=m;R
z&N?u^wC<s=J3;A8Igf;dkT^^Xj04H1q!-XV#3=#-htu%Q;a>#;B~aZF`VCZw%w~7C
zORT1sNhlQ&Xv70QkuDk72xqVob0iRwkVih&T7;5y+z5I#<dA@e!T3p4t96RT2l%lN
zf|gYzWW%5>;Rz)$IRI;6T86{)o|+6Jfvi8nBqVO4Kh-f2+XL?ZLok&GyCqZt*aM-T
zk0~bdnG8?|2qFwc-Y8MGQp$lFB6|8+gEH>y_fY$=qe>9C0$6}k&4g&YWU$VI(vyk6
zcR&=5KvBw<MP#>-$0sqwog`%F5lBlY_8<w2t>Gobn#u=G1B#F|t^J%)u?2PuHAoQM
z@ssBYArMhoV-9dbS|fy0K14$u4mc#0^h0ZgB1v5W0Se}`J5QF=?_W$FPGPdeY%$a`
zd_^W(CXx_@JTi-M0o(IltE2;h3-kd38X5_C70~-K<eaD?Bo3_w1JEKKVt^wEamYb+
zUOq{JGXUCCN#3LKi*J2FG{bt8%V*D=Sc?VO3vFL@aQ&G=Rw2|#%Pc|!6=^C6s(z|H
zz;p^cn2Cr|AOR`{#E=r>Lap}6x*y?Ki0C<&Y%DUTUi=Mnf{uxF!SyK(DWZyjL$k6#
zh#Qe2;A}uVQhFc0&o3)m#7O|W-9;{tNy`KJb0+ergWI@3AjUZ0t$-aYhf-ibAi;82
zYNr^#U;U6;DPwtkvzvN0xUM+heC^>X89$*_el62Nb#t(BCNv~WNRd5+T|P%bB2gm7
zhhPEG&6dEKh*Y(0dK1N%w9<-xv#H7MUXAgi!bH^!rd8hM(Wxi5KbN#^Y}j-5K8DP5
zdcavJ^M@OX*A<%s6J|^dGXI5iIDk1*$i#8ME3gilkd~N;2*|^yiIoW0y`d$OIP?%q
zM6+iUapVZZfYl2rVM-_Q$+%^v1Ue76h%L5qX~|t5C`!Lc-uiQlN*(!rgN3?Nsb=$Z
z!^ug`c^%R253I`(s^o02N;*f7W@R1w#$|0VQ>&?RDH2@aTUHtEuWK3|`2LuYnj0%C
zY0vYf3oC<*6$q%FEY-v^Th|>4e}8eyW3@T{hPiXd);Ric+q}N2B<ISW7{TM3xTMEs
z%5JR4@Q<nV?RW01Gs*0Uw3t{4coV4klF|6>T*vYtz>w<Y)Pw~5iE1M@;&IsSwl1WY
z0t^6vB_u%*2!b(~M#4Nub}!PiMKEJ#A~B=@vzaLtd()7t7mY7T7Pue*<#cSZNU>5d
z_gT$de-JWyGX321K1ap;yDU%hu5H(?7iQ_Lvff@<8M#dM<2QT6@B1|@+Cr_iWCIUf
zJLfhz$s+1di^J#KnpY*twM$$)C6-6CZ}xcJ9}ORGj{fzak^PeHcNRzQTivAKNI`ct
zCW_VPJoQjcX9~EeYpwO3=c!|w)v&ty7{h$7W1VSc$HuGX+*2E@eeyasu>AHog`22*
zfnI#XmFqZu)PUo0i^&vVG^HGVF@l{1=+m%f<3KRC^+m_D#7M$`iHum33}tW}k$@C+
zrpXe>%LPfmrP(@)lBsMqv8c;~lm)>F2%UC<JPaMSJ(5tP8<R1~o=4x*c@*~Q5MB0u
zy}UxlhjK~3_DR9Vdim69<9)Ryb}h<oj-DUE^D_f15_3nj*djagj^D&}8w}}}_ixHt
zFXP#qx9zw67HY05n>ym`zwk=NUS8^16qmiHbI$#x@49QlKsl11ShO+sT9lP~SdV<^
zEz60d;7(hVae){0(Ns^Ir~92Z{+4PP6<Bp&V0LR>ZF016{@lq8+vT;B-j64wF4A^M
zbKI4}{KCUl4u~g)#W)k(js&|ggh;khB1Z7bJj7x#=0J4;*o<Ky7(56JmWDwtV(>BE
zIf)COKx{&^ljPacQYB(UvrL5II<pA@codtRmT4tSMtlg#E?b(CX%}~%lp>>seWc5F
zvg5|IW~yw(Z*0sAwchip?}kzN+2;gyt<RC}WSL)ExSogWw12v_&+W(t!vf3H)1M2(
zyk6$a=?$(`MhQ91!Pd#5-H%!p@VnxAOZ?t0u)EhiI#Y<lEJl31^FuRzPFw}OaSt~z
zJ6t|hbNk1RI8P_*g~^ZO#GDv&)w!Ab=aeMf+ug@9<PU;2EqNIg&}3j#$HqgUcSJiE
z^JzIbFs6VJ;2;Ep@)oM+P^(JV1H&OR1M=Ha1EmlyAcny#<=ipagQ-}MhMJQIhAL5h
z1&BxhIAKH+;>`UR+g_1W(ybFgctkYtk^TuahpE1U&~<g)@-8<CfeZ@plFsln{Rh@t
zPDzlzwfn1nnY6w->+PUl;L&ZKU(^zkE7utotoq1JjIiF)Md&m&M;lkFakqvVtFP19
z+_yvWGK9y3aZQN5>Mb4fG57MF-qh^iE;k3uJ#JYX0G+PT$<=;jrM9h9;-r*}c)cq$
zFS{LEiBgS@fhI8It52WMNy~pLK~y7(!9G3@rV?PWm#(B3)?X6Pz6`QT2MkCDARrS2
z3MUWRYz?JydohL)WdMyw4ouTwTuH_!iD1lff)y`f;0KX+v2Gfavbzf_<G(Rr7&z7(
zWz}1gIMYCVYO81Y>&Q~S4-+>dL4tK`m+9VvO|$&Y^%Uc@J+&!IYcIw!tQYQ1UEBZ8
zuW<qWUUy;A*&>8`d)l1h&c5+6U)(jSj06;jn4=D?>{qI_XM<(z-X?uktH#_CBV`)j
z@!K|{byocTQV&HvF@daYU)>ostKsIME-D~`AE+CRZx^Z+OJFpE0bFV+tjeiyi77xZ
z!zGe=ARY1o7Cato3BaNlJOT*_l$GJYoC<p=POxOd*&L4?*y`gYgC~eCJ2=Jcw~vwF
zOs2y89FPptcnw|T*&1rGHj{}%qj7T_?&NZhh($>{?KL+}9#4oV&HlnKLTww3E=&5W
z{^%=F?ufg&K_XG*ZV`X0BIH+83qB5Fev>wag|M<e=1UOjGi{MQ^i?(e;b2*7BFr)|
z`NH2(xU~xRL<nol+kEa>59IkeZ}WQVb9K!dH7_5!zruR^{bZPV(~``Sn?3bql3K@e
zsF^L8+G%torZB&_82le*7qA!wB}uRn_BA*-M$ky8k3n6-7y@M^(ALsP4@?pEBQaqM
z{>8u<zd4;IQotR+o3i%1#7-64U5{bH#7DwslX^PR?v<Lu;s14^^&wFNvII@Nzn)BZ
z^-cRkCS9T9F6#+2U1tA{6NsC?u`bw@U*jX_8u#udvHyzoDDLmUsIR?q<>m5C^`6EX
zV}EVDcc(uUywiH(0ns{h71x#Be!kF01yd5zU!BhmV%^bKt349T%5GG1JG2hbU*)^W
zT>VJj)8@u{ITLx_v3mfsnbF!O6FToS>CDxBU`06t%qu7)jukJawigKK@X#$FV8XmA
z84)mTORR%X0okO)hkgTC01(p4*#uL89V0-@GLOU@`pss1KVlI3Z8hqW5&ixKGcQ=_
z!=5N+Uzk)UWC$>X%L?6<F8<nCx{!qOG-Y{|>KA^Dq&F$oq`nkd+g@L*T`y!AwWx5l
zBV%^IiniQ0?o;Vmv8r2b&hxnCG1kT__DjAw7&_iMlKRks_Z3_-V%avA=LI73g7IgM
zwDxatLPE<uQv|zv&TrMZ|7!V?n@!(lWpdN)F2r%Vh_H;D$Rj1$@97C`313}br6%P>
zn${svQnR`}3|(rj%hkg^!a5X)v+>S{ok40luOpIL%n(f2sFE)GAw*K02!aO;lKd2C
zXA(dJcN!sFXn7l7Bw<rq8Gs~0WP&Bh7JG=vLLD#ruEIpEG7;@|Yq1VE4EPya1pf@`
z=z$CG9xkT@bZksAWBJ*3RU>Qn<gQI*mtFIoaphf}wpd>8HH!%!f}5=u&MCG&SAiCc
zd6n}rPC+FjC%SBn^|6fHz%vo6eiNx-H%w5#vRzBZv{u#0aQaf`eN(sxnnWI|x@)Pn
z?RMF#J)BtUn~&*l_pn!PHMvtVCgEdiT;y@L+>KZ7_0&{l8|*ZiXWfQ)s&46Vk>_^z
zX=kZr1U9ykly<1}PR-fCD?3(sR2EC%l7Ui~NJ4QG&x2p;LO`7)NJOQ2i;GYSbdnSR
z;wyMo0*pAiwEP5u0b!TiqC{r$P^^_u&3;j_W&Y-vNjWFdFqFoGZCBXq(ZE(xv12zS
z$f!VAnso1DT^x;!wzkcKNx;I<D{fdFt(!}z>sQ9iQ1{k2F+2HHtg-rhD{Kij*=~QA
zZDmcQRe}9qiWdb+rB8H5&zjd9Wt~p?E!!*ZR}7VjeDh$&>@JTS*vsl1Efaa~fpcZo
z8tY8dvdr30&9&5X3s?JOa2LeR<kmGa_oSB^J-PpFIU+V@>pZRxG|G4JeA0DSLa7Gb
z=j0J{N$jqB1UZ}_eFy-EK=Xsm7)%$SKT%gWiTFy}Q5o)%%zzC%J`6a5*;BP*aM(*P
zLP9<&0<0h$QTt&+Ru;clz_k~!Ik9NF3kGX4#K4i#-EXyV{#3P=>XtorH=F9Bj8TAo
zK6|D0cCPj<%e@(aiRx2tMVqJx;N{cDG}8BXjXEaioPss=thd}JUwR<?Y%4cCK&NLx
zeV|}VaM#={_QsPM-G|!`>*V!zw|@Mi(HHz%^&pm`#&ta4+%=j+0S0*-=!o3FlfkWO
z+|71RMmI+QLUpEzh^u|gGQVayk8aytOHfK7og|CbdpWVo#0msMLM6~()Wu<pft?5v
zh6+B?AQzDkQ8APP*kA<a0^I<@D<Tjg&q?5yhj}G$5fNgTu}ii)F)<%61PifN6k`Q@
zI@_CIpC8sOq=0(t0}e9YBohil_vED(=A@3f`6|_p5Oh|1J~`FAoMl+B^qS7JNSI1X
z`meBV9n$ttOK3{<dp92AX8Fq$m%rk*#U-5@t$oKbf6A)!*-u{eIaZ8+TbEyn(3H!Y
z!GChMmP!!AT;)!SVg6m2D*#Hq=e?%ZXu-#>b;dj-|0_)GGkapqGyAoVQEDLC?#M%Z
z$JBP+2NXwQB()Y5!Igx%O^_TWBB@oH4i##F2(TO0l{_?hj|8?B7%{j`5Mj&df+2}P
zV@eS0M7hu$E8qzOiGCx%31}!ML3XLVtzBzW)vzlFoFEwSikJ*m7AcNoWnW)!<>X%7
z_o{7!^`4LI>J}&SFW-N)!D9SeLZxw3gvCTR2J7U_E<cH3UXa}r$tSfrZ>{+6L17Mk
z`>UQPP>UF?ZFH}$6-I{?22$IMP#bsokDUX=@y)%R#<aR^ODRvy5fPt+=>x3GVg1iU
zr<kzV0t=~|pe+NDMN~s0=mBD66-YuUR9M86<s_4OFwj8e%@kHZkdFgvPaGw05pQS4
z&iG*-1IiOPWPnXp__O49c;0i0u*@kIj7~j<y)joyiG>~gYq78c1lwv9l}XXYfde&$
zyCXGtONq9-<oKIPt_Mjy^Mbu1k^^3ds0|0IdML=_gT^!-cx$Ev%%_ZjJ0OY2rr(z^
zafIVSrXK7?foz%*jT$b2z&wf+m<Z1>!6?t1csi3x;lY}p#4uaP!K1E+2?x1SPFh+h
zq8`_Qol=6?B&9YbUrNf&$D1Xd)MGCok|xEKf&?A|-~g{ZHG`E<)=*kVer&;mW6`Lf
zixGNM3&FDtX@$@vlo*KwaYUpo5)vYIa+%0OcICTJt8PkRfh~Ydb=bt8RDyH@4j|YC
zrhw#7NDuFBLS%ynh}16S2*U9HS63>9Ab9dm`2@AK3lWXYfh7h=0o<hRITfC8LQ{)K
zD$DjDxFxmM!iC8o$6^984}&ftcK4!0Fqi^cAHstgd7+&c@UJA<xC@D5Pj2e`i&~A+
zV0j{aAK{;ehm<^8Ce#Rl7Pj@IZHUHFaRw?KCJA_@lXL<^U8_t!)?Cd$QJ+JDKE~mQ
zK_Eb;lnhfj@B^@b+GPNFI77NT;8lncNhk)P9Y_2Vmw6l>xHt!gD~T2pv&T4cqRkR9
z*vHTXB-zO*1s+5QQ3q30Jfo{>B~>M*3bj{>B;Z2?AR#3VfrHaX;3jAeL^UfU0E1JF
z4u=#dDiH<jl8H)#2a7^4wt^GMPvL{WYx`j{k@9S+HE2i!ijdU9HUMvghWAUfP9Z`J
zy5n${kep)D#@Z85Pf-!IJz50MQU!zr2TEbjR;*iWoiS(KF==`5BESSX1OxV&hE;^F
z3#k+d2+Donrj)2og#!n<Fz_~M8vO6mm`$5Z5Ih5LfVBx?ibb0h%cRg!Pjg_IUIZ3{
zvLpSKssSb!l>AI4->Nx9mIrJxU?$Z8ela*;%mhc#NP{LLF*bp%>Yr%vw350(wONsr
z0Wp$c1_2^KFtd~HoV!S<9X=pH0y6!C6r^ESmZ;oMfE`^S30D$0Qs98a5Bv@j6E0!n
zI0c8I1l0+o*7J0?@Rbso0tftu;W|`%iz)31_!SXF0taecC1A7v72OfB<d5DYl?wap
z5aOx!|M6r9nhpd)<^K>204iZCmii{<7Z*~cBw)R_cqAr5KctbAjX(sb$Oqh*WuO5?
z_Ap(S03wMakEU=0Z5Jd9ZxPTSM7VP#5=E$updQ}g%#-h;suoL&5z=31c6%WKP4<IG
zU~oVk#svzGVWRkdQx|9HY}&N88I_g?0W54G_?(j%UIKv*2O@Adp!G#%JRv{|dOAQa
zfbaoUFfcxwy@Pm94NnMK9*BT;mb#l#qO_MFC(1IC>|oPkj~-~4xQfa|<&hE_6aCN&
zu0aT(1rd}A7N~5*5-KuNO8-RIt!4Z6IiU?ZNotlRZD>#ko=bDoxZ3XU+>PorB&5J|
zq0L}98*G86Yzbl|F@bZiAa0C=Qh@;Islv2O*czs)q=;%^P!Rlqv)FpqoTy_|B2N8K
zB-Lz~rLLeXL=Nx<;Kx$G|A{z6^seTXrOP?7yFfF_8VLyn(GCK<h7DUjl+)q~AcjG^
z=Z^RcStk%fJ&G8n?py#|inp|rQNAaE;$l(aNCq~zgRC&tx?osaVkG(BPBge?z}B=>
zF6Dy%k-&F%^qqaQ8~uNTy$e_rSGGU8tEiTyG0<vb0+CKNN^FQG0-69q{+mW35`zu+
z9-~Gh_(~>%sKA(SgS4ViLB%Ktq`+3tn5a=_@;>Mod<-HQ;_I0F86`*z;~0bHn3?l^
z_qVFiJkGiIE~Ot;yLPSjUVH7et94|5J*G<VxeAuh%9!jlPLgHKV1WJWOz`yN5UB0|
zdHO;Yi6EIYpOr9Hu3-L3W0~ZDvzLaMt?%Caqpk{s)c*!VFxm_9zL3yE+xbmfC$oL~
z=If(6nRQnW#=S65CYaHPRfCXL*&={~3O-W$HCaas(o@N20wjssczh>t1JKP7??|ix
z*PBu9lFT3(ZazY)YfP~%xOsNhr=$q~Z`d-9h7ctb0u>S@8~Xk`fCBQQ``a&WH}9H-
zMHwLk8o+S)VG?!%f1rR2TIn{Rrw4a4lG6QzvP%!9<=)#d$ia_8-F3~xAjve6gpJ&S
z_b2~9dO*|?vRKXs!zj+v^drB1krIHd^zEB7M_p6HUPRymswn@2;n;1M`amEGbL;x-
zA)k%$Sb#vc@tGxHrZPkVKhV3Phz1i+FFUZ88O#PKugt|j!F)}cS%`sWcYSH?#w{Mw
z`&;fkI4lSW{1^WVcX5I#v$k!>5r$by_}i@<OG6|)#x&0xVJew89fALcLlpE8fF*tQ
zL#2ioj=l+j^AZ<(6^R8XM&Klz$RcRW?hxEz7S??~$4A_1L`xNah$`sy*PwX5K!k)P
z+zMg#J0S8%5lTeB%A|7g%ph?RP*KD6=!Tk8*_RGLUlTYFJwfb>e0hAP3wV~v#pVDd
zym69jQ4}U5(*dSwmcugQOb2K~tj5fD{6oFpg$pY<fSw2=1g(PJegkG=44RWNFoVi0
zHoX59=qkhgyo>9_@L~DozUfAwgd-YrsbVi@x=CTZQ#CM?rVGcxbfC}&^J^h0>dacX
zc`pj-a3hN}ne75w$YVQ!7AS!h;=`DRG3I9o;iVtY<s5%xPrnDCYAAs-!Euew(Tx1}
zKv&(b_`h+);eHEP3Rw=yGl)d$*ueVDCnKY8!VF>*yW+I#K?moEA4{~5qB3OO6S$6s
z&NV0lU|_MwxXtF`WclYq-}Rh$mktv$=!9N>1E5NAhU<#wVg3Ik2go3in7RC{YMauQ
zkB&f^f*b^6Dl#w}O2mO3rZWn(z$7JsSV?wz##y>oe5ZhIV)fm{FzLVnbb&|ZCon$U
z^F?K4vS#aVuy+6avPXBp!3SUw4?fX5XL6&d3_K^o2{ipxe{uJIe85`GJ}WInWs1gl
zouGCO0j)&@Q_i>sW+PeOhHoYzmvIl@6$8Ry^G64Ug=6R|*A>MhIt-ysGe%%;;!H8i
z$mjRpV|1{Bwx9ptVXx~Kj16ip9{L+-J-~`Yl2-VD_h8R&LBIF&ujih4)?&cpIT!^)
z2yp-^y(HO^;FDBYaAcrVRC?GFpmp4AnN*mNl>P-~p)&KTI!si9Jrr#KROSj$bxgtw
zJ^n>eY|Q(;3A;YcKY;V#!~Y6U2VYFlTugp=`T^b#4tUUg6X>!>+nlPM4I?4^dWJNH
z)e*A8$wLzjSUoctCvhP|=qX^Gut_Wpxl)AigfRg|eG_L?#v<^|jFke#`?Cr49%PwH
zcKW{nm|G{8HIR+MSUmpd_w3bh%In6B8ByG;?a$EKA~sM%?1iH`kUTMv6`^bk>`h>-
z*t;yjBn>QdYGDr8n_`~IhIzEZ!!&3O;^S5@8r3cwWGfuir#54P6YBekM;koM`>i7A
zcf+!rVmIm&0~&+@zvr%w{)nt;KhDOqc$mTDW1%>JS{fMR9H+G;mhMz-+>pRg`+-GD
z)OsC;hm*}_d^IsFM*a%w1l$C|1%d#k4zY90AX#XFTyyvNnjda9(s47r{0iK?n41z{
zXe7V~02F=$9=;r{&N?G3@6K^?oRo#jZ4XB+jH0Ug5T~-~rtsoACNOeX)?wnHX?oo;
zFqgQSsheAImGNP>)Caf<g-RCMK(Yi1bftf0eD_c5K57N$r_%c`kghvwy+2)h=!```
zk-;ng#RX;hC!z)=TGK8pzQlaztmU|e(_|_tOfdH^HiuGfaTor<aiV}w#cl+Y2bjuD
zb!;L-87hVe*??#$fXj{gBoi}zrl3EJW5Ae`5m+$sK?rG<oaQ2}@Shns|LDOPZr$|@
ziMR%(qqrNaOs3$UBTUg*>vam_02P1$bl$4?$~`*bWzF3Ji#vC7Sh&#BpcuJJFo#8L
zC2}^!z{JEvKnemNfZER#ZXSc$o*~d}!MV|r(JJ;bp_xgSrW=LF?kvnz0(Ahl;RJ<K
zF&G{FHH2rqih;3>I|cPe(H6ZUc_SU~%GlJ^<u}&J;>xV;$lRNyzpv2Fudubk2`pq8
zBXeWHbWH_k?4BVE{p;$B*S-}O#-A7QW9yD>LcJe3060}U!ZI^sgqq4siVugG%1m`w
zFe$QVFcX8ySQ?+29{0ETAdkAE5<@aUWFX*Qd^ya5qyS+EL8`scquSWN!c6EI0Po-?
zLNGIZ(ra4By%nR&?w{+~TjDh;KYEAC3{xk%!?qUG4?ftuCh=Ueb=@0b-+WjzxAxY7
z6T8}w=CwFs_WB%*ln6tHM?lQ$+1JP4>^Hk#S3(j*XqXymu)r2E_)*(UoE=umS-1(h
zW9g^kh+qRmBM%0(WJz*SiI7pN)Q?l`G!WHCXq5sE*)nk*23iYvDEX_xdiVLae^`uL
z%ghvFnMn6|2u_|@*xr@>Y*j+rsUT1kyeKkg+@OtX5|8za@7yd`q<#?}Sa|i`+|icn
zFHi8EdU9dZ<KycS4=}lqhagn|o25ZRcSm3^;7s-0GwayF?gP_}j5Q$gL$-u7UOP_(
zjaYF|1ThmB<3HmQl_7SgFL8Nbu~3*~cNs8_N)yL;M1|uJ3-TbKM7ET{F3DZ$gJ;1P
z2dD!RD@pU*nA!p~bN#A_?m;izkgLZZtow|61>{COUaCYjUAHx~EK<E!ZQkJ1+&RBb
z;<WJrt-<5d_I}j#*TSxwxX{F*5)ch_$Ehpa6#oC&+|uWoKiK1IeQUUZX`0HoK|wTN
zM<_tz#L(^#Mh$a61M~-@zg<-u*hG^^r>hzJ4tp7(bx<+bIHFooI&jS*@J0q|)(Tkz
zNLq?`Kqo1|q>b?m#ArY=#iE9}esDx{*CX{I!RvLJf)~XNLfz7kl>=<&*Q~&z#UQ`>
zEl02Jj+z>Ap>FKn6G7h}&y6m>Nx`dLkt$7e;YNs~0J<Y%QCU+Dr5I7hW``8v4w9x~
zC<Ez6djqCUz90|9!lrO=pYeeLn{FE5zla9aE{g*w<E!O3MGO-lZF&*uAQ(r-Xu#!8
zNwJge1&`|0z5ejDTRBbx5nQB;-^~Ukm{x4O+jTbhG>e#g=xE2cLF+TGAKm;HNv>#F
zv+7_+`||jTm(**Cr>-_Hjoz_wY|yO&K|~dy9}J`tvI?+i1h)iU&*zptE3-cjqwO=8
zI>2K2RI{VZY}!TDrJEZJs*9C7G;%12Cpge#>@bL;NR+3b2Niz=b;Tr9WTkXufwDAE
zn3-gBkblr^NmdLE>QkFK<0_LViD;b=F(C;T!eLH?$Q;&<Lw9N$vpx`g?{?e`es@#H
z4?(5gQ7^H=_Vx^CRJGrls9MvKc<af$cxy>sJhJ2#miSe&%e!({VaKij#6A)qVh1yw
z8ll6YI<K{S>SFopLYc{FvX{u|iW?CRw-`_-3-mD2vrY6#@=2m#nomW>5t!hF1T#&S
zxru+!2o?z^{3agP&scw#2xS5Tq=#{635Gb#52b~b(K7ka3tI|XkN7W<{3kMNYFerb
zeXF$zrm0(x-*hTQ6Q9npKv-ry+TP<gG~+8~7y>p5a8qS+)miUD=O%c2O>4iM8#l<b
zJlaw}rxmM5b_ESsZ!&3c@rQ@@+md~6#``RQg$P-Li<P;=X%3EApdACLK@plAC7C2}
z9IH*Hw|>b^He6*xG!7#hjy;W_7ln@|BZ_fAiATfH+6@JDn8~yXjx`Fn{c^MXVesJA
zn3{1f&pm1TM)MxrR^UA%Z}gxEL-TLv$5baY*C(%D*ycYZo)_;Q!|tNDjwQB+p1Gr}
z0t{6>(IV{j)schzZtNVMU8-eAKknx}cawDK<ig>KJq=cdizN6Z!X*0rz6)jcc^$<I
z3EW7iO_$A(0u`~7z0mBGD92ax7`PXG;}?zXtD5H1xML9t+$Q`7TH%kBn)1n>L%)zF
zmnjy(2DS9rS&<rWA%05o=|v4vWnA^f`u@QwIWb@LTXe1^{;Qbk(j~>0lRNYFF5Q^5
z?xx?XBj3iTy1I78*(DnfkDU$AAoYerOumrZ_D1}=8-vbhO^dv@j9Rv_KB{nEy2p({
z#2QdwZ1B4FrdotUZ<7)_%NbDB&5F*EKrE6p5z@l=Z^tUmUX|_joMA{x=*Fi@3gn>9
zl!HhYdU2XKQ&fYE$)TJ}Um|w%m34Bx)?WJi%Y$RwXI~!t)zbC5KNf*=4|dmk)pp;`
zbU#&lY4_Yy2gL3b|Ck(Y`Dw<^Bl$5^kF4nzoWR-1h4z>!+dCJLKgUKd4+de|dzOr@
zdL%Pr+W6I4Vtm2I<B>ssOV-^A8X$9{;mxb~!-M-(_W3oi$?%J%K(P>tgIi`UO(fB1
zR0$2Q<c1MJ#&9kvdICVjIoKvZT7a#U0w1`fuVo@&K~)@x`y-8g=)>~u%(-WlN1n*o
zu)K8aA=?LTd*;Ry2eRX>XT5`NWyV*RzTgNt8(KQ4Cibh+p<6Z-Y&_Bx_(;8dO}bUs
z#;@MIYE;3-KJDNh>!pu{b=+AJ8`!dX?st}x3sJ}D9kr*cWxoBvOa}w0gYfZ-yw;?#
zXBu`15P>YCK?OA)pn+|je5}FD$^{uLa%o|7Mupb|GfY43NcJ>@T9&y`rvirrN(6>G
z+9N1omWiYpmPx+M==vw21xiC5*0T}*0q4DUR<yG0_?p7zV8g4UD_TyZSCnRL4&HgW
z@w=HjW4dp=ymM<oT-BMOQEf}3gHIbXBHHR>J9F1}{+0R#mp$N)3*0b!choE6SMMzB
z>df6gBAQTg(@Cr`07rxeUghvH5y{fcSsbGiF`L8KfTspkmsG<ft_~ES4HIL4iYXCP
z*6Bm-7~Gvbol|2h36yEOa3PghAmUaq0fA226?9tMz`S3cu=CSl_{>D_{Ta(Mgmv_U
zTgG5tni4RfAL<;o`W45N-%LKeFie^daD4TuP2a>=si2`2!MBYrW^#?6cg7y+<W0Y!
zvsc6gpH;jMFh~Tx1Lapqc!#2FeDST=O)P9)kI9HqO{YGb6L(2AR8y*h3XC0CP-NKG
zf^Ly~13N%GOijkF28&?fW#~<{sgyMmbppZ;lN^m$+bhXy;HF%7%;N{CD_rh@HF8-i
z^zkUBl*zscs9Z?#H^iUVQh2q;VSV?M)HRSuFyzI{ma%2&;9|0r;6H|Gx6^Bq^P!Bn
zF;!>BIy{?$oZQ4QT?ak}zw~P@Zf?qdTSzLi*P;txDCo5p1ikCd6g1(BjY~-)DIKFV
zCMTHRVQR$;*g1k=-l9S=^mv{@kO-swRhE0<&9{HVjfnDIcy-jE(kwaXWR7>b=?ClC
zA@SA*CFS(eczMvyv_|v!oao7iHZ_GV->9LzJjLsyRPQ-s2b^o|VKG&wmqnNNERCyr
zWDKj{f?**m-s@m_P;=JY!iJvSA&OQY^!bG)Ci8mBr$rX)u3d<purC~$ID}fN{Xqn#
z^OkM`0OAG}c?|ljvJcNUGSP1w_BcY-NTmpzW1z4)BcKep!d%RPkG!+<)8jaY%X=a4
z?ASw>w+H!UZC<tOjd<(zb^nT+*1&Q`)SMYQcv@3@#rLnydDeV(`ua^*k5rBlm`bpY
zSKn)IS`=5~cQ8d7e<CML>sS`nb$i~kWpd5<%nt^g>64h&clDP?yG|}Zn4)trjMdK8
zXJ737?7ugCD&(Knr61)g_ER(zC2(C0NGpA?X+@8pCPKxOemGJKDh!w?8o3XoW{>9@
z{H#lb=L88wh%Z5iMC=LY5E&v$%MfI&jLiGBtge4X5QTzX2-LWW4QtHH_Wf{WU*<@7
zfyQMqJCY>-s->eUUOISk&AKlGA4!Uwe6l&?-0=%64*8kHh^NMNpI+7{<J0_Dv&9OR
zVQagGYTqjkb5BOt)!T_5t-sJWvoR-P$LU^4VqY{ikdyAP&JUODxAB4TGW<Fs;r<q1
zzIL_>Y)Bp$*pX(1j3?1|J5n^VvslG(lWo!~T-cIV9o8OTuBr-#P+dPTC{A~&W~c>X
z%rea!ih}3rzr3?^-$Q}w+?LAE&l^>hv@EJRC2K>U9B1T|oJ|d4P_ud^^SQD82Yo-|
z)lrr&{;>1Z(K#!gW&Yl)&cUz7O&OeUsby*Wpwi6wmvUk%rS_)R<AVFH{$Trt`lwUw
z=M!Vv?j#;8yObDs_zqla7p;;Iq+2`!_FMGh51`k&^W5yi7?bTqNFJpROv8dd%7l<`
z9^yg?g3BeOiOxb3oIfSl5gl^_s1s)F+o{(aUZ2lLk6<72F>Y9}5ucJ2-{hh_*pF^h
zaZWOYOU%b9%=NF`==$CxUg8Xey)3L|&o$VocS=3WEzR;u7@ynP@tWH2+M$}4R_D00
zM+bgCF>}-^;GY*~eQG@Q*4**wzFYI7ts9HO$}U?Ix;pcts!Nk1AD2q5aFd@m&mLXz
zh&cXQRQ329IhU9wa1D&klil$;*4~8hx?r8tV@sH_3iw=|sQ!h9w-FGazQ$}e$P!E#
z2@S;SlEZDI@Y#Bj``y<tWv3w@`}dK)GHYumt2AU9P01$gVAZYGHlODyY|YAtv{JBP
zP;nVRF(_dZhV0EBVnShW?D6;Ch^<K-d#E9*(toIYF{6!|c$;n(M3;AE#aC@u0z*wK
zba!B4E0wo#7H2(L)sr7<IlVBt>SzP<3t%_^W=G)fyDI38#bvE4=gbaN1Z+&T%@gEl
zG~kH&wPBOw>nPtdxvatdHWi6sTgO{tw>N}hP{BH<#71|P5Pclp3;GK?orJKIuC;MT
zqvt-*;$j;M`@eZV2Q_oFRq*Eln!*AkuF{)D&?`2SW}He{VVmG&Yl?y=u=={HmSR0{
zhx>4Qf)o_<^iEH_j{UoPQS$+db=@>5Rbkp}o5~Pm3A{kPB!khF6h2V65IY(r2vA5E
z42<%BO7m>SG>c_xJ7%L9yXRw)MB7y`f&hslszK#OU#^P=C$L}91<n`~Pb$CA0QvtG
zG^i(yv%O?EEtuJxxPCx?p~5kslmM=Ij{5t;lERI?b6XXK88)T3TsTjYPg**&s7{T`
zaCZja1P8(l#z|AxCaGv5AG3sNL%8v&wuU44t`|as-T*phhnm6XKx61`KnZKYvL*PZ
z>YsJSGl=Oo;MTuU4o+8gWc<LSUxSV_PFuhy9QS_;I74x#%{M6qY^A~>0}DV}UenpZ
zeW5?td2ntZ>Y=0r+x(=d{W$8_P-<(@ENZ{m#`qKxHu9YzG`~d1iExqh4kLgLBqOsB
zF#9(Tc{u!gKI_d4s;4<!COjX1`-%l}jw5mV&oq7dVzSh=tDGwRRHO!-OnunY3^WD4
zI#Cm<8YpZB)9x^vG$Fya!z{i>Y7WbbG_gZsLo%5i!gDD~Q7A_MGE@e44s4fsuvI-8
z;A!P~C124C5{Cjj`10#_=qf?n{<BS=8d!cQ*YxkmGkD(N4<CxIkh63}WBdTaUV7+<
z0l|Uf{{}Xbar7_P5&aU<A_0nItS=^TI)wBy`}6X)|J272cpv@XZ_q{a2cRP)9(?=d
zoyyw?{GVDCi_mG>O6E*HjLe}%w}TSGi;(RBarkuH6N`;5U{2P=lVN12BzCFl=td-a
zh$M5Eff<k#R}7jR*F*75A!K#b1*yT2|2ypgzL?Czr<kzd<9x^!W%%zf{zX^Ns|S$k
z75b+;A4csrqh10L7c3g-ZM(2wP=?51U3Q~NjZ|7WC}u)qcw!FZ=0*WWLb77=wae!%
zG||z}NTy;wvr$k*kQ|`EI||Y%Aw;pN4qtF#`67M1@|ap2_Rw`^Tuu$$zr_E=j8omR
zbpq`eEw*(Rd)Jw<KYooSQ9%NNVlEeBGqY;%CBnvWFAtD8d-!hrkI}I|gc%ImS_KVH
z$l#!bF`YZz!%{R3n}>QCWJFsa%YyVh<3_517pm>WnVXBa`P~6+do4`$#U5~>_iYx?
zk7S^a3sj+Cl?C^etqDfwf^U|2240;I$#3l3!tUG4qJ!6WmgUTR(%rF)t<KqV@;r;F
zNzRQk-=5!RZd%8hoiT2?Ds?O>v}d=nxEkZqc;-K<qX+&x+37AFgQw*nUn5z{@8qMR
zl5i)WYFFGI5c2PTe}CL9zgL<h=&K3T0N{IsRwNQ;gZCC}4=a-7BA%q_*H1*T6r}iL
zK)jO|Pakl#4r6dyIy+44Qt<QL3c`%q1de;K1XB?Tn{gcSgHNh_uDa6W6lEM^i5uo`
zi_M?qbG3N^;({~TI|fu-k^>)ka_$1Zp*azk>b)wC<kXao+nPHha>tY#L(elWo4aT3
zvj`;(QP!3<13ohKJ)(8wMh9=qTD7OWJ8Q$F?dL3Bh5lKG&NX&toI09cf8^S7pOamQ
zr&{TD0P?!MWVg3`n!<f=+TYgn<FHvuTN&t^f)C6Y0z#mh;VjgL`HbU2t*cF*R46Oc
zoi%ScxSJgs1FcS*C<`B4Ql#kyMAw@8!@_9w8OkDIn~dZ;{YdqdUy~iPE4%(kXZ3`k
za|0XhWykE-Dw%EL$76$jHb+*jZz>v;n)%m4<D;i@_oW?N^3FKs6>$|*S3DAGh^jf8
z8MUObCpY54Uu_p`-yF2G<;83-b<chFZT7e6bEZ7jI(KBv=~rJK*SAT_wsqWc0<#C8
zlWl}S_Ai-8`2{+TunMo-0CJHq0PfNPgOE^(VgU;*vP-^%1PNz_MwH<)Q+%UI&vjvA
zxzT`Pc94oql6cC4r4Skd=As)JV%DbZ&v@PAn{{7dKD{Wi+JDqHVeKso!y0>D9XYu~
z4tyyyYJo97qO!2@_M#YTYM)iM105@-zRnGimHsabUQm83p9L=3m-*fg&Gm-@7Domr
zFORg&P2T@x)VSsg`66m^$?n8)jc+yVS-VVWT5O2i_YwyN4?uUN^3#34uhLBU>!#9S
zp9I+n|A+crFhxets6f%{1WF}|2k>$5jG7B&U9yJ7P5H~KaH@);)M5lf97Bu8C9IbG
z6Ru3)oRPCbs1suO^Mba^9{;Xejy)pCZ>%G@uy5w(8vm^Rm8{QzSGHeD9JO%$i4Usw
zzOZFeSM&TCYY*JckJ`am%kucZM_$M|*yZ-FJXm?;L|J~sg+pPN?ktP6zQk_Mi8Z$t
zl=S`ToA}3CM8_v7+qcagpV{d5;((8U#RJgbI6;t3tDOIKiDrg3s^!daju}hrDg?ze
zO#RYL7LbVI4h{owHBbylxj;vekvWr?pU@=p9cE}yeYnXfg~%)gV>#AS;bc{FfQfQs
zFn!G|N}R(hk5pHdskkXaUrf@s6?A#UwC*y*1v76o@x!Re+t+xVYh4p{W_k39oWkxq
zxsjEPtYuj|a!LczQh2sDdaQX_<dhX>r>72`J7Sb&chtC(Hw%XJ{p!-e%FnDDBfpP-
zGW62KlM4+o*3#LyM=$6?@;Iz24b`qas$r^tE*9+XH__<Qu>J!bw!pI}n8OK2g9^_~
zdPySmff}<itZE6rGFA-18iwa1;uD*&xMIhtUHIHzQ30r$77XgH*ciK;`CeA08){9d
zxL*TR8^ST($&V|)Q^2s)n-_jRUc-|H%FTBQLRXZI-MM}L*ezOEdE1isnqogoT=j&0
zSFaU>ZOXNN8Tj?WrFx<XL!t)^v|<6kgA8ui&136IP=zQ@H(MNKwOFfF45}R{j0poR
z7`AX<L2iaXg>Jeb@aaC>YoDWEvW@SNO+8L04}l%Jup&&`P2~Qn0VIiXy&?uf55iNq
z@~7(jX7u$Nvfv%<L>Riqy>KHijNV)&Dehr3Yrm7PdXMVLIF>)X{Ejo-llZ0<4+$6@
zdt;b=SpG0PFa-*OjI1VrCB&fv<RZetk82c%D6AHDG9Gpo`EqGCqCp`yB50iip~&Vk
zD4XanEEJt!1AvYsW;$a)b%Ik^j=VeVt9^tU2>1n%jp2I4(u=+43~1^Fuu(BEfp5?e
zsDmJ$%llheK1xx5@+6kv*bHmHE@ucIE`eIgav=@tG98n~4TMrL4KcL4zdmh&Molkb
zz7yp!Dzogu+K+>!pa%n@KXicDi!{%P%$@^d-{t)olXa)a(eFd}1!+B&dw85_!t<GS
z_;`?Z;U+LS8pGvxo*Vt`Fm|x(gH~v!<N$Uu8NRL>@XJiG^qUMdF(KpQV`9bzX?9X*
zsN0~yI5?i<<-ddtcRI$Pr#tSR^umeYb2xaZU;votD#T7JC%VGDRK<oZ&zBEP`>KJH
z5&VYV$S8Vvi($}%tNuy`sDsl{W)bv>qH)5`4yOL;>Oad5NY=kW!#UPAZ3z^5;yCUH
zK6}cD0){}#o%pT+iS$rV4@JZ}M7mMvDu;<^N&@KwHo4H{R4D*0gY(9mms##+ewLM5
zizB%D7eJm1hNOSzo_qG5^Qg=(4GA&3s&f)-j{SC+6Uw2P9EfQ>{b5ZE_jKa<D%kCb
z<kF@EA+4}W#g`2Wvj}L+M6HZjrW(p+i==RL%(0v)<V`79vJeg6seUgb&=c?~HY4Z$
zlNVpeT{wo0bN!@eR!jR-baRAC7Gc)({|#=Pi^SpY%MWs>Mj`cu+t5>o07<YVJn~SP
z4vz=@oX(sHY~71oc~`c$XiEZGS&X>L?FK9y7^h>&896|;fwUN2lRDDvg>0A9g2hk!
z%d{`|&{>>%2#`M3j>wIGHe?KK$<Bx8{^tkM487NPM#o+cBcuhal&PdD(0dhk=FEpX
zA$KUVuKlQ#qJF4ghnbPfVVK=QbHO<Y*J;d`!#?$Iq)~)*5u0cuSAGUsq2d%1^g@6e
zxFH3exHjR^oljd%ovi0Mv#@zNpB2ZIo1Zz{lQV1F1mAze)s%+Xx`bPaM-Csct_kJ_
z2%{GQ^gxD=ow6-0AHKw(E_H$qBj0LlfUd}5Aa`R48^}mJjN_Dx+Utr3tl6DO7PQbG
z%m|vmeFG@L#0~2S30NiozxAmIE)>OzhY+m<y)faX*9Tx%yjKN&dh`$1BfovVa^CdH
z1Xq6b>2lec>YWqT>NRd{!Nt7iv%90~=e1WqdU`>0)tSt>wa0H|RCL{25?6in<h^{P
zt^8+GZ+iV6-a#SkR@Zm=o&OVic9deu@861r39do&O)3;7UDe2!(j{n&Yq$wlkOq)C
zhLk`s7B8TwyAFXT5>Z}dFJcR9LlB5v&=}X@80`}$mHKA`*z_wI(HCUq@*;vxE`2`l
z?{99-oL*V<@9Sxo?aVsfw`a?Zplz4eo%x}#An40)_dAAW2W<;l{CxgjyXw!(s;jyg
zT)OQE=b+|AJ;TmeOA-q&uM8M1sK?&aK7GO;P>0?8`}??7`HNwP@x>Qa=O`cnV8o!)
zx>G|LQ;K<*A%rR6M@<b9FS747v31jphb2|8yLN_XizyI0L%|HkwIGD{6cE5n$YzG`
zI8q%{2A)u>-+<SqGkK5uZ_UbEI@I;#fj@oKU312OB8JNUm^8xLGD`a{@X@t*w{&j4
zQrQwwyK?A}`?G%z`*GXNklL>Mk^lYbnpORJCR8r&Gho-J*Y{i=sdo$)MquLlo14S|
z3X5Ci6LnjYu`k6l++fd*?w)04luLwS35S<<F>@^smp7N-!E8<rPe=O7;<@|~XB5CZ
zw0ek-02u-r<Nh-3+#cOykh|pinckPY=!+}vH+^X+ShqZ8N5YRUjm<fShe1udyZ+s-
zqq9;|y;p_xG>=`B{p5f3WW?0InZN8WZ57|&9OJ$7`V+6N+0yyM)cbdL$3OA>$1g0r
z_sOemde0sT;jzG646`(5?H=@foM!sGPQbQoXe7xcW*?R?iOa_oOCj4^E|(dMhoNsp
z!dGFY1RYg_ruz?g_`5JVbOm`kbz12QRM8^zcXWrcAoA5q7cx$VNPQv7GWraV{NMS*
z;DE(()ulBH<6hXj-S5WSOU!&f>q12JyK8SnPsxeU{vPyRb9Kv-efM{N+`M$=g65x>
zw6DqOcxURGE#JN~^~B3NznUFi?O(Ifc!g6Uq_<~-P9gZ&OJSY7zoC{7Ns|XA(QV<7
zj3%rhzU4o(EPr_NkqOCD`@1at6_6?5hhbspj4J%00|3%K9h{!cQuypr^#jxnV`*HO
z+3Twvxz#~}sH0$t-ZepA02zgohdOH3pU61cRo{QghuzL}ui)3#Z^dlA>h&1wQ%4*A
z7Tlb<s_@dir4dthUhep3aOvACN963^zH;i{Cl7xk_QW6F%wDvC|IU5tOlK^-ecQ8(
zohhGdkeJZ@p2CDQrhaE@KY#vKyWCn7mds^~b2>6N3BX0eAJey1lZpaYto0#r!BAuj
zp?K(}<9gUkq*HC>-OR7EUeFj?5kST|n~~&3gsG1Sc(F~uUg-p>s~C^%p--laqY3`A
z+n0y6pP4>8dw4;~QmW=6$E164n7*!!NKT@QK4JSVA6zK^T$+^bR5X_?)1w-F5*)%X
zftSr1qJ(uJOgYS_q8xZ4cE$tI;c7V=k++bGBm)H+*c@4606u~tPF8(6x`m}dD-{Di
z651;K2|3Jj4NTP!F^^KBA$r0Qj8x!|OOGHt4JcmI`PADU*>{kS_BIYS&aFIKiCD)_
zj)(-#Z@3#X#87O$KjKu91S%VAnRyp-i2YbtLK4HKF`>fIVT5j+Xi!P3Oo}~(g>NPn
z>YP__7zjTA4o&rN|3(rB^Z<e~{lf#4vHM!j(e$3~RXfwcULoPI+jEM_+X&?|I|57x
zU$j#|A3jNX#i_9plhF?OMgCXA*2Ruhrxx$R!+tFNoEU}6Co-QgSP(<@o|K1rKtVrT
z8McrL`V3}BI!(kC0u=|0+~oqF^Zq&e0X3);(a*xS6OCA=6Plpl<hKxBrj>>L`cQAQ
z6Si^D(~F7G6AjL^00<rP1yd-DT*&-xT3V9HjK?EL9qWYMpw>CYq&X(Rz$puB4=J2z
zme8YF^J>LPr3x;F14oT+qS5_{uaG53LiW;zIWeN18=naxj6fx90VWK;c!N4XYc+j)
zrG6X-Jicb28$q0hNc{w}2x8Z}n`cz)!T@rVPa9*L<TB#9S;;CAXSkcx^?_@1T7r}i
zuGY?1jF^7!a{?bW$;pbaDkHb8vj<RD3Wp%@X4g|s3PT103WIkYG^m`B8#BJz@m$S4
zo&9uJpG1a{^1{nQuU$YIb3<KG+|&S|QAZ|D1bF}$129L0>bxJ5o#n8Mtvfm&)w@o%
z2Zh>36;JwsBFq^+a9ol>Nyc+UX-+zjE^_G@3QSQX8OBKUdFX|7gTRbG;+bcwH67s?
z`rtADQq%jJ6?6<ib}uhA24@9~Jyp)V{(^c$wF=Y&Zk^i033)Kqy;h%T`#wSp-+ci0
zHNp>KsRX!!0S7zw4p*H;W>b<z7Xf3S;!$LL@#TtK<R2-ht`m$g`j`-nJ4&3;2OYRh
zhJwAINC9~5`Yp!j3)9>(xrpk}fi86mb$S4>P{ejPkEs~fW$}>LcC;akp&GTwkRJxW
zTn%r_t>JVnOcT$yL`DN0K;|AE$}Sa(27=|2wF)x&Ttl*=2n(MMaX_Q1Fw5Y?6O7cv
zcG-xD)kLP-;Xh{UBG6kgG{P<5Zv^B-Ox87oTo8_vcTy0#!4Z<)0BDObRjGIcI<N9?
zJ1o8f!wbnPPBPlXRNvVY=-K6Xmb7NG_4M0duG9U%dJJT2gbaEm7k3vQ_98mK0gRF0
z4uclz+hkSr4_3>ZCLTY*1m6>Mwii^gJe|27!3ocCa09xHKk|X7$N`r6A<wYnaaVHZ
zHY;35`e}5*hogieA*l%I6%!s|=3QrBv+7*eRI@=v4X)FNyHo|pVdV#10-rQT9lh^$
zaaF>IiG?f*476cX5i$gl_vi>E7w}A#G8Vt^wF~)|m`d05f=c|vg`f-|^jD}EkR;t;
zsvQl421h~s#c~ZDB$hDt^n&cnFTC2OzRPi)h1ndYflS7>94-`RopnOeaD_Y20a(as
zashu0KTd+0NL|sE9nGwUq97Vi?!$M94iaekD?F&VSjDjZ-#2!Fs9)eg@J&asc8ynX
zhH&(RL}bdO$fOk11Qts`i~z!&Ux>Tbx&|!fQAWpz(7{q(nB>@lA9R7YBV?V9LPt@N
zhCT2uB&%>9n}2(Y31c~iZTvBYhc9Br0}p__7JUc6HK1AIDM_9F6L5a7ywG#}f4pLf
zM7pP6hj)z{ILg3Z6g+V72Ol}c9y-hnprdn}KCY~T4!aGZ#A#s7IB?aNDNJ6cXfivZ
zno5Wh0HYIDA~#@38Q1!!zlE9-qth&kSdoAY*wG_6<Uv}6QlKLnY4bmB>?$vzbN?II
zI2k#eQ9K$;0}=QReOD+TK{Zw~o%L`zFFW&o+KZMCif|WMZ{}k1I($Gk36d@k$P-v<
za;s)x<t`O;DjW?~wH=@f(p5Zafq^_dKdWgB9u71K<Tw!^sc}VSuid-<J8PE^XzUOY
z9_L3j&*;8aH^^||_#e<oJR<<|6lwI1>2Ou_ovc+iz_bZzT`4dwl-V;un?GPRkiUMJ
zJe>_pG6W<z@ZFH6Or#5ekp8y0wI7H_?j5)R_j0Jbx^^Od3Bq7z0~^o;M=7i`W_+{z
z>#SYHq><4zEiV*VQK*PDt`E$3^i$XErE%7E%)7{ZFXQyTlJji;_*1>6XU1XdHqvhb
z<csJH1s23_C<JZ#rz0Nkrw=l|l{2P&KDp?NKfnx<s4hS=?=g+zGS>S9VIRI38!F@w
z3j#vnq-!4nhTbvIB_>efq3Og?1sN4@c(eu$g`l->uWxq$g9rhGTb4%!rM&vGPj}PD
zrd8*}&qGE$!J8LEKU035^~o;3mml$T$LfsQGb^hzuG}8&CM$)VLIC=_PPTXi1K(pK
zdSFm^1GvJBpKoST$1ZwQ0Osu=3FGE>X(^~xL%}-q|1f$A#bO-Mj|f~g%Ed@q;7TX~
z16d}43P#c;{8+?Z8Do%cs)wPEXatAkD8Wc**AtU80j61e?84VC<j)W|9Z~6*HEB=9
z%Af(^_eX}>s+ZOu>U@G<TM#vDl*Te+_aClbRr%$W?3aBiRtChrpEw7XP`;KUE7qH(
z;y8qZpb3TQ5MTm*^&VWI?Vi`QESW`>-^`D$c?lYhCu7~%HgNIOkgaJP>Ir!Tj7gtl
zb710K2KJ)hTNwtV^%{(UFk(;zql5UJ3I<4+(B0ui)9{)(WdYox;vP;g#0HJi-4;5)
zA##g)W)BZJ*6EY_)_;<H>Z3W?vz|PDI&ap3b9qCPIxD9{$O~pPp05#o63=yazY+Gu
z()t0(QO^|LeB(KELO^2Tpb9squeu8Wn(kg7>*u|-V}5E=&bg~%(8Y?c4S0CzYz<r{
z0{^%*DeOBuuRtT7nD&@hKReQGSgS(-4?nCLWW~-ygk7<$GjO2GQ*Z_z47RUm#`1t9
zu7ot2O#KDZsbodi-)nXw2E{rE<d3N9p?#f7hHLT)d~Z%>d{t^zYh>l4H+Fy9T)l1c
zYctk<JLk2CYTxOz@x<cK2AnIo_0fh?)wx5W%DUz~FNh?784618<fQ=|f0WGcjfz_G
zrZ1b26?OC~TQSqSyme`x#Hk2UEbOCiPwe<MX;^-L0WSa@)dNly*8)92=wWb94#Du}
zB8e~!Rooa^>0^pSG1OFfof%VmfW$eRoq|3HU@(Z@%x13(@ONLIO17&Y%{@rc+A~h=
zdNcdA=xP*e+uZkeUq3VT#vfE^d)H^?xv@{@EGfTKWy+tqd+<WN8>pwF$m6acK^dP^
z&v6Ff;Vx^p<^`pGhwWNy(8bF6GxjuKhZA9!Rx=~-^MuJKYCU=V!!}-KaO-g|kko<8
zRDh!5H!V1f52fh{Ls3d$gtGX*fl41U)e0!&k8uTPl9-)nlb}RX4}&^UY{TfY=aXsQ
zWvA2Z<=T??N>mA)I=OmPVQa_onBehEI|HxQ9~!k#^vPM<eLic{ApgusH}g^(Z_XP<
z80dxvsnKTu!fIV;Zl($P&R9zxp4Xh86J43m@l5&c<ueu>AmV_LVNmE7J38!lEgv*G
zjVauWL^O!9y_)8LkHMqBSidAl>Hv)p26#R=N1B~#_$y3?hPA=A6~wQC25+`@qJA@d
z;tLhmbcC=^nfvGOI8w7MG*n?;NB!rqJBgPRTi2Q`=aMr{`Hu<t7q`jBdlJvFkn?VR
z?#rkLw;inX0LcpX#`uN^9`^yv8hSQ$QqRsk9UaTV#1+)n-mW}ac0nH+(%;zTm-{Wh
zYk83~QxX+})G8h)py7E5RB)?FbDZJ58<Kn`>q?f+Xe<ZZHB&LaYF5c%uu$n0?zvV!
zNiVXD8IeASodO5^ve+Uey6Ta!2NGl2FU*VEAz}g_{kS0N>GE!ly;gc~fP^ggMGl{~
zp7eG^yirL$G^!KG_v7lh(TSpVx!gK|6U}(`(T$moFWb`W3|e79ppVPVWDpIAViNF>
zh;<U0g$)$Mo(BML!}_<$Md==m^L26=o5XP*0)$k6lOBDLx6lQ%@y`Q8RVEKQ+eakE
z{C(wa;lF*^`#&_Vt}C*R@7!|w_DjXoGw$QjSq@Y04V~yXQ2}>r@Yp!o>GIJ>HYPRA
zh`G4cayP&kX2RCeFjhkCjTSucsfeXv6C;9Ol@4K1CXLV&a2a@`7{aR!Bx{nhKf_)+
z>Kw8k&h^H4(cpO*b6^<$Y0rO5{cg5^&Tc>NOf0livWS{RufAb(YlF@P)J1whm}qWb
zf>;lxBETf;cmTvKm;pH`xD{r=AEG1RfJa8|`fGJw+MKNpe74g8=#D^NE>y&nc|4PT
z6fbb$`X%&{1SbmtO9!(xyx>t%KmvC_6C4siTZr%W{KpDB@0YRF8B>+0szskJXHD7f
zZK~>8?tSQJ)%m>m$Ijr|WQhaoBJ>CVnh$`yrcMts#B?_!aKj&RKe~R=GqcjhHuOy6
zq%Z)B(C-cmRA3^21u?^7)YpOajjlM(@N;M!u1gS>xSe3)u#4D7^L;oeh6xM?c%$T3
zx#OpeYjI54ex613|Lnx<jKFJu-hORI-xH4Q+tU5UCUTqv?GHfw@)3t*@-V`}9D-}$
z3gJcyfM+id&%XfA=boDf>!%x%-$C$(cU51K1>05ER9sIpQ>~mSuq^C85oqM<mw;wW
z(O6O~)q+gNERB=e$YvY{RUl1O8XU~8@}0yLT>=Mjzm0BIzpB-iY|8WAd9)i`QXb$9
zIZ7VxQ9}x8Fxl4Nv^is3GdU2_2s(p#Jzk}j8PhIq9XM<fZq_>;(uG0yhPu>_LFJzT
zL_<$GXut)e4um+hAjQ^OBHg5@I1BD+*!P7{I|PYt$(U~XF7qGP>o}c3OJWs)$&=Ef
zk1oA8`Xoas2#iRD>BnhfCL5w3(;0z>=fXpFBGjKOdVA9Isd@_omQq=N9OoL!wXwLX
zAJt`mr5_r&WDPSW;6NTtw%|PNqSj@FrdUQW1K0n)UPoG^*!+`h{$n>=+^St#m?XR7
z@%uw=*q54?TyCeDBo($tkCUXkDCh>jF_GdJ%*<4Mg}FS;07nff(-1wzEXkTV&#P^c
z$Nu!_3%qinA#JsVUO>BG3CJ$eHy!lNaG0x#VwKbh5~lecJ*`whr@d?^%sGE**N&e`
z=i?|#91;Bl4}3{ISrGSl-!*l?8-3f`YSypWdHLM9)<J$J>IS7IjVjYt&$~T8u738X
zscQxtJ<zda0yeQWcP)>p#uGhzvSY089$sNf)$m&wJ^8i{Ysu%EEyc7%N(3fAf<i-7
zD~!S-Cs|a+Cz(TkA<=S|YEg+1y(Y{cO_}wAfu0n>K?JE^QIUXHnK<x+-K63t)g;LW
zd@#85r%Qr+xQ1k*!bJ-X`N&?o>s!Bx*C0Bn?~8|9swccKc#n2+akTYf%;BK{x0gjd
zxwU1HPh)UlRrlwcw?EQ}t&d)te|*Weq^EM-hovtZD(q@`ZTyM3k=-M9G-n?7iu|u<
zhpr0Ly622CcRBmkZ?49ZOn3Bs_h3}@l)&!C!C#<Fl2r{@Rh32Aplxd#ZdXjmTy@D1
zv322s8QzB+OVgcbkh5)I?zQRZf8+Q}pvxH9dPZ8u1ZY(iR8@FRXqtYMShBg%2*n&o
z^B8^fkr^jvh#P}5Bus8JVmrildwQNLogaqdiT$RW`Y5jI(bJ1YRBjxb+q7uH*wB`m
zi?wa17fzt<SAiS9|3{y}39S~dqPFh5$e_{}OLn()ewv!uaVM#H@%R^yL_FQ*6>|Ss
zt!Y90JMnK8F?RB^FKb<M{uuXkPQ_=DH5V)PonFXsBim#c3rb5QM<WJ1CGMU0SG0ST
z{(a_MDf3$U^_%(g!+#pF`>Rm2^p3$$KMz0B37e)XS*`x2lAG7eh=x!a{K|Y&2uvj{
zS%a;xV)zM6Rwcg9WI`S<T##pRflVAIsTT;GK>w;M_j}~F2fCqAeKU`Au1L<<wD0)R
z*x-w%+*$LY>vLNcji}!8QG0Mgd24|ZwSUWW7Cn5lSI~+y{}--TG~RyQ=g~kS{MaXD
zi6MWTj`GF{nF068FYvzaUXI;9a7V_m8*z_siTqb|<x$KUG_%p-EP`F5j*SF1X7!}i
zTP|nF-;Ex^td9onv3>Q%?mGhLtstSNO-NsdxJ;KUC~8ogVYD9}|NFrY>CO&Q3MpNg
z>cB)bUH|bO(ip$l%%oQy3m3i-K*jaM5iR3#D2+xLKjTyH!R6m`so-104r6A#b@<;O
zPxU!)a?y;{2bP2^kEm?!)ZR$D99Gs^U^`Xt<h<^h#^_B2^{ZlPS3Z{VdR$=Xlh;r8
zUG!1J(|_&v;m|0#XXMbF@SjrGoPF;=JnQ@JSEHYZU+|Gf_WKWQx^#SHWGE?TZx{|l
zol^gbZDH>7*-?+5XY#+VcLhXlUGaeaKIDqiTR|+Gm|*08k-<qu=^_adnhnFnhBY(P
zTIYlTaz?t4h~wq`?5K8OHW)cJMBrKgCX)1qfe+tp!-@>=ANh|J_hj(Ka_lqb(&);T
zv8zrsoLo3!|EoP25m#?77!kB_&8njB;-8#*x%FPw=x3DHeAfBtBQL!A?xmhN@s$(2
zqhfme#&w<kWPIkA(uYsZt@|?ONaXv)J+q$^3{K<9-06!a_%?RFKK|YA&PC6s-AxuI
z4NqXewpfS&C?k+g&Dwem&om$Tb109i*%t9gvSEhLr#O@By%r8V6yA<drpC6}P+5to
z_@?LSsO-RuFCK@F4*-M84kjBhJHgPHK+K>F99>KUiQXJv(`75S#9sE0&ithXU3Y3q
z9M+OYPss%_HKp^SE%!1u{bQlO_m(~W>=yGazvo;UAGon|ZpDWkIpGT<8Xh0eSzV~r
z?>fD3{NKKge{5Yw8jq@)a(?-&p;5;c&g9o#j|lFYb-`#$KCB2&{9?rI5s3AF>|%^|
zt73{{X>7&1`OnW=_;!8H+Ee*KM22*2M&QMYSGqnpCpiSrR+tdL91^3SEQLvXK<j<S
zn?7REqjzA_So45CP?$!80*s$XhL#|&tb##C2TOc^_kUlf)VAdBJ?dU{GJe8m%I1vA
z-{qO>4>gV+y5;DEG4U+1)f>BJ<wklQky9QRI(4Fq0OT!XF*Ix^NElmIbHQpVh+_qG
zFCq<?K|3**k<5?U_T8zvc&_VSO+weV3w_!;^5aPrVaA9kre?I&5s?5WBQ%%1&qJUo
z0yU+--V~S2SxSH4fFX>+S+wpQkkQ2Gq;NY7o=fukc*os$CNr#jn_HGf?J#|q@yhtj
zt?iW;gO>aH`#f6PdU`>KqdNJG(bkeoTgMd^(t&-5etjC}_1Rf_@@C@NOYKX&$FzLh
z-?C;^^s|8}841T-qs^V4VdHB3>^e*<*WyQ%7?wRS!4d2*CNnm!@LAp;RF#IcY4V`b
zv56HOUHHXMxaqZTF3bHpxhqfs)_c^F8$tw`ErJh{B)Wnhg0#pS5gwA57{+8xzXG2H
zS+<5(G()mWC1UZM?L3Bz?mx1=!U@?ca9?w!+!<e=+q7f+jY-X~*&2gUP`p;#(-?d*
zbELI!{iUqfn#P#@t3E2Zhf3>IS8VwcBfeRRs+#2cO+Uu$ihTOG*gdMPYC`6uD-|2Y
zZm}JBtGOOyTQD|CGa11)H*uhEW)@1br^}nds#jE>&K#|@O5^J@Hg`o<Z(QKUae-%7
zF3Y|CY)xr;E9jxP4`3&~Bzf7m9vVWAHGHQ(UB2|Ly3ig?dI4dM@aU_j87uLbnSgxb
zA1|!L2*K15H7(xfV$2<tXRAK;fAX(4f=U;3G#~C>78%^KwmEp+t0S#PHxvXrO@1%>
zomdrfwI?&y>)?c;gEt-cuM-!ZkGS_`|E{>o{+7q4<mcJeF5f+OPwVkdGKx0d@GAUp
z_1rWt*@r8*4fe5_j56geG@-7ZcPDXL*J`gWw;`sc`Nhj2U6D_IP=*z6wgOMZU2!JZ
zSC7tN)zDqbBx9ON3~)lR;ex3zr8m?ET^bGu89!e&LfMPC3_~ZF!vNgS5ya`gfKhDq
z<XMSsiACa(9;+H-cK;aO6Ca%6b8x}&`niGcMLje3%8Iboz|<Qb7&@z7N@5+Kr*_S#
z4|&|bPtN6rmZg#J@cY^mNm1|TbcM3Vwq^4_ebHJqrAs^hNqXRv%WX{yV^(o!b112t
zv_NJC)nrHX1kq>Cs6QY6sMOG1#&Tw4wT@qO_1HV($-54^cmuKZK-B8#S_!3uH4;G8
z1=2#CK7$bg(iacbpAc~QkI*QnqH&0gK}#n#JCWpjfp!k!G5(l#?Knb4%((01qPU@<
zca}v}X2iUg9rR_t!-JAPTfcT+`^WvON}u|=PsT^xgC6}jE54dX@89Cco;5Tv>_B_Q
zfk;cs$|#@K)$Vsot_W$vvih>wFE2Q{Og;9=X2a@q>*)p1K~gym=eP%aHxX8h4Rtl;
zjv8vX_BskU{(U{WAf{i=4m@B1w!HQJ{fmJ~rP(D2D5xbwrO_M5$OAc3i6^KfSEW-6
zi2_K}H}FUmOlVXC=0~{1U<@g9>nA--q6vu1w;DM}zWWYEYaqEGWd6)?SMmFCTOzdM
zxwh`+&r&M;zIZvly!o>&wa2Y3O}SlFFAdfFpQ^f5k-BEiGf{g^c4tIr*OuZLSPS;;
z9(74bYnSzlu9}iPdzMyq?2|3)zN}rp`kIVIZDukvB*NiV>>=wo1K7&$tayI&pYC2Y
zmEF#bvfnE2x8(8jB;$T>zMk8Xc9ZH#b<|03DGiBJH0z=O8vg{TE2fV$_ojTH!G;Dm
zP|?F#8p>(}gdH%Ba#=_@qa`$i>xRKtv8vDt!pacAScvg(gKe_9rVjJw+J%WH?Y5M>
zxbmJ!oi(ZL?2WMGg82PjwcERv#s#-z?c9tdl!V4x%sVP}%e1Yz6&rj`Hhto^TU{Rg
zIEr4-znw!Q8noF>0|n5pI=*=16LIJ9D700Z+(*Z)+`Y5#(Y?7iF<-g{>&-c1eG92q
z{g-ZVm5LKY4YXHZj-bJm8`FyzWDT>xVZDVq4D!`mdmJ2n0W&b~C%Fquq_CHb#oq*p
zbplSk@An<|<|lu;@zl53^@q+~YH{YqY%Q=3i1=#d(7Kw!_8w;*)!Z^=YJNz%ongSm
z6Vo)XHpw0C|5O)Mr_k?nLVm1}s}EhLAsRmr=fGnKonx>eQ{pu0jJlSFrw40T{VJ0Q
z835(a;e7?JwnNul<dk5mkPh6jM~lUGv`JE@jjNPpaG<s07?#thod?XtDwaOKzH!BX
z?`2J&^j&@UEper0TjZjM%JrQhTYBZI$fw;spZ9;}RZNV&vsLwEzq}whJ#ez+R>qaA
znb;0<sWUrlUpxM;E#o0S#?}vx{CTALc3$d@1ICa1@VJ!}Oo=yU?W{O?GuL30UU6{5
zDmtQsK;IKN=hk;Eja4ljnKM_Hw||~-?y3gWf)wA`u%bM_2&Se-#LNKl$s+N>;gaBt
zR2p+J6Q=o}st-h@s`g=Mp(`Z}5gK3`C>I4Jn7Dq_^CEsj^ZoA|n*W$OcI(i2CwDv=
zrZra{tsZ;mbZ*r9?)rXzT+k=u<9{U-U&UsZb$*LZp3m}5yj&hsI+QJpu{6IH8Pxnj
zn^)xzqtxclTbeRH%4nUMQPJLmJ@uty2Ui@pmJ#vc$wffTrRYo2y(mZ}Ez6#@`hB-G
zSM{<rbQDBa_HEiMJd%Dr_xi~Ea1Av$c%}-CMLJ+!!x4Zv1jOzFA2{dW2lNr@LrTE7
z)9H(rROP8ccheHEH3A=WW5Mj-16UvOU-sHJD>Yo=(|2fY^XpO7sbPEGTyV9kvafgc
zs-pX&g9ZfkyyRW<%+kP(fBQ$D+}q1jiw!Z)M5QD~-c9+)|An@<)<2%S%JfgD#tS`O
zv3H&gjoN=`PsyeJE?2ardv<)~(e;m~D$$Rh9jamK=kX4ndz)T~3wlY4dEe_hO|~3g
z95uP&#2-{P91CD8X6?XWYhBlX<OTh*2HKN?b66ax!A4}J;xp`q072RVi07U*6#W8(
zh9shl-T2Iokr&Ki$W>8>m~@-eCNX63OeMyvum`_*bbEee(9y=8&(>$UPk-E+6LK$M
zd-vz9{!fm&Qtnmx<H#X1Zfve-X*#|-v}bWhhUr#hP~XhktK^?zhbG4UJo>TZWh09$
z?S7qJmm7NKgdOO}iw<h;tP@bTUqM*aG5q@3{m#QTvFKTN_=}m%_2<`_5*&prN7w9v
zVH&&{<<4|uA<S@^`3M{!slmAdR189P`as9zM)Kh}sK0_d_-@YdiyrhHk(oqKEmz-v
zd*vfp9)bfK+LxwgZaujyHFHb951STxo0mmAeZ2YO*6fgbDOvIRC*r|kUYA?Fj-Fcj
z<h-7O)VIX$_U&&!UTUB5!jB>IXRK~7yZ^V;PIt@a8=D5qx7$+LtvO+MKp6~uMf8r+
znw6AwhUe)83?t#0GrbSZS(5fOV|a{NsbODcmt5u?NU_D-ne*y*aY&Z574j8An8ieP
zfN7du2}LybjwcYG4kT;;H|Qe*D=-(TPHix)6bzkE7LsFV!0cBKBQX*))?9!|$!A+C
zx;mCeRrbwFw>H;fT6^qc^U{d_+?`rKcwN_oEVkr1Va%L)sv^JIZ|&9vUG=Lf8t!C-
zAs;YeM`8QzWsy4u6n%M6T*|kz_QXBqUAbY69j`@~H^2UT%smdI+=Pb*<l_3XGshn4
zdhC&gcsxcUCpdrOjYB7+g1%7fipfz$WQXo5H{my-@Ni)eO}-oBO)rp$DHq7$?l>JY
zaXnp>z&(%hRp}TkUZ&TH0>7n&d=iE(s6vs~DO$e4PbS|J;E4H7Mr*Pu$rbJ&s$eh(
zqjB6ariXdE_831e9d{~!P(s_CoUp6c^7?GCCA8KbD!htq<lAo-jI@^c_sKZe7Fe<%
zZq-ruwG~)7(Qik$1lT#V_I+z#&z#trFAsKf%#WN>a&uXnT#Z1GO)C65e)#0tbUaZ>
zZ%bj43K=kiheW*}az1cUa91?1g4hUR$LMeL;`RYGp;8Do?qF?(!_~v;sq7LzENJ$^
zd=|P4rX~8c-B}!0F<`;BORD|HE<Lt1?NW>D!kA;9ZCo>E%?DK%ub6UXdMAF&yhmA2
z^R2IRSdW_e54WY;15_xCi@BJ<z3@TTA-XR-g1`zUlKz;8=V3(`EAwVnKkEK}U94DX
z-9>&?XS_<$0Ys+*B&X{k2@YmR=IGL`&=AXe?b1{hAW255%EqvzkrUEyN%Jk5pi|9`
zhoHa%Cl|*(_Q;8y;YdTdPrl~OKJ)8SeEQ>~XX5t1I_S~bbA6I-&3$gDCSMy}wIOR&
z*x{y(%dK5=Mp>I*_%h;ZQ$c*t_*1)Pvq8-_c2141E-YKamxnd(o_o&SIX@~0N<~Ru
zfZgLRr1sMv`!z)EFTNm+54aT{w0=$0K@DVxurqI5pE>Vtz!;J!{geU7)-NQyLAyn0
z%tJnO*O_i|n;~KW!!t3WSwUW1L0OgPois=ibwdWOgAI@6#99|ADb~RomqtG^zCH1j
z*7$nllYf<O#Sb;KXY9GuyeK>yGzVVXbFPo!YMxKG*XbWdJ~^*#u5I73*ZU;C^5cdl
zS6zQ+_I>ZR$j3`Ed_H83ibpbYYP%a>j~nz-=HPQ~q^YV)=OK^v<f==Gfw@#@3=sFA
zFx&K;cI>l^WW$KZ3;Q-53fu>(h-x^bp8_}(Q_%_TC0f6x*QbGOI$EyU5uK?92;bzN
zW<){)MK)nhrqSa@ioi3o<}3oQRLp?ThQm=6t4KG0LR2%S`S{{UEa%S+JMw&X&&c46
z8$o`>v(%17&o8-pzU<Ctt(h~|jIs9pzPqi=>*(?Gj_r>gY->(0Z_SK*y!7q)Bi@;K
zdQrS}i|Llib`Mx$dc%9mLGd$dT2Pv4>dqf+ZB7c)dT0rXl+jqOV~?L~Gk)B%c2j=+
z;Hie^=M`*h+G6Xv_D8G^$|7gRV{O~o4x^obln6ZIsy+pPO~7qvf(}E?>jV;#$B7`j
zAua`O+Z>9LVZ9m*4K!-S>jbm{w+E=U4xuM7px(@$XxrD3M}uufL~YLF0|E~Z8dyDk
zwY%lxqU@>561L`7oAct|pZh@pyZz})VWvOby_I9Tbb96Bn7%h!N7VQ=ZSlF<wQT%b
z`^@gJlK(V4wqf<)B^}q0hPjnBdXQh|PAOsEJ^h{r5Z;UXpo`L_OWQA|RgV&Q>R(iF
z-`zVSh)}ueVxLi+v$BQ01W(1QM7fzRha(mA4nT0l$p_>29KjomNvQcpV4>(InMFZK
z3^h0J<tiKn0p;SP4&jQwsgJirv($EH!T8R<#{MgM_aUDnZ|1ohVxGx~*>`Ml{LoR?
z7RCqqt(mhs;7V5PP^Z7|_Dgseq5s&yd-gRii?;e@UJq;N%#C;_!8WDek&sr$_D61r
zKmAyH@@FOXv2}~$pNMW<8f(EOqZ>~+HU0JyJdZ`tN9k0y*OJv|UpzKpFJpgu*}tE=
zX)zs#FT2#X?P{Gs<d`m?%Q6nq8m3D(XewdiBu7kYa2C@LT26$RAyS1Q9j2m?H5r%S
zGe@&JG%?^Cc?goWUdyLDGnQiM#4`tv71)dgan_cn%(>C~58uncPYQmP(f{Ys<;{<T
zeVbGK4eMj*$v80jv2_b4Jk@yTwYb2R%pW6Ko94$?ukW+vT(`VE_1x(rkDcj^8drQ_
zUTn~Wth1xdKLs!9*MnU{`aPMs0l0f$x8dx}j-QL;20b!%z|n@rWzkP}G%byJT;QxX
z=srIyMSx(WplNW$u3FDk5H3j2>vR}IFPWg*Fj*)OKyK7OVCxipgF95gU#JR&Ax37#
z^I>(vra~PuY;$|gd$YX?PhKf9+{sV=$nVCh=G$+)9G2=^i#yQ$4!FZ4=K*=u(e682
z^NNBu@_IbdH+AUFgC}q1#12ZuGh6BhUp`)+ajLa*Dcup1A^`#64-hf=l5FnD^RB}K
z1P@hoe*6U;tPeQ4VcKGi^bm*<CG<&G*<Bz5;gf>Yj0O=QMu3;2YcP99mQ_-wbMj22
z=D3d0Vv(SC>IFd0b$V#LeeZVPUd~!pj12B?ozjV4YU%ey9`Xk*D`t6*d8^|JmR4Z#
z(ikMTLhw*hd6K|dxEXwUC>-ih(918NyBWD^+l}kvR;GZXnv9>$v}hg*D2UWFJYNQr
zYv7oWz<V#W;N6XXLN2t2s!&ZDZ~&<OLx5!<RfNN0C@j$!PnT=4ZHVUPIKsAW?=E1?
z*`tEf`@Q+fk<pu$>~5FZo0HDJ`kdRq1a54q!=y^!Sza!c=4W6$-K`glG$;QZxNzU>
z8PVtefDIl{2zwaNG|RXz=^0`K#P5ofg0zSp{x}3xTLMrmYSA><l)gBL%`TYJpb(C%
zvK+e+(LynRC<&h+1v6@kJVxao(AdWN&Vvy8tq44;RbYK`ZSuUAeYzSypWucJ0CEZT
zp_;HRidg~_!Nig1q~jjo4?g_r9TMm>a_caLpX)W5P|Aovi|s<d2uxA#HP9uqVjDf1
zU!`$@!f}fy_{3>hC|K2lJn%f62LS8nbqJV&g9c+BC4i99CH^KpvUFrkNng=gjT~~?
zqOPW>X_up_w^eMvI-sKb8?IlAja9gpMbG|&rKDI7K){avCz!G^d78HPfJN-Wh=*q#
zrQ;g#PPSOY9fzeSPX#mYk@->ig9^HI#$uvGI%W}@g^_A1W&Q37J3@oM&PxpGl8NIw
zF6nhqci)+hFVPpb?CBQmD@N99elKQR>FjeT7SEV|wD#+Wr+3!jJnRQW*C0Xj+zYuX
z)IjT<(@C?Y(%bLAv}DF(SN;H{fP0}^LU*YLOR-JrvY01@B`2@J0u8tfXG{vOL0QOP
zJw~=!pkEne;Dvjha;z|@A<;~<qN%_7P>T$R=+af)Vr^K!0e)yC!{JkVk^y@84aPLX
z_#c4smcXyixFKtSO9Vzfj;X5()XWxX<~Zg|FWVK0LBn$)-Y6jh>PZb0jgz}k;d>wy
zhmi4{F&vSL9KsFyNY`YBUkr}Fsp}0HH%Bv=HOd-+FA|YX3;AsY{gSvisTyN0dI5j{
zv*%Oqt7r02A9{4d&u3^E3V`iMWQ_t_{$%r={CF$;EDFkmeuYNqC`9Mz3iqb_4>_#D
zo8n3?q40<cHc?(@oQUxR4DSWAHQA__HJ0Fj5@mhrqf2YN5;{&|D%Ff%+4W;>n^x@l
zf4KS<sHl!??K=IS+ZLhG#!(b<T6}a9oht$g3ar!JAT3d8LE|GY6$AK)5fLLA?>f8$
z8I)Eq0}6OR1T-c-LUJ{enE`_c4q|H<ANffxqXY=yW(*SZn|1%S{%@Z)$=q9s&FRy1
z>ePPk+O?}HD&TIwQME6RTsLfqD}hP%ifrOb0tK)8Md5%J0^RtvV7v<qK~@)h7fM}{
zi$NrgfNvR9U7}K(Es)-LM7J<)=bmg+$h9C%7#Fjs8-gJltMsQKBMBSP=dVcw0b*(F
zS*KRH+ruYv069S6=5^J%#h(wHv={s>`pD7)ANe%3Z8f(()@}**i}y-#R!*!s`fsJ?
z_MbIU{-hF(4bx7$5r4&@!3|*zO`V@6u*bV2{B>-evpYR9cS&pCzR;=5Q#V#7#H|O%
z;7#cHb3#$PCMlxl_b~@fx!A+hMSh+Gfh&el2LMl8kRecYDrC68YS!u^hgEOL<B_r>
z-$%j`w2_lhYzUed+28^@DuRnd{EQS<@Q0c<%%Nx}J}3he1$g_7#q=$&1ctp8t}jS7
zUufGJJmqTY8#8m~ZGN@Bx$=)k8fRvX?zZ+uXLt6$qg}H;I=gr%EA(XK*Bci&CzUMz
zyzhmTZJEdpde;A#t9+apUb*bL^33qF>#nGik@r&;1uQR6rqK)KLfi*=Gw|%)M+=7*
zRo?VS-ohmu|5N9I+<9AHJ=i4S_l6mtgHvVJ6Yy0Bj|IxeA{gz7`Z}WkEP#6eokJiu
z27secxN%@cEy)(oY7il%T9A%99J$S;U%82ATK(#y>^Ble@6U{w(pg_Rx#?Mf?MmIV
zf~3pyRyPiBj$SKuPdNJRn3lL*zt1?>ILL8%Uwh4(up;m44fCtB3lBVbG-c_5hyF`p
zM3E#BG3i!qM#u?IjVrQz*}#Mj*WmgEkk{wM4KqmL5qS06MY%_gEf86B*15-Nb3eR8
zl^VjX9qMs0H@ng^60;&eu-B|jER#f0As(~yW=K4Uk8PUdKOi46$MD+?^%xv?0hmz~
zhMXLnz~TnhXDl}46mVCt=;XA}(J)zotAg*|zqPV6>B#8P>l^)xQ(UPYx!mQe+ckL)
zKg~&B@qk}Z_Wdq4X<TpLrpOnB(RV+(Qa9C7*POBN`10#Fq7QZ}CU-Rm?|y!!uV*h-
zbsWfO>F(bdRd$!&j{>eHVKRZptbRuO{t1_U_sz*A$ooCA`E+!at8c%aGGXWJ{NwMU
z9;^q`=7>Z=as+sqYYDK}y<@QV5FH+am;nqK7oS*0aV-ZNb`=vPurBw~Z<xY!r1`B*
zyb{>a(&p!=0UFC$cNAGbcF*D)eS1TWcQzzdG`7AKJjK#{Ag}FD6a4;bOlR+>S9h%2
zZS4=6dU&RU-D=a#ougTyle{I*e#bUM4f^i9`()6mn2aD+N8#7&7vv6{j6t4u-jJ>4
zKANUN1_XadD+$JqjDSOD4(&f~Wyo#%cW1(k;NQDM3`fVldiwgU&9Q_VUS>^bIH)`=
zY*Rd7<oX^901)ylhgrbNT4Hc=0I)ZUK>1AWNekRfBy8cnM6Z=Ct6obiz5DI2_wpht
zTy(lJ!>bO*ooc!NFmvhH!Kd#eoZMbMv^$~t&t3(|169kfg#;X$m3ySUXCLUS30%>9
zXZ?bcrk=gwQ?zeJoD?2rgq9y_WmCOp<{I07o@rY$*t*?V_~rUV)m;+$PlZDsF8&bA
zTH}+|EyLKMV)eHM%07SN__pN*rwalXZif@5b*y;x^fds^BvGzSKwIphD6&NTvB2Ay
zXf}ImiZ0@GF>^eeb=q_r49vr>;txbvL`FRq17y&fVDn^^h6G9Z#&pCN&=>L%zdG?#
zbt8{&lD-l7PAb-<@HMW~OIUEgk6Oj7$m$z=qAI)U6RX>IYf2AFBUuY4-^5%?YUzsR
zyL%$4i)WUuEgZF79DGVmQXyFbqGb>@Va@v$Gku??Ey5PgJ)duhsy}@r=~Ft5JYVg5
z<JDJxdQ*8@W<WfKq%0US*efAE&%Du@XC7S1t4rc)MS_Y7FyH_{`8<75b}O3kCkMV5
zmgi-|ri*jf^syL#VwWaSC*U?!rLiR~Ixb}W>6&DT^`(4xpmg%!(C*;s<=Aztr*Gf9
zvhIu!<%Xn%KQ9b8tXfo^fPbQ<v}3p-<aRtfau{1G=-O|E?8O;;l^}dt_IMxJAwVFB
z0ITN4tMC5wCXMI`6*%h}r`uwF5DUii`D~(i0pns&2s|p_lZGq#WY1-GR7p&&Qo~E3
z2LV+fDj+x}hZo&rZUwWo^HC4}lr<=FodfH5HqY<<?BR!7LXLB--`l=9xwL(EWYyB^
zNw;0?+n0~+ypx4<2(m)_y#^#s9I`DD?_|u^x4jind1!IS#q#>kvDZKLN8h9QD7daK
zsiyM#J?7?-eW4Y)v?we`IO^X_J#=Di5KrU;WyIJUo9#PZZ%f`HKsP*5d~vadhsThX
z{zvbq(}p~9Yd}qWs8mMm76;==eh*DABD~MTZX;|;N|q3SD`uKrE1Gcw9w<nQ9m7CA
z^)z{LBakT%eSrvLA5Fddzkyn&$-TB)WUZS&Jks~KR|03yDLb2g_s#JVj~xF_kDRKd
zi~IUMI~1qzLvoSylzCu-#G&-C@|*R^NIOeWep_z2{5Xqt6r9|Jw2&8;wkB=J8{KU#
z@7WMmiF7Ro>rBHemWYliV)+_DAS$7s{O^@V?)P|UO5=r9Gh-jV72$_@bnsbX(bPAg
zn>Le90l6)J3u6a*+91LqCMT6pK@yCn3}kx2#5AR<?5Sle$iOMEf|+hIcyd|=m<$|+
z#_#|h#a4p}tA3+&WTm&qse#oyr`R{mpZ;nr_CHLU>wj32RP$BD`3WnU+cOc)ZNMJk
zlb0<{s%ds@iq6g(&NMzbG{CkSmwJ5eVpBa+N{v4*jZ40g*SlZY(cH5q(lE3sEXUBA
z7VcLf;Fai)K?RlrjU3pK@pj{=r^jPw&AWJWZiUpk4T&%6IarDj`Sd+eu;ySwl%7I-
z9YEm%s4`#y2ijr03ZUzd6ej1Gi8(p4KKRLm9~s~T8hBzVE7%1HvokG5f=#^M;FHD~
zUxz!qdplGz%ZPZnoN1G?y?ielkb2{piZHH!XR7UX<^7e9KRYyQynhkqd|lAR{mHk>
zhu?`R4}2|sQQO{;ERQDxQ;JiQU}|=RRLAdE29NUyaV@M|w%B%MaCpP~9Hnu&<~A_R
zL{?@Z4wYSK#g11!POYCi#idOe|6TBjV!m(xbZE>KE3ug1)@Ff7>n<Vyz!jE3T9{UV
z*GS4Y5$+kyo})C0vPX_-MYK&ti+{8Oi$YQWQ@B~14;_t{&d7?)k#%q(3-DSc3Ce@R
z->mH0rZSdyd#1oaw2$lx@h{pN?$?#7T-*LQO~BN|eQco>eXwwFXK2OE)UJgA%ld5{
zolo~GBl<?RGPc7AYK;L#mW<r0-7&v>nR*d?lKgpoT|9ZlJckX|$UJJvpTKnDxQ39N
ztt|hI6R#x)k0zZhp4mrnh*cVMdd62}cmUl5#egfB*h~}|t6|geFvW%3h3u+a3CsQI
zj#kSY@xYLXhrtN+W42WkXf!1$IK|3H(d)hKG<Td8aW9Q|#rdo|>>+w-CS2^fu_gGp
z=XKke-lw}TAFe6Q>U;dov{jurUr)CoYH~^44tURZ$;zfAUo7`8t}o60(0I2|!)FE`
z_g>#_TPxm4UGT!}pJ#<w5Zcb4JEd6OTu9RDv5ThyIt<Yz=qs^vU-a2GUq5)iV8&pX
zx^?^Vms$_BB0kK$B#?h~8bblAh|mRf7{VYDOr)vZeg(}Ai*>*UG?M@o2N{Qz@p6&1
zB^D8B1TD}bz<?fMZIP~dIB9uC7WLq-Sx@PPRyPKF4`PKyWOc%NNu{2yBkv|4-gdk9
z>7FS4XvV?-@6-urnuqon`x<tNIBT#kYmuRA-@>UkG9m+()rUOjeu@JE=7QpkN3J3M
zFcMoR>}TIyiJJPyvxxz#?--J@zY71iQ#E83-E6!QF=r9omww}B`BCJQ$ONOc`QUBL
zGs0`&0gT9Idx@x`*Tcmic7{KNgTw0xR(Xy`e%y#ihb*+p?B*X|>$;$tcdzupS!0^_
ztia2WN>_MA0yLYT@$K5YP!2dD)ParE`ru*E<A6M${*z@wW%7aTquUBXoq`|vi*waT
zbv9>DY8n2wko#Iz$C}JHUXA=|2Bwh+VmFCkMKS*dS;jugR%VUEif3@BqD&N-Ks&<-
z7`Vgs8sHeIxb@6vNB`YVGj;}HK!KM7P$tT5<B?|O{qozyqdOyCE@U=y<KNaD9Dio{
zimM-27Y~$P89tC~H^%F^L%S*u7YNh^rW|vbJR`Zl(Lywl)>goUwPx!LIc|U`KA=^s
z>U<|C<JJDsS~k}E`2LXaDE}OYhc2k{o6m1;PNN27CmusLbXb%mpg{%vJM;n!1;kLH
zrz#@F;$;9-QO#-|7F3|bTalh8Ya~4vSv3qdWH8bx>dA_J38BB1CaQw{UNnE!wmbZU
zXAH*8vvODTOgePElyePk)_VAUc`V(pxOM*Il3Ax#wBFc^vyM`2mew2ll}DZ|^Z4BD
zz;g)9pnC=pGNLNtxFv2{Vj(ehq2bEAc7-~JQbQ&E`^{b3=V2D3Ny_J0?U^o^4=lIh
zjo>9k&U;Q0F+`_C5q&f<CX8K4Ds`GpJYun6ryNT>@f^k3MN!!!kup(F!v_->l!?Is
zsskhV;!8KD2Z<E_SxZ93(E?oY*^d%(Q>KtlRDRbBN)*G<)R5y#uiv&SdgoUc&&ti}
z>rai&#(Za1v1>#0!OA-}gf2$jeX_P&q9%cQ3BxVKoPg^XlLS7+J@ga<3!{Bw#FP@v
zsf&3%If~yWW6~SREO&AMNKMxpCk~xfGy_Lg3PZ=PQIyG;q81hSN4s4G8H^T~4M6}X
zB#wFbr{)&|V*S}MxOTHAB5pGCfNOPHH93nAm_AYbg_s>Z3hXTwUhRebuwC^TYvA3n
zNz3|^vwQrGvdhZ~PH~-2lg{*gxifO=A#5H{_IPuca^R#e=}_DSv?Lc%(Tt%m@(E@O
z>j<tf2Cyweh7*jI_Gk6ESB8gZb&9a6ZS&mvkNb{ydi3f9tgU|tAUoEJYIE!wgD#d=
zp)q+07vrJrWCf$v+Hn(u7N}hTBt)nT_tkLUrJTltcE~CkLeWX=m!K_pDM*x&q@N*s
zyD~9h`<Qs`70&^ixv6urxv(dxqPg3#enI77kB^K^2X&VZt-rBnfwMmu6dc_#zv9&C
zone#4H9s4;?P0T!;e)b*^x=?59f>BQ{uJ<p#sLr0I=6f#PRj0cjq?cH^1F_~);%Fp
zu5Ni16T}d>*gb#rn12(I1rgzSa>z{>ZK+5#LahQgTtF&Kcm>acpkPY`R=~P6P`xN{
zG1eT+mB^~n@tl%G*c=*JV4n?DwwC$!<*^@AZ{OX!h6Rjy4PO1~+$%j;yX}`<IJ(tX
zIJ8+a;lf<KD~%nj+T%YiHN<CJTS!gs&mMs-oj(`+ZPC<{I!~Wb)0-hDO4dv02JDYE
zQIN1Z8eYMv)KH2zy$1?|dVvv5L)Se2Ta6o|{110jWgT}lUERJkHgPMYk&S(Q#<Tl9
zH^Gfsl1{xoADoV%rp3b4JTWavD?q1N8d_6|cR>ZnAJR`yNjxSlNQ*kuV|9A)J6;SQ
zP(}V=B2Jj8Q-gCybQp2p-H%QGSmM<08nKC7=@}qs4>o*j2)rLr<sC!QKa<C;Cu@Th
zZ4%x~WB58Vn@#6YksY5dZ>3H-P4Z8m-}P6f=|^^rndpn{&0-s&>Q!pr{x_-)eH4Se
z8Rey$iOmLVk!FskUgLlC3%Xc<UBETem^?HFjs*C#Oj%cJjA6!Dfi)GX0id3MRjt$G
zJWd!H^p^Q<!~%D#BJRlRb|LAjlW}XmT@ReiTe--;-LoKftxMl*%gUS5Y!pX#hfMKm
z1}qr<NP3t$Py_l|;2BoCI}K2pU>?4R*oUKB3kF039^lPM9m8u)$S*GMd2hU;225r+
z9a%Qfq;`V<8bO|*ogi!iDqEdIW(+E1eVGS>=n}KB3k11rg(OlA=DP&$nc}DjHq{B(
zo7e8hW8B0>wb6SzSkOf*$OO-@<!z#8Tk_hTyy*1%uJp($-a?mWTkhn`-=h_S2*rzC
zWxGiMgXt+9#}1Wcz3P<U2+QIHjDkTJgPVX?fqE!}>DXs$dTySUXG%@JU=_GyOa)T~
ze@jcuB4A^6pav?!M0|uNBmjYz-q7+)g!ZFqSg@<7#<W3&V%`cCUnie109o|m`fIS=
z-(QIs91FhZNHZ>NICcBJ(%5-_N5qMEFIC8tqKykk1@Sa$F*D+CLMHW<sdSc=*xUzU
zZ8NAPV-zo=&1_`|Ko}s9GN0$)1c6BQKVG7TZUg_D)V$H|5ClAzkJ^9;8i8~KsTzUq
z;L+(4d`Gm%;~V@Eyp@|sza-`{VIBI!6xdQQzz$F9ZJ7?7XV|5XyO#`Yd&2xJo*JBK
zaNvuM!M>g0(B9sXn~{K<9L2eK62a$N)Q#mF68a7<#;eh$2#ir5i6rs`K>J;Paz-dU
z0{WEAhJ8XkvTAB0cFBYyRx?QwaR&g1P<Rjz=D_?IQ!z!0KtBY@?-Q(QFrNXb$e4kp
z-3~xeD#kuh{Dqk?@uggASz4e>cPS_Le!bCnxjl2y3v-9DSZv%T28=+u5gut}brNtS
z(BlZ*DICPnz_=WL;epEMzX{J_fy4IiDGNeksfQRn>u4ATFpL&~AOVPg22a!c629PU
z0@7RHFM$($7pKV~6}?F$P!KwTMgf{~L_9^23UC(k9}#o-uhb84kF#3O;fO$L3f!Ww
z=Bp?^$9NMzThIaX#Qkb8z6jd_g{#Fp3VL9)#!6rtbOJT-ftVOP?0g1pM$hNp1l)KN
zF%kX&Jg8=1nC9XYdOQMrqE~2CpNHQZT>QGzq6(R93KS=#CK36_*uRWt*faw#u)u)I
z#XY`)?~0Ad&|5qz2bN)}c-gGxLw@xabwSjr*#fpEB!L1I$Wl-PCRIE*QIy0wV2%2v
zLn4pdZzz#}-@af*3^UYWzMe;2H9i0=Cg)@(Bh@2g-pKF+Qt&YQuI6zmIL4Y{)bK*<
zRygp4h%_e#+lyR7K_@8y%5$<Gd9efgsE>wKe?mP_tim)SS|LL!5}ug$W3;6MFKWkP
zBT$;|WAfTSx`0kr0RJ59wFPKX04tnDubaq&V)*X+@2@WiNrT=%Gbr%Sb8>RPBhX#=
zs)jb;(NnEXl7I(`fNN9{0S!0DC<qLVq2rW2JO@3)6FmOW9tIX_LItB1uYU?7@L%hR
zR-UUaKiuOnVaesTZAQ-nADJ)cl>^F}FN5zT*GSKr@1})2zu%!ryZM@@#x__HSe2B{
z%!QA)EDG>SO|D4T5&m+H;YNCBRcC+h!EUy5QGhOkGL>vC=aeo6s$i?(v2fMMTmz6?
zbK~UBwo_O1tJ-!(PU)HzSaS}&Rjc3l<Dn1<3{RU3sUZhYu9!0r9E3I%9ww21O#+7q
z(BMp>w!oUq62tPr5rUSI(kKPm0Peu6$O407wE~obD)J6|Y&V^-KJdsJOItUGmM=R!
zcS^f$^6>C3GuIY*x0LG{-Cc6>wdDggpQhF<%zd@*w@tp>{I;eux}x~bUe>oK^kn$Y
zGec-T<y4br*k|Ii>$m5=5c7ZIdgh)e)Fp=oaFt(g2%Cz~%@VI><db2Ga*~yN6*@xp
z=H1V;yMLaUw1eYUB$V%0`XE}>s{k{D)?p!Mp#cTZjJ!&M9<UhkL3)6S2q}^ya<bZW
z0tqlNzzRd$RfG;<<JHDA4kQLSIc7vFveO3@5)dos+>{1NpB<CBwuY9E>qVGb-@JFe
z=twW|t(|)>x#fQEzD3n73Dw?G^1K&fy55>QrRd#9gQDOyJv8R*ric@r-z>6ToKlF1
zujs7Ngrc`20P+TpoXI7%?>-S0m5-a1D}dON!WY>;k^($Qo2{sV%5`ZTR==)6)e6s)
zFUEY<wiRU2ajW~2mkt6j1w>H>R>0t0KL^tzi2Os9Gf<QYS)%8c__o3XN<%S>m+y1>
ziBJMy4LJUi0Zs)CKWJ8$3*wcxM?c2CooK*wRo~Fw&`HaBuWnaGFX{ey25-I*akuc<
zkLMnyMproRY{G*5fsbzYO^x3^cS=dztp%dh`7kST>dpQMmraWdSKmVJ<gM!c%7L5{
z?Yi|i>bamZwBlBs&60O#XBhe6A|jBAX1oeX6n4z;DaX`(zay;tP&>Dw6({-xc+L8|
z88Cpz+e5qZh#%D8a-7DqPJ$s31S?L8rDP!7Mj_KJvkN#ISfn!hf@-298btVi7)gSy
z6bf~X!ACd?oWXCx0UlrmGv-x{{gU0AJR%HtIko5FOH-FER_>WQrNwh*>5+5#hDjr7
z(fiY)^iNZepI*LgUiIMyYx&{&N7+4-mR+A$oxOZk?vf|o(z~T+q#>`~t8ba+o+$Qs
zGUk5s{+(=nXvMPnz`^oyn$jg=Hy~GtF#U`bp@Tk4Rzg&9Sa(Fg;nrm@ieKlm_4BL6
z=&aAMzaw+3dUpAXg|vH-0((K7W_3IhMG@;AbXbB!ybTcCM20RTR~6p?JN_wR%xc*3
zeCtuBSLdl%?a@^;Bp;@#v$_}^@`&l;L|{{ep}E##RerhOKlNh~kLlXCMO^p$cGA`*
z=bp9Uq|Uat=S~q-(OEtHY3A;>G{CJ%-Z1>E&6yQ`yfY=&f*jlAEK9=7z%$~{vyvV>
zITSYmXUE^zH`lLo))&!PP0!-)@jgdCt?;hH4B@0@8cYq}O%E-rQ>&PSrS6hWV0vGH
zSM#jlW=K^+{f4l~-nAJKeipGO=}6y={n#fLBe}nnoH(>}5ZTeBg3)CV?AKX+AOY}#
z`tm2f33LHgV2Wl(jv2d{;CGnU@yx_=90Vd?Uh@uaXVX^*NM5wt+WGS@?;k7kymrjW
zm>TV_h@9$@S(l)C`@LhLaHdT}b+-5V$BAYANk^6sgtQD|ZH6NE!C=FL2hC6S@mPu$
zH=#!AeJ9+%gQb6XJ@87|o%Dqj2{Usa6p!v+gpmNHFifjKB8{6<RgjA~g?~b7X^X2N
zM2ZYp4x8MlQjbljI+WGbXe1qku}eG@8Yv~r7Z6Njlz=w6a0BV)ah_)e&IG$l_Yqf^
zk%J4_p21CJyjl&tkD4Y{P}+WR{GU(!=oW6S>x>I0*62sEl>PSb$XG?d*4u;q-W%18
zYX)8~iF4W>c=M;R5ewX7oDU5ygK@DZf^kJSTTGPjLck=Em0iW-jjW*7Faowr>KGi^
z6BW?R2Vwt;FZxqo4|(5!{FO4QLWTFWuaN<H4g`rNV*(&jkbmJ;+Q_$p)1lAfd8443
zqk#3oMxxw|JgrnTb|omytL3%?`!Vn77rf(c-`$6i4CBLB=e9KVrI|beH!Pi%k=@jq
z7FqS^+WG|nodvd*3;CK5^YAG8Hck`%gKbIq(4J7(n~WvQLs50pBLnO3-cfr@NN;c6
zvao6i#7NGLBenvgHY$ol!Ej9Af#bfbTNh5|mSJVkH!rMd-G`ixI@ZX8cqgm&be)qs
z29t`!AvL5GU)@@ocoHP41G#7`3c!mLOrm2_Ypou^!_$p|um-dCt9aGeX)IrAOr*d_
zK5AjE0d^f@6U+x*@Lv4?Qa>PM>aV?05`5!|-wkh{qNb`x<F<!a7q@*8EKM3Wz|xA3
zonxts0*X^}vpi>AzWo{}e$C|#W-i43A%ocbZBqMN!N*I40~1y(UtBuovJ>vOi8n)o
zpj{@g**<nGL(#57@}Wt$KBD?&K`DY>R%d;NakSIpiMi*gpvz&a3|I(T{C`XeJOQTz
zECx49lBfw_j2KR@fU>s4L&OMpRDuGB!dJG|!Un@*6pS!g=GZbxf>x6d9pdma{0|Xg
zg3K$<ds++8cJF-?nhSpUj-hR<DLm#wk0-`uno~Z@V)Hl6pK5<clsp30TJpNn&HZyQ
z*a+`Vy7;7I@rc6t!=(2&509=lj(RRGy}ev>DnjbpGSBZJ9MB-R19CxcT5y4M2<@7j
ze2)nSKO6FlE6rlYvXQrfXI^{O|KtG}f6U6Hfz(%@fR6#03V~2K8&Kjhl#oY30QZ_`
z3@ac&K|s<TQ?%747&^DJNvfl1K!ahepi=Rug_W5{4YCie6BytX8F)P}@(=tQOW<bR
zB{}w7)r6(|($y_<<GRlW_g`9kV>BaV3cullf${#WCa$>q&4@Br`uy@oH`5kP@)|gG
zab%pwKNQZ7y;4hSn*F;p8TX4nuDiN5{CM$!OM*c^F^@@*SQI_ODG!*LW^6hbyzf~<
zQo0?fN(_qK<5GZU|C8IO<VYGA0JwlWQ6$<@5ePCpI;%qwwDLXhGY`cen4o*iF%VG3
zD$XjLzN`#nlo<fzS8-{Gyx;*i;R<IyKxL}<0DH{D|6KCdwf|ht5P$J~D>A@qacwT=
z>f9gL(tTrhq<?39ZuiCy8pZ>b9bDf1q=7O7z}CLE=pHiMd+U`~db{}-Tf7d0JZNr6
z?vS3Yd6wDp@a=HF;*_fZw@DtZqh)j`HdaO8;<z`qV$Jf@j0+3PZ{i4p!*LVL7rH0-
z4MkB<(vjqmd22pgLZ^hMa0dfSbQr#cRHhQ4UqLfsX#z{F(glG-Iqev|gy>prPHvf1
zhmZ!$sk^9<^6gIGbQ4=e0}e!78uOoG*_9jOIawaLq3ecsh4^*;F!WJXXWSPdhKFy5
zv~BrvZ}f)CI7V8dJhEJ4%W8gX|7Kmu;*bX}*M<d?#`OwM)^z1Xp3o&n1uVY~2fOW^
z5dY;V_j4Lmv*Rz|;7+**(4%?+Iv-Q`2#h&9_tFGnN_0hE>)Sl!UZa>&h*WQ;o(tu?
zO@jUfTu?@ED#Rdih(rL4yK>wUreroeB^^wX727-_2gMC&gd!X0C85Qj2zmk^PGC}}
zC_|V5bVg5jp6PSyykEvThmU2%78Esn)EYm!w?y-)C>lm|drQK4NB5PmAF-L@W1O{9
zB0Qevc#6%h3UVLxM<uG}?9^NeXz3LjP9YaQsbYAP56UqJ`6OUeFu~v^uAstVEv&)o
zQ@SFn#mInmcj%KPd1c-TGl`^`U=K-&N@^wv5^kt?z7erqwHgy$@Oz27Pn#`+aK+pu
zsSLG^0uPBFL+@(>SJ)Lrv2;Urx-gbTM8|u2-HUhLzPs-MDe|DW@%J`P^(wGcB*cd2
z)(Ae!S9K2W%kKK~wezziZC2Fvpn#H%@ZjsvB_b8L)8JIs(4+1G!eOqJO9MKgIBgZK
z0!8i>=CLC!a%w{A4#q^u@OjnFo0qPN0!Rh{gU&(Z#AH<fA~jMl^lYVyF9a~=_{1Cv
zeEn9Gt*|`?qaE>CX*xVZqc8)MGpXPjqajca8|^~>Vnv`88~6WM{S-3Ilo1bWq`ttG
zvlsoAr<@wV$5rjbzogekynO%3+Cw|e;poL#%5BLXHiTUH9w9a$7I@^<Duuc+5$RYs
z3>JbxI+2`I(-DDp>k@TF^oGx{@mD~}CN=nXmhE62Rx430Q0(&s-4c9r!BAv%S@T$)
zios!1Nbd`J$*8sKMfkpDr$I&n`FVTvB#$uG4-AUNJpRa*Y^*N;LpuZ<lm#e7Q;h(8
zfr%5fWw4~Q$pQsU_R-uHwszXu(%JQP1PlxIlxW+AA)1zt|JxMLze|t&w|2ki)1)-^
z_H3Mg6e)qo`Ilr>i1?o>h$RZOH8IbH&PlysB5#6skBMNG98uMUJchwhA<3ud@s5Zo
zU8%KYKnDvRcr|02gvk8(8V=jHfLF_4SR8^?e*~xk<;o<vsoy>@f&no2?yYy2>{?dC
zW6+X7tf*ESj7*#Z9}Sp=a@Nv`y4V2FF@nhGpD03s+t**Q1de+<;(Ac~?(mZ<k-*t9
zx`ZVe2fuqQ(0BLrw7D-o_H4*qTllr-fy?>e2I8zSI2UB1q~?o)6jw#+^z)*{-$c^v
z#!74#v!wX@o%8jEPKYknuvRn<675ZM2iAVHt%)LfW<i1~Q0Qa824^hEw*@BQc?GmA
z0{b#|k|+UR=slDbrY)PZW2n*?yC_Mdba|C}25%w;67Y|y{$R(}o~jz}FcyRn3y$%C
zhz8%iqx8A_&D8I*KWoPt+|HC!-JCt^T>Tf{oan;(<tF{rOikdP;#Hr3!6|Oa@B$8j
zCAd~}LG8)~a<0WU9TWD<LlV^`7#gwQ=u`~Ov!x*bZct9l{CwdB#4D|e@jT4g-_`2b
zxQ&OYQPL?ih5Jh(LnjwyE4Oky@G5^-*DwhzET|dplV)Fup@A{Yq2sgyWH8bxOWr34
zLIA3lA9eKhZi|Rp*#B0zGI-RpA*s$`%TFuz`1DzSY4+%&fz>X}1fy$oXLNR7+x9|-
zwMeEMNV8;OjllkeXy^q;vk_pqiK4#YR;2$p&8ahaJzl9e=B@XRcjbPZoq~m1#-(fT
zZX0SA$vve-UGQ=w1s;Yqh3}8nA$Cv~h~SdSB358(`e5GUMJ*2%%&_-T9_^AN5E_Zn
zB0M9OCmQipntpz|io*5(wVwFdb62Ea*NpHJOFc>}8k;jCPPFP?9Psy^dFsg7;Fkyg
z?~hHx>5Kf6$LTV{{9_W4BnB(WFed^#C(8ncAiow4R0OkeLJimXZ1Lymd8g+qeVRr#
zEv)L&OsHuDaZK!bRchCt(JdHtMNter3JoiuO)_9|ffoP@A~&9LhusC+@{AW{NMSy5
zObBLjX-MT@NT`I(u;|~%kkSNsS19yP97?hS7=<|4j4aJ1eZb*VX)iL{Ri-g|L&2#Q
z*TC!9-j2wk)2eKNY!O4yMYQIrF^nRv`>hBfItG2Lcl2dqj}fG)vp)zbAk?$s?z{h)
zSmds$1!C2#j)9j56nrzGD$C>10faPvZH_l;y>}K_osxpjLpT7!=0N-ee>js>KYp5v
zix*bn36mD23`p?pQ{ExD!YqhF^h!P6kW{(n?5+s?;I0tE=<bMWOE(s_yJF1!SFi#5
z!#(DnCjs9ESO)dAXW9e2et7tqerRV@m5P0sY6(c&i_xN?YnR5JU(?r@88x-5pw#tC
zw^Zn09r|nBJl=V(zWhf=`&JWs{HDi8M_Lc`Vk7%RgYx&8Ct(OM<g{W$NV{49NC$FO
zfgHINtlvTY9kdLGK!yOhD*~0_pfa%S^jgHs9h{>{Z(MH#GEG4`j4W5G;I4?VbF}t3
z7nm_&+8mDYMjfIbxkS0OoB;f_rhPaosdDf6xFqxaUM#h}(20Rk%hH16BNt*cfj7N9
z%CD~zBdeqxQB|BSzCLM5xho}9d#rsh3x~-spP{jt&mpcg&o6$>wbu@5ln>fn>5C@c
zoG$kmyf@L!Zu;81*>9dp{b;ZnvS)L#8)*^FuI(ZIx7LFv*gWXZ_-qRZs$wGaAWfu_
zfoG>dTO{mY!Vor5ZH7MvAhEe)87BDz7!g!*c<n60R42X9jB%F&vwu!zl2nM81a@d^
zPt%J3#OeeAgV@K`GUd$}^{5=)gc4KDJ@H*f`S<%~dxm`)KJlX|U7Azb7XmyFJQ~v^
zG$>XlMVy^?_flL)Ra#L>6cV{isi9N74kWJ+*Sk^<#Te7x#5~p2?eHR&dOS|j<7n=<
zc@;NP0@I&^&hkZGRti{N|L&6q(&)b6<2M*<-5oxqxIWhfK6ZFMxV!B?;arNe*DJ`Z
zQsEP<3<#*G)Cgw)G6@EVOr!w5UkU1(g(5aRHjk~HR%TEuFr;Q?>}{o&ICw?k7^MFr
z%#S3ufDKmBhT)`?h^T@b&%O(t%arPE5drqZwp{gF#UJ;yZJ&qL3F|jUr`PqpJ>9Y6
zeDym~Cj!5@d&&}cEG@h8+t;RvK_~gUSXzz&=<Uw#HxI?AJwm=p3qFC#izJ+La5}j9
z=Hig`ny%1Ex6Yx5R5+D9<KbY6y!;)({)ev7(3|Z+%Hj3giVH)~tHkXmh}$8TL|_q$
z7L34SJdI&B7?Xp>HA4v)XnZYT|7w-QgT4+-lB(ECR_v|I^M#I8JhocWQ}U6P@jONQ
zIXjpH0W8IPHCSV(QW=CdHjidj-=#RYa-eF9-_n7k2Sd?sZT2|TH^tKUtih$8TYdQY
zo_vp;_mT%rTBG;BJ$p|C4i-=TDzk|BKDn71xI*mlipyQPVS$sS)=0EsIB8(Y4VzDU
zPs##4GEtO+Zes|7aWt-lq1Wl_kq_RG>~n6E8K2y|e!+Io^;8CX{`PG{f3s*ABoBZw
zS{a5s3t~oaK89Ds2(qsVb(@8gEi79^9#)L7YRwU_@)BZUFeB2Wu7NZIZUoD$Ii&BI
zAlOC5&T>Xu2PK97R6WQ<iFtbboR5R~29vnYw9kGVHQEMVuoslKZ#$QATLx!KaptF*
zz^ctYl|Q~FG4qq}vc#V?H=~bu%__A#NjS~CRA+YupMVa2)^NvTeNx3Z>=;-6W%7JC
zi&DRN;&P+KlsMTl?$l5~@qzN6FeTFU=6rwqo{;0W1X9%@DSch%YRf=k>ata&{xU!;
z9*IS`gsx#Y6?!N$%wQk}NCZ^_BnEeb3a1NlDzr^TEjCsl!%ok`zko0pc`j|5BZe$3
zau2_k9h&C1&nV3|{)7a|e<;Uo*PpN4YIJpOD_(Wsg2t~iGpg#8Z>HJRl^GS68<lzZ
zdZM9T_toV0El&<3@y+@DzD2F-5Wg)`538|NLaaJ^bNBGhuz*uD_C;I^dpTr-^S&VW
zh-{Duor9^>JWc7CbHms@@$|go%TpCqEqx8x91r%y^TCym{^}qwm;@V+=d_#(+56;b
z7-{EWoeJt*_Hzk~?gC@1o-Q%F7JevBM-WNn;RKAJ@uoT4Mw)~+_K6BX(XDB$&GU&5
z(m#e{zkl4oB@As?eq_M6y2rZ2fIP2EmoV^pOTw(iV=-Kx!O*{JhIX$HSx#wn`6!2c
zk1eqhUT74=8EZI49q=g^Ep--J#6D3J`*-~TN6j#<v*A<+H_VVT#Vx&Z<>IF_i3<_{
zd{`y`W>u^dXFyrgLCA8UAF0UGr5KjU?h<qxPFctM=~;z&>`Fm_3^8I3#xi(5JD#&}
zV;s8CleVm35f6LKDjLizw7Nzz7RFtg=nPq%6nw5fqr~?@U*LnmM3_>9?*y9I7W8$J
z&lZT#|0*Khh=m~%#$@_u0~sqS`^SmOit+=w&L}?&{D6yFzu&%3LM0LZPD5_>5Wz(F
zLnic7Lb}9nM3D_LRl_YwS@|(%uQNf~iT(;mAIqEmNz2AFaPy6V3N<7>Bv&g;Q_MES
zT6H*-D)kb*PwDgJmBmHhPFm*M-IY0gZ<Od=^@?Tq>bA)1Ni#NQ55h;Z<`7O1Jv1R5
zC!gih8~>#!WW3|8V;%J&M}}P~3(A)@D~}A1?C=ox@ieAYhF-Zc%!*9lF=pg~;I-4_
z2~$Ia1W6C|Dz~Oq6WutAyJ_@{dF_zd6t&J^>RDhoHt9XJx`y+fJHwh&cABzDBCEAr
zOw#Q%9V}_qjC@Lf*e8LU0+vzz!+&c}^4V>{Qxbx`W7*a1buU8Q+*-#|QW{W(lp2QV
zc_=G-*RKU|;#V!^wCO_46?Se9oph)zaDxCdIj?LAjE3OgmOi-h!>4cqta=dx(L~WI
zLO}o^cOgeFM}*2$L!E%bkY58R#X15tsZu%;i&=I|Jt0X5bP7_4Mu&t%x6>x$7`+bR
zSFL~*xzJ*f_m1i*MU5cJ=UHp~ftJZzZ(nHMet6aQQ8|X|D@Qk%99!pgr14p%G_)Jp
z%sjvXy#ZE^8>hc?i2mYaL!)3Mpr9KAJT;|*732Ct)(RSB#O25PBJ|%w00CIVm5d7_
znK;TfbGZ3XeuN?61_sfI=mHFv8&)m-f=i5uwbMkT#bEy5q>v=cNkFP|&>I){u8ucp
zYV{KtJfxLKiKX+<E|IDNjYi%KUA$%Ft$8PkX<+3U)-*AC^Jvhuxi1e&iir}UL?Xds
zJu2dC5#rYWQM8FjdUw-_i(Rj=-ERFlIjrpQKFzHKW&PMDczG{D5c1L=r(94GZwVMP
zKQEYV1%)8<_zqOb15z5yfI-3MxaBoZG-Eo!3h>l$%XN87an`&|uZWePFtu5jqwO<{
za^G&-E0`ocJ4i1>br$F{h;3WzHr1^QI6TW0TzzP+DSH2YMJz=#_C{y!XM9Si3(;Q;
zdn`E2L?`8=1OKydI)Rt(K#0e#UHWEpcKJ~H0>jnq;U`1yW@3z=@_8fWaCwe=({@`>
z$yI46h`|xe6z6UXgrfpNK^b9^6@MW{SOpU)!H2QLgS{G!f_9o=Vm?hsjV~Hs-KfRl
zOG%Of;Pc}2pOq>f?!}fd)`r_(Lmy!@kmLjF<+TDql=HL1U17q+3nva5o(-IRx;1)N
zX3d4gy05jEEeDEa^Fsp*`}1D7`+`L)FfH;msizM$NiL&hd8)0Wx1rWqAdx|V@>{xj
z4!TX0hPHsNCiu{IcaiVY5dJr`>9k{Y->_OHV6bdK9wD-#Cg$@h;|!@r-I#^l5LDby
zejyzqqhg||kd3eDZ=8d`+8{{JC4>#l@D3i%JIhTI*NyDOs<OhQj-edYS&RmGoxqsH
zOi<_ltEh0)6SRl>u4BYo=-s!_si1ult2uBPRBGgv;P*|)qLM%p^cGz83kcvh$WV?U
zpNG_dXmTsErV%)AoPmM7CN4-34+s^<Op=kQiy+l2F!3m$U0L^QwL;eUS;q^bUWm(b
zQR~WM`y#43Fw%>tycL&b(Dm-YkX}<-_s-ma!EM<l1Tp9(CT!G<&d7xzdtS!;0tb@0
zd!G#?uWcPnoqtsduUuY$Xm<lDLKMb-({`>2gC0CXDh)-H=##8FiUgS4Y=|!c!q~tL
zTl2$;7$c9zfrDejh-X)Ti2$)&@Phyi%Y1P%bUazt??uo@8T|d_<0UD!+x6kq@3#ys
zbk_bQ>vBNx%=;tzLi~aPBonkK5RU<Sw%~S9+^zl6WImc@YWiTSCpZ^Qy;(5f!LTSk
zPSIrC?xsKkd-VHPx7>gaKr{tQ<5Tw>2wW2B+(py_Hu&z)cbr}<iZmdcj%b|4f)3>1
zLC_NbyEP9O9@m`7*3mDHp7@U{U`fj@E!f+aWGS4oH2Blc#z%=s1NU6h*ER3gip~3x
z+m+{&1|JKUcM$IWXCe6oXHwTt)H~<-k4v#_E+5*(o96l>?Y-mv5D+J@Z+`#k)@}(e
zqx=mwZSbIoX8;>eB;X3nA$zYNk)#zxNG6vMWcno;SC1qY-t7TWU8Wb0sW5B9&alLl
zk==IhRil{z27Or?clIa$Efs}Zf=?uuFJwvmt8FzO&Ru=!L}1@NSJd^hVl8SR0TzeY
z$qoJM2Rg`rI!f20vN|xk7t4hzN>bNz?VWG2k)4aKNj$ulvMcHDn<OavI^dO|9S3S#
zMSPG$0I6Zv43Y=-VOD9NLMj@{4GF{e5ncl-kcFi`YOP)TgI-rh`VcSI7cCL2EA7^M
zx9{%aK^h_W_=~rT?;yEhX@M(5a50Rod=e`6`^A!aPQ>P5)*QH%PC9nwe-%ABiZJs$
z1r^~o2%FBULMsw#Qx;ZXhp>*q`>;9fJ3_4_+{#MQ0Sd4S8GoRF3OGzfe=*SvvnIkF
zCu=HCQ3C`M&m*dNyhcV9AR^o`UbRLDYMz1JMfG~4h1bPn9a|n_wvKORU=Dh~3Svu2
zW9^?O-HI^|KHgruE36qx*~MeIyVB>A8cX}&G?DQpu*Cm{hj1qQ70AkhYXWyjtW%w@
z&Wb+R`FNagZHl+=ljR3W&y@Qm-$xpu=(Y1v+qqg1qkifZazF%`Ea)Z3B3g4q0p1X1
zp0TwH$Oo}fk*tuU9W6o?0tZ+yYU>Q@8ugk)y-J@C?_45$&@>%Z*>?2lM&%JA2=BS^
zN>z!+r8e7F>4yu>-eyT-nhedmaGHxIekjf7c$4!2^#KNbtbhjor4UqjjUA4=9r_Y^
z8(4z=jboTGGHm(41ZdNG`Qmw+KZ6oMU#WKj@F+^;v<ZL%Uogan;Y2nx7Q71pLLs6G
zXF_e2X!rw0N_4(~n8&H1BOHVT9^lFL3EqKl9;EpcfU%^Om+blG-qV^=ml`u-+Qi(B
zL2T$!L^bsoIfFz0wYd2N?}s3uXBb91>wtXB*m64k#WWK9AAUa-f{)(At49?<YilP-
z!SF2}V^5Rauu3GG4?G|<jg$ceZb(rh7n(VEBc#UcFbEv-R|V1+=(q^gsFoR1QZOV#
zzj$Xm*?eegp49<rJ4lYH0lZGcUH828zw}zg)I7eT0ie1LZvtT)K`=Rdk-(7C0Jis@
z_m8%75^iIRskH(e;MX8?3f>K%MCvH6@F{Cyn*NYP!p0K>b`vFa9?w)rypqs3Rm2k#
zXr^L$xR4Csoqp%}C4V9_f?t^#uZ8;|HwK?q;L$;Og)@GY`4^f6-K7F>pgk#80KY|R
zf|F^HBV9V+EAnpzewVU!7KfnEga4?PbLfdX;8x1Mz1#wD3t^^iL)k*-=+NZRY(*>^
z|AED>gYQUhGEr4Mvtn`<7hp~czK&$o9ofJ3*IwfXIsZ;>3X3Mm46G1v7lnk*3iT+^
z7toXlf`YI)=GHj80iA~5MSXtqQQcYrPh&bOA2O%_i-0v?h~4$^$suCFAPH2!+q{hP
zqeY^Xczh*($DiMNEkax2{~UPlwhps#KNfmDNAoYt0aunNmrtQ)k)g{fxpUM6^Z?qT
z0t$%u1MEpFu?B$0HKRmxNfCFeYj|xex`MiSq>|8A3DA<m$e_5Ma|Au$E^f}ah%ewB
zv>@QP85NW86)lJ&RU)XM7(4pgZGXqax0UYk^dWOI7}4sK7zTnspc-hD<ew1HA(T?V
zG8Kd)1lSZp^$eZGT>&-@qV?qB`*X@TxvCjMCa@cI6czCgjG<D4B}94v)#;BL1^8@&
zwig_bM_?S<!rddNUIf+A9Oy!6RIt17PEn{q6yO7WM;wJQfgz)tirg+Sp8So0!)aY0
zqB{bl4&3i9$a)tvU2R8y#@fH<x1iI|B8i&{B0_Dt3-{~jO5g;n=3^?s{R3vm+;2e#
z@iuom7@}IovX7A_BX<;hiOO_-F<=19P^+l-FPa%W1#PI)dI<=8exa-L5oS<>YUICz
z51ENNi{Lonf>4A|3L2y9v(v!ZOrl@F4g%f6H-YhzsAdL1zGGsAc#WtaS)o>W)WE|E
zkyPB1*B!qi7Ho>9;XEJ@A8!S%@fiNr(gzhUp<n~&=+9bId27~3_p`^o+7WO|P!_)L
zZ)Xt!%LnEehLX>xt{F}Kpuse()tRhU#fP>F#SL42GH|c9+b))!4n-r1vXOU(x2L%9
zE?J(SW-!3U4@lCXmr#NICur2vU1OR~?gjEUsM{DwV?+cMK+lL0fuj(qB{4?LB<Wb6
z{zvSLeTiDZ(*y`YEno=nD`#o%`s|Cf-9RXwqY_ea$fLBybJiWbZ}U+4qDt(eBF*+%
z)!AUHV<CPoM%43~dDS;kLZ_CjD*r#qh@(yZ%P+86$)qW*$aC$Ax)$Nb=B%IpUgW*N
zD}`TzGU91Xr_5H=Rki#I(&Fwfz$OZ8KZpg|8j1upPt=0n7THXa#=GI9H4-B3;<R#w
zE0*U?z5)eSk<Ag-XZZL}dxk~&29*c}jw>pt?u86ruNNm+I<aE=J;@~BP8!pDXScc1
zPsLS#I<>1lW397$jWW!8Rz@A?`LWcwt&qc>6d$B}$8~+sX7PW&rgCKabe***AItwW
zfoq$;Fg)DNJ|5|+&%N0C7@JF@v%X`WWSlwirXdfgrST>TG`v8ZB1s}@xHYoEr6nlA
zVsb|mI8B38GFEA7c`!7OAEu!)LMcruf(^(bTH9_~6?eaS1P4LJ5e1+KL_x<vld_1Z
zH)oZ0IPdRVSn>Y)dDpiLq8c3NrS&0M_a_~?s58A8S-Gq~qo(X<wHUnGR?+wq^9-w<
zch6PG6@3y{G~nm?<G8p9=Qe!MpEP;Y>!RHeZ%`=B&O5eEIQ6S}&c^wZ17kPMFTW{)
zKh&{*UEo*om(zFyB{-5a$VHK(AB`7*a40s=mncc}l|kj{?-Pc_>Tpi6H<`c#MXXE@
z%t+Kx0s8OxGq#&q>=!nb+W>gbQ&2x(Yl8I9;>_O4-+kNg(Kj!DZg=mH!%QI+<7VcR
zAF3-dVX3KQxIS?65WBiB@?LV2&NK587CZpX*pRB?%*)zITX${xaONq?G%;R*!^qjg
z-Jx-3+qUJ~Zs&cO8dZ67W?&k$ux|FH|Lz=0`MFFd2<ST?L2?w-jKE+VgR(h0=)~g(
z^cqDckh=su3rczGb*vFQfS>EY_gDrZLnaIGSuIbsnQBHBKj*|d)!)*?xJ8;&Wx@ri
zL`i(SGqn77dy!?cbX7vFx8?%fXNG4Dxt7MsH)r^?{b|y;1{}grKVfqWxY6Pz8lL_A
zlVGeiyP2~0yQ!v(DCDnT<Cvd4l8=mgjMoXa$(8r~zcY5%7yiR>XVb!|<5DJEY3~7-
zDQb_p!6fILC$2={nNA-oF?ubj2N44xXnm1@m1Gb>bG`HfqDR_Q(@kS_4$SERa)(6c
zM{<I@5iFn_1LmcGr1J0Ezh1;M6aip>Gt|@=SX>hOu}OsG>VL!qu07s_QTr(g_c1S(
zjXg0oBWxv29+!$elD|%m@L%4mEl_sUv81SJRYf@XY569sr`5y{W*NO23Xk>u6YZ(Y
z+Kq=!pPyz<m@&r^kdWC2|AiBR%>Q<?>$Co!%jl#50X^44amKQuUl|uv!oy7XWjcVq
z(k?gb2Ze8}$)v?Dk&Gu9gv1#N-Q#aKT{^5W#hp5>Bc2kk-uewh<`CQ<s=+K;+k?Kx
zwg>Unzef_9P&|_DKr0+KM4qTB6@*6-^3|@Rm}N#(Zu!Eib*!e+g?T6M^<bs=puar{
zIUn(t(=_;lh#}7-2Y-QSHIL2;xW;G-ZUr^WF*4I~xJbm8$ZHODP@+?zfK!ZmF!&ss
z=ETy{v4X(yXe1{asW~w^%b3?Ofd7v@-;0d24pUUe(_jfI7}hN{*8-DcZXL;tavE^Z
z1hyn8?Ql(9-i#kn6tO^$Y^g}eB{EFmBJH<wSwT{0{I~wiiYqUK|J)XGu@zZclB)LH
zUqcrDaAIg3T9)L;NpgTHc`F{eL?4O6p!hkrmc%^+R=cYeyd>ru3ds`&9=IhI93hV3
z0fi^<1RPfp4A+fG;{U+hi#Amq%CuR?g_-(?*ETZGg7_c(vzbPD+1NRx-4G2KPE?d1
z{zGL?b#@va21Eyug567(Xq!+V7sP6D{40wmfDRwX_J7~7jp=Ueh%Vmn!R_il1&Gl6
zigSNmuy6nn(U5?qT;Lj!Bm)+cj7$9G$=}Ndnh2#J1Y`-Kz;Vl0#Mc_0iUJ}O7K8N|
zG$J57LkPfMxsYp!CH5VCh!_7ChKQQd8E<8)`8PjoNJR9mQBS98V5fwcS1U{6YzGhg
zM_f|=j<D+&HZD*eOo)>>OoIkyO9PHOdm;jsr6k`Tewu2=sej>9x)v+bDUA*Vz2)T#
zn1A?!MKsNTgDjH1t@b){AB&e(=Vp&fFZ!fy%g`>c+e+P~|L*y&`t37>Hk|bY7-g1|
zm;kW=@AEMCfD`0}Z-M`yf}q?)LaDNwQjm>X#z#8ePgoL9Lw<3h(}X?2h~gLyV0Mn7
zYLJjugPD??KhJW}xY9ObTglA5PrWW4K67sJtvLHr&to4V`#7^W^T*`L_eXZ04?ggh
zyuINS-)%}PUv|AT8>gG{9?HRhr31FpJEAI=9mqISe*euym4};0JyV)DETA4k;F*B`
z+<F(eoh)C37Q^^ktS{}}J@<rt?3#pGf$5e1Ir;sXZn#Je-Mjz&&-dPXF9($%a9G4G
zOCex|hCo5^k5|Je65L=!Y8`qLWOi5uc(S&e(`H+SM>)k+#W64?=ICl0QCZ8yvgul_
zfXdjy2-`_zA=hej0o-JL$GS((59^axlnw7H32W_pXJPZrb?aGldeg0So<CgZ+V|l_
z=Us1&(qVpbt2TA+y^P}hR<?10{zm#7+uF*WO$#f>)rYKY9?4t?Zw1)E@PS4iI-Y^X
ztWM42jo1%guT8{P7T=s@zC7Uno#*z*dxb+g&d*GKHFd$=2`hKiV0I&(%U$%36N8sr
zpfG?3QAIpL?WpN5uAu9njf{4J8x=&th+tvfg0iERaf2UkTwN}n2H1IUzD39@gcXw_
z2tg1gph&sCJSpl|VjtF=9o#g3ve$L{BU{VhKb8L5M0WM<@ah*UE`?NoSC<mze4IL8
zKl$cj+k^Z4nY`I~cV|fTxPf_hmETTYHq*8_&xKWskw3GpG~@s{-2(uVDBEzn4AKXM
zi%ughaS_#KmQKyAsg1Z-_&rzV%8D5OF~V_X-FR{1@mq)X&Xxp17OGPr^Tq1OhC|Fr
zM)4K*1Taw@mgF&=NjJ?kFh^_NGM=q0%rYjVN*EPt;Ru=x7;xrMZU>K*@M0eD1Pvs5
z&nnU*goUsv1UdOo)}qPdF246^Rbbe@h=7~=c`vt%%7NqGMbw8J-|Ex!)71Fg5#_5l
z?l!{L3%<8vidMJL+}E9=`k>%Ow796;t081_-e8hr-tp|>(Jc@tyiA=f!-~#AbkXxf
z0$VqE*$kjV3P^}yO@J*&Og?_|P*S%<#Fa<Sl1LBWTaHNTRs=JHa=1Ye0`i0CTG02^
z8a`RFL12p0_zK`0P<q&JnfY?85YMW?UwI(7UO!^7-NE$!r`S68<;G*hqq`Pe_v?RW
z_RjOuL!04CD~;Wwnc)H0B5rcm^-EK|u4yJ*jwl;t=JrACD}B4@6DQ^be;IsgbryPy
zyQqBupPE<To9cahK&0ki42TmCk8YhAHu$57G_Ey=JQQdvflmF%=Rr7tCPF>abRRk>
zK~QlEMPOcIv1-a-IBgN+8)~g~ngY>RaV%^s*9RxZBG6nbQ+9B)nG&X2>(na7z0Awi
zpPNCv5580N*4w1^=?8ppI=e*u#(<WvhZ$T#tXz{m_&FG0TB+ajL!(}@!%D!p7i47P
zDH-dsl=0vp&`!R_6$+0Z#~`H+%!q$HZzxo|r(1Dfz&o&hTM`YNUI*<0@0tRQDw8yw
z9{sJ<(Ixx>3H>DZNA55h$mfY^P}IZ-^a^*$XIS~|aN{{N>Th-|^deb6N8QMkMVqiI
zp*}3Z^T#MwN`X+r54ZsY_)c>&tXIOT8ONhrH9B%RL<)2Ru?(mA+K>bU`N(H0vn`m&
zKqr1F?l}Skx=ABFe}Vpj{C<OoN{VhZj}FPb$5%L~1~tYV+}ezDd;T8vk^n>s7?QYk
z7gRS1jM^;lGINXIHxW2RZ%DxLX+oYl#DJG=x+sd+&q9E2I2~=GuKYC5Gt4^B79H{w
zD7390>kB+he-$}0Ditmu7x5xDmKi%Ds7M8);1={P5sivNzzs~tA!zaPtKZEGd&vZZ
z!S4p+qP|F-wIDKqLh6=7$B|=((S+Pv`6g1BQBi&hClllqSsgL7W(r88x_Ai1qEmKI
z(z$zXQ0}Dr2YkY5W&}bL-T4*CA`$L+Y2trrhu&|ruS4&~2p*fyt#SQmLTnS&XaGT}
z$+#qk4un#p^XRF28iSc}6N3S26AJ-k&gE#T4+t1z)jHgK4ZMT1&>qho(?SQz3(MzG
zM@=8@LYKgqySKO{^8~)*w!3g!s8zX0NJ=s|Ft5tWSjzE^D*I0+1(BFW@93fce-vCw
zh2ylKx||_SXoj?@;Jt}p7L46dM0X!d(*QX+5ZDX>uNJ{P*(QM(tgD^2ujKhkfoeY|
zG+m;g$8~C%bLGB*Es#f|rC;z4{A~QS!(dW49WFO^v(ByQ#<`u6BvF@%ZSg|H(VXZk
zAXJn|5-L)G=8~JooCGvhr6ktL-Zjw715{K%yK)6|8?~Fo@ilNXW%d<+QDtBWsv?Q_
z!22+IdR}mPaBO5*Eg%)?4)HT%HB0EzDiZ(kj6aJvurftO|HVf(`5YSjNPPo}(O2*T
zcv!$aCgES63ir{BeD)6Nh{{SzL8#-wa8QO6UPY(YF=0&^^+_OAffL}k7dsH#LZ^4}
z`Kq&HWR68e$UDX^Z=Uf-%c$p^C?XqcJ(@@7^zFNiJzcvZj;lZ}8Bllc<*M{S{Go&I
z##p0f%(;s0634pd9T(xa2m+LtyW?OP#zYYY0)JpG@r{@irJuLrQ_837CikGG5dex1
zKdg<O?qGFVOun$JM)0$nL3Y%*p6)4GHKaOg!7~PN@;Kk`!GK<=e*g3*&Z)zfG(-e4
zLXMZjl^(3ii@=f38bR>Z*o^lJlDW?WBrzN9cvgm;UhFOs6IeX)gj%bac`-RLY9clq
z!9ApBQgR1TV1<P#=1iQyGq5*LKoJoLDDa}pv^*b48XLEW8}xMoKIPd5o}X5jQ-%pA
zEdCG;EPf=*cf#$_cd7?fs&n<9cT|qPGkeeZS+;idY->QaUAyUnS@wa~xgIC0PqIe_
z1TD^;e1C9H<h{~!dm{q2#$>_G+fX@TTT=G$t%a0loz&7B;H^pO7#`ZRsETXt%QSam
zx>65hJJEyZTt`<_APwL<$f<#!di>fU`{*Y;<Q3F}NYL*N#n@PVok+LvO$170nPws+
zK?RIJ2n@<rx~DZ0hY`-P_M%QSa`ZaM00&rW&?)p{t$B<!AkF#oZD@Lh?#7nLwxFTS
z`RV52E?dX_Z~eXEKDy6)B%jVKse0vbYEm~di#?u;Ypc6v1&!K=b_JLH^YtQI1ooev
zu(^E@2i_X6twxA{N$Rbr6U$QrEq$Zf^$V~SIr3cvJ$i#8%Eb6ZD5?J{lkAQ+Chpui
z@aXl8)u$z7gi%%e1{NSLl&u3)8Vx@o^i;_6279R03@LzggK%MDMpmfiSq-lni|Dj_
zewUtr=!r3nFR>rrh+B)a)cDRVJ{a~s%HOGutcu_1v$bwFV>^BC2H$P(&BQ+FPi{$`
zVSh2U-`I6Nby81~@9x7Z!zLa6;pks#o|H(|>ZSF~N&P47?P=IMzy1mKhO1ufnpfR5
zz}}gALLy8299HzIj=JU+Is8#Wo#fBTzjo+Z(b_byqpY)NV$H1tX!-|r@`<Mz@;&tM
zCJ;<DzXYER_mxl=A~;<Tm~o8GpjG6!BxY4KgI^l)5RXwT{;_;M@@Tv5tNn9QX_!4{
z+w{~AuQy(Q9cRZ^-I^5=lT@)>V(C$<?blxk4DKGp-n;1uC*QCd-}_T_a@Ed5*XFLa
zt!?j1jXp9`a_-$P1}m0-6LLIz>C8vwWi!W5=}Jl7T<l8f_gmhOWKLLIo)4}08?bIx
z1_w&As4+le(&SDI>t8Xn4s#liNnQ?<@*wLV^b08UyudmkyM#0y87?a@fTTx!aO3j|
z1qZ;6Wt<fo+94@B-^gQcG*}iM8`8963X1E?Y4Og)<WUD_>FszWCgnp!Y}SpYM;@QH
zI<R|R#63F(d`PH1HDg<G!QP)3m)AqYZN2$=LaP0)r|+GJRbdrH$<Y^`4W;SWw^$1I
z^PZWpqPH*2*gCv(p`l;5&)7GLr8P`J|74nB!WY+Uok;W&Y5rHjm8R4yY=-X&`h^PR
zdSXXAKp{(~(ZYD%NGc6O?i`2#EHjuGScnu9IQT8UFp;Ar_yQBl#Iw*%hVi}@ay0@h
z7CS=aOqf_u+A>@G@r{?(4vUGCw^p3{Y3g?^-AAfFJN_uU8(a6h@26v$kO#|szmBNF
z$=F@_5mlI3^9^i_49G^#wgfGMvqlwnFqFYGaST{R*0S#2VrG!~j8o~=b5B@0{RM1W
zT-tK(M$(yX+&2UyoILa_8_^;3nNEci=qcG;Fe@n>GY2~_fTPI*omR_YDq-s=h4T@@
zPDtOCRN!i}0y8#}JpBV5B}#cXMCao{Xs9vck7(obK5CCby2?HtF7I@vS?Ir!ULqI$
zX`cRa0`U>VnY0K^S&K|X_uW*{p-rsfdpYTd)Sc9Sq9kRbUki)G?h6~Mmex!7$zh`d
zJqteSt_Aj)NaPX8&Z*#CB5Ms*FgeCqpjhmbg(v&Ac&rZp_y}QQ9eEc`kTg+UBY=1q
zWI{rLih%4G#Ciby3c;$6JN{&aF+pZX7P)%Dk;Xw4cf1A!m5X9(mqdvt48CC+W<F$S
zh^G*CM%>6*hsD~0yDB&f^;qOR<&krPZ$<?iZpb~@C@PH?u5JmgvYbODN$G(TmIp7P
zmb(Z<9&ED%z!+2q&zrEW7&je+8FZ8i_7BElIk+NrE_N&gNdgX_Rr05~Ko}uMGKpB*
z13yfy<E^WYA9RZri@0~RdYsCX?fLP!I7CfHgnh=|hD%d#>|*-QUtf07)xYxtpTT@h
z;4xd%zVPu)ShtAHMF=h)Z>P>cc*wQ;rI0@iE%C9l?X6H)h}<iEy_=b#b#M5I<(lNl
zqXKm|`rk7LLNT6~pAkj;fnost!Ce9bgFF`!NfBcSpb-UOaa=qy=&;}rM572M{8Ef5
z;pzz194!qR&>^IDi^$y1AuZ{j`YCn|da-42c!#n3TYbkXffb!UtUEPuJbqsU^3H4r
z?_BX4%WQ|z=U*>9(*1C=(c*M@`C<uj(uU5#bZomsZx&_T5`)81xpg1zk~1s7U%k0A
zqzZclTxtJ#dWc`UPQY5qmb&KcAQ3as8u-|#cqWQ^5qY{IXoj(c3=p7XMzRCcxa6Ka
z#?(C^V2b9Zi1G3ufzZShdTiDP?g5DMFar__^r)C2FY;*f!qwPTNUDpN>K*2~vnkx~
zP_XB!zeG&hS~2=)@@M^6Ueo?~Z}gkn=Dx7z0_)58Ah^@tbMZqu2e&xvPUPyqJ%_{E
zzm^d%CX&>_QANL)8lUKjouEXY2+B?2{>uhRE86ecf*YrLHGI^9UHg5c-nR%&b3zqX
zj*FM>Q1-DEH!zYaBk@CvDzK!0cHn%8_(5(Bf9Z=NHxFtnKO0|>bFQVSY6zSKgqA>R
znhT-cB(ki?A|F|EOgDS^4nN(rsM<2l$Me`FZ1CNG%G}hOb>8Fhi1&$C9{O)-Rvvuu
zvmIvajHFOhY`roRy!zBy=Wu4RuM1&hy~IR^0Inq{2}Bgk4&uc?e-W71#Pz|$iBfj7
zfA6K)S($)C?s^gOsrLT=RDB0P%>V!Ym5XzHa&Z}bD%a8=E1{*kqj8n?AgfeTWTc`p
z(ojajHHb>;+DTGGR(2sxE0Gixg(N;1pa0|aK7OC?|K(iodym)q^?L5{8s0!1nTzDT
z(yMevCho^56rcgd01OmdL33#sPz_j6Jb+&s;`5{nEIp7hA!-b}BEd{z6Y>+Hl(;wt
z8H9N%wpv`t0~3|T$zoA&gjMg-7rJCOq;>O0SS{A|J!E?9_Qtqr+54qs%XH+%Cg;DB
zJAfQxkz?>&1#8oKTkjI@qy5`Tydwmn+df7hN*ayBW}$9cI09i%OQcAKaq-QR7{jN8
zlEa)#8b9e{2->R{6KtdyhKo|*$J;B|U$nL91YuwzGS`#CA*doqhD3zxW{xAuUc+Bx
z@*y%cxeKpBR1ydvP7VyhRN>fhsdytcJ_2w^b|5S)QVpw7BD_|}T)ba+@fgG3@R-e3
zY1yoW^6?|H6yiLwfA9VF-cTB#|K0&<fl^y911x>zdD>fNe`yq62U{r|4N)@Zz}EOb
zHpF>$7$%9Jk7@TuHE3EzXA9EaY@m(U+u0j1S22FtHQXU#wfwRClW~@Ad<bTA%L1(;
z_(OH#_!XW;D2M7pChjK&2>Bini7zg)KoZ~)5Du|nBpFGpOmOkx^&ssq5yXwen@b75
zSKg08y$R7-Al@jth7#9M@!BZt{;DB0YD6i`HczN{0QCVvv`jpTNGuIi{4T^xktNWT
z2<uosKjC($ZpPT$YZ6xbviEj(Y?v!oOD<XIJ=$46<4TyOg)M$ExBxjDpdr~Bass`=
z|KIqcjx))L%vXSsjPDZM$knj5<>FQr98wcJ{kuS*?8S<S)Vk{;DZQkjV2zSFB=HhO
z`{(WAHs2VD&aPGM5=WaMFfq{?mW+6sHo8LUU~wU*IKS(n{$!HxyoY)xEVR9i<d<Uo
z{4gxnc0bx<?~~fifI3ZMuJ=kE?%;Af2`LUB0~eXUJ3(+v><n~-M9+vsM*-+*bZHI~
zW45pllU5SC&xJ-M?l0Cv;lSL$O_>x-BQYI=+UA2Ry`&^}vFA2JmmP{5?gxw#x>T|<
zZ~D!3&b}XPl<>fHjAGq7*O!XpPOOTRsmG?^BA^(xNYbP@c)tX&gJ{^F!UnP_jUHCG
z4qHb(^F|yz2V)PU<|cIdV%)9iM`BB`@1=V`kW6u<(K&dKbd@y!x6=SWo=bR<i|&yb
zxCqIih-eJSh%k*p3GQAEKKAMmue^bGBT7dWqyQYy7|48_K^~XvvlBxb*i!O#gw@uC
zkM6!+E%CxozRu~K1|6~Z_JHC%cLgim^12Q2>JLsADOy$3_aPrj6bSkxhBSpTbdvo3
zQ!)WwTK-lY#Ruwj`ttFGA;1|;gdJZ--q6ekT)CJ;^JwQ2`-~hs5e6ZE+6ZD;R4=Kj
z>9Ew;sD~8FabQcvut$U`F--R3M1=*xkAwus7I6vuK$~!kq7lQe#qfX5@2>5$cpsv_
zWbUKNy8(rPF&-smO)T226&`aQl^z>6GXJ6LYqbqB;a4ITYX=x`;YSm|py1_UVw{m*
z{}r62)3kRs?SC??E<8Fe(_u<!bVT*@k2}bMKG<YE6St0-A9!LOKZV2_sDX}RuZIlg
z0CGs7x<_8X!7I4b`?w><*C)Z1jg*4(z>$=2l)P0ah60Bkh1rz$jntD8SpAQP2dvK1
z<0d&J%I5$Qg>l|m(d#<`bRNB;Axo&EFnx6CCxtp$EKblcb<veWi**C^uaO7=hk6rC
zAXLRq;F1de2w!pxwN-!_Ag7OQ)|X7JYkw1<*%^H{wQ~^08+N_>LyTcLDPe?p#09ZS
z9FUHZ-$<Af(TNk{H{pmgwibN^`ATU`%qXV8gP}=+&skEkJZ=~ekHQb5X)-JZxPSKE
z(0NoDiIhpL&*@PyuUzj>``k>M?c5}B1#2x_+gO}?_xi5!lVfk{G|w~3La)Rv01ZKO
zLH)rnMwS2DEf77ICRL(NWiZJ4t5Ikilk0$a%=5(LM;h_F{MT>>&cMGUtVT359)U7e
z(Z9y!+J`6Oq%D_5w^imRWaem^a3pA0eih<PJXVfI2l=UDJEW_@*bb@v&35|2^g1)<
zzQmi}hR#jJGz>@*1?sRB;y8obm9Et<PQ6<n!pS6IvnZ+=1VK@B7A-RjAmvb}#9x{;
zt!preaGC&|GzyIH01_fh`v9zRih=OR@4tjO34M{<2<pXDhaZw74wd5%G`dNcs>q4W
zE<g`~mI95F=HxsnSgsd~FH%fYk#Ua6#TY}l3J^3LC?awLJB);qk^Dk)#5j=TlA%0e
zrA{M_XW~XAR!Y$~>H&0%dRA)gn>TftM%rN@IS^JR04y#XN`_@pKS-<5XcPuogGcpX
z_(IJ(!c(Tu3-jZTo<c7rIJi?nD%dcAP0~Gu<lX;&kvfU`!ibEd7=R|jv^c}mgmb{7
z?j`Fm0oo!B;E-+suM1QA5LSUkAVrGIgAOa|C2n{aAWRBmC@3wAOOC|`8ZObvuP_)h
zTr^yIZUWKqPB?+{Ebx#NyhU4cCn*jmhDK+jwy-P=NmN^WjiQ#`@C5Ss{}e!3L^y`P
z81rnHTzg5ycr~S7#15MgJ3Qb}GtSeH-5qQ=;$&TZyti^jb=}EFN}WRvHSL1|hAt6L
z-*-qy=`ng2Un_}DO#N#V)`RTcs#HJZulYVYqUL!fUWM7t|6JIXZ=d4&Kf|@fBn0aj
z;6nLLI>g#Tf5Kp+GwF+uW|B+5zd6ZivBQ`qvgRE4$FOGX5l@vA*)zF_Szw|}0;>l&
z$s%wT!x{&g<XHe0`?;xUp!uC|LXFb)b`Qbajvk+GzD?=*M}K_<X;01ix8pCHoju$=
zI8MqaA#9M35ka84O!AM(4g=v9GEfs-eWK(ECSDyst`8}d05YKn3ri>i@Pxfk07~=#
z5(q_%M(R*v1cD$ahj|hbl3^A&0=#h@N|FOUBFjV5qi9$ejn;#pdx)y0#Qo^A4ec)Z
ziW;FZcNJo`X$KmbX$=L$Bc#$;8BrXvFk{2>fQ1Uo5WV^#htzJ7Qs?_%!-ywJ7JN1)
zAWLyDP7W_&;s5^>pqWF{Mt5NENdb;1px1|_KZ-e-ndCQliu98jJ&c}Eg}PIGNWSeQ
z5#^Lj0Le*)gR@~UVtNEfFpOr;(Kg3k0bDLo22z;RSknnzViw#R#=t(@%;>Boh4xDd
zLkzEFi749ym65*?vWTW~mx23fnKTaf&!87v{#~dRVDUKEpkk;3@)REO3b>RQqHsee
z2p3-`xF;Nk+#U@hV3=G00~U&5!hl2O3~&q3go|Vr85=l|04_)KB*+{CsgrOLEJ8GB
zW<tdcH4@S(yA<9QqFpq!3?0UQQiz2@wVIirfm)z?2_6YO<IjI3lN^$G9G7Yz-l}rw
zf=)g&<oA#lF78Ly{=k`-A?P7h2RTjSaA=Y&1jsbIMXIDK2pG(7AsKgCqzu+-2E>zC
zOXv~;4dT#o5YZgR2_w8vNCX1aVhTDq2Ed?eFxy2$cmjfwYOuMOPjFpafJO*CemMS$
zi-xMhfv($`qX1?xWGNX;RYh(<%Lzz`uSJ9Sq+v`tenuIwBQ?h%u)}0XO7OYKTn$ZJ
zSdytRgQgrlj*}HRDk}+N9rwB^ZHxrRBn{eQz#(NTk_B1Oateq4!J51-GfXiCttQ~2
z#o&$57wj+`|0z-&v{{0SkQo=J$ZgA5?2Zd_@|nxmlV^ycC7CY*#QzHXcybDjkR%KB
zg=L1ZS;;oRG+B<tM9C;REs_H|2BTu(F6F%}EJi;FL`H_G0o*vrU}D;?lGslKVIoyY
zn)F}0;aB0USQ{V1gaCi2Jkm2lc;Z5A30Mr0w%~yO(=`%qT!6ezB+*L@JQOAu5TzhO
z{iH#qFc@5PlW0Q%2~>*B6ZjE#PY0io4A%~!MXK(=ns#8}C}s$DbqXsalm0mzO{^IR
zn*;Ah;DSt9q9}!rd1Itm|49p5A^}Ldv#0|E_i+*8@K49cy@b6WwZH>^z&VjNLfA=~
zO{f#s!w3zrps7d`VOL7=xltTEhGyXr(teZ?aO1p7`HI%+2~67%qofh{O$L@^=E(Bg
zFy@zowTv8^U9s38cF5EaxngcUhensoRJBjCmmNo^eI=rj17uUw`Uk91q%*M4?L}iq
z+aPDda1R6w1K;AnG6hm(qd~Hn;^ZwZ(QWw=;tn9L3|0oh&|a#LWf3K$IO0vWFn$>|
zVO|g6DJ4u7;j+S@4?901V8RX5P_?ylq?yBzph=xvf|@vCY+5F20pv5WR4kq5l8CVZ
zBybV~=unrWIAPRl035jI0F0_FS%4N{N;qAphlCHnVU$%u-I$38Jq}PaRV2WZofmWp
zXb>x5IE|C|uYnc37+e|^|0ctAxZKfPCX9dBSCDuK2)!bV#bw$qUO1|h5nvRK<)vrP
zC8V_JOk0a6<mOc|GZ+q_k_D<mF{eo>yMWpu5dV6cg{TQ3JQNQ3LahYqT!bsQ5NY59
z*lJ8jRbt`%vDm~=ccq#xNhpgHI+lg~59(ZU$}pk^xXjE!g&-hFN#a^yB!T#K)Iwi?
zF093fn$GUUcd_^HU@+FpbmMU`#DI(Fp3@TG2e)V?0X|ARuojzz1Y<b0<zNowjuMt3
zg^&aa=$L?8L&`9W2xPh><gBfv5cB~_7>*u@@0w`XA}K4;J%B}g34{EBKITJMCh8w%
zE;mu)(nPdKB?jb_IFrqx(@CBPOX$zwg4|$pgOS6+sAds#hOx?3$8pBfC3Yz`w$R2C
ziYf%B#*p@hNl~$V%1l7uU`giw)i%WEJfDFoe~-7La19CpxDYNRc@OYbUc$Ihwpf)*
z+MAD!h0dtl+F_Kg5)~S+?#A7rH?hi0&MZ5B;o}`siUhU#F0>H+I|uUeKdl~T=CJo>
zeL!5`-QGZ*jEAEf_7>Orpf(C!Fv9RoLS(0BkILajVy$?Zh7no|Mqy}Svm|V#vL3q(
zW`fH=z?yVQtYS+WP$9BTN*L1356g*D&u-u;Cy8XaN^hjJZLyXcB@8oTBz#yjzCB0N
z7e<7nEEj5=k=#l4KJ9}dqlHO0f>A(ekfMM*)?>71zZvHiow=3ve#_{d5VF~FUPQkB
z6dNxk8#N>Lo9%iD@7)$m)OqZ1SoF`EZp^%hbtqnUN_|54*uX-g#}5}wvF_AQeZ6$+
z!s9A)GCwZP@zlLu5c6c4Vfua3vu{pd1vc_JiWBjNF|qPqg#IVitgI9aS%JjE=J1g;
zN1B?Q1>N@Wde~8PHj!&llI+MZS|-p64$j1bVUpq4hdM`7ddq3EOvbYD(MJcyd2tGv
zwqzAYG4){cZMw{Q2nR{gXnqo#CBce57WO-?0J^uuFof*=Px2;|E=<o(?e18w<FcbJ
zRPzZE19P5qt&AOIPusCDeQ*8U0E43syd~3?<}SA5VME+?MiFN|*S+;<NEq~0y4$f)
zPd@IB^0Be&&Q>>Lll2%TLt59mA+92AoC%gr-ud~={brB(NrN8xa%0zB#{yLwEie({
z{=oJ**zYXjaCsE^f+9A&*so{KcvnMrDg|KdZO3E(vK$Zw6q%$YcZsI28r>CEOAIF9
z0V0qMnV=IQ?oluSa&giXdwW^z$vzZnw8RiQAd@f)4<WJx)E(3pmm(@U#?Dg?JLJ0*
zbOjyb_SGlwVj2T<j!DZrxwRs@>v6-OH>Kl}*O<SOd#Ic6gf`N_cHyzH>JGff7yf#R
zf9>1$w3)TfV|iq~^z9LqQVWlpkDYd7PCWudXAfid_OA2?qbfzTuEpg$Yd<iXw;0I(
zWjdi|X+!55SERSn@88a3VwqD+)JcmU86lJ39&GscT`)Z<T<Wi$<>$|Y86;uSB(a&?
zOk9b?-ck&7TMctY;Y3i33DHdi1tlzCc6A~gNz=N;5m`hZWwVCJB6d<(k~J(Jv1tuW
zjs+EvVj;Vn39$(s4Zp1_xW^43fK062pqFO-c0^5a*LbWf;hZ>n&?9zxOz0xN4SIij
z)YMkdBoZD;j1|p4F}prkcS>yZh=%47sh=NsCzk4;FzgM{J61O>Y_Uvfm!e?xp}a;z
ztjN{$<^&hxGv^tY%C$~@kZUox72D_bx-H5#x~<>8^K(IFr=L=LxBL8S6^EW6i9L2*
zw&wfx%~I$PEs7=<;u6+#fBblNpBb`(s6+{_8W#yxpc{b5gy;mYi2BDsZaR4c0+x^i
zN2wFbISlY4n)xzoHQ8Zb%r%YJVPv|ZCYEKvvaU*wG<;~pP#62U%Gf=1T#(-j?9;`y
zg?2sU)QmVLBBF`+x0Ez=SFi4HuDrJ@`)U2Pz<dmC_r>s#m?!q~aib!(etz&~ad6mk
zZ!A=3Gt$A$kk#v^kKN8aYMP@*?5}()A1_<q>8tbi*lDk2N+)t?ubd5zZV2$0KY6;R
z=c#=<KA#>IZZ$YscW3d_gdv}Wigmd<%k)y0dY?r;X#>bIW0x{PE*w7+B0+gOb1cBI
zeEmrd389p;=rn|cFy}>lmE1w&00t0cfD`1vIJ#1QG2c<9JCY748JQ(kca|$HVTPWz
zTW_8f21|s)wUy9hCSWkDB%4mO6>LvZu;M=lk>kF%IbpQ@eCXM+>+VKpYj0C3|1$YO
z<eX#HW*&R<M@SS`udz2g`ffcfSH<1xZFmCqBvJ8l$7)K;yV)PN%u9OXJul5>q)@)v
zW|YUA@-8GeYU={$vj2W~G)P~L!<BaH-hxSqW+Mc*FQ%yogcW8Zo*1V4rp0?_A8yW@
zkZ&j-*U*0<BPPT!qt1-t4F)j-2o2EmEFn21Gm?o&gehpx%;bVb#03q8)j;VG$VxUl
zERv4hfoEXs7Aq&-Oo+MD+$c3}2u(zf=jRhc$ehM!4lcgWKajK6gu}!@9a0axOme|R
zDBjUMg(k@XHWn+DC(LoKd^5e=eF66UCiB!@=qqB~+eKcx>`nEbNbr&d{SDG$y(1Rt
zrb!&_4>C%SRFsZVx<BYpU9V^nY&6kO0fQB~qiE~_TU&DxCPBl1hae-YFDYb8(aHD%
z6j;e(fn{w^F@Y0TOz;e3VWc=9U#dvaGJ$i_R}KsW^i`6@%scwgf-O5{NeC@R<47$-
zim!Ng$EfV)aicB{5}Ba%2EDku!r8)=OOsen07b^#aA;f_)>P$i^f4jlof}q<2a7Dc
zvt{xA=O4#;kGLT1?Xb5Q3ypg0e^TdjdQQhh8Z+1s_K{ISq!C7xG?4W&_~hW`Af~!Z
z+e;fKdtXl4Y(zu;^$>%EUWY@0bTpO8#p_6c5zPi&lS7K`a7G&ad^r~*hQtyLZi|F3
z8nu_GJQ6b?PjdkyARBoJ2LD=KW-v$AVw|dyrD849Tq8wU!j_#SyOuSmAt9ksW9b+f
z@1&ezZW3h^5QFJT#CM=Xr344y;)1{_izfx$&!o|KFm&NjAy3GK|4LXxT77ZD{ocjZ
zm=R`ot;5F%Yi+7DNZ+2|p3sZ^qs>DMxFZzSLoMt(D^uL=Gxu1W_u@GvltL1B3v?+-
zNopF3*cCt*+aG+Mf$@Au=McJsu+d<{EZNJY(3uJOL^vIY#5ftSA~Zhw(jA>3eWp7s
zO46{Njf@a;JdHJ86l|oe%-~~@QMQPqQKT5A0j~-FNIi(QnX|-J&QHR?G<;$Rc~J#6
zFF$OZkz)KPm96dVJ9N{eZ_PPwGtFW7p?<FgQ#FQjJ!~-t(_{YL5mKA=GVbTzka_;m
zXU^Y9Cq54(6`*~jkmxC#eX#~67UF*xbmEeMjXW++X-IMK4HJ?EH;t>I1Tw?6c9}Ui
zhapX9ZFtmykWq;bu)N+t2xkVD21(751nI*Z3sF<9kBnC}QtYPVG1NsES_7^w^9&a4
z8xSYyo<JSo6NT$kk)k2)G}417z0RXs9$R<W*?7+RJAR~x`=R$cUvItk!b3m(klD8V
zPx|T8r<7tr>+3-}3U#KR+1pN^q@D+bU@Q+MIvA9kL5HcoQnapFbq_vk<gCDb7-~TY
zT7Z^}Wrur)v^^43A`K3}fGngW`hjjUCrW8DcIYEvhQ;6|D10kn(20FQByB{SLya=r
z2<R?6A4b!{1F+IzoG={PFx~VVo@YShphM)yo@B0g5<F}Wa`Zu-&few$f8C^;euh&X
zteYT6>vRbI92@N*k$-m=i#~mH<p_zvvFaXkN)@cP#+|=LtOSZCMfd}pal|~x52LD<
zj7Yj_yx@LTMouQ7H*o>>I2<J&Zq?AxW^l*Q2p%Od)fP?+{@OVvX!1Jq5}$UnKC^92
zfK*T&2LmxUQX;qtojHYc)PpHAm~$$wkS;-bM&(t1qfE>a5=`zGE@?^L!$o0+sY|01
z_cjk=%IN)RuT#^WV8`J5L!Jv#{|dXd`H)lo5Dm-Tbd2?GToj!D#)q~~XMGs@LZOGr
z1!;f^AXdaLq`o!u?nN)SnU!Ko*C1k+VlGe@TsI3E%@L6T9#fJ_s5Fz{2S;SLN>&&<
z(nyi4VVe+1W_82(qH~PG6?qnjkJ(DulCJ$98_*;;jNttL?HKpB_r^i+98}KS`FV#<
zTI`(()$NU*y4XA>yC$vQO*dod$PvK<54_LrEq~2?4`O`YfrXHZ>#&&!ccg<O;VY_N
zsDfIl6(J!K?E6zbQHH=d4FH~wIWNQs$Hvu+(R_knu070{Fk37@Nbp6VLJD0<O6h)%
z(ZrMHi)%xS-o#KRS&<yAdX{}X)+@dS!a(Ak$)Twca>3gX{!@B$lBSQ@2JhTxM5%@$
z5=9ZDB!(D)zNc7h!&YP76MU{I{^fQb4ndTLKK9jfIQf1cP)VhQGuk}){hJUeY!^X4
z0RE@hl9xie5)}s-0ZJl37ngD4+v_fI%Z3ro0O2F9BteFuq5arjqtF&vS0Fc2l0<i4
z=Almry-G;%q!wT2hKv#>l{I0CG>8*!DmEUI48v-}VO%Chic6=-E*V2yaJ&pDLbYHe
z2^+GjI@v219|XD}`SB==U`uKgHe~8#4N5Q_s=IMo0=ojzSowBj7A+n|Tks);;08R2
zumt$&cNuZ>s<Hkj#;mOHFbN)MJ3EXfg~jI!k&B~4UfQBVbYhrgQq|x9?jFl=znx)*
zbaYnOavvL8S{Py}=rEJZN<v`PFpMvmG*OC;paFRyDM-biR)<CMA{QGJg`r+V{@@s!
zlU7Jbk#AcrzNjJ=TsWjC_77iV;zSsfh4>ljMl6`5b9#>C6A@TAimi(K**P>xM;Q}L
z5XVNO99cIEqO(AFhXkep0lWhAn1HQ+4I4Cj?RY$oqne~54#xjv6>vu-UNa1HXs|$E
zNnjKg01O(J&cMnK*f70-Y`;U=L<h4*L7aq%;fY-sCjtdjV}I)vJO&j>-66$6HC$4a
z2AU5wjq_9)H~z=_cQ={KeSmsQPcY#ya9ToOj#X%+uxA6n%>u)7vFHqlMqC$cujs5K
zFsE|FQR)T6FOZB>>d+p?845vsh<b}8JrnaSvFQto78a?bMUOJ-!RXs_;+9MBNL5IO
zkg74-8SjAFCPl0faUgCvkN}{GcYT7L%7sw;g9)0H9#-=St~HwIYM+3p5=AFrWHstA
z@CfV!7EP0K+A@KDfSn6@#p6oRpm{ko<9~0`W&yk-Mo3~O5Xlh|5(LN~2bmG$l~{8*
z%CqO_&?j(hRor-6<}F^rS7rbE3`oZ+{!)^!go5_)g(9Iq_0OkKFjLJcl~YbiIAeI9
z-><?wyUS&YWvU}aj-9MFdg<>wPaOGu(4(68C3r5+ht4vSPb_Bf==e-bK{@<LxI1TL
z-><^med1fy1;3(e5>*QlrwQzL2jGS8$oa$;Ue%t$!cD)5g}YPftv{Vywl`3fkgE8s
zePbujcDrM`NARbCGg0tdd-xbjOY=tp^>?ga#)XCb9pMZPKCzt0*>2Ubp5*S14$Fi4
z$t!9SrvcVc<T|-!&afxT)s@F^i&R1#sD%?gc5)+HGA$r{(Pe{wX-+L|qCc;jW|gXh
zhbt@{2}B~5NOP)VNv*=&hCTW<szRH`53L1gQ}SHzILBl)E1~G;yHHf}$a+1eI)Z*u
z!(u#c7HC<d&h>nF`**#800c-OAxB(9EJXD9lZRtS1LFZ&o3RrU%v#T1{j$pMr6hS=
zeI;o}i#9&u!e^15M#u^D1;!#9Y2b`>k42#g=nHRTJPHVJTokwaZ_=a3Z6#J1YRd#n
z@g_Vsa-L8y*AtN0K|QBGSD<<X?d(w%ddD$T;snE!RXS=6XY=SX)qqQ)>Q>XkL-r0I
ze!8r=D!&1^wjhnd=Up|O<yJk-DuE|gO`mwU7$+0!!xwszi;k!s7eEnHwTvy&=knwn
z+DHX-q0PkK?bWX*mnoV_;3+KQICk4u00M1FKnvftvNBzB)32h_FHeeJruxH2e4uXi
zu{}vW6TlNqp6y|I@ZkP6sp2cf`KXOQS#1o?IjG9q5a?YkeD{@50M$=MPiIXAp(&3X
zv@LZmKf#qp-%_(xHSwdTFr*=O5k2uMI?XaOMf-P4-BNq|tLW6Hb^_W(2MEQ@c1xPc
z@921K7>xQB(}jX*JF-ua_U~k<bXL<feZUAHt!jdG)H>+6oshyee*!D&V^bZ@=)iHt
z6m=_xkf)%jUhw-;;^1MC7#4u22Dy9MZZ>rSoYyE&KY)iOb0VY^UbY18SFNl}Ruedg
z$&*1yg6iL*aK;>wFzP0#Zt`rE!%|ySKXryTj-3pqnd`kw<&|gran;1e@I^Dwf<f<t
zs+zvw)PzWNi@1Mjy%*cA^b6Dl>Q*}igq`BX?v5f&``B5-bfxGtpHHZtG|eLQl`tUP
zpC^?=T2r;F8SMmFfLw|b?+DrDp!=;`!Rq?FsqU&j06mH}DZb1&M<~(9$Hy0>7(o{b
z4A}%mEv;Wm0eH{43%FTjuv-P)B_P`YDkWrV1|ZC)SSi;k^<6_-i*Qe#@B8ls>Q<>b
zkHL#Fx5(?opPe=82f**(0=$AqEj}s=c<79-rRs<vRmh7^>Vw?PwmWAhd9;{n#qOwZ
z^b@T3)Xyk8CJ5S$2TrsmTPD7BY#pXU#1+fjkP>aNu?M50s9AkcJ#n18Y)g%5Vgs6)
zKG)l-2Pfq1LH{>!Fmm1<1%)<?xJV~EwVYa&Gs@dT)^6oBo#Guh{9}5utSmkCf?cDw
z$@%Awnu{FXi-dx2lbC}`EQF!a7qkO;t91ur`wENII*gg$91CV<$OcRjy$}|~Z}~cY
zSB?L|!)wx4mO49K_F`J6e41Wwru@V8_@sUc$P7NKn!yjqBjhaxvk!d(0GVi{Nx+$L
ziDgD7U=<in0rB(bljjV41XQ&=YtmIB6yyvpZiw`M-hD4I$>&F5s@?L@Chk86Lublz
zd=Ivjy?U^0`*uB(<i_~jT1C=d=T^*3)?4|_W{XuG^8!5}+gDkrJ1{-d^-_VoY4OKd
zo3~s|`Q_YP`PKi~k@2fT^D4{I%k$qkf70?=BY8F1S1aqnrXa<~F<p}{wRxX={V&a;
z^!&Ctc9z1<nx&=3PFee0mhag-{K3v(jmo4Wu35#Sg`GqH(vqLAKD+arsb6D$)${x!
zex8=^TaV|7&ks9_x=n)nm3A97%XS}pTySRhF@d^RXxI=m?d)l<D;0}Bo{x`Ij)-Me
z4VfCDO6Y|@`IT_DXX3}%9yl{?hsO4vtLq$xZhEVkx=tJSw(`+|qS4Xz%Qq#j@IAf5
z$aq^TuV?*nhYIcIPG8T|U;XZ+w#4rE4&QPYZr!h**%b;=tDOf972SXQ{Nm%u(@)B{
z*Y)3e<FeFjmO)iT@bU8&v4`(59`i5Mb_wjCn(5jWb~z@?_4_zqNcDCtREwAKFn^e5
z<=Y$8^|AHrmW!+QHHV+@)5^K=QTLHBP9|%Pd(B@B_Dq}GanCxV_7<efZ?-?R^810v
zL(7&YKYMy#qhy7J&(<G-C(Z;2iA+=#hES+Bl85SvQleB?1qkGhyil-J?PlVDh`QVZ
z(LZUtK;N?_;}C%dMIea-p2FQ@aUE)hQyp%b1d?$uVk!)8O5QbpL+`o#{r*iGi&l(g
zEXr8l*VMe@Uc%t_^5RSPl51nFH_dYBoTS`UTzi3)TUIA`YiX0RyK8C)w_tzbnWgpR
zZQ6}hUNWDu9epjHl)dRbkW$ER3er8j*I~@w3R(L5p_N8Oh06I4E_(?q`l8g?1E>5h
zKH7E4q&muT;_EUkH|yr!=}R7EWv?z$vCXrdx6Ny!kbiEJXU<iZxh9)@ni{@O(#p|D
zcD$IZH!&@*+-cJ!iy7zJjo0S7g^wXR;2VT1x<j~FU=IytxR4SL$q+L_c6eiBBh=5y
z{}h+Bpc_Ih7Xla{$SD8u)B#okV=+;mfQWiT(_ovj`ly1ZFKujII{D^a$-8|l#P52s
ztYTb4zJuo{$G}skjz!IX{#a?L`@`j~ix#P$SDxlmCev5BYlarLbMO8G(k=e+ZSqOC
z=eEf|Rj^5^cC~Lio&Ub8%Pu=8IjiNvroUC@2fmxX`l7wHpKsS|TMt!NpV8J*8?`di
z57%r=O5YZ6y6W3t!*;2BqqO4O{7Yl3SMQf|Ogwu;OFf{yFrnD<^tGUtZg2E_;oFn+
zDMaOi=s^ULN~5dM1M>NCSEYc+2h!_Q^q&BUqJtj-^=xuOV!crCiimKA%7N%>Uks!^
z^nCvmazZiV#N}?st;(uNYvoe%#^rzd+SJrv_|#b#rLMZw+1lnug@4;cN4Jh6o*$M5
z2kd^`bp3qzsB?CQqJ0OG@)O>9u5XJ9Oi`%a7$5$tWxTA7^@(`1i_JlueS$LGbMHp$
z`ow?uA>i2s7FmU)9|%;-=$ItZ_0qR)m4DW_Gq*%w;TIoX<-~Ybqe>W^GOudVJ*<e5
zhy*=DOHcxFQ3*rk-4>6I8gXY0?QR%OAw3)~P#d?mH;9{;_u*FWPEYsQ2g5rnCXPI`
zaqHuNGlg!MY5g+l7Bk;;oO=Ct+l5UBgf5;_|H->>`oR7t`Kgf=MWaK^Dx-Rgn5{Xs
zR*XxhcWPT5>aShpwXA*h!{@f2dRD11eP)?vZFOq<F>RdnrDw)#zuHYo8FL_1-}0&J
z^sPC%OPoS%uAG?gZFS^er;sAVqcu!Ev6~a1oJQ&s%vh%mPeH4HD?lQvMM8moXHB|W
zxNckutQY89n2IpPaAwiiodn7Gfz6@)uKT0eYkdEu$?G;Q(`8#&j<#W>waBn|RXd#A
z{A2cLoAa|m*XAE}-CT9xbZpQ`n>(ihWW7zlZUP?<E=`Q=Z@U!uIQ9A@?Nfg5dbc~=
zcy_i~nBDIC{a{fWgV*!J)lpzJkaQ`{X?xYhvCFp2Pd40BBS<Z<Iy>)NK=0gF_Vms3
zbe-A;Eeh?HdtYA-ix9F#NK8#ABnaEt{=!IL%t8nZiR~|$;WVLjq6jw2gqFC`su>KG
z{ej@EWh&u<9y9q3at-fJd-Z>{bW)0)ayGj8wED+8K*C*<>q%?;w+tswT2i2KAU}EP
z`m0I){A!a&Rr}5#;wLsZtg)|LH92O1&zjrH>_)S5jC(<!S_T$e)Rn*Tq|?>Q<S={i
zyZ-79=_1VqiYFqP4OHT7IvOkv&Jvh-lynqa9rc-KJTuvFLk|0guaG}FE_>D1(X;ZR
zIR%pIFLz&@^vWO(yg~Sdcn3smG%GQb7`-*BKQ|^0euN7I!%|@iu?x)64MKZ|CoKbh
z)K9D3x^)wMe<wdL<jYIB)n!FmFPcx>S_Swoy%-X&ygj>iaQV{JrO7F8ZXXh;=ATzc
zzN<8M!Dgv7lcV!qW(^gVnYu6==4WKzuIo<qaC1EDn$C=?n55NZG(X<&?qH3{$MjJ_
z{zWeT$jz~fzdTd^n;Ac@F*@F5b77CBk&OD@)roe&4Qi_suB01Ji{Em&c9HhV($Lsh
zCfj+*+PjngzWYUcg6js&kE_7)Xmv+ThT-@~V#^6d1q6eXi4_Psgu6s_o}sdmGO*MS
zED8^4#OaUfl%xQU^$PTPbf0kBN|=H4!2F@!g5HU%c_UcsrrvPnYz~6jSdeC=FC5sg
ze9zx0f*a#aXIG42Ip4@mvOkz}Np|NJ9)D(I_AB|br`XAjRsXC%_jbCRx}Q_MVCPlW
z?=!C}9(TQe?LMQY`$9}dHpibgd)4uqr|wA>{Lod+IqPy>OUb%;dLggBYju9joq_F|
zWo^6ooz{a#J|!MYR%$6UkGrvbYfIoMxxrOhRmuG>Gtb7`RJC8dI(G7FN(_RxC?}CG
zGvJw_!PnnH1fsn_y@;Ta`hciy=mD9KiCuSGAs83wDz>?{au)xi+3y&g^-!`{cGcDq
z7UPl{zMFal<yN$5U9KHH{gfyBD97Ve$QEPY?v|+BPp6L<ONYAG>W;H{X;Ia)sKv#y
zY15HydXM%0(d>Hmc%`29;-(F2?&M6`7Ep9P)lO>hg~*vI7Mc&P4W_jC>RQ)kJ1Hrs
zgx)N&n9+W^OXaVk+IT0eMN^Xv4%gm%y;9BfKxbNGU`gOcy$+w}PBL?Nhd&l>`cA}n
zcMa)74^a522A)dPt(%FBLdv!U#S(iP<~ilb$or$W7C#v61gDF}kGY}RLznrO8S7zT
zbn&!c^Eut{rU&7wdi}y9yL=b=uVZMu8aCLFHKn#IWcso`Uy;6)#^%SBZ_0ySwdP+9
zzgRf+AC_5urOP7Mi@KE?v>z3EroD)1{rJ{Zw|-ws)(hst<mGY>mb#~3|8m=`x%6B2
z<+2vHg6a{=RaT6?VRuCPg5!Pjf-yg)JyKd4t~<e{Sl7QW+EZqunO8kG#<ullUrFsV
z5_>p&^q%J3=6yfps;)fNuci~ndJd67Tm6a1A<jNjMAY4tD_0U3^xH_Lx@*?#Swr6`
zdbQLI9{vvW8BWNnx}?yZ45cx)rkVTY6`wI{)~p|j-?(&AMbrTm^SxERV-7n=``9G+
zWH4K2G=w&OIGz7|I=w&Q7RNMc#l2G-D-WICkY;sQbMm3aPu*Xi&2$Rb@bP*1?q=&f
zKlk*>vV0#3Rwy?)%0GXmn|EG%$&&76!l7*@h5hamG@fcU$*Ao8$L~Psj-jNXTWb$m
z4K)wNZ%z^Nw>qYp3i>q4n};6yEy-0|xcyw-ITiQava4PqI%9pIY1)o6w^SiyiK8a&
zV@{zfRU(z>pl6DF5sbJ_4tB>a1VN~DHuMe36mM8-5)-;!ZBBu_?0oK<m?zp3N{ghd
ziW!j;{hzt2={c88RB@?dKGD{W3W0!I=IZb?$GJ9(d&=mH=BuJ?p+oD#^T~PB*V@?K
zH*w01Shl}8bf(eSR=HM(*V8?H`8zLGU4cnS2|!|pXiAA#Jcacf63VeOr#x7db{pb(
z!~4oZqbNbr3;XNKVAZZ&`P1zzSrv5i274R7aT_hxtBiL#I#fS&``kjOve4k6`ZHr#
z6;i&MYnj5s4e^hrZ=$;C04W76L%FYrvg{HOLfA@?{>Rq`rv*}Uy%V>7tOkRVcOx>T
z-rE!OtN2tEF&aQ70ur;b8vHC4iK>Y4z@Dxlj>Bw^2-FYwapEhf*Qus^N5j67YZnUs
zX(X;-x<((VX*uZ=ENI*f0}J7oh1;s>GPhI{sjv*46|EVORIPzG5lfxa7rw}PG#+T0
zX7%yTO^P*DGeodMVcU!sBpR~PAdq%069<?Ch5=Q3oSi?wN>#%{j(=>~G3M_mTldc?
z!ug!|MA8(}0U`OV(uE&74RxiwkqiC35##fLE6C&lGiHcYQNAwKzqn-Wt1ZbfoVZAF
zKZQ0U2^2_68UG~SAh9#lT_hAdLf{}{>||9T(GHyQ@I|nc4JqjfuY-DT23yk)Fd_c>
zR4urq^6%&}LyZJH#?r(?rygT`U&x6?<j*zbWpC}}s}5A)%!a>6mBd%*GG7-`5grn%
zL?2FBZvNu6>gvfF(zSOd9(UqT4N*EWkA*u#LIEr1{BT*x%hGgQe{5(cB2=LI6Je;P
zYJ~vOn4+`qnr(2h_fOyj6T3*yzBSs&<8`|l+2{5+J3hI(obE&4N${eY?oYrZ2l%?K
zjDkil0N99SO_v!&XLrY{#^h+9Xo-Jv)x~f;=^DH%xEBm@G@_I8K0pi!gey_MA<L}W
zpWFue(76#Yl5U^!u=mc!H77JctrSfFD3P=MkCF(8mPOi~IYM?6Vm?ZM`E$}7=qz)4
z27)uh)*v7hM}#I=gc=tqRD_g~x<SJ|bP~V`w-sBQU_cT<W0}224a5vpx9Z7|pWH#Q
zCaz$VL}UOSe6^RAcXs$pt7AGvA&5ocedsIvXeEZ(d#Z9`trFK^(L`e9@#sqQ08~iY
z@|#Gyv8vvewSB7zSkY*I;*yXD^b;tPf)%OmDj;DRxMYZois2&j$(Fo0bk=)^LA+Eb
zFq%PZjgAfhVL`<106Ugh<4M%){qgE~!ogz+!-8eeSyYVPA`IL?_%1;-k+QDMo@G8b
z(Ngl0qTI)FHK<i}&@FBT2nkjk+&>bWJ8_A3+_-4fmjts~gqnf5SeS04KPU1iPe^5u
zKIQCi#Y-&+&KlK3gh<gF2Blg#1Os~<$n$Gn9s~_iVgl!^evF1IiC@F}@lN%Xe;O1S
zm>J)au<l{+<F7a^&>41Q@kmr|;>3w5C&e)`n9JV#Ky!#)!3H6*x%dWk0gHO(JE|sJ
z18|5mVRo8b2MXaRK(i64&!caxOVz24Z`tBudg$rT6%N~5c!)dy)&>Yv#_8dC2V+SJ
zFX8jz(gxka83-mCn+PeVpMo^J5}q7^Tg8H1h5$n<=BiRwAFd))YiX`Rd@w4YZ)|4l
zO^zu~B~kS^&=(VQ$hlohSm6D|Z&h3HQuH59Be#L%Np!;8o&rD2=+_8d?1y?9Id(j`
z2+9xFZ^WHMVyV2UF`7I~wKYB(F+BfC<0%aX?CApJ8=T_n3;PKL5$ny>HHr}E4bqwr
zf2v7@NfCD^l#r5OcBeWb34sleoc<kW0QD@@ZnWu-bnJaCkTHG#{f_>~;>^=>JuR!I
z$~uqPchxM&d($buL#@M7tA9?x?If&tBiupIM9^(^Qy6mL+iV^L$5ds;a4O*eo!^NO
znU4fhg!CEhysF)OTMUloybg)EujQ)lE$!qo?Zv1PyT~U6t-0^7utG*ZeY#<G^D1?%
zQsui@S2cdhcdr;YFmrr%mt0pxk#VNe2N4)zte6jgXCjvZC?PZ3?KVhKm1Xt}Vga5M
zl#%L*viOvUpr5VHAtw3;3Y}VhIi-cEOE1Zum)*C~zk0V^&!n@?0aXKA491x>ecR?P
zcZ6{`t9a4V$q7y?pBX8gT5&Axiytpo(Z$*|)R<fJvPkvE6x9qvqDH9TY)dD+*92a~
zNQ7F|+wp?{qNjVo<~9RHomWxWzB|DOj{9t!KD6ltTxP$W_nhr&rYX(-@apJ}I^Q&D
zfox&aGsmZ9>rV7Ho)&(wd#9aqrZE4!#{0@sWf>3awZ3*&Ph~QOeSk-Mfsr8z;+pm4
zqjtw<Pop1{d&&G-n_Ihr`SbYM@bg>0uOIS1)qP#N(8=6ombU60dT0Iuzokh*J-_<2
z)Vm^OS(|-B?`R2JJm&A-yDVIPfXF^ry%d8t5kbL1KlEjODu(-guPvw%==qj!IOCu)
zUe5Xb@g39q<u8>74HfFG%YUC`c_eD5aG`Zw>r!_aZ<B!@yDUSOv3(O4ecpE9xLy6D
zw}Z?}_K-wk$xN&<L|HvfLCF#!PSF8!J->d0`+b|aa^1}6Kz8L}odK7#TcW}jxQMcX
z+6$k`$h+z)DK@f)Zocm;ugnm-=A<@OpNiFf^XPhWgS{ndNvyjuujr0g{l=ZM`tg*~
zOaQ(q<JVr^{2%NA*$IC=vk~~Kc+HSeX)iii)7oJ(e#rUN_z7<>xK(MDUNop!uu`=D
zx&JkxMLF-`gOc%fyK_%JtEQ9G8;P3NfTp52@e!!=UH7z?;cq@?=ULw<n{H|_^ZoQK
zceY%1|NFez4Ap_`vgrqJ%6eE<R92bD+V^?4Uv`%}kacfKt*pX>!Oizx%kJwMReLe+
z(+ZWH5sfPyTP=MT94%JgoL|&>l9)lMoVYZZ0Zx2(4N_gi`{K!Q(yz#7Cp%c~zomOD
zdHxx%yze`v`)x>hTH~-`Lh=fIzGbGUXS&~7OQ%Yap5>@>BWGlFDYlySt!_PNI@2-A
z(b3U#Mpl8adv%n%!A!gG1Fb11HfQ}y+q%NbD1MoRxx;|ZuRi%wry$K&;TI=vD)^!s
zx}_cZv(b?RSQtO$7{kk7)g|zz+Mz8-gVDFNlOU*SU24a*rf=UgE_GSglpSM67*+%|
zJ=t?U{QSDS89pw4MiHg4N2K!(DD)0(YRq{1&%7nET0@&|&5XKCkD4B|t~xeqji#Mi
z`M?YPwx)B*^P5_2q=&VRuJ~m4y!@Qgy+kRaXM3tl1X>H&k5*(iK2~pT=^Of|r*@Yo
zvNnw2^$;BR{4<x5nk0VD;>PTmD)nSVbeC~tm+X_gi@CQq8n;!i*f8J6VZ;3Rr)FhI
zp^x6T<ru5kXk<Iv7F8T)Ur*uGxh4;tb&L3sJ=EBGXPNag=l3m3HXIq)HC~i*a)sBd
zzt68Mz4hamdXm%KpXEVy7MByxH#_wqKche-*Uy9SssIv5y2anVO&2z+m<VJ_&u!CF
zJ7vz-`=+Iuy`^nx%2}TuuBTj7&mGg!UR_hZc3U%V60>ZM60>Ug8UL8>nd6<ljCZ}`
zzS3SR&*;Wu_Q!_0s8_`c%98GMJ@UVt5_@lVlt8geSLxhI#vJ2mC;GIun<&0_FyGWr
zlDu8#N^aif2-Tm82+HoJvP9~=5QUFb37SYuTfM9N*P>H9++tnypSdcVrih*hkIJfU
zFxeEayR+|FXHtdja=U`jn>)2sD=IhFUr?JNbNE_pw9nK;OUnsnstMt$yPTF@b;{5{
zUx)F<g0)_T;P}3=K9v0mL1)7kJO*2jLGCzyKNUTP`3LQU=%X8@4%*}Qz0CFh%s<p`
ze0JiBkL%_wohcLO>)oGZk}m63X=c;n`2K2cHC$;?M%=uNGy=^?bvCFgai9}I`*+sV
z=&IF2&fvp!mxFz-Ot3t=o&HXFmaJW?=6ffZqXB33L~DAgJ%61-4>BuOR?g4Qp6F0m
zGNJS>H=<PKte&;(GBwNi4(A@bv&#9m-oEm`_SWli>~)dK#DD@d1%>1@qXsU_Jz?i?
z{zId{5UMI?K~M6f0p+@6hm_-9F?wzrOx%}lxqR%C>w>do@0J!_`=QdNl77!}!;h$?
zr&?`hYW4q;Red&aeqG<n9O<8CGi1k)>|5P}uY$4@?hDthcD|PCrJKP>_C6A2Z}DJn
z)tYlD4Tj;l8r}Pu&URJ1vvOO8N549;7Yt=I<^M9Er2kF$s-y}~3VkXw)b#oIV6SkN
z3Q-_h?VB3o1N+RDN59b9G~xWH?}hD~9uw8#mseS=?*5Bk=3a8WM5DiV)1Lfi4{~;u
z4>>-$vv`wSS(;X<)ABW6&bVDLE$)3dOLo_(Ru?br@ZhLEGZ(9>nlHNZPeXJ4HbsDV
zmF!s6Bs7Hb*@$HY6a{pRCOPR!VlV*vjR^Q5nL?8PB>PV^;h)U9n$<^}d+12U!p9@W
zZczPcx%=3$&&i>AzE#kHew(hRybU^8WWUCj*=#+jSY><FthF@`ccK5hjJn*b%z~q8
zL)V^HJaKh|(1cfz!92*gHN8Z0<MfSrXP$0YanRKL&ZnRfflE%wFPT!_x!l;#(#FS~
z15Q^Q_kEvJ`Y$agDfHcnv-5wNEG-)FC#{l#6+{L_fUX7sa2UfM#Ulqi>r&gp7lHJ)
zwqvA%&>Ft6>W^;Xo{q*C3nB;OOE)*^U8*xV?scQlHKw$|-mWaXU@2#HbKv&rC3E{H
zbtT#QTGkdM1!-w6RVu3eXX^bA?R$<0<hwOQfr`RVuXRa6R#92TgFS5`AHC^)>mE$)
zYPhmVt7^r*D(&2hI=b_+E31y4dU|}($?uMkYqN<|S*^dUr-pL8Dc?I)2g+ft89lLD
z$mpr2@4oc%bgZ3<lYkR9O_lLv$AuX($Fv0NV`Epo*c$4(Jo0*=Z`a@Dc;~IYfJtNT
zeVbIHS8#vHoZ05Wig`<ZJh$n~DJ?5Jv~h!AuAWHKv#NJZ#cr6^1XufrZ~h9I`$?0=
ziwf?0M-$hNWm-mlTM$39L%7Eh-2HaJmoK%`SdPMm@e7^CY};hAF(u~Modx3uC%ijx
z235h^LWp7ORw5GEzXy%<!cv<|90}sJ@tcX1P)*jUMsE=ip@vyBXDqaDgpI##@B&PE
zyqL<ht0sP4@pAaW_enX#fn;zZUm;^P+l+WoW6YHpiy@HX_tIq=!u8MA3$1=SLXq|m
zQr$HHLr`1{d<ni)PhcSi0@#LsX;UXseHHG&@EF_b_$O0H<?I?<CsDF`qFszlh<~v+
z4#8`k;Tz&=dg!~)BGZ6T8}N?iRKwK$>6QL3tth=>_jShbLSzszNC4!OiXPyF<eiU5
z)vYdrBn(1gig^ExNaB?u&J+3S;nRnFD#t-OZ*%p2qROh|_m7#Sx}NqOOYFXxYTwoM
z-e<Y5_>(}t#`uDP@2^y`_Z;Afx=D59Wpllk_dJKMXMB<X(*DGrqp4duqk?LeFRm~P
z$}&x!Z)az*Q}g~i+h_fw+}4>`E~(XX+plo9Pd50LOhDC%y-l;0t-KzTsB92r8MSiS
z4o4GpL5twknwX1uj}3f(_LJC9TzL=WOz?NEFz+RAxkYvFyXp3SlD^6s%!)hW{=!7V
z_vhurZ+S2G#zh@=UX#5_%jtbvQBuy+U;L71p=@NUcmFMTe#2$p<Kw@}y%P!(3R+$7
zF>Y21wpK6Dm#%Y)^qyO|_E?RR``w0baw->U>mse6wJck5^zgip>pde<R_;&kTk7r7
z-8H54-KvV+a9NF+3{C_**Tj!t?{v?MD*y#=>Li$zkv}y5z<)nENcy_7c2({5Zv}qI
za>9~6!MM(XbJG$hrNoc5eE;`>V=BQ1!h`G+K4`jlDI|~FnAM}T(Rb~Z=ID=){w~k!
znjYJF$}{Dx`{D~uZn<Sqie2u%^bJOKobtPI&aPJL^T}nK+BWQeTBC96?vKKth74X%
z!=o(AWtI6^LBlQ2u3D1!14SLbLn=mMPj14iXJ(Cr7_EOhtq2uFC@ja#5Wa{9#Oa4#
z@sfQrUHX$<tDHf_U%q{LueWP8v=$DXOxIF3{&Z^N=yH+4ePciE<OQ`RGaL)fq>64P
zZ$GjRnn*L);%Dkvr}x69e`)iOppiSAquu47cOdS@&tKV>pQPDanY752KC4=O-nAo0
zvw!z9J=2*1o0EQNd|V)H?fuigZw<n}HhhN4Y0~xrEpIq7gy&|1qHP}6sFLupPRZ^P
z#$$}S@_a}S(ER#OY-hoiA4N^VxHg4naie8Ci_T@#7923xG3D+*3P-vhf0h-x-b%UP
zt~qTDe6Dpj--MK(yJO%r^KR-Ms>zenx>glGy(Lh(zCYVA%Ke3YsLj`qxbEk!>g$4&
zFMSz0yRPE<0w1%$UFD@MT?JXe@g32-*3>37y?b2tsm4eH$xkY@lmWigFNK4RFqv_v
zybhdyYA6b6@t+`_J(O!p-pe;UpG_`j?I$?qx+-(~A6%d5dNZ;;)XOPrf&TLQdt{ge
z%5IKA#$f}UtxmJA=!crAcfC6Gv})<;%I$M=CxmyH`Z!b=cJ~d;(s`a%Y`pJH&hDPp
zCAYlz=N@f*RM}P(ndyA@_N$)z$M#eLtWHlJTwQtjNIPlek10CUQ<eXW<2!+oMV<&c
zDrgc=gPAE6%V(vN-yPv8AFjksQvZ0gQRjMRlJ#}b;e$J}r{0ben5uFL)I^@=nQ8e6
zM(4Mfb!qjjYun(rp|neQj6W+OKdrb`{ku#+-d;zO#**89!yjIeMB>mlL~q;>_eXRz
zBaH-rDLXC^K@XMFlxQTWx}29odpcLk3YKh?5qYl6av!53r{b~K|HPUfv)*q3#ljSM
zgto#Wz?ioUgcv@NP&Q(JKUG;~e=Yi#*3$pek~n)%6`Xhs585|2fQ&*U0jS<;gxW-u
z6kfffjD$_&_u5?j5_**g+i?ADpF7ErM&uh1!^1d(OG2A!%7^Anox@xOLEAPgk9P|x
zqv8*=;GeY%U5N79fwVV?{DM5IW9J|k3-tTY14S9RI2k}iLRqHeRnQ1kUvN9lAP9mk
zp$0rgOc!)-G^9ygB^A*IGN|U7xe9uykfJ!Ef*1o`bNjE;+q+3yh;}2P>cyipkwT)<
zx!~XM30)Mds4)iVKENj2xkjS6(fOU7ntPTRyceG&0V@P)_>=Xl?MbAGGOAFY${>J~
zY&2P=NI`$iwk<^|`T=$&$%>_OHZ6ted>BIu{fne-?$63UW_@tD>X##-9}<8=8Wy~Y
z04o^)fB^v{)I$Kyo4O*=Sc}oPh%K}+WvCR;cY12>x-r!B!Y|)xwX?y1Y|7teu3f5K
z;S1S&=bn3W@!IWs@=q+L*E+8DwVa_?I;3xLZ-ny-trCBYmb+iV6Z0(MBL1b#Ejy^q
zTQx>GXj`UPjGnt(k2*uIK7Xc+&e!-4)xeV}Ap+7CH~gY9ZB(X#F?<2`!2lv<M1OJO
zMF<*dt=ZC<a>j4DcCywiU27NRT9>Ci-+g+w+kg6G;3rkn8`}7(szRY|*%Hf_mU2#O
zYIyFuY=XWmuZ-K%k*B+~!t9?3XKO?B_RNI3d*Qs&-pbL+-l2Jus7I@Ke%$h}n@Lun
zQLHoJ(Zo|RaA;8P_Ub=bKAG=w>??1Y?VmUs&wQ5l@C(Cs`R4pqaT}f$mZeuJoNxMI
zWEM0_xw%UewX}SJa;t5@o0!0jMH4<T-*?|{Id%D<c0uPa-8D#<2e=&l@^`_v-futW
z|4Vatas05tnu>8#+pc2V*)tFg1qzI6B%w+YRGjKGRn+)B>iwO2(VN#yVC5GU^c>M0
z3fZ;Gx>qkx-m<mnA7*UwTK~)EE*#&$-r_vqusqM@p37gmorS*^@*W6}eQtir(<`d(
zIg(QLInQrZS3|2tu#Me|BXN20O~FPs0anfdd3m39wSNEc4|zEijs|1NFP!ZWjpPrN
zcp@}+3PUxwi4uoH*jHZpAG(qJ<avqB9l@o-h761E>w^xDvrTK>YQ5u#-#De^_2uVW
z_y=m8R#qG>ysI%w+w-k+n|154`}U%;*A=csV{@My9UZ*)V8z|^jM78>dd4&3o?@Ow
z+~7H5@4SJK^r58-GLGB<UWlWE59LaLTBzZ?h|F8=cOcRb<{vT;5s*vBB$W|@f!2rd
zr9OC5bJv5^(hcr)6P9@D^=K|FbyZ$>*2_-SbpMSXb4okPESk(cgewF(c9(jdmDEZt
z?Q5NQId!daMV*t0hu4>c>@!dORDUENzkh3A^BYZ(l0(gpajPwMo;I1Tn)Fy$I8blx
z`ye^@L|pc^o}pcB)<22<^EgQvK@?+8Ze(;nb9JlUdS4Xq8c|<s620XhrWKIgRXw%v
zq~&1q(n<QcQi2Bmm;^0PfAjS0*~;9`Br8sAl)u22*|xG~m3ERvu<7+GjYs}PBj-17
z573%#AsGENaDfjI5=aRWr3KLf_qw+;LV-czmtxZy9Hlk}v0);2etu!DsXYC%xvd$O
zGe%>u0ZzvqN^GFwYlOXk*I@<bnn$t*AK#i|5bSJ^eT<%V82ICD_&e11vEwHH<1bLf
zP|jzzIX~SUwv&jhBoW8dAROY6*N_YxBswf@4CJPQE>HvCec4xwwYCr6?X5f09I2$g
z^?M`xUs|zNKgLLXE5$HFDipsvkG^Y0kFi^n@jGVK6=%|?hs*c4Q-;nXhDx7pUhD9c
z+z&5Ee4AArE0*8wXmj6d5qT#47aLij0Z2D8z!V>`94A_H>NXfqH~A+v_YPkkuIDNE
zsFNFd^Uz#~m$~Ez`Y(1P%;4Tn5!(QZ*fN6*hrXX{T&~u)-?@|SDGXUg2ER4oX0(C~
z`2SXBBOM4>LzI$H?!P0~yThXvy=?gs`jfshSlkk#KS;`42E(QyTwsR^8E|!horKED
z3sjy{X-8&!GVV5F42v{}ew{9Q*-f5>o<mTLLNpnJiQgo{0F6jV?nz2e&d8;`0gj2M
z0B|6|g4+^vt&pLW`JzX_g#kj$T~~{GzkVmpAznMtCGctKAV@OKo{ar1uK+3+1&~*W
zRRO{{RMv?MP9I*bUM%ht$$KFSjt+xvC@l&_q)M^}SlRl?AMFQq20L~o{9|HL0ZSYz
zZ!*_2E=lGQeql((Nt^R_b>*q1j_FPwP66wSB!krTm1`5TiWwyMAlPzH-@45FpmOo<
z?=BZVl1ys=nn!GQ5EjOgpjvSTit%&A@4?d)<zYZd_@em%HRIvbDYL%^YB8cruRSj>
z%%14#tXX`fEw%je=Em+qgJ3&%O#^vd0~N1xf0;LCeLk>h&Bk?85eJN2z2}AR*X2~!
zfy$#$>1+Z;eBR|z*%86;A;fy&@Z*%qYox)OjBTS9wOL%BHU06}XW79mX{V>NlU?7s
z|6Os9QQWomo>`v3MAr|>akW*4pX-J$*mR)OD8ExwcE>OKNyxTGPi=1Rx%%b&MY%ie
z@8fso7+qRm@nl;3Zq4E^g$G4FkQqd3P+1UwKp22wr)Dh_BO*6#qeun6%o+es`vPyR
z3fDPaCCe=Mrz)7~z3wfOl8<X?{<=6lOZ7t0xRB5JsRhe6t*BmccGX{3%}S=E*!K4~
zhx)AWQeV88S90<EJEv8Q&I5j%)pV0qKlD}6>}$_2trf&N?PrIyn6$?~2K789oI(t;
zNTr1Xo^=U)B&9Hddv|1nL7>)G>stFZmkq1_+Sa+WBJ_gVdH<%T4tpvST7DGj+*>ub
z>q6}uKbI=uz8B24V=q=XShPLsjB81Mwy)yc9AWGD$%j{bDXh`>V&bKJ%DgczE~Rc%
z*LjblzIEkQUwjV^6VikoKp8H`w9Xlv4YmO6)!WS;E1~nH>7v}Ux;Ec&YgbtME&i7O
zB)=`zGJUPIU&-HH{wrhK?A+yQOmd@Ux@>WN|KQl>?d6WX6|Jc~gCWU5_D=x+ocBZ5
zZU#66oOHNjwpnUv(z%#R!}cFq3TM5nLBM?n#%=waFl~oNUwvL{fnC|MS8`dZ(wi4v
zE>KllVtgPrY11<wX)UE6y2hM0H+ucnRU2NCO-w%RXgSbw_-LUs@<6h(@h-7%0+%KY
zP3`~bTlVVRy!_Lz2ZxQl6fSCAix_fvqJF|+)o_K$JMfqF1s&1O<pB3@efKMwe_~To
zwf}_caT$@HY)n~^tV0LW-F2;eru_7Ia6$gyUl%pYZcZ(WYt#B*T`N4)<rHjkv5_>M
z53&WYJ`GSxK#5_~%hnv~F)GWbzi;eP&{sG%u%_sQl>XuDF6ZOPB`x2~{ga+Mt}v<0
z{eE?mzopN)M?KG%#5#=+h;)i<UlLcb`W`jABVuY1CWc%Ih|~Ki!Q6FmQ{U2s#zSv4
z&CZ<?6!hJ5@KP>49^ja?B=TRHN9)$vI`@7~aw!|k#MG%1godE8Q*{O#Fr4FyhC^Rd
z{<O9|gBkip6~))1cy0M5TCuG~DdUf=ANDQDiTZW0=qa2MB6Ay!MeXq={EQbTwtw7n
zHZ^C)mic)`1#8!(OMh9?s;{|Vf233FlsDvoJ*SA2IuRK$fAC-0(`u(Noq2}ia_%3R
zFxsv#Uf`c>5xS<L$hTs^+1k-ROZU>thCqW^efC7~d-s6F=tRT`j1l|En5WtIZNB<=
z)c|e&1WSkE`ME9Kh@W_OwL{!z$Rn{P^JOj=0K-y{J^%KRP^8@0)}Lz}d`}Wny10`F
zZ>6b}cmKr=BQ%mI?+sdUaJ|lE=RNs_eruk3BECdCJi%5(t>9?LG$cX2C5qh=nJ@V|
zo9*)gJ(AU$pQrK*ueGG4x;*<N|K^hasrmuMnawU5U7@D>gLWnUX|FdOd8Ju=-s0f;
z=jY8XHhvvjH2p^Z7)Or}7ICG&MlR}d+N-75Z>Hb${BI{&vqep9npgVddRDD0ja`wj
zts(DX({o;tN?PoS`pXNQ+%AlMB=~8X+k!EggJ`%%9|7tActYRG1h8^2@MYis&0c~U
zfYhxW+uB@s>E*3e<wMeb{n@coy0n7U-NAt7X9kYhZ5DFPt!C?z&z)Vj|AB2`g09!a
z%YGJ`F}$mBdHL_#qNb{uyp7}+d6lPc+urj3n0ga<sMq&>yx;kDPEwr{rPY!xjO{2C
zg^zU@S!T10tw@E2va}HXN0i8#p$J)qnPDgy8kCPpLK0@kSSnGLC}qpm|9-~#{{KHO
z=Nz4s_w#<1`@XOHx~?Z0qq9e@b@xvxkt5SQ<*(fp-`}`NFz6Qw@HeT9WZL`^HHlh$
zvWIupTfspu?s?%v7Ws{$q+VP@yo%i`M<=ffn3{r>;_ZC=JJNiCMLqUOM_Fj<u$cE@
z=~+iP(yo!z&yP=x0!NyrTfW@~ksqL}LMv?yX0f{&4jX-FZ~^%PMO-wq&*2RqPRzgD
zZlD6K5MZ$E!C4Jw+g3p`^yk!Fl0y?L{iX@S3e1XId1aC6QY<_nm|2lCph^Wew;5k<
ze&kzsFZP75op8LW%g^TQ1?}H|x3X=US&3WuwJy%k<|3K-a>c-+`z$z*#PiM;e^mbI
zcR4Mh-KHw{?(C6m%aL}K48G%vb@!f|j<N5~JC%f3l-#}XKXdlZ?^l7jiBTh$TqC?a
z*vw;c;L(MV!|qX(h{@siaBwI6ql_pxz^X+Q)k|xEZPf1NS_6k?oE1tt^q;vF0)h3E
zNUA7~9MWKnU!beqo5{?1b2Elnbqh~;zIS1euG>7h<*bBmXD^*v+FN{{*Gj9?slrc>
zjq7sUM9U5IS8*j0*e^3HcUCRp`?@X|{3q+o&tO+YLtk1KyrtQ_&L0gdzMi+>)85bS
zRat0>2pv4JWS}dE2;yDp`!F-`gx##}l~;Mr#5N~yWWV6`PJLcshGAK^rrOOm|AM)4
ziq*;<)cSxh44_j|5^SJbjnH>gSV4he{HpFOCZIFp;_w`nNZM7=ec4gb&@q0bvq*4c
z!(~zd+b!*a51{^MdRpW9oF7tZmA`1TMYIX<Kgiz`p_n_goYK0!HC#96o=xSI8d9%S
zHyztAkUmg<>a6Nli~Jox-|w}4PW;jRJ*juBmiTMkzFBFg;F88_NxqLlUR~s?Ym>~#
zK~gDSA1GIi3<aeZbnIJIJ9jhirez5)q7khipie|>zg4)REyfpeEQK(-uyDZg+TH%3
zjyoJHXf5<R{%hTtmCCB>7Lw(~;0Vl8%*e5dTWss+pTcSh!8>EW@Oo|D@PcD29KSRL
z_GeO=issJ<d)a2`c-L(@s+0bm8JVmgxtAl2>z{si#85p0@gs!t1I50ERMknZy`EjZ
zX@{f>{B)%YA|<G4%ifxbA)llkc1@oCxq<Qa-S1oISU1OU-%Dd(=rxgxN92p>?)_ws
z-~V`eZol8G3<qRGupQtNacmOKm%`V#=23GRkd35{t)~X_^HoITEK;(?_9dPmm^}qX
zqCMxY54%!){U-A~MR?|cI7Mya-M=aPNqw2tSe#z?C$%xz`FZ8fyNRTtJ{(hdpuj~!
zLgB$n^+U=xHj*<E{E}mW;?lZ@vH8Rz20cEcJPm8zNy)<b{H0pRsTKVZBt0}6J;@Hg
z$=n!}L!>gvio2E#3+dX|NWSl;?)?nQ-{x%>M05FTk!E%5fd+}9-O{?bXxKL_R;S0u
ziC+0@oq%MrA~9<4JuS;O1jYwLgIp7_(GdOvEv-T{+(<KE*w_Jy`O6y^BJ)yD0jEai
z@uoL-Gj={5j4IPKmd@;tc|PJ`$Pc94-I3?(=Vj)*KaPjr6RXFlcebhfn00<qdrlya
z*CwTiZ=LH@GgGrsGZx!cx>E8Cx}(DW2_)r6)CpO>){+|J1H$SgJPn@V%KF{ajMce)
zp9GAG7;>uvMMon%Rzq6zR-b8~@q*XhAUC`1+2<ZJ>CHX?IyRMaK8_DBjNz?a<w|_d
zaUCwV*rt|ylR4y3tsb2|yle5_&Hqenq#IyrI{H*fD;*<jH`MUvcZipnM+Bsl(<Dbd
z&dVm%k^}c$9XLbVip?+YZP(4n3q6l%2vv^PSIrBseK>~^ru)zlLnu;C+L%xz1o119
z0LvnJNf|<_FPxKst0wsgf_yoWiZTGr?rpuR1DG!b_vr%zj0r&0GaUp2nM%o}w@*2j
z%ys!^VI04-#iteCAAjVnDb11SAv*h*zZqh!TBZ{(>e*#4OJpb~)11;foxQ~U8u033
z-V%(0E3rKDP(fNuQ5?aE(xIzZB`5N5&NjmT`h!k`Zo2-(O_oJd5*Y@&VlhsPM@6Zu
z=hsR#@k?Y7C<%l9t)kG&LeiL#KlHMr_f|8Bp#?E%q!xn{DwF;gW&ZW5!gvk$mtV-Z
zRC&7Zm$UgKGAVxyx+kNW#wF*`ARGm4nIvi%wp{`;NKqbncCS-is+N&DHLWgG^OU#p
z*S&x%$}HNpHd9#7l#4Zg_fY0q(3y6x9iR$0v>a8R825e4>ET(a#!0M^=Od~@tLH*o
z;#tUTqVH9#ZVGgaaBQ?b^K0E9{L)Sm0ggHdPMsq1b5@I`J_;*=Tl#pB$)Py>q$@5t
zqczUz)!g=ON@NvJPPA6#?6%Z`TH+}+WAP1>r(ye_Y9n8Bl56oJqQ~yDDRU?s<VBV3
zwvjBgj-GVxrmLrxvgqqSabJ?pZmfFfUBQY?zAJI1rj7Hszw~j~K8RUgx>eVQwz{1k
z4;#EX9%z5|(3hRccjJ<-Pmp`m_d5;lN#wApd<n%}qnakCs^Nj-p_OVZ;{~GHR69Yl
z{xHHNmK63@GoBcS5he$^_sz*|Oz-GFl%9X%Q)gKVoYYc+Br1@vs&Pv$56v#>IijVY
zi+`({=SMxY8?VN2nC_S!^N@FaK)XCPuNtYz_>vu`$9zU?%U8Q;RgSN#lww8RD&79=
z*E;-wigZWcmlSWxEpkigdGeby=NBXM(mYQ7gY9!~7%4=?WZ8RQGe_EkO1!-#DIVo-
zS5O#{8)(xliMBMeJwJABlHMN{m`yX1)3Z+Z{o=_|9M`IA(IIboQR7h5;{T7dPA6wh
z>({yjj9hF5#s%*hOLE?mJm8utuwLJMhzC7q3^l(Xm5{z$U&HqWrdV}{vnh9cy#W9H
zT#xPLlrX&DRp?c3>bB*m^2D5mCz1HtFFjNx?;J-q9~;EuJFIu^ESXfP8kCw11+FYL
z8^B+{z+r!_LWLc-x$pataqfp`S&^N}31+e)k{NPDGc_x-1U1t>TgB5laa!p#0;T_N
zKiMfk%OwlUW=r|+Nv*q*1wNt<XLB2DWczjY20341e0XG0{25be+fs6%n&aSD;oAAx
ze7_(xx_x@i&rX*;Gxn5wf{|DwsXHlCm8wC*lP4cpUz-c*F}+(#eeC$fgV{7jK#|}Z
zwh6L{MAsjiBZ3_T`LQD!&R(Cb)5%YbFj~*D{o{Kvu|LIn$^RT=m2QMXUF>vXut&!7
zH<R$815!U^9zbehXt>!Ah{-bhMHYpT2GBo1AuX5Az;7k|abY}?gFW$UT?0|$6kmCC
z)$yXm$Ttc=u3fY*j|#`61aYoTQ3`+cSwp^A?H{*HH}u&3?w@1aRY98)MsLAla!)<T
zOH&Kz41W6KH<XcMoR0$=4mX)3cr-WIsmmxmV#xPv^~k47_LQo2caPLb4>2!ghDDLK
zsa!lfd#zeM?B4tir->&I`#+m^e09kR)`84&RIFW0U@&n=xiP0?PNM>}qT;qtCZP5-
z?tFuc;lI!_hzRrvR{HIue;dmjheam&z_H)BHR#v6>yxJh;k<iX%neS-Pi=O^Ev`iK
z_T-BJc73=km9w4DH|pX9=3Bc|JAO#hjq5G(@HX0&6Iz?UL80mzzs-2;np#!v!MI0}
z({}A8bDbAB4{Xz)-scTRe~4(IUgK7zrAv{!3&i)ZE%Phd7e$}wN9aeUjM!qs$>xt+
z_}tWnE^qDs@;|@7sAFb{Ic`(={Xb*4Z0r~JXsGd(x*7oT=>)cC`{wYY;R;)^gZtVZ
zV-@&w$w6v+Dg-PoS@uZiP4hIa#NOg1y}W8SN3UJ?7jgDqrg+xflnO*xM2SCb_qiUH
z$S~<Wv*Ox-fklg-8uyRWJjeb>>D#~76_{D#q)mAjPQ8dsVb$i;V}7msJwQrRZ2Yf}
z*QyiKeR5|)|2ZQ3&O-K3!O+Y=qnSrZaVx=#XYi1XrL>5b53%Zgt!va6u~UK*XJTJc
zJOptg)jxD{^WaP${i%E~fVe`psjT9+{3xKz&hO;>IlC=Ao?dXJhTLm?LQqm`-(4hd
z(Pnev)%f<IPWy?DU-BN}mO%H`ft4tk+F4z+Uv<LtWoSa_^uy%@XHBM?VyUx9ePTxJ
ztf?l+rR+WXW5JNo*sN5S&)Cwhb)BT?H`9yZ?IFe7w8>cw@7F;mMnB6}MjGI#59)Mw
z?6!GW)HL`YH;75}TsaBZ1&aTW0iaNtE&oy26LymS1RQy=xqw9EB`g-B0r`PApB7#Q
zjdG~zKX|6Z#<(szH7z4Ki*<zar;r#*t0GR1#=j=FIzM!z7F34OWSkY#mc1ZRE#Ucg
zUYoz%_*Ca)-U!qfm-<#M;)rDPpnmqulIWnmyFBkk^|k7_q?_n_H&)(Do{vl>Puou~
z`cErfGPkKGG+F0<F1usfe(nXO&YAJ4a39-zU&+Y0ktaw5)}D8*rdYiWzXVLQ8^Y^h
z*aC8p0|Y6y^pZ|2>)ih^rhFrgnjqbb_hHraSv&YDJSfHE$&^-dn3m)<{O*H9Z`EYl
zHqOcPe6CoDoW!<6%H?E_kK`7jwfkp^(vgWT#UG8{IalcP$E{LS;+TF@mSnC9oo$PY
zWzfC%KCEV>DL@fqYpF7n=h9^TA~dh|lHVo9LjSPwUB2I(A6?>2nmx0Xf@^IyH<Hr8
zvKok(tcW6sq}uO#<cVJPMZG}FskV^Us-)`&*S}81Jk`;>NGNLj2pLV3QSmSunGpZm
ze7<4*`DTT0uoJA560IhR4)^F&?&4puu_DO^8omnJ7C~jiK7$A1Vq@|?4X2V$-mu-#
ze6Dh&`Pc*84Rue`^#LOo#aBhiPiy`6H062a)rK%Kv~HgFen#L0y56v{pX-M6`R3z2
zZ|wrG+77DebUiazMOKpUUM84HroQvb?Ad(s%y3#!UvSugG@hSi$0KtSiDxbW<fWaK
zvf8;FsE;dDL7Z%VV+W`qJoq<27<zKA!9v=pIrqH@F68?evlk=3y=bI^9PmkeSca2G
zwhF|Yym;zy<G5A}Yb4sUKU437n74Ki+n|FZc8T`JJ?gY`_pqYQ?G?LU>pTt=HLm;_
z+?=Dl;M5u1P7skFEku1>goX$2OJYC_$eqIYD>zuT#o14hpB8n0szR9CV@7F7ZD^o2
zvnsI-$XXve`r9oO$lfeIw{GLBU+bvrBZnoc=o?z&#u<u7UDEX@RckjpA8%Z4bTrv7
zNaaw^F~*c@h&Nh)8`w#yiJUNFu!3Z5zLTA{J+L~-<iyUZ2X~os;x$y&HujCX-q$ZA
z`E>}oUG1H1H?Rh~>nv(;iJx3v=!bH2HD5qO*Xt{+TKPO-Hv|MsM|WvKmsF?(0m=-@
zYd7x#>@yhvj#OC3<mIx^dx_je$gc_R-pVySIOSPxn9dH>{!H@QmUJyX_l%Nitv!F~
z%$LrpmIwpMD#w&bZmG2kYp&pJnBs_cDz9j!dE87u|NWE1lM&pNnj;c_+1IDUYyVm|
z*3)%yh`c=F(CW+y3!)5g98?y=3nM}&exn-b=M-!y{@9sBb88>heR!Hz%45v^z^Lo&
z7Z1Q$NveGAxX@|yH?4-ItNCA;X`oL}7UpJMl@f9UQOj5XVE=?bAr@!~%$Pt}1V{MM
z(r<TF>lUq#HftpBYK<{3x@ks|c<3nD)8|msts~i(|F*D-+vmhrJJ*akTksIqr?M~;
zgl~-N;9T;kNTPaV4i-?2WSCzBOfS~Z)e};Je}1jA%%PCQF<Q6?qs0JYiJf*f@8Vap
z{nWI!ed}=5)1V$w*U2_;JjGS#{$*3o`{I?dTZOK9FyqH!LTT^p5lm@V9Pa{+S0x}F
zIj6Rio2tSM&XSpGu+A(d6!BLAVGlw86_FP*(S+^z<1e_VjREa-nZkQ*kL~6Gm=H{b
zQ2ZSPrS4;<wVi~xrwKoN5Qp(+w%|GybG9&fK=lj#V<(<edc$vrOUc}uNw(A!gC9`T
z5}K0M+2SLC8`M`_SixSHJe}Q4r6k0iefw3vGtEoAGwS@WbzSVUJhSMc`r9OPwNHL|
ztHm6BLhMTezyR<csfvMNzOvX?-I$sxr?y-|DOQ#4q2YC{kRQ>qskOn^l{B({SNGto
zr}7{r8<*u+uCR(=rL|rR?AP~@<^-b-kDn0}&K2{<fM7?}Yv3WA%(~0&HN^(<r))Tn
zyK{ylE)YZI{msQa@B<9PP03g}ny&ptg`%7SZ^h-%BdL|NjMR)IN9xQXH0R^}e7fo=
zBKTqExW!~?XI2?CWUsmR3|Z8l3w7_y5beO8UYk>yzt-I<2Z=vS;4G&!q>WL{%Z$Bg
zUNm2`+ongG1dDT4j%r|(!MikeX@)$<B08TOWJWKhmqm`wZhBoy$m$?V-msB6N?y{8
zOYDezIZhH3j=`$)nmIcr1qmJSN<{2>?V)VuZ0F?5?T?Mk)G)KO;Os3(k;?D#EwAiQ
z*~e6w+;`JVOiX)j>)f>i{6N=tC+v=UNzg3@yE{k*#hWd0rH>7&pAv(=J&x<?|DLIu
z4IJ!U%wOv|#{ZZM^QUf>I9gD(7+wjqrhDZT83_{-D?IUhjWh>oj+jK5`3H8(46Czb
zM5U@5@Y8!FZrVGqa?9eMn9!WIxAxx`6Vg};>JIsw9_fNhUmO=%3?|;4QF*)kLK5`O
zaV1*{{#L6huIZSVGSqXLu)9px2+hXTlFf+wL(tAjyt_R1FC+ijW?}nJX@2|e>caW|
z2)JJx@ZANfr~h%>m`47`?>|k0nXHDd`}o}7S+Vs}m*m|WndgSNVzQDixMzJPtIdjH
zS<eA8s1r>MQy0&azxH5HL0?2-QKAcBPan48N5vLTotD%v8-t<*Ghe;fNr~99#oGN<
ztkeW+d+2viYcN)m+J0g+s$zCN`3g#eGJdBBG7|vp4YYpckL~I_wyX!vP-yEQh!A;N
z{a=kxFueG2FX@I%9&9J2gV`@r5V#+5CN50-qW|g5j#_T7UJ$Ka(~T%?>2w8$hpBXL
zl0lk7CmH)|ojG3s39lb`DZPHARk(b27qUzNPAFI^0iFSRuADT;A3ZnsbukLW_m(e?
zAJ}VmDlYX-A=~0{RBQ(CJvq84qn@|f<3+pCjGIl#8zj`xZb+<y1ZY^OHuxy)xti`7
zew{K0UI?@2uzYw-MW?;5R)6oszx>qITfz==DzU@G0SMGUKIuu}Tc`Nwi5hy0)?|^=
zpgk@Egml!yf=G_qS*B?(DvCn8I+Ius-~ro*IwPqE04$oLH)l^QAA3&ca$f8~c*q;-
z2R`(ddT~<=9I+%0rCJf|B+xJkcFf<N3JwzHJV7CB<Xhe7eA;$Zzlng@OPlX>Mw;gw
z1~v)&o{-!rR-`X-Fu;n};WoK1!QvaK_?*B*VSkyGpUW0f+MN9y5!=P3q*oPa2HaOX
z95}Kg_J!cxXie?kUZ2_<gaYKnFB7O2nhL2nv|XFm6d_T-hH7dw&WbD<KMMvv5awqO
zAk{YD)D{~Ra-X1~U6s^>rp5O+r{0Y(W*I)-Hg+<%0Q%VIp?lhQ7-j?Rs827Oe?h{I
zmtPkRYYBUXk*}oOf3iSp(&x=5<Jz(fx^q67g<{IqlhDhIs9JBU4Cn5?BswV2cEO87
z3R2W406}ADZzpPVi`(_WF=ji<Hg0>ch~c_6a#{AW8|aBs32lD%9LQWV6TNX((%CIX
zIQ!+>ee#KBlq_W`#7%btslMKGjByjS;@=(B`pBXIA+vwKU<1j{B%c3t&THJ@RB4wZ
zE6IhzPtW=MNwSI)ALO)@7FKH3axQGreE!X`B-Txj%lQtzjKx?u(`K}BFUw@er5@o0
zl8M-w7gZH1Q<dhq52@C<1ts~KPomVm@fbG^rfC`-vnj>(Y0l%kK3H2OLnzxzjmQ*{
zr-Q2q;tXF=`(FlGAwY8B6_66#E)V?kGqjaUd(EI+qo2LO$q84zG^+WLY;yIu`(Unc
zOx1}=NmzZ=zpSV`N?aPoZYc(1E`=N2yr_P9a;8p#{yHyE#ksr3w>8{4?@8|S`a~u%
zv-<bblosvI9)d4@!>JmZ04g14wLLeC%ouuz$1G9J$<dx+9D^!prJm_duW(M`+x+`q
zAB1BKn!|bH4V3U2QtFN@QvR23I`ITLTV|pEv#Iif?VLXXdE=zMp4=o+V^Cn`-Uq=_
z@I93LL<`RV;fx{#w;<Jb={Jr~c`&Qc)Ma7t@tyM;CRy6f^QZD`NbHZ^I_D~L>@Vzm
zD9vD!E%ZMuouPJ?!w_+=xSBzKJeIs`nDvBqC0H`2KW~^dHF(F$ygP$7A<;6X^H`tF
znR2wBb`<b4gD>YcjO!jeE;y6K{w(vbt60U&NSTY<<eTK(3ia)7yXB<&Zk2LAKi}DU
z(6N+u+tfN@xNnX=;JDDcS=qiz;SmJPgp=7y*B~<xksq=JSx73%KZPb|5lIT)fU<W6
zRZjg-<9Sn8C-Tb3VO5KP`8a~}nRkY?y5K8b%dF7wH&MqU+U=@wwaN>#q_hxa3$^O2
zT^C_US2d^>Z`r=u?mXwvB~gSdQ%@Hkb3CTwUG{6;{u)*3Sf<JK8w7%DXBIW!_sG3R
zLql`rcnn2;3UiBU`EbWfhZ**7Z?p5`SI(BjHkPHo@i%Oo>Hbub@+#!z1*xAb4x{^9
z&lSZTCK#L?H-3thkQ>G%1_0R|pl8G~SkUTDaD)NzKxB?KmrufRV4S0kmUDBpBpYeJ
z!D!9^A`Nr(C&2usDqo{2PCU`yb4bFX{Bvh@V$`Zua|qeMC>-AJB>Wndq2@69`u&W6
z6jHDpk=sPBJ`g{$vzq>u?mg^j5Vyq3XOv$2slyzki7SUj?sjZ@29xvxXU7;kG|lg<
z&#>T175rK!WrzPrx2X!FE-6l0&2Rp}cVdNJ_e?mlEvD7a=TLt{vWd)02!@JGpg;~H
z#48V$ihYBI+LarKrUWwL&F2t96j&+g$siia?Y@5;bac{(m5N}-q-@Bmg4ox1c8qji
ziui7us?oWZ+3BezE7c-Re^J|Twql9bD)@UE&0UhsgSm=qF1>xv2)X2=U+L*bqh4hg
z9TF~CZz@!)>fzN!vTwd>imx6q^;$!2Vnv$tO^=WQU-=B)8T$WuH!N`_BKHO9$G2`P
z@121-IpU`hj}}Mc5_QbwnSz!Nzt*LeGp0MIS5oKv(zCMg;`we41@{H6Q)x<Awe15J
zXb$7CY){ur%f(n8{naYnt#B)r0*g@_#39ftFtO^~WdEW9mr94e8)_zs@pAhh<c<>i
zW|O0p#Pbh3=va0zMO4LH6?f#cnt96cycY}boAxt?)4Ql_$!6yB%JlGJ)4iS!p4?|Z
zS7Y%)mB#~WGdt-zj~T-%#28lGr$d{U4rxj1W~vH!z9jP?LR3y`TCpu(x-az>^-QJr
z;D0N5$2M8n`K-6JrMuOMO`~%fG5m~yTO<1?0ZDbHEK=>JAaGpa?XLdkttMWYmEim#
z@Tq)`&NRJ+m4P$59p2fbN4{0y_FC^a{Qv%Z63KzxjHd?F%o=>&9P;vUmIIl;&#TOQ
zaw=yO3|Z#NYp&O%^0|j4sA8iv1p@lZi~JpzPS5ToM0ndgMk^R~_79f&AT|+dd4sNz
z2KpP+ZTY=WY>6Z`Yk4qWQ!Qcg>B<Ocq^Lp=VN$gIi-|`;&-eTmhXLSfcx44@oec+k
z{aiA?>fVeQb$io^C8y^O1^D%r60sw!ZVua1m6BfkTlu43>%2Dva{E*&hUXsGP>Boo
zYK*)k8myb~Z_>Gxk)ZXHZYS`VE!Zu6CEeW|@BAY_BzANYollUvV=HJo_B5vRnZDd9
zlUtv$WS4Rm=p#!`m|s-&$YtFYTOO@ABB#@HWOLz2&?DFT0z2}#&#L@gLz%O_lZX~p
zCd}%=KN2yGL1YcmeQ5OoB@84{+0I}abTix#QvQkNqvx=r_WRoEG$ySBw%9t#ci11>
zUQMedzP-pU=+p60Q;+yM-SL+E9Y{kb2D!whu{nt=!B;{T6DpUlXkUprf5kA@$K^8l
zO<X&5pe0A2tk!uwJ>Ta1SddTQA>PHIS!vB<`5!8e7BgrDM;pd#L%oMfJ{AwrUr#TW
zhd%S!zt3~kc7s|S3z{&qGEd#`-Rwzjd1pnVzIyZr=Ic+og7RMMncHk$NwpIs^TTKb
zS$$~VEY>Yd$bk2U6wL~`)zA_|=a!M3an>cbs>DGF-@_pX$Wh*Vl|fP3Suqb<(rvG&
zaJ0A_{Uf4J09|v9WsMcJlHl~MEcximjipH*Qim4B>ph$3;q_yM6V4Gq_>p7$visfM
zNz)u735m?$qa4Q{e&g1uhy4~9f#r-|Q;8No9hzP{nUi~y{gM7*MxcssMc9whPf3Z(
zGGSOA*X7y@cG~7_nCGljcSZlIgcC0O^N;!_?D4pG52oW;=BuEsLqO(3Lbd39Nhu0>
z$YTWfb*OMQ7K(v{WN@L{$jfR0QYldRG7eU6s_K|*B2v}UI(?fqEL8^Nlz2-{bc2+l
zVeF^sS$YM_R8Y~eq)f#ujG1Tld>N?I*0L|JICZ_)I%YWNguQ1Q_Gxw3Bn-YHR~;wq
zJLvtTdS7GuV%m-jc|88G>Lqrro%MHvZNJtr@;j>}N`k^2ec8(aA%i(d*Skck^2shG
zmzbBH|5_)1m3MLuThVCFHKhCG$M!ntowoQ?zs1oy8EPSy%+b&iLdyWIHX?OvLlW>6
z5$zFzi;$B3Z`TpR=hW#upr_z_QbUB3m*g+uVx1~I9Otu{KHP2AM&0&0|HD_OVys`b
zuiR2nNmWL|m^$pF)5QXEFSd#8eeL1tEizi;7Y;2)+6eNk4fNxqy)?;1RUIzwlTWe&
z7(Qclw7L;*v(v>{2FA<hdZzK;I=rJUkDv8@f^i+nP2y8;6pq9NA2rOa4}2o!(rEo6
zlK)>Q@7|zhPq>ZNjh`i&!_xO5PRD!+X}-ML613;Yaw6TWUL0259fcXK!aE1hJR(XX
z%;Ta8I%SZ>0k#oFH?DLQ$s~Kw*aVp-=N?yYvZ2oLR(HLPDEDrv&`HTZ^h08|MyU0!
zCh38eZUNcKPkfjB&H;>kKfy1>o>`ID<_Fm02S#j9ScwLiQF6^*aF$JMGCf{Dd?cWF
z5~Aq!TfF@)&g!Kjm&dw&7Ry$LT&A>-6>*LR*Lf=aT4$OK{x!wU9^~Fy18Y0G5A=n>
zGBaE`mmh??^+09&&xvoOVYafvMOE?caO^i;klPi51jk2Z1j|#j_3wYJyBxZ2{B6-)
zn8iycf&C2V2XdQ*BNN%{!1!j+p^8eH_lR-{T8c4jPs92v&v#dUvI-6ElH8P|V@B8-
zz1@#$jF%uQ?n%${bk|`F?0FlOidT`_L9e{yNiM!T85m|MXW~1<P3!Zsj#PB$SMt22
zhB+|vg(m)qNWIBuZ9l`rIWEl6+fHtw*6BD0hWS?u+WG%UFlXKHr|k`tGdVMyvo~^I
zWrvxk5~b`VQn3NQMne;9B%xA{GMs{wW5^2TN$72hj5X5WFhMy06As1WO%s6A|88aG
zsIT~@?-BXU)XCk%qpE_`*JV$2{B5LLuPc^~@<Q12Y1ratTpe-}Z(6_IWVn0So#Q6t
z>O{-R0QVGGn;`d>wKvp)#05Bjqr2dc0(_Fx{7U%duKpnub-=rut=wH51AYh)?$0dv
zEFass4_e%*%4*`72BQgdysTt=$+t-jtF)o=nCR>GJzFFGs(SK?<{3#irccD;()Gj=
zxrZa;+>KjUU(b5cAve9^qqq^tT~S~=D3lB8b7et@=+k0to&1Pvokr73^n?@$e5lt*
ztZ)FKxeUBp2qtoH?tn(tSmq(b1I<2cfanud_vDhP!=bF2m*PIPkH!^sl?A?uu5_H=
zcIL|$lp2RYOv9gy3F(-fm~V?<QQUxS`Ao^c)b<C8R3tL}zLH7IR|az#byS-j_h>tO
zo&*Q6_mX>PQeJ+f$Ij%lp8fB-I>_xYl;(xO*=t?L%;*|QYB#Bz$T&SB)tzp0H&Ce5
zRQUIp1OpPD(4FvK^=8gZU>zf7bmA_c=JM2G?|pzdg;E*{YBW$up^-@5#zs)9B*Oeh
zI>ILOfatJOp__6zMCW*qYuYlfi2xZ^n;@Pp?qgF&BBd~lgn7y_{(AtPlz%=p`N1Jb
zi3<MYmbV460ts2$G9NhKzv)~&)}LXr$;Mv}_o(!ufhqov2UYvG&CHco<E4Jm7@r+I
znyinvANq~WG~W;jXQ527z^I+ae5j=L$~2yY4hrFnJl@pWnwBBir1eU;3{h~#@{A?V
zA=N3!2Vr9=&8jU!k`0UG3oMe$&^oWIqmxI99}GM;qRLcV8}JKIT^Q@c9(rr65_B%~
zpIV$@Bykvrxn41JRidE&CFas7;CN}X()xx<_H7&sQpT7|YccanyUh5XUhd91drEBh
z%fWM-e|$46WMM?>SN<YV+;f2gWJD-xjpuP~y>4@?tvA^VeQ7OF^hFOfR+t*GV{^pl
zRo!p(RH?cHIWu|W{>r{Bx&HK*<Uov;Y)M~NnfFfY<jh1XJG{}NzeUqB&7k?>28PpK
z&TEcxbrvr`;<(q8^n_ot^JsU6rI+ImiRIurlCOI)<Jm2r@uQic**B+Gb85r&M-Gmw
z{#0EU{j&FS=V>gS+id*o_LCuERkl<Cw`3M`_Jh!lCP?!6VNN5mQG-<=8wLNsyZqN^
z9%z+-<m^B8Yn@3mJUya3`6d(lK)U8dwcXjl4gW_)v0!XaQLM7S!|P5O8y}eS!l#Sm
zCf&*j`JcDvZGmTYyIh?W<{v00r)c5Mn-^ozB#G373*n{=>6UiS+U>X`z44mAE4wr4
zr_<TE?8>s-Te!*e0~G~M6J0a&a-AVlHQ73$zt(N>@ChHoQ4>_p-h24p;a}@U588x(
z!F$enJVd$Kzq=6G;DPy2vT>9>cT+Mfk+l^9>jq}R9tMUYD01dPa}ntS{|1Oh$j=8H
z7t)43fPH8)_#ep8eu9w?$L@WWHzQr#l#?Kae_3W`7ZRl>#S77yk65oX?%p%ff!}VS
zqeDs&+uN{FDG(SznLr-}|6OF?6`uy4`L3v&6N3!2f_9_qWBV0&ooQk-Dc`-u&=Wu&
z25f%Npg1E?FF~tN*iS^@+etG*&fqP+KgFI9(#%@L)?{{BsJHR?_gkKS7jRsAi-`*z
z4&qHD3a&JgG2KNE^67fGsI>%6Lx7G!ak(GXUW_{7DcaZ#QqGH3P|Zb<p!bz}T1Y|2
zf#IX=5AqLTm-uuZa(V#f`kGBE)>t$wl&r|vQG>ysJw$TBkIzzNq~is5Cfg2q<`z$m
z|M3ON((8%!w|JNC;3uj28Icx8Jo8SR0ORv(`W042`a19s$cYZ}Wcd$kX559h<|+^-
zxZVf#I8f7xBrGXmLUzNCvunc=&*GhJBBvdjtee=rfq16K<dfWL#x{Q_`HzMyCsJP~
zSmYOfUX6FF-(fW_^i@J2*?Och$2KLZbVgh=1cND|;&x5i0)#9$<$f}VeVrHY3^305
z%TB2j9wf2l@5b?_?VVK14_qLd9id%`$iA%VJ$_%V#i(s%N3G*WB)Nb}Yi;g8l>$f|
z*hisY{S%r>SKjVREQIMBLj|#jspRg1K#lI}3_p@Y?Tk%$&*1xn<q)$%!q=;GG!QZe
zY%>QPeOUx5TgIj>MD(0NwJt{PKu%;n$Xs7TfI$iw6i}d#rH|!-j`UxRE_hXeq(T3)
zWfvIg!Qy!<YZuU9X11EMilm7kslx5tR&;Fn#?RZJbAe135~VBfgy1i=*2hKQ$~7w`
zy21pQ&Rz!qHL2Dl=iT_PbsvtcoP-3+0w!ZER~FhULGl6V3All20WtF#jC5`8%GZZ0
zfHq?e^p(HXZJN;oV+bVc0NqJ@6M{Do$A6P0IxYg?PXn<9N$VSAoE27XqAg~bO6cR&
zsM2=qHT9<WPEQ7rmCKSmn)Byoh7vwUik1w1J(XwP5_49oG%UJsM)?U=P07I68BgEY
zkLnmzHC43(x61sg-UM<n?s|MYUR8rO$TrfhC@K38j&++NX9hgp0N+x6g51)r<$sFi
zUi!X-g1|!+TlylA8Je-`z6toA^OSCOThY=Clh7vRCPw6K3Dg7pu_P$sf$yrDjPtIr
zb`{mTcmE2pE-%&Rn5nS1Gz?4MH;tv5!CcmC6t=2Rt#sftfIEROtTl=uxC#Jfh20JV
zU5@?9?G`E^j|o#4CrY$z3#3pQVvPomaZf1z>7m!<*jr*EAkI6s!tfitp{4PSJpIPj
zq&&SoF)7yLyMe91XZs5s0+(#i(-lxnnl;qKEd<Slx;HVWvg3~OabI>nq&zA%*V{r%
z!fTmJwp?dS<EMSbIjhRWj}|Zyn1UVB(BXC5u%a3IH+v|*KE88o9GA#Rb4u~T`FT?Y
zF{%qs*i#``ZhWXJqa@o;g=jv?TjqG~Hr2p<=Z%-+gv%Kuuh2qme+i>PlZ#HwpIUSF
z0Gw(gdGY!Eww{uZih+ZkB+7e)&Z9z2AEb<8IG0e`-jGvzDlGbhsgK9SZ{+WOV?Ao0
zna>=XOi#F7o>&+o`4Vw#eBG=caN2w>PyGCg__0bwu1&_~?X1ZK1EZ|C>bUDkWRJG-
zb3^rUN%3~X^64?{S-0ojukTG{35Vyt#wK~(P9T5ln3kD!0K&tHOQcl5qI7G^6NS=*
zfG0!Q)?G0Uy!^ZG8aJ!B5kgsTt|ecGk)d5?Eb}MKVTjPxYrP*LSU<nhWyEPlN8rg%
z$ny<C1)SXfny4B}iVIUIVTQMUYm~8d0R9R;tL-d4ud@8z0_H0}fgH&3G1x(jDo#v%
zTAy0=qQl(Wn`TF>%1!tZ&O1H8{5aD=y;PBu1f4jAH$`<h84sYFjQ=t1FdZ5x<!B&D
zlL>sE_j-IxdsNT0We{ViZeX?SGgh0!u;m>s%?zm@hgyQg;Y-hEZa;eL$!IC*5&J{l
zkkNR_$C*{5l5p)nmU55Yj^f0t;`;*B{r9vkj1~SJuERCpVe$)VN~TXj+7Sh%GO}6B
z02SJK3@){%0mObsA?rpl9eMKQL0*r33v@z4XJes_JRnXeB7j$4wwcQ0xaOTU?4P|b
zvGv~^-jx>*kgIj$ANfLlp?X>MOTkA*Ye&4wc-TSDQpH1V<Ek~?x~YlzAG^460Aa?}
zQa!BSvRXTiY_Ds?Qo2j4DTA7JA0_KoBqYSxRO9d-r6PCADjK`vNkOD^mK{mScjHgY
z2NnJCWr@qGgpg;0DK=GIGkxu3LHX6L1=!^p<w;!&t7Ta<-`eF|0_mImbo{E_Ukuk`
zYPiXd5Jx6xa{wQME;`KdNDHV>kBXH+Kw7~e0x7Iylqw(?Q#|HhjoN19=v-GPRQZ7k
zPUqHWP*gxr78hm}u`%k8&A%u0nfkr+aVRYD^YIAtlgk-D))E@tX!c;VB9nz9Mpivp
z+-x2;ahCpZcQc{DpNdJRW7SP=kvS`N7bcZ2mn@84D&e@cTN=zs_*v2j4e#Qr>pBZo
zn&c=0t{J*n9amW&#KHw*YNwv28<<B*>`A~~tn^E1V%l18<%XkoJ?Z=EgCCq|k#?l!
zr?_0@P3H5cM)q$>lVkHaB@a`_4YSV&?rU#$FHekmreV5imb|AlvZh*2{i0b_)|VY6
z_VjiVxEO^WSxk_@?=fWQsBqj@Src#x2t<VABZ~Bw?yiN*?DGxTUNdLpkpUH=RYRNr
z_eH2#q38oV7`;w_MgeL;f#}-pHL8UyU=$G3he$NN*z;=(%4FAQo^>_hU;#l~%o=Yk
zrtMG=?PU-AYsP@wB^o3D0ZZ;bpZa5yqZQJ1@r=#`p9(k@`KqbIX3@J^H|R))spa}G
z3o;X-X<F1ht$)jvU6epBikqi5L)VEh!+r8py3=&mO{jD&<diC-PRQ6uNB(?MY_+Sv
zr_UZDmgeuL26>i#y%A|+Z@-rWm_1dO4)Aoy5{n&9r~ht{nx(c#w~}su6`Rn(6O7DJ
zbgJ$#m0QjXP^~K+^3#>v20&0UtZDLnMDfL?<51lBtdB^6ZXpV5uq{qTZA5MZDCAY5
zV^CNCBFX>XUjMskgtAKH^8-5!^rW&PULR$x104I$B$;<O96n7Bl%3j_Xx>5InnpNn
zB{2!i3+t3kZ@xB$;k|d7J875<cRdI+;chk0)-d^$`{Gv4ZZi$Lt_S-4*CTWS_s89b
z&R*b=(&09Gp;+nG=m?<B*|}SN5T;Z>GENLh6sD8N?gte}kRTUrL|GBCbKmTP%vaGf
zLy`!-pr$f=tdt62d1D!A*aIE$>a6=Wsws(|?{o%~OjeQAAPNpgR!5UOffz47&i2%X
zrf*-8-}av#|6O6z1JzZ*)m*iVki6(rG)BEhd0re8erH!7p`sB>d^C5BUV5)0zV}AZ
z0-&yZ2_KAZJYDp)MMY|>Xw!5_27q1A0g(PK`9f=cWg^?xLx=J+D7{-S9@Rm}(A@d0
z<dt4PXh+7STz+IE4XJnE!}|MQx$?I{1WKbQg-m1CI^wY7P0qhgwfisRLh&~g!7CK1
zbg&>s@>XP)%z$opD*7A#bxsDOvebz6VlQG09%C<?w-BgISDPB}J!tLDD#kf6NLFTG
zT0p~-Rz)K`>l3fv4BFZMo^{B_ibeDLYLsOCj!!+EWJi4GVKZ3!z^3Y9RMl>~<GkgM
z9T-w!&VVUz(xwJefF(Mzb<&Lh0p8f^@>NO)!#hi6^;y<r>k2wnBlg5+j>=kVf&yf)
zU_My5tkzCzp*<6NYy?Eh{12R3`fxl$#(}_+|Nr1d)dWwrSS1P(%u)P;@P%BMQh~%F
zBA<6uYx+1G{=$YKRQ)sdbRcn4{pM_dM2YoQjJwKYam!y7C3%(MSUBHzdM-`arbeJ=
zq}ma=(#hkgX}f(x*=gfdjVxXl|030a3j~dvY>OcNhfvp+i}(GCDej?9ym*tRlW>E&
z3*m=+pBUMElz!JbAFp2e+2)vmHWS}e-0}_cl<>)eAUniVc3_?x9Kg1{DLf2SnYI13
zyTvMj!b27T5>bYdNTLS``1*4&1mQfP+LRKlv{^`6oQBJ|AX<4j!}!p+5;r1KwH&=p
zs|n#@A7y=mKx)p}wV+Ix<Xz073{gAlg9Bag#QqKn*jwyP@jp|F-LwF{fqIU@k?#0o
zGW?J#<L6`BuRaochYcJPX$llSoBN@8X(^1f4(-CeyTfyXcWjjor6<o`vsaz-Yz?=4
zakjHMqPdSpnqCap^~GW7m5oCqwyN=Kp_sS7=n7L6ugX<S!b<E7)W6F<MM;?CF%O~c
zV6?1ZoY0J3V@)(7tV%bBOTuhHwFIMw`2zlh-S({(xBp(-qmXoJeMj<LV;MbBx0>9q
zQi9A@zm3&EGN0pPi4XAJdugg*CFI#cTRL>S9R_9k0<&uS7qPci;)L?4YUUt$B+TK{
zVy=|tjvRVWd6BB_C;N+<|C26u2MBtrc_-3t4@+6${!CxYPUxkR9iQ~P@Oj)@Jhtrh
z!{e)1heOS3fIZXCK}D-)-|Vu$HI8oK8g`U_$G4-{pVMsy1FIcDTMzI+p{^Y)z<>;;
z57{I1b>S8ns7v7Ps6bfCYjITQ*B-6(JBlD{BVLiV;={mb=M+zTO-bN9DX4H5!Nyh`
z&;&I8<3&Z%r4dTvmm4kfByr!Pl;qN9f{E*jFPlns_Ky_5S`}q;)5c9D97@8C((991
zeRqs6hubA)pYdMsZ5}^TRkeeKogSP2)6T#Ax55Gp<=q9st)Owc_j&Jl3%xGVLl~_d
zMI)_oBJGNHPJ$;dH4an%RrT+S7rKJdwO%!-dd8`;{_Hc?<IKumc-+!~F(6m=l&%P`
ztwU>1b~9X(g=_%i-=qJoA@oQ-FEsgkL^}|)<Iw}_O<>+GVp|shwmH$~Wv#36)_W+K
zdW@mxu{QFzU1O8uV(APD7`=&Dz*jd!csA<Bdt&z=W?P#5Q)(7iINmMLV+=po`d}wk
zN7AYD#b`o;gwhQszROm#=pek{LDgT>X3rh_Zi_ME^fW9-sJfLF`I>NH58G{Dw3s}}
z%yxrf^EY4r?$C3_QXA3Xl8d?t!s$XzYw)<>Y`_J~;s){dzc(p?!$4C1R(fB_R3+|)
zT&eVg=WvR9EMu5#o-KvnqiDg1KWZDQO?j!)<C~aqCN93Ach|(5ut|^e1+Iy9ZzV^&
z7Uq-Z^?Z2skb}<$(!4><$v0zmmvDu=C1k6C_HE^wU<(!j0l6}7OGFz(6J0?DhZyPp
z{re9#*+*;50_)yTWNEW_6lQyE2u6Q>zQKJh?{a4n{krZB%Nly@p}%%z@aB>Qby`!r
zxs<>5HmNfjKXqOY_pRA${4z5%vDrpFjH*)eX|?XN33qC@{RP3sK&Rvcvk#@C7S$3i
zan?bB<LUQiIL&F}<ag+PGbx|7`2!$$0-`q78hJ6xUPfv2-RF=igL^e3#L*Ekpfg)D
z#T@&brwp<$VLlYC#e(OQ8Ib~>FLCALT$i)o(Gk~2`FEc!_x}lMMF|(xWwN`+M>|_C
zf0-{Kp4E<rOnPxXbhfVz|Hy7<v8!sw4*VrQ1s>}72iy&Hz0t>Xyof`97P=S@kR2p&
z1B_0r49*Mibr1qwt*t!prVxTYPoF+Vc<zcELVds>T)*iBIz0XW13=^njovqe7ui3F
z>x;PJIJcU_?-Q%jOUje9`4m@gj;Jju?Cyt0CC@wu6DzTDWBC5rTdDE%;^Oaq%A9b!
zQ>E3s`BhPaU+df>Raso}7q<F^mf-k5)XU$q|KTWN>Th+;F`J3+ObSLyk+yh#=S~&W
zse`*6O}38c9rJ;=3I&ENKpt@B|C?GvdW1e?FcF{NekLLI622Yovb(Ml1uO(2{D~S!
zP748cC^{C4N2j{SJkTcv?Rw61wTh(Vva1)qmDqc{m}zmpL78KkZ?m(jHSaTSmdNP~
zH+)BYx1c;!KG74@WA>y3&pzc$Ke>EYG?kJn#Fjr5BvI*CbaP(Q-M+86&m)K*>Hzpa
zZ}j>OCQLb+aM_JBfp(({$X4p$%uf*cA-q)S4}_zsL(e2}m+5c9ncsfu?Q)|{ny%7a
zc12=~A))4(%EGKW&++#wtWqLXHRG+Zkrc5@(~w5Sr;S-w+S%R@WxspM(w<qWcMqef
z@);T3teyiIH_EaxpNcXRCh+}f+cI!dj;e&cJa!FAYKTgxO%f9WHF(j#CRbuVM$lLA
zW@U7(7I(uQ63U=^-wW?_1at?zMSzh+;Sp73r#`%P+-&L9;a~qJ`4TUhOW=&Nev*7m
zeTGU{r7U%0@ooS3d)u9VCca2J8j*W^X8D<f3t^tZCmWdcTE8Iv2u<5drUrS~7q;$Q
z{E?;Tdx4uKrX3`iXckQLd`H*!O_ESlCy`FdU10IgdtK`^b7xcEAkqtz-zdl#N+OpG
zfCV8%hO_9Fkug+RyAh`Z%ZIBR$GkV1pnC)eq)&RW!ZJ*&(ZEdQ!nhe!qJAPKR=#e_
zFVYn|-bC$lZ?iNe%IPSXTiM8Kn0dT?XZ!hzpf*p$?gsjM0j=JV94!l^DR4ozg-j8u
zzwu^N_@SII%03swZO}AE&J-FNa6<rO^XOoT23s1aWJVx&gcpodP0*ARk%d>PkQNpq
z@h0qs^w62ll(1NJGcj+?PxJQJeCE`TFt{af^mBpTxw=(@X9KNnD*r9n7?c#Ge}LXp
zbAaBW10G*gvzGJnbk~Kwgb@k5pp1UW`|-p(Qi(hD`5im3YQYBmAYs~9B9a~BVizRx
zXxBugYk2r+76x;s<agZN=DvSkSAHVX{8GFzH9V|7FDPYzQd>h<>%^^9@F*LOfWw`(
zR<^_LLF+&gz6S~y)PMx%8ChfmDm;TxD@&3fEaqTi36n38(FVY9@BR(fZ|%QS>^wG1
z(;$5RJ~Z?}5|SxK-S~6qr>U)e7rbPv_GRaf=Lk%*&fKw3<l|#wEI{H)y8;trGD&|O
zM%_7G{yRA?n9JnGCd^DIPgWjCm-Ut7^@6^uTcxY@Ftg<mBYlgrxii~e-6a2+iUQ|V
zTG7Y{*4@g1AlIqjNso(C>K*0GE8BB}Y`=0gDAo}B9h~gg@Y2c;nKBccaS2>ZMWb!M
zL?(+Vfaa2OK<LOxU{tvTXNUtlCqykoP!bHh`u8me+CGVVz7ZAx_&Ulh|F-}F3Te<A
zNR=z2^r4i!gajDEkOav1s1>Xt190I<DZSN2;varr83Rtb38Dryw4?cR_w?}F7CMfO
zxTRHf{-A+V_OEriphvF+HIY9DSZV<zi#NwK1MIGiLC5(NZJBh%l{7C=(S0_zogN_Z
z*RIcvwto#}hMgG<{me8<#7NuV1Iq4ln&Uq91zo7PMw<_(npbd6_3LSR5xR2<DlQDo
zisH}e932U<w1tlHee|40JCS$O8ZvdorwjI0;znpc+#sSt5ArZ1O+t4+=vS;&4EnIr
zVO1AywLsY#WZQ7iA)42O@XesxA}XUhtqy*JZQ!n~X#Xe;sA?h<<Mkpj8Uwgi2~FRm
zP#0q@kB2fUQ={+nxm!z@@s{3)%>6q3^eqZ?D;fhXQJozp3zzU5#S^ZpqO1GenfUmC
zVpmOH9d+@-61PAj?=6{*z|A`E&LsPmwy+Q6H-?0zI3X6;%&OMQocS$oGx&Y88>Sm=
z{kM`i%AeTol<fA!;8hi_y(42DQ#;|?_1HJRvZ&1eNnwEhPJ=y{-EU)bysK#`xG;TM
z?d&(4aIWEE*pR~9>xwALT<a(cizAd@7b^<yM?ykBfXsbhhT&}ikdc@M&YjlrH}5QH
z6<HBr{|{ZdhkT<BoO~S%D0W$IBq!;lFSv_W7(B9(|FJ}P(rr;S{^Vs^9&XSDFJc^q
zx1vmfqX)=>lWz98^@U?jX@@>jJ_|R`ja_=gEL@CZb1Q5AvRjSX#~y;g{#rNutx#|5
z<qw`(u=3^L3wz-Wa$MrvH5Y?(JZ$J!j!Xo2MDk9wKU2BX4!QiLBY%bS_atZXt^80N
zc!w@}G+Aq*h?wYqN5Vy+iKB4qCto{!FR?O17++Dz-zl;pYG?~U*GpGBH<W8v`?=})
zEX)MtQVgknmn$#q9eX_#*!{kIDDR`hE%H)0#tWBaVAy7W1qe1S_var1sGMGQR>ws}
zXyvBX|9;w3xmX<^ml(j>7O=C}qq&&Ts7o}X;?97_9j%|!GBx6BZe&-?a9yD~{C6=!
zbe^;;y>iwdtJ?b4I$5_jEw=5CGWCvQU(VJYQzBGDnY!&gYIUGAT6od6*fdOwBJveX
zdQ;eZ%jm!+VO0xPRaI72v(RGfn=R)g1?tXkA>+6Gn6NR@<<wQ^LZv>#fP<1UsDDtD
zMWnJGGXw9`WG20BbfD`++xm`aj}6Kl_0-&Aq({!bz=;1!e_9YXxL;=^j=taNt^|SP
zO5ZPdM@ZEwhvu;ExiB0}x(QQtnx}9Bq;(|k)<5yP&pm#uZ$V3Sn`LE4^!*PdlTU3f
z@iA?x8lEGB=ZSAnh<$~nR7dCu>z*qrsJ)2V2cs~qd01{@9l0sUgn8$-XK>2|ggA(C
zYj+(d*Z&`wm&CK*^cZwz`45(^R7sY?en!d*rt<Cg7B5PA#r3CiZ{N0E{~~_&s|9n}
zN2Mun=l-~1pRiNI6XdVol~7;ra3SjdK<tOLFo%8|uZh<Hor?&#xtj^@W3%@3KE`6Y
z^mmb;{PQf9>=eDtd5B5i*Gez&)4AUi$s|gglHU!I(j3PLH!B)~7m&_RJ}O-G-%#`1
zD?kdo3e+$7(hoG|u-$*bjqfn9kExKCqVpygl-R?~qR_!Y8MHFcHqd2<;~(Hes)HDN
z6vE8&)N6kd)1vn!rlK6=hK+hC<)9Dw8yxr>f(-j7!N^>q(^p<>)1JsppQ|>B+s$Ld
zmfS*A>7_U)dPRng3Cn3!G~tn1G;FF1<SI^^*vrSh12b@M^#49aYkXS9Zsh6b`t5i_
z5*=0#3ik>tJd|q`o(D_Q=S;*QK!pVdeOh*-bp;=mW43po-T~x?Lzd05CKP17oE|(8
zc5usk_AvkL?4IIVb1e%Yz8TvGa|8;CTRt<kRc~=T6=#b1NythOlgL+_^_`C%O0|YM
zs`UoJ6__%RcZkitfx7{R(LISmJItSlc4mmz*#iP)dJNs;3CXYR1NO@`gO^*~%_#~6
z@C0UoRr#t8S!I6dyu5)`;%W@I6&%$`A&<}R>tQj;p0)0?W<g~Vuc>B1eCLBp#P_qC
zV=B6dA(MG^vj(@Sf5!^a1l6Lw1h_OrpO$e*yciKA5Xuer`5DHE$Xmi2nFfh!0lKDR
z?Y11i`xWx8u9x+NG<QcH;$g`q)@-kuRlUJ!Rxu4BUn4NO3s|nDy?wktsdU4BiHz?S
z9HW;Bk?)qg$#H{pZ%Ia4xKU!RI-#QQ*Shtu%dA;v%FGB5)z6MQfC3p>N5aL3xK03H
zwZI+fxirZ5#Q#5V2i-Qk-}1aDrSOWAI=Tp+pcBQ1@RaobpuPU_f`+K|PgS@ASiC^X
z*t@JTu=d`)ib2sqH;>Of)r|M^MrTS}WB%K%nw~}V6AXvBvs}V$w!I`fc^>skD_n|$
z+j=|j-)~w55UM*68nOP$VRWH+_-53|qX+>B$~AN&NGM<~;J6AG0|OnxcuzBweeekz
z=hm~|875m}9$$uINx@usZroAKLo|_ai;Ahf?amVye|QppLWWXtUjx}lgR|{Nkf60-
zAyl)LCdscgE2>`Yz*opwJjkY6NhqqByhzz|+@Y9K{Sqxu!y1vrSK^RooP3?q1ECr&
z2^xhPR8osB9e_*dVi=Sd00$`1S1i{Y^k8a4jQ17vF&OdxW$Nc+%?X#_!jywce(CEc
ziyvwAC~acpYt~#POC|<u)`Z<2?;aPzF9+6u!%1N$3nk&9@FPLvGjTjhL=4iIBquXx
zyPzzrn!E$Wpu51=tAB{GM8;T6J@RdJUvisL=5J+ub_=m4zqS($R#kmgv3ir>l3MVA
zYD^IHOH|=cNm*^bN5`6W#70$x^3k~s_n`+P`HvAHHS}Ods<51b;?;9I$9fZDVgOnh
zY7g6^V0<gAdKZrFGwGMRD@#kZU!WbK#F#xha?Ww^#(E!eq=NAt%qqdikI|Phw#8x<
zvvCWsQt=A@17w9`5<7OAYR+zIpHdji_I+H5(D*1Lmw)Yst(06g39&_C*?}$ugYsvT
zfgkS?w9zusuXVgj8B@EAkKD&p<|WGD>nY5iAt0lkYpL~P<>87|9q>DU(B`^&pE>)7
z>~Nw9Sijc!Y`9-kF-W(4ryKDJrG`BUu$aJ9@7q&!y*=_37W?*jY#10IP^!a+@MwoI
zBUt4+upDD#&CucrO?)Nj4y+b=gtN+g6sShG!(`m+aeFpit<`SkmoECmJ21augpkj2
zF#Wm0Dws6D1rkqxD1aRYRGx(ejDB~Qbp~8NlHm_e@|zXhPFiKv4UYpE3u!I$h28Fh
zR_cuW*Wpy*jckA#Kq2c1*ah0<TaYg?;0f?q_kVpn!fp=>QJp*WG;<XMFf3=cMxAuh
z59Lo!?fV+W4;51}KMNcc7GqZ`>O&Ksq>|#a(<!Ssx<9Y8JL8Aw{TySqnei0dm!o^@
z1kmXwqO2CVO6;Td!V5Q1F9fPTV7~y`9@5T3+Ie&|QHC1T)BrIoRzXdrk;WG5M1*eu
z+zXW~G}oO)4Jb>__vz9jA$yO7C=jBOe$WPz?E0?q{JX0&Jm$Z-Et$|%p7~S^n?)7)
z>*Y6QOY*r#NUeLCHrIJS%L=3nI|bgfCW1bLcRGGO$j_eb0S!3X4Ua$@bO(}t;Bo7J
z9v_CYNqCiFA(Vx1h$H?EDwbe_#-Mb6?m<femWw+UQ$C=ZR_$;$U-h}A^4xpq|8`cW
zFTW;Z_x0<RNJZ-0pLA^(`09QlP)kRfo<$69R2nhf15kDDbu=Gq5Fz@RhzOHg0wo!2
z3Diz6lx@Kwj0sYCIG|*HL*EP95d@7v5L^3vZ<0Z7OBLw|@%>LtzR5X^-EW28Fu*8S
zo|UN{ZiZABt~%{bJSz0sgv9B3w%14#G(rCT5dandRD{3-l<mURP>^q-_z3YDO9m%d
z&ECRWMIlj#*SdI6M}`R1Ylgt3JAD-u&WlIT-^5pk`Pvhh+BppLP$46Eu@}K*T=c}y
zL(-hs2`1P>RtKBFlT`Q$0U5a|?6=$*J^A_ps`vFz7bDZtI}6~I1djmX=A#E-@>@`0
z2EMDa-_V%_q#$^xx2%|#!V9H@kqX8TQkabh`Y5qV$`+c!-!j_vhj#8&nQ&NA@R5;`
zu~P+un}uU#v7ygV2+YD;h0YOp?!pz322P7`ibn{nl$hwJ!0A_Cy3=>|9$2m!!ZkM^
zY`H6|>rIjZCN@w#27n}Hn;5q}ED~?$+_#enOZd*g#TnT2Vv7;MiHGs<LnakBnk$Wp
z7warH6CAItdIV}SCh(H^`j^TcIh&cMy${N38A<1!j&W1f*!D2{H&%aY<d)EqgIcdf
zI2!*iYi|M$_51#hDv@kawj^sZ#uB1X$U2OKF*7tHTPemaD#?=UWNU_!>=R?MW=o{(
zQI@e3l4M_#t*o8rJ$=8||972p{?|Fzbxzm&)1^<-d!G06exCce@7MjhUpLx8wjxHR
z&p+$*=hMbnP?5rn710I5*%n=dArPPU2R4s?0P}$|{S19$tgl@NFd-mld&nd|dSm0O
zMv(!Ekl)h!kz#SKj#oJUj<FdwOIvb>*Iye`Tlk@l=QjVJft*4@;8`^wa}!1IRP^w(
zV0#b5HNOn>|BjBU;J%UZVv!tlw3hSE*o%bDXO({$ic)PXyg#|UG{#LN)5daB&LiCN
zvNyM4K%>Rb$E&L^x!0Bv^MJMl1r=QH>C{-LY3R64%Fw{CdjY97iod0ABWgyj>nHCp
z!~H9?K2S!^^9~%-7tdg_=C}vP`lvi}x}%{pcHYJI=3YTA$LRZUe5iL#9+UTEyqCK^
zk(T{!!tk2HaJu5kT+}V$u9mNPpk2ZYS>HT9w5Ixb_?_<#RO>>Y(n0C?ds&F5Bl-)q
zS0Frj7TQvP!-oJ$!E+w~BMg|t$IXHW8d%^UD+{0i$S(#!j-aiJW$WyYLvGG-ra<#&
z@E6=(!iy&Gf=^l^@6bCHCpKy1PFm%j;9PQL>d;03&}5n{U8LTk*61x3Qx?9~)Ia{q
zaG&3w#QoH^)#72F?CCuWg~z900p%X+3Y!{67q+*knYm0;c4TfmKWkk+@OdloJdfqG
zkfd9@B|$&pz3`j&Vg~OG0t^E5GJL%Dczv}WD!8I6Go~j)8$Lc>L@yGxjZ4=uEh*U6
zbLpnsi~SY-BYcEcYE0!f`j*xfTr%5DFlAt@coK*Y#Wfpo=SNTs4O~C>-+qGohSi*2
zHoKIb7tt5>C1<dMW%L6W=5GzXxK&yI5%ni|`Zu0S>Rrh&pF6K&{%L{ey~uqN>9UC&
zK0Zz+TVv*e*ji5lr1g4N3y{4r{fcM{Q3g>noy%y-&i&gEwYUL+f-TsU^t@LWd`|!#
z5rW6>Dv<v0-{C7HfykiC{I|qyHT&}O&O0irWLGpoBI#99yRh0L_^mpH)B(3oFY>dq
zc{Y|UY~5xiylkHiDbjwta(5l;N5O6vEfFpo2Q*z3ENJ!5zq*80DlTl_kXx@IBt6=3
zIfu?|6CO_O-{`URM!7oU-GrPdn`yu1{RFPKVe+$$Gx}^g8c#?A)nu?s0I%+mR1Vbq
zHb8dw5-_Ka0bUL=8zEY{V6pc7+6p>$RqZJbTSn3UtT?Z7w8EMLpL#D=JtO`*`Z@`}
z8IX%OY;^6o3&0*KJm9@`8r3i6kf)ANF>S<t=H9R|f5~s%QHS1b-BFWns7QWeP^k>c
zfxDa!7@PO?bZ#VUbYGHq=55vU(x70&K}SRW*yV38d7k9&%86cbnn*mNDrdyYZ8%?S
zK`+7~jlBmTC9wBCA(};+1(3nt)n~OO)TJi$cw6;p0R3B>fxeO60^IP4X~>j`M}4}R
z(4W&r<G=X;5Qr;LX`7fUHid50ONZA38v(Tx@}D!Hb7U3(p{x*}5hCs(pZ-z0{Q8GZ
zN=ruEZ=8Rfl!oBjc?XLzHT(Se3d=py?hH$g{2bvp{#R;4bDl%Sa%295#w6kMspQDh
z<3X4a7KtCN?0Xc@y{C+x?*lN|0AR*JX9*}B2g|03+<YtN>&MrYVI3@AtD>1}mf$g6
zA5~Ur(a&#FhmtycqTnm_Iaeg-jdBKEQp-kzXQFS9WETU5@m@$>ezvr}+9DQHP}1&?
z47A}z`_G?yLQWxt0?@xDV*u1Cpg}y%6aN5GL<es;#7GJpH@<C2sBuEz)tk7^i5E*I
zfJ-4114_jjwTOirN1ZCTnQIp1n-A)YijVQ-$w@Ux<l!hP<;cjfr3X}RPwXo1kt?%c
zmxrJc1y0p!;cD)}ucty=BBOtNRdhU9i9E!6Pz7zE?}F9RHX`JZimvrc13}4_Gac_H
zE2Nc+zmjW#R+xI+&-X%S1~Pglk^+U+Zd%Dn<N<rcsx1PYi$In=jGpb}X0>)*i7s)R
zeEr0mg#Ta!dVRo#{K~jfX)B(uN1u%}tfGYqrv2SS+Qr-9Vza%9WWC9?*~m^(vM|_B
z_;!lKRo-fu8zpsh=k3p_7FH&898G<V-+$vd2ORVrG71vPj!Z1PFwpa*c<TsU?b<5b
z-=nw;h?4~#He0AH=Ox&nC|(F4hg81dD>6#fauK--PX8tlj{#>LI9>ld1jzM)%FILm
zWoV!dQU81+sM-OL8g+gcoiu`N{K%VxyXBSJiUDa(){^KQ|B~_%Z>PTKTKHp|7E<?%
z`P42o+g3!2nf1(W`g6yc&nvvOKHsKNYO_E4*YaZF8LclC$|`ya)Z)sbaqU1yYQf`S
zFT@!nZU%sgs{X==nZFDxTH0633v?6zGCW_Z$)VwmdUJTO`BawAS94}_QigS2ag$Fq
z-uFe$C)?K{pVeQZ_hg3LCYlO(Bqrd^0kc0Vau~p*bY&QN*D9#i=x7Ll`~gmW>)2<o
zZOp(wfe!QN3?ij6y=M>gb|_G(Xy6?0@ebV^!~qx2*1rte7LJEcXemNPYUU1AmJ4g!
zzWHgb`M-Kh0crCt{A9oIn!~;8O|2BTtkSb0NA=_yvw{h8&n}ZMg(>~ba8shdJay)?
ziHUIO9L8czibgWn!}f;Rv^OK#3wj+3(1RutAvK00MMk_u8gDC1Mk~RdihBupjh9Os
zg^rju04x0cJ5U?efljjj^70_|T60q*S@`RvRZyue+4O9Vx3bhe%)M{a8xsgB^D|Py
zurk|`=ztGK;?Ks4>%_^%;jV%+#%q-EJad7_&8zTYHFJ4KmGa+w?a~)iuDa~D=5J7+
znaDZbFl<%B&8Mlzxt$s@mrfrtVkN1OV{N@q6C26q#nKiPLlzE|MI>kI^%r@+whmWq
z*0?MeEbZ<YfsQEmDUjg-xxRlrbOHH<-rM+J5-L=AQt@4q`C?N`<65aP&x?K&VAKaL
z_B}x9IT$~UmV0E?akZhGy!Xx3P`D#TiP%HUx$?{c+p8yGDW#d5&E2#y^6qu-rnLBW
zo$2(N$MjVH$ny1!jFY>10nES-0kl&BEK&N*y@s-2z<Gk)PB0iKUkXBe5CB8=3K&6;
z`pbVAs{2QPzV#~b3h&=H{L|Yias>SAyTBs^-ajW)Wsm#^7IQCm=j=M-@8DRr{J<IW
zy|vFs%j<#~*ez^^62N{#k4sagL+ZU6#{f?7m*EO_1jHL%oq$aV(fy%v#Vx-XL_biW
z0)7b?n?dzI`DidthQO3U$wgpHB@F4snUIGK9zg7&V3zS#dsPtVnGNObp}UyoL!?d?
ziP`Ib$qH4SgS1srD9Cw1Yah51>Iw*U$j1heC#Wz0X|`D?i{(kbPhc!JEy)M^Bnyzr
zh8j$QJpv2?y{%swaEdu0>i8cs`W>LZc(0SPPO>Qn;{_zPxd|w;1*4$RG_9do1?dF4
zR=yzUhy_@>dP_cParYWN92#UHghyI{F9H(;B$(;MX>bmKtOS&i)(4;0vOK`uuM<DF
z%Je}U{$7iU*`k1O0>~e+g18YZBQPPrtCI$0W<+`yt*t}ACpQF9^kzYRi0oQYHvC)I
zO2(<wuKnx&QfERPM8B#!d|hAEr)3miHfd0j2O1!td4}Hq@DGipbECkR04CR2f|!j&
zEjG$Xr<@v_=kV|d!@oIlyc<m{Wj<L~Xpq<ZnsZBIX29$=x1+wd^6y&(b=)IelS>uc
zUx(n?z`2pozcX1-OSDM1xhYbBUIYyz=-FB56@W!(`i#!$k%JiX^d-c-21+B`z(WHL
z@(Gy4%P_EMSCY`)R-it*GrGDSD9u<W6zS<*>0z?6;k5!-xdfk?ORnXr%4*A}Etw?D
zxNXujjLH%t(JyHGu`K$tAvN9k9@Pi);FEVn5q1{iwU}Q}-h_X$VFG6<68vJ0i`K&m
zl??gIN~<GRWpBz1A2M-dorQl&#RTNwyD;{|+G{~=yQml6z@+{&hyRIzK@J&A7BE22
zV*VH9&4|WRp*IXEsl}%-k;4F~Zojs174OMu-jgrCNWV%v0yIXGIb`k>WW+slM_&C(
z$?10gCiuwM*;OC~ewO=_LC)=E8qjrW8yGS$PNls{8agfm8+cBOPkPl~gdkUYMlq&6
z;;YnX!U$AZ3J-@ioWJ>EvCvk{_S$Wp3l;YpcYMxrKj&xpEtB@V4gn<mJT|oo&|m&}
zAs|wZe-4gWX#E2O82s>Wi6Ov)U9eaO8~SlGps@l8bc8T?*g$sQ1Yqrxcj37?w@D%{
zs3MKAxKz@czF)LKgfGyLH5Z(<frF-;u%Dm}SEryRz&^?g=B?VjHJ)T9bU};)kAgn%
z2NyIUCz`6FQEo6WK?`>wET%uJtO#lYT73&Jzj?p*bgj#^*{_e{ib5XSlRZxG?l@`^
zMhmn&J^{P|OfOiAMCcR(KgQ>P*#)%|Ax5A%4&dCEfrcDU0R@%*xwzP6=p#`l*So>6
zQywGVTuJ8LR*CYcR4kd>opa48i(XZ;tB}MqsU!o%awEqu&dC6JmH7~Qb$0Ij88=Kt
z`m@%#b3q65Z0^0;CY^8z&;$(Xlk1%N`OhgU7#U6OXD2Ej72c?*&C7Q!dt5K_OPd<t
zlM2Kpo;hD-mMtB|eLiGvFVea6$<)>IgMQ^+!EXJ2sMZNeD${8wbRq(<L-D!bLBJw`
zDsygHiGt$g#~SxNAzT9#OyutNK^qQ+LcjGXtV~@YS+G+&mQI?JeBJtQqXu;YEOjb#
zNpPMUsxGcM<FLu*e@{`PrLvZv6ibB5eOznD{v;nj2lyPnm;AcDsiK0?Rwaf?&3<)l
zw(t`IUY7bbh(nt~eWs${E~z1?@z`un`lNzCcY73zT3PZ*T<TEN*2x_yx6hDAK<9k}
zE&zZ=)Qo^zV*!!X5Iv?yNQ3Pk03^ol^}Q7|a{<Oj6qxv=yJudygg_l~CBg&a?)d?9
z+lcjQH74woZfdHUWce^lNN!%$r$ty_X#OA$5)HH(#w?OHGB-jSJnQ838hD<%#~WS#
z&@eD<>{n4CW-eA(?(g1D$#}T47jH7y|8pnm%dI4V`ReDd@f}9<GJ}#06$bL^aYFzE
zhP<<ce{>jdqOGC+dGq7YIO&b*!I9DXUq>If{2&dZ3Y`n_P>W2+Z-ds^+zLH2{z&RO
zpdw?s3?kPsySwu)$Kfqf)#lHw-#>NQkgA&9_IyF|+<JDb++<gOh;BRsz&--UCYT@Y
zWk{xwo@9nv3UIG^Y6C706jywYD~FgAv!=}S_QCX}ogT{#{yq4WV33a*em5EFp->cd
zt>`kp5;o2Dm*pXo&r*w=MsDZC&-50nycV9tMSU=~+a=z$tl+<1Soh(U^xGbSLlP)r
z7z|v3UJ6}8o(GEJ)K^T~Gh#%JLx&4U;OIi{6X++sxaSQGRO7m|+xNgB=e9zfxW^4$
z^3gMb6IZ3=eY0g}6(n+@PQ|qd)5F&TFwq_lPN=JO+lVjMzah0h@{aXwt2X%Hqrqps
zG!0Cr`|J7w^%OZZOO$yBM;>d%sdhZmuyH+WFZx!ZzM)Vh=zpUY`6FKLa^5m1N;{dZ
zb~5A90nlLw%u`Rsi!LZ4+Y%6c<VkN=U~%g1^o*TO@;BCJ&c!kXxu^wkcyxlr6%*-A
z#L1j15i5r(mTY~G0k}K0{j}1M7R#qs{-nt4Jw>H&tdtM4&mMGeya?q4g16Qt1y3AS
zQ?li5iW6zaFs%ME6Arl1zR_^h6-RdTBCc}CQW6<GLxyKBW|!6aK32Klb35bbT(JRI
zzyKqD?s?5Rtp%L^x`IX!vbPB(VxE3xR!;H(>6wq<_Yy=m_^8pTKtRl_hP&xigF-~>
zBBp%!Yixv@v>&3T&ldrIjlsO(?*#2CUw$%2U2bb_tm{N**c1-G2k^?qVJjD<hdW<|
zGu_{(dzh7sx+6b^q<(G9t?JzfBn0M9ynbox^jqwsYn`{aHuU+JN+2GshC3%G-|?LB
zbphPi*<1Jr`&1C7{a>-c<CEO)fvn?6fT=^#JuvrAS+yar@?Z8|GaO{N13}V}bY=jW
z;5X6>T7wp_0m&i)Ubm9?IE|WuXF@mL{>xAcQ~`zpwXuEr#O&epm-@Vsr@jGgU16oF
zxHv1!swY-3zwn9ljjH^GX8{O;jo4p?*zD1Mk|a!OBC`L(E$6d+KsIO~$NhMy&>;6~
zgO=o5{@3sxS6U!H9AX=Q22cOcF&1EApx$PJfMEgI$Z{~zUVsxDILERHY~XuKN0o`!
z2RaHCU+m$En>mLpzM)qn$nYA>Gl><_ZSybRa0L@O)RtVO42k}kha_uCRmh9>%M&LC
zi%O{&x$LAR*4Wz0h8*j{>|y+sg0VO>p}jJ^p~jx#5`=ZZ#ecsvm)Hsu?|jTOb7x-T
z4Aju7@nqJ2XrEjFji<ZA0PZ!K)fQ4R1#W5pk^2S68$A9DBKezE+^qP!aK}QI@Lr%`
zNB(eKE=fg9+2K~aYh*KDvh-NwH?LW_7YU!sFwRk?5>Mrk4k|~ZykoCRt@t}xCG^D!
zc9y%3xM_b%u8`BefBrOe>s9(Afj<;j&43S&cTc|@?DWwQ(hYx+hJx2^ZO7W`EPsRE
zS@a4NYwO(VDJ5;DTwl=s0@UgP%exB)4(YvakS+oSZ8po)<H`|r@3gClMQ_6IVUBzM
zWe`G!oIE{89n^wFso<hNvxUET3OmK5he}%Yd_mxS6I2jAvmX82vd9tm(-4<jhr1s-
zq(H@G&j)qr90p>;3ZTTT9iUR@U!8sYTt3AYgX!lkh$Qjt9lV`gmUj5HZ$U}RWPMo>
zyRGx0x8zlLHG0P;19$!pKDybrQK@CD%QwHjrWmtbLyt6oa=?GR`I!<RKT-xzZ!3C}
zd#KJnBT?N*>tdKj?comR)3D~6%=J)*d2jv~1rypu2rtwphwO`AE35&(OkY27+o?c>
z8+0#^_sy@!L&!?rY-=riiLo8TRPlW>C@7sPN%VXZpyhUKVyvL25S4=+xcg-O`3@Yn
zS7<UKOZPmXh(R>I@}J8Qq6z*0E~uOp${3MA!#2<q7Vv}UyeRJF34ov4>4Sb!sK}*0
zA1tzW<*<+NH=!?rW$}v+)iwwzcUBvn;v23pbN){7p`u}ZNpsQ5i0CAZ%6jB%H2&y@
z{-V+K$aR$}4ar9c<DzKq?Ay|tcjp&x#>J8R*9q<*|NOXF#GbCwf}vyfAQklDUo}Rq
zFXTc;(J4y6gnkB71(Ez!uR}_Ffb|Nh?@T?|{0Nv~Q+FUgaCz1YArZPjWh{6pQ55v+
zJ)#Z!rRz9Bw{Bq+zL=hY9Qn%-Qn?FBhwhTht$uoWc%1L>ro@|$8jvfL?-1PIAY_HV
zQ~hyil=h$A2k;sFls8xYRXvKa2WFhyc>*%gvVgb>A!ktYHZW%TAxw`?Kt=*e?}IZ5
zxC&qm#ih|ZUO&EI2-U|&V9bvkHSj}R$9cmqHaF`}l@ufZLt~@!wmnc?t|qUt=L+M-
zy})LuBGmQ+Ie=IjRMwRI>AHgy<8k-WPXNcsdFJ^@!cb0M&(*iUT`?CNu)7A3TL?u%
zpq%Fk{Rn>Y0c0IP+VKT2gwT41&SG#5{(F}ol0$+u1{XJ8ze&u0ou<}C%edoE*oN_`
zqso2HtxuYOt4}wb&5$(GL&m?h<e7QNpIO`QQ#cA#W9zsdXAHVcZwbDsHwn>9|Hp=e
zj`9S=f%*SA1W;>{1RcOowO9r@2`XIYT{IF?A>4{Lx<qvYG*d*@wTRx3B;C6Lo!9dY
zR;KXXqB8e(ls$3~Tbw+`*|YH4=N|HGs?NzEJ@iO35Iwslqy#-A=;8Wb8Yu7`K|c?O
z;u22<J*84scTFvQAXNVv2wsB>!73rq*<9QksLP;1Dd@|ZkPXVE*AMK#>%N1`72x4O
zCDb#^V2dY%nW77u00clC?0LH4vifx+>BU_nh%v=q9toY5Y!mN{{(}#_LR%R9DD8LW
zZ1j@L>o6a=FNDs>S>6Um{xVg4bEP<5Uy$gL_t8IUWRB6}`S2X;YqNP=>H0?%2}AR?
zf-|LTYDdd*qoZXsc>yS{e;{xSI#TK3GSrEk-d8pOJXa3nR|4s6b+B+YTpsWlnea0;
z%@sYc-2ak)dQ+$X*>EYXp)+RC-0+I_NFnMgr;)7Olu!~5|54_8i9thhmf-L+-w>4$
zRf#qA=X>`bZ|nX08i@J|d~hh*JixtrG9wtQg>(Fld_I^EJ)UYy$Mf*_eu2>?;x9Y(
zQE2zaL`-PU4Xp1KrE}H!Ws#?%ldkJMlF(~74}0|B3LI`2iGS&dSi+p&aUP2w0KJAG
zG(Ci?<yEtYafn&^v3W4x?A-4Kr4P6q2r@(eqyZ`_sI+}aub<@SKenYxCwDcUOrj%*
zP^KD(!_ywwfui(zc$z1O4Yq*Nga%B)P+|`7f9b0kKqLszxe}5as|GW4=4A^1(H_tn
zavuI%TIX)gx-|ITVEM5q7^NRNS)CDfK><|FKzUR@Z+a@x1&RiNlOQh-dSm{todI0n
zL>m1zX=d8-^%*JPuOy+9JA~!4B-l>*I6N39po{{%gPslU=m2@s0jSe2AQVI1Y}N^g
zbp$SnF~cy8`T>lV0SFO5Rud92JmCtVa|sAx)syKU@kIY(D406~fl#2d8blYsX9wNQ
zlkOb_b}v`@0Dygf>KscytAEhTf`G68E$`w&AX&yC5@XZTaq#$(#!(i}vi=PqLC5nu
z)a?@x`mC*A09Q^T8t?)Cai&2MRS5bn2(|@dP3(cDSZZTy_Frn3t}Km!B&RbT1H#06
zNH%Mx;4=M7!PyIs%wVP;5Fwg9^zqyV?HnK@2hB79u;{OGy&F^z8zksNehY<6G?ZQi
zF^J)S5tMKN`rC)Lc%&Ei$|%E=^ghy1c`aD!|Ll3-6iBegfL}!Yt32gZ5fC1967bWY
z)Hh#Z<H_akL@0r5p#^MLeK)@pt8C<}+$rCmYyIZ|g^Aq-N^}E13pzhQFao9&z+eEV
zI}0&)0Tc@E1bi+QIXRI3OOL((1nhF~L9vHIiYr`sRcmTL(q7vUPn%$Vyb&+bZ;-O@
z4D0E9V58ETU9C@Y9eKa=`fGkc{=>AY8}9vAyU5+YEYqIeHoFE6UUAvXy`EeCswMYF
z1@SU}87eE4FBMereARN~cQ(1T=6fDE`JjG<o)(AP7wBmr)bBq7m4bpsdQc2DWa~TV
z1ZDx^9}+PqixR-6rG!$BZCwEEVMODk>1Uk?6RUsV_1WQ&yJU6#8;h2Qaz>nrg#LKl
zIj59zYgvAvy+G=W-&G5jvvIavwyxqiPooJKABkfu`HGHOU(vs4Knl-smUukn-GXXP
z%CCOVu?8_EGHfVV)?@jc5A_1x57=2pc@Ii9%fi;DMJm&2d4NtqIt$t>)(w02>m>io
zp;n+Fm=8>@It}719sx58P5OOlAes!RSc3P22B8AYptSMi+VyIZnDHgCJD{071;Qie
zft&U&-DCRPrJ3Ib`L&Hri$>nq+sDg-<O(D_RR)D>myCGHDwZ;Mq0_$kErwHV&9oX~
zvKh5De-L>&xvCRw$nAca)jzsg2GD0)gA#3B&ayY-YEBcJ5ACgh7RU?t8U;sN5hbmw
ztb<;<A6g+@>A2`G@S5YMjwX+mX}!DUmMS~(MvmuX%=s)frC;{hivBzs1;oAUf5KhK
zC4U*b7xi8fuSVOL=1@S}LNkc3M1O_Q?gURA9|pX4a14P10x;ENegi{#%BQn3dDaV_
zEAS7_4HJ#81Dh*&u>mN>f7>WR#t>p^Z>(f@43}QmDp;*>6p79z9=)Ivx}f-4{P*JB
zp9;Ig^JhT&AmdhQ$q<`520sc)5L?dgL~omqMiJQC2W$eaPTxL4O>URwnJbn#i)x5$
z;xo6}L}ffq#gB#;$ECyuJzt5{tBV=}9e!db^Uv)>tZ4Olh=w5)y@A{{Q{bnC$(^sf
zzvuT>Kw|j)w>z#`!;U!ch}4Qp^q`%7TpNnUKLi`+Kh{wP*cC|T>zgW2qLKgJ%9G{g
z{Zs;P2>Bdf<0kfjEuJoYGW}(eX*2Y+yNz$m7wRa7ri2Cs#?;a*tUtX(H=b1qlKa3i
z+e6tLs_nlMZSmu!lpE*>z+PxnOP#W;!L+YA;9C6Jhou+#w_$DhImZrD$I0U&dkD&~
ziIx!V*J6s((-TvuY-5<)%O}W4<aUVcthw0a$><$lNbnWp+_O-qWFbs@slC)n#v1mr
zrR?ht4nl{)W`V3?J`k(qP1gDt#J*rTTD%?oR{cEFDrgXGKm@S?cyu$!Z2iYhK}kbU
z7KQ$h0ZPSB0;iJ2uKM{+%E?&~!GF;GAF#x^5G~+IA7K3Y+g}*ilPuXIeI&w!o?f$O
zJ)Z%uI5KeV=JtZ%7h1;CS<d@1%`n+tmPadrb}n|H?l(sVL2Am|Ap=g_nFP&Hgo{DX
zIcE#7IA5;|1L^$}dfeA<kN?TdsT~I`0Iz@B=xuftcO+tY3aT#TO^wt0k$gi>dVV96
zV2usjKy@@4EMhk)PP{qjg|SGm%S3x9{Eju$dArNIQBXYj<|)cN>9Y(mFQsTm<9TxT
zO2CQB9X;TUftL%Z`#_#6eM4$4`G7+*>F)Oie&7%=Lwb->;LO=nKw!a1kOe)ir4KjR
zIyP}+_zQIuc|5ZKJY@xW>YmYGhEh}0(eUbLNSD(^ImcKwc=!;TeFe$X+f`jKEauje
zgip4hJq-M0X&m@{V9(Z~;+aS?P3HD*$$|n!QMBz{&^_k3iI((LN#b&=iuqiec`UvS
zpI8*8Z@JtD=l}@~onEuBPcI#d;b*6+4uLisN;X$I^HoFv>F+#;TTYTsxe5I2BBo46
zEyu7e>fmyepoIM>!je!#&xR<zSHIphvkX<to&+eJI(MrOH&n<6O2R3CN+mCakNsEN
zB0&bL9=dj}gI=M$=YSfHEa#%Jvt=Oh9IyX5m*k;uW4Bo~G5jC$@_se~!M%C&B_B}X
zaH)^Kg8ALlCscv4(_%$H2Zrs|yVnM1zp>rg&wFIzawCdKFu@>Wiyzh(q9kAX^5(kw
zGs`vU87Th?fHyj3E)i!2lz>k@ZVBZ=;DHrYxHAu|zJR=>FO)U<4p~`nskD9qeo63D
zM@3bdbu6Isbd+DSM?TdXdYNJsma14XcICOimmeq=uGpEVL&rXEN~t3~v)iim+K+y$
zt80p(nbV5!=I}#im%p->*x2R_9Qn;N3YY1}oZ2W>(GSnF=U)kYW<l7kO)@>;EJ$}C
zfI~vZg8=*kW$4f#2!9U}MTDCCs9f&%C!A)bP%J=l8Z3|UT79C_3foZXSB>u126^%c
zd3!$&ES|w|6j91jJM`~N-yz&@24Q$U>Z@AG(B;>vLRD}3qU%!2T5i4MRpXuilYdI|
zj+|v~Won$28|Pn!7kxwZUSKO4L42;ypw{x?L5Ktho$SE(2UVCqKmr9wQ{+Adk`r1E
zz!|uC#j~Xm!+JTfV>7}nvl^W`dG%2TVpG4-$r1Sl_g39hs50l)m0_oor_%EPPybvX
z^IwJp4auV%S$X+F_2J>PJmO_sMM`q2ZjcG3;_l6<F3^UY{c$VMF6xB*Mo_9q-!<S>
zY%0F1ll}v^i+~Lc&^!n{fn6^T1=%R5!S>1Y1JWDY>ZR0UPCr!_%Z~`|;Xqs<W$70I
zvckwobgJtmjm9LI;v4FRJ@p^=c_7+;e=sWfcHmQhlZS@oP0)MNAn$Z__N{)IT)5oN
zUG*3JAnOR>GoXA0@QvvzX3%#oPXPS1*Cuo)5&{jXfy<1~ofWa8PlaEFt-@}4hD#FS
z)Csw(C#FvTQBn=Wox44!NV2c@ZZ2l0$b|^!U$dulScatsXvLd+UDwFi)5>#~(=H4H
zrKN!x7KCMB=?tEXb%FdS(6U89Xc|D*ARj~xN@gZNp7*~;w%Sa<X&-9TB0%4b|Bx*k
zx5_!JOOUOmg4RnSQ*hU{VN>&(ytw#Bg?hfG=Qp<8hwi1%4i#K%PWA+y@(NaZe{K0j
zq?7fRw|pl0hc(X4uAikq&8VTa3y=%{PoWlgKJ>8f-!Nz_e-a=cZ`ug-S>B8L_thi3
zcDY)-{v@H~9zzd8S=ddaidgnb48OE7>$HwXAs&c({$)7-#ylXe#v-6Kf9ZJv#FS-y
zG({+gk-t0DY`yLDvA$K8<MW@)#H|1mYA*;>E<j0PC?N|;D?mE}#3rG@^@o5vaCAh2
zfRb)NAda}v)}hU#>%Y3wlXy0EpJUqwPrx4;B0nYdmoAVlNemTK@)?(-cT%-IWbka}
z@@~rE+?$paXK_|4qW3OSdSd6&K!X_2CgHyT91FeP-{^nDyU+<upAKj(LL@t=hdS_%
zJ4@H-S*^qk@Y=Xn6Y{DSx#BWbw;!2MUxMv&Ob2kRA4@3$h66pTqywNGdQc20eL)l;
zE^R250oV#?&J5*hfe`t1D9f!`3c2|i6?eh2LD(t;rW2EsfGZ8Y6f#PnsssXA|BVIO
zpm30&CW~Bw2-RRDzvz~WWxzP<M|_!1GKl7$VL(^`)hB`K7U)^x)ac<Hw5`M0AwK=T
z+#Ivf4B`{yXi?3NE^(2ow%T7bgL{G0GmzS%zXM1`3)&D~pjH7S^`YvZ#sj{YV0*68
z{x&fPaqru=k6}OKK8FAP+86AZS(5Os`}^M^29|w{`}S$=n=aYF3g#sLkL#i`<Nw2@
z=1h6*Oez20d`v_1Fnqkv{O<qtCKLNI1^>q-%ysxZgA;QwxPnCO%$OLU>mUnb!GQ_r
zN1tGxeHZE1EW>|)`uFGm)9-n2T}?DTA{jDS85X*r`yC90hBkQpKeznPjhi26KCA-M
za^l~#>}UM%tT2FSNyseewtM$)UiL%t^5=cBde;BEE*`s9@PEC8g8y%acFcwH^nZU5
z|MOhm<-urukDmU2bsY(Y?f?IFG5)fhW;V@*ru?2(oha^qy^5a<-5me_U5C3L`2FX-
zLT}@LbNMmcFOmCy_|mO<5Aed|xfu2x*tehQ09deKL4)_UkAaI@QFOo95d||l&x_jj
zUgp6M#1oV<kD{{ZD|;UUv)2B_#BByGyMqf1zpel`@nE*)Jg+^n{zfqgNeX;9Y%5NJ
z?`9Mqj{mZ6o8dwVVbk{q^!YKnmCRqjLPkH_yE{~WMcyXJW)dl{bbe`tOwK9&ZNEZB
za*e7j1)I0UN<3z{L%waY!7^oxm<-l7PB%vdTnJNZ@g|34I8yF!F~l%6kdXp&aGt&W
zPe<;Typz~;G2`T3dTZ8z>_gj^I(ha^i~k~CQ`0tIia({1$|G#>Z9dH1oI@zmmX+_5
znDNrX_V6%H0U2GI&{=#DkEYT{6YMv^e2IeH*8%>Ia&`n+;|K}fzT<>l5*9X}Xhgtx
z-`t-P%lggVMiMYS=BdWW6~Ot4RRLDj(Zc;;kj$9S#&`Oy*Nmy|%R*emT@)Eva~1y6
zqGR}Jm6CUw=t`6wnduej?ZG4k7S)CTZm-`D?l$E&m3BB@h>ynxIKt6X8EkQL-shjD
z2VhekgfsF&$*Rd`I%1iX0xskb3rjX{QM;2bJe)Btt_W1ttnKKMn(UfRvJ#8(h)`{X
z3&7ZBtJZai))Q(sPwXnL@le13NO%8i?VCAkpLR_m$=Xgq=t0}2C;5QdFl^)9G}~pz
z4ScERc!s|mm19f)CZV6&PJ;hhV0i0ipDXu;{bAGT%EQM9o(ini*H{I-Sw~#AS7U_4
zT8?MllDRR`5w9x&%dka~;F#xBY;mhp#=gtkAA~3ux5rFgg(|8=d4`Lwe7H7~r4vCC
zf5qT`I$h6Bcee5NaFXYUX{u?iP{Qid36I4#HKDd?9ioljn`<?X+ay5~#AKy7uT9VK
zKqq!Yeg1_4!UV2@uyZ#+bfPhZk>ywfQA;50vSbJHFx#wW*3*uli$Ow(541v!0)DA7
zr=;6<-nWU|ytH*OUN%!>;Lk3_KDkX<NX3CQVpxG3<>?vUFoKv?%6Y~i^!`EGP415u
zN@wD&DhRv0;1LAT)Oypx;RIm;2hUN)PlJ=;jECOYX=(&6*27f1W2M(4N?qQ!<|T8$
z)Y{|{TC9Xxcth_r|M|EV0=?|wK32@5H8yOajWWgZf$5XeiO4t6rEfFuyC7?r90~n<
z(>5yYWMu7kuxH`{e)IG^{TtUZ*9n+M)Cn5-itTLTOvNMmKq!_Wr`0jzujb^yt!ZRr
zX+<D=v1GdbB(hctN%HgDdndBZ5s(UQ_3USZ#0EoPh;86ue&L}C!ZI0|-oJI+a#**y
zwavFnSYE@RzYdNWEB!J4sp77s$xAugi$y<*?0ywe=lO~Ua8=~16q}OU-}WMJS5Pg-
zOFtVq;R{Oy!sA>O+SNQG9BE{$QEKr~&bNw8?79ce#NJFmCy#0Gm7W+SYz5rOoKK3g
zXV+mhZWye_c@K7-I_`>ybhEu1_vQ_=JXe#V{>28}%igVuA4IaR6?clU>k2VGXjkiU
zu|2!_`4L#}Y?fnVvl3Kxsr`A+HBv9+YVatp3Cxc_eRFa4Ar)885w9-+YfyhOlO}h1
zAAdNz=Lwjfp^sRwpOf1_TWu^ObL{VlSfiP-+?+JgaWba~95eZ(Z3Q9vT_9BH81seq
z4`#~77+MSBl`fx`o@9>`i$2CK_AvZH!oyI$5drC@@6i2fypJu6F~fu&>Kg07GW{Nm
z>5w#z@OU%o^2HIg^dpq-9(i-Nj_FRCVg_E*v7VtdTOd|K+G4)yU01d$<KmWn{V74k
zQP@R=HV3Y2of++%f__S^N8315)5H$bbjCOhKPoshKa4e16@7oe@8anxgu+8+S@Q<}
z24GlruTon`@MXnh76*aS$xg|b0|E|y?+>!nbJ|T=+i`vTabA2^|C%3)Y;%Tm;Z-{<
zo7OFe6LRqD4W2s4>9;TEXsFni&s(BgUZIm5hczRAyTKd!so=&_mZ{T~i^YPNnbY(+
z-XbIGaq$cLT7YUw*H1Y(=22j@@BrELu4QP>W6gsa8DRb2xx2pl0WL6gCWZ8?jj<-)
z)iyjRPGC&dE2L4hReCa1Ae^X{pe9;=i<wK{yL!~pJv0SXG$mYKV%y}W5U;GL9mf@&
zeo;H-fTF-b$?yZB%`vfzE&9=_?`EglB?gk&HltcyW7u`r&8|fO9l`3h2`dMoY#xPo
zYWlna#Cur_^S`*(dWaKjU80I!A7vF;B!YDr9tvwC4q5be%?M3a-h9F}Wt=9a+kWhT
z0*m5BZ%&E4j2`Fp(hmacjG^e2gNgI&8>FBv+^#Lz3Rf!)nZA1+yL#Xc6@I9f#eL8g
zEeZ?D<twy7Q?O@?*8}^h6Q2{un=6akb;UY3)I<gZTKGt^*4}KP0*YqMTp^q|E+=Qi
z(a=5I1re*EDza(d>5=!|TYgL}O!d{yW_e<Xw}SO~<<IZ+uToo<n}c}r^%T9jSV~ak
zEhoiCVfn3F(+n}~w20gEEo@^Yug1i_uK^t>EKnLj)RD(;O*XJp?;5FI7G3cvd6yOJ
z<PpvedY&@1FjiB^L-i_%5I5Vu4BVgFUV5y*_{(5gJTUr~0YTaty<)rI2%o^!QY`0B
zP$y8}tf}vh#)N?%!Cp15?@ev!U$2?|?A)q;LB$VMeD!M9m7%wT91>#fhmSoBZV^xv
zDHvlvr>9^ob*M>M>`7bp`v{Ntj)idZ@Na-B$7bB9n_%Y5A_Sfyj(d1`=pacoeT|H)
zll;wcD_A=&t%aPsqQiOoh8n`1iSyQ=$AcyHmVRo>n`O~v14fQ7T+1u1m$k>LpZw+t
zHN2JBwoylIsg+gG3=yhE9ly7@JQ;WGX!m?Jt|BOR!XIWUduzlsE_*udLbIB-6I<wH
zBjxwotzx|kZ=_C?ic;9kbdNbLJN<+c8nzR}9&B~QAjeeM9mm(R75C(MJVVIQEcU`&
z!uEnsI*$nJ#%6ilRk-5|-@RY_*~M~ugST**gX7J`Mn~132@OP{`6GfQ<in-{$YHsm
zm3n^EOE=uxU4ezT#~<XaUN>Ff`@!l-l4}Uzpm3-n#=^;PWG<VK#7Z?TA@n#h_Py;y
zIU6=t(G$&ztu{5@w4JFfTzMqNREhsB2XENf&nx%!?V>>PpfSH{L}ELU%gs<xpMQ{#
zHdMm69IA>7e0KTTqO${EGA#Rj5Vzeubdh-Un2ckYQk<v2Jv_F!J!|Yei+G;+@<uCn
z=q!E(+9+$$ssuNfL<CnZYck7O60(fe?CfS(Xu5-Vp%W5?+L2I0nA^2}(r%B<EV&g0
z-p#nuatOPbL!4v?o5kD~CKzW0$0R4q7=b&T-qCjrPJdce_Ryp0R`9mm&VZ7tm&2`4
zPXS{CNnIAZW)7qkMU~5nA*4Te5G64f!jYFCkOIs8%g|(X-zR&BoR_%IXf<sNxL-Le
zEvpH%_c5+j7d6*t+$gI?W_G}{bC0_ymuk6X#)T5pAf67Z$f9MThDW^kq(&Kb5CXKS
z)sO3Q^TU4MY!pReH*U$neuy~A+^=-<+i1Pt!W~<fyc;3>b_MKOdqV%Fg0g6fB^$p0
zzpV8SMWO=SN;EVdaT)O|7KSkPh@{b)tBF-?8I7KqA&Gt8?TqlSrg93Z5<i+<EXzE~
z@g+?%i}#f-co9F2rI2>Dqg9EzuEUu$o;|PL2VPJ6%3se(8XqLydJw$UKk3l8zj9L8
zu+cS+aS``DS;kfK^Du8id1l^-=TNNAcZ{@I^Q3zv4)3Uyp2JEUYvAwnM#G#pr!o{8
zjrn4Y!{xmWb1JBYHMKIlO<u4P9V$Ee0FR4p4<qAFpb!2~Yhc{>Nz+kIeDnAZw6cPq
z*8`T>(_log%uMB(LuG<#Ic<mY3V6)%-gfuSOtq~a;T~y3$|7DF^T`UOo7?GEwq0Nn
zA=UKuXG)(X8@o^0j`&x8UG0_W$6Ai5s=0Ps#i8$nu0LC7$9x<0<74+@>J<B9+3By$
z$F3WMol$NNAFMX!j!hnP|1REmqKw1=zjt4Au+hDTd>9Wrg{$(lI=C<4EZ5Q5F=@I#
zpEjuqc$UgBJ-8Ns=$!)0ThsE;$IRGm1&LAeauLss<aVoA`?uJcJ*&f(ffli6@mZ9Z
zWFpu(ngdmBIl<T=>mOF4aY8ai0?fzK_H$4=2yR<f(tu{!*%!jDszwey?ebm%i%~5>
z@IJZI6a%J+ArZqdc#r^d@6$#k^zqs5;J#mrPcN3DxH8V_f~Ro=uQ0Y<mhm8ME8$(q
zHqN}Q+6`u)e~{&Du0Z7rt_UP7AhSD9RV8vxnc8~v!MpNsN!QjRLM!`pl{88Y%lIHO
z+6A9Hh+sXI(9BE=*JrBvp-nzg{Ultpe9vswmMt=W!`3>I?N(@!#Fm82o$wzGN_E8K
zJ`B~420ouv^)CD`g9!&gcu40+FpEz$yN*C|a@A<1n{DX#>VQ->xAu{|-qCnDTY|0q
z!Gy^Nx=PJp_Hp{sO?wK&n<Ed68`!O3+HIX`Ovq57^2(VKT{6>h<-75HRqlLoj{=?{
zNsfeXBp>yK=gVTT<*;{^?dJNOVv|`0CriQwu}MQ_4nrwyB2@=ZTxm3JXZg<MaNKy#
zETorKdGUT>26|hf4VJ!DD?=M~Yfm%Y-ztA3daw$?wOv@3{5uN4@SgdLxak{yI0prX
z`?`u0^SCUOwNd^d;1bvDt(9@4mf;V3q-Z&k+;o2*@E2E#t2Ufv@u4e>V8+8U?ju!?
zmEX~Kwk0Sq9k3jVCn{bj%^TDa)yT=stU9k~hXu8x;^kR1vS(J)ZI6%QaY{5<inT$i
zmc~y)e+3*f(pTA5UaG2%eVeGqOX$I;F?PK&kR;fVPv}>hX?saqtMbY!_J1JB?3Abz
zDFLvv<X!E(zAQ2VDZ78L3Xio9(m0_(U8!?qbCn#IK}EXQvhx0rk{A7||1gwCg3Jud
z9=m){vCusehI@t#YjDmjgq_XR>=s@uqcxH{y!iX-i{CP^npL@<`O+UowHzDywP`z_
zcm<Wv(=C~qkHTD-eq($zRxTk$(tTTQIOYZRi%xJ)+_&U!B=1E$t=IxM^X4`j#M>>%
zP2K&DOdEV-oI-c3Wv;W1E&Hx6&RSlE+-#d852ChhRdpm*B3h-Q08Ch?LA^Kg`G--{
zZppjt29c1j67pO<V}yoX4&Z-}r}MJ8az4Mp73?mKsiol4RO_!yRyG6-{2}t#!+DfD
z+{Rcje;IDuW4YESZDB~_G)L7lk9wIysAkjw(P*+^VZ&br`0?9B^9dZY_ZaP~dZL)?
z@b=kHd#Pq}@0uuxl!F{Y3WG*-Ur#go8oLH!v?eWsW!}gc=!XdKOAW>l{Ukl2WLa!m
z2YIQSTyf8BK8fuQ-YGwPThMD870@q{#f0?PK25G+kPbzjq>(oo9SvvZ%*k#*IEhFy
ze)ubX3!UpvvDUA?|JjMROb`w8SgEqskHx#kECcHzkUhPRl{KwGO&>SNOTiNbvJaPO
z1&5GebE@$b<m%*PScwBS&5ayncC8P2#wpWN{na1DyX4}U7$zhVR{Sz1T;BXv@*sa-
z^nUC|OsSZcKT*X$6Q^B%b0o7!XssI0v=;0pZ&(=~?^{tM5nvpH9QxdWj2gVyzR%jA
z4`#N1w3k-&{_X19Se(G?fG4lF^gH)Ot0uGTo@<Z4-S=w~4}<UL5aB#S>n<<t-3&PW
z%5CXgLs71|K0`Op_9X0a_vzA~tji}UW;h#ew<lBwHKvQ+eZw%b9FcI5FqYl#j%%;A
zqxUM)D$W0(;<Aj5_ix!^le#sYnOeyBHihS+=TxGcHq2F@BT_GHUs<9aYtT7d_QQck
zL5;-9mmKQoN_I|8sHgpa?`JG%7`!-fUm>8+XAZqlObR?TnxvX4c>KC!N4w0fg(05F
zRG?Q!?PnOZ(S0EODY6VWufVBOC*3Vjkmb}bua!TtmZ?l}$bGxOX?7e5choC7GNBlN
z?1m|rRpH0vb&ou-qmeH<{dS_|C7w7lR%y#pJf&ASyi@vT?KTJqNYoYXPBZWK>_~Mp
z+ty<|R+D++Lzchzm~uuqOepJO;c4w+{DL~_#s_p{effq*s_WXr8W$A5I|KVidb2y-
z5F(FI_C1a*Zn~+_7tYei@rsF^5!>D)87*sjo5g$op0eB*1>D=@7k%YpWR4XWL38e4
z%}8`ItvsV$j#6cxahyYf{2YFBU!IcH79Lw@s{_Th0Y`Qi%qN8}ZD+mR_a_0I`I_Br
z5qlEwK>nldRr-W2eF7Vb<BhYnjU)?xTdnWG#X=ZiM_QYJFP!B$z8%@S6moFTaQ41f
zP+nT5otREs9<t7tYS~uINc2Mxt#&E40$f(vtzwDx8}%_L*eoh)6C1gcT_Jy#HoRS#
z_|&03zc7JBb~azG4m86J8|kw*@~E5Rk`tPRqW8I1EzF@Q7DE+*0xUP3l*=Q&r7{2U
z`g1LzO(*xP!>CiWKEW(nFnOagbI44MN5~l5XX|2SB-|`K{6}k4br4(~yB3#84&hW2
znKW8kqq1R(JF4dQhjBDNfD3YK8<=H<=Zc`%bwYVAcbKw<EyqsrJnn5h1<)YO;O#=&
zLh{u&uYN|MVy6O5+&$I5aRP*R(>8JaR1kkH+Ipv(1w2^-XLrjzjI?n4&E6sZT25{4
zOzGK*ZaEtO0D)-#6y#CW!CSVQa?(1M!<+&Ib@4~Y#t{jH+O0@-JONGS{lj*z?SZR}
zpc$UBc$mCvhHVdmb00e)@2h;T24E92zh?^Z00*!Q6225JFYp?{(5&rNN_$AaVcYU|
zU;zVrpPG1bG#Ay4<5RdIoD$I6qE?D6mfXJ`VmN()Z=C5T8OOvA`z}CnZUqPZfpZ8S
zM`l@TX^sY4<FLH>@@h1}u2q=7FMjmG%4yaW^9dWiEQaQGRc9w_S@BX@$jAvp{?j6$
zh@slRN{55o&^yG$$1d;#lMjm+PWAJJ5R+V59sr4s5_Z|C_^S4bHw&0LioCEkHgA-h
z_4AgDBP-v}Qhc&t$cwNh{yp@zY7m(#uUrb5;VT&lCuZDHyAQWUq}udT3216J>g6J?
zB2bKM^UB>|-t`V%8b0u~Q-1u%MnJ#jysPAVnqT=wA;3+LW%aB>wbGV4V<=9ATwYZH
z78X?ApobMwE3A?7!hG4@1zR`^b}z2N1At_6m@&=M*Kgb2C1}A-t`%UP*fwE_LM2Z7
zJObG@+)4I<FSc)k*EGWw_hoLoT|Xa|U7k4^5>mPBJZ+l`Zc@*PDoQnP#Af(188Dme
z3*u&C2pNeE6>EKXSigd>{o@<S_Zi;4nXEH<2!->Z$nq_!qe&di)b|l2WNJTk5`;R%
z%IQr29Rf~GmlaqT#Sm!KJBR6f`v5x5SVlRbLog-ps<_(sVRI}l74M3(U80cQ2x%r7
zn0?1r@2mE86y!afENSG-!p|MbX1i&=)LgAN=8TGOJe)l2iv+mIxv=pO)N`vR=&FmP
zMl$l@C(Z0%FN?EL#jHf0Kq1w4`N#O#g+xc$8Q_v_J2`6=XNdwGhUEbK8OgM8W*qTM
zIh@BxwJ`i`|2s{t$IXO{{IOGDE!>vXtEw|*=IkYZ=0kAhj<_+)3=g$Q8La=a{ae>^
z3vr2d<NI?bY>nLuO~(!>z%-xh^kp>k4Vk%JNz<X}M$je+ZnB6m#JC2m@&)aEmx&T6
z;9<k0<xqIE?Ps8UvS@2Wwvi&5+l@1_1`vN(s|tVQBid{03mqub<^v1;W7VdaIc!d5
z);;<#`C#7=9;I&J$^8D9$4X<=!#Ue;Pc`TNGPEv*KQ}CraBGlJ_=>YK#ga`3{l!5n
zw#6enR-%pjyS?NEMBgT8lM$27VkbVcF-eRN>>d)FFMhY(iY;Er%pvoLAP-Z@ORWus
z49NL~_X5%4LtcILbqXOveit3oF~GgRjD8+?$b%y(IH6*^q#L-INz$?c@f`+aK|I)4
z8@MN)6lb2V;t2v=kup0&LY8uQvyuARZVB4PT$rrfv$@zb9TAvF2OlkG>;$~0{|_$F
zlJ+O$iIX0mddxR}+>icw3aH)yAmuU0WULJovts7&pedCTv_Wk%S}&WU%pDS|h_7kA
zIJ!tym8dM2%HKGYEVc60Y9UxB;7JLKG|e8*RR!-~`xJg$+5U7*u?d)j3iz@caDhI1
z&=%XHTE02r1{aN6?hIMBkKl$I*2AdSHDF)sjjw6~z?Pk%!Lor=7Iv8JhSXqOT9sam
z`BF$l8QY>w`>?SqT1!}uRQB+0Y<v29(QI^kh2ZP+{^wd%*9e%fjY)XJ_NB~lAzN6u
z(y&~Qy+BGDh@t}w0}4;)aTteQ`KV}5Ca`n)$vk4Bh*UnfdYdauEJ(<)fiGE>YbxFC
z-tk3h!<%3`F7FoHz>=Zu%ymCOvj7=iLG(6JbGbL1d~!l(yAj3Cm+?|7ve0<z-TFuh
z|JcF546vu8$Mi_#x5L-f$cwnzSu9^Md4PWyO)X(&5ZyAv+p(tn@or5dGl}MFGYLKk
zD6XPCB;``Nu4|VgrAAC^n3(cr-|ghP8E=2#(inKW7|L$1ms#ne&FSoK8Fp8^&TUPi
zOP0%PnMTHkQ5yL!Jc0#?qYLGl$cHP)Yh#qm2{wZHCMs@Di$(6VppFQsnIe)KIBaUe
za^P@dL^!Lq?xg8tOJb*{>4k4%1fl}^FT+=<L4K{UvOm|Bt7CNjpz`E;#_)x5Y!S=d
zRc<n%@G1r-vs~Ry6DppI^nO=7Za`|fHxpb{&RJmA?4{TwT4|QM=hyn$qb6RRbm78S
zz08dwbJh3gDz|=kW1ybsr0&9UR$g)3^?MQ%t23=yt`do|=TnI-D9eV(fP|gZcvL;_
zQtmkxJjFqva{l^(pD8Uz3P!OaE@2^|hD$64=;Po>t!&|y=BQI%jn}JZ2iO~kQK%nR
zb%j`oR=RuwM3C|j;E*wPzV%g!?=M5Bfqk@=(-z~5G{E%)fm_vBPQe14PvyayW9;cW
z@Bory6W@}c#xRy4GlVu>^_~^G5uMcS#l$SnDn#pp4tEfU!Tmw*MVbDGO};{5r?#r~
z>Jq_(PBR%_r-Y#xQeFOUeP{3UL19c{2L+bMJlF-kbYHl?<#%;V&pexD$6I}B)2abN
zeEi5e=|6g1H5DAxL;76hnN_$q!MqHaKZNYA-4{D-3c$$I*%qUrU1!bga+Ng{^j#mo
z*l)X^G`AZTwLY86E4?H)q(f@5@#f0h!eQoJ*yi|(s}^l|XX0~`)1;nXh{FFC(6A|~
zXSM*-VpF}Er`jMUM5bKE5U)vIcFc-Fkyz3qx-!#CUoBO3qDgXwhp!-OnkP*Iuywv`
zgtpaSRQFr!$}dQwQg0h9dz$;HtjYn_Fm@ynE!2|S+a|%u!z^SgS0v!G#J%v)RKoR~
zT|`!1MF4a9?$s^idLKaS!=RaBU6A<1>9zOWlN0-b7A_X0&o8qb++{hIGsz^pWbJ&+
z2sPv<SfYj0NEPJo#>F_#(jKt#DCjEi563-4z6pnXlJTqIep&dib&%f5(qVx+M30zu
zO1hb_WKRkyPX+e7l4I3uzg-INZBz7s%~@NKN^o{^Tb>t@AzNk>ZFXieklDTj&fm}m
zPR-@Lv0eMb!Tmh!NRC~7e%06x-GlN<eXQ{v&J+Af(eaLXhj7n+=6M!MK6KWVD`nM?
zgZt0^9@G6K5Zk%0y{}v5$VznIq)08ZW6ofGk&V|TDqz@=%-*pJ7z<fJd@mUDykanh
z2N<2T-`yJ(5J?Z9F4y^9`y|k?iC^_q4porb$JfvHr&&+#+mlYo+_N2v7xkTYU|O5D
z_n#~DJcR}F5s#yYV(yW+XQrlHFNEynY7(c*{C>?zmH(c?dwixgNzs;a6jtIJUa%By
zSMCoV!&aM2O1gFruj8?O)Y_l<=}x*Vo&pnsZ@NPLGkJCtWm!Jft=&U!6wX5H09nfx
zz_6;rDwNR>e)z3Kg6Zu?h6j$Dxv9PL$8QyY9r1XHVi`;YcX4*s=fpliuQcG%f%w3A
z)*u)GRdF_LqP%={)&jUaSlUA#w1VHUeQg>;d?exJCO#3NSDl9mJB}WNEiUcoF*o7`
zq-rd)$hV~<Z+YY;h3Bm8rBF)M#Aq{+D!IHhHZ`m5nX`Fu+fM9&uS8th{s@K8{1abT
zHYRB=WzsI}3#=sh9__waZ>Jo#+}|lINWN;eCEa)*Byw@s1^bBG0dfOv&brr?k}9p?
zTAIOk<==3vBE;1gIfcZ}SL+-N)>SfFV?6#d6*F`&O(B9YF|RIo<D!Go7(TPNQ}|+$
z-&9!TflMoJ<-*^dqwHp9Msmbb_?SfC12<E-ob(?uOUbEn{$iW_^TgS#o0jb*^RtIm
zRPnQ`XrSW$_PSImmmBePc&H$E>t#Vt)%01C)@deuQZuh`$32J4&jc{j#ruSv@TD-(
zm8NG1?qIR|FcxtgmOIz)aNNFXA8-T!I{j2^VWBN-^Tn>?y79uTaok#E;d~3C%{Pp?
zrV>W|#r-P^_?X8nXMeI{#tNo01Tv)l$A8+$+WY=CJl^LHzrDdjX@Z-12M4z^Wzrs?
z;>Wv(Mq&8z#!s528fNaTzET68vlR_yDd~FO6}N4llC?+zU1Q~G<h|pQrVa8~Qr<KW
zIu;3Q{GclAx55j@gi+hpQT4@TqluakK78M1&aK}c*8&RkivVK2mU~}RBl%?@*QR+S
z`Ylgwyd#@AUlpk#ivz~#jL#u^*?Eb|^(h)~Pi0sjOX}GmXM0y{t|nbTPCjVYfU&!)
zj3}~6^v`tUgzIqIvWNxaG!s~M9bti9w25=9BQdV_Z$Gy`yg%|R?4WAWyHdF-<;@H0
zR}D*qC%yc$n##*fc&@F^8k&)r)tF_rwy+s5l?xA!?dNXAF6Nz($B&<o)nXC*D67l#
ziqKU5i%+Px1Gum8U6{W5!rZeYa%`dQFl{%}L@r@VuQ-j71BtdjTAryesEyyA&}C5E
z@<xkq_cDis=UaRXGj=@Hr+mPAoTDsBEw2a5KmNGiSMxcNr|J-1I!u;p%`9RpPV8;M
zP9~4&{-}>!g$8G2gN}vmyR2M!clu|qvn5E>@0)e+os%k+NUVC~LQXIFKUlgFc&7jV
zuX4<l&3%+G%u()Ca%8q)t{f4$a^&cebU4clbLCDHGN$CF1Ldq_shpKcUlY|xwKQkw
z|60HQ9^c3J`z^M8KJU-#c)p&m=j;96#LgibpO=5r;Sx=K;QTyOdt24k1IFL;)s{YQ
zi->=ISA2EXn~&+PH}CsC3G7Lbly{>Y-eDh}gbMb%q}DOH&ESz*zl}uIi9t^Z9&)tg
z$6K4zwvV2d@$@MW3eVdjiHLM+Htvf&s~OmdnIV2`IWlRlBGaX-aU>b}RBXl4o+ti@
z()%#&+nq}d&l4N(3!O<9<TkeU^89kRy}`WOFLHmwz5+Fdj@}-Ficg9Q+*reo5AK{Y
z*jeP(*iUlHcCt&fHd$6b-&X1|cH79*EZB(~eP79eFvqTW!EM%BaoJsa{VCzDb1u}b
zX!H*9rCrUI4^K?T$w!L>iZpNQ5e{|OJ6|pM3qQqS?pKoYg&TYmc&`L-i~qSl@Gu{F
zx&O0h#pi~^`O8in;%YceAK`n%ch+mScL`p4oHH*zaP(K?c}_Bz7tuU6p)YOk^uV#u
zIG(AJo_TusI;0skW+Aox%=%^1fDSgz^EvxMpQ-!3*zck1hd`@v)*<(#N2he@_77l@
zv9%7deK7&L9b>m1RXI<D%a3ijpByrLt7FNu^HfIl;PZ#4?~kS&jCMJHa+&+9{Ef3B
zck&{UgM)vjnoJt@Z=J8-Tym(4+~E8IvEF#ZX(C^6t@cB}Hg1Ren>JkMxO{KTcVu_*
zs%ht8vpQc_4X2#`wAJ*WnULHc#+^fx2{&5yo=W4<a4;1w%RX~$t!l}C*+;-9!dHD4
zyBn+ND(h)Tquo1Zy2Ih*geA=((k;n8ut$3R=4awWK~*oFJ=vD$TJTzYP%f~iEM(I?
zU)P!~+;UFC?>n+jP6W2M_*eg82=x59(l+^mA+4S~JTdBRsn*i(oGgGksdezG%7c^b
z*ZL9W67ylh6E^3SZ^HXR%y-b#v=b4v*(R?&+YXJM^l^p){^$*QkF47WaqGcg#DMbE
zO+snu0;>lT^5@4@u-PY&o<zrj`0qEO_wM~sR)*c)6e$i1sZ_FBs{zq!V4uXp8>SKs
zXJ2Z|*!&7?+8Q04zGVXa+5A%7nQ6}Y`~9xb?#^cL<9CoP243-HS4yAl_p$F-t;=U!
z2`}%`xyQ4s=l;KD47Uqf^IK8O1!BX!&{d%*0bF>(sn7B%YP=<H-o;P_Mem2-vv-p0
zKfmH<{Isdy#*#Vlojt9YfE>Ohni9EFS40?rh1r#v1Lw<5+BHi&_?-HBWY>%nT!QKv
zTBL2?3G&U)S9olMAWvaO7A|WXrrz&(t#~R0a#iaS17oyiplQ=<_CvQsljXzI&t;Eq
z1fH4_%X((ps-|$+JoEXd?`Pu$wk@91V0FJ5U2727tmSbSE4ca$5$tP{^`xAfXw)be
zB=~u|y_wq9w;yi7tw*|%zg7Gg?9kn3Lx`Ra%B@89itp+APKxAHi#)=cFtf{BV(i?f
zUmrf7s(M~*HE*?33WC^FV)Z=Pbo|rioCi<J>DyoO*z{zQJ?AaUj8}Ws;)PAgcPE?7
ztS0wzm7N}MD-g)AQ!A%uJ<uR-nEuz)(5Y~Bmx{$&$IB*}zg%HE$gbZPe|=Pu6c%nS
zIpFgo`q-b@hK_(B%?Q_*8us{}0yC@oH=dm(=I`27wZYo^ScZLXBEP?`H*a(L(fbF>
zNYVwuJL9rXYUykr_+^~3u#mV^yc&9_zpi$#<W|4D2g5Qw<ilK@g+hF12hw(VLt&FL
z5P{hw7!wEwhPx<@zzFI3g&SwfED}YwWi(6_d%iQp)tqii^3UDtQ1r98`s$@Gfg0;U
z^|~TRpwYaxza(}NEK}9pZr!_7^Ln+^r{UbJ_M&}f%v7#)x?yaq;!S@MA-Mp1AI&(+
ztvYF?!({?;&IFZM9cf-+0m<cp5APOyBNcCL#}9l6Iht}+$TK{xbnxQY_+WGFgS|JV
z8=n|7o>JL68`AHf+}JgFJKO>l?<bx4T#@^XmhQIFS+(0smbBp-?Q)Mt)%UdTr7FfB
zjaT!t?B~;m+C2S}Bfgg>^k{|BB#BEYSxLHouI!%fi-65!3kk)Lz2Y{!(&=e`xg0IO
z$CeMPyIFxUW2aBV!JWj9Ruw4cctki>L~YjTgEp-R@g*)hDbEwOCec%?@(tt3R5n%%
zNA?Kh??O-$gb0?KQ$KA#|0rGZh2Ed=@^>K!KbQ2a9tXoe=&azw8qdVzpG${0J?44}
zji$K6NnD0?R;RyN`E`+X5`}k4leZe3N!&_QYy6f!efkTRQitchql-z7@%==hgCRoF
z;<7H>ygb`zJhwz{5Y2)M1T3um9;+Q3iSIcqS$aT3HA-Qg)z=aZJ(TV4-S`AaKIwU3
zjpMHM-H-X7Y$F-x4x(i~^Y<T{WWCgY#(pTk$EVkg_Z`sH?)g>NrEtGLk(77j^f|Tr
zc`izs3H1WKuO~rfHrQ&J<w|`lGON~6V4MS-@{aq$^1Ps$2wh9}q~6kJr-i6h9Z*C-
z^-DtMSKY1ul)g7>{h$o6yZ5_v@uUr34>=o|zR4=m20GrdL%V+q*Hdu|B2NPR!d>bx
zNv-4RCyN{luFh-u)e8sq?NXb(r=Y8G5Z|NLk*EJj@Eq%dcD{=Enb+fcW{*R6bn49I
zkG#UaZWySAu=U5!%^VkG9!lJxIo*2n?G2C(uLXPZqmI1lJT&7*rTNt5Cob&(A>>XZ
zh#xm?;oc0I-v7P0HlY-ex{eAUfz*jVP<Al8(KoDni1lmo|7DPy%oN{H>yq7bOss3Q
zU$x8+w%wy;^ZXuEAi+>S!KjMpR8X1MNq+E`>-q4Wa~?g&A}q2ft0;>?DN5D4Sxwq&
zBH}!_Fr%TVsqUU@qEmzwb02YZbmWdfOW}iS=<YD`HGoq@*eUXl@Q?i486ANQCbbnB
z1wWN2RBP&2qjqpf!Tf@@*C=gxLrBq-e}w2n%_v=~MTd1C%;J2$eLOx;BEN8YKxKn>
zEeuW(s##NC8Kr46u*l+5Oh2X?m8nKAx#OKOKTM_8QJvPfG&SEE!6xhqg=RL;b0Tb1
z-!ck-eZR3Nu}PYG)M@-O+*TfKxWKbek3b+`FY$0LY(t}fT5xB^X5^eEOrvx5a5o^U
zW|1+BA7;+*DPH{I?cy}`OPlq?x@C57$A~)y7Q8AwJHBK>pDuAwO-I^^SPb6gyl7gf
zPOvrj?-0(HsCyUc51DGf|7qj3s=qCEY%`H4M61<0_#sj$Z)yg*ZQk0f?0?$)cEcrO
z-APeK1~2Yr9gRYjagX}mCh;$~8aU#`9hp&L$p<8TROa0<tv{kc(~m=O24gcq5_UWv
zNK+0x9xX=O&3(LEHXysoK!bVS3-Nr2e=Xv^&&sjeov4lHj5-m5J@@1wjC0IO(B~|b
z!z1rvELRKS9k&d!O+>IQYBlg0cuaJfAv!VJ)T83z+S|T*7S0UVM;*(BRRcS~Yc6YJ
zscbcBv#M=&q2x_39+?0?f{=u?(uSKq3<A+37BM*?{~3w!l39e*Wo^>(s2X+NBz@k&
z2&<%|gx)Sa4C_I>YuU5Oy+i~XhS}0RcnGd08bp`bbYu2qb1B}vx^N7_Z~@T?Oo8)I
zmbtLVQf?)<R}#F}J*VL2E*U*b8bYD>E<6jp<NPf6OzjPKbOc$UL)Iz(EEU0&PmGaZ
zMdfB%lvwn^l0}z1;h&zDKPhu-HcCFxgS4zkYSGjLeSC@Sd@MX)J1zKR8`^Ri|8FPw
zV$L6i5Oxyz`T0ze0CS!|W(uH1&AHRuP>JQSI}Hi8>;V!?wVa!>6{o33IOWmZId2+f
zdrAG}y1E<odN$_gAH=TgjGi%0&MSglIawN`65aUGX!$2G+4GjKGW}pA06|HI2YCB7
z34tkF&F}V3)g(AH&;1#WWTKD>wgWsd@Qz~<oMq4JjNeTjB`-I=CVCOx<>=|PVP*8I
z8ft10DuaVh;7Sz{_UUjXM4!~Ww&A>X@%BA7<w0<i&V1)D*y1@GLpRQ)8<#x6!IvtH
z@Cr%53u$d)HtUEFJZ1~Ie(ukR5nMEiW!ZbHaT~LY=`E{SdE!q#{f9PM{*!a-6|1ke
znC*`FR8%-vq%~s!9;`7B!wZ2?YE@gDft?_!$l`ta;M`*1==LeYZK~DO0xGgp-c*z0
zFU$&t`xji&M)zN?&$F?c!7e(4V8|?7GiSwQpb<8;(h@2E#_=~arf<tvT#mU^?~u|!
zIJgz1+mYxc(QGU;LmuU<Scna0LDoh)Rp37h91rDZeGK8OzV;+m%=-1ehi5f$Qe*nc
zT>Pe+PxX}mTEIdqpm_ySDRBR@JnA;{>v=Q{bPhtfBy-k{C#XzD2G-|$#d1EgO|-8z
zzX}aZ4kK}FL5_8}rmtfg+#j+Sme*?8wkl3;S)8VIx6P-A#_SY+!-nr1RG1G{U~7_!
zH={yCekA|p${)4gtzeVwC2!4n$o8qOuCB2L(K6Ems*$E_b1CI%dyg#tgl*D{)giW9
zw+~N@RD6_txN+i|ur>OY1#?xu+ZcKB+%Q|f2_{Yg6#1vg3KTeaR9crcJnK}z;4DDF
z=eFT*|L*l1P5m5Q1|LD5r7^`FbB$rY0)*YEYZ{g}zs~L0aD#Xy_9m~B8$2n3j5J66
zdm-$LQd2or9=A6j)Y>4BHXD`EqIm&hZ>4EFtH3r9V3fn)QV|3{2u6rLY=nrP0W9?!
zc8{17vS@h5rgpiB;OOBJeSAQ9HiQ*yUy6Osi5@g<&o@Z2YT+=9s7MIsbvyon6K|gX
ziIpN^=7+N^`=5YE!LT76?uzrm9Dt9^kSNr1t8yjt87b~AmIx-nih97bT#keDC|KEZ
zO-=n}@T@%8^BB`3%q?~>!<5Ey$HSYjND75gtV52qEU}u-AZnnUn(}C~LWO8;Z$+`W
zODo=QFbSh<v9ODon;6>Lbu2y`Kb`k6GP$SB84N<ed(-1wMwGF2{KFZc#RHY(&2n$r
zK6?1Et$DB_PAzJS=;_R2B#skciB3%r?h4TgR6RX0qL;QUCoB~)dh^;h=h2A^WT`%g
z_4e0C1)f8wfqxFs9eLD=Zrx00Y^LA7H%hvkaI#~=wcw^t|3{>`BX^iDi4zN*>(D$N
zoN2k|%rKm9Z~8woW`ae(MPQE;z0`@`6!;!H3|{*&q|@yOYHH{?{CCtk9Q-1Ui_ZF7
zl;3>O#D-C4pf-Scw|J;30=DLpv{X`ur`}}ZHqU0$xlyPX9wiMAI=hW9?2UC&C5O2M
z`0!E6$I@$&<du3Hrl0?9hPqAFxlUf&?CkuJc|!SMHL1>sC(qHn#GxIo3Jy2h;SnX~
z=!R)DF0l*ROqp^@eF0uP5*$6-77`V5DV-b`?)bB^6@wn5%j}aF=#^oT-+Y6NHbs2#
zFSXipEKS>PxeTm{pM?$e?ai0^#O1WC-_5k=9(43D$@D{^A_R=NSy3f4)<wbF=_S_F
z&r}8AZD1&VQ@?Mv`K%pZ=ex7vk?!boAf28wlM{k}<0Y!d>va@i4@;DuPNiVQ=myaX
zaWI&QKvzOkFxy<<lvh)EPEa#<Y{~ee;3J1kpG+K_An>QMgK|u*r^%3xAg5GYmNDJB
zBNQ_6aC>J*ABk+@@yH8kU2)I45-`_F-qp0!YrCm-UBvX=*b$#&>cYW`gZa<Sa3k?~
zq?t=o^^A$^5ZcF7SB8B}-1!Bc=hzVb(VFtXI)N7@4!GfqXu0;iGo&mR-|j1I6y=N~
znpha3McAi$aa?y8955;fuT@~x_gvGVRT(9%p)rM*pf($qMlre_w*}V;FzN)BN7G4q
znf@P~Q1Ekwk_>kU|EX$ACMkVsguS{W&GQfC=cu|3TV1CEC-_366|bt1I-AwJGLmBh
zG4uci=1g4rt;W2cv3K3Hvv<cP*nh$Mn>FtHFKWhS<-!ysFH0rMhWt6twPilOt_nfH
zO>}L=F$Xj{`^eIpG=GV&jcq^1rg?V?Ba7YK8pYbv5B2@B6DvbIbYE_j+$0fJ*pkS=
zocpr{Q6+dGgRWJbs`yYE_m@j*J&7LSlW()!m_-f<3|=%DI6GeApg66X(1q`o;`T)L
z=Fx`SV@d-S9k^Ko<;<EW`J$u0<^5VsC<-}367tDao)X;=urDAOcI>rj$dLf70HipO
z^614u2f9gA{%`Zms#dC`fmpR^@a7h!Dci!4pVp9pq(^uWSS>`VF<%+;qF@>Kd^9Q?
zhci$3R=;LyS+4rfJaE7M(ON^ZoA$fY%ed`M40)5$;~(v(qS`dJ2+cVvrV&Uj+u|ti
zmfoykYmQr8RuOf^iR4ykeF}Dz|I5X5Izj2r9usDxKC>q&?M!v$b?mMEQe?ya@Hg_G
zZ{_U#Njp~WdQ^P?t?S?ZIA=*0Pt(ud+<UuUxwHG7@wI2#nWesS#N=iRM+UvP@a$a5
zu0zi@2NXYPpKxXMV+41WAVLiVVPJh#MsChGu;}Q+WzDW#XY$8;1@AvRb<{0V=g;M(
zrxN#@5kJP<azl7EUdtWYMLCjm?TP<p#b*x)pVZxvjkT?k(Xt;{C*-n&AC`{VEh@@l
ze?0yD<N7_H-JHzP&Gt#kmykeJ5nr%^NKLESxnozXG2Sg|^c=l55G_Kgz~Qifl84;g
zz0<SVn=BhWOI8Q1xd8wH(vyHR^j{W)A|6TaM@;hs9+r>n|MClx<mE!sqD48)khFxt
zVncGlj(w@AEr2aM1eW!y;Lpdt*26B;6?SP|jq}e(&iv^uKS3^NoJi{G`Y~9s@asL=
z{>gV|rI)X4{%jdfiR5?rBVzP2^}C3Y@bsfAr%Kd5$k@I!T+L5xP5WSS<%yVL?R2N#
zSSz9ProRIZ{zw2Z;6LdWtz_Z}TE%r?tg&!2;`LIb^!;meDYgA=i@IHpkJgtTo(PxY
zF<VEK-)G=<{mN)o+h@0qsN8>O@t91gTQ>9SSC6Z_HGfDa+nymarrgMOhU}LeUQZh@
z=mvDvylr`iwf)G8wRlkU)%E77_4F6*{oCksiI4PjOxk;QnIyPok0zO3P^<iX$H)A^
z=f+6~v<g}{c*wEEbgQ6W(~rY<gmR8#?7bi-S^H$W>x{QdoV@ENM{PzB-H|)oe7h~}
z7;Mio;#v`rXgV<WEEfW=egzvNx?|th!D#vEvx7KP7`{?SB1ZT`4b)Rc^$_1j7P{SY
zjj)j;bJaQf6mpEVOB6WDzzZ^Lpj5qb7S6Z2rhmjeR|!IW6c4JiKWe)~_e@z#$&=4j
zEV1rcAHK`ws!<{B-==Zg&ZDK{r}R7`!;J|ZCr%pARyD_xCyx5n#4Maw4%F4L-ulsQ
zRZ^ocL#|L*FWua$ggg1QW3Jk<|9<S;N(ti0QGCO<<omGwMYGJghP?Z~>+2QoOdNe=
zenUk0hK%x0bcucMCd$F&@mCqQ=-e?WAy>|fzx+?gb57;7zf#v5x$9;|%1>8&Uxaw|
zSoUG}-HEhZPzv48R!+Nq^H<X0+SGSabnyt1Zr@B}aB^<V4W)0wWQFeyDJF`Ujh_rn
zqjcZT1xb2)9U~+LoZ9>Q=$l3H#O}1)J+b|Vt=!`*v}*ryZDBtPed|TR0&GYAi~VSl
zK;rN+g~8U$z$BEdr!#$hy*=bNM+a9Qc(nTqX_Y$%N~{_3O2{JM;}UlOPFHqoG9C%u
zzL7e;XmS^ZcHwbAF#E-F$`7%rvi*jzkLSjXx<nc%u=TrhV%+e6g)j`rpv(>taH1WU
zRwJ=^rVAVtma|*k_IkUiN#*{5xQtGJ{?f05U*z8#{}D1_AHLVuuemM7drGXrrq;4=
zS%Ilo)f7MOwuSM2_M?1-CuMxx;~}~Iy!bW0a;d(rOwl@8m;aU)``{Z7nKY@K@prbP
z8wFUWroA%lFQ3hX>`b8;zEC=8SG~M-fxE9Mpul`-*&>3Tr+ag*P+v;rbr<>Cpwy4;
z-_>wm5BrPujqep3W{%C18HJ)z?yp{8@zk#dyRvG!d93|i?`S_wN@mRM>fFM9`4o@m
zR8QQ260a(hbo_0od6hAx?~L_=)ehW1asR?7+v9^PBHI*LutYsNEefRx3*!Gu%eAl~
zIKJIT9YC)PB>qTDP9!AZ+qbI|2I~a79vN|?=6x^^9ENB3e-T{M-jL8dtf-j5Vkt<M
z%uX%=Komhl!aemIz0{dV1xoMTzg!|+&*qd2<xJb&tS7dr|F>O5#v?Szdg?&No79;0
zOku<9$j1>s$J`$bQkNFKuL<u=)=xDUj}>Vb(?279(NTEc&s#@Nr=*M<VzQ)Uonw0+
z@e{3{pBX-}hew657+04<X?oGt<{xObNumRpZ?`1liAcV)b1d}h_3ZoRY9r%VwdVTZ
zSMiFy0Vk}k?_s93+!sU2JsFRgYW+yDi@8)|@~ly+va#A#if2gM8QmNwGAn~n5?!tW
zb|l!E8CB5DFy@)8lzDN3@<jg0e6;-2ujQXGuofby2uxQz0viLx>8BnsTrh>w%_#wN
zT$~D~Xm6U$c<@>+y7VhV?ag%trMuN_SfqcoTrN&Y=MXi&P@yhLS#dWFua#FGh{vJI
z&AI~8`kIs&KQDg!%XJh{zeRWSTVPJ%H8mB3H!@}g&Qi%aLPYM4BrE$(I`l8A@1uJC
z2GB!yP`Ab(A8)zi++E~2kZ$7kD#QMGYb%_bcx&PmUh{<A51Z?g)jPPnv9*jSbav^9
zgq^7I?Hw*ZBoDof9$fhH9%tv7(rYz*Qpw_~dP>Cej0yjSWZ*&hKT3YEJTG{9Rm|mj
z*T|`Uzp$9dsPtEbKa*^;d`>fn>XTw!o%|bhwVcwrJ5zP9j)sKb%`*Sqi^Nj}t=?a+
zEt*iKuQ(Wzk*m^EHjLF-8V)tYtS4CYK^<D~OFe2wm8Vov%FQ~P>7|h$=G=p~p%Bm{
zbqK6-2Ly(?5*!isH?o010p7tJl}JrhtG|=izBRpY6i~U?q3j-NnGoc{U`p=;Y91{L
z!Ue_(hXW}5;OmKP<<r=w{O1_~B`@sDzwsJB5^_;y+Y~g*-YF+!zM1t1I3JVIV=srz
zGD$zhxbk=!x9)R?;Kkm^#A_eTyowPnt9}%mdTs4wd0XGqi}TOE#qfV<KC?4|ZeDAW
z-rtz6dvGEEJ2f4guCeHSve3{mw7S&q%N2RMj5FA_;uj{xA!&U}sGKu-OT$wS9bWIb
zYSQoSo9lQ=wMS0N)Jj~Ay#MKyDZl%CF4u+AfBZ@DqwJ41U;Mc2b*){8jKe|lMp`Q$
z_u*k3WSc}h-Y1WJ5{zu`FZ(6-F2p;25^AJZmRTKeJ`#e`vDcxGtQYhNh&}-A^G<69
zB!q@S^-aW-ZFVlB2wXOU_#!(`k*C6N$noG^@5EyffCO+MHFQpN9fE%qDWbRu!u4NI
zBxN!@F4ufWUDd_E7TVv)Fj9`YdX4X>oI<F%@#fi3Yr>dk)?u}3hM~gCkUmGO-{T9C
zAI&QE#FvZKe#JRYTjg9BdYY|d9DHv`^K2sl|K<FYteFiX`sa3)Ir*KBv>xx=Yv|3~
zCk$Ia%KDv75dGMwZ;G5a8IsMgOUkHQS9p6~t0$7bc&d7TxM2rQy59}<&5@WUeA-T?
z+-sHUKFjhb+{OE<Tm6xXHR5Cn&$8Ob5t~8vg<%;P27B}dz5nJW|FJ7pqqh>eEFl}T
z4Hw*WYBuWO&;i<${-4@n1zZE@kZ+CZ@@Q+;Z+42Bqjm?vOWe`j!q3DiUFy@0!#~K9
zH*~yY&~Wa@?KZ3!kpdOA;BcD^dI({TKK<(WZ%N^<)vdMT*M)}{rLRv4f7Biwx|427
za(O6&c&U{d!+m#qG3!(;)v49ug3zFz+xW4>&{*e%b0=rt2@{P>;&l-6s|FuO1??HF
z<#xW`Yng$fv`0T)<X-t=*D_>!^1SS+T^$0gts)n;-|Kr=_Ti*qPyw>ApXxE7r^LG=
ziNW4kR5?fq(pC(m4k|w2mgu`z`DOCgot#e=-ZzQUn0=PHCw8^{L0HdWTyt>`w>I+e
zv*fyF&riFB@0wycwzGHX!VVTju`*1E=@Kkg*zIBkjaFRr&zUzaI!L(AV99TJ>me|^
z2DS9F$3zOJvLAdT*tMXV(C12-PRr49@M9Wtheb(r`*4a-fa@sf;hBg|AxyugM1Fw{
zZMOXY6wt4Yv{Hj^Gcr)M22skpD-?xL<toRYxV&+mbCjnjq}|!r8GNEWyI4Hc7uOUK
zYw-A7)v8A$<B68@oT<ZhL;36m@=Srz)-_R7dpv9UxOG#K-DADQnRUl>dD|}~PiYuQ
zRzoy;S9YkOGci(iQ|bk)v8gh-LcLV(`^+OT(tctOscDK@LMe;XBJIS)AGRfviTW;q
z4lUV%7{-#fP;_<p^%Ftc_3mBOzXN*$-TiVSi6zl`MdTYJknu+_Y)|m>=c0F?D8+th
zy2>P;@JchZlfB)#SD{#C(fZj*o(DeT#V;lcslLx;P=;=2kL`TZM|@bDTKTb><@$NO
z*J_26CO>CSm*6~6VZN~(+Z<wJ;`fPVZ%zz1dDc2-Zse`BS83{3hz(W1%MVr)-@Wh_
zxJ`hKIo8~`<xmBZfKx$T!Dp_0%<}={A)_V&!Ob3MLu`nnV-zr}j!}@#WpKk0yQ0&M
z=Gtb{w6aJlHER;TlSAIixP0<=7n9L3zrPdlZb*mTD}b`3^DKE#BN&INexmcmGFwiv
zALg7^wz=Q(o<H=Md0<cekv&K2EnY7V9r<As;9T{SWSQ;LvJh~7&|&uNtkd;`6FXK7
zDo?p-A}=~ll@#UFSr?qQ^)Cy4@d}^m|Fg^f>tsrymao^&oWL{dTh*HbjgxxpR;8r7
zKa>mY+%JY4D?dwrMNK-sf8gloFp1l^BgsLvx96cxz)+ai`9R(!#_3^|oa$Eh%^#{}
zmz8CoikIt&Htn@sMIG)xj;{D7M?WI&BpK)ERO8TnBxRkfnznpn1_2DTa&|Wgl>iJf
ziq0Lnp310C7<^=4F0(^<9#6kwiyFW?4r+JDqBvygwk$$zm|OP@HCQ_*^JFj_(>q$!
z)YO!!Mr|<$Xruo&>KFwp|8W=hvYW+FNGnCFTt)DRlFv@GCQdcwcdAuQemLSu=tMmI
z>{025rj(Zy{Ri8V=z{uUhdMG4R(HFDG1-BRh2rv?^!EQTlC{o$HI#bWwQNz2>2C5W
zDU9_@ajEhdX6%#v*r6MFOA*I|caho`{^X|w9LU$|vs&i9&^9#1Q_xo&_p%Qk9<k3j
zQTlQ7U#@uK2fyb-`l1r9dv;|nSx`+?_RA<Yk5wBU{&MAkoP1ahHt34*&3k8G2QDW2
zogAMK*(BGvN4QDgNQYpx&IzOG;>6Q&#8uI7Rp#0KQ=9kf`IxG8!)IMe^WHbgMMEXI
zZC$MWhnwWF#al|{vbJe=V<T*q?>Mo);$o;K+~FogWJZ)kn2Wj5YXCigrqk1|sW_3?
z)5+$t0*qhsn`c%;bUjPHwkl^?{zKl1M*rwA^-Eg|^{E1CnM9u?$VI?nnv%w2m9uxr
z(~jFPIK(FcUps_9W}BU*{Fh76;J)PCMQ-^X!{cpoGANU*ndU8s=lEj%OE~NCfY(We
zfj4U`vXMCxmI}Sr&P=OUpJ<k$)CZSW!X@92?>|43uhNw}*n2%^Z`oNr<nijTz=0pT
zx#i4tjehaqV$&&44Ib<>Qrd%fc@nGdJID@Bx%UD0!FT%2X9J3<vQIW``@=J-t$J#@
zNhi{3D_-e-*eTMi{z&kmE}x23c^SVHoXT+8*LG1bTzFZh`RH37QQ;Qj$Zeym%=(76
z!K&xg#yXf?ER#pK&l8icZw_r7cc!j|b@FcTXnei73Y~~oDSeqxiC9nfrrx7@X4AKC
za^S0)jmk6TOR$eJ#jcpJ>3o-ep};5svDk4+eGXUdg&NgRYm>n|HqvwT5~~NkJ5U>k
znX*rzA~SD7h2ygg+Hyok<BiZ*Jsi%GFW8{u%X8rr^@g!hrJ4Z!vXBQox(wYrVhVqX
z*{+kN(jOWp7<a}e95u(rwnzpaEV_s5-&W$GFHClMc<P9Y^VoWFM&M8Gq8BcP_8AWZ
zu1X%G(?8hCWH>(x;JzbzJti+~E=e-xjHsHr@<DtQqp`+j3Tvjizg9&?>eJ~6ts<ua
z52q|k<#}t_5?YuyVB@pKiJATrwplnl?1`GAm&W|>StECrIr!=}B+wLhfJ6h72nhmq
z!Wx1OWK<j*Dz4RGJ$S)7yfnh1vhy7A;f^>k9<U%zQ&5q=g4ua}H_3fN`@&2lftSpW
z-Fm&m%HXO9-(09l%RLkFUIt4zIPE&SOfTpvenKu!C*Vo;JE2ieen*w~lY_$n9*^qR
z9tVD-)iif`^GsDr$0T2?bMV`zK#C|&*mg-Qf;L+rMps6aj{<>bvKynM%gehjP@2(I
zZY(oO-*Fh4T2(eI`x}LtO!?%p2>K3Z#eb1!VKX~Xn;sVmfQzQy3B9CxCJu+Et_eWe
zv{GPelkN(3u}yC^>FLuKWAkXN=fI+t7!a@#PEh<?bNEeTzRQo!t@laXOqf>9EQ`nc
z53+~mfV_o+AD|XvK`RNE)l^`Ce4g@Gz=}wZ@Mw2!JlBMNWrq;Uo(Bp#f&j*$%7^Ee
zL-WtNI36tuhC_cxDF56*NwFpvCK?JP{*hTgkrzemtot%^T8#_0WkKU}n^bp<#IJH?
zXmCY`a`5u)3{DsB7xkjN^NZ-*@ksuDk5bgOki&I?&m3B(wR0zIVFx|PRHlMMB|@+-
z@mn4|Y&U2ku*^iw6R^Q#+%nL+sdvqlgV`hC;PbKH0PE{_%V+v2g^E$1vb@w*6w)3V
zlL>oAfI=T7#YtMRqs6l4ypCe3VxYl_7X8cB-!#QMuI!O_k$;N#>@ut1%C#VLpN~<4
z0~o&4yDK(*kJ4^qGPE}_wDP_jDEpyY%CRd{H_A+84!XD`1L(^^DN!+Pj}MR5ABWBX
zr8Ivi0#@HYmlrrLNW^F~rx>x)1zL^w;%H`0q%M6fw!aHX{)?eII~;;4tuUXU<}XPu
z^9*+W;fCpezL(Y=NB!uD4Uc>E(CR{#CGg<!IOxZTeXty32XeZHQb`mbco)$**08}X
zgF-1WmG*t6Ds=8+sD=U!Si5%lXTT>k&T&Wx)R~b>8s|pgJ7l1zj8IZCP1R)3ITUw^
z!%3-lS%EpNf~*dQaDD~Kvq>km0H<)5k@08%-}e~x&W5skVFb!<2csW*)`8f6Txr#(
z0<C`zJ&xvk%Wx{F;It^vXb$P&(B0VuV8M|aH4Gh?fPZLTvG@$#5x7>jUdnqZ`j2Q)
zYi^%Hj;pOqY37bEDUDL0idvTZ@|UY~?ZHJ;vUaN2X9(n!^fR^U?lvpUR`9F3l|te!
zmNt&=pcSxML4O5brKGUn2dXJmKt2YLv>72ppCTY`biBAxy@`t+@=vCtXL1m#xt|<V
z|9x`M*bqN;O>S-5a0CPR1j~9$Dztcqa`eRLz-QuraoirRib8$^!2;;_PVV729(9}L
zhzeA#rxZ3E{nfIO8(id<Q(lHRO+yZcG~c3e`VW{;LFbo&o(hQw0ZsjX){TuEf>oZH
z;(T7r|9_YSv=jMFBAk%L5%B3gInWx)hwcZbFABW~Q86cI_rTR-umK;Jz^!4csihg8
zTr<Iz-Uc#seBi=LP1GX@|C}%=K{)^bpDsO8>M9w+1JjKs%!?m9&D6nv9JYEZ7Vq4x
zMf>*{zYl^z781QqPyH|h1=B9#M?Cb7Vq735b_yx-SMgz8uo3=8IB5XH2BYBNQNam)
zZbm4C{O93|R)Ysw@}e_;a<gMfrh)D96q|>Yfy=gpk4@X)&AXkk&w>4y>qU2j_jT4l
zNghpei0Ff9+?ES_$n_@rv4%vl+W8u%P5Nn&!1oT;iaowHexrZ748#SZmf+gYekmzw
zPo07_viSx5KL_`V@<rLuiO!aPuZI8dm&@RDi0N_YrdH?F5$`Hr5y392dCLar_uobz
zxUq10J9pZkvC80$0mUys6&c(jSuxT`TA(5M%ca8UpTIgNQK|eU4%ilLVdN{zZ24*a
ze!RFPUuB{IqwO;X{35Mo1)SGA=O>0|>x!JR>B=+gz_5M@q2y@_8>wLLv2-Qj3Dmw?
zH@;SIya{X<GdTgMC9ho_Y*9HNK=BH(Re=QR2+b)%ge3?1NhuF@vtuP#0h~((alzTj
zbJ{3wsIJ9nX!m5OV3>WE9&8+0^=H<TIt;XxOs#b3`$D~WJfNE+SiVn!9s?+cJLDqe
z-`)HkNSYz~Og2GNRMpBs8S`nF38!;JG0L@Zff7~R1Ov`(8Zo-B?A9|u7$=5DmrTz`
zLcK%U;qjH4Wo;sxRt7xA|DGQXV?PBVj9u&47uoZKXasCbX^t|gvRS^sV)<A8<=X8Y
zfmUl8IH?0+L#&?Dc9ehCsvL@vo^mK>uEI$-)73c!$|WL@MIiC+fMo}k6YK)R7|zK3
z_rbgNfI5V;*XmdB+S1Yp+QwG!(Q7&`E-qf;GzbW}X8A<O-&Pz?=B%D^Zgynope0{b
z3?RJ6U$k%1;q-?XHW<Y3ABLCa-##3y*wETg{`SuWiVZ4lv*$T|-R=x+I{zqtr2us7
zcbf=kQNtmWlvpct-L~dL8otYIVJm{ubq?Xy3OR`MOH!tiZ#Ry-TPXN7+P1y}1WJ(M
z!H?T~(%>2b!NaA8935p%B8Ko19I*o#r-(q4v<x7b<RH{KB1}@0!M5OAp~<QUT}!Cc
zv6m!>UNYw{t-Y1{;#A-{BZC0x2B%-x{Xfh5pZS?=&Kuxn*zt%takLy78s9eOb#`$m
zf|(t_82NiR{pEtQj{7HRc)YqT%d#7I0&Yn3Bu+#eW=~P`+R#q<5bp;`0@I6wOA2Wq
z3YhZv_u25THDW4cYR=D%CCVpk#lX~U7#+ZYLV1nAqIZ;nQ+>1fZ5Ol=+JBZ;-2t!#
z!<uC-;|a12d$N2Y<4`vfs%q$!PY;+$07@!k4Ded@4fu~OVTut-jAwj95g^VGIQOck
z6@6Qr&eza%HyTOEgOq|g4h`oh1yamicv7!Hx>P{)c}yM-F=OlCCYp?R{QOd1aBLuk
z6D<Dc6ga|O)67~;^`(`t;(wu5+!5HMi1tkXv69Q}e1jdb^0&%7rQ&lj?_#p)viZNL
zdJ9w~X1jPBCs0DfAt|9{z0_^i+SP5S%u8zz(MxRw-9+a<Z<%l;N9=|dzz$GZFxV#i
zx1p0CJ=-nS4?3m#oOU4*NOP`b&ilUcH7aMB8>vUQ^;X1W`9|xj+Rm6UW{8|MBmo?Z
z#N=+9b<yiwIdA#nTx@BSgYsG*2OUO==*HyJ=Ny#J4%(U)X@J1OX7DtpQ_DX@^iHW#
zhJ8N6EV(^{A+`^>bC_#1yr%1jkP-XfAjNcWQKt=I&-`!#%~-z8t$RxAhAFf%Q>MW-
z#r|>?Wd65SBdx($)lT?UmoIi}v_pXLNWdh!<Xl1a+~9#mhPG&kX0iN}ERMxwIaoO@
zh&DAVEArQ>Uy`I!1yVaVgDpHyrJ&1(@ZzqnUN~`AU|gUVCr*Rq7~%<DR08XnS}>cR
z{#M0?J2m5;PZtts4Qi;)5LnhT$A@B-&^!tynU-bHac9^Y-2$xYKVD*G%ts+<A@0(&
zQWV&Msq2|l`%o0!SE=ZKTs&Jfb%8-%LFeql*oB)QeBmu)X}R0-B{+UgCJ4a|L7Igg
zJAiV%U^2#QCFI<zhEgsNytD<pYAA7N-d2JizKCEfs9n4pvIIgQlRhOSi5+Esxs0Cv
zzCgOL!FSP*|7Mw)3e$u#n4yi@hB_XnUO^B-TH@*ip1&xZp$)+S_t3EK;ON9QizpeV
z9gf!HIJ`p_?X7IaWJNXvl~Vt)0p)pfloVc_7(m~la&{10KgfOMVDULvBc!EuHE9;U
zdo;&N3G7H2Fs1~mw_x^?vEw-^GNVmbIQ!)AI~TD|jV0lc*r8nE|1#8~FX=ytBh>wZ
zX73wUIAhc1utEt3TXz8V_5f7~wk$#-xIP1w7yS+NG?`ZbC-DkUhHSvQYzs}<w`s>X
zi6V<b5fUN+Ho*xWMS%H)&MbpwG6@P*km{&Mw1J0f_uCArwNd&u^fn`b-Z$cOLTE!E
z2RerW3=Ib^7{opJoguNf{Z?RzgYeTxL1*FMyA1(Bvk`QQW-p85ytg;-S){BCExW5m
z0XiXe;ZO2{35u3Kt!*CeN7~3K1{RdzRO0)7u1Q6Cl)k%S#Cy@ip~@g_dU1$5yW1GR
z&iqFQr<TB&OzcH@EcD)|7e_AU0~1%M2RPc3Nz$GSA*QC?8p<`YPlM$37&jf1i>U+E
z0s-8g6NHeZP(98M12BNt50p81SPV8gM~{6HotXQ}<#m3hk(wpX&g5%YUNp1B|3|u7
z+`X?>PWk)HM{eDUCDE~@ix0Ctki8<F@@O&MSIYYLU3L8w{7}z-`$)qVf(S1YclTjx
zC&S^l{59Q5BcQ5e)#E@MG(>b-)&`Iqf9~wngz@MmYXCt{d}PIvlCO1GPZvp<^HqVS
z!`&UC8_~OHdB{M1b>3X6+lIEL&hmXWISjfIRV(`MOExR_`)uSIK5taMeRKUwZ)ro^
z1AFg;_Y!kBmD;qSD?=HFWowd8&5dX#+sovOQwve6&;A_$OqIu^r!|?}>}qXbNhbYK
zJVEX}y}WC~f6*zAdJI7KIS3mztR4ukfL4qeoFLeDs(^sWpaNAsSf8L8Reyl?4I1KV
ziwXpgJHRiQWr7@ngGt-NIBYvs%r<ThyYo8nlcutv&fxsh_3cXQ2@6Jkv9J1UDEIAD
zKZ)i(OvU}K5Y4E4XKFbzHo_x!-3`Cd?h(7$@@QgqK-nShCrBcCJ^c~`R~scwQi_SF
z=74xOo^GxJEEfs64DbrbN#L9xx5p+TP9lnsH0JC*t}!|Uf%Y=OA~-_3SqkM@Kfp;A
z8G}<aJ-(bp-f&x;pmb8K#r<}Kr(O_!^C*K~YkD{J($Ur8y#@{lEO+1IBWjfW!S}FX
zVu2%>qgK!Clvi{t7Rb91@2GLBBVxQ4c8-e1KC|4?Q*LxTZmO*7yZ&FUO&^$1!%U8S
zncDGB-?;lx8KgjoVcte2lohd86iQ01R-vAQGAf+jmpod=_{=O3T`qZoSb-vl+b5eu
zHSm5&`^$B2<JP3Z_U9>QaX(Mzmu*WqOxIme_qCO^>lXfVeDC3ET0L*q{i%xtuJVj0
zyQ)Hhe^ERQ|69Pje#X5N8-~+TZ?tcZ2-&0W$REm<dO)~HuAC4Pn{C^rekrXauhKRR
zDg=o4bg0K#mgj?Ts4RdiP*+r%Mx|1wGo&6vC{`e~g4*bcRr)y_C>7Jsz(qZ1`>0hq
zx~5w9@k;NzmIVT1D46B6urqjh2~$euy$w6#Q9RRZ48{}wn{@K&vcy|?z1Sj|abx^M
z49Esb$gyb;rRVo2-KGj|JU%n#^JGq>Dh>}|B5wakj$Z3<iDg#u=pzKQut^=L7@_%~
zoJdTQQjAzNbeuV8fUSD7e5}NZ15GqIPZs`J)eA_^%>8f^8w1_}%4iu&GYeMZLz3D@
zf(i;TDRS+SMR9Wa(||eMyW^Ed>TWbFf4e66+t9(f<+N=1BaOUlqq{{<GMGcII(2C$
z<;a3W$K!F^8tx2Pe`1sBc-G{0b5}`i#wFmHAJh)l%X<HAO%@$`ggcRgb7z@2MfB?1
z3uk^xn5B)f)qVIl$Q82gJZS8~L6Zd)3MD;^gMLa7eehaQG%Di;C}BWWdMoSG^Qsww
z>c2LYwC?N7Zdm@5<xvQvud+%3bAhsp^8o&}<}H<RrL^UfCw1a3$UViF@|-^?Z<@z8
z&18#qdH$+g@eq(xy!_*#>aj4hL6Yb-{l`scl5EW>0x6HM-{P(G;0Qsi_T}KkYq#sB
zLVgatKXT)_mAlWDjK<)?aNRPYvDinKxUYN{%^VfGcJ^lTk0C)%gNprS29n5mlg_lv
z!?**93<9Z<F*IjMhY}sMts%R~!yL(hX{Tx=#LVNFJ!A~WOehWmG<7D_TL9Tgo;9M(
zjPf`TTE@#43-uzPp%jcA%Z-N-gbSF*qRvuzn%_?DvDwkk5~Y3m=w`;Ah-+!g^)lz_
zj6HT4vNcb`W6Qz~*JrN%<?{FZ9WQ6#(ZRkh(AQv~mzt4YT5~E1^@*8uamR&8In%oq
z*#GpLH-!C)`;ttJX9`zDMHPRwuim!3&bA)bunMMMlGx$2Aq&8%^c|Zc3(kUDQI3Ja
zttry<MYzkMG7b(6N>H`x>%+#htO+>HIu%Ubq)>YEI)&9Q@o_+Rj46_n<{+ru`tO>K
z?=xNy@pI}l>zja^mPU+da5D$aTpJJ3dO@`MuC1hbd8PHboW~_7%^pwK-;Y?g+g|>e
z`EGIN!GT5on?ugKZ~e(Mr&6)w0m7L<%=Cywv!```xni|lf2}@7j_o^!)wz_QuIHO|
zXg%G8XEkHsOzOm$lhm1>-ADd=dZK0I;>pzwHB;J3E5x+!D?(xZB#s7|12gpD*5<(o
z9d?qDoE9IALmh#q18(KCgMXDF@M{Ll?#!Nc8Ax&fBh%h4jVL35-|Bq;O+WRB+LBIw
zQG%~kKGn%wF~Ri3O^PP|I<0xp&-~7iqF=3SO{;RU)neo6hIKpnYk~I6Q*7O1?=$~g
z+w3TwRF%3p7MD$~S3X@gB^9h3`*gCe^FyWb5%0Vz;aCghrv2xyt?^mkl-<@z>PGtP
zS3;JL{gRLiZdg5)?UbJUA)s)_Ss@?M>*{N+m*UT<6Jvp{>m*zRP7oZ3(=-BwK!hV6
z2h5yQ_rLlQk_&4F_#33PQ7mY(IDg4$n!_s(8gBF`Y@tH=#4SZL$vkMfapXvEiC+2@
ziu|K<8xnaQCQd&`xDQRaSZ5!#lifq8Kk($vp3MXvRb1_J%;uh9pC5)%XH>hz2)@a!
zH9LK}{cMnQwtLjl-Ow^6y$~VMVysN(r~ApH9Wu)obwB_e%#nrlrWfh~Tq=2=WiG{y
znf|ENj6;=e6N}#~&3MM+DFyOqDaRaBP76#)X&12tw8@VQOss1(b6PYRHS`jPraEFP
zn$CQ#IfawQuFULJQd*fI0%y3OozPtk)i#V4gxlefmV9NN5-cyFc}qG-nE{)DY33K|
zmp5&oGP+C$zoRWIkJ+%U?cuE^P%MU0PrG>NeI+>EL3d=bcF?0vA32jom~^d`Y{KGn
zI#b@QxI-9C2eMDmEFS_Cr=I>y)5^p`5P}SUt{d+upU86S&Q&=UL0_VUAB4m)$bz;e
zm3ajKb7>OTO_b_2-EL6t0;n7kn)ZR24r{O~e->mFY>Za_GGTXgCHIdL%Xk**W`LSf
z+R<W4GhtH2=<s({i=QkLx&Bt$0i>4dd#@*Bd&HfC>Zq?G7B^Q?sSYn5J!{O-otd`^
zroEF%Zg<Z~-ZHL@b1CZCdO`Z5>FwZ}nH=4@z^_MZiz&a$cV8EJUn}!6M4d0;>vD19
zg7#<d`aIeOFT`ClzHQeuIBM`^J<CBV29N-fQo39FlTXjbHU03zV$Zs;8Hc!Q!Zb+O
zh3Pwwu5J5qcu%Um#r$CHF$2*(w^ctkeDqF=u<-PhKJW@Lc7z!~9l>4kdfGskw0gO4
zZS#=arnTJH9oN=3o#LK(RQAnSx)~Ko$xgO8#=>_uln2HRS+I>q@<(au#M^N<f3fAR
zwH*nr{Iag@uZ1^wc)L;FEM;s73TgE?n_57i@I0B|jSx-)2E0Jl3O!@!fWW<Qeii}K
zE^uItrvUsJFk!rMxTSJ%!r%gD41_^v25=HPsEIm@IXAEUGy>Rf(0eN)9G-k!(*&+r
z>BM=-Zw|uK;n8xHto#SP)#$zlqh1W(cA@-&9Cy*aM!w5zyR#C?7e*w%3|LU_jllsj
z$qiGq7DG(b;k=gQIbp9UhcPo-;$i=XLnX5oc3MP-i0#2Y^iNXFUYDy!VZO<&D*KnM
z64o#3AJsX9UvpSxSK;~xL2o8vhl~_a1cC>G6hs{yekel%MKdbs_LLVs3~h%FhqGfq
z6z>DNmXVh5t(@0L0Mstp2ivWHopLjBrhhRm5{rkB7rVoQ!J%%m%Ljp+j=-S%BAH^>
z-?G&X-2@4ofaY|E&Gh1>t)ROur~@OIHJ4QfTAsuEg6oB9Xbdr(Kn4Goi@*Jnz-z*m
zmNtiBowLIH2QSUYjhtFF44FG+c{$;8ahsk><AvOY_1=j$<!9d|8(z1ZiNI40PX(^d
zN(i4(Ow%e(IAa|+{WMZ`b)gey#q$H__-?=O-2u|F<a<IOoQCp8;Mjrxwj(E+iBDvY
zT71BjtseW??2G-&6?OjDK$%OkncFWITmiVyJoii~N$9w3p<Zkj2jp-pAR1zT@sse#
zHhC5x9!|#DoT?8|0)AIjfMOzI$1_O<AV<=&?zj8I;p8M(-aIlx|FRp1fl#JuSQxtk
zY^*4y{F$281w$Z1?iL0>dWLXlkt+h=)T$sJCkTba?VE;m4<CKR51fPzgC%8Q44sc)
z@SF7ilGikKrTxVe+jn)_TAmJnuNZvRnQ(7iP?C&(*60{1=eIvXdr&vMI$Qa^NfrJC
z)$8RZ&kw?amcDuT-c+F|@sf?hkCEh6VOB?)>5XpL^9Ju0<C3aYcmL%I@_yUs%i6_l
z9%gv%esM|d<4VU7``Mkq(u!?-3N=R|b5=NYSrKGxBY7as5glsq7f>I&BJTP^K$YMC
zEUk)gFed1=7`($zQk$BX<~(AcC2K{4s?Xe$nRZJ#;FuLu&vUy}VH%D@yi=gS*QOJJ
z(Aq*YJqRZsV22fN9VC1@!?<E}!A>SHnW~&ll1n?fzdx3>tM0z5uBLENlDlBS7cbW%
zi$kX*#N$tVY(rUD87SYUc7+)V{!wz&IL`a+IX88`M1-o&pSH(D`rT;4^V6q4bxzcD
zYSdXxllYncLH{V91>-YrgU~picTn92q<E-zQSRgN@pN-!TJD^G2J&DX%3H(_q2pqS
zxt1_gP`0R9d{J@t?F<R=HY}3v!!s=A%1U)wcw&DC?wkWjK(Uj9EKMYu%NBIUgaA#c
zXq;7=+%B?l^3Ij_TALK^T7?dFcG?Nerbf6^Wj7vpxU-Z$EPtPD!=~hQ@}P4f6*7qM
zxb4pxKNa!gyEdP-Dj|@}X<@y|YlHa~eY$jsRqGJbKVgdzA{LWXX7Vh%WcZAU6=T7~
zX0ROk8zw0VB(%H`NU-sN$87Tl4v@4c^x<9@Ks!d17_f46gI@y5W=ArNUIN`DPabrP
zDOnA|20P-yV0;ZUHc)H-3(&!rR?aivfKpkdyf%mx2IihK#fX;C0iHq5_zF}5IS7ut
zk`oGnWOEEG`Pz~429A}<1N}DKI1?qUTMIwSqvx1dDJkwIt5zs+Iy2JD3Yh3)lptDk
zyIoTm<-B`g-V4$#sE+wpK!fPQ5$o_oL3w6UIM)Z$H28VoT$cf<t6sPG%k}b%fs3WG
zSUnRuVtRxipUwp3K#5z>!vzP+(0*(t8G9FSypIXYXjv;S{BJ6wMV}TZUvcJ^om6F<
zZ2&OGAWZ=~CJ)Ete(g+YFk5@%&LjWcxv%xE!i|68a-gqgpbewDa`sDyU}cn|(ab+;
zI#R!XGy@449$CD@a_EYI>fL9m`((u`AB^3InnXAxiB-mNzu^$jup(<-Kt_Gxm#<LB
zMj2@zS}XElRAv=ncf<l$Ff3kA`X}gsp)Cj?#FZgqRQE-RutVJ$c+cSa<tQ*Fz*$1s
z>|t;az9Cr60L;{!$X_n4@o(B!#HKp6Ie#~AXN8SBgy&!ZWe{C^ML<sxe!E0qMO@7r
zXo{#{j)=r4Wjhu)zDVeJb!VnR=j<p{-RyE-GsY<xx<ws&d#=-f<JIKZHm3qsNYt$y
z7#!>nq%t7@#pKTq-r(=I7|T=+!(qq)^$rlFXG$VE;Ia1Ri2|swPeQyXWZV#91p&3(
z3Ed)230O(31ym>l3JMtmP8YqCixcbBys>^}FCy)2?BwDf9XT_I;VIw8leVq*))ozY
z&1W_wMLy^FU2h{KD}EcDJ~Xhd5l3VFdrYNi{L2q2VMQM6cYiC3QvZq#GfjI9JTLbS
zzm~u-SMAL#6C-Y<olrq<+H~7Ay;v3Mdp91I<2mx$b=EMu2T4ie446Xov*Q2>kq*-X
z5Y_`o*tmcNor5PzNjzn7&0#Jxo1M)yM;-juw)D9_)NgoNj@Nl}jE_I<^1J<s`4Lxs
z1<Kio;9|F?@h;o&dQZiOH5){;@5=gqRQ$xgt97l~lIJOMh}oL##~kz$;bF!K=gncx
z1uJG;%Cv&D6~(YPNP)I|w!4JCD?J?u&v$@Hc?IA}zbT+c1SLfwBMsX72GbH|q<|Ra
z5P0fha|cRuB9`8m0^okG4bzNw?ZOETR|{@$+$xKafM1A<erqNpcj?-T>I_-O5(}x3
zu#^5lu6EX$eZ`i;-cNTuH;{b5cXZC*-~=*DGR-;%-PmE`Sk~fkb2LUc{8{6P=4|$y
zpovGoBtQShOs}c3wBjdU``!qITtxPII3?zMF1&i-2v&L#6X6$=m2@yP&1@60)o@is
zu7hZUK2-PSz2QCK14bXr4)^`4`!bKQLPxJ4X<BK^po34~72u>ui$UN4Mr`V#Q>q8;
z9%*eWC-yq9yd5&^7c7_%Nif!bH^_^RF+s6je9LL-NZBXxisF#EfYq`}DT_dL$Ha`%
zm~eC!;4L3IWZ;~0Ls260|2+u0UpS^UiEj=&Tf4OaI~LqiQLjhnWghdCn>YLk(&D?{
z&8)^w$SC+hHZgCM{3I=R-pnQBUQo<*?Ect$KC>f1jjJD#K}S1p&KXAOqH~ql4sS-*
z(--8?5hjT@U*3(nkbS3oE-slc^5CX>i^x%V(-$$Sl#?&b4t)4tKPb5%=gMewn{Vjc
z!-GRw<_>ZC_+6{=y^-k8z}#>~(V(iCPQwq`cUF0dyE+w=&)r@P8o_M7C~q;ggtpkW
zOADSJ_z+DiJIs7;q0MQ7=gpc<5gLcYaz-=R<_2#6+AK+XVHG*06zIg@wECGHPO0pF
zJs4<$0?8(FCTiHMKU!K?Y*CMgo*|w6@Y~G^eNsu!^a^ua?rf}S*W6I><Z8lL><#JQ
zxQ48gFVTn22QkwE{0?WwC6D|`W`=BJ==lUO*(w16${K?&*QHGPCyI5vuO~0d8SNC3
zCj{(aTi+<@dos2d79#@ceZ%$b*Vsi->5sp<5q>@>U$^+`&3u8qbP4$Es<%Fdz*Iw5
z2(&c0#)Z=*J_kw=4p$)C=7<oWt;Uwh_ECu`oc;qe(nZW6({z?Sx7gL>W{d^XTz1r_
zyEC8Y#mRl~GgP6%p<J_g#66leP7x$n|9n^EBq|(Ftpl5?Cou!ki<A^m8P0~6cB`NF
zw-Hhrqk9X#ybvpVny$p&&%5>?bIv*UUG~w`0{<r2+vn{gt}Uxx+}W&>CI;YtdTf8m
zQ&&FgzJ(&jgz8jKwg0=-&B{NukFcZUpX?WkdfjYm{9Z+KgD3YZ;lYY%&3}KSkB%Rv
z*k;rRW$WxHAM9S`3%V7YazQsHD?TJ4I?8NlI9#k%<n+{EE}Lp11vqvPtH6E&e=s@n
zU<)iD-Z_m%Mm=dc7*c16yYVamW(uhnVE|UfvT+ViH_(78GfEF0wI(Y7S`K|U{92xB
z3QgTpV%GKlh`J7Ns`vljkgTq~vk$U4M48v#os3lW$|w;jGRxk3MaVHqA}bjg2if6J
zIf|%MhfYX`?Edes`+J`M^W5h?w|k$<Ip6PRyx*_&e)%r9<X!X|P3Sjmi)Mj=fD1^;
zC?O#sP84gNy?g4z?z`#<DJF5YdfE5OB_e~;Oapnk*rm9UY1f}mJY2YH9S#MdQ0#)4
z*Xk~HbQ`c6#fcJqJatsa`@go|D`7;26wasF(O;)$6CMi<NM<eSoz@M}N{OzR<j#_4
z&}^u|WW1L1fW6SOebg*I2Yr-D=x}@Nlg02*m3IHXhlpV>;N_QH=A_c14zY|l%@<ea
z^~^%#j%XnxBa{)!q9*U)KB>|bGa4EKZznBqeT4x}cVr@(Wd^CV!GXaWeZ!7wU`s94
zO;g#^$i8G+_stdtdXn#T3*%s<O-xC;b-V;2e4U+S_d~T?PtBt|eyz)W*0mOh%475~
zx21NxG<nDhFLhb)f-z=UHlm3|z)AR7fKIzY?!P@M9*ft@8{K)M_q&q;+X-^Jl&GdL
zZn(6sw!B_LM*3(VlS%NFCR>!<(h50b%{QS`+UMH0(-k{V1o1_nThvm0?rKsVzYFX&
zv~**9$M6(OLD@L%aqJS>{Zsh_+B1TV0|7vx@&O55kwILXRsvXQL%~LZO9p)em7>P_
z32(f3(ocLGgP<vo5dH=&%W0ns;dtg&vAB4wBO42RAhH;h7Bko;4_xTETaUH;EnSR2
z29$(FkjH_qX`reRJgBL1%?<G=2FLz>D6QmhDGH|o0pE5F$sEmQFr;9t6qf1Xq{5(d
z;E3=<P8nqXWTJ?04Jm?KeShp-0>m&lY?{D9d;sM}ct~Wxm-GYT`V3IFh`bLvRTI%n
z5-viB!rI-4UJJVk1gs}TI#d<c%*hh~G;msvYXXM`{0I`AfE;LM$Ha^_K~MSJoJvq`
z6@CQplOxh&TmGB-NeDrW0Ff}r5?oDF5ztN>ng0MN3B0xJB6ObX)f415dXESloYYtY
zA=~>xV9$R*-lUo@u_0;scSC!`0VZs~$`?06v;8bBfp)!!%YfeX2PokfLllr72iQYe
z>~!Q&ncZ&E-mTnkP&R@ewi)m@Xl=oE=s%HbF<B;Nv_6JQR!=Q24|1=-Qi+=07#=J+
zQ<OP+K7B!?<({r{D@|@?R;wu^pBC|vKCsF$Wg(e^_ehQ%x_}skfCug@D3LVbw=()e
zd1hy10o;d>fa+odXzu|lN0==A!Jk+KO@`ck$J?P-wk%uW@rwirXmI6LiD2r%>ki72
zh0Hh4(zV~jPCl$=S-W#tVNZegNVBTw8z3`dTPS#sNUN)IsHc{yUOlJ7NX49!00|oQ
zTuy66&QJzuqS6qx9ULprG9rESLc|?Gye<P&&Vjs92;0^O_PrYzfP&x+Ofavyr4>i2
z(*WVWDU&Q;f$4}lmjodVnxZAPyiuPqJePxHZVqS@1O1DBibjFTCa%BvkL%i?Dupd$
z$PIXgKpJt*r$!{4YF7shEisV(Fffzw=t?f)?BftTGY+tImt-+0s}MEdAQbJ*|KYRn
z%{i&Sr3NcQUg__lE_+BYuu5|Hgh}`j7KB7502&ekG9>zRUY+xGN6xB7fjxDw<df?0
z1j#_!etVE+fRYdRaquc3p94{38rdg+{KEb7f3N(GT2Y*61T)OcgtrZ10$5m?A|~mu
z)L>b0s3*>YE(y3>)Nd>!LtNphx!oQS+!Hs7HhhVGn}NpzuIi{pqrP@}E~a={mZ3jF
z>5t#q@t?N#W*8@#)v0kb`T&i=mON&>I+1G)RhhE(IBfH=<0WU9&m++vnS%@C`2q2t
zRP`&D;EH>B`nzCk?H|`I7~>2g5Dh@OejpzcJ~eD=;~pJYNvfIC8Z->~nZSBk+uuW0
zPkI_~raIOR#}J}ZHB*P7(6UoL;P(m6`(L{zT(?ttxfs_x`QOheudvmL`=6gHihg`G
zQ-HzfX^R1Up|A(1MosNzw6qEf6Q}0wG9!B2Vsb`BQP!y|pJt*O_!FGdT9kUk&uH1M
zokkxaT}yjwX6|CPM$f+Wyyc~NR+&EyvN#Q(zyv*=+TeFB51_~9HC1`bfgZSFWk>I<
zZdeIgpOcKttDm4vC_LC|S>TE)B|+n%9E#mi-(E2=Z|99-03qW=l;}KsUF10M2g(|;
zT!3EF<x|um(kT9~xfGq3`DPidEL`iUNpjDwykUGUPl2bVb6oMk;<nr2%<TCJO_F&^
z`j#iXxHNAcTa8V@pALVY4W6RdXpS4&F<t2bZFgh9Y9@w74c6u^2~*XtVfsW?>MSFF
zt{Sa4e=ot#{l4aVfkSTm0AhrTcU3xy5wcO>>5j(n$3@AlL9;?f?8_7X@8uZTvFrG0
zqKZIvrXF)g|6}}u%U}v;inV%rF_%onGvSR;w9<5LUv?HOZz#PoG#OJ8ci~`10Kq5_
z6&0_2<tlLXp6=STjZNCrZfl^~$kc^VV#(W)&y|JVhH5<ZBE~1#OT{+Z)Ge9)qE6S_
zpglD=PZ|l2;IDp4We|O8HRe$RyCj{h4Q1wKCZpHr7oyiSbc`>|O7ZRTRXqyfJr<{W
zWj;hL{A|wW1w@!hEt3F8bR_U_h%f*V%D}tnA4Ha#bQxo%ExLo=)MTyE(Iv-a@E{FD
zCSqG+;n`p@YIR^!h_zddi+ymV5(HQp_bQP8A0)AGJ~3~;@!ogg5{p6UAARpE7t9e!
zq3jt}d6>uMi`Rrh$!$YU1&_Vn+K(P}t7t3q(Zi&(<;<L|mpJvWAK_8&Mn5(?DR*JP
zi%yGI{pp@4aP;iV>tj<&p`85`=l3pM+a)CLmxJGn1=fb<$qYl#`64Rp>f>o3Z>AFD
z_*yiAU1hfB<os0%aSvC>u@FTBItyDy-(MgtwNC<}6o5~V)PqQz5IvZPf4LYWDnQu;
ze<~EsfCjfHt8-8(T=u!P${zKA)ys?vpQ)Y*l(^40%<M#ICI-dKEEGL^!z8aKeIvXi
zS|#rf;pM8S>SD?*y87c!d2!VCJ$XJ{DqraX`z?$^n9ul8eB>=D=G|_5nesmD@!ZEn
zP6dY=>9DAtcQGww@4)carQzlNAgTk+GHe!muGKA?1yI96f(CQxE@Vy+SFfG}p%TV-
zjh=8*cmCqB_&fX`u7H}O-0t6Jw!3GyD$Tg|qeD3H6s}O-zsqV5>xGL>)>0GERuwU3
zY7BNW6vxf(4(q$lGCUVzRoG!n6>y`+#cT?lq}L||Uk;{Lw!h43@8A51i9KV2BYT5`
znswf&TL+&z-aoE-Y{sarK00^C)#0qhzlML9=qV;~L43zQIF)Ei{|OS>IM$$3;g>Y7
zEn<>!ncZ$$$xR=3SLAcQeRzQ$aT*K|04snSC4IDX{C{kTn+Y+WW{Iz586fX-I<{)`
zI>Pu{o(<9q0j}nipA+X*2^qTRP`IUGPXLG2*kQn*3{go8qt*#$MSus&#bwANgcu{o
zyM3i5z=9S#2VlkSR;+wiB09fah!3I0=+e^a+~Z<C{GK|X+|odJ93SVtU=r2Uc!81n
zF#QEb6)7&+fc_gykM7LWht8<=$Xz5DBoAWdei!tZBw<@53h%(}6L&7(LgW^I{=wUz
zLM5gqzaS*fq>vs^<bn+Ba_&sBCxvGXy%wKz=77HPhYAqC|C(HoDU?MLqIL;dXoeyK
zj7Go1JX|(SD?2QYwb?WZ^yc@u%(wU>4Ap4o%Rz-<zH||n6ASlYxfsUoEEr<ZchJ;b
zWFcKx2*B`7BLZ}2{Q>BIT`q=bd6B*#q+xZ@cozX&!oRx_r;S`KBh-zcVqC}vf(Imn
z@`MN}Pa_9mJvxup$Z;}w4*W}9Sak757Rv0i=~8YWSon=xCr0MvXapjHTng)i=VAyL
z=88~8jO@FR?E+9Lk#cnrF2ef@shI=_W)c+bKagy)Q-FfR#1A84H=xSJbgnKU_ft7B
z3cdnGQ4&~^A_vP=+g*NC2GvJ@syxf0mI@+>6Jo}2L_ptSpmoi^T;>}{fZJi?!6$+O
z>wvk@Zu@&kc8+@tXXNk-`DnS9v^)%D5;*kZGdzkZ4HqH_fgo&fqHq`m0pdP%vSD`b
zmR6uN5sxWGB>foF@5iYqTju^TzzNjTWaPd{1~W3x2mTlq$tso<?blFVg!(@ub8NL-
zSBKRDZs79WQ<$Oa;af5D+l=a}0FEN9lf)@3nbKs}WA^VMRz%%K^$t2kJ-uL$2tk=%
zxPBq)N6h$HFnrdJoc`D=@IenAB>XL0y8G`T2{SYJ=(fAZc)k9pMom;{`EMgRAqdmn
z1X7ySN+Ubz7O>rKE<)uBpC<^=%+?PM`%WF|(a@s8MUGtVIt;qF(*S%G_g-PU)xXLD
zzY?;0hRffjzX>|k%Aj}AAQ<%UC_ls<b3Y}V5`I|jr6H)zL8hus`F-EHqtyaA)hS%B
zfCPo=9Ds7>Q22YaUo2S{(D(=Bukec&H&W|s41@;N4L=pNLeX~>G1p>3(IDJAvn$E+
zxO_E~6O!k&JU1Xwiv~(84~WAmG{94c!ZiVkJ(*GO@NW`xC4g1_=*yGU301H(-}M2E
zjia0h0xgIx15%?ZV@)mgE>Lu@=7Wf_1DF1&fl&ndgTz!^xef!Odram;Ybro7j6r2;
zVWBsY#;O|xGr{$A!xZ^N%qGxI<<3X#J~2=F@Ark7KphT+bBJZR{IlthY`+VGyapf%
zgt~ck@+=akhWHLp&?E3kiJ6^DGZZZ$T1h~C2x1P~%)#X2b*Ad_xN?`SK;uVeGw*+e
zPfCrYio^7P2-Z#j0B;Zt3}Tso|MzNxROYr4{tz7^{eck`3p>ODg1fk(Q;2gLN%8@C
zA|$(DbB*!-WkA`j84|}1U<x$_2pD*jIbf<BqEXtV2VrFTQLS$#;B8QVSd%eK7&N5R
z;9?13#o1Q?3J!lMQ><!*AJ0_f6hTBHc(|YaUugVhBn!eFJqT68Bw+6FwHY~fxy-5f
zGr=&vaDH9%-q$ybgQo*@Rn^4DWq9Q8FGvQ<Pp5y^3KWevBNU~(>H-V2buI(>U6{H+
z-d2zrPNW!?{MUy3!cuL4APDeGq>|@?_Il*Ld>IfarauZ5fGWsEE=f0&#J`8~I_H6v
zwo_>3PcabEFM*>%LjKXm+?s39Zrl06sR6>=bJm`%)bqb|>8V37l}BP*HiQO`$ALQ}
zYU0*^n`GZ^!V`be^Wrf)BvkwnElL+kR*z2nqD4fa$52Gxd{I-p@;WZi!(;%1n1C!$
zL21P-qUS+*D|4@94UJo<?mg`<H9q8b8&zt}(bVYNHDaG;W@TsBl@&E#)|d7ms97}K
z&}@eK?fKrnhn71B+!4!eJotT?2hTF+B>7^gfGS4zVwsMP^+YZdEl|#-`vtR~52O}N
zrUl?g?<Ul+@bsB+6e5}M?O_$<eu@(Pf_wwxDgC=C3NjIeiI2_}X*Jqwd4Ng-Bnl|^
zIuh`nxD=@BDdDs%@<##|R{^Dm=Om9+*GYe`e6a!{I*=ciRsO$nev`XcNa#jXTAeC4
ze(*8oS9!%i@z6W_?)6iJhDi?7_y-JQ4U9aq4y2ohi@Xg!rhir%ygS)b`^BlSKz{Gm
zW#iX>4;8l7sZ8lsZ*>k_*9plUjS~VXodTE@3_v(%@dmN@up702k;xvh>9kfNzg)@H
z=lkPw9sg(+AZgSAXfRwr<m_VL2|2i24{ia3Qy4Aif<U9g{zz}uICv6d^qH_9WfB-7
z1aH}aamoYHrG(~ve}n>f%KowLqbWT&;e{Axx2)-xj(s*CA-E=KJH_ie(uKZ`POK23
zZ}>rZR-;}?xwL0QXJ}#8zA!bE`^FGcz^OfL-wI!j<E=bJ`hO289j+!+#H=xV)Vs{-
zKl}KX?wJqL?glr}sw=;}?PH`9C)jmwAZ%L|Le^-og<N1|uwyyEHxV`iYCR;Mu`jFq
zN_ZwCh;GHf4F;mJ$&-9~foSOBFIj>#6cny-TR}fYjuyLYe;<F$53Cqzn$(+@cDqLi
zFCyi~obMC`)8>%Q$~5qyT$xh3c{VD>^CGNNnW~$I*h5eKn}1@4p?Tg#d&>CL8tVLQ
zmU`|Rfj;lvX@?lu1f4#sD*(LBYv&ezbbZ15ORK`JW+*lLTyy}A*1PW&mVR${d7TJV
zC|tC^H1rI#VDn1@Jq)Of@zy$6dyA4fI31*@m?K2vRSghSWpDy=N?fALrE|~u@u&DB
zJo-tNR`A(89po(&9ZbmwEF-Jk@?f#fmrtSuG{o8|$4XJ^M}<;i)Q{KvD(x%iBNL?k
z`7E?uW6CHVDpFmaPQ1Qwb$>fN(Be*h0pIxtnauXMNNxpuxYI3{KNPR)!+-no9kIzl
zhpF{S<d1DEI;^u|_hUF8>j*WF!*jpQ0LlK}Z`{8EXBBo+Vj+mP?3ZelW{`|x?A-@+
zbiNRUnJI$lEM9RaC4#Ph2-eXcDgqpoAd*LP{(}oUz|?`rMI;^svklNcVo*SAiy4jF
ze;39Ip^Pb}O#EMKH)EDxC3&gb#J!ZN7@WBzxjd!Ws`Z^@bIB_G4o5|_twO#`In%Sc
z(WLN8=X*5-UcM|@E$7IOadF4&_k*QGn$K7KZItco$c3Rfv;B6p+5}rcx5-@ord|4$
zphcOSgDnn6Y=0!rS^2ezT%D`114NIY)eRpe=I9QKQl6s(I*jhV{9vIIv<?uMqjxS2
z5U1}DYq+hRitV6*Tvvm^@4y$L4kBb%XMw(giVZ}KVj$H*>u=_;Qbf;$e#^imbK$RH
zqj_g&or-Kqd@;URo;vw#=dZRa^1NWU?vc4NiD`~Lw!sjOeKe;_aN%mBTB6NTBs;-c
zk~;!%uDo05aJ%rGrJ64SUpyJ6c6d_8rp8p%LK5I|M{J2;z!8$jxU(q_lY<$rAqoj)
ziI^Y+ojE%ASVwS8cLB7#^pKo*e*r$9mZv&$9xNjNDANNVPtmhx5wG6Vk~qS)N8=62
zn%I_w+j!$(@~2OQ#8f5$FFgxoL4#gwoTu9OEWuDJHcQZ;F{vW_+uH>#4OjECdOK~Q
z8YU*~KQ7E)>WDfm_xK6(CB-|@Z^N#6cqhTZfF}iKK0f!~=G^qoDxVYyRHJ@~bULEi
z5V}SV<&=L4ABFjs-$B9KSqL8~s#%~u(7CxIMmTxs(-vB7-T5Om=F}G7hERIx+u!My
zgq<0|4OB%X1p(Hl(!^|@X#3m8sZO+P)fU!mCwOIRv2VHAM>$zppi&FNl&UhOgNhij
zUBM|qU+w4qDQeTxzx_P&2$(k$pvFN$H3k*>AWpc#0C|>za)S{yGzc9rnu*O_UD3N$
z+-s#6lCwa(`W<IsXx3hJ!9&+iXZWLgT0!cYw-#=acPxkaBTQc>O>g}@)B|7>8V#Zc
zaMV=8MY86EtRMl6IOKki!I>e&f>76avU<MzamY|C6ack;Mizp1=?Ea~yAgohVqQU%
zpG1Xm?bY1hmIb7LD$$@7OUbSBy;C_UC)5Ump#+Zc@_76hJP|aT(y;(7jK*n&QNTB1
zC?4q+l=-&iT69ZLa1#vI;qghquj5&WEfwfrthl9j$F0VRrjs)KcuHf!p0o7R-LxlE
z#tV%zU;vDP3zB`98V(mwiZc`|?B8A_yGHGD=&sHpRVrxm00>2VDDRHJ%`r51?twLB
z-UzwB_8(6ie@ai((^W^Fm(c?^xolE!PqVr%MfQe&YN65;3}Hd!P6x@h$fMX-M5=!y
zhX|yKqu@<?mVRmMcm=?Lh?lTo#b(Paz34p%pAh}Nd!afR7qNSsh$sVKr#VrXdIAex
z?q)FXYFD7_jjQ6)Yh(Sw-i&FFiQsrt?P*I@%Im4_k$(@JAp9BEMZ9C!18SQNiRO&j
zH_ECAMEsJOgODGS)h<7nax0Xkd5=MjW@res1TX2qB?u5~m=s`v{)NclYX1<BQUy)@
zx|3;|ktf;V;&(-u6y~R}dcZ6A!G8M)_-_-TYS`ueJ#moz3K6_ef$iS;p9?9H@{D@I
zrWzwjPa=JJ@OL?dEytJ7$m$C_0b2K4M8N+4Ou~=7;*pqnB=xW=Z}fP?^98unWuvy>
z_Rd|E7rR^dR*loh8=ZwQ5(Qz_Gt*6(y^noQ;cf<q43;QV7>m#Wp+U);1PVpIp1J%j
zadC(s6c?91fHl|cJVAc=<Fb7m;?~R1R8}8rIGKIb0tfKJCoGK`*^naa`tD5MReBwk
zk+g9at_0Pb9(l&J+i5m@P9$NcL_a#W>Xl>HNs8Z&O?4!kAFlmP;;O#Q?OtW3aQftr
zjmBms|GMFLMY({4Cv%as-0^Qhl78_wKK)AVdsXnQZiVr;OL8RV_uiXAX$1?xRGB6J
zTu;mQaN>B27X8CiT_T|P<vU)oYCO6BE-I*JIV!ayeyS8-pSZbg-)hGh=0hgRL}W8o
z!cIl*5D<}O5;g=yVrUYYWg;MaYxBLOrfNn|_CN^{!#s3{X1r^SS$jM&LhxE7B|W?u
z$VR>53!xY~OR_FTdiMcy;R0{1L{TCNnnmOSCC>rk4Jb4~!LVh5oQ`d|{X5xZZ1?(J
zv$>b+3|&Y&v2xB2Tfp^YNogd@O`LSy&ALuzeQJ+>a!R9iAn8Q6jsK!lar5jnXJvys
ziLa{Ec_Xlg-9swXGb0DZeVf!reVoer*7f)DBsNS~ztf7ox@PoVqr1Jfm`aYR)p(_l
zu-_RWr>&?Arl0r*Qmqmv??2y6muodm`W3vzSi7x$ipezjyGfIR=jo(-V09D!9!d(f
zYA;~)VJ1Z_F+5p*ot&y2dKxG2dlEP4e`Yr2YkUVyd&Od{t0Ma-T3inUL8)n)3NV3Z
z5qw_HB9;!s#LnPSWCaI5=tGm4)1k6EXYIq>4fze$$3CZ4l0YB|?Q7&oxK+R?0-cf$
zq>N-3f_4|g$O5{o;}0VQ!%m3#mx~iwo<*YEg6j2PkAW|9pV#@k#l1SKKE0kB?>R1!
zsb_~iWgOv^eB_TJ?3u~f4YidGgv7j)AQ%o5PDoDe$1X0_T7BWjn=p9Glpvivx##cI
zpZTYMKO=r8^>@;qo#?iLZQjJyPk49t(!IhHGnYzMF1?@n^Bgn2Da2<c?nxS9BJ>dP
z+oB{|c9Q!~&Xk?e#Lr!S55>Do`hC!ED(VjYB-Eu<9w9o;Q`|I5?>{r*b6e2e&piDs
zljF>sZ^OkSTN>l;K!HRsr}IV<^d%1hL6L21q0tYtYDm!tlC*SS=?A{i&k&)zzF^^0
zKaB%{8Wd_cqF$;+7sSMlfdU9}PzU&v;lD;gkZzd7FQ67epsO|`L7~dHz8PQ)3GuG5
zw-+-|wKBZKVAEK!64&<`g|w2ZVw8t^O+%OSGI_r+Ez?hmU*=98Q@H;;w}nvF`%Kxr
z<z|i8N@?Q@LDh(UyX^oB`K?Ut1zM*U=6l?+HxJJQU0f+%f{?TKC-?|^M<yH!ty81a
z{#_PlW-D$DONn1r#gKZ{#vZxD(I~X{eKRI3eMfu$r=nVc&_;g*U>^ojdr^`N9o<%{
zOR+m~o5?G0s^W)_TwW9BnUEbjdcJ9(1G7$)PC%qvw;<oS(207NXIB{(2PJ)7h?;)g
z&ngOA!CSBfKt=o%)KX^=z+w%WBvnqt9ct<cUUCGhLaSr01I+h0nSC=)o01Aq^Gcu{
zhUNO}dEbY8Cs=s*`cST>bN)QaeUy8W{`^)j{)S>h`?$vBg^r)W1btD=x}uB`=<EfK
zK#X?7TP&v5$ZOA3;@t+CYam$fr`qrJw5jE(jYzxJyvxikPAj_mzsZezzUVH)Tf&PV
zTD+V3(M|m>+2_5+RK-+-+Hro{VvSd=ro3`0WSR6;Wpx$$M^4ndQIZC;>6AZ?7~th8
zWHR>>(0Zr=G8o{HOpJ-#U!MN0=DTo1fWmzQDjgvqI5}=fCpZ)_e}~ITjY6dH0A@#|
zs`#@Wtse7aRsBeam;ldk234mBL-7f)%;>tAEK*_m?o(WlWRK8U$vItO=*Om*s_*!q
zSma9`eFGD9bXU%!L4}L%gu%r3!Dx>dz9-UNT5p6LJ7=UX`d&9vR>&TRDAF<_Ive|X
z76-{$jb(x;VO#W%+yM0@f_2c&Kei@eXX})o@R%?63!7Z&tVy1wGN@sUovv!E3~>85
zhw9jqc7?r2GPz5}a%CY&mF>y~@9R!Ks&+LE%l`MM&O*h|yocGONw($USHQvPZbv@0
zao9icq>U|ubSnLNm9Oi$#p0J}5eF@Kk?F1QDP4TX6d+!jJOPk@184P+gnkpS@gRh>
z0vl}&9V@#oP@xCDC1@dhHx)qH43``>*JfAsqi(NofgQP6G5NwVNu4_iT6!W!(XZ5E
z78xc&xX|Y_tR<F(83l`lLKy3E&Df}HUS^jK^oj&9m)doYsoavIWItxwmTF7tjawC1
zxy@Ebq&xM?>3Eiq)}_f49O_$vXHJxwoKQX`5$xwS6%yyiCe#)C>36B%!lm54gz@>`
zf}@hD7uvTK_fit;Xbnjv0d5NQarHCuXDaK2g3~mbFU!3kxw4>O`0~Qb{=lQ8+3%ZW
zYHb<1EM8|QE8@rP`oH+5Buqcrq@5ehSw_n?kxw(E%|!DK3~x1w=}bdCk>7o5;w*@w
zQOUtbgffu^A?Fd$8G?f4fD4Bx$>B}e;$o;Quvs1S)i=w4-BWK}G)j1mMANSV;sU*U
zA*T>T0!n!rPeS%OAdqZu<bEE=a8__J-3Hs*fC*DGCpE5=S%B&WuO`Hz3d^@pKNB=j
zDY$pV&Wl*i+-})wC!JLOZWmMD-|<X@k=+QDOsy7S==B0mQd*8`Np8Gz-d2~Rp6p^>
z8kD0a{xo&7dQ$6Z*)MFi=m+iY%U`j7;vy!msE<x_TI2`5&RxC0A`Yk+1Rj7v|4W5u
z$Sc2K=D`J#wv&ADrthndffphH)P(@bAb9O287^F)UB(8f&J{m;3UViytAH?=?%0sG
z{^0XV24%SmY!Og-202mxo&$>uoSn-6#>(U{IApnaEj#P@2lBVeHPcvVv-HS5H61Uw
z!%au&9>tH^!atrQ`7j->QKez84K%f|id|p7`XFX{n>xOK;!V;W-si2HWd#!q<)=AI
zzQ}3ciMU!2fGd3Dsv<yAx!v06tUY|i?L{Q|$}C4_V<^?*>$s*bOh!qUwJom4))ome
zXL31xJF7&6)to<9@P+gAy_*ZpKIxt^UzsMZPTs7&FFa6e)cwsPtx;%3G|X<rrDZ$?
zBXD+c{ghrr?A>+sR?hJqH`P4Z1D6jX#{v+_u@&MauG}yq%MVh&y#)f=J0P323rhlp
zn@E@u@a{jTF!SxHes@r-jQ)1u%-y;pBmp+L)u`RTK^~TgWm%DgY%0z6;}ah?PSfs?
zDfjxmeI7k}6Ti>Hym#w8WQEq`+uP?|Yvru{LnKjs?MT)sqO)75v1Bx>6Icl1w%_Le
zO*d#p8gU7(OY{uXPVD+6@JER|gHJz3PBzOWo&LhlNP(~ET{#=wERD<FM4j1Ozk8=d
zp5e_pqL1vX=ozVZ7g9N8iOKC|O-Yo#A8rsrZE@F`)1rs-iCX;4s`Y{!MR%{Xc_gvT
zcbMXBiT~yvJ+i5Rh3Fp0R6sg^R@OlNlzX<US+KYna1o0p#E65-!3Dbsa#0-+VI(j>
z3F2>_yNLuB96#to*7?B$x^_7+l#<sw)}vACMRN;%=<!qH$(m_>VzL4S%nht9de=`J
zts)3k6sjGXB@3Lo%9QOPyK}9Mw8wKPd6IcB<|U0V+pMGcvL2yymmyu`4pZj59j`pA
z^29J_&0AY~C!2pi48KJGNZ;<;NiUo)*3#dK5oQ;hPyMsF-={296d%@YE&RdYPM-R&
zY3|6cq`KWid^f{~;I@RBnUlivTHkBqso4lFvyY4tZ9EoUYW-PF{e#D<CN6xOG=2;v
zSPc#DkxiJpQ8a?=f53qr;n&NM{teN@KzH@3e6g4X%oE#)aGuW%wmLM+w<dCR#WFQ=
z4}^k(vSwD$Pa7OuxBIlu*W^)n=Q(lB(+UbU2Ei=CMhQ>j*6FhM?l&==*4xO`d*wB9
z*tB4<*vXiY^|r`&Wm2#PrsRZL<kbgp+IAXy>nlwhCPIvD_3BgUOB~~d(XVIx%$Fti
zaShj;YG0K31ZF(DxlVW35%p$`rE&4^A%^v_l15>jz*iB5)b)6c2A!IewSYP&R7BO?
zK|hXwpE}~NU$SVVGd)OjCe-Vty8T-EV=k?-9!aXz2(ii0Rg9}Lw(9DzO|{fkR-@NO
zxw<2_1U!j@*sO3-U^4!*z}e>EQ|JL{0#Z?Qno0)~!E9g(;_(Q7ynR|RFCEA38F&CN
zoWQuUh%QspwrBy;r6CUE?;*MSO^QB2Y;rHR20zOt(KidIrbai#;eKPsm6qi-9uKAw
zPInb+1kI-otaqI5Ncj3jT_aSW#$a|J&-K`{=!J>!re;<j(`c`>o4zhXat@KV<YK}G
zx2;`^y&tCx$h!(TCNw=QZBLF?m<wJ1h;NS0G7ZyzLLxL)*$l>8ohw6S2(z!LMYr{^
zeSIZHQ^2SunxnW*<@!OWk1O!r^BQ(1`L*aZ=>|csYIa@L?0uaHX3#-*C?H3`V!LcH
z1wb^^tD)(By3iuWoRNX_ADK_x%$3wIO9bJxLXQA7x9_8ICV%|Fsr&#+)#0E^j}5P{
z2j$Q9PhR2(;Ype8t#SOuxYXBo+S~gg2Pf~y9^-Hf&5I-!$G?Z1?dd#eOqdvbuqGMP
zX@$IA3{S71&H8KaTex1QOT9HrNA=Oz3mu!mQJ<t`BPST=!!+a?BXiASUE%#Lx!d}_
z+=-peYktmWA4NFZs98B&D}OWP{k*BCo79=PXC{Ve-r8|Z?;`ZsJ!NC`&z4VB2;e_w
zH(Z@i*P<J}DcH#cTJlAp%of4MiE@Qd0fypuV8+O<qn+;4dzsU4|5d;{rP76gOIgp%
zuVt`8@=w&h6@Vm3srtb40AezfM0nMD+TYV#R<|pjn;Ml7WyhXmP^e|5{<)|3l(REA
z_JNo)2LW|%?*hRsfjU~=!FO{<&T-ZIA*uV9&v9iRy1Pak>G4x+3FBHUX&>2AE-y^8
zQ|XH-y^R-sUqV^%JpS0sh=%gzf|~)7Wdzdze|Z@v2?03bXS#;|xibA@OSe@&%^X)v
zb^pY(m-DYnyidlfd3jZloNJ<d^&VZ}7A7Q)jqqKV0v#)Ww5_3xt9sg`9l6mYU$8(6
z0rxWzvI$V3ypSsA>cQlP#PpQQFx3J8YiKyu2EV35khjZMq&bD*IWUAM!%)O<w^~Ne
zU;x}adZ){G34M>Xug<@wm7_O}Qcy%m21g1DM_zP({P34qs=sATAMsVv*~@I>wVL{g
zPoF0ruevEBp+gF_++edhabLwFkK7t?s*pH)HEUC8Lb8px_WZT;I|(Ni!X++D*xV3y
zwD6r?;#+xr4cg#2T@eS-ruaA8l_3XjAWM^7S;iSE650}k;EZcl?|^d)%!?9eJ*zmE
zPsOPaPL$z4B?wF~3BH?=#AvFH?)-jZ)V>IU02&NIrb7fciQ!Do==-H5p|gIhLGOBh
z$p@R3S|ehVY;KS!`#?fn6_X16$>n3B>3JOkvt&`hGdf<{{hXoCJ&O7|sy!B4>YU!S
z;m@?6VAIX5W1N4%A(v{kLvJ_wc#Qj_cKB0kj#>k1qp?uVS`|O~<>W=HZSI65GNox`
ztK1E{?3AASR{cdWC#Omj`Yw^x_fEVruI<XGds51+&yJ11a6H2+{P*JZ<9A&L#4Rpx
zPSJ0+r|l$}(Ysf0H293Pai(NB-ifv``!md-RfuOD3D%Rzys1aiarDCKm=>^Wx(kzm
z#8b~eWP`>i^CzZuBQ#QwbqDZ~i!iO_j`|>^pt($^FGSojhg_a4L}mt3L+F#PUH&r}
z<fNJhMG+s105weY5%Wjf2M278@z$Gi8B6A`ocHHL%5;iUXsKdT*s2G6V!6=gdfd!r
z_FR+*<hfJXf0SaJ6t?)PSE<80^~tQ0URLPOWHm|?^4(4S?W-Z3+AoG#_#cV*xCTTu
z{?=Ah36?Fy3wXcE;@WHH{x&4^?l#3mCPPPuhdPg3C}+>s*Ity?spsFY&J-cV`^hy4
z?PP@isR>?^=Pf<S6Su4)^Eof;eL-sT>mv;Tv~gp;PwpSP-7ZmEq=VAYwz+@xj}~X~
zjZ9xWM~CT35CmNu)77<Q?#mw<S<@tufz*iJ`u{!&jB9uz<O2aFRhhvMbzq1E0WYZH
zMTh;jk%?%*3{kR3Ak8jI`KG5493NMB--E|c4sW~#*Cr(WBC<g(pi3Rc8aXVix>@k7
zIosz<UMr@^|I_{6?kLZB?puB5zFe2Lx{vy2eusNM_FC7xj$7b_Pw*6K!!g+%<Jfi5
zll=XMMzC)4BuP}Xc8Z{=vr51Z->m0du|B)s!&O-!Nb5rWSn&6d=wdO6*u!(zdHRR$
zvoF>kF5Z?ZYONEE46>CGDxTn7vEvWlkl6fSu=AA5`*Wh@x;Fj20FTz-S+vx-zlU^C
zqB=g!#nOMyF>lf!=9eOjGxZdhyM<2j*;&GXwRh#AXJi4ZzY>a^g^rMzbOXW(94MjN
zl;89(pCp1-;3sg{op+H#0%1~28H&4)S4-!_Tlh`X|4Y|M*ViT_OOlh=nF=IuJk1%I
zVJcx^vEeIosdCyn#izv;C~X({>No!Vs!uT9JG^<1gyN~J#axP-_Mahpb0)p;2>N?S
zk9<XDL(RS$k57GOY&sS+d#|lOg1TsP>4QCaZ2?RY+uLfIP1fQb&X{FXN-y>6;pd8a
z24KZ&%e&86e^!##WT8z_ws5;S-6D{olqUhL$fbAFLHc{UtV02nHV>?CW5_AwPca0r
z@(o!Hk?|k&hR92u3PS6hOF(8JM5H3PtZD-Z+}O4<xF{s$g%JUX6F0dKZSQ^fv&>Mr
z4)Zfx)?bf1!t4qt6Y1>f@QLc*?=l+r<a`@1h<f~LwU57tV9Fs|VI%jRu;<iA{1rJH
zT*#QP!o~m7-(K+gcb$e-jb=9CMbV8=q5AblMD8CmE(=1}Z@ZcPsR_ILdt}`suWxgf
z&N!1~r_}z#`D5>@06R|chi>KA_yoh1%;YP74+UU7BlS`8>l*Rej{}Y`(LEEHDn)C0
zS8o?2B;P(-8(n>J!JtX8D0<B&_+5IP@l-@Ea@9rhD3mb-5zoJ*EdgqF$76j8G8Uo6
zeN{yFLp5_49>y+-LtI?F@j!981#t&16}f9-aEBuA9YFs~(B|7iDiIN(wxRio#Y5y0
z)zJ4<7DCb^U&fE0%^1EFm5QBQFDtIiGd4`%u<9+lX&~Lqe$hi3I~5TnHA1JX6-*r}
z=tWSny+UuE+%ugh@3L~jqxYFhsFIw3K_RVW((t|b;DT@A58tMm=%W%_I4}YC7;cMr
z3K$V;{e0R-TJo<jOHWWU&zs7%KQ5dp(koq-x+thi2=|$Q&`|9cp$V;#Snagx$tZbS
zyIyCOy`K8^5pSG?VH_%#^!)X!s_WUE1ACay=LzaM8nQm;K82s3>DcR?z$j9u75RI*
zv3lhc9h$k7d@8-1QPGU;q{@e{e-D|;UXuRb)WTLVV<1!KpN`fsUJFkF;6JxVJ#G_#
zC+dYl$cF;**T~AseXSBiNoi36jMSU!44SCLn*XugaSM^h@nrDN*I=$AnsKq9V`5=m
zgQz*1at6cday;40)ZWq=m3XnqpjmK6OGn^D?!{{x2KRz>yRr%!BWts?F1Ju?S>I-A
zQQ|8N9{p@4+!Y_%C+O%>T=z)wHdooe7)jY92mj(j*#t-CoZcW+u4CFzR%TntlK>W5
ztu-e(cC$LSl9ea{W;Dm)hMp&{vux;mS_aKNc?;E3B>dX7ykUGn=Y^#1%-Ug7Hj}*k
zw(O}=><iwv<+B&l+X_c|gP2Z5`zmju3eFAgcK*cDYDRz14f?)T(0-$sjSjv(FyBbr
zm<C1265}v`O!09dStcfscHZGAIBs?{I-Wo|HU?zbrqHVX5Y#7vU-Q<GfDAZWh~@7f
z6z<?;4bz2bPes@UP(OempHU$x37J{~8V4YAbJup*Dp@l*%jAz$^tru@=_hbrmlGA#
zZ1vG%{T|8vKAG2qZ+zIUcPc`FRwgCdjatUi`{WF}w{rSR+frS4a!Twj%`?m@)Msyb
zwpqUYU3$|fbHhC8vbA*8FSUy{yMvo2B-_XzChLr=nYY5nwWvKc1hLrN%W#`U?OG=Q
zUuQ-}3o#FaO9(L|a7~b1Q~$GACw5gu;vfZ91XI5Slu+SKfFL1B15(~i{2aI6d8UHM
z%ij=;w}I>_aa;Twf`XI8DAxmlha=EX(2nJY**eoj_0;0M)e4sNAA7ltnTWP*Q)1e!
z74s|jG~0;hKA({dOOFhG`{=_%qXK`C`0AAD_(DmUo4lut>{aNDZJd>EEGt<4kt>`N
za@~u&*Rg$FMs4_XU%{6?J>vj{kAVV@oG!dDxcaV#@Qw9dr{n7gU%|3>)^D3~W4_gJ
zid&v>GxnD6@$>Se-IA-Sjg-;#Tp?*^*WH=^qE}U;SF<lBV>c4(5|OiWIs%t_b3Aq}
zV8J?g)a1Jq*Qeqm^2>F#8&5I0dz7aVH#PJEMeo@XHG!1H8NUq!`G$Z7GjSX3j)S8o
z3Y-H-*6G(`X9G~+cv5MfEPAEj0IcTu4*DZ9_d@7?SCWGGe24>M$FZXKRH^n=CYURi
z5Elt}7XZ-EB)TMnpc~}Q6rkZccy@ze=BW%nj%M$gq{*<Ktm_c=aZf6BVISZr6?!Ik
zem;?Rj6XkEO@U&{*K{W;n2u-u7JBg3+~Ttx|C6O|DBqf|a)th?0o~{YwUaBizsT}@
z;3~b+jK8B4n)-%O0k5;k%ji+WF#Pw>in4r?<y(o^qxWw}*`-eWsK|ZHEu<g9ov=F6
zeoT<jMvni!;IxqM^yIoi)n}dxKPMvI<H5s=gs|LSr@VVJmxV8P;1X5^%6VU@q-BSY
zrADS#r;WlM=6uL?jCsMc??4{U?J=GoeeS0Tu^lX-R9j$Vj^Dbu1|b9pi;Q-!A<qMZ
zktqU^ex$B`Wi^HW{&nVxfmRw=+k@C~^Aa8p8ig`E?2my9g{ZlJM2Pa^tYCL;vb1yl
z7hcxghxw_h8-H8dLor|AX9f@Nx7RBE-uhdd94#uxH=Gg^E_n5Ev9%vx`Ul+NrwL*4
zJ)@l@_s4j-9D<{9*`5Wf=bT>GD&)@!%^B`27+=-8L+?}&r*K<@=a1xYA=)9RS0;;p
zlMq(t7v^sx@yNSAHtSdEP=08<zm4PbHuR5~bbaTGWL>Yp7Lxug0})AiGWl4Jfr@cD
z-%p16)v!<7;k_kps29n-AIhzRGh#zuz5PA@Lge*Qe21R$9|^;sC+UTKC@pTT{XJ9@
z{maf-uD^#0^SvGkcp_|s9$|?6@DLif2rOhnW3-~_frryPNyOIB0{Lq0DH)tu0vRhv
zo@M9jiG2p=dj%CM^<N0W68ssFckWml5nGz&e(X%$comvta7hX4Wrrr`>@Cf`(7qgF
z73WB26HE@vn_7CCluJrt;G1-Z0)?dycj9_>zBJ~ZZ-&ThF*X`AGuCN--SwQ*!)MEV
zT3_2rT&s$jdgiRW^7IeIy!lx0gnOa3i5*ko^YW^FH}2^yqYRA(L6dWqFi+ALzi`Kf
zJ|HNp_?Oj*5&d)@s!LT2eeWsvGm1tZq*%LM;_x|(b7<zdp7C+`Yj{WWDoAw7b$FxQ
zf*KHThxvPP-^6;Vtv|)rk|Qi@f(xM|1vown!h^Ui_XhrC>MoqWmBZ#sL^5l*;85T-
zVmQ!C5@EHTZ3s(v?nuaAmU#Ub$Ghd``ZEGwyLVVY^mvkNJT1nSYQg^ded<RI8gFcr
zi$i6LZ=^qo-r`6WChqyJrka)v#6DL-x%)}=dxw4In-oY9$h=8uuq>0AC!w&U|8D;8
zAx}4wbL!x~Zbk3flXGppk7l3o^s&|JxjEB&OH3@+({-QRc1~f>&5M2#Pv>tLd^ng3
zs|$61Eu90YEAjDQ;|qzYqp5ix$~hWwJ)JA1PgQq2m(lF^9E9FHNY|p^kef_-f*b`w
zpVTF)cf}I`=K$L`wEnJ)*`O}FwE!KCniw*DSph0}1Q>vku?)rkge!tmsa&kMAE1x#
z>rd~;3^aP0kH%%lofIM{7;<D8-jGGl{Gm#0aQ1rKChw@XSfSn{cZvJqS_HeC=I~Q>
zsXn_1?fBGNGaILF==3$1X4t7d7|4fOB+^J$IMj`sY@<&eaw<Go{GD_AbGRqH3qCxQ
zOVy?Cbu&qrmy|Z}PCEO6$&)iCL-*;#o8p`QESRq-Ng2g$dKHX`gr#%H5SorW5kX6N
z=xgb!Xo?h$L)H7QvC2O`ea;OA|JbIyHoh=wl9uF@@4h`=QuPV_4p-tFH@%k3KK+Rk
zxW<K$^Z@rdU2rpxY0-!tBtkzgw-$qQM85*;FPVdMVaIeB7dkkhKaFC2cyHk#Vt`Lq
zX5JD~waMa}Xvi>l1(FhIGvyymZ(uZprE{M!+j^QABOV7RkQq6);P!^|2rxn@CO>_<
z9h{6~(()7!Bu*cF&}}E{2D;PRlCK0p=u@JkT-&0E$v;isIz-iByYW83aYfr@IlVUW
zPmG<J1RJm_hoVJ#4bvr#;>y~5_g>3P8NK80-<8)b4Zxb&cXu1!X5`j?$XItw&`ZmZ
z+N)7>C&57eLgeXN>C}{B%hvo&KQ>?GEfuI;(`<6~*BKWO#9!m6s(syv-!Ktm$(fTL
zYZy1o9*z3GBYQ?eD|$YZZY^}tA4vfeJf%CQm*2>K?{ZPI76-&{LZo-nTH}{YQjnZx
zIx!R?sNr0JBuhvpHLj${d^BFNq<t2#OM?;c_f?3K#E~1zcMbFq4xH~Fb-J`D2FMS@
z^!z8+k}H?_tT8C}$f<#wyZs+G*IFfBZ9BJ=T^hBVF9JUan9zT;So&Ck8u;8KTT3tS
zax*;1#-X{N`UU{P_^h*_j#SnBGOFl^G~HG3G^wWbAN$1K^)fwuBeYQ*w!H}?1m-h4
z)Kho$bD94zvNL?U%A4<w(&N9kAV3xz-v-vBC=P=w<O4zAflm?2c`ZhmaZ3d=h5gHS
zpWYXV!=P*-%J>2<ga!H>VxfPlee(d6{84&zA>9y(xg585&p7zqw~tL*oLt@Bfwo>)
zlEQ_Q*Dp+Cak{@~jpI$)V>-Trxaazg60J=&iVVS;raD@DreF_^+7t@`sWb?ZSA(vC
zL{=Cik-dHhetNO*QCf*=krv&JAOkHw(VWn@_<1M><se-ww+2zN8&5_s;~=P?KVn}Q
zB(?KN$b8g;P(SdUTY+5*va_It+dL1wD+~faaG&r)QrFmBQgS_qYy0hk<ZnuLkm&qY
zc`7e7&mFG|RgmfyUzxG7X^Av#Y_srQ315BP8I;QO!kK0v^XSCn@<rd3@_E7ecDcdh
zRT7azKzg-@O9_rs4>#n;M%uwC92e)iI<b4c@D8MdnKz)Wu)ypT1#zT#-1GoMO_NbQ
z1ld*4Nt>A=F6#rq9wU{D%_Nnn@2b1uFW58tG=R1(TwKE(OR7b+Q+cJLs$pE1*C?g!
zAGT{vJ<5L!v<@#P*)p@`l{~l74o;=g^z>wt+2?#gO3+l!hQZj-jNKZ-WEcSL9gJ20
z_+Gv*4lK%XIv5^?{Any%k*ui;1{h9emMYkk;2J)tFT>!aUxQHJUr01Nl4N3D6oV`s
z)bC0PxpE{S7Sdq;w+Nn&rbUT<nn=@oRN?@tDMTq6*TRpx2oo=l!Pm-Nv()sjeSb6u
z&tON}HBeTetJaqIl92YZFMdQMwU+vhh$7CTg-7ek?16?Xm}^1iC{OHos<nAGOS?jA
z=sLW8tAvYq6VO4oni~3>GH9x%M)J&esv;A&Afr^xK`&4qM*)`M_;D`jsF3uu!XpqP
z&(gXuTI{+80|;mm=K%>sKuioufPD;dtswTa=rlhP)e9nYz;|G|!DfLfDT4u2<|^T&
zv$vVy-`K|uzKPSOgmH4PTx4+YqR8$w%tk(VDzZdm7mRjrIKs%hfHX3&`ZqzNmkK`W
z2IyX*1dyCv^F{bSz~Kc$B#>gv6&4-_!r?*H0|_FaI0PRN6dBJV+9B!&8ScyR_t1ps
z&I$N|NU=ddL?+pJ8du!HMdV^(60FxFf!gUXF7y49Q(UlEOg)1{O?Ba{WkjOlcgw(|
z4yj}Rl3^Sv@Ddw{Qx$mu{|jwWO~3hHJdKS0+%L}p!4#ELj?K42Fh+-UqQ;%Jfg?9L
znN=xAUMjY?uT*5IByHKbQAa;L<6>>j^S3R{g%|uUz+jx8JG=|9dqQr+z%y~|*9A&I
z0d7&WQ0W8%HuX9S3=g4xx>@U3E@N*(cEo`?Dni{ra3P;X|Ja7^Jh(-)KyXnFrXv2D
zFd8tzZNh6QGJsb^k7#rK^<cBF=~hB|4?B$OL{cDickA7+q(=Rldb4(f@5NE`4+WPh
z7ut9QUUzig@{2dC%Is4$OGsx+c206qniirAqd7@W%V<%zExK~8Rn648`ws8tky~lT
zx5<q)W04R39;&114~}3R8LFg^^5*^>XrS1m?UQ;0<Cc2&=If8y^OJq~f_~S_`Y98?
z`Zr0aznU{~jjzsorai+i#gOYe?<(?LhhlA)SL6c4;o*{pT9zwo{zBf{UR8wjlL4n*
z(AqRSqU}7Y{!2#DnM0?uw(9SpH`bQ>cAf6wy~Qj-6WII2)_xp#ZA`CPca*QfOwnh3
zFy}0BDue}GD(wb{AwC1m+shD?4daZFO9URzjygFGh&DbQS8xqMKjYC#1wpis5NWW5
z2e3_am&yIk!1vmo>3;^Uq*V#Nztc~-Qc~|huDZ#0e)^fr!$&(*l6A>5UOquY4IPJN
z0N(yo3B9itcu9A%SbI8aB`5jGcU)Tl-(m&&)!DjjHB@NJW%A+DwurNwB?F<!saFK@
zc=|%i)pI}39hUl_MJ*VTbZ(;s6(XpH%@l4XF*|OZ#<IlYP$uc0O)_%0OMH#5h!gB-
zGfF017)sWH&AxB(WHb3DWk~k~q)({im95uN_M%&j|G|m<<as7~IQ2qivr_H*BG%A;
z?~l)hUyxRisdzCEV3lH&&<G~)2xN^678Q3Cl*{+UX37dZG+)=IgjE;*4`kMHOk}lc
zc@$ng=u@KJ6e?MR`QO<O@6X&<aH=%{Nj};?2YWZ7LF86=^z<D+^&?%(mRH>un7S+i
zECQ`cnyO2lA7xtpl-SPqYF2RQfjk-(NiIRDoo(<vdz241Uj<Ce(}8EE@`~B>wDdx0
zIjMg?nREM_<HnWjxPTqHJuP!Rsd{eWpJmR%*yAnyFH{CS%$GkcI*ko7^0^eV8%>1O
z#-L_Xw<%RSkLJG6>(t^XGNby~kF}aQIy`Q+KiRp)p&cmhcx7g=V*qpAM}^O+I`~UD
zs%iOp&h?LGUrb+AUthbvqq2KR7Ic+1zq%PB|Dn=m?~yqHG!@#Pm57}X1Y1o@5C~Y|
zs*p~E+}w%WCs~D8Kjp?0mlRU92+Cg6hnQUs7*tV?7{0)8>V8QJ2H7lGhy?OSz;sd8
zo4>>N7SqC82yToN|NKD5a-AJ>A=aXQJPdEj^L1QdroZRf_Zbz>haH<H+VHMpIqn<%
znz8=w<$qhl!C`{g;pPnkPYZu<ZY)aJ%bq+tFgu=9aSU}^{&hsy7fIT?lfSu{)BR)>
z?HZfezXg+DV%(RmMP58YGfzH$fu2%|6Q!p3=CZN3o@^1#{^v;1V-I57AczB%DX)V2
z<^8fChPf%ZtU7IO{Fqcq%`(L!Au5*u4O}9d@tXCYkhZYVJ(dV6HBS488~g}VW&m!@
z+=?G?27;`0snA5YoHfhgsPw{GDA|^!Tns!ZNMl!d)<btR6&v`b5E;scgk{1_XKlv<
z6KgU6?*<hVWa#dzg9W(-kzndhARPk4nzoE50woa0dlZ>n4r+XG64I`bOl9iWEJFRx
zVLEi4ZRj5n5zBPDAx-bpoRE~v=U1ac{t@n+q2#%z7Jg+?Z>-(6xvmuQRrq!wqa}E;
zA;8Ut(SA7H(QU+gn@Cf$8rLM>N0EYytz)Vn4yaI$eT?7k?=*5D?U}NzXP{i1uwV6>
zm(ExHL;mQTwnv&d8gwNp<HHtP$|t&@JlCDdBd_vZ+I(5Pmu&Lp@4xq$AS4y=?d`UO
zjkINsMyWq0dm1hce=0hXdZmO<vgk6e0gPxtq6iQrD_GNhVnNsdMzSi8R;{ZaVDsG;
z<>5vG{0!$U6I_!vCQ-W9MNv0LU={#<MXC-UEI*9#RG3emVb<nQbwNnQ6FbDNcSnF-
zM8O_ViT{l{!k~0?IAG)|sGXtA0ej~z<yrw>;7&~{-Vsxl{Cdu4at5W-*EF;&Ok%Pq
z977A-o{)3cL*GvF9>(D6pXTJdrbc?bOy!EIy2KXwuzt?_SeE=!nt2;>p=Ksei>a-R
z{Gu+FL3+e9+x!>Zc_n$~X_A+_Sn65H7fSrKxuqz{u-L7bu=_0~Ihdo1K1r`S&d<nq
z^E_nFc`t>2_Rq3bkg}(X%ISf1&PmASHD3CAC~I2m##yPtp@`6t!wG6OjyiMA!n4%e
z@r!(a59wE_>zL^MIFcl*%6XabWR^{xbxv?dXRXN!mg>NV_cR1F{-p#YPD1p^4K+ma
zi(KmE{BV^gX9zQwx@q4IRJ#;(;m3!cFtno(tw~+|j-me;-oJPimfI^T-9gBm0xdCE
zLD8=HPy&Gn1B!Pc0g!Bh!rBOCO$ohRq|(%gxL6z^^@*MGePC3Raiyy!%9+W6S*JzX
z-#irG!gkSj<5&mI!tS}>+>NhKrBw4$8$63{#7>Gw)_ZbSowx1fbfRlyE7nXc@w+Wl
zn&a=JM5t>WW$2&xsT-n{aBAH*>oCq4$RE#mVT2y<Z(3Gx&Tz3U7Ubz9i9AzJS<JpC
zdonn28(a2sjRbTReA~Ei{q;xFe^$Y{3kJxO2u&PR)E!%SHhbpZ2dUa8JtMLz9Q;!>
z@2RCLdps3z;*zeTM#`(sbSe;00O;A-H<|$LIL*+=4`tbdR&h)1Ie(Oc<OD<U-TM{P
zXVS`zg>pIy8+@OJYdf_R#(X`?Y1US<EdCKM)ungGhs)97`l$<9n=~Klz?B=}5%c1P
zU6(W++D&rGSxB%q$LO5F4W<WIH5~}^RNCHSDwOz&D7=5Q1oe`$XDUqr|C3sgj7LwH
zB}#8fnyNJ#Yy@2xo7eb0!QU|YVxd!Ut}_M;f;q#<OC^Eq+C=vZH459zw4CC~TSr!9
zn2G0f19*F_C$_{QDc1BQ|L}_FOG(UkFxu1NQR$2!<A~vZ*B`>6KzRmR1ygVgszkWl
z477lHKESgTAc6ANurI4(6T9fdNILIlLsYu;yqg)DjIwrs$GY`|(M~(3I*mBdgR^Z_
zI;_|7U-}I;e!^gxyHlp8$26z$kq;f0G6*qAsMZhYaouxYA3k|sw{@XT)=u&lPd}Za
zOLo@l#E1<_*ZHYW#>#4ayT42OR%1It$=5!cXumrVPuYJu6Xn88;xBb-ot}TZy!a-!
zv@p=vu;QCZtNZe2ub2^I?=Qh=gFe}~wpkiwN%nl<j{B6Pk6_A@!oFMeDp(MZmC$K9
znMzopbb6ybN~Jv7&3?h`<qOK@a53x--#uG)?P&(xnNgjuk50lylC_dKx`bgKxI@eP
z$znFM{4+oJ7GSy#%oAI(gt;{eFOT0%O!f>?A71DH4*wuZa8;86g1(tSt_7+->faN)
z425SViQALP2k|X}(4IsQ5t9YvAtA2+(KuNbnB6Yx^1IS_3RF3zaQ<wt*X#)eSAIOR
z(Bj4wfr<;Cg9G_?rPpku>x;}8F4f^Z@*blP;lt!rFS+oty<}e6{XK5=vGFAbabwfu
zg{vd``|;{|15|7B>lcL+2e&C+pZ0v8G)FK{Ki-vNFYC-3TmR#n{}-o8dJna+fKqt^
z4}<DWtoyvpGBfYd`CU0HBRnhX?8Db3;mrb1WzVXS6D|C7GHki8oeJVz+01SAw)ZUO
zytcfoV8dJGw=BOB@Me_#-b=bvUg<sVq;Oftp9EhqB~6R?0tDyKht(j>i@|}doeDh7
zi9p0stg6OUvp4@%bEuq$>_{WXfX-Zb0>X_&BRd%!==ehU9R7jaO!G9o#Q6$i5IG{!
zFW@h65%-a_1bDz}@Lyq3Lf2yITfKW6GWw2fM_SpL^P7f`ORe}mSyig1=9FY0TyqN2
z{b*n~quG9j{m;Aifjf{=ucya;ub^acyPdYLI=14t|19>+9VVH6jB~c<PK5u&gDdsU
z=-w$CX1udK6@3gkHd#pag{WVx*FdQD+{FmG;MR=dLke6KUo^SczE*_4&-i@Pc-4&i
z?;(tS!-mgz?auMR)$zCQ9rsSXX&*6af90jroaWH`vVPU<><q~>oiX^@Ns_R#<;JTI
zv@+}EIQ3yx&irucxw;KTn5@Oi7<B3XSUS(Jrn0sRJBrc`5Q?CL-UXzIKmZA$mq<~R
zDn$XMC?e7ss`TEg1VR}Vr6^75RZ$3_0)nDO1Vozj=Jy=mZ+^@*Gh8zhPR==d@Aa&8
zFSY9dNw>fLHO<c;Az7Q?0p|Xh;UtlUm*@^#(VapH9VMVf7l}CjToyB-LO+2?_c=7;
z=1SsIa(O`qfbB~NKw008;Ys)i9=UXJ|80$crIzptUiHZM942~6xj;gd?pNd3De#(o
zmOnPQaQ#%xIez1TpC2Dys%SgwcKa}gG|Z-L<Ot&QI0sdm@+|rN+m5@V5_dMV+Mj2{
z-?O>cKE)+cdxs)ikj-I#!8^s1_;AL~C`vgn*Xdb{p3A<#zBZ4^OT~!C@2Rh37ANR^
zchgoY^LYC`C09k3@6Vyl(1twT@kzIyA35IHKC?3SrOgyYo2VaftZ`>2yv^*M6uc?z
zSWm@XP`+oj@%!>yrRVmRKEvHwP1i);Wr&Y;vk5arGr8u}6;dWZpLHI;QmFSmIU35;
zXp;UpO#R`jZCVw|NG?Tlk{~e97lfky3i$foh$gcX>p94C--ZMrbMR5SB+y&HN2V4^
zJ7NHKvGkV;Y^*y#%|kNpXsZul#m_4BU;V6EqaOtfNc3y5O*PPf?KHKTg%ZE8sqF~v
z(q5?clfzKswKP);6<vP?Wu%u(IJ%2SANlN2RHowR6GUs4b&jjG*X&xb_tlURqZ9Q4
zHnXGC+P9vg9*w=e!JOuzxWUxc<UYeg!F;n^<K)O^PPS6L8&Ug4@hv4FX#^c{o7o37
zs)#XE)7UQA?{VYOJCTf9rI&Br{c~t~M=}fF+O6}nfA}8%LXlt)BO68F_NTStXvqZP
zR8O8QgxmaF<?!v>+de@D+|W^|aTMZdKplXyKUAjQW%GiE41njl5OGSklceL&qgd(>
zf%g~Wqqw9uNT1f>*mbt@F)$P(I2b4#Wmcnu%XF{b<bTjqIQr3Ol_|hb6R2czAMs2^
z!F^C|C9nO-L#h5hCk@Hnf^TE`{G!i>dOF*~FUZt<)NWy1^EI~!Jzb7IdA4JNA(nC{
zVNGLVmDAGwEOxCg!SWJ@D(CmmGu5J!(YIIbi5*`5H|BMX=Ipga`}mt}X-idBb522>
zn7oWU?+4ofj3fRuH`rIk6SzDkX<y&Y;Lv>I)1BobHqsDTf4fonzQDL!@cZN+M<bhM
zcaQy;(QWI$%%0Yk-w@^4%>MS>P^{I$AU|8^_n4xt+((h8GiaN?)V$sBciYXA2%KLD
z$dh|=d)cN-{$v2%Z&aYxa#cc+<NFNeA%rTHZ}NS}?*zr~Q!}b^)-j9;03dq-BfwMK
zn+M%dl?LffM8M6=;YX#r-e&dW$Z8uwi^Oab?6#1_wcwUHbY37b15P?g7@99#kp2ZY
zKGG#E->ll>-kFdb%d5Hj>*1Jg0Zpo0-F)7qo}02x#+<^D8ZK)A#RQM9!v%G2Gbcjd
zeib<>!y)dx8*}bfz@hv>2m5+C$HS~yT5oUxGt8%aSol{+Z^jq+bjI{)-O@+&%I2iG
z+m*ubbth}ig^%am_V^K8_%7W^$`oz%wxBgOR%7FFoS~%Zt*{2mFIy~e#geY;_dWRu
z-^Vr-gVsfG^wWlE<b+QZl?Z217fC&ZFoOjNjHTYCfUCujp!fybL^q@(lAn9;-y(!n
zy!)Q01o&scxfk-l9h~W-{ADhKJcY<t-34lctsN0RTeD~gk!R3~)-j_GIg5@*c}k55
zWW1#PSbpb&hQgKnv`aN_mH)Z>J@aeWB3H~J#1_?CDz5=)q@Q2B-(P^!?(m(M`NEjb
zd5+@?=Ql75MIzBF%)RC!_+dmk%re*-17YasGJwspO#Ofl@t9?q_lG}023zQbGF=g=
z8E%o6g<xgcN4%v}th_(w2$<jv(4V!D#u`yy`L^OkJt}R0pB+$v&=f)qVnD=)qsT-7
zVI=IuNI$UN$t9>piQK#$Tui*=hnIBcSw1TJ+^1sMIlevrlQy%ejoIhRCX`jFvWHLe
z;{!t92kd!PJ^rp!BDAd%()YM`wXi#%psc@)I<#8j<ewKGpiwE1jTcW;KPJ(%9rE!D
z>1We$dLlsvz~2x>1V(2VO@Eaxq(=S{vozp7Lw52l49Rc=dW+2s<dWvT;OlK{NzeQ?
zedr-cq|%?GSi$*h-3x{8TfW}A`C$R|Br0ZC_;*P2MawzEATmM^7T_F=Ti2RO6eaKX
z&^#R%YAYkFb>a9u<!JQIUcy#&9PzajV$N4#Iy=tzcz(>d&-DFte(k~y>n~<cn=W(A
zk2o$|JdyP*FzTx(odM;%YVy4JsW2_#)5UH|znkisomcp_d2>yAz|^T%;^X;}W4&!v
zhH|t#kDmyBetYjgr#8Hxty?idBAvUhL>3#B=<!W(c(>(|LEDoW##bz9Rh@0+%`h!}
zFs=`^OtXb1u#XpcD4iKMl#3DzT0{49p$$5GQ@AqPr^T9>e}A}(bjsu>@j(SCAC0z!
z5Xwg<ZM^s^mL?8h0WlD6JZ4j;%@4BHjf-DO`pa4%GJ--LSYQE}&J_rk^1)drT12OK
zZVztnRXf-qWR<{%*rRogaWV4(?$oDmk(ZlepPZ1}l{d6w7*M1ps)p#41{aSr(gxhB
z_hOzH+fhV)7@-OfYQwd?Jtuznn!5>07yI}?@kg2G{c`-%7XDl<v!@GoQHNQ&%xbut
z-gk&rAK}vDs<7U2%E{1R<=6`t&wsEeY<0)6cr5Bvj$ML^o$6U%#=D`*25IYq6C*qi
zzkl9Pw4{Bonl(ufNxpw{d1$#1GyoJWZdT(5_I~53l*byUPP|`x19xq(Ap8k>tqPbR
zln3F8^~13PKR2I{d8qi{m&V^P(ee*Eklc1&HY+&>L>pKILA>q<0?+5Z*2|Nd!-EHP
z`(`2I2MK%-DIM(`m37JaAhgA?+r++O+y|WNTa+(E)A!UvkK1?r700z&LhnfS9RCti
zj}{#yi#`#ovINS#5hl3r@#kHxQ6r0hhBfOw^=)4Jg2AJim!WUWe(8}F_yo1M^M?cF
zg+AXn&*eLMcK=C^qD|7V1P|TGLeuS`-!5BgCMp7_B;jVy0HlnUPRBp`&Q%9CcrR#Q
ztrc<}-w67Nb;Yz^S3tz^RXg=la|LILUIEQyZUge>dxOM;4+kl&!+lOOgcI8I@}G9$
z&NLdQp?_57;pu-SDbyWPR3(ftyz`Ly+U%VRCV6)og<K}h=8m<5vXZ@DN$Ne0kn~_J
z9UFK-qH$Ge?9)gO)+tjg{qPcPvhY#Y`SuQ1aPpd#PJdC!f#%w==A8Ox+>W1*>Ar|3
zdDVTnQ}z*(zYN#EBs-32DCWt%bnDJSX(#UOwbmJlLF^b4xoqo5PK&Dk!d4PGGqjAp
ze;#wB&6v74k=dUCjaZw3LW4Z8524ntfBqG#T2vtOY*0Kix}AsWuMo=tNs(KCRtaP!
ztp;VVlIG#g{oQew0KjwNUxeDUk&p{NC2l(RXUd4@e+7*1yYW_!$$AIzN^F-t8VyL%
z+^`xa3X4n{8Q~U8azAalTiH?;Sg+qJP5bwTJ!d<LSDgR$3D)cA*4wN+x6?<=WN~^d
zR{V)woliJI+b>bYTH&@DMV}l_m+y*bATLThUGvIaMK=q~n{6v@=fzjlJK)E5xE|@L
zEc+-i7|&SRe$pZN%o2B17+16JaA=)e+#9dzvTL;IawQuSztCE#m{J!DEbMJORoE3U
z|LI64euX138{3<|f)X4Zv_FL|L-8qreuGR(VBd@69<*So4JvZtqn*yRVbZxEN&;>I
zZIwK1>tK+z%Lny&t!bcjgB7rC6e#!Q(CIc52IYo7a4zr0SuaT+E%Q2gr%ab$fKu^Z
zBxOx%(BYrj;mJqWv`UR}15I|N!<meiQn6<Kb}1P)|IJ1B;`H1<eAWKEDiI`mcKEd0
znM#j}E2-@G0@PIt?bK^#0=I*G=6FP^xwR^$*Ci6DBX$aPow}U)HdXC#q+7mgE}KJ1
znL$a_vjl2#(J?=fObw7k;rK+7EA!q}TPI1MCrLd;!2UA7KSO~sqC&&I%wYPZ0_$G{
z6|6q95XF`@+S50sg~BOYuB5Z+C_Yr>1!4mXodD8FWeJcMJ*fQV{zraree09?ZS?b7
ziqamDOQx3!LUrC%b=rs9Z{M=%Y2n0A6uUhEp;<Ypz(l*R%){rbf*f}Oxl)&;WcBAz
z$mbc)nzxS1!z*M0%PyU51SU9|pn`SF@8bMQIP9yl=A7#1#793%A{zn_qR4q1vN*7%
z*(wHBH;s`vHrJ#R(4OB6_iD?*K&V4xiedn+5_p(75ukyt@dIR9H3yY{rs-g^nnoCl
z98o3EJ>E~@TmL(sOaf9SB?7Mved-C85@dN``tSlTi~oH_jx@!|7e?yw0YMDTPaWH>
zcz!)L7n1LO`?ZnPvg*jbBPBTL{LL(3Ov#<u%f$lj_YY3&n><brK)(eV3#4WsRLAUU
zqN~b@G!Ld>4~_wMMyASF;4kHZJS({F{PdB(z^pMoFzGrW^_)OhFjb(7MHmFv^olLG
z(n&$kSOAY2<S^FZ`2-U?6onZeHHN7)Kun5DMd6T<TDUM^`=D-;M#O5UN<q#jaMn2N
z)*0TUo+AH?Ky7Sy8a!<J>bU!_dL1{BZT6rhzF)^E&esjRCYYsWzU&lG?<bL2orouj
zH1{n29CF|J_PK0F%p@oasQsyO{6MBlfQ+pHEM7U3^UIj<zPguPUVXs$7YTot1tV8K
zIWPu~Cu?=cOp~^e<Xo7nnhD&R(2K&rp+z=$hiqyf&jW*9a4V8&N07gtCqa21)Vsog
z({OFNCq2wx$hjCHlKIgNp^PxSNDKvw?tzY4k#S|$5WyFO;txY=$`VHxz4M+cCr-SK
zH~4z>bA4m(G~aUG0$=G*9a8vRLE-1hlKSY2xw2Nmk&*#4L|^rn)UA7|2hMABvO%!s
zK)Lu<@T*2~=MgORgD2FDfRykpX-mR2ILB$u`!mE)bWRuIr~fj4tk-)xkx(&2M5Ckm
z!4RdveK@^ouq$E?tO0<_L5#lybw*^*&2t`DvC_IGw!}e)fF$93I*D&O)rEQs5zDwh
zdC(Db`-j#^Cn*;T2YQTC;*F!!KVf-e2K{AC+KuFio~na%2Ww%84pU{`I&A~8V`drA
zjz(STd(vwOQ@6|(&aZKF>g%MW4dfHd4V>x#oRB({)Xd|+FNOLEh^&iZ`p`H=HlQHa
z6WCIS8-L`d-Zz8qCn&wb41lnJtSUr3hLPg|04P3y6el2noVj3b0CImFX_obE^q_+j
z!Tki%s^wgbU85mD6gJ7=ojUJ}iZ?07Ss7HsqOHz-oZ8VgJJC&Sk|w5a1rxa~F$&*}
zYRZY}9m}Bg4yFj<41hwxXV@&=`XV1O%tN*ieUT)f+U4U=?Te~h&{oY+(}nU&rd*i2
zo43*w+_K<87>6HieMJ&{f!o6y1{?_3M(051zzt0VUL;%vFRk|g_JI9k@0ZU}fyY9=
zC#^9ob}!5IZy*x89m70n4zO8j1CYc8{<;o%Er@xg*cn>0p#RZdwt5!YYtiR7ol-T_
zED=P7b`mm6yz=@2q`ItSX#~E20cxQS479fQwT2-0=s?381>Y!*CiVLqSxsPE!2ew=
zL`Gfd;N-)LoTi9(RlN3lukyELk^Rs2s!D;C9Ow#(F_BEXaU5lLZjfK*9P^sU(3)!y
zP-c!|dUVEVq8nsZ-o`@p$<HfV%xh25UgKBi*2P(EsfWV6RC5F&E(e9V4H9Ii+`wzC
z!iW@VH^T2qub7jh(gC?sE;zR?5!w*=kju%QJK}lcCB}|pG(sLn9Z#F~kNJC|Y+5?C
zd=;ZOjP#rwJ4z;@=Wpg$Bloj%Zgl71bzRx0kE|aZEOf(svbP+$C+kX$<z9U7zH64l
zA^&qy&e*E-@+G=<EV)L@!)3&$XO)lE>$ASf^l8-cg)c3skB50Cr3&#V$L0_2H9qgT
zp4=)N7Q6oH*07geIeMu)e3PV9+G|?S<kJ0`=y%w*AnP1jDTK3{F{i&JJ&kV=ve2tZ
zHe!6YZ-;{3xOHYWZOtQ^eCT>o4X2_-NX9{eIdsTki=5tPJ(*2Pp!=ck=g(xz1?^0t
zJDilh!Ye={gz7rH7Z_ybZMj}i5JS-h`nBd^IN$<r2a&VhXF)?X;?7VLHR7Wsf~nxl
z#VrK{qv)K#oV<Zp_>XEGuxv}G9xJc3N%1X=z<9yP^lUzEYfgju-@)&Vd|Kl)G^>g)
z=id)n2LHP@spy{_@Frz+F!-UwZ{k5(z{IGqgj-SF`7Lr|e@Ttcj7_*@$kY4(D3}E-
zO!<!Fy*Zs)&+)e4_yn%ZYpFCqLSD5e<QVfxd4c*!$)al3V+?0i->2XBTPo9F0SaFw
zP6`)Z>X>E5Q48!va&(_G`gBBO(qq0FKk9VH;`G3=qVDKl?lKBZpS{iyySU#yp(#8*
zb+hAfujxXYP2c+%TaJ*d+VvHIp_dTw<|}CnsN?9$j{)~Jl0JYN`x?Tv7K9RC*R%y-
z*|#80dkFF3wYGf45nyu7-ZNGbcry^^0h1Qo+5a}B0<o4|%>Rb?qDWT7zWh}v$ZPZU
ze(R5t!Om&S+A&fEJ`}Ce)_a<<E1C3Zbi;pSU8iL1AbyGkRk5@?Lgl$nqgPx_)iSi2
zUr@x&u<U;2{wcEOWX9u<1<Q-Gy>qVj$Co)CF6lnLseC0+HtaNM_S5FjrSWlp@&Viv
ziKJ`NIxl<1CqmZ$9BN-v|8uBF`}YB#`Tp$H-@7L_+Yk0x`uajXZ`J8@e?(!=(pv`5
zhYyvtR)4|F-L(nsv~%d)N$gP-+>N9B7)sKwJN0C<q;m&*xq4csX{ZqI2cSr`bH=gI
zk^SJ3)0%(eTq*+_#vO*I*0F5M;BbQ(s3B`>II?MD6bp7SelaH>9cQ(**2*Y$3a%q2
zazpE@#&@J+-XAqB0<0$#dl_)ep#f5W3f-?9q(8mFLFlv1!G<M#PRWt3la!Vbk*P}9
zBv=|Zk6FpHC{c->4O0Ivq@}1#8`Y>kc!nmklVHDd<jKTdLW+9{N{?FgBL)D)lN77@
zu8J(LrI*PQtv{pnHttTV&)*irTpC!wE<4RomL-!~bc@~d)%${HTK6M&<QMjD-YZDb
znUg(KS!eg>P)n01n2Fv{N!}HBI=;&uv+9@qA2d5m?F<QcLQ|0Z$D$tb!K>-UUqlH4
zrN6pLA8DF$+Vq;Dg;&?p76%e0+}zbJKi|GG0?#7eng=NZnFG0Znf=NlK_okB%LR@E
zWL4*XF9lCv$}4AMenS1W1o8$3WURrK?-t<DO|w@1<AvSGmNcVhGW5uMvSdT3xXVTq
zd20Lk)$S+%W-e>MYJDragZ++j>@Nl>un@AgjIg_44r5$OS-9%NTPaXu!|~+Z$fw?U
z*1&7*gl8i=!ftzWJ74Wu;+Q%OtvNy`SDk-~wlHWmJiq54oIUBcDA0FF>&cStu8GHc
z&lm=_4(7c5IYYnNQm{V9xt=>IL^%AaBwqW%SLVKYl6l|t&s_KXIh56R^Dxnhoo9_^
zd-Fc!XI&Q?ZqBe7^U8kxq12CYQ1|9M94H9~JdZp>*fq_^d|TcW%dSOr8s7R*Noc^~
zXKR?AZ{GuwF5i{ZhSRlRsp@;NzNCFiW?EH<{4G<9vcc5bp$1#zfnAtd&>OCTM*8=3
zCB5v2jEW&(P6QfU3SjNMKZgO2RHrzQ(?>WLOlm*$PZASRW?eS4eZ>H3J|PvWOAf>w
z=`2&4fgfIu3r7}32wULs!-(>w2lHmk;hYDxV+7$JuUabtipqk&geUY!I7u)sSUQIt
zzNvlLaozhyfGa9XOL$R%Hi=jBRlVZ+n<?C6k$dd>o(`eUla~qgch2mqvIm80Z@2|u
zT??rAJsTouCaLBLDS3*=6k2oyKk?13hTe6Eo3Y1i#V>IW=tf1>dTK}X4bV=?l-NXa
z`Z&@rG0b;xJ-Wx)fr3c{R>odXjaht;vD&-JPT(Iu83}Ack<QQglfu8K$`uU*N+qS3
zEcG=SektB+E9`9TDt<S1QYvpa%PFdEL1=k+OuM#;s5VTN5NfD?-1s~?=vUPv(|~lP
zvt2?-87J3X9<Ehp4=AKi0tAM93IB8@UGa@CYQ#o2GU>GKFt6&oCUa0Nw<ZJP2X})Y
z%z%_n#3JJ$P$?Y0RwRbnMgtxw<_8XiJBOu{iqiXm2jeF;m$`gs1;edo+O;|KBI2u~
z=1?IO1ws^DOpfh>$#9yWj?-MQDmeKalR(|9ed~;L=#{i3!xQA@)gCit1;{1cZ!iy_
zM5O1cRmfLde=3+7UCc5&;#KxaDs#QWi^h=mU#8M=*NNH^$FqhLRDQI33OT3aDolr_
z=yX)#EdHS!eXiYS5c;ZO{aLFi6?3TGCeG<iI@9_myBWb%LAK*~1t%6<$?IhnnzwG-
zVxR8WZ;<r^<O@D!7W>F`(^$#5R4;6dba2X8^Y`$?aity%im==E-~4z-;_iVMJr{A~
z^hJCMy+L_vc(V2SmAHWVM_gX(A4d)D%}93j+2SdVC2g?_u}K=f^50T8GPOwg%z)Hp
z(V;224<1<Bk?@~05pc9c8@ifU;7HdHvj`&X*ys?UB9N^gD2UUKXT!Et`zT=oKxn&i
zrNADp`*4^%`KvUG>&>)}dP;>P`nKYq<$VFnmH6ov>g+I38e8ZqEez11z#x=$h9(!h
za1pq^c*%&?z~B-1+~uqYk)`&oKbLTp^3rJ|SCNJuk+Rt?65%kua*o~1`}%y7Vu?fW
zfRS5}aVMWUCMIsuXuP0-T$i!nc$Q%@8TVa#*j2G1ccPgyH=eefF=FJS?&P4^<T>Gm
z+<Mur`G-0QJGbB7eVFm*(4h&ZcmBU7U8ksqFV@?LZ;;u^>DN{t_nw8#;i?{$NM?Yl
z*mPWk;}%UH4zY)V#;z8BKqaCjeV&MrG{8KgEl-kyd2JdN9SX;o1*GVs1L!c`rn9*}
z{v#4~1)+lGBZ@W!<b9X4%-9Pz6=OvbA8d>g*<AIwGAe-kxJXb&hlZ|!o)EPDzZzMY
zeh>{uiCNQ{cfSAHbi3M@ggShYJM3b@;OJ~D%AWsSNvLG<RAV3~&Gf@|D)&eHXaUj8
zJ0*r$MK^j_#U5|SoljA2<l6jTpnb_?lvR`Ss#o^*Wgi=2<}*?%epx{6J=3^cHU6qL
zqgzS1H5n5T%Q@yU7*6H;Qj$D|Wng1WSNPRFcWBe$;$RS$l>8N)Z$t@O`Gr?(3{}6d
zvDzOzf84RE$%t%%-`Mxig*pN6NB0T}UspB^hAvHN67@;FhY70m5)`YVI_XY?y}11P
zOV0G#P}N;N2Cul+`sbV*z*W$;!5bjApm7S_+l^Qv-~qh?v7En!>AdYGRI7T*H^Y`y
z<uxo*ilHV@%-aTXZvVlVm_{&6A&?80%|gxZU+7^3+RN??`_=f363=KTzrPecaj_rX
zpwOvf4QDmLxU9n_JM7#jNrLk7o^?tF5ncImwf5#$221H9qD$pFO~al~H{3bydBnve
zJx`c99aA4=D1R%;<+*F?ui5T{+;f%BG-p3OytBsG8~0J*t#ED0q?bHlbk-)bpfF6V
ziDl$-tAWNgW@qh&!$$)F_b!>>I@cw2JM;AD2(3#U!U3nVca8}*ocpM-_f=G?tK|p1
zWagVBzM;U`jiB<277M%++T^-7FW2lA)HUIHHApoz0;WiyKTvMWSmZ&K$pS$zdBL%r
zA1GgPUA`vk&@|Jm)-|UAR)H__?Re&%-04S6oXAl%9&S&K#Q<+Q=_4O>`T`0k(x)B&
z0!>VjBbo@1p4N`^6E}!j8wd+)25Li0bD0P_5GcG>vrYl;fxZoblDmd<`zW0OBf2dd
zc0m^!qylKRC2#{km;fF#3fDp>&EI19ds$J<?l<b<LFf9quSIJLN7;_jvoQr(v8mtX
zSNlv}cBHAy&$q1eEPV*TX4{3aFX&DhwnykhQTHWmUVF|sbj8B^j6sTWn$3W3yiN^|
zX070dhPYdj_uoa){pu^?&!`h<SiSmv!nOJL5c*zqE`!`mQK^SZcj2$AFrqn^RrUQ`
zC$S-DUF)hQo%@dY#x0}uADUP9qx~Z7t7-S`_$PmHcX#+F(X`BfGZr>V<wGyn??05N
zb(KcX8yjdWiu6N$BL~kZB;IXu3h>HoD?}G$#3qPH^psRRRsC<c+vHVt(D5|8fHf`O
z1WWmWDnf>bLx?u+Zj44)RVDx=IkkSaQ{lDnEZVVvmKgw*%EaVwk`Y_R?^3)qGg?=t
zR5;1;BYvoiK;@fstP~Nzb*N^G*haVaTBk9O_Iof1c1%@-V_!E^Ph7!uzY+Et)>=?`
zG!=f=Z$$dVq9khL=YXHOq3>^M?}0$Y>RH|U9VWDfE_4Ztr?k$pvslw>7JJH-czH%W
zAiX}KlREDz`{=2o8gpSR-)c-@CpKH#!=FWRTCjL5`~B#?<{fRxWz|22@?UeLJ*PX)
z?SqckCYM}zoS~L=r`klwHR=19PRhCEM5^;i!`n<_)-hX!vymq@{4t05)^yOly|iUW
zp!K@pPZ}J*tG6E2O>hV{slmYt?xtmQ=>P7woLRu)`@_H-%)KCl))_~OT*R|zK-a{R
z%SO8VWPGxAZcx%c+6a`&t54*}kvZ0;F5^dRFe;xzGBqx6H}}>G!iwT*uvNR9^5+n#
zq}TV<%-BjmQRPGVvRM(iNSDPgrYCL;mgQjvRFC}ki})_DWZgS^?uPK-OX2@^c(gz2
z>zugeE~lC*ssB{lc1IhNaOI1Nw588=!r}BOH_L_c#QZS|KV5SMuf7HK$(Q5!;N_x;
z(>e(r-8jp~#@}u@dk;$O6yL6Wx_RJE7ZBj)rdpgX>vi~e^N9OrF1JXg-$bEZNqMxx
zfAtNx>&x%+quh74nRjtN2kppfHo<G{;Sf2|MNVeias=zbAB~Hky@!D<a4k~W90YXp
zi^IXlaZH0V*k8RW2H0%OBxtvw^HJ3eEJtbqQ2(&k_-Bl}+A6>6Ujml-o-dr`;Mj%%
zfQy)gAd~49=gRMEy)FZ^*ZNuS;fn+aE_1X2B~Rrn{K*n|M|2u2JUWbJ+A@#U$Qtzs
zq*vUyB!9l`!=jUA%uToc^5n_~^-Xh^dM@>dKNUU0r93HlyjyUx@+6M~(`zv_itgHS
z-MFWtr?7_C>G?#w{maBDdVhzau`zv0R&`MK*saS1ytd~;`q^XjYE1dhniia>Jr2Cf
zbBY}{LN<;_bsgM|*R%Iwcajgpb*j|zzV|&bZqLwpv^{svTfir2^pyQr8(FricQs`<
z49};02FxX<>0t^IVKopbP#yHr#ZpZ>M%ZadUuFYk3#4j3H5igL2^5>m5uqLAF6=dq
z0lms46)6I<ppJermM@QujQ(+9g&9=7>A1_d;G;iVCAY38gP(Q#8whz$+rx=KHQU0+
zRCKqgKQf_mFlOM)Jj;@YrjS{*BQ@OIOWHKb+2wJR71~P^g0I-+Ec;u%CAy>d+r{2J
z4?8O=qEans`(NecFJ3*jp3WO(T9M3W&$0}gB;X3GK)We0GAeG_7UFF$ayN}`j<s@k
zt@+_1_<ZIoDUbSCrXN{>QLVweyY!#y^xLz<wx@MBK*fAdJN}R^{UvRQqy2w<4cp6`
zbRAh*QaOP6uLa#2>`6pgwcZjC6o<e?ib%m^?<P=!$&j-eg%WTrFFWO{sooIqQNs47
z>HDhc2gPov$_Mr)Z_I*m%&vbh>6ntZxb)wXLmw{WAMMWufy12~e-0fpxl_vOmlCU^
zELMi+V^eWGyh&z%c2^<(Up|?~{jc7Y7}RUJCM-mB6Vo9inMJ?a9MZY$?R#DR98Y2-
zqu%hp=aLu7qU{7mOqz)Ks)L>d@;_m@_NJ2{Q&C`d?#G)5lM-jJ+Kxgqa@uGWeW#%s
zuD6jY6$(Y(&*_S!D)FX4(|Uq9Ma`@a*=S<~x{PUhm(5AT=m$9Ta*13!ZIw4<nh+7E
zC%sirxOS%CR=6e@mgkRouGW|?T1lrYZX7M?;JG;#`N=VBB2xDveM7a{jLbK@lAEUP
z<v)jbH|L(KwY%Ss+_*lcre8c?RBPq&i|lTE1SQf~y<2*I>IoIc=Yt?@W<>72%!2A2
z2N#}yrzgxvH-0okJUo(wb3Yk(=jl`53xzL|s0Rnw;PhsM=>h^iN4Ejd4-aF)RDanm
zi4C;~3WU}Z3HH+<IG{=(Rw<HtLR!%8nW(D4H~ms~bs6izt~-x>qo)x{v9$RCl#78L
zD<St-%n)LjZFb>sh<vRzt^JZMXBK`gT)QI9`?nftSF|F4=1W%JBhz(!@J^T##z-4g
zGBVR|G3%f;T^mre@7gk5_wo7FC;om)<LKZV0~J(o+zqStr<?EMXqNT;WEj60cPq1Q
ztSL2{`EK(_l)Q3zQgRTab1{5PyVA=dU#e=Hm4kmmt!S3(9@d3(ODZO=ErnwstJ6aC
zw?IEh<RSJAM_J6kZ+2{No<Wxm6?4qAMFQ&fsWeS;g${LP;N4@clh6RAG)fv<1|9g|
zMtUnF{-&vaR!=05M@4G+nAcZ(+c`ODoblI?Xd7Nz4PN)oV{0N|$j_hG{~L<Ss!Ga!
z$-0TL;fR?QZBaiYQB?d>MTv8#WJaa(4uf}x-^V0-p|QJlMwfs2NZYT~Jw2P%s47yx
zI$S$erc%<*bv3eH=k_j*xczMs>`?WtG=hj!S5CR^M%U&jt@~<hIFPfI_9}5d%ec3R
zX~Zh7w7DV3|Kbq&jM7o_ef3O@`O6g~RGL=M;MoG|FzG*szM5&i<1syovOo1wYxzAu
zJ+NA{6WhbjhV+;@ASDY*8Cn0;pF_vfTl4M~dIJvyza<7B9IVTg#Q-=cUVZq#<?nzI
z2|iJ(2zTE6SKg!BVkyzhW9rH|h{p(wdbWRg?}1IqRt*^$L0xHHnB3rtQaL`q&1@M!
ziF083?1=qP8L`svGX7*AZJf2RF00l31D7~@yHVA5p;43SzLr6Sb@-Uckioj5e@YGO
z@Z((NQ?zsSOT&>dCLp@v$}f)XhlYg<72ztjxprDC^s1kDQU%F3Qa20e`?8%I4nvuZ
zE@RLO0VuJ7LB}@_w^09Za5F{hwbmeeP9AOlr5_;Z2*Z3ZTPjQoS)+4)=t43M-alYf
z$%67CIJ0q1?_hqK=qTGBX@l4F(Lj!<nAaMCwF~}oL+d(2h1wC;(T)D-f8&y}&u%7g
z2|JsQlYcy#-0&^)c8hwnms9&G8n^h_SijHFz(<u^aEd7){0{Hbr1{{HF`|l*<pk?9
z3vbii4PK8m7l}1`2~MTccm5p0#)@6<jPosX$dfb+Uz+>j{}o+t?G<uf_V~7=tH-O<
z$I}`{%}n{y86~ODBod@eUG^eTj#H$gwp}3y@-L1qt43o6cA20Af{>Zvjbn<C>9<fu
zcMdP}nBH$5PK0u!QPR(dz0-3WriJ(aZMu^J4iBfWz)ge@qkPMe0;dkmqZoHusBEX3
zeo3cB8G`SeQOb~wqwSNLOI>we5sS=h@}jVtMS*PBlcO9`dbAAn$8oxW@g5FuC*@G=
z-X-WL%WsaNKO8-$_up`PM=8BvbL<c{>a@Qi{DGfvIWqrZRtv3XoyJXaC~R;CA1+Ns
zWq3#bIaERHb(FsvC;R%^V7Z=|b-Kr(p~=bSlZ{yosSL$b>xTS4XAIU^ZW+_GaG^U@
zb5Z62M3^n8=?R2$n}bOA&V&jR=}1Pmt-2aVuxf}vw=t4VLst=r2W<GXlA*_Axz|u$
zG_4UpiI*TlH>oj9OG(m$v>y@){MZQ=|H31kL@#XjJtph~98v?3VtL5U!bDR-zP(WP
zIH}~7(AY071B_lnTJ=KTv;nt~lgloxWw8Vwi@vL^5!3ZdN*9Y>cCo#Y$W6f${Fk%;
z93pCdnu1NJkc-w={%^~!CRBwal;*r2J^HnPK9>LiJ|nQp2o6$Bv2nx(WTb|`&Hx_*
zGUEn6s0w5EgRnHv|JC}c0t(~TM*LIynZvnRE5dXkF$N9Rf~<Bz8N*Bv4Ahc>UrTNT
zvGr6HZ;E=+_N@;=g@kJSs&S|q*Tu<v)vIHU3Wd@+lHb&8qRbmdjvV<#qnCC@$pz}K
zjy_LdVo}XwVDW7_6K_rtzEm%&bd%b1R?lN!iJ0d@&vNt5o$8dtncnz~)$3%`v3nPN
z{waQ+92B4Zh$FHDts(4rTW;vODBoG1>+@AP2`C2*xq!xHd!`H?M?PZVRQyA&?x}^-
z9&BZ*IVS^M$RSC%>Uir}QZrTeLVU_vJH<oC$KxM~lnATr*qNl`zkba8Im8#Jzzhy0
ziOlqG<*>B4oQGK#Tl(L2Dt>+#MVjO76!6G*rq-KZ_3blGnhIsQOW(E08<4C;hDBQ6
zE-_gTkQvk0{gND%;KsjQX6R)@U1@jp%U^x_H`+MTn!}p;Z3I;O+EIra$`wZbenJJ^
zZy7lPKlGL0ARuW^bxZFUT@sq#`>Qo$Y|~<LzU*)EGOH@BV9JrtnO^1%T-Mar#`G<Y
zzPfkaeo-kR*`Zm$i9CO-?>Q6Ycf)w9MXDX&{HI<A#^{DK-VKUN*N)8?n`JrHYi19<
z$yf?2_lYu<4++PJkEiguo@P9Cf8p)jgRUD*$&Zc2eUc6AomQ_7JaQAfOr!f8ds!^8
z-htSCs$FKp-)i`{S3OH}F~2CTq}P)9Z03cmP_Ytb>A-56c1iBIaW~_uyH93?b1Nbe
zvPWomR-U+BojfOl*=%Icv2Z{!+(;r>o)N_J7xt*cv%fwaBPu3hMgG;WeI*2^tE#_C
z+qL>r><3N=3dS<p15scbfCiH7|Ax}Y834#K{zbcN1oj{(*mX`7BSf!ZxvmDJ>Tz&g
z?8heSfx9libv0uutzXIIVFKioZ{CIY;~IfVfXfD8M}h3+#Z-aGS8XlPJsI@9L@^AG
z2PR#z&X&>h-;qWWr=+r!{-d7N^8DiXbf<_dLCKb5G)}9k)doRlE|%~yQ`yNbHgLXh
zXJ@cdxb#BaQ;qtGlKOb*%&we5AX&X4U~Z>piu3%E@9gKVsW0Mhc-sGb8bA-NApujq
zstNmc3-+VMAG@2m1pOE;;avKQ=ML`d%g=<H2cd2gn&gVy70*5QpV#*Ds4z{cqmPSL
zNjy#<G@<DjKFf_Th4c_3HnF;O5udn%($<@7lTuc3!95K7v&6tu^5d3^c0(rP;C!qC
zxE7%X+C9kru6N<x>xVIt_6b|jX{2Ds1(|P;q@te)LMe~o3B;7~4NrkWGWHB+5)_&~
z*ZQhCqdF*@RCx+h{_mzl0h4yjG6eSdnEM515e%X-A4eNp?!d@Rf56`yzQwodBHfU}
z(NaW<b+wHIBZw)DN2(s#|C@BGO3hK-LV{^Ek02y>gK0kPoM`ceK0f&f9_j}Mlday)
z?O44x82fF^kH%lWbJ&(8!)WUq(^_;2zdVkkaduEa6$JH=UqlDI()pc&F^PF%Mcbjr
zmM<tAu*j)6xqntd5zCm!ko(a)_i$x){8Fu0QSrAZ#k-i-5$_rP%^E(var0o`{)EC*
z=;<6U7X7+x-3KmR%GI>>&JK5e66+U&PG=Gr3Aj-kmR1e7_0dQJEmsMzdo}luIBPL5
z&S}VJxOF}BeRmM5_Cqx3#U#$4@Y{i6n5N2U>{H>~o{STV4fAPU<1c-1IBU+}tG<uF
zpoA9(N47({>`n=coXX0KSSh~TLfv>|Qs;E)EkIsp5z&^=&4iTKRxvZ>j_ScW5x^!n
z`J4G|p=$KPX<w9=*N0SjgojPS|1Jhn0Y<jSLG5mk_YN>axi;IUN0cy>wBl3DC$=uY
zUdYD$ZE}n%eOrFC)+3Q_Isr1D*Zk{80%6MBP;CJ3TMOaMSA8nQzD#<#V-3XfPt6$g
zFY5xA3*+eLrvLc6S+(O>#}9tYXRXSBQqeO)K8issjAs%u7EGCI=Jd_#u5YPelio2+
zY;i3Mu^(02Qr!P}Q1XdHvpi>X8m*eTKbs}Zm|!COhG%Ea2y=fpc7i|0aczWC*2lIf
zxjLA(MEI-*Yt`CV^<?ma{JRuKSMd$mD08_pzEbonMHJPMBXaj@8dOJliiamJ$=n<@
z%&b=H(#qkzwe6!y3n5)>ZZLJllbEFkEmxFH0IH*TjIPd~ZKaY&89Ab4O;;oYV7Vq1
z1q#I4z!H4PDXBo0keCAaSkm(4K^$~HgHe0$=GI+=BOwQ?FMive_R(DXn$(K~j<5cM
z1w}mJHR}&^TQG$#<KONLkxzY!Py2bPBtpfzCsi9;&<umbcuTk99VgE6lJ%}pkAO4t
zCI_`?v&Uy7W>wb~u}r&c`c50=Izutm&jr}E1gkGpw|(AFH+xLC$0oI-I7^$9cJhjD
z33s&BJ%_#w>~kfFhRX*{ZqRgeGALZ=W$0I#vBlDN!Ott}Rb~flx)nOl7YGnd+w)7=
z={{#Usz!898mDV3dpYakPC&sI(kI0Qr)uPPum4&RrtnbKBqS-Dw?LuTPS|f={y-RZ
zO(3Jif;V~(6G__@rV5qo5!)3TBB_CeM|ymt%=#Bi7okxG7mNw69QXg=ITpEpEHb08
zBdVV7{fso8XQ}v<LR+zwHRhzF)zXQ_E;np;Y~6q0E}~zN#yl+gkAAyzhaBFp!G7au
z4L_w*-8$py>NOQ^H%DhdnG2uP@F%^Q5*i^kE;%8Q*LaO2Ufr^El=B+%v&BB%(@#+2
zR)uLd#lyzw37WMYy<beU2z0dWVR%9@T}HQJbTd6wT&9vhUo}QPNJTaNp;JNz|4D*>
z+cV=?hfCS8BTKMluzF23C-C#K@zIFPib1k0i*}vRv*yx>h#0e4SfrWx;d%NSgg}y7
z0^m0}@eLG?m%aeD^XFZ~>@4G9lSTXfQGWrSvY7YVr?9!cVRL7McIE8@oU4AUDfH4j
z8dO)64j1-@ZCg>bHQSuNh7cSWg+|ob-mOa~{&%b;JFxV+zHL4A$mea7bF-Q!8e|O@
z_&sLrZU@l>mb4jiN188{zMT?(svFT)uZa6;$S1*XWY?Bz^ahil%;5fhhl>yr#-USS
zqJb*rmqd#VJoz?eJ1T~DS21%`S>ImQKQ^4zz1=!sjzn0&2j*(9JZcU&Iw;mDnFN<U
zeDjC_pSe9S=^sKhl>O-lGKZk_ZKL~M&56dJ7@31`RVuna(cAoDnfh6Nes42n-k7nd
z3F$$am{R<#bB}NAw_>}{1GKml;ZUgsSDlG_fuecSF+yVb>G-7=Cr{&e_ze2aXy_-s
z;A&3C)+{l*(Pshpg``FIYy+*5I_!(B{+h#S^TF6;^JLCBuA)Y!58jR*C*5C3x__RT
z0KgvDci<Q%`q+;*u8gRJonV5lZwQ{axLj--q}QpYh!*NLo^cGH-2ufVd#|r=XCOjE
zTUT=|PN#U*pY_t;9VjryZbgW6ErN;}y<K?<%6p`>B4#zPg(dUgPdr610|eyGEuA-+
z`3;kj$-h2+nqAv?RAiFC(LvF#ELHuUH+Hwkz+w9JJ#MKlqA~Od<1r^Z9}B-7HSi8u
zcYir^hUHaX<e9Z=FYB~tuKy$|STjXD*K!_Of6AY*DB{CN^xJ-}%GxBBCV0&o(zceS
zY5g<U2xzODNLloALHFg$Q#5!4nQ4M-ORx;EUnt<0*}f35bqH^qjdM-ZkA|bl6W{^B
z`E7Nn)9&bX!#(qLB<5_{<T8>9TjhVU9J)9CZH)_DD4i{h47*vwEJ6w-G3}l;0Hg_U
ze6P>6J6|usbFyUp>OH#75r&V|i=H&w-10*Wv%Y@e?^u3|?BBraJoHesm^NqcG}5Xb
zo!PJIq8erCzc@AK@LYr5?BVz9HQv7@L+I-ClP~EHs`LEkS4BzMB!RkEmB(dQ{ChT!
zJLeY!#O&v^f!I}8Nn7?b#-Q9;3uhy*n*ZPZ^Y4lO55CO-R?zs|WIX;VSlHSPR?TB{
zimj9WDG)Qv#E$<MjvHG;6Crt^P?>2$*NDwECD;%`;~$kz8hoBH9+nWh*B1>|54YyF
z`fe)M{ldKcsF><sr>B!Xn9@{e_Lm4^#Cfv_bG$@X(y>2>9yU`;wvqU`e<}J49Q7S_
z-<i9^GL<EU7sAS0Hw2Dmk!|X3OhsetDP41e+$W@?5DE!+(EzWK1@UMQdZe;eifM-c
zBm^fBJQH9NfzAVKMwoOvNvx0oiq^qwn*z=yP&cQM;Is4_6Ziuvz$E#Ula6Nxb_0cm
zMUP8kMp~+A_ZwC@*fdaEEpzjA!|64;8!nB$5<*f2{26BkUOKp^HFM#}52gawo(axS
z#hkvGYDdK{UUZk3U|kdJ1%8_c3&o*syv-U1_my9s(m5S>J;sXbtGi8if^bgL(4fio
z_Prh2EzE12ANn=_s?M=<2XssYR}mU`y9srPDt#v$!jzLE_PEUczgv)>65S4ZEZ`;z
zF2EUiPCr^Oe8h;=pUDab4+|w#-pGSAec^W@?t{&?thQp%jtoVW3LNAW-jbqm(=E#O
zG+K{EdzM~Puj%O7c9%TAytUvbcl?sjw85W4B`2othp$En%a|?NsF=(*W`$-4tr?wh
zOgr^+qIpbx$4BU5<I`%zt0iU|LXi=!StpM?y?XhAW|y76!tdM7Zl`NGj_6&MVeQ~*
z@GjiG=kGlXogQ>{9iXAUr(%Y3)Fw)Q8a4$;-^m*4$I?k^@qcUTknbTQ<8Kkwpg#mj
zU&w#Z8L$&+#HNU_#0rFuXkGUg^B3nx!DAwC6VppL+KU`|^IHtG(7M=yLmE^+P3ISu
zLCQRzh5tzsrzC3UCqSlI{^sh3YR2L#PpipIyuA~3)-R7=TU5@A*EiP8_hox>&*-x1
zm3_^8{^^_cRG30z-?SPZa&Lph0cq$kn#(@sBHr+4OU?+hXmhln5A9h|g$t=-XJpj>
zH7k>CqJkcub`wZo!m^BBsnkdAzG>jOA!&4^p2TjsP8$raq9tST0Zj0|o{n4%V17%Y
znh%a0!E#|HDDh@fDQ~l)SI{J7bi4BZhT9OthZ;#B`^3kWOhS7wC5)OZT69MDzd1QF
zZ$bn=SB1W4lN%v1>Pebj0hg23pF>t$tI9V%1fA`Ba)yvsUftBE<cPvE2A{#{x<5x(
zs%Pyf11g74Ig|=<S&slFp%OXG6Z+w@=Kzfzx$0Y>jM8Nh=k!ejpRwpcZ!&!`gb$qF
z&lV_Dv_lFU1c?!0#CiZko<>=CGi%?ks4~opQ!~}c6pks#sCl1s*sSS#&xlP_txI=$
zks*h`?JMlNOeY-W#ySmdH>+v?7ei{!mXbl+kjF#bnO~Q%3D3JN89AmUovkRiz<dln
z_slgXsQt)TPx;OM>Au7_jSK;_T52}M{N6OYAa28akV6yh7^Dehx>%hnz4g96jH4mI
z{X|GAc3_SM$L6IY0@emI#7e@36ulKXzV(y074(5;-nm3jI7Tp0dCuUG6PXsN0+$}>
z-O||YLJ63b^k0wFA1AKtlHXrz$VlvVdBNk6)jdsccAR5;dQs&R3yt5%)k)|%cA2;p
zvTjye&6(=UPHEu!+25ott()iU@4z3!>}H#!8f%|Lb1RSidS2kyExDeUzT$;7sH;BU
zhg=)7_h4;O%auwi<pcWT?3(DKL6Pc#K^f}eX)K6*hBJQjlnp475g9Zn2MaF5dT6)e
zUosX+P;EH>lpXH3D1U$x%pBBNVdadL<1k%#M!XX`zp}}*@2L7B`h2V}Q&e*vN?P01
zS;+ZY`BMt-%%&{e)3euJWP2?qY~D@EmF~@*u4sbojNu&ZY6tX|8C3cI2ScB%ez0K!
z9ijgyW&-JsF!4V+BIwUb5UJPIpi0I&E(5y-)~{`mR5bAb?jFi-@bC$QkFKkt*uy)_
zydO^f#{A-8ul+)PbcK+$(REXyoi^ZQcX!b7ie2`DiUGIw1G6JXmapF#exr2D@qU|u
z@<)Hgu^5;Ac;Y=tL1VU0!ckwPck1*rqsvpDO=ZY;F0NT74VoEXb4F^l9}XXBdqrQA
z1@*m6u#!INM<{FF(D`?8P3-QVHr+EZf)nKO!P&f~<>Fj21IC2}v0{tMut0&;D-4X+
ze-8b_LFt8sk5<)SdL=9LoU7j;-y%U(Fs<zc^A{{Pvj){If6V>L`OBbL4XsIs)&;D_
z?SB=T!wdwJy_*ZEQwiJ@sogtlg5jO~JzPH|NWfdW#;Gz_@%<1m^i}zD=-tC6b3#Oa
zth2k^Zi%<5rcV#4&R;<_p}!iROQv}dQ^t3{-Tve1^Jl$66zxcd7P^5|=TxR5DK!_|
zO{5_cxWdo@1c6G@YLWxk_uF~G)bzg>E^P3o%9qEOFJr!LVQF5Z*a<=J@EL^66j<2B
zjQ;TuDu?LcCX3z{ppXuZ!&j(Ex1mad%|*#j{AJHfugrnOM15aB$>Gj0NdGLx5llK%
z00RVf7S6j4$Ipja-VV9?`aRaeNqs?%P3*-M4f~bmE`zm`l>ASZ)5>rQ&ybRnj7fR5
zgac|IdmCq*c)gT*SoJyGM2KO<es7eO`t#__ldQVQ9UaDBmHV#SgqdDvt5%hr>;U|w
zmTq+XT^v&(0EWT&@jtEz&K217OufR$L=67XH@p*UUbW>lVDj%=9hkGKt!)ve26VE*
z91$%`q&+>ZpRMS5TpO`PK}6s)9!%u=<{BGNynq5BqxKk0gIZEm&h|Sd<JK~gPBqu(
z;}N&~3k7;?l<)noaDA_r+&sh6Bq5mPNS&pF9kUc;p1$@;ChhPP)0Rs8VQufnY_;zn
zKNvr5zAUHN!B)C?;<$L><@Dm|1-^{l`;FrvqcZb)HV2pvt+A9~j@^P+U}Z%bXPpL>
zacD|OF!GlOH;bPf_E=v6DRPZJu_Zo7OG))rrQ?Xo>1(+{l?)$R<0v$1${XWX5zNZD
z5oJ)m0mco;G2J!N_B^mo$@3SkeKg3KBPjd&U`1D4TIe^#k|vQ}dL625S6#tWf(`ul
ziYlRiJCcsALbXoyel;|@_SPGCzQTX%&=s_DiCKEYq#s?1(bDj)^7M%C5N59nS)Xb>
zwy+&Cc~kHwBDKHv)}TUH&c24gipY3iI(W1Znp-l=ah1(dG(Vf_2^P|>e0o%*i6zP`
zz()lu83<pHE=ev#J(j-sw}?TcnnQUQC<7o`!w4hS*0IkoCVN>oL81NYh!V%!9NxHZ
zBq<>C9rJenOQk#~66Py{hb7~#EI&W9Yl3ZZ?%4yq5c-b?gSXt-m^+7_gvGByM+9tR
zxGh>Ztz|(L&df(O&RzvqnzqhEL1v`d8<=*5409VcY)FVFZBW~P&KuTQwprlDs>UEq
zsbFLW6FQBN!{54bZ<00qb`W)R4i_m71UX9F&L;qa)23<`hyhcvrT1Gch=M^DJ~RF0
znFx_L@M<Q(G;~cOfnGWsAp9s?mO;s!I{;m*!d*bl)}JoHf};ei5LqcY#fkFH337$8
z&9fSU%qQkcy$FJsN5!*2C`aWJ6a+))oB(Mcns`?#CV!+b{B7f~AD{fg_EX70P=ZP$
z!Y|9zp5O^3o8r7laGFeDwGequtOCM={H05mEXuK%`_U^I{R1uF>p*{n&<LDv;C|Vz
zgf0MWBP1XJA0triakjYliH}fG-yF*EH2honmUk!RbO6ekG>z4bj{8>rQ7af867K-l
zHD2B)3HlJ}j@DY~kfQzebw}2zO1HVj^>59@5Ks#H2&t|#jeu~bF4ljwRwQ^!>$hDe
zN!prEO&Pf^sWbwFVn}nXJXpaTuBoX>C;d6}_Mh`*tv2^S5QT6AQhJ8s3oQImO_i!c
zWgV=bWeEfH5oB*?#LOI+-g*iqS$xXk0#GD-1$Yd$(Kz&FT;K?R2LUk;Pk~-3A1$=N
z0WA}LUE7$lV`P1^dk5FlFHZ_XZaKG>1nSltX;0ktQI1G~uE?|v;1f%r?wbyQ7l3lM
z4O<g1+DE`4@|`j)XzLxW3M8rITr$)%4BVe{vGF})ZVF5$C-A)ir|>7_J62gsnfGKN
z#O49%78jfg+<t+a-z=jB56+8g*%W4-ajADml9*|ZT>h3?<z2ddjgA!oI^xh@0Yw28
zVrY*4b4V=%TsYuDvcN-yRV(#T2=ho~Emy*dS{4lLhfflie-j0M>KLZ1JXkc!TFthK
zi6y&nIq1~xZANq{Gn!Q1ym4>Vo$tS@C8w+AO)^H3W4}FXtGnlIkJ_Kyp=7--gNfS=
ze`vqlJF_6h!N$Inqor2@Sig2~fArkJENN4i(_UJ~yi<%8dL&^hDd=T>bFzNhJ%(lV
z4Y{WDc5!%hGLNCt#T*mS-ay8(vwS`UJ-@EgO$846d<}+>Evz$2FieN_fF}|@@%P-T
zNSiY0Cu#c&?C-V(U^f-mxb<Wj>UZtcxc&zik;esIsG|QyET0KLXGU~}h^kX*#L3<=
zEwjDfyw@R?8ucru)bhvmTbnquO3u@n0SP75Qn@ix?c&)VX~ORvCx6nOY{iL$JH=>=
znEW1!4|A)zx+I5J%d*`Uz4@q&#`LJpP!DOr)XRx1lSi&!Jeic)JKj{c!T7ANo0<##
z;s>R@kZ~%ibm?jAyixB>W*6Ml01u<ulme>S)atR~ZglJOgfmO42M?kz^m3b7dtEH%
zk`7lt&6jBCWo~ZAfH0Q<hY#<x!XCXJrs!pnUT`Mb9sM9UTHuXT<o^9)!O;h1w4^Ae
z0(dhFBI^}ECjm52rn@CfEE!vM&dz^zKV0L>fARsGjK{JB!LV<6HsrKQ(<%Lzu|5)*
z?x^G(wO4UsYKBzx4tHI?Ea-pVqQCvh7+<Izd1oq;MIy5NV@$@i;rI1Fj_^a%Z?S$J
zdx*o?q#(Wpb%PR@n8}%8=$@Su8d@#q_f#98^<Z#v{e&u^Kj?OBm(uP|JmEUIL@LN<
zoO59RAs{q%QhR^gEV(qxu|Q`>B`|h^<8W))8B|cjoZO8sC%I1RaH~#4k+0F+;%6wz
z;KDsl9##GDddJel;F~r?1>{l|G!R*=stO|a09v+_L=|Jf`+h1Juhshn=Nd+&nT7#h
ztvW>KXD=ZK{(^g7VE+$^sS4!L+pLlYhb?JzgM2qW2~Spl-PwO|^Fsg>kk?@N=H%3x
zg$6*h7LeZ`m<19v3QjaSf#Hv8rOY3&4b(&Xtd9oyH{1^{ayD@7yp%JFkid<INIL~r
z>$54+JDt6HR*%MvJGC{^P9}`y+c?qWwCAH|MqAuXrmIiy{re7GaxI_c%Nc>#EndMF
zaVP5?r7g5_&u+<&E+(h{_D*wK(6}jYP-sPJkL7Stt+yw>%oS3<a}vGa?|XCj(u&D~
z*~I4~M;P<4YTDmSj3ys!T7^5k$V+@1AHGx=Az4kTq1tSr@~EP46@00OYsS%A$gm8w
zWxzx!iA2Q$kq6Or)P?ElcGT)}KOgl+3j895iA1`;Z|vwEi26u()4l`mCS?97yNF|3
zi~KjDoEuQM!=wQOk;|;`#<P)YKj&g%4Ut}2BnDjIhY;)1#=z2iPOV9nATkcU9D&}>
za2Cuxsh$yB-L_O3P35t=a6!GoBbQ?8GsLa-(NyuI=ySV&6D72#B34baxbxM`^fznf
z%L>QNn1#yTQ#|}xh2~8C1w)BScJiv2{70|%KE3*4jSjAjiLZ&g?Lpi{#M`UFQ?E)c
z8rAi>qh-cv>nH7(G<)iKdfuc9J!~O7H8)jwricq4Q%nn;&pn&Dp#8RkH__Yn;FfaU
zQ{Pn0hS4VwD3d6w(*2?D>kq+glq0_%dYpE82SV*t`gYe~EmmVlK_X#@#3Dv;M+}U~
z5dbV8?j?iK;qR6Z8U)gt0O*I4b%}s6Rj}BIf^aDlUh5&@Uw}#tq@<a`(}*;RfjbAr
z-|#f=^1(J5U6zpEI&g%b{qXFm5vTg{c^YSF8mpqvMXbWVfwbLr|Hi&q9I|Q?VCh=7
zov;XJ=ID5AyQJ0PA7ag#l3}VWXm{F`qaimycJF+|a;&#phm?ehIJR&)Mo!;0Q^~49
zEqamCK8=s;_&=Ji1Rm<PYY!n4VyvO8V<aIGPuA=PL!^xm4Vf~uXt8HWV+%7OJjmLH
zvNX032}8w<6o#l&DqE60edq6azutLz`(>Ir|9d(2xzD-IbuoQCcXd}v2bP~xy>lk~
zjeD_NLGpR;Lp(%J@`8+p|JK&-GbecO=?dR8R>);*WlG*p^;6fj<2ttQ!4;YH54m3n
zi9}1vGV{L6wxSYYd~I9qPAxvhrsPT;30Pm^z#iNQgvsAmca29wJ=5uGu4$djWR76I
zg}Z~L2Y2|Em%qv^PNY7>5<|b67;hDEiGP~$9HA-!{f}6FQlQg3(%ljR3g1raSP1v9
zP=PL<J<9D(6k`BUAT~-ty#eIdT4PwyD|#Cm&zW$ROGl)s{g(Yva;~DHa?i1x&y%&v
zBTQAUX;W+YEB=$2`Pq2C_=ht~J74(^bclUGZ>}I0aSp%N^>1`PepUEXn*-s|xjT#7
z4q=q^w?C7v$Vla#opJCzcP&H}NUz8Cym(J9(h%uQYR(R9!*oBMY1lH!8T}{T=2WVN
zO0Y2RiQ6N!_swu8(#k*fa~m0*cre~R5+_;9To5j!nqm{I<*|K@`c$9v6@NY~18wBq
z^EM5dnK2iF8jS>OYOU@(g6>7X_ISl)=0w+rjL3-%ktAp*1)NvLL%kjv$ejccE%}j^
zmMO0dV*_%VaANOWLnR5BpE`~x;)L6Hv;jZ^Nvq##Ry3dRTR+@fMtOWDt^*^Wz>wd`
znzm%!J@4d$DLA4jml4onD*R*9sC{c|&W`Ei=w_hDK!@uMhMu``;)-m`8)6j3XMe@5
zHl8N;!18z?E&x%pFu{_Tb3LX_f+caCi+xy;f{c%Fpn$#uq>4N2A##DPY)FYMWg5-L
z>gA(L^a7}`J!rT~rM7o0emJ}=R-^*KL$Fw~sh~XEZeP(-gU+~@{#K3db<71U7ZD~C
zco6n))Po)+Nsga|aJ_|RwB1Uf=XF*{U;w>h_Yt$W-N_$j9tS+_bn($$a<1KS=JM-W
zgNQM@^F_D(wSdk{*Uh<wN*B}>%cMW`9Bq;zz@5TSuJ$q3r{_ycg&~q60=+gLU_HXz
zY=9Z?MLPVE033n{<UVelxQ#X63Wn^1mS&aQ2OiH>f7}?nX}67)am@ZH1ZLm~du*YJ
zJ`LdA4>xY;QA>tfb^_JV!s!NV6ffZ%q#YbLBVi^)qzE`WF;pX8T944CB|E=hxqS11
zM&#2QZY6^k+z8ewgACrgWqaLz|2b9rompurkvcP5d}KR2b(~7*u>M%F<&gdIa@g(u
z@*Go#;8QCiKP+u$)_JD=4;sD*zKwm&U-OSb=O_=_uRiy<r{anxxzwCgk+rgXMN_nx
zwYDcl9-hs4<KYnE3BzxIA?_mfIR5}@RxQH0Qx4EO28z}3Wk%IKmi%XJf5NhQo8bgF
z%X?*9ZA<0V_aIyD7zz$P7}N5tpt89Yn8<?Q%yf8OpQtMN$?6XKjZg5WBP<0+{Q{jU
zqD;4>1!goMd(%WuOK_I9cd<Ty_`O+NXuh5duC>>#XrCl?NS+b9VEW__TF?qpmy%8M
za0qJM7N6esb!+miv?{TYr}Et+Q-#$f0`d1$bynO2IxmM*Rn#TDAdG$xDAv86{<=r|
z^->Fpp5^7}U=k9~|7CP(w?vg#WV)_pN7cMV?P&8|y}cR@%#>6Y5I1R77(NkKgrh*7
zHw`8kVJb%il0ebCc!Ns-R`3JJTNEroCdr}c!6Ch{ys(Q5U}rzMvN3$xF>edXBLU(9
ziT8Ln_i8UDdT<gijAYmQ8IH*3A+j$F1tcc_&>VLO8Z8Wl+|>jzfnT-iq&sVW1cU^e
z*zcBaI<h269=45)$9U*Om29b*@Ht|6zO4I%WXN~G+^hOhE@zAm3@Uhbrqvf<AMWU=
z7tv9BGDZD|`8wpeUd%3+PX=dN3odkSi&dO8ia9qEnOC}RY*=5<`Ed#FrOWznUkkp(
z`!aMc@{YZfJUuuuRo;E^s@+k!)b>N-KS#|C2u}i@0Cj<ILZqtIu_GeyKke)eIIl3u
z3H{wiP50;%N)Os9SD9;;Kaa4_s}9eD>K@dT84IQg>ik!s=?D=u+jDA+kz2aWD4`F{
zywr1o7z`xScBD|z=Iw7G``s+2UO@xmwWwQ-LNVV0Ac1)u2<O*}Lxq+>?U>b(^v@@=
zmcnVIQP`oS<*0JXFJE8|wi#;@<hc6X;KhXyA<K+w&!b=N@eNLEF$?8bb+EkNIAh^>
zVKO$H$@Y!a@sjlokD2z`y3*+`Eo+lYcXCU*QIR556FWb=u+MSw#lEs99dC78ChmTh
ztQ&8^4hi4CmXcICup&E}WhOIYrl+sTl*U&aJEJd|^utR-n<4PzCCmSEnDzpfy?bx8
z0^fuL!TK$ZGLk;xey`zy@zVA%9_K<dWAWrr_CKz#2etAZ|8$tw^U1u}uARS&T<}4=
z#&md^`KL^3$s(bnGchbn<3&v0%qMWcfcFD(2yho4GFn0v@=h<_eFti=V-Om_wo<Pt
z5Q=#N@{ds&EjEB>ZWIkZ++HLuk$yY%$P=Phr<|j~*+&XSfe@1xtLFuuvN>Hi*`f6L
zz7_ZERld90hhN`TbehGS1HoKi0RglCfQ5k@s)G5^>|?^cy?NKd`<NI1S(dDr-+Ox0
zL}=LU`oPHI!`tD(X{S41tXKUJ{7rhg+wrqt$aQAV;SN*ja?8Ee?-UG*UMbRM`X~>+
zq-!*7R&J;m_qz8~A|@xM-nBxn{fg23iEUe{K|(%h?SeFl8hP9=cK>9s;_%%^gF>-0
z>u8DAr&O!S7mlAxytm)tF8!ED{zl6-x>1^d8B5)HK%}l#BTr529gkoM@pKG$SjSX@
z%8&`hV!sU(u5u}+B*0*o+d%ndi#x^s-bV8@+~CT!kdxq7gttbto}yv32^t7PFc^jM
zJBC}Z@==)})s!f9#+?E^*B?y~C|L%yBAHD-d-}uyqu6$iufy(Zso`%y4g<kE%e;ns
zW2d<<@xTAKQ7zwMds6k~Z9VTXhX(8?4HkHl1LdFle^PauF3+|Q4Liynzhavj7B#-W
zpDeYgIlenvHOgFnTWn`qb;(LNrMHy5r)1^6l3T1&+|48DuJgyAj7_QsyQgQ6+q<55
zoefFm&&;4dcNrA_SX#z6@GwZhx5_7HTEM5QW8t6*o8R{GV~RUt_+su9?@4|3!?cl+
zq*ov<C&9PQy)EMOvwGo#*(-tW^D9P8QXWz8{NR_sF(EAsG^pv3B<w)^uILA>B#&5R
zbeOiqeLo)VkkkUF{S$(15uoUvaE2dKn1L-(#RC8gQJfI@2@CJLeP$mtJO_T=`!a+S
zg&&KaaJT*?WJMdf^VH$~{xa^l{Wns)W_Q6HEG?lr-aTQC7Y!J751f4ewL~tY^utyw
zH=#`7)1CB0o`wLI#Fg8>Pg(Xv4)9~CLLXg96sz7DtO(19Cmr6L)S_^keVD!|^W|f3
zaq2@Sc`V&cTY=!DD0<9%!B1Y?>|VpS(REGT(1){Es_h_S;t5@P&Iqg_stP<WaS2JS
z{ke@0#LPNSAq|q6nvx3#FVY?5pr@eulLdd52?~Ie4tu*yxq|NRSaXkPM=oIptIGJ7
zDJB*fub%><K5AZXteo%J3N6)To6t%|?Dg5JTMXkm%Z~#+9wuDD;Qlh7tkiO!^5}Zx
zy0DK4;7@izmMs2dUNAvkFCT6#S*%;YWKyj91LAW+cs=Yl<hmVY2zNn=7Q0Ls@El1k
zs{&~Ch174R)@gj7jR*L=u9rYetN=VmFKk)i*S&r0Qq!0qgh`>)PiIa*zTDjf(|+Sx
z6Bm%r0~8pcmEWaa&B%>)KHNL_7nP~InOm^%$B7H4sD+UH0k)ih1){<?WV$!ffu>g>
zG}=O8dhpLHM}oqv#rVJc=i(uk0Z@<K+aRAwg<RatcsQBT5DR0L;gEv(g#v$6h6wEk
z>?#XHQ`OO`0{BJoW8j1k@&buSpr3=xBPH%0!r2!p3O7GE;cpm-T`*A~<So<Ic-t}!
zQh<<D6pldTBW?TtOC*3^%&uMt?w>>dTM*y0;h-L-W!|HmL4{hmx*ci5-^HJXYZ0<G
zKgE253<T^_V18HJSx~2%TV%#p7evAvXuSYJd0Of?ki~rewkT7a%()RtDO1%f`T*y+
znNokgA)|Bg4d}cGj6sqxQ~LMLOn4ZGJbCRYgmZ@(<Ji@|D2sF5Fyoo0h8+Mz{=+Vl
zh1sULB&ija65}CYeufEh;!eB{ZN(ZLN0p##4xJWnlBxBPy!MG0pdADbr@XMuDY=&r
zaqcK2C;nf-osuYq=)7H(DKc`~obpQ~H%swG{;{VR%OJ(^RblA48%FY~*)r7*1>n~q
zkc5a_qj2D+?PR&+)o{kAJ55#%NUPuPpHMG@d3XUmmZa^_Z8PheSRTm0ny<xMVb2@|
z97X0v27QGK6ADc5kfq*Gl24Rqj|Xl9UI%Irlgknm{z3sld&C{;=7w-_ygu)k3WN+=
z0q(KY?RhSvI}6H>&vz<{wW_w5jg-CyvBzL}z4FYxOu&{2K8jis6Shn?YjelyGA6Nd
z2>_n@ww*RU?+qKpSG7SdOAHigAqqy)Nx$;Z%Y6l*7zS7FT>$9`0OT=8C~vvIe}2c<
z(I_n=2Ty~lHfX&;=*lb|unIud+X`A6?bVHk#f0tQOWcg*Lno;Ce&a36iqyc~rw-*&
z%fva0u?G%<w%6QuE@&BM!jo!46uNh5kJW+FM@xEc7eJgw;@OI>wDIRxy6Da;B7{1&
zEp;T-33xO=H+O~|dwZ){7Kn6Ag%Uo&`N#ZzReRYM5MxyIGE_?SoI(=v|D8`AnFtGn
zs>a3=B4W<>qvv|!3wYnaH-SD)%v#Vp+c~FsR>u7w<^mD_j2w$;Mt^YRxZoYfRC8L(
zec2PVCK5PCd6u=hfT`Ovo4q@iA+N}%a`M5;k><~rH_whhPd0WgEYF93n@aF~JT?g-
z!_x~b2GR4YabuJ`+EMHU2U_nVNU76a{Rj}g3j8OB_KUry$Qx@xj}Cz0eWTPo(f#|t
zegE`m?Xgg??Sqd3CdXizC;X4xuO1|~A52n~(xMEmY5VhNe0W@^<<4K|*Q!{y=CAf5
z=VBL*?%VQ5V6j2Ct%3CPyfsODTF_pQ7hQBOU`0i~bi7`>OV`D`tnGVYx4P}^RPV7$
zvrZF_%l<~;qifZ{8pJtZOMhP~CWO~}(lVwYK)w)?VjpAfk!<6X3P+8ANM>mAsh1Ya
zTx|oSzhRllr96vh*)}IbhWf_6s6b|*qG6|I-|-y9H!(}w;a-m0FDcM(_GHG2H|UO&
z1*XcpBoLospYNAkd($*3#j$pub7V_>qqgj(D*~w|`GV=sH60Wyul%@lw&{}7dbe_I
z^+^M#FYoxx0>Wa*#4d(!`nfu{&vKMUH#ZOZw|&3s_i*PulFlQxzMyNgXxQ}roc!;c
z&#TT`d1m}9{5P3xRp&ic^2I*r%I?XdT+XL`@iOjKB-|zWY&Kud)3>6o*Z-mOReiO*
zIhwIn-aBvcCMpOU&CF;S+sphV(^0XR0v&v7o#WKdi6e&Mlw||T_=0-EoAND&&<QRV
z6~STKzu2e81y5dd0m?~tuN4je3N!9gNocw#w4k>{2^=@5+%NZAq%G2j%UdVKWJ|K?
zko(6sB@BO^E6IYrK5x$Y3EX(D_*;<fZz(soUR@CIR9(aNFrWT0<AGCmFAJ&gZ2LXp
z@x5P6LrSGHX!{RIN$(K>mkTPp4X<M}iTli7eyBKUCRkW!TRkbKbWx(#awj(hWaaY=
zce}r_zD^MI4R|^qK?%v&ej@LI{IEXJXGv6YCH7N8<Bo^@_m)Ka&Q%D`8y_%j@4509
zrMf<ZdC_@r(P%uyfh+WJ-fmX81Mb#rm&CPsc-#;u`N$guKIKN3%wI<FHb*ud08d|#
zx%XYp=^kEFlHLQkXs9VcFfzHi3pN;&dkzrQT;|Xc5hP@VXjNnyF=pN4tQ8a{R$;5f
z@`qZ~FP2*{Zl<4Y@O~%luk+HfYgSYHk7Mmq%FN^89UY$=D#A3${q>6Ykcw<FPD6R(
zxA(#is-d9G)|tqOQU0w1&Fk^wCz29f4R=u5Qq+#hE!(&LG_pzal#S(nIeK1XRe+CZ
zXf%KNl+3i?(1cy?joQAUeF2fO6H@0+26Lx9?+fz1p;$gbNxrzB=^FIfM&YwG`m5|*
z^R%IO#+PC!0Uk-V43>!e!@I0OdzI^@!F{MLSgV85q+BPf2&rV{Sy5mlJRvp#mwy|%
zc`ZvvxBcKr`6Xs&ol@+9`im*K-)Fyu0_BtGEE5ni5QZW#eiTG(jc~$i@j?Y?`F^;9
zh^0y8l5E-DKQ4b!k6+y=dDppeh4=Qe3<I}KR^gahwXdRuFTEaa(%gQ@KT}pwcO~MH
zD+qNrf0oHM#=ba`dg<aoe7J>D%<p26vA}zixcu^uyZC6IQ!?r5@$S*s`Jhj`K8s}E
zADIoWdtbNZ$@&{f^dm*tYd;IF%!?W-E(X>V>`ugK?!TbJ7%RQ98|{(uHb7eN_~Egx
zhQFw;>12z!e#SUlSraR+lI8%V4g#@#4#e(Wl?B6U1{TE3o#v$>8}(buAm($)PsJ#B
zYF@dlZAQsuT28l08C0Ph37gaAv)_B{`nIQCaE!DF_5C+qW_*GGdjU~>0j%dG5T(c%
zgLH&Y)>QG5QG1>qYqML>gRiGv$LMMgET}!W%<6u2WOmB8WLUO5;E(LIr3|mixXrRF
zg#==jS<~!?p?^^(>D{)zSHgTU2Fs->hWg=0p646x7Ox@SV@&U2mkug@vdSJScPdq8
z#u(QTRz0n*`q#v7V~rh;+LjxVvD@*>X&vT+?$jrf6#|Q+#|e|w&Tkfo*`Ess@E*!h
zH)<Dyjg}{3SGm=6EWZ`(`rv3bbYosevR=Y-?e+ep2a2huUFw#9^-Vra#$`OdY4J^L
zRn04R>ifmEdRZ>_=;6eWO1#`nKpy*Vp&-Oqw;;`hS4_ftR0yUlDCY2Lwt#jq)d)ZS
zZ#Cp%U`%qYEg*mN9V)8cGA}N8U<0pvd!Gi9zpv(&e2_K*K1N_afm`BHU{Za#u02uD
z`d&|51#e6HmWhdL5qXx&B{^#~79%A!G1`B-gZ5nd1BP6agXMpx@VRE#RYc;XD<7xU
zPacC*0>6v*D3+&2!ZT_2RPAlOih`Z;lY4c;G~L<?a_aB=%JFG<a^@b^e3wdXQLF3$
zL%DU!`}B^G;i-p5!>jLF8OzT8$u<8Kj7hk)FTeJF$}=Z=gHg#<!GVd&k@lq9ifLEH
zZ@y7!_}u+w-=aabT=M33&bg8^y&fVXLSb#o*qgkcmZa(oZrp~JT|tu|$??@bgwbWJ
zPcY?|s9_w|LHLoEzCuMFSyt!PoV|t7rrt;9Z(6v1FsMf1JR)ORpQr`57$Aw4d?6UU
zU8Y_%L~=Y<PbGXC!PDp%P^TQ4NR9%6g&tF7ZWkE#WW&z5-oZ=n11_CRUH;)MvnE9L
z7;9?!i^^cH?-~m?=$h=?bV#36LEFRBH?jOI`-HV1@w>pSHfQ{gi_!b8ub=;z$ST(^
z2+MH(qddFsf-`WLiM32<sG&dSz69TgRMcXWk63pBf-U|K=pWSX@(B_Q6UIxlAd-Wl
zz~seO1iUOuZ-@*)D&A_ufR-Lcq@+OGsS@r84wdEB8nH+Awroyma=)$Io=^bAsQBgY
zF>kV<*gUystqbT2^{(joB{xn=X^`F$tuL?nJXC8k!v96--7Ptk>&dfcAiob!V(fO+
zoUfeEHe3l&8*6W0(FwHUKQr*mGxE#evPW^RM+PXQ1g<X697g1tkdjvt-`#PzLSxT4
zP}9!9rF+3lHrnQJ0pukqiDwZ)O%Zd*9lJwhAN&hRXKBEbXluS{+%oo4vD4jZ5dH2b
zHpuY$IYaL85J}9O;y85bK@QEROSkvSS39?X_2=7NcMNG?KXQgq_tn|ou)Hq3BcPDb
zy5>+^AYP$3x?Q3vePT&6C%JYx-AeCgQ<_EX;8$;Rqx7_a_M_*e`k!m<lNx?l&PGpO
z7TYuNbKGnDYwWRzx$L;|as9H(f~CYeFNW<iqkkJ3A8fMtUcGv=X_C)7>;{v!ig(Ab
zmB;6h#z2Xlg~JR-uJ4Bofvj~wb?i03=n8IuxMCIy1XBYYr$c<HR~4f`V2Pp<D$$a2
z2JUT5{D8~Mi?(qBI&v!p?-R_RObihi;uj%_^AAk*giwM-uLJZY`$E8-3(8pc9$Fy$
zz1UxX21v<^v@Pl_S+rkD6Xd=*uTE%?d7fA!w3vSM-4!;o*B;ark7p>p8Q<9zdqtfe
z|IsnrDla8<XDnOGl3_OKA7h-pIYT{;kQXW8-&PoIrRA?#*8bRvOW5c}!tg6uCFR<p
z_&aOKyF1gQ)y88(ZXML}bvWf2QlV=6St_z;Rg*K>)~74ij_}6${%wcz)tkSI%GBqc
zGr<V7%8ptVmq+l_!{p}#aF}};S>LD#?Wz-fziGnWcihN6D6qd};s^zy^3Q=J0<7sg
z?h_jf<sgbD2M#J&CN|LOb&>KoTtnf&|6~cbIbH=^ozd@=?=egA`?B6xVZY5ozWY&b
zz0)_iD)NU)xR!)(Pt9EV=AN^je#?kG=p}uf%KsN-hAq8p;GIc3U!7SuH4&f95L#c6
zEmxkd{&Onq+{<^~fhjgS4@*}xp2yrvJ$o}fm$sly?|AsDFk!UNzc#F<K`MssV{Dz^
zZk-r4d-85=--yQ?iy$GWYiVzG?3XzO>xIv2G<ME9dsL;Ir!cM1qVpiNGkH>lr)WJo
z<KkZFuorXk5~kcAk8X~(s+fd+qS@0iA&LsTb-{+bZmB0^)TF|C<Yx-3$6|1`eLoNY
zhgOg_F7~H^-Rv8=qk1QAqL%`Tmr9D1$2>rJ(kJLGboxGn>$sa}v#313s0my^K!c-H
z?%mOgcSG>W{O^2K{aSDjj^U%UGqdOB*C{sH#1j|Kg{P#go?Q&1*~ZShZOgp=WPxZR
z^`c<^RZXj#Y2xe2gqi#t4?XXB!SJ|MF5X#fo55>->aI7@E!8_Ns;5nHmC{ePXRjYu
zw=<VHhwcB&k?bEhvX2q<Zc}fH%OQN9@T+f<b2jYr5i0(>9$7F!NA#nC-upu@L(bq0
zgj%dpl9dZpL7>JONGl6~W(x4(rk9-N55JCCHhjVNf;1=bI!@Uf6iCzH(g}!&V{m@G
zT+{s^2@lJ<Rjff+E{EUEVVwkaw74J?QnN>ngr&6nW`<@!@UJ&&yHJ0H%e3&=>a$rn
zUWcbETX!vJo(SI3U1$6_<h#h}`w{ovmW(Z8XdIF`S!a4Rt?a`q_@13p3w4?c(Dn46
zEsB31V9WmJ#Ps+x)w<I)A<3tR=05t{m)@^$4zNfa?vc`1CK~AWSlFI_+d(hw5xier
zqv@~glcbAoY-64J?3(7q3n0_lw@|6I?N1x5ZYugxEzz0lfXn0f-Zl9@P{3PW>^vd`
z2A8e^@Ofgs^QwE7s}auUInzL-@oFQhQOx=c;)!|z&w1_RKmfOJfRZJ49+csaKze7s
zr_ca|e4*;7$ZcU^0crnSZDR7_I@S|2^h^0E&D-AGW9M}=2UE3Nx@{%&G5XDl#UJyf
zM6+_#Ej3<Wx1Y2Oa5$iD`0DpZ=EcI+Aj^==Nqf}Jvq1N7|8(z*>)JLU+a_-1$4#c>
z`^m0VXAiqt#i|wd->i+s?R%Zs#UJn>rg@-z4XrRdwwHr{J(|84?yM8ubx5IT+)}Y2
zg7~#y$4!p!R|R<>$>neXt?(gK5^7w5ptr!oH@33mL!mUwK*udy_D?Z=k;4A^E~uK`
zDrWr%mX|~`4KTAnMS;PHNmxYi%cFaty9JJ2tw1P&g}F^<2zO>U`0?>M?+u;bcQ?#w
zyFtSJdLy)Sj#jqIQ+e-82WOQq=jvIjye}Z4OfOdJCJnTiaX;>FPLcr#wdZc3*mne`
zWV~Hq{@Y__n{h6pzKimYo1ASiV;Vd5+yjbHxP}?o07aF9Q>}X9xD{j4R6(QX9we4#
z!+4ifX>8DHNrIFof{%jwTnw~6M9rIu+@0v*B6?qB%A>zRJK;M7FlQR>%nH<;uc<F>
z8nU8UJ-+L@e>t6dtTsGl{M1;OY*FxLhm!?X#Ok8<!n~Cu57lj?HR_%nm#WtfxTtaM
z^1G&l`-Li+_IBoBEQzX-@e5J|?;8`wJ+Li~$`5zCOP$_(JE$f<nVX3R5Z=5{UZz3n
zg%DBa4Y^f=Z$@Y86s__-q!5m0sE9g3WesdqW`yswYLNsuMAK~m^EDnQ?QsS>MLyiS
zY4ZeFu;F55%u#u(P&#MVEN>{qU;Y)pz5B;K)S2=l<*o1DA9uP$3`;yzJTm4JyC~C9
zC#N(jAo*o*sFZK<-OJWfc8|Y@_+u&A2^Vp-(8gqY<f^zdwT0~|9)i|C_sTXBBrJp;
zWQFC~=;H{hk9U?!DUMzCe8~Ul>T-?IZV3}Z+U~1c!aBtO=a|f8xh4M4506UF=E&Up
z6I2Xu6(eoJrXm0W1a}XDnQ08jo?7Z5yU3UiXSasB0|~^k;{d1o*vrU`)WgxrC!#K?
zCY{jjv4o5y5F{%j)Z)j3J<-MEa_<-W9<P78*xL}RxDxHD?}k0Ipc7Fce3srp*I-y}
z@_zobls#W`X~s^hqWW-yJac(_*0Vn*Gm}vyrMmYwTNmR^{oW=ys(GY}pFLd>l{cSP
zBwsll{(R55uSG64aRu2|#FLWZ75_AzV+ZM&j^}?=YLMt|EER^a#sFnH7wmYz_=}kL
zA@IXIZbX*d0jqpMxhF>?)og@nBGF=mXF}grpI$4(cy_{;nf<@j2A&2kZ}|#>L<rW!
zEielG2Fg{bz-1Vu43R`FFc?C^@HardGpLuK%o;3~013$4=vOp`RP3Uy^stO^s`o@2
zV>wv$_P~iA#av$JTCM=qg7~)m0&Gvcq`=RMr1j@-#f(yHK5n&X$fPy=u)HbW<Kory
zWM~mYC~CI%osG;7q3wy13ff|v??aW&OD})RGge*ZSmI-)-`+-R6-<7@;J;0XDeI05
zZOC#;2^KZfUfwdk*b5+=-)o2aaF=4%5`onSx+~$P1{24-dqLUMF>;)BQ*z?i?6-e`
zP}Lx7lmq$mT&~x$9Mu|v?vF*7f+awY=0j6Owadx+KY(2gIRjYwqZ>7CQGGxuGI&**
zXNuzv6gpgx6_9KW3f>hP5>Ncnw)8pv#lQpcr=ENyW3!TTKku&i-aB7gb}hDhcrw~e
z?G8P-Z+@To+}HIS?JT2Voit|1uZP7S9vr^sv`a~4;KfA$-N<_mkMCam@_zfyd+ywV
z9HIPkr&G~Ozr~=>6|`BQqyd?;)=Tf}ouG+5D<wwxW2?}O`zJ^@UWN!6)N6>e1&g)k
zU{*6;mkQx|__|FUj)cU^6v2I5Bz5Yc$@hb^z;O&ujip|di@`n40uVaDdn7;t?!R7;
zjgUy>1Mlf%RlN!kpPTLl@>S=i_xT!i+RvnFxpJFyID`{36pKZNd2Po0wtbLg?y5Ht
zuU@~|jX%-*7xl|wQONwC6A^v;mSjK$GgBx=;tRMhetBaY(jl}@9=$A-CkrTEf$ggk
z9+p%fwv>aSiy`{X!3Bdh(Vhy!>w@A*U+gV@-IwZx{Z(!a`zshEXU(BWVC_vYGb!Va
z@#y_env?qTX4CTvoz#V`S3GZM+^W2udotkM`%fd?`Ds3(RU!urlBHb@JNBhdn%8)$
zbY}V<B_t#q>xujpON3^qn560WIu#HG2Ox4Lf{e{To*sHo5WXOaeY~^Nx4aJ|#3w+r
zj|L~;VL{%YPvn`^=JNc>qsmR0ib+B_q6ldRBDW>c1FfceOD_4w$a5LLdw{l&sbq;1
zmlqX=3I0Win*b9fucO5Ygb&^au*4(;7pBp14^pQHWD0n=-0!dLm|%bwK~%w|AC8hU
zkFY?hlE)>FUK2Y}uwU*T1e=hN&kY1vE7CP1_I(Fqy?H6ggGar#QCw(sT{vSrgo;(r
zgquw0o5o9+LV#gc4J7AA?koPW(x*I#o`udJ6$b<|i~a@m87Rlvg$C9`1pzW#<-oTY
zyI(!Z!Kp!zC%FQit^$h*1bAARH8YR^6j)=Wg)bu>tz9=`TV}`J&*?1xTp*%wO1MVb
zH`g+Mn;J&JF0$6<F^lt(V}X?NxOaY*4R^I$^|OQvlAAqe=$J$!lS0j;PNUm?(dLaw
z52dOcwr}ywPpT-qpLEIO`|J2>`OMn>Lnnumu0L)|)Eq1(kJ*Uph}Z?u<0cDrcf@|J
z9b!tw?sM$aPdpfCwPjIqFd^VMa)N;E;kypu?7o{r1r55LqH&k<A{{~SJO=pfmyoRd
z9ITmTsXDnR2V@R)0Ap_wbjGngBf)geFkWUz4rxs#Knd%=yk7Dtq!$UHJU4o@@)sp`
zGv2irnEm40CZ%+}x(Y3XH6M(g_jbxC+UF4_@$|K@s&yGtm8V22gKf!v{5CyPOEMu^
z<exp5T1$SzuIU+zisZ5KDO&yw^Ly6SqQpa4%!UhBm=7p{+SRTk@i>uyoRbNKc$$zk
zor$F@VCsdfv;6Jjm;%kWQlaXZ0Yx^J$k%%h;Y8&Dtx8i&=sduP^i0E{ITjwv#^ckO
zahS4N_2iVMK!c%r<>)s}e1)>gRee|tIPe#N3{XJF)k0zK4Rokv8ue7|5G)XR_poG<
z_tlvHF1tKsnNQBz&chklJ7*G2j~L$Hlbx7D*{mgEm`!?i*lNIMzBoQ!SJA*XIu5Je
zP54rWBTyDenP$bIeA=hfx#%Y(H+AXA-xW$RJt@k#C9N-v2~4mbA`~(oC^>kMjDG-c
zQXqqi;t(771{$1wpbLaTp)>^}LVi$$Ag%2)$IP-)pm~!BZNQALrof&6r5$f?pCBB?
z+~F02Ho3aTjSK2ADnY~xhi5hW07lgLkoISV<+smzcX=FsI3e&11?ymKdg(P#RNWv4
z%RGEgZauy#lL$cV+#YjFep2YAH|}{=ApeC6j8XoT0oP-aSJ%#b9Y|N0BEyLYeCCdv
z?bq7{@+5{^y5@_SlO|RPj!<!X8hAEm55@kjWwojAp@_ezy;JkL50J<vSj6T(iIAYZ
z-SQyj6-okNEVSbBI*faQeoZ}V{`y;?80ZYaaAAZFfX-Z^Xg!cUZ5w|`V-)KhZn=b^
zS<y4);N~f0!qp3a+$K<u=4<=68an<}L3$L3$G^}(djcwpZ=e{Z2#?*(G`<ZNwEyyI
zkmZ?Yp`oBZ`wcouCUYmB{ZvI8jYDfBUM@L4!?sKvd>U1&+JZ6ZCaqyw%Y6B7!dLUu
zmig0Bpf7udUq9opX+R0Enx3LwFkIQ&28nB{#0^4hYz!_DCTHVmhopgxfV1dy1=qG+
zD!)sDmB*X`PzOKFOLIJuFjfju3L90~$qqgBsH+o-c^B&5wW>~1{&Y_>6ltAS$~Dfy
z_=%(K6vO0;ou?jJNo5pHnz5j|)xz7gb54vAnmbfglo55wG!TuYBALUucY*bdDvq17
zN0)P-4cb5%*?21)bwCkiybJ2Kw~8SH7Kpgdd7y<r{3cxELSPGn7zaW;^)^00Yh>Pf
z#E)27dWANbVEK>AB~|HDMT$A1juyRF(Q?04qrRQCeF(+WYfl5ef<9qt@AXfQCao9?
zmQTLjg9ew)`P=7YyxJ1=a|r==ri(1my3M}~DG}XHX%a15sSw9;%XH8w0Ahtov>^qw
zY;J+52gsE;rOjIoOI)iz+3jNzn|8kFMn#~G<xywoMmrU~T}<qPyTzMz#b?1#aZ$ZB
z0V&TiLChv3iu$0XthLO<zMUwxI04;ip}u|7L4sr<JSIs>woze^)w(ZP@CyhLI6;f9
zvzuiYg*kCGAs-59x8>3OdSgj4Sx7twN<2t_CzNGrZ&iU6({AJ&vruJ8orBW}Cic6l
zCnsNlP>ybaP!eeVK{toUM<6?wH|8Nh-5B^@m_K+m)_o@h=b*yQb9FB$sd1Z=qS-D>
z2HYPCJ42{Y0$|w0*9(Y2Ro~&JP?r4zrytN>^iSW95~0o&p+{08&pNJMxkx?(1g>5E
zSswep?9|XQO)aTi0`-wJxFdcmxv2QnwqV`$-3)#`%wwTp)AD@~QAkA)fnq`q)<A}2
zK7Leh&P0M?`&KL%B5U5bUvTR-ss<JZBr~I=Vo97!JYCrrDs$u7=?jzrXu{`x62N&C
zn*W;gd_zQ6b613EE`ax-v=`t%nS$k#^60V?Y2jw9z$?a6-@les*QomxAH7hfellxT
z(I+RewB%E3L0Hyqz<NG5ibpypK$qQ+0v)MXh-PzdB-jB_c^wuSy<FVW5h7o39?(*h
zxwnN2f&hcuA?QkzXc_ChI{?telss-kEw~&y)7=t!ciOq_$wKUo0U<imz?cb0<Jx?+
zn+u{CUAJKQ=EbaJkItqhP9Hv*6(aC5ZlpSI&vnKZDaEgLO73m=0+wpGWuesDn{I**
z6C&u`z{T5d%Rf2IKh>YrRLqnLE4TEst@Rg}QYgRF`ir@RK2wxt6<L+%rF#?Y_I~0)
zh4pKnWMQr@GF8v;&h<38ACGhPwFwQCjg24k#&xB6PfXV6N(YBiI+D~g)^N`{O%*2O
za%^|IsdP34wm_HA9BV%|$9Ak<1Zdj+Xzq(yy1zXQgLDx{$t@Vhbz5Ls02~3uF_1)t
zwzv;^k=i{<XzSi+Zq>IFO4ulCRzgYA!6mVmKO(jQ*vmT@Ht07hZ*{RoRIP@<9tg@8
z;QA+&BlxgxSy!(#PCCJob;SRQiWDv<r@Ah>C5lOWm|OB-bDRCd@-})OefZ%d`oPE`
zv2!{xMY0ZmOyn12vO8bZ1fTp8aiz#{qQx-%hv(KKpASzb+Fn+!_d55bb<btBo12SY
zE(^T~5EPf+Icc&bo3c`A^)4ZPXH84N*jn71Os`VL`AHiu!v)?~f`3f1aq*H}Br2^{
z<@6^z_xrh1Iyn`+Ije@cf3l^NC4<BtKFbP=t%x&@k-|#f13vbi3wWm9qr>JhUO@Qw
z^UTR|hcI9HeR(@}jYrvjNSh57g3w6<`?>U|#m{_8ILDBH%mMJCReG&KOmx9nr~z=j
z2eQ#h>Yia7p9!>J#H?@4c#wPD*WEG*^3#y^&!vBDZn?<4B5k099hkRO!#jWR9Z0w+
zCNwWZt+FbcHDC-!2?gL70`9o!n@$Y{)`x>k8Bd>|TM0{S;rS6H<V(9z=xjT%IAZ*`
zGo|DTR^Mo%XRt7r%Y-=?XOyEOa=vOsr}|Y`!KZ<|{$i_vx+eKbty2-wsh#tyP2F4y
zdAoA53#hnBf?=lo#J5n4x?67Cc7?&9ZA!zng{H3>_XeP9bi+K8@tP|=qx|`&9zQX%
z&bfm5U2x=tn#6Rl!96VVQ`nj7>+h3ghjgk*A<4bc-Nk7O0#_qq*!_+M--4?Kj^=~c
z6<Scw`e)-1`#1jU5k$sF<jcJ%K4cI@p5v#Lsb?$N;m=!^0s7Ij)8Q4cA<QLS2%}kd
zc6F)VIc-+tXCf~wrM5ZqV3I^jU}(KVn260G<8D*FC40VcpRqxC8XqypTM{L5FO=_X
z#=$A8-Ge8O{}3s7(IvS)f5t%Y;whQjociWk>qk<AoSA~e2>))Y7eQvj9W4K4<98*X
zt@h}^?<J+vO#-pJ9=(1c%)h8QflU)>B++qy{qy%}5<`OK#HP)GcOz?AMq_qIqa_>Y
z6c6=uiF4fzq%*GLuEmEQ27gO+4N}z!@UPx&Zc^i;t`>Wuwnjf$!RQcH<Z+tHK&p=A
zju{6b`AppeOV`XoT$Pvgl9cjFzu&h?8HZ?9?i<e{Ce}qx9v;d4?1qI(#F?*(!_ZF-
zlI6V2iC|xZJ>@GDjD`>XM5;nDLJk0lV}*-B%UYqTWmm#8;2@lc-4YL^;}{IbeN#=P
zY)pkbpM`nu-2}6Nf0Ml+1rvqvuED7qSl@Jg={o?mRBTHEV_27rI|OQ4Wj^gP{UWo6
z_VVicb`?{X^44>rr{2l6`RsRI%`3rc)MlDEX}t>a%1~6Y3(4M6cCJ1{dH#v5>*Kt>
z6!%x@N&6qmyVY#<eJS2FqHCBPCfl^*)9Lg5K-f}tXJD})!b@Yb^837jIGdyA3?+#P
z8DG$W&>qe3V{hBCu$Lp@a)xJt=CG?{r!n@2@A|y8pRv-yw9f?j-&Y;`R)~jg<QT7d
z_m{sslUbD|d{k{ufV}7x1&hWVWsHaQ;`*6|!}~@9J9;m_Ykr^iID($C{_)L{&T4iD
z<?<-sk{!+7TG4#@J{KncTY;eb^hGV-CHz>qAjScyaS$|Gw6jdZm0%d<^#CGm8Bh^4
z@$O>hVMRi)<>MIWlHxs?gsI?bfwR}cfQi|MHgqs;^hCddSTgN(CN}CH<uN0Q@&}>$
zt+m^p`eV7$FZjPZFA~=66dHF<-t@dbhT_$4)cG`9&`WZqV}R1N>nVTtj8qCyDg;)g
zWJdWPH4(yxNuS8wV^7pyCT;IC96NlXP`7^ivSjnczK|b9;RzzE^X9}iE>5%B9;YX^
z-Vc$C-q)u2;fD6?sd3-a6X_a?mxV>Ov1SrXi{X}b#fsXMUD-*+3H{A82S-@dvk4Wx
zx=Ee;Q}ar8u76pjIsc^Aw*O1u3&8xGYYQ_k94{k^cSar3Ud@p@qoTShANrJ6{P`a0
zc$`JOI3u}qZ2E<t?FFi=Y%6(Qf^Y(8_@F{~RWjEMG=2C!_BLq)WM%?$3CSUS^mrC~
zDM!}=ws|9;j(38kn*`43dgX5Y!p>*+iZlyP`KLUx@Vcy|{J~V@I06y`W_;T+mEg_!
z9^Bpl2ZOdDp}cD=*?HI6UQ|4eCVEXfipeD%D%CoxuNFKQye-o}!1syVq(CAwJI_bf
zyYqcNWv}N-qsN~cP0>V8R{?@}NcMwn(+i2omU%?u30+>3k@2(NcPNJ+w@BBP7~7i6
z6hEUGw{4-yBC_hn{M|{}BDIs{It!TRZr5Xy?i-EoK7ZncnX%om=$*d`_gwZeZx{7U
z?E8ya#<-P?GI*N4V$>ffh>#k-mX6)J=vM6{abBTz``6L@<I<Tu4a2EpFT@j0>)#IF
z+V-iX%DiKz%dU#rcWuJ{^1Jfo^*5)RR=6mQ^ej1!K_@wvvcmG|I<8sGv3n1rdlBIQ
zMEgDd66AX337Tap5fH9H5a_|J<)Nfe1*z&IlsdzBWx|4%YL1^g$#R;;0B#afNkZp%
zJ1rk0UYcSOm0FU`ml_Ctkf2hhr?SATt~q7@!pnDeIY?`m!`(;>s)SOhNO=pmjKZO%
zM#YNN9OHMR+@`^KPu-gI`g{+wP&ZfGw~D5G)llxXcANmq=f_}0oY8aHW0{ON9Vzp_
zsC7wu8!L+kvzo4-xAW!?FHkIV7o~E#kKO9wv~W|k=qwHo79M}~Dvp1*@UA|^&NQ^e
zSm1P$bFez+wwMRiF8Kpl!YoyW-uu>XR^hQ#g-46OE|)zDRxuyU4FoHpII-jmjV2II
zK{G1?7Z1C&LKISt9FuYBXmtQQxb|yv<iaij`drbS_|)-ZdtONHE#F*`HQ^|9-NiJ}
zXQEx@n~k<u3>CN~%;jEx*6Q)vawFeVOtA&71(4HowD^4)`jnZFESv{5R#<ni@IZY8
zLbx+4kjszu7!TgN$IHx8;&tvn%v;sg+a9)<_ZK(`CziLXj9%8f)e&AK{I<P(S$L=Z
z(ZW5vGZX#d7~O4(TUW|Wv_YW!ay^foUuf2TvuGoMne$`Y_R+ISemvRv=*+&tX3L;D
zBfHQcU7q=yw)c~L{B<`K8C}_%UAfnmcFNM}J#aNN#|}-D#}z4CKy)5|tKI{2ML~0i
zijb4eB0L3I@)F7wfiePHdLf2OjugLt(`}RDrh)nZ)7stz#joaoZj*58%_`waj1xJ^
zTlBq^ghZxj`-X>2jq`qH&Xks`rdMjygpy5{pg<O_1bYs8e>Q?w0Ba&dUo4UQ-@3;8
zeEXO5*B)(27Hhy=^wfLsb~;!=*#9Hjf8g!#`>&pLPp5qziXLjeUjJD6H&@@GkmT*{
z=Su|V!>h#Hb<ExN3di}@iVqlN<Zn%=h>XOBXjw%%Uk=8oOrIgnS(?mPHv2F}!;}Kq
z6-44Y&@f9JVkU{#fP^LR^79DA^l>AQ-AurP(ij-+mLOk%=t}KnY({7#0Iny~=5-;g
zdgRzuLCF)zKwL5j_5LW4l0eo(wY6`GuZcayQ?v*!JO$dj(-ijgy(vwk+w6(&Ap!<t
z<ARa%4>0sN0e1d(Pt>i#U?L;cW3G@@WbSa@X_@icTw)?5?@q+!llKJ|q$KPx^3#Sg
zJXNnV_g<Mu$r8jFAAG4lk(gqqAmSE#Z>l$e5Hg}DucI$rUH(y2t!@#rp`}3w17BMz
zj3@W3U4GYiY@8q_a~-I2Ujz6IaMi&2*R-YUW$Jh69YqK-5c&x!zpPijRe^G(evDg3
z-PK&hUeGU4|MyFf;KgC8wOpR|#JxuI>gfQv>yH9jHn>qwx#T`CKr?qn0ioqh%foaD
zO+{0^ePU^uNOiAO<ixf;9-Hs1ZGlwf@#eKczfj+PQ^|~r+AESPFRYZxLTIYBrG)fs
z+uMdcqp}Spg4beaEk>)NhTfN22)tg6w9meZ*3WJgIsV7T(u|Qsov#jm+n9X+9{D|2
z_Kow$m0KuFOp1MbPpb~lVX+rfc;kM8^M6~gWdqWVvv~tK%OmlSd<V>IrMh+ExiDEv
z(A?@!`dT<<=)lu-0d!ZR@csLs1G&#Yax5`s4yut9rp?P+ATb5`2TLRFI+{#^wHfcM
zuEHAN$r+L~gIfzskg=)oUvW|>GQuGQ1i(Foi_ouID^cw*Yg>_7G4nW?dAD8;%dY>6
zqMYh+T&*|4e7;_{7^E|Gf={B`>PO|1LCimBfl<$$pUcoS4y)}3=#nYF(?!kianEj;
zwc)<`-hUTq+_RREXkDI?@L$FAs-&XgS0_bo7Y&2}ZW%{`9w`6-8@=2-4s5@nzHYp~
ziBalV6f9JWL{cY1j3NVR$fXGis5y|cKV2?T49L6WpU@zb^@`LRyTwgn*-E$h7DHyA
zJJa~!B3wAChORz$_+e=MG<<UtCABHgMX0PLV<6b>q{Ex=Y(>U{SiXt6Ad?=$m2&}p
z#2r_>Pj&}{34ItinlRjwr%^Z`F|P4j5|e)wW*-2{*7awR6$KehaBk2&Sv3h)Rc(*W
zfZ{9<4i+hZCQ=~B$lnjdBPPnxFMJ&*t-1}X%}uy)v=gLDSul^xNk{@jK`D$i(NQ>T
zS3e?AjP3`LCg?^WxToD}+*=Mf!nK<I;q^nmxyE{9N{@z^y=aVKos(0as2DFFzMb8$
z&Fz(+6OY<atYW1|o4RGkQu1%+nMT>tL)Y)#YbWKC!czm+b2H|wZWE)jo^uYIlKIp2
zbMuph9gjHc>n1|tmvsN4WY-(H`$2&8D0lE);C-AZKw1@`v*Ub$P?ob5Hubrod(=<G
zT(f~W%*@?qNSCbR<Wap@bHW+O>v>ClGM54Nv>SC?y;0i1qNR-E@C$k^Pv@mg6*JG;
z>~)3GKkSu0YNKfJw;aBo-*>4%asiNguxNc_`gpd)R|C*5u*V@}*~-?EcQR4yB}-Z@
zpyFEou)Nd3R<k4T-5u^vm6<Wik7CC~+&^PfE#Jk_<o6|}`sNh<ct)2wSYe-$y3_8P
zk*QD!pM;gC@>;~oJVtV(?Z<#i&@}nfAP{TEKq3JWRlD7>My73ZL3R%4F$2sP^J$M)
z-8YIE6y2di{V8wgjS8-->Oe?iBcmpG0WuP=p|s5t>g*R_!7J_GhDjK6coyU4+{owJ
zZv!bisLVkI!ijf-N5!VAt82y6Xfu-=7M-uPx|H2SzdmW}%Bw?1m-*HY{jo(9gejDS
zl5)rY>SM+8%7)4nUlbNqW^)Au#nU9mA4SIl4V{B&leRsGW2O!-D&|A!Okrul{um$w
zc3dnG`Leon0kUub?KA~w3n@jKb)42j>mezmbtF^ILcLDaNo2`J$LX6TWY~-sCvbJ3
zNcufES0S%%QkA-YzZLxkCDQCgp=@0!(#v8|wC6Aa!oT(#mw6&}B=_Qv!e`Y6E=J>_
zx#OvrOQyb{stvo-VirN}XYO4(4jifb<#|n(e&65{0bwg(JS>`D!)%lD8W;?Q%yX9$
z6p?m*`8t<8H~L3;<>vUJ;w`1lr}I~<Wd!tJ{@cBgK8LRy2QBH^{O^f9nqh+mxdiPc
zK{!=mE{uS5$L$~-#Dp`78(=70Ad>O<uY)oQDp_ztpc%+o%$|ZZ1k_MgEG<*u=7A1i
z<YZlXsB+xH(wwJ0V4MYjP=pNzsQ{tx0yT&>08s$R5m&Mw7;opiE4>BH9YkIpRUDpE
zF!23gfQS<oc7*^3Nuaq!Ke`A<v#HxCTtIt-Op!cj2+3L@$w|m)b7C-OfA{Z$4vuAL
zBL~zyqSyC#2*tmN1I3BjghHW&rkbeHXYtD9Y83I08Cp(=D2GwgjyeEYLewU5CEioP
z_%G@&>JNQcXNZK$#_Qm*Y$Baa!j`17Bj{}$MCHqU<RUg1$0pPJ;Qf%q#inAT;Qx=)
zNJsyoR&j&MPz%hoq~S}lEh@;B-B>C^t@6-rDoUVo2rq|1$))<CqWsbsG;2IDo$W^#
zAz$-*7qa%Qh0Gx*ec*8d`c~8dR4O(F#UUr+%|#<$8&wf6`_CpkZV1dlz7_r*n~bHW
zk0L)9M|(>9Q2*bmjh};6DTrz5qrXT;YX*7PphvNqBg!W0IFVbO{wVwLu>C5JdsiN`
zqvKUL=q$G35c!GHlj=-5UI~n%4tiH#t0X7+(QT}`q^z}5eVz$9tRKg+Ic;6Pd21*5
zWzCP%o10;r<_KUD4jBge|J%eyY#PIRX5v@`Z1cuQv6ZNJlrX%@j*3^J|3$5kPSseW
z2G#zeYHN-wB`IR*L;RHtC2A*kB?ATpn?9&iO&=n`rxLSCII0f=FV4VHT}`kN>uhp`
zxHY+w;fv)&DTFl*wVJ~iEU*J=<ZM5n{!CO}sR652$vQAn@O@wv8D2_7^bM;{^6ebN
zvA~2^$o#Qk(V;kzcfl-YMEpfD6zSHSc=F6+9o7NyXmZIQk;;bGPGLM5#**4YruS0G
zC4FQ%3y-D23rr<ruymbPvQD)Y?IdT{Nd>kX?l_DDeArYoJE9Yn@J;{fNxB&n_5Z-&
z4<X<5Kch%s6qpX;GU%<00K_OZk@i2MkfE~HaV4)tOiewcTf_GUEa9Ao2->&hXfO`s
zbUd+=ZLOl?;upnD?^f}%l_Tj;O>is{Jv$tajYIp&IkApwzfg^`#vYfxjQ75Vvc{bY
zuf)UH%v|H@nKb@R5`Mn1reGL`@O1b(!p2D9m5Sj#8&++xw2hcUtnx#w;z~iGQs8X|
zcw7V6VN+A8bZZp)Y(Rz1kv>upjf|xosvd=zIuM0dDiX$*93o(|$kl9;&JejMn_M(B
zM`-?*se&SAtdeWmY>8CDoD&0&isDBJE2SJDz+lB;*`?V*D%vb$q_W`)NtOQ@1rLgs
zFRodIzsK_GxkmgV!FD9V7!D#PA=85A>egWgOfB!!Ie<c|20YDZX9vLa;^xRDR6-FA
zuahGAXq8OQCKR#pICBm<l}L55Y|Ej)XOq%l*c-TnM_XO7=^mG<116|QPgDwjHPw$D
zv5-!~Hjo)ytOb}QWW(=9z6^$mm<|>x)vRue(k2!=VD9V`eGu%Z&g7zQn3RH-V<RRt
ztf5ex){5jO*>727>=0rMy$`RQO@>zqi%{K1)?wl4@Uhv%6p^zVTh>~zSy=mB#2-AC
zQ(hDv4Ta+3<dD0@o6c6lkt<4EIOG!fS6Pc23x|wDzIDSUa0hs$GMKt`hMWk?l?>Yi
zOk$Br;2nrb$ohr75HJOAL2Ob9Xso$FBWqKUwOpO|=fA1araAnflC=M#$R%u;Ogx=N
z#tjkZRDu$CkH0ADfhUu47{ByqCOpm7axfm8sLCO06bg+M@~eDy$Wsf(X=A7u|8Evd
z6<&U-d3BwEx(P!KR@DrG11n)6b0=(zSk+0zM@)fBK*8D~!;*2X`lizd-wZQ0Ix~RB
zu}RI%1T2fNuacnC=T_Xz(1w9tAw+QyaO><D_=n_3xlO)uDJY!~G*>FN(i)A$^Kwzq
zJou?qdRhA1C}NT~vclm7v*!o1hn@mt_zGI?veSPf#Q>X%V4T1vydU`TE-GH*4}7#7
zmTE#$d(th@JcvcSk;hBeD?|7q_%j<%@77-f2M6=XLN>OoS`iA5e9IKeW2#mBxEg2o
zNrh`@u!QTHis&`84t00l)L#_7`VAL#oxRYvVGq2Y3?@<8)_i1SIn8n?&$)oX$n=3x
z`2RNwIbdGX@ZiKNBfzKwc%`SWtiZrObT<6ERrrs+n@x%UXUih2mt_QG<83e430vcF
zSup8%9XN}q2Ub}$G6{>uVzXgEm~ajfu!F=rVm3aC&=6Kbs=-n#-wwi2hHP^5hFQe)
zA(U$X<bC$=8d8T~L$Je|@JitK{x?!s6dwDMif^u|jKHFV51_Gw-$p(nPf<zGrs0dS
zBZyffEWG;gxL3_#;00(z@CDUWcq%*=e992sR9N^pHYEz^=kTb5N|nbscub4){L`tn
zY7J9OBjqW%)VB;+7}rp8$mGFDk?AgA(mF|wV=bMHSQMItS6e0h&mwCq3N2?x#g~Bz
zY{VoEZ0g`BeN=Y^W)1m1*n-+rhcz}{d&t94t)e>X7i<wY#=%>FWw`gRgnfa&4JE{(
zld5*-xQdRoc{={|;A2iMAs>1V?3|f@i-g{Ga>7zYVN*D;*<g_x9R`WKLXr~%x8MqY
z5lF==u}N51-^f10!tW=j_{K(<PS`&jSlX+%e^I04s<Y%NDjQqGMprWY{-Pp=NU#I%
zYT$Jz?_&qRE0E!Fi{Od#rlL^r8Jv80u4^dNHET{dv4qvTNlHx^-D`18tC#;|?*UJ|
zVNMEa!y>T37O_YnTuz*Ai?{(8Jg_JeO!@B@rwt3Q)|d_v57n9j%N}iGy?0$_Ox3K0
zRk5|%Q|oMe2`zv@tfs0GiiAnh#)N7%zPg40!(Yw9a~`-R#ETV@(-wgP0nHVKitQ#=
z+7g@D#?+YwuECXR;pUV7vxfr*6NbEOZ2zoSD-X)?u<>Az4i(weh&^C9@{7}@53g1V
z_FPM`M#)J5cxoLu*P-`e9bW?936n!37qO|*ewD6sB<(B*+biHSNpQ5l=j&vRavs2P
z+R?#-3vob9@HmPCi}%G2V$*H0=>ry(o$1gf+aQN5BMxi|8V4h?VUlauD!z$<5+A}V
z)7cyQWMk4yHVm@nbZ6N{AjfP9GLf!-Q4l)80{DyaqvEv@=Z(w`E;ao)ct%@1w-Sq_
z<VP;*bB$nm!@;+OO1Q==M<qP&_7>u_OYaqUjN(vmB0lCkKsQmUbo8c2DiY+E!6p<K
zi;Tv9HgSSipNqJRN2L$p1Hh;u{N+`W4j8}zYZ-;YbGonsjw3!ZMHohIYcp3pdkV2>
z2E1Dzd;}i$7M%f;L3QQB>-3SH43TSAN%SG`6D*QWIss*kjmDPXJ-MQ-QLv!4zz6ZB
zP_41j#7ZY?9HS4<!GN#>Rv8vv*>I!~>8`?HQ}b#i9@e=nxtZ+)zyC#r>GXo_8%FVR
z4fQ!O)Tj(4It!0NaY<Vt{s@M+5;;I{Yy#MXy<IsNzToPF$7RFbfH@6uUHFS4_J!$W
zpM~X3g$UU%g@eo`%zsVj0M0}$C6%Dly_?>P!g5Fx_z;`+E3MjbGO_7u^g(jEi#XyT
z{|F;Cxx!Z=?h@P-1Q8j><v8SU9CA^3l=UrGPnAgcLPDYVrTeScQwxKL<H;t|hw#`e
zybimPP%~%KZ5_WthAD&5E6HF5`9K|<5oL|qR5qnOw2ne!qj+ODc;&8{RCi<R@k-Tj
zi$Lq(qrfH(^K_P7EG>lIpU&Q$?vlX@O#g!tCg9$)XTTfJtb<wLpppL#T|<+eMq*30
z0FQbs{s1b<j>?9EonIE0e%6nLj1~O=&v9PxiTG+7K?l4(MT-%^KnAW4Uk!%~_=!Y1
zneH$%k0<Yx8zkg%n)#(1ND<~h3893$i7}jPELwN@y)_u+sl&3nW_w)DFn!3SvX?&R
z)%-_zuj@>+2YK!<>Yxl8In^TmqTEuPNZ0{X5sh@$`{jTcA;pR6ikCy7W$TCV^xIhK
z5V=l`j(0`2S#fo4H{58};grhuV{zaHuPkF(WbHme(U5J#FIVK~o?0b5-o!;U{#Kx~
zIt1<tjbq4hxf130{*SKrfp6lv^F>EOP%DJS5m~yFw>N^0L})F@QlotK<|f#L^lA`O
zx^&CdXcLoGa+son?q*Z6ga`?T%?-__6(cHwl>$wL*|eddvN*cIK}z!SwkR$)soWZ*
z6EnHHG%Xtx?E@mk9whGfjO>u!eUA~+Xf$)q_x%3<e&@`1{T3?OV{OVH3kv#w{nkf`
zLQ}h+)_US|j^8b%x_w*32lssznDot8lxP`abv~*!u0rw4J}G)mWeOlAe)~lw6H31}
z@J>>c-Ek!Yi=7*LM2e2r*^#|syi2*3lptz@Md?My(9t@pWG4E?^u}IGT&(Bop?<+0
zyMHLyW>JHOJzwtyAlvNM-aH`2)^~NDN)!%ifo!$&Nl$w2Ln%}5cQ&v^3D(H#S8kHc
zqp~wDUOkKK;_1mX#T_MU+Dq(0+xyyv>%z_fWl;6~JSiF8W5Sh6?`Z0gArRF*<FSgH
z$tk${J2%QHRVJn7><jDqUSH7HCa4{X$hJfmT4I)0u9@bQme{+l?%M!JZNk32;yboY
zCr;6K{*(4m{@oSNcm{plbh!8N&+DSb8ZbU52L9%zQQaS3{lu=e#itYAJ$CQ?)JF~r
zvw!quC>>=HJdk$qAS4^a;MVGc@O5E4lO(B_MU@xzo+MWh)oI$c0VzB*@3qC1Q?>Db
zLG<?UV~uR&!+jRvF)1&m?-LWJn_5xgxzS(!dQo}fx+5>CJrCS|iZ<wJKHSXcj^KJd
zbx0>T!Zz=Yr`L>n9!+}Y>~fEsTe95+NXLbnl0Ri~XK@QjdoL9^d;a`BDV1emI__kW
zctlKh%}&vhYkwt`l<5FjrY@IuUh6m8`vr4B+;`e4-ooS~wakXv7)-FIxT9!=&$YZC
z6W_1}9mmpd|7QLxS<U^rEB&uq$RV&}u4b+ywJ4?7<^;<`<H0A`>gRVYhmM~JK*7>x
zb9vGSZiIsS<@ZMwF|N27QSA0c?V^ZK1YSl6J!;Uc6IKfmX43XV`Vy^CuPDHpn1(ja
zEQXg|+M|mI357iazV*;H!M^;Q#r*l^k>63vs|&wW)9CASIzLdHRNJ~Hav-mhW<wke
z-u|2zQ#aq-wK^9SXYah23eZwe^?qmP<8#TNO76<Z@2l``s3yp+ChIGb6!3ujo=_hw
zL0o|sfF+pQ)OKcx^YwmVq1{Z>8rz<_bkWbqsRP0@;_f{W@A_`waio$KbzateEPdP6
z14DFO=Jfo3`sXc9%C_F8%+FsJLI{LMJ@ZPUjd`4Hrsl6KZQDvtnHEIV_pJ?&Y*b*N
zQ)vWkTdJ28ItIT{FM#8s3TZ&_B)N--oFq78eC_uvh@78Dowgq^S;hy}A@>{<$+n{M
z&O;*9<Uo1HDqPg%a2G{&|IoLd&18P};-WO<rziKe>}cZV-ZiBK98z*MePi={&2oFv
zpE>srm&k5mSxNlIG%jWjvwyWA0g;?#(pd!0q`^@Iyy+Z;{Z<B{drWaB4Ro;RNoi%l
zIbN^Wq3^<?UpDm*2i6^5Vcy0biJ9D`ZXy1d{su?%84<Vg97-Jj+td-^KqoB>o-N(`
z{xWA-|M<a)pX_8Lsi>@L?#VN>yy^XOvz=u3ZmmIga|*Da_IG~`;|EG8sZl9~W3SDD
zeMU+LhH<%$E64?28kLSD*A2pWNzQe2JnGrq-e!}2cPP?)bFRl+{%uSM9JWbx)zWbT
z&zDpC#YB1R&A&bTi_kYxwKKFRpIEabwd?N3XR#IcLUgw;XYZMFAVSx+0Xa)75ZsK#
zz!NC_)8`wArbN($zsuR6D-$Qkjt0wc`(zLq)c(?`xD1==9@w%1RB!CRE^IdKp7D^y
z)@?^X)DG{z#2#OkfL}|CzbJAF&8y^7&kl_JP5kiPEB`i=m=IOPy>WBq^xfBmnXGdu
zBYC+q_nPpyH2BWU20i1rLvXSU8GsNGG5qN8gKQgbhI1#Q-k&M=y%kr)`a}jwHF`#5
zJ~t$2R2J3yQ>ubkf+sKU-DF9{x&@vv31UBrbk=!tK4F%IwtDPE$*y);<}W?`!)D^m
z#A`Ed|LbK*sJZgUz~;`RZ`p6z`|l$(BiWiDkRL5-Gc{`33c<2hvYiG7f5I2|(F;kE
zQDx{jI-D!IUIwF7;>b_fZ<0Eb@?cS>5FaKv2jor$->64U=zkxZY3cV^7VHwAA(>k}
zV!Pl(7H$)d2Je4ry<~yrtCGFF%QA1==~<U}`D8ewD$CP~JI-olUw9JQERiB2wuPN-
zmStDS3xcJ;4K`o(A=L}qb7C&Z3`x!;sU1?zoPhG{h_MPORYtf(<8Ump8&;9Y?sA02
z;SFnuCF<wSS)3asaaT~hNwOp4RRbxtn<qTa&3plomrbw$k{M-#ZnrO>zH%QlmC#zn
zJ;wcOFbzmV?Si#OP=wHJ=3$;zSfYZ8Bzz>Wa)pog+zDMT%b5ZAW(NYL47@Fq+$!lD
z=;agy2SGc8lVpgF<kh#@cyq~rvYUnW&wJ1o59m#b{iofX>ycXsmZ%j+EI`J{ag?yY
z0op8yb+C&eyd_k2zXlqCO?bIFyAbD!d=&h|FI5*1T7(&vK_(|8i9M7mDm%|BeMLFN
z$Y7kLl177Nz+yW{`WytOTj;k6s<enbBt`{GJEYs>ey#fp58d$!@I>1TJdzMYJWoJ7
z7a!7jz|hf9CC5mu)EO0oZUfIoc~Q4ulk5hb1S4Mv*ktD@p#uucAeb$xvY3?JjHJ%$
zS5<N(2`(xZklJ#qL{t-vJtXdZY@~8>wQ!$vy>*e=Z5Mdjy<8X#T{iR#2?EA$mMk%o
zz=9UFYfdK=yUap$cu5_e71=GvgZS7hsvG6wf?%`Sg$^(bJuE)DQ865=5~=GU1da@Z
zk_JJOR?Jy%0O6515GtHpOj&`uidGC&Txpbx6Tf=$v8K%*jdcmDy2NLiAGcdgKUsLo
z3m_7bF%C+M3BsO-R;$t+3>u&!U>aIR&!HOQXGI~)n^d2CSO5nFp(E|XZ%Qb;e);(v
zyA+pJfI($Bl~i~*o3GF&qbMw?lAGZ?LP4%FcW$m+Az!cb*u{aFwBXoxUAWu(QH@pe
z(1Fm4f3rW`LWKOz3PhGCy#^wKOyL@u(L-4sswO@sSYm=&_JB4ZlYkt~K{=Va!rA{c
zgxFO}6~{#|XMphjlQMb+c~WHsSa2NGC1+7u+$=n?UxxwkA#9?Im)>c~fw-Hkha_E{
zGx&~ps~De#WU*QKm8!2xl}yeg&H~$D8g*nMI?1Bof<#JCH%fLd;q(G52Sqwk_IRII
zu19zlEO)?2ayuz!5OtRoH%F*&(FVvk=~U&^*<v-GK8@lz+@3@IHWDSfC1zG0Hwkm=
zll|`qw~9kkfIOhU%{DXK1d@8e!U6e6jvH4#Bp|A!R4__NuaQLS9oXTBOl~QFpzfE!
z4;W5_iI(BKM8q6SP7Ppi5F81X(Os6gR86Fc088-He8je*+5=GxWl1XEc)5PzP{O;c
z<letpyb^r6`{ubF@#UO#7C?y!R&k&QSuFMikd$q~e-KIIc_qmeKt;P?a+U!LpD-JM
zWe1_&^8EMU3%xwXgtz4`jL`DB3LSUH!*|uWO_bc{Sp+T(%#y^!`H`jStcy0OBXJF+
zYE~LKR>?!wQ%13h*xQAUCtXPc74k_T1{qhGvxSNi3*(2%0ak54BRL~e1R4<G6I9)H
zZvayJldUh1YLgo$C*-FnMT*_mRC#b<3(2uyVx*Cl@^$^vWbkx6)W}I3!-sL=E=ng5
zkT<Ri4<k9;Y?g#}aKL02tQ{s-i&kO`sBE0BB9sfPTEJ0dxyAt+tYS_B2gNAPkU=G!
z85IPW{{;kB9)02kLV=<P0pGF}3@j5aN&{LXwg1|is<1{$6=lRqz}uN5h@{ZSvO1)K
zlTTNU7Cq+E7PH3^HK9${7ee1zE^+I`357m+w}51CLCV({scQxzURr`=iZ+EKvbJQ(
zINh_j?_UWVN$UhRRJN#eGG1z-liogvJ0NgjASqm4l(O8oGz%aWSkQz!WFSvYdT|z$
zl1g9GTyX83yLP!3^g@G3qj<ZSC?az^ChcH&jsg?aGAgJHOg;D&K??}PsSJ~5!U0@+
ztMez4dsGpVk~1mO%Mw<*UK|`yoTIXP0LDp%{5v6LXiAkK{}56*-ezGXCsGBrrAAy-
z!{pvq;n#O9{uLzJ>NPF*`ME3^rC(7c6#aQ1!jM*SlsCmvh|97}RP`xTF_73EvreIk
zdcOR`X>bZw7J45M!K1S5RAu8*$qhwhP)Cx1Ur|Yokp|ToCuoUQ?uk1dV@jlCpWzo`
zqP^W?!iB$Z4K4Q?G9oFwyEyV13;GjGt_&kp-e?I{5K`q&LOC!@6$TK&gaBV1g{v_k
zDExBc`GOc%p!l#w^~q8v>!DCjCi8a#Sw`(mp>>QFg{mg$Jg7BGR?%j;XOAUzzo5=M
zj@b&_4=#ykr7yd(>m@O2g%?wRD8&(pQ4VP)>XcQCh?X$EJZ0D7*M&=vZH@?SukP=%
zM?DrpeLvUy=dE*tst;#FL>g3b7G;=Cu*uCS3}CQS2ou0KZc%H$*_}iS9ZSTdC7QpV
z)XtzB)5ss6=cliASujVi0S?Xzr>gkT6A(p3L+~w-B*QAfpcs%UfD|-@0+5_cguIzK
zu|Us1r~N40Cm<l#g%pb5MKV~1)t1C0R1kWG;E6a@l8U3nNneN-z9R|opk=M*OAX&o
z2gpZxEpmAsNk#j+1e?h|<Bl2I0LtV~pulE;qA?E2++|5Hg`U@gXxhL;;Lw5tvKxX5
zOp@5&E!f(?Lx5W7yooL_gqEYapep%Ml1Ul_%IlG$4B~eVNS$aN%8|qlsn#xDJ+DKZ
zE0f&2Ybad-uC%<8vY>!2NcO*cY}*?g7e*wJG#?#q+KWM-I4-I4JPSU<ab;Va2@Xhz
z79prkFAp<gpfPkoc-Y^+oiK8u$@uknoz<UU+r^}uN&=tkWXiVlu(%vwhHZf<&`^V}
zQ1;!+$_jjg4)V7O4PCP&6_DRza(`ug3l(O)R1Sc0Q)EaolW2kk3)W6htHv$59#j=T
zN~wy|7ofmnpfa1j?R?9J;K9LlQ*zeMq@ZQ&foxjJly$`0U@IUGOPxtF2o+Q%cUkHj
ztAfSHKw`LyLY|gQM~-8B+{(;39^J912uTO27&mDEHCQk}c`T||q6fh?X%3>CN>1mf
zz-7wKpFGF{n79-`Q(!ZFJM*@eWEDpZiXhzuU}Y&hEHxnZRY1B{?5!lpUDt&(xK@lS
zVp;CIZ>ue-Aoeh{8;-Gd+O{=jLkzU`)u^yfSRz8OqBqHqZ4vqlR4w862xc$l7xbXy
z5?^z0y=rY_7^f0&1|bmLW#_msL{m@dm#~NM*m;SQOGRyR$A^gc5DMVTl@zry(wWdc
zq<S_UNl`X<sB%h`Gin6Elm-WeOn-<Po%mMu*ad_Imbm}6v4R+#jq4Q{8$g5J#jI#v
z==YKgt_n&PI32Vwy*2XVI60M6+{{7aem=v`!R^TWbLc|S=Lf1XNew6>dM@`iR9~o8
z7*x-`uS$c^;@xS4hX<6EFK5rBZJ-z=9(<@^w6r~h-YQ5P3<kauIKBZSP~p)HamD)-
zFfq2k0Oq#>m@mR%8R+2HdHg_NtlbbJC;%?t$RgrzA@sPju?$d>H~_#%w0xng7H%>!
zIWuFGSE$>EBuk2s6)z>+DsQNXO8doSe2)j+)hp<WkzvK1?7hU4R~CvbaOf^G0NUXd
zG^i%Vz)%+1WD>-IUVb&0#Xq>k<Yy^(=Qf)M@ZhVj^S8dvgRC<ROx%`qmP^j!j-40M
zr`w>OWN;Lb%N2<tZJ?|nrrk{jW>0U%ZbOY!uH4d~R|q{bcu0n~_=6l9<CK%p6&#My
zEjMSaV403aHt~FBx{({76@F%tpAVx1=M**B1tCKB1|fqikn{Cnx&Ssy^<5Z19O`hm
zasb!7><b)<Ov)*AK+BlStkj_MoM?EWZfmQ91RBg+18Gkvi3Z3L*EGT;5OR|67p#ZX
zu1YRGnXpN49m#3aEnIHxoH0&>m_npK0r+=~SH9Xq(gpvat!Kwq49lI|Nz0uB^2)S4
z+>xjjnLD|zk=Lq~oc0pDx9-rD23CdeCM!*?!gi1vlv*d1xP++s%&;{x#;HeAY4oAE
zr?J6enSC;6$R-mxW$@Inh{hWfa0(kY78sKsL7`xNjK1~&tPnyXKL5i+?NXVDqcSae
zw?Y(?gbwufjWtIehN-I|>dL1VtY24LJ){Kx$LyxY<z-^Y%Q>xrOZ$+)B??R0EF4>$
z9O2mD_~k(kZ>`Ynp`VviW%ex-`5B7u*#<ob;x##_8Y6D-hM%{I2Sv8;3acLoU*Ks_
z!LMyqB{xD0Ra$vnDEjgYp5s(b3+J^^P2;xi!-0`EtoVa*5Z}~rOWyKePrk7@UvM(d
zytw3!D?e$mGmgki=CGJdM_~bEM7+*}p<rBZ?pb$t4S=!}$MzrM(VVl4ws0nZC5aoc
zxCfl?U3Bx#f<s-ilw}|#Wt1tEO7_evjzgsE#S9;0s;Sj##_UNL<jJOvFJ-T*J0|)V
zY3h8ORQHc@6E`j%s<@-p#dmaW0JV~2*-(o+KxP<+`-pQa6ih~{ZW5{M;WV&hdHc1(
zGUF3Hfn`93%DUi@MD?$QwGVm&G%_i54n)Y+Jv_>1oMhJpPzIhC=lk3m&+jqH<~o?P
zN;Rif@)skp1^hIS)956_O0YjiYudFkvKj*)la>u}WO$)~kb*;>O*+r+*g8roHJCIf
zzdS7=W#L<`G+kV%aE&vr5>`oA`SZ2=xAotz5)9x3r%{D_+jt0MVZmXTMZJO+LOC*I
z2%A>rjrrpMiXt$&-Z^%rlILIEkE!6ufEbgAKRa>QdM8LIYEp#vq&VS>0dZ9693{?Q
z?g++}@MwdJXXtWI9SV*X_}muXE~erVygx1yY#g5LHya8h&4M39nu&rd-#0+;_aGqv
z>q-=qsW*LK!c~=1kRgAsjbV*hB+_CwKu(;h>62>&fhnRI6=o<vv}nnlWB3Zw+>9Hv
zOOjJuN*VGz<-hztdNAoD#oI$!9km=x+ZovMdL&yA;|l18=`Ao8udh2}=P?)HGf0su
zpQ&rL0cCMr8XVr*(;7+^6NwNvMlSn86lC|IJDv}M(@kJ6xi#W?ITKBwF+uQdN8<~Z
z-1g5*aWzBlBSJYcOtPFX`SBw<fnp5i^{pCuyFA0t`bu4cd2GOzVhO^ys|La;i&+E~
ziQ{vtI}T%EyPtN^YQ2__hqt_6<}#d_+&VJ!K=#TOH^c_!PQ`4JauhZM0#}7Cqo>h2
z+eRySr1wdwQx)3HGK|{7C7qy}D;xPJe<q?6s1?^{r8BKK^4jk}ER6KBaP$*5199Te
zDs>|%tf1n)vPG0hj*b@fbm{v(qQXAiRot<aJ1?Y>8ZC!vqcSY97?<;9C6ko1In6MX
z6}*$ngn%G&merav;O7Pq2B8>0I^)uhyMzt-f@?T(6+OS)M1@$7fKjlocV)>6s)KF_
zN`+ZXRKv6u5*Twx$8{lBLDc|s<C_j2i>BXp9{7I2kuEq{C_f>VcU%`nd5s7Ch{z3+
zJjdbn_z)!sOhv#2F(PRe$zs+BLW35hj(8C3OfXmMhUxBhZ^)O3u2+B}N0ntc{fYe{
zI#%6QQz&F&C`_`i3BAC=wViL*YUAOozLHk?)Zz5j1nCH6Bb|GH?sL$QECUTFgi=L;
zm55lxa8joGjYc9xt`s<(zmRl-n;0PE1sq8tw~G^}D)zhf#Z(j+w`96(V|gywdgG+t
zkyBRU)ZbyW$&syB<hwAS+S`@vIpID%{0W#wRtx3XJ-~-1Z)}hk1vR47xu5Z47Y$1Y
z!~_CF)g3`BAUP><5WFuZD2Zxpl4p;uCTgbq5(5V8A|4!Gu-Fj4c%uDG0)?q(-dPY2
z7b4FbiY#k}^G*Fx$ocROcIwa=R;^VlR9_@SZcq%RbC`q%7}cojktv{sPR`)1;&VT%
zG2%bVFj$gFf{5G=hS>uv=*U)}h4Te*G1GuH>J7u{VIa_4G<MQk&FWny5#?V~2rh~2
zT;J>Hn%6sxubK3%tV23DR&kWL^VOY~<;8VagHRZw6oq9ss0JYn<+YED!Pi-YU((1G
zIM<S%s=fy<kx&iY!zRm_xIz&|{5UiLf*@jtl+giiFp0v<bz!@h3T8KtgV))K)Tn-0
zGNZI>>$W__dNB<P?XF474GJ1`7~>t`DnwV2cm&odGAUz0>m@Zra_rHrTdKxcH6zkN
z@4(0}09L3uhlj<jmsVbR7CuI)P!8AK3K^jd0}{SE%QE7X6dXE};j>XaF{lTl8=;WU
zibX}O!f5`wux&IH(kh|lTON_szpm??`39X0E64Q-)+Jwv3@DB|>l?*tO+DCUH7Hd<
z3y=sF0avP^cmvTI0AiZrAR~G(a{1W|ZwtX^MwRgDoK^`ZBM?b3t}L8Qpr6iiaL?nB
zj)aCNvor1sryn_!V;{0Dm7IoH1wpS6m!1sl??qQ(gAfqkDTuu19Q!@`&N9iB*E_F3
zY}nT4=5Tqd7!>Aa!GIgiX`mr21yX5fz5e+7>YkNovoDT{8kXlk-axyS1>!rjC;{l;
zD2PbJ0cFELV|lU{<JCe<h6d{--Fq#6ExJmw%dvh>n<+5Vi`EEw;2RL)Ez;EC#|E!H
zOGl-9_Oqjx8NQOyC5{|cC9rT!tJZW5vytPc<I+gE1aN5nzSMzZijze?vg#oKNOQI}
z5o_<3I<(Ayyj`SnRu3L5q{=%aCE8iz1_5CQixja~N3R%M%$7LZMvWb1c;3w1b)SOQ
z0@-5SS#+1xX_+(*>dSIK2<0g*S&`@QWFxDiIap~VYxrNuWB(v3`^ND8Z@Xy^6BDfa
z-gD-3$wl$<2GdeULUNrf5Va?e51o`!aC2$9wPU!!6_{UGLl+t(D^%Bmv^043Os%|j
z2J^)i-m3i(LQZF@1=TGm8!EI&!uz#jNA)n^T{>rXHnmVq3uAl$r-d+HPWcw~a{z+A
zn!h}3on3Gyx{|0-CPsoUI!7%}R;-H_NiWRa?%WFjM2M|mN@K7XX~dNbuQTf;+o;fn
zi__<lHgdIm(TSy-RO{39A;*ed1D|e~7Cr#FE`sU)w31~?1matdw8&chrhWH*K3Ha`
z2l&ZZ{0Y1!f<chsy6{W8=S8p;g62*lASWVI2FDY}@^WfGc1Oh#e=HhM3rnodtna#z
zBq69i(;L}~q$br)8l^MmXh~d&VWX#<e5e1<TL7nwU$v>VW-*{7S*0=g05>HSeF|jb
zoTdG&9?I#E8m>jy1xOGVnE7UCzJc=ao@4wInn5z7%HT^p*YmnCq*~<*f!(oPbAI*^
z^*Yl3-gmLHs0_xHP`>0ICCh5CRoXSmS0g&pG688RyTD}pwJVL25geP|npy?ioc_k8
z48-GG2%WUH%Etodbf`8}G=TnT2qg@>s88mM_JI7CKJkl^IK0S8QRMJq8J?IZZSE*s
zIOyw83x3+;!h*B>|NBMhX!DFXKq_!tW#Y=A<^0IjEXt@bol@uA3S<h(r<hHDZeXQ&
zZ*8Cter!oV>j;65Ywck~pfFCDu-kA!1ls+{MG>xuxR5mQ3($hrq2}*}z0YRKo)Rb;
zOna>966Qd3DV)c8vBZQ!{k0=Kxpb<DYVegK&52jheHG$_o(RP+HMT9OxXX^i-J(qr
zurfV<2bWP!X>|DMZ5ks##<3Xov&zL1U&45RA&wXpA3tcHy0(Rx|JG#FKKzZRMvp~3
z%e@QT3xevI8pM`B$xm=1HMK1r^xV4rFJDqwM*@|BMk>!QzIe~c(92Yh0DnX$8P%hj
zEHj9p24ldTbgdXSq~lZl@{9hzx29gLLwjl(nLvEU&#Ci{r(_<?X>B@$0iAIr<vY!P
z9tr+@ge;Ok_!?BMEQv`(Z;;B1_ZSE%E4k7=K|WQ;f=|hT<I9?ju&w|S^Q%2~6=8pa
z?a|NjdjHJyDyu!|->Y%Z=!5tF|GzOXi*Fcu%jXY^YO)vKS?I<^pTM3P1}8Uf8=wDj
zSbEltFR0c6F?ImGC>e=0{~b}KI^x3IM%v5eMjZ*@GFl)*+!X#w722FONODVRFFdD2
z#q`(OELQSjFK-IqYX*3{fr9_%#ugir!Vib2MgTmHhalEk(htbm{|Ot?f-oPw(F@_B
zAEqvQL)?cVP?TZg5c#c8y%F9TsjFUx+C$RqrUY|fy*b?)$LiK#ynacwx?Ih*w%HQk
zSC*Bmf>j-r>{i#V2RLAZ@laNL=D2q0XpWobBUugo0%S#Zp(=6%WF8|!O9uIIt|9Nd
zod$hROdiXpecTRZ5J)4Ga+kdo%KJ4aqX%dS*20m9yN-LsBZK?V(&g$_H7H;$g7xPY
z=7JPS*;YJ5E_X$R&RNOLR@Y&JsyLBTH4EqI>*alNpds{WP?KXvpoH)&DOKSLuM0au
zITkUnarMBtVFLorqi@cOKZ6fnq*g(6;yhGYQS^n*vq-J%4XB6#MqG)KKP`P}mbc@_
zN3&<QCB4_;mjF;O+2=9x0#<Z*EV0?;2UL4hvd`~0-dn@ifUSxDbp}9_<2Ahs5>)D(
zt1M(l0p6-tWM?|WBRmYG41yT|NfV#|5EtrP7PGI-`Ze(JEreA&ob$+>9eGwePbF8b
zWOY^=hqi|jh+k~0%q;lj6ZyTNEW31FcvC0{yI=49VK!K7v#74j=TMvUBjIQhpK-)`
zx$vVcj`A`bU2d|sri06oV<$sLaokXhyjHm)VXZw%4}hD0{DX(NZ0Ug-;~)H>0WtsP
zzS#FkmPPc53~I-yNx<WZMK7Yg=9so6aOlH>>}R~N;W%|+cyW#W2x)%2>oX9hxMCNu
z(hD<8>vpR0petjOEDHkt09WDQIE(WKLwUVGOMx6Z%$*NZSOeBGr;|x<Us*=<JF+sr
zkrNC3@~wvJZft0V<X$tk|BPo!7!$@^sTDG+;#MVGkPzcL_#jd+SzEqM5CY~oK`4?-
zrZ@}Yv8?`K2@B3=o~VV$u;z!7@#ysqum7jL#^7;sYpbz)aLrcA(9x92Ryc(t^)Ti~
zXiefs8=23)-K<wzSMZK*YzEM)v96!-aAMd;i*o0v0wa`BWurO2DjfsC5S3unSX$1M
z@*vIS4QxxeCPN#YPYdm<^lbwc!7Q!NWdxd?v00vu?*S!HO!d@&5b`hy5p&}Tgp*DJ
ziV-B&w}&QSMWoLB2i0o_<@$7#bps#Y!`RT+Rn_|gg27)G1|-KW`^#m9^##si1YW+6
z$qr8c9H~KYC#B9Lq`pF9^L@vfp%0mq;R)@T7rO1{HdhdFOnJ-Ws}m!spZPwGk_XC$
zsjM_{!-p2W(hEjuAZw#8hfh}A<u38inG55&anzUo7~k|a7Uo>P8f8Cd=Jg1D+<QR<
zd5uLZa&3TIO>%=o_*hhFaas}&@e41g3?vQL|Ioi?NhsfNm*qv9#Xo~@!w?3GTYbvB
zlKmDRw1x{M-y*OihRx%`hcO%9v}&De?il{F%4O}*_}WhGVm4oX_V=u%Yy7Nf0YJ>k
zz}(+N<q#@W7s?OR;z~9>fuJbNNG81nU4!C^8$dY>0J)35zh9w|H}v1<*GAXqm0ZC|
zQ)bU+D#K`ZSRPDn7TPRNy3!meKe@FTR~ujr%C=G={!^o-@sI;>`Gzk86QB)<HaY)&
zz1jq5zI_@nTD{tCc~AoMw8YCrNl;ti3a^D4XNj2LepAV)>+H+UP&iT-tx|muH=mwI
zH~G@y$x!2!wb2=CGeueO^){$x-t5!>7ARWV6jm6>7FO(3>slqtLO?dX8NccvI|qyZ
zJlG?jz%`5)1$gin%!uCqxTMU&S>Y^hwWX-Y@syb7hq&m1N{+_E&f*ptTZb|O3#b!$
zmO8|(?S&sS$4@miIv~ws>moI+LiI;kmsk8L?9-=+JB_KvcL0o+W2!ZcM}AX|@sqI`
zcpFn+b%eE;Wt+YnBLH9xP3UO+yn6p5SRPuaVa0ExbqaQATyc->e;GUICR<hxEZXr(
zP*~tkj09pR!G~(4A7w9^o7^~X4|iP{%xcbEc0siu6pP;cvIc?}FsD2H<IgIPS9oPv
ztF(TYLJClPtKe%W&mz2WoH@NU@pDTl7-^nwU%P<?sE<377LjU2>0YMLxP-ebc7&KX
zAYm2Qxh1n0NqB>a!5pjPSW_K;pss`Sdax*~(#=uHF_<j4+1gU@&>^&AkdpT|C5f%B
zn_8s>nCo+h!<do~?0xFwDIBe1H{LmF{;h`QwdJeV_XMaH2#g*o!#%hce503~fLU<e
zx#=cL=GT_5yVTNVtcBe3qElVL2aNJ$lAF=@qT&)>tSTZa3PwkXSPqs*{&on-ClXt&
zf1+d@at{48wJPEQjvBw=Z3^@hsws`X0r$opPLCL%?_9mDfJ#Omu8d1=CT9w14{i^?
zkS=W&U-MYJ@6x~Q7#b|X7^5D`mSy#jO2YKoY?H=;T{x|zg*#Mf`5G=A9<M)nJ6g?8
z_Y;}k;h$XG@Cz(*1oM?Syt6Q3uRIhGG`zu{)-_YY)B2JHI)ek6-LiqXkQa<oq~1wI
z{6}?-$J$v)6xsxvWS^xXLno-Bcg6|Ox{Pb|NS0MF-fNf;e!9&T4+>9PR(Zw2?4`Y~
zAKb!;FpUzNYi)pp3mRFy@=>z-GRJ%7<<Tn7c9AUeaR6{@GmPGi0XL?QHnr&L4TxPz
zeP!k(PDaFM6D1SyAZCE?FJRToAPgP~8|BUYZOjEyMyOWFl?AI{>$2||#b4YKR&|SJ
zkL9>vm&gxhv~U^({RHMGAhP<Uk01^gk}yu!g+E;0@p~S}LJHZZ&IQ!*mM|G((?Aw*
zb*}#rUjVnp3G;DBLz_#4ae!qrvH*Ipgq;i`;CRU?xRJELu9Q|{YG)ms>F>5d{)(5d
z#5Xo4)-K$P)j6tUDUoCxGW#D%`UzHu07j9E;0djfYvpgeRI41V>7zTU4<dRph6e?1
z)EkaZnn0*#C|HdNGar#U)UiE66rA6$r8p@S5I2+D^60PaG`{*BCRn?Lq%{gd-P*h%
z7|~(0OI3u0SOA7{Z@;l{8%q~vi$AkHrriJwg8j!$*S~lWBL#+6bw}mh74IkXfxk1(
zoV~JjwjR!3Y{3O#x#@-UfVAfY%jV@gsgvm_=3YV<uzIUV%xvd+No?!7ExPLdWpzHE
zHbkh7C7=k|AO$e?oNsPb!c;vbyv$u#s@yHwJ}=J}U+xFV2~Su4Bg*L`r-oUT)2mr~
zHh*np?zxZZKr=LP^|96ofLd|j@+$WcMl(=odOo#R9xO^I4v6O*i?e@F?@w6$)=Y60
z7K*o9=VCY?Aj;t3jRemZNEl@@Tuo@fiA9W<ziG2zLLQiHU08|e@%ay_JNeIu3wggy
zgXPl@Tc1mDV?>?r=Er}Y{F}vf&k@e|D;EcigGK;kAm#3?k~fqwR6GBEX>F`<2Xbdr
zjY_U0c{Sm&x9vMU%NL;FD+F0`rQtbvisMKe<qFFON(-3b0$7W{?`d%1YgC9v*Q_a#
zLuv#TiLx+14KPE03{qFX8W6CmUp$gisyS`&p!P2(3y-wUGLRa16#rh+40qn3aA3cA
z<ZGL~exz23f<Nu?N#T!Fb1>n(bbP^P3;LHja8g1gs&oW96bt8&ijllwOD3GJiE6iq
zX!zk&3MMV;#UoUb>N*1^hSlmCs6i5FVXMzts}xj4Gy=5=)i7Fq!q(hKC~siH5Sj(d
zUw-OGcub7;<NtPGbLvOAk&dWYS~4g(7FQA_hdS@{@{PECHkdfKlr@>mBWtHdz03G?
z8RVuFZ!`umc0MQD1(ri%jg`;k&0g`)4YUSM-Ywr?7p8Wmh@G4MB?3}b>spn6uS_3o
zeU2Nbs~8f{;#ULpuimqKXC$Edt_!bIhUF<gc3eoaZmILf!mR=IE$b4>sgP2o4iBE9
z;v>`n)K@WzAYIE>29@nFvt~>Pm<2(VvB0B5PO0m3$N>lpvp9|524EiLafs8u7~=w5
zPC^}Ddif80R4suwem3wnYW$RUe{ugdub3{Sp6(J{YEe15YM%G=vU6Z<!I8MczT@ja
z)jQqe;!AXV!NgZb<xGuap#MpP!YZe)S3YirIRLz^rI!w<((Mql5Eq8*G$O-9LV|%o
zhfg&qZGgUU!l(Bz(Kh`x`^g?I;)BtF_}t~Sn>TvCOq$KIyaa~!$d%>lTqza*mK4m~
zb?KzfpDsCK2R37&DN;#c)>wg3y$^V5(Kak?<B-<`qc<VbmBi8ScrW}&l~zIzd9pGU
z&hp+Hb>0=<0KAbmq?TtsH#TEHTsQgldkBFc9rWz<8OAWH{2Lm2=RM2Djfu$uEqWF=
zxXmqY7<QzKqk^p{eV5Z~QZOrS?QQ>Vpi(1<AygE*O0)o%YcWf%6r|Z3jAHHeu8va2
zkJ2s3pm`Y3U<CD6uLcAoR|B2Vt#Nn?V(W|s)o>h#0JybfjNj2rOO6b|`c3iX2Yt2k
z?BOEA1mOjSx>j*G7~fWu5f>bTEao5%(#jh1CookPgZ<*vjN`K-=CP=Li`v#6#m(p>
zC+xa;WYb^tqnyUA55JQ4+GG*_fkU7?|NIoOr*ZMm3e{ScxMqOF%&YcC`5f2UkH59{
zK_;pt7%tQ_P-hw&+p|=w!}}%))yA<{8&KWoTPrNiBpdXidz1{a?T9Bk%0j%0-#+to
zyU>pN%IT$k>-;F78lZ3iUfg&_Kq~@iHo_|~>(dCQ_jR1hXA&Y!T#`4|?r=ii*lPT`
zkKeQGENya29bdY<O2$ey5chJ=GDe08sKC3J-fD;<u-ap9JQ7sHxyoQ&&$0(VJ=nB%
z?sYA{Si(gopP!uHc2CTT#NI$lrs-tx*GL)-^8S(Os<t`-qxtYCCK&P28l4sfbb|k!
zMWN=kO*6-Q!StFL-Ywr!kO7Fb^BbKcg!;=g&emsom8AMex+Gd&cnv1LA<{}Ji2#nI
ziX`n-I+K!9l~Qf8Fz3<%`|~kSm`9Y5#2Wa|w<^x%JyT%%<E9O;4UU!DIi0ft*f=LJ
zW`kgc9@bd3_V1VF^{51&dywtiLmPD^MK71W*oy|}yQ(^ZntTTkqKEWbt29Vn2Nnw;
zzQM#_bNqx#-n-?x(B~26l9cL6drZQu;!E6DOo;-wXe6&209)_VFe7{?0uYUy0TOy=
zV61VBO3UW*Q}_@8(R5i7^SVFXz=FgJQRmqr+hoHkn$p^*e1UxhSNg3kAr%uAx>UqB
zA*zH<YA~des+Wkc0GK$KgQc)h!4$=_B>K&N&qD;Tnj@+LAa{mb4mgk&3jT8?tB1H;
z7(~2bm2M;opE7&KSrE!^dB#`>%&DYtf%akZjO9qWCy-{C-&iQN$82~8%K}Fl1>}(>
z;ybwWla8lG<+QaOwa%CozRQkK0?SBIi#2~<I^NvbZvf>%0l049BU!D|0&2KTuIqqp
z3<sYq%=bb~oCEg=AQ8iE@PmBQL(SSI>v?{Bc<ZF(NPlx?+Dj6?fP4LVi7r^>4XWNi
zqST?z@7QOv;O5KxwfLr2<H|9tp5A5gfs%qHYPY~0ATPj_>3?>ed~&+&DI^@kC&VP+
zEy5riF+{G$cMXJ~S4(j(a4d!>w%Td3NUiwD%x$_KyhHSv#=20|b92;Wn=euyE)UX@
zYq!n11XY9ROf;Nt<k3<Zf|8a_RA`SwZ0{0m2m*j{=HaUYN`ZVY-fA0WhBWm`omF6S
zcuOpd1^}wXVNk4s@03l^9H&<RyQOPf<uR8Q!P#`|y;<Af-;tPfGDUntnD3$pR()Hg
zFLWG^@~i)_mN!EnatXZe+JT0WJ<g)?(gDG?dCAg&hujJ6Dpp!s1&gY9JHB_wAOP<l
zGi>M*Fi<SNE_|oS5yCI&QxI!97lIt)wK+q>s-I$S3uu8O2F8ECWu6o~+}|w<_k9)O
zx5{I3vinm>VHr&w`(3&RZV?bg6{mXfn1QVjnj999N@gPVdv;ul%bjK1%WtGsqjF)<
zPY+G8N1yoUP&miV0BoeB?YQ&yN*);j@M!=Y%IkdG5X=@nKsR=H27UJHaJTZq2DbM;
z>n@2=^PEIslMuU!43-tTUp{Q5a#LmgbiqLjF1pGr#$BpU5m$8$@#*mjB2$ET>Kc1{
zd|imIA&h&>xJw|uc0~Dp3tWVjmG9=a*R&C;{r1|xi`(x8nlQ9D4~}y!H_#dU)p3|p
zx`8@Ee#o%;ks3mo;=hEJBhx6j5>MV?cgAHb^OVFPpGXU0+8Y#-Th%pwq1|IQxzrKA
zI^)_8B1h6-$>Sb-bQSKhlT%P?Z!iXruC8U`Q!QMicJG#zC$}HVNuXNz{BYm$DImC#
z18W*zIgtP{46y^*H0N1}n;Sxa8Jm_6)`|};UGDR%!1qp3+BS+?K$^@?3wAY@kWXdF
zL=rXddF#9I0FgwGi&~dXo~ivN7SI~U9k@*aE_3s1qf%~E!mv_hkJRy$j;^Zt;YvQY
z{pI9Vs=a;c=Y1^>0#t3vt>_#frxdV(gV@P4C!k)8;4K9XbFrgik9>Ap32A#u?S!*U
z;*+~B%8--^!;-KI;^s<>;2rQqRoce3Vs*EFblTOY<&?U)y==@4<Gr}HB)*^nfe{Q>
z>%*a()=^`<uO>R*%M;)MTk$K`1)gVfr2O0)8^AK>+4CPP(;C&`Y7M6)wtasCvPf1)
zx$*Q3uY_#eT~rWy+4hJ@F<lT7JA$#fAq~Ez(4jW6SQE1SSREtRLwZoy<+4k5AubQ5
z?dAc(TtC;*VI|A>j0uc$v!;b}lf9RL1%%UJUI5KNXAm*MflYZ}K*5y{gYc0<p_&Fc
zdDOe1;9BnUSWv_Gg57Q_+EIHGtQJYcB8miv3sVq)UhGpP7-UQr>AszODh<M^UKF1e
za9`0L+`b1-SA^9I_K-mx>;kwDu;;YuHhuhxUK&HkkTea+OBFl(WnhCL);Y9DlRY+2
zey}-Cg2Q3%7?Yb;JtDc5C*{~&QT2(5lIsO5Xe*JQ<Bq2-SiZV7%?*evgSY^eY-O4p
zPx>nKOtx8hsbCkt<StYX!D_o;M*jE|WQT8AL=8Hrl?iCVYsDmsC=s|A$4~((d@($+
zp^^|RjK9ggJHSSoAoQ3XZlT^b1K22A=w`u0iKLn1P=(4mIk32<H7=;g(K#xdfS=aO
zhiX+WEifQw(CQhd<Vxc{o>h{&3`*P8RUU~P-IC|%n%W13-Q`K(U7WLxOriwJ0H5Jf
zfo<x=$u@;fe{`UPQ+2|0q(V^U%U*X6p1;*a3mATSqsNIGMD-p1BsV|}iW*58D_vku
zz~GUhybR2W3?X8wBR7Ofhuj`ayt`6yzu*yUo>@}xNmx5h98GUF8u^iz4MpUK{84KN
z;PM<CN0v3f@l3wf@*5aHw2ZG%YJNV4m)AFa$tgMDnjliNN2GA}-nu?LFz7V=b++jP
zbP?i4m(>`m!_A_EqPyBH9W4up_{9O?z(T<+%^Gq2xCO4Jebg+4YZ@=S^Zxu%d$m^L
zt=7;;tNAH;gJ$G_3Gi*W_Z)QRbm7qWv42^FB_P}>u?}_BT9~>tj0oYkH`XVaf56+v
ze<Jv!DnUBJB{<LcT4$IH9&_eLQUyV1a}D{IkD!*+W$`+G3^@bsn(X1`_{;Tjm49n(
zepU8Ec7TQ>yz$~(1y#dYzY#SUS73~K2uh8QZ+$5tb*Mgn%-rSXwpcuL4G$~|LFjU6
z$pOxmmWcbW3jtmwD%tH7T!J8vo|oPKa2#t=k=#9%Tm!O8ir~(VGZ7^+z5ziPh6KHV
z1H=3X!~*e8A0UDu;Bt$(1}Q6Jtt0-}s$K&h1_{kDJluOr6Qze1c^!O2g&_A8;Cc86
z?q7zyslE!714q&?_Q`gWxz+bISe2qo3!=?r>4yoAZ(ttN6Ui-&Jn_12McbS>1|5L|
z1A<g>14E??U{sEC7zZJ6dWCn2X?Wvk7E$c&tqzQ9xln?|`WE;F9^3J99}99&SjUU!
zd%_YhOle)<Z4?lLiYY82C#nw*w&@&|Ql;BdiHXeqeX55)z7cfC%SvvAYbRM_p)(|8
zU<3~!9qs+_=_Nv!{T2u=Hx_wLT&;r6sb;8YmW5xjM+Fxo_0;<uAV>%rLGjnY#jAym
z1o8?APlgMsJ)G!VxI#7Na&cr#RM$%=jEerE;vBe`0l2v2E?A!Ivh;UZfGIi?`Ni4;
z<j*0F=s9s+aOE|UIfiZ;p{y2s@z2z@8Pz6M1WfQWte%f(LtCbxfrX7<h%sTwzpt~S
zwDqnT6hR~U>cmKZ7L3jrEN=j-5%f@9#{q>TN&)(6oKp;ty?tYnS;pGqyPC}XKP<c2
z;2ieF1fk!<AJ`{gTE+E?7Os86_teWhzk41jrT#!x-@4+J1NP5VU2*cT4if<<@!*>$
zAaa~MJHX&39?IM`C;b_X=1!jbeoIxi{~0a*1D*nTe71{Irohi;5e^tJOS-2cA?409
zIf24Fn8mtGb2}2HFDRJ<q7DARDlY0qPL5U)<=KT*%KYF7a0e9IAO8zF+57&!b#>e+
zRM&&|>TTOk41M>?!%LxOdHwBbZT#5hAQz!y3jnZ*VwKRleci{-MBY0E?BbswC``mj
z<^Y_$U*%?^qWU6eulkZ@wo5(jS0zWf2T{Ogc@PS>F4yQ(70f^48B@aPJMcUM6?IQV
zv~|s)Og`Yab-Mv9&u{%HZ{MnpXCvf9GN^oUU~}dRxBaUhz%GK7XO4_LK%T6`W`r&+
zXi(PN17x4u;wie%W-r2|7^M8hZnMX(3iC!HlvE%aHAV)cFm4P7Fa7z*m9n>kVo^~E
z53(j=D!wi=N}Rx@Dhv1UyO(jpG|<^3YbGxBh<~CRxU%fqTR)!NR|*{I4INFAH!eLX
z?OuG_3y(VM-{s6bRy;56OT}hzGeWgyfus5hYrN6O?ikY|I77cJ6*Dadp(H%z#$8;O
zg<qut*7yrfZ(jm2lO*yrlJFTLejzhf*>+w|#Zl;#4@dDs>jp$I7F=@Ny67v-Zx|vI
zdX*!qv2{?*Yx(LaMjje_#+q8WdS9MgfdJ||E?oX<&`sYl;R~yPc>D$r;WHPzj?fF%
z4`6MelRF+b9uv$AxQ<cDZ<$q=oTFbUpjzUfK9~*%@7o8uxl%Mqs=0Kb2GgZrHLqC4
zUAls~ONe&OW5Gnxv&gsbPYhh-P&$;yLu^^NHXNz3pgpm1?Y@_TJ_rb;12DG#Zs0yb
z`<`3KqmBxfXmrxG_=ClJmg>m~H_R?v)N{QV%sF+Z-<TbOBaC%BeL)xw3i6gf5W}_N
z-510nBI2T@6>8}HvMd(GwAo_z$o4iX#pDKpyCwF5<A>aDRpPS)gXf-c>>(O%xnO>i
zu*zil$v^SCr61!#N#oMNw=Da_=t6p))d3X5>hGGV_w9BcALiD!+5ES6Q_&P(dq<q#
zQvrv|lb8S^yb2;M0FIa_IMSs6EH7DEUw4Vgol)0elfB)nh`c;T=$Yu#-E-*Z+5?WX
z>X|>t#&2n!B%#a*Rwvi=tpPjMW&CCL35`#{@a;ok4`ewNy`9FTyFiv^o7gYRd|YtR
zv5S}TIv$zyzy4&v$BSEo1&}2`51@vGBnpph3MP<{mRxENLXRNUg+RRIP+mzTfj5X7
zTHSHtj0X!_q*T(QdaS8p?*g$O?Cx*7r{{l`NBp<-<a9WmtwHDK+bs7ZrmW==w=rKJ
zX0t+TdJ&Um!OY8T5YCxjly0SoB0HovavvYKO)WSQJq}J2iN91{^oB|GC)R(30|6XZ
zcpL|nULBCYKv)1#l_~7}u6fCz%L(yPCHH{duDrkz-JFpc%Y7$e-nv#nq856{{yjmw
zm5jZOJJjoEr4Ea&OHA$Ox*(sZ;eJu8&~A0TxXI$)fA-h=0qNQ^f0h{GM8XQ`baWSm
z;DA|P4j|!Kt8$|cTjy6X)8KUC*<4^2%HTXE%CO}KvH6XRNf1`DoAucXpHfYNZ6=kj
zW+ys^d<j;qGP20+TVkJ!8_06WiKC+*-)w;gO{tK6OSajFSuHk6+&shXE)TJT+^$~}
zzCpp49&pMKX)pxwGFLc20WyNsqZzA<e?{<AE28`D&axEF>+W(81$X?AN`gd|m^5d>
z!hTp%+dtkrvT}dq)XMSaS>=6S=-ad6OYFA>E?07gCSJZW(QU`%xTyBln0OavvtwNX
z5S_g5aOU0Q#OFAokK6x>_Mmg#q3UxbKody>(Gor^g$G6`9O{6aVpN&~!5j_ojFdmB
zuaRTU`ZzcUN9hJ%1yMEcuqrPUO?KO_?)?bGlxz97pVz~Ktn$U)8K%y;TPI2W#0P_;
ztGZzP#h6Wslq|TpCDvtcx7wp$6$gL#;?TW@{7>5c3bM)--sew9NHaLPG+_Dy;Grrx
zgFZjdsRdX11CO;j=?kSx%b=>0;ry^-h!xiCh*6Um^|U1&Q6Mrux8U0+Bcy%^31sL8
zECgxsI%3aY7=+f9Wn<?&qS<6%dPU4KISs!m2+7sYzf^qjBGvGa^RHTOOZYmlpjU;O
zVTlm%lw9B(Xr}fAmw`Mz*bjiO38}B*g5~F)5u&N#85bE<PC>3QPq9Y;KZG{>e3E*{
zY8Uq2+V<23q1+F*G{+ln^aP%s{Z^Z$e>KvL>%xnDMRS*kzygGF7F^J774Z)U<+0}u
zs=c_UQT?j-0f<bZ#|3k!`bc^N)h0jf5ff5S9f5%o)q<$@1O^NKT#uMHk~<@QXfgz)
zCxB^;3_|+~f*ESpWl4&ebC`B`{;s#aj~L}vxwG{4+2Km+slPYb-wMnmvsOv>G8KW2
z)e=?hcNIVe4k-q9dn~A;?H`Bnr0N>qe}U;qfN>0*a&aaiUM_X0U}&KO&gI1n!#pyq
z(u*Is4iQ)qzE${;8(0-kF6H8DQQ>JdjsPZU%9JI$@O*a5Q*`jcfd`I{uN&2ZF13V*
z{4D$o`AYFGx7piGu>EjL_PGUn45&O}j6czBfJgQQD9<x{(*;;(m&$E7q{)$Bt_0AH
zU<eP*NtN#uZccNr5or24F!c!^2|Pwr(MUg#N$S|GfCVprKKN&~C#^7pcHtns?^r(a
z(J!zv^zDlF;lp+5za9Vm9g71?FWa`+`|q&#ckz>LV4O!(yDd=M+2U1_)0ct|o$5D#
z_70?&%Qegpw2pC!riCTq9+K|r?qGIx{py5N!B0%@gRW-1T%LfAODzl43LR?T6j};I
z#i)Q{qJ58Ipqe$v+z9tx9eVrKlAKR&U7VJdAFe<20$H>kbvGow!p-0!XI|6;OdQ|r
zJNA>jj`D|=NxbX(Ar$TSE(e!6PLwuF)RTZwA0g?DIOtbPtGRUt;zC+l2@W=<A}~G>
z{xz$Inlca}&WUwflP>r`JlW^V7awDx+(&Q!uq4;@&`;-{US#?9eTfzR%k5AJByXDz
z$E>2wV~KU4ZLCBWUwn8!;qFJH^xc1ZQ#~iCnA4NQG?->vO?$9%i~tsYv0wt2I2B>U
zxLK=#$%wVVHHZNAj++c-UhWn<i!$nS2gqSYFcEP3oUC5SYua9hNP$IulKu`7AL;N6
z|7Dwi708zMwmW%b$c0Ocb-|pQwv%L-FM&jjv;OYnvBbHxs^nLhi%J0%UJt#2jhvZ|
zgwK)w(j4U4^Pma1JvYm1kTk@J1Yo#7m(aH>Io4OQ-w8IXvhVR5n1mUQr|+)T$HQQu
zwo4EF<ifupRt%<D!2D#F_;goB6*8QA9=iZcf+e<rg&jjxtR}DON`CU(cm9{hBqmA@
zP>OO7$ITM$aRC<WXrd4Vu^{$bFDF9j2*_ioV2uTaNoIImWdiqpihCSUsCjHuIzV=h
z_+x@~`Hyz*-?H>X;r4UkJ^A2kmOI)6i}>dMd>v7V-P{d1GKx}?*xtrfPejoq2R!u&
zKtl8HKK^t9qb;~pFNDXDD8Vd131Jx^{T-l|>LqE;Sdwn;SAe-?6u`zvXOTxPZf2pU
zEm2=EIzx7kDoIC*`D5*}?ZCPr^6FmSH$FadCK=phw`~^JTP^bu%|oI*mZ)HlcG+RF
zAWjPs8miIZ#kjeVg@?Jd8Q=a7X~~i1HUY$e=zwQ<?*QErkT6(M){&X(O8bI|f<MU-
zu}BV6ky5It(0X9GNfv7bXHiP2Yw?_~*{;JJ(#^}UnnSvDoAld1$p3JA<g>erLjS1U
zzE`jao({OVo7zO^Uq8&I`A`EXGuBr-M={_~u3<nrmhN%z-eJLkjSChOE&zC79~g(%
z;TViNQ2})&q-Y0mLU4(D6iMqAJU9Xm7Px?bGivE?lav14OiT+m+ehx%U%9Yv<oVOu
zM+?Gd|IjYccIohEtYSMDC#eIRN^k*o;j_TQMU+LBVZ1{;p@-3_jeh{G>qr=CrW(eI
zGfHwHwo7=qLPGlYse;8v623rsuS@mi8kwe6siQ5!Qju&8F3iG(R!|l=o+>%gT{CC(
zJeI8-ix+mE)?U4*I?JQpF&Gt}`mM?Gq+stisq;*S1^lzxt*8v;0E|t9vSIv<Dh#@c
z)fQnF>~(@c;W(uX0s#l11sC>#EDeboam2R-5<M=gTtIoQ1qPw?Xc%qqVTRq!g0UP+
zP?Pe4jzL`2b^qt8&ZhT-TKd)Q*HeG?p)l*YhX!A};|oK7Y7tDpWuMh$;kX20PEM;*
zt&-Js|9@@N3cSq<=1orc6Iw0-;&_9q&!KuRz=DFCX&*`mV!Gs_1%IOFGtB~mDi^d5
zP6qjZ4Z!)c(Hi)ZOTc*%kNHrgI|NnP)A2%qq}N-8BTfD7kI+M>I&xwAhpF8+g`ZM4
z-Dxt3w|0wZlQ^?2WfLq>!Q#qkd}NKR_V?THHycL7e5AlNf#<xATu4g+2p0v@KqIeT
zEu_>D_+}6;nApjBH=9*2B<peF;<HY8el|oNr}9Vh05@hi+(nP772K*kEOK0uV4s~|
z?7OR3xtE3OdgFkly;C)t=lvSN)M?a2Ec*(U{w^^pEUHittCfm20kEb8YOInUKPswL
zKLjgXa6uKR7e*8jFakeNCqUm~=NS|L3R}rmRb}>Li2{fN27;M70aa5fNvX1%DX=+M
zmymQlg?2n9yY=2%ckFCF@UV{BN@bVWe#<%Y8mOHGX29IWm)amehze%Hg}tVMIvOO3
zmu12^pemh!%z#x(frN((kn$dwsQ@Cb&Cj4$PCQz0<VvAu@z7vj-iQ;p&jF$7xQGW)
zPV!Sa@z-LI;4aoDo^cFf5Njm+f|Z}U2ub{4iyi{7cSK>`rS_*qq0NG<Ou#=LN!iWT
zxh-Hu-4DL00ZYnV7=-J)eom>4H`79p`a0%$5ny$QCO8){-Q!Bncco!1uo^myn-#x=
zbak7~?HqSDRXToGbA;dSgfKLYs@qx>!bJv4F4fa*-?MW0jxT;#U53+HB_;~DNOxKx
z<rcNuE&&$Yo=1oZc!U|Z4~?c4YK@gw#j#uZ9YMH|=J8ad+(XjRvdIz235QsK|MBBJ
zp$FB|mP3G1qGxH+&9ANxp|y(PdGHWe5ERgm+c23HCESWs5)<F-5=XN8ytaOR>psV2
zF@FWP>;R;8317oUw_rn5Wi@bto?v4LR;Q+c>jtb~#_r*&SMnnoR!2H$p+oI~Km{mb
zc9u#&^x0Aw(@@)kbeEIuAQWu?3eq$P21R8O55Pnw4|^IXOGp%Yve6?xL$7$3{@c1f
zoCx8v2wwq-L>SQaHlYi`#UoLKdOSem(pNW5!+xGKaD2vG&t6LcHq;y7ss;*4n}r1d
zIs!8af8(G&|L)8HdIx1lDuGtCMuL+G7dJ3Hl!}wVQ6eTe9}!eN<2oz>RGau@v$TS;
zDsmenV*;L+Trvxms36$7TcAV(lo*1DQy69rilal(iZf&WdU;pSCUKjE;uwVoDlJ@I
zfGLNO2&hoZf<tr{Q9L={6HueF>Psa<P(r?iMcm$YX_U}nID@A~sF*<}^IcB^l@F0g
z*BmjSk+F$T{i59@M%%v0g$l1O0so?cU<NO{MPa9oaq>9|P_zaQtAl;x&sbC+4)_y3
z5h&<Lhgx!QD+(^a_rYfL1eb_$<m(*9g+w%`Vc{uhYE?Y>IjH%E$Y4>XhBrv5^4+eN
z5XiCsYLENoo1Sz1<HzexTkR&er2p$P?^tXmyAAt-1pxyf*}6sZP9p~kXAMg6Mp-C7
z{><v9SP+C)VW_-*ZXO_4l1nWF5<XYj=cjLMTt~}zh>Z4Vl4z6K2f}~uc{ovh36h^H
zh@*sSUct&2gB9gr$3OXklzIGrI`gR>5*u0j+}kq`+uQ80`&gkXnC#IWAzD}nA|b@H
znM5J712N&@S?t5-r2tIMPLK;!vU)59m?*@W9SPv^i3zVr6>bF}=jo!AeiP3Dh)+1N
zq(GIEF7?+y;K&+|JwhvJBU_YUK#?}!_hG~^L@g~j{M<~N^sDCz|M<MzEiCewqYLe?
zFF77Fb+*rMQ|I3wZ37dHe><`E)80ReRp;z;Wit90nj7C^f?CciiL?|*K-Eee5GfUW
zC?F>7NX!P=9&(Akhc8^EYm15%-C?D1YZ(pVj(P=B1i=QYU<jzHEfG1So`-Jkzj>a|
z6zln`=g%y(+XT_J#Qr+TykoMcvxS&-Va#gp2cj-K9u63=$3m_X$Ld%CISY-IZt4<L
z$*Duh@Ny)AV4Pna4V(<sK`}bwNO+-qU}c}|Oe!;$(ohX1v&l%VacW56aEj6sxGjN~
zg~CgaIY0=M!}ispwDz{Q{_44}oW3B~XMFJ<=lo!ztIckCM~FVXIcnmfB$xy<5dAxL
zwC#OEmFvQPubzk9JF{jDIo|CCmr4Y-0Ny1%5&)s_N|f&EV~)hmYiVCl-TG_UaViaU
zfcUi(j@xpC$1K#Svu!GF;nA*P{-)Ub7i1P-d6(ItP&du|#;SLht~T8!c<kR(7PW~f
z1oQp3+M|DJZ^w>dt89RE9Pt(0kuz1VJy}_Ygr@TN4BQG%M_{-KD3g}_w1nCLp<;52
zoR@=Tc_lasqve8tQOMsY>fRh3c6BUriExaZS`c=uYwC<|xtqKSv6MgtVMgKaf8m*F
ze+N!j5C+=Jg|@@MXOr6YeNAW=Ed3r7JM89&RsprFcYm<#@TFgQW*>q@yZHDFEIJmv
za5Gp^)saROVZLLdw0a1WrqQ?ziw7_)n3e$8MH@81<r*;%FmWr5sU+NDS;l<YY80!G
z8^G}Tg!WjMBPT5U#Ww^&+Rasej;MN4Q4<W(9P_V*SewNxh=|ksv6#R>wBQ40`hOGk
zHULhQce?nSaKJ>M#=ugQ?e;+51lT5p;uzXmB|&ua5+am#U1F+$O^`S_OxIexE*%mA
zbw#)LhBx#elNBh2aotL#hj>yQnV@z1CuXTM(~BnOLR>qzr4VNqH6w(?`@F&4KY?K;
z$>coe_woEb-mkG=4Tn{vqm6%wLI&t~>Bplgs0a=8?4zBu3B;xw_=zj$Kw1bp5zgoy
z^fsz}utiYjM}>1R_tKmX+xY?#f0!I07MgXz|2kneHCuMs$@Fw6Z?h5`{KzYtO;8yT
zxeJV``-fYlq?q9F3FceQObjP+r|VCF5Q@>~sUC`GMqFKub=9sy8R>x4j+~b8PW8xr
zxq{eLtmNSAE5>H!c|KJOJ4ck^I&AXve8f(~9C(_dF`Bf}B%zbtdBbra?4HZ5)1MCZ
ze7k4=vgJwplx<6v%66YvW0QhXkAu{S*(;@{=9934#(zPFH-2r6_GqVm?TS`pc%q)g
zY)elS;!@C+f|hjTzefC5_6YahIh1$->C~aZ!J+6tINI2^$GtG*E&@=iuh1TA8rL-$
zTIbFh&TpdJgpCPy8Ie@@E35v=^p{Uc<k#Qd{`(|SIVri^ylb6I$aLQ<Y3a8jNvjJ6
zmYkBmI=k<`U~V(R@1_%j`n`2yy*}}=-dZ)Mg&pvC+$=IEz0}dly&WUM!;x2lP?Cxb
z&v;O;-FIour(}XljX8!AV1#a0K`pt9EttE}SE6YlC^8}PQN3}1MRDBHH|<lX@5b$~
z`=pd<mf7^ByDWr|zg%Vg+Hr`>x*V|5Qcl47BchCjguS*O{UmkmGGmkGcr~*;1SAxk
zslxpkh<uI)6*7m`=w7?i&YdwfBgTLylow5c1U`$wt&QDD33Y)=;(yZ~e(Op$QP_IT
zH03Z<@*bmSQkHx&F5R&@`Lqu=JmFsXTg0+3ZJK|~N8j`>P2I7<>?M1^5{cFZSe=Pn
z@|rgOlw0mKny*X*s~9M#@OyAF2Qo}8tmKgF8IEGH*xT&rpa&!_ld0OU_v-p-BrD?4
zL1e}Tn}Pm4y@{rQTdodf*g|&mH22~D_m8pp7n0um@1o=+*}wfY6ru~Cy#C+**^6A*
z!FM|E3DMMK#AC9@v7fAH5dy|1Ra183u_FO{{ry+2DFKaw;7)hCxZx7iU5=>7$U>u_
zu|ukGesG0I<g!TJT9k+AaBTJ}2m=!ufa>Z538N<?Jhdmgd7F|>blzigMw6!fDYs;v
zdN{R}rYAq3dg%@1K*dCCX3rbnBB?tBpIepv<`tikqzRzp$#*<;x&G4eX=@VNHC6Xv
zn+y6PljBGV1Ox;z28WEGS81*SuhV!RyZ2t~LsNuqbz@t)F$@N`>WwD6-)J|3dBEjN
zkd%^!u{D{yvIE1*Xc^r51xUbbl5DUbt8)Rq2I&{p#?t1KkWyAjni&NkVj=GEj&sKs
z9iJKKdu77I7~A}vC*aV-5uqWVF%Vtc2TE=p-l&(l)ek(a1w|!{HF~Y3jI5P-eGFuP
zd7npB0~Oz(0rte5pjg-?)cP)a=z(LFWQwg#+wV{MQBO%od_mvB<p@HaoCG2h2`iGo
z1nMH>r7h1VhRfdy^d(x8J{hsU7SYQQk;AcQI)p?#t{Spzii6Uc-CAc1v%X;2iDK3H
z0xHk(+DY?uNXJ`3pIw`8>_f8Btr$jFzM{;Gz|21-ud|JBRF=zV8=CE_(<TTKa|I<}
zN_xSuB<+>bOa`fj-Nft@q5{7kxz6VPY4Udy0<cd~bs+ym-Kin=y;D<eJ0vow&<4T^
zXX+vY*D*1B4MtSeGW1J#9eS<~%4~_bPom;h3fLDAphWhPYcUMkyzDrv6y9TtBT2tB
z(@)3}|Id6h<)gMcr2SS|u)63`(-f15iHGMQ?IeUqHb337d;$I6>4TM+_+%^?QUlQ3
z*_ZnU9)cnP5xT5WJ&92imt2mkEro}paD!~`S7O#VP*%B%NBlC{OSBS1tpM;IWH0-a
zn;m&YzZj5N(${T7*_UpMirA98mOjjgy$gzaDYu|sn}(2p2>;SVE>GzhJgqOPfMmwk
zC-1}=PI`G5J9CP7)8#9OfD<c~WltFlCdMK4K5*(lqU5Qg7KM6;fz9}l(T^pFw89h!
zF1H)c#J=kO6+ZNGU>5RRDaGhkk9V76t)0ognt&#PY=-PmCoD}1jV?V*fblpIcLMjb
z(r@b92j+<NR^#uP-TJb49z|VkRqOx-JLrgaxa2#KZba`41UK#uoQEW_X))K0kA^NK
zT5Gx{1vw$i<as}|cUyrug$r(e*8R_2P-_u+z~&@5Vd<}_Y_=wo0T?<ldtvA@UkbY-
zB$y{KGvKno0q7IH1+$m(mknic=e?Pf2NuJKi>vNC%R?82t6$04*}^g$Fs_DYv-Jb9
z#jO?X3u*~M>|ncu-3BbbuL#R(K}-yPbC~*sS|r166Egg2<QK*9*VzUT0fp@(y{@PI
z^za>b+E+*i0E7^`SFSn;!S~B^(47F~mAOT_(SP`B-_67fWj@B1XFkGvv4MbFt*#$W
zG%tqZ8W?`2qJ^W{@nylmj&d-QXq<)Y3Zq%JIFAKC00dxTt&nE`pkPsuIA|i?WYHtB
zTcZPW#ye#kU*MO(QE3yZEP;;u1rGe<K53@~DwD|6G@cOO;7P1Qw8$Ioz2%{in4|N-
zjySh@XP;y<8gJKQkVX_nt1&1rf+*-dT2yk`MB^Y2xPll)TXt4&c|IlcTbSAODLJGm
zqeg6O+(;Dbog${1todqOpM!N+$7zbV%#4_m!q<ow)4{2@KBk~xgvf6rCTJ~<7x9vr
z$v^}RCSVRZBj#v!jz#31R;U~yUqzu8Kz8oZzz-~<Rvp_LL-vxQo?nEWzZ7?knYOf)
zOt0`f{QDFembSY6mnJ5d)xU||n0V|<vB6^J%sc-MFZ;$5pvouEGC;tS_&OL4gaS3u
zjL)`2fRcd8Y@#VE6)g@z+v6y&V~)aQD}`_A%a#h!l#nJk{#;rK@&Z1!>L1#|*TfKc
z<|!pHo8WArQ54D=u1xIa3wnEgVb|$JEo?^V<ZC{TYXDVC%g9y$;xI%&a81b`TLASD
z2!-Li`Tf_LBtaNN;wIQ6MLn8kZp?9M0+ik<Q?*L@GD$bwJQUT+E;XbHiG{&Tah=5v
zJhq;nf9PI=W}2DUu`9x^MJ^?J2uicw8^_IL8pAPI)tpZ0s}Y>%S-baF@=c}*K0ij>
zx+;S>l<8*EXNc@ew-W(DfCVd<@qt#Fw7LsM;(f8##AUW<3&P!-%XK(bIl@C5^}whO
z>4e-`$y3!wE{7aEOozsb*y1Y|bCW&FOp&fMFO#KOXi<G^xI}9~1mZkrLoVH2s4VkQ
z3z@Xj|Kkxf2fQ#Cd8e`q3}DBl8MQ6VB_SAKz)N5zCZ11P5e^>9W29BT5<qGWBZA><
zxfnu{;OvO+sY(JdJ^N6|jiClDFhL|I=N$XZk$teCdH_Mow2<8n?`K$yrGvk-BJj<N
zs_tUNKXnGUCJBj5m2F-bR~3LPatham$t;BhQ5&Oa%Qp%5+X|F2zItk_VK4v`a_AUL
z)D6&3L@mYj@*&s{Jbq7pK+*SV0q~?-J&Aq{_K@1)&{~oG+_ybxnVlFmaB9;QVNi`D
zuu0G+quKD!sY=gOm@5snq%gr26hz=mQV18e&ZzihsL1RECJAp(o6rYB^#a2tfan?d
zuKfu&{P>D*jUE!T!^3Vp5al6kGJGv_XX|Ap+n^>~3A{Wu{PtK8BmK(OcrLjfYYRHn
zd;uadF7RNx!t`uxp&8MFor}LhvH#Cp+%@7W*@07yE#zq?GyG_^*O0j|nu6Sf0m<n}
zpAws?YurS}&oe4+ZgLkopRg}>jo3M(>Qa%)O+6@c2ZW}FhhsF;?x_Te!)%ZTkTQWa
zJR@al^ywnmZR^-_FAJw&B6}JUJy%aw8S?!V)h~j*cQCrjN2TvYxL1Bsnw?`(nn^Xk
z0FX9Ke7q#58Rd(B3G(vdL#Wv#&)ACjH}2;;pPZX<dNF3&&ki40@=*0bR8?w;9_hiY
zbB67@J+)vi8|$|H1t_CG$769#!5|83eV`9W&%unG82}4X=(-kr$o(Wm$-dz(5YF#>
zPO@{m-RG_NX%c>Zr@Y!qCnPEHnV{J;Mr9WxCFz=diuGju(xl@(Pcb)besU?12$tqm
zzZ{u{c`D!o3se#tJF3Mn(zOgX->Vcaisv!kN@s?+Ck!z(a(u-Y59TW|3K?aC5GekB
z&3`-XRM^^wueV6A`^`@qsEf%eCN}!=GP24dZ#zcqJKnIvlHmJYF;Q-U<@Y>&C#A=o
zTiLvP^Ga`J#pn*PsLD`&JeX0g)b4>gsu3QDaIj`zBXBfv3J>E3_PCtvtRJeZXln?!
zYu>LjPc7!I#0Jqhz)_s7p!bzcv{_;6mZYuHajSWN%Ql~!dLl`Dj>j!>(qc(Zpyt=g
zC=^nJCOrm_rI?gr)aJF(Cy3#rx#pEv7V~`%`~B<|EL0h(gmP*ig1&$lvO_|L_3W7b
zm{1P!K(hs#32*6(BaG0xF@GF^x0snz3=X2vb4Jvj*wboLb>+-(POseSpJcO2t4x!Y
z{Z5*8llR|_%Vs~B{td4=X-U6DlTW_^(IFX577Wg4X0DXB(w%Ovj5>!%dgsdd(VRUS
zd?ZsU{L5yg*=$Ah66MUL8WG-#>D9vhfeShe7;GIwUfX|Jp+{D<JIQI0Z@j<Tc=W^A
z4>8|R%$W=Kwfbx96VvBmNdfPS*PeVMiLTR5$!4+nj++ixOuyJM(e8JV$LViu-jqoM
z5VFYNKAPD8zoav#f*SUN9R~Gx|1^MxWvpkQx6wEyz|d}!wT~9y>Lq8#7)5cV1(YL7
zLCIy5c(%TUZ0CMy&*Ca<XmI$vYd5;mF#P=9Mt&srbLymOC9ux6%_1e4?Qt(bE=TT=
zBr;L+Iu<e$H+znmNP4Se1Ej^R0TB?yRd;%c%}2~wX3-h_hi7&GQUXbKl$Qicf)w;B
zedZ<17(|Dl0cgsw(SVpBs7rb+NF18iiu`)^vVp*+TF_~%`Ol#kQYTMMoVGah`M(1u
z7Ln~1qFQFWYnk-=EXmGfGm+9JuqKKsEHxrFFY<0FnP{OZG&TrKY2;UaVPnqD8%GlL
zmRmO>Aa5SnFgqL(n1sBXXJAW0T8&PvI|Mtd8T7Cj83V*dRr3JrsVKMh_5CQT?3gF4
z#cZ1V{Jn{;l;TMfF~51&<G-^c1G5+W!P|4=Qa&@9^kWT_M;uNKIE0{GpN&vsE&-yH
zpAwTT=@W=P#JMyS(H=2Z3ji8Q0IC`=*G8e+Fd0L6H!1>h)vYn*a4D*Wh_u6@Pa3(a
zdUoKE?88gQ49o-fKm{YQFHKA2_^DECI+=Ryft51%y+ACcozD{zUYcg>v||&~PkO4T
zpM1bW%v)O1+@$J4mryXJ&AUJ#Iz#@=qlGB$UX<YH9B_NR3U!IFhj9Y;<;{;F^AL2d
zf*DXE?1`u%3`ZaT-Io>T)6^7K{eF-s7MBO}Bg(VWHs^t8mW)o}t_katrd`B3>;3nZ
z&JNRQJ52|kAp28#Ns*EQ5Fv}})1o}QmYCi01YeKc+w(n))7~w8@0&>3)Id}dlA>yf
z7TfUqAC2^e;ByX+sN4oe&Li8Rx)D|PMAO$tv^RNh@mc-&wd*`PgUG#|*;uZk^bEv)
zK67!7H%qe+xo#L?pF3%>5sURZ-}BDC*|VCgm3`aa-R3F9rL>h0lVo1JW}2dT(XCP>
zlb4u-V(b{6_e=dEs@RTTAjsfRAsy7T1vP{&6GaM+r%^-dUcbaAA{SRHzMD5@LOFf{
zQ#JKW`27L2=VBg1@qjLC_kDfS7b!o*W+APOr*5z;@gdQlm;A#-vTTXhH+8@Nc^Qz_
z+Py?h!#x%oKXB;B(5rWAg?Ys8wN3QF#K}QL?=N(OvKFCJ-a#Uej-YA-MGA9AS~)E$
z5fj<^;E_!kr7?hFM9HCiGjQl;=@yQ`u^8fshYPxVe!>^2S}Q%U;Wwm5uaePaYf0b>
z_J56fR<B)FU2a}IH&@A%JMCTwgb^{_YbyX*NXDUwm}doj(KI2u<2+I5yU?i?3Vfwi
zN&`w9_M(Q}oXwjd24S7eB}-dwY|xp(X-C+o8p|paEb>JE1m>TJ@SrCn!ZTENYjCJX
zt%V$9m5U^q*-leQuay3SteaBzep@zQPj{8_W=SZ@eihLKhk6W$A|b#L_@A7%&=etW
z--NMA<+)}h<U&$FsuYsZUiV5M4i?crZq{Mr7ppD?4I&vB>tLj>p;r=u$Z1A_#cl3n
zQBx`i(Yadi=fRmCUJFg!ZwKRe7Y;glh;>2Et)KLANQS^prXIEv(oJj$xf?A}ND0Sn
zVHTE!z_hQ02YTiIz3RTB1(1wdRp8zdN*6$$rffGO)ASe%u4VIp29b?JBsHwWsyhdQ
zm{++OJ&KJL)-!@_Q<R=e4A1sd*erL^1B#sQ{mF6H<Nm|HSo@v*<PmxnhZ*oC$nH$e
z*Qe&M2SxfNyI)GK03=}_95+1e3E571%)m|LzoqI5_uy&?wF^o4C6Bd-vrJ=1QC#TW
z=eYIDG0Fo@qp!&@8M0AmeS9Xkk*{eC*E`@KtWVm+gM*onK2XZ>@Ga3IjuT=JN}1JD
zI6M5OyT5<f`o_~Q6urd0ijb7W#;2yY;-ZAueIj*A!ojWf5FxvENOodN`H784E8<#c
zva~fOVxEu=D57&X05wE}s~PCbAg;`BA($i(QhFn;Mh4x%R=X+uQVSL@_BSpE2O-AV
z3Dj<L#N6aFfpwF1hBx&5L_hOA&;ALsY%gSO7JCS%l6V(zYkDWbKcU=@Y(&EUY!;>}
zk=<U|ZnF8j5Tx`rrfw82AG07NPbU-QJ%X2>u)24%2#3Q}`m`|OuGqA*A-bl9qMXHI
zOykhFflRX1H{L2@ye=EeWgxJ(E>uoCy|<$L3dE?%HqRW?D%+MO%^SY)!oDRIJK`7d
z+K`1Ku-Qxq9TG?0JCz3j%Od@wMN0X5?(hQZ#9xcsz48Qdf7NWYmt4@RYAE|X;o#Au
z_+Edf50eb^01T;(I0w5a<d7p>t!|CJ(e8u{_+5(?GAGubx|BO9{0~PLA3Ab5fCG4M
zFeR7sFTeTM=_MB13-wiAJ7xkC%Yq~UGKp9yVxkSwn#e<OEo_mlu}WsVL}<z@j|_`&
z)~Pylrf?KYU4j&Tj&4)QQK=PGd}FK%Ky-kW>AOrUJ>0jut)a9tnnF~_0YDdZx56Kd
z7GKI<W?zci&1?4u(>>*XokVsF{(Es1SvZ*<PMW4C(Ih}>(gj4|{Spr;dYDSvz?~N8
zj&iT$<9&&<dUc{+^K!(xWmZh@Ut00jU{vA3N*E3q)$m&|t&TX1GL4!s<(+9LlM`H-
z(?KGty!5f||GWBVg&pvIQLkTS@>*hDo0dp5qNVRVjDrqW`R&G9TuMqx$;n=sfR`X9
z{E-kkQAMk19|UAt4}47eH=(*j0~W1^-P+1!3a#L?gN2M95FP~}!&(@4z~zd5ipN6H
zUFG7%C05KA$^ZEB3Yb0fTtVYbh&cvyJ#4WD!N5&c`lZ}4d2r&co(Yk48`YJ*o|Mh*
zeFL6n$yzOTq%l`PO+?`2CD-(mvP`B<7q@GX-i|!6P0>m&b@eJKb*E-5yVnZFjO(5O
z?5I@>pH;(NCyVNDVL!F?omkmVtgegGUQkO8sejx6XE^sZj?RoZj|2w~y?og_a}jE|
zuPi@i59WV8%l>2i!0fg2oa(9@)(%Wtl|DJeY@Kq`_44E2l9R}dyu=LCjc^PQe#`Vt
zVx4XmoiXX|_~lUKpUm02pdB_4n{Ql(B;U_~ZH%#TWYisg%y%R*5Olw)mE-(I_pYRs
z9*b``ZR{;9tT!vM{BRRHHi9P|x#FL`eXH2;v(L2iuYZZwa`Gof4*aX<>ksfHH?Y*V
zD#yM(*KC<<ekY=?l+6}|V*rBaVJnmV3JC*L2^R3vzf+DUN7R~kHn95lURsQuFsn<z
zrorci4P`Uai+XKYknTQX2HQrp-uV9-D+qZSUuh*eC4D3&T(M_>{o*2QyL8k&1U=T+
zmz7&u=RDgsIR9y<^V#*cevf;<?eqAWrCaBwt#i#g*3S4AJYRI<rU`4%7@hP}%k-p0
zTID5nzx4JNWBn*EyEVhDuXgW18Jo-YKZmSiw-KquE=70<r1Z!DvBP#8`cs(VHaO5i
z%lsU$06EJc8O$n!MlhjY>dC0@OuH5DhwmxrAF#nCRDVHTaKQ8RFY{&BN0Sq4ft%)H
zRy27zvP*!HfD5TW06tEoX_lT8OPaf&9$eqT@j79o+^${fNv%vvu%R9<?Dk@n;ucQy
z()L2oMnel0un$fP*F*=U(|74%k-Oywg3gNKtORLMJYQLH(Dh8CSYxi)nq`t@*`)t^
zBW?S`saapT$gcO<Rz4g!8BChyrk-5)`1Jx*%1X9&l^1Zx|7e=TmqUF*Qje}zl*0H&
z#S?UgT17+~h+W9WdK?-k&v2ahXrOgIg3IN^WC6$qP{HHkORF&+6cJ}Y>FM^C29e3}
z-o8pB>Uc&0D(Z<?XI#H5_O<Ql+<VLCv@Vyv;26B=9^<t;J%lD_Z-y;%KbS!3qEZ=^
zcD_baK6G|uyI1=A3iB9>6PHz98bs57WDQD#W&Z7^xE>Jt#`8ec(960WTGT2d)<w{P
zM`DZmBy_aZdx2xAJsbP5_mMNz=OTNQg3;@+J?@*ab;H`fUaFQ|qoznL5IMl&+OJI#
z*g3mF6H!$}8EWlbU|F2-cTu0v6q8}<mwE7cr2n|%+(7W94iC)Jk|rc!S_wgys;4`3
z$febyLy2)MG~XyW@iSsd6H}%{_EvXbzb+hSrn9_?8o1u3H!AQu$L!<&8LzEf`@){)
z9v4QB5IzZuN=r;aIZ3N*N5Tek^r9ywjC{$RWWsTpw2*5}#73mAV>2Ga1k`uP;Yt{8
ze~G3|5)X>l0xM(AN^**`>L`9(G{hEOCk8_fL_9D=e?F_^vVZSh6mr7UM}nS#V1D@2
zcQ076`ky)b@Auxz9*Vs5e8r$@0IuaqF_ZMViFPreJueG6eTE5yxqZT$$n$!VR$?d8
zzZ%uNj|ZZY0I9E2EA{JTzn!`88=S@Y>M0!RbL$9wfCmy6QbKgw?8(HOnZQ_vT^M1s
z!e(b<Z3NvfPi-`LSoy;HmT#2i+iWq^gdBiirYnhL_R=JgILb@t$SS7^vAR)$A%o0i
zo*+p=%y2qFdf`|KPc-;r4j2st+=YVvphK^^F|U$76<14?m!tY`voUr7UGRMhX2R>$
zeC-5~P##91;H)Vg6bj6(G&Mtsy>I(PVAhhtkiRMePoY1@S6W1>OMJjX?7#z}HetU@
z&A^8%`JFomL7#z0DMH?hsDTKc)Jnn|yMH69$Q7a-VQ;5P2N>qoFzD5OQDf(v7a7fV
zb~i)Kk|(3!7+K{A(p_k5MW+_H#r@0)T{V<>@4w?JZE!x?%L7(wauxe?>MD~&e6r1i
z4})N6(o54s&c;=gLNFxLF*#-34?y^PBw9@5)N({z6@aQ3rHBBEh9ASw5}sLG)K3{>
zh953#MAJLmYT2!hA;qL8ck26;{H|BNyRY8P-rVMWAG=@12GZsDeUD_PRA=41`Ag>e
zO>`%Cqj8p>gop@gFCdbVCcvRdlazJi+$6o~)<6F|<vad}zKNYkU-G~#uBaiZm)sge
z162VtbE!gJ6k5zxiMuEJts!MU$hzuM%c75xZRCb8={y$R66>i$Mh$FL9yb!}=iUZW
zEIXC@lR?9DYh<tT`zNrjx!4}%UWS@STo6iw#*)D2!w4@fCJ<GSm|OEGDdB0DY7C{7
z6*Vk!Hyv><;08FnjWdS^`I@KZC*%2fG@T87<W@0z2uoqsD8*Qwt820`s$P0g54H=f
zE0aGZAA3<->4uyuTP9(jqREPLrJrP@Lt?{EFSG^AWe}3gy^xkn{N}K0IE)X+k+NQD
z^NCh81i1*~5iUKvTxZOV8)R|SQ20sM5r(|toGj<dg88g+bg+cU&A|rv8fV><l{8Pp
zdeHUioHzt3fMW2aD83>!jb3}7u*sOEvLtZMH1|1u<0|0l#-LFQ&H@av`)C^2{7Bhk
zLKXs&p{N)D_={t7iI3Xx5|KRgu%PIOTEb00J`kO<OALTVR3ok?f!*%sKC35#yt;RR
zPhV}BQH~q^#r8qHb#)qN0+pu7&HKk&j_7TroN7cjOrO*Hm|vcc{mi^@h)u2bL2kB)
zbdq~<89<^*0?yG8o0<l6qwwh`iJ)%bj8Y*+pNgnq*nz;oL)~A2fis5=N`dR50g%Uu
zTLb8?zFUrLKW6&!7{r3P9+*<vl~STxAV7>lSTIjaNtxsQ!uc04bk2HyJ7;Xxf*l)I
zd6$#$?c%ju8iF9;V%QeIF_^liD#ookO_)2xmVSDv8McmG!9;I!>D<3?nW7<MsJd0W
zPl(M8L))G*BgtliwV*x{d=b-6##`*sR+lxKCtmdyuQm-a&H`bf627IA>Dy~KzxY~g
zR5#}T5B;^hr7!=+^UC+JL4Y`Ir9Z!&5X2WSco!A;)5=R!7;Ediq5+k9(u;T4rx5Ow
zijjQ!jz|*FAtu@tnWBYgk{?40qtyor;-G8>;?^rM1$kVOJX0(oLUtu&^fgPxe0@cs
z-YB;EUv0QN=dgEOX^zs9y1le+@q-U+NFug$>WOBOz#zn222qT2zm%N9AptRmqpzVu
z9Iw5+)NDzTpbXUW&2+Zk4#<WqF&8hndDwx?NYN7GhGTCSshfyroS~nj8?0ib8P>D!
zvwQ^-D%QTebZG1aHH3oi8KI`g0CU_LI+xSgoaOn{1;e$wPm#@V?xrv!NDIiQ9h!u|
z{AdyKkgL!hZ?zj5fSs(yO@E9f-8uN_i>o%^unxRG83!u8<RZpoPspLNh-M5M@7ENy
z$d&Qa7#_=W2Ik{ly8?}1Xdf-`vzJR0h599G#Hfms#>c)IxiOYg*;h7P|ASA;?jcys
z2JceLysLwB0ZIla4mJ5<^^)Pxm=HN8;^c-yF8yAwls3svx`ij6wiE2VjqVS6WMHi_
zdOd5kP((x?k!oDK<-W7UM6-=a9zX?1Y;mjV@KkXNk=B@ALNCl|1anz7IbQwQrqKBp
zRKxLjD2^CA%p$2bBRX?ptTV}lltfBkh7WpY7;!EpXNX5Vq2s7qbu}YqA3Zq*-a!9P
zKr*@bx7`}XPbYrJ=*U9m-ZgVBTtW<ZmLIr3>=aWk6-OoL%qVl-U074LuQ;fMT&)CX
zU-S#%4X3r`*^sRK(siSfQ%fL0%z||dBV@tBf$3AS51b89LO!@MW^SI3XFrp79rH8-
z4s8kGM)guu2r`IS(6dmyVFTZ6eXMv<C@{_giXQGe&Z1Uq2^&(>L;8&NFhl;Gj9{H(
zPGg(U;V~c}h3;j}Qz*=5ltGnxg*leqsEi^k=;0_%eHuo<3Q#Qod+zL(Y+lKZZW4oG
z(o9?x$<I|>V#3?SaFkcP(lp?LnGnf{sv>ibt0g?+;^oa4a~*#ZHO-ciZXEHyvzs?w
zz=<e&bxR$K&(jYkKM(GSphu9wP#%);<tM$?muF8Z`_&SgckAE4;=8a!Q+=}=XO4-`
z6$aII_U4e6*nIRVu4!?OH5Gzd5<sVOBmn{poZsDpZ;s^zLzNwR;47y}`YA{XY!z0p
z*)DEUE8|lULmuSkH^T*$8!@Kd6sIuf@JSg3PDy5(mzgFliDvtYu@|q|gE%ilRow`!
zy`a|*f@iX0!>#6-9IQAE-T4<n!kNfG7@r16?WTZ)NM~b%_^ul3f%jehBWH>hcECN=
zQkSYi@ZreNgE$S_t)Cdh@F6^05Ws1CN<x^X(mAbH!cdrkmZH8@TWQgaNHTKlWj)Id
z*@F-YZc8=rhOrBp0RxO=`-bc9pGqPKnh8Nth1Haq2&y7u46w!^sbRp<(o6AGFGEsZ
z>6;foOnM*=-&3m*H2|5%^-=^m1T2CFJN5n?FKvWD);rlCt^~UacSjhM%gj@uFu)~j
zx+*AE2=2vIV{L$fa0Mt7qa{x*5Xm<5`bTF#3h;Wt+B*pX9S;o~eL&UMG!h@FMns2S
z7;aE~n){_!b37bZ!CzI>rx6`jf&4>M1D}>7NXqM?2GEWEgZlA-b9DDrPsTkNbPpXo
z#>4UyG>>(lGJ;r$P_Uz#fvEKGgW{cqGfb~RRzS6>p0p$3^@<3G1&n;D8s3B`4~aKN
zVpt*PTewV8OYUSv!qhCub%B?{#v~O&P@y~s%JAwX5NKR20UK^v(J|(Frs~H2N!nGP
zfkN{|EIww?Sr51~t5E0?vknJ>`n|zcM=iT_a0%ov4DT{xE(5b^2J0u&KqI(w;xA`S
zOnTL8KFa<Ou-ba{>-I|3WHV2@RGRBDTO=sCywKUdONY=wieVRE3EO9|JIcdR4I5XL
z6u2hEVldW??jmd$j?0ShLTnIQVE~oS1an-$<c#=#`~-HTmK;zG{$lcbKm6=Pq^|k8
zUjIFG!>u&!ocQ6oS+r!&?@}K|VDA6QiE`M3^S;ajiI_YaBIVXe_aiz)&dmW7w+06^
zBCo)9O1S&na22YE0@Gl41BZEB$SRm5gX<Y>A$t7HS3nKOkWY+37;g*l;DvbYKiYZ?
z$P#)BA{2k|Y@_+seVzX@BV3ti+Uj^V9vfWEflKrs&NibTt@4B{#6E3PY!=%ET;4gq
zY3Pm4MXWi8RfWNOijHVgrPtNsI&<k}1#5rAynjsPiN&+%wjlUn$3ijOh|QOs2Be1<
z6GwuHrZ92=z}QlYj_Ea2aFDEcTbD1D4(6D?)o6N~$nKb&yu}f+C_Vp!=YOw$6FWye
zczNANf3>jWms2;{A6xq+6{PE(qwY5&c`qMQLCBC+K;a=PBnAt^z^n#AhPTwkHLHt<
zVoI?wjzx^a^^HeXjA4Y~6p|-g9olQ*(hc1Z*<kz_YBY@g3upq*Rmfr&8m0K#bDG!d
zr*^b6{(jaT|D6j~Y9H)-d;Zav+8@ptzvnJ=(NneW+{B_eLhRnZSo*w#E_%tOGH6cG
zK&D}@Ju4`ThvIty_SASwx%eqL*vR%*nA@EdmxT0IKzGR#-JG;3mfJ#U><}zV7-9Th
z+=)UR9EKn2Xn%-j#RKc_nrHf3Yl%s=d!^KWI=0>r_1s|Zhf0edw?_wJxqDYJu*uu=
z1~4=67_phYaL>QSZiNp557<mePZMi(ks`ne^mFtgI0S?io_${q4eAU;lH0LAyL5t{
zXcSsVD++~{f`rUaZZ}vP7m1%=Ct%cJYxRx@MXj!;1<F2bR5!kKp%*PhLe^2TTH1cx
z94xlnxV1)8w#`DCDA|5=jWO3Qat_lMe)A_e<Ur=9GAvX^5$)oF=%RWR?&#3&InKlD
z^}vYYsg-iI;0I3-w9J%ZD<Q6_mm6@|tO73ucQ@45rsuIPsv}@QR9D|4aJB~VBUW$?
zkD`3$J~3uw!?51c8DcE>t&VZp?F_rM00#T4v+hu2J2F61`<^ki+$D&R8%aDOxCq}s
z3TI%-j~zd=fjgbG$DBMyGs*$~%>H3;6lBv7j-VSSUaDWoVE(5lfXeZ=$FB%z;sHxA
z##;&W1@YRSA2^>v#{dK+cgmtLv><&~#VtT+E%?G9CJp#8ZR2ZS$JDmR(FHKISLSpm
z6TvEgi$SY#y&O>mkwtkp^3&Tc;#8=Tvoa%qF$W$<yin}xloLQEP-6OLt9N}JDr-TK
zpfB+1#@vp+@j7_h=qp23;WhXohEu>V<zbTc)70NgOd~Z$JYRbCV9G|Pj$8!e?Hq#&
z$*l`6xn}(u%;!D5WR#<Dt41{lE9}rhkX^TAt4<Xv4jVC@U&@yjhLFv@Htz$51uU(`
z*$mb)f#4DVWr%-&4h|rX&GgXAedFhU)DDtGh@OzJzIhR_cTFyFBtpI9z;&i1AvTPS
zN_@qiWh(w3C259*T6F7%RhMR2KeSM8IUQ=~B5+33a#R=KTpCQRtY{)!DSU~f1x%EV
zVhV9J&y}`VSHWB1aYJbu|2ba5IaU3@+r(Gn!6JJ}LHn-ohvpeC^2xUV3c}mH`N8p)
zu!QzyWztH=u_d!aV&*3{i9@RRxUNnc7-r((7)`Ez`21RSGgyommqwvP9)jfHsZ=u*
zHAH#H&C78f#+J0eI!iY-wAq<BuLYs5PP^X!k>^CA)hL{X$@g^{1*kV6a`sn(2cvu1
zh5xU4>;bhtmKbmJt56rhS`j1^ef;~8&-ooF|C3CzQhiOaK6X`V8wg>8xK`!q?_Q1v
z&)f%ZF<ReM4HujesD$xX6<`yRD#I__+EY|5Aq<~#I4i6Nmq-+hzE{RX;88BG!KXGv
zEU(3`1}r)dTtsec=zkd9>YToNY^-y7@xFThSUn-8J9kRbjq7h(LZL<*cHp3@;O8+N
zube}wh80r!9nXL5hB&|Pr^qRdv(q6HTv~`I97j0U!c;{(5FQbchT?qMJgffg*dT0v
z(@0}Svk#j`8GS>s!DEhg3)vX+7=4e_VrY9K@i^$a4WI_Mn`b;O^8emP>aESVgIkKC
zZ4jMpGN-(<9jO_iJ0keNtB)>6$3zyAjp!A>CX&K+*pbHg_-{3YrvN%^5e?8#Jb}(c
zgvY#e(>za18O+2ywV*R?T6j-A@FjlIR9s(Cj_M^3XM=AWJrxMo$W?-RR+$s~x}v}5
z`LVnnW9|)M^oX5U9MQF-Lf(e{Ad$dVcB@pPT|h89AZ*P7|BA8U>QeZiZmg5Y=ZydS
zj#`eZfd~i!`Wr66!tf2W=-~ckF%OI72jV17mP`oZWR=&coct@uUOd`*Nu1Nz*TjxB
z7<1yvkFvpBjm?QZ6C1x6jvsg{0dP>3U^L;NS0BntyHLQvFNuYuOp^ivA3i~3+N~I|
zGbe;pjl!^0j1`+dXMFr!t*i*n1=RvrDGF>_SQ#uUd7kY@Rz;R9H!E9VOrE6N)D2uz
zT-LZUT<WQ@97lU4mdk8@WJJmBzC=ecT!20KFum@xn?TvsO}V=c(rUl7D{Ue+3%n3P
zh@iNGW5TcNn`5$_NnljN0gK`#Sx$b=82b*$R)Zj6W<cG4ke4HRSyn43ioJXXO()H9
z{@}}lMsVQ-S%@FG&utiva-!_X#^x957_QQh*r5;~3HF>l8hn+{bf^ZR&zO+N|1+tv
zB7S-)mxN=YEk-16kpfVWK-Mw(*1;Pkuzb}egjSj-le>)<zpICZPIXW&i)%ZgDtwZ;
z^^(k&`3Q>nbQs-nq;D?NGHxL;_#~LZ7(=R~DV_l|>w=yUL^3v52_77MxslVWko?s`
z0I~ui2pSOZ4MdpiVxywtY9ghm3cl+q;k+Vq?+hWfxHiGXOH4Mes29JW3Tg%7C<s<J
zsP+`T0;oHOnc8$D)-!N_$&F_~BuSA);4YO;L&+E~b&y^vJy{Hkz*UfvJT~X-Gw40x
z;{t#q5{)sa6Qm)+Kpke66kRb3NqUKynFF|kE->BbRg$P$jEHK)ahwoG`^0>A{oG&Z
zVK;yYWme&~aW2wrMMouL3*TLuE6-Grfb)=hJ2B@PxN}HJ!Sz6HW3TkWFFh5O2YF<I
zC<>laUS=-5u9myV;3$o&(7O15t&v3bP4EXJ)WpsraDe{@?r??*WQLL^E`FSmB>eo7
zzeQz^XEi8}Awt9Q;y|Y^sRfINxX49Z$l}y;<WLW14dnU&2ua{VdZ3M3T`WBUd0>R`
ztxY&eBD#kOsq5h_;%&$yyXb_Gc3S-xl4fKb(57ew^Q!v@#EJVHM<sSR<Iua%_M>)K
z?ZdR6`nkW&*}Oau;UHb+<}3<kQx{`U>Rf~~JcYjzZL{KjDWlAexzqW&;G%AwG}i3)
zHx-S!Kz?YsWSD*mU3Ax;fHDYh!kZJ#7J_Wwe~MoLy!H$GiC@Ug?2{2IK~$&3^L=`&
z*OA}m#w|e#he?#5w89hWo2Pe=C5ph7^b(LqgKgNMLkmM*G@W+F=s?Vwz28uBSvMR3
z85dfy9TZT)g5VT9_zQ#OFoIQ6oG_kf^k~3AK->b^7$)H06UF86&Q|+^7}m3q(Ueb0
zO|XM0>M%m##=Qa-44ns*JkCt00GV~)C!beC4qlFeZuf`0$}OW>i0V}eC)P{1z6LtE
z<s&Hr$+#g1+V#;sNCH;!qup&i!`9=&59#%TL#4*>=Fv{;hwfLIvf&AbWc1a?#@+%j
zNbECrxFDsjkyFdJBUwDj>;absk+noMa|t&B6M*3;hsi{3r3u!!@4YaDr5f}p+<h3~
z+?eGtjt@em<@NtP$YV%pnq94`VZ~#(2=>Pi_5&Z))in*>c(jEm(z>=*Ph6~zK|E_$
zGiB}L(%R`W>SLW==iW|YJp3tnCH!dLQo16?eEY+GlJwhQy0{}z2G;xVDP2@zA$m|3
z<0k&>!z*9@2qWa9Dqb?;`jxGdKp*TOHos|vyJi-^^PuARuAD_e-$f~I7CCp9-$!wB
z!T6mZtNPgGQc-7~5#vjN?=n~9-JeN4SV%tpR?lu2TE}>QxV6y!AE|4|9+Ot}du)Dt
z7+Y~cP%wrrEVtlQz(PB1i6280H+}!6WEN~e^{XObR4;EVD+4eQVfO~bo3(l;hE{s6
z0x;qsV1e6c1qy7gXn%O$a~Io<4`T=k2vrZRTFcx^U;jGkIkiT-`GYTZZ{i<Xk<M6t
z_qz0T@*c(mjc=s<JP_wAU=9u4fWSb3;2Ohm)!k$>yJ>Wq9#%BqFS%euyBbG3z!Xrl
zK!oGTGF9K|)ZZULUcq$k#r?(H7lV`so2WmxRG8-!Qd{L0Tv=FSw+zI7cF!^9IQDuk
zrF)^J&i{O-31>KR>Gt7swC4NV=j40k!|93NupiR`akcB&Z6H$wERjZUi%`3e4@lVl
z`yo9H>Q}TfwkFg4+aOMN7kX5$I`~`T^~R7gn2pUDKnp64-*MnVjp6DdeBN__Rf{do
zAYyKAJrF)P6kHU3H|%)#*xf<b?(>OH$&=Tq^|2O?{A-RmU%Ru1(P?+)5G*k!#f0+~
z;ir&hyC||%EXw6Rih=DEx_)f--~8NDqC6Di8{8ZeA}WB0kX9+KpIC(7$pjxk*r^~k
zF(mN#tk~diOD`HAT1<578Ca1U+33$a6qv?5Z(myU#ZiMUJvJrK=pJX*dF7~HN9PmA
z6(qbeBGOHFy_+&gW{C%e0SV^SHHm^2f)d?HZYD_!HG5HR%AdFEN?r@ORDi?ZD(oqu
zha5Wal?tuI5OVU~-iSs=Heyvy+Xx~Bgc}+^9O{jrI0mOySYItRJivN(hsVG8hOy3D
zQXXKU-G+P8p~|jx2}Ao6rrde$iSbcXnzkUOGl)+TFQAnLmjN<m>+A_I`6&yLSpwO>
zJylRGQYk8!ts2)u$TE<%xiy9Xiw_?v-CnC=AB2%jm@e`u*(<`wd71@P48(3LDhH7F
z@ZjL7&8V9`$S9GUyy4i=ZPXd+oM{{!ZbMa9Q^s+zvY6PcwgfJM?V;mZ|2Sx+3k>iv
z`7waJFuKibrYXO?a1|j|GVz0Z4B(CGMnr`!4fl>kqi{v89Ljof_`u8F8L*^`ES7G}
zbtCKP4v~Ja+i27fzm?ojsizWtzfw;$>-|G3+7rBMK9rM@?JJt%darg)Urc;FTrYH#
zv>$>%rH8q*qIcJ)pT>tb1ayQE%==*uc;L3RJ_*f@{azrGCq1`G0E$4Eg)#6@C61vo
za7Ms!6f8f8>>sOJrpS)*b9v6jGHd!pJZ{!8XGrHPk9=&rgg#6T>Kut3nYQ%4@re49
z$2-W`3R7^sFc|F0hs+5UzY0s{erac)CJ@#Uq5k}67b^_Hfj+O5%|7{tsRU`X{keO}
zVuQ<(&3&Cxp2waW740+JLVbf9X<$#4*!bNcDjzMh^|wpzp@h&;Vi-EW|5uDyZUnf3
z`T%;O2n(@!VQ+ivkGoDBcz?WJ)Pn8iW^wgar**eS^idWA&M~u(W<fmN7w|EyZngr!
zRRV$$e)jGSqU(|J33R0Ja05Ba)i+;0eL;;_k!U#JE_>^(U?##DB`GM?ppJ{k4%tVQ
z-hLQK1OBBHt27H^Q;?SpiyTb>$VDdaeV3!TsV^q<>ilO<w>z8h*;N<csH4m@Er#DF
zZP;N{oNx5!2jIZ~0FQ^JqI#v6zat76pM-1gJatM`Ieo>}6}_PkEH~c`a9Xv~K}IB;
z+Hsgx6v`=(=>uWhvrvaIHs0F8F$k>mpm^azt+~DJw%)r@Ne4fS<%4676pY13675bH
zk~v)*@E^v^7@q)_BVJSqFKC|$1eXRSU48J?Q}Y=nr~RU(1q(%_+I{NO=MQGZ%cq^Z
zcW@x5M>u@fT)Vp()62>rW>|`Xw-y6Ph^IX)azNZ@J2!B8u+pLfLBx0*fxZZ>(BX5h
zuHAKdoDH5hkJ~WdjF|D1=+qm<4YLWc`xdv4B5|DG&Z%8&g53W6XlU3=mov$1B@2B*
zc4-K1@q6LL^+u-?Mo0>f!xQytTw@5FXx^?P<{CA#g;r`_I0?T<`wMkQzpiwQ4-6$=
zgRwQ<AHiwixRU4;AoWnv_`kVEBx9ad;?;V)h~8Gm*c>BiB#Z%p<cdtJKt7sHh@U<a
zTf5%Apx3Opi1~m?T8Oe9vO|5b=?96AaA96`JntOva3U%LNU9p)7`V~k(zs>j{0PIa
zOr!`6<bnp&6QfAkMwojTwR>>CnWXJ(foXo!onND$@->avSN2^w)Vd8;@NrSAF*Fc}
z`3}H2PKoIo#sz4qD8yfHMFGE|uL1z;#BW$+6j^C2!LYsQ!8bu8;W+ngQRpf47*Pc%
zs+HZ(J71%G3N>fs*`Sr0#7m3&;g5XtN7)!8=E=r9r*}9xE<~}Fu56lj(49_<E5Gfm
zi5HBGpKHx&h7W#9RNOlRMy`QLAD0EFZmri%8ta=K#j&&6k|0exu35|Eg(yf-UHRzN
zhd1JqRj2M8frqJlIS)tm5<Pl==v9+6fk+{Te!Ux;n+EgfzgJ>z5C*0h)Y!r$3{UDc
zO8oN%cLom=>yh}=u=rrS71;RE@NL<!XMD|~Hsuy61xZa5dh-&S$zvp1S8WSgY_;OM
zg|NJ^De7q66oCe3AaV53ZQpnZS1w&C{%ro?t;mLyfvQU{fjZ)7VCk6m4qxV`9!%P?
zb&O(XU!z!ZUM+qzK~Du_f|U?41dsW{#-kLtFcab5JH3y*q3z2CUOn}9N7{1QL;)z(
zwI$eu0Cp7@><fLb#_q8OX7bRT@qgT8qA4#-hM3`%4{ii89g3&|NEE%+PC!vh2&P@u
zHJJIB6&*B*WGTPP*w%!Tq%rw0p6#buNG+KAP$^tQ7!f&5>xW{qH6)8~HhVi0wcB|3
zmAC&c(}vM^3rwWKP^QlVBk8hrj1Kj|zaR_NGtHHs4QKSCdDaz{=eOZy8}Y`f8$b&A
z-Pou>gihv0=S7C0k`Xc+Vof5NVcLUUo0u<MPzDELD5HAR)*v_fnsreQz#Ho0F^H>K
z-q`Hz32M2Emkt%{w+z!X*c>0=C~n@0ddXAnf@;NFnSKl@fd!EHI}Y3Y)DDpfK}p-T
z{D~_xDe*91c^H|Bh>3bvM|I3guGD0joPaukg6M2)zAd_cW&#KbMq{+t54$*8Fd);0
z))&Et7;7FCn7V;xJh*YHcFTCHF<fsSqG^k`CtGCb0WV1g^}wG+<woJE3V8(c0oG&3
zl~Jz4-mw|Q?eCjw$eqm)589vtlUuJc8^Sz79GNWPofurhS7dxhbHmsM>jw}1L-|2N
z5NCT+3<LR)@vaIh4A&o=CBU9)bj~Od4UCmQsOfd@Pg|s<iDaCH*GtG$E#!u}1t%QR
zAH%gE0+tJi3nTX-#6kdX+t6Kjb^)0r8UkHN0SCm23JME+HV_jv6Df&M?uYE%Y0IvZ
z|7r%>#wWK6Z;2|-2qN(HptA%O&WRia6Ls1z@PG@upn(T~eY2t;gJyhWr}Hvyj!fJY
zlHh!WdiAV+$pss`kgpJGVp7sCrI1z<8Njw)e87`-GDYj+Rl~4#2R1J`nz!1B8SfF`
z(^GdP{jgg1HK=r7!Eo~$Vc&iAbC?o2IDjrj!wElZ)}M1f15{8hlBc-Z&FulJ71t5(
z%w9>%Jb>7e7TI$DP`fyEU-56+hgKw-m4%b#Aw{-XWHSa!khC!}@lSW#+#0kEVU`pv
zj027?s8w;I1EWbIaW6XG)&<uJeWxpHnj=9S{cJ2(Xdh{fojG?D_u5}*6^r!;i<|qJ
zt%HtTr@Od<_`G}-0_+k&42?`8b0*Jp<dPt8FP$(0w}l<<i~4yb1I4Sy2)!;R5*@f9
zNRBgd{m<{V`)O6=4{8~IBO213Qr3^E0bpSNxh!q~Ed@lRM4scj&Hdw*)36Q^2{E$S
ztRUX$^$RsdTUZjCzE8Ze!7Z`qyA@;tFJTn#kHF&PK6z#YdXh~M%s3er?}9Kp^_Egp
zih7yReT^irLzl2)?jzH*4YrAbr69F3tc7afm<tp1gylpC<V6t<AvdtYj#)}E(O)cC
z$6HsLEhIl;rANH@7B(0}is(YN9p4p1Q~6G5<<wo(@iXG;eGR$_A<-w-{{>PfB&wf}
z>PE4CsR=|w$|1p|xLZ_hFljpwdC9aF|0D=h(N;%z$)t?#&`S_oWJVbn-4y*ZJwwt)
zJXuqI=EA<=M!yb%Xd0MrLI+r()A~R?(Jox-F#;bMYeu8_V6kJwX`0E_&Y8MMt;P|5
zfykV|O<9TThSsmB)gtPHw_DJ$0(Ni6LKmP=VDduM#5(rS61M+FEHd<B(@g*ZY2`98
z*&gIpuR9Me!a)51Mbq%>am-j1G8ZLv69Uj6rZvjA;=CACI?qsS8_8ptc~1umxJBH$
zJV7`la9vUa#bA{K){-GLAYEK~70MgKOy-TaM^s0p=wX-<Ob(j4itPA-hq#vt=ZmXQ
zAzBorE;2%TvdrzN1$(kV0J&_qG{ntB=k(Fyxev06r*>cdlu;_SkwZ9(+HXL8M&L+n
zu>x2zu$(-()4ifd67VCjS!Y)*@MVINFU7ry&N71kRSLNnio=m8%G@O0V8wl?Z)$hj
zOfz6sVNgVzukw;Gz~PgFx-nOo7guA{I=IgmJ6LE$lJk$8J~S{DF$ZAu?3$gMT0hJx
zwAs)HK&h8*cKk{~F&&fBUU~BR{~uM#LhS7GUJJ1yNV{<~ml&h2kB#ew=ni1sf1H=*
z(ezbN5SL&5;J1HD2Zfba<9wL~tA8FxIAn9-eH?7T99DfKyy=vo90@+A<kw`+(YH5m
z<m~nu!z{!g2Db()r*$sw!D+Aq>@WKZyc(lPlWZWwUUjr{LIQTS$m|Cu*$Pz6x`;i0
zet`ePgNI?uvuGN*FoKW3#^Z1OsU#9V&4_?Xols~k!o!a3Ww&0AXCfaeK)6|fWeqE>
z+4nD2mJOb{Ch<xyZnU@ERMwBQisRMM$_g~ds?X_T&$6#6I<J~&5*g2D=vu%E)AG->
zd!Zz{1K=PpSiE?MTpS<Xit11pUKdFa`~&Egs~aExfPsWeoVhU9$IjwfD5~8HnMI3<
zaj=<QcJUSC*q+0Px&L7>^KgEzksGOxHS6#1Re08LRurr++*k>(2Olh2_qAZ5V|XbT
zD!WCDyy2n(n3N>n6iEw1P|?9x3Gs>;92B7p<b@0wss?8z0So^xug8D$!AfE`2YJYi
z!8w>c2QuLW4t*J$jZwmK1Or2*7mK}BN2R3j!AgSP*>a#T)U)WQpBNl~Wh*((7B|ZU
z(+$DmDKHx3wP(v!ZMN_ek<x4z{42?3l#p)>0}kXgc2#DO9^_@X4?}}jf!Xo@iu%Ui
z{%HwxN<94EH3XuNi<cEG>^e-ml3fwGNXe~**1Gj#u^h#Yn@kV?!?d;N)zlB-E^)Ju
zi2N()90(Tm-Z}GL4Hla<hMS`Zgxw7BGNU#ClX7YB0L%3D(%W>SA6YnF2fBLuk7n|~
zTKF&_67B%+#pZe&|8ZHF>S6qXOU3XTGE?|7da<T>+U~OlT&Om9*}-c~URvkLkDs}l
z!!pK=M=I<^C;?*y`7Cxo!q{8TvpaCJ{A%zyEJXyZ(j8sUXu~i{yvQE5bKe5=Ps|hO
zuSR~}Xk0?y`us;<SDNoi+aFqMPRj7sB!PDT6sB>$^@-?bM$rmGhc0h6*)eu=J&H<`
zY+J6EP}own>QJi*?RRW8R?Tf333PZL-s{j$&+n{hUlzwR$9h^Z%&_~a@@g=L!B+ul
zYN3ueTtDoW?S3geC&%qQNrE%Ue3r_Wl^1?8dg`sgm3^~c)OP!#scp!V0Z4+Vh-J?E
z*ZtRIs+O!;Na6NnvS|)bs4!M?T%M_@lc@$vVh@EIm5rgu)=j!tg^J~AwZ3MB#X(xU
zFP?dfR#rSX#BxUX2g<KX>+?t(36rwO%+S>_Y*zUI`<X_;-`Lc4gMC|Uc4dD1(2hrH
zvF&>{Y=1)fsW-|eBR~rWL;wP^f1JG<WcH>Bj=8W(Q>8g1aP4x!W|t!zs)VJ7qWn7u
zTQ`e7>t9~%^_($~$Ml@4w_!D$^;GaRh@`@k+5AgoS77~iI8fTR^qR|KN-MmOp)YgR
z=eF6HoMOv&Jr&EHoOc_R$VW$%U3=fzwS$s%-n36ndZlz15WdN*f5*(!GM&Mqpi{om
zp$F!YZOCcdS|Fl=CYVE8en>YKD#cN)ai$$<w)^2bed94VNPc9+rVM;M_>sXdm;PCI
zep)0+7<wF1R9o@lk!4LF3La?3W#pu(vTM>?4c%@PV*bg;uPToZ-8Vf=uGtW=nOsSj
zl7*OI=w4~-gP)XLszJB!j0$!wIY0{<d<~XT?nXuo)ibp7VQp?b@xQ_s{u_yRv80cz
z0i~Y4J=-^EAp8UeA#ii=vve1ilISB(KgrHsJb#-WxX4Iy()Z`Prze4F+Z26oDgV^e
z`Ja-X|M2#H&j9m0MJF+D+B-MpC21=l5kT6x{<kaR+;DF%MYL*EMQRtIg5d)Zj63RZ
zHb}{E9lm#|w!)?7K8%gkS46T(a=g_(m2g%9=sM&wu_Ti?x$ap(GC8|nn%u&Mdab_U
z-&W8@^(}b9MUlywQF!e`Gsv<>8ESaihfZ}~_sYb-Anlca$7Yq1K<}qJ=YL{go4m7H
z*=sMOE<;x3LgtQ9Z#4jdil~Mg>_k_YX4e~;=Z?hkub=M8C=`Cwte<{y2n#xDC=pru
zIJO3<|1$C`f>VGX_?s`GzjWKh3rpZ>zHeqCl3*UotlQ}J=_8h8boYHPX|PhWNlN;1
zNeBjigpW$ZPq+5{dl{W(HxGFAHz@}e@RH#`=pGbb6?H{ZE4e2;OxE6`EgRENWw)K#
z7%jBM{^gP|h)bRd`lyU%gKY$_U%K~Vb67wMHIbaO6#f6a&M+Pzcg<^J%0KI1Ja;%X
z*M6spCU@9^1dnpRlp<Y6BY-U>E%Jm-{}b6v3A%FY$4p{E3X7`@J|Eq8PQi_6r42L?
z*Eo@XKT?@L6uiB7IuF*>hw7g!Y$b+EG$YzY`iKZzo1ewWcA`3yO3)dG<A+bJ^3kM2
z-lHT_K03kbm=n!$!;&hiFv=Fp9*zq8gYt-!Bwz@LN;ci{Bn~LKJ}#_z8QVV-HWU|k
z@Rn$?p4GsgWtyrwq_kCfW1!DYt~-=z_3y4+*u4f`e0r>TEU9--c@P3?^}ka8s?Z!k
zZ>*skeL3Xyl~q=Wut_0D607i^@=h7NGiTmvNxg-<2<pBH3e+e}+6;5TFGHq}JdFz*
z6gI4u<5--LNi@~-j@HM;R^oyR<y9568c*+V9QaKp3`L>Mr=Zg}HgxV#0h2Zb7DTIr
z5l{1A-`k&(XX&d~L7mWH_Q(<|$Ro*eITYERPA)j<pQ2BkV&675^F;yb9YRt*IgMYi
z;-B~fP&ThT;gNq1Aw`ifO$mibHyRFK>MLSyKWG4<1hV2~Vd?3oBI;d6xLrVE0a}K~
z;mz3jyHRW?MbQauLx+n6#G75a@bQ>!Aj+^_s`U;3o>;Ip+RUQ+h2@y$PM_#9m*!6R
zu3o}I6UbEnlRpDy>Vku&ANhdPVi0D}zF1_h(TA4op}pale7$=RsVKZda1~Aka%*qw
zJ%<TX0EMw=)TCl!AyNV)PGG5X%?w1ts9GPI`<b*7O@LAn>fW$fae8E`<Q6S#`;Azq
zPwb^B=^Js^yV9_29~G!fa<Yp^m=~5G-tgx3N)V%9y73sr%JQQXV;y1hY!f0nN<WZR
zSkdprx+8~LvoF1>u{Ls9WgHV5jm9gi51HB32Vx5>L6n1k?;6GxxZf@ol)#BT#dc3f
zS6nXzf++H6aw$1Baqfa7<zL{TtH|)1Ff9{ACag>3QDCaOeluf)&AEtCR^{OE#W%sI
zIGYSjA(YS&ow;}zU62SbyLGQ+USHFIWkG3`H5b_}LqenH^16}BD$Wrl|91UDwinq3
z_wqG`%L1l*I}nkn`IA1HvWR5iX~*NKm6W@93C3mt4|EY(fQTlqB0{$>@b^ymFzSFF
zF%$%+VI<QvI@H-n%|j6#mKAcTrPyR;{Nb+cp!%z~Tbmn|pq^3ON0hP4YkClelWvT{
z9Px|8LHSZ1Dqao=FWPTs@>ZV&y5a<z7f3{sS9#@aR1HK93q%YGBm|%ss{F2)Fj?RH
z-{|x&rid~6@pU~ZM$LYjG|_2pv>-Jjs^d7WutU9f!~H+X<}t?vc4G8(5}o7Ld^Yyt
zk-k1X=t1j!M4|QSFYg?LRd7lFzpT9nSW`{cD2fP36+$mklF$-*5u{59q4zEz(n0Cc
z5s?sDC?TPvfCLZ_QHml0BE2gFREVN<jT(6aA|fE~*}UKR@Av)pJm=nXpXcs}nLRUm
zF|%g&+H2ODS+hcZaX9f0LiA>I*})R@W|G`=bli-<ECkT@3C<RhbpSd2PlQ;2QIfdz
z51rovmZW?G+*1xFIXc}H+TCEP1Uh<KWM_6rCs~Rx<p;!_6W)^9p(P~$8aT5w@XZ~b
z=tjV4E9r4rwrggO3GSgCfZ7raAf-`qNRYY*G*`gbJ_Wo{Kt8kpx@{<#mkrRu374c}
zWB>>oP#L&@{`&(Ie!{430p(_N^rjMkTySzROX342-9kzLEBp7Mo$o{L0lr5UjZZLn
zA3|^k1QuX^_5;pqeVlYYM3I2)UjUuX<v)GPd+(|5p;mPO48bCnV5Kx?^}TO-Kz|Lu
z=oTa3NPL+r$zXmF7#-;=W)j@MZ~%=lp1dN$DG541pi;oarzE`@12<3(000O92|9*b
z%o3>vuO|H4E#yl8O11zPlYA%HA0QuY3xyH@E%$&4fT_Ru>-Z1A1SL;CuwgoJ(5e5r
z=uV$hLw72I?$#BM?L`T$g4Vy^0UonXpE_m0_2aC+zuokM`&z}tCaUd?pPZd6|1%@x
z?dYFw^w++Bx?KsM$=`JJ{dAO5`6+^*x1Cfw&K+yhMbnKL<X+$ZCmOEZIy^ph7(FSa
zoSLPtn@Tq;Wi3<`rGr~`IhIE>>Cx%Ysjgyk6a^dG|IcsRcK&O`sS)GpuEu92WsT2D
z`2<wgh@U(h-Qxcv<MHG73BdpL&|?alQyz&;`*+APnBo59;dA$2!S~&tA7TEZaG95V
zmLrDKUH=5+RIa0C*S~x8AE}FHLHfOoYNzj>xp?-LDBVdP8#+KX0Z^cTHBA}L-cLDY
zb6pv>6~L%B-QTE2e|H)v{-4=~XJx#<)aa`|U2#~N=Ct|sTI)aZtV!jm|Nc(^lxw-?
zuhalY)wHo*88%6Uch5(~`nvLehW}RWY2MFj|33u%=$!!As{p3|=Wv7N+qA=-eVz+H
z@BJ5`?$hOnPgi{I{J(|I167Ylv}V%ug0a`jQ~x7mHE5LqS>SeDxVNkgdzDgmwEl+m
zdiB}Fntyj(<-d2+pgT#8_wz|CPTukR{L3cb1Otcv-T_A4e~M0S|0{l<f7--4yh?HW
z<^K=Vjn&~mgN|SO`rB<Y26v{-uUGRY`bhoyb?!Lzl&-*{Z+W&&%EepN_+QD(e)%a^
zeo1Lf4+Dso-FsG6^jVF&3iywz6s>aBYvlzEdUT0&G+?1B;plVqtB}9&nfh8|g6DQm
zeg5$;5FiGU2Fc5D%O9_+@-C=>t*-w~hXUIKNNLR&RF9tPOr!(odECKc+R~vs8+%<v
z#X3M5Ha$FY{@(#0P6xMwk@My3|J?rT0nKPp@LmTPdreiaK@D<e8hfoAZ*_eY=L%GY
zy{>Aoat|ooJw-V^^I7-!`oAMJ-Z3>5%j>GJhXE*9{{JrcD6N#;)5>~*_9^=WIcCB7
zmX}*zR|DYvRi>x!X_Xg^si~HI3UD|#J^8;r0^0jO1^uTMpOTZBfW2PUf9g6s(*raV
z01Ny79`qgoJW&EfM^8@=u)S0NlCe&hA}GT4it@!<ToT}df0!cu2~%7{YpV@OZcL4v
zN}!pcs!oyfq#`A?yZHld6rkXZ3^slR%?zYo6)PWnP6eirs)PpfvSiKulrm^i-o^Vr
zmNp1xZzmsV`<M-msoykzWI&UzJbMjHjxc0^{&vOv)Q=Ni_&ju^ci+IS72byXMgdfS
zeABB~!qs(g_8YhYl&8vbNH~x9g0FFi@4?Z&?*XN8BY(pUW$-8vXpUee;|FiTeOsuj
z?pP$4ZB)Mo?Ws6i5&RK#@=S6Mzz*6XDl3sE>U2+Mv9ZKp7-VFvufy7+as)iCP)y|c
zjaDYT)s7)WzRiz6GGyQ~33eJYvSI&u7tQ>2#>wfPE;TqfwpQSL`J3i#K|VK50i9_s
zlQ%nNGW5Oo0hw*jOa^O4Rl3L)ZNS7Y|4{Op1Nl<quTQhqJ{R6h7z*<iK7db0Q$aYV
zvae>tj{v1r>%5ELK_P15!cP}LirLy727)!C)2Kf4qD+TD*)WR3X;gbH3I0%p%F}hf
zYb>b7=|L=(@6wo&$)ZG_5NnI?$GdrMtPf-T2sW?b)6`)lA*hOh%)NBVsrXf($J*fd
zHob{yfn0kV^}@s4fq1gy7b)Obs^quLb-!TaRdp`ZTc?@@{Dc9zlV=X-qZQrd3nqyw
z#*|Y|pO?}4s6$=g)qs|(2b^Y-D;@8D(%WH{$TPkFN&i9{_Mz`%^Z#40=xUPIK>}c(
zn4y|_7gpfU@<~7_&S@O`{NrOo|ERVwb02%sYtV|3c=dEj!{<Wr95x5vK{LtUeq*&W
z`RuWY57^@c0TO~vN$ck(qxDTH*l(t&yotOjNELdNn>0<K8Q<{z@f$bu$-tnD`qNeL
zBz)Q5vaxfP3A{i2<+tRWs0Y?FPT6m^Z#iYZRyGtIajjGm4H|$CQ_SYx_y7&q1m@K^
z%RIHN7L{_XtV(xPnVmHh%^CwcpPfNG5CsbKP?{U?0W!rYtz)+NOkj;Odu$Zn73%JP
zE8~nspU1!lsI7(!ApoTK>Q}p7urqt&Yk-orm`@7Gra+v?40Ww^V{P^ADEL=Z(SYZp
zs7O<f`?WyPGbhv-%8((5`Cr*)CLOJRf&T9jt#wuE@PYind?XORLOc+p0+`~ywa+#)
z89WLHnkK|VD>{{V@wXnxCly`G0x($cuW<}7#+{d(3p&YVRphamxLb((0Hg}9gI`XP
zKk3|9FX{m>66A5W%w8aN68V(kS(0wo7Z?ciDK^30qIV2Pot-(!oq+#Se3E#MYD}+f
zss9I;uj2#%^?ZBY7G=z4ntx#>&k48Q!S(ojXWLP%4}V6VLc&hl7SAJa&^20TiH;Uq
zdz-vh`s22*pcg<<sPEgrjV%CAu@(kU&p(xY)sD$c5NzOTW>O(HI;9CxU&9A-W3TdE
z88ujP9S0z{89?{JQ6s4TiD1w*TU^e!KYcl5(&LjFlM>AqW_#pzGtoYa;&P&Q(wB|u
z1ivzmfv@-!{KsgswKv5D765QB*lr{X0I9F|ByHwbqcGO$WC2sjJ39jY0qY*gMTJc>
zKGAjX<Xz9I6Jq&S0N#QE<9BLolt6;LWo#6GOjt;AlVrsNn|fc%OY(;N*QI4+@vMgc
z1&pGZ8yc9?UG<TL-r$5}UC<(8{c<!o;VAuM+yOI`zznU?<`&4|27Up9kg$@wCQy~L
zlTM)@9Q`S554_x9NyYoy-0B!br=NdP_#>}!N>;S772kAuW7s>$3~51ZdKLe?$$xxk
zG9*c`+0uOW)#RqxI^yg7a)z)uGc7mCDxv_^-eyZ5RhGV1blp%`Tfjg5yWnF*r*%gz
zLvdflbbOM|cd;}fGQcTGw$EJX#Y3roJx-#6mhYeioZ`gCMi1o75?Ps{EDWG5^m#`m
z-UeqxtKB~K3v0X1c{b@isS@L$^a6Kyca}#p!_i&DrtA5Q9Rc`?^PJ{{WHmsaI~UV8
z9M@Lz{glA97t+x=dfLOF0<;0MM>F2P7%=&IKl{uj8$X9u1r4T8K9cm8F{<KKbovYt
zfH&C)NFLus!;0YJ01b>E5NsH@4Ee6|WdTzT_5*9X!4i;_{59JYxCEe58C|*>AEWhJ
z(FxhNN-6-DSS4c93HBz+nw|kZ{%ilGjok^#{|b7q5?0XaMeq?^3|c){=5BhbS2{}N
zZG?#rgFNdITqXY{V(`8SuJ-&IpU+r8jk7#JfB!vU8S5iZJ?>Q4VcUCzR{>t9Tcy<n
zKlsRzjtXmsjtbk@gv>?6Ha2<#ZuiS;sDrEba-JdE%!U#lFtI?Jgoc$23VM$bb*10u
zP3O6Knn$vxz;Q0<!ZFEBXlwfCos+uDDYM3`K)q%E5wdleoeIUjhWMQ@fA~cHG~z+o
zH`!15vis?7D&#DuLa`Ca!aKP>{0W@beLJ{1#*=P~mZeX&+iYyfC-H(F*EpXbUOK;2
zJt$B$Mf+Dvms7wAF&88iu?I(^gp{Dm@h0C-P&EL)j_F~=NgCQr=7|=IOgdY8H`_qd
zoNbRB@#sE4dg=YWifHC}+ir)oXo7WC`g&2&_(@$y>Fcj4rcO0mwuZvp0=Zl$MP>=y
z>q0W(!6NlB{3R|VpZU#fvxeToD1ckP1`1;HoLD=2U;8Hh7lcu6Q8X}}%>q!`<_6bC
zE@0p<^QVBPzq(#?HEZZ7uqi+7)CS!kL?h(~haRL2_5dYM;xN87>qx{@TX;y)(p6@+
z^zCr@q{jfWa1M#R#1}B;U*jAS9RkdY?R-I$#TjsbK=QSRQ-~Jv1wXJq|1t3VB<>np
zi}916F6?dF^}Upr3LpkBTMd2Jg*t-Sg6&D;K+%U3K#Y>-0yIj%b{V)%G5$-IIt2oA
zsTkUMM@pnB<7?V3+77JJwn<ka{t>AF@9+j(xb_6dW`XW-&8}Q(*CETGcaW*0x^Ufj
zYLHD}XVwkE#W$eE`F3V*wQ-lkP>$RlfyMBI>~B#$n5DJ^6{#t%cw{~&T{nv^gP^i~
z=_f{*1&z0U|3#s%?;O32%Wt+nPAR_`f5)-u4NK%&&>M4LHbJ{~Hk;+#v4~9`^5@Lb
zUAr(Oj}c51M|Yq;%MEV90Je5C$3%g-%UDVu_DcfOhb$$ZJ?KaLo(WkzarP2dR(35D
z$j*-5iy2BZqwMOj0bs!2H*+kna!mqHTh%0V6TdfpC}+%ED|D&Q)XJ<*y5VWR^NVLe
z?YppvkCWfC>Lsf%jad~MS|YOt89Q2DcwQ{cU9MwsPj7T8zydG*)W@@JkBi7aGOjl^
zt+d>=$4+_%-(<xXn%dmLNVir<Oaeoc7gm%Vlp?+pdHg8lDCuuHXna$WrDps-uV)^U
z(`-3SgVG@U@|pNm*kwl-|L~IJj_-IL)2!U+ynPw8K3AcrsvpMhZLs=48l!1rz3hVt
zH#gTuf#j+Qn&3=jJO<Bz^aO!8ARs~ID;0ymiWod7jfq`Q+1C;%7(mxDqS*QP_;~AK
z(HbOUIhl2gMDXl<?QE5LJ*==^{O<T~$z6G6b5-_hpQ<czZD2`Rry5pIdoEK9SbYFZ
zByvhWVZhfrmkV<lp5}Ei$r_k&s|4-2`0uF>XCsl^fC1}BDd5j#DVJxnsE;X`MNQg^
z!H-4dZds@Wbx*vYwB$T5GZMufY+;@&c!N<yP7!OLWnW}j>(jDj;Bzh0q%qH?6FYrD
zBzp2dj`E?6Vb+OXf%loHoSdrPM4?_e6u49aEz9?3Fg}*Kxwv)?v3Fd*^qxQ<K<CmJ
zjZy4^FmwCjqZ($zuVh@wkgwa~`?Rx6=|%@4A_}~H$fIoH!{PKxS@p7?jbmy|+Qj|i
zwiz%%Id#bfmkN!z3_&1)OG;i!i=(PsFKt!~9`yrOA;N#t(QLi4|E5dtGTS;QJMI=#
z>pzfcX^EA6{E)K8ckg)I(^)sKm=9mY<YM`7EdATlT$&i+PUX0g%1;3eP?UrsO8BzJ
zSXZf%*EU|m*qXhMTLj#LaX_D*v%bS7`$CE`MXE}V^XbFd?<=LgU<{c-nNLn0s4Wbt
za$(*U^S<CEVr7Zx{=9y<nQgiSS^r!Er@2h}?TfigU#CLlJLPLsN+EU)r)PKPhP=Cs
z-BZ=fbnu#f3XAx&pNzxK6QJnw71>wWf-s9FH;a13dd0`}u$wMxi;F+WDqfkuJ6qmQ
z!hQZCJJrFnD;uJpHh(zxQ|JOupIvfZHROjhPGvJNz+#Pi<)`w>b31)bR$Wt?5(#$Y
z+S@W}qLN_`Sxh?SUVQm_O6}lONO>v5dVO5?Oyic$r8)Jm3Si88vZzV8Lx&FyYeX9c
zN&+NWAhoNpDH@fzVoU<1iGre&0eUpogWXg!MQ~+hRjj*Oftj<ikO&@A1PByVu~ILm
zsy<p4c++CDJjxn-dK)+Bj(-|p^qKzhMEKo#LvG!?b2WzgOugWz#?d;|(RIjC<#g3B
z(Hiq45qu=Gc1Z~XW>!b=xEkjAE~ulX2LGnpw)hiyEVD1ZZ|c|<W#p$y$WDT`>&QHh
zGfv>s2rTo$yNAD)-kFaVpkC%Ae=@nszXEJ<sb=VWsr>7=q;=ib>?8qT?aV#p^w{0^
z8YP_q8?M`|5jf|<63UK1o0tEN`-HiD5iq|urjIYQM?aR#pYSU%JZ<F;x+%5mf=W~)
zQKokTYZNgQZR~-ZTCKEYIY|h(kjaz^n{*h%4f<qFlBWP^%^!A?_i9EOQa>w%Eb1GV
zXjv^MvNhxVaIC<ggNxc9d#<T6P2?q;?u!lc57{M?d>(9fgB37YlNzJRRXGflN@1DF
z8}?w9x+67L1Q$kFh$+j^!tZ&?e%*@&uLI$wJ~j2y9u9*YzNg%VGHr9!wF6$ruH!R;
zq(9<&(BlGN6><VI?2bufcx6`@L!7#<W4<}D*?#%^5j+#g$`=nO*+cD_W!gOo7Z0;#
z+HH8=R6Xbed5W6WuS_>cdL)=UGMY{>el4#gmU%pOH|vK1dFu~W%aUfdD+-~fiKSDV
zNG^k3x9VwLj5PZrC=`4$2F9iJY&%|4*GZxvq8&rw@+8k`>sN!~R;CpE-67xp%jvt2
zqFwQfv{0G>J;NG#Nk6i}qRly9Lu5-&SQR2IrSD+&04&xqzzlSl#L~o^3oyRc=C-?}
zTs<QvqxZlr31AIPiDTtpdd}AG*u{MbV2co93ztpM7D7iiv!EQ!fa`Q&Pe_lwu29zc
z=uGO(;g9y(v!343f74wuI3OmdyRwy<nSI|&ctJ=iY^(@1PF~)1B&V3T6;M3h#T&Hk
zVk2-Y#zK}=FPRU;-f_o{LzIzM(5z1+QYoIi-OhGgnAd0s2qZWMWn$;!ZA3G}fh~7Q
zv_8`Y$+N^^g~nOy6xs!=|0tbQOx{Z1)Y{O+(1ddb8)SFIB$iuo5)&+Q427XrL-@{S
zMC_X037V+1HRD1{X9rbCrV}ARrGm1JL}utlx03I8(hX053vv{Nxbn`Ep8*F>F&KlN
z>BoLYIg#s$UZmNPEt%MCF{8NzlIQC!vB0+czv(_DVRU2+O&HgY)Z59q>715fYhvjA
z(r&f>)OWh;(|YB4F9OStcMZOoiiv8`Fc|B{jxv*TI|?k^5|ucVi?>?AK?db1qcL);
zMzE5opCJx?UY756j)lMIEK<YR?yb1%Bl?{z88Qn8i=@S)5Mqa<9~3HGfZ0JL$Pg0r
zvJPbm6>VaZ^jMST!OSQGSBBZ_Ae_?NBCos;7hH^+xNNdC8Rg<1E4P8y-pCp69V6vF
zL|G0xpW9h1?r-a_l&He6w|#q9I?N4Gwu9Q0Y!;KL&R+g<b4A@QiFpIsdIEK13!;#@
zW@=l;XqIX2b6^9X`x3BMqx^z?$wjS*q!R@BC`0w~T{4lMiPRPo6{KZFWIZ-a;@T85
z{1t3^Rmh}p*3DmKU$x73c;?gwQp?Mqdbx!idszPcICEoiFYiwkyY*=NnXJhhkRzUH
z(30`@1F5GprL;Fp{&7B3Nu+698rbO5Asbz0*VS_vB`M@yO|!6P`B#l3a5U>vIHif^
zTYFkWOUI2WMfFHMzfpy?5inaxK3(V4-?dx*d~&CtO2?$EZ>)nm*6<v`Oc=~UWR@<6
zN}$!G{JLd{8OJE{zNrMjNWsh#HIOU$K{yoEjH8wpPc|_lG7Ag+s)zk*@hEb<646*c
zS86QetYW?!@`9x{QA1jD<E9Z(Gl}K4ik%WfTcNrCLoG1<HKqI(YX1FbsJ+`Xe8L=w
zj()%4<y(VUM#pG_KV5pcnKNx8pHTf5;v5g@Z#o3zZ@K|4T<z>GECLY6<Z6dAHhEpp
zM*_7=d(dtZ!kMM<?_ZGHL=Om5l(!pUjau+Wa!)ft3f@lPmW*PZxHXW`Cdh|u6b~!d
z1dIOT{@iQ{Y%H{2v33+w&|GJ7Noo44eNjE|KI{FHi|hbCTMx6db576fcT|88R8=#Z
zysvUs__-}Gl2E%LuiH9gTLdPMqm%YUEnhD22-ktd{d~Ni`e*BbfKMfe6y(aGuDmN(
zix~_w)?{(Jm-n4`;dB_+n0RDocxuldErFxx7e}}MrekET?uJ0PY0w(Q#wOBN-SOQU
z=5+*SK#;UCn%$M7JR<gCr_1N1=D+hc*NC?8y!%t-XOu}z9{&NI)`!?QC$<^*+`-bu
zmv?xKm#`a%B}Wc-XBBE%0M)Eb7?j$rrkHWKa`!gxN6E_cL*4Np2%5}!nu%@gN;9|5
zCjk9pAyq>m<JhFDCVnW^EQ$Pa8HD=SrN+$x<2DsHi|uZhIo7_^+`iP~2BOri)6QJO
zzG&IvOuYyX%!0AT+>?d!PVsTLFi-#DKBCQD!_F8~E)*1`ILHZ6!Z%>_(?)Xgna+6B
zbbRPElS{><1@<pha=d%=HyuOvPRjB5qv~~&Zcc7->08e*A9BX4X22W!Br{P#%9t_j
z(n_*Pwiuy(DG9uPzI;;+Fi>UQ#L55QypS{H>j9(6MC!~0H8868vb`-_t>VYi>NHn(
zy%@A#&TEDr;Z*f<wYXWBONBL=n(-3-tCIOWB0Zf*qFEH*`Yys!uAtyz6D<P|if#gt
z8MsVa9C7x`jEU0dsm-c)4rIM5qZbIF*OfMiXvh~1#jI}RW#(S)U>84bLtsI&OWDyL
zP@sVUK`Fp2YW<D+>rL)qHGto=GKIE1A-nhmJ;YKT)m$zY$&jJReXkGf%NEW97$|GD
zP??PhTHMZ2<F6=9=*y1yQlcgMG;_x6Jm%#@t#|7Qw_ue7j0iXWdYQ8g=DWcG60SQE
z7_E<X4qPSNVE94+ezvN^c3TWOjN^VQkzoiz6I&2W819R+1oLvMailjN-&qc6z12;>
zh~dg?_d$`zSX^}MqnaAPb%~Kn2S?RN<=st{4H;IuzKmr*2fRTn{6kJ<!OBn?18!GJ
zOwm>m^Z2l^>vsELV4Eb@O~z;!lQi?kk4n7!_mV(l(Z}b-P$qsXX5Y;7Syt-#0I^ww
z*If_iyM$)Okc1U8V9}TCwt(+ubTm<qRS)~Td|lvT`J`-8Rfr!>{ksfGQs+;e4(x5g
z{sK+0sd$JpPhcJO2Vi~zWFw#U_UAk(CJYA9Q!KkgqMfr20QgCzb{9$OXyuEu-Ar9N
zA}@~4aRWAa;|bN^dLI641NJ^v`%>54c=&e^Sua<}#_M$88%leQY-qjjqFuE9JDFj|
zEqP+ONo()8vg3I{?}?TCI<HiXm0@(c!o)`S%I%s?X@zx!H3*94L4rZ2Y7j{ZycCQ6
zUiCi=IV{v7<sVg6oRNCjR2PmKnwXw;)o2owdZ<6%1EJ=0oM=A|K%|JIDgS^Kmtc-~
zUH+z%HXSD$x3$)d6Cc-J0wkC!dk=gpxYHs+-6MsOCe8!ZA4!@^ZfGqG8~sr2bE66z
zeO3e_z?_8{5c1?BttosnxM04~5A6|iwjP@;wcHk?zhD%G110l&1!(80xJv|=6tht+
zrsj28qBT?YWu<5@j`4HX8`brZG>0Q-c=`bh;F2O4i8C5CzW^O@TCOADFJ!FbW+j3}
zj6>;Ep|@OGe>^8$DMAr1aH7#g8<X;L1>qhilCpiN#RCay3}A+&u;ycmW9ZR^{79~e
ztvp55pczTqhAGj1q=bxoMpaNVuX9H&J~O)$yldh&HXA(<N-#dVpp)70`YxJX6O*f`
zuCn67OjGN3CacEy5w7dc_b1wD>yGSs`BfBC>H=&hburEuf+U0+$>n%)T9JKNq{wk2
zl)T{^x-(pj@84wsftV43aJXVwga(C%q)2lMe=mbW1S5%Ea_v3hk;z5k-lrPCJjYZT
z#EdeN;brW+>H@7Beq~Vj{1G{sA#bK2bybGG=8@612uF|F;%A?_O>z%#jI_=M*gZQo
zuN|MWNfU+stQQejx7e3?Nxac8$OQ=VHbdt2d8V^FoQ=s|19o1~*mH5e=d%iIOMUd2
z$A=}#tzKp?4vBH0!D~4KYO|GYtFuqXd$Xt+ScieFoR!Pzo)B#$=xsGO7EmlFWn0W>
zm=NL!n<dh;$gH<^wjE1$ZCAe2K529bB9Rpla*4P8t)OLN2Ig^He2JS$c1}%pJGroF
z#pmWku>e##ES}l(6|0LHhse`I-HppwnMI5D`&{JAplE@04s42Vplt6eD~~qC(F!GE
zw@7Hgg8dzVC1-9;7ONjt%ipFPU%X!UUG!p8+DqQVbfP9$Na#fB!Tff#FLiI(w=e3p
zFI@$C`^6lX!Q#>iMm}WK4y>lxeU&i3_}X+rAXX|$UDF65WZc&EGmN>}TP+q^$<~?u
zT|aYh#NdOpmu=s7zLB^P;*?(b7T|8Qz_?yE_X`NbE(k{;K1SylAs}EQ`g$ms5UdJj
z0E1GE82KVD7T`0ciS6x}_agpsFgHT9jFtIFtXJS|nU1LjjtLUqg%?3%ZHc;uGDh3R
zyV2m=;WBq>N<^tS?Xk0Iltjp`z&*7=9}ffX=z%!sbS7!u5YSF`$ecHA>=e(u7M=#b
zIL+jgQ7!MasFHHuV6^YrA1%O$QYwaArjebCnY8~+q5}O^g<0HF2No)3rlc)aCJH43
zEJ^E@<!TSaS}|az4DGUkPyVdbl0ly+o=~~*p7hJE;p7*IQV4{Q1fF@<*;#$G4lIex
zg2PD^mll|tC$AiZ2H6Hz;$(IbIKdW#1PIlkgG~8Fq9z5^8!KX_Qh7tVRoR`SYZ1)C
z2-C*K@BxBqj;!fZiAvUcn1YK(nPokBI<rMn?F;M=N47`}(BVY&B)1z+>LD&Nm1U~q
z5P>FLL%G%|xw7TV*~OSf&E?<FsQkBX`5wWVHLp@?hs;*RYPaQxyGjnPAF(GUt@s>;
zdlREJEH})5!zCIUU$EL20V2BAH)flkNyl;`u#{0R-Fb?}O$I<6Mg*%@6TZM?R_aON
zJKdfUp0H@sr;PFPA-kxJodQRmXlF?t5ssj2yn?%FP&Rj;VDoC;>}-GLtLjU7=OHQs
zoJO-h<uLI$(5fm;)4d)0onxfW##_cm2ij>u5G-|lbOwVH7UKZ3MBDY)vz)K>zX0p2
z_D8X^5ZTXz!Zaf$)mDpi5%JF$V*GY=1SmgLbvmW8&kh?g(DXhRc(`C3_bP@I#(HEr
zJ{-;mu9^^)kVdHu%zI+&`rFp5ut~4LM9?@@uClDRy5sn9c0cYE^~`KT+OUZcB_HFW
z*_N~2sm<@g%Dd;ZM9dG9u9fo7VYYg~?)jkZzDSAkjSQ2^tQiuOm(=UDB3E)-hl<Z;
zdz4E}l(TjT_;5evTE@%x%&c_3*u4r&YJgr|Dg`swfd2+Ezdd>D6^p`GMt{CU-<iAc
zp7-7xz3Nv*FV=M5M}@y%v8><!5PVaBHrL+XUR1BNhB^HGk_&0HQpMHQ|6#7a|9#C*
zG5V2M#F5Ysh7V{%RkBLM8@bl*yxGf+7Tqk8H8)t?)y9E?5nG96e0+k+j!r8-#T+l*
zdY#_C54wD2Y50|EFIguW%fmKaWyu!{xXUv*{fWd-)zIXx+97Q^<C8!d<eI(5<aOrO
zHXF+Fm)~^TS!6r3ig7kdV!3`d#tJUd=^N6no3RkmT;gRwws{aF7?N%^+^Pqil#8&=
z`Bl4NQYq%GdCci65MZKSDKo2~=O;Eq0}Wh8%@cY%6eoYx;a(@;zZ|G6TkaHC6Qd!N
zEQeZlE5`jMO2S#S_c3#apGA1{a11n<Yct`3Id@RAJr=nc!}QSCEcQm5n^(bv?Y9Tm
zOk&<`K44ydfNhz_-0sQ0WW!1Zbvh((xpF@<pnsstFL=w?93yvs+~=LjqoRBBL8gYb
z<ePE8Td>d0S(&siMfd{?P8@l7@|mtTpty+&V=lkydpf~|Xp?i47m@?ktMxvzejj&r
zM9PgRqvWPR1RrwM!2iz6*2VcME$I2AGR_Y%SF2p#_JYmKEn@bCi(S_?96~E3T$Cp@
ziug`9eQzF3hQlVaZl{%Am8xSUbnh$gEBm~Uf}asHFV*jVCLDfWeI8jtHEq9Ry#<56
zcdWIl7ldh$q&6(={d)5^3FwNsjiB;QQxvd1NC<PqO4}D%A1D91>tgK7Qj9HE+pe%b
zI$1TdXtKDe{zB$2SCb9gJrmN3Y>SmLv6{ZSSOfp+>5a7MnbTYX?xx)g(Ky6c1C^(~
z)MKWZNF|6CQAbeV&y`1+B$uz&ka)-vvG%1h?-50LYj=GB$Ixl=hai`sR1=RK5u@#h
zcjmH+nbiNK5;Cz0SPOY#4w#Y1D~*lK#?czGj=9fy$f9^{5{EoPo*k5b574hhe?uZI
zX+nE}-kT!3hv2Z>jGvSr8c7+?1<;}G%EKnJdFA4PZOlD?3bITU7%*ocxfyAfZ9nt}
zyMvoCf5dko@c?C?#lVJDKkw_NFC90uZuvg1jH70<Z=)`VU$QT<id7UFGsw=l)^1#{
zH|(U~`qARCcl-0Y@dZuBYrmLe_`hf9qk6C(l&|IBg}U<h`JEe}&hJOSn1q`G3?e)X
zlqcnJq0&;^gNXUKvD1MT6S=BZ2g0_r_u;-eiVbn9PQxnSX@33h%IoGt{n6C~MmIQ3
z!5~iJF~e6P(k8xY&A#{O@?gm?2-h0dc{3T+VaXXkBEbqJ`;_9wg|T6!>gT|x$gbN$
zF^SRV8goi#_c2;;gk|PpTq-dVIjqjI9|~)n>-C~t@3c5RT#qG4J{<IoN#^#;o6LA#
z@`fZ<0CSyN8WxFJwJON#Xd;eRHWsN5F<qKkezWM(2kP2|HBo7Qa-_+vrttx9hyVH%
zfV6Ubi$5>2IXBnKbywK@z|!_Ju<at{ma@+t<AnwrU+uIbM16pVyv6Tw$S-*)9a1cN
zOjRR!2sWu*4=jOYn9xnt!>M!gp^+;2tQ^j_b16vD0W|7~lBw}p9mhH2<eKyEf(eiC
zuB~Masgc~DYwE5s1+yS)eKenmzR%Rp3GQ|y2{z{T;Aedeaa`?Oc$ITJR-0M<Uu}$4
z`q<16ucoDzrtVq!Uxkey<qgk>S<c;$?1F2ry<eH>6KQj~SFSd`9#+>CS|o}Bw;8;8
zI!L+|VuSF9RpZEZcD5D5+<;w!OwHwKXE{Kt9ZQsxQ_6#T6eQqOUeAEV+&KNFR?ijO
zeHS5ct8f0j9z%@5cawrW4gsH*LZ3Chm5}BIo+^awnwcoiDl>vvhL|W|w}#x!lE}(g
zMHl%Th(L#D*mPxTIOh@#0`KmCV~M-w8%YL4os^F8ktN`OffZSdaV~yFSdRg8o^AYj
zbd_ljI8E3QAq*noK$Z~R^;ZcqDxW5(rkkP_k@3naRh&({owDuNL*J>FkRQYxzwm)7
zK%K8uP-`)4+{HZ)GQOM%4Qw-XL`d8hictTJ>xa!-tk^EcLVBvGd#dNunzFthaCQC=
zy!};kHrjgpR*UF2gMIN=U-!88&V_JjbhPE$<d?v{nhUq4WGJpH>y7{DoD^6T$Eq8|
z4AmyT4Czs+n`}02(-I9u+hqNcop^1pIF6)Ui7V;o9}Vz#aN{3!9Hx6{mW%RfOj7a^
zV<){FQ0(4c{C!zmTGjNyEKb;USL4$RwdwiXZ`n8Hr=i#Q1R(C&?s2kqt`$idqOJUo
zt3H75b($Wv!9RI*8yfqOj~NiAA8r>oMCu7!sqgvAFg$I9Gd+*rz0`C6GhTxo4wj=a
zuq%3PtffBS)b)#O>Pq4bjeyzb)L1H1SlQ!Ol<5{AwpC73EDT|`J=1<~C3A>J;ex>q
zPY%T5w10>=TrCd04Dy@u^8Ow*w>*6Z!fL~H?Y@)$WB47Lbm0ikcWVCJAB!FH2o`x_
zxSFdI?w*)A_bf_BYIk@$sdiEJ0nR60xm~^NjX;N_Ze{39jmC!ynG41uwfNZl=Blh9
z=LAMQ;epjlI^Zk!lzgN%gcrM>-5k;yjl`kiLl-+mY-2;u27L$`@r@I{Kw|yPZsD9h
z6XN@V6TYQ|39V1g-xMN>y$An_RL$MsVSnN-a_zQ{;M3K?B7B5`=FXVi);!-OkG8C}
zprs}+PoCQ~*SfWsTI2!$FSd8CZv0RJ)(Jgd3~b3@W_S}zNk)<BTLx-NPSf1qv!V28
zLk2PLyYmCk!11FO#ycvDyDWRE57&~}CaOuhe|R3y@F|S(!AU9j8)aG;%@kmT{E}i(
zd_nhEsvu=o#OO7k;WH((-3W5`*;CuZ<BOz8z?M|BvkTdc@NLhDjD9$pZ?}2LZa@z-
z4;#mZjKup_W|JZMlizt|3fW@7i+|JARC(txuY{94q}}tPaEKS2%{(<L^VT1#5Qn<t
z#h+d*T$v2<l4%itru_NJRsFon3PsTfV3k{(2;vz~AI+<|5BTsKiqg%%v(!m<tgHlO
z3EbK*$<eQxz1?F<wb%4v(>PLj^08p`CQj%N4=^YsDsRLA>J?}!dog%tv^0PnqOF>h
zlGC60`gWGlMA0w-8zLpYVHBtKM}&vRXUbcV(W*1bu{e6DH9)vs*EetZ68Ns<JL1?-
zR0+o2+M+3BUMRxO@O7nV+n`ap`xYv}Xpqy{)dy0OdN1O-{0k{1UhZpn5suEc3kUqB
zVT#k9V<XH@as&U+4Pas8)kd{bWNJpo<421lYb=+R-tj>f<a{8bno0_VZ{!(o_eER!
zvQ#{~+VV!?{rV7s03D^>-}L8TUaGETw6s(G{s$}M9-GpIZhG2n<INv0@w3E!akf4o
zWZv>p!|m%`W3kVSmOj*~u?l;ibrWy1Gq|8YxY``(n`+@><T$}wT{#e=#%U~w==U`#
zol`cpVqE*HuMzWO*ge+`8et9%uest4&#!YmXRQ$RHXJVts#fIK&>Ta=R;OaFbNsb2
zz;7AvAMi17aaUqqR;40SZ6i9*og-G=>TUD$Y$I2Zy4Wm_vWe7MzNsORvge(~qqiIP
zeC|ooL^Zp06Q0fs4N=}bC0d>*_&ElHzbGh|-ZZDXthq5Q_~s5Sikh32+T(wbV3}Ta
zNAPbteLwHqUxCkxOr;RknCILGNi-8)lpx8?Ey*n=2@Amqi~KVEf;b0F2YoVfp5KjW
zpnQ4G*PpT@u(%Jp!Tt<s^@gXhpl^A)d1}uF|J~OG42?RJJN8LAUF7(5VP6-gqcHJq
z7+C8wo0>D7HJ!sC5iCLdSp#{Iw0d9Pu+e2hOvop(0Gp1%;lg7^Q)lfL*|S+w?P|~{
zyl<v74xjNW@b+&#;?B2W24;#4gW7~Dg4omzeT*u$z-8RCLzzxeiKBn=4uu~^3q%wl
zUoq_?Zq&t}F4)J5uzp}b``qwm(oi#Q?=&G(r+3E-%K1mleLI+#tpq1Ma)Ye1vilqC
zv<L0BAxg7UuSngtO&|M3(-mv9h*Zs~@t_R$3U*COsD@8?dkV2V`gm-gP_`HGby{A*
zzN2143~mpe80pzqPr3&jlpX&y&tcMGYxE$QTc(CvB?x`3gIPpq(3cM<;F}cX0<2>h
zb4Z%zb0mFSf4?>gZyT(I2TQzxS8<G}M?{z!+XOhu3wm$si%--u<^I;aZRCeb@iC9#
z4U8HOLs(TC;iO`Pr>)LE%eM(7ZhPzA7Jna9%wy$?Jm2XjU)BN<4vpIfDM#Y`J}+g$
z`|sU*h3$rSX!y+;g?Z?x`H4Mo=*al+r`-8bCp^VSdsnNTzbjk~&+S^#7>leJ-l&cT
zOg~;NFMmd^`Uw9#BV*>z^UTF$boET%dGmNR1EVhy!@3#Oi0A7uN|F~|mL!*}zE?uF
zg`x9W+6(G0!+FcK-7MN_lYFH1E<ejaDU5hdPNJjVY_3Fy)DfhI&pXdE&eq!H&I%Q?
zr`E--+_}jb`pgt8iPb#f-pm_oeQThY!w`~K^~OOsEFGP7s}CYbBx!PF%{dwLU%LAx
zY;Id6M!+zU_mkaU4GyzP9g9`FGHk^90nwoL`HiQG%ibbl{k?;}t89%iXX<vOb3b$+
zV{y{39<}K|RVIh?+nv#{^_*it8u%>!teq~G2~sSx$)AnOt}^087<Qh7r7(~rqm3{&
zC;J5;K@m@Z!N5Q4jp1W^{{4CVebGOrxHbWLqSC~?Edu`V%_;P^49^~6qda~y?&%LF
zKwo96E>6i>N)97}(fN5QX@T-RUF3=c)r3TcRh4~0C(AxwvzI3YZl8-WvH}Uhg>Evk
zd>Z+Xlb+W>wVBAz63){RFL_e7XnA?iTWcay_UcC3f|Xuw`aAu>r7vc25~-LAy3E;w
ze*`I3ii{<rp>>=o!#}oz)*c<3l*8t-lZ%gsmR%mTnGX-*pi(Y5UbFYRoyUsXxo_tI
z2LW?Z-}5E2N9zw8KB?Wezw9Rj_#tVZT>O5w)4EXbdZa_ropH}_p46A|a=WkeWl<2T
zdD(}*AQfSqzkK46nP^E|A-8uRrD_LRZOlu!mms1V&rmTab&kXtM+}Y9JrtW{Xl>lW
z*2)~wCkmUXda^eXRIU~I+I~mXzU}pI#BR0Snw8=7Q*dl~z)7$~jTH?wpWAZ&HU^#6
zzC8Mrft)P`$xgX7e@zDs`Ua2lQ?cNMe$k8*%-|sKXR^~|$BTOb<E0@=C9CS#vsiy}
z3D&G@d;!Lsl_T|*L#j;!uQtJJNw_CMgbs@>$nh~7*A>YK*DVpb-ffgs7QNx$6i>au
zua0orDtqZ?wTwd9a8&#93v*-3rB-sAS}mhMqauPt)TfkzV@ua<?k-R)!lK5wMRxR0
zS(@?0hRL0s@;Fd50dO;D!f_zLQ|UE)fY1wLKPTA9vlc%RC4Bv8!aY}0;l2Ve^(2cK
zzV^+4wO4X^;a0-54|w?Zc@mUXH%!g63o0TTK6Xs7Y%^!?2glU!`#qeZ{(|^ZOcGzy
zq|<4F2SMm1w7M7L{I}KoNo<xxVd~SjXm1T^XV0r16f@CcmHNe_%=+`V8=gMwdIh)Z
zl+t0*-3UH*<)bWWP)4mi!*e#GBGqw`CU$BWh3OF>Q?)iG3dOiKF&QI$46oG5VKd4P
zJwa8N=PXEY`h;GUpBbZ(VPkb3y#+?yd+_LcwKK5YR>iiseRWQCTU89VLeg#0gwIRs
zgL_ysJyuX9vrDb+R$cceyG0jPc#A&ph>s9uL{wGUbr$!6AEw0;Yvj~YA(E3S?%GD&
zEfGzltrjNAPl4U>8j(L`gYEbA0B@<_MYmxr;|O-r5;-^3D^hzWrwDhZsXJpUV0^yk
zu{Ff~^jVMseAX!^pRqXR^GXh^r<9}8;dDFyQ7OA*TK>cE)Tyv?-QOT0XO@Oca+D-N
z3NWhS@#np_7{*?3ThF_R`mW5}@p2}Q2Ct<L|7{-al{K(AYzBpM^pPc@>z=%a)K1D2
z!P^fti{+G}%rP*2EI0Xi<Cl%b;f79?)Q3YH%`|7-lAGjO(v7*Zi$3pQC`oS5J9$-u
zr3`s6W_x5#3~9tETA?`{s}bQvyS~RRWG;i(69lyA!o@c`f72oL-LyKbYsCUvch2IL
zs*ehAc0QD-o{RI9yu&xTG&rpu+7xDq5)D&6RJD6Jh!$+~#T;?{=8IKA@5@mh^x28l
zPA70}b_+vC_(-dAt<8}NQhQ0yxsbUmUtQ{d&lLgIHr>Jmbv-Xi_>chkaq&K>0d{Ss
zbx$H8eUg0E(sb7bts^j0>Gdq-w;1;>*fha-`9L>4)+=S=GAm7SfR~5>543e^i!~{(
zKkImgQ~4wXEG}92sYkeHNYkWQzo^%7ofq<ML$nD#oRN-I15t$8*tsJdOKiUwqfGrK
zXwu3#^A-WrzVS``Qhzf>X5}({c#S3rnWnJ^n-JM7{@D99NsNkb7-X4s3o)5}@!}O}
z+4NSVnypF1!(`@+Hp#vb%w-f9_3}pRAMGnk(XGJ?y38V+kOGh2Q0r87j1OvJreP}k
zjBINqE?-U*V?(qhM5#2(b@TqtFSH?;?kC(FttmHoBWb9FZb(K4Dv(sPY@o*_IJ0-F
z8kEKsCVQcK<_X#AiMsdhKFdSIn|eErYpU^poMKis0yzXmaJ8=EY06fw*fTts)h28t
zHR{TYKOr}+=M6WN{$+n%99wK8@osd}v`Z9oC04R9Rf@m0O~=<{eDSShkc-QE!Au1w
zpEFXgKIDE@u&u&bBNf1RYFI+dxtT=%a10Yek<V3Mm0hq-%@I8tRiqRsF;}a(f_(P6
z3TOPom8<BD#f+`HB}=v8hVl4s)!ga3?-xERj9L#LD&_ARESUxDwJXv5O@hm_4CS-A
z3>Wb$CaE{GNu0Bpk_m|K(A!61&NgSzfp%Xab*?PCP<y$m7`03iS#DR&{x-m`9FK<|
z<)(ZxDwzSztpna#b3WYd9Egr~+f6vXx4Z*wcP`DM)ed9wK?0^@s1aB+M!F6HD8h4J
z(iq_|a|i8>+=n=%b(Sd#71*KIVZcJXYN*FE2PJBgm=QM_SM1yYtGV&Fd3EM<*2-eu
zdz@WC8~9)Ri_ZI^5%CPeb6i=zHt5^JQDX|D1^4ySUWd<qXcuaH<~mh^6uvSdMT|6(
z{sw(KXQityhq29-OUm=WU*M(TPKh6hBS*g%@<jdikv(8Dyc*acin_#+1nkstGvKSo
zjE)|BQezi6530Q8-l5yan|H_Yn1SZSe?GYOR<^>{V(|rYq4i>wUG>T@P3+Ln2Yi-R
zd*F3ICd*?^Ig@%Zr%a8+y6z(E=kE}Z8Cmu91Lyh})fF09m!Y#dSie!Umc7noez^a6
zD>vffJJfrm&Ej^$>9>01<Mht!{ly#hp~4?`;#Jh~VehKj9sPNJ-POAjHIY^JFo}1G
zXxiO-;deVm52TnE77$pO1Nbi5m-Lzt40y-G-TPoKsyx(cVZ}E_Wo<D2h_$)<T7B#}
zm7#own=cj^VX6Kfg;o58!u=dYgUu&KNN2z1aA@m-LL^tb9Y+O)6a~!fE~o{vE-%)|
zg=UaT*a*yp*ntq}`9l%7f0R==-b}TqUkYts|2%75Q$#_yF<knk*HFgsbSQG9UiuxN
zZEe+gHYjWgj*k1Bch(@rE^!rUcbZ7Oy?r272W%DoP3MhU3?U})42Tlcp%Z(mtEHT(
zipzeU`dQpJSkp!<W^>l!AICE#*9zE$Z6##({$SZ5Q+1JuucFKr7{mT$0<oq6@7#v^
zRCa@yB#vOC*t{(U;oCI(kypB_y5&UVe!q5f0#<+Q(nN`SaQFgXn*@27(HJ>U7EW4G
za!@>PW9V6MVgkz6WbE$`nCqHm0Gk6Xg7d=u?NdtaJ*peO_}8P&SH(`na2e)~avEix
z{$}v)Y=`o|CTwd&q@5Ar01i1X71VE>BkPwQVEfh|46ey!pke)e$Gv>)AS8N@eV%9;
z3IbePWP@**&}H@h$x8;T>;`|y#Oeo3elK}8+nbAim3uRc!C8Io4_icY51Z^l4eWjV
zqDz=Rtn2N^r%BynRndl4q3;s}zjiPphWCRq?nybi8M4;apRo~}tho_WedeC~nK#<|
zuzRk4%gk2oc1UT4qPWprErRy7hnTOIZf}^y?Qq4ra|W|CsQf7|Ur;?Zn8JAz!iue#
zX!dpqW<|!N2N0Ho8(B0a=@xl@&&<}Bdb;;AEEMl<OkT<JlQ!^g4ba01ni2+W>$$zq
z0*w+6W`|wh4r}ZU=at`h%N99N>Z(Wh;Sw-*od$*8XHOs=2q^LYO(!MqzLkp{1zg!Y
z2ykh;Jzju=tT@%%1Pmty5^NTN5~v;7-0wNuHa!v-xO~3xMx}DZYLPm^TpQF%G+8sU
z9TYkj<oxsqQ_(O;c9B?%u~%Duri1^%n+G2ro<|q@x@E=PZc)y7Gk#~baPgL0)OxDE
z2LGqA$-L2Bu>AW+M5BUG)4j!E!M&HmtC~wI6J<+=kDH!wz3*tv{pJ7H;`EceKPhQl
zZfoI|f)J3Rx?|?}+tPB*<1C!g!1@6zb*cxmy$aqX#t1Oz2p`M?3x%F@idV*B{^V4Q
z-hJKr7il#R;p;M_&7q+Wa=n~&P`vR-{%KF(GpeDm?OY|Z%A67!u;MWl&Uk3@z5&>!
z4D1kC?3_+{jVV*IA%D?j@SXAp_OXSMtADBL<m!ky%+9FoIt2a^V^5HI=(288ELO8I
zc0CYXSd%`GF(Uh*R^N2GTor{vO_ErC)bGo@Rjg62!cHo{%pDEc2s(hJCv>$jL^Jw8
zc^^W_&nU^uy#OmfImll|vX`OI2xPO*0FR^j^T$_Y`rqUwLTmbGwc{NhN@rJ5&BjZv
zR|sneN0g89ed7O6_*_6`mHPsOg89qb;8drxHo%Cnq&Qv9otkTrENLN}jPs(p(HeR=
zj<`5{QIzArkw|~Dyqo7n`H(W|eE6+ljl)yNH4OUiY?$CYH(sO*FVmAC=8_7~7CctX
z+N|#WryVC}_U={nbC#2rO$Zj=ykk{Y`nm`<Pwoqq_8H4>$=?*IqRqt<9h>Gtgzy`?
zdREU1^4fH3{KAI8{yp5L#@zjTx@w#h<Qc{bKJ}%CLRVHAT7*o3VRjGmtcL+pm|eXH
zrg@6ypnI_2j-)oW<r<Xh3EW|y{r!akFHPPTsr5@)nB)%|5&J#Zzk+LYE^U3vX!5m?
zSxgVF;tKA*8aeU!fZ_gE1<KOXSA!LnV5QTbF86W=AN6lK%f3${Si@@V<;L*s?|rAn
zD{IZ38j8%9eRiN;bC2D*AkUelt!VGGH{#?x@to_eMTB(FQz;#F7c!c+wN!|QPn0J@
z8X4A3cx>Ls-FV1LXj*z@+wU5igL_oDaEY9{_Am}AWH;6~l+qT^C3t9Txy}NaU)*Hg
z?pp9s5AdnEn9%KK-J#WCu=1Ua(E<{SvZ}W&|8bzurN}Sfz}B0v!D+0YHk)g(aXDKE
z2j_B6F}`q&Q`?Pg+dCcjH{EQXN7YzmA(a*c(KVFpG(A8J4ZrO$z!FPMzw<0EwCix$
zpghYc*|e#!$1J*Mi&Y}3iN%0XnjE#-?`B`dO!>y=^5^1gfAFhfQU}ob1L=+lU3~0?
zXK?Bdu_1De_G=eXNv9gA5)S58=LXgP&WpKQHBL>h7yrq<FA^QkHfy<iD9BZz^0X#L
zBy4gsUrVMF@h%&KC81t|6jAxsM27JSVb2`-wx2>c*%HUg*r{4=%B;Oq+LhVa*7R2+
z6WxkKaPB;E2{nJshQ6XG=2fYK84(m5ZCMXPY1Y1xGx?dKvCb#d9MjuIv|&#y4RHMN
zNE+PiAwsKbqh9?wLNmJE29Fr@MzD^mgvdyJPiagvzweARvzF<gF2{>K+xp1T%#ZW_
zg;(ag7naPzK(m#EAbWXNy0tvV?uTt*3O<ec=-6B*A+4}H7WK`OApc9X-eZ`louNvN
z;%SOCdt`clMSR>0y;O{}R~pk(RIcsyA^BhZflOzU?pR-ZH^fiM@ebIlp8W_@3P_h|
zEm@z0c(q+S<#|)|Y*o9iB0;QWy?l;d%qaNvCU?d4vkZFnUFeW^pM6=cZuom>A>N95
zrEhWjLh|VkuRA)-kuR7HL<rrwV{hJMl@~O!K0qJp<2#}MntP3JA_lDJp%dKY+Vabo
zy*aY?^bU5FIAOqWYPGH9Ky3W#7`{Yq>4=peGN{%S#%q^ZFOuaDoq`Cx?J8XV^v_w{
z%51bnVf$7@YxS08y+%Bb%G2-dOrOHej$in(>-XqtAE)`dRt@u>%(}A0YnRg1hD2lx
zM^}s-!d@*y-^EjWADR%RB;YdF8J{_6Y;4vB?kn7SV%z~DF?g~{tyCbmfI}Px!}YKm
z0s~$s%R=D9g1*J~+fT*3$ti(PjaLK=wIr`dH}T;Y-%RTIzDmf3POQB!s(E2j<mjG{
zTXY#9Gk#22%-PKLZZp`0KL?x+>M}<qaBAcVmBs3;7VrK|XF}Rd7E!<j*^sEZ^ykz8
zlRwagxC6l8JT){z#&Tsw*!SVCKX6c<*1{69+<0$8u9+BwnNs!GS0#-pZ8#n$_>M<|
z?T(9HE<OqnC6ag>L(j$<CFQYD!^xgSBXK4kcug5II9NXze_y7b+!|*&(Mp*y8G7@q
zizF0+srHlJQrQ~KDP*qcZIi@e5j-|T9<!B)nX}5ZOTZ^#TEU&1GW$23BA3<k{HM~U
zU-rgCc}CY+<BhR;@ft&JJtnEbU$VcU>n;(4R}aJjh%`wZk+#TcY5!GJQ`kGRh{a3T
z`u7<7Y1yLu$J=>1zN;v7$XJ$xXK-gP$8?*wJU?SWXX@9x#pnCXi$R@d-gIUc^Ij8>
zmb0eOB8QGjnrj!$RP>ah`?Uq1#cG#}EM~|#&cDZ~j$7W5E*Bnu6XN$fDJNcq;aA)6
zbF9ad?8gUTRa2t5;F<X85bkp%$0dtQ{kbnRn08Mlcm}6ac88-fBOE6C5$9uUA*FSu
z)7!k98luSLI0oHtKbIf*9Tbu%6V%vC2={g$%8=quq5mH|eRo)r?fZU13mmz3Vu;(+
z9BC=)D}uPmQI>m;a#dz-({K;OnVFb4dn<F5tK3^t(T0>JrWW>+Imq1oZa?4S_;DPi
z{}enA_j8T&I?wa!68LEzS#^=~`iQ^Z&$qS#kFF}wSgx-kjO+6}&a3F>*sn_I{_EaT
zq=`{D|K=k`py9TlWxvu!WGwTMMUf77m1%z%zr1fag=3m4X6ymoMCwrunqy&9zA_-6
zym<4jx@Gm-4U{lW`hAz3Wq?EQ#i2_bsUi<=+!3fRx{^9e&Q?3`Aj0;QGWpWMP?DU%
z-Rl}!=XlU>OS(_0o&W~bCvS=1q&(MMi3WaV6)M{xi<>=R#b6Xw(1~qQsd>%ByRYt^
zEF^@_yGvPpz&B%)hvFmrSM5*w3y!fyhaoaIX7`d8G0<OV=Y{L8)a-94Bhpx$VU^i`
z@sA8-mhzVNp^s8m<X3n%_P7wbRrq8lp9H8QAOn-_An}lN0RbM0#F=xWxfPP0Xh4$-
z!A#_yvz~Q*C-=Fuz5ItkAWnW!`k_&ix2ST~!)~>gLcP!gqm-+4oIQp%0X&r+@5!8+
zK5ashf<v6V(*d1N#KOIKXsHYZ+N!|3pNr3ff_kgt;sPN!!$9sW?oXPF6CrAOrbF%e
z=OveUZt|Gj%-}2`iwtF+x}|bc?0tFfHt$nE$?6|oMyrWWXl+HmjI>73e$SWT-hxpN
z1l#Z?Xxw#|{E?8eRkmIZibVS>mj|B}dJ%c#8CU&f(21{ohvBCrZQ*0{^j|lRdJW@k
z$LiCm%*LP-VWf$v?i(IAGd^gG7hkmNENH2ksBtW>b^Rkb6FKe7Ta8*CHp7;?+O?g!
zx+jB0C18?iZ6i10hMJ8{tzU;~YbtAN23-Dyj%Rkgzi~WA;d9x9W5!#(S7#rjaAWu!
zPA3y?7GBwOtv~CUH5Dny8FtR7ir10rj<v_?6w-k+sR?pT)YC|+zE$M>x%7<#54TGP
z#<%I$3M1XESKNbdh^xTZmC>Im4XlF;J;F%Y<iS#D|BDo;`=+}OY4IyHzdtW<+U}%$
zA3WELDphEA_2&&gzx>F#IAx8T{*3RmRQt0mU6)5sKToTx8uyj%pO%4p7-{w{h@~5?
ziM$>97A!M%Rk>eMz-CXnxuebYOhGoF0P39w#GX`JK=F|enX|u46)$^oqL@f!AEFJU
z6#EIZVq0kAM^=-8O_PcWBXh1^MH(mj*djP>nN#p5-yG_v50UiEFi}r)pWF3o#7WzI
z0q?PiM}18}n>N9P%^Q5wMgKKAZ*xi{TpY5@n*wzG;H+^Q5peffRXVXA);bgMVN!mS
zX!R#*kJJxxpLYQQn|n}~#`&@s_K8(!E3Z<&>csnCeYk6DLULcLRG}nh6)y>gSjhI%
zzp(WG%U6lw$tXGIK7#BMS+}hFiev$YSH42I7Vn#VqQKDjcInFx1nYtRlO5Q;1HbhM
zEBRh~)g=w>Doue`V$aR=yW(Ur*rY7763Q(M+U~ho-I6xpP?X#JsI9Ble5N#on2|-s
z$GN~xD0&3R-9UM1ph(aRsJDXi(V@2C0NT}KE4)v`TFe6%A~QxZeB!&yg(%PIRfySH
zSWjw535u_Gb|Je6x6Lbe#WO+S(mduFmFFq3ja@cb)hBe+=H7dkVxD%>6h~$CldoP{
ztje^_Pi)EyLP;OlD}H~elrb-|tk=3ozWUQSA$dX0dS0JPreEy2^}U>3vEgV{PPd9C
z;&-P54lKeez$<M)22ZxIzTvNWQ@S7@l|7;**mT_FWt`+C`|z&Tbia#Xyqr>Hs*+Y=
z>iylOMevS-sF&_834wJQ&4$L|vv6rqyv14JSd?+muf^$9i=>iO;(oNe(O{ve^?5`+
zIRheEHb~%(Mhqu05S0;XyBXW{pZ%_dgt2mBZrMG^q|y#B@4|>QQ72(l3e$x(N+9<N
zEpbovH>rE~?B)b%blq?3jt1$(!O8S6p0DUbg&^Q^pTP5M%mFn+{R!8&O~l}N=(Qii
z^fi7OIdynfY{P7^+M|JCMAff{AfjN$AAdvk{i&K;n3bTy3Kg{`xyRAj&1hvVYX`Y7
zCM<1~$X>g_7QTI~cUDmLY}xcCMJgMZx1j-N{5gMxqRQ8f_yQvh4OK`o-?8zhd+)bo
zo}}G6+p<Nz9;vElu6ZFPP99Ab6|1O+q8yYOtsPVeDvM%a(<2b#1!}T`I-yl_?7rLw
zxOs}^j#d?SIu2J_@sPKVyCqsU4pWD|*2kUowL2vD%NT@4eRoHiSKhUY*9uX5@x|>B
zqkc_M3uC!#V?*?{_ihI1Rr)DEN*2D)-{b`B_m9Pw=H?&mks9{+O&m_7<r?Al;2{E3
z{%?0)r7k1q21@{`xi(Y+@r`dOEuX{_0+;oaW!d+OFEF)rVO0PxF$@M{q+!ICnrc0n
zSD^ib3MlgOrcOfE`J4@mC!AM)@-F#4-1KQ)Q`7d2%B<WMqNY)H^~z(cZp3}%Z-Uv&
z9w*;!F2e=UoFN2TX)0em;#PPnqh{P)$8#G<fXICFrb!$EQLK~P54#{(3aK#?dO9~k
z{rv%cKjEr>;;cYBq>nwa(0B7pUbW>zg)`StqB)h#XUk3*Q+`+xZKT#M6y&YU$hjSg
z74moFDgdESGc_?D#gccbq+{<gS@88<v(P|CPv@)cwP*4zcUq^_is)JE9sFaB9rHi8
zO194XQe(M*VW2cLaL(ZO-y^c=t59XU6SUlVOH4}<P+K`d0p4Q07>&P(&E4Qz3|ZXd
zI;&YQhKw)SP+T<(M&5r<F097-XYzT)c7{uXJ=u#mmT+tNb>-2N+Z9e3I@o)-Tr6=n
zhHz6iZGIdq261f8+cYkwXzOLZlaSlc1$t;gEI;AwY=eq#37cZVzOM?ak>kl9SXye)
z3RZdC?S8)Wsx;x)xz?!@9Q;q~A+i|dtP+Jbg~0H}auNZRbngYZBOQT|tJmcPRGx{P
z=K{X*CWa{JX+m#M7q%`2{!Mp~u|Kj*GDM_#ooS=YTwEf4=H1>ecqAPvJYeL&bs-wV
zup4}n_%VCS{S6P1_=n7$7{iU1d7z+xsC#nUPFk}}`IUJfE&7DL2lta}T2BG`!Si2?
zz6<g15wOug@qKK3M`tXp@k^gMy<>v1Z;+>8Z*n@tF|9d@Wn1Dw^!-60SHKfD*pdvj
zKehHfxjkI>GxC`;A`0{O$S7ftB+*NElbMI!wO72Qk}rL$KPiQmeu>HxW{5>Ymk%=E
zIru34JtE0;sc+Me@9fL|;x%;4`Ol_CU{dS8T9qPmXnP0I5R+M@l1Q=hxDScu(Pj&b
ze7nwY#%)Ft*!6$7)1i*~Xec|Tf(i0Zu75j6i%=;O5dhitT-TIjbT7UxT*4M|eXf`>
z@Ag_9Fn+_kD_0CI%^EpS<kN<DWkN%gqnjn^-3#^yD8R=?;{7ykKc2cjJ$wV2F(aV9
z$do+ax-Sr_`8wN>TA07Zt<qUkw<)4F2Btp5Ms41GPBE3m5zmA4t35+b3C(8I>=E6g
z^_)e|oDvCQlHnrrY??-1+sO0{cdKJ)w1@c()5C0OI=Y(DgKE2PiF-JJ)c5fabrzyG
zv0OMfL2#XAAXx(0*_-c-2!?<9dqk`x4=iPw+w*xLsv27bk1cZmF0c<shR%me*Oybn
zQFeRqTd_xjo6F=g^A()9s=X$~wJ<y4iA61pA?n9JWjtT8kCT~Bw<msKyrR4H45K;<
zXQIA3z7YwYqf9a?8iX^#bLS<jA8|zKjjSx(Z9BC&){5xS=Ck1WdH`AG#y_l=h;TyC
z$BiNobfERtqa?I)HLj;Ge84NSQmEg$D~sS!=#sMT$#9<sks4BAgri?kn0iV2K=+>P
z7&m^BJ4RZ)e^}%;mjTB*X`Ei{AZC79YufTeFrog%&Aa67o4+*n2?fU-JRh+d%hD=$
zB@ux{-Ba%j8FPkC9iA55(T0|zRHO?9x0OddQOx<eY?u;=8kBv=<oz(1EvcL?tTN1J
z&PlErP8SF@%ni5tF5JC1Oly~zZsImyOiyUKtprBiQJ!J&`g5xF1t>^$KnWALf?;Bl
z2&<uh=$*EYSivzu-tLnm7FUZ+*dD>dB<a#a->_LGe~<Kh-a&F?7XFa=&Ju!XVn*L*
z)$?Av6vDA3fiE`nUTT}Ud#+97m*^H!mB~_7FW(LK2M!1gXrLUhHK8IxoXM*cJ0p$|
z7uF#omF${v_{u)}E!F4Uigl4(Y>G^RfyfDs@lf9Nfws+17+*g@6e%g)&F3UplN!O<
zo;vI~Bq9PWhwI%oHbmINd4#;#QrnP8MYMu)q;z=FgY^SeEzdvNSIm_M@N5^{+ZX>H
zncnGJ8JI@TrnWU6;M;NZ{O;}`xFzonVm)mo*|DZCvhHLX#f<7%M2$(A=mKz`;+D^X
zl@$w-OgL)6Z6WkvsAJvbY@8qLeqSk%oo9=bugTX3>r_IsgSXdev`BDz**7HGdM5gT
zXleKIer)c6%DM|v=%A)!vPNyWOa=m|Lv?Rb+Jp*q|6?~^g>iBjigSeE_XY9_21y-n
zRG(ptfgC$o>Ktv{(;0e@M#&1(&sqLc`IML>sU!6{7biRY%jdJy5)mYLptTV)Vy~c>
z;&S2K^T0ph;uNcV6d+p}0bvHoxW%U~lKZ>KkGoHP*kN?p0&^L5KCen{g$zmYtPd?F
zi;nwF0S<cj{SP;u$=!M=<Tm}M>)}X>5aNX2G`<}gjBeKJU;%`F?uV!-HwZJ58(6%6
z_u~iwPp5){(x*J;C==TiBx>q@k#9EVq2cd+dPE8lp{!veE<7JEi4b)Z@95F*ixsEr
z3GJ`}b!O1s%iXt}*>6IlF5wTQpefI`T8*$&CnA*vcDyrX0yC+wEF7FFVk6h;IYtmI
z%Zc{s=XRIBGm4NSxTngsQyAE@#mQ;YL2ai~7-7(5o~Q2!!k?>DYnHIVo4P|kv%|QK
zvd4!>F5eJ?i(vAvMckOz0dwcJhM_lHB6{3ps?frr*80PtFbDjEgXz?n4B*00xV@N8
zZV`I${?_2F@cO6>bXYs#T#Wl`y<ukhAIviU+;627L_R=!clXnZ;mj8k#NgysipG6I
zLwj2r{@K`c`;JxljxC|R%YTo2&u^+0H>Cj9V{#WEU!y`n{_u9mheS-K%{L7%yM8hI
zx6&<Ji|k7>Su{~L$7a8f{5kbQI`|(ZCV)-8K0<~2bLVu%(4tEHE^Lv6=nR9j@;XV_
zqj<8?OZ5kUX%GShIcIB>B5O18+;np9HrW?kOv>{~a&E;|0-&)MBH$dtj4rE?3dbyi
z=^^e-^Y+Zr03wO;cq8p+;mW{_^8H?<F{v@gJnNzDKNH0edO_*q-9~d*hHAzA*lppC
zy@h$kT`zfb?=bk>u%@dhSAsE0(?C*8<Ng+|C%%-7jF&s(#6y$s3M->q{zqSJ%%ge?
z<%}gIuR@y9yu4c|zHcnxYyu|_@R{T#_z3qZuUxg!MqQ@_p1&1_-k#W0*Cj^Z00k<T
z(U2i$O71<Y@RHBbXKOnXZ;XZ|Fhe9CwICDLsVlu6hwQ8N@lX?~!=wIt`;(^^;2vNj
zY-q(C$|oU-cupkYar~j0(g~6d_v|>ZH`mo5Rin5cz)wgSJfK8~%65bB>iX~u1$l5a
z6{X3WLCnl5u%XFj^I5mE0V6%*fdQtii?;QyGNDwLtb@(U9NO4XsSIc3z|$kxmrCc3
zg{vGeT(dURc4c>@8E3Wr9-(|lbV{nW*mfyvFl6xeh9XDTxM{{T<7PNd=F2ChqXkzD
z0j_}NHK04pla4W`xE@sM4kC(80rL^oSRo(FI=KgAz@oW82y0!irkd^n&whL`@b}0_
z^ihhM0T~LB0xBA~=%If`7npKciE|r)C)}>mcyO?{4|>4&LmDx0#u)AX0D~fv*^=$g
z{MPI6sYf*8<ibT3J-5}UYHdP<4c&4&{vfk=pQ~hxWPCSP4Unnt?&o~plJ1E`YwEg_
zy0%dI`;ZP+#y{;VHn$z?W~kqM(pPh6C@No<kde0D%Z{@#pyW-K`mJj^FI1YJ&;;yb
zGMQ3g93ieiBnij@IqSHgkcYsNfJ`nhTW$52zt3vuTOGhzTIIPJo}UXpx)$S-0c=UM
z2y3zT+z-xL=Q4pr>l#K;QN0|lf#Ts#PO}4wM_XGgIwV4c>dbnI?I-6hwpAA$0?P}Z
z-WxDfbsimASMXTHuMjQXbLC!p^l_g!x1AFHCVL4sM-flTzH_kv@z}waPZMpoBza*)
z>-*!OCBm8Yx{x<oc(XVmV^F4{H#|d)ULGE`+=>r>j}SsYVPTycgn8%j*?Lz#Mnq9*
zu{;##yAE@!64hm&SMMD0@8dG4oLT4GRto1t5?P&%gQv&(M5{#hf&fkD*Ivh05XolO
zt{6Z-3VUb>D$;ULZq^I~ri{d<>WjO3%ebn?BR<F+*lFDPI$rrMep={1U}m2Fd8}VY
zpU79jc;le`2-v@i%Y62GPqo6a`gAb07wyhY-73kEZE5IbnpN=8q!~&3kjxZXJbL0a
zFc?9*`b909I{BF$AmNZJS3JkNW0TDPGrbI?!JH68eT)n9(0B&F?onSeeo@aT=7iy*
z3x{`~dMnGE*B(7=ggV37q6X+PzTq}+rz%vRCPoeblL+#D<E%`Pl+hCO^?3bmU3607
zLdzQwzw(f?<uX3tYVW;Z_E!2cTv)v^ri^Q|CEhxx8JQHt>_}{~#;wSd<$Q5v37<_n
zTKk~$amuGOdQGg&3C9(Wv3+8=7v1doq!deW{K9@_IVQ{n9fMw}+Ba{X*HRAEWZIc5
zAms=U5|JHm`SLah%24h^w$x<s%>a2*-rlz?0GsevcmXS&lK3s_!!TkF2*hJRC3P4}
z!S`ouz;d_ZUkTK(QYji{!R0O_n=Ogtk}0u{5I;VyVzTK&L1t~AX^7p`sb+Z6B4Sz1
z4y2bsDxhg4zc{8P_nUnsYW;I<4xsT*+?9S@0x5yejd%F}p#&Pg9~KS8MVy_<sbfr%
zof6W19)O_E9N*VE0d1g=0#ocDrqk(#PSCgGa>dll5sdg#dP03n+HGxDCwYNZ#m<`V
zLn`^|i{T<arF$l%3Bqd-diOa%18n`^MFcNW?>g@604P#Y0gs(kaeEvO*JoAfE?Bjr
zM8L=sBo>C97r`3{-Bc-=9W0Z4xcWQ+P9dpHtPc@}axAS$qG`_S$|kp(u7_LpN7f6;
zdgr3bv1n#AE+mp$%qG<1mBwt&bJcGisfsXe%i~N4bQh%fj>T9C7qNuO1y{M~lK_MB
zxk>5ack~`!9#IfgqAp^yHhA@Kc9_37dC#I*o)U<=P=$EgK7KA&IyfvP7-xv-q(IKc
z8R>loAl194>zZT)Gf`#*v&eJT{fCg1G7sRAfZ-AFaRBj(#kXS9jzUTu-P$kwDG`B_
za%upGQJ5sKZ;&^mwX^L^$wW+sf&3dTDJkn7dhRmC%8C`(s{EgElFS{v<~A-*-gLO1
zdhOb^92w)$ERwP7Lyt9=I9~;qoR=iGW7dvQ|CEuH(*=XUzmh;nXd^2^UA#i23wh95
z5m)Qe#b6qipgN3?X6y5gRG<FbFw+Nm84qN6{$i+kl)2AKx&y1Gz-#_5xMM+5Hld2O
z1+T*KHKU(-XoqpU89@${DTG5;gN!CVi5bQoiB$m(*KMb!d)=SI;@z|;bPE*^I^t*E
zXWcK2Jj(hUaD7lC8H$4Idk6aOt~d?-gguapJ=4?4X^WYIK-WDT)Ut#v7DsRI^H6?-
zky{mSK4JL@*_AB@{Uv7AM#}T>1yD6Z39kCC4us_(Bi=h_<xGw;gnqwUsrQ|9c|HS_
zH-#_i#s1wz^lTaut9<r^Lu1PZ_c_1xZCVnjFc$A^l{TG|4{EL!WC!M6Kw|G5UI^a;
zw302=*5v_z#bj=Un3zP5?*M3*4p;!M4RoUi4yKS+&3RwE<c%B|^W{6Oi$fyNN*&cN
z52hsk7vX`=lYGqGg=->g;QQiRy=TBZA77d{Bl&2X^3kX6Wb-OtskRL}K0+YGN&Z?f
zi0V%KR+yC6u-;9~^g_vEG~I8}Ifybner96L&}VR!B(pEFDz(?stJJ)HyS{x;VUDpO
z6rpB8gh5tu$7;^D51gS3yr!*l?xwBll$eC7ZliO$rF<<DIfN$<fVd0r(>mgBKn$L9
za^FB0qR`|;7qD^Q!2a)%l%mGqY76LM5G&uR-K1I!q0vR#_VROi7cb=&*={vpPbE8V
zKbaT$Pfq1<tIvn@bS8FEj(niY;w1b5s3|)&<XPpN^KaSO=&I?ZOsx9dMa{;fG1fiT
zsk(&P5~($XJ7vH2eR0qOf}0aeDvY><-5-2Dc(?YyP-a=h4&pg`8RjdGG6x;8YU8z6
zuq5O_YQcs2G_R|jZelpeX2jszCCE-t9mD0-V<#S}(Sj<B>;<UNr0*>(Mr^5vA??~>
zDF8~+J&m%6bjS2_52#rO13Q4yy~^QoAn$ellc#4_W~z)V5l=NM+n&3>ZCx;An%ZbU
zci|W}FNW;svWBKHY6fca#@Ayf>3yp6{2(Soz*>X4T#kKr-j*vJmu;0T6(@7mIjOb!
zG(TR8m3{on7YAx79}k%@D<N{p(RYmzLKJ^;+S#h176?i?A6Rr0dK(`IPv!|6)^cUy
zQSu>1Z8p;#(#+9zHo-qWbqu50L~R9KHo0fd{9oWYay|c7RipQ8{}-Yw6t`6~IBwVy
zWYwd9j;v}ecD^pQT7SKXy?~M9%yK5BeA4%mBPh8ZZ1Cc6Oa-c{ostlKRfX#Rq=kX?
z1p6p)XE!9pg?%HFmq|&fm=HS}xqW1*#M7}Ph-!zyp!skiSf3&^^vh7g;J#2ZC^ZnB
z1T{=V{*5{z<<O76EVS{hY^@6aUB-?~3tBn-u+GaJDj|hL(4K&~qNK!4so3{rgDTuP
zc%czpe7NphL#JiQy<9O88<?tlZ5qEAV2u9s$@+t&N|9JB6WOc_u5Iy79<~m^r^&kD
z{}NFF_@z)0ywp4&cKhNd-igXoJ3fdm`2o<LiH`Me_bZWowBcQm4PLp}9llm1eKoy^
zqscw;6|iJNs!;9X02r^`agXS)2!8WG%2vP7?}}YjaU(f+t}~MpL2bdQBU`Vdx1=ZH
zK#YXGGn}S4(!U1koZirsUNc@p^0%KP?+bNP><fmXz(J00PX~E&C`Tp)IO?Fd<AqfL
zZkP4LlZi7z!m3JxJX3f1XX|GOIzq06>(Yzi4_;T@1H?ULyzVh6%{JsiJkSOsPlDJX
z8xJlNy7<t#1_UD$pi7@90)}>Sc-d^f`>Ysr6e|UzGLHwPzO^rSCtMF_qUxSj35YG$
zu?MWN&jXzY?BC8&PnUx))&;tsH__S%ofUsBRL2kv3Z=q~+o0O83yND;4nHam#buD=
zu9EW;KDe&lJt$1SRv(kPgMC~xJiQy?r26rD>#UlQ3<a6OhnirjkatwbzhzcT{Qz)x
z09jEs>YM2ULsvlU<fi#gW1#-T)9Oka+9zFK46Lh2X<Z;S)P@%#ueewMST?;`Ih<wY
z!Me1sYqtWT9>Izh2F_fqNKCYz;iya$sdrh*@1M@<_#ajv^_YUa>Hv`i&m_2$vG+mE
z;qUf%Gy^Ki7Rc$au-35#*Y9@!dqnl`kyOA#wk`R(;@tP3h07GKnu_?(KD7-{cSJ)E
zQWG{^SUo1nrU~_`$a-a}&><QV;!*KDR@i5CvS#on3>{9;OS-%-(4Sc3?`C&@(k3@l
zj4Swc5%2SLvv@~IX-fPa{Cveb;tP*;!yn{FdPz$|ZwnbJMeqjIn_6^>?&70cR<??-
zIg4uBYFmOBX0#u?5UjLHF~>*qsAH=pfQ9#O>Jp@_Y2x*-db(wAbk?Cj7tlZUY&-tv
z-tfO?_J4ovY5d=FD<8ueRxqV4JQ{{G-gVtQ@#v8S!+9Vh-IYKC>rq?)q~0;%oc6Ub
z_IschuVY8u=Q=R&CvHl0JAc8yuG@4+5~0#vc3@hc_T1b209Pgz&BHXhR8-mo%y$}w
z(}MRo6mu_aw@iPJgn3JEZw>}#HS?;{4$`_*DDU^U_<^0qWUJgj4n4B8o+_@0dPHJ8
zD&pJhLy|tZR+VA|dSpza3UBql{<s;Y@jAHVWE@Nt2*mi9Qu8Vl7z9h@;O{s>4IT*|
zSI7K_{Xnt}C)l8pf<k4R80+f&<Iwr7xBvf=yLbd^OCX@zvIw@dE}Qzs>gDAPx~#!n
zQBRrLU6__PXVM!w0l`3Wl5Tzty0_EqbF^cC;y`Q*oudW}xVr8p<z}Y;$jpu90uZ*#
zn%(w`y@CVYnQO2f+`MpB*30Ur;Mlyt15F)Kc<F&E`S*nF8JDB_kjgt<kPZ0b8-1~c
zG5wt1_s!Sz-IvW6LrD+IPLn0PJ;T+Is*}>Me5?h4S=+vG<!|VJd7E~j`voKXESUfz
zrKbz>e3EU%-PjB#z;$ybh@^Dv7`MsEC($stIwn*G3>Q2g)fr}0b3<H4FRLr;T?15l
z(Q4c!qPCaG5DM_9a6w0lY=N3U-iQhMN<;BH&YE%MhI#4W2!37HRT!c_jALk~<1B8r
zgE~!RoNcIn3-ug&7xd&bVxpaZK!u1E91mppH+ee#Jp%s&xzFPh9}Ko8)xVGN?gLGd
z<Lk@}ZKLy5(}hlIbrJ{IvQ5LgyD;KKc+?l7__Y3+jpT(*?Ti?<TcISka3WRP=>!YB
z%%G}dVn<*~smq~&zUf=ZcR{ik7IIh6VX<enZ-#eQ=paK92N|ecMTqJ$2RvL5*x0hZ
z5=^K^vhLLMw%syUasQ)_)45B9NuC#c%t%~2+8XJ7JIG{5hE1EhCAtRc2!&Q`Gp@CX
zUf_&6akJb1Rc8^y=7JgZy9;VbXsEL>XfL&Fyh=CrWGXPeK*0Pr_$--|q526(o@gNo
zO%1-@FzVt{_iYH2xh6m;>B72&OTd6BP<?$mG1oitKNA-2uoQY`aw~v!jkHzb>*^eg
zjCU35d+j8{IkdRH6+`a*r?kRqx7+yhk*{V~!>#Q!hIE_H>yl$Hf_{JS8cfwxg7s}%
zr|h}V{K-BiAf5+z=kK6oD5}N}{~nPbFw4VsP{COI@8S&D?Jh`43*Wa;&B%l$B<deC
znjzi;pGGH__{{Y%20YO`uxu5KlL%Q`7Vm;d*rL2i0r4lwC@oauR?@WEss*7o+$+2=
zV7H+gY52Ti(_?(v<^DvA7`<|RG;6(1Y884K#v<(~iR1}Q7v4evSHd8(#ixx}-=rbz
z3fF|C`N4SGF1#*rXh&QBUNg>7Xi0Jd^B<;aX1lgU42g~72}uu0D+TQ!&sCCaMrqCT
z>tx<+l;k-N@HC7i4>JE3Q*k7{SbOXfNSi7(bGS|6CoF&~taw$ygx-^A1EvNacS#zl
zi@(?d=Nev){M>^t2wCw`Jtz@IVc5!)PXZ&N>l$rknK(JH?wKGe0DOy93rlvsx)<uP
z|6AIu?)T<3|5G>4IMK;6+Z#S>rZ<a*RXhZ*0Cz2+fjivgP5dHkr=X|#drpsiw`9|l
zEpw01BC63BPhC)Bt?oM@>I3)x^{k(-5+rWv^}byRXUTu}w5}{Syq|RZm+ojyU%s$8
zn4&XfJ;PkcVxYaM!E&nC)t$5kcB|2e&RptBL!jb0DjQO4a)t?xNEb*a*REyj4J}rF
zPhOP_69`Qzgt;S0&uaR(41ahwa4^Npaz=3?O7*(tElIxZ=*_`(V4{>pW6F_%Q=4dD
ziDBsXnaB>GT2hK`vS%|=Hr<jyas8~)|DUOTws~o7QAd2uChzUqO$rMd6qJo#PeLw}
z_28zXh3e&7*3Wqje%Kk(#V)8V785Sc3^VF40;lIQ@hs(GlD_0kj9s(><4x9=xLi_I
zslolu^wu-YH47{Kt<J|H5EZ4j4_Ze8C$C(!xG0rT-Qqb^xSH}yX2xhR^b4!J=JgEI
zNgEDyCMr{P?XK{;WOld2E$<iR))^1Wy!eD4Z1Row<*Bls$+D>;8EW>t3_Lrlzb}#U
zJek^_qNRB5`J(Mintks#VSMF710;0j+L;PxxsMzT@4EVJgtAh(S^UP=N_Om+@Wy6%
z$OXh3vC(aJ_fUN*cgbc#81#49L~_qy+OM*Y$~dJI$A*C#XH^>7jF51M57hc<FiP~x
zx=JV(#mqYivD<k?T*sLgJ!{6i#y5BY`XX7C9goyUS+c$|w%Cy<JYGBXpzW2iyCYON
zjRwkbvkm+n->QXQ6d3|erhKFfsI%QeWKnuVlzAfe(tP1E?3ns@3^)anp+0R)7jtsr
z?1PObx>H^JR2W=m`k|>xM;08qhWCsn_5RPT7wrV5T6qH8CO;w?dRK+4PG+@_y954i
z_!)p-($mPw$p%Vp5ELi8G><d4Od<C8en7KNi&g{r^WEd`HwE-{T&A~FgN(hYi!@a3
z^U6!11p{EtdeqJ#!4DUlP}I~(QP&hX5^HEguP$Z48D=6H$m{ipw!|My=_je~ws8O4
z&iWO`YqneGr4%p!3-LbIw4-L}C~npHVXfiDrbv#r%h}J&hf=Sr_X`G_kpg#4v0LA%
zf5T(Ikjf&l&Rnqy1HFjYQ2|lgVkkv;nO!M?Wunvy#s&X2asM4tF-im!z>~RUo}no}
zIZr>AlkI%>JGN|gN3n_i8h3E~xlm_uR#EFg;@i(myE7lNy4v<hJSy?0CAB4&V{r9_
zYF9i7YZfq~u)-vLn(99q@JfIl5$h#r(R2$uUy_b5s5(5$gnN#D2g&oP6kTjq&9x{x
zXji-hmQ7_C66p#K+1nnlt2y~h=`LjmktaU4`l&Vju$);2z!4aRsOf?h@X$yp`5~|Z
z#8M6)jV)CU0qzI^N`z1)!DCPkWJ8ZKSUO&4|8{o&?PJpGbLQ3sCb0;a_MYZ#&a~**
zl~^4UryfP-)-x5)(OdzqM{KD@Ms>7KkFyZB3zPAq)=ul1KYh-HFBv2+YF+xR`Id{;
zq0No_f?v4C@SbHqu}OiR)v-wzOGQ7cbZSt`lP(@8%^7kz*7KQ<eCHeUe0<&&1*>5k
zb?}Sl7ZiF6+;7OIQC<={^ziXj?hY8`Q~lO^*O-@Sz=FX{BzSqL*O)9OjqK;ntp3+@
z58Vvhs{K8>A|-W@16D9x2Yr`+2q^l2?d01*1pwQLXKh-fw5QR4{UdP+>d+i6LI0z6
z!Q5h|?hTiuS~xtu+lnO=>yzT;1)Q|pVnJ*tE~{9JX^D^8%#MAd{+v{f<E2$Z7f#8T
zr_E>c86WUhY*_>UYLwskLkKuE8Syp53M0^gmmLLQ!8`;7Y!G=uEE9mQk-X^$_@hc#
z7}yVf<>mgTfVf#^vGR|=RUvaGzZUK*fBUC^7U2r!R<Qal31B=}BqeWNJMhovG>@ow
z%Yoybp_d+N!{Wq7d~X%P+58`N?d{oP)o52M)$<_lFLwo%`ya&fyaL<e>N*poHUksn
zrOW_H6EZD~@{!V*=fHlI^4Bc5R~sg@vm+NDSRu{>g+s_*@}|j*4M`nYj9%})rgJ(|
z))w<cGK^<4qXmPy!dT}b`&ToS+7fi9*FD~r3HZxYtRV*r<f4;*yr93>N?VS(Kb#tj
z-S>6O#Hs8`5A74Rz36Emah+zL&`XRI4;-V&#;+Q;r^-De*SC+9w>7k9QTZNc|A!-Q
zeUx%X!!_*R<rpI*gi$;xu#%$?BP)zwjMf66aa4GzS4?Ri=1Ax&H-5_l3vQj$VN9#|
zuw3#;rWRO=sn+eOdg~QzX$10L*R`AXQrSLrJ`XMC$6n|^C6MRodfNWrl=Q{2!gJL1
zOHJVs*IvZ+e$u<3MvZN;)9UPOPEV6@;G0e(cz5-_SG)Sla^JW1MpTNFR6FFUZawpi
zg=(osh#*`cD$%Qa%D>KJM{vn%E7p+5eRRs$1|OaBC-3`dvEe-Os_MV{7gdQ)S=X05
zx>t>}9XYLy<wn|07PQ3ZJlz7eUoOi!j{@b5bWN92iue~w*f(u*?P9+=Wt5QF&*EYi
zJ%)HARcMZu1K;Ht-%Tn9o&RGR&6&F>5B|(XqLq0PC~k~Mfr5iV9my60P*mO(BJS#l
z#kv=aY$|KU5TUXE{AhTr$70X1lMA;0Mt5(!>%@!=8>f}K9^No`Hg^fQx55(MIPXNB
zDRsag_92T2%d-lo)i=nvWvi)5MxAp72bsHsy)yDcZij*oz+XxrxQs(1JC`*urco|%
zC=(ThWf_~7^u)%G{8M5@j!BIFG&iv!Vz%lc6)5>Eh_jGf`Y%ZJ4B3*-B(;~&&FTe<
z^ER}h0J-MARb>s+M^(6!Drq8`s3G;+Zj*#vyb|<jQK5q6cTxiB?6c3!$$u({KSj(~
zH%}}SUwNQvz;2BY;Ry`Z@b2l+G8tnJyn2x*uc0==t&%iq&^Tji^hH=If5?DGpj7&7
zEIp+^a!OkGg(pfQCCWh_Q^_jQIZ_s(peBZQ>O+C5<nIw6u$WlMqt~QS3ZBr=fz}g$
zik1j&-(u@j2-B@B7kEM%!L0%;pH;o&-`I(D#LI$DD)j|24zEmPQ(#9!B$bKU2nlF8
zFF*qDry1)iRHaXpa0r->CBjVd`yV0RMYv$^t_j`U)e#2rRH-#pik*GGUQp@KHH2WV
zC2Nr!l097Iq0N^JKjt7#6!cl?p}_q91na+OUI2-=9**wyPkRE`=+Gq*G$T3Edp3K<
z;OL9W_ON0MJvcMyWG>u%N7wbT2q@`mH#|Tfap7S=(<{ucVB?}N!AnDGQrDlSw57?#
zvMG97gnHL0Xf$o{Gw*0x)68iV7v#RMKj))lq7*NmL%;Sz4_l`^@-w|zCO9XwXvkYv
z8_DS)dB+yvJ9|!MC8{)N`ao{hcXv;V_M7IuEuXO?$r0#oHTZR1*IqT`0qolRrKIFf
z{R?K!YJIgYWhNs}*!OjQ&1e_rb%{&iY#PcaPHlCQ(egIimkpmUVm_<_PqIqS2>|br
z!F&uX5fC$8hoQneSEY&s3VA)@hZ8A!MTF<0GoOdCiS774MC>%KhzCv8N9)sKCkV$U
zqI|yG{}2u=pWBEdK@HO=n#rB2DfL75b*GGQ*?gP~PkhZdy{FJ6mU1=HgtCi-M(5Jk
z5re6`O)%Wzqq%jDvV{ZaE#6THEu;j~&i$1NfAT);^3G)C;`S@7Ki&kOpZ~d(iN3Qn
zkXmLc0;O$9{{PU(z1w^CPJxXT02)=?LS657?XqG`NLfAT=yvb9?7>2<Ix)5Rpqv5!
zDk%3`bu{V4H?&Q>6)xKg+Hh>^!I=ZWgNmY0<cN#^bylWd&iq={;C{ynXI@9n$3L&Z
zjC29}T%>`{8`{6JagF&Ho)#SefthQlk~qo8C(Tb395_=`|4iHFdiD=8Nd49RWP^Y$
z?r$-P`CdfnP^;04wvoJZXO_#S{*7{Z8|D0(@U3snA^F5OC-F@RRw?`{vb%cq@@FL^
z_Z+p$Z9*NTIr#w%mkrMD<wdC_T;X^erAq%~lI|Q}yD9KANN6GUy9eE4u|odpgRHzE
z7TEa2T&1$0d$o`$CX=!#o1?!YdCR#^AUJ(B1vqwe>-a|Eph?HH4Jm&_+6!H>iaKA*
zhA(1+og?7qN#k`ipE$m5BOU%$$_h>!@dlDEfhw)?HSb`ta95=ORfOr5qM~m4rjgwq
z`w}4(r$qAE)nA!zwbCYsxI8x(rgz;J1@7VGE%{iENliOomwN}tBhGWk9*EfAlaINo
zfWVeU3s78)&|aZ-o{SO&>jU!`g=9v2gedfaN}a`{%e7S}l=VG+8UEm8T78N4w@-vU
z1`efRd=^i;t_A-T2`oD1rFlP%Bf%@v>skHVu!GT2>i7bxd+fc|^^wA1L^#x~jwa|x
z-}s#tdA^I2F+~Hk!iA;V#-@p*`33hixucT=h01(Y>I#OD+x-9KL_!4x`vl|@6Q92n
z?R<qy^HHdKboA1@<!nZVrglIFF`qioq;31hRlR;+{U+VhhBnXc-~RYxFzzCZ6RbNv
zG4NI}VgjF*s%4zd@;p+!*;IPz`6uCogL`J`^Y!VrvpQwxHD_J#H-$xKcm?q8FE=~B
zX$-z4trU~k7fI+QQ%p6VUQ}f@HWgDNp?~ZJ?fqhi4dP@Ri{BR8T0S$OJ}9hh@1Aq1
zqY2EhSbb7zgZYOSt{|mC8ZDq|y8ISAWR2G>PAmoNW)<?gANB$Z7Jwypd?3}SVK(P8
zX!3V>LR?!d;DQD7q|r^B6klj0DG(QZ@?jPh?Zo|X^acbsU=>nb;grW;&J7mQ;caMd
zu1Un*mfsAgOsiZr0K?}g?QTGWpznOhtu+7da4AWiKq%Ie&Wt59y&d^VEs2YmP#6E*
zv`*%hz9!bgq`9T=y97J@$;9iE=Ar}}&9x41u@?S*3D@axHP|l<ACdTkx~H|JM$Y-!
zIZe4myZAI;V60p<qaYr7({1-BJJ05qMB+=|+0~Zd4Uvc4p2lJJvMU(R`8UXas(;+R
zy(LusE>JaO{0~A_&U;K$lXpS0#OOveyoIS|^jX!=u`xDXp#0@Ir?<U@t*djy>u-iu
z0|MCU)4vbhd*mK@{^W6LrsXdvKkcUOhIzDud{@C@)62?OeofT0L&l1djn&*2E_D~i
z!n(@+y`aS=%_~!9@fok&QxPL$`ZM$1<8_y&u`tAEh6BV(tAIf1q_X5eoi69A7Ic@C
z)@OdIoKkEIYyY@o>71`@QZ5@OJwf7rBajD5o~>~yfY<>B5?Hy>Wn~wyb7;;iPNc$O
zF>5;lo4R-vHlbUF%P{zH$J#h{qR;^-!|Il>kq2Bms4yHetYJrKU6Bli-_F&flVFP4
zrjkGI&covexMDi0RKP-!{2vS9^29^=5ReaDvA!e<`0r((fX*GHrA<V`pm|U2YnG4b
zJBO2KQ<r-kY%Uq{rgl?`%cO^(l@B!i362Ow(i#6o5(?Tm;C?J-Pa3FWsIEkE4F<5a
zd5qj_$5s!&*fVzEpOK&RoA_-;Ck!L2{b$nyK2M}ZPME^|V_DZ?iEIsLpPf3j8nDM{
z(473D^-^S4rO6h|+v0>~UF)Y+C7ZCpgGpN-{o{ALsiCEkQ8zk%@)5>TXCgVCTyzkT
zZNAj_%4IC?w2}M_oU<w4P~zi$#^gr>CCk$ng>Lm4Wa!eWGzSOIUs<Sd@DqBFZYsNP
zkox4QL*~bnM4fv2+@?s|SNsU$`mdvpKYB!*800S)eDpH+nc!UZ-y<1XJ#LE08X59%
zW&9vbV@jz7rtX&i;a|roSBKQV8{VHNY(Dw2aSJ=BSjt3AK)q~E^3?zv&CQbKi_!6m
z_%m5q@#rm-%36D3Bsw|=C=B(kt?ZMy7K@GdAOaU8{v^=bFN9H|%x1a&_1|v!I<TKk
zn)^vQxhte_fSE*^F{LEWH3J(9LlRHS$pn%0%y)dZzBijpXATh4T9$#Q4v)|kCBH*{
zM439ZK#A`&A~y_p5u}w%y%t{ecv)@^Mu_LGxM<Jihoq;c<7J0=NP}s06;vTlZr3z&
z81HH{Xb<~yE93x;-GC{uc^^9%j=Q)d<mN;&C(U_=D))?rjSJ^>%c4RgodkRVfUY2B
zXizFetk~2~Jaky^%2L$i9lML}9SjOqJCFavS92cT&o$dB-AbO%ShRnGthTyf-xTJS
z9{wf*J~qT|+QHCJ{%I|KR>R}l48l0_SqfIwy?+0y%)0>H!BB<@qVi7=I)L}e;|!th
z3s=e{Eu`ZNLaXC+7U&Ug9>)K%@Ah=RBP{SD*x>w`n(}d0u{NWw4MW5R?BCmE6poAl
z;!T}<>zjt^w%PIO(8ZjxYly|j8lk<EtxHtcpyjQa0rO|iw~^@jO94vc`<EmMaZ*I~
zTNKA8OS~TzHF*oa(U))XwN^ZQUYIjd<jyBvV~gyPWgWuVxu+C}i&zB)y=efzW+94;
zhpvd4&q9oZms1`HZJQ3C6_-=%E<oQ|zMwAa*dsAatMyxF{s=X#hx07z|2mbo$M=~T
z$u>_wK@=P94%pe>9){YkSY@4awpgF64M|7R?a@fCyYx&x`E`kq8eNba!(ayCB*q^^
zz4B<N+o2*bmpSev@FQc;2>XPG2TY>?ub>QVaI%!{0(ns{|G4;H#Wzr6srVQLmtqAp
zFV2(K5^T>RCD7eIq|mS<z;0c)9<X)J42~ra>-BI#T$8&vUmQDNJy6?Yx65N~q%<2|
zgG&>0?A2bT$(bl+C79i>I_7SoIXNPV<M4k{c2Vs;tTEZNl^0q+xmT}sSDX#qI-U41
zQnf-;yf~rfm<eAIy9+!(`4~niYf~ieE7atgiEm#;4epqQmK)ENz^@dxb?q=pdi+m8
z&dEJHhcbE0dc}u5xx|aYWw-stPX#^!9My(j<eNooYh!7*JY+WNES*hwQM@zSQe+L~
zr%DHDG3xs~DlgND-OhuW8W5?_7+QY;=(xG@XrL{t=Q3cmNN{9#fW?_^A%0lWi$BEg
ziUa~5smd^}WBaf(fW5fP6XNZs#|LWdL=Hc|6_m%hlll-*4Z5FYhh{Nj9XY*BXE^O3
z>8pwhJ#l2`tRDO;Mq4GW0MNpnMMYOaH4XE_jm7BUm13g)OBQl7`tCkZm~E&Afr}~Q
zzl2K=>Sani;T(9h4?6)DcvFNC!{7B)R?P3^&3XP$au-+vaR`N-7u`I6>pTYnKQ}SU
zkmJ)7*piK#R%xkm%)B~^oEzdJ|3<$uUhR1@y+ewNe#{i=BurFE?)o&yw&GAI2r)CY
z(J22-dj03p4)ax^QcPv_-y`?;2DYpl-e0qKY-sP<4f#qE=1urr#wV9lF_uf-sxyC@
zOVs0t7wbYbczH~F=A4_pe2Q%5^+Z|pLjEw)-|FA8%Myy_xjqs?s~+KkpR|M%lg-In
zuQX*VzoyKJ@oKT?!k*sua1v@s8Jn+DkK=d0HpW_;s5zuyBumODS$Z+LwYO)B0!UJ{
z8dJp<OJ|a7+R%-q_a!c%Kqas}x2om?=1aIVG?=B<P@PDNp}o$>JHx=&p)B;@0vJPG
zE~V=qr|u}PC^h)z{2;jNUCX>?ldl|CRZH{_gb^cz%oui5hLLRXm3GIZHuXW}r;nW!
znU;w{qPzy*q<md4cPboExcYF}Yy>MtUKFrfb&<(gA5l!6iMiUW^2vUHmpfOlOMcGm
zSDQ_VO1b^Gb({@}wgk(q_&ZX$Gl!&B`6%L&VG_^PD~)@HH}%2D4{e%cP1Q(y#2<J%
zksi*$8@NF$udbob3d5&410E^O8GaIlPCS2Bs_|Z^rAL{La@(`#ihM90t+FGfFE!~r
z%J$apCGncV-y^k8ip%>KmzqSTyomn+m58dpb<SeM=SR^TrH}GY9?io)$S-=G67fNQ
zun!J><$GXd!y=zI_&fATp1-vqcQMg*Bs%J{#p>lo@!^YX^<Ow7Zv+~_MDHF8dE}cI
zQ`&j6X-L2y`A7`@x?9y-@fG6Wq$O10{-@PUcDD#?OFPN9rxuWcnOaj5oyD1P1HVwy
zj>)^x>S1$+lWC|}c^a7W;-ujJ+d-$GAaC|uxd3!Q#@ojVx~}Eui>!fB8Ca1Q4`7*W
z{bU2sIxR!D4ml}GUh>82RAt62(T?~hL$^N6z4>NiL`rwexQu0P2-@RNY!BFL2+B%i
zwgC+ta>#o}8);N@nLROe8GuATP^DV^FNT3PI9&Z={`O@3?}HRKt2!m_>2u8=Fk1EZ
z(*`Tn9r8lro{!%5j=#F{JU#ofSo-pZ8cd+rz)8IYiyim3t)H|`ZJ&OhM5z1sHu5Za
zC3AI`&)A}*%Lma0HLq=8E$<2Q{rAIIN}#2xoC&{+=fL_IJj^=eTS`|^EaHRiSAN$A
z!<J|dAzArUvJJnPvz1AWp4Xap$ZQ?2gJ`S&UTi7BA+aSyminz%tn>Un;n|fS$uSe3
zo4n}CT(V$>-hxB7zt@+`d#8+kpFB6}hd35aIscPAT0uAP5zCD0ioXdx79|nrar3(7
zFA1`gAEUqQ?Q3?oV;@gROz{b_=Ku7V=JyHB7TyRc(b{&|dN-u{6wm$7YgStC98hEk
z9F|?XDml{?l&B0H^ycBJ=^qjFoWPC>6><qH-8rc;AySVy;H{&84m$(1*y}ut@#k}3
z0C|@J=;;U@aIWYzsq(kUI?}hWzH3~dlgawSHzi9Jp}R&1OzQH`8>wEoc730C&%@hF
zSRm_>5XaPc!nqfc<ll16<d|;2=3vNnOIap&(>m1l)K?Fyh)F*^+v@5#I@UK{150e{
zA{CO7cMyolJmJ*{K_sgPaa%{ksernI^+X$KpR+2MH&P|gGD;b5RG8*!N{=LGPCuZg
zTR#114Z%0$uVk3Ck~4nVXkL8b`EgyxVzsYh&UK)6z`~Chjpy#Z>f4*ty@fRF(^7nx
zJ}Ld{q)xvEs>#rKm*3s*Rr*Bm+%A{HCwo2VQJ2Air?)D2$|s416a%iRebto%h<_F;
z=PQNN@0onri2zBMDRvKXWwu+QGa2(61D^`4v#yYaqV~5gh{+{CT|e89!H{O}xs!hP
zjA@?vG^*fSYjNcMX)!#hw39+KYIN|^vyT<p7m3UQE7$gmvWB?ar_ENx&cDR63jW)J
z5EINDH=Ox%aIfQ3XW}}b8f2hzsM5@_N*&huy}>=H@JZzH7M@Qx3UpF`!a^f)A~fX5
zI1n2p+_zvQ)a5gPHv?HFr3Wrmy-@3534{lnKfZMS-OJvhvKorswfT3DRvzJH;Z;lX
zfV;AkSuRnD)BAj#O@qrp6sHoVok3WK*WKhby)<OOpwBD9Aadc8GLJSjuAOyNAEV1<
z(@Efd_sZ~x@cH%8P4`{4Rpj-8R(s`TW~k68FZ<h95e)|u#?!h-9fgA>;$Y}dwjbw<
zjQ<{)uW{))?%N8XIPqVxdycVJ>D(pthK6&gpRwHXa{HQmz-8iQi<ADN=9u#mFvYa$
z#MKYd+X+W8u>MFmD?P1DoO4SN80xvCsn9&c#3tzv3Z``nUNA33giV>-gtVxchNmxv
zh)w%JS++wksVk|$bokjcqCdUkb<i7*>ltKn0Im3EzM9kHV)yQvg03-?oz%L{tuQMS
zNz8fX<G}QKre{iR?%hga=ganMdsm9IXkJ3OroNIz9_lw~5vA3By=eoUI$FJ3uara$
z9tv*rQ|>G=1C?^nvy0b%9X~sq(87QDV*g1^$8?Qk|4mJ^oQwXC-=#hweo)H7R}jw|
zmz_H+OPGG8B{-po{Ja>Xc`*Uws2Ruaq#A3tqgh;<1l$@tpl_7TshpQXAFM^!?I#CG
zx@RxF83<@{Fg?J1{d;8W%7@r@>cN(S(Z@o`X2hfAxF=zfTbLX1t;$KQ2!ZLTx<${;
zGYMTk^m#fUKaAf{PTShuMNTzhV;2oeK1=%B&WmH8Zl#Uc-5ybznr@{mVt(v%BUUV@
zdkPp&2geVxNW$%>>Fx`91SoqRPE%**=ZgmMkL!c}9llk2=r<j4;(+}7_25_xaaj>;
zC2YK_8clFx_$04#ze1atRVcoHI>flsBAf5EL^>K$)}4H&m?%zf)6e=2R!;4c<Tlx`
z5CZy+l!o33Wy{1+s3ZP$Aw#HSQ=6fe9g;US;asj>=a{@yFRIJdN{QmH1Vbln&-6+)
z@$_)KWvtf^0yjYv8${lv?;}23`Np%&RXFYjDUozeapa*iv23Fv5tJKx8lVakY??oz
zp~fAL-O#5Po7lXIXI2@HWE;ko(gUA7x|%)xRE*uDZ942zPFiP8p3^*E&?Cu{`8@%!
z#LyQ<wKyW#*@z5jo{)B)sENdH4fh3CdS-YddC^x5{8KdRb>}s<wGNzwr*g<9+GS_N
z#Ib1;hShbv2TXra>3r$iwb#|DgZlk@F-hx-9vh^t@TU0E9Fuj5i!lY78}!Fde_~Vg
z6^>XZ^`Uzh{?ZFt7&}$_dl!B?PHo(gaBK$qBlaY)<Z;W;GLX~F2iqQOYJCf5&gKay
zYQ-4E8VW}1Pba9kxf)9aJv?PH|My6Feyy2*ys>|WPqLFS9WG4~8s^4RSa@;h-{go1
zhnqoJd-z*a*xw^Qgl(i;5+o(K1LMU1?Ah}|H7wO&W&G~0=*B(?(Hq9Yvl;yW&>p}>
zDj#`mn|E)}VKD??7sN`7fF-$Wit^IVuOgFjt7PLy>6?OEN%#Ik1qTRziqW)5+ClQk
z_oxi4YXp8|B(&~SvX+O7oVIHFj8z}>oC#M|Ohv!C<R5Y;=h9fBL;kG#7rl<fgxU|1
zp4aOkfr^SBh*Tisc-Hy>lmTiusW(IPHOUBGW-<+LACoSG77!3xiYJ#%{HA1mmcQom
z6)$zpb)Y-y>aeC0d#}eI#N8dmL27;fBLm5J=~soX94I7Y+VN@Yi(|<)+)1lYMMy)8
z(j{uOGFTNg<K-WjA%~;-(6HGH6GWbuYDn)C&9fpNJd~Thx6EQr3O*@S@KJR0Ti#X<
z@nq|FwrsptVBk{$h;f<;ue<miM##6QTcRbWRrjFQ)OyQt#oHmvxXzzl)rOZ-Hpk<h
zYKnF?K9eOt+-Ev3zWA0BRRdKlN7if0KT8fFKsE7CgF@4M=#c@z@Eg>b`eukuLM5+>
zf5(K-{A}@wfV@J!V6q4FaQ~0ldd0Sl^!nvb@Xj@YUhPz+=V}_gUAT7jjZhNqV0E*g
zOxDOkB4=B%(W8|c&qg+Qnrr<3SUU4~DA)i0cZ#xR&z4~phQvfgF~bML*kvinR(6G)
zl4Wd%24fv#myofvN|q$bzH?$|F;NCnjhbpOO_rhG<$QnhUwJ%myRZAYulM`?dOcq`
zO6hGr4SPEj=FGJBg)RJsn0yj1WU;Km2MS(+AqGqCnM3si$FNiqs_n|E{0kf$6(Ijs
zti+duEJ_SKgY-v^NP+{jOk~XdWc)4XUX18Cq1_|oKGe1t8)-f>B_*k4@m<y<tHQo_
z7mXlIFhMXmpqBvfA8j1awoCqnSum8Lv0}w5r31h0;!&{`W;?kYDBIhoLxzJFcczT}
zkomJ`-%O61EMRw7h{(Ok@@+q10#(Is&o3WLKpFfiyL2A1d3_~ufBuaJL)Dga?*r^a
zRjUUCifYsJ{7j3XR!eAsigt!n{5*e>!?~q%7aAj<8VNz|UWiYf?Z0z(*1U~;gnoUs
zsvST2LDEK0*5%{qkF2F_)czQY#D6#vLAR^)fL78C#I?p|3tya64jQ?)sTNU2`b#nI
zgV2vpQIG)l^y{+7Rh5l^!yhxsgd?3Nk&9k-%0Zo$9{G%l%%e|us*96bkCPgZqY4WS
zmwrmoQGYEGIdW05^o*apSAEPM0zYP1#RpjT`zv~*sm}gYpOv_X7Za2h6MDM-&+bBu
z1BvOA_eUY$@eUNNaQ%?fgsJC-1#{kk&k=m7zE$gHcHCILUOS=E4jJ}wGht@c6X3F!
z4@v0{yeu(u_iw<6k4~Emv4d6{3eF?T3>CL4EKFoAguE5}3J5UdO{Q0dfhI0S!6#YD
z+U100{_9Rfj<dAGwote$A*uz4f$7`|yYj3xAX}f(Nb?&BDIqr}kfNq^8Lqyys3wlN
zJjCRg|L<B9f&%h*iJsb7zU$S&pMZE^(tE6uBMa;UJF4sVet92ER5s?{Ay154j;mJd
zbF!K&7m3ha78El{{c!x@|HcRxKgxVa|CIkR{je@(C!?qjw&L8K`2kYz)_J~SsF@o%
z-D^(Y6i1XB-Mt?#eGXZgAb;&c!oXZ6WY^7aylfR7O_n9N(1M%ZOAkU<cg<=?E&unQ
zWI{%ik04dMUPxt{h0;8lVlm0Cd=aR<(Qqaui(_KoWJO2w0duwDZBe`-ifIQAC+!KI
zjQ>kj!vo>ly<V~+-2dXhKk)}l<-*A^O}Y}y<P*St<av2_GHe^#Y|yt~K!?8U-Loib
zA$ze2bUPP%Vi6{HkMGnusC1N}aX@$^E5HVsfhTL(z$6IQq^l4K?2|PreWG!D-DH1P
zM}g`YZ=8ARl<a43{v9~)aKhQ0td)uGjUz|Eu+EGU+d`6Cs~+P?F-WRhja3<exgac)
zR7Bjez_t21(Bfn`8N;)%&8OEm)i1^VrlSM1r6@-{w4`gNSrwM1yLSadgU5rG>L{pk
zCTz2U$>a=)R>F(kSi@yHT42bO6CYf@E3O8^5*hL5IAYotQS>uTv^@RGllp+kbBcJ0
z#dP4{y8m$?YA2I5&~IY>D)Nwf_iQo-2Gd~!$Hl8=i}X4?yiEEvs_E%l!g|(jt6qYd
z4bqMtLIoTJOkpq>DjQ@Rc&EHwj__6PQ6AF|0tPh?-WYBN=zVn(OV9JsNs?wP%g3%9
zobQO?U#Mot&>5TbApx&Elo?DJk^-gxi(c+iWIECm2zmdrCRvZ}#H@pR0lxdL(dS-I
z@?*e8XL4kw%>8AW5%^rfg^Tx+?xWT4g;cj_gtxT+aPYJ3RF9^(^5F{7Ue!207)&L2
ze{LR({lCRI{c)r_{3hSKAdV<S5G^V!5ZZgn*1aefIHFW{J3MM&94n873f5on{Le-g
z1gs6UsNsZKjw2q3aI!!d3w{^ic#z<&AweXnJO26(cnsDmhh$n`4N*=^_?Ul2QZyQ6
z1!n6(y1I6TQ^BK06N^C0QNY0nfj61PF8BcaI04}7D`=?3$RDRE`>h(of_@>B=e?W1
zZNY$?7}$20DC$p&z=Cc&swO%u36ejB7X!gY%GVRw$e<UDD2oejRt0nKuC*&y?+{}4
z$H_z+NTN-ie1Gs(ig&bh>WY!vG{0rC)wHK>n=MzF>-RO1oIm2~590ewF*Y)dOcREz
zkCX)M{~CM=XM%jyp4h+=IAMSG=<=@i7Sp>1zR<g-yf-s=nTyj7;34qgfeFE7tljQC
z*uAVK-UDET=3)fjK(OaKX{=!+#S)bG1Sq;PKtzstAGn9&<=KAOrPE>79Y<Jfz378U
zcre9g;Sb~%tD1w4A{TeG7|xQfiS?#>cA3YgPUpS@^E6wA)enQ)S8?T^52uY@DUw_}
zQ#zdA65ZnR0m7`f*z53|IUWf{su6hUS5F8V^S9NA7el3M{qQQuKtM*RYHDhn{%K{!
z$`gQr(t!EsRbgRaDQiU8pX(af*#Z?@^#{xm6&rKozdLxXPCW*xONSiW%&6g7ney$5
zQbRpW#a}>k;$dP&;vr9W)FQrt?M+|eJp<0&z5>t-v@{h25`A#^5C`{i)f>EvX*NKv
zM&{t?kExy&Upf$Z*cu7#|Gc-tY-vzfpF0=0#C35oVU7WjgaM9*Cc=wGX!*k&Xb9R#
zJq)dnJuu!E(04y2w2Wm(@0<{Sd?vFPIS>nrGd0N4BgPXOni^we8)J`S3N*hL&V-J=
z%^@pBINuyq3u4Io{;P%%A4`r<%e_5wyXC&5&7me9)W{XsT`aMtaT<;XW;mJ)$F=vj
zSDf5Z?ZB>LUCvhhfD_^enq$RG?_NhW&)PBxZvh+!7=?r4GZc;FX2Xr8TGo`Z{z>&p
z9M`oPyOfMf+D=-M74Wf5(=N2VVmo@mu2c%mxYE=3BHN9js`<^f(~9krTiuyk_;e-2
z!Y&Ku5O-TTsL9Vc8jm4tw1n7axy89p-reUfwcb=<wyvk<si>YhM$vwVtKVq)X8wRD
zwhdob0D;(kxVs<oX@Vfe$Fv<AUorF?IFu<s!WIWM;o=4G+`J~JJ-HRwlaDFn`ZiIj
z`uIIy>!@s*!osEvTq7nuu0o?Y$xBV2&o#0$<LBO1V0$>a)1RY34LCIFY93$83cf~5
zM)7GYH+Q4QzuNxV77E==qZ!Ov581qd+#^k4_@i-;g7N5g%*VZ6;i|DJ$<;4(xxVow
z_kY$%L|kCR+_mh2-O`6Kk_Fmz*1_Wdx>W)#)d@n(S6yHyjPR|(So@J0>)n7qj0gg8
z!3tU5AM{epxi==`6M)#!Sd%NcI+ItsX9@a~AqJmd!apq_4C7U<-08)JnJF2fhZ&mT
zay^J};)O`#B{XS4XWxtGI}Av#TMKECtg~S;h>58FvSh4w*muU~-njg4r~&Dvs71z0
z;e-!z#0lZ<f|ehmx16J`F^11DB*?IjQsc#t)Sp#Wr}act6ZfjTxTg7%tfrxQxgPrA
z{g6fa!($)M$uV9mgt8fem&zuYEXjf|Hq|rMdi|Kmn)}&wCr3@MzPD<k>4pkF{7h8E
ziPdMTsl67CLDWnGkGvq4FYrDcyKM1dGf|mI6z5GRsX+LD`4m{(-)yk#N4@{sDuHPc
zC^3o|LXPGE)?<t<jrig|v|RbnKX`{D7Zg5?9Q;gMK*Q8V;`k$7{kfz1%)SB#jV%Rk
zv@7S_EJq?OLy4I2g#&M#w+w$(aKxK;0@VU$_i@D|t|3Dqoz`1{&-%6G!!WC&9eM;b
zU%$#5$(vD5=EP`;r3*%W(fdq10oq9^f0%k8*N!GHu;WszK28SMPC<C}jH}NUWN-Ux
z(7akM)wM&8@tdP1yv;&J@^p_Mv;tSKd;1(oR_sJX;-~Ktr}c1MjYPBX!I81;)T?Q7
zANl%cWx0v+pn+SoY34KIk~6Gke;`|G@T7%(oN=mcR=2HWsk8{okJxT{F4^(>mb`Y~
zp=X5tCuen{PW>&wT-s{|*m0#dgt_0&`<#8&=yvwVf;*jkE6a9h-{+aHE$R+G7t@=6
zyK!U=XWg~Vzp^2TY<x0!=-&@<3W~SH%RLk*W%pqECz-sgf3w1vk8MNE6MgOKk8g^9
z!RG$H(dHfcBoR@1aullrX)UrDf5}Yg+DAi+CacVi(-tmg6fA`-UgbNgpmXF(IZ~=G
zMp;xnQsA2gkriqg-XV~|&=JZ%6`lgSLYCi<p#EN60#7KmZj|`i_7XTLzk2@POVDNa
zv-?Jicm=2kyRStInPm=6`iZJP%rb5W0$4Rm-OZotXzmZ@AHuALIzRgq@Px4fI}FRV
zE9RRHLvdzjYwyiB@rp))#|x8P+r0`|G)S!8sc{9WFLJCy@ReBey^)aK2Kb*n>J?X+
zwhTK=2?0up+uvNvjJB6X_UueV#9Ol`H}uA1sbE%<d+TCyUHla(H@~Qw_&R%Ie=msz
zGp2WSt?>odo{0ZC!$6pb<vIi<8zOD~twjWJ{3FJgof&MEN3WjTqW3uRiJp1aNQosq
z@#l}wWnlzII(RRuTka0dH|8)ZmLyW$J_+edNNF(x%BK923R5GV!A_F>yfELrO?8P+
zzq5e9A=`6AXZA=4^uEu}DiII${^u)h`Do*+>Jz$60}=kq8f~1Tg~1tf@WUMz*X+};
z9~BbBuup5~>c=@f3S5SWKhI}eLnr3W*Ro@+I*Gx{`&WH${Wt~ra%tabLL&P^nciuL
z{`o@lCXZS&%*Z|97v|RR_kho*XB~cPL(*gZPVT4mhsfS5tozp`z$)`O5BKYVal79~
z7)tAwd%5<57zM|q2f3+(4}#llVRc_t!<|h(rr!S+&WS(w-O8qXpp=2vG+ydj0W;}3
zaf0OJQxFXQ#b|e<GgS^SL|ys~A0;I+TgN!6i7##_3v;9S1*x0J=x)FQYQiTrOEKdG
z8CyY=c8v%ey<s!s=Q^eowS)8Bu|4hG6h~2BkGQ+3z%H4kq@{161%638Iy5yIZ|f;~
zR53aU)3i#9=q<v(T_xqcvHs7@mt{wCJeO<2&1X4+kIDQO$knt;-|y8dymIKl>vHqm
zp`+I)Yw~UbXK=MYYN1w<YVBxKgSEUDIKxY4<$NSuize+w<d8uyu?AnR5iyi%N5<>k
z7HeaLw=x~zuSTxz%+T?f{Sk?xzTU|g1Dj9zv#QsyN5s?RYkM35G@AUQb*I)U*IODo
zNKb5vVt0lq@yB46MHW$exu^9Li=H10W93&iF(?rVq>Hz+W;C(|tTY-k<A{&i%|2!#
zSBtp_#mj@Ug7T%;-U!F@=e#+P!*}=2W4q55Bly#0?P;BjWx`ilj3>=vbS)#p+$srO
zjSr+sBn#iZ$e>G;B4$2gS_(Nnag304=#_o*=jn6YKYLn<GVu<{71s!1rm@kGncJbq
zgLQL%i%%C~5oM#qsJ=i*u+7k@56+5Jx~)8#IFow!M94eE_!3vC^A_4a5t!aN?~92(
zZ=LrkT=-~k^4E^?ywWMoPU>$1nf+ARPBS#^<-juvMIU-B;;@)p8?Zq?ed_%Kqi`lx
zr|}gUFaEi}21#F9?cyji_7BhZ0GGL1zFW;S$G3@#%Qrn`SJB&{%muT#B=^E<L!nC$
z%jJA3M$ArXM1Yu1-!uk?K4^FF0;PPOxe`b%JLsITh5753n6P=#4+q`Tq1<R?+F>X&
z{glP&l+}w5>u`cQ8J~YrziaeywF#XA<HWn*c-TRRO`si=_~*}_2iR8IoXl7)A<eZ%
zY9+hp(ub7C85y23g{8;Wb{A~1wVtyoUPmeK6sme$-Qf01Mgqx!RZc})(S-jM@B<sY
zuK2Bsp|Zmx3__$zk82Wanja2-mlpvuH*;K`)s$$t`ci=PQ-bwU9z)<Nh8qm%TVRdM
z=19KkXg=np5WSKdk?o~<-t%yqEhT{{!An3(Zxikm`_$pRlzXqb)ze5htm@UE6lL2N
zMc*A2Y<@0iYE$|AvOX7H&iY-qcc5Ke#N>sTNY0PNqUdPBj|bEjgRb}!h6@?VMeFbr
zH&Wq)%cQ*?Q+k71XU*$r0u8D@Z#1x@?PkMI&H8k&y+3);aoTHnRG5yau*mn@I1^K}
z0a>bP!gRZ*zZ=^+rNQoEX^g#Bxo;T!Pla;NS2fk<Ye%fBPmv+oriw9k>EE_{ghTRs
zDy!K}9%DC-2h;U$cBg9?6im##O?TXoIT_`Y{gPYZKKTqcbKt;K6&#b;HdfcK)+>ol
z+bO^GB72;1%jLO>=TNTIG^)`D_;sR-SsV(uLxGJ&EQh>PkQxq=eS{ITTbX83Q2h3e
zpK+zYk10t1<<He>E~a-5O=?~%u7?b%V+KFWv(n~Sp4f<V4>&Ux(!XlXsF3%}V6q{6
zG32m7s%ILlEAV>4@K0m^MY1vvtz4y)K{sXvrf*?jzrCwb>xIJ(T3yYwEx^wNy4oC2
zcI|eFck4C5x|3V$ySmu_vw{oOiO8wXESu+A&t0#HM(BUzL<q6Hh!y-%J}jD2;}Ny2
zlautvQs|Hsoo87=q$boT+W8r0@-7fd(^x2Iu&I<acm<Ly;__*HHa4ft5-iFKuW`_~
zAfSTB19V>D+Z5lBh@hKTl~UQ?B;MHSF#7{!?a8W-G~>&;{MmpzHI<5T6_+Vz-tjYA
z(%&H~sBYI+tmSm8KB;4lJ__8cs2`;WyS=4Dvz&u2&K77%MLXVTD;Dd?-x9d+#`K06
z%0H{+(ebCTclnl#-621dWv}d15sQ|z?EOj}kBlAgtDP~8&r5!@)R~%bYKeVz<ka(<
z9sCC^f&(;fUI-f!e}-rzRHWpXJ!nu4RHIayg;Y<>6dCsP50FF7ImK+t$2d`ih0Y#F
z#8Q&jLklHwvKypD<Eqfso9o<*(O0xuPr7T3i%E<v=Ta9QXbpIs=t|OAczLWrThYuX
z%4$xbImCGS?bhCtBMi-v&km+uxhcJpP}ET%oUjmjA^9AgR#N;FbN*BF!S`+J=IiEc
zRMHRPp$&;K{A-Q_j)I1=91yEm6Ho#OjIu2p;HxlDkmfVJrFL?Ir!j6{y)AzQ#R$(d
zwCZi)0h`Md@NV*6US-sx#sF$6scQGg1_WfOfV~7gS?!m5QUQ`y1x%aikzGCsXUYJ<
z1)9{@NUd{5=1Djw@BgbY?r25`3F{hF<1lO4Q9{0A!-l+95Vc6FqN92i{eAQUFLr`F
z7Kyl0_qMGxUxVePz?AKBegt6kWde+tpz=R7Rv$AUPU>++Ou+e1jz4!|XO<RTHK?CI
zC0VJob68N!R?H&oL|w1=>o1mh@!6K4Ds8I8&xh}hp9y<$I4Gs<@oL11=`EL&-DUQ2
zdE~+;J7OA-`z6EEYcE=6hKh4lpNBDt8U1ZP5u%Fvow$QVzn^1I>AekbICI89ppIxu
zE=Fz2y|M4VJUfx%HySM~cwbt9l?PSSX?`F>wTq)ZL<?MQdy6P=3!y<r^*qk>PxS>K
zv`m`19Q2e=Hf`N?1;zfer$RwA|2>fZFubt0)x4F5R}Rq8M6N<yWAV|-qZLFZN0NQQ
z05Sb)TNE<^8Y5V>EI~$Qvr_SJC$6rz%2V2v<_X-hv{##0;0%a(^ayy)ZFzOizTCK%
z{GXl=+!O8m3t4fOzsp3DU@9a;j4mUv@vsZfaRTb1YNY1VO=?xPyaY{@Hd7icJHvCR
zy1N6n*D=T*4dDuv9tX(t+~;tsd~CFuJ1y(ao}XGzCW5x4uZszxT)dL;Znge+)1a9F
zdHf;m_g;G~1{DQ7EO-aoNLI1An#J%&7J%mwwjq4@6G3PKFyI8G#z~wq<0xW9tq{q8
z`NiSu8uX!xEp|#O5~J635<&+^d8fqRI<G?;5ITyBha`^1$%H@j8*Vy!;V2^IS~~Mt
z*U=SXfjyb*?04+$DSf-hG`-@yoR)iJgFI?T#!)<q^XuXF6wlOhNLn31Qc@BIgNn$=
zNIv&#gz^d2yB~c#`sqLG)b3Js;)U&I1?<)u?Nk5#N5P|p^PsLocAEk{mQgSF2z<m_
z#9}7;9m`9Z_63*aW;M19iMqRh#zsWKpve>!X0jl%`gO(RyT;Yy<Nw=gxu<rG?p(cJ
z$mzrRpL`<fHX{%XX_p_DnnlXZpH;=#uMK&TG^IWA7?&AEcNcvfk*a(R$s)n|YB#dj
zfDZK>O@dR|O*ag%8dK0XlOY4QCR&6_@({xqKI^I8co8$dX~W|s2v+N#hA8Q;>x8ls
z7wU&7)8(gRd%ku$tTx;P2lqdFz-=lyan&hfT}nU=L?LXU$Xjnt<PZ3x%uzrF_<x3*
zZ@_lr@DcpW0;~}KL}2EF3M6DjX3m=8{USA4{*9Gu5I$!@jN}_bR_wvg6JhtYlMG<H
zB3}~&m6YynK;%gv1yZ&&nIrNxQJa&*SN-3s4m?>+^S$d^k^YN&dR&SI^X3U#JJc0M
zI`}et9YY(Kg*{(mGa!%7A^l}9q{?>+g`6Dk(t<>0#x=zRmVQkTqb;f~BhnEkqQ!CD
z=kzdHr&X`Tv>-<oOTQ+*A796E&?)*523HH-kEIU*E2v}RONsk?@2!&(e+8~bpMor%
zmRr?2lqjXV4m^-~13SQ4oSt2z&vt<MY@iBxZpT+D(N~O)5!6y&P*8lHfPcCR?GYhv
ziOKz9-V5wnH7pBEVoFFx7!?)N8Jb;b;`&j<oHloCmJfORcLKQZW`Ng7uAz|n<6(%2
zVXM7BX7xU1LyhYdj_Z5PR<BuQT2f0pdU#c07Cr);Qn3~9h3};aWL7<IQare&je{8i
z-->&qbDkk`N@=B%tm(Hc9DVO>J2d?78a3k2Gndj3rzexe&Uu?FrWv-lj$ywGJ)cCA
zn7^~50p_mSUSJ|+Q)9$#W~2yJL_;^Kvmr1K2;c5=Z|xkAU6Wt6E<-JH_-manuCbZH
zg+T^b&NR`#yMO=L<7!5Yqa}Fk@Nal8eqlEzKJNM~=Lxy*=g`ZUBdls{Kn3r!Oq7DM
z#ZB=@0*+qc$OXd~B$<J3r!cq4RA!d4r5D)|BIuEa%QKh+om94?Kcu%yEsi;H^Sg3l
zPY#WwW)Og`$O^TqAend*Vt{xuv4=If6*#NIxDP?##dOhwp3R_&Y@M%}D5lvfH{Xeb
zZ5qB~#%2?LfZFBoe26ty(ylAX`Rbxu$+C0zRIar`wgAGaCtljvOkxBs`X)wo6snu}
zR%#l3zV1~{&6Z|3nML-<VAy*P_eG&1hp1bUGwTJsBj`ZF1SKlvneP+tYlojzEy+58
zeo!yHR=f7sLkyXwRG}K-mGx50Nnd?cW(+D&Y#@2SxhEm@#lh<MGRHeVd~W_25&xg(
z-pBoHdg3=5`lyZ$4>F&gc*nTE(}DMH9l5Y`GZuQ%q9;Ml*CQ_czbf3DKLRwk;M9bt
z7*arBGa|5&&l3PFffo~?<}zuW1?>qt(hpkzHUl_Q9*zLF1Shb@VINH*V^>=37m2+T
ze&SmCHIh*}`<xk#@V(2#-rL7+02TC15P~%%#{&R^24kowDq-<z<j!!JV1tih;@(8M
z!M~@rt6CRCo|!Wre=ci0+yPj_%|pBj=-XtgkUWiwai01YKhOMTq{T(2O1^DUkNh=~
zgy;t5=6;Xwfy{?b4U_<y_woGjas;wR=2N^fHe_x#j--#o56nbQ{6#l)<+CsVdn3j0
z7`%DhXleL{UnfbWez@*59kw9}9Xs}k*@^J)^{(#8Nj2b45NICPxHCJ4`2JE%0wS>}
z2;7+#gM*VZllY8LX8i}T?U=*7(>&p|hKWJaS{0ad)K}J>Zq@I|M%aOE+9SV*6EDx4
zz{UeWF&SV$8(Bc@i3LV+QZ@4yW5O?<%(W9ISfFKOWP)M^FhbIr3y@R?UpQtOlRt)O
z0n?8L^;MWh^+?Ua$l`$Bwa$U#$y|5R%qqe)>EqQM^@=aC3H{}3T*r*c*Dixz8Epqs
zwR%L$;ZrS9bFNSW$M1<gXDWuHWto{ZzV#Zm;Flm5IO2C0xh4{Pq1kQH^b4vn=BsZD
zxXSR5pQ=4{33{}n)Xl?C;{(+4aL^8|9yjW#E;tcs1+j@c{z|h6nT#fKgknoluIoj+
ztC3v=IGU7J;#_%_|8S1kq3;zssGHq{9OWuRtBtkeOH26D>p<+Jci@wEmo`M#j)Xhl
zK<8WFajN__()jCmm^3BIf?lpCXP;$eae=U7Rbb|?-V$=b@$j+@+rg%G!~0k5`s%}$
z{=1EhU3>9rYG6+X98jtN_F@m1G2p7=q&epP{{Po7cg+Kyt!jR=E8Wyl!&_(~Ft>qZ
zh~dHJ0C(EN(fUSG_p=Lui9gUjb0C$(*ZC>!<9w*pa@u5<za1crnW{GJWJTG847qjv
z)wMU$^6{6aqn}1%TPG4(>_#C)s0goAz@g-t2_KjwAK5TyakOFFZH$*40qUZ2K-vJ(
zcXpYI4M|NI>Jxqd2X`kYnC|1K1WCoKmlL=A0SDX?ursrD>NWWCaK2ugcX>>Fe!Jl<
zP4Z!;RJ(q`*>1K^%8|erO6PmTis*sK6CM&$wg>M&|F-Q(l2Ir;?z(R_#6MG-`?rXx
zf4^puPWipK*rH9Sx&DELr~}9I>WY>3XJxBaJN*6mQPssMiqzTkC%2voI~k-xt9_1^
z<Mx@K7!eHWXT}Ivy5rH1EK_a10$baws^aYaezbI$AhivNd(!pxcgg2Z*$;5$Y)3Rp
zZ4-49N%ijbR(%apH<+F2-$87F3dxv)CQoBhxEgDkYsg^wf14nct8sWavU`oHS;^=l
z;K1Maf6OjQbRyNmnAvq437Evu&`rTeoKMjGpI^~ke}9}|chS7~P;%BZS*onJNCIf{
znK5HQQMcRHHC91$HMkg|1+gSC3RR6^tb-x4ZaEl$nb(MXo}Q^VfV53qpWz~stvtlB
z>R>G-7qo*>h<cET!qqxgb2>?83(lNJnHu6DYrR=oWV&X(kP(69>iZ!#aVig&tlDzq
zm{YyAE7dJer|!fc5qt;fxR+wvo@2i)pB<gPBy4x&wbY`$Zgp_=cM+?mA@_D%tlVc=
zZTS_lh}nr6_T<B}Hz*!W8gE9KEm<#ezH*-*%6K<~{cF)=RDaC2!Kn8)DK&L+qvxN(
zAu46xC%w|{=Fr^F@98D|0cvS2L&_9*liZe4t!|3!<@SLmDHa)|AuolIwl`5>F7FMC
zsX8TI2e0+XXB+9Fvra{aLEpQJ)~C7GrDo(^z2<N!*8FsZc#o`5hT$8QQ+|M&uAY^7
zW2@;@*~y(}ExKNVCpz+Jm^>9roUNyn7Atca%e1f2d<wZfai@x*`UBn;-~T&tMC{J}
z?iGO*o;Wb!F21*~oAxH2c~OyKzDi@zl$t-(E&`rZG#N;WFZfL)qN=~y<^t^`i2(@5
z#GW%$Akh4_Rl(pIF#wi2eKn9!##|Go7FXqUZHSq~3hmlOnVqSbg&Y$!{Pjfgy6s&N
zj!%`>K`u#r<s^}nx$;57b$tKu`qMZyuijtC^!J>?8Lr>8=#g+Ny_W`CH)?73%c=)2
znj@3Rf*1$qJMH2zLM-lXN=C%RTJ?|FSwRG~jQcM)|Li$`A44U0g_lfKkGWirxnLPx
zLcj7+i?DcJCqza^%d_N-LlQB{-k4SJX3XEkJq|Kt*!JR8Yi7@w@ogP1;J)ckoGh&S
ziO`s}8$d*x*}e%2w>jFkd3YeLWB8ZoE#G4==l$L^mBeVNyIy6S$60C^2_ksMtB@!T
z#^qX=;5_B;N47$xS8ldGeY(_5-s*lFR;yv`qflF-b*FpUL)v{tYmQ;&ak0y%!_rLq
zb-#4w@yL^E{bZ6|vl!RV=)BiZldJc?pyw1KU+i1lqg=grUGT1eXOO3H4_fT+*C<M~
zNE`0XY9YiVKK)Y}yjA-<Up<kT9D((e)3~t(7;hjl#KK#op~@K}LC#HU6d<--tnMbN
zkchDmFz1Q|!>+?%D%E!|`Q!LbPV3gO(5*l!H%6rbZqvmK24`x+W4wuG<DC=_B+Kqs
zeENF4496RnPiix#rOwWrx+Q<Z1P~+w3kCE?Du!}n`zyp0&;p!eF8|R19Du#Y<t%l#
zI(}OyBi(89hz&{%qj?|*o94$xS^;l|t2xtCD6R;=*mAMH|5?1xITe4gcFNAq=9SGi
z{K`7ePd?sqQ4eK|Q)kySobdZ_rKWX{hGxM)PyJ0K?N!%#bYJ>wmDN|rbXvcAXP^z|
zIa&iLxVGe2piAeRUketVk(<_0lcY?~QDzH^XSlJ7u$h#QmaeL%b<CYnpqrlyjeRir
zr222p$r9h=F9P9y^bycg{wZJcRyz5up_6<kH!KmkFUU&OGD^_AUDa;n*6NR%`zF?o
z11vKwB%!OSjOQ>_IWwjh?Dz>wPI;b{((@wVco=xMH2&FBA0oUpv)l;8u*7pTB|b#E
zw-^{oB<!OW_Q|$*nO7O)KiJN7*f&ht(+QHD0~?iceXHz&&kQsyHLP)dkzsSeG(O3b
z0P}#U@Ev!s{T1jI^J(!51^s>t!KnTuIPA4xL(P9uxmx0p7<lJt3tnDhf@!gkG1Ymj
z-RxdW-qEW?x}dv^7FkJ{*ozj(AqzSGT(EZ9`EP`H0#vxtk28B@+Ej9!^R`f}vAVG#
z6##?;+<4{y=LrWbEFsTpexy7@6(Uf($^bZ$hRw?Gj{g7kJ3OHR2$P;=>nxdrMqXWS
zA>|@}HPsk0ibFCI{q3)Ea}vzp52ta~Uyf|&<hzsy(cVg%Equ~}zPg}s<ov13rIGi>
z%nVpxAndW5dGM#VCH>XecMqE6zM@21R90=(Py4-mS~H}3;{^fXa?7?rwRI`pv@z>V
z-5cM{GNH)P%QD{`c&y_N4jPV0(}zwPG}8{*jS7^z40aZrfHYq_{6%wnNCBbL$g?rc
z@37f4{vIQ7qLJj8wP{`XbKhF}i6z*fEt2!Y4c^u5BXQ55i9v5~B%8|VfF>PK^8vU|
zFBJ&X3rkUX#ZHC*$rY3{9NZWdU4|JrU5<L_N`Y5#tBWKSvg>RdjSw+po5N{I&Hu{1
z&tk`0{@R3m<0MigHhdEOqkSSdwq#G(3y2+%IAx3d{W%bS6`&)taLJoG^n2i@Zvt95
zHvhg^RF65B07k$kDH)iFYlg7b_eU5Qf+W7F9V2fbZa-X}PWNq+MMPTMRL2N2JBLbK
zm-r$5EZoPm4wdw3ejuT&^$)Q9c9VZy$Z1CbJaMOZV0Z$q#+&~*6*QcUhEFoC3{L?m
zOZPB%lO_I&tou-9q|LAQF+^>jhP)MB&h(m^S^?y3!PwT(ComhqLTqc{+$hhL)qnBZ
zUW)7AhU~z#wEWwUFM%;ziih8wEi%7Za3$*2;dvcfZ9c8S8ms0+Ut71vD&bKE!umwz
ztu&qR(0MZkCSyx+_?2Sdq1W8-?1pFUhnY8EWdjaX|IYZCT?r`WxPLxr@zwI(OocgS
zF7(}_%!{Kz_J+qE%q069j||(Ea|rtKz{Y0gwDE9_wo`_6iF=N1S%>()51%IJ?^Hb(
zi?5E+us&9D>xg!GO{7tOOpC(biMS_Uuh()STakV-2dRJd#A-ZWor>)ue}azi*-PLh
z@X}QtP{WH2g!D!4^cmN{$*ohVKUBj8?T7?hk!VyY0@m!%UD&rJZ7cAI62zIziKq<@
z7zOyU<JE{&2Q{HxFj%S47z4Nxq>o+Pl+tii%44J<wB_j;3oUQBSl=gWDB9-{R?bWw
zL;$QLEon!$a$P$~lK%I6#%lH4UC28sJtSv<J&^d>Dc<UX1Q(nQD?=(Q&3F4U1wh+*
zNNVX)+;NON=r1VyM}nsGE_M9~_+qINgtXB1kz@QT;uOqgO7Rd)_Xy+efuw5|(%J`F
zb;z2-_db^ua#T8HM28V3V|pQ}DF^pSX*cp@$J9*D$;D=$0SQW%62t5yo|&P72ieh8
zf>s@7@;{2%?oJOc-Ave5l6CwlmCslG<B?Z;q%S*OOUQDLY)ppD1w=iS)zu5>J^S2O
z?Ho}jzR1}By|sbXU|v7^`+KuqIeRNEQt01pF(=p4UiNpa1){$j&lGOhnzc(aM+!ro
zxK`y$-XlWmfkfl?X*<L}dj{vyD%LbSbt}g4f`6NO!2+1)S<^2FkU?jr^Uf0w(-*Rj
zFqX=f7|;=q*4US)4{hLXNw#Ij-Yb5-CLdqQ#YwR2B8k6hrz4VU^(^!Se-v7$w0ctz
z{_5`yQ!up24uN<n|EqjuM0dZ(d6=%`M7r~b=5fzkQg@2Ylwh06WQ-ULot9fsrqx{7
zdSU@KClN|t^?gp#U&VgekYZzy|Kvz%CgCA@*Md4f&ozDuS!qJ<H!x>sXtElX%GDb4
zEgg#UC1wg64WuYA0LL(2p%jFc9$($PfxEcZU9lDsyy>O^1~yBO%G~8nRsadWhEC)1
z$+@d|<;~8%l^Uy=6ZwdnGt3czyq90obqT4`%A{g`NjG|ICbpvYo>um>XQru#l%m@y
z$w3<N{L#@jB8WV?X-tiV)uk}KgAZ3#t!k6Hi`<Vo`4<ioV2G$sa&os~bZ&EzM<)<L
zdH<f48h$;E(z>mL$e#Qh!aZhM@CNZDJ>8>6_S2I)aWdG@Rh_vq$rwa+>UlT6`5QZ}
z&rH#n0?75@GcXhsmEWyh-A|HOgj0Za(Ok2t(e^hy++p9OQSa^LNAW}c@OhRMy=Vkn
zvT76z<{=|%8B0`c8xA?Ycr+wWnVdbVP)`ocQ)65ESx8_iOnyrNGGA}&e51XDeabV_
z4S^rM_CGgtn3clI$E%E~QU}LaOHur+vFMDeMmc?lTl+gF1fd2ZT6rjfD?PMyO6rp%
zARd8^ZY5w+7Mv?AeviR5O(o_Pbe&%L|1eLf3r}kbz|{CvvHi(TAy2?D#R3Su$v&}e
zpk&tdo748{de=tHT-@<E%-nGZ)u-ML;w+ubKHur_;SBBC5!;!pd{5#7pI__6n>h={
z!5!8ESZJ^%E0_kY&VPk#{x12lWRqm}0rLI~AZS$QSiA&jxBjhvYt`gdc9@2lvO@ZN
zHr`O=Z@vx>W*ad2@@K3fKq?QZo79iR^m-^&i{`RXR5UbiepNXSuJCCm{Zksq&lHTd
zF_&3c2f{!9fQSitD!+)_+9t9Iv{3HVhdX)+{WP1&p+OvQ_gt?MAkW`Qfll};0Sm^C
z;L;Q{tR0)ap@TTJDEbHhHYM{kcbk%)h2VH^!iP0e9*g<!1njN)fakD=@m@?!3f47V
zDP96oGo3K>?=0KL84I%amt}pXBjrE88dUA*@?&oylm6_n#0hU|hVn#SaRb|u4oW1~
zSk#c9v9h<di+gww<R&r|m5R1yW2}j?`pJfhq^xCp>-gEH(vjOmGs$aW=M}Ez8(E@F
zze2;G0^ww>(bFwsf6JU<>Q_83AtneMX7U+1Qm8twgZQq%*H@b0pDKO#1^t5f;kI!p
z+lSWiwp>ADK2}B;dZuC))ESvI@i6=}YYWEOO3YqPh-e28qS{@|<vWBQ7`FnN5}2t1
z3}1y+9q4$+9rTq3s%m+80YIi|o7+yT)k(}v+xcf;r7K04hug_{rYOR=+>64-i`^+c
zq&kii_;NOy|20WcGU92icD;C7qv{FFU7{zF=wRtsJ>%=o#=+}vJ~nRBm+o|zJk*<3
z?3KpHO!qi6Xe#HC?iBRi%zpCQGiF0I!XMJ_*$lC%N_`laHfOBT5CGt`JLBsKqa0+|
zLnhc^2p#Zti@kAoIX4n5+)sV=z32J?7G=3&KZxmHon-;-1ELK_X=>U4G0`}`3KGZF
zww{&MhIFwXip-p=_;_1J{486#!8HOrb$Y-QU$0RPhn|E2)B)&g{6{XfsQe26l`jAp
z@JmaTWG$l>&<qK>^7$`gFt+B|P<Ij%CI9R(o8MNFZ0Yltl9G}J(<|Y66r<fK*P`^B
z!+pMJ#oaL`>Sq7%0FN$OZ%}Ibqhrrdf`S!4r0`;Q8o++6-RmSi)Q4azUb|P^0kPOp
zk<{CPUH2DqrGc4P3=~}sx}N4!+i{nSrNM6|i+e|Z^`ue=Ji5EBrCgASL;!+Xooaqh
ziMOC>Q2%xv3`?`JKx#^66HGv*{D6oo=<4)>vy0KegQ8hZ1?R%_HWMFC{MjSj#Sa*=
zrY&2*{ZO8&mC5vLda-EVAIT9;PXTi4^<E&4UDsrm3q{~3+?eQ_KXHW+oyGDbl(tRL
z@G9a3Swe0~fIt=TdDJ^sCWpLT;x1MthC*1$egaD<Ss;BK`BJ<xCjp;<*X<;kMpw8*
zS5meh<J?4mJ#$~}5^~}N<lcIHc4mwG8GSdM+<Ujfx?TOzgx{*^74Qv80BsgayhsGJ
zyjoOG7T#$&aeJ>ZyLgkr3))F)tJsZu5u~3FK?HsRlIp;Em-ln+nDcWtR5l$K3?wnQ
zJ1@zxuvT`b!SPpx5l+R7K#bJhQ~HNIOs!(oX+yn1WaHl2b;d|uGr~_T7YvtcpaNJs
zqo;sm(NsL*s0@M!zy}Dig{f$X+?&8Io}+CkDStzpYZi7WAHW~c_gVo+yxGvWc{h4S
zZ|0e`=hfc0%STj_#mpujy=q>tS*tbk6+6Qd$>N-8Pyz8zi92PaUoIBIv{^7N-PesO
zkPChp_v;Q9*T_O?1p|&&A#lQ2n$!#u?$CTx(JvK=?H{hMRc1&n8OcxSO6;JT`EmKy
z_`vFINET=THX|FALn<s@kR#1$vVm6tP9+{O_k4~~0mLPjs1Wj-{i4FLIt;NporE7b
zJ)wM&6U{U_+R{o}B%(_MIY_?mDZ>H1W_b)*8Ne6ct5{KGu~-~P>~o?B?2cQ{|HM_D
zfcR7*pa0PHm!uOxZwhnBb^SQchWE1TI2fiTW&o3c_rE)CD-4eh|7Crs-MPg|j+i%(
zz>C-qKgPBF_FJ9lVoeAq;)9kUdB^EjOc+>u$9z?d1h=WeD&w35y4E=ae4og}Pl>7;
z5nT#S%+1RVems7e+Rr;e-wU}Mr)*g`4v}9y*?>&1*!!WdLp`d(bR+Qj4`XQ(0&O-v
zhKG)!jhp*Pdtn|Zf*(|Ht;dc-z+0R!-bI#}@w4M~Hf8H|<MS<ns2)gtT`X#){Ya)M
z+!$THng|m#fYbbQiGs=y65DnY1~`EsZ-z1F<PQFW7*BjdBjSt~V2f%}p23Z9+xO2>
zzoGBl>CnM6#pSkQ*`~qPM3p>CR?Y}G`K$#aVjZKK=2OkPmt4C0iOFV~Se2a{!PZ{C
zI;}g9%_CKZrD_3EzEMPiTTE;Hv#0Ih{aK#CT0&~qy4YQgF-PpccJZ;^guxEzFL?07
zcPy4sh{IL0Zonach3in0m%%8ubjpa@0hJJu(Hfgc7RV7|og0jky3ar-C%7q*BeNoo
zfB~0*6vkZxsEU1uTgtOyn_}@2Q_vnI7+M?zPMmqKVeL0!UVL{rX9QX$YawCniFzev
zX{Mk9S_578jQ{L8{LA~>7|(pexHkA=cMB{yAAtvZfG6apV2KV`9G-9`6qN(>V_GoT
zln4khhbloI2AcSuOU3({7JZS>6D~M`*;<s(spC3u^9Un^SBeZ58G)BTyV9IKxPBy!
zPsR{_qeWYH4s5H}k8wcLvTxk=+VcZtWZG9Rsb2IYcv4F55oYawV^pvucfea>$czh8
z)7!Gs8qG3XNP2w!17t+Bcv%m_lC|vs|Dq!TvZHBAHtWh32HCCVVImnnv{v2SeD}Ni
zwaJ>$RhuB092`aJ3#ivJFhqDG!cK^A`5gH8W!=iX<*<vbsh!FS0iSHiK(MA^-aEhm
zxxuIARm{=8UnTVgcotmRJyF@PI0&@k<^Zpv51irXSg@yO$$`Vdi)DOhqk`^p#oNYy
zH&iD$6USC}(o&kok<UFe77FQ|q$(4yE5L8}vb?T><p3Y%xVq^q$$eY$_J6w!md=DO
zdf&?m#|^>n02PN`dZj)Dk=a5Ivs;>;onmhngME2}1T~tcYE^O~P77b1sq5`fOze&q
zl-Mu_SsV4T{l6tcK|BHj(+`0eU?ae;iv{#xz-hAiVr}wyYEAs)g{zd?5M))wdn4y@
zZB_ynge7&!tb6xad!BO((dDImx?hAcHR|fMIlF7@kWX;iku_#?^><$7$?ixVd5L0v
zocd}C4hQdy74zz47p1F1xF8}j*$M-?(^)B*chQ;5p#r^yj4;_qbng$JpMlw5o4!H*
z$|2`uE;;2o6`MoDsPu%PDQbi_C>?zi|K21Ws~cQ#`UNvqUm*$HZCkax${Swe72=V-
z+oB%cRnt7jNsW;Crmf%`!nHE3LBZ?7CGL@$#mHt%8p&o*+l9y?)3KI4H<e^BctTGO
z7ToLF^E#OJS^3guKh6+%8oJid{`7R3RKMHP@!}5QaVsT31CZz-?vM>$B)j#*f`G$-
zx7pRIbsfUjXfbHojhaqe+g1zuf<y%NVg&NXg|5*quvbg`*;9ReLwSAJ76V;%O3y+O
zL3Y6ia8Q0PjC651TRr(WLw$s04|;whtREl|3bHo>MA49H;z7hK_%Q|dY9fKLzfqhK
z-?s|soyX|j?wsJm*(PL!YqhOm9*?Z?$Ut+10Q)9epA!*pjw#PfnUI$r&uI7dKLziy
zwww5hEK*wP27k2^(GVmP_^avBlADrP*v1hR(mx>S%5S86R~Hpe@u=d8+`YT0eC`mH
z=H(Omq}<VyD)Yf&E~k$l@MA5Mx^6yNFKbhRt)!9RFPy7BgW2BrxEsaFx&F!s8~C%1
zcSnnCf{YpXPg9ec-QGS$0F1LK1?{~8Pw!nSQeO!=^E0sZ><o40b%H7<TB!f1PU2wd
zh4WXd{i*5H^kwsM_0)lK*MWlC_WEF8-+l#1Kj6BAmnKF?SYtLd6kc+$gEW$dFohy6
zg5WDb50G`5Z}r9CzLGJu7+#f;z~?pGP?Ukn<J&IosUVJnTu~|Pv8?DL-^J5efs+kY
zt!?H^BClTNv_H4MVaM!C;)v2T_tK~`IEB5CT;CDAWXycl{;E}@pG}-2%&*IQ=uC@k
zR3h+IsCRKQAFGCMOSWMF{v`pt(uqhm!>O?F+iogV=F0BH8XQ6YWcB$z+?3Rl;g2b%
z2|WwHI=>++TEU9anPZzDiQdkih0{$OL}J1MOJXe9Vzn~Qv2t3vV}dauB$yXJ9l?fE
zT={H(JH{j1f*8a~dkXBb!wjKksr}U8mX8sNcC=3h>CC%Ly6Vu>7M?_b8>lE)u@Ymi
z(U$Q7!O&g$Vb=JcJx6^Gw=iK(F7*!Uaz4Jc7Y+G0+3C4rH7j8MxM@B)#ipL?t4ET4
zU@9tI2|pT^K~!{D?DwOJv^-wIPpDU4*>C0dSX4A)ZhRHhlw}v0Vh}2x(o`3setk#x
zzx;*bZm_cH&|40skAht9e={09ZT{Q2?PMut5S>nK`a6*(&QTIbUyL3KU-gO01%XD<
z5;++}kZD0>##_ys<C-@r&T?^;j5J)D)sjJWxWT^z>$u6P2Hh?BRY{OShlY9orAfFn
z!iSZRk&##xt1=IIG|`9fTLE}h&;#^0H}^hTdi?!y8aKWmS<LzMuXUl(aWb&jweN^+
zQKxz4iU*r}3(opTY*?<0{)i=9Q-EN%F!HWfWSyfAKf^-&&Ln*{T%#nHg6$gPXh^v>
zd3qLo6ZzV~%-;0MaQBAl50q0PqjLdz-0aqPC+LbaEby5E+ODLD0t%!Qd*ax#?7=}=
zo~}{<i;Tv6<&h9bng8Ljf@9~EhPSEB>Pg?thgVDGvOW%-`p_NrwEG?F-_KSK0?^yl
z!GVts6c7%_9PZ#)aLyfA8Mx>vQnWv77OE8*g&Un~nP=UX3>nPeYAdzI+?EqvEr!Q_
z`71?R+;1-J$=y9A+C!^8JZ2^TPAspU*T!?8vq8+&2^V@?#s2%}#+$2s>a@Bv+lY<?
zJWO9hz^d%Yi<!cgc-7aY3(7X2xx4{y?nj#q$B=x{BL9re3WuP8_~Uky(f2|?E-kIA
zm}lC|VtT)!{h|duK&?RntBHuc!Pav$t984d+@J_-HKp*~x(;N>{V#t!MiB8!v;H7q
z>`iI@bGpo(oS_cW2#kcQWe&;i|I8j_r%z&y%G|ULPIR5O?=Wg-=yf?&C^8>d%G9WN
zv+Gj5+p+*^(Wn%k{lY}!&z|HhszaVpukumLpH>*DDWjA~;+B*Vt^-jHT#BHKutD>+
z!E1ob4`>I|_N^pRIz+d0@Pft~;va-0w9SPwGU*dopUkOTFpDez8=KxZXyJ+BV~vIZ
zBtV9vx2Z^P>0_V=KblwO`pTaQOAmRmA5!GSZ-N+UduMjP#1co=N#c81+EiZ5aQFHw
zuU>99t=b-~Jdc&GY10yEc_H*AJNr=elIgrwUP-Jw*-0v}aOIIv-=lcJ4>E>|qRj<m
zw$|qs{Hb9j|Gw(iInw7QfAU2QHHsBhrT%GbJA1woWM7N_3hs*(A=VxW<(`~z*2^W7
z1w2%G^6|oYYrfuy<M%JU;g-jSL~I`_(SlN@LgYS*4zxaGJheC~*#40c1~LcQ{W%~m
z=LWA_0cdwDxyS(h)tIjQVJ@&R0HOmM4$iRfnt1pxHYkjWKM7J|pt~k&?6I>n#)bY|
zeA7^g(?l{TqPz`IaQI+G1SGk(7bAgB35ebB&IYXM%AhTyiB|Jj$<^Vf2MNH&1m{-?
znk!ac73)~XR8|iVc*WE>9c_LurFi+jk*1aWU?^3{S5BM43Dk`gPH4n`>%Vv}85Fq<
zh2~h1{{d-^F!M3#FM0qT?SlNLLrlF{DKl)QkiizjQ;knRB@86wbXJ3W*RSfw?>Hgi
z(hYok2aC2MpkE*KiGLpyy;&r6_`~-hI|Pm4hj*DvNPTdAoE*}0R76OpBS5#=^4$2W
zT)OAm^M`&xzKyC(f!*(VOfUMFBm1b|p%3HUnddyJrH}WUKAhjre5Zc(G0ykL`%S5l
zGBJ3^3u<-n(+j6b1qP>0-;kf2%;bb6%I*NY^4{I}hvYdQYKS&V!L@qtRI~on;d}=4
z)p+!9t#VhUVzgGwi9_Pk=N>4u{@PibU9HUxeP@k8+H?bL2(cYEr5m`~d*wu>{JJ)o
zGu+Q~P2MtJ&AbpJ5A<1SRXG`L7YcS8kAQNNT+pQs(B0^gPYp9$vMjc--v-X6ldQ~E
z=ECpj#u-fjJB8T*#fTA^GXyU(beB9u#rYw5d+V(u5HTl6m_xro>mqyB=cmM?*Uibi
z{f{jUa#15@+QKpvOV}}|<pjB7)*wl0zrgFe4pp9lohi1<mMVEx>*f}gBY>+?m@Y)+
zy8wcAe5Z7Eb?pA^NzA`Q`skx-XQxJlH&LkClfMmHn^N0P2fw#@F2IwQtNEyow8Gm*
z<q@rG@+GB2L0rY^4aL;fQz7rIE|ooLx%aD>#8oVTbB6FY)G+xk8g3^2kN3mEawd;<
zaH`$lSVutLq0vmfC)}Ld36lc_M|YAt_kU#07qUXq{)!iiT1ou&+jz0+UZfST*xr$(
zdjE4?tPeB_e+jJkfOkdfFUdzw7F*gIv9kN;O45s<|Av2QzPS0rLIBM32A+NTn7vsQ
z)YZ3(3WA;hg$lohbt+X!g1Sr@Q>Qco!uQbnN$Gp{E<eXcl629-YZ6~L-Xc3Bp6SYt
zUzqYsS)YS%wv`$<G0#}6KyGZ8&(DOBkU&@FZ5c3Hbxb&+3DC4G*O^YYxTd0S0MHCt
zeXhepA(?d-y>(aYPJZM1%phZ#hKiC=cL}K@5)m#=S^t|Pn4p13HG?<unxmBU*z8O!
z#Q9Ezls+q3q+KIj5Njb!$v}&=OEcE|NUF!wUkbRxHebg|0Q~9yO&0vx{Rj$j?-j9^
zkI}Yx*%!~x+5Cd`yiBVKv^g~wCr~Ima@l2){;Ix$ub<cA%&C&HY~DIdSg)?X)T~(V
z=AU%SVEUtSyO3^OPA+aW4CgEL97}$R^SkUh#7=G`-QYK+pDS{&y8YbEyt~?g_Rm6=
z`}{tc@iVU6>ik(qulqZV?Zj&3^-MPAJg7Bg<Y(qIBu)EX1x*<|kEz^*mKv9YYq1B7
zUX?7lTf11yZ631Id^!O)udA1?^z}j7n1pO*UEa|FkI3Rq(8uieRRVTb2Pu@DJ<4`^
z=Q92aAQTt|37Ugjvdu^0bk_s$0eFlc9XCF81*A&cK!OC6HHn%0F)bErVka<Z3-td<
z#&le$=Bo<d*{KPhn9U}NPl9{Fxe*wqC!TfV<ACN$B1VVNlU#ocNViB7BO*v@i$>_;
zs>};;ShgK#&v+|%O4XqYBlOx~EZ_Uy0ugDGWG^C*B3sgV{$@XD!K^dDlwef7bHewa
zn^-FmhD?UVDN@~ME!<VBK#jz;Aeb_h(|vnmvm=!+Df_|H>gN+P3hYUxmtI0+#2+h%
zU`;C?_{CP>&B#_3Fg>4~+#1c-1_#K8&GQ(6e1T=x;mt@*$datiEy>72sSUgB!VBjw
zeH+;)KNIjB>N^*Js~?0On>9OyAy)^y6OWyH^j*aD#+e7HKZd%$><>T6ZS=D!eC61&
z=BfPY3!`gNdHw6zQQr}V6a6k$F7?Hr`M^28ZaL>^ymdG@HK=nivrk9nihD`?{0z<Z
zQQgF*;Q#*YDczP`O3%DHmZ9~&PR(iAI>S@}<VZlk*&uzvsRLpaOZ<Q-&t0o-G`27#
z7Q<78+BNbK65LpU?_@^WeuE>+aeP|ytGJ*mpkCrfJ8w(RJiHG7zEy!_b(sw)t{G#a
z&_kS<q3llDmxSREinBjPB2KJIk<0?g!h7wB)LoLo?unUFuWUwxe3|DsFYIJ&2H@e#
z66=~x9hepq&|9wEt+GI=6;W%hU8~3w2d<GGK!F7b6TC|TCKO$uiC(N!*sZ|LjFaQe
z4;p7Ec7mKm)25)b66KlCYWE|2$IH%t?9($-)R8}#)JC(g@Pb}+pD`M;zcmWIpXZbA
zL%d3EwW0e4r*OjyhD@>j$@a2j3mKPw)FlD`m&9uC$vT*2DzacW8G0$Z&C6_JU`KVQ
z@<7)}Vy1pi5+d}hCBpsj$TfIB=ET|jOJa%JsnCy#dLn7jJ1Nl1iY3Cr)#B9pP5F@!
z#ht{)+s3vjo8GeP;eo^eO=d%texwf@=@<p^>G0<spL2+^f`}qKTZ+MWJJNM#BFciy
z@#ZJUM8jKffin#h&eYidW9iJpp<e(0-zkZZJzI^8osfO!U}o%OiDWDLzJ)B|WZxP4
zQe<q&mL*cQv9B>mCS+*Ts7aINjD61U_WAy<%U>=omznq6_iK4RAJ63CQm@63Z+YQe
z!W9^Zh>9gSwU4zJ7L%oAN1S_97bV)v0w(Uuv1&79vomZ0{!l1SO@l52&2`XOD>D_T
zHS>sc7Qqb{4+D)oNCZY(-XKd7nkHJ+&N|vS&Xe0<e0lHynHeYY>q}zE<o$v`UYDF0
z?pGD^2~WQBP-ZlpIFjrMp{O^?^HZU2U~zk`x6Wzf!l#!corsdBrqZ)}vg&;qgVR@g
z2K&m{&5;BO@8qk>?OH%Z_&bh}gHPjai*sE`oxygc`!oim*+_=z-g!OEU0ULyY)@MQ
z{<v`aJrU=;z~EnpSMm6m)IM@`m3tdb5|u6~{wx6Kyq%fBi9P3XDGBoWdH?vE?D6&L
zEeuG!xa>^Vf05|f9w~4&H*p@!CeEHM)b}&v7ck-@Y~c}-9cwv4fV6n_5{XV_Q5ot?
zb}G?yrV<jGJdQaxdd=s}UR2QhzJ<?#-xEh*?~VfbEG|9Bxj--Yt2$3=ox=c(l*7b!
zzy*$)Z~z%>c?~ja_P{|Qr0X?GvT=)-^UCM|8v9}}7$CLfY>s7Z#AUq_5uZTkCNFwu
zl3msnCG@Ox*P87wG8MX0pssDkZyLmTO&m%eA}_12kNeNADMp3U84<Z^%IS<%ud|An
zh#=BlGvnG`{&(*8wa20dbEKd^s_xa(WdEX6snIjOz;bOMGTbfv{nrQ#Bs1KA)GUUJ
z<@IY%pPB)4XaI2b2B9l^2!znYl;+&S&)#`-f+!8vvo!WfuvyqT`w^Bzf%p~-UcY3Y
zCumM@(ZE!T+*qbJfZ_kmIaU-o&@ppbcob97Krx%VzEN$1Aw+>mFW{kv)w&b)vBdXZ
z2F((k8Ik!W+!Dli1<>9ha8)2jJz6B-!Q_SK3G<-rLqyA&dWXKxO}qhImq0A5eF2h#
z%QT?=caG0v-*mgp`np$gB&Uv8t(V)$Lk9?g@1b#9h=QD=$f5EzgoKT)u;L-VHTJbc
z#La%2;YT!`1`62Ew11jZFtLD1@{kJpcIcs0#9|I`8SovuHlpzwTjqYoD+En1twft0
zFP*ww84(>`0fj`7+%i0<3{X>~XrR|OcMF9t$S#~kPn?1!0sFR*Gg|9*Wf@CEDoC0~
zs5y(*0CdGDT_*5Bgo0`UyjcAl)v-qDSrefBFfas^_`(9U^@1bFO!nsj73dDj*sHQv
zD%OCC3swk%z@qZMqsz!qY-MF-W3-1jklAdLESB%y$fgy*QqI`uw}JeAXk~w2mR}@y
zY;*ZiU3nJAC4N}d#Q1^$({QMUuBW77sbe|R;O{pjNGm&d%#&Z#Ye4Gi42ccg(;gZa
z@v0;+#LP6h4hj);9OV$Y?B*yr5Dpm+j!9ss0J2&~AS^G8lu35aNAvJ#<WFk>IAeD4
z(AfkyH~t>@pnHQy^`7%n0pLCW!AWz@MRIcTORd#65jYbI1#lj4+#s=r9YaAV0VYhW
zB}X^Z%OR=o%oj9)w6LI?gDWoxUUU>_r%kYo=FQbnlY|#oz&U`-lZQ4r`9Avg#OM8k
zPcLG4N7tfH<m&43lZ;xfgt+0$`)==>@eVM?s{4FX&oFw0Z12j_F<o9o3FY0qDPS+x
zL-4^ry|z|vs>Pu5I3?XNVSZseNR!Wo{BSh?RV3IMB>i?K7@yQ16d$ybzUJI?pHF!O
z$QJC5sRBrRKP1N{s(4WxO%S8Nh>`Nxe0-%464ZcUHHSm&2C(~M;3OP70dG!W!F!;K
zth@n(e>$_{u!g&WPcN`G0U}EwaNM+Ydxk({x78%&R>}N_mkOLuXIHg7=gAP(FL`SS
zSFRy(xL?!UoQMIq$j&}Rw3U~h|AF?fey&pE1S!I?!lazb@ZB6CQDn=A<2r{94T|;|
zD*9TlT*YL;W=XhvtO5exKNpA3#$JjSBxF{vP=L6LCjZ5XpE6%o@;%6d-t<Ko?L5iY
zn+6SO_(gFH90=k#M9^<7*kI->qn(ndf9cVaLM)U!vahjhtTl}VDvO3zGSX<F8p~L%
zxcU0SU#Cp}oy(>=@*nAOR7bLfcV<?2MAPl;*vR}<K}dciwvVA<Wd$%|ko2o&`Qv|2
ztLbEuo7=&czSs;LaK(s5LsJ@PcEjVKiIfTGLaC&5$t_T;fuXIrZo#CCPzdi=Po>J0
z>O|`FsJ4E#S|!;E`>|t0t2zAnQU9Ed>D;SqEX~>OZ|E`M<T9a8L5B}d0g+Hz0Qe=i
z{qb7G$bP*P(##oE5Wo<Uy=+Ik-rN=n5l?&WNb5i!QLtV;TY+c!pOa)JOAde>h2m+e
zS@hAT8(;gvF{#e9!PL6{&W#M@#ojsicv$gIH3$(9a(WO3xd0mATW=+ZjqU1!lY}{h
zCm`y8L@I=?^h74u-2+n1|AqUA+qyd<QD855Ch#u=?*Im3AP75C^!zNj>33)0MnA*j
z_V$4+aT7wtg`fJ52zM*)KEv+IJc!cdpPBwA3mwBiRZ~w_)s~E@n8Rvm=egVfmX41?
zV8sjPB(d);S>M6(I_Xpw7<(Q8=*}8MlXtC@hu9Ca%LKEx5ZAUBu(gWK*2A=*QmK@`
z>pLJ~w%DQG;`QQbwu-;P*l>o;(yz*LLFW#Gxtg4X0<+v2o=3HJJUIoSM_FGG6N!xo
zkue2@t9jnH?n}(4OsqxS9!pL}oS(Ys$-8@?T;;iiHF)bl0J$9SXX2*sFFTK|RDdb>
z&JX+mX5pCbtwuMy<Io?s@VQ{XR)Runhn7GNo1I-{7Ep}GtKUe#6!!BI$n}Zj3b!~c
zNFF!CZo2)bBAk)jjHXE-;CeFu|F-;<86a0h+=L%=k|1K{F-i?$%2lj~Z7{s<v!NTq
ztV<S~%sUh}5D!&!sA~1h_2N<FUm?ILCe!h2InPA{G&;4QT<5La+;TegaSA<GXdZbM
zRs8ikmN4h9J|YRxKbD*f49YcHhsX*hg8em1rJ7{{m~DWP1-KLdPzl%~a=p9~mqotq
zQ}4vva<lR~xK-P$W7F(A?(yd7mq=v!dm8)S%Q<gRJjz*^AN=gvvW|Pm4vz`aMqwt4
zj_LaI{5EC8;`w@>JK8VW^Oc`6txY0g$X^=*x^N%g)q&rWT6I(EjiHUWM;`0vyQm}|
zTzi}1|Cchs`p_Z+Z*faFJtvwQ%f_~9@+?C#v@ne=huLLSgkqc*@FqfNJerZ!k4)RL
z$MXO(_vraCvS#%_Aq|<@@*+Xf)lPn_b!=>AqiSa9Y0XQG+}_AmTE#?HA;E?dj=D+F
z%(RAHhR3NA-Wh_NtoakaYqAcMt*zLmVGwts)QiMP-lx<{2V7He84nxv1k`^ANQvFd
zlC9OAT45TMPTE6XuW*Z`6RZa-p;N9``|J&yAQPQnx91WaJEy&;_usittF48Ibi!N6
z#3Y!nfuO7(^etMq$A^9x16+%Rk+rp>I8Bd!sTZ#*m9fi{Y#%sp<)P`WV&4Ar%D>vb
z;Ctf_W8W_aVt=0I*NW;aXwZa$tRz-;lqf6;TH6juDgu}fxrt-VoF1H~G1I`s!<5qN
zGkY%jLV$}eO4nxrhc#oR4d!E59V}A2{E#{=qqzez?VpJhA%n9}If7v*oKy>YJP0}f
z%R+Ejouui-&<P%J3vPz-F|{oUMWx2i4Q#9G+Np5Kp?7V<4OUlHe6DQVT9|vI<e;P>
zi_LO$oojL*B)4!s{tPqI^`^*g8<j+`eDDr;ApERd;_0f|>DZ=in*X5J_4a8VDKU+0
zjOM3xenzOKuF|IiRYH^R`K!%k@z*9B+PpZ0ot^*9;a82*v*{n0AC27D<^D2L4i##l
z|8CNH1@Vd^F65_~SavP5S$ZBdrLq7?w)h4~s)BR>Wcq+xSx<jN2x#prMS?*_TR!Z|
znk3=g1N8#TrYIZ6!z!AZT(8q*9;^(v#hc$Lrz7@oTj&v2>7fT2d#z-RD`l822PNKT
zL&=RQ*gyvB+MFkg%B#A}h{9(DqV}i5gL$77L~MHyI`-E#k8kz+{)0Qn^LPApm_d>$
z6EfQ(K6t1VulLi!&isKuhfz#&>=l8rU{AE0p_B6Ab^i*FnTbY+#O81#`*#Kudnkio
z@Q<+kRba{U(<*&L*IGSq{#Q73G+cAMlsYG~2Ue{n5$0G3%^T-xE%sB0I5pjGyvPBp
zasn(qOXlwib|tEa>RktwCspOF$9n-Dq`-^fv0M%n6I1U%lerf?pr3FRJ4HGz4MGyG
zSB_*mQ^G-zI2E|!Px;B!;9eho=J)=I*#+s_&*G|pSo{28cyx}uiy?%@>>|mzgrioR
z2B3gtzw$<Ylo9_40(2{cZdKuzoS*50C`Gob3*7@V;zXGFy)WQ60)0-yMLk^*d7eEX
zG?#4UA&=XulAiFU7*0!w=Xr_k3*{-m%+VG$h1{SKpyhM*d;iYH{1@*@G^434(hX;c
zQ88hGJ84~d@l{rZX_(8%-m5V*kgZ)Q^>GXLxb&WI(MsK6>}E<)6x(nVAI-WQmgzZc
zXtkciors<H_B%~AdZt9H-5oah5XB7~unJ9P6Gbfa{4Q5h-+thUwx!&3e!wuBg8c8?
z4k2E{fTex=!7=B9?*krR<CJvsd(01bad<iVQPih2g;V}zjQ^{~ekC?TU$;`hXc^B9
zG>3>N3GQW1(JZQ@oX~4B-I^Uf=3B*t-UHfs(ukKF#t5b0SuZt8QR2FDtjs^RL!7^i
z?gkF)s*u$qPT}sjpQ;c;l>OtmlMj2w4Mf`~>m!Axe^glcE&_k%{nyAx@9sY)CGTQc
ztYzc<;VH!(no6ZI7m?q8joo)?X+F|ZyB(~<LM&}Ppbv&dYC{bdlHc+~E0{^Q`<-N6
zm4vT^yo;09I<LOT(31V%xvfZL0_@MFn595QGAJ-Wa3RqCfzs*TeRTFzzY3_tYsD+a
z!;a)J6**N3wD7gh`9m6R1Wlm`U9DM}YAV@kothTip*heNwm7aKgOTP(g(CB`rij(6
zIMIAszV-yC?+Skq&E=*VSGMVQiYQmHz{^y8Bt9Zz(Z4ZSd6KqaJ|Bs601uQpNU!F~
zb%gy}WkJKL3TI?vBU^&|m7|sfaJS%Au~v#^$6oB`34jOCe>vul(s)rn|HrkzJz54&
zVwK7#$8fyQO-+OyE)>JLYrUG@7aQn$162g~s+^8v+6J_43mi*vG~>pX=#1q&dF`rG
zy}K`8Ftg*~Zr-kw_kLH!_|Q&I;M%pnDL0g6g(O1d(o_DVhS<DkGC<2UScmq~_bkdS
z@rhjE2(vp8C|&mPgXohwU-^~s*(({Xd~g@kD0j<;d~km(`O_K8q}h^cX->sq`AmeR
zFVV6QWLrgAg^SM*bUfM<-XEy0oJZsyKiD~`C)+_PN^V8XF%w{Tf8{?f*N9`IYWLUs
z`sqJ4=1^|<Y4RsfF_DjuTv5?V$D%T37q|rDGkq+W4Q7`)=D5p~@;+tqNLH?_#i^e%
zr%*}vK};GUpgp!^XAA8YtHO^Qi9??7<MmFm{$A)|RoY*lIPqB0iu=jWF4>6^`WmL>
zJYN%JxJ-_tu~GMX$6f068`wb5ixS)xYf1CSa+U6gtLQvFZRj>ZULF<6K}1{PpUC3C
zsQL!kvjLc#-D4n$tjPW9iFhb978Ko}V<DaS#*I<MoAfoZzt8%8I#I{uMkER!vw`lQ
z?-Llf-!e^^IzedxU#CMGapG&516|@%;UjH*Y{k^E$0>DA$1BP!UV0*>ifn!<LHC_x
zdj_3S;86wQp<Ttd=fkdQoI4D!OtJWhv(g;-xwx9WizjU>M?yU*mG}gbu@yEv7$IVh
z1AGJennNJ^tCa%?#40*??V^xdOzxgr$T%izF99(uQuT&(W$Z-7-{$5q)iiO4ni!}}
zdv3?1!=LYc?%D!Q#81JnRDPOc#Ckk%q|5W6S6y{eDT98K#j|9>)Dc9zMp|XljFG#K
zuRko{#>XWKsVj1uoffwq*CbKeO`W3tJGV|>>UQ<@!Oe~1!6X}!MrzKRQ)HTl3~^2V
z?dOqLwVYQ2)el*R1^f)_lE|YdXLb)oU~iUe|Mt6@Wl%WZ@{_VV%{t2|nl5}5RF<tM
z0oUOF&Y^g`dxheg`3_`0rrFWroiqJKCgc3+t?<s-aaE}3xE0UeahM3@PB-k-WL{a?
zS9D28SjOPNzo-M?ugZ6wzS?2nF(q2qHT<`lVQf-?dUwYirX9uRvh}R$k94b82^X({
zo%R@sAF1A|+d8dsT*TY{FF!(Eu_9TO`B%x*&U&C(I=`!#iU(esNQG)NAqom0pPGu2
zk^sVaHc1f8=f%BvLB@nycG`HtVhY3WU@a^tn4Rhmv2!pB_wWdw$1`)xgO`ap?UAxr
zTcL1$FJu)>MP`3GpWmDt2wusb)ViH0){=(HF|x%v14~I^=|W*)g)SkIwKULx4Y>f7
z5Y&tB%@~ZT-ciTxv2?qBg;U@Jz&NHRH$TfdkRI)Ra5pmr<@`|N6>^p>0!-Sf`5E>q
zQ-oYhD*vLd5h;<FM~c2Ux|#$w^m}tI1`bTp$-T@b^mHp(195p43v#*SwZ4*18<v=X
zzX=Z3+QNPMKN<IZ#@|%?#V>$vh={mZqN8e+Wic00Bh)=j*W~etW{vfP{>h@eMKas_
z$CpoPIdYnjD_XC=bZU6FCyPpJe^NV6Z7-P~hDm0lYrG}mTwI1t?iEWFc)9j197x@p
zeo*DM%gi3=>UMGQL14;hf^Ec87SF$HABdH++26HourUZ$<VnH8rBr^U3fr+;qxw=a
zW4fJ*UXvQaSRd3MA5o$!(|jetz4X6x%SIxqh=Ob1IsD~5(g-bH@V9)uZ3gQDDZ1xJ
z$n_WImqm|Dg-*C|-Ew7X&>LA#UMj%~ET_%zez!Tnx!y2`@cuc<MPr2P^mEeQ^>f@H
z-4wN-40Y<h8n2QQG2RYo-1_g_kX}^lvt`25tT>mdHT<}~VB4E?JDUqOLkvaB5H`+>
zLK!SQ7#pwNjK}oMupgVJnR4>i(QR^BWn=zQFR3upNe&;i=Nc+g%Hync1zsFDj0^o3
zR1jzF?w;l08An94JsL1*;t>GO5ij<eW3xb$$Bdc&3noXb<(|;cc13u^L{P>Dh+yLp
z7!=B#6k^~-w(c>dKXc;^xn;o+HBl>bXaArzuK;AodV-Q;63kbSg_Ul5aI8AhKDQyR
zbV<WqgXf_FL+OC~!((<JZxB)e^9bJM_JL<`n6#lNkWK<^IDqBW=v-&VeD?(D<zeD>
zaCa>-2q*-CQ_$3<nx`BO2qz*+b^R3?VLa>A0h&pu3aklkQ>B^F`^CLtHw8SKJ5hRY
zaA8e=N8zUoJY{8^ZrD__)M7yuL7j<5@dPCwbjPm?+RI+iw^vIgRB4!5%vk;98(s2S
zXvn3(@vTz7ZzMljb~EoC<(3wcN5J@P>y_<*!EwG9%2%|$??^2HyJ<D=cw)tPAb<2_
z!yy%6nGbkGz}t7sZplPzyO<y$No&siRnLj`u{+StuiPKbGu7M_i{d17?IL-&5XKK7
zU!F)h#9X<ZC8Z0M7&vBW;J<ivySI^hCB)DD-hp7q`U_G&Ted{)a8+x7uOr{VS}a4>
zm{rW>Nmu=C2CH)=b4z7YK?YexvqNj%yn$C`9XZ!j5rJ2~5#FpDGgP(pnS8}UHj=2}
z+9ah!ozCZ_pV1i!`5Qom=h~F!E<N=`GKnW()6LAJ(LUrxj0)zFJHCEprdvc8+n>)T
z((sO!W>PG_=q5gVLdcHrfH#uE!f}ni-E(6by-`>XQXuEq0~B<C>&F~wxEIX=gXvx8
zM;!`wDg<dr$^X<$B+pq*NKTtZ$%?A;iuddzSN_Pq9|^3fyx_^-teg~+_CJ}UCMsI+
zPJgXt=~s8HsZFOwhf;7=0MG@Ud7-#5De=HONxF@!k8x&8bSxppYg7y!a+0TxG*gc?
zu$qJjI>HbR%(-dF@EgE(8ku{H<YOIWr?4yei*Kc%Y$x^tO+e%>vn+AqmmZjba5!*!
zk%3KGE)vjXK`aQEd6mio1l9qj)A9%MUapF2?e99%`h6-wFq&8S4X?tmWA2LDdy#-v
z53vNFcY0$@$@gdKFzdY~KN9SkP0T!|O}zJ30;(e)8amS`mLAYnJ*LguTd`bB^{Q@Y
zY_yryC~%!(vXM}t&-p^X`*-h^TVr3h>K-C)JaNCGNmgE2ld{LvW_Cb-N=b)_h&bd5
zGl_=SqiPImJ`vxrRb*{*1iZ%ghLN&O?UB{q{CeR_June@zuckPu4<3VuSr2rS9e94
zPtm8DJcdODb1dVKDS25D#6VJ}vr-^c-AP9`$umGs_*NIrU%pdT@o9Ft7FTr9#NxVB
zZoNA@LD%VO=(f%AWv``O#3VPak1y*(=-*$J^EQT3UVfyQg1qiA;^>OPF6Nsa<rH#W
z{VK)fWSR(kw&%4kRd8&_zMmEKFd?_fyAiOsfrP;Y<SP=C0(YTs1E8(uXyE+=E8<bE
zG@U~V|4m@X1T+mxRC6#UJ2NzWYA3H_V<=aLhI1&2VNt)0-}7@1ut&JL`<RQU-*TNJ
zR37k~<e=WdBQ%qrrhPii75j1^u~Q&ZJ?}J)h?ZX!Ez-}EPQxXhC`zsIhwzdjAesH6
zk<>Ze-r+We=r+w9dA9=*Hq&H2Xgj#nEP;<W=uG_rKr*4g5l2-6N;Tq(eg<4D4MCJ(
zlOCyYUZ3{r2DeU)f`YD>pC3ccz*L?HQJ^ggv5W}pcAXJ(32OiFOhmsHL21rXc!3WP
z)bnQY8wVn=k{{+p7QVnCArK^B(O2wPkom#V#8?#%Oe>dY@*Bvv)(2>{lHi#|&dw8G
z`Cg8;I8EU`%*8O%g)~1KT5^A#9S`+zPV43LZBVpzomh-KVG7Ig-?Q^qc#!oV+`VNi
zSRW?v=x*$IV{k33>LXAFCA=VHCP=S&joCaR3p;M|5eGaOqF}Wrf7Pwm$LTRd$OXc8
z#tKZHejg<`jd_%m>O>{~aS+HW!{(dZ{~^Ntt9St_@>JJ)ihbpt$9h`GN+v2Ki}R_~
z=lqNLBJPIW^TzQ(VtIjW?vBm}4ktzVCwI>F19#d}gMIVs_IRC*CqbT*n_k)H-5`yZ
zJ;8ngWwzG|53L?4`q>BKL|wUX)gSm$bQ2zD1c0voA&|yJ4jGT0mH^<L5q1D!636!L
z8T6fieG2he?lb-3--6rJ<YM9TVN`=&uICj=K!!=u#%#u+x*7jFSF6seG1J3N#X_=^
zPT0puYV@5}iB3ZgIj21fp$HJb$qDKk0)R1)X2^58p^^)dNQQ7SkT}3eQ{$)JgwuSK
zu0xqTsAGUv31Ki{HuX*;;V$xLl{|on`}JefAauoWxTH+pp%ykr$jJbB9-xnbF~Z|j
znW-$PAMx{>tC!vr9MDyDTBc8(^<qYr-z~EW3+Bd7s^v+E2u4=0RE*L{HD4)ihrK`w
zWpaP`5ejDg+>H;{T!7e95$Ai)qDv`;Y7_J7qtD&)Cf4UmK?93|P20l+`*&8s<qWN5
zV~f11m*n<vg)al$AYXW+refmvp?iuwa?X>JKTjFU-KPXP5{{fdKlyH~z+^2cm!9yL
z&P)DgIj(i+19Eo4$!D<oWq@j*q3HMQrGSA?Zr$QJhD|kWX&qFg)vYa7ONZqALO*+y
zgwoN4g@2tze`j0~dhzD%y;jj_E}gN&s)A2PN_hF7%i0Mss+cPyE52QXik-{;Z@#`1
z?hTR6yI<kL1+CfgAGO<L%tn=j%p?e~g)d8Kn0<Wm9iQ-iI&R`M=bqfLa~>k{#@m@P
zaeQQ#h0S5L)x_xmM-bMvnVZT$+Sm)k8H9ESRQ<C4tY4{n{Znv8HP6d8EOLv10Y&y!
ztD2X69cDM;XeLsJr%u?GVwm%gpWk<};v|uiNPEL@tx3T`Hydvy!|u~M<2?!AnCiUo
zwh9r+HLGDVEREHZ{@>rc5e$PGD>0elIrss}oLTu$&hbzLHOKZyGVy(#44R<1oG{yN
zp(c2RmH%I^VGVW}U8PA#mT-t*an+Wn9<TF7MsntX{3pT36@L1DzvTc++c#fGv(sD%
zy!95{Sc>tqV3+W`S-doI2YRFlGSrF)NIKI8;D$OF@}>jnhvq`dWDC2y|Nj{&7|_Vy
zR4h|UPfV#Im?aV-6g#T61i<bUWRvXJ@$k?u!fxpiDzxMse8$F{Q~nCxKeT#ErU-E4
z2Y>L|FirD%N&gMWswVEV)kV9?V#(=IXnD+28o`$IK+NN>syP%B_ee)Vo*db;No|zp
zPH;dbR`_b=EZtS_(MxHzhyq1^sPl1%q&u_uQ7q#cS!$}y46RC9)VjRjeJc$mivCvf
zxWm$&_wj|!5C5IBNbeiV8uo|3<3GP$>E*VjFe2W!I^V^O94pl0h1~O>omq58_YvNI
ze>RwZ$2Nla;*Z<Lp5I4vLV<8boU1Doe#~<G=&$byZvbv)?6(xr!2MdcAncI3z2@J%
z{Tz8PFp;bl{&7hfp(^Z<-C*BHv_Hkny(*`@T`C?s+eha`%@2xd!x`>HwU!_Lrd~L4
z+UM#hyclUMqh;((B9tXncDuZF=wkXOFa8_AvC%SDwtY|1v?v<6R=4!gRTlk~N@MiK
zPvqs*MydK`fr^~pcWk{Prr5*9Z&phyXzzIDMLKs-i=y<KW3CU3OvkM#gv=c|%CHhd
z)8L$;Q<ALYbdqg+$w@`;9!xC$(pFT(0KM^nH_Wol*C-k?DhL7+D=*){+D5_g*`-?o
zgBgjZ{Lb+#WnlTPiR2QCE}rRkCrS{V&mX`E6DRGIsdqOj(m<|DA^=4=RbJLc(;t9h
z7^C%fg#rIR8Otel5Odv9r_{BAXbzhs(8W0t#^F0WJFH~S4$a&|fh_4Fj<?Q<Zf?B1
zyaIGON+<8`BM<b_3{=XdOf$DWb8zysb(M4{8OeFx8`DQ0a4Z@&b?{}j+>#Q@_m^VZ
ze7+T&ne(Dx;B}l!(^sOEox#kPhVRv!=S>upteE(i7N)+}AYc2KIUm8%zp6@nCGv#>
z`U)Bfn_R=RQ(T|q2VX?8Q|JjM^;h6cT36Vw_SBWDPmn8E?=P5k{UW8cb=mvP9+qw<
zZ|L1#E#UAl{we<u<I`<^soc(J07cDDi75qbH)css5|zQD4PolM{jWZM#Ky#jKIzZ2
zzx|aKInx!}{8HlUR%vuuGki|b6<)ee?Gu@GdW76F>?90_x`}2cR>_V~+@${exA!kH
z8X7ljsq&bUhFD{K(W(%SDSlIP`cl|2eG_Ag=|rHl&vejh(W3PnT>E5B5nHIcmhGkT
z+#P)5#E$GPQ&+nQ-xd9XO4q0=q=CWhl4pjChayk-L!->q;30xVqib`)UtD$s+F^vm
z+$!66I-yv#n|`3>^*dgqRUEiR2Gak74caI-|4O0}jHJ#^mfnhHeq}}^%h(bcb8xo)
z7X!G6{j?4>`V)J8mB50H<TCz3H(I3#-L4qr->){;OP(M(qXYN}^&U=u9u^N&9Bxgm
z?fo+pkUY$l7b*&i5CQ<8Igc>lFkIxtXQI+&%zK1#;g|$kMh-8-i$SH%#e^@ZA`sXk
z#_sDg6;HogV_7jEUKZ6M60J4YLCeVf<M$94`p~(OWOIxxGyRWG%IV)lpb=`m0XBRA
za}Vy6U?W7Te~Wy`Ux{-1nJlsUqJcMB6Qjqj$(S?Gx8~ysy2%}zeLh+pH;J$*FXOM`
z#z>}DuQ>+TQ9LLyoldV0dX=mqAEdX+TdWXH<JGJuf%|OPyF3qVU$59&|Dh9?%8k-A
zb<Q3USMoYd9_T;@tVJ_CcvO6ztnl6t@@4UTOpd{UR5AZ9*UXBU!Lf@$C^b_<_Ho*2
z7Up}(s!6HSEHO}2HyNw&O1P@JnmrZOezk~cWkjQBc7WhcOSvxaVsVO1^|EY(vSc1i
zlmVay@Jy(_D|@c~bdrK@ujBB4>a|4=?K3}5huq7*UH!E#5Deh*FK9I4Os?cTAbX0G
zcl>v*{&Tm1T{OdrctQ`m)|9rnl&|-$>K%Ll>|RYdl;U^d_gZOK!z7I$ajkGs8svz0
z>bm^EH1AI1`$0e4+CK1f0ldTtrUpn`uXy@BbCW7({i=&S@I<7oy96c_%>|4>xQy0J
zZopgD61ZL>AroI%NXeumR|<mMzZon}l#23g&+F2CD`Vb91Sk{J2MVB|`nOs@BUcwF
zDU3P|PG_d(Xh0nyGZ|AcQejF{Jt+3()|)eCYb3Cw0+&F#W8fCp8UR04shp=BSVI{%
z#v)WS<uPiPN!Fi%4c8>q!Atj2fdKzs22QHix_U6b(z?p7trJ%RnEmf%hlW@OXy{IK
ze7DmLb4n=VC0+3kO|*XF-<OqPM3EQFK>GH|N&iGEmBW)rH*|q4WwZ2Ijb22AxCJLl
zRcy$=<szfPvYqt@ftjymZQuQuj-<)he6=;_jp3UJq`b<70cZ(7)V$Q0xY44$qRAd=
zX;~`KwI{nIXnw%@KC|m(l#qmvK%R)zbz_J=S>H;!&8Fr`m9f%FNbF_B*!O{=6m5!J
zobM@LRhjSXX=W7a#;DLnnUIn7_=vZh&pYU|>1fHWII>3Fs;newRH)d`CFr4Coh>2o
z)>r#Rwd=z54ov}3*Za0s6}v`TS&QBcVZ0~)J2#qelCyWGXkQy~rP-{2eY~dkX1b|;
zO=Ynm56)w&HU)Fp*zNfX)TfZmhikfPc6OZILpSaHr}2^Wm=Q%+sGXl{DaZb1h~c4j
za#Gm$c!145!_U+%vv1N(Q2zn}tzhjUyyx0+y=3nUzyd9d0PMXqVnrnZkn(LYcdOK@
z7b1~gI439=I^Ec7r9jtdV{r)O5~9Qmz)91arI19DAW8qzX~leFL-s1+31QCa13w_=
ziPUm6^PcHk0f+$(p|y2^Pd*`gR*Br)gN(ObN-&~>9#{@M6E|*hJDd2WCTnqR=Bo{X
zM5X_6n?X3FXCeYex<1SKxgbYg^9$|UE8a#nO2c>d5BU<j-!5BN`J|f&uqhq06mkl2
z>V@Eu(R>L$1+jE*SNvE;kx&hWD+PW5VYS5HhF_ouq`yZWXMH!?6Q32nd;ClHweKv%
z-J%aOgq9{b@;+So7$Q1BF~%j~cq~e9&yH)2TJVKoRtz4t{Vec>iYU#vB-pLY{JHVu
z>K@*0iQ6@NY{J!aDu?m?g*Z(6@*(#m+>$b0r1Sa(n^Cx~GAKH!W3t|DrcOTYs%|gx
zA%beKz4SWbfUCwZhs(6V*UO=H{BBkb-_RxsBdwGXq1Q~kbio~Z6vHm`Gp>tiG)1P}
zR(hqPuL|^bPQByMm3jZ2OD#0jEX(xvHC-+k%zfEFt_Bgy`P3QVdv_@Um%E>a9*M8&
zul|L#R46oSo?#mEuNhj;b2V$Gj=uRf6Mn?%GPAdmO?ViRaCLd)x=0cttF!T0V#Csw
zb^UfESzw1~<B59Sf<&gV7B;Zm63GNC9w4yh1(;UXSc6~i7+}BZ;Qv>CtQc!tpfKzN
z+dfJA1U{`_ob_8b>(y#0R^b`=X!`An8E}T5X))yFyea8V3Ff9?=Le-gYJcXxWeSZE
zq!vW-pI%ry*07q-c!Epqr=!5;*|9*2?V0*p8%D_FKJ?tQbz>)yZE7V1iw0cb?=%F5
zG=3Pb9|=^6_W;Lr`&s7#m;+~b#hEmR{WmbHxSoysU|zK`JUjVyh5J>+$!!8iaHeal
ztxBJoqMEp3PWZr<K>SZAe5mi$ovo0Te}yGFw@Pk(aVxLl?e)BHJ!r*&>91TVJFFS&
z>tcMzq=1H{V_KBtdC(Z|rT9rLpgxHn`ANxn_a$lZ`c{mpQgmIg-;4aEMyjyQVl{qS
z_Qge4ubmR2PawnA3;y)HLke=$)@o~M*VtB8I$ZCY3gr!91`p+f(la~GyH9MZsj#!m
zjynB!j%hhNHd8PBkwZp^h4F;n9j=oMdcCW&`)tDg6zR~3H6JDA5n%!>O)4*8>4WmC
z04igwK2`683HeAlK*qh=i2im8-@0Gs$Zmm!(nnQHo#{*loxkDWGd;uIbfv&U?RZv=
z@*w^BPdFm|c$;sZ{DL+j8pHlMJ=7^*?m!+F5E>1Qik<ugC(Y?7C-&{IPsdzuq0U5m
z>W+RqC|-%s&X=}H?+H=&Orws@vpE27;Tc_h55QGY7h^a!iyHyX8HK0L!H-4JYJv=2
zmX#jr9(1Xt*MYkz(fZEkH&6jZfMEv?YYU_Fy9z&J$6EM1W)$|cWv_{x0@;-@gc%zJ
zGOZ)&oOJN+@d3AsD?5v6WloqZ>3<ggwmbnoN-#H#J=2{CdFh*`bq{OOFa7&0$G9hZ
zJoF1%jWbjq2*2GuV%fg|Ipn>wLeu=$NfUarQzf%KNvdTm?wmizSO%kF2Qe_pRRuAy
z=B<Rlm!~9-7nHD$<t99Mv~$hF%B#mrdM0Sed6u=vkjWGl@6LD)eRUZB{--0J0YX-M
zGNr^j=cc5d*qD5N;hlE81If@mXANJz>hd!*LWVFHyzNNntL>q)ZFk6d@bg2*FLdBz
zFGE3Z`DeL3`A_)?6AfJ*ms&3Wh)sOZQ6lNGpRL)$S!+B=n|`xpbh<|p7cB6T+xP+|
zJrJ({)x9?gqh{{?M~SbQm#;=us&7&FoQa~Kp<~@0dWTEWaZBo9Zd5wo2f}xsjz#V;
zSG~!5oty9ZYEon>z3WrJjg+Qu*XY+7vpNjcWU6+RV?=U4UL%{RDeZ00StN6(%g$S;
zbRWqbbiG((>6YQFesGP@W+CFAuF#QJ+k|`27*-k2Hu$V8hS}0B@-Cl#xfi^x;Hqup
z>%oThWt{K&9jX;|M8f=L)!n2?&0HhGG#c8X9bX<>EsYBQLy4FmQZJa@^#TTe*X%&-
z6-EN}QU<kCxq;vE5hA)912O+=D+U=<Oh|a24;Flt5;`<qsJ6b~1qtDh<i-O1IFv@w
zfhW}>8WcwjVBRsfSN$4w#IEBW9&$G_QN=AZ=CL?26>u&=#6;nuOzp8uO(YkgKMtjR
zbJr#E;rqGB`4RdBgDHYzGVEu&jCQ+pPE8cFy}ciscgi}Sc1z;V30KD{gySeq#r&TH
z5Zd>@C4qDeI~FYA<i42C*m?qrUt|uC8YHz}0c012fk&MRxtw|<;Vi$2;9n{z*)j8g
z0t(Yi(iE~@D%=S1F1J{o<zHqf#6cZei0V%^Y-RNE?l>Kh&97FxpO?;*<3V<ZeZS=c
zK(D-({&`n*4g4(LCz#*&>&2l~sPydWAAAnX_jx<U@=)mJd&-l6E3=#CRoH3B3wy~M
zh0@pVNtFdNMVrjIuG}BEbiQU*Iy!<w`KpO7t9Y-{d#}f;O2i%JFi|N&y3nhlp#nu~
zz4>Wz^cEjo?_!w|R+u+s3Q``n1Zh?dDa$}+&^?Ic86g^?mWrRu{s3iK9!wS9#2cR_
z*l6{Q25I7SQOHP2mAjeEhA)OLU8HU!C<5K(VLhHPxV^89)<%b@i6sO;bptvluM`Cr
z6V{*c6`qzbto}2b)wz>qBNJqN*Z2JCG)drOIp>6D&nRnG*Bt-t6Bi%1ahF)c2uZqH
z^CwHfwhrsSU@<zt#M5Yzn56JPOz>5K;eY3P!m=9WS<3^q1uikjhu_pSl~iT_+OIvt
z*z&~}sV$%tmew*!(M$Be#wPY((v3~ulsl1S3=Rwz-W?v<SScPMNEA(SmVz6>8+>U+
z3ee+0C=j#mDC2=f14Q+4#n;nkOK|?wis_?Us|TF(d?bxB(U3#WcS5AtNy1?2xCotQ
z=mxLtM0|$kgq|aSDo^aNMx64y19ghKxGh}V7~j}wt)F)gSFM?bW_V79z-)CQD~8SN
z1!CF4Y8<%<WU*`vUF!W8F9!ZHCV`NuRPL5i2eY?MdTYY(I(7jJI;|bx3NY}K+kn#H
z0K)8O@_t_?-i~59yuAHC$B7wZ=zZW;t(C8r>7G3);QD83PCXOh;bC<vMM{8lt>E)v
zQp<;)(9G)xG8dz>9xMoDbuTJG$MVftbQb*AWFL+;rI$IUYM`emf(5J7a0KN+TF&$l
z%W1M)ofhI{Pm#ch+@C0)X2p+3E=p4hop-N|D(RaBtkhqS$0PH^fnP(R=<cZ$&)t#t
zcVw-Ron1jD=Ay_*d7dM2qO{fweusaCr#?nVnW0k#hwfQv+Ek>!EYSG|9YZJwhP*xZ
zZ7a@0=4W?YS{f*leo0`dJ<y5q>%+=3cOSEme>cY^(2(POX746f1>UCkthz+P4F!k4
z)0y6z?m1<q+;}bXs@d(c2u*toeE-s$Mj4;j@yELlLhY65DB^}US2jb7w6)cgsk+}5
zIj+94eY%j&2j822{wwjOOqKuDVGi#uqi`#Tah%7>0SuGZ=n=GW{`0}jDBfU)cJ|PV
z+)ExyGxAT}oM~k$_vU*%whWAoWTnrbFdT2!rSl#B2ch|&p+LdrK$CteG6P}~O_Sl~
zKO`4Z>51iXBa$^l|9M8M=7TlT#RHq9cLWb%=D@w52Pa9G##g}d?A*Dixd1)#FMKDG
zk>c6TA;m4^(<ACafiVL%07!Ov;y9rH)TzFvZ+e9S1FOaSX<4xBTmY+XY{g`=)^sh5
z9YYW~k|ZK<t?(W(eLo6B2>6uFNZ8JM;ubV)Q5ChJa~EXvw3r0~^VgABA!H~3MgvLv
z6EqrBMA!Ks)IhvpZ-kfMzuWrZm5gt0IRqn7|DMv_=L^PsU~i~WmghPytg>sfnEJ`i
zGIPv)-i7+%G<&_`J&#Coh#`ThE?~ok?^KLQd%{MBdLYU>B2*U^a>U+6Dr&F5*$HL}
zQ&}Y!>kL<Cc~23=){<W9Fw?-^D7y0c1^NJ4x?Upk1HSZ*x}N;~fEJ3BAtj%!Dx9@e
zx!oD7z^K2>^hwA)TQ^QkB=88G6m{k9FA{UiEzgCT96>Xbnriux%7-V%_ap<%VB;Qf
zo)iJ;D^9SI0?f@#>5(nw{3q|V7?<8TV^2Qrqeewj2-fra<=LLj1bc1$1?EE!yH#HT
zF$k#z6RaD_xRZXGX=+--?qMX*y1Nj!BN@Eg!*v~%;J0zT<NlZNyS%j+L6V@C4}$A8
z@#AEs=pA#AufF=3=A|sRgnnWVZ-#FQLfro0X$B6gg=`&j@2%v~y~qQ{icHc1OC{ZF
zAVMjqGnfLRg#_S{67U6pK%kN#U0MU(5fGDW?_5dOH_0x=fvU)TR!NH^^U$Dc3wW+N
zi*?kv?F9*S!|sl`)*vO5^^NAm@U(6A$!zSEu?mylA(elm+6|f~p63713k=B^AV$PJ
zQ$Bf;)DD*ej9(4ld>0gJ$oVH=4IqHN-edn<y9IUNUO1D-P-#GgWkHjV20T}P_BKZ#
zsRmCD^v54-%L5s(B@zQ1v>xU+F$qqWFtT|k=_7dD%xxrB<zycF#&%V;iz$bi-b{L8
zU*<EDf&xc6L(Tbmf{T=VCkuX=H;KXUc67xodg{M(EB5iOW_J+&?+4mkg0O#~gS_Je
z$E3%9(`Y8{4j1dE>bG=$k`BddmL<#$953AWop9$s)q2;ZG{+VXd)X@8OqQN&hGZHH
zv@h7#c=;rP?G{ZKpi4X{YZ;|(g`Ga9$i|Pxxd>&O`ViJsQ=3@3Zzx|3kJ%|cEx$I4
z51HH**d^Q8c+d>J&fdH((hTdm9<cAM4_#XjIL#sIK~?P^HeGH+7s(-eUr#F^gs3}X
z&1M6>5$&@i_%tF1$@H7=R5*p2^X4$J#KB;dkBNWprqN2<cg9W3{5-hm)9Ut0%1{w+
zoA;tjAK{wQ%r#v|RXZ`vI~hNI{=xC~U=S&gETWZv1pV)vqSBF=p0`|Y);IgQv5rJ}
zImNEZT>*6gS>fMooeiRI9ZJdl?uT*Qk~MqceG^F}Fi3CpRL`Hxp1MKOb(f#dm(0s?
zUCExrJ0&@S;NEDaX#V@t=+ZzBwYxE}C}S+=KIhsovyBh{*i;1lVJNK_9-HIY0*z5<
zy0b8lLw93zBUldvi4+#dFYFdjq}0Snd5687y=C#Z3b<TM!hwS|`m(<<-m@C`vgy|$
zuy7tMMegtnAWL#IvK;s=abu4ysu+XAFz5Eq#mMP!Ip=}rD6|%+<pn~tJQPDZMJG<|
z<~7=%>aD^9gR#&5CYhGt-l4q@m<J#AT`Ff+zE!E6?%mS~&&-|-s$sLjOFl8;lL{hd
z=Jjy&BxowD^7cK8Pf@uYRKz|kf?>wUXy8MpBMt9Ze2Z{nFexGQ8RA{uRVVP+*aV~%
z^0;5A7O7XrvujfFx9f(X+|XEKx%=Ut_(Zs_H>Z5^-&@%znuEYMZ1AZVm%B$C4{8=I
zWG(19tZ>T)pVBgo<^=oFoknCobZx-x&0w+PUvorhw;yIl3EC#bWST8ooyM2PhiH8)
z@I6$fcl4dlb<i3S6?V_k_4$PJ{w+EE1<F`M5pnxr+BTzXR7nWI@W&xOwH)#~c>CSR
z2&v!+%X&9a$#BG)=)F+cwwYi4`m{8nl_H&-@jltm+VJ+XkhcB~!|6}diMP#<4_V4M
zBY17PwO$!;K01N#ryGiqoN|jV7BX}lqQ5M0j=!xoM0_%B>_XgJYP#+AHFk7ss2v}E
zT?B6a1=yRpj?~@y?|t_5#08DNuIYVnIsU$F*cn6H|EqQ|LxgunR{LY!#qy#9xwOq{
zNVZy@j3r;TS2F6|BHwf6KNCl(pFRi4MZ0-NSSxlf5ZyfT#n%r09PnwI6A%>f0DVEC
z6`KA;R(rgUW|Sg?k4aPkq-Y~zzj4F&X)x&~$hc~@i&o)e7udP8ZZ?3DW``Z4b-PRt
zOlA-8?{mvQ_6fL+EDNR&{O*83&n4m?_EA96i_O>UgeXL1{N%#oHB*a)9Z@Z0QOm;6
zI}4hu<XemGJ@d_zl(WeaaXc@=Bg!p?eYq59VL(&I{>8$Ah77P`Uh(fVCyCGK2=Ox0
zv2oyl#7E^XmCGbLgnGFKkUvajn|r1BJ+*@VIugXyjtVYBTC8X$hLd~6hEdnq2dAad
zWzdcVc6q=&P2Up&->wnqQm-BxS^0c-e$#@9(Qm$aaE-0QR+F0^0E3Amzu#0CGSb=9
zx`sNYmcE+EySC0_xk4IAViR7tu^?178V|o{ZCt^@en&F%u^;c)RnW{tG5Klvb3!UR
z3C)QtO&x_e`>c{C*J+W}SW17Q-xEl=2y$Y#Hb}(Wd_{LQkFF|PEB)LH1E(n};%BPC
zzH{5_)M+WJbmhhCu6K7E-ODout5(SxnJ!bAmXFI%2?HVJ)*}V)lqP~JSJLmH=7vKH
z;w4Naw<zka*)?g6ua}P@m{F&hYt|Lv*Wgcju1<3ADFz}L+<b|T=cGk83_e-Olag2x
z^XUbe>DAIQUrP#4bUKS6af!}2nRx!G@_bxB%NJaVf`Ji*-K8U{T~!ZxOGx^nLybQy
zYIzARvNEPTWtLeHEFzGd-l95pUMcua-Rwy7rh87<TRR>>1>?yq6kDc_v3T^!yU&%2
z8YW47E69mAm!1TkNbcFhEh8#HVp`;x?T5;L4@`8+#&p@+kPElLe8$w^9dsE?eOXzo
zHDKhZoRK~w=WGMWg~5TSn}_C@|F6_N)$0LU@c}m#-7a>5Ir9;wp)1yI>of<-)k=<z
zLqrdtB3UC(k)d#mh9Fr5UolbMzTQYq9o8HGp8k&EWAEaVzaIa0j?GqkIZd59LQtHy
z#_8*2JvOhZgmZF2bV0JP<PR7`_e^60+%~S&{F?ya`IT4o9$OUWq7;cNIaw^+4aA(K
zKQ4|W;q-?zFM*TsQSBE&v|9CfR<%~6PRs8Y#yy~>V2ST=-Pl%V3m14@@#<C4%on-w
zw)dC3@56p7=2Uf3R;cpo?l)7rR%3QUn0l15)kPoJP6{Ury5|!bJH0JgVlt2(S$lG>
zJ|TKtHalgILe#oW*R8@ZI%y$FkW1eJSWoR^Uc$-)-1I{p)<_n+V-eh46O}Y3i-8F%
zCcoSrxc$wR(-6ZekqGo;;qLFM$ZobZr%%V1?l+Jh);-8`Y-{8PF~MRHgQ+bR%ufa;
zyx?io_gAkpvni`$(%=ehB9ji~%_^#kPc%5yvJs`x42Vu&sf-KW3}U@pw6s94&^nPQ
z`URhvA#5)tO;Z>kvG4l1P1^8;HIhM@*RO1^<QW3%=OgDCH{G{G-KOa`XDLb}CBmS~
zzVxJHEimIr4XZu>DB;YJi_phjijp&%$h7*LCb((jX;@F7Ol10!_5Q8ZKdsUbJ(5h(
z-t~MU(;*G}b|lzVxuE@recE+obsyy@gHh8)tU@cs+l)7n|DNR6??uoZ!$MWMA5se%
zCO7L(#heeO`0w^#TG~sFWCgOdstP@hXc7Iz8~V6`i;)I<QtFHq{6xb@Su%=WH<k{H
ziEqikoiVzRfaI2ZIUO`#`ht4v%&)w3OiSK8&zX{Wn*{>t20CV|*|2h#2~HQ@r$?s}
zsX0oCg!CjBThvZ_I<Y_QGE6{0m)+qCV^j>+Smt-7)0{oo;%Sr6QG`)9a@#r4!%2FS
z+e|D>y`JX9Px<@rwbufn8nk7uw&C%uGzt}85=vouZ~X4xmB_y3Q7J=s`Q5-UoR2J&
zUDMDJ_`D@$BA|`V{j<Pv!++<ZG*cQxREmC4(MI+%iLb0Pqh-*WZnmZCQTE)Id|bF^
zYlG4MkX#ra<a%X5eG3;eo*ps!nsRC2?wCM6D;PO0*Be-Er&spzaR}8%OvJN8t-ZDM
z{eRq3Hh}lsH4k~SW5fJCl$3w9yNXFN+RyjP#h2Ck6#$~ud(8Wk>D_ooy6pJdiUhv&
zyxVtuy(F8;{ogvj`m?ebiRV#nvAnwU;vl<Y68up{@>CP{>Z?~glBfT?se5U=H~Quq
zuW6G?UgJUi7Xx2h@d5pK))r*UVTfdBu&Acf@4YdP!KKM0ZyqD9^7WF3n$TenkyhhS
zJLpF~QJI+OH}!=<@R$7l8NaKkIWPrT#}0~iq3Y^4vS+d=^;??O^lsz)afjeLH*NC2
zT<NSINl_fN;^w+-pC^D`7{Oev$*&u(pSu?Yg)Z;J%74od-UFn*YBZFFaLNfnEFy)7
zo>dz3*w~yS%$k#F|C}fBcbw=GhAN2gk!X$H$}#ICi&A+#k(dcG@@RT)<)@rpRf{VP
zq$YgP?ai?(>=~Cj)+jyAPtdLs9@O9?9#JRO_@ku1YQmsUKz9eCtS431v(#Y}tt|mf
zxz989u1>Tu0(%1MNBMJq)K{{9R`3t@YNu7u+CwQ(%nIV~Ab#&b5$s~fo`7M+@H6W%
zPqa)~{-AGKIJ0fua^2*IxrbX2!D=1JeM>1BlF{n9F`Kd3>00UGjEcOTZ0+8SfcjrF
zU(2pp@pWHPaoNrNC$u_TQsgV?vxI$bj(ab|*j7O1v1zYM@8xY7yp`cw9OB+%L>EW)
z0pk&^#8#aetXl&0e%ullzOH?-<zv-}{^?FXtuZ(W<_pu?5iVKi#)-uP@#Pu6)OE&D
z+1ze5IX8P6sCdF8THG<OnihzGADr-$Wz*}bul>oYbbT3`m7ByTXYkC>^N|{~bi0*k
zfUGN9Gg|f*C4SZc#&*jF^QxMSEY&Dpu&=FC{tYXNFp4{G%#CHt?Hl~avu4#a{E%R1
z%U(YAb4A^Mdvu>x8Br7GJ}%XjI(u31#usTer?kaE{$b$!V8-nG9`gDZYY(=zi9Qqi
z@HRS{Wu?qtd<wNG%q}uz5EIv$ev<Rs<b-R>2sg}v+QfTvej~g5HCgrtHbhoc6<Y76
z8B}x-?QZjaWO({msV}uqrNp@H0b}~umm%`~(EMnkLh4h$0)^t6_OarqjU&>?WPY#!
zl%;Gbb*KFuPf+T3Qc>Ej;N(&%_+P4P789cd{yTS1rpWUZi4{W@-Wx04)?(P?p+^D7
zHafnw;D4j=%hr0$kRdX|(jQS5tx3gNjNMtVxeV-kjyaC(z9VZ(LlO*<EEMnsR`7?d
z9s6YI_*(3dWY3q_>8;4lzjIy{7e#AIMf3LXR8M<adIF~vcIn$azKGL;HKpy#0W-xn
zhZCO9@xLl<n|{#KK5w;Ve)Cax?Ul>DP<t_*ow{kUrrKpz00S1FOHaMge;FV-MD+$O
zqGD2S#0vsaQ1Lw~OX7|Bc?_KnYxl2hEqhp#xfj19MuI*lP0Xl=<tM~cYH`Ic{$?**
z#M?+N8H`Ik@2#>xtz@L7w1RvIufL6m<8?m2$)AHe-NVi+Q=jDwFf&>yYqs(B^j6!a
zSA2*ax0JATC%aHd6{oSxJs!o(UC5Do@*De;Ed2rEJ&+EQfb|$V)a4Z|AjF(#7jt%b
zPkj6Y!P|w%@-^Q6<|gbXqx+EETO8t3@%(9SzA^bmr}y*|FU2Po(lv(JUcpR}BFt>V
z<cG%AMV58=mvsa2Y5W)ZGajwI>7duHYU#?G@~?cpALY`y`gSF%-D_O4G)`{~Hipf(
zEKJj*^kGUgTG{oE4_ca4<do^^m9Chqr;+Ho;6Q<EYh~Pa2F2>k&C>O4`9ABX<>UXI
zyX{DW$-9hSo#?r)<2SG-Qtz9%ou!>a*7_xR8kb5n`1U7|_3bAgk=0G5lweQVjk`DR
znWP8xGNX5tMv~L^BI6ZJQ14qsFteC}JmjU#E00|6aJQEc8<1}wzUs!c=4M&2)<~tG
zPSl<M?NH`~w{?qtF1clWzY)%B;z-LUN^mMB_yb-;MJ5;!B>Ct@9>%bYJA<u!0xz`S
zEyzR#-xdI+HcO?3w_U!PEvacvze?lsUd~BM$KS=M2>q&92C2O%Vj#X!yWRY}hL^!6
ztC~iF86k44!fwm&cgWM68@6aJTsY$YadaMzRPX;E?}{=rPO^&3V}$H7zs7NF+1cS(
zWhG>vo7J(Mb8I0ymA!ZNv9d}|WJMiy=+H^WyuY{KU!a`h^Lf8tujljec;2kD#kwv7
zY9kki{@exONJZp!aQCr_7IE{g=d(mD>5H=uEkqAygPsDDWpOs=0(c;TTcrWu7U+UU
zGXN$(NC13r_Ny*Q=5!33Q&DHeWy4bQTmgoi<>exaDOLR?bC`9rt9PwjueIai9?hS9
zQ>?Xgf9-{F?VlabV}|wqIZfQTJ<H%L=H@o9=ugm_5Ge5*dgeC-Ywb%#6xo~(E&e`5
zA=Z8CKVJXHCXdy!D4>?%sna01zgFNK8Jc@spjr@YH)EzGO8P5O{7*~w^VrtxZP}k0
z_YHi^B!BmN&9(mM3NZfMR1?LxRJ1#w>7kG*<V$<>)?PfeuJwsXMjq^zJklTym3Mi7
zZS3>mzK?6AcXLf}@hMk_g8ioDQya3MSKCc|S<#T?;>&`ejTfAoL6zU&dy}b)evVyz
zWsomQ<l6&YVJzEL+<Ka~T1C!=jqS5;3DF2oyK`6Go-=jT+xBzK+|#K4n_;=@BeLtD
z*qztntFgs@vhspq?K%4gGiLoyv^G}hRE{MAS~L!W7hcrH?}bxh#NO3|xmYKsfl!rJ
z4azvViv5@~p_(M<S_)O<y5(CI6)2%0GJWc!ZhUmg#h*6b1<#0)ha);>G)?3^8`Y=r
zMt@3WliWF~meX)u=h*-X$P0;}&dZ(WFol6kzy$YJI6ofjfKQMP;LiK?X1*5kel=#I
zR2;azU_ap{5S{T3Rm5G{FU7Mr-VnA=1zyk?XKd3CPHDhTMb9C|`=0pS(;~seApV50
z=yYncO^L?d#V?0bBK7BAj=Ol!?gHZ~K(nKOF$NzXIC&db08*VN{%-}$N%w@EtTJ<Y
zsmn{$6K76b{98upOSWQ{5hBb#q7oK<HO^9M)?$l&#GX=ix5?<aB*v@Xzl11mDQ7|G
z78C7Ee8hn;u3Vj2l_n$pw9-5WDoiTPca0{xokpvVoo8ye?`F*2RYj%pY@~M53XSPE
z)QqWq+ZXj^o(=d!?*X&wYw%@FQrEAwwLcpcvHZyT(8es>wQ*<o=UR(x@BvxD-^x$v
zvan2bx~tuQ^kQ~=`^=KZ#@##Rgm1-3Gr#B0QqS_PEq5v-9$lXKglY<r5h@!;>Rx2B
znthn0VEmm%QHp5u#8@C!Z{7L8?t`ydPx{3O<ULT(bZyGO-|##Lg?yyVeN}eEYn%8>
zlRo#r`$|P?g{M(vHBaNL)@%`AvM*z-5*5B-GO-Q)w~}BwOPUzdE#P2E6c#Y#m}!b)
za)~2FrRBVGIq35gSx2F)SGibn^W@xT@Q@Z3K!#+Bq<lPaO5aGysIlEay{L%$MY;Cb
zI(q`2pzVWr@U;JpSP-{Jji1<&NVe{Lh-AGm9k0`{r9nfB65Q{yevt)|=$IGSSCY^L
zTyicxM#(^3<zqTBCi&lmR}&rV0n-H2McMn7PhalJ+Iobcx{=Tl4m%y8&hw#`&1<hz
z-ii~u##U<tFb4w1a4vBqW7OR-i*;T)rZH(`<-D#rxs72hc6Th&q9YbVlYDA<w1Oe8
zRS7?5eis(wdZ!<YQG*Q(KpTV#emYbW9zGaiBg6InsQ&pn2EKEEDG;hZljvh<c{_%7
zHk!)&9QEQ^wg`UKc_Z|k{DxJ_Q(wtIB|Bgp&Sfv2MtR^m%O69#Q2s}vlzdLk-PaNX
zL2UD^AG55f-nyKwd9u2K{r1PaZ*2B-W$r5AGE+t9WvvLq86gF!64lsYoX%I3JU2FM
zn?UwrMuhmsM21`UzSKxTXjfof#=B<84~?szRWx(QntWBojtdN4=Spy`vWs`vl)&sq
zdu_yH+=>QGJrql`emz$GlgC>1KF33ZXM-pyR5lT$e+w^0GJ0uWMTbO?(<5nGpHI7e
z-=X6WG@cC_9ci#dC3>_@vXSQU(v%7amX#1?yL`M2(QBEt!$0uodgk9}d7+mYZSB@I
zmXG(cnevp5Bpp3dhg5Q;NRmK|*P=YEmO5F#N);tRQv0K$`KTsLo9h{m1*XVWlEQ*+
z{rj|4e!<QpdX$&@whWYqWh(K>y94U1-a+qYO1a#QGJ{k*iY`A}THb1N83d9?Z)I(r
zFxGj?)DL}By>g&`7wvQXZS+v$-pytdHqKt~>1KFCB;VN2_iy0k6hG-p-?dn@zHCL&
z2ZbqbrK3D{m96iyJ@$Y86W%4iB;Ipn>{Whd=)Sh@>Pb>kzkfa2-iD{RDc)B2h$X$n
zPbd}Ot{a6L?Uin3Qu+!Z_TVod#N0&zayGG}j?X6aPISNe@4{vfTbB>%l*&*4_MhnY
zBj%e@x)IsXrbXC7s*36muUKOjZ%WbSmnns`(b$e`Xd_e}B-JSM8mjD4@MSo5XjSXS
zSKMh{@G&%&_*<}rW7Z$J@l;M_@o!&KNgra}L-{N|RI8gEZwsO-J<{W)wX;`D?;e|m
zts%&8crAv)r>&NC;C$r1Kgrif5p#}dQUgUY(|EBUyE)<QqJelq6z#FDwkaBN>;E;R
zrFj%EtDsS%KW|N=aN|V4SCPWqZoU~qP9pR$l)ys5T1`rnS%X{xKty7$w3Jy(9*orF
zW>Wj9{HC8Q7itB`%zW4L-xzfNwO5&zPa2)i<ZNwn^)ygv#M3==@yii(Sbn$+K@D`U
zmAX67y7eo$2WJd+<#g8tke4GO^}+@c>p#YN*o}>6<lBos%sh90tXh-T?)%Zc77M@b
zU5Pb)79g3XqrYw=$WAwvFf?%+>(|ZHX7ib?JKgtnM2V%jY3cP9qT53Wgq4gUw}WpV
zmeJYOD%}|!&>pw;dpQ!3vSWn$9DOErU<Dz#yj9>}e(Y|asIw%H?$WU7pyEJn|BnFF
zJQ&w1Z9Zq9uDj0l7>B9XfpSa1Y@YEnmUv9H4a&}qPPple5MJ?JSn&>*Q6Q`KjrHUH
zo;*6CzVrEYlRJ#rpITo>|A~V2qO5?q6nX7oko$xqE>E)0YeET$pi{ylAo&XU-$iDj
z#RBvFZu`yuUFdQx7VyO!+}UwLtM-wf&UAAea)^&w)UCDA>?HnnlN^dd<(#FOtrd`y
zi252dJDHkzUbNcmwB%{Vn!{6aDD^G_@lQ?z=n7)TRMB|q5Y4o4P}JdOQCv?Z{&&GA
zc);1s;R6i!w}==jM;MC@rJ1$C|4@Pasj77-vMu`BazjsRomDES1%SW$5FzO@n|!%N
z)ltblBf16Uz0+L0%(IcbDIqFUaKS~X1hUgVEN&Im22tMX7J6h5`&GpbN|e1zmz925
zKM|cS2-SO^bzpvY&df>pMB}IkR#04mf>76EwQ0edu{YS+0e~axL{RV?83btcAWa4E
zAkLS<1yyPGfgKN;9Yzban0Zq+%RURJrfS8=>)r;6GR>BjR*NZe$H?opk1G`n(&&nQ
z^Ha;;wH(x;K!w^Nf#}ZPZyzPNRw%MdUu1jQ&@Ho*{J<I4dv?gSQ#pZen!JXRb5xHX
zPtVTyxt}b;dQDz8;jPa?eqAK<<CBgPj`Wz1zh1kBm!W^v37jmQMmkZS+2{mn6rAMz
zmLAQwqVPwu)ckkhO2aSk4hNv1TA(Po^7*vsY{b<4l>Lq&!K;ItRB&tWn9rO#aOpNt
z>Awp<Ms_e!sa<ctQDEcFUnD-PRr(UgGqU-j^RsZ~YCcSeL^f=ttv@;dw6$&%j+Hi(
zw)#AijuCelhOFk;g3?QDJuiAhu<3njR$njO%04ivd>?^Z7J7Pf*K`()AR!Il(zX@I
zAg_NABE`x<cO=^`I3rRZH<ofRMvM!)iHd`6rW^<mN3-23!9>H-;!Al1b^>7(4jotK
zBZbR$iVaAB6<%XkcoJYoxrQN%eEGBi-ZUuu9L%gfMgq^G#=i8=(wNL%M)KH9OozB#
z1e?l5$8op*4?@ncAg4IUIVe+$7C0=tgT-jXg2swk=%pJD8xFA*5S`*LlqDW%Pi`&C
z{M6n$7uV~dwoZXm(-EAn^3Az<1OFG#z7YQ$#Ef5=-C@&_oJ{l^5%;y%ufcHZ!x|Tg
zgv=dspR`&^rZ!yJEhvB9K1|8!Ck#_OUZf!Nqd65~W7Bgq3JHB|!1oa?Vz!+=)!m4!
zPtz7kYOj!0Ps*WvnEgFt+JpV_h7+DI)512!zK@u2cr4A9|CPe!d%1*TLdygc!f3#^
zFBdZNn;93HNA(%Z!r<A?tbd?EPpdb*>R_AEiGld5-K;m*XMJ;h+YO27bH^$i1aQK9
zm+&12U4rs$N|NGy^PSSBKlI**%ZKJ>7*hqF(SVtv4ZKWZ8$3=mLE1-dw)m1!M(N>#
zKBsidk;Ghtz!v8=N52DqpF2P^E)iGJ!(C^v1GXUWc@iY|7=Xy2dYsQyHEz$A0Q^Qy
zZG2p$&aaKN<0v&QXCpH7Uox*&eBx#pm+Hia=TxD*qI;6`P!Wr*65_PNOd)OO?WE)i
z(zzE2Bo_gyt=MhJNncQB?t~_2oZy>cf=kaf6=JvGYuj*RU~_^KCyL$P*LfiKSan>#
zh-7?Rae8Bo%f3%xkg1v>`K(5%@EAipUPA=@ON`Yf9!!zz-Q!gU8q2A*%{3w3Z~a*N
z1r*M7)$W~1e|=x~^)ks$a<q;QbFw;^UaYKk_hhLcI+*g8vUm7-9yutZ6Q6^wx^(uu
zHO?j^qa#;uRmc&2O4kAYocv$wrS{YPcY(<A7rEF`creYH2yUDtn1qCMF6hLr_GeEX
zi8Y;OQB0f0(>}s6Au0j^i=AEDIkC}+#ALF1X7Bib;zk(IkOEAWP2q!OJo`T8dXR`m
zn5zIWhW5B@K#90EmB&)RYPeYey9q7;rKKl=gwjOt#<iK%jzX;pSYQt4x3dGn`P<@v
z>A>@i#OOYdx}Wm;AF1S>7L&^yA#AOu^u(a_3;vBd-=2I)ci9k(L$mSAcg}saU(X|e
z+b%&B)v${rqW`-fqS~m+%Oq7EYH$Ktc$YJ3rvt<0V&pBA*8TuOn6BQqMAP}1kkD@f
zBb`4UWle^n2&m+*wzkf|C?Hnks*VGoE=#MDDzAsKCkCsG(?*3?{>6e=DdfH)cW#h{
zgUAm1rKKMqbuzKM_A>eoo|t<ZKQ`wjF}Gjr`C_V5?3)-r{FNc^JnEDcZn}H#(Z$RU
z6=5tN=KN^gBmy$Pra`F{ChKqMJIUPeByMDYy<iVsOmdz=A?XUkPF|cz_VM6aPk!q3
zJ_P{ScYurrrM&FBe42hyM9XC6xez%V#d|3<x2~zP$Ov-{SESC0PS0zaTv^1v%kn`x
za+I#@!Zq5XYBJW+1{vH!;drvdLUys>^1dRRN{}Hon$;+l#{b>?6F&3RXA8o)!mLi0
zR%z~@nTnqh8WPfUKLts4NV+q)@FPl_BWfUHfH=NWnmqHSG@Pq(Tr6BYuPBdCr;x9V
zJBsB%Ef26=ODUYav&7`hupM+5ApWM8s&S=_M}Y@HfOWZEKX@Y&EqLX7OwHp7;gEhv
zi*&=cp%9!Zb%~I;4$x^5rnS5zn#MAqN~BU{cCf7Ga?DxD1HXTfHn2ic=%>Zv)CZWU
zZK#&$xdPDNRDw7+51*s6LOHnu9;HG)gxy@uT?AzUK7kUerw5`Bn^s<F5lf3k8rbL`
zCO&?B4Vxg-b-g^^eqT)w!()Rk?r;fIrxCBrjCr-V!}40c+_fMF4{_mF+BDF(_Zu(#
zUKfz~|3ol=Y}=<{KT{H4|5dx+BqYagWa>%r+UmYq*dhNz@Fuc_o-jywK%a1&nL!8C
z<KD;RdE^t*Y&UgSplk25gyi>*)#Lp$ZNU)b7X{ikt126+-iNKPZqPxk{qg*_t~q%t
zr(eepc8P{i#?;Qk<r;$1HZ041q-;u65^_8BxYcct|4l~PpG}J)NQH!~rvn;XL$uC`
z$P|h5C=t+{gIrKJsB`5J;hWwE+B8~j@3hE*-S%HlTTKmpt!@LSsXSVDOv(1{$C8Z+
zk!39)Rlm|4g4@{^IM1#DHX?X5f&nC`rqeB2N?W|sTaB}U$5r83BpP<1V^Hw1N+kj~
zoFfaMDyZBn%8N?BT?;<lu#9Jsd0WkfXzcRf!x_rYq(AUZD8EC)jmF3#Dyef|`VHPD
zr_sK#X!*9Sk@SnxgIAbUk{{<0jaQ4gf9<c@yb<f`M<zYr?^~}+K9L6Lo2lZfkKde#
z=Q|&zz&3BMt>KUP;%de=-Ft0a-KL~lO=EzM^$fMuNUI)CKm&jf-^V--)3R;vY&c$L
z-(=WC%zn|7TE4?w&mVW1E4+wd`K*(4lvZc=Ao{`Tzf1Mj<KAA8>5C(cN#77@)M&sm
z(E<Pe7-cnjpiBn8RWtW(Jss1_Wg-#4LGi!8{KH3r38z>2sN%z5k5x2aYD+>zpnq=|
zlg0Y_lW~~DcJ?8Epvj18X2TC?oHNznz%0<cH0<iDv1oGG0Z9A3lw&~oMPTG21;`Yp
zzh(7dyua0T#UGnIQ@v8MKC8mcj!PT*TS0{C!sY}Bz4RySStqsW)-nS>EzkbD02&~?
zKi9A~>k|^1os%A(*%*)3X$U!!{N9k%{Brhvq_LDIlwu0qBY9`>$2BKQe1DtfgbEnm
zrJ;VV`IiF_Jw|;I3szTQw<8!Im_;UaB7(x{dRdjZguh^0QR_^3*;@1{0e1fCIi(23
zFs5>DyzZuxX({9~q2d!6LNTwMpft_O>*yrK1te`}u1$xV#B+)}WXKr!g$slO9jmXG
zDP7heuRu3*1IK%S+X3n+P$_}I;8?-7k_}b87-f;>3fMyCvG`34ogh#;w8N8i|3U-R
zbs!DTeps%hpXBtv>@g#&N=hbvH;DaEa-CTdfPD|#cgHqiO9C>a3X1u5Fi@Q9DO|dv
zk>1xgRv)6Pbd@4Zr?zs=TmSvft5KOdmPU9DH;c*Mk*lvZtaBM}RyK_7+7dHt)$&<3
zEepa}g})_`>P6m)bZKjTDEg$RO7qpV-ri*!t>huXz+LkQwh4>-<kk6YRUo}k>g^~#
z!C0z`)04`XIt`wlPtjt@11iu4!Wm;RJB7FV1xPOS4Mc-WYe@++Y0Ww3>{o&?UEB&i
zg%*`qYm~RcdSD7|poWwX!hs$w4IPtd?Vcw{p(Q3Vv92YoToPSc^UjJnMcF^%(QY-S
zdlg$<uJl7`=LQpRNmWdT){GWJM}TAsx^%PQpd}ZTT(ywRo*BVKeAEiVKNsfxC-~|Y
z;e)6A;K+HgBOSW$GP`z0>oZLlkRFl<QsFGZU!vV%#cx7mXA2p}eKw={eLus~kHF0>
zejM7U$1TN0cwe$EK{grm(+M0KOY%qCFxtE<G`n2LK6U_E+;4;yd+eYBf`BYny1|6`
zwHk_d@-O0WCK&~ryZr!U&Hxz_Ol3UR)$dXoQa~oIWTtp}ztk0CB}Vpf{&6D_>VG}-
z+NZTe(T{COGO<B0qcg}^!fg#{w{eQd2&3+hi)L!T%#g)w@7J@#+(8hIb~Bv3>hEO&
z7klX3&@CrpQ8j~g;585_vd0#;Z&wGMwRrt^LE>0+)n{$71byEzUHGD3?eH3Wr(8@~
z&Y%U*pc2uo%Vb4?&s?wkb4Em01<c(nbQ(~aZSB>N^`2E>5u*Hokg(DnRdWQr&_nyr
z%#ZgKROgHkp)3paO}5`%O0~tG5X138KkqBDE5Wvy+R+wfCH;^3Z#Oz9y7`XIEX>2H
zE>_lPzc$EP@$pRZovfz*VeBsRL{(DP{};sk116FQ5#U=;d(mysru6N!X^^^|Er(DG
z*G!M%l4(C9c2J}X2|&ieCy(r3U+rK^3j8uZ+dQIU)v8n)ph~CHwKB=^r#hO>%+B5K
z2K^#tUqv?ZQMfy`^PPnO^I^v!*{Bgo=QZ8Ivw9hzT@e=39R~)xaL}?b1EmK+ux%Il
zsoBxRmqCIN&_%XsfmQ>e6{UXtfimJZqqQ|lSTas$Da@T!$we-6W*mBXyHPw>E2O{1
z@ly*nRC!!VL#=gc30~r$35fr5lB3Dz%6F?MT~iA6B1@ob#>c%;9dn2XRB=?*7<=wt
z%7dgT&<|zpfVExwE4N6>N?i0tC8(&Ca}rJYMU>f8QNPklRMcJe`|!`<t~Xni*@QeB
zY6p%r!_ghEr|e762DKjMe{Qf!v4UtKW$qjy2w<Pv(abhwoRgg5(1yO8ui<P;ah`;i
zdvyZZ_IZ)1Izc6&qvmpIOFkjdX(`^%6;A1oSZ1`p=>BllYPK6UNL2Xu^|sgw$AKDC
zIWf%3fN31)cs$ycj&7%O9L8D^+TL&G7hfDnjCym-`Z^@sh*r{~1bK1djZ5`kU&dWi
zDvE#hUZzorA+ba9ib?bI#nwqKX_A1cQYSQ~V@t~N6uLCq#+yJ1yETGZZKpommt&xd
zRdnMi<sjevX6BD?98@MNynZBR5cj#@*v*#d%2nr{5sbcYT<P4lb%*RvDp$4y?Anuc
zA$tb52!8l|uQNU2>jj*&#@+pXk6dUTVv<nru=1_TcI+ZaSIq0PWXBeV#8-CcR!iAA
z*im!VAN9TVWx3pxjqr^#-+q)&S6;3{ZZm9kFb%s(<yZCHD&aL&KSAfJ4Em$B68qnm
z$YvRfwDMts_v2m|2tMt8@$iK)x#0Ts&p}`8DWX|XEG}Xc`Y}3}LpBAdSDp()P`Qz!
z%yz1CIn?GWd>2{oA|R{AJ_~HBA&D=|_qR3GyziH+No$eqpl)i(gt~rbueVB@{b}e2
z!v&41=vEJCQ18&2(`VBBOo{snT0M|3N$zW@(Dn}~elsC2+}rmJ=bJ_V70@czEsy|v
z`mGEX*6M5vY3yQ*6v#AME8>P3)X2wNff2mZMnOW`Y=8%>mc02UJN-gUUeRR2hP1}m
zJTsLtgdAR|xA)%#npT+q3Dliz5#sFg+vi;>Gn=Xo*cCViBJu~K!KMrRG|)pe&zlRQ
zILehT)XswJbfzi|Sv|JQ2#%*kb>5b}+kIw!c%Zk-rfbLkB|!Qm_6g^{`;2leBQ|=>
z>azyn%iFhuijCf_<BiLCC9t|%F+O@~QzL$F$D;e1ZVVj{vj%LQG)n#(_FDQEQd`pI
z!m^0411$SuVN1dP$o53%#hv-)KULIPRU*Fc-it0QX5Z+4_(o!VWIHA2`QJ*j2VY57
zwXAJI@99|;#@i%{2U9#VCXQ~z*eefZO-{<b#TtjQaTgo(Axbc2M@pCKT4#mHwx$;f
z@LHTj+cV?B#;8ht34KP4B0ef9^9R=H4)3c-jX#XSAAC{{HwX7k=TB3LpIyV$W_<JB
zf)dKueF{%<<)&k<55<yQQw!cD@ZZtZb{znxRgv}V_q&nIAwAJzXG(loODb1B5kmUc
z1LJzFZbc$$Fh;M*BiTNl70>SVHERV`bNnWmFxu-Btv9ElZ?)Pv<?FnwsgNq(N2-BE
zMHt_eCFHyO;x+g_-z=50C=fY7wP)UxTg^$wR~w)}{nrV65oUFIG)IcH42Zu<nu_s!
zacN|pK;y_n<5y*|A119*!X9V}6cBnZX=fYaQz=^4OO#@nhENasQ^8K^RfRLl0XTB%
z9Kbzb;X=l@tx6i4-N8YW)9Y=Woz`-uJX5F|1JMdUQV^%GXyOH%Xj1<cYJao7xUq{Z
z@DE*p399H4JsNndw~;hr=XwbsuQRiGXR}exq6w`A^d}haP!#$gLRKXl3{%-z<cURq
zgMH*q7>!JLuYp;XUOcpeJvq_eJh=4X*S=bcLZL><uDJHAN%j7HNfFP6dS8hpD(OUe
zTw;TTw;B%#FHI!LKVG?9+ng<2deye>h-Wl5Sf^$0G+6@+|0O#}#DA6vlHfGK)ZXW3
zb7J}lczrioLYd*YuSXADb{}<yNV0RMSeWV+=z4@GL^C5Rp{HMdhO$pS>C(q=1&l^X
zAM09OW2BoNJd1n#g|WY2AX!S;2w~2L$4Y3#dR?u1R{PX2W`g}yAz59p-B?lggd%k)
zxMEg#=}USrQ~EP0=&k&ZnfnRVY|E{4c@ZVIkRCd2k<WcB5G(v`Ay+~IBp0N+3wq<$
z<`R_iZ$|^#!ON(WORa{wcLNSA1=QIhIgLN^Z1J2FZy_(Yt#|IdKGFM<?^S^E2mUar
zAI&a4Rz8Ty#xb8b8Wtq=t8)D*sT2m4Z5dii40Dq||K3Y+5#o|RS+aowz^2_y$uT*t
zIgK`=ktx3m_0Q6oXKXx9xdP8arhlTRW#P|`)P|&NZ!`v#21PN3a)(&0lB_mIF|~OF
za8E}HlXM->J*$MuojxCf^#hYc9*(i70+R4go$K6V)OnhT$rlAil1RzT#zRMtDlC-+
zCUxy$nGUuBfU2W#znP!WEt(9S_5~_wHU;bQG~B?ashk<9C3!X&e_(l+ub44AuKJnH
zaqt>`s6U5g$2EG$oD)=}={Wrk5+VUGq40S_?o;vg==X7G=gf2w_bwmo5L_?0T$@-S
zAtBXd`&OjsH=tQ7C!t7iyC{S0th}4)z<&bb76PTE^fH8TqIvp&8azUe=v;c&tp{Yk
zm^yG6WsSt^JkRHxS=+W4?Kai&c=ZuRkk!r4?mT-(($91)!!?iJLBhLVeTDK;zlgJ+
z*~824AceYaE7+>e(~zc0N%8jC>(7{6O7amuk2$m6@0Bn6@v=1y%9_rM@psq#rPT{v
z>3ivJT;P?CeO7jg^bFb5dSG6;J;zIq9ZNN>8hkn%4{dvv?rE@oId|u=IY)bf8GY^>
z-IlI_YgH3R()}#>nb<V(x)Rsc{7+QX)m^rF@7(Usf8nb5?}E<VoQe8TL2;u-zDt|g
zYWt_TvJ}nKiTJ)saXs7o`$?f`q;KLU4y;0wC3j}BND7tk^5T8fh9_%?R)3Ocy{B6M
zg8P@Cex#^}t5xe=+{6T{s8Etn-%xnXods|XynOjG-?KFPE^hHo^LK=VrjzWneq;Rz
zMQx5-_XWN=_DJ<In?Q1?ygnnoyXp%FVenqVoS$x+-+e<jm0A*V7T>mECuxF5>a;ri
zPW**LP@EeRRHSEq6qi*ymxA1r#M$1A!~l?{qI6d&H{y%Vch^dI;%Sk_rcQEPT#jB*
z3J2iMH`E~xJZzYb)Z(IHohli<I;?X7B)7)$|B-7meME?*CEaw;0k*&r!NnNe{o5aI
z7>q-w;UmbIys@i)h`rm#O5Z6NoAAhEL^}f_XHIUhuHW~wbOtsXfR|HW&f-3X*niGy
zbt_@`#jUsNK7m*xI%xz3XsJ@$xJrRx+Ep7oMt$0s<^Oj?b@7zx%!TtepBBsiTTz&`
za@K>w*It1F+<Eeq9sjEm(7(mSKhzNOlzEvXh`slUoMN6zqxr9UD{+*{ZAr2#sr!W$
z6`zVJXNBtAxp|zvTnpiOaH3mWHlBWApH6Dn5bL(O`GxZsQPxYn@Uy(UO<HoKL7DSQ
zg`Zl=%E_F#zv00hy+`6AIt;OH_HPUn?SeeS-%8^$-+!7(`1Mb_!e8p9i3yzDwb?42
z>V6xuuZ>*Wxs}}7$wL*KIg*{urSgd_!StI>uFLjV)&8;ZJq_2NMdtrv)JI=1Y4Hdx
zNAWKx&iS^FUiB4ii>A>skZ$YuWJWJ?T8FWs)-70-ZhTx1^lcC`yB($_<V_q()0e&i
zx&22@pj9tj{fJGfpjoLSP)H(<t$i$BL-<>&TI>^^zb(U@+2W^75tMPiGAC-!f!|Td
zV)Ua3lNIL@BT3w8dV*lJ5kF7M^((Qg&MTMnKP&hw&Q!RDVD??RXRUc*PDL=8?~G!$
zM$snG7Noj;*4>2d=oCMR{#&goZ+lci`L2k=*Jb>RHcZD1c`&{sF~+>2i?LDA@=K&b
z5<jo7@Bmx8%!5F87ttD(3D)~&)<&2|^4`^zT-o!tF^W~!CK2dp_R>4{pU8@v{9A3u
zYTr{pK_1Mp$$hYWfJ~j(LT|o_n<S6>jQ;!QY}#~QqNR-QP_S+L+E&KO8JYK=J4mY9
z;!@>XJ;*3&TcUz}6Hq@1lb~xAoCaGgo1oaJbB9G|L+GssC-wZMQ6S#prSYzm0X#ok
zQ8H&A63uEt)Aj_P{L+!uaPDS*xVCFAZY8de{thQGNo;@WFOW3O&Gn8SQ-Ep#9JHWY
zplB2g!kC7Kel1?}Yw0}@JT)uCaIJImeLD1IVd6KZ@=GfG;mrII5J2h;VU;(#NRp^w
znAh=@ApYk4ikp7CF0)o00EZPZT{}h%sE;hSMr@m%7V}V~{9brlk9^gJ80V^aKjd+@
zbEY!Nn-`r|VLMh_tT(Nh$UG8ge*9|g(O5Cf?iDub?4jOd>io7Ft~mXrHer8c+Wklr
z>TAhO*Y!1T`FZ>coCRyd%e#&WGzD@&>sIBRGJ8Dq{XD5}!(&+DBA$~<`a6dkBQNd-
zci!bui>cE!ZrKZC%3Xe4`Uq{MHJ<;;J}Pa=PopVNYgOpSXv6A<&+n{jeO7S;AA4+~
zn65kxikJ=U{I0?-zlX(N>3mKWc|9A&U@Z{o)$)CP^a!fV&tLllJr&HP;W04t?n6OX
z57&kpQ=^IO09#S1kc>8)ajQ&=;{&g$6a6InJb?;X;$tm&iwzd2-n$&7;LXRx=o_el
zJIs4!Pd1e>c_B8gw&)GdMD*M#c9dv(k=&bnCyHY%H?m<`cp(S&;RrJ0J#Ur|?@r@7
z<rt**SuK<7d#n<qO^8k#1rv0%Vp%`OtnLsg!PXzeYfN$d(L3A>9*o*g8;sH;@k~RB
z?V0;CANM`Rb1{(&)pQR04AhY{?nrUxG^0GZy*RKe^eVk`GxaLC%Ym&;6Y%Ijr!T1o
z`Q{xC0QN^BeNvZTBdwYN@5MQ%p(t$6IS7Qs6t>M#LeTV!Pj`(7hm>&W0Tzj2rO|<w
z*Yn=;kCpNxC{6xABy!}>qedQ_0KYOmAOTl<l_E3k9m=n*vlK<W$QVLWEptd5{;*ez
z_E8{?JDLjKQ~7*HoBq))U(V`(m-kg!Jrht-iQyb+!@G%Mh1DTXT%O1u$k^@<uX{&b
zvDXp~kt@*hTWkMopZ0e{(UqpxD=Tw(^T{ogiqT(wwWwHjp`CU;Y^<UB@T_JX^~B*3
z`{E^HfbQz`0D=3%J!~$s*aAB7DO3;LxTY(XuBP5ctzd(*foEXTu2rXvWx+U91imLp
z4uf1{O|4x>lf?HLygl?wS5_m};e*zCw%4XULdR}vCVw3ZjypkpC-v|qf9sXF*m!N}
zm~jpHF8Ju$<<f!KRzGqjWMYgeXZ>BTg!8cz!&-&tl3ayj$$<bjH0$j7H#xkbvxhPb
zMQ|wTL>P$MPFTS~T;=}Zl3G(spQrHo!UO5P?hMzCMt!BQ4@RdofwqMDJrde@AVL;k
zvj$*f`5mTkCO%Oo9in_>i8X{MwU)i=CXv4AsDTGdKqEyz+*TpkQHa9dA454r<a?HC
zP)E~~B3Px&BKUiUSvSE)XB%Ac;q7@w$NZ#$6qlffa|ul9(qS>!oG4LY$8U}a+5;dd
zzrrX&+Lul+mAxO{2zA-Sb0?wf{>)3Pa|LKFPCthF5a&n&!ocU=OK;xTq`|2L*#{FQ
z9+KktXvMxX_E=p~XuRHCxEWFENGIGj_+TnyVQVH^z<?g{U4e$9<BqTub0cS^xP*^v
zYFUS;+hB668+Xlph@MsOKwsJI_s0?1J<pnvv!iy-W};~UrC#tSUZn*|t@Ha&n&}zC
zAA%)UJ!!E?+<TNN%Q&=~o?EN7{rTI+6>xXjhgfx0&RQ$C0iHlsdYs{%o3FOgO?{Q3
z#l;?P%>@cod$jFqx6=6jLT@XD6wmx&tWsR|dG;|FQ*)Vk%+=ggMN^sf!eCc{v$8Oj
z>M_rlTKDx-`*;luONg=nb8s)7uWqVp(-T!AvR1E{r;>;3*I5cNY3jFWN2U99XDfFn
zS?^eC=e?7j`OEu<%@prcX6w5ACGO1)qJ*0Kks@Z^mZTIdcQ)Mol?PEw)84@;7Vq1G
za!vY9<6)ZmB1wCk!msp5^1<X&X`0Y{{ZF_NN!yX+Z^M+16N$AynG?8ZYv~Gi3;K@l
zA}>%wZN)G6Gx~-79g%{mtyrVq@V$rT^0##L_Cx#Gda^e-Vb!DW-Uwy0FO$6b*cNo{
zR%bqtr0c6LcY+*4-(&y8W!?zEvD4hgr#VT{d1gUM+k8j(9KMEc9P4F<s}v5Li$+E?
zvNg01bDf#~lyyBCNVwz=xI@$*e8kPi;T`TezDRBMl5W?ZV8~9cJT2nCqA|7*XNbsH
zEGucyJ>lYcK^v)CC|_iABoHXHE_?9M>Eef{3Khf%FYH5w5stO=n;%Sl^iqPHFw&!H
zTok~z1o8n{H9gqp+a2P>(%QVpq$756^00tn{-GQ$ftZfvrSF86iRg`s#yMt5nXU1<
zl_-%zMS_{jr%OY`+N0f;4*5GEy2{eAYLF6D+Of^##xO`p8f&B^|9L(_dNa-rY&J{f
z&700quU_EF3oH|5{w29X*EVYhEhb$5yP%Q07*w;E!goznSV<5%u-2WbI9pl5B2-87
zY3XcmF|O^u3tKOGek0w8Pj#8&$#O22YAXj{<0m3HdjfJ(&(e>$6bq-_bS1wY1DCO`
zfmX1!tC@m8nMvIx<4n`sk2jD_aQRM9FdtLl`O)(0Y%NiYA)49kb|G~!UbNb<H6Z+>
z#9gvu!5VsuVQP%0H|0XZt;a@x+$nS{U5Fm8<4R#BONg(E&|~hb<Ez_OX)zY@_4zuV
z{WWVnd~b+E^z}<4wBPfM^lqJ^3J3I^;90Cqy5HL1&f;=tLnSLsfpnpEmR5(lgX^C<
zif*3xnR6eGF%X6ZjDu$pzT)-IeFWB`x+^iqkj5JCwuU~RZl>&u$!qc6StmEqbUKM_
zw%6Y{GClNv+)H+j-|m_v^qSh1Q5-oFqvdn6Al@OPup1zHH1nlTa_ab(e0@qX<F^Y{
z^P)8(ch?^vX-=3q7niggduhlCLSOhH^q<f|`65i84wt`foH6)XALl`r|K7}b*MU*h
z9=g5F`_F$Dv{e&)rv+4TX>lQNH<Rf$f=R1U>!g!t_naU(jCK908eXlbzm%k*V~9ZD
zCeai@em>m@NQo0GfiIwnV|MjKclzlO6yx=!XeezJuMDQSPK-skxLtW3b=U!~Y{xDf
zS7h2Cuju@51t0FJYffoq#ZFcT+%x0vxrJ6xErkqUn~o(Os}JP}>2Q9c(Ruhlff)5X
zqX>e8w^#KOlOv+HC>%razKaLZV_Qu#re(aWp<DqIi{aRJg(cdH#X?~M+pJ)&2OS4z
zeRXyg_`7iSS*Ik4#$ymqh!Kzk6<=<!2b7QnOg>uhb%d31_FWisZX1{N)xEuhKCu+w
zHp$cKz78?`A{7`Y;Wc{QzH1Rvx0c`?^Ird|J%2Kdo0!HOC1w}<vr3xHC4ZuY-x0zS
zra`<bM^wub>mEK$8E1#wie*%J0E<YGR4|g1Z9@|D{!Gk1ytapQ^>L_d(0w@TDdh(h
zmkW(ix%>6#`p@8sM}gPg=*zFH3>~<B({HV#$Oejx9P<ot`^sC<T%VuzX4xNqxXzc7
zri)EG%bj_I>T5Z@k`-%9a<6H#;t{YeKsI@~i=P%G3~_8t6O!lz^bg#E_V$7t;3ueM
z?j(gh`2cP$J@2|QH0~RkgJ=@z!FzAP*{$@h0_$Y+^l1e5-CrEXJkwM^j+~c7%j&N3
zKaD$<`B1YSacv_41XQ&@`pK97^dw-vQ8Xt-J~~Hm8u*cy_WRxe+2FEWk5%R4R{c}@
z)f&kl*~SK?d3w7t4r-D8^aFl21I?4GR_59@Jo<`O`Yx+R*VemqssfjgLNYH5rLtT;
ze80>Wi}p$~c2FXVfcm_|G#O5?2|+JE+pi;LjS_bKXN^IB0NeFAD0&oqKwe=5x)ZP|
z)gei$aq-KKdeC=$xLG^kmvkf>!Bn}!%kfL|8?sDliX0l>>e?kQ_Cd=$wNtR`58TDQ
zD_sscIIo?U1_3@%cXtmxZIPe~F<l%ndxsa{y8IV1;9)Of?;(ArEKQ}x*LeK07|S&3
z`CC9EP^~wu<&@!NZ?i;TIs%SwZR-y*b#ofqZlp-x>fvjY!wV|$l>+*c4{5#{0G*EJ
zAlhoY`JjBxPYs;mj^MbDyb{g4Zbl8MGkV)Yt2%rTfvo^p+_|j(&mxx6{)9=$B93AE
zR`|>3;Tu8zODNEyWHi2WdS)!PbaZvKKld(PM^Sg4<5^oI^-#4;5&P8vmkxOSTvXzJ
z7g{f=#rx6ELH-$ikmSOe5>tz;S1V{DB>IXWecTSmeZK#AYw|_6VIC%|n+*m7ZKre-
zExg&K?aqr>DYF9)`6{vyMv^@%cJVBks8F_G`N^$O6|q0tURf+`CW|o|N@N*c%c}$x
zroP>qn1^q4hOccF#7A(!oE$&1=&$$HGH+>=usI(jh35N`WLBcBqDK~Y^P(rBN+V+L
zW>xJA`?^TqS#F9Ib^FvA#XyF!ki%8Xp17Z2Li%>OcjRoYeqIo@$YC>1MAd$;lnD~~
zvUjQRi$PRy=91#z$FAbfGGAWyOF@rcgx8M*_YbUytm3K#<CXY7@J#p<B7QtiS$HK<
zfVao>wtk+*Zv>pEM*43c*?a3_ImT#x{F~*z;?(b4TS<Zt5qIme?YS_5{s$7%{kKvK
z@AIoV>FDbVD?^JOJsFHs;2zc95OcNn^ABp%4&3G-;3*vaFRXX^N<Jairk_fjY2%Y$
zkpz@wW`jmRxx^B@+r*BgW4FDUkNCNYaK}3P^2F))=|lX3WVbk26u;8i^njBse=PTD
zu1p1?;wDJhp@~rtiAmoBNU3w_VWyqMxBG$pRRP74r&$`CxIMv9GW=2(J3i^L*qq!J
zQ&)kXdC0V!>a9KO%Tc|Gxf~-h0{7-TGxD5vl<w-Wv=2B|I_qxA;A3eV3FrTu6PCES
zNO7SUQ}a$~t-G75!yyqY`oP6;gH!wiIC;u&XM!DfWEQZ9s6r%&)={)YRE|kI#esB!
z{qf`C4kb{=ig|VBrX7y<<K}xZQM^>s2&dh%Th=jFGsEt3`@Y7Ci>2nK4fcLOGmYzh
zEy=$>xatrHIZ#$sgr)^ZCMlkzyCx5`oLpnUKe>}2bIs$w3zz;GlwJ!MyF(0+WS}+-
zS~e~b3Ul#`fTHJQ^r7!Q@Rc`g2VlJK$IcO?wLd0T@_s4uJCuHpY{z4)CP1TH@$m-x
zYhO#wKbUtz`h$JPN7sHM%8E<$)h7=W?~tqz>w?~PBK)NR+}qpom`ANr-}soLFX@Aj
z<NVADEB7Fl{jpz=MaGg>znVysrJrNf3A#9j$??0O%H8fdl)QOZK#2ldI}Ufvn#0X?
zi+aUpJnHK6;JUS^t@ca-$rGc;&#_4h>Ui18nyX0-$b{s!Q$(O?_+qm9g?OQ#*Mgrl
z8jhb$#q5Atq{efNXD2peKT^{e-f&D@T}l+Z097I0{C5tCGm2Zd|F_hteJ!N}JjwJH
zzgwrubdvKsrsV1~@qU;!Ke?ea&i7kWCINRo`?|A`5T$&7uV_$Fkiw@--?z*RE5xNT
z+jQrEnq!~0^hFWEq`0-h`AAi8s^o^;OfNmT!Y<R=)9||iz?|PZh)#GRGO>HlDegze
z9oTryt#;!+ow_qv&kF~?-qE?!JRgh-3v{=ytT{|(ak^>bzYCtRhbkG4$l0~%#I^Uy
zhLmli99{CUdh@GM_Z^ZzxSHWQ<F5&c5DRR(@n_XR_^sbPP;h4jUw@MVLT%E}Rk8gD
z?E5&B1HBaEc{DcjK(L1`oTJ|v=I$O<dU!L8ZtahU!}w^a+)vHT(JpA&kWdaa0pxq3
zus`KJW;~f8jUBEkpa>`<fEQFlA7~Oa*PxmebDZQjh>L5)p|C%r=6rB(^u4i>!irB1
zXZFnJ+!<^d?~QL4z~7WxNSWz%C^pK;CGIdeTtTzW3?pvc!Uku*Il8_%!BolsX=t%1
zY=|wy`jm}7eAWJ4_s6)1(zA<`CM8@J&Qi}i{3w<meBa(|@ZT_z-KJd*gCq(M`ZSH*
zS+l%~<kOXyx{6Qqyu|%m_f3DyWJLG%P_E!x{9}@<f7r4~H|=9}HQC%T8#0gX2%XR`
zGOew&w(QDezP&9Yh?kL^Op3}(sxx$X<oy#@nct#)Ra#)V0iIR0!gyU`f{jxQokyFH
z*x|BZxHogC^zLyw-=e%z@mN!^OsZeQ=E#6g_iXj%S7rnBLtm1s=$s8rZc6PD${${`
zFGtETM|S8G?GyFQW4F2TcX!iC`;OMD)F9b1m|lSOo7Y>p=hj9YbU=5=)5<Uw=P;jT
zL$KzTKW5x?6I&nTYS}7%oTQv5(5-PFSF_F$FEZUQqPOJQ(lD8+`uET_f^W9HLa;gP
z`Jw{%p<rbYP3vQ)@tz#&ommDn5}x}s#s&(8ydWkx7m0+v?B*at8vq~|+ka9783;9t
zZ~T80QwqgV1RaFx)Jl#sPdb(^WD~@`1Ht<w@SXU7(;1-aYS_tZaq__Ce;SkDWGi-Y
zcj%Prb_;a6hp-=>#hszQW=@7BT2HZ0==7S3WKs~Rar1Eo>(R}}sT>kOSqx65O0#CB
zTDic=7mIWDY*52!UIdT1o1Hgf<d3!+37SAY(S&Ca1XAWuR$coRK)U46<j`>hp=~oW
zBl09UoFy!23%i)M(gHNKtkjXh5)wQfMya5+euW7HEHiIM<RGuk8nLP~zRs8$5|%EG
z$6?3Uz8e?aoy)n*=pZf7ru~`u;yyH0N_S|jolvf1>iO(;1jRUufxbX8es5L5;(?Kp
zFjY;xKv<KpmvXXH8{=m)YW8QMs0r$-MnPjsicnZ^ilC3wu^XG-M0YY-?OV`77-yNj
zHO`e}u(y~!)f}uTvZCRhl-v_`9c0F|eL5H0uFt&-{dp1ndeW`RPDjdbN;8G`e$n$m
z@sbG6n}51GG~C5LbxGAn_Far{Q=JMN<rf#74>p%W(4<W?--nIgP--KDYqD(lnl+?0
z2yM!>Ta0WN3v@-pN%!y0BIA15Nl*p-)}8$)@o@{sm?WzOhun*)MWac>Z{N6E3LC5{
zmt${!3h&v!#q{(cQE}7HC~3oNjA9Lz&7js>mZ1riEE@PG1kXB#;`Y?hYS>fWD|e-d
zx~&6+0)aE#i$q-KwHPR%rt^S7*6>`8Aw38*$lV7p9+u=LZukN^!^|-YA60a8Gm&%6
z7gXqY<H@F8zXiwBZy)E)<Q_FFY_%e1c~CY%r#W9V+@D%ZQ#b=%x*#$Uf^R6Obq!q=
zt@jB@ezP#ghUTu*M29lqS|?p8UJI6!;&Ba`J`&tJxwd+E?aRGxHkHiUL4uD(Ljulz
z5e|5$5)$g<a@J}X$odnD^ia*wDP+(u(QySp4FLkk@}d98_le?gxML_rMX*so5&(vr
zIu^X?_P}d(9Mqm}&Je;`Vz9?2oJBv{-gVDk`;Mz6yHcbuHorCkKakRI1@qtPPI?%r
zJP&!bFlGMZ`3u&(Cku1;yI&AKs8krrryb4`|L3k``S5!;(}x#p9@?v$-vycy*{Aq~
z+BrAe3z^~dY3fQ)Zj0b55uBy#oWYAnCF_K@p$FbgqVE}p6z;9|qt@)bpQzJ)P!x{G
z%m+cQ!2PIwP1Hla#_XMODD$6PK{8Q%62=m|oi4Y&K$!MVb@VCsI9=h1q^)jUKA!dO
zy__^QN%!ge(`4ue2kxgY4E(m#2AJQkdY-2IPam(LLjod?JYwp;G{COKmET0AjV%^7
zJJG%(OP9UtHEpW|#&=MQzHh8Pn>HCXA3g_mMDR0aEs9vQKo80x)j`8;N1Nc*leaD1
z6rJ?vNcz7ElI)Alp0Otd;MgzN)u`HsP(ix?gbrizryT3#us*l1dR^17Z|OMU9qji*
zZWD`R;c;!}vm4&nc`@QJcucI;&9r057D1<p1t{u64WpvNV{wp(OivLbDi7=UxvbJW
zb(OL`WhFXN0~O~b#z?PK5T{^R>!qKh$kohk8%&}zZHE4T9b{$(R^B=5U@||K**O6D
zc2P_El8Uyg{2L!N{SC!STVhjOky%W}tzDGn{PzR%h6>DOZebLe*M9W0`KkQmxH(ZZ
zPjIw;){o{_ac8~v=-o0a%f6YPP$dHCV;WG?21R2Ohv)6QP**G08*eBc@X7~Q+9T=G
z5dreJn%Nw0d*Yy#r+(?}kA^Ya;@BmZDbFl62L_q3QV5j}+g%Z?k-TlVCdx-4U#a!)
zrLVZJL0R?-G?`Ja1Jy=LpRS!TyF{)Ocz57PXV<5k(jA*0txu1gUET|CD#q<U$Q-G)
z=|YGY$bWG}35H{BNiwZH3qy5x$l<%gKC$MSomr3Y@4TOw@tDT_Z^Q}Z$b2bKFnuIT
z6<P01cTP7pTqE1b0#^YjnZ~HV_ML@7G*#ObNT)>?KH-j$u|v}kIeaql7r`;6Pc}52
zb;C6mADEK2Z&JFghHM(|3x_`FH3}TfYDE07TuggjLNrCBhDs#xL~{_U#BQL;|7$`1
z5$rk54f*c^!djY;u-Z>|aZFsI`IP<0xwN2;xO?X5R+xtKTViX!{)=@eu7i)}^y-1@
zPevOLXI-tOij!+!Dh@>KUmq*KHVr{I;hEkhOdO!E)B9}2`|<NVCz>F4=?cmwr$fQ$
zKZiCk0L7OlbP!q@-UGCofEM?5tO%ziNPdufUyBuFa6ldt{7H|W53V+f6V8`~s1$b2
zPXN+nN~KP|8Ahw7x6S*C3}esZcH>6ybLH9|P#Vgt$w>Z4)S$j=-2#ZA*393HVC&i#
z>ShvN63|AQ*q}o+w>J8}*sJFep<}xDcHx^3C8mXc2`FLy_w}KfzP{*-=48XmJISJ3
zJ_$ID%*s`_;V1E@Om76Xu=g9WK?(y*{dJcN59B3TDxBqmV?CZ<{Hnw@NNmhO|Le6x
z7_m0^gew;ps@Ix-N0OQGXqZ~M2OsM+eu5T{<Zqs7+FejMrL_{a5Hv%{7t4sp3E{Sq
zU8a3Y4cccthORp{T((^sA6I&yUy`6J%(qM9&E{<>GoG4mwTGnH*KT3<<)0o;B}WiI
zZu|4DLLslsVeN*^ZBhKJXd=_StUYua)s%+o8!wQ+oJM%eGF+GQGBO+;X7rTws$#&1
zb<aV6%zbVy$34Gz-=%U%ff{CSnE&l9e>6L&A-#%`*Zd|EXsT^B?(C*Du0A}+cRy(5
z2+*=(j%iw5+sx#=`rzp3S@+^iXF#t3G0uMin;?fiY5!fotM|E^u-?ef4~5TsY(9c7
zn+~{x@si~AHZLM%j19ydK>a!`DMG}?Q~GQ;9FDSh;=648ls3HSLvId|;LW~~y!&%z
zb{FJHx(P)b$84c2p!5i&zRzs*;!NFOxZ^w#dpQYkpMxr*eb;u+v~47FeTG7wlA&KI
zh4=2kN%m*3wlkPg!xNdD5NRm!U&D3QJ1`R8V^!zb8*(3uD%9x-Z+lS3o~?0l0<<>=
zOWtILy-z;kCWH3Xm*Kysc@!aS?ysb2;5)04-ikvY|9>$p0LBFzBS~OKq&HT|dD{|>
zC`vI)<u8}HH(zs2In`oyxTr{%*ZTP$|7VlttR)v)C-Ebh+Y;3@&8BYANfQm5iV1@~
zBfdSqPjfr`ux|hJ6Su%tx#d_N-ak$8q^?>9zd^~<4N=uPqAQa{Eq%uc>_L)s>&Sil
z0-vV_QN`yQvV^ND{uo*?=6Rr)U<E<xFuw2*rosqFS^tjHq+@9-9gC#BGSwXl`y61?
zuMgp6NPKM|nV`1dzxW=5sWtSxTsm&vB)&A6lEqB@{;I*$WU)QE<N?&DVf<P~k}2Bz
zv2@IA!Br$f+3Nab6Z!T9QZM7uiAl5#xEc*rmVfmwpGvK?Du(%TgJF>Nl5gb%dOWg<
za-*ePe3KZIJtlSI#dW$DiT;f~h~H;goru1Y@2B<!=^EQSjd$5+dKorawn>UVBz#Zk
z_cUOK&dnCV9Af0BcW=xe1o9E(e;tVk6Z>LJk%WY)BW<S}$L=8(Bd6E>*BLs_lK#6u
zN#6fNvVVG7SUg#)1uev^7hOIWUI^9FZ-l`};q2gPAyF#Z1@G&Jjah@sf6&-L)K8=H
z5Iiiz#<OEFl$iybfAv;x71>^Xb^|5kgE;~MaIim?;wP*174msF8`a{D3>U|o-S;&-
z=LOf(Aw<))eb0=<F`LJss2~IUHHsTSmG%$&0erEr)eX;~Loya~rU}>RHv~gcsXRdK
zHU=bjK|WCsU;skI%(oTp6vVU+!ks0~v1phGzuaCD&S#$dq{~>XZzbp7Q5cL1jnatG
zbUmgtR5dHTlY4-v_<G9xB&&X%eD)kql;^kcS{~b#JJa<Q&YY&O=;*zP+6)&jk=Qi<
zzC#X_5p;0}v5w*ruC?#S^gEH43ddHSD$X>oV<~scThIO7s>l)_7_$@yF>pm=>4t**
zG%M3<za<70euhEn_<IX>9^Z*}mv4uZfnu5Cz<!`F#X|o0Y^1V*-C+M#OQ$V^aDrNj
zZg8niA6n3LvBpe>0kVDu6`UpdwB+?|HavP?w{JRlHdydpgm(}jsS{7W)$Z@<x2gKq
zkVY{ReWVzVD!6h*-e9eI8{u0vOK<xgr)gbeI5Id(l~gRy*KjI`&X@5$<h@IpSsyy)
z`(N$q?0*+n;@_Vp`5()5=zwtgKN2AxQb<=_@X#jFw4UmGJ&OA_9>qJV#rA(3oqIgf
z{r|@MDsqfDpT(TTB2f&t!<<<p3R%pt916Fbca!rl#}wf<$C8{+i8-Ic=&(8D&}3?a
z)*NQLf1iH)+v9=9KHK~CeqGn~yu?qG+UW{idu>uYxum>UfwjC?nZpFuT^Du_Hb7P{
zoy2$*3gWm=|4a|5#KRiSjEszwhDFnDX)h?Q0>K&{$%XgV8Htu@iHrtOquJ?}r7ap}
zlvzcb&vrZ-AD>44{W(~9_0m}GOvYQ-UeZZQjF=dAnz{D)tfx|yVwy@gj$^OPs^Qaf
zE$<{n;9%m|t3DOD4b%-dX_KjH*j_P8W(<gU&YW9F1m=UvN+$uEQ1Q?aq`Y?N>7DB_
z>>ndu#olBsP3|}zlzOw3m*4e}CZz-HRE)iAN<*a59~hAACk5()i9RJ_Jt&i9vvmlu
z334UW{NDELiE@_W&By#6b@(9#g=Zu07fy7OA7v{Vyj^<Kl=<8X`S)zRt%UNkXQSay
zDK>OuLMh?vgl|*HqZq-C(Mm?T$IYzq)29Od?sO$}l97qKqR6AKp{_S}{-;D9OY~PF
zj(5W^&(o!^2IeF2HvfD}Pm}(H0g@~gGJ`1JP4Y9*An~yoHd_V3Jf0t8?|Hx4?8Qi&
zenRgbgdLD#$Zc<%PlaD^ZO%>|<H#*Pf^29$;z~)rS!sZ%Yk6zyl(!D^D70lH<yj-`
z3E`21cTr%164`&QZb~)esRD<~y^YJx*D9lYU~3Eag*Qt2^BBSEeFA$v-+43~`Q57e
zQi&Pek~|X4|9X<un>t*`{K1e0gQIdkB%(PCOo=61m`DIbGMg{B)=37sn>HC_6*_yC
zfo{@R1+vNwxjK$DtN)he;CWn>v*$ieUf$G}HYqJrgq+pj+?GzQsjbNXiW!bgoxcOU
z!+S9Zvxh({0A15VCXgyn0<Tn@#p{O<MNX3_#5KQkvY|XiOFzdWY}}ygM}-TMQ>gz<
z9^D`p{@<BA;OwlRX+d-GQgVHloxNwhy}X#>QQ~(irE(*jZfybX&HPK5Zzd_30GcGt
zB*5B{$-xu^%No6}hV9+Gm_4f{ehhsbFbenlKzSobm(Vi}QQ<A{83)|1JEe`svh6j)
zKRv>|c}8;{-sw0xn!W&6A5Y5QeJ7Sbo|q16x$$?TRZ~gL)-~az#RnpK+PSr@>IF4=
z4PFcP<TI^UUZiQwuk*$!q`ezU)-$UcNY)=vT;j+Yd8HPdvvPC~hHGZg5lqic&j@_h
zR8^F?c>R`EnC6|Pmj+fV$qPj*goCONuTNy|%Xip%^fb{-GWzo*T}l@o=}_dfJNfe9
zAbD9idHGAtT>PZOPf_Ru9Z_LdB8LDa$1RhlSN^5?mzxfQCvLj$JhucPy5dxH^3fh?
z3y3Oe7@e0<l#~3PZMO(P%ZzT%evLNEg4R?Dw`*2=nt#=8+01D0j~Wi<G;zO&WUYJf
z&&QW9?#+z6qad*z4<0Z$PYd8%=F1{1w;uqLLP9XY{BuPocNo#jI!n&N)u_MiYeXu7
zp12}M*E2JXI59i)jpC)jHp;PXC_xhfq)kd0Co>Q~T=LXm6M!BAvxbW@T-nTw_a8>?
zCXc=vH3~b$7%aJNR~jenN1jc=!GILDI_?0<OI)E#7p}$0hP6`d4(KAh6vc<*+oeB`
zWso$}Tws3j`31#BN?BCHedGARo3jz{a^eQCgO0Y$`tMBD9wVEou#&b@mRW{zp(GEg
zx?`PZkHsURc)<+5E8j6@2$OpyEj<ulpM!C_as!`sk0vojC<>B$n+h?D;~-#~B|HAg
zBkzF(ZoRQ~xWdvoQ8qa_eY!RDW6P#q-bc^@@jnniU~>Xqjzu=Vs#Wq%UrD6x`u!7?
z0vhz`gGC@^z32lL-Q~n@s1^l#P&@+oThCT|5Eepnj;8=~V#Lg#Ul#idSycH3!M{+*
zJ#gcO?e``vLq*ElU=5c6j}bK;>vT7`C-!*LOh;d#QE^i?k@X+ac##~^=3#}7b(dY#
zCkwMt#b0LId<su|=F85R5f#|Rq;+Z9rFM!*n*6`9NUO>G2d#0s4hQn61Hl3L-gwr^
zQ{`x*;r7S{l>)>C*uIJM38F~PDDjjJzLAt4<rG&@w?zrK_wSd>mlO7&EJA^ZULmK>
zi;KkuYt-=I?A=TJ8FmL@_=(}u?g%7Zd2hd_&jN|_)@4qX%n}VVBW3uR7vyK>)x5{b
zbSP@Wec+L-P$xt9Mno&wro{6qNwygvNF=4YmLi6Mhr-@bxo4#h9qh58=XudGXL4g|
zkoRObDt_YyuEpj2zh$f*XMY-y_MV4$PtDCDCaB;Yfo74=W8fW=r(V_-R9$IRuUi^l
z^7b*HI#)%&qe1$bgI?4)0X(!pBMTI63>%1sMXOjlJ7#U?X<d;I#13!ur5gQ@O|r~R
zW&3%kfp%S8-7lWVMlv<9Mol_t-{M~1SgU*<d;42p=*WD2d>4W#SYS1ii&2<K?J5&e
zX_d0KSfgUm39;FD!@&H_XA%4nQa82}d!@&NmT#`4QDWHlqUE3!jRjHKGOR$vx0Q#^
zMsry!^~mG8WL2{HYsne$Tgok~tFZh0-z8gLUOR%ubaN9<ruc(LHmjDNeIgk?aVr^r
zb*t%=l^8-jCp)(v30l~G*PAFvDZEc5WDHNJE=q=*hp#VnsaUk5e8W@TJ4{2Zjv7rh
z3?)u=2&S$<2U~0HxZGGm+bZd|tI;?!gcmsDhhR^ZzuFBWp3M8%253c{rkT%oK46SV
zB-*B)ClXJ`gMpD|OuaOCRyLm?K(qPp%s^oW221ZgfxKfBtkR`T+>#0Ww|$EiP+qr2
zD77uXPb6V*bLD%ot;+?piCDfaGg+Y6*l-owWm^%;TOs402_s|^q3Lr&3Ac#SY?912
z7ksIhgklk%=7Y6NleaWx6PPkZsQs`ap(-27EuKJp&}5f+i02n#+3zG_&h0yZt|X3U
zce;oD0w-g-gU_wVMq^|6`8+3;CZUxZc3DS^Sq{N%&>?aXI<&G)6_9bb4lAP^dgtOn
zfA%qmG~m1e7ITCK%2*X%44U)Dd?EGSF51Ap!tPlX-XnY9TTCYKWn?oEsj`l%9B{$U
z>v<NLHbpfnrR{;PnOaypzn_c$t{d3)E4Iws(07veQJ8MUSEc?U_Bi%O4^yVxWp#6B
z)eF9qmNGTYg*fNjurV(@xTcU8z^hlUZa29#u#lZbELt?+Z@>@8?GQw<&viFNF>$D~
z#n;mAL2)#rC%}yvK}*$azi%G&2Dfy5MrSY)-=F-T_)VSBP=B8aJpahxIzCnyvw;)B
z$3~sn#JS9nqH*Z!^Y~NI|IX-&<DUL^<~mlCib#-fUGCv{QKd1U7OgbINRJzCs2U`4
zu7lhPJm5!al&F<vg(XJ*0)JPPO;xywz6{cvCbaBs9q9-O_XY2?!~m<lD$j=i0X6P4
zB(X0uzrsDM+uv$3e)@V1kOc;?snJ{qImev87y>KZo6in#-ch`%My9}>vHy((6%j%#
zN;GISbfO^Uj^nF*dg6h;hh34=^>8CI^*p>TaM9@#V;_er1JI^<Iym;qjED)s%YzbG
zPfiedx7$b=^M;44MHk$OD>Lbv?z;mV?qvA6Jrne&1_8Iq&mP!HRtZ_qzN+$sbW?GT
zF~6#r0P(B`=Fx|dxO{<gJ48D?U#HrOzlJ>LVsCsm!70-K|JavKsLh6UXcSvj_^hx>
z)0+uq#>!LDLFw2W{9*o-pQ)w>4^45Mmjy&i8x`Iu8wAWhSH6o~nO~sLB`FVsGyq*%
zM!9iy-yo&4LAh~`Y+VT-&xkIR4JSQ9El*zB{>>a@p%^IFG+fG}CCPuBIhPvfz0crG
z4|Q_=iK2MM<37DW$2U@JCr86%mP7(B?~`xelDD=~$k#OpnJN{=u;4=)7E(uJi-yXR
z`_qpUjRuMvM}IZovmCiLsE7?*(RuLE9%SH@yrXj`#ApKDyv?0FOS!sPIdNjPkANA%
zd&Th5PU#bv2jC9o?EeIk6yKWh6kVe8YQUX=%EXqw3{9?xbil!8#{kh7=S`(`DZk+Y
zR2LO$n)1&>&H9K{;zr+e_5}M50=DOIXPs$7^{BeKbfUhCplXJx?c|)p?a7A6XdOp8
z76EWAYXo=W__ztR;Vkn*l;-jh9U8>QjTF+cidL#}h$F{`z?6H%z-k*>#%ut?PXvaC
z0V}4!=WBldv`lX183P66TJ3y!mPd|Qb|?s8?yePXpb!zI72|avWh~$IeBFed4&KZS
z9ZyT0s|#2Z@5ToV#0pLYW_Am$;h04$Ahj^Cq2)UrrJ*B{Y6b_jaz*SBcyqi5w7~Ba
zScj=<XQ%T{<OXD;u^GVnN&<Kth0^1dnoD;Qg(JR~?_Ga+6y2F916Edg9}u->Y#aF7
ze|AK8zOL!AWXZ)hw2ps&-*@O#NzluU5;E6Q^wIY^soLSW@dnj3bz3!5^FvbD^=kXP
zOC8`#OZj3w|DBopa@;Fi7p+hd&lB6YW^#l`Tv|i#ZGEGiqSwdXlTqjQ$_q~sYhRu0
zt|@IX1E=@WC92GktxDr>)8~R(U=w$<`X(puYK@;LRW8O4H_UCgY=b&6AEN4s8}LB)
z9Rs1s`1IhRoe&D9yhUc0f9bo)t^dvps&{^u=oTSDh;9XRrHkcz;3+Cky*ge0GqIFS
zam055ViHfJL7qg6?77UXn#xBb3+}CY11`;fBH`kUG8}OONRSWihEVnpEt1@NN(0cg
zsJ7Zl4PG-h<w`?K;@x~0C#Eu<V=c4zgJZ!PK&?Ux+&6F51LHyS$}O}1i{<s|QOX6N
z4u7&%^%g)NYQaEE!WvDnw!R3|z&TjAd;0~SgqXZD!pkJ+=Jm96JI0)g5vcUJo0=#s
z4)IQ1_elTZf;1G(&@;&fm9qh{d1xyeC;BB!(Anh|5LOk0%iqnC1>IP@NWBthDE;4=
z<D<1ojIw2f?~dWQnG37Y*h|*VnDGo7gJAVn9jmvCF`iFPZU(8RsNvZ7t&}@dYwxcJ
zep!sU9h_^Ig_qQR56x1nlsUBNGVXcao4=gfZOijk`Sxe_XT;;)CiOmj%>JXw`GAhP
z+edpL^-C7DMi(p%?u@+$^aB3n7$)1JW0^_+KHl{fJZ`uTH3{rwvUw{sTPdv(Q5-R<
z&NRQd=Y{m5b;FqtqQSVz=DytJxc7Y%|DB;u_Mig#rE}etsOVQEls6NQz&G1Q<lMh3
zP|g3HxgVKyR|yaBFI|=x&kWIP<M~2-X@9`L4$|i;JtiV+TNxq~h2fKLEo#Xzjrh1!
z(rhP@>^2Bt^Vlj;@-^QqdJIj9czlIm7O1;c(W%12Nfp%E$*&NxsLF75sRADraQ(Y~
zpU=t3V%!n#f1ebWcEA^@s2C~`3|y>VXH}DCz4Yac*@SC=4=4?|MH}R38;HeSI8riw
z7k{>C#&*_xc;Vjjz*=kYyvJY}1s+(=5J^-rJAe^EphWB`fD5WQF^N@d1@w;541g$6
zbzmf0&6YDy8O7}$AqL_Fok&i5bD!$xaOZ4;iW8sqsJGc!dflrQm|sxK8~i#DFR*F&
z_~23X<d0jazqZynnEJv+>!bDSP0Yu)_7UmK@VnA4;<?@&&Ynm_f+z^DO=C&iJaj;6
zfgXKM2aaDWufQe5Pp-seMOXqeKA2(wp~<*V()%4yd$71d+Lgyh^F{BXH3Ml(EoE@i
zKL&FaDj)#B&dSc-%{_Owxi;tzB;Liwn(jx;Qc8^U`!~szuZ)wtR8F8cr~^BKOg(`(
z;bjpba~4E%CR=RuWz$a7%9jD;5AKkE&BDf!&luoeMf@Q_Sc>^zSFrlB2Ug!_DjeLB
zeDItv(6KmQ8};;PkO)6$9^79zDJKdd|Ed)JvCfk*i2&PA$R`{L4MH3md5JQGbZA=N
ztZy8~L8X7s<dKC{&=RAJ$R9c$YFR8K#AjZu2DJB-0S<XHzUY?3#zZYcDxlV>{E1M6
zI~Gisr=$fdmdx%RX@`=TcJbbFYp5rfq*ETQJLP~ijTa%bmv{vAX9{eM0&GygKN+vP
z7yVvoFt&X3R0cSAHq5PG;3}cGoSvMZLqwJNVQ5VJ*hG2YK3D`$HTl+Kh2DVNz_Pp*
z$vv;fY)^bS-|^_nQsMzH(%jOXM<PvZGkv1W|D8jujB%tXp+Tg>iLObo8OJo0H|;CX
zq8!z7rP6*fa-+d>N9o%IG_{>m^e5V^nf|RfsnPDx|3!`OX(X8t<$%@i)g&`>kW}S#
zO13k4I)*Ko2I)vgBu0v?X@{Jp#LPZ#e%NF+#L*?q=e`4Onw2G?T4Ife>oB_twY*5N
z+Lr2TdIDF}dzkiLNzmy-DK0y*!|lXOfK?Z+IViTy)qksx?Pp$6=rBkGhPW7`Rrupi
zvjkt#0-K>s^u5t?Q<nrKv+%Soo7-?j?gzHp-5fRSaeU>FO0ccD_Gq~;@Rg=(@B_u8
zrEn{jpQoxtOEm!Va2|?e8$w1#E&Gm`V0_CPRD%^8&3<mAEQBkn@~d0-iiU&+j6PHu
zvMrW3`jT&!p6LHfu_0FY#1>Q%z%D1?)F;3qoo|H|8RY^mgzPIeTI?-<-N&N$uD<p<
z8jP@$J-?5Vmwi>PIV=~+@7D~XgWdUFhNe-<FTHtwH{%2_qw+(`seeUE(1uGv57SOy
zJ2}er&21S)3h>kuCAK&z&7Cq{%C4pT;lpe<*&ar#G|CH_oq8(qu!L+sReJ~xh|9vU
zgfZCZpe3lQ6&@6h!L|U6+<+VeB7!(`fj6Vg&P{+6$G0*N&q*mKyVqbK2Y!={E!XUU
z(=VF_quA*@rUKO?{6{xO)b`XSO+#n*FO}#9r9(NOa!~MvH11+)<yyHOCG5M(?jl0|
z(_zPT;I{xpd$jSi%-EL`yD=^7nLyYB3^-?=Yr#icyc>v8xeHf_TAc%VT2XS8(VWp7
z0Ezwo%-8=n{0DGs*@_xwxKyi39C4~q-i)IEnHu7qzPer5U)S~V)|F!&ft=KajAHHY
zUcb3#kvfY@TZy0hdH6P7H%EBgZzDEq6!6*QTQRoRn0!mAm-)XD)r5MJsEbI{dE-1A
z_2z*^ExT)}!saJf^*o&?Jr!C>J=ty|ZkV7vx2#~<+t^T=x?i(EfaYuQ5c7NHf<@qK
z9E!2|OYIF&^T&>gpG_!e*-7vA6AiWDiO6|Z+0N$Mg4&rsy9al@GdtQI@y-8Wq0PCZ
zUHv@3rdyp-J>ILHKKgIB)Vn99RP1HhWa|NuTbERtDsmgO*B{?UKazbY;b(DG(l;}-
zaZR}o&7b+YlePR5k+>Z4p_M8o-#`DPxqn&U>Y=W9HG8L8vsSOr-~3az#}^><HAlbp
zxs>m9{e`OG(sj`f5qZZU2j%P63ftarLn7i6({IA(3X!jb1_|Q|F1`df>+_@|^)XFv
zZTl_QtmXx;$|vn=60SO>{M)iN%P$PuqgzfL27fXH%szWZ%nDA9?3tSw&YIoTRKTao
z40v#FAt42E1+N^s{5r{Y#NnADVmy22TmI#|!yraFR7*_`AcE~Pm2P(;T4-YX41xSO
zo+u+E9qQc$QqMUt4=?}oVZ^tu{gSPUvace_9qBA`XY6hfR^Jv*WTVm-Nyo?r_R`+;
znu!yYz!1P4D7cc@8C1qb{qnyJGaLVViIxfhEoe3Wa_Y>`a@n|?D+$1-alE{G$#nS_
zAF00=&IPDU5~o)+97v|7b+g`+JJs8~s*qN|WD_C3XP+pctZ8%9bZ>0%v_jiQRTUtY
zh$|}aUQQ^Rp4=@Y;>_KXezJC&A;KhZ8w$Zg2&3^6?)z(zE@aeii>$N|?yEB8<1|?m
z*O94gi*3`E_6JJl1AUyS@X>+<B$cY05EY5?MKlQ42Wnm{{)hopPb%{rAC=bq%_QHv
zu7j%}EqxMa4&&(06&Qf9H!J{7i=<vDLhT5?@{tgBTG6U#$GHiF$_0Czds~`ya5Z-^
z@~G5`UkVu1L7ykJp!-GyJ^w)b3*9wWetdhFmhSD>*UT0{0`5f~H};XE@gt9uKdq3g
zbW6Xm<rQ!}dFUPFI}v&<?S(s(H}uuRwva_Y1C*EjVq`?o2rkxN$$UUWE7#f8Wf&&O
z$i`*EQ55m_8!M@~oPP%ulYEZDj-=yROFQh^cO`y*7@5LR^t)N!$Dz6owbXpXZxqW{
z$|8Ar6GmT7-|%cxi=h}15QLZ4f{enWA06@I<J}I*F^Yz-9vL<Fm^b=STs$#CySmrX
zk;Lf~L)P4?LDL!~OxB4&P}R6<49_~WnWKde4eBfic`vb434Q7wBl4bWo$-^r+Enu6
zNCJtq&q<8GkskvLSr2DH@i?E*x#Jg4cj&S`Ov{HjCmZ5w72?cFTZ|8@uF0`9rN9hn
z3A8>w%-S=L!_GL6#CiXxR1t(rxRt^Hy^}_eemX`OB1LTaX1VlOvT`t00B<i>3gfxP
z_~YfWs7f0jhY=InJTfNLeqp(GgbgrPKFN{C%=S!Tf#XXSnk%Y-Eo&qC<pdrt{QR1)
zM5G~Gu#$O^HIv+3vIhl+{33rt<uf(30hdo<0ZtfbKGyo9xEt}zXR}kv@Fdys&nZ3+
zA2ju?_D@yVJF0ZMk9TwGchZ$68P7NZi^vbwaTT@a|6M=^C5QHkG~*)a6216eU8k8v
zY?ADOvXhV@)0+2>k9EYH9hvEZNk5UN>3z5VJCi8i1t<F&7BSG$IW!U7R9f)K0%`$V
z!s6a)Y8845|KP)#FQAL=`cLrJ>;7b(%bXD%Rb3Qq_w5`=4_ozGbJ~a#n7?|*(f4P#
z(!}KPlg`izMpjL}o8+v(c`?_xJSn@93&lci2omY;scPCNHaBpja1YlsZTayicsi<s
zi6;lDkCLU$17dzKs`UR?^QxYF|MVRE19Rf{DrMQsouke5_nhfy`0SA`Y9TLFF@%?H
zf|^@<FQDN5LWLC}Z`CAGRu8ViOd~}e6@;BI2l)b)Yms$fm_rH2%`GO|RyYGgLU{(O
z_-;wBHORn^b>habr@U*am8ZO?k{3KDvy$(;d;_mwoNs&Oj_P{EhWY1HxyB6pgDfkp
z*o)PZMc#Iy=^BqUI9zTLwhtQ24oS-<n>;uD-g7yr_dd~(*P!;zjLQbE;(0z(oYu7@
z)VP>(Jn#C5cXMgZE7F;`4IPq{))FO+Y$6%xSE+}z<txs;pyA-+Pw@j<+4|wIaNV!g
zzfsuBkusEB#Q5_bC?U5;#G1hYF-0ivvt`30b16f0@8BI8xkbwG)9i1<ek~5dX&0+i
zY@f+(KO>iEH&(gSSi;kc&jCbGU;WEYjs*@dA73CuIeGscP@2r`KkL^loC+}4W`r{k
zr7F%igsNjEV0HfE;~)<eL_V>6u!%{yqW-*aMXlL(Lxilfi!8kkiTEph5ZuBOgJN-!
z-SaP03(fvXiV*$QRugHfCp#(L#oKx$9x>hFwWRsi%78>}X9P6!7ivNpRmvK<e1~#c
z!a92AucjrmWV%m>x%&4?Qov#Q!s3-D(LWetEx)ne?*x9@5e}Wgeksq;VEf;av%BYV
z-&xRZhFK|)JRYpPO1@~PWj@kURc<DNq6ko&uZfJDQy8^L()YQxu*uzjcZD$I9-#C9
zj4!iXH4T_2UQ2DI1SbA~I!@UH+&;WsMwD)CXSUhncu9U|KLdtLI9)N0Tvpg=S2*F7
zl0N3;8rVn`gLXQ`i2j~NAxmd?Z!l3bqo>T@ZEt~%)%%r7jZow+f6hCjpXLlWBeNF*
zG)5lRQU**Vwo#uR?VH|FOndZ{56q)89lUMCqV-){b!Y}QZ(9=ovap?wW_&8Me{Ru8
z$uyXy>+)3cFTJ{YXhc}uYb=(Tcp(9N>Qcr8Ofti<lS^<EojLeg;X(bA;YtCTqDj>H
zh3bn9UvzAKY#NlPy~}UWX{4B>-|S0i@ID-Ho3K05!O0O%eA3U&==PL9ZU(24tu$mR
zGHBD!=Otv!qqWT8dLHl_*=cyw_(>4Wwie8!{(=y{rADSuOdp@-tsa9G<5R~N(i(un
z4E%{MaSh(w?j!R@<LxghB3si1R?_tty>d?cDV*<$s$5LxqBudlgzP1`0~kIVQ%O!z
z8s$s!o-e%b0GND$-cc&XH3MDc?H$iS1SL{6_a`I(0A|vu*gOC$mS*(<0GQrMvY<qO
z_UE_$gpEBk{lM@SA@YizWZ$R}>vhYHZ`}r&8s;{$gd%>rLJI<`4_SLK#6y2d0Q7hI
z!fe&QJxSPDFagNtbn(O><`y!h0x2CcZi&F%x}?A2%|?{LjQj!d8;9HVHSo3^fRvI5
zomGB%7!CjT#lPc+rpv&0SJQuIWCP;bt{M*?@d@Q)Dv#3=b$_4;xC(}(DEtDdYsKnF
z%QXb%_mh!kuS1g>*m2l_Kl-}CyCoNGvOKwDL37x-_#VHhW0+WtF#X&`NiEgQE^#w4
zDcuXScKQg7<KgUI=%>D0+`iyV{+YWqvy-a$_D|6$0|U_oU6Z@hvMJX7kz`lPqAT6Y
zby)dnx7hN{K92j{6}kW<>~XG8>#>cO@Wpl)-cU(47q9nshU3u);)T0CbUaUN_@X#V
z;X~v49ln@t|Mbt`l6NR`pD9K`Z1Bf0?{2ZTwzJYHzg|VD!>`_qm5Y3}>+Se=<Lc>!
zJ39G6jdC4delv=g6?cv&8nNGT+vj~YO-R;g<5HXq-H><fHzHxzz<E`}j>)_CQ&ZeF
zlJ9_wODN7Sc_wjBlepYYwu~s8FsjIx=JrqO8(D2JO)46&EvyQPKJ~xT_Z-c!%?>}O
zSQwE;$^CcsCG0vlLW!~(c=rKK!yMy47(UkpB4<w>T_&hA5|<!^r5;3Uvti1rIu#SR
ze=Q%g?7B`H8EuI`V0)N#wcQFoY5a7$w~htA{5*s2b0CcPL`nvx5+4a@XSeqFEQavr
zw%dR1WW^?YWtKg@+~K!?r?K%Vx9GtwJV07?!50na@kQ~<8DPAwd)s?pE)<fZLC1*7
zt|(1NDSHu{9!vwhKj7lsiG=^TyoOsi{oT~4wS!_we(ZCY`&cC=ZFe{k%J=n1PYe(S
zzuiQ}bDj>C6pgR8T$c`$M-FJDh2~1%I6A*Ey7Qt@6X&mXq(9vU*jR58hO;I3J04xS
zS3Q@$kv)oo(V$E7`vAtnT~4A6^@IB+CC)a^2>GvWpvu^uOPg-+Dy3HGUk8dX97j9F
zcz)Q(8aiGjTW#}pSbB%4yNtKHndQ}x=}zH3?I<}5j(0rv>D|*L_wi@5ic^fy(=_ax
zo=D*=%ht8tXHCdN-BJJ2hlZ?Y=E|{ak-w)*stuQ<hkohUEC|5-SDQJU$6%>eEqguz
zt<7u0bf8uND?fIkd*h!tlvl>sS)(Rp+WZN>(q3}(ikWY+-Qu;8Ab#{yt0TprbDw&j
zpFcT26>z(o|A|jItHGmiUFR!y>dBjtlMhYpE+f1*ll&-ViDV0X$s1>7LYd=m3Fc@P
zB|Y<FUtq;32Qk+2>uZj>$D$>H#tzD@aehq8S8MN{H75`W-PZF3p&<pU+s|m9#^ySv
z4-8fZLW*mTgi{>!$FV*gH0QVtX6o1i5WbLm+fdKv*(Um{<I4S6;yYFZ)h4owrl@13
zJ8;^N>w~DN-yD61_Aul|2punKwzSxF9LBM03q$H$U9NYcU_qWr-$Z0<1;EBzY}^jP
zV=;1pDL>EN$hQAgQ<8(>uY}A?&za~2wXAy?|I_|YYy+|LtNA{^r?z>6AjRs)=bG9i
z>>%-0*MsE`B)aSp<>^PN#YN&<oxL)`F^FcdV)$%+vn>0^iQer<$dZUD#U?4*{4Nk-
z*;V<5YyO)#{35;r!bSDNjj@}60x-!ywjB@{fn5fO-6UEzU(c~D6^jR`z@yvF<ra9L
zqDy4$4W42mU_zE{Q3%W+?+bolq%F4+zZ;+R(RdD?$G(}BWiby7*4ixk7NgdyvI{Qd
z>Q09`)}H$-62lP;B27#H?Q=jv4r7WibzzY#i_l)zT9qrgZd$S~xFV!K++qKd6osK=
zPftfM<=$P)za)oO+Ov32^jPz{FLTs!T_`r~AoYt`ob7%P*S;Z|lD5ptU|GVE+;LrH
z#$oYk2T9oyM!%m0rv78Zc_I;S5e14EBSl|$!*?h-yE*zT7}^{Hvoldki!3GvSxD5>
z4C56yg_Q%0S<ihMyy*ko<YTd^XqD~sZ%~l?+d)3btixs<_HoJFK|YjkZ4;YbEOGxh
z3$HA_DRf<}>nXV~R6BfT^m^JQZbw^D&`rT<n}n)int#fCNG@OdRWq}d5kd2*cCR!m
zvx1~*8Ltt`L8se)cgONV)}GDDWw&)pk(Uq^pP-;<RpH7yD_I?Db<UE%-4BL4Mp;z6
z#AM$s4hskER;TcoiEv$QAYfM{SY@X-OA>#oOf0hLWpDMr@V|HD3tjNyY^pAcknYj!
z2^HfF($Pv<>lWzx>PcBK<F-+p_kGW?20NABlWC?-<;Ogi2#Fv1)HM%IRP&>Q>V8)F
zM%bY%cEeuFY(!&GSsZzkF_0q7UEt+j(sZK(%P791U`j}*lNPFy)ynsDPYr)<_}}rq
zcI<ZmR{y|LxymZ?^nqOkh9KvCsQI`Ug|we!@W`CPCNVzek#OXuOO4K#&DxdJ%%ZTJ
z0bLH-$Z%t2yGSLlK=!h~vknFh^kn=KY3sK>dyd=eCQ*N?<?8C*Y5*Hql{@*008BQu
z8Ja&K5;&HRFZ4h846NC!7XS$4<aa>_*nMLVt3^Tp+~VKjEJm3+=P<>^Cn6@lf|a)`
zG3KJ*hwrcs+u2lFL>Y?o2C()tlcS#=F{v-S%@=meg2#)3ME5+KsFexImwVH`P|B<*
z*%MP{C*s{Qr3}%gXQyHtDy@Q>Cu$Qxs|Scj+23MnDdot4^11xhIP7f=*5tLitVzhb
z!YTCgW;F~6!@hskLfP~o{q~92F65t;rjF6rmQo<(0I;h|Og!#<QP>Z^s$MA6#}Y8N
z^h;YQgboHcH^bo;>}BjD*PyK7I*HeTmQxP8zU$?!Bg4T)r30Lr-H_|%HeYvy*kA&z
zI_qYZxH#JFtn0-V<^QD_2M>!@Rec%W3Fx_E*7g9aq>|_nE$k_9cGP;U_9>TYd4on5
zsXU#hCq6lM5xRCYJ(6<k4xGcGNYb{$_<#58p24n9sqrN~RBDd+t5|-sQ)pyY!KqAn
z2DD}xBz^Cr4u*)<+7&iyj=?Jv`8U)4E#6`GF3j_=7e>pG>%PL=czG_!F9j$}i67A3
zoN}6dn)UKI4Zc{vd>fDS;JjIS?b4FH{oh~L+7MioAA^3Cm)<LP6y54S#J39xE#^DQ
zcuhSBWZp2DSQ`<qYr~2kojiGhMPvoEQ`*gMzx3|;L9uazw7ALlq1~=;^ANAbFc*eb
z33*DLWCIXj@^5Kr)r)(!2~^tjaZXOsfnblSIwS?bJ!=*Zph0LcTnk*KJhjTd0wSMF
zX|cxfR+LTC;!YLwzj9y!E4bWwXxSpvVGbj@>L^Ayklrf!AotLz%Zv~HWV=Gwb}E-V
zee4f~y;bWv;A(CbNwQg&IZCnzIbO_txm`1E77k~jIv|iCPvf3Z>W%9wCsX<?K8-D_
zr6yin`b<9I+Qs^kt=DK4;R?MnepDM@h7|YT1F$5aO|GBP8Jp>w&)$`{j+~ZQru{yV
zozz=`p7qFB=S{*fg=m4_9D#2#yUFzHS60#^6dub3CjX#{^!MgcVvtg;=kZj(By*_3
zaq7e5KC`XU*-)|)mUG=-m-6?Ns*dC~uX);-l3jTrgZDtT6Na8>zUgY9P7FS}bXM-Q
zr9Se=z=Q87<wC^vu-f^=m=FQ@=%%<tN|Mij#gar%A@!cM^--elwgHg?4AliszSDZs
z@yw5{&I1&oe{p!j#@^w6J;koL+k}kFNU1zh&l0LSP2<V_NvOdKUt#U;@{Q!ntm<p+
zDir@rPU!7DB)J$l@tA1U<830w@dF<P^5o?rmZG${?NPo2$&V8^ujO_YE7<O+K~|vr
zKNNWgL}O7xZ>;lxxMwrjQAvViIi_Q+Gi8})@D0s9S#0x2t2pdtePub&f}QKB?Y{-<
zqRKU(OanIc%7(O;#ck-K3iD=5tNnT4aDjnzwv5IBSl4gYD;NxJPgppMsc_`nh8(c4
zu&~@61{lh=$a`Mp$r`KzK6%8~+3pSM7L}j1DV2rY%f~SpSa;6gFy+<U=g4C&3GaN0
z|J7E+xjx=CLd@k`q`=RL@K(y0kGb}VVf(E2L0q9!wzLm@{^?%Hi44;OcO=(0W4=v`
zW*tv$%DHdxM24Bv#p7N;EC-swYTnakMQARycti#@dz2w~_?T$G9D8DH!2Z^3;paNl
z;!|xiKk2}S(uUjSAje;-9a1(YL?_SbJN+`4J>icEO!%G_9>q8C&97_Uwr^UI@*s_l
z9@e<62uSIr@5{iWkrQ(-1IiPzlga?;V|?NOD(zWWNSE==x90>b^vH77D5m`?DJuKS
z)fJvv=7mV+S)xhFxs|MVehtVwvE2@Z;ITiOPSCngM3T1no?%)XY?86vh+j^mG+JG_
z-NT-{w%2B|DW?xD25kZEb{kTC3k^)%@&{K%gMUL>$Dt*ckrnpnD2bclPTj7`rIe|Z
zNX!{Ru(o@aoSf*zwFt!M^w?W%ShqK%JoJet<PuurZG%c%(b;F~P9kjJcbAM`!RPFs
zYbkaG4jfqB84rQV2j0Pp4upgkYUOt7z7?&QLJ&K56}?}QPjpXJunqU}`&5iPxwN~Y
z#z`JS$2=d3$v^f+?TPYOx)Gf5q=xj=#jBwF(tl^dct7A8q@8F1MW3+D(RWbG^hLwC
zzAI%$1X<F~_zO<Zdz}9_PBWXT+lC^|=S{XwZLzyX_)ldU81PNJ+0s|f(3%hV6$`wY
zK%Tj@6m=>p;RDqwe~0<W80%u*e56uoDohm9L*72>WrlbL?VGw9coc24qtexc-fX^!
zuyk7vLxDpyn@2hAHWAI+g>ONUg^yyLZ6-S%*}t*h>%$RTvoLhw!IEQJ1FBKuL_1|e
z!+b>p^QADtG+`~^WqX{+P<*mmT%sSMwZm{`uCoXIc4m<|;}}=;@55pSpfKata3=Kl
z%GyI^p|K>L1BTG-F1Uy;ZA40C-)o%93v>bsFsozbTJ_8w0N<U0y#opVK4Gn`$W8hz
zgaQ6}ijh6LY^*25>!6YD&W4S}j1(L}T0+QhBF@AXm@pq3+FH`$XL%27{>h^xspl<x
zzk}ebMg~sgY4!b;1o)SMaxsH6z;7n5z`tyE0GO&0q;M(uaaBPl3=m*Vx*7KL_AhqX
z@Vs@QYi464rHUh-X9O3)G9UU!7eZ09QZ(FC#Ih+`pdl7K4smw>CGFog%d?I8!V3(J
z;TEP;6`Z1ew1f#<^L#7Sr!_ul1{N7VS_XuVCM}*H@PcWkC`$dnR}(Tb8?)g`L}MJG
zXWKjmX54^e435>;d^K-7rmOzIAn6Mw#E*qo7;?jRRv=54rNK<~%4XVIn6?X1R}8OU
zGh`G+iQxyrrs#%h#k_CA99mXN33}=74h1RJ`gF+w0UUq)*o{2m@8}AAkM>C%`~3ZH
zEp<sJQjS3(ymdH=59#gr+XZ!vC*gt!vq%I*{C>W;ikfgkQczW0#H=~gE$Fr`$k!V+
zJn_{kFY7Rmx1+9S=<&#v`ui2a!}ZC7ks|V0VT5$K-ZlL51D*IO4<ci9VUqZ)|HX@w
zE5F;4yKReUhB{ZAl8PPgnUhf4k-{7NzI@Ba*w|}FznCl^zt6DlBeMKc*~$)o<nBCX
zrQJmr+hIg+o4E6+=U;uF(<E1SG)|LFbIX+M&9V+GQ+B0WFrL*m(sX5z$yU}G^<`-#
zBzTMUrV>>Xy&liLNV~*Y$^BZ@<4*8R%B};P23lTd@9YQ_p}@Y<6oA>SY!xpVpSXm<
zzGD)IfQhMVhXK=NILI87X7#oh9YFtRCS$NYboKnUm>4Vq<E3waUOr@Ez@~MYego?&
z6LBM<7bAK_v8x3Z1@M39=a<L6X)R_@HUW=m>A77$LN`dNygwU=ruL?V%F~qk(^TCB
zuv5y|jLGR-VDHRsr`cfG4!~`JG_f~N47ercV%~{Qx3rWt&?%&2k;m(-!IOSBhrbKv
z0g|nCHQ-XOIK<l#uQavy7+`c>jC*IS@WfUeX1RHnw5qfgeUG-8sA!)Ut1!H?>z6ZJ
zd4V-@zE-xrC-Bgy+3j8zzbjakWXzp8>+NFF4O(K9&K}GCWNWA;XHTA(1Uu}DXPye=
zFCZ={>@X(|n8R$na*-F?jPqOMWSenkk<q%OkphHWyt&5Ed#<MsWzzze&=>ZFU#AuT
zqv=%@Bt`4^oxZ4L3cSnpG7;Wi^?;-4UH0?P&ypF9ww$s$HZ@jz!`!e))cAM7&8mf5
z@(6<a>cssmGP+<{R*`c7PROW~t<-lcpz1f}8<}V-KOJN6a@#C7{La0!r^&yPCl_#_
zc;4pj)2N_DpGI?ozO&M|>+)}+^KLDN)z7n9y-x~uu2Bf0^O>GkVDFrKG`I{iF%Eln
zM>Ahn(b=d%?!->`tKOKD?t$VOjIVZho1eHA{s!&trb|^Hb?Qn!<#_szOfA2_RrN7e
zX~Xrd&gb$c!vCFl?%yfws$lN>Z3W`3*b#RzsYJniNaD32TgGeod2$SQ47+vJ%1XLC
zW=ZgVq0+hM9Zy}9ha>~C-D*|5Z9Xgo^o=}!BlCGC7Sv_Maf|znj9=Z&zdbSl$FybH
zUD{(8noSYKhG_V{3w(b_%Vl*?z|!U=5EV6@+Y<yD1hNidTB7*L7G^fU(&Viy{<{p{
zT+o1BZak_&Y0;nvAWU={d2PYJWNae8%u3PuBUFIL)lSF81~_oz5GHSxD-CjU5BGtQ
z5paS5Hm<jEKzp%|S>9hC0*K_Pz>$%75K!(^%1TpuN?h~u!0^ihcz9LSi&sN2S_Y}l
zs2{Y<U0`IZDcK2@TIC{Tj2hJ<gdA&;h$Lv*Tckr;SK`M{LGERvQZMs}t3)AB&Zm=*
zDhEVhz=2_f+`5=n8vL9=F3I9q_HzW%0Cc33cP!I%dgpEi9VNTfVvz^I0(E>%Jw>^=
zKqo>f`wIbT5H^Jv?i)UEk0_v{=B1R!qjk-=q9V;sH?29$2uyI{nmCo+9laED<r0WU
zzUf|*;_){#EnH)+1VPANQK$&XW719LC?ENM7p(L@6`fLbbD7IBOw9Ns>V5rs;R%TJ
z#nMWt2e}+!C%H=5b+J(JDYU*Opr^@8(jjJ?7!MOA#DZ<R=w8Ox{yX#NQtBWx>28Dk
zX>-hFaG*P?8Lhl!R*<*MAi=Zm>7h);g>&^gy}aDIRD(7Q82r5Uu$RlzrnmLJP|RAu
zajpoWPza=1D><fiQpF+RhPy-vB@BpnwHbWdzA^XIERy?IiXK-};6R?<hZo!(m5X$9
zBWj@R#ExP-ShWuVtPL;3LsvW+H>g`{ziWUPGYN0{5!xMIyJ$FE<%ToW9TwfIYUP&|
z(z$WU9i-Wtvr%FXYL#sE`qm-Hvd0r3+e)%_RpQNnb(0e={$fqfQ~F(b8ba!`TAMW^
z17!pzV5uu-$u5=vB;F^kFRjP>z(5R09a3FgBjCtgR3*W}wht7CGgAs>aMIvZL=0V}
zr`#`U3bi4>nLIxB<e2{`E?ojlu5U4Q1b5h1`(rbvfo$3l;2`15cneAF%QzKN`)M12
zH0}TEBM&?IjcBJF?)*zKbW_)h?zPWPHCdCt8&KvT;!lGFDfU@~rnEBW0tS6_JOg<+
z_uLXyaKBW#t)@6^fMk`0H_im@RsQ=4d9?z^nD3Dr{NzUo=tN;(V06GwXCJLWy{#x2
zj?Xj?l8&9CwT{6%SuuW^Yf$zvb_Xobwk)LuS0ct13|PAYa-R<H&-+<J+dU1r?F1XK
zZMPN~St3B<-e9OVAJX(gqy(_+2<eK}jR@T}o`Tva9$&7HnR5hvc$4iFDj2_+wWo4h
z!iDAI=xAVlv(b6E-fBjX?sI30Dmz(^vh(#_U)~i)Sn+ds;6aY$mSgF9)hBY|{C1k_
zutCugIaduMw%i??q{>DXNSzEN<KLB!@Ah<r6jF|)Pc(U*|4DvT>>QM7jpul=)#}xe
z_a;U>?XBs8st;%XdjwEQ*wpM_RF;y&X;t;*Tafa!$BmuvQ6PtNz$ubre9M_HcFGrR
zMZC(~w465GQDPjpV%p;+?o36iCHqNLkj0H)1lrq9Y3AtlOL`-zgIVbxI(*Mm!HGP*
zo+o@y^Jbqsv%a#t*m5mIUxt91SPNjXy??EatBC9Ec-#Le;0?kUY%PmJAMaEjLGnL`
z`mG%<lwQ|AYJjY7k($(OY}Z58Ur$;esu^pQki2ZQ%94;;S?RsQ4m0DIJ1okmSMQ@)
z3Ew6o@1K%C-odN#R22t3bnJwa8;~~l-P3lod{E%v(GHL6{AiiPSY3$MYUIEkUzu&h
z6fwuiSW>!JJR+sHasr^u3>d&X;1}~lZ!Rm4<jbVa|AXxO4o)hE9iB>W`p1h0Rt`@t
zUc~l%0@{NcfMy$U$HMynfY$`vt3t^}Z!<Fj(G?xVL+|qmPi>3XpRgqHU-T#WZ^n-;
z8xGXk#tq~xUAPjDrmTAT?m(J#F%$~({NsIo-N15gtBq+E4Yamn(N`#6q6NF<T9QGV
zXZ^guZNN40(b}O|FJ_Ku1>y@948(Srdnb8&2PSgJ%1Y)vKj(Uv1D|Ce%;ru14BUaW
zk1T6dqyx&<3BU?TlDEnb5S1%B&n`rjck>>2n#HBx&al^hk(npx_*~k<v?-rGCn>E6
ztD?qc{9Ez)xtYWdIcrN}{_@Frkh9(!AN{{ny|K&8j~TcZZ}m8qqu+P*iR#>Hf{9Wt
z2Nr%-xKH|~+O7L|H*1TzBxHl&(N*HNFC8j1ALRw~g$I*CR5?d}%jHVKy-Oc?a?R|&
zHebv7HxSkQtlRqKppCcsb>R)Jo&ii<o*5)i{*vU5!P`gcYlP&Ntr^MgLGo~ux9v!i
z`&d1z7$~jADx%(8;t8|)zVdFRDo3THsIt-C&w*H(sVwP}yVB@eeVAYFqzboN0_T>@
zH@+yBDbB%!HQL6&oOV4E5AYQ4U>6oEem8}4H_7?+Dn?o>A~e{`Kpi)Lmxn%Zv2Das
z@n!hjGQg;6C`OK!G5!9hi&3AE^I@Hz41nQL=X&es)qp_Syw<uLONfdLP1Y!!zFn41
zw!#BG=<soJpY~npMOmI0Q+ps22Tl8>`4aA{K8>S;j%+Bk%JAi=+LAyI<G(p1LGueh
zfDPFyl}o^P`+ucF2aeA#%aqd|B@$39x^g|;Gz4m^qVKld>K;VyoO>dI<-44_U}H<)
z)bTFRVh2NNWEGKexlgyt6K#vkKFR}+hdGFn^52;agGTZX+Xwv(+j-w3gaF}<k~UyH
zO1wu5Pm;C1dHz#V(!~mnDDeZwg8GUh$#W^I2^|OpOW2duI#h1WdPT|0+RGDyJ>w2l
z3w5ST-u3)8Y9%~@z7@NAjDuBJ(2wztcXaS*q>Sq`{*IEHEYe@sqE)JdLOb)Wn6D6O
z1{xNmw?z5Yb1Ww;a2`A2_<^<8gp4NA5a;rVxtA@F7O{FGF`mK{RfJ&&nwXN?7cN-5
zS}x+Ymb-1^Ry7RNME;naAb+ag3swK`%%}}mtRu_I=u!;NILutvcm9pHwf({)RZUfZ
z1R-;WUuzZHKyrLupR1RzrSLe=+d~Xwn{iEs<d#<2BblX;P(KLX1|9u>2b%rx<J4N~
zNCO-4ULjGea8qt6!w@+G4K%mzl1{@UvIAAlcLAd(*}pSD+O!{v8~BKV?Tv)_L@*}s
z(odQl{_5!_1s(!sH%ZGqgt?yT#kxev3To=S(!`E<#3lEe*h?tbW1r;Cs~Z|SCn8t8
zzQJko>3jSM<`aU|KznqC<Bz-tM$&)of4Kiyi%jY$<;j81`mcUXJXK1uSb%buv2>?4
zgY=Y44Ds>jOF&DssH>Tg5;u`^3{JX)4`fKuHt)N6(sMJUnVcM@de3{Dam7H3ub(b{
zHw)XSgog?qK`Ih8Zp6(U2y7jxj+~e($MWtkJj*(VjV=w~Hm>>dQ27(E50x(ogREr*
z!`1>@FrO{nS|R3AuFR^W8+Q?E$zN7Zq(_n;1jjuELz<s{D~Qq6wy{XQof{c`B2|<9
zGuKgXZduiMK$AzvGW3Cx!Lvi*>3#{NUzK>?`%>Xo8={Kl(4IzBpS`BTOF=dH9L$Zf
zpdnL)V4E99yzVP$0^jei_N+%m8^}AtXiKAEY77@D)o4A`oAATLd1L;^gxq^M)#Har
zyO0wJ&aVM+k^uWZDDKk0t^r8))8PPLf}%6Y<I?jN6|XyPCTe&aEGdj<nAayhrAeXA
z-7-s`9q%-K$G(?8Of54x_G>N&k55EF&!_XPp|JrSU7AFPb2*s*(IRD<fAd223^}J_
zSZS!%G&IF!ydwoZC6zgQ&w8`c-R6Y<fw$7lDPAo|FZ*WpbyF80LtL9mxxaN}uv$UY
z{O#H48JFA7#9YKdXo^2ilg^bNiGaDr6&}hz6tUcw)5sJ8;lCpiGboK6-bqwD5Fij`
z1=#GDr^+ZBvjcUOSAUqK5c7FP)oPqc^7PBa9T2|<(x_p}RjNYQ4budJLh--ZmgVSj
zo58j;WGkEa$E+3|^Ry8{B3~jOy6s+_oro<EKy+vUu-ZWpQGgM`3C<mUJz92s0AXCK
zLSKlIAaA4a8}QUaJ-*(vIp|MMpY9r(Ly+J6cCw`7fk(N{?$L?t56R?7nTgojqZeuu
z@+;?p7kMgjow<?&krS7kuC$%ij|{>suUWhH1XQm*5cr~M9AVW+3Kbj+W22$XR&G|G
z2tIc}DEf~HMB3d$h^BdbYUwZ5u$><>7q1+OmgzTV{v25RWPRHFVEKLeD9ORY)ORA<
z<O2~~5Udlja_y+%)MbFlW+#0jiw+sZ30@Z&`TpM-5u>l_bLnnEE^8)jg&0b%z~u6H
zc)eMn{F0*Xt&zGLRr~Sm_kYAHAqUZ(^YO#q<_SyDLZlAdKVCnaZy6@5PFg|6+h+W>
z{M2YygE1{Kj1;zPHyLs7-i)R5InEos*>53lxt`j8W@JVAwZs&D^V-$CG&BC@oDFd}
z4?LNCuxb`Y=QxNbN?Z48HUepj$rEuM87ZE#YdlNxhzec5Q=x93$AKk}32z7EU4SLu
z+uOJ;%fKYNoH3Il2mmYq^uWU0T;QsG^oOQ0Dt5d&#eT1@wx(Kz-IOJTJs2;w2%a?B
z(LNRJn?VQuy9BI?bPhxL>`40{-inTl6M;Ap=kyvq@VdeIbZIPIGW6<zW4Pr(!LkQc
zezUFj92wc5<^9CWH~I=*Oa^(k9M)&<9+xICD6%fik+S&)PFK3`1jyZ}e+oPI^h&5V
zeHoyWG9fxF64KhswTUv_GD>KC%*ayGr34Vq+d=??=2@g$Qp)RL54cP&#W7Y{*Pr(A
z{bx8`BD1%gjD&O4IgY-*@x(0*N6}KAsV5QQSw}mA24e%{_XwoNTm~<D%t9X;jFwDV
zPDvFTjCHd#?P(NajxPk;^K(>x(t9V#aVHT)oCwtwx+oed<{=R7_p<6q)eI6)t()9x
zsSknRl!61);p+x~zz@e2m5&)daGqb?oE<=a9V6j55#ukAB%2FGET6;pZ8LL<`0*WX
zU9&>ms^W&-EP3BXj9<Qznt6Rk;(xk1CXDg&<}kIPIQDACsIc5GF1O;(r_@_rZ9WYd
zIE4&wIwDT_)k|m^vi+z9^#yXR4Hc1ocWsyLpE3%teIQi$piI_$H9P^S-=t5LGFzy>
zur2zSiQf9IPu$7*dMx)a=VL41CLc8O0V!m`i0ngSvL2zZMn3v~9O%O$5yJ=LM0hhK
zA&dhG!E1<+Q&}|H>4v`t{-h_c4RDhm&ec*t_ph``-{kZU4iezu;>>8ThPi@akRzb(
z%%PL-^(?Z_)P4!77gz!TLhg_(<*|gjR|pn@>yXxYgdBI&MV&N{x5+dYpU2oX2lKy^
z=K0F?E9c&z2c@_Dx4X05DNOIjk16S&DP<G!X+fvlgb;ijn)wOw&uoE&u1UhHA-19j
zl&|Ex1Z3?86YL>F)p`ZT|Fsm3(KhKF4}7^(KAHReE8j|0|HAF-QYreL5&B-=XaxCQ
z&@TV&rG~M~>8;HZS}*lHh8vzJYT5SvCatyg132~1fiy@Zi>18~`GAVB7wo>VJn%3m
z?t+TY*TI@;QpB^T4^BXVI(O34><4X9z|O0{9pd`+w@fo^v@yu9W0k<<cJqIJ1vTu)
zvg8bz{<wPTY;f(Aul(ws*Bw<G^)mbEA=>o0|0C(l!=Y^7|Nm5E&%TFdu_Y$^&SM?B
zLX>T?Q+ATwgQ2mGG1hDuON%5+)-v|BFqDiXM4?6vO=B>8ez)(>e-6jtV6N-Fuj@R|
z_xtsRIL**MZSLPuh0eZr&3_&zPNZG^H&ha4qQN7`oIzWNPXyX)3--CixAF36ZIb5T
z)Hk!IP+#7#Bty&e5Q1B|4bddHHsMpOd=CW<JJxL?nQc(v$?MGFcw?%LRe~t#6=Rew
z0VqNJ1`_yp4p@J!0%6z2wiEfwwc&teF66iMC22z#kRqyj<*{M>nQsT+p~xw}l*WqK
zGHnOR&jg5nCg@x*i<2o%mhi7MyW4=<uA_&Z_1vUK3eRjEnU07opTk;1C4VG>uRlgf
zyB_YX8boJPsH0niCfG*WSlM2iY>9=`%v3UuO~BT@&#>XCBC{+y_C8TEF@`bXnlL;e
z9!X0AMFV~HAF0;;K%(~2q*(5W_=$=JCh7$K-%o;Aq(1xp$OlRn=@@bvn5!R?KAxEM
z;~iui@XkMMUDDw=+2M{ex4U^aDTgE<#4{FGFpbpO5?v(S-RZINzr`WP?AY+(@`4-*
zDL8rWl7Gkl{_*d9tnX#iFTyRbc-Um4GRD(*bS;_HDtm)B^xk`L$e<8%vqCW|K%6B|
z-F_!abY=)G6+wFZP+Fq;p*%)6^TjGQ3S3jQ?Po51Qra_+BlGN~M5tp@TcVzvP2`&M
zZ`%AKK4q?If4<FmKdgRBLTbYAkN0|rcWp-4rE~mxL#*e+#OM{9x1J(J?@aVf6<KJK
z;=E5ET~K~|##bg|9-79Z@{L}vb4J&Dzr7TC%{5oW1tm@^#+XGJPJi8SaQrSvQSg!;
zPtYgM%C?-c((xwRh<>-_W8MP<(Rz+zxUxFaRq}ZC5H!;Xr#$F<b@|mq@3iHt7zG$R
z$(@F~Oojv&pWp61@BbK}&x~8xOIqe@>!dt1o0xec6pHXNE55elM_60+GpznByZ!3B
zW+-NuCz>}!*7DT@ATL@r(G1usZ}@rO-hy*7FY5qgJa;7!XWVQ5joDCHm$z^FTIaIi
zW`WvzE_$UhUu`!*Y%1na4;5&NP7U^1Y?}+2n6k6H$^=xOKx`_(!aKVjF=)UIe4QNz
z9{I|ifTmcf_dN6E@bsKTk!Uf*eDafC4*fMm>atUd`h`{MWYW%e`B0I@890i*JvS)*
z8qf?aeE3?%xi2H-Sh#I>9CC@KJ{6wNt?X;dxvxGk?AthQhAj<@$bT-t=xGTpu73Bo
z{G*prH?yBUKkPs)0rl_K82jYOI|iyqBsAJg_o0#EIKk^eW})r}3UVmJ-s2t{GszE6
z*iNi4IsG1KS!mRdGRV{)oMq$X75c>Io+9QvNinYeZ{F#Mn_ug~d|@;km0jUBKLXnW
zMl)#veW$Mer}p1ZIloRZJ{7_DTOKZ`wwk9~x{FrsM7WiCT<0<G!_7D3U^cm)r+m`6
zsL!F|!|rLl|0v~YSe>Oz74uz(FJ+<s4zk|$Tatp|R<p(}C;#+==R$I5jYBQ<<kPI5
zO|5+yqWBjQaYv?BI>FTA^eKtJ#sQn;p#S}2_hqx9$Lo!f4g=OZi`Btllo@y;0s2fg
zV0x=7uNh0Pv<=fymTeV8N&lMpYZ=!pzm(+Zbegwatd)2%T60!*7V_XoLc5ZbFL+*|
zlhl7HvL=PTu7do`XyS(54A{Ic5}FyO(LP7A-r@gpY2<_@#@EaO2ondOkb8*1L||e0
z`34^y$swm%)^)Bm_N#nac#Bp8Ry=Lo$(zdn`opfcFLPpI;ys}M0ILAy-oDAg;x*i{
zi2gBqkc-;sqrNiPUHVjN5{*AA93X0ZkJN8{9b-jpaYX`eCbBTO&8yi>9JaY(olx9$
zKP-+Z&db94{^(%L$+=ULo7Q?CUMjw>J!g!&$!79}VLyDe8f@wQb)bS3q#kPdVrnpJ
z<T%6FjUz)yI4Thdy=nYIlm_Auy;a@Q?|J)bjS_R{i+k*{kSH+9-CQxHS>*m_)7E|t
z>u`mNXBg|EVU&4!ba*h1$1u`O$Q=4wZYt}~-X8?hp0nM36+jjbQ9JzoLU7>J%(hIR
zqy{!Lyl~*UqQDs={^~(h%T|Ka@O#aqlz6VJS8wc2)76_u=E@sK*W1k{0K$v%-bm+l
zHI!?`;I5JT`&?0q6lox&?U*O{5%Q)_?(gLomt)D%{YY;60;_-lzf}L{1EvYA!9!7g
zsI%PaMgTQjC0$?bp26MCR^oKcR)x$W*)n^|S`2aAR@oojlqU`AuK?L!9sbtK7RwQM
zj9@(!5TRzoX;e^ToQPHZ0B=UNPFqAsBu1iyEEuI@_G;`~-IG5#!UR?Y?p#cCjmg+y
z>6S6o|Cmz%_+^J2G?jYsh=Ldf^Fv+2FL+?S|BV@2@4RR*MbN-B^7tz0(S-pCwm`D4
zlZQKxNYGmS9u+LvqdZ*d<eNA=d!w72s3NsW=ROo9a$;igP+<6cLATU&>m|!hLdN4~
z_lY7aynWki%z-lmqGo-p{>-P&i3uOheR%f3BO#RpBVUo{A7zDGn^I*Z!?Dk;H?5vG
zdC$*(cgP6-zF{+nB{<zH4>X&+Tx8X$cJtX#&ir#<)fm1UD1pVb!Y(6f!@<API&oB5
z+~g@_PSojou%?N_3#FId15zy|<mrM^R<A434FZ#S#nlZgZlmen0@Q-uK2;5<720bU
zB;0$_VlvaAgn?hX<e)p0ua;jq1a3m+FAvr_`HiT9e@%#g411w8FT+w7sGr4M`@{WF
zPnnSNH9e%xEa)YBoant44T*?C4ks02+uB4yl&Glf#o{1AZ=!*MfQvPApVV?e9iy}L
zpnwG$-g5mBI+eS(X??6Wgl??s&hJ^NXxhaGb$rpLsG4`B==ZHn-wlqYOe{}G3xB|4
zIP%*09cko{)NKp%<%aD|$c^{m)3T)@n}+M(BExDAW4H)?2dAj@Dibw7f{TZ{%u|`)
z%FDlkazvXpU7?mugn3&NvX1}0JZtc157k!wl=7B&K)h3yp)xHbMOQjs-Vn$&MM$vP
z6?8jfgsbjHKRe0{qMLt`>rVitD>LTdUv2)LR2@BHfz5mh1XwDa?}~`rRgBxcq?qf4
z{R-SX>q1vWx{Q9$=IAjN&;SiK&pfXg2@Nqnzr;q8P9@dvJ4wj_IZ$Jz{fX}z^I;OC
zgh@DLpF_hVtSczxRaS6X$BEgusgWelEpQ{20>9hIx@QvFP?SCwmEOyssWm6zcD_})
zZi3GL(>#c{DLBcqKkg66BpQ(FYp>dL3Bl=c%|Q#$u!Q_|D%?9`Gxgn)w%j$|4r7|J
z+gB;IHFpE`YYTOvssFx0HPqTd!iwQ9o=%8md_H#xwkXnWo)h<c0n#GB2x&b9P&^ze
z&rlLZbd0u6?}t?B&Q14dAw!{`uDLt@^8VJsA>ha9i^RNcFt<vn=x)Eq+kH)!5UHUj
z986Mxx831bV~Ir}Yd=TnEy|)_{%y{^LY#S)x1cZFk^=7DWnb5p5+C08@)$hbT#>j}
z{$n`&vF2kP4T&P$N(k({!O}fR#D#*lPw!K-qc?0mjis8j?`Gv^8*9#LzT+aQiMHLA
z(VLn2)Z<<*z-B#g2~<#)bJaV0Vee9Sw5ZjH#hm2S#b^UraC+Xy=jyyvU;ovX%YmP0
zD9<IN|9w+XR;F>%LgS8se2mU%n%Cv#(k^hwp}IL0d#GpO>4=bPGXlFkY0QwERlDw4
zs7(4I-*}Q0tz>E5L}PicD4*sT^yV6Qv-QROWWJl%5zdOfllMy1RuS^OA>TWMr#EnP
zc4z&=)Nf@>ij_o5(Ug$!)&PFo<(43n?&HPK)l@+X?5W<hC2*COWMKAKY2aZZL$%oG
zx%TU|*d;y#@b#iSQ~j=Qhg?YY78dz=4UZyfl>9o?DLH{4xJr@zHLw}3jo);6P7UKZ
za^@Zo4SnA`PMxIa<UQ3VgkUbWpALUA&PtWgFbZa-G&DB>R_2d8)}<u3q_M^pLw&{_
zf#oA#dWTgEh#QU6MfJcu>f2qUoYEck{qFvP5ieEM0s=3N$|*zuRgGSY2(yh6RP~NF
zbR6^GaCtJ<7f{wrmI@KP9r51~5KO*7BG;-Th+;Eu#k9l<$<ir6M~(Mz9@Q@*)b0dZ
z=qlrh5j1mw;*AY)tUVGQ48Y8hzm)%1v(5RC?Ds|7;;C;qh(h?&X~oL@xZ7pPL1+uG
zx-8#v4CBP+fiJvWmPia?4jjnTFd_LB-CC5K!Wx^+v~=qXl=Ic!b<H-BHO<0cpR37Z
zS<wCp?;mNw`_u!o4}8MMOv(hmev*FM%Q%uDv-A44xceJEy?<B6mieOvW`-(FmSfUC
zvm38oa2n8Iel(zP+t4%y@tfu}5c(oJp+%+xV87sY71@JgGGZi~<q6o-(9P9Pn%DQL
z7LdYiHTv-Vq}St80bL6x5lSbhfC#VC;Z1`tx`1-pux(1j@V0|v>cfti4}YWx*S51=
zX)uz*Evb>~1=LHHH&Ji-)9%X>?V~N-0+b|IIH=EqcJw`icbo3isDt`2%FMb-jS_|(
zuCnx8dXjNHN`;!_QT1u2%dLc%d>O(-wm>$Na7|XSrAP<3CDw|Q1~KLO)%uo>Vf8=P
z?!8u7iQfu)xziQLH7`^cIf`R<c&h)mEFoPZM6NjfweI_{5lrTR<NL$0UvmM%N#1?a
zFdye*b^n7|4o_Z*oS*IGr>eA>{E1J$xeS{NHP`mri$>zA$Kf{$vDsa^vxxa7i~GX_
zpX;xWl{cQ1jWAkc3z4s%LdHHheopRCX+Cr_{q#&Z8RSLdYWMuQ3{_jM378I-kfz45
zd>l2^G66)_7KVb&RxN06k8Z9N)7SL(mBJL4kcFq`BvL5$4|P5t)E5u)XbAnZB4<-f
z<**BtRzlZc={{rwx5`*P)#zCVbt2oEeT@9CQ6K4k1$OXdK<D6!{xAAD07}H+_Y^u!
zI_$c*mf^Uko^f(Va;a_@;YiVg2>B(?A3Pd7uUL#tOl0o}3zOin8s8yrQK1t4G$p;~
z#S1PdzX|Pu;5#~bYv!TZkqH47W7~TnGhbehxb}A>KrU~Z#g(#eSMuwai}Pqd30~H|
z{r#prrs3X^R(UUDv{8uJzLB0)lV`B;yUGX0l0ArK*-%@g`YvN%tZ=A<dz&{XA@di?
zbwhx_`CsuoW-HxmQN}(@2cy7kbtd@8Oba&N#R{_X64RaC)>IF9ZuGF!6*UgsDOwaA
zhP;l0DdLu0{`b$5mDfOu{Y>qFr!BzegLtuv{xvT4S(vj7VPTIOG=9gjt4}TYG7zG2
z)3hCrQTy~(?fLxp!SR)QIay+m@1aV+JZOYA>F;@_b;}?4JvLgt)L43^wu+CM-(6U%
zOTpcM1L_xVl@3QpSlD=$Z#9?G<433rOc3tJEj4_z=f~*#$yy)YRsHgc&1aF?QKt*(
z_J(F_T_o9q>dU=~Y`i6rKVx1!>g<3>x43M^rFgB6lo9-Qxv-OB`LoMi!b|E@8GG+1
zo4MEUh6!^nA)6uANk4LPEecLYCiX|B$XAGtpX<j(t6D=+XUPKNoj*O+U$v;GDA)VJ
zA5jpg`U^f`bJ&0wd6}h<YPX1(xJdaIL4Mr+D~+QH`~uKKhDT2^dhb6h;`>%)E83g3
z?}vxNHiOt32B9>FdHqCLFP;hL?3Qlr>p$iV=ATaw@^vaMUu*9GHH_x(z91-ZLnq$9
zg8hUyL3wmj3~elnB+M$iG2o3`Lc9*2lk^zlifp?1Yv0W_jpaP<ETfvt7PiE5OPEB$
z&Pm9Nr?^x>{4ACGr`p^n$my=!2}>0FWkCo-rR-N0g$Z{ylSoHzv@tgr=L9e1L{WTn
zu`oM*$$cJqIVDp#Q3v~Bwi30Cdd}H-UxewSTHGbx%P$kk>-0&6?j*}3QkmHpKxNb_
zquP6$x^jOnI0&qCk?t#Of!CK_$*cRAd;og-A7PAUFG3#l9Vs)g@sJ2VT>b;N{WBOM
zkrkXb&&)eOr&AWVTx_^TUp3YgA(iYAEB~c224j}fbf+Ucsg=-Z_<s~ISdl=n%i$0w
z?N#zcSK%(hazsNj2i~G3n``d55COV}Kd-vaPF8P1#$_<C9j_H%ot&6xuW3a|@*84j
z_(ifvsN_y@l;^`ip|#)~4xyE9j^n$9fO~dYK!r32hY@)z7&83B5k?wxTaO0bqbx}c
zTN0Wh10(w>7~&$7WHf20ycuj6hK%8~oB+QgP}FV~GUrPoJ4I?ku3&^)?}zJXi$%6`
z2i!$8D@hKNv12VBFM^gbU*Y%{Ez3-0ESyD?Et4R#s9Ub8Ck5dqMXybbH-AaK9uRRK
zcN~dy$Y{&BG%_a|%~e~XW>~8g?DBTRVDW(l=q9;ZPaETW+t+>MO}$O~k%T~*GsW|c
zNX4@?7j5PAM;W*q<z*%(Syf%K+5wUXvoEJC%y&P?(<G#w^DNs(g!8-Q^f`Zi1$>5q
z7tkF9tnE{iu4bR68ehKg@>Y&$-o~Y$R-PvKW9vce*WjVt31#<(vJG{3BCASE0Y%f(
zZgdCjvlB19z5zc4-8;36G^meDU||T;wEARPf1H`LdH^H_fUr9HjoyTyMVIp1Yjl>3
z@^^9~nA#Z6+B;sB_KlJNGj3%~2^FKFv2wBE<FQ^gMtSz?dQ*Xm_pvC=@Im29X7KOU
zu8!SOH9!g&n$6eC4nW4`thhBzY-|<9ZO?s_3pL#`D=wnEh5wf^aMc;G<pZHl=NTS6
zeV)V^6*AM1o)w`t5h#``Nj)8Umzh)0%Ty=4AH&oN8P^5ya`R_h_0i>fD;FzmL+Ab@
zzFvo|Q3X~CR<e4ypky@936Lf$&HzNV|6kImAK8v3_kT=_esuAdlGy;el{}Hwe!=Qu
z2Hs|sI@v_7Y%8_&xE4(ZvrZ6q4$!YJHe2m8^hb(_7=L3DtrpwbP9IB<7|HH6i!b}x
z{&u%UotrB)8sDcs+BfP|wQ&0{6OQ|h1mzlB^K|sWZ9Z&8V)~zUI5q{@%DEX(@ARI~
z?O!#v;VtqgN=0%a8bRc_rN%e<GLQ0zQ6UrG?0V?+pmSumXqV*y5l`{qTM97PIv?=V
z{KJs7HGp~|2=BI1ECRo&3woappR@yRsf*Xddr<DCBG%I0<toVHOuaJE!dH6Pi~N>1
zp4Z{J5>U>a!E?JTeD}aLl*;Qlg98HDAGrJ<DKX(^MbdPGGt(!mBUM7sRL=14#@O2r
z3Ed5A?TKl%^+ZQHI`joqs?y%dg?|!yx==00+WwaFL%_Je2C?PGO;^Pv1|&G(*R$_G
z<^@^UPx%kl`)_E~yPGEqIevXuh-LZt{Vzh?a&I|FEo`qF@mT&%;r%9~w+@oS;x%>V
ziuc_)MvIaZi)Ih1e%eSR4_En`W8`gPqYb5_MV72pFCT9gje!CiJzJElCyu2--7vE_
z3rYN@;oIv%T`&_}U_9^zVBeo)#6fM+FRAHe9_brKK}d~U9zAfZ4O6NBapJ^mSrYfO
zN9+Su&_deyZnvyeDMUIy2CTU7iEzmgYb*H<V~vH7hrZ;Fpp({!<6x6!3IFLNT#UD|
z+ZntQ*d|KemFAm(=tHd>BVn+H<~m{W&xREZ|AOvdW_c*tu^^Q?*d*x~mj7xZtBz*!
zrLAthpz@4bk-$y4>CQ$vfVB9#@lm`j`-+i(e)VI0B+w|0NLTChl?wc(Rs2P4m%Eyx
zvrwFmqkG92gSb_uClu?|_zrZZVs8Hq0RUYQg13Ewbj5-_npYTphI8DzG3n8izHvAP
zv}Wa=o_HfhPTx3w{i9Df(ePysD3?Wn`SRKY4O%v1!*9Qv^VR(@WQM7YoFqP17<=lv
z;y%*+Y2jh#o5eDa=m`i4mJ3iQHt7tzRn`1t*}Fp)_T;<lyH8^oE1o$guLpZ>;{!T>
zEcWDFpC|IkNlwpXkC#PT+0D=KWx1-QTX<_(YC|DnY0djV;Z))NA$5?d$BmMgq^9`+
zq3A85!5x*^-;5JYlSr~;fx~o-H`r;)wAB1{g`n=tua6+pmXNi$0xQG7nGDTv51WkK
z08rgEA5S&kO+A6PJTbo37O&hlD&(BL-Z!>;(D><PwFubQ01S;<7uRT!J+8B*NQTtA
zIe5C?tjJgDsNaXSPM4#&&0qAmqDMu6$9--EMp}pu?8cXi_J+Xs@d~E+QLA>dZH!I=
z>S?QGSo@Irq+GJyFtQ$HusCTsw(#;r-bQ9i9uUly?rc59yH=WYmG5b^P4zo|^|5yG
za$pw67Irl@sW<%t#UQ0ggo~8RE5G~1X-9mxLp(61jJSSk%S*}_q}-Y%0Q6VD?g!uu
z3pJtnDhQM8c3X>qL*6Cm9*YYUDpkmr5@!U5kc>l}tU4I?bWz8;ZAOgn>TRbXT};06
zSbnTAm$9*N_Z*}1+Os;qwG}T+5Ed3z1-#S+b;`AYHWSyX^<EZ&RiKq<yH5Cju}ml2
zmE+&?P@~R(iAO)pM{uFBuNcA>WfZC75YJ~=#SYwX>&Y);IYis^Sf%5Xd4tnME7N?0
z=YRHFR^vnzL`6dv-H}-b6upikEqkr02fFYtGyfI$@C8#8?^p~`Mz2x1^ivPFlk(T=
zti6*)3XH5R<%lUXNf9G+M>%dezOF>;SB`TL6(XNae&d#J#i;c%y-}_aDgoapFiI%;
zwEd~Ze{sQEDqHqBmK&O%V*DX^$h^wSJcBgz9M|}1ZNM!f5c=eis48PuNSuNrL(K^D
zJ=GhJGYopzY#0<Z^tq)DMe1)f=f^~BCbEYme^c47I&YQ_%_m;IM_DT=d#|P;z?BBI
z@!{x6QE2ndGFAzZ71C7KL1AU}y#Qm8A<ony%8u=P398KqNrhc_KzI6DrKghg)yM->
zvPf3Hd;4!qfU-uB3-4pgIF?sOAcGEOt6LCvhB;yuM_p)SNVO(M0$a$~qX6)l6B2z@
z8{Jts1_ZfcoPdBaN7J$KZ@lBV6K`atO@MaGdclUpF}$dnW-B7ikeQ_d_fLQ=0MJ^k
z(i5?NSDW!>=<X_-t;(;nC)+uiE8hjScJfnY)a!maLQ>!F|BkV<tJ0Zclq(&OfN(#*
za|hf1@{XrxjCYqYDQTWP5|}tKbS4h{$XIi_mdxiq312X_Ne2=Z+UjVA@%w&T@rB#7
zeyBar)%8SlT}@D^%MgEP3`bdW2$r-aNVqncvXkgg6FtvE(=kiyn~fJg&zub+jY@M$
z#Y(89xjysjEZmp$>7IAg(t?Q2qOrdn6ccq=iZ7&;svb#Yx7Rr9K3_Ic%Yn=W(quj}
zI;=z)|MIyW_*xRPoxxwSidipGW#agQul_}r<nVmv3)k@4O>wX-Qv5~n`ot~S&R?P3
z>XtKMM&p%L1~_~vkF*+dBfAxqlTr=Uc*MhS!m3>ae+pdsI-c#towwqq$~nH%CRM42
zut|N9|NV1*9M9G+X;a&>Z_n$X_F+ACVwNYcwrumhj|<ZXQw_oV*225YGCDV}ePBB|
zE!4{M_OmiU`1`GVHP)AGAz6pjTQAjAk<wVeZDbtJ7T#u)y|;4&_{U;A;eP)C87Kix
zK1PnD1doRQj-S!6vaLl6bIA0r7oz(JWhLoAS5O#FJU_)<7OW2$JIPzYPHanKr9F}n
zhBRT)SmhmU?nqg}-?PO#)e%yBWp?!OS?hv15gN_Ijf<8}(aTu7S#|x7rwe|Q3OmOZ
zvUVeAt;sa*(b#vF_LLtG92hnZ2m=c0fR>%NPdtNE70|GjS>mSp#IOP44S1O%B*i-W
zOf_(~-$HelBc4+)^C6T)LaX`Jy8PA=k2}u+r{)IVrrPl%!Z9F!U$02foe<ep{b(Jh
zbXiF+T_qi($>Rkh9BP1EgZH(Ncg#PHEfPPHx9c1~^kNND;0orqA0#eX)I7Edd%c&|
zT-kPN_#k(@hQRX0@-ODchf!F>U7a*pC?;h*OKh#qGLbcNGV->$e7oqoNBNOMQnwmQ
z(mzL`Z6%oM)IXu{X<~GDm8T4ZFUY^8gq<;p=Yq51H=~>(2qs>9ORH%*k(2JzxmZWy
zD7c$8kz(X`D|2LJS-mskTt|_q^tD|YZ|WY|qR}`=%5bjdhAavo`~Xd>z0`jyv^;r&
z3RCIvSz#S1ZYuU`?wl0(dZn)1w@Om(v~Ma~%Bwe$z?5r_s`u;@Fz)^Q#+!GOHBffI
zsFA=2P?WSQDx<UzB=gYs<=k=bYH^-p2_J;lx7Oy49p1KR&0PLCn6Yis5A~JAG$J~v
zwa+Gt;#G@aly&TaK%jKq4Xh+!l4U-|5}^S=ZW<ZzZ<|B|=mE*`SQW39B%FrV6WVC)
zfwWYN+H@CGfw`BLeicx<bJ@{+$O$ZZs4+^s(Lko3XAD4Vp}U4oUlbrr70KfPbOI{e
zG<(Oi0`*HGLBd{n0TJ@m`uQKmi7G#OpWBSZ$|_FuGc^)2q<3*BEqNw8)d&hN&)MGR
zf5S_S1NN^I*d&dIoBSY0!=Zi(Kzd*j%@L)CH)*%7Va=gjcS!inh*;YLadq*`e2*o*
z<$TL8N~+y!R+kA**`k$0wlh*PPea`t2ZA@uk42^)J~lM}jtiCay2F*C^+j@o2|t}L
zFJ8`OjGHazbx(}<r_SD2aSl`il`{iRC)sx~&^iWS>Ij}P3J$XesukEb-w4K=stJ@^
zmFWy{N#^X4eDym&&-coOwX3s09S@ZTUVX9ngHP!HWjUlH>wN1iE^{SoUw+u<`Xh;P
zt<0)(_tuX^4r~`b4JAxd^!P?YCj;f$FdZgTL_~a-HL-H7(=Ee^N6@yFE$i^=PYRmu
zd8n5ibqrb8o3<(!N#H$Iv0r{#wCb0=SIJ;PNTJOZgt!8MP0ya?z5&vWfmkwcAHK<J
zR)5LR6lN3}r-y>qEM&Db$x&mc(h|`ZVI995*C7gw>cZhC*d?9waOSL9+uWo}7I1Iz
z>_+mm;RCq0&Is$RO#q=NlrxehJ`^Qy)|ZR6<}c*B$q;GRQcAPe<<5BcB$_)*<}pna
zzs%bS?)^<u>Ue0x(L44{LoyEMxO*(QwIndkP!0vP7`e~%^=~Q@t9)J7bAlWmszL7r
zqR%~v;nWnrJ(pGu;Zal>Cm30cTQ<xLu9?M~3x}#_8{eDy#*WAt=<%tH6VJYw<Gtiu
zo<>v`W&PIu@q}<Wjt%hlJdjdZ|0+K!+_&UEP?a1MV0J4~SDkCX>eY?K5^9serR=|@
zvxF1jq2BB#?c`O)j4`r~(n}r<FHMODcM#R5OfSV-H^zIj+-(@huJqFO3ieJB+&*lq
z>O;K5H^jG3%Ku>w)uP0xcB};{#*+?;jP16h%|QDi0I`B#?1FxKMDQ>$O@Dsr#jWRx
z<(bV+uS?OzNnoMPey3%*whGUE`)CPGxn=H^7TZ(vq&YzoI$QrKYE7;K_{j0hHa~ZG
zTPdb1SzA=`4<{9eIx>%#2i7RGG(wW3Y>o#Yevq-bQDPXUi(z1anh>O!2>FUNmu`M@
zPMjtHxnI865e%Y6&oX$m6`*yVTjS{jE7r{yZ&7b8ex&%U4Fz+R5jGM6V9m=EPqO=?
z`b;42N=u%fdAl-eyZ6i}Z{enZ_tiFWt2BBa(^i|T#DzK|c@ek4F5Ebw{ky?|xrKCp
z(Rjn9fpWIsukapk>hLDP&7X`}XO+)4GTjybr>ZNU>*|3|%&EmbP@}y&!LBC`RbD9S
zKhhfpdTAC!jC=)<_ei$BQo&^$14~|6uw4WaHw;4GHQ@VR@;4`)7gkl1#PBl5v3oVw
z%YXPl=bd!PpmYhh9kAq{Pkb_fmU)@g8@(C3pVLIv)!>?C>uLhK$h=<2x2bhD9JdP)
zS`IVlT0Pe{qyE2tQ2TpVnt6=qfg5B*Dv#z|@v#5As*x5ApT$EU#T+WeBi6R|@MZM3
zxpyl>_86b-yz%*mxyEaLXo}^HqU^VdmjC_Jbm4^!$CF3HHoB0Kh_Rx@WT1TbnUnG2
z+SAuXmxI=&yoMSA5hLmM%oSVu8i97`WYm$7uQJd$SY9F-TBZp7NDv$WGF!&;TMl6(
zr4XnjbnYT+TlbdPP$|g(2+=b18@zVp50p!05FEh#yQuvJP}WIYzyk?Y#M8;ZETRjJ
zGeX?@bZoma!xgM$?9Q@ZM{|>P^3SD69V=e`$*DEtbnT&{_{|@d-WC8wwI4td>k)PE
zUozH!wDW)KOlAvm$4P%Uw40BaLxITR?LB^?(gFWE%I!nq`%-JpD^=HiUa1`GA@m(7
zc38cd(qmBeJLFr7DkhmXE+9uv`1zyYKDHGgmPVN+IVDl+@$~6(bVkj40DcqErQ$Yr
zobirf`NgqxPYB8Nz~D-rAu%f+Xd;`M#@)Lf+I_`D{CMWncEie{;uYJG*L}}mv)}r@
zW}h!44!v(EY!O*PcHR->rh46QsEz}D(;a@{Yr4EW_0VO+jq3+v3BL7Fqrk9sx|g)u
z4QrO#Y?o+}Th@a?B?@D`qsiV8X}Yg5^eBO*KsKP>7Sc~SZ29qqNdNV0g}zF&^!(pH
z8B^K3+~8Rc;fP25dNySU!_@BJW~AwM1;5#Ax1bPx9JeSWM)&@0ug5=GSziS4tXw1T
zv(K@T4W*`N?o1EU90Y9{zJQ31-)X-8lr-&Cvz1LA3)r^ca9V!<<)RH~hTYd1OGz)s
zg;|=ubSVLGkP{@Tag@Q2Y12E2Y$U35YoBB&s@qf*e{S2+F_yQF3gM(drm13?$ASh7
zG&Mjm(_|H3VHamiwZ#%jNs=GkmHS}W%V<r(6bb~_tz#CrjZ>c<pFv9H0SXCO3*hmr
zTpj!A`8J!D)Cl(dB{ud@n4}P;$(rHH9GY84_t#|%ZG(mPPHuJ=kH_a6!aWnj3Lrn7
zw$D7Yp^7HiZ(K5r7OdQw*1{F33f?D`ipw}j(Pot{kwc>$&bJ4rZ@R;3^YUXo`C_=5
zJC1n(vn`ZV<97~fxMao6q9vUZ5TAn~id8qHCo2+lHfDPI6KCl!-}n=?MLAo_-|z<Q
z>pHu@QXHR|)5bO|g;K6>fa<Zy|7e7GfFa|50wROJ&%0%0U*w}5_Vqi>^@%;kcc9d9
z0pZ8Ni^efp0eH{OCh$I9QQCyprq1;Cn_N?z{8z3oZLROnXyjo;qd-AYPL{iBci28$
zUrqAG23Z&Ob!&~XMHk$|X<N50;m{z(#$k^xL2vH@f(HwW&Mtu;a;>4+BA{+|rR<#6
z@V0EmOC@Oel}xOtSfX*@U(dF2=C%flJIyAMQ-CmaJMS@+X1FceSMKgtrn4wsExG+x
z3R#&J=}6N<lHf$Q%Ede+P@&Sc_UyEQHM6C}v4+o-do0dl$8$&KDsFDyB5XUQjyWwU
zFG;uX_A?vuZk<tZCjKG;6p7dRCGPo*b-^X`b>OyK@e-I?M^y8siV5~p<bbcwlDZ{J
zUDFQ*HTPM>&B!PhsBKmpi|Cb3c(zlu-p%b(i;>B1&|FUiQ_r6vza4a>|C=1^mLQoE
z8M2TDLtZUZIM!&v)*48#9j3_DMsxq2P5b84<yX4xD_x|M%=XRGrMV^woFGS)BBO@a
zk}y|gdAq<EXt5BV^0$m%ccosmv-^K;Il8S9P@f^o)gkqo#t)WDl)cv??%wu=%R*+B
z$8XKZY7!Qu?tM=mqA9k97dPLFji?3YiV_oOvxpjrjVivdQ%>nvSe2Hp8shPWyZK{K
zdRY)y?Xlq{NH8NLyV?7tjnvOOK+ze;$^0vIa=20Tb;?9Z{R1P=*Y>hRhGk5@rP{qy
z6P$S|gv-GId7${>tyvFugLJFq41y;?3|=GO>faE?ybLg|m`VA}s9uLUI;Jf6)}lyM
zwDgf2DOS9J8p9+1%dd#C{=8p|2X@x$#1hYRKp<vgpbL7_0#ObGbOD?#3lza33s+^d
zaBu}RI?(d&%4e)tP8|?CjA*7-CNK?Ux5H6ZM84lUybZcf-?|P!H?JC3<Cgt5Gmxu<
zgpyZ0+j(7gyB6iz47s_5uG=v~Y@fUFZ&P(ki*`@{o^aOJa%JJ-pIbO0F0615^2T{d
zsFQ*gR$GU}*(ZAv@9>A>98@~UMv-A^vR5~wH8dsF`M&HjZ$0RIGDkAe8yBAQ?G(Q9
zKy@!z{#u#RA|45~v+A*7QZOA3A3V;HE69bm^G50}1wN?*(wf)vT`vyjEf!;*-OH>l
zQx(1A99DC<+8?oxPy1O`L%*E$CxRAzC@GLd_uDgKD&^*F)+YRmwdvCEF>#Q~>LFZ#
zB%L3By==I&Humq_N#?yDUq|QRdKaVXIE@|H{b%dCOV+P8Gus9B-v64`87B0(KapX&
z038PNRsl!G-24V2u8lg8VmYNyj+=ls2-|0gGy$d=Ejx*~=wv$_-9zPgRv}|M1!`CD
z^h*IC+eAvWT1DM$umhs6PDSgkUcLIUiXo~kVfrS%sdWV~3msM&Q%)D?PNXJA%~SR+
z?}M5-#Lx-4Rwe5CzG^5@EGtekAuWpg?d9+PKQ}3g;cY?Adn@gPk|6Br5qITPT^V=p
zgcfC}XO_|?DUeK%_EsF2EvWm5Yf+2g4A+LHKtzCVWi7JySb^IQR(f?{dmad~;t1f*
zS_pK<%jwwakGX`T-EY6HC%+Ud$Wr5RWAgbm^uk_LeS+Y5Aq&x=K&=efZ3`?%psui;
zEZL|kgp(*-YkU4WPMU=Un#OZ^uoINw5#_D4LI}+3H1v*op98AASRn#v_(zr>gR}nV
zI)0@^(C+k)4s5K4y(G|!wG0xO4&9oi#-2xss&vR;?eCm07v3Xzx4b6&V0WAt@KXE!
zvutMy<@wlEk=eoU%cb5y)w`M5Hg~Z&XV15L#-Bq>5BGz6XaiYa(@V5@pC-q?Z#Ne|
zPs#kX3dh*W<6D;%utSC$F8g%~ZFO`z?8F&4lM_dqld_^3G3v6~71E^#&&*kNrOLG+
z=Ir|xnVG$+Q4R*soxid`I?b(x+29fU-<(*4-$}Lg?;}~N>G4Ss=oj3ZCz@p~=Hj#5
z%(dv7?sl@Z-^{WZbTXdOnRn^tn{sXJ4rC{c+j8%()5&jm&&`LHL;so+1PAWrkr8Q>
zIPT3};oFnQSXL5h6c`Xk;?ueRzeidKkcNh)URO)s9^|VrRYf2p4nf7i{j-8{SMM%H
zM<mO|-O@J_lMGy>PLwj>wjh1OgdEbBA?4R;@C%<L04HX6P$X&y^rpZMbM>t#xTPn3
znPuS=M*eVTbi*0(7$FTLTqkM&i6!fg;LiglQ$AOMVp7Ju6vd+MmdnB-+ueHlx!kC#
zJtYp8`qRsz%KX2k#m)MqlP)@dq)sPasYRBNp_t5lgz6rAdEC69Y29MNQG33_rEWG3
z#AztBE7+og#1LFrm%$W*urt`0Zf^&MTx3sVsWYK*nmRfrupiXw|N5?tXdOeX&-JTQ
zoJ?psi$)HTJ?w9dHI#`g<$P;@XLfz4u$DOOaf*?B@?Ty8Wrhy;@8%#Rw1_TKp@1$j
zYV68n0?ZWI1`hQCW;Is)0V%kGdIp|Y&NN|i#@9BdN!A!iZV00@fXqL8?{OievY)-f
zDBL(2<UC>C*C+he8PM>8#ES;eEq3XgoRsJpx(wCtu2bG|>WX=DGDUf0S_99%B7QOa
zCTyY@ir*EL4txrG1|lYzb&eO+^nGe8EJ(}zxzI!^Ks0}r`FW$%+#Z3ai9te;a+vjm
z%!Q1iHc#%(^C6^v14Wi=B{UTyWl;vKAFoCE<Qot6TdFT<svo?ms>WtsqRJJ;hAYw~
z){`)IlalX;?JXcg8)y;aB&NOY4zk~c!)dY<%I3#IZ7jpko%eGIAm4khbcE$D0p_DQ
zoqtmk)A3a~Wsykj(m(}db?+sHgntv{tnKltdoCHplf2n=3ROOC1-;Yt@Ci`vL>n|0
zeQpNUUj>!hj6>}<%cV$P)B=*~w}Bqy9HYtXyWoOuxmpNB8Z$NBf@9WYJ(43xc<*}^
zxesaageAe3tfLWL05vgme%A=DU#qW70|E1?%Rqn+?pXRrcfj{Ibuu_*lWhzdJQkiy
z^J+O|Ap<6Et_56=T)PXY2{|^{m-Vf3lRCBd00(1q0~a`Xg|xe!J$ml}VFLSo3wwPb
zYiHZANQf98@TLN&lE4q=NDa+30Nl4;C8;lfs7}#Kt(5|TO~X$FHxcoItA`a6PSVY3
z%S1f{37}jM6}Yg2Kr-E$?W677@gp7m{_PWhV5>wL1(LXpe)EV>XM6QafMgC%0D$lt
zgE$s7_{z@tuUV4ZzWvja{<Zu@;cS78I2U8dV{uK2q$*4bq*7iHE3f^HbrHr4p4&R_
z(0uy{gfD8`9s;{e9xKIn;%vs6oVB-jlGmQ=MG({H@-~fc2cWCB%p*mX4~*#3-~Zw<
zFth~YQ~h;(HUYhM%C~yw(7nka(=j77?L#3)Bt&hM0+A5E^1=T_;OfaSZ>nYLgR7cr
zE1C$>9X~d=2b9pDw%5?E(T7pP(hI2D2hk<zPm@tLgO}@>oW5D^CK32ogmzh8Z7adk
z$Kn(}I$87Y2xp7|`cY;~N{vVYK$qS(AJ3ikVkA#^k5{FUy~iu~aomFKfcqp(k5v~9
za(%~KUfj$~)?ow|c<*9)4+YIP+PaHvTA-kXV@>pr`3_FLyH*jiuwzsEx(Us-(YD)v
zBnauNPqy=WAVl{IZG?!TB=29-3=41B|H{qI#&CkZg5*)%bDqVw4VAb~IO--OZtROs
z6I}Ij|3VaQB52}CKLM^R1OieW9X-pM2DTub?^#2OX^zw@7T25jOY${^)DS`vqxz|5
z2!fb58O3|Xfip;G*~=hR;aroLa?l#36$*Fm&Us^OJcL|My?jYd3h-;dZQ;vwaB=qD
zJ9T9AYR#||34+rwLuJ*fbc+~XY}<njAYGwST1+qLIxF}?@!<hbb$bT;Pq`pu{@SZo
z!)nWFY-8KVNh8Qh(^*fMu&Xr#;zDW7?{c&BwvOpMvi<g_(>+Y!W#^3U(}DU9AhZ^n
zZ1MKAR$EQ2xe(W{^a6IA7136;p2SbaR|M@j>&BQN#8r~kfJa&+>{MSwe(GvbML8;N
z{kIxmvbXqo#k`@5we6YJX8r=TaWpw6-|1fa_J@5b(_iOm>x#WKZtmMASIEpVHmhLk
zt$(ygmZxY4lm&<XMP%?==u9tkPxF8SF=$2w;3&4LS>d`8CoV8m4@+KqEF5k#95r4u
zgT|#5-2qN4*6M@OuST~PdWCj`qa^|or91C!mJ@HF88sE>7h-f+gt{bfRj^vy6HeZf
zN=YD&1Td7;d^j@sbBx+QX@(#zh)_cz$*#-NN1~xyh<&;KU39&KOh`wpS5QyLk{0WB
zT2!hYk91^K2ptf>i#5x*0iCq8k%GEIQy}YPhA{pM(lpb1`PC$K+FKfwyV?@vz3+M6
zGWq;Fus%Zq<^UH52Hw^2fGQp+`)GzjpfSSO@x^D8-i}1PN%W%<lCOvK8LAr8molOY
zBD1Ye1m15GTuZbkwJOADYm3IZb4w>Ha3g_@<O5K#Q6p`@8^ld-vPML3TqiHfh$geO
zuWhy6Vi~OH^`QAS*T`797WM_8*G66jwR$c=9gc%?Jfm6+f`)EdrxVvWo8LAx-gwkL
z&vTz-R++bD706$uVx4y%%~!z&am_*V=NC&ll#$fWb#TyxGe)%g3T%y{Zd)pa?<?fw
z+c))5w!XH{Rq#yrlN95fzHPbdT|T{}XHS-oR-8}NNy?`73-bnN1l;Q|#yx_VPe@K>
z<c7BeXvDEDnqhhR?>c5@8u75$j-{QLxRh>wTyJ^$-%tJ{-SPLPhlKMvW8G(`+T#RK
z|9iyJMCM-UW}{ZHrl&SWpHf$YUd!c$;mSF?I69<~enR4ZOkaDXRpo4pivgRMJ^KS&
zm}`lXPPe^Jjb?gWw0WAfJ`r@xN8~1D2FrHE>HDE9*4&uWTBK9B>FTt*X+1hmjdbk2
zz9zK{&{)d)UGXH6AJr(<-w)}InojA_S%Ri;A6+FijBihD_hQye_H@T|vVp+Fvj-Y~
zk)GpHpP7&oZPmNQY>ndWvLuZo|A;jnLnf2_{*&B2Vh*A+5dd+qB4do%rueANLgfOs
zf1Rt{><H)_Vf9&+a2MazbmfRD6OVM*sY=I!#H`25P*Z`KU()qT2+z^C2R!0`7PY^t
z{?flfU}t@HoB$g``WL<6N^jVQpz?470zk0p$F{m}H)X@%*X}xCUOc{xm1{D=d41n}
z7CY@<E(!90<W5N1yrc4%o9VR~=N|0gIZn;|M5^f$<3W1}u{S2u8(U=EXo<BDiSGto
zue;w}Nq-X?t{dVNoTLEs_5RD})?-c+D22%E>q@kgSgSW0!BSdi%4Z~iXLYVEuf|0<
zT-x3>Q=1l!M;FH9TYue@TBG^Dn-z`E+=K_BCG`yVd7n`#-v9Bto5?m4(vL9lYJ6-I
z+5|rIh*8q1{QadEX|^Wm3RZXz@N*LR`6zNbg%n5cF*?hduC1ACN=(v1K1`?h5n1qr
z;MIqTkCZEdw2$p8&h)k+BxNmuGmv!5PZ=k$4P*tQ_9Nc=a^DmKjL7CH$WLz;wM*v4
zCIlvmzFe*!QG0J+kogNoVu@?js_Lj;m!zPN2Zxo86%Abat%ucFVnR${DZt;61>B{9
zz$2CZ?TI+Gls&VMK6KwJ*TjhTL%pIr+Pohr{IfjX2^gxAUpv~@6cvZl^5)*m1$?TI
zC#V!qEm*{u2VC=&jF<C!$92k6Rwhb<bj@?FjVPx=G(XWbr9W){U2_g<sCY<wo5)Bm
zJ=E`SFGp`ufz!H_rl-1P`gb!{DG~7<#5C5q^Kqavr2WB*W7%2do$ZdE49yT~G(KN9
z@qNTiFarp;765qU*NZnPu2n#nP8RUH8NEGCbzM2hh+!FC=x!qJaBrg>oF3A}Ju7#y
z4`XjEkhvoR-DDa_<GC|)ZP9~Yb?=a5+%x!Tya^W@qTpxZ(eJ=aIP%}gD6qR#WP&(q
z+W%OOl4gd!08n3!19W71*|wkF{G<7OktLpMT`1B4Zbi0gL>t>)7;_3dAnf2`rLoQ1
zQ|XsMFg9yGwK%XV%kF8${wdnSEtBOn1(NO+HD%B8zkf`$%Ih&3#vKY%^8$bL{hqBQ
zCO1S-^s*fS223>Nx`J00ii?JFDV9%woW^w^U2m3m9GD*a|0Mw7db$o-$sm@OK<V%0
z+zVoevh~)K?3-(L70)~5{(`>VJMa+5C$(L^HIlWC?$4pcR+i;#p${a!J#LQY8j{8l
zawxD=!l}5i<bJoV2Ek{LFPMIVdf2oF>p$fHCTy$wfJ5I8tS0m~R`K2k;5MinOnOo?
zR?`hH<X+Ot2F`4^a`JSm?hG0$h|yMB&V6XR(0A+ti#SpyE<EKtg0U#Ay&B63=6xtE
zhv7t$GcE(e^~_91_mQjuRq{#p7pEg7VwFi$2(gNB7u(I-M1ikTL|j{sMZUvIIwdGl
z>sbI~Bia}SHC@)*&1y#B_(3%l3boT=wn18C)E{8)Vf)an%eFyz#@#2_Evk+ekU(h+
zq#d;bTuBn8rZY-;=@5e7Okt0dDrTy-u*0c7x7-8}ESu!Sw2saiQDo+R@k|$T-%dSg
zTrd+Dz@K+X;s&<0jz0y^DKY~Q97O}2QTCo+B<}Dn&pqUc8gf^yX70}ku*l5XW0UFT
zCsJuhFh~Um!UM_^yPnreE`@Z)ghJwO_4m&@(mA|OmeT5cH$lwlXKl4Bgv-tEF6^^(
z>gQ&hGHLxGce^E2A;gokr$5|#-G&fX-PlNUp#vtkOioLiBx!x%6GxAI5@@o0rztV)
z{}+J~yaRn-O}r>}BG8$bHY{Osl!G^WwF^Feo<Iz!`}Os*Iu)8U_UP=tfT9h!X0O{7
zyzajFLieJpm?g<^Q*UlNARM5ly|OuJjzJKeLcT(S=odo=HjI5{rKK{1br<dDk*0c9
z$SY)q3Vma@04&j>=c+7%*&1zZGgj|%!daR~yp-7kTXPMLvj7IKCM2}dQq?j%glZsA
z0I+4xm`o4?)3=Kiz?hPMOW$OjZssC!G<%f#3!XQfk1~*lfP~q9Dx8Ni5v+hRT<XnP
zlDn^M3ji{fPT@jI4-~Hf9Al_=(m9~BxDg5W>(a##b(#+#V`$&{Pl3+{M-K~@EbHG(
zRUO^#`gp9%gb^f?)JcI_V`Ii9M2<L5P5RE|vkx%2HOFNxAIM@%(fpi>In`5LLp;jy
zo7ynDyS3iY57c0I!6tq22T7%6^3wUI!HIG7`KG=C&#Qob><IGHl^+PwbFhmu`jI#-
zLtn+k23k?WmoyD9-()*UUdvEYlv=dOtZlM<9>A5*i~eolo;jMb!_Kc@tE7bjN+}qF
z&w!HOodK>AxppJLo0nL=vX;&V<N_TB)RkwwY-_-g`5rsYAYTB|10^#%bQzIma?z$v
zw&q{Ye1^bPMm!C;FUJ=WzZ)<}p@V?_o8s98$AbjyA+HS!;J2hiDw^a>b0e%aCrFim
z2MF3b5jq|kjaf6?hG!W@nmB)1yb@dr+BEJGwdRExEG*M3|AZYzb5ZWZn(QSdC(3_l
zTiuAP|7lKZ_tZGlN9V*RUb^Jl6w;prrs<EoNm6Y``H8^DpCkW0lftM0D?0}Tc6ISJ
zs$arM-WnNi_yd6h-LwPnZJq_)el=27^n!Zn3b{kBBOM+QQl_U)2Io1pl;}e(xf&32
zk6_C))$K&~A7cKE<J$a_epubHHl#FA48R63jYeJz6`9{qyBG!L>`1q8;5GmSF()Ki
zINYM0{oBCh;#>khAyTfT=L)Jnw-jDo#o+A?5-Yjx03VVeM!uZ(b*&t7!vIkg7hRMg
z)&0yg6HT+}+r|Z*RcrFXRaj@+0=f?R*8T=LR9@iq#<%J9jjyL{7gjD8?}bO4#?hIp
zlH;9dX8u86u?!%NuIo0_@c(ij5x{`sQz=%Mb(1Rc@X2RSvd^}IUUtFp)ho%F6BOqV
za#acqmJ{whox<VW3akx~1q22O8Y9@qRrv+a@EJI+$9D(5aW6vJ^frD`Y^?yv4^Zb?
zmC`o0^!U2H$1eif0pM_s<tN+$UQ5OO7IFZl(e6<8>_xI8o|kTg$vc*=u=M6xj+l_B
z$ce6Q-=V-KuXjn@ZA8W82I_z4!;g340aVWURXPjJdG>xHugG+jp_TuVyBTn`W`L?d
zO=YP5-h5?^S83CIix$njs$s@oHV@y3h6F0@Le)MIUbKsEsGiF&d7O}FEG;cE;cQoi
z5xOk8iF{~VO-(xA%d5Q5U4o~h&&$0VAB)S_w>7lM>`dd09pHKlVyW@@)KUz13}~Cv
zX9*>yhRBu5(JqFW2AmemqE-L@)VOxVu0K^D5dqv}j7n`~X1tiJ@V--#eUVe3yEheJ
z-FBld<{?d0fONxIUMnG)+%`KpQ+1?Wb2~6!I13477XmmM_$xrJO+KH^81O~_?9V_J
ziE%pjT5jG_#KqnBx=T7z8kcf5Rb3RxnV}f>o8Y-SOZs1@`wk9u6(U`J;4f_in&*f3
zEc9j0#(1Sy%5O*vskbvVeDcR@(~L@NOdG0sT;4e8vXgKw_!q{{xm6Lt9HxZrWOr9N
z80@13Rp8iOD9MLYgWp4{>epC(K_#4g(`P+gv$RGx`@x+Gd&?T}sVXm;rdE@p!fmBD
z4zo@iD3)6jhng?6!(s+QE^r06b?ocoelti~H;4(bzT*6zA>S8LE8oq=39c(qU&#%C
zX74B~k^17!owuI#$L9n<m+mJbEk${FQ>|5I_B)vSoNekh>kESv&7<!9KxS{lJhWf(
zU%rgK3eA2`aJ4c1VA0NfJMKtgfIdH<e(BloCn;axK|tiQb?GeLH-WuW;2@&89gf~~
z5*OS)G7Ee;9lHkj(#iG`teNIanet>cfJ<8e)M1LXx*3`}rI#G53}<^Y$xcKnBGwMi
z$mi(P^rb}Bvfc_);9j!JIfeuZ5+T<@q4}2HwzH*lW;Ta%12ev23-56WV}fe;ZM*F{
zA787|`C?Z9b9BeTO&RFPVtyEMJ?8^7kK0W(VE=3H<ScFi%~+0t=YoXbMJFpOV#!<O
zKD5DiBgge7$p$oOv}*HKX6T6EMpbduz$vf3^h;XxlD4WS$IWYl7r(yrcne*`gevDq
z3@0i?K99eh(|>)HWp+fAWsCIhC#C(hAV^x?<Fu-AaLtQl+I3aCKO&-y#THtPD|Rr?
zw-5a12L$Fb9gn^28LDGZ?ZzEu^fGXDr3uJ=nh#6Us11Txzf$|(KRZs5Zwq(_Dp@*C
zfLtt!b67p(6|@105dxa8qEjZp-0F7J;Mi)Ywo<RDz*N-KjH`W~==~{`3X2KYg`Ym$
z3Hh^g1@)v&6N`_C@v>b6F{}3W57x?0)^T(;M;+5SD>Ae^eCmWhD+Oo)HwH?17fu;t
zeiz9s_CNNLzm0H+1qH4qqkjWepsj*;MI$X$V23x`GMG?8WLwb^hvFgEn~>(auWio7
zDh<I2Vhi<FWKAW4Q}Wg+r{Z^Awm8;h5P~5QNKXtvSehAfrF(Q~>H9sYQs4f8Z<G|B
zNnodD0Jk;nzMw_OOdPI~1kcyvQBe7RES>dVQ;qw^eH4|H?vjm0K)ON32BW0~gei=Y
zR6?4EZpP>wGQvb8q&r8K!2|?!=+J=!CXRf*$It8g2k^@{=RS8_*ZZ=}p|@`C1+W!>
z6Fmgl+Na#u>WYs><G1vF(QPV@0(O(i(!ipqbUFwCHZZZ<qqj#+Ah`2rC7-=);6hFS
za#o=<o%AQm&z{vuVY1K#RwOtCB>!&$2taEwVQYXyf{i|AtfCIR43v&mG*6o~TmtEy
zIPXu1<fP0QVShmO=vJ%4(mPpFfJi09Xkq+@qI2A(8Y+v9;T$q!9}Qb}cMQaH?c8U9
zlQe!NrBXkB{cc44_|56R3(hB8?p0(1_nWwiCb<{(4_n#3BquU^pQ2(MnJ<OIz-4=C
z&9cf%Ij$q_g`({Wt5d@6E9B8Y@mQJowBqELg%H0ny<ds^lWr7i7jK9_=^%>G>gL}5
z=tu6ynMvwRgra4hg7{GRk02G<zGCB?ML!KCL#emOPZ0^D6FItlJeuEqzEq1m8~zPY
z3@UOP!hW53Xp*^}LC5fXqU>OcPZ2z(HWPBWMcp}=D~-Q4X69GSp?9%Eds2f8aOTIE
zm;I16#x+|!Q&F<ZqgR_Mj)Z&pG0S6#v2<8A^{otJb9Tml_5iik&lY<te2c4f^LboT
z-$u!g&?y-sUSgfEc}`m%se{Lk_<r3qCbB($y##r$l|Yp(Bj1?*^qVl0tY%5a7%Z8V
zlC6w;FG5_7QjEC%7J5KNh+(;2;#Socf<~h=RQ&1=IeR$PFLnwUMmc_ERR?VFFi(?p
zK8-1aAK+152<Iox1%8mb(0|MuROPk{Y>+5nYD-$ehY^;1+RXEL&Z5@YM;?dw#-Wg8
z1aq~tD5R7ymbZo>M1cq2*EDUw$n4Rj;17k576=jvnjX4{mVwp?hMhEhQJ~2IJnX<N
zc^WuW7n1kz1S);FIRMr+AVvYi%<)KRFc!W55WJ#Tz|EiEHISz%bN?)7+&GkI7Q?MB
zlZ9<2(1(_y`}dy37j6BGvEnEExt`!1$zhAQ6ia_>RLukE4UDoGFhbBMREcg-C(P<4
zGF+I55v;s;5_f3~ZT@Lndpa;I&Z`VaIWLVn=C;MKQ^Zu^zRd>a`xba^FH$52|6Q)=
z>npW-{}i7xG$);zo=Tmw9rX$iMr*U51c{{wOkojDQg+?`avj{-w)$N=jyg7hrKfQ>
zoC-4<L+Py-#M{`6RZYBar4B!`>B_GE??Qe;v<CHnOn|KJ4z0B!YhM!Woy0$e)(U&S
z!}x$;DsHQ{{fYR8gvdFK!i@)9-zF4Ou*ztUO=jgXr(pU?x%?oM%KG<*<$;C3akH?^
z_l((KG%mjA52e|Kg#JT@!!7x9U6PdI+~hC;dpy4l{&BySKJAhB{oy{SXz7?gQBe5<
zUQFg$Y3!(id0~u@Hl&<mXuD*FGq<vsB0E)}YwFQH#5pXMF5YhN>FuJB@K+mtkty3V
zrK@aZkEv|qEl%qR<n}jz7L(^{o}fsIKE=eK_tWcc#(wA~t&G&xuwV~dy29r#hRtZ&
z_nM~M{V2_b+c{S#y3bn80P=25>2J%6>rN+Z^Z}WRGSmBUc%6l_d|-G$A>gL?expPo
z%N1>?1NKny2zecS#Rjza+DSe{_ln(jirw<uMcoWAm|ymp#TU+2J*B41o%8WsV<E;c
zl(aVKkbTG_D06|m(=rTDYhq~hbCNy+J~UV{-UV11dYYlp`Ov==?%p#BK-Y;BjRMi^
zBvrIOAy|$>LV$il)xN!1Y~K>e$|CNO1`JJ}1_LNt*ECGJ$ODxR&2M$af@Pu=b*HD8
zIkbu6Z~AO8X0PE;r@smcQl9Zo2&9)4UCQo3L|oFf+FFdF+v=77N0YV#LWS*TZm}oI
zRDFZZu;YH8Gd?_}<sOU%^<O4=wdhp&0mvT{t2X6hp`?mX1&5kSjnl`D<nOMrrU(54
z&lj#(!1G`^4{PLn(&BXGnjF4h*hNINK#i9iRN>=?uO!>xp5eyFsw^*QR;)8b*e@co
zql1HS)NR0h$!Gu-XNy~}rt{}-mrL_^hUa&u8TwOW%lI2zmXjZ>=D8dI#$*^M9R%xu
z*6f*PZuZxPcKWaI6#R=#Fm5ECvmbu|*0WoU#DLDS3RUB5dUdkNzH6!ITlGaEU@1Zy
zP7D3HX}w~Kc`QqjC#Gl_fwr`_M5Qfj&r=qIeMD#bA@53~P7Iub{6*Lti|-Yxr=#u6
z$_dUW(|{$YZx@r8=Ze&@kYJhqWvF?-x*7wFsy1v&mDCR7e3emR@E(h(X3J^am|wc@
z7t7(7U?Zd)e;K;FV;Fd|$L?%_%E=rgxw-@zyz*nD_I?5V)QRk=spS^a&=H%$eBnm^
z!>1u<7}jgfl$@e7&f%D1yhHOD!$cf31}{v#v$(sC3^CG|XI{|u;cas8ChK`+8RbQu
zF}B6wBV+MiNTr{@t7BTGj)d$^bo8*H(tc^goZ<mvL0YBF!X-`C-x6+}ei)oKZ6)3O
zILw865GG|x=reHvn&PNa0n@NdypTB^>2@zG1-u2IJ^^u61DCVpXiN)uU~0P(M%76t
z<%}fGk6-fmjcpP2%gV|E@>eDzuBX6plOEo^c2D3t-j!(fh?DDYa$fa-)Xy(@zVRK@
zB05?&qdmAAMRTDr05hfx3>Yl#168-kSSkjGYF4;t+N9G=_Z4tbO1b`3=fMu;6={Ju
zm+B{<Q_h$&#zW#ZPG2Qab>yiI&QFEos~%<McfURJQA#w`K|Xk^m~~*8ATkIoel+yS
z*<`J$zOcEVN(NOLza{ZWq<H%UHox`13k=lY*PHmCLWL-ZXly^hiKFgVpmJlYJw8H_
z$*>8qiMi+Hb8VOLRwjq)&i~<j6DmFtZg?hUk}76<SSl^!Khq}Z#}2hmEPtVyQIyNo
z;_-lhFfR@WQ;L_8YK9L*imq+OZFIa^m=nHS)_uF7(*bon)7G)-?)8AOI20KFtQbav
z>md+!Q4hxhNSsR!hCL$2;!7JbFSAEds_%^)EN3`-E=840oG<K^P4fHSZC{m0+g`73
z;kW<#knpLBc9`+)pzhV*C0biSlCPClddFlnsJ&s#_wte%I|R0Qp7Y4sJf~<L1kblF
z2H3tAq;qmwA1l#N@?yWs>t`iAkAd~GPXWE4`GYO<`BZp7kcQYz{rvz**JL}xswXHV
zJaJ{OMl`~w$#qjWu=?d%l~>%MWjq^R&*<lGa*6a7cbjGW5@Mlf?pV`#CtZzmhv+jJ
zY4rGg;!MBW_qjIH#%6hOfpfGFKM{34N;%PnI~4R~4D0mccL0yJ;SRWtaW$jEBQWs?
zN}e2%tg{k_{PUz3EBCezfV%!&F&u$KxREZo`m8x~B45s!xnekn&x)TMnlA9{1A%%%
zX)kL8xh7>~!1FG!LM*M54t-I%<f~UeNJWGKo|W5BlWMO<POzA>5*jz{H~JSd-qEfj
zpVhr==3KmrkTdenSjAi#OX|%t3QK~*2SfU4M1Z^?kbqRIGw{e+Vs(U)VmPdJp0g^m
z(sxM{`+y6x3>heL`BMrjH*M|aKN;w|#u+_X!FKVrDGm+^9nVevy)Y+`rq(|TpD7L2
zm}~MnfGyFfV!!{HmA(@sy~zvRZ@v*PmG*>s<=PGLH}^}!S&p(rsI6h^to<uX3B}}t
z7rdgiH3421Ut}Udv*D~?i%+!g%cTao4V?(;919l1%wL!n(R3^VTgB*#JbqJ^?*oJY
zx5p6xePyWPQbS+X15x8^!*XnAXtto#G%L1Gx_8E3<#D~_)9kN)`=wmm|7P)^cBUJn
zu|AzZfOhraRkU{77;z}TF62CM!vDzI-@i>uCW_zsTjAHHP}a-%$S{$AbG~--^w6=A
z^~2z|NAhb57Pk1+O`+4cf^R7i0>4v!#&GOumAaqG$Efp3=oid8&<?f;vDm**?nrn}
zWXVlSiYYk}QwN3F<q;(R>CC*c+&6D}2TFV%xD_{h{?-bUW@t`V0XuZw0#ZPk44Hvh
zyB`|%m}1FjJ5HJ+p_(hISB}A-qRz&{&WY>D$bN$jVCDSpfJMo<heQWP7~ZyPc|&BF
z`~2&OXfdCQiLG^E3(3$3I3SVI(t~0%dbk&&3yzgO?#JjMNojMr5v7_yBuO7Pv5zZx
zKEB<fF9aC=>a>KC)WJ3HxyKUkf1W<PV~4l0ie{n$QqJi^8@_w>W2d~tIj9$&)-rNQ
zILw3yyEu=f6A`6?(BcZti>Htl@z5%I7Y=)ShgV1uU6v{V7Q==@0B#D$LF%dj-|BxC
ztbR;x19|b>W2_YAG4C`C-16Z&ZhF)~V}kHZbj16wfDmj(bj1er?qNRKO<&VJ{-<@d
zLc6rtEh+IPDaQKXM++V4`7R1<XlTa*UPXoGzA=dV+jn#P%~zto&@wuFueu=y*s{g@
z6H-{af_g=FHH8nOxc!Iw6PEe2neLA&PD>NsGe*2m|87SuT45aX`SwV~{<4f+ML<#^
zUqzRD=*vh2cd%_oL=5kZiiIc<|1RsXHlc(bthv|3gXUh|9zTEvS~CJnK3Sb-_x1zv
zp8*Ks_2-j-Sb>{eH)HdBic}ZCQn}#2{CddMWzl5Acu}I}GVtaC{J79LC^8xYaQIon
zr9+GoEIIHH3Zhpm!q>7zuy`N`EyzxI{dcGan?=yyaG}%!iEwo?k}E*Lr>m1s*v=;!
zg1a&1Xxa`#dM^`x^c0^<K5l4ZJt+n(QxHi@LP^vBJ4cw%cvGH2dy3k?$pTU5VgPC1
z4chJp&l36RnCC--PAGU)CuO=<cJEn&gCv=5U-{sabqU-Q=%r~G9lH$3?ZIzeyVKs=
z?{C(o4L%j5J5T@mo8vIJfg7(+fZf-k|8=PQi<BND|Bts-Wc~!(TJe(@3ehZwx6&dU
zM&=WLBwXOLg4}0I$|9%kC0#$)j$Y*3-JYM2KGK*<tZd*y8CoYhzQ((is;X4jO1qQ=
zhO_DAFI#4FT{?g~stG;1q8LU`c)%PY3k-}(T-K-`AL4fd7U<oORYl{kUbkw0N9bK^
zEqz+asuXG>+Bk8}G)fIuUt~Y!TH^O8lJ?Jj6(r(V-*zfuAOn`Vv2o0$C!H^uXV<18
zJ-zD6)t%<XQs!THyXxOIMZxD9Y0nBN(e#Bd&3qCfJ`lJmdnyZA(pIG)U1{gA6}l<E
zN(o<=!%-Kt87LNMKoNGtw&<E2i3p{9;wr#CN})kr)^@C9IMgjB_09tgmv%*Zhd!W%
zmfjRfmdl-LxrW<f9dgo9hRA&+4Tog_%Vfzz&**f8D({v&T6yDAA~0ZU=cWxNR6%1{
zsr>t`;@#5w2u4&=78NYhEay&2)KVhZE$OflQ=L{ZIATs3a6iW3jwK)&^Jr}aHSbuX
z&To_@Fx&ecu6&c=2I=bw#fT>#n=Ngs(?>-euoacn67Fx&g!2C`BrZ`h2ida8iQB1Y
z5Ws=F98ow&5_}Yc*$l<dSVBXfQ2qE&wRjAp6`3K(b;<b6QIlZ*@}Ni#@R-xDUq&L`
zA#%8GJ_x)l?w@GaZ-w3_FV>bRuf{FW6iK>b22@MP>;EorC~X?`Y&L=xNMbbp-8{X8
zV>;*Z=!z)6p*0apaOTYpX9g)9p{yt?bz<m{HP0Uhid0@%1o<I$3c8v&MrXJ~h}Pl%
zsp8~b;7D>4NL$fRrDyc@V*snilujpw3cVd#tM;&~6rCN0oaY#^y((@~<9fc_{3290
zB*kd>H7$0gUO~HMYs25!S9q6^Gr~Ky=95p`OUFHfWrLNC-(ILP?osbQ8tPo@U#vY+
zLGRhkB9;f+Z*(~PcY)@&kqOnTojY+E%^<yzF5{JW!QCeuM{)8me)Eq1D#4w?9_Kbj
zu@jL0!!-|$d`^grGJ2|~@?wS(cO98jRxo8^rHRDlgC+aLwoz9Eu-es6(R&}llZ+N`
zkTa~(NH79Lnkb=~=1r)k+W^+9AaCPqJRjXOg_1ojJ>~DWBlokoo1!y1VE3tDA`0+B
ze8cO)0W^HwA@Wrp5JmnHcbe?N%lf-U2ZQGZ+)XdBR9Lu7bKf$>wH1e^6{o6$Zt8<A
z^K)gHXBC~pBpuzCT~nSnt3iV5_+8TJGg1<*Hn?exJK#?xNl3A+^Z?To+j;HVeMQCS
zs+ON-VOhdR5$-YI>T@iYn%lH%DVrBg3nKam8v;iBMRqt0C_YV1Pi2|JzZy?zC)TGN
z1FwpJD3~61z&%P57iFvPsUQDIn|`0aJW<VyCF$A7L{b!@HZZ)C-R#G9;l&ZwJ}{q7
zlK-82=B6K%X%TI;H}BJ)OzV`M!Y3PBlt2G-(I?1Hk5eD$8S<tZM!1#j>d=B0{XFlP
z%A}+e#l2w#`Sd9}=jg{Rl#hVyl;)~mWy15{7}`&-oeNycf8i(h2!t&{Td`W*>MbWQ
zSF3PZ8MPN(mn>{?u8=oaG;fOHD@qxyYT`rwwgs+9MlI%H&Jv8=aa4`kI-58xRHoR<
z>`jr~FD46_;pUJ`2Y?dP49jS0?qRrw8tagcHNM{(iqRu2`N52_rXSUWo5a+L8@@xV
zMAsf!ovKsylI1#UsJ+57nWv|%i=Xe6wBntet9`5-K6Of?YdsV{o}Dl~9DD<^Rx2h&
zN`!PW$Z#|rpUu4FlsQqQCf41!F(#Ya*(D6PjSw>XpA>Xd#o0I?#cuKZ;%>7rK9MG$
zG51q71BrWXt(hB{wO)Ws07j@}hKd){u|Y_430*SX9G#1<$_udGiMv>_i3BVyKd^&W
z8%q=IhRuhHhtqQp>?8l4offRbbkLE!-#`{u&mgZ?*bfmj(I7=#3K&<JK2K4YZ%Gm-
zJFLj;oC36&Jz5YLYEb!urGv*GaLU_TZ<_#SIf~H&yqWTr$1Q1<f4<c_DLX-}S<%*|
z)i*1nT!dP52fuv?_-qa^SHfFlnjGf)je`0fRL<lp5cjdP3#NN@fPOhok$scaGHFS$
z@R6}#?vc|K@vAGR*LWJMDfe4^_&d(8Mh^k?3Op+i@d)8Ck+q)pYv2#p;W^D3-_FMc
zMN*BKZ4uqjv_OR4>`I_p2OE`W=+XI#_HDG1!1pcO?CI1iq^jYG&~eK}5Q}#IrT9zZ
ze)`$sHUq|@MQp|*52U;PT&7#@zS%X!gcsy6Pk7aC*Gzu}pJY+<@b%lu80Jr$q9tY0
zYS`6BzWi%D+3+6vh-Ne6x(SSNN?~-i+#z{LEt)H20dk_oxbzVab0)|e`-G(A8{Vj(
zvQ4DDG3MHM&aUW_4j((12+^+X;PG17)){LM)I2wMYj44HqRRc&?hv=5A~rlb!JuIr
z+D_09|7U2jT_9Xmaop3^$(`%xU|68(Ug8z1PQ?3D?$O5S5pDiPpv4ur2hV%J9+R(3
z)XB8d&2_Au`7Kn<HxD*0Y1L65s6^d`)Y!?%iGR6X#>1VX%qzzq7pdjKsDnZb^zpln
z`zRGf(jA%_>fZ4s6vs6SFyG2g0<z;h!aQcvq$P2aH*Cs`yj?IA^m6oFx|tgdg#{L7
zSqV@PlGSZMx57y^2yBv>;tPiJx$xWt5noMhrZfACoP3}pBlv{ey3gU|jif3vC_+h)
zb-oV`1EQi#x}@?&{fzU+(rII8aR!&6(LmCCmDdX3CCg}y<m{sz;oj(*dlcoy!;TeU
zi~Nz#znvfgSFG!n+VGg1Ut9J?nyi}3ww^FKfc2}3rV(zHBkad?oxyJ`!%#O<)`x`o
z75e{aK^k}&b9S({Kl;2I)p|Wigsz@GBO}9M0XRBGw_LluG_Xe-Bz*7m`<oop6{(rZ
z0B3)&;>(c8r|x-wf~WDDZVtk@!hzt|SJ2CGHUF0L>r{9((mveH2#XjWBK~20p?%5y
zv-6kOnZw=Y@lMzvS$H0%)}}W|r=_|;guf6d^^ndn;Z%OFW>vBU)WcRL(w<KzvDNoT
z>D^QgS=_<F3nrTL(uL{n4r*f}Okbo!&La1OA5#Ccl~Wqt*it0&(q*K^xyYz5%~t9T
z>PS6O!Kq=Bi)6cKB3B%5d&G{+`0~6pU@4xp?czaaM-Tm04q5r9&I!v4F9kQYZw&I`
zRxe;v^m*oO)A-J7t8tz~dM6!%!w=FeHYOT^geb|-WA+uWec5Cc(UnuR%<8Ll&dftF
z9jMrAf>e6b+l-6@dfjxS{R7*_Ku_NJqxE3FOm<=~oO@yid;0TKNSF43b%RoZ%dPfj
z%2}@_GUfOpCb+6Mjs-g9t31GB1#eQ?iWGsrOiRvW<YV`$iMi6YFV+UmB$_K@^GKF?
zpZg~4->qggDIeH!d{9P_eEk>Yfrqhg%x4znu9|fSd<K<76=++NTv)QQ<~;T_RzTb7
z;fk9K^Oz7jyy?cVTq)^fap-Doa-fMvm0OHrqr?k9xYJmI_y>8bBI)4fGPsfS<n<z8
zq(SI41#*+dX@S$Vi3ll*C586w)3y`|b7EwQRETwp4#(6EH-0yY6_+{GI{&d=nR`*P
zH)Y*pUyV?=uTOL^bFx)&Vxq#r+<q&`F1P0}5yZa3-@lpnEEi$b3KI>Og@Ri|T9~Lt
zkVG#~d6Pi1^PnXI#H>RY()S^KE5GGtBzHA1XM?sFW|Z^AhU1KEDlO}evySAu<C<5C
zR^_G-gqD^Kk__TEd4kd@Dvwki>4CKcDG09<+2$AwD}MpE0_pxioI*_;!PWOPTPAQL
z<vY=R?L=l!IH3Ydnnw$94-%)FDU5X$cJ{C!x&)4(HPrwQN&o*YOdu>zh}$;1tP)9<
z35JD^dc<&Eq$bTVb8x&*7spfv-$A?#+x7_q|J(^|^3N-sQA@cQOAS%i-%WDeW@~PA
zF(}vy$z94+2L)B7Ma%b;r_e4{G(zO8n*IA8c_9m|Daz=(5-UQ_Jo&^UaUUFeKEjKu
zz3v<9Zu~%de_ml(jJfhCU7loJ+b-TK2r;cAV3${$WxxvJwhgo%H}R2G#~$4R7yXq^
zJ~M@Qfdtyd@cXJID@{7dhI<G){arEdn6@8Tk+ObYYy;;<Me7eKZsKi`{%`(JZ&VZa
zY8Jf5kE$l({{8O)Uj@p{zcoryHD~;?X#uNB;qu5`W0hm6^m^#xCewpe@gtsWPJe60
zB1`G+W&DgB{M)3);kFIeCVzkx^>&f<-2%%pHqn02Dq^L^-CdhyZSb}CcMrnULgB+>
zpt@c^HJA9xo*yGjY1X3vcA%_1W~3KdJMYmnXNY^vk^#+14bfr@K0!F~ra&p~Pd+h=
zV8)CBTL$ECIUKTt6E|1dZJ`7x5-FZ^OolCJ`yK+D!N!tzz-N<J*eJaS;nl^BNdsf$
zX<)FA+8nc0)hC8p(mQFJiXGg~`D@<(cR`F<JWN$zkTnDn{xJpTd&9~2neY6^nvsAz
zvxa9JW4bi5qnF7Ur$e7fI3K$$cD_}xEZS@eERlc`ap|Q4v=(*?CIIE+h)-v?fT|55
zN*FLCU{YW_4IC3isyjSe^xeQWq#S3Nc-|Ts2UtjcWRm5@!sTmpbNl}kMg-z5!$4cf
zer-5_7>Mj(bH1|lr1*+*UmP@OrV`OH^(BDS{Tl|@tmvbR`iRDrFUn)bmj;p&?EEI0
zoFQ9rSb>i0BbH)!$-GXkte|i{$)74;<6m&zbyD)b<!aQIX*PHx-1v-z=9f5D=39H^
zY*Lsa?Lm+#%Qm}ruN;5<8x5|S2zXdM+1cJVVtvq--REke=<Bc9SI#mWo4h}yJE{Aw
zsR{Uc&~1G)YNih|;~)g<^9A)V$J{=5mbB?dPuTL+Mtj|e8h2v!8a89ut*^P<b6q|J
z>nj^3x(Yt=Tx2uP{$QkOyZH7G42^C0j@EkT7Jx1-hr35xFMSH<sj~B)v)>oUw2+>n
zWObl#&gQr+MB*rlmWPaP+_f*?0|V1l#^(jqd9S#8AHSDR)g(WBN!1&T4LtSOgP%3s
zyZ*eE;5M+R@ZkNHzMJu{8kP^{8Z-mARz!R>OD`Z_aX(~pX_h}@v7|*bxT@9DcgfYs
zEQunKfg6V`_a_M2<LQ_$R3Vy>VFk;_5>3C5E_4-fSLdH?3&TR|VhTSep=nYV^<YvC
z(<atHZ+5%yCIz%D?A-_7L9YOJG)gHNlQuz!T{w(Tb<c?5B~q^U#pSQ&(Hans4NW`>
zjA;z%!7vIk@#X#GFg@_Hu~LI^)s-)IlLZ*O`E^AiIZ=<go4X^eLvH2W<@xDkea<!M
z!WWiC!c2&I8xz32vQ)FS;a4H@8*_Pcpz?lF-<=#>=Bw0;+9KB<EQ<^i=jMn+a|+@b
z_xU1<;Y7kJ_EJxu<*IVPIW7=^R5jQ@ED+99ez`AZl3J!&7bwNklRvM_HDxk|X_2?z
zLd%|VoTxENEq}Wkh5hrOrOiP+kOBwA&NRv5%HWIVlY#0y)B9=<-)m51lR2!G$F`i4
zx)e6ESEtyZ&;rQJ`h8Uub7)qCRl&4q-o}G^5rvocl4(8!qIX4R3f}Sl^zTgQ6kDFE
zIeekqZ-DemDtT?XD*J$k=0;E-?X!;O-5c-{crb^V0jJGCIy_xAw>Z#iyY2?3qU-Cc
zSsCW-tV;IuAoIMbb@)$pkBlJEBI^Zu+0yb;rN?7EryWaG#d>N~jipLJa)0sMWx4vL
zyb+aeka4K~vBv4oV5)Z$@%l>3KUy2F8g`LWt4_=iITL_7uRA2nxaWU0ox53PhZk7j
zeay|D;D)*Ylt)oR`&!l&(7pxwSkDLuNUV-xG3L9`3?SmCUU{%j{FNmwZ41{3mh6=5
z8)mQ*J%+Wf{`-0njMh?hO}+~nRp+eCY7%IJS2D_?f7iotfRfHO?e8*g!8w~9_u7Pn
zi9>^Ev*a>*m{eML95Y`?H9h3aB)LLt#h^m6v|Tsv^-G%HRJXO&9f`Q|+UQf%#1~8*
zC3OTHs|+Y~SSkm<Yo`Xe`eD{dcZYt02R97)MbpU20GdjoK2$g7l(kqg?78`Ja<_a-
zU*ALka1dpB2LE)P<{ZbwqzdT)g`;i&YXYlppJKryDRs?!6g_Vs->7(A=iKD!FWUkg
zZcQ<MPuW(a1Ss9zp^IwHbuD!4*pbi$g7<Mm);5$brHCSgLZVP|DEDk`iHpmLQP^Xl
zw8&qHD9wj8RLjh1?qT$~;Q8HH*ts?P`{#LU8gh*rtq=cF=xIC=H+`VCR+Hd4a>Vj+
zT6e}5=8KZ9Bln-KEO4VGAg);)!A+K6hins_4^@vCL+qZDCPXRk+HS8H+?b$+csGi1
zZyX@QpfEz3*hbQRS~s82h<2&}Db0!}IF1sIc4byg`vo&cZ^&T#`Mks(Znz>@?j6hD
z?rpWn>(;?N%AysTp(pk#Ta4^SvGQz}5PLyA%s8}8UbuKDc-~_|i~DoS$gMm>MjkMX
zjJWJ+_7j-npiFQbUbj_FB?<I*k9H3szryOT^Lq@>;2T35{@jBm4%og7B|K)_-Z`1j
z-_mk~5!uk>#ppH7zg%8fTKF{es>$-Pc1g~1;)q<U7)5;x^i*=6j$yNqV!aWF)lCWb
z7Nv`GZq*%3R&?Q=8vhn0;C+5~8_?8!iQ`A*FMc<C>$moku2P+gp;G<h+4!4`V6xlM
z9ywCvbhi=u-vvQ+`R`5#RNnqCZoHvJdT5r4UEEVo|K!VeC5Z6+Fue<9hdNc&D#N^_
zNIhf#n>McTR7hXf@*i%v%^(mZ%TA6cELDsFd~PMAm-Vh_idVJxrclC(QPZLB`0L^q
zGJ~mL)1rHQG{=0A63H>3s?((ZE^K*3##HtX90Yf^<K<`&=NDnk=M0;i6r{!U(+Qni
zAnrDiI9s~NlG)4o*9NzsV-heY4rIXHs{?=)q6PPKle;BxuPvV?(E&u62A{kL05f>g
zb%e$E2Li@n@71idVhv<^fv?b87bNLlZHg`o)2Crk5<E{oIM#8D)>zbrH^NaJufGDY
z5-{5zM=)BBO@a7wlI=*ioMHe(-5V4P*^%a+oZRlBZQC{SG<|%ELE+Okk!@z%ur^%<
z3-s<y8}=!8tD5TKs^ysb<o=ESF4Xdb!_@qf^vD|2xyf0dw!oQF`|@b<wAC_fy4QE*
z?_b9<<AjAm5JHa^a{apI%8vSrFLO_!%HGFv*DP$u0x3p!9v<D!-soVOL%MXGaY?j=
z^9r#Fu{vOl`SSOCC9$nYaqGS;G||edQ<P$4RP91H7rdH-t3YqDZN&o2*o`hFV%z$&
znvoM^05_@rM=z+n3afE8?i)zjOV_%d=S(l7G$m04<i7F{2c!Ro2T1}pR$<mur3Fp)
zf%f@>7*WoUNDeC(%&^6T1gtk>wMEz1IZljwxvEHClww%>SJRHb>Sj=DsQ+(l_xcJ5
z35`WDsnvJE0|<~t4Cm*^C?(>7Uk+X}C*Ap($~W$`zKI|{4Q;<4_w#{NAsN5ir<aKb
zrU51x=R=iWCyLH*PeA+WJ7<DY>6?@yyiqMhm^$+7v6~UNCnaum@EdOj-y64~Kw_T&
zQTdBaQdTGYS8UF_5B|j6a3^-GxTZSpY8qyaIPJQ3pZVa%>H6k()2hTf+=Q|{GQ>zC
zKhdA1C^U}a065rFjD%D39#Fp8%jSg6DjY=0qxc%f+Nk}&x(akUZz-ug#xP|d4*{ru
zb)HcYBegl{>HtOz*yAIR6&c+562ph4@wyJ%`9pz_mDs$MSW@<UJM+S)^%`6^{tUw<
zO2TG(A1i%G2P7{>e*22z85!2an>6*nYtc+8MZk#Z8;08DV2CJyxs=(-Ia;h+_-0_G
zDb?)pP@@=6vM5L!iV&f-%sRUFJP5&e1KT%wB0o;*Gu03_IE+9O&Y6Rc`5qoKL@6wy
zos$fgBIrqfMB%`{#R%#OsGLplOxn3yXM&F@Dw~;p$B}?dhbX>_b3x&CevwKwvB?_*
zU5|&SnWG|o{!DKw>*8<*TNrWEr10C{3oQS;a1D4V4V@KY#)0P;aI9$KuwN-Q4+Z+0
zb<XGb58da;Mk!|?t;@Y7d&Qs8+lDbZ9FAe8Ywaxqx^EUq#S7d}Hr2~n<bJMo&ZOM-
z$hqvJU;A-P6I7l)KLaXYSO82THmHnK=I`UfVZnQm@)<`Bt#Q}%2ABJ{n#x0oVqstq
zDZS(6m)t1OpLaF&fGzn91Bl`USpn4+!;3g<d7Pum)8(hqhy=h}ZZoAj`(b@KS*Mr+
zl|aVQp0XYSI0R1CMzdG1P+$I%gXjntN3Opa#DJaF-ro9%Pxg~Y7ju06kcUo1J<eg9
zbFF|7p&CU?7Gvqq0sjPqc&Q7t5IT0_+C^+J(Ho#K)Wj3O_2J@i^K%y!uP1bAU-oA?
zqL0qUOZ=Nk<3bPIQvb5!>{WHrmu=S4=CvjFu-c%fG-Un|8zf5l42S?((OJ&TtP%7z
z-P*=vm6DBs&iq)SP0kPx3gG|`m=BWCA6Y(u1HOg<TiRx0wveP=Z{vEvdzUE|q-@3m
zZm#}is2jbFJ7)%P18j@2x^eXHdxm4UhoTX4Eis6xcwlGM+8=)^PgVMM&T#7S;;8z;
zAuSukyE$p8&V?6Y%+&Xh^xuWf{7L2xitS-wbGM$~o?hhDN?p_`ZvQOR9jFO((Q%MU
zRDuj3aEL_nEEdB&=7CFw6QE;9Lc)RHaNLs~1u1AW!%TJ}HTdB9OWJ<*Q^zuc9BYNE
zN9w6J_SiP*<kDJ?!NjbrviwbzI{CcpLJH!<Ff***D=Xw<t;$!DrtnUB1rRZ=M-~ab
z5j}gBx7LVHmHdNo*G7<t0Ge$`bqDB~rA(<(X3^gY>g00EaxF!nhJSV0)HNToQa}T{
zfJPUXgC&7r4Nf4ZV&DKZJF%I$Ih*f~W}5v;(cNqb;|m*0i??>$2^XfKT;rHE{rWGR
zPVN07ufXcZZ99*qQJZt)sATFiyKr*mmsy%tgt%V9LbP98POOPYtep85TgaFUprh-y
zKGuoDb{?0S<9E^_cWEjMEc}u=zE2S3P{BBhBn!q6TYyZAg8-NK9AKcFgU=bW0(*X_
z!pedJZ11^aKIudOPY{j-&lUt_0QN}Wl(Yg;FrRPRl<<S*4;1>L={AzCCmAw^xYq&_
zYJM_OPPk=EWae0eR4&^RlU^=;0)A8lL7iBPpXIOhE&s_1uY&(C$I7Ek7ypI+R1f-X
zIKbsF%s$*00Q^w3$W{L~l)lSK0*Gt<jA6DY5RXM!w2&+#<}(UzIY@Wp6IwUFQX}SW
zJ|$l?r%E{$?SPWgAP;s^YfT?r-|;Ixz9=yjA1#!A%GGE@-A#*JFCS80g**MJsWbia
zW;TE_QsYsxK+h@U0H0Zh2oL6eotWE*)XP`r0ygWZ<J8Q9BwPa<MU0r)B1i;8HbTi9
z_J9^VUjlEEf%F0*(=LZav42i-0Sn3o<{ioBmGHw4z|$#~IOR=6%(R2o8Sg{-fU1_B
z*VA%iBN$pgv>;28(n6((@>u&<TF8TiE~e<KtD*JhS#1nEC%mLQt%5dE81st6<XK68
zNOM2LRiJIXi6}$A3si~^q!Y2_!U1DB<%uFA{0<fblGT0NPp{8PxBoI9r@XHxBKTna
zui-iP${Nben21R$rl|soEfNq;v%?{P28Wa5<zH_5U$Lxmog6k?j$YNs3%J8JE2;rG
zZI(?y(|)}ruo@oBJ>||&NR{~J{coL})MWofUX3<KR%POh6W5ooC=IfXXV2+}S1aFk
zWxrjqYfdJ6=9xYwCvF6PH=GeB8m~&CwB!v8RhM}Ev^YN7#2Q-MHFi=Gy_Naw({S(u
z0r3cU+_lS%nhkf}b~X9ez8Cn>KM?mtk8HOz%r1&cgYfWEZ2vN6qSdO2>3ELo>2xbF
zB4AelIGHZ;d0G-pHRwNC64f839q#+5!1M@+V3Jn3*IAZ}NT>*C18{X_0mbnkNraU(
zTQ`&FyLR=z3nhS*nR8e$Zn|Meb})ux4W)tx-cgRC^?{uA5uzHN)xuh;yI^gkr;Sn<
zgn}9x0bPKl@P_*)pN`>~psC^EU!~Pl*a-G6YY1$gB`H#br>E~CD{K2C7E7OD-9M-T
z9Hde#n_uH3T8+hEi`H5<Z~N)fkXu*qbj=ydBz_X|-ndb^fIg2Ff=hD3<V(7p6i-0|
zHsDB0HQ|R8{=>QNzAn0}{))&SGXkMR2z9p{#Pht%;KSgE@i#)iHm)_%n2@qk0ThR^
zEO@xDJ>>ZiuQ*?%Sn9%Kvk$6V4lgDFaZ|{<Fl4cop1~n5Do1JPN&f4+h9+>Vvj3Q)
z+5Itbm+#ntmu@%L&jw$%E-2d{F$>R^8F{R_r8x$ereTJt2M~T=Ige(PHn(4mL(z9z
z{HcP#r<Is=KN7@jjbY$C>ITD}?C&@f8UU*;a5a+mWlxervW%YO;7Jk&f}A0(4tRtY
z{t)&#E$G(26ED6S<vOEXp#cm5n=z9+Zr}V&H-$tklaJZd6&7;(^qe5<<a^_J*qZqU
zCI`u4(cXNI48N*oJ@^}7>Ux7g92&$C!dRJ#$swK53h(M*{$Cf&*NIbM%UyItK&0^3
zv$Zi+ydU$^%8KSN;!4A|Fqvh!ZsBtlpg7Bt`0fPQI*wfadU#&Vz9}1u4&L*Z#j_b0
zam27tm!z7JToGQmDZA+an&N)~q3ibRmzJM2hCzhftwryMTMxF4YZ-(gnh*3TDU*BG
z1RJyXxY1jh3-&RCwlA&I3DK<W=7ZL2DBcasHRkt9*E53IF_b%R6`}58MYJx`?-RJ1
z3fxHPMT`E#GKk+O*=BX%^>F0T>Xd^N@k>RRcN*n}8QOC+NoSmF3O?MVi|Nvl;`v19
z=o2pg4ZcM3@fxbj4rkwH(rJ58<@Rkyz|G9_XC=DZ_*V|Ct7vBAH2ue$<83d-&LI9o
zPvrNUBq_yI>6#nxhhff7T`w25rZ7mJ7^7gHYr7zf`PrOz+i@knJyp$@+{%M=<E0$j
zLjqqcN_^mmaQ~d`A3*nk?QWy^9^U6^^I6#YEvDyr6BxJ?XELUm<ql-63wfz|>CSA(
zib{_HAv6xMI>aYi+l|Hk%G+);3?Dk4)E>~Ww(77X%uONo0}Q?k^bh<wy@@=nU?B7l
zyPe+k)wS8^=hF;XrvwJ<#$-VV-j?16;4$H#)3R$%Wm+;{aI9fh%7B{$bbtEppTqD3
zSDYL=Cu_S>_jeLdwpa%L#+YJHrFU32Pk>nL`Ep|W)L}Tnuc4Xt<Siw0UweH*a6Yq(
z9bZX@xrXqOA5kLO0DGxjwhnMD1YlFq?fq?IiuV?2CGn1qm}%5s%OBC{wnf_Y^uIWE
zTObqbf1<U?-xF0?;%MKTX&z~Ct8zdpT1bD--r*mm`^8Arl}y?3Cc(Vxz`E5CH<Dxs
zcnQ}0AU^RHtnrDiV-26QkA2=GDHZsV0s+PAGvA+w$^pZWA?}|<qa={cim>>7apyzD
z6%F$|_5m;XH-!APl6Vv88-KoZuo3e7@6Ug|R20*pX}C0~5yf+9c60Nifbq_o$q>7Y
z@J}mm-nm|>^A}~}gXdAJC-!f78vD;eOS{(eHkS2nh<(MUDEaPdzF>M3=;@%x5$x)G
zMx;bcgo{c!m6FDidt>k9E#Y64q9g?98Y1#sn&1(_XoPHj_F(Epd0>b9Jr#6e8{Te*
zhe5QEC0(%meSpixv)3+e+&)(K^s@%cNeVhNrVsvz@cJ#Vd0r+)m3*a-O~ws%NAZZh
ztz=a!e7N{N00aldr<NeY%C(&I7el4hzfMe~IlKD()A<i&Vc$llpXQHUgK11#GZ9CA
zv<@`9usPvtDK1I&cd}%NL-W;dHYYz@k~3I4HwmPMFL<8|pvXFiEwlb_cg;s#S#bmF
ztig<(Xf+0$Wr!Fl!Egd3dT*?hI;5L+EcqG-g~Ij#xN%K@^$8!4*mO+--Ct8ZZV=P*
zOx_wyk=_?FJ8O1+3$*q`qyH`>h1AQW$TnkpYQ2DCM9k`e7k(GW=w_AU9`Xe~kWSXc
z&IkPrggii-6yM8=*ObQ+fglq!Ylu6xk1Gu~xsMnP&1(F&3}85g+2Q44$Cp)tCU)vK
z-vB`a9O7A>_0FnA?Q~pgwYQCw<s!vMsDHjdL&UIJXE|!##o!_Q{Af@lD<`oR_xXz!
z(ooYN$P92LP1Tr>gll`nvj(N95PWuEe}u~cKrCMvPKUnu(u+6)!l?ha0q&3^QS|BA
zNDAzR=SZ#dKX0TnX0jlw_C+=ruU#hBS;c8!kk}Q;Cu({)!K|VLpl#&wi-~0YX_Z!N
z9<TKZVO!^$VPi8*Hj22cWi7=&-YbGuRCD<_O8+V9<#`BH!$-92th}UDH1*%*h`JUb
zBa^v}^JMH}%A0!vPtU6JBZ(%%Z5!>1t;?R^REmEhLj_cINZXwh06(rdy=w*{Y3Iab
zmr_a&=~c@=7<Fvl=)X9r{DUv8dNn$uNhMr9kt?L}CzzmP^3|7MEyiwNZ1f~E%)H9|
zT~ovPG(L%XAJqmu^PYuG-XE*9Ubrcz3W~U98u}D#h!=iR*IGf<3eoMdeMuGi;zX9m
z9P8yn7uk&t5GqGiTU8qkuX8_(-z-V?gafKmDOwmlkq^OF0F-_sZ_NBjnoAznVX!#M
zjHg86oJ@EA<m0ELU$Nt7@SK@riG&}Vc&qzGKYF=WmDbvqyHFP;Y9s&Fj9RLB01&b(
zqv!>vB#AfS1@XEeEqKMD9z}(_z$E_-z?>-ZgE}~=9wFQ+3j|9D#Q1vR>N6u4*#XFd
zG9&h}lnq~&exxsm3c%<;y{uDCaRA0yhw9d=fVD{Bkf!11vBLen$qk!S5^56m=TWEo
z^hnUAFhpI48G=Hp+{KKAH@WqQ+JEC*n^TRcI;;(c5TO>_#Ai7SudV~IC__eoMs>A<
zKSbQ~&M(ifI^VnntkunB--`;W1a*%TwKpH83KXJMl`e-23yrO8GtE~n>4Mvj#qBvk
zime|ICv0WNyzXn&vvIBNnQB+Ufa=@CD?~X%PI7iLntRgmU%kY$;=b=aCvzfVmsOP1
zxJxd^Uw&Cx(Dq?}>HJ)BeCeze5(rFQp99=hL%y$-bhwFrVog!RjI#K*mj+0vop6e`
z8A}cGepEW#{d$%z?Guc4r%QcQfsRd;_@JOWeH+xj;jdcd_Wp%Qtf9F?5U+fm+lj||
z`2*>g-Y~5I!}9u4d*@XI<oedzm@f7TOI2o=PbN!oPx<o9dxXJ4GXPmqRq5fjSh9=q
zgHhvF@renxM~oHr6KDMN&{-8rm6Edib-&|R`1392rvhan!5kCH9|#7n8_L6;LKZad
zJDCnc#@)^f9!oBIzx_%Ui$DFmPAgAoKffrk;XR%WhcA~|ioOS5xpIwghWhWqKXSn5
zNvDz$az3ig+F1ot%j@(sI#1uNb!8?MOJtw7rrL1hugn%>b-#Qj>2GLXBtp=Q7ok84
zSttkSp`(2(Wt8@6*^IB~Fj4!fTLRxX1mg3XS^q$RqQn)}eWIY5yKV5~!w$z5#;fK~
z*FK$I9Z_vKMU^y9T8H=14p*uzSTM{doam$hVMg{D4zLXr3|mLG?#nm^QXJ^87M99<
zMorm!o}$z2c#|{GsspyW>6b(|l9q(FK?3knUcc$5pn~Au28kxFCUiOr^gbu+$APwW
z@eo_D<xjt?i8X%N>AH??GEuDi4l)Kd!OZueul)6om=^WBCUZ*%4Oa-Q+WT|O4ifQ&
z`-(w3KC!$*uU)SQJgTVUh%gX}S@`VG<L@DXj&&P%QWeePA5wWU9rj-5$bF++?a=S@
zj>NtPLmxayI`Y|i@a<b-x1UAG0=K(mkYsvs`(*VB0^h!4IP88?L6s#ThmQQ;g<ks?
z=}$&GdVrIC?&s0Yb~_!Z3jcKiqZ5!)&37kY{n^c(#A2lArvNlN^L&$AV-LEXU=ozp
zD8qTDC+*N@cazRUQT2-=W$G(lhy>p0gMYQRoD;pYeCHUXnm_yj_xOdwg6UFuv3UW2
zZ%1a`5Ah&GHM-+x)9n1Es=HP+A-uWb7v5KqJ#yNx7>Vl>rFott*O5@qFJsR|Nw<{l
z`P{@m8||wF$0n`Y?UeYWR=<cfQ&34wQGZdIM8*P7CvZ%YAA8zDb@IW+c&8>qe~mG*
zwLa&Kr?YvMY&^DjsM4{B8s9RdFc|h^$j%r2JmFoBIuoSpYZa<LEBbmk{^nNx?xNOR
z!bJe=ADtbG+ZSm;K-(<0Stn<!mS0bR<&yv04`WuE-%A(JgNF1wp(zlSvEo6~Mtt<P
z>A|rggx#c~8h{tbZ8o;N-^O-pmk?sze*j;ss5wdFf%@SQj_(w;0H59a9jT(~9Dcr0
z%se^o7U#fk?h{gQ7~aBxUMp8qk{rndBwoZ~nP7aK63KBW#Bn`~KSD@<9&n+g=vZ>4
z8W%%?!<^S}&TTjzati8HBH{B4*8C9ONW1G1jE&gd=AanF3o0Jp*ynE2&MgBCHrLP?
zUfCBxdMDH(C6~vBIIh<zF0Uf3)L@=wu{MXgKScM}VU}K~8bQ5oZWsU$N;61;ZP)~L
zL!#+<lxM>fTZ^<Nm$DtFS@B*DH>zJrS2Upv*Rl>E?snFSCBM6GJ$VsT{rvg!<(C(u
zR!wGE?h*#xIfi>4MZ{)FC66bl4Ht<ic}0!|Yu@*9DMD>53xpzF?>d*ZuHM=SljSQ|
z{Sw2loPJGLIz0WwF+WS*H%I;L`<)zUu|^zaBp*z<nY``6G=7I&Q>>UQq9~dt3Lh~v
zQ5CNgOJsgnONA&v&$A1JOp0(_e&S`usy{2zJ@-NAF25)4lXA)ikpbBSOI1nBzG|)(
z{WwG_ifmm|l-*8*rDA+J9@Jyf;s%5Zq>PhBG&&$t?9E8-QC|Az%g5|qmn83RYv;Lr
z_MTDh9n9I{@h)p8DTb~dM#tLuGz?3W$rK)nY4@*i?VXi8MKe<k3^~u3)7N3W2`3UO
z@<XK$BVib4VrVzBQWUrT7dn0sZd2f40Twhoz~sYVGC8$he>?<y^cdy-1{|28ei{FY
zGs+Z%l#6V{fMr)Jo7kj-xNp60_R?<vurP_<`_P${Lrb{`$JujS4*D2Eh%<6NpBjV&
zN^z|J)~4_V+$R*xhUe7UFSm>Bk5s1h(17Gn4N``9nqRd~yS-XiFaWTW7R2TvjG#Ay
z8fux6dYyRvN?n8!EEhiA5F;6dL0>LNBm?=_3em!sV4sOo4>c5uNQuCN>B3=MSjxE&
zDHzRh8_=&-i)>%9w{GBfd*jpQ7@gaRXc2NC`C|yO`R<Lm-NYCcOXoqoFt)H*{N?5o
zx39cAx@j@_OV^Ku#ybfNo!MM7*>bp&?`~I9Mq{o~{Dc>ThN$MVbTTZt?hdHre-QET
zPwfjm_mL*qqusm{Ge_+&7UaGxFRyQuyYjvDN?@#6h~{dLv$&>=Kh;}xlC8QD5G31*
zNUn(P)S&7Weyl2ITR3Hz@8>v`ztt8Lp-K*JB9);H)(A#E!#c!ICaX|R9}QQBGuY%x
zd@a7yVDw|5l9$A(5!qzO^~hflT!_pXUlAJd6crby{jp7lYdcN17yhPtFTco3MOChP
zla31qWI};J@`aJelbiPdm92bs?4*kbYvN@CZ!+`E<8>ltI98an_Yv=!xGfJ4?PmJ?
z>qKlnCjBlQ3$Or}ueGlWnfz(_Zrn`RY-(c;3Y5a7rkaVXv;SRasrZE#%sXRW4(0Px
zr)LW0T19}g0`<X3rU`OsZ<D|>)?hyABfxDBut9(>$<V8tq4ONKyKv2c=!>_BSGXJd
z9l}&0YQI*xi`P=D0$CVC;bWOA3uwoDkk{pON)tz=<jqvTg@1{5lRIp_x{u~juwuK1
z%auMogPp)diBv%ZZZ6?OrWFBekE^wc>ych-zDuWcQ9=^)i(>jf*;(UV+Cgo^u>bDL
z^~QuFAbcgVAc9TJ-D^rX_YYIIzET)$Gu_W~gk;A}H6pwF%z<6mJ+<2jSuS62wGvJ*
zmd}~p{Pn<<{5|L6MP}I>PBYvS-lH{>Y=d7ymS;!MDy2u-0V<sDeNt!JmC(os80Ao`
z?*O?k2tlKBuBK_W<~^|=r>n9~)^T0GOhpiWI%UEQg8S}d=T@8n)JG^XAf2(v963K}
zRb8Du7yaJ_cQ$#@O{o;!RB~8l6TGYErSw*{Yzhd0OVP30xmMuco^TXnc5OzQpnh5^
z6(;ydRa^nU)HHrB<fuC><P5Xc!DPMWdDWH_li!pBIHaJaNYO4svu!C_gUj&W9x<E+
z&z@y#J?7Kk)-u@P3-_`Sl0%e!$Y4FH^7s>Cvo_M(@=2~qK9!i<j1b`D<dK%LV#vFj
z7hW^spp$kaG+oIUl!3^Sn9YU1pAfLFVy4)vrEg)}o2?l+&1wI)Ep)nWS{Ek93Zy+M
z>)A%IhTJ+0ZN}l$B{MqAdIZR}4h@QtQ^&HqL%>Jf_%rJ^H(KFUF^$eEqIEf#7)M5%
zuc9I)kJ%n&g~>O&(y8mTrUeKWSTbrG;pcPvFBARex_>{p#kM2;Xx{u#=9(JY{3mH?
zR-?XmdMRE$^eLS@@2&LTL0e?bvLGD+bZIDA$Bp0EdhJm8lN`t7j!8ywJ8g45w@rfK
z%Z6sI8F708uAwafmhoUKqdwpGb53J=z({tpL)jLC`3;Hq#)MC#><r@1M>ME%PqGxB
z53p`Bf|<N}wo~H|E!KrUS7|53(LC>139`Ojuu<8`p26`+BJ3{nohFp0NB~$T*_O+d
zr7&L8!$bn}E|o7-Q6|1e*4a^}H`=vvPu6iPAB5oA9G{83dX`VF?k5*-nD9kuUwc~?
z?6a1HuzI$%RM*d6Dlc`<M%^y0BI5;FC+ALm#@;UPY943oxs1BT!$}R6Gb?_kNcC}p
zlta95U2-iQuZePQ-h;=&Tj#eB@v?j(iB}i?o;JZ?vc;#f?7A*ip@)ZUx|fZKv0}aX
z@?x|&+fGFD-RiwoXWwHfvMjEb6(JhdD)c2%+iJdb98WR=#!qkqX93b;RM+vFN5-z$
zR?8SzjLfDEZUE}@6}#p46C@z;Z2omJSth!;ZBArIVOj1;IY1TeRkVa6q^MFM3en;b
zmt&bhMnMP22m_ymYZhN`TYj!mO@kUFabsDjLh3-cL0+Wrzvbn;o>B67MWy_nh9xCt
zj*&;+5O}}>O&oc2`Q4lG^ymkX3tyQI*z<+v4ne*HIp6=y;VOpc))I7*?zXv4<fN?A
zF&)YTjJI6RtgDkNXiq^DgC8=F=Cr!4-*^z!6zBcB?rmTpswKrN_XdX9^HIllcvpG0
zd|E^9uh~zVBGc_HLp-M?kNe*#sTuwe^0Gj$O=FBnmhp4#rd3`_NlnB%CS(PSeWxjU
zX01E*jgRPrSPyi&tDxkKGgN6UXEy-b+-N|vL<hTK@gm#+|HW^SxUqy!8<`vnx@X|#
z$TH-(s&N1y!k<{lzDu*tCRfLYz|JS1)7us-?p;v;dt6O6tKZ$R%9~|qiULV5cg<W`
z3P@G4t?TmKqGUet5hguO)6z4dHPv;^4Ojp8xKS_JzefF3-u3@TI`eoa+xPuH6*3|F
zE(~Me_p*MBv2RgXvSnYgC)qt13}YX=kX<1LS+i!DK{1qOLYAmelP0FId%kzSUw?Y_
z$GFXXo!5Du$MHV&M1_Pl!^m#jM^&$;knBM{$&C!iv)oEH`$ictl}n!D%F(NiSI2U)
zw@m>$TDM`z!#cj;|Nc2y_Je0S_+S)&;w0^Y$Ub(vdA-sd61kWzd%v+)kxN&ixIU>P
zoKc@s_LTK&Cxl>5?svMMh@6545@rSmxy$7>GCt2!BU(uJ-jA?u(6xk7jbl&wT&hpA
zLqI1=D-^>>i`oDFG25Hj*@gTavRUO9C@CfKB&+I!$Z%v>U4)x2K%Y*8>6D&x5zId%
zAkG>niJy_N-QbY}rFtriDw7gb4=lL%1m}{YU=h%`c7RF+h@eGwvYW2Fvu2!$)gu7%
ztF{?HwSfpe^h9@!%hyAkSsypO=8Gc$ryt7+AV>3!#Il)8<vI{vL*?#_r4GVw!HkRZ
zmwfw%uUauH68h}BNQI_6(!vx^&YXV3N5(y6{HzgK5l9?LdpE*%x7)uTq5;ftM2Yvv
zR>{URMBm!Z&cfzE&%id0n6$yNc+Vp-a+Mv=SDj8)^=XRkIHnr^d(`SVjN*L6$9{h{
zPHaVP9r$C!bC-P_(3w*A*MP$YL|_p00tdzwL6S&D>#NaSowicH%aK&*&+H7O>Cgui
zBTS4~#2o^%;5kk4`|nUbX{TZOM+|X#9t~nuF2tUIff023U2CC*^!gxpbHBjugSKyk
z*O7_;I5>W)EjY&&Pf24UR;)y}&EP?HX9ic^>A`xICpNJqhHc&TM`mnN51qPv0dEJl
zUOK9};7mKg_`>7ESrsa>LwPm}AHEuV$1_uZCyB|sze~1^H;}7BC2?`-c%5k1+-gOA
z6|JUSLWo~u+Vxu9Qmm=wXqq9|UtSAGlRN=W^HVkKr^<Z)R~r0!;+sD0`EQ<_2ytri
z7US1zswJ~ubIuBmyX10;Z>PB?359<wIq`isjCrl@s@FeZ-}PM2kgC2V=%JxU2wLq#
z6%*!#NY2oh``TrIMHq!a)`D{Yw5*|_jLJ&}fa&5}8&$ZaJEs<MhNu{KuEs5x9_!AK
z^W6&X<oh~h_$rWGpfvLI8RU_IS~Xxk^QAxo^P||<_*r)iW1PecFxqX9&n|JS_lmKl
z8xm}K&eQ>DjWxlc#xNY4APU6?12j78x3RpcchEI71!BaiG|9lZGG!=>xyj5DVsQ@x
zkPLMZciiJ__JGyX11Pw8maf|`*)IN5I@zBD6-Ni}pRz(7I<Yz#O^(4w2EXp8{y4p%
z&#N2H&zyTGn0kLdFOQs_683P#XVq>m*q%f`_5Jp6hj_L}VolAry7TFF05H6-7nHwp
zOCXY}&QqD({G2bgR?rAhTAx`jZQ6On3UfrBAwmmB&z@ETtz^gHm}FrHtpX_|pxCe&
z2#rbQ&R>HSw2RJu3`W02JVRu_l`jI$<*^fW^S~+jv=mkU9yvTQv`cb+dKSgia^{+-
zDKx%2f7HI!9n+S&UtXcK{en*c{guO6>_s_QeH+X8u-?Sc$>n3jYbDa|z0}$}olc;N
zS$def3m4r~M7eQ%W=~saP`0#VfKq~!0ttx|geM8-dWDub>k%yoGyKcE;WNQO{lkeF
zH_#G+0BVumnTN@_rbTG<M|<5Hir|Ba4$A>puu-ZjO=Lo@y8z7!-m3X!3Sbh<;%LB;
zY>%ZS1`@E3;%`U|r!Qz4+g^2UAims-yd}^>DfwSM$OOpuaPc>DQJ5+`<y_LH{SCYJ
zvuduArAl?>OOE+w&G*?K&@3ucA={(~%U9SyY_A<=-C@k9#tKfb^1%E%GkNN~<+CTg
zPA(SKgH;jZe4%&MoaE$o4ZQ+WmH%8cb&C8Hk;mmi1|{k#qD%zlwN8qcTzUrcD^BW-
z;1vlY!*L;_I3~406TiOvXD($vH|rNS0z-^RCthUTFxt~$I{Ae1i`jiCL(A^z{7+(c
zZ}yqY=(J4Z{9XlZdu41TNP(n(%V!13%lnK)6f2%KyqEE>kuQ=tW!LKD`O!Fgo+{{0
zG|dI))wZ>xGdCWzN%sM|0-VkAHhx#UB@QeZIkk6f#jHe#Gn6r?U*pyWx^9guX7sel
zV`<)-7jZp&fEUcNsy|qZGpVoONBS!h9ky&P#<`ym<AR(5L+@K%?tbwhSa_z@@@Wmj
z!8NU6=FLPPLFsCB#Z6pd{RY^FNp+YBv6%6++)tZ9^m6xVj1Egr4_$poc(dL9n=;{F
zU6`_Nygc89OHIOxiXev!DRnXR{bQPfag(qRx;7vZuj5pRn8QnPxYu)!YSnE2?=Zmc
z+ym|XB2{0dx_={@jfdcFOls)Pkm|{XhxvN43W@{tyMCYc1oaO!C=uemU4kNMai}gC
zLk_pVS%Ae!%r&(23e6O}j4sf{=qE^87k;FH8M@L7=ZgIuuey0t-53FDG7ykQKLCgV
z8hjjG%J%ecfWZ*r5M6ps`m^-3!nwN9p)9YvSY@@8St!ecdt*j3K<p^M_NWi+!sp;v
zb44sk6*A~89b-?EeYUkalHk>I40%CEqBN+VFQiyfqR_M3vr911=n9S^ddaYAE@{i)
zJ4{n)C!V<@!C(UaF~v&s5uB&;-B&jsi_%#8G1QF}H+-3|<NehKqBgKo&&A4MZ%|Fr
zsE6R&>{V<?q*RE-FvUA&CcEeKMqoTjx=(p2u>7uYR><(e8zA|+#Kws(lp{F2#afhQ
zao&Sd8_7@h805Ozy$MG8Pp#Tq$IDM{@=Lx#naN{mrGi&p9vb`2d(+mEGQIrwm4*Wc
zG1L49FIe!afzRzyJS5Jq<yC-pjFi<-5qCbYM>zUJ)Q-NMdOW+IVnXI>-<&=Iq`$X5
zR#SS(X4(4Jxz`C9cf#lC#<>q-N`|#jUtr;&FPP=v#)Ur~9DaM+CO=gZru?;?aq_;&
zwdWv>jTVhpR2yVh`)of;dl1urM^LA9kHL3^+S8GRW0trdr%L-_3dFoS_@$7oJI_$z
zC4|o-U4wtLryqJ)bC|DCcy<$z0P3qt^gb<Bv}+-Tf{Rg+H{>N^shNXH-zf1aItiGw
z>;3S^`OpFSVr+syKFY04%Dly&eeL26N!yYV)z`S*rSQL}<!;6gxn~nQVM8oH_6e{x
zjEIj@0Ft@rVGo*H8r^8)IRRG=4q^=fa{9~$TR`2>ERcNAL)*;RT6801X7Q!TPu9o;
z>}E#ah3~)i$%8uSp1yE$9zn3q{g89mQ78}&{!{PG;acsX<rS?>bJK_wh@0P_@JGb(
z3PMkPd&)#>O^L~(3<i#1ZB_(vr0a`2i4BO1GgwJ#D`iMYR4{6O7Sdnn=g(0opf7b_
z?Fmbv15C`{b?&zoDd!uk==Ia5NYAk@50HMvkD==fUZ;ja!UW+9`WVBY-!Pet`*J)J
zD(&)9X_dK#O`?E&jW|gjaQoeNq*rjHM|Lb;e9oZ%dOg3&qT@B_tfFv0<7R5(L;MQ0
zYHworU59lxTYq0a^b;|ohZ;t<KweUL3pMQ61M^!l$a5%$e5(<~X4|&h(3|a#6)3>t
zf24%A`@zq@{481{6z^TY+X%?Ei;0&sHxu9m1i9dTWhjqRl?3VGMua^jN<Fs*5`XQN
zkny9caeGa{6j~!ah$1jl`ujSXM@BCkwVW)tUHCXbsP+*jhPP0&30Un{R1D=FA(8eq
z=O4x+6@nhE0zlz}38vS$nY3c7FsFdYGE)DB3a;=-Hy!Pg8>Z9w7%hxQHEF7S5VSUP
z!^=xq5y$P95$=h#-9)}&o;}r8bZOA=Bosh)*#>0N^uXHa@!Sp0VMdg`+a;O^!$mGt
zEJ_^UM1r~cq@~R6K)k$gsxZTaMIuGU$Vd`}AHFqTs8#oPIpQIvGtWzS<XYcqOsH&)
zobYnS!wz7ZV2*a?e-W(wzqAew5W8W^!9^<Uh}q5a)wbPm3|l2E$GAT1{jC@IwjYxM
z-qF;IjCIAev3)=~fCbK6TfPoIw6foD+jw*&adcaP&m6}Ewym<N=4EBFXq!l7+c77o
z151o&Xx%Hy!|=?fdL$=5%Br5)V+pE?SQ5>7=e<oSr0rxw^tFrk-V(3wswY^Ei;)Cy
z^e`|$V7?`W1whqNH;32fe%$q*F&sGOSd}QWjyv_DIXMSLJ8zzAG<0scA%OAAc1cN%
z1PrVShaHWJD5Oe7Gyr$!gBrqFmWDNf8CYH9h6Ub%BMH2qojlX&H+uAz0YBL`PlenC
z9Zt|TCG`7ka^Vwsh(Y<3g;sBzmY*pk0$8|y6vLUMs>0b81=Bm+Y7WNYQwfM|^Nng6
zXblY$h_^O17@%BHp>ejk9$F$m`G!51PPvh{R;$gdbhYMtj9+VOb`gPO4Jg$00TNa&
z{K&RutXs6^s90o1XK#c~TRi_GF~)LgHn)b6pgB@ETI~1xw#%=%R<aI8P4Sfs-_*m<
z08WWw#WJxJ*!Yv*zKBGM7$)yV-g^81@qeB{<BBihyF>KDg+h8M!S_w)p;cSbE5B;C
zSDrJqx#Ycg(SbRNuXnoVVn+WSsUR*{Ch45{l%IQYh^y6uZ<!@g>;*8pJJw4F5`y*K
zBv_xR9o->0iunA9aHtP#$At=5gK-BBelN*PZYlF}2B(&WNv`p9#iScj<%apwHHNop
zmY)Th=Og$-@-q*ci^Zuc3zfS{9|PHrfe{`uOwVsdT)}(awvNVtWFDFR-#-ygnClvl
zzu9k^PyoLkej*(IIz1zP7mT}oz`CTv>(E|r|4Evaa6>tjnClhGO9DinB|zYTOMcx#
zgS7A;-xkfhCbv3yQ$pD<w;pS#YhFq6-xcP|YZYVdfW8w>u=bVf<PFU{V0fgH-tlk#
z6tXeKxK_f;-gi(qSw5eudtjX9RD=}70T}9a=!I+HJ6!T}RhWD>n{G{d7bYzOyPr()
zIXxCH=K!tWh9zJuNHF`EK~GE_qsPpMBM{vU(xRPo)-ST73=&vfMv^QGPn4wwFJNnv
zkr1{5h;XuqNWh=+4+NTi1AEBLKx)aG;@w}+Oae;OB2Zns<7PjsSHMm)boZQjbC7tq
z%e{_j+9}=;TD0)>BO3y0Itnn^3b?GbufKolotIi)A<1d<OW}&?5-j5yfA${2FHAgE
z=bmdvG#)ygdy(Gp!^ikEYwx#H*r{+!g<#29mRjQvh8oGny?o@S$q$rUJ^2kc*`Jw*
zwpp%V6-4k8HUTV*2B(4&_4-c>Vyx*f8Cm``YRkFC`yNN_fMl8<_i%bF5ji~;_w{vx
zy+1NyVJdK=pM7i(KVngMy}RhVD@;~7lVlMv<F{TQ<ja09N=_!i$iiW=`(k{>NFu@?
z=JWB2)lZIZ=}F?pj^_xeJ6!+$lSmFa%P1WBVv%L}ELCtCUoG<8ID=KL=DIy;U+@=B
zN4XlT;+6gtaSteMCd8sn(M|R);g^55a}a<~MH)v0+VA|RH>))|o%Hb!L0@cH;7dQg
znAr}q>sSm(jR{5~KVM+UVxfjOt}<oc&gdAh?k$yiQOPO`K(!E;ImnrLa$niF?&GHL
z^c$t`+8BTQX<DUnjJHlY|4~&<iP&w6wOn!h06d4Lp{LNwCbsTfEd1kNpGQLXa~4Ff
za=>;1ZD>x3v^e$pXxSitIKI#s^FI>;do2rmF-o#VX&cA3!_));V0qr1-prQ_0LC&Y
zfhGcugj<Vi8Cnu_9e#Ex6W(<k-HTP;yt;H^5KPhtk^F!;$ltkEp^qy5xIwvX(O8Xs
zF4nIAEkH5@<NCe;6eGSi0;PxMT!l=D+QgND#fhcIdN?+I{hQj%?)>_#?Dns2X-Rr$
zADI9^1j5^kcN0}f{)*$Z_g_EZCTscjT)>}~`itBLFLU^k-4>|IL1ESCO|eb>8`(vG
zA6oZNFOj<i_msgZ!Tk55gQz8-f&TUNl^ei8M-&)UOB#h}bnLZY_Eo~ETfu?61(m5U
zKPnJQ$s?--e)v5mNI*Bv2}AKS>b9CF9q>vW<d8Eu#B#KbdfGPL-(u2l$9u4p+4Y=#
z|7--06bA4!Os6uTu9!82y51vhpds`M2R^hwQ~f}lfRbbt=%s3d!k545C;t>aOQ|>@
zl$Zoy-ih`&<Ar&z_^Js&gV+FM4wwI}xZ|vO>7Cl;HsjuWp+MoZj-;HN0s4RhhTkKX
zNy(8Umr}DeTp|Gp&sZ@2_v6Zm$n~D^L1sgy;f3~7nsdDW>;?AMSr&JqSNiBg==41%
z{B7S2fDVaqPFv~g=jV0=Qf}FCtl6C078F4CN>wj`fLjVx1J%A)UK+^{*7GfVGOLAe
z!sOtKLZet&lW1KyBW8{S;auB|IvD4I9hNPFAlkIGyx-v7QqjFH!On;5D-xKBLdfyu
zS*IYCa*GO%>_9<FjISo~JlEaJaKVljg6h^`0GIXFOg8?AM;aHCFQ%PSJDz<rS!<_g
z2APTHM?7l5-^ELtAJ$Dc-2*6Sana}WLto;HSR+ORKtUiQH<rn0Z#}5ue)*DG?>jwX
ziKvV8!aHv4=V;Q997}*9+%T~HkW66SqHK&vv>faG7F-upb<?_#InsNJPLqvN`W0He
zaIDriAWkFRSq%-{Um<q*2HNk-bwg36E#YEM2?#qoO~uTbT0RH71826%#_b}IyiL$C
z3sU9B316=^lKL32iSJ8?#ux7Bc0bdp5XEvsr|DbUe{cUqN?>#504()u8yD^x3P77i
zIRWv86zoE!s?K~k5WL8HWNBgv+*m1p<@Z41=}a7l8$13bVcQt?rwy=2X1dYJ0L}0S
zV}7As-J|&Ip!`JI!@F)0-PSRu0^k;RofKEd%NbqBh@pCu(unsV*I{54{`ERqAt|D6
z(Qt=tL+#+~h6csh-P$*vY15u2=63&Fxc2~O6q9qMFQ4(hjdO0^GUFT@$X_QJ$DRXS
z)6dK{LH0BeF)UTgmTci^C6cj4TAGq%sZ=5`3m{Xybsb6bKJjp!ga>MT5vo6IubCw#
zay?I)yOpMEIj(7R5-9L6?c1=?q1cgS$-e%$>838mj_uid*t%@^{W_ayrLdJ3vI8$X
zI}uhQ3A&87UTaOHcV<>B&%0|ACdCNCKZaEM16=r7UKvL3@5jPt)$Uu@yborQe3<Sa
zV&V+RNGfeYNo{4rN=!uV^uD(Cl>N|_C<0{D)gQG1m!Yx5tNR;H0vEJ(M~@B?Q!lO_
z0a{?1V_A^0&RX)P0IJV7A@(yBk+bFp%Oq?19Qe`1g9a%_s+*Xh;o^}(@^Dt!@z{oh
z0j+$$=wG)!EbH+x8PE(iNYPY#0Qw6kMZyDEYO+OGAw&Q*A+Vp+#SAKX))5ERlVRd@
z=4h)q!21U9#mG*P1{d}B?tff3=IK?SO6<+V_CO?RB+cK^4{@Uft?Jk7bV`TUhfAX7
zR{F9l^Vg4AcJ$Eg-e&@q;2_j7KlOsP7|Y4+$hnqE$Z(~&T)sEXCeI3?Q}CRU0HpXx
zccnPx;e0ofVH%S0is;X(`P-F%F%TEW@fUoDrMJ#rJi+ntd#Z?UN2XQgZ`)t-dUJG?
zq@ks8X^bm7Yk0d=aN2a#_o`q6W{j>#eO-0`Z2L`@-{Y?O;DUY$mo8Lx(vOaJ>>UTA
zl*XQJ$^NFuMKeQ<%i>4cU|kTKOb(=!=XRSn!q8L)!H+Hxf1E9Hp1q8)mA1-K5<i0S
zVs8o#=57J@1N)=)jp^g{O}F$XzwaNsIAcsK)B88JRQe~w^HVlzuF}=am}|VKj*1?8
zKV%o38UX*Q{p*bb(mZBKhhBL=n>RuSSVe(gq7Q@{c`<+E3IG613r!x(fJVMs=NhrS
z4Y(pjOm#UJ5|Uzc5k|(d;1pWmP1=oU0J;FTe?|PF5Kc;#-I=5US@vFKM_U;+s{5d8
zP6^q-rz(9S#47|a>HVf#Ro`cS1U=xk*#dDh*ox9WXR$U;);5MWK9L*LA?#rRD=V^&
zAvtSc8!%k512RM-O`!Jyaxd!^T8FNkaRcT&zI}-vQ%HuPTsEF-ohvsCl<iQz=b8hr
zpQ7#NgXUqQHwbY&J>srgA0#)@J922Vxvu+Ud5k@Ju^e~n?BSW<;}o3VyT=<&52xRL
zzpzf+U*FtLVW~=l00})bYF$?DO#P4gM?74A>Y>>S!M}l#&YdcWiSw$?5eVtL%JPC@
zNCvB#{<Sc=$>srKY4~EpjM$8JifdW%TvWc5b7!^K83YD1QUyxO)jAm>kY*4oQ2=UK
z`b)oKDF+f=v;PWb0R=UnU@A=hJV!BJlt}tZnWY~TajtELobpB9B^y#=XMWyZ+51-e
zORi`Zq$R5V*j9ky%|?wHaMHcn1`<i#B%z+0x*Wi>v2TpP(ob{!1oHf5Syy$r)^S^2
zkutff<ZEXlPfT9I_UFNAwMWfoDO=19tUV?!dE1Kl>}RyCZ;yg%&jil+d4`!Cfzq@&
z<a@+?CebSKw$@*@^w=7!dfz$)_v5-dO~pfbCf|52vZbzB)w{gsx9$meEh<tYf~BW6
zUbF_uO?+?htZA_C=rOTIodxzC891$)v#ue2(tsrB7$3Ll=w1dc&b;&n0QM!<V4<~j
zeq5jqV+<;JbHiXUYk+IpJ}^lFCL+>K<2QV5c841qUlcLB&|V);KaKJNN-~ZaxB*4}
z9X+(Ev>w?4I-EL~u~$`VJDQlQAJ-Y{#R4+~o`TE6PLucbrYx!KFjgL575n!AzaP2e
z9|7#PMp#~Eo+t;3XXuG*XNwcn!ZX4j1DS^>Vh6OE3TLpOdC#iA`I9@N2Ji>dhP%AP
zO0LwjzsLK#WyI5(rkW=26PSbhpGVDZUise)!c`AeWffA6EV;1>;-LVd4gy#JANK1q
zTNXu`1VJhSbkY{L*F8P#hI)C$gGh22_vGIc#^ycj@-!1i_?eQPPh|hBy(boY2IB8S
zNAvf12b<mJ<a^>c;rx^)3%Wd2J(f?S9$RqgHGFdS_S_!3vSL?+RSQc%ek>cU*AA}z
z@m)*iNS_s8_LNW7RwZu1yvNE!Lt}W2-tCzLG$fA+*hA=xm@2p=Gf`%m&(aZrwm~09
z3_%4`C%V!v%1p{TyiwxVA+pGP+Cp%r?hSt1JF*V5;&qOejOKUDnEw3*k~@@s275dU
zCD@2lA7Fe?dY&fO4I+x!<k2)=>5pQ%7WsvNqw9rJn7W69#oS&+0|pLH$)Yx7WY(oD
zS6$;9UcRA+LSq+@z!QV}2;sbB`P}y6YlFf$Wk5DY5imLO(JocACB8&1P({P%#j)D3
z2-VDr?&yq7mC4Ll<{A>>RH6VFLBgzQmNbowfXl7)LN+JGra*TdXnPd_&G|jT)jIZ2
z9cC~z`o=aDMs%YeW_64M$Yr)8@WnWH;5`yK`bM@CFL6+-SM2q4yBt>;bdh@{(U80q
zJoVaU2GJzhU+^*ix=AQ_o~#-ffsRuoOi*fH)320=YwToz$8_mM;m{~iVij$;gm&50
z;;d-x0UZzu?`uWhWb%0Q>1^blu`H!&OcOj!@4fIR%Ubhl$1wYssgCV!52KT}QR0j&
zCgqm8*hZDgw2z*PO~`!R6;M-Goixd0$^ZS+S|fj*G}YimmgmEc1I^}<tSjO?uf|`}
z6$DP;I<R8J8}F2N=#m3JDEeDY3Y+%Ki4J<g#hM9zo`Km(sGJ-c8K;f^$?LMbRhQti
z0Jh+{^=KHH*6-HsH}r!b#w<2JlZ0sjM&xFoC41yrX+A5LO<nRM%huIw&bG~5Sp(6V
z={f4{`39Oj_a(E&QGNUOuYDBXJr;a#k6vQLh|wzPl<qyra~GB27-#l=S;S&kImu_J
z0X4MLcrRvBaXrAyMjQk(oycNY^oUie^Y=o+YLV{sWo(QFe98~GGD2}R^%J#=UKzVI
z2-$~hE9SM(56q^&4OU|mFugb9z^!@kNCil=Y7=%(rnc7@E#%t;#97b$RZ6k!2zqE&
zF+Cfm`R|`s&kno}8@P>e^cX0+DY-eB*>*q@liMXnpn(|y&g=$-ao_;0%2?PJ#a7N`
zX!1CLl?)Z8*qwP6fs!PZz*7D9KS?EUn`I92^$CWx_Za&;zGcPA?-w)uu*yO#SaeAm
z@HU;^P_+*qOnCFfl(V1z3D7Ql1@F&a+Jb%u%x*tmKp*g+c^D{uwa5+E>wdAyBBI0?
zHA&(wZ`?Jq8;w_==#Fj%K4VsTw`6o>mvbWmFl5T+z?hs~=%x5eOVzl8oTw9O<*>Xv
z6-h11{Ea6^LQjY<mBNBkiifs64f&m}$I=I4x!epm@7*k93wu%~+ug!d3|iK;P`t@g
zZd!a7))M@Me#umJ_-*F3Hy}{2Q=}&OP08qI#;u@_z4ZPJPrU0%llh1;fhNyc{wX#R
zR(FQcr|c<%s8fh1|C37PyW0}n4DDDWy9~pPJ_)#Ih*WRE=!#!lu$R6P2UToyio{`3
zkU@|7MDi)femOHdQ2^^GeskAxqM}zsVqhvGZrjVB<~gCzrf$O@Ry)-dF6Yy}P1*jo
z`Pcvl*!U8dH!frjly)(_AL71kxM}=$%GrwG<yi{158OvfqH5)bl~7n!C@YYoqy(X0
zc1y;k#svCa`W>j?qiDJ5ftAwSVf~WewM^tDAhO(HCGg4AQgkv&RN-U=vO6z5kPkF%
zNVc={bp#InlKZm<aJGGz=VlHjJQw^G&^GQ0jsOPo4G-cZbL*)T&T<;1GN&aaKk1SM
zHh14Lw!5~E+4}sm8ZChUrX{s?#Pah}X+_|cxo*J;{No604J!Jsdd$PrB=JaRy-uC?
zoO7(5o(dxqQK!-hDPfJ)SC@nuRfr*1B+`_ki-bgi<Vq&?A(opA@*1nA5Hl99fMwXr
zG09qd+V$tfcH_0mfi=Az_J&^XzT9AYmapxvVp3YT7@jr+g4B*u#a5GT)@-1bY&mB_
z<8l7eW8dGYn%hzCkCfvA{y62@4?kr%YuZm_xM0=;kv)!dhGq7_%8~4lMOsE4MIbN!
z(3<coo_#!|wokt3)W;cOA8D`Gk+Z&{jmdP)YH0%Iz3p8(dB`!FZN5-9#`f9~+WpK_
z5++V%)O%dQ%Xn5NLpTj+fZvtu&?69k?d9t;|NkYCW>-i_pvP_lym4-)?bOi&KNJ5&
zVa1T`cv8Yxb~ac5t9l}Z(g9X~>{tlT;u(6m@D$ex(lIBJ-zYO&BSC)VM?L^77cQcr
zGF_x;Zx=wD9~o`+LN2GiTjm>#s!FPw?%1o3t-4CgVW+qOM3Cl>`SvAe8K>~qJ@PET
z34>D&%ArcQSLSwUJMO90F0DEOBKQ#UOT_j!<90o|;U_YHk(qNj&JGzQ^!3}9h}bip
z>e>8@VB=$<Bki;P|Nc??g3cSBIgzPR84HFToTjy{or(<@X^EY%^?1xfxknG#FZ8$9
zNtFfC;RqdG=rD2vv0AjE^SPUBWbfNgL(%7U+kmqF1~OHEgrY^@+=$3(w4q0g#eiOR
ztIuxx6GRObUS>+RcO@7-9b(2ZclyO;Q3bEOkIE|=XSK{odR>(oVdvYWCvb7o8X}@-
z!;rGj4)?7Sxv9?1v!Bjiv&rO%J!6Vf2at%9g0;Pnz4W=+U&m7W(V(3k1;zZa8?62;
z3zf`aU+RA6t;w-|ITG#ER4vk$uT=TBVgM{&>;h3^5<rbN3}55r^Rx)G70<3re{pb8
zoKF6>F>LP|vF#}I%`LrpblCY!u;pHp-WFM5Y-I96?SjM1`8o<u@s0!QU5Xa81W?^O
z3G>pL^s(ZzsvD;|i!$>RH@%Kry-pkM6guX1awn8WT3WsgtS6qx=;h_@M+LCafHUxq
zY!IMs3QQsryon?9q;yfot!XEs%q}-i@WsT(bmPhE;yV029KqO7(NWa8e*Xm{Cu_|R
zN8Md3WURP2{RPdzJLFESQ_W>9bW8CCywB71O7LWtH$z$HN-wF01lj4-W@VKt3{uyp
zKwL3ajtY5tH}iuhe9Y}-1A3Hzy*YU3Jtxo-=%-7DptqG*6(;~xY6huHE`?Z1(!^Qa
z!MXCj^N8HkT7$R!V7saFSTK!`!`_p9^&<StnrsWm-A5+ODiZ%quV0KS*-G4cbj*G7
zsNws6|2&dsU$8Zf1#6WN4?IU8eMZ^}IFdx8jho3(*09N`moTFI!B~PK9Bl84GK<kM
zBenv;yz?IP8NcCRVAV3qfWkHqX9a1HJ2l5ule*~(hdNkNCC99&sQ42RV7e5?dBh(-
zPIYiaB+2U~JKb_l1c!m+3)~Fi*#^5<tA%Pq8FK<bWwf$Gsv}2IUGXULVohCPG_(8G
zWaO>T{NVgJ_|q-(#BIx|5wDYTCN0rAib85quW{eO2#j{Zn{2S^LT|EU3OMf6l^Sm!
zW$vETC8{XZ>_sa@Dn)=u$@j)Z7NZmmL}K|^2|BqOmo_EF1g|1;@>v;yXm=)cX`Im%
zKcBqK*M+*VzJrT(?)`6D!`#0Qj$B_~@E(2KO59K&^PS1;kId@fzUC?^ar;Dxj^<VW
zTOw{Omk09%w6V!YhRT4HvAWW@hG@TxSYwVf$x9YVkp5eIEzSZ+IWw8BfcY6sgt18u
zTu4mENg+*})+16B2)P>+QAm<gpiw8?NqJu-xzq7WZMJF&JEZ?uP(%~z%7Wvyc62m}
zc+#3v3WOS2C9i6of`eByCEtiV3M8iIGhtbZRu2ZOlFiuU8RPiV4zI}LTvp`vFgwq_
zmAj9#$TvDA3)qn9htfy0-cmuiz9(#d*tX|8?MQxz-6H$}V_#=-&u@kS&i#giT8pf5
zsIZ!o_zZx*knI@h{9}+HVm-i6s2$9AOB6m6D{c>M%{4=Wn{gX?Q`L^>8L6wozI+;q
zgqr{%`zkO?PihBsexADx#DsoV<m*-bKJpU)E3}`}2I^6`o30!UjV!pmi*q|3S?3p%
ztb$#HR;LI_?KddA-8IY1{Ohac_68srOa^48zpYVV4!d{gWCUw3$D!vs)L8^@MYh*c
zaClru!7a}Y7htF@&uaTO9Cn|5DpT1DesqsO!xFPQoaK;`{mH<J??zzRIB1QgB9jUH
zLM^G|9aBM(PO>rME~amedY<1<D6)w}kD*h1iqML$DaS)J!=O+RGW+eaaqTNyx2t92
z;Ws5Z0`N0<d{2;sF-7jcV-I3PM_WGhW}k(C2vl5*F^s&RC3CHc?`>H*c^K2Yq|dkA
z_-z>BJBH7AB*y3DPbF~gh$^|PQ`KWTO3;aW#<e=^8QfR%V0_M<5Mnwp1kgfH+}Hcf
zJ*I#emT~_vee4_;1zafddym?;M`X_a`)6I_doL@{!p$3*U}F7%U1RCrSp)C`fP@UN
zk6&i4(*P4&3uG4)2*w1Hzm2=PYHiWm#esK2_+pSj`28ScvsoY1HJ=`?42{+)1GK#^
zateVh#nIfPebyzRyqCxb*X;yDDg{epq}FIMLwY$M7R}yp#p)`>-Y0L&I+A^4JJ*B^
z3Nm`C#g9bcR~2tkJ+wMweDuy)zNQUvtTER^Hd#Idouw_aGPd|IRQlAjUgMGG17yHn
zow58NlKW0f0!Nh1mjpy8l#P3ujvhfI3yQ=*t-7HiEMveglR^g|rK&Hfrg!nNHfS>H
z*YbA@JIv2do(ZmRjbNvb+UXN_LKP}lE@U?6k5(D(T>!Q3Vt3?TUq&M%H}E9>_PkD_
zuw6SKmZ69s&_&?*1OCIrK5m@P#Nv<tmG-VtuEuv!s=sNFVNNU*5(?M@L|Do<%HOLR
zmmaXF7+%}gH;T=m2eWdIu>zC~SHz%b7to?-B1G?)RF-Xfxz2~#8yegIzsp+@(}9gA
z2|<`DvtK9Dx+$rSELa^@d)+fD3Nz=J%ck3sHkXZ`AUmYMR3$r4gv=E<j0m8)ebKJ}
zy##21E&19b`L){tP8A)`|5rh5#a0H8(lxmV$|sBNqan?%2vSlwQ$Lt>T#hX5>i{^D
zfh<<rl8p-{PXns@prXLfvSOh6kL<hVo-=pgfyv|jM;G#+^xmqu$-BM=zBAKD4ByVt
zk6v(fHvlgJLxC0mH0Xf^=Yl=)`+ol}s0HbHx-NV~x99({=$X51PrmCG$sI|ML0|pX
z1qO2y@He{$X!7;1xIC2grs_NYcvr{@^dj`Y=P+|a*ju!ts}K@ki5B1f8~KDAK(h}s
z32hD{`}WpJmOXg&?o_CBnMxldv;dg)fKNwerz*i5>`XaHHfhR-%3g&ADt!Q2^0NP%
z1e{!oe!TfrXTd9r_65Hh<FDW0I&_x^G*1%QK#X02vo5h&kVN*8pYm3WoKzIj6wr;B
zC>GQ&n$;mG`awhULe)_1R3L}u*qzaDN_bJURvw<068Pl8`?!7oGd7@C=*R-dp=7D%
zuEWD>E1aMHP=)3ONq1RhWgD7xW#8m#yC)WksIs&PmTm}N<s+z$^%u&bLI&>IVduev
z^_zoyjh`o!L6Zh+eO%cjT;18#qsEqRXXt$wc9JPUr%LJ~C#jm28f`W9b-k-QU9zFQ
zZZs|D_O^Cbg2(4MCwHeYm$!$H-j-%O&Hz|W2qIz%joJYS0&0AIAxm;&n<_fKo-*iY
zWqNQZWS8AURjeK&qSU}ITo8w>_1;L!tfD!eY?}zL?ri)XE%{;cL_3>-6x7i=N*I(h
z-j1fVB)FikR7t}&9!&E$;62X0e5$JZ9s^Ou{|MLe+)tqN`0E*1W!*Gz@5EcIWpCK4
zw0Kw&1B_t>F%RcQpXDLSZMtY+Fc)B)R7)>x3Oqg=R@;p)^c0Dl%)yQ`!s^7$_|CW*
zRYI<^GJaNSpYi}8)3WAOOJ}+Wy$!sfQoZl?4G3AK(a!)G8jnoR;w{IN5dm-Svn)45
zmod=Q;;Ant1=SKXbKRz!cCubcMkuS=4on3yqA>Vts9emOHMB9`E#6*Y6x&BHv!9JP
zrfri6@A{TN#9dtf5bW`MgcWEqZ!u-O+<IZ|$l>VgV2}an?G56crS=YVEXPu52Ff39
z$RQ5YnGGW)k$Z8Bt4NRK&rV|<Iq}mASpF!3uyD|EeE_#tJrb5dQF@1uGB5aFuIMek
z;)CZ*c`tHO$6|Qh3nHU?uRQB1Ydb^x7w17tzJfb?pS211cCXY0yoGz`i!r}O1XlQL
zRm1`Xt$ky1S?yv|jI`uo@Khx&{Yc_CafNf(_^Wjy1~mvZmh>a~;~(GxZ?6$XEqWPq
zvF?`RLo3=@&RY)E^vL$YqxY)L*#gXxA2A>QU;Xq^m?{sJ8x$PZ3N**v5Z$c>%b7q~
zL0)NUsWaM_4s8VzQhy(c3?_Vg!LEULX<(IDCbvY@*us)VlAV-D3f+R}-p|)Ro>zoV
zux(VDF5G?x@F>zOF*C}NwSzj{H#eTZ)2f?lS9;#r3c-sPa?D4`EcUhbmwhjPecvi?
zBkE<~l{Sf1O~qmuH&RDUTt0CZ-y_iewb`irE+1z%?f?v*LLPU_>}prSV@p<-)-ryg
z*|uPPiU`rh*fb%UzaFjBsEf&WTZa7^^ZR#qSjQYoutG5?BjBj;yYM{{;8yw4bMhv~
zE=SBMgNNNoDZ8lc?PS{IgL2sNoTY6)%)TfkGOT37(X_3~sCl^#tQ$0C>VydJF-fyn
zjCKaSgAB#l)}2n2sNSWw`^a;PJ$B4dH{w2I<pV`a$J)a`L%T;jFQ=(vCDb1-Y*`S3
z9sH@Nm!ih>JpqBtag61W?*>%vW(GYUAD;4p+l7HS53efePA)86q+Yo0y5r*icAn&p
z3)|-O$PK*RHm9hXi;zNEC4;rXWzKTDM)h@b$M8ij$>;lGp}pQpRL320QAP<+xH|h+
zW`m8z^9-JkBBol(7jkyWu>o7GfAmPsxysEe6ijpXizvH|)Th~kn7kUffX^)Oy)WY1
zj`(vUnEumlxwe%=!;ZzV*NMEW>fwjydGBNCsgD3;u~!>&1jvEvft#?$VzpYHt3sZ=
znvED)4gWl#{84gZY_y2)M>qE_%HwJ17RT;TBEHxvtDTrGM`gbIRQ*?)oK2VE+vlwb
zKr!fFF}z^lC(Um0M*$cnVi~5Koyr~dHP94@Tu~D4a3$C?0`6|o$wzl8xGBFmYU-(y
zGu_YBC?GYo<ny(DR(#=Id=-$g9QD4_QsNDt@k44<inT=2NZmwB?}4ZsqDjPO&wViA
z9p%*P3A71z^MdYm6O3z<>7x|y-j1&0)qhPSXdbk2mThDgM&}dg5g9;sJ2V=%TQwbv
zGp;x7hsEUu9xIgaJu%tzPvUsCOH{W$-wN}5%<v5eC7?H{GTo|_BybbgFnhC@=L#TX
zC}up2_vJO)lC#%0b*@20Y02}@bg02b>Juj2uT><mpSw}?V6|a7Jl=ghRRDl=eL1@(
z@{)yMLGY_U_tVJwRO|-W%Rv2`drj3H(qGv#6K+Wj`_%ptS7>F7*#CHn%D+swwI%rA
zdi^qgOl?5%Cb4Jbv#b>tNSr3>98~e#G;%jXy^e3WBTGTlq7Cn2^l3$p6aBud8vS$y
zq(W7-u^>C}&A=rx-P1T%GDF2uW~61u&6uQT0vKf;UClVWpzD<CG!`N`eoCZ!p#_b;
z1itsYm*`O<qBQK#D<yOabCh|C0sn$<PFp2fl>Y2t#us0Qr*z+XUBO<s=uHKOId!Yq
zyPai)XEQ5ye!tNa5jQ@uo~F}U^vlpLO{#pxLGMWNdgId*AFY+InGJW-x#y{GcMcx;
zJ$<21vJ0*AmT5oL{=EsZSU)ZO?ptbotWHzppL>xRsDY|NJSjU*75<RX)Th^Ei#-+h
zDG#(U<?-2x&%sW5lFV?%{rJjx#`s<n-^(^VQEcd-I!?JC$2@P}1SV1~y6fF2p7zvr
zYPiF`Ui1BcHEpFk{+TE2?gCA?e~Ll>{iCokq6!44`D$^3yA2`{WbQ}jpqsw9kO#x#
z1LNkTN|T?Dx{j~Mo@cluAOaDA+5&Hq*E(A50UD1`xR%HFM2{bW;2XeA7w9TcbE?23
z()eDY!IR$anr<ZfH$O{ubc84(p8?%^&VT>R75~)hY<H{W7mq-1$&7Ik)yc|hxjy2d
zus#@cs^qdKr$kI1JViI-hj}|L0?W(us#+52R}N)t8cSfo1*a?<E>RzdRNO%`F}SyM
zw$rC+ZD+jUcb!ZZ;h^wi(s^&O@^t?o>!c=cQ%cy0@j`VAyAf_)MxR6~GT~<(ujd#U
zZ#SD>m>x77<IaE_ht#97nwpy00O8fciz_lguVU_6WTG@JnxySi??xyYxsjzt&}`>^
z8SK%wlTS@zz1bsF3rqta;2Q2vCpY0}97u2l#MiyM)u)W6O-{TV8phmc0M4hzhSo!e
z^J%@gNV{wQ$klR~{;^hbLj~j_ZQ~nz;MNaw)2(L6^CHtIaR>60uEKx+u!|oWTMz%B
z%7jl~73$)bm>)$iUVm=UPY!Td>eeWvbyP3G|9h|CzkkGK8qV_y&f$jGP)0N7-fAPl
z*mwGMLANvhO$*=w&&oFHQnc&XmV@eN64DE2$`H5~Ouezw<I~VI1^1H84f#*LGLgvQ
z+jIDdJAJ(ih<?@$r=0XmJ&!wTQLI&jiD4_$ozgV)!FdHzbL<t=q!4-8;nNf|ounb}
z2j4pmtXHKBxACr~Cihp8baGKYN$Liv38ni#y5(PtL7Q}5!EP2!FVv?06jxBxf6nHu
zk#gzRrF=PHTjEAHzvzbV!KqLok!T^pwiClC>)htZFXMn4di6@#*z~yKU$`60ivZZ0
zXKm%be|-7cDYX4Q&a8=h`C5Y03kAQScH@Im@p(-wvn8i;{k{D;(1Wf~NGM_#xbm`z
zh$z*I=hcTV;ILmMGVK{>R-`oRFyebcVV_321Wgh*bzyDDXZBtSGX`;4L8}LWF1gSc
z80`LAFl|k72Cy@%^H#zCDL0^%e&1T#<Av2_|5BAj@q)>rFMYt|?{!CAG2jLDHCwjf
z34+N+s0hSwnmvoDL6In?)oiWoB&IxSfM~%)YW@otS1g<3s;P&FIJxY7*_KrAZ<z^E
zGW$SAzmL>50%o1MSml1`XKX>Ov~u&P7`QK^C`?iKw~}msaJu*sOV)0xK9cM70w!`^
zRWyLAClZ4oBHL4ro=<!RzMnU=Wc6aMDSq&8IMKTU<%w5`+D)B4uV)$^FRI|y0bRHI
zvB_Y!rG86D6n?Ba=uz*zLcJy1TES$II!EGhD6I91_8I#OO~f>63;PX8=n`HBTYeEl
zjSEd)5BVmg&{-k)(<n9N!Lw=RNZhp_cDly6fOf`Uqlr_V4<6Y4TXT>X){lmK;xpU;
zj{0)0k~8DR_rY_yn@bm{8qUHx5ZeV&iO?(K*?OUe{t+7__RSBIUA*F&F@O&`&y{~s
z6C^O8b(I~)V*5f`PSo1q+E)hC;VvBU@<m#kqr6UqDQ8z$T|ob&nD~)tm)z^p;<tI*
zt7J7RtdeqIslP*^n;5`8K6U$?-eRi@7rqPs8ky)p*2=lNbaBe=IrS}HjF$f)Zsti?
zs1xuWOu5`u9=;_wG4#~LQ*NJahSz6D?Su>Kq3?CQK;cFv%X29mu=C8)#k6bp=RQG<
z-#e5_w0s<WW1Jp35Eq<!Qhl<-9Z^jvkM(=*%bR%hk;aX0L`R49RKjiBzSElk%)#Ld
z%|X4kMUxDJvdM{xqfF~{r;vpPh0*iuz^ZOCsNS`CL?WVS7M+M;-QhIKuAo>x7`|ee
z&$L?an)v+IE$tpkO=y<<H5oasFMt*oswKh&2Lguyi&9~qR2)}fok!+O6P*c^XPNa+
zEetGW__!Wsz{$jVgAL5|3#=5(E(q{hWWtyh5`1wIaP?U6bB+z0;QwWL>bLvwv!k^@
z;hydaB4+<<>gc`9^~D>`7xcLdLgxALmD1ciu!9vqmpq#;G`_gbrfxB_Iq?o$pu=X$
zsyY^@qQhdf)6cbS9k>^0rV^O@k4YbJCrFTh@d2P=VH-{@fP7Ni{%_~9$m~tJHSeMP
z9Gx7?hT<us1SeUq`>pfc5U24xySehrk7RiKN3~x~ubQax{ASIO`zW`>Ke^`DcX)yn
zqWJF$grGsQ^x1ImO3FW2F(<0i7q8}i(5odu@$KS-3#%nV;0*9QN+cF0n1qDBrbQm}
zG0jyBp;Xrr%mqQQDfjp*K3=94PmCBmp|u|MhnRcRAKnrCGIe}0J{j$~{OBh@Gj0QL
z48BNxk=JN&6u5U0+VWK|$+uaiCrmctyQMWQFMPBtMO;u(OhE2eaOUxDN*J%Y6@JFa
z><_~=nI}Z-n-yXTr2Ngo&lackZI``A6!nVk$mgZNVTQd)M%9He`jxTAF~!++)%#Kt
zdt^(IxffUa87y<cP?~#O-rJwULnNQUM{7OR4_{|Pr~EI{3LvHX8u3NFR3FTt>%Q`5
z+kEoAGuNJ^2jY328`y?3ml%+(TgIz0z?~zgP$+xn@kDcB2?I+S>UITAR;|<e_$JGi
zbiv8>q`)5XlxFJ_3V3-g%w$t)Avu6}p8dZSsbj%*B^Lp2?Z$iottZs53#K!3a-t4^
zx3%yy%b=H&cCPE!^E%e4uGQjxQz?EbTq0r|_R{&6FBP(%B}h!e6u+Bi#)lACT{BP1
zgTD<f=B3k{MhWQfauTyF9APZRFfueoclMhF_ZY?NFYJ;r^1(1eZReVJ27PsMAB+_R
z2qC0}liIEI^)9c|3WR_hnyYTFzhaLM9HS3ec+(CVQ*U+tKAmLk?E>b1TD<>>ZGQ~Z
z;-wKnZ?|tq42)TURR93o>?2lnes!bmjpAdmK!kdJ$c#U|Ww?=>H_@*u;&s-4|1>O;
zGaegSu_|VBv~+^_SoDcbE3n^!Z2f<9>l3WOF34Fx{VJYUZd+K#^GUh3=A-v8@`7$W
zQ+iCHlTB*0rlX!~B3#-6^C2?P(@Jq=H#BwR@<sR1T=#$f+$9|OB^Im`^k0>1j1k;X
zx$NBd@+`~ll2%4YDEE(uTX(<QwvgtF#a4J4$4$uuh6tTAZ)5{H8wZZ{hv!U<8gMu?
zJ2A<|Xj>xkYZ#0Tma%-!YU^=cXzpaaFXpbRojdCu>^$jm-Ht0{+xpIx?2Kzc3g1wj
zT+EKMuQGxhKCQz-<iA{)lI~bTCbzX2hjtB5#CnS^a2^9T#J%L3bYE&3PqCngCVYX<
zC*M&@#jBp^Pjug1byk_&|NG~*Ia>v6P>^J@MmHTik)F^fqHJ@o!`RFD5xRGwwD6(y
z1*Z9R9&Ym<Riiq`qO9(4YlL7D_cRY*T;m4WpJompoxb8ZcA3D&zQ%7(^borT$FZ-A
z*zM<>=e3l6AIeCws7b&WE<&<PG2698Zqo7sW7~5iANqmC7ZLnyES2RNm}eY@XCNRi
zf+?WL%We|EdXvyBn?~DLDnCtjBQo;>P$>iUH(VSmRMuX9&{5)O<Bm)Tsp@CKyMLP!
zaqKH6i&0AN)Ao>`SPwa8>#10x3k`9^Nb?w~Btnnq9!qW$OpWk0GTnx=%JFOr$kvt^
z{_R-&r9ia-u?mEog>}!-c{&-m!y4y8g7K<vm;WHMct$Xb`GrwQCm%WOZMvsaU(Q6O
zpK`1S#MJ$2aiaeI&1}Uu+>6mq*Y@GP7Z6veV2s6~=n$VV!WJK3&bcc(3fp977XA>J
za}7Kzm;Yj%Ef5M7DQpl-S$Ig0JYXKq=9)qJu!W%9uhrf`)L3zoT$%BMHeVIGR;i+$
zvD%;Jy>uG{??xBrqM)F~iY^rn&8rB@BOGrPVy@bkkW7U|>R(}0dYsJkeh@fXNy;IC
z1tUEnIFibPBSB7<iVB8c=!f<vD84a&;A@8KG_TL!7GL5#&%U1cD5=1k9WOQ5ADgCS
zF8eI9VIQPiO)o}aLVcF%`*hAwMj7c!CF6UY1kCX(Y0HpQ{kmB)tItoEzJq_FXW_w>
zM1%3A3)w<rVXAf)3+b+U+Mn&(IDM{^|Ne2wP<`Ex|B6&Vg(UO6C{Plw;*DpH=W8D2
z(#$CGZuUh>2KYDT(n_k`RX%TS5{JXzaop#%gGY^qi13|iGbBvb!FRIAUm*qZEFbbx
z8^7^ER4OMNv)DI3eY4mYgYFukjIN0U<6g+p`yJnRkOj-)keB7gr!w@v-K)b2cdr3i
zbh7Kn<B@%(7yXF_E$HvJr!`rr^O3Q+EWlzuHMw&s1tKbXL~kirszfkaTlP<(yEZIp
zW?P2ybbzwv7fKv$8wu1p=O(GVN4tph`*R|G0(X{6DumS0k*+qWFU)6fjmTR&o%~8n
zx%R&m7TV_b)nzNT|97A+uR8HGkyQiwgl;KP?dn7Yhhl-NVNn3o(*SIV;@m@sZrcRI
zuqn5$eqQy^GB6zE=t%v*(hEjFyNxo{r|Bn%?;bRHd<IQusDlT&Sh};h$EtGm#iueN
zJ@zQA(4bMgTo=bq9p9nogy%*IndY=do|at$MUjU^v2N(levK{K-dB2}LO9i(KKF$k
zoo8qnmkGk6KN9y*Xzh7}ORADD%bD4h_N%@PYR?T5evw)9X$;K2Cd!=AiPofRDlfSY
zwiTSDeRYX^C^qxC$v^=hm33<@^j@wuk5Kv{e}kJ=gaL2l>z>Yuk|v<#3t3l}BLwc*
z$>tXO!1c!V4NPa2_#WH6t-&EGyhod~8m{&g>6^h!{NC?_Me&BTXa)4Os@)UGneU+%
zMKZ1{+2V_;y=m{1ZC@Jrdrj1`)dkfXmqnP}a}ol>Yr{39Xp3CM0tcSjZB;F`mgaB?
zrVSNZR3~4hd#)isOeK(;JD9nv;P0I`)KHo*6EzN}(LI;Xl*whOzQ!H_D~AJLN4mSV
z*c@}^jmE>*cg?yK>zjzNx!W1}NUH<}V6J{ub4X|c$hM*V@1L}4G*IX^^9(eN)HxU6
zFeCb-RL`UMB2JF6K$$OwG4fr&bRI5&(cO-8AI+8cWDF{L14LKFnPrk<J2kzI&hyt;
z@BI5=*0*EdQO4{1hUtkyXZnKrY4`9<@JLsm(5(zCa_wR^>d8X&ou8cQrre#b>9J|~
z-Xq>LO<1{n_N(<8oRd#HZn5>&pg>m_0<jkGeRw4+UH-p6E<LvIXs7L%U`GomLo#(!
zJF8xnDf47!6yPoXw4ez=3x(q%A|eLf{DN14L9_hLX;_P(#Vb8X4L&=gk5bZ3?ggX9
z&Vp>*W(u$-C`HrOC=S)Hl}U07JVmWQ3N(dt5JZ4$oGoQzB1{3qt}incELl4Tbr3Ke
zrntm+Lf^dJ3J&_yXe0SlgK;1pYO(>#z8Jr!VqQXogPR2Tu@A;bO*?h1AQ~!p&)vVz
z@SHiU?j(fR)uZ~{$ftQ(uFZM!s_mH3x6X})x*%nW7*_R{sIqcLCh|4rg~cc2!o)rq
zme(Eo$*{WU9FcG4XK;V-tx`ZD1^UCNU}Qk{Lr2=HZbZ1{FT*~VBhf}H=A1{oU=vig
zVsNFu#w-isD$b&_{D&$Qlrn_k&**HmabbSGqZKXiAvT`LCrX+43Y4gCrsfe8dt+Hj
zck*V~93!b>V&w8trNmog$awnEXTYD42%w@9)etX@>F8WQwG|8z{})#fsPD*=XW8~u
z%SG&pi0FgRtry`5q!Nbu!VsP{tfC_g+B~Fjlhu|f!SzHc<&SaiA#8}cLTSn%A=xdH
zXMv`|xYs}AM?c7d;qKp~p@`sD2cb&cBha*d1N$Oxe`B57V^0l{*mO}`D5lE3j|m*9
z`yS){Si6Getd|9ZI7IZcLkPhU9Sso@m_!&40XBl_3G`itA+^Ar@HErI!#tRfconGv
zNWl>oUpi9hJeV4}88vKV8X08sYZ|@>SRm-zq4+u`@MAfG+!twNZo<IQ-H7b3(Qjb5
z5WGa0MHE-UMcEf*UJQ0##qFp^N2ZE;9))yPWXY2x(Fj6mgo=iPPb2Ud(A((rC;rKx
zV$09?^YBf+OHM}3(SH*&iS9?~qoNMmWgL!Gcv&ylq;4=~@aaA6R(cfGBt}lyuK0*)
zb`nbA2C|Wmce`6nzXIfDNg0e-RgH<|-4?y{-(+g#w*J{++N3tlU4J9~@ZFK3f7sVX
z;ksD=0I@>6B9CELM#jNW=*$`*Qc&Pi#f0dHq)70Hk9qp10_&W$F;k?{Oh2Kw2A{z)
zpyY*@{V1uN8iS6a$$o`O)nawZRg8IGH;<Ee{EPFi@E_-21SKNJxsJ>Nd$X@3_AjBg
zgnCQO2>WF6^hZRT5Jb`wSVA(i-lRzKB3zvbo-B9{lzWxrr@JGQz$CpFvE)RF=<#9k
z4ERLuQ+XP48s^(CMm>*Zar{KzWFAy3urEySU)a*_&+MMn?3T;=A#8f|CBN{`lm5nc
ze<Haa;de0-Ujphx^oT)YS)|f3G805yj&@bvgblZ_=#wL6L_~zD3m|w2oy3JJrvi||
z+-kNVG{KE2b)_DIZV^RxFU%?K#mGy9KEWU(a}h8|#D<1i#TyK`LvCe_?qLnGCvbzF
zTuL--bs{=F3hUrS>}Rq***uVhAqnH)Aqa?!7}x7a$)nizA|fz73(#s9FnAAEg>FXd
za=ZTk7k|q2Ulc#Xbn@wYrwKlVL7W(*6{Xw%0AZ76KHv0Nns&AtlXCkIq%K3m%{6O6
zP)lSXU5lXIGEe;=zSubxq?tKJQr_|4W%6o8{g_0`Z=_+d#qfHHzo88hrg($Nv~lc8
z_GE12m8<NnF7%xX^Y(j$rm%bJg|OCF`-2*&Sa$juxMnf?#*ICUgU01cQXBT*NkmSb
z2W)3A(99C-Y-5}>=#ogb*N>C<zsUap1OEU=UlmU${UavqgTRE3gd~>*()?p1yA)Y^
zcab9q@+}{6;@S@kJ;tAsoU_UQ03>No0w;*WWKdQ!?1@aYBINl@_K`MWLmvoF6Sbnb
zEpssB&jVyOH^P?LM$QWSwf0|s<aPz!h2&ya0_Y)dh`JIGi1HpvL5B>thoF~;^c4A8
z5z8<rmn%XzE2F^6<W`7hsZLKJ<fIwlq&5>p6D!bEkxiJ@320j`lTQJsxrs=GAql1r
zvl{&$UkWz+(T3hd+!8q(SB56Lg#G9=<%}AwkeRT1B+fuxn;cymUJ#p%K_5*>L_}c0
zg9R^w_$%nX1r$LU89b1PnpR}VlO{~}rT9Uj40N9eANGH$Y)pUjY?APZEFo4c-ewK#
z!QNldC(2BZQ7m+ZEExE3b9oc1MvrQ(hUO`^vVZ8Ba|a|+>6B_TghHDi*mmV2*NDyc
zmISD>{{T_WsT)!vx@11xC3ny4J>r4C(UaMhzQmzTScq(VIQOBFR>rl+XX!8OdzODG
zk!D4S9IBg5e?{w`(CH;mY>FyQ%e!Cobb9hX!v6pY{{TsUnK1YM!eo!5z%)!RUVA0G
zMjWN~KGbzqc(K$x13r%_ZP@dq5a{}7LgAh)S7bKieoKq&_B@1A^(x?2F9bE+eGNWf
zNgZRb$F@sk$>4PxY@pl8IkCT%UoDAF!F3Yh5om_e=zXYZ<nS>+4km1fUTGBH_>}pc
z;R~1hBJS7hZF(ip`oDr3H|TY<={#u-y&h97*pIeHLQaI8A*Iqb=ue`)QMpP|vWodC
z?P+>5(33Ax6NMeYlBBOBq$Y+f9efG5B*6;IneSW})|APTGK<j>g91xKM9A-xE<~M@
z+=m1Yu^P>p1faF1NXSG)L}0;#1u09WesrZNU&?%_#y73d{zv$W{64kk@bD_xs%Pzw
z2`-WmSv{MMZJ1Ow7AF&A+u%~k+<&7AMX+fxXMj6F;pmKtztG%2lo-=yf*_XdHT;XQ
z5?}ZcF>b5+LRnQ26hmV+Z~p+mY(9*FEdJ{LP>`+=?*^WT-5j!Wsu?-JZY7Rv3evS`
zjmv`ej6J8a@2#taOi2&iI}1S!T8}egaGJ>X_$|k+FX38ks{a62*a)%oKhfwjf#OAZ
zNYDNQX70~sLE=PRB2BcXLM!nPa)I_|V<!z23l+F4V4RPC$i@Ev??nCNHbn4_fo+9(
zgH6b<A-64vK_v7fCxPh>Q9e-n&V=95Z)YtS{SNgjmM4(VultocBggP%!!yV{DIRo+
z+{*9!k8}GWcl$l$9>VA(z)A3X5iU`M$}a{~M?!akK7_=62qx^JwAq&?e2~&gKkG^K
zMe9Zf;i87cLS_tX(K?JpkcCW5Ey-IUu=z3!k+MA=#9fgX?5L=MdOc4;w*v|ehu9|U
zMZm*u=qD^~vm_-bcouV{A{M1H_^EX%N;mjt#_RNFNI{_nh(a5Y{K5YKo`>@1@^CA(
z!e8x*a;c)vh@wj4tc`Mvl=>EGGPx9olxywrYxF*_?ibk?EOKDlmGUJ{$$w%TYkVfm
z8!kK`?M@CQ#1Z#Yz}q<OG@q(|zx##F`H5w*i#MUOc@Kh#v0Od!dhs!eai7^v+sE`v
z%Iave>on>)InfKMA|-sfM9+hqir8dt{)mz;3|+Ppnplrt0{N*mkd9lb2?S}rezfOE
zgW6<cEi{C^WP6F~hob!NVSLxY{#X7>`D}&A(FX7@M1Bvnj4n#9<9(itc%<|u6xW1`
z-6P3T_IX#PcDo#MNS_Fou*DKQB|hahWs%5`^%2mQqYgA-65XNni1Yjyf7l)DT})ew
zJOp>iT)*t>{zowe)nYPxe*rH6yV$-Fevun;`vtw*5^_s}sSseOXx*ef#Cv~2;&;&A
zZY%7h-4YXeDZv%z^e5$>nLM}I<|Vu;Z3y$Fmk;z;V&4aVz6X(hk0LzhkgO<s4EE2q
zL}Q~p85(5C1c*i@V>wJtuy47FGEC8@WT6}j6iXYI!aW5Q_D?TJL_!ds0SGa-(BA|!
zaLPiy2qEx&2>K?5HaB0*N>Y@iGpr_%!B=RhZqG61B%Y6PLWib=o1@NOWgg@ac6%cB
zLti5{g!Foq!b9*4*%z^2W_5of{3re@-(G%C;G=k+ODXY}*ojdgi+WtTFukjUJPc`N
z4W#xY{6S<<LIvPawmpn&l+sPd(Bn+^4jY*EF@-lb*qyn}9j7O0!&VJ_x^9bc+`C2k
z7l^5$GHyFewpn-{aNtdwwM0tk2~_#P8-mBhC_)KSI3_pQ3<V2gDPp9QUj`FVFP!{u
z;{N~()&5ub7x+`;oTgsFF?bQ}3Y(eoJ<RC8vKuGad@A%?Y<flPBY*TeAM$(YN=HIo
zguM~ak|INCa)zzZ9&_G4mS?jKy9;cHuPGJZ^)O}s0A%<6WWVNmpV2+Pi4C7iO^~+`
zppCm7jPy$2Nj5}y5~6n^ZbOuLiSj=&(B(K3<s-b?-H7i00Fg15_9NHmp5M{bC2U^+
zuEx70irf)x*q1ZSiF~p|I6|EmK8xc>I<$n<GU+rz5t0yuGFCa@L?HGbN0K`cSdK`K
z11|H883{?`gd!sAL?H-GDRiYpLujaRi8@3$A|mL)f}#lQsqDr~>-IrWb*PTW()B${
z>q3ayPayMbp7gvYjSzYf(dc@Y(dyJG$)3#i$W2H>6IO&F88jgndq2_F;N#<EB@?tt
z=}8n4ZpiXQ8NqZ)K$NXG8nkbLk6}}6f>D_vVQDi*u#`Cv)fUC9T66J#25@2VN?(fn
zZ|3ypSN;R}FTlUWt-2}L@_!;{yJAvB&@Q9Ij{+hfq#H|j3Ge#`ul@+~&nfge*~Hz2
zo&u$^7kNeKw&TnBos0W9$VdMG32k3TLGk^Lr7P0<Jmh`FIbyBbF%on~NJuzjf=2Cr
z$nYO&(3=zR;(p(`AB2l=J6uW2limAdP9L+e{2dAYj~5%Z4g3d@4EGlI2bqhi`ys+F
zK~9EOrc9K`LK9LnMon?>Q?aCDM3*8i8Bw_JMn^{Bo)2Q6#q%O-cd@!Q<TIU#x;%`O
z#GsNsrVMFHT@mtSW_8D9MO~5EE0HH;b~SP|oSh6;;7#40OiGv=NN6>NOHnq;mf~-+
zDZ49{FCs`Mf)n9NUyLW`nqa`%8$$>lh1Eao{{TjMug3-~yTThbCW~X8In{5f=kTp~
z)1QkLi&weF&3qT)L;RCYf5czr?u^P++7SN$=uCV607Tp~8*piag*IQo{{U!Lb~#&q
z%oVNnC2TS$52Qo4^iOAJQyoFSV^0JB0FWDZVw;8LCG1@UiEu<gQ+9@)1dkJDT*_=+
za7DRDT%)o70Ar1B?1z7{D3-~7gs=W5h;G_Sa*<fycN22wu%8I%MmqRprY2?`2?!vI
zp`sW(i1Q!zno!4p@DE0hV?B_BAv_S8-vwP5d<d;@Cf@^1laYrbYr95EbVlGu8<I&c
zfi2*AiIz~X(vXiLF997DnIj`3D^I?pA|fK}$7J$HWWGozv&c@=N{HY{BoHv<XcV?H
zV9r?RqHIaBo9IY|@}+uYjD#UR6s0M2r7yrzl%*+3QL(YHv_S-urq|lP(fRyHMH(+B
z*~X7m`FZ&BqG#dA{tM;D&3qU6i~Mz~{{RtN$H-@)T%-FtKVXjPB<|m$JT`fcXNsLm
z`Y+??uP@l$yMD*G+3ZVpE+HvwB<Nlu7(BF1$UH(jU(r3^MO+28f3Y?R%#mHrY^Q(n
zC%^JB<rCpQq8l+h8tgB`DZq`m?{O>LwtI!pc#x9mX;~RGCPjQhZi>4{koFPNEO-#w
z^j99tkG6TSAtFN^M244WsmeTuxSklXBjF)FsPahkH*9$t#B$%EK80K+O6Fe!z>w4O
zLU!$w$sK7)+3pgdArXT7sY*vjsa+M(d^AqQ71>?r_8^GCf})gdm5s3lX(n#QF|i)R
zIU9Si%z{;<M1;~qDd_}=V_H({N>Y@%Qk0}>R#sA!rT7m*x+|iFCt<f^*Fca&e<^>8
z^<U(qHZAD#{%c3A(fl9HAsp++$!N|rdOwoUh+ma{XYpU?FYu+qv9FN(wk}F8fg^!+
zAt~@hB137z_C;*^A<ytT;=YEVTW{npsxQz@-hPiLM(zICle<37*X>H2uh7x|00J8a
z)yiyJctV^C`?-DHi?d?#61Y4^fe|fX8!JTAnj%}YLv$Z)JWP(oa2xnXl+Te(+t84N
zAhZ(ldlfdxdSbEYPYqdIvF!9fmz1=;20CP2X!=JxX)i>h@+X#Ov&2LnVq*E-Dj^L>
UC!$H{&tzX(PZ4?|^k1X@*;Pv=jsO4v

diff --git a/docs_old/cookbook/index.md b/docs_old/cookbook/index.md
deleted file mode 100644
index 931262053..000000000
--- a/docs_old/cookbook/index.md
+++ /dev/null
@@ -1,23 +0,0 @@
-# Examples
-
-This part of the documentation provides a few cookbooks that you can browse to get acquainted with the library and get some inspiration about what you could do with structured generation. Remember that you can easily change the model that is being used!
-
-- [Classification](classification.md): Classify customer requests.
-- [Named Entity Extraction](extraction.md): Extract information from pizza orders.
-- [Dating Profile](dating_profiles.md): Build dating profiles from descriptions using prompt templating and JSON-structured generation.
-- [Chain Of Density](chain_of_density.md): Summarize documents using chain of density prompting and JSON-structured generation.
-- [Playing Chess](models_playing_chess.md): Make Phi-3 Mini play chess against itself using regex-structured generation.
-- [SimToM](simtom.md): Improve LLMs' Theory of Mind capabilities with perspective-taking prompting and JSON-structured generation.
-- [Q&A with Citations](qa-with-citations.md): Answer questions and provide citations using JSON-structured generation.
-- [Knowledge Graph Generation](knowledge_graph_extraction.md): Generate a Knowledge Graph from unstructured text using JSON-structured generation.
-- [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
-- [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
-- [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
-- [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
-- [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.
-- [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
-- [Extract Event Details](extract_event_details.md): Use Outlines with MLX to extract key information from a text message related to an event.
-- [Generating Synthetic Phone Numbers](structured_generation_workflow.md): Use Outlines to create phone numbers that are valid for Washington State.
-- [Deploying Outlines using BentoML](deploy-using-bentoml.md): Deploy an Outlines model using BentoML.
-- [Deploying Outlines using Cerebras](deploy-using-cerebras.md): Deploy an Outlines model using Cerebras.
-- [Deploying Outlines using Modal](deploy-using-modal.md): Deploy an Outlines model using Modal.
diff --git a/docs_old/cookbook/knowledge_graph_extraction.md b/docs_old/cookbook/knowledge_graph_extraction.md
deleted file mode 100644
index f4b21c84c..000000000
--- a/docs_old/cookbook/knowledge_graph_extraction.md
+++ /dev/null
@@ -1,153 +0,0 @@
-# Knowledge Graph Extraction
-
-In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to extract a knowledge graph from unstructured text.
-
-We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
-
-```bash
-pip install llama-cpp-python
-```
-
-To create an outlines `LlamaCpp` model, you first need to create a `Llama` object from the `llama-cpp-python` library. Then you can create the outlines model by calling `models.from_llamacpp` with the `Llama` object instance as argument. To create the `Llama` object, you need to provide the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames or glob pattern (it will automatically download the weights from the hub):
-
-```python
-import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
-```
-
-??? note "(Optional) Store the model weights in a custom folder"
-
-    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
-
-    ```bash
-    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
-    ```
-
-    We initialize the model:
-
-    ```python
-    from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
-    ```
-
-## Knowledge Graph Extraction
-
-We first need to define our Pydantic class for each node and each edge of the knowledge graph:
-
-```python
-from pydantic import BaseModel, Field
-
-class Node(BaseModel):
-    """Node of the Knowledge Graph"""
-
-    id: int = Field(..., description="Unique identifier of the node")
-    label: str = Field(..., description="Label of the node")
-    property: str = Field(..., description="Property of the node")
-
-
-class Edge(BaseModel):
-    """Edge of the Knowledge Graph"""
-
-    source: int = Field(..., description="Unique source of the edge")
-    target: int = Field(..., description="Unique target of the edge")
-    label: str = Field(..., description="Label of the edge")
-    property: str = Field(..., description="Property of the edge")
-```
-
-We then define the Pydantic class for the knowledge graph and get its JSON schema:
-
-```python
-from typing import List
-
-class KnowledgeGraph(BaseModel):
-    """Generated Knowledge Graph"""
-
-    nodes: List[Node] = Field(..., description="List of nodes of the knowledge graph")
-    edges: List[Edge] = Field(..., description="List of edges of the knowledge graph")
-
-schema = KnowledgeGraph.model_json_schema()
-```
-
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
-
-```python
-from outlines import Template
-
-generate_hermes_prompt = Template.from_string(
-    """
-    <|im_start|>system
-    You are a world class AI model who answers questions in JSON
-    Here's the json schema you must adhere to:
-    <schema>
-    {{ schema }}
-    </schema>
-    <|im_end|>
-    <|im_start|>user
-    {{ user_prompt }}
-    <|im_end|>
-    <|im_start|>assistant
-    <schema>
-    """
-)
-```
-
-For a given user prompt, for example:
-
-```python
-user_prompt = "Alice loves Bob and she hates Charlie."
-```
-
-We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with the Hermes prompt:
-
-```python
-from outlines import Generator
-
-generator = Generator(model, KnowledgeGraph)
-prompt = generate_hermes_prompt(schema=schema, user_prompt=user_prompt)
-response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
-```
-
-We obtain the nodes and edges of the knowledge graph:
-
-```python
-print(response)
-# {"nodes":[{"id":1,"label":"Alice","property":"loves,hates"},
-# {"id":2,"label":"Bob","property":"loved_by"},
-# {"id":3,"label":"Charlie","property":"hated_by"}],
-# "edges":[{"source":1,"target":2,"label":"loves","property":"love"},
-# {"source":1,"target":3,"label":"hates","property":"hate"}]}
-
-```
-
-## (Optional) Visualizing the Knowledge Graph
-
-We can use the [Graphviz library](https://graphviz.readthedocs.io/en/stable/) to visualize the generated knowledge graph. For detailed installation instructions, see [here](https://graphviz.readthedocs.io/en/stable/#installation).
-
-```python
-from graphviz import Digraph
-
-dot = Digraph()
-for node in response["nodes"]:
-    dot.node(str(node["id"]), node["label"], shape='circle', width='1', height='1')
-for edge in response["edges"]:
-    dot.edge(str(edge["source"]), str(edge["target"]), label=edge["label"])
-
-dot.render('knowledge-graph.gv', view=True)
-```
-
-![Image of the Extracted Knowledge Graph](./images/knowledge-graph-extraction.png)
-
-This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_old/cookbook/models_playing_chess.md b/docs_old/cookbook/models_playing_chess.md
deleted file mode 100644
index b8d89958f..000000000
--- a/docs_old/cookbook/models_playing_chess.md
+++ /dev/null
@@ -1,90 +0,0 @@
-# Large language models playing chess
-
-In this example we will make a Phi-3 model play chess against itself. On its own the model easily generates invalid moves, so we will give it a little help. At each step we will generate a regex that only matches valid move, and use it to help the model only generating valid moves.
-
-## The chessboard
-
-The game will be played on a standard checkboard. We will use the `chess` [library](https://github.com/niklasf/python-chess) to track the opponents' moves, and check that the moves are valid.
-
-```python
-%pip install outlines -q
-%pip install chess -q
-%pip install transformers accelerate einops -q
-
-import chess
-
-board = chess.Board("rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1")
-```
-
-## The opponents
-
-Phi-3 will be playing against itself:
-
-```python
-import transformers
-import outlines
-
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name),
-    transformers.AutoTokenizer.from_pretrained(model_name),
-)
-```
-
-## A little help for the language model
-
-To make sure Phi-3 generates valid chess moves we will use Outline's regex-structured generation. We define a function that takes the current state of the board and returns a regex that matches all possible legal moves:
-
-```python
-import re
-from outlines.types.dsl import either, String
-
-def legal_moves_regex(board):
-    """Build a regex that only matches valid moves."""
-    legal_moves = list(board.legal_moves)
-    legal_modes_str = [board.san(move) for move in legal_moves]
-    legal_modes_str = [re.sub(r"[+#]", "", move) for move in legal_modes_str]
-    regex_pattern = either(*[String(move) for move in legal_modes_str])
-    return regex_pattern
-```
-
-## Prompting the language model
-
-The prompt corresponds to the current state of the board, so we start with:
-
-```python
-prompt = "Let's play Chess. Moves: "
-
-```
-
-We update the prompt at each step so it reflects the state of the board after the previous move.
-
-## Let's play
-
-```python
-board_state = " "
-turn_number = 0
-while not board.is_game_over():
-    regex_pattern = legal_moves_regex(board)
-    structured = model(prompt + board_state, regex_pattern)
-    move = board.parse_san(structured)
-
-    if turn_number % 2 == 0 :  # It's White's turn
-        board_state += board.san(move) + " "
-    else:
-        board_state += board.san(move) + " " + str(turn_number) + "."
-
-    turn_number += 1
-
-    board.push(move)
-
-    print(board_state)
-```
-
-Interestingly enough, Phi-3 hates capturing.
-
-```pgn
- e4 e5 1.Nf3 Ne7 3.b4 Nf5 5.Nc3 Ne7 7.Bb5 a6 9.Na4 b6 11.c3 Nec6 13.c4 a5 15.d4 Qg5 17.Nd2 Bb7 19.dxe5
-```
-
-*This example was originally authored by [@903124S](https://x.com/903124S) in [this gist](https://gist.github.com/903124/cfbefa24da95e2316e0d5e8ef8ed360d).*
diff --git a/docs_old/cookbook/qa-with-citations.md b/docs_old/cookbook/qa-with-citations.md
deleted file mode 100644
index 213942d02..000000000
--- a/docs_old/cookbook/qa-with-citations.md
+++ /dev/null
@@ -1,256 +0,0 @@
-# Generate Synthetic Data and Q&A with Citations
-
-This tutorial is adapted from the [instructor-ollama notebook](https://github.com/alonsosilvaallende/Hermes-Function-Calling/blob/main/examples/instructor_ollama.ipynb). We start with a simple example to generate synthetic data and then we approach the problem of question answering by providing citations.
-
-We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
-
-```bash
-pip install llama-cpp-python
-```
-
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
-```python
-import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
-```
-
-??? note "(Optional) Store the model weights in a custom folder"
-
-    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
-
-    ```bash
-    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
-    ```
-
-    We initialize the model:
-
-    ```python
-    from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
-    ```
-
-## Generate Synthetic Data
-
-We first need to define our Pydantic class for a user:
-
-```python
-from pydantic import BaseModel, Field
-
-class UserDetail(BaseModel):
-    id: int = Field(..., description="Unique identifier") # so the model keeps track of the number of users
-    first_name: str
-    last_name: str
-    age: int
-```
-
-We then define a Pydantic class for a list of users:
-
-```python
-from typing import List
-
-class Users(BaseModel):
-    users: List[UserDetail]
-```
-
-We can use a `outlines.Generator` by passing this Pydantic class we just defined, and call the generator:
-
-```python
-import json
-
-generator = outlines.Generator(model, Users)
-response = generator("Create 5 fake users", max_tokens=1024, temperature=0, seed=42)
-response = json.loads(response)
-print(response['users'])
-# [{'id': 1, 'first_name': 'John', 'last_name': 'Doe', 'age': 25},
-# {'id': 2, 'first_name': 'Jane', 'last_name': 'Doe', 'age': 30},
-# {'id': 3, 'first_name': 'Bob', 'last_name': 'Smith', 'age': 40},
-# {'id': 4, 'first_name': 'Alice', 'last_name': 'Smith', 'age': 35},
-# {'id': 5, 'first_name': 'John', 'last_name': 'Smith', 'age': 20}]
-```
-
-```python
-for user in response['users']:
-    print(user['first_name'])
-    print(user['last_name'])
-    print(user['age'])
-    print("#####")
-# John
-# Doe
-# 25
-# #####
-# Jane
-# Doe
-# 30
-# #####
-# Bob
-# Smith
-# 40
-# #####
-# Alice
-# Smith
-# 35
-# #####
-# John
-# Smith
-# 20
-# #####
-```
-
-## QA with Citations
-
-We first need to define our Pydantic class for QA with citations:
-
-```python
-from typing import List
-from pydantic import BaseModel
-
-class QuestionAnswer(BaseModel):
-    question: str
-    answer: str
-    citations: List[str]
-
-schema = QuestionAnswer.model_json_schema()
-```
-
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs):
-
-```python
-from outlines import Template
-
-hermes_prompt = Template.from_string(
-    """
-    <|im_start|>system
-    You are a world class AI model who answers questions in JSON with correct and exact citations
-    extracted from the `Context`.
-    Here's the json schema you must adhere to:
-    <schema>
-    {{ schema }}
-    </schema>
-    <|im_end|>
-    <|im_start|>user
-    `Context`:
-    {{ context }}
-    `Question`:
-    {{ question }}
-    <|im_end|>
-    <|im_start|>assistant
-    """
-)
-```
-
-We can use `outlines.Generator` by passing the Pydantic class we previously defined, and call the generator with Hermes prompt:
-
-```python
-question = "What did the author do during college?"
-context = """
-My name is Jason Liu, and I grew up in Toronto Canada but I was born in China.
-I went to an arts high school but in university I studied Computational Mathematics and physics.
-As part of coop I worked at many companies including Stitchfix, Facebook.
-I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years.
-"""
-generator = outlines.Generator(model, QuestionAnswer)
-prompt = hermes_prompt(question=question, context=context, schema=schema)
-response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
-print(response)
-# {"question": "What did the author do during college?", "answer": "The author studied Computational Mathematics and physics in university and was also involved in starting the Data Science club, serving as its president for 2 years.", "citations": ["I went to an arts high school but in university I studied Computational Mathematics and physics.", "I also started the Data Science club at the University of Waterloo and I was the president of the club for 2 years."]}
-```
-
-We can do the same for a list of question-context pairs:
-
-```python
-question1 = "Where was John born?"
-context1 = """
-John Doe is a software engineer who was born in New York, USA.
-He studied Computer Science at the Massachusetts Institute of Technology.
-During his studies, he interned at Google and Microsoft.
-He also founded the Artificial Intelligence club at his university and served as its president for three years.
-"""
-
-question2 = "What did Emily study in university?"
-context2 = """
-Emily Smith is a data scientist from London, England.
-She attended the University of Cambridge where she studied Statistics and Machine Learning.
-She interned at IBM and Amazon during her summer breaks.
-Emily was also the head of the Women in Tech society at her university.
-"""
-
-question3 = "Which companies did Robert intern at?"
-context3 = """
-Robert Johnson, originally from Sydney, Australia, is a renowned cybersecurity expert.
-He studied Information Systems at the University of Melbourne.
-Robert interned at several cybersecurity firms including NortonLifeLock and McAfee.
-He was also the leader of the Cybersecurity club at his university.
-"""
-
-question4 = "What club did Alice start at her university?"
-context4 = """
-Alice Williams, a native of Dublin, Ireland, is a successful web developer.
-She studied Software Engineering at Trinity College Dublin.
-Alice interned at several tech companies including Shopify and Squarespace.
-She started the Web Development club at her university and was its president for two years.
-"""
-
-question5 = "What did Michael study in high school?"
-context5 = """
-Michael Brown is a game developer from Tokyo, Japan.
-He attended a specialized high school where he studied Game Design.
-He later attended the University of Tokyo where he studied Computer Science.
-Michael interned at Sony and Nintendo during his university years.
-He also started the Game Developers club at his university.
-"""
-
-for question, context in [
-    (question1, context1),
-    (question2, context2),
-    (question3, context3),
-    (question4, context4),
-    (question5, context5),
-]:
-    prompt = hermes_prompt(question=question, context=context, schema=schema)
-    generator = outlines.Generator(model, QuestionAnswer)
-    response = generator(prompt, max_tokens=1024, temperature=0, seed=42)
-    response = json.loads(response)
-    print(question)
-    print(response['answer'])
-    print(response['citations'])
-    print("\n\n")
-
-# 'Where was John born?'
-# 'John Doe was born in New York, USA.'
-# ['John Doe is a software engineer who was born in New York, USA.']
-#
-#
-# 'What did Emily study in university?'
-# 'Emily studied Statistics and Machine Learning in university.'
-# ['She attended the University of Cambridge where she studied Statistics and Machine Learning.']
-#
-#
-# 'Which companies did Robert intern at?'
-# 'Robert interned at NortonLifeLock and McAfee.'
-# ['Robert Johnson, originally from Sydney, Australia, is a renowned cybersecurity expert. He interned at several cybersecurity firms including NortonLifeLock and McAfee.']
-#
-#
-# 'What club did Alice start at her university?'
-# 'Alice started the Web Development club at her university.'
-# ['Alice Williams, a native of Dublin, Ireland, is a successful web developer. She started the Web Development club at her university and was its president for two years.']
-#
-#
-# 'What did Michael study in high school?'
-# 'Michael studied Game Design in high school.'
-# ['Michael Brown is a game developer from Tokyo, Japan. He attended a specialized high school where he studied Game Design.']
-```
-
-This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_old/cookbook/react_agent.md b/docs_old/cookbook/react_agent.md
deleted file mode 100644
index bd3d51d6c..000000000
--- a/docs_old/cookbook/react_agent.md
+++ /dev/null
@@ -1,233 +0,0 @@
-# ReAct Agent
-
-This example shows how to use [outlines](https://dottxt-ai.github.io/outlines/) to build your own agent with open weights local models and structured outputs. It is inspired by the blog post [A simple Python implementation of the ReAct pattern for LLMs](https://til.simonwillison.net/llms/python-react-pattern) by [Simon Willison](https://simonwillison.net/).
-
-The ReAct pattern (for Reason+Act) is described in the paper [ReAct: Synergizing Reasoning and Acting in Language Models](https://arxiv.org/abs/2210.03629). It's a pattern where you implement additional actions that an LLM can take - searching Wikipedia or running calculations for example - and then teach it how to request the execution of those actions, and then feed their results back into the LLM.
-
-Additionally, we give the LLM the possibility of using a scratchpad described in the paper [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://arxiv.org/abs/2112.00114) which improves the ability of LLMs to perform multi-step computations.
-
-We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
-
-```bash
-pip install llama-cpp-python
-```
-
-We download the model weights by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
-```python
-import llama_cpp
-import outlines
-
-llm = llama_cpp.Llama(
-    "NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF",
-    tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(
-        "NousResearch/Hermes-2-Pro-Llama-3-8B"
-    ),
-    n_gpu_layers=-1,
-    flash_attn=True,
-    n_ctx=8192,
-    verbose=False
-)
-model = outlines.from_llamacpp(llm)
-```
-
-??? note "(Optional) Store the model weights in a custom folder"
-
-    By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
-
-    ```bash
-    wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
-    ```
-
-    We initialize the model:
-
-    ```python
-    from llama_cpp import Llama
-
-    llm = Llama("/path/to/model/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf", ...)
-    ```
-
-## Build a ReAct agent
-
-In this example, we use two tools:
-
-- wikipedia: \<search term\> - search Wikipedia and returns the snippet of the first result
-- calculate: \<expression\> - evaluate an expression using Python's eval() function
-
-```python
-import httpx
-
-def wikipedia(q):
-    return httpx.get("https://en.wikipedia.org/w/api.php", params={
-        "action": "query",
-        "list": "search",
-        "srsearch": q,
-        "format": "json"
-    }).json()["query"]["search"][0]["snippet"]
-
-
-def calculate(numexp):
-    return eval(numexp)
-```
-
-We define the logic of the agent through a Pydantic class. First, we want the LLM to decide only between the two previously defined tools:
-
-```python
-from enum import Enum
-
-class Action(str, Enum):
-    wikipedia = "wikipedia"
-    calculate = "calculate"
-```
-
-Our agent will loop through Thought and Action. We explicitly give the Action Input field so it doesn't forget to add the arguments of the Action. We also add a scratchpad (optional).
-
-```python
-from pydantic import BaseModel, Field
-
-class Reason_and_Act(BaseModel):
-    Scratchpad: str = Field(..., description="Information from the Observation useful to answer the question")
-    Thought: str = Field(..., description="It describes your thoughts about the question you have been asked")
-    Action: Action
-    Action_Input: str = Field(..., description="The arguments of the Action.")
-```
-
-Our agent will reach a Final Answer. We also add a scratchpad (optional).
-
-```python
-class Final_Answer(BaseModel):
-    Scratchpad: str = Field(..., description="Information from the Observation useful to answer the question")
-    Final_Answer: str = Field(..., description="Answer to the question grounded on the Observation")
-```
-
-Our agent will decide when it has reached a Final Answer and therefore to stop the loop of Thought and Action.
-
-```python
-from typing import Union
-
-class Decision(BaseModel):
-    Decision: Union[Reason_and_Act, Final_Answer]
-
-json_schema = Decision.model_json_schema()
-```
-
-We then need to adapt our prompt to the [Hermes prompt format for JSON schema](https://github.com/NousResearch/Hermes-Function-Calling?tab=readme-ov-file#prompt-format-for-json-mode--structured-outputs) and explain the agent logic. We can load a template from a file for that:
-
-```python
-from outlines import Template
-
-hermes_prompt = Template.from_file("prompt_templates/react_agent.txt")
-```
-
-We define a ChatBot class
-
-```python
-class ChatBot:
-    def __init__(self, prompt=""):
-        self.prompt = prompt
-
-    def __call__(self, user_prompt):
-        self.prompt += user_prompt
-        result = self.execute()
-        return result
-
-    def execute(self):
-        generator = outlines.Generator(model, Decision)
-        result = generator(self.prompt, max_tokens=1024, temperature=0, seed=42)
-        return result
-```
-
-We define a query function:
-
-```python
-import json
-
-def query(question, max_turns=5):
-    i = 0
-    next_prompt = (
-        "\n<|im_start|>user\n" + question + "<|im_end|>"
-        "\n<|im_start|>assistant\n"
-    )
-    previous_actions = []
-    while i < max_turns:
-        i += 1
-        prompt = generate_hermes_prompt(
-            question=question,
-            schema=Decision.model_json_schema(),
-            today=datetime.datetime.today().strftime('%Y-%m-%d')
-        )
-        bot = ChatBot(prompt=prompt)
-        result = bot(next_prompt)
-        json_result = json.loads(result)['Decision']
-        if "Final_Answer" not in list(json_result.keys()):
-            scratchpad = json_result['Scratchpad'] if i == 0 else ""
-            thought = json_result['Thought']
-            action = json_result['Action']
-            action_input = json_result['Action_Input']
-            print(f"\x1b[34m Scratchpad: {scratchpad} \x1b[0m")
-            print(f"\x1b[34m Thought: {thought} \x1b[0m")
-            print(f"\x1b[36m  -- running {action}: {str(action_input)}\x1b[0m")
-            if action + ": " + str(action_input) in previous_actions:
-                observation = "You already run that action. **TRY A DIFFERENT ACTION INPUT.**"
-            else:
-                if action=="calculate":
-                    try:
-                        observation = eval(str(action_input))
-                    except Exception as e:
-                        observation = f"{e}"
-                elif action=="wikipedia":
-                    try:
-                        observation = wikipedia(str(action_input))
-                    except Exception as e:
-                        observation = f"{e}"
-            print()
-            print(f"\x1b[33m Observation: {observation} \x1b[0m")
-            print()
-            previous_actions.append(action + ": " + str(action_input))
-            next_prompt += (
-                "\nScratchpad: " + scratchpad +
-                "\nThought: " + thought +
-                "\nAction: " + action  +
-                "\nAction Input: " + action_input +
-                "\nObservation: " + str(observation)
-            )
-        else:
-            scratchpad = json_result["Scratchpad"]
-            final_answer = json_result["Final_Answer"]
-            print(f"\x1b[34m Scratchpad: {scratchpad} \x1b[0m")
-            print(f"\x1b[34m Final Answer: {final_answer} \x1b[0m")
-            return final_answer
-    print(f"\nFinal Answer: I am sorry, but I am unable to answer your question. Please provide more information or a different question.")
-    return "No answer found"
-```
-
-We can now test our ReAct agent:
-
-```python
-print(query("What's 2 to the power of 10?"))
-# Scratchpad:
-# Thought: I need to perform a mathematical calculation to find the result of 2 to the power of 10.
-#  -- running calculate: 2**10
-#
-# Observation: 1024
-#
-# Scratchpad: 2 to the power of 10 is 1024.
-# Final Answer: 2 to the power of 10 is 1024.
-# 2 to the power of 10 is 1024.
-```
-
-```python
-print(query("What does England share borders with?"))
-# Scratchpad:
-# Thought: To answer this question, I will use the 'wikipedia' action to gather information about England's geographical location and its borders.
-#  -- running wikipedia: England borders
-#
-# Observation: Anglo-Scottish <span class="searchmatch">border</span> (Scottish Gaelic: Crìochan Anglo-Albannach) is an internal <span class="searchmatch">border</span> of the United Kingdom separating Scotland and <span class="searchmatch">England</span> which runs for
-#
-# Scratchpad: Anglo-Scottish border (Scottish Gaelic: Crìochan Anglo-Albannach) is an internal border of the United Kingdom separating Scotland and England which runs for
-# Final Answer: England shares a border with Scotland.
-# England shares a border with Scotland.
-```
-
-As mentioned in Simon's blog post, this is not a very robust implementation at all and there's a ton of room for improvement. But it is lovely how simple it is with a few lines of Python to make these extra capabilities available to the LLM. And now you can run it locally with an open weights LLM.
-
-This example was originally contributed by [Alonso Silva](https://github.com/alonsosilvaallende).
diff --git a/docs_old/cookbook/read-pdfs.md b/docs_old/cookbook/read-pdfs.md
deleted file mode 100644
index 1854aba42..000000000
--- a/docs_old/cookbook/read-pdfs.md
+++ /dev/null
@@ -1,367 +0,0 @@
-# PDF to structured output with vision language models
-
-A common task with language models is to ask language models questions about a PDF file.
-
-Typically, the output is unstructured text, i.e. "talking" to your PDF.
-
-In some cases, you may wish to extract structured information from the PDF, like tables, lists, citations, etc.
-
-PDFs are difficult to machine read. However, you can simply convert the PDF to images, and then use a vision language model to extract structured information from the images.
-
-This cookbook demonstrates how to
-
-1. Convert a PDF to a list of images
-2. Use a vision language model to extract structured information from the images
-
-## Dependencies
-
-You'll need to install these dependencies:
-
-```bash
-pip install outlines pillow transformers torch==2.4.0 pdf2image
-
-# Optional, but makes the output look nicer
-pip install rich
-```
-
-## Import the necessary libraries
-
-```python
-from PIL import Image
-import outlines
-import torch
-from transformers import AutoProcessor
-from pydantic import BaseModel
-from typing import List, Optional
-from pdf2image import convert_from_path
-import os
-from rich import print
-import requests
-```
-
-## Choose a model
-
-We've tested this example with [Pixtral 12b](https://huggingface.co/mistral-community/pixtral-12b) and [Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct).
-
-To use Pixtral:
-
-```python
-from transformers import LlavaForConditionalGeneration, LlavaProcessor
-model_name="mistral-community/pixtral-12b"
-model_class=LlavaForConditionalGeneration
-processor_class = LlavaProcessor
-```
-
-To use Qwen-2-VL:
-
-```python
-from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
-model_name = "Qwen/Qwen2-VL-7B-Instruct"
-model_class = Qwen2VLForConditionalGeneration
-processor_class = AutoProcessor
-```
-
-You can load your model into memory with:
-
-```python
-# This loads the model into memory. On your first run,
-# it will have to download the model, which might take a while.
-model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
-processor_kwargs={"device_map": "cpu"}
-tf_model = model_class.from_pretrained(model_name, **model_kwargs)
-tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-model = outlines.from_transformers(tf_model, tf_processor)
-```
-
-## Convert the PDF to images
-
-We'll use the `pdf2image` library to convert each page of the PDF to an image.
-
-`convert_pdf_to_images` is a convenience function that converts each page of the PDF to an image, and optionally saves the images to disk when `output_dir` is provided.
-
-Note: the `dpi` argument is important. It controls the resolution of the images. High DPI images are higher quality and may yield better results,
-but they are also larger, slower to process, and require more memory.
-
-```python
-from pdf2image import convert_from_path
-from PIL import Image
-import os
-from typing import List, Optional
-
-def convert_pdf_to_images(
-    pdf_path: str,
-    output_dir: Optional[str] = None,
-    dpi: int = 120,
-    fmt: str = 'PNG'
-) -> List[Image.Image]:
-    """
-    Convert a PDF file to a list of PIL Image objects.
-
-    Args:
-        pdf_path: Path to the PDF file
-        output_dir: Optional directory to save the images
-        dpi: Resolution for the conversion. High DPI is high quality, but also slow and memory intensive.
-        fmt: Output format (PNG recommended for quality)
-
-    Returns:
-        List of PIL Image objects
-    """
-    # Convert PDF to list of images
-    images = convert_from_path(
-        pdf_path,
-        dpi=dpi,
-        fmt=fmt
-    )
-
-    # Optionally save images
-    if output_dir:
-        os.makedirs(output_dir, exist_ok=True)
-        for i, image in enumerate(images):
-            image.save(os.path.join(output_dir, f'page_{i+1}.{fmt.lower()}'))
-
-    return images
-```
-
-We're going to use the [Louf & Willard paper](https://arxiv.org/pdf/2307.09702) that described the method that Outlines uses for structured generation.
-
-To download the PDF, run:
-
-```python
-# Download the PDF file
-pdf_url = "https://arxiv.org/pdf/2307.09702"
-response = requests.get(pdf_url)
-
-# Save the PDF locally
-with open("louf-willard.pdf", "wb") as f:
-    f.write(response.content)
-```
-
-Now, we can convert the PDF to a list of images:
-
-```python
-# Load the pdf
-images = convert_pdf_to_images(
-    "louf-willard.pdf",
-    dpi=120,
-    output_dir="output_images"
-)
-```
-
-## Extract structured information from the images
-
-The structured output you can extract is exactly the same as everywhere else in Outlines -- you can use regular expressions, JSON schemas, selecting from a list of options, etc.
-
-### Extracting data into JSON
-
-Suppose you wished to go through each page of the PDF, and extract the page description, key takeaways, and page number.
-
-You can do this by defining a JSON schema, and then using `outlines.Generator` to extract the data.
-
-First, define the structure you want to extract:
-
-```python
-class PageSummary(BaseModel):
-    description: str
-    key_takeaways: List[str]
-    page_number: int
-```
-
-Second, we need to set up the prompt. Adding special tokens can be tricky, so we use the transformers processor to apply the special tokens for us. To do so, we specify a list of messages, where each message is a dictionary with a `role` and `content` key.
-
-Images are denoted with `type: "image"`, and text is denoted with `type: "text"`.
-
-```python
-messages = [
-    {
-        "role": "user",
-        "content": [
-            # The text you're passing to the model --
-            # this is where you do your standard prompting.
-            {"type": "text", "text": f"""
-                Describe the page in a way that is easy for a PhD student to understand.
-
-                Return the information in the following JSON schema:
-                {PageSummary.model_json_schema()}
-
-                Here is the page:
-                """
-            },
-
-            # This a placeholder, the actual image is passed in when
-            # we call the generator function down below.
-            {"type": "image", "image": ""},
-        ],
-    }
-]
-
-# Convert the messages to the final prompt
-prompt = tf_processor.apply_chat_template(
-    messages, tokenize=False, add_generation_prompt=True
-)
-```
-
-Now we iterate through each image, and extract the structured information:
-
-```python
-# Page summarizer function
-page_summary_generator = outlines.Generator(model, PageSummary)
-
-for image in images:
-    result = page_summary_generator({"text": prompt, "images": image})
-    print(result)
-```
-
-### Regular expressions to extract the arxiv paper identifier
-
-The [arXiv paper identifier](https://info.arxiv.org/help/arxiv_identifier.html) is a unique identifier for each paper. These identifiers have the format `arXiv:YYMM.NNNNN` (five end digits) or `arXiv:YYMM.NNNN` (four end digits). arXiv identifiers are typically watermarked on papers uploaded to arXiv.
-
-arXiv identifiers are optionally followed by a version number, i.e. `arXiv:YYMM.NNNNNvX`.
-
-We can use a regular expression to define this patter:
-
-```python
-from outlines.types import Regex
-
-paper_regex = Regex(r'arXiv:\d{2}[01]\d\.\d{4,5}(v\d)?')
-```
-
-We can build an extractor function from the regex:
-
-```python
-id_extractor = outlines.Generator(model, paper_regex)
-```
-
-Now, we can extract the arxiv paper identifier from the first image:
-
-```python
-arxiv_instruction = tf_processor.apply_chat_template(
-    [
-        {
-            "role": "user",
-            "content": [
-                {"type": "text", "text": f"""
-                Extract the arxiv paper identifier from the page.
-
-                Here is the page:
-                """},
-                {"type": "image", "image": ""},
-            ],
-        }
-    ],
-    tokenize=False,
-    add_generation_prompt=True
-)
-
-# Extract the arxiv paper identifier
-paper_id = id_extractor({"text": arxiv_instruction, "images": images[0]})
-```
-
-As of the time of this writing, the arxiv paper identifier is
-
-```
-arXiv:2307.09702v4
-```
-
-Your version number may be different, but the part before `vX` should match.
-
-### Categorize the paper into one of several categories
-
-`outlines.Generator` also allows the model to select one of several options by providing a Literal type hint with the categories.
-
-Suppose we wanted to categorize the paper into being about "language models", "cell biology", or "other". We would then define the output type as `Literal["llms", "cell biology", "other"]`.
-
-Let's define a few categories we might be interested in:
-
-```python
-categories = [
-    "llms",
-    "cell biology",
-    "other"
-]
-```
-
-Now we can construct the prompt:
-
-```python
-categorization_instruction = tf_processor.apply_chat_template(
-    [
-        {
-            "role": "user",
-            "content": [
-                {"type": "text", "text": f"""
-                Please choose one of the following categories
-                that best describes the paper.
-
-                {categories}
-
-                Here is the paper:
-                """},
-
-                {"type": "image", "image": ""},
-            ],
-        }
-    ],
-    tokenize=False,
-    add_generation_prompt=True
-)
-```
-
-Now we can show the model the first page and extract the category:
-
-```python
-from typing import Literal
-
-# Build the choice extractor
-categorizer = outlines.Generator(model, Literal["llms", "cell biology", "other"])
-
-# Categorize the paper
-category = categorizer({"text": categorization_instruction, "images": images[0]})
-print(category)
-```
-
-Which should return:
-
-```
-llms
-```
-
-## Additional notes
-
-You can provide multiple images to the model by
-
-1. Adding additional image messages
-2. Providing a list of images to the generator
-
-For example, to have two images, you can do:
-
-```python
-two_image_prompt = tf_processor.apply_chat_template(
-    [
-        {
-            "role": "user",
-            "content": [
-                {"type": "text", "text": "are both of these images of hot dogs?"},
-
-                # Tell the model there are two images
-                {"type": "image", "image": ""},
-                {"type": "image", "image": ""},
-            ],
-        }
-    ],
-    tokenize=False,
-    add_generation_prompt=True
-)
-
-# Pass two images to the model
-generator = outlines.Generator(model, Literal["hot dog", "not hot dog"])
-
-result = generator({"text": two_image_prompt, "images": [images[0], images[1]]})
-print(result)
-```
-
-Using the first to pages of the paper (they are not images of hot dogs), we should get
-
-```
-not hot dog
-```
diff --git a/docs_old/cookbook/receipt-digitization.md b/docs_old/cookbook/receipt-digitization.md
deleted file mode 100644
index 8ecb07969..000000000
--- a/docs_old/cookbook/receipt-digitization.md
+++ /dev/null
@@ -1,287 +0,0 @@
-# Receipt Data Extraction with VLMs
-
-## Setup
-
-You'll need to install the dependencies:
-
-```bash
-pip install outlines torch==2.4.0 transformers accelerate pillow rich
-```
-
-## Import libraries
-
-Load all the necessary libraries:
-
-```python
-# LLM stuff
-import outlines
-import torch
-from transformers import AutoProcessor
-from pydantic import BaseModel, Field
-from typing import Literal, Optional, List
-
-# Image stuff
-from PIL import Image
-import requests
-
-# Rich for pretty printing
-from rich import print
-```
-
-## Choose a model
-
-This example has been tested with `mistral-community/pixtral-12b` ([HF link](https://huggingface.co/mistral-community/pixtral-12b)) and `Qwen/Qwen2-VL-7B-Instruct` ([HF link](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct)).
-
-We recommend Qwen-2-VL as we have found it to be more accurate than Pixtral.
-
-If you want to use Qwen-2-VL, you can do the following:
-
-```python
-# To use Qwen-2-VL:
-from transformers import Qwen2VLForConditionalGeneration, AutoProcessor
-model_name = "Qwen/Qwen2-VL-7B-Instruct"
-model_class = Qwen2VLForConditionalGeneration
-processor_class = AutoProcessor
-```
-
-If you want to use Pixtral, you can do the following:
-
-```python
-# To use Pixtral:
-from transformers import LlavaForConditionalGeneration, LlavaProcessor
-model_name="mistral-community/pixtral-12b"
-model_class=LlavaForConditionalGeneration
-processor_class = LlavaProcessor
-```
-
-## Load the model
-
-Load the model into memory:
-
-```python
-model_kwargs={"device_map": "auto", "torch_dtype": torch.bfloat16}
-processor_kwargs={"device_map": "cuda"}
-tf_model = model_class.from_pretrained(model_name, **model_kwargs)
-tf_processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-
-model = outlines.from_transformers(tf_model, tf_processor)
-```
-
-## Image processing
-
-Images can be quite large. In GPU-poor environments, you may need to resize the image to a smaller size.
-
-Here's a helper function to do that:
-
-```python
-def load_and_resize_image(image_path, max_size=1024):
-    """
-    Load and resize an image while maintaining aspect ratio
-
-    Args:
-        image_path: Path to the image file
-        max_size: Maximum dimension (width or height) of the output image
-
-    Returns:
-        PIL Image: Resized image
-    """
-    image = Image.open(image_path)
-
-    # Get current dimensions
-    width, height = image.size
-
-    # Calculate scaling factor
-    scale = min(max_size / width, max_size / height)
-
-    # Only resize if image is larger than max_size
-    if scale < 1:
-        new_width = int(width * scale)
-        new_height = int(height * scale)
-        image = image.resize((new_width, new_height), Image.Resampling.LANCZOS)
-
-    return image
-```
-
-You can change the resolution of the image by changing the `max_size` argument. Small max sizes will make the image more blurry, but processing will be faster and require less memory.
-
-## Load an image
-
-Load an image and resize it. We've provided a sample image of a Trader Joe's receipt, but you can use any image you'd like.
-
-Here's what the image looks like:
-
-![Trader Joe's receipt](./images/trader-joes-receipt.jpg)
-
-```python
-# Path to the image
-image_path = "https://raw.githubusercontent.com/dottxt-ai/outlines/refs/heads/main/docs/cookbook/images/trader-joes-receipt.jpg"
-
-# Download the image
-response = requests.get(image_path)
-with open("receipt.png", "wb") as f:
-    f.write(response.content)
-
-# Load + resize the image
-image = load_and_resize_image("receipt.png")
-```
-
-## Define the output structure
-
-We'll define a Pydantic model to describe the data we want to extract from the image.
-
-In our case, we want to extract the following information:
-
-- The store name
-- The store address
-- The store number
-- A list of items, including the name, quantity, price per unit, and total price
-- The tax
-- The total
-- The date
-- The payment method
-
-Most fields are optional, as not all receipts contain all information.
-
-```python
-class Item(BaseModel):
-    name: str
-    quantity: Optional[int]
-    price_per_unit: Optional[float]
-    total_price: Optional[float]
-
-class ReceiptSummary(BaseModel):
-    store_name: str
-    store_address: str
-    store_number: Optional[int]
-    items: List[Item]
-    tax: Optional[float]
-    total: Optional[float]
-    # Date is in the format YYYY-MM-DD. We can apply a regex pattern to ensure it's formatted correctly.
-    date: Optional[str] = Field(pattern=r'\d{4}-\d{2}-\d{2}', description="Date in the format YYYY-MM-DD")
-    payment_method: Literal["cash", "credit", "debit", "check", "other"]
-```
-
-## Prepare the prompt
-
-We'll use the `tf_processor` to convert the image and the text prompt into a format that the model can understand. Practically,
-this is the code that adds user, system, assistant, and image tokens to the prompt.
-
-```python
-# Set up the content you want to send to the model
-messages = [
-    {
-        "role": "user",
-        "content": [
-            {
-                # The image is provided as a PIL Image object
-                "type": "image",
-                "image": image,
-            },
-            {
-                "type": "text",
-                "text": f"""You are an expert at extracting information from receipts.
-                Please extract the information from the receipt. Be as detailed as possible --
-                missing or misreporting information is a crime.
-
-                Return the information in the following JSON schema:
-                {ReceiptSummary.model_json_schema()}
-            """},
-        ],
-    }
-]
-
-# Convert the messages to the final prompt
-prompt = tf_processor.apply_chat_template(
-    messages, tokenize=False, add_generation_prompt=True
-)
-```
-
-If you are curious, the final prompt that is sent to the model looks (roughly) like this:
-
-```
-<|im_start|>system
-You are a helpful assistant.<|im_end|>
-<|im_start|>user
-<|vision_start|><|image_pad|><|vision_end|>
-You are an expert at extracting information from receipts.
-Please extract the information from the receipt. Be as detailed as
-possible -- missing or misreporting information is a crime.
-
-Return the information in the following JSON schema:
-
-<JSON SCHEMA OMITTED>
-<|im_end|>
-<|im_start|>assistant
-```
-
-## Run the model
-
-```python
-# Prepare a function to process receipts
-receipt_summary_generator = outlines.Generator(model, ReceiptSummary)
-
-# Generate the receipt summary
-result = receipt_summary_generator(
-    {"text": prompt, "images": image},
-    max_new_tokens=1024
-)
-print(result)
-```
-
-## Output
-
-The output should look like this:
-
-```
-{
-  "store_name": "Trader Joe's",
-  "store_address": "401 Bay Street, San Francisco, CA 94133",
-  "store_number": 0,
-  "items": [
-    {"name": "BANANA EACH", "quantity": 7, "price_per_unit": 0.23, "total_price": 1.61},
-    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "BAREBELLS CHOCOLATE DOUG", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "BAREBELLS CARAMEL CASHEW", "quantity": 2, "price_per_unit": 2.29, "total_price": 4.58},
-    {"name": "BAREBELLS CREAMY CRISP", "quantity": 1, "price_per_unit": 2.29, "total_price": 2.29},
-    {"name": "SPINDRIFT ORANGE MANGO 8", "quantity": 1, "price_per_unit": 7.49, "total_price": 7.49},
-    {"name": "Bottle Deposit", "quantity": 8, "price_per_unit": 0.05, "total_price": 0.4},
-    {"name": "MILK ORGANIC GALLON WHOL", "quantity": 1,"price_per_unit": 6.79,"total_price": 6.79},
-    {"name": "CLASSIC GREEK SALAD", "quantity": 1, "price_per_unit": 3.49, "total_price": 3.49},
-    {"name": "COBB SALAD", "quantity": 1, "price_per_unit": 5.99, "total_price": 5.99},
-    {"name": "PEPPER BELL RED XL EACH", "quantity": 1, "price_per_unit": 1.29, "total_price": 1.29},
-    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
-    {"name": "BAG FEE.", "quantity": 1, "price_per_unit": 0.25, "total_price": 0.25},
-  ],
-  "tax": 0.68,
-  "total": 41.98,
-  "date": "2023-11-04",
-  "payment_method": "debit"
-}
-```
-
-Voila! You've successfully extracted information from a receipt using an LLM.
-
-## Bonus: roasting the user for their receipt
-
-You can roast the user for their receipt by adding a `roast` field to the end of the  `ReceiptSummary` model.
-
-```python
-class ReceiptSummary(BaseModel):
-    ...
-    roast: str
-```
-
-which gives you a result like
-
-```
-{
-    ...
-    "roast": "You must be a fan of Trader Joe's because you bought enough
-    items to fill a small grocery bag and still had to pay for a bag fee.
-    Maybe you should start using reusable bags to save some money and the
-    environment."
-}
-```
-
-Qwen is not particularly funny, but worth a shot.
diff --git a/docs_old/cookbook/simtom.md b/docs_old/cookbook/simtom.md
deleted file mode 100644
index 53820258e..000000000
--- a/docs_old/cookbook/simtom.md
+++ /dev/null
@@ -1,111 +0,0 @@
-# Build perspective-taking agents with SimToM
-
-Prompting strategies like Chain-of-Thought (CoT) can improve LLMs' reasoning capabilities. However, they underwhelm in tasks that require keeping track of inconsistent world states. [SimToM](https://arxiv.org/abs/2311.10227) proposes a simple, two-stage prompting framework for LLMs inspired by Simulation Theory. The authors showed that this approach outperforms zero-shot prompting and CoT on ToMI and BigToM, two benchmarks with Theory of Mind questions.
-
-In this example, we will implement SimToM with a few lines of code using Outlines' prompt templating and structured generation capabilities.
-
-## How SimToM works
-
-SimToM calls an LLM with two consecutive prompts:
-
-1. **Perspective-taking**: The first prompt receives a `story` and a `character`. The goal is to understand the situation based on the character's point of view and filter out the rest of the story.
-2. **Question-Answering**: The second prompt receives the character's point of view from the previous step and tasks the LLM to answer a question using that context.
-
-![Figure 2 in the paper](./images/simtom.png)
-
-## Outlines implementation
-
-To implement SimToM with Outlines, we will need to:
-
-1. Write the prompts with [prompt templates](https://dottxt-ai.github.io/outlines/latest/reference/prompting/).
-2. Define the JSON object each prompt will return using Pydantic.
-3. Generate responses with a Mistral model using the [transformers integration](https://dottxt-ai.github.io/outlines/latest/reference/models/transformers/).
-
-Let's dive into it!
-
-### Using Prompt Templates
-
-The authors have shared their code, prompts and data in [this GitHub repository](https://github.com/shawnsihyunlee/simulatedtom). Below, we define in Outlines the prompts they used for the ToMI dataset:
-
-```python
-from outlines import Template
-
-perspective_taking = Template.from_file("prompt_templates/simtom_prospective_taking.txt")
-simulation = Template.from_file("prompt_templates/simtom_simulation.txt")
-```
-
-### JSON Structured Generation
-
-Outlines guarantees that the LLM will return a valid JSON object, which we can specify as a Pydantic model.
-
-We will need two Pydantic models for SimToM, one for each prompt:
-
-```python
-from pydantic import BaseModel, Field
-from typing import List
-
-class PerspectiveTaking(BaseModel):
-    """This is for the first prompt."""
-    character: str = Field(description="The character we extract the events for.")
-    events: List[str] = Field(description="All events that the character knows about.")
-
-class Simulation(BaseModel):
-    """This is for the second prompt."""
-    answer: str
-```
-
-### Calling an LLM
-
-Let's try SimToM with an example from the ToMI dataset:
-
-```python
-story = """
-1 Aria entered the front_yard.
-2 Aiden entered the front_yard.
-3 The grapefruit is in the green_bucket.
-4 Aria moved the grapefruit to the blue_container.
-5 Aiden exited the front_yard.
-6 Noah entered the playroom.
-"""
-question = "7 Where was the grapefruit at the beginning?"
-character = "Aria"
-```
-
-We load `Mistral-7B-Instruct-v0.3`, create the prompt using the template we defined earlier, and generate a structured response. As a reminder, the goal of the first call is to get all the events a character, `Aria`, knows about.
-
-```python
-import transformers
-import outlines
-# Load an LLM from Hugging Face
-MODEL_NAME = "mistral-community/Mistral-7B-Instruct-v0.3"
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
-    transformers.AutoTokenizer.from_pretrained(MODEL_NAME),
-)
-
-perspective_prompt = perspective_taking(story=story, character=character)
-
-# Call Mistral 7B with the first prompt
-generator = outlines.Generator(model, PerspectiveTaking)
-perspective = generator(perspective_prompt, max_new_tokens=1024)
-
-print(perspective)
-# {'character': 'Aria', 'events': ['1 Aria entered the front_yard.', '3 The grapefruit is in the green_bucket.', '4 Aria moved the grapefruit to the blue_container.']}
-```
-
-Not bad! We will now generate the second prompt with those events.
-
-```python
-import json
-
-sim_prompt = simulation(events=json.loads(perspective)["events"], name=character, question=question)
-
-# Call Mistral 7B with the second prompt
-generator = outlines.Generator(model, Simulation)
-result = generator(sim_prompt, max_new_tokens=1024)
-
-print(result)
-# {'answer': 'green_bucket'}
-```
-
-And this is it! SimToM could be useful in agentic workflows, where agents must act based on what they know, not all available information. One caveat of SimToM is that the perspective-taking step may remove important information, leading to wrong results. As the authors note in their paper, it can feature as a simple and effective baseline for evaluating LLMs on Theory of Mind reasoning tasks.
diff --git a/docs_old/cookbook/structured_generation_workflow.md b/docs_old/cookbook/structured_generation_workflow.md
deleted file mode 100644
index 524283f84..000000000
--- a/docs_old/cookbook/structured_generation_workflow.md
+++ /dev/null
@@ -1,211 +0,0 @@
-# Structured Generation Workflow: Generating Synthetic Phone Numbers
-
-This is a condensed version of [Coding for Structured Generation with LLMs](https://blog.dottxt.co/coding-for-structured-generation.html).
-
-For this example we're going to be building an LLM program to generate **synthetic data** in the form of realistic looking phone numbers for Washington State. Using an LLM for this task *is a bit overkill* since we could just as easily accomplish this with a tool like [Faker](https://fakerjs.dev/), but this example still serves as a useful way to demonstrate a workflow for using structured generation.
-
-## Unstructured approach
-
-Before diving into how to use structure generation for this task let's start with an unstructured example. We begin by loading our model:
-
-```python
-import outlines
-import transformers
-
-model_name = 'microsoft/Phi-3-mini-4k-instruct'
-model = outlines.from_transformers(
-    transformers.AutoModelForCausalLM.from_pretrained(model_name),
-    transformers.AutoTokenizer.from_pretrained(model_name)
-)
-```
-
-Next we need a prompt for this model. Since we're focusing on structured generation, we won't be engaging in any form of "prompt hacking" and will be leaving this prompt untouched for the rest of this example.
-
-```python
-prompt_phone = """
-    Please generate a realistic phone number for Washington State in the following format
-    (555) 555-5555
-"""
-```
-
-With our prompt ready we can now generate 10 example phone numbers
-
-```python
-phone_generator_unstruct = outlines.Generator(model)
-for _ in range(3):
-    print(phone_generator_unstruct(prompt_phone, max_new_tokens=12))
-```
-
-> I'd be happy to help you generate a realistic phone\
-I cannot generate a real phone number as I'm just\
-I'm an AI and don't have the ability\
-Sure! Here is a randomly generated phone number in the format\
-Here's a phone number that fits the format for a\
-In Washington State, phone numbers typically have a three-dig\
-Here are a few examples of phone numbers that could be considered\
-I'd be happy to help generate a realistic phone number\
-I'd be happy to help you generate a random phone\
-Based on the format you provided, a realistic phone number for\
-
-As we can see, none of these outputs are even phone numbers!
-
-Let's see  if we can improve this using structured generation.
-
-## The Structured Generation Workflow
-
-In order to solve this problem we're going to introduce a *Structured Generation Workflow* outlined in this image:
-
-!["Visual of Structured Generation Workflow"](./images/coding_structure_diagram.png)
-
-Let's step through this:
-
-### Real example
-
-We start with a real example phone number, in this case for the Seattle Public Library, that we can use to verify the structure we are creating.
-
-```python
-phone_number = "(206) 386-4636"
-```
-
-For a simple example like this, we'll just be using a single phone number, for more complex examples it can be helpful to have more examples.
-
-### Draft Structure
-
-The next step in the process is for use to define a simple regex that we feel correctly models our real data.
-
-```python
-from outlines.types import Regex
-
-phone_regex_1 = Regex(r'\([0-9]{3}\) [0-9]{3}-[0-9]{4}')
-```
-
-Next we need to validate this regex against our real data.
-
-### Validate by matching examples
-
-Whenever writing non-trivial code with structured generation it is *essential* that you first validate the code against your real data example(s).
-
-We'll start with a simple method of validation: just checking that our regex matches the data.
-
-```
-import re
-
-re.match(phone_regex_1.pattern, phone_number)
-# <re.Match object; span=(0, 14), match='(206) 386-4636'>
-
-```
-
-Now that we have a match, we can move on to generating structured output!
-
-### Generate Structure
-
-We're ready to see if structured generation can make an improvement over our initial unstructured approach:
-
-```python
-phone_generator_v1 = outlines.Generator(model, phone_regex_1)
-
-for _ in range(3):
-    print(phone_generator_v1(prompt_phone))
-```
-> (206) 555-1234\
-(206) 555-1234\
-(206) 555-1234\
-(206) 555-1234\
-(206) 555-1234\
-(206) 555-1234\
-(206) 123-4567\
-(206) 555-1234\
-(206) 555-1234\
-(206) 555-1234
-
-At least we have phone numbers! But I think we can do better!
-
-### Inspect output
-
-In this case the model *did* create phone numbers and, impressively, got the area code correct. So using structured generation did improve things. However these numbers are pretty boring. Let's improve that structure!
-
-## Iteration
-
-We've walked through the loop once, so we can go quickly now through each iteration.
-
-We start by improving our structure:
-
-```python
-phone_regex_2 = Regex(r'\([0-9]{3}\) [2-46-9]{3}-[02-9]{4}')
-```
-
-Before rushing to another round of generation, let's validate this new regex. We'll add just a bit more sophistication over our last check:
-
-```python
-re.match(phone_regex_2.pattern, phone_number)[0] == phone_number
-# True
-```
-Now that we've validated, let's generate with this new regex!
-
-```python
-phone_generator_v2 = outlines.Generator(model, phone_regex_2)
-
-for _ in range(3):
-    print(phone_generator_v2(prompt_phone))
-```
-
-> (206) 867-5309\
-(206) 666-7777\
-(206) 444-3333\
-(206) 444-3333\
-(206) 943-2222\
-(206) 323-6789\
-(206) 444-3333\
-(206) 867-5309\
-(206) 466-2255\
-(206) 222-3333
-
-Better, but I don't like those repeated sequences. Like good software developers, let's iterate again!
-
-## Reiteration - with debugging
-
-Here's a fancier regex that should give us more interesting results:
-
-```python
-phone_regex_3_error = r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4]'
-```
-
-This looks good to me, but there's a subtle bug, that's why we *always* need to validate our structure against real data. This time we'll make our validator do a bit more work to verify the correct string is matched:
-
-```python
-if not re.match(phone_regex_3_error, phone_number):
-    print("Regex fails match")
-else:
-    matched_string = re.match(phone_regex_3_error, phone_number)[0]
-    if matched_string == phone_number:
-    print("Successful match")
-    else:
-    print(f"Error {matched_string} != {phone_number}")
-```
-This prints out:
->  Error (206) 386-463 != (206) 386-4636
-
-Ah! We were missing the last digit, let's fix that and regenerate:
-
-```python
-phone_regex_3_fixed = Regex(r'\([0-9]{3}\) [2-4][7-9][4-6]-[3-6][2-8][1-4][6-9]')
-phone_generator_v3 = outlines.Generator(model, phone_regex_3_fixed)
-
-for _ in range(3):
-    print(phone_generator_v3(prompt_phone))
-```
-
->(206) 494-3216\
-(206) 374-6218\
-(206) 494-3337\
-(206) 476-3216\
-(206) 484-3548\
-(206) 495-3218\
-(206) 494-5517\
-(206) 375-4636\
-(206) 384-6216\
-(206) 385-6218
-
-Much better!
-
-Now you've seen a quick example of the structured generation workflow that can be used at the basis for building and iteration on much larger structured generation tasks!
diff --git a/docs_old/index.md b/docs_old/index.md
deleted file mode 100644
index 3692d32c9..000000000
--- a/docs_old/index.md
+++ /dev/null
@@ -1,8 +0,0 @@
----
-title: Outlines
-template: home.html # Note that this is managed in overrides/home.html
-hide:
-  - navigation
-  - toc
-  - feedback
----
diff --git a/docs_old/installation.md b/docs_old/installation.md
deleted file mode 100644
index 04ecd8237..000000000
--- a/docs_old/installation.md
+++ /dev/null
@@ -1,48 +0,0 @@
----
-title: Installation
----
-
-# Installation
-
-You can install Outlines with `pip`:
-
-```sh
-pip install outlines
-```
-
-Outlines supports OpenAI, Transformers, Mamba and llama.cpp, but **you will need to install them manually**:
-
-```sh
-pip install openai
-pip install transformers datasets accelerate torch
-pip install llama-cpp-python
-pip install transformers torch
-pip install mamba_ssm transformers torch
-pip install vllm
-```
-
-If you encounter any problems using Outlines with these libraries, take a look at their installation instructions. The installation of `openai` and `transformers` should be straightforward, but other libraries have specific hardware requirements.
-
-## Optional Dependencies
-
-Outlines provides multiple optional dependency sets to support different backends and use cases. You can install them as needed using:
-
-- `pip install "outlines[vllm]"` for [vLLM](https://github.com/vllm-project/vllm), optimized for high-throughput inference.
-- `pip install "outlines[transformers]"` for [Hugging Face Transformers](https://huggingface.co/docs/transformers/index).
-- `pip install "outlines[mlx]"` for [MLX-LM](https://github.com/ml-explore/mlx-lm), optimized for Apple silicon.
-- `pip install "outlines[openai]"` to use OpenAI’s API.
-- `pip install "outlines[llamacpp]"` for [llama.cpp](https://github.com/ggerganov/llama.cpp), a lightweight LLM inference engine.
-
-## Bleeding Edge
-
-You can install the latest version of Outlines from the repository's `main` branch:
-
-```sh
-pip install git+https://github.com/dottxt-ai/outlines.git@main
-```
-
-This can be useful, for instance, when a fix has been merged but not yet released.
-
-## Installing for Development
-
-See the [contributing documentation](community/contribute.md) for instructions on how to install Outlines for development, including an example using the `dot-install` method for one of the backends.
diff --git a/docs_old/licence.md b/docs_old/licence.md
deleted file mode 100644
index a74661fa3..000000000
--- a/docs_old/licence.md
+++ /dev/null
@@ -1,34 +0,0 @@
----
-title: Licence
----
-
-# Licence and citations
-
-Outlines is licenced under the Apache 2.0 licence. To comply with the licence you need to add the following notice at the top every file that uses part of Outlines' code:
-
-```
-Copyright 2023- The Outlines developers
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-```
-
-If you use Outlines in your work you can use the following citation:
-
-```
-@article{willard2023efficient,
-  title={Efficient Guided Generation for LLMs},
-  author={Willard, Brandon T and Louf, R{\'e}mi},
-  journal={arXiv preprint arXiv:2307.09702},
-  year={2023}
-}
-```
diff --git a/docs_old/logos/amazon.png b/docs_old/logos/amazon.png
deleted file mode 100644
index 85e1dc81c405d777b6c5ce1a9d571e87b892d91a..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 19489
zcmYg%by!s2_w_JzclSt13lfq7gCHd_bW14>(%l^*EiE80ba!`)Fw$Kgx}@uO{XXv>
z?=w8a%y8%2bM{$#@3q#6R8^6~#iGIjfk3$L<)P{z5I6w%y&D4!_-?F))C>GTv651j
z0)eXIu^&uPfxj8d<kgiyATJgWC?Es`x&xjH_zwhf<pqKEO+X-#G!Tf~F|$QY40r+U
zqoN!X^!)FU+g_9i0x1f=hf2Xb7LKw!eGD|0pI28!6JHw!RKR|lACEEBy?y(YjUmH=
zP^u3VMF?en0kQ^`=-S*=Z`@57lzR;1v_sN0canKK4-+Ir35~)s<xl7OoQu8u-!~Ti
z>-&4Npqje6_m&e`={~EK+QuYl-ji42s~2x}voFM@Eg<9L<D!UYCKyxPP}~m@S<{*b
zB_yaOE66pt)A$7>5MCj>Wf7FCEq20;8^nAW+EY8prF+g7W(s2-Hg`tXjFp3Jq9Rz-
zvVwM%-?>Ib#9#*H8bJrh#lvk${*$TaA<$(${r1)Y$ta2pD=TYx`7=Y6o&b(n-3wEz
z-1hc%`^zZO6Y0w=_8QhI@?U5%FHQM)kUr$!y@vS5;Ngh<m3BD6vhj1TYiNkW6tr8A
zfg~&=Bba!nf+N}a2ocq9C99}(xT^GITk2K+qnMq&HXl6SK47mmIP(zWmO52AYg-PC
z#7Y!PD~4|ps0AY6Eh-GO#S$Nw&NFjVBrh1Y#@PkUSjWe?P!rs}A>N;-$PY1<xJ-$t
z1FT^6GSH`u5{bF=QE@W?2N++i<@iEHXQ#h5X3#R>ncA|7i73?V7sDJ9A^v~JMrsM#
zC|V`^9L1I|Acl|d$!+ImO|3-6`$kIs?{yXS?wZTE2ur9N)OFwFo)L~J=5SPMh6dq@
z*q0nsN;`FtlaupT#^_4f*;z)Oyw8mod*|wJ(zeOT_A3u^{lZyMB+GA33svC*B;gb@
zC?1Oj_YggXcEIKwD<0Y1E3{{_#mh^imveedhw}}?IpLaPpgW#MW*KP$I%be}Noi|&
zX=$l<-GaP^#`M;!ssCOT)BSYB_(dWum=ZZej}avLnyOtqS)sgo#U@0odPx>0$#6<%
z;@(BS=TZJu`qJsFO|p3CanI$>vvlwL_sWV<qji7OHFyvlj&x?)VJ&|8wZ}Jsp{cyK
zwY3zEVq(MDkP+phzMkJzhGkR49P#i?dVK)qe!Z!=#dLKUv?hPqHt>x?h5!XTiPMGq
zB8^J{LUteEvGS&v&IpaV75_$xJ)?b8Q@pXh9&z1%J~F6jj_3F`(wLs@0S`g{AzSdM
zp+A}&f2Ku4i6v3=lZ1azmmJ}n5Pyg5VwAf8Z|Ig4QS8dX{Csms6pI16B0*;S{vu&I
z@5tba#2V$=V2SP3Pxhh8x|^p%kOY{9>#NIGOICR2jQrM)i^yX<JL$-M<o-Qjy3x0H
zs^if^l7d&)&jpQ*?!Aty)4zY$ytsYXCD~D5;GqcC%f5%Xu5W#wwx=IXQ?o~+c;EN&
zggjvq7j9ev&d}q8r@CxA3xPwUT(3Hwl>;l2N>q&h!d4g4M_qA-|MPuT4UJ~2h)JF|
z9HD2%=$K0$<>KBwwA+p%{Q|uNL%)w27IahI%}zHjDbK7$)t%h8dO<Frm#|zalo38?
z#eb8#JbQb4y1G*vmCd)o8=j4tk@2pa<y+JxN|t`imgV*Jt}$2F{z~E}ip9r-S532-
zqsxCyl5Ib{AJ|-#y?o%C_tE<?b&Yd*pk!vY*%WqnhpqZZ#5*A+(6$ztxtcz_Wb_JL
z^}k&@%98_!8#88p5P5M;dCCb{JsdIj)BEg6O`><5GC%(;kUZYtu*qrrgB(Q?x@A6y
ztZKCOBT}QtFMo!^imSsvqNlLSp7HhBba8h2$TnR?=$Q~*{K;;&keaVUGTAz5C@N8;
zPm#1FzT)p|{<mW@D`XGQ*Wb$u%E^KyF2GU7q%7L1h86+zm*iO5fB*iqEi=hEm-uj*
z&G%vG<h@m`9Ioluobfsu!+MJZ30;!3sS3PkH^&s~EwCUt(p%JVDOHA!#thDGEQy{K
zA1x(1{NFfQ&J#yLA7Jds-irD0{R^5~<faslo#n>#-&$HWTU%OEV*hF?D=V8F{Fpk4
z+JD1X`IAKTab@nP;oj?w;IoMCazy_EIN=0LUrcCx3OOxQxGzYPXZoo~&}J^;FzNxJ
zr#ohUUsC&<C_1YcvWo0hw#I%yh<M%~*{-PU8+${A?0Oi2iMzPrPu$_3-Q^zmDFRVL
zCJapEO3nCyP7T_77AqFtX9>dF_qdUg0li{~PFA1ijJE~S=XJq}7NI_V3;KAcXE))n
zt##FY`hcNsDT(J#Vs!EG@>OGLQ!w&KF$g^8#lwfsgi!9KCi~@l0O?~9Go~A79x~A<
z{-Ok)HDP`9T`FC%?^in~uOFd><_gdO6I;8`0>(ozV0S#dU-dVTttfRoL;qC(nPuKQ
zTkYRGGi2xWPc(2TSI!W$v!gjBI=hh$e8$#&K(fVu_|bp;?4UnbsgCg$2r+VXwuh{Z
zmmPWqMr)<RI)@cHoPwCa1t<P|s*lX1YdSCWynkZ{g@<u~i6i?KgFpS{0iP6C!B<)*
zF6iE~B=VPbbL;u|`}1Ht<((SC{;bETJAXnUWBpidHbzXb$d|4-uz5lHgmGwSIHJ3H
zQ1aByOm?3su$d8_1r_rPEuJVT5PO-K`uCBeQ5Gdm9@Nm#V4w5%QD5~jM}ZJ*k!?*+
zNP4RXVAbNMhP&>-lNPMBv$ksA@8Sq=OF==ry&28Ylg?m<rko9|56nbGsvg1#@`u1)
z;~~q)KLtTV<AdyZ-$U&kBKs#Aoql;lzjD@7<aA?&8yAHbFB$Lz6c6<S6DvR$|DjHc
z)&M24-%r(a<D@<_<@7PSkp&!X_myjWycHc8*}u@k))#iU_jYStYjxEdM&+w9O$TxQ
z(jVFB&u^1kaZFy@_{zWIgcaN*XQB|Q{rH*(<JdauRsIH^Ca@r31t+Vy5bVIasXA~j
zguMRdN!w2N^~&pacTZ0uyr)8>p2o2xd%fBFX8%4f{;name}~~bG|ElIQN@{qc*Yd7
z%lylQDB|J?R2TF6Du<-&gu_?%Cr<7B(OH|mBYl&1ahPQ5;&o@Wrl<;wdqkHZ?8%P?
zL_!rTzm@Xy=g;coZ$$$G15R`gR5vr9XmLn7(qjgBUJsi+jA{>T){^>_{?aj%%_z*i
zc8oqbk+kkN{&I25?(F2Mq_(~2fY6d53mD$nabNfBEUbyE2=6w>3~DX43x0|~YY7ZI
z(H`9u55;b1P~ecN!k8<BQ%RZU+{tg%C=8AK-J2d1nn67NjOuMo`nC4AGx6+(Rq2ib
z2wVSLFChfsfwS;o?rEzzr`udH@l|=b+02FCyGvWoX!fP1!VlLnL;;b|jU$g0UzVq*
zeihg3>)QF``!{vwOdqSS*)z3db>mQ~BDgY_RPLTi;9k<nf`c#rT|15PlY>Jg_3ST4
zbjRZ6z#p&WF5ytfnOkYplp>JG*c_?rJ6Zk2Ae}0GAM>rx$+s?ciBxRo5(ut?*1}&9
z1+<Q(y$WS$BAuIPxjhi(J-8QWC-f17%5I^<R=JRmhCLoK#}}N>2EJME{04&d*VSUV
zUdIDb;QXIQ<3VA}po3mp@)yvv*(IN^5liw%*?#`D?a>b{i0f5a>4klf?sR2ke$_{3
zSCtRj1^-qyoEfyZu&{rBuy0>pedEWRd3R9{g}f@&8cmHJoXquL^t{ZwD;M(nc@qDQ
z=f~e`ei3>H!3tygkto}q>q%+z?8}Go8mrY7S2sS7B*OT?%?-~^a~Ikm@F$G5KPf3G
zmV<jD#xX;35OTHloJ;XYXom<V(p;nc@~o=(P|GbD<qN6PWj#(EZ*Ol}I62ITkYz>|
z@=WzLQCV3OnROPrS)3id`HN`y=TC7IKNG_zu2dH8Q(u49(#WaqGTqt+YB7xZO}6r&
zz3-FHUUz@@HkaM_-Ek_M{c2jmvBh^{u{HH+#fT#Nvatm7mqqARn27FY^zyVSbexFR
zw3}7GCjzL9Ob|I%<aTA-P}jF9C#=LH{n|?)-yzv-C=u^Np9*R`(VdX}118)%=G=PD
z#z&^twMjqyOAENTUpP*E+m7kS-LM^5KrJJ?l%qht)ra08Yaxsdo#1|vQnV4?6b4}j
zN6np^@77sDF8|fQl~*On5$?JlR?&=ey}UFu5@}OPoxXMW!y|l#tib`BMAdV9^;*T<
zRXlPtP*a#>{W!Qvlui=&lTa(2G`fdtVF?-rbj2`GJ0I7*#baj;uB5DNNg@#uEDG1x
z({qE}w78zxSuRmbicZ1vKKjhV)jPjfYE0uqxpz@j-svVhE3zb3e?PCzcIQTP-d#Ir
z(?!=iXo4CV8)rYeo;+d&iv;CbYNet0!DQ_CrC>H^5~H+wzMY48azQ>%iZBn*DNtW>
zahrt3th;6iKGx5aixNZ{pw2CyR$6&uh0==>D5+e5$gn~i)waE#1^IyzCF+vwTDshz
z<AERY4dYGaWAPJi0xN;1La<3_*M8Rpn%E~M3^&-M;t#K{T}H{dzft2*xdkLmgsGQd
z_$b+&v-ehlf-2*^>3ONd>ctYrk(^pOkSaJs4I>tD@P_z=Xvw|=7+o_+M{~{{xYHPN
zFRzp&L>{_Sr`7UWdDDI9$_i<rk0~GBGY(u%OO<j~`xnYl6_nG>OC+_mW`EaF>eSUl
zNqJmxJ)j)(Z9IX{D{Qf_p?cICU6u^PcCN|3&9kQBY<5Stq1qbXe|tq#MC5M#gi6KI
zr613*0kI%$xJ-`0^U$prVQ_-leVNs4FxVjWYi>@sBCq23MOP}D3XaSdi_lMh8P<xn
zm6#HTITi1oh)lvR!^Sb3COJNjP5uMQVn2>P$To@n^_u;UsUqyqo^vIZX`!^n;i*q9
zKZ_f@;e*9P>%w!8bCzE6+mk<k{^Yp2xmA=qU%V+qYwhc(WxQk965;^q`}jA65nr63
z!X^(!wvQ>oh{r(2(Sw_kr?5`^Oe*#5IQ>q)$CWO#9WPwDiK{FNpZ22lkn#D5q3b^m
zOE;a0soaK=N$9|K6aoo>h3fk@Sxfiv10&aq@x^nOzyuhuOKHO9?1lMK7`wf4(KLSY
zdFbWTf4gd~+zqH+gtt3=FE#wyJkB-(#~*(d6<KYs&0SD9>K?>UDoZh)b3s+yY^gP{
zk&%tT9hNLQPBABO{Z~w>`fc+(=vbz<MjbxCXG&U&!Wf@c7*Mo}YUZa&RM4^VaE4@y
z94IOxl5J9QV*2rrzSE=@^<oE!tEZ36hO<Ad{5R6Vmu~Zt<O=C`-hYz9zhp)B435z*
z+vd9r-{M!n!B4gr&BkT1WLUdN=b=E|me?B+4L{ADK^!5HhdGzN-7L~%DW?1_aBkK>
z!K1e&P7S(@u6JRp`FYyr-eW0&BqSgZiyHSOE&1*G#h&ezdd>XrH+{BO`zbn{KOopL
zX(jE8oMVpihKh=cX4=}??vyJ>E_}s~9||gt%?ck}lJ@s*tsn`)(4aNUrBzU^++y8l
zVv!UcV{JD=9fLiuVj>)8L1=BqvubgN>Q<2YNoQF!l)LZI`kc#8@cs;)QF4jg#iI7Q
zao{658U%lh+vF032crX=8LXyTeB7psB+NqeM%~mq9Pim?&$fXxoTbINS@!*E4sMe~
z>?|J^E}S|coYV_y&J2G9YryIW6CMK$;dSP|U8{t;EWe82n<zg&Rf`dGFege8Fa><W
z>Y67QkEA)MqVT6>BZ*VCs-+%fh^u+ODoNQi?KBO&qoUgAn$91l-TGL~6aMlEdf8|=
zZ~LD0`Cl7spIhlrvH3%cC?Zps=<J$3t9gl^tK-=zbakDLqN68sXJ_Ym@A&wbmNbm#
zJgN9eP}Mh6ev7$Ml??^1!9?_#rcL>Xm*KCb8fkj7E)qq=7@4cv=m-}u?VY3P0jZNY
z^w4~I@W7ig<D98(v^A>BHu}Qpz1>xH!p6GSiIe_DrgX{lDh2d6q;PG|%S(KBH#hr*
zrTA~dTS~*bl4bqgPBP-Jc%)R%C+p^+A005k+KJuUPze3r%itUDWSF40_V4Qxv^eK|
zTk%PHWpEYZ`WkPIgVEAdw);XtCQ>GJ3&;jIkVR3>K|P+FXAkS)Zyn#bqReaYfwEW@
z>v_8v=h8J{$iP}M?7w!f@yUq!(_gYq-CzGY<NfW_570}Sw_<&$8R=E&!T~?-=;U;D
z=p!<f(>HHjxOIOcFC8{{v5tQiuxX_dx{;f?a8tCXwQ^8?Kyq<)b#=F>HHHw)<9hrE
z8g*q2?KuY8RpJtZ;|?yM3xKg-N^5gpcbc&(&B;cys-R1c)(_z^`&$vv9P&OJD;BCZ
z*RUe5V&~a?j3J_;rXECfBxXc2xQ41sn)C4_w~*Eg8#Ql9q7PPnBeMI{grT#qYCHDX
z#-<c;@M5-fzer>L(7Ux}UefMpFq)i)PTqoV76o!bkYF61c7noCNgQzI;R9LQQjrOa
z9{u&c@ilViit7=XpnaJ|k^dLYjI1n`qVTrQ{K*R2;Dq1+boMeO<YGU@<Z(X+#Pr2v
zp4i;f)Kt!dWhc+6MATW+l1&T)+ii#p)Bv-q;=RUCgCnLq+bb!CgC@^+qvS|b`}-Z}
zV_0@{DB5n8e2uAh?ObO!_W|j6(ZAE#&sYz&J)drOjgg*aO?6iqoR(YSQ@nI4L0|mU
z#7_cTxZQhI>Br3ZN?ZBT%``8szMdO(j>b?Xt~P<2c=Vg;Dxg~j=eP7t?0Z|oiS(`#
zyxuEydM>9}*L^kfiF7pG*+~F4r)rC<FlxTs_z=974=lk3VN^7Fz#CH%27;H5QMK?=
zaRW(;k>LR%9GdIqqt|+g9c#*pSdo2%Ar!m@jHGs-gey58IWTtKaF4pbG$V=rudC~l
zq_L%meR5vF^0*Xn5vNo%?N8q=ij{GG`vJ@oq81Q4^<P{iu{WVy#sVpwpZ)N+R6|mi
zvyFaatawq{ufZnLv6Nz1Jf%Y~@l#A>oVio=nN<n3Ya@p;UEJN@r)&c2r`*LgF`BDq
z?M=KygQfvQZ@ro5Q{wbrn9UPbdwOiF1~WzlbDMC;Ek>qrQ;Vj$p1yt~Ibmv}Us==)
zVp(+p1<m2zZ7lyY%83flgW2@2$ZjS5qeT)riU1A-8J>e%f>TKzfO^A^_i;B;rq_EM
zblW!lSFCa;vm?T%Fe&E<7(^Bpewaa71~B(Filv|09M*4$L!+PyPYND0ffV=pZ?4{G
zCh-uIG!L(3r10$@@DzE_8tY8#iDu2)>K@krT78x>Fhc)%chR`uZP4Sl=C+~9=2J&z
zYp7XyFk)`cF)y4py3BftzH7*WVZ;MMcbLARVb)pq<Y;4Xd|K9*b>^PNR_KHf3yK@P
z_EM(wn%-ti`E2N7lDk#<H8=;A&UGPuw93`&^l7!`IF%T&nAzy{`umq9K>;q1{>Is4
z@%|Zr1!n~>H99ojA^B%&coW{H6Hgs`c7A!IM%s85I@&+b0z@vRb*SRfWV(5)5H1z4
zOP%qr2%C~N-aYQVw1Zxwi?p`38j%0d@a=c<s;>tZduhJ0EZl5ZJa$`4=8a;i%AHB=
z>-;IRy4N9G$eb}Qv1rV8$mJ;Eh<@sBqKlb0M-obH(C96Z<8JN1&KF13fBhH&H=_AS
z_J=sqr^dir9zFj})5F94NMDHty<5Vswb@L6Va~-BezR(GfyLG?L?IqG-Wme2j9K<Y
zj|_pCfr~buJ}xm#{lk55^DJ_TAuTU^wycU+>Q`B|kW7?2L=Y*q>yMylCzOegCnijT
zuqgcP%!QaWMl5v1Ip^~~f)W5tWC<GS^a@V>U96zhP0v*)Z3?dFcH9-NJ8>goiNb$`
z)9k9G*<05zC`1{)zy=uy$HGOB0p@dMKtCL;7J6Ycw4F(h@*UuETQoVeRFCX5kUZ3y
z2DbZ=GAR2djuZ+<8Y2C2D^D<TyVv1k$awfrBr%-BJ@#s8CxJ_Eg!So<mRy>Xqr=0c
zq`!}=NE6@_bO>?78!8W$L|xo6G?9kC86ILQs`Mza$Z-AwHWF&tg7+q@Fj3~5&L8|4
zN0|BRFxRYz<wOZ3qq!6v{}-P!(FvV6K(g5i$xdrJtE<mrH(_Y~h@Tz_U785o3X~#`
z1C6EZll$oxQoJ&Ggp<*pRT>Pxnc&F9E1479$Jo7T<UB@VU;jQTH!aR;CdRQ+#^+QX
zZwz$I+jWTpm@&d9v}Bu_;faipY;g@fZ)$11LE@_TsZ|Q{Dd0~%U^6oli-Gea{nvf-
z%?Lel*!-@+)C9u{zVG?SjobcHmCI(VqwC^nr`etw;XQhLSE!g;@!{@CQFwu(5)aP-
zw_dk){xG^Ln$#MB+_W@jY*u&oBnn12grtRTvpZv1tn~hboOI_oL`4YWe&dlkZE7*3
z@O0PL*Z2Lmpa{zLeV&meGJmHfvdj9r&mX_U?s}0K@z?sDwGEEkK7|D$UC)|FHeq1b
zE$KSJuqpCu{6)z9fleBna1ijz-dz<jJ>r|t#m#T2Bvn7YS2lc+?UHU}#eGfFe?MXP
zs=;Bo#a%0ZnS8|O%OFT>vkpH{$QmX0doO*UHdZ$K?g6@)w8%2(0JVFSmTA*AKRqH1
z^m?f&*yjlXyT;%`hGS1R8_df@;?5vwLo%WXp9D;m3e{~4gjj#=)!i;F^THI2hAeCM
zoLRn(ZfhZNEGq{x^Lc53mPJdx6J{O=+IdY#@!4O8{h4i*pPV1lQwA+;O(OwRlq0Mc
zXKh7O$nDTy_<S8*gezU1r9i}Y!Y=%mHAz6lmF1NPzmO(zz<`!+t;?P|Icu@ugwL!S
zP>s8OonWzI#sU)uywi;w+$1J!{!ZX<j{%j?j9ZittkhVlaK;LOfchUb&F?jrnN%((
z<5-YNX;hSyK~(zN%yN6e_b8lIeQN3xD7obzs_g-dc5p<fFqS_@;_$<~t`|h*`uSyn
zOo>*4{os2QYkPb9GM2^o2;uo6?y3<}|MMR+1htF{i!-m=C8_M<ijPCAr%l+tQ(}1w
zACC!~&C4xec4^ZzC|71Y6=oQO($$CuI9R4q^8du~SW#)CYK7HPRKm>BxXoG-0X*_|
zPelRo;c20)EXo;Il3Pm#8$|Jgtm-jN2j7&~1gKHOh4B)|xS|dcZcK|zUz|fGTRumb
zB=^30p{@z$1&A#Y!W^HKYv>012C@X?6)eM<m-bW)@xv8Is{qxXG9j7sOVYPd83x6y
zf(_4%%+RkGL24E|c6N4`zBYgQW-ff7z$3J&Xy4`)?_<o9P?T6SvIwW8puqT0Z@lza
zj~682Lo?1-8q6R;S34f~XixD{*M*ls_&U)X?u~D<OQ378)7MdBp6td2L|UGz=l{kO
zQ+!NT+z>g$9dQ_At@dBZaR3=U&%uFcoadhN8+19v-(YIP`X5PG*1cRs#3<H3%x>y{
z#<*O!!^aZIjd%_>Aw7WS>m0zBsz4#<dh`r}^BkkES?D7T@RhJQkg<&AQ(t|$Cwo}2
zukfybSK#LR0eu|bEV=gZkYr`6ZFV7EY}a0DrYfq57+ns=ArRUf5r-UAsc=E-skcf?
znqaufRdvatBesglZdAnXFR#*;{IRf4@h6?)0go_qV>RDhI4OtQy8;#Kl3dMqJ=dKW
zG*sF5Q;UEKpgEygDZkX5B@40XDc*c~dJ=HO#kF6oZvyLZah4;Vj$j}`LBTki;!;r*
zVTCnS&1F=AdzgOrw7)w#rL_u3gsdJB1^|Rn;84PkyAGN*YzlM+!M#!pJ+tZtPzUWt
zvJ}F|G2Hhaf-Oev<Oztp;*5=QS<gYQE+M2QN@~ZpEzK)52W#u^2IP@G(lAN+S0MA?
z$o73kN9GimxORpyvisSs5~w3vKKg_{&r!_LTvaU1v?G_O-`M6L&BSEw9SvQI=k&w-
z%Q_bk+L8U}E|rQSK8&DxM!{P>lbYVYKk@GdP3ErFC&-N4tgi(aK<xl`R~(U9;7G$e
zaV4$EO4rm}J9W{ASQ^w#hxH%QeH%89@p&SS7}JQ*uY~)!n5->!Wiwi?N5Qz=<vwkb
zduANG9*m)^m8d)Ku?ArNKu=Har<<!Q?)6%<;^2aI>0?^~xXNy!4wHH?UXm8U1L_z*
zEX^^w;jhcPwA4&3wNO#g5TB$T&*3Mop7WO7^1%x2CD7>6oVFhyM}O(_s*lqpZn<hb
z=xb@WD+a}%O5+^-(kgWWBp1;Jl<vg<i{ur0lFTd#{H{IHkIaMt<p6YIAtUeF24Ks&
zMm4LNQOqpust^8X`1tAUQ3kdJfvs9t8X5uw0??HCfuYhP(h>x3s?BcCT^Y4xP#Q~3
zL9buGMlmJza76uO_Z=>Z^jy64-e)Rd^lkaoH6eD_ACCu_?yjz9;`=(YG!PB0#R3G7
zLXW$d;a)LWE0yY4p^BJJ!5uw5MUB3;VGW<B2kde3P5@vFYX&z>HILv0?P>BQxU+y{
zp@w%8rr2EoQD_*nR`d2!_~QFJOq|u}1!COn-aG8tb0*4oe>#K%^5*#W(!_!jdhAS!
zk7Q3tKy^>;CXO?M2z5PW9$WOuEGo|ys-U_HjcOv<N%qGm=)cCFVrU|q8(#XU*7E+7
zU11@7!}=ft9#%700Td8=yu6OIaJh6$ZJvLE1y;25QgL&(^g2IdbZ+<O3?Y^fSnyXx
z+JXH|odZwCjJvN^k=1jkSl8cg_+zltCmG8?4SgSv{t@3@ea6WdAF_50nR%I^$P%po
zS`dZe7-``irb8KZQ;aUO1F!X(aSnjj`+o5#+x1e*F6>py=g2|HGPn>QMcaJwMng>h
z#FV4#gI$Tz&UdWu%At%(@A5%3X6=<&C?)`bVrbtVl>;8sOQsz7kA*Ngi<j-`hZZ#=
z7Le-LLz++Q)1moaI7(Eo669yFtt6e(W_1@?*VKj@e86#Kdes}2#QtHX0NYFq-C`m3
zM@-sAyn|L?J27aj%@Cp>bAR|rv+Uo2ibSE6cfnYoYt_zc(6HY}-0>_E)#wR>a0Ko`
z%_GseiNtt{q^kOebpx%>J{~dX5TCV~yxk%QEG-+xHB0-a4KlpM3X?~*tfcqA8_H{e
zLQ@#NA*Ya;Xfk1(gQ>;;<CsC<ev){m$$z+T<tDqgOsl3s<vtj(JSppfV{Y$+&##G7
z{5Jin8uTZewJz3<FaV&am0*)4xepTMkZZZMr>=`^ATKxD`SR}Fd+UCo9^maahO8?R
zEWO*dT{5bGUwb?H6GT3Eo^K6LavHUHm|#4az|h%k&mJF1C-BuC#9N(A)`AquG8ixf
zWyP);#d5MFd~p<0!-%j|VOf>`MGv7}l##N7^nIUc1NZTzC-e3tWw*SrLUY_M|GMZU
zQika-EG`~sRqC8+;I6xc2vyPDyo3uCIX9zwDi3{i5wXLqk5$T%@NI(|MazF`^!-GT
zxwN>bp8$TVDWi*&NfIl_$+h4?VG-BVQT_^p@usdC_g^v*LV*i~i!73R8o-{5I8s})
zPR*Do-xX>!kd`NNX!$6cbP`}+@tyVtO0P#{XJ;q>nR#IbkjL=>Su%f*j1^toC5=ih
ziBY8J39lD~F?Ro+nIwkrQSu_`M1oae&)8|C>e)s_3D`iJbbxsRO;VEq4^T+_Q)xIt
zE=wvvoHT4C9o9XSuj>Xj>9K&4W&2W|J@Ix`p(1}}Wd)G!PSE$|qCrd{r6+wTCfk6H
z!}lgpnvpZyKL?th!#Q2TTfHL~!w7etYz1R%)~z_hBmwdDxI7t&$K}=P?;5kOlRP-D
zun!8<?eFIn%weuCA>;O>Y93%!Qv(d-E$fRaQWlUv&hyWRCifdSwtDl@{I)Y5e7|{R
z(}V({1R)be6Xp#mFg`uM?G;Jo+<SQf?kkplozqE`MyH1cov8TXyTk)lK{d&xS}1V|
zJTfSvjNqh9w-iti!3J1*@2DY6MvB#Bph~W_$fo>$#~;HOJ9}PE@cSr+3@bcE_o03G
z8X+?5;X`*CNKor&qs{Wg5JAzo+fE|=T|?&ZXt7nLMwQJaBcq%FDPq5w5u-31ATRhw
z%l{LfyeMnNHCvSL`xuVwEuzx!LHwlYA8RZuZ!>cf{AXu}3?{Sa&iYF~XEH}k+4ZN?
z=|>ER?cga^S%o&}VtIL4aeP`U0%ZI7^Sl~Na!2?P0fprBbY~OPlfLbpyQ&_#=!tmt
zG!n^r3P<?urQuF>UD%_rgfMdjOy}s=Y;Cibid=(L1zsZ+S4bk2cF1p^pwfziRk1@q
zY}`3pCOp$B@a_`+HW|00&@Tc&j|wjLMy26J@;J96a}))$_Wbt7StaN`OepcIVoS|@
zYpDYo)3ZyKIcQp%qud6G?PHEVhXM%RvU1;Odw6##=`6JV%jwTjvLXRWLrJ6h!bg)e
zk6G~hwNl}z>vTny!T2yY+sOU}wMXYqR9M+Zle^Eg?Dqjf{1e2upHcr+a)VAo)TEvZ
zVs@bzT>yLcZZ^;e)$ez;&#j0%G2Z;d_r+q}DXj(Zx<QdE)ON%9!=sAd(O1_Wie0_a
zBz*5YNSh-arEfqu|Mu1Nv<6@#$?iR4RUpK)LH&g<b-17vuRs`}meA|iCZ`g-W|eR1
zW<h~~iqScM@EqHyPO)iLu85(KX`6Q1!Fb9GUlU86z36>y_S=T|@&uqKQdLH@5g;>R
zWEKH3?|2(9$hUzqv_-A25zCJu2k)x|8cl8c8Bw~CgjaY<r5uZ<|5`7R1El^nBOhxY
zr`S^RsBh;!hk8M;QxsQ-lZYYHj<9U<-CfbxBCy6#bNWHNoA*oT&VS2+n6b&$&7`L&
z9^K6;J42eJ<fo=C@`BG28O;O>1P&FIgc4^TYnM;~g@*<Pv+FW_i-@S`?DxiU5R&lC
z@Zeq%7}M(3U3+jZ_SFR+-0_!p>n4e}(SJvV>wMjB>&Gv<j@k2k43|l}7JK;V#YKUJ
z&pMAIVn}9U8YgHuvk0?1ZrXc>#?FEu7Jf|CgcrHy`6hLE+}8SgLili8*5H+;(K}K+
zuU1hNzTLK%wU!tayj*7AXV_Xk<M&Apj}@s%kB3~Li0`0D{TJ{962XWe1z4(e-mgu8
z*KIR^A*PM!zUQa(di9;WdE)4>PwoQV(q>cfiKCce6L~%#N^aCm@n&T;+{S<<HU1qw
zUS16GSX&Hx-=!92QAaLNy!dUlxZVqMo6V!H0al~*@MrRxBU@JZ7eW3~5W@fHc1qjV
ziAkE5_wDa%AtcH*N?49eL2s~?dz#2)$v?|Zh-MmqlK^u0gTuprpez`sI5|4??L}3S
zU|j>-m;|YK;Xg4=&7#$3*~?!9L1RbeI|3nsMAgCpjMEDXExx?GYhD{0Fp&LRm8m4s
ztP`lw2CyMWQ}!Cko<n9gMbaGhD3Hg%mW=)7(MVN*NeLIArG-zcdh!E@UZg8o)^Aqt
zGeQlpx_qOb9c$$z=pzqzo62*X948yAV?C&P{0IkUYdo>!!c-euGdw8~Py&<*-@piA
z@k=(iT==Wbd0tcL`71YOpJ!>B@MEVw;m~Wg2Xe>oq87;BJwB^hHTJjRf<)f~EPP=)
zYHEG6SE=_PeAlIQe~0XbAQTgfT2ryKE%tIK#7UT7U>N>C`P}kjSZlRe$YYJF<1zMY
z3>Fbrad-|+z7rG12p^;E<zqW)hlGNy7?>tt$o(ZrHqiYi0yjkutcC@4u9)XS9eSXq
z3hT%0YoqXC+8vJ+Pr3JZiT7-f`j;Tg-8cg)72}{ae$zLlUXzQfN@K;4rF-z?7ePuO
z`Cf;<rbIymkEnSE?vsYNFUrjw7M^x?cfI|fDE4##l-|P3Pc3zT(1QeXaJS`9(JImZ
zp5b0}v$s61gVPScHd4$8VE|Fkh3ceYi~QjABg%Sq6hP3Xsd15rO}2kkRbKm~C=c2V
z*J_5yX>2fbM(P~u%)T<<Nb=GYr%PC6^?gnU>>%(ySKsK(bD9Mavv>2X?Ccv}0uL<K
zj1uNtm|dVdX~#Pi!R&YIqj?i?vJy%Z3Cp-<9)Cl>?#!5LXv}YZWhROYMzsvrX`J40
zz=&4|KRc(SA7;#<Hs$qS=hy?qoDtUZ0b=%<&e79xYFfUaY@-ZtG~o$P?V)IFz2l$6
zA`XDmFow8XQ{lWd6x=ww;l^SOH|RWreyJ@7Qz6!I&Tn~rkoG;dvkAP2P@ZYM9^u)B
zaQAw^O~T`tt9wLdi{++?$VlvXzW(RlU^>|cxTykoqsVyYP@7bO7MT{e@sktuJ1tqs
z3!udx=dA7Vk&q$?o2DlY#80g?pyT`jz8dhzcCScJSHp$&q~ky!^c=hQfSG}YeKYA$
z4y~~;CQaJ^+1rt}t5alNJU|j~;(Um4{N^0G>BFj3u5-3`?&|8)LMv?=nEnwJ>n*7+
z<phcHZu0p0{S)~}-JX|l*mHKlN$0APoj!a7BhhbCpHls|91Bro4N9(xpukbgzHIW3
z>KCftlI4*^O$0+w(P5SUo`1DOZqAo3`Qp53)5pJ=CV}Z&_1wHLj>gZ^^RSK%n=u>?
zRZx^QOTpCB8rdpOGqDZO!vsdNmwZEgsszGwWo*3I94+{;0#AnGJ;8%qiLr>ypR%I}
zs68tCp58!wec#6yfHfsDy&o^Knkd%K6{<;xk}5LsMbVe({7AinH$na;!e`X!4YzTN
z%bpj1Hx@|%T|w`(tbyi8`~N&6FCV^P-M{`(br9xy(D8A{5*Cby^KzL#v07*Ns!jlC
zg{l8LXeIqrT6ISXt)MAcB~CN_DFT%&yUY7$Sn2icSo4%bH(1qw2qjyUEFj^r2EZf;
zrTeI<{gq`td90E3AMHHwX|i$#ABn>XrYr@fff8oVTP7gl&e~Qtd58WI&|ZadRqM2-
zXR49ivoC9Ul~nl<LT2weBsVxn%&O-K!an`AH$ZQu4O)#Z_Ol5&scIbvSMil$;yctA
z<V#n|Y3X#B*Ju)$UwWuWXfyw3ZV-y2zW!qJVet0ZNn3lCkG~y@G~l0N%X8}U2|0B?
z6DrmbuuealT5yuXA?3(ZC4)Nan$<J@PnMe)n<$2nkcFrDC_*psUql9`*xLJ9!OUMl
z(6v3pb28+a;wZW~4naeXXxNj~Q{{}*cX!-qM$VZ>w5M&Wc=irde38%|`qf{jypBnm
zK9vcd{$^OZl5&F3v!=}Q+!PRR0O~C8kB+T;j*Xu0^D+SABsUTW0$gMB-N25H(ec9=
zEH{k!uw}WRPk$p*bPk!qPAyC<%{FHWPK?8L#j%*+d2m2}04yZE?f$xBddmR0_WKfR
z6_0W6J=QggSv3bB;R$lN)B)x)M#^Yk<)F2fCY1y%Z&n)+og*wFY+ZC32^!LF59sKp
z=Fun5FsS`-uLFPjSO$2SC2=^73`-7Fgm!w&UUoqY5c%H!3hS|l7=?V(-mpPnVzqNU
zPONp5v=a`1NY-UPPi&E1|MRabw6zad<TdSiG9t)K<5{`Uuyc3#aJ}ZWCNI4ZW`>hB
ztWm_R*Zi@+8T@|mJvQWmjYJ|nrr%x*(F&-)4ka)+I)+>ND>DTNqI5csmF!}=_Ufh$
zp*IEA`!qtPJ<OI|nhYQTs`hYcDGn1Y-J;XanvZqpzL={gr2nBBC2Zgnr%Ri@tN-Au
zd8xj7?VmbsY5bbl=O|<fBO`8mbQZ`q@r<kA&-M!(=>vVw6NZ?9iO2LH)QQ^nfc~@(
zNhd;4E4Q~gT`?Dv5K104F#V%V`>)GlAdEJQ7@CwB>A8v+BTZ-~!D8~Ie|v9C+bQrs
zmnY?D1W>@8x{4N?NCoO8H1K$BofRrHD;;#O+{;6`qfDO+UJvCgQy8QXrsIb)hdD<r
zZ0+d3YCkg~A&j{F$7GBh!oyv-ucNUJSd;6x&qrO{uPaB>->SCAIIw`C6YXvbHN$0c
z@a1&4D3awRF#JM&zqt&@g-V-#C=Fw$7CYFV==0S{Qz&As6d8o)ji8vUdCnqdlqqDs
znu2!~Lt9!~y$PGAt$Q8<t1W;sE437EczNpp4oq}zUScneVkZvRbMboVn5Y>^jbh)Z
z$IJdlLeu|K{aVcS<>h6(o&bz~53mSCrQ%b-$z3gK@pWWjc_5Up3^z#myVH|zBWSnt
zKFRcYTKc^!<oj>vSwtZ!gSAaYzEn;8>UKBBzrU<PgtFDN_hpylJ5PZn%F*o(%Rj39
z;HuKr`_F3&dOUSqwV@xOe+@8(pKByB;$DY4Bzftnh54Nqs;N3Wf>}Y>((<ix;tp=?
zsoryAjb3{?9X?rJE9ik!R2TMVcO4z^;yRh~l$w1RDpT)Y-{pLtq?fZ;g}?fyc0LSf
zg@lIf?HoX$mP9<z$*~s0w!XQrT*%r6x2vH7P62!--)FM5CMvFtu(w-tN-P(P0H&=9
zLi594)YN9^7K>Ft2661$OW3<{+6qo<idvLpjo!I(WjF2!bnd;t*w<u=8p1>ArXZZB
zOT&8&c0i`G9{&1!&qa_jnl^mM_w6H_Gco05TRs3iEh}P{xENg(6%`hM0*eZ7$^C=H
zajL`MJcHfdUWx?tg}#8(5qx8to=6AUDdbP5&lqk`78GKyn1tpV&-gqM^kB2;bN}&#
z3>NUEqa%3%&J!lHYK=fi-UpkPjme95clVNr#l_aVpaUy`DxLMpZ0jOv?KMxsbfurv
zY%zbg|M@u%Wc6U?<bj5S>8NeApUSm@6}Fx60XOJ}=pkHK?>Ov*RixKSf~Iifs|F?z
z5?~ey1qN#h_wvC_zI+1KIlOh{bkEh-c=dLZnhFoyGUes60Pw0v>V129hML_!?Y(he
zvj`2|s4hAiU;cLuIV*iE-+`tJ$i&yG?YTZtdfX}&e7nig(B!n;z<cOc{P#&76@Y+9
z`J(=3db*ZK2eXa)>2Pddpyalt#h3JobdFv)fL)JEY1dg2DBHZEUV&}iG?*=52VJ{a
z)*9|G+bDs+#k*Njc~$6DiuL5`vSiBHeA|aWrQEr<Hu$?xakla9X)t($wz%-P>J4U4
zzD9q0a+q4O3m}IBX5eR(r~HLc$SG%N54vbHcvo5JJKuI`X$p@dN`kLKri&D;b1r%O
zsaP8RF$=!(GW%LT5_){JzrSA*qSXFtu@&6(<rF_QHg<F1529U8b|rj2p{q=xu^KQ)
zSOE@U=(_gIT+y+V2wKou1mNZ{x`;!JlS#S)Xdy5qz3#}M(Z9`+>t!mEnBMy4?79+(
zJHr=XGZZtgB-?zR0=S9&-*G7co=WTh;P3%UaVJN_;c7kV8G64>$B$D?6J<;dqe>}9
zTWxOghMPr<B}O=F{Bo6P5r8OXFqSXMg%O?OQ~!d3Bcok3dDsWA5gc!n9LEg_1(37P
z&-`nPQRC3GX3{vCBnjU9>NAeTejD%)%!_wvG5^9$=e5uvasUMs-SGKO%CG?^q?2u&
zFZjU{h=8YmPN1|%N`at(nt4y=gBNiZcn`eEqwSK3!(}TA@y0{j>RlyCJ5_(8f81rO
z8^;#P+=sO3tjApX`cA9QAHFhdUb>_mO|2|__u_DDTLkbZJ-`G-ISA%Itg)M9;4>7%
zYfp`VHu=#=;u(+2&k*j072yN4_jJ`<zC*=GNElyLigI|mB)T@NUjqJ^Kcl<XfLEcD
zHtK^Nh2#@-ZH2H9S9<_(E*!y<)NI5sZYz{_%z`<15j~3VyF6fhxMc{G0m)fAB2%EJ
zPITsCWjT2?dC471tTs94FVbxB?oq~w<!btbWhGo7Cu!LHX(@R;TZ)NK1}4dKdZ<M=
zY{PV(q|x1C{ysiuoGxZaR?`5XsCiioW8u#kS!}gV(g1<YYv;)a?dN!@oAP=_Yqi$@
zRu~))t(SGUy14vBhYY^pqhTdvaJJVL6y24I4mY3S{yt*96Fv0ut?Xb8Ea3n-!4S|a
z9KtSq*b{XbPFH)Co{S&~$N<DND8=31+e)N8qQRb5-xAan)XZy4@BKBlUS`lJnfVTU
z%(n{L24$!OSgaa#s*NFrxD;Pn0?2F;`>(J{toIM#URm1Ojp4K(Y5KtgQQh2t<Ddxr
z1Df^%>Kv+OyULHL<>~puh^v<teCY_}(4K@v-VC3tY0{fd?1@gUzJK0nlX5^*v~Xct
z7FI@f7&$TTblf?z>NN<24VU*u%qzIMvWq2PPWfbFFClUOoF&TLHN!bF9<dzf3Fz<7
z3Q9Cqa})~yB$dWw%^xOa51l=D$Io+MttNhYoiMW9A*^uy=`+Ap>9c3}rgq&IQv8++
zlpt*`c$4w?u~=Flwbg7X1he(W9cu1x$!<wTM-u>ol&q|*3S+M$;H_uUu35cxnRaTb
z34x4*tqQx&oiwecGR~*`zXqyA#T>)ER#t$t#$UKFwgm80qPbq6DlWo1?TCFsAmGaJ
zMvwM8`rpbPyU*kqVn#AJYO*9qHhl+SE}gVoSk%tD0J$h0?ni*|*Xn9I%`DE&oUmM-
zPaa;<V5d(Uv7CH&DcwLvCX0UkKI=*=d_sW*%%8qEsFKF}g1}SGKaW>DYdm`ufe_Gh
z$oI&1>}=)8$P#z!4LfRYX^GrZ+PT2j>Pk(Ii;H^#6d0#Gii=aUsK9y-dFca5*~=}M
zVCx5M!TTZwJLAbL2i#LJtJ0k8)@oX{$D;_K>k|?Z`s2N#i&cayDq{Q03>J#suLrD_
za?M1aZEb00RgIrDVDa;|s6Dlw!Xg>tKMfcESrR~T?4^NS?r7R?Gv_wf%8BV;)&}3_
zz__n1v@NV72OTsJ_$j)d{mdNwRX+Vs>NDziAjYyvW(*n8uf}ZZCBA)c#qtH+VktlA
z^l`et!@|G?`}Tk5FwD<K=PWZ6n0`j@rx$0VD&fB3h%esn2WDRzHZ0Kz<-F1$ywxec
z{R22n%Of%kcFgVHThtDwGF92w0D>3Lv%Lzw=ZOQeBzw(~dEj~IX!>UZhhH2m3jU`!
zpmu$(%-HMwZ~?dacUp{b-k)jx9ys3<9HI2`tTojWf-MTny&?7@P)WOx-a1mUkc>jj
z)qu##_)<+5KdFYm$5i3~-uc5zmOZsK{~OI0Mb1AJ+3sz{z)^!)KBj7E=o6G+=?$pU
zaULDeuU<jUHB6Rlo-X*QP%eHiFZ;iKZ-x{8&zX9eFvN<++7*suM$2@c+2a7)8cbK(
zX=X1Ie1G#)0v=Rvv#d&LLovt(L;<XP10b=!2MqC)BbPsV>;Tsu6$6{6`7>ZB?Ats_
zV3?DO=0Iu%KM79QqXjTWxa5D=9E#Y_IlvB;`yV&T%f$tp{UKKqFJ^oG`{bN?$RDHE
z3!FbeS&7|mg_-jNqVNVcVn{#hP2ss9FSCCmzR8Gsc$uXlN$;HQ78T*6qsfXbChFD`
z0=RTV?#k<9)&Dth?Ui18rLj84z$e}ng#I@8(1hkelkm^>v01)hN|lf7^mzuz@YCXx
zrPo-Yn8A_Po>fly);STjW82O)Y|4jRy}ay6^8`nB*5BiHV=<vDaCn6LT1my;bsGal
zkbo4y-%H@!QQZQc5&_UxQpX-g?H&bMXONvJW-FXU_fVLGcjr`~{ejewAq3!eHRzKJ
z=SlA%v?VZhtNPCrVW$7z*DF1OQgAOXj+xDWlaW<Xv}$fBmTQj(C$1Qac`>L|g8U<8
z(z>xVR63zmI78Kubhk7aWtO<TD7nBz2pfL2hM@_b;e`k$>0;Cd7_!l=MW*n*Tw_kM
z3M?YX9@ymJT>qTz;NAWI{y|Nm+nv%OKW-l7FuaQX6fakzkFy;7gt0UXY8;wxvn@BL
zU;*Lw^2Q`!bliHp{vGH1+VS--qi(Zq&<HKYYD)71LT{2+2}A|lf()Tw^e34&xHl-L
z4c4O)J%xXrE5`7j6@+?=QGaNm0`CxDzvDLb9yoPcgC+TvHn9ac<B3+2+{D$FnI`?v
zpakAg&-_sFk<U{i{`8H>M_f`nR35%%yhw_Hp=#2%sVbgGtc;oHz)>RHZT0j)UoHMH
z)1<yZgAtN_;c#${oVCJD`v{C`p7g(}KY}p{rW^sseXI9`$4g_+E#%vZeDR(7T$Ew2
zIG@#HrrhyPl}%cyc#ezAxrKB~sRxEm%dj#Lf2;LxB*R}kXp%eE8!6{&CIPoa<R3j3
zjd;MkS!W~gXHcru62c)4K=getuj?(@@{}UyGp@91(@Zs`aCrH^Z5qS970&h{_mJNU
zYl6haw-?ZJlvNJ&a{W0vsqlTWV%u_~g*A8e?<vBnDczkZi79#?yT}8E*f{+9IJ<zE
z)WYRUEmD<GJiE&;pNeo65Ixg2?D8(Z!s+#(RqMb%lvH4yCPV+ltFZA7#~tEGr+ona
z!(f0Pvm7RO1-;4Q-XO&Ck)7C*DwXZ!4M<dY=s<KSm>^ftd7$cu-keZ3GzBmyLDNx`
zkSqg&A3Z;Ufh$cX<6INZam>dDEC$ks-~88(VVv?t%)b5;MgELGTMtAea7)W?0o~;V
zr_|oNJtQL(ErvFfGLOsnF^gO#PyHd`q^!Vp!WG#wc(YDq>6&f&sD&kFm?^kQB<p6I
zDbkUg*M=f~72CkX-Hd90?UEk*b6(T4gGkT2BzqG~X(6QLnQ4xqnl?z^zUHA~S8AY{
z!rujAbM8)F4&Hw=eb86=kq^}mwarR69=WriOMAOk%mLUn?_}>=Wog2lk$>^_sM$1(
zt>bIuhrQxEB#=+rzjx89iM(c95>q<!6C)IHVVvF9tka973RA0ovd3%8^>OOs8JV^N
zuEP9Zq82JCFF*J3-mS>#C1D%O+h*)|hXpz*V9WoW9cLX;F=wNrpv22a>+^UM7=-v~
z8iS*cRE9#zT|8zX(eu()Tx%_5QT+xqV(dbQe*nR2{4Ri#|HEh5=ls8jACv`350qh;
zT?=30zBG|-w|D(cP_BIg+P-f2D!C6{skG*32Tn`le863Oe;%ZnW10pkQzjIS509u3
zI{k!c<j&BO#kq0P5|HT9(eL{j=Y`9k7!&7vp;C!|fz=8*eeIRrXS)_f)b5NQaxfRU
z>%y2bT`JcddInr|zj#^*jMOy8{tJk!KMfdSfqv<uR%#OCDKF0OlgZXTo^wuxa?Wtf
zBdiyAaf(o0qy=Do&_SgFhC9WGvg?H0gY>o~q;eTAO{QzQ<?HGJIIh|^D9lv&W4NhA
zBXsBd2Qc6yd1D<?7`oR>jpE&%ZW+k-+UF2Ook3h@ZtC&@T(hKm8d;sL)>9Ai5frhM
z{3VVr!?FLhW9x)C4)DG!&9K1BsY`hUtaAP&<8vlP>5ZVH6y@TvMM$Y?MN+eiZaMV;
z+XcU*B^7cecr35UZmS-pOS5bjyElt%U-}pwk8<FXfZUE_Vf|NN=4jgSya?s;43L$0
ztoXVfF|?-!%g^5k>W)*t!@W;NrHL67OEL?5rGI-XKDPb{`HvUH1iu#>W!XAW9hl8e
z6L$(fi5+nrlY+Oj7!2YalQX+(weG{zUifWMvD8F<crk|TM!oT$Q`y-1AI3s=RM8Dw
zj)~7II~)4Zl<)ScWEc?erAd>QUxcY{#9|oAdKzAlZX1BJDRH=czdZ$1|G^4V3;pqd
z(6Fn!CBWuCE6g5d88Uy_IJ$FU!iDhN)230qiGN61dDpi=>YTb(&GT!WZ62Ah!c45a
zC1BJA>7*frH-??S4?A*Cx?=?L6nTmsG6~vzcRG?MS$fDnR%c&J_Bw{h2BS2H%IR3r
zgmd$j?sT4Dp&ontotS;n8#m2FQ&^HQ#h?=gD{^>>$8Z0Rv2AY|l01J99^FR~o3y?-
z=boyQ_d+<mC~ItvH`u~x>((ZE^13(tj>Jfk$^5#OTpuaoe7k_O^lje52fUq+BYC`{
zUrjZqyHoQKW5a=_>RUNIuV-)Y65ECFl-{3zx1;Y~CyJ!amnkk%_#*yPG>JNJ(cFI6
zp8pljW~{FfVZ}#dumn;jsI5F^7`4lZ&;3)5g)8BSOtTdLK40-OfW!!Pbiv2M7A9N|
z@0$5qTSM1`y{-*y(ooqXwV~~08#Z^qU)$_r4G0NlnnlW?gWph(9msaw%J>{w2)^p=
zQ3ZD2TeD)7MhdreqBS$pZ-o*`-hsvmyR*)MJo+4z0|a+}lk1323D<NSy>hPep~CM~
zioT?Myu_q9f4Qb;&YjcvKfoV2)y9fLs{noHVADqp84NI1T}yRlQ7kLO*~ImAm!K1^
zmh3@C3x0iaD2(kHa+2?(xx>e+^vqA{ff<2wHszK+YCMhw%xvzfZ-GEO{hquIm}aj!
zLJpBr^TS=@`r<jFH%3wKTkqj@O%Q^QME|3$@u`h;a^vjto8bNz?2`}`*=DbZ_(z$Y
znj+m`AlhFm2s9ZCOQ9H`j+L$1a-;g8KcL-YjakIjXR7^Q9oHJqblb;wFq>Mkd&#jO
zVNsbOqvbGJH|elKNHw;RL`Y2OZjLcT8E%r(QgUdd$a9M|Ioxyn<PZ-}wpd5inkSK^
z`~UBLo)^!H=hgqsb$zbi_4|FV@AvzA@&Ao{SC}WPX<JXyC;c+9qkkT9lvr?1wq-nL
ziIw>XEBg*42VK<X9Sle=Y#3h^*HQZU%-k`dz4HT6(^QUvfQ!RM7WU8Tgt_qFtw<6o
zZK8$c1P9$ywYhcY`6Z?cOPAcm@wQXfLUe`+DLn;ExRZk{z8@EiN^FJNUb#*dBxx`g
z$8MBH@K)58lDqV5YU(;RQKRxd?G`R-@*czF#+3>6JhLz7jmlIa^j|V&a$EcDGC?Mw
zwjyr_LEdIYVLd8GZ*eZta4xMw>Ur9lBZ7NO2Pks=ns$D5ovW`v<_>%HXOXfBUHMfj
z8G5%RfB#%YpI;ku<4@0`mF)f5p>-i~EvIqJA1w)$KvR%!UmLBpWst1f``lkf8H!wl
zF2D?eA%#CrKq%^ysjOisw)Sz<T;Dt?>?02n!-zmMY42U;%@bg&XVaS&Jxm}Dhj4Hl
zVzqxC*=#|rt|}YzB`=6wu#+1%g(KmYg1u;%2qVdYEWqlyTM0+eWJG>;rDrf&b$4(e
zm}yiP=a1{IYL8OHlO-q2#|@RNLmO6$ptvS)3ZwSBbWQ9Qk)yC3zXDl-U~3!jL-5ys
z27KIh@#glqSV^)gd+s7JuD2x>d=kU>R=8!US4pzMTYRzO)PxetMzdkkPotmX>{G{r
zq5RR0mJAqe5%vNxziY|tgsjUSngI7DO>+iaYdULx_e|7E{1%?T+yvboTipk~(|jAl
z9ZY?(JxU9bM|E#JxyLVSFA_Q=Y5e^}>BFHO+u*4}>kL#l>I44627$#CLVo&II$WqZ
zx^TWQPxrWxoMMqC?|shPF+<F>0jnreFL(_vgLh|(%-IO3fr@eyn~So9SWc8ynK%4v
zUy(>V)|h^j8~5g;`aHt3erz+`3HV7(hJH4PJ^8c~k32IwwCirjdizwzCNGeky%fFo
zb8m74usiVcKE`#)mD(z$d*gA^ctx9GNaJ8lPtoyl%oDS61S^DY$E}SI7hdJ}HaR89
zP3)6<D4|y@+%<+D4luIhCyF{B2R{LGSX~NjLCEw_W#zGsMft@h&Str;y_GTiIgb8p
zZ^6r&dK&8-AU!uuZKmj-T7GU&g-E9@+fDHrquJ=3W#W~Yw0W&p@RKv$WiwmOE%W@E
zLHyT5O}AceIYM8R$iA4WT<P{O1dJ2&&;|U9)HYb4j`;9+Q`@+pOE*i$Zry}hpst7U
zu*%az#h7Be>S0`%X_Jl=Ig=YdjlJN*l|~)rbnI^~yEBmFs0XkZ|68i$-UgC^^vJ{=
z>ER}I-$lzPR9_+sUyB*9tV51V8w+4zm~o>zzR+SDBRjLx;(`uV<6SjlGot&Y6^8YT
zXb&%guyioTUFgLN-QYCPo?W4`cIOG%5NamZe)_nTBOPGg7FxQwL1DZr7kf$@JhU=Z
z!m`0i$tfd=pM8uqpQzGDBr``Jzv|H9JQOT{{f#{k658cDAzk;03!8#sxL+r)2FBpl
z5T#$c+dFd0;5}zM4+?Ty6UK|5fjE1JSJ5JBk}k@JHq>G|wRb`zox~=@oEb0X<qqD~
z(MDz@MeY*=S?LYf>DvTY5odlus^m-7otzQcApcQLA0aE8f$fUi@xwSoekS$cQaegX
zw{SXkrJ=09qO<s?<nY~4t0fnF6kAi8@T$|mJedWcITb_J*2roo$9qTGzKmT%@Mf8~
zmSYRx0gMOSC#mz9az7rHqYA_XkqGdG&xZCZS}%W?tW&PGrGKa4CdKLJ3i_qdVL>1m
zwcpr!GC6rPN92}zY8LHR1G!#`+nBC~pH{SO<hx_axn0CUf$tHpXr1iu5TZd2Ro$%!
zuD|bd%H6<G;!tJ4UY><4#>Xdo=5g_d-jTin$}(y#DS3!nK0XIRlh8~q4>)igwZO^a
zy!A1ru4{xY>EJR&|EQ^-`+nZY4E{VqdiippeZu5Z&@<X(g&X+3!pg$+6MHi72Hg>f
zgql2v%!M4Wr`7G%{oA*{y`}sHD_QFPk)<dj=493_uj4Dy{WY~Uce@rH1QhjAn7pQ8
zeR)Yr<gG~o4MB!0VL9*T(9diG1{3X+E2e;{@9_?XrO_&8YrPX`*>5!a4Mlrt2VjfM
zx>n}tILT3!0E-aCT=korLhDSFQw2M2!9b%uJZcq`yg*)@-0$fLyxCzg>3V~%6~z!S
z?IEo1pFP+-GXQ%0G^r1|86$_1uewamoSDWO4;d;|W$Lfw=^2301yDmZNq>_1lw(Z0
z>Oz85xA<F8ogrkkX_v$sU^Czy1R}MwnrX$h&r6RaMWW?K^ozBc3_V{}CG=4tKCU<P
z-~RbOiIY%6Ll2sX(Tlf)Qq&bV63VCxMsw|=$yw2sn2)lU)zf9MK*I~v_#qwza41H-
zdzQd+O>BXej`dd?uHUALqb}`!&;AuqJnzgvusvoxP!pwkBGjSkddk<HwfV(?a4`dQ
zdC#)(EZg%y(BuCs>tsJRR9cN_(+zl+G!~(YP}5@rUWk1frR9%AUkT7m*2f!38}-t_
za-tb}d!%DFMCV;NBm`MGpMCs@*3Me|Z-bXqSXLq|u~4htAN_Y?;#H)ScXIYF=m+6F
zVMOweGtK|pY7cidqaZzyqLUf&l$*o0z*68o`9^3<p!>dW4RtRMKcw9Z{ezC&-%Q2z
zR*;iHs@JGUKFaP^KMkY_MASQM+FBjtd{VA+C^f^_rn6(UEu81ob(m*^3+35RR8@bd
z+fNBX{!S`2$GI)VY7}csv4#$y9K~o}DBF`#<oVf9yoH{C*chlO-z*}+3JkaDDZdsT
zD%iZu#k~)B;bH&1FDWProrf{yds0-pDu`-TIHM%=i-gM24l3a)LRa;NBeYsixuVN`
zUYgi|sX8m4B;y3-b$fmH3H|BoaA6o98htxv{Qk8u(D3&GL3!=?VK8|<?&GJj=mub_
zmPw48ZwxaqhULJBW&s}vkwCQGP9SY3+IbO)4mP$9)<C;Sa3Byo2O_EevmyK}Gc-8<
W|8Llx+Y1i#g1EaK*mI9UPx=dL>UA3c

diff --git a/docs_old/logos/apple.png b/docs_old/logos/apple.png
deleted file mode 100644
index 345f6277e05770142af72b27d334f6d631782c88..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 21938
zcmZ^LXFwBM^Y=+YlOm`HQdEjaub~)fKn10%2vS9)NmY6$DuRLvNbf<fh|-j5=w2=g
zgd`{m(k(zhs`Q@sgy;YD^$XeEQ>Of8=IqR@kIYRCI9T~v0RS8qFPys!00aEV@P~!|
z$B8&JOTTi{HPHp&T_W4g4Mun^bo0Vx69B?a1AvPGU<=;DO#u+70Kgwd0Ms)9;Juw&
ze@PQQKsp&2oCCDD#O@XVEJzp6=~{#g&W!{%+JDQ_nO|>GyLUwBLXXhhA9C$t?#aKZ
zuSure+Mek#jVw|}uLenteYyMwdz6)1(x<)feFG!QVIg6k^}Rh@?S=8KWLb*;b#z5-
zr<U8ox(#N=zt1Lo;VCeeJzMFd?@TC7I$G(KPP}X_lHm4NCb@um<=0P@`{#gp7}2%q
zahC^QRx)4K(OyhqOWnKTjktq{K<byJfl~sm0s3wuG9$#XP(|W_gNy*XS(-+wx~SUu
zk?(WG*mcF<F&CT;06aSn@$`V^kjpDX^!>4G2pkgNr()ec&8Fs6Mb-1K#0yM_yn7CS
zMA3}_Sq&GeSI^>u!J3!^ILuwW>efeR<Djk|r<>Z|=;<SXxuB6t!EBBq-{|=IC9tS&
zg#gw49ka>1l?JiZ<0D0`my%If4ez9i$?}!z4Y&6{5O}S=4mW>aV%%PTl<E<kM_K@s
zo!Cn3{%PIv#6Uq$6c2!Vo|r%ODcQ$+EW9SQ2#WwctoN&epOT~aY#zS21n48`Yy2#p
zN`Gw+W(C{;m?$f6Q|~ioE?MnA99ZAuq&780U8q0;D22{I*M$h>@S>`>24{>#3*qd4
zUM@FBoblw|k%;q<fb*YJ>YH}_efgd=&-LU1<nxd!<v?igEp0PqPX+iG)thyHxzo%3
z%}rGll$IlQ?4Pkbid(?+Q`>#%6~s<{j}R_Nt-g(b+ZD*>Cr-r3Q2kHB3wxRDb?UNp
z^H-{Z0U8Q*YUl@zP-?!loLMG$34j&yznizUZGYX0Mgs|m$6_=y)z-R6iF%U>Qr?Pu
z(n)h^Gt5io0}d5lh|Go&KYzKh?~4W9U*5vulP-C=`NT>WMv7PbN3nMdU@^#uO5D8Q
zqI2?<ha)81SLMvfPAjF-f8B}!6yqWf^JMtZ!G#L?#arx)rTep2?k`s}<4&m}kf%C-
zZ3fN1ZeR3e0%!m(K6~0IK1PnO!HiicV^{&MW+`RXp#&Xe{UYTW68!PHt)0{0sk)Pc
zXk`cH{u-12l-jl}TLMQWP+aD+lw)_5_){mVcvBGtzV-@l-bt^R8yG@ipZ#-|0b3?h
zaFUwDUl47^LuMa=yJ4FIvkRO<%8rDzFD4>EamLZ*gjxG-Zik*jAx}(jSo`MfF_#Uq
zrCb(sXsn7dI>c#feJ)n8*M1n72Zm2%sMLRkx`t$1pAnx#w$TJMg^}F=T&;81OM3F(
z;_1W$`YtucQ(6S}j)Lw40R-ZzvSv$Hlv8Gcn9@a-Q|SOFHHc}Uuqbx7hqUR?91nov
z^ryK=M(ng@M+AMc7zgFG&-x3x=)MxuF<PSScz{=vL4d1&PZG$B6@V@WaW?8Jj3qnU
zN+zAsZ|^M0U9nCIDD3t3oWMM2q1zX2%-*D+k>FZ%EaZt6z!F;ii%_KU?}bkb!%d-a
zN>n^pN)VKcb!fQ4AOUbowu;f6=}uGk8_q-ENi>-Fn_3>akHqa&G6RVhZGL4xTPX8~
z5v%~;-JGi9dPs5Q^M8+zBKKDsdexV>EUdUPvmKqt1*w^+zGk}qwjin!1)*hrS$cBE
zONxmRz`3TX-4CoUva<sIwldsZzwheNBqgV*x_owGiPMkkc8nnPLo?yuMIE6;zC8$9
z7k_86-@q8Kmq&6#79wO`xxPKd4e^Up-(1+~5Wk8V7DA48pT^=l3e3kk#oK#8h3j<p
zA(Zd95*GiyjmEjO;N<>LfP)67_)T?nh87&-{|`y0<Wzyls|glG<;SwPM!I|swd@vT
zuRkojQC}5sI{vO&RUMDS=ps#^E&A+;_l5xiaMP|#TVV{@Ib0BQu5KUSy!iCc+4}>?
z4!iFREZny?bTA9NXWQ&$>2Nr|=G_*3L^DbJ0bQ8o@)K9|-0vBk+rLPhEYULuJ?ir)
z!2RJ8rk(d$QD&Blk~MoK0>r!1b_QQvO0o0Z-?BG-vq}0{_N3Dx=!xRpNx22*+k2FQ
zfCA?l0CUQk{cfqpvRxz6&FYRRz}1K%nEu7F@da9JG!)l|fQiJ=xMSIWF7MBt)EF>h
zG5L0rna;iYQ;K*@k8?DoN)MupyVG>@5m)xS%pnEeawM?v4JfBvc5>GxbFv~J9xutH
z_;b)?Z)g03e#TtzzJII{&F0^IDPp6N?PGclyT#K*!d=j*sOSDD&VL^S?Qe~mPcw|s
zpc+>z4fn=hN6W&!*?vI9s4!8xyu@R9Le)!(4CrhRE*Qkpc($tcH(>ZQJ{HsS1qpas
z4}+cvfr(fdk=XQ5P0wt)1L*NFnviLWHroBTn8*k4y9G+ri99Fw!u-;#PwhB(r%C(9
zhVMI1cDl-ghI-1PZRWWT^$0+-o3vK0bFMyS<6)!k{G2YZ>RwLBS8y$T+a3pIYKU<4
z@M{r#Ff=vNJHPqvFHY|7+^Re#AJV0;t!-<H-l_ZGyAP(IEvFzJJ%MWg-w#JWsk7fp
zms6}BRvEa&Lg&XNd1r8Oj7NuuZZMI}>6i4}@87}Lve0P?HPAjCQ@dg9df@J86*CMG
z=l$XqXx@q%qRWr>srp3Y-y0e?ZqsqyA6@=#x56Z-TZOKv>NY)s>`Z*2>^v!U`VcDe
z#sT1-S|PAEG)d#zgATAdZko21m6@h=o-S4!-wRPt7&%w=g|W8b+;>DTr@J#T20Lbr
z1WZ2PhBF|ZPPY&*>Cq*7P)BvA?(;ewrS5UTE;_>1r0f0-TvMDzptV)zSvdMKyZ3J<
zES<k!(y=tv`p$h79Pgvd@yW#2VuYxyFe>$InI%bARvjChyjxn}qu8SZCBGM5@w26@
z*=6Gs+$Jo6T39L+TN|qG37r#N4TDs6n)j;KKEr*s7*6P2s!8ul(_TB#$?_Oe+Nt|S
zJGRf?s3qYu)j&Wit45X{0Nk4v`1XeW5M!gE(Uk?eoAT-jXr->AZTH#tywEHcQ@izi
zU%z4E&a%SQpGP*UxQ?AM0@yK`ytu0BV}tIrTyX~a?6vE$wA!t3GA|O!-BKXgz>YR%
z*+5kirtecjQ-(xL{BNH+!xLn4NMf#6dc%Ko%zbNi-0&^A{(mHTgi4(KuSYedGARED
zioVu%Jmh}w)4sG;^(UIy51$><OkY(xyH6!;cWiOYv_N$SMq>A-RlhoUcJBcDpbLrr
zlTAjXP4%VVy%Bo2a&J=cd&@OA)lV>?L#b~&pgsm}v~*q=`Ue|df}nJa>XzQe19UY&
zes4ex%3?0DDBmvuxJ_GSGgm&v>dss*m)(;P&>9gPi(~pa%KQ7qeF!~O5?^TY(Y48#
zhfZ`~c-_yIC)%$2D$6%}I8^!Q*x>lA;3)>^S~MEbRJKBs(=JV#vP{{Z4g&>Wn()2a
zO(<!&aCm7>c%*)Hr|iIK|9$Zs*Y?`*P2meI*jL1fgo$Iq)t$V4b@W&<cgH!MThG?6
zV1e!KB>Dq$_ej}*F^ji?O3wHVMY`k@=O8$iB-?c8mZU9nulBvB_<_59Zy7-2$l_y?
zo|8M_(SU^6Bf5<?G(u7Gi2qLy%LyMut?k)mgMAKi(N(%O(7QOsM4{f_H>BQoBJPBF
z!8~d^9qt8Ul?7Zj^D_jCeG^qbmB3QiUfyX%ci-+6FV6KiRg6z{^PnINd6r#I_1s_Q
z3ucJ#Ppx>-D;AT!<v>1!oco~-S_MY(IAVJAqA6=xl1xxm;ojLp@cKxl^$A|o95mPs
zUIAeHTKtm$J;KfPr>(OTK4{LbilJ}SqG+wlBI~c)yf5ZMf~W7Lhe{X}*|JxsnBZD<
znh$E;WWt|6J#n01MaNFOqefeFLICx`zH}rs6Pm9!=Pt)!x1o7EESu{eC}iAwcK?eg
zoeo8>%!ZgA<hol<qZCs-bgfCJwiED-4&<o)Xpt-^ajz&Rt!=6J7}QGpyUVIa^m=#0
z<#>^h=WAoMIcAUAlar_DK#Ok|!;yk6YKt+%pZ~@wlf{d^#%>Z|R!ccYSC~Zl2s_PW
z)ob+Deh8z?i?~0xS|)3X(TVswD}1jgEA6ab-2N_AVSQYBDJ$bBf(1Lov&f(kgURwh
zt_U4ticb#drbNh&9YVHGC}E4Y-CF&vP1#KrQz9(rE;imHvMu_`j@-#3bOPT89o9XP
zoAdjNEZhZ$#GA^Ie`bj>#gk8f;yO399t%g{#e*FI`}9@TaL_n{ZjHL^(>GEp;$H0O
z=UDUz8jX2?Z36zT9MK_5guV_RPMsjM?e^xF1ZNH?Y~~@wK+8Rwo746`%k6UXY>U3X
zBh0+7NtbtNEd0ORFj7@?SV$z0&h>z?uDFraL?tBz%WBPQ7N%nH0`mHQSj+$eOJL2p
z(;E!OlCna5_k@_B`<bYG_cIz>H2%X*hAyaq?+67hic8Bdoe=D%*Qt9^4dUCA7ZuHY
z%8|Hwo9hg_0ii((J1+p}dM*0mL63Y(=V+D?eK4jhSL{KL+sJ0bQ=*KnIpZ3@YL$uK
z)6U6#vCOLo^~VhYl~kF>gA~n=rvrTcESSn^aY64xXs3@`_hfG72-<<&B0&t80|lMl
z<<C2Y>=uc(clC~QkgEe874qhHeVE@cP_ahh)cJ31ETpkjKH87m$2}i9-gTw6RQfs`
zVz}zU3BG>*BXSM*YB}Fdl+NR~%6OK^nM$vHWLQwTL^59uY<jG9``^R=kwIXOmz}y7
z8lr44p-<P70?J(cmCKbjoxq>kz}z6EGLvi6$w~X!Si;>cX#A^HF+XYEa+G(W`s`mH
ziGkvRiF1pmdzT2aH|Q?QJdpVC2Zv8b_;M26jarP|KGf(Frk!z|od&_v4Ze1+VzfEw
zfgNgjD>IwoVz3{6)UE5+4&-Jej^8>?cPLFV4l`BJ+Misb+jKG!xUJU<_Mm>bAwooc
z8VM!Ih>aXhPaHo@;;ex0M65XV%%ns&Y5Gw5qX4SaCZ)0Xr!E4l*1DIG*ZxFk)_Dem
zNE-ojowJ_)MP}6RvtAp_aGz%Z-+swlC)F)QXO$D+_(s>;6O`MQQF-zcp(Y#<;M!<F
zrRprbb&cJGX9n)QtA1B&JE7%5TV0g|<&@0XUpCQ;CLpuH`l#lshNTwF+1`6dkXiTo
zs7Ari(zo1<6C7-;z%lAgz^&oHO-r)c#(ntuLt$^FU-afeX8v3T0o2)uPeD3x(MZLn
zaf4|6=6q&lQD~X<-QY1X(f_hTUkn_%<)bH=xb<<TWe0;+ISC){MJ38*@L|++I#&4)
zLI~Qk;^LMQcBFo*@WZEJ+yL$8V@jSV|H}+*ECczYc1OL1D4op;hib9n-27alQifCx
z1LeUV-51S%Mmcm=R91<nBC*Sw)^7`GUuQFfvl(qV4?{eiwXWg%^w)_@Ebu3kY(q&^
z|0~MZJ1H6>L!~y_r48_-;F>e8@C0^wFWty5>O?;aB-lm=)#KssXCOy(Aq)oSK$?OG
zuUr<CRn(||>~aK@tY^(n9T&mb>=92Eyv*6y(h8NRn@k_hv{f=1rn|J*{NcVBZ(Of`
z^gPL)AXew{JPB^iO_AiwT}4N<KKKtEs+>6SdAJ~a$$gTAr|Vxx>Odu58cd!fTT2pM
zH`ys7%R3DzN{UC&!oQciK28j1X1rPoab-CK$=#IAeo*PGt~9$!+v}R4MLX3$K!N}7
zKic>$9B8k(olc6kRyA_|E0cA@TGOa;uOP+w^{LJuGHXR+o1anyu9I?&yJn5arN*Sk
z#>5!oj{l4|s1#|WC4fihJig?8(u74TBk1cib}nFqk0@#XUYZryTL?{zqK`6WbIB`B
zxn!+X;^J&cO41YjmaAFdhL-!i6mWv_J5=VkavD6NJKv6VdCg9IE0ZN`t(@TOp5Be`
z7nezSP}MU-yJb@(K284b6dLk<y&yX`^HC_K>ios?vMC&urhT3D*5abQ*6}y<^FJ+z
zoFJh>B{*G6^(}z8FVW?kWw)QTwzM<hafhF^R9mI{S%(I33iW|z9`BNIJ_Q{rlIU`;
zL$Zuqnq^XXQI-7Bny})A7UP^Ogi7$YIzbd9l>q#ola3{>Wu@s?m6W92(n_}|zDs3G
zeka0rsXvEP{$|G<j%Y4VDXYAyO49coD6%?!#sB(zmli=Ic=EhSIlxO<c5Lyb)hJ25
zsML?;`SabOTPsv(eSB}Y>=@oLooG2AW$RYLUh*weC(YK+$fvU`)LPsem}|FH-RL7S
zS<4GJPd^h}{yv;->u>8rTINqu<xe|ad8L-ZUt{*`-SD#<|Fb@w#v^(N?7OmDk^{~9
z^;ha&V;_<ULAB2LjsDB?>Qzxe-3JF0MPvT*^!gi>QvwO2nt4c^llAE#=NYdwQl(cr
z*~`+0!a>mPpWq`I_m-xOD4lL_=DQ@*bDtn*H5sod&B_Sodec7B^Kjdxw)fXgW_bEr
z%XMAh&yVHHw}+$M0xzu$+$6|kOvE>yf}_jO61L8-W>e>KqI}<p)azJHcG*qej6)oy
zEOclONat*f3*zU8v!wm*l#|(SNC5ZAxIUf%F%6gPE{BI-!~Ok?<jCg<12x2zPQL-^
zZRhnbUXNuHBng_l<9g@-`0nprGn=ecIW)t`{`0y6*@Ym%OZXtkR(<1Ow+4jA{C6wM
z_3?XlnKMd8+^m4f_2=v}tttck1gGXOKO=HyD8Y!kd!p*jj9}*d%5(`ES3NStnjq18
zt`OiCV%=blu*rkaH5P3)`75^|!Tz1`vwq7hPLl4?_b$E|6XFC$l<NE(=Hgb!TLbdj
zZk|WdcGvvKWIn?1&4(Udg^xRR2gY0?85Kk`sHHpYwf2l)`6Qfw!`h0A<F6r^!bj+^
zaVE(;8}V~L*m3x|L@I`$^lGBinCAf!r#7S<H<|SK@VU08%ql<qa=$VxhfFrg?_7}Y
zW?R`*)5T4rs0`vxxCK0Y&5Rc}YIw3ff9=qbg%P#d{{Fi7!+_{QCxRm^MuoIyx0gCL
zv2f2EeC=Pth#0sU=~ge^q;~HJ_rEgWKDsqLq`J^~gHJC0OQ=vw$LkZCrJy+HfxqL^
zu6e#w@v0%GUQJYM%02utB(%^OUYRXq=;cc8r1*!{pDzT(_gDP`Mtyz#hDNe`mu$D6
z8F@?y=ANy}6O#IswWF#-N`c#2^k2p(eR4aZ%l)x>N9@(aVGWd_^R<JN<bhGD{X};s
zBL>5p$p*(!4sMzstkobSeVkawhy}Af|Dj%GJDPKUNBT{*!P4&?^$J!!ggHmTWJiRu
zz=7vK<~w|44siF`U{pW39@YK&?>{^0I^;IEW9shBncYp?tEXunyK_@|{qBn0MOvTP
zp1eOmqS}%IsPQ`z)CMO5ggLqIyYxxQyv@s)6U2(898J-OwKRc2#ZMDl{cHW-BR6lD
zcCzbAfUcVto_4&dx+`S3xBGj1r`<bd$oZvU+Io1{$%_shX3O`Dxnhv`jdarX(t`CR
zuCzrv(@y!Nhe9?s+6!4P{LSmneb8yj9vc@3HRWQ({o+qEqq>KuN*9S+O;9mJWoNe1
zlyz#Rj5y1rlb<P>W}STnaDUd{j5xSj_OkLLH8nRkj^XnuHfmC%j>ZP}&1To}uMyIT
zbYTeQOfF;2?Wx@MG$r54I{h%#M(xvN$okT&jCy}p?FyLD<NhGGsTXwg?32dKv2kko
zC7*}w{^v#u${2JD+>oZ^{<#C@Fnv+GxiEX0$se)OHo56In%B;#t7AfvU#e*oVqwJ2
zc&E_1nw^Skto&@EoTVx$>D@m!>%62`n7sT<yVaB~L8kf!l&+ITorEq1%TYQppLUU9
zn|w7lM+?r{NuEVLp{V&g6>Wbyzx;%(=7%c{O`$H$>}V#kIccX8-(j?8Sniz;5h>{8
zV+Gl_1|oz{b_(e#^b^4P!Q2M&Y&Pd4j{Oj$Nr1n0z5q8+Zu_-qto^fmT&DH?j#Nv>
zk$WK{T^P9q;sxEFXCgb*J8t<`LC=Nz%Y_l2cRQz6KlQX@Fwte3cCAUX1?w)HYM-9J
zEE>1n^f_KZxdfCyW$p;gB;3^w!$gxG5lYgj+Ug6WJBF@<<5@dyrX@r7V~zsShS0Xo
z_AVc8W{H@a+zIo`Vt(gFGs;vny@O2lWY;CZRK<-=JMH|!dycrGiJjbBqBQM^^Kyj3
zLZ=gdu3lzZ!K4oQJOX<?a00b?GFbf8-Fc$Czl}rbIH4?C)Zb43mNV<*=Llt9X)u)_
zyVUC?%*5~Gx|iMSr~BZ9Y@_o*^5+4?JpPP(TQ{{Y#l2#Z0Q@(uHD@Od5{!?SbQ3V5
zyw0YtDbJ-F#xn)Szg#n1l%FSV-sz+@NdlXsAIY0O>AVcGixyD@V-xw8Pi$Laqzs&`
zpIo>#oXL50J6n*kwe>OCtg7mm`)iHmp&emL>STx&B^lGj5h+=iwOJ5>t~v@Dzx{|%
zz5e!hLG2XzDs+83I|p4MwhwkqW+xop4u4A8Wd)<|DLel%3wQ@#CwGJrYHok3i8@GX
z-g{tvpm#YfqAq?L!3}ob-EPh`8(1;8p_#tyuU~E!Je2qkI>)sI`Y=y!sfpM|HBL$J
zT}9l@MA7`rZ$F>CXC;Fxsml+luB;IoDxb&VpB0UxH*eMWkY0kR?O>W&LsS+rCw<1x
zT8=m;o5Oic^rq73@P>kuX~EV{DOXr=VPOWQcc@B+zaH2n&-i)I`wR|Fs1GP=xKc}H
zygY6?X(yrKn+*EYb<(S{?JMdjerL;xWrQYOD|VTDld90#&ZpM8@4tSzxOpyzXOJ6s
zzq@2g)OW&<`y+$L*M@WAmM?r<pH^dban^SWP$f@Ve>!}uCfMpA1qILXQ{CdrEqo#@
zQsZfwx#CN94&7Cuit^4ur`;@zoP*#QM<!pGFaHo<@*!(QPgmL9lcUat+SymQVI3eM
zDzB*>wC3cdCpuRm@%3xnzNG<31zygH#o<e%ZFSjdzHHS0?A$k)9n1U39GegMaDN6A
zqC2*nfMc^Mk(n5J6^C@3f7)qQ_3#+DhSyjs*R*bKt7TXWVw-Tov@ddEE4-fBRvK*{
z9D3qC>M>{cEb>2qt&n{tibnTKg05M6j84fT1WtRpJF_$^L8(DA&Wp3Pg`G^BpdCSh
z-GIo0oo5)a1MHM(7rQnD9-B}4KzJt@=W*nRXKmM9Y&2kUZ@0@5?YdLE22OQe!k~!{
z5jccE>iyp25^mt){-%N&&R0r7_HZ2^z(`a74dC5XnDGuFcRvl21{^`AqbaXhanJeF
zmaA$VSg}Wjb9a~Y?r>I+bHJx@*EStsQhC*<Lmhh5O{`pp!tN1Vk+o>QTR0;&Ibb3O
zGw@LdWP459rMJbabN5l4Fz8RyX`s3lFPQ4)*MP+Drc+)HTufxZ!_(3A(>x1%xIk&p
z*F7<S^<}_6P_MKKTT)}hZhNIkSH|bFV2N|j9{6V=@mgz?(6wDu&*>qz0lQ+DNsA^*
z^1Y!UdQ@#)2zHq#7_riksyd6~XThKs@o^WSJ;V@uoE7&pLgRYh43NNapw+3|g$%`e
zRiNJQ@@T~;g!wV*WKXvRQ0`C4wfq{cfH1dpqrUC-SzrGxWqZnP!hVYnFjo2WbztU3
z4qlr;I{vlG99dBUtnM#LlUk%e_Kin&xpONDfYl>GCM|&*JgIe-BWtY40amY>fC)oJ
ztSARjY`}(s#QzBD>b5I^sos}<7i+(i0ExRJN*Fb|I|(W8aE0$zV(-E~P6k1T(6W0y
z-(=-9Fnl4p4bQ^8TXkj6#Qa`&Mgv1P%e7mpZTCtnfXp+tsuvCn*aJ=3$Ne&pcpV<{
z=D<}>B>tcLPSw!*KwqmggYko%(+~=VUN3V~+POg_S|YludFvo}S!=zQs#_LTC1uCh
z0}I9@TO8n}nw0c}9V6B^Qtjk7H&EOm(^_ix0!%%jb@Q(90;!w*n)w<zG*y=t)*s(@
z>Ydh=0Dj)a&kE<qna@l$I~s*o1c55TL*u%Pt&OwUY)TgL2>iVh9etrSS-$3@M9@#n
zg0x6^1ekixAG7Z}uyTkXMVvVaEc(=^JIA<zZ*7Pi$vqN<GaQnH=X2Ml5WaN~Fi|>N
zaUGb4rc|2Ddy<VZ?r^pyveUrVe}7@5jig|LY%}6qBEX;S>y#w%l@M#0f8cY+A^N^{
zOig+*$yg!YUL^ZZp9CtssU)1smTHrp4_Rsp0cw^#YCTr+5X35G$HH8f4Q70LdPhOk
zg`3waHI0Dyz&0n=TQ+0D-{CkLjxve1If>&Rl1ZE>|JY=QSp%~v%Z58&VXiNxb3lpn
z>uYc<%eL}y5Ho&lzOyfDFf&7H?{KRKLAk`!A9VQf<uA`%20NPJ9(TZnXI{zM{L2Bj
z=kkc1@TlS%#E$|8SaG*qMgBNWkUCqf0vbPzReIJlZ;70DKpsmbqK!3BtvQdo{~0zP
z;lq{NW!qOPfksldS9&<}+Tr%5`aL5k*UW_J%`dGB%;sPDaC!dL(<iR715)snr=I^+
zArdgCw{}s!8_c-6$aR`st!MX(yUa%ySJan>9`|&Cp2rJ&?Rkg|r=Wg#k-M;0Avt;i
zm*+>Ajs_w3ZT$m8nQ0<qz)qBe`)s_zY+BZOuPk0~)B-x49)tYAVLjw2Nre>h(}@HG
z^MXA+LCDChfnrQs^t&}s`b?uLwT}sRN`4%LsSB0Ao-2kc-x*!xgeBu;f4h%r%N+4l
z42zpy6-@}Cs^xEjJ;Og+U~1oXe2cwr5_}y!f)P<up?SwEq7;-{RCiEVa_3~(vAu-L
zAVNjJ-5+@PSZZgWmi;_OOIhGFs=68tfuGFpK16sI=-a=?hb!EC`ST<$TSFtK+HXT%
z53$%{eZ2u?ezR(sAeeffEV4I99?>Y;zp=-OebSUE>UVx-thR}<RgZx7^b7|cGh2e$
zZbN@R&e<b3?!krlH82g;ZX3P}?5#`$m9y2_phj8aB<0au`1YC>8y>BL-a(mPk)KGi
z`$k5&Uq>t^dzI`WieENVMd!OO2v>U^aBLsWxe*3{%1VEuct!2D&|uAhMR6!wkLj6#
ziru#8hIDYS3js&eprvW)*N^@-AAHB~_QE{TBMbpl(2{Y~f1e8)Yq)}DXNcRN{LW^=
zi!*V(TYR`n1dX$+DBv+cp3Qlc5P&TkW7s$Xy|s$JZmo3R;@B4Q$KvD9oS-;uY|=jK
za`k<Wxx=`*Xz-Pkxb!H1cj2v#25QlTKH*CWWg<r8AgwG9aZMw*VOs~(C`vC-UZ2-`
zZ|T6aaU71RQB;YWy!Y#b15DKfbsh~v0To3t=J6#F_`=SLBkySCOvCKqzcI5Z0`ZX#
z%WRNmpxUt}jY(Pw0hE>?GLXq<M(aW2Wy&Yc6JBk^4P^7-Xi89gY-TQXnwMiy-tFwI
z9m53=Rt^BEg?F>hUcqoy_Wn>0tRo?c+v{kGE_{2oim10Y=H1&w;{d+qqbpU@GQZ7x
zuV=@E&c(64U`bdo*x|z+_p5P0VB59w@|N8R2MGg=t$_oUEdc)|pCU(S@nRksCIU_c
zxc423%PUQb<~E;6cC6KW_f{AaII61HFutaAA_T2Ab)4=hz7#nZ_#LE?auHKqk=C|j
z0AJ}kX-{}t$~@M{hr7u^0~X(N1?@+FwuGX={bg)lj|M0%&X17LkzYQb?8)@w7$olK
z+JGM?jg*ZzmdjjKKZ?Mfj}$ME+?G^!WBSoV_mne_Bg%KgKR@OJV>Co@xo4$8Q3g#p
zTM>1;6yh*GOkv^8zl&p!p-&htxD>_$jw)*}bclXCllGAU1U(>`&x%}VZ`f%h37}mT
z2h4RxFiL6`)Bt{a=)XuShk(?Sj$crk>8sD-puBT!iv^@D=LlMr9r5r&ygggqX8?+G
zXR171EoM^r8yFTtBN=dAp;FA2-W$HN?5!UOO^*_JnQ%dw&*uCMR!vKi(SUXX?&=ig
z7jZO>rcVZsJC5)@e+A4XuXkFNggQORL2NvvQ;tE??7v;{{L0??Z>Ut;DmRcC|LK)%
zRTAd+BpY$&BLvAyW8#f($M*!Tf`D-bEk>N-=4}-<no2GBt}*jR5?FL}VqOl4f6vqc
z8P36xjJU4%T0^e&NJm|Ls&+5{c%{?M37R&mDPI2_ZjGa>L*MLa#opo=Jg1#s4v3wW
z2NwUWFTGaJoosajPnDtC`RXQgX6`&k6(HWL<j|l0o9dO%Jagv_Di86!50c~(@m#Kr
z*2;$CoRNjIF`J>0ra_KIpbf3cHmwLOT9*gTQ2MC>%&Mr{xsXf4sZPZ}{yk;~0I)}Q
zgfhO)#4uMfytOa?b`uoi!mC`0zRjdEE1?#(wU}_MnOVX8-Ww+`Al_a(NRPg_ii!EH
zyZ{;-uC405HVQjElIGj@w|e_J5UilU&aNu|2v;c(=9}mBq90z4kz`jj|L*aCg>@E3
zV7TyMC)nPsDY`AhDxb_s5AJ<=dHyE$X3em1=q3!GsiTipkMC{U0>PJXx}Vu(%b$p?
zU2EhuWr~ZU7BlYhkOuF!&6RC5>NXzo@=T*-v@M#(iW6NG0EXT++ScnVTMj@bLoo~<
zCe~U)^$<N)3!Or6=Hc*DV@LGH*Z>;<sYA33KE0`#tH5*oQX#_J$!(HLcqNkQhe9&E
zxUwuMyz&a1v2X^dnRC$<Ged`Qu<5RA_NSNoZi@~Y@W7_BDhmJJwi@cTr|J<tAQpIK
zH2H+$N)x!ImAJP2_XFsO(ipc9-Z^_6dQ7_CwQ&>J+lDS|-Ah;j7~Tp}?M69G!Th-4
zfSnyk{Yv4W?aG2t*!+QemYqxE(Tiro?LJ_{UW}Bd9Qa8J+b)ZG_n2aS2Vt%!?_6el
z_$<v8tihOLewpUIu@H^8jVlHcFYj!=<l3?Y>|NT=k@(c+<Y~Q-hS531ca@9;R@~t*
zIZx;8V_O=iw*`1O*jHeWu4HrlN36Ind<Eao*6h2`rH(*EU2LJjhvr~b79RDcUgn;P
zxyud0;}75tZ{^R>sNNsKL9BBG4=+3!qJ>4#m)Lg;5Z}pg;(^GqMV+@uM|oj{x%w6V
z=Um(Xdhpkw7g?68@LE86e853Rkqg8ZiqBA^6xp4b9L=D=dFlV?XEdI%wOj;=4``OZ
z$7tQNzAavvodhJ(n5%rk{!>VT0R%c8!wp!>)@emL@*)uCN)_XEda#yYt|smh&&5-E
zvpIUr%aZF55L43dbS^!{q>6f*PlpJmcTr~%*(!zusU*rdg+-*J+h1TVl{g_K%w7$q
zI@W$6@D@=M$kx&9_0z3^OSljqkv#vn8s+!{4TcUt%n`oBwkUv)ijurHnzMem6=iLp
z2~s~ASxch-Q|RHt{j|0{15!IAV`=)_U`TEea~GJGB5|->76j=#tu`Sito$6@=7bpf
zkEvBVd;1(M$+%nZUj+VCk=p}f?lx%as&s9N@E+R}xB|Q+j{(E+M3H^&uK3;kpBM+S
zL}K9j2`<%+K;ntcBfKJG>*Vxdb|9vxn88x0$hZ&E-#gvhur!0K$JjFLD%J@gw9>yZ
zqOg+2#L8Nr>dNH!`Y|^f5WDl}XUiqV){prp7VNEDv3vSQK@JNe?pbC^3v1yW*nnn!
zP;F977&hk>3yjgW=gMYw?<;kt0EEP&C;k~+{AiTIARQ@>Nh5mh76?ZO+{z8lFnaw4
z?<2=^mrg-eO3Q)5JB+PNA_zS9UcOaKPchQ*nkDxE@bdO+7Je&GD$NB9e<>ZJ?~)@)
zxX1!~VE;^JcXR3eX2q>CAhD8mvPeja1ekiErp@M<Y@AJOy9@h^iriis^PYX-3i=eG
z)L7N{_XMoLcLc0XUIXs#2n8EKkos?ph`Tf5ZTcDbKI5|v++pY_qW|7re!~T;h30B%
z3HW5j*2~t0a4?P$7s~^hX(&9bN&>8~2}w(jK08(c&d^Y@WpSmGICQ?}JF@jW4iW@`
zZGIv6A^y<bU**=qthFL>6cT%gME-Sx;cEvU?ke4twMIff@WO!+aNs+}R*&ZZ&oGR_
ze_(8VL;pE{ME52eFrxDZ3Bn>wNGW=-nur}4F7WxE<k<j+97rcS!O@E+xagx{z?Haf
za1PSp7$XE}{aOL8>2krnZvGD~ztOt;Qv8U7WnvtEm<7WVxB@Q&c;Lp2ZDRwluhl%_
z=J}o}B&^ZHn&RM5z~*vM7q*)g+tRfND~lX#IG1*S{b|xwqDPn7<6~fPw<U_P^@|fE
zBma=@3pU)({hvqf@<D}B84oGfgSX6ZF5DBaPKDAH3X8ixkgX?RJ)d&;f23TZlLFhB
z%p;DR>$z}zx_dT&1&Pzk$rGMFbXL?2@X-Ba$Po@2onkH|c%Y_BpTWaR?*)VY_kThM
z74cZ8=w0J|jBRC|M#4v{=&SUhc1$cdBy8#d+(}}%SYD98La&tWvf>`ZKz^)|?jZ;T
z3l_jV%6<seQ1MZd58+afIJm-o`>h0B`FMI~<o|ApeU{x{y%*XCFE|@Rul+|c;w~Vd
z+~0yJZaR$@7$7thP_R!2VuHSEybJ>VfmG1J+<Jb7Lmqs!0Ni6D`wG%Yh{u0s#EAiV
zl^Cu8`E&(;tiTE+6%AIUf#wZ9_yf*`dlZ69Sq8W)yj}kQa=>u}%AO7w?y)cQ)%Hn+
zC4kfi51`$(3eu-}I7-sRuYhWWfo>g23Sr)cY>kBCyYF%yRvU3!e8TrW@_?%U<EB+v
z7_I{s!yzHxa0NyHsz0$#|Idl=&6<KIvY?0+fxCR|Ji`+iP{c<6X+r-=*X6;I1iHBR
z=mzn%^}_!&Pj>=LvrtOF{H&3)q_A>1$P{4!hEB4F;Lnh&EFkqibrE+vgu*43|5K~Q
zi}P?67J+REbiUvUGeVa$v(76Sus_K{WKQ3~!U<ezC=WXXR|1UQvFCA2JV1_t0WYJ<
zwU0r!oBK|rz~mf#@2Ebqh#q}+HG69_mM2oc3(0`joJG1{Ln!F7!yTc(8&5F-!<$DA
zDZc?Bm)M~<pX58m0C^9ry3og~cas(O7ef48?cx6|?|Desg9&fIMW6eR=tU((qlZY`
zvnlfB`5V0t`2d|L&)GpcmM4<H3&nspu08s{dCdO<S}oC0n2rq}9%9kHiSN!z&>h>?
zM0>y?3G$(Rib(_Y2QWDOu-1ja$;jpAz6U56IZwaW-pvggTygbBfmWy~H%RR|Xl=oQ
zk_9qIV6J-B{s5inr!q*oA1&1SKM_#6^o~JSNFQ!&uy*7g-Bb+_^pUYaq+fc+t|1m5
zO8|+yLO#)$yGKEVb2jvK#}fbwDi$uQzp!G_dY1>7v7&HatmTGSVM!XhD=4yW5dN9x
zad3_q3kr9b|HdJR0POMOQumr?HmK43uu~F&J)o75_C+>!j|?>P27Up2jAciV-c&R~
zVQ2)#kk_K;dT7YjKq$64a7-#@j~AFdMBy&msD~f;0PT0xznX;(C`y#IJ_|e;<#RCu
zF_r&h;hwNcuP8E=@KkMws+*Pl0fv@;`Y4fl^s|&#14Q||-{*T;kV2v7gg|QFyLqOb
zG31XMcY(PLaU8p^gD7pGebH0o_{8=V-h1O9e*IEYG8+ORQD;?YSoE;40u8E0*jaFw
zZKW0t(a(%Skm!aZqe=}X6|8{1P2Fl$%6=HpIeA2Pj13oe4vFoYEBF>eH?exY)I_!$
zx})`%o=aaDzQ~7rmXrH!sq)yC5=!Bd2mmkZu4#+JG#>?jWn6&NkDJ!l_I<W1K~R)_
zcb%0H$GYJRo9|#VLsTI!N4)SW=9-+Ft1}YJ?c@`Ihl7P?uS_4sl^`a^^l$_)KQJ?a
z6<%ouZ#lSdZ#JGYqa+~`3Z7K8&`%DTD7t<)yz)^o3Tpl(39-(GOplmwhVRt`+@2l_
zgOMDa$jXF^zor9uU0?;l5Wt4bzxGs{WWhW<UZlO#f65CV;56Vl;z5LP0t+1qBc+*U
ztUpQWlYd*9b8q~MfNwT#bB~X--E#$Des%7kIH`T<s<6y^@RkK&zctDW>;q^dU}9+B
zGIhdT<b`+I8c!s@y?y$nktVR{5*-kN$Eg-R+^x`SXMjbw&1SGL{A7hb|7fJ5^nv|d
zW%7OZf9y7jYp~oH%>gM#mmi0KVtja|wJY=$8Ezv=hj502LNwtR^5)Kf>@NiVS#z!m
zH&_lTIf2BA$>tX9hg)-<@X5Df%sq+)I=0%BX<=aT^?hl9RX6A+E%TK`1M^FwOS|d?
zQPc?9%EA^iNDK5jXb!sbym7!{z^pPA>F7hZ^e)TNg(r74CbwT1x$PkLm_Vwv*;@BY
zR$N{9*RsQ)xZ17ilEa&^TfuTaDBf`Vpk=rGpO8U&wvFX)RSnGu?8_%)G=aoUj~{;~
znfb2PU2*6+fjf^DzQ4i_q~`m)a<T%|e=^D<y1PDD>%zRiFH2*$^5WhD!E%efb`@5@
z992M(W5ub}>lOe!kLa)0f%Wou7>6CI{7lO!ok1(Vd^HPPR#}0^&+oY&Og~h#CLuVU
zq*T5}{y0fDvi5&5T5a6>;mWlv*CvknXW;6ug~lYx%AW-U_GoDeAEEY-_&@_cHQlRR
z2bc?t42bhuZF6gcuh!8$g?Qd(q`uUC03dzYyb>z~h7TUY)elIGRV+=f1{M77wZ3>A
z;9aF6Z1H4=y{3{?B=DgA<7K~&{K4M@aD&!osBR|oBd1!MzmLiS`}fZ?E$epay)YY8
zm{8^G+8@Y-f3>xK4v8ZSqIJqfFRt$Kv`D+kM*^J9pv)2fn_6yQ776g=<_y%EK)t5u
z(Mwhz`+p-I0R`Tas<ZROzu8-J?Ov@Sip$#l94t$3B7VSk->*Ymgfm9{@4#C&*hDlW
z#OwUoBiGR2<lg*`#>;8FOt`3enQd47CRgAZDx&~u)K0g|_fA=x+edB<mAG2Ta$-r|
znLi152gcUYMta^Z-1a5fDs7uveQTwT*(D<V>pVy@o230gqk8hZtDdSd<7SiJ5t<s~
z#`?Y4Gdy9&^-FS!A6->{TyO7Lj2_0nphVLKO7SFw&_yNbIO?o=H&_`rEaLz!1&mYM
zW$aMe*l8o&Q*GKuk5D~NRI5PhZDITZ8kkKg%5s2-e8wrt$YWVJs?!L!#|<)Wxq(ch
z1GmqG9;|zDvIVV4vHFR*0D2s?HWPR7f^mc2634Q#QA&E!sMeuo%Z0Q7LeU#Wtdf$3
zy9><Q+E*FNnVPAv;jA&HY^YW{?>rTMhx&xbq~p6=vWbjLxJ`_FoSRKD=iBI<P1go)
zE--ns#V$$gjTJb>d&?GNCc=Dgy8Qglz?J27OY${uHn6;UZZvEAjS6Zp;F1{&V7=*F
zPL%s5MUObG_h3C@)A0~zsV8U&uMUE*Gg040<&CJno*%a~Xo>cB|LCST#|oNmhdm{&
zIDHe_RS}Ku%?CZwaTAs775a#zUnXt<d&4_JpwdBw6~|U{8mtW35VW@~3qi;!&RZs6
zsz%J;rx}BPS9hbHX=O;Xs;rLz=P=W;HXsvCdaOhNr>M6s!gq~bKkq*GXfY)&`%rD7
zJN-_Ee~6qX6~XRx052}!cD9TKL{{G(O_?SXmO^=*igP~?rmD!Z`6T~4MK#np4W^B4
zqseymzIf0hf7Ab(j|_z1<Jo85YEFWc<%E@RSW8Q*02k89`GcHb>oxjj<wZH4)fA*U
z(+=y4;e#xIX{t~#UBlKknEh*8D7FF7iYv{_SsJh$6@7cicOnQ26wJ>qlk9Sj!iZ?D
zcGJ&b&6i{wsQg1^n3D_ezZ4grNFlXoU`1WQWF`vWVk1>(w|$63q^O+o*ha(P(Xw`c
z*Nvq%O_MpMeO&WCvz%Ra;^}t)J=oyMCc=yFXy9LYv9JF!ydgcl5ipUx!d(Y0FojBR
z0p*C)QB3$`{bS!9qhRH!KpqO4^#vmYw#|gdyL`X<)@*|6pZv2aepx(7-05p)U<PQQ
z@}2zPx7IcdtN;Rgr8Mccl4ifF>kEgsFK)!u8s`&*;3?N+IYc`*L16jvlknMABZ3VP
zV2_mLC-~W<d;ez?sIvGry+eDD31oY`h!E=aCt~Ff&bt2)E=zsK5vc_n=jS^N{CBYq
z_++yaQK!{z9RlumP5it~Nq)Q0cgtf>kS`CM%>h`p0O#M0Iw}6)wESZ|-zB%NF8y;J
zenwZwy!<+MZ2kI-*OcyK(i0o_k&1g>ZiZ3SjqA~6B_HD6g{*zy0X>&J$5Gb8YHl72
z(JwzASrc6fxbJop;6^iZk1Ve?4(9XWI{8w*Rl}~q;UupD&VSH#E0Z5|?Kz^i&yPTr
z4mu}4?)(`lye>bbulr=Iszoo~N6#7oRz{mPgQC=0vTz?czok_9TV`m2>d$@{g*a!w
z`K*3pEV{WwJ~dK|6L{Bp9qca5QeGFq^VQAq<=;8M&JGy=UTaRf?h>@@28vv_OC^1Q
zM9j;|>!W!C1o^6LTX5=ZRZAlRuaMrUu}pJ0H^2EQ5}EB!-Fu0aJNpj@U_^^XcX0Z-
zWv@-;`**{@t~-*Mk`ve-6?`yF9&mr@8hB@o25b`q8Oa+!;-B7})gSrO7^8Y%ZqBQ=
z&>7j1A_us?zC3OcK26rlN)uxT5B!5wJx1BF!Fv*L27Wn(e7d<2ido`0Iw$xfu75a<
z<0Txmbgu7$ncJ1IU+b5daQ86MS|oeqg{T4_3<~>!6{p+)LZtdT_yFrTVYp`p*{Y8P
z5`*isc}s5a%I!@lEG#o(Z7u=hZcHVd9wHg-X@W@Fu^9yBTvX>EnC2%gr$V;5_BM-~
zCwajL=ISVX%SrJIongg|t4$|kK=sIVrKU8-))xfJNzfA%z0n^%>(wbR$if2V)|})#
z0};2wPIn#$gMU`KJpvI5Vs2gZ3r^o1`8?UMEAY(UV`?_B6~qnt$1e?_utPh0r2g@m
zH(;t$vXU8ByV;gQ^E4oihNNlc%Wa~t0>ddBV>>zlwX<cPCPm;4+s3)_#yr`G<pSbK
zFo@qSV*_QfU@zNkA3|cG^?8h~vpa?Y081oIyo)$4s|Dcqe)w+=q=&}dRCcVAYKA-f
z!sj)5RC}imzJVX&0ycAB3A>5MWwjsbhAK&WumSgE@yd@m3yAL!odfys{ZMvyZ|q78
zAI^DI^UD(agznF4@fVYO%K3eq?TJ*SIRucsv)Op!WB=2IU#dtX)^o?tA-d;t?0vpm
zaW4i2Tv52Bw2S^77!)U*y^+9VeJGTJHkXCCpg-Qp-X{loEON%lp}UvD*M27w2W%KY
z_Mf6qPTH*@#!zt&Uf}-bI(dOpV=(Aiu&byy3YaK~Nl$+JMFOXEzNvy9<>+kqi3OPt
z*S=(&-lGkAxQ8aTHh1rAyMohH?lL(3&e%>It$MY5M<;9h_W^)IZ2i5rr)OZxxllP<
z^aS8PoOXM;xd1NY6cDLSSm53`Cg;B4yUyN<*G#`!eFP|fJ-yt(XnqJXKoz$+0QmoG
zF>=!8<iN86m-uv;v-|TcFQ|SnBB{5^u(rn%^a#mM>}J`+I?ed@be|2pZs}SX-KkmQ
z1)LL3WsHF1b4tPbynGPP569@@CV&@~>uj5l1ah5`Htc{O-=8=r4Rg>p4+6md<5BMQ
zBBT&T_IC|GVDCv(dQ}%)Q^LipsnJ$>KL~|GWahp=zpz7t&PQ3VCx9b}N_o^oBib`!
zrY`d3J*hqpR<4+h4rMT|lK&Ol;G|y<JqX3iM>m~eY?UQko<U%BR{zG=laa@_f~!w}
z72<tDbt7Zz1HxH0BvvOvM=NssLAl=Xy=uW?4g@~Yq4Q4G)ltY9XQ-hx5_@u2^UBti
zbbB^p%G7P`>r>Ff?C<AwdSx|3oi*ri0`)a}Z#wAlj3yUG(q5L|%lofeX1b^qj`~EM
zZl+x}_-T4`Q#1KNH4M95W0n5VQDs%P#g`(YN>mW|jIB`Kyr;<gE_Ff$?!f`r+>El;
z8TpopxNxF7)P@^`t?sroM7Mmu)X_U2rFQEIxQ3&mm3p_=OWJciS}+-&SK*z7_23Qp
zKkV2_qw=0|pl~v!(qqgQSj<8Wy6j3^5RTfW^R&h5!w}pBJ6DI0t(x6kHmJ~hrg!A)
z8ieXR>_KLn&cRtkm_4I*O@@Ew8Y7Mn=9@3jQ4!3&@m1a5s;~cFu&B!GR&lopo9qU{
zlq8uwZi0{;&WXS9jzOVWg*2zLY#BT7^}9K$xF}rP5vC-)?coMOi0iaC=B>lW8FktV
zLzSg(-Gb%lp~~}NZ?sTJ-JP?uWo59M4imr3dm_gl=#U;P-L~u-gk72hd4aX{12~s#
z#06K_LGgJM-opq7tNn6OH@yAk?D9hl02h|r_1%I*9g0gB#5KgxB<Q1`nN8MyY&18J
zG!4FRqRYePI4(O}>V;C3qWm}O_#^pQzPAE_$AZ8U{ym}+Dv6zZmhbfhqWIk%YtdeT
zRi`{L%(ONAC4k_o=CdJBfu7_Hr)yvtEP+0u{K*=ApD`mtdPM0`qnD6ZNAwTUAboE+
zVd^MM(Nnn=Ov!xiq;y%W!erBQ&XePNL={w0z(jfJD_&rZ{#|KVq`D@iVRW=LH{zl^
zzy~$=P5o@~LW7G2J{^5wNT9N0(MNu)-{S7psW;WhyUGBo!qNOXI~l%ALKo#I<@9px
z-*bJphB<IV4OG&p0<*+DfHjFA@U=x_+`u#c%9axdEQcoi5U+txe%QL}%%1ef76VXO
zd^yQZyJ#i(J4D#iO|%J#i*FD=>DgBqY;aMOFN+y{jsmNy2fMFluYR!?d*ZC8?`))n
z!m&1}6nKp$1gEN|yDVQwrx!#^BmHhHJ2aM5zcVHYt)pNOwVXoD*<3m<_SE&KX6Dru
zE+BQQ-)^8gV?pMQa<ArzW4L(rSA@1`0wk||IG<znFpzRJRV6o(A`{5Zm{%}OutK@%
zQWZz5gon(|mwYc^@lm)Je!}OomJ^V!U1Ua(RRC#Ad6S$pmuB|Gllc^hIz}eEl0$a`
z`o*Siul3zx?SfldY~bkMw8p{sy6HDqiwf6HuUDU7c7T(*(I?{_Y#~zKzHDyU;|V01
z&ic7}juw|3zC-FR88E9r;KzNsUfHZ#9zXi16s_jwzh6lc?yvRvQ5rj+n{<L+AJb?E
zY>U2{4xTxCsdk5jal(v?c-&-rUz5GtgXh};MKe0|LY4n@>9O(>z0W2c&j)m2Q%}rk
z>JMX|34tI>spbl*btDTc5{+s0530?+v4}t~UmktQ3u`rr6I0!{=4^v3Wd40RoLRPV
z3K&k{a)-7MRUWmZnxmT1;LXgJ+0R>Qb8qY(->lMC?aKiYj|MPj<g53*HKdPCxGCvI
zBk`w9yGQ0ihRRv>F=D+{8L+MDKx6LNp=I-Jvf;<g-@}Dw2y-MYD`Rl^<JA$lI{Wf>
zN4aqK8q{)}gThDaOP(hWva(?R9?$<YxZ3w^H@z|9M2A8o8}46qt+=`cQiGJhOsX+)
zG>jK8s>m|eF2AWEvl@gyFT(-uzPjae8e`3a0ux@04EW4^@|5u{Dk9*j`sh1rt(++r
zz)g<x3lHbS{~1e+qwcZyQ-DOOXXTZ!!EhG)qlqCN4l6iE!0Hlou-ku5`ysjOQ-q;h
ze?O2&S{Ns<ojiBEw&O{cz4e7?6rL-;duwiFZ#}!5|J<%$ePzN4W~{JaUU<UoW2^5k
z_+0Rzgl!1|?{%u6&12MS;?*UMao>5EB*`)9st+wX7Wd_9N=HGL)POMsg)h@iA;$Yj
ztZrI12Fz3ZwoU+Vf6rQSF6`2<@^&gnzk|83fWn*9```Y^L%pr;$-TSaSGdQF&DlAn
zc69popBhCUdQZrf5RhHB>+`0zMMLFV{x$20nJwW(_V_Ijz1!9C_4fSz$q7{rQ7l*^
z``o5@HK9II$8cWz%6VYpYNOiy;=D$2`JhT$jIpf4aVG48NHHqU0{%}(a3k3+Z*Ii^
z*!*ht%FbfNIl+z&=NplSaYZv^Q$p=S=Dw@M$q>mZw0;ExemW{OB_Y%?FcNEO<<pVA
zB?e4ZomD%=ole-endCT|j%M(|{}09FYYHcK;6}yFVeGTUBtE)ej_gY5NX&l-{QCZe
zy8rp0FFFYK_&e4*72Kg*r4{pn85hGOfXUy_BAm2Ucn<iqd{K)VcY5OrS#Avx^tJgB
znWC|J1s)XIy^?rgyGUM*??3BDn;yZ`X(CUw*Co)H#5w&MDEFN7*Z<2|X{0f6_i>j#
zOdbn6bnj(d-f6357P^?;ZEp?xo|IcZU`p8IZXFQ1czSXpAOSW$ZhK>5qm2M<@fr`M
z7B*>OKf2J2r9TzgAUEPZ5L9hpt0m%$LbN1t|Bf}Lh&@LYNR(brbK9t6Sk#xFXrC2l
z0$US*1+NuRyQRSBRBdSv2W;6a2&CdQEs9XEwbPZB{t8w=3L-_LrTY;>3>%gjldQ`k
zV5M_z;53>T*b!S_7pBDK7On{6dvBH5$Fw;-34Q>Jw@dMcE6GR7=*W%fN`pyP(r|uz
zJG^6;t<`&l-ZrYBRxwTx6k-BlicYh6k6biwGqw(oBnMU&!tPSs`edx%iT7f-P`-4H
z%Isq>BXDJDwDJV(hZwD;Ty?zwGGD!d`7z0v4+1|X=U%gupF#<g*DsfAnvb@rp%irP
z`vr~)Gl9upLz`Y%oYxTw|F4eg3~MUs!Z#q%RaR^?Q7kLHhENO;5@Askq==LNiUkso
zrU}JJ3#-5i)*vkiL{=#hU^k&G5EK=W6as>Pk*1#{5dl#`F%<KSKffQ}ulr=4bLO0T
zX6~KbnfHBDt0j%f0B0I==}4pLaGQLL&w|9W-;%H!Aok1oUEW_ljB8VrTFt6MkO=52
zd{2)o6r<z{ZEAFh=CquH-p8pEW-Z26JYc;ES`%VpV>m=*7u=i+l6kz?hK*t0@nbb<
zKrS9*#lw#;?%zlmlO<Yn!Vm!MbzOGI%Dn^FnB>|qA8SAtW}Zae4IenDnh#qXG(y>u
zIv~&UNMeZ3&(Tc&_ApRU@y$B|DL&4<a~n2Tnd>8TSB4s5!xxV=DbQ<uIlFP-|6zml
zn6==oVeuU_fItq`q`@vsjz6|q9aij3SR%ZPG=)<FWaKBf25;uN-h@q}^2h^l%Dc;!
zMfei~?%ZwI@#A%Up?e|YcK$A}wh!amuyXbt3F|3H?gJ9V%BlcR=+a>(-dar&D>`-W
zwvV)98uqEqS(>2o3%v4Q3G1k8$*PUbA~ZjAZx>z<!aj<VR6gD;&x7FqNx#wa<Q!ND
z>m=#}Xp~Yrc7)atv}J>*myP7D;pp;G%`{A-;N{%Xjh}tHb(NFLxO!nK?d@)j+aWsD
zA!r7q)85{8CtvCgASm3UGT#9YFPQnorjVIPeK>%{*ks~CQ=c~Fu)lFr$ZlZtd_J<c
z^>Hd~5wiTF7|=Pt&Bkf`Fo9WerG*6mgY0=FRUr9%sca+Va5G9P1dV|1z56J%=?6#{
z;31$kTSa8S==J!J){%hxoQC!EOS)pZo2nu<yx6n!Lolq@KS6ZW5z?>1hzW7FrxKx^
zQ(ta$shtDQaV9euKR9`91{PNTpjkH%baS#iWNIX^&bx74Zy1tzu`?SqIh<(v*^Tmf
z((gdCfo_?S=RK9s2V%eYcr<=1IAR~kbwUQhu;+t{Zz93qrrYdm6{XvjfcY=}#95gu
z(113|!;_P=Ya~@h5KW}oq@<wFmh*6N=r~uXTbfkDJp{60Xpv%*ZK8n1WG#c*4Q|}U
zjIhKG8t+f|jV=z*A2NUekgxlq;8EP>c5RB7R-!nnwt-G@2%P<yUXsTEIsF$pJ*9MZ
z0IiS+(Pr<ub!&ruZvdz_AMEEO4G{)_U<$?YvcdVeGL1-u$eZbNO_4fqN@Gzmn)pa#
z8Wxs2N%kuNw=}$#`C(icTOE~CZE;c=ii!=mwe@)fFet8&p4_|x7>wjy{rr6d10d0(
zg^!pwHGp`7Y&t+<CpUoJP4z_<dYyy9nsGw2gzOK&AXE(sWq2DDPvnqSo&NQ(HLP*$
z(@25NOQ_fJDd&>oWuz-qW8<JQVg5)R%L~eP8ohT4=DAMgp<XL+(kbcsLiSMx%f!FB
zf3eauAX@Z^_@0mFF57K?J;MyE72rXz$3mf*`t`HX06%Q%&5Y>;gINVMJ8Tp^e=nM#
z8SBC+x-w^nz3Y^)h?!#P^b05~E1OlPB^MMut7y~aR0~~|p<fdc76q0FeJE?H)3~Un
z1M2nc!8ft@X2||o*PC*cSb~En@GDk!Vs`|;v>JLB6Tg`^17hF7q4)qPqMvz`Oeg`>
zaav6@Rr^DMZrm9f<52|<qtpS4`+No(5a7?7ay$``a(f*q=}JehR;_~mEj{U(&AwFF
z{ICpVIC>5cmU@y4sb~<~%my##Z@H>KzaQQ8cjRwyibLzLQ>rQ?ia+YxU49Pwl=F|!
zq$m4<8RMXO&-+!K6;P|8PQr@4K3oaVKdG*)8V%@*XkeNa__EjU3#f!Q%j3%yMXZ?{
z??*<e5;o3_ZzsK`0jZtAZ|015(4k5yfmNZ0gYk!^ca?1V*+9#*=$uJ|iL2~hiOoQA
zCtr^W?Q%N64m(=rl41q|`HbFAMuIgiMtXhKZNo;2VXN)>kQNPSY(gq~0M4O&0P&*B
z7y+@*ewUYTfpcppy!X%SsR;&y5cNQ5S3)|7lASn*-F#93njpzafJ+=PMjQOfkSDo<
z4wrMG+ad_i#skr)eR4eza`$M|KA_TslAUqi2Fmk<1%WL`TUzmw$OfeNrgh!rH;&+w
zq@Uy>*y40P?oax)v0MzkQX?!YypZaLM0i`s7PP!P64!iBA}y>qe-B}-)C7)4sXELs
zwQefTI2-Tv?ez|CBGy#>ssiPJ@HYZsE21|Bt(pH18Szo-Skm*IU7&8^P$wSXYxYi=
z<~m#0^Fjgj;j3|x2x)Ibc%-})wPw9<mcIb<`Kn?!v1+^zeg*yhcJ~Xip@;aoeQt$?
zH8OA+h1_3_&t*`3SOJGaO_2Wc?~p>fgzQ7GcCDzp*~}Xn1r;3f*M-u|xib&64F__4
zSn-3LcR0H#g7fj0_M`c27QR6BXzaph5G}W(y@*X+Ax1=Z*EC~BA07P9S>fIyLK~yB
zJ9yj=s%u89gtCFk5x<+(tRHbbjeTE><Q>hOAwTFKAP#5|iXw}nnB}@U1d&URqYseY
z;EC#lj>=X8#jT{|X#R1sK3a8ZD3CBUeWzwdt+Hl2={eN&unekXznqgy8)O7baz>jC
z{luTyLEn(44i4ib)6^OZf7a0B`Zb1R4X~naK9hysQgbq~r)86zfB3e>y>@1uTI7%6
z{M0McI4BEA)*%=+X80RV#?&y345$UHYU8Fs`%G2lk;QIp+=={327!p3S!q_FI4t`=
znH?0U&WHS<J{9jQdN6h-EMAa?+H3S@88KP_BfEpLEMK^N|4Wfq`bfT%m{ME)cJWE)
zB)?n;jxOr;4L|mv8a)#jfLFU;OU!2IX7sfhq;iF1$1QEl@LT-WQTNLE7O5!dcyl#G
zmuE;A(|a@eS}+xoC9=zw<dR?CS|5>YXnJK@lTQjU1(T4Hrb!DHyCCif`cX!NiGV}O
z?T$k95eal#+pyS<cu{*husK}+SzRiM-B@<uG+$G8wc$qwPB&B36*<*kW9iPxL>>oO
z?+hKkpLh10{=XcQQEFFcp-EOoAo`W)vYq0_7`#JC>CYmazsbc3vC8Rj-YJC`JasQE
z%yY3jlbI>}`}q~)SwT0LbiJ3eMfoT<<#gn}3Q0w?(-srviCf$w%*T5-N_u2i-9KS8
z(bd92ZH$tvFegI#%(yJDG03lTVxB*^ev=ewrS~Zh4}~=r&$TiRWHct~J{8o>#70+J
z&84x;c)yg*eCcYHQZ(qY3`x@X!FS7T^3~}n>ux24<*Ny`Hs(#tsO&+ey^|5)ihGyx
z9(2@qvJQs7)Ar1V$P#apTMZ(~x9P_#ho&%d+Kb=%`$_{DOX}}7$|6e!Jpz|U<pf;N
ztE=(;`=2iCdsFtS`%|u(?R_N}CgwYNf9vGMUibA9r9IBa*T@Hfj+|{kRb&wcpZ8>{
z6?1Zwh~lk3kz>8GW|i1F>2|c{eoP4a^vJ!^L{`pbxbvmT+bwXp8^4EHd&?(HDXn#0
z3^&V6ep!;LpO;HaQ>lQjjfAs;(_=&21iW2qE&DJM{6kYE=h(|73tO$|r8pGY?z4+(
zy>Ci^-@;eenth>S>;e5+9jQ~(kr_#kR93#i^8*XvnJ<I=D+@@}IR9MBmK~$z&BFD*
znA)$w{zFyiGo%)yb;}3M7Fi7XM=VO2t2dXWQhPdj(eAZVUIZaG$)sm$tu1JwNmK0@
z{&LE6qCmlBWvwWpFd`!C7Wda5s{AEM=XBX$Skz4c*`Y(cj6^@}j<?+)w8u1|F4kgf
zLd-@LxlwLkzMfsovs+2F=6%*!?Jq-Fr@A(!YROldO%p$>c`6m7EL6UGx_anJY=ns2
zE$c}C_&yPPIZkpKgr^9U3#zcpLH34!b@tSc^(e@TZ*Zy#1BL^GOY;%7u#>Opy8?OZ
zH(-ilLhHp9Lm855BfhLu9Wzl8&bi;IA)E3SDKX074PMNcE;>XjPVv_Z<|a3MPap<6
zObiZrs<%#zWQ}VjyVr6vnK#Pu6l=kd)H8kdk8hj&USQW4g?sQs=RH2lxs2MhaJm)I
zv9{(-oC1j$<?u=T#534cbY3-JOF_<*w2JWn7f$t=d%bGqeN1j9DoGL|*TC!AQf_9J
zzj|7fFZm1S<=v;SDRe2xX6fG7@uLaRTmq|6mti_|u>Yw_gCfZ-p7)-*dx_L+u0GM{
z_U2msTdojghp#xHT*)FH7YKsq>ha!C`-n;I8u8*{mt#7nPqWz|U;KO|PrLhY;LWaE
zLd0;$3x&L%(dOU$Cgy^240axafzJ;AMb0y!k?W4HZ}F16ph6?ksqzc@8$%yNQcYN)
zjeOfE`L2?j6;fPnS1uX%LLZ{n!9#h|c?j~6j2-X8P77`}8`y|5-1;ncjFkPP{I5k8
zEHH_#nQQBTBmj8BJEbuoN?PCnyAZcVPC2e#G>b+ghy){4E$N-Alw;xkgWS(JbVSGk
z<F1d^X?G9K3-SRK$=#9CR=0U0h_2bfjBf186Ykd{gl!)u0OhF{&2|Z*T6@!Y!$lc8
zD36!*Qzzl~Jf6i*5zp(!ch9zcpP2+~#R(a^;G1T?x->*vz;XmM)+uE{*A4DF&0H_d
zT%Udv-LHsNAr0yjSJ_jK%w6pKr63?+e>OYCX-uat_o315!eY%rT8W#Bg#j!5ngV6N
zCM}fZn8#z{9C!qw@x8N?U1gfoeRXC#-CR7tTCImf?gQ!;+oi{JCm!A4Ax+;ahTV|w
z*;z3|Z$o1~KoF!8>+ByJayB;9_B=5ZTp$Z`3#5rT%EZF_B;ug0xuvbuK_hc>TXXX)
i(hlzbI7G#SgyYHo-(kg+0f4+fOVZi>_gcp@fBY9z(lzt|

diff --git a/docs_old/logos/best_buy.png b/docs_old/logos/best_buy.png
deleted file mode 100644
index a273b3647904ee2342e4ce2d65f8be40f098a316..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 22328
zcmb?i_aoKc|Gz|JWEPPT*UH|#Z4s_bWbc*iy)sKigli`&7oqIEm6WY(@5t!dTlV+7
zeEx-R{ZQ%kI_JEe=lOh|=i~8woG49oMPdRP0vHTNj8Kx-hQV$y!eCg#_&32%<}oji
zz#lkvvTCw0SVbJ+xdkrxH=~u3wi*oP&kBQuyoSL}!H+`z!eCx}FxZ+o3?`NagHgGD
zXw;Gb|8dh&MNuAh1^v!xDNF=E!Sh6@J;7VNiI2xb6DDZg0E0b~K*-DL_)cxk`n@#v
zYPs4yUR6!GJ5`yS-;G*VZR|r)Miai(#xW*)!n>yHcE;gb#$KNjV%&LHS<e`T&7-Gj
zJVE*WPVz%@+!z^wVWC8OL?UA9F~<;Eq_^+n{PMXhy9xL!SJ~>Hg+B|-sPy#XldOHo
z?f)`v!9vK%W9cR5UWg75ymB>U6qhGhH}Q^ax}AWLNmO$`fhV$dSI5D-bvd0buC&>Q
zhmC?w<Eu@z+3hp?#=0+}xUbCC;YPG?7J3*hIZIAU8pXsJf}&Yquz(+tFqmeHZJ!eT
zl$n#--8TqAE==hpYd=!RhG>4wzu{TWXOp`;hkQy;C<{pDT#OYxJoXt?h+*uEU`?8O
zb*|aw*`cXQ{tZjKsTdn8V<GSH5R^#b<%3D-y}dI%FH{P`J%tCRDFcI9IsN)r+^8GZ
z#>I}q)^py;{asGj=7=L3c}7mSG3eaLIiB?*KOP3d#ewlDvkqoSR@7?l8wv&|s+Og$
zhY)1A>8UiT<X+@HWH=!nPlSD%z!va&HvHI<Jtd#a`5EcKoyWA1ME|<f{^0F7OizoF
z_U(v!UY-X!oDnF-<8&FOk&G`rtM*D-Uwx|(N5B0i>%Z?y+esKN<O_o}WbKDauev^$
z-U#6~<IKRnyOsg3yu1jHi{20Uz~2(&n|svI2T5cGy@4xRr;5^OxIe1LInuKBS1lby
z`oq7t`NbLRq#^#(o0zsc5$Wj}8vQ{*sCq<-Fz*xnT1wkgRdk#(D|R>I%H6=X0O67>
z6vfi=6t|nIX_~mT=%{vXZbj|;P63T=zlJ|)ZKba!Q<F09pchw5_I+i!J7h_(WlF!n
z=~(SU;pwXG^Zo)A_8;^(X*5a&I5yGR>G?OpRYVOg$)azC{Bm_=;?K2hpL0SRXxE3x
zzu{t<jItH`b+p+gM&oem_xJMK+;UMH1u={OKYEsO-hDhRajDc*g`Y$IPq%mXwin5z
zj^^m1rNjtHjwhX%!{vj~gq=3woGKPNd8PmxWS=x;npCyT^q1||v<_`XT#8R%A=qKL
zkxK8`CM5)=tCU~S$FfbUo*MmaY?2oEC;l!THbLQ4g(Oap$@U80o<7ZyCMugmmJ%Ip
zUO5Iz_yy|fF1$~6aw>_$1-qW#1@HH;&q|k);Q74e5XrXfg&FAJ`-geI%pFHY7LxfN
zmv?7}Zbv)Yve=J&H@Lv!Z4pdW_{*i{=T>{Gp&Q6j)pY_(EX`tUrcb_qjKvV><WbqQ
zk~p?2eFhgXhtmVxGYaC6)shmUdJUm5Pq$=ZsfTrgk$e18FP6lZJzc<zWtjq2TS81&
zGFY-jB_bH>M6-;tfimaks&mGooPI<Vu;-C%mx2>@!RdX{y!K|~_okT^(FO^{N&$gf
zyEV_5+!zyJIUzasM;g};7>5FXbuCHf#j4qjjdIUK=!f(OI31ne`{lbw!gCgdx=GGa
zx>=nb`1n;yY}b1!!N60`cS=~`ZPog_Jr|AU^OwoQyGwQ?>Tp(40?o^PIXSDn#Be@U
zguk#G<)O$aEZyvVSyp`XfO2z+42UmHb=IbX{3)#U7%&<INETU2A$w0n1^I<(G%hDi
z%#u432GRT~p%*;h-;K_;mJi*awSKq{-&YY1l4Faw2;D;urTHq7Rw%-`nPQP%%A^%z
zZ@M}atCt5NV8A4kkCrCCT|WKOofvh4-Qw+C-6ymZsg2VbESnmPEH20<lUY)gxnWY3
z0jAq9O46hZgWQ83Z@QNHy<%gbH@Jsa1~Sp{Moqr!U%a#wW;{P<T#k~e1ceAes}G{c
zTt8@cn1bwuys1e81;gFFgy4FNKAPPqapvISxrCqipf5I9$bVv*TjNjubc>T@3SW^f
z!lK7VeD|zLz6P+kb-3F+;x@|+&XLN}m!Sk%ie1s=)9tTgZ}gGVKG;v#X<Yp6*e{-;
zgHT#pi>yXpISim@^djmpb-(l}R+Ddh*O9X9_E8DSLp?-)Qu<T-0hwIyyzn)Av=a-s
zx;sU}N&eyPl7^o%KHt&WpKg*~&pZ^qy^ff;NB`-inTH8J2(2EE_la{0e?-p{qHa@v
z+apuSvOo(jPLlqs8(%8fKOXH@79qujo~w9$Kedc!hSo`<IZG(y*XYp1fvw<0$fi`S
zl_}}p)}o3|iI0%w@9n*AcJ=wTG3jNE74z=7)nfkEC52l*o;O-ct}5VTRpNdWHVB|)
z_O0;Zro#SA@o@GKvbmBEl1=K)bt;QATpJ(GE~Xw;ykdk*oz+J}c=B`Gs$v5E6{esd
z<bU8xQWDG@HB4W@S+}WIBD(6XB0_P!z{&Gg^BG=>D0yNn*tfwSF7*m%Y-Br+LXU>~
zVd8v(dLiCVXyQA3+nCpfH@xr2!oHO}m^%ncoYNqmxR+Sk?}Zv1TKw*ZhQ0#Vi!^hT
z)WuTq7B%az?O^tTd$gqYve7JE(Nzj1IW*tH^4I@_fBmPX2FKdC@NOY}Bb_7^%3dzF
zlDaIW&5*eC8rlP8VN={Ivhi22fI_L1e`zQ^B#w7`|3+ei>gh|!T3D!W*)-*Eqp&6f
zA&;}P&1So`@#&?KSfTjvsn_)rwE`Hy57D?A>E)yi+T?Y$T}PCvN1?`wYew>52JB|d
zWP$}@qelFY+~_Xk+nGZVWV`o4ucyC{;yTH-r6+g-o(N@0#NH2?HE;q4PrJ9H_FtC)
z8|^km0^s}Rs{v2V%wJ;54U;<fC!+EY*wYh#?|yi1lyxbo1)>htv=sHc8AG=OBeHPf
zZ+lHy9_u9#tgBJ5Jb1kTBEPbpL8|dEm^cWK=4+epORTJv_1K_Eu05w*TvhFt{+@3b
z%o{zdwwi#QWvBalp~PPRqZkQo$V48wCBp+8^Rt4k6&AR70=#^5uUKi6-()i1{@MaZ
zY1smSr(qY}SQ7(N76~Ls>{>(CxA*58HWt+X_rO5?OSMotTG+G_oLrbEm<l0N8L?-2
z(PO-lKwk-&`3>G64p!~%K~9S-x&y9-nUxDJqlsyL;dPYz<JJwZ2x^`!Hq|W2047-&
zuRibPE7FH^W@nB&HJO<t(odk+Y~LxSGy5bKN*%&&S)u>Mx-ee;$=MeXjhC}iM@Pih
zmWYAX{npop)l!$cl_d#9!5=23J4<<LZ)SKjuw)>vXDPyFl72%O!u@y>fzqsSUMtNw
zFI`z#k^KnMEG~|O<+r24jR-xIXw1WRD3&`lL#I0&n#8pkwD6{7VfnGljO=&*6!vFH
z$5LxzEI~*#ykm@t=FCj>_CZ~$yu5WIhsYI2ev0v_t-M)>E;oh|;f0L(d8{f$w2JUg
zf@^k*1Z+MQY?UZ3-*YqINo);)LafL3TWRo$K#^W-m&zze{)&IjsPyH3WE?zexGR9;
zW2Kee&xotj$XFi!A+TmVE9D?QT^dee`1oWh^5m%nsl=uQwLF2WuGGgHVM8SQGQ@YM
zh|+oElahpjDCY}RPJMY2FInFb0TTzH>gxOwQEe_4^H&P1Lc#`f(#r)ti0Jqs9$U-f
zZ%P*W7mrI7BQ5$$4!<?kDXuBHGJ3^HJgA%gCL?-myE4N)t39#VR{mT|S8n@{qA&Fa
z_Wf}s|9pilos}?SyXDHF?KoZYjFMBmcf2|+MH?&RgsseeCb(sU>wOm&%SX9PC-1#Q
zJSa(tWStEhN>ImRqO7gX&f_Zn97P+}qW$t%_r?$t!IQ%YtqE{t=MG`Cw6}|$igRj-
z%U9j2Rwt+Rtc(2bZ}(@>ucr3eRA_AnOZUr9Jv}nF3Kp7meLMQ4`Nq~)$LwOxi)U`i
z3uT?Z2p=p6oShsgn-DVRa-(-6``e9T=pNOOdbX09(4{VtE^!QtSLJ_9ZK#_#Ev)-W
zx5|BTtXE+%{od^6W=q4n5m8W!{h|Ep+R(~oZ_az=*w)aN&-N|#2Jl8|`?pl99_a@s
ztO0E(>`+HLm#>P4GY&a_M+Vb9;!zg}%hc1E3#SIh^-0Ksg=u)x95iZ{DvL-O@T%Y@
zEcV0n=jpMb2VGl^?_CP9G-dew8^sNSpRcUOcfL&*dPk|*&2$nnOzPvr7_YwBUEo6Z
zmTJ4ZcFCQt$m8ZW^EWfCCAsq=$C&o>8WXyrCjHmYnl8wZ2(xb(eclyYKW!Nn&U(7X
z1mbop+3asJ{|WaR(f8UI4OYwWCRSThwOz`SPDPodAGC1Ef*cf+ZfrYe$Fik)@b%^*
z?JcZ3EgkPPWxOjA>I=maFTMF&^0tmfq&?`0tXMLcG2h2?)1DbobFCT9P4(^FkiqJp
zZptUUc}eEfKq67AI+B5pb%%ST^>umTDcN5n)2yw3DYKXEb8nP2iRAXGNe`9jWJ@#4
zcQklYR_m{m#=KpD5yH~p$`vGaQ*N@2=Tv5#0aL?1REHvx!;PvLZ1*`atOz=b4zmQK
zUt=9#q4(VSi*ilm2nM&zCWcWV(WZE+!l?2<+gg1V*Um?&MwS1KC-@&f+7>VNdD8fn
z1rHnriy5CQza0Opeh-IxY{PMNUXuK{9k%Pq(K7ES(v+2d6Fg#h$Ny6s`KzT+Eg4M=
z_XlA9G-5?1=p0$d(6doN$Ubz%!Krp7`1$T*f(9!tNY3W<tk3p;D9MKZT#2G2)e-V=
z<6NUtv(2?l#0H_wXDm)<`+$QzxtDuVmFSb>fLM8+wQAva%L)V>3z(fq5zCEEnIK5|
zUJm*WlSGUM8|%^ZKaWT(XgA-PxSWJV^I8yGb|}f@mpuIU6jeU#wEUeocHc~v93<5a
zQ~i2ZiyD-kuK#TdczxWj5aqgtv1iDy1LE%k{&6bfgB6rhjeXu0*F|66q5oEyWfNyj
zP5j9-u5frz$xww{yrwPSU&N{Y9??e_I8cwCpG*{TO5yH{Qxmj%(6fY@(6|)P=2a~5
z7C=?PBdC0T@Z!1Zc`7RP*-O%`Hgh4^ps>68HyU>xe~$+Y-3p+@8~D)RDvgZ=hVy2c
z^7U<L3JiyjB&_S7!uX+w--T>s1uh+>hbmp|o%#+_6>4bQSAatwBT_rl+v*yDif$`x
zX$m$%3O)?sR;^8<Jo+ii+Qx3?lu2B%MEsPN;s@C+@Lh}?2}&c5wQhAiHSt2#90H^>
zGDPidcT{JP{@Wvs7|Gme$zQ&e+%yhi!BZP2f&enGGv2X3J?OO1AemrL6-yU_H(Z$M
z>8&T?V4WMe6x5iY#gNo@J*+R=!ugEJV%dg4J<bBA<gx_cqV(i>q-cwP8J&=dp*#%|
z)z8zEL>)q<=aKGaGP0AwHoiJDLgu78M-|I9Sd@|M$aoh<9i%JIS6cwi3TB48Jl-ew
zkWBE(<P%*<I`03%s5>I_r(RJ+-GW!x(Q6cnzUV8}*gIosNy+DXT=Hg}8u0@sv^1tG
zQN^Z_g?*x|DWy(?f&_fE^0PpR3dZ=rdeS!xd<Mc%&FvJ$W6Pa<=(BR~z|D|Swn#}q
zmaGqT#mu83cgyH{wutLRzdcU}>-*X+4Hj3047zvovUl$2fE@8)7l3DL67R1<K`*R^
z`x0i2@P9PmCeD6U@DT6Gr(pR$5je3R@SkX^y2m^5^1MC4ij^{lv14f#`(QM+E<Ylh
z;PIJGphQF0l26+Dk=zmOaVm6>E&_A?)NGOPG?V&RX3pBpBHCItQ!~F(S;Rnbqui>$
zzyB4mY}vBdE1|A8hzMoxf{X#<s@AI_Q2d+Zo)3h+n7#BByCAxS)&9<nOf0&aVynZT
z6Gmi%H{r|{{;{I1NnNy#_644!now4=FsRU1bN5%bRfK>G$?g)^cq?e65Dtds?x;~*
zGTJjEC^G}$ExLOd#i9{$qC%`LrEgj0>RlHemmb~#E*c%x&C_jT7_3O%KC}et*rs)L
zB>Rg$vtIt@M<t`{XH9v#OUcBsQK_Va;Ay^<bGBK!FJ`QG=Axp%WnGCmlHPES$NatT
z6TnEL!SlB3J<rO2%)rswCsccaX;v>7Nb9O_yp4PcmgyTV5#%g(wSCeIIBSjotro`#
zXO=#BThBoR4NP!pIH7-5;3zV!LD^29;w$(rENSQd1{z@-<WX(68GYKdg!@Z=QN|L0
zO)P~jmukPob^m}2Q8L`})fa*L8YYHv<Sb<!#%8}l4i#t<Mb)F&R>^NcBO9wqLluyH
z!wj<jHP%KZn8IbREK&Z%Q@LRvQmF{wB!@t&MOO&MVjX?xJvcFDJ*3Z~AtK5y?ugav
zu0g_koP&Z&8YC%W*@9>J#Qi8C)Cy}_y%m0WYRu{t)p=tfbYWz2a*_?hgk)D?#ZZzQ
z%0PP~#!1FZ4%N4*=Yv}vVKaLN+uhjI_Zl46Shpf|kDe@jA#k)8%BtTgpCH4dfL6O>
zIaq)8!-Ec!==b$MzCAT*{PRDHDPzl}hLwUPj;;oBT)Xpqe;@<5%^LiqLP|~^Brhg$
zn(KJlppP5ZY2jLAw>_&IKE`$jK&5OXRPiNmka+|<l`<c8>BiN%X>sxO6z>lj!|gc`
zI{4%Kxwu!waiW3zci)bZLtyM7qA8<IbUZ<<Qat1Gh8xT(ud1qw_h4rhSwC#N{E7eM
zNuWq;gT>f6{2C88r@ybGjBAL>Qax30yZn!D#lK300eN76{3MjM!)S~qb@68~fPnMN
zYoxPJot1m~h?Y_G4{lsbtY}4WCWfm@Get!Ub}V0k;0(SG;qXTw%r({j*oLOJS-r4Y
zeB{(12@hd<D!kK|j=Yp5zk@NMX<OKaXcDO84RhyzivIDqF5u(t(jI_%+FPC2l&J#)
z1FcHB3*5*?L>?12KK860g9g_%)=0{@=YFR8T(<Z@mSbrg%YdZk*tuLJE_DFs^whI(
zS=UOGE=ntbM%7yb*qR2a?PU$zF-n4G3SOk%*YV@vC>Es?QZYI`ho-;TaZBI}YizVo
zM}H8&9v_siF|?iVmYAe#8{nLzMgU@G1c{oIkk6Y%KD5OUZHWFf<WxDyO<Yj`yce=~
z2exS7RT0*DG+-j00*a7<2}{8Y6o?ML7nl$xD6JxvPM6X+HjM`mR?@*qyq$cK-_d#z
zr|_qHzd$2f>YWb2!pSD@OKA>LyGK-Mk_{fKHvTu2=F0(rryS&cCd}T;!`A=^52vFs
zWMFg+J3Ifw-}S$J5k5$TCp5fuX6|oIp}{|XnkjRLtdkrXdUSQC6DuZ5n46o{<1kS}
zaQsTyoxJGL(=eGfu}2AM*JQz`M?7>G`b*)dv}XSIO81qR+Sv`7AI)c~nXp%hM!p^9
z{1x^d>x_pVT7`DHvEB1_!+I4$zDD=-vH9@acb*ti^rvd#vN^I?mU%UD;MJe351lW2
z0Y9a%m4X|S1<1Fptal^30^%xYLAbUS_tF~1<E|<jKD1Pk@q54sB<ivKDx^30UVJq!
zI6Q6Na82K0eHUF}_8|L;Ula!lX_Q|st_pAtqeyIONt5)kI}NI9@YFzkV_I#p8BfQ+
zmy>u12m{8mDmMb|=jD6$5Uo2$p4Fy$!uoeoPi&AkBT>!i=Whk=cxK|@z&Byt@Fp%A
zji5w+!kE&5frxAXGgtBG={YcUaP6_%^pOX_Db(P(kl-_YIj)o77$froFPqc?P&<m&
z<0-0K+L3b@t;*XC#8N;RxI8&yjv|(=jk8<n26dq1WJ>Uo7@95Imd#^&u2Ltj_~#WV
z^w`d0hRCkOdB&UO8wa+AS0lp-{!d7!v;;1cv_o}zUa!*~R-9;s0e8hR{rdr&By3Sw
z)J7(~Mj0OIBLUdQCjx<HXj^R#ztPaISFAUoq(r;b==egyPXGaXqA6dA$p?0MB^*A*
zfgDIaU@uDAIT9X<Ajy;tHbY*I%1*8<y`qDa<NI0v*SRN1C;i5X4_DB%EVHad+1AQm
zLjQ&o_b=STxn=|=9@sAQ8&-dcW)JyDe&mVT>hnTf=DjFi<4e!9DLsqKM>W62hsEG{
zPpN9WFY)_qI+q{F5}Ud~+PzMV3|RX6W{yp2j5{!sJ!Els@?_^JvNYB6BFWd?x%0dQ
z*`inaO{p-gP?3%TFhNqOey9kD6MHlM4R9kL>)$cbNcuN#R!QA^BQ_THTi9J%OI!Hu
zbNnX*7lco3`&yhv^PD&lH`q-}#I&7|1LGf`P>mjc4vbBo;~CF@t1)GMq)=M0N0zdJ
zaElj!3Rf;n$Cs-n0pQ}#Uv-?#TsR^Gm7W(K#p*mMP4FbpB1*t)BcG9W(mP8y?bvHi
z5CUKnfSOl2RpYu_GK-qh!wbz+zsDi4Rrpi}Ms&ubB$RbPFn_B_JQdMW{AvEr$0xeP
z4K)nz!z{?#786_6rAg`-jM9azUK^CP_$KA<WG0^k1PEYg?+LTB{`&jtk8se%djE#g
zme>=t6iT4rYX@IirLAADm5pH6Vs{LRAZV)Ta7D~;{oWL{it%exg1fu}K6X#%wPFX8
z&)IM)tE7qs6R<TK-2o{8&YI~)!J;~c{m$;(vmzq~&-zf$E?enbG5k1Y;3x6QnQ6lq
zuZDLAP1fEiC@k{H(o}gn7E=`>8NZ~2Imgp+;5U-GuLke$H49W`c`A(lk_Tn7HYU*>
z1f+Ry&Fu^&tAM;I9#)*3W>0VLBgwLu`!AebyvGN>Cypy-VW=Mi_X=$k)Anh=EsSP<
zFzw+v`x#Ebs@Y|O5})>0)DStm?Udyk-_)Xp&P48^1hI_?3!$tBF$XzWs=e9$hi33y
z%LEyT4ysQu=$MzB95VtiQM89?xc`6y!3`Ju`~r`O;vF3FdnIIJRS|=$#p&Ay%w~oD
z=M;Dd!9(VBMSoxgP?Tv$;N{Y>oE$Fdj4n(XME;Y|2FvSc8~vMy%wmjO6(yw`vimb9
zLn>nfVZRUSO4)a&+J(}kc<t$mHvS`wwbKHm^zV6h%z>ECl>ov~(Bl1^I)RMMdcXA;
zpg2^z{N{-ocoM1;biB=;zef;&tmlh0FSI4LOFT~~=$6UYt%NNjl3|kFTOh^zwl8I)
z5MCLedLIj<vJMfb;z2Y<!A*bveq*pt(Fyx)(9p*$+3;@6r|}H4sFE@Jo01fKqobXF
z`S07>6Re4myiB;T=`{eKiC_JuM$Dr4;SnPczYb-n9qd2f)Q**D;8Eba2=GHT*PUGK
zd6x*w;y<Sfd4KV5Q#(t*2D!(SF{A%TCFMaB)uS+c*v8Ld#p!1~hw2Di^hARz-&Znd
z%I37*6V~nU0v7Tp8cSF?Gr;|3Hj7^mO!|4rHdU@alPqy8@hhQxR-a<MFM0cbPpRCQ
z2*(CfUYHg{$d|X=`}&`Js}l#|H);-8fHQ<t0Qv|nig9t6O?0}5K=pHNW^VFR??aVn
zZC#xc9_iP&$_5`uLnN<xdo!nCkKww!-GYpdzw&;hZl*k5vyR%fj(ETYAS+1l)1yi_
z(<zm|86U&E=c_08?QqDNjL>7`+gC-qw3kwW17R4OhNAdzu0uM)*q6|XVmSm?)Dt}!
z5spny7bp-|6OvD@_Q}Kfnl;dUPM5=BFCPR4Uu<-i^lU#pd%$&)xxrU|Qwa-9#Y<XL
z>QBt!=#m{KUI%N+;}N;~ITZFCS+Mb2?{>w|<pimQu@2Av&*}z`LM5@Z{I)8!qC<d}
z!FQ$8oqu^<-hy}vw)~Bulr7skVy^SIsP3kJqmjL>)J~44Y9>~Hmb9;CP9E>PNV(+E
z&y6F6_VR;$)#_lpCpp5=Zb0Z=2ux0E{?Sz@W3{vr%_&9MtQoCuznoXB8cWLsf1Qs2
zws*xR=1~nKG-uXR{BLjdv&0=4bzj3aKECWsEKG~=Q4Y|V$O}DV$gAkATU)!EGV{eS
z&*S}jcxE5_M8nU59k=ViNTG}9>Rdq10VqxGJ||FK<)p-QB9iEwD&*`^+}byA`v?>8
zNlN5uIsf3~%WpHid(wZb(}=viYNVbJ0oT`LO8CTI$DkY?-*x2o{6ihlhKyV@=EC+g
z3bnQ{nNg=uwKMnpGiI&mX-4#AgFNI9{}q<q{%(c*06B2Z;f*6T1fh~8+p8H+6CcUb
zW(;_r91EYDnaf*WCX|l}{_5(Ci1sSA2jCU0W@ZQ9c&`$7{m~;}3zpCk2unSR!ZJ2o
z^Nc+*htrsuw)_Oi!{?PVrEy)zW6{nGQv=oPHs@!1)T#5iE#jM>k+moT0gw@j9G@|`
zoPO+WRic!`SGfcpdv*PTK3=--$vf<+S2a>u&1_Q>2aOe%=Kw$X;Dl}=tEKZSeu@+J
zBCy(B7q(<rY*+fi&xmm@q?6A3QT#&ASb&K(#HCqOGxIrVu-dl?d6#(R0vtMG=3Ci;
z_3Fjr#(cgP-yH@%ywZt$IXXM{Lvunh*Wn%qsIPQ#-slf!-28k+gE&N0(27guj>Ti=
zFw!_Qjx94h3@gKZ$5&<aakrL!e~H*acj*m8gea_y-UjWm7~1Np<{UyeU8m+mU;^Vm
zs_XHzcOjE4o4KCsqAKv3#K=|`yE6|2nW)bmh%J^u6lnG8E-jq5Hez$M3G<kfk|(AM
z`GvqUK!1oKfR<G{2k?{u;<beXFT@0h&LwGvnoPvvb8CMXMX@yrd~s9+gkO(I_3n=U
z)64#?qX(%B>P!J#BVAE+o#mN<1a-8aVk3t0kM+umdfPo^KDwW+&-H9N)#ywxy`|x&
zk9)?_3j*vwo~f#%dvDpU7SM9;0i8^9^q6B~P}=t2Cdg3B5I-&l=i)tI*Vo~f<f57P
zdkXo!zR6=cYwP6pg=X)Kxz{bXKPn|m<n*rF>>=U1WU9OMF)oG2-g8Mlhf67rthaPN
zHa3Qz9T;3~B2AxKx&(e5v-9`>_wZw1E>ZjC>n>9e>n?fNOt%<?JmOr^rI2VEt6ILl
z;{N=Y9K_Ynovqwf2^x)em$Q<pXG+D%>mjlT_h~~>)tA47V)&b;l(3Tj{+73VJR}WB
zAHC#XKZ+Ad6wp{)3%@>;hDk%|;O7nvul))&B}2@>h#y3esb)@j$~^krmVDBSxDjCg
z)TthWWeJsMZ;TH<V24EuSSjocbv(SvL!F>PEy^E(yiIW<WTEpFzSi7fUgHiUe~z|l
zhPY3UdGba2Ug*Mf^qMzx{$jo9qu(;9-MHm0!;&Ut*KzPz$~|uB7P_5}V@dad^M6zP
zTyA5*)d}O+U7C5B>za)+08@Alu5HN0T^F)^BiQ8sww17!OL;`LBO?#=y4RI%eCsxr
z1}6|GH-8EiK0B@*Jj0XC`5-LBYD%u}nu>abxbbZxy+5JD2rz!21;Rvf!G`9>MaoR3
z#Q6?$&TbAOZ!-}ST`7zNHP*HS-jD^Z4wK6iUzTRYO)3F^X5Hj6z{!u`5#Pq+$r=<W
zJg|vonOi3FNyu*WXvuN870<Ygr$l_h{x*P;KD0Hud;-CYcG89uQLs>`y;+z6>emz@
zK6hFz6Ip!dKZhirNtW-fc80iv=Imwak~C812@UmsN18q&UR3{GFJSrT5FkvcFq!|N
zx)N6#Ff_H3)Vrq(3%JM+5JCvaH4v_kGf^XAFS*Vts9u(P_U5lIzCe(gA4Y9|zejPB
z@0j<Cz5S#TIeY2@69yI22m2c2xC*lD`}_YhMPq*Ai$h6EnTDje=MpDs2nH5{7=pU(
zRWWT6QXG)A`8tb+<)p<k)j^9(c^;E+`VYO00ASlZ=`-Q0OTwsYd%9ngZy$1L%&L*H
zdqv5AT)~9aW-A}vMYliapDgqjOs|1snOD1#i7C%UyG@;QSjtRp5WU|+h5X`VIQ0VU
zyQ_rjsf7G|0(?MGqCv@yhw=2!5f?npWHnKwp~iJ!eig(D0X&$TH5wNbUM%p?J$IGd
zCgwwn)K=BI)UwW7S7lX#E$`W{4O0`x;N^pARVHQElcXzpYkj1Jcwle-uP!&p&qw5f
zGwj+IpGjRx@ob!t%I3^03w+V*-IPS$1dj6Dt#s@6+y7bTeG=Zu)2c~FX{{V4Wu@9f
z)>QSxPB{_&`_G8pbWGRy@{5XJ55-7Z^su?nZnBas5<3}H$b3kPE2dDIj0$Zj=OGFk
zFEg0iqlQp2tm(IQMu6+rp4wVnRH$+iXjRA;6Dq2BOWT{19HisAY=P2<?Ydd1r`nHO
zCe_L|jbtX$sf6Sh#WzK{keDB+&{@9U=wMGf!>wzmZay9~Z1;-o^MKUBBQ2P=ot>Q-
zPomhj{&`M&!6<Vof-SK;Xh9#Lg?eum8~>OQ{o>>@FGJZ*cxa6leSflROcj<%?RxBU
zVEt6|y2D)Vp^<^11C5#gK$I-PYRkkEgEVV`P>42GVn?gIhPjDEwfSojqJ2Q}d1?4;
z`p94u_C$@bSW79`5>|kFXNsweBo>_603Zm4Uw_b#Q?*8RJ?WFQEyoxCxxt#g2Z$M#
zy~UADY}jIc7o7qM$B8V*b@KnMv)%ak4vM)^T-$7ALBBoAXmW1vxsT@_Q$wISIm!Cx
z>bnnTb%pg0$(?HmK`?IA`ycddeBN?gONP+a5*E=ba4vw_kc!e@+d23_yfTKqajhIt
z<MMao<v)%^nC1rPL0Xx&sLoU0x6}J<Z@=)t%N%DZrLYRAUhJa!!NS=r)swU+DZfS_
zK!a!aRGdBuCr1YqczO1DsFLFS7kxI`shUzDWUG?t2eG=ZvXTyNARZ(nCC|$j6YP5c
z;(-h|L>rW>G1%+LCq8n-#<LmSpIJ!EJ-xgO^i)oRy(;|l@b12pL+_%r(PO<E&45>E
znsPtA1Kby8-}`+zcOygc*Q=n8@5}8I>-{`|O6ZRuKJOw@5zz_C-6v0|*FMZGf0>22
zDQw5x*BL?dw=Uk7I{FvEIaJtX_`AJO&Z!5zr~1Q2{pY>NMU?izv-*PyKy$c6V#^f$
zp&=!T@DjLS_mn;yLexC<7buF8(365umXxzi`}gK1zMycvG)2(KYbRwTVD_gZ_z@>C
zO`b>CVUjgHDDlUH&YRtRrd+ey@hDSwEM;XNKg&VUQt=r78~!my(%ieesx0=wy34o?
z<se7<k9>Rppp_hLNghm4!z^w6<w#BQp_%{I;*Yy%te(8cb4@f!p$O|ckI5f1UNrli
z`jjMX9g=(Bvm6P1#&7{O_Wh?ykJ3@3fp2zrYdvR^XHoq$fhJj_049KCR=lRd-u3~J
zw59EKf>64(LOcf1T~<ajlx(zZ2m!IAFEs)+J)@STcRd2&?2bx4034DLYhs?Dyuwi$
ztmILIW5;-99w%qk{w*jP^gW>Z^BQgc_gf(=76;<}%C*p9h!Pef%oZnixw_9b(bpUg
z58bPtn$1nx$Xo3Ls9_8i!bhnwue8F={kQOT(g6>GH^CZSJuQ;H{w(!^G1ciJ<543~
z&LTnNiA|^G<FDq4wZ<RVpVoK0Vst6R@~c;uT|yA<uYaBz*Cn}K+G|^Nfm8b`)a*_v
z$5IIo+pU<Bhukha`NvI%VSh{DCU*cCG57Urku#7mP;@3FWOAf=)-?q^qZpfOw0yEZ
zEH5CU+Bj!}r?<B1<IzPQZ6S#RGvDq6ov;kyt(v=kDH}{#in4S@*f3=)jaKcirpLHk
z{9=@#GYv+K^~)eCt?!*)&Yxv5S?y{jsrrDfzDsS_D<_NWpJw%?P)8OP!aoG$MB2Z3
z-ALZWzYIk&!;??#`TP(ZiCaFC^IX*WGqZC8f8wu4WLNmr^=*BYkkDF8!qtIjxk^>!
z>0Tq#%O;I<&dnA3ZtbljjrEKjSDrd+K^5rG1o}mhz|#Xai_um=LR&$#_Z%E+XzT9V
z0{h+WbSQ2`jobGNQtu}nq#+!P4Oxa?cz}(GdqWmnErzf6+0nu1I}&P%!E!!k(x>z1
zIl1Rf9VMf}A`+P_NN>G*touA63Q2jZ#^t%GYgH?EVwNB3x5jeGMtbC_sAGzVtGGLm
znErqKihh$kjaQtV{HTLT_QMmS`w|EJ$U|I$GeX&N4tz<5dZ+mCj8`wcT<liAX(XUc
zV~%}?4GMOCM$d*cNd5xMrv6%X-h-_3cfGHoXVraozJbh$T<!AW+I17`FWRUP|L0Je
zcgW_r{KhY%dji=s<zI>6yhlu}qH9wJv=7-$Z4pn5;Np?22%c09*Hl%4ewOsFYt5kA
zwMCTl)%2*9upBnpCpXwVCPcS*;n~#qx@EzA4v53Dxh9P`y_u4qDL0lAN&Sd`zt!0&
zt+kIrv0PosqMpzPwN#Ef2Ok4moBG+&#V?6_b|-Br?GcaX+54WI-tk$(`K2-&ArLpp
zqNZq7(f*gtg;RAKhIm`iN1y35xhA*J=yr)^a0V`2W;v?7`x~-1l4yV80Yn7hsNsi?
z-UJeae-%cW4n|G{To)2pO{M7G2-I{_6-Jtm2|B?_6$)4r-y0u!BLoQV#Iot7V@y*o
zIPeLN0r|<3ywr)C6rn`=vXb;oS)x#>r5s;?Nj1NpAy48dF0<vfgz)oh0ejQ%A462P
z!Zi44G>Nd8WTM_LmJTlmnLL+$NTYEy@&hUTwhOtPB^B6l*?A+WHhGgfu5qK}4#Vs9
z3_Lb+3Tm<deuKU=@&4umjnSF2+$B9%Wd=~t@eT6Z`ZuXc_vo#Ti*zE70dMY60Cio3
zWb7;ZDO#DvdkErka|`BqEQB^1lysM95NFSPi5kxT{j`a>3CcOL4=zA{hvM8HO?fNA
zdPVN!D3<6#7D_&C)x$mM7@BRi`=rlP-NbQ@EG80k8dO@<(6rQYL38g@QetE9!2nth
zncXuPkJn$|eP*N4Fs8~J9+BsgvHv@O3TCx;3;&^cN8gps<7g}DyJceE!J+bChw-tO
z9a`E>9t3|(ECQKX6jmosn|SrvMYY-PT3&}uc_P2OFo}G4;XT7!6vsAbEktCJ{i2Pl
zsP(HAe*flux5T?(W~SeU&z(g-8_Z|Niil(ufqonFeN9wjrOGf*U>(|n4}-S%SXH->
z-!S~Pv7{w%##<m}GG5bMlTL<3fUjv>0Hl@A2d7l4D=?~<1N;Qollpy=KU4@qqXqsA
zJ$88>JE4sp3|V-iy0(qWpqfS_v2mX-DCF5q1($+MgyxRQAXplCgPo_r(?2uaOnX<h
ze$Hfk!jqiY^2eH$iu4*f0d7HcY%Ac<jYUbC^RM4$-wf=}_d}1{F7VqqzMgg~CY+dr
z=<NF0Hc2S`#6(hMi1=(#3OXBc+1r<H<v2t^FIL}o3IF=-m;SKbBW^<WyLFV<`&Hj@
zYj$6bz8>VGH~kzRXUuKi1{s!N90=?baM&4TF!y-(Rq}RggGKt(>7U#O$9WHGTX>io
zI20zV5fDUzOqbjM!iynX9^7iG5zjbf{txU&QY|k1+BWfS!IecOR+lOD&Fmx{42f<D
z@z)(gL|0@yAWOEGj~$dAXOu7{fMOo3PC}>YoDej9v+Idj`D*<ux&Lc$6J@v8l9SlI
zZ5baHh$qEnukanuqOsXy@>}M$%eXI!<;Itt+Ew%KYJY`wsp|%PTSn4<Ik^9LQnWzb
zNh}8`_&iNmn&ae$0q2P1BcA#&Y7m4yRpBGuPHsT^x(TF@u@Fc>WEUUYR~4(~8)(z{
zb;f_!l(VRD(T=rFqZFWzpJUMSUdH;2Yvk?1SEsP8;NuC6CTNC-t*`5)j2Rrn8p)*t
z!0x|t7OP$Cjqr5Qs6BjEO8^RHAYxQ$Y<`e(PM`&|5{haFevCHH|6#r|{sPF>l~0Et
z8SC<;%^3Eb&dnlOEUbrbGkAc8An+AD<kxtz8gKG}6LRSMbmotp-q>x`eo?rZEqSE_
zV7V3bL1lsU(CLg<y6!>BQpO{5qL}^Gw>E|*nE%oysn)Fa^OAQA@+u1cEc@$yCkYx(
zv=|D$WdtC7h%)zs0>X8xC=UcSwaKKTcZ+p4^DQ+Y!ZNO7l5{2b3+n+vO}N2dszcy!
z_fEk_Bum^&3*@HJ*n^vzCjqQ6pLoEC`1{4HqLa4w<c8F&3zgFQZy#$teGY!FLk<cH
zMKL_FV_T)b<tu*dBd%y1S<=3-Cbv2`Uw2>H$a7_;UUgM)hNU6<3n+7OKnTGQQ;NCH
z8BofS2kkVdj!t)Gx~2AIk}|0FzI56Xruh`1QZ!Y0y7Pd|)3^85z@mIx^;;WvOe|pv
zdZQuPZ1#cqtG^5O*JvpsQitt;!gAXi16ODeQ4A{{B)tI+=Gr>K9R?uB^|8lN&;s;F
zO$HCkE#2<T7s=h>Q&8JVqoCK|GjfJ!)A-1hO%}MYIi5pWu*1|$!&7|F6g$&98(+#Q
zg1n^kLCVYE|6ZnqD$Oq=CsLGW7Yg2&kwcSs_j$=Y0Fnz>9y@@XuhE4{kKKpwM0|hu
zs6I(r4)W4&sh{X1_?54V%RZ|+$N}v<6XNT3Eb5GS<_8D1!)+-CM#!O1T@n4pLKL^5
z4oltogjvGUQL3@xsmL24FvXkNkGMy=yF*;&p5a07P(fisXDWv(@gV=isTj61>By2@
zJ>rK}|I=7ZfJP%rfxP{1H@AizOW<3$_1y^{Ac}P2Xu&k5M-?G?Y<CgJ)=U87u1(}!
z7jDe-E)14e%u9V9hSG)SPG-Dv%VsYcqrBD1Vvqe+eO&^1C2>7EiS1Ki;Z7tQ1rDG8
zSZ$pP-()!aLHc@dfrv;ySNJiFlWLaj-cy4Y$X=ceh%!M!-<qWc4|TiE%hv-mY%cK!
zB*>wEnvDe$^GlF6V|~la&JBU2-@>HrOIkvYn+lNz_d^^sDGp!HtaNo1OE`R?EVhfn
zg?pt=ziP70x)MWVgm7L}uA{`q27VXt?JM6)Hhx%MiG!a*m5ONmYxrcH?SOH;8Gmcl
z-;Jf0soDJkNcfpO&-AI@we?LsFtaFFOzTVP54;$A<WN$_&vuq44I1QYh6r6QCMTYy
zeX48Qyy^V_B|C~6e2q}W*{5m-ztz3W@a1j4bY<GyQ#q@}HaJ~BIzja^83+izz;6K{
zM3fO3Q5cWSciOnl8vIl{8cunO(v&Tep5u+EV4(jZpU)Zj+b76_=hm)&^q~9<`}WYY
z1*lH>%YEIXgtipe5CU|gq(=VF_?weE9yV#CFS#fY1sdKE1OTZS_Qo&hSMS^948<HY
z)a_iVG)s8BWP3Hu7vMds-x~2oe{a;*UtC2FIc)|M1-#cLVz#gh{DX$`!>pJsPwU~W
zHYu!r`hvEFXYzT9aX!O!21H0ybFnAS^(HSplOlP?h->cXX=A9D_BLV`fjoyUACxg9
z>e06?R3U9Eh~S9vrIec5Yxh@hUaWDw%>lD2EH66hMK&ruMxKd6Z763Uv&x`>g;>_W
z=p7b>hA(`OZc3BCe%#w&YhrDhb)L_iH|K`6i0RfaO9mT;QF}h|L@G^%pnU?f=`=?f
zD{8`cqK3-VZ?&B#ga<k&nGSL>1Su96eoVCvk{ztH(I5rpV4tO%4&&K0$l9X+GM+)M
zG4Ms8Wo~wk<-fu^#9pll8Z6*Dmc<N%X@K|&d7#WnA|9+PaZ;sAS5%u2RP@Go_{{k7
zp+XLPQm&zIhC#{xqs;f;lw2D%&;whlYYw1)#H{AsJahT@q1+3`2r_KGqo&#1h*jl>
zq~R}^&_3`<rB)@OghxZmsbt4vQ*4=>vtR$+0A%%3AO;Esno14N@o5%GH74;4Z%@8;
zM!q9R1p+`I`~gBtibkc?L0&ZQ#*r=pkD>;SffPi+%urorF_fq_`*|o1Ir5lBobo?#
z`)IclM%i(UB^;CsALvMh4~wp?_T5WxIHRFiZTpHkCaZ+oKh@s><Q5<lsh0sOu+TM?
z$8wP%<UJ&FdJ4{Te#bI2xTDl6Lykms%jT~2e!EZvdi>Y`-!BUPD9ex>Ed{)XrAi)D
zB4GZ7vFYjNB9IYC5-*?7Omi^W3Zr0@h+CMnfY?Vbnrv3}Pc>&hFY$jn5gjL)@I$U7
zl)~+w9JULBHavKptP)U?a@Irhj2wzIygIH7jJP@aGuk`t((1R3j!3YFYY3@$KGVOG
z7P0A?vF9<7`D~PS_Cs7;{7eaiv^fsMI5Y`Xv^~e2@f=dnhD9rC1f23BXO+p3s%mjg
zf3GH7&x`(7j*=H!qq7^UFlGtB9onC(ZcHA05jc}XNU-%Sr|b|@YdUh;{>_fBj)KE|
z0xeEI;0_<JxCjZc&huc1Q5=XKPNIJTw<$@P+W)F=;BNifb;*GL#}E6S@8ef&RKN+G
zxmrUid;X8Z9Z1sc6BtXK&FvYad%TxIi2w;VW~DRB<TM_Ihj@(i=+(&aZKF@~P45-I
z^%#liC$8u$3fR*)6T0L?uHDi?9_V|up1$n_bO|4;NBR~cE}}s378biY)R^7-Jq)i1
z&j?%xwD;Go^Ncr>-FeoR7U%#Gd6NmK&yNF6alX6L=Hr1V#@)?_nlt{DU+N-PsU;m=
zY|89N%UvA~GZT1aqVRtqST8tJ)h-wifJul+C~oFu#_a|@Aejmcq|w$~D%&l@p2mnM
z>urV}m$4ryK_mIh-Hs}*mA{P<545gd2|OMZ(~j?xkRy<BTkk1I?YS^S#=~{f@y0h_
z%v(TbW8Vh(qQAFQ8G{eR=-~tG#9z&tUB-gj%~hXG8A6`Mju-Zp<qHd#Diz%AxpJ{<
z)B34cno!MPfIw#P(@JDh1H{{<A3N@<NQv3KzHbGKzLs3_T#z6nre!zKa(4aF)&W->
z*JWlQ8QQ}ju&Ak}C6y6+(|;)3P>EpiFP`ROe=+Ux-a`$rLOOklf6LRC<)zLGQtoYU
ze6y-veh@K3K=Sq{?by;ai-LUIws05=3z5lcWJH8A()cp#bTUe>#XQ1t{!#_r&zNZ^
zh;Ep4tj&f)lT27tFyHpp`%XnVNEP(`IZRn)dv9c(<?J3p3E4+*stz;^$tcimPDBo{
zq^$P*Faf-+I075cM)*2-H>i|o^cMEL#p*tOdG!mHw75q5femvnEZ%x-!lh;vK_EeH
z%2xDV6OMRAkHf$MDU6whfP^loDeyDe!}Wys9<hvlj}1MBAES-*^$xJ50t#BQlyz$K
z#QUiA;<=*~ZzSLzsI7(zH5gWNTRF3$X$}8MTg$`p9kgT39z^A(8O5a1)};rQ#@nko
z>7PBhe%zx_ac`6&7;cYc@b|c38N<a7%jX1TOWL#_Jd|o%XYz#enUFM)9O2?v^?r<X
z8b@3*$UkT}aMFe;oB7lp-cG;Bg70iks`rCjB}$G{?w~6lk|5yqVi+sj@IrII$)|2W
z9h3ELY(C+JCOyGccBB$p^xLur4HD;(sCf1elRx<Q3i3iy+cx1-J7%msKRgb<d6Fek
zXi7cvLQTzF7{Rep=i}G)dv4f*8nNrKiPi82NcF@~KWIISSo)6yaLC(?H*=Z~6!%h<
z44;tSagx*3^?kq(ll#G;KjCnVz&n1+Hw?s|O?PIN*50|fl1-_hd9!B#x(Wi?q)(hr
zrR;S+^Oia~<h{nGjPJ1xozrA+@VD$VWyC+`$E8qPP;C&Fl+4LvnNMA+8&n(j1`0wX
zj%F65jsOU(3f__=^ibXV!?6kI)Z)IzW(rgk2{Ws0Lbx}?l>2-snZMbfD5W{~nDBs%
z#H;O5O~Y4KV{x|WgiSVl^W*Yd90m0bx#0Q#-AqZUgT(V{N6g4i>}`<TR0VQZcBUYU
z!M%goC78~e;^trla+2H<q8k>ZFOg8(Xv)vZwe@+S${6u;+$Bz*zldD|`>kxQcT`(f
zod^W*SkX3?jcsjXgQPw!*bfJSgd4TV#S3iu@`Z)NX`)%72MS)IbSYXdcRf$<Rl*)>
zh-K2_qvDYA|GBJO3<D4FD5is+`l?NM(==11aGVTRvjGyKAQ}_lJ@M<=fI^PD1w0*Q
zH8}F&c<l*vDdII&yYZ^>;?d`27M<S=3E(0Y7@}*6;nZ>G_2^^-J?If2M>5s8Q%{oN
zTF-*gdQl9F!0oYP_}ioa$4;4bh-GecF?x!$zJG_CgqBG+9CS3GDePO)uk1>h15<!l
zhXOCeNis`a`p?#CqC7@$cvY{q<_ps;XqZtQ&$h>A6Pbvs7%td`Dp1E*b#Fh2LZ*Bn
zJ}YIto+&&QDV4dN=Hck<C#G^A_NHC#*rPMNH@|noYSJ-<_L}prW@Bt)=K9wRE^!fH
zQm{i|)Wvff=!m6LEJTudK+?s-RQn4$1h}3SjUSSEheWvPq!8Rk2!22W3wSZ~YVeuS
zKL+FxTP%`4N6D3*ANDPdnX%nFAmVblv~E?P2r%%ljW5uJ3y<T$MEP6&H%iM6xdPyF
zP{dJVZmqTsU5sqK8_|-S9O*}X(18&}Wc}^X>m5}l&*!`M-0RCJ4-3e**fA)UyEoOq
z1s1oK$%8Tudf1y;0N&&NHQ;DrKh8|;w;iXz0p?dg|Du2077gTa9{}a0jwHJkvQT@N
zG+-7luVsYnEpTSKw#%D)NU6Jxvp{SoPHWj+rV2@f2}2er@17eYGL$|(qKfMqyrb9w
zB9iZBK_ezCvRa5nMT$C>O<D}P`xS-TKDTC)u;qR-uJjSVAt)Z&+Ur$v4{ISy0lTF4
z1SpTkz+Gm1+;&r^XCfM>W)07f@>&LF$;Vd@pfyk?y18dsR7TKPb_dW}s=1JlN5dv=
zKu6=<zk+ZriJ1$J%uLKE90Ie8A3mx6XguM~+qG5!0%b|V>c|Lx><MnDr^CX{8pSkG
zpyCv<DY``i<vbY<rT5dszp`qah-16!-ZC@6yzR)3b{r9YNNS_Ir)=*tGJwtzI9fh@
zr8FDd20eoibkvP@O;4arob{AM#5@@1_2?aLV}rgasW|oLOE(~5<_1?7B;8;W$x7KK
zaC>e<OcYjlxCR9je?1*WOX2z8g6{n>L74dP=;$e`A^06!AzMWcx;Y*XZB5U2I6`mi
zM`Cd7zz15k+c0r-W~NUjN>RyD=a6XRUp`bPgj8Cg-J<2+1Eqrx@@B{;xJi?i&7IXQ
z`QB(%LmGsE+-)dZ=uxn=aR8}l1$6Rtf+&LS?rb%t!xi!Z{%(UBHk<r<#_aVN^ZC#K
z=ynt{5o|g=1qk+;|BLxFw6q$Xr`DdfuwdW*w=SMp;<I0QEC23kmg;+tRxF7^(9m;;
z6g64d<Bji%bb8e`o{5mwugIRG5cc=aX8!|<hMc{E5V6Erx>(`Nl-}CQWC@rAr$GHe
zG!WMwtGt(``PO$}E%mwK)5qQ)6r*+Bdxd<|d?}A`38|~nW$=EU`y5u0vyt8mvGW2j
z2XRV?i23P#jD(U#*~Xql5(%&bf#@7|lLw@F{s`Oo41&`e5zH86QF8!#LG!sU!5Uy_
zsod#IuZj~za@-KRJE%AnAm3WB*vfwH>|9zeF>6%=4Fq!>3<?eRQ3tJlK+nXZK&3B&
zm}iZ$a07i@6F0@A+N8K!b=p#3hsv3s5}f>W0!PXlc9;_S(RP!5!6>{!@03Ya(`|f2
z6MBYEN3o1yBwy~z>yOlt%stQvS!ZcWgCAKc8V7yvIZ9FFU;sPRew*#~nJ*x9r-p*W
z?*pkp^RfVDTg5*Fe#F%Zk*P@}IrTgD<K1WerL{Wk5nCOeuGD5*|B*Z4O0-o0GN{UF
z!rirWL8?oe@6juTKBo?qLCoAaGvQm_rRno*QMy4sE6l@%W7X9<eTmB)4B+n3hKLNV
z$yAlZtO?h4;e(V9EEr$W7F{#~S>I(bu2NCjR(!>Yl&Y+nCqQ8hbUci^Kt<DU4rFF|
z=d(>W!htg(N_`T$NcA3$2LkPA1_y)8?gAxOOg-p?f`0VjE3gB^y04_vR?OlLaJo0#
z_s(4d9WQ%*#1@n;?zB#u?wNgr&}G?yQw8USUF>{KF_yFj+8;@c9GV%JtV~BiTc%Sz
zFmSMD%zDvK+)<o{;)E|FY<{d48vHvnsZJ_<D3N!e98_5O;vy(nYKXH{X}jy_sq>Wz
zxg5>YaW8Hcrvhk<-+mOBZyG?Jn0@<kK!b`Cyi!haTv#Nt`cL|ghZjod*8HI9#g|{`
zv@x$FnA`-zb49kMcGleo<2%*UvOH-&MT$o>{CH90OipqS2LLxZNEv+gsw1`g{5E=`
zLwRlCJ7qlG_6rH&PYo;=Sk2Y*SLN2E&zT)ghoCl^=ouWb;VEx21A3)j()Jz%-{`#$
z5X^PEI=;6rprqTGF?LhEpez#H|Mf1vS;ovjXWHW;<*A`_ihUoYw4DO_o0-TU74qpC
zAw8cV;G-bN5Or{eZ@n8MNyFT&oG``vH94L6n)2w`xkBG7z2u{hqaULg=0tWLah^Df
zj(c?02=daal*T#iE>^bI#_vH{dgkd`LDaS|#!P;fsSuLJwAV_1rvGGWfa$0|r^Fm6
zakL(HSJ`PKZ7e6Z&K=v{4o2#}Jicxwmoi0Bjl>y&>y6UdQlQ!>4@i9B+`SSu*#viD
zjivg;J+kHmitMWWg+jQb$-?A)N+;u^HSL#w{tyTkDi@dTEZ3<w=C?m=Jv$BS;8h+|
z0eG`tH{qw>^_OiV%&D#mr1P*vIX!!JEa<(H{qT}Kaz<JkxjlXUic8$9(tA4{>Ql~q
zB?OoXvo~NvlDet?@g;QTg>|@f+~N81AZk@jroalicxEGh#i1hRZ{q3g9>meIP*mY}
z3-LqmJjg0X@m_0y^UH}VL2XT*mj$(sZPou90}1K&S=zGTruAia?}kxrm%lHJR0y5n
z;mjAkZ@A-!c((i9ur+L<Y4wGJLndX?79b<!PrxzKZdyXXbfrizJ)ltg_lY#Y4T7I3
z>8CXiA*!<QMd8oBDJzDXOQ>7&H$J}136z#^b50x4CwC4U?z5pbXDWr4@PD?6KDV8h
z1{W(9gDU?4b+h)8_~PFPIESUhdrDXj9t6{!-dw5Q<mC&1RGx!Z9mS{Jbc>}HgJSip
z_-%T9)3v8rKX_Bl@$ZhN+-ZI&U7Ae&zxT-aTW9CH%+;2{H3m$Z|F4lNkB0JX|5Ng{
zC$f~1eJhM*G8$PT%h(etO9<JCEFm+<p2FB-ERkesl&G<%<(uqf7a<u;V<|LKW0~Lm
zc>A8;dCz;^bKbw^JZGNkdhX}Gujjt6<@34J;lH)fS1EG}t+iT3Te`|^#Dt5pC#ApE
z=rcwdp^_^&U7+nN<q_uIdfeP{kL3jVo5;TP7DO4x5ut<nAwE-dfL`>cs-k&vKGbn^
z*|YAs53K=Tt|gxJi+-K;O^wJ&6V|JEM|)zxa7U(yVgmB~=oq47a!1w+E6}?R7&u1k
zFcK3u8yeKQ1WZNa{b-+-^qI@N+jEU(LklLPrJ}&TwDs)W_t;g>n^WCAQMF}y7bW3#
zXKCqsWJPv2eensU#HD(_=5FcUPgD=+{KqxmKZKe*29UZ;q8Vh@i@9m#VaBgMOAB}5
zZe;AdpGjaz0u<G#Z(0#8BW5j`(C1)j$~5G=U#`Wo>l1d<y_aF6rV(s+k&-rq?9&cJ
zT=rdC8b5es)(n6RrO{uLu^L}(BAx<nPnx9VN0MuCn-dz9jDRM~TjuY}#_Vtv7nT#M
z%x()pJ#Vy!LFuhJtvBuN2cL_Gsg1rsmk@mSw8@+XTh|u63@pv=3{?1rPsPJUlS}y1
zUzZ2ck``h(T{<oVAUWzlXO1w42<Y1I0`eCWKq=&weOkm&=0***eTx_eG(b-VsI;B)
zrxhY1#i4MIxXc|1@Xnp;Ts}$Oj`V9{oRA*JpS5#`e_o$ITHlOy^OjeBqE+-u7g4m7
z9*lS3NDehR)*T{G+JEu>O&EVw3ajN~(pXQQWWw^1ipo4uvi?GUq$e?tY$l1=2%HDx
zG88(#N=e9M$E5savu|b3vt!>ntJ;f??>l4Trc|&fFw#8ko0mU_63Z$@%xR9XWio87
zjCg@=u-=}a9=VEp%e>fJiE{c$xfz1dvEuSt!W_;Z%c2eQ9D;K5_?q0}V_)k;tVAD0
zo}M~!m3qPo*h<A5o1m-;l$Y-}b&^gNKB7dKp6b^C9fphtij7$Hmny4bJTQiwe;%`8
zNPj6-86z3$`zel1qa~^`kC7B-`PP4d+WhAZkmu35GO&rh@`xQG7i!ScGGgw$%p#+&
z=qj`KVvN}<lGc7KmIgvwQ|uME$d9+!ripCHS`!Tg!NS7AD8<^$#4A~p-kwJMAxj=A
zHb{}g1j74!-G+c|rns5Wp#1v4Sia-7L7B6*s0^PladkH<4eb{9@=ixm-+~My_va-c
zG!O4#W5)N6(4Qxu<3&nAOr9^JC3a@>Ees&6oO=*&^QRDskIkF6TFaD!SL{@?rd0=#
zXb%0k3lNGO+%*#{m>NMQq{%o140t&11kH{e%n|Mi5LT(BQcqGguj5MH5^4kbLixpD
zP*&ji0HB;6B^_(|IN6_$xMM^u#WWDPNi@jU1>SA7J4UY3B~_u*-JW603HZH6<BB%L
zsOPnn5%o@LJqr>sHfCLh!8(!~l_g%~XhU>FvI9&0RwE1AHQ-x;M*_Wi`}voX&zY!l
z=uEc#u2R7aCo356SG&R)i26vq--|D=NtRmHv`ZnI-m%JUT<{@VVt<2JGY`y%jUr?3
zN0-#Hh&r+cfZ~0hY&*^aax&Y1*x9qPb?cjMug&Ru2VCC;*CGU_K~E`=^QIO}Y&UOO
z*gp2ee41=KbK?*b_nb<l6E72l*U0i`YQ(1P49~MUWe=HQ4h?@4`o#<~L%aXgC~Y)T
z=cY-3zlo>t^1iK5-i(4$iTn0I$hL>o4*>(*$AS?U7P`WPvrFQ6;4F&J$UOvu%s4BL
z;lH}muE1P3;~QwqX>haRTXpNQ!ZKiIL$WRBJwho{J9k!0yEPSG#vqhqn^>r9ZMx8O
zvRZyZHBFpn9KEYOr*@;LOeTp>LM)i^9x8kPOI)g_%tgcdBk&qdjy^*3ztJ*xL{dR1
zNSY1S-uGQH`2H`hgkzi(uk#?p;|3TAFPyp9WzI|+wf6XhcB4x#CPOw?9cfyJyoX(D
zln}zGOA+KRF;iPYgXh-oq(gqGsj2oV*|U?2R(!{rzfXQnf&lQMZc9>OWIabm0svGh
zwG@rdHT|kC@!yQ9P`dfb`%#Dm+HXA8zrhSeZqUhfI<Po%OIBPKQxMC{7sdEqZ7{nE
zawhif<!R|4jh&kB1iCTMxj-cBvz$5LC31W*XWHLh>++HXEdfPhVpdocM*??ysZcxE
zV}SeakjQFZ8Ag%J@;0^g`%Ia4p#^9A_pXJ9s{L@!@wJlaQ>|flDBHz7KHGn-fdqPc
zQDC0a>DedG!e_^Wp7SE9?dLN@Z$yGTg($o8jqOFjl4Ca|Czr7_Pa>sDgpc(A@c*0p
zO^Nd=WEKN@2*lsv7nnNkrCI5Uj8iPEeb!1kR|<+JB=&~^(+vw!vk4Qrem4@1g7WLr
zNVPoQA#<Njf+yt~jUiKnyIUeSXF&tpEJ-%!U(e;Q6EHXX8QbmA!8BaDC?sL8Nbahy
zIn*n-1n48#l=ZV|jzsS53XM#<&>5J7t5?SqR&Z$K6%7px<xo>1pb5TI7JwIYm$%K-
z6xXpn&0GtlH9TF_+qWmRAjD#Id-~!g->Fzozw@N?)zUnF%x)cOZL^<#OEaOf+}=BJ
zqT}o9Hl#=0IHnY4(1+Pv*`6N5^6m=A_Vjfs_y`YbSV1;^fy=hld_Y3}!0G;!sxw$e
zH9YAmb2}?u*>6=uTKz!`3orDLQU5M?yC?7j<N%adUU?-KL~yIsu?lkz4FWaNljmR(
ztL}+-Y4fq$h0Qvs9~aqKp(^LQp`-qfHS6w8Sy{Z0TCmBu&|IIuY&}kT;~<ktHRFjk
zwsll!T_r@e-3|q$8a<iwI=~E{O4MX$7=5|Z+S-dxStdZ4j<08~wYas)S9B?e0NV}%
zZ#0U1Df3~7{u$1*X@S#P|H8Tl8fD@!4yMRu2zB~5{jbOg`%iD$FYB{astad`su!(+
z5<c5z{?2Mj(*8;8qLiBhVWaSgwHzw`0SJCzV{IWRW9Zhr5%<4$3?NyX_wd>L^pqdY
zQ`KoCxrO#I?*cFSPnv{xejP<uS<|~|yXm-a7$0<4!zNtqmgk=cQ8X?u-<?c(an~s1
zizk}94%Qv+`OPBg)l71A9k-^rQ2Q9(E}&_h#iohQ!6VFA(-|IS>;z;aZ45A%KPeFW
z=!CI5SysB2@78f~m_cs0hJr|2lZU>&VPz`~LdF7q!R2!r&P33K%cO3}VBu$?sZdzx
z_3Pg0^3k(A`v&lrg%{19<gmvljcAeC^#)(D7~3p$mhg2Tcwi4Y9B9LH%cMqLd6E7I
z&`s`Y0j==(VA`cpcUPTxk9pL9$4MmeyjHZhK(bMyxtiIjILHl(c5IfzVdu*XMa;6d
z{QLk$7z?3pO^b2X&L49Wb4Ttm%Rq4{?yudwGx1JPloIHa5eC6?Penb2>R~||=Xggs
zvcv~J+t|cQG^r}k4*a<}QLpYn(D)Fo`zO=<GXj>21C!Ob;u@Hn8f#kI?>=AT&_j&=
z3qp!=e!kxWKe7-7hsRGWtq`X8$u)Y^YRxnAxoGxFTlxeE&YxWncRZNmQU4y8fp9K-
z&zs(juoL?{Lf`>tn}Lq0h#zaeLGv)}!~t$e{llF4Qdt{>Yp%=jkdXmix00<oh@gs(
zA+Z7A^=byXI@Uu%%2ul6hNMaKB1w;neDR@~h3OQdLX+Rnp%pzq>@yQgIO1~zA`*_1
zbD(num1`Cyn|}H0nt8(ed*iZ14r=v}-@E0x{V*M6P$P7PrzB3coifKi=?QfkJ=k%3
z_Zijp_dc{mE#Q|<qrV!t*N+ig#J1d4Q>-uC1+5~%nNs%Lgx)vvYbMgi&Q6WDqt8#1
zv@lB_+oCA2z7!ulT&7aG<MOQk+FK?F`D|=2OD2|#BO_G_Sb;~EmR|t)nfX9uY3&?0
z-(q)=u-@@E$N8I_#_(ngfFXy}pKTnvT8@7c+cw>Nde}@n5}s?OkQ3|+lIUn2TDJoV
zXHD&T<G*vKsHX+rKv3LW=JJOsj1bY!<JstGlcN$3kb`s?G~6rcWa~#J2Ey=cDDlXq
zQ_u1m04k^1W3r_Ap|*;$JmF~W{H3t)-vkJ}d{)K#^@=Y=7`Fle>uP3Rg-dJePnHB2
z7AccE=D!SsW>JX!|B_?QJC;My3Tq~!L>jB17&6J-*H80P@ao1Xj&g2+H4y}v30Iz;
zDmkj{-c(y|8TmDZzt`d*D>PKT^%{J=?tnzlbJ?dXo4zweyLi@e^$^9#hVujZcJ+%m
z&(%*z|5l$7igh^g`?AWYS@>X@t^29tEYOktxAf{-L8op0kbrU8jf$L@s#msM2FZ3U
z>cyEDF|r`!cRqPpwc0E2ZCOPFZ55J-`Sx(0{!pB*9~N14Y0{5N)-j{XKmLm`dXN=z
z1M4?T>}1AKat<<ss76gOV)n7P!Wfpq-#0O&Dz8K;*v3x>W<Mf468<boPJp8Aqk^`@
z@}m^Dh{7^DZrizBhGgeEkEP!H)dBXI{W)ru$9%2U?VO0;v)8WGmK~Pddm2kW0TH!K
zG&z60DB=6dY38@(i!(U!nmruPNhKKD#S<0Zyr!*N{V94&(Zcyc_*~hGbLy(vwl!Z<
zB1*#u!DNv$&fG>{G(YEsGiFB`$0p3kLzm3>qfSshtW$0`wrn=VpO+KU$rC&P%O<!`
zmr%5PX7*vp!7n%@+B?fyH6_6HfrFp4Ldduk(&7R)U0XwAu=<j&4yhy#*c&WcrUIY+
zIBlPrME$jd&Z`Z<Pj!Xep51cCq?aY*q&xXCs_h2i2#iq6a*r2F`eRo+kw$hhUKF_)
zJCCc5E{8(%!o$kCnr%S}Utlx<c7InVEV@=fEWQ~u<q`sG1_R*Qy<pb>exMjIJn)-!
z0GI*{rUNRcqTuq65X_hrd@=ujX{igmFz~n3zW<_RSLZ)U{^<NqCA&KRQDUhpEB3!=
zyC1hZ!vEA}S-V#pj7b(A>U=#{J}A}!P?f+e!eFUZY?Y;P_kX>U<^fk(aO|uU*V@)+
zm0oeKf5!Re-JK=3un!&<y&pE?=!ae}SpQ<Mn@g}e$~DLx{J>O{l+_iK)D)D|t(A}{
tRRl`q<OwAul#<d19JBYo3HSxLd0h?v?*b+Lts_7HhSoPfTchh7`+u)CYcl`<

diff --git a/docs_old/logos/canoe.png b/docs_old/logos/canoe.png
deleted file mode 100644
index 3876a3fedc342cbf29b3b28210f4201eb2127b82..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 3719
zcmbVP<zEwy!<^(`fG`*#B??FjOr$}OF&LsS*hnWSB}ypW-6)dMA`%-lLO_OejP8<>
z5+?nXj-UU*^Z7jY;_hDDy||Ybt@jj8PkW0N007WyYiS?>fE%cPJ{d^<k6D-y4gaJ@
z?};Jo`ubW}(hyKpbx*>8o&$bza&}wj*{!=zxCM22h4k0gw-S>+2Zp8!N}vGc<#$E(
z-?&Hb?(W+<hCp}`U~X*ySc8^Llbi*%xU>eKSD&3*%E&6+KlrcVd#9{2=F8V=sJvNw
z$3RScPGA4n=lOYkw2QfgXGK+$@?)zW;t1l|D`(e;ZW8Ig^K-$5hK0jJ06@u>NQ@jA
zG3)B8U0p@AwI$EY_zVyK`2Bn4@X*NI!^<y`Zc&!*->w904K+irtZnm6pI`OhPVqNl
z(NfYLl_qpy0*ngt%7p&+$AJ}2)o6!mUBAf8^UVt)T&&@(VX1BsC;k74_(0A#6+?^H
ze|E$>uWx!(45<eAd-v~85o}KbTXwcL5`Q$WzYUOIUI6>M+0dR{7U#VvA8z!4x>2s3
zC{>*u6T;8KrU`_rbqI^PFBUOPxVktbk=W#OsZo~!Z=_V|@s!u+-w~=euocIwM+O$h
z7x0E94nM-Trd+Z&FNg=;h7-KUtS1Ibo{*<d^mW6xmRzzxBWo0AklTrOPbT71aQB99
z?74U4`Nr;|H_|i&#f?L^@BAG~yMuXG=R@u>^kSk(<*MmhR4((7qyGfe(!)Q^eGU^d
zTiYkXg(y2LZ%gKQqe!O7zJ1-cI9sgkE<yQ|neB_IPlEFX)xAHjgMRMU%x`bYX0G`n
zuePa`*2L4!2LfhDf@4LT3(lR`!=-Q4`eHsd$2wYA5wluNWkn#@+>!Kj9W%vFz^F)S
zS1xX|FeCG_Aw4VFfA$#Af=%5!?i<toRQsAo(T(i8tT-HAAeH+e7Dzo(B9)ou?3Zjm
z>prq1?9E2NXknGivOOIvn917<Le~&w(8GlwpI6vo_pEp?ix<lwQZ0)EW7TuPkMjT>
z36r*7e?QhFn!R+6o_qFj6f3wx@zuFfIV>WNnyKsu|NK+7d?(;?#!9q#7CJFy9T-`6
zb2LJ)-wE1B!7nF|o%}A$*aGr}^&2}U;)yOkd@{3zUh&6Ct;}1sCfT$8YbaD!*%@J&
z*~53|9Cl|&qu+r+AwjfKA<(5`zCPj4JUotvKZ^&mNkJPg*EdR27~0c4D3K=n)P#3Y
z<FcG9ItO!>umF46@**Ne6H_BH;5eFwd`v}ab&=Dw%(VWs#g)LGfCr84N{UYdNc#2w
zAIo5y)1K}N)0p^>-rvtm1sR_aOBWMZz>MyKwqYz<?_0m#g_Jl?h$x3hu%}o(!3vrE
zol2^WT$Y*9vtGEL6ge#JrO_oK)fn8EtEWzK)Wb0};04@}zLXUzEzhk}e5~K5?AE0a
z<7q=SEKFY@?CWoMr)`M{C*Ni%HTyj35*78lolPpW#*o@WL1X(WuN~r7x0NO!Qv~TM
zuWHkd;<K&QFo0Wd&8?jS`D>1dA3iyd$ZF^K&<(HN$w}xs)+GhW{+F%p-ckD_8`X}f
zOp+B0rQN=p&orGXAX8oDg$V$d{%~6>E$o;Dbv+?txB3ICUT#QOam(l^3<cw68443b
z(<@y%wXW{asPezz1&&73EW%O^GV&deX(T~)O`wKRV&HTD^GDKD!E)ZBAMgh(ctvd`
zvN795Ge_>yWyaxN_3{2At`1hbBfM2Q<4Mwt6=P)wnEc!_w<u9=-Qp4NZwrfqD)4^&
zxbI*>YkY{av`_TL80sFkBx}Nh4}dAjpg!~luAT2`YI}G0`m?ba!TMt9=p~zP#s!EF
zS2r2|F=qota@I;m#d)3e*YOW;90HgJ?@vcNQIU>H%2$TwNI{pXc8|r#hqAVt+KIuc
z+ExB83znn}5ccPSi0CKMK&4TX;<!px1ZzvG6xUNC3&@mPh0d@V!6w&XEaOVOaI4vp
zWqA@)(<wZi3~Dps9PRoNt~|+S>YQasq1O^i<8?9cfK$ptVD{icfK#9(c)GM>TK~OE
zpB=_sqFxVw%MJ61{`5|sHaaF#(YpDOjGZ*auiJe754@x;wmlNq_zUO4uN95R>tChp
ziv}03{0sv}jCX+(r&!=KoSOM49F=o3p;m2ni)?}U6*yF4QGt!QNS1*pu#o?@gEfub
z+P)`>ruMV~?pmkNR!&S7@VL)c?Ul1P%LEsf7ftu0#O|QSTbZQ?I~wY`N)d~&GET9w
z4nV#isF$@zM6N(3i&bv;qvaZYf)W|7g=tU6v(nIzdb-B#Emg(sQXWLMF}Qq@s-G@`
zj8c0FGY4fD{oVr?a|A0DhVz7dS>oLQgP3t$J$=N6W!vZNn|qT7_lkOodQ1xO`h3|B
z?&|}k>G@NPCj1TijvoHXS>6>h)g4yjJ;4U@gwKeK$|a>PzKUTo3NAHGq^rzA8Q@AY
z@aplOHw4Di2k{kPi}2b@EA|Z++hdfk6n7quMd(}#xKvJdoBUC1>~}VP7X>Ai6pE<)
zJtW#WA2l!%LfW*NWSm;S2>&rvLKGvnc&sGvKb-bfOd(4zij>T;hO!A9>FF5|swh%W
z#8NB7vZ?2g3KvMP5(UuRLs8zI3C3T|vWI?JW52i(BBEa`k<3j>o=~v-mfx~lrcK7*
zA{s5R_7hEd`0&wAwJOFv);Y&_BlYHZZH_j8_GA{b=l@#wT(~nK9^Sc2r5>Oxb5Ij(
zu+-0!i(#{0#H~8~4B#vIyLUbh{Ad<MW{P-^;T8j=m3id3gFqUg>FNS$HVj5V0%G3)
zZs`0=MP|8>WUadc(7iE(*2IXiu#TUrQvc$2R=&`Rk6UTBG4lWjBrtl{JeJQ2;;o}E
z&$&z!<I8Y237Lsd`i+Xl@WOntPW5p_YHXawz~>;kVI&(CYLB9w$K5QebVtM~c4kId
zp(_I3-D|SW4vFO66#NRX*DIMFEf6Y}Kr7<k!hufWm2L9}inoE1cK(-tLshF|cv#0D
zeo1dtgLo~80gH1{#%yX;3G700isD3PdK|4Q6q}8}k%?ZYwv=9~p+O%D1AEDETXugF
zSo3agST><j_yInmNJh1SqTQ7;y+Idrt{DNP$8wO@gUINI8n-s_izqCdVFGQ!iGa=z
z5XfqmtR^}{9Cqqb{SBf-Cj|+gK#C!3Za_?6H{8`Yf_nRhe*HqGPKnVJ4aXG3)k>yQ
z2X$Z2UUL1kD33)M!$dr4ikCQm(rD?|r?&bwyL)kLnsfkG+-p|vScej5@Psk<EdkL%
z>O5qgFTTxwV;+}>gzf2X8t4yi1kwKIU#lfVTN;fr=RmVGx%FE{Uv8<=1%t4Z_F&NV
zU4UC&otN(+1+PNttTC)v6LX*f*w!>p<Da#BN6Y5n{Z>!=hq^x{n_n@)WJJ-?8nP|P
zPM!krlKtISskdB)1RDmiX(U)>eh5!&1DiFr_cHu=h#omZiFT;tT^;f_6J9rD3$(Hm
zUxpnB^GX-&<o9<Zry1`{MV}eN<_4)!u(Nckzz5VBfL8gZL5@c^`ZQn7Oyg>UuVp&#
z!JB*4yCf27N!AnH-AZTpKYk%oG1SCVuf-gZrpKR!N<oquyAPTSab;!kPSEpN<31Y5
z*sNuO>twe`-BgAHWK5wI{*x;?ZH2u4YfYOHlu!dM>o1WTVk%qv@PV+<C2PDl!xSnP
zqHkh$<Yga^tE*h)XEZm*H}<;@1LG|d-o?bc>7ze{?%<Fmmhl;5@7E(H8c)}~3W7J&
zo!KQ!lzI1!>K$!Tl$&45-EbctGu77wO!7jDV40ybve5S4{eaBp+GaBasS#6=Q^6}X
zOxc^&-NW|o-QC+RpmMF7aNkzSGzAmqFa1qP7D};0U}INB+h-7Wv!5!yE*W2)*+Isq
z@hg45vG$}??7UY^!_K?e?WdkUzV1y{5F%<kW+J|D#QA<4d8_d0Ei08LfZ&1N_?Z##
zYk0pIs-r;AMVuBa+(-*8o0h=yJj|ZxQ%YFbw8^B(W5yX$R!$46GCFOf34Y>nkDZAy
zYKYLti#?q`zMWUzeO=*!=48G?kn?qucyO!o8gQP)Q&F)T9bIVLosZIyzM>8!gX#-<
z?(uz{6kqubwW;3HC}#N%xVbE(j^fWdN>iQrJ;e~eQ0k<l;(E%lW(=D1{GiYd?)PY*
z!fozP=9wI8SIvX6{%8`%4+_v~X$HzkKNTHGSn$=vDwVwqH~1KTF>YRB77$kbB-_gQ
z?^2HSJnkXRHUmrG;5$IVda)xh7rGHA`5g=CQya`eYnO*yJ3b6?FSc<nNrfGk7T^*D
zauVO4boiMzWz>DBlNXrBsju+eEg2Lgfh!t)0c0u&n&=uA9R;SHvr>kJec$T}8UCv~
zM~S6Qc`R@w+xu2kd==_{LZxxKTOOs~;aD(X5K!1c0`0t)ief_)-Z@&DJi|_W{Sz4=
zL1+~8bl!=ffh^kJ`;YH9<N6)CTIi)%zdVB<#R^i?bFG&uH(q5rAn$DJo>MFQc@&%1
z^TM-pZDJx3G`hOvu@Ysz<UNH(dtGJ=4OL)x-EHjD(3x6qohPNzY(x2+@H8-^!_1dw
zSFJI1GTRDFwL=+9y>0@~#(+JYh1a#b=DT)wr@E5UPoC0XxRe1w5BW|^3b8Cof~uC#
z+<59%nIgrW_k$uF?ezqexL?wAig&FQOKo*TF@!gCL#~H4ZwT2<$xUHWTm!5o)fd$c
znyHAu)sG(EehD-44G4HXzyG{nGsW>U{G{MW#ozvA`<{|DLATVusjTuCja@sRXdd-^
vjb8Csqi3&k^V6cjzPq-x|Nr4a=j7%GMqqtK*t*rfX9>`TJ=G{zw+Q(kKhq?@

diff --git a/docs_old/logos/cisco.png b/docs_old/logos/cisco.png
deleted file mode 100644
index 31422734e45f9c2977e8730bcd88ffdb17148dc4..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 5784
zcma)AbyQT{yFNq5&^gjd4>1fd(le9@(jbit9nzsv!cfwDN_QhlBhu*5A}taU5=w)B
zAPni9@2<OkYu!KYT6gWU&%1WKXaDiM@B2La#Odp5P*L2X004kW^S+uP-p=8>FBu8`
zE%BPm5pRf`m35Q>pdp3w%9a>E=C;3Yr~?2&`~VOh1ppWLsqifT2!H{=4jKUDvH*a^
zJFne90lz?E_dr7p`1@}w>Uy1upCR?v)KMp0Bc)|#mND1Y6a@gLJWVxa<7e~x`N1J+
z{cVF`<CkL{JmwL%)p@K4qV`B!M4!1rOH`mHmfLRzZwYY^#Y{96HaDG470ggd54zQK
z<=G!`s}Ffbl!)#O78K~!%$raPUnqB*Jnnha%10LQ@R!<?r{B+H!=k1ra`yJFdqVz>
z=zh<3-Z&pskUSXaj52ii^Y}-!^(pNq85-&ilz$o;Nq12D_#RZ#<+lcu^09ks68UWS
zYK*c({X6sd+H;qJ32V0v!y)OF>taeBN;a~3rJu7)U-q-ZgW_HF#$U9~P|@30HDmMx
zcJ#n~9&mA$AAO_c4vhwG9)eaU{>U|gaDz3Mp9cHefo{1zrJR>ApZu+D(eBf!awk7c
z@@n^ohgvWi4nZmgn{sPKWc`m2gz_s}N6g3383L%j`*7A9S_VtAj$M)cU$Rd7weMo@
z^SZ0HD%B91)GP9u3i2ti*L~$Iv}fGS1GO+`-7B3Wm~v^tO5GBLCEY6}u9$SexJp@|
zZzbKM)c(&}5`LqRact+$dFJXyZBEG3>Xu0<A5()_G?cBj!E%?CmC;!~LOnnue!id?
zU9RpVO8#8DxeF0x;u}BD`FBPlgAOV0Ib~^P-153i1D1(|n-fL+g@Wft-QDRO&d9<J
z)vNBnYgiY?z)_yG)2o~>@^tx~-@&#xBx&2ZhUH0FO6x~XCkT6ro@V8ZCXZU%S+y>l
zkg_n^FPPnY3cO7lr1{TMFaBF3`^=a_x!Ik=vR8~*P};YCu{OX~W9oD6_E?V1Kqzg2
zg)tQ?+4F?(Y2aShvNi#8g6}azBq5C{Ndm(W8Kpc<TM7x^<qC67<S!wuLE&=y`NjA*
zWr0GubC!bbxjFrrhJGVo0zcFlaLpOxFDZY<q(iARC;dtpxqbGuB%zvT(NKzdsG%-&
z0<yfBvC=`e#cm*S3X>h4JZNCybXyouEa$7oCQE`iab?1j>RV(<!W1u>`q_a<5_gA<
zi#4D7AwtTpd=vVo`{A3nD&K7#OIu;m7sdo3n6?qz{q#<eK~^J1>TU0y<Z!HgygHhQ
zGk?OVNN}$q_Mx2X)I(3x^f5B#)Cvndw_Nqrmn3X)9>pbr#3jZi#JQBc%<5c*CqpIK
zk*-p&@B9n$zX9+s=Km7_6}&Qm?+)J9y4l8Bz|Xxy1`A9l<sO8^mFb(cc~q~*+YqwG
z4vq*+7*3T!%TX@Y0@V$rW6ni}a`1?ap(O!I_lKtMVWSA35X8C%l(f5vU-m9H<SL|W
z-5R-rM=^oIPmz1vo)(Dl9tx%<^UUZv)eZ^$Y+$W<-V2V}q{Fd*IYB^=6b7}G@`Z8K
z%xubj1+%|tR&gFu*s5zsM2B?0P^D_Wbxb5l-oiw*)k$-6f0MFht^_I+W_DAbf_)6+
zwd?scI}R4UK*ASc>LJ^x1$68dGT}E)U8wKC<E15*(EZvrk@;pqr@i7ggE{Od>||L=
zYKGVOjWU*n!M{{hrzIe}cCvHipG`h(RkI?fFio*-n=}J9n-D>@dW#VINsUG_tI9up
z1N7pd(z`Nbe;O%xEvg+<`Z@fwp<?Qzv2jwhY^|~(+?AOoLkx5Bz0}c{x~7=_q}G4u
z+5e>0zj>ykM9i#{_6fwLC%^z_*aQJ-PD--967=)EsEcOW;lwgcV-ph>E<Adm#)qM!
z=_Z$gZfmWed!{N(@TV0<SzB~fq)}>v!gfA9xBDkzwc~aZS@XA@qxR!0bDAsm(^hKx
z5XWr&69^wyb)b))SDu}NUT0S!zj%cZ2)ISW+uj&t-#OGFD_WKL^<MW^M7{W=s+oRp
zE9<Svm_NLrC34}Xf%A~;wT-ABWt!P@I$<^uR!`<M5|NDR{*#7&v6vSOWfsvFLJdt(
zi(B7_R?>;`_k8dCI&PwynADp+t<W3@q4jBhGCa4E3sP9tGe9gbFPTYU-g{m7uvU_8
zp^&Fn_b(K}v^&O_XF9Ro8D|H$kUICuP@(;hkN`ap&IgG+xEsFPP=>G&8MPnmYXLof
zUsza}`YYaYDn&g`XCaRi{S<lUIrpzBkGrV8qkAbSaV$V5Csqlt=Dul5w6kbxcw*@k
zl=1e_c19+{4|#6)o#7WY#5mF~*3UH?5+D8D>b}(}{$MKa^NNjxEi}9Z2+F~XKPcRM
z0r6vAF7smTqa<sh45z~K>!);&DJ<M|)vWmaQt3cc+g}<x4m?XBx6F2lcJOms+zc(R
zG-0-Iqt<K*CSkoBd6dbZbWMTrO7oOu`(yuoSPe0`5ogKAHig9zvs&unPJ&2ATi7}i
zP<?#w85Y1%8Fc>~8p(fY9w-h}xhvIXKYi+EJMdIkgnH@mW#<E}e)lfzy8o2tiZtST
zea(hFJU*ez_O%Ape(Q*@1jCV7-Z!0<PF9<N4}wFI+s2NW$;fTa{^?Z~tta5~<Z>mQ
z3|HiqY{q<=<SaZs?BN+m#!O+8&-V5w2%yFmVq3(=<0Ly=9=&;0U5a`f+*1kwp}(g;
zi-j}CECCc<HXh6{*UAxPWcOT75d6beN_Vl^T2W6b5J6#S1?uTGDxn}RdWZ<h$XLW$
zHzs}i+#@aKoBsoy=<u8_p;O$2Mx1+IMUrUJpKy_5_edex<O`)9o@h!ZU8Md$W)#cV
z7YM@_ci!f}7(%Q2iE-{|u!JzkmkuXLr~$A$aBHt{&&`RDW_sQK6Js*T`GJwzim+Tl
zr$b;KCX}ZsPg+BrobsTZ5$k_DgfX$08%$gY?i<9)P^@rE1kGU)zJe+aUh!?a?fQD(
zK7uRKsUGjmyn#K7=~YM~W5_m8NAaN|x?fb*iBiL8Z*!BaCBcJE9Gk3Y$KAe}c*^%S
z?LCNv5w6Gq4CGcrFkw)#q<0k73vSTo?ZQA;mvY|OJ!Oi<jL{L45Mwhz&y1d_FcxK&
zoqwM;!yVacl8eyJQ?BF@J@iXC3G^I^n#!4eDNc{u1(fpFjFRboEFAG4ymgmSHxDiw
zc#>Dwjs1g8)a_14COAoRKe5W3jF3o$#D)~avVdACQ0A}n$Rf6QwkNJ#-d+bQEn59l
z7@W2oM|hkc5dALKw-q=i#!AvLI0DJV0OTi)-am|<q~4l2DZcNNHyuht5IC{id~pn#
zds;69MnZVCt(`FnR@=}I#);8ZlpSTs!IpVGwLvfwbq?B#$r(ASZ6}z#??bl4JxM*y
zy+jvoIcvht)PlDpU*YN?w5BF1%BwU>ym1{_`Y9^yx&2%av8&a;{4yU2ztGQKzL=u`
zmg&SGQrF#KLS1z3=4B(d)~f9`h+p^i%}}T*^?N_3M0<BLd5-R?s*79+hL1D4bSY97
z2|qAT<Sr4wAH&%;b!^?_^!MV!-kH_EJfrnI9`ZW6wcb~cc%^!uGM6H;8J!-R-lZSA
zMvqxZ2VahxGtMHN6BF5Lc0ScLG*3NTB(P)_VqjWp0ml1WBkGew3NaGIruE69y-}_T
zRA}GsX>*OAw}F&GvUOjA$hvpQanwzX22BK#(0v<bJFo1nwoy4y&s_@NvGuIS*sIA1
z=mMTbWv(u*;&%&g_gWcnw3iERqMd?q7IvQ9u2HP=zYPIrA^sH1w$N>b?x_73%ttnQ
zgN6DR@G0-xZ+%ckvF9Pc)ED8|)`_RxC*L9%KR}+dt`wg`iI({1RdyXRevOVKW9K8%
z`>uu=|1OZ!zr=|4b+V{+0fm}srI3?`h`s2)5SkG~^Bw-81y~+5&Bmc_(Fin);Soxq
zO<XW+5~M&AM)vldAY%El#=@1SM--C6bDEf?->TZ!PMk1m6y_6AGq`XOssf2JlXQIY
zb#U7+6iv0cF*D|kf<*k<_T*U_eCJ-lu|B4qBJCu{<BAhrZ&90M40Z^K5N^h=i^<<A
z4RPUQ>>HLphW6-P-|P@i<QnqwBraezU)b<CeduML6d(zTL%IZbcZ&M(r8bbQ9BHoC
zUr67TKPly1?otO?*v>VI&M2HG(L7OG7=zW|*l)Kp4Y@x;-&9IJ`C1}V-*`ye{KQl?
z`OkA`+xOV3=8h{iEz80Xqtkh8B*5t(Kr~J)97Ks%NhU!M?U?f?x)diU2dM^0WLq2Z
zIWK9jr|GwxfDqdyY5)pU*I%vb@k<BEt|w%j-_bPw#c%6<eH8uu=HtW{L%&ZXoGQER
zHVHnZA80lf9?nA8*TQ-&9;I)Kx<Xj8z6HBeqk5-v(=z6D3giukpfmbsLAL#|Ay$rO
zknT1RR?@tXA!m$AGfI1;i1<nB=Bn}U_`87!{=4%6lE_VG2mepfjb6N_Sqr1D=v$}2
zM7=~oMB#5}oAqc1Ro?L(8ap{tW&i5u$f^1+w5pt4(n>)`vTt<qlh}#7l>n;BHROgN
z^Gm(I;C5-qVEV<cu6bgx_@<p(#NdgdasQv2q0YKr20+VOQ_$$}Rg*$*l?wNWp?0XC
zr_9}wC|DMPZ!*h<No-Gwu=p|>86s;j>xjrX)QBO(W!bsjh6nwu%?nT?v3}X7e<qC^
z;L?^B|F(ZF6h6zyqU#iyO<{trowVd9lrd^B$Sk(~(oxP^-zr+r7tWP+33w7PN8(cO
zQm~MT?0qgraaDcBsQ<8CPXTrR6uowc#!O0Xp281<#loRoN28^8+W{9pVefW?cVZi-
zyXPaAvOHJ=c8YnC$%ghvQK3ZDnnxeU?TRq$3PM4>^3h0tqvhr*#!}S?3rpIv(c0y8
zsWkqaGGnO$FZfz!<USIRsg_#%=H@t@fF&D{=`TAN$W^E-C5l@(H3YUf;?RjCtAizb
z_dlJysMFo2V||@<IPI@eZ(KoFW`>B&Lpt(~UkFKr&sGI!8ynRV5Gwz1-14OL5uyN(
zrgnV8W=@{<C@eipom9J$rz2Y~Q0aZ!+@C93Qz*f6*Ua@#PN_z90XR8lDPA|kX_W#z
z#@!!9YS1%B^&JUIe0-sM;Z(FhAg*KL0KQe0CD!6$8|T}f)@5eMHfpwffc|LL^Sg(M
zHFcf?qUK!O*AV*tuf5EG0LcB5{<@eBzZGxgdt!{=0<+JXpp<wv{N)9{nZsIvxTo)h
zKgwUNA($h5-<}{ERsE*<0U}7uwjvPtQsDQh8(J@7Sg|4CmKqf98{FiSdBzHqGY!K9
zEb`mFCCqoGz}c-GzYamhbo<j(Yb1(~z2#7S_;{uJwv|n>V(7F;v$Dhpebyw~*Y-%r
z@(JT6y8;u3Wz2fD(uFSFf&LCJ_(y!=&o&W5k{(C5z~9Yy2sPOaxf+vpz2%hqzHEkm
z_x#=b>6hXiRBN8{^AqdHiW}B@U2BV4??ttTC2ph#-#4g_k`(09ePnkZH&C<{E87w_
z!}Ko(SbwHu?XwvqTo~`(zP~FX#MTwo99p1GU)um)q!c`Wel&BXcF7UvKN_(#RviZ)
z<=M4w-=^(^*W}GKYO2-5zd2LO74Q^Ylr>w{{rM;U&5}xAhI7zkt;toku!+;aOjPLj
zVzvvV3BUGx@3jT)9b+~jeNmF*{WkcwdXM#_F>t$hKCKBmrFit(zOnX||A5En2_d8j
z5f{E`F-iB~xQPV<Y@aSt^~h_*Strj&%jwd~^Td91eoxA>-ZR7C+?K{1O2#VAek@V0
zyZ&_oM%xa;(o9!jpOvi;J9P{f0$V8h_|73W8f2apu?1!#I@{{=DD;nmB=p1G1c=sl
z8O$`2JN-JrHJywoEa$VwgWI%PJ2x6yJKlH38A&|6YpAO{>@DydhMCAG@<&~t?$rCw
zDp?*9s9tBCp+fY0OF1_vQkxS{xL`8>WLClbqUr7<6YeH6T%<Vs5We?4-f$w*h2~Lp
zbDB13eWk94t%I!&jZb6no$cTm3A*O1VyEiBP;3X>>gJQfFju<)k%||9^MJSYEir+B
z%U8I}6j|zzo)@an8q`(~O;XU+%T<$L9Mq|tfympBevJTgFg9@g_Z#^rUMi`!{$vL3
z7{$Xevt(a~u*)Ddh?O$dj6~`ATLvvlYSrVkYfV|dj9(Qk1gl=~IjXfy$LMRVWzJSL
zN>2Bz>Eg$6TS_qFu@{Qa&`0aJ?H9EN8+U=SM%8XS4TgJ+JY!RP)txRdkDeFW@36V(
z?0r>Lhh~VJH#}+V;&vxKR)Rgm7D=EBh@yMbxq58M7yOhFL>c)Hyztt+0ofWA%c)~<
zVgyn6^Yu~Dihx^RAz%;6;!cheBq*kph|q2#{JAH<O3JZu1|HDJ@<!0T&F;+v8K?Ix
zF@DsmxIdVyu%I4X&!qkIHToqvT8?n)Qn*lZp5*AB7ph6a^ZwEI5ju1tP&qCkvtCo<
zV|U@!{NYjR`^O7=743r~C~rQDZdKL7cMV9mU$(ax7GS_dMc#yYr$*?pXMWMTkV#zJ
zzHY0ewjo*Zi=O788%~}eSX~hZVm4ocLmn?~E@jmWItO&T+HcMTaFSBlzA=A$*2phk
z-F!6hErYfLYF><aI#l16`lLt$aVpf2SB>45@7y<DE?tNJF|vc3oD&1%ZUJTcRF-9Q
z9y;fLHw%MN1O4lO9OY=!vsC9hSs%lBpS~}<`)rM;Blziq<`f;ruwIV`59|DgtFr2~
zs(2rIVRI{NM>J1WBe~G{stxBCBRO;^&KAtwDd0mm(NE;BSF^*GAzp_fBLnKCZ7g$f
zFGoC2#GKZ4CtKi^2Bx!4Li>=(f%y@46^}Qvyljf53@+ppd%e*|74xP)YPY0RoN6g>
zXk@fCdeg7*&950N4&3PWOS+B*&l-RAE?_B%;9A%bbR<T$oUXBDC#5&$UZ>mVQ@Hua
z6&rlJF_tB98ZTC8&jL3iToI^T$&;_W+#l|Eau)LD@+;;)@KaL~)+a!s(1Us+$>(!e
z1=yIz@2bN=;5(3+$Q<t_gFQOQY;ctfU%sDUVyC>nTFu<r-LXKzTq^o37q$gOe~tX-
zbGi6s+&>0`={f~SP%8@8o0;0JGlC-kp1xw3H3eDRu`dywEldtwt~3YyqwZxRthJ)_
z!k^tn{m5#I2=5NHNqHR!95|QMkjV2mJy$>73zeN_wP<=F9xZv8)KcMkrk2`<yy0qg
zhw!u8!$ASyTdFG!U`}$@UO>zlO$$m7eR8A>9E5Lsfv-mCuNLEWE06D)7gfz*TIYB|
zB3K!)3K`_<L|{3eqEi_55{Rdj=lSQzu(RXzPj#56?|K1+EyoQbI7c~A6P#$j;!dy#
z?$l7z=$;#0fsy69yVi~4J{1!DB|Q`7zR9QgI*k8y%jd6d=I>zZ?<i;Q>xef193~+n
z3cD*RfiQ-{<?xR*0t$o4!C+ODe$M~P!NbeJ)hY1*c9{P&ynuHAG}U$0YE{q?{{=!i
BuZ;iz

diff --git a/docs_old/logos/cohere.png b/docs_old/logos/cohere.png
deleted file mode 100644
index 85f6e6d9904b0d49302a2d933181279e974a7c86..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 113536
zcmeEvcRbbmAAg-FBXP_^=qR#M*UCO+Mksqz%FLEMI@hfZWfdvAtn7q{aJo?<dy|p9
zPxj{b{;2M~<#YP|9>2f8e_Z!Gbl#uW`~4cv*L;5jTvAn_AY&xkvu6**1w~o)J$v>i
zLI08L2fumH;CUAOZ?Chu!udVfb<E>?_OR`_AbU>J-FT{p{Be(t9{-lx^b<qRM217O
zWOCukwvF0Bw&onfiEZbjI-604o;-O)LgwyoMkkt(KqQGpO<cZAOH?VGmm2)pTrGNE
zuey@C+O5Dx<-~P+_imqLneEtfo1^wx&fvnI$w{?SPd2;rmbZN7Bs@s2Wx18&EwV<w
zd@S_hxXnsLLdAx@wP!B^<^vh6B5|8`6#PH#tV5F|Ng&{setRnBJTZ~_?Ky5$0tCRf
z*h1&_|85I(REl)f8E0Ylx3lgRx|)gL2}r0;YNFY+PUZ~29PbBN1eZPG{h<>s?Zs$5
zeG^DSm<UPI8^ZhB^FJm4{rVph{>OxG_Vzzb_@5^HPZR#}7ymPd|Cz)8%;A6L0IcAD
zO!)t10_HG_-`bHhr@zM)4g|+MLaC~qaqyIt@>y+1j{j9vOXH-(glc+kIYJ?N)hxZ@
zeO2x;ryHNRgp;b6*dMi5y!Z0!U7s9Yv@#1U+Q!G;kNuA*v`B_ll#<TUI+n-RY)g*H
zAjdBSpAC0n4y=~kHVAfkPCXuHlIpEc=jfHJF8bJ+Ia6@{t@(*V!t6c13Bt>M+KoYc
zJE@0hB?6H~r=h2~BWF*A{=Wli>U$5beo9uF58EeN_X%kspCrIu$<98iAfI&iL(dfr
z<3lm3)3G02l>Su>GzL*~uKlf-?}%SWco~)Gp^+F$ZB!2n*H&8ReYIpI@AS(VewjF_
za1IDnrD&-X7r!Jk20-Uo(=nYK>UiQWT4yIeKNZttrct4XoTWfIY6owOxu|_TZaUG&
zY5K>%$z0Fzudl0SDY=H#L^OsqxaLwdg4Osa-ls}A*lud{v^+P4V9Rt9WO$Q)2gJT*
zM<#Sd?$)E@%2xevEQRY;!s4j7SQtcOkBUkQt9!F^|NC$vl7&!Qs&j?YrNy+ECQkOz
z5KZ;@w%g-O(#<jgmJ!f~yN<JkTGJ%`vBSOuVvL}mW2K0!l&4+N$_VXjZO!%(<fwcF
zyZ&=Cu74~CeDJ*7Zz{W{m{cT$>eClxEM!?z`s}G(x)ou}EME^j>%|{?F)h_GnW!CY
z%b@NM6Ls~iu65Q>8cm`d9ge9BsH*wWJfcOHzr(k0;Jg>H$@7(xYi^q%ZSISCMOGSh
zp>THey6kqa4aBl=WsqfNuM6`*JE3KJvH4mj#&xA3Z2p<LsK^XyOkK%xvud9v&QXc_
zKWSvbMTDDtMPG_y5cbXK7x@ci(H6_TWf%9DpFrP2my?hrtqpV3o`a-cNj}wUbYyMp
zKz-W5gNC8bWo+t(RHAhu*blURMKnLQxWlP_I}CGt6%rp3zMQIvw{MwXWtV{-(ZkOc
zI<Y0*bPU>Jra!KY$4w%$iFsowe4f}t)IHGUoUBA;WS7C+8i(Wm<tK}N|8`2%dZbJ`
z?d+DnGUsQf?x`uG28d=t2@#Qc&rR0iW01Dpr0x*b^HHkvIG&o%hq{9rh`DK9nY**6
z9}4`!9OL3#=kb(!d_-KBxyx;`0KYjGrOdq}2^(rE#jJpOb4W?X<S<9YU2m)nq($6|
zs`;X!+KsAokd=7DV9_Y>-_4OQ>-1-1H_EUs?;1xMnWP;ygSID&#CahmV_uPyBz38#
zvq1}7qH%Kd9wYA#%!tk9V+E<Xq=itn)2~<-cX;#9CXZ4p&W1F8a;{aH%3>D%lEajj
z=jJ+g?E`e!7GS{YCgop9sn{Cg5e(DCQy=$PeH~D&wG|y0pdJ6XDVmn{lO#-ofOrFa
z8b66J_k`MQ;0hMivU205A!1+I5C|?ce@M+e^-XKY3vYC6+|GM7q}26@-Gmug)3<K|
zDJdB~0%+~<W@m$@r9@IJTy~!t-^^Bh+;(c_vo9X5@)Q~{1m|0bzJ*Tjvulw6-#m%I
z$?52nJN7G<@uDQz)xF(hV>JKF4f+N>KeiZ4w{lh1Q%AN%*k|URU#@*$S#li<UJ!a#
zBzRW&OtT<()|fJeuXQ58KXqt&pQt3ef&@D%t|ii+F%a7Nk4F&|tFtL=7$}X`tIOJ6
z@thgKRvC3ZPXYhj0HNNgGUhBKVBjZ&Oo?k8Q`+2prNE=216Ih~v`;MNoc|sW$jlC@
z7+)KfQ=g_Me%j|9Ur$qn6!e&B@EE)DW1G-pjw41*e8w+uvi7)KsLN$npTaVS@O&J5
zfBdHh8PiqG24D?qMBC|e%D*I*irp!L1V;*rIh^6KJ4W?^c1L07%l2>&3}8F3Kc9(Y
z;W;j==8L9MZP(umLGj~_NAgv`2LA>sf4h<UBBaukv>j4XzH?tN;^q8+(WxC_sJax{
zAIko1J8PFrO#&AlX;88}=_zr;f02r*R|Jf+)I+ni{_%tY;$cG|&WYq^JMO8)2-ZR&
zP}zqpoler`C($Q-AU-6o<b_I>ww&@<R`B$up9b|PQx(7#q1n!z&3vHNJ@oXDZPEOk
zQn1CSqFM-6CFvgl?R-FgB)0n~y8Q*G!|k<HU+a@LME{->o$wcU2(@u>7$A7*FWjn*
ziD%OV<>>cUS07$N)?=}?axsY<{}nb+3M&k=NdRqLZIz#Ug7EfMT%6oIbf$CQOo&Ix
z(3!p#APmx5X+&Zk#6NM8;6X_~r2P3A;MC}?Yr)mg!Lq)Sd<ARe1uvWRA?1q#%GYjp
zu^c>Q41o#zIFaWRh!0`bFlq><|K&W=B%elN{OldgeNChfn(YUxB`%@ahE*V=!8sDY
zdgD-Yhr-1BQ?Ta85|n}vN>y}*gZ$@b?7_5>c%M|*=o%0;*y44?+hwQRl7vXh14R+(
z4U7X{Zy-ht4i4_y_8V?UPNObC&CwypTTWRjlS2pnSLqUX5%G1%a>t=y_Zq&%HJ$1|
z@<mCh#!SG2wtKTag73cI#(c?mA9q-JRRL8=34E8fRt3j@%m{j1f)S!{6x-Jg+<}2e
z{miC!Nnkp_D2yrqGcmpy3Gj87r?}vw;=)HId`xwQs^{6~G)i8`@d4BLkLF=sk~6C!
z+ea`989|dHQkPv1&eO0pcmYxOmIxdLx|3jwc&B5$mG!!Fqmg!1$>c$fZ6?_-&p;77
z$$rvtnr~mK|Lou#b|4Kh09P`As%dHT5OkqU*H>0P-rbLnQ4}pbt!~({kIyFTCuAU@
zI{OND*GVbIC??M902{RX5+LAidqqFMJ<Er<S4OQ*6C>QO1iHJ*#<Zloto|=cp#x3w
zJJ@Ir&;>~@t}fck9RRD&z5+Gh8gN`REBTEh<wGqL{+tq=a=zze*V?35S}lJT%Pn-p
z-Ka3X3AF9j=4nWpwh)$-AU7dqQy>gD75KAdhl!CM5-3$mnF}X=k|{VZ3sMp37w7D$
zXJ$V0PO|BxFew07a0H^r8?p$Z>SN!0{a8ceXfiE77VBJdrhw}w+Xc@!z)*$HJgDjG
zx_#4`H5DKPW|1B^yJd>A@!*6uw1>;Arvh@rO&Gc2XpqtEx^=F<YDGCw6mq<{T77<W
zxxQ}O%rD5C5%B*4FgWGHWF7GB&uXUEtf#EPqWM3Dr6a%8JLl|pm;ao`{UnB6bf7lk
z?OPpw!}U5pIXSQoRq_?`PpbFCL%!`JDpN{)c;$(r;}=Hd&R?tc;cEStf>(NKqxWyE
zmr47!znu%RzltDg1F=tm@mw7gw2*(Z`nU%sRH8;S$PoK089Q+BlWOMBR3Wz?p^Q7v
zTGT|FEzfYE!2W6Z*h0DWb$!6c%ya!|gW@cXE~a0jVg}@pvoD0gU+#p2|2Ykt123lA
zsZKigtUx&>ge`#}*riL$5C$K$xlVhxcF>q{l}o4_(jlX@_1+8r!q8i|Kq9A#Zl3Wm
zW%?BRjZcBXx{z4tK{U(vrye*t21WczHP~2!YPN|vjS;zZ{}i=>?Jp1M#~>ipmYz%A
z=rWjC%;^Vl5L%pg<GO{oP%8*8aT7#M4J{o4iBC+T)DLE5|7G-xYJwPrbVEaR_4>W$
zd9N<*%f}EzN+3+*-lW+BR(q1hh)9*qw!D?8zlV*Kt$5g$K{@*0_;JVHgZV-!1vRz;
z_BD%qK97aO2wEmtS`EHnv-u}aOGx>iXO!3&6rHP=tw_}WY$pmB^j|)r6CNT&N8Jss
zzw57|&-NKDcbzk}K(aNM05jPtR%-?pQ~8g_U3eCQ-;!-%FD}x8T>5_i2RyY*o9)Bg
za=53KS<03t6g=n#-9lRnSua7BMvSpFGfNdz3${Ux+MovR)4$UHPhJDI4Zc722WFx0
z@`ZXU$j23Do|y);hru>Zm9tHQ9hD%u<Gm`a?o#q?x+E<OV9#c`aW?KcPQMM&Bh<iR
z>QTe4Qa60>yBpxJFQ`GtuN0mM$%h>AZct5{oziaOb@<>1RAm6)je_uR^#46jgA8Kz
zVT7i~+pY1}<ugxH3Y2c)#7Ie!+DLmrM7t1=^B5dtVik5@iK1P#@Ij?-&$P?_%+*`C
zTw<pRR*IAn<z8u@chmR5HolsIsLvj<OB)77cAC}<rxROVL++ouYSMmTCE$f9#;;a%
zyS?(s>ZDh;GX+3zB=95(kB^=QxdGWhHSKa9R4)~>uPH|$>fW#We{mH<i?>ZlW|ewj
zfw$0k^Wi+utYbj4DutLH01PKTINjx8m#*mx!nwwou8Cc;^fc@b4xi_tHy`oiL^*}j
zKLdvOZ$cI>vy)45&Dm&F%vYbu{T1-`A~NWz4mh#I=*D=or|KBIl>+&v0st#%H!Erc
z5gF~-?FnTyvDvYa$W-&KaYe_g7zHliTz*!~F-{EsaJb(MM^9fhX^QLg2qJw3D#Zdj
z^8<Sk*Qq;7Oa9m>xAX8Fa=$4nk$y2TdKW}A$YZ2?^7B{kF4xxeHolKi0!kzU^ogaI
zH;bCBK@aogW3GklP`g5fhM3pO=AWzr+Kdm^*Ku>}EgWW?$!@%6{Gi+|oDA@O5>!B0
zCD5jp?RCx?$#d_qE2qx>B7CsemQJ^t%eY`=h_9Tw#7(4IU+)Zps8}B`p|w3x6hS1$
z_2Q<})F-97oB&#2-%6bIu8>dtX%iqp<QmxAw~r~#XZ|LB#qi09Fc0Pr5D8R!uY<t3
zKLf!|lbG{~s}zM2HiMF|pC2%W#Z4n;WrFlKjgE@e9s|gl0SwY;^{6TU7xIH@_kHb?
z7E-V)DT_0et3Oi)0Qi7x<86wHNwIk!D_&f)X*EbCffg`N-7xPMDV2KyuBzpFU+6pz
zFJGZ5_UmHcwoiSXRjSBaYM<~cpNX{u&pQuA&M&x?ZlMdwZureC)!sc6KUYmVD**BU
zgY_1MpZ5+R<1fTrjbH}ePc1ooU#8?a8kPa|vhu6@sM#FiabH_H-0N1`H7Ytae`47V
z4~jKFKGHi<?=#N1Jl$Q`5_@za5Q_9jlC%@efwPc2aXQ~-^mUs0YALp(S0l*g(J!21
zyJU=c-;cXv(c|4CE~R6Pct|tG56sW6C)J9Cswf`!$a=AYg7X7^3U-?j$<+UA&;l?b
z95>2Sd$`-r$0s+nxQ%x&;sbC{9*4odB;y(?#<Ihkauh1e)($M6xc*}nzzHq0%Mh!N
zIUJWW%)6Aq@3+V#4RGNK*aoFZTpKCXiYLZaCn&aAA}Vs5vU>R|i~>+%!Uxtg>`m6*
zmU=C-%!CA7JeS}N*f`hoI)D^xq{k!<pR{s2d>|Ed)y(9O)yZEJ@D@4%pje+)Q)3MS
zaFZ7Uj@(yGumDa{((5K8P(Ue6SU@AIWy%5**T4<#-CE$o=Wtq1!se;=c1BrTlLOHj
zh|gQ^Mp_?7Tt{{YpK)z#4DX9Uo#LtaH1bpWwxbKW^i|T`&uh2I$U&$Bp@+CzvIc@^
zhV!(b=-LMD>>AdE`|z(D-h)vm`y_0=Xc6h;BttWH4&s(^)FaYxJ=0Z<%e5`u2C~IT
zpAP+;cJ7cXGDtuwy9EiG_s!!a0Hf`|Qn$X4nH_mo4UNq^RoJg-hFk39jz1msYdr84
zX`(9&w<r4+F90^U<bVLt3oma0;xe9ojmvR}UzHo=Cq!G4ax6yAYef_REX4w$%)DVR
zLb5sJ;2LBXUvu&KadWHspa0dS+0gyRhz_OFme8k6;!|XAp_hSMD#ok2fRC~_VwuGS
z<a})ciw=_PqPPbWK{lkwZ;)AAYL`5j#g0x807UD~ir)rt578aYbd9h9w;j){x63mF
zs&<%|c_xPPRdEwq0B2#qH6MlcTy(&Qij4}n<5MZ@RuY-5U`Y#+CUtnVM-CjYOKUHG
znIa|({24u9u)$%vV`$SXP#H)US3l89j||bZ;amSDeuj?nnLTvN)ym51!vPEX4F(jz
zGQeYNpUqg+gat(R`}e9=akTc>s(LQT{d;!n1cIo^f@h!O$1RWx8KWIfkW%SB1<P+=
zdIWgyi+LK*IL(zT32LG185uv9Np=Jcfc8Lin#V`T5nMTc3%3T4r0qjb8i>|TO9(C9
zsaoY!qTIV%R0$dgx1?~_Qy1$?b2JW!-Q9!PhCl#h{VtlOmIHww6vWoDX&uigP`Vz5
zkr(z|*nHjlPa7tG12oKogykt7gia%t=Xmcr5^jMS7*dxx9QP}9p-m-;gOg|n$(tA5
zax7@-I4%hTMt9nMenZrD?pv+N{ezirbL5!80dIv}@gxbea>5ZV6~UBI^I0I|BVesX
zOzioDd_<Y7iB(Q-|4*t9d=AErp(?C-o=L#jetJoMQ6Yg90R5)JJ#)5(WN1yK@%>l3
zA-aq^RAX(na{RRj>KHmf7f{J9OQ$)Rg{!)@Sm<1FPnq1*VYeLPkX2m9xo3^GhvTet
z6Nx+_IPs56xsK@JePQdp;8^>1Asf50WH+N>|LZmZe_7bRN;4L$sa*yX4+Dy4veT(g
z0yNH^02P7)tcx|*-h;cM5VnS95Fd-2s{6DpD?1d&B0Eo143WJZ`RoLu^yKM3MAtGz
zoBSp#dlz@=FhA1;nj_lV@;Ggq1PKBt9#GNFNwp?Y;&k}3ay^8RIrEJqDk$8}ec$fs
zqJ=c{R+hZ&R+QAyI$feRIgn(XnmTl%ffe!YVJkim+lh6T^}BeDY!{REB9=*#!i`H$
zOBP0hNNE}gVx|w5=@8dx6lFA8<5fg`&!F~K3%~K$ZDmTR&^e)?@zp-=kNNM9UwRHu
z?2>UMDTsH!SBqnUv9X&Xu>rf~N#Y%g3jw$*-4E=e3uu7Nd|os<jY!_rX!b8dJx?`R
z&+}>fMSvgtnL)SEPe7Hyu=9DV&2{<H;+O&uV6|~uN@E_798{b3_8h8?m#=uI^vBCH
zySGd7nbSc~Zc6`oYhUnRz`p||)E+fwz^Ienzj(jwkJN<;?DKGIhj+VmoN7f4liYGY
zmb|^aj~w7%8*ph^(VgcK1VMH%$XS7VDjh_dn0%xw;g{V)#>QgUf2q#n!@#EvDNxGG
z=Kw;jt+7TVBM(0*o?>!hDLRIVQ;&K|0f(aQ2+^_!PB%=7%C_DP@-40tk!+-bYK9E}
z*h!*SDpTIp0$L-e0YRva`)@k(J+cEVeydbbxo)#qZXt6ER}Hw#<NPuZZAEk^Ro?Fp
zr?o}w;ghuu`#~oDD=0J+<H`kjlo?n$@|N?m><j$St6=qTi8Bq&*%aK;boT*D$+d8i
zWbEj%Z+(%U>6^+SSZ%c*-8nSXxn4<^^$H8(0wVPjW!{N2MQZ)eMkVII>{AH-B$tGQ
z2?79*jyp7)MGE`jr>3SFAGzvXOt=Qn?152{BdNqC(`)T@_Ey^5A}QFs@!f|87*vc7
zUP-Lf-@L2eXLSHG2f(-5xt-(0*P94-jn*1xffV-20CyOEQR$YVRc8Y$$Jh9C%KnTc
zS~CE4+qY~ZhM?qzaC<&%K7;-5_<g;kyR>o-MwlcC%c=`htM`t&64jXaUO!}yn~pUb
z;se9sC+T%V%q79)O>ovnq)%N{Ya8^0iC(|U><oq)Y*)$Qqx^9Yv_6_uR_!_K{&z3-
za0-lPzlA=AC=}7?^)&U|@?%gpx{Q$I22GpBd(FZaZQ@hR<x`ja{r9bd&WiT3zS=Gs
zgXC5j5znGni+A5%peQWNueplw;RO#EtPx_wtbaE}tC@RMA<3}R6^Ix?+PpTi6Dw9(
z9Xj@J_N|+Ndlit2zmc*LhdpbSRWS8a+5S8GOotJo1Jm~k*4$->BJeLCp#6X`ugx2)
z$bsk~OCvE-E6Pe(JqG*Nui6QDkvj74pj9%@DISJ)gTk6gDHj<~sFcx2)Jcf}!NERT
zo^-0+vf7KtA{%PCZ^3J5;%A@1l*QSg2Gq;9SS63DNXPVA`r6BwL7iK$szrx5pdq$q
zOWrKvo5EAvNQAjb1*#|eirGnNPln#%l=l-?_q>d;IJw&`0B%nr3gcWI#-Bm_vh&F}
zO5|w*6l~qi5&<B1r(w5pcPUAcMiEXo$D}aKjy+eid~Yt_OkoEdK9ZzB5myA}Z(`S}
zj@<G7G=D#OE_rh4PkVnh#cyKI)Mp#rFuNx(UtT_e5T+ix4BT0Wam*`H0Et_zk#R0f
zVJi3l+V1DDr(Q%TM5T#(o37n4C@#~Wptei|NFVfA2|1Sky-gO(yXhh8?Uh>Z$^ysl
z4EmMSM5g`xE4cVK^V`|Qe+ChGUIb@1V@NSVER!0ZbzJTAZ`{aGoRx(IgoPMf*?Uog
zG7(U5UosqJ<aj{_Dm2^;H;%pN1$F5l$uq$t%W@To?IjGOyImtF=!ju}NEs|9_)U83
zN=OzSf&CZGC>Ii4y*BaEzQ&`UpMBsUGaTa<UZLy8=PIAPe%%{)ewip3l-tD0!sv*K
z<@<Y2Po!Yob=w7#fv?^zvNn{1!oehadqa`sfCAdwNOU0(S@Gtb1%-ZNOqfwCx$}vw
z^M{8oz;Jv_9AhBjGox$(fVQy3?cotDt`uqjV$hVp%S|x{XnLs0#(ds4ZT)NY9P$C6
zDycry&et+x(l#JmL%MJu9M&RRDe*d8k;r}ht!pi$*M&}t>=qM%o}sGRdrSf%yy2$W
z?2kayqV>13h!dXVpvrwaHLjkM_a|b-n}6z_e>*lti#%!NgSq}%+di))U&mIV8^=K`
z2MTeehi#FdU%uoMQMHKeko~p9ytCT}BuS6$F1F|Jf*67~4D3}l(zc}+dARJQ7-OyO
zp~Q_{ehqL8I0Vtr0I1ldEd0uG-4N)~D8r36y)=iTx;!paM=*i>6n`_v3{EgzY+ozi
zVkr8&cWNX|)?DWjcz|;=a-@i2a|i)-ijHQdI(z`O_Tp}#f^KO6u1r2_OA;BToMpOF
z2BM|KS}n$;cH&pl>4$_gHm)a~&26`6l-|u>5L;1IZD(DM-Y#&}RQCrZ6%e|t?~hSq
zL%mX+{;d8n!UyU22UINDtrDMk9CQSlZXd9AE~GxzFbQ~TTz+nI`3<h^W9~dpqqm8y
zSi$JsGcZ62sM4==Fgy}Aaex}})2A|7FscA&#nPsE5tLMiy9(w%Yo@)CM%CT=xwrj&
z8yGY4p+4)Dk({S836a2wmwJ17;;kTfR!bjRD8ssIs!=}kfZMd#u@|M6Gx}Nr{A11~
zkO06zrdv~=4F=T=x)8N-OPRyNX}b>!Z1p_G&$4AgP4|m_4T(!EP>9KFS((jKUt4C6
zhUhXLmZ0J)%k8reVuj@}jbfNd;iYIly>;$%*#u|6s5(9I0Q64ZC&-_Z*YA%BLs&zn
zvb@}HIplH$YTNM5<i+5MA@OXI30|V#X=yX51rOto6eRNsL_dUKGngLO*aR_)$!T$M
zul(r<y5pA;z}C@*$pB(O&!=02Rm!{C3pHQ`Q-1;ie?bpQFT)@QGci)8@FI1A3V&9P
zqY<j6>r2!AcUgVrda*7S#_x!nhC8+#$>)f9oaj9P{F)20vi10-zW{9#Eg^`um)rG~
z!9tKYWZ#1G;vjM%dy@mkfOME9x#WJ_UWha@bx&X83>%zyDgGF0I{m7Ym?iR|Dt_I}
z+U+arJQ<1!n}E|x(dml7bAS0bVZ%N!d=^HHLCZK4_ZAr6=(j@IC7<i0`6}S3!b-P6
zWQ+T%xp<fTuAji$L|CKO#W3S7KefbbO&v?8(FKh`>*+h(R4a@iA<xYXpA%<)_U9d#
zz3xTaCf9t{)u9p3zrlQfO4lDOq;P*l4HYGa#vFndbdb4E4(&c8_>I`{uYTMns%l-*
zBFo8#Vn+`n_<&mIbL3$_z5bPvkgMvvrSW}i^Wj~EIOz1CRl?O|&)YI$-1Jc(RxW`K
z7y*xd8?`$XChMV@g*IwiMK*@`r?H`b_&mwxaD0E02h@$tJ!B#N05qtH)IXef32MsZ
zF@B55oU!@&R;v#$_x=6L9Kv^-nml<44D*;xx|b!AN`TC)dXVkKehly@lq|jrTmati
zdNM~0V3H&dK%c@)gCcN>E1kw&G&Hj_{XjD{93k_t71Zywl9_uO{@LY!KDZ>}`b4a6
zn$)H-*{2Lik|rr_?YUp>3Dlu6QX#258;YpmRKl_xtvddJA$7F3W2>5*TDKHPXVKO5
zk*?`V^@OFM*(ub9faZXksmm@M+>0>ccJRu3xv&Dcom}PhW^Gc?P4gt>@2Rq(*_qQi
zF?Urq=EW-|SGcGEoxp~nE$@7sz3p<YHR43WOJFqD|6r=-{hn<zuYeh%blBn_knJRK
za0OZ{$Fb)#ClnMtIuG1sppEb~aiVVlnUy~)7CYa4S&af_<Dh}b9cor~Mk9yb^UeT~
z_DQx-{bS+RiSBS@_&q|!$^8kd#*)XsA*#%#b%7dC2OrPe_trVMuh$k^k+bB}S?P?h
zN=(L3G@Yv)T^}DCE2+P7;Sk$bDiBT7`-7UIJV!=-HPUZCybhI~{84mSx{CLlVFl{D
z{W<vp85NYdb~~}?-p8sChwD6m-E1t?iXpG9?|C!Vr8-sSk%Nd2U|`FGfrv$Rr(K&X
zcy`4NIco_`x9m3dvFjjUO9cU2d1sBlSZN{Tczpbd`4ztvq*f~PmJ|ZX$;Hm=!7|t7
zA~ay6-?ecz{=vlI*ZwTfy|LpY%rd5yFFkDoi^$-ueNCS2gEEuAV*7Hf#Qt$Ea9lN`
zNOSRHQ1bVzYSm55@sR-BZrK{R_5woT`Od*70HIS*cYUln*Ea2meIR!4Y4{V0V3^0R
z-z#|CzSwOzGVTkAE%${T5<l+nrDE8EGG(Msu44SCSkoE!)z5KrsY$~8wp;Igz|5J;
zg6Z7|W1xU*oO?|5`ZqH8CC8JJRaoz1odaOF3$(YKc&=X9dcM4=ParD08`yY{EBl%r
z1LjM>L^pfoa!$>cLoj6lki=LJAF|W$@_wau;U<0QR4Qpw^%mk2>pKlI{=gW&1H`Zj
zSd9v0H-4>~QJ%s(ALaaG1XNc6^VBG@FO1vj3`#iDj42ozLfy3<ol5Mjl-@jWLRyVY
zYwp0Mk=ergCQcLY-_qFK!{${_nf^Nau=5Nh?$9QFP?PjcOKN5+c?X=+A%F2~8=|JK
zLt|d-O(%?+!{p&5sRv0i(=APnMloV-twwwUSQi6opmmt;^se6V4k?s3KUWT@+#=Fc
zFNmkw5Xq;!k9*|YyUdCSU+xnsd<LH!T<A9`f04gB)opHp*H4{4YCyf$@0iJhd%`1<
zh;EMeb=D~xVc4_&FqH;~L!UQuX0{d0inloEe+)9Xwv&P;F9Ygo(IjItY|;xRFW5eC
zG72nmX0^}O@!bVz$Gm)_8hbd3aXlF{Z?C{=$}L<wsYKr8jPiTrm}fvw<UXc%YkePs
zDds|P8?`<*hvy^(LR6~bahgdBMb_9;X<)&d9yuw+@wlZ?W5=#9?-ZmNks%1!m=iR=
z=&K@JW(VS4Z+er^LXGJ2TD95KhodrkLs6$D(^CXt*J12vs%B^2U)&6|vr(4b1nVY3
zyaX%#LDHn{HzX}}WO4l#>|W9RMTpfgQ;9G8+S{aWevY6A(IALQP>+X8*<akWUbqku
zV;6+2Ep0APh4q*(d=a8^(F`{8{D$R+Vm{{pW$-QSd{BS0cV*?h#sSJr6<LYoAzCm#
z2dCjY2)E?7Z}S+SX^K}bKk6es&`>EJRja};C*i(UjNxUK_$NM&2*K49z!)uU6*5OJ
zv(@BvFyGl?O2{j?r(I{Yx~8+(WEij<-t%??7D05Aanrf^t7n2bJH>Tg_X#6RI01B>
zTvPf{o_n+<g{qs%czY0@)v9D*rsDm=XC^U^q`dEyGK1prJDeHL$jf{^{rlmq*^5}J
z?Rjgxt^I}8YEWz9o}XUsoO(xNCQZv#o?QR_4|rg1lS^p2|7EkiouPf$;qDTY6yUS-
zWA+*C6b{l>8?k58Vh}0tB%_I)-Xx#V@bit%;MP$GID3J=Ea!bMm<L$MHCRfemGKdZ
zyaKBhFUk-mTXX$i1nf(fN==OWgkyRsA<-69KBzza$C<V2m?r*+n71pi42I$w>Cca!
zc-gFzXXt+Gsl2HlIoc((b~Nq5A=wJk_#ks;c=v@t8(X%F6y7|?u0Q4ob{t}2`lOqc
zg;n|(W$$VsSJT590=P#)%XR69V&9wX;@2Sz9(TGnp(*Zd@tJPYyQ%5rNKxN(`vZb7
zy(0iB#pl>QOakW@=_<qHIc5Z?;(LwK#92GFN>A%tT~%K6c$e%;dIwI06ZRqE%O}JA
zd_O@m#)=2(oMhw4R%$k+N;`a6Cg04?2BjYA^)LV>ed&6uU87YynC}vVuIuZ2{KU_q
zeIYR{4j91V=Ljoz_m^v3Sa(rUST<3OJ;qT4>HFlHrHVes18nkfOWu@hkSmerKm0tY
zJOHcYr)Dh%yVNBDo%f)*bKPQO>vcwP)5UsFb@d!c7cBN!UwZp4L~6JVdp2_aE@$*@
ztJ-X-?v(>Bn~~yP>CJ1P)6&W>qW7mL0Djle7!$2OFaI(+{armUGXR~^06Ik%=bSmZ
z_$N?e2D=L(XTFUV=xw>QupNTkER`-R#_$3I|ICx_a;a4Dg70bgi6#AhwI8zp6Fz`d
zPq7{9&%%OW8VR5(luc8`<?H2Gy!Fti{qdb*5G}MW>Lk5|!gKkg_ORVC81x&=O@@_7
z`7B>;*`b?w*?p2AyqQRw<TlBZq0y0$N=_`4s~{bY?yZ2e08p}wnRnx_`-Gn?t8;0o
z1*T(HcC@q-cvMh0V}vb+gGM?vCfv9DjO4PJQB#nxr!3efJA>AHVFiv4+0|h=e1lWw
z@OI8XzOfE6aA_x3*Ck7a<>v*bfEL~Cylpwkg5evCj45{f`j;AL4n2YQbD)GQ$Zq5b
zZsCmfv_{w!Llf7vg&G6B0Mh)Jk<z+X5sC0wgx3#%Sr&etX9p4O4jSWx;(#UAK3e-#
zFNM|JLjXITpogo!X6fc=sf(=+A}zZ#%1PMuyT|a6K(Uyp?$v;&C_)s)g_k%j>1Sr|
zl2P+xICL8X9~|j9AGv%FkrPneQd~FaXgZgaDO0%K1`5cNeU7H)M+LPTb#)(Xv;u_U
ziQ)ITVm0sg+SYX`I~%X*KY0qIEXzC%I3K9N_!|41diOlZjt2EoDRO1&KF%&L5h-pR
z_JsO%8hxE6bl>ZOcsbar1T=JW$);sOL$Wp3y#0GO(<cGsIyV6n#VW(+`9z*d{lP$8
zmDFx#cHRWFxh(P$89e`KK|LSGz!h8A+A_#@y$})MWft2DpG}sHN5y860An6@)vbwP
z?xniM`3mYt9YLlJ5<1>L6({`7?7{Dqwagq)h|Juw3tVx81{+~)t050j2-=_Ip0B)u
z+#nrCn%`wsHl2)!jMuRewhzMYJx^d^3Bq4lIpmFpD}L0=4H3)mZ4h?z0NvrZ%wAfg
zn#?aZ3Vx86JQdnZDPlg-+cl^Gd=tQTOp%gw5w1#eaV}=i2pTwqiB1yuiMH}DXW%F9
zT{w6d3m8MCEyt?#V=O=~9$cni2ZO<I>bA)iTHf%yHl79fnl<H&%sL27rbV=7&m!0#
znt`cf6H54`lW0wmO{7rB+)j5aG2ng^G+57cCPJ(XSE(VcCLP5SAAr>eF7G0Q9!D@S
zeraa}w#g-%bL7RLe%Q&vcXv|X_Z}AAK1s-r2w<35Z+vRupU6wQ-^>9hMzckTmVqud
zzjRL1jb29Nz7Hn0@VO1i!eexq&3Ik2{gcgm{0q8$UDLM!oAy-&7b%#!zx33Ql?VcD
zbvR)zq@q<<(<{eUB{xN5)7ntLpzg($^3Z=PeLv4}_rDFd3@EHKXq**;4(CLijp4i7
zpdv-a>55xfS;1KxIOM~tE)hn+buAR&MwtrM*w;PJzsphrBLdvQ(fUZBs%2;8He7kU
zGGTK*Fw{=>T(i!8=PF6Qmn{RhX~#{X-RVVCSH9NtP*uNcAq45Tp9c1jRAYX?3xx22
zc?jQaHS52%fuGK@)a!l;O|@mHE0q}ZTJ9sjpCokP8;hsHep+h6Ipg1N(nv^bEeKq7
z7@SK@RDDJOR+Et{7ZbJpysc7%#sUF7o~th{g&>&fyN|We&TD5Ou+uA;<c*H^QybEY
zUy1bX^MDv@6RLaIUo2K14apKbuzd_Ae3}3ZIFSy;aMD|h<nXt*r1+fN3zEOfSZFdU
zL_KhnUfu9A0g_-$s;>crJ&tZ7X(iv#!{|ZY_sU`!ABy|Lk<Q2PS^)0Mu-dgu^zFS3
zFc=I9Kb?k-2yO_{+45(y5nyQ`I!{keY!eq(YBCjZXHhQ`98$NhNvluz%u;M^xU_EZ
zTMC%D6v)C&*h8ac!OYquzVf^&@Q%Yr^R-OfofhgrK5fgF4j)yLEIh;U1EAAcygKM<
zR}QFcrX*(-+3MVUt`{J0wfEq!w+6nW&@#hH257UCDw6O5^?u|33S_szoAua@g;#W%
z(AYW*PdxSd>p+)KHMplFx>Vw?-fng<Y7t^Ko&iQzeY}!3M3V3Bfq}T4v(Y^ta{03L
z<Td3nkTuTj??M5eQF=_yu<g;+Kv-bBh4v#!@>VhP+$vl%JFJljxhQ%`FFgI%@spGo
z_zi;D`>mwQ(d*~LfKvA2qKLs1-J;@|;*s-Y1Zb3#dKZD(yh1s}^|<dtCyvs4<Adjz
z;o%w#`^{L#<PHA?#+@&ueFhPiQ{Q$}D(e$wAGct^SA+?fkPVs7zp0$GU$HbfWbvF1
zFkSoN^)LH~Pjv}-zD1wd<U0b#a<Gf92?#F9k~5Mb%Fln$OJR-6+_I9rpOc2Ig^?3j
z)X)wPObE38dDcA@-yPLKv)wdAkq4hR{m0Etks*n%l*r4+2>3A~mWyn1aR$$8O;mpn
zOa}+dEE83b@M|T$v3oFdCk%N>;#~|lX{L8aXfzkNB4ha)gj(suI%}!RLRBjJRkNCC
zTQ~;@V&EoHZ;!8EKNcFV_wq;DIqdV(0?L0TpBD-%d;;As)G!fRsZN$u#ZF$ZD$jLS
zWT6KbDdp1>nCbe`f&6!OL$;h{V;Y*|@4;3SpzDv)UC&#$6F%*^9&f~^wZzacumQ?!
z(yXG}SMPQ6!sTILuYZsyEtCX}zh;Pdk}#D=4pry$AN#p`Z~S#wB7bB2d9!|pIZOw4
z$DZ6x61BHa9UKN1LL!!D4C1`qtrs}iQ@0!E<eA{rEnx~_0w9Pp^61k~^MIK5I$xcX
zpx)*OT1Qucd)$j?%;b&rF|F^de|8c~Go{BY_s+G*&?J_qD+%z~=Myj<00cwc7tGAJ
z_^*U(fKq3jm%YaX=rx^}Jy`gr5xUI>XG1S|LkCua#@7r?e=M?^2urIe;~T|L7Uq8h
z_mN<iIY?ueXQP0Jjkr|J1-#v8X`$dEh&{zn90pz2&cMPCH2C}7VbO&<fjGC?1J4uZ
z3EiTPsH*9^X6H~D%?q!c(B06g-10LT<%chd0_7ZjvrMNZ`Nw4+71dd1o?g?POV}{S
z1DEBbg{&7p{LS@^k7du&+;*lxto*l`krtVj{&Mi=L-)TWo=d6su^KG%S%1V9p#~UQ
zCT=oYA@eY?UM;Nm%>V^G%oyB`CEca6OxYOP*F3+?r2cIxY@SX{LPD>U4^?M0yPq6h
zpF!V>SUv2N<fZ6!XT@QVBt_usbOdL?_@<Xrg-Nz$E}@;85`)?rIlaxiU(uh*j~1Q)
zg{eQ4Z!~yNO|y@MtE}8;v1PYUUh$ji@p2#5G2-8<?|R~@kI|sDztvBP)T9b-*jFnO
zA%i3`c)2gx!qSxt%f@DWh#RGDd1TmH?%btwvS?6~J<Jt5b@8t9TclX$lfQ9axQ98;
zTb(1z&rgn3HP7i2kV}ZP;`}&uysxIlZ{5hIv@fTCg(coMu4hL^80<!%uiK0ZbuG>5
zQ0v}U)|a!6u>tXxa^QZo{@slZ=u#pa)R>MVdVC8-3Kr^p^ceQ;ODq07MCEl~Mde1F
zXt_Jqpo>7fOu|exwp>zrhW0zTKHtIGXoJcVUWYcsCO8{lNdcrdMsC=PUz!mflC;&?
z3#gG9HtEYVPv)zAL~g%)WFfZelF&ET$;cimV(hyZChj(4c;lj(oIUO7&^anQj!PE{
z`0q1^R1nDKh_q!)L+5<l)ard)g_d4fcn}{<s`e7GiMuh?@}W<>aLLcEmeQ3#)FP<B
zMj4E+zY%Y*dPRkj<glX0fj)+lVuLvWiuLCR5>#dzT`0Q!q_eZNtFLSrY!g^<rbjxB
znuOzEgj?@d6XdLP+lIpdcs)m(5vdQQ*1C=MHa-43u=k`X+wif^Cr`H>*$!eJuZ3@J
zgG+eUWJ!vCf_<@@HLViwdWV4maQDS#DnGmW(&uv0Szn_=ZYSZvL-0a>#HJmdScN+?
z!$i?z-vQsFLN}Y#MkVItz$Fk^A(|Ett5-yil$2UzE{QJjN8U0~jF^;aJE78ORgOY!
zUHc8Oi)`hIeX(BW-zkZdWn(m(<8dRkSpyFwBjqcMqE(;5POpF&`T*SedQb1W_Mzi<
z!{u@4bU~e5sdMsbw(*bk9pA3S%xJ>HsbvB#b*-m&lc}SmF@hqA(PtAWm%8~1+tK2%
ze-R!oOs9#;+VD%hPSe58>PLuxvN(`WhJWc$2@xBqqEPUYL<Q@nk^~dTS4Kz=0(y$`
z)*>aj7)^hC`QxaSo~qTG6dafS^Rkxz3Zm^?|DDRonKu@bw{^1KH3l1U0+YD;C;z)#
zu+uMC_e)1!RuPCDYmtX9_<NZMmTr>QT4knTG-dj59cdw(m7lGB-aK7wg{@XWg?r24
zg-|1bJXZY=B*DgJfS%BP{u;;1YwP@|O4f)IA@Fl5Yz@aTE?3LN*XuZh>^yb$rQIpJ
z^{lx}5H-`=ux1TgF9J6Z$Sy9Hl+LWlJFMxG9ahXrcD+z1=4(BG^pEM&d2seBtbW}>
zzaw%Qvfy<Y?QMC{6_jnm){2|t=jgT%q*b#wy%_;7xrBoVu5DjG&zYPWD!Dra$kU<v
zOruRPPE!IQk^@y4_@!EGk(t$bQUZ1xL=8svi%$1=c_b={yvf99zGK0SwN%ik*V)_a
z+TJ8UQx=4i*U5W&?<vt<2-l(qG_Uu+v^BZ-3b=^4ORo&)2plkx$haB5w3To<h{?DO
zoqmq8|1Xv3nIgA$iwCxxeG~+rze<NUzDs{#m{(cpi=FFv^>`h%FXPbGbdEKzZk}oN
z^&E27=yduB+cw#YsHIj6tB=LkNv7##MfX57m!-R`%K7Cao}G)(BB0MsWHEjCCDl_F
zf@#_^fRvD?a*;*u_!((xrr<Xy)MpP;yuxu<+Y`LNCiljotlP`wp_oW%#ur%`V|1gH
zG`I`e5L#?a0Q$g){nSMLg7523$m>kmw=gXZlFB9Bx~ZEp%(EHE6ekI!dV-2<o8`W4
zOXn!KKd+?kSEq1Cd{Ddi*~;PWMi%z4721f7fb=#H@zZWz%Y1$pN$;L8WfS84NR1ch
zda|-eyP9<(_$2&P1Bsw=KMz36)F`djowfi~9Jb|BC7F=Z`!%)T#v*~Wryp~;Uf#X_
z#acGP=`doy`nyZKvn-V<DheO&Oy=CjvMV!dKrb|b(G|-_xH8|u(LMu_u{w$*Yv0bq
zhl8mp;pW!H8j|oUo<Bm*xOrt+%RO3h0PuB4AqQb|Pj%fE++IXnmV*xiF(V`^w>6z-
z*50S=R~IV#x^CO$GIXYIvwrcY(O^_M?2OSZ^deFAq??ztW@!g?jwVo!WG6aZ5i!v-
zVd}lwnCB=0GR`9^>%lJtTWJrQ7Yc=@&u%h2oEDJ_Ol>cEW(~^%wo57)djKM=Y343z
zIw(`N0_uXmOfIo$9@M}ox&WSEUi$GPz!6mFp9E6+5Uru3u`eyE43n$TP5Cjas3>Kz
z1@7tqh)k@>y$kxLdkVVfl1zeu_BrKqFW5F(ZClC4=sm4^M&LkZL`}(?GgFd`8aj=q
zL(|9EPl`=bDpdGe>#Ok*5Ya0f18`@h=|<<)hCx)pr;vf_lbX7qRx~-V<rArOJexq}
zr9lw9GikDZVWv61OJ&A}P3v7Zzu2^eTm_T0uF44l#O~r!0WtoT#6Wsw2owX=E6cbz
zx=R($80U5j5Ioe}hxMM<2ZU3ypKaaI?CHgA3fqd!J*H?kIf_(h%ZipFAS(vUo07g*
zrYvP9$wPeo4C;4<(_0u4V`p6h1r5>~{w6@KnW(oNzr@m+-?VeAyl&R%{B2=S8mzTW
zBY<a%)MG#?0z9>vMwWn{TQ5lks;&zl9pj@~)$hWWL9CC5veyHzA|{lcRDBFuv>+QS
zK3x%SO(E?4T0TN^cRat7c!-e=t$f(Jh!>gERY~<HKdjg~cKyEVdp`lWwY!X3>bD6{
z14B%2T(1-_$1rMZj0^9O6AV=KQkIp-B}_}7Y)EIJ#5uM7Wt92>I-pPW?-b=MHVlyC
zkJmJt;BF)Ue3^hqIRH{~Vy?~O{QzNvJdPx$21}nUXo`z!xePb|3@zggox?Auj&2p#
z(cLV0eG8+hVqDRoxxOm;fyYK&0e*gr>Ul0Yc;o)<na|PIhL;Nrfc~})U7*#NCTIV^
z{+m^WIL-3z$+nVbt|HK}-g=##t)LMx_86tmk*j#m*m9Xb`fYvcwBl-RN$TjS=N>AH
zcEJ9%{6`%OEe3Q9!upK-xw&Cb6G#ockEn@_MXWq|`7OZZ*>Vu}TYseutXrvJZz8w8
zBu5NyEta=%ij1GvpN&tOSfaJ)D3T`T_XmDlk_#JHSra74eqM)n12^4z^A`W!=NI^X
zTSykipTzCk$SJY-SK`u4X>%(w_vwIE1%c9uImwW3VY{`w7GF6hINfZa!$wO5`+m{#
zoB<(1Yeq=Cv-+ktuQ^Xxd!j)=Jb*X7VSUxPZncsw=a|%g1I&ND3P8k%01+9!AtK`&
zOrz?$57z)v==rnV_kMSDT2Q9H55%*iwa?Rsmx(scr~hsL<u4l5ju*?>B3ql!6S)m#
z;HD!1G*~`B2hXnwlxjq2>gjBMEK@4SG&XYttND9hPF5y?yAuQm)>2o^z6W-uSIeE1
zWumCHLUCWUKut~5So;gvGxz`#M2hOck@fWp#vLG1%mI-2h4YdNNAHm^R*l~-s;hqz
z|HRAoo%K#0ayuRbC#);$m)$ZQh#Lt+3s(@L$r|)27WFo|)9>4X?;4{y94$5;>BO?C
zC^vWyc1&h3f`YE<wNUZtJX<%JNs(8tFq(2XzAjE1h3ZN<IhNkJ1ZX`)__qy6y2h6b
zHr;8nzhS#CEzsDf@ySXmb|XCSJwr4+<V10W0;pG%9WrmTx3d9jdN(aCHXhK9m(d0B
zLvNh<R~i1X8bOWzIp0FJLUV0<SFnVvcYBtG#*OjwF~5>O8|4V(5FF$~E2e%a%MBd8
z%8ju<W&vmf3raAW#P(Z7X|oUYMQw9TU7)vh!K5z<orqCTI(QhZ((q|%zdGHU<;Jvt
z$fOt(&9`GH0+N16tiIO0wJ?Pjev**ZLy^SoJ`aFV5;=aE?*6Gi2pmqIExs3DSNg1x
zh=R|aO{?Hu-<#K3hA&6EuxHb)8;1#mz2@Xe%xX{K$76eL+%|tD5}Mvo%8)8Kv-iwl
z@64;YmGG<fBEr-?S2h-=)-MaH0h7GyW_+<Ao<+1*lc1a3gV8no?AK0dSK6O(G72*O
zTyg#sjx7$St=Pay`bc;wlT3l2b&FX=rVoSBtUOgVb{`+8XhwIB?-V=}s8)1A@ELzN
ziVGs(1tJE#Y(*pC?+Y5Seel?P(#xe{VQK9LRHDZl1VP@i8(z0j_h4r{09`v;kX}TA
z^nz7E0AMY*3tld@CjdK6CP%;PeJ9@)!rP54VqRbrjKb{wXN|=`CU)b^Zx($%*`PG8
z<Em1>#S?4}GD@wn-g|W`UZ|!-ZdyG8WdJsZMBG)oo_T+P?y~!dPk?)*D_(vk;Co$G
zEaiNd93gQYAX-`~DNVd2r~{;lax=NXSnhR0F;bggc!uszLBx-_c~PR93_(^M_<3_|
zs6tlB!$58z@B)0Vn#hj(1&BE%kr$jHt~o7Nwp1Jmn~x_P1ETv6Ah=>X<KuitAB;c@
z7@D~U1r?83bcV_*2*mfSBtzk<dXM8ILw?W!)rA0Tk>782R>uFr@5T1$sa~hI7NI*l
zAag7CaLy>YR$qp-)H;@Z9}Z4{x%YD*)`8aula1m73Wr%|m0Zj%57p07(_%}<5?V8Y
z;F9_rKXde6awjC8qV3#|v)d0fy7UQV5_>QPBuRcVZWew=mGdCxIc$UEh+mb<HXE&p
zgvXs*xLRT-i5%%>FS<wOAdu7Q-}=AhMtc#3REnHSC3*OYrW;VX^0Qx~mERwz8J*wk
zESmJ7{=MZ6pP69gTtNe%K8R%#m#)`{GXxSaK!kS1wY*z=KRF=rpd2$OKD|zflDa(i
zcmO{80;eZU!Y+Su^9qp%nR)t<y{U`!04djKVWdQgJ0UJ>0~ZzvUh@!#xdWoTkA3LI
zBij$L3xN+l&_ZukgIo4li7nZEW5dOsi4h3qgx@{@x~%L?%}Oj9&`bW{;N5`-!-Vp^
zp;qWd{9{e(raf}DT=_R^$T2Qf<>FqHyhMjUNExwlls0xr2Q^@KB7_U}Y9}CF^!j^{
z@q!AJJ6D<hi74BD#{)LRD)HvbXP(CIi?Rh~aoR&6kABNGu7R-mp5FC_Z)*;Jt5&Q3
zPCDeg;H9mbSq3>W-^FtyeRCt`UHuVdUo`?U#bpsmf!x>^Lyp0d9;Y3q19Y~JDQuW~
z&?w0hz>h<y&_UL90m*IO?!lEZAk+X>nMN!8eeB)Hv(J8g#o!MS_iKauK@Ueu@<tPm
zeU$_nM_qpVYWik;*kBytvIkXB1ezzc1RwRJ^dJRTGgpq$bm>}W|GVMEhT91Oog!}&
zHovmI*mXTNfOMuEmz$QqhJO6eV<yAsot89P0iTHPMI2!Gx}5mtzCWZ13wrnCgdM#Z
zhwdpcs__z1xl4Qfq^sYZeGE8op|7%|rQ%xMYl9~DtAwn_pN#j^O!?%2LgTMm642h*
zw3?PrgWAs^?P|U9ZwOIAg!0Z7L4-`?&xmdnD~`MUZh;N7AfpIk*#oJd+`5%l%vR5S
zbM``NaCV8nZfc;@<J~7yy1Y;y{C`J)2O^LNa;gp{P<q&@iV%zi1a<n^H|^>v`J%kG
z3%ft_Cn>F*4pfN_4yI`LOZ}@+mY_iBs<=vBu6QmTlps390hsl-+iZh|S}|VHX6OO6
zJPrI^G9nUY%>*Ir)%Hid2o?y|J@%Pt&}@s3ZrJ~(aAVA7(u%L!bdVi-ud(OsbEv!U
z^$M*9O*Cq>FyeF+{KXyMb^y2y#N4*Sf3zg|{0NALv`_ww&}>QA{G`%h75FM86&t{+
zaCHk`PuKmBfXGTQKLSz01s`YPIf{Gjn_uIy7>rvD5w(;jpgIS&{#`$*t|NYCfeZI<
zp6v>Qheoui9(YySvsF9%l^Xq_!S|#<!M|1m2QG~2I&q(a8XkkGif*HQ&y|ZWEAbN9
zL9yju_{6dM{IW@dzdM-#=3K_j?VeG)by|9e`Cj{g)33y?j1vwmqd32If@Gj?q_xfc
zHNy8;fnl!DO4(SxJr>K})_+m~4(&lShoK`M3O6o3E$Q_>P~^pnGmxVwnHW?RpY_Y%
z&y>v#D%G%#0`#r2h?Xel_>JCNpGmYNVOGMSCoY}0*{HwAsj_insy^Om(CnxJ0qA?g
z89L|;;#8kUg;nh;ps2vzSsS7%HB%J6pZvtV6YvxLz-uJqctDm!{(?$35ES{;!`r?_
zKNFa>L3(^(WEtP}X0h*>`QirV9OFh#PjmM&uh&!JAgHa^S-=qM%85!*rdNiarw2nS
z`!KX76Kc!8jHH-k5UufPSqGIdP3HTFD{M?N9tHP0;q~N&H-<R}l)~c*H*RYSY5|d7
z6Ok5cm69BG*ZX)ifIv-*sEj|<*~aCHdAHqhqDKJJ3N+J%*0RSg#?XbSM6_3R!(ZxR
z1Bc@Wp{^I7R|IhoXku`8E`G1dM)E#73%^5jzdR<qDZWmv-tPT5Q&AxDnnrzpdj;`n
zzYECJ%(s_=;SxVclr{7n<;Q$S2J*bNbEj_83+h9sYX|WWJ_u{V^&4H;i&+Lq<K0BI
z7f^UT9GEq?{H;mWz|CsrMTnC(Ftn>y^F6N%cjU$ZosI_J<UVQVjcp~ul5^NBU|QN5
z`%ve*TyT<v7vuoeudilo_%vFXAdzl*!)7-KR56HvUJgucx!}2K42EnFeLIS&e3GNP
zT|SVNuqh;rBSfXsr3IFIQv}A!{{RP!c%9Xi{;at!<JIdRLg8#R(J4@acQff!&wxt!
z<Gld)#<OLi7Z~rhRY7C6)P<(o<DCMGlzYJ+dI0u)wYXK!lC3~2+Z8?w!bFhsj<iWi
zpUs=D;pux@odSa9h$&*|$N37_aw+;9cnd`5>dfas#=3;9D?sOua`eB5WKyIsXz_*H
z6~N%L2!F+p-kw;W`Y%&g_#j7F3jHv`!t0)8K6=X=pkZDR$0gsBK=Wj-F!yqRwn|<;
z4LpD{$DylNMdt;ab0pswWpYqmVbgl41H9NvawL4MSOvV6J@#$YfRP(0!E~#fXz+3e
zbl&ALSF1wg@(J9JV@^TWtuB%aIG6i8?$%Z9mxqoaN%Xt3JjW%p7H`iu7y6`wo-NBu
z07;?eDBZ;OrM;ue-~H0G9W1=V_971?jaNz5JO0E4L1^nm0zobnJj^zrF)>YC<jr}b
zLNAJ|GmZmXP<p`wdVrv1(*_?P;X;YZI>3~MP*Uj)ng5q#JO&5v1P8ab1kF+%Jzy;n
zn8=b+&_t>9CSWKsv@6=6FcEfa#)1bS0u&`;>e%lT>miZ4Se~@x`U4KFwwtz~p5FG}
z<6|no)F;(<;T7w&A!bs(segt+#eTcZp|F5yZ+R$p4-<fZo(K}yF+Yqvc%4#vts97r
zkk=y&oxuWq=ecg_1^`hED0r8`hzgyrDy28C+rBC@k^cHUkcZ7d-{*s%`F3I`9=@7|
zdqM0p^yK!8p+(1fR|swyoJMJh9{g}UHn7`p!EX}wBCJlk<2D{m$3unb_6vZ+vHkXw
z9nYFs{wkl=dTBsFlZEINa}%X|#M9nWXz^g6u;|p66eQ?Fn8@BZNf2xkQV{D0RwGaK
z{;}K`)uxeg98$8^SbQZw&Qhe~HNrj!+zZUj0tvSF(Ts7;Q$YWndK5q^2j18?_Mnm)
z-r9&JxW1Bu(sc7;W{TDmz>8`ZCgg$r>N*M)M+rOE=iqg3zJ<#bYpqSs??d}5iTGaB
zxt0r$O?xn&qz6$0TfRrNa+LX;nSfpPsDf;s9p#%ucnu?IQq^dk+i09-qEdReA>^$X
z1iP&KB;__N3G-nQdH8<PTc1-=y<4|I=O$@tXNOOpHE&tVZpac|Cxom*+Ut!i#c4kj
zv(}W|p{C{11qNh#e#44@e+pcN|1J{(o8*%FoJja+`Xt}GP0s?UBcOz%^CsQHX@1#2
zuLy#U|B??Tr^i8M<5Qe5Uc5Eu*|}Cod0owbD^A0%+$JpK^i!b&BJ0hiewm&Ws~J!+
z@}(Eg_w<IKAAbI7XM~4eUN+m6Vmt%N(_TjNaC9@s1rGj(A1}~K<6q{R^YA&Kp3ie?
z2kLpn2Jr!kA-{3K#owyZ8B8W~Vo>TiMgKiACm2m?YM7eXoY|hr&T=+MH}_7Pg!)r2
zqj|-q`H-^}gn7YzG8DDh>g~4l8WgXcYIivJ6mYN&yRt}lRXXQa6A@lxw?0V&{x|8K
zjVH3BfkA(?%YPi@?lZJuKsf5vyvR1}UcGS<KLvvS!jv6d>;NghPS_wN;UMUFI64mR
zHX59xdF_e~RLpxhc{KV)IdAIOQo=elK{90%lyqQ9%wp-mgzTgInH^$ZKs}FPINVtM
zr7k?c-9pQO!MD7T66g9HMZ+^pJB++{N^H6>1gk?h3<ZDnmU(tK-@T=NvCytG8{sH-
zBNXDvDX&a{Ehf>X=-(*GpXfTZvHDM?iDZ|r-+(g4k258YnWaV5#c~KA?<ILsM`Qd<
ztme|=IUx<mJeQ@~thSYc-R%h*^)P)$$zy}%)&5Vf_#?$rTz+ILabjEjElOU5C2%qq
ztu)Iy-&`uq?<m{IxTDAiPP<MF;$scgrF951Q;*om`^EPBhv4@kwf-;V!|y@I>&4~9
zE`_ihnf;B9RdYGWFy8-c?_0AP&u<}-M&b=Z-}D>&SLR2w9Y4lxUcYzZ*FTZ+y`CR_
znQhw<gk=oZFLKaeq9KR=Rq{LZ6D|={qa6CBTSc11PobRjP#@3rzk-ZmMHqDRUt)`F
zND~8K9ra$sFjbKhg}iU0UwRI6<gv=9#K7O$=8u>FiscFw^*O6=4-u$_D}nco)f<h~
zZEBtPo9qIv4-9C6%E@2}s!3P0`2vAkWs_Xs4FFMKNa0o7kr)-zf3joWOo1{!g(zMB
zL}miFf?BD+#!qCl`CM4<gG|Aj3{2Xogbj=n&i6ADe?(KdU__7O1q!%Py8px1b;ncr
z{qee8GAf(CN^VxRmSp7?GDG&JGO}fF-G)TjBV;5aWQ6Q-E6P^(CNq0wuitsDk?QvQ
z=lkmWa-Zj%^V#G5InTS)so2;Hn{%E&A2bV=PhpRE<j$ccH_R=HZikgKTD}H6$EW@H
zLvM{pp1}QC+9P{<>;;AviM|wiBg<0ngj{!%$U1v=)F<E!_}ELD)dX6?9OQjm7RM&c
zq0e_<iA~;w@7Y>)b4Pp^Ont)Q)<o0dYXBiv`goV_A0Z|+A%LuM_eQ!-QPYlJh6#g2
ziJEo9@)8@<nN#E}?hcm=+y4QrU}c|qnnLNzx2T|z+ave=SS;BLPh8b9?k_OMvvxh~
zr7RRTS)SXL?$%b3Lp4scS{50Nh;A@wZ5s#c5*c=G1D{T$^*Ud6R+YkDcW9=p;1zPd
zWiWNL+5EI1Y8sOBpN!;@vuyMY=NO-#`RjK4?bp1GMS2PF1MF)T>#18NFcT;{Pw0{o
zdix`rXdOO(MM~$p)ISLP80oW>cWcX*r~@DjQjEWe;e*sosfS0-rh}ip|NGDl=Krv#
zJ#QbNd+xqJ)gWZw`9<ZHg$AiA4;0Yytab`HyuIKi<71;PG4S*s0-ti)Zy<Y9+QNxD
zC+7@m0TevcWj8u`oozBVs)G7g{(aITYT(Yt)yHA^y{G1Ka_Hp<fb`01eo7U7X4g>~
zN~itrvMl;6j=amV2fy~Ef|2LWi$|?BoJp7=%cW-CPs!_5IU7hHKmYG4@bM@#j|Q)x
zZfnhi{-@3c8F&={^?}dbExU#9g1`P*cSU{2J0dUX`vUaUwk&h92vJrLU+1;^Z^=4+
z)S_CNzHpRlbi)%Xtx&h{2F2@je$+*AKP#TN3MOcPnwLa^)Sc1i%s)VcC!^4vph=eW
zB=>PMzz2eT_B&HgW|s!hP2KGGjQxkoRKc&<jIdF;JAUz@aaPIcVudwBQQNF5MdYH3
z2>HK@_v{)7Ehpi@!;_<nX{w9zgm{Vov6oy%rDD^eLD&EE|2!z$*XelnxK>-{MsH|t
zO)ESZiUa!B*!ai{MF&TM_cs}WS8?igu>ilBiVPdn2FYv%2-Yp6bo!-W3s4(nR?tW0
zh5-K%Xde6%YRP^RA?nuQVn4+pwcce2H-`AxKhB<TYVAcus(r}*#eZKA4x)lkfXk)j
z_^uF5{3agA(9`q<fJ<1*wpx{bO7HcLe##TligT^eqc=oEt~4cX@}deLDOfk~mg(F=
z2Cw^+(PIHK+(%F1O*}1seX6CmV1Z8(9?lMSIoHXR(6RL&@08j<n?z_1fO4)s;&b`-
zM8CV+7oI-iy9JP!)&W=)Oatei9VtsPDCqHjdELIYMjO}T016LeGNlgyo~M?jpXcA5
zxn&fZ2js)ApnMoc4!(rL2JJNT>?PjK*q04b|L4IpbTh~YZ&6u<tucd}BHa+sgROGX
zIf-=T`TuUwiD=R~&QxsHEJ7cY=z>zGL9W|xFK7>nZ+=M`Tw=i5*?Uga7L9waVLha$
z@C4Vv`{=7sK(NVvYBeZu`h2Fv?7zE}Hz+Ge?x_Qq*`@!k+QV`oKm}^)rcK8d1&Po<
zA*nI<z?GhcTs1YtukMvoFl8J8=1N~TwKY-Huz!PO<<_`ui8c1}Ti+%4PBWPn03dY0
z>0n%OFHE;w;~WfdV!g0j&<{%GxcJ*C7tfoFWVkXA6;OkFuc2zwQA$gau3bnL`HyfX
z_lzI?)~9Rsbwv0AWH2en9Md&9xhdN)xpB}<tJWW>`v+EMi&FfaoqNI3IEg>Ffg<A~
zI9@z&>bS}sg)<vAb@V!#t;EOvflNCD@*(%BHig^s!w10?At50Ad~cItyOH@LJKfjw
z`8f5*@4u<tu5y3^U5VwuQ0vP)>yf(h>lc`1!7vrBJR4HOm$6*YS$5E2#_8pxG<J9A
zxn$AhuyDt&#C8NlOSCKm)&;rZd+CgYRhMn2)!`I5t%>#Fi>iIXC#Za#%IZMl)hGis
z20tbRS_&D<E#|XrgUjAll3Z&4CNatAp;q&}%`)G}jVTi&Kr1UitKiPm(W=-+Hy4qj
zJ2!}Iu|v?q+5K*Ri7b8ZFbBSPw|FTGNZ%)5<ETCQsV8Qu6?`|sHu4u%xGz_cV5Jv=
z_?_**kc#YNSH3j~7@>c{rHim#_Ato-tz6pW>A;>_wa09*->pT>;@8^+x@86;gV$SB
z!qKcCZ%u`b8Jnrue6|;jdk4)R{bSAaQF~{1H0l8AJ1}xbTdkGsPeh*FDhz7KQduC5
zz*@J8c?EQ(7L@tiJD`~lRV6?mr6HQBVSv0S4VHJ>u(St7Hms}}403b6<<wH@zCe!m
zHULyPSwEzG&bbpgOD7ht^SpQxD?kR700voQBLIRcK51itDFf<bnY<E+yKriGP^|a+
zG4B)U|25d<U@t^B;-u9-bolr&q{k;g_=hZ*S~?P^tl?0xMin_*FB<1e<fze!)z8M<
zf~_Pgn+qC)K;j9tvZfXcO6$RA4cLUTe>{t(@|V$GIi`#i$_%RbqYW+M0>-VpS)muT
zuDQcZfX{lb+#t+n2vC3JEb+C%9d(zma^!9}`V1A-lF-;g4X$P5Mz3)x@P!Oe<y3IT
zG4!?j+{w}~K@g&jb$kVtK=@OmHQ~!dcw}+GDDf(pstMrCdrG02`Oi@<bG{?x<>@rT
zmqM^JRSm|Q(+w_!P=IDQbJfb<Gs+zUk6U$p%9IDd-mPaRmo!`GXt6&GNkl8Ko*ObE
zgC3-BZs>MHd)^=jp^f>GohwpC7W$9t=kYE|BfdZwQ5EjD?+d`UCV;grVB@uk0ZC4g
zr%lVN(x2FGzMy8atVIkdfRR3}uc`;j@q@&H<rN8i*Tf78C^MdI`AyTVSbK&jCj3E9
zu0&5U&_$BvNy<%wS3$0NxHJE`>FzYfQ%_}_0dK4~5ulXFE9TdHFW>IRxNGHEVhu7x
zhBTm*{)6D5c~IGN5Lv&iHdlc)4+Ni51PD37xjB=l0>~;Z)b+anuK{KB->7cdeBL@o
zCzu*LTaML=gTQ!kq{6-EYvfwWEpuZo82Anjm{$>~=I&1VksVDzVxih7OoKJ1LCxX2
z@F#u%<^N7DcAbmhlNifUAj%N!{n*wdQ5rN~A;wCrV~SN}PnLkh!;}0Bz}d^vZW`Eu
zFS&9<I;3FFz0|XZqcrCBeTU^%Z(s$gSp0gp?@l`RPg&c$6uw-Eqk|A~pcCyzbSHAL
z(7G&LIq$&_th%6@hOA>LHU)r#8qbPqtqIWGYY_t_Z`ZCTR>qJ~r0o)-sWP~PUGrCP
zN@!9qr@EV4q^;j%WZ1uI`k?31zgJC0Ga8+pYg^$Mad%51bP>CcbOgail{HnhjMYa4
z*~`Y$J^WfUBqu_O+hABprYMT|()C_!B67X0eYzxv)qe=AJ`A1N9gH3p^V6MDpQY}{
zs!4siu!kq_4PVP$IxhgnQ-n}s9ZOVqBA=Gl_-|$W3kr_rz)6!PNd!2sX<=b7lLPp}
z>h|Q0!sM5slOrlc9?MwCo0kMWb_&KV&RTC$y#tpc69FOi>Kzb{wwlwrzR00{^Pii%
zkCgG(b{!n*b=$0q<fj7PSCXRz4^wyPgE~4Nom(2Nk|MUoFXTD;y;=}55xou0_QLp0
zOmdKkqSl-S2<7K24Hy1}$q#4AuqHOLvbSF}sePNyjar4~FZz7ED;E{xF6aJhH8t3p
zbsd-5^2*jiKwFXH*SBLGq<5hb-_fbH(-AVOl}Arp=S;s!L4mEJan}bulzLU*7J#i>
zvt~7sjL8QK<j!9M8m=eOmCj__g!P0yuuidtzDg*#RxUO%7Q2sW>ln{be&(BO9ro-^
zk)Ywmw%nT}WZmo&LSP76v!m2OoD5;}ZbSiWk(*VVjrNVPH`ojcauPoGWgZb-kfFa;
z+L{)A4S;RrL9F8^(!9JkOsFC%l0yA79$OtO<}{2~)M=rI-*rfKvFKiXiHD7HAcU!e
z##<yn5$y?Gf7?H%TcAlP?s%Iq2<)7qCF}%)9|J|gF-sh&>=F#vAX+6>JQho>2=P@B
zb#lZ^z<6+!zt?Fq7&KNR4xnf;YjTrEKGpHA;5Sy3punbE$!9EWeSSI<cDK7HYuh{p
zpT>eZ#mxa)Ph4dk#_+`^yPXqL=x=}GJ2@?m?fGy{zk!EI@7){Mf8uPl+YoHv0{|hQ
zbgj0?cD~j6HU!HD0^xW;nusBn&+9c^Qrq~&mhIk~%6Qwmq{Vq_AC<J0%|)#3bAUE&
z*s6ZGp+KmSv&9cxWP$-E)%IM!*D-L4_~C~y#>!GO*fv{r#?n>to#ST7mrYUA9sy<@
z0bsb91HpE~%<qV+dDWGL($C2rHRA8YMrCAQ2#l6m;-LZ;=sw|KK5@DYbzxnEX0*X~
zf0X`#Ntbg&nJCmH)`O}zsK>4hNQ&P_szD~uicGlX-1czar7{jPY;_Ri08YbUd3BdL
z41pa84#4r)ffX}u*BjEAL7x#?J!Y@kwfx+(O?5My6kGHpXO41F0-w-<W{avK6~GOO
zJN}vXnJGxdB1u-xJ$d)|uMhQN;?1_6s?o4Q0|PE$pe(3QV9<^4WHD(vN5bc5GT!2Z
zG7?d2se4Qj?C@rmODPy~t>dx~sl+3G142{9U~`lI)V|2Lt`}RIY>kAxD@(*~MNiXh
z4Sjyzy53%wJ5_WFx^2s2XN+{}LdSol459@V|EOOhE5T=dOON5MxsUgAP`M=NeFGiK
z7+lEC<dt<Mp9Pb-tfB(BpDC2EZM$y<hf|70&>q0nhG%j!p=+#;L4m|T=sBm3vRNZP
zuH<{7&Dd;>YCih5$a?MP^4E23`L^TSY`vc)f&85W7@V=M*OfNuKq0A>zMoTx^E5w(
zu76m%Pm~q&4_hiOfXS3eg-<>05TXMEx~%Oj-!<g_7)=dfnt5I^>brI!%R;+jqy&^_
z(a?aZ$AD^wmQMt=P*QN?VgoMuvt5wtzUnLLJ#kwoT`{NDi3t1JI_+Z0!+g+48)lGf
z#W}AROwyjp`o79W-~S_AMF|^{8TAl9&S4rcmjzt}K?$W&Eg&&8E*@;1>56UaAtjO3
z{6tTKO{&577zXP{x>i3yH#*q*(t&lK0}zoh8Uhi?h)DCqBzs|#4iU`=g(8<;8m={O
zL5r>|BIrYhRZ!W)Lqm1w$d^dF{;j)U?C=8Jb>$W2PHYiDMih)CfhQX%1kp?Do}uyo
zKDC2|@}OFlaF58hFb6{sH)()4z>+z|+yvhCwZBHa`iu6Q!kgv6jL4YtK)8bQlRpvi
zfpX8YmklR~ET~ej6*ic|&^&bIqZ^}9?n^3}-T+sCiU3ZRdr)+JuIY}ugS)wFw|h?d
zX{xlpM{kkcIOd)w)6HDZY_zAfx2KG2D}w|+8RWA{8~mD955LJcJ9}L=sw8y5RviH!
z`?4}UVRUqJ1{%O)*0~IQz|GPUn0egdXYcCD0<8wKPAIA=VX>IDq|16D>oNdOlFJv|
z06vvw0GtZ<K$L~4<9b<HB(8mhj<@IQ6N(zx9u5X|BNY)zL(Z<H5AL)PJVdNo*BxMR
zYET|)&Pd_nBh#Rt+cLs_3xMKHT^*72Nq`77g9ZRol2wrgV1Jl4$SsRoJD_}?w>bq^
zq$HUs!_{w&b|NZj&9wPJHRw11K?VDs&^ozRbI14c+*qM2J>67O#@HAZgZbtGh~nz%
zHC}{usy)s)4}PJW1eh!poczrm*vn|TIfZ{XQzfcp5s{bQfYnQPB3S`KV0jiIa*6jg
z5K`<=N(z<~D8>|~BhGsXOxx8x_0-I<gf2JGRRt@43V7RgMKsY>RHQ2&`+Srf6)OFD
z9oJ-i&4!s4!6;;-JE2QzeykmjG~TkUnsq;+u49IUuj7m};NFwBv?W%(I@4|M?o2y9
zwuI)KT7bnePA!+R>iLm0KqQlzoyc&n><?oF5nRq5l|nz#YTD)gQ;i>Vh!dvt7l26x
zVe^qV>UAd2Weg;_SBgva2FrbLpkk2bOP=oQ%2SkGx}epoIr(V<V3|;vP*Hfnpar4w
z?LT*I5k8ZnK34F*ttBb}pFP1d*XJ3C>JQiPPR}I-w?ZRbxuAAuJQH!;d9G3@ii(QM
z+tVNS0@n3kz<9-a&bRr=lrB&7xF4~;;|%H*F;#%#N+rdo^}vvHZlY(PM@BFQ7F&N&
z9b8vdo9a)sP~W{cdtI0r+B$@&=U!I>m-|Q$zbrgzD2L@6D`+nqE1$>y;-$GJ28+UM
zJ!Rmd+;~k;@phQX<&D$cR4eckK2KI#gs>G4`wU>Tl@A@hgW}6EPKXbdYhc&f;$UpL
z_XVBcA*`9b6zQ^Q-O4l2Im#6k(QY81g--h&Ja^Rnz~M2fKwlc~rjEzhK2d?{q*#{p
za`_B(3SL6A&~*flRnI3+wu0GnDfzgymYkT|SQJp1bLKhuD8(1e{Amm5^BML-ovOf|
zqM*qt?xiV<mZBHb?lOZvsj=RP;-b?x=IO8iyy@ock(Sg}ODG)$kV#O2>q=8*^X>lm
z=8jL7<<C~|Rj5=i7A&Y;-4)njH=Ko4)uQi=8>!Q)!*ih;m}nmok~yXluCsjYZc>u7
ziVRpbPtBSkM7O2D{yA&|z)y_cpgO>hR4ITaE8R*RPIw*Y%g-7A;Tm_N6qec4?T_F+
z{IJ=`1x<J<mv{$b_@n^{*Ymj+4o0QF&@~*&z!G2Hy^r*%7!flWEZ+dI#iE0*N+7{N
z7XZVRdcu{L=u~95gI(3ibXAxPD;A9wCarkKab4Z;xX97UMyL{Y`O$q;&>5=>sKZxQ
zzS-w}>i?WBT1JEI{-`&vkX%+y_uLH@>A5EopkB5l)WPUTQ7hF2P4#K|@(Zn8s6Z@E
z?TD7H^Pr8#)uyjnX2A2H7{O*ao+hc$QUsC_c?Ah|y>03E+rl=<Bg%bR5Rv(iBD|kS
zBnK<9LckP%Z}ssPqJ}OiBGek=){$7MO}dw1qgtsaxz2l1`y^;V3Xlf~`-Dxkabfog
z4{Y~Cpu!2jR9D>9m-!}A7OxU0nn9R^WQ3A>N3#^#)t%vM`+|#fZnqa?t+C8)qdt*~
z&-S!zI{060y2Yms%o<7w^a#}%+xSfRL0EZKv?%+Smd)R$@=YT6d}Zs&m74DEeDAS#
zIW|b^q1mNVOL?puUN?MK{M#%C^-12{Q#zJ`<tjh$+2&=@Cdr|<ka;Ztb|#h3+4dMs
zq-iN*yo0eVsnp-2qjH9*6I<66WfH+q(S}^4^BAaLp&&VGp1l5~EAhok2f4WQ@24$H
z__4@QHM#h-@ep*(^X)FsF$2Jw!Gm;FC-1y%+yrMCI&T>C{I6&dARAfi$`HjFLKoKj
z!!Id9*Nr^~wR+<+Zd&ci+s3yQ2G?^wb(R}rTavWx6mR4$;`i(qplT@i)-bMoJh(j@
zhmCJCU8JnLDp->Gug6dw_=7LJ#SN)Nnx_mX==Lm{RH_ibR+#&&C&#-@V?7{oh5wyP
zKvRA~23-H`Jcj5PNFW{6`<`>&J99f6bW_<qYA^pr*~*<j9eVi2{s?%*#1;F_MB_$D
z%r)D2!9d}}-sz#<RW4Rrc{PR>p|JsIk8%3ZH*_9O{dxN#vn+J;L@h+7&w@npL!T~1
zF5mG!acR9(?o!s@SZJgIPQgRv@p#;uH;Ubg@rmyygP2vp!tHUfchb{VvAWXtABRHz
zF1=q=&~=Y{;iJ5owaJ?mZNh|V0oLJe4tO?e+P!^}c6D&E_&VhPdaLoP`r>WEzkOaf
zQwEpSy0wY>W|o%5NJ&%n+$yBhXd;kj6nW?qf~rP{r$c`xmHKX{>X#aPB}Wy5?<bG=
zZ*cWObDI(J>aq|o^UMw&VoAT#4*O!ubgIp3{hK~l_P*3)$L+^V!CFNJdCZUV&eaQ9
z);Wf@7Qe-y-7{=}IgZy@nPY<OOsmWNy>C2`viJT23nv!2q83o8{3SBm_k47LdQMe@
zsy`A;nN^67ex%^Y`TGXi7*1Gg+{yIXHH+oy%@%vtEYA3!(5X=0fg3v7n|+R-Qejfi
zf8fJ&mbZUj0S8J{CBfF4<F0Gz?zPZB)8lQ(&CVpw=874?v#5?M=ic|jL?HL@JKbuD
zp}5&g#aA;>JJ-Z+am#R82A%bG+GWeV{X_YoT`WM!5?MeDqT@d^tplm*f_!w88Ub%E
zjK<P1d}z58AYx|b+>z8KYQ%3QJ0g7{angq5UXtg1LQV!<n@8}ncqQZ02Pr7nTaHqF
zHBPY=W~`p?>UIB=JyPRt{FtHMipwG`CA4_Vb;~Wfe%xLK#Z|vyDkWy3PIvzR3~x{v
z$j%jVg(BSO0Lj_!3N#1d#K;5t|I@%Q!+$~R$3}3dO>VA*tirv5?FYrcgKj4k3MG{W
zvW*N%*xw%@7yIo&%zFaZ`<$+!>ZVUDbf&b{>-0tsCff)@KVyD-;84=X)r8v%V+pG|
zQhetr^jV&Q4I!Xi{rXSbRkTfMp<8B7YL-?&;;ZvCB+%{%*nrW^C`nQ+PDt7e<m`Uo
zl^G<0WpR!SNy^Q<Lfdk)$}|an=wztdgaBBP<8Xc{3#|KOIY{w5QK<611-F;>b3mbF
zksN`YTcg@tv&jBVd18nrF$^%!X#W^hbi*V(k-NZIV=nv~-hN_j(@y*?8f|~=J?kBz
zA0;INPs?)RjzeccGQf|Ei}oKUC60|}eB)-bK_77J2J!^!uhZ`RtCGU9m~kElby!F%
zD{mmTl0)(WX7|kaXZ^eyjLHP0Z11>cNqgVcez^SC=I<lrakN9%%Jb()sCum@Yu&|-
zC7ez}bQMtmOuJam7?rd+uONS@m(f;4fY6T`%L!ud3yu?TO}KyJ)ah+sQ6@dnfDio&
zu?h$f>wf*6q-vw$mPDxeGo`y9G4imtb<B32lF!rzMD}*K@N)StFh7P^O3Zc;SN<Je
zn(CwXfbt6gK2Jr}uVH(XEaR}&=Ii%QZWP?);JDR4b`wKkL<m^<wxj8{V_L3Gu94Sl
zuk$mG{voOmf8+2SRPL^RFK8LcWk~(c74wUJE;N#L)9`pDuF~^`Zi8emz021y^!R<#
z5~OFlIa^EZb8R=zOT7Q_0NS-5JJ@w4UwuV(Sl7kLGNhZc9*hKzg<v7~rAib}xs*EJ
z`*Nr<`T7YAyBG-0`bqXDU5`_+RoKu9Fs3JcavTf0Na301-{YV*ENd5Ne3@Pm(_nVp
za2o8IW<)a}<jalpbb>>+p?cPrM6pOY<5#w<l!u!$VkRA-781(b9hhUqT?M?RmdSra
zV-h}QVLw_%dR^B9+STuyN7TVA@-EivkEEN7FKfN@_QO~JL^q&zgX<=ag--Y7$>U+^
zp=DXhmS&g!eEI+ij$d@!AT%iSFjcC0-p^+ELM#DB)+KR{lg9n*^HwV?c(+Lqc15Q~
zu$}PtrJ<d8CZP$ZR!24HoklE*rWSebbfDU{DYX9hz#*1XUJnRgem$!XK36?q6lTY^
zY+(AkeZU;GOrh^)XuXpl|6IqD&d#-dHB{5iJ8S{>1ZYG_-WaEC=?t&lx_Rji85SB`
zfprJCb#%AnO9y;8b4dbYV~CHx1QzOp<Glr9oCN7G`Ps23eP2mxp)y7=sQlME#J$ky
zJ|jowt<i#<p_D**uT$GQC<+9pi1LlaSLAMvHfRWlHLJLeMFO(ntD5S1|E(=ejq^u&
zDqn2amP|QSu>Z#k`U_i3a?Pr}W(lmie=+a?iX6X)v%Sdl68~DyqK3BkO^f7hwVH7U
zQ)shi$(3RX?sc3yUt0PmBJ)3pIloz0gd8<$q}JwMU8Z#_hx&?>5nTq`_K&_1>mn$g
zTZ=1sjU^bhif@i81}Cup^?aN}A4!<O!t>S1_v0h!(|ZmAG1Cgn-K;6OR^OuwGTl58
z<Gt)gdf#{qh5sTRGR6`%&bSd%m}+hGJx5o6`v?6jK>njfDA$^rZ`k(wPxRi=dGy!n
zi2L`mKOS29)*(t`g5Hr%R4$b+qCD^>DT+lQL9wVnrTvc8b2f<s*p$bEzR#(8s8v<%
zY~k9f&l!&_@}+I!`QZZ68(kjqw0mnDWtW`aOa9HF?Ajm71OB&1rq7W_OFhAIYOIAK
z9r<X7wuMZf_yW*9$w*l9jfK*npG!FEQTp40wO@%)0{CqYGiN@Q+^o+{jV>(_lJ(9!
zvVF=V5?w@a_-h}7Z<UGBRwBe6r6FZm$7|{7DE_2^5v8R>?u(0sDXH^y7lprW`*46+
zISW6YhFBufh)I4ob^Zdsa~t8i-&v251V38TWI&qIXM@=dxs`+Z6x-AXtUD5<*|$8L
z<UAzM#%`1z@Z|L0567wX;aEqL!LK}-ej+)42WbyIjJk#geC#7ulVj>J$w`}ROj_rE
zoX9`;x6&NXxi2(mo~2W-lyjx4r-x@^NOhOMo7e86$caSt;K>nVhfAc!Dk|V~cz>Q7
zk3kb{P8d(v^4_+~e<rpG6IdW)1|cWy28S<aA#DemsMRuXNwA1q3@mNsYFMqdwd<GH
zINeXmcX-pc1#EWzEzcmN?R<8=)s}qRHGntz^Z5Y8j931!fmOS=TXV@=IRkm@q3zAx
zBLzB>*eH3mrNhJHX0g7YuauK4!I(VBU#}qGn#r!=8tXJ5Y_Yd1QCgmC+oe#C3^;qe
z=pfuUd|dWkYeNMNJ@H>hFB1cSk!IGh3gw=!Jv~w0vX%imf-4PqEP>Xx++_2&r4k7W
zSk*?(6s0kACB=DmPKVt2&f+%Z4xmao(MYdE`T!X1-ODS2`ubrgm8IuiddR=sB{B%U
z%4kA(Xsxn9wEC2f{MpPMyMlIy2q@wC772Ww(8g?G9H*}-^1)yJ7MTQs6ce{gp@n^r
z*<!XmzPJ-V?(U)iQXQ>;WO|wa%fGFb-a%*ODJqpgRxX|9p3Uq2JcgJ6577De17NsM
z$J~>eh5PW|>9}>Pj+UjzXJS`1f4z!?a`}C}TgUsn`2svc2I?Ie7d?j@k9P&3B2Tlc
zMkcN373PC~DQ*lBwrP!HcW)`SZV0F2{;p=npn>cW9I2W6M0NM>-3kGX7U$?evL~g~
zY=2ta`$%cH6Qd3}ZWv(;hx($W^m#I}9U0m5f;RA66;JxgMiBjjvt5WJc1jff-_fx!
z#d2|2ZSBvc`W2<MDyk<1cuPBEMrK2y=D2|P{4to)jfGzB(Zh(pIY5Ww#*s^?Q>~=A
zW$Em!Lw7=+0J8-YIS7(LHt;2tFMZ0p^Qk}8AJU78B*>G_^)RPOGa1jjlc-{SaEEqf
z${^@4_2rEG;ApGES0CB(i`bZdJtb@7Y_;80yQd~%r*|}KnGo{S<nPGdm?v+kNNiCS
z_Q9|38iQS*2mI7VCSry+MMDL2C3lvLgAx&^I=*u^OiWCwotC->v@6drC6J5mKDE1d
z@$CD8%!FJMCl*{fS69Z@$h3m3cD6T@1K8$p_BPARhf!NGLzwf~kni}<|1ft*C%E#C
zk^QNdz_KeXk&uJ;ZX*B=-n})Bwk+qshd-r7Jof%L2a(O=YA&tTg7sMj<v6mPr;lg@
zm|tIQ9z83JgX8ML>+xy-i!7}E6{tVg_QE;Evm`Z{+Be2cL(?fgfwbKroWlu-aFp*T
z--K<>J`}9jP?y(#>rH%W_td*S(SaaOFZT(hc4nIGiloG;dg{0x@}jOm8BzAxaVpK2
zgd{f)DQkifk`sT4D>@pd$FH6H)kkkf4q76do&8S$QBBgi*&mM4mR`oBOLJH8J^$SN
zTO5}xy--|>j^LsYu9m|R4%>V80>&o!$DpYy9*V5)vTd(v+7^j~Gx=@TmOpbjL@kV|
z`#XwrRJ$qvLF-)nibM?m_6~%BtLL`ipL+HJm>bu?xE9ZHx??+=+dIQO3Wvt-^aUi<
zoO(&#snhw}{PFYwd;jqgM`8{k*R5Krn4mD1K`yoT(2i*M$sBA|Fn3EjoX-aIbj68}
z<nFFutWW|6ZE>@OEeScPM%$Z}%d>PsVw%N>_|l(=Fv0-F+0B^i-NHHb!&A+se<$rh
z9sycf-sk6e`7)~VhV7TbnT3qAE<`>)DSt>%15=-qRKoa!XIgo~t3G0l%F!h|+mjPG
z==Q0^p=Cl*1eH8&)XTHv&e9)ZipjY4#d*k=@F;&58Bx*p2TFvYCDe(1dOMi}$^n>b
zuaSIzC&L{PEM#=GGz^_v`eP5o;{IsXe4Pf@<#*Lh4WtQgc$9X25mN>^T$cl;I^Y<I
zvb+eq>Qe#PhTrXO?L>0FG?9)gqdQJCACYI-9ji|yc7<ctvLgJ!J4hYh2CZhH2Zsv+
z59LyL{>2KugN>AdoNQe6H3RQmlK5Q-=rKnm<^?L?&rb>)0+CMs=51?N?%wgtQ-4YU
z2}NFKmwS%8`C#qj<Yd050R!dEPvWG338^|J@Z8q-?p<bDg_c5@xu-@GvWCCAeUu{J
z&%CDLt7pA|2=*84lq--E5EG{=@C3+F$)9~`?^c-w80Eg|su$d0(H<K_6wpz&^Z6~k
z_Kx;zW?W(y=#%y+^Atrcd({}8Og)|u(ca%gp1}P3_K^WN1zH7n%``@iN^nZ>>}*UN
z65+v@j5qAA#uPYq#9Ij&QgIHE{=V&l@K~m=XBtV>EZvp>(M35<I_@ruvIpTx`Mf_X
z%EQB>rbW8m_<~Y=^5tD`wrhX;*=HUU>hxG#GIZ4#k-Eou&2D%7bg=%<C|Bc%)|V?!
zT?Ha&mIL+H;wVx5f1Ldrm1VbyLFi8(H}gu!_P(Zl@o4uaXdNhAZ}gW$4J5RxJVq%?
zRmG9sZ`l3wjxo7O@xy-Q2FgZ4rE_lTjRynOE`82IInqOlfz>iRV1jEew&Ofy5&0wF
zLtQ2;`q8kidC7jXcrv$$c!W-GPepg4tz?|!6XO;EqfLtRa(@`lm`AY17?vWvnw0#O
zU^bP-kh6Fi+mZ)Zi$ejL9dg=~z6I#G)z$7QF{9$RPp_kbUH;hrQ<94O?2Bg$<5J!a
zOqFHF?gg_m#~|M$AkTF+KY!{g+Dw<3n{VOxfj@+wT)vN@GQHMm_<=JOSEqW#bJY{O
z7Tz-)FyT!ko}(n4T;pZ4<OqAo{w*>_-h+SSJEZKo1v;;4+vneJiUK?QVMLmHhm8(m
z2i@i)55rMH_=-HTYrR%mLihhzAK`ZxbtN!YD!j2dr-+|u{2hFk{=_DL{_n0_`8c8x
z15$maFqzt=g11k8_W^#C2c264UbdRpt*qwMg*G&VNA1!FT?h1MQrfdLniVCnuuvH6
zJNG_-COEl2=#N^*aZqBh$7y)Ekrgb}+1Y9RN&Q*Gu8l&$p-M~ND|sVuu|0RwbHeJ0
zgA&u!J(69!#(HNtKT8x(*G7S~n&oWxIRB{|4k}}NanUEB4BYHj%w>|1H|DcT?U(o&
z^#eNmwRd;Ab0lF(hqCzU^A;T+Ay@2CM8LCS&JVy?z#D>NKP8OyfxKnBgelh>*Pcpc
zPgb^H#Gr^c(>nLnw5}k>ho7{=Uw!xe$Pa)L6wg8)2LgwrSt_13%wY3sC$HH34Z06U
z)2T9!E%TJ7-E_ac#^F+!{V!G>0f|>XFH-jmh@5Rr7$c65?{aC|2!RkrvqRTB)P`#+
zk5w?$$ZMBN<RIlx{BEkodm8qIN4A@d3U{f(M*JG<%|p9VwjD?GP2#Ld_n5!ZDlm{3
zQTN-mBM0GgXF#&<9`ZIw7ZpAn_@&UWMZ(Gcfr<R@r4sQJwDtLj){!grkRdS=jp5nd
zpPV?5qQ_18lOp(v(QZ{Y2|k%i6rM>wuK#NZ@Gn#oL0h_}o6*pG=jNT}WTp0|^KvJ5
zcgPHd{d9;dQ7d?Ic%oaAK%SC36L0EWOdW5}2*-(|7ed0+t*=^shY}Z=o7Owk67UN~
zNSEc^F8RJjDm7-;zi_#AGAr=E-S=ZCApQ;=XjPl~rbhFf@%IK5g<pZ#ZgIFS;U3i+
z<K(S02_a8sqPgFhk56S8Hf6aa)tbYh!&ynz6Y8|fSfm~()`%|N#iO)EjeTBqJnwpL
zjJNW4HHwq#vkgqcXD_;IJMMKxp)I;<@99AxI-*nJc;shL;Ryx$oDr0XHM=f;=K!Sa
zGbeb~+BL$WOn9{I*COzzBz6q|5@837&~1A?xK9AmR``>6{m(9=gk<I+I>KEBVF>6)
zhjI(;4Z2rqTH8-iD!Er_0)Ko5CB7*Gbf)AuJ{D00Dx62A{73zmhv|1i-EArL6FC@~
zBN<K=6wL9XU~0Jc%scwXT^6GlKy_QP!Kfi#55`@4#@D|7ZQ#$TfXBLbesgGu6&PnD
z4U<;OLWxAAu@QctPMF6Y9nJ$fFb<qFA~}c}Yn0`cyw%2Gw1ii78ut4+7*L@%a?WOS
zy=+cpi;|W<SFoo;Mo`ryBrw77=B^ZpHh9LGRxs@@fl(P`CeGA_75yO!i)>K#Og_`D
zWVFXB1`VJ~nf0M|jKpt@-$s}lcuQq)YwXzw%zGq=+osYOulg0{r7VOT4wUme_UrYe
zAfJH234jeyT%xJmEIuHW#qj&?r4P1}8Q3{GidqgOKbF$DVIZNtXT(N<3<vjQB|4&T
zk|PqeSeL8`ijG$QE`|4~egsWqPUZM$3}`=E|FyEa7AQQ(9)>+8%NQG8H5uf~ved{p
z|J&NcLI*>&1IDXfKHy#A)Xw2{J1YKc*N7l?U9l4!m;P%5X_G)sNwZetVe1pEzn>nT
zB2R4CdeMb3S8u2)xEOy@oMVr}s9eB+@;fLb3Wu7S`h0!JLKpMs08WqB(5|h=h%<_r
z;Afu6J>o&mDdXay56rj>zvK-09jFpT&kGu2IWsdewds=Ao&QPBp50?HI55%}t|?pT
zl9sulVwc;yXl3lAK)$EveP&SIR2Lz~fI_`etS@;T8fIhVpQDGV>F@n2rVWT;DrLvZ
zv~rqU@wwZFb%lN%;w}_a59j-_zJiHeycp!m(phpX)0Zim?)|Q?L+?RE!qQ(%jG?Wp
zm2=Y^&PT$tf2F(*XMjNzpv5Csi&=czQPtD)EmvZ7jn1Oh?iatHgOuQ$@-vbVUC9wm
zp-daAT<YKFdi_p<cU1C9g_nuz%DZlu551mHu9Wq>X&U=GKio+(Oa#BKzxU>ALGH7b
z=rrI?Iw%;^uErdI9THCRx>-;gKR}ovAocs>j+8UCR)w1RB`F1+7S6o&-dQL2WCr;W
z(2{ZeeiD`0K@a(FS7;mdBo*xbzEuDpM^jfCW2C*fmGnN+9EbACS8whS!5qnTnS;fo
zL-v<mQ7-rbTt-*>E?Ly}`}7W*+-$kJ8>$kEZ-=CUwe$>VIo4}Wu9EL@<siI)t?zud
zQCj{`y@f`OadWb%6U$wnJ@NIk0OVSs)E9(%FrPge#7Y}4BSCwrpR~(k?<bHIJR;gh
z5k~1Tmp`r^Ds&|XC-3l{OdLQ(0u$B|M(j?Dq9bUEQqpMe<afdmPNIv=+F!LDrO5lD
zWzEKx%c1?=Qc2e7zkR!qI|6o5E@wal$32afGbUEm-1E|q1V1FREY?p@Rx_7T0X2nV
zQwC|jWCp4G1DJ6S^DmKj<kB$(uWrR%|H&Ra9{0fP<H!9{13G@8LX;JpIXsgEm~w)`
z%h+Fk_Xi3{5%wB^2itN-b#!!O_PC1WN*beouYw+bfVQjb7|M6-_Exbu`rXa^7YmRG
z3$Ts3>y_8=$VJegfScwwVU42}DY{>S5l>;I{wqZ#n^BL)JKZ<cNeQ<*j2K2D+`tR^
z?F|@YUO`pF#{qRWZrQT#H)7~C{GZ|L*R7*-4Q9P`CObGZ_ockZStyXI504~19)eU-
z*s#B``XPhzyTLw5AQle-Jz8Oja22(icFh<os>oaQ%S13p0wW&l6L?+%IP)mGocwzK
zvMu{Sx3-@<rBaCJed&Zh!$u7PYc(1>?orA8A|YPc)%(3oKA%|pNl;sKa?P6d1^HTR
zEC+rgVFN>7gw_X_>P78u-RNt$1iM%Ox6W!KV;Pmb&7zh;fyQg=zXrr(;1(AW^Gu8g
zQWW-uB`9)&w&$-dx@&2Cp<!W{{grg19svcGthg#q?+W9-VcX1^8e}TH+R=D?pTNcu
zwq*ryRTx;$Pn(qvJ&wzbr~Re8bCB|GJu`ZtiyEkVVW;|HzLVj(tmTd5eM$^M`_Kuy
zN-iOUMteJ`L`zBCbNXc;7<2#~=jva0;8R_Gb*rkUJ(J{6>D|Z|Gs!#E`fYyu!#Ioe
zGE**u6uLm64VhYi{4Z)If&4IdRqVih20ZEH^xjmLAg;iP*AhR35AV?ksP^eTtpwev
z7b<I=DruVS5k7U<zp6;$DZrd$M+p>@o%#CptDj?<=^8gDzu>(g=6ywBq!m0Te{CVE
zQnSGPTjrBM1@HYnzbFf;^AW>Ip-$YA-~|&2RgSS=@`50?M?IgqvXn~V+G++hG6%Fh
z_IXykBcL$aq+6bM3UwT$sjZRb=B?sYWe84^eL5kYj7*oB9RA)lRnArrpdPS4@z^c2
zQNwRDTWZc{6Fr`y3AkQ<_KN?$7kH-Lucjp=ui;9^OVfOBIhyj0<C^6@8qG0-Al>%`
z`(tDioC!qPBv+63a2jpB(WfWc<M{y?KG}~8jkQb>xe4zF(jM8q-d|4p1WI1qrq?Ku
zhz?j_L_#sYU%KB#mJSJtz32zHq?4fB7sfHOS%T9&nOl7F3$ETi6BTnDsJ7Runp1c|
zs@t9|>f7Y`=HSZ&o%{6L4~}}PNi!VY`aW*{m_*bgd7FJrvlu_{3WB8S4m=Slv~hSs
zt<_4+-Ez94`_Ax=#Nl&1&Q-Pc2WjKxo35tvWWU@yxC;WPg%|oX@j=vWf~Jctt{-YS
zL&?dlbr$zI8AJk%H!!^-Y&g|Kt0X{Hxcn#i(_guyoE0$Aj6yINLe2#>Hk?Eoc$i0$
zT4}wzeP5#obs4WM%b*~$RLFS+Q+9Y#ruR!Gk%%gAj93>wJ>i)9ux8NW3lh-eY|Vq2
zf7)OBD3e6TYrq=U(^~5E-x0{D?qeq%sGqX-9~KdS6R^lRw|rTem^{kQAoR*_E#uNY
zTHK_xG&0e(3toCcSkM_NL^bqQZ=VfAq(P?E$!hRALKb%P=+UrLdMAU3@f1h(fc?3b
z=R>s5v`4_<`NCuFTaxdyAKk;daed!j`dPtV($0vP!{^w6g79?Lqy;NRjl`=R+h=Q-
z@dfC!*s?a-aa?#5gM1#d|I%>@NXLJYCD6hi62IsYu$Aw-ni%JL_}+=oU$z86Y>Rq3
z7<sDW!Fwq_i{QbVMABFGZ6N@(Tv?6INYcVRSS+u73M$Uh*JoqAK|L(r^~=ZKM;hU1
z$~wldJ&<mv&}!{oUu*EVh3Q)E?6ZLbyrp|Y)*0TT+~?`xF?WvMDa~(O+H+rY;u(f4
zm7=-&W<^ViZ@RBh?i|11Zb=u!2MkaT<hF0fZM-8oI~8@0QquhmAUR!7$jq-lN}*Yt
zmO5aM)ARhd;8qusZ(RJ(LwIpFFs*w6mn6-BhL*_PqAv&o{HZD=3d)6RhYDZOxhune
ziIsRhFm*vf2b#omRp3nAE;iCDp8jT5H`;^8^Y^(Q6pT=SYl$thiapPMHZv8c$NRS}
zAc(*g=I&RJ;>C%^B;;>uOs&3(lUC<z-0e|;;PpI+e<O#(H8$6C^iz~mtFANscXc0;
zkh+QB9d?psICzK|9Z+3MY^it<!BBkozkNLOLd+Zvr7Tz}vQ4d~<@)Gm`di}z2LKS1
z4k(6*AVy(lE=sraYbdvHj?xD71iSr`oOse1PU}LK8k6=D%E0hzB$H2g?~m73KsXaj
zE2exE;R+SFo9`@bMdZ7ykZ!R{^pb&3$?j}9h8Wh_wc%8)H*CkHz9@J4lJ1S&KS3^v
z`T2p;r<&CEcA|)Lu;%BswEpc2MIX7&iAnC28e_;|r6D$>5&Ffg7$}Pr$$i0n#u*ut
z0q8+6UOk2vcWtMyo+s1Lq)49dE4@R{;AmP@y1bK~FVG0>y|9j>w%ZiOY%`V}2xnEI
z@Fn<d;D!VyO|46j4HYj$&v5SPJMtTR{j}OyAqU~sx1&GWl(Ws+e?zEA0!2-ZTQ~JV
z&h2(bEdp%x<~sHP*>pW&f&tx#`r2{rqF5786U!5OuLqPZQw8$<jL7!~o`{leRS*kB
z`jpz@9SNPlG9klQC%M5rBMLg)o^g(@?UO3$7F#POk9&Q#flWIUM<m16uisOY_0Ys)
zK*kt2ZAhH@p)N5kgQVGqh?@vUM#rIKYwE|%$Jd9_x<4ub&LfN9Q){!s=O}`e+oP5A
zb%!Xzo~aC9Q^I|v5?(}iMA0a%ipE_@?t7Z#h2GiHeCKwVPW71t=jDR1;+z4YTT}15
z&^p!U)0Irq9VTZiCTrPb?Y%01m;$}JNd|)7_hhGufE#?#q_~w!A7>)X$9t>!&BHyR
z0@5t4(PieN@0|X_ymV6E5*!HihCsmPTj2CDrFV%>z8|k#I@0w;yrc2ZiTN+nqNix8
zdh|1C+-1|wou6bms(E3i^yvCh#kvEyAc*0_GJMRZwBQLpt}5rzb?arN{4E)SQX;`+
zt>b#`?h}rOuh545gkz#>anNf_YNFwNOv3yesWhLR#qm?^{G@U?97G4Z9FjYLa+Bbx
zc6D3oAF`L6gH3!000%>J5_Z{tZ(Q@itNvX0an|5cUH9QlKJB?#&gVQvzFOMs`Ba-9
z;(GHZ94n)P*BUc*O^7`-K4xZeh(#A}y{=4^*@hp=gS3`4{)#ib-+5GJt?d(Bb14Rz
zz|66xy;3bH6BN|0yR9gVck<~-s`7#u3Hb=8rM|~WJZ$K5*x>=;cGWOiJ@Y$5ZZXn_
z#V$K_#TGM<vIIt1w;Ej7xOCB(X`qC$X2FbiHR8$o74ydx7-g1o0{%8$=nyIR;5#NK
ziP~6Pj>>lzlV?&IV%gL8G7Vi|j-|Rku9#@3tUmx=d+n@*FR}^e(#x%%S3~fb)@S)g
zeF(GyYiu6HEr|-+d(AC~chHSW1PUH^%3}${zjVrWs_i+;`50D+o@fM!e~*$rt0qAr
zgkao_<3fF?4#wy8^<Hm(rO9%~5B8cG5ISk8%o)A;h?m6*Gooy>+$1b~qF+q*@GDbA
z_MHO@U%oGiNE9ZvGxl_ET(vhNc%Xad$7YOP1XJC~HO(S^e@w>#K?@n(R~6KFh_ehR
z2tOjIFBtmbXZvXm^bL%3V*c}(6rAyKfW8)7p0$tTlxf#|?h|!R{;Qd<*M*+Q9@^rS
zBP|sQULWaatmLUOeVu=a_2+Z?>NPjlA1XwXH{5KgF&=eY?Y{f>d3L~^N}S~Iv9TL=
zsd0i*r#ueqE`&_RS-z;4>>E9YH+WNbz&EO$O=XmO_-b&aZ{S<Wh6(}m5AsimXT<bB
zTzAa-Tr(YJ*@NnS`bHEVGMlJeP=+dh2&}n3@f+TGvNh&Tp-%loF*@SYJK7Jy*7985
zvE><K(w5{KN^s(e!28u^11lQ7+2|DEEaSjd@k=wVvg;)ZQO^{zrAtlw$P^32hR0#^
z&YM$3v%W8C6!ehbF(l$0Y`qNePE6_?yq`txLh&X8(Z%TL5U-IH|6OeBZH7LlkF}IP
ztJC`nU){Db7ZF9kU?yCSF1}T!qQPM*Y}v+PxK}nVGQqxi`bncy7&SQNbqYGWeQ7Mw
ztBo>Qn&(p*AaI0u4p2e8Xs$_O&v;Cwlx=(iRb9%&LRL-sXfMo-W%+8D5HS?q+uY)4
zc#%E;JQ>TB@qIm!C{tQ)d~Qsr*lGIB8+``ZWOz=sZElz{TT?e#$+@X1#rKI2!jAY3
z5chLe1Y+s^Y@mAjnSxL)j+L@F`(MeKoGwc2Mh-fwSj8SHV4;?7I|4LiKH62NN5G)W
zbIw=2?~KzM-zpo+8a5K4c33ClhWu164yxC#$?@<eQ@bsuU|FUDY`IIl&mdl$a0~z?
zJO139-&~fNa}UgR=96qB-3{I<ypidZ7B=k<rgG|ZhV?}$uO2{&5OncOmyiruloWpM
zmJJ#ekqC6q5T*0w<A~3ZEa>jC7$2|iahu2`a^v7AABE5>BA~vg*d*%-dSe;FF$pV!
zvuQOh`7#fF;c5G5a889}$HdZQSvG%+PrnT%^+l0rkNMp7NZ8h!8eF&MN^C|9@^}o2
zx|2S#vXvG>W8;VI7_;f2!Wv2y3hc|zk$`<~vdeV~q+Xp06?Y<H-IeJBuq|2iptn`S
z(0E5k`ez()yya=|;i<Xxo~!P?U)VFq7NXO?cM>te<<ImviJ~?%(>kj%+bP60=dW_b
z)!qLHUWCL4Mj@CuI+X;edl&Zp=+W0s7dqV;vi=Kr@GAAA*6vbBQuz8Sl*-(KVXjVL
z$;L=g05iUr<YD+6^UPOKjWZ6%-5T_1Loc58mC7QANsu_)?OSkr)joHp?nEIgOZ*Qg
z4iZrU<ihwzM1}$?dDM0~7o9dtrOY4S58H`N#6!>#RvdOXge{;L%c<6U7kH^p6b^kK
z4bR}-T4o8OW+=AE8^0$6I#bIJBlr%rHt^`>w$_|16cp`Ff}94qC|Kj>i*~>Q0wz$i
zRTQ%8RT*4#m3U2<Lqxcjk;PM~mJSrY$_(~aexF-<bM-O<<nr)L^JYeuA`Gdott3kG
zF|g)WppegSa6NA3Mwle6y0kx<e&){jSqlIyz+tX~h4g$LAb@c@i6XlAay?2~ZCF0=
z3hztXyghu8=L_`sd6VFF{ci7&qeaS|w1AxGW2nQN9die(R}yO`6asn(K~I8Q8e!PL
zHE29OK3%%fW<guSW<_|t7O=<92563i(5=<ez(qlcnKcpRi;I`^o0E2(1U{){t9l@Z
zTv}{uVR57iF5?f=qz7Qm{*va5X)ybG&|q|c>AGc0zcv7p1t8N4m8*&-IdCG2S}epp
zx<I*r1wwuFe@f(oNl?5IOv%YixA3(nTwB`LDfAOZnN7L5em1QO&#Ald)Kr`Wyek_m
zOeLU2rapYJQkKGf#{8>W*L~zcoSuND?^E7PZm#QduqjODhrA8-Fou~~fREYI(w!Ss
zIEMEzlaL|OV=ogYzJ*#iU#sM<&lc_BHN!%36jU1pE<2BJ4u4P7w+o3|qPS<0bL{}?
z75=)+oC!scmgCR-H6gxi#ykjuEnWg_*kJUzqa@<R$mY>TpRBJV5`5=>HJy=)a3=<i
zdql%&T1BQ>ir4%>7f|v{Mv}gQ<JOOF(Pu}q&POt{Fl$lUT|WSB_I%5|lv0&d$Z=&|
z!`b4*XTB~-uo$3@V!uz&H4y#hU;=OYCD(5?ZirC}xxENcl3^yVayVE8u960U$4T{d
zHGmQGLqE*Qb(3otxp76S;RT*3JwTKnk&XG%ZIOvi=UXj0G<~v19@E%mfTmZ@*ZETd
zgW?Sfq`tzJ5NkTPq{XKp-_F_dS$|g<0ekmVn&-Z=e=Y__)*W?pMChIGd5Y+BD{7oH
zBwN=n0;#qqw=@f}u~p-=AM)RF`AT>fs&yGbU6q!=L>?Fm)eDzZ7BTp(<qHzDGQ2no
zJ0U)RFxMBh2e;5__jH%6=YKT&1e$Zihf!<8-8{F;PE1EJHL3+{(Yqr#k0oBe4xi)n
z3tYeCXxUud8Kt`TDuhUXzT^-g<e5o;d{KP=bQ-`B$+fq`;$`+qAJ%pA3Gi3UiHwLX
zN7K+7nQ7(K7Sh5xe02BG{Dh`27i!|pW@Oda7DzjPhYaVsM4xDA8qvqDb^PATKS&fI
z3waT$2!3K|1Z4_4Cf(**Tf>4*B5CTvEbo=IJ!<I?ESs3y-C)Y7Zqe+WWd<ztGe;$=
zM<sV%;%oeoV9h`iVScoO{yw@6XM)PWZS-dP^T`ZKLVGts6)1+DB?l*{x2ER?CqQCq
zJW)M@CB}2p;vHEBG%|Lx9R?Znsf$Eh^Ia9=g`J(bXVOkRCPwPY!ItWstXw`<m9RIC
zFch1}@mQY&r$(6FTj?Sp7^da5o~Xuv6C^;S9-K`U4@y@@G%)$p<ERy?P<J)?l?*Dq
zUB@LJj!c*4f%9tRfQEHPT<~k?=Q8M_obL}f6lzU9;!x8b{Y&b&VinCmP(xwG0hCWD
zHEn7f4-4Qypj$ukgKhyqL4k1qZaApbcj$TIGY-O!wAIaJT&(kT)ep95R(7*bBD6Nk
zbeaNGyk-+a1{@yfT}SxBA`C2CRZ5@eoXz5r;5&;k^<b4PxOB?UW+VQEFC08!!cbmO
zrUvBrrL6A+9zu?M+u<9!{Y(^^R1Fe@Ofx2CqGRY&XUeKB?Bp|lF2@=RJke8Nw6bk8
zKY}yNH8Lm#G2?d9-A9hWCrBi+s<hoV3g|U_CZx0=oD`A{GAjMYS%70HCw^o8cFnWK
zjm}8^+~OUg9bdwng!T?dt8ubyFZK>K<1iuxFg8>Gp+NJT;*iOm^R>f=9x@k~4x;b?
zi<2lUM;$A<Mp?!S-;Sj`DTz|Y^8$;o1J_tS*M<iQW%1;1jNYy(-d%lF*`rc*a$>i3
z=17I7UvT3U*JT7x=Ii2O!O$fdRWMt1n=xbdqhZg90@wDqdI@T>l$l4OV3>!nh-F2;
zYN4JFlMRgOkAga5C8!DnKeLVl7(~z<%gM=kYhU^;I&j=MV3BQZ(R8;G=}3`b$XF_=
z7TpW!V-aWVZJ>|*2Q6lG2Y#g58t{FICtJDF-bQm8MIs)7whJ3xU|S!3H$t=Q(oxk2
z98y0c$hEacMAd;s0^mZ>6c0Ar3fJoe$hNY6lHH!qW&yHpkZth8dfKPnQh5%J)2bL{
zSu@yJv=CF>qS9Mrlo)78z3L3!@(SszH^0iUM24$yPC$^3-j5XcDi+?m#(-(Jw!7vn
z52wTJiWEvRGYa4F0r7)0{)lj&k#*^L1!cFP^t3IaO>)r6%%?!b;R)D$XyoQ?KXSpu
zPcbS;`zKKECW4>hS4w-=zZ^-BUppdq5>){;A}78P9{~qqKAV)ZX(5(Zaw{Q=agyZA
zF}>}86%ht=%Zkv+$+0c34{Y~U1|Ir2Dy>~Z%x7-q+1H7afj*~diZvnsStg2(m&&5t
z<bRly_wqO|$uCcWjDf?x+=#fu*Ci+*mO0Nbsqg^LbQiJQ;LTZEuWudN)9F@Ch*4f5
z&7y^`2%Z_`Lt97EyM`UX)`ReVR@mfcx(aRcnlUm7JWc}aF2%QprvXxrL3qcMeA7AG
z$TL{<<jQ}El7x`#f|B}M`}TV6dgW|EFh)T2ut-GY018IXW%<zXn7g@2@tRMoiMt4C
zDV03ZA_Tas3IbD}M$P)gjYl68_3Ax!@ffu4M?YfR^kf0h)tkg`Qd2p`JMEjoIM+hr
zKE?VyJ+jRR2I&}`hUarL4N!fawB<P;AQXJthdh|$s`x1z@cGy+d`u%OfO#fl(0cU~
ziICvpiM8UOvIjQ69shusG1cw|m_f=a*QmP0lG|4pAS-=W)Ofq0A}5VPsm#{a8(3RA
za_F3EZzlxL2wWZy8vv$CIyy-4RCgQ$RJ@gBo$;tKWYIV`{==eVq&9%*mNoF}0Jxsi
zbdnqN*$~DMc76*nIvz0QU?$!q^F#o1hHq3ew&ts9TW>}zRE69((21{?fk9x|PFS74
zI)C=L8g%h=Ik_SPkHck~;SW{Uc66Pmpfz5Rv&-PI76dYR_BasRhu`{vVjyD<z?u!E
z`OB=`qVnqi>=9;)DK`L7VGss5n;NRar)r=H#uB}tSfH;*#B@K2*3%f|Pw>8KH<a$U
zbJ^q`7kH(g73z!T;^fl{UwU3z9UcRg<RGvl6vUzNz`LS8;GOIhuuXs28OpHc1&YGf
zTswFI0Osm_u{M-1Z4R(JY&x<SFc(?|Zl3qy!V@#s#eqyq4?S93KyHvDro+a`+gg3D
z*iRG`rBwlAaQ{HMXa9dtUm^C)*-`Jn>xxAsI<F&Y2b~jQ?XNIkf)r*@eyZO3q!$Ve
z1!mqY%p?Jktg`N>71@b_FfXpKb?<;hUsE~EbOS!IWov`9Oz-8f=r3m5RV<_keEE^0
z$gu~|)E~MNzv*JLyXstezA5#!ZIOz*hRzn9vR>0ud22E5!a@<r0)&2=IpbbSNGZ{w
z$U6;Z7@pt9jLpT|g{j%sro6!`SrRI7LDPEWu$?%4A9)+liE;5R3usYGkYXU<-I*g*
z6IlUvvhqATsFt$`-8s>;zEueDK>`($KsDs!6v65h4kHg~_a_)6B61Y2>C_z;Ju&NK
zzIc`6@InwLK0n|?7i7^l;FZ3>=97ZRCrFT8WzjLPgs$;w6058lM*ZUO^WqBi3({gs
zttyij7kL7d{jLBpaX=lV610CYVx@C#URLUo(+ry>2D1-%8x_2+!&Gz9KD83pjq?lS
zfPmizHShBBazGbAHFfm|V`5YMq6Q<9t81QdbhtoBfGsSfo5{=9=}fQThQKGbD&~FE
zP)d&0;;;|isQt0zcFo{|X7UX?jSfhko)YKP>qSSLwVn`udRG`96mcP>WbDl>z(j=H
zaR9S2`ch$&HuyZCiZF{X1ydG5oMlIacDqXG7Nn(^IWA==_k*sn_%n1u^p%AuQ>NvB
z2-k0Mu4D?&=%9`A!86V6K4_%5r-bp^IrFhX39|>7O*YPZ&;=-lCnBP2)tGI6m_Udv
zmurr%)bk|4IgBq?ezEouiryZ<fNU;jp$}1+rykWD5#PKgp<jWq`ekb9=zH}pcKxkH
zd2DqY6(ENy2bFB4lZJK>3%yt^H3yo=tO%|cCg4gGQrTKx!ZdZn^Y07k5u}X<>tvOr
z&c_Gc0e4TrHyLIsWVKIrUx*#KfLb>V_1SJjBh+A1CCuaWb&!pkoiEj+;V>hvfLf(R
zR+-SmGBL+a6{_>Z(4yl7B~dmSBh2;#==vI@P)#9OGvT(WY%OQvsRRKFm!Z&Y*b{B)
z5q5>mDEE7M1p$U@B$N)#U0aSHTfcC5T3(L@Qu`}~1_RjyiZ`NHg~;L2JfqSQYvk4z
z+if^7L``5HQEGQjOzidvd7#6%9G1OBYuAH{-_<7>97-Q*#DA!L{{&_~{hTvw7lcec
zIu1lXb0z_pJER+s+o9ILX2?CL;z4@Xpk-L8Nac-wgz1%dKD#fG!uTGCGAGqsYgSV4
z0W{Pr$!WPIn=_%JoWPYoN*K_iFo%R=`i{);XuAt0j(S^%I4@6LtPVbp=M2`c0U?#>
z2)p?I*!t?QD7Wry89@h7Iuwr5oq~YU(%p#CA|a)Klz_lkAW9A0iZn<!0wV?uN=U;<
zibzN|-+pFL;r)I8oXhJl^URL5@3q#w?zJC?3#2f4HQlE9>y5*LS`6_!nW?7K)8?2)
zM?@uj*N|q@)W0niQtr~ke3=B!X&j`^=(x}(5P<43TLqPH)nGSW4*mK|lG(ETA1%Nm
zO}hJ|IK-cs`7V)4gEAo~AOyBT<v}3m53x^lcj(wHVHHvy<kA>u#&Ez|a3Hj$D4(X_
zHthRWR74_H$O_!Hf%!a3=;Ec~`d#juPUVvUjTQLKn7k#3J}hFcL8N`Z%zC<iUcNt+
z2OwoC_!&ko{*Mp>s&HVcdEA?MlXP5GMC+J{fxQ|<XKaYPRO75}-)JH~mUI?)oMjlT
z%{>Xu$U2ARZkG9rTPR%YLWlWFF%85icK)}K3xCQ8QaK835bTl`4M22SU;>D&M)oBe
zByZ<y9{k>sA$notyNo%BzmO;{<d{RTyoqCy$3oWyV~~4illAu$XN!wjud+>vDt0;l
zg&-fo%2q19bpziCz~|TUjJ+rU@_#u<h*#Pf;0mo5=p>}gyukFJq>Y9}ieBJ79K+!9
zQy`>bUGHnSf2ef$lmyxWe7>$~QB<8qJRuxDh~U|=YPl>sZ5HN`@}w^v(@-Ey3iP+8
zT{p>XLTYLQhl+u!ClacMJSX;_hkVr)0?;*f@VG{^;=%`xx_g`<HcAkHq@zL@B3tkN
zygOGZ6a|X}F^CD7=bdkYY}0MlvDv0N)-iX-86hwQ_`!&_o7*TZxVRI%XG+sDKMWFt
z-;lVnV1<zfI+Pq30(^caSd?SJ&V7IbsUXiue+%>7<Z#!LxRK$tx{ZSCs;Y2&@bM-0
zLz;6x=zybpfc3&7GVp%^I?7Xc>fb$B`B4yBn;qzlnTRx)2>zD=i1kxS#e}?kbJ@g`
znKcfEWMUY3PT+GCYfLof>cNc~%dNdj&1z}@Rl`xA2aG~pi0e5^td(n`Fj)vV67=+f
z?X-J_3&W2MiGCsgg@A+_dgEqPxWGhTrNP9W>Bfzc^L(MGO5uFI_#gZ1^TMkjZZJia
zeTKPl##9yAH+Qu^UsSyY$dhPPMzs3W-W$av>x}bAePa7dcMoIN0@!}mqhej)*AEm|
zs<~ePbUD0c0S<9!%tq!eD?Q4H(<f8nX*MrDx{Vx(h7@$t+PJ1}TW<w;_#(!NtKB|v
z$qz)S1Ka-wBQ{wYzER=ytW}E7cJW3GgE+KYp#7q}^;fQ9ipY8pb4(g7%ogOhQJ-})
z1;JmnWU&29`wn9q0=XiyACR1?=K=sf2+RRroUv~i!O(*<(9#Qaom_d!#sy490m?vA
z-?H7zy3ed0MG?u0UvubwuHbm^T+qJxU^;x}*UD$~!lWHIB{=|-vy#tLL1;#S6~S1P
zYOHFpwXQmw&c{5G&13|@)phfz8qt<VP6MD5S(9kpIv?b3s|4as-*vrDnXcnCcSgHv
zd9h%nW+eA;{8-QAeUH80I6}L4F%5ha2ZVLvr!lq>PI}I>EamBSo8oN0U0hVrt~aNE
zqAUsILH+BeyHq0Fo?LVP1jSr@gWnv$!F*1F;T)|SmbmWPAukJ%rn5t<Pao!v@U1#n
z`D#Yzr#;B<BlKuI0XSUrlV=>>HiAq)t!^b*vZ>HSeaY#8P$xO^-F<VoKBz3$7`!!%
zdJ3h$=U$>FV#G~<U-#Q`jcX6P{D6PM1v(GFlrNFp7#QmtbH#AY3USW?5*`3vlOhx&
zo|opvwae&BnVeoLzVIRYg9T}P82A=y2(*MOo5c>8R2kpxl~BF=;0!wyhp=P$2lKe|
z*6UZ5{lp@L4)uMASbPWPn;$;L-CleSUbX|at3U=DK9U5!K{=mwn`j_i4_!ulkMxu4
zGIjoS4LbM1Af>)uHDEIBYv5rt!<22Sp$*C$HU6#huL+Zr$PBN1u+(M7docC{4+F;}
zf#uTIGZz1QI9hr0YlyNahMxA339^C01Znzpn0|1kS~*G+X>Ts=Ci9C>8?4nhFaJ`l
zW_{!qpA7K=1p1d2o;HEtBb~bcviDp~an}sXL2;UMx~AMAX{^FqSghv$Ib5)@&Ul#_
zI5>bs-~xVCH3(q=zEC6ZZIVlO7H)kC7<Nmo(m=X;8W%@^!oLMqp~<oka>2E8_^Pmh
zubH>EH*dKk97HDJWS;0$9y+*85|eFpT8q+Cro$fgbZXBItITDB`Yt+Le3E`jU}80?
zAB8ZxRz!)Z3_YnD{!Qgorq-yO6XJOO5%8-J`}mf2lS0qD+zxUu+V~1YU@f1aba^!O
z@6smOx(;0m6cs-I_729z{6XKuHp{{ESHVUz<H8F!fVyp&5g-lUOKuFtg9$Ow(6nq2
zc#WnIzEdcsiLhgyr;XW#{4f%}`+m~i*zm%W^<2{WA@{Ky{oG;}<%SGkx_1yeHWF7Q
zZ(Er~&)x@`gCE6_m=H^vehiio5>yv%^WXm4+KMS}+i$Wb5W9d_YYAn_#rp9Z=*KqS
z=qM)NeU{PZD5R+xrJE3y183|z;jTv4$bzrU=U)c)gQ6m5{gkgmpg(GQgp%@${8-Ta
z;06ZKNzPTHbXmo$8rq78&pyOo0ArTHUm_ooX)eap{G}a$vVuk-CvWC4(jY-^z8KNB
zJAYwF1{eI82bGfq!%B_G`;ldIj6F5qLmB}HdA}UeQVo3ZzZ{e&2x{r(usnYlE6)9g
z?jTI#QPZ`lW+XWccFXw+ivR|hB8^{dzqAdw@hw+YH>%l=IpxN(7GygN@J)N9W0-Ai
z^MW?(eVaZvkpo{ip)=s<p(5s4YR9BJCv?s0-OI|9vW8C*u#<wcA=CRMil0No7E8Oi
zE4p7)fkg-cyOzV==78C?u*;ZkApMZ?=GOrcEVQ1V#&DGbSl+TdgsNO9wxo-ZtNv-h
z3vs8nR(0DnBRAn6_0ZYTnVe>{5w^l~=j49dp<^#H6=BZRILDIe+?uA^X2+(%`1zC`
zII}8nb;TVK!Wc_rp@<c*Qu)N?Z5a|dfJ^9pCX2PnV>haAFKWrSUsMP-WLe#4HIiTG
z{p6ju+Z=)K*SPqJOm^5;ihj@Uf{RhV{v%$y8Mhu}t)hcX<H7AgnNhlXU&sF2nENP_
zP7fmm>L$wHCmFFJ#4eu3m5H8*J1ztDgcKUC1Go_(k*;dH315Ld5;#Quy0%d$=IS<9
zV!8DUnE@sC(r8WC==l53UybJ4TtigZRL1*8#4lim4v#cN^F+k6wf$F-x2mn=FYSf5
z`P|v7GkN-OW>6ct1IMBm_k`rN*K|`SQ!)fDu-N+G-!2jYVuya-S<(op|KcSa@&g{f
z+_SD94PqM{8hSwN)1&c19G+&3-d9Z?6N{KXc$DUcyv#fGUk|o%jvfh{-rj#w5W3T+
z^o)rD#sQIXk7P4|9|GRe2iFc5<yz#%#ka<uk!>LaY52T*Nof)j-H+1XXAEB{7AvZK
zXN?iAI-LL`7Ix#gxt+#Lw`w;GupESkXn>9R&CS-aHy@`hGu~|ngcS4#dVUI=lJyY*
z37d=Cy|Uf`E@B%1#=n3-|LjPJ7#2U-<580twu>F?J_W->_*08zD5B_{_s9#}vm^3^
z4GQoxswk0Ghmx9PxTU5SI&`|Ys&V0_OY7|qhS5+j8G@0}wS837B{IOK()O7p`HOcF
zL)b0#2d^W3S5wa?HU+9pyMu)>766g^tsp#8`}7n$KsjAVb8UI!n|D?f(HXUXXBRIz
zP`xu)J~>6h)qVG?nnn&YLFMkH2npUTau<n<hD$eC&m{jcEOxz5wgW|1T`WtePe#ON
zT|^a^JC|msT$3amG1Ee_gP}GIH;x_zZ%9Ubba$1Z4+1fIswbx<hNX}ORCDiHI`5d^
zrk8~G;~uu})(Xt?4Io&9-E6tXjG`2Om2yvcsaS_9@6*O4yACihaaTM5I$MOnuisFE
z7tlhV74e*2nkoNQLwqK6l~j3GR8*Ty4WraTL=wI`r^H4lbbt6EhsP^Pm^>+(txNuv
zH90U$aB5R|v_`Zl{7CxyjbAw<9P+sS<ElX1>P4~To!Z|{z9T|!Vv6T}Cb6`4(nxlo
zTfkf6s^4JiBOA=62c{bqmmtTM3Zw1tv!KJ9G7Fncao^nugX(^m_(2bGF@grNZtiTn
z*{8rYfRt;Bx#<x2KjKMxTJK%6x=KsS`Ja7$(E^Rl;H6TUO`_cTsM&kW^Q6CNY(q9q
zD<`n{X%HlbHBhU2aYULDq{X;o#3(eG8IDwsM_p$63SF4e&+pxe-vuP$*`#N&6C+Uy
z^-p-{HvZtYUv%>4%kt=h+l!{CiM-a;(6LTt2?P&k183NC3X)P`auO0PExAwRsTQyC
zA8}<hX$#m!P-&+!wx_k15>|6@S8ppKrTb0>H5BJ%u_tp{8s@MUvP3+h4@w-=iEmpO
zEBwhw5zl&mvvzT3-p%<f@8XX4!p;h9!Hea~%4>Kxl=R0EpJNx~qanJCXL0{@x`R!W
zf>|O`qAbg$B1wIpvDqVPs6(JutLw#nF^a~>qV~!+b}K%cIjjBC%EV1;r<)h=mo%!^
zy*Hs%D}S#>JLT1ub~;1N`_}@U&(vwnNQqOYA(X#F%FbQur>N<;?;jV;lNxWK^8$!b
zFgsDW2dj;9@Or;q|CtsX#fvypIuqLC7z+WbmsN>+y)xa--g_^3Dm^?Y<HlZ|c9f~P
zHDBZwH-}K`hzD}OHOfb&#X+U{-Tu9U0sRz#l<{H?m)_RP@-#cGb8D*yDcPnAy3DyT
z_CcTbcXT@?*8C1Oqz)E$#MDxkWRVy5TFr=6mornQbefH6Exp>NLNg=1@+Q+28>5vp
z_dHZ3cvl)fM=P^vIx1~7JwvKL@Ocp|Dd}dp{VXNlxXPAnz>^Y;QyOgxdrG;nVE=~6
zCb~`Xj!zU{%Af5%N6v3lWfgE*Lwu?9WW@>Kg{COY8T3sxHLz|U*gDpG!Ec~{=GNXF
z`@Xgq9@P={OC3j%<<I3IJsRkM?(()P*7Y4G7jsQ;S(!k`6qU~J8eN)0Ac4#pT2XP?
zSB0zjCbf458`bu;m1|C7qA%z#4zlKNf9l?AO)P2jN&^mC9LAm2GSS3wxD%M|0RO=u
z$TPhz-k|FnV%dmyE>!98&m@KZs`W9uJnaLY%WxXOttOzKgMl(Gm+^l(%2A%U1geW;
z+SzQH(XP+b3NChviT~BOi3gxQTwfLu;kmO*DFe3qDwspW%7o+L3SdYP0GHC>CccG{
zK1Ib&;gtxMg!4+=7FcWXmtMsaR`T*@O{lxNt$)WKyo%$7Qg26c^Z(2tf}%2eE@L~%
zro*?Ot;1<HI6^$64-4hs@iS!o4%Ul~cJ`G-AUTeWJwcfLyz2L#uKhp1bQbU1dY<>J
zlF858ch|KeOs{`ne|f0HK}^D5#;)#XIBdsHLRWT7k%JADRMo(u=HCs?DnvQ928_kJ
zByntcGxr*J{@3_d`Orzh9yide?s(Qf2CU8y8%oA*=bh&tS<zuQ11_BQn7o*{^)Zr7
z^3HMzOKob)#Q8skD7d&0%1t3dVvp|&@K>AQJO_^jvHh9kp;6_qJIxipHt0=vaKK-~
zQRd4UzoYQiV1}s7Im#TFk|FEmPTRb4IP*W2gv(U;vrAw1C*9cEL+jgP(V79L$zKAU
z1=dhd*Kvnk?d}_Gx1zz%mxG^o913JR9QdCl;N*6)&#yFkSSvZ55vt^6`?S#+llMYD
zp~Z{i(H(G8CF(`Q{c&!T3};7$>Yob;u!cbIgpgVMqAegM+f{`82@6#f|6?cS_Au`B
zO=Eg~A_y;}$N!3Du|EvnytMmkT+F7uUto>SD!ZFytiXru{vkmDepwJlq$DDKkS0Ix
zChQawNhL4<6knj{pQ&LMK-w7h`b@w~wa>5I*ZbOCtCg#%7@8US<B|c6hg5b|?Qi=)
zF!vHL_utuvw)^KWXT_oh^UHJ<-Qyo58r!INnS46s%zt=9=}DeRH=Dd|pP$Wb2$Utj
zZGcWvhe_C<iY#^!1A|d-tH$rTt%=G`(}g?q^;5k0+uUF1h|s?8?V;=aVWWJHU6USk
z<WcD~SJ}s7V(P#9NJ)L8#Nh|e8*7``*>lQnlypre{;VLq8iK8TH%Z+wSmfFHP(8Sw
zng3J0wf4B<m#lnLLEBJ|UA6xlot@eA%T>7`Q2$6b6FIeKgNkUi?X_)1a;Zlmq=BJe
zAvcFv0O9e+0KmTxIRtqf#GP(@NlYczmBsYju6+3Ocf=jILu6)1OX=e8k~<$+SYT1m
zLAq+Bad^%S7t?}6tBLrbR&l7ZD04I($KQ^uNCR_O6|ae52QU2xp;MP}Q(1r&8E!zO
z)guFV#R{|cT6A>cO8UO6%QCMML2~A=tTdHtDYGZgRIGyOp@%D1*nfy-{KByKM7n!_
z&nd<xiYT3`D6gNX?^O}i?Gj^2w=e=K{69D`g%_0F)Aas+XVRYZl;R~E#1g=yrygwB
zO3&Yc*l{`yExYlR>Wpv*B;D;A6*~LjZx0Q<1}uLlGq~-F0|f6cba|@(k0-=FhaTKv
z&to=bY`~c`pFyQ8NTwAR^mia@cxqDlev?ISrd`_~6mNfs3;h4XqjWpo!n@55TiOaW
ziEGw2N?wlbCrSTqn9?Kv*&+4N%kTQC^CF*AyULlX`9GydIeBbN=@HO4HkKnZfuE8^
znj)oP|Jvw;DEYh7)g5O;doSaDAk9hzM!{(zh`j>p@9{_nl1%l*?TVN@*JsV6yd3(i
z;%)zS{R^AGz^eQedT=DDbd)QtfE48qDuFa074O@0fSEXB66aQb3Mwh-&Y70K`-Q--
zwD-O5y`6)yAU%yJCM*o&6yv}IA%E8C1nd@kxIEQUD<@5mC5hhF>s+=HHtxe5?K23T
z>s<*6RB3)s&cK|hz{#8$N<Su4XVySx<R|<zd)IBcGPzxu*2hYOQSf150~Zjvmsz|Y
z>XzS9$DJTQk8lPiMMF(a{1<jH1Gz=`@}W_k_SX3YQIm|31`Sm}${gAX;w@YpUEFus
zhIT>zG+=%U!-6K>9_eBPTSdl%f%SBqa#R{s9gmqsep2e;{78&&YHsW9krzlzCO*6v
z>?j|^avdCPi-P3we=7j!7=4_@WxZS9wr{ITt~+jxlJs7j{I~x7@&0>#_cr<`eoTZ`
zX#0=^Mu1JcGQw)|XN=H30#Y-VxAtBD2d!eqW@~4_Ak#N1j=$cC*$~CoVM*J#zeHM3
z5*Q4=-*3z!f9(7Du==gt?IaPYo1%TjHZ7~EmFsDL?_cppb>r>Q2@ZTNuy_Qp_=eJ_
zDc0G4`vmNvh@9Q{oiFj2j-h3SE^~XJ$u)w1Yb>6EqT<_1h^M~1e~uM*oE{(pQH9dc
z&OVaBDCt|DYInAOG_ke+!c|@?;#B+-cLLO!C-r2Xa%;ao<Uu!@9OFraAOSG{)RXHl
zH9B-Oas(91pFXgdF%O85;Y-{8+d~_|%DZFc?n}1@o~bX<lUM;e)~=H%{%faTF93mf
zosa9Hw#oom1Rd22naICbLBd`D4q;sRYSENLWH(7xK3Kh<UV_Ijz1U1Z{oGwLj@eN1
zqCL{_1Q$>7Z{@wyA?$OC1vixw%sWQ}v`-ybcyNW@tyu~09A3BZw4xN!Pm8UL-VuP<
zIXs&=oM;xV3nhL-t@TE|*=J;5slj$XXR3w$_cw~eAfJOOCeD28`_pFFEhlZ;QKEC~
z?=wI^d>$5RWV`w*1J8RDaR%ytV4y6cWrC{me~+@92D*G|&h$=tkA<L^QP;~RO8UPp
z(qH-tEWc(hQ}bT({ptMB0P-MaxPy`Z)1LBRPYi4MXSgJ>g(!31JK@SR%EI>>0+ujL
z!68^^k~?jAB$Na%!tfC6;1~vVIfi4x^*>)`WYYR^K~%*$MmhdM+!x%xet`tk@v0<;
zdXo3`EMerAe}nLd?11#dpRZvSvxP&e(sS=(UApV@xKZ7X`&W-#LyNE>>>m;wcs3V)
zjh+B}In_zk@?S3%urBl=1S@9pB8}Lj*f=-m80r744PX`nOxAx+JA9~h=8y~Bgb-kH
zqzj`r_%nnYX00DQbw(MK$o0G~rO*Esm3}GGzsh$jvvEQ4S5rQdBK<XbOJIBqV?+Z-
zY(Sd+4x+a;z`7=rH)1#I)nl@M{dC}1FN}5%k6u}i+TN9+zX87d-t_GIV_&vJ$lUua
z6DLXPqEP(r%rPsX+D5<jd~-5!q3OPC2hH34_T;}s1@n$4pokt@VSVsZ6=`bPO@HR7
z#Vq1&KG5-Dd<0?c>Ke(v5rH|&ke+KJUc9@^rz9#IsW$oVq(Fuy)oj^aposVXKcVKh
z`E`SHrbWJ+!v|aKx-n_gxJRN`U?>jisCQi5-!^TXNZ{%KDes8LTs(SQ;0osaQa(;9
z`}cTjIb*f3Ei>2wpttZ;t}j1Mpt`(Iz>X_J1_s{c7zW9|{|@jGxW;~~n)+MRPp^?x
z2Ibv4jzA*@+&??*kE+Bo_Q<^~6mfI2>+FPYz-iK~!V~pHIsE+pQUwSqN&!^R`PS6<
z>38vdRna&a;p}UotYU<0Xf;0IBDA2;!-f3l`ao+w9PaHjJ)NB($=BB`bhhDd+yNH0
zg&ECJ`*v%1)h}}PO2{>H9k6SxpBh^MQLTp~wKLy+*(%dSNS{x;*)&z}=O-7xB0aBg
zOEKj^<@}q-&d7lfq<f_O={Rd8Nt5Di<mY`FZ+G$fi{YzjlbR~)PmB{R2ia{O4~r<R
z$Et;@c*!j*ruC#)r3dI02ShGQ<>+m^5|&tfW#Te2{rk+sA^IX&L8SCuBtykFU9E=Y
zt=PkVIgJI+68g?w<ki!OE5)a8k04#bNx+&vw`T<6Pd!}}b?1=$8@FJU>}xDSO637M
zgX;S|^{>3*()_|!Q|+udC-S6JtIfI=r?zX=HdGAxR@QWKvSuYMP4@h_C&WVzNv}Hy
zH<P&xDPh6h32U3v7b1^d5~_9sRUgwVTz9_#lucn)V8``Yl|n$*!X5n0)GfZgta_Q~
zL8EDJ)$9ReXLM<18ns-m`apNc!?ND$gDrRY1D@bLE01Q^%<>-^bMv&EJC}*DxbR8=
zmK$@nHY2haCt^Kc8~5{(h)0VOM<i)*IC|ej|Jfy6#0;48Cq+Q8S@56<jW)DRW_yvf
zDt8NzyHUt_t}UV>UehWUTILakSrO!cg34Qc+f{cXuilX;+UCPNInw;a=lMTc00Z;h
z^xO=!fkOM3G!139DT$=+f96X9*D?R_gQ|Aizhm{`wG1XbMqptq<MABNrFZe(*^lZu
z<`0>@88pk*Q()<A>RIjSygP0ngnKb!$1XjOe&bs3=BuD|Bgd(huXbl_4%IdjlsCF{
zhjaDraj}f~y<1%RcdG}eNRuor+7t&XetJX5lgbX9S5SolD0qeOIXGU=aSgM)?N_U?
ze9|0W4K-qA$fCKYlHJ88G+tk5WiWE_hRQew2AP%2-Rd?OVtBt3=Q022%Eky7FivMl
z4IwBvuqyT8ku(pON3Tp$98h?hz(%j$W8-j(Z6k^rdW{STyX;m&^!>XQu5k03Y@9D=
ziP7qp+ZFv*V*}?@uDD=+;-fBDw2gVK`dgD@YNC?KXKnu`B_JWJ#7jQM6a32q^=M!>
zamhL2w|N!#wJ25OeZaan!0I+@#ac5jFlb)j-TICs;vV|`A4Rf%C{Lzq$)ptW-fsAs
z&rZ>j3=(@keGaT*J-@Rjd({n;0_;w^9El*btPv@ft!Rlv;$pUpLuAnpphOf07-Kx4
z_}WTJa+Y6&idRJ2DErnmA!2#|LHx=Cx#((Tx9vKU9uM95NBo$}9_W9<a#ubH$1H0v
zUh_SITrIrl(P-Oy#@g%muaY2UK#uyU5g45vBYLLnvTe^8=Rm@gSBaXN5PriK6rRtU
zw!3=G`?_v#MEa4rzSN%P;~2+)OJC^K{N=^v_EW}Q_9vVJ|2lsFYJso>72)liy`lAY
zY#*I0gEB#w<2lOpmISVWnrYAN{<ZDVv?^RR%4+NG#DGtMhcHvEHnr}%=`j|p4Odfu
z#)|k!aE6!Gt2WMrrTcDgQrqPt6JTUTZ)*d7<XK=k>)j_{ZR9tX14E>L!0J3ld1>DI
z#k7sF42UtS6H5qgF|sV#D33*LmC?kxyydB}kWK3%r-E?&)i2oNt?wmjb;ddm2Z(2+
zk34^z9WBb#oxC4&<@y_4Pk@^ZHRNOrB(^w<SDcBP$Cy?;RJ~NwT1g~DVGFz^_LJM?
zI_o@h3$#Ype`7)gdIp$?+lb!5_)TxWo68YoI&twwf}==Pv~$dSU!r4Wh4E<@B3LI#
zZi70(5mGZmiA^uErjN3%Y$%Jmbe(oN0gIsC3~8%>-9=<pk<2G0yftwJ)89+4hZytm
z-Rsblyji{OSC>A=@Y?#{4ToSW7zit8r4oj^c6WQj_&8VxK$Jbnn2nGw!+SRo8Q1lB
zg2nqlSBg5w4E}@zjode}=6YLm&nlKw8DsAJ+6W`>KAWGvvaNb9l1n_^ID#9}=4z;e
zOA(H{oK$c<9pF}3+^*6h&JlRp%oQO!V_e^*>7C)y;i!WVt5&N_pZq4V^my}I+Iip@
z!_ENKXUSQhUZ(4%<>d^PzrOMWjFQsxN)*k&y@je0x~v}*M4F|*V&u%Erby4F`16T>
zKXEfhKjXSGAMKct_*3aG;3k-Nnz|Xrw$!FYyg$;OVEEQeF`Ss8_}%Rn#&vL-BfCk$
zo-1v}njl$}$L4YlrU~NJPmN#^%Q?{|bR*XiE{Z8Hur>d3<7WT~R*C(T@6)O6f)Gc=
zAjrxukybKopxfGYg-uvy=PzZ;{o#J0FyIjx{{U@5@mZ-eoD{b#m;j~;8z__FwlJYz
z7!m0!(?uq4+j~9Gn4-XCWN=DRGmDC<vrx_7Hen_s$0%mPQ1PSQv?|xDuPyJ9=_TYG
zu@SwSsD=LGiIeK5liktLoW$0~FbJN}$m~E=Ij*sQTbq=;HC_1XASO5mb^eTdt@EwX
zb2Y5%CXgl>E+>sHr~LF-hsA`=b_V{D<!JC6PUW;UKJ9SZBTbb=QVB$X*GvgtlVpW}
zg}zmg8LeeaH*L10*O<r=PF^XS(JCadR`AVj>P@(+ra4$v`ctr6?||v5fj(yS;A=-)
zujxtf5WyS;L-+3PEPT+0tA73g2k~u(NjL?f7=N=NTBOM}UX8~}Eb=i)U<Il_CfiTD
zD?NJ7jCL`n6xzWC(i=&1+oi3Nlp2i4BXKW~KYKjPt&-RqwJc4yDgz6luNr?9j!Taw
zoJpZC>9+Ech44pO53i25gwGHH5gutMF`{?HI(*UGyV6}nuWeZV-Bfb|=1j?7;UQcm
z;=`A}_e1|7ScYB^8%CHu^-@T*pe3M#a&sgT5%7=tD7TW~dUMs&`9b#bn75*ok$|Fy
zy+N;fjp_29{iPJ(YW{)Wk_civ`ocls1|svqmGz|Cn@>?`0HM_5!%mXGK_HV@`bLLo
zrpYxrZKOd>nG@H7p-^F2m0M(~pa>&5m45XM;L_`>tTm=x7lmHFzjM@*H(&U-Qj6Ho
z*5F)w^&h~cme)<^;keA`7n-9t_a_^t)}570U_lBLK;y9lF6CsKJ70NaK_2RXfa}RX
z!ERx;>l;f!E@w%bBT<My>Its!-J-71)8)0h8{}RIF(jVxCkQztiQpgQpb_W(R2XCq
z-V@?sEz3ekQ{a75HOx0YaETBP!fS#qbJH_2FZ9?8w@zPFiPxSVXggZ|K+I4(ZEz(F
z(MtJ~!y-U^MCj(VQN>rEVWYjvWR9zN0<9|XIulo30{OuKd8MXK!PNH3E6jR|SD?%n
zztpt-qTRiJCgtcIwJ<8B0fxByw7wUeWK!j&0S4jnJ`aTA{ThwLboDwdPM*k@K&yTG
zk#r%<G%_>VP;+eNaj+6USW6I&J@s*}OXFfmx1&H59o;&9UK~1ymJDM4^J8bNjZMW%
zg@94`@Pcj*|IP%1%mH0dmc`AKdxYh>AIHh55zK_U*YguWA<76z`UC1%+|`GcIJ&vf
zcj00^+xOc?I1OnE+%`$1t$n39fp+t`j1%=+ge#^>Pf^$L-o07lj^jg?ipgQ9o1OH$
z(vQ`r>(2L{!mu{1T>P<ozTeVT*u04rk3<Om(n5H<#5765VkJxIv^iI!ZvKY2=9HQ=
z<(GIq+Nv+oRwljNUV{5H5tjI|`F&Ytma3=kYFu|7WQ4S7#&<zvpvtnNguH+m`k?`X
zBh<Dwkg>j+nHrxmTn1IMjQlc~mfpd9P3OXC){3Mgc3HE)!ODQ;7qqCRl%*StT<Cs=
zkznn!YBf<e*3~v<haEW%Ku3uuoUX&<-67$zelpxqdMO^H>pABo9+P_B#WCxz4%Lxd
zXFDqij04UrkK*llxg+UlB6Ut#<+c|XsSEDsIsrTN#igmcqAXXB+&>C}3<tcr*yaq)
zP)D&KeZXYueY$h?9Ach&DR-uK)<Cy9EwhdjxHtN+D>z&&wXI&;2RzQb9Kc+QD9z7P
z4&|m}Ua<tW4Lr&PPeI6Vkga5CY7VI?z49F3R(v4MBE1$4&f=Y|S81B9JvR^&R{?4>
zvX~#FgBL86ul`Jt;}pIHG8lvnyzu2X)6v~F#@n5XbC}qf7z{Cu0{fzW-atd%d>A(R
z)vK^gifQ(TbN$E1kVC}T=xkN!tQc-lL~6%I0}EioiPm<mR{wXsB&bO^AR=AA04|(T
z82-`#5G9`Jo<s-%!p#92&F+^{z8V?p0(9-p;LCgnsisx@xrKUh)3Q5&)N!T;xwV<y
za_Hw;w!I@RT&WP%9_O$fh`W)w?G8AIK@I+kiU3f+@KW<!!!2iB5*yEh;hn(gTg$2V
z<gPtt4jY8tNWsXBGEl$-TbOmj{0HSu5W8grYA*MNJNi7uUjtYo^ZMPBAMDTPFrt{P
zQ(?K7s}9Qmn|TIT@^;lep~aWS1Yoebh~C;yKsFs2`|ZuK3f*oksuPgJuU{RZ%3J8s
zNh{P!-*~d~&onxNEE~u)wepGPW)C_o1;2pH5M;>NCj$(W_J4nI3IEe-g_$uY8_Gag
zzeD&gTS-{ZP;eL96|kieH#io<=gz)=3cV*o&IFyZIul1YU))+QR}0?)(&TL@a9%Ol
zc34gL0vyx7r7PF<FF+1z6sAdb$Z%$hXfEc!JsWokEx+kU-8uu}^E85CrrG~m*ZkPB
zPpksb?jYfr=@)c@vu1WFPjHW7jTRwJok|YX&#wv1GMru>3gdA2=6JNM{_;D^06EMp
zU-U$<>6AX%l474Sz5jS4h``-dufB-+B(@o-jBikVtrqP%0^wR3(Bl&0FG;)@yZm|8
zB3oY!XVGD+=U|sGPP2~}Cjw0BJq&0ZjDdAXzCQ)pbr75cU~Kn<W^0f?bFV~e9~~XZ
zs}p7{l<0I^s^?Gj%&BD(e-Lawc-v01=7)5Zb2S1*Dd4xzhISHr7m9w!;Nz0OubZnc
zN(4l4uSJ-jBs<P$aiWWj;!DmJ<o?upCa(exYFhq_!q)_zimg=wJj6nDK%@#TbFbHG
z=`3P7DWz%YtpX%{@_46KWbdnOX9I5~8u;NKQ2;)y17r2Rty}&*Ae=zGKL>&n8llG^
zW_Mh1@9WuCWv&$o+4}*C<IW^YPorsXPzA8Mf>X*55iRla`#l17ef(tM$EZdbQvTUz
zpy>NUYdMUG8=(HJk<+%Iee#9jO!FQg(*3!XLn<&}0ld2G>T1FH)Pg<9xHVhQp*Tf$
zqwH5*jGD6JO_rOtt2!qCJ7O#(fico5qId2J*K(S^y`u|NG^eyCUw*{*w6Ql`F8@fG
z2}n`7n`qy-lv^Vw$6mVN0~Q%ncD}j4O%fHbpVnl#pjAVm_K&;=v-pCiwr#B|yjS<-
zKC)mNoVA>L9UtL@n~$PdduPp<FHn7_kpp|nOc)z|!{gwAGAs1n1BZI3CgZy~*HOVS
z>MK6_m%{>!`2e<%JXlcmFm#NiKr}c-6DkQ2<9_BqU*Mrb&gv-Ze%>Mog+k%i1iP6E
ziW((WTzyYz!i>QQjt2f&tD{Y{K}G&pybU)%<u_T<Y7mLgHQO<uWPq8sy-f7yi}!mz
z+!18#1cUroC(jaF)x#dI?K&K&e7ocWd%FAOG|LWw%27k~N4dizRfMbi_8q-POD-%f
z-Dv~C)TiA?*&<hQ%!Z3oRJIM23sQ!a!8=%Ba9pmX^Q~)UBo3mWze-vi@00;WXgM;!
zz!v`OnCoQ58(w=ly?)uRNeWqE5ch>UjdP+U`x42L{M!}gdgJ_?m$GE>z(%6CaO<+q
zcWV^o7l@yMu8UnV0Nm=g`I+o$eWbhGF*=YW)*ls7K43OjTKT(^+%v|&0wOt{#i*n=
z6K<w{<E>huJQ<NWi3`NpA*e;>gRrSncb{umB7s?@<9^HAT=Zj)`sv4FyP?vKHa=!V
zVDm7nKkL4_?Uh3rP25Mg=g_<J@rF|Cf@C3@YPLZT7-&-SOZ9@_;S6Y+U^YNU?=HQ(
zpkFOtJUz+z<?pT6(B^{_#SiSz!E13FXO$i(f!;gSY$lAeD9*r?UdPFyTP}MQA}In>
z{cm<u?@-BK5{(15S1iMxPJObU^xBxz90wCQVn)&eE8V~a@O^+mA=4?6u*nyOArU(V
z5e}!NF`r8&+pM-0_bqmNXv1_7xR(DA4a=_yC%;7{0-@9qh(&P1^XyxMe_06>G;k~6
zLh1o@D}$jMu8(%BOqExTU@)X=k$!?}{8->2%4N@<DKb6_=uNIcy!D8HiQxOb2gj=^
zAk641id$c;U{EOsNu}kO-}sRC%&uSbuhoyTC!oMxJjQ?SR`0;`^0Y;7@X{M;AH2Ip
zb#+&#ZcB-%2#yaSt&duymMkIxFioUFiJd=AfK&v}312R&&fg>HifV$&L*(eqy;(Jg
z_+w;8`&O@=@wxP<`8fzWhDa*c#>W<sGy^*PvsUK+sC6h-iXwB`$_3R7AL?io97VJ-
zl@4hoGh?&?N=(Buox7KhLw7XdDm1IW8h?{JRdt%5YgzV4l#r+#RRi_7?ah2C=k9CD
zKh^(T0SZ2w^&Axg+-QKkK-}wVAfO=_r-5tp;4H@aMYes8)~?aBX9cX)34|y9W(>Co
zN=p3moQ`dTIih##HJ9-^0+(D1`&zI=_m{&~_&|4W+JyM3O^RZ!2vV5=058F;aFdQU
z&SE0qjG`znG<=8lh77kYzju{7(cC1)QBB4~l##>$QhQVh7+vi1T|KI#0R?rn7GbOI
zT}kG5x!Ed=R66>%td-v4y(?LpKo(@V5lutK_7d6VQ-xoifK(0AX3L=uJBJv$PC={^
z<@zFR`(xzs_g;Wks;Aco3_LT4ZL<eg6X{cf*vrl3)O912n2{9M)>dzFD<m}m)pn0o
zpJX@U;b{MSJvn+|j2frvK}*J%(x8NJIy**D4A};>J>jHIvp=RCwx2cOvfdeh2)|(D
zV;PF2j&Gf+zSe;QO^^`lqa3<ZpmwZ!r=BiCvImmhKlS--gbWFOs7-LhQTt2Bf?VBZ
zuxmG&xos%>4kwik<<!)YA)rrMDUq8DBW3S!t$`QN!Bug&U?Z`2`v92$f>;u$6t>A7
zVMgX82m0X-2L`S7E}qI-_O`@KVvW(IeSO<aSLgs<v|vN8O!wLrY~O2ITj-B^1Hv47
zl4Adz=5K=h>j5~MfaSp<7#^&V)U`=BQ;8=t7iT>7)JM3+)rD6R_AMWA>gtn$RPa<0
z*;0tEbxZPQ#-!pW>^1v{A8?;Gm%SJTHeO{?ptjFZT^zdA@+`azX`1}<^O4(Bft7{f
z8E-$F{`yQJuT_O%hS>;WG^6?=y6xD}gc^hT+2Qw^UGJQ|&5zMZp?a+)_PG4P*PL)q
zt$fM_L~NoQGv-FKL(Bn<{*^nI+UJA&SVtI?_iie&%$DCC?k!oNz5FRZywnd$e4z6Z
z;=jW<cfLuUjbE8GumJUQ^mS<EfYonDCBdWc6_|w*0`af}X}iWRd?@xmK~wOA1wE5e
znpytRN3koh5E%|#g{V~R@7LvWlA6^7u@CRjfs$UXHWfVOZu_{$14lu^AN39Qr=o(H
zaOUmKov+HzAlw<g?_ji@O8ko_*jF)s&%y|40c~F--f&Hx4Z2%9U)miMb<rlthPB;G
ze3M@4dSYRJ-`D}o+JYDDYnoB(tlVa#v&aWL|0cWKbrf~^!kb7xgD%N%o@-F`2T+gE
zb2BMd$z1w+K=U-e)GkLv2?^1RiqbKY?j&1!o#Ei68FBT>Z)&s|(Wiru+XGA41uvJs
zE*8BEnp%Ti>^08egqYfvBJZ>-m7y0$2ys12P+>WUO50xQ3O{y5AfOBUqk!0_c6TTC
zBP2K4UA*FcHf1|bn~7;`Vrcc#f*l7Gk5Tg{-5c3UdEu?T&7ftEC^NCo3~7tk4@<mf
zYdqI#E{Q#E*oAy4iwz<!Ha-?7a|T9DT8~QmwDD{AlzPz|bGx#k$2N4VE|LByHTKg~
ztJXS+pvSw;$G2Y1Mf-o=*4jzcUwW;41uEI2KZ07b-n&JS>+?SEm1}bDPI@WtG#(l>
zXiOf=%k-1)vg@Ws*unE=YiA0P>Au|0{W=OxT@gI$@m?9jc!0e1m2rTxMl86OP^itD
zp;r(W!>(o=gw1L}m;6Ct^3o*`lF+<hjg+P*K{Ryjc<7Bp)JI1WF_U3Cd;dHX;Y`q9
z7U%Hc(u7ZJBO_GaPo<7Y>zBM|nq=KKMn9{52=D;A*ly+@*yI0DmdDLu1W7kvjH?|~
zKRQHe>;u^gP+5-aCh=4v`&I86T@00g#mH2jEq2gHQI{Dr_0@dGoTtFo{&Th8x)0u-
z6MjPm)2BSwMm=WN<)`Py{#VO|c`U+(pt7T>H1i-mQPJi44e+prYT_<&S%9ngVXHXU
zw~ShFX(5La-exXl^b3PG7*kSJvPg;FR@_QLx)&_5ha4#hgO%ZA{gW5;pCHPw0!Jj?
z3<oD!6+&vhtvEOak3O2B8RF^(s!!j2@lcn5Q40sSdFO03LQbd4ibqlP1Q@!~0~0TZ
zdDOqqQurq;C|5iL{-G73PRtkN4AFOk`?bFfiBQZ^Jy8TgOh|W?xJ=eM%vmFNtTfHZ
zgkJsxRIv*!i5=Nxpm?o6M{B-cf{S%e{+P>A>78NqUtY#38unSylJ_Mxz}2L&THm!f
zKaQ<^a?d1#KO2l)#}jxy{1J|m8&_~f)vhuYg>eQj+C-|O#AYf~+^q(PCg~wy0!7DI
z6Z?^ejdEO<ZAr6&rC)peFplFXXm<s9^gr+ikppR#X0J%cRgVe_D~*{RLy*kpq)V0@
zXn;xr9ZL!AEzBlExYg`(l1`%<wnyqIiFm)ew7^=g+jS=G_}F6ceHbIu%_vGVdcr@-
z1chEBfsHyha9L&3sSL6NN<YDXZ&ndD2KjaccY+k;eNGR4pSoYm((jLZl=j2HW#vXy
z%qQ@hp!tf!2zDPKNjfp)Tsoa{3-fC}>}$C}nt=(DoD(%y&cU8TQh*EhigU?yV^fd9
zDWGub9L>ZF5})dcFAC|reOEFC?)`kJclai0J!RCGRTX0Ky6w5!Bydx7$nax95zT>B
zF=f89lH7k@7zRuZTG{6Wl_@cx_$BbQ?*X40z?EC0sunsA4g<o&*JuGwqyIw=t;8^I
z_0RQrpYTS~z;tN_sdqBVp^e4=r5|6ldi*8Fo$<|EfPQN<3RYq>>nYaCPY(RZ0vf3S
z!q4NL5;8pSGIIa)Y*5}@09HoL&dpP2=w_(9!_0i>&JIs5C!hJWw@{hTzQ+{XrPqc~
zcE{45SWmtr&(qkvwpfVe)%YcGcAiabhtQ6dfLe$+NR&TFUD~?!+d&cPlf_OGdLVic
z&)D!wqWZ=-86BAC@2b|O%fCCw1&||*14QBY>ROof7$mF&^r)ZW{NpfCb0Y+T&Wh^Y
zmY<r|M=J)9vi30SP3S=X(cVd>_*b+iyv|;eii*63RRJRAsrVm$tlr@pTiEc#LBI)z
zk#xyLZ+w0Y>Akz$AQTAEq}W+NX6#Vle~+Vg(F^@(WWTbOZ`mgn!j*S1{Ta(0(YeeQ
z=!PCqV2mp38R~oV+<~*usY@pjB)9~%pV9?*;0%Q4kIwf5!j2$zPF8(>wm|PbYY8BJ
zWO7`}6L-(xnDq>@Y!Lcm0|;{$m~V{r%ersX*Cj7k-S?zLEa5?-bc6J-WyqfmnyNC#
zRuPm5`z(uRHzN1+i1MC8C{@UIFN@DEiESe{qA&3?W;#)%OZHF<9d`XxjP6M?V8#a@
zCj;;(p0gjBEwBY4JmGIr+x;&vI?j06IQ{crW1^g=dNT0=`?EK>U1eNyILuc*F!$F}
z+mbiyR<7V93ULWq14j8OqHd)hyS5x`j3bVp2kX#glAWfN=@t}KjYX2DFrbroM5s$5
z_5Nr9RJ|nlv15reK|M@fJG1LO&^aZI3nO8ld?AbM!jb!*H{NK8;cT=D&Ii+Fo19>c
zQ3XD;iutNdF0Ag(K8uSO;h~U*lX<ByZ-IVpf6#<dH03sF)^l)7R-K!#_gG$$Jxbb~
zK)k};B?g?R6fWg+esUnM5MmHEr3v--DD_dw;0v9(nS8iCWU#Rc;{)vo&lCev_#c&W
zQ>h!$^gliRhJVHb{i}@ZT<>sO!|Hq)P3Q13h}<|snKFS>qAtCht&z*Dj)SEF5kZGW
z{%M9E@Lq6=Xlg#By)&i4rDH4*N#GOzn!$vj?k%M!ER**EmX3eWCWUyk=V>1}U9JRR
zvA0kN$2XZG<;6MhoYAt}pBW*0zp`czDT9B<?-e(A1?87Wbvl#i4_DHxfo0n%(2czx
z9f*&4$5VIK7W0k}k}MRR{^IHWnN!bBwkvYumU9$VS8TF~Bo()xe(>*E(7(XVjnfx?
ziY%6PUfprfxIXV$X2n2Ru#@+I)~rWVbHSib9FBuI4oCp1Z`aQJ7tyv{8d427nr(yF
zC|@QH$~E4}<1V;`2i^6xePH$Mlw{0G4s>3;z7?fRVtKS7=6SN6VhW6i1%A`KXllDg
zmPkn$w;b^N#D0p)9<Ptf1_4ivBRG{X*frlT2Uu!7*G1mS$<cx3v&xHSdCwuRLt}TU
zoZ9wF{5b9!k^xmeiMVds{jzadKBBOLpbK;X<PQnX&sL%TJSuB)^s7t0TI*v1`EwZD
zVkR86uG&9|JR`kv2cKeDwnEAMPN_}=rnLWe;j;3j_nLmRPP_hk;OCYp@YWdYLJLbA
z`u^^_Z`Q%v{S`ka4rQqf_<4oKenv`&rbi%l2T4ao{3_j#<JJau(0#=cp-$|u!f_D}
zRldXe$lhv!DatA<%5e$VV(gua0qvUy@_sQ=6OQvY5i)g~y2=4^|ND+t1Sfv%Y*$A4
zYGcmjkV{rjY+?N+?ZQkFPYQB%Hf=2fBc}xS+6?z)U$)AJbPIB1<bt<gPe)(Bv2?%P
zV5j@{3QP>-$G-AE5RPyhzW0c4#V{DaK<_3jReY4X;=df^-RCZ+vHDbIY1FR3+{G(A
zIO^HwuHn>)az2Vtusn))S=lF~6mv3mBWOBg>~wsZeFwy54)g@-YSoIRPL)FDlgeQ3
zRqwXpCqZ*dIgSGHesoPSKB_JyvTpUL4ieIzp*q7k$t}x5Z!zdQ{*G;13E84<%Hv{n
z%fggONIY=*!c^AoZ+?5|T>s<F8rS3tsj$^&chXr8oAeK%Q_Bd)H$=~Ik4-S?X^-V9
z*>=81=|%396t+^yGR%sDtZ7|-ZdObp=Xdi3K0q~W%UY(Ve9onG{2%WCU%H=`^V<0N
zdBFF*rdbhusNG6DxobNbYbLk`J_=rJ8mZFQ-<6vmY;)C6UinoRU7hywo7Bj-yyGa&
z4g+lgsx!%AI|#rR-}V7U^?Bx_t{>@A#}*;85y4?IM>qxlPMO<A`*TZonMMWwf{>tg
zeVt=Lx*apxx>2Y?$mfm1#Cfr6z!#7|Q60ChD|Dkv-xWdrME0k2V1BxY^hdez!FYe^
zYo$qGoU(oKzmmcUHK20*Ut5=^r@-keR`d3{#qQj6NfJD4;{G^?^61rrIb<#I8SF<H
z`NEcE9hy%_5no!SK3|JzZOiTY(NDMZJ>|n2kB7|1$Yw2EHK~r$bYfx_qN|_TQ1_lR
z&O~lzf{LYFoFC6`k`tKY@lhVl+PYTH!E5z{^a2!Grn4U0aV3j5^8O5Pq=DK@xBJam
zzCvCuytoi8BT!oWbozDz%y)gDyY$23`UvjY_JBjgLo~rRZ${H@%2<w*U0w$2xihs*
znFD*s428Ngct>Y^0zpG~qnX&j@XKZ{O{4j)JMaL5M0shFKQHjKTH0ya`FUY@-;dYP
zq`kIgHCYD;keJ-if`mVsal75uo-!U@J8-zcB-%JJNh67ln?7<AYy!eSf?~R}sp_pb
z_g-Me#^GhkOKcZbDl@uz*lSygEWODxI0M@bWL;~1quAulfPc?%W4Qra@U^<{J#iYa
z<nc3*_+)0{0bsbMyp$C`U_1CgC3ASpLE_{A<&ZB0p$xD)N&V&4$8cHr@%<$Nx@iE4
zwG&z_92bBZWGaL0OShPzBzw03>k?-H7OY5E5Jud62i5F&toEz81Y6?;EVTk&&C35?
zx&}pFPq?)Bh8}2?e}C;nj{G$XEO=8KWQGOF2ATS69<4oa=*BDX!6Is>BwLBSga3At
zDe@rSzAM~ZklXC^?06b-iRzFcJ7@4xUx-SlsJdf0>NnuWRD6B6G$xs`^LRw%sYR>n
zJ;Y)lN#m;0Ag}W$cR>E{yR&h*4>K=+CbD^K3!Atqmg_hmI;=I2v4&L``*}Br>Q*#>
zKA<4)sGn{IPaB>7wxL1uqz_ZWDT#_Je>M?uQ_}10O(pro4$kC$Owx)7BDOdTN8jV8
z#dCX1#iK1q*^8k<gp{(oh9x1%G18m_eiP;9ks(p*(L~z3?gy>e<=rIzls>OCg}t4_
z;5<wTr(0!iVMg;h%kpxo*>=Z^#l-SkmV<O^@%sDSSC72x1>@(`%v(d&#4dUc-lhEt
zhnSO1553+ez@?ZWNA-GNpTVY7^vpQjLvL#iw!;|1S>nO?ma>s%7u#3Uj-FNwG8c;o
zThaQS3odQe0CmNIjRTMQhd>Nq{P3_Q4=%u?Gs-_$dP1vJe+DXYmhsl+oKb`-sz-et
z>OMR+(y=8Y<|IWO2zi$lv+VEzTGCXRnv@~V9I9z=;OxAb+TJ%S{0*mi2ho$}5eMR%
zuduf`{lz6lUi0MwAuC$VAkSdAk2VZZf%c>B%Af?RIvwM%qnwQWAU!F}<3|2L)bl$k
z%|UQbmoyWem8HCzcszy#_q^~HvPP(jTq8Ri);Rvd(RZTr`clk&3zT%9PaL8gllhEc
z{$dIvdCaT#aBRcixw!Sy%H5Cf<w=eosyf)4`jI!c27fmW-SOf2{N_ARDh^Mi83H-m
z1Ek+woq(mo_Zb$EFYT<&+_bIvRm@YAtMEDqG`~>~8Mhd?!UK+0mw-v_s+*Ti-Y2^8
zWz$}%0TR8h@9BvrVM_$^%3}IH=T2f_EHDvEC@Sy@BkVbW?the}+f?Z6dx*(90_@jd
z)4o2s-V-gj`XQG3R661B%9o$6|L^g#U^e&ClSGySF@Z%-5GN(YjoLiVCz-1s2tv@u
zR7Akl*}!9W1=R;W6Z+#rbXfm}?pIexJjvO6R`qkINLJQm?40C`P`staV{C6m5a70D
zkSfHeik2C!#Qsm{$Ips3>$IdD3zhE&$D*MwB`5=U`w@Y;ydNY)*q9U4)HUzOo&7d+
z(D=jP^)(i}Z*FfB_J8D67j-s=L?wPW&Yyz%AjQ&{b#d^{cQUAIQ@EIEF^lEvENjy^
zC}|tsV-5DBJPPP#b~<0EUhvXWFfXj|d=z(z^S)^2ae#u0hoXKEttEI2&m=w+enSEE
z-8d~~q&B}Rv3i~+e^|NzTlZ1XyhNe&G^l>35Wba*kn#OdgRFf+W^*5t%pbx9_-WJ%
z;Y+WaE&p{1f-6tC09iL&lYK+L%xAfS)3M5cOiC=rqUApf=7K9^yJzxh4tD$3Qnha4
z-mTr-sObrIs4~;Z;Xdv1xA7B$R-oL8U%C5a+F_!1XpM~6Kp*Ule$|I_(=Vp|6z=Xi
zt@GzUroc@6D``J{@UOd7QK52d0^njgsN2QBx;0lQ^y;yb3>K1XQx0@h!(*I4RQS|e
zgi?7LFz(uJOdHDa`r$bLVMWx^@9=mFjSp<bO~T#DNr9H)UX;(l^LMRc|MRp6YNy$;
z_>x&&RzET}J*b+WZh2CoPt1_2^<(svFFVl1t02)`g86m53?Ip}pNe?T#$w(JYVSrB
zDVaZgb^qeWU;FNl$EZ=@cN#!YK)EoY#~kNT{u1O=BmZ(P&@R{wn_E~^m(dpu(FDXh
zyN&gQ3GevKqvr%Xk&^-4wU|(1yFcQ=;?(NON|-HhP2u~w#XDDJGx+2fxXxp9QqYip
zv2|_nypv?A{>@imeEBh@-cgML$Jjqir<0L;-q>$cxBC!W?U0)YN(0;ksBq|qf4GVb
zall(>EB*StPSr)GwYZLegbRa&|Ir;g!;lmtO(#0e`|gLIgKP}%zd>@&$@|}dg7H=4
zS>`leA?(Pq3jIgys_OdmJmHOs(&kQEP8C%i3iO<$*gqel0-~R=a=^LW@$6XO!>@r<
zmJF0%Ukr`*kA7V~88=#3yZHN$6BG}SZV4@8IXU*3Z2;1}Dd~s~5MaXie+$<@!e7#n
zQT{BMzbkA!4xC@IrMRW1JomIzq?DxfIk5*xO9E(t08hOB^Q@)L9pE`W+&Af{cm$uf
zXgY=k%2b;cKx-?}e1gHlhJkW<K9VDW&u%s%H(jmK`A@L5tB=w-ZL|?pkINhfO3Jd-
zO>ep4<{>@B5o>$Gjvi|Tm~!4iKtP#0&0?Qp=_%W<biJSXv3+TndvQ*{gvgfoZr(+I
z5Tfbmry&Fldqo8y@kSHpfSM3;uYdound?8duI3#-mWj9h92nV;Pmx^F9&L-!=cVq{
z7{5xyP5~u|_j6STM2R@&NR~qPPY-9qyPEntL&yKME@DOq8rVVsB}Dy<Jb+mfR2V5I
zIv<BV#51;OeT;MuasKnNLy)-iZGk1L8hmE%RR(Gkbl^qK#$_`C#K*=4R0eqEzrfoP
z5SjDkq?w03##XA67gjW*2BxgU!X6*a#GDIuz0}x@>2foK#8MHfgO+6($CD-ShC1!`
zc5bz?i-@%6FPdaHKN~D0@qkNb&yDdxS6kxX!pPVMewUZtvyy%usB;ZV1)G_@C;jjE
z>Vt|pj-a=quF(fKa-XC=P9SE;c~n^@K8VdIv#m~nDkZcUKnqmg;c{|ONn^}X#H7~e
z$D0Lr+hWQ)&~2vwLJA-<@0tZn1N8^Bm|uEeH@yx<2L#roD!VNVLk+PXe8R?$jK5uL
zurK<~oS74dKkxlcu=did>YRc~wtx2$c&}o(ab$Cbly-E)m9#uyuq`d;Ki?W%`+A{t
zwn$1j{-Oe-4)&+R;0SQeJy=t1AE}x?EbE)E@!ix`7GsV@{5&!^fLBd0vv$kwgL5vZ
zH0#?8lphi{I--jTUXH7h-S(`|!vs8VL$oi*bGUyzd6G~neQQhN9x`*^zU#E}v5A6b
zrQ_zeT6X&w?WxRajDuQ!1d4D?e=+vpc3drQkI&>`^i&RGNx27!iXEkbvcDJ^b9Ten
zkIfQn6(#>3XLqzD{}<QiA)jP8t@{=UHLpl#_Y_!Me<6yUfSfF9&v5Si6;_aqUnK=}
zgJ(_JRq_Wh9{=;K9L7RP5IE^sk;#?=GOx$o`RPHSM=5tiw#054Gd6Pvs?KR|!wr<g
zCC8W_h85F6HNEOzYo9BC`$Uee7D*i{0#v>XH!nrRggw=3TNoBIT8Q}ASr9l-tAioj
z=v83r_^oRu6zLx=FQlM;PRY`_Rj4*@T-H2VWCrWjqw>n~4c}nBV`qT9P)d~%beP^~
zKJzHQ$hB2rq4%8gaFUuR*BZBt&R|m~vU{q)$#1F)*4h|#{@yr-ilBvs3!OYz0g0J<
zh4H#mIDlVPNWoupn2otCr4W^cEr9}&D-mEAL*LLO4|h^kZrc?MdF(B{LAs-k2?s$f
z)f2%46cp|6Gu3WLQ|Vl)a$3A%mEq9cA48vEq4S*PaBv2w-IZ?=PVTORVY>j%e11DA
zOUWF0a~^cs>3wiS#PX}&ey9D;rhZ)pde}Xn#K>vm4OPBihc#C4vcl6mef9P|s$mBQ
z;EIYDGc~I>kG|s=Is*+rO&uy!A+jv!f3gOM2M{JrauYyG=O$3=n7X)fh>P$@Bh={a
zsyh3YhnzFL3WrRFonTV-(TxP>cZdaBRBm>!{qTVjQcF0Gw-7IdY}GXX8g0?EfF16<
zLjbCeyU(x7UA*%Rf?&7@ykqx;&HP$y+AKnW`IveWght)C!?l@JwMFESg}}=_b)zF!
znYi%Qrg;~MU)!d7I07vW1RGFYmiz*$OFg?2GuIW(*OtAB+%!jYl5EwmT=3sBAN$V#
zz#fp?Uz<}t;7?J2VQRF~BuyppjK#QG27ea(!}=xbq71qFL{4ax;+^si5N*#B8{|<s
zjzr8#C<m*Ix8FB$g+k<@Sh_M}hG_Nz+X4gES;cp!eq${vFc;MuJtxJGKZWed4de6G
z>>7&bib+1UdO**5x6QKq-t%jrH}rww{+^p0T9e;}J;(eYi3ftM>^DovFQlq)W9=Q|
zod+?Q*E?x^<+=#R2g6PO=Sg_it%_$KTl8~vMFBC^_BQjPL)A61gwen;ChUkTxV4tI
z%3aYp=R{t*mG!nhGo9ctqMSoU9xu9Cu%f$60M1HY%B9>A<xXStd5R6|4;UK$A6r))
z)`a)<bq*;NCLkz1RJueRB}}DTLXb8PK?Ugsrvejc=~hY_MY^U^qJ*T<lG4)6-a9sR
z{J!shK9A$>oO4fp&N+8<gzgIx0?lr*_24|4bdQfWcdv%|Z&PcX0t^nQR-dj~dHuuR
zTl@U{Y(@u2FW952n0``#3Bx_5J_uQj8r!^k26xbj#ae4c@rw~-Tto|Vz{G3x*j}m@
zV991YOM@{2cie`ZC${sHSF2`Se9Z~gn{*hG+9q7IoO%DgP;|&kS`gf*l!nziRog=T
zKJBv?GEtV4lFPsPwE}0vZBu&eqSyL>^M`Lv{&rcKI$sW+!fxl#M}5bO7OX7KzUXRq
za<!$XPoB*AvRnREK*vj6vZr-F<_}=?w9O_7X4__|ptD&H-yhyk@b-{yiEfrK2n<=C
zMb~j&#M=NlP5$e#%~Lej#TOkY@y(bJ!7KYiIe0zeFEj|fPE~Y{GQ%`)eB<fH>j=RH
z2RWG$8vIBHYLq->o7&*q@1;??AghrN1Fzzu9b}3R;@?>Q<GN8$O8<zv<|nG$jh>J^
z$pRC{CA=q1*nnynw9PDBYSQ{+N_Gd=1g1}}+c()!gCTQJ)yEpTj+B@TUH+fAX$qB7
zmYR4>5>S2YEcNA@D?CqFA<}uNipe}OLgFF#+fc}VjQwhR$oDJ*n{6lM#;}~C2%J6+
zvrk_SK$H|oo0`&eFT(<Hh1A?FFo0>d3;7%oxJAN3fPulltrCY|K`QS&6~__o1a;G<
zMK|Lci?m*$fkC);hVcatE_hNK;r;Bl0M_bJaG}oW13zplQ<6Ct;u{E;X~1U2*&ABL
z8yY|zunI8nmu9P-_fo*SGQ{UR1__*GYVn1{*j=hSgr|4>d7nEq#bySqFq&mDc2i2?
z5w`=7pJY)cVCa_9mj->h<qX)dFuv7x13z={k1`o8jM@Bs@Zf?-%;ZRPjJI~pTqk{D
zg=+3J{6uUC;j4dFL)LEOdN~Pwq{n4tk@k<)6&3eCw!;QmM)*D7bK*a7qj(bnb>TG<
z`mT5MMRR>!{>;*Rq~QuTzO9^Bb^#<nu4PDe6MLCIx<Jsg*Ste*PsSgVA<B*pM}l^E
z&0L3sffxJ6W!MYVKU1S1*|k79tSwR19Q1-yD8I<N9^VCJ3c0sWRKWEg7vlf?KDDK{
z>-P4}6mE(-VI_rgV|Z;YyKPxoZIkU7A?HHrQ@D5A?#ySJ9k?r94XBC}%wXmGAJzIr
zQ8M!mylY_b?IHEvCEBNhPNCb_Y5<w+SsubQNg9|w9osZHpi%<*7{lJ@Jm7uLPVaYM
z)ZdF9@3@`q6|(MaS&+f+dX*iqt0=f2=wfg(aQF`-`X&|0v*K!CJEm@V67{2nAB@3N
zpPAJ2|6h6W4GjWdBFNA7X^9!AM_w_mG)I^yi3?o`!9DuBClUjiNAY0NP8mb4mDFlb
zp`2FvQAM8TfB0J<UpGD5TY_Cr7|XB(SD5Kzd-AUHL`8hcrtg<WwDJU{LFSOC3>M7o
zI(3O3UP|Q8FS<Y55pqKWaeH5p1XSJy1<E8VfA<#*py*w}{wQB^E^o;+-kugyq~I?=
z)<zVlQ=a%Xa5ufskUIR@7rv(v2{eaE!>%b6yiACHq99{Y(0(^PLXinc^6qNZ<D`b9
z95#w876t0E7|5(xI&ErXhK5q?9qDSu$~7lcTGPqZ-#eab64*t_`yO3eKPrq>yG|-C
zMwPnM(ZJ6R=!TZuI?jfCd;p9^@;$n3=`eRN!x^GmNUt_4%e0+q??pz9|6eHXA4z?X
z!CZu;6Ym;Xr;;_hPUzZKgLD&SvC4O+ZE8>NodP|^8Qhj(u{Ed8JQs!aD1r2^ZVSq%
zNHbgM_U}8(^Z=xE;En2$q})(m+WcUB_P+eQvGK3I@ZkxceqCwE-WQshLd?^yPZS=?
z%KhyJEt8;c4x%-<ZvW#9{=ac#gU@k=Vyaw|9c4lw0d6NKGeX@Vwszw~^XGnhC4^Q)
zM(v$6ln)UTw~#j<x!EEnq>Zw2`)SQFym0kl=E=Py{xEsu%24@=+-$>T+Us^^kQ17`
zc8pUw?yGf2g+ua~EVY{OETAq^8pi6!kC67u?2M7y_fk^Agi9+7)M)-3gA*F~?TYPV
z8BZ?xd24@Q8P%|{wcA9%+gKymGSPb!J;vBVyK>9DT$>WTEW~u<nBy?D^b;4P9s7rO
zKR9Sa<9Q>T=)?8939#70zS=$|1n4@mjvgKV`oXNjI*j-{O;5R=stKbnrNc8r&@y_@
z$@<q$Cy@Ei|I7jm6+E3zI=7$WYeQLpW2vh%>EYhm8`6hLOBb|Y8)1vM6c8`EEvRi=
zq@%sD8d#jND?gDKZ0Bhr+@^4uAbvcFS+wgdfWm5>u1A$YkgBVhK~Yb+!i(cQW%?>s
zBQAGbe~D*^p%mk*nPu10phfoE1Zc;{7aYSYqWe@M2?7zz&9lyBHb5!zh&%!E15ZGT
z?6m$$4WgG!b0$cy5n2_mA77p`lWDMX;#U#902<x2lbgk|k2`-E(^o*t0#XXne1Vqn
ztdnxIs8tJTy59b9+fN?PK(;npju`txuign=*Z#^~@~?&T1JN7_1J!;!Fyafvyn))*
zOd!r%a{6ahR~6IYj!TCl%y-$~#u~q)h6x|$rBj%U1_q#feeblHhk`Szu$5gM?dSV$
zRE|qX*Scey+4>pP@25HZDHO{LzJ8(R=DT8=+uB6L>9Hlo>bn`g9iA3gRZZ$OUbm9z
z%djS)gX7JOppe%n+$#p&Y!t6ttHMw%7X|aQ-e$68?~L1Ul<7+y60YUIfBL$>Z$w6?
z%VJ|*`41}87r~NUsp+$;gRMdOmc(}c7zu+ermoa=+<p-(<AXf`IQ`AZo>;l?EUE!(
zJcNEtU^VN60%b;Fuiy=%!jUNoLzkZ)uby>Xc;2aF6T6RBKX@<)w#lb1Oz;4I<5N8a
zx?g9G!OmF6d%BJ;H{VJR3&ej~uMh(O-IS%>?X0ZaWQ-rlxMOo-f>xPxEr}t)yAasD
zLOzOq0dB&amZu}icY#@xWcR@6__DQZ${_zEnQw$od6KErB)7T5Q>`V3o#LhWRWxC8
zB1FLO$HU_N>h>7ob=|tdGxc$4-NjIEZB(KZ0#S7dZp}x0)Tpy!&o`H=(n=X`&!I|J
zhh4-LyTbb&U+iO)f-`s|V5O;$o?f7vf%F{^+sMTDS2~!DEsDcW4zF&@vTT)7g^t*}
zCQ97z;Hb*-69b{P>y4#!lch357KOS(F_Vq3G@yunpmOzw0^R!0rLF*kY_mL_=<9oU
z6F_=F1ypmIZ*OTSg@=;?H#GkO&6z!#D}n!S0%aOVR9JB*LUl*&;LCuArzta@NO=s1
zdl-(|t6m;?E=wRvcWAR?im7)nG?QN*XjuguzSX=umL#T3OcI=Obo<@rc*YAqCY!!@
zfT-o#r61;TCunF`x$a56{UdhDKHGG+R7&hJ<BmYqKpS+mh(1>xCjIVxyV^d3LVhE$
zZEWe!x0uS2Z4p!3;7g&{>~V1%uMHVqAqn6-fw$yVYg8OdjN{C%O!b#cCubCs7?3ZO
z%ropoDalA0Vph4lk<A<2j2G2+9JoIQDk^wQ3jDe{F>Wsbf5<~j-stko``c^1TSnI_
zKT+DZV7I@$9Y<p&w-q_}>jfXAGR$jyR8<SB9Z@k7+<ql9DK9RTagaGAQNPvs7J-+Y
z6Adl~J#Xs;4Q7j{k53sn{2bmi7Icby)B)QcrJiF@#!?Gm{94%Sp(VFrxZ_&lVnys=
zHkw2iVZfGJoHkpY*gOAjirIK5lXBdv3K1gZy#7*czlq*}9-Db_Tiv27S5SGtron<;
zr&YJPDcV}L#)$HiIv@VKlLjH&SS=6}%s2PAo#_|D&V+@U*t<FN?BiDB6v{REF7{e}
z_aYmKTkunxMX~fjgzNokP0T$wfzzinB~pdr^~Z868jCqw`^CKOrYvtg+b=V55S(h=
zlqoZYD-}n7OlonYs-F;wQNUjl_>t_uiuMqT-b7EQVE&G(9L#_bWuA(dag)zJGveXM
z*R$@h43AEO<GjUh_NA`Lba}OTKGX9d1SPvnF?xYYO=4iR^m38LRuL|ckigzXAKmBu
zwdIBuU6WUX8&$MIE(7ga4|Z|v2@2pIDo5i@?XW=b$ou!nORrFC--fn8f-!;R(%f|5
zBi1T54La0v_e$NUR^~ZXopIm`hm^;EL^FpB&p_m82u__L1=(Dk;5f7Kl0c5p3wrtH
z^aA#MUb+$d&SvGcC2wC6H#J~~ml{5Zx=erVh1VBQFsPo!-<WAIM9f-QVs*Qj&ESoP
z0dO@RNo{K^9_$zMImkp8W1+70-59?kx7c+szF(W7l{tpLk@%lIju(ND9!gp%r+JCA
z_sR+@Er^d&EHJ7wSBcE|Af%JJfjhqRX8Y*<EMZlLG#qON<Kth=;lm&H9aOED)BF;5
zP_IDB32B;;o#M!#!^D#20S-R#?JHpi8`-*yuM8(NN!%Oz;x_bU`m!E>lOcqv!#u&c
z3OW7KVy!3q2IWkx-F|M;E^yr2&)6vlcUy87;;gh`fJLT=H;DZM=c<><o&1vLHu|5`
z9{0s73PT`fC8JZ!Q8dSXqZ)Vl#T>%|GXHy&nELj2pQbhJnx%?XJU38))reBzODCt>
zu6(%loFIxL@=1~U@fSW>q<&fBOF4IRpUu7r#8j1MSANM56djl7jDB<KB~Zs1W}uEQ
zInsG4PAD@GE=><YvlXN^VPyL$zh{!>+e1M;g$*`hdrcCDV6z6^^cRbdXqz7gmtSd?
z5WlgZu<(J1sR&F>&GX5MVr(xs<u?16$Ug1yBR9MTG6v6ah*Tynr#xfr(RDJ!&5ekR
zdyvVHjKVFBtM}7G=7Sec5~nZ7Lg;Shz?(}s`W!m9PvI!IZDr+tO~&&okua3m#>`V@
z-3BgXJ8rHY<4z>LWHN0<wz2J=mDu#=>^2ph=-Mlprh?J2?SerHdE9z@HU!JONuIJE
z@V;`X{*z^Tkat+ie#<@+JUp|R)<Y5=lL7yP6DGTzT^b#%faJxQV+6(JL(vvkfIYpN
zh3)lM8qN1<tj_TUu{6Yk+{$Dn5gjv->_4fzW7WKUS*9Y%^lc5n+I-FiH>>9K=P~F@
z1Z@9@2?e&)fluAm;&Zp@PGy`+#?ipNH1l;%L0BtaZ<ugrpN=;y_yETciX*~FVdKcY
zyHtW~Y0gJ!A(+?3c$&1sZ{Rd5)WvbLGNkP#<v*oxhFB}-G~4-ftEF4Hv1<`u2AEhT
z>tMX#7W70a)tCGGCy>mi!WMoJ=|T*0GpnO#=l!d~_J*LDrTtgw0L{%rSWah&s1=?C
ziv-`@fYnyZ)L{AkUBgdu!x;Ve`eS+VYE@VNM~cAi#Re*G+RaXOMy0uG;Y6ScM4%Hg
z>_y8LAxd!8zoENl3dN}yjV0pCX1#Y{Z+)Qq)FI5+)V%UqujVE<k2vTCe_*+YFqdM5
zQ4tYwlYDf&`+c%q8_p_7z3vP;!G{%2?Dwu#D5iziM+16T!`%F)(ux5$z2oD{4zS7Z
zMAAsW&}(KbC4ah~u2o$5I{;lmO1{rfxr#9|f{Bi85gAnBM8|y9g%4K|sg@*aS}SYh
zg_h5R$8U+0?X_;|%0l8q8iRozTV~Oa-YU6&_`1nY9dbQ|9KSqlq@7)z?yuEE9ofAv
z)qtaxb2@rF){c`1cyxeLvJ9OM<=as{MWm`IVR}&w>yEC#(#x;vaZD|Zj;jdho&IAb
zKPmd+^y8C#dTlGp9eOEE&}dTs;wO{jM0%#-ifZX#EKKCup`~L1O*qq_`|LXTq%gb-
zJYv9BUj;;4liwR?S8e6CAOhR?F!I3EkytKaCTu&hmBKEr8xwElpujD=8Q|g$8h^#$
zSNJMBrm|0Z_AbnRe<*0racgItA}Ws9VCkq8Tns$gJLa`bLmLAD`ktR&2$*?_#(PgW
zwqSi`Y`{gCH6VqTFq<iA14}|iNnm8>gu=%1W?Dw|FQ*$To18<e5}W2fB9%^Pdfe20
zw)ah8`rw`Pl9uA)%5`<_#ZSJyItXGGsDGtnWTtj%6k`LoRHW}`cA6Zd%-D6!EsOE)
z$Rwx38&xVjc#n6E#<6!PQZ7ZFsyNGPpN_T4qXsq^IlEIIW^~ubD6QN8(rih};x}h6
z)W5v8Cv)Tm315oOmQQ~?<$Mrr5aa*C6-hx5-A={})r=)UNl&@V*e{TE>U6vh)+b5G
z+WV>`I7oaxc{BhyT!Jlx{dDXMAeJL^lPf=c;7!x=+XT%+)PUFnQ<h5l3ORBuMtGGv
z3b^`Nb-TFISy5&;_5b<v-+rLFwql*;zHk71^k7nMT8ejv<#OMdy%^DnJ9AFL(Mez?
zkhUt+s$`~&IMq(FhH$5I<_t=1R8`Sw-(v%`9=d7da*Z-0yth<wPi-^dU@*TgGOoAu
zH>X%_-qHWaP!LtPMYY}Vyr3Q{Up<u}!T`wlP?B4c>D}QsNvO}lVK_3h9D6D0WopL%
z%LPYpwWhVo&Lc0&jB^U$?=Mf~CKY*iP_!E2T{j&K?(d&EIyEQKaBtig`DE6;PmDXq
z*Ly)iSK~>iW&Q{-@wC9`ni*vV-CIKG#B&D6#6(>01?+ZQd#ulYbm=9g6o4Ic9+|0M
z4}c3lj4lst11QfmJG@b8E8}!={zrf5`iqc7lOz8T0zE3uGo#97QE*FBIRr(ar*JgP
zX`_}uCBk|fpAfx?0WH=?Cd-XRFM?tkHs1YL2JFF_o<p0eS|&5{{1*=M0)q7h&2)={
z^)E(K;#J9ZPnFt0k*O%Q*!QS>icmQcOv;101R!V3oxb8qM=k?;$r`>`++;6g;y1@@
z@`XK3&eeT?BJf6mK-Kxa+<m(67&ArEHrUPhQo49koae|@Z~{Z)mtWw7BYg7s6J+;J
z$nSnoKHiKN@MvoKG@Hq|&-Z8Rv(?|9ioIW5I3-STL>8RE^;OyW7JH#RFDL#C9o!k%
zcevCHKAih$KAU6z0fubYE0w}41L1>Cc2PKnJB+UgQf8cC6T7mfk30fEGTWqcMH6`e
z3R?gDdu>w_>sb%=e%r5D{Yj5CADkc!=RY|X1dH>99Ks875qGZFFtAB2m`&qviD?BQ
z4}B2AOIwl#5+YM?CnH7_xHhuZ`8TP!KqiAv>x3XUqfDOfFZ}%>Y-GZB@P_<8&!iWE
z3{>;5dpP1ZS9bn5_-si2pr;nGgN5P<ke@H27l#R7+HMvnK&^P^glU8>-T)xWeBMm1
zK{C#Oy*HVc-|us;fGPLoN(p-7NX?2n>+ufzVGufUl@#C3mYD5Gx_$?X)8NZEZdUK4
z;ru)`KG0yupx_#U2Kp~xGaD>=|0@F68-(R5BQJ{aCo|Q>E8ABK0d_>>-uJ#A1Q!?B
zM$muxM`5Hv6(XNKl)%dY(6v+SRX}>bYw@;Bg7XEtSM8mer7vo3mj86>3fbooBlv?y
zE&!}#P=ECjy&MntCe7X6i)#SO`5)pvI|IE*SG(%|p9{!ECZOxBMYC2>V*_@c?|}f4
zSE^N={(A-04|>h+d;OZDs<;5(UR(IRu}_i!iFLg{Z$IG{w;q~YT^Xm`9B7O&DUrt>
z9*o2VUMe^~H)X(U3CHVRV?8wb^EdnM@ck@tJ#wkM=GJh$Mf^G&&<Dt+x({6cdmD0(
zcLymmf@Er10@KQK@v1;LC7%|>^!EPM<SxC&8|Hq%m+54$fuiiY+n;dD4x~h}@{}^8
zRywGm<)o^U(VnsWA;Od!!d=_mZJk-^#%&(Y9Raekw*#L51M}<z2E2iqb9HcxI|@{Y
zr#p|`E1Oo@XA;UVJ;uoa&WmTOi^2+;X@NV;)SCxe9QkY1cn|4+1=u?66wGec+oTC*
z&GH`-$RTzsayeDc>;h>u_L7!iCy-uVCgkQEI5j`$Dl^mX(x0p~Wh5RmSGM0MZj#s<
zY781mY~@cdai!q$$^=}Y@t-Sj+byiGk9jWmA0a*xMLjo`noZ%?VioTb!@3CrM?|qI
z@ZW1-fki$ne=tkczk@24dxn)`X`=Z>EDe;z@shDOg~fd%&qUNqNVLRAw#~{-=kF3F
zVg5)U4d^dyx>@lhuZJ=*r;H%+`)*`N1HwK0begfS{u_55j+uex|1yKcdUt%3-k!IP
za2@@1{FlnU8}X%8uyORznuUmu;SlHVy#RLmA8~GCElQobKN@HTnw<Q92%w?K@NB#B
z($2mS_(+}n;)4IRaL6*Y{avrIqy&7P;Xi**4MBh;H2E`_E5Rv}IT80qOon1JI!h;h
z_f9!$n9h4s1}RrMOYg6sxxz4sAa9axdaa@UZd>St<`)FxucMS1s6`&NJw6-R9$mF`
zaPRC!8&COleH6amRU0YSVQjF^(~5>68`HZ1hDj1<ONlF2c0h-R5oQxaR3gM6cQq|%
zOR=pz7WRy|22U1j;p+dO3)B!&D?j^UU&L8!0hv?m*ZEA@3vZQuFa+pHo=lAgsg{ur
z{(F<45EF%*=AZ~Rb2E<SLVAHOu%OJD$A<|x`d+Dv%z)zKsBUJ%M;1+-cv$`W;%@kp
zna~;KJN?PiRoUvR)0pF5`gEx$_mX{`jjQHLU)DUo&+s)4pyCjn2z4X9$EV@h;3&)R
z$2(OCvF*C@r%mYkt@T#$-{=CB3ek`Ejo`Fe33YVh-*p|`mhX)HDnoo2S0Yl7S#)<)
zwcf;_YV<fZ38!{aG^Ju9^sp+PGSm4#&g?bD4D<rNQg2zT9D?*y(zBkb+_n)SK8j7q
z6183GdVA_avf3*zFAz#LN4^*Hl#1E?W|el=u^*Hnv?0#WG;HLx2VPR=zA>d#JDY%I
zb$WiG*t1;o>dAe`-Wcuo8`d2~VIpN6(p)%YY>@%gyNAqr=C3D}F9=o1Jwm6hb<@IL
z0aZEcou&1RHF;)<xZi|=QKT44+U`X3$jdQYo|Xcc*b1zGHPzN`q?}_>YwMaaK@F3_
zAl3i|qg0NrxbkYEhjRWaOg*nvL5>)hLzY5q-92{3Kf9$Xlp20Wt-J($_Y;qcK*w@m
z3o}GImVgQDm^$wZ3i_+$Ah21#2s4%?AX0?tUhXuZ0}%A4yy$PHKb|9!7sRxyUUW*v
z37;NikSy|Q-ft;QWqAepSuR2EC9Jvog2Mk6lvwY>G}$Je*FX5~j{Tx^()ia^z$M+x
z(~#okI7YE%BFN<xc?x}c*Bs(lPQ08iZOP&qy?T^rFqWBWU&;lXRnRo!!ObB9eTAa~
z&NrBC?e1QL>whU56~cc`gBf*RIKHTQhq?3Qn~Im9p)eSTE)R2pd?xm38kz^CbMsQD
zoa*0#uGMOGsn`x9Y?6uy_t1lgsVPf{v!1#x;bCQx)SH0aLN7DO?bob3s1COW9_==f
z$4avt4}libDWTuCUnX)ysu{$#tDcN~6jR^x51H`bMQ)3&1m3K3YS9x>pgft18xA*1
zxV*p6RRg^7_P^x6kni20nd)ON0J=-sqoeeNM?V`d?#Wa`6y;GG(E~cfk@t@rnyq~F
z(<Xy{p7qnW0Ah!W`bNHP<B?pz`p6tqS%(CR-L0DA7+~14KDTB<dzTKj-i(-wGu|pL
z*1i*?8MLtQBDZQQWNUtT1s%n;!uFB44GbO&2?8k=Rxh`qO#qiCb1Pkyd|50x@k_4&
zEA9JTlb7IdL?I^XIn7jboPno7&TJ?XC-dY_Sl^R_5B!l)<+WHQB~DczTxx8;_z3~r
zma&X;>?bWTylrHj{r#h3YTLQfhi6F=oWjU4?EN1}$+F4zR8<MPsNGm>ER{T{95m)~
zbj;qW*<B1ihX}hHL#Q%jkC*TjiU)CPCfT&SbDG#L&{DEyb|PIOX3mrWLh8CvqgAA`
z(<|gUDeR8Rmt$FrZYRW-p3{6FI&+zn018@?r#QRy47fNJ`tYosw<{lK4Yu37Qkou|
zK+I$+byOC@az*nrovkd$<#=x~&ZJg2j9bI&X-Ioz9SD+o1QBaimk=-T{&Ha)1&}F5
z_^{ONW>a@!%Q>RWmfB%hZDCuFjOCKj#O3CTtYt0$a!poPm?i?Z6;N>|5sovZ_dq|w
zxC&W(2}=)~q>uD!-yg>GULykwX$m)^*Q%|aPcZYo0#5ZHD?jKZ^Jw;5Cu)(<Dr}qt
z+Uj)lop;X{`l$PY4;o>!V}tfjiTQ*^oDCMtLZof0jiufUUkw9iB?`1yxf`c!*YBr|
zH5Ab1j;C$yW38{8pqKjwqJKY>m`Xlsl7UFuePCf8&y$6t%tXNY(#ZP;(3T_an5fLm
zJ&j!cGYcRs9`E$|!=Ub{ItnV_)jl@e)o;Iy;23Uz)E@UHrd9`5GKourws2`sD$k+l
zkydayB-?jJf0$$SQVu2Z|ND{;<n9uM;!cc)F5$Wyt^K9He=01XJ3Agl5+4Y);tjT~
zaM>B}S^BPR0^;Ul>$u{b#`uwO11kaIvkb%a<5!;Y<}F&{S~JD6N_qOKNAV8y&v`>d
zkkUvm$4?O%LGLoE0PmLvMG@vzw6J>G^9g!*mxCN6X){UfnoD1)%zokl`|Xyp`SA@I
zv2XN5vbA(GYSrI7NriL$RdygM-Im!_A?nK+vCS*xTnea))X5iwh}p*67xcUYpZga1
z#E%HFLq5h_M(Kil$?lrJ?;bH?U_O8ko?x)AU$_topk*J=NS!<{(7U+82~XS7y|!<r
z#lH|a5b0W<P&Qd27*zO?;V784X{~QAU^F?&U0s&6P(wg1H~CwqFA2nB1F2Tavt4&+
ze$B;sS*6(je&MrC9N`Y2Iw>ovL9Kgp#Y~&)1&%Tkp2SNcG&FC^i4O~2TmZ^@AJ-Y(
zB{#JB7CB#j7h9FLocCIZ*n2aWpqM64GM;)lf64^45Zebjnbrx#R@<K&HWr`M(GYs#
zztWLPgNnUQgQ<yUI#2ln+wwA#=@E*W$|TPCO&>saZ|?-WtWSHGfw)M+cphZyo9qn*
z)sKo5oJ!vkQOEnp%+f(^(ps_eFuPE<v815<fH=Cxa^GQwU_@j7;k^!HF@dF+rD|O2
zzpdENG`WG+=Ka^ruOdt72W^rY9^9!7NRn{@J!hE&G!@nR=flKYz9~W)mt!%)oXsp)
z)=rjo`@NK92$!-{NW)ol|II;oz#8&YaxPVOM{4>V(H<Ipu@u8vZnY2#`zg<NOk3ZE
z$TMZ~1`q05t6a(`TZ*1p1sN5^_K$`Fk#czL4U#~vQ+TCpr9y=zh7`>&yfEClDfa9X
zD7Ku&(dksL!uMh4ZxWD!>0OBK0j<gpD)yp4b3c&>{Pt=wF!#J<yZw!f<DM$MlsIxV
zDUv0udC+4;NPqIJE2`9Jsij4%(qbiCo7j_JzLRg)1DrHSwO7-6<}4zAkX>iA1)1Q{
z^$ETrb$!(T9RX9!I#I$6tlY?fj>P9|KENy=___DDp0eCGzu-qUN?TodQ)||V_gtWG
z1ox0cN(A}EuOl0^>4O*MidGFi&1BHy`4zf@Vqygu0~M>S?(B8$)`x=I1~S|0JYP!i
zH3A`s*c>mp&>uQqyc?I17u4Q*cgG9ohp^{C^JD@(e3>MOn=t&7Z;L{tUZ3#ht)3g_
zg09ZRS*M2GmH%d-PIM7uVb|E?E6lDYuiiKNDqx3CZbkEP$*qk{oK*f{3PPkyrfb-c
zwWv>!W_U=}AIbY@%$t%b`d2Mygj<UX@p=9^n3@>U?iTUYA%y9=NE(hdul2q@@B<aw
zF;D7ggYYg(F9RdvsV8pTQk}hA=9UbmQs~DyUSBMOiW5gx@Px?5JCJw_n<?^6m;yC9
zJknE~xiJ%<?*Dun;lcZcnvk;bp*YqCxgVB~cHQz0y~_>pd@!gP7DsIRz$0x_0+`9|
zhJP_I{11cDi5{Cfw{FcwQnF}C;yEytg-rT^ZqrNNvMKPy^|(M3_OoP&0ILi2LA>k<
z-}V4?AM|Efqu^t=>cgyXVc_YBxvM*UaiwSBfzwyH&k%0v7WwuyONRuLY=t&Hc}&o(
zL107yk!krSBLkP)js|@04j7rayXy%w_qaP8q}!qB_0;1sbOfMCF9PZ5X*s)*Y-VIr
zq5-j+tzEJm&GDu<a|BNgzB}s2@w-s;kxY)~>W&qU+G#~O90ts@VYb}V6#WBzkAK83
znGgm}LR}|E?RV!Y&Com}uSm+ub1$7)tVv(r@4z3D*rw<{yr)tXa}@OyaCl~&Y4k@x
zP+*uK|7C3$J`+ZzQucxFngst&$xB(f-?z8?HqY4-0q#P#P}Xa0sLwAEpgRJaXsyB_
z?j$Z~3BQoyOJi?JHgc;{-$Jd=!1ox@^T7-Pb)AT`8Tj^uV*RZLi3h`5BDphUAe&w4
zeVVkILVQ{)G{8~bu7SN&FJna#{n1*=$8}Zkw>b0&40~9CVUJ=er>|IJcT}ZB-X*cr
z=e&N2+E<8G^@!?Fmq%qG2rpw%*bDCiPth5JDEwP}dGMf}-S}&?vxVZYAHFR>#&kVt
zt)|1ZevoA~>QE{*;8kfs<qDD06*`TI8(Cqz*zIYjF#DeSInXgh79IpbO@363_!v{{
zC9UaDYy2_4=JZ7owsBg1`NT&dP4gjJYpKSw9jf&nF8L3&1!Hli8S^33^JKH*1r-g2
z3G$aXQ6dR?lPGOw6t9y}@lE(zZg#WG*?oH72x5W3#q*ZcMelm8bZ|W%nY`Fo5iARg
zGmPuq8@y6r*h$)QJ9_#bYw-ijV6TbR8PmA0I@e41TTqvSA<v+vfTrihxWIlI{vbkm
z`R?mIX`rsoCohV$tzMdc4bB2(YZLeI6H8UIQ^b<zGh$0$Z$BDqSzWfws22Eg?<k3u
z=55=Q_4ebkCkS|=Xc(r(Y5rxiTet2{ZRH`q7gg?q7q-?sxZzzo_s_KxE3By^g!JrU
z%!qoA`^<2tn#6fWoX)x~Ryb^yG)%_)JF_GJxqAm#Q=(}m@USgdS((jROEq&IV)FW`
z9uG2)wcFrh3EhlWOH*`<%(FpNjlXyX-wn3<4KPUS_<4evFC5dCTgy2bH>jK0`jJQ{
zkftk;DhbOKpYS|T<-g->Zkj~Mq|k@+6_pUgj-^vrcm}2fJlQ&F8$#Oqnro-wwZffo
zt=dF>g1RIJ*|;#i5_`Yp_SK<4;0BdeHyR4st7*B7G~az6CVV3)i7gX4d6hN1$jFvA
zc#7idtf8}m+Ct-$+P!nc+_@({o)26^>4gFiG&IW3j?22=Ij>>key!qZa=b^$KZ2r4
zxjwP*MRfiiIZ1BMM(Sr{@qCSszu?5q+4KiQtc}$QxCql9^hOL4IQJ#xLP#0OlV9NL
z$vQ)}#CUx-0S3EgZyBn=;HHMmcAwsn`wB!P)!i^1tf|@k{Ib~<;}t0K<VWOLfJmKB
zzY^cr6b*{+L2hlXn*^WwiINHJT9zc@wEk2r%m~crMFpbOeoO=>{z-*#vEJ!)u;zOJ
z06$EV0buY6P33Dm-=ihn>OR3$-LwBqNX+RCy#Uw`Tgn1;$+l`_o`F;QH<RaRx4`<E
zG@=%R=@AkueZ~yOTesWtqE3FQH~HLvYnJh1;=yo_rVNwTV=g#B7nNZOU|^zhDKwSI
zA?*~%YPaX*Q$(_a!K2MY+WPwtL=3CuvD9%<02%=x|3dAIC-GmJqNSgsO?dy!F7(o`
zdv{D#(6_FWeKmU2^mM5vrp3Rd<}AL~xu+ne{?Muuq=9cbTdD?CZ=Wx?rppbEQ^H&`
zYQgKrxy;>7f_{8CzHa4igEeYD6Un64#qX;uC{%3slSZues**H%Ux52v3E^(ewazp;
z;0<~Ld?^vhI!OgCy;^~O^L;D5As-n}{4$ckI~m^-4fD91Pr-Z}WEe4CYK@aE_13mD
zM2e!J14bLh-1?k{4%S;s?3U}qe=~dv%m2#YKdq5D{$p}YPD!0bNko-!!Qqsn>xG~&
zhzO^DRFDn%3r=^-)=)({94;jWVF00GwM!QtWL8E~pvC|$#IKu;oB+<71hzm*;&++?
zug{OTP|tVd)0YHF4>mhffZ@ME74C<uEt9H*#1MBi$Z*_P^K`&x(o}^~CA(-dQJ^=U
zY*-W`w-?Q^3QVGpc94F&1YmqwdFU#byUH=C?ch$tm>VGaW+NOBSa}&HovP*IM7$%1
zu;eLm3GY3ZMz^A<^*~hUQ0kzt=R}wb1fb>Vxk9-Jj$n;qbh!<TV@@ae7b8I`c~i#o
zhcj;!5iWoVrNj+4?`VHEr}laVh~uc4m!AfS)jPENO=-&v0xoz2$)>K7#%wLUF7a&p
z<ZaZPY<Iu(>e@KRty}&l{fQDu<Z}|+x%hWCmuz$+oT1<30SJ6ym{jlKR)x!mc6L*D
zZc*SCfjA^o08H@QKCF4;TN?oU_-CFonTH9KV2FJb`n!>JM*_(cj3rQ*>&d#o<sgN;
zAtRZ|W>Z7Ru6OeuvvyAu{JsRyoh!5YBr~^Rc_op8$PYB0&7a3qiOz-Ho!xv#5?QPV
zxU6*zqD%eU6dfs7Bg<Ph^JlD26crbwDzUmwwfccPGS?PN!%)E%8rn$cqyG5;)LS}*
zkfJ<ml?0KZ1Bd5vL3EHyUYDyyjg^Mm!@{22{wzLsvf>CH*|CtHAiqQAo`(ZJ!qCes
zJjC2_J;yj8WL;Z7F}_4Zp3aLR*&7BQ;p5Z?!(%2gt3k$JF55-yef1Gdwy}v+FgzyN
zaKbWDolRa1WNa<+HHR7_jc*Y#Tc^q<EBVkPc~y7hVzCP;2M|9)_bc6|0+H%d+}p|B
ztz#GPKFF8$m#oxcf<=97b*(Rk3i$OHFVE&l)jPy4MdKc*9o|^yUaD$6>CGDsSj|Y7
zUi~!A-%_fa{R6v#*3%6M@yF{(BVRsep}2?{3la#=(BoUD(`NIo1oL+PE$Q<m0zAa(
zYpMe&uIYKsnv>{)oxZM9YX%tvgzCHsx(cel3-AP^S0VoRMzs=8a6*KB`HsTV(5eBU
zTTZ={g4jV5h2@x=%WeBmpD!mGHpW((v?lns5-6c_pmmgag*B)z&QnunbDARH4(*<d
zG~`f%>DMQs&K-_)afT#%?f{j+M$8My)m-xJ*5M?QBN?A8o7dcZgy({nq-$^Tn&8-b
zkw0EFH13Omd2Rrc)|uzLXZL&ou2_r*aa(!iu*kJ)5?oP~7FM)Yp6g0{-Fg8?zvVY8
z0iUVz(tOIH$8bH!w=J>>@nQ(gIIXvD{qb`@jU27hJ9osf%(`Km`6<cA*C$qoL!6#h
zP{Y<v>Kyj?A5X=-*xQ9jOStNdVZ&82uf#vbY{Z%G$Sk`ivfPB`?0p)uPBFTz4$7>=
zlO_jV3hE}6-@mCmD(Y3b7RO061&KKTt+J_s(MWHORJ&mFjwKa^LO+7{30R->#qE}(
zm%md4-o+2nYS^Rb!RnBzEETtj4b(kS+h*X(@qg00Ef_pf3%bvUz{*39QDDyNsK2AV
zx1HUH9mir~fB>p_Pb#K-Bvp*^1G57by{w+*eCdc~KN5gMwfKcO=9D~su}^Y2NI&(Q
z!SfrzwA^_`VOvlub#TG269f9<#MU2%OKX2y@W+ov`EmVD1JNxX{oxKz5hlIZV-xM+
zRNY(1JuL4RuPy$MMko(Y@CY(HFVSeL+ppO?`w1Dq|3}$3nK=;Vj&d@|gASd9ew3BG
zDV!ytdD`8d3KdQ^(_gsW7PchITRKgcsl-e0I%lw8CL+zasO*##jc5gM$v*>>T;PaA
z2Td94+_0bCc&{)<vmV(oC7JF1_KZL%aLJF1WS9(6RNfKAaSa!gU`}OwJgfV#a^SAx
z80Uqd27tB5?LwbY2FHrb2chea=00u2%Ln5{KDs@PtDs+9HoX7=6IOzLwAOn6gXHD&
z{b>jYOb32~UI*==!Yd``48Is45?I7c41KcC;nVEnl<pVw<G0ets@Dr^g7lvKBK3;w
z*exHNokG8%V-Gq)bSc_~Y_*Bqh)&PaVXaenNZ?dST>UT00H-P_Q*-F>H3L<=u!pLv
zBr?MtXMQw3K}tix)(KWlZIbF=6A6!^^i?5u8=LN1U;25|JmC??8OW#OplcDajNZhp
zVtKJiSj)^762`IUf30|)2e8~zlLNz4s4CoY2CnW{PK|;S(MK)%4B0rTIE_%RqB@4U
zi5n-<a)VYbinUZaEwA=d!v<WTyZSh`k+me6_HC=pb$;p=73`F_G5*o8HaFkC>k>rz
zQYXtMpI%iuKL2iHaeySIJ@hp)NcasG=+O9}mtSA*)WyTF{y9iNmhk3zcjpHah;<Bf
zyoZvt0hWF=8}*ThIR7k6cbA;_&zic??m9Y5UovhsD}JX{d}93M3i^5F)%$N?d!B`X
z6nJ1Y3Gtv)zSo*W-BG2Kxh{_q^p>Lck9?qbC{t+FO{z0i=XhG~6%FH>S!)kBWB3r}
z19UL;>HF7e_xw6WlyZ5|UfioiqAnUU^OP>CSIq00+GOg}+1KLutY%M}N5IJWz`Ix1
zwhC>cyW?sBaXbUDM^0ap+VXF|b!+(0A?{o*JbBV6FA&6^k06Sk`*I684Qy8~!X_z0
zhK)bXWDx+Uy%WP}OU@o+veDd`+3MH+Jb6Ti0wngDl3HF@z;vy&TusYzdJH}`n);L2
zR-B&wRdLgw#9!((DE4XxRYFW8QXG4TjYJh?++3-Jcpx>`Xc=gp`3{(G_*m*@un4|5
zW>w<Gho!n3d&@Tx%?IwXChRQ%QdL>m2*_{O{4Sd!!qjk?8XlS`Qxb_zjRX2T=3Y#f
z4z@l@(V`ZAx1>hER>K&G>J>KTLfGHndS3*64Z+@YW_BQ&bxpbg!M@wA`|O+kFhloM
zw7({<NfIj6r*wM)EK<p&NE?EwD*S~+QT8#AcCow^aL!6By)@4H>}Js5L%|A8c*41v
zs!bx;L&Fjk3u;kTX^W+|FH#e(tTzIdb;h|Pr4&GaNea$V0*YQjQjR?nyf~jW>*g7I
z7<H2p7uu1{la!z4vRH{Od5o@sTiWSUT1IgM=t5?J;z*~E-sq!bsn}8*y5t_YJTsx<
z-%Z5u3%{NFZa<LdMMr_UhBG+a2;#NnF2dd+E)_%Bl8-ic6}%E&8MRG~D4GHS)baY+
z8!d=HG|P|nL<L$u*t4em<EUAX<(i+N?`H>CY)g?<NiYAjmDIYC=Gv_Mr3*y)$D1@{
z;JU3Espl|gxO*kX(8OeBgJaD+N_EDY3q;Ii{fx5sM(~QuK#%gUF`eCMC$ZjXEIIO^
z`4$BrvmN%0`>uEf(V1M2k!qFB9h>}xL9}cy$e>jql34mnQ@c3cW!F>*6b3y?oSI?U
zpte@2%x*;jv-w)k!@K}CtwO-JzA-*EyB_~H8ARS7aRftJ{ANc5oeF`E&fKYT5;aPQ
zmlDz3+z~!2geOh!39{<U*<&hcpj5a??#s+B^azzDusuw*=6YY673Re3EOL~X!XRXz
zm(f;^_0H(?UbTus$yaBXIf4GzxXxbv8WSr*>P7Y8P+i=I^*w(wbPU=y5Be{<aJU9A
z7KsUyV9m{^aI;XNHWMm@qAzBtMv?L2Aa+nb=87G}X<3lZzRrLRVQx$f+ZDcDA$>Rv
zmkOXjb$FLa?3!}#fG5Xkh%D%N7_L7`ahCaJ`<ZCH1@oK4wpp+*-(^Y_H#`UsyY*uk
zsRlZ$cR+KJM{_%<@z^~f{B7StEd%h=W6m$HT*34)E=M0Pd0sIN7PD#&^l}lGN(&^8
z-He6ibkb?hMsQ8oxx~ll1_Oi?udOE|A)g0pL{+pVebV6qrK1PvA+o@K6^h2%S<zic
zG$it_M+}EDle-2MgK3&2e@393+8$$g!0lWo*9)Hpn*_-rZR)yaEOvYD1f@essFAK4
z$sdFqqAgh;t{AV+EuVV{+Za7qbS>uZZ6840qHwpHt?60eOmO&mMR-I}c4!Nn>ki*l
zW<#VW(uVAwBE=gRHH>@r!o&HvnkUGUU`202T_vOp_BQQ^iy09DkOX>NsqQ^sT-B>P
z($bHb;R(nZsDr<>!4c_T4=zDJqQVv*mQmy>-=m70#1CcV38ADYfnw)9M@O!eX4?{R
zO{>i&@7n28H=4D#qJ~?^UHq_)C0XneVL9$HOr!9};q2`$1eT%}l(~Rfy9>ZswD*HB
zAYIt50sy(<m$G3Y6a<$U4Z-|q*+CcecFOk0zshw)$5ho-Mcir4?CNaBNt6^Az$sfz
zw@p}|kP6cz>ae4RDGWW9dJNP&zDINR>Gh#QVWRIJdxxplt&xL0w%hPpdXBdER{VVk
z(>fafcw5BBj<FPuOK2wK{rKrl)-w>X>Ryd0zJ04HW2GSQ(t0~=n_WX37)aZtA47-0
zki&#bpdIO;BI%w2Y6zi>T{XnL);dX)&m2RYp>%icO%M9+wN~!LvG_F7{qyC$kL?91
zXd_#&mXf-DwZ)?K7wYSB5MPjlES8F<Ac2QH3*Wx#)3SHJ{vpWCN`UA$3ZMG^hK{ae
z>SOInrSYz|Om@yMbltx=Amj)J)WX4l+8|^3hogqvfJUWD=2#z)Bbm$9NNoP;OKO;+
zn3im&c1hFBjct=K+_D_`cW|a0%bhx_Ts6c-<g~r%{%n_cALU{|ox!0KFiWPOu^-XZ
zezes>fd#R9%k6b4Z1s65%<<BU1sMSbH^E}3LvOdod09_q@A~tEP8iPKrvNU}9jm#S
zNM)@uV7PNrRnGxXnOXgtXZ%Z($P|pmP50VI{M?qtVLX0kM#jT|8Y45b`2gl(RI+!9
z(V~`OYjH&QQIZg35PhgpYs#F}N%iJ&@++ovz-?z?7=5$AN?vKmNSsYnr$6pKh0-8m
zm(I_A-+3+^Kk%K79yDwc=|iw3!?#ZnS~|G=jE-bAwZm<669CuQexIT(B!SV@E%irA
zIt&y&gjX$<dIix^87`h7-E-Z@8&KOB6$x>J#^btb5MCH?Va{GD&V5RAWyCJ*9#k{M
zJjYsPZg0%;h#Uw5Lb>6Yz`?foqn)%Z!7sJ!g|!CCh4?KguFBQIVho#x8H!h6bJ3%F
zrgyDW!6uKV#p2>W{po}$RGR=`KqtP{!1n8(SpXN6D?_+usr&JW(;$B*rPm~-*V@G&
zZb?Sim(3FP=O*iWi!pTjnISgZA4ePyDlop?gQ6Qz@A+3K619<#FSsWi&VDapX#w7m
z&30EshDkX0@Hno(01u##I;C}i=1`l=U3M_737~RBI@;tC7}k^^QX>wUlFjkL;<(Lx
zFm_;bhy46mAtbo>gTvQz><7dvxos_V9||ntXas#rCbaS90%+}I)mzyf3M#~;n_$Ds
z=c>ob69dsbdRH0#!AVGA5&#^!rgybS&BffRd+7(p;ep?^7QIdfUTAoSUViiTR8Kzc
z;x&X(?v7xhFzMFSlfIZxTpq<a)&N>A68w`-t`eF0k*h;pOq1q_^Jk@n{qMR`GXNB#
z<ISjY528ga@f0`nw)_$NI~vFhsHVH<!};}(ev6fHH6FBJKE&#4$j<J2sTiJrdow^4
zD;pM;H^+P2u3!7414#u#vplsTbi5#RdFrIR&(X_B!RI~7)D4rbEb6UHY>J>f@|pa&
zi3=Zvh1Ztq7DRQV8MQ*Sv9-#s;>plP1=B*Dp3^b4Z-Iv%Zo@Wi-WIQVi&M%Zf5PIa
z+R5ZEDqwLqvU+mSVcJT3lLPj%dSWwx=xGID)T3egj2Bg{jD~IM$-;2$qh)o`n1eus
zKVg%Ja_vUC_}FTQ+Ft1BS<?DaIVK4z0Q|P~-F_X|<gfjT=}oIJ8M>MF+5E)0SPkgm
z(;&LCeYfF?Ew<`AAy#e4boavE0zsG|q0-GB)k3Q$%>(b-eTh$VJ{g3r`C2-n&^6Qw
z;g;GgwD$b@Eofl%*AAM@AGBLKL@QSh+-)joXAKDRbUlz7j{l3D<IXyk4#xd?x2CAy
z#Nd#ycWC>Yo#Wue#y_?L7V=ZpJ9;?gw%^d0*e(}?pI7Aqu^1FZY;&Sv%?0i0a!QFz
zzBiiomJCv9*A*-sSZ*|qUZG)l4XWthk+21S_M#5Izp)x4O5uKYFlWnisw0PLdtCIG
z^<8i^QirlmG&BVmz!x{N1XA!H{#A^Lv9|?5YyixR1R*<d^ExZ?Ipr4EdXC|g%%;06
ze~3H?A%T@_VtO2;JsP8>ZLeNlL|-=j1&+PSnRoY-q{PmE!sO@9xaPesi-r=4DYh*z
zkoD^I#(El8T1aBm@6Ddwf!$JcNj=x*G@%2C(~Xai7hvV_Rl1e<+!wlpZ}*_bz$~$Y
z9-K4+9$c)(N-yTY4nG@VJ8*K35Ft~PQ|k5lAXY}9cZ;1gQD7a%PAr{D?pZab_PmSg
z-;zcMf&<9ylh)c-3WCb-LT3w4Yj7R(`o`B`7bc-%)eI(*2lVH-Q})t63bH=x;yG!t
z*|Je(x~N(*sE<Hve_{R^+jX0-S0lT@PZr~akn{v%GM%GHC9z@lQdjrUMGlx=5xDnv
zpAAbmP+*#DC1j*F44-E-Ks8<O_Ec?!hOBf)AxUr#%nVR-8*8Jk`Qw8v?qjCf5K!EC
zI(pn*%)Ca@8j1V{KrFviI9<cbSVn2O5cLcUm)xEkZX-`I$|x+qR`wl2!xe=bh~ONb
z^XxZ~y;IHc-qf1oFBXI3hE_3WtTk+=jpTC4>a&v{Q3Re{p>?F=taY^Ky=+BR*7WG6
zJ;dFcdVK;+^x3nAl3fPRqG0=sB^L!kq>sbWubVP);G7?xZH5Sp(=GU=G|#Tl>LXdN
zq@aypol7SolIKP<02mczT#Y{8da?JG!x-IndeY$z`^)CJ6&wm`NzRZ+lSexqU8*zH
z=iqZD752yf6?9y27$;jnX)D}x)d_)UB$R>;ocdV)jpZ|Ns<O;xcxJP?08Z>&!N@=p
zFu`HXtpUGA2KMeVjk<T&QIZP_v^HyvN6e`IDd-OUy4BoRVzK{}bhck7SQ5!>HaKAw
zy1hM-_rk4z{WDry&`=bGlRVg&=B^faIJnd{b>z-*NG%O;ms02CcQs_Pd5y)N@7>`Z
z`~HSQT?&<)9SCNp$%dC?$RrHVcoNSz9+@u&EzSy~&(*bWo)E!F3&~C*1seB>{CCd1
zBGXne&8;jEmiKh4tvmPkcEGF@Mc;8TX|U!sHLq*X-~rJA+l85%l=P-mlI067V33fW
z&5vv^LYzU-BINvh)0V^&bK3OovTNMI4W5r-x5`tDwf@b^fX_0##;C#;o`L1HpsO<(
zbf`9v6@{-(HquD~EKj!P#+VX6KTd&Z_1RL*!TS$_-1c%m$+R_aYx1CxtjjAJ6Y`H?
z(1exm6aDrj=ti~Kl$LVUC>z)}1)gY$3L{hxdbU&mt)MDMH{K)i8Y;|_^Wk7+&g=G*
zFOWY;z{q5E0J@%D>dj-^|KnBYFqon+j&Yv(96ML-Bs8XTMne?@Tg$5Z?CFhpqC4M3
z^H*<&+<q5)&>NRK;@)8e>#HJ-$FHkFbZ9{jv75DUHSG|qQV6SY?Z(%?7LEk9UzKZ8
zB&$thM21(g%wx<VuB|pPw#);P{^)v{W7R{};eojTdIz9Nna%WWqEfj)sf+(nf3+lP
zNTfCP<E1X<w{sk@w^h`rXMgpbO$fmL-Sg^`A<{phwzrcEGSrj&z@_WBcRpzeGfy{R
z;I4d2c=rbNk$iMsX4E&mOQOvC#aU{UFx~ps>D=<QoZ{78F*wWlG-w?j`NZ85E9ySy
zVvKy;`;G6l3nzc1KsSM>>oigCjNl56Wo!v*Fo<yO6W&v2>X6|`fXSJ>2CAwDA?({p
zvHYw*M?Y;wh|u83Ux~=g2TUbnAu9!5%e*^T8OKozI4rFiA^5chU~`eM$&=pXr}qXD
zX?59ZAM^}}RhJw*!yz;Po?F6A%DVHxnJ*0--z$Ijyc_`Y%icw<N`uEq0O;Ua5RJw>
z0$GgDbWZOY*{Etd>&fXZ_w$<KB=||05)agN9f-oWt^fry04lXxLpzxavuU}-8BTb6
z2S^qZO0vs+jA@}{^T%4b#b*T5Y2=V2E-4h1;|Xqwc6qcsJS6z3YF^>#y}fRlEZur6
zz==!W{2YZNL6w<X@#rTzP35{TTGa2~I~?HW>Qm5EETyXvCZa#T(pYgXF}<VV0xfv$
z5aZh_1*18PUIMnet$(l=K&lm!*aG_f>@4C690?wvb#pRa`xWLI8NvBB<)6M*nXpZ4
zuO5v}PiXQ!fZ7Gtw9>6rNJlaHBiba8)~oE}bS}=1Ay=SQ=sveQJqv*?+Scc3U~j50
z1My4hYTT6)@GfmHeHsEB1qI$<s1KN@pI_4*CWtpK<*E@Ov(W(fLy=sK%lSP^L5^<Q
z0U*1Y^YUY)3m70eaTi|S%kgngtZ~af^iTV!PBAG`YrD=zwGe@uQadB0a?neRDXqxZ
zmktE>S#-r@J}EO!lTj3uqvoky3H@{|t@ogHa=;t?&Jz1*@~eKl_UNAb2mEs7V48q}
zfcHnXlqFC@P$irTej<OW7!5$MF7$5^{rcITO~AFIu*Vg^X>)_VfH(&zu<}W?b1dti
zZam@wVY~sEbPIx5ZmY52v@XLA_%)dUMsdJt-%B!kz$l_h5H5|{+hg^ANhlv}V!mfX
z>tun>EKG7Q*23O88s5Fg`6nqtl7r2OpU461cda5x(<-Qi>}d^G;5XW&Z(bm6Fi5bA
z)GL^uhqryo6nnLYZ~--Go7TF&X06GNOe(`cEvt$&F2hE!x4C6nG5&vJ_<4}K#Z<+5
z-P{q=G7(h0dzn1U5agJnRC;6U03L}_n4E_X*Jnwl|G?b~+dzbV{KO+*hMN#wBO&Zk
z<@4nMbcxnV`UUpCq@qYxXUf0mCb&&es`cv*`M-v}i33_VmhfKx(3ss_7N$xtq0v{Z
z|1x@SWd-0RwQj(=wJbmhC<Q3hckhsCCXLi|VouCZ{c}5P8nh_djm~LCtvntj7O_7m
zrlX=cNHIY|$m)-F3%7^vCD<S#jjW?zON_uFLu?Reo$^p4B|S^1>w**a@Z89SQj5LX
z5GMmUu1CMiUnLk=cdYsgJ?(<l;dbP+6RRaZO^nyc%Z#qc*My8I@^j;u>&rz32FWsQ
zl)N8l8hi~rtTTmS>2%pqn#zIH5chB2oIN2be=sbDObo(KP}&6-(2o^n-G!2YnHcXd
zDj5qXfMC>KNG_n-peMp`Bv`2rH3Fx<F!`j|lLP?$Bm9p#y<bebRk~Zq^tn}>QNgcg
z0PzbUQG#_fF`Qjs^wI9n%&YSUA<Dpf^dX@fwXg<r>FLkH-z3&0NcSd^!LNzBi?5Dx
zrmQB^+SWzJ96)(e+6!*C1<B51d^v`Fw~P!f?KxMReF)PxETCe$`xn3Z#=CqBxGiuK
z{8l`{GT$<xBqf&KD=g%W_1Ww-pefjbhYQ|1&eF8cSFXfBQedb9C<r+LSq88DT_*86
zLzs&Ij^R0pX3Fd|<I9OKFatoYtp>EqR$ZLs(AbZF7IYZaRx<6v86XB8_D2pW7t~dH
z2Uj<gRBn?&pqm2t9zuM(8@#?{xw7QRU#~QB`jMU5<PT66RoTV)T;xTa!$JN(!|zC@
zJWdUx!daFQ$0LX3wwG-i<#__Cw=R=YGd#zPS(mx@m&FF|E}^4X6c4EWRRbk~;0e2$
zmlw~cNf|9(uHn>B1h%ClHNAc$yjV^8lI>#Wv5}=wFcy>a07B}eVv_CRH}B8W`{}c?
z>3c9P={?v8tCr}YE&Jn{$ih^HK21h!FRx6fcvRjgtvw@j5d_WKOC6oe=Mm-!$6@ms
zbI;C#rXtdUM2dO3OR5%}H@iiyW|{*0FuxSXSl@FrhukR{yT5nznsOOFF=&L%>B(v@
z>fm+N+aU8Vz|~pRp5Av`%v&U-&VsP<admG^%8&h#*~}Mp)!9ogYSA1d2O(y(zl=-P
zTYrsJ2coyagsZ>)mE|G&5_VZ}50G_#Iy9eh@E{}$OdjZ8cm<J#H5jC`(un~FK6fol
z#TM8<Xw%xd^SRy!rsX`XTW9Nem4>ZV7o~3E?D5dc8tV&&gFX1O8F1Eu%Xg4JNTezp
zv8YKJ^B`={J1XGxY;eoUk&!Gf#g@d{sRE2$b!E02qWmA=CM#)GhuRvjCT?picg=4k
zP;Ps69|?X;ayj!l`9VI*qo)t9w9R&D_GZ>P&O2r)9KZh}N0XOM@$u=_>+)I1X0-nO
zOBz(CKfEs*d)=jv(7XITC;XZ7cRJZCqFiqtK7ojRbuFfOm~;Ki=daltUB#ZIxrV3S
zRjpOd4~lr5Ub@qzf-QWvZM1fgKS~lxww#Aa7uaaKNy=o8>CA9(#TEtseXYav3Di<`
zK*)T{zuWmm75Q<8BPbTo)?Oj&xll>-W+LA6ijO#}g01_!l>-e<5GkXX?iOoHjXp<{
zaj{M4Bm|LDHXC0!T6}iof%)I4xa^JTR?1j17mV8osZ!uQeGoFpY05r!=yf<`rH@2J
z#1Jd_`KA614Q_DcdG#R=vma&j2*C)jk@#NKw0){qKxNUVXr=A%lH5=|W!PwPYJuCp
z^9iOko$G+9bl`=(npLMNsoyB8N${OyF;LnWHm>MX#$A?aHA$mE<GaG@;=v#{T#dKI
zmA}x|3tk1hkvRHlYyC~+U*%p-#N1QbN&e9rxXWSvt=sF<%P5Q&*qL2iXmv%E-PsLS
zH2%K%ngXz9sswGE{_?KSOZ1FxdDsCI>cskTqru%{4nEkkpR-Xu8h_xT)?28UzHlp#
zpM6VrjR%KgGGPq3XD?7ES4`=1UQGO7d)FD(bhh+0h7~Lzc(H&2t^_tB1VNNetQfjM
zK@!?tL=X^^D&6R+YehpSQbJT(L_m;834!dYumR~sP$0N~fly5h5L({z53tX3@BRLM
zk}vQie@@PsGxIyYnKLt~Mj#4z_(_ZdgVk<<4W$+PjVr3po#6oH;T|C2e~u9{w}(1u
z$*W{U*Nlo)4?@<tV-!{8#)pOh(-OL(d^0&q%r)i>uvI3>kJteZ&@MWn4vg#Pl%r5#
ziMxda_g-DGz&GdM^?L9zQ!a!Dd8PX`k=xh(lk4hz8*SZ~N*^9RJMR4)*pvuVi4S@F
zN`Z4*CA_jbOE;gn=n2iR2sV@1DfAT1iHwSB4WYgo16#XFzIJBVjxXGt)X+0Ew_TNG
z_PKsOV(5GKy$gQizjqa=quu2IZC6{{@wj1I?)J)I$JTt`__?hOHv4vi1u3Y%Leh+?
z_gC}y4=mN7%yi;x71ngq>WZT^nir=zT!+m8qOZyy6rRvt2jC@bt})}F%~`AS2qySb
zBB(9jG4rd`Iw=Y`KH$?X{Q)|R*OdFtMbT_!0n~RWLcM?9LeTihdIcc}cVu(%hBPhx
zZ#^FV$;te*qu0>(fUTw&?krbxp*&0rJw9R9AB`~yX+;GA_<+yhC{?umc+uUe0u^I$
zi=zJON=G6hi*%0|{8K^YpR2>h$mR-mzirD#rG<39t?ouW9#TcSi@?Cs?8bOa*<4Go
z>w2awpNQcvj$C8HH<I>iCB4KS_#?*<V$vqEU6q0LuPf36PY^T4SU0u)*`+Q^LPQuf
zJ?VpKSk3ww-dfaNG(`e{ujY<t8wR%L-s5DX5)%f@SW;jj-uKW9RMx`D^7FlJC0UP^
z;F6oNnH+Z1{M*opu__b`wK@A*=jw!Oq+U)Ywy;euekl_5m7Ryls%TTexbbzPQ!#R2
z+{H-}&kG|xnbT7YslFWi5Tcj8zrLX7pp=&p5zN{CXGjukWY$ZYz$p^^vY1ssILSPW
z<r57!O40J*ZtEv|DEiDP5X5cCy%)Q9g*7~c6e>T0po0RNY-;TFBIzkkC^s+p0POuY
zL6Kc1m~Hefu;OAg3#hBne99eItdU_9Gk_d5ZQI-S?y`=oO0y!cs6KsqRuh~{l$bol
z4VZEAy8<b!DFN$>*K#t_sXAs17!2XUMrWhI@hREcCSOCSPQ=W%v69x+V8xtkKiik;
z)#RejXGF7)gZ-}21R+HKqw7n~Qnb=bwL_XdSImO@c0%YzlAmpzWj(_8a0xv(6o_$_
z%&bYklU*&!br5lXFK2Z@>q5eO!xARKZE;sWxWorm9Fy8I4CVS(DL!E`{sl?)s0C*n
zWAY|xrUEK?GWl1TX?o>L=gJ9@V9Lf3%#0!n%cvatdMsz$a4OCciSjwFvS-yO#z7Ys
zJlE0&r_Uc4&xP!ED^$2}n%H<TA8iu0^mu1uhMLeUUy};Xts6)?ZfZSZ)XgAyhz5LB
z<_!E$LHAr`dqg-!dgz}Szp91_`bfNr0KfUxtfpPzUndq#-Uk6Fb_DE(@w&tBx&*gX
zSe=pb(RQlnAvorlXC%ig)T`~aT*HKCoF{A`5I^nCe_?bXxPDk3LvDE&1$GEfF(6L`
zO}oIMse+cE3!0|W8V?$gyMXFIZ9B3Sv?dP&fc+zgSlNM}6?`5XM4<btC3_yVEciQ>
zPpFBLCZ^aDhw?uKhFUl!o>x!?9|URG8iF#)j@Z;OI25#Pcl2(k7S=c!kcS&m!f2S8
zik*4mRqqM*w@y`lKX{x{4X;Z982fCgZa)+(i9lT*NI@y<p1G}6$Z8TSg&tl#w&srZ
zD`pLLsQagW>(s$kL_a<UvaYj%2jNp^u7a-K(2n5OR!!D<E~dEo1J$<ZCzApl%FmDA
zBKYL*hja(|>WbhJ3dOpsDzlSwT4^-}t>gp+Z_#IZrvB#ZZ4I+0@fIV%*9rj87%?!G
zn7E;xZ);|d-k3@>&d-wX0N$%eVoAp+th)GR_hoaDV+$ll#`YcaeOTK4s#-@&L0{j4
zWkH|^;7di>OVd3kbj?1CjktL`FyiO~YI{<(?2<lI>$6(3Ii$EXgtc39oz5i|(u0|}
zJjm<A$d+AfNYtg%UjD**8XrgiTDrVpo8@y^E(oWCUM(%7iVCMeL(sY7Kn-kvLR{cW
z084AaaJw8Wce58t<m6PE0|{}Sxs@-ROF}*BBVT~d$OQ6+sEF=MaSDbilQz72CX6o6
z)jgxiUR7anU2box{(`CI=gw!VUB^pp@rH^SQnRV`q>b&mO~xuRjG0za<-hCRN&-Ej
zY^coMkDsYHqFHzt+POWK0XJ~P#_M6EN8d+Jtf-iz6@j__B%QuJJ*w{tKD*!F5zGwA
zOG%V=$fwE!lM~@AoYLDpat_`345;%i!W=8O?pffs3foOKUJ!5A4uP>3eW+oo1|eK*
z*=i4M?LT0886%-}8n!t7f#+MJiU)x9ZP|!ahQrk`6fNUR5z-2-Pq=wmYInMd;F1fs
z(fND9Ir$L=p*Nv#V^2v0=({Dz$HZjMl+stKlT@^S1tr@&Zue)e1<X!luar^85eQ6B
zVBbu4Nm><ylYBBMzW7gdq}t9;-{3!;rVJ{#YTm2o(L@J-7lZ&aM%%-=r^fwE$KE7$
z;%fSI1hUpF#LD$N+8qU-D9m!6W;U<eZXy^PD4l#i%79v?%>!Vgv4fRVUj@KISPPz*
zZIhqvT_J!}-lv_?p5@s+mk<S+7sUXU>JT;bXw!J5HtU&cTS~muY6c`TcI%G$$RWC6
zdW)Iyx3qz(I*6e$fN#aUS!^U3<w((bGE+hd#Q2ccLRlK~>UL(2{JkzZ6-|F#jO??A
zR6)BN$a)Fu;kL)j?AxJ{`0??T5^VPj2~#^?&h>JRlHg`^Oe@rlX$L5WZP%YNws4P?
z;QlS-WDbG}&CG0S0>_m%Oz@_?@usKnlg3Z^<yGM6VMl`E#~#yCdwMK!5V`p22(VJ|
ziq686CsjFVWy8a^v=<3FfIU6{TF1U`(fmfL{RIEQdHsM>NR&VRaJ1x>WOqW!*>cfe
zY*au*+e0bS?kmt3&Fcu(zHcg3X!jY>{3|A9jq>7=A?f|AtbjCE(Y?_t1`frqvY|d4
z`vkq#Dj4xj^2{+dFo3lL<t?-Hb{@AlYTWFxSVNY!ca=oc=V9x8v4K}brD8jWR<VvS
zimliiv`s0<TKPPvY9p)#0#>-TC$CEx?A&#J)uPEyuBq)e=Jytn;F4A;=!V3A$vijv
z#KOp#%-%2V(4v#&2)3@F{HJYU$|b6-JmM}L#2)9B@I#3+#WmG1_)A()+}!#4d%$eb
z84&01X1=>rie^lGjNL~Mi_0YGe(pU3@&sc_(BaYKGT+R~%Wr=Xmm7N*GPC8~qORVO
zj=DrlDlN?d5djE3&;b%$P-^LzQF8ffZDeHOXYlH)Fre6xoWZm3h<yN&B5_#P^g)AV
zT(HVw&bMk7YG`!!87J)N5r-QX^0Q-EVM>g=iI_{&o%2=CjxiX$msZB#2oKG<a11&0
zH$*Z0k0{Fn(o+O>^4{${oM>Zv7T?CUbm54F&9LZ@0${y+;RP(a@aS^yaoTUt1l=!7
z9CC@M9cb-cyp?Gjds+VA3PMjG=}*Okd)=z9Gw@OlnKw1&WI;020T_aO9(I7h08DFp
zS~?bG#rvHRTc-hFs|3yRB08VZaw;uoF}MjO)B@=9a}F8Bq_>u5sx?iFyw_Suvz<H9
zuR#?QyU;^L-{KQZFW{?xON7$PJRp;omi&GPz1zS8H^IJggqX!%Qk^D5%i)0s8DmEo
zd>CAh01AHi?J^-Wv*@qU!^alHSLo>3X0C?IPw{Tziv1@3e9k@rWsWa`)3u{8K0eZ2
zf-B51(MkV;z%=eW2Y)a1J%{-wbAOX2JuzXk;>r2f*V`evHv1B+^WK=6rAH9npD7&b
zUG8t6^0QqbRnPY;*@(~0RwY6M&x4RQtvZmrSNlCIWaRn)(JsyB#+o$>WX$m+XVNlo
zZl>=~JKI78J=ciXc-IQ0QTLbp=U>H4fUqdrRTfy^R&QrurYEX5av6)Gcr7c9o7C7o
zs-s%^(Ch6DT|JJhn;_qU83#OT-j^ekCJC+tUpb>CEtQxw9WMV(3iyYPJ}&7`Th0#s
zy?Lf2Kg|?cgSH+lRf{oj#WOhgil#81is^S)Hhp(Fi8on;yq0vQD?H+JkhT=L3xYf(
z7(PWVG)XS`DxqTbdj@7?zZI|ej#=*=xa8Ok=0jCq9mC4(mr;|w^HXU_Vtzni0C&$b
zZ1KF;Ou=hQ-d$DmF~JSjmDX?vttqL4Z}kMcb#G$%g%=)t_|#o;jR&dRGspw7Xf=Q<
z{A}0cb~(y0{4BP~=alLJ$xU~e(<ELE>C557du+bo2r^$BAhpV8Zq5U78hzn7(QsNO
zzNW9WTV@5vWZ*u$_gKCcjbxTe_paF;y~XJ`c<VlBvgFhAo3u)^6<Ey)^Q>UECnUBd
z_m2iEq&&evvgb+4;T)qVCnsTF0PWdhIq|M1!4w?>!oioonX%&Mimx?8FQO@nuo?BC
zCd<+)kLhsp*q4hTP*DYQ26Rsn{tg}YhP-FTg4||al1u_PfoJMRUw}Tyunfxvd>t<&
zwx`~*cR3$H$*MoQ18+-3fY5TA(ikmfrQB?ofMyBw$``J!%VETJI?0GG$oK~X5(`%V
zbyyvL6-l0O4+IXV5M|v^A8&jm(Esv2fPT5c+wF6m#s6MRSl8?zZl*fyL1-%Ii+JbF
z4NwceXiaXeJ$roFmKT4yjJNJ<Lkf5Rk^5TXB+f&`_c&O9bxAH?@;d>|aQ`EyZJhjs
z<Tiez>}NUeTWD6r{jZli-&e$XpHxALm6F-F(GM3gZ|^%ob|QvjvL_egU03e4$DT9i
zuv{Z?Xd^nR5pu7aq2P6A+_@t#GaNVp&3wn)KE%v$j-8WLnPS7M$T;?69NX2wS%?rT
zE1JTVFr81ND;G!BK5UCOoc3boOh+mU#S{#^7j98tTSvbwW9-c+9rS&799&;fo&u=Y
zy!-w+cmxu#+vLRg&;<tz*TnPwsL9angj*XH2D;?-)qkwcQ>8uK`F=T1Ye23mcw~2+
zM#%RMIKCq0;jWLrDlK1Fty8xvCE@yB(<fp$pdPc-v!AO0st*Z^;GFp+b1-g;_iG@f
zCFDq_E>UOt&K;Xn&`h#_JzVJ&Zz7mC0THL>Wk9;2`s!jwCr4f2i41R45p*sWqK>7a
z8KtVRE|!d_#y>p@x=Xzn;#@;;DFiwqPaTus8h9BWTt9yG3f3)+r0(%^Gmd~WBV+qg
zN6fzU>F{!9Xm4`<EmlZCXH03l#=W<V=x%mCUvmaWmI$q`lyLFYM2WNen8wq<vxeo|
z(a-+EVO?`_Ieq9RRNlbM1SgucLjDb@5_5{H_9T*;E<p_;lm=aFa$aA_a-lk<esaNs
zEg4x`YPYPM3w8us6n*^kvA2SoNKaQ{j{CA4*X_0Ap=gPD6|@Bbi=#xl=pF&pwGl8N
zvY2w-y>_d2D%!<(QUW>jT@~dGI0O#WY|UHBe2J>>TnZWeyl)vJ)a(FuyP0O&->zA@
z(%O}=&davH<>B&2GyoN+<LWv*%N9D)UdL51c0SF7&Qm%7k578Gmn6Si0lqo@`!4+`
zoR5iS*22*ytx#D)w0y464l6P?m|gS~(kFU9)-qN?P_r>0;Nu_b#~Qv)`e0c%uw1N8
zFqNOhS(qmKm+$LcXL^vcU>=ia8~Hl~h*ba(xv8h6EGXrGV_Y=Pkv2oI_*mZkuWAW?
z*Wd-#n#!jtlz*xJ$TmPT&Ms%fnaFpQaXaHJMk-aPP1bciWY>x1!r51GFoe*#v@_2V
z>Dc5H{5@oTwoDvwW9oy5NR8AmXl+sD2K?(VD<6)d{b*tK;FBoH?{@u@_Vs7kOXs`X
z5{4YFM&4BVlfInya$m}?>U*Mjd&?JlIwB`A#Vc#@GGK8KzTTCXz%-a!@V-24X_0Bb
zMU6RT>&J~koF3{ku+AiANXn`JZ6Yc$zu-OVh#f6d9V*92C-*3(o4UhfR0TJvL)=Z{
zRTa*aU&w1m3$*|i0aEc2y_=feaQjlYMMb*3FN=yHC%<ExDH<L41AJ(+0%1MT+08;Y
z-$IgUa+w{rMr6IoPX8JF<8aX0*geDiwRdePZgB2j7uSES`ohT`Od=Yqvkoj0T^dhH
z^$>|No@4VzdyD69c)q>Wax!b{O0%=sr?3oMIOj2FUyDk>)+E|D*#xRC7g}q_5F!G6
z%J;BBvc6V~te=>z9cVI*shw>Kfk6|tx&>(R`CP;JHH7u@Al8&l!jTisk8x15+h5CA
z97;yz?zCo^6XzapN{~sq*V%h8c?(V)v?GK{`Bemu+WX`;Y0*q|k+#-4442I`F+Ih|
z?TGZcKZ+eCOSF&iFD%uo_70qbq>~H>aEC#MfT-@gni!6D8^xMF7jW!(LF{PDq-dje
zd3aU3B?B5LW^s<cNH0Y0RSbQTo3>IS$dr*3M`Yeg8%?u%aW+V#89rX*L_Bt6nJTnL
zArT+dhbn>%KQ<X~T0|D&gzq^h8WHQIZb0bV(EeCx{*{nMO!3O>^vWc~<M9qf2MD&~
z!uk^<u1UoM9z%auFBoovCM#xrHC3W^>)V%};C(nG|AwYSqL*rTSN~n;LYhF39ai(S
zr}mE05g(IUT$Iy;mTcK3+T-#N4Os(jVm31$omFj~gbC7gl~P8GG7?+nN=QYe!6RfM
z=}XL{9z|GN1^VtI1zV|tt(cix#eQ$!tfkqle3wnvMnVeCA6FFOa7$;p!#;;J=oN1q
zh*A=WeJY@645JZ08)~X=ae1O&RW;s}UvffsBEauFq-kjI?R?v?18Tv5ecA7M#fh}b
z;tvLo#@qAWsj6F_kPb6VMhk<nA_P^G=FiEfze`h89+RqGy<@wo4TXJ84bvlzp48Pq
zs0aJl<RWZO1^#uzWK`V0nvw&XK?r-gTB-44PUr-k3gNq;hx0zR_GRKbRRs_d?)d$H
zB#2mEmEdaFn*7ohr;qfU>3Zg4gXO3W9nT=`C>(P&6Xmz}kZih1{$cjF0zIS^7%elg
zpM1?wD`|B#NbTu_Gr%)5d>w40KIVi)y!Ht^ij7fY9T_n08-O-xYCs}1X>eojpf20V
zcu3dN@leLAt2LU4tUs4I^a@qp;!Q&{`oGs^CpuA|B;d%8h_&jGF*X4q-!K}E6PDa+
zwIM!I5QJHDK53MJgxrdYj)tohuG%()REHShH`=DA7wL1qVe7wPOte@Yl>)Z@uO=X_
zyk<kYCZ^m#?{E?^l}_t(zI7{r9Mt$+6td>P8R*<3M1!ooPv+Vvbu?u+Ai$L##JR3a
zej9Zt$9~iU8|YWe;T-$a5?GCo(q$#Z2U@l=5VWWRwE6d&5>O<|PId)5Ct@|<tXZ3&
z=m46{{pB5PHQw$)<aHI9$s~%L9TIBw_7~_wP;!%*WdE8*iooZ4S`v*ca(Zw&5u<lD
z#tEZPurhIp=qD)0q_5lO<TYl@hW3p%^Q`;Q4OZ&9tqu1xcO}>;C8U@mgjChIQ-QNF
zN@g8#i+Ezu47W-H5tf~56dw_3*q!44=7M*yW%<>&Cic5q7sd7hR;niH=NeLbA7>tC
ztsY0EHNMO;-cM6<ts|RlR3ZmU+j6@^%JclE%HVnL7feTD&mtpZh}=Hm*f!D=@6Vtc
zk?7Jr1{EQle|yxra+X6A$p<0Scpjp?7ItY{7vCgLT7T?|fwp-z{-IS)&~YD=pGTC)
zq_@Ofi7CgxB>~cNEY(+;m~z9urf=UX8_PNZrY7bgBQzk!&$3vTY81~y`%Kufx(?r7
zbnsbpz)o1Rrd2-^D>O$I@D{?86}<tO_SK2$LR{!2I=ylvAZ)C`)AgS$X%GX+i1?ho
zUMAM}`;-n&rpZp)w&4p>U?fU4E>UZ<WR;{emgr1Rw;<ZEqa`?Ay4%MhH%?@=P4<wH
zS?$vP5wNCpNV`evx-K?<dcmbo_46p%?c=48rdSP8ABGVIe)nFYC5~r>g5m~x>4VOv
z<KX(O`Hhp8Nz~YG?S<vg)dN7deBSK<3~wbN?|=SSjluqt<M2NZHcJAzp1gkq+%ZFl
z@jvea8W9Oj-A_jNm9L=_pK$;6;E##?x9EOI<cCClSj~Uvzz-bxZ;1UNkslKIar*wJ
z4*cLFz*+kbD)vJnKP0l8N%4clulgPTZ+^|Jh@8IWRq`#5xi#PibHe6$!BO|?{|Dl<
Bsw4ma

diff --git a/docs_old/logos/dassault_systems.png b/docs_old/logos/dassault_systems.png
deleted file mode 100644
index 1ed0e876d39c0d88f482ba032833f1561217e94b..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 21865
zcmYg%1y~$Qu=OtP8r(g>B{(b^G`J?Xlc2$!CAeFH1c#u(gS$IyaJOK=T^1I2n|tqj
z|NkxD0y90+-KT4Ms!pAXR9BV9#vsQ4007vE3bL9200J2PcOM!G{2MnT!5RJw$wo>=
z3IM2y$Ap<7!~dtXP|#EX0DPDM0B{HZa1TEO{sREGa{~YerT~CQ8UR4zoY|rw20wvf
zt}HJLc>epzZ7)s+0JOo1vQpY!OUGHBUX=aNL5b6Bw<I>x405z_$(I5R=po<ml71l!
z%<ccdlU50qd^xAcOj+a3GWaD#-TeI@6%jHlqyX(=fFhcyy+{$`3oZL7iYZ0G%ZDvl
zaG(B9&(`9vj{({8YMGW!9v;>nH=?I)mmRq;qw@dX&q!IcEg_P^#b;pk(GNTdxo(_Z
z3n96%ASx6r`I37`y>1yk3dysBgq$UhE^+W9X-$8i@2<ne1;)>BYqJA7FWzXpc!O+e
z!b8{2taGlR-+;LI;)q#aV8FtH{V>dJzOc)$x{CG!V*sS~_Nlu8N}^C4>+x^@tbky1
zb8Dd<N3>QzLC`U{O0Fa<8lSp<0X3rkq@dm`+vD3m<75RqI?MXJq?&y|SVvhzTqw$^
zpR_}(&~FYkJa~1i{&xTx+JMRkOCM6^YydN09r#f<k;Rv?hUBOHP{oR?QZbJEe~p-H
zGlCw4ohi9(i2|ksrHZOcE<aiqx4I9p{T)*6J$lzRG9ousCrSW<*N@3$iQummAJoJv
zdyW70q3dM+W0lb9=iYHw;*Y#LBh<}s`*uD4aKB5nv1|wTyTA7sIT$`*`mCC%F`oZM
zIq2@=+UfE3rTixwpXLsPf4kVM&{gw=DjMkz+cs*+iv^H87-<)%<o#U`R)C)WlGdL2
zF|aGR>%dRsW*N)y9<|zs!0?~$`E*a?1L5oRtA=pEG44z!raat_H9{85@B7q_^)>)`
zT#&Tn1j=P+`i{^{h#Nt_ATS2-fVlLna#~O7UyN9-qDPm_-Jb1)Ud*u!Fzf&18DcgS
zL=h+^wQOmD9DX>KHa9Yu{PrX%s`B3jW-h;DKCV4GQ}o5ikz#wWBsl#d2`YI*ZmX;^
zt@^jk3nH*mp`tN2_+-Ke?<dP|f7~NM)*1OK9#)AYw<TMae+&&nvXD#I_CQWsk!onN
ztI65Px{+o6wH}4hf4yj!nR{@F6@gdzzvnsgHqZ1g{rU&2gRKvQSA^C@)by!K6)Yj+
zorqlP-zw@8iT*LVT!J@C87Bjy@%<O^lA}b>GJ(p!_EG}S&R4R#M{GIv+9=pSqD8DX
z```9f*8J_!zbN&+!_oxG{<m#hS_Y8f?K{5`?6Q9s;cLmX2$W+Z3#{a8LZSWN5Fdkc
zHJjRVRZ9PB^CacK-5-Gxr28KRS2YOsq>@srZr*|OH`@OloJqestOD`_zg&_2WagMe
z`R8eY12&V`1M2?L8@rVDcbfSK>iZ-J35W<U3kioCKU%Q<J%}O6bnrH<R`~1~pwyBv
zcZIr?`(N8gk(#n(gnfxk+<z~x^`TordQ6B+-}Gw3z~cD%on^SN1IN5x<L|7kuBAA@
zk3Z+h5%2<OGYfvJ{hf}k#0<j1y3pSR@#aE69AV%8<WlMc#|?h&%Pej^;6|{lEgF7>
z>2@mX)8DzgLCz&`CrEz`Lf;{=bRB!U`s4q=dYNqh4)M+(z~ui{Z}!#yH_8D+4Eq<7
zim$9yd<^~%Mh)gkKXzv?DDrSY%n0PyKZtypr4UD^bZ1swCa*dZ;w~9W0-=z}b6%JI
zyGG?}CCJ<(F*yQARlkVAcZl0UM1lW6Q38nJ-LI2(-()t;xTvS;WJn$PQ1B0U`3T%2
zW3g&L9JJi;Ecya>Zyo;KB|wEr8ML2Y;QD_@c)3qiuILPTm9<pz&&~Xzs1YJF??#Vs
zSMb3?&q#JR`TiH{b^n0)1}fHqO!?|u@cOW;0vI6??EF1S;x8)jZAUJB`%6VvD+Q^c
z$GaZJCGzEeaewo!MqB+p8(m9)h4_K+>?@$<s&QMh_~ZYLuWUwT8SfR0Y@rU;b2Ph6
z9*<hHRQ>Nr$XbwBQe&KqNnWkcMBM#!p(<!mRX;EOYj{ddCCJrPC3eO%iKuAwzxV_O
z^DAZ2vrgv1{xPjWLHNRwFc#ZK4KZ{_NH}W!dsP%hmu+-s1$T|9-&48aU_K20zg0aC
zwniu|xw9h+<-bKdOfgCsf^L&fXy^PpIaoe%my(8{yG@KYDFyhhSG9n{J?y`UD@ox7
zY0JW_Sm<w<a~sBzFuI1f$+@QG{`Y;LT?&EWjqZHuT<v(V$-P8=I`!|VLQ3-`Br~(n
zdoOY}63r*whtBSRRKsbLI-`G{(N`q(fG{aus$z9lWb=5O+&bZfg`he$A1w;qzi@w{
z{Ut*y<+|X_z<rL2KoBUH3^^_O*QUSFYG9fsF#crVSwZ&r!K-{AXUXeO`r*ITg_{cN
z_4s<X5?i%yI#YYFGq*$R!RbGc!oeBRocH6UCsBV=T$le7MKxBg=yaQ5Bf%}6K*!H-
za<`u1=HUY0!EvydUMDs7mjqb71PB&_W>7R(NA5nzklL5sIsbAw5iZoCkktS-wbZP?
z;m0y`c{NmHm0}sRzPSVW&Yu-OhC0{QMAq<8RICB%fvEhEbx03v_JSiz#s|OM+Z6u{
zbVi-xiT_R?c4`ewnU7dZwatRfv5Kn_j6cw>rGZ*IYlsu$@qUt8n6T`&OWSQdOcHr&
zHRP-Fqwo|+HUK)Ld3-0hqXI_lDmV0ux~8%_q58(sUg$Lth8p<~4tGuY5a*dIDfYuP
zlvm~9UKHT+NdD)?j&h=<M@-|CWvwby&AGn)Ls}yP+FI&4d+-o#b<QDTDT=62Cf66Y
zsW~sD{(A>$AC<C?9z9FPFLyt#SMNEI_6t=8gYPf`pr0Uqx6Ee*a~()+l4D|0hl^Zc
z@Aur6EF56~vBj2oJ3A4st%0lGzm{u2t%BSQsotf@Y4fUfy`7Kaxcg=c-v}u?%4qn$
zUy@fUdWKy19|i3AdRepe{a`wHF_t`L-}2n~Y@>p#cL7MdCmNfg91?_YTG+C2MI8LC
zYB@G=FM`^a)(>;^Bn`SORZd*<#{2z}4;@bV`sMtk>%+Tu2&6%>*00V8&Unx0Q@qo9
zv{pdSSsjMu$jq?%hpS^=$Y&xvAR;4gm4!D_&CfHG6~#j?(EiVRs^m&5ML!c-^rKL-
zcJsyua;I7G&mTEV8A6;$BoGkh+_dIlg0htbY#x~K4IW+bPdth+&O-2xQ2}0A*CE+O
z3PxokN3aNlU0GdBs~vYvl=6u4S3NXvu9V({mqbBd?Lror*tQ@jSgHdzW?)GZN4|3M
zGRF?l;(1z?86|S8Q_6)_FVW6R;OrjqLm0QF*vvPJRbrE4qG}ASt{BnxR)mXa%naP=
zWSi$h($^ql)Y{~OhXflNzTzH(p`Mc>ewJyK-ymcx2C@PW^EiJ}gbA3oUx@Tie5YmM
z=BqDp%C=YvQ!kM(LG=KN-CMKcq6}Qah*-k(u?QCX7j~N?#()nmh6=XHP5y)<I1@&!
zFxH^=Fw@GLilnILv{PD46t{nn+99Tzo~N60=T*Mr;TG!?)uF2utrndPB_<f>@G|Pe
z9$U&DibM7VR@=x9EW?s2#3FZt(^<5@91J|x75=PWT&6$7+k7Mtm)@bEljzs$&PKV>
zCKx|9)!d#45T@lJ7+fHvstO2O+&{tK5NY%z2leC<V9*=}RFJ@W<^}?hV)pfNa!J)H
zyv(&>sy8g3AF0^ZP@JuJ=!>W4ysT)(N~DdfpMOW?R+CJ6Ad=qa;NBL`zO`HfJ|V%-
z2{XpW6e03~zZUrEjl1deeMRUc@+*b=|5Pd~6sy>JF7p}nSKmPrDqp84HI8#9m{|M{
zPZ*<Cvilg|9bIxnXIjPP3PST~6Mr|=mSEeAcLXWmf8D-;Y-;2ozq=HT->OykRagPR
z{2|E}6lyp<yF>D>9a|{6MEENY`|GOH%*!=i$a6yVR&`}gMNe{fr?FpSPAM_w8peb1
z7T_Lg%k<5lYvQGAZ6_{-;*CLB?QgEEn-V9`H`yCu*%O2gndd8@5v~wY(~GJz%E+;w
zjJ(*4)4lYDdwUOx%Q({8^x2V+?XIgX^%L%ZB^0cx0K84<T&~b_9MT68`1`hHn;?m!
zmEq>zHd3PNqLF~OAvvGr2}~M@>vo<o^qTfYBP8qoN?fKgpECnvBXQuqseu>a*QFJe
zFKeF-=dW3DL1#oIr;gY}nccv2+&T<N5@zc#KCu9<q3^s(@3345nrAvXvW!QW5}_!r
zTk&Q)?>Fpz=%3mnTJoIhl|oeMv*LcIf#T6lfK{@RF~u$S&kDxev<tVkOLKjg@xgw4
zywT`Y&Vs|l&(c?YWV58cOmEhEUrtJm)!h@;E+%$0<5NaLroLeepgKd`h$4@$B~wa{
z{36E_Y(Y(9nk)SgdL=P{y4v=KdQjtCH0EP0+>YqZY+ppCsJuI;t3x}K@V!m7b1xJ;
zzv368F1F=i`zm2>v7DrAWIQQ@J?xe#dn4Fo;otVV0?E$Q0Ta?^Q)?sgz7o56Kmx(N
zi|^59Dq)p}DiF<)U3C$oiko5-1>bi{g(no#@A>0oPb3h@5BHm3H|5cTb~{foAwL_l
zSWq5kLekWS_|0<)a4*AM{<!JyD|e-*=9)@`U$jyNy&Ye8CiHZnbxKPIHA1I`if1Q?
zb8&|@>;27W>0Ay0I$msfoUIHCt*^3=ST|R99jaf!{rew&tHllqdCnB(p6@E|$9S{x
z`CpkT^~r7=cOAq1h*W)h@iKd%(`=3zQ%jBnGc-H83!U;+^aY{_DEl!U!J_C;p(Cdn
ztiC*`#uLo3BXi5sL#UsP=#9^V$x}<2xMR8{PNx`VX60JF1*Qth1DVXwaP#oRrpIzY
z@rLB@8us6};;A_fj3`I=Q?Yf26jv94hY*N&_8vSg)Ygc5$0WXov~~al!oSW}Ul3Dw
zQ*z=hD>k;_>!VSCn`b|SXl73WTx-Ol&#<Zii97nBqkRZJK8o2p5;Mv(cT?A#TYbW+
z2ds_qq>y7VJ-duHX*-_}p|Z!MgV^zckKqLuu7EzI)8}w+O7nHLOMh{s#4W0-4y^C(
z-yghg97v(`6{sZIfcwG@sHnbZvU#~^ryDReF<u3SW4lj@<$6qDp+{ySuIWc6S{;k5
zDj;B_yvQ@ilPzGqcY;C+6zB&D9}s`aT(}fs7XMkU|9)T$0Mj4EpcM>mC*yhsS53DJ
zW1^`45u&Uq4sN=??T;9a>>@j%F<xg0NX6$9o%Fe}n<P)*ItFB~dxfag_?>Y+(gR%R
zAB?J#{S8zS*#2BB`J?Wrsji}LWnoZ8{VW{G?uz95)&%>`X>#i$0wRE1KUnT*e{YZf
z28!4*VxVG4C4fCUH+Ctap++pRJY(MJd0X(h);+9zu;KlA9#q|;jJy$5GCE_?9H|NO
z4GDkQ7cgg2ti@v8(02wXFZx|2q}Wy#eAn3ojR}>+94?)=Xj}FzC*pKy1}cAyL*e9d
zp!+HeI2^sd;9k%g=1*xb(uuM0A{zYeWT^=IxWPU6rdQSmcxQRNr2SCGi-9Yr1_SB}
zux-lE-(sGpwwVHowr)?PV{U_(+Y<Gv#y-|woC`u}x;AI59HpARnxFky(@(|)-3d77
zbn-A+WEn*lO0$4d-Q%Ri2D8GFq+w0(0UY#dZ`>?f40!D9r2SE@fiN22EKEFv0v0Xd
zjh`6&!49UH6WMt_Ezq?k1;TD?6H$_A!agz+N$9*nye8(aZG`9?;VQMvXYaGJlic6+
zGsTPqrW$HNJCx?Xi|T)oAipc#;u0vRTs&?s<uv<b_0V;2Xla&kt($HG&GGTb40}j9
z@=M4X<~P=;y?YQlj{ONXX_}us*<D{S^~(fQCIG%HkkaTqhy!>@7pBe+^w_c=2i-M(
z2jE?<i(`5cfxqr&jy0Zb<KzwuR;>YrBj02#Po{i4ZXGemYdJ%W`Vg&40z%3LGy~Zr
z3TZPE?Zm$4cgH+1z`aT&;nl%~rT%x6{@I(e@`#ShQuKStcCdH3=cv(7&U;OPZEX%&
zm4{`lR+?)>w;NVpdy_Ra5l)DqJlXI#Ec%c%5L>IAqFK9I6m@NAG-!TxfO`3k#Nc-a
z)e_N<7(m~h5YQfm-X-nlOY?1KSRqb%numNZOOK)jS*sB&VFllfxQ!Ztm3dqsq=4Ym
z&Rmy?dN#U5@$@y#Q|%Pa5Zox7=?mZjE?BMl1>SOUVb=p`4Del?0?V975c_dDp2U&2
zMT&Vlr&uc%qnh5Wy#8=9&AoA_a!-J%4=q3c`ryrOunkA^%-6hZQkq%eWbboN=TU10
zuUgEFUb00DOQT-q$7Hi!%{;MUy$9~1g6R($HnGdUKg@k^EQPckP7it*5G?*2KL4Q*
zt5~B31&z_WAY6|h*YS&!E_#lCBYe4gdaR?n;VqvmdfDD}_{+yRsAU&5lk%zJ&ZNm;
zZ@S?WdSk`@7)r6##L-?dm@@%EjCb|?5p#4?6|dDt&=R2oyn9tssa)NZsc8tfk!~pq
zSZ6^M)9t(aKp~k|*yCx0ae+10fpDX<p<{kQZgjy_CQ2eF{X%|({*wU~-!p!>dwQJ-
zVl(~B{rBlG*<G#|m!~b<bE$y0)}4``mMVERsJ51YYfuJQw3>qDuBI5(%BbHV_RaZu
z77#XN<1oRL_%3577!on|yRiPeIr~#}c)W>@#g6I$FU$0rmk!cwj;TSMCn3EHPX%@$
zwi>Q?;Nbgk!kjr1rMloE#}Hq)C}E<^ExB+#78(|Czw9)^AGU}(cxcqm&X*F|Zg`C~
z#DulqR#Z-yXr^u?MCCbZRqBRtYq3Ku`xMf+h#7P@QZCR|C-ceuNVZV7%IWT|=>Dhr
zi*0>D#G)0I%FV?y5;jhV?TV4a$8EvwV@0OSx@R)P)PU%g-JI#QD`&>AqCI}|F?(1j
zIyu;x={o(~jU=G^*rt;pU9W<(mCdnuqV;_Hv-+XA{%`BLr1vH}17bAhLm-VrkJA8e
zJF!pWwA+Dkgfzd0`yQTwcyKO4H-zfX2HDzU4sHD$r|dQ{_k2<Y45Bdi7;&X8BJ`?H
zv@N{8b6|vj%--;OaN!KDlzrx9q__bc(Drya;)lVvc+voti4yC{d_<}@^PxmunvJ1R
zEGN2$8JaXnV{iK|@ElphB_J<lsp%&(3pTRwZO2+QZ5{5|BFEl!L7xWkwj%p)=n|Xp
zxYjArM>~a36DC4)z7C6UCRIRI8Q~aL+{%V7JsyH~sRLbC0!U`NrX1j2Mshlx{eG(Q
zCu7Q;g_P5KoIl1ROKP2CFB*F>)}GKhfuh*kmSRq?3tH!x=Dl1j`)MR?jvGIzp`4`J
zm5w_}`A$PwpGI`SAl)M(7t@<QNFLH{|L+HyqafvSDLWG0mP0inuPJ`2PNgSR>;?gl
z9M9O>NKu3&TI-ww5MO9P5$?hCX1ZNluF&mQS?pin*iHri%Sn8pBi^+ND3e~0r@hLc
z=MzJq%;2D(X09}Xx`#6fSmz08W5a5BlQNwl|0(b^unGGF^)1^4)<+=hOFFh1VzK|H
z&!?!@4ZE#$U;U_?2l(7gkpx8V(H2eFUI#6jozPABaj@OG!-OVqu4YbCw@42se2bAb
z-9+8fPw7{HSX_CIh}Cbcz2|vm_;bepR35U<Ftd536rmi{KFb}wxr2~v#gIiBDjVGt
zd!_T)+QF3jbLi5kB5OEVAlS3BL<^q<GlbhMQaq)<sS)dp_1G_Ve-e2oH(SI7wUeWH
zwUB@COv1O7iEEfXLJEWvA<v7PA-p*Q6pOhk6nJnOK#>zRZqGG0voGiB{yivNld<(k
zgYihl@_oMV`1MPL&7Gu^$+bJQxg)@ZEgAnRUZMmy*ApXQM(4bZMZLKDEBbY0rS$*;
zkDdz~g7kF8Vr>DF+D~!Cf?P+WcKtjjq_)dZt+e0L$Z-5-Yq(e$!fxK`(Zv-fakwg<
z(cFIWqMv(yd}g%-Y6BNeYr1V<)i70DFC>M@D-()BlFqwN35lRe4_BZNvk~T;1KZG2
zXZTahY<(r>_rf}Xuhgss{uiD4Gf<A|adlnemE-em&;j3Vf<4(R%T_%He}<PI)`IBB
z*Y&goF@p4HMeDgr^<yezQ{(|-{6Wy`lW1HB#8Eo5AuwFeDfm*VFUdp5uTShKt_b;;
z9=WInX72CTnsXQdxw`f)hAJ&38x}h@Y%Bnd<%gyPj@ZTDWjsyL=546u(9b-H0-nm`
zlS`{Xqs*)m=K0rTY92)^zbI1td3IkBq_ahvWwVZr+psgvzXsNX19yaG^bjTbxpP|V
z%UB4d)BG5CX%_NLiZhU354eI*+^>*wX`Uq@HgRj>(>h=SFETq4m%(kp2=e_p)QQVu
z`E7F6Bmqri(`S}DEzi%?cW74}$wL*el8gqr0KYlL-G(<z6PvHHf2gyI4JJndF;3@Z
z2-!zYBNPzN6v>$S!CNJqZL;*PoTx>px_(Cu{Hvc+*U7iZb18$wUeu7i9gLo_io@8+
z9pc%4Z)(^0i98^NUj)PA(`Dh2Wd%NnSO&xnuIV8pB7YTt(+stHs4wdb`{7=rd0A|-
z+uw*}xD^8;`bYpE0{?i?y(b!@W8g3c;F%lT)-Q=cR>v^8y~SE}rPSMqGoKHW&>->W
zTE$UBe%wsUI)M`x*lKbNRu3XL2my*>PwejF>s5Dm4l03?oYyi5?LjVEj}9TuRFPx$
z@HkE6mk(zpS?IFjq$8;i7wop9x@zky<PIz3L?p*g@^E{eL156m#_bp1(=Kb|E>2Wu
zctqxN4n~&_kj_i%Q{y{53IW=w54eduae9IJHR#Jaq(p-Y%M%|mGFj}nV2X~DJ%qqU
z$^mA^Q7OBRAcxY^&*C_DxnolTQrnydd=>t+9GS}Ek{}6kOOYL@Fv5jOTMy*{pBBP0
z@lE-IM<ARlL9eek_%=T%GHByY<2!t1>O4|Ct>uW!uGlZ^=)2c&YX`k$0_;DNNwV7Z
z8$R}QQNoFpMFFE9V|1u&u3dAe@2$t|al|SPH~`tP4naTmO;~O2QK6N-wC^6$wgz9t
zeG>obI9~+Hz}lINAqyEg>B2Qi6~j8lnxtn;q0~zrTTu-9fHtVE3ZqgMpABtSe4Qg%
zx!({Kw<kJ18uU})^*VSbDcghdt5C}DLR_%oYnEaDagM5WvcoiY3q@NCTtdV}tkHNd
zew>7h3D#oSWD|wsuD0b{F)EtKLA*iF?#`wfkIaW<bNY|KE5d;y7u~TZ4nVvDE|W<+
z*}yX8(SZ1m$9mtVxkAzykMZnb4{r5w0x&dT2*AGoZkQ7>1@RQ|6an+42#8B^+a#h5
zFQD%(iKU(8bK}-X<=yZ@S^4+CvK2nCHY<2P<cQhSKV<KW+PSrINiBEU#9U^o<+p)>
zxL)E_Tt)0D;-F?~GsdQ-A=R?pn0kVNqUAu6-CG?$!@Y02pC6gs(`jfg>{c$PDkC_g
zs0%oK8IMcTHx1q$RqA_Se>nS^i|rRdyybB0hnYBn)|E!J*IRhWlM~CJ;1RRcA5Es?
z6-uemH;d4OCD&1caCRA!(*T@ZEr=p4ieXhHwU-+#PF5YJNmA3}u=5wde}o(q+@%h&
z1C0UMy%&@`ry40Zer=HC7EjNobT2<)yv*RH`t>)ZSJAjd(!jSCH|i1vz{Ay@CQ`C_
zwTF^g-Qb%6`31gq#NOW$GD3|nca@5Q?%#4a#1`vxEcQ)J@Ka{D2dgFng%hV%qfqE?
z1PB&oh6`bVJ<*^3$ldnV%&X`%w(e+IvhAH_@j&#3&}zbezT_V;DIc>O-6frK&Me!#
zdWUztMzZ2qqo5p}*o1-JWohkzarAl6v1>j$j}@a=W>|m2g#85VC;~9h)m2Ev$QvL0
z$T#r7ro$hIvD&!Z+056Tc|V4zKX@hYXs?!UW#EDh$^Y`BX!T9v`jtZA{Jt&C>2xxj
zq@F%MeBF@ap}F%I2G1!5t-~o72W_GWd7-+ilWlE+^t~Z|<4KkZPpt!)jlmhKD!C)8
zE*z#5ZR8sff^n{gH5}ts54)Js<8X`9et14{eqwC}I`zJaK2H{Jaru1YcA?`?*!rG)
zIq@p%GF1L*3j9?3(SQ7I*S*vZ6efFv%UH{0>V><B!){}?7(F$`rW5tv6gPoc%on$4
zOPnA*L&1{yDul5FE35$zLtMAtfVL^)>9I(j)gh=JGa_d?bw2d096uwsz2m`o?Aw*V
zacnppC0g4%IYpAI;qC;;bhc0C8)#N{_M81pCzQ{Ii-dyXgKk#$Kdu+o7hEB0mm2Zh
zxZa@QI3z>Fk>?yI%&hAgaNGg9DI^|0x3(4kD+w5$t%37m2N~0FkhM54@t}wc7BDE&
zo22d+$-NMGcxgnicvx!p<nhNdcD6y|3BC7u7b%IuF89j>{HbUvqLaz4jM!gKxcTIc
zYW#$<cS|l2RLBsVr*;3-rDd>XG_%`?{NSr}`GM?hy>I~}r+^g8S0p<I#X41gByKF0
zX5*CQarXX5s}8L)==t(%DM;et)437TN_V>nriM6>J+a<gR~4X|*@1Jts5lng8L|*)
zx<eTy^*&T5InOj+cAJhM-J9`S{Bfi*u~@&u9t8rYJK9`!K5J!pBisA)9|!L{(2tI{
zNCc<8c^K&5VY-nQ4nwW1Ro)XsmaU{2zpdOKj5xw%zj=OnbEb~Kp8A@la5q``c$N@r
zxign2?y6Mwz8Lw}<zu&}`*Imh`w8Kco%u0(oE{hRI#Ce-7J}Pq98omjNuA-{HWAfg
z6ncu7HfApq^fR$CQh$%|^AGanAjAIrH!3rNIm!lG*WKX{FAcG3a()wgueonzEAn1G
zWKTy&oer<YxjtWoQj$VT_Jj4apWEZFbx9Nk<{9@q?r)f6nf1DO&oQe-MQ22PmgO8u
zA4*1#;?HbMqTNPY7~cyp4gSE#pHC4e(n!6=%Q%N_Wo1>pEf(M98K_B|kZzlo@wc~-
zqjPD{ir*YRf7f!D*GmUa-Z9SftCTZi&z%F39WM6{71SE?dqm>&AjjuFVu^=Ms>1h;
z*Ut}ZY*qHy-nE}0nO-!I-t4?0osiY-^$KRQ<|{W;&fgjITFjaJhQX<#c5;-eXz7TI
zgInDp^O74|J+NyE1h!a>rJ+NWGtTZDN<Gcl`M9KYBHhEU|6>Q=rF8CB1}G6)SRfa<
z+o0jQ#lXmWi}-M7=vI!Yfe#k<`z4&UdCnX~regRBSMHQLyi&5`kYHW9?{;Cd^6+l0
zQ`#<g{jF-{ydM7P8dVd^7n^dYHOL@NVt*g!g@nIHth>px?94O7?V;wa9*nGK8KL2l
zt4|3-8&;9~$zx!lQzKQpm}Syky?qf2$rsC5C=5QYo-Z@w6GhJxXK$Z%6<(_$3(v{t
zbShVT_lqN(f8o5T=F*R-Z{k{hn~YB_#k9Sh;<ufhZ<`Js+N)a0=@r%r@qguZuGa=h
z`vtn2oAB(M6BYV0s0`<k*ygu7vYcu7C}0nZOKG_0jF0Qj5Zxy-=@($i0*u^LqHhbY
zdw=6PdwEnyD%m^u;jl?PAYd%834Oe$J?522-maSO<;Nxyqlns<@a_g2@TLa8hWfZ}
zvOKBcz5j$WzoA1bbevRnoJo>Bb*&vDKRNGjY?3QtJ|~g8=T^H;IKtH$*l$SkUbD2r
z<MGeK^9Cm9;?Gh6f<#E#&{6z3b)UG1JP;!A<z|Nm<X2Mg4cIrx*LMW>jAft7SiAN^
zpNO+=s@?>K10vvlmo-0nl8XM?*SU#{%eC8@{Z39%{ONN_`oRa-PG}wf{U}0mW(&Gd
zmTKM1cCW-Iwi>vP({bWAR5&#wNE78dF%KEgV3H2JGd>^=h+OXtpvOh#YO=qJj{hyy
z=|FY3_1rF~a|?x%)4HI#r(am~QNMO)AV@a=x1PH$j4qMEUGIgH-y-22HGY5E6gmb^
zAnLgcCg-+Dn+_0<bQ(>jMtVQxS;;E<;1?&m5m(gkr`%9iu_AdrpYxI;WB`2VMSi{5
z`ayYo?!?9CeD(49Gxy`Erv^46NxXDha5d#Fo9U5_dq?t{tYp9Eu>nK9=BgC&j#ZF|
zBZ7MVuApV8>}Zj`E0*rW#k1>4^jJZrP{B7`2~w5rYz^}SyIUI`kMe0CqgK#LQ#7$7
z_HU1c$-c+<&h+GLmc(?_7#CPSUWll|>u0hiB&V*diD8VS{qD4H{$14QN4S0db2rE+
z{SektDqK+~Bq&TSPETTfkE~g@Z*#H!PhEzP@uRub3Lz^Kz=ppa^GMO(Lyi3b(HFcL
zq(rH9niEsR5Wu&9I0B+u`Rc?>ErGT0y6e8bYpv&Q_t^Jd_Z@V8LG9v{5dPLlp%V}c
zE6fNYVH0Pg5442lGXB(hQk!}wIx*bEK;-J;K?oku@KS?syT-TZ`mOxGx12x6;~>oi
zA6a*S30%k!-7LP}zNH;b(I9nwT}`%AnESws|30>wE}GZ@OZzMFwz9Ex>jY|M&oYZa
zM-&gkN{0uLvjpCNP)D#6@8#I7QrjrBiyy>xqq32a7UxKiP7>6Nc4~e^kWdKl!s_OI
zVTQ<V!EKMyFk{pc&n+-~ZNtfSOfk1(gK8C=ji_aI4x*p`m2%P8bstjWRp}(Te#ada
zC|O|3b3gX+b%(c6**lUpcRaB>!)%G@&1o&wxfc=M(Z0G~qfdj)i(dTN>DjPGUVw#N
z5Px+M$~j6f6;>^EwLs4jX{$hwPnXCGJ}=qC$I?-1Wl_DaIwM*8J2Q=$Vf>A#V$D^`
z^|z?i>$sqpan|sht~!g6<22spn9bfm)syd)1f4!{EMT@XAZHlTr&07n7r=MNg8_y4
zb?`Mkb<LHgqgs2%h0TX2(7+1yK^u)MMWloC1nFir>(O{r$j{m}(#G2wtUy_(ZR|8X
zRiswC$aR=QkVDN`Xnz>?M-el!Y){(eFq@H=y%1Ql0N#nz8@t=RJ~BJk=6GSvG5uST
zd?ybE)<Ee-Tk!)?Oihtvp5omf+BHlA_Mt<%Xa+ZJc>Wnmr9wg3gJyfH-SC)@UCm2y
zMJgYPlMsv$KzLIC@)dE&EG9oyKC`($RR(prupY<G>&YIyC+q-Zk}OkZN`G$Z2fR4N
zGHs8z-J0IMY|mLD9wxXh!J0ExtQ=lNofRntXK3Z2Ivb2dzFPe~4QKl-Z%|38_jzn(
zBR#t)73{VoMk!G>FptDTu!-+RXBdB5kl5<f1LR_Vk6c`5cXEAm!~NERuTzVe;zU=A
zTkTCV<E(N4O%8G$$%($>={Z5lQd>;Q<Gdt2j*Tx_G5~hGefF0YTsZubUvqQ}YuUa2
zBz{c!15DVV9osFE&Q1Jlt*&dc<&gojL|r^ez3}5%ou}la#8x7z{Tpsez|qg%RX>M9
z0)$@}ma9m#H5crlic%<J#%=dV`-1ukffA4LN_1qFr0-S+@P&W?nwBNFZC5m&XKoh3
zcK;-Engn+z&u@S#d7XCNjpA5Ef@xlepjY<|9W$3+6SD*+<B^i@G(veMp7dkc`A}l!
zJrUJwwihBAD$3V9&SWnd=;qxFX*)6}cow|k=Z2j^pF)_sNbTWH20=NELxLOa>zbFH
zKeQy2^xRx`Fq|0HkX%#-p^)7TSY>Euzd)Xqh4a|#@~lxue{uZ>9O#13#wO^bTMu72
zsbS`+PL0(BMwLkYeu=I@eXykS+!d}ipmi+1;|SckxsqoE6by_Fqz>^vMHzlnj`2hF
z%9Zw{9=NDHMb^~x{mATWcNT3lSGZB1^mfoa7pAr>yKnITN};H6c?pG~Y?B8{pB&MW
zyrvK})=MX<6e>nKT+M4Ov&Nbn(QyO|>#O#T28K2a#I%}T<TC?A@lH^E(UR7`G=z!z
zr6#m-pfqtvvuy^FyBhf|>@{0lPz0^G!WrdZ{u5+PTy3G42r&%G;MDbR4KuMp(9l3l
zQD&E=zPpwD$M3GbD^s=a+d5o}r`V|`%(<`e&hW0&#j6d&T+54-gaJ+B2}u}G8vN)}
z6noO62MFiFop;wO5y=k8LUwR8ev0`!40a%>=K?1?3%5<O(K~Y)62N4sctK0e1u&@4
zW>+lqtM(yUSC(?LQ3#Mb)%r=pa#DL20l8@f<&zG&zN!Y?3C=xHoM#s`T+&#DOqPZK
z*a7|b?+hd@KS7!dqjs=m*Sao5YI-Ez*&P@A3C~nT%mrgIeyc(e&iqPq^pIxQYnOm=
zyWbzzUFp{5h6u4Zjtp8CO&tLWC#4&4C$R@e91#Q8t<cXrH>h8(kex|ixE|}1D$j`7
z!tU*tj6LOu5xLBF{9H<JK)&pcx~cHoRN1>LDZT|hnYK|sm)P$1E;w}|zg_8E_A#D*
z-Ypmh3$D|N;Qn+k(sv;WwvVpK`0Y($F^-oYMJv;dvk?tVPesT9`+=*`?AE(S*t&k`
zJ8>=tUgQT#2r-S60NBH2cab)3Lo6y!tnbc(tg4ba_18y(xcv}PgID1hbVI5ae^pWh
z{xhi6lrFJ)raO9cUyb|oZw1in`skoqU~btZ$pq1~G_uNI>3=0!mG?0xnV)xRoPCN#
z7QnOc6i}CF=7uXWqW$MV{_oQwMgJ<?Hds|f9L|ousE?eaoOF$j@ry+dL1aG0B6<}S
zDWZn{73#@mJnbPi{VVTL=h@3=?;AByg=Ztzbw@BFj|`en`rK3YLdR{zAL`Z1GKY-U
zlea}}vCMIEEB2}V+y8dM^zDiZbEEIb7_RK3AF9w-1ny~-G&3TIED`u|Pv9&ZtZ!cX
z?ZpR?NRw&lIx;4%zBS?cM`;Gsr;dbs;YZKs>p?dCCodxJ6*m%hO>(a{q;0PqRJe{K
zb7t|S_w9n(V=boY@1tJ*MG2#8s$?`1<Azt4wCO!4N{-1&s(NSfELmnxQ%%bD2`pO8
z7|`5d{<mp9LN(HPsyNan?5iP<1i^J(Xn#-&%UU~8sVAqqu(N|gA<Y~A8!!Lyf5FNX
zjoh7#c<fa={6OX6ql=Iz1x%R>RoKb4Grd*|vgySixbd9fkDu22Pe^*BT5E-!K*?>&
zq~|_Og%6A<$2Zh~h0cY=v*}(RLJY(5;?yp^P?S078;KsY{;nHbSjtvl8h{&f^ybxS
zFFYs9pkcyhfUKLcIT1A<XuYMZFU<3ON^mfZ?aE#2-}WY|`B2jk8DA2kn~^`-OOAx<
zQmcJsj{C{-wrZ<U^{HXnK1s5>gZCr<f%9zTciBSANx#i+&Ttwho28bifqg`v{^54Y
zXHL}rhMhq!6XU{Gq0pq{3H&ccXMv?D>j=HRo$8!u{HobBAj3K~?$-&m_O>i8tAA6r
zEtuiv$31vTRZ`o!*QoypDmw~UHS0?ul;E9&wKqvdL%9bMvPEkt=)kXa5o##wi+d1P
zIOC~ciHqpVg@@U6l_>rMMy)cg`bLyAJ>TzE5xl2^_-mZpOFWvp8}T#A!0M`+%fn)o
zkWmIa*SZSjD6M77A!*U?J1(7<u`6Skl7W8rVdQbzDP<LHM}`OT9=W&F>usF;HSS#l
zz?W%x5qf^8G5F?d-@0+JMdJ+%40N*}Im^t>1~VjQH~z#!PjGd=jqrLK#8m1D<S$uu
zn@Ii6;&QcEdX!v&v!p4rj^{Z%Kc{%?){gi(ug|B$)GZp0u0?|L8ZL=U9IEp>INIvy
zY8E1FV-?eBhJd4%qvUR*DB9^k;-_bl<uiUB6rwK}7o3WR5Z+TfwS2_rf+s7?z31P=
zaqwip4S_h)=QO<%0S8k-VPYd&ON8Ovuz3b?hTjHHSTtlR`U%`TFsl2mT@Nlx6m#M#
zKf?0BvYa1OM@ASRt~&4dbxSwysq+AA$LxxnZWwSA3mNoDGp>w1XWPhhUgG#kK?TD%
z`xCgdKU%_>iS{L+>1FzA(C0^-9?mX(Q^Z_y3y{Qqz?^V47(1_^oKmUk16hT7YE~4x
zI5Hsx<r)~><;eS8?x)aa6M9i%zGq@{`^7`AAFrv-D?Y_WaMwV6*%J-A%Xi%T>)T{`
z+hzJM=jZ>bJXeCvL<a@*Re2v&-fOAdb<Fpr9P?Z{Nwrm7-{a^7s5_$G5rW;d1m)Tu
zYijw-J&+cWEKoMONd_w)s{&DAM{~vq<@pHC-g=nTrM+KIc}nh+hGu+a{lrq-5nq-P
z9`z^s=*f8&rjx9rFR+A6aec3Zp4JJ950^>9eKM|q7tXBQ%wH6{p4m?3I;I}^60RvF
zjyA8;7p5gwR$hs{%A?jTu@fde;^OvTVMN#(SX!G_VCGd!7znrqrEwk7uq8Mk3&fjN
zq;AIgEqb4y4%c_R2aT)~5O%&0YB5)PUE?kK-5Iw!FWv;ZubS&D<n(rQbPdIxth>V~
z4xUn+(3@9TB|D!$!a*>P2<^z^w%^#dX84Vmmf1qM;UzuOAUklcM|oO|dWGMVeN%*-
z=wmeXA$U-uD}(Nfaj;h-oJLE~k%p-#R1Nc;KLpu<dGJeT4aUSS#pQRd-8~bDVY1rN
z;w}0jf5=SU(D3`?MIN#Dt$nW6lD)U(S>B_nB^wS2xK&N-i5gU_^&X6fl90^yxZyBq
z^{_F0Bye<jaXtPOA3XdwU(W^VUoXB}XH{v4i7vTAw5&8(f1g(g)r+`AQ<pz;RA<sP
zK``;UZ3l>blH(s}ovKMB*}%T@w;a$i8Wrq5A&+Qi)FF|*$ok;mM^l1&L07WJQF8L6
z(n%5s7x~Mu2XLjd=ku=%SFYW{Q$sN>h9&IZ>o?ao^;vm)gZ^rh=2Bbx;F^UL&s88L
zUd`ks<9^*J^B<hYql&wTWtPqsbF)9kl&c>NZu?_-3c2XZ4SV|DHc6L)GWyN}yS_0^
z&_M2<VtFQy4eVBsw2gLGsOO7cQY9UC9b$<Y=&eg^^x#ea_Ttr2=1lwu@|aGl5bo9Y
z(uDV+LmfiR1D(O*Y>Z!B*sm8YJBh);DVd0V3@+Ea9!1mcb~=HvUu{;PD=~DN@1DiT
z4lRbatEd@1s-kufc27?G(hyZ_Gd#?S(%qOv`C7BxOAN|2taS|Nyo2zN#v3f0=6_Sq
zW|)MzmbW99ylQ6p{>4+g$3Hq4xgy}8GU9FAMps+LoZE(6S_ct0@|6MY+;u{C=bh!y
zPU3!i-wYyt2a%-W>I?JE7+6TQ-95)I@;G&#)Fjxo&y_q5JaW}ljb1Zk@#Hh_i&dQ;
zaE>3hb?|x$oV$;}-|yA@>xSxQL~Z;@A5ZW*b$#E)9aggE@J#O=^LVXv&sj#lN!W61
z1R5}H=202swcWo0o6zThe5r?=Mkx8JQoP;^lzlQ5UAdpjKuA2pdJ=QmOy}!;*@TN5
zq%T$5IVkd;y1FxJ41Lj3B<}n@!pCqn28YLop4LPOVl6%txN8;t6+i97uKVO+M>n`6
z;NT)0v7QNbp|T-d?ey6XsevuG!4U1{)siM|b7LXRXa-6CGYu#t2sbco$LRVCvAGTK
zj6FpV4~q7do6VVx#?*?FkKzeZ8Xx0EYyCbQ(Sp6Yu>kLxt%reW%@&yLDOi!LcQo(y
zwk3theoH?-6irXZx_g^LX4}J>zbVvKm|Gn0rNm?Lj?GM?n1VqsRkvSi?Yt>sq|&+N
zhg%zi!5<`2g}mOwF4c{<<N+*+t5|lOMm^Xm=Esz>#xi$kcB!a*YF;_CZ6SKDzWSm$
z{w#pQpM!U2={}KfjOO3$7C3#Jz<PPT%*)=H-}&pCu@b%Oi!+j3k&UTa=0x#q%0~{H
z>Sw|!-6^<3_WRF+4!sI!fX^apzHN%?H@d0KVGvzP;L@PHhsnB~PvT|bI7=vn{o3Tq
z#cUr+ueNdnCf+?!Ov$Dg!)cW~v5UV9)E>?zzKaTWqTiw2*0KeHFeR<ND?h3VRWCe7
zdTI1m(Z+v_mq1g?VfN6d-cBXGa$ZBjG*4R7QU9>9et*kd4@I}?e|w~yp_;S8pZ<;_
z8h5%;JOi#YRjAYF6(z9Ok}S-kX-@8JPqc%H;?1R8zh+i7Db8t^DeyCLfLF_8sA22D
zpc(u4Ht@O{IU-~9Lh>iritz=Jyo%oqKD&+c-u93sIR=X7;YYM$Yzc;4i)i(H3i>j<
zlB#9yvNj4|^O2N`6MHaEAkG!uw={yKHMNX|V)}(21%A&xmjR0={U+<^T?izfU9WDv
zV-7w>%i0xAQzQE_RpMN-*?^24lshT;3obh4nj}J!=6eoBs-9nm5Ow98=6hW-FtHP8
z$^ZT~@asa`wx@n1?wW@N`Y7bEad{u;s?uUHI~kzsjH^=jMFX1J^CJqkQ<$y=LEZjW
zhodjI*onlSr@jvcT1H-2q*Ud+FJBE5;hLCdM{M}k20gE*kvv?Ese1#{Y}ZgX(P#+e
zSb67?Y<^<{sbGNZ!AX(HBV{y-m`C856j~w8Fx=aHETEbo>MFKca(pX7MHe0biMO&<
z^`g<{8<3~CuLxgX4aSxh&01uk@3hid2_}-Dy6`)+XiG1qUnvV<2UjlNCYRcI+)61o
z9L4F~v7PCJ|9IXn=y19TdA@icXl+y+%OLCqahRn}+Arl4hm8b8;Jn2}YMvPc`+T`6
zYGs(z94-|#LsSa1+7WRnMZLe4FNEusdF*7)Gy(C-5B;c3cX?M786W(22h<`HUNNvP
z3v9j%3-JaW@_qr1lIy#lG3fRjT1cEThkAa<g6T3E@TJEETPX2wI@p)v-2K}3JsQSl
z<g>f=vGOm%u{c2u6aL~$GAAltcZ(U-9N0gJL1xW9d9mBm{B)foSh!vBFsllf>w^FV
zE_v`W=qHu7GUofegq_`;kFCz;wR0xQBn_tB(SCmBT5#;|Vj8@S7O0*w{@f=XFMA1j
z$PzkhV`0$1-S-*U6V;QZS#&LdyuPu|xp95_TJ~wF-eBtlL}+xFQ@S|!Xh5qr3EEfb
zd+MuzE&_V_*{D1tk08}`zlr@fe8O<6WRB*0s1Pq35m1`P83@KZuG`W#MES1QVj|X8
zO!(gGz_TG^_L2P%`BTu6=Ff+=P`Ic;gQ-#TJ!=*bKNRv;L--@cfsW>Z=S&0bM-6kF
z;=J4ww6#6DX3i;aKr%pvo-Y}$OPXnnHnthG8O@DLl_`~}$uUx!?Nqf}$~o-5pwmNe
zx^pB`=8HTMhZj6~KL;p^dmfgN<gaIOG+}BGnmtQM1jH2WalbK`<^UfKAHbP1nE9I@
zDsXkVY2e<~F>!zl=QSdyMCs@}eb<+bXB#GR-CY-3hcwh}n1txHcE(^S<fIhgZ1mLa
z)rrP1N4Bsi0mo~xfB@%!Ktn0t16^Gvc!?4Po0<N{d7Dq;HIEz6R8!&1B$#}f<kwsJ
z$X#MMX-ZV}5VRkpzh_u<8s@=muIT5xBBKd?7Fu;xVll@-R*Gb3!L3=iWaRDKiMaap
z1Wgh_H9mn$2q0m4*fi4G^nM5n5VLTIMEKq#(p~D-Y3+>4Dikv;LiPqDK{3ufQ_f)P
zLr#DC3z||S3nny4no<mvkcYW5i`)<eR1dat|1{ZM)g-O^SQAq#BvYdsJXoO_yee$?
z`OoECDTEq`v{S`T!}W}*`^G#20-*5YwyVK@>>+Wso_?g2=zqQeb^#I;)9(0))FMf$
zR$tsU=|xNb2p7sx=Xj)5THcnI3eeA2uR>AH<9iZwPcOKiZ8(Dq-Zla>AJg;eU)ELG
zJvR~%q2|SR?dOONO)Z`A%su`3lwBg%Mj=Em{skTcaMM4@ZwQ626X6A1h2BNJp{YNn
zw1a_;lh@d#0L`7fb8?MNHUgYI3+P=BOexL)<{S3Nv1ak^9z9oJ+FM+?^3Gq>+38bz
z99pT1*ml7fN-O9t)l~~tkCAXXV<Xh8W;H@;H12@JAwk(FZYWQ(RJS&^=0@Ov_ot$@
zXNf{q#re<rW^1qOmuAWpb)Xh9A+(pr;&t|5xPl@@UeVfPpJc1HY&l#X{k%6*)t9!C
zepAK;yXsA?_#>qU_)VSKQ+|I{`@WehryJ~X*R8Lru;q+}ecknbB7zdWG%;MTs=No|
z_uZ@Ep6JVad-8r#&sf<<7+>B)fL&aXgC;M@6Z2b5u}lC<cZ+uyImy2FEBC3k)UnPa
z-F**;LecO-Rb49=>|@$M%r#q6wLgLmzfwN-TW}dRdg}_opu_`D!Lm0#NgO6{X}R4)
zi$EPw1#YyGr;hijM&)_Dmj`ZcjJ~Ib4OV=@&<e5tvt*<k;E#9B_@`Jwbl3b0f~i5X
za4tfvQ5625%i$~ZxZfXN4FT&4p9D=>Y1T(I4cVj2!sgsgmWPnKP0}?O_;bAcNJ4x^
zc}?|w5hITz6s$Xl`UkOY?_A7Bh(Wjk;QDcYwkLS0Y`dUj^x!qTu*>|bJbRKj>gY4F
z0lR3T?P7{y@6X44viB4;7ScXgI_GTFGpQZiKjqEsKg;emdKoK6MlSju@*Xl=YGiu(
zIc?lX9y-GdNr^dJ#3L?neloWk1<>%1)1|Kg2|G#eq=p#?llsh2HFBC?46B{{3l-8U
z`-x600?c}2TKAy%bOnl*Ph{pl+PX{p@)_yOt=OABPICz#`@qZh;MrItDC$YTIopFZ
zC!=-eb72_aVXSMV>#Jyh-5^vUx51&rpRFx4rW})T;R#McBN>P#FeK$&y`(i6<j%Ms
zV`onu;_s;TL6eL$3c2R*dX9%n67I|S;V*jBmJ^`O;GMBC7WGzg_v7Ga9Zq;QRKYU4
z_ffXkPDZsWu&ur0FS{w3)Uw;_+u*CD;v@*L?MmBHc%3tPhM$UhP+-N=!E5q<f<7yo
z=nqD~WKV?@rJ64)DJcI>+$SveMbWdPetm(LHQ$9&v}(?2)aO^mSY$YN{|3H%#H(sH
z#D5O5Da=)1mJ_6LzC|UPFZ}~9z7AV5onSN<HP7%1(rG&G76?}BJp2;*zyw<8sIB^P
z5eLP;?RAbe(m%*-;K0nmbRN}VKh9LkhXy3D1yPNbEPI&-9(>~(9DrEurk`Fz@44%e
z2**(^7UxYI+DiTB2!>Pm&st898?Z|Hko(^9YSILc_<gEG&-?ZGV5L<jxEf+cQEsSb
zt!Ni)IF^sQn>h3q&C&4^7OP@Lva9pVx`DnN&duHJq|s1pO}{3fZ@uBT^bX&S&V?E=
z6{O~WnU{F>pjZ?Rgp;|C2%bf$;kMYM6nTQKj%3kk8@%Z;CoOnhO)-03K*(}dIvajq
zf4R)EXv!^I@cN|I<r5RFz7Fa=a`W6M-2JxuJ9Y*6hzo3G^+h{$NbRh}xQH`@-n$#z
zVPEcyWXdZ}y6u<K>611W)_J4IMwx9C{ka*kr(|vA_xWouJ;g9+2H&a&4Xa(NGXlMd
z-dcb&U;G*|W|hLLe#~P5BTbS15&#cKkknfV8tA`Awj9;N%PAVKpk^aQGdA%oiOtxF
zn0{E5&J8aOC_qK+=koo|YMPFY5S1DiKKFD9z8ddV5S4F(5Br2KEDv=rHVMG~iEi_<
zy0;<p>)`m?V(isj)Gx&H*Pni$i~Ee1oV%^4Rh|kn-5bl5I5~I)l*20*hWL#Xq#A;2
zjk3b%1%mgz;_k<QKT-f!{%44IANDhJT&x{Lj(gEWk+sc%qmy;T)Hd$lua8KAazmYn
zMn5(j&mZ&t+z+^bYD8qJeaKS)K|E;%Qd^FMtwM+2f&K`0J*U(z6qKew;d+!pPObGV
zLAK}18cGQV#Dk)fw?;+T@TN01r}T;5B!s^8dm;V=x0LmG9<~V$1D)L8U9oy*l9qO&
zFBea|E*K+sQ#vqFHwO=_dmfyLgR`or@)~REK5g$+2iH3ae?3iuq0Bwa*r-r{XS<;)
z4nZKepv&E!sWN>h$SZCTd6(E&Y|Jv>#FbK@tblN~ZvS~ll^~tceOjR0%@vsokgV0v
z<5C7O5mzzjeI%lR3+4_j(qda9+iv;4>@Y4-EbF(UgP3!7>>fLGPTq1AbGV}nsRA?8
zD&N?D3sT-o8uBanP5>{`BvdZNSq#SaeJ#zlpr>Hjs$fa1k>edLyNgJ$h*O5?M7L&v
zRuQ3MaeVt~m8O<APYPa<_-oEfz{{crRysCSCdG^&FEP#X4gqFg6-|=eR9Q(*QhNJq
ziA4!!BotJ>oZ;Bx5^d022e0(FP3Qj*xqQvM!Yw&RK~dUKVVSUhGjLd1x=4OB`(+P=
zpdS7rn;{YUi7A8aPKLJVqlkV^v6bcf{`kkgiOd>Y@G5;%<p0!hR$)<n;Tj*hyQND&
z8kFu<X;hF-0f|8dke;DiK&eq+1O)$z#L$vMg9D<1Fd!iz3JlUcFlU43=3JbseX;kq
zpKq;a=X!tdn?O3lS))Ut;Z=J}@Bw^sxV%`PXg(usXBj5(OSv6PfrINEx(B{P-}S<v
zwPTaL*@zMu=IzwpmO}KZ7a!IxQ~(GV&p5gv&^NX5$ZBd<3xJDRxV)=*J?}pKUAX<;
zf_-DT;Y3v{xieW!Ct~~sfAd?Y?0s>~l5zjRd@^4t(CE;R)050hrfPxZSvtvxa(B=U
z1hwMe`|twNu|r&@ig?`-3%2=68Nqd6RnrrmZ&@tsO&%Z?44-3vsj1O0a<_y6b&4&v
ziFiHc)bY9W^JHo?jLE7ARd5C~Ar!Y-0mKET%EPmXT?UyY8i#V*|60_8NRo;aUqp72
zD6xtj{M6#YQm@PTe4GCRrl#F&O%UGk&hdv=Qb#?%W-PI3KNi1V)VE{><2@pu`?hME
zz4srO$JEZs?U-tI-#L9RrpY?Ov^pscAtN*y-9;V4D#fGGmt&^c3lQKyy1^5P)1P)j
z&^CP|p0x|l+Zw+Hp6M4><o*hMmCINRfY${(4GCP!ic^MeWq!M5SonoXZzYs+SA5pZ
zvDypE)&2Z<Mo79q>(VdEODI!xSm1fi>)m`9Yi_z3`Oat*8{=_3CXu7odOS~s$@0c|
zs$eK{%&S<eNTs_Ki*?rEixI~}WWI%I0fEc4gVC)olPl5d$1j7(Qby=v+CI_0UUV~F
zNQo^~Qh~eg-Dh60jQ=KC@xXGzW8sZ@uy;#tH8;_~_kNL;rr>#%(Y@g;(WHn#7FBL)
z-o*u)3ihS77}R4s#b7tR*$`6+vwU{dv0FYCiqwUJ?~N@6ad|5>;Qk=|HKn?A<zO%+
zw>42xLR$qzYS{OrGj?jkE<xOqwY4eEBNG(}MJ;U|rLU{-w`OmMj{tv$aoilS|KYrH
z{hR+0x0HtxuY-_2zZjoebuEFv&mJ4buVwoqY*LhAM4M$5%ITlCXanVH-|Uz8N;|tg
zLYr?q#;rV}*&*HrT*sj5<(=uF#L?ZS7#GJD<y?nBI4`~(usH#CLLyFO$MZ<c<cB{a
zzdSp1t9QK8=g2}k(n-!s`eBO}Y%sun1<T3m)~#MTmCwKb#)@%Nb8u#ZLSXmQ)83$Z
zN90_f(WHEv(pv{9G|gsx()8*ITjr+xk^cnbA~T>mcvLfUO@zNN#jf-;@T9=ifWbPg
zSB{+{<HGDM4Uk1)BmX}Arp2v3a>eCr1QC@&-}rY&4SNPlu@6(ch~AFRj$%nLT)OPT
zMyrzQWVp$Zd~LXD)<T{}&7DN`@e*MyxdDh(#f${EDtU{!)ib^;@b2I{v?JB-E9Zoi
zqRR#!pPV%mfzvDya8fR=f&IHC7i-@uBq5$%#u4j3Y)-_oq)Xy^l>s<sq3f}a8Xa0O
zri&P-H^-MLaA@C$9Vo?-t8M*mIURqvO8R_8%n0y=r8Nzl!`wvsU44|3-LJQQ5nWnd
zb|RN2DMXmxbc`qATPy_8vK^`9FS#y?^cm;$M-|vZSLFiVS?yqoz6JJy`?lYomx_Z#
zfKMPK4X|A0=&yN}YtGr_Ka<G3RL>9*gsShp39P$_sxL|JNmB0NV}E_f`nxTvSix4~
zm~kWTaH2f9$h>F$2|aqZyuZ;G92sxFthj?=@{#U`8V}O`M4(uJfw_kCqHFY}8ae$;
zyl1X!e!>PR4je11AG^D^iA{w{_Gk!`A3T<bh_`&}>T5I=<O!CoT$?jWrWaG=SY(rH
zq2~Kq33=C<AKe#~bEXgpe2kWfH;D`TQ4Dr+KWO}<C>eeN(Z(S{#iH$d$~4+UnEfe~
zj64Wd?geJB)f|@A0HZN_O1MH!t(q7SspbFB@5e_#r(=Y3#+2bY^6PHHnI;i1O#;pG
z>aZxaCIYb4hlf;XdcF4yC**Fe+Vezb*B{I;_2Md@CACJzPw>r5#sM$ItRZ_^+{gkK
zyEAj_n0)AT2>b881B+phrL&evP8F_NN8wr$cPc!*=b(f6w*r#-PeJ_*RNm7h{FUCG
zP?lX(YPm_77>Eg8Lok@XjfOJ(Seg&u8Kyjt^mOr)FHiEdVL+p}4E%SCD(6rp70W?W
z{u$O7Zt*rgENw!Dc#u~^oYZ)$tr=I3JwEBZa0_x_z5!(O#gO;j>EL1Z$X?X0L(iwa
z74civp?PpXnqja1=vMCZzCg3ZDK&}P-l4Gc35I_)BbYZ+xZP>_Elo`9vWq6X>vZ#H
z<XR7oW8qBMWt7Q-N-KM>wkbz0nm5&KtZyWs9UE+ONY}D*iCp;aYFZC#Kit>9alzuQ
zF2gC1PRv1(-dD>|oD#FI|2hk;J71crk(hn)V@^+<uJP>f-iHl%^q0HBg^|tRM0v+_
zAFLudz&i}k7uomy5?z0yalO;aH+TT6#Jj<O4}Qir;Ovkv{CFMTEI=td@+<Gl>h3p^
zOGQ=vT@eoYSq&1ky(HZxEgQj-M?)xeqD;9P<&VIHyOVyk7$-wERI-{)qF!h-`%3m)
zOQ!DG$8pM&XA{_{GP~1bVv?!*>#o-tj4hZu6R$#+wImoQ$(p%*>>g}TEN^3NgJ!}J
zUF!y}nFfb&oWEIP#a{N`GWT$qfzymBk74U`<($*xOcdHNx%1&bQb3(TtA^|OUTqv5
zqTkXI#o+&J@jlBUF}@24rL9yErqmJ-own2*q049f8tLKc6?UOkE8;?AEp=ou#LMWF
zgS*nj21!Ga3^s{ucr1Cl8D27u-j1q{G*6qn5p|6`u2voIIR70IIT(vXj3N&qc!Tgx
zf?X$kho;Ne`gF9t`fbzcRN6DLzNfRFL;z2dQr*}upNLcu!Fj`1uCCSu(M%sT$AGX4
zGAG0j*MgHW?ixouAd*2BH**4j-`f<<Ye^!oIWx)p^@uf*u6kQg+FyFB7^27{cm50W
z24d&)h8+x}JgX6m_I!9mN@A;%)jbX6#Oq56PJ;~Pe!O2~Jczty=(efn%#3bRh{dpt
z1%LY82ahyu7-sWnc)4R3n31nz)$sV1Y`w5+*R#;LI8MJ1zT*A9w>N#9J%nmW5;wD-
zv)-CAX!VTpGs|@@GICec9Wll5jg!43NVc~7@SI~;`6wBg!)&50MOE%MTIobYYOvCa
zl>)h~woQ0GrG{=#6Jn0Cs@uFwp!C}i_FbfM)I~HPB~i5SaD^iE7Vh#c&13bWLVS-M
ze5iZcm{<HCDI6AOYQT9CQ=m*jsG9{9r<x0-K0aCs{V%iG>xpDu#}jV30RK(;hM<aZ
zp1qNr=Lxy8PxBo%E~QWhBe!E^j19_fS2nNA8~pNq$nA)VN|~LA?r6?_I>>f=uOOQn
zQUk(}hlsL3lDWEw+Jf<h2l;&)p^Z~|?a<yI#qZHK<NQ#mY}og=W3gq5HX46Y170@C
zOwMge>20h7i%_Vk*)Hnm?`LmVqwe8ZxO%paRj{Z?05bme8=P;gJ{gvk7ua)A5F~lG
zu$P)ZZms^Q$Vzl0OeB-Attzs3$g~f$0qb5IWm3=M6btK=30<9)u^Z_E6<R~cF&4t}
zTf=qD`|EW{d<lC`j`(*Eg3OJ+L?5DKauiBidD2ofq0(Q{g5Xhb7Nq2e-Xo}lh++E<
zWwN@)u2{aWQHeH?C*Wr6)^|1}Fh9Ilu9lYEWvix<$_E)Gn>T(|f{x@4NgjIemKV4s
z0Byv_fWPTRPYveN+2JLLJRj_5fzy9U1tgzj{vk&Jjv(i8kL2ZV^3*`2O+tzT;h`$O
z30hI^zf+-VFtrqOOm>aIZ#Dg{pW5x8<KVyAZ&?kGg+%q8)xiP?{Y>Z#TJ!oh-;Hu{
z&N+F}FQ3&Km&@8j-1wNNs?(m-27bVQ58C*IVT-DHB6f02o3(Tq|69d+nvs>DxQ1yU
ziVwE=iAD(#v`Ivra(vtl*FU+3QQQ#$8TMuNScWq~#4o=4&Na|v>j)+;qm*0T)f`X<
zP;O>)pP5W!ii})tm{-kIv+cyHXX-n=;Ep3d>VDHRB{&;BMlkqo^7v@zr~$Dl7P+e1
z^e9QL$*yOWMeSKx<C@vxTo0xDpwhG7&C?p12zbf;N4{?cOCbKXYtXk239{8yu(e%j
z5RF?|2m9Ymt?YA^uTijmKF{YqEv1Z7=6?xg!tAK`?u12OvkVwZPy`s9&=V)EMdmw@
z;-JN?&_E%8HY>@|fPCAV@SrtpVSPG?{hsmBM!EQKU{KJm_bU6$&Tva`z%A6iyRKU?
zp}CxWfAID!_j*;ZQx2d!?715gBaSKJ#s7>rjRBxy$-H;Z#3)}=#SaV;%2%YoKY4wI
zRqM&;sEt!D-nz&yPNlNam|$);otj+cMV(Urxy{k{Vixv=?J>EPD0|b0WkRF-D;0kU
zYh~~O@w=u!Y?C74!&q8&h+5D&NJwIFDr2Z7YgE2O1P%uK3Qq6Rr7LqE^M6%Tw@y8A
z9`m?2Wx@|l{xlerUI9LbtHdb66zPizPE^00i&lAdP2yV}>~&C)#BwQz!%&FS;WUPI
zzsk&vCzJ`z>5S-K-m=+a{|LjS6jAfa=F`0*@OM8>-Y!G~7zmKw_Cx=y=-H9~$qk;6
zfQQFX@xKcGNc0Gi6|V5D;Ory-0hgexRK0)yW!gt+lGoayWa2e(Q6$)fo%XKoRfx`%
zfG`yb$iFEz$ptIqg3XPY@%oo5O2fRdLr>V-qJ~zdPf*zn<u{rg9;7!OKuimtkWy*R
zl29)6zu-ou3l#QBWIMd%S4+>m`*0$RtbTlM<y5<>4ZCqKVu7kF*f^%Z+8FMRl~V&N
ze-EJmeB<`PiCT-z^k;$S;_M?1u2);CH}#+T@)yT17?SF;^b_5gn-6^Bquhw>pPksY
zSU}h+X6j<mKeQ;(3|k(yoC|c;l+wxCqXb`z5{$dW8V2f02#d}REHR9&u4DR3`&*s{
zNxa#Gf&3MZ%W|zd&YU{YJTelbXxjAJ<9hK7dHY#0r8*s_w#*b4t=;~U_C3K#TPQ_7
zVuA&}8rXW{Yh*2+ctAY5T+0^OPOOkPwCem6)Fg;AIUyYC4=sA(Z6cj90yC*dviNVF
z$K~dbXqF%Hk&uAk1CFm`cs-${EGoaYtFX8nK9bigjMK`fKhe0tJs8u_0g=dw89dAJ
zh0S*(Jp)&Fp&{0OTzd#rYFI?s9mQ=fOc%eJadDFMnSJLa!t-s5*f|QVZ3BtWjOsM2
zl}ljg{haoEy3rR9jNT|TF_3_XvS_pw`sz2_Y&c!io%O~RX0#;YuSt;2lZ3I2*pW-e
zw5Jz%a!jb_`+Va*eSAq0-GQ>gQbsgk+-D1A)0%OGdeu7#-&>PjNcxy(@f*3w6m-+u
zF(GxeU2Oz^R-JsTT;4{+sobS+l?@V<=&nb<-llDA+{!IbF}lQ_a-#V`OrvM8*&k!Y
zYuvA6?AKd|${DW!C;b#q$D5Yyis*d_2b4bGbrF~ft@__Qt{j%VUoOt;^%1U{Jm)LQ
ze@o4kaskv(zwIi&j=B6EftMe&_F}Yu$a}<C{f)y_j9u!|Kx!0-X&UK_NDt^>gW)K_
zd*Y`8;Z2u37%;3y{j6F(W|>~qD|GZ{s=N1p8{XJ${kKnZ3F_f^`TXmt)C!JY|HzCL
zW^T2;bH}mjvNW|6mxONwg|FW8b?;jU{CUSXxZdio!4=5HTyEXzMW6=A2G5SWVs4!}
z9uKpGB3W(6!VTMQUx8o3O27fwbOdH_2`n2xGuiZIzD29e@ZY{rjryzi?otit)xNg+
zJh9Y22Jc-++P}_1BOeh&ww|hnvamsGVCUSrt&ajsT6NoWuj<R66PmAJKFv*DaBQ3>
zel2>MWy-lpdevCYCyQA^)20tGvRj}={&^pcJkI2>{4X;yX0NW|ADPR+qPC!_IaR%#
zBOpF+xxc!kana2$@L#bgxmGU$f#=0BdjUxx_s2vs?GFl#|2oPcevRoM6cwja0|gT4
zs(40+{3lpm_2Fuw82R4};<w5ep$i0u_}b^aD_avc{jO$E^y^zn<JsTfRkzKkP~=(G
zGqPL$h<_)<m;O1)ADX4Y*2m(Z%6V@Yc*xF_io@b3>z&3Q{kv4wwg~C@WId>S4^ERa
zmUsB)?t^6j622kj3g5e$3ZU<D-aWi`a}CFS+Rhq~PAVLvZxiJ166B%c7U%(NKyos2
z@{%$NlCnzoWRz5H%Bjf9i_6HU$jH$2pAh_?0za_3_oJ}?zaX+ey8|cy-PSkNtJZOj
F{U5Pf%~AjW

diff --git a/docs_old/logos/databricks.png b/docs_old/logos/databricks.png
deleted file mode 100644
index 15f64018e6b41458d250e2b9ff925b3e02370893..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 59004
zcmd?Rg;$he+b;?RDk0rS3?&WHEs_HQGLljPlF}d@DhfkMcPS~|F(4o?(jnd5-Cg@e
z-tT<hK6{_P;IP&#7WK>%_kCTzy6}IcB#VVkhK_=Qf(4b6f}x<?rA9$PO@D|6zSGt%
z5dyw=VlFB9%G}5Z1?6d&(c8DrY#r_Fjq%>TZS7)VLbrE@!Qnx$x8gm&n}2trp!v7`
zx&I9nRgBt6&r6T_F7w-WZ%>@Cq@Ho=4Be0W^PFr(p8Y8^YM=MVk*D<K0;Fc?sG&)^
zdN33M)5N>)0_lu>M#-OVq=r(PQlCEz?TUG9ppRM{LA}99?S%Um|M?Md=$RvDg8yC1
z$6wg)8pfonf7~#(?dJH&+WxZVsS0OVS(&Qs%Qj_swTA_)?9eVY^_@<IZVu*F4tZ9n
zB5KapDs0SqH0TmEnWzJ1KcK1gkM6t@d9Imqw_&#9sS_HFFG^d6wDBtj+6XQKvBn@l
zJWR|h1yT~tD{ACRH&PO!6;93_WodXe=`+rUE8LtL33pK%OO>xQDWO{6eo-6A!4y$Y
z+~`qIKKh`boL+;^D<~+ihbX9onkXnjF(@d+Hp6eILc#B%87jz1fzROMRI<h!e0pRn
zr{#cxf`NbYLPd#BBmv)i;0RTeez1)C=rNDbj`YknI3fyEN<!6TY-7^ZC1N;^|LS~{
zjudIqX|eEi@NE&-TU4jc$0cnq`qGsmD(p=AhUnWmg}Gk}dyX$8O~2f$*zSz1h+v>s
z>f=*-rj^L>h2brR(QhKX(BsMGnuRBvdhfK|hNh`1pnaB_%_iqw1@8+ghClvwpk!Ol
zb=YzM=k@=CkKd-8BzI6yPIxkMeF#G25*yy)T9JCsU-jdvjZ`G?izJ5jbjR)g;uG1J
zC1P!S4(`dFhbSm7gC<m!gML47QKAS`d-FuOW09WWa8KIMBNMmYB*_m0(@UaKm|ePn
zwe;MB1@$i2U=OKLP&mcnGjVZ?Elg_)=gj1B`mgXfG&JI)hKL}Fw1y;*J~$6#$eG=a
z=Evj9ttwTo!|Eo>=b;hs2~3|tuuz_(prU+yKcgfr5!CQjC7xW5pg-iCrJLxktB4fq
z`F(R~dP70OWe(M^vAUujnn>L4mSwF$TKTgl?p(1qO9@=1ekiSsx+o@0-4|MrFG4r!
zFc-PhjAG?t=J;p+J=I6}JX{-ZW@)&faxJxz{*iX&czUBqjP<$`CMQ_%A5l<#Kdjv3
zvHLM=GJ2-+O9vK89YuI*arbhAaXi0SLd`#pFR3b{L$^9smv4w730bIPrAFV&wsJsX
zd3*L%uiKt%B3~w$z{uD|*pe><)R|i`UI~x4<<9I5^909|a{EtLjpo<9&r*X22|?7O
zwGP9y;YoZ?e6VkCp!3Tevm8e2;(7@Y_jU;3GL@<YGn@Tu(J(qsyQG$TFCOCJ3VYK$
z!VOch9&KMfuzS$Q1GTrsAexf_D;Yeu-&Ry2cVft*#&SrwdZM2$CY3(@z>C&3pL`O~
zGJGG`mxkR%aQFT2_}3Bl*Dl_#p9wqbVdG{CaAAPTcn<}|nmf646=|CY-P~5jEA2B{
z*uzU3ayO=P3F1cyDzJ1>qcYq9hltazJl^87K5b_#Z;;6KI;Wff&;A|`3QE;r@ZtU|
zG7HbelDT_a`>MbC`pnchoRq1blfdIP6~_Kjn_qfPS&jDb7zS~}F?81cS_MNFhkU7s
z_^hO_!2$jqL=BqiN22#IwDj3r<lKd%p;9JvbBiaG6_;xs6U#{ZXY@V1N{bKQX6yXo
z)AvXFGaC1c<&Bag@TIraRY-hewMx|td=RoDZ<UCSCL>Z+HAX%pRLr2=xqZO-&sW!|
zn_FZHj}AoWla&Hv(Dyp+jhVt&-HGvdf=5m<{ILUV?G^4M_iu}bX1Yo}SQj3qP8Kr{
zV-CP(CbnP=D0N+!ep`#dpI7hyC7(;lda6<CauRz4+36TF`&`8O>FxfR$eY-%m%cbt
zH0hDuZed{yR&I+jcI)Eh{9y(2I6<q#sqJUB{3ZrDJktK)V1iru78e&6`pv`;6D^`o
z^SI2BnOB{;n@X^GIKfIaN1kp4zR!>N0+&lDGFJ#MGAdNt+_IlXvi|3WL4!VUYCP#;
zKvZ9Zx-7j*)KR1SexyToDRrFMEm_FOZWgBipR%5>K+^60nAIPZ`|GeoxBGr5UV6P{
z7pA51Sx0Xp7Ynlp`E36Bm^YZj=hI1OxwY56uKHEM6X=T_`-HLHwK<{$+a^>LoaZPg
zGcvo0ij3-o_>r@-rVAx=KH-~Xg=3R#;W;iE%MEtO`)stj2ag#|(O(icwTz{2yo~XR
zS&a7yjrE$)-(2%hVXn|#nDf}H7L)3S$BFuZ%Zjm%`P1CJt2jXLCp$2xG5Eo4Ey$)@
zG6^rT3sh<E3c@Wd#aj%(3+!9t3{7>>#q2)odj0k16VZ$2uD0rxCrt7VwYm?rgtwM6
zEy%c}P+?Gs#K;tPx%QX**_-)b2ke%JeSzy1#hxs(Y*r_o!>Wu~A`pu-D^z!GD@TMP
zo+uJs$j9TySn60QzV!GiNk;M1Pb@4q{bjEAk9Tv`g&k{HvoGR{nl}0fl_2(|_^}fq
zzBuZ3Jf}Hb!H<%I23#^K+Vjc;3(=ByMZe3Z_v2T-{h>`<+R-K}^JJ+@C7UeFNAe-+
z?WL_NJRze~_M%*U(ZmY#>Xz7@APv-Unfi`(;I6yV^)f+eB8QjK8u}aK?D9CPmHxg{
z$`|B1RXnqT|Ex=deK8AEB#DTYc!6}VBj%U&$JP}`WPUI;4Qpe2_n|xN`gi46*(?pe
zu$7P=SdafIvx>Y1BiCn~8K@{xE{~4O8|H5Q?8&6Glo374N<*zV$rn&K=lIv!^VgbG
zDa1S><4qjeyspExr;y@0FPV&5UYT~Wgr|#%<*kIqXjU$*=P10M)$M+Ho3q2djO$>Z
zK{TSru#QFG#F)1mAZS?gfdl9m`iQPL>I_c;4vomzs7$x*3eMm`*Up6Pq0O>`J(?fa
z1%ackIuXVXgpGzGbz_A7$_CkGN_6)W!(A(+Ry~x0`j&XYS9aYnNQG_0cGn>kjmekU
z9&#%N)>khaRsXY2c6Z(-Lrl&b17Y5?(p;BS^?d_x|4uYb5!&J73OiFi(%uTMBBbeA
z+dh?fV6-EcgR5&|^OtVY;o|9|8l&L3-83!>XVY+(tstnmIYXc7_L`(XvgS2?j`esl
zrNL!0?Z$>+PmT}!BSsU{+aeNU@-@uJGUKyO1Wd(EuCx3P6I(E9{QKYMR`T2?c%VTB
zPS*`*22O+{+4!`|9}&gsE5o#K?K&holRFPw5Ut77DLqGO?(D^Sg?l?9cApd`Ur42^
z!KcFLq%2F?2s7AeAb1?0CEb)m5?Z1K+)B56EumApU#jnvLjyMI-sl$&<J-w((OR#f
zR~Ffos*T`9AoHi+y|&2JX<RGkpDfQ!Yf#8$=4OgA_Cw|c2_Om>o;w_d)W@`G?mQ#&
z*DfB*2plz^ue4N{yX%wB97g_Mmp0Km(E^jJ^epUkxG)Q{FXo1}m&XnGQ>b<9BjY~9
zwMh*Lhh!*vH^)rX#<VA2n4Pe3QA%=L1)Llvbi$mLjO1n4BkZ+s&6^~pSoynUZ#ury
z5yc#BrTj!mR1f-(^^hIhgumjT*MNRA)1Bt{0Z-WblY~%$br|{{UdG%-n4u-~ppL|S
z)({T@dD56%#bdo1u*_3A&UEkIq`gsE^`sOld4`j-*do-G_7ywFxA67%vUj5S$4lVw
z!$Sv*rF<nXU8F(n;*_dlbY|HJCCa~=WepmuGhU6EzludpNLWvjVq{NBX)7-wTG!?g
zD3Y`=UPF`4SEnCvMPH2gzDrt6CP3sg7EB|yqp4Uvd=CNbXh%~#AfhASN}|07>s8^w
z%l&GxFUJ1*bpa<kSWaHslT3}_zu%cGvF5>Vp~OTo@`A!ppgUb}6^f^<ZInLeannm*
zV4a@8<?tY#>P`)fj33;$t<%OgPDXnhJzyNV6Lc{|A~cI*lu>(@Zso`AE0u#TmDx<l
zlb4Lv(s1oc#)4q|@+D|~{d7*?)HmYWUh~_E`;!TwX=x!$Et)!6kd$LspEFl?Q;N)0
zZMk9P5Xr7LM^v%>0X<9`8)Kb6!$Ruu5#J@DcbS|h9F8gPFpr*1ef&K#3O7yX=QE4D
zclB(vj6ieN$DR9YHJ;S|)%VHL&=<i~1|b@-vu{8`vve<yA^grecc(<1NN>5$m;6jg
zn`n5_+h?wvn08kTbL#oU>&`71MG^`Th>tUr!E&{)8M+MZ!~m|6DKF!c=>(M|hCp*6
z6YXwW_;DAh=cTHbgCll*Lt@@|9lvyTVext+-^_i?e}&oh{VP8-|J^u~9SYAD{aTMz
z>Fm6bR8@iqhJY!>aNgh%#6BDw%fUYL_p+n=QI>hhLY10QMLZ_Vn8_}a+Q(&1GEi0B
z@s+uC(bH}si1hBWiDKk^v@p<r+t@T?f7YC%CPaoU^-W)*=>5GdmHZsXE!+51WY$E=
zUKF5gF@R-KH#S4Pe#rO3yx$*NcbO>FN%QpQrtCmmo%)s``EG~0L6N_lW!3W_B~uvL
zdP?*&m=QogD2IkI$HC->Zid7OI*m(D6RZ_p;!%+r59(Z6C&u2ty}W*s^|u9%&hZQJ
z?UH@j_;&WLb|2k}$JhNiR}YX6*giZ8)3$u^j)0+2UowZ8mYw>9I636%*YkE^=RzGd
zKIaz=J*H&4@0)Ng{<F^64ZR*0_XmF-b6+IQPT%<+`=MRK>b)^EjIoSR&Usjgv8GjA
z#rv4wGIg^|p}ymD!&OcvAKo2PPN{%p&34G4A%8{PZIzo4150*p*q1zG?lTF~xn-(q
zqK(QMGAz1|;S7>D;%_k7zLU63i0NN=x2Km0VonYF1rc3y=o$=bZ$S-wH0T^qC7e}v
z#cy#bR^mj76dEF<4YF)ZA5g)~C9hkq;(+)i{@7{AaxSLew_VY2@8EpM6UDlm(#7wp
zu$>?mKHXQsV>0!t&E6MWtB%Oigo(y)wMqv4o8ar)qbhR#hzi8;$Y>Auk#G8Hx^SUm
z(1^9Go^Q9GJ6xA>z2say&|ICen)(-K+YKhGE{SR3T`0|7#u#9l#|49aQmOHHghhP|
zYOMR98_%h@e&Ng%dl4|k5L1P3AQ;M*!wnYj`vW)CSrsn!)R2BQWf*o3U;_27vYPth
z-{5GR{z(+_H4!7yS2EcdOfucVF;AD@y;;by_u&&+_#=Tz7M6p{ZRFct7`TGl$Va!%
zdYs;>M(A|e?Ri5PKzaYTORl~9_w3vwV<`Hb;HAqioj32IK}D;WchO&-C4y86fSTlY
z814R=SId6pT`m(QL^lc9t^YRE{ux`}se?bP;glq!%bX5+;D+&}#+D0kgU6Ge(<0r4
z+SlAxwnZ`AWLzV+3wv0pvzyLj>1Cjf_F<|E^PqlqE|*!84=B`Aq|9M5$nIpx#vv<!
zt-DGuN}^;>;H{JpSn-^lQB_DrsP(yC+VV*+$)ce~xMdF)Q@6=>V@3|%uE^=dCVrBy
zY#s3r9-S3}p)UrZs3^^!T5wQ)C&@ioUsh2wfiZVH_O~%bwte6QP4ewKn9$VDj~rxY
zti(h+FT26`R+ty%rv`Px!#n(m07B!2DX_B059U8~m3vdntjAu$D7RedIXK0#F%mGy
zYcLx2cw=ie>^hR}hP2&q9u54C06B6>zQDNEUHdRm39*#TTmXNkr5uAYad$;_J2Vid
zhY5KAB|<GvWD;OnYlfer4U23b)kRq8=JC4q*PotsPe&r}bBNQKm^L#7e@MBf{2xN6
zC(;rUUL-v9lv13GHc2>v&NzjOiLFa2Ug+v|xM~z2l)8A|*D5wwarvyp8aZ^Y^7<y3
z?1Y}6oip%OlPxl&3WaPQgm}CddK0<iODw%EdnGTKB;kPVt>_8oJ99JepB3P!SOuip
zYYX+`c=G>=f5x_WEGR>c|FrHC;ZW5z&};-KE=s<@cGxEtdNiJI&7^@%MY6IV6;F<z
zK2VQj88kLQ{MZUymlXTJTIh9sH@kYWcp8?(cB%6m2Mus+P{2RRS^uOrP8Kh|6eSX5
z-2WcwFa*(vbaka*2_A5LnNqO#V^;K{`mpV+>iqRW38Y5sNDT9cO=)$FP6p==GH5{a
z^u=L6%&u<Lf>u(Iwr|~E8&9Aw*4BqK_|ffF5>v=wrS~PigM7_sNNC#8Z*Lq2t@W^b
zfwfvyj!XANT65HOKV^UlE-P6*9wfvgm%In-%hxmx3*3D0hkKPMo3rP4cdWEdD_M=#
z_7#DXtK73nTe}dMbcn~l&iXq^>=VW6moDaSCi>;s><vbj0($0o+$>f~dpW;#)co&#
zs=6OI2{3-*g>P%qFF_$ST6^1AkEsd*McU|A>fzyL`LD}VKR#W()a~pJ;gz;~_+ysj
zmcV|@ouz%w%mj}>x5tK>_AAOK^gmA}R$P{`xC_@^P`()3L2u%xWHK`8y3*Hq&UxlK
zVNl2fiI<R)luCd5=Pt_3*>5II6k7#yHab4|YRlzdK{s)RVU09<H=`nf?73^vylYX|
zrn<wWoMwGn#<+RgoO+%RlKs&DelPvK5aAB98Iss9uZ&*hv@0%<UKP8Y_Hog;$!116
zcV|o+H+;YQZ@G7M0{WIc`NX963;;s)Na`5kbI`Y?@?t(viuJ?wOS`*dtt$7lx8(8e
z_*hxUCScTiG#ye8s^Ykcm~fq2{?dt!p-$>KGB)jYlQK6iv(q*VCr~WS311;{bi*V<
znHbv1M@Olgk}AlL8bNq5Akt^dR|otxvjxPTZ(1Ioy<@?<DaJaDnO%Ga`?}B<15UYP
zlZ#74tjM=bNS{4(mPftg!7!OOL35T+ywaX{u=2x{)})XmrZ+2?O=`1fZ|UbXwR9Wd
z%TFW%|M*aq7A8aeOHkgSP4y?I*o$lE5UrzvZXaWR?)$hl-|X%8iRjk0y?!&8WY%Cl
znB!{Y-2Kk#1}AhxLzE1pE{x)3Y;=-jAZGV3O=T*tioN#z<3yKLAa3qQ>#|%QQEbN>
ziC9sp7Jt@?rn^pX4-|BV>4&wDi;?(_j5??q+=35Qu(edyEH(hzC(()J1acXY#S?j;
zQvqKt#g-p6WwkdClW@TDu+!~Z>HWL57$_CPVTgkI56EV6R}8i%uxW@Wd+h8HO>T|g
zUyL2rFE#NC`L{NGwfbGv_@zzpP%519c}Gd}<4U3A=#BtI3+LtX!xV{p+<&BE!j}iq
zx9Xo%{}J4D671L$1&f7uEb8Ent5GOEG91&NTU?Vb?HjZIkV|#D5AsFl&o{ik`LTxS
z=u!$q4kSHAZk#bt5!rp&-*p{w;M48f9tkY|YDX%n&i}lCJtP8s@3DWDQY?XK;HTy_
zk)j;be-NZnY{-mmt_ro;nY}l|8lp&VSg@;M;!6Wlg9(po^*pdSZ#&wc0v+}1qlB~w
z9<2)K=_vc;&?d>dlXy&#G4G|aLH$j{L8?xuX&Aaa)S;$j+8gMBpwi4NWrmC@cIwnG
zdr*1>iDW4unR%m7EqWGv9WD!YiI;n+Q+fEw2DS?B2Beu+_AQm|s*LxBpw@+tJH;=s
zi2k<-M|5U!={c;v4C<tSFENU>Zqqg$KxigJ+_3qm=*+Bz09Y-bz;mSiyf>#Jtx5CC
zIeu%W^{p{;LQs`ct3=h63TdqyAeeli6TM1#y=LL>nUjywQ{Yj78?=|EJezK4A+r7}
zn}}s*&Y4?S#to|$g75D3=4!Yl<Mk^e6^<-Rb+mc=$dQ58D^gM$7hjQufwp@rFOvJG
zFQ3OVC_J_ni^rQPZb-{cuKII@w!6U;sG8{gs-U3gsp8=jc+6m>JVyaF-iOW&KKm8r
zbiiLABlRFOHo+KHBaB6-Fsw(lw@Tf}c_W{Q@k|s<*NA_`dK2B*eYj2oiE7|W5T55%
z)0LRmQ<KT4c=nHu6OOXUd*;ttB(*GnL;r&l+J8pue#%Sc7wV`dQkx_PXuKioTKlur
ztNhE(Yatt*3yo^-b4<-w-C`@9=YJ#O$E|obCPNzPx?H<6@tf=zGU_>??Q+3kx$VJv
z!>&O|LrzxS*x*Ut$UfG&kJ9<IO?ZMyQp~|geppF$PPc#GgpH9<0akhNAkMv!_YaGj
zI`ahS2**9;!f*Zt^`0aON@Z@3)eojt^3fNx0aKR&U7X0y_<VHoPSHj*hi8@j;i*IF
zqlXD;TqBid?~R{T?);hB|F*oaTLXHHjhUn*kS}aonz(*}U5uw*NDr|GJ1JW%MBL7I
zpI<8-$WQWCQPPw)xVvU=d)*3)=Heo}_P(x3zG(Vtv*z?CGrOs&$NaKZDKYxPI>HSu
zAv;b?WyqPlvzsl638W{+O3Ouuv#Pu1$syCup{%loW&fJkjMC|M3_T^Og=pcikVoUk
zM2agUgT3NeLfWnUqm<uOBd<dBE5%-o6rI8|&D8<W2Ml_W%y$~fP+qei7-Dy1_;rK3
zMF1#2xGWuc?>Iwa<0K~f7D;UXp}wklu^l2d28eyNRTe&$_*siGS=)`$eCfH7b+>ug
zPzxeWgDCJNkS&$At}4<6MH>BP>#3CkouQz2<}QC&@#CeOv{5>Z7%DQa%f+f5o)QWh
zY*ICPlQBo|fDvDQZt{s)xZ_pYm#aldRrVZBN=XeIpEQxU+`fMVsi9L?rn%FV_9Nr)
zdM1C2hhlV#%Sa}xZzxt~W*+rk+?eK+{B87K(J~ObA60ehv@4BLg0+jDs&w|)twGW_
zXZRoWj5>zzm)tXUIWe`@+LdT?8q?U^dIweemAgM9Fn8zk#KqbNbx>Qwtu^u#+k2es
zI>|8kZV{XQGMX>ERZ;4LVc4Z?wR^ptRA;&8>x&bgRr&XUpG1;(2mLrP%{^=Bl7!!@
zva7b(NpxK2P2(p(h6m{!Iy8QGw_@E$gE*vC1j<=|Tr9Kj7uW3nBMqu7|G+DzD8?Yl
ze$iP#DaJbYaCv?|Zw;5;w?%{b_U(CbAP|4}Rpr#o{`Iaz!!&NcSU^lJOJ_joV#G(g
ze2FjcfdVC%J9|RZP&B7*XMvO$WUt6}xLZ}apMv5=Yh;)dcJb%#?>av#Fza{aQzIit
z(|CAE%6~M3*$Ni8|G+hu$}6P$KU@yFFABGi|2?0O>jpg}JYfy{9DwI-S0&1N?aj55
zukU)|@|~Q&6{1d>dBVr0ENttqhtTUP*61_ph6=_B6H?8`FvfO5{eWR`R~#By@E9fn
zDX!IEh0UJ&%ptgm5Iesl$XBQke)F66)0mMjQzS0NY_W++E<Rg3>jda%nJnM&`7bhn
zfAI=aPbufA5T^GzavZ?&C(c4v#gU#vkQgFTp5fnB>ymORasaqw6pXQIM1FB-=pbG-
z+E%cuzmA^VnMotrS(h6GIsj(;ioY6cDt0?N*YgcK2Okg>Qzk-o|MEbR8NIrM@H%^)
zCZ{f$HYlcv%VrHida55{3fw-0!-h3SDXH4<71tNe)BIQ)B~e6Tz_TzmZGPH?C1WLW
zNFyc1dgB>DU6&8!b-QTFyG|9ooGo(()&x4w6aK1rer^_u&5hwWW6uBP1GBs0Hk8Tv
z&cpw08@`x}{S$BZPVqi*^#|=rK}7ki->K;N4TUNa5=Eo<*PPzkKctONtXwWTjGe6!
zZ=E4ZFeETl2#@s~O6TGKI|x~N--yE{wYI20mn@7=Jv?~vywd8}XDpLV7GQ>C!hHw2
z#Ms3rJ=v;;n7*|EX{XXmw1fKAKk?=Eg6dORG-(@LF@VEiF>Ey*(dZ`)maPmn?(>sd
zS`mwWrb`5Y#HhXwW8MhxgzMG;p3|ui+MTouRKe1D>vuoR)1AtSw~nXxBK91-I}7#5
zGXuw>)BPzERAZO>tR}xnT#w~(_Vq|IxAHO(X84JWn$F)^T?Zt_HfOG#@roDmoH0cX
zDZvT6!YZ!`80$su`C#2PH=~C%@1cV{HP$g1kw8yU)IInXGwFSma^K$S!^a`#Ke=Z(
z8MB`yHqC^=?c7jLWaIN%ub@-x&-*77dUExgT85vyb+wbe|8GH(V|~(zC72}XTh(Ij
z#ES*>Ty{5Eyc+vp#q7cltLlY+-rl`MIhw3Zp*iN1L(`6*4*BWNNP<5vBa2-!26rQ2
z+8On3$aP+EYK3_B*oF3IX>(@Z@YgJkZCTVsS@JYJrTWV#tz6{(_-oOEwQp$B&DS=#
zsO+wHshby>LjJu%P&|8|MET^^VA!4gnMGv&8zaI-$hQ^4?Q&)%zXZa2|5bfr9H_+Q
zgQ%srN<Do}Tl1HBQlr~+MBmC%y2IfCi@tC6PPdK~24`m<GhIskQgxhS9aay1Y1=Zc
z;`u!V8KnA&t@}Ezgl4bCqv>lC=p5Pd8d*NtMLUllQ{FQ3?o93U)cpsGdw7AsM5Tky
zZ!dF^sgL|ofKR`5S~=cQo|haF%c|B<@*SS-0(=Y@mpjJ$t-WL)f3ZgB=-JCLw+0$z
z_B@<Xz{ubDb*gL;KP2<owwj>ck`Omc_Q{HH=e43?K#{|AE=;}INqLP+v39_U#M$#9
zZnkI)=A8LDdPeaTea4^x@!Xbz%DH<)9BJv+&u?qMM+8!Hu)LLVknmxX6qEQcL_>pw
zC&c~YQbY!$GcK-FmBvyMyWZ>dGaCfZ;aJN{xlQJ<ZiT9yMUmz1aG}38L?KkcF)Z?U
zKpfJ~uMeRIohsFZgNiIUORp|AFQi8NPs9;1AdPiUZ>5bOt#pg{2<%6S3!kI6j69av
zU#(drhwXJg0`NKI5y>!t6i;xZFGhNWV3)k7Vyp$?n3}k30}2cVH^vDgE^)sy7EuOe
zU90fOFf!UGV)F7SmicK5UNW&*^L}9EuHf)lK^KRmDToZ61pQmtkQ?zAfd*p7`u`Eb
z=)om(p-NoOK*g`%Fzv&(^rU&96qsaO=!k>H7CL8km085Xr?v$uN|(OC>Ls5{r3r6t
z{87p54;%7GCsWcI23D2smmxQRXzcm*$-1NzcAMHF+MnKELE6=G#<~Ybs@ZhcoK$tI
z73);bhI%Bk3lG)URq3mgGybrAV<;UP<SO~hwe@(3IZRJEk?C*#cx}Bjd@8w<O*>9r
z>ceX;YolSwcp=lBYq3qq_yB&6f3<NUU?4d#9pMMvL|i~mPI!E2%4;H!{X-~-inRyT
zR_>Gp|D`slYB4|hF&lAqA>#wr&M6QusoMU5Nl*P$roEIN&|J2E#^|5jVZSOgId+c6
zuadgNark7#f0hYG01(Jzs1>G5TrnG6pB0c9CPP4-B#h5UyiA2>l?B*ImI34gC7`*x
zxssdJBj+%I*RP(HwoXgt<h&f@^<VDBndi1D?3>t0)zMxo#KNBM!noxVbiPlLVz|>i
z&C%b=o>>WKYaZ9Md6kHsw*WiT;KjfZog$u<y>zw$9G*Ck#S}y_zZTA{@ER+`V;i&P
zZLxIDQ?Z`|>RJS*9dfiy?=RK7^TZb9vDXD%CzJ~cMCgVGn;iff%&FC|G&HI1u%*5l
z!ibRc0U9}X)wIW;{s*jIPc<UnECk>I3RwcAeWxv;V&*&Be`@GiJ*JZDryFfx(|blg
z%&ndP<o4KwfIdk>CzHfJP2%qewC4H`w=(yPIwlM4CEo(ZAX>(x>+^!#(R?1gqFwFG
zg6Za!fIw|~9;*4YO5&foN#3C)N?J&6M(E}miT5>*B5elZQzT!O|2}`lPu<NMicCA)
zKe<?^rqpR)e@z7JNE2ni=Azn5?w!6UFXR*9xgred-Dw~(4J~4C?3ylbVvgMwnnSqs
z%70Eip_h1VFKVcTo4sXF>L|qaKjg5fL&jS(vFcATtGT2PT?_%q;F7Pvj{>I<ieDF6
zUR~9+@tKNktFRZ(akArwn%OlYTOZuFF)`_!pwo!F2>yR;U&pSm?W_vvt?6YvWP@AX
z0vQOqtPRVGy`>dU^eru9a|5+4t5>a?1_?j&WkCzF1$Zv}i02pD`%n(|GIk4HaDp8<
zhu8fv-YnszR*rvB)DAz{QB?05+)l0=@wWIpP)yKgHU0j80!V1d6%1`h6cKZlR|`)~
zzF?z#?D?GqL7A{eOTNbL`!jVs^tWE*5I#1!I#{@f<V1GItrK->CtXy-o7_5Tv*TF>
zM@UOE9U<(QRL2@MLi&*bn?wkKlxAJ3XHKH=GMyCqctbY!t4>;$0B}etsw7so{!Cb^
z^FI|tD)FORY)6@<fBxlTsP0=hHrVj3#<YEe*MS0{OF81&rna~{_+KZB`Iv<Jf!Y!m
zzK$46p=0caB#Fm7B}@!a8B3H2NA5nzRxVh-Q5LM7=%-e0lm+yMC=;$?gEmbLdna=B
z!Aoeu^~YNT<%Rrlae}9yqEL<JTfDg|c_3+R<&s$Dwn?*USq(LJAd@3#(V%MNOh444
z;o_crI;`3Zl*%iuN(cRb4Kw+rmlFJC-tgE2Xk#;*3=<=FbCMTC<%UjvV`v%_!nPrM
z)P4B+YmL*<4tXWrI^beYyAGYyZDQkRdbjADdhyZ*RH^1&`Po#u+e%tN$+|IZ)hOmM
zqx4I+(CfYWIkVAKjv`!A?)=_7=MADk`KaCXMdP^`KdLHxH%@6H0H@MgS%<vdqATK!
zcfA^Mt*3l}><!@4{L{OU%Qr~xl1J6T68ym=qYJ7xi2ai}Oi6a4@sIW7QLU1f<buHz
zMN$r~U=<T4O+Y5D!n;s%cFl!}JgHtMAsO>LehW#TseBwWGOp4koO+A$rk%f)H6y<>
z#NS;i4oWQ;WUW;Hzs$4nJ84cCE$3HsLQi&Q$)O`vqwSr$o-kl!WTt;&xCNBKnA9oB
zIAo?c^$_Hb@e1K63@qW!!bb^xUgW?H0@6P9;<5-tBkGz1@#|ff+v(gFhN_9Nhj@*U
zyqh+=hK*D{yXSlWL>HhSvka<HsjgY1gl&v|=KH%nSqoD$%v5Bw9wfuAq?@^ad3Bdb
zXGF#yZsBicR?^fmctzK3qMwn&Dk0+VOs}B+t$NLvKhK^R^;yD1b&b<x(Z`1<wrOuM
zh){mV_JZ?)N_+&(=MxU%P$C0iAb)NJM96eEk;hH#gM{+V!{YLBt08iNrfl}oqm0@f
z%?_gBQuYRvOuUP)`)Krl6l-5|cS3eLID$j<NqFF%qD2qA!NzG2|8v|9JSsAo7hgu%
zHzrdW&rW`Qcb#}$F~>EOB2{79(k6ZC2ZAaQeWO=xM1m%X!;(@PbM(l<hsEn`$LU24
zEa8ZcVQluSjSp_Y8k!8F=dj4>lU!_sN7$jJ_))}r=Ep$(-ktdT)j!95guyr+ao{Y@
znXMc*xb6@CmV4Lp-N};kw_OPZHp65#6`>Sv*aE9t^R$X5o-i^5Jn_4pbub5`&yi;_
zE{IfQ9uI%$@ckuB$&HKq){{yK98`E`+tsm2ExcOEC6afV&mvsD)y)^`)g7d&s#~fs
z2Z#&Oz5R6w1yL%57p>pDjj|DeYkL29R^ij%Z-2?&YH?1Dor%)vIXqi3wwRJqWK4=!
z=@pgnBY!dpGP(@<t^A>VJY6}6v&jfiQc6#cN?>XjiHTq6lO}pU4;q_{190C6_9ot+
z=flJ>59yOR@yT87GOvX~o&i#x<09uLo&NA7;gBzE&d`Z~Q2mwv&J-l)5Z(dKVZ_*=
z!T;_Yem}m1g4`Sx+0s&4h4yJzFZ9LRBE?NJhh6?zuC`lm2LCct8*%om=56n}e8N5#
zNRn`F-7$-%R}7akx8}^zePSnXG^z9h(c5k_8J*4)ptxy~a^OXt#=_2?o%d9Y*!f!W
z<=>M35iJNV?4A6uhI3v1q`p@0kp+;%hVvYD$B)Mx!?eSz{xIb7NgwhX4X6FbdTjE(
z@J!Uvqxi13adhS;_ZPPkxq!zfYr0y$E62Y~wZpA=;K+#?$j@-QFbaSM6pmASNqqJP
zwH$V~>jzYBt%7Wd>mh1i8Y>q#oqK_V`A^~BF~r-H<MOm>(L%Mlq9oWfmUpw`O@>>h
zG@2jWqi14y@z`IB)Xv@$2&5Rl82{2-RXbf!vP9}vH?mnySQ3K3lHEH0Xyh0(@x1s4
zMs23S#T#GK6b7A@;98F?92;s6hlGf`1P$feN_4>aWd#wZ7<gFNo-LV=?xf*jg(-#b
z(}NR!`w(TuDGQ-PNk(~5-ZM>i6{lUdXN}0*yM~)8b%TH!X)~`mrKrFBCgSCsc`m53
z@Oh0d7g{z5Mrmtv`K3Fb-O8Zs43wUh{J4WywI{h$q?DKBrHq>#yK8?6vLZ2f-K<<f
z;fu|?A5%7QdWKauXEh2vqxEt35RZ3<%jYX^d@|rh&#C)rel5q9b*wqkiaJu~?@;PO
z|D>9z>@k9t21JDUwobXo_5fCSg?YwHXtVZm!#|~Ltpa(lL*CWaSTI76QPAAI;^=~Z
z2gT!To!~}Wi{GVa%B^{CnuSx!@?z6+n+GTgNSv-gUq31w<tO@}(TmY8j@|=}TKP`1
zi^rOPn!&kVR*_e~0ge6p9GQjK9Jl+~=^y2?`ror2M@AljeOdOGb$HlK|3Xn}Lr1(4
z=QR8AZfZ+IXQJn&kk@;%VhrL49Nbl$BV>l1f`&*IVp55Q+DyKA!@@f}njtXJRDr;X
zl&k$KSsLHUUY)qV9UAyG0nF(OC1RjNNP;wn1!V7g$^(?&%Le!Fp=^<){g^4qF)v)(
z_H6xaGojO*uj~D)yHqP3A`HTU^99&6S_~P8-tQHc=3ilIN8vDSb7PZpv263PNS;Xi
zAsYx}@8?}lzpu#a!{nAT6$%YK54n=%_yLxfmik6XK;V@>z-=sAakpHHph}{Ay4`I7
zhjsInGU*n;f&8(z4srWZgw-b0A<AU0_BAP`-d29=dc<7U-tJRjR>{KmKHFJRL2->&
z?u9`Y6w8W}0X=28SgeTe8OnR)m5#W*_pQ(qXm7KH6J!#|{3^6>tH;y&>EjfaWy?*@
z%L?~hi*H2QHI>}q00X%Q`tS`wI*q99#o@e;*EPZn<<rf}1?J7HpeX$i$GJu)#<>t9
z{x$c#PgT66k(;j})9xOwA}yjK9EcN<faYc#9fuD&xEemGr-gvXR#qtnky!S)yE8d1
z;t+#KxlL4~S_oQWFZdkvA*BLl@<(R)o6056zMK>VJ5AXu2Ncn90$j4rtv!VDPX?Lc
z7MRC`K(Z!OQ&#p*@X@-b-{E5PP^q#HXcCxmhp~oSBlCVV^c;-;-K-T>tQ$E@<P#BD
z)ee$jtjkF?kHN$ZyQZ(?iLzr+zqR!hGwd!7X(-IM3}A1_3iYO|Jlq1u{@M0JO^s)Y
zAOvTGu3TYQekG=#T~04i0U{PLgz<~M)&5p0`-cwu<ef?ex+eyc1y!nw;S?Krqb`!S
zaRi<D2L?O#5J=^Mnb?xiH+B%TNRC#@XqNwTifUm>j?1NYJpO9!y?1NA&XcEk!2+pz
z2+Ny{QiP}B(8{Wsb+*8D>}-Q9cU1<9QDk#9-EPnz-LPukAY|Q6r>BZ>4O23aDHQxr
ztbbQKA@$qznw;jtR}_kN;iAW(v0)alH4B-Fkl>mWKbdsg;?iXtDumixbpFNHBEv&m
z5MK%56G^wPbJ%6L^Uu&CsqZC3&sy$mlS<hZH(q+yH@Ao}s0=-#s2y@P_81S5Ik$S+
z6$~2Fn}O)e&4AKT5a<a@ho>rLvwm1D_%Yr5rNDGyS6ANDow?p_UlOz$m(>fzJua3e
z4J(I|^rusVy54F#e|Zd-yMbHK39xHQ|K?~)W?Q&Xj*yrBrZP-#qp+f~?&;FIH|f6|
z(y^+4Z7V;Ul$&VBiD5UX44wvou?J|{gG_76w*g4o2Y<IGWaBeOu|lpCv)zhwCEwI@
z>6In%b!2=^M@-HaVu#TI+XPEKVgBQUtwCjxRiz>^DQ}-jQmg1F%N+pyZmjf?x9mnS
z_vk><$1go-Kmd`q5+UP0Oyj|MV@#4S{m<Y`^@vPk4)1IGX|eGXI&q24(o>6CQJyWr
zykgEj`?QzPMn282H?4BODM3el0XX(^ivKxN&)}=-O%>kFqN^*vX(OMQ7U-Y*;O+1k
zG6v15d%3U9LM4dXq@B{nG>Nl@)Hi3<^IfC$EA|{N`51QHCUP<dSC=?Epq0H7ki}&N
z7qOSSQMv6)7Ct_C8o4w%H@VGa8+%}NSyr=gK|S6@syQ6id?jsWLTDt<%R(!Ut<%nX
zY`i4A@i^Ah1-^}Z4OE+CR|o+W71!w~(R5HE25Cz%MRMELBLa;{V7y&4Uu-<SCe^Qa
zda8DM*ND3azhq~7lUpWa<}q1M3h|ZOtbLJOR{t<hblLQ?;95;U>1vVJbZz$9^Aw*X
zuJ`sFa%%o?_RyXKw8?Ltg=U_$r6=X<`k+;?$xCzT?mv1PFM})s@desRUEw5PP5D>X
zP!&N*%?q-};?t`qkt>HP_F4ure--MlzB!7yC-j|VU(xC1xfSgb--H!?rj4P5L(&vW
zOB{}PVMwDI!dpD8>Wc$`sVUY-1rl9n^&F?{xlDPxCrnK2l)=;tU>wf<RTZZ?o0WFG
zGlHkP=ozz(3i7Yhbm79Gsb#|Ye|`N#E=trQ9N1{<=dgpXdQheN*}0^|7dA)j>{YIA
zC{Ma$1-aZsID=2xr@$m(<wbtk;~nzOwzEz5*O6yr!tZzD!q0z{3H$3V*hI!B7!LT3
z(Y2iwLzycgKneOJ6tc=pRZ`)nPoA7N?yL;tZHr}M*2X|!j-mqV(f2_Z4DlnnO4j=o
z?ZI(b>wUz7MF&$>81Qh($tbQ@VL3v5Hch>j(d!nC2zt~&N8#fP%-XZho@2);@dxkc
zNPKw|9Ud;NXn<_bwnKvgK4AFsi)R=n<+(|6=X9LGA1;<Oz6dyuJ7wAbFz`e08h}hN
zh5>oz0;OD3Mmp9(0x~`a6d7I_<}x5#b<k?(|H|8+Y2ptr@_H>KJiHCPygO~t@9Q_=
z8y3;~pIXdtGDD%8^y<M`-o>Yre)%t_q(54(B#+N+b}bT&3YJK1JRi!&w<y{}Kk1q0
zhy&HEJ{#~HwH(%3p$1dNdNC>bR0Tt?ZiA+#kc^)iZs9;OBFRqw{qs<V!DQ?p_euS!
z+WX{zv-A82&%_-@L4))Zzd5rbH8Ku5IAcgl#|K`M@-$#X1PA8eJrbHDVz<ZW6;AX3
zglx7T>Ge<wu**Pr9`cNn+qCN)rdZ5FEX;q98>NnqKgf|Prt0H&{i3T(aoY|JPucdi
ziyyI`e9z5usJu&seQELwW75Mvqx|W;xa;L`;Pb-$eTnt>7D0AqBm0-Uq=WZ;(lTLv
zTaJ6UOD+s9u9z{->?+yJfyT?-EQE89({}=i7`eGkk6#Xwu*A!dseBZvkY-(vgsz&M
z@39%`C%ps3nQf1eopYpM8b78nUo)wbO@C1G^#C|&GXX*Wfo{H1v&$2M=HYEIr?Bh7
zgPJ2nMJ$}#g`OWw$^>)9Z_!;C@=cjh|B_zXq0!tCWvmP`Acw%EKS3W9&&D?RGhpN;
zulR?i{OW~BjjdfO1C+1VY3xQT7<doj@o>0~yoM=rWxu^|@dUoCkS;E6`OCq3u6Zeb
z%O?-6#-fsm!KeZeqHHGua=QB^r1y%h6JUY7K-<-$tx6EH_K$}MpAymly(cXg7t6vZ
z30lA`%yaPumZ+mm2$JFWfE|U8^}QoL+EP||qpkN>WDA;$dfRVe<0EZ7-E=+td-c2h
z;6F77dHOa3*a$m&bzjJ>FetATcL8mGW>Cg=-nCmQooAr8sNZ1jb9Cm@d^WzxrIoNN
zngXxmcYdhW4KE-Ne%vQX?c<BkbBydKRwl}8e!KRr(9b$*Te%3dUt*&HTqJO8t@S1T
zY2(GT3i?M99hRVh(Z5tl@>6{+v9KHaXQOo2e)jYp({f&++bTqu=;dPjGgv%XC(L-&
z&1*FPP4TrjGJ)vURhEvI0R}aKN95q{<dNv>J{zCQbnjO0NABC-W`?CLDX*FW8{}oj
zuVEIs2V;L{s?PmFS;Jr&MylF?6TQAv`jOEFIFS>wf4jt<czng_q};>uk0fHi3jD!i
zEc#16C3ZISR~$~Ro+X4gi<;H(Q9{MoJqOkBBF~VUISTHJ0gpphC2c5}k%)ZOXPDj@
zM8|cW2Q{~*hDG-H8b;H1iM|9|PKljqr-7XK_N>(~CGB;~zt?##npQ_=Zr}Le`yhTQ
zpV~sRau9KD6LK{EmWaLJ&~@XyWDU0_;{bmbRy@f%tm>vg^-M$xHYMl-({TUDAyWT|
z`@UEUXx$8eBxUAMdpx@-AG&Kq3(aRK(67tb0a{regM5q%0dt;7eoUeNW=^(#NnKu@
zTN$4DXRj+iyUonAvlUfMyU?q&cL#ylsUQt(#*>oePY}o3d=1u5X7DW0j*BA+Jrw^a
z0Pnb9yhdDlbK6sbobt#vVqo2%Y`S?Siy_1Bv>xgaf}uymGORAjwwOe|IwYFxzr0eo
z1VjeBo+K{2t4z9#+@^x*Cozyp9tH>02GKDBqSwlinu*FfW>qR|Y(b<~DdG(6^!JGn
z*;CI;HguC7G~sZ&e`B`fJbnl&d}EzEEOHH9SlD?gQYU2MmBL!Mak}1CnH=d**nC_^
zAQp(y*Qf!T?j|6qP0)nHgx*F-pU%S@yV%}m%g*T%2K>zS9pN^o6dU@NXRk>_xjKH>
zUUc}@d!|Z{*bhr^(RKmZ&h!<BKAATT4dl=03!r9enOEU3SWYKAm9iFP&6|$Qv522H
zdHf%FN;{r4*@(#>rY;j?|D*jU*30kRO#(huzcU_B4#lckTb&pf^s6)CA`pt&nG7C0
zgZQ)$QTI~3h;s$vQ=)U*Z?Zh;!&J9O_2t*FO~v%!tL5@6ly7CT5&-eLWz-#o_AOmx
zdO7Rl^z}K$Vm-9S3rmQ*@FE8WiQ0&V2E%*wvZ5ab;*mXxLE*XWrtb?-A=VV4Y?{$L
zP{=?2a}D^<>)mjeiiI&Ts|T#VtfH+hd6#T`nTB}zJOGRMF>|?Ej~qe^P~@+0nr)$2
zat}C=WQR6uYJvgZg<9GE%?S*O?^{<99`_`aef>H1hU``M<&{+g<+c8cuLag)bKOI+
z%_a3XbVk~Y?;9P@ntIb{MD5QPWRXBQq*!GEy2!yP&Go|$6lgp>5y^Fvg4P>tN9y?X
z4J&jy&fh7|y?jt2y8IydbvCNJ+9;!J*SGD@aMhd9H&riPsSunB8ZA>ms|#|9)}6O%
zcH`{+N4P!^8+z*cFLbLTOW^%%pV5#6gc<Pb`9K=xNlYUen}W&mnc^AEU5!H7emD(!
zXnVliLU^O3Ve=!PD0BUFNqFTol_qs?G~qz#lJ=jpT%!YDl4h;aD__5qu#Nz1`HKEJ
z3ui`!oEJzJm9kDdB-<lfkZsbs!J5^&K>ZaT&ls=vv}Y`)WfACy8>r)&qpg<l%9SNS
zfp-PQzvASgYZE#x{=azecK-<aClLN+5!S{_A28arH~*nkP)NW1B2DIBY=}-uN(f=h
zlG5Mu1XvMmTyF1nRCObOYS^d0x#gl#vixSD?@r&u)_VCMI>0gif;Tco_!5J?mf^V_
z*sjlD^O5oGDCNt0D~#g;;!~<ozzXxn&;!zY#rVn3P3T1Xpt>5~WQ3RD(9^cwmtd6g
ze`iWH!^4syz>0!bMbW4mSRrvI3p~{`Y6nHt7)|^~<B*++<|=FN^$!w-W(x;YNx}Ms
zMDH*8;&U4@?2_}>d~n_@Ll0AtMYcc76|=`E?$1<iJFK$G^2fm9?-&Z1TMYkv4?qqL
z@)grUholD}pG*A_1!ZF6CK>~X3ad$aGYgxdG8l4kv$B0_h(CD<1k%l#e4uj65>&Rd
zuYTasVl=trMbt%)2-Xn+kw*Hs*Ey?WQF_1Z6KZ#^e?ca0T0d^3<5z7U0yA6sL+6HU
zN|?&k*bNTH)~_a7guGbPj2SXYHbP^ga2baO@9Bs>vY5aYAj6DdcV0Dg8r5`g#1sG1
z@M_D9amb9Pjh9f_y-yk0E`1dCcBSix>t^N#ysAW>>OMi9oU9D8h${Q9GS1PvS7;UF
zr{>C5e{<+likI2=Ed4J6rVt~ZHIAQ<2`%wS`Mfo1q#Xk{E9hGrVM-A{j*N{Uq>Uj6
zU9xi*4@=>SIN@E_@#thvum-14Ir450J~6zGWX;ZlS1r<*D4V$b(gE{5J*MY5qA{9<
z>mtn+Kn$@Nl(td8tFW$wesztg<K42jnGo$Jc#G37Rw8Ho=RwpMGq9PrFNXyF>&!3Z
zZs#YJli;tJ<+1${S|F6v|GZ>D*YjZ$4l&6~53?%W`t4*UHYZA5Ma>=@R{+dNvP(A1
zPdA_uI^Lh;4qS%QL*OHV4p8_Q!!C=4ct^gT`xkwDz8~wm6yoxYTbgMuU_9u*)lymX
zOPhu!)bus<x#+@Y%;iyBu%l6_Z0erWfnl;=LKJUVJs#Z&qoV;!0W62#n`V~xfj)Q`
z)e(S??0mf|uaKryJ0g14wk+)lZ+Nb=V_#XsQW1>t&<*BLlkZ!5p0(hn3XHX1Uq%RX
zS>w}~3{JeSNPg=4^sjcC$c>3Gbei(0#35ZI7_t${8%J=cMbO_2l$DDY7Jr*Q6(qWu
zg-z=5zp-!%8$mpW40nB9_9Fhn<H>H%Y!Fj=Z7;HC+Bh)kcr&kaa7+&+&E~RKL#V88
z@1Jp^G@FvyNhl=5nfF5rSh4y<D8Y?>j?w@U?wgk343UpPe4>ZvS3*vjzP}j*Vlo5~
zp^vw?<eL?hgC60Aa9^08NT6=CNBZT_&@@`!fHwnA4*=e*vFj}dW3G4sONn&7LLMD!
zE+lsPz`C^pbyQ3Mefl*`X_6=bQXKsCd>3q&lG9tTgf(vlDJdoMw}7#FM8&rQJ;3Pm
zCpFSl<!3~iJ~2-syAv;xV_nx&tjuu!6%S5O&1({Z*myhhMsskAv;Fj`;ySk{Q&NXi
zZz6Y${2!m9<KoInac9sR6IN`ONVB(!K1~sRAJ&eJ72mO_TSUfkpjY19#P-NVqW=1d
z0FnNCD<bsdv^ekolJE}u#TLq#%qYf_2rR%2P)wPCtS{SYN+v6H_jPaA<yv$t&;&Bl
zZQpfFEe7I`mk^f)o&4nlDep-p%fBV>x^1TDhp#VBwGB<qXvhBI{Hq-|+O3kEZeT!5
z1=neU9DHK!?`DAFyhfqB()#uXr9pEojmqay1e)oM`wwoS*4fOAUE?M$zzfV11h?F2
zWos-73^7qHsTWXc&C(bW%9a+6sm6dQ4~jybp|~i%tQhUzst`SDr>hs0>=rc<fpC#d
zoAIQa8zn`l;wA<f$ERDUSA7!Ma${PwiWN&@n3ah`Pnt~(ViwbJNTX!p-6E6tlFlf-
zW5IMd?(=`)gPCtwkl}qzH8py?dviXZiCNdG-J}I?hIea67MGTk6&*w}mY;EPo7@DX
zvoiXJG(kA~W8cJcpzaL&btYw@_X@k6U@eKSI&E~t1a3HhUFseE>_4gW6;IrYU#H0X
z?B(kf+tAr2NI1ecH#5&i5VzOX9&dM@KmJcd%Y4+)(7J==K9G~Wdb$rg24w2=Cs~_r
zD)P-M=tRc|Q33%m4x4xCn~L>VwV$d&d99NK$G~i|(wzF&vB+qdVO5Gr1t~fZA15xG
zU^PhzjzPDtXERMn0(lRltGg)j-@vbdPh8{@TDSr_G(xH{gxAERhsJBtEyj9Wk6hTn
zBsKGCcec=|MV?(opQb5*Hz0?cZ+DwEGB&(~7W}ZI2T%K7ynUG=6U-As**OINJCV2&
zooE(9l(P=v9}gk}!n2U2j7G>r*1ogzU&&_pO+iF??udK^8|s@Oxe@yqwMu#hf9-M;
zd7rptOQB-7r0<wDDPe)#RFO9UD-PX67w_=F^}I&lZ%-CzCPb!OW?VvVI;~jXzqu7l
z$MKK^_!<&&9^!cHH4l${f1Y`=`=B5EBZIraS;>ZFINIy$^s5tEX)tZVmiuhyr8k2L
zV8${|TkKRjSksqAO-&T+IO3BUCx((w4eRiEJ|?St@4^A~Rge<gz3*QvF8nFqTu1Jw
z8XPW?I$qq8Q&x$HYkNO;OloM&@F=WTX4mmeP>Yk0N&$Q2)BMLxgZbw3h|J<EIt(Ju
z0`W&p^SK27bdKK`X)SeEHnk1Q*YiG~>uED@CT%t1TPAIc4tlbQYvgMCjMT>x4Zy!F
z@YgwMZu02tR)rqCv?%5ke%8Q20|_U$$n~T=pXK`CQl%g-j>|!qM8#C0uC}c%Rp*o{
z_C~VA<q_aZ0*>cf(D~?m4<&Lk`)6(`Oo)*16&m%wqzmn^4qJOzYNTnPN<3MRM$>Rr
z!Fj|L9k?{fdf~19OK5s_jy5VvRnklZKFSP6tKuQ@p~pv%`O*ZLuQDBVOOSpp?6E7Y
znTG-S0AvJdu!XJ804)HcFql0F3rm?A8F@jQ$pc35ZrS5AgM8_Q4=Fo4wQ_K9N1F<D
zQ#!S`6jhiNHhp7tU)u-EYdm103={B~f)Fk}A?QihUg&$_vqZLIe;)Uu^MDjth;?lH
zsexm~sU@Q|%J_yp1g#qs-N!(gDJ{AKVp!5t-x@lgt82EL<E$L#oynt+q#m)S#wo?E
z8PfA9f1hE7rH1Ojo&!N&3$%je&oLZ*EmJXbUe|0jKwD<KbOZ4O5Q}7_!i#Uc&o<d|
z(yjY_U{Hj1F$sU5gIL@(S<g457%lj-piq=Pv6tpZNf%Vg+906O?<OebdMC2qyg0e}
zENgMk5ZH)Z#lNdeo;+GItFgYBgZZ`%!k<Mgrn>j005!T9>W(l1+5(+hUiCR;`@5L@
zq8%_Z%_Ph%102I|HPdJv;d~1artL(%>|DWHt*DMs&D@-9yl*~v;|vF4q0YN$+$1^8
z_jI%SnUkVsYGRyNAC1>{Z;#=$93UDZ<L^vX*E&~%d_?lVg!Ot1p^?6g(UVRvbM66x
zCijlpLgp9SVWoY23%b1*S?3X!lS@EBp;|iLP#%2O)}W-~jrR{rp?vRk!}`A^EqSiI
z$T4vx)Aj297kh6RRb{_*4`UY!2nv!Cf;0$7vsFN(OS(isK)Rc)qJ$twHz+ABE#2MS
zUDDmL|7)wybIy6rIbYt-FXJA=F}!8(eO=f3tu@zNbImOT`qeWa+a~*!zvA*`{O9cD
z`T4!H&=|@Eia6>WhySE*sy*oMBX>!gww;ihQgR=%h6tTJ8$z9P4S!E!)_EgR>|WiA
z0qD=P0;Q4;=p}bx)t3|*g`U)3EBlGK0yFO4`R4Q2ab0o>3sH_+$uMDaADB&@nyuQp
zBG)#!LAJ!>kefa<*j9wF8}_AdIASe#Fb-vUqvlWTU?is<ZJ4ce^FyzpClEhK(-d+i
zK~EX2qicJ^jb4F8gkGR=adL;@vfJSQF_jSYcU7A~+}oR2%xbh(p)M@h{-V?FBOgw5
zE_br?RX}~tBCCzjX4+s)<`v5;ynBN`RdR4)O=+`PaQ*$CjrX72|4?vxW2uJMWZu{*
z%R;M$yUZ%T+T@bRgcI3-a|%(pwVFTZ(}NmXB{xOITf9|-ghr2>IFR#OHHcQ(e@sv)
zNv7&gBh4>@w&E#|i|L?^ka=-BxK3!@WWP7P${izVfs9d+LXa}!&(l1sJ)rzdS-L!5
zNq||ys5N{0VQ4x?9H_JC`v(FKRYOduLjRb^RA{x6i@d?e9V68`c-g;a@HJ*&uq@L;
zTuDN-{5s(Z*@MW)ZOWLnlZIvP4egSyMZ2b97ovZXe!fYyxr**KBJxS^w6*+Lq%7#r
zi3;U#V7!V`=+Q9CZTX;c{zy8Udw8&Z#{c+ya&CuvM8ON?e^_0h`=BD)p>QhOTwFHe
za8ZMw>$&p@<24|RJJ++@C++i;wB1=z0hG#UTa?s(mqf%B`pb6mXD}qr5LaKrC}5DD
zqXpGR0SJ<rKhYpGW3LxI!_OJ6GzE+h5hu?<afwsdnQ+EU_2u@(QXanJo@eFPVGf?Y
z4I(jR$*SXz|LFEUn^>#<F#D^aiG_-fQJ8JT%i2<3#DBz=uq@I4OEI^RtUJ+R>h=*)
zO->R62$~-!Bz947_Te`A<6oGII}yFjMD5}__zgBgK4YmIMJ`%!9|(KV?C_Xj{*?_*
z2_132JojGGe`h+WcDY44A%XRg<#lhT@321m@BR>bfqTRHZ=6&l5TxYM`E9_KM9!@|
zB!3y|r?mtYIoGoG64y@?b!=8kM71Uh{bt~^{U>jkT?fptJ8TMnq@&$|do&A2r`0u#
z8XLiO4^MFZ8HuKvlN?;X;yC5+hANLahMjDJ6ZWH?`mmInZ2!?rg6m$psg>aX=8jl9
zVPOfEERXG+?7}g3DeT3~$1HIq5C##rVwqPqKlLfr_)Bn%m(AxF7yKs|o7yE8(wJ=;
zPqA60kEFNrSE9G4eGrl$rK!_rzm7WXUvud4G4A~h?te~yyLcYMggg8fc~|b8p)X5b
zG~1v=UuU7eHF$!jw?z)6OH#Z2^m$?GGz%wkGG6-89jN);7el$rG>VLiQ2}cCoA-$b
zT{@Orh#17-WeO9!zUJ}-gZx&j4=is$J$?h$W~3HAINUr(l*-<7*rZ!~CHj9mA?KaL
zdZK}6nKY}JM*$MDFOK37jendpR1lv+#ILsYV+;J1xPq<6a7+IwcV4j<M9tZmqHuHy
zwU<#Y-?+@{*cQJxOkj8@B1EO2)I+f9_^YMYh3K~|bhLkOc>n!Ur}{aVMqjvZ>_c|>
zvo%&qx{r?bI6v*n$I%)E2F;AN^!GXiKP6sh0GWoEIZSP6y5ntZ37-Xf5Uefo{d_y&
zcS@DDmtc0LQhqE(h|ksybRmvGZr9MKWmhSj5HsA-?)WhnvOza8al0m4C|oQh=c9Fb
z#4^q7pbLYLklU3`UVnDc+&`ul7IHCF%H`LSv)Oys8}{0MT)V9n|7ps#=_B((f%)}G
z@ukwLPkK^^jrb{8jf|f+MoK<<zFE5!5gNAofY_sa!PKj(_p!IHucoo7qZk;ZFyER9
z1IExV>qY3sM_=X@X$om)r2ju$J2cWI2~GdjsJ9U<zi4HrAM}~VmGynK$tU@;WN1tO
z8Q<tQIC-gn{DT`VPT@U~EmncodY$u!#@+=;kXyIi-n6m_<wDATf^7}lhsJjbqgR>s
zo<ow-1nE9SE5{3)@)K2XLD~;gS4r$wKR3PhLf}x9U`d)nXbrv7<NK`iMDgY}%vOXL
z8Yr#xI~bylxislu<yqK9D?PKn_>?Ms^5pvOYR^gZ)b#EY7sX?O!L@et{?ibZ;M{MI
z0djl4u@ofMk}q_Js8b=gI>C|bPMoo#CBN(8(XjA4OJ0(Ok^lek`mt)&`|yLok0Jm(
znw}RA!S;Y#RCZ2nlZdf+m(E}!SA+(vGDh5G^GU=0?(;f8UBg9xe|CO)CS?=Ir>=$s
z2if255*HXPPppwQV?KH%&4mB6cZl7pm4BciTe5N6l^q&wr^LzYH!uoNdt7d?NyY)8
zUgK_T*L8B(?AustlQSa2ndL%M=3-J=(%YEn#g$et5&<2UE6KtQ0bN{6w5Q8Z??LQy
zqk@Lo1NPO!FIh`qLIF*Nw#EnX0^9P>J3Hb%c!S(wJ4{-R<*>_^+JB{KeVy$;EGE4V
z_G5Bz4`91MeYN<NvHW0+5aa-x)3dID0S10;G2($&+v@KdOHqo(+N!04BNVK0V2w}c
zIG>PGE3`(P`XZ`{NdC}r)PF2>?nX;}4rmg`f{}B;kYyvwX5(CKP1Jv|21%n`DqtPj
zXa2omylIIg8n!^4SK9m4o^OnERPBey&x#ktF4F2(3tn13C~)cx+{MDNWCN*V?fu!;
zbsmE%o)*HM!Uom^&tywh=PUoS{}D%Kp$$!zS1>_B;vvDo$e<86q8_dC2VBjZuiDu_
z=VK0T1S<X2nQAd;vA!@9efH9;wQ3^bzf|L5$6Z!|Zsh}~{LJS{&N5Iky(44jzn3aK
zMU2)Bn+lTG>G7D;KN%_FH#a^`;gEiFa-s+-_~gQ9KkO<xA8M_<Le|OvKSJnr!h)nt
zm+s7;4>?a_eht6SnzcA`<WSQ6<1dncF>=lSX17%Oq@pEU@HnOS28`9|%Bh!Gimg6^
zN8IxBkIBc8kZ=BHuE~Xj&hb16w7TFCl@k;;eH4GXYMR^0=o0zhH?$jJ{j>31z^*dG
zrxvT<>><n4Ly-44WDi$!8guGjr?>JvlR;>ArH>dJu<KkqxS0$4QgTodJ3fS5yH2Tn
zk|?@_oq6NBKMlgy>YMg$TEzTUa10V=;xp~Icvz}M9tz%Y3Wj_m{%`TOkJhC9#IG(3
zT=A5*z{=RAF@JbOr=WWURueZYPA<SwfPftRs^s0C1+=Z##}&sYP{zA|Y)B6v1%D`X
zBT8-9rTM?QFv_TQsXw}~xXH!NjOoZ++-mHlrKHV>I7i?IbU<=TE9U$eg#6loFaBw{
zpL6MJV1zEP-Zc5Mdbxsp1#cv?VKD;_6kaBF>z$tN6{}J&?Z_(v92gYQ*a!?hpc@k@
zHz*moiGNe<Usa@}E`uqw*YYrB-x{_11bA&?c|!Om<cvOw{WTiQMe!2<02g$(83iK&
zRTnd#p*3^t6hXqU@C0nxTeXPL7{O($xUajK_D|_(x$Z}EFQnR7|CnAZ*a*8@brYXJ
z=*hMeO~DK8*wxibMsiQX#gTjGU=q6ch>W!^>>@gfqv0-GZLbH}%+s6vWBB~-JLo;x
z$~7NJ5HHyElAi|njQ?p&qz^sqU@sihrFI3nMhx4nY7>1cBmK?pA5M#zUeB9L1}81c
zyzMl0zS!PrdkpQ2<{vG^M$L^2p-UnlQ(_vG^7_1`<^R1g+j9Ha>pvGUTsyEJC#K{M
z3V1#$-T1o3+1fLUwI*vySlj=>;$<~CoY)bjieLcti~IDk_fTi>|4K`HmlQ7AR8lAu
zG<Q*^jaz+);-6Qs?yznpnc~-zy18)PYtgFYel*`|CVRiH&GB3vO5MXF3O-mcfu|PH
zX_`s<D2cakyX@y*o6pm+e<ECB_B+gS>^7%CQ_KE1BYC~|;w6>Lf{g<SN*CkMi1DJ)
zCQ~1&zit4a<*y>b`*Yg$0ywhp{?*JzvUm6He<7rmNbI6vB4$Ev<f<@097QE$T+pa1
ztTjM>YDrR_(_Tm0HiywVyD2cx8Y&9pDwE5RW(1Rk_;02=@%z}^8<~|=;wvdhQW9L>
zefsQwerUBFv{G9Q-~4(N<1H*fnc?n_^o9GM^^#M08a{K<M`*LkH`SMBX>ko7ku;7(
z!j`p9@{nqrk7Q(OlZ2S=AUP{NrIbn9wT?pv8X2;eR3^F};sDR<OkpT+&K1#3t_ele
z1jYg&>FW0AT<}o)H(_bW4~I8Cy(i0xrB$vk`)IZD(u3~lzE|7HH?hnvv&2PfNvRXb
zVzSd6{<Pej-?FA&f*DSvoSa!hfPGWE=5n~Q`38unnYFB|KY7(|4c16-ss`3wV|GX*
zJD}%K0O7savD&w1lwBUMIOyR!LTBErD1hE)`!G)ItYTRl(g0g*UV2h>O}oo)uaI1C
z`A<$;^0&tR&5Z}KQGNCGQ3}vlq%MLw*|5kYoE%Y94QLCULQE`_e3jBOHxw2PacgqG
z8AT&q1vXSpbRq{fQmT@Cl%lGPh0^|QQWa(AzW+HQYG!&X2D2S;KP05432j_3<GgWH
zRf}$CKJc3xjC9><H!vxXzkE47d;Or%yj8_Bc16NlPRY1O9T$<I;xQQ1<PJoDV0E2F
z3oLn{AMT~Xp%-nl4&oAsGFdvlHOo=LmeyTR@VECo43)oh_|&I5@2~OB=GIf58b>v<
zD)XAdR!Y!~fcux%ipGEU(vc_Rnw>hUm7G1p!=CmW%%W_n2^IfpGN+ywL=cpo@>)d5
z2T3J4%L~Mi(4)8B+F5X>Yni80#JoGdZyHuP?*H)6ab7N^wNS~@AafkYlm5BoJnME1
z#w#g$@godtG$TsVgPUMU;p;&KH2&`qEVy725j89qQoi$UfYpR%MQpPC&Ni?=!U;Z4
z4QiDgY`;$mSbvg-xoi;+m^)P5jgl;QR=ewJp<5`NkxIRG<PKXdYL6i2+|af)A_Vgt
z%c1r4>w^}+eo~34pp~?AaSvH0MuX!v&a;$8!y4h!{CMHtBW(5^HK0KM$vVhNBR_s2
ze`Rr2@xw7>x;@BrBp}tdoBqT1N|;T#WO)%xV<HpsX0hb@47}<OabO<_m!aqemC5WP
zk(?Fd?Ri{$LX+p>*Iv~u=XiqbLD(c^FXPuC;S>CMQwmsrD25UCRia$>uxpp2hhmSH
zC14RVw;@k~`BReINl;Pvl6+vfRc+3pDPLEm6enpf`^p2}F!CFJT7b3(4lTxCsU04Q
z*ZZ3If7(LL6_N51-3h93GtH$9N?;nr-yRPA?cKG7f7RmE2^*|7cfQh#VxVoRQ4|7C
z8&*)!&q%y$U)WtP->AO{<30V1ccQ$(ptnN@eMWIR$5``kKR(=ufwoA{=wpZccoidA
zFq=U%Y`12m8mX>|2O{^)Rfjr4HaFp)AJd1^j{A(|$3kU^ja%-Iy{+^9<0Z>tz!#R}
z5hgGi+oNSCD}jJy%s~Jf0DC8bb1nDC-$a1@>j^`F!wQaz<UrtUHpi~hCm`g`q1man
zD=eXUoWW4Kdt>9bo)AB`=N{<UyN!LKH+{y@D#8I0Lp<1{gqR43UuL6DNH~xe-kQGV
zR<l*Zq#VC!v$})>qpYg7=&dxKU;b8pvUVGP_GcwZv!-7EUt2}+R}p2lAFMI?w<%ct
zbIGZCMIpX=x#M21QD!mS+S>`Sq3j#tJT=Ia>4MdaC@C91WGvvO0ij_7@afbQi_ZD4
zCPXKqbFDNk8Xp7)NJ5(2Kahgu%M--k_>OSz>J|0Ad)}_gr?xR@);<1ZvFOrLRK-RH
z@dml!vwt<0WK$Rpt>dX_-d*M@ypICU(s<E@H=OFmJY2Vfo&MnlH~z;B?wWkhh8irO
zfaO+P*BDP%CEt9SaK?Q${c1tn?^~f6o8*R$`+1X%Mb6-^6_j98&5t#8;l`h=vnBi6
zJ+2Nt_O~#eeLUECjK;J2=T63?J(B+J^VDp=H%Yn&FPSHObSe+1Ly5PBH{2<&Y;3$}
z9LXz@!Jv_3lZ(S<_;3TH?EjGbQIeD=E$YxGX9p5QbJDIDi@uyoCMaZS|E^g!<6Qj=
zcA@|J9TmDrFOuymMnJZGog`U*jp)y5n4|V2eR(13ahZ&<?8+ddP%G~)Wz!zFi=|^X
z!6QTR_!G9f$x-xkCiY{N0*jqM$$b&&z@f*2g0E>g4U5%Pf2vqwl_w69@Fu)_Mxqj=
z{ZF$k=KkK$UZ4C<Qx@h$J}V%V7_63HNA{rMJ2)cIfg{p{4*g$f=JP*0W}IwKTiV0|
z>6;nAg~1j!*NRPY%3M;gOXJ+fe{!qfmr@EF04+prNMk#dd0$9~7^n5FXwP$zZuY64
z^g=VyDU8~2la!B|+CK+~UA=+sZoruAuIzR;Efr4j3s%@}8?pwKrCGBgwPmwlN9CVA
z{E5DK?g%N&`pe-d{4gsg#mjl%)VnGb{QRTG#>-6*X#8J0*%n&*!)P%juPMZ7h1SxF
zJya7@K-(~QT6+?ih{NW113b`}Mwz@mM@>h3r~QOW1M>&;xN9UB*Wj;=!V}7Eqk`^l
zFB<8w4O6n0+YTUyEBuf?Z1mpV@FrQAftYV;)b_;g!mD@mVz{u4uGPv>;ZXXy^~GHQ
zo%D!tr;wm>LlJ-Ml95E4+#*=0DS7k<x|?W$)hDR=ZWpF33ri1)oI=+Iwn27UWDlHq
zu4ogD9&7NX9Ve#CkL|A&`TE9%5s;Auz;1GGs0*LsZ>gHK<KBz^;`wU{R?oGU|Ah7t
z5wFH452srqmhVIT2!C@Ou=6UsVPc45P}3GAJrbTLC7^#`dFV)rR+>~^VlKCSF9NoR
zULgcE-877H!W)lgpX3Rfv-^5`Dp`|aW}Q;s%&+4rw7-VCV~co!%hs*oc5cHC=Witq
zqv{mIY`=7WFwB<Tnxh<BI;j<`0fhd_3JR#c?Du}N(O7n-HF{+B*Rd;`)4)9MAF?pN
zP0O2T31mW6+j>GPR;D)Qe=%#maOw9YV#Lu?<Zy@{C#b7(b)VGHKmHWQEmq1^w9*hU
zt-IqeCGpNgTQ5!CW57!J@^~n$8=53yKXAwndTd~;T2o@_*)<pdK7*fOjKAl?8xD&o
z@5D)p(pi&7hSsaA64QUZ_>&|sv|i8DFto+*?$LOvEU;SGd&(}APC2{sgAww&BvojE
z>fnJ_bPse8HgPmFuhcqbQXXg}6it0~`gVASmPTAAQ+=1m?2olhA&emm(1S8LBGT-|
zS2mLu{;bbqXo;Al&@h=L%O<CP!r3zN1O=cMs2D*8u;v>b#VUWR&vbFJ-&AalaWC;c
zI*DBm{R-=JHS15_+Re>8L--yO2Y@dmOTP7+k^VYx31X&kS#c=Fb~L@ZW-9i2zD@9N
zCdOE&^I+KJ^w5)5iJv-l6_ZR5Hc2B(#@fig52mV9oS`H|qvi%+p=${$yT&8wQ`TGy
zK}J>wS4|R97qQ<Xz~NR4$y=YdpKK)liuQ;q|MJf^IpI{?X4!uP;usQEn`50)Ex88N
zApY)tdAOmRyzcaNXiQ8qo$xyyPP=a!VcNf#FgRvQwGcY&TmV<eh+4Kjuzff{+dlL|
z#>?CaWeot`z(|z_Z#q1je-ao?*Z%cr?)RE^dYDE2?e%Y?#K@Stx8{$)GbgQJIOMPK
zL{%cAA~>eqETHR2?CCFp;S9~OdT(c)SJTHg0dgwf)ysDatqB4R{WLNouJWH9-O(Nw
zE`wsJVl#6;djC+C7}WL3BG$Cb%~T;Fr2fl;=3<tw-a-;Tsd1TOKrhr?DV{bOX;JYE
zkLB9o-E}x2P6)pWc_iY06|hHXnM0PI_Y`(aif`59;A1ZXCQgGG{Jq!k*UAI`{BODf
zL*kb0;T1?kUPGOymtZ|bD><DCjH@%ixO%{T%k#IH$Z%BE#}Cw7hE;V6Ei2r{_sAWJ
zWt*OP$wo4YzT8y_=o*>OwG<BWn1#nCbL*we4#+J<>`?dG>G(tLY3SgY0BwvOv`cVE
zc0Wr?B`3c-5X%#O(+xG>`zYb`Ng#TZ=j@S)_D->!FZ}JDvO|Qm1;bhJPPxo;0&*3!
z8}77(tGen7zvV(tITJ<iu8|kpbeMkVy_hnWnoO1Xj+K#+rc6y+P&xh&KEU{?@^|!W
zN5`mW0%U`_CX9ZIOIFFA@JR!aRAb6Ayc2vRr(AB&tS{(FYEl_1ecI}#DmQn4J3Ri%
z@0tZx(|nt1eD)L9hg`o-Y*>K7=}+G?t8CWlfDtUX)(l<868J>)$2As%lL-9V^bav9
zR1UwXtL}f8pT<&b&=u}x?|&Qe?h*OxL-=hs$Db7w_*QUnklu^i3vNC==ZigYu&<vs
zqK^8z<KXM`*-mA?GE+kuLd#}!^(P+F-Cq}Am)bQT{SV}vpnm9Z`1~%;!LwWU=mWRr
zZ-@<sv3|;_<?@R&RT(d=Nxy9V%T`5hur^W4z<*wVbtTUIkCEGhEJiJtt<gj0y=@D_
zz7t;j(vzRm&02-CV8)NevJ=!+dH0WVVS%7QJTSGHQ7pB3LPK_6nX+1#4mQO6VL&Wz
z{*XwYLU;}Q?y#<;cObU)Y0H_W7~(jI@5MHM-}+-V^Hwc%WK67Odts|Lzt5_DBhD*f
znXPOw!Ao-wi`>nFeRFjUEYo452_!1w1;*01@v*Hu)QUBwg%TKG-uOFSMh>4ky8jSj
zS29Qgm%AX6o9)WeGvO&1>3g4ShqLQYR34o~9%?vXG(4i)3oEv04`8%cRpVj>Ab<3%
z+6<RKN{T|j$yFsc%3i`cpO5y|peT_APS(f{s=mjM2g_x_V(Q70^d~|yIHNxNph#@}
zJOOh=^zYhUR>1RAekfLIwLuGhbeEd$9mrZ3-vZBm+n0S>M5%XFN<Mt}96<uZ>Lhe?
zopk*=dNO8y4&x8blA;2fWk>#~?b&B}v8%rYaXVeY=V&!A`gKmgVq?Np96x?1A)N=m
z7xxppGRBENpMfh!GgrK$%AY;Z()Z^<_v~Pet9Ms12Uf>_ah|hG#}e{xXNTj?bKcU5
zuAJ%k#}8d1o=|vK^bvDn-SdpA`4K4^6i>APYakhPoRF)uG`?q~)WG2M0<-P4;+Mmq
zd7$Aa7=ANQOm+@(4*F`>d;hbHj$QV6`Ax$w8+s;|P~dxtU0_awmQC4YXu)?lDDS|<
zq7{HYU!#3q5;0sucm0IvvdIEr^FA=~B83VCh1<XdFApvJ$(ax)q*MXTL&*T0OH%n~
z|NLGkb}|f4#{Q~BU5M$mzTbF(&uFt`vzl|2sDM~@gp@w1AN@@Q7a>65Cj+ZJEC8H;
zW8F^ok4zn#KCLRTx8C(6#RLP5|NrOz%N-brC%`--A!KRE)*8jzFQcwbtfsEMT+K4=
zC~6S?^mUH?fYHMwjS{=gQnSe?m~<7^{t*%4Y_D5b*f}_AA98RUc<+@a_b(Y7ZZB|$
zC+6|(4!k1h7_OL@xSPd5G|<&`Q$a&R60^F)?K)9|&VyNO*=9RL^<>U-VRAQJ*CAE4
zH~|3{^&LCD7Q4FV3JR+EMt6+I%lJiP1B1CO=U&K6=GVNs{DPvWxY(dCO(n<c(<h&j
z)5E1{DXj2BzseWCMzVFqg-lINX`bRu9MUo}Ce(W2w+mR#wfMEPw3wG8oSmE+`uq1}
zCA(Ew!V5B%myL#gn#Zvs5XfO9Y7dF5f0K7nDn@`mrMx_b!0lUMSy|1sCv7K}ZsfA}
z_0wF6km&1Le4|?K!hf;7nOIIiVaIl@>}WN3Y)q3-8pBUF%w}y&@^#LEtV-}k*}3!B
zZ<!ykusr${E2E}HbnDhF`yB!j5)vUHc0NAuGZ%jP3=I#9CUwgS3c4z-wHtINc8(NU
zigvSO&M~HBXZvqExGuQv$@!nI4;ixFyO&T`Rwm%Dd%%X2O6nE{_1;tCq1*+g<}&AY
z5~nFocFCJ0-e0Tl`kZSv->x`^#kIRSI{aKyvuw62L44Z9dt%~+ziyb2Y|Y!VXD?Ua
zYZJI-^!xJ%pPtFs((%ys)K$qGSHrBvt~bAYx8Qr6h;LU*7~2VIzaCMxdAO9RMo36l
zw%z_I_L;G<anIP8Ns-fnoPSJA%ud9{hHrmwZ(h1`)>&82HMZXg0At?*CeAml67gO3
zij`37?ucbeF5>>(J@(ISx_SY_(^>ljn5cPm8ykwJxEL9nw_%MO!;I)7mSmhp@m|N#
zT0%`}#O5l8yRS4o9bN2r`L>+;$mxk%x+49>TfZw4#)Oxyu5PE6RAi`<%=6=Rd;9rP
z2TTQp<gZ^Vtd2n`x4SZ251&i>#hHnsN^oAqcsty%)sB!J&Nq&5hJP@K)G!ht#b?%x
zUtL{Yd;R+L>0#B!Yg`G~e_uZ5>5<I&NCfh<R_k~Qe~Wrk8M9hOQSo+OZtj=n*{X@N
z=nSilfTxN_Ow4z5vNw*Jy+r235}_TsLF!m@vvW%*&~p2iEav@+>vL8-dJ;N2uOvi7
zE+K8zM<TIMoA_yOe<x{mL<qB1B7`Z?8$-i2g#G5s6K3Bpd{2eA2_)YVxzBPH?TtB-
z@Y=2#QN%Lh+A~OYH`Leb)m3k73=i}2q^PN>u~z)<RQlax@5dVudd5Bz6ci-DeycVR
zIhaGs!t&NGx|}rTx)SNhsx3lDo)6<^wAt0$81OwkI)}LiI;y$f?ch(`G1olp>WdD?
z8}$pPM{D&JzpoEUh~|2>?u`DL-K*(D$q?#IF5>#mwkY09r#o6G1^X#a8kQdzW}|3R
z5qQCw2`uEZI+YwZw~Lp&a&mH}A;3!=B0m#x%E`!dWcFofIA+Dho;`OBqyM_%BY5N{
zlx(lt6XEAnqu8eOINj7)d^_>B&Rk1auH}4NfG^+Xvp26_Q`_6KT=-MbDx&UkafKd-
zFo-&DcZ$RmqfU?RJ;tuPB|wTi-tfx0-7JsVnv0M^AP|`MG1Qd?-ox*iML@7KX_YMx
z{xqLP9W{Ovd~ZIK>sl{x*iev}=?jP0i{?C^^qDhftok)<RmdEdvnZGPGi`6;oV1Rg
zGG6>MsKA&=g^U=&E*U^kV02LH8p2eX)i93gIM|w>9vT`t8!1I|{rYvBY8H0OIf<)R
zug2;&2Xo&b=K6B`&Ygpiii)?m!x_J!0DOCaF1yfA&+n_Fw&N$Au2=#DJAG=V7iMSY
z2+0oms!=Uwd?W~$W2Iu76$}2&2Fg}=U2em^w91MzzpFZiL`yXmLP97o&Or6X=bpDN
z&xE#T0;Dx5C~nb{{%Kk<`bYIUWIVqU#ht<6JS$ZR^%Z{5r5S!6VF0H2r$7Jx_5b|u
zPzNrzy@y?WeNP#hFZT8Ki_Ff<$cTzw;b3R)2%=Zg*Vp&(@OTZ*>6Mk0xmF8u4<cG~
zU)A5Zdiin#d2BH60p@;X<>T;&#^K-M;wbwd84j%2q$DSQOH8D;C{(BVL~*dc|0Oe%
z`QYeCUSI!ySnF$>lfxa!5GIW#PjBy%gv=Q2&{WiXx+(H#7{|2iV9rs9YMd_fo@B5G
z*O=MtW`eSsjZZt=_T;oQZ&fw5<fNqeMdug}!`qCSC81T-)dtgbpYd+p685^ySTsG(
z+0R}V{!|EWU|@j2damUT9RoxB;^Ldd<>mNy@6J$Az$YbSRQ~byy{jwjg9rXL%L9+;
z71J7I74meUhS@F;Xp}jHIXO9Dj*gCI1qB5Oq^GC%3=gAg-q$Z*upcroq%A;(<t8U5
zw{5X&3Fn|sfh*<W<WypWcdDy<y12TU2=&3AYGHB7pe>Rc4;%Z@yLa#KZr^@N#A%`<
z_~c3W;o+ghTw9bE=NzM13_U%)E8Bwy&x=Q;Zn(QTI6A@wx?Ay(v_^6VD5k3lrz)n$
zhK3Rh6<IH>LkaBiX!zW&a~?bN_{!Gu$cXw0J6TNq)vMPiac|zNfz&o(zcnYUQDo%_
zuaqejt*b&!5epZ-xU>{^c<4ZP|NhrFGN(t?Cbn&hy{Qq>NlC>QFI__3Ds^^TgZ#nV
zZ9Ed1^7ZRm#Y|29qmTHS9hqtct68W$)aiI)PtWtw_ZP0c%v3LY`D<a}+v%wbAu%zP
zsQ)vC-qct9gM*^h*49?5Bivo=>AH4yT=V0nE90R*fBtk44o2pgcPDB-#?CM}784Wm
znC4Q%6Gr&fv#s6|jq7~2HP^b-%idaQw=vbOL5XumM<qHvWl;%<y<QR=9Q?36Ik|C3
zp>twl;=|(NBB!IHW3h#WMQG+#UXy}|!=s~2DH%N=-XQ=M;Z^v<G(zE~8o*=9(Q;hj
zIB5!^Cp(%&p7v9x_NGMe8@@Jb57MljZ4OB;C<t+Lb9;tYo4SYoS3RSnlC7<+fk8pS
zrYG&FFCuuBPXq-c+6Rr83WiToTh;Qn2|w2v7Oj2${5igRVE_trd0APRx$nXDLU&_(
zd!|x1`=0+jaccfZA2y>OOd*CDE6am}7hNY~P#a{ZPGuF9J31XpaYaSN2Fs<hHpQ7z
z1V=|l=#Rm}!}A?5fK(*c<I>Vn?&e@dLzk1Cehv6I<qp5TxjQIjYC5YZC<IOaq&<ji
zXlx`W9HgBPnI2*NYCw2!cqnIPW|o;~U_wR~lX=3su&|K(x|MLcc6HU5Br3}Hq#JP(
z5FQXPo#S+{IRNLidVQS*DhFSC_f)OdLh<;?>Tp6r0=L$#U2Scx4hzBSy`7yqWvgrx
z;x2(^hT(n1)=PavII1ck&9i$+h!fEWHodmQZ{NOsSaV);q!YxuM=v%3H4S|w5KadA
z`o9+Ew+EKcp^Uz=*x2__L_{FR-;0v!e<H+<+^;56evIuuYGPZtSGLFh6#D~j2F~;^
z0wQ?8ueAVgE|H-D85|_lLMQGVtAkAw1%EuLB%=OTk=&MU4~+&`j?m$?Jvlk)don|V
z7_^#g3Tmycu6`F46@@;q=gysj<O3JU&@88=MYglEBWrEVp{u7C05C?)KR(xS4SZ~w
znVDr2cBHknwfW@_?9Tug%FEAxs9Iu|MNUasEvKavqO7i7vY>9W>n$F_bVr>D|2h^{
zw~})wCKi@f)c$9j^W=2!8jB|gl!``{#q7Czv$Lf+uUmPHvXtdgQ&NZnxXB~m2n!42
z-Mo2o2W~XeI=T#)7#J)h&c`5K>Lw<>Y8!POSXY`Z3wXQiUtF{*s(>uw01v-+eSQ6s
z&B{=3F`SF)V%s&TQm2FL=~NfvqX}26XPNy+JKNjwT7?$6XK)Pq9T-zmx;AH<2cvjx
zsp&p#PHLf0fS*_~afb&7!`R`AH-oe?0Mh_8HOv`2E+H8X(Q+Bg%+7u=+TX84a$<2L
zm(8F>VyK{C&>1VdJd|gMtDtaV_F&Z!Zrg=Rmqbbr<|5vq;jgW@nApRU&5-g)Ru+~Q
z1cZbXzV%yqzMXK>S5{VH?8kNV^h8Rx+IZhU9Gr8sSzFb})$g1fb=-~5&&%UJJ~=7h
z&T5P17l$C)PEAb}&rD4v+H=ObeE9_>B_)k<iNGz1Mt%0K{?YcVJt13L+edZD^5%e~
z-Q~KwUibI)w5*U3yr`%4(9(VJ;w_VAiQNv;k{Dk>Ru;#4b4Jt}k}g(me*W$aOw8lO
znA3%rA`VIMrG*7^>0Rc*%(6qe&u+d}6omcGa(EvpC@2Qp$y~T(gJ-caXDt^B%1?|Z
zVo=BPYdt+ZMyZ*ZoVz<ae#0w?MMXvJCsB~<KYjWX3qQ=xBY}Z<;Q-W_mzI{wX+OoX
zM;q#?D6$ZKUh|B#GfSbVMOP5h(w#mh;b0Uu8@=Jv$B*AV_KzO~yf$M~yr5d$133tF
z7=9MIl(PVzJm>A3=UO911zlaw2?z+da|%CA99pq)A2ceXoWl{KisUk%(nBY@5(tC-
zNpa|xewvXN=ucKXr*L@hak2iW@E8s!?fSGY;?zV_b1ac_%<h%j@XABX8#i25oL85Y
zj<#(>D_^jT#Ds*1PzNlOL1`z`58q8i9)utl`q;12Q(nG1kq!vdcXw^vl+R{axuwu@
z{wAbTMu6XL^hy~L`&suy36D75S?A}sH8(Yd;}a4d%BiY`JCBz+)6TcYNI#5d<u6!d
z7J8-i2I^-b9B4u2<5gR8;o90W3JMBZL|hrOS^T6}_I#v87PD7}J33gHIXJ8}?d<l&
zX`~{5jN6Vo(?TU_Yz|>YRc)YST^~xbk)h$zaejXOh3S*M0;9nvP*Vk7(W@g*4C$&_
z1mK;vii>H=C(UhbIUToubsmfzk3tUwb#Lp=T4rv)#dy>W_vFN)!;UpAC)T#;7&1Wj
zPa$9Q&pLNQ<U-a>tDX!E@$+Lf9WAnEb~fAoeiH4tDoRUBD|!P93srUGvb(IJ`s2l!
zh^Q!j1PzVe$euI&ORCH7ePizjI7pY9sErLU_ujjAuk6m9J68-03^wxe^5!--H?Ncz
z;0cq5Gz&9@c%f@t$u)eY7PTKQ9?O}UW;z0VQOp^PWM^mB;kBIeURo_)iw;~uO_QOX
zv1f0xBH7=1@9{r&+*{o_$X@Rq>wCI2*Uky(QQ)K-9@{`mTH4!`l$6ZthW!r6F~>Fj
ztC*OPK>-15wk4VNKGXV@?=R%2Gxf(KZQxkS3kiMLI6l~7{rK^tr+r~za8)dYpoYdM
zhi`43RPd!sm+D=fg`Z5ilXZFG(%g3Y@S#M9ozQY>dRh%GV+`}!wPrddCL3W9k#QSK
z&(6K&zU=I5r2Xl^JY$kn)B}kb%bNCfUs9*_%I&$}V7(N>Hy6q5XZ)ZdH8wVGuUBIc
zX*nz;gnEsRiY_Ecg^7uaXD+yZ%`D28oYW>HB~4siH4&7R?fe9#gr-oPX)hXxc}JU7
zF>tEt>+GC7TpB(57UQhe<~5(p%*@G|#(-eGXnX1nsJ^TKnxBR=TT7OUIS2;p(g76!
zsL#^g-hN=L)Dai53>M(aaVZjFVyh7`$82B{8dhu~5XbA)``t;>G?9GvT?M73QKivZ
zZ;-}zcA;>+-*a<&A_D?&cRK{JNZ{HXEnd8+IMl`UXuN6@a4-<Cp`IB62|*68)`^Fb
zQh<?>akBbVy%V%(gPu~K%0J!z^jul_8FY+uRaI3#pxkh>GBK^185tR+BpUb(uk;sm
zw#Q(k_7dvlsbV?rPr4`i_}p+EUWv{{=A@>2b#--Na&vQEdh+Cn4L&~p@akwe0@woj
zM*j#LtAd&(2L|GhkPx-gql0;@Xdde(3;rj=J5DI+TD^MZ6%9?z=vj0!n@W~XdI>lB
zGe19n!0ySe@Y*n>{3&~ZMQDFzm-^C+ZR>qV@{Nb{?-_)@j`|dopZtKZq-DM>>S#9;
zd9rklMmjpOgX`$vpi>ZwUkBZWk!61WuI~uFV8(*TH|YJEGhKF4YYN6)coWOY#<=_@
zHWW~bP-YV1;|U(Hv0+6vwhWz&t~u4`1LDp~Pj5X?pznE$VF6Rz&zh+27!MDI#>`Ti
z^(fo%JfNGZ2X22&K73)r6Xn*g>1Zr$ZM}o8D}AjIoQPxtRie?E<}f(!zX8r*j^3*q
zWCD@fZM<utboBH|%i_ii9>65sIbAA0HB?bn9%zPSXs4>BHEt;>DGA?f(LXILa0M{v
zcX-+v>V?^BaplLey_{oC-Oryt7YZnqTQ?x}bq@~@$1F#ktSR+8S1Rx68+tREZM`%N
z=h_(Ra2aFWK~J*$b2+&wEoJ3>3o)@zvk-&Z!l`BfJA(tzLqQ(o?Ca^NQBY7^9n98k
z&MCH8696xL$#OZ~!#nZq0=BF5rX?G7Y-=S)8S4w_RJKif!>OX8?0oDxqN1YQ`<0i-
zl;Lb)a#|4Us3ZdP=nP}i^M+Gv4o_y-pkO$CZ^W{l5!HM!Aa)O@m6nO_aJ>9<Cvyz}
z9nH>VGMcr*`DijLYikX|{SN%ww@>#!Lot|Z<s8R5b?_R|(Qwq$qbWVv9j=?2n)-?E
z#^F;2tGy(3#W>4ZLx$QH2}dByW)Y{;h|_h*NQb<{KD1GM$ncbuKwtoSdwU(pAQnoX
zI*?w$!a~}ZnaS`VPmd)zZGhjPF4Z&@gWG+0vW6rJ@bl9yAyqS>zA3c1wRH~q5~MN!
zGCZgoCrX<3wFj<b#~94pO9#S~o<b$qUm~JNj*iArrweIm4BcA&yplT~=5!o*6rL-p
zGTd$^J3U=Poua=1gTjJ_ifW_@_l_Xc)!FsQ>Pylw0>%ec8ZFPEBF6>pjp15sZMyVX
z&WgPsUUBJKuu?Vc+sGMS5fZA3fJSo*ZpbtZJkqM?ZqpIJIPfNTO*NOLq1SOagw#Tv
z)NNl~HeU7&i9|%f@sd$hb@&Q_Z(H@scd)9e8kdNOC<-1->1>^onc0^_>6ld|6<NR2
zoEVf-Q16?lt{~s0CNT<%$<n>Cql2Z)@>AT@UOI_6rVq@HqY6q&0o@BD!~OjsdJn>U
zjK<4O1_5_oNhQC0_YtL|4Vl9)^67EZ$$0&9MMY{L{|?6+8n=>mc6YPU(D@iZwTr|0
z#AAcCu~O8+{QP`jSlt>>3t#8vUZcFdy_LqCwj{+>R0btCN}eku3=OFsR+F9nFoodB
zG#V}1*+3p9A>W8gOV@jy@-#Fz$3>r>LA?_3EjzA@Fja-yW)gly($SIk<e(jy-xAI5
z*zdj29ix%|vK9%R5_TptHO^;oMVuc$uJqM%IhfTGI9?ajrdvFhF>B9Fe52abWh+7#
zYJH~BSP8dmH~TQ0Qw@MhB@1<H5Q4{|a&leLfOtMe7<BR9vaqyFPEBn_Ffy*pa%v(+
zEgrH3nDk*J8W39B+RDS>%F4}UH$k1AAYCBo$gxX0zVqNKXpx<cNjcbWly=E}z|XIN
zq*N3T*q?BXDcPU$LbNtD6@ANe3~8<{103oCxeu|ju;52(H}MYD!`|u|gnj-C@6fNX
zwM48jN9Kjd*4y%$vf7a;#@G9srZ+mWUe$kq?k1*KooVvjCDQ#p=tEXs+ieM;j<=vm
zG=|uA-T{;~*5BD_E<ZNCy=~Rp(a}K$@U@(blyucdU%$c&dDe#vA?51onmB$u$C+8S
zQAc#7s;X*IG|vSjF*><?oP_s4RU?jF*RfFW-zn7m{opWX0I&fb&l*}<jt|y=P>mtx
zwUPKaC=dbN$H^=+k`-=Cl8}(l!0z&3z;}c2t=IcTg!~tuKa-Y59mJe&C&i3-PEH3Y
zX?_Eo5ekr?RB0TEI*HhHhT`=rJe&}iN4A3ee7Cjna-^b_)pkz3FIfon(z@%?tXsd6
z2Pp+bWpv2i7G&QQ8crT3Zy>tm<mEr2?}#HG>7?5DVP6hpU^$?3%4BdJt@1%$VmBUQ
zM-_}YECoSZGX{Xrp<6&nL16(-k^{xzg7+OZA)jf1<;BI9rZbif@ErBc%sNnzMKcGC
z@}2vqt$s9ip%W)$H%=hBm4He!GXu_l98EfQ+-bg6ezIaJ3O!F!e0<3!Ae7am#l<3n
znAG~|Z!`T`{f0uGowBmDSp54FCwfvbWyx#B49#e6F0S?UbV^f`%ON_aByqmlC!iF)
zIY?fThmV2r>Vv4z<L5>Xr^}&S)Y)IWq*ZuS@veSYEqwU)rHi9q=Gkt$K#h2^R=TGt
zBPQ153*~IbR&Jl*=)y>6*Je+5J0C9cc;DPoTDqOlmIPYPZMfYFz+t-_pVX6K9Szr$
zAvVW&*ObW|_9r7YzxsBX_bUX3pw`eK!I?G}mzJy3OP^U5uAirBn$lmIXb>0{maIoG
zmju0X5K#JFzazE%m*`zIn4j*|LLK%WJ>3%@+t#83$aNFg52VS1u+vqjKQ5e$g?vZL
zIr05Vk}4`HnXh$qSBm|PPBRtIDU8qFVy&Y?W(=CjmW&LhGix6<?BTkAU9R7rN1zUR
z_r{UP@dWV@lf!vFK0b-e(%o<^fZ=uU=w2%-Dvb8*jm}#5b?P4vqK;;nIn{rfO*)GB
zl9n1&*CzpI*inueM;T6DnDh-kIVwlSDL#3wa8DL5C^R(m0_wOOb^PSbo5d2|<hfT?
zRkS_*y(fIMv<ozVJ-WxQ#<Xi$^FzLHn@8>ypA)REjzFK(dc^7ebVSCSe(p~-I1EmQ
zoaG0SBO+wZLb%*hP9^kYLPGHkLd0ZBa&m`@ylZ8=dJm-UMmC(ZG&S7}W>DE=uQ#9d
zr*if%RfNO-)K=t@WW+<`xeZ?dvBMQpq@(x;-_#i9PfrR93*CT7tN!e8gSlk=tnp}3
zFX{k-cowEJM<xl(1;BId)_CG>Z7>m(V<C>e<eQ8xqL$!E1YtqO3BFABijNd4A}GkZ
zd1OyMNeY<V*%N)}OidCRbNCT4r_)O1L;W2i#7b&vN1KpZepS3XCx1wslA5{z@sHFi
z7)G_d$s6McN+KA;)a0@FFxMJGW(6$|**zMXCDV{4QxlU!+cEok1RvjVC1m#qy0_lx
ze36h)Y1$E}V`*P91e3b^x_K)-*{k-+VqYh(vMuGR<9>gw2uHS_PtLu3b5I*P8nch%
z&{MB=gvHFW;Y!y_SFj(MI3i9Lqzw)>XFH*u%Wf-K=YDKAG&W9vhaP%*RE|1QNbN1k
zEGWQtwYXN?H){`F4B%x>=yFOA0T--Be4_TUeV`J&py+HG#qStQrE-cuqHK}<{^0pK
zi(4BUeA{8QaDMeAf!k_sdlI^}gi>kZu#(rW$>nmqUouSVbm8K~E*PkoMhQ3{mJ2_9
zI>z3<Bu`HU!<d74q{Z?8t9Gp9#ro3C$CU2ZOE#1G=g{QC@qA?D$Q3R8=4^JyU88V*
z$A=3$4lwcpa`hLGTK%&>{DA0cYHQ1nPe?e4L32-2@?>Xb@?(rxF$EzP?CqV*BTx0@
zw6!N&8jtt5d?=%&MBd-UZuSW@{GKBADfSDb7By|P+N2%zl4q93$FE6`)~n+I=Q?WX
z>CMNVI=;XAMKiN}DAG$;1*ilnQ3_eS1qYk;GEx|C22cMoMeUmE_4o8(H)kuY9fio^
z=~j`=VOD$J8qs7EmIV%Uifh$yPcDYn)(6NTeK-o{s4WC4C$q57-U4ygK6=}M-<;Ui
z`WTu=m0%QVw~aSrVcK~@n)*!Djo}|Um$@)-&Od*2Ug!ST_l{ogQ&rRRdU;fhkF0x&
zY87s&e~k@pN_o3{@9tTO22AZW9RDvfTStQH{ifv_)aPSM_m*nO7%d7%MlEgI`LA&i
z`_!}^k*%=#PuTM3?qPchRZN<#?M8n)5m9%pVZqGK$<D=XX=Pr4Zc5|Tt5+j{#zlJc
zP~-b7{(*t#A3l6IB;rF{7#J8h^YCWVCBJ9qDg5@Oq*yT>3q-Ky54*a%U9S^8tZ5Bn
zqxn+%k`>s0I#DdHH3l+>UD6w~)1?e#f;IK^uJ7Mhr2yOQtnGz4WyU3=pm2eWEk_DC
zpc`JbMY8hp0de2HJzic~s^;P4nQ3ipu}O?ipe?i8c#=?9c!#-(->vdR1$;Dqfum)%
z0)Xmu*GD_CpPf1Q<?;FRM<{*0Kw(6RiiyQ~Rp-5vpmqfc7xJ5%hK2?S8z5&FXJ=;_
z85xYX1?HE5<XxX@B{JZN5-85d@PX;?1%D$ZlBr_U@ujpR33D#wX-RdeQfAHO9pHps
z=<8or4G%YZs-e*`koD>)Ej|5ue0==dMjb^Uy(WPvc_ged7~MObsU|CX28e;1(~0l3
zL)D$sBVB;1A;GNvQ7-q01@j!RLq}6nFEzBi1Q!+;t6+xpMiWL7rvgL;kM3e{aB!4l
zXIr$6Yy67gcf81JvuuVio2U>l@BQi*65<ZorHX)&(Z*U*vPD_2+U5H7&)+0uI{Nx9
zLUw(WAh*Iz>|<hY&&@)BNw83|i~0c9|7&sathpi$iVYrKMNQ55+}zv&U`ts6W1hgI
z>n<GiubuiAfq%m$B2r6^jeYRBdh%D9{nkYwSKKv{r!5N2P^ZpCweg~QCT<_O1M%<+
zzT!SBf%JI^$e-Ph?(W52%}+${Vn33X$KBr9c?>rjUbqXW?^I~jHVE9#c-6j#MzW)~
zGpl}bqey}p_syF(j{;x1Q&Us_fI<0y0!%?*zM>-)@lcYYPE8J?WoB;fePm>$uCVY&
z8C!{)p67RWcbD&Z;nxoh4}SFKc+eCS9K8Cjvg)1S$6GKlprnxwiyi3c$(@G8N1B|R
z%u`h~{gU7!vsPKWl9EDHB@5;XE(8kK9fxEt;JrHX4Ewho930$sZass0_rk<vY;JP0
zvvJbQ2MPv9M0j|em5og{Qp-$_1tS`QgX5^~Gtmqn?WNPx)84SKuqL<+X3ux8(y#A4
z`X;lyi?qB?#n11o%qRkl-PhRIE^f$MK9Iz&J3F6@!1&5s;ra8I7;mp*IJly63Eit7
z1C`YvmprbhqyR*E!u7zQkaiDG&u{S0zC&j>DI$KSvOMp}HbzATa&v8ZYO2CvbLLCC
z?V2WXq3hl^0ZZU<TdLi$Oz$%=c;{tjJ7ku_H|@RM-rg=Rm5`5*U-p5yBZwv>9l5kP
z8g(APOy>$QG4T-e?M&@aY?sz1UbF`($s7~zmz*K?+eTa}1K!hF)1_>zy5U*M`;z0f
zU!erGv9T0EySp|e@%O(G-?%hjwCkbFWzI$1(A4xkJX|#?DT&5iEMbi0;lnlvs#!W(
z+S=pQ&99d*e%P*!y*c~tMNDYucl<kdL=|LY2JR3Mr7CM^q<4*uekNnmg~WZKEeesD
znei=av0n?xfA}r(mK`4{C`qtAYit#(z7+-r2VXq<4nrt*hIm>qvyO?0$plbCoCxsi
zKk}t1{2sx{X3?(t@Q_BNzD@PwbZBa-;a;vmw;3HB-2#Tp)^NQq^j&b|e=svK^&$7y
z$qR~$9j}u5Y5}VxpA3()HX|bgd29v&aAUZ@)Hecwba^2y?OSu7%wOjK14G})=scjL
zY&g-mEUaPY&z?CxKIVc+AK&7yU;C2qXy1B!x1g3eFz#Z!y^WzSDkg!+V=?35xE(;F
zCmqd`prN23Y(Ktr#=ZKTzyD1uYujoseCDs&**7&5{myP~`1$*vrMlTj36W!JY`pSA
zx3aUd^9&JDgeqW8eFcTy>9)4ED2OOyjJx|e_yz#{2uhR@#CQYs!pvHlDVdq$8=!-5
zq@to;EOFQod-<sd;^*xrH|h<*G%kLAe%q6BU!GCj3=I#@1`6vH@IAyY4Gh9~0fpN#
zGnIckEI>UD3c677V+|7%^Zdn&7dr+9?!#0xy{EG?YZ@NQ&i(uMJ8=49S^OIs8s6f4
zw2%-MuEoN5{?ULE0Lv1P>4>Z;W%|<6(z}$Df+dc78CuXeIR8}VxqSYtH{}13k&!p0
zr6sYw9PB<3p`krzF-+Fyk%6%00puCRTPw`!u5=pdI$LXN;l+g|`iJc7%lkY1TKq5s
zw5nOTeP8rTqfRJ`_F2cKK-w%A7?LY0Di-$j^&zPcf+6nXO|m?0+6yKoCU(>TcVOIz
zH#|IS3%S{@gn_PWfsh(Mv8zk&<+EoV<M>)m=JnOpu9GV(jDD{?t0&PJOF}~8tyn@I
zOcW9w?Cc2c)6)w=JxYN00Wv95D<nAB0>;m5aIB~j*JNd7WmHs>32)xCphh5K1k|rm
zy?cbg#>N&ZV63{rpwzEV7vwO~)|L?<mT)?I@Q^CW)esY7V!kRw7^PC+pit2~HKm&*
z6@$b6@SzkvGc&50t!-$C5cs*<MA+CrcHu2suHaHXtFgmcb&ZGj<VS~lLogQ5^P#-9
zCPt4QRSZ>CJ(+{w_~6I({rmUzz<_`)ygE7~ZEe;cCZk2wK0ZDo7!za)w6<0jXE8jI
zS|T{}^3oZ(9z57gg#(UsDv0gWpO~7eie}T3T4ulGzE$Mlz!;~^@$~BDCuw+icop%Y
z0l8SDd<21CzC0rJc?fS^IW`tOJuzYSdS|hB5<+kgBv?3$6GYG9OTn_U-}?~rzrn@9
z!C7A(%xQ&?lk=BRqhx2tV_;y|hPkz(f}+B8@d#kC-+w%FW@T(_Z2r2pcbUN(eS%Pc
zgowl8ab?1>s4Ey9;p^K5UkXA(9930^2E{{|4e{~tWT?1Zu|t|KJd7Bv#{U)o$(p}h
zMWnhZU9I2+1jM0qG@lBGde{^^qzf=sSpg(iy1KGLa(cM4v}`_IcZHH|O>+w7lla>J
zLWiNRAxF<LBP=Z};-S4lEb>bNrTfFG(ccD++*WV^4N_h~h5LoJHhCA<>E>4ZV*$(g
zB0b&L(WOp1DK_TjtinCb3?a=Qfe0I7ePbvX%%qn8Jt-$gO-fewWu2^))C~x!0Vr4A
zaHX%HM_;zKvB{X*I#|HFD!j&tbypEX<|+n*=%y^0gJ-$J&YO<WV%r^MhQ1%U;&Gkl
z(boWj{{H!vFx*oZ;xF&z{W#rqs%G)Nj?r0{__{pQo9!0R19s4&uRH^&!^z2sYUjxb
zdvIab<v@3YSZn;zZLfQ*P4yji^K)~nJkE!?e*$QIch|?##^y%a_wU$ZklUcxoH_gM
z?&!$K@z@>_l-O%??U^>#)}=Csi?`}~T9!+9ePM#s=X|v5Oat5p@Qk~IT2Z&Sh;Piy
z&Q^AHW%swYGuO7ZGU|-{G|xhA>%V{Ggcm8`Oem&Rdez+Ayb|Nl3>iEyqw2#QX2aA}
z)0YI)9O@CL9T(lTL-7I{M>c>h(t)agB_4xBSx!{EyNFrM4r-gz(}e0y9GH#Xij9w_
z+MEZh*fQ33quA`!sby=vz5XS^(>odcN0nLC1$lX$fUe7~su%ds)6j4(zv8rBe9pkm
zZnYteIH8h5Li~MAPnVp|X%{fW{)X_Xy~hc!;0{spQUA@G8_ho@Ls`!7TK$rdFfv;H
z(Z-8W5y5GS{tcKh7`ta?n)aEc^&%Egoa0V8`_S1T2SXUu@++X5&YYZ@>S7sju9)5k
z&B<Z2nLgYJ1@1koK^n5kinh*-k~lu|TvI~>rg%9J#^VYqyF-*S#|OwElIs}y*!5$V
zFbLi~s#soG$vQdO)0+iCHZC_e&`q1bg{!Wsq2coD)^KZaRn^1NBVnqWtWX`)1hb@(
zBF2)}*&nj+U%Yg|+>C4F$4Wc!AwHtw;zTyw7cgFADZkYwfEbDfxSN%E`{H>s=vT*`
z%V5xPn}~$ORXfz?T)=_v@XpqjE8K{2NCr=wZOO>Uyc>0xG7bYo@&Pkeuy_Meaj+9B
z>@^@nAse3s8MjU`O&JIG&Yk2FI7VLViRf0Cm$$gK%E=ZQhUJM$*Ck}!i&9SK($Wed
zf*-}Z-O_mg^N{=8FzE=5iqaTg=uW!!$;-<XfT1q@UB?CM507tu!HF*}CYqU<X@L8#
z*xc53yO1BSsB*rM$}&g+@;j}q6ENOxIckZ@5;$F$0KSw|tJuc)XIa_lESYqqS+n!0
z)APDDKz)E@b!~rjkYmZFvkFn*BuYl~IlcggNfDZ)>+p&Cp6CiPKOk_+<Kg8c3knHg
zHd_02SxQXI0)?%fP2~wm^{dQ%n;7TaF!%2!k`Uv|E6XqDeWa&H`1#JkM|LlL6!q;J
zIR?f&boKRm?a?F|$;tKylhy7fFq|T5jf5)oBP&arK_z>pWAhgMON_6I44Z5f#ycB3
zCoUz0cSOUUUMsVh{q(@Aw#uu^#z8gQX)0%CXUX+_bOzSdDW5sRBmi4>_V%p6&)i^C
z|H<OkJll7psMbYhJ4gAEUvRMRWh^aJa{!Hw?5&UAcJl7uo#;qxjd=BzQ-<Eho{*3E
z!T?5rHIQl|TQT#u-@GxnP+wbH2$S*3`C^;eRx>Ni^B9S92(t@>S6mDa_BZOuY=I%v
zil6!_7>uQ))L+rpr#QDdS}eS%B`kbl!i)=>IzQ-X+&fQCQE7es?;XKbDs;D2$BRQI
zxrh1JRV+4xpg<a8spJ7*1DU`s@S1G3m6emn5H?L`zLH@m-?(w(*AEiyb^XgveLj3R
zZ?`$)si&>|E$REadBEO3AUhAhWGa1SX^Fme(u}#r`U@ce0S2bo9U$H{a!{$IS@Ma>
zzWW2E(2e}R_TD=h>p%V<z7QIURH$S}Mpjm3%LpH$VQ<M^*?UXNCY2Qt$w*da2}L3!
zBO{@ZGBd;dxW4!OyU)4L@7(|1f8C#RKIhZvbGcmC`~7;op3moFz0r5!z={gMa(2@i
zWTV+_di{D6>W5EIR{UjIAD?q%%4~O@j)vO!4gmCzhQg`VW2DBrq_wqGg<+c&7%>{f
zUFL3XuT13RNQ`>!+=HNv@toVKFP<a+MA5z6m1A}(DR%#L4a~8X{opH`_WY~Q%^?t#
zzOVDVqV>E5-t8B3WdhOx|0=H;`gUX(9>5?XJv%oy_wskVrheqhwLNMI7VI1+xqQV$
z#e9Tq?UWrD(K(K_3DlYTEwzjP+PiO`0CWp}F|Bx7(Il@9i<sE8AD_otlFHcbb4OhW
z0g!$s2{X$dCEfn7e^2IYtylb_ifk<57JeBGxbV2w_jgfho2zs6B8{J2jBY4XDrg(J
zMwm;ZL0n;!3GnNv^<7nYVmW&JinRIGzkgo@Q`D7AO*2%s4$ZH34h^X<E-$~L%`x$l
zW=82#{51G;^7sC<n?IVdCjuQDwpQwdUA|xFuk$}|0TCYwM*clm1~TK$mD-F4bV3aa
zm8SXR6a<Cc%F2Uk_Wrj$n-~Afq|T4K%P~JYdlv}(XoDIr7gII0eK)ZT&FAWq3YkpF
z409@?olH$+HW%x*HivOPl<faK$;x_z<yK!<=$QEIYSD=sAA<zy*4WN*adG)%i=`2d
zSYVY8(98sbXiCQPGju8D>*yhRddeR^e!Rl>oy4rnR(9RSj3j+-=;+UhS;pNkA&D-~
zvqNbT1o1{reSM%QketclV!?{Bk3QnZ`q3CxS5-N9T{dQF8DDs3Y~&|+lkNtfxa;^-
zzwX?=BNKh+PU#`V0KEVEp8|%L&U?*&t3D^fz9&^9QlS1{T$SIz%Z9GLqVfBTXGw8+
z)=m4~mEP!(m6J2`*4C!|FTrEpOA^2446%Ca;%|~6CV0~Ot9_)IQIKF-n|-e&+dw*?
zNM-k-(%%05-9Xk|A&MYq%{X5=m5>=y_8H6ecAB*R>wBTKTV(@GOk)sk4Kz-3htSB4
ze5Sce#i~>IcFT9wGVN)O-fis%s`~oJAjJLp61szSOs-TnZ!R6&JRm`P6s%73{K8+O
zQs=R^y2tK`nWr1+Ym-W)rk*bqK66#krlqr!CPdu+R_f%;#^2Yy&z0Qvqx*l0n|Krg
zr@>hiw-95a?KHc-LOCCw*VnIHNl@X603lvrSjOW~p4|5^E-vn%|7X<ua}$;~&Oq8y
zah8Q8Bm4rJfANw~z(21wzrMm(KrkhLgS}&~Sf$dZO+R^3YGb7{^SxfFRaAmbWlasE
z1mh<Srs!OrLLcG42poX+124l%>_5bb)+;GCX{1T6hbC#=?TEZ}YqyV*QiKVJXRRxG
zdR694eNRuGEHO4VwkBh|&~=)5jz3`+DaO(Rl3w#Hnh%ycF}4LYj(LcQi$8B}YKrQi
z=SkdK|8H?$<C`~TABZx(HGE<nP&UtYQS4p_;$vYC3hJO_WO7=Ef}*3ltLw1hPzF(|
zS8Zx*tM6cId+z&a(}xdx_>Eq_V-$BZ%(!>2eK=P$eQ=M#^s<ymz<Yp2*X$e*)``(1
zs)j?Sd0nEaau+J6FrdHl-^W)6Zlnj_zO9N;eK8!6k`eB#wu=1r#y5WmBo_hNXa1^H
zQB!M#RR0WGXVSp+N%|kw*(OFXC=d?xJypmR2Zr<hhNY#S^AbN77bTjQdpB=-jsgJs
zA{9qXqwsNMo<K$xdI<GDAG1~Z9)qyMje@tLqjh>90#m<u@n}hi1tx4LvKyCISE>D$
zmpWYV4aszC)<Tm$FW;qNiC5`HqkiF~Dmc5w+edi6Un|l(B$e&PwmrHCY01r~3O@p2
z>enx_1CeADp$j95%<F#Al74HKoX~H`5eMTCGX~`Zcfra&&jZ^m98{JhM@gKWcrPUn
zReMHvL5g#*9QFN#ZjH6pGA4}!Bn=JL4xNzu#KgtD-AzSx7fPAKzTu@ax5Yd0oS9jB
zr`5wxw?omtIQg5JPGIfh`EGuMn*A?%PoF-a``T@q)zfozD)6D%k5Lt1Uty7v-GN(w
zb)>ZN=`|lq#A1Xve3Y44#<q-EkkGIz<eJ5y_>?=Oq)$L{%T%sAdw4je2w!jg;Z(ns
zBy@~;E(71oEUHMZ@cBFQK71d&@TtR@6_=kqe}4JuwQD11hZ6gCZrdC<@a^ZXrN5sX
z(%*f0BtF+z&$#4eI`>W1FX^#8WC^F-&U4W2p%c*W`7V6($B)R^SgLHlfAw$o60)C!
zgh=ITrpD>EwK%0ndJF3s7=#v=+((<74emw?Bj+d?Asgq{E1DS65Spa46|rxts|K|^
z+$D!|gr{f#&)@Lz@-eQw`SXD1EAgrSj*A7&G0t_0KEGB5A;_J-rJuqL!!G~)q*l)_
zK!46d38yiq#$zw@g3Pf!&$T?f1w}=}(bBu1KJa6IRUN<ho;e-}Ks}r{B5bkNT=w%-
zj~_Gg8NGgGiz?U$m9l6){T|6+mP*%k?XN>;N9b`}uVwr&0?q$8HMI!()N3!GF}SiR
zpBNuXJjNSGMgl6ZygR*HfX$69jJMc|JB@HeU&y?QO?bx0s(HQr5wpJl8>uvf%<}gN
zx>LZz^!Qbi#4%=kQm0SehncjuE+D{qiHyK_AaZ(Y>YMQqx<AqZ-{1ZD>iI|I;>$^k
z-PF|77-=ZX;vG#HD4;;n2G6-O+aJq*&_YB+QjjVOa`!C7;KMZUHffI>+D9F*tyff5
zc2&7w41;x$QkMUV%*=y!H*Sc}jM~(MlM{m9FU~Kv(&uVk$Af$r&$@;IPqfrKDL*EM
z1~(!--Ne$?mZzh+*{Y|-35?3Osi_C&AmqPdNxmMcaH}O01&g-LxW4@G$${Y7TB&~Z
z6j%El9g{MuoR$+;>B#wZ)s{rqpjv2bUMgTggNZ%Pu#b-KmfzBk{Z1br2xX1rg2D=Z
z{+ze6s_MvYJM!sYzd}j)`$~%{`bukSVT4bUa52A=pFjAcp+T17{y!HlW8>GZ<Tza@
zE_S5tF{+S5HK5wRKMU38<`jlpviBPcVhq*Kp1l|2;{7!rJKnN7ROA2X-K4AHRq1!O
z>kMxP296pU89hI8_%MU~+ztZ;q^t=REzz5^1NAchR$ds4&DKdXQ#GFIC(r|Dm;0w6
zr~lH8``^ZLo?+Q;JwFUDRVK3|!-fWz$<Wc!x^PcJ%LC)su&KTMW7d-=)5yvyD;XZB
zbN;8L<x*9=-2`clphffD9^x_?ProKXFRtVy&<Nc*oydbnk5<_)t~>FJ@H`TEnb6~E
zR+{z!IAd9vd&KXayw8;N+xl#Nt$BO*#yooTRcc^x<2&eQnOpIw2S<TyJ`uO?k1?~f
z<T&Wx)!R!>@WL@=uvvGtFK}O0;eEyvhuBXXKh8ViovUpI@pvlEs@b8{ULEFpj}JIY
ze4CxBta<W8-a1|KT<U@I(p#^*H&=ceYW-NWc#@Gp0AF3~uzCN4$xxlIjxEDb;XB85
z2B^o$YHDlGVBXhOyqI#&j6p<Hyb-{}vuf#`A&Zt9>B9pH)43Hl7x&4#zAU}B=2ti4
zd}Q_l3;9G=c2NbrAk`xW8=ERk-9u$4d<@5r%V_m=ddc^Fj(sr4iSu1fPVO!taL3l=
zsYtEo91|G91Hi<>978p7mwv2o{LoCwn57SW#M##0zt*+3Fm4k6!IFcoehdxe{M2Xm
z_Z=Pdliw!KN=RG)%N-}g5P9+-PMbymgXsW+rDEgb<JYP>ySi*if-IiAGoY&ew|iGi
zQd0BF;^Im(b9=)(^sWI%sVH!8Q?s$NHw}(LH5ihXHc}E46hwD%uu<{p^XlU~`T2a<
z<WX$z74}`#`12qtYR}*r#h9~1Oj_Cjs3cib<C0@v`Te6)Q&XG&`?tOC%NM33hR?5G
zpYvLqIykwQ$XnNjk$N?qUAYUY;YS|Z|JL21i?3sd(28LT!S0<q<^6YW|JGI*;9769
z&=cH;RV#9Gekkhb=oDtB<f6frQ!_Z<)7k0V?>|WHCv8~es;CdyV^%?d`c&D>w{M-Q
zZ1D<Q1_o;Pza8AIps&&h?&6V5*qterxR@9Mpe}-0DI+3!9|&fa;d5Ovkmq5b4sIM{
zkvMhhICPw)&lLYEk#ce6eu87h<9^TNWO{3d>E!n?o9^77yxiP-dNqhZVoA9%6B84^
zOv90Ax!V`mg7fojBY-w9LqsK=KdV9PSB>Jh5AXf9ux5w0;A)(C>b5R}4eG$c;v(tg
z%iVVP@vCpM$F~}lDXBuCmh>{Z7E$jxmo@|V>fWbx8D^r2>#ISzeN}401Fs}4gKF!b
zw$x!m6KWM!xxMe}(<rH#8Z`R~r2-eehhLh0<SfCcjuO}lg}6+dsLVDuz+_WX(-<c^
zyEfWx<-0c@i2W9hFq2Qw1zsu}t<fQVM*N0%Rh8-&#lz8*=$!|F-L^w3d0It(n_2A9
z{<O5TBz4ZqO1f1puC5^v;4=Kqn3j1G-4r68nAn2K&EYcwvA{PdpIgBF7}z^FoXmIX
z|4Mn{OzzP${a1Ldp4w4I=fKF{M@z8e;o)uS(j+G*e<0=SQC43sqX73-R@`OI_&rc|
z%JZ_|NXf{{OWB#uO<!z|W{wuCGm|$qCIZUl^nAL4Pm-cp@n_Sqf9t0`%Eel8G>&xC
zy_s|!!>^|~cu;Q7{{4TAYOUHMUZnfEPNS=%pHoySR8BhhfwGt6;F<OH^-Gs8lVSbH
z+iiJxcof_9@ng?K#&c<^_N+b<v7x<`*YgEm8IbQl$AydWn=*q!!GwT-fOU`t*BK)A
zB)+!AxqV5?=)`M4uYW*3stj#7dwNoW7hr;_y6%+R3iEMb>fnQ3K33=4cO+9Y>odC2
zFjCoH)=v$$c`?<R!oDbli|`LBg{V|Y^hOVaER>?5Eb6ooZ6Z&T4Fa_~VP|XW*xlWI
zvmh_;GtCOEk+wD|m^<@nuBZz$Iof%dTU*RJ&va0c0n*1zyS;u@Owxs4Tnv1X;hXCi
z_Pq7@aZWS>yMaj7d>V6B>RA-pg<EmAjf5pO>v_K2cU4v6swm#iWD#baI6lFwz>Z@(
zgRw_mR7hyghxZ>XzxMaPd-yOV9P|`er0nJU*N%C-OP?LH-u+Guo-&{~LB-q?fCDaF
zzrI8gB<k{ZSCpTP-fK5*cmQ}!7>j&S9dI@Gyrmd}izEnt35%5)TVEvFyS0rCA{`o2
z{4WAR*=L1?6lUETu$qx1x3!O!-k{c`H8RdoPa#!OQsO&$G{x^?8YO@i=bJa<b=Ltl
z2BU-oftTxaz|ac^O)ywi7pbmdm=>30%IUY!Ut2plt-dNQo<IDh_Q3@XVfFw1dw%ls
z>mAn1Iy#gnd`T$AEcdUS*w3i9myKD{)XUq~$+$kizfGm;D7BsvRKBMu4=~OTpk|DX
z25dU4!3Id3-jRFd_J^)0ufQ1c3m;*dCc!n8P8aAI#r;?KS5zO-+|=R;6^`13=d03E
zGO$@_j4l}&NfkJ(uB^zTJKOEskoW9a(2EzRovY$foULoB&Z=CzxB!Qq&%9SnvkIMJ
zyyffaYUs1%tjWUw$e_8n6(^<ixO3V~P*8BHdCWQU3QuoVMTHLAW?uHQ_o>3(uJxya
zi@(2vLmU0TI^?cW1eOHy?5r#@Fk=}dC0cy`BI4qape5x1bVGqQn|1E13P+qsdIvpE
zC`Gt@@R#!PD_F*N$E3x#e_zqjQGk+6Y~Q{+Oz>mJ^4{KeXY{N2qu%3Ris~E8N<EEg
z+%DQm)N*X^rD_Cvx?fmW_?dul%{)YFp8Vh^|1K`>eIv*>3G%)+F0PlFn`L*2&G-`*
zUfzAMUdZzbvIhfq1W}cto!2<x0Y%fRii&+(TYCQfROeES+v6{hhw*SxN-!1@5hjLk
z`8{SOx+>O$peA!U1A|AMx4AHLfgV>a_`YZU%hs;w76t4Dl_#VjE#s``xqj|Rk@uEQ
z%AnG%*Aa1%FuW5S{6>iJ0R3X!_6RyLh)Oc*>X^V<N7M#_<FmB1T<q=X(PcKcJphFZ
zORS1~OCZheU7bBWVQFbhug0Q+KYmo#hWwZbxDv7S*1r+7e`i+!w4_1MURKj&XCC<D
z>-_xZWVDA4Ip8ZQ6;Lxo$l)_(=j3Rk3GxHXZTO|YIMw`yQ++V#S&q{eKh*D;2svrh
z)7#5>>Xb^<DNCN+!YNd^VYEL3w>H%X_ms$7vg0OQs~XHp!dk6^@zr?GpHGy}w2aB-
zD4p*!jUz$J^GhzzYloRBn$lj(U`n0SQurtwZzuH~aURQ|u{}lWQ^iHITb3<$!$#zo
zj`kUGSRjJYf4JcW;|65%bNNZ{40kxQ3kXny@5%%dHq2~5I0q6Rw<z-k#%vO6kwImx
zgT}N6ckh~CTd>H(nX>C7C+E!8a<^ud2NtBnr{B=Md>tI*?|)@;IU+Xp;5TIn(Tx~I
z4!AO3|61yh0)pa-WpB;5_X@KVaqp<#bPhrbeGU>7ox-dvg=cCau__5uJ@+uMVjt*k
zNK|_J_t=H0b7J;9zOY}fi&~Q@n(Gy5NaI)#v=hOyvZ3p-XWW91lwX>TdC7agE~TbM
z61*pAdPihPNJzBftm5#aJ-bK;SKkp;zu+tsju1W(kv=0253xox_E$C*dv#3Y;D}PV
ztT0|aUFmZ7%#Dsgx+6!X?<c5zaq|8<a|DxgwX4Kqb=WGFpkK%`XWgg><uLg;6gW?S
z^l7_Tf#J8s{3SGY!AW!<+P|Fbk#SjzIuvyQ@Ur{~H|&3BiN_+3+Bwx+DIS7Z$Dx#L
z?V0)cQtPuT`JKs!aD0HLWwAf<H#P&hER;YFF%3l)8p55XxGDFjJ!_XtOsLTYy1xA*
zqwpsSKzied45lhL9U)R#bCIH%Uw7X8bm9{3-QjB=m|gG5ZKip_G8l~Us2cu1qlc2-
zFIRy5E58uHyR#@}KIS~Qdtd!ft#6}_Z(YIvP^}^brYHbwu*42RnO^s-;9395rH@T2
z2USRpu9Gy6EBZ{X)9`CU1A*TD)_9!2j3QCemWGd>_I-Yi=>QPLO!xvNz{aQ@-dWhW
zbn0+Sax%To2%vPFWF$x+;8;0-4QulH^^<p-_kEKI?)VuBNXZzwWFte8!6#jVJyZ-4
zd4&e0R=LFs@_^(ra&jo%(w+Ioi`v665`Inh&rsdz=UG|frwV?Zif##o19s}qABr$9
zGeHLDzcW`?`?)j=_4V{VD}SA|9*a9E*w-f>F!ufAWzE!cWuCtW%LZWkWpMG7L}w$j
zwS35><x^^a6A#bSsNd3leLnWxu5W11K;t6I8J}5NdMIp9jjfYYBp|RVD3%)p)!Bw)
z(Hmjy1xYOWUCz6LZsO2jVh&uA|A3U(m~ACAK_MVT<j+xSlC!9EcckX<e)1i@hz<aU
z7M%Ub7l0Nx=eEGVy*h2w%+h=%OBqBF1K3L&2u>?bmIQi-hYKDI>g?>{aJ#RGI@Q_T
zedrD^4MbQxi9990F6qwiA{9*eC@q}wPen<|DpSzGxytZaK|un)fB+X6fxfE#pAbH0
zMs9B3Xiv{B^wwQ(d8h$BI?v4NRg7#0MA!W65cq+QlbG21%rEy%6&*p=4wecRU*ApN
zXeOzb{`230-JP7~W)QG&s-mJ|JpA5{A2Uu(79pX}8ag_wUE5omeHsr%qRx8$F?=;2
zL%KL-0mT%~xcQC0cA1B1=<blCXF7lR@@00D`he3e0lEy)l(H-|tA6WGpGpk9vlcnV
zaP(-fAW06v>9Qgx2ZxG1T?y{8sc;JM9Wd3%VaOw@i{20IpZj5?o%k9@R&jB0%uxo0
z;)Dspdr0HGR%TP)8E0Uufjyl9kZzcx0xmBcIkJrIdx6ZNtokRo9I(yLm*t4fAK2QW
zB4pIhjV`VG35tj;7v$$JPPkT4e=c0D2eTfu6GA=!Tf8U6yz9UvRssLSg*iY-f^l)E
z?)TS<zJH5t0wr<tCqR*t5NuICY=?%2ndv8_>O~qR|I~Q@EYw#v_&VG5Qdku_CS9Ye
z)vsT(@biBf{`AQaElr15Mq=dQm}p|}@0mx@*47s8UlmjBzbPh1%X>XJAt@xK>g<2l
zu6cN)6)&zR%Fo~1+}w=57|>>hLaxAQ&Gj~c(ZE$BalN-M&0yP72n?(@badpuowU8b
z$fQ03NB1SVJCo#uIivlxd3k$2wznq-t`2+F!~N(6V(4Pe>YZD+AQ^Itnc$fsI2nm+
zYim#Z8LA?K)+d`oJ;2=D+!P97*N$`<di{WAYr953c0s3d!pNEJB2LhzP+>J6MS;q*
z6!tC=EMY`LHUXq|I=Wq6gc1DeTQuF{j3-X?^?vCQJTN<2l%M;&`ui`ZIw9JX5=}_P
z%*~&amy|3#)s$YhG{U@#Lui%bZK*s6D56wqvY+4?dgsykr-kKTYG>OY?JrVQEeP~p
z+~sSeuWILD*SNa8OjT`I8?bfk&q(dBh2IsUgAV=0tm77Hk4{HrFwh7=wm?7&&qmHl
zclhvooI+|BKR$6~7dcFfy|9^{Hn&L+Y6%4_D;;%#b|!4*_iv((unnsz<;I5(wBi9+
zMyBTGFH2F8j$CXDG#SL@i7&z-aK)4=y@Qf7-UjyY%PG*+s0^(Fk|f7sfXc%>>XRI%
zCn}4U`@>OY4{WuxV&DDM)YPO#3toBC)KvD=I&1s~oGj>*uZwu_(blY9KrhI5`gFL)
z8)JwO)j7Mdy6QPpU7sYzLUwq9$iE(EWUQfR6l{K6b#tuYP3UQO1xH~Qb&WWcS3O`a
z78V}<c~0>Hc+;5#TUYJ4cm9ElQB)9rN+~J-{0v+<xj?_xhbV@~#r8J>Jr_`k%!l&u
z@TOwU`)F`t;^P&dcrO{nZSM48B>A7Z{77s+q(N&73k&i(Gted^!JTI!EhtEH+Gez<
zu<*d_>}*oK-+CP9{S^a!gE6RcGewU5?#8s9*(@N&e?KFqkRQfNR;}1qp#l;Uz0Y-?
zXPO)Cv_r$v;L=Prs_8j<$!F=uy-N<TFm{*kq7)GDzv_C0_U9InR35M|ui@xG$qxnR
z${i6N&W(T()t;!Kfq^|ZUnUSF^lZ=Z_qZbz182`>60-*GU(j<NnS0y(lk7lrg%`%F
zi7D%``$YTUw*{{@lUkp@3K$sJkM&mQ>g&Hn+(S|6fV~x@HPt&9&%Og4FD-oaYFS%d
z{qFr@{v2fFI4|~a*G4oqXD|n>jj7-yYU%ISWm{z?OfmlJP@-xS>mT@f8T;P~NgulR
zhDYnq@ScB^78fUvO2k%e1E?8gZgLS_2s1F+&pwCGyeDO|LdQjjY%H_9e*MHDu{Kp$
zwunIb9PX3bMW1lUaxrpDVG<6&OlYoU1rY%?wuq^N!)dqUP}}8Wvz|YBde{a#9{CeK
zfZe=RZ8@`7Sy@;bXd4Be#2n4t*>}jo;?mX2zE;4+URlAkcEN9LsUrz)U0P9%yMy}H
zw)V?>yu3??BfoxkZUg&V#S?X5klSY9HasD$IIG}QY`fe~<p<D?1TujKXKoxgR}eN{
z78lp$KpVxe{Nvc6L+m(mEr6Ffh&&9%8B20c>A0(!PYYMNAE}0hMn~OXwRm%H@73%M
z9Ov1fO#W24PPQ`POx4D*_%6E5)JU-W_U+b2EDe(bdk>WFjIAyB<96>SQvL1D+12->
zqm%RU^Iu+|riF;}8*8kJGj!w5E-uptzfh;PjLShfJRC|%ElC6%nsX)2o;`{QBnMpa
z08CZ4C(i&TXz1#S!|hpPG8>o0P{?}vbo)4L@E%kY3T(}V70>rU(>s~|^G9JvTifO-
z)eP9p{Ld0PBP{%(O2hi&otvHgchT5bf?W=GD*I`!CS)VUMmWJP+|@#vnVOn;OU<S6
z?7Edz0L{|K;Nadpn(01aM)i(;ycgK^mz;8;qQLe202xi803Y9BAI35f(W;v@H5*b=
zlAUIM>yjx+Nf1r45x6m(yALCq16zT>Ix22Kec+Dt(cWF_PLn&UzIh_yq(Nqtc^)2`
zAftiRG7EIw!0lIs5O?lVYwMtcsVU3f%y+!|dDuuLnMU*Fvre3h|F92r{8%yA_Hjg7
zFq#>Wk`Q3ko7Cg`;(CSU5H|}mu|9tr8`JvlsdjkXR#HNOEM$J4Af4+)E!5h9xJe7Z
zE!B50B77+0O)F<bj_r4rFjP}eXcWQ%%@kimyS)J^%N8nzLPFR>9?!RP^6pK?Bcb?4
z*OU`&o=RO6yT4@(F(nP3vy{CdL0-2|oNq_MLN20`-gSKV5Or#g&=p8W!%_RMa>iF(
zrj1wqcMyDa*BRld=)izkE?xo6ca@K|=DIOTW+EPH46>2~M)Ew0S9p#%d;L44czu#<
z=Gd_yuRt-!q1}<x%n@i{_YNyTZEI;|<r5JZ>5G4}<hGvQ5oe;7kW5QWO#J&E$jWb|
zD_8)eQ?y;eTV_AkDFomiKOW1SqpRTp0-E#+kEm~V&VRmjJ1busd36dHXs!{bP{^iv
zwXLLfY0Gj%T#V-oj*ixhc>LHfzc}1u4b+&=vWfuGWZr-M>U+h62MTZjfuf6|SFehL
zW-GtdeQy_(z~B|u*4IT*k@sv#UW0%fYtJb0@#BI(xW~1OFgd;%bPt|rkx_mTnDm%$
zr=(yBW9hS^qDWLHGaILYxRb@M>x%*G3JRG&Yca`!UMz>~*@3xzZ-c@T6C)#EVruG%
zWfZDfYVPyez-rz#Hx=;+_`V2S9jX0%^_G|9+P|fa-+LFA6kvi1Phtjg3$)h37;?FM
zIaUtfl|bSK;GUBYFS|lYnE3fK9@451KY#zSlVidEb)}qh=XuxEB;;UkPYSW8Foqcw
zYO0F4Ay#Io0Lk5&9Jb8RCtO&Or)u2I5b<GupR~lOM#e*zyXX4IDFOgU^7g3hfvOJc
zUB&I7L3R+!$UcMP$8~FMkD!0vJ*<RNkNfOdZzmU*5@or7?PvbUw;|NZ%nZIc^YQ2s
zu(R8hGEQ&9_6S~YP2j>*q#4%k6KuwS&y<JOHgy1^1|vg$i=Av{f-Gz!Fmk;iqrO-Y
z0KANls7kJ--iPJ|#KJznYK>i7|2a5Z%K4o3EI&``;8P{U`j9TnXv(Z3=BOLaG$n_-
z=~7!qy5i%dTp_)HKO*iMnLG<kL+E>z-h`f4XZFzx94%9AWk@FTA`O0!6gTTRIdk>(
zsc2@*J*OeAZMn2pcMOJyP+*4h1qB6*P&tG<=GNCU<J=lbZ*H0MMNXTw=?)1*gq=e@
z@77#9(lE6KrWK3E6GrrlaPfyM%*_i64JtGWHc>)OLiC@C$PIf)7T#PZbiRqmGw#ly
z6*jf7xcPoy;H>_GO*+g-Fs)kK*(uzi!j`AZv9c0$c^h^L`#GuD0dn0+&<yJk9m+*0
zDGk20Vme4{h8XOupMg2q9LL_~+I#%px-{-c1IlI<F~^2ZG5FW)Y}ov~a|M38SAk&_
z?Yrr7EA*|Apk-2QfRRKXmoJJtp9Oe=rLbQ1s|Y9;X_%R*AtY_|cwj&LGW__rxzU1r
z9>1jTsL_>yJ9Wcoj^$sKl}(->{HpBd=QL8Yhx5mHnSe<(-7*K~kI=j8D0&eF;!dxJ
z@7$X2FL&pK_}TW|rt?DURTWj0P9Vm1&#{mdZ9&lKrNUk?gK?kzT*ifl;o%cMI?~hg
zpFMlP+Iq6`*|Xgmk0qzarY0uD+zwv&Cu%%0f=S8K9N7<CRvl@lwGA-SI{*4|x&r5*
z8fyGt<22hj>xG8ogoI9nlayGj{uw@q+AMP%Ld2K{4<vlMo9_uePKtrE!3C9H;QPYb
z>Zix7G`orO{3Xe_H8~qE<y#O41OpZUAycoSp;3Ow<L2*~nUM6r?E(LniM_oi)shPv
zDJrQLZ6^Mlsc;-7H8A+<FYY+xT$Go0w4|fAtBaJIJI?}Qm;g9g<VTetLYBj*Fbdqs
z*4)BE`4z1_hD-oD(U_Lo`U;IMLq$njDlOX6JIdqBR;dL){Cg-q^wCo7e`F<T_Wi=8
zOIIYVFy`s#>2WXD;QKWIQDbavYdZn&z4Dct*rjAViK_L`noOdKzxIEBw1YN2E{KP#
zVy;_L=EQ<(E3hPBdTK<({!`d!eGyM9d2d4l;l1S|G4qfPW8m(w%<k!8>oNJ}Cxo3D
zdUA4)d-D43<N)2Qth%zEJSha)@}d?pEWTsI^%1I1Nh8c8F^9fK?A)>B#KbE6yXSWz
zaVV64L+AWGrLV~;u|j-D-fV7c1htG$L;lgge*U~~3lX?Rdzw=`U=J`kqTu7l)EgbC
zcRqHs9WNId-P+o6#*Onsdv#4&IZ{%eJ8sX*pu02}u$-qlGYtUI<+gwPc)|8LG%5j-
z>sbapS&_*>DqV#|d33%zMMl*W1OmW6N8gp%1r?^TCjkZ(_9PgZHIUJyVRLlMuELCg
z0+(YDzomKtJu=07xo&3VMNLgg{Oxb-`(0ZwW@-VMi9I!N+3&g6D``0FAMP%Cu$Y>f
z+NQXmye*|hCqS$z?i(9xZh*q0dafag#Sa7REr7_!V^tUr?(2sL53{k}c{aB1T`0~&
zu>Z<T`Kx2N^7E?!g<HYj{!p|H=ZnLImy3W80JPSvjJHG!z>4?!@2v;4{f;wN{*=Ek
zcw~pUSV@V3=82i36-E5%(+ce$+Z7)&03AUL(;y<MYDK&kEh|9hM_Yes!PJNeFQlfX
zW&jjQGGJ7H5x$%`Vy7PW{rg=#e*R#SfCX}kyXEC7zpq<bHb5#VwY>aym@p#kv-jpF
zSOzgj<f2FW-Ct&>1BQZ&pP18uT#kgdR;48092z|P+aLMJC@Yfd>Fez`u;)!tr`Q`#
z`u6d4{4!v)7hIK<{+)>-x7;;Q=gtf1j1Cer&Ol;1?I&$ARSt}7>|18zAHD;tGhqDs
z@F|4?($Dx+G7KGdgY0M)_h7aiBx&B<TiL$P!QUDhl7bZP+#eY1595fsRGZj1)&PIa
zG{N~V#7%Y`kI%AyNiK&ngC+jMG7czX7}K77btbo=p1Zs&!1}QZ#u54TdMK#n?ClFi
z;4qp3h6Y8}&DiRbYEMZIMA|{7`xWkSFDSKA`0Wi~ECoC~O&a{o73U-?C1oT7Rikaw
zksw2GB&cD_L>;8G_hPm|NmwuYLE`qQa2n0AVDJDr;PxA-c}<)$p}y2lGc(_8L45xP
z$l9ML8@WiPdJD&Jnzgl!lw5Ate6iz}4-D+@q18Lfi!3|igoi?iFLGXAyzUbixYLVo
zdB@#<%h>pX<;zcKes18Oxu@EB%V7BAP;4nKG6$Zyd9%`U!LLnnC;SoKpPzYWj52ea
z5$YclpU>`YZiPR7{w%7!`BS5SM)>&>ANL1#-ZM9q*J^#2(L}bU>jhx&QuIM~kMW5U
zsmWx9e;UUasKI;bmf0pwVip|PbPon=ipB=?5=njp{|`ILV<EAj`5A;a$HlGIynL1S
zJ<?vOfHUkZ<i5kO-v=Nx(4H^CEMY9*bSo$ykhg$koJ?x)DO4~O`HSq`02VC~_=iuT
zx{{;A)rA{a_>nxcNYYTYcj5+qZ`w275^2hyRK)z_7;9*etTqOn0T;+0VVzyedQ^P1
zR@wFe2H@Rq={`;@cr{oG{{|o=g$->gK{^|m@IQi=PjB9^>Z3!>Ua2(X=iuPzgw0O@
zZVP_Uini1h=7JJ)3D`gPq!0-DlwJA0pGa$qlRy-1*C2p{8fz;nI>@GKtE%)EQ8$|&
ziKio!SLi!x?4L$6xrgW6ZCBNmTkhm)rd=2)ir90n!shN*aAB@Ca6mjB(>1~I-i>uK
zp=fj>m4+!<?Z)jlJ`&5?qn(|y7J>}tk@m#wf-H%cwKcB|q`{iv&rHKOya)u01#v_7
zjz^O1>_5&r`)OE$8U6t@<T=IerGycwE}NZyQ=gxPFa4}Cl5v@JI<19TOuWu{z++GB
ztZ#4LQiwcYBr*TDPYES3_g~`IJm8_gfEIB~gnIeatstT1LRvRw09T+`K1|W8BY$A&
z))uts;-C?&75Y9$6oEH)x9#J5QE<OK^-Rm$$AQxPC;GHFg1<#F2ROg55ETCU@KJ|m
zaoXa4u<18oX)D?$f3OU}S~)2!+_i;?`AtWf)J+`sOuZ$IyDv15sGK$_FOc9A`ywGK
zdNrHM5a9~gP;M)KR|-2mws!qpO7UfCEY|(DzgY3!^NaueM^1O=`XM?xx|oE7SL+`H
z>Q&q?>Ykk5@rh!gVd3Mmm(|xlrwR5X>}moAcvf+7Pa=M2`)9OX#$F&9xm8djyW8V#
z1OO(<fAXs7=hE1Rc&<yv#((~N|L)p^lw!#Uvl7ss_F6A)!7GDCLB(7+#RSqsaw2F@
zr6zxSb-y^u5?Z68?&0Cuk?!uxp3odSO;QG(o@v^)@nc-}a32KN#m>UYYIf7!p3mmJ
zgTi4LI!)jFTenflv9g(Ys9)^<Rm|<w{%&$FF#QeicEmh?{wpjbgw#Z<bobxYRRveq
zmm12-EgSun5=NLs%;LK<m7gI)GIfeeQ}VVk-5;;{`Gd9D*<r#7y5Su^6+pNfsjAM*
z;C}WgDJ!e_skZ{vgUeHD1xk23pnv>7CnrCX%UeAlqGwC2tZwf6@11EQ4Q0eQvD3py
zgyP{#$cDiG53ckuMv>Ehr8kW+k)b=U0#8Z|A~2#qc<{~q>MDKyXpx#4%)7ohxw*c3
zODVvOsi#YyAhPJi#WZ`wzmJ>EH-)<)Fk}nUkp&4sj8O)Z)Nza#J7`^*VW%A+N<XeL
z+)?rKlf>n@1TXCm7qrEZ&=YTA8T32<U5XBXrl&K=>_M{(_hI~Q5{8I_eFb_qFDNQr
z@88HkNM?Kb&%VHgx8aeB<m*iSTU$LCGR%G?Y3pg3CCvXSqL&`7tF=N~X?~)bn8Jk3
zd<UaDS8SgZSt!Zx_W$YvFllRTTzINog3!J%Ik2SRoH;EZKwsqq9tJ**1qfr6fdWN1
z{KFwC1Rdix5nwX2vdU2gRQT8bc@A~EFD9NDjLT9;)?+?)^yuI51+S9!&9FU(iV<^T
zwGd23x&6`!qLARZ`W@rOrlzNIkAHz=KWb3vY!*0m?>Kdvs+ts%m;U;{ke=gHK6z5f
zMM#KVjDoTcs{I27%nM$U<u=`%8Ye=>Hq#tqjuAQq2N$`iz_SaYdp3AGI|e&N_m?kq
zf2IFQ5n)qINKD|5=8k>(dy<#;4x&s;193a3nWcOtwDk3Pde*NQ8d8CNTtH2l!P)e+
z1o8;J^XJdsRp;E9ltAK4^$P5?j)eK>y*-d&yg?;%`fM$dSyvPtP2FrMOuz3@l<m*^
zC+5H5(84~tBe{Z)Gla$?Fx>1xPx{FK2sJ3F8n3L*jhd*k(1dOC5q(CWbo88B+dNGf
zRz`$aF+_)ht^F<KyYg=L`4uc6T}KU#_qksyvi3cHfKx}yi+kqR?-LK8O@zf#cob#O
zfJ&JL9C#$i1y#gxYNFk;@YXx3jBJNyOS@l&Vvka>we7Y)I@nx#?+z-%A#h{a<_%Em
zYlrFSJ5KTQ=itZ>;W(ncw7MGE_w;EE{@0Jorw*f8{a3sU8N@!O3)0!yIZro%J-$|d
zD8j9$p}{{3<V%!_f_=EX{kP^$(g#b@VA;-lw{O2}fB(MN(az2wH6bC}@aom@ofz$P
zo4ZHJ2=m#|O+NCOm;A#0U0`>^0o~C(TH}4lSljTiWP}DaY}=yf`KH~@ZE5zIczRa&
z+t_fQ;^A32e)MRQo2O@Kq+ucGfXvsgA6CF<^&Wk%B4CQ;F@~?i;&|sy7Wk=kNb;&-
z;gn91&m6+|0~-xPHkoXDdhHJwuSoD2{h<VOxOOec9lG<|z@pRupm1-073M9gs8I0o
zk~n$x>_7|NyXoy)Hoa@t#0pRp5w>`4xT$GUbEh0W*xkEA91&aR0CBfFOeiw|ekQ?}
zn<b{h9<pEJ+&QK$rZb$!Xw1!!01Ag`e7WQA!B_CoMvtlX4)yg>!W5E#@oN)m2xTWt
z&HIlT|3<`qs2swr=TB6<7c5g!v;Tt)HqjQ+Jy>AwFxJ!7Hcw4X<|1a=u2)}KL>mU;
zaS8{64wmpMF-b|o@26Rx_Z5&JouKH~@#F4~9z7a9%2r8_{xTj?xK*T{#Gvu5vvF`Z
z0r&jQUnf4!|40c_9<FYFEAZn!!o}P3R}YgT4HJ_iPIIXbK)Tx#+ma$=?nSqGxVR%b
zW*;G=CSYrD3x~iPv~?aZ03Y#4x3+PxJFj#kbLB88Hw$f>>6gT9xF+clr&RC1HlO+3
z*=P}yP6iIRh4-Jc1QFfbfvrU8CoN;7ppZ3#HhN9SqB#f`2(#Z5CUZFNb?XCaA#gBe
zj&2-Ct_#?YsZ{Wo3)9YFi$jM9K|N5lHG{RN+OHFC2l)n~7Qp*F!*xvq!JSr8QX*=|
zFTNG8DIH9#=`d77UPClAgT3~Gc6Ape<vzd~%NR+1BRf>K|DGmof`y5RNnyuO)fNUU
zmfLM=P(A?MjsdJWgVFtu_eM|N!;OdtH+uSH&oU(<TU=B`-F*K9?<MUkJ$>SW#Xwub
zAO{dThE#%b@U7eB|6$zfFE&$XhU!xsyJraE&?M$UZzoXgST(lTt3RRLmclk5<6sjq
z`p>AAhl696jK~~_)K1u{aYWD@KFk46cMlhB)C%UA5G>8Wzjg53LA%y~&gdD4f&cl|
za0;mHk8Hd+T6L4_K}t$+L`3uX{{bB!&~C&yKn@yp18vL-4nvmv%G8kC_ZJ8Fmqk~s
zhX1RPyx%+VVTy&83|-PCtm@x_4ucfp;uURF6pap*FR74BfYAUltEVuDkb*5a127!G
z4M}xg9(QwRXYmda#fV~D7Mk6iSmPM+1p~MJpdU%E1BtL@da!0ydK726r1#<l10|&&
z+NP%8&Jk`k3Q}>1M!X*yVtiRz3f%J(6%v6bCsBECi%?_I)W)aZc`K_~#C4i|e2Te!
z|AELx=tCt-UcP*I3A=)8cKvgAHyQd9?R9VO^JzbSY;-ug-mZg(WC4*e%S0Es9^bW}
z-%jkMCcGjx-{+;;U(f#66oGDHmFx581P|6T$*~_Som0-15}TOro_D+Iwm0lJdVt$k
zHjFJa%=~)ZD=t`B1qu3DlzpbNXU{eU95oeWXfPEFQsEfHS-A;vg@rZ>(>!sNLD~T^
z3;MT>&CQeg(`#FbXJNfIv#=<&f{jB5DHX%Vj~$C}7$S^Nj2)q)!<^L(!29)6r!Ce0
z>xZctK{Yo3hX9WS9Ysj&Co0YZ>UZzGc{3tyqNdhX4_WxcR%Th*ku;=kNHA(7UtpU~
zN=~jn|B9USO|?q_@PnVBMOHw6$%14wqWzrbm@Y9ZXvSx%JQWqM@S?g=!bb=~4|NLk
z-JA7tn8wSwVpZlQepD=dnsg0rx5^5!36zEC1ZD&RK3^i(6MvYgciuxl7KtAu3Wvs=
z-*L*I)JjT89W9lT6zgzvcQ`stn!O_iK%I<B^Ra1V`@^!TtlHDRS5m9<8~w|H*jWPy
zd-IO+FffHtoPzZRCJH}~gO}t_-e<_KGx1k6K?)^#!xnpP5C#?1S0~>!H&gH%RyNEp
z{T=4}w$xoXTA9gHpr{4F;L4T5u`T24-abAfXWeHGTOdDjE+Co<LG%3>s{)aAZi>Xr
zGn0^(v!yaaDUp#=e(h)S)RgSG)IU0ghIg~H^QwqALE*D!x8jrI<EiIYNA--gSao93
zHS_XbT(x&}q?=wa@5;WIk@wX7<Cibx*DLqVu*HzFxx9bhDi3MdDw=!o0^O4DuKxbD
z$KyX7Zo7g0WD17vSRZ=!@TX5tqCpdSQBXicng5K`-E7fd9Hnk7_N31aBulz>O*ZFJ
z;lSV^1zP!I`+rK*&5f-p7u`ALb^@jOPukc-&6^gkoke(qCpZgP<}&Z>lYixXqjW%!
zk!0en^fmQIVlR9l2s(~7haPv4;?z>&+Q4fYc0!cli;N8C%#u8B9l$CH5=P$2w$X^(
z4aJ4AL;Q=eJ_7z<L+aZc9ua{p8gk)NBc6&vAmk?|BnpU#iT(yAPwT^kMa;_0{Q-jE
zr7q+Vlw1P?R^>L`wX-4q>q{Z|nVU;?@WKQ`dw#~2z9ga*ixxt?aTK85G?~ByfGtAh
zGreHmQGiNmqUOfqA0~2{ml{xB8%IZF277yt{654^P;ESa`%>POO7A633-Hx;faxuO
zx_-{6p(NtTUj|$IOK)gSzy;;SY(AEe+2?){9)Bi0RYf0Y;e902s7^dcdGrXwf-rl-
zXy6hWYXj+b@~&_LKl!m`(e%lYXLh)h%K>w;4du$5GMTXUZDzf_1k(nyjpatHoAdTC
zR?L_9RiK-w4A?p+b@uG~EDl&=c;iV=rV-;Olz_1g2<X3>ng`i@;URkS1GjG9b`1nV
z0Q57dGcet_@8b|7M5X$$8Hjg06ul!f<jV+k>gXsn(2K@GXFvdH71p1hyhn>C2z^^D
zP4v67w<1j4^zft^L!{)}+_*sx7l+$fY3Y&Q3Kzd#%)G?XAQL3@#o2xpfTA@{ZT7;>
zPbS59I93b5jw)OoBhXKY8<z|-(H^~NY-E33P(TG_SCZq6{L`$grE(VfsGg5jrBynG
zhOBDZv>~h%CL{R}eij|v9o7L4?9v<EJQz$GdV58EI*W!Bdfc9Vk5I~507RmT1JAY5
z=$FD{U0vOVPbbcg8^UTws0nsFx!iE(d-5p2rTg{uJFB3aIi%C;q3$zKrCp6*4OV>`
zQjG_x>*=j~08jeqT(xuiCKGT7qCai3Fe5{@u(WjPjF$Ec<n?b+Hiwq~F3U^S*3gxS
ziF<bU^aMPgZve0}&Iz;u;+P*A=-FzrJMz#RpnU6~g<pDDo<s*t-mX3KyJ3c-l{GbG
z0n2j+Av|9Y^8aqH!EQ$m5U<7X<^_{&)}5<y8CRmm$f6Zh!{jqoQd}%i1A4I$58fa{
zU8I!>uxPm-j*HyPB<16F0yr%|WL}|@I$*Bld_O(_3-L<<3?S{g>sg68Nh$xW_o?Xt
z8`iz~S7)i;&_01i3{P&jC4~<2HO4DP^ehgsZf92)7jN&cb+Kp%1Y{?Z@ZAv16olJ_
zc*@(-mm2S-2v%Do<}hdC#E7R!P+}t<fD&eV9x+f9yPd|#rB%>m>;oatD!tq;?n@_N
z?57K8)D)yY?dTkY;|d09`<!-0Y7sB?hDvMSYhXmANJT#%iQ(h??AFZlo%OB2x-B|1
z&50|$X;L=$0JKtKhxd}x@f*H(c6T3UWMJqw7j|T%HOvnBSB(@j37pe8Z~|n&Lo{T1
zT`#IOfe+89ICPkCxwbBIY;_lDX1MCpy-LqzfNSdV(cJ)!FgiC_wY2FSQ1>6&Qd&C-
zl;u#QgS4QXB2PQ-Sy+dCiYj$cJrHR;I1*h90depdlo#l0$afz+z{7g<n7;mY`tKoc
zH!vA5p#LJhnCgj%Ko<^O9O9RW8vS2TmHTiXzJ1}*FU+4{Q(5eRR<9=U>%H9Wc1Rb*
zH9uKikCKEg^8n@Q?>h#syj)oBG?`-zB{5${epfj9%uf$R?dlK@zPSo%{A22N2k2Y(
z!WiaT52ADmYvS~`15}JHW6qDMxix~I)1VIAKHG@s2xo*>t7|dO2$#N`00pH$x3*bL
z(hx`<_}s>p^iI>v%EL1^8i5RchxT6})<T{${!MRh;`EOnbY7VBDq&U4>ZisvK0ol<
zofn*x;1{6XRsaP`z&6u*v+7~8Y0xp5m6javX&pme$D$xr<Gpe{|KjEkLE(wLV*>R<
zrS4;I!>8k(y~AnOn42DOj3}|mK_HmR*jw@_!#g+-Of4;Y5TnaRXEI`TO`f9p*s%am
zfalzsZXa2IfReUP6a6L8xoA{o+xrl%W+li;{Z190g5dW={GN(WkMA0ly7oTw_gxi)
zg6-sDoCFs2pp?m<s;S@OrG;fr!xwIpMbog?&yh3)MMrmxZTZ#>jZaP{z7U%#XP?HD
z{`SUsPhkWT9xG;n#TG<>(3-!r1l?)<U1(rSz}(vbk7RhFn+2^bFqO|l9_34KZxvDR
z5cp{V(`++L`Cn#7YR|(fdhH4SYG^AZ(a>&(DaIV;_t*?J@`=%Qyhv6@C%FpLRm0o?
z=KGjs|0EtxX2ZIf$Hp@%_p7R(4vSJI31Yt}Du9mm)7qXxt>y6S24KuD0S*RTu~MxQ
z%#ToF_|}iq%lN=+6^zdPW2RdvtzQYm01qSp1nviX_9rRK<zHzXcqy?|jiUY<@0%F+
zeX}q%`JQCuCV{{a$t?Y~zMexURHQ;w78JX7`8jXWg2v2{+GII>{=8#TelJo>(gTN(
z-(YSW)PhC{5}A)gIhp_UtGyJA2U^xH@mKE9QA{RW-pKjB1_uqDAD+ID2bag6<emvu
z?2HhU?Q+Mj<-gZ{R(lFSrWy?eb1l?S%(IMCQXF_JgRtZ3&va$Y6k)~e6>L!GtZZb!
zQ^x%&(DOJ^-{vEVXaNQW22<kG3bw#Ch^wbbrm*f=J&NvjHy(U|4aHH6*8ae^7hgHe
z!}{2dk$xab6S21$qyqv-$o#h5eto#cO3lXHg?LdVWwoYpvyE3aH2BBvtoP#J{S}WN
zolWelta7vhpUC&CbK37r#LM1^I<#tB<LB)G5ehnr?q|zUum_+}G<J2lHcAHWBmiAr
z_VM=ir?FdjUh5mo!<B4iXXkr}Py>&7{*q^6sPaDi11#LMz^Ga*1BcWGx<vUjNpE?2
z9iLzI2wx@vqj+ue16TUH%Ga-%Y;El=t^K}jHxUz5Q08XQR?{O3<hKuSVx4Fv8P0px
zzEhzaEiNm&PPw+(!mnSm<~LF6TY~#S!^n8|0PUWp(NSg$f((cUVa0$Hg0^E06Q`Tf
zFTc$!(aEW)-N5M<+|K!TLz0m~Xh7)By7doJR0ct$nqVM5=7xZ_A;rAh9;j@%G*h35
zi;8caK$Zg+8(YZFxw)M`t9V~$2$J<NblpLf=>_AGj6hsX;=?Es=Nv69ciHt7m{oS9
z)&2o>c@HV;>1H_lR0Ly5Lm15G6Q8cy_!%zON%_(tK#!b*^W+P-{I2{#WMA}!988b9
zpuDh0P*g;2WXnSM9Xdrar=WG#U2e9vc4P!W1R{zvO8fOoiHoncLa0&}9~VcwxED5%
z()XsdE6&2++|>$g(iq_40p^E+d5>!(O&Pw}fUIA@40|^wCMFT2+iMid!x!CHk4dii
z0C+!o0wGbdYHGYJ$P2r!p>f|87^*P~P2jCtckfaWhcg{MlTU6B#2hZ)ynK1KWkSc^
z$qDKam%I1wMI&0{@XT?BU5M4LlnLBsMB>n`WnxHTN*=fR%i>#q=jSErG3fO}M(tsG
ziDnlCc_!I~w@2@?@4XWh6^wbUR!Cf&8d0r=DKRk_L-0Z{-+NLC*$pyF`;iD`2HDjK
zPY;j#6hVX0LYxf<7yuWSg3z{LOtSREXLfP7R_PMuEoqVi5Tx~#LI%dp5V4i)?uI^w
zK!E)4(9fU0juW3flLjE3gPlE8nD(a-=08ZCiMFU}fZ2aV?Y@K&DJm{bPSQkGMbapk
z1ewRa&rmQ#Tz49*Sr5PW!l55BL;>S!kN;|)JyXfe$x-49n84u(H_V&<{z_}qM4|^j
z9r#i?S<p9yHcpEhHz;dUQ%kn@(G{fr_*N`D`{Tc1T-8lr57ssZHA&eZi77_P665Ue
zQu+;n-MQ~Sq)*={k$`t*jirxeWu?p5%G_N0APtSPma_7Vag?WeRdsa^gn@@<=S(*V
z@*#YfEF4_^UL2Go$jKVhGBz$`=fATXye8a{#V=pB;{0%_%*-4^!ejeNw2R5H8pC4r
z{>=Vmo=$1$MROxNjAx%*Gd7m4uB?2(#?PCAPX~)1u|RA6I^*eh7h+oClba~^GHx9j
zIHD!ibowh!TI4FdgrRI7GKG(QpPVeiLGmUGUE@zQ0qjpY5!)9G`w_GVH>#_vBkk<$
zO#1rz0)0pP-~DO@m|$*U&QLAA$(JC@B7qcFeWiquOu)9;h^U=Cal`ULd0EBI!<L!j
zEUo`MsgladL<tR^NYXhPm{#O;@!i%Y2?80#kjlcs&&vZ73p;z(U?T4|k(37y3Orps
z0>}xEqdz<eKTH#*OYKY5C~-MQo%<0y`4rGJXlr<^v_dNNG?tpZB&V6P=}#JyY3TsV
z>^^~Tc@lI(oqgRFnq{TWA0mP6Hu{>Ed+)7~67+ZM*^^*uXNx@rRlc2=kkA{FhRrlx
z3CJ+52sF3Yo4<)~-N!xgK;3!`7P1_>?F%XoFP?_+6w0_GzCJ!-g+BjbDh4G^3>S;%
z{LgmjfmC&OfB)1PpH7AKxBWfaiCsDHRN)vW6;2s~j&jp@yyfRoutx4U81D^l$+R`&
z=n)E@ml6CNI3%Uel5RqMbrBjp(;FaSAk@g(`>4Gc`cssTpg{A?#_<K<9>FRcq*(Mg
zM&UgYOmJ80`77vlks@djkj}0Xl3tyhxHK6EPtgO@d<F7|YH#`<Pr@ODgae$It9Xc0
z-Fj(peo6%7Uu9c?-e*asB8-~ATw%F5f*6TipgFyrs$k&BC@PB9(bAgUsdj;mDT4cQ
z&U>9JS6<<1i?tPgU{%M!HE#$}g`$7j?wF$-=VKTc7&uORlvNbQHRRhL$I~_#FN4ve
zyQ2vE_Yofxr1RjMlLMsdU}pB&;`;SOJZgveHQEF`U?AU3ROrTpu#ga*ZcmUsLr}@~
z+de{cU_oDY-QIqv*{XE3tw5H8_9DdPM(9mqz!!O3rz5n2Kt3`tK5oHGN0;*zS)VnK
z|DBVO(JXxNBDD)c;S<A{NQOa+XIE@<agZoqyf{sC?ZZoCK}YDJitG}C+XE=q3?x>z
zm$kIKi(Gy#c)_i959j&T_iKdVp<xn2KJpIS>uPKL^*|s;wCtll#s}pNG<E3@)BZfq
z$?<soI_LGe=8xAm=K0RnL8R)7`}ses2jt%KrKP1r?AHZ+Y?gwLKwx9A2=}|Y2gAVr
z8#o_IQg<UB2t+~1pdv_2y#dC1YjreRj}Q}W#()DnZbP>q{l>XZG$LXZ1$(O9)nN6S
zg=NMa1up~m)a&Z%o#k&(Q<EIKM}ZoeQJCM#^lixk+)&kJ2Zx(cmW!H75fNL221>$;
z$(~;sq+R)ig@4XD4oUpm+$$Dg1}7@px25$L-w(Jl%^K{FFcSq*5`E~<p=h-6Z4VIw
z*75OU&*XyF3gi;-hLd7XG+34H%$tK|k;OBZARCfJ=YlQLjpQ2LTKk13#*9pX9|H!L
zHE5%HD0Ujmv%zv;1bhKg;#EwnwH^ToNkkC_df$CkyNN!Me;-4H6dpgsu#MUni)Q^c
z`sniz5l<4A5obKNgN>QcBLSLA;A}?jvBDC?7Q35+?I2QtZC|>&cEPv?O!mV$JUkN4
zFiGr%$1h$ua1n~Z(?`4*1>8wI@NVEt#LL#1VHE!bdsHGO_Jg=3zS8MMUlG0o86L3J
z)yO4j>t;&u7WlRyXgsbs2?#t6lno-fU$CQ}`#`u9xa^9ByBEFU3%7H_iEWd0t2Iz-
z660KcjvCD-pElYHwaGXbuB`{^R`e<A&9%|?h<*z~m@i%s2B-rV8WsgnjAbk$E?SkF
z^YjBy3wu>OF4h5oqmD2N0q9ZC(V_pBa%X?k1yUSJfR348zw1Tz$c?k-&iz2^J_1k8
zciaLu?NbQ{_2xoTb{mj192q%d(`NntCo?w-*jN*QL(G0A6`Zh|p~IR508_%l5-2z_
z4XMpEZo}vQ0XSUy3Nu<CKD^FKFHo0|r7iW)K>rlld?HQ``O&=2=svXBct!`zNul}q
zJYZMg512-*p(vJJCc2+0Wba4?dVZ=EfF>>jZs73P*w1e;i>$V8zZ+jLg)<tP_`LFz
z%bB*1A7{e>PzVLf3MTN$o+8i2Qf}qt*%hm$^&;@b=e6J7vlA1`5_9WIYJSpzKes5U
zpP^y%K{6`DN=(Fcj=HazsJS<M(XoTLDw(2_rQK(GX3)8>;IlAc30P3A{$q=<*~(;H
zlOVE#kaKS0nTA^k9+SgB+C1qxS>g1xJYJo%&4QE!DDSs<FGF+rxAz|O_df<!PAerT
z*>JPmxBg*-71g}AJu-2Nj#s~FOLTN_c(DaF87U+NL=(auF=VG|8>qD%!;`5Z%$&s4
z84o1K#ksQMIi)R-FiYI{#3(2(u1UXpm*}9=xW;xb)%}y<jbk4jVf8}?>65#Q+UOhD
zk_FX8*Q@sY;zmkIdkHLLgmD6kQ#N%JKp;%kWgV@pUD46e^e~L>Ma0wQ&JsUOL(C8d
z7uP5RgP&3miH>Ko9fz~*GG+H?eKJ{*a}(RiXx>0bt>B2>K$2_-WP0X&bZQXWPW`SE
zsWe3*!2PyKk90J85ANPouXyz6_$RyxC=|Wkk&!h14`s?U=A)XHNXj7?1FdR@em`j&
zX{@X=GR#>1yVB<d%Up45cTN|D{+f<F#C0F>BT2<Ry*#=1KYfZsuh9(jFsD&a?;K5|
zCRiU>E`~v_Za#Z<N=02g?B74XH(Ul4g^Xvc?v}~2e4|d|3Ip9Y6rZbMRPpN7J8YrQ
ziVAUrVdU=H6>=;g<Ty^q&%>1q1liCX|EgaXlAi{O1l%UfD{H>6{?}Q<9ua6l+hAZ9
zO1LLv&M@I_Ej31w_<-qEq4aBSD>7NGhdR5!rGiHJ|Nry<{U!Kr@?nQ$>K>IS?WXTG
zo>6<!O;Oj)(#-ApISbe8_zyu?Q25MgL7~%vXRio}oD&s3Cn~`!D0og#(B5kC+5h7Q
z9Gom|ti1o<f8fM<)x-D!apG^l6m`zR$=Tc0=7zN!LE(S>iA4z0IE$R6llgU)%icHL
zt~=ahx#(!&<m&9?iVtHcWWVEpw_y2SZ((`;riH7Gvzv{RBjNx2DQ#tWmQ#2a3GU+8
zGb?y4>;HPKgPEhdm6?T`yX$q=|Lw&gG2vZ!@v;B=#ZH#j?f<t|xgEFf#H$Dw6*Vpt
I%A1D#AKQeeCIA2c

diff --git a/docs_old/logos/datadog.png b/docs_old/logos/datadog.png
deleted file mode 100644
index dc758f42c58eecf3dfe904fff6045a16bfa4d194..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 32779
zcmeFYWmH>1*ESq1c+uio+@ZKzOVMJ*Em*Js#fp2O#VPJm+=CYn#ob*C#oZlV+TQoG
zzCYiOcdh5=d$N*4W)3qaGkf-2*R}TvQF;Fk6O9B7003ah$-Yqo0O0)q0Jta=;B(2K
zld;Y7iON(~O$h+-pa%f_J_7)^&qaRQ0Duc80I&-J0EChN0K$)HjjAHg6+jb(cW<80
z&&Q2)lkf8h*;!8ME%G)z86MXQE4j9~=OSz8x7yBX#;#N!9qqwZ5OXSL_mAdO#vea&
z0nq+@qEY=lxOvcc1bGBs8*_4U3Ov9699*27++5F}ub=<s|69Vz^{4*-bMWx<fuCRe
ze5QQ7e~Qrxj6#zE0G}ecH<BOR=J%Ie)ATpoubxI+GLz?)C+R#i@`%E1G5CP4c!Zow
zn6&;5a;LcrZM4cU3{xTM`#;7i;pBKs7fBDh`S=XT-^8lSSZgg<Y%Z>>P+EMdEeWQr
z8JE>Rnz5;EnrE^YPUb#KZ_=-4)2mYi#C#kU+)8u$!p@CArG@I-##@~UH|84Q^jx|~
z%_QrQlbrr7_H**X;Nau*V1KK9+Mk*iqA`vzad{yct(!JWa30=!fp&zffB&*wRFDyq
zi-H~e&~{Vq@TYBIGfeZA)w`d6x;c#R`gdU(vc&V(=W4joKjp*j0Ovo~I#B@9|CIjk
zRAc%5ZFBfPU7i89|8^+?5c=Eq|7!LB`#t~n)AzqR0slX<YbQJG>hCm6-Y6!vE{usV
zJx5#Al?%>NlA-QCo-I@0(A)6r75sZ6#iFB1O~M$oR<BGrX?~+M<cffX3V8^Kut*rH
z9fG^j=`o-U8l&N3%<AZzKvqmiUktZrqy1xB4G0ItMOXMSUJWvxy4fU9k`Uc6YZL-{
z43!kBBtZB(MMkVR29#RRr2iCxOO}VEs0vNZk~a~)<d9a|rqibm>CCtzm#x2MxzO`|
zjped22hT)3<FavnRDZujNq(A9UI{!6`i^6X&m--ih`;WC0@EIK3p_zgc~uODCPBf7
zf);%0XWVh9qL-5~cey~r^kq!{Fe4f{)aP^B-+{5?-x1rF_VxTSh1to8jg~OdB|##s
zVk?!cpV&QXU8wP$H%QZDetLRh!n*LKB|`VB2%H-K2)GoMU+i!J<|}wqAmIhHg-zw|
zVGg~9&&kxff#bT*qxN^%)*~cRJa&z$5G8&F$yMgZp{vT7$Q`D%DUn)t&T7kcZ^-qS
zhBH)#6al-bbq6QIOXtEjOI?(Q%-FmIxo7KKrvZxbnZULv8ON8PG6^<lR}}QGA2@uu
z6IJja;yhhiRLN6@GpTZ;79K$Wfv@C|>&|VqiovCq+ll*7qf78T*muhK_qS*ry3tB+
z7IKjAZYFtSZp2y+FJ29Ec6U_szviMZ8pL<27<{fD=7|*8ii+lP33BN>bUAHyM*B~4
z6{C}2tF-7zpLJ5-n8?6T%nYjO1KXo5P>iXJrA484G`$Du{hiTVZ%WQAg8wy`b)u|(
z2Z{IQj{7Mgg&Xq~H$IM*dh%V)5VTT;><c6uw4bx`)U7SaQaEK+Ccz?8qJlme&Xk$=
zBYTqXx>iYH3F>B{$02rl6)44`6PD-n*O)Y-Y%IKKj6NtUMZZ*2$CTtHFol!qsBZKP
zH|IU-rg?q^f!2d`F-RmZC1~<JsvKM|^o-8x%Dld(JNHnG&zp@so8%uiYWL*9;i@+f
zZlWZWEv8)>SOZr|Q3RJ$Nz~L$6dubp41)iL3R;&aOWF<Y-j(AGo&rq*j2BofT1#AW
z=ggj4_-i;EiTf*&O{oQ1_(`2yHBor)@P+noLiTKs$7iLAr>SZ|))7^jrITQ=ZT`4G
zJ5iv+oIJq?DGDU0JQF@$d?qH(_0gxdDvNfU9dF3BCu-k4b+(=_rZyAR{JY1Q$JKDp
z(zZ4t`OndeK;w4CN#<e7r?@#p@Oh-s(9e_L2;X_0Z!r<5qpxy^E#YkVa$Z`>&?eO-
z3#j%nNw<}po`J05$PI1*?^fDv6M}vCrRRmpq1KJb)uKgYGJ5mSxiCrLznjdeX@v(B
zQ~bQm_jqe1&9FkGsZQBI9io{fy)+C(KLJ%vRm&J6B6VVWtuS*s)z8+LXtL5{O5M=o
zy7lC$g`TF}bX9bSlDiBg*aoLWV&v7{E0${!OaV=`5}kOF7c_7YINtvmGvPSz>q*aW
zo@~X6DS|!gm=qJ!@M+q~a1-uARP-T)E*IM@hp(vp{n_E%L^hJP(bEANl7fM0RWsl2
zr}b4)J>xeyB&RpDr*IB8a>W52Fbc{04`A{5pX&K)T7tTRI<l;sIs_}P+%nAy0!gsU
zY(RA3ThX<sXIF54eaH(L%9K+Cg_MO@Y*bn++`u3&mG9A$2|2n_;kE&MEH8VAYBsVW
za@%;7roiAb`dagWZ)2e*=pX3SC9u&XuxV-ZDn3zwg4rFuj+71+Kj{=0U%i_n%!M+l
z8axdttf)VJG4)Nd>D^e%WcO5WZoKF@NU7Q+=Dx%HSh@NT6A4AA31p_jz*9+;a~BW%
z>wQX05-HE(RSJiuHc@!Fva#PR$EcM2C{h}C3+RShhYycZU4J=I2Z)e}&6de3I-N>V
zg2G07<C%fJ+@3{ASN)oEwg|8*%sa*Lh5PM@k#r8f5{pGs7TXCfA}_OP?~z65@nYNy
zj}@2u^fs|l&5?rK5$bZ8lw?dh5h_nXxNgqUi|M~eLn~<4I@TlH)#QX%GUfB!n=a=x
zvkM8#d|fCB_mupkU%*^1LDM&uK|4wjk>i6;h@Go)9pzEuc>G#FDJ6_-Bc&=AA))pt
zl~1HFG(UTB?;^h`CyOO_n;Pf5+@~;smoMI6S}FMP4Xz-kdD5Y^qNfSFX+u2}kMM61
zyOI6t;Va;kcqhide0|}3D(T5V<QJSCum%2KSP-^Z7x$98uA#6BMy#j3gQ#ZZ*JYji
z*VsSqQzi)UgP$I{gA(nm_77SqtHt5&=n7d6jt&@zYKZag-Gu&Fx7HF46tNCfv7=|r
zTyo?EXmDcG)ylF=r_RV>U8UDYstRs3vL<fvsVnGcZ)L0tH^{CNJ!^&Ks)xJo;m#tn
zfqt{&mpE#-5y?Fxy8oSxFQ&^ie~RVtS^uA{v4J@u>80eJWS%wJn_l^tX@$!gVg<#z
z598=DE_>MhemXzsQcmz#8<qvU*55~IsPu(*N#Xfj4T_x%AL!1Pd#*)PM%!H1>RjW`
zDz{(iq12Q{f1)<AC1D<3PxH(>uB03$|3Eos9ZGOFol`n+ETL3IPM{{_{}QO|CmOoX
zq?isAgm%QaD7R^t5JZz+$W=(cO;VCTqfIJR3)?h5sMzPyL-f@DJmPFM9@><cSvyfv
z*IHe-I6qoHkKzA5?8X`Gc(E?=?x*Py+lb)Fu+g`u%#KOFeV4lvm1z{+3c-^*=%O;m
zmZ7RW!`~@$Cg-k$%*q(wwS26I3!;)^ngy<U!hAu$qSW`5-~M>7dDW6iV0Z#@dVPGB
z<`A`-iR~MF4L2%YT&Ab9r_ZcyI#<Uwa%-~#tiv#n>Fl4$2yTmmjQ98ky%KNr9249r
zFLzowJ5u+1A5O7~h}UztU&YVXR>P6oK>11xS+|l>PYFG#JNM?jFUQBjm)&}Rbm6aq
zZBG#?!Ztny#7lRkZt1}VZKZE*5oRke4!<1h#pyXrd=EpeYgzUcb%^)FRk@x1!9e-L
zbq#OHY{PPR{)$qBx0E*Qfg}kTgXlgFTHxQ)tNSVyNfZ_<!N!SD3iYhFHD9f#zhnJu
zzx3mN{Bdcnis%DRDS8qZ+~eal@9J|Kly|gf{IS_#r)$Z{g7q3)<3_g4fqik89a0ad
z(LHAe{q;P>DY+S}Umbp0u4vVjmtCNE609@OA12P>@v}91Jt!Dl<y1Um1eqvb(}$TF
zUBRR;f*O@2@88K495%uC>M(&pc-t>JBdCA#$!~}guyR*(&<x`WA-Dm48i1J<ZYNmp
zEmQTX+j~T`u~dOJNGtvyTTD)ShZH!L4^9$;UujcD^YfH&IM;#~8cq4xTT9P!xDOk%
z@#QDz*sD~1BI-(4uZZZx1Q-9>^zI8ke~MRoEDq)?D`L_UZVQ+c)GW^TftEY!{2ce7
zqr>I*4W3SA$&s(!z>n(HVNM0hH>PyAtldWFk@dFBxQ9~-V5e*?{|5&7&83>_@e3y~
zs{q+L>UZzRX#6*~B;0pe#4B}^jS^1&Q^B@_njM9#<`$Dglf8;rhIpi~gp?ES{$GUr
zWsR2HOAVsdt`S9D(WACN+zK>FJlB^1xN<$OzLXO1(#cA2dQy=2Ps^cQO*%lzcUQQ-
z+nLr9YP6vu=`VAJ!t|iF7tj?rariq_I)V#RQCHR{P~$s(f`Jez0eDq-&BY)4<I5#r
z*6W01-EU&l``0bj$csV}<z}-aP-XSvmNM|$an$Hbs2I<VO;;q>60d2Ko>^l#{pJjt
zLdHSnq$i}c4!pV-?8As-MoWK6<Q{9yWzCG9cafxBz|+QDpmtO}a5n*&TNlM+Wy*1O
z+6Yxw6?#a)+Y<MFI2~{m4Cby}Tvtm&7ScEQxqX4C*Z!vD#_i)jgGptY35)0d5Hs2u
z0n|}Lg%O3XznmfusGYIv)2y1sbNP*Y8)EP}{bJ12-t(~0LM(2%>beS3Yxu|6i|Sdc
z?JK%;Nc8@&c0KD=k>w~mg!lNFsS12Pz91zq80B(n(93KqMZt#yFH{KI>pV1Ow>jj>
zmqH6%KT$eGT^E(<>kM4EABYhf6OXHeH~b(_=dr?=ZQ!ExK`mXHy0G&)9rrJJwsslo
zPrz+>72`NvxWFvI=n}Zw`+dj5_zv+4L02deOf6fhN-5@G!iq|7APtIC^J2g9>jRN?
zY#644vi%W9g1%bhcXEOJ573Ta`1v&j{y5V|yYMyyAI8}a$p(YHT<*pA1_Uvou-DKE
zv`53LDAYSroyT(0j2|_o77HKGc4s0M-*N1FWG?+UB|9iNMpjWY!Eu9QGD+xh&i~4w
z&wFFTk-TYQGD5P5E+Xzx9e^a$!pe7-ys^j<Ru2$zjkdDQ6oOi<nI3@gzwH`6DymWw
zgG7N?f!<-AKH90u1m2Io6o8kPno%&G#HJaA4SaO_dst1yP?PD--7A=_0{d?3?Qh|>
z`1~D{z%&2)!1u8b+ytfe!p|AFQ2gu4zShVM3OSA4MwgR~jlf^p@lBq<)1lfCs!&`(
zD4ySnf^KM3b-r2(n8D8DbbNe~*!5l?EDnr>+qm>*$%31do=0n5wMcZTAX|ptUX<LN
zAT1bBLH=$^x#dq~iIZmbjRx&xqh7c+e@M?v<Q>%M^jwMLvHp2@c$S;DLtnnzLeK2j
zHwE+ctR5+Pq0d!=umT<osFG9b7mc3R<tehAuJg&JUcHX%mBdXH|1#dA$L5Ud4YkKI
zdNs{#F{?gmApNMYshZdscfsRIqKJoWI{6+(s}<DbJmE@Rbac%k!Ks4v&#ko=A>@#A
zxl?Wiko78FLi<Ikibf1C9i1`_FE0gLH=F4Kus(wd3<S?EFY4KESHA+T>Z1(stNl9B
zs|%JMBv3XS5NqECE$i#H-3I3!?NoFl91j8q63dQ`1-TCgamFv6y`;%m<XQYQnkPz9
zFsj*v9}7W30UNyL!E3YDYgB`Ys@0jIE=ttMYVSof%ek7&+pqKYagKjjLwUa(glqQR
zv08$!;wPZ7&w6PISN<gNV5Ib*k^f=coBNb?GlPc7p65KFbDtViz>8>a=y3u#HrXgC
zE~8<qB_Y=q!Q0<&;NIfkMvq#89*-aXN-Qap6TSsHw)wk#dn3w-by>$qk;?7VPH3;|
zJjaeuuI9pQAr3I`OR7;Og~yDZqVdAWa;o55M%rJL?1jDuor%fDz?b#>fZIQ^%{qo*
zDtm(l#L#{HAo@-1YKjFpAJ-TPmb_(~54~La=qgpLt(7N^qeH~xD8soPIyEM?l$%U;
zXRLu7S2yp(;?N#KAxOAYB(u+q8t>&u^YPcC{nJ1<YhJoenqY#%J;T+0$;<z2Lod+!
zZMkNfuX4^mS6S7Mr0jD`-cAmMu^V|!OdPFj*<h^HT#Mv=W1uY|Y_5!QT*`JNYOT39
zxIK+gm|>vyLRg0juh$+GpW3`zaXA=l>b}?N=x418TEtylkobZKCh*&7zwW5(j>M}V
zf67a5C0J=bmx_<T44`ZmGX$sr3x@~)!@#C8pI#YAd^vdgDp~^DpAKmBb3e6cOfY4^
zhFSz}7p}$sO)OUJy2C?k;-@~eE7Hp&qF9ytnV9$|$j(?>(#e*^J@ew}Z@P!}eM3+o
z(75g=@)6bcRT$nv<m}WzBKm{3|1tJxfp;=owVN(Lcay@qqK#7EFB}{izn9R-zW{kI
znpgmVq+A3DJ%QzDpDT07bY|cNBzw@8-Oh}j4x{xvFE@h-U)t+C-(aa+<E>iBQ7VsZ
z@m+{|K7J`!PT&t3ufr4gZ0(2j{n{5i(ri;e{c$-*i%Gr&n+1}&C9YVgf!XCDUi}Z{
zZ{oR9h4uooGenita2ZLjpEQk+fC;qJ^>#e<Gb*TV7=;MKk5dmU-4w+=-5xeR76gwk
zs_oI!*>}`$^nWli4+Cmna4cAYQFsHyKF2?ThMCSH=ih4Y!Rrai^JeSe|Ie^0dMmw#
z<VYSF0t33`iSDf8ysvH<*eRywJsQ-Ec;;Spc=&UVCuMjQQlzG4v`}d%d$IODE<x}7
zXb^odQ_vxVUM#Uv-SCu}$>OQp-j5-PiY>lPGn?A))!G^3J1Yrj|D$YZRGV1G%oai}
zB1Jt*_<Iv{)&*CWO_<0pVpwL_Qfa~oB}@EsI6V*I3s?&UKSYfTjTnBLj@u>t{+k=L
z0v{QB0{--RVR}%456&4$(swFu&-<cQ{fH$T<v|jOsMe8fh~NaUy~y%)2G04AYI34l
zlGhfKG+-=N31M!X+U0G#J*|&Jk2<B|RS(}lDS{Q#HpZwRHNERc6b|vrn~ftj1BbO4
z%8H?@NAyma(FNCw>;(+kc|=N1=MPLPzISQKu8+;p1cqd?cTU<~c$_wIeDz4`ftfmg
zRU3`UA-&e&*>eQa%2^f&M*>3Kr$A)InxhI^KB;;|1xcng4FbVONNd|EVYAYgU5AwV
zd(oX#os-{V+`d=J-xq<+K8bgX!qM)l`f}XV-c>}m8P%WRw`*Ugn&ni{rs=|-ajPfY
z$@7wYAtk;=s+O_`$9L#JP2}GW3}oQvU`n{0;rpWBs$8}4l^qn0zv6iQP4q_kL0c5&
zP`p>+`_<AH`BE?BJTEGla{@7t-*gbpTNca^xjboQgLFn9@ee*2e89z=hJHa+nY*nJ
z{TX`Yu;F9l-Sn#x{UxZyn8R;_bGY0}K6SVv+2YG><?UjqYmZ9ljo_?>zd45iY@vz+
z?{##<`=*uYyYItUeIv~O8r7FKiql5eS7I?#r<#14aeuhu^+K&R716r6&@Sb1>5tRu
z-W!czSr(@XI-;^eSjbw^?d}5RtZsL;LDW>R`8Ew=DNe5t|23u~$Y@ducbWfLfs`Q1
zp!FB&oeSB=@H!n6^90MuVj|@|8v2Sj)9FvY=?>cWDezXRAg|uP9GeYcV>$H!|KqG(
zjiH!hgO>sT&E4Tzv=*`aYzJRECnGHbe&}?SZF!GCQ5`5qtDozx;wZ)On?>dCfu9)?
z8hCJ+M5$3B&$9F%U9otMgD;cjdl}q94uH|0_yV@4&Z4xCx!`qFoKTgmBx^AD+g#XC
zF3C~u8p|_}te%6dK_sI10TYbQxff)O>Q6!#TeJFSC&VxWb~gl0Ca~GYu+5^J%ZAS_
zsd3J2U}P6X8-Cm<auxB}Go{>KqFm75RD#{JXOT|Q3y@7zUi6UN;Iq7lSU(-JGOD)B
z1Ja#r%Zj`x8qU4Qhw?<!{Yo%TjGrzwm=kwD{UMB*&atnQBh+K+x=tA$@!I}*dA$7j
zhX*C;>?2QCyAH`lz=UIX(jJg2T1~}~uw&n-a)OA0>8T>OnRqE*v-$BwD}0YPBGh=q
zAK&tOig7T-IQbOgcA!zcN$I<v+aaYNTgt~nIcY#A+P*Hu?t%VKFhlmowPa&FbXS90
zee>2}c_xvvt&L9`TO#|^iJLR8{MRTd#pQm|hkQ&TLB%hrTyukwaKl^^Mk(h{n4BN2
z)olVXC=e6W$sb^*CL}FLqyK1zxW6<G1QXgb(4gT#ZX>bvPGQA+s-|N4l8dvL&4C52
z3dBP)P5W7rGf6GyFhP1FDctu#moxIsr#P#<L!N5TEdFR0_ET1y-GGDqzTlt_8Z-GU
ziX~tufz~BcM0xXhPQHjyB0QrLa)83C@(IH_Dvjw(a*p#ceKoH&toL5WI0@X=HV7O7
zx2bSM$ThR8K-do+xm16b9GAAlk>(Z^?r7S{->eZ`Nt3RG_TJYy4m^i89RwR8WKlaP
z5Jr95A)W84`73xQ4)NzY=$Ay$x0Z|kBxp%W81*7x+l@mbu-?u;_y4{&f#agErh9f_
z6dXu>d~UfzsWa5r#;_Md_L5`roEJA4zTW~*6+mR#EoKx9xh!>Gmr$o7V|T8ZvJBTL
zil#M2e|YwwyiUvO`Nuzp!Z=tC2ks7zY#8+pB>1(9#`_zTW?$3J`=Q5x&c2`UtTo7U
zibdmo?@r(^SMu9_Mv#9*PmSGbWkYO>k!Dji6`A76>xD2POElagyq(Pw<>aR#Qi`_w
zpt0HI4;wUo>c%z;Nh;Ofr;frF+D_w|oa9)^1D!KE-C@aDa$s5v-U?L7iezt@wKo5V
zyHl>^sx@+l%R%=r|BbKQpT-L<V12KJ=RxtSNTVfKlV^<R#R*AP%~D*(N1BNae_iYW
z2QWQ=%?ywE@=;TTeC@lo=2jZ<S*zQ`8jI%Pz(KS1rl??%fHbF#C8Ei+lqR-^PQ}}!
ziid<+vISAejqw{PLU;>*-8zDnxnZH)6#~{BvUt<)W?Tu6va^aR4|RIcwT;YC0`r|F
z#HlZ~V`crm0Ze1OWbaEI4@b9RCDMeq5wcu}Im9KBh=Y}|Z#lbbXEGXBkG)g4q};1k
z*|HouNSUj3+AphsZMEdp8y~CB@q!;o+zO1#z+mvgieuh_AQw#{fsDn^jxRImlm2t{
z{yn+rkp<`=(}`9u97|*t3g}ieaRBYlS;y%A`kD0bLsp6XK>s@(vXocO8bxW3?Tb3-
zktt(~PgT<6`&RW+z^-}(dpK*E&Y9At%I5ld(XHRYq$b~45S?I=yw%W>2K@Mh9;O&}
z4W^9ikl41?5$RRIRIqGS58_pT?c!LBz0R_xZ7VfeDLRn=RGF6UO@z|K<bMGdC7?{<
z?jxDrv|YDZd#&n8g^}yEw|Ygo+koMh04&O@+BCX8J?u;XIj>im2_1ngm->^T=qJv&
zjBw|doaR`X(tQZ_ap%(}^Wx0BD49m*b?i*8_3Z`$-w|Viy^V2^#>ukmM*T7Ubj%Mc
z^E?03s|F<i^{yQ0qYX+Yls*tEb!SDv)MmKF<2ImCv?aY^Vry!{=5CqhnqjmLOc4A<
zJQll-;OFM^tVWGA3ab82hjDpMhd=<P5*A+_hb+kwk%@Bre<o<iDL!MCJwL*B!#9^?
zP=ak-mB{w!TX*8JtUDDO_9()@PlqpI4E_SH(IqT|jknV72s%5F-nZHBb+WftIs~Y7
zo4tnzZ>}cuJ#NnwDrU{+l5Lb*88X$xsPj*7zdHGMe0n`tVFqOk32G^Z)Bc_Z!e};}
zHLO|Fc|pXWVBJSQw(oRCnaZ*n6!pI8&ffI<$h*96mPK^Ty#g4sxF)S+PSeQvSijF0
zg@vZE&Drde&wJ-7%SDH5;QlF+Wv$TP0pt70XY(uuU+>P-%JEV7vWujAF`COce%Vw%
z^J-LYTb8qkRD~H)_c;}nAJ(@9Awj3V@@m}5S4~Fu{HMvm;{MRCq=%|e(GvTSjv`2%
zcMgbzNLV1krNVK<XZyI>Zl^09_LaD2X88V4J>8r9*Ga^ef(l_1RenwYvK>xjKSe20
zE#G@>)6xr*(SNXqzmhlyeG)n%{n@`yQ73Un*k3EP%Sg3t=Ua^SYtQNX+AJTuT9pjC
z2|-*7e{vwLaIq32Dzp|jlUKFkZMQMm!Yq;&&HVjGVw3<lo0TFu;d!t0cqgCdceR()
z23!O0Yz%+#nX~p!ka4+n{P?d=VeW;rDfr3Uv4+5G?POWSpU!Z{nw=@EZ&Y#dF_hrw
z&D!nFO7!W@;X+|TotUV7>Fa!C6AiAZW{s~K3X<MT9~#_cHhyC(aQ3rU0joMAwfi8a
z&=rbvD*V(T0OU<`MqfQgwx3tvq9iulf~3J64Q&!?-*{>5H@vE>#0RB7q>TD7ENPrF
z6cMyfHYk0fGaSt|4;v7rk^J1rMMxWc35WJ<I+1|D4WhPH_lUK8EwTt;^K2u+O0o6T
z+zGCI$7Q2}=TTJY=Rs?vvV^#2K;ioJOb0ehj>b@B6)U;*h=vJfg=?bNmxIC?&%xWk
zGwJC<I0?Wy19e3AlTcsuSqACXm%*fjhPPD;Ga|L8o%q#9@GL}J=(?N`Phh?D_nFR#
z07_CXoa-UN1f!3znU<Er5Uj4;UK)BaTj;SLM(e=zMhUN_g4FpHaO;J`A3B~5b37=l
z33}lCTIxv1GHCrM$xIF>MC^<^TM5(7qJRL3&>`edM2MAya%U^o$ny2z!quEmbrhF+
z!9sHotSsh%$<nKCIK@~8ZJ$pYi&aVljg_xPNM<_rw-$Nj_L|s_@60cLxeQb=4vF1f
zX6gk#)Q9*2m4pBY@HXwbv7JqRe*<JC9^XdkCgFNznus}Vm_m8q+TQ$BPZT{5iS;?}
zm2g*5z>vcuuu-CBFd=ypV(Bf}5J{R~u*j{O>NF9nbv`6!)h?n=hJ0z9OK)6)7)5+Z
zqwe*JxwO)xYA3&dW2bm?-R;bYa{p<&RJc_)VF<sGplJEh9%az9=CjxEe?nVY2OkbO
zOd&tFk#}XThB2r#%H!a@9{bMv&1`V&g+Rht_3pekY%_Rk^n=}~KHyo?IO)62p@){I
z=Evf|{M<elbn#Ixi_K;|7#MEr=OgoA8lC4G(}p&Etn=&SjX#E^|B_oe08`><8wi6q
z2Ke#Bm294@G_M()T%lu?GLlbS%xM#G4p^;r_i!j+%@cz26#A(f49AUexd)*ST&)ur
zv?+e{Ra>!tBb?>dbPbkPjX0D1jc4yL6^Rm9p1p{KZ|#1-6I-i}$lI0dj{+5X_u1zu
z>RaN`2}yM4?{L}f3K(VqyfR@CK5BxlmS@R^!{qSRp`lq90n1~<sPV-d4ceSW+@J6b
zo@O`5!F-hacfon{l;}=o++Hd0Ja;%B|EVP8T%4+(v)Nb2Mo?8uJ3Us;a!T^Df9-%4
zU$Cxl^zhoO+J1<87R~;0=!3`I1}Ag015gXeLGX77k1m3?g1W+&M0Db}PdBMoZ<s_-
z7VS+Yqsq)L$Wl3ZZ<byE^l=EH^6<ZGH<CHB1B_XQmTgTw%wF9whULu<^|mrYi1!ow
z1|vX!448WOkMk}>MEdZG#X}!Rdf?PFHP{_Rif3GQbl2iA|I6UAF`sfW&a|kGEG`x)
z)}%;WdnDj(;)|C9t>!m(<AUl(`ts}#z*(l4pWD;G8jsxksPGmRFZT($v8cGeXlRNI
zkY~cF_eB-BA!Eb-F;90<TRJwitm3r9lSwO6s6vo!{u{3PpmWk3KRt#7slSX$gboJ%
z?cT2~r~IL(TO|b!N|F(s4D80(%WKimDcJ9^CAE#gQbBP#g4qwE)y?UDdlAMSl+SXZ
zmrt6CToxq`CV@pR<Dy|oBpy9{6LYvlFb09713Tls8k%=<kH_SjD!w+wundRN0|xqc
zVWYQwm+$FxG4CPAm#z0_VZ5I60Tyy=9k8Q_-R{)nEDbf$|72N|7xzgqVb8?4AVrn4
zRw$xd3C$s_YC0`*j!YWAyS%9I72URlm0&8t>-wi9%k0C-#Ic^rO*w__+yIEZJ&vAC
zl??u8-yk*9sC=lBEUGR(Z_<T7lVW-i3b<MpvhXFS%rcgW>~i~8Pf6sH2<`XRdv#dA
zV;#mvklCvGVxFu>Ci;Vu=LEP1%-1g=sW1!S;nR7sPNVw!Km-{#f^FiVjuJ?QYgqE1
zl!#|LD4+L!EINBR2F|h(r>p8PA2hLznn57LP?kKx)rTjuQ)$|Y<wb<DTQjKJ<X(Wt
zbDCMx{B=41khMSYyaNopLaf;zzK2mlt1i^9m#-e9P5r@xvkUCHfeZmZL(@Au*d%6r
zl3XYDpD+Itqzj}+9JlWll4(Iz0kp3=bCbJX$hqFxpsEy2)hEI!Y4c^AzPrha7Zzy+
zi|%Jm8R~wEg8O@?hgsM(-ne_03X?q8HT}*QV%eoJzxIWLA7Ew(o{gXE`K|&OFz#*>
z2@F)1-A|_<8Ji|Bc&_1G{#rX#H2sYj#1|398X7%hJ;db3B;nT;7G+5HQyiY3X4;a!
zTK5x|9>;!$f6e7ide31Jpl>@8j>j*+m@U)$fx|l>>K|;U1So8X;q6@wq??aX*{h*%
zIIek8`V1u(qk6nfdSI#Y1^Unxxy~GXg8%U5{;TCmv+YihFxz@gX{+gwWh7ipsRKVw
zxK5ROdj0-?2C(jPFs=#BJXS5BdS8vj0rA`YT|>wF5~J%-d#>BH*u<WW;RZ;2pPg1s
z8(b6dIZq)gCcUJ_r%Ygz4<+x9rMPD3-clJs)#03~SSf8loxc~;kwv`I7y2jX=ft?*
zsR}DOg4~-fel7vsT({Md)?AUO2_^QaDSVSXc?MA;Z?9M$1nwWUl%5%1zRy+<STu?r
z-`A{r0kMKWh+s_O)5rd=JzRH`u|A*H#}z`G-h=rFESIpV{qQ7H^W+{)-#yZvPuFeX
zhHA2Yqd#C7=Wu<jy}J1q`cZmS*QL_b{$s3_W55H4zAO4YL(fT2%{|^KlQ>#32Zv{Y
z-w@2(>3!>70l`ELP7hoUXfg!a4mIlG(Q#ZK54-#1qL-N_#4&^X1%2x13n-IFQC`YJ
zvC>`1<`tO4n*H}dPvv3fm+kY5%zs-PGkDs{lIG)@&kt@|`fs}?;W?<R&DOU|HmAJ(
zg!7K%nRh2a6!^?`9&0h5vdG<juNN$*$HfOPs=iBXJHD7Xm=@cm4oB_O0UDp5Wxn)t
zeV4Zd&l{f-m8$5BcrHwJi}Tl(%0m&ym~ssEzkGTH;L+?}@Dt)rwbRV_z;3OkpSp|$
zAsO;~oK@v#*vP-_IIaA&PDs#JaRm~TF>>2q!<TjU!g{4+xK0bMq>L;frXQmn<EaTq
zpQM^3praH%{JV+O5`S%AyV0-o79Z{|go^~mDsEBF`_Bw~X!O`@7Pi-Remi5ibh2sU
zWT~=a`}THsYdGbpEJbbTEYs+dal?GnSS>ffWC2*RC3~yeg2She02ZZu2QJTK<Cc7P
zRH5A#7^G;;HIIKh&iRatB9Q*Xl4~G`OwKg(bPF7Gfz=mwY;YTjX-GoE6BvqG=$=FG
zBJVZ}k$@L*>T)*b9=@N&k4CW41fTg(aiiafhWb${2e3s{!z&R@bGA%{(t`wZSP(W)
zJq0ME{*rj_@XyP5Ic<2VRh7^5`jC`lnoA%EEUsz*T>_G&j@WrV%|uQ=EH`XaeGSSq
z`4lBrk+{~zSc^53A?WMKOOc8-l9tGj+4_FGHX!HaJ{{KRqLw!e=!8hlo@{<52PZ$T
zf2KF{%cC7)mS9|2_sHviO40-{%ffd#8!jL%t06{hH)y%AkOrJt1L~hD@G3LsIlS6@
zo9Gfm)P~`O19^QslI5GuZL_RHDR}uPRZA9K^1fxlu}UlXS~`#k*75B(-5zgns|oq>
z3#h4)CamOMuSitj{BC*#MHAi=&JBCt4r&h`q1K43FxJk&(Y-%+JY3x}eO7{K=R8#S
zM{A6ZrhqwANy&0T)G`KsRtCC<wk_7T%{M)So-<Dup$&I6R;R&Tl(UW-lSuVJ`uo5K
zrr9rahOv_ui(;;u_nuA-i`+&Y_iN++LL^S_Gby*DD!SHIP&^Ti<FVa-=La2m_B~)r
zI}mJpsRSNu9t&dzo%VmAvO1o{SHeKK#W^4vfg3^5GoGz}RgPe*CM(3cYsT7&#=56A
z!mP<3^D^htKLTJ3SOn~r2aMfDWTC?d5GtxZFI!m8+aX`xgfHnO8cg#(qpykp7qmxQ
zH@0I!oOC_;0arus@c?9%C3JpPFZ)JuS6Dp5a)Oq)Mc^Ze0J65}@kd)<<AJ)whr4wj
z0!{u>TiHBtS503hQL-37$Yo1)r0$7Q4kY?hNP-;fOr?h@f-FJ~eTi%Hb|0;3*(@u6
z>$ZSnpVCM8Ihvn@TNy*Nd;p0!{l_Bo<Gb5>Wsp2BjhK?sA@42^%tBDJDtp;g8O4Q?
z>!0i}R~EJq@b|EMa|?XCti)9?5pW(c1W3>6LyC&d>D5QPoeJ2l-+DLrtXpLIj1(fb
zmlQ0|#BJUV`(O6}Eld+)CmHw49(WN>ignHsbteIwK%%B`%?OQtvnuvACMk8loIxV%
zhly!_cX+%x?GdrS2+$tpbYK~RW{R8SEIQY*Nq`HZHC1nDgTL2F*oc13z}&c03k)_L
zXhlWxKSG2WM%`FeDuO{J4LYWanYjsuTP>65(><Pe&2{(2^Di*Pf4cI_BpAKxoAABw
zSRQX=%RRwrUj_=n|M<|oCp&(0MZdI?%>6L@1^?`TpQGtxrYolJv4#;AFG5Rf&P5v`
z521hyBNnNxEFtPJ_c;$Z9#D7^FfEU>3Bvv$=N<Dl3z6CXWgnV+1yB>d1f~gZ6H>}G
z*H|?doHj<8Tj)=_d^||!ssBST^gFR$UEOQ64idBfA{}O(B6J;#jYNcIHW1^Tmi4!N
zvW?g~ka9oOQSykWm)#9%rsuWRmA91+!>o1ZMpg*;6PaasXT#_79)j}&L2kl(J_vEs
zNpQUV5(8U+R``={IXKIrk<|yKS^-no(M_w&T$mP15t05i<Q4QovOEHvd92&5Kl`XP
z2x(RwR!v|VQAM}OM_PB2PnURbrW*!kh9eIg$>`VQOR>Df8~lqmnDI?ygQiX5{JdV$
zj%WLtr$M@jt1XtAjyHpjC~NIzU42iT;lj#OYFCXgMQvt#Bqk8uM|nCT2~WI%SY4YD
zSI=(fND{s7G@pUvXl3E989bf7u6B`*<?!OzT4m|DhePI(-ak1WhddrSMLj1pipH(S
z?xuU3mMlbjOMSjlwO!@^;Mh6XpEcAGSdwKZb5?7yY-)5n^|o~|^B?+slh0V$UB#t;
zQ#{AX9i2kgBBOcFtfRGA*l7OCd{5l*r}qp)P^&qJlK<p|B%Jy%sLJdp==%LC&Rd^9
zd}xK}vaoBh7?d~e5>_#5b+y9)UbGR-7Ei3^^zThB8oxqY#;i%Q<p*<+#BuV;mGbSs
z<<^udMf7YRJ;OGX^(2|za9Y|R)zbHoK+seyGYjF1Ditn7980(?3NCdS|EKQrGJIk$
z)_-B+Z}QU45CBG;tP_W)f+c1)W<4#QTs<9cqqd9?>zaJ5G<ydxIN)J`9xww}L7+>c
z4bHA%{TVaIeWXnvQyY2?410^7uU#i4%afh?Xc_>~jC&84jzRW<fC%TYl=51zn}b%*
z%7okmVO*xG5p$0}{OBN6C8r3_w}WuH7It<dM%8XVT+)_YQ(F<1=>v`gNU<i$mEuuD
z7=a|C#cGh_|2l_;x?j9rt!KY#r4TfJxVNt+f%*hMU5`IH#6|8&8#00&%U-3jeoPF<
zKgotTe)Tu6W~1Hkqmg)rvJ7wrq(;=tz|&<hGSzlk`vWDpEitt5!u9h^wXg4}KR>+_
ztoO>QuH1^`J(~5ov=tVCp2NtavB)ll0x-|X=Du+>qnyKa>bH`y9)Rw2x!GP%#g_4{
z|5Kj{3Pc4ooWR3IiAV`!dSVd!5BY*MZ@ShWa~~7FJXT*7dylVOWbe4nOv<wLkL41j
zQrUKvqLx;smTN+}vMQxa>4f1KKlB$u${hI}-%W7}s3I^Ae8;paW{%Z^q`u=LdIXeW
zU4J8fTIL<SJvLBoq0wA-+IVkQd&KLUNwp&)<<kxVuwtpOPmWg(k$o>kiq<}|Nd5<I
zWhF^Nb{F{4$gUABsWwzad>-N&$&{;=`=P(fyc{ta2Ui8S=l$wROG9;;W4SQQBnrJK
z(`I4QZ|yO;D|5X0yy9aMXV3gD{JhokpYnrOmTI9w9h5NF$blJJf|W?;1gK{)g}uGB
zPOY90@uH5Nt9;n!{|d5dxrwEzhY~Y^%c^J?uk&eL<UuQl5{j`^bQ4L9O3lwf=TXRS
zTotJ}MP}4SW*qx8vj{ze6}w<tSl#<eZOp-r7}C(2gn}ct>jMqa$8+%9BJWuwLOMq&
zZ!qV1#_?dMzK02ubJBMi5l=H)Af`{ie-bUF>NfXu&GSHL8g=fdCi3i<cgMWF(^v^?
zecfl0b@(LdKm&=yEIjRqN|H)6D;FyFZ$x}D>gh3Vd}~TsRx$Jlc!R8&wb=7eB2h@K
z0cr>Ea~a|TFvva6|94`2*^k`n`AnLb($=O7&=y+D3sVDJULMnX!l)K$hT*gsnr+D%
zj8x#bH8pzQ@IP;5+8!D+jbL?Qn~nj2kSG{u(#fNzexDSD44TS)wBYnH8=LNhNWV8$
zPwDpmARic)&tzpY<FI4Fmxg21ftKpC)-f{md*7x={@q*CV5EEnvaq>t&uP3e9lI<_
zCQHCQSOy_$_x{WqAFcVc73YN%1CDA<%+2OazA#ZC5;sC^(eHhV7e{3`%1yqj%FLqc
z)Asr0rKC%R;j5+m=COwVo_v`igbVA}?HVs$6k%$srL#3!q#)NpUs16!;BnKZzWHP@
z<{{p2Z=yuny5G%5u_ho|lLiM`jh(Q!U5S|#mn=#fpfgU(?)cmd;`xwJhcHGnD~nPR
z!_%f22ei|7boYRY2=}K4&Fiyg&<K=OAvk@VcEIaFHtZGSA=*W@CH9aZmZgp9_A%Ks
zxV(9r=*DIF5Aw+|^%0hdK$o-7^l~s1-!6QS#>*1`K`X~M()pPHrW1}S&wJ9#6DQA)
zdYO8TdA*l<#M&A+Lb;!_m|nFaK+=+<S#iRPtoo6d%esF8A=&`4=3)?JjJ2ngq-`f{
z;i&@d4_4+0sWivi^RFr{PnQ{kPRYeASp4F_RVUsXD&=6AFhK?54aGEzlgfCU&6>AM
zdEx2o8L1^dq=85lCNk-zxeU$&ZRf;VxW#I>@239Fx%Jsb&kF4T8nLUE&TAw_Q<3tj
zOC5AIlZ2GY_oGHk`ncG!0pkJBEee6xa=yTAA;f;6ckRW6a>=n{Ai-!^%v#Lvt}r98
zivGKnVvEpcp~^k;2eJ1(^k!|lhRZc0x_}8<Iix)t@4%^iBs@A1zzgoXxEpHR5u|j%
zdr^sKly01u>?f9jU1XQTRd3TYsdMT#|3D#4h+Ux4lscjcfN3@`AL<O(RzDJB#g(8Q
zTgJVOM}m=328`H>Dm1@G^Bp6^<7`1D5;q~oqv0Ylm6T@ldt0)kO85~#8a4PF82yu}
zUm_EHAq!E9=YiAgepMd7zkt9Sba8iDXK!>3N4<c!JG3i-&b^Cbyl>p&GlD>M+|4A8
z%C+lU<q{cNHdP)}<#N*98J7(ZfabT#YDzo&o<QqHp?r?*{gMA%fH2wmWZNo(IPtlt
z6D;WWY|1To@#cdu^vj2Rvn~qI1eDIxMkNi~MkqG$<ocnVYXwtl<VVHjuKM8$^3rTo
z4#`Xn6+3+QLY+do*<C5Gq_jcP6~x^W(FGIzDDbR~G`gAKUl4YRQBDf$!#i{kodhfI
z&-8cilSG8fH3CA-x{@`Bf!bkQ@ACs+11=|BBcslJITJjVMLl<$Hxr%oC<QmmReBxJ
zJ|148FJZEOD*z6tKKVx89vpA@HnjtTGo-+Nj$nty+KPv``qF<Qf$e0<9$_6SBSkMp
z(o%H`iauo}<<mjbGXA?QpPzo=Q?KLW%aHWLlKft@0^BwRo6rjH`B)!o6*C!np8ljl
zokyiRj~FVo_D)UEPtd`B)dd2LrF%=+yXNcSjEF9urjgF=?&jc=w7nite6jw+*z}(U
z!ah%a8~QDR9l_f_-5=jLbRF8YD9X>#SZXn8IEo)iKb<U#NnNVRxxByXS*GwoOtt%E
z0F8f0R#eD!mkFx@pG9}GgQugOQ@%G{*0uF2nMzg1mejPFuBj?p0!EYHr1@ym5U*nv
zD(^+@BgNXHWPibluxoGi3u$s#?gPbBRbTxB>>j(J(q8=_Sk+sSrQ4*ad-^LY0Ny>k
z67^SB3Y62eJHSJy8O1F^5*98mn97ev9&)lPSiQMI%z8D8scg7-6*J;TD=ss>fDyTH
zvjsr-z>>pT#DPqu`PifvPXII48j=g&1H+zGa$(t<0{*QR7uOcfelCYcFF<mBx7bDu
z^qERF7mEjbD|qkD#Pk-N=&CNF6atlWfowQ0ic!(ggbY4XkMarVk6;X%WtsZTNj>3f
zG&&#3oiRjYdlyLVr*KoE1jKSJ(|)KvO91)nuxUGea8Kp(gOI+)<0j2ScmMBzGj||5
zqqxjJHnquL=OCZ3L4B?EEDgQRXehttc&|K7`#8i^c7;c>j`(HCb#fsOoTLb!4Qpm&
z!A}g<WtM(>NKA=drrkMud7(u(e80>@DRff)lcsd>kTX%3zXm0pB?(+tvJOuAU|JZj
z4>iuDYMk@I0AYs@Y=r(zXzc$p1e3`^(JuRKolJf{9fu6F1$Me_H+Xlv$1)e7QvUUA
z*4JZ<>scs%GtFR!uPssO7UNA_!{}^!36Q{1)vPG=_r-mA{D!{w6|DmkHBsT=J3Q^e
zevJkZ2xa$qn%ZG&(}j#ysRR<3h90GKtt4NakXioZOkiu}=od(SsSB<i3pWBy+s@iz
zKN=YkUHyI>Y1-}57ACSmREp`o?<#3EoGR0x6x)5z;v(+j^;wW8*67wBEU5w(TwS!h
ztx(Ip=g=BSdrFBKPSFA+i8}eTtS-Smc=pjSz^wf>=^Weu`|uYp+Dr#RA^b9QL(h_T
zbv?Y0EB2|f{V@5OJXZq#ED8pIH;3S%DmeQB#zx>YQ=xnMXwjxnwPG`Zcq*#qZmW6c
z0d0z_H1KYjBA?VN6HK@e>r6EZJ^+`#mY|PTireYOf3Mu7wV774YM9%HdxS`0bPL2(
z;Vc!bRG~_3mW~Y-T}n8bqu(_BPVqB|3D2k{W$Rw#U%;Wb6Z9HFqm~(-GG<v`6n>+J
z<i$l~WVchQ<GBup;tSYNImCogm{a#dWr%I*r1JQjFOog)1Kw<$YXn>c?%HB0ToM6+
zF;jl;5rh@Fh7r+{$Or5u0sR5ANtPI6Bz|lFD^>H7T08UVfG<}hQ~f43-s{1|OBH{i
zM~fpu)>mq1dsfezy;fG_qz$n=*z48CfS<NT&qlC`aO1YO8}`t({PR{mkzT46WGgR~
zQTnoGHy25XX9qW2hhyv7EW(R`Rtb0~OP4e?r((0M@I@kvE58y|8$kt9__9dI|CLHA
zxf_Byxn;Q2pIrPGQewc-2I&V{%HFVE`jBOBU34LqvltVR7nes|wDmq#WB!=F0HgbZ
z_@Xz$qgqzU7!oW5q$MZU`Y!vn4Svg?cgj2`ZtK^nER%|+6<y+0elTd+LnS&^IW~Bx
zws}5~P~aoEsMZwCoMk#8)BG_t74bgwuB7jn683c>^J2X|@esg0(a$GPA40vDd*9dT
zT9(MBHWhV7zLY#AN$XcC9N$2=`7vsHm~z{LiSh?V`j^>{-uKW2{});1xSHK>!iHe^
z#pv6ZL8dP!2HGQQy;Jfa_EEMu(Co$#Z(gNIEN5r5x@oVzZnEiQOKErIe@`N6r0yl^
z=RZFiR7<V4+qYbAWQY>?*|-HZqxfqg+kPK!M*OVK`Py7e<YBp6e>;;{@!gxExA}cA
zhxj8?C49~cS%-BAb{$^N&E^f*tBwp=DZ7pfWyAX_YamIT&ghF2MZWT~h{6PGVLm5|
zp^7N37NT<?6<N~8tqjmJP)1R@EDxQ+JrO*T!as}DeBmfV&Y4g4KS_(SqX>}smT#aY
zMPDLkswTm2-VUb|B^ftg#y1v%Y9E(XtS!HCm>AQj7C~u=B3|=Skp?=X5nK2fBByDE
z*xCdo<W`=qH*g=_^5t*Y>AUS1)?)ORlCNz-VllIZRGfIpV|PT1O~$N%S}3}u9*!S!
zR3`Y8vtDEgT>s|6kZaH4HDbi~i^(+q2R@Rk@-_f_M4DZ$_CGVByuf;-n31%m-)p>y
z7PXpnPBUkO7|}y$%flVLV&WHvjnN!S3K%{8=k%j`7+};Eg^%=cE9?$e&(ZzX6$<o{
z*1F%{MXhmI86oVti_I>M#q)?7cmVOoZr_^6i!i-tF;?fF{?D?7*<5Zia8?KiJxzQc
z909j@yD|-_Z1dko2>!QF8*__{ZTzc1QOLIDu}qMcvqP$%I+VgVbi6wZ?fQ0-v}HI(
z?|kXKU0r69B75My7*dvBcvKh{ptKMtp{0!z$k!VPJ{<Pg8#tcWjjRsos9S^#r6PfT
zzTvL&#szTL-dK;jxjiX{ueWF9r1^^F?j+Qd^S9iDlzieSv`fmUR`I~w_ToO(xHK!`
zFFol;b^W3@V_|RthJ3601gG_g6|!lWjiY_EmE?p>nE9ON%p%d67FW%#$#Ud=jR0c%
zSlymzr{gtO>|JZ{RDDT)lHZkR+hBZR-lk;w=&Sf^MAC?cMam1GQ#k7>kv5NiB;Ccr
z*wy~ZDQsB>kq=P$8Z}pwa}!VfNZD}sayaAhYT8Zv>CGw!?IYn%n{4P)+H{<Fv9`b&
z4L3n})9&Xc>qo~ySJ#tyW>S>A94>Ed&IVoGr{K3uWTX1r-NBcxU#H>lzi`mzK*~12
zDmj_P!{lDgNTUr}evU=#wYm0e`OaADl))~6L&$bNtj0EMWNb*EhfcNr#|49v#gy2W
zHQ*{ju&#u?Bp*)yjg9i{X?^SY?|1!8?y-CAw~`kUFd}gkqC&V!54mVd^}PTG|Kw$>
zLM4;Xc0q|RA-o}ZBihtrKODw4k+&#5%{X$qEFX8~(KNmjcbvc1#r!Dtv0P^=udN-^
zI;5ed9JjMCA)xw^Je<ms@4s3Dul14az&d@Ut|yPS5mgvWE0*?V|3X>6(YmI7;)fdR
z?>RvRwrgajPjY+<;U}^MXpb%UPPLTDd{3lCh8{PEpMUUX*-46GW6;OsyLgfU#Dz}w
z<U-1;{cT|LSycoapzj)^Xelq6*-|`0Hnun^Qbf-pjc5oM2r?>E5As`0R&30@P>O1A
zb)TxH()_1+x`jYa1YR1iSW05y`&88M%f|WeWYV0Fk&fV>b8%K5FR$J~KV@mnM`$$B
z(qgkp|C(l7a$&jJx?Dd0ZNAsuinvHOmX3rctU8;yB_(o{vVK&&6hT!@v0d_}l!}N$
z{a_g~Qr6pV>Jf?91fMjl0$+S@GnXVVFr6`gz?K#JtsGt~(7{yx-Z}H)DWa1h8vMcG
z1Yep*p_sIegJy30<3b%3|G@JM|6{y9T2bB!FF!9Ev}7SA{Pt;DO}@&#^Ym;AZP-$E
z#5?wbI@YBnzmGrZbEFns&tdR2O<ccWea<r~dQ$^PiDpBu-|`VUE2pRP;OJgxH*_4A
zp~FZ?-so+VwSZ4VeGojhgU(w+9!fezJ0>A@ya1poUy-RBpftP;g2-!dmtyfhh@Kc=
zD>0|InivU~tMBo@dw#Cw`yoPQiz`Q)w&cxQ5I@{NRj>j&$ilRdS*a;%b|Xwrkq;m$
zJ<uNF0IutS`#2naqz!cw+QuF26t!&_NHFIfWc9O_;6;|)`MrsA$Be29*6Y>uox2T}
zLJCaK_IZ6|k@O$oMLv|s#`SD%3Ai#6%PJPV;I$ru9+rDc%*m1Z86bhMZ6HNBoKc%)
zXE76@s@L0HRy6)@K_1$&UB38Qb^HrX!SU<V17OFhWY*g`dCIv@VUt<0wvB$~1c+sM
z%a6L^9sjf^DY*ttP*q}Dmmcw}>wr%F3O#2XkF5H;%{(pV?`CkvRA1$NQc%1&Wh<85
z=R=uH*i=gO3na|eG8V5W(=R&w@(2N2fl~`C;3;k!dOJ)%)dYFxfIPi08}93MqO9Q&
z&VMv~^f#LlLKg47*+r!B(JsPI3Ko{D4^y9KTIaU{=Z0Y2@>-ly)vr!EoGgZI7XD9r
zZxt0s7j=u`E+G)yEd<ww;32^sg1a>i2~Oh>EV#P{2`&wdTd?4+jRtpjyPbUh8F$==
z`*6<ldFig&WA7SOySi7cnsctX2Zr>%FW6Me!JBQx77dGDm9-@OS>fMygHF~wV|C5s
zo>gVOam3EW`LTDS?91deEz4x)!kLfhvHQzj`|^LVKEC@0S4Sm9$f(NwYn}ivtOxse
z`TJ4}sH4Yq+q-X60U~g6EGDeDnH-0a5jZVgs8trp2)4#n$Pt<2<}6nyvj>I;J)9pn
zQC-eitV2FR9FySe+tz!La@)Ts@Pd_8JUuuXX?Te7ubtfL|HXfYj@Fhk_3?VJm{zj)
z0cLf6djsUNV#bOd2CO0nxjb(Y15KgFMOB_m(bE?|TRc<D7iaghkq3tcz{BLYYj)on
zM!uraOFv!)CL32Z%2vM5RopVgS@ZL|u}PaZRRKQW;#PPN{bw#PBi~18JSDGRu2+3_
z!Q!htQ+ITcY(UQDDAwhn{WCLZi4M+@HW{eTZM_vX2YxZQ7L_fBf5Vc;fT|$1XmOW1
zr1WfvMyUL&ang8_=Fx8c*j=`?{0@pp&{OThaD@(%G8}?ig!`KtQ9mZ`WO{AwB0!M@
zef#|YwOszpnj3vY=S}rby^}GVqlu};Kxzt{*802!O3S0)8lnhxnuJzhqh6#4>)**W
z7NPJBQW<?T>|BR@q55m;7F(|o2jC=YlT<-ipkZy6GxB-0>YOmnOW`mnymH%H;axiu
z0Y0<ohmx(Ou3i{A;KhJv5CS;jrZ;}yU3y+`viw~e0M0=eVhKmFz$YA4x^UYZCQ*h0
zJRvD|lCh~3^Ld&q+C#-kt)`(3VwQad!Sqg|yHCW}^IhIQL&s)Eeo;xKkc`3Mrk$~M
zSMncHbyi}3yFuD(Bc5aY=Ew2MR|DMaW;Gcqi<ryS-Qv&En`>)6dp(iRk$#K5z0}qH
z_l%Ia>>|ed3=9ya(E!)!-Ny|FM&6DZu86Bp-x8p5ddKSx%g#OXbzJkQ4{uk+wi+#7
zM!d<;<mfSfArxeZ5BnAYiWKx)w_I&aR>GjqkJCo;-Q~i_w`f3V4ZlvZ%xG0~_oVK(
zae?_JA`)d?hH!e;rwD%})7tzmTT809i+YK!4!gj`Ipu`L<Tk>b0jwVY5Pi+)qm=Le
zInhv9d9t|Ra#o|uqCqQzS!^e*UuS-{p}584P;)itLT1QfkoRLTZRs&dYG0D=eE(7l
zd`97kl60#p_`yYinTbHJqn(#b%t+X5sYzM|0ST0o<KA=bUYe9WYFb>XYc2}DxeU0{
z0RSC;ta+~KR9j+eVnD4hGED2-zyYHf$=)7SjMu~2ZtfN3`Y9s_l!PMS_DgckzVb{|
zlt<S8Ql&xfuj>!;Fv6t<oe9InE%C)4W$jR0kV;#_K4x8Bo?UilSo@T<Z>rNIX*wOT
z1?dv_0ed`bLdN3Ah^OpQzf`jby%*F~jePuk_Z#&W59HaNdj*k~a`&{MKE-u6REF?e
zuv5_A%|fy=r`JPs9)o8V^WgQqYSXySR4wMi5reI0^kvQ5@9C?k^!zJ+{7FJ}J3jDN
zwLKZ*ka5}Wgs%{zB<iS|Wo^2Je}J&Nf2v(VEvG6tL7o7~63+2qb8q!!aEH=YuW6qW
zu^b<Q<T7gf3cfEO3TjfYOj=}v#rW+$J5S6z-y%;0T74>z_r7!eLFm!id=XzK)S<%|
zapAW2D$Esy#X=Z${m#GQ`#4D$Vw}oSq5^T_OSrT2w>FNG&F5#+B?MbXN;JefQhBF4
zebYihNyS)ei5kmi`ang1%eNl9U(xZr1GzZmj1MqAc8jwQ?d~D`KlVt@6km#H;kBWV
zf|d+E^4uj6KbzjQ7U;`<hpt8x^5-UC!q!7F^MO<OLQokDJ2n-x`>Fea7h$W?isixf
z>e_WcOyjwJ@Dgegz=2?0XnX}6$6W9b$;KCU-0cmkz@ASp&|eZ~pQ|d6;uropK}iwO
zq~7srH#;?v)JZm=u2O{mi-+8EYUZ8{_)naKIwxa>b#0G%5TgQ3*_pnl#QbUWDn(n&
zytYMf(|<B7je&VIm!o~Q<w$Cmkb;++5=YkCm&$C{lM@H++vie}HNUBIOgXKQ)SJ4r
zHZ(yu>6nj60Zu`P6fA|x2%PQhfRtixlf2JpvhBZl6ZWqayDaUO!KC^+-D)I0C_du+
zwS2_N@bW=5`n5CMP;z5T?fxiC4rrN%_VaC`SNy1UX;ePDWoe9Qeh+#b7I}{L**5|T
zYfeI|-f1k}0A`&r6fJm?>x<uG$U-oodf@-URNueUaOq+9xOu({@^@kHc~N-x{BVeW
z;=8uJMRfCdXant(^|W<O8w$AnFfc5Lml1q!9Eu&S$}^p;>&`QdtKy)H;)LzONn3s7
zxf|3qE5f3u>GuV{PYR#xci+B9)WTGkN4DC0;vh=m_IG}<r{}hGtBW%dqXAs)$~U}(
zcEXd5qvxG*zoPjB0N-K0I)S8-!c4aB_^>JNx?Gc_MS;)@O9C6Iysk0!{~p4LwG!jY
zEqB`vxpRTf$=)UQqvf7A%XO#U*=Ekf?Sd4@AfoJr#xa2+NUc`YhV#1PpT}SD$PM3-
zOtDIkE!+y)$i3FF?{Tu=9$-Iq{@H9;SAGCttwXkBJEaAdDU?e<NVP5InUn0Z7`6-x
zlS5PzuI=sQPPAWUQx5x{hPutkNO#_9R@1Y^y(%Y(2u-ubsD1Zjh4F=$6WN8n?^AAj
zZL}k;OsCz0xGi1u@zAq)TvB$4YM8yR+kd=Fp@agYN9CBD`A^M|yF1(vO&0L9*f_XM
zDc${sP1$hbYc5r2lDt8wEa3Ab6*0p&BB4_bwnYWMTCZBr=UwZ5mv>)0ww&0AD}ovl
z#v-^Oc)Zoiw)x^JNLL9_{D;U;IN=x%LJxSm`e8sIAgnsCsIY|RXjg8kPN#v@tJuww
z@i;X4&I#=*Si(?~#gt9<r%}$z``$G)NeT;eX@3~&^J>Gt#&{&@s6LNQgQUtGZkoF9
zX~DJ%{Bk=Yz4gBcHAY=sT0Pvx-wq17Kj|Z=xe)hauvq*&`Wkk8EY4bRtVZ}5?yRgZ
zU$G<*B3<^UUN6;ce$PuAg1xmLB+d)4Ce@`{Gr{#eqv(A6nvnYmLh694O*D>dhoTu+
zAr6|2+(9D45)3NmdDUF0=d`DE(_JBS_p$&h(%tz%IuxGHz-JQ4Hx(x#jo8oJ_k*hs
zK`{`hl3;Sz6;u;MS9fw#g_3YJ{++7(SUGlv`mpNT1|Ro?Av5gZ_YX_g!1&Yysup$g
zLaDHr`!t33VzeU2Z^(7gvYOSfOHqh;#W_gB9Crm@LedpWBTu~m{FGr|lK8pAMXMCq
zio$Ox0b<fCD+jfe?VHC)Yfq}DT?8lBAB&LnU{37Pe#^&;;1~7mH56Hi8NJn&e@FP%
zcn^t(QCzq@bC+pNK%pPF+@SgPhjBg}{lYM`>kRZ`p}aWHu{Q2S<1TjmT{-?Vx*a@(
z*iNcEp~kQl8>!7IoUPLObVl+&j$tyvum<bE|Ji`>W*&f`N$UG88ud9ZdZaB!_&7Po
z*P&MfwLr-@GL}ARkXdIOUj^ekT{HE^wCGf;r)p}bofNdqPc)d{mmr%NVk>1BlkjcT
zk?-LCvODGYY$MI<zKWGilt_a5x07OWUpH%!w4_U78ii`n4H^$h|DhrQZ<>ir5@`tS
zgXfoll`i|`gtcOOUmiQ{Ml#Lz_bog+1U-YV9gm0)SnxFg{`RqT1CeeYgvN35L}W4&
zj59Q>_M4vL+@-}b{$=e#GTN1fiTAS~4xvx}d3Mi13vl^HoD<{7K794c@Eqi^{c+fy
zwg~4zc!R_DkxOB4DOAL}Y%@FjvGu?=ofxmpPVTFyX5gMW)+C<(e7*lISJzs=qTI%k
z!%}>Ox1V5)t*Y?-X^g_YXM<h{Cn@^14SpSb#p@^)ba$+Xy!mRHRK5P=Bj3%b-#CkT
z+d_lHh>YKVq4?*2pi&d+0~uGC+hWhAEeL)RQeEag;F?x^zgNNrxbi-3a=)u}_=n+8
zFawl)8g#Q;y%2$jrNd71jJk#59M!!l#kCF{O4U)nXZ<zJ$^*@j82}VN`&DIZF5;fK
z^rjN>3mjPb@s{g)w~MT{0-_e|4Ei)^r{<qC_e&B37p&)nfW9;(G4>1PCiQ7tN?Ye9
zc+ryN8AXDxd2xQC?!R))eLU4i`N)!VsN<Dg=d)>HWo#&>9b<vH=o;jzMc8nF1fYh(
z>1yrm8{y7Io5n=aCX=bunZ_6NCPM~#m0un+r^hh<#ZsKFH!EpY*O7N*1ASjz;}y{s
zWB^ThKBt!9uEqzQv*C?hxq7wSU4wZA@jCLJ1LyaMfDN8f0r=D&mC@RGlA}U>-%oJm
zdnbAPDNyryVBXhczTPN?zI*qq)6VVgi;M&-jIFxNkFfH^a6(?$+VG^b25E6&gjCuE
zdKEKGEyvx6xOz>yP_biyE$pNyLMBa0xpMETgy0jLZb4*<PGC@Uo`X&eI!wcgO&IT%
z(5&98_F3j*q_u(u<7{C=rAJPv@IRwM!@a;g$uu7n)tsoSTE$d*jDTzJYPMc(29`9Z
z8dW3E{bMQXfk~fg;E#~23O9tT2G;g3J=XwsJ%zDhbpYu;@%3s#e(we&;cSfZI@8+k
z*d8<g@@WhOUNi4PGX4e9tvbHLh3NC8SuS_I9d)dy#P|wPHGoWT(g%iWoSwcx3^p<_
zH_u`$vXCHN7Na1St6UzP%2X1!%J)xOY|6@NWuD5d{qj`;|0dn7f`(9V2(iN5?>Vi3
zJQ4AZBkjCCbDIARm@YJlStn?|^~mBS$#05$1WNydPH&soUrTS?kB-5Zjd&ysNX-TA
zautF3g&i5^H65;+BoUx$&HS9kX9XS*K;d+rI@xQ(`EC7Of6e;OqJUk$yP$)ReWWJI
z3flEQw%^Eiv^5@%ef~FTOQIKP3`F4+k<UU_2zIt(c*HxJ-)YPiKCkUimqRDE=$fJe
zoV?t0;QHXqYJ)%2=JL0a%}FWD&EGp00t{j1G-eSb24QST{td|U#`B~kFo~Fic8a#X
zW#6`%97!-Q%H*qZ?+g4C&xZ1U3G?VtbWAq4m=^1SMrT_0entTbOJuc4KRo9Vm^D+Q
zGPmMjO~RrtSS^wXuWL}}=af{nU1Y%oj-AA8bPf6fqp@>j4FyKVkhO`kwS3L|v?kCe
z#YNk*RY?!pt-rFf4E~4pzrS%xd=m|r#0(hFBRi@i)E0Igx}`{V8epK2+{4VRhdE-<
z>de{U>h&*IAVQ6+Kg>%NzzPJ7nmJuR?gmJ>H`845N5?U`&V7A}F!N01G4`<C*A<ic
z@7N7v!`UEdHtTz(HS&*l<$9X_BEdBtt_RFDHV+5b;!SY42<Y1U^PmVwK(114JpWU2
zlIETk*Ph=DrhV@!mB+zozqG@t+4{fe=ltV1M)}sHmZ9NUArF^WfkYPp>3fLpN1d_f
z{z=@U>Xv%>C+ji>T!vcdk+6c`b4B@;lLXfPR0<A8P5aZ1slnX*VoqJ6WG|4y>OJ!=
zBXp&1G`+d;>Y3l|QClE4x8(;S-%sKt=UabA`49>0T-T?nWv~z_r+Q63uINu+80ot7
z8lqN}S*|RQ%B~W~zL0xFr7>`R9%Xp5RAv>^YP~0U#<+R~^YDa8+}({6@2Aw*b7mzk
zyW-E54s|YkY4&sMGcQ+;U!L;X%*)o^t_X--+RqB8WhBM7V>Y6>OdES{4RTvgRIjGR
z>pAL!tZ&%#w}n>T{}+4`J(b(4ge|v^<f&PaMZ;b4CLX?=q~FDiW$`)c(Ce^RN8qU=
zvZ6$&b~(vrFj39SRhEQlnF->%kgMP(&zzPKZ18AfU8k9nCO=TR|BuMz%SJw~+ejAR
z#a50O_a1S4!T5l_4Q@|rXtv%Sm0q2qNt=UrcAOx*yQ|NlN)<9<Lq>q@_4?5HSNm0x
z1rd>tqtL(qhW;DQSMnT=9iToD{vu6ge;_6W^5<kn9(Y8A;Y9N)X>PlR!Z~Jxb~uBO
zCx!VD8b+~3>j4IX_F|an-x%^L@@Uh9OZT4$zzk9R3t)p|<hJsIJLEE5p-7i!#&l2*
z2;JqMKb*sTTr|&ly&f~Uxb(yix`o+Twr#V)3JFwinxr}Qo_nUfPL7rD{oZaIq3p=I
zVQ?bT#c`VbTvrZ6b1}WnBvB?6E+_U~gjAZq=-oEx+_>w1lMvLklhP)Jz`~y8vWl%(
z;bWv#JphdSXoILLes>r@i#h9H^}|?Q=gyUbwb^PahRaBO?tkuxlzU}cZjY`ee3@^+
z;p~t0WM#}k{rE(z$kd_2JM0a)xhkPCy?dC2>)$dP6=wirb}`zmg5gdkG{zfN(D>Ku
z74}x~&*ByKgZaDhzvn8u#B`P<o(YwAR&Hl0{=7jy1LswzWM2&zP|LvF$4ijlP5hI>
z$msF#X61e9!Lsk;bQK>~;(fo|e6>9QQKyv6#2%V_Ahe3-?n<^Z3gRTU-~fd30)p(6
zsU94-yYDZoeJ@gy4Q}JhzUg~TQq%{SW5o$CHj4m>KZtBZ+pj#9JdFA2$PfAa>2ah-
zY+6YUBfYszfRJ=0GWX?6R3-wr$Awf8-B<jq$&YsSj!s(oeV^<~ir&@(BWhMy31Tsj
zoN@_N<C*fusW4>wZ<A|$Uhcsc*F()iFPHZ%2}FXEMYR)FMWnKxCZ0i=$oct&so!Zz
zmEVIfgCT5Wa}+x5URw%AM2mmXV+!K$TZCQGTt&y$Xz{Y~-xEpv>%=H>%e=NaBfyrU
zetqW3g}sgJ3)-uG>irMb3uMcDRn2WGtT*lIr=nP7H6n269SZ%9?O95Y!+p1C3hlU_
z^d%_g6J>1V<S}2kWJxavVwsz3S4<`-CRlCH^%VyyX{Z=?08O$TsQCe~)|3MN@D#j)
zlGsk<F+?e*C2mm?x!_;2<hVRLHD2KgwjA-&<+^cUub+|C-MyQfI$3@jaN6BJHWr6z
zFB>-HSJ&#1Pa>Ue{aOkfRu7z}bqXZlg}XGXd7Hek=$bLx&IzmZEdD5y+(UZ+Gr}1V
z)hjyr$a6fnrJxwzYxc(V7dUH+&O#8&>f>~P4VI0+Fm0e*E<A~Woc)ma>h_hQVun#F
z?4A{~Ze|Z60}O6@3N8Hj{u%K+C`VqUw!Jwl7a8&kiUJNUmJYv?;}Z0r;TFNic${av
zd2|^yxyTW{59R^AKn`wGD3GR}mpmi{Q9Zm0Cmb_IYB$O|>E@K$PWX*<-MW-cs`_N^
z-zdYgBM!wVAyTI<#IUW~a=d{%n4H&_I3c>{RA~t^c*;x^eejEJwB33sZ#A#dgHKAf
zMbCuSuOWi}vYYWLENmi5n}XFxbo*9uxNtRx=ctl_EE(Y9opo_zQ*7w<*w*y?*y(`@
zs*lUY1@V3PR_mWM6#ak$84hu>K|1)PTf%E#g}&MX6V~VacCvLa@E$Nkt+0)GR8q4!
zB;|-G4jN5x5$u~tWh9=K<7Up0OYC*~7GQmwm#9KwhJPBEDEc{i=`qyt>E0~ouFI=c
z@*hEQ#1FWv3_iL`0^<yJ=?%hfyxRDt)OUzIq5aMJc`8)QiArAk#tCp4^wx(@nOp!_
zz)TvdzsHxwzHzvs0>au|weY1FwDP){HxlG5iHCDLGWB(^s1Vfc%fRCULVldOHb(Nh
zG$eM6$<3Zb{Af>!nGY$PM6&9#`k+80I8^P<GSR@8UAG0yUb>haWKnKp(_+2M$?nwP
zUO{jxFu6<dz;OWkB*U4}N8t@4jA94mWE2|R?oJpzUAc}KH1<~nP~|)y>2vJ~yxB7g
z8HE>8_&HnqDb&PH!~uG&NA^6z9Ag>ZLqG1Z_EvVTum7>|cazB9s{FE2nY2P?o<0P}
zPmgA=Ctrjtw;%D1Mn|?IeApz2ewRm4d@GQH%7e!<oO?Fhd4>;N&!Ji)e0OwOp~M?4
zvC#{j?CLm|D3V#48mP_Z4EQLDxl`hPuYWw-9!K6;dYdi=E<gK=TFxF_<uU5}pXG3G
zrydG?ygRySb~|87B}sifPOzqiC#LSO+u<@Q>kc^{_|a<9*II*Ua$!O1i|6v@OQu4-
z&RQ9g7<2-H?U3(wJ4^z?kVh}Mf%{v1)Wn>7*v2Z!e7q~ako)jDR*u@b*zp&T6t!7u
z_hfrG)x8JCVNod<y^<JJLK&675W=-U`7={=yBIUp2(AVXNh%my7{jSbEY74rmib?6
zL;T!8B16~8vMn<2{CKZ-G*|OwiBC{$+|#AObzdUaMqt8|)z*`khF3j<kdu24I%}qb
zpS5!plgV~)R}w^MPtp?tcZ*!B9s^v-&Pt6us&$yJUa&GPv>nVbt7`2NX;|No#!iQp
z6HSEj#$v0I-|5$xz$^awyhc@U`8{Sh)u=%)?{X8lf@VG`&@aK`YVB=?y)y_RhGk+A
zZYNB8iO`mM1&^5IL(Ip!BSQzSb9RXFgfR#Y030|?gm${axU5zY5|s!m1VscPDN|a`
z(a6VZA`P5F5pGR=k&DWgCX)$<Bd=P?Eqd$3KN$f(>@(K5FEt&5y0@lLS}i*N!~|Hn
zxp;|?rVvh_@r1pLjU@ib$E^&3KOqtv4hhCHhC6KeO}X3~s?zPjh;6blN~z^FwW$Ak
z!?UxFf{%3J6i2HZWuwA+8ewhS$9p`YlkOKKoE((gS%Xp@J0C9<P7ouNbpeUT29<-m
zZUiB1R4%-O>aE+$5NiMG!{yP?N=G-f_g%p%BeSq;NLLP8bOQw-R|mrmPLrB&z+-S2
zKzD$MLCn?}*E!J1n-?zbYp%lcEm~@LgoLFk+xNlUQTx1Wa@6hj43DF2n02elsp=Xh
zw5zya5OJT|E@@3CsMud4kc|<<nUFhMz-UbSh4@RLc|rhJZkWXn!hf)(P$!m(YAtHM
zQ-l4BmI;?`PvF*M=7$PNvd}#O6367rzQd?d2nsO|Jzt;?Cxg#!`qc6L%2kZM+Tt?X
z{#z0FOHi)EsT9jeM+Wv_AbK}@Hku*bUgG<=t5-{k%KEj-(1UL)$#umg2VJj00#e{C
z-sX3*0<^EkN?X6ipeY1WF=4wO(!#}mQyF|S*e0|*b@jEJ#1UQ>oSc_;I)}U;D_}bB
z%R9~mY`rsk%kFU)F~lK%v;KjDh7y1M;=4&pO#=EmIkXTbwVwg=fm|bz7vheZvj$`#
zx<|ELU*9X7B-T_%-!kM{O1tYr&tW=*N))9<w?JuzHAnb|rZgYKy$|70NFcQ>x!`AS
zQd_j-%N>Fed>g*QdYKdCQ9M#Hfj;p0GW`gF_ork%Il_5}b2r}DV%6PD**73!KQ>iL
zNnWz@y?-zY?yDm&ZcP+$CG|b3soG%Df!a`t>LS&kHhV59LVcH2(pTY0+hIG2&P92|
z`sdn>znM~L5lQvVEJKRj%E*wI`^d1T@~4f5KzDg>1YX5jPHrS8f+F{nh%IUVnHJ}6
zXQj5{5VbRfF4>QcJqs;g_)ja!O{A%QVm()x<1(h5MXQWH5%11l174hG(oCx(9Ed8j
zAi3`t6B}@UqfPq^)eZ?>yP^8sh8?hK-)X$ITntO2Vh(yD^+38s%ypRGutD(BO5cTa
z)Ztb<@h?M^Lug4ERXTK<|CXR!os7Cxrg`4P>#qd_Xy2osqIcQW%Cgc)ynG63ke2Vj
z6>d7Rf1HV=Vmz?x)_Lv7!^&^UA>(h4;=nIoNw@etgV3n9%>Docp#R=^#TorNwmQaN
zG|*)paPbMFlkCSH_S|smPTB#HL7hRYZ}pZIHVP57QdbD-V_x!TR42dUI-$Bp;z+3L
zYVGx5iXh=~(g}y}X^&S5p;IAM@T>HM>HBwcw=B_rX#*hvgO1(rwAD_>=nc!CX!Zym
z@ppIQ)E&?z96te`%!)^iZhHIK{RT8yljK9qKtbVL-K!x1u2N1!KW7yjR~8j*{r5ju
zN|L3bJc6p!PubrNEe?@3B~{tl#0ij1adjh5>KM;7&_C3tt>9z|J}OxZ%SC~;6g*$r
zT0NkmtQ4kRyzjQ}C5ci3u5F7}zdDlIC9ItU%;9!K4Vnqgux4L9y!EvocwzTj4dZMB
zEC{R=@LST@Lp1>`H0@GYFK$zl@>(ayZp+QP-(Och?vq?}5JzQJ$Iy{Bps2@A15dMw
zhl~&*WYNa8d%5n!v(-n%w7Fyp-fMo2y54u{Uf^O!qvvOW-BRx}RD8e9A8=VvOi}cn
zCT6D%bt+P{&0|CdB*2OB>rIo~IiiT^j%oW5(XGm!aqBQ1KH4v7oxX3(W26~6;%||=
z2~leDWbx0I=Z0^cW_)BGf2g-xZrYcUIM$SVkDlJ1)dg=_S`Se29ZH)UEOkf4?TL)A
z^>S(8J*If)W^Tsk5PNs6kzmKjEpU;%-HTi)@OwNxXuF#207QHh{If(c>`Qox27K|k
zm>kkP8FAY>KkJR1AL$&kZ}AjJ#s`VeWZi|A7_D@9g{>fy^LRr-UoqnEX2kojyUhr<
zRb8)J&SmZJ{Zn2qb)AGrrd-N-QyjY6iD5!01<)g~Ib4M_SsP?QM4To+Z-_X5bXT9u
zX)iZdw$IiZufi`jG!=u^MoPAn%OOu-1~-7QpdP2s7h3r(FRIgz5!Z)W+0Ct1r>^79
zL5e|1Ix`^xWGo%mL9PuNl0y?$F-E$h{g1d^zQBds-|Ny7oJx^26kO5gyUJrs#N@b#
z#=u5FEFZ--h=*gtUPO$aIM;}n84&Z^Ra7cP=21P+#4-_-E;X#y?$kGT5HKY!l>V-j
z`PH$E5X?PhHtFPM{2x%g1aTS^vpuC=z?xbQpoYic$HX^0mr~%ttD@7>ji|tjl~x2D
z!jwzucJ@BZA-OrJ+O>+OaCRuCeGzXtWJ)bMGG#+$x9GmDVDVz%yp(zS!7l`DwL{nz
zx^ZKIM0lO3NOQZ9cL<^f>otZpm|6Ca9?(sva4=TMt>+8y@Hliatx7F;dK#Yn2&WC}
zUipGnGTl4M?`rtjntu}X{0K=tUK>Ns)7iRj$78Fexu!%U#gl7DZd*?&Iz82zZJpgY
z`X}*jpqt}p`iZJ71u*v2tBA*Wm@nHtszW|V*SWy>wG#fdCv3$?mEf^XjByNM>Gt>A
zMCM9_cEn3-j}tl@gtoR44RzR+DCArP!L{Oiojjf6kj1;@!-#P3*|x{aC2-R7cSJbg
zqt(#C5s3L#qDAqtT<U@C-Dqoy_si5`OP82rzT%DDxkm9)!_^t-NFe^aFS+|dbuqoy
z`bWR1({sHA({)J`sG|75HZ%0AtmxdiWlB|p3Hqr(Z9E&486l?5KC^vv%;7)eE}t9h
z5|xMNn3}|@MK`_FKT4)XEBJaN${1UZND;0(TY7!ReK$Ujq_u(d(b%S|9kZM~U7}N^
zY%z^yJuxQx`l^%|Gu7O7v7Lmt!&ME;S-!m1B~uGLt!-d$Wg@-StL}fb_uJ<uzM>${
zKt2EG?hVjoFBrLI**M9*VKy7#CqrBa1!5V$-ML$W_3(G@Y~Wi9vukZd=eOoVlh{--
zrFMhz@;0P5zj#<6D?Zg+;v3gA$%@9xPY?GAZC7JW%<EwPtTIDAW+NywY$2Zs$mkOB
zwsP2P*W-ma4<s#>dCi2!OM3{ywnCmvb`~1=gLuMX*hBV20Ttxi?N*`wVD$OUWxTAW
z%YIlc0*w$Q*l4+VGav5?k3lbZ&&#+fg_wtxozE21B5})J-AIZ935O@iAT}fc9iU!G
znZEB=afh4-eUJ|-3%6sFc&62UX`e45qx7;BgsJK4xmh%`-SzNuRXIy)wu7JZc1#Kl
zEk>@D<qssv7qT8&L49peTv(Y=0@sD3F~;k2p>uCEAD10DF1dn7k7!^rk4jx-B?cek
zr|uRGmH}$VT9m11EnBGYQ>AQR13a5v)mwiY{8jI%U-He6Hl^j%`{P0LSPeQLCE15B
zhM#yug<(Zif`l-slMpOlK_M40{Zlr9`^Xyy(QDe=?oGKu1sjoqOg)&mI8EM~fVyYE
zgqvmS9{qN`(GG5$M}*T%N%;QQespwt+?6P1^X7Vp?`WIVj#`kN-b`ceuewFVVtnG8
zrqB9VqEg$3_u=b7#iiN)V#NdaPk5ML(%!}4ANgIXAA`lNi&B0y|GZK{abzL(>f_^b
zKSNCC+b1>#yHDg<cd-((40J%Qr=SUAS&8GfREu4c_Ad`p?()gP2-o29)bp@T&w?%?
zXQ?UpWvb*+xX}^aRFoS$YEATq(PaH)9KE*4C8sV#EDrrFzHFsfkGJ&rWgB7r6+%7_
zs2o#avrC{L3SmTru=c(>2P(IE820%!11_q%72yhz%?EJ!f{7JHNt{QHh{!>5uH00n
z*~9a#$gx7c3Cz(tGxiU-2!!YS2{(NR(-<wN_(dNbh==wC00_qLNyS)YXmvPFpS(SH
zmK*UCAoZQf0rejb_JAX*SDgzXSzmImd;5!5bNB@?d3{<s5j$!>4{eLxwNnuGwVSN>
ziZDHr+YN*{F~%AT$}(r^yy56mMOQPh__iEYj8N2WwE$1`qRcq_U1uKJu0kJsz#8JC
z`A3TO1GHVs>Qtvh&B#}~a@HT)2nZ_6lx*9Kr${c^2#!L#gJE^+u!8ooMI@`MRGS9B
zpoaKJLml!nz(d@RZ3aS@z~RvmMx4mTm5bWVw<1|>g)t+ynoLt%hV+Wnd`xysu4v$>
z+@#!Y=|rLrQ?GWs|7f_2J>N-lupgfosOG!puNjE5(P`HGAiD*oNX)&NI)*IjH}Nsm
z(a1|34+>CUvYwXa;kE;*afn*)N_E_yPm^IQ4T$FU3u#Nt4mJn<8l6pGS`dswc0~u-
zOE;KK&&*sibhhiBG!m8rAF{3GJ{?4D*6~G-o!~Loz#J*x%mXcA?R_>ktF}!AIk8>(
z#+!ft!v)*^q%Qi~+su%N<4vf873pMu4)Ggr@gwx(w0qi|Mv+qTPEBUa8I-)oKe5lV
zqw1fWGydB1&`jbP{zH1n3_`y^yoUW$IQi76{@~27@A~V}n^y^Ol$ZpQzB)MM>H63r
zrO?@O#gOU^dvw7G_YXI7IC|>n1qY%Zfz{x^$prr#it*N*3O!*}sC16ic!dhjG#`~J
z6_7$m+nxihUTUfuosQ$e8cmV&JV?fZECP5*5E2E595=Hvet?q+O6xm5_m%6_O)JWX
zqt)ZF+q6aC&CsdnVESvf+w^g9H5(MsrafsM6ix~l?x;PMq?#=HI@B<W(r*51j+MT*
z``e0|C)na3>;u#oD2CgV8kMXZQEb|3#5pw%u<mvf@Go;RE0(`rEGt@V`;*K-bkgBO
zNmKWG@`9@#=Qv#MU=?oHi?B`*lWM@^6$ji(jXgVzvbB3?!1s3+6}*D1>NgyIL`41L
z@$!6$JZt+k(D=_OTlr-nVO;&r{#(c89WJa_b*e!ZUmFI4C}>Ke*UczzCf`UeYx%bi
zh}SMZf6K2vUNKDxLEax7d~)D1{A>n!GOk(v%0zVkvE8F%oP(sP@eEJ<@#7k~pT&Ri
zp>Ad?oiMUQMzOow2;kEVBm4JHP)8|28(AK9=?8mqV~D^76-MIks2%v%2V|5_;-Ia8
z3){qf;kEfJq&A9<_Zc+Sw7Xe^lc>fwa-ddys%$su*Ak%DyNY3)-mW^;#YkdJX9FxK
zqF0rnb_*BEImnrHZ(}fd=xt)zS$E>^I;UWl>mI6NMyL$1TJD+-;X~5&+iR#IA*{2*
zoSe$#26kI=&G{Fh%BTA)a=5qH*go{r7MaqS0X5a*uZ>NfByE*9R`QlfvXdsSusBfc
zjcZY=(_q+d$9mt}m6mSO;09q=6B3X1b3XIyav}ZzfTIyb|NEO-QN0KDFO?YK8w#K-
z$V;r|Ct<QTQi;>M2h2*`5j&~YkH#CaTSv9sJwYbaF9>k&q!AsL26QW0&H_{|NzM{s
zQi~3wm2g|kCX~RHkgwmQ+V4^VR<(bN!Jl4Uj71`5AuFtu32_&hwVo{|5E9_@tOJP6
z{@8^>j^0{BF~hFtZX#S$YhHl^>>2fu)>6@U437a%CR6-u^y^lU>h7HDfI)4|I<>j2
zVe)&k-`2rh`*AUBEX{ONwv%=4`-XOMe+K@Gl;<!*=Q5=)mBu}hS_!ZA`?M3m^Zd6j
zb4pG8GN6hcYpj#fp_Swmnun%}QKD5+N{D|^Kr}oQVR3*+kKR?Dlq)=UvKF9o>X5j~
zu~B0hsm?GO{}UWn;m1Mxc)2@6<O28U4|?b4TrbNQd8>#5c3Flf@|&;^*na|fvGGL(
zx62;*?7W>8ecr3zT?J0f1MQ8fxvOcJBGLSF{0=8g3C5@mi6_?5p15Uu+zmtyD*IBv
z1|p%Z{DXg*I@g7M8zFXXU~e_!U7=W3_*7$R|GKu-4&ogeZ<-?)>zEjG+QAWq$+9a4
ziMO=S4X|cypak%|+V-{5?>0kiRthV&2*(#X*Y)}_F{<;|X!5rBuUe&vAWJzsI>nJt
zuy9jZrWtl;)9bhPQ$KX!vdBJ*ZXbcSIqp_|4M|rJLZzcy8l*@)vP^h*r<ob)BRXFX
z`(F6*V~>hsi@fT<eN<bt@M%O6z!#9G#bSG#a<roUX0U4F+d02gaa~8WyWgarom=aB
z(5D;MWHo-7Yi;&-u6FK{XRVR_u|O=ts>w4WTo9Nk=js04LC4M7KDw88WM$U<mi{{;
zW%w0KXshyPLd>OahZ7;~U+F~YAJ-ROW-?)-HgcEnf_HQ+8x@o;s!dWcqgie@V=&_D
z>%f@X-xTuZyBlD&W+VDtsJ_pb3*HN{z}Ac7$x2Xp`H4`pUwg++Cb&G}lS`A;^q*bY
zT?9a%TCqAo>!++#?XCxg4_1?abfuTwY1#f21;MmB$dpwit(Ag`c!E__0aLzdlK-ou
zi0WqA?}1L?L#~cH!X1t;9;1e%$IJ~g;nz>;3pydo&e=P6B9=Ga*Y0=E(`J9#3YZoy
z;z=#rxeJ{?h}XH)S>kV&mH$!H)nm}JVm?ePfY%V3+Pz~aI?X>R3vrMs9g#mtd(7El
zb{QGqmnEg?jZl(H;LqMIPAc+YtS|j@cAQQ<6(>UIcNlx{J_X<^6~Ws82>Bzb^){6L
zw^=5uXVzx75C~FhB8Tn1_juitJ+^qfvElR(v54{}wB1SH<%K0Ji^bLcXp##$VqT_r
z{U=7t2wGMkubZ%%S(ivS;F|3<xAI%(NQ!Fjm_M4d)ymo6d~GqOV=DQ0X8H*uFop6)
zg=tIfhrzwGtXYND?M41lcFN^_M~6t>oHHYrjvB_EBLmf+O78~K&GT}su?AB$o0ghm
zzvSf!j}N7y7zv2fa0oe?sFy{6{mmj$WxLkR)#OKn0gPE&G9w@3$$b5v`%4ctx=)#{
zEiWYn&Bv=oD-0eME_$A;ivyCfpf3kkl@&H~&%uyw>*7&Qx$`muQnTgkx#cnoH$z*c
zj@Y{O4Xf5FY(IJ#whi-V>HXb@LtUh5u2eR5`cU`lz)`zYwZ!YHIfun}xa@9IPB4v{
z4X5n!dwwAv88ZjU^ZBXO)$1h0#?a&Vqnp89vIDAt;#twfd_QR6yTQTC%UgD+{Uqk`
zohbD7nW#9iUGzB5Qi8ROelA5Y)EZob!JcKgUAVM<9b~Ii;c4Ahwz%`$x%&94RllT7
zVAjCT-ocOYf-?BGL45k~VU2!GHFzvpX7Z@(N;~mc`<!NZY{*>WxLM{ry)C#1+My|S
zvhXlxbt0<7^4&lT4uSgg-79lZ51XgMA6|*OOgFhm!N%{KFc`6&Znzu-i(4nJw81Md
z`$yGDM4B0}dqqnfldK=VYpuN=WvF~QYl*IiTtVtVoc<*nsu!NDs7_L{cl(mG#^t6t
z{6@s_*FSt$&a0j<X<Fg4L+7LFQ9T#Sqj0Nj>a2NDe1T*<-%8W|5UR%aBI71^s0L^W
z+#J}=LYu-XckT?qwl&PNUu^P|T@5lvWDo<wPI315M)-h}2RZmtQELls;$e@bV;0k%
zzAPOBg9tdMsbwfT9@<K*d(XaOx;@ZnWS3aXV?a9P^Sl1BiF(hp!9$Mu<*JU$d3$IB
zC1x>6yK-c!Bu!$!hFU8MS*UqF*;i+hAzmuovpuHC59V|^$jaLOa5dlqc@eR#WLKmT
zu?z^zVc!E;2o1QrY?%AE<T%#WypI`M{<5FhM5Aq{uz}MsZu+6HhhS>jsi<yIMC8LH
zHOYba-${O`kei&@b}PC!Y_p<l)1Aa)$4+=;YQo{s8LjY|>8M0ocaT>AIL|ltjZRWR
zQw%AD>#~!A2{DsmYP1w}-ITjEPz~lB`dCsLl*@8>jLxfodes_Oh*a<H-rxN!rnf&l
zL5cicoVdeB*c@WbAgi|tDoSKTZL{r<F=`W5s_;9tcJAOM6J$FfjZbQRmwLz=amFPp
zkGtlZ!#4LSp-=?D0^$z#_FqDkC1oy|EevpaQy%kRz+Q8n-n2ZUiRC!z(0csQs{>e(
zndxh`nQ2svOts^%>U7g30(u=(Aeiz}m~TY0$CkDdCYPbVV&i~3fI8rk*%A|9uDLKu
zCGxRyQ;xx_%YtWhIriZ<x46kuwc_4ywi~3F9QPZ{^4iZ)!^2BYyG|!NcU<?tv<{Q+
zdT;85MnKNC`hvRW3G`V#62|^;Xph<8Pm|X1eb@JLT0alwCq&JAQ-s=AHlpTc822m^
zo-eAme>Lk};j+sTMSJwHzQX}DL@-cUB_+J%D~$Eflb@g3l@Mq#hCk00;0?N!sx+y~
zZwvCo#bkhfjR960)|-6~kTV5Ah98&>nD$Z)CF4eXH;U@TtQ!x<qb)&NwZTp=AKkGl
zQvd9VE*A0>$7F+Twwue+Z!Uaum=?m?ac3q>DQ;T4(l!Qt6bCt?rBa@xNv|@_)L)8t
zAA}yz_N$CwFqrSC4|`P-ujKjS@mL131V^K7V?iEcmf^~$`L^@c_{x)$E{*s){o0oA
z*6UlyUT^(2V-F**WqlKTo0@g(#khN0TXNVI{0FTu7-hp-P~?5T6(lg+*lnbZo;|BL
z7)?@*k1-)4Ub;W>3@CIdG_{MdS^~q<ka-UHYPX!4T9$lL3W9;x%k%}94y&9^VE4T`
z6*RRo7oL}VP)t-Ci}hJsS>=+~&$-DZL64Xw5q4{J$-Trwg_tC7Q?(?aQX{DH9-1sZ
z!;FhTx_qrje0G%VlFQG2kDc>8dqxqug~id`!f0dy(oNt_)80H{<LdSLjUJU}2RLv!
zev>U;Gxb-8?!eCS>X(`0jCQbR$9DUcZ;F?E_7afr*sbA82c4?QYUXgB&^(zR9GIq&
zB(^iOQ#>g7o1*F?pZxVV^HuC)<za`^R-?AThN`AIZHt)*#q}Ke43?6T$HaYK2c3ts
zBtQ1^QsIV7pWSWV<!4fOaX2JV`=u7Y#!Z_9yTajP0NC#x!LMH=Wh@O$wu46%UmIkd
zD70!m25rVyxPZ5wzB=G`xW)|~H5}_QdwRt*-Fl{F6&kx5iQVP-dJ<*3Y&!$P8RbgL
z4xf15m034#J}1plo)2ymJG`G8IaHLzoUt9^m8Yj-xj5sG@)Y)7x`8gMrOaj=eo$|e
zB&5u~TI6fN*?fNjI;r(IS!7v>{YAm}(4Lm<Hg_T@p}zw!c0VkLU^GYQGI<bxoB$P*
z8~BxGTeadtz2mx*F0ad1ZR7c6*=Eu{K3Q(%^BHLM<cea2EwP#2Hm<y|G_8Sa&6pDj
zi+nJ{b7y=yfJ5ytE>%|nGEsOHZ-_y-O;qY!cA_s@IA+*-?G5^##yF6~F*d$m-My)t
za$d;x<QgRjt_Zhox42k}J}Mtb!A6{G>b*$##3AofxsA2eaGd=zXK5^KvQxB<;hUEh
zy?gj>P-^&xIfu3`7}8P6QX=Qpx9t)a$SD6D9L+NfX`A|z;#Xhaw)It3F{xn$_p5BF
z?0J4I1rI4hMck!%b@lW$WxZKn$OX>*o@GVTpC__DvCD+&x$MM81g*#Tf{E-m&Z4ak
zwljjyI`7G|n4#N_Z3CK(ysOS%7kpT%M0S0GE!kUm<v`A$rzNW2Q%21P`e}+P5V&m*
zcM6kP9n+#eNfS@xL(X(Jjh7Jfs)0J~A`L__nOjMuS8xH!oR2#5CP1<H!-=5p(Q2ya
zZanoYHYw&O^cZBN4M(0Yq0piS>C2#b{=+<qiAP^qTrK){7e)GL%A=sYbYGT?v#ENm
z<f15_HJunYFoEdmhKFj=)lAd)gGX1o`FbcLg*sl&5|sV~9LmjHL;c_=E8jde<Xkec
zl$|wmngn@nu58iLKc0)ayITw_Zjdotb^eI$zN#tL+(%S)SS=!H>(zr>dKuNCUy?>-
zFw1UT)zV=}N&3-l!F$Y>{}M_*=#^q%s0QJ*?5nH66)xq?+?}l9#^x!JTG(G}m4DCE
zd+7Qz-@F2N9pJS@AKZ|*4OoGG(dnqa8RpX>WT+5&N!20^!M4dvkpE5>cpxwkom|ov
zh~YFOJDBn!6UJUh_&HH8qA6XmU2(E6{+tXl9l6lH)~lKr3XkUB`L-<YtY?`nv~EPm
z)Yt}9P+Fu^k<9k|Y$AoR5cMkP!PnJHO%E)(BWuAcY?u2oH)MI~rIQTSs+h0ctUK5t
z)OANb03`0pX^_Uw-J05iq+|NA7F-sNw-+s(&D2y$Bnt~Pj_%le3~zv9=TCUdBbDr$
z$Gj<WhTau`qmzUCh5dIej+Bedxvr;^!KGM``%c}N#pxx_VW9@2$@4Zb?RKk&%56)W
zTXE~pmG{3r7Iyl2&&K?mf@3Tznt^45k`>C+>=~RkA1VbE6Rw#t2`@g#PW;$1eaJQ~
ze}8y8xorDP|I|3PH!xD%1h8!>IEvSlndFy#VixOM?9PTSjIDduxHBxV-`is*NKIth
zK%Ta8eckiA!`1)U3>zEU0r)~Far@BREC|_#hl9Jtvaz!p?kDcpIrbYm)PXIv{$h#d
z4+jT_Bo6y{1qbT_3skU8_*eh?jR@4RG5?MR%Y|WwhC_qx|DQDe-;;q&-2Xd={!iHc
zO3tt*Z=fr%O<($z%Z<00*^dbTtSP7Kawl88>4L)&=yZb~@8EhPbMdz)BeHvrF;E9L
zH!W-Z*m9ixZ&w50g7L`<JT{(JGRcW+Caj6UMN-?v+{DE~(9FpK_65hy$<52g$<4+s
z@R{qqAn$uYE?yQ+PC-u2qc^tW|DOT&j^@@t&;S1cnKDy$umNzgl1iT{C5!|AACglA
AR{#J2

diff --git a/docs_old/logos/dbt_labs.png b/docs_old/logos/dbt_labs.png
deleted file mode 100644
index 3e52f4707dfdd972597f22af6360e5cba2d843d6..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 13418
zcmX9_WmuHY*ItlDLXc39kPtySMM6SCK$@jHBp1b{OIjqPyQE?1j%5iYq-)8AAJQ$c
zEY17ye?M^T#WQne&YU@O=EObUG(ISj5IrRVfj}h6O7dDD5H=D7!XhQW1HM!wX*>ab
z2wjy7+&~~^vHL%)cgg(bz(*`MEk#*S<p}*Y@DGlijG7DxR1-^lZGj5{3EnBo%jkGw
z?W0gGI^guXqs=+oUbT%XcQY9<>9K-B<m*(zw1mfY&n7BTxjN~iLh|GIz=3|awDvek
zet~YdV(Br0b~+z!^NPt_g&dCEJ4dMMjV?uL<@Prb$VvfP1afdm*Kz;mM#<bt4*RUK
zY}p@Kmt81;b#@VI!0-5tow3=h1yi^!C9&#X@H{<o-XmV$$TV|)?2Th7NZ9JTo7&ux
zOG{r*AG|Ljz*h(rtA?qGzUF^kyArTffhNHMfhdaw@OW5<nt$-6gyRTeM92vJCK`z!
znJI}(vs>r?@s3R>=aC;(5tYXbqM};+lz*CS1>BZ$gFxob-;*HE7!eTV<l&ZFS7kh+
z$#4ymq`fr(EIM6Furbsyhqnd8GqBaloY6-J0$G9ISC2e^_EolF47KpEML5mDq28L2
z%}0yh|JSVhv6@Fx12LkzVXa$~y*QP0Rs@RUkkmMQ*NF!*e?>%y^wRJ=^DWFR+>^No
zgu|MUU|fRP*#~G-PL+dCwXZ;++{Y~Vjz;IdAn)6`5(tQ@Sj<`U;e7)nPn@(ZpJ>&T
zp6FCzfjE|wy5^~+;48bX6v@Ig^p%h3$%FWt6iGTOxA&{SI3Unt+Y)nBmSr`3W9jCt
z>^rlU>?NYxt1g~jy=e^PKp>8(&aPIPvcti(l=`bUh%2jvxiR~aqGYr1YxJ4`T_6^S
z7WKAEyjBBFcQeD0<>%jpzr*&8MFa7(mRVbUgt8Loptjs4{#OIO5H>sYZeC8Kz=aXI
zah25L94$_|@)Q&pho*vYQIwhG-95#P_!;X>9^wcBDa6xb=s0Q0B<s8|5bv$zgDPS^
zre(z}^<U%reMyAV*#j^q#3z8kPK}rf{e|vOl5bw$t~^i8@*(4Anq>0e(=+f4u)K$B
z{pMNVFKx^}@#?wVj|KLb#gwXI=NUt$5)I8qgh8NSaUuPMB;u2UI*&q^EkrL*(r*62
za?$LR0Ruh=bb8onZZgoZ(^YU#?wSy3;<ul!2xo;B*5)-KM}fn&HQ1FwgT9EL64mih
z!@O&@Af{nh*sS~<YdCOKlH)902`xEOwRK)ecn2t--yFEcJ<k#e&y{-IF6*Pj{pG4;
z-<xDXDn?d?0BD|PgqbAi?U+<Z1yx35wJZ0RB%AIgFbaykm>osw?^IOr5Q9Ks6OLW2
zjdaxbJ)Rwrv(t+E3(q(uq^dy4nKN$!9PT?z?MfocQgHh0TbbOOW&$L+BI3Dy1zwCc
zGU^4eLuIDKR0%<WR_!o-%*N>w&BL(U-I6gyN$d>ogk5@2U=O@Z-^0n2ule9;EA};m
zq!Vu}^C*OYa9U2eaTBls^qf`toA4t4c3PFc`9JT(-=f9Bd|!#6&n_W=xamviU~~2Q
z^}<a~{?Rrk13L|(7jphm<d1{ZqJY-21)`tx6p!sI7Z{GWQND?Ey?+9RomaZyOW9bm
z(71R~7D6nLn0!^=ycFf*s)Yc*oJqPqmc#rr_(AJFnb>nsa#t!n4ye;&N-8IVYGI?`
z_wmK;?ssoTKRJzN(}i$|odeMyo4E|{O@uZe@G9kB*$p_W|81H{7f2T9qZ(t?4>9vn
z=|RV(wNC-n!7{E!7Mc6fj)EghoEc-E<cQjqomM;mxdw*N3=r}g>|`y8@zN?u{qBto
z--{XRX-;$f%K-x6i>;WldBtUBX047(TS+V*i3#)5ssZy|l@LSO2}1t2d7b_y7*PJE
zlm;(RP;S$VE-;<rbu(OtoPdV@$tWiMdg|A;j)p9tfdm<cHMbwg<Fn3w^g**3{w_!z
z+cUD?uF1zdCnIKI0R+-XIPi*!8Y(xnm$lT)IXAH0S|8r-bhTxzQzB<G$BPXDogGy$
z>Ey`j<xkStUq#5r!vGUBtJp@>y^kQYtY#L0);LT%USWvqKFdP*_E{)0+X+COUys>;
zw5KF#P?Al*uMY=4E6gJo%(7@eplbCAzXZRR5sJ9;a^B(X60ACk3j4%Z4zM|=;d4eU
ztHTbT%9kM0sS16M1sMk7NgB(<NVTlkJZ}_yBzw4R@Sx1KUnxop4!<n|39km9o}MCM
zY@Y|FHik~;M`Iaa@1xK8vRVQVW%WY^7nAw;5;XTh4VXx3Py^qzzhs}mt@&o_t>m9&
z(!#guvY8%qo~luy;JjvWd5X;Z=IeC%g-39oIvXZsRd1Z?(!zIB<KrQ<OC<;MDKO>)
zk$(BR?xU^ERq3$L^-ZV3YRZ|+ey;n4DEe%Mi-H4Q*I`YZ<I;-wM<S!Th<Rm1mD_C^
zW)DrW_Yzl0Eh4~vxYJ-q3~n&}>0m-wDil9>nG@iklG~c_cCEveg$BP4&@RWz70uUG
z;r1CxS$uvt;Ci@mq23Rah2bUf<jV@C&aMybXIy1S-SNZ^?$T>#<TtCAVI~(SgYjSC
zyG2aarQa#KDuY15#}=1Ubn2TZ!ErlVql%#bKFFsMuuW^ec(Fr{W*S~(-J!uR?n&NS
zRkN2F2W{;?n~*7NZ+VPG31_|v8k})UV9W2J`M9~C$iS#}Q!K;>c+lEu(%4jv|B>S_
zWiNE&N;4O{ETIVaa#Kq@TTVgS4IN9x3HnI&q{!p@OMM#JA`ce*2jOC${!`>`d6kZ$
z(NUg+KV57XmA(<HKO+x_%(=M={eEhU-;atSB!47LJ@QnfX>!9tc;2(w$|H3|)plx`
zXzJfCNVPjJR_KEjOf0EzY+lZ3h3w#WwVOIQQQ34vclMJ(?w1A;RJ-w<li_)mB=moO
z<>m03P_yZSlLz9tjZ?`=CFLP^GwE%4Z3?jgOxJ&?$V*Tb;=7^SOHu3dxdJ31oBfKp
z#HJnIVlPTsg=<P3^XkKAlZd$v{<L$Z(kJP9^t`K9&<LX?21hFE!aP0%(p%YF^69|7
z%4Lcj*Cm@!su{{IJ5CROa~W-MIEtJd$e<*fozzx2cDfR@5L{ieegdB+kVTQb^12a2
z9}xb~Df;Tlrsy{T;dDzX&j?Dden(S%xUfmHVH0#czIjk|(7slDW3hMKW4>s`m|ESm
z{HAF`wqbdyomCh<p6$K5ZFMKiXI7YF9QyQaPP|n>xU^7Mp!N&YHuL?}kknG7sw6u5
zQg`x5<@H0nyfJxyZm41^-p|iY)g$HMIGN%OtvhI|tLw?O1fp5*D=wOU_1v_Kh_oaJ
zz*5M)?y%&3gVhJF)p%ojLg`HAWif-Z8Xx8vNDj*NzP|nHv2^Epl!Y19-C0cR2!h6p
z8$r*?B-c)e>jV@o`_Lq$FP~6$jDP|u_DLtTi<@q3i$r_*ef%>oNs->0>dwjQj-&W{
z;Z_i`yitTN>d3ns8}#<;_0UCG_%`aYO5X>W=MatBXP&Wncq|oAG@1DQ$Qu+Gbk28k
zmgY9&mpVe>WsK&7F8~g~@{~0W{aQMFl8$kq`^7rrbNT63i4Mczuy>$Ng6knK#v&A%
z^z-AIt1&H;?o*dwU6^EmCq5Q_m5_-*8`#~4KzDbdK&KoF)br2cvH~MzxE;D}IGVZ4
z?^@0N-+ymWfUPWS4A*GjVOT|jKo2~h6fNKVgGi~C&+D!ftfk;p>AwMP1!Y=9c<^3*
zcp_hRY({H0zbP5Wl~sCRwUcjn<sH5_bI<WAMyPyhPv&vao9x3Ab?M~2g7S#ULaoPU
z4gQUi$!ZDHeNnEUPJby?b#=(F;qw&LoDmxy`Enl%4+}%r`aDh~!z#uVX0p%RILo;<
z(&$0)(TMLU!s6&`#t|D7c*;3xH!{3wb+qJ}j!|vnkW{zv=99zUr7!-Q$H=a7ym#93
zcV3QSUPj?JAea*P45`zRh2f$@UCx?JD#N@|m$)=l?2*sb=*V_7nPN7>=dy8X5q{Fq
z*=hn16uPw?9<nVPrB`zXEVFCKeL*gaadZatt7MrVEUxu4x(ozz*vt}STq^I`4o?vJ
zTz!CUuZ2Z*wTr_)nTEwZx#{XLZ)X5aC4U<sdN1t#dwzVu?Bb%k6cjvXPv#}ziuU%7
zG?x_X-NgcF{FDuXj>IlM2%Ds@mn@het!3WI`E&2o6ux|=FX&IkvS3eUaBf{wwVd<B
z4|#LEigu}RV*%9XTV(nF>y4(c2~m4oQ}`=i0gs_WHjcA<<pCvDj+tjcBDGd<>EC^9
z<=R;q-)A;3a+SEBl#lwi9-tO$LaP{!L>He3X=B@E{tb9_e}j9y&LNB!xZ^lXR=U!U
zd^ozIq~ZJ!vbi_e>cUj?F;uJ|P;b$BWW!$6vK0#p(af_^jJHoeZ;;%_jQd%m)L(Xq
z+v!wj#=H3RueM9F*6FzNiBLWO@#z)f?jD>=u8@-@_E{_IR1%$?tfKG5-Kn1ht~q<v
z#_q)PpqJYQkQ-<f>`;9Wz1;Q^)T{WU3Ln(TnrKvW*1C$(!r~)+Nk@$ZN{yNbhD!O5
z(q7<3g_}@<KzxtZ!LZ*=A?HL-S22NF7bM@C5Vk#hKwu(}Y@TIx&7i>}@(LTor)fou
zyf~Kxc-J7Z>Uj<VC5UuHHi$|#utqIEh)mweLbCy~rxZp4!n3vp$X?mF0i!U}LV#T~
z^0MKk0^yE{+#d~sp4njNNc<qS`4{9$4G!on6FUX$BsvNhM`5*Zx5nZTD9~KB$Z{Xb
z<?LjQ6P6$V9Fo9l)mQqPNg^Tw*gcx}TS1;qDWD8osesjt_o;dbx=gAC2u(2uZJX&n
z(q}znWLj9gAAfeu$1c}#(IlIo_bWkLu{N8yVzPB}>974R8HZJ$)}>t)MqVT3?WUZ1
z1C~y8`Q`Sf6WO&<IyGvP7w@J;8uICsC+2)lw|MEWRevCgii?|UD=g{`WN$jnMh_yh
zpfvxYuAu%VcKFvVO{1B+9jldgHLTEBEa@{*^i1=W8{Lc3cYn~)&(#_&xjWdb{idm<
zpt_EZgn{|@g0{!%+BGh$Tz?R5aQulu#Jc=K_9?G41fFA(yJY9lKN2uH;&VWmd|o~A
zqiOlhRa+3%FOP7$`*wCVsl6}cT&IXvI7c|~Z2GlF0wxVnbn@Tz@$&<dQhbvs{2CYX
zfc>Vnwm&p&l@R4Ul>ou*vPEaCFrg#!EJ1*UAYs>sBCnor%Jr3sW{<R4|4aMcU6(eX
zDr%mZ*d0oN$${86fj&cbPLtbDmRj6-BVaX)G0*?s-DyMeq)9+$8jNgS?xrtQc=)w7
zasO}pQpDFUqA6y#r#DTG*bkW^%=h~a=aPs;O@t51Kw=>%>mEx#xFJ!>uB{QbkAe4Y
zoh7*1L}smb!A@%473<EHB|(IlF8G*>$IC|C%X=Vv+a)=B2#T8g5)kSI{+F>#IF~*^
zqf<M$+A$Jq6ZAo8(^8XL#zF@M-!3g&Qs{kxRShm{LYO|9iXoe-l=am;-TqrX_ULH_
zbY<0Lqs;Fx^L5jXY(Eem_iZ=w4M3azPW<7T6dC99OO>#0qwBxOqU+CLl+>qsT=9}J
z(I&dPc5`h*|8PZApf$%}1CR9QXH5+*|L!iML%(_fbg#DAfWiD+E|oZ=tZ}FR9`+e3
zxJU>jrGYrPZMixSZ-MaWs0g)+{^)4EM5ei&VAC_Dn(qTKuAY1ns-@G#zvpiSlz~Qg
zg=^{{^vWAq9bV=xoctC}gGJ|avCA~_7OtKU7~O+I1#?R?r3Ag1t=4BbygmsBTa{}t
z76UE-;$$iOd(ixRi$Rox#vxk3Jm>WZZymN$VAnjv`U_+>lz0uG!GDt^G5DpB^Q(7k
zQR&h@tdJcj{QR)#s^f~P>CfUThR)a~-~OG4?f*ED*)|;x!%;d@&*-QfJHe$~`U7uN
z1y%N#2RG5xZq98U(m$Tg8e|<wt_95#3FvQ9i6>=XPd%Zd<{HLmA>1~4e5fk57`sQ=
zM^Z+sh&gvFFw(7>2*)=wrz`2ZZ5`b(eZU3o&`KfFP;}CV2_cLnpx(QA-$s=ExZ!>q
z`98OJavD!ibiHmK)35_cwDGbq72tyO9U6Q#2)wlDu?)Si`MY%NsqHYj@gdvaJ?Eb$
z2|R6g|Js!*0?~0pBIGxNrf~Uf=q@7QU=%2>X<*x9DGa~4ZBg-0YYrP+4cDqM=!Obs
zdRy&gPZi{NgN99)q1qb6GWffhP~YkF=%|wKI*^JgBF;X@8byL`msH5^`MTsZdN%M)
zl!UQhd=*0w8PdsGzGCcFN7<i6yCaS`{7(;>&PWtz{Afm0a=Oo<<41_A$%7Rz41S=X
zP=+iKbH;2m=gazDg`Yp^5KB%U5>#X6fgmt(%YkjITfrmeu4+a~4euV^n?JzzLyG{x
zq)L1lyX7xo+W_@3lSMed1Rl!UZM9T3iUBvfuA|P=>7B9xGqHrdN^^4q%cz&%-xhfl
zw!Wm*S#lc|HdPzPaRlZSYT(65T9Ff;lvc<S)Gp5Xq)k!nakHJqnS#^f@HrZ+EkWg2
zWz?Y1B7&Mg*Xn<kx?|pmO;N|Sv7}J?tuJ>?Y^%+Oogv6C;*HBmnMdPqErJ>GVXWP*
zv;3eW?>i0TZ=Bn{*RXMXZCsy{=Gl4g<1whVK)tDFbJbGbar7Kal+>6ja7@S)^^6OT
z>P|x#H<WLh&Lz}wHJp*59yf^`{CnOi82oyuV+@GymMP2Jyy!-z22Otdd9+~jU*hjS
zTh6UU6re5b<bIROigK0?b4kDM@HZm-)C1u=Z_Gv+(MAL_g>yhphb^@3q#&WW#;`U-
zqOhUp`A=e`S109_mPVA6ESj_%N;uW=NC%se0`zcDmF9_{0m%{I*G)E>e|ZqH^DJaH
z$R=7?=Ja@SD)k4*90%s9E0@^?_!xWwWz7F0w-8A)F(KhC_@oXp7g8$<`NjJO$W7U+
zmC(%-<ay(-v=aF4moepRst^+dVyFIriU{;_k3=@4jAw`^V&9K|j@YV-G3~I3^Xho}
zM9a)qBk633(Mwad;-}oh>lGy<oeCD}Y5HAB0=HmJKAppUyYIXsS#~36^4bFK2*czx
zziG8nY#N=ZFT?Pg5djrSm1`%|Y4vh8<J`gRKC;?i@Cy=U9>7Gu6=bp3diFHAfl=~O
zWx49mKcI(rvPe09T^7?f-Hf+iR&H2TV4p$w@lm9hzTJK*@qraswq!}BnPKAeZgqpt
z67GO*|22gBbTFRxsjZ{*^}o$Crhs~y)|=R~k-p>VETwYrZwQfVNINzvZoQ(y%=sbR
zb06l#yMBzR?&?!tg9cLyc||tMM%ON!J}!P8#2LaMGlSU}<E#y-x>F$MUN+h{ZC_pd
zeV8(?262&2<Eh}>Pz<-(agpE@<Ngi*XMB4x9<2K$7`R5u7k4uwE9{Fz2c?|H%z}|W
zl+<M#hWB3}yz935ircaoR02imD^<M>*k2}`X>xrDj%cC!r4m}!Ki>PeOp<dZIG0z_
zUGq%u5N9Xgy6n_u!}Oy=n`>Pnn|G|pBe7++v{$69;sUUg8FLJ~SS?ZGNTvi_cyq}h
z>FK$Js%Y6Nxd(6Pjz>^WG{4TFxtvuF?8HN+-ZB&CZwe|QCi`Z4Jh<@Fnad^Tn0jo7
zWCpl^vu)*JAHRp(W+i;CwtNczD9_o)VHalPa4u_UIn*1gx!5%?kX=9~_*DZQ?1~DP
z*OsY&ReOdf+;+BV#+b?LQ0hi2wRmpMi%L?M3T;dKth;j}9dNfV|H7nzjxIym@6Oy*
ztxB17aoF|yB<(7t&u?)mdTVE`dj-4{`=e%2aJ^CT(ObkPZ=>(<uKhXIzb>gNdI#$y
zxrbb6QSF2hEk>XLVR97vO<zx4yv!}AgikNuTkAF2x;6vkZ5rnINS2i|KjGsmLEA~n
z<Seq6Q=VSy|EkC~P;G-joGN~SS!<jF7S~qY(jt}wGi7sZq*32g7Q4i)ls`$gm2#~N
zb?jxwzZAvwDGcH$owcKAH7)f1;nlgi;R7!<2J2Pld4~r$H9L6YI5Ue2O2U8lE=Oq8
zR6cbK!N{h&0%3XH%;9WeXG^Cm|C&Z!54cjzY`9B7IE10hqZzSUCSU#8k}!kTO12s=
zP-@*Rt*43J@TyNUJ7=+#1NNUz*fIy5T*=Z3_=7B>MX=`eyv$#j>dJIhN9B*=&ys62
zNG{?AWK>S5zek*BJAzjpTAC3k<UcRZb4q+BpHF$op)@aJN#RlbL5$v(F`>MXb|lN$
z|7re~ZkMT=jLwqx?(#LUkv4X+`kRO!Rn-|BKK##%4KnGp{6bCYSho;wCV}aGw|d*I
z&Cy|4L0%>PjVNkBnoMas^{h^V{M=oKXy2V}3jBU~?w#Gy*cIXndE;pU5j7T?D;;w0
z7*cEQ`_NiIrcD-&9~6A8NB)H5B6L{YLO&_uBAGm3v6V?-+$_K}-k@JwHhtJ*R2aAG
zdpnhUMo@6tusWwPW!cpa8^j}dwtsRK9T6>t%~n1Rj-h2xUt?it1<{O|$0Vd#cURk?
zK3JaVm}Pi<Ca6b^>7J^9&@3k+#2@BLRTj#iyDMg2EWRu_{c3S;r>xxt=*!?}m9qD#
zm$s5{`hh*s&p5Ufz49B@Fiss}k2PIK`Fw|&<+_AXJvkF$enRi*hmzxXy9f1z-moiy
zFjut_hD?(S6K!%_63?M*+i7w57`WoK5IY01zax=o0IViJf(F9zZ0Pyl3Vnp9OoKN$
zTbQFP`bT^LTSQQ3n|N~X7g|*WMa}!rhMn5soLfu)vcTbkb=Pf1(d}qLGSkN+JV(-J
zPmBpy(lsQB4AU&fGPANCO){0syFBUOrb^`PWAWrPglp_Oi5}0?Gi?()69rL2s*we3
zw8mfWX`WAxsUHOk!iAshk!VcMzjYrYr3hDDqbtqLt&$eu!{c2tcUItdqK5@^Hi?+~
z@nPUs7wpl&pKc`bOLcyig6(A%)5mC7x2W`2(*P&)hh~Xg^HMn{ykBtQHKWJ@nc>b$
z4I58%+9V6Hy>*!QFzIe%C*_zwNgeX4VnXN#yc>QSwMW@3#)UkCvC*Q)@23&dGsR9R
z!S$-J)7p=8>G!K7HxpmK%pk~G!Rf93{RAf5!phMLy6=gd0BCu)ZJvv<geK5c>MTl|
zwx>`FCBjO0b@?Xii{~(b{+F)#${}GC*R#pTkyDDz>^dp73XQklc3*$AkXp{%(&yAo
zv1b$Q)0Jf@|580-5dS_fTP1iVD)2YB<^zyc|9!IJ4P?Qudu<Mj38vSz+IR@Gq#SrP
z)U(^;-lH{ZNOibf_J)F-Jd~rPc7J63m`xqS)5?B5zHb1AyR`A1ivDG6Pxt)~$U~X;
zJ!la^9`7@?YN4V()`(xkPO805Q}%SK`$I?L@zIjb_K%x?pu;DtVt@zL9w&7<unxex
zajFL-JxJg}FT)GtE+y^Xt;&bDP)vgwRSC)!dOD{XQDdl9j(Es>fpAN(%O#&)?H%FX
zE|!tHedo_bpl4jc^nM+i%I0&DTGUnwaYjIjcz-vH_9w$~3M{TBGUDqyO4z=%|MIcQ
z2DZ+mYL!Fu4_1rHAEkem3mCLyT(92;O}#yqAUs*Nhb@M`%8rD-fT)AYzLyDKu5ls}
z1c9}M*bg}(%~9oglXyX8`3@v|wZib<TD}q3s44sqrWk;LZfDDxNxZ|sud`n+H|AJQ
z9K`gz3*V9bYp$)We2yE-H<r>*&zV6X99uv+=s-tI;(@RgciZ2EoWNik!`NXf5P3#5
z7xL0qxkN8NMyPF2K>-B%OW{=AILbQ&gu22=YZeyL{KxblCZ!cUnA!oMh8_FNTt5Nm
ztwhj6WTl~>Zb?6=TR~yUz|B3`okLZ%IKlPC!n_N2w_Vn*n?meIUGjYqrjv*B*r@dD
zc7=VATHw%{sMdpMg<f>{*HCbcZO`-S@FjDWRcq_$sBJn*`QLo12fYU~K?5s+*}1q>
zu(@gi`iST#%6JL&?uSUd*uM{`dic_i1?Nc`Yd!yUm+9A8JQAaQA@CujJQeQ=Y?s2R
zHUVds&w#sw*C4lill}NRlYSSD;|G15m|x$bD8~qzs0v;}$CPbxu_n@N>+tsNzg9l{
zbnGx2M1ah!CO8eIeA2fMxR3=DB^yAPuk~pIJCjMNSp6>lB~`I<YDUO@NMVq9sOCNA
zV~iiHGedN+g`fiEzQUrKS8vG+HYZB1(RP+Hw{*d%tFMc{f}u^2M3V1sb7uR++@G#}
zPwf~h(lIXS!kd=95zNVRv8YUSa#n*71PNb6Hn=R)fNrc<Rz1`FUNCD_d=bW&EbmZ!
ztp0kwnfHGAR28Cze|cCpEaa?RLpm!TEKEoJHE*byZh|hTX6#oGj5JM2h4WRpOu5RG
z4yT`$T{It9Wh&7D`X4*>y&R*gW1^#0EtIS2Leq7ZIzMJ9sF2q>m<6OYvc5eCyF3ia
zRjahja=Pl+DFV@ch}Udj2wwA|a0=Z@G_;9r1INYU%DhR_@tkI@VXBg*awHBCi38NC
zTrC-=^WS~Bk5WV+`4wYJvBgOAu)K5v(VwgGxn|n-f}f(&2bYea^ULa7O6Tm_H3u;>
z5BLP&Bj@nrw2%wSB8R#piIV(3Nn=F(r5lmTFUHO9Omwpdv<e%4*5L0RguMfDF}P`Q
zc93dto>P&<2fRmFuhHejCtCbP-H!5)zpz&w=tV{1wo+Kr<$u7@Byr=p+b+;f!|pog
zLOSW=5ZQh|yTDBzRw;=7V1%vP;7-1bXQ}X8oH}hsxAezSw5&FE!Pm-6s=W93q|w6q
zZ8QOlWf|DH`bAFGug2Z2+PJ@drYg}ZEDNju`s-btUebJ=K#~`s4*PL*RDUgBaiq?6
zXI>5N?o!x05ksuPYCz+)<_I4pB)d2@ug|!azf0CBhIWVI>j~%o1mdJ7&ZGHkNIFR~
zNokz2kpj({f>YCb-L&ZH)pC6g(O0ztL>$q;d9d<=1v+RNEG|3WqKKSsy)Z*9UKYB7
zvhEE;x{lxq(|Fgyi;$m>(j0yOk@&CVNLrQZ0|vxU*t=Y7wK=KbG7aUYX%)(y#?#`L
z?|G+p^naT*kircsw3HeQDTcKOwV)#1F~yb@3mDwD7pj%+5Npl)itUPMjZFg|co|gX
z`MU;jcX#0p-Gt8a>7Qx7%(xoFbwIXEXpuiWm_t(Eqi^@fnlNf0sx8ij#ftD)YF8DJ
z7zm9Q6Ro*?s{qi7PinSsJ@USS2V@bq81BbUZIDSzd{lW<%h$K>RW?`1S^qfcDeUCW
zlT~!!r;Iu-1cZ<%<s<)Kjg4l_9FWlGYjs7HQ;Xngo=`e0c^&qY7|OPCGbxY7KB3~S
zCE-po-yQ;zN(wno1}pSmEZmY-1L;VWXu5d{CpG7a&_&X}Y1@{n+|2(=X7p52Tap)f
zuMVXJe6wXqXWNj2YGd0*!>W^NeIHo&b2}ega24%tG66}@B~Lh8mp|v=0hf*v;<0PB
z?==J66Jlz)HxVgp?!00C;5%%Gv`PZb^-!Uo&9_TZZU?Q(|NP+F>UUStzg;pLS(5@U
zc3*5H|2XDE0+j@@h`FMrTW`~XPntod$uf(l;;Wd*lvLRJe_z;D*8KZddhbxz#3i|r
z>3L=WQaWv-DiqB>i@vt^?!6&2I$5h~{ORYy!3Gri88>LFJaDGTjzyd99MvFIi0!2|
zb{wb2kgy}4--Du$aF<9YD4m)IM7dfK8DpIKHeIFxH6ae_f{1RDe*uCu|HUd`ZZ08N
z7h4ul3AXQnVt49$*T)Gnm*D_nfXi~u3qUA{CG18mbA8=b`xZ<rdiHN0BSmD}vVaM&
zBk|z~*%2q;F-D+o6{T`sDdS8PnEcA)^RZcoZj|qY^d6W<bAMC1&q;gx*&o~lE?pc7
zLC9)`bgN5E^WR=d{CT9$ypTN~S;DJtx5#x0H%)T+d9CofUCa6XcP+nj*Bj4?nH)0A
z!C`ft?KteO4QJuw1^?TNJ3PsRfY|Gey)41agI+DT-b)RS_9c-)pBs(W<xNO3lb8XB
zMr%G%26oh+_>=omj*P{U_uZl&Skpn-c!F}G<$P=0the>?p6E;W`cFN7ue_i2%k$h_
zZ*`SLeI*WNPqLs>E}|JM>1JRfL+@ZijYYp~tH(05zhwJ%dNCM|)D6)8@W>5*gEvrW
zCh6FMuU~G_s5J%kl8u}~RnI=AABC;^OP&5!JA_ruCoMO;gAUw8`a`1>1kD<F_LW;>
zo<1{hPfASw^&3L#yHzHA``f-{FE9b4RsIT2-oi>18BUvgS$NZD)pk3QnQ{0`V(Lqt
zq2*r&o=}Y5VRD3@{ZH<1QPLJjiZl;VmD1(e?GD7UkbrViW0B}YJ-mttp}jt$V@KJL
z7PAd2yGGum*uHB#9<|3v@4cuh&<gCXtT%1AtjU#5_AB+cZL8`~q^Zt`7dmL+V?aA8
zs;2&_HDWWMc+8G08zo)t*-pv$UfFS(t>oB`F4HN&{IoQAXyzw0$s2HayU^V`u&&!b
zX#{40X&ffAr)Bb<^sZJk-8$>3+mhk7EX}lW<8tum^BVul*W~;(a+T_|vyWbWc#Qv%
ztSdrfeJVi|TKG%_V)E}e+7q=+{psW;Ik|pVx8}opJF79hM`l#DVZ$v<CtvMv8z&^E
zaCU>xmD1CDbo*Pq7SnyfU%>CaAbB@meDl+Y%V=58VUX01k8g1tq>;-A`WFP{{Z}L7
zpdoh(pZ~iU<>bu!Sw|u)pUeBI+%c&8muZ944`T83?XXgpve_bHZbbWa;<=3uBaffw
z2zWeQOAu$5e=BTXXa7Gn_o9PVH<c@@_HSGbP2Qp&&fMRAu(dxAhVn}yL)uqZpP8*d
z$c`1i*bGY#yqZf|u9ygj?wb4Q=fCnjUu49R??r{%Fos(tsk!=ZkjkK9KfH8=)RT-O
z6AD?3af*(CL0TDYFnWlyJMV1R_CVCNUUgjGehZDgA!V3NPxxQ(>cv_uUp84VpUsHx
zSKoz@@}*vZnWt}?Cq7k#c>3>Uw~5*l6A5cokBGiXA0iGDwgUSv$3HsXX1>^t7U^g+
z{xxElZ|dQRIw>tyc{Ch~OU|?$!(@f5{%mVAGxw3ov3EDRa)DGpfTlX7HGI(3o8>Zh
z&cw*+laiv$M0B*;z+Nq=(>1wIa(Q7PecmviLj#r<48^2c=Oa9Hic%tp#CR9v07C|B
z>3DjrpLjZDzvaJ4%w63E!&qoE$pa0#ZLy+2`g7^CcUbvJ6-qr|*lYF{4#DF+@QD@^
z{Km|&ZC*|bi`4RSQW><Su{gVC6RYzr`#5&(!r2s(NO$+|Ln_$Gk#~F!eS~bH;9UL;
z!{_WXa{6x66uc|uANIvgtWR748_rv^sl9nt!?*4yqekg$dy|~~uJ>!Ba&CkTTLUcW
zj1VOW?OA;XZf`vZH7BpuZzWtTfy>Xt=0&0b*SDiw6oHJ0JLH|DR#!hBDKct?>ziSz
zD3`y>rz**iLWlOsHOH!$V}E6vrkj$I+i>M~rLHkNez&Jn@i-xUM*Mh0_au{^3b_|Y
zzV+e%K>?fDb2d=%4imA;l+AqMV5<2dFVC~1d*I<(USUU7>$25C-*4%q)B)+%F3x*e
zaD4Roe=*?j!Sf?uohoLa;3bpSdV}l@$zH($5BK;-$CSHHPzaRNd-%F@V*Rl@KFJi0
z?+6xNMfY}EBdMjHED+SYbQVeCv=sUITPSOA-2I(fILaZqTn#den3mW}376Yw!i<_M
zyeM$vpVHXB2nxJ=LllM#_-)wq8qe$1>1DHGF5HQX--57YgcboN>PDLjWemwYq5a_D
z|0V+(yDrz@ljR2#q`I2IFA(|Zrj!p0tISc)s$PU<l1EsnS>qU*!aM3ZQk%j{d|#HZ
z^d2UKxT2cYHHuhs1DueO3i5g0ic50p2(-p!xxSz&WQbi9o11L53Ay=^tduKY|Duui
zakQ%+Y+X4^Dt!FrbO!$<WcNX#vf?O^T=TfT-dcFqQ5^h6L?2T&t=8pf@9q&lc-$dr
zJ+mEMLis_Ld$`<nxc-ME!8Fy|=B3THAEKOR!H3)b#XLlMtw%bb3xrU$*OF<~OWV;9
zf$XXez#7@_o?lJoLq=<dE-CU_Wqrw4=cy5P>)-cZOjdC2(J&d4ZEQt0NFg|N%3t*3
zW*&T%phP=JFzz!qbmj3E4yG75c5DwaZmO14X1AIxTtIJbf-gAL8V7oWGjy&#e*_Am
zGjt1EqV(O?1JqK4maAKU<|<g|rni;St#0+T9_LMPQ_VoKeIW%5bX4VB2#Z>ni`(KP
z<-}{F>D~DN^)%|w?LAybRF6414t$q$rj8=Q)TiHIUarjLI-ha3pO*@v?mS&2EU!E7
zFGy3&sh7j}JLT|Y^-j~H^90$Z996{@<xk}PhDzMuu^x!=>3hi0hj-yv_e%1ki0ut~
z2G7W;XsW#eb*G}7a*7-z@oqKn5Wc3{rOcJ@_fz@<u6Kh=MBOa-F0z9T8Hd`cH3#&%
zuNB56T>@@z7k;<WJLoi7cIJ{+>bH@<AW+l(a^?Rqf9w=_iB~MDM;qQ7l~nb&bzRSl
zq)dl5KeFP2Q@DZ=jqb_EFNf3hl^HLTwq%4ZMOzv<*n;^3chC%1L4Yy`QCl~knPx93
z`6x}hejLJS--E~a68aUVV^;H}@%^F%d-LBsdU5QyP|@C;lj66~H+o<z<4}o_^s9^H
zqJY2y4sL;(p7&P3IcC`pV|=nNxXB=zfwD}vf^yw8T%8|CHz6#CJ>g@x^pze(;ns^?
zdPL#9s@;QZryOz-8}J5b#m34<e1JQgk!99PBD8wI%?adsSjJ>t>T(^-YKpt(FtmL)
zCmFzIDT;FZFXi^;I@safgFmJoP_<dQ^FGY+;nvr#h*1zV#;?!&)O3F+3$(<9Y%Sfx
z$#qZO9BUJfm%q;@W>_|5)mRmKGjsNV>10J$_HU&<;^n7J_BbN+Q7hKfLY=?(`&XiB
zf7Sa`#r;qm_8X9=dvyoAr!$tVhof1ZH|iP{F-nsr^%wwAX#-{>t%Zy?0cfD*2GmW4
zG&Q=6$?|qAwWVmuKn0g_+_CUqESd^|Ggpz^-zP9&89~NiaQv>v35aQL7pW7Olm6RU
zE;N_^BDY&+=F$XQyLWwdwiEH6R?~UpA>x19ft)brcG^!mq2-&j+5QClTODshzGHA>
zX&UQtJ9<fxYvZoHzXQ9t<2w45i6Vhb##bl#6zM%i2NWn>-cK~Yu$9r98^>?R@ZNPS
z3c=|GAcQ-{@w51a_>ZLLSj;vWc<GWG%B2h+Can7y6#JtKY$iS?dHNOe5Hz3~Zcu#8
z#KEuKHq2}V?T1=5F4Krh)G)?c{_I<l5SujrO-O1Z;Qd@Wuipnm7egkI9XBr!+uAMV
z?1Ki+F*lq)1KRfWgeFX8)EHpSpIj8OouwbozXfo5hgjn1FS_9F!;3!)(XCu8YBEm$
z^GmFnM|S?;$8B#{7=Vv5pel)s&3%TGtN{Vl13VoSJ=~wlW)*A)cu5*a1`5=+Itwn$
z&${bNzlqenvS)WdoXuRs1@!x7#JkuQ1*pfr()DnDF)`<PycV5kx;?OPksZliw~<|b
zt(JJ#+!p6>=C5+nfA<#PAs|+_b_#(OeV=TQu3yXl=LFeyx2K2yM*L(Fvs@``XJ=Q`
zSWM&B@4pErtGYeE$$@mZZR97}segbY-Ihs^sA@Sf8)Geuq>+2lj_x;dtT$RHM;B%C
zXJ*dI@gu6q!lcA=rFR3mZX$$qAihkOYSZ)6j>Ejizq*V=y)U=5_L6$qCFSG2ysz?W
zO98kn3xM@z9PazLeP$(tQ_s!2mwx*RuH_efYNpux|E~4_fTi+pLel=V{HZ6NYC?#c
zHE8UwuU@$LL6eVs<CMhR?h%>Fzr?v^ntd0-m-(o-l;F?d67cc0A|~TxU#9yhz%_cN
ztic8Q2k}=k-=XQ<KfqN0q-CR6w9NM<!4F{oX}jkZL^JT;d1I^%9Bz6+gx2)!bcbxD
zGEfv0fk1(N*Kr$aGPDj;XXCuf3=?k{P>g{I03!Kqu5Y!i>GrP+d&bX=F3ya>u2?=V
z4-wQ?vV8WLjQ~gm%QqIvnVL9-`}@0Y)|qB~``zW)%n9n&aeY_#W%3PxH8Icm1y}Na
zLEALk-mo&2P{+ejNr>0&*+Hm6;o<-Qi~}LKEAKwTuIpp>U=ZireJdfK%c={*nD0+U
zfoBEuow~5-$ZgMpi*t$Pv+*p?uR*7mRihIDlI+FVjrZ%yzkOlU{7)2PtDu3meD+C@
zEP%trU!qmypcP{nB;eGn8&vl{0Uqb5m7FG7*!eQ#c{{bz(Idpw@1V*+i4g>H%`L!$
zx&M%zKAHOUV{Zq+Zcg(ae2-36@>t4yRzO~Q70I((f!3AbYu#sVUE{&11Zj@pVS_p!
z#ityWE<S#^JAE!83w1Tk2pGoeTREb~0(Hij>cURSOpo;ep!wN&%TPhACtzBd<m=iF
zP(`_BrP3Gi{{u*GADGQ{7h}?j#+nX)#lM4}6gM)xcVF*r_rWoy!9F|E(L*^5*#KG+
z?hT{|xV$j9igRDgN^Rgb=*V0O<4)7hu_5K%48c84?qtG04M{+qn(-*amY&(RjO5hF
zJzIsYokRS`ulGU#X~*RINRRC{_;ctSy645V@}x4=s;9?t16b?6iD0RZ#5p{zIS_(t
z{<W0t{7Ocn^PWtlK54vkEj!MXv@Z`pZv#%N)n_+zq#oB9l)i{zR;)zQ(GR;^%EbU9
zoU*+>6*SoVxkiU}2(#+18LXtFPboBw4c95k;Q`Q9TNw=K0q=&8b14UtqWuTqrJ8rb
z+HXN1+8_oD-J;g{d&px0-Eqa}uR2oZMxV3$9sqE9z*q5cU!T;~EH{kkzbdi09Ehx}
z-<>%Qiqk#DQA`-@xOoRap*t^$k>A|}Yj3XoZA1hwrFOQ9Rd5_0=Lyi!6HWv8tb!`@
zeO(sgNyX^n(+{qvTyNCAk>vgFt-v!FJfcU~Z;JLi!Bv%?%9QXuggT#&L*FMZ)Nw=8
z>iKiR*^Fb^kzK|aQDBKtPXC$1S<da5Pc>OUs6bTn?vkUj{vziG_v|<T2=>@+%X}a0
z{k}>?!<P}dDex7Gnb0AaMKlXQrdbx4?}rM0j*9+9upIKBk$A%001^Oa2VP#d0xvHh
z-8TiLSKwE>+#+Vo3|2WzvPl4c6I2v1kEoWM#0RG24cAA{2P|8Bs6IwxgUpSINRV$@
z2T>i~Cht4hHS)FI>mSWMZs28ggpEJN1$CY$0+h1|v(zNY#;se?oEEc@{*MGxCG`^K
zcH%Ec&{TUx4~KM%*{-xcNSdtk1FC!uq>VggEgcmaKa93;8an>_p{D=#W){eh$eT7Z
zUe?gl*P7@WYGY>4dqa;%JbGGsS08`^2cD`_FO`D_Tw3mYv|n9x`u$=QS?xKqJ?sm+
zniaJyUer1><a`?qg{a<@`Y>s~SV-)hVvhyh6?C86VNqFMS6MDu0#JSsNcr6d`AS*y
Gp#K9FV3WH5

diff --git a/docs_old/logos/gladia.jpg b/docs_old/logos/gladia.jpg
deleted file mode 100644
index 4c5e505eb43da98db8c9042bd42e10c1c0efd795..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 11198
zcmd^lXEdB|wC{)>JwilJ5D6I~Ix|G{P7qxp6H!L*j4lX?(TNeG8@;#DTa@TV@4bxP
zYq<Vv-LwAZoV)Ia^Yz~SuC>?mJn!1i-tUKZ@82%B6SoThN+mf(IRFk00DyD%0B)xN
z(g56h|Ij}i_m1!g@ctnp0s?#jVxs%^iHV4b?~_uH-X|d^AtolHA|t1uq@<#}Px^rR
z0VVa_n)07ZaQ>0R#Ur{Cq$D9GxjX$I(rq(<iU`nwlZK1)6mX9U2bT)xwjIE9XXHIx
zoI3*i2N4nz5Z%YeyN7%CTALE^kK#Mq35h812?+oBs(ZM2cPk>Q2TzEp`JPHWd@Z7B
z>RLg=Cj8F9#jTu{?lHg6TP;2RsJQsDO3Wk2n6zPOo%aEO=>nf1(HW)tceg(LPul*s
z)_+c^{<YyhPNx8*xOW<<aH#;2fJy<GH!?DK{|_<Eu?DD3j2*&FKX+2%MU2Vz*d!hI
z&BCy(rXai8&U#ljN5tNneZ~cKbq#g^G0xNU!4CKS9q<6uINVAZhxHMIEGO<JNawJs
zO{Vfn@r)|;*MOqq6l@gF(L?YhnX}8o=z{^7ArN7Q!mcPI)yxA*1*ra^?*nlBBAR2G
zvRg1|!41mfd^&aMJFzpR{w(`36Jr6^>;<$2W+ZCKdeFrV{Ao0}Yq!dgcMCA#ZL&)w
zl=pPY(eue)bS;!_w!H<o@LkP3OAjbtIBZ#$V>D1t;5Iltvt$%aJ=Lcl|3JYKWgF*X
z?mK0sRiqWW{8LWczNc(!lq0>*(tlK^a@6tLQou9#u2aM<p!l5vE?NqkK4cWrp=?RZ
zoq2j?Q?Buv$kYP@wW)G{jF)kM;Ieue6Y;0nC+O9V9c&6#_dR?mK~>;NA~mg4W_--p
zPWxe`rjTyRCxS+<M8-`8tQ$@}L3Ok|a7BoNijs;7=YPxyo)n0>^r#<42oG5AWBNXX
zFDGy}ymTSs2|d!nD00*94-;ED0az*B*K0I(huxAgr@LCsGW=s*ud})2U*7^k$G~Oi
z(iGFZA0jxKYP&_E%hejtW_Bm8tsVGg%*Oi+$a^LEQ6t-lxTRn?0fm+RYDV0TkC)27
z2#<@b?Y*{MV;HOElNLP4Y{G5<7D@e?g;~BbQR0JvyG{xmMNogJ|C8o`w`4U3^b_G<
zre3$@T)Ies<q)3*$q(BMjI3SMK-aYLzU+HfoGCv`XQvp?`K_jr{Z*(uf!d&un$aN@
zd?0p+J*Z@6;}XqRbGGSz-WwBOu$P?G=Tfl43A@y4G$Jp5Q~75=O7Z7R;Xr1M;g`i&
z-3?FgTuXIOnP#W{_hkWi)dPA@ZfO`yv_~i^*KwMONvDmt{?CO&PtWu~(xY6|Pyc45
zgu8@)TZkVH(Bgwo_!a?4F};)d$Y1f=uxaoT<!Olhv+P)(3%owOW2H{Vap7ojM`Aqc
z{;IW`cLv2LOIN2Cu}P(c>nn9~Bhp$LQ%Tum-ms@lQ;#|AaTau%DNMx)S19a&>@Xf{
z>Eg%u8863~Wd4+1c8pH#i5kx>N^1<hK_IQVKQJ_dC`pVceEtra+yX*Yje!JR$E~%s
z*<4muP{wjqfdSIb);2P~`<)^_0`Q_?02)(w_+D$iW1{qNn;MioP}Kq3NvqUJX;-Pj
zk)FF;O<CO(+jewPhlc8hTKSw6>l$ZU-i%Znk4zWF_vn4mvB$oQZqyQp7K-w7qZAO$
zTDf*}-5x{M5Go2JfW}EL$FpngAn}{9I^`h-hAuc=XD34U^7S+FU58^9QC7;HDs<J4
zE$7QzDU_ZwFDeU9+r)P7TO)}voy17-@oMK%>92q@N>SVdlIdV$z=X+s?sKTE!1*pm
zcryf&`DD~htc>KBv^;@lq7m@|#m$)T<hK{L49@!*(}m#lMI&1@{!mBSe!Jduyu8CE
z&Hz+>PkVbC2UH~#b&%;>tpQ7zWzMrr5sTp)F)V!;y9)!-TDeilL;X9KWaeQJWZ0-V
zOcx^(_YFzcq{%we%^dZZ8KnyU4N1%Xlea@#`|9w*P@V?NKe3s-Vdj+O0cOSNvRH>_
zlqk#9Grbfx6dL+ch<+zQWT$V|+8Ek@3#jyyR;SRr1a+~qT)aO_efsp?wW@&jfrQ8E
zX9v0AvxZ%qLJP)d#bdzpCP#*FOkas*)|X5x8pwBg9}3&*xALvFI%b(=ME++;VyB2N
z6%TwMvf{|Oxtd+X0blQHVjK777zF6cH2-pHQ0gIubsZ5(co^+8kBn^D_N%>~_F;H~
z!FJnRJZoGg$R>2|VXD>VQG<l!tG7Jn#DjI_m@s{BkNKQP^Cqi$Yqp3JJ<B>i4c$l_
zxEiJ8OZ5irqfe@*ZG}B+c24u|Bc^iIsan*jGox++QH6MMXLI}8b@UMSt^;U8ZZtC;
zcIxp*^6Q`-#?;N_hrI`+yF)a$0HQyIkMYY)dtKsdC-PE#SG;V2l4A{Vy*xj$yV({U
z%CyE==t2xbZMzh|rfQoRE|TG;dAw5p`XtO~?j~&G6nBZmVwUU`##<g9)|cz7Q7b18
zc$wYsrsHt)BOWr;0EhQ`${Ujb^V|ZmaXmTjA~C@%osp^rl~ENfnu5Y_1FHB-d-&`z
zG8Bq1LpFcl3RTxmmwqkEjv=B=4(Y9%u;a*Dp(5;nz|Be6a4;e<u>Tc&q|w>P;9Yj1
zs-?-MH@qI-(V0L=PYfb&evqm2c8>ZM04$I`@}A#*oxk~C0yove>Bgx-P;r2F$7a1_
z$5LnM1(4OrhysuD^R|@!E>#YlP?kP{GAc1BiK8O%17TaCPU)!BS~Wb|r9%PB2$u()
zk^Jl}KiBcNUevto<gVEFykOzqlI^*Em73xvU8^cQ1y3Q}(qPZcMU+95ttvJBtZZo*
z9S!=^aSeucM0Y^~4%nr!I$Z(Wjz`KT<{B2S7;gcp`TP8ydGN4l$E$*#@E3{eS>ZVn
z+G<}#+XX}jD;|h_H|z<iTPm%zQJFD8e$syyXE^U(_i<Ek0MSE@n6%o=OQJAb!m^4U
zAYAtcek`0hAHIR++u2^>&tw;1z0=!VhI>&I<yvF%z_BBriUBDyc2axZjUxnxz1|i2
zxO)ZeX+YxVO!w^@(Ko;U#_}JvKWZaJgA&4Brh{^C0j_INqtyj!$XA92%X3O$tY8k#
z1Thx*LuC=SlD{Q?2Y2Fc0V1TnH#F*(idSjVz#DtITfj9}%5kLAIAek`%}}{Chfd?N
z-7nE1OJ^<207D%mOf&)-GAMO_*_9t{E$w5V<u?6gt)SvDxuSXI=Uo>qcj2&Slw+d;
z0VG`MP6)<RN_9gc@P~H(rX;mCVO&**QE%39BZHw_8@psJ)get9;8muVzGy_Y7@V-2
z_v7;9P1+6yoQ$<`jj;7qbMiEZ;NpUvfXN*2?6F)4Ud*XpoqO-<;ob|(Gme37m31xZ
ztYOPqF4MbNgW4JjXA<HPN-Pft$$i+k(xG1)Z^Sap8Oaq-G-Qwu7zQCd457<q0GMd?
zr4PkPq|WZs`(_vl{64oX?oUbDN4*fLs-xZ?>COpa^J?y-_e4S@eqEp<OemTh%NX5k
zZ)lQS>%Qoji2b>S*T0@u)p7VKw&-0WLe1!?0N4m>;DII5L8bN7@z-N&@Bh5Es1$0v
zgDZOd37z6MYoFfJj?tmFq1@~oyO%yEG$Jg*Jj-zPCz+n*`T952>ux~#k%9nN!|vM6
z2kxuabe&+z_LeD;DGD-+ZOIqxJGN~gkuvXY9gLM+PRoag&CCac?F*zjy2BK%TG67`
zAUKmS(^ciKy4>FL(<jx&6`w~3t>QNYt8pqOH#wj?HOVH*iZrsYsl5ojHcoB(01+x?
z#+$#!zy*n!Tw$jQgZp|G&xNV7jz>m@s_0BZ*wZ0w5P34F0Za28!e;U#;$ZynsI%mk
z<Mg*BS5K%8Vn~&ViHwF{>}dckIsI5f_uh#Fs`lie<duU^+Vx7s_|<^?O?osRV<+{%
z0am4&up5|g#o?Q{Zx~*c8}|@%3?6DqmwPbo+yJ*+I0drJ1Ww3k!q=Vh!uhAGGR>W}
z#OQk@`0WvJxy!1Y=CxMeCDCs$xZ5F4G=||bvh%w--AAV9&Sm<1=Yfr<r=>IM+JtHR
z3d7P3URCWBzduN{okHAZ=Q8?KopSQ3K2Ge(j9<kv5*wK?t-fnEB04fL@L;Fx2w3jk
zgjR++6ECJ^jnJEx_p-Jrb!{Eiu*4W~K;8Kza%VMSE?!Kl&0ps{Yj{3H_H1z5=Ot9B
zL|jc=u7jXqq=Z7kX?hA_?!;cS4Mx=<+tx^U(bK@{JmMYQ@7{vmG+Lxi8144SPjMlm
zMG${-7AX>}42|TH@#1)psfdf~;_-uy9znq}x$bDlNa0dk?B0{~qPW1bq7bTax>GUZ
zk-_Fp!uN3M#Lu2?36tdoa;qHp*fv8NZ!%>H_qFhrz7<%@asXx`Kx$i8<Iyxz3TK(_
zvIl-GNl?+2Dv-v8eVZaIotR$le2yaDXjqDcPor~(cJl1VwY8}n5hyVPRIA>Si~2;L
zZ~ax8T-^dyBL+}B2Ni!jGF`bmR$%V;{>zo`Wh+TW?{Q-Cp|Y#6nz-eFNx0*=eLXcx
zMwwn=+ER2tcrs-x3tOf=)O)63*KL(UJ0ZMtF+*J<zvmQ9DSh)_T)Q@KQSxxAT~c`V
zrj4sH7C+BmatT^V_MlcZVsy0<ZL^g4JMN)ggE%)R0k;L34VxrnJ?fDOY76Wl(FYj_
z=X!PK!^e-S{sQp^P}Y+{HZOz936usmatsb;(|VsA+Is+D3<LNJqeU-5S0=~g?b=^$
z(b%k=xct-_(im$0jPRg97qYxzPkDpG=|?Q(q}TLgZJbwE&Twy{(L17r-9Ml<dsZDp
zRe}^-q3quzWUd#C07oz)JFT(RI9Y6RyNDCWAImaQA&ebj9c+`<m-Fm-(kPp>2c809
zK2w|j_8YEWz(&feuHv%4qdDcDsxAXs4{UQAOGb{)>pR9}2X0U-N^l!dpL**Wtcjo1
z7jCCY!=r82vT^&0QTtEM#i85T#o3Y6jQ+)|SqC+hy<1E(W5+?sH`_b|O>up`r3^QD
zGoVL&<=AIb5mp2sbD;vnsG@t?(c}<k3-*b1=}C3W@NNQJo-+XLWc*@Be5EwQH{%+1
zU5oc1&%g4@i@|Q%MzI3R<HndUh$AGf?Mqs0!4~-$qOJmh6CD}#doVg?usqnAAKyt(
z%q{JuEoni1=`euWNGo8ghF?)g6x^fWWm1I@+F7p7Trtu=D=4g)v!&LZ;le`m^S?OX
zA!B&F{;I<Yx*wDv64T9z=D_dAxR2Vz8%DIOHw!+y|JI%6SwK*`+sR8Bqk~SuQqB!S
zm$a+zLwy~P;T2EWB+a?BY@m(Vt2M4#VR2-X0L9|=dSDA%)+}d$J6A%*ax2PREHnbM
z64Aax*8n!etJwc6%n~@#@bw-^T6nRS`7I#+&A99Fv58JO^(;$pd^H3S8o%W79_u6;
zbf`5-Tv+|*seFo(*^l@`!ySTp!pfG2%~L6<LJ#x(yrYZOj1#!7<-*CKg!$O_gmY8v
zAkQNBx<@furjswu<z4?+-3TElk^(bpycam$b&+Dg^I<WFpw(8==@L@;%VaVE8&yZ*
zJ=L^pBU=6Le#pg~;mRq(XpT^sY4+_oZ&=Z2CADy8#Vvq2hDX>4v88C9cC@`1b^SE`
z%WoFL`nKUY6kMIqFMXaty_uHeq}e+x?L9fBQTtN(@=a$`??<U&#EsJc{>um2BfVW|
zp+ScVX9ShuxY3GWKis*dbW{pW!}f>Q1;fi#s)7xwia(YQiC%|=S&}e06|*o$qERt;
zpkOo#o-2*UlN*+{uQdM+wU{z?MD$2S86SC0x&2(F`^~_`3JK}c<X#SwI__8x`fe)m
znnAm}D9X^0`@K!rYu)2(p+%c8&oB4fSLX@h6$vrzW?FK|n^kW4qQ=Z*+2@r$Qr^X!
zUFqo&xvLp<$R&3u9iihH<sewW(46-x^TBc%Lq1S1w(_k+sg2o56EWw3*moJwEdc9q
zoLb|pFe8*;i3}UFC0O-(<i!<Dum9USA-zPMNmj=Bpq3|rpxz_88cAU~(APRvJR3B5
zc8D%MW%C`?vDncaBO@MG)t;Qony`%(@OYdWS5>>Occt2TA8QpO@bQ$d=OGUvDsp)h
z#EgUW({PcN)+cKXLpJ`5AyZL7yp|`-YNqPC4)S$bS@LZ$6kfaZ9j)~2`WIh`ru{8Q
zioUT!#Qnt--vSgRq0d&DlGd_(owsfQ!^XV4@mnrcdOLA$`J=6ffcZN|Z<Uv4X6+zF
zO>cCUk+2G;-c#iCyGL#XpR8sbG}>vzY5*J+e{)~?p2rKKMk>iheprR?gUi3KGRg@H
zhY((DGqJNs57^S4pZlFB83}643K-g*y|A#ac2ad|>KtW~UGbE7?oI04If(3!bOpVN
z+@Nktk=_-Xtu1()Tw&{-^dK+dS-pyZ@?$dYsq>#cPV949BI`r_;5H@lz$hvlO_uLo
zj{KgF#wCOW6wN9{zrt}tTAEk9mLuhlX<%X5*U*Z0joha+H6+H=E9p6zBERS4geQab
z<}PxLb|sn0)aKL=Rhjn#DmGrtG;S&V*)rjFFUXEThURXVMZp*L-JJ}iKFD8z25tfS
z5>kH?CI51{PQou-9$8GX<!<U-ol-a_7T4)<#QITuw0iLiCi}|H?zcNY#1)zZLr4}(
zITX`@Y11rylvUj;-dBgviihTAEj*G2`O?2I>{>0lXkiQ>CWrb?uKC^q{0}L%yKe!T
zx_ygLwlxM-;Ovn{O7vl3IgQ&Mk=imFE9SzUt)fsrwn(Y#ufTK}{K6Zyr%q+b+7GW!
zyn9fbft}_1U1CkAhgDK?hQC_4T!d{eBm4(tE}(WIeXaL}1RG>WC6tqimIDV~3nG(m
zY&S$RjbH9qW9qDH<t&v|Bufm5B2XCAt`~^MGVT^|ZiX#v(iZX!mZJIWI8#cL5^iJ)
z4Rz*0xl!T{45)1|9l5c3+T&X37~1W%b2`+y&jd}YIM1z`26xKarRIjlxn>WrdX<DU
zCB)1)>SPugjL(y+HrYzP;}%rc$akFB6Rg<RQkZu6^ZX-ekpGXxqM%MBq0a`l^qGlU
z%q_tCQ_2M2Enp<2_NYGPYU|-mOT<l9oj%1^wZ4Pz_yZhuCj1*YHUdt%+Ff=UfUa~?
zSogD(nXxz}_5eiyxq9ju>xN<5k#YIo-#cRw?<%>{!=NK)R#@Ri_)jU81nOq<65_p<
z69cMchjR<eK;GfsBH1b=<vbhIZj1!AMPJF=vAjQoO3_)pGWz!97C?OGmB*f3H13@~
z);Z3j-A}o&`NWd6dkZj)Guc^sj0Fl-92Wr*29}Q>ldXr)Mp7`}Y>56Gn|)h2>ix0z
z<Y{BpQFqkb77wXYH<(XCm+q(NF%+_4!Mrd?lY9=eFpEzh`BP2oni>ZE?cFkQlCDTc
zlAHSp!Xmy5a(ylFz9hwjb$JkJ1qv88SdbzNpg^1KeT`FPyy7X5Rbg1uGT|xJcGy~@
z4-s)XJ{C<mltM$5>vbz626Cs3YS*pPeyoUju)BHEN6t6W>w-&1-x{9>3w!+X5da?V
zR?A9@_vfOrrc8=^v`;JHZ9GE(dur7fZ~;p!x|wzD#2Hnw7u^wl6LGM2xbaBvdlnew
z6)#|S>>J^m@FpPfT8IC#e%>Qd2Yo^j{hc~9tyHt3E^@4jpzk^E?AhkpE#Ru8r~b*Y
zSYMv_x8nXXsFz^Xj(Bz1vN~xY))K996J}0}>`W_6US5OIdH<oBFVK0l$M9ZXn?k>9
z-G2Rm#6bTRP*G_+?w_SY_oZC(5ML`iz=R?{gN1^r=?VB2Fh>Mbm8%OMGV-<e<cqa@
zFRe(Lz9-As{4A2(k{dyiyKZ^&!rDLc06UkB>}Ty(2#iZik0$<r*ZVSBm=>JrW=R-u
zzuuRyIO<5E!l<e3ewMdvZf4;KBeE>;77&a)GJFZHxtd0K*j<0q)%-WIN_{<_@&S~5
zD8k*=xnJU%e25xq8YTeWK<<_A{okVBP`^#IRPbWLc4Ka88>C9dC=<`A$24@9<wDT$
zuUGv{nfPeilloDK?8Z$UppEtbs8&DI12*Fmy-xj?JM-^=w3cR7p*|{OeeuD<_?m2+
zXM}$0PhtPPy-u?}dXmz8ufZM@vg%wHc$t5S+sVB$2$Qtm7JOQjI}QG1&%(I?A${K5
zd7frhdsnT#^DTw(o7xpmkcQTK;j^4kYp=84cmD;O&in-xhDjMWzOH?2*6sXhwCGq{
zv(vjRKdpL9TM#*NI+but0E#dPoqcR4f&Bc-%S@F!mEz(3JSCfOtd`YW%X-!5Lsd7Q
zT(aR<Kf-<y&P&lV9^XM1j?h~GzV&?4<}Kj;=oT+0_09dsCSlA?VgA^^<r1lDE9_>@
zi!Vh5$&naL$EI1-dT?ntBf9;%`JN7~-FQ1M&&=<&I<<YJ#O+#B*?0O(;>6Zx)pC1Z
z(lc)XoL%IfB?8Nnd;B(Q{`NK;>@?|QhW{?IrSb;agD7S3W?NHd3wF4*t4))tAGhR$
zGo?rMxh%8hg?7Ma9~C=Fdc{VBQi5tO<10k_NcrWQwcfd0iDh8=fF0{~vw02G$5*e0
zTZzb#9u(z_HY<+HSOvf%N;0&Du7&Vd6aP*-v^Ht+yVpfr&rm4)u=JgG&2N^Gs$fQ$
z_z+4(j&?%Pc4FqD2wRg!;tqMW1{+VK8x`7ByoxN)D`&4TgX59`(m3pqgfc2v)(75s
z@?$#|{na?Ry5U+Q;(obx#+i8XbijakS~yz(ZL}u@0>c?#KU`7{Wq(EnJ)ocUsnpfJ
zpR82$`{girPA7vdYV172^oBdj^k|jQVIYw$q!{4T>c2K(EEBEUE#Vebt1FU#K2z-Z
zj^0)0Ia#$!304oe`&D&n-r3?B6yQ%7lLn~xjK7;>@XUck%x7AjWJ8iBn1{uyT?JdF
zsB>6_!QBRffJb4?fHUul4fbT=Ioge5$8}ZRP$-Wso;|G+aGZXV`+24*0B*U?6A>Z+
zU%+j_qngO)n|8evNb;%4>ao6xyz|c1A#51}z3|5*&u#meTC#XK2DM9DiXqCsJS>0?
zY*`=a+e)mwo41=3U3?X{ORL`%(vih1Y7f~D)Q$W|`{D7~lRg;dtlG1NJaC{CqqE=V
zQevzRdN4X&^d9SX3WhuLW7HYYcvNjaPiwsKC<9I1BFcbDQ)=30Myc@!fiqUdObJgz
z1Jg~U>;uv=&T`wNw!IUaqg&rRsfnLB)5=TI?w2&1a&h<^>yk`0yFF@D3tuHse^P9*
zI}TS@31946xvW4JRVGL5j^ocii=7eR{ZLr-9cd~r_XMr+UT%oyCOp+apF!OD&oGX=
zIeyV5OiXGK!GBj(A-68JK8X_xMF9)CoQJbNsrUyg<v-@EZ~=K)H1wsEtWUZ_zbm&c
zpWXtB=;D`-5x?`@_2Rimdex3zn6viDYchQ22v8Gu9It@;{tz<~r_7X1Ly|fw{6(42
zUob8*+t@>KK*13Kn#_c~qU?8g&v5Oo(k6O#Krs#Mw7L3PFE4HD<sNJd8ZYp)pKtg!
zx0{>me<t>r(Xg{vDztN%H|5iGU!x77rWJQy&X+n<54teYfN83GCUnNDBP|1VLK4+5
zcE8h_ibzkR>AD&C^zphgy`3(~Gw7{5`+yDl?a+RbnJ%7?Md%Gm;MB>3a7LEa76O?%
zsocuaPW0gekJEnkub2dvaVkr3tN%VC+Ecr^c2;8OCd{m{UwVYe^gPR$+WQpoWY}A(
z;<G@%opzK~A4W^fm)3%%y_9*md<y)w*G{;HN7M#!6gj(5==`vc7m-($7AZ|F-!R}o
zW^tY6h*8zik`?F5!Eyvx47`lyj7K*`!WX@T#aFS%YZ3?Ug+<PgV0uKz9|H-~5p^OU
zL;LskK+^7#Y2<v>Rix12bPpMieAGK}gn9sSAT!&S#rufRylWnk8edVDb3FPf(G^oM
z`*D`XkS0)_?;^=6_<hx;Z*F*BYPFoV^2#yne0L(LXZ8p{5-inNSP!W)9oNd5C^R-U
zu4SO9Dy<Y$BTx$c@JMAiQd~exZ^hnXd(NHt8&JHJVa3Mgvc>r?<bAgLOzE<|^X^;B
zutTVy8&fEa`igI_ufen=kzG_}^%g$nBwR1#d(}ialPRXxN$2=<NN{Y`|9+Lx#)}MI
zB-DA%@JE5g?h$qRN`>2I)b0%z8TEc{VMhAg+(RxbBxLbng5J|DHzr4V;%^q0>V*D#
zx(n!!RYtq-4Z{WJ`s{O4i!<WNF*;gG8`;5fNP|;wrf3n~Jt9z$l@A3&@CfkD2$O7c
zCBKFvV_sF>ki2^;;=%=4tEaEdpPpFY-0yH#ihb}#^7nNP_m+#NmSK9*w}Fbayzr5I
zLpz-ZhAoyj$NVBaDhJ1OA6N!jR$%eibI@)ALkrJYU+uc?H(<qutP{`ARaq7PD6s{-
ziL5LhaNsKR$xB8*V+Mq%(-e94`A_Tjt7vXA<~&OqcJ+aSJb5ie83oh&!2pvd&`^LX
z!>_}BcPxdI;(7f$m&4z%4%-t~QuvS|6g{Y5nwL@NKRAFFI9d$xb5b9COP(F)=3c)W
zLf>8ycjee=TQa&T>J_Qz7EDaPY+Lo4&5YTOFuqZ@?A6%8Y`|N0Ca_K*bhc>YWjw*s
zX>yAdK1Z3zG^vJcq;>??D-qK0S#K2!rW#A!AnuK4ef(gZVKR-GwVdIZg$};$cSkN2
z>hI60V5_6C!=lk4)FbB*->OZ~#XYG04#g|L$jLAg)2K`@d@0%4)RSs7#uIy*e5g0n
zG;{;LF}`;{^uT=>ThHc7BbsmdYuXiNm5G*qZb^DBLsf<Ng4@X?-5tK-dm?pKAl_!@
zL~B#IOM8F2zeaOO-;ALk!8756Rx^6xoJ)auO<{1H#fjbHjN4;w4zpcXIdpM=EAUWM
zHf^`KWB1(=247d5JnSuwgi7J5aVLnfb6*)CK$b_ZnS<a<82gAw4%5}+)mB?PK6$yr
z<~=NT(DT{!4E(j?DvUfimunNcX<C$)8<vg87V0A%tOrN)wZ)^0n7*+F5N%5toHT|H
zLFzsaZq3f`{^oZg8D+1a>uc&|_NshG9`V(g{Kn|X93*)H_N4wqw0`J4AsXCl>78*Z
ziK^}UsRx8aPtkMN4`E7>9xuP=XJsBL0s%)YXZXA}k~Rhnjmc)g2MFW6ackRo&4*|;
zWtRek2Pw|1T)b!|$3Wt8<RI}TUd=wC{b-zr1jProHE#>FZ!nEgZJ<?J$vPScA14lj
z&$<_6qJGrWr9WST8Hj>nnsL#f{Jj6vgK71$kp$@Rj^mti1?h>aU1)90Z{jlF0{Hy(
z&%#j)B>0yL62CR_2`%TjyczzoLgv?Av2H&2`z5_=TbI#urN34@i+x0g?Al_*6cTbG
zQuL{k6Sr~GC-t*|OEgA;1S#s6Q#DfA1~;be@eJu72QX5k-Nn(OG#WeiRBMW-qkqP2
zAIZMlt6ACC5@HPNp<uS@AkuGlc_lo0R2ojocc?ry<_yo?-#-K~xk<xEOkI*b_1N*f
zGqH(_F)EE8QJ|Zu)sHYb9y*%n>Q-LeJaPKsfEz55FJe>zPT=O~9_pGxkObka&_-@#
zqY~lA5^%`W;xXmh*yOr~;l?h$aa#tBaDaI{+uz7h7iUlBlamvNPFvCCuM8*447D#!
zHq9$-l1Lk5jUBSlLR+!5$&0;Nhd&qb*}Aq(jGsjHo!BL?NYeT1C_@jzV^^K8WJi)m
zv3*s`*5^RrqwRO_{Gu1wd9XJ>dvO#B{jHO2s(BA?w0_WR>YBrfRsK&X(nO8(nx@>x
z4)xbix8UENQef)<ABj)u!UZk98`9mv(NM=YJ<q@quOKY_k<8hv$T#-TZpHx7<x)M8
zj<43Jw;TvAPZf27_RY(gi59Q&KTCj<Q&djeHMt^>)3;r~4zK_wh-r`<$^ZP`Shl5f
zJ8vGZHJeL$w(R}}<flM|!kEhVT?b9X1!sVqZ7vWAp4?3mn+?v<HWyWaXI2a)z9^t|
z75`aSVKx(d-5)n2(9N0u4Y%B2tm@PL!Gz1Slm1uzv9u}Z3xURNULBRpS~iayyDWB|
zla2NIXyHNfNa^GrTdowV>5rL-aRPPr!y8|I9Cov{?d&+I_Y;LgX4Wbx-8DjTaxBxc
zQT_oUX`%i<!u{;&$KcI4yF`-7dE+*vQybgyRz%NjPS4r0GP6)WUO#^lo>mYZ&V(4N
z006cDBtMy?Z&+65k=(u}CnIlH=QZv|2rh)z!{)~u2xS?>FP>YRjlg&(mJAe^JQk05
z)VQme()1>y_r*|uww?K+v13<Tso@#*8R)pe-Gd$f5ti{f*uPR*_sES)tmk=#ulq-n
zO?AVPB&1C9Y;>5kjZ+>LZc7oCBjkZjMnq1^2r0ZEUl{*vHg&xXr_1MkP3f5F+-cdk
z36sG!zCBC3pB>`JT8}qw>Y#(2SBHk(F^=GRn_Ivubhl@ZGrxMV8*#*Iof!uNS0d8K
z@q&#<B5<U*m}KxTEe2MQFe^Sx%1Mh>Cdth#C|f>`<!^i-{o{)Q1o~YOs%Qst96Dk>
z7f_aR-m-hoJ;$4MI$2l+9o>Z;q&hc#)-G3msM#puu{+UtNMy+wu!OVh$FBX)u-1pr
zw6Ju}7LZEx3wpg%aB5{ATL6(nYS5JoWxQ#bcec3;MSj;YE}6GX^1L8ZL&a;*^{$Jg
zl*6Uss)rCS#cbkoWq48b&JkU_98wyZ1hn?WFz&%9j)O`p0vF3Bwrgg~NO-^&>6O!)
zhC}!qG~%qm&#wC*0I2*532mQ%xUAsQ?Vi(giyP|LnI-><g2D#KaU7<K7Ts_CNioAI
z5C`_n4s>tHd-cmkV0-(kAH3=XeVwlu-*$%^=(A<p;3Ma0n8ZzXckSWNDpONDmZKXg
zWT@gztH8Bb{Gqd6jC2X?_aEurxXSlj=Guxk*U3z3H_`4U9k}TRqFL{HG_w)9p-Y$n
zmbjOv-@)LQ!{w2lAJ}ck?N@MHI0s*jQa<#~z493XT0$kKvg4N%Uk(Y#)9BccFlkZa
zaAw8$u|qsKOYHAU$xzbrl_V!GJ$k-7X4AB^n;X=!EjZL|G*;tD`6r~IKB6moD>?YL
zXPNf0a1s#T8(Ee(;v#g3_@b;TD4sODkwWbc{b`^8WD5|tZ#<Tw@fq1E@D)ntx>=iU
z8hWRbjj<{*$SHWwZN^bMZHt0O!{O1iZTJFI6+$8z;i-Z@pl>WJ?s|qKTQc}R)@#jB
z1}F<!CdHNleQ-l2I`+(f2&>N!!Hf}W*>E$Rt!M6eCPTlaxVTz|<vN{7o4x^UA{O>1
zKj*~rc&yi}N0z|nl0mT8qlo328;&nooeoeYeB|ic78*e^e16sdS|4t<k7O2?0FyA*
z@7UAwDzVGl{H@c~tP?@O*J760NtD$Mhv3cx(SANhMfPZQN2R`3+mDDab&}gK_5Ceo
zeyazyzJ8NC`$Jk6@a5baBUPwp1a3?xr(Z+66#UDj9m+G&E&!8rPAeKWINZ1Y?M1O`
z4R?B&N%TBlL1Gb^oAS36a(8yv7_a5@vuW!`2HZQ}?GR8~v|%1wenvrN`^FO_!hME`
zv9xO<@GgxaB)mt&N|{-4m$p;6X79l4#JaVwg0nVG`MioF?JqAZ44EJyi~uXP%~(Ic
z^3+aiXGwCz4zD`T{(Pw_Q3q>{mWU$WiAH1P5Xe(Bg~;jB2;A;=-%*CyL_KWzaV0#T
zuD7(u__EK4o~v+NOaz;YBD04{FGbRdhX@ui2DVs8q{+`$ibGlYP)becs=Z_Fyl;|}
pZ8RX20HQ=j%?iFBkT+0+yXj3ZfaLW5m@fEl!v_C9ZUb*8{}(r?m_+~p

diff --git a/docs_old/logos/harvard.png b/docs_old/logos/harvard.png
deleted file mode 100644
index f6ec08d7210d429e8f11f8c3173ca3557cbe0aa3..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 31910
zcmb??Wmr^EyY{9*8YyY%l9p}<5G4epn~{?4h7pt&5ET$491xW5ZX}fMZjtWpvv|*W
z|9?NeOD<=CVb83!*Lt42R>TW6CH#9-_aF$uSAGi9gdkJ|cwUT+3I0!KpU(jg=+<(o
zau8G!i+g2`0lsH|Kh;!)AYWDpLcE2b3-BYv8U%UpLC}U71c@g>5QX#CS`A6?3rvgW
zN-*g5?q5bj&PNFHYfy&Cz4V^iPWAF89lhvmvbORXre;QupPj^~)gVUGdC9G}w^0E{
z!<{HLYe330oaT<ZYLm8mZZ~iVlS+^$2V>vU{5bmsKdVZ$^!H9*eUnUO@uYpca_J<x
zt8{-`o7HJgkC=7e*YmW~uZN9LSlgnrC=J2?=ZAH%d|caDbZ53A0RluA<lmn6=I(vI
zO5J}#G4u~r)G^%VwXlHPvbM<Hp4|OE{%J7CX>V%b?vq%XVar0zk3$?Hm{fk(|HEWZ
za}soXG~L4${J&SAd83Q+f1j2D@!t>51Bw0LQzOXr|Atcy{r&IAP`ds*NYuFhju(UB
zzhlOg{qOMqUoUv{0m^?V<!tvlom`Ja9@o~TAv+O!G?yYx)KnxsTe{Hie=m5A_N1QY
z$j?oRE|(h>iWXv7F8<5Cm7WyyEfl?2bjAGnb0(8QkJfdE4%_t_QNaKkqryu?W1L3E
zA8)ZkfB$}v^eO300Lf6?!;YMb6d|(onm_ycPsB>mnRciI?tMXis0#eg`&2{GsS{@z
zjid@`tl`mecwG{8<FwDTKD_!xQu3fEA)&bo9gkS`DVahRPUs6?$Y$iF_qP_Mw=^L*
zIaw>MRx0uZ8|_p7e1&W|g<f>!KHW1sbH&GbdLQp85xS6#bl05K)u0h!`*gHi2&P<~
zC)_1~V+b9MmCZ8ap>rBu%K(#kd1DRT{9~PO2FHwFBHl<63s(q15bPsyovXl}iZYJB
zG`%JUaR}_bCPvU?$PWe~JE~;m5XkJ6ynvc=w?eI1GlBa3N<XT_hrwY&-e@`sgv6yV
zCIJOB%_MAkY%*y=Pw(X7!W!vU0btU-t8*#G2}*;Z>c?ae!Qu9U=uGgdh=pIId857z
z#1*gglC#MqQu{U<Dth1@(aLUhvzUy}ETUN=kv1>2M>h}8H&j`*LS`ME{yl>CTmS9)
zuud0i{qpZ=DfcabnbTKCVrJz(148d{g5z+Y+*~S3`1^>6h{EmZs`1_LRmsZnpMKgJ
z2q0d`5e;F;J=M}4{YBZ0{XFr(SyKXkHn9dR@n|G+t$cdh5+!1&>E778a%(j2)18Nz
z{H>p69FbKUo=bwdVh{rp6B@iTrhQ{H&&xE9e`I9j?k6cCP92eCTzl82^G`HLDc{$%
z>-zeN3kV8+`0(LPRu=QsUK^oaiNDxEbi@yh2AiRmcn<{7nUrE+t_M3;Bcr2o!=CCD
z+26lE1lMW{Bkb(?BFOKu1f%hsFxRVbwPaV7mq+C$$Bm4NaxofW==syLo3Pj5I-dR_
zA}%>OGT?NUB2Tvnb8yG+W-*L@>1bFztaG^~hE}53s+Y=swwBDKNaPbMxb}kw55ARB
zkdc|+T%GULr0|c#K{-_qsIW0ZP;6~&A;iCbR@JjR!ajd`dVVdmgz$T8yzEJSjO)~R
z&U$m&bX)OE(D%rus}!%?dv9;RrE5bLl4$6yuCNz`j8GR=ZmxbAa3?egc9@_pU<?qQ
zP2p!efWo`lDGxEEa|AkqG1RdbLZt@qnuYWe#!qxiSd?qYnhPLFhziYA1nU47QiTjq
z)9<4&gO@{m?^X?o-U@DcUY6cC6EZ7P5tu2ZsD+jI(GqmhUgw7t-SaOb^z#`<8DNx-
zdQ(ddy+WT-vV!@o<2fz(3m_a|<#I%CKbsVJ9rvY5r1&mMOX4?^*eEeD^1|~UI`h6{
z%#C%89SZLvS~-7;vKPkPeYm`bcS{f7Vt^hko--qTJCdjpbng>?hyDGG(nYGU-@3H4
zRBG~i{t)7}&x}W3d{H;sn45kSlE~LV&e!1&`+ILhcg@Y)5c`NL!aN|F!?_6`8ge%&
zC6{F(cDRqe$g4UWI4I=cqpTcJMtJ!x2uhkap+!$Sy(sofr(yTPMEotjsbuqp*xy(O
z&=MD*6&oe2Bscs))yn~GsFTi*d(L<B(ciy+Gcq!u-rinHNpHSN$Jxkeye1nu2~X~h
z7}}`kyba}tQ-|I!^z_1IZm;gYd-u+J<7W<^&%vu=N4<jhk01Hn{;5RMN|4{42Hl2H
znEDg1rUyyu6q&rgpF<O&zg7z!#bJKN*2-3CsoQOgGH(ktpQ(0X=i%x2`}^bmVZRJ~
zZ=nT(KK_2|vfOd^tJ=QViHY!O$2$DTq6c0#hiW}(Qq=Lx6}RYPV~m1=Wxc0)It8tb
zmpf5k1g)0}>vmdeR@0&=OuSJlDk_phT^Iw4@;V1OLQZ48<uZ6juruu;=}1;ED30Ao
zZ_o#(32mQK{Cc8fucKX2N<SOAj0{kQbsn{`-Y50ccd1Swmi1hie-mBu0EQmYvXH>M
zE&(~=2JZ#pJ$ko{LPK285JvcdX-7k;Ur9^S5gXovxiHdqFx=wgnBLFUQ5TSx>x+j7
zd<bF4?Gb5U65~Lw?laSA@i+`0`Gw=d)hwdrNEP%7#_(a~dZg`%VDr(s6o!&R>o8`;
zypWSA&1ilU$`(p$Iam%>N+h##UiccW2B}Q(^mfYfoU}!NyvRA0m9CKshO4XV-n3KG
zx7^%&V4HSeP#8fPTXTmt0$-fR^sP}K$HsFj8utOYS;u;wMBLLkKVHYCYtOIlL(1&D
zr-s~w(r0E_tN03Tgu-@TgKo|#h3zTF3Jk+~Q$&`J$4xui2xYc{+T6>whO<@Cpy&sJ
zm|n9Ux?xMaHs?`_PoAJT)~>T&Ez{pNU!5O5c`qd>c>iuS{RyDkdL&8{)BgZ&1sy_L
z8yy^tEXeCu`g7_DE6BN?Jnn8>DLQo?V&i)m5~<7m(IxWw37s5~2i7vR_`}oZHyKw~
zS2sdK8Rg~WG*IXL!IZVSpUKP+8=AN3a5)L6>8o8In1<m4jWL*f_|0rmlA~BK_c-ou
z>&{GpSNe@A!{s=zBXeCWd5GuW+<!6>xwZAa{!8yaW=N@u$w+-m>3V_sP2ip@WO%WK
z!ZbzS-1JiWQK=p`eRA)=>tzoI{RH~w&z}8ZMBb0Eagm{Rr|YA<jLo(UQjRa5d0o->
zySw}OoEwIj*~Mc~shU^6NqUm_(c+mOva#h%Gh#!S_*Cf7Qx%mh4LU9^FKT6;x-9|!
zqd}!_P1MxX@KUXFu>RZj7Ta9@eX(X_W5a^X_}u#!lH=otHg0z`wYAAE{8v8fE|v95
z<EqMt2n!=4P6{o#2*Tir+}*K$l4&@w@<!Y+o==wN)!vamhlYmkDQK7s32n<lK_1`W
z0dff|OrVfCUH{y1b92*zXy#2i*&>&euQ{5iMI4X1O0O!A>=Z<U5BAiDJ;SJ=zUFRE
zT1et*xWRa5Cw$E@wzEtjKC8@EPf|N!#p#v)jRfM$L_<#o*9>R(fqo?NimLx|$z5E8
zJ<^pGQRhK-@WQ}g3SB?Jbo>4K5#I+9{2SX=8=9$fOi?h!Vjq7Cp)_C4N~$s;Tp@)w
zo((K}Ub^Y`-|&0ZGGK2qEBi{av9Ud-^C-6LCQn4Nn{r0FV#4j!L@<4Np-Qwf;${(O
z4)@BBuDR?@&@LwWKLco1f7zdsSjw7@GBRIhQ}vB>xo*Tcsk56bEGzqBcQacxg)g9X
zo+vIYZO(z1lqbWNkr5>$-~Kc{ZaGGRKxEck>kOBSL%!b{ztUkklqp6OHM(Trlb6>a
zEw`8Y22bgXn3x#A#&MPpm=vS;5>p~3s5u@zD&R}$6?_2S9Q-ZB1<w!W_cSzJ8C`eR
zF)SEM314%g{cVhcgCl~Z=iyOPQTZwdqb*1kBW&KhE@@zoJowOVan1`@=@)OON8u(>
za4~%Tyt|SB7A6;;g#!_wIWr@)Am%mE+ZWKTe!*ByX<1p3X`N?56`SQp%X5#sQG+Ki
zdXI0nyz!RjEsgbT0F@ZVXC0zMZ-JGQiDZxM;CN2qHDZYy>@1A1zR;?hYYgy<nzT=A
zr<4fiXZ47$f5$JZJ6!I78Degyw8?&WaKMdZr|Y1Uhz?g9cG=0;U!Tg#%7XV=9onkp
zxEa!5?Bm8Wr=l;Y(8}LZ$`J{BSkwNTw5MZ^Txv<qru{=HLD2kgpsI&0F`Q2A>C@`T
z*A;UBRnKh|)P~n!@SU;VGl0kxa8&h4G-7`l?oFiKYfeZ=7_iP9#7^(0xR7{*!&g$M
z8%4xV>rgdq_*sG+wK!+Rv&N|4M)-@$i_uw~CZRz@16j`Sc^F+c74h}Jc%4vLSy>ZK
zw5SB!GHBLE38w2F|JOUJJ$~GDI%?3yU(I@G-cmoXsk+)bfQ_6jox9u4Qs0ER`(RQr
zQ^O*9_UE{!EX=LDyL&#SJ!5v!nHy!`v#O}a^*7~*e?pPP6vtZoBtceXV(I*z&)8ZQ
z;{&G%Yf<WRS0ArCi<-P7UCHVD)em|-nS#dL2TPwcBs&u&+?u~bXR~KWI+)?gzQUlh
zy@qpc@MG8J@y+0@Doib>Y--Wj{~P-s8B2X>6>aT^aL3^J$lS1Ru%pn?*Vh(jdyAS*
zFXdG*%sy`@q;B9h{(>M?Rn>=!NLzzgm`Xmk`J&GY?TXLX0s`L*|9!HEZQ7+5_|-I@
zQCP_HsehsH@6JC7omP<p@AUrhXIE4uUUhFCBGtyn`8zl-#MAA@3y(=OYGSfhu1nHb
zlsVCzUs%VMF<`r0+8J_ia>frf{M0nwnkdPfS#-V%3qL{I`<?~|Ljo<S)57H%)LkW%
zwTjy4_p7y81jUuQskymxG({@d!OmX@9tOtsT1jNdR658J`MQ<8=?CKk9E|`TeT`Pv
zK;iz2i19*zmzP(<#3REtnP%6QmxIHFGx8vuyF2p%+*)o`lb%t0aP_dD!k~bEUUHlX
z<t(p`z$?M2UFMJQa>r1Ub6cV656uSAzj06E(s$>TH$CefV#`0pJ`UqpFbhORWc1X3
zt8TK8kaqhFw|#;U_&S$h+n!+QA%4+V!ot6yCcY+n!NSQ3VnQ8qb5NN@_kZ0FVG0{y
z{hR@9ZE`GhZRX`tabAAn(AtxOi_PUvNAOZ}7MFiR4XO}2YN~0cYo<KhSfd<~Vkw%~
zMrJjQ9*&AjRC1w4tzU9AIO6D#TIO(fv+;M(4eDFFJtFV^Db=vk@t#@=x>2`2U=C^F
zYf%53J}Qpaoa8w*-GrJ@V3Jc@Jb=9*kjl7x@n)(J#fBhk`C-Szpg5Lh0*~ITUo?B<
zoOFKBD455%j~@l!gggFn6tSfJpb9g67VqxRM1;`JrBK(I*7$2HAIw4+RJSI7byyg*
z4B$@NYWl88<KKO61Dxg?E0kzbgv`|*8lB+*u~pC>=Sj^m6*qN!v|+b!rsC|Z$y^a{
zdm=s=rN=}|^?u9c{#0jF2|)4rrcKi28c(!Gbyi~DLDAp$-g93*9ybapy%4WA-D!T`
zozU#L{oOmi29JIuF}Q6IR2DiIY;{8=?GbjXiisVd%r1h6arb*$N=h*_Uj~Mw{*+!l
zyuzJ69>^IpP#so#^66L$RzA1rTu)Zy|MKO_ub8sCL+=Fr>^2<LTNa)3M})Sy*Cx2-
zVL8h2LJk=ty;Bu-4UtGmPVClKsVSuR5FO#!<7k{&9nH;@%|(`0q(T&**UpUa66012
z{`J`KV?+swJgq7UmN60vG!3nqam9@h`$w67nMv7+?*F&ICbxywYvzsLPEAf*eaABh
zGj<an*UWn|q%Mz5B?*CZpS9VdWT7&zutneH-j~kFLt=&fcs1c5m;hn3TeFF!{Z0Zl
zDf*(T>TnF~QY%$1)URQ#45}a0dX7%~$Shi_fN6VBIAt5Eo4xTfeMtbf_ml{ksv}l<
zE=t_H9el|j(!W}V*(`(U+U6;q`kY1C0FsS8+%rV+mx|lybgXi!oU{)k8Z?BtC&xUw
zv8yo94r?pP$Qj?M#3*K55Omxa`MDk0^b}=cm}6nV$CndoOO?9tZzMS6jb!YIrH@T`
z(e)Vrac-3s{aeM%lFxhF5r`=58m7ao!*g*-Ye4S|o??V$R{bOy{A?^~MPD}Qk6I+&
z>{ult*x|0)O`@S9{D5eXYoI2fyeM19L3kPkx|ex%r;^Zja?!^#zqgK7-Ci<xTUy!3
zl7aJ4(hbx!UL*wvqi)Ljp8%jS37HI@&q&);7^v^;2dwYB@lm|R`}K~(B;2h}3=7c0
zVm)qAz*3(O^CEF%HApo~O=Ev*<+M6hg$;er&bDX^#V`0u*>p7Y{A?+TH!eLrwr;1E
z>s7(Mdq2zi#Bn|AXvxUPC<R_&1I{}*Fn}<w^E3ygORv`5whk#IgswBH(`tYCuh&^T
zwcsZ3uvf$-d*QV6wHclm=KY<f+rX={<=C$7?kA3pJOq+^vR{2Sem+d(rPYVHxVTE@
z11~G4-qZX7_kSa=sN<`cn|vo$jx7_$U%<JYXFbNhRaTN;ZdcJO#?rk93kZw#<IiDN
ztZurVVL;PM>?cbHoW6hm?tQ+J_@NoyS=1N6gw9Mf5n5V2=R&OFPvZ=rG|qiuUPpoc
zsm0TM`1tXofriH5z{gFoRJE$n*8#!dM`x1~<5p$8Rnyxp#uUihjV({i8Fi)@ey@nN
zTt{^Ju4ig$jo8f}(Q1qVy&l!)PPkonF8K9QTZ*{(7SrtU`78ajN6JGCXR|O5;fcUD
zdtvpR2;5Xh+-8p7Aq$$?-1FsRvDceC;b!coKFxTHjx{UfGhXun4+n`GAGDhBF+6@u
zb|(zfdTbg*eCXoi&dptFIGuX}REC|}&B8dMdyxqNCzDw8{(C6UtwCEDoiDbfrKR`v
z@puWqzMnpQdMlh+&i2+{+VrX3_EZHf7M-_QN_Fid5(jJbBGDDwtlDzVxn8#kHzl#C
z)#@yOp66u%rMNr0u*1~b{CsZ8_Z^-hZ(dniE>6H>+o_$-tertQ4HI+w#{l?5)qJ2t
zVfE5IZj*XS`hX*15)zWBPZC`~K`1eA!}M-20ZRc5$-Z>4S=3o<-Zr)Q)Uc|uxHKKC
zy_=+F;+`!<qgQ>+Sy*+xv5CyK_1xXmK97PS(Eq<s(`vc89^61WM`s1ghf*w&O9eLb
zB`$I3+S;qAXV&$cOd=v8@A%0bQf<omSN$HEoKp%~sU%Nt?~M6sGbU3I<xR;tP9WKN
z28C2(FTb^-7^%JnxXy?5qpcX+UY^ZX5M9ojxLIS?!Ejc1mk$~qc6D~zp_CMNr<as;
zr@2Y2;tj>c@!qLfZKV#P#?4^B_P)J7bpg{FHY9bn5YqMMHErnOCoQ!%4A@Che!@e;
z!_#`uZsVn6G|jq*D9~}Rv~twi0V&&C&q-g(PH~ji)YQyLD0b%3v$SNHaTLk#VBShh
z3RPBqASw|+xzkJgeuQsz;JgE7BvTE=;?agFg~qgLwVkwAggCH2dW0YG!%VELh@vrc
zA6$^&v)9SiWc7XrwafbV#HHi$>qeJE9^*)G``MXCkT}i!@Q#IpZeGIcfKFPka;_W3
zZdwi~K$;#e`Wqhy+9gZ&V$!FU>POb-YAU^pMX%Iy(6b$o+oZ;sWfEUH?f6Hmc}BlD
zzT_^{1)myYK_Qj#_K)|@%pIPDWEE2|+a?Spymq1uXEw95vvC(=$!I4lkMoUbS(Jx<
zqcs>$Z=aZEwC%FM?T=&&b4p7~6+0#{2C5AGBHJx$kagpc>sKc649MnBGp}<6Tt)29
zoDxlZZPyL~JwU#{;R99S1>S<$MCqOzdQNn~*jP6vo$uPW7i-lqZ{x-{9m?v?Hso}A
z`sv<KV^S=X9~oucOQn9M^ySL2G+Gh<a*_JWZh5)T7T{<OZ)aUPX~P8jk0jV{!4E(u
zwz0#GmG<=#{R0+ZPMf4BGla6RkfkysS`8hRDVcdDU|h0h?N{+D%@=EH7rm9Kw5*oZ
z|86@qDz5HgE_sbOw#LJJQqq^DUsB5o-_n4jX_10Tr0uv%?4i?ToHE4yeyx-#)GKwM
zgXA2NitHue<6l^$TH_x@-s=gydTu>kUscE725)SRQov4jen?t_>)mvH(l${7M}bU(
zu06^%D4sog#%DW}6_*rEsv>tX>t%XfP@MW0k(<k?K=i?B%Qt`VU$a=7$S2sn<{06W
zw{9Ch14T?dJUj}}&8$;oTgTNmh!Gzw=^dUe8k6R9N7IN2d`mS>?iWw(GYz--oVK4i
zXj30Q&MFJ5WhCV5upqDPTy!QC?3f$>(zN+z?v#pGd1!H4C8QwSddn`V-IW;;z_byV
z*S2^@hcM}!QKBFvRiRvSNa5GX&J2jCr+`#g0o|u~%$x^`t>Fy{oCi|qn5uw6)vGYY
zF1N~1prGqx3SlS5?TjlRhqT{Z?$S?<5cQ%`W_^>pHCXF<i~#iPTfc?K<>SYE_le3O
z#rufGCJz{3+lW~W1@ThI)Ao$49hK6gO0jV<WEfwPu746nTBDPpqdjU?{k`+pzz_6V
zr={cdpX2(bmL>H(#8UZ!>wtr}`qRAKoAGRntvcpSCc6}a0PP@gCqnXAGzUo0&%~t_
z2U{+fU7l<|iY^J-u+J26XERXW!eG$j`Y`^5Iw=D}N5o2B{1|mg54sVH5nT$`3C0L_
z#PQp%7=NLy{SIt+0-#y9u%-JWpJV^br6Q4&PAAMyS)N*ukK3#F=p;1r@OVuA$N(Ms
z_H%cR1l+?)@}_7R_pgo~8-pfsMHuj9f(HVB10Xfp9IAh=HCE1CIMi8!>wS7Ax(O>4
zP~+(F(%$aO&f`!5+T=D$3Bf<=b)eSBgpW61!x+S*G&8Hp%h6{&YzAa#1JZh}sMplA
zkIT1>>vfNpL2nzJ^db<ZdY^M#1Ci?4eUpcMn=Kxskqc8t3&WLVY>j7m4t7LLHo`RE
zSo@W=_3~Yj2NF|>+p5|uBRe~!t&Op{*rVt@*V;|XgUy6f8?x^$DSCT#s9|Ijt)(tS
zrfck>ozq&CgUVOGg<6gJ*Zmwr>;u2&Q&NpH`#m>?%6&m`b+Lu<XZ#&Uu;hVg#bx>i
zhJW-!Uv|)y<4@KL#QCV+C!&OtZ~v{4rL6PDcjQ4ci>{WT|CEpr>Ob{GC7S}J+Wjk!
z+;DjwruSjr@Ze3*s+1kJDaAz}&#q={0|Wz{VQ<fAFdOk^R;SrC@En7Mg#|+3QwwL*
z*VBa2dg`$oJKqzXdbu%E<MyyyU@q{|fd<qPz+cu}s+o+xc{j|(k@ajIdm5r*#NsAM
z`ESGblDAb+=6zOElcWR0Spf!hYPu#@m)u7u=QWvm9p-pe180HSx_7D|SCp>x16+YA
zp;@5v+(g^h^~=u}Rz_Hro&P}Lfe6Px*pC7+{trCBK|!O@BHCeKa^-le&w+Swb^&Of
zX*Y{_5Krs@f$it8JGTPOTYpuosC?2s0<FY2rT3p4)Zvlq?cb>G-A|SM6Ma!NHDKT6
zjrh)UlyG6KW`L2}Fc^pzmt(JEW)r#fgE_Ra-&$;j;{LQ*t==l_HY?HNgLmNN>y@Bl
z(KsV`h)HcRHeLG3x~Sp#qyH#6^y&tm)SK`&o-N_p&((|l#TK72m@q_*3pa2;C9c7U
z8^*8~He^K{W?qf_V=*=`Fpz`QiVMgQVb+or4}1TB9R>0lryLoep!96LPn{-@T5*al
zvEg@@c}Z$%UGEqSXtV_4snU;>RJGMujf*jbfz>DnY+_?RBI*hT>`sqtF-kZ!D0Gzv
za~>iDVVQbP`DPsx9|AZbg|Qo|X6F`SwKJi;%iSh=5(HbCx5eKs2yxR(Iv7ss<B%ig
zC;BSd0tK%sSaBH{#dGf*CiF!`L&WZvJcQYQ|3o$v_A@tk2DNK_1F>WQeZY;i6H@gn
zO0`~gNzIXNw<w9G4ENwv9+HkO>62gn3^Vv<p{%CXCjn#Ld^*J?QI%O^ghMpmOFx)D
z{pLHqga0iClHdQ^Vf{E7np3$dOk{0M>jp=y+<FN&b-h-|Z-`5)c@g!I5u_MujLT+&
zVN_}VfSkuDTsu#vf*8?K;VK2nF(bP_Me+FNy8%^VQX?4(6X^#2xLC=9r*~kF^6_iZ
zO8d!R>+}HOe=9MhO^ZH284YatlP;*1c)?8WAT^w)mNQJdjR01}F|1srfB6?>og-pw
zvq@qzF=8!d0W)RJ44^v`2=_QLOAoNGl-^9DS=5U<?rxdRO;(^BxIi&6XSMpn<+>6-
za2wFR2CyiLdonzJoi{2jHD$1f9SDRJxHGb%KwEs6cIflrNhGw@g*q{(7`|t_K<!wK
z0>u6O?bZFW@tc!Zg0NNc<{g~o1&x+^kIHkFuK+1Bv#=}zWP%CDJ6!E|s)%4=hx}@B
znCrI&vm3#3%A`<^k6}XE&|Ri)7Y7lcH}L*`9!f}`#^1bg<O*4P#cz24p>%z_o|%nJ
zhA=}%X~>)>adSq`SeLXtkEmU|gE9(~a0BT%LYQhqB+w$!CbAAm(5${Z2@Guf5^l1=
ztp5`*xwzB(?UR(IHJZ;}9Ffm>RZuc}S&5{`oA+$ZAkp3W!>KHsd*3pganfYiOBv74
zSJR&?j!fN0e4u_<Yxq#E+uScj7WrAi<~?5MTZxki_?>M6Y`Vo{yFy#79UF4l#P07B
zv5RB<PjADZGb-7b8x-{;+uHP3`a63VYtEw5C>d-+3dy;-ZkW#kfG0LQP@}a-7R+FT
z8L-&t`zzs*q0wGv(P-OoNwt-UiHQKQG0jTOYi;fM=<$ZG-@lbH__!k3{VUyOpGLWx
z{9{h(750S6Hzk3)fXSB?^2!nL_30Y7oL0Xv&&yt8=;#cKcI5MWhEP50m(iuTdb=;z
zCau%|Lxq5|muT#%{}P3=u?AUELTp%Rj{<R<HqzW@Q^WgSP*l|UJ<%kxwYAk#;l+!u
z?=Dfcc1WFHNMaPYe^`krpBo(?UoL98L52JQ>`fB);DpRtf-%mn4*N^I_bu-5eolIz
z?^8%Iua39rO5mvG=g(#RZ_fMFs*tM)9|YgxQ;w%w!_t2+W4dVr<h1zyRSA}7v+Mea
z*#aGDBg^hd;f46&jPoWle__h;JLr4r?!`RO;p70-_viljVH2}lRvaGg2y^%1UQ_`k
zQEMCcfoVqlHbQAiu)$VD!@yi=@EijN2ZzDcqAJk6M<z??Vpl8PzI}T+;3F!SZ963i
z`4t#aD#sIG?a?93X_=>g+D@&j<s4Eti;up6PBzP*5G!!G|1#R4MWL+3c^h6^{$QyU
zm-iPCFr5E>P(o&8Fyum@#7F^&%EsOv9h!0Ness2yXv*il_Tr_s_F`U9WAN}go|sk)
zYjE%qn?I+}x0FFX^iUnt;xgC%uVVL)j|1gyfddTRnW^dO?N#vf6ej1^Lqze^VvifS
z;R6{Y$rEG2Z?_@rbZ)v~?xus<=~pLx(9DkGGMGsYf8dUVprPH0ZWA)|D8Xh-A?;{>
zV*=j?a9*OpCX_7C_z_UV)=RmYs;WlK&CS1+AdyS%kQXCkj~==SbjlIwDsr?f@c!p0
zkg~@i_tz1lLfe>-6bWLxnpE=vT)newCc30X!|GGosGGgMzJ4aNaz9$Ng3^IL=wP>B
z)vS-_Mk;~sd=tztgSUX{F4=wdG23=kEVYkhE^kAAkO_LM`h3{0H3_%bSqAZe-&w*C
zR|FWv_S2OGvqF!6E)N?34CsZ4NzBrc1?cG(z^R%jx4r+`&hA589O6!bxRd<DQ27L?
z6sv$|vkiaY?@@UvCL==+h$=&5WaNTporacH$jr=4oam9a<k@~F{i18l6d3|Y4?1~|
zk4{gvr#1BTBfft9dSWu8@osmA!;ByYeg_9#YmD5G^+Kj4)7!QATAp<cW3MY3r%Zkn
zVHXnW5nr&tggA@oN=MB2UV@r44s(0St*l87%tA9pVp?!X9|6QyHMLtQgq?Vb&aNuD
zoL+|Q)b}<2nt}NqtHY-`!!->O<0U{HvPmX)Sg7~~Fmz{LeLS=3L*zA6qbHiTb`>#<
zH2tOau_CSU>rw{nL^SUVdSM4J%Q-g5I)%1VeL%{cYA!|S{0E}f)J(QgPAd_;4T-Q3
zN1%xh|ArhtTb^tGG8bzzfra#03v2J<z<D8kJ1)J2m5k>Qs;sByrQ>%=I(07MS7npM
zV22hO@J?OPmU9Bf5pDB^HLeSvXVHSMbVgn3_~jZfoCHbk-#4TcvF@V*43+KWjt%Nw
zf}vx}qpF#qf>Nbv^ww(7_;HSpkF5ndT$w!L7_j?2*M0WYc@>TK=v&OevM>YVu9X?X
zf^PxKkG~@$m(z%cf>{~$>chie)>Bd_+c{6O{Bd^+iX`GY>?jEo>SK>{VCGvX$4g`E
zovPIZcKkzG(|O^mI(+eNGto0YX!m!iQRXR4dwV+pcnzv^cFf;#QgU)0?5+A8b!L<^
z7!f*lFjO^7Yd=m0y_fs2w(qUkhEJPXjMjXg_2YUw;?Q1ge4yemJ>0z$J!X=Xs45%z
zMabU9^ILW{E3-eemWH|$-Nmu6L)43}@&X&+Lvy$1Ok*OOPAPOl3u7<y?Fz%jH}O@K
zl?)&WPOs`SyP_Fc``PBVhSGup0t}IijrKnfVL~@aF5>=jitEmSz!sky(l&3QywuE5
zSh<iY&K+7X#Y(`CjZOLF0qqP4IphNk`V?!QqT`A9bVobN>Vh%jf{OUm&T4NdchLPJ
zd2&+Hza)cSH2#cQrbih&<9sv)%k6}2q)^p-+aUJ)&#&PoucA}AyAN#i6DVBw#{<rh
zhhih!jJe(9&Cc&lGky0xXL<5n#*tzwqF<%mZR${q+4`cHau9x;K>Fije+!rjHgitS
zV#^=KENn&r45rYOc$a5WM1QHhqJ(&fr9D3{hIMeh%&Wt4#G-X40s(<1NZy223R#tI
zIv#&5O-<+DEQt%CJ$C-drpm&>#N%haBgrbydt<DSlPSF?^qtci0|Z5~N;5s)nlE1P
z$!U*rXaxFJrsVk3G)IV=WLy7RS<$~(=aG}||9bHn=KdbZu6{a>C7-ae63<)!IqR`0
zYoV36(Qo`Y$H#25haxmgUa5DK744t<yHUbFSX`?EofVn4s!6({q#QqnlJp}pz?5*I
zw;|sct@#wxmYh$YKR<TeCJ&*O|CDc?pf+5NIooeSQ8*8sJqkd~A9$#@dwx1Mh)Yd<
zpUI}K7#JvhaCqo)JgQqof}mA>cTxf2rQf3k%H^G<e#hg$ss}ntI|xU-338fwd+FB6
zV0e_{_f4@yU3FA+<42}*4}YiG+Q7FvA08+b<R18D>hGNVw3=tv7LfO3p*yw68OV{D
zVE3)8tmrlPKBn{eOZ;-o$2l<6H0(#+rNz9-RUAho(tmjTJdXMI?EHqJ_GlimK8p)b
z=&)I_^tMGkP}0dExVZ9bAxgY)sJen)_=^It2H%Wt(z0Q<-t0Lsh$n=*y3_IUb)>~J
z|CCKwIS)TnjOCG}Amr~L-+b!z2%%IrpoASxcRKz#De3GPO%%I8M`8F-<L!8exZhX)
zllYjdk1Wc`B<f1zzH_InAAf`5Dyc@=el4Ub!9bLGf#CVNZ1eGVxulHDtrOTTYCMa+
zm~d`|RYY!~2)Y>yR4sJtcCGsy?T`N>(bE~X{h{Z)3aK5@6&M4{`)|4JOGprl<jsiB
zXl2&@1lIxc0jK7_!imosv+r5w5Qc;aaXaO<^oq>Sbaa~iYNL0<H3Q_L=TBko>l8no
z7cYpW0;B>i+=U)LUM#_gUD!G@=M%v6IHe*a45Gmz2U=+V-nl;o%yyO6X}+=Hd);nV
zvw^G>`FOXho7?VA?ATi*lJNlok3sbNBVf?rAyS~P8HYkuPvbvvbsw-?YoG(SD?+Wr
zu|$vC8a$0jJ2WU5`xyR%;aQz8%T#vAT&qRNSJY;beUJO_t>c|E!G{kQAH{Yan!0bH
z@pc_U7at**S+1wYg6Uo$8PVWps4z6h!Oj|K8O$EZRiLx^VSlElERl-M^;i#j0QYQI
z-TFh-9O;Zg`3fj@+lA(v?;j(6{`@JG`^zQ}`WPQSMNb~dcyMs=VrRO*hZ3>B-09_=
z+IIzcDnSJJDg~YL5U>-}#_7PK;<f#Oui|#ujIkiwzItf<w$d)}>2Ud~uj?F8;dC>2
zmtYCMU)9<W{^SASfFfXReh9xIkln=$){kd?w3TBy1g%ZEuwe)9uu~*w8`EwZgXF<S
zm&(GP_Xo2Qpg&gDt<C^*8kcyzf+5YS!T|Teugz1GroEJVKG$}NvSWf!oKW!ope_am
zMom52wgz>@##0@gmoG=S$eZs$KM1j#HL^*WaT)*hrA7fonh0PSNlG{fTi8Fwr{aJ@
zsA08SZ^frrX2|%-c-{&KTuIApZf@>_mJS~;G<L3=1z&n)R`@!FMiFJh*Nxn&D$Pl6
zNqbTyD1xr`vDC!ZnZuFc8kveM_QGL#$2!L^bals{k)8lw><EpL$3H#bME}lEVCUjO
zgSLR!4I&k=cm~Gjcw6yv4k$+loi1j_QsFzro_P$|I?D0+QetAn1$bN8+bViJXnS}U
z>q53dqzdQ1?1kyMv0FQ#j&;$$2Fmfmz%pyP(IJ;3G8q=yc7&ct)TCp9Y(bm2p#+=q
zkTa|*E)ZTdEz+51bbfxGi9$p1oyL$^S-dbTXFgjn75kQ8gZlKX=eE1}^Lx0sIodip
zJR|FUwpqEkyA<T)bi;&RlBa&CgSJQ52p)%y%}B`p6=M2(t}+zI9bRn}N7hd^>2k52
z6GrkLB)CAS;yD2_nlNf{Ru*~`1+SvH`9q$XzwtbVhrX{-xItzsinnpC==&o>AORTY
zw@Tlf!F!TNSy72FqgZJ^)=KcTt>}}GYftZYF~u%?P@+RwnE!SOl%K`iyuy1p&3mp~
zk-^=(O*9})L;-4M-6yB2=_VTciDL6vo=advyZ|0ifninXFSBY4zUmv^c)=mpCxtFl
z6>%6PF*5OJVX_MsHs*O9(3v<UlVO9rOF6Uhu%-Tuy?U+;Ko7mIE9mlUr#L&3vX<i%
z;+d~Gx(^O+UJ}&J&e@mfrI|%vD|I&1(ltk_4H>oTz3j}3TGHpyAYEm=2o0;gSY<+r
zfg;e9$3^~iu1C!p2UUItkk~({JgMUFX9^1KA(XN%ac~8Y5Movx7T;!n0GIJtsSo9;
z48NE7uMC|&0ViVJ+kQeC_$>n&=P%X|7e3Mw+=U%C#|rGoa;j=;slYmV0IZBVhsk`d
z-lwal2d1cqNkvUPt<tL)EqO-@MN`%A=p-pO$BQ}vfe)*&8*g27&L}M<0ddN%t}a;}
z9deNU0WlYX^nR?^&XA3b*H<93HBs#XU;9cZbHx`Uu_G!ZEF8*ho+5IIm1+fNJiw1I
zgSb1GX%fXLf41)A{^3t@K8+jvq0{g~s{vT<XN4vA#*Mvb`=l@JfmZiqxIE4@OV>1j
z45VGXfwQ&`$P0s%6z%%@`hd_Q=AEhKADJ<&K)b^H_!~W>1@WF*#G*I(F%X)uo;`nV
z&PN&d`SZJ@BllNUR*b-3P|PF-9dOTE^6mxFw^wd=-VGwCtW)M@pHN6h=<Oe4fWvch
zIhh4emPs_tEl@SliLqELEt-DKZW5$;;x-%HJSCt(<=wc6Pu)cEugn?!*0wldDV(Bi
z9{osZbw@B7c63L27eQ8o>rxBoKk6t+MVWXxKu^|-0wKG>K%tag<>-><$aNL{1esV}
zA5hY@prgd^7`D>syCe|qWg7_e@JUGcnSdM34l6+_`UL`$pR>UZCKd>~*HdfL>wqF0
zS%|D6SA_RUlo+oK4+kRQ`|}C4gL3ZWr?@>dM0(%_P89j0R<5G@V8MRL`=Cbrx|>B3
zqf|gFifslNf8<62A)XAVW)KwgmngkI?Bc?gl#J}1XLsdc#`0IUJ~=r#6u@Ejfumpn
z615YR4%7s|+h2G}tMnbObwxkE&>x@VJtZ@h26X%BY{odKb={p&LV_Ch_OacV8Sqpg
zL}{r2Gc)r7h+%dB7ma#!iu}E0cY<7kC3M5FP+KdmSD>SkB8(4mf#9xyt%n9+HWd{W
zGQscQ@9*m4F+<>1hJe2??}&Z?WN9XORA*2_i5|2Cg~i}+yT>TrcfnndWs;GZ@1c~a
zptoGtE1Rru8eX3%&Kdm<9vGB|M)&H@)>Ukj6*KoAL&JlyGduu^-^cdw46xoMPwxjn
zB4zr)NO}nSejysF)_YwaKpy&;kUfDTDuwUnQFC`Vwta!<G)K&&XtZ$b8>H0f5)^WJ
zb`J7^9so8N`Vi)&EPG_x4ha=J=ZUK2gr0M~?7yL9;d}G$t*NFa2_-y9*n!Hb?`s6`
zXD%mbWCHzjQrH<mDSPfbq|EkCSU~dnXt?^|50~M0u&v|*0;Iqs07J?^%sXagkI!*N
z)F`N-iLh>8@wPN7IhpEdN0#Gk>RW85wco#gpMl&mJ_L+t4xpD`I~4(CyQGsY@3I~Q
zdG03cKgrC(LIRM+{sr7(ew%>|#}&7WqxI_Ze<?1jsa|)fRa9sM<XOK&&_>E+jFLJ(
z3ZlcpeDgv(^yX6z0>1YA^w$Ll`e`f2XQNcV=(M?kf%Icp+oJamna=<aV<6d0@S7)o
zN8OZuU{ac!jai<zNs*ZWOS#=UL1A^oQu2W#@Z~*3X`(Jg>5Zw}|C9N#DoZW0of27-
zL^sK0kKDy#ysB<lCrar}0P@8L@%X@r(TlCuohC*(7kHf5782x3&S3)pPf&_D(g4pE
zME&kkBO_xQ!vw4!aljgueJZM&EQAAQ#JT1w4Fv$#F`c(8%nTq8IjdK_XauUf_g)JY
zko@}wSmUOq4E}lwk~^1;`8ErK_;9{a4GHAjFLk;ws%!dgJkM!+bKu((mkIMQ^FN<B
zY;0^0oy&4oR;&jHS2t(LDJftJF?0)nCv~6tM@C-K1fDJG6`P@ebik7p;dS<^=ZA8T
zWO859+pkh7t*!EKu#dlb1~%f~45*d-TenI07M^%=4)8@483je+cEoY=r%yuT+n)5O
zW?2MpT#I_GE`K{#6>adI-{qE`#UJq?aK)9AclNz`Q1<lB63L@1Nf5s^{#;loD=TZ)
z|1qyHAY?(>?1$N*7|r1(azlHjw5L~?P{1RouKr(}V-@d5=5_P6!|@sr0GD$YbYGB_
zA1-$jr?$uNuE<;6$%V@87<9G|89Rw)4gRuLXb<}=MX)*FB|u|kZkhpdi3QM0Si(hu
zavctUZu?uD`+GnlS^}-Kb-QXl=5#)Y9*Ed7itnl3frc>9r`-U=1||*;9BplF&?uU_
zKHLOW7Z$1|5=ENqj8WfT?N5*O*c>wlO3Ak%M9`tai(j5VlBye*Q~xSTOb~_!c0x7S
zA+_}pS*e3!AT-e&2dRu`>~N0aqTT=FQ(gT=&d2Bc0TAQ0#0fxF7RI~P@tN7yLe|I=
zCg03)+;sZOced0rsvE#u3O&LJ{Up)MY+I*E`^pF@7iAJV;EU@Gm*cuKHxS%(iNBsG
z8C=?yCvC4*T(EUI<%DmkhPCaRIE8a6$Mc{Uo3gFP#4-PT0jw%NgXzBCziY}$OI<mM
zY4ew9-QNQbppafq;!<LK?P=MQgq2-iLx>zr_WOpdE;N=W?fskDHN+KUZ{ghul@cFk
z9Yw&pd2)*-11zn#Fbg=GN!We@ak*1Rc$a`cBb0<XD6RU_-rvpXUAjEQsWAfh-0pQ&
zAh_`%DG3|&S)en~*lnbxe8~v<wIE&(g6xX`g>D^HKb8BD%ek?+SrTyR`PHT40Z{aQ
zDNq=bmcEq+`ra8pR4(0*suq6TSKq3HJ9OxEtsjD*(-W-;T8-^PgzUY$oj>TGiQ$dY
z@HbLnxSS0`u39;aulA2snUq=8bjs<kT5VFxyCKUk4hc5c0ifsuSsy1}0WW->Mfu`h
zYLV*e$KP#*Q?Q5!Ycz%)fka7DC)e$hFoIsWqq8<>Pu|H^tz$x-amek4Is^Vi)@fsT
zho;<f+n|V9c@^ypUpAD0_(on%uElBFft+ehk2~SMJ07Jh>{d=>?7io6DB)6)GLcA1
zPSDo5#>ji&4VE()P&wC&`1trZ5Ch52&nMvZKKRWBA4DqLg(bMWh^@c6VBUdkAQ~j8
zux4eQrN%2s&=Xfk92<dN4?L-5u)@Q-SLEeS;v8pxn1$C<JJl28Qt)hj=q86Kg=}7Q
z`M6i_Hs9@Z=Vp|yzP`}zmEc)NLRIt4hepnY`em5Yl@4tm)ukgA^WAXWK5x8B{1Gmb
zfQDS2q>wH5U`u9?<UJkpNL8~|HuEhhOPnWS6sK>F@_S7yK@)?4BID+ZeD~G1!k`Gy
z5cQKxM#-!v5T<b}HzuS(%KH$R#csQ_anb%tpcjQ``_V}qu&4_tu(X{>2Cvs7#k(ef
zD@70T<uk)m72Mql;5z_1)6P2db8{QD_4K?+PYEY5QN$DQpov2EXH|@2?%EJJN+7P8
z-&RUoXur`)_4Ud0e){cIPG?^f2qfKwqkyUN4M4;YOng*g^ZljvA9;Cg)h^2xr_QL9
zub-}sl`x^kl#J)=hX6b7S2<bzUMp@jNKbu;j}La3t~8%4wLIG@oA7O=98nIJ;<dHe
zGZd@AaAr7?+*TPzIb$WrJbou$K!14h3I#jpS5(wXZDYcWVRn!zZ+=}-2x4EG7oeX|
z|3?v-P#$7J(@Nf)nshI5>xz3w&-rYomYjHSnJ|R^MBd3{0;I^E#~%qHf;_S{z;dN(
zzTJ+%H`Mr41)|wPjI}VOgC|^^oI4nMgmZ^Git$<9kjeo6k0h+?k}5xDXp{1V;fCFB
zd)}772;1t{ab%k_9#1V8CBHx3!!TiPG0u`EGC<wzN`6eniT!pq@LxSF+{jwT*!Vrj
zk68f4`)*I|13BeaeQVBlt15|#*#|#^!wt)n7E(f=@nPEhrBFr82izmhcq4CzA(gd#
zV>~R&@B>yLJczE}6ex8U2fJ$@XW&%RqMd!DZH*QNF<X{ry-T*{G3F||_lw<zub{7H
zgV6m6`UGx6Kma&h1X=<U(vYDIWAG-PnHDhqD!<pp6rh)YWR`w(3GaZ9O&>a=uQiCC
zzT1!fIQ?77`)SViRm`%<D{%y|MxdW?+88~E4G17?2JMoywFAnkig+WJhC#&k9z)5;
z+6G2m-rc!W2`}O82#30wn)4~(IJ6D;T-iW(N$VRBC`;I*qXa>-Oh_`HX`LEao6~`U
zb@hm7L_z7eaX3Mxuex$y>!-i@NiW|TYVE_d9?#|Rw$le)5l5EuKNVV<o8f;EvA}ZO
zO{NmvLRqpXhyOIi{hRe1k79B>ziS-nM_0e6ZT}J!>oqrFKZ`I7s~qnf`-h0dn6DqY
zL|8MFH>U*hUwzx2+qTmD@R0J!p!5_xQKWq2wP29ssbjs#t_ozU5tu9RMTNeTiSSK@
zrmqqG474yURnthOaQ?@lM6#8#PG9Lj1DrMLM-?qf|9dS392zvUP3Cp6U<SO{u~D30
zqVZy{?E(kO=tm$lIqE`k?Ih44?q(@8VaC^kzSksGVw?!nIUuYYhTE_p)vKy<0~vjR
zsvl&bto$1xEb#Z=zijN_J#ulc%EB}M#!Wxhf*tlW8P9VAR*%2;SQD5r^0mTPtj!g)
z@WDyKI8I7u&DUzN7=}R|ztF*H%up=XP74P&+A)#w5NuISOQH_uvvG;+ds0nm&vJND
zu0#VKqEtsb&uXA%tSwDVH<<^j#&gL7aC4*+m~F9;hS9BG0QBk<qlsxc`72H@@TzP3
z^XJb>Ym-vSuvM^zfcDyd9Fl(tNk(Gp@@nv*iVFQX-JRS%8S=Iq1O73@SCHx(@G$21
zgdOpGAfk?}W8yuO)<#EqE50OVVqK!=2#=ORGWk?8VDFLviZ6y3M~ehJt?0gnIoo=2
z#9B|9;=PTSw`WgSNS-x3aw+&M)PlIL>Bi?3M^+|qSSO`jmt+14FO1o8DUusUmmr>x
z=V68VLCNqmDseteGY`*)6BB#k>6g%@;w3nB<6T%aoWaY=9U?7)bDa#xsIRtmD$cdE
zFuh<(@8xk=J?N<QP=;Ybne5-#>WA2^GoEvPK-LMOFY*T`AlTgX3L|XFM_)^st2<6d
zBDtB7yzZ4@PdK@_Zlz=|=YT^u$A}yj-=vOM(^Sw1+}=MNphdwhFU5AjgjgD_u%p#)
zceZVk$I><vO_yl{A8~N3E39ElJcUt!b3Jl?zO&$Xr-Lz3nQ|~geF|vSQ{(46TwHd{
z$Q|sOc^ZPwNbi+ldto->!S?N|D}g0@OL+97k&isZQNHS-KpAa^o_QpAHMU!b(V>Oc
z*Vo&t4mZLs3nN<F=V?IbKk_%?2X&`C1mz^+HorkdnOJvV<sICu7d|oJCVx(cI8jV@
z`XwO)98j0_8M8P>>{d&^HmNwvbF#J5H5p;>VSnDlIfb<Bp(A)2NW<9+W6~8gBnf5x
z(B6$N8(V`wI4<PtIimiN+4d@JrqF9^LWAJ8FHov+1HEIy?#+!2!%<x)HC`zG+N)t|
zh@QAthjit;o^hQ7YH>V@WocsA9yR-?OM?LwDzrVeqW29Hazu`bM{!TGayPI7KIHJI
zNSIoeO_H33UM?J-7s{JuD_{w-C~vr(K6q2?U*x_b#TdC{=+c9H6F!xf<59|e`YuQd
zKd>8a%?DL&VT7bOMXZ^Me<U=I4v0@lNhz&5sq*phAy<w!0oxN};M=lUhu%8|>|s&`
zAF%|VwJ~8-RAKu=q1**L>`EVb6KtK`MTK?u&L{@#rL=h64!s`o);sNfYJU)uQS;)p
zH8@aVhcdFsflAyPvA^=sJxe*p;|oZ5PiEOpeG%`*hfaYMCU{xflAC*heWae*NYu~H
zqm$=BKINj+N@+N$>yd4nJm(Qcfi7Hplw=cAJ#BlLPm#8IiL&N#teF%Sx2Ndk@r=!3
zbVB!Z#$Gr!o_XCRq@@_6S%vnv9w1L;&$<Bp9<1wG$(=;;Q8t~QGcKpHW_3BlJx;Zz
zRbdq0hxfQ)pN7?cndUsS6;ds0&!y+1#8c0tZF&`5vK^pDfo>MnM=`w(CS|gu`p2Yw
z-#F6_CA^`OxNY4X>{H`>try@#nt^Fi3MU80<TDuky3cJS^30vawE7(8Zi>~fkzTQj
zz#jNa+8n}PyTwjx8!epTpu$V4KP==x4mEh(cqzx<P}mW(V+TJ@I|Xfk0I&*I<@iPt
zbaO$3od+HtHDtureL$?Bpy2BCH?Qt<tpGcoD4HB3+|oRe4xLQ++}77i;<|flV>yHK
z%}n<%1d2fW%GB~#>Fe*OpB>%=?W+#@IwKd~Bi$Gfa(vG}q=o2y6bPSEtdXW-fn65j
zn>heZq45(N=L!0dz_;!LDyo<sV+7lONV|P9?A;znE9UwNXw>u>Ay2uYOQh}Qle5<y
zrp=;LA1V*kr_C{TOw{<2<mOIs)or3RlB**EV@ccqP@o2fNlVJ7)=QJX61M{aHwLgc
zaq6FlA4%!z>DiPU6kvea$YoYeG~`ZLWl^4zg{hrgi>VC{%AAL<DWvyT#d3E?4Me4U
zthHO}MSUSA<q+;H*MZQ>5i{G53l(Dtio55F+U((H@jwwAPyj-L3^>~7(Kte{L;78P
zZ07IkDj~VTJ{#l<rzsSqde4SQuJ3hjX;xvd%ZQwoSjP<y&jva^yneF{Fq&@4%?*Q+
z@yDulGwbf2cFh*TDTE~GixQCUI6awQM+_j_?F~W<NmvWSsjXv5uyOFGD)a<qf#}XK
zEajLerK5yzPs08ssbqUWG_%T@EF2uR{-!Y956tTx5QY095ii`n5HS2kiCsUCaCj&4
z7TH%A$MgQUwu}L+c^cvXoMz-<TtA8td$+49;pUb6nTWyhtH?T5hMo=7CCT^U(NZ+r
zRgkyRaQQ8|B<j%(q9CR7aocW0;K0bI7QC+w3EOF`+;+JZdzY^+;e_3dchP-K?PDe=
z%-uXIK+o6r;tt9?fVij2zMl=K8^goHyhP~c{Nk~>M16d!!#sbwx`@Cr$G`an1qH)e
zK$8%cVq#iWg#_K*-$wI$Dq2}(YwGG&5ktqG0pMJ5mX)=&X+Ndm$os(8K^+s(Jv}|I
zfmZY#iIgArU0G}kdk!e42ymO8jE|jIq9lF(@@1VGKz|Jf2M3iL#f^;(ot&H;C-dl3
z?2IyxbX543)0Zz_uz;+11KviLdN+)Ri7At0TZwa$x!bjfpzu!B!z85h{|@90)PXWT
zVQ{+T<M7as&dB(9VtZx`sso@-tDY6=XwI({6+7g>)lPsSCWZ1z0U$G&mDRsPsBbsY
z3|OQIX=&2pf`Y>u;Ekx3J&nb@_ZA3&5J-Zcdp^X@D9+2o)MjF8I<KAhD!AR^%cn12
z&Y>%pB)?q4c+>!x&VuPZah&aO6mUooMRnK;1RF$vA(6LyoPUx<k|x6E`5V{g5;yf7
z@$y!U{}N}LL?FHps_bA|ywiVTB@f;6zi;-`X4+2SG>9>R_P~l@BbK@{&|MqR)cs#m
zls5w|^~&!UWuXvJ7eAL9i;E(Oc1>O#^Mef;#1(m9!<O*jF+>pEaxBDnp{MaFNxdC-
z!_gYr$3BpjJKYk>9uA@(rg7D9cXqCRjw_zo0Y=ItOl(}h{;a5`kRNn++~8u8+3BxO
zpujPq?c}*59}eZA=aAJy;ms+U<m}XCck!Tzlg!cAsRRovdif5lj9>VLjT2Vhm*mF+
z^!z`qy>~p-{~!PTKDO+XnLR7(7}<LzA<4+*P-av%<rs17(IACLBr=MSb&QTaNyrSD
z$sT1(x}K-L*YCPre_a1v|9sp&KA&>D$7?*s{Uscvl2bp$u=G4IYbOx#K@$$o8X5f#
zm?0$PY#)X78x5i|Vlo&=sv&^Aud}Jhbc;pO(?K@msG;|>6u0d&CYW5^qRCA)&MZ4A
zbYT!Ke!Cz{hVVJ=IZA^RQyPJbRDR$=lKoJAy)gpKa?GCENiz8xegZ<D|201TdId2g
z<%$wXRcFO~WIZ7!P(dG!(;--eJV_MhKS#}(CVbS?TV%`GNB>xCp$tiGlw{FIT!b^~
zV2Rve;azA3KvxXUGar}<q~%8W-^YDTC~jNJJt@k|jtNmZ{D?>NbscxaF(37H%B0W(
zZ1#U5ls@1N#Hh5&6`o0egHJMf)*$^8ZG#xY(l<tC>}%xlvlokjNODr&jDBKe?9*tA
z5}#|=Zdc1oip9g!o3C0f9<j|l-bBX~B|$MsrO8OIfuHa$zeI1UH>v2g<)t51M%Rzw
zU}7ifcm~V^l9NKd;n0lJC++yC!5L7!aS**Dm!~FqaM047h=$0>%^I<Y_4FFJ-3{{_
zd)MIgN%<sy_5?DC@N1cw-KFE&H51HliTGL1KTh7VFN{<>t&uQkgtTvPEmKA{FXSD=
zYU#xDQU&mjAdX35osaUmd7RgpMM8a-MG+el!|r!q7TWM%ez^(LSuP)ta`g6bNM#JY
zACG|fkCiMBI`21baBQ}JCB{v&3WdnYDk~p((kr6dJ*rt4du0j3Wc;sCXNCBa(1T=8
z($E&v&E%ornsJ7BCOK1DWSa>zjSF@D>~6eeS3abS!)2i~)<P;ljVY-6dM&EgO?Gf-
zNZ$O?CE0bBL3!e8A4b#}HVh@DpWYRX$7jeX){PR;B23-@Mo8#eIG{Vs?7uG5L%565
z=&SmMW%KzwTC}=A!bmI1?R4(VxSWe$^Fz|@)R>gT&Qwt<g_HhNiNF0q-db!qU&hD*
zb|i-VDEjj~+REJ7E%`t$mP^M;@e@<I=cfPCWL~rO+&$^YSm762v=Fanq<c<%J$Ff<
zb=1FpT^n=!{CN%7M*R{gz~r`HDxitUw2eagpWBPt%iCr##GTiPpZt>{hC6%Z@K2j*
z+=aZ53*S98>$;<wtvM=yp(0)o5uxd#QO1WVDq`?{_CA*VC@=fn{Zcjgs#-j9AM1Kx
z`j=yKbuF(*xTp^iYjW~M2@AQl%gwkpv6KdYZR9UtxhZKgyx>U)Fk$w%BpE};8AGHj
z%#|<khBQ>L-E~9@x3-f3CPpCgNTf|)&+}q8{^LejN92pwnet4{(gXo!LCbdS9HBP)
z5*yyvVKx9he^4B+GzI$H@uS%Dy^;#2!?*>d`Ko&G`W#dm&d&VFN{gbUcJ(Z<=`rmT
zjhu?EI<a}b1WFY0c_VXnQ>0v<c#i)C)ogQzc(D9RMm5oONocE(JIc5RmFa0#7E!;@
zY(Kjs1r;mutT&?$(|{Pb4tCiTWCmM_fwuJYqo8wYUyxI9We35X@osvm&!<?|bw?K$
z7hZ6iU`l~S+VdALv<wWWVW08iOXwQL+=%X6nD}$^Mq&Yp*n5Eg87j5(V1huN;`#IE
z8rs_0XLs++u_9!e_~FVK-(Fo-C$QB)cp?yU8w-6!v7h0Ma-3?t^S-;ggXr<-Ca^eN
zbfzE(w#y4T_gGW^b_)GR8=_@IS()HJIDD;+bx3z*nXZm}**_H7Y<-}$^kB)SHHtwE
z2ks0vctdiZl-E=mz=se$c@l*ra>~>HIha!q^=quVIp;jx9(Td~IA~(+)kF3)nz*K@
ziKVR=ohQ9+Ivy{=V>ip}>cIv~zISg>OM;IN1&KvsFof8eIsk!$O*KT8Nbmp!P0YHA
z%KQao<quq5?tP=`mK2_ij_AJtrN#uSUB3MW`weR4eNnM7D91i==YUc{^xnqY3@9u*
zP$L983KNh7dKJT6VWZaFA2Vp5@rHnR4i`3|4#o%KmEa-b2tJZ`I}cL(^%~K<oaq3X
z0B{Vz;Ig#;5jvbINUb5M^KG!jBT)Q&qBm(pn!4~mL=%ebzj|c7q`u7L2+?&31GkZA
zNfCbusz@v-AgRYCETE*$i$<fr?*-Ne>X(c_m{}@x+skWWHRbNtu`yai3e8vM|F&<4
z_RCUC?7ZtC?(fUlM(!vZ&sH(T8u)UMm7EAS87y~5;nR-cmF#Z5tkowACh01ZJg+Up
z-HeUtj~+dEWDPM|3Pg%<87Ui9{ryWKT<6~nC6PKs)?l-`L1pZ7bBreS!2=5T%`~F|
zlqVP_Xg^QDBj>&m=o_zSm8>M*q$Xr}W&r<wcnx{8q5t<XpO8RCXH@ac*TlGGJq|@r
zfy5WtJD-v$5I@a*5)s(QdiaoiU&X1FOo~u3Xcs%MY7MWiv$>i0xe!w~*3hgAi=Gh~
z^Kyql{|22vXIyXa8)?Uxl%#=QkJHp-3x9~+hRD|YzUXrU34SeDGGRw=$Hpo4tINqX
z*QAyujxbdpjsDHf@3FH66Z7<cBDw9%G(2r4YCMY`9=-D(8ZnOe%__dkF`BYKi`w6L
zH(H9q{~@$UD$Mvu8%&EsR!2X7##T}nqZQ}o%8rB$S;g5kv9KlSozGR{BIYUS&#)20
ztG2!^<at$TEchp`QWlpiEiqs=z^m8_y}eK)M9-S6{O#f?TmJ#`yk<*4#1IKC_xY7S
z5_D)~1lz?Vxj)0WDaX$CYf%5&w-n01OfFj_fKWoa+udy>rzJ+_81=m^bgO=nj}A5u
z(3WF>)%vL9*-Y{aeX=kSoN)(TkGMc|ZjGkyB~R^$;<>$xf)C-jVFSjw3O9p&=!%>z
zw1&_nRhyfe^yH3$kSCFX4op0VsCl2~;))U5-b+z>My?3=ObW&SKEh&NG*%LB>tm_Y
zG7j^_OX<Q!Q6h|*Gmu(dyKsQYwn5^kplEh-IY7D4BsUqNHZ}WS=QgH@g-Uo!`nlRQ
zowng!_dA7vF1ilFBW7!HfK4q-0V-_ka6=r4O)byzYJ2O9?Ch_?xl`Z{PA`==>$$t1
zgh4D%^vLb~1radFfrEzjrkyxr(So-sPMN1Ubo<}K{>0wiO?ah1ZdK{TgU16;nP=I~
zzn2qy@^L6K{BR?ecoQIy!G82S3=?e10n4kAWMWtyGJcO4zu>)}v!bU?wWjiE`+icK
z!!P(v!}vjooVc~w>*nqrA=U{2^0-rnL=zCNPHqDQoyf?@=X(;8k{tl?;p@EdXo5fR
zfW&Q9Ov%<Qz)vC#ZJYWq#f2|m=|o^q)`f5GBnN})_tiQ6zVgeB?ISH?MI;w~n`^tK
zn6x9da%Z)YTcQaBJz`^lSldUte-=Cs&LXT8+Zuv50uu7E<Wg>$hK7co{!LHH@x)A>
z%+*irMEcCC#kJZgiiZ8g5?Efo16I|zuq5XhFAdlb&mTZUA)tQm_Q8)1DCp1w^oB^T
z0;fxUJV*ffqttp<$(odyyViQ(ic6e&QgHrpYF3md<>-2Q#SaBxSKsqKV?%WJ84*b=
z>mCLw>k216jT@DUW4Y;q{2vwl2&o^UdxS357vXe0*Ah8)eR#^d_00pl<#ky;UB(~!
z2{eE!vkzGiQrXT7ue0tUVLNsqZEEC`(vnyrkGh)d*3ykSEqBPyqMh5Rh(orid{7vq
z{GiSZ|GYlxU8CEz0n^?Q0vqvfN#u|_f<P3P1a?-Fu{^ik?yBABa;TESQi`}vFKQ0j
zdy+|X#hR~jL{4OzD29vu9lE~MFC;NCn~*&5adsppJLt=PYIIUT(VMD_)2nS+jK<iQ
ziv$6~iuQ6(B)qiP_^Q&`@86xFdY;!@0#j&iqkZ$!S;$3I-w@@JK<2t0#li?evTj2b
zH8sLtTvkTG<TUpw`(;zEdbk=Cei9_8@s2vFzmI+nsb&hBup*aa>cfic3jDox^sLyg
zPRgWeHshDHHDkLfg@r1%yIiCr$Q$VOARs$x#M#6xDT+Lu`Vo|xnTfkY!%7tU^VXn#
z$|3g~o-`B!xj#KqQ_~_>NQg7v8K)s1I?tgNXr78%W<^f#{Xah@5_^aH^YBiRd1Jae
zT5A*KJw(j9(RVXR>?<9`p0+G&Z;xX}A;hM^{4cjb6(m#Mha2oH&~bg=^Tqnr+DoZ4
z_mVmKio8Y!J?SQwV$Rjd@@zz`PNtClZ%EjVUL&heSN9PWk#uiNU3fzxwR_ooz*yJr
z+ZtEgyYi;}Q?mYEufJ$1r(}rR#C-q$hVKgmDx%;KX9ZIomK=EKfyDx8=>!m*l0yz1
zZ$G8R_iUb7;zA8;!7p{ct$kPtaoFhi3rmD4Bs7{Lq{?`!GWU^(^X+R|Oa9?g5Xj9^
zV#fAHTgW#3-nM~kbIhR{BoO{GMRuVxlh3_T)BKyR`QRswvBDG?f+yD7`%U$At0@-0
zkbSn1Unq$=*YBwogu6DILh`@gj1W1bQ;}|B>sk;IGD`wK<OVaYW2RRe8GQ;dq~^o@
z(@+qXAWqf0P``#yKvvIM=L2!ionhkQa^dqc;<t&<G=$I{2TL#SLfY3R;S0SpDz@EQ
z(h4CRRF{^1!!ouH<xqG?S+VHWl1OeyO$|$~zO}E@(Dy`?p{A*I`aoqW?02;T>t2`|
zabi(8+?_!oAhsb`Ff}UsRJa4ju3Oby<DysHya1JO+kh{wFAHx8Y_NYNVGBjERUD!v
zd%X|Pa8qIAp}|}IY(OjyO>dptilLJl;ef||@xw9Sg1&iP!v|t5#Ci)k4a;P|naMl&
zz$wHuk7UC-3DVrHUNic2F%t}62rE_XOTZ$NnB#)*M`iHFcMFU5_hAGK9V}q2rca5G
z4wp|3s{h|8Msyd){Oxz6A1w&J;!jvE+yz>NnMk_K*iROAd4=6y(2|sb$QPmknH6D^
zoWy(gjy(>Vjw80VAi25@QMoFmYmgJG{hsAH^4O>6iRLgYeh>tLJEj{<ClPJqml!Y9
zLBBc6*k0$$@Dub!mc)%asHTQHucU=Q+EXz|kAZ)6f~)r5t+W(kFv8wGM~Hi<26ksG
z2||*9VY*+l*hs2<qU=X0RY1qDt!GgT#vc;n<WAQXRtV5I%9^qoyUt$IALuQ;W0)bK
z3);+k^7D3fTv=IJTFq<Ce-4QurSb4Tw@&!?o93-S-s%TD)sFD7^Drn!7#*y@bAx~n
zjOqnFJ83^6U}5I_!YXm9>-j}htLYcKq!@=iT+?g{(g8|6b>tJ|Iq4)t59Dl}J4sT>
zc$u&^4`y%G-&p;(2V9*+E+YL>1P<JLO2Dc;(9<bStW>0r6?oww1jzxCb+@Vs+`JxC
zzE%ys0CeMsqb!yJkwA=E!kA1sFkY2nokM|s^q}{=fDQt48Vv{#EU*XO%0PJ8HI$gL
zi$y=AXgUc}weCbzjcagck#vYqEUL<^?`dCtP`;$McZO7o<T=CRW_FDbUiJTwOP3$$
zQ}#j9>>H@mmb8y7Pay(K1Dg7o2GBX6kV2<+?`wqqpm^J|sqyOk?$bMk6d%4*kJaBm
z!bx_MaB5)fUlmIdg8oC0CgUecUp6e0(+VSWtkMMsn0g(MGe;sT9^y_QV?^Fi&8Vlc
z=!P%SOfjN)bCSK3^a=UMc@m5moyHNSug^q3^tE6Bb|eHEx2kWBFk&9@M#srpYV#a*
zRg_zSe(-<``Jp_DtpK2v7+Ua*=1@TZl*&L(bh6!KPD{YqCZZiC_YA1s$nb}pHNB_*
zw#S|>qG1TwiKJ(*CY}Tl_|#hEju>jVAo9@syJzrpY~f;Ndxg0_6@Ghgi~Mp^Sn&J?
z(UR9@P?g;bG&N-+a_A5|G*weGGKg6>VlMGBL<~uhoZI{ECvLpwtDTIbL6Af=D0DLw
zZS-wWrX?m)wHe9AU>1bfE-`JD1iECvyQ@$+YdbkvvlP&x&%xv?;B$Ba0s6-^xRp|v
zWn+Z3`tCAhU;dPr%#A|~g$57qQW?d%C=-Z7pVn}0O@N0QQMP|DJ0xf@LOqE_W?I96
zUexV%h80{Sp0~d(Sfg*o%}evYRpzbI&~Z%k7RvWD67GW)nYQUY6Nm#|M!12B;-+7E
z+f9mX6|^3E9*Z?jXtU-Oy33Q)3iEpK7xU%UK~9n<pePI)P&$4B*t-H9y*iszOyabE
zUe(ste!TK}1#~@s#92{$axhr^3S+N<4p`}?1>rv~NMO@Y*C`0YcOg-5I;zFjCMY|w
z@25Ym&|MN4F1lL);Tvt_#L*ldVA3G#Hz~av&ZZX#pj4<FL+*v*uoaEAruSUAY}G{<
zXCis2<lV9Yl$pLRcMH={PfJOFd9)=_HDnk!FNfS=QomK|K4R2RGrmFQ*mERkCU)x?
zxmo(_%(yMD?{!asm(=6iOAKu01b=_nGvU+lgW}4^%m_1MMREXzWMYzGTG9ZeN0b9n
z93DalG`G7`&(=dIfmq;HUacjCtDrFV_$Jh>e#}2jnNHIWU<mvE@nMwms@sZ>;^z#<
zs*{drJv02MN|M!=OQaMk26*A|14SqPE0ov-Ap;KDNgpQlKhUawhbqykA$YwKQ<pIK
ztcgfaE%InG8owD`>B?4J1*=ijZ@%X#n0Z%_sY(u`NZv3<kmmiBvPa+Et;D*1=QBue
zaC$({GdEAUu9PjLWTFkFJ_YA?+)ShW+8ZfD^VGFtx{9Z*e}7pCP{snAZiIZYNp$Y^
zd5AOAhdKc2CJ?!uD(J3Vi*-uuf>d5tz+%BM{7Smu3XiVBZR*J>`qNRO578EQHsx$w
z@?tzfTNV2b8f0;}aolcQCOUu;+bnBz8k1N@N1>fHv{a2I&_f-wG?5>wPVlKR1Q&>L
zs(Itzz(F$Xt72^{#oR}7dSq8<qFQ)E8|}+`-9GlpiYq(Tsc_+Y?HZE-XM#GgOD~u}
zC8q3tnc3JYTpO!1Td3kYD$Y#i#Km;KdZYdhI>5husF~EQJjqp>Kseev!M~+7`DHIw
z(BKHR<LQZ|G}paqtdN;csGU=#l?9p*6FVCmkuO@a`Q*>kW+k%u;)#vNr8P<b4PV#W
z@5!OY(~Ds>OD^l1;S<qB3dw)}9kyHq7$hKaxQ-@6$?C!}a=7P^X(pS@eL=^HbqM=G
zPVCuj?7?!5P3aI;TSx8`jlOcFfx3GlRH#ne{D^+(%tT3@Vlb9NegPLjet~@I8HWg2
zh4t}GpYct_6%|eg3Z-AZO>w$&hB}_?7*?{f*LtzLG^Q{;!l!;9P56|u>s}!mpG&Ue
z=+Dr}b#U_*YrDfP{*`TdfdlFU=k?WY<q~zcy1?R5L4N5TxFolDCnf&3_X?~QGyUvR
z{`|WFO{2hjYw8XD+Cd}>>TFinqu2-^MEBC3eygXz(oX6lf9?w|NbkOUXQ0VPHF?`0
zJyWc6C(!hOi}FvE|Ahl@mqeqR3McCM>wn<?sFPX=0_6g+r?(QUkCg187a(^`qyI5<
zUZ?0}{O7!PBTxZ2p^IFD2dSBhD<d~o-TK(OHPkm<_xc1ugREnWlT(@SRkZWxNSSJU
zsZ?oqkeTVdN5S0<S7|hm-3w$kTX7`^lLBTGm0rF5q^RLK@(Ore!A&7Xs@5*>o=fll
z{%gI96GVwjDJQLdOcW8|3DxPf3b83qrnZfVa|q$d&t?Eiw=Gd%mr$2B3cKCcbdS!~
zKVXY9lq=1y@tX;{s@!@*7PA}$TiM)->oa;Lv>(2XeU(0H%9X%Y!DQOVokm~_<&7U$
zy7Z_kz}vF@h9G1sEzHfM`sSdd9CGJE${K{GtRM`032A%V-{EOZvqZoHd14Nna{%R_
ztxbw2Tq~_$0-QBbojWNeCTn3~p=O}dl0h4nAEbHx`VR@NV5@7dza;m_dXr-#CEoB@
z-Ye(yJF%bgsolW@1T(Nht-FQuXIBL4xban@JA2j$>r!JfgbmVg1hS4x5Hqn{m#Eue
zoDN1G_lwGqZLq5;(Vx5S>A{)CT1TUx;LmxQxaPl6=WH(`CnuN2<-m!W|J<Krr@b}Z
z_}0z^v7w$cA754s6}6_E_Vv)#Zj-r{-Qt@zY7VHz2?2h^=_S6b!mdSwoG`VC$5QCc
z%aB1C3>u_nQa!7)G~=(TYpO=USa;<!+q<6w5+F2oF!s(@{CfNHv?n!b-+acq;`am4
zH<R4n)`h6T5u|gJsZkF3uDiF;3L9+dkS!{vubhL|zC#g47W+DoN&_h&o4jvsBscu|
z^YAlL8xr&#bN=_Y4J~5xCY{9*h(FjEp*rH<7cW97Xw2<rTnR1DV8%2o@eg{el0a6H
z4df&I?5sAfN-f;e$cQL^I=TFeu~#TvuxqWHzglLn%x>v88Q(~Md(zjH5*iwsq{K>J
zR<VR>J3bgP3q(IGD>!W57oEudOwJd6QD=1VCx_YP2hHRaT8v~4l7lNSicu2NIfSPd
zT`%W+9?ycc(FAFQ;IyZy`G82oC)QV#w&dh?R~w2%KUg+cj@Fa$%1&O1sC=8@as0Vb
zK+x~+Wo<G8(kz!)W+Fat)z;ileiL%wtvW`1Q$-BhPqSJI^+D+8ie&p8g+d&+;Q12_
zzn+I3DDt!N;O{`FF6iw@M99mH+pZAq8LVB>z&9diehPPgG2DHU-SV-~qGao%+Oa%U
z=^@S}!d%awc?Wd33j^oE05n#F*?IfY2xa<e2qVj3slTV>bFlkEIC}?ugYLo=dGfv~
zhzbm71dHmLHgMdwJUewl!r)@8#WK8gSEDcg$)cqob=awqhpL|Wsot0IHSC5QAC%A6
zz9?oXAG|+VROsR5b=~Q14Rw;@<+JEHC>hN_z@PqEk{>BX`k`}{JM<_taku>a1%Vul
zCqnCHJ2Nm)sUUWipmOU=@V{C4$u7k~nYwcBb6C*JH_;%YMmPZ8K9pQ%lN&qwbjw|5
z^92}@{j7V=iSDY3w}CvfrL$>!FZb=zn}x*+rBD)57YfIYWG-{l4<Dd$vh~)$AaVB>
zFPONA>cL!QwzkVqkU9Oq<WKb%ua$VW@}W=>EMLUc`3R`JJskQzeiCp-iloCM3`>T^
z{X>UJGw;eSUfNY5$}WEnchrGfu3>twN8&4>o5jPv^`6ssH$pPZN>Y){N?UmTA<6M*
z>*^WbK*>Rc-Y0vq486hY%7(MP+Z-fisyQh?&7?8QcOj`@$m3$TvOo<umzj0t4Z1lu
z7+o|hKVi;1+M)Dynd~0<{{8z9#ko-+wYs}0uMGF3X@9qn)YGicA7|kYp`E{jUk+u5
zTn!po;&1u{t@Ks5g^i63eHlU--NEH{wou&A^td=8$1{JFCv870>GNu;?EIf=#Blz>
z&(6`LFEpEn92x|+i)I)el*oL%uq`4iEUcHDe8J3&wEk-{+?ScW*n{p4$I4rl^6GNt
zGpvE)lQjn~XN<C6%R}2$hP8L*G6S>jk3Qmd<(-VWYBlJ~rlx5pDtz^}>E!*2Z?Z<$
zIqij<9H&|IK4Vh|!_`~KkvoO%cb|?u@_FkU@0R?i8^F`-jeYM)*?5@jm|Ue~<ws%|
zO^@+JsE<(L<_S5=L37LOB#g+RBmOJkD<?-|K0K#B%*mlatQvyQMEYH+eIp0ZK;IGR
zZg*U1WcmNBD_7in^XAQ&!iWBKwr5RDQi6rZ1d^g@@YiPoC{;4i7z$}=X(_iJ-ecf$
zKmy;YDNOOm`TrEe6i+WIc7}-u)e#JYg%8S5JL^|IeX+R=pxPlvsmFBT1<~(W56Nmq
zf3rAq8dYkCyrqy5TkF-jn(bL>^>Z^`5@B6kwWt#Lr4jv=5m)h<)t2d*F;8EpuZPFt
zFC@RNq;_of#>+Ip;#BopD^Eo~lq3A3wvXueU`M309B}5))sa&EDng~<hX1WcxE=oV
zn#fAL^zks?i^89bc_?y+5}>ih6?@_am9SA(G(@=>VXI4o1d4k1kkmpX8^Oaju^`SC
zCq25qdJ!!9D{ldc-e|7^1=Apikvi3kYO7JiiMlij@)PkB*J?OvH9(Mzi(4R8<aTJ;
zliZlhFEQqnJ(wa7Rw>9P#dkbXETSQ~lJaQ!)TvW1;AZ<E=bj#fy%IE_nI3()^%%P`
zwkFVY2KKU@_u~rPSzCC**%37cosc-zd_e;l)Yu5?3Xl_SB9EDVY~6&R<>B~Eud&+R
zHnvAm^Pa*xq!s7V3yNYHcqD$<3n)S1AP~WAeq|qFya>6NV`O}6w>ku<TDKYorxWs0
z%GWlIb~!(p$SDqBJ5+SJ^r+%0<W8sG-_sQRaptM@R=)j)Q&K(s=+fdkTnb_ieRD?v
zMKcNNLZZ#JI&va^N}_S|qHV~+Gvdn(TD{ZjFK`OL$d!?50irTqiRo~LIu6)sP|Lt8
zuQfzHGt3~?6f~eXLxOsR+U3%reCGyb4K@s;Zsq;0imIw(#9k!~Gic92F(8d{|JfgE
zUWa`^j94ae!PFF0K9mrZE`2@t0l9(JTudUv;<?JMX2X4HWHg9*j+TswMu$DnF=&Rm
zaWeSlymS7N5Bu9tSX{4D@wdh^^&W@&yaW2+>*A;xD_#oU^~}t=+kE*?g$zP!vJ>F=
zg>xFbu7r9zkXpz4dwQf&aI4x&)gfyI&z{-!4QxO_vl~teK>PAo-3y$%57n8AkZW^E
zR0Y;b{NCN#N8&bbYQ<POt5HQ`QwIQQyk%=@dL50ej0o5Qmuupqr?(^2!FHpte@0i`
z_t!eMTus~BiQTMYZ^pgvon+s+<Q3aj7YduGVH`MxQ?O-fA>vhuG)bSvEtUrWR)Vf$
zeqJ6bJ|91~QGI>Bbm&|ypcTK(&bG|nnEH0tgkWxIcU;L~-ND;i99FzqX8lP?Nm38b
zGwL#dj1P3nHek?EzW$OJMXyBl<;$1n0AJ}ZHv0iiL)8bvo~&TK{R?)}K6W=%)vn)f
z5u{oRl3kGlBoisMuIV7e{5RM!Jp9f95bCad=X`Rp%Z)mM_2CCAuf*rC#3Uh@BneyO
zCT_9#X<cOG_-m-ZSZ6ZNbV19UtBrqz<WK&6u14D$ZCEt>@t?vZ%E9wq{35H1jq{w2
zeywep3=-{gP>e|g@UlXVj(nhc=c|Su=gNC%LXPGrTQ@d*W#qn&tbTeypsdlaa`5V-
z)RXeA{uC$~*k@wOX4&>HD|An&Q2xGK79CC2bjck{Y{C!YhYb>zIt!Cc&bDPnCl^^?
zc9gQPZyHVd2sQ@~+-&aC7NyAK1lbLEwP%1Vx_aXZbi3f#5gZltFxvq238jj>zi<iS
z@duCLexOf>0rgVt_omr0$XU<J$_TC*b}Y+b2ohJi@o@j#K-AS2TrSR9rU`{z-;Ey0
zXgDbF7^eF%WAz1oDG`R1Ay6j)`J=tI#CO7}zFo|s9j}>&AP5%j)-l4y{UZ|1;yEO#
zjPhE1a^S90B^h9%-jRM?r%F)tOw?;3)95)pq})Ag+ty@SJGQ!*7@>cG<;-5-YA}TK
z)l?W<jvD}&d<h7=;|FnMu+#*QJ2wq)nD*cur(B<}kU&-~FOE13lUgq=DJERi)SJi8
zVU-OIc%fIi{O1p>Fk6f%!RgkHp>26#UNY&YMC6`;`(@Fqg*1=eKg$yVV$j)h=R5#y
z_b?Uu1aQ6xR2845sBsXnkKCclXd<j<7_{p&=zBnu3*G=(dHLAGH5}MBK`P#=uYd<i
zG}Pg-%#?PG56`^!sHEg)LxGdieXeVttuN)9^$aqJ#L75QS$XT}QUAV=ntToIz7q{_
z^2ty8w+$Sfcc~-SvkKk+XyWy)v1s6s1EiuTZNdA@BZwhS=I`&fx3_1;+;&O41)DXA
z6;UH1mZqWMfR<oY+IBZMZhit0MiY_~7>{Y*234nP%6B<__6})Zy$GZSnb|NqNamRR
zai(4&AbaQ}ypsk=CzoP>!|LM2Y1kXmBfV#)N+G%HKkiBAg`z@c)$(axn6J&5SavQy
zSe{x@vOPy-)eZ(C05$;u!+`9<`$=@}+Ep;b*{5mmJ102S02OuwnLlXz%T?l~g3p|~
zOL}n4F7lB=qMpD-Gc!Ia5}0R^WqVdOH6uXFB!VG53BaV*hgWmht)D&*?EBBK+W1&#
z7BWeKg#WnG#Xc_@FKv=8$QiIEMe8nDKJ+UMINXKuV@6VhDl6kVBP-iLK*=HVwQM($
zN8E_yocg2nfvU|7fY!HcSi1d}-n9mxu3jC#Fj|wZFpl0rq;7nBPnObgQwUq%8{)C8
zVbo^<QP1Wr!X+<9>K02_Dno3H0efr2Kq(QuT@3;0`?0d5E|+g#OfRTI2s$c~HP~2@
zSrRGyMO2XEt((8Sd4@8ksy63YDg=dNQY3sooGhBmaJa1*5)vZ%=Gr$xf}2H?8=M5w
z6Ve3mT{<90u^(>NG~aow!2{4d1PdVxrq@@xJTOj)LK9&#fi|x5jz+z6WjA2!LPr`M
z8`h)t%_yr4g*?Aj^J@$e0X&t;O>U{$aq~jRT*v+_^J#ghDG2-*(NeU9AyZt#{|C){
z_REM@+{o%I4?JBao3nHPc`%`ltWb+7qE+l5BJ*jm9Y>BFdGO-J+ha~kN7*ruaXk$g
zSI1KEh?dc1Uj-J@T>u4t1fKkzK~ILj(Nk={E~FRycD*dYG2bn~wBnqY6z3+OGEK>f
zbrP_eq!++YHaavfS(TimW@a8Ot2d$>W&H8sbopdX=xPhtSfODnAbx;5PQ-Z(3=9zG
zRoJXo4^~w6s@K28bkKug$&>1;D0`giKs790zBZ>qwe@AJCOmyacrI9i9PIh1hbszU
z<2}o*HvaXg@7pCG8Dw`qlDTRUfgW6XdI-7;G*7}H##Mr33Pg|8LB0==KMs{^t<Igq
znlwz4<oNq9B?>^e%sjrmLyLTg_<3u=Pw7`bjRveZj$^Hi|Es#J4EutHcfb~g<7QZq
zsbtgGqrJ*VvgTS{>&4@`!6)VAcfJzIScq-Zo3eq6h?tg*p=>ND{8jY*hJJ6-Dz{87
zed!9MWAFM%7)(Q_lQHpQF2t=v^NAA3LC0Wq!SQN6ScMyeVSWJt0hc<OvISCXObik6
z3!sj?XV1uCJqDZem9mLn0K6jBXNi4hUERZoF>u!tLf2M)ufGlNKEdLDDy*oSJ>kWQ
zi?f%FO!3t%LPD$z@14t!PM#-No#*FaOyjH7et=t?Hsvnry9{WVb$N61aIH}}OyIfn
zD)9%NscaYzpXC^H^DR^8fTb|9$-w*{rfV8NaY+9%_!yZ}ul=w}<I30!NPl+pywV{b
zm|%&5(lx!d+I1fCPI9el9`mGxVM<LVZbBS13p0YkY{1^yfBxfW_6w2{R*U10R=K4)
z{&TWWr<jcH_Y(6r*CHi$z6J)ke72?0D1xY2F+bZ;a?}FynXZ1~CK<1dtka_+CgUPz
z;t5gUSrXkgi@EbGI+^*8<sZi3{3xBs5*8a2FEhff>Pe)phO-hJmvxbxlp;azEa6Sx
zpxg^q0OGe8UYZl~=D<>n-%~{Mk3`)&E^z%G=XRa6J^3k6v3;touQK+EFx}!NU2hOH
z2tb6;?2R5-EW~#T26it{zs!(uhP6SG<h(zim-9s93D8im0FaK~6Obif+`I*25T%qz
ze%Dus6<aEuNM;C8z6q*<)rAYbk6*kv-5c`O{(||I0wVcy^-df46Sdnz8guvet;Zd#
zT(ACl&pqxTK1R;Iry!!r0P!_8cAB~v18U}YgGeMVwz7W3UOuNR0Dsy14jddsVGxO<
zk=dvW*W!p_XATKFEc;wg98~d&O76wbj6fCzHszEeItq6xAD2~$R=Tyj&l9TBISUFO
zmBO>6&D;#c5zX|jz1%GED_&>(5qV-|vwm2huD$(u;}nw??)m8LwSC2$<lA2E?saN>
zd`?G5KHZ6`iN@Ij(7d@rfa4rUQy~fOUSZJz`t7$~t&iX7n_%1zK2&qTA0&xO9F*%_
z>wC)19`z*?XRA#T!*-s{Sqkz1+JBt+K6w!Y$8wb^YObW<>7r=XcYAI@{fPc&Qrvac
za5a7vd+*r3(b7GfF?mIdypwU_7aUH`EU~ifFJJ%eT_{fok!G?S*q9Hv{*V9yRyU(*
zRfX>tWE|J)+He2yj|p-+Za1W&#j%zhuNcv62UFc_W0iM^@_r>aC_qnD!@jT02NR*D
zBZa3t^kfu_>(!%@=9;@w@j<M?5E&$?NN~eP@kg^-$_`9M?SEE#AsF+Uxliik$)^LC
z{KA$Z)-P3)hM%d&MO;Dcrpr4Rm!`SaQQ$w|%KL1rttD3e3kHP*Rj*y;eG>%uU-@D1
z2t?uXR~~dHz1s`zBVpss!v1E(wqtvOo;52CR9lJ3z<Ot)Q3yfd;Ya|OxE=203fI||
zQP%tNIKUPE72x?~b;Ddo(Sdr+P65I|o@Za$xw|Ycz1w=W`!??e{;MF#dRS5<s9U&5
z-vkb2Px4siq`9XJ_|-R1Ao!~I5H{Gd2hYyy?Ci{DrsE`ZYz(cvat7>uNtyY2)OAJ|
zHXPNBDWaz0ZjThwCRS3CR7TdHE>jKp3_ApUM299kFdL3>ssH3@PRU>9Cz2~L(%?ou
zMLGV`w-#Cc{=VCAGmsh644t|5nj}(vvZGi_kL=yNN{+b5b)MWp@yV_SeVzO>{)neQ
z(<kA4Nh0*lPBBB!;3`w^#bx_rf9mr^Z*dfjEcm^JYS3|w>poN;?LRd!;9CQ3|9S!8
zY{PIbGeDW}AV60gU0TPzFOe@6jh8_wjC#myU%Y+$aNgJ_ztEht)+)UKQ-5&rc=TD%
zy>i%7-fX@wMM0XtZ?$ub3(x)p<bpBo{297L?~r0b+!|*IMLS(c{WTFdU=T<)Dql+7
z1Ahv4#*uO`bz#+4Jr$k;mJayxe_#xBsd^pf1*k=K818JLd=fPwr{D8%&KO7zr`|4P
zB;hW$*;^BGj#YrYMSJMDS&c5M*EIc^&D==??9N6u@rlIB%(t*PYf)8=g_LlOl~@9D
zpv<cgFy9dHCg<(S{Nd4Fc7*>{s4F{dj0rSBDvXUyncXIMfU{)G8ohc4xe}vp2(Zhc
zJEuBvKkYt_qyX+%)L&KF`1d(ZACi30E9<Lo)*M3}b7CK|*-QxxmvOti<NNPn=t@^`
z$u$u;GDN_DF~`?)q`IN*#&n#Z!3?S)Ww8EKn>$&>agF0xJ%W`m(<~_)(=(qvqhs$j
z>OJ#5z-dos*l5EMHtyX;cCdaI-PyqcbRi5|BCX2lYXIhaHA{4?<MMZvnRKl)4}nI}
z1Gf<q<a0<f$>ZeUo#7fx>+BLi8*Tkd^+RI~>8|-)=QxR@U_fkPNrtPU_S~&HksBi3
zj*d5~@<r7H@^X^N*O{>WxZHT;F&w%?2uKd)I_$~PRHIn05bHdhKT!G3nH}J8ln2+`
zw9I}YWpFT1xYBD+FsD>UVlo7fJ~+1NjmVBH)FCT)sV1Ze!aD3KAFnH%kleIzaDM%z
z&?9SGoIlYxu~HudXE6U}Nh(g$%1Lj()?>T=)sa(JXvHpyie%l|n;Ot`8u`3xqou``
zksXHLHc@9DiGzwe%8s!Nd`H4eFh)!uhoo`>$22;-<e3#<tg)QPAP;8d?|e};_nOhr
zD&>!W<2kFt;}ob+E<vpUoCyORc<7T5LiJrQ8O-t&@C^b%KM7f`fr{Q7P2>eK-rCn&
z6ac0pOg`X*=k8%*rA_adXnT{(P;rXLrh21A7#`5^yRN(@A`gJ{-0q!oAh=s(^H-#z
zxA~cLB^=fnduMbubNUu+nI)FtZ-DBNoT;%T^<?1gZ@o8U(W|@oTMXSkir5n3u|cm7
zl0Y2zcB9Ui)R!3>HDd6cV%>o-{MwGrhgh47n;*xI4DoRZ{-(e;!?{A-C7fzt8}iwg
z@0>zv@~eH?eRE)0RJ>jh1MwVX3@)xRSJttAD~iUdD)oBki(Mk&4Khe)nP^kwJern*
zWo9>-`GVj8tUlz}(Oxy28u>D+A_sSyiEOHVd3o8SUf`fL`coWK87ZD{Yj7PHcQYde
zM@>&*#7rN|a<5Hxfw>QlWJ)U!PtavcvL`TlUu29P#l0tQHm@SDFLHZLvS&n*`iZR`
zw%i9S38BuFR;yPMxsGVau+h!|XJ+8tbMEoT@FV@CPtw5cuUw5W>m*BUWiu4Yr^~)G
z{@VH_&1B8bGt8!{n9?+yiq1XwPgIX3)6m;@!7e-NBMgwKBv8v`kjHe66L`l$CXJI=
zb+E{fN9?vDT7$2khIv9qN1s&TSVZ<<wxdY?oSadpFyoPOGr&D{<%?1jZ9O7&E~m#5
z*bIkc#Us!UD8jAkGB$A^e8Ee;2+*=?NBZ5luq*^C(n*JNnmWZT`$G8>K+Za4@f!|M
zQhkW~PBEIsOVt%IC%Ajwg>2QcvTdq1ZKn!mhC5=+81?7)Cd(@Z%N5^BRo_Z+P<30J
z-YqWXFo>+^dFd{{F8pDk2nPo~X){KjH|o2Xq=(_cl%8SXFOY=8-Cj!((l@O$6iB0p
zF&T67fgp2p1|(IIs`Uvz>yrFI@6Gs2=D(j66u5BideK3kXg5wn0Mi+!^UERgVG6U^
z&F%MEL1){T2k1-vhhWCP=G|XK8NBiD;Mu;@Pj?%i&~|aGKS-;b8Il`f*kE!Ba~|P-
z=EX@e_4Q13?R(yMOIgnm{&CrV*;`-KFj|~r{R#<&(#OM{&AvE93Vl*_HE-mt_Or#7
zTgR0rH|0W6Ki1SPPJHi5|96rtEH>Al<2m`I+E@KeUWJZ{p8tT2*m7VJ4gFeM<4n0I
z>+m$g?MOszP0-(j5^+9d;&+Tg`e7wzY&B%ty=`({KfdQr{NBHVYGYEWUS0W2Pq8lI
zll(8R4U=Bk5i$#u)mM$h%j##$IlTxj-#Cu$8y!A4b7}IS2Rn6SZBtyoHcGVN<o^!e
zC8?P65^+3<x_U5n>F#4ZS)%dTg+AI~gEkpC=?AqR(j+(>0^FO-CU-mue*<B$%<{vC
z=!_IIh9?L|uPUZUli_;VaWQIm2o6L_F+bk6dfoG6q?3K)_gj(j|Dsk3w;~jf55rGL
zBxy7vm%Vdd71q&Xj$TaV<Cu96p`K+I^yCJmhq(`FMDoqM+iE@`|L1qH!yR0w4-dE-
zvKUBb9>;;bPF>vSK_Y|@kJt4|iWC&S=E3X#io$EUEBWsgHg;&GaC|5vfZ+*_6JCYB
z_~}=IF58iE6`C7U#UAXOSxadrIdWf^HV&aK1m6WvKK!J<AslJNM}Lm>NW{%mt(!?*
zZlK~sRNTX)ICMR}@BxL&g*wvuFmZ8%d|%`2MH_^h=il8lR*#@t|7jvCz5OP`&Clr~
zg~Wfd0wt7#8uzKz2G&}ymfa|zA$0j??dTx%;~)RVJ*PHB+EMn07wLlNv6cNpH->kN
z24MIph4V06T^l-wB4b~~L)t!>g8e+h!1Hk#RR6m0zl&bf+!Ay=I@85vIasncI<3&<
z<40Lc!O6P5C@Oqll6lb6H)Pc3pa1mmddoTP5aj>w)-IIe8jff!_hYZ3?hUv%J#F$S
zmA((K5T|o`7nSqAoib<OE`K#s5_7~Qj=<m{yOyTr7GQt>-^JgT86#oK=Hp*?pv!c1
zAyJm_mr@oIXI+RyY4j{w#>smIw!XLG!YDM`o2$<a{O`$lJ@+~{qOu%yfrPoL`59ea
zXiQ?^4{G;z<bzz+I4bo+N!-8p;OZLY1=Tk50YUv7E3G*@MH{-o`rife-M*cS`0`){
z?T`Qa9RK_F|Ic4OX?*{K^2@f&XuU71iy(-QpRSdki<6(L66S_0d_kn8q-DjW(Be`u
zm!xHs(9%lM(x;`Ql%%AlZ}_JFzy5%?kBj@2;Q#MGa98rZ1pEQSK-X9YI^L207lnhH
A+yDRo

diff --git a/docs_old/logos/hf.png b/docs_old/logos/hf.png
deleted file mode 100644
index 9ba7a6f25bf89bd94fa1ba973197e0ede96d3f70..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 91529
zcmeEthdbMQAGS`4qCHk?cO0u$)u`EWR7-7%)DEX+W7OV7BP~UHdW<3zMP*3T9+3*F
z!ybtl4X0Mj8bOTr=RA)-@Bi?=uB%to)l0tf{@kDYzQ51SO%2boUu0)tVL5y6?jM#c
zEUXDEEXP^cP6I#jxp8bC`0I?{-3I|IEc{mw{~Wt#d1VFo;jsWqLp_$VF5x-gA1B>!
zo7`q$DNp6tg`Q$L_E7ZRAGfVTk1gz-eVMZnHWT3#(k=T}!i5KH_$xi%KR>(X?TA^>
z5MmYlu<7>NfwSyvv}nnspeZN^n`rg@Ywkitxz+g%^<nqd>Z)m-R^E@hlpJSTXEJL!
z-PF#gKKK;s&`^8rM0NSGZ*CtaJFmaJ&311sIPYM2f-$&b=6zCGJ0xU6du`8JC4|_;
z**|_DI!<KP%PRxp{QqD7Pp`nO<!_uU>C)Y6*Ltd>Kh3)b+0|xx|0b^x{-R7_sc9f;
zFQGrWd$K91%yxdit$T22-o4__Wb{F8@Q&}vqlLpK405tK8lJ1<f3Wg}o+z%WAnz{P
zvE#72a|r{}qMmRu)^i-6u3>K5`P5yNUCQQj()Dh_slW+eDBt>UtVh3wg@tZ;Zpp2)
zWP0@48Mh3L#nBpE`q*-yQ$R1J*LC-Ut`0%2Zaj;4=V#Eh@tut?QCSCKkDeb0*iFv-
zUkCYp<&`m*^|X9Y-ZT%mr1+&;1mR`P&4H@6K|;yN_D+Ux)A(P0l$f&h8Snj(U@u7I
zSBk^DQVH|GsAY9}!o5G}@RtAg%c0!*^Dyt*k?VTBR-zf;t=~tl>i?Lmw|yLe-B!9M
z9`*GhGCg3TI4|EntW-R-G%u!<WtiXQYzf~$C+O>X$*<p}9q)W%7P|dqo3MJ{(M(xr
znLb%<z+LVF#+OkzCkPRY9Tr+`FQsQg;#x9y4|%zezQ-}AAvTF-YW@6tdALo$X>7}F
zZpx~DLL_ADm73E;q*xSo-Ftq&B;rL(=ej#1PxFoU0`WJ8{AJ?ofkQ@%wAr#<BzDVt
zW8%$0Q_@60uA|fL{>6ZdV@ESX6}Yj48`;jy$GV+kr%#6>^K$YS;*@wZcy4rBlWDlv
zDYOC0#S@m?aP6GVzkJ_1s}efnhu1$W^+!f16T4cR;e<{n+)Y0+m_boJTfa9We|%y=
zdpXRk^FXYOxWAJ%;TW0xiLCF4?8E&awy!fD85sDlG8r@CW->`a-EV#kgyp06O^1lt
z$|`^VYyvoGwh|i<5HYRwMh#=dX2$FzcWKONAW7Rz)4{#bcHv^k#I=3GMap*6Ze)0b
ze`1E@@4xT#<3=Ni>+UALW-+G3PNLh+0nB8vtY39-+jE+?q#`xL%Z@Ox_Gh5S(d1bi
z{xCmgeb+L1zNmyM5#AdyC=1V)we26ZG8lYR_?2v@AW*aYa1H!*E=uq1$d(J0I;<e3
z5hyiUALKoyMB6VXJPv(vmC}YS>)(@<l$=_e+n9^$|5i_|-VtrvtMH4+)ghyk_GW&L
z_Tk2Ng9vFs0Y4rGaL|iTe<!Ua*_x`=Oq<HQ3Lxru-W&JOM^?W5bx=Ie(I?EM4L)aH
zt?(D<<n(F6$?AQ&)8KZh-ner&?Z(Pk36H6r8}@Q>1J}atH)0lBTb~}Mh7!8G=@~(J
zJhQ4zZnfG`y9J{`*<JK<4X39~N&AvPgtGo>wrASHDY{9B_AlN2p_ry>(Zam)Eghx5
z$p;Pl#30Af8!MX+TIQC&bqm~X6t(MLRy^61oIdI)L=UX?->JRw4~~&1KN2c*w1T&?
zEl{TeEiC=#*F4SMyv{z^n*7bP%1`6BTDB_{mDknxq7C>Sk?qSz?YCu32J3aQ)fmj;
zV@tW(5g%T?di80FuWu)a)~GjbRc)9eygDkNtwV6yLH}fKP4(reoGB9065mv;-99j(
z$SQ7n{Zm;#6r-@u{AuKcyW+F&=m{?$jCaiYk)yD&%U6*dRU9xCjN|<UBT_=miak)O
z=*V9Tr*juk^Ofl*n^sl1e}7%wGJGDJER=B-cCe6N!kW8dOVt=@>8YGo#nzfg-;z5%
zXuz;AM`IDy(K=^?_1_ECw5N)7Cm;|3u$Jz(cuL~h>VfWFbZ6=?G3KCv?ywUTBnJOl
z!i@3cdKuI(VP<z~Vs817v?qG5_*QLmo&a}3QsU&uc`Q+Yz}33E++Ag1!FjZRMz<6>
z1E18yZ$7A;GZ7V3?Jh|xn*Pq8Pb+xS+!^pB8y!cuPCnsWf_ktA@q98*E0qc4tCV@R
zu`;MoLWy8mknr8o;Se2Ag*UNVVA;a@duq?|3h2g6RTp#y>|Abd_qK~n-VMC%zf-|a
zj5j#;LAR`Kr*@n1J>P%zpoeWZBkO_vZig|L&-ka-bSK~BocHCM%M>OZ@lx<;gq2U7
z1wAh3F-Lh%tGzq+md_!dU=nzd$9RNaqtvpq@5y!8+5#+(a7h)r#phsi@_n6CdSxFt
z-}hpfs?F+rBWtDTtjzbpFnB$CR8RxSfjg`pPni)ap)EiEP4dGDR<6*y!pjxjl(JA{
zwZ6VSutAyHkE$v8c{jIn!va1MtOv`tmZi=-o0`%V-|QLB%b+lCYV|OxC-yfkI;l5`
zW{o~qfN!@Rjn?_9*Xg$okzE)kpN?#k>94A`rz4~=Jw5ug=?l#Tik1|_tq;ZKedxHN
z_be&RWh5x~^{NeXNZQApXwj#|89_fkLL^T$8mtI)%i3=|0a56Y=jMy^gTRmE>Imod
zWyUscL1y|7)Et(ZaZaXUdDL{U2(!C#pu6OC7u|cCQB@r8(nzY^ifa8vV}RDYW-EUL
zp38DK*a&KLnAy;J${a|XQOsS}+6FU>(T)8Ndg@aTu7xi}6_<DAb)=X@?H$d7<(k0U
zar5K88?qb6X}f>VJNXKXt=HB@5$3<If3P`KGDzJ2!OP2pPY)BRn_<s~2}!nEYBfIm
zpjEGk?;m5g)@DEG*}@L`V|L73M{9h=w3!PPAzWNsDP*sza=bQi>p+O2YOi)t)wOLK
zS*^#x!C|JF$L+-YvQqW4uC7k9#WOrQdVi(dwY`|Qzk$;YTCAdolB4ui#<ni*SI|Q)
zM6+^;iZeT0aT=TSwoyCgvW-Oe2ucHcZ4(8JiSAq5kJ4^68;P9hTqI?WCYGeNI6`xU
zAuso{8s~FlT#lyc#~ufLds6hI8X|Zuh<XWf?5z!Z1l&*KGGxsQ>v3+e;!!_d2`RLY
zdT&(!o8{V*e(?b`tUtZWxXK&6!kmfOpDE0+fBdCOxm#<kZ-0LuNt>aTzE#VSKRY87
z>)R&Gb?m1Vje=Qx&HhT?V^<E|o~pAm+u|F;9Ocd{!&V)L_R*ub9{1^J?l~rZgb4JJ
z`z<Rs9oMzyKcM3JhkF6!Z11GZ7>hW+?7Ezy+j>6jkkS9=t89V=1F?`8%jB@PubI{H
z49!R&P0sYcR6m1I=03O4^X|I(8s4Juk~&{mlHIA-?6R-<zpd_E+I>OcgXOpm;$*y+
zo0v;T&ESRFKxgOgk?^LdsHltttV|4AaJMN}2_#Ryzjg4;hLely(_;7DSod5=wEEN5
z33h=gaK68~iGzO|Uwzl8L%ZLQoeMqu)I9rhZ4ZUGNrS;5gRA}*$$8feA>BB=?}0jj
z=PcWo-++1SRwKjcb2ru<F0ph)6|1DR98E>s;K2p_@q6E94f6f`?Z%4JK6*aHPng9z
zTY1y8u!8c(X$h4V)z3Uzo35+Xz->2q!o64bCZl-O!;NPnmg;79VWZU!e0u%Hi9B29
z>76XKkCdcxv$L&&R$FjQJM;RqkUhCQYGamdh^0Z$PQ+$<|Gw6hnSi|Eoa%M*B9%*P
ztm!-M6o|D1msB)z-+R>ld9Gtd4&!C5Qv1kXeix61iBUM&@{xXbt!jES-HQ+TG-P$Z
zla|}f0anL+RL5RS<6)Oaae@Ug^G--JeZzYsCwRJw?=c5^ff|XEq7kE`)Zi}(XNAhn
zN~~Oc`can5mejJ$+{Z=!kYyZDm-nc~^%7^t`@gZj&}Jh)#7TkQE_b_+gxk&#+A81S
z&Jav@8*8%2u_vPmq<^OUYU{9d<_5oDoF&H*e3DeI?aM~<ibi%x)_9^vz1pHC$;(p|
z_${-jWTxJ&p=CP<8+)G|JwNrgR{ncWoXFBA<K~}h@tDSI<mHlr^h3sjC^r$@Vof?v
zGh>QT6yc)&<sdh=X<a&FH(qS_d#A+TVm~K{d6a-3-LI<W0w7v3tJ@nhb)m4<nF$5v
zJzUy)(BQgl5{|I@5B=KO>oQr>bbHmK)x+<bD{)?v&^1Z(8{rRMrV==4ACA}Ar;+OA
zv0PwK2}(0a$CJD9S$E!sOkd>&exe;6XXOQ37;HvaM$zB`Wfw182(z&`7g>~F$j{F|
z5Wcg&(ipW0^NkqYqk7q<wKffEx%g^g(<j;@Cr{^bEaHxVc<AAvm!p&XJxMYNZKJQz
zwuDS4q~tNbM-gSpCTXp!T>G|azs6TQ$ocV-=xKq$jB@S3^ShsVu(ghS+2E7@$^5+|
z;iI$S9QnQ``AEdaHnN%0$xU$Y$j-udzpshn{=MBL=3A~}%ppkaUX57v7EP1X&D<5D
zt|Ta&<>ZWfX<n<z`<c$b&Cr+=2m7lljgi~4y@ka6Hic+^`+(;gSu0PR^6}Nt!!`RC
z@jKqXcV%sc_3w{ZBzfi;IyCYN!M;qLFxyNq<^J8{DXB~L8P?PE*Rnzy?w(}v?r`et
z`{q-6lN6n8<5hecALq*b+<@k&qzhhqx>j)98k$$M=GL<3bq`c3ZGKJ0Y-*3lAS^FN
z_GeSqb>PI49|`({tpNc6PLZo^0>rJ9kj5$KDQ5lt$Ji4i|Cp>j2dN075qqnS2AH)u
zy5Y>8PG<n3DAeSW7Zhou;=g(^g@P>mkU0-MT1*zqJx7U!HAs{yo{uk0N3yooDLg_B
zDKe(@n=111qqZ@K?gJ$?^)~{-W;(J%GXn{VTO`-^l)V^z{8pN<w!+g^D`jQnn9YTX
z?)}0)<pVV?*ZYS(f73cgm+ZRi6ODavJ!!}kX*?FXSX}X8Z88WZIdfhi-0A4!pH#(y
z`yiB`S9r-_(BCM`z?Wq!Yg@Or9@-~YQ2qn!wNLFkoV0?y<%0Ak)J{%L&I?#UkfkQA
zvI=i1!C6+CwNAG49p6PRN6y518AkRR2xzNnY3-hqc}huGIX_aG-anHYu|y&6E&A$~
z1%H>=6{D;zRXjsFig`)Pqb^my+o-iQMo-!Y>NHkcD+w!<;O^O2bE{1|(^VF64;B74
zv$bl+XY%(SjTa-gu-#t&mw;S8$Ah&EPYC$rLs696NHMlRn+L2|qDmAUlK$wuzt6rC
z`3Vy$_+D}At1NCeqsn%Xawe9Nx4pg1bc)&-BsKz&-n11#WJD;aAX>eYsJntyVwCF4
zdq^hTG?Dycwb3!dGs$yRwLh%y;y8E03d3+U_jCVO=*ayO{~b$D!+XI`;r$*NcGiZu
z=h+1+Y$~R24Ry=%f@h0gqvDh1J3I8Dpwr&Lc$<h~Ene_SiA=A6V<-*fM@-#2Ecap}
zf`VHUY;cOC_Oiq*VbNhZoD#sTv-SO;m;&#<lwJ_7?zl-wbF-vZk$0<j*#$uKaKa~n
zDzVSsaX$(wHl*|M@@k;-YBGpY@uPm^ZB2QJn78e@#}5rw&>XqJ+psYIh;+%Q0=}2h
z_pOSGSeqvo!?i=LIGnT{u?1!fWvL!xw61T%Vs&}UZtB4*4(=Pan8g<7&XuP{xDJ}g
z%`Edxy^n01khLn&ag&aB-LUfN4~^2C97Relm%R4zjlQ4Cq0lawDx$gBTap_-zn`~1
zmlxfK+Z=82mbY;hcFaTZcY$tVWL&rt`ZxO~K`s0Brre=YO+M4G;(nn<LOszv@k!>_
zUl$_$KJgg+);W)do`M+Vdegz?FeU3jJ3~4z>HDWWOHp=Xwiz$>DLp9x*_$dHs-?U>
zV`hVSjIn_3ZJ<DJZ8SGGb0?s~CEM0NGTIRHvCACXgSd#*d<s`+!|h|`h*mp!H0$SR
zB;uR4-r6&@M?)(D0uF(%Iy*Wdr&)Eknho`DX%)qD@-CU0nYpua34h+mvf99pn3ODj
z1GR~k`C9X!jmEkCdMI~gC-tA$TS0QbT+D5k?CQUN=l!(_tmimgO@8voA~Wh+g=vCp
z2%e4;lXt#3z&#8iX73CZIe{BIa9sl%UNdMvQ|smjCqq{4W_1-BS4e4vg@t0JkwJI_
zK|#x&J0^%K`y>J$?TEV{J>O-Z1}}-y7=1kK!TF5uqHkTLC_j;Q9bF$A8_R${*@^;S
zi9!7NRrX-jJ$l-`Nl*_uTIFrS?--zi)<__R%gtBK;f$+P1PIYd3K>j#BU@qs?)Ob{
zpL`hcUD{?RLqh6*GvMKL?Yv9+-uO9Epp(b+#xyouT$tnpTXXr{yU`&n-7%<CaYzjP
zw$qLw!v;16NUy`mKqWl$lUmbinMt;46#lPRJc&B<duNK4=jOscUwA^#awx?1ZXJ|b
zUA%~{htTk*qh+W-2!_%(J3E_}s=FWhNSWY2^{IQd%54KrX?$2G^DK!{v_0l6D-)IU
zWN3zG#sD5U6?_l%Tp;KHqpsD{+mqNNc<27zquPGUK*?3mTd()bNVmcT$LI303tm~c
zi@uf*@iJ0;Iok{0*SQ>}F^UjaL_ZesIA*6LMk9C-WJB=v^INGIWFJE5<uAVu8I+CX
zW5ccTgbr$5i=O&J9gg1`JA*C=<XTms*#9=egC<l~x5cuD^-;&-X<HDHtsX7ShEV&p
zU8t-rRzc(QrP4;T?%z0%P{Q28nzL{QS$2I$-&yv!5{}EXXtB_W8ccDomVp>gv$f+g
zs*{=+)Ce^vfF_DzI(D78<3MLI;<Xtgw{2o=tm*5{{zwun@l=}*B9yT-ZWhfPo!IXf
z7#MJm{$UgS&`24}$4-~0sbmXayDAFW0yz=-QaPS<<!3>xNChhQs>;%`K+N*{#d<`m
zNvz)(KdT21*RRpt21e&zge}a-k{+|ay&z$uWK+93q9F4OboI^E07&PF8&y9*5AkKP
zG73R?k0@mho8_rdj~4O{)?%o8w}#?7+?!m2zX!*C4;B)xi2T-l-@PFkcH?xLm#)Jx
zlz1)@x!1uRv(?i5J`NJ_EGpX1_D9(_oy&HKH(Qgz{RS$l1y5=`ZblhxtemfC5Mun&
zm<F43Bi66^iXl&1d2)E3B4{j2;AJGVuNb4B5~3=*BH(!5!L;!Eg`hx-tB(%Z*mM~y
z$&qjOALNF);CNwmKkAYwvHJXAt0)im*|SmBtcp)zmxblIeDDQ{5LV4=>`!oj<Q4j0
znL|YJRg<q}xUc8vm`nZEesaahUmdFT3GBPxfGl);68$1Tw?U0cSYCNloE%Kjp###%
z%O=aVsR*%2P{b5h_p<kC@sc;`P6s=<JpWG@#OA(oLgCF52kT$>d*3Grh@tYDn9svE
z@FW_?#6;Cg2(A+Vs?=-Ot~Ht|KU4D=+q&Q&rZ)+hbL@~v;cIP+<)-@BTFVJfoCsIA
zfB*iqYkXWxrWvdAyT!t6^oFOx+5gs{TPo&X;f}9=<layLNgF9ysT0Q7;s{0%=*kAg
zX(F>6WnP;pG8Sro?zCD{x?6Shl+Z-&<Njh3u{k=up-<x;EMKfHO|e!0Bz}b-#3L)-
zwBBc|Te#U*Z^U}@?nQRV!IlT1sn@oqUNi2CRwut4nsAD~b;n=)j$@=5h)((Is|)ED
zZr49^7St~EFdKb;ax=a63oKpbYMX&y(Mz;H(&mgTue=-DP4vY!Cz<D^^loD8iVbmR
z(74)Xd`D8`>{Gte=<AuPz1|4D$u*oZ6QLFa`@Q`5WV>j@LY_*u7+9szroDvqE-&Bu
zYGX|%*)>b{8Q<wVw`spAaV`nrU!WD?aW&b<t^iiaYT}p{`qh>9bcA0TnCDOkTze^d
z243j0<i#1!$BzE71gjZW-Knl2nDCK^Fk}JG)+!I1ZjigS@TZY{)$d*^*w-k;L}dJr
zhUh4b#vDVe{(#F)L6Us<Lc_$~l6&-ep;**<N%VY4Ze>Aut%oeOp-nB#hMO7DHlaJ>
zGhRP9VHalcDsTMpH#x_vIorViiBF1~EIy&m_WbCK&%Vt%##g-izLZVf7!YU8ChfT4
zLD>iriXWUgxLd_o7%nuTVLUD%w}qi_(!_RGh|X4kE+Zf{=n5~GW%s+BTZp62rkYK*
zNG;Ml6x)A{#fi|oa4_Z)e2~1+QyYK!^l2)iSdB0j`<tX`WUK7d-n$c*dC$`c=a#az
zJ7B9XYwOY~YrHAUo*LAw0yQIf9*TA+C7;kdqFT3RdW<fDsyxzHq}9kRlN~3dxl8(P
zlE~{j@_KSrFS}H1AQWkqe(+*pNlA$}x7NzX!{UW}a~b33H!mOOJirHP-JE9FNtP1U
zdkYN@IzNB@JTAH=BhIxPf65nImfjZ~bE<_3w$0L?w7%{%M6}Vu#j7TQSN?Yc1KFAt
zW!mBkCnL#P6*q>%{p_s_{Mh1LKk|0DQ-6YQ<KNBjv7uoA2nfqQ1u1PKVzv<3F>@Kw
zbC*w?I3ZRUY$d+iX_qPTqR4Z9<p;Ba8VWLWKiA#W6+Im>!K8}qqGL92#R&-s{HLnf
z5mlgD3#_2oMq4Ag_=}bO#Kofix)y4e_Q{VN)ms$_b-%jsz+=F$rh?+WLmp;=*-HA1
zpm1v)-C7TXkUJY<EtwfngqOKYTut_Eq<1KH@49w6RRvoZtT1#%qI=8&8)1s!R-waY
z?!HaX0cw4X>#=ea;$QxhETk6g24HOZ=~hv$B&r#zkrH-_)YdGRrIsVG+^>ZdcSJoN
z8Z{AEUTrLWsCBeaQWumHEY3J;XC^eWr}WX02ov`BAWI%lWlDOce5oEFlR<!1*tyee
z@Ln3Jh8j+vHUv&@0I*qk-gZ(8Yw@}vvBM|Seq(nXCXG7ZsJ=k)F3}HowxKhGhqgF%
zbad>tjfYGStXNIAVBr&6VIqIh(pqQ)0W=&`x6JBMyZ(45I{gZ7N=DVgb;YAWE$Tn#
zuQzRr8-a8b@poS`8R=AYM0*K1RHk$#s`k~Ft#0sId(sC<rpix2$4c>ZP2hyHwrr<9
zMYm%_c-eFxdlU3RwY#}R=W)JFuZ`E7n{vzL%4?F{-3kL{0d2vAk)Ze>eB9Qd@U{Q_
zFJt-b3P7T~;dUm1xq(m4AUuZfwS<cSh)YxhY1=>jSUwq3IuEt6H`Mq=7le2?3%wI>
zspnShcq{p2+je|1(g%skaLU_u$?*6z2J`)~9formBgPP?k)N(u-X@nhG+c+1Q0KXB
zlVN}awEQ6$q7aKeKGD1%{u#Zdu|b*7&_223>#863TYv#1htkdbVdGA>BQ}DA6vcD%
z@`wkAU9GQuK&Oh?{NoYFkrt@MEzGT<nX}9d_P_hd%bx1?jg+X<<+u~)DiB9NA!S26
zHva1LEvgIIowJ?8@gAf@2Grv%%3cD+OFgqJ^1E)RzV`KwMVl*IGH$%mvmZuIk)c%y
zvdNBtzDIm|GW@Ce^$bD=|LC1=%jcdxhDFe7`EdB%-T96$J&>6Qr0sqBK<txR9B%pR
zctc3cT=vB7v{+8q^qYfW9D|Wg4e%OXVtx3mCfH^W-z{uPbX(2se!Uj03X?<f*_`M3
zMJ0~AaTdxpf+}TBXVg>Yf^n+Xju#7RNVDYvO(RA2B|BwsK*RP}n%fyfk4*GAcniRJ
zmZt#)Q<p1$Wd<XXk3SCWjE(YhT23Ib$F_GVYPz>BblP>w5Az>y7m?W0q%4@&%~R-`
zGEYh0iV`2fKe@whXvS0w8@N*rV&u6GoCJ_oMAHK^q0?CRAdl1KZrrIaUW^qFw*WVt
z@d*@*OJ9K6tugGBx;15~84vtFOb@rT*{>KZ(j3;9Vn^rpZ-ld;$V}Q3_SgYS)wA)u
zCnK01-0&oEX;~EalsA~Kg*}dsT_D!_AK*Kzb?+9?cRTKI5LY<3*Ecrys?!$hzJyCL
zK6VlW*kr$hh9a8NUrPWp<2H_vW^t6}exZHRtYs2bNUNs3UV9k<KR!vU8(C<!!ypgw
zx@4ZlDjNFzE3hgZqecRQ+p8v8N=neCjj)MMzWqtyMEj}{4;(CR-?#jAiq8N=FU4r2
zMHO{Io9(A+pJy7rjQqm#)O`)Ud@GF6HMcCcrj0}cWbooh(PlMKBo3`x{-N}`6<-#6
zLIga{@l=z0LQg=ydBn`Y<|Xc={^|ik(@a8SW0UsdF&Q(QF$f!Tu*nmMSj?XAU3@!F
zQ){H1bqZf7Z`Zj<%@!LB3MV4f{V!E#=sf=CF6VO<-`dc-jybsP1={2uk`FeSq#)E_
z9AP$hl~^@6^XsrY;eW}M|N3HD>=)tQT8qKAK4x~t5$X#t4?cEay>a|t7Ep(3@(kbo
zs6}rlp{Vs<LtUG2DGcg6tBy79d>Qa<q7A4CbsWOitP!T7n7{H<>IzlI*7tT)n1uzl
zdLW)pH}}r5RU2RYZdcwLANPp4L%Q{pF`Th<DCO2;=Aj@ibX8@w=^%+l5aOO*VRz{~
z!$|L&zxbWzx3kYsDX8_axT<t?Dj+v)0}B@Wz<BnAqzBE(f0wYq<?6<28;)N^S`_6U
z{~dIi{@CKGy%o<j16jMy-<sBQ0=nw-LCN6XK~Ndqvd7;%;LYsvYQBn^u1{EMk&ehx
zoAU?%c<e={?Zszz?$b=%vOi%(NKEFJ49D;}HR@({&3Fvp4nVXQ<cbjp#8mK~e_l}-
zX_{#}CYsD3+P0!)pE$`=50X{;*;Dqa-~E;O_mK_sgzwT#z4p-2*#bShk~^etr9CmY
zM`P;PUovL_>O9Sccv*W6Xrc>wHTE{^W^$PumVioO?-fk9@x{1F0qs%6V-+!*fH)AJ
zmL7VxdQ=f|K=9Tnw{fM{%j_&k&f{pyF(V_*cE-AZ(Jf{nbuDWZNo{wJmE|Sh7LERh
z(4+N7J=kmK7VjOQ(G5-KepLjyd>26&$eKg;vu6i?dV!aEu8rL=&+c?IRg4aE=kQPW
z>s8hGnO^lTF392<fBvV2e~>yE{SSGkgN(8e#H9DwsmbL&Pg>Gca^KuNSCNqdr}X52
zy2^w)fJzD*Y^kWI$lD$z?u{O7ghbDTOj_`wgu#P&DtZLv0-=<^Xe|+r%B@q|#iqwc
zZpbU7v3kpw9|7%?hZ2H!Mx{RYnnQs$4ILMi{vOfNx%SimTe#O~dlzj90U@%kE>I3O
zoDMd&;DDB6hy^#ael6H;W=wc2=5&vo@8qNkAE(I)m9j0JDj!^z-3aanT~~Jt-qxu|
z^8FBO?mhP}r*$~20L9p%{_4JHr)9S9X)T_{o~tVQ^k*_~g4Ylcq*R&X_iIBR&Z@mV
ztHmo{T@n~$0(6WrP|!Zg%bYJhOT@l1+2<sw0HqHjm@NKFsg0157#SL?@0q!ENz8R?
zk;tWI!$YSl&YnmRE%5R|c8xv?-pU;A?;m`BiA>*(&4$q$4pvJ?6r#3wKcEpC(PKci
zA){&chr$7oddNB7`9WY4z?epjGzs0@A{xB<F23BjjoT%YAJ01keZ75E;>Kpsr=xrF
zk?xwRVFWObl)jI@pH0vn@^`erpl-ukbYgClp6}`lBeB17ir)U=BzED#_LQH0$5-<z
z87L~AQj7wftEmeQ1w`~hz+4oAm7Uz2hpuk+85qyqw;P{mf$ClzY+3a;Va@*MfBdjf
z^}OWYMuW{djQ$f&r_|Zwvqud}82;>Vki~oRA>5(*G1X{mA_gc!41K9x0Weq!SR7(q
z{Y9YK(2}~&lvH$fU69~)^tQeo+00*ITs<h;%-S$kng3q2>tGYvzZkm5&BYb{PwVsF
z#BPSKdvwa~1vw*Z4g}I#RvT^Gfith_AkbGg<++94xKejt<_MKP(w*Hkz7DX(Wutd^
zC*Q^5m(c8h7AUg-608iOlLgts&<{G#gP&bN34-zKvxTYq22M^BObjPG$@(hBi{@k5
zFk7KYj{h!E-+<FZiI(YQdsh7o-dF)Z_+fT-b_3IQf7&;CYQ!6N_Ff=Cu9DhM(<Z)U
zG{~!Uif*#IICNxl{r)8?`5q;spy2-!*G%KJXCW$8sq)io?w*jTzi#HAWsgtlF~l`T
zcIRrVz?*f<YU`rbu{G5C?RhkA-}2uBzOrZ5ZuAABG%Tm1yY)yW9y)mD*+Fw>I&smg
zq-8XCz=)R&mlTUwD31y5RH0~<egkx~75F6u+fVHb;&DLu3o5M1Hyt&y)4rAh#p7;%
zQT^p59NLN)TMZjXmdxhG#X`P+-Wv^-YC!DZw<LJJY6<>StpNHVp*g_2a-TURV82qd
z5)?AxCCvt=&l>7L9rVRtP%Pn3AcHN$NT(>ff@!}{50ASdCtNoChNh#PpCMX)`|UTB
zj3X~k!}M2LrSUP^FE}c>BzWf=K;4O^v=`m1AQdMK8+9a~j3{SP&Fc;RJ})6`&uo&@
z?qsNEKHk<aBp_WeRuNn<dTjMOgiy8KruN_sFPHA#djE3WWTI?S6`)+f*Q=-fhV-F9
z0zJ2{j>zzAa-9t8&3#)Jq&@hnb;wB!WJjlkZOL{5zx9rQM2fAxx0^?f<Zg@DzGBlR
zFB|hd^dNjERgV_2Okni&=|QVfn$~_U!AV(SkzZ6RV1-Xf5YbL#(~U+q`JwZABk~=&
z;2^>X_Mn^l!JGAHYEu+5W54}iOU$?Tn7D=~awU5&+E$>_arPgIv@rSNRI^`trQ`|<
z>h?>a!r$0@TX?k-P^H+PM)7+n8ACu3$*O&#>+el4Br#tRIrir_BG!l9^9bvJ%DQJ+
z>(v3K$8rbZg4RF#v29X${#ueh83D1&eWMz&^ckA%1lfylf&Vn|&FYHYUN-#ZBe!>W
zls*BOsNIacy}dfF)6&!fj`@p<T3R{uC$I3PH%5#QuqFE40Pnd#D*j-|!0-OxuYJ0s
z_=5b<G2g~|MTI1QyC~nKW_wps89oK3rP{XdQ2Mo?$gnLMVxk7xlAv2Xm_R~>HZsN~
z?q^q~Scx}A-FSrHl<It57k&$)+K((Y-^z<_e)R1jxcIVRO62#$_UwM>L3`D$W98Lf
zM&54rJpcv~s-LScdyAJF?hs%+H>q{(hz2-26qd+=-@WHeIdlKIZmrV1GHk7Wz@Fxk
z?w4~Jm3Rx@0`^rF@m0{H5%Iu4ye?qLy+|&&is9C7r<(xMMC!%Q(x3L&qQUgW{>7q=
z+<N-sZ({r0{a`?Upl$+-6z%8ZBfn-|%R}LUlbDQUCtkKBvWQcBqWq~<XFl$@AO}nB
zUl3Do*ru4?+2Vw*af3AmH_CLovqvP^a3}Ch1By{AQ=l2orwn%$x35gCDl|+Dmul27
zqk3)*3eovmkne7iI(Y6#mOkE^d)eRB@xaibNg-#`_KlmHo6bdYe5#k+ic0I~J^HQl
z+=#0f*(@8;7O1byGV8nd;gHcl2N^EMFK={@Mgr-+i(je#D~#m-r#t!YviFwVGwiZG
ztHUHmxdr9h{b3{&FMzvDX}0j~9~s}yz{MKd7M3T*ht0QM+^$IKDz#z%jlJ-7*9Zh>
zN~wM=I%ebB?>NbP1gNr?q1s!|y6Dd5U{#L14yRmx;7zksFxBfCLsxF{v^5B{ir3y`
z*^IY!^j6h4DnZ^N(){VNYwTJ$(wb^nSG+S^88y9Djf2`*NnyU~f>?I;^xwNS1TQA0
zkuG1pjOK=mZJBH1=XtA=$-=alP<dEkuQt_jXHrdc$s=N}WC>H_ZId+jmz1zh@B-wr
zF!>(+k6x&_PgVMTQ`1l1ps?q1?}ApI>$y0TY_tQUUAafd+qBm)>sS58%a_Ls<4UPf
zDr;*=KC*flcLc2}lT!vGP}4v>4Lw3obvvJAO9~&v>*f7f>_@k*3I!CLw<xPS3HpMJ
zgm5WV>o3av8=HN|51wNS8Ovy2<MX5m?#2zQ?_RgYGPa112eDCDR~5J<a;8<2|L%zG
z2)EY*mv1~r*^Q6&w-sOX|NN@KZL!-(at6^nv%47UFNbE%piyQprz9_ERct4q;Ru1|
zC$>JRQB^cFc5d381<9g{7L&Li7>1_dr_Y=K1-W!B^ooV}Y_5FI99Dq)g#x#TAKv1g
z8#e7u8n#EC`qRThic!}B&I=L+I$GC0XLTy=pI35W^$F*G3C`rpanNgD+#iQGc${h$
zEuP)E#8&-IL!mlAZp2W{Esp<~DJL8xy-=YK$b&e?$3SEuS~^Fh0xZqDpN%$#J!EFx
zFy88Px48O(x?0$RSggyii%G=}@YNs~pPATejyEwVsKp9>aTcV6Ibrs@q@-&`-1HNO
zol~K;3Z|+9ji_>*Yh2P-{B?E}?$+Y(uI-v|TfWQe9<I$f-BMDRj@$Ww666}apB8gy
zEOq5_z(Ep``)((~zlOQGs%k_I<%arWE)TW<irY_{yzDOdy_!OB9AJCwtwu4%87pBk
zNCuwFRrR`hCstC46D-x}-km%0`SaaJb~rx>Ysph*b)BZnh$RXCrPRS~e03X>*P!j`
zQIU9}f^+-W=K^;axnm18vb-&fla0$zx7zB-*Y_8S%L=s=w9i@B@P0}Xu=i`82P#DJ
zlC_l9kZ*AB&8MVU2Y0SipeN2oR`qH~j2;|ZsABakzy5T?J&?J%=oAh>WqZ&~jD$bo
zOf!z0dB^5>71~TJqq>grWAZKknhb3^>{oII!+Np>VmbY=x!R))C>tp$6lqSdGv=!{
zw>$5E6c2dNk<I6ERgj53ZDk|a7OioCJ(V-tA<}1v>q1Ze!2WKW+{X40a^<^@tg>+Q
zH#_R%?it;KZGQ){?WEjX6?vgF<mxkta#TXLSlJVG0?9pJ(m2QF{E?u5FD_RWK5HN}
zcjxY3V2&iz5-%GXxgJ4X%MCS;9jQhcsh@07%3NjcXwmZX^X(U%fPzWw1Uhlz$}Q>q
zl@6%4CWs9kbpub%$yAP5ZzsPZ_hl?|ln}rP_yw(PO3WTbSBRXO!OP{ngy+&YLpNTc
z;itI-7T*cUy;EV@M=tq|h~_=`wJRGyysq;)cl_(b`p{l9Jc%E>+NeicH<b`ifjRDd
z0QMo~-$s=}p)lmVK_!Q!JG!4A^|ZWNS~z<CYDG&sX3>ZfA;8byv|**YVda~hmDN}O
z7c*8;#|i>S;q@<bQbjltWuP#j#dkq(lMt=8Yk$f8ulkr2q}^~S@_T$ePwpC3uU=gY
zXC-BCO=~lPS^y~j7Z~7`Y-krq{z?duIj%eR{19K0Dj;Xz<>b<;_3WD9;;O*sup;lO
zk*k2+wgqE{2Mv$@!OhJ*6Wy0(ox?v~l8F$d)D$ppH#*Sm`b1)@`~C~6PgIgbzCQzW
zJLBq9ay1)V+4mc2nB>Xy=-e4tM2)S<K&!&FgBOOD5%}pAGRhQA0_f9b@5k`9J~U$S
zxj_4)=;jer>sQ*FH*Zpxs01c=adGkF?qd)a6Xx0G_jtTO=DX%nt$v0OB}{6;u+{wL
zult$Amhh?wW-Tz@GhjR-yYxphKF_8I@5V06!w)Bo92y#-6u>~7!GLi0-;9TaG4-L-
zXIu&Nfic?o78-tO_mkbk4!L`_k+sD(esf{?w{CB$U--fE=g%Pq7$d02tR*mU$<mAI
zFa8JE={-i_B35jD!lwbF$rreWBfxfBal!{lpc*Eo`;d&fv%!52QOPHy$j>Qs+bNJM
zU-2!u&#SNr+A}iW9F?xSD^`p>nXL-jdCgpEShhskjD`~(>90)Fg}2_BLqIL_p}_wa
z<+Ss6evub=?LPPy;UNnKN8?<*#E@O*sSP2xvXwegdV(ZC3HeU7)Gx@rOJe!`r{A6Y
zmH*J;s$g4Lz<R27IGrSZKUR>%##eUlwG9SaRy)4XxeVBv-b1e^_BU%{R{M+<qAUXt
zK!#ylm@F-H;lh_F{%}ak&m@H=bsKdIHbw-*r6zXevh^=!NIoA&3qoZ|3%{ln+1{1q
z0f!cdtDk+kkqNBc{l&t)eSo5LH%6_;o?!g|NLCR$J3D2SU>^1p<BX2V;r=4t&0MO&
zxcla(Gj>=w9PSIK$WaFiP6zauaSsJW4UKy9${I>|nm0k__~aNx+4nN`gv+kKXTD-U
z!ckA|eHQzUl7EFw2PSn*Ld$dPy`q5@@1_xy8?G;oIb23DpPRhk(b<TJaq55gME6zR
z(~jS&Eh-iZ$>10<DqTVAPc#TUC|?IS<GNaEYBleXfffLZU_7u@xz?21rg_EFuU?ct
zfO0eR@Bz``7v2ruAF=nfF)Fapocs3VBdD(gHW5m!m<Em=*KPbrg%>)E^gv^DjfN15
zo&ch1`y|^U^Q}@%EUat9Be<FW1pdBmSP|8>-a$1vd+%q|#-fY}(1gaErFX|z8+$r{
z?L70M<ZUZ3)K&2=A^Cdof1<@-REH6W(Txkr2txY}*I)84AtjL(^v4^-^%R~b1L>J1
zZ=`v_f@8w$nxOFq{)(|~A*f4cAkt&Tc*rtmeutPHTC}l<kiT;C@>9}BBV?6l2eok!
zgEJB1qiRG?<FR12SlyqkznR&if;a0iDjk|a9_aKwpkuy?IpL6qkh7iHWqIQg)v@+E
ze&`<jXdARsokNj#M-9<N5D9$sSFJ!tH7TB^t;wXdY4a_F?k>~>n%K<;{yZ6*v`Z7)
zr%fL!$<jG6><9Dwm3SrwHo(u=?L7VVl&kxI`{NoV;jV*e5wxg>F!@6-QZr<KB9A#X
z^JxGa4!9pm!nwJ*Wu=eRpEldJ^YSQ!=V(aIMpK+(qNW{wBMnCLaK;|p73WUp62{I4
zk7B>`Y%`)Mp^BKq`hsz<X>6gvVqq1C!xO|I8wscwO>2^eqCzEDE6hEL?9mYD0+zvx
zHfp$T&Zc%w3t&Wkj*De>QTM-qh!5P|-E{+i6kYUNB2ky2VgxH{sV`RDDle);A%dK+
z{v)b2X{L4yYSB#viJ9rY5Yn+tbSa3`{%0|T@0ekF*U+`4*XVH5e81js5Y4YK)rMVG
z3;g{9s*dD!Ib!BVQROR#^ZAPAWvYml;O!>A5pQ%+Sj%BRHg7KSA{YGw3ZWwFFq5<a
zIQCP|E;w<wxpPUbB~*1D8NW|N|B#*A$nQN50+Ae7!5u<AIlQNtd)mCgs1IuZ1?qX+
zt<9*^v8inxE-=2td-`h)6i<K^iH*5$H!I-kv6a(jWA@kb#B7Y`R##Vnpi-xSK+x?T
zu-wo1YmdA=IreS{2Yad?q3|t*OM;G)jj(k~N|@p`3E>n;pdXdSctIU-{CyhITD@~X
z{B0Njq@EpnTEv6{H)1r_T+kyAnBDL{5qE4FR6VFGcGREg5&N^L`yTh@iu!xhkg1Ns
zyMLj*hH%FEv`SlbYI$_7>Ct<IC%LU$ODbJ<cNKcMK*#Q07L>O&l$BaFs=vdW-D5HU
z&#%!D_Ta}*xhqi@$eTgE+v00lzkaM1=RK`ah=diXC1*P@-VQr^w<BwZhlh2AgoSs<
z@zFm}qSRXdsW?<Z%Lu=~r|GbkHVkx)+^6F`RK(X(`S+2Pynaz8fufk)&Y68bt(iTx
z%0!0xq4K(EETGRNkx|L`Ry2JeQROLVwH`=&vTE9K!HF>uOg&n_9X5EZsl|_#>>69U
z_byv8(VxNIKxVimFJ^yRc93=}twqu_$3ys^N$hIX^i{fJagceXPPh&HolEVx(*kta
z|KJY`i+14z{&(rhyBfLEZ9yfKw?WRBd3Crrf=k3}tXQ>gZZB~-XBgl+n&8TpjyK`j
z+!&pO2YnBOI=dzwv1oYw`FSBu&dsdRxk%HmVL1I;hkVl)uq&O6t=Xd->^Q}27H9i}
zf5G)1Q!oLw`0uSf0j{!w&ciOe-Xx&^*{;<eA>l3DTbIzYKg}Vx@p+A~%22uoGH`gt
zSBCct^j?lE2LA```Ee-x3b<0_FnFYUzG>6cuJlcdTTDPulwGy*X*!MfNtgG`?F-yv
zkG9R4lwjz$8}o;0!R6AlN?TflW+c|v;MN%AfRvSqc2P&(v$sAg%iBeu$6aYVVqPrk
zC*x=;cPkX4QvCdH`XrfOF20~++&@}8?}N(Xi1k~lfCHZXB5DLT|C2+Z(eKB{j~`jF
zd|9%XMB0b5o7oP5)yHNFRIB~F_eVDh(b{kSkSh`k{T?gZ-rhcFG@>c~hX&gnk@LJf
zXfKE)HWf%&z1B7Su8fEZZm6OEQmVpPG%G1m-13t#dWtz<PGj410(vh6+0mAVOV@Sy
z<|I#UAs?@`zcrOwWi5xPBRLUzjb`#!ctm<*jWWaM@FE)5-5zQlXjFtmHiIu?o}2C0
z9BfWdH-QYPtH#q7|JZ@CKtSYEUtn#0y2jDTy`jvWI2cac3ZF(^2vkft!eVoOYfjv4
z1KBx)e7sHlbJAH*>)gBf+Dp7C&3D;e2r;Zc<2^m*jk}9Lc%EU$4j)yqsm0xq;<WDE
znpP6JP~E2;<PLf1Q#IJLv$3XouqHNK?uxKAYTvLBIj6@X%>B$lUF-Vwf(;)FKD^CY
zZ*<Ur-QAL@k);1c^kkBqLl>VPgOid1MctGnE1J>NKVa5CL}d!-BB$u2(Yya*i<qeB
zKEA5JApovom1!>cK36Rdgqv0vjc0<#Q=&aQX5+bacUQVoQ&a6Ud?A6(FUX|zeI4Ig
zKwqJIXb>G($7eq{9X%4dm3EF3CxNo{=8|kgb)u=fkDuc8ZzHP90agAna4k@!E=#~h
z_-IwLnSQ8g{2|AH$d0JkgpR%+o>u0(bt+6H#_5Zgg2u;xs3D_XFlytr!v1deLFE0l
zgguQQL-QYa^2b}=t{7o1I!)RWhiGowHc&?Vf?M{>f{yx)k2|}KEe83`*x`JE@8Okt
zK&e|nsN@}Pc0I8sRR>UrO`5*|j<inSW1#qKr7h;119-^kw<PP%gJ~?dUO?<RGY{=;
zg4<XSSq^RYo!bRG<!hLL9btRk`2>|d%BQUhbIK!oEJAq-V}Uwqzv4Bw;K1}n-qfh2
zRl+M-TP&?P+lNW0lr?hc;ftpMBM@IpeQFt#30TZ)`p&y{WcI3;@}n2x-DZ4MS+t%1
ze6PJsS&k+x4BxX$2I7JfP4u#cVAN&Oz<_nx6G-Lh9WZOh`$9do=Tli1u6ao>Ld$Cw
zy8H5fT|sM}=9P;yWq6P(R@f9Ui~u4;{vrk-kwzhqerN#}<gaxGV@5M%Y4>ieX9pN1
z0`yQ7))mIb*114ic~xJG7=uDKDh*ng)ItFuI}HPLoFw*VRUpkdzbMzJ`))WjUhEbO
z$jOBDzRK_Q(YUGpzn4wsHy0WmPu#Fmmjhc}%|`IGh2cV#x#6HQH9%YaDL&C0;yx7w
z#POs5?F1eb@v_A%9hV(%o5_L+-eA#oi*8=f?1qbnbQx`M`=~O9Q%AiTfm=(gefTY%
z8L~K%=3QJrIC|R~_QFor?{`c_9obyO{xt?`8!;8DPR(!bM!<ZIfZ3W8SctceNZo31
z-UlOkwsxmlyu`4oeO?*!b~ye`g$g`=#l!dzkun_Q8exCek!~qyp3|cl=Asd;4|j@y
z5Q;h!%kKT-bqup6hFNhi>9m`aal^_1dH#kgLH%y=IUeGiiLdQ?yf?U^`;pBLe3>zK
z*Z)FJ*5|PAA|U73+HYL)6cIrqqN?#E4rCvHA1W?8GxPInKXpz2$v>t8R1a+y29O%A
zDTJ?2D+6W8QEgx%H>`y_@-tYUKBh@s<?6IZtw=~hK{>qh-OAckEG3;ZK}riye09C4
zHx@^v8r7@cLY%-IO+Ag2LKmHudq>cNBtE&+b&(8SQ`FCVB|vEt8r0J`3+IcI;;~SS
zJY=H(x=3JQPT<2Ur`eqXgSD-X_b{W^{~8W3#N6M`J(vg#o+HAxXJ6r?P5<@N43@*X
zBW_fxQ#9Uhm^Ux4toY@$I0im|tfcSOcQz;u5L>0a{YL6yf1%K=t2~?x{H0I`FBdx8
z#LhP$s}kW;`HBoAW@y1yMR?m1FHJSIk|$f$O-)UVVWaV|x?w=gDwWr0fAr^P7R<TD
zOf@aQnj4Ql^j<-S#$nwGi9`488sHeJV}dL)__(~$PaqY;`sz8}a>&d~h`IX0u~kdC
zB5dY^U(ATF$EaTIA+>lwAL=Vc7(fp0SH9`);=Jt#=vWerkQhdYCV;^JJJ`<0PXj{d
z32$Skt<Pa=4>UrCzFz;WX<$!n#EX-YvvE0=RkYS}M#URqr~CfSINu+vgHILJMV%`E
zzi3P!I<9aQo)p}-`ho0eYjh-$pyHfs)x!8}r}I}VZ}<*%BmC55NAf>?12BOxItk``
zOk0QANmsojKcZbPh2IE($5wf^yq!Lq`ru7&_Kk?(KCtVzbUCrytSrkmd})MUt&xa;
z%eScBk<vt(-`Au!Em>2dn#zBExwNpj8wJxD-}yD|l5?Dl+IM<bAezb!I^!)`Tn!j4
z$$#oeAP^*W?LP)dfN*B!JxH2pEE46)xEBCFhH;pAk(tbCSK1WrN(iqtfbKQ?+@8rJ
zUaym>9<>>bLTZVV>rsi&Jhg>^a)~#xkzX;;qmx&kY*{Q?SpDwy(&9ZrRNx5t1|h~`
zn7`Boa9e!EutreW%=C2AJLgugUhK+@Jl+zR>>%mmNC+TVzM7d2Kzw`rFUiJc0U;b+
z%yZ)VH2{CGGI9MkTcVvnTqU#>;F&gUv2Y6P9RPW~R`578@W_7c39{y72>M#hXQ+(!
z=|hpcE;n@V#Nw;j;XuwZ(Of-Ze|rXJ>o0ZfU`-z^V|(`mz1>`hx||wG`uUrJsa;Yf
zmlqj7a?g(7qRvpUqL+{@oy-WA<4v2Eyw0X@bJ7EW*WBTHM^n9Z)m&-Jwb4=?Y4@h<
zylIy^u&n;SPXo_zCxg5YDDrz0zv#t_mfv3ik`j=LuyK^z-5aH2?Y;$msSbynblwYH
zZ+A0HvGWWbcMkqhf&rl}E-fz~w0+7Lt&*jLm!fDccOp2$tC~<nm2pt`M_V(3SM7Jc
z*45*nBdmYob6Idn$x{)-1Bv~^-)h8C!+3_hMD87Ek3TNu6U>P@LMhV#9w{!Dr&)5$
zU3cf`*SiSG8|k&&$yid3n0i_@->$pAs?f9u^KEdNNt&mf60x?j=AMB@Iq?3J5hsJw
zS_}sLrh>8U^9yg+{-=8S6r=^Fnu{+~$PjEH(D<^o#2c#@TG(X(E@kB(Re044h`088
z9-i0gG|MW#H~ETk?_xKM9gcU2lv`?tCSeoE2G~@<U@GUUM35pj7kanv7|*#RCv)Is
z5J=MZKo~j8MC#}S&#K1JL@iE|*rO`0JbLC+_al8_hmK1vxs>ynOGM*uKDl?wRl1<e
zd?VT<uNdw@mRu*l<HOm<iAi0mEc8r`)0`V()gQFffNcA_DkHiJL=9)Ij``+A#Q6<r
zDtMTP0P;#p7Cj#y^5^GBY6JP9?z47%m*uUa6Mah!|GHjnHq0`}kMm;USLlRluKuZ(
zXvf=dH|Io|(Y|d`y=iPaAgn9_O+Lnat2qjsRiT>YfNx{L1ONdY?VZN7>9I2vPEFmm
zppJ|CGqguKIyycmQQ%Q{y+gBKcubJzUE6!cON;xj%ecZc#6<u5$xloRTxxed{7Dax
zaYb9M&fg&B9f#uhfsj^5WD{Nns9O;nb3DAbMq9{Y;ozP6YhDtBWd1&uxL5f*v6mi7
z=X-6vanG%@1kj8VeTpa(SCu>et_sAEwKbZt4XKi+vxsmbw0XfBczO4Cq@xd&pI5U6
zw&0F>4mfvvqs<;#^{NZFK3LO!?1-W4kROmmB)Z9N?3ra~+J!PbqMD>wK5u7tI%>G7
z5}hAHww8&j){8sKdA@QZfMw!_k12XDm_pgcY-&erdNldD!ocdnfEG(jjt{%RO$9!+
z=Yi{#tE#4^XqQ8=ar#5o<k6qI_`#pj-@FkmV&UAww9=oSvJOXg3vNLeIAN-m(TDD~
zMi%=GjT$1JkP^2^{=XOdtknT=-2Db3vjN!<vK&2d2rm=%4vva|MQ>blf#urIL!GN1
zoDL7!r<8cj>^=TQTS<-CopOpie8foXiqQ0gw|I3bc*zlB4)#CYz6ou3S@}9$kX@2>
z15zIN8$W>Lf5ZAyp0Bu*t7xe>I?2glN;dQ#{nfu8=2hEpng08v5@S*M*U~xTAaT$)
z%RE_OJ*YPq969aig*JyY-(h>M`$!fPT(y#&Da+}6X7mns`w7FZKXrXS3=VyG_zh7N
zj6zn<-pVcl^yt6IInd?rZ~cZ49O$;dKGSfE!^aa(&OaK9)>qdS-0!LHe%ufS&&R(<
zZ6$1NC6DGkbK*W+C9*hhTV9<$!!Gy-6#rDQmOt)H)pdt;K@f*`89t7k8xH9D-KmU?
z4MI1#U}2b)br}^_z~YBeYWRM(sOl$L<fh-gm;2cXz!=dSG_|F_uUB~Q0x6lB^NdRP
z6t)ETaCP;nTT@z`<PnFCINztO$G!j>&#epA5*TsHf)$uM<eQ=>QS@Ie{E5ml=h_io
znb7uGh-b0Yx$CniLHYMaCTns@;9^gXxh?fm7p$zT>hzNHcK#1fUm4J3`^F2Rf+O`M
zrS(Ng5s@4vC@LzwF-8ugTe?*k_@bDgg0$p<9v#92q|1#4CrS+@M(4Tlf6m$0eR-by
zx#PNibr+<Ee88KIkebJIS?TntU6j8KkQ&%5TE#wCu34a1;~9tN9wAKZNyVdjZe;HT
zK(ZwDo+qnXN&hGbLO4ASCML;sfY!^_U(xd@2)BY~tX{h6x$$^=+Q<4YFN~uX+oayC
z4a(>Xn~bx)aU;6&8tFHHDB5;DKxFq{PESv-?gNA{q7#l;bft(|2jtPZm=7wB`ca{a
z1@xi65$Wo*cMYu`V?D{0N}fq&%)RMnW@p{Eg5_iWgyLu29}>00dIHil`O}rs%D9F#
zTsP-@q*{F6abshqKqWc1yc{{@q(Mv4+&`E)I`~kx_q1>;$_tPh2dva$wgID0N4jWl
zQTUn0RwNtv*1wJKK%skRZ_j4c>6w*-OU+y1m4a7hp8s77uP$U_i)U04Y8HJZ-^Eds
zhm0veyLgIoG!8TJU2hkFJV~P&5mbw*5F4fo0i$~2oVmy<e8cK+@q-8&%bu*w*u&e_
z>fvXP4s$ypv0IjY<A=Y;UXAuHUXjoumP&7_xV8FLoN!et!cZ=X9i*8mjN4rmAb0oM
z>?GRk;5k?#uh&~M{$OEeS0ayRx#^_r6pqFDtG;Jmo~Ruf0@nEZP#5qDi5Q@gKX(P|
zvHr-4yK#AVoAkd!B!us`9*HZRpFMU^Ql35b7EKbCnmh7dl(Xl%Hgf2)M>hYGfsR(#
zs~Lw;y|6z|*Pfza%r?a>hI!tr6}7tdgn3s9@BJ@?!GP{D8@Y1!Uh`N%71ZV&4W4XM
zQ}o+A2!*0ppbqv&0AOaxMMSB`55>|AV`E7@nrUWkN53+w8Na$-OvU|H;W|D_rOHul
zJkp&-Wq*Rdz0?S=K5pN&UfH#JzcQptiD?#%?qi|c!e_$>AbNrc%_GKkY-8!+ifwj)
zGqZ1+$XEu=Tt3)*qN&;zLDNhz(fcxI<2!ZK&@x%O9KWj8LF$=ZOrDzP@eNSLoEL)H
zcj&#<TxT34XCd(c?X+IaNOTXDIjSOgs<4oSNF-KQzT1Qk^f)!|h2Z_`lz+6zzB}e*
zrQ2dWLTd#eoen-Te?irC@u!ryH4%g<3~rk2FuI@$D=4rk5Th?P!~S=IxzWSbwIc!;
zr*?@I##2Y-i7_M1{Wjq{_xDH~lkd5%D2DDAH1S@KSi7RVU0S}UVj|w<yLldH&K%m)
z>I1(!vRPm&+(*_Z#c8eN{oO~dL%{3-C<I+y-ABvA8BzXzwTxe_<!}m?j9bnRhAOPD
zr}P_x9q`8nORVOXQulNivAe>hE&A_JvS%3=WW{~S=+~=psVL6ZQ-<dip#7LGJ@OK>
z>f);4u>RVXpSW&fpuSPh)y)&JDq04;x9~d070AA<x0;GEI?IQnm#7I_4K#;6rRDJU
zz8cM)MaMmNrWmNhPkhm6!_n_k{3FG-chn;0iXRB`mtC{`tC7GSuE`!f8e2X(Ty9f(
z&sH-6L}s9cPCegzUVvv2_$?l+p84Jcg~a-cyfR}`BD_<43W~eO4#QOkCX_}cg#Z1~
z{=A~UuH-p6Qx-Fmi=3MsrR%L6*S~gJRC>lgc`8YY^F}P?bYq9FGnzED9J6^g7C>k@
zRZY_Ko$0G#L%*s+KKFPA<%2vw@qqB2yw%nr@oo8`xb(ZJaSj_PHKzxYL5?Z_i&{Hl
z@qn~awBJ#*|2`eGlD*zMf*(Ie`CX@3!H7bTQ2zN}U()gsvnpE5V3Y|gZ2>Y$2*Q0s
zh2ttIgXfKgslklA_+T7LLL067EW`DB_i>@{VhBxlVe*e?w0PCAL!f9x&$)2r$r%1R
zHLPN#;mH2p*eFM%5aIn{>QNP`_*!b=AN_fiivpfm=ERF)^|9LGDMu^5GQH=8czLTX
zutxc>#It?t-?cDhZ5#XEeWM%%07T*`5g7(GJz$3ZP+0HQM<_h_TaK^uzKxA*fBx_G
zb<_>}h@8<o)a*{tXU~)*6TDkqplXjXjm?b>D;KHN#q31K)L!nxUsm=dkn&HQdRVb*
z?{D{-S8Aj3ygw!pA5?tU+WQu<J@UTsBYChW@_=dv1Y7WdyW%O>s3^AKvv^WV9ZGPx
zEZ@Perq$$lKvt3`y<iEmLw}Ftb>62Gh(z;~Tl7zw8N4(x71<D;I&}G1V`QY7M1tX%
zapSQgd<2L@w>2yZ00TdOn|4K^pI5^E9XGl6%aieYy;kFANp+`45(Q39ggcT}eBWVJ
z>1J?cJ-Z*KOFpB*V`^t=qy*s5BaV|pF9lTIa!m-FPb6yCBa&5b0FI50t}b`+8aB7n
z-My~v&E@0$yN~`Pk?WlH4_!j>QwDaHwSR9ZTRr2@{_j3bras<}2|c|5SP&_<i})`$
z@!6flFR|$Kox70^Cv@}Plglq_s2_N0Y(~PcrN|=D`uB$7C9cD1HJwP()YOs2sb=6v
zseEUxT^JQ|qFdmlTlwf+H}j!mV^c%JKgYW6>to77+LXE)3UzU@>VGR#R172TC65t7
z3;;%gRE!w_Wz}jA1}TNS(?(4@M#PCafunkAhb~gx<`W)mV-H{~>zrW&Y1`|jsTM7>
zixskKDGK?nRYd{A9V-W8({&Vhw7-A3^X1fnR6jZ2s<VrjM+mC#E-!v!Va~fvFmRh*
zYc(KUMc@|z@*lU9R=b-0BD=!yj>owEV=<#l)3e?3m)`BX(@4J-^EFv&5r5bJ(+o2<
z-p>$}dXM(ESCLH88Y4qPmHY4k4CmqaDR^8HN3m5+H~Av^`OyBK04%2VXwGFUfSf|f
z#%MZ45|+)!!Vh1F204x{&bvtIL~P9FmN6H4Nd%`SCGB(q8tf>>26SM(qykY~JM;X(
zb>HE!BPq_!_X8~wtWjr<iOysXP=0MM=ixA}1pcLOuRQ-^&U0ti#ql5Qa*Ib_%vj&e
zrMt0o8|DKy{iHBswoAkbNj}6_nw<J{WyqBv<sk-_^QCc)4S6NnlaP0mYWLjumz0WR
zE+Z_Y8rEzHv$9kbz@Vha+ii}YE!#O|fnN+gIx?#L5#rVYuyO?u&JwAyY0<c05&K+0
z34Rir)7TTM0A;6^Gd3d|r_bZBDw|JDBW+%=e+x^YQ?;EASh5wWT-Yub4BrjFmOg-9
zYG(E?g0=r|U#pSmZDXYinP^>Fh*<H_sf8U)h0{Iue_~su`p8RWLH_c)&3+dJ9~N}|
z0~aG10Vj(z;2(hF&9NTYjW)*ofU|RJSWzV$m7uVty%EP}OQ2B68&)hknH<-+yY>t=
z61g*)%R56$NkZNo8u~E%Y8c~^TQho>y|7LvbM)>}SMPq87-jJwSwVnoP2*dJn;v0E
zjm;?hh^T<4DO2OD`f@?|AK!F~zVqpRu_6M0^208uJ)eeinQwAoYj$5Qp8?bP^6y|3
zdwsf$Po?hCIH!MNHW$#5z@JJ!<;Y@o(@2inK<{*%>$M`q=UXa%bB6ahOeJue*`h2j
z@!3QP_9r*%ULlhtvyTgKus!C-mutVhmZI{IN>b7mw<M?#y02GmuI8O)^{-z`c>u4&
zRDV|duvPo5ajX4(=WJVR>!ev1-^+>f0|AE(8u$7qZjhRjV%y~Cd#KYfga?kS;Q79g
zTuE%!*hAXirS=cljSLeJMiMjl?0RI7(MD2qMy8RRnwpxx*!vKw@u`1AIl-fujqC2V
z*W#{^k-hlm+r2h*g94|boOl;Peyd<Z#6oVzv!G2>8ec8WH@XEk8AIoL0gi8vn()B&
zLt=>%g}pU60@AI5Yw6HmS0<0pD#fyi4~zp#OUo|Ps{<q5oA<Z74q<BYJ{2U@pA~}Z
z*F(x=A7po0ODtMMh8f8$mZ9+k#eqr-2~IK#ev0k)R2WVPlY`YTseC7NWaMRPupH@k
z0RJiL;(M1-Dc|I&>ZNk6rR<}2&7I31wnwoqSyfqCS$~6~!ON6RN`U-g+pY&_FVh9+
zK(yiGLB3)M>^0<FFg$c~a$ql{keYwO9%3X1+b!-&nXW9zcNcyS18z!;{@{|UDE_e>
zZKiu@7?iQWtD;gZs%bLPuQu*HcO>Y;L+zq%w+ra0r@&!aPF_!3-$S&7Z5(@FkY3@e
zFD<-ymL>#lex`d%sO#s`_Ko$|W8=umxeguzs>bzns0036sN%E1hjn&Pu!4tSF^;(c
zc9;ka{J&EAB?HnYAtT4O=w#4uvr!K<8n5Me0|Qo3s7JO^(UjhI>jE|Z+b>r)&H6^~
zq1;GLqJat8lm|K{@EE7dcvw_aRAU9m02%JEVVvQ_XY(^I&d~`O<;B{`yka~E3c@K&
zYBIcKh-Nz05Lm*cR)uk59kinVdxC-|F<}BG7awbqy&UHX_W<OasZ7{hU*<ONDOOR$
zH#Ki?dZ9U7QM*8v{s%{L{V=STK+;u(ui&TDrhf}zXTlVtUyN0Bjf89>UZHn-8~VIz
zd(-7OCxd3H9tfAgXaNJCT0L^U>+c@hp%Z~+Rdb6=@Z-CaUm=nFj4SzrQEV&yNFnW?
zqh!1VEjzKCyXPYtPx`4Yey~d`&-ZxOv|}!H{sCk8f4|}nnVVdHSFKTtH_NHGk)}7J
zuXEY^Jwyd6fVK4@g|o<1yu^35N6KAH&*Nf!dmq#^yvC_AGZsLz?cWSR^j}01T<WOl
zNIlhR*XuU^)F&mch6^W~w}jP#4BY}P8`%Yvi^a!bEi3clB5)C$+zAMpr2kjInX%H*
zba$*OTr!2fa^Yj?XM?G~L_nWCQ%tRN>g56vkpdAD|D>l+f@)(S7*!Jyfv1#K4}F*`
zu%9lMC5mI?*KH%!vv2hBjuh`+Wt~&?mkc_5hKFyiPsNYs9FdXzQ&UBo2a(0#b4sTS
zAq9q<2niJz#4GITH&uvP8`d&8BqZ_!u>F6EeFwhg56;Ty!91L*=ik<I?nXko{~OQ@
z@xpUHO2%-p%&DepN{F)b#Nk~URse)v2p^aqGUlHr&oF=Zez4F5iEVCk?6PKR>X(;I
zR$tw3A8bnhUKF|mD;o*6@M4M~c2=}SFnUDxtXr#GuU`-Sz4HOHV=duUSUq~TSD*qO
zeRVs|!mHk2+MVXt<E`fwfFMrq76uF?E8%Rbl<Anj7c0uKPKCOVv<S!T#N)JESI2K*
zIP_;AZQdD7&-epZ<%J+_g)6JmZDr-%<&MZ00HFSO!U*SvNS!p|HI9+2cmsd7@%@!f
zzW=wymJh_9X!~7<R5d}UQKSBoq3^yMD^vuPW)bRPy__r06cf4>*!SKxNVG^;c{R$3
zx|~QXl0LBR-)o<e8ik9AYz7QCAZ{_DfdC}+s1)(sKSMv4rfPc`H==+aQO#&uroxCw
zYWVv)a)if5l3~~t+|-jL@Z9(I47c412p4xcHaw3vlgC2G?=iYsBuVjB^i&ssBQlbj
zG2(-Fzw040R{%KMHOuUV{c(fKwKSME<K`}wb`Y1C<?4}b(3Pxi6m6#3XbEZLd_mw1
z*+8E9Q&AQo_05y<fGuP`xOSzE%x;Mo=%3!$iepPbAexqMhyv?74}7Y*n$z{N)zkfg
zBUKL>NYk>|yhD+|reZ&`@!cCbx%G`6j)<#<XPNS5&m6Dw@%}tvNMuR)w;Bl$Z*(L8
zo|>-j6PTu57Po=>Ui)W|6%)rJG%-RG#`3%w&7mlb;SDH%!)6_lV=bmu8Rx2twl)w(
z`Y7WbmzXt^0Sb|s=$C1hCa-Rjqv2FizsoGe9Q{ecki3T<BN4?2?|RY>*76DYjg3di
zK-(n^GPMAEzPe(@xDh>=m9#ovhcXgh1U8x3_G^DLbF#lTn^WAC=JpyMd8hw9wAe55
z7Cy_eQ-;5EkmM>03Ci$B04cS#yd>8Z7FJdX>2I7Id(&R^8|41uo;I$*#oQyiO)q)1
z8VRY17<kk)We2H2GQ}3v?0?*NIx$<6@irtrl=|l%GV^A_j_*Q!mSH6ero)R;Z;UA&
zy^F0sEPF?eD|TPr|H;!Bbof#bYDe%^{@sEMhX~I7qfHs+x{xY|BTPYW2mL;tyvg5*
zBl>1+k<Sc|cg`=!KCijt3rF$SowncH7AoKpy}`nwEiT}7T=L&%YF?^trsOn`VP;l@
zQdbI(FytsBt<>0EgIJ?4<I@SLNr@%SrrX0UtZxrtWg|%%d%w?OYt(CB@gE<)pMt1Z
zc-%D|RNhucnQZh{#|T*g85{BFJAA3S=wrQ4GtvhrlB|7Jel%T%$>2Ir{sD&2`7B}-
z_nYy`$x|0#<p}`y{g()yGtYN@-e7p8IuL0y6;2tbBJ`SW6Bk=KAO`<~66sA@0{?To
zEy)7UDK?97;dU@m5GcAcW8^w`=aNB+a;7UKNaG}jYPuI_?HroC&1vX1Ne+lq!tN@{
zjXB(dFlLo&?b5I26MR>6`h4+h*0lO+O5(X0>V{>2_@b%Gcy|)<A`u?PEfwS7?3`b6
zET^5W!6&WgTM0LhrtRz&DjWfv-Rs5Km3LR<?ZL?YP!Ndplsx&BC)!sj`NX)fL##b@
z?#UD26F8nEII)vNtBlkG4EJV5W=pV?HVT&HkgQ&yh49GsHwLf3EUfG;!hYFfdI;E}
zI^pBGAzYNU3!kL=bdfiPEcwTVg`h^7GEpa5BX>5>VYqFksRnf$2(_i^d;+<7Xn)^m
zhFK|i8EPivybu^f*(qDyDwxemEsNVw*W1};WH+~%uOG(09}x4u@-00z<)uuH#nr1<
zy;~|g#(t&%?+~?|ZKgjM-F%L`$-D2G%1-ao?ZmnlM#=R&_ZO5Zy5$?IF4E&Z0KeFI
z>4C>*#$DNG_|NPHgn6ci;PvR{*pPPv|HDLcoyfn2V?vi@#peg#5dnB;d_Wdp@w`ma
z(HWxFX3<1*;1BDNjOk34#f_v*WOFS1W;1QXoO{%^c4!m1kt$o6qp<2_4HZ1m^}*nb
z|CXkrd*IR51I3*md;z~mLz?7ihGn}cU3qn<$_a(tnytZQYtiCAk(A&(cx>cmr_*S9
zSU_3gQvI@cX2;g>{ZRG{?^qkTvEALq&H<O!Vob~{Tw$rIbld06W*r&AJJ9}U?v-is
z*U#sEVsq;O|JT<>k(U7MDeHPsuwn$9?9=DdyFPC9M%6JrbCCc#iedrCvu@AIl2tii
zUz1KrXUe-rOgFKo$YNGa{-is^5;ePoA*i)-2f|${_0zw9quYzgXITQD1Z>W39QG%H
z1kh-LjH8h*Xo0ziqK40!wB^dK<(>lWl9iP&4D6ODsmz^bd9nS_z^ip(F*W<=jG0Jh
ztQEpqsj)=NV%^Nr?WI3O9_81>!}f>0gjGGHfPC$pHl>|WY1jtsy$0&>_jO(_iFs}>
zBH9{vdoEG}EB!Kxt|FEyi-Py0ip)LmKdgIO7!T)C9MJ2XnE<2Rk*{!-_0R7vWBWe*
zXr^@b(T9buOc*2at?+*B{9^n+GC5;rmOf~<ft@UyD{PbhQI7L;2ju9hE0QlSuk#rS
zTsJ*ZS+GBSZRWV7$7!jLr%LM|Pm9MVo4f<Rhu)3FOA`Iiqqq0sZ}GdN-$*Jr$k!KQ
z3nh3*Mm8e#oWMMha=&Ob5Jr)&wUm)=L%QXSzcq7u&&AToSg2mN2$>wBAv3-8p5#2e
zmnwABpAo*==oM$d{vTJEWYyP!h}nACRy^abmzvvTAsedLk)!BgL($Crsm<En58P}Y
z;FR$4Xug)t5{{=dvzO?NOlFUn=@g0!=pXwSWIm{+gOC6-0kxo&=Rn(PcyAt{qLA3|
zo?`N9aavncI3zxfvR(hY_cXe><;KnbC#}=s1BcSrz)hY++Y%4wvn*d9BQm|^?3E{b
zbxx+kBp_%4h($w~^Us&gFu(E-ZYVyH|4$iNPA=No`nfagH)ez8ItDp%Wcp}!PEVDf
za;S8x@4w$B%KK2U{Yg%?4;%X*j49!`=(y>`xnFzN6t4LV@4DJ6Hg@ohE@1kyfEFMq
zL!T+uUdm<FlYRb%#-#o(fczs~zZRRg`29gM^xo_LFMrlE(l0W>Ju`dypBW3!ti45l
zIl=kDq-=B(=G@E7V#xtcwIGfXcPx`Sw~Fs@&BD!CiwnK@#k~SmGg2)+IfyE^_k{pF
zT{i<w7k1K@Rze*r7|LFJP;}%!?LqqP+SuZ6O!Yy0dT_a$58fGRwJaa=T!pB*ecMX(
zXRsje!<O?!z4YB_TRXHv^W&ZH`1i#XtcTk(TtsUL%PyXt@485?hqnmc_fN)=KUY^?
zU>&o3veFdANYmW?ta+Zfsj0^ya@H(&A&+FdJuvk9i(Mv>V#16XTCUs5L8#XlW6m9<
z9w~7(*6a61N^v#jXdb1qxjLy@W~PjW#Uy66yx&8q3)Rzlpn<O8WL3N^B2_5PlijD^
zl;mAQ8nuZgzH#gg<#+jC!*%?89okpNPq~}|wdf!ar2HinD?KNYr8ur=Q-W;Rx$9EP
zwPUe`HLeb%W?41&oZ+jsR!TMpUvBzFYHl{^NU<@(B#?T{D8IzCqwu3nW~q=DX<G(|
ze`sn)c*)Op2zb9ar5Q;4F1TL<TAPX^2|N`nfnd3N8WG&vxL_Fb<y0|ozW+DV5`a2_
z`7L1bs503(AC(F^C|$EgLYfh36~Zw7`I#cFr|3WBZ~knCq2>cx)~-qX*u#eRUa5Vw
zZDuy|l$TTr4mXO5dMR#iL1Ey%VHKIOr5DN5HZ;VJRB^`=;>;I8BlZE4SU$tD1*A{M
zjxx8v;pD&wj&w2ET^7SOrf{L_gjcx_|2AG=bPOR+DQuD_DAFVI%cN#j3^yD7jlyCr
z(nAa{H~Nf_xN~c<bHuTDF|pztmI=zuw^&xlFI{lWyE+DW6oC;qaA%5*r1?9xO7fcS
zp>Q&dv`HSoXZ%!H{>*T~)6EPx5n?~=B6&2C>qj*u(S#c^TTaN%4|}`2piGtj;R|D>
zX8VT%lNJ!>dE%%bP6C4dCu#I<oBWOK#SGW~_n5`5Cy9clCxO5>MQnE3BbguPVWhzA
zwa!O;rM7m*{o;~`uFmjlphB0LN8&5?cbbiLTfEJiI$c@U{S_ZC*o5;QK5=7DSKJ4F
zyNW@?CymV}VOev9!l7o5qs^Tlt;cL4&d=pzaGzBV4w602pD#Q2OIG+}1vhB8Wxzi2
zji{RHLM|q|0USuijFHLzDInGv9{WkE-a@E4RaN1lV%0b={r}J>oF30!c%jYc&NA&H
z*Jg|e0;M&RS^V@w(FGdajYLQINrqc)#-~_Bc?xA36U#AEj_@fIp`0(xf%p7XXh5^7
z-C%Lz4Lo}J@N?er=@6H9v5365&}l;e#K7;(Q&DrOMd!C>tv&>%D-9;iS}_{s*V8Ox
zXSluQ9$mE@IG?C~L@<I<o!oPpM;mBUM|)r*FCR`cLU92Ds?ToMQKQUQ^6=R$?3|GF
zQ%I`)9b7Q+*2dN6$6dwo&lU#whAxYzlAKp0o7S?>&eo=oHacb2SguR?1NS+Zc<|X}
zm(zl>pfvRD8iCzxMXyEMt6*{7^^Hyvt2^{t3fA=1K`>ZXRPZwiOX*l0I!n03cd$nW
zs!Df!CO=I-+DSK4iY0<^b&jt4uNApG&EWUtH5;e#7L(2Lv>?$5{qkQ|A&~imU~Fw|
z^YHW>FGb5)d|Q~fTIwFwAHEhQ>FKtE0<1Z{F|vcr-~k`T!QHZV3K6D33mY$D?VbI3
z<?>;eqj`8I?Z1tO1zYXLZfQk<!NWq~HQA{G&bIX@Rze2+e-aM$xyW&opChBXddNcu
zza!Y@<b!RsdOgl?xlkV;L6`j+Z4NYu8ZEaXX)AlT+QFb0kZcwr1p?RIVk4R8xi@s+
zhQLH8geC?vv0f{>2IiJ|qIGx$Xt!t{V%9;!dPX^Mlz1n<+;hg6$aww2Ih5+NpWaN_
zJfoIDyD?{}s;35BgEbpFj)rvh@Gy%C_ux*jQxfK8kY)Z_1_+|Oe`H^>hML;A^JRS0
z(|tP8dq~|-UP~a$9WMs3Ut0W_Eb~#lbl3??S7y|#7I%+DuHdzeu|Xr+j*A9_JI?;)
z`p5V&{|L#j_<|LJP^t8hUR8J;T@FL}XFtzBr7LRlD)rxStCm;PXMXzi=GjaIX%K0=
z3?(1~h;dPetfbujQgpq1z5ea2S&P3{K)`;&p3H`j`JI=w_osx-1I*NVk(Fwxh@!4b
z#KYuRL{Z;c>Cii|Zez%wH;}S-hMsi=r0w$)6f2xl5j|4{^B(ia%pcDiVYPWebOzFg
zN_W=w6*bd4cLj#ePL}QHtOj>KNN1q#`R*k2j)vLwH`0uYwv-HZ_h5!84mST$e@QW3
zW_6}-@{Vm$-wU1`-YM#I6_tUU_Ok}T2@D<w5fIKeJ<7Nf%W>G96qYx_p+iruYS+9r
zXY}0D{8^jQ3ZqlM_<OEr<hclK4d-scS*i6|QQ{ces-t4TRC5xJvY04bZYm>vyoLT5
zE}51k>4jh-uDUA*F3?M_7D3U(V=in<LYH}tCK6!^5Y*ATKs@*5yfORJ%gytn#p|)O
z;1XKP`YQ&h+loSm@<I(WH-}MyzN_K>*N=4M)cANmW+OkJY$$76bDLfc3D_((SvKvs
z9W*TQ$t7>V--mHgd%OUX>iLdR-SAvcd+tm~V^@zgPfz`n<E(yRIM!xsVVS**`cn+=
zpETiez))G*BUV#lowOvQEkc&USw<e!2D1nDwKp12DX@^@o~%Zrdt(Yd=BvNCkMI}<
zy{?AlDU639`%{8q?5uL_pNDkrVj7LBW#Jbzih9zAkP$;Q%2g$YTRp6~k<^^X<(t@p
zPkGx1<n$t{-A7B!KD%ov>t1!x;3?Z`uo$xQH-YQVeIt(2>go-!I^0KvB#>?-jX7^(
zVMO2|`3pgKDs3v>rZ-uOQ|JU?6c3zX3rL3w*&%HLN%^T?S}o1q8C;n2<j@=|uFbLD
z(hftXsL%2hlJc=KnyWwk6xU<87{7v0D_60JkFwG&qaKr2kn+C_l$;Rh>@7YwgW<?K
zbQxk7XRVx=X~LC~e7tov7mt37RV|&tzD*AQ_EOC^ZzBsZ0SnOS>6dG(@09%3tI}7q
zTBiIr1UJTVaTv_LrJGal{{Bs6iScAaU0L2|?Uk%Z_8FxiEF(9j_TAE)JkPMJ$LQgI
zD~Hu&$%~)L0ovxg^c{aRTdHXvcJL{+w&}a1oGW!SFUro7jcOU><(|$$os^*}zIa%V
ztiDE;%HRpQ!&|HP{Za>O5mHe|U+JsKLB`v5@54a~Z8Jc}ZJn>KIHN~EN%~J$KP571
zPQ~h*r#PEuoqCvJWXOYXMOjFc(`Eh1zLF{0O||Zx2sNP>4gpe(W<NGKSOV`DEx`<;
zpOfiYQYb-*NL??G@;A%U_Qm$-@W5Qv25|U5;m6)Y?C3yrew4G!Q5-1n@VXkhX1McY
zXyyT_;PFZLJs1(H9kloTRi~FkeU2TYVTzn#kN&IqHukX*%`v^J``jL+L(OVcS+=$Q
z^XZpUMbAe{NOD}Mxg)(Tu)D1V>a2;f;;V2|n?jwDDwrPL0$OhIS@5L1yb@Y-w@7L<
z)|@A>Tt<|?vIFzOw_bd_qix~#g4yrS=1||V$&g5#sV52f{6q3*U^YF_?Co`~Pc9`5
zgkI|1iW$6l5LZ}dbLr>SRPmBNzrcQ9&B%?8kW;o2U`QM%g~`$X=?1|HHYl9`td{%i
zS)1JY_2rAqNvqDjeXrLGT>Rr45rzD5dOs~fRJU!`epD6?PWhxuvK3aPZn5g;InE_W
z8-%oA)$_F}9J`LH1+Mo2P4P(j<mu|CQ2@dXQeKqFu_uiHNTw7(J@ciFdtH=W7lI-S
zLM-?P$>wJX8zj2dwfdd3;RA?QXGD=WXCbn7;}{Ts>b_f&N=`0`Vq|wzo!1_2H76nm
z{mp@bN$2aEi?>T3_IQ7eFP|%G?wwVsz6zWqOCrO?)65j%X<4+O+@5=mRfM4of2CU+
zDV!3u-GD$3*T+FK<kvWOajo9lp97~i4O+Hdh~fXOm3S^ApUy<TYHiVpD5AS=awnf)
z;1)%0+MUY%hiuV`x4yN1b3csl9<q$`?*V`Iq1skQ<Biq_^fRqX8@;pEA=U?hRtLQn
zlS;}mPbocm*yf+@*tR&WDjy~TIVWX)LBV*3@LM7&JiakY21n}#x67n0B3#V@mCFLJ
z`Zmr`Bp>5W;`B=`8+okY;+&BJY~&A9<l)_X<HUpC`$puSqXio~hkMJyQziDd3sr!N
z-Sey}$DMR~4CiJNO~ayUrxRoc($SF;{)l{Aqq5ZFg^Kf2ck^&t0U0Oo7T?<Nu+fq+
zrL7E-#S{3efCRTxYd7d^bMU56g@03u2{&#GMY$bnr6=AwVjcL@OA88%@9CN3(d@<G
zB33M94-35zJdEE^x17Z$L~rew*~dm)yM4`*3zF@73HBgYkCM<#;a~ws!NKRq^;d9s
zOMu;|o)KZExm!(uEI{IUiK`=&YdfHu9ilt666fn5HK~YhhgBb1n@iN+_FK6LoAufv
zwf?-nzC$8(_1FZ5s7jB^5ZIKL>u5cu;sWGF-j&U*Ev2WE#SKnC?oyIkUUXnBx2Ptt
zDT4=hp=6auWhic~+C^Lqv8q47Hz)cqkUvY;cEDZx_j1=FOZC5|_S#RIZH2krB1_Q~
z=rU7AyHOs9PC<s&IEHY~;wMigZ8|OdJHExFDS2){?9d;UkLexL-|(j(9(~2Pd(yqW
zYf8yHU^+Q@^~HOj;Z;dy=bME*N9ss%HXQC!`B5dKcc}@dzaHB)!dd}t)d@slaBkBC
zn4Bf`jhzC>VV}$Sv9O_S4O3q@gd}ugluw7CG2v~!_(roB>pWQ9q?G+9{RT74c}Q(0
zNq8?oa3`Yk;FpruWLP^`INz5<2XQDoWnR^~??tc))Fr&O_|R$6NIR~CLBk1rcrM@8
zm!ItX1-P2)Wh;>5tLUr~w&RaC+}uaF2q9DM<A<w4VN)-7C@WPP7r+=u2xjN&eHp3F
zg00uMT-0p1f)xp^boH-pCC)c(Z=p4oYdju~`LEVHkjFAt?oO`~bT(E%B8hjbI0soj
zoIPf_YML#v#3lD(A)VF|;BR;+T7|Bq987IX!0}?2&rz{sX+#gbcJ0xB_^(nX@HmVa
zi;~F??4UN2seaeT?w%n7NEPfszR2Q6^Hp2C$rq@1@>X1Iqf_+Oq1LPLht8LFvEcN=
zBA|PFd+Me=#GT=d8~e8D+7-YOqU?r9r$4SA=8M?Q8;dMXTr!ohEPr!J@bqKJKDMNW
z5d*YCaC6_@0yIp6>qNwi&xvcApR1!yih|RR%RK0>M%#M+`yt9<)q-iN?)6VMAdj(T
zFB&u#>#8#+vmd4Yxsq>oLx1?omw30DkSe3I8nwOT*iXdSb$=`B%Gt;%3N7FUBVAF`
z7B_5*Dg=rswBW(v-3jq}*oJqAF#0Pzniw%*8Wj2R)vLBn$bX<gJ6;eEJK!Gqu{KvA
zPhn@mV(5cp+~(^Crt$>Uuw$rezEu(DEq6YgL9OT$-cEugKEt@-46kcFnIG@hf7Z?j
z>LMW-Gks9H+kZ2<iOx{W=kj<;Ge<y2(J72B!tfNGaHW<`+Joq$$>7m3l1$T@0b-e7
zws{GQ=P9<P=V<*X?XdLX3?lCEKn}WBQ2fq$octjKw$10Y|5`A#!Z$MTdc>4lgH&6P
z+*madU)z5c2vP;QV_!h`fPCDUD_Iak?3yHg9Q`j2)8=OkOVXbV@xyFe4yJUt=Lnya
z7Zad^e)Bm=7xImd;rC*(m(KwUbv++>S4w~_4-}D9TM=|K5KMviiujGj;GW{jgxvV+
zbf(eGuZJt*_3&wCUQe*xVQx7x0)}bmu&{RO^q%8b9<Jh&HcIJifU;SrAZRp1Hn^M9
zOB*;6<XRs})T>|m;dlP%5YhIT!Iki;ZGo(hh-_T;!B+l>Xhd0w^8R_9b;{X(wxRcR
z9ax^z4rn6a%{G^(gt2Y^TD}Z~@Aa>)Bq>^l6-tN{l3ZuU^b-2@LW+OVZqkd8YzA(R
zGvfuX^N?MNcwc@1^kK#uI-ducK0l+$d)w6+XHOuH>f_7B1jTNq^Cv5~(>?w(AdpMo
ziow~8mS6;9-F&a`7l)V!kR*l_CY455E6Cpim2DhhUqeob^l-69#%^}2^In;-1ykU%
zz@~Zg{+>>*yKTh4=z~-h^78UtKr>|nZCk~_3E6vrk^l`n-`i1reL%9Eyly`b7yK?O
z?i%5o;Q8}$>?wMxsOX}~eaTZ^=o)1GxZe3MyS}=2X+)=7JQt|;X1&HLWD04nm+|kW
zW~eo~r{`=%G)3<0KNj_pz|%sv{^S7kqUJ7i`g-h{`XUWsr06{b9g&Rlt=vndX3jO6
zQ%b^>7*zd<jD?nw>dOm}u>&QJNg<j78avfcR51s{#v?ZUQ)xh^eC$O5>)UsBRlgBY
zQ%y%{TXlJ7&jsxng3Idu9R>orx>qyCuC2vDinDuPM$e<VBcJ=d#><OzeL8v=-X5E9
zPgUu!QlF^SJV*Vx7)o(On}1{e`AtJ*U_z=XVu)%64H?iLZL56)UNA=Pmc2nBoUVyB
z3|i{bPY?M$QAHo*L6e7E%v<m!EiqF?UguXP{Q1Q~9BKw74)LXI-?bVCAvPP&*gU=b
z_X`<sJcdTJHzn9kia#8Y@nA<gsW3_Tgl>;5-wpl!BOCSL;yjgOsc%T)-dxBjIt<&O
zi=AC_-J|=khc)WaOx3---04PqLJhtnQ(?K;y_%9S^y}vYyNsLZWkyzKk!R3=1fR1_
zD<~)kvh58Tq{o8MSFl;i-r)Pu?4)B!<zfr{aEAREJ)Pi9@wy6*Qo7E{*H-*b2R0PU
z>fvMjfgEP^RIbO#>h1_uG|{HmIt#Ljc2^g8K~OVS<U7$3PQC)J$&wj8gEk31L@Ov^
z`SV10|72v@4F_&2VWJ7JoR2-&N5@xFyaWCjCu0C$HKPq8*4{}mOX!!APj7G@)LGK4
zf4oM<!vrj8JH3$_LSLtSV{S42z1py0rmgoRjAicyB`dL9GKR-LX~lel?lGHDqCn=#
zXl4>>ah)FiGlnCfuu>(|UQuc2K=Cckj4m74Zko)!V6||Adj6ZTee?xg|L8w3KN^oG
zU-6p>lCi(lqRuFeLnS<}Fdy9=l!`ehm2Bc<3({Krjx?pA0W;MPKKT2AN0Y&`_{9e*
zw6}?N#l3VHdFw-LV`nT#!q;}B7%e=+1O)-7f01sjxVzpC!MNv)H)|<oJCU4{;vNP)
zxJ@Z8-0FQ%r-7S{qw#@Iwm!#X>c`%Qj$+09-R89o4SAL?kACeuK;53ZTtf~?0Q>%+
zDkyhl;4OS}_1TlGKYLd8Qc5vd_xg=T$Fys#AN1Tn<SBW|%kOTj)Uvvr6U$iQ+4?p0
zs8c3MA47NVcvqw;4xH0|IzG$H<B@GIj;l@(5>N3bn|FF$lqphk$(g&%d%Dm5^sZUB
zpGCmW%QCUez(T2JGX`WW0*o2Q0T-%tcPPsjU#Y!~?R{L193EhZS+CCQWf#5+$fL&%
zypUJ?A?ezO$xHHd2q9Sexb4K3kW(2W{duq(styQNJe8#W!*_af>mP_xza`~Zz_4Vz
z+$Dz!@-Ppu>NOp=Z^`HNl!y7d+-|5mxv@}mK?E)?u(gIo-lrTB+t)yz7;nBSNYxWR
zeJ0tUARxQOeWLF+j|rW6Zez5+8y7xXheHjpVTxFis}JYnSRyLitc__RXO&N+rAfP*
zk+_Q!*U+r9Q@$v0`my3z)xl}|+fAcJ>^m+|r)|Y1h4oXtX4{V&4f6o|aA0RgU3mFl
zFj!@<{~~hL8Yf;SyHYJtFsDyv1Fo3W|2*6*^f$gS2I?!m=WowU^%r$qUD`L}$k0Mm
z9#3YL=a`N<U~HoNTu407p(ILJ5JX>!()C5{*6e{xbd-6q@z3KZx>;C%E5@Uf$fIe+
z@Dww)=NhtphDO8ZAPicytZm%tp{#BUqMDD%mfV5lz;j@&4Hm>Ey^A5W(_CS!ZT$+H
zr*tkr-5F?R79gZCbzw2lxJ7+waR^u>*j7__j&CSIe;6gE7vP<Z?&e}pgm*7z71E6<
z&0oG~b$UFYx3!W0V!Y`e$<buL)PXq41%a1|L{&t7@MhTR$PEx7xkOgN8U~FQ!#UO%
zoQ)z<{ke~y!Nko9M-KAg(BF!We}#?=9Ouz+n>$x>XR@KXS?*cUa1(v~bj(ZPtFaZI
zS<1WPHv3Ch^SQWCLM%3s2OB=Kk^Q<MM#@*!)$w~X$GL1Vz%l3E^78%&hIP%DzAwr(
z37P7*#J&Wgl$7leq-Nlt@aZYmWpMsiizx;sQzyPiy7L5YQ8@$57OMB=d@mTx4!YfZ
zg7RpmVVi|V_HZ5Ci^kJAv4ksd=0KOie>=>kR+T*I`r_0o-u=0cy$edpqT4Vcf2G0s
z)Kl<#ZhK2)j}1tdkIurTU2Z_=_nGa~BHc79UBz8u@{Q3U$6Mk8V~=-nsQ#YA^@x_R
z_8_ZKJ>;=|v<?%)c=~j}d7uhsT~GW6;?}cbOSJ)^&;$;M>|9;ziF`X{5X?|I(?QJ;
ziRWZWKUMvT@AXX2xfUlqUFEtN=fb{dnE9FMnkOC9PfLFOyra(-pmbO{a#L^u*r>pL
zH8I%3<<<r2A&#>9VvL`qK!X3Hmkn&659(PeJv$YGHR)EFaOT5jLoaT1P_pCwrgdS@
zb`Ch%@&Ufw{?-*mU0&Yecx+Wh*XII>u~-KggTIdXu@-M|ba@X*@2H(@9svTF=@-B_
z4d{z%4fgMPs%q{=iN$li*}V6(fJbe~=N$|gpu8XTGt;|9e*d}cwKQqB0|r+aY^5rH
zrRyh*W^6#{!46BW%;UKEFZvI`=yv94U1E=9wDafYuGvy*ROOVL-ZV{?8t>sAP#AMU
z)Q6C+l9?o9pgJV0Jj;1XvE)~6-V8lC_A{KkUx9_Yxq*vtk>ELZ@zv?RsG4!T8aQ@o
zyu!mG*OR>=0`ME3*!Dp#0MHW-g{pw0=Ui_DKHFa<^=@O}^5p(|3(xj)s|DIgx?;Ui
zx%Am-xC$g{iM&GbsTE91My_4que5P2nQ#e^@$G;FA4Xuaq?w_9v|4un0nZL(9)1fR
zYDw43!_d#wj2D&$*w>#ZKF91UT-u;9>5F~i@Q+XH@;tdBAkG(E@=TI4#s#@Ig=@Zl
z7;nmki5cRMj2V+olEVq^n=7{n9=dX2-kqR;Gz9$3!Pg?uP1V|{RXp<bs9yaw2Q0bm
z-YG41%(qbj5L~qe<EIxO{b9oRgBHXSU2)ty$Fn>`zYl{?(21LK1$Ies2+cK$>=Q$$
zTy#(>yX!g>zo_nBE1@glqay#nC04Zv?C9CG1!d?GA>gLaPJb#TVz~mjG?6Q*PB-Ca
z&cZ(G)+~wMfaI4?i)wY2TCif}9Lvg-5P2kvJFocl2X5Tit<<jt-m9qoirR+)EEVwA
zn*?*@NRFN?s4l`#>HPDFs2VI|Gvt|U!?qo|DIN~QdaZ58Glzjr+CYZ7VDtps2RQ=n
zRQyb^kjb9*KhOFDRm4dHHvNjXZ0YRgJ|adeF);Jvq?SCWksJ%1ha3u)b#c}vcz1>i
zcFp;A<_SCX0ClXj1-W45tPMK?INUn-_er|LC;W!?yP^rqVC#XeZ!O!We8X6|rE*%_
z8S+%j&c#Lk(RGtW3V)EP+zR}Wq2fn}&mNQQi8GaT0jXToMXCEhEX7}+3e2rv<4U|k
zFLJ<hyqTgEJ{Wu4@ed~t4Jo<OJ$~8FGm}R=QTLNj3CFfX1)Iv@xCd@23uyBAU9zGL
zdIy2Ll2y7)0O45srS&Ckzp<lx{K#FPvCBPR9CIP2nQ_Y;yFUB?+S<&h+v0SovO5~4
zJM03zI<(t$9eqHYwFV*mu25x9I#t=;OU;pE3WL7Ql2IEFE_tYLy2rR<z1ZN55U`2Q
ze`$IE^<)j@VyZ_5n%^F~JJ=H>vE(5N!fk{qI|iBrs}CChM5UGjNl#~&sR<4UklwE~
zJj$(=b}z#84WBNmvKhJYFds<&6s3?6ie|wr?*}l4#>dt_QGstd;>Gza_vuU1WwlXW
zw>ucyReiUH6w!map_`XCDi`^Clgs|=e1#swR>-8>-Vz8VaHppqykZ@hHehX~DtaJ_
zZNg{I*3K*y@sgi~Ssx*s=QC1@J!dfjWV2)IiRBPUZx;xXT%wyG&vZmmu!(-#mfr}6
zSWMM6?wL)V^TkxB1QrIZc1PJVj5s!RREp7y4XSgUzwG&u*GM<zqUR{GUkU1h@1Sk=
zp0uh6+kn<_R=AHBD&pCZTlbR!Ch4jD+WfP<W!3~@LWwp0>_`=5eRoe2RpzUN-tzpk
zbvAZIvMjbaB!`=9(7GfM*xpKvlvO}`LrOx3AG^Efol?O)9E4*3On?7*yVrcy>@5`$
z6Nr1g!QlR&F?>BH_($|%{Rf$G*Kme+ooZK61DwqZS}`N$Xd=w{ZUi0}&DFrjVtF7#
zL`s$&V%FE|OvqfHD4JMZp<!X{8*_a7$SrS<<vhg!%=LS-rgq`m!1WO@X7!ww@1(6&
z2b^|$!%j+L=-J+{9LA!ncbNX|kAOrq0@qsjX6MD#U_?j5QTTK9&F_y7K5aETL^N_T
zQL{S$mZJLYvWwV+^7d_dpKX?BL-RB(3p8;ahYI)POu9;HrZX&DSo|nkmWq5IF20yw
z-CJ|1cX4j8(xT1b1MJu5N3JbP2!q?c*-D;6`*WV%%|ew=o<N~+7=!8E_g1bR#1PVH
zoEbuR6s|CyZ?~@(Br8#sjBNkRPl0ihAJ@6cfB)11>ALGy8A9-8AGkrklP7}(_Xv`Q
zLS(?_@eaBW=GYL)cVpGmViqzKnZ}O`0jD;+Q}f~NG*3M9-2>mH0DcX#-Zl;D<Sg!f
z^w#cn+z_KFL_P*Oov?5P_QVCtcYON_Cz}AE#G&3lv~ZVbdXx)>S8=isk)zHb#fF)1
z(;N)Pn%P!;|K@?Pg`M+@%|6;{JI8<xt#)i_YJY@07}NZx%tm7{E%?c-y*dp(kSyoE
zAO+j+&h9hAXN6bPd69IHVXfXileGq4@o_P%bcVS;*b{>4){amDu(u*F2Lz;A^Ai&W
zpLsNmS*et{JDB18xAOM`ddyqS=LmKr1w>2=Nkws|`yC`~#H?0y4cOPa3IhU$2Xp?4
zgG@ZEsWOJJ;|dM&E(2zS)*x>ismxc+QtF$Dc|{dFJIuHaxA4g)jWByc@q}$X$SLA3
zLYt*EsPw#vtRH&H1{t8&EfT0x)7UKUt!?G*s%)GHAz{$8mB8LFUw#MtN+P|=H4J<S
z%4Uv;-b$oWclU(toD1N`GcYJvaL$D0^VCBhgT|+ivy5@5p+c+GQpwJ88Oa#?oRR(g
z{dY0`s3Ut~CtHYrW9`viE-&&N#|J6aXs=o~X>04kb}!GX>kT+B?>xdB@Km>xmTp|e
z|4tTOo)Af%Wybp6?ANa`l8{a|I6h-7Eg0w$q_Mi4wQ9h3w|s?~c7|H+UZ{F8(6j`8
zX~~=F*w^?vo`0prjyJ5cU9f|3N|50+?6i!={4R4`YTFWGuR~!hMr9pmwnPj;Z)92D
z7~Xv%*iCPHkkP#ModL)Eb5>j7LHX}5>87$U<~=GUG0LoHfsoO*5UjLkSbsu2u`;l*
z_(e*d-ii<QCHn}LvYb{5HFx$iRMK>lQeb1ii|~4uWtB?NTxHp#2!oo7`(FEp3Kqsq
zk1?&<wy&{v!$eeO{QB-XLar<zhv`{oNwU_k`r&7Cnhs8I3{)EB?_&w-X~qkjs9S+&
z;(7YWI+O$r9bRx~iN!&qB#lcHu4=mE@P91-!&j+~?Xavx98AP0=9+I7`y0$ked)I-
zpN`BpdU|0y)sk$#0Da|EEEXtab_WkX90lRJ`e3PPF4#jaeK>mjlP>MalWr^<-G_VW
z%oBy4YN-JN+tVQh$SMt5<@ivP<g?k?E0ym^#L5qf6+Z~Sf|V?!Wpk(8phxMmvz`1p
z<&Ty$i%)m}O)x!>D_yi9f}0I9Ku~FrHJ2R(4O54wn8I=&SDh6J+PF%7<z&XUaE_x=
zw^52Po3{NUScb9raw9xF^EvmL-lDB)(2A|Sbr@o3IvB?4_E5`5V~Ae+0sG*RIZ|bj
zk&m8E@0vx@=XUnW-*Y>kP1yYv+zWd7sVUOfZ$1CjsXuvkBIQrgRa+MmieK>f>Xhr3
z{M1|m?h{nWW~^Cz$Bmf^gFfIbPLs~NP<fbES28zK+^eb{TRTRI&l`Nn5NSEJ7!9-D
z4%C*33><R=w+dLwB#<y+whae~(~xPJGPGuIJ<V-P+TkFg*8Gj8rcL~m0uqRP9D&g_
z<rV62z)Gly`z`dx-b7Pn5a%r*{22sz$ko}mv32Km&CJZettBg5#^D|jO?|aY&pxy<
zpbiTE0N*9?eBW<^6mG=8g`d0<hC($BdD%>N&$3|cqje$MPEQ;;uw~H|HMm|@;k#NO
zhE5ElC)Fz*`fc=5da0+&Y)?hbubnE`4Y*q`O$9pEI@C@*pC?I|_Y4|BWSa`1{vh?U
z5HS^rJ)_<nXTcDZSfj!V%!2i=1Y3*>Rask&tHazSQf13&UZhXB0lyXJ2!9D*FPrCV
z+J@M*fXdhJCDE$~zNsodWO~jo4?Ov@6n>P^4QSY#(6tYvNz0;-C1b3Z&YY(dVWMFQ
zVL+9u7X3AN2op1V<^uWYb&-e{+-<BUo85-BRhDkhp2+9nPr8aNG44N;jCsBTjC5@b
zv0<l6%vjtSiX&sP83pL+;02Og7Qy_)G3igGo&|a<{Vq!(8nLmp6KBvFr_falYdRZ(
z^<wbU?%w4x$qtY2WJ$7=t%oz4RcA7fMeY!!xYi-_K_r$ucE=2Q9V;`A=y+m9GEM|e
zo4|jWcn$>L`_F%=8UH9hP6TgpAG#GD+Zb>d%OL_<=f^;FJ-7#G#{5Uhry;7{Jv|yO
zUnIpv4oUT1<?Q?YDx2}ZBwRmmR2#tv-U)FRC{i+&%FE^R;n?5cGVl9pn##&mKisOx
zTcCI3T><xO06*_}`MG**v>6K&gQm}Txa7<}8rP+#Lg#Wfc(S4BV0S7s&mW%GXyfEx
zvt(bk!-qK`Vp7taFDOrY*_N9B4z3W~Fw)WXzg6p{=7+ZvsY4J{bsn0BI=<cNhS#<f
zvi<7}qF3`s#O*WGhJu&n-PSmZ-ro*>b8}Yf2}>_5J+i}N8v@p`;rTPHQgq6|qg7`O
zGEC$GfXOe&+(Y*{A93{w=9*hU;`+C$5#I_`dEc0xvGUA%5AQ8aRrf|dePGBhW4DBQ
z4AmWQS+H|5O$7`&?;o|DSAtHFG!18FXY(pO#&QeUeg4(zKM;)}s@}Pi&WKtXs!8Xf
z_W?QK9vU{v9-@L3d>~n+nEY^J;>qNdr86cN+1{ZUnoIcN_LG1B>xmi3kF_{gc{?`(
z!BVa(<P?qzI7j~M-ry^8vi+g|DfJ)$)>cJmBfzapJkrC*SV=Ay)kjKPQF&2Xihm5<
zR(c{gL5-z_yQ|ZiX;&q0_c63nrt+&CakRZiOAC590Z33a&S&^rVg<b9ln$~sYQ6jo
zYiiP~(Q&<CnI2J%HP`OLg-)kd2xV%Yeb!H!b25Njm)XWzaTo|gROSH`EiQPqnr*+o
zeaAi6|5erWOK9msw(zK4-qUm$PPpG!e-l_(aQzvlsN<$7m&gC4rh24SPi7Iuxzjfw
z2yk@;<la8tn6h>bmwm}Vc~?<h>Yab!Fx4+^9vEGVX=$o4EbPN1Semgfg>JQYe<~=&
z4!6d0iIy~@R)HPhCb$OgA0i6&&+|$SbpWqFjouDIL)u;nhF%+=dhbyj9d&C)^&dk!
zIoMjjTgI;-pRLhu8Pv;T;gc^mu5}xy?a%OSjsX3-z@Ks@&&<&VtmEtk`)88OQ!GNq
z$<Z;FSTm9TXG3LmQunxs{x9{yzPuK=d>0+`sYWmQ0q%S*`Z^uX-%7bARHmbp2uJsJ
zjhLqy;M9}NCjaZpY&^<b5h_)9wE_7OlF>(K=n+>t&m7tkh$IDgCDgO^^co=iCni^2
z$vsKMe?BePF$W4FhT>mWs4oro2b?5&z?Ve?$0BxNqwVQvlN1G9$JNFg6fM0c#b95t
zn!(>Dz4UA8U;^7pOl_tK7x7-dfL&@`_j=ky8&t@&Qjfk^GX1i!+GpxgR_%_!g^#in
z8v*cML{$wz^3r`uG1&|M8>pk5m;_J$qBhj%I%AV)Sy3_TDw(thk+XCetgZxNh2fLl
zh4E1{^G^<+6PbinGe6N1Ow4urPL%6WXssS5l=$nOZhW8%BxrM$x?aZqftWM>$Nf|B
zq$#1qM0-f`@GkPXm1qYQ{$>_@*Hkb?`9fHGdR|&oI`|0aerO2Xd}wyqU1@!r5f9wS
zf?2Mz>1TsGWi}KLcx@Qp48_}5$_2(E#Y=Vv%6=~Sm?e-26}fPHtJa~Cu=M=$K5#Ay
z{1h9!%LC6qD&a4r0jq?f&rXSV2;tF_JdkG3h`knc^jXKMfX%w&4=uFzXU%M)oPAZc
z!gU<Q=fO}&Az%cK=0d>X3YC+4gH0@tCH6$2j#y9)SLqU*Grc7D+iD6^k1~3Oj1(Th
zBv(q^$461y<A5MWE2svlw*WMi$78%O?rWzVU5eCx_Z17y!&d7g^GBIH32N`uda=kG
z)d5&E);K5i_q&S8hkRT_^|*v0kV;o&B)aI4&2{owEIoupP2g|vsN{1l$lkQ0Bb@g-
z>wDA3836-2F>Dwl`p+*VrLXTsxW|0^eL%B!yB2vDw-UJ4#2Nsa#Zk};!Y{Gtd#FbJ
zAC9hqA*ybR(mBA;15!hWv~(lgp@4J=AKfJlLpMlCN{DoK2}7qK-Jx_!_q)D-F!#(k
zJJw!%t?11veLX#Yw*9Yj)X#v7ujrrMl5gL5@FHfF^qDHJ0D@1jRDZ?Yio`2X@Z8*6
zGnO_D!0lC%Y$t)bsZ0E@t?4;b_^9S^F21z`USIV$daS~3UUd5T`TPV;@nVct*801{
zm44m_AoN;Q=>8}x#c~U<>iv6ECWmpftxoQA%2#7oaOqJK6X@HWRyE#r98~r=UZc@I
z%<tGe*Y?%7JJjW^IxXZR$UUNqZAoEJ=c_II@n#L6=u!UVHP9XaBw4LQQ)8R!DzDDN
z(vrM@r8Ec26(w$7>SWuu5iIWSXB}t1%hosB%uRmJ_ntf<^xVATb;weF1SC_=eZ_$m
z?Kse0FudnIFlkPw<lM{;+pIUhO^56I1K@9jLx<X7%bXV~TwHr?7j#&^0WiB4fB!BT
z0RI-w7_tU&UVT79&@0}aR9zgPj9}y1_7XkuWAhEzJ-coBK)Ne$HMg%*I*Wny*RQ1U
zZL`;>KYSr1?sD&!UN@8ezlcC&Yi7ons4j2b8BZ3ekpRrpDY`4l1AX(11+3RSrDe->
z*jD5^a^5IEp3QvzS^|)Uf=+*OPV5@<Rg`g|_6b)jaW~?#vkq(lzBLq~rZ2+icvNF=
zYZ7N|`TR5YRW5qd3M+J`$H-2w!Ybc`!nN=3a#;Xxh$R&^8XdMH2kY^{i#EsQwAOhC
zsmii-x<Om$BV~i`43#kK$0SLxGcj9;$!Hj?cU7Rf<{s4<-jvoo@XzKF4pi4MhmQbC
zL^$u}5u_o25e5ewdh~uSQHN~2hQWc&im+bc|Ar*kzOVuN=pSR=8>~u<7}hV2F;=P|
zU!;39YkNj4_W;pMrvPc?sLi=;Y9y_nvT=^jN9||G#CN5vKc42Dwwypo;zP7IHepNI
z!f(%{hSrl=NzFh-qn|}x-TV2Aw!7Q415-o!18SOeQ!@J74G-XaBFjR6-NxmGsX$<P
z|G&X*hS^?jqB*kn@stt)hP!Un<bX6?G;cV3u6*KB-qAXAlW^A7uEV&``P{DaeOV=f
zKqa%|H)rqCpq2x~TP25O9;dy6JAx$z^gA-A?1{AYHAWH=l3Xl+v}jhjtPkJL#jo5g
z1E{v)^)=UdfA#$N7lz?37S@>KCAy}-n6^&L9^m7$&#Sd6vi<6n_R`N3Y79LtI(Qe9
zu=tIVuvpjXH^14c`{jNFXfL*r`k>ahk@S$*`BNDvH?x)HF^69hFLb_jMd?e=fA`D2
zE37Q-zvh}6#R3Wr>S9e>09Wwl57W=QH)Uq$|0X+tYU$s*-rBhDtb=|N&I=q^h(xae
zx6K*pjWbJN+vl(TvpspU7H~?k?Mpmf+7(zQwFm1u-#hu+_GVq<^+qQI+lvF~|H}cK
zCDoDF(3}{K(>){;_v(LGgVtRJDbH&IOp~hEc5n%S0MN<^-+!8O@VO-$7`REDQ)>ef
z9@^ozs<EGA#Aqsd^Qjb<PJjwQI&u*UvIfswIe#V(LQFsZ)CHlLe7`C;Xkrd<UPb@7
z=*=Kve0fv78v5~aDL04DJ<0fHPsmOpbDziQW&dl9=RriybEdw6-EQ{|hz&!7rh>nq
zj9drx13nh#y;gu6pm!5x#il7IsKT~il~3t%x>TQK+EPBEUmvc&UUL3;F)8pYn}340
zRQ3=GII{9-cR{GZ#TPqbG-;L;!D^4#xCkR}xHMAf_6B<+`p0$BmQ-%wCnra+GP&}O
zPd_Cb|18?A&g>sJnfjY{+M>^~ZP15Vy=ppc1g6u{%#Wz$eLc)}dbzrmbH}uJsh%4)
z#Iu9v2zpR}ku?;Pb9Ww676b&+-NtQ+>@A=1G!Q`0&;P*|?vHj3uvT0fsjp0Xbe8l>
zV?Cz_1@FJTnsN1+lw|^S%1$mjN<W>jmX1nw-6h-D^#IL(bHjTPo08UZ{ZW7)V7l3`
z$E8*IvUCsxJ&(tiO?BQW1h!=v%1@ay8(4=fvXmBus%<x4FI%t9=HzHxT+CwL13Kg%
zKmNous&_f8s)hu)0+$UyWAnRF1sq@AfneL~Fe4{}1jpeECLDw>z%7x~`TVIEIAFdb
zk(Yl)ZERt{?nN1x27B{ILC%Y85V(Adh!xn@`3ac(=;&~F6vINTcfLbT(y-@EMfMvV
zCmnkqtJqi|6X9z=wb=H-AoV=A38Bn3(OVDU7J0i?>2;%aPrjWIKqc9TE_yN;f<;P)
zzP0_gV=9JO0N^G{-qlsGiX=6<J!99l6~14>Ul8!k^Dyz!pR;kaejgI>{?E+Y<T;0x
z=VvU~BaaQjNdKo<juZd_yMN`neVq`|jhcs%?nnbZY4BSq`=(>C%sYR9^*k~EDmOG&
z257aJZgvfJMDBnA-T<&WF(;~siZbj6DA5Y20kf*}2B7B*!`5yh`1lS~&d%l*-o6E}
z`1vl=qMH9^d#si-xC?o|Jw%@!o0zaA>A>9?c9%@>?yp%=C^x^oJ>UK?r6lYb`*Sx)
zSkb!~V`?`X_chjXhv3ghKyQ@UAn=8$LYM=anij9=osb_75mocaq1A$M+;v4kV$p2P
zpIAzPu$MK*_cdwOzYtgZKfUJ%5Gvg@o~4S8DSExH_A6zKlk<+R_Jwf+5cDOvR-pZA
z{yF0*l+P|ALPBK7=I>){UO76wOZF+B0uaux&K=J{=8~vO^?s{Jhx?uU1PfRZvNONL
z-NDwC1k8SQT{Omlx<2>QPNp@mEA*4jjD6BQKVi<D^09CVvCX(X?5W6+0N4t*#$4S&
zCsK6H2Z}wFtF*zqrUMAx=9zz!%LC46Tk~YXVV3XDGJLEooH2F%W)NtF>bdlNPL3&-
z0g%s~7R|bmo^eFFOdC7M(O+9G{{E7yitet>H($<V4@lO1Tm>rpy{><ZsS<P>vN#4n
zUe92(E_Pn@XGp-?Wfme1&z(V?sU_RN?y_zof(RO#lSjZpwh1fpz0kLxKLFvZPcwjp
zu*-F&lmDS&81^WC%c9a(A^0nFoSb3}@2$TU|7sudGIgovruy$yr)|2*|Mhf5^FOMK
zp3B(Zy}Z=ia7sEF-ed|qI+;5DcJ6~RME`b{3Bh6*{pP`Ytgd7~T-c#j=Te~%tsL!S
zkaq2ROyBOqUYfnsH!+ty)@3)(=%k}ne5BfuMcnvtu=ag?X`hSZ!-vA5JKAd1n1Zyr
z*Hg6KuUy}Bm<p0k5I^PaUfzX@Z6s14tj3|8J(hiHn`!}}2?|;}=U#{gR&x7Y?rBQw
z?u0f5a8bamb_=tAG|n3v$k-qXhL7KxPFjExQ5B~>RlW+VU_f0qeA-nK9$XahgzCj?
z81FaM$ls(0Pby4}zQW&pgxwq+1>JeN)Q?SwK~$oPEF)_OFZgL2#U+XKre0t^Lbfro
zacPJb0hDh}ck@wVz}IP=>7Cd7v^N>p$jOX;5~?wb7&aW!iza`@cAr1Yn}}K5Uv^s2
zvYND|JrP6qd)ltF>$pOT2qobCJoLt2)>GF$Ye3WxRQ66+T7NBf`M2kkV$R3Cwsv*~
zH!{r%*-uJIW@Kd4>>7efZ_eO&qyc|)15l~kiYudAd1wb~i1nG(=8$CwwIPn6lYkpo
zqA;Os>>ItL6x&u{dF%qJspkq!0T&dBh=|OxAK5vW9o0HxY;nnKp6P8K%)F()X2umD
z>uNR-l@Q|&iQzSEdV^^NlbO7olH$dT$NbaKc$J+mXE^2i)A$Sft8Jx2b!{1P#vOb-
zagiHmtQ=ls$M&;M6}Bt)K*DLPqR#GK+V|JGup;cD{;`RGUze|trKhFl+p@va(UvW@
z`S@SCvccdmJ~i@J*}WN$5e9mNMJ}=AaDtXTQ4IJ_orJLP7b#VxUGnQ0KEu}}c-B0b
z#r`fqEx`&indA~B72JjbU{%oW*#bS`#;*wkJL&nn%>V9LyKfu{<>HN4h|TYJ{q5$f
zt^ikYKjMhg)>a`r{}$tLgN0aRemU}9t^uT(uY(Z$Cai7nHiqmbBRw7G@NKP8vQ?YF
z22+^X^yAp%X`=nISnOswX3G~Z<<4O2YG1gT+hs<g{=$3R8v$q~!bACY(?D5$WSN%2
z-wOI85nwHKKM5vSV)WPE&j=HLK1%t$kui_M4`T*d>tz5EnR?A!E9N^3Z`qal<m^7b
z36d0*kmy!t{Yn>dSLudRr__KM%>0Nbtf@WO^y(XgE(eW~HZqCddvxyIYb<0sPy#L<
zUOI~5<?0?B{vRt!5OeG~f_Y6_L~SIqXn-U*+&~Wl3hNsrV*LD+BW(*6yP{-@hw8g%
zC|os6UR+2j5Y%*+QC{m75lf0~zLqtUR){{!>WKG9kv0=VoH-}1!Y|d`lphkR@Ce6e
z0#31ZUccAM$tgiWD-t9J5vNbB36Qi!i{HFs9k+E00H1!Ej?5aA3q6k0rj5YvIX}x7
zT5xyAuTdQ`axeHsHK3p?>67Pd*<kvBV!y<cEu?Fd$Hh%J4!b-!umv6VQtpyB?Y>bW
z#{@65-}hiMn=sQ$8GXVAeinuzhT6ouw73W1#l;0)=H};ft68ZVN~mJ*a~a8kNzx9w
zn%Oxs79qjF#b0TRQiz-)lVl4=kzBBF?ZSX8u#VwPP904Sk4y;H)-&oK0mNI+B)jm#
zj*%xOCa4Y$4w4zP-&T&H=1(}Q?@l{nL&*y8P}XZ%KJYk@UHFZ!H`!+18Ck7o5)-s&
zuGhK91^BhRRk#l-&*Gj=NAySRGrmMmA#M4@No5B<A^en{HJU!9P(rI`s#VMk0-)BT
z-W+>=yTYQ?kX%&!=`SEe!_N)J+uV$hU~j3H(*c|T1wv^BEmK$VtUrAQ#olY$rCAvf
z*d4y`!ks-uXzITY9R@!1(}D>fb#C*rAEskP7(s5|jUN;X`0g|jMK&pF4p;+)gM$wg
z2R&>PrZ=_^Own<D)ld;IkVnkbcs#PmdU~S&sSR@8TwV?(f2dWYH8aQT<?5prd}hTD
zg^v--Vx{z;(cqfQQrfdm_&SNnn+X<!I<~WkBNWNNK~S`AAril)a_9<ELk_n|SghpA
zeEV%d(j6^H@yzqbQYNNZSHQw49g0zI0Aa!Vwk}ucorM%kfS6xiUQT~$3wuLbtm!Ff
zHyyE6#Dewjb5RIIZZZo=!(OL$#8R(9Z7jDII6h1T#@uOL*CZBi`iSgz(G}qW*OTH}
zcU;k6xloI_d~l{Ln>5T#TA|q6SPdhJ@Ol;pv#{t8!O#StW;IN3^@Y<ak>hfA>N}we
zknz*e+{%HP<}L6v7cg-Dxo;*XB%`IIe_zFvx}6mm=Qr$R?%3P@GAQh=WVx<EBd9*9
zT>dvHPel2gyV&n<dggs2!ZF@;9+-^0(wK+?Jt34n#D6d(;A@9OBy;>)_rbMhY4UEa
zuVY?xFF2Bt%Dkx2QIrHhRaJGdi;^##STMq4Q-f@cmI3qM6YTG+?cCXYE1f3y@CZbd
zfdp~$0C5IOmLlEXOhv?|f$uSosueC$a#lHr!W0{cV#beooEj{Hcj!2ap#Q=mbTB=-
zcMcOx6S7s)76!nZJEa;3hH;x(kKgA=CoFlM(IswsLhv!uIB@Bd+UP=RP5thB-M$ll
zx>dOhaTbdF7XIUMdsC9eQHlmD@OF0qRut2|&I|R|9mmASs1MU4iF}QQVt$G4OwYFc
zE9p^tuM+U2c^AB&E-&%4uTDyX5n%*xVND-he5S+>(2y7RlZZvf8bFQ3Uv744PDOtQ
z#!X;=fu#u$CfEg&hW>arKv(pSkHH4lSgk~NGck{_2Bf3vG^7W%>BrZH@}GW$N|U2V
zaN5A_?qyMl!VfX+L<L?Gg%e6MT|u!d3aAGG9jRvCM=<gR2^`>!Q<)UXLb2f7D~pRC
z>LObOwR}PIXDcf!7h0H_Gi2d{BGj&xJ(cKYR)SKsTBzoNkx)=0#2u$3p?zJ9gFS%*
zycpOa;3*{tx(`LB+a-;Oos_93uNxtU1L4FO+t<oTAo~|NTrgZP7F7ZqF*TSK(Cuy#
ziyrxTfC@+pAmY&f-`SiX!l<pY{vJDCmm?P&6s)@td;=O<4f|HX8}8)PM4}(hwk$Q>
zwBtCmMZC{+(>30x7lB*GbSmOdJsD=#^8I<8%%}Xs-^}+yHdZ`jSn#VNJ$uz%V?s$Y
zLFgE)>%A5f>+p9Me!#JiAe;t8aK6S`21(tb2#O1TAC6(u60TWDu*JpI8!ExeTd&Rv
zEfd_YSG&~Iywy6Nc0rQD(5#E;ALi!<K0C%iKxMe8nVuQ;4ZlyW+pi*V4=KsDMh~U;
za^c<a3(;I+jp_ZLU0^C2BYd{tw@V>Mh9HqQiGA<-`c~?t3!n_H+g}GnCe95?#+e;?
zK=GX?%*x;}uofOOu1RbOIY;kB#{v&?QS8`T4(S|lGTMTP5fYF^Q2%Seuwg^Ni;n(5
zD6gCOLg<uwZjrDc3%eaFFtM}M+RMG{$STfcvAU2xkPnaFlq*(MpluoUQ#rFpxKEnH
zLDSR=w(zzei)IzyDuQBUjSGiEK^CFgW0Baoa-Y8(;(2&^ao+h19YhWwwIHT@p%k3C
zV{N%@l819OevI=mJ|+;k&C(on?}FIti!|1=FVtl&<e?kFO`amC$g)%>@8hZJaWak_
ze@#~lywbm@&MXn2ZwzIv%)1U<KL^A+E(&~Bm>15g)fBgW<cRQ+z8=A7?%}@(@oGSQ
z__)Knlok|^!Oahb9hx-Lm+iWI^l&ue;^wZ!E8yxg!@T#S#`d{{@P)fn#Ds1@=E&jt
zUTgNb(eUemT<b71BTE?!Xc4X?!f~rrHPI83s@K{|V_VvQwpGZfl#NV?;&8jA?8~-Q
zkiT9e3hii;5s>sqkE%(6pNMS`H%QVJ(HX8JLDzXa53AF2b4b0XA3uB=9{fF|=v38g
zvz~QV8e&;5eE^#zhsR-sZCcihj7<QaP-R?n8pb09MUW&4vVcr*N1`|<OM$}$k7!Gt
z*bSN##PcEqk4R%=UuoV0hSi;Z568`%cTqAskZ=qGg(&8Vjn*Rx4%czkBo8qH7VxH8
zTue4!;LYuf5B`bPcdlmJ%%sw^S3QDDQc6Z}lTY8uc)lkgO<ltSfOuizL-C?m>*>-m
zDW;a!4hsC^l)2tG=*ROsR(|?-U&ao5+bu%BEo`_r@VX`lIR@ApGh0hkiz4#l0B?%U
z!5sHzXR+ZWYB|CkR^#N$PURMQ!<na^4i38!0L*a!8Gs<na}6LDDQr*_6xBvz2x{t)
zlHmi%AKgOUQ+d9v)9Nqc;f(8a*$=VR|I{}7ArkkI?oLBl694%gMn7fzOntC?9Lfhv
zPh&r@nE@4B^bhuI;|<Vtw7No*f1<fF#^j}T09J|p&JNcwAM^OoOjHLMb=VAMUw=wC
ztk)UoKYz29?lB!#Y_<a;Z5O>@=2$TlGv|#Th`@`L;?KGQ!Z99|@TbeY@A#m!Z@9Li
z2P$zEo}C4S>Agw3!J)2iQ9@W?f-~%ua<?Jgu@7dUg>@yPk#4I!B=}dJIBzHJ`F?dY
z8*K$rDVU#$1=J~Rmqm~I$2<kxv_pjhhee-FW}%ihDcuwoFWq31h{@;<p{52@%s(6T
z(KuC~S{AF1aAmgiNm>7k_4`5+Av1h`qr{i-9B?@ya_T3PGQ3yPi`^cE&w#T*B1fGd
zD=A08LGIE*e_O^3>a_^-^?T5B#|5%9WoF$KNFWO{Qs+2^nh#0VVur{VE0X_cEhum=
z_wyVb=dqhttsD~+2MFxI`{z?=Ht?>_O#1N>5-QEnccaake0QUyoz8G}VO92;iQXrX
zNRpfYh~tF?T!tnYKLsA9J;)i34b3Byu4lMGU;!*mNC+1PogBGuOBhtxG+`9*{+RS~
zclVP_!^(DMLMtvMbM=@kCyubghjHF#sL>C&g_e`4ZIK7t-TtB$rxSU;R8k~}Y|Qh-
z-7H2pzosK-@nu8pbzA@Fs7;VJer+H}_fks_uU^%ToMbmFr|E!zU8bJF%^gi8_mlGV
zF}?Xpga7i%C_w%MdpSs~&d*0cY%bTYf4KrJq^}6Qlq7{5yAP)A=!02F&6?bHQLU~f
zP-0s;axnD{eQTTU+~l?q`)k!+y>BawmQq~_nY{np&@4e4X5oGZo@TD;HV3X+=+RMU
zvKmz*_Pe3}VM{ElEn}?+PHF)O>LjcF(IY-u6}5xO{JYo{2G|>feHOK)qW{d_FkTta
ze?KkwTFDiwpL_t(p~3a?@VEoxWD6eA1rSNT3t9*ghJW&<1HBZ!lYZwiZG(=qP&+kp
zg3OJ3vZfoc9Wxk~5q`TGT7nc@DF<|dE%xnZ`pH{#I}G>bThp>;e^Qje-lILIx6Y{D
z=m{^@{gU)mmtOC&P&U=qw9Ov$S@kl`M$YqbCi~Cu7$~m1kNn^EoR-0K@WVvPl&!Ne
zGkw+~+mT!#G=IBAmB_IJYDz5@?3=DdNzlJG1hrNluaDQB_WcWjni*h%4dM^$O~$5L
zUB1ZXMOj|vAGyJlt7`#vzoL5C?$PN*ZVH6Ty253)lhR`Y^D6Pi_vEF?(XVxZ3W@F9
zMUgnt<>sV-xdX+@`qzE8Yc-|(m-L1i&#wMCH%4Hg&_d8^#j-1nxJh$*YfGAY6K}8v
z3u6Qz%)9?xKbhr@edyt<P%~@!Iag(Lm;-cRe@;%YwxZ_WlLFd>@3ttyH&A={OEc_c
z9;81n*D-!aG(Z-4o>jaXiodFrQz~ckGD!UzC8qVE=#AscFl*ra>WR+@7v~^VR0QSA
zx;6o}UIv1^6M^JPKW$`Js`5+hn|5Z9t)8gc-`~Th1>F%BUf$k%*%k}J{<D0s;y^9s
zd_)1)!}xu5c}ZlXh)8XL93%)b145aZN>6P(3t;<{_pa)G;_s-HSP*k|jzi^A?|15X
z{u>Q#IWJGA+ZHknFU3()J2!EtgF}M0DYU2uXg?JW!Su*Q{~<=+iAD#e#|kWN{b;GV
za(uuptw23}&~xTu9#{B8+u{2$fGYmQNLpd!tA(H(mkXy9yTYXEZ<A|r0Jg)n<gMfc
zrDw`1A`AYc!P>rU-OXKQN_GCo#?F;mEEK_HjT~V$-IxNcIhHVPt#1Xj=mxTL+?1g-
z!Zdfi4B~h`S7LVbEpsqP-)n7jnXsZI5b}LL?l@-07a7@o3^|fW%4ViO%S)P9j~+AG
z>hHa<L)C=L)Ohaz9+lMBQZ;~0$GIwXWxN?17Tvy?qoW$><A1FPlS1JR9}w-qVM&Z%
zulC-?o5dKIS`I=4Ccx5LeK)@ZpqcR*6>|l%GG-XP&f&4t4ilf^M<1G$M^zT|(kIny
zoR;m^CMk(NVAyainBY(ATUw*jzrPj9;^1cVoYuG2J8>`;u=D+7d|}T0F+Nb5B%;?@
zdavmD`gmEn{vqR5mj;b_xheIxKzjsp_T;W?PA`Hv{Oqiz^BEXXlkv^78&Sx~9JC?Y
zu+aGRd~Jsa;<<QM#XIKPnpcnBa`lZsXp8YFaf12E{@bDK|AFJjLPHfONGmlzM|mCk
zX-pHW411H^QcZEH>SrdM(QCOuW&Dofng;8p3C0X6GZeMBznk&}yp)muIlb_I9pJjR
zZwZeG4?p52xojba=kCut1g<SE+8lx`c>TM@N;IcA<d(Dtfrf!Q{(I<xluJgU4*W(y
zamBq=n6W5K#%%W|LF+75@3uU8G1-T?mm1(UDbypoP&Xp5)PSd3fMT}j>a4?;UhK%+
z&z;zNO*vWZ(??Iw_BV8dcb&oQV#FW_qBs+*hysY|8Yf#?UZl1lWR+{XoXteRqvOT;
z!8sDS&u<JZ7dALTnJGN6>7<3T#`Uju<R$K>24jDo31q75FqXn}+;EU*F2i^Q$1>e|
z?%sfd1!EO#Fw!*<n~zKAJXfxVi;X(?>x9kYsAw<J8F|u)i_&}Z9+x+VWUQ3GY-K`K
zI2mzw)fJ6o59y`+H;IKx(SCJScN~<JKaT>MJrbD#cinz~9q|H2FvMlLh04Eqf7E5K
z&_tdV7LU3`g2b}JEeb#nn$taSlkYF)hUdu4xP;UCAcYkaX=a;GnhR?AM|CHrBGAC2
zq;H;KmzS$-WZjuef){fRC5}!iJ6gcA;|6L71-s;LmigCrInu05dPm7RR7oc|s;pQ=
zL#3+~NTN2ZdKQA_4c}^D6ov;Uy}zHIZq^ePG$h`cEUY;$H_@2t78c=imn9@&lVs{a
z5tE9XbDlQPC0gJZxZTThbDk0J-~+xg*UcLRoRP)`^D*vL$Wk161VG6wm^~g@QLfet
zWA4~C$w^j!e%3D~j%?10)g#iOnGru^H-5SK72tmIn`{0pMOyxVOQYF%_thqE64oit
zZb`KjCyMEe&{~IZODw+-&zf-K>+1U#Q=Wov-K0n^^74a&YfmR$>o)htoL*jDfodV-
z=huW`VdCFTuZ@8J7Jlsr82X%RT0rvUxckg1r7G}_^EeHiWX;m0r8F}V1v9fomHF9N
z=&UV%Y<-bDUOUTZ(4t&bPRjAu`f|`%Q)R01MU?3~_gqEQMt{(dNw#mDJ-be*#rmYw
zMM96;^jg>QC8iG)_D1}u8pGz}G;~R7#eTq*RxxoLSE)9>^m97;`qfGD&Ni8%IpWo{
z_fJ2Q_2Jt1A~NCd!a?lCHa{=vH@NR)&5od0bkd*f?Z?)-N=gRT)(oxtk?I#m<@A_m
ziysRCS+3Nbzk&MQ0Z;2X>{!Tx;UWhWS3t3DkA1DkZ32P`3P#lCDQ`;p_eHu|%*snU
zwVRD#Wi*0Im@6K?`s`P8iZsI?CxRh?)x;70r6WR*b~i-!hF(RFzb&Du=VMHR$T#;x
z-d}Y{ef#NGnLfLXbIQ$eTk&MZjiz(H*w^Z<yiC44PG%X#Hb@fb4082t0tPI2ud05X
z(||~bT7}`o>JhJWj|%88`lK;z#CCR1Qw1?>$;4@8OkxnD@=8$7$1}(L`A5hIcc)~&
z{t4CZ;itOSd=LrAm}s|uXU!JX`$M?<38f3+BUK%4%LZQ{>(-51v#$g-Pw$9TreNDh
zN5YLziIFh|bR6-4O<F(Y#1Pobo5U=QH$6YeFG@$e5i4mu>u@io8ec*MzylKDaJmA~
zfnTai)?6;9*+#;xr*IaK3#6dNEkcTZ;s<-hR*>*;Ll+l>oV?9#wap{YoXR(421~|F
z=36f0bKm|x!D7uc;Qf~uDk<XF_mek;v>KIy7`&ySEjZpG?z*;UhzjP4P-_~_dwTna
z;6H)_t{=P)Wj@O<BMrCIS=%PpoJ7&v2wP{=N*ff$tjMQ=r2a-`f-aM}%GHFVlhy2m
zFBb-D&^p^ph10qXy`EhnY<e1%NqXkoYU;%AQhEMBqtoqrveNp}V!v27+4@VI<ncdR
z*#5=r-C_He3#x&Le+-CRIMr6ovS0JUP=FS^@emWr&E-MMx%!IUgD8fYR0cs_@dizF
z2NxP_Fq*Xr1(a1%e&e-Ss8}6w7$NK49#!~I8CS;<Uv{h%By2on_+c^r731UtP&(ht
zr0iYA5!3{|nRB}KmEs=M`OT;7toHt{yngdjhPBC#RcjtpxAqSFcUVN47&g#<xvI_E
zt+Nyp>&MIIqEDiyLcip374)T^?oV)Tss4VRLGJq?Z`6erW8D6LJ^Z>tp8&be{)x`@
zMDWl(<sE?LB1toHh}2Sd*)tv~J96;W0-%_i+NPG}o;faeK8M*%k>7FS{Bs;W(X(WM
z<*&|RY`8xgvSv$mlL`&U4DO(&s<5Tj{|3>T*Oq$Sx$)g-iG-j20HfGld7ysn?+2Qn
zk13chgrA!4{JoUx?0LH{n#Z1BK7<JRJq?f<v<qo5T5tBL28z06$KUspM8bmIAg`(d
zuf{F=Ha(DQ8Cl#tJf2RDj*fCTJ1?i1)&XB=s$>m*ie%c6v$2Js&MwR@4i2pJ*FUvT
zuVD#q6KJ`iwHw9z*xz_)fO?VE%^$7cj01`|Jge8`ecU9upgKQtBZorUJ;t~h62TmJ
z9+h|5E{_4dVjO<P8pbCL1tu+5P(2W{Y8;23eE`~cny2o`_KwJgauRk!I?}eGM%gaO
zFA~Q%Kkg(Z^~}Y(2nKdbp5<{n(TqJil_3JN&+cy)KKO7-PB19~ZaYP_5q6L&u6iGG
zC?#Y1sE;!io!ot#z}oH8kgQqFK9wR-3nD3D*k~{W#pU*!cv5*=@kg>Oo?idPS!ZPN
zr%hg+1w*XwhwXH;3|s}M9`O=&N7I2^2beFLPVowpe=$rTGsNOt$pM;z9ALuEyAjhV
zc)!dh*B)L~rxt;kwrFXEnFi9vKhx56au@Hy!|h{Qnj%>mMWkVTfL;=6upmEf7E6qK
zx?hslgsx`~Q>BD|?N`vD;83rjfyqd5=S(QC2TuEaolvZNeK?rDYb1@-CLqc@JVQLd
z`eY=%#gL^JyG&!W1WWv)P}fZhE5Z}J{3T5>$p8+kz2#vVk2mT#EPi^C?iid@Qa5=_
zF+(_?U2G*~V4nTO#><<aczKlO9z1`WmE>ogKq%OvI4Wh<A~(Xc!7Wvta=E#b@D)SD
zY5WUcR}6ZiB*n?Wf%LJNFBxrcOoE)lri6`?(~D|UC~>Q)a&%ldYg;QZjD}`2V3EY3
zEb<~JH{CWk|E@Stq=ViCLZF@3K$4d|WO(Tm&iu%<(+C&Wick1fU*$~jgTZh-1P^1V
z#c8D?u$msRpxCM7e4tyz`)p$zu%&D2p8t!8u*H>3pU4^vi%JwUEE9-jzx4K8!DLq|
zu**pNe$HcKw3$_(!ZsT))fatyf-|W!VBIk*o%~_-^nnvIinQe*rIgU2>%+q&fcCxv
zFa>Faw#YeEcVi+)N<`N0^|_v(iue3^=f1!p>umxQ4d=6)L>!at^Kt%DNorDEO{YgL
zA$LK4p4Q*UbILs%CjzFGg5*c^qB168h*E+_TfOw+E2dTj>vD>4{2hK@F&F|J*22SE
zddXEqI;<BT{;a2T0;a6w4@8z>tHmwIq!tbdLvtRql<XOPRY}u)`CD8Ezl}jN#A|aQ
zS_B6ar+U9n<5h{mhXS|k?b^c5g~F9vDo4!FJtDVJVk~u-Zk|xKYJfcgzzHPFnXJIs
z+PdU48(yq~84R(IyL(DvisNPyooo4Y-BRfYOZ<NJ^k%;K2kAKJar#|xz#^zCTY_bA
zF*JFG-n}p}+hD<u(Bimh;ruB<S?iXpEesPn3xAFVCv!N~nmZCT6fje8F~>=3xZMsb
z^p9?c?ABZPwYMm*+!T7@lxQd7b=4Q{IFn;cqI(C>;NAr+6;3MlX+Wwd4QB9D&h-!R
z)A_b|LFSU1Byuo!L2><lGr<7_88f#>2H<4sWX5CfUIIm_XGp|rlnIczryCJ&xMYfT
zh$8S=#Q5185nOKZ0xwkWh9-?5syKs35e5q;7*daPU%LaBaen=^7KxV!#HYTJRA$Bf
zcdJB*JQB6c4KlkuJIOcc3l}F?R`GLVwEC`x>EgZK<u61pYkHoz|L_E?<LYaY7H|z<
zRlZGdLH&aPsNjlfkazRx>pK<VnoBDb6?HOdR6p@Z*mW&zj^0G$BxF9(>HG#Y$*;6p
z7q`E}KDa*TC-y5in&<Y#RL%~2s$tk50BJya8oh62Nmk_23EZczPULwq8y<ZU_|reH
zLY2mt83WI7XRcX>BW+^*IN5!gEpyDGuA@3&Y8rbz3w^YT(mAF`G3I9otARU>oPMDQ
zH<=U6%)|b}O-*LY4Oi4Dtwd!g_SP9tm;$bT1Ftgw%_lV>j%IxdcCmW^Ja~<sBoqd<
zl{|uc;s6WLV9kgAIo9H3#qGZTcGs@|I{Ieezq97~As{y_MfU~%9lO(?%)3P1wPy>R
z->4pAx;)7fu{-1$?42<*R*T;)nhj#^_`I~0PWP%FQ73Gpe{T<GrB6Xma>X4sVIBzw
zOtP0>=FYJH@5B36AvPm7m*g|)k1KGPrX9qFkenI^k<0@7+Dc`LTP}swkKF<so&HTp
zMWND?$A?8u&q>@EZLB0<Bc%Qr6}>3?$2b>_;^yeM^cG##hA*ZlCli2Y;{Y`!lqQ9x
z=tyKRRIYDi^!HD-QnZ;W_KB~apX?ZSQAHWz(tn}^Lr$L@d5uNsj*mmNFl>!Eu)5xk
zPiCCvulbH&=i|$>`gZe@(VHCebkmVr4**CZz92e7Wwi)xgSGW(W3?0G1<XJaObrf=
z>A#6i_p!u0#aWA4P3k12pVffG2lZ!v5EgP`Jaw6pQ-HI_Drn4*>ZI3Xo;BBBx2*Q{
z!EOqvpZ2`yN?Aa`bT29!ep!k=)<R|EJ>ZDyc2;o2?&0Gk0YxmhBdrnryv1$jE<1~u
z?m5#znpV~bK0DwNar;{g^+Vun_T#3m?(kI#S!>n>TnF+p9WgO<B@w7&zO4WX;T6|C
z9INlH6S=hy8YWB9{`M57GV{V4qTj0-*_&S<4=29=S>p`yI<FK_F(Ha81=UW^1MBPS
zV&{E?&Hnel@;GR;$gv}wQ(wK+h@r}RXW#`f=bCs+F|dXS4``X-B6xjyzDmm&FpXIn
zwz$}tJdS1Ky0eQ;{~;t&Jm5SYeX?CXS>-1)$$3H#2Nag51hi8ah+TPIFPNc`o9N?H
zYMFvfhjq<^FHLT4%?o-<w6Io^yy*X&*4gXb64<|2E;CF1APSCTnO=2T`eu?_Fl+Zx
zwYa!)XKVGobKM)%2a#aa1cr*LuP74VgwD_3obu&Wcl`Kp;-fB0OK@nd@#uTB{Scsb
zWVo5U_G*SdEYvq5@f*<nr<xWRO{n*!+lrcN8Lt}LtOs^`AZ2wzq#!AGP3bZYW**qv
z++!|ixV4?WH*2f2vsb4@Bx+gW_hIs!Cw_KlU}gW~RNCqUg;l*V)Ww|x=H>Cofrv>N
zwlUe?<Y->qzJ^VmAC(4Pe(BTEG<gdI4ehDN4PCvmrsvl%XmRFX(hT|1-pU3ji^oi&
zH<kMy99EfF<-&K^idm%o_FmvcceK#)1VgsY9@bx;U+w`0h{VPSmY!~nF`g9UBgW+p
z#2`@;+9UWI*mniVV!urNl2w8wlGm2UAE$5X7$2`a-UdwLdv)^gB9r#}OBbN0&<yjK
z7M@K|1{wn63W9$T=f{encyWxTo0adh9LN9>`=e)vCX*#;2oX|X@l=i@H|-w#;{*{B
zwxsT*gqd*`TJ^f^`mbQ@NqPSD%SfM`+^8|Jm_fyQJ8!5^u=4VaCK6V9=wt?G#F({R
z^;33ZGbx9&gB``y;~2ZvW$Z0mmc8r3k^P>C1Mz9Xsf0b%gA0eTV}PDTi_(frdt8L)
z`Bc*Y9c>Z*DJECxDMfGbgA|e6xI$c_&fwYl)9SiI+g3E&v!v~&uN-*I-%eO|b3d}b
zbnRo~{OiiJVwvq;mR1iVcQqS?`Ln(X8C2Un;}`TnnG=Vpwq^v}&2<sSC9T*3uE@3~
zuBG`jKGNu3vyB@6hqn8fKPS!;&&;LG4?f)QXrAwP6wL3JmzGWf{-o*#5Pn3hQB_y>
zjr=?Z!;GZ{M?eXXUkmf>Em$Z+76_#(eO=6JIV@>mGqHKs!tf%`<OI(FrSlUU&bi8o
zcWx<u?CtrE3!YthutcnX7ax<P@Ei5#N}oRA5Lu1TTSebkHyv0ZvmYXW@nr?xv(y>I
z=c~#BgRQvXpa)U09b3fiJzI$fZ2oMx&Se+b%G?@L^b6Cc9J`D+LPX)^^hYDTzkP3o
zF?uK1?Y9f@=_=JYBYSrsS7Bl*_gTY%Y#h6;ttFd~6x$#gL@DpU81`$!Wx~Ea#wU?<
z@;Ez->Uo7CB1Q(+6TX+MM~OkCUHkH&?klfwS#ei6OP`|<3#kbexf?v1l|5yTEpsy@
z!<|Kx_D@ean(`8Gj~tz!=x-?kYOh~}(%B8J(I>?pj@aGY+&m4*C@CAaWxvnWc;M0n
zBgMpPtWyCAlVZi#4%^~wc0+V;K=PD3><wG)O?KkWK=rcB9Q(=Wb3K1qweQ=%NK*Av
zjT}r@NJ%lkh(KQ#$M#Uu9SSqV0{BIVVb0vP^@zL0OzIe|6v!M+qU1<7=lO{vn1*m6
zs47ljQ8~>~S1-LN<GPqNbZ*!uNYQt9kF`I@+%*uWn8jCm<pi7f=;9<F+y9SGp~=X<
zOg_-Ml=qLc?qyeD2l?I{xohb1Ic|g;8DIuG?WcrCriLf_y`~-9JJWOLGk{AnhY3Ee
z64SPiZ|B)hS0U1#O(vkKZT`U&(ok2sl-u!?eYMijR?w~&OoGiE&+r=*ks=ZQ=|}xV
zs{i!?P@VQvHnO{}01kfh&&z=iPR*UC{-<pJQJU!n{7B=%7*w)AB<M49>6hlB+HWcH
z&9=^d15M)kuL&&yfi4|QGuPvo6J2I|A_T&%0y2Lb%_@0cljN^ih#wTa`{%f;lHF;N
zSGq#qHAJDMz`9Aph`(bkH{v+4FZgs<d64X){ZZxYdGqOYLLMC~F(?}V0qIpc=&uy;
zY0h~9h#J3WM!aVFttD!OGQ67PkDi$qs8F{UD$Oc;C@w)?Wb>8F5c;QzO=T!6%F7p|
z2}!xF8aNC2z`%qNLkQbsm3#L+0%JcuK7RePZ4mS0al4EaxoMY`FrxtLN|Zqr&%sdC
zZJ`x(sOcjs={FkVJR#CJ@3?Lkdtq?)_J-Itt$+$D(z#cAmOM>W1WiSDZmil*nO7BZ
z?y|2$xk+BEQM%E&=Uo!}wi1|w1Vz(GTQ3w#Ns8oG0Ma#t^x27C+1=A~DHweloxe{x
z>-^iLRU3v)*1L0{I`ak`k>+DKq}$$^2nt^Q_lmKIWNGPo<cQ)BltM2g!LpYg&=(Uh
zpXm`J3`g>4nIjg*BZ_1DG5hgXn7$&t57BWqdJp%S`hzj;z|Tl-uqBUi?kN-Ww=@tz
zwXDwq=Lf^yQ0Lwx#-K5h%3H*GDG<@*`pdpd3|W*Jgrbkz-Mq8($WuJX2Tsq^O6)SB
zc8)Y00_xFqLMs6{i{H7AXeE`sxC?5Y@E!p7!aAE@-vXItBjubi;4CW`K)llaIWx4Q
z8^OT8F;gLX781i#w(DiZBW?!yAr|Xx6WJ$W$np0^ERGF4uMI}~@1}6K03U4Y5bDw%
ze*U=ST+ddh9Y@zkvG7s6u=Pte{P_6zWnLL*#{XnFb&;`M9=8N@qeT%$ewOQS|My^t
zW3p7IP_TLVKe#N+k^_2Ngj#eNFkmbg^#u{Cie(pQ1e46aHd-F^`f9pGMR@ING<Kh_
zLSb=fc2FBK$-X1jUfeLT@%RvRgSYIqL!Bfgd=0ecVZf+cb7$8}ng1apaR(V`Ir^E&
zUWKz&5bWaX)=qo4thw#undW$Gi-B?_1y#kg1L?{s5)E|twls0U%vu`2ho$FJZRpO}
z7Um{QH=9Tz9tSfyi}u$hO2v}E&W7Oe`P82uwGdLU(sqS_pKP4;>s)GExLnUL=0$=z
z(h@F;IYwDz#<PTa)Cw0n4stHr0s409+uv0Y%%BaTglM~TJ>6ULin#L&L`{fC*f!Et
z*x<LKuR(6!fmaNcW03m4|LC%(J0ZzbO^ls{feuAL3`4+9G%RW`D*CO(CSy_Lp^gy+
zvxdFlBrjA+Q<~_oy1}Ad9~-h}JitpkcsG|-#S#6O8j6a5_vKho_u^jmL+vK_Tn8h)
z)q=@;WPm%<@ne(u>vxZ7!-mFV@aHIKx|L_sJ*3OkKF&HJX=s-F9~=UE!$1_y2Is8)
z_1)3zpZkX81_6IoMAWCFD^F)HGiq1j(UDZPmz;NS>;(Q?W)N0N@DD=L`3fE+(}=#_
z_iRO01+Ln?VOaCd$v?y%MVOGj5{DcPWZ25hhj<7leAyME5S{<j3WI}WfBy_S(=S)1
zs6EtVS7R<&iaadlAu9YOB=<;4MV6{Pn^m&3y!=v3CUOF}?Zv2O059?%gERYbV`C$v
z2je5iojW;Ru`LMGmsCI%qcrvU+DQ5f>AKSxB9qu}`tRCz<H-suuPU6CC|>a@p?k;u
zD2`y3Abz5*Sbk7c;DtZ4V*R83^^S@%6bXG*Xnj{eOuspI-2uSwgEyI`KZ`UV#<Y<e
z?pZBMaSWs+e}y!c@dq<9<SUrsHxHy^8^qCM%m9qz8VIg<j+-LITE^?deV_SRBPJCp
z^4FaC!>uS~-Fd0#%68G_oy9ar&rfl3ya>9t4gFE{9@RY1Ue?YpPBva?ZM|k!`9oJ;
zNjX~(!e$;x6G13c)z*s;qo@|IOlKY*o%;8soc(V%fI5Fu0>;OcW|ac5L|Bn<tNP|p
z3kY_~40kNMs9KmIJa-ubIt!j@cttc_!G$Fx?242m=<F(Hke9mSzAU4l1`u|0i#DVc
zW@@^l6aU5KxcJjklIaeOB17`71aZ-}I@fJ^FO2R{T{!zc^iTQ=E*6})Al}Tw+>D3H
zx25rtA9x#eR30=}6tt3>ul@p&#1nWb4{S&ro20M&R)XxjUk@m3O+XH^<@W1hSx?Ao
z+blfdJ;N11Oy?zQM0Z<BlG$Mzv0u8t_Ju?-=X%a8-PT(j;Mn!jiH$hkWMQ@Jam9BP
z`W#8%#zHE!oi4~BR%l-Ao9hFg%Vj<pB{<(BuozbnVj-9<h)9F`s)NsN?pEPD1Ohpm
zA*-+;Y;wj#K`$)EEJY27axxZ8=yetaWklszS4w6dp*$%|Z%LxQ3^KQX3CNDBxBd}h
zR2bto1E-~+<Jhtz#fd4ct^c78-}$sOUP!ajr7TjFDyVBo#V+QJhh=Kq%yPf(DH3K}
zVR(%JuMCUw@o@f%MjyZWyORcq=%M>@w0i|M98mh1`ow+=W8W2*{M`BQQ<A;PH$Fkg
zB>83oSkr%=Id3<0m3EL;iKTD5@56kf7R^K&c0<ay;XHeTmb|g-ZbzO5{y(g{Rbdzv
zuc*`$UPTQrhTxwTl5zQ_%g&>vH!2t9KQ;EZbp7W&<_G;^^B5?44bW>=2W54B7m=^s
zblp6?s(<Ka_|9i2ygoboagh48T(38BFr<`tLd<@Ug|e{sBbY1AOa-83c$kI()0iH=
zFSq^Q7yq+cW~+DebK17}q5AVQ`T9CXYiO6j7;XI81?qAh=_7H<ce<;_8{Dak0d3@B
z22KwKoStY6WcCJ#7o5k(H<?}gvY5%ZL*}>7aJj@4;>8=ZnX{8M=e)ZhvC&gM;;BQK
zdrIjqXu{#?ky#4OgT^1en#p*Y(^3tBL%5}=(XMDTUl9yOA{zk{6i(%0NwDK@jD4=i
zc`=fD%(Er7!Wf76>5rK5ZtoiPer=F<QLvuSu92OXxA@}U#2(X?$UaY|><Wk_EN?WO
zW`O#(<OsuhYNg$AL|L1;BI|ISO~yRJ@S*;<DsD4N(w{H!$^{@z&CM?+RuMcX*F0KG
zV;S2P_V~Nc{R_Zp4T3|tIfPsmLS$6~f;5O2>5)WB#j%2lNM8}3{NkgKaCYvI@{8$H
z^cAsyY%?)M+J7`;gEsa&vtqjTK1aVH(2G^Fwr~qQrMkHA!d%}~o&sRZskxRgItjS&
zBr##+CDE@ejDIidDhz+JG=%E_8_o_YNsI1XDD2`|>2>xnqovl&2Ws40p{fb?6_p>D
z-sURi$N)=h^9{jAvAORngY?4RcK&J!rl&AVzH|HB>-^5W54_lgP)#F2H6V5I2oDYu
zdp}3czdbRsxw$FA9k9VeLQZy0S`6~5pM;Up{~HJ?*mwo5E+fBoCy-WfXlN93d?=fP
zavon~smf1363%CWWM3?YGLL0HtN}El5Wb!idhi2F6IhG#s5<%IzfU0=t*eShQK3WO
zSbCb4y6JN>68(%Am0U}AYdkus`|<M2Id9jgBfSeVh?7#{cT?MjQe^M<^GFMo>8$b4
zEM)-?ki^9pZx&DHn5q`$_-_(i-Z#-hMo~C2XL9d<2`GU*C{wdq@eSdX14+CDWV-Hh
z)UCPWGtZM&EBHGR>NiW4!B&Og`tWj{$~KX_N#$0-*f4pZ{H|vJy;0H^cx~MKjW68e
zvA`lD5#tAn<6%oU6l<LUmDdL9s_-SpG?|8zo=nt9zG92ZE`&*OVgNdDjKWm(!7MP9
z9w3TXfLp@R!`=vhBl=4O{aE~76r;Z6H~c7+X5GZ+sIa91J_?bP&LLQPA-m4eEdT12
zk-ym?!GQ50*6Dip`iTcm;Ak4VmqIZG_2+MvUH_q+CeiTlbAY{;lj^jVz+wC3j%o$&
z*r8nz!`2HvIk!PdsQ;FI4<ptaI^gj#I5;?X0m^L4E-3qie@s`Vm~CR^Cxq;){~0Rs
zfMMJLtqL~6;h0fKhvbf)h*N56E4wmgkJ>A&+@smq>dAurW@_IVai)#;`sj+t7fcv%
zfZWfl(PE1VcCzm`<Iw2D5c<&cj34zS%Hmc{eb<>`Y(kE6I;Aa<r~h^5_id$+Pw|mw
zmkj}h0}5G+KKuLoEUOx_W(>e^f|8Olb)Gcfi6!cDZk1%l(Y|P)<3MD63NNj74zh>%
z*S0EWRB4K}HZ^l8x?&T`v@(|a45B#xX_D%|cU$Rg;qX(YQoOrgLQz-^N!}<62g|N2
z_|t^EK0;bUiplxmg5n}=x<8q6DFl{bGH3uX@2B-s(7R}SPl2c1m&ZI^@}%a5keN8{
zrU{^8BV}>mv`|@U^C(}N0gCT#$D5N{7?=Rj8OO^hC}^R`B$fs%YZy~RaWf;2`m?>x
z7pe-X)3q}fw>LEOWHh)pm!0$$Z>I)RCnqK77n?{`eXUe0EcZOPE1TQA4W);)fUEEZ
zxC-|@FI#N)c{NvJSp*)#b?fz~9`kMOV>gx749xA)y})EgHL#QI?=$)O)2AwyjJp@6
zMcnkhptx&Xl7=KSfTlXCK}C5?0iR=`DC0JCwQYsiEY@2mtWB(;;pgX9R#v(Wo<&JT
zfm)ux$mInoj$j$Ei{C1FlGkO7eHTxpxf&A2bOc@OIy-ux0ej%+k>6!#2=Ui8dY&!6
zr=jVfAdDn&)-7jj^~w3`N$qU>X$4&G<8jMpM0f~3#)`ldYK#_tm41i1j6~$JOs25&
zmQZQhKl7O8ZFCYcvu=mwrfF$<)$$*=q1>3JM^l~gMv%!BX(oWLD<cJ#;A2)Dq%Gj8
z;kITcaG6^R$aC%ws3YoRvT$+yx@x6Y$Z6)v!~0eM5t<5Y7*GK<#lH0;b9VB0#!9{J
z(yk-ly2xI%juKx)*59Wpd=(4TeqN_YSLuuYVjiEpe8&8OPIqB%M2W&V6G|M-a`a3<
z#4v`v;*Lb^QNUlB<OSR07D>2zM7V}oc9V|{OWE^5iauBYq)9)rupNVVW1pQg<cy<h
zFn?pH;*nVBZY*t>Fukb2YR%kx9V3hE(h_F9Q}%++I)Pym*(*UslF0@b8*h88Zjp4q
zzB=u3pox@!PudYajQ|YMyLtV5CUeEec$*gV;om<dEL{sGgekS@*#MFhxg3^QQN1K`
zN`wUX&NpvT^*-Tiw!rezbr9I5j*CEc{tOD$nc0CVTL>v)73-O=bG@qnkvw3DBc_3=
zqQDiR<C0jnk0HqeGHbj2FnJL{(uE{NuApc6epW8H;P){t61xi#`mXGq6f|brC-mAM
zgJdWLEu#1Aqee5=wUznm{Fk`4`C+g8#xQQhgwc$eL8=pu8A!mc$huF;yOGj#dV&q_
z!_!Jb_h9CNs@4`%a|}Bs^D_1_+#XxW6%UG(5NCS1WAfaauo=Y2zfCzZGKMbW+rMTT
zw*~2nb{{2);{RT;q36}VnnOc6En}TV7N&cAP}l9=`xX1vN^lsHfVsg@>;OkyUA;UM
z!1LEJ`kue=+4+4I(ul=~Js-t38rZId)6ipm8C^~BGP&fK{UCceX@hJ^IW6A5N#FBZ
zv|xD4^l=Y2njKU7atr?+)5a5sN_Y=!$(~0q#jYkPk0$ru@4C)SSqKDg^F%l;p~2c6
z!z57MU}C!EZQ`(0$<N&}7H6ox$L(CxZBo1#{)#8c<a7DqbJ^euyw7YOIeAYcC(6XQ
zvF^Pc2Lea}K3jgey6FixgBdkWc6RlHVq#*c*zj$iD3nNY2fmI~Ma(|=iN0Mw*>|3N
z(ytHL6Z(nX;}<W4@A_<2B*HQ;u#?)uw#!K4^k-&+E`#*AURJvJ4I_bpSAJ!Pb7eKl
zeF8JEkYerqt?gyMe(@*{40}(3=8Wv+>6tb_(X2NdFI%<d9Cf_h{Gj){2s?KV!bmUu
zrDxTm+j)}nPmHl${}(Cx=3a^Z&AzjzV**klgJedg)mKzR(h7fSx|PVUb?KiL`m=B-
z+bJTI2?wdV&seMP+YH7**_l#(=lKm3*AMfB!H`U6O7Ga7i;Jw{KZV8PzCkatG-LJO
zoPSA~-SWT~f6nq55`+`e-QI)40I?MdPN2Hr7i<k$v~w67>F=-YkdcJ)a0|Q!555nD
zOw12wY-I^lS8)(r+rt_y;4;VLa`Xb`K8mgs)DEC#uKVCAH!bcxk+lDD_1*DQ|Nq-1
zdmj6cy~{jV2a$2?d8Eibh|G-aQO4oOu@f>XqsR`~;vB1l$ljyujAT>y%jf&K@89ox
z|KrhL^nSlz<9c4t>$;wGx*}{hCxq!G2gMt|Mm@<o&10MxLkuVOo1Fi3+aAo<{CT+p
zRe%I+6GD3f8im#|ymek<Rqr4+SwQhj)zco@X{)-Jmn`V17-;vY<gsGK{<@XQQC&Z3
zG+PG~K5mJBVy8P40@&jbfT)~7h4z~sQ7>a{p4_M;E+$4revi$!ig-KN+Wc}&`m5%_
zC1df3u}~ig>bglD6l2ts%+z1?6sBcu9wTTgyztXrC0?f;p<pM!{qb*DFw0_PCbll2
z3><pLMKDT1W-)sLPKW9VedBAAv>{tTjd{%0ToAlEX)>Ig*$N7`v0+cpWGk1UJ21iS
z2j}}yao*(T<L1J>gkPOm5jLpXS<^l%-E3H_o?%Wk0_ald`J+pu6K3^v-Fw!v+{s4w
z>7rO-?jG##rxE9eC_U_k@={Of_d+jW@Sc?E;*Jrn@a!lR(%6jXtTt(zqZ@xhZ-i|o
z^UoEQmF-8NF2!gV`CJ$0BO~TB*tv(k5^k`=hcR6hESUIz7zO;E<YoSSk-_~>^7!=b
z#+aXn$!`ms$!gPe(`9w2H#k0&{}c$DVI;?u5uz93Uez-D60o0Ee$x@0<ff*frk;d4
z$EQU5P^A9GBx`I6<`qaERIfcDjm#MRN3Rd3!FDW#sq?uN640={9_iY#PAIa%+uz<*
zG<~0x@07>Rwd6x#$~>5`@{MH|NAqO$=R#FBBDPDzzDmbJPtD@PAGKwhKR;|dhIbtR
z9P+yj5iGM$9%UKzqUg6%bOfv;?DmT`eYw`Usj!8sS0}9o&o|TYOl6m&?<}=DP{v-J
zOPs5{&rV9k-aW2G3Hk}Z!p8sU31o;=3YVw8K_?S2X>9qjRI*u$fO#WQY`MN6N{5ke
zV~wsA+IQAyAw*io1FjYVfnGdp2(!>pMUEQ>{V}S35qPw8*EhL32)8@C6(q(>DfQEE
z_Zy;QX#*}ctLp2XYwBwp(a_VSg@BHQ_>uy<1a&jb57|XTD)rGTYvfo)(z3UL`92i(
z$?BFKsE}xlfjbA5yCI83g49U(`T!kMu$vY0+(x<*&|9%vbnsi+BIynsMSk^eQ1G@B
zEIm(f#XbX^%+7t87U|fCN$XV2?UECUm+8*>V-l8#iOp2i6fVIQF~4hKZlc=TFz(%@
z(``mot$lDZPy>~G@)I+302t4xI;6f6YF}LFqEp$wlj6+!+87kCT@b!3eXg!WDvQI@
z)$8E913zEftAqv5s2UA#7k$lobXt3RI}!a}eY||bJah{L8b9r{`$4~}C|8{MDboR=
zO4?Dh#!}eBk`AZ5Tx^^!nO%R|Conb=eE5KwXpM=?ZlcL0qmD&K%jCXSFIHH)_hQW%
z?xssH1=a9NWaqrVE=!Ie&x0a>(`o=KcmVB3i71mdMOP`4#ih0N^z;x>yr}#ddv7R;
zuOs3G!Wn|u*FEHS4HivP8`Al`9Mo5EX8-0&=-pL5orL4#%duBT&t|UfShFV2CANO2
zzK6YhD#sILLVQ{h)CNi|CWtq1!tns<n(^T_n;0>4F8tkTD@>^11`~Oj*7F`+`y`uH
zqcVrfD&%HIjkoyXO+0!=AIi8FwuGixH6u@_FF708^^m%=-po;mfwSeB$<CA0B#1D^
z0E3+4;X3PR?1`lqiD^1xOd+d0&&lK89TNfeE_h|5bt{2iC)&xGa>Xo+F>`i=ulh?7
z#JMXyi2DiaXKUH<{RxqMCF%!M+Cb0w6V&{@Q)mPwVgDNG<Oc#L4=24D7-BPZiE*Yg
zOeQZ(2^oj;BwGsnz+d0Gw%PDN{yUJMGVu~JSN0U;=jUUHG#%aA<>s(*knm9UIHN`7
zh$`|=Ob7B<8(^MeW{z2CvgeNKpln_BE1SCihb81|=h8ACl8dyLe);eg#{{|U+(Y(m
z34tF@E3M86aP4fVuwABDI=$Zm{T{ME{nw;xwI6-yU;Z*Ms&|#Ez<Atdgc+<dS{8+f
zUlL7`QI{vgv_X!}mzB=+x7eP#k~iB|izKt-J!k9EDv{H_>SuP&I}bUfy0w&hv|ZW`
zjsIS&o|}=fbJ%Izes6IYZSlpaJ!B<4n9sMOiSmWaZZ>VC{~)6tLqDkvn=qyQkY#?s
z2#j)`G+GGA|LAudZ&>r*RFqW_#~erHQ=}ehCUT@@W@d`l*3~`MyS}h*#RI^)bmU})
z>_drD!Z~l!a!8~BR+~?`=&3myhs?mwoy@0OJYzOiNoiKQdej_p^fMboM!GDtO?>j?
zw*v#u`?3*acH>KQ(T7xwSWL_BgUN;Lzeu`5A?%9(&$Lh78ZY4%NxbOETZd+wdsXD8
z*^sL;ZmS%utnVWpiltWvi`-n_-29E~YI_28=3Gjljnbt^`4VtHaYVG#K1<<EWJVHS
zyF<Jnr>qT>giof@d~VKNY2MI_$TMxVpqW(g#|C=)7H_KBp$@~NAKocB=r+!>alPjT
z0((iBHM9ItNxJwy@BRQB-KVV0_IoQ<(CZbMdaEPhPvwxaTcJM9bGMlJ#ltCpHbrYo
zCc?qm99{0~C^A5uvAuLiSP`63%z2lzYf%7EP4jC{Ta1_;&#WT&#fn`6UQ((%KEibO
zEwK9EW2#3ZEueKTb91AGHu12{4pipLm$w|EP{kNw-V?D)$p>by8z&n_2w<$WG3ma9
zc}8Vh3-D~bCws0Oa|c2^P3>*){f(0V#DR<DmvjQyV!+Q-?>JcnpD-2bm;Kt$Iu`ox
z5IFhV{~Lmsxyx0=aW#IK^VOYRENjsH_hP6^WBbV%rs1;kRzumPlsjW!NKUCev^((d
zv2+)Qo0XN-hZ=w&Bn>8ZduL`O?M>mA@icMxs3bzr!q=_LgW`Deyw4Y(4F6GQM_LB|
zyl%UB{G&of;@YiV?c>I!Kv`Xe@mOSI>qY{oEg;?iK@?sA>zReLm4<S1iD>bazpgJ2
zW9OS_zG-UZO51s+y~hW^?g~X<!bg=m@3wKFb&8(AL&RYFbH6vyIUScg7WD*7PqxIZ
zD?&M41T#n=f0Hr|$6X1J+Mm<SYwLf7g?LHnO!`-4<&HATq;4OOv;Cue$<5_>4AjDr
z5?H7qu$W3<r`n{kKUW9yXF>6I%f;iAzxbK>RWDzGyjBKP_rhA={7ZRT;i=(9_k@uO
zF}v9ZFN;>iK5%$iCJ&9&XNY%??i1ac?H5n4_1<~(qYAwBEmL^o5=8#(3dQ?vv-{g}
zJY}*r*OGUa&8(?&K53WPCX*3v)cUSARYK>}mRge4W9AO(0|z*ty9hDGot8H&zj(>P
z`aXFN8dG^hx>&g_W>5LcR|Gd6<>NY9y6Zb_JZ&c!7nk}*RZoj`Rho0Xp&>gfX;Yf>
zYv-xg+(4H2bE0ps?|tbXZ!W#!;q!EPoojEP)k3!ZXbkuMPYnOkEwO?>Tti`JQ1+uJ
zOUeV&wF|n{@A<DB{Ne67!IX8_+yn23f`V*!lJvhEtTl0abLtcfJg&;O;sRRY)SI^&
z;%4vK1P}D;T3F0-?D=TBJV{uHvuDR$Dn|j;br63)x5lJi+eRR17X)2qi=BGaUzu4s
zc6wA@T|IKY7>Dw+Gvc|7?1a4Ly2P*jmcA_y^DKT+4qJa8^CJxb$D`QN>HW)s#UyD*
zD-|W^UiLQJfy>dB2~o<FVs#n1oC<!SFNM;6SSdoFFe_tLj3EZ8j{P*&-9j7M6^|=S
z;%KAYS0?7>wmB7pYWTj}mCAvT<uyMs9aj5ijOb*~a4KhJ2nL^58JQ=+?aX~-oWpF)
zNym(K|1QhCWtE*hjQiF4bMaV#4|Wu0O=kl|A$mGNsVD}}u4BxcoK}Xi<+Lzg;L$xk
z-zeNpXBBNWas1-%VNt9ikDR84#wOLm7o5A1&3I?vq>!^DGBF6RUOm92Gl}UCudr@0
zzwY?0WsnR%KVNK{_UQ%Vr^mEUb_l~)zGhBF8Ez=cRQ#oK&8Jl;iS^|0Ch%ullyE6(
zl^0149}`{^+I1vQ-7rE(dozYmFPgog#o&yqGFa{xNF#m=AyCaqT5?+Q@~nh;o>JFH
zK0me8M-BlC^l7DLU(j&5raQ_qS_nINP$Y@%Acsd&$$TOpDUzl{)lW;ad6UQRz9r)>
zkk+2PmYsgk(DX%ItY>;!RBVW<<hN|5wcN^LXVA~w)WfuIEelJ_E7aQL15?*PKVm${
z?&r@st*VUbZQc$v(Y9dDYE;sV3t<URN+w0U4&51gHjNusgZPyT2N%*tE9L}{WF*Pf
zRnh;(Bx?NZ&ekW)sII%l6o+}naU*p%?OH4*8OSeI7Yi6kRrcR0QVgz-$h!J>s#?bB
z0H}|XD<9kS<qcu(*Dm;ecg#seIlua0k&#!~Y_HdCo8Pr)plbfLLT<X%7fMHVp<pez
zTZwtiF*x{Ya%|Uox)JX-RcQVtWAAlIMeB*h@>l<I6c*MRmvVQdG7;`<sgXxyi3g)m
z58msxR<=nu`>Yty1WO7a83dw|fSyCgTN)iNbn|gKgWJ#I&&tzgwOnr~xp>4d*P13q
zK77_T&xGGSzIA-}`^#V$Ol9WKulM`lX#wlI&|6MUk(iHMeqs<`mv>h)NSnT#9w$Lm
z`+-gpsFi`UUW@&e{Sn=(E(w7zYPHc-b+@=6_BchUU5^1g%&h~(Ne`^u40o~cGh>;o
zmfP1~<IgHWW+Hh`In%Ymuh3CX1l4OP<UsDY)<{i<c#djtQ>#VE`-ov;%cJVS#X$8@
z$|N)H*qiOz@6U|Ac_jnW<n!n8=3EGiI8BD~tp<$|zkCRKgN2<+z==X8NOAp;&GOKz
z7=8P#0lRgRpD4AW<nwVYXHiqzuB!@0C`#2J^i#Z}+xxgo3CGL4)b=YIv!r<N!9{6!
zZl(fmzyBQqWn@8BahB)QBs>4`afe$im5DRAXP6+r%=be3EoA1ibxghh=m1~dQ$J8~
zetwvm<QRQE^5poqa;<uc@nY>jQlP#)D{dlO<oG+lJ4V0eX<6(F6%&y>_OQY`-k9~t
zs|2042ZjjkCD{D>jVxIyDZ<xB_lu*;Xmmx|_?sRILDJSQPk(^Csg{;fXN+0lm1wym
zoTPH|>{GES-+)THD_!k4W%6-1ZOh%VHelXZelh+-`DiolD3m>mDey<Y>v!h$Exq}%
zHA;&2itaINt<g0ste?c<R%s5wTX!WFHk*zVTDuOjA6#u~dlH}0?^s<(1G@Q`w61_i
zA!o(8Z{O^hMbVtNc+^3JE`Db^LJ*ictma35HM6HM<jM-scl&9-$bf$sbEElucDDR#
z%fY4Stbvf{u|Z0HQ1UIYu0-m|!`t7H+Vr`IGec=^D(cgcS07pHs2za{1_Is@8IEN^
zcRm{zu77m4+2ABF)pF5s>kq$=wjE{x&MX&-jq~260$Tp_U%ldwZTHtlY*Jm0X7dM|
zqSFin?{*^h9$iDwin8A+mFf_MbCz7LSh5(^*y!opFfInG<+oUGudl1Osu(qsZnGT%
z*FefXWxJO)7}oFNJEKYKtSEDy(k$Ps?Q-uYHM4Jvi*0_c=N}WG7V4@q{~I1DO@?=*
zJiZ$od>S(k<>25r07!x__0zZtd2RGwJ3*M`L6uf<Nl9-EywskNNzyqU#fSmorNjZa
zv#M6_P|IbHvdh{}fkhNc_49cR@ZPHI?P+V1KuRB4{VI!;A0*+I`?Y-R{#>d$?{8MI
z*P^r4ZB^3Kl1?*^6XC>vcNHDD|DD13xW>}qYgl{I4o?RlCwnA_K*7=B8d_=$8pGh<
zmey5voioOwSzF2DPb4T*7^bpcHs8L4W5|y1q`XU!jW6o85Uz!1?8@W30t9fAZCR#c
z%SF~$Tthb7wSrHyDPmI3<y?Tb+0!Afk{SNUl@`5g>B5MnZLy3K!hWEtYPQbvod59L
zOdZ%Eq+@YhS$(s+I0|vM8+nz=3>pMPLqZG)?feiGx$ER0B*<7)>Kr!{B^^!1U8V~`
z$#H!`YSju+PEPs1?ULTSb9D24^`}wIo%3|Q^PAkx?ZRNYXn|lNicbKRUMvf3a&kO#
z`(dc~c+N|4+0>2Psj1$g84T2!mt2kQAYN$G)oLLrve!;vH?$tDATO^ZFyc807`|KE
zspH|d4*(juE27N`OxUr8(bWg@&c$S;cQ%3wTp(j1wgo|BUUC6ec?r*m1^W*^$HKVp
zw<(@1EvSC?!XY)ZmgH2wlorOT?}i9C?V76?l^K})1U<$eKE`?Jn;`<2*NMFH@?T<A
zk{^d?xwK>Ig!|xo-EmeiRm4jPVtNr4&|oamtWE014ZChJ49=5kLsd{D#O}EicR}Iq
zxr1wEkoAe+3yMJAu9HA*B1Te{7P&L=U$6Mc?ApKePg`0KMZ8?dGf@REqw_*%_s<yu
z7={QRqd<?vX$=oMRrYdWw~Pls1Rje!P{M2zLM{l~#zaqwtd`K8J0*9|TTQMY79P&=
z-$?!X_l?Zx1cv|k8@h3rW3&-J;_cn+=LN0$z%DI+ZDE-$=jpOOey_MwKNOabP3D46
z+52v|C<$o^W>^d>+p_$zFS~aGc6-XFbpKdyfZ4_Sj`&8AjUvjIj7G3tVzI6KMJGz~
z!?f4;o@9l2rf64aoAhSNHJy?SqtQ$DsiRPhY$-xJU#io~c~`lkYWVguG9I!SFG3GU
z(nF>uCcZCeN-|JiC#Xe`ac@?zUjEOC;Oxy~eC-7dJer$yd%Q}`mD1w1*$mFy#>|!C
zqVF~83rsk_7e3?CAW=C&BJ}k1oYIL^+Sl*jd|4{Q=IGFD+HWxO_MnncRU!V8-}}Z!
zJR}C`T5lfCQof*H;*$90&!<n;6OBqQUf?m$W*_r-@@alD{Ng%Y+sf5DwG=nx&cL;-
z$#SLqm7?Z8iz?yK(IW$w%W)hofl4aFeh)|TB*G4*m-j}V?@aR5N5znTBFCYC(LEz^
z;%B0EZK*}wR<#DhA5B5(V9>@=7eRP0R)(s5ftr|C5{^|qSN?kRfYL*W5u{j^NZz`|
zrM+|-|Lfyj`CO+D5kiI+0%MNJr=7n)zMK%-V{<Cpl8UfMPAaf{TnoQYfrE`f7YQ)x
zD{1BhhAs|*DuZ`wAdsfl?chB<x(tey$D6Ov8i>^e0wV0xs+a-OzE6iv^ZdC`FoW;S
z$*(kD5$kwqzFu)4Rx(paLCaiW#n8ccE&kI-)A%0Q%m{JR{`Z~C5Qp74R30`dH{;V3
z>5h4EA-nVW50|C%UGrCOJcsf;QP;)AIfaMy8SYvKPE)duqz^>69$h5+<dpYI=r9Fr
z-PB>Rtx8FxN<a~jF+*Z}2aWGNs{!CZck%Q}%1yZy69aLphV0|OxZ#)03<lrRLtN)7
zx_w&j84tLz&gvEmF=Zz?#+;9+1C#hatDhm8)|#ziT5tbCCxqp)^rp(P*J}&*_{DBl
zNm6&As;V>E8<eYWu67y-D}DQRMc4Qmh8>L;5mnA;ZhZUUq-n&qbM2eT`Mk>$$el7s
zbkDsGm$A=aOdH7~hPF;iO#29Vx>WAUFF$9wXbo2jgy%Jnb)d34`LPt3`Jzu-^tTvu
z14!q5YK}nFh{KOk4LI9;x)!l6r-|?<u1B-X|0pBAK=;m;`?4OtHWM-LmgskWCZ)px
z2PR3>T>R~N1CrXEdB=FL5@&M7fU{!bYs<pDJs;(i74D0Ib4cefT9BWecB<*iMokSX
zC1lFUX+Y~2LqSWZk5RfDny@17I7ZFe_~FfMF^y_Ep3Mp|V&L8ZZzl=y+CG5xv-(1g
zg8#cG)^tO@@7cX_GM<RjqeC&c^8QK@o7OXX<7-92*#bzYGNr2iWT@Jw#u)B*?Q)i7
zTvUS<=7K^Umhc!?ex1SHNnzKD$IVbrSvA6!nGefO)V!YJLTN`CbF)Q<MaU!LM_AKt
z4l{_E@RR%34|wL~RR*?p4=}ljE=a}VE?t924_}RTJ$igz@z>W$Ahl|L(~DZ0aeSC@
z)65gAuJ4GSy*(yj<dp$YHOB_Y`PaQ2m7WGLJmq;~Z|>o6%0mSUS?<fKU4Gia?LNKQ
z=bvsnaCfx-aq;>WhVzlTQrbi%LEY?SXQCS8lQl#H`*{sm_pj3OQpG&SH$L@x0$t-g
zn<vS7Ke3r(eC_X^lC~%$`hcARk=r`^V0}f5XqXSZQOQb6lc0jQZMqbV?xpQtFn2jM
z=+|<XmwvKDzaO1X{CFvhx1hy6)O*`iG4?itb*;$c>ULG}q+s%HX4&1MFnds!7+hCZ
zU+5={xf){aes*pdv?Omx`4qI1Pc9cnkCd3sS&1;c)Xc7FD3Xq5E_G3DxMFHwZ$jKY
zUn0TA6a|Ecs^&Dqr)9by(<SNW2Uxw1%1z|zp6}msvrg^iXE%|7ZD;PbhGjA*_WWv%
z(4+qOtj0FSADGm5pPofYDS*5kpa8i*P6c9~Oh=2hi_@)2p0?Z?tf_1DrMR8(@r;>m
zU~AXVPQP_2+Y&fpCOK<>4&{vrZlej2TQRHuTqvhJ%q7J@d6-nuQvHOEaHQ-Gc`lL^
zvpp%4Z2WVhl3}N{^}f4L93FgsFU;$Jay4l%RPxW*ox|*Rp=Hk-<QKF@=yUfdymuT_
z$XC3FfRYFj__x}aN~4Fp&*4TWsK-_$TqyIS!dc=rP3Vn9zKq}JElBnp{Pu0B;$C%E
zw;6G>GSPurgoJ$ozz6F7b1J1J$tTI_2YwiQV6kA)wd@@Akl#-VVm>c^F#S!FL0i^%
zS^9d&W6kRw&*y8~PBT+$zAAle$O{^bl9Jr4xTrw>c_AJWIze7cm}V4koKxsdH2S)(
z0gK0nD<st8Et#bsEw2CACpIs5u<<unN5(qH-Oz|a)&jMMzRI+ZTq*f_Ye&H~_F;e5
z4ifLv=z{Jbm^JMn9*`}*8}(qf9)Yqbkt2?AD_guisVhC;?9gm9;)?F%iH*gH$(;PK
zZ=vEPzv4?}8cyqD3IY#5lH^IaP&IvPWeFzCDv@+c6*W3&!`^H4p$yjQ*trs+kf~T>
zcX7A#c~W!MNQVZ;!?UC^lDfD1U#ujFA>l9AS-&vj_xv)(=(krQePhYD8<wptWECLU
z-YFb@NC$4d`;{LXq5ruRhHe-^#nc45eQ|7FS>w}DjYRaDj!x`mj<}gzJffLkjLA;%
z?@7hqp#03{IleTBT$+vu4Yj4hy|)yxfQmn-7%Vn!XB?<DuY)aSl`saFsvErxfEkH1
z(N2I(H9MwRu$GbCX%~HzcU86&D=5>zWU91fp<gPk)T>F6X2>*u=-zsR1qrPh6sKX_
z+{-TE-fFRCD+=TO3RIkrSuR9HA-3?nno$WR5$--5zs0y0)ep!eV~J)X4=G--1PaP;
zt`aUf^&Ma)Jd*35Txw~482mu>kBT;DTGRO1hZ=<U%9n$JFa1Ojj}7Nqs$#_gEHS6T
zGB*Wg&YcR(&fXO0&eJ@2X=vym%I*B*oblhYk)ds39wKJ!6E1=6g+zzbMoD^-<HLW5
z1$#Q6FBoC$2@Z_=KF^>{{0LGa|B@wdEAmk3r|b?D&758k?+u@~VuCsWYP5TI9O^Y@
zh8{m(&r179tV?8Yd-$FQG1nuUv=aL_SF|7JeX9lUglt+l*xf-Z9Ugb-alR+yHYMg^
zHvC*N@{g~JI|(dhDH}#mOaA@U)^2OV?U*g^fLN&82l^W!=WMkc6fgOdeSLIa!2J<#
zkp~eHSakzOy*=MS`UE`9wJ6BYb!9gFrH-0TVT=viGreX0-KW8w0}KUaq94DHui1Kk
zG+lP`NX|=_?3T^yZk8=7$#$7%niWYK0``N6O%-kNWyRln%W*$^9PwT%lT&-T<uEFh
z$9_!PGos5=r+tn!GJj74>I(jx|8CB?5%wtNpQ8SH&VR7DRvhT{sy6R%(64}Az#(VA
zd^n}Ra35dOTi8(16h;wTz}0WwzAY2N`iyhl4A^mH)YN`_TE5DThw$NHJs6YdHWt|&
zccQ@F$=UIZ;k&$8+;Q+8adpKa@2m19Ttn}i3<OQFT#Gr3gUlK>6Ys&~1bj*;-aD#K
z3et-kN~^9IYwJL)Tv}{uyfDMAC1#h9pSH>9>m<s#WKKFJj@WSlRPKM*P9zW|OVv#K
zE1S&mGF4&9?-s=WEQ+6SgdL$2kW1^7CG`~HW}~I|>|d6ahyxIzcC+i~?&nm$fLu**
zCk?SadN_VREX<OXnwM2~myZ6owE=6Ixr+i0{XA?Tm==4<N4q&&N?;`RKQVzr|6af~
zY_NOKt4srZ$)^(3S!9pCzd9-E#-svB-jFZ_$~3GL8-9TmkJ;PY-mXhnaFYp6HNoj|
zn}b5@Yv=k?RU<iBVQ)vEz=>BA(qMos7XI9;SGA+ZDCF?g(Ro2}{R`hl$O4Z-$I8g8
z>r$7NrEmJbHX0tuO&3mI`M%E4vDWpskj87sb3bp^hYd&<YQ68ZlRCr;Q>%SR)Mc<9
zO~wjQa<y7du#?us)hgByA<{|Q4c`&;uj{CzYCgKCuNFoOSpIor{l?LUjsy}B+~?}A
zZVEAd3`_qwT9oj)@bOXRJ2XmCYzEbp_3_410%J7ykzMi-NGIJiLkl@SQ<TW7bn23_
zVaH=qpC_qhcnw7*F*#I76-ZFR+Fj2&gMPG?-MQ_qDQqTX`xKL&>R1r^2bkw92@gz*
zBmPt0nnTF(_ieq|e50*OYYg(HHn_+y(ookKK)VJq7AxdD%pw!~*u#T*9~~@QE*=t8
zf#~{?9Sp&8h1gz!e2MNRcztbM%kHMwbgfsHu3B3|L&f_dD~!TbQ_~ez;*%4U`!(y2
znSpQ$3JuG3>57L2DZ~|JjLjuQuR{WZevF`!%Qi|Lx2%)RZY+@VW~7r_E)?C$;d`(R
z$z%S|qCAexE8jLE%9Hr>$F<_-Tzbc4yL_6L*L#XTWfSgK<`upbUIm1?_|JX36V>B~
zCbU~`-)*0!i`yg+X+1{UOA=48;>(v?#zsbd*k7hsCL_e!fMzfY!HupuqXxoTCI;c<
zpcdGpM>vpc69u~_>>xIaw`KhTV;x^RCcjJ)h&jI#^dBIc#KccP^jgjc2DP=e)TMG|
z*T_L=Vx^`{>*?sXKo}B;2i_9V>v4VdTLFX<e4fV559<Q<Q1p7d5Be)%H)qG4dKC9F
z1X+i0{XOJMW{-|~<)5}^>=XPlKa;&;37f;h@s+Z4sB-i9ajfuX=Iw=K+>i@JAz4iw
zZx!ap?pUHV`N~u-V)UTv&2vA6-}FbPOAKi4Rgl~`7K<Bvjhw1Gc=`A0^}uua94-!`
zrNH9ndII*$T-=2O6scU~uJI_MJWr_Pn=~E%OV4ykt@P8tvBrw!WT&*t)Xw>&dwLdX
ztk8rW<}}Q;%`s3Fqc|;*0`1p}Yh9EQjUt(D2gWd2Sug7sH$fkSLtvbnT0vzz(gIqQ
zvjBIl1PK}7k!=O?8j4HqYUy+7Fsgpi4;@qpp-8n`Din?{E+fR`Ta1a8mW05T?aPto
zG>fHs@;62=OeNB>Pl?F9hV%46{tt6{+;nu&QzEF>pL5WY-snMJQTlvXfo~`q{J2nU
z9GBs7`wEX4;&!COG|0QTZDaE?2C(B|3yHfD0;rE51(eT{bg`hX{(iS)nPh2R<tX%E
z?FDaC4Nop`03<+B9JIN}`dXpiZ^K)F4ucl#gclz^v}{(aQ~&4Es*z%dR*PcC7qu7B
zFGJ8dS9G2F-4TM+T%<tv<D%bf+~pA*9Q*^&3X96IZUXX$)=pZ=vF2PFCO=zWzjA!Q
z_Mk>8_irbmUGrTpb&Ihqc;|ww0jo-j3WBd*KYcx{EXt7y-QK7>Oy1>0Od)Ou!q^40
z=k*>33Nfz*JI(aDMuuH}tY)(1GVGHOasBl1GZB!E@;(8T)ca$T!=9;25QE=(bD&0n
zgpN70Ml&0|gJFMDf;wb0yc}ROWXDJ0AyUKok6JA~S>kP_-?V#-g}zgO$0Psu@qiAG
zJ%f@)A3GJg7DYIib>+&HgZC5ZCNN(haYHG->~j=mj?T9Yj6oP!62L5Di!3NKoh)~U
z0$Q&_)SyEcXExp&P501>kaYM>(@oGM+)MHM`X<`W^NlPBF@t=f_A<LAeq=HD+{622
zVf`OJ5rzr>Qd5KMpJ_#AqqL+rFLmp&2KEe;%YE?a+F+Ewj+@J*v0h|Lsc-JNO?{v$
zM{Pif*P&v;B291MdZf-QrM)J^Xg;9#+D3v>%`~r|?57v^J{Kqa<K81>y7wlal}+(4
zlNl{=SLQki^+he_sM?LHtCR&C(8CxrgJI^bs3zJ-MR;@-%`GN@jlv0|aA*WCVO=Uc
z9a#VyscOJ<JzI0Yw4*PH%~Pc@%KU^aCF$f;Y2n2(Q48S*`AAKzx(K0CWwx&8gg$TL
z*sl{(Zx^k4+WAetJas!CGY!26j(YU>{BQ-hJww1>>9*s4@`wU!ih1@NbOftD(@@~c
zyDHi7H3bBKhM~-@`}Qd0iYgsW4lS6VAxwTn?{~#&_i2Nbmhon7e-jS&R3yrT;6M6@
ziw)SJ(!Ha46QR|l{fo|%;wo~yv5bp+BaoNX{@0BEKHGf{kN!xDUN`HRsD(_q*qqA9
zdf7bao%)yswuEnr)9=jMTe=#*#d&?#3;Jlp*dLqC2KfF<0W){XxXA^{n9$AMH|3Gg
zZF)jfhm~Lzl#k!{GcA97RkvI;vn2^~9D`S~E9fIe)=$$~(xI6f5+JHCuGU8@CMM^8
zN&;OT3toTn7&6O%KyK|F6R!7P12who02vG2zh8wLemR6=V1(hoZ)@7wh8h0u3t!Nh
zw6Rx!t*!DG4g<U39CAE3K3U549v~J!_BmGj>+pwcBlBf6lZ1W)84WyT-h!fAoUIrc
zI|P}XH^IL16C~>ZRUSiTcCyx|GO?Z)jDwwjG>Fzs4K<s!2;#WS2Yor1z3QX5No{86
zt<L8QJ$D{S#X*$(Gz#(?6Nf0on|D6YgzZGe+{nbz)K${PB-?RoRk!V%bvXDbb(I79
zd)((j&?ffez1kd4Ap<b_Mb%KZG+1ZmgbuR(i0X)3q0*U}R!bgbXoUhd)t4s+rAbFm
zzJ&qxBX`s=5Uw2k%N%-Nu4$E<?x=rPe3vU)f|3M;Utw@T7WL5!Z8uaa%V1ctfO<5H
zdv0#-;h`~=us0o<V1L(bivzm{dOjuuFv6BAyt^Yk?&|Y-)82Qus5RHF-*Sm+kygYI
z>LtRle$ZN>>AI9W&u4o5OkbRuW#1o?N#lXOFEK0FO^4@o{)a>e0=2mSm3g%ldQFCP
z>5icx4U^QwRGS>z)07W&WN~;pY&f_E<l9t+Yq!KcIl|;#($k4*8-B(U&9*EcFH-A^
zK4l&5ahtx(f*1?`#}jfvgY~G}K6}KH!Pd=`D`g7e9~LAa=*xBrGY-x1dr?{-pZ@XP
z>-^VIQ0nGm`7ax4aJBQ1B0{NQOl`lRhOVaS+mghq_w|Px>?)cAWvsmP)M&?UTTkJM
z$dM6$od51V=m@sh3v@qUSIRF_-}NtTRd*>g6A?3e7Dc3MQRh*cj<g>~8i&!cgjS8O
zCMko73}twbsPJWAjvu>^rPKR7cZd1yXh&6d8JB(@K2@-w<@_$#QV7b^lGXqFWweDT
z6(CtGxqSW};K&(W)$j0(hNW)#U1_lF2W_eCrtT6;k)~gcv4kDOAT`vNd;J6tvEt%s
zC(>~t*)|;JS*aRa_N7n@*vuDVBf}(wh4O8myqTQDj?%VH{FYSAF=no`{b7cPgOr!*
zLoXcQ8X$!)L_`N?1}cvf<x+bFAx^wRD=J&<n2<5D>M-=>X9>_ceGO-$T@HpI5`qjr
zlSKT^S9R$mJL~2^g&L^ZMV#L;`In!+N05D{m>4QL&(y^i*aLH`Q2Xr&H;}iq5BU4%
zJhw~R9^W848?baZH8u74_VpjVB?w2JlfTkb$S=rIPRE)@kJw;~)3KzkP1gRt)<I2x
z+d6oO`H*mf3I6DY*GEULygvbO)<KR~!XlN~yWPI}$Ulnn34(o2t13XXwCr~X<c^C4
zi5!aNk4z_efg@utI4EedO6(DVC&R<J)){}h%BQ%K3~VacIbv2D-YI=s5}KUM#@Ct(
zLc^SRYNOJ1cEWB5G~03Q^wJNF9>9s1RO!IurTzPO#f0+F?5x~WsvaoI-^9ekOdZ+)
z!63AOIyt?=4uiv_XcVk~X-lZ$Xx5H!3gs_rRTz6;VIrRXsWEll)y3`Nbo$1LCq1*Y
zujd07-3-?I#vj$>Nm%nf#nnnG+9a7KFuO1zf$XJfP>b&Ypyo<BlU!g+eM*dI*q=-B
z$uhy>n^0nK2>gyvEXBQe{(4a&AJoaBG}(9{)qU31lE&;Ec=O@c#u|YH#b?)%&}{#R
zTl@}WMN3WY)xI!ezz@g)KPKtFm9WeFW(8g!v$0VHt~||3ay$gRe>Ob)Ofeb`?9cn`
zeGQZXD~37pzK820B_9RSyhDt>Kio>uQi&f1+Qcnot?)_<xS+Jx2stlx8Iv(D#2DBv
zdd3?sDDw?w;n1C$o=*TnVwN%21ZqJ)zvYJMvv-5z4}RP4J9^*!-lhefeVt<NG|wcC
zB{+Ak#`)VvWk8U!JZK5-(B_1DUW!p-!-I&vdq#AZ`%%2qr)}|}VdK(z-4fhU@Kugh
zwx%zyc1`fZUYmdgxyHmv-?+;Zo}i#=OmteJ`!7l?U|NMLCB7|3&E1O`kuL4O-l55x
z@J+9Bs@Ie9itMqqH{}&cxxepCu4LP@)4ICi?Rie43Eo;LhM)dwln!62V5+;`aZ5hN
zm8ACG!fL|P+rH&jLLH5HmZD40!&>vDW%>*nr-*ESRs`yIevv;17@P1xNu*W+jKUbe
z;D*M=Ot0U!eFCIT(#>Y^D9Pu=s~d9Np4VWJfca)w=g56I=O*x)1g%vRy>op7;=zD~
zV%aC$bN3v5ekXmU;<!{NYfU4+^-rF*7_dHF9CBKGMo{YTNt~F*H-mEf_h)}jx9gBY
zo~GPN@|G4*VAFJa+gNjQdis;_T=8%^RhIr|Oabc$C7$FfF#EYG)tUT46^7j*h;v?P
z^3=OF?cDgHtC-Ja79-m-dA~!b2ppg_h8kL92BKcQidJiwcw$JsO{ipRr;v1qR%C2!
z><Wlp?v?<#3v!z`$@n8|VYkf~WR_lt1^mLVK@SLt>{Lnbv5|Hxb#J{V@Y2-?F^2{C
zLzqJ!-N7677_v>>p|dnM-tk{lnjP59_n#PxB%t7415ftvlEl20V8iJpO|k4&+C!09
zj;bw1<tU#`kQkii*uT&fVGn?P5cQTQ=mB@UHMzp(z+T1o>tIs8#git{VC0a~8IxSJ
zKrU-rTYl$EbdLe%La>c*#7t-SbXs3Ew4%v7Y4Lo2sBT2lG!kxP6EB24z;u#(y5R?{
z4VTYeX`Hs&zX{dX7J2pN>?u&nfZU)DG{GWTAPc+bCj`cUYNz|UOCI-P+RDwD5Q9e{
z$0=;%!HACrHWZ21H^ZUeqt;5@{-)|JS0Vo$<0oyEMmqz#dnt8;B`B|t6SIb@FIWyt
zBQFSgh$>Hb^~aUnY%a^Eql6w0t?+Q8Zi#Fh2ne!)uEfJP+l$i%RuqA8j53St^zpZu
zne$U)D)B9;De728()XbnPqx13%!DTq(mK9Tg^?xTgqD&lXyNWQtWUeVnzo)sSbob2
z$ywm!h+jT_Tgc&`U;Q_#=ewS|e^rGd;y6eFfDoFOjDbFSE>GKA*&tDKf8F$Pl`KvN
z_p2vf!eUxk{7Wxk%@4<GIxd5<CEz~2IK%K&86045r-+d9vkgJ`LPY@Y`RRtaIV9$I
zwrFjL^%7s4{Z&QMc05vY`q2J7Cu)1vGcJ&QOL}`LIP$O>StSi8>1+FwT~oLHJR+{1
zCrXOu-{zGTqv!G%?U%^lLTFzl+)ZGl46pOhHGx?Ow2AJ3H2OrK#N2N^8L2o3TUFzf
z7Z{Pa4LLqpw&j#{ro$`{aAcY_DpPVhUt!OcjvP~bn&EZza<*rvbv+uKMp{oIb3ky`
z-Z@_Pqv#~165iR9RFIogxXwd%in^Fc@*w1ONy*0CB1l)w)a!bRtH*{qIf1@8$SY7J
zQGq}5x4C`eNr_}>`ZY@6Totv=vnEun$}Bqc%1(Wh!d#WGlK#bK#+5j+1-l}WFFWSi
zI-hy?RTdXF?WWOVF?8+B0?Geya>_cqa&A^h3x!T%x=8Q?pP%#|Pxi5|r2*6rIh31e
zd-&zMyd^#;M=?<!4Y1fr{qnClU=C0yWVH@3`JUFz4;$k7$`!$+%5T|`9*OgmYICO;
zbN|Wkoqg=Lj!b(w(f*tKqQ6gx=saJgPTqSOd9`Y0%z7dyqViXjq;xuE!2XEoV@F$C
zTN_X%uEheulJs_H-Gu!!mA3Zz`B!bCSo>!>Z5;OOh{dl*g0D~MKo~t^PHBwR=Dp-`
zz456~gGIrI@&^$)PYxZ1&#d)2wx#KJ1E03tAy~MT`blo^uVee^gSi0&VMa6KZYY>F
zue%)hoj+?v7m0?v2cF%3R(uceMJO4r)wX><>d#RqlW-IGS|(JA9J-W%5+t?w;5HEv
zgyRrcIz2sA?9f`e=|Yv5m}tr@sF=%qd|=hC<ScKMGZLgJC9clZigG_k7dkGC$mSOp
zHu{FDJg*b1yK%gqs*#ZU>;=l=X@YJo|1fQrT|zDG$cJ(!c;7%ay_|GzDOYKmDCL-E
zT87uMCm4n}<$3ewO*eoUVV8=eL7NoJuiUA^koiV~>8C~n?X&$@r}rSi-*P2pozem%
zTWe)1qAw=;L|8YymvWMywf@#{_##xS9;Y{X>b*9S2nmXmZ(Hr$YnKy$D7Y{+MUfq1
z2o<@MkwZ1xJAY1@5WfAear0bpJE=)@_FIHPdisYC%`JM|FClb{|Gp!0R&qQwXmJw*
zv|`08wO_l7g^OILPK=MA*h7KXpAO(m^n~jK`T5DD_Bo#4I>X=EJRh++A2I!6f7j9V
zygLR#y_wQT5Drzn*JA%hTma%<yYi!JUygealD2p7;ttOS9mDwn6_I-Z)4X*{gP|5e
zglRi|EI_}mXGC$ogqPBPr=GUd5UXBIn#VNH6gF==QjU^jIzM1EwcI80S>@FTdhnp=
zJGI~bmTm$-K3Oo`VQETY5)xhHP(Mrr`qwiv0lcDrmiRwEoDGC1W1hgViUOT|e1U7V
zgRB5$08|zt`I`F96%S!?2L7^OnsU(&?Fy@=(O@uqPG-h$%KfyEZk|pfbU`XS!x@g*
zSmTL4B55Zhfbr~lmjvCTkx|__QFSOwlNSCGTrJPi?P)?gj!jO;<#j>`II2Idiz_v^
zwvj#ft~XC@ATM^}Jv`3}ewluj%(+ZL6)9OTsBwOR2mu03V0y0a2dZ-}YS(P3a{sv3
zH`969nSeayVvqj&LyE=sy?Y@ADegM?6%3BGBDSSOTm*s!!%af@4Je~s13+KhlVXY$
zZ@|C=uWy^0J@`QLkNtM|vj0r`?Ql;FAkg`!5w@fOvX-L1o=Y>J^NA~*7%JAjyLne3
zQ>*F7%#M`vi2A)k?#h<GOB4bhA!~`~s%y(%M3Q_GfPtHjEgjx8ChA^98b`;((DJ(X
zGAT23jpy51lG}HQV(%&O6x?CxGK(po#NZ&2eO_I$9VY+v5Cvqf-wTYg|A?8vDd$W3
z;vE`kF+bBBBhDkbJ*h=Em}K%QTi7ivFCu+`I{4gB%O$*o;i_fR3_DRn<6BcO8H)Ld
zyz1@B()<In<%Z*IzRnPY>fC!jZhahXvi;LYV>vM|`6=%l=f2FD<^4k9j)XhHkALTL
z&YX?bwfc&S#i*m6NwsA#=+smi*bXykoDe0<_!ic$wn>P1Ld4;Q|My*Ay!C>>SN7wV
zRiIY|x~&S{6uFL%K%;Mu-$t5kPzak_KvQ?e@$J&LZf&!89T^l8c4)nv6{c$M(5vvJ
z`53_GNvoE`9pCqhlyGRR;X50?_UX8ee@qfF@s{myPqYHGHa=$H%>a7{ZL{p3QahyC
z1c)jMPBM5+#%Pv7Xc>yKuo&mJe8waNM82YrhR}v_oti#{H2<|6xF1obF9>wwtA_j%
zSNlLclmCdr_bNCunt@%Jh<+7<p<N>z+qtOa548#r?7Z=qX)56>nhseOB$wL1Pj4PK
zW&!mgBOFu3{j~n!{heTiYE@S3gT;p6KFuBsKJ3h+PpIRTMvAPM8TQ7F=jnsRmJX^C
zlWN;Hl7FzP5c<`Jz8XM;99JQeGcip3Y?Gqi-KoY@Ao)g?ii#?Yt-TF{1GKMnG{4!|
zeOHo;uCIJFqeuV&CUuk!54I?aZw&f>mN!6@-&E8)UZ*Z@jcd-RV{l-A(V=BLO5u2G
zT?)h=Gj}?}=yyMCRhEcvw!=a?8BM48(!5t$Vkt>$t}?;$Rk)((!`<Btxyl*)iEtie
zJTSIwPns*kSFD58rNm59DBWq%q+j|p56!$%?|Z7MUG|r-%Nr1P0ZQ>#RIfgo*PadS
zCvh{C*oMqIa50RLyHN68vr{*?UYMW1M7@~<f({I;K7O3b%Bs$*t++akgGBeX9|8CM
zLBAMG4a%$kzn6|a1tV>hY9;dv24+U|WdzFi?q$gakB1Hd!=WA88wK*&xCTp)_N5QC
zweBf8;mQyqkZ#MI-xtdRHm3J&8&5VO$csZxnnh6j4;h;9tc84bZHj&^F%1O+>24qX
zu;&5!cd3t_I7IfD5}(g@-AQLc<J;7mf4p0r>@QybOyU4yu{9VDsvu<Aj1Y8f_GL>v
zbyEh}qJ&)m9;k6``q?+vjszM0duuwswl^!i{TVn7>apJp*bWxn;Q;DFx*pjjrVgxd
zQ=fXewoAJeB}z;Wet3keXCvi+=x~XAloRkivHY1_4Dqa_iYHFInzjRo$*?yok*!PL
z>k-i!o8os!tto$}Z3GciHQw!_(XKr^e)%_WOO5lyh{Kt&2!>VHS%y-iZ!Yo|f`k|H
zRv^l$qVJNC-Ae1QttcG_Mb`f;lBDJEyK-b~c#nrOxN$Jh=I~OUL|y?tIlBl}T8<JV
z5AL+1MGl1l0k7(HqN==FZ*?lW7=pp27k9$rpaeRags!AgjzMz}_73O!N_p~S6MBSH
za@=|E7<KcAB6ixpz4-`d$`PVeIf|}T`8Z0iaw}Hok5JL;Ih&7Rq0Wqu4ir2(**d(6
zY^AZWaq~z@M#fP*h8^#y_YLYi@bf1e%JlTVk8B!-Xns%1CMwm+iq!97cHao<JRhje
zjJkDn1cN}>TQC?hh#dw?)H1wTI$^p6O^CS9PR9blEc`+z(WQL@&Q!HC?lqJc52{}D
z8yq>rcRzq=0~h)uAczP>gdED*r<yfu@cyA8yL_o>4DKu;E&Zci6nlg8eqQB_{cb9G
z`)gMGBP#m;cZGmGMz4yh$A0ao_ocb263|ZAPy*XTy*~$+P@(owh*F}EbMRFRJ~T9R
z$mBuCqi50>ZU~+Vs?H!B8@;AAUHz#6t-$O3e&8EdCV8b1VJJafe!f}3lY&T5*F5GZ
zyx!QD*X-b6)eu$=Omkac&~PD=v8;RiH_YwN{5+Rfu;ccHz(U3^K9$&v;>W1k1R7bM
z4M^Yr^??N<rK8!+?ag|H-^G2#;Nb^=-_Rj|f%~y{w9G<LuX#f)LpxBfXNsa|S7<-a
zQ>-YIg=B%X<HME{Ve;32>`(WSUHg_ColQ!F12|pywMp(<AY3ipe3_eJfBh}jq<^Bg
z`RscEQYNmpsc+%CI_J3Q{j&W+y3Q4Tf=UP(KjriypQyX7!tINpXjrceFw?2FjuTAM
z;DWdo6N+HGmpBz{|JVB@b+!|t>^)}1Z;i{Ny)2cq6IkOlxVj3Od~Gvh{WdseL2eTp
z+_P<#*aHQxtGv_sG`lCj+QrW8y63E~BfpFZSE~!U=jvORif`r_41q0B_^H-8CU4e5
z)504z#7^srS`nfgO9Lr5XGeY6!%xX^riX~S2lK|1iQ0Ux<|@KpzvkWy!74x_pU}pJ
zV)Y?tFBWHJM|{QykS6-TePJi}f8H(l`g`p{&SWl>_B2kO0jg{s*WFkfV&&sO&yVtu
zQD9=B%Xog2-0+KO=eh+S9rep(g!0XezR++j%MvXTl~V*lm7<tUL?9zO`Nf)xu7r+!
zUMG#y+o<8>E4;Ph0?HldqrTy~a55zHWqOYbm#Ixcn4@Et<1VPtIqhi{3#wEYKty@X
zC#L#Y?7dua90)^;M_~i&H+6)}{=WmXj{?1L<53>l=G@1RkFewVl9?V>rXCM@<=?)2
zi*y|~zyX?n)_Ml#_`c=WVuPYwEFs)8mEj)uX9PEdrfSx61{$Zu_S40%A4VDTfMIZC
z>6_|0;o_=m@wcgO{NDJ<SKHdxyF&UOyuHQoyr&4IlEmUfj!EfX*<&6LfhnPAyrA~a
zN12I1Z~S#PF2V1=p`<Z@lH;wR!OlZ$1`8J{_tw$XI$jCF3SP6Iky;{jQ&x7Rwzl?o
zd>H1|&;DHJo-9|g*LN)KDIHTY7lfBO2>%)UXO8oZ7x-7;{3~)k20YwrTFduo&y+}0
z?HOV=RBQ7p$#JU>lPr_?Zpqy%(*8mXJQw*VwG2*FL8Wmi64WTSh<~_DRza$C2}?W&
z7(@a+fNiCzYkq?P!uNkR6o`_eOb?Is_Fq+D>ndoFkLh!0ymjM7#3b&37?T2e*b=nh
zVarAJ`j_5kx>Z=tTmGF@=F5oHx*8hF=Q&hnE)aWpjAPNVLW*a%m@GLtws5Q*eZ}PL
zJKI~nOA$9Mmyut;-O>83*qkT%VkH5-RzRd-9Jnb-h!VM3+#O8xcBefxP%apxuma74
z%R=Ku`4mZaB_YN0%_DSG4_06P`Gg{qH1|&e*w8vt*Z2ou27#lmlt{Qz{MVs%8-LH1
zWVy4p&GQJl=m|Gw#Z`{75CdoE-H|p#bFNQqbc?A?=)P=IQ`2Lc>p!OaW3!>pv>M*j
zacz_rCC<O|_+p=*&Wu@rM<?m05P7wd7;5)D@@M$&zLDbY^uVG*1F{jPPPV<uoFZZN
zYzNOje81u?7p2kdqc$xm-+KJ?soe;22wW|BI|B~Y_k=gW@Ik)2N8F&*6kVl>Q#UkB
z1kM(${rJO~#*uEQQ`gO_BTrBR@sI!ivll1i3}C_2HouDUYkC=15FU0sR=kBkjByU=
z&K7PS1#V({p8!m%5?~vZilZ&+biZHtNfsnEqKfe$x7a$&BcX#8@P4tO-w-m&w=iuE
zQarLUPx82vq<4z>p<x<9c#7B1r^rK(rFWebw_!)4e@+k@!Vj2>uCdh$HS?I>4n^@&
za&mA`GpSQjQHj=5FQrU`kmAO{<k8#L<1dX>COCOyNBWG*CtS>H{GWxSwWr!$nav!n
z0@v9CMBhNYJf4-`oybnI*hBo{p(C@*mI`3BFMEI7tdHTQd<<HCtLJGp$7$Tf(c%?P
z#SG*r5tDULuMd*Kz9c-oUz&R?VFA|I-7*@aoX6#;zd_B36|K&H60Q}qSkxW3aQj~R
zM6lu*`RCH^kg&XPVrBv>LX##lEFeJ7ZXrQXy*$|H-F<KGUp`~j!9Q29@Xd>;ppyf~
zn-QmD@@Q~IC~k67E@g|kkT;=cqNPCgN$wQI|JoH%Nz0K5sAQt5%<|#983p7xInF#7
zD$PnlOpHsoZj&$HN!C*&4oY#FTW6k^kuQy@EI5<Ro*V(6A!Ngn=Fl_a2RlzoF98Wm
zec^;Q&7xlPcG1)+^bVCfruPxm7>Nk=;GavUL2<YTBo%Clh78db^~8V$R|-7|?9eI^
z%I^Jf22_8~Osv?mF5Y}24QNDF6@9|G=h8<j3X)g%{}<?yg1M`TT}oZTj70gYhm&wr
zt`iFMb3?9R@E2iEveH>3rc7e5%6X@8>(H%@jlJcSmA|mmPXHdj-$x~n`Nv{c_?e=y
z5+N=;7>{U(nmPTjzM`C1^4F$PCF#=3lM;DD`Iq7ZR#`o)=INBog?f6-ntwKQrCfFl
zU~O4`;reF~D|v8(iT5&A?RBRPKWM6qGlr2K!dpj~1peow0iCjGzzCD_KX^WEeD}Bd
zw18Mfi?7*2GcCHzY}3-Lu*msRU8T|++NkqXLBlZD^OOCL*V2XtuXbSn^jXne^`zmv
z$q3hwwaH8fa8401<AN}jG$n^xb9zC2U1+bct7q3VMz7gD6y^F4XGrNay8B^?bQ%Zj
zcLQn+7ye{>>S^3@$gHOGmH4EtKuM2d4{rJ)u}z$Sh&L{SxaLFU|Go=Jor9qK?DI{e
zeMHGam~oSAA(3@fP=DsQY33|kXfjWWEynfwEe=OPqDPB=>t4<Peu94l%xiX7YN1nr
z;2`2u$xv$JAy(xx>}S#GL4o)0OTrRRkUQB@aHiL`215fq0V>u?cPlCQm!HxpY(Kv1
za^KBu(_)-AlOPzYxX5%3tq$hZOKHes6MP^Tmj5mx`l!&T7z2!J_M4_n`=OB2gP%&-
z_=~<P7O0Of--l08i|qZvApU|v0STsjFD$3{tn`Pfy!+8Fc*Xx7U7MU<c)f4iDI23{
zL7P<^e9aAE#mc0cEhkv;sVa0fN$by}hnr$3FilzdV6D?`cgJGxWK`Tf1S#?7ySH30
z7Om%4O_5&tgp4B7lVd48icaj=d#Llp2-E*L6^04`1R%^SQ0r!`y5^G43WW+;C2I9o
zRl$O%M@G>1eD4%6A#@eIjw}GJc3f(5g{Sau;=`vmeM<O<JLudXlP5teuo*06A`x;C
z8E!tZPn;()xw(^p0P`I+cD8OL>gJRdD?9w|F;g+fPV4Bfz5lI613JrHdqY<CFmbDr
z;Y!ZHN{gk~3h2=&*ANnP2|T;$Njr4FW-MOWLv}UqhTrBma<+{{_`hp2a>hhRzx17_
z-zW_;T%Xls>K*<cuD&~-?Y9fNwW>ixOQ}6Nlv-7L?@q0@R>cUdQbl5~P&2k_e=V&|
zjhfNgTY_k<YOPukN)Qo>7%|?j=Y5{{dHVV*AAk6K-1o`3&$-TZt}9&}+R%TeDbR37
zs`EzpE0?k9)inY(1c0_E|2+3crgU$x-?X-<FBK+LEI#cUzI4)UO^Ds-GHV_Z8km2>
zs=(OfhmB6bv%sg0n(#ft^YqO%Mo;D>MDDg>crmY);ivI5d-PbL;jtG)neVheQAEDw
zAN@%{!Z&w>N17Xz>=DZ29Nk{tg1(Dg8(<H}pPR`&X8ry5q;)?4EHsvNjJ>SrM*^>z
z2~XW=?}WK3xeS0BFul2X{W%-UrTOX+*<6iEqcTEHPLA40h3!%80~VQ8*$LxMd>O8x
z@|@*&e$?@+3s}1nJU+h<^Lp>|T22%ZSYxMovAJgIWjqIb;tSEVvgU=&XB)Tn1L8pw
zXI%4x%>ley3lFpZhGEHvG*nil@Rtok=1jE^>-0&qS)F)psE(9tM_3kbMUpV9FetAa
z&hYePuiwU7sPwiAV3z&ahx}MBu(h?-_V)RrY?s^oMdCncMZZxzX5GGCw7SHa5`5|y
z%=~Y1K1S+fI#@<|y5ve9!p%Yfv5xqLa)>1P(gfq|+Tx;9!Iv*TQQG_-;6O)=jpMUt
zuDw+p(MMf(rtu=x_uUaZ+Kb`$yHc0B=s5@`Bj?;0zfaeeKl|vFhwv#D)_WFM7T1)i
zFAfR^1fc`1D1axQp0b(ASiBS*G|_&03RB(@S|{#on+J>=9H%gb7pQQxt1o}xe4Nj7
z0<IhX4gZffzcQcp96XJ<lc~qqC6}wr6^%KOyLNa$BimS8K_ZVGOyxpDW;<dA0DwVr
zzHB1DTsVw}Zmz{aw4&ua|C`Nl=A|5Fhj&(tRof1XB5WsSb2%;^&YC0(wf^$T1pvWX
zd@xN!;Nc_A{&^2Gh|fRzuxy7%D$)lS6i@sU)<C1I4}ZTB@>OkbVVAmYD?|Zc1!_d0
zH2DW$!s!FJH+Ff=em1-{H&!2|!G((J6rKIp2|cO?;*b4CS@y!(wjkI0n=y6<+=zxy
zJL<Rf#br>fXj;XS_yAs=2|*>gwHUcijSzmvsh8cfmh(67!WgR~wKj9u=ZQpOGd$vY
z`A-cv%iAF3$gOng7j%z9>UyQ_nEoBUfW4ha)nx7gp@RVLgC*b}!FbM+0SB2VZ);#j
z$YE>>A0Ezvku~1j+OjkmTbbv>?{pPRxHXcq`sOPBd3=5#E{Ilcmvt$if$k(*U~I?I
zOxun7&)yC(iRP!&vfMoBTqF=A2TVLYCg#pu1eUI{jy`4WeL9ME0O&>oPp-jP^J<VL
zE|Y1^fZCg)7N<x;+TI7RN?;TkIs5k^Dse_}i>&Z6IqGVJfAwO<LSpsKu4KM{S*vvY
z&05i@lXYimx=e+9{8w7F%jjc{-K|mEBPBpv>*!`KaQiGtvHa6gpBU&66pv)no`|8-
zL!6G2zSp1h83Bj_P}xUzjb(kfQAiV5otY3kTfeFk^V+`BZ!EgqUEYCt8RU>WNC#6+
z>9HiDN4TYZ0cTuZX;0RPlLHr@+~@h<xAxVgD9oCuHcuB|uf<y|ivYuTwD%Q&sId^7
zw1A)p{kl>{M1JWQY@G&`_oCFQLRLn`0iS$w2ui;*POZ1*;e$BCySCM$V{e{hi}P!6
zqTVo?48j?)0_g=siyjaGB=Jr1@L4brqH2_;4G18`gaarGOp>eiy@Slo@dm1wawbDT
zn1;4bS9MjPilLalPf_l?2=~j=e?EWpFfmc)ml8M=iR6kz*i2cp)kk$kV>EL=5;g(<
zKEP2vm`z$XCf3R~S{*KX&P4Rg(9-rYy1OMu8?DcIHRe>bb4ANd6J_v1oG2~nl!EI`
z_3Q(15~J%*rx<{NeKZGjch+cG3He6?e0(%Ajnaw-tmGpU495XRl}QcY*$lWLeRfF3
zqQslV9F}>wTcGbG{(22Qo_0R#ak6~X{)%w`u|I<<yz=>V+b#*9`2J(u_UWlUfdx9|
zMu+#DMWbSn^8O(^$Ht2&w>m=9QL8?q{^@8eO^AIGE6;fIA|%FCyx@&;z@=y5A281G
z<F1#0okjgww(1m?%VDbRj4BF)X#vmRsXW%0naO%j+osh9Qp-yRZV}f3K$EGt<f*$0
zi%UYFxh;>$k9?dCCQ_$I<f_G{zRq7?rN^fXZ_4P_b)k*3OCIvb4F;$w{c;n_$8~kx
zA`<vBGc%<-Z#V#o;XCs3^6&4bS8>GwntpuN%>>yZ4OYa?mh}(Ws;Bw^DU1p+fepP+
z6$fHf6eX@*m$;yVt|Lgf^l3_d)P2PZ_mLP=isfOoFTTj%6)zyeC$jTwa#jo=QrBf;
z)_{U2j#nh1hrr`J#52O|oGtewR@%${pX>jV9|2BS{A(MA4DoIH$T2}ugsOe6Tzev*
z<8*vQ%29efWJW|LFzG_40PC_PCATkcu;p7UNtV~mvb=foChJX6XgC?D^jDE`&Feqr
zd%5Vj<tEFclZM5gDBFF_4@K7;F&Gd6PCTQwIE0F8X~_IOBP`16x&_zyjWS|Ud@Sf%
zV{&X^TI<IC`n)}7q={o&{6!&5C5Nwp0qp8k-9HQu#L<x?ki%p<nL~&T@ZFO=C*$l)
z@6D1{c7d7xsOf1cAYAd^bRThD<zN5KWt1xqF8Z}uGEZT~jrgyA>~APQlC853b*S5y
zGv$vGW)0l|Bu+BluWV0xXKDGZWsd?w^YbwuhHDu_nE;&t2=?Zrxc1ct?abx9PX$Cv
z-o;#1qRo?aFF<}JD!f9E^lHzjD~SB4eek>3P2>!7JJGSDOYdI)Keaz?oG%D7=y4|c
z=dkG$LiP$n_Bw$E%T~qX&+nURJixLBx13|D9-iMNsH)lc=at^N@QkGKZ8!_?Jk4H=
ze+WV8VbEMCJ&qqfMgQB{F^k+~{}LdRU~Us)LSl39n7Y#tXv|roj56yo@PAy%lUNQE
z!`l1Rwen~81eaM|nqlekI7rO=Wv$(>lrB*)%2_|uPwtU;ORhX5)~ZZXj*;LMP6ceP
zk{8Q3<=D2bk6(fwI;BCv`7-WqGgS_rz93U19z@LW6~sLKc&qLYJp@&m&Fl~(@DG$6
z5HAB~PGOQzjsmy`dYonm3u`Q|Py&J#_w?dfz*6*o>Vz9_u;NQ`C&Fc^5)34{OFt`h
zc)C8Sb=Z~7IZ3;a?b`L5jk_F@ZU&jVjG-KO%4M?O90K*opNCpo50B*DSuDxIo7;8b
zQ<n^5Qf*k7??>KD#y;!l8Vo*yLo^biACrt<V0N7AWJH+eC>{7aCd$p(NI`Wj{W}iq
zl~upKTPzaflC#n!x;1~Mz_56=pqh09*ydZH#{Ye^<7l6|w%vMkU<<7KY5DQ6M7Hhk
z2VJqGTdd1*a)o_-lpPO@1e(EfXPM(LJlIF9BiBwdW&Is(RZL#2wqYI!<&~NrAF#7F
zdPBTX+WTF6?*DZ7+!+yDIGYGlMiCblR$RR)xjF`y4*@j_h-}jOulb&o5VP0Wmi!%Y
zgYQm5JK>zz>KF~_T7~;xYUy_BwO0xtlVfcLtj_`|f7)(NI*}o9QjWdBK?TBRe>PL0
zUqX3-{8_7+zKA(5f2nEz*ius{OkDliwON!4$YuAWHS{CL4~b)mUm-Q2^}*=mmZL+O
zcg_z0*!E8#-2ciDi}{}~4dlnM2qClEkIA>^!HL*WEkI5GeAz_w4QCDZQS|%lTAX7T
zKDHyZPO0`S!0BPVL=_O6wC%Jvnfpjq5tE+ik!jukbE>>$Z&^#RH6;H+J7G%nh;};p
zZ1|5&ZQ*{$bD%C@b!Vx2q;}!ch(qL056p`^L)J`5Pz+CW2;{(+^0t8vVd3yygm<*j
zO>tZ(!nUgQ*KSyP8d<>#UjA!BWSpU6rYwm8clw&V{Ki|YqYbSiGJdZN&)fMlx^0Y|
zogKw!g?BwO?EsaVi=%ww*5-c)kFSbti<Nc*)khwaG5%Y*-e4KL^uz=@Ss=t8kZOmz
z&7+G>cw@#5s9G=gV><SM%|{Jj(IWrs>WB&}j{7;A%g5{Gvbow8aB4#83wAfaPR|{=
zk%kZ2zIU{FFJyw5k?*3L;xyk2I2fzLdKvL0m!p<Npe~e`A+(&{N>MqAREH&tTzb|#
z6QrQ+mzJfJCZsGYQg?UY6b=&c^LJYWe6Pj*1JMGa@|0m{U2kFQ+bJ?VUw`5G=)f^E
zynV!I`fi{iV<6P*@?xA?SdER$_d_&WW6v_&ec;^3lCsen`4^V1;ilztSKNI!I_=z(
zGI)Y=cC7FQXKq%yX+Qa?5qi5$Y*)+Q-hOr?P09U`|HqXXNSDghbAYDf3g^fK3Bc@g
z);kSvkitH`x;>`I;|)TY0N&;QdtxpCb^B>s1aizeKcZunF%f2WX-h*tAPi8a%%SEr
z2LluB%&r^Z_di~r)zuh!6LFh0>_nmn<X=|Aet-dKNI`S+fy3(ykfTI)MfYAW&LWGT
z%rK6%fP`+QVXvHTLpLfLA;u<FTMtYcAW^2mUBN5&;p*1Pc$;tA)w%W+ZhQN}JJc7d
zA<{up^F_7FP!&e8b6M_InJFoyqa^*QE?3W+^g25N6X8e!*}1Qa1T0Dm0B>3veaPdv
zvF!oP=jm4}8&KF;xUX+5@(aKcYQXS{mBszjJ8f<<#?JKl4^g}JuV<u+hchbW#KMq?
zIgc^nGjl&n2kg!_`XT-<{R3dUAs@Zh(PzC<*zD4+a@EnD!h+zT0RVrQ#tR{q+|c^I
zXJ<;(py1hj+mPaV>y8U{S90Oi+REi!&iuY*qg`iFHx-8Dq*Z$X2?AfX*26Dav@a<5
zug#5%2s3V`P_%eyb)-p!pqpJZ2EZ5FF?nd&7h)fE>cvZwk`SzcO7D&}c>a4o07>Xr
zeiw&Ud$%Sh&W(~2an~ort44J;cR60euW|sZ`N(E4M)p|z2IWT4;Zzh2t!r+k9782{
zkGrckG4E_x8=}+Flnn?~dWbreLFIMThH$RB2%sdP?M_cm%goLKD!TgCm<Hw>b01Ge
z0IOVMJ#HsCo}cDU)`n5J>moac1xLh}&<`zc3rA!b*N6WdKO5Nhg#5Fd0^rja0<f6X
z_|>l&!!pVj)yOgnsKmg*dcs})VrM3hAt514Clh#1z{YIM5Mu@jppJ;}Q|!Stjg2(<
z2ls{yf}DK!NPVw+=+{{lBCSj~&BVb<?B3-9P~+~4L1UYQ7V5kNtC%XlIq6oKd$}ko
zGL(a|;$eva*g_AAC84XFU2`IzF!_({SRc<YbUlr?`VbK6>2|EV{@>#S7Ry-y$@>!<
zZyDY!QN3AWuq-w}jb3whD*#L<cj-?8<dq3xY=yiP{nn=(MjPY3!gk$k{LUf3`0*!1
zax1owtSNiXFTJV1xy&Y7AaX|U>ZJsnw-9$ddHoTyllQ5_9imXKSPREs(c%`px9=&*
z<ui(yRqlJ4>dw{};h_}T<9j?2fXtCJtV{}^)drSHvVIf$h+Q9C^-Z}Y6meUGnep#F
z+)|v?{AJfI)GfuwlTR8CI1=BCkgR(Pn+1(4K7G1!&xH+pn#m}2gnwj0hyUg6F=j@p
z+<jk<;NWSS!c`c2Z*wWB99sY9i+5ZKIM3uKXAB=|LXhQo`Rk2rfe@N>EH>dphh1G5
z0c;$@Q<@rlD0#HidN_*TZN}4@O`w2WoCe^9EyP+?F1r`-lVX*ER*huSJjZdR$UI4o
zL7N90gUM%hhMr16zf)f8VY>P^7Wr!AqS%wKcDfqsU%}jF-3luu4}p-lw8F6-(+BNB
z3ZJ8;_M%8Vcd0aZm=3?;LHGBH&o~IXFi(+q>DFb_M|FVpc@wZKA2OeHxRe!)IL4YE
zxI^?<lbO7q3n}wNCq|23TB4^6H_JTWv|(*hbZV`2<9@9x%;fG5hiSbLJzh6Hc1V?x
zvZ`58(%kxV&(eg)0XJUla)1_(Kb+iJYN2jLCwI~1zy$&{+Ujx~x-&Hbg+W$w7K06#
z04?Q+t<S2F{~rkNq6e$fm2FS_>b)YjE~x0xp3hhd-=V@i65`{p!X5JrFO1c=rU5pi
z=k!sSE2SK{(Pty=PF#K;i-D))tfEim>T=j)kHs*sQ&Uq}02)QC?eFakGWc<#Lm@g5
zB6Q_UAxj5&7D=EXoZY54s;11!8`IBy>zq-`V(pmW?9Aeh>%>AL070AS5Fq;=<w>(*
zyl?oyZ?dy=ozgdQNNSm$tcY&F%${RX*{;xtdT4=TOVNgL^TUD=VetAHE8aZaE8Kt0
zr+|G-)Y(qc*=4rd%8z+*(UgwLk${e;-rnzT2D#hWwF#zuMGo9q6nntC3JVk(;`bT+
zA=m+XIgTCUBWU9#2cp|&J79s9B(qr+k@A1-78dxn(x$THbg@PkU1i-Gg0*i1;;2Ig
ztBMQ&7$)@nvc4lo_lzk2*#raDD`lZ&-Vp#7vDI(b)cjYZ9wcl~h9n5+5pXB(zwFcq
zFlTOuGv~G9rZ<nMs`5-@k4_)&&0w$apNY8LYL%C#qpYY3eFzvh8BQ`Div|2Qdw%Q$
zY`hpHK$7#`sOklDEH?Vx2x&&Ji$RNF_81`5eB?PceXo;nco@vy4~xb|^N>lTFAAt?
z!F-PB=2zORCo!6W(5{7`g-C9jx`h!d_k$M_jP}{0>p9Qpl2dc7)Z5lUIBG#)<j{$6
zQk<HB4gTVVOW(}~jUQUb1KbP@&_vnpFPpdSbq)kA$mdBkJU{TAWI$enVhRSN*&vT^
z^GL%s*T-CLF^GP|+RMl~=ple<LXA87C>K+J6<4UWHB76?;$~Ld|7I{|XMZ3mgr1F(
z8x$;;d(qyZerbs?_5kq6Vv8TUC&p~M_5rIAkMr<GA*_^fw`#+0#FIGW*+O|D4D9U^
zH7@H~pguI95c_@Z!Ja=Gr?<+zRHL!l{84XqhHR350{;uMzT(i=NYZQlQ;f5VFIjlp
z?3mjj5q!cme940j4i2{04|dIkndkWT9uEcK1lJFh3*&sg8Nk9Rdme$)^^vY+{zK`b
zoTpa+0UB38EKms8bbj{_&Zx1nz)Si9_~Bpg8^j*CqKhe>Y%c#?rq-2m{7fA(A4N}l
z1ILh0BLi%uX*UiSkMtMD*0TzdmA>mAsp$<-dgTf%`f;s0ajkU=H4|>PE5emoJCpzz
z{_I2Xa#m$2{2{vAP&LBAfN@r@ammM40w%Ka&5BoIB*@2y+6YW6ZRL=qWpjyRPGStC
z;Tl0%`@@sZF=`xR($hMz^m{Jm59@wI40#!-kO+j;Rgf;(A9<e$0c){^Bs<_c^>6P6
zC)57E5tuluk#^lQy78(6vt+;A3d$iL3${;qYp8Mabh8X3+Eo2}0QQz<)_|Rm51Y#Y
zj7wbDFv7Y+xi)ZnH31F*HuAB|=lJqDUam12t*;a5a#FWqxX+a1zVTvMw13M`7dGl2
z%${&P3y35_JWpbEXHFOfHvMr--l7i`pS074XnOqaCb~D0YlsLHsGdd7_AHRNGQ>+i
zoyiTXi>MZ&EH6tuZ%rgs707B^10A_?l=zFT_}Az-IsJ#?+Ws%9)M}j1?KSp1l=^Qe
z4ScGSPM1qOp>#eZG<<A0uu91UhaTaNa#0Qn1VFh+f}oEA3aJ6W9n|}Q*HU_<>ewR~
znOos`0vZ8BETe)W6N_5``~m&co$65=S~no{pfw>&hv13ELSg`p7q;&0{q2jB?GQX%
z<>aMHmo(kSXNj$cTNUn2)NvOjUJ;l1Kl43GygD!PowrrQe16qDp^m0nyYq)#oei3T
zwKMw~^*)VmW!!kL{G?_OQ1%fFo=Z>yAMN0@r1xndM--!0YITKc03{GlBacm}jEgYX
z5B0kZQ<*QX?QLHgzm&%mclr*V#`QDm@CCq!7V2*oSb<^`b8U6l2<V4AfeE|5$S(kO
zEMf-W<S=SotQ_q$N=e7KAzo=1N(F2*o>LC)Sq25p@?&GqMqgq6-!_D~{zN~&QG{pW
zTgS!f9kgLUl)gQp8g+Z{$1=Z9a!$UJvo||dfpz@^ot!a1C#yRFkr6nDLV&{<X7jh|
zfu1<pd9sT?F#4@@h1%``CFCe^i!>TiC1PJ7!<3!|nR~IhdE3x+Kq>1B5BScIeWW~(
zC=`R4SX^ABq*8Y}(Jp5e5j*_%jV@9luebKSN3DJP9t7iec6Jum=WqkXfFX1U^%?uy
z7!Q_AdS<PpS7~^oGV<{&d+1{yCoFshdoy0ytE$H_HqB5?pGNx12-K@>Z1;tBJlX`*
zRaAD%j@GpfapwPeQVZ)14N;8apKBZk^4<ZI3F&JMzt4;us$OLde=BKz&RWZzn=vv2
z0|>5k+QkdNxRfu@1;P+x);MP8|7k;*MOd<B`9(w!%2%04zk|NYu|lh=riUMLw54xg
zYEc$1k43sTl!}RcPptxZFzB;_@T>5`^DvC3H;V9N<N7$~tgJpOSIn>uf4R~AvFcnC
zA&PHRmd?PTA)1>$mtVi(p$ubgC8*n(ds%o-_k|vVdq$^{vN*8Qw58@AI5E)@@3Z>6
zWV{9`hp^a^K*poLJaDl`<qcqon~JMVg>>m7-+cz!NAckf2O9GOM(#-(_CpEE`_Bqo
z@fU8M^{I17yQ4NwC!2^U*U4y$3$4)ry4fiQTq2SKF*x5p^T(&`G;8e0Su-7rsrJRV
zEpt0hWDS=;0Wv~@3b`lAfyH4Ws{A5lvvB{rFo>$r-(6ZaBVe(d+PTa|U#3!89fgQL
z9SrTNjO~C!TzUZ`I=lKno?JTn0*%YkPdw_OOTJi>PxcHQF>o)hF&$RSsP&!iE+Fpp
zd}fvoNCJ7<I?5^KA9yJp8U)u!dobUo|6=3V7n=a!%$`}9dRCzuP3mS!P)~2UQ<V*p
zT+FS6nnfr%4%Edo^qq=bUxXwej5{5WZlFvtssumDGc!GX|K72p*yw|M81VMG9t*eB
zg_GkMH5xB6nnQO=lE2@bInlzo3|2%^%r!UO55zb4xa7U>v0h62fnmxHll`tr;T3S*
zdumHMCHdIM!T_dklol+2k^ROzHj$eNcWVF6I4j-7$}J)_4~Av027&&Xf$9QH5}dWj
z(?S3{=5S4DR7bc=jw<~iK&SM2#A)s;Y=kxzgE&Xw3w3+gd_3djN<pBP&*7=eAJY@>
z^4Gsp;>nYekydx%n{!7M$iMBh(@1p`^K@<N0hzM2x_Vf|p|<+UOc$7YDJC2^!8K-!
zWlnY2e6PF)!DGVG1q(T&pM~t>LiPzqYeI*5Fpgw%jB5du864M?4PEpeY_!=`am~6_
z7wob_4bZ%L6_sT^cC#tfNnKO((DG-2mYDN}CS{$BP6t%KtrGIR0ZT@XPh=!hc2l21
zGRgOo6M1G1mvp26<jMPY1A7*A!_gc3PHBqd&(Sr7<cG8e+4i!6HZ<FpV@}Eup@us2
zWkN|U0?6j$1wOsFu-Je=ZxG-UdzkpcBCN;jFFpygC5SU>?y1g{QJESmrZT(k7UL5j
zx_k$nOzKdTsO%Gnx{{KT3uhWro%v+S?BR7xm71v_r+GS5tSiB3sdF-O#>JL4UNGU_
zdSDA^4+LjB=N5d;&CXl{63%uc{DnNT?Z^#}1b-h)iS0;ckZ-?@^w~HC!6}=hp2uQp
zLGeDtfnsLQbQ1}T11iwlYAez{0|=ceFY9ISn~qSP7kiy4ceS?bk6T%Mr%FN`xN!n&
zY?bb5y!m6Uj`4KO-KY+`t6i0%(tFS{k<u`<CZlrX$vsyh*bv{M-JyVNqU?p(a<jD`
zme9eKRmTsk^KrhTgIE44eSg1x647DkcLr}wo$5C)Ov%Jn#?lc`-{}E{{{?NR_>pya
z@WtnBG=T1(AkQ}3_^8nFt467)07Db};109cYS0x?nt8~UIlB~~qE_4I?XKmu_?`C5
z(Xl102qtoue)`@|n`!s)A#TKZ`!~{@+O)Q(Rczi2E@|z8#K9JLOLKE`k$0p>!Ka=x
zqjSo?8?Fvxe7qmuuI+S~MZil3o97Wjpx`7MbLr(Je{a?j`ArRq|3Homa_P2KxX;y(
zpI%M~#PJ<&0%#Svz`9u|7hRH2rVv4{ls>23Xc*|}S~FzkZKcoIh2gX1o^!etKEmO1
zPj4;{xj6wp?NFISorklEz11O|CLKKDiTc6EE2+P2p&wCcpoBd9D>+9%=|EVy+_PJ%
z{uSO)Tp2qk<#-1!Mqvmap$*WA<7vC;L#`e@CnBnC@f|Q~Wpg9~0Wp5p+F7}<Ux69R
z>j~1K4}*8RTAm?J+Q`sEMEYXGV3!4?^&a_;0yXG<6dPYKc9KY5_=6HDZ_l{=5%xYa
z^AH2P;ztKl6hE9=ra6I?QV%MIiPW)5b`<-SBzkF7I0gP3FNKVt-%oH_)2fJYRt~T$
zx0Sz?IoG?XT&C6){~~4nbFpO>rC~z3KdPmUy4?4>igq(vgBN<=+Y&`lT}<|X7}}Wo
z(V)=5paP)A=!3vs)?AMXquMUsL&USqqR%Aqup6b2R7MwEk3K0xAkH61Dqigd(t>nD
zr1C`*p7kS0M75dvYoXo^$+s*oZ^TD1r<^f46L}X;aW0MsTi3&#iS8sA82q;>0X}s<
z0G>LnU}a<8q8lrh4w096QuI&OMJ3qXkFYpV$s_l(NF#>#9|$KJD1eO~y?D8n4b<|1
zM8TTM0ytWQgPx90UpY{%ww6ff_Oj1x2IppuAiUSxF3MR(DtOnaD1ABt7A>|aM9Qs^
zTf?KNxIJcr*)F=U@uNfh5p8M|>e)vLqJGI5rmJZ&T%HX(Yj%74w2>}$os_;v_2}Mm
znR-XSN7gW`n_H7&rR|svGABsYeajdoPYYHJl&Iis>+ZPIIKNDZbCNLbs=9&p<N50|
z+rvL8Aq&SdHO?Qh*<|B8cY+QnM~4*m2%aDyG!D6kI=Z|GAKD3>Qe)8gDyl2@)l?m;
zpKQcIa5X;SMCn!pB0hL@vprM#JHlSpK95Z4vKQ39fxQxMpy|hYbq8_aw4~DTnyyAj
zhC3LMk$7>{9d?EJuOQ`WkD33y_Kk6t#SQW#Y&qg6tN{z^u&;{&)^l8$PJIA%7)Pu@
zy*|}7Ba^BRWg}P<)FZP)H6oOwg$y5UJQDU%u({o8&Zu4XE0(uh1V#?~(6QR&yE0bE
z4)q>3ZGF-~WKY#{&wuskBx(gHwTQy8(&r3HFz6OafaY)3b5kfWpHNW4ct2dAe@Sq$
za=NQtMx!}H(|C6-N=TBmzn%(JEtDJXZ|4s_%#;rLzSljTXzNFaOgN5jO^0zZF5)9N
zx*cd5@q5QEnxn02EUj>6;&s7s@<H6@%{{@C^xwizEN{x7&~rY<@qVvlw(%?aGfPX(
z69sVoWlyr}cNv!k`j?P+q@{qrBtu@-$T)^qpsYG_s}9QJc=s=8N#w0AOj*S141Rby
z^7dFmwcI+phB7pX#~wX#TiN*004g#PeIq{lff(Q-_cWRlBAp&+QFYEyCGNBbBnBbC
zpvdf^Y$kWtI#r^0u~JndG*&ixWw@1gjBErR6JJa=8dqgSXAMQX>D}DKN|Ea&wg=1H
z!y=`k;|968w+9=Z(tSt>wz;$74c>Z?_B~+K_(IL-l1hFMfo}Xle@YxKz{)qHu+`kX
z#8r8PdY!mJ{+>wyc?XhWM+3ex$y^n0+5B;vB1ap@?E`({bE;Tq4mod+UZS=9iMbbq
zDc~sLVI`JH8@8)(-PuA8AD#|FtBy_0ro8Tfo>uuqwwZ7+H#Z;L%&!SC2m#6Qj5by=
z@o|^k1jA3wUD$g%lbeUM{+%ZZw}3E792P_3HA6TG|H6<V54bp^ND*ga$%x(GTHhsQ
zF0rC?daqNBgV4BnY5UVE$=Gd$TvRjbNw8ZTTDjEzaXmb@zXv@-chWEt=FJHV)`*#b
znPxAr_`{Vi1JiYY7xp$!<@_p@O5J9@8G#nwe?Ji9I5f;*SXG9kVBBsQaMRroA5px@
zB=fmet}zx=7huddy+(o8Ug?}$c175T4Pt9Mw*U>C(uc{G6Ye}!Fm?QW{c4-}$h^uP
zgl;bKZtxCPXd325v#c5Z{W1+15U^*3OA<OFQG$qSwC^_F$uT3XW-trr#83-Uwt&nz
zWVxF!WOc$zC&I;unqH&fic&Ve#GF@AoiaEoD)HEXnlM9(jDSgkTz+5guuC89)EXZs
z(Nn(TM1UCl9cv0Yu)1JB<I~coJC->ga#v1+wT7&jU?)=)M0tyVa-Lp)7T#M`uaWL{
z2#|0KO$vEKc$sD&`*(Ql1bEv;G1?hecibvcMBHQ^d7QN@JTfucFc5OMA&Eaa+{RZv
zA1>bJA&|ivG<VK3R~8O}f&q%3q`QBQRm+*gQPazfu}-kde#+91@!RM$9#`Ew@1Eck
zbLsLN<frZ^(u_SWm%-I#p=KoVlhd7whAHe!126}R*G$a2Af9q`h;~s~i8{3$ZLZXm
z$!=7L5{pjI_WyaD^D2&2r6n#rKs&W>J`4fs{V(hsz)P}d#7=0cX++D{sb6|*St>y9
zT^Ek-5r`1|dK(OTsTiXe+pKA@nRNmwK_xt<L??{2C|v-{?40d<%M3?03D9Gb2PNoI
z^!{!DZg4({7>fI$ua(~R)D0#HRz}HI(Mw*?<^-!nBP=8B^=~;)pxFGy2V(cz*q%qS
zDo>zM=%~Kv=NDmYxdI*6#a@j~V7Hc}Q}j8?L|9-3N{qVkK$aZxj%|+REix^ks|@SO
z+Ud9qQo;vq)GE4;Z%!Ci(cWwA<!Lo^srMTJYr5M8Lua@ia6!h`cm-9x77s|H6-W8U
z_^g3}@k<AQTXLR~(X}G6Tr|Yv_|hlzZbm)CE(1bpoOm}Le>RdVM_o`fXa?h)!t1hX
z944948sro{n_GW4F}g&-Y#A$4u>&&uN;A6=nGcmE#y}Aew!xb9TGZU2=_~!l&Zv#!
zKHlcq!2D<eynd_qy?iU!yyR7Jo6ECtVpc*z!T}Y3bb!YnuQ%I%LCkg8Qpn_kT!4uZ
zo*}*htQ<1Ba*sNAZlTHkO(}l199(6TifN+>uQAI+Mld8|(56}=&w6QktGY(WfQ@kf
zQ5CxmAf63pl#!UxW$m*0DJaYY{E-#CxX@dom2Q7M=QlX-^OTFZ{lJ?Ko=Cy=%kjIA
z39m}VY>u`nLZ(13BskRuJ6KrG@;ZPx==&9cQDR9QjDek`4UThbixFutq(-@P>fQ-W
zjk=rh5duTM_v}QYof65O9j?L)%7bGd<uPdUQ>0Xo#`dRs$mg46_Yhib{EbaDpkARU
zjc%n9*A52<2l2-RlS?du`;@(GhgbBwe~}ZYz$p5t#-BY4M=A;XFv8IlNTB97&H--8
zTt9_#KWv|%h$Y6sE|-Hk0<@T&`ga?(S4`K<WF(hc9vR8%+xzlKV#=WPKR~!TuszV(
z8fUrqHx75Q?M&^d!_}%_2_K?PL@UO(-sV<~FwkR;+1pN~T3ii5awI2?QdS!&nylN^
z{uwobTEY9=eJ+PjYkK1H$Y)r%;=5^lkgz_IU!HSTYcQ}yxX%qhjG<U(1~;9USPaW^
zI{=JhZ=66_ar?UsMoRisY#~S3Tu&;54+^5$WnRd-4%DbwtLKqf<6c~s4gkQyBBGlB
zrh&((A6H3(gIuh9(#R#=h<nOhx}AjYv%&*X;K%Tgzlw#AW=;mTPTBYeIo033!)RY-
zbTud1`7)T~&nM58Y|&}Y%niAAeR(MQ?0KMk4Fy$>O;v2tu6sNdD=@vl#4BHx_zm9c
zrCiKi9X+fA6_hFa2Zyn*sE$E28{R{e*GeyQ+xWre?i~!>n;M#$N?2S<gJ`+B)joI(
zNtjWX*Z`<?N-bNz27+rKiCSj5)!`akV;GKa5Gt19tZZ*Sp{)$&HP1#ypDovwBi0;b
z3lz_pXVIU#7%}G?TEF6)VdiVePWErsa3c!VUXW;1TqaR|rZB__o(n8no2))~F7;&S
zV%O6utLSJqp~O^afks8wBgd*|+YRq*hya{K^F5X!Pl3d^+2GPX8pu_w_ZTt<pRfFW
zid}Gjbz?v)q<iz;vwK;Z<UDEgnKQ`h*2inR?8BEwqn&8uE@gH|eq@qvHTnRa^)Z=9
zsA2gg)>|!gk}Rf2uGrjdpFai@!XoK`(pX5^sMC(z<6RLUP2*jc{`-3qQFZzzJ<D^D
zp0Ie*yBZ}>MEGmkos4I^DhV499Mt-Dg7g*uT-Wk-aMwix+^T)F?5UFs)zOHRh`RW6
z-_T+Q@Mub$ghZW%jue02GDFRS7m~7~2~%o=XJ!)*2W`gsBtjP*-Jf-2*<0RquoXyK
z2vEM>I-i+#ltBr4_Vai*o6o$|;$+UwB~N@ppGC!3y-b#t@9Ly)62Hoy{RE}yoogL6
zoe>|=GlH=<Z6J*-Iu?1kTgVO?sQnS6Lc$QEVZ2P#@}r53<(0^_64$Tyue%mjEuIMD
zJ11gSA)4lbV(GA-LPga5$XY!o^Uty)R;M(#@)x(M`pAT?FnIKenR$`4$Nh9>w+c^b
zKbgrWEJiH7c#P!2SDwe_TjL*16mbhCS>YQN_DB`2SSg#j%JY%-RR+;*juWRi#;R~B
z$JHbDIBOeL-(7YIA7$7MmA9BAj%h>{D){$k9l?}pLG2PCRNhKwedJVNK5{UO8Ihf$
zFgsao9cWd>2lUiZ#%M6?=d<y{dtt$qPWlAKitv#$V3~e7Vo>$TMh50V2X3(p4sN>2
z8}Zlwzehb;4+jsp(yRmt#&Z0W;6-&lN%+U7yy^#+Ax{m?xDJg0mZBO)X?ndZjVdB&
zLM(dYAwu`ol5bHOr_=B2aJQeY!|U#JC%=+~dI*th23))<R8Na=l2lGqPN*9rf`c{B
zmfNmCQPh9Y{SwTfh|Wu5qSIs3)8J<a_dMlzEM}skqoXH^2w<5ZR9XF61L)g)TK#D0
zg^vUUy(*>2c35qG%YsN@X@XT0gw;1+wG*Lo^<$(%j!0Ya7w!n_cei#t-qW9TJ!9-s
zvCr?uq9c7l)M~U*&9|L0nCg|T7YJ5KIl@y{1ub8fjL+_Ty<KZ_BI*SD+)a)@=TBTZ
zYp3bQpx5@4$(4ik#KM}Nbkr03o0u6r4kg=i&u!Jm^^&)Lym__Qa&Tnu$lT0v#C+kC
z%jUP{fyq?~zmS3*%`~gKlB@$P{2vnsmPAjPf?W>eG6kE{aa#TD4M)dGKJU1t(o*Z6
zD)rFHq~YClElTAqxW)E&cQS<pal0X!2b({W4%bjx`<AYFm#h8AkUwv-O7Ru+Go^d?
zEc>K)Y1=#Mi8MXN^$p!K0`ArwGoCpcxiwTl>|yt=(o*iXDU!KDqs;ZSXP*#WXZx3R
z)8^bi9ZqWsI#3Bbg@wh0bo$^&-Dx2elH$|F$W4cXGHRq!MEi;V`}5MBt9^|Rv$5{D
zg;q)!Abw0|IdAOu>ty94wFLtmtLpdDxvFl}JKo>l4ScF;wu+K=8*~Bxk`T^pM+r94
zjlt5w6i(^5X=QLeQT2Ox<55pZh7bG91x%4?!s2}T#Lm5ty`KY#l*3gH{?_e&0GJGk
zct8=t&uh3o>|9^pX<nD|So8^^Bqh#mCx_LB9FS~<kUIoah#74PzrUcA)f;Y(&u^J1
z7VmvGsHv$L+&twbFfql21`Rby4cMARszJ;krT42(IhQMrI+U5Gnr&3{w~gqyx%Mpg
z)b~-&a$iaJlQthnkrJ%fRo&V*?JOk@2bL=i4_-rhqOHq{xPk{<1$%zkj8phT_IPXE
z-e%?^_X6D9v)q8mCnb1uBkO4YU@rZzR$@wFA#sA%;H1-or*^U+_XSEcM|RX&h&UY2
zK0Zn7a0i$X&T<Pi9C*ezDrBuT*WeD6ENK-Ib^=02nG>DD8+i$;S(&x_q6{`64=1G?
z`X{Dk9KY9X`41q{=W4B+(6rh2pm1FD8r7m!I&lp@rs)fB&AyyipEUmox%uzgIx%Bq
zD7zokEh}%Hq>9Q68RNnZKi|8~T{&?_atKnZWm>WnWHNAb1#}<f503uQ(9e|$9`%s6
zUw2@11AFm<pMAenPS*=p)>Df7u&3!d)wCT{5pegEF?eBe*}O-taEaB#+Ia;zxmQCK
z0Cxl*(eSO?tAyibFe&kbf5`5i`9|D9Be*&#dr2efa%&W(dM+^J0WBV2=utkf?3|rC
zE|v~<WI}d(F1G^mK8|Z?IK2(DD!fO+DTKSH9#C(?A%0m&Qg-)n#Nb}VLH|tkBzOhL
z81yrsEzNWhjMx%*Wc|smPsiS^mNn9vYjTvNk}2P4V_u@vY8H~fT67W9sxaayXm6gJ
zXm&b_TjTfsFH4=$4HEIm{mIAKS)27W$iu31&z0qk{q>}fAWEUu!BXRKe?j!40lz4{
z<<NZiy^~eN@$Q`Q+(1$iv9Hp%X|)hHM*Gdv+9)w#WV*TFsj+I!$nSvId^ayhL6)!W
zXM#gW>#K8TJrxIV;FriCm#JuP&p<P;QG?4#V5%gm*@=A%?ut^C@QcOWL>^7ef$)?A
zYtPc>E6*)kBU7=DL)5m-fPo8hdF;#<a&yeJ!1LCBF7!mpZ3e7+X5pXD&u^-lL(TL0
z?(4%UiDFb}N3o#K_nen^+tGTENKr6lva6P^wXBw@+^4w&Bt!J*Q|xG<NEKMm6RDk#
zZF{~*eD|C-d|EIe>aLKo7+hihF!blMEF|?6%&J1dXFfL!=9<*z*Q3}s=X^N#(DX?Z
z%)DZ1)cVt!b0#+KdGAjbi=!2j_W=Q9LA3w0X|sXAM6&0p#s^~?lm^TL;v>%;!+F1&
zMh(+-Ls@k^&h%Xavj}FgPiJCSb=~t19aF{jF0in%wLg9SoSM_f&TzvEH8OX6eYC0b
z-hCW_6Jll7x&6(TH1{JX#0)|U8hJJV+D<O}N#ssMAY_87OaBq#+FKYC92{4{TjjVv
z8p&F?+p?XBL`JogFK_R>;CAzyT-^$!QbIZ?AuWk|GXYJ+PL@Nyym|F<!A8@I>Z2;G
z{nWX#4@h@c>ZnTq|DNvYqdrerXbAu3JzpszbUkn3b*njy?ce9#R!fG>+|AR5U19-o
zUkmEC%dLHr2%5DdFDxrX3W6f8KPwQFgdot=_ESRYs#g)&v(JrBy}%vhNyu?wl$A}|
z%*fG5E&6T;tv>b!qx|JesVCLxIj_P+8Ji!Elwm$r+f*czUmI*JcI`I=GL!}~S%Q{1
zg9K`E9HE7WCKd%&rVz*bK_4b+&5_jIVC%1+0;h6^tFu~1pD8o;1t<8i3rUiRlqPeR
zMgKbz|D@4&3+9$2$jUX1L|ZP#991n)gg4mxu&PCuP#;nqn(3fF{I-0iOzw+BfOm4J
z9xOp=DzKIHWyOdkdt}*2AEH1m=xW8TwSrsqR0R=ap8yMSoRT~-<U5$^O><fIJtVM)
z>=ey?#~LO%B-kbmr2h2T@gu5_gS5TlWW5sjvhde%*|?&^!IlsA)MT{3t0ssfJ;Uz@
zG;EJ;b~4%6o?HtmMXC!{-=z%d!3JsDaf17STbg@!D&0yb$USS?j#IFm?GOR~;)qPW
zKi0FCW|FhkG-@Dz$TwH>y+C%1_>lX$GN+(I;z8|ghmT#@{ew|Ci?UpAKu}>_Q8FxP
z-e`*Cz8q{EydAfT=dKXrWcv3t?iWH>TkEMlN1NGUrx|%4o4rI#Iw}`DHObC__vyHq
zw$29?1)BB7&~dh=U||D2S3owL-O3-oX8S$Th7P}L4Da3_0|i$Z*A}0dl=Pbr^Rj{!
z5pmx61tSKg>rmqNuc5^w=4}dC(^KWU{3dX-92)4gz_QB69*MiBFqx-*!OHcN%6=lv
zY$d$n=9)2h5N+R|NE{+H{`ryhkWgDJC6GyKRMX}bFhV^KhT<kgb}~0kN#5v@mC}32
z=nOVPW;r~4RP;_{&9B%jYg^C-@LT+V|0R;PD)1-3a?VVtA_>v3>S=6Fd03(3&UJ<Y
zF5)^d!pJWFeWaP*{#29#`+5U9r>he4$2;B2b{N|z1HW&z8Z4(5wbf98SgOxLYWkvz
zH4CfVGDA?`&VMUz(9r8FdG1xQVe#`2?5wa5loNrgm4PnK{o*!D#eJ-$O%E?0-t|A?
zjdQtL=ep^)XB}8uZH>EMVUF|LmJ?Jkc9|yDx2_`lr*hinY7bX;EPB_?9Q-qB4Q)@8
z{7cLs{8Aq`D6{EVQ!(z|_G1BSX=&9fX-V=h*?(WyjJLMr(z7;#apW1AQ@cTly_Hl!
z^HiBhp26~nq~esv<DT-a70+*ujZ01n8Xk%FT5W66eVkTk=|_qZR`(nf76LM8rrre5
zE$n{F_Tz&E^+NNq5SQ}VdYcKEaXL!#+v+E2M9s?{S?rcQ;L-z4w=Y%$mQub48hg;b
zg9GdHrRS#47cbQ{&evLzm-nA^%V;LGFbmV+Z<j-Syc>Rht^*|atfuw3rH7!Uq}_#=
z#sgcy^>?e9HG2dpt@R4tWO=LmNrl;P(uQ;6wL|n2U#(yX4iWTSWz7JbiSipt{n^}a
z-k)E#u_(t4TN<(o51fQ1<j71^bPZ3}(!R{|^zN1kP@e}Tmh3nG-f#V7*eFH$#jBa*
zP;-sa8dBCn{Vun?agzz-Jxz7;U!C#^!KZJJlmB=g&hVptNvNtxEmDez;Fdkw2g=g-
z3~io&AM2WGj{VO0=-BJ0M0yNd9-<r@Cvqficyd+NcBYrRAZw2L&b5JK8;iR(|1|k-
zW$J!)I@P#P%N1z|j?idzsebLPuzo*(!9ciST#BSqot4=G{xTW1XJa9`-VnG-@_#th
z^@}T01y>=d;j+8A+hil<y`f=Pjq-WxTs>i3kuzV)Bn=XI=2w{A$5m4NIllO|0L4gP
zg89e7Zts1IV$;pNJD><tQ_JBR`8(bQZqv|TX*qDFhOfh|yuJY|ARigrLGg?20I9~=
zu^f2+y2N?rI66(+nOEGZLt|4=aw+Bg#xs|CiRc4W(gvx{RJ_z-kLT@}SjiNU|4W09
z{)A_Sd6I#7qtX1@-QhrDwjvJEsNLRrvNdaTAQoYsYi{~y_1$Q2hRs1*mbH$1R5MvO
zo`W>%J~;Q2oK;<l_p4c?`b1)YvvAY!fhv?M5(jnnuDg-ei@~EzM+qLLfrkghai>No
zTB-tm@%{4+PB7QF;+d_C>J{14`~U9_8h|@^U;Mn93;HAyJ#ojVtwti+s*}j7`#HC6
z;Ar`lz4`3?v+n1;#U1Hf$YXv>4S;=SrG*PHx*MISGPUpzND!Nu{4Dx&+muo5`Sf6Y
zMX>`7(Y<@iy4vz>VoB~7bL{}jq0j#6k!@B?&E*TG7G{K%ASv^(kqyW00^u==zye9a
zjt0Qcg5`AD&E>T!RX=G%+m<>Xwg+T2!hH~t=p<h=ivXd>=jA5`Vs9Y~&_cGROpUKc
zVQz^P{O-uspE{r9CTfY$@ul1_&jg30iT!v%Ta`o&w<Bt8tE)HbLjB-a-}aC7AILsS
zHT><lFNw6(=k{4~JZYNGJk`4_(oL%^P0|TRrZdZO?}q2<Lqb|`Rk#wrOj~nFTJ?TI
z`STGTA2fd!!a7~#P}974%11CxYJuC<!CVBLn780tzVU^+j%U;wiM+T+BY=Esqec`n
zWX$X0Ceq@p3<9N>%}h1N|MWrI53LqQS4{y^mQRrXJOxc7%u#qcH-#x{M%=IdTKpqZ
z;<9}V+TSxZwW&X}&$Pt(lt!_%4{2$AsLOZYDdX2cGoMH!uHFpY`arTXm$84=$F|$s
z=f5Wue)pMVS>f+>v{*1}I)C`o^b5YQc2a=rzE#cEutTd&6#O92|JU%l@+wZ#TC>#p
z(RHtR>Zb>F8CvG33NE(_Nz1+o`Wb%n1I;b8YYTvdZ38=M!CU-O2MyIZZj{C@w_q(s
zlz*sR;wWeP?@v>k&F7Y;1<MMI*v+SSr-(K^2eXm<p7j;Dblbd4QYV+@HgQ>$ejsTQ
zuO*Y+U76yV&9%9Ax5-t%h{VDfguuyxxmei;eyzlDE!Q?1&1mLc+-ykg*w9JZIoOWl
z30nQU<zExo@3WA%?$?i)mmtI;>O&4+$^O5vbrj|aruBMe>R3nkW(vvcC6~lDys?=l
zEzhR+JS>amnk|y}BC)SHK-@FfXy^*;C7<^*Ec#vVS$)9S)ydyk=rDu68s1QHVx-CT
zIU$pH%_het;;E93ovrVdjivRkr`7qJbFW=<1?VoH!l)Q|RiI1fxr&{G{O$}f6no-J
z%U60-%BwUbg9672(2BL4nt^x61I|hlX|_lvcVEAZ4V)uw#ztpS_m7I1HzK^tJ@TKt
zNDQg6>xq^U&QtJnn4*7%Bw&g1%`~!{A|K>4-0*vh>M|La9$JBQX?<q9>hKwbw<)B3
zkO|txhil59!)I6{C<&GWeEj5o_XQRo;M=Y7het4hSkez71n1nTu#kyp`*&V_sdTP=
z%}RXYj_>)V>NZ!>$(&k)`86BU&J7*+;S$Aiu`VdkAq(8GG3CUmcgkP)VOcJ{##x(}
zSUkJS&)E4GenTO1{HR<&g4uHmQEt_4wOY8=fAYQLr7y8c6VTDXsj7f-;tS*vbzDsA
zp|65FylUUZ^LK-F75&b%bb$U-v=!N=I=+85Nb;jbb$AP}5;#&|S+Ru~RuM~0J7|)!
zxlBJ7<ko_pl6!If3~AbQZA+kbL;|_7bH_Ur+r6$ji-}@ZuLbgzhEmCd6wW*`ju|=1
zLzWuaYT;U41*ZhsSvkBABo<4AEDjT2J6y+QPG~nuOlAFg+g3ec!?o`zy%?9h?{y8@
znTeCd6&-2rsHt)$YWr>1d!X@QGp5Nlu50GB#ZsSbO{6q|N6n=a{O^-HF>S|A@16lt
z_Q4Zt;%b^C0Z+82mps5`O{=n4VU*1I@+;kl=!YT`tl?TaGxQqIqO`df%7{+8e$#oY
zqy^9~>){9u&JhpBdx6z4*7&<RpXo|1w66U25PwHbtu_`J<EYZut8@c(e2)Br*9(QR
zk8TDGdid^W>}v)m(rc~L5|stG7MDELNB)%8TOhA`QDn!+h&(n*H-)roF5tDmkEuQc
z#9r@8E(n52Edv}uT!$k(9~{c2)WImB?SGvu?suQ3n7Sq-@^K+rhTZ|mEl@H2Kp9t-
z^tP)4hn9@_@?)w;#nw{*mp-w3ovL1-?gqWP0yqMwP&buz32BJe1Lm(tqcQjWebv0B
zTxjc70lfpzInQ9Tq=|rgAtS$-##^t1)Xphuj<l5Uj#VXvR8-r*faq!u6u@#pB)!bV
zOIz7h4;uuC^OgPWlSE^2bz7OPRo+6?0R?GlQ)+vUidj{;v~1n^lK-4~(^nu1dVn>q
z#W#Xb0;SGr;Yy~%rW3=aCM#PFzH190`zgsuljw)T*f&@q#*_wwx7paa;aekwv{h5r
zBKW6w_wXgUMdxqSH;fIZfhvk;sh8HaJRbe-7rv8}x9rL24f@e^6*Rs%Mj^<SF@^^A
zGB%l~)=ZD)e31+@JzW-f3XgbEy;@l;O{|OtxtwZmq)3iIl_-T9gxEPmV9rZyjYe0G
zQ;+1061dM%V)%8GuVAAUvSsT$C-u{3jj&b?34_sK5VuS3YD;&Hd0Skuxi{3CB4J)O
z3YemQ!YQmSc=E?CXMrAFMl}e^EA&&F#C1sOBhOh~%W}7`cO@PyPzt&h1Vbw+qpAI_
zTj}0V4MkF8I>sxc)(3{kJn5hCb@%alU`8vk?mi5txTXU8pVcalI@tDRdlEh8=Q$>S
zX-_~;rgP29nHh&ru0aI<bF%94Rj<49odyHtTS?6-){QT;tfdOJ(<_vTe%_7F2!-tN
zoq!ApW(}N6&03%LbB(f@E4okm$PNB252{T^kTo9@WxYIS-nzc5F7I<_YRM8ZDry*+
zZ#nR{xtz(ajMuWF1RR=1x*kezY4E>F+H<3s_ROWv`Rh+OWv~c^m_wHjxIQ!tdt|Zf
z*stQ{Y$%G26T?AR7l*WmW`vxHT-s!y3Q>5ZGTxzK++%x=GO}pWtl^wKLh4VF%am|K
z;Vj8tY~u%;Zp=8w|2p!u{&isa2f2R;02^QW0+S3h4qyA1BKbeN?(f~*CRXE*c)uKG
z%=>M5?)!E*_yp%reV6PHGN1kiN1po6fedhRVe2?eJ*}!1yq_+9C{&PKTlU}3#2VC+
z``&f@Dz<=j7B6d8czS2kUx>U}9&FnE``avNdUFyt9>VnCF+OZ|I9OFev3a#%*@9H;
zx&_LC%2^I~%R70BPdcYroqopWaw=dDw8uH-eB`91zAvz4$q7y>Ebir!HRGG^ld4CS
zIMlZkhW8y>SCN7FWs>*dg3E$BmSm?N`Sj6;)_&OdQ?4Kn@zQx_l#xvb#Pk!%+gk)%
z=~kmV8XI@tZ?$lpm;9Y5B{^liD<yatxA(9Fu0_>sNP7O~q4u)ed56+k3)`}c187p{
z{?+l-ZI2sd{5`Y$Mh|>zrC``-CBl1)H?16!IOn~+O@WMs7;Re83@(X1)^9A)IZAs(
z+Q?J!%T&^Z{pY-1%12(Yio9ww-E_$8Ea)MmHPTE~)6Jwf*N;?KX^v>N2qd+R3I%3R
zlHkkpmHWCqMl+kbKzDgi8SGXvziaV-+PU(tq|-fI&25>MvF?nirDiqR9lL6_xCFwq
zSm#c+@pK$iM4@6#LxFGsMHX9}rm>)zN$rf=otknKfm9X+0h^MqWk8mKf{2QNNkT-3
zir96|xpx1Dd+tx~kDujzp6By<S6v?H&^^*bTNIN&hzDj1Elg(dq$~#-t0qgzD<m?*
z&YJgcm4b>HxF>+GH0+SRD5bT}5*&7g8*DldCE`1T>hw~R{VZ`)a;-$FNtx<@-Mw9I
z&IFV|FCbLL1-{Q`%-(wED{fUdX#0b6ul3O?&=DGE1PF7DpC9gUj7`gg3WE{jFOR1F
z2*kN){^<OGmU*|z{_16aMz<_Buw}mTrf>s#;G>bDxMhohgGmd_(fq`|2Xsb@FbA_6
zf;r_}1UE>To-}%P4g7H%vZa$-|I1#}Gkr6a4P_w)XK7&A2#zCF5c{6*+N!^&Wu_@C
zk-fN<N(b9yY3;mku9G<YzIN=^koxc3ZPFbw_*Z1(*^`{hFn1<wCB0;0&rE+PsrLA7
zf%b~qEx%R8<PX7YaQ#rjH=Yy=csyE@JZZ%7tCWKA6ob+=o96Uzjv(xczV?RCNp6CW
zyiw(>RH+EJj1f@8q_7h|+>C8wQ7tZE2+a{SL4bRAAOfSdXogO7Oe(T!L(P|^m@K44
zV}r_Lu6|hSvZu~0K^^9e>zmW?%7ItUf)}`9iT$Hyr1)6(47i)v#Lj=MzSOGCz$FE}
z44PO~ueLbv!)hOg(s%I?r>m^|tu9qZv@W5jIt6aY-_`}<ivr`fI#bO-t)|los*6%J
za=Sd%%i094<*pAqB<h<LE)(=MXqzX`^G+%qB#hsrNACEzB;A&q$*fC!v>kC*yuXQ&
z9=JwWGvz>E<COYgJ{k4<ej@dYea`83F{EATwKlMgh7BL1NNSEFEYy(;azzu)YDruu
zIJt$_5t4GVDUzv3F@%2=tvJ|dGZsGZ=+7`r-m5@Q^Vc_~d&(DZLz&tg%@C^|1k#V?
zG@kWs*Xqm>?b9zl;Gr1*<N`uKmSnynJEC=+M?uEQa)7-fNAPQGXa*=<$d#nn%<lw3
zPvq@%YaDB$a!O=$2t;*i6;fF)rWw>fO?UQXDDsmgTe(`R+*7Z>@rKjPK?tj<&|j6j
z<>x!sRoUTsAgh|0;6`sL!fz&Ne`TuA#K?hk&_pW3?i#(FPG)pg_*ihrop_1&Olzir
z77@&))spD4?J2jIZrOV=9jX-(n$|9Y_?Ih7=Qg*IlU?jiOP3Sw*~wA%N^meZn6)BH
zccK#m+W<-ApSIpvE$7pDWssPQ>|?Mz(Fy)kZhoC^^oNR;SWZX?AGxq^3ImPn%Onez
z7I?Ka6`>`3c1WehJ-&BZ(mqtCO5xfny=4+MVUTMaHV!t>uOJ+vS(vV+thsN7q7E(a
z1hhQwMe{-~$KI1*T?Mhqfjfbg_JZo)(}mWwHuhvha7LUMw`niLwx2d9YWZub@n3>!
zNbYrETPhmolAbJsfai1c1r0Y{(X`%X0~DZA840C1p%(G@J%j4^rV*P?CLSXYjyn$Y
zbJS9oTXl3<6neMvN7hYOMzVgfz3Y&zWB+Dn*RL-Sug+>%_u8ax=i`-j5g;hE4eu$(
zH^AxCnmc(a09RkdF6+cLI7KxNh0B_<SAq|N`KtVaMD~S@T?ippofV*qpS%>%IO80p
zYRFf3-gC1wK3O@e7)kNE7?7QqWtH%I(+*}#wt(`ur{2XVw$Aa#+Iia#bwg}9SQH?W
zkFEY&7H!m7<;(m8rit;7XP0SaGfN@B$*xnRx4xUN1Lt`OP=2#A(n6(K$2^{aQE|4t
z5^m^px59YSQ5J(E4m9rBH8<}eeT!v!hGq$C{xj<xNpamL=74Bl>rzvnNE&L*z<lF<
zzpgo6^(m_;mUyf2hV#$DgH%VBpg8`@`%iQ4AeU!u7*bZ&*fc%lD;aG=r<;dd6Oy@9
zsG+vc5YEuQDtc+o+5e^@K(p%g$C|KKHj&Ym0=4A^jusjBgn#=c=BdMXqyZm=4mCfw
z=3<$|o?_rPV33PeqI#s%S_<3Vt1i5r1qE!CvN~;s!d{tlp~}l>UJ!8=(CKqz%opdg
zzY6kRPFXzRTSCFkg@ce$bs-AYjfF?dM#uRVQ+4_p0jTTAmmDZ@ci$my<CEA}Vh7Rt
zU18B0I}XW}#najIez#}e^5R#NW8DYB;QXujNA)ku(T+3!9Qw<X`isfoDtCLw2&VDf
zsp+q*pN_m~B`cHQwIhUV!`|0cpWxIs7sBb#`I@#ZU!+j6u*+1eD64Fu>E#Jsc_~S`
z%%Ru0m*HNXn3Ou5z2B8(u<u*x>VtSK{Hn$Xnl7SIZ77>_7)TP%rP>Dq=VsnQT-_es
z+i~|JB4sZLNMuNg+)hVynDX2f5>~&d2)njgP#6QCAgb;ezgcss;SX)OF@JF0Bw*R;
z8SRW7c@+AyMR%4&ah3&#Yr+sRYT1o~BjfwGt?fQ<MJ89cO;dpo)bf&<w6|*L@%MwF
z-?tZn-&D7)esv5~kz~a32$5l7luTy)(l<Vr+q_W3gNLQZHf#Lm8H5eXP!uV_DH-iw
zEC2k1A!*abuE)giMsbp&5Vmxnf$ktSH2b@D1%*E}OYvN-|Ar~O&WtSEv2>7r6l9I^
z`C?E!J&f!ppL{@rj}$orQTtZ7lSO-_!XjIb!8{uML6w#agMKdifrDxHSq02T9UPEs
z6dAXj7yGP0mC!wH_8!~v?>r8pQ}b|?6{2O`cdpT#`@4aKv#DO4ev@frp(y9=o6v)a
zOWgv#ZPlW`$b`8okoBMC1>|v?k9i;zD8{4Be`MP!q&}e4X)rc1lsz;x?SB&~E%BF5
zMZk1Nua|m1V?=HYuMumqAnUAzP59hGH|0z6Q|x~QbP2e~GkP_lZ>)^vMEdB+cX#-l
y>mcV_lQ5b=xmrU-Et>!T_&<|@VmkBvM+Yu7i-A4YzWXz(f5d%#gmdWB#eV~YoBA>U

diff --git a/docs_old/logos/johns_hopkins.png b/docs_old/logos/johns_hopkins.png
deleted file mode 100644
index d1ba7012a7083fdb14a255bce47f5981b5d22c6f..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 76335
zcmWh!1yCJL6FmIkZo%E%U4y#>cL?qddGK&caQPru2ol_aySuwXaCf(#zphT|>h`L3
zYHNDBdp1f<MHUt56A}ObsPb~s8UO$d`FPF*!GB!OWj-AN0FsEUl$4sSg#`fkEN5w2
zYiS<hiQKQ>NGe7`k`<jba6#l6k}(jR1X@N4L=1&UieF2(>Rp(hKf@9Y=0?KA#zMa1
zsxzZRBP}BAP~^pi{Emzse%|u`Z9m_7zc=<)w;;M-cAZr_fzSy;NtNSL=YgOUOOj%4
zh79!fZ|$&(2O&|r02rWJOA7Z_Dk$I~Kv<ZGvJ;^lfby9^1_7PQKRelpBi_-EB-0I{
zAfZs59tr%4AP5d1?iVXr2#Cu-L9+g%)dX^30n-r+^KC$l1u&%x+?xR)Sr6&{P=G-)
z1uj%>0zi&!6)6qa2mxgiU!p$)y6gbXSH)fdV37r2m)H3!2UIlxo#PnDH2??&U{{X}
zV*p_N0n<S$Djy&;9l(*f(h<I7s=(N1{D@R~jc^mSz~>+X5Q{5<jt(m+{kS|n6ArH#
zv{{BYYqxJYA!iUb`sVd00OTg%e$4jz)n^>7Vtkx8wjRxd<){PxozmQV`)zlk)I}Ts
zwmbr+-k8~HKS6}xA&zf(G-t3j#>lyz_c32<(8cS3+{0y^YnT7W=JT)krlqCbot;I6
zUP%M9VIA<Bb+<{U&a1;?fbh%x?PmKw>L5;|AX(Ul&F;|~r2^uKM8r_@)x9{`mj>kb
z7m^vuUU~B-T}GU3bu8Cd>5T8!oCTCol8Fp`G}G^v|2CMPVMI50fGRstDE|bOKSmZX
zcZzfvx*nbX?g8Mc#kp&i9s~oi3E7(Pem@s`mCdCAAYbJZT>-#YnwnK>xK?Zc1OU>x
zL5x)rL^oXo%$@MWU2v;i$S<b6p_0_S-I5rRNR|*{H&cePU`hJ$u__WqQ?^-g67EiQ
zoA4wTG}dnIMl=x@^d~b=b{BnX5F)I^APC!xVmT7ZJX)737KEQ3_CV1k2a6R$O+FNf
zqfV)qz$3?@5v@h8Ek|-C;sz}cVJOp_AT$VY2ET;2%5o$J)u@ZiqSQ$>{o(r*DwBsi
zY3ad{5r>!ibJFeaCz0>GzxF3fZP1tFM5PAUI}ZpDxOyGfS^Ik=tNED`lPpH6Da%kw
z`#)8EDT8T=y0w_7_J+j@`rL`g3+pXOZbm90r>mr^vMBe5oS2#$a~Y8j5hhf<n~pN+
zkK$k2X3WQ40&`A|2wgd5s(Gw9Y&H!4NU2^vYOE+JPP(ReX!+`Vt!b=jnQ12tszc`D
zpWJevsFH@3E$wR%1d^G_nFhmlvUYlQxORwkD6R~UvdqO@^B*+-(Q6FbDG=^r>~inI
zTm8o=q5V5UW9`o`Eli#`eaZ4)^%|Ax5r43`Ju?KxLh$4j3(E6SW<J^A+uYk$9(Xik
zuBDKwq)qBfj~vP#q#eNAc%z_(<Aw}GTGJE$BtRgrCiEt#{7<Qna3+RuK#QG^<M_Y*
zbo6v*ZC%EC#&5U@!$>I#DYPj>jBHxPC50vHCCJ*>+9O(V#m1W83K?xK&Fx~u+0;_)
zk}S1jO(IR}Vxd0~CDwnOwZ!rR^j@Rj^?H<gRHgFObE{3ZTl>k8EeLCUm?`X0OT#^b
zczS88$Ckc%F}D2oO41S|Qd0a&K#EI&OQ%-jI{B*@!uEfz51Ae+i2L+KY1aYK>p15O
z{}_l0xs{qYtPJf-Qqxj(OXW({OJ$GScq>kevI^Tix9B|wp-BuV&>7OnTE~sUW$kM&
zX|5F^72zq12;Nq(me>~%{c4f+?r<tP`!BoQ7?Q2zT(s`nw*N#7W_#6ob$lwil||8o
zuSKDTUqKls5bon~brvRJ8(6timon-RmntwzF-zMDO=y}D{yiu2Lk9i3ZNK3*^|oWb
zWB+ejcv=ctMdc&qvkVR8rb327AtfQDfw9rCys`Ooo}VHd3!M8u_kS+@+-NY>_t9sr
z|5M*+C110nU#NFh-|7^mFQ8YTtF3QTD^s3TPBY(9-cp2F)TEN6Qkp$)-BWd0KU81b
zKv=cA2)|NM?^q8uvo^=KZs@D%IZJ9w0yl$yyow;8Ml7L~lIW8Jxc_y-5&R;!o-v+z
zCI;@iNj+q<)}6qZSmfHS#~xwmwawD}IU{S0Z{6e`avd2Hr5HTInkLHnlc;}fux&@d
z<JQ$^%WwKN>u2rH?$bXjm1n$Xwr3HmIiuV;4B2{I?vp)(Qk&|#Q)0+DDeU$eWdW*E
zszKIu`&4WeYzaE$3snsT-X-^D=lG&`qHa0cqT9lJ>zwO)CZl|v4TB9Msy)uxCSe;=
zfmH97?|$!7KoNun${HprNFvAp`uV}{8%0N~nBa=lUCq*|#P{4L30#S|;PT+zu5t>S
z!ISuMiu?Wl#uLVE(ZbXM4(2Bj%=XM;dIC`B!Xm<S!o^};B6UAaaisBhPpwM1Z%98$
zr%JE?wB}G1sAN&#VH7Q4wG&?EYvyhi>u3F{H#(=Ar7Q2CNG!n0Si$S$(7y3<dL8jE
z>mTyAKmLDggGMgCzs9?7!iw;US+SS}Q^l1cRrr%HBrq8N;C-jL!uF<#Me^+?>9(e1
zo8p~{T$D0oBvrYRV^sD_JV|Kf@ulwLDq!Ai3Z*NKZ&gB7TKg%OjhEp??#L-3*v1i(
zm=NzOSwf+lRBwG`z4z5?q|7R;=B>{urh$z#1!2g(Hogo_j|Hh-vhYAlpMHT6p~l)u
z=WEAU)UZHmX*Sc?>1gfnN<3fYdCvPORyWpGVtqllq7Dm*?p|A&oj{GStM3ZZHZB}f
z_!syx;l}8CM#nkT(o>zZihwp+KV|zM<F4W0t~x0DV`Ocl-7xU+>cXj!sxF(}MxEJH
zcovZ;Zd7yy&I*1SZr~#Qs?RP_g!6Dm`fBpu;?mwG%ZabkUk~@zS+Wh#QcKeI($Jhx
zmzMvcZhZNBdVBYIg)ME2ex)U(UaIMLhj3%7RngG+n2?xOzSfP_lNRsU^je+S!h_S)
z8IgbM9rwxHib3oBtLRYsiZ+zC^4!-st7rM=+Iz8zsFK5n)MrPdaur9{(*c-im{$ZU
zwEBgn**TJF5iO0ld8T7ov;ngAx2$t-7F1qT?H`%_|EOG~eRmG;5jXBhQ3aR1zWbML
zA<U3@kyw+k*-2(xWyt*0%y>31H)vKJ6`h(tTK9XAT_Hoy<QCTRRk+p~vRyt5o=LL4
zn)sfTk_8T6eOxtN$*cF@w;12BLw7!J#_-sh{`?-ZFVreDZg|z`aaFUji>=eBlTj&X
z;&|k7B>a{%qdL%%*^*JwWN6=X=GRu||MmWmD4TD5?WWE567PL}E?Yokt?k;+`M&S|
zb^~uMz>V&8e57T`o9?W<RkhXRM*LJ_Nop-(PR!tKb@6;P(5}-v_j+no_8vYgWa2I1
zHV$11E-pQ8BpfyTIGbNoK;WO4+w0*=k<^6c1Z_6{Ti)A91wI~Iz{A|}!uZsGQ@_%#
z(vmtHzZJY5bsiH|uJm{JleQ1}aB&#le0e_Y@?P~OIsd(ZuDbJ$Jdpm~=CS%#bF%0;
zi7+YY9Tw)l>lZ=R=h}~YlH5X0Lm2@4XaE2b0sxQiANNB5@MH&oBU1nn{0{)|ofA#^
zWIn!yH&>LE2HyXF<hB(heY`<*k<)Vr0A%d{pHM(XCc(#WKwerx%V+rnY~@3;P<N;A
zKS4~(q0d=M60r?Q;-1WR%w@=DXkgedUtX_oXxptE&D0o<FO*N>9?gU&X0Nr>E9(*F
z@~=D}7$qJP5QH8{Jh`*WmaWEBwElW$d{N=eCjQ?hXQ((}DhG7<y5uxIz3hrhQ@}ue
z-*TZqFvwkR(aE6K;;XLJi1D~#x1va-F18dG04ezJWS2E?i<z?t(z?sIxy^T<U%7}6
zm;Rz4>7P>Y00N)}czo+q<@~pI&)%OdE&bmx4GPpp1Q67yKf+}FJ>Yp$)MrFM@q#VR
ztMZeICEk4Kx18^VZq}&1obe0<Ryn_1LZ9OCmx;Q2F+1YNanBkb{#reExX-v#iTOMa
z86q)^Ex^fx@?DI*oB?raa!H`#;&L7v5SM@+S`lTj-7F1)4l?*G4vm2%{<ahM(l$vn
z(KCl%2B}&=)lfM2G~Ria&0l7sg&Y{*!^Uin+Ouuak%<h6gax3*@gY#a(0)s;$4htT
zgTEcfU43d*r$1d>nmi8zwJ8aF6;U+nS_#PMc=_9halxoYbfJ4rPN9Q?ao4oD<@j3F
z$N)ft0v;;32!T*kAb#(92b04Y)D;go;d%OZJ25<A;_0S_;DQM-0^NW({N4opUQSeM
z!z-zZt|7Xo1bVBBDXfa_Jp4l2y9aF;tfV_x*+jwRygzhP{e27+5Vr1BJNj~#QUZms
zQz&e24Zccc<_y+M<%O~U-AbSX(BuG)C)C1v9f@!Ww0{LSYXP{>Kc>ZnKm$9G)QPar
z05z<*JDkFV9Vkc+2!#L;^fRTXXTI~Wdn3PZr{$_AdZQR=z9JQDeHxE6Q<iAv>Cm~a
z04xB?We=6>MCGAi`#TR0g&ENNY<lBOS7}Ei5a=kipnAS)OFZhkHIl|=ntJkS-u;8q
zM6@_X`0E{yLW$F7!eCPTgQh$}lmqyn<Y0g`!0&nGJsw`#XDvM@hztJtd4@fcv2)2j
zLt+ahdx;OW;bn_3mn26Ws?%9lk#TGzE^s$|>?Vqd_k&M#3urI0KPyr~QNa!oQYoJG
z2m*9~+E^PK^S6PVt3kjxa9+bSGJg2jWo1UhWyv;6jqD0Cj9CK*1E1vkFyue}@{{@@
zTYAxH$QjD@bG8kFg4r;>PGe@L7sv6E!ZQU*7bqyCcm@RgX}zipc507)eI3d!c%7QP
zlAF92qZ(T6@6v_m>{B<8k<Mitt+B!^c6sP_p>!X>wXRrm-?mu6E=R@mxs_)@qF4-%
zVHSWrR)q(440~K(pk4b<-f`IhVSv0mE&)m0^)%2Z$6!tm#*ruBAZ|CeZXBKOy_ASm
z)D?>O(@>|p;xUEhBt;RTIPAmL1R0#)6Sf%)WGQuH-2x6WxaBPRR7z*6?hb2b7hKgy
z16}0tq%huH(_^!uFNu-5s%rUD0Hf>G0KONFXt`Aoq6&ur#0wr~5bf4(yrQ+fTbo48
z3ySR^35I@SZ46W-Ypr5~uN1MzgcW8)DZ0)j!U}lpa{mti8BB~2il4o&wesB`j-*QR
z<lLonIEHVmx8QocohVDdh+`Itz~YL(ch<!fU_!l+!jY~E0!6IOp)Q?p0cl7X80;7m
zP~@HH>%QSS2p|jyIj{#mX!JG~;U4&YkiMGg55#|`B228hJ`z(u_jJBF2AsGT@VZ)y
zFx#)viuu|sX)plpu(r)F0w;%xT{)!d2uM;s?;|==^){Wg&oZ0Um4Tp0G*}d1E`SRW
z!YE#Gc=Qtq+|l@*B8VE;Bz`~$c>MmbC8TnB0En*aAp_Fx6Y5Zt7#xpp_GDqYg&^ZJ
zXlYm`p`*>Q{W3^Aa_8wknut85p715Fp)0aMek#S;Zg@Zk;B<aFgXyc<fD@r2H0?B%
zcLxcxqM<vSM!3)O_Ly>_S8GANy%i&`L3dh@lt7!l{0o9PY{*%QdSNB^-D<oeiysYy
z0s7NvD_SZM*1gs_7fi^`ji>lB?HtNbTg9?wSmS8j*^@^@;-{0zM&76ZSTN%(e5)G%
z<v!$qsl`9<o~)yc;XdO>*0b1cOVM;u*%?Gp)kYd1r&SK@Nl{<xj}ph#BZmTZ$Vf%2
zD6?4%Ln9RlEzT0XpoeT2JD5^gdjY6c#-i}5G=}d%6rs|VfN4256oE#!O9|jR29fI|
zafp7<A-5of6{oPGO&Z&H5vvrrl$ep)w)|<6M|LdYftD|4nwd`Yv_H^1esrXo*-PMm
zuSVIT#YHVO!ww|aGDB56P+$NU67z6vOSuzZu*Hw_1g-_o)#sb>w{@ND+2F`Qkm6>~
zW9e!#I6esw-~4`gzR|qvAby7wcFmq(`#q+aV-|nxWFd^*?!^i8yecl;NyrIJE}w|n
z>pmUd`>ChBFn*(U(d~qq`Y^LAX`ig)VLpW=`=Qa_PJ|5_V;kjp8l(9#fyDY)GP~Rj
zPg0x`nDnslA<sU=(!l#kX<SI2WXZxUWtk|41w~!^IdVT*4gYFxF06F~LtO$02lXKJ
zh_!FGTCD&O3Ye(DO(H(gz}SJ)_GAteasVqx8de5&+Re1@m-JU}&LSGz@y2)!#>l^4
z)h({(irAW1cHpt3r=eYZ8pZn)?S-l+0LV|id)d6=7b2w379v2x*T}7ku34urD4sCz
z=&6R90@ht>5;cL!ZV+Jj1M}THPSFv7g{Dl_KvSRyxc;TMa7Kq{Dd1*CXet59bwLZq
zSmgw+jKl&A1D@Bc=U*?gPuZqu+PISzj7gMtVI;9&?4|zX*Ki}jj=*9BK`a;du{_N-
zYCo?vR<j<gy**v0iec$qhTxnAh$UeSLk5YV#H)CA)tD#9mPI`RIDd+fLG=Kj$WBxW
zuC<V%{V-}hY!`{eeVN$E;L(QEL>vCB0mH8NvAVbU_=z<{$m+CJDwCaoXV1eJUBzO2
zRZw<a|K_CfvpfOxpiliDL+_0<Md6ZTn$Ah9XI5oC^N8^2mV}@(bidByFKP;u<4S+2
zqF^y43UhltRv>=X<mIm|%S-WQ8)-}^o;3oy(rXj%T|(rR0%}0B!4~=_B$W&6H<Ohh
zoTT{pf3QH~_G1vL5E_~@Cv2+~d$wLBQR32kOeX^wP12Vdt1XBS$(`@piN=!1K5556
zyQUb}ne%!i?RTgKIUE`t+&JK!Y5kQKXgzCRE6Ts_w<olq=$#s%x9d0Ekb45i)fSKQ
zAUY|gOK24G8NAZdYK*E)d5D03RC^VJo3QmkT~5Cr{)i|Bko!@!wAA>rG&c(5Rq8G-
zg3n5p$0{HD+rnlR*uv0_7lTrbxtJa(k#V}E>-7|%r9lpdv290Z`~LVXMs`dXgvAb*
zo#K>22igLqh(7s>X1y5cRv`c)#C@lk!irn7)1fQ3KeCbn>xQ*984Kgrd#Ke|$4Vo2
z4H@PDa9p`owe%TMdD~<}#~B;IeMyGHO|{GuzZwow!*`RK+=h-FgoD6BPoqGAtC<04
zW*boXP1BsrGgnOWdhN$du5pW47a8-Zg>=8Jeh+JZyWdycMC#~1)*xcJlz;`+FrwOR
z{B$a8Y|J{4?O~k6JEy#90eRTADP|1dX?1l*i%9nuJ05d6C-hM`|Kbusdu4IO-Q^6P
zKZGMZiqhxpapz4lJgEuU5Xl+3in-CT8x$G&*G8~|FJpKBCss+0GA?F&7|yfJdHL^b
zrJIi-(x(~+{!L5_pk8EgEeG-Kx)`SNveAf+f^uS~F66a(R3cVRna6l<FOz_hD<21l
z45t2`kM!pd%itI$%Q$V>u9S42;M^wm<u1<?RcP|YHaZO7VXelIy*60g?|vd1ujAWB
z@evu>a^RB-r*|yvAq&-|gRBI;X3cv+8b!QQisnbk%I#kIp)ovWP1f)*w6Rv}`QE}R
zK>CHR7zMGDFrE>+EU8>BlgTjP@M#R6wzltlmt=}o*9M1VE90vifjlmRgQ?<=kA^&!
zhZ*IPWj!eB``&|2@aM5DMQ>D`_1e58_X+CfLE`#23W$e2N++DhihD@;TFB)+3Z0v?
zoF#IMhg523ep@`FHO9psT6cO_OJsOC160tW8YdaC{BGqq-fgLWM8o>u0N0s@#_o5T
z{J%j6dNR#7c5>KaRowz0DRE|W&-CC2mq2FUs7Lv|fu#ddLh*VV2n8sUR0|8j3JRq@
zh_U*`mr<Ra1_9+#oU|PEy@vP#4=auSNO;GVpF`0_15GWsQ{Ma3I2R&85K{x!nl8p(
zuFs}63G{OFV(AlQhtLp=ykg<+)b0gDaBZwM6_+%CDHDZUIB1Qdtu1$ijpS2Q7Dd;i
zfZE>kq@Ttb`|hkbBzw!_>Rgrh;pQofM{I49upWBs(1_QCn%nBzKv#lQdgfiZR>c7$
z9I0ja`=tY>j|7|h7%Xf?yeR=zGrXqNM1>2yDE<Z%;Lk$!p}B&B;vp~qqLdstd4D<p
zObFQjWP3eiVQUrCoMe^X<yIZ~lvI2Y&;ZRkGPqQuV5P75#HMJ;lRAi}y8S&Xm5Jc;
z{N=PM(D=PYXFYLpk)=;k!XDJP2m7*gb!M8igp80!ptvi#!1Ld#4M>mTzd~Y5BpKTc
z#eg7CbjF8^&~AhGlt&ox%1anj7<rF4zSt3(taTSh`@jT_gT=x;RO{Bw9<|#z82^fg
z{X-i|28hbs<7dd_^so}!KArG@q0CeZ-U3Y$T;4-(Rwl^(wu3efNFX8jUc-$NO0k>v
zR*%cd?<%q4?hn6CtcY)%G>jX^rt7fuM})?Oq1AGP+d!0AB?6?!tc@}-({~^Gde-o=
zosbpSu_Z-k67gLmRC1XYg$2wQL_;JT2Lnq(**$}hI2M)pcKP~xse>s1W=zd<lZJRm
zMC8T)7RLT@R$4fcXv7;DD$>SrjO{=ti_2uaF13sDw5=Cu<?pqAJ-R-iYCP|9kpFpW
zMZiLJ<D$~I`CmR>RcFqt6F(-oICWg%48FPEcE(v(vb>%EhM`1S{2*=+RcUipB7oEk
zAOay#%qX;Ow5JjPX<A9$Ji;}QP^r@rNLl)bBC&BgS9>eH+ZwF=V2K=F_eHn<zP|X3
zR4KI_-5SILwQ)q!cIaJchfEt5Kl()9_}tg^d2rFY8A0Qxj3a^|5Z?6Wt1QlK#Efk^
zBNiVeJ~X@FVPZbJU_kjnTJpYTx-Hjaj5y@pjh=1wOV>YS!N3e*+o13gfSnE69p8%%
zvCvScpX)saNJY)$PivLT=))|3hf+<NEe8UC3JBi%`*en|6TjYrLdAbh^i<gx<OvMW
zOs7qc@A=g9N6?|v{QSvok@P_m<%tue&A=y7hp48%DyXN}bU!i3k1a#bRybXz`JBF8
zo;KyZ)kxPb+Jb%_a#3+L41Yi&040$as?-Hsi(`%Gpa<c4z@VWO;8FaOg&|Ej^Yc4@
zwa1fox@b;57O+5+Qn1a@cMDSm0gmxBVHBGd9lka6h$!R#X_!me;P|DY{Ep8hbko8s
zak!zcji5cZ4^T}Mmwq|B%hg%RMRuK_2Cc{P@?#7VWeZC0mpgllR2MkU?1_ZrjNts3
zx&EBB1ReH4^S*M=J@|)+E>!wO%uIY#-hYr8xISk|2B?A;&zqhmU5@^G7++IjV36Mo
zaOS$*4J%c9kBL+IuM{gDw^x;jB!;peNK`mAEPzf>v!a0#6c0ebWjq?_3=*$SC<}x8
zkP7H+WWv~bIJe{On)8}sGk=c@{5)9AJfQ+hira!1kGex5jmQ7vel>1_N|cTtUqNsJ
zS<x?l!8;D*3cuq*%#mH1Su(6mm!a~(ykcjxql}B}EgPVtBNT~Mju&T791RPLKVPo;
zb4G?zjgM(o!T@cBY${4F3#~>VM*-M{klGS(rZd=h0)tBFG#FSP6bp*Qu4h*M(;C80
zmHPEhgu&hFFrkz~Q&!b_py$RPuF^pAhdN^7&sYksb*2V88FOVgMCWjjS1K~p&moX~
zhIG(zJ;DSueAsnv8EDEpL4DE}cW=619VNJgwW2PdB(NlyjsB?fw;%SwNa54E_iY=y
zny>6b#G$)0jlwbRpa713HEo|@3Os)6Xqsnwk}IuQI2>+u!8qnxLhh#k%OHz>jL6)0
zCIT<&q4L3>#Av}O<4hQl9Iz=|QgXwb;aDI7%oJZt`NA2N>rZ#rx)Sl=z?dayXpCD5
zXL!I12EvL6T0CeQaFszJo&*7#;dK6HGf%t$FaHMnBeKum6etaZc`6B6pI?;?!u{MG
zb2rY_D^Fv-W|t!S$|2>I6-q6h+oC+Ko*%?njmzzGvnq8HW2NWm>^}6<)!$}USMquF
zJRYyKx7UbdQZT?_$#Fz4ImGc*h*(6H%~yVkjF*w_B5<R;K1s2eHH?S*ewCpv^GPd%
z2KKKd84G6%4k9!rg@A>bG&{8fw-Es7M@2GuQf!SE-5Wp7s!YV<o_dP;JE)#))qa`B
zFG!}QhLX)>H>JM*?<zKXOKHBKD(*Pg5a@Iur;lPbHA4HM@HTPB^_v*=KfBmO4tgXC
zqwOIHMWc95Ulu`vi7;RaPMn;EA0Nu?GZaKd@8a|ZA<_o!kQ_9rCV)iwsKn^J6IB+U
zqf~s`Xe2t`L`H}aqOOqm=S5A8kJ-eSkmARs)bIMqyN^LQo;^azqnhrEBb|cKv_Meg
zj&KxF-MbcUA+kJ%BCftOxa7HU%wKlayVV0u8UH1S>aD873c}D&J|s{|Ctq+&|0qhJ
zXqnPP`uy^VfIXZLe_I`aF!YN)9RH$C{Zmh18d;NN`SJe!$y+ZB+f6I0!K{vbGyNf?
zJAIzu^0e~O|D7{YPVCsVSWpVq2z23Z^s8#(dihOv@1Ld0H5O(mN<I`Y?MuQ1_uA^w
z?sIhdzWh4fU-jNrDfHVy4M|UsUyW8A$*YVO3NWc@f6ckftftQxHpv*nH!98y_chAx
z{x65aVHPDwLw`Qxhtr8Jo#Q9xh7t(%kvkLff<BoN95_3mLl#`Yanj6mV6{S+N|AHf
zi6RT#&!6O^jk^B2`pettob<&-u#%0vmws~RyE=GrVP%QnsVV;EXG3-ib+nIdtdSqX
zpusxg`RifNMbdpH9Qt|mx+cB{i|Ay|iw3vms_rC;m=vs(LSzy0i1!gyjQNnE#h0`k
zp~tzv4Ot;E*x+H}F|Ge_QJSs#x6R3f+)slI2>&(PFnag92(KiJ;=VS=kA;)}B_Ywj
znfmjR0GElWg+D?NPJCIi`UlnbI%_Opqf#HE$sVcO8TyMf4c5gv36jr7=8(P4U0gnI
z{`F^?BtEV2Ds6lX8qNjGtbL(zVOUr-fh0+d%7wldPl-yKKd9?x+MCPD$1BaF9^~<f
zk#i3|@5QCd?IMYzbnq6%w?nP{NEt3%y}U+p@D`u@Y;!40W_3iWf4^sJ<A;m!lJt2?
z2f_mMp-rtLD|YoC`WoAh0_1bj?*)8!1GS9StE$qs&GD<n7bvpj2-cuXvsDZndyp>!
z@lC#6NVZ8YR(eY&gAy8@vkWfeB%H2U<?K`BJAcZI>ZFIQfER4saEvF##G0bMeFw=+
zQ-g_zq2@_#ppU+f)=U60!&XmP*%qJZlDnn>_+lpuAwmp$!mjARoVh}Z`-e59kqx6w
zKO7y4<3r#NiUu^;X$<pji*~}cGTncfJjy=L?pGu%b#I=*o5@-#LLwQ1%*0ANFTi98
zVQd|1mi)gz<mBmYl^2<;J5|GzHpQso7C+witQ-^0S}PJ_m{Mem?=JCmnVID%BYjfZ
ziYgo)#W_n&KG>`ZiPDZwyG<j`cm}srV?V)T^g{$ia2~sq!Y3`<FE%L8-@_x{&*B$u
zkEQFYls_e61kLQIDCpZlO9Q#H-=OMGAM)RW@cq^w{<R0GS|ayTN@>_S6Xunn`Zli|
zg>2lnJdEjN)is{?+B?l^^>1nu;dfcTKcGyqoQFTxqpf-Fm87zX)>&@`8%6v>{cHq_
z356n$F@b|;b+sIkeb;2HsPuTQUH*3Ljq<6gbsc4dS<>Q#SR5YE_?IYS;&2=M<UAMg
zx)hW!yEJ4tl{9>;w8cyvT88a*^9?quvr<w`zycB(A`zG26#2m2u>c03V9D9#JHssw
z`woK(_hPyy2zU>}rOil}Qbn0faMUJyqYbBYIWh;cnyQ-i)_>_fDOL)Cr&#-euw?90
zn54w-nJV{AODoR@;Z~>j*xf|o+e<SM+j`#WgMJkB%*8L+Ru^+-Wc)q(-Ai)*V0I(y
zn=_D4N}$i$QQLO8wY?F07MmSDn^UoXP|D1@F!2O)q!p5=ltxUv<&~;#lrZyhDu~}A
z*X4WNT*Q;U9W=WC>dRD+ja}(P-(7a|UzxPM7I#1`Kb{in;!J;9<GLw4@?3dJ+RLaS
zw+UPChX=(Pl674(#jBTExJvFZ_j_y-w{#>D19lx0-@_^^A({{R=iqGri0yPS2VE^y
zNriZgn!U&dx*43u|0W6oI8#Si$y9cS;&<R6GE#E5w4eh*DLKm+Cj15&<CX%UyK<gW
zO!@6lY|T~&q#we>^%VxX@BP6lWMq+raWRG0pvVANhiMOv<|bTT@hwnh?SILfm{-k~
zn#!hCO}rP!%enh9fT>`x7Gl2VztB_`Q^T!)LFHga)w548=FC}s;C35?OAw>=<~_a8
z7jNEga~S`E__Eq8cWT(y)cGwunc}S`-uE+(w}uczeVrH7zNv4>?S-^a9Wup)-)Q~Q
zdaka8qjJd=yR6Zwp9D>E=&F2*HDOkR{^++Mu>NaLK0vlIKjLX58#nAL)q9Ne@Otb;
z5^YrXfvO4;_(Y+ZK4OueZ%((S))ogtb0nS{-@f^u-zBB;{aQK2`$!nl_L1@c25lcA
zIU#_L+tmPma}E}Jn&~mByL8x6jF6dj9@*h(K-C8TL!^+xPv^JRmB$Yn{$NvYgI+5D
zkfTW7#Q*^yLKkUT4AlaI>?Vf)#_i9S{7etPhpp{ZtGd4&DsJDedke-)zCEVBef*T|
z+#?%t(vn<_0Gb+;9?2T$jBpc=b3Wb`2|;ZCAf~PpXQ%J__oh!uw!fbk%LpuT`4%2>
zbmx{XbUGq7ZZ41!upMtDe1^hREPQJx{-j!Kc&(bjV8J@=Dh}iOPoJF{Nqu8aD8Npu
zn-l5q-jouV8f|g^OvRpHg8sO6Ij{e|G0)=gaMxpiPFjNOXS<yN*YAo9-QOmp>2Y=i
z=%ajRPckB0zy8O*u`V)L3x_&k*pj>5^ik)Yk*aj4;A^BSc^yfv%d)OC-kj1gqnwkl
zse0sXGsV2^&95s~ks+yz{-rwdbKL-(qkUG(fAAdVtjq-NcKw>whBKPQ5cokB)UEfc
zj`!js8t%43!MfHPtsnWRBvxP8R@>lL7czMmHNy44Nr&tB#ks17NtN%$X2XmxxGvq_
z68<x0M^Q(uz}2x}$5jdwFY9Q|#ZZgQ^#ZG<Wa8hAFvjj1gY2WT#buIrJxYUC`Xa*L
zW}bb`539px4aQY>o~8P)Z2ANFZmdA=Zx9Sdkm%FTv75DqtbA&u&U#PTM^ompH7A{w
z?DgJ-76j}x!o-qb)|I6Xdbq}fH~ZS!HvZo5k>a-{atv4<if2@euxKd?)Q3mh4sx?5
z9D*dMnX9>~>gaZt=Si$23BE~K<olkBAHlSs_pN^b;ugnN;kKSIBa!|r@#FE)c(}=#
zAphTueXpT{)7z(hD@>S9Xkys<b5g$h(L`Ap8~BpIZx7-hI{mksPVs+P>8)S3TdQNZ
zN|2l0ea^zeOwb<I{1@t;3f{BsHLt9#(+;8dY~^8rd-pZO^Kq#tzV=4vgWpTR>s!>*
zTCtN~$ePa^X~uGzAFEGQ-#4?Y8NKSG;FL;_;_He&gu2yl?7xNUUgJ44hr2Uk{Zy$B
zHXcsS+V5{-Em^tqJyd&Vxk4p1djpS`X~x8aU2Jxio+5LOx&wwVeu?h<V~sr-A_3R1
z?tHQ9{&oBO{LguPgUo~)F?Is)R*-@wxe=O4t^P(c99-czf<GJhScZ!HzJwXt8ehYF
z;JWi?nOEWMqQ3m*!yWBOde%Iee}zFx+$qUxwKEC4QjBtN4JtG3aWJts*Z8q<+xtYr
zvLKU`-oa*zCdI!wr)rEf<@m1V$P#crDe7~Le+J!r-Zm_jeYT76WPZ}St6W4%tg!L6
z+HO|)HPV#<L)%IpJox<0TI#}p2kM^W1J~<tSwt<iRs!i-0$#uVy?Q4wD?aEl&6LoG
z@iJvU{S;Z4tW**C^Zo|`n~3jaaF^v4?!@U?i*U-q1UX}0i#A(Y_dn#HJ@*wyf8Mv{
zgkQ-q5T#{ED+l<kjhMpLat%KF-wh$ah&MZ4Lzt3=7@py=Jo=x;X;xZfL>n8}?lSUN
zyXUS}GT#cT`xn4Z!>QONH5CG1x%3RTxaEhteFPwXTv&6C6*|Mm*+pB#&X(Lv|J&su
zSzxx)ZzvafIs>;|x+67D=`}e!rARIAH{a*umW!oG`r9o0l5Q^-rK}YGEyd@4MFSng
zUCXW@uzz8)Ai}in@^wE^w*D6E$e!KOO&~H1-%w&H^t1b(C$w$B&~vXFCXy^(T1=XP
zhCimpfeb<yn4bN1?f>mwc;h<CQsvpF-qdHn>Kfxl`53=-d#=dyMP81ftd25M(z85k
zpVqO*P6uHns<Q*^l7r&2Xlu}H-a#mj`(+b0Gi%~^=k)w?T>EvfmDr*ey<?ff?eA-!
z(R#1cYs=C6y(xWZN$6x0YJ|}9)pqO*=<Ey|!S}U~qMT(Z_)EW5lQw*v#Due|$(+ZS
zfW7|qd_SZX=%ihAfccR4h2Xh$m-0(KZ0QXegXZ5Rs{6ME+IjVKABOPNh7)|8lfL%8
zQj#gQ_N(?M_0oUi>sojBbpGStJnSbN!}G;)_Xsh74EtAqHvV7N|F$e$&%<eMC-J_}
z!HH_$)oBHlxUj}hLlm+9xWK!x>-PqVuDvd-wN>jN$vY5-;T+@f1-!MQB6rj`8smO3
z4)=gIV=^=RGJ?((D*n;)RcD?5_NyHHen2SiO#MlWo&p;5G}|5gw2AP3Y*iX4Ax&Ox
z$XRhQd>?F=@;x?(Vb@YHhSDLufw5;X&S~Q*p$%O8j+-bMOAag_hr}5LJWVg0kjdSf
zG`3@6V4_eAizc~ngTtZREJ1<(FP*+@CuEOaj78-I8qwGi%9AVL)#*aE)_=)&8@K3I
zpRKbN;19K$ipIh1AJ*lES;~h_Q@cR&N#2AzgHONR398a>bGTNg<M3g;(?W7s!|E)P
z7%Fa}7aFLtZccPFSQzvbp-S&SrK#bW%cRB>AKYKAbCWo2dB$3nI&U=<n0;*0Bb5F9
zB%}wG`g^l$99+vJ?W*V;OCWx+zTYC|Z_0kjk#{tJ<`gzE-oIHX`Vh@?4$q|z2P<CT
zV!IZhmyYZOD31@faO5x`0zt25Z?ARL4sXve)hEZbpW99fo$y1NbNADVDjl9|6s4%^
z@|MIZ+J)9rIL4`g#<z!y_K|w8&mE~O2Qf@!`Q@TFi}FnRqF!$qPl0+*p-E-CjVax}
zS*G@?NLpxF-ooI``gUC}RLmbq@!x0lj{|r9B?%g1j;Q<1M9VGh>yyA@0I=J%sc%=e
z1_v2BmDj5`x(m3^&)Yf$*UNoP!Cj?LxRj8p4HElO^f%EHckr8|*f-8!8y8OsO_rM%
za4Fga-53uG11$_X+@gPr*xve-B+$3`>x)?r4oz0E+W+$I1n%t)@Nwi8;~Qz?96)V5
z49V3zyqJ&*ZEm=_<R{y0qR6zeW}8f6G~b<|#<T|-4&XPVoH8Mi%Z2F-ZrKdLApVB*
zR;2QRg&|r5JqFpQzs=Q!Y`vZTRln)>Ya8Jsu9i>={ge!R*c{Ob+`MeNoGp&&QNof(
zvsnQPk%TmMTNPw^;$htu{5ivKaDD?b8#T(QBr_lb(cPQ_Y_=~Od^xo{&(u+ogHiIG
zTMKlC=L&1Z^k~VWN(JT}t<Dqv_ji3A2}8%EMM=(;$g3+F$AK2s?qi!vOm^`~4$|GD
z-mF%`OnCgCp)E7JG@s35&iv0d`QWjebP#LhP(Xk%YqoKF?pbq^PThNl6wKwXG-n%`
z;|E<8%l9Y{=~6-`KKc!KYt!XPvnlVF790fShx@#@p?hEPsaW-*iyD9B*<z_$!L*a}
zzrl_q0HPu1Gou%?SgK-h(`vBASB#yXi-|D{4}SjOE^*M1>Y6zd#vL{O2EMYB2Z@XU
zs@35ED=#Zf${J5(%3r#??`{9jHXm@Cn!Tq%I0rAXz+#UYc#_Feewm;(ObA{g@6(ge
z(wNsCV+DZM5@kgvy_TvsY>Bw$jw_=li^CriXJ^lhoB@>ED4+q$T9u4w(~buAP*lsE
z)Yp&}XyZA^<%1sV(iA~c4#)cI6}G;@<^*13F~4U;a(NvCN;l3oadHW9;|?7!-iUWt
z;1SWd^MFL<&BC>fanL*kGzRI?Bi6ih7)^I&&hO=kIl?|g{k5U`Wibd8KW>|y`Ssgo
z{r=N}5IcuE0f7@CY39M&sb@_`!)~Xa7j3>Y1C5c}@U#i<!h?T$?KC1q%O8QOq=Fyo
zZ)v0pTIMF1Ifp)h@PO_l>SF|gSLM<&?yCyoiyo%(;C^9j_+(urf5EG&QBSz;Dh~aS
zO`VA6TSr=%JsgE?Zf)27RQ$v(dyi|J>B!5N;V!T|q5o)AiAhnVnL67^c#KDU)0nb9
zU=e@|7v%19$Fd*r_)FEe9?MR6Fh730E<w{5ZD3u|pTxn4zk&+*ER|pT_%GzcfzWnE
zhN4HpR`&b_h1{N`$DrmFXiNBByvsmQwUSMCVa5Gg`udR=xZv7QbpNX#1)na5)#XH5
z|DBEg`X2gp_VA7cbkcuXCgNpz99RzX%(?PUK2uPf<4Y8LO*aB>v|6tW$}nAYF+7Bp
zbfpYalWq{24k`VzHgg~Qfg+K!3EQRnu{Ov1<;d?7=RD^`8~(~hf;8lwM(r0QCS0o7
z|D%679QDjP*PF(AlpZX-(}96R0z(eJ?l#q~q^7jqo;w2GZQqD;NdJ_f6(zD%pb;Dl
z&j@Xu+>&@>iSvCKfvT-|ihQ5CRyp9msoDMsVFrwBP5NP|R2ci=ZCLTspeprQ;Lu-4
zX*LYEpLYz#O>PTr9oJcZfx}>88KjeAv%J3ij=KcgWJh-u(A9T0e^y?eu3ttBOXWH-
zT2p^N{tI8sl{%LZ;Tf57&X8HGj(aQ)6lRUt_sh8(xx}m#5O7bqCh0*i7scYgy>{7D
zqY*-z&5<W5&#?-;LQmTuQ^e47;&F#NyOh2J_+x{BU{c&*<Aw8$r$nm|>+(NPOyEWN
zinNO`b|npnE4%@EzZAN#_k|+Og>_q+qS>TA(Zhmm>2gQjq4~jd(y~GhSNf~0^x_nk
zTv4UngP&&aGxm5%=9`)43CE$NKvuypztW~YeW_H}xq3vKQ<zQbBq2t0^W|ieB9WIq
zveF!H9&~a7ZkHU<cW8iUID|8q`Az48FP117v-uN;lT!uUD;F`_Kit@>m=LFETxEzu
zv<0D`MO6m05|nIP-^%(LDvVZd+~X97bY#ATT%H`No*g{S)=Le|-@iPlyAj~<D?$Aw
z)Sh1=@MxlA#{<#7l6pQzAxrUVjyE|(l#On~vsIB8^{67tFH7enjN0Yl&{T26R%Il8
z(#AXP2!OjGL6FTReKpEpM<Q0)nI}1jA$cG1zvCR(=lG*B#s7m?!ih*RCU~6Hp`r;B
z@GuC+?UkHRdwDYv@o`Zg8zSrxG9fNGRm2I%9<;O1L@#t>l`_6%gW`KX3<{ET)&8xo
zj85x8>MUrD+4^)tv$Ej9ZHRx-Y>pjblj)YrPInkIZ_fQ$ST8ZTB-`>wCnm+gm#70l
z3~4JrUR*kZv3h}pASPlAOci6SpfDC5{m%U%EnEubYA>6;eohzIN(%@1x_zv00;?#N
zJ+Z903Q_+5zJ>TkZ~3Ui&SKoH2V5eOn6&MONmY-i9153tOG}#~g-}A?!atIren<Xe
z3u#&+LtBjY#5CDh=H=>HyM6sFtwH(GjqB11!yvk&{u=%7;u8}Zbyr~#T5=77nEFud
zlDno9ujPQ8lrhqW$&s`k#V-rD?a$q~Vup)2KJJ72f?Bw+(2}&3QYOsTg=JsA?KXzh
zksgmVf%9PbeZ`vUBkfJvZF;lrI<Oq>2BzccBaC`9MWTrjwioY8_xmaqi{Q-(t5#T+
z1Ln@v-IOuU2iE<%qI13-&l_>r?lGh(a5_$^6S`;*+u-`r0G)8Rq?nhb#jXM%X?mM}
zwaGBm&{fd(X_h8ifAt;~U?bl)jK@27Vm`>3q)sb|wG;ic+>(77&|pDE2#_cJCTA8A
zx!!Xi4AqcpcV=HZ*M9V>9O>qf#<*#TFgOYZ-vww5)9dilBe5B@5CefO_V3oR(_stz
zGdmbi!N#5@))t!<Lk_MNe?NEi)O>Wno(CK%_;@^<dX423+8Z1}v#IPoW%~CO+BP`u
zVtW0>a-sOk6Dqq}l2}oAQg1n+@t9pmt}3DD<$W9B0bp%uVT==q+(=F^HF%@H>(u9#
z^6Sfg;B5k===g?*WTVNnCpdtDrdH68n3udtRI?J*Ctw-pPNdsK3o{BtjY*`i$J%l`
z^h#zjk98BcH_!U~$6PSjS_Y-tOy8zeiyw<?>>oyINy^Ec7-wq5L8=idM2-=PJ%4&i
zFh^-fK^6y3b6-6j@22Fp_O{k;!{lhg5Uz*Z-rvg57rW<;Y<~VD)iQA?#K!Fh&*qjO
zIdNpWHb2I+XCf%uJG$^!CvkDtTLF?JFcu83T~V%{K90K~`rn{qJoShs<*LH{2M_Fc
zdAVhy{?g!6^y7DEyn~*>Y0k0Kx)AR8_siI|*pEfZCSe3fg?|%9u(uVXpd=6C<{w=~
zh6iW2cI$bv3&NzO+<vqg>jYc}JRDmfj0Xg4bRuxUjTzKf%!^P3c)t&i_vesyG`?-#
zuQw;CdYfp+OzAG-v{%$JeB%=&%E1as4h2t*TD|)?Y{<4>HjJ(N*MBaeiHOD1YwQq9
zsGzT49Vz(hwqID+EC=bvIB7adzN<*9Xb)7^yf~_wnA7>fazd{)g6DBUs?_>+xb~{!
z)_+nX=<zo?T}`EpHCdW^WFbc;ECq0b6R}s?U|m7f<8)?!=trFxrt?PQzUkvEAT|pJ
z1;iyfp!mK$mua(f2tRW*e98&jb!$<H@3Db9_MpKlNZwR;-N#;e7B$nZ*AY9H$qf{p
z@uU`w6ric1-4qp<P9MVHq!e$r@p(2+k0D)L?%*E(uj-rXP#`8WTx{z?&_@l@ex24i
zUH|0Npg6QVGE8`^KdaBh-N%aa(v_v&80fxHxMle^9n3mg6DU2KRzLUp@*xR3t^LjW
zs*i35f+F)7(zkJrJh^FpxPK+(5z*%2VEL<I`rDTakIOpylmo|VNd!0V4<kx<z=GM@
zPbFs41~=319FGGC_hl=7rWe1{TjzOGGrorHMV_<mS5yQ}AJwqM?iht7pX8w`FWtev
z%i2L_uV0nuLMx{!2$zD{iEr9Yhm$hqsMJIOAa#LQ&yb&#^|M##J)!DmRxWYvXDJn@
zOA-T~j2?zU46rU?WZ{p?CM_)iw)4SL?!U2jWDrnDlp&b5(rAt=vpS_cnQPy{PAH2-
z7C6}c-PY3#ui@Gtm~q|<P;9Cf)gWL2<eTbv=YsrKGRpW)`<3)aA@LJ+uisNp(H-93
ztZYCWtjC<`tK<N22m2pK-Q^df%9sIyakI*_!`i@UC&>3vy8@By9@jxwER3|;g^{Z1
z^D*jFyzr`LwVDVonXO<{$I=F`xNiamllVd;(aEq#){p&F_uh1@#T{$rzJ+toiA=<h
z4>h~d)s}jOz%B3>?dBn^wG}orfd=DYYk)HQP9Q$ipBdoiP8HRrM(Is{H^)9OgEj*S
znr8!<nlLkMc&$fAq}9*AP@Z_F_>q6Hv!<<_J3k=x+c-a8Y0KvLdT;(yNxcdsQ0wmf
z!!r+xr=GJt$O7Nk1e$ijRsE^k`tDQs;L1=8n+kTKnIpIixO3dEY`x33LhVQ1Xzfm?
zqlKj3D$7CxK>!LgrZRqF5&?L%F)vD02SFisVOcReDQg*npXyb~SZMnrY2)4ktg%WM
z8Etnbt#^lIK`E(}By<Xp$Vo%&MP#6QIstb|&Ob|x1NQ_P=q|r%{s7(e|3-1rzTDN7
zA^K@7ecP%uty}q7_$qst=%5_)lT`@3>nKtA+!fW*MFItA@gr0clxHNO{Fz<LFjeZ@
z&n*|%CV*{Y_3!<eS)-Pr^s(lfkNbYj2U%VQ&Ir8j*l~Ze_z^gAXusSZK|ZB&(}3g9
z{3C=080>RIMfeay?X_dJn5ttz@l2l<@kflg2o!Sw(>omOaDHJJ$6Z?x<$saBxma7g
zwZLS;yV|fUqTif+-(auB?+@&|9B-V8rcl7$OF4&Vt5qhgc~;ekOIL*S9C~Mg4Gn(h
zYh;)FU6;ZjP4MsfqNk7%-d}|;X}mzt?8TVl&Q80zloheA#l3SC{R-1hUR&07c2||Q
z1pDZL`GcGrIaBS(K=4x!`X)8_!wH7R82x%P2!h4LjGgF>drR~?AvRMT8OmVkfrYOB
z8vW0`DcDt?X3-qw7wZlQKiKm*Zn;~5LB}yOv#6)WnrOk}%n|e)u=mBcFNDG}2A(L=
zN^D%bf{o{@9y)30%pm<agH6z`x%i(PjP<Bfbzx+hq7>c=5!J#1d>eXcrglFHjX}!r
zVNvgO|J<u>?Vph@T@`q{A_vt%rk`?vw<Ae-xkz#0M~jVWK61uDD!4;-h(}uZ-F>X!
zIPlY5@$@jEH?))6H!;U0a@Z^O((Jr4+)>?@4X(rl+1&d*ThvaR=cZWRbQxgK9{%gq
z)6e&(!S;grZ^4`@{({Gwf6%^qnJQ>VDo>K6B)XwR#TQH!B%OFVhi^6+f$g~jUQ@#3
z3luz`?aD4UyjBxsh?G>2TUWH#B^7p1BJj9F&pz5ndbR#F9Fjt%CCZV^eFnG_@=TXT
zDXC4BRLqi;C?wE;KtCJ&+aJ_?Kd9Z)GEf<rVg>kT4D7)h(N^Cv_1>6gV);&2BU``j
ztQ<-4@k+lKMD}aoUMdA_#XQ4{c`sbUI2DO6J&GnlNg~5?iV}KS@R^Hoe$H_*Id%4+
zK)0Scd5IdjB6r*!C^_H!&$ox*llUAV0R!T+K>6(>xM$%Hk`#BGgW|w~Y&V<Smy%Qq
zl~2t(ZDWQ1_{03=B>s}8vS#YdqN+yS{cQ#l1?QpcYz#VB16<Wb`(=((ucWsTq$&9Q
z`<$`qF`zDbd2|&F%hBC!G65l;{%QirPWF@%_g(jAbBB?)w<HAfbj9dq$nbZ#@IMU;
zaP60qF$m9t<}J~(k-iC#Kjfwop}VPnWt_@U{c?)?ull$iQ?2iB0`FAgynLD7!ARnh
z>x^5T3}uFQ{oy{cowGuYk(Ma(a!p`)#hp^phGmP1!BXMsxzNyG(s#Ff!hgYG+7u$`
z<4cyZ+6gYa=Gi60-?!1Jp~R(ni&op3o&&=Zo2~~@Bqes0?8|QhAAhHj+#>^%h!AQR
z+Y3&-Ol5?!FV_+uTh}S1n9u(+Y@(;Av~Om#oyZ~8IUWn>-p1!9r<B%woD;}m9mn;=
z=VUGgM1wcK3+!30ScVXXWf#o#5B(qtir+!#H^!X6p^Vh!U4VizWlw!v>9_J^C=KcN
zV2Lwg4l9T1{9(WB?84CaQyFTlnu<HgN+kw^KqZ+Y#6mX)(bnAKuR72_d3xzdOsoK=
z(VR5W-NK9nKJ~nblKUL}I_<HFc4<WXXOaC`f#2SiK*eaKUD519Cr*)Y#vugc!k;4c
z!R6K?f3HbivKv(<$I57G@gR^`&tNBVC620Yff^NNnMkd$`P$3bwL-Psl%YwjeAc?3
z$R6A5zP8HAaD^6G^MkjfRw5SIW2Hf>XC>T51Lis*DY6N>w&~Kahpn`<oKJUt!cD*L
zJ;a5}9lDcv+}KVucjsYx^axm-@~GGi3zm?T%{IpG!Y8S4r3wJo=^-(9G~kpiYP&S*
zLunYV`DBgYX#NA2mTNqh?#qeN?)P&`<mLqd<pJa5`DW8V>dp4+p0tWO{}hEZ=4#lP
zi^lmJMoP@CJ|cZD1{5dH(p7>O>Bv5=zBD}UON#g7H=B(l<tdZo8`>i61~RSPk1eo5
z(}&z9&FzS+A_cGQ&5-{ANI|#0dyB#tjoC+QnNvBf-s^F{^@ZNF&&t_mhjX9m@#<kq
z*eDRR4g@%7aG*c&+E}J|m1aoQ?A6E$i<(2<yZp38KSNxlNg}WUU4_fC19eeBcuYI~
zwPZj}viP*KH$g|RccSX_EQ?&J3wl%Q7d77QbE~=WK13H14h^S1eCKlmt%zv1%#uf-
zy_L^{0DvtK*WQ9=j5zaVX$mKb3g__Pb6s_roc3l3<!FFRRu?!z)^XqI5&&e8X%YZn
zS%WLOq_y@@g)zOtkn&3VF)aW%37{G;j<?5U9nR*A=2`vq-7=Ogyd|D4wpROWl{0Xw
z?CYnkXgv7&SFboCD!yv)i~wzZ5XUWQy4UMcc9k^Wp>p}65!}%|^p=OV4=*)`ZjV;?
z0sv0s=w>p5GFwj2S@B(JIy}wKI!jD#?U5YCKIIS>B0f*%gr&{jYK-`=cPR`inFu=v
zGgr{pJ-xgCEdc03D$7HhwriP6m8_V8<=240?n5IrnY<Qo&i1sQ(4#2W*%0K1bKgmH
zP7)BFwWcX<2yuhLDmsWuFyI)vY<WXOIQBCZkNC1}8T;3^Lp+|&Up)}dc2rYtg(3Jo
z%6HCP+uRF4r3Tiv=pErRFp`7-Bu-tm;7*@g-8pK@sHB1UqT%l9k6-$`51qA&u~IsL
zgz`2{6VP)AQqF!Qk}X5FKVksJiq3{V0YYjb8`?LLdhIjY`@;Cq-YFdy;>_gKvI-z=
zrvo4l|Bt-;xt^8V4h+2~ku5fu_5ii^hR~m1e%kWycGUY5hS&yHmV~tOXU7~v9x2&z
zIKVcVE*=TfSY=;EKQ8EV-*M*Z)(^GT`tM*^8G(=fPApSsil*}K1<;&!|A=zr9qY*i
z5!CsWlp$qt8jctTqAAyKG8?FpISMN05c0ZtX=0f&J^SxWy|7QMnrghiWDJvK_pDMp
z%8pb}4##OJoEPgt{^ViYn^{Eyc=wwBcS6@-^6i4iRk;a70Cgcxe`CbIeMMVH+N~~T
zm=oYc5?^|J^V6!rwyGr|9_4c-0Tgw4(hCQMUisTc4%(9@7+8H2Clg9Mh7$Lu^DI+o
z$Dvpy@)Q0XB7}mTXuhZqx7PUo=<}$1t0_MsLP0lJu3%j9=STLhx7%013=u{)I5`52
z!mBDuM^P7}(}c)*Yy=?WIEio5ic$b1p!BLBYeJrVwLwpmX!e7L`D;p!UjhI!&Imlg
zMF$2FZ<NxSd}HT@T2m})gPswuTiG1;sk$^Hjj|S2)3j}5CCZzP^(EmstLrxVz3N`o
zED|Z3@kK<C*Tu%}q108@sg87{SkO%PgP90e$G#*1>DpZZkD8uHt)iK{Clc~+-1q5L
zoalLH3Q&~@Pa0b!#~9+T{_*yP=MKcaoX8f>LLlG_p|vjf+(|3izOlHD4GJNdm0p96
z9-1@0?uYXkM`9W)`>~b`lp~*fMT;+07jpk^ZOA)Z&dH66pCL#+FqnAtJzJ8;Ie`4M
zSis6PwCWC+EmqUf`-QEwD+;<`W3c0BUdxP`WoKH|U?SJkQtLl3*&eaEw1TR_SvcTP
zZKOJp<Ff=@5CH9rO8MfR{;P|+XdBCWm{qxw;Pws6+cwWMh;p!YL}p5yT-DL=h)dzg
zN-Z$k4QQ@ZPJ30@tFC&!Ck>?nnRy>Pbty(5w@VpjoGDkHx$vRdpyzS7OQCAK9wI^_
zTU-%M7hhh~g+B#sjBz|_ufo6=^0^lfjSX0U0OHvqC3Bix{Ba~tJUVht5(fZw%8J&l
zs>1IEoY^YY`D>c{<0v+GdROoJvPHd_9EQWWM~vY=)-7(@;;@sY;*aSTfK)nm%P1Mk
z+Up)xI7^rOTd55VLawMowy0nE`-k>#IPudDJ6lB^!z)cJ;P>VhG={dp1S2sK&^s8<
zyz|NJ-Fz-hn_a0&2*3;=#u2yl=3K<s2cqf1D}^;0vOeS;XsZu?LO0}2Ara8@BghT2
zTGVQ}^&dr?kDdgW+g6Z(&GnJ4h~NE7yQ-BcBU%C|=%W4kz5Q=L5I1)m$tmK~2dEZc
zKa$TupWD@m2&~feBO+uAIt(WBrCy*%7FjrTlj!0T7wWC`!5aZkOawgURGB4_P5@c!
z+X*chlK?1zIAhc|6hAYS*Ur}rp;Y2KB7}fPl@Y&dZziv$XAP6kXueuq*!Krrh(pz)
zPXK^;x^VKrf%qBWppUcBWY>8&JOu<(N|2C1s)^?1x`p*WQW)#Taeh-55@M<Rd#-=1
zYe|Hc9lIaZOE{0BD^Qpuf}$>zH4E!QaOA}wWnJirfL6EohBZy!>Q>AYS_DKw;PaAp
zRQNwL&yk3p+?QB8oXnr68PZ)Q^;V*z-=j*m!gmxjBSB^jtusHA&w!jDShHEBkj(2+
ze~Sox)dFXR2(ffw`CuY@{69W=no?0|d8`@|anjN{E$sK)fk0#O%S!quiS>`9TT|)G
zT)vA0Gxv=F02eAS0HpnWv9tCZioG*$wOtK*)mTfd?>%>IjXreFs))(_nU5|cFWI&u
zDs=mI@CA~9^~G}emb$>5bs=w$-GXhj62$z{4XLIJ`h~lDhEL-s{aVgckm6{LJ*_KY
z5Ew)e1Z34zMl%EiylNOqEitF*XUu{kveE*W5Cn}uMH`H!@AbJHV`*_L(J;qOk|Yro
zrox7i>Bcz#@E7+EoKw(@rInGg-C96V_^u60S{^xNQD9cgx7~uqJ_^{nuRq<{7#b=c
zyE57zu%HX33wp=SuAvs>SjwCEH*Q^V6v9DZiM%cVfELyTH+M9I{)G(1SkW=%Ihrms
z<}~r{y@N@&<;#wG8x*Ci(z!g*9}z)Oc#x29VvJ3ju^iLMp8)ONVeJj!oZsUzGqI3}
z%?BYzEfN3#QaOizZSQ?$K{M8vEHGA}0|KvG?OL~}@%c?F>VSx_g6TOTnYPEnIgW9q
zRw$-B8^eQDM@RvHZU~SPKKS6)zIw(<Om+l(41fm=dARYyzJdFQDo6%rBYyzM77f?w
zYdRX|)T~+T6cWG<|LT)}d;Gxa&3gvFma`aPkBi3_HibTU)uzRF{O-Lc+U0Zxlo#mH
zv0&e3W&f;U@QA<!fqe5>9k8Gw{Dezke|D*y#-;(8e`a#p^5^&TUvby(E)PsZE*&n5
zZcm??7TVW{NkLwkz{xe%1kOGbEBKHxJVLn|B|c1`M>Rozwl3_;QE6J`*rRba$I4mi
zq&Zp60wDlnjPBbqu)HYbMLEqFvqI+2AQ-~fb>_<2e@P(^tN-4ViOULfe$i<wWoJ{(
zp8=pS#=2O4>yM_7kEV;qli9)MKQ>yMO<A#<h_%<bL`Os5i$Sj|J01ZciEw{^{59Kq
za_bo~s#NwPIX;`V;Mm9i<CUZkg!c5tF6v3h8IcVuK#^!M)$NVpU5sI$eGF_4u$dRb
zxsPSc+}AswA6S$w=x1dMI>#z9!4RRS=`XzWxaKFFY5*|f%v<U1o7S~aW5oYkA}Ch<
zK}6_}rB{%obpRay4S<K)OxdGIpIfQIDvTD8sS2+fN)=C^<7YE{E@#5$Ryw+eGT%-Y
z^b>?6aJ#tZY>NE;wVPJl;B%>(5E5|$zcE{l(2wr!S)xfGIhuBCB$+o>E~vlF?^gFg
zmF^!Vn<h~c^6kCD@ssb~GF)QDY&_uj`<!)@s`W8LRapN-IulRlFTUr6LqV__K_3Br
zv2#lW02qh}%i1CduS<Q#Of)P>l63k(Hn*CIla`m4Elx(0&+Hjon$8tZ$10p50)R{5
z&{`LaF|(i8j8Dk0=O8<kZAIfhD4)+isB%^u@Aotz)x2i>*BzVtTy`_+SpcX1`OXs&
zydD)2Ic<Myt^fDqF%<v+MP1hJ>5YECw6CmE#?4CX@<!CNdZ`G8lvEfm2~QV-9U4w8
z40$j#WAf@qT3@}e{z*+YUSMWLX3^_r37BFv`usJs*B#IG!;XU^>kCEW+)->-9`U$U
z4h>;n1`%=&2VWWrKNH}~ajYe|Ij_{{ize>(xVSbR(IF+c3Yvb+9h-aIU^dA;mM;ka
z;B%?v0I;&-ocZ@)B3oQJ*U!f|0|W^44yWEen8?0F(*?K`4owl?GYc9*KVpb+Atj*o
zE)~Ufr5?Qh=>&i+W-s5+1ecxM_DFr$cb`k)Vm#a!OXX|RdHqcQ5Z>C8E`dD<@Z$i$
z=Tv?PCTylV5X-iy44W$hvuVn`(^kOg%WFprVcgBFG8RLKA?s(DjYW>xgkFVn$Z8l4
zcwFbEbDCq+b@^&?hKVJub&m@vE9GYP_LN=?@vN4t4SI}m*<k`aIG9||5$fmN-;rHH
zO4!$*F0|AIf8JQ*?<(Jm(d`yO@@Oi*>5k_H&vPggS%Al^rkul$BMLxfD4E?g-VuV7
z0Fq$C-yhreR%PioW?XKInuxFNZ1_ieec(`gLujYZjeCWZbJ|mR{sLf}vDKaRH7-@T
zAeAd}tU~`004SX0I+`Log^*<|kL~Pds?yW>nKga}ez&JsAMzH)V~2?d4-6#NDxB2-
z0y~!F$Ikz9ySSAnH(Fs`Nk#wwGlW!U-F}TSJM_qoVFCd7&3(IHx_==4-h9ze8AAyB
z)jbR910Vd*?_b>G=MuSus{U+-tCIhH0HKh$x})xAE`{}sCw!6!24dMOhGOZ(KDWC}
z1?}P+q`5tf>mEuxL#15Qu|c11NKZ1aefHrUgDSxEX4pFegbLZvJDS4xE^3ZE+1V8C
zMSv|(aSVLS0>ET};GX9WG$gadx0bih#tIfiMq6vV4_eAUCmV~1uzpEXp`phA1dTJ@
zhyb9Vi*Qz_*Et!ed3QLyCI9Se*5TsT@D8`ae&$jbIrqXD;pa7@{h2+zue<xjkpPT#
z$G1H$0nMq*i-1ST`aH@Q#Wxe00rHv=8jNRN^2lv(ZxAzlJiD}Bh4W8X_@~CO@3h6u
z;dggL+(oEriRQ1_W_=#na<Dm**IqwSYdqbM0}Go&cg(eJow2SxUmNn@2jfgdBLEOY
zk#tdBg^2PP7ajrtY^@LHLVj<L*<%T-fm-wBX?Bma4I%8jKjaDrkKYeJyL0Es`v;QW
z%NLDCK!kut9a*=q;ghf5wB)IM-@T9n0GB3^W&an!UNspd1a|!5nr%zlYH#4I`j|_r
zvNn`07@rDy(Cwg$$SUNNR#p96`C(Zo6ysLolX3~Q00R-RL&ND42NU^a|Jac=X3{nz
zB3yRLf)`q90+%eP51zTMvu3Z55}k9Wk68d1asx`jj*(Q}SDL?7;kpn3JPIppJ22R5
z1FX4Hn2k$oa#GUmP6}%5bP)3G?~9%H!oe)Sd>-Ht$ua=&<JyDzlGeHh!hZL*sy_h$
zx*?cm(5rif;xE19#UX*C6~TlAgh$p$_plXkNo#G)=TehY5h>WfRR{^0f_}-)-sEc-
z%zTyZ-u^^xYfrrQ6(_cih#6>)Fn>)?6jIq@cqEbakM{!%0T@DJZ%=gF9NT4F(d^9>
zirRfd)xuayXe0%8bPvDe;XSE;ha5mk=ZmT?M9p}^G3;qgG|Owv^8k{WzG);Up{VJL
zQaSC{shr+UMBsNTa#?%*wQCo(-ucYFVSAjJAaIOW{Xc$LlQ^j^;4(<cAJharUE?Ay
zBJ3JSY`W*UL+9j(+2GMUJH%2)k2#aY#jW+l#z<gbbUp!<HM}MODC%Otp^@y@E?g4~
z&h!Ip?aPZR&+Htz@|zF#u51hG)<cxA$T155AKM*sDGGnPkfe@pG!YtW0=aO&bHwvs
z4_mPr8vFa=FQ_FpYV56bAz?U?tK|$^|F2T_GZjghmDq%rpWOOv*yq00<5G<AKvyzb
z2*)yoD?BQ1vId2h!9-;a_u+)a%?}$wK2&xLOrtD>`MgEeGn{(c4Ug|%`js1=olyXK
z`r<JA#+y&#*6{fGYV@OUJ-#Qd`hD(K6@@HoW91AA7(-~S4Nx+t&DNans#igm5&?{!
zk<=cy%Eo$L3lQo;LLysi^t$-MV?2<R004<pUd<J>5MuTE8JeR3)6xUhtT~inFGgP2
zJMgjoSazw91ggTIHsrm@=TUyVqKOwaEUdL{o!yd~&FAYn_Ei$Q_~a!cjS>IPR6b65
zj5+VQsVAEGct8O!a8n)xH>=;RX6SQv9lxY0=2qG7IgYxZ%9J3@5ZE8focf2mc3%G8
zpFJ|8y0xLr2U18hZhRBB(&VHQiyU<T=p2sitb%!NZ$IxHNu5OID5P?wbqGaW*tcut
zg4(jNVzYhO&)5+1px>=Jj3taQGOEg#?(L6XIIq7m00OO<sB0HB{K2K*j&UPK0!U;E
zuk7lNt-pIq(%DCCP9thz2oRD6>cie%uS+TCC`MTbKtzaV3MXwn82heEQEINeYuD)S
zSE*#Pza|(g50x|yA+UM=c@FlwVt<GTwjCID4aKsTnD!PcSt#3)Q8?i3X{ig*Tm~qr
z<(b;*0ub`KsH|yu1sUj8`RcCz_{GYxAIPFEq?}aV$sX-ZGcE^?t^>pwlaK8gVBfpx
zg{uw>B(BnhbTb4BdR>n-23?<i-RYhEcG{<1=;3fDY+kdhKAJ$#7U6}OfakXQuzy?C
zT|)pvNJ!<18-Dx1zN-{83oyAka=CJ5@~l7il15h!ksgl(Jo=b%ruLtNkS`jI-9yRG
zEous#Atb@wFZ6+K$q*ICVx+^c|KRX_WBlK-B3uk$l$s?XSkYcvKa$M*sbch3Mf9vi
z{GNx~YrO_IDqH62YH0`;f__gzjs>Qr=hbu(+0!$!)|hX<am4Z=i9ia;&Rg5KV^MR>
zxD{al0O_3O&1>=_wSK7E-Jd2KquUb|=E@N*Z*k+rCoRm>hrJVu01-esuV1Wi<(Gbs
z8u{t%+iV%<j-bPtM>^Fwme3p&E?_<lVYY)pdw<FhlBzOaUn~<G7igHLk%h(@-(9CJ
zt3N8oehV6F(teM-$2r=v@)?V|;ooy;q+=jM$GWYcfp`w$*`gC*S6lQ{L)f=-HuT<l
z4#pVbozEU(k8U5jd^lbBvMyu|Ab{Va_MNb-<(qFif8{Qc5*0RPVw41&*H?hD;wLGI
zVAF=qJu1gPcvMz3Z5k0FUo`3t4yP{I(jBiQ3*uYpKnl!pF8+o^4NtpO{>REze^h}4
zb46nb0Q=2#59~SV%fJ5@Fe?vKn%2Z_;>@(ejK3V!<6}s&V*voNib2L406>@QV#Y8$
zu7;VdEsR{D_`F9g&48WLRx6D){<NyFu5mFp5acxjhm!da2XM{1$s>vY0n92XAR=7c
z6#Q-^;M!G+2`cL42mpG9lA8_;Bwsd|$RfbpPB9RMut{OwTI2gt#P9yIOJSnA1u*}j
z=@NQJ(yts!WPj;VmDV5MwF}5>n`A4$bEdQcQ<!+d6>f9r1OPOD{5$(hMTM+kYhCCi
z01$AB-dHe50tgN^g^!Avz-4VUaYf;~$I}i3fPy9<nbm@R54ex<J>V-l>K2MgCa~*5
zB<e%%7iPgGFfiMVARtm&ljrCL{UTQ|Y5@SeDu)x6Huj&huI&yfB$JX*h%wH&dh+7`
z(aSETAtKh|2mPc~ZFffk-YpEP>5|zwLes^o{<VAfWflNJn&Sy7hzL?h*tp`Yv6i~P
zy<xvQ4%LRUTmMfIz+gPPbR?Po>l-duc?Lsf4(-8Gk3S*=Qf0~hbC_ar>3AIkjs*Z{
z56E^_=U~rJ;tV42Po(CYp?GF@f3L$4&S~=WrZo*w&e)3xRiK3#)dLYh$nRd(9W8du
ztML&<)TO_ZBv{(s(6^v5@<WB8a8^}0uc9vf`}&g~UEW?FvKo-jX(hHj_&wmp%$m-+
z&)0>#JIh6w#&kBy6?Hr`l73ZRJo~dH&7t*N;l#|&YwWY!`dBRK4LCsIq{^?%=b+9u
z^Tez`*4`5W02qj8&O?U2O6O4KumoujZ#lSmt{uKL5{{O&c+*nKgFxde2oXU|&~w7R
zzU<;-*;ef6&$u|}uR}y-+|6PDUYC-+>CxT0XJO4H01)Gho<Epb*fp5^+)yG{V;}W5
z*ZOyETGRaZh=*)2Qf8N&(|zUxw$h1Bsfo6bQaN{J-R9;R?_b?2Pgj#zE3HP2rt)vN
z;nDrgoHJqw#%x!8S0@8*!MB~e^ro7C=bx&=$!Y&>$7O?9Aqo0M(u@0|+237v|L#i|
z<4Sje#*DtA(cE@AlI|zR#1XJHYE4*4ngjqi5KT7=DcMAjAmnpFQ`qNJy17Or*=#{O
z1UPA|w)(mK(WUb~a{BU7G5A~x0)qkF(o_?8p}cs}QMpJGV(I+CdtT_dnwbd(bDIZf
zjx%9GN@%JH9Biu(emme%4X2T!3NV0(P}C(1r}CFPyLa%~o1f`>_1<9(8E5hXSqApJ
zNBceI=ldQrGD?JIg0(}&(6VJqr6gsd-w*m+P#^M;=`iLz@t?U6Qt@>Dz_>$6E$tCP
zuHDu(yzH1o;fV13-a)l5nmQdE`Quf7PA+bW>^WgcGfj8Y*95Y_IAh|#NTKom=MH@4
z&~SRAB$M7i0NGp{{Gk!{d`+4?fw&bg%rkux69A==tVke)lyKJig}(`S)qR{{)egYc
z6p@Px63-(tn?TPc0brd1At5kfh<s;L<VRt@s|!n`j7Mi0J7rK3KrB;Oytgm$<KI5G
z@BKnjm<Qeuo2_mlc1s&u){f+zjtK^knd!Qm5I7RBo|wiGISs@l9j8=<Nw=9?FD>?P
z&YZh6BmqEiFrMwHY%5gxU7=_+92|%@&+G9tg;bUSY{?Z49YJ6how{U5<?JV=6aXyw
z+|t_XLWU0vBwyAuoUJQJteOWgYl1ChDe+b}gb`0>{v7eUzUo$);XJk=qA~4_3_&oG
zDxCJr?*1R#`&`%8AKg9}Zf^N6VCIMuax#JQ5^`){K&Z&GK>#S~1Otg2SQU8VUVgU%
zwE;IenU1+qFY8Y|nJva}5;t7XjNm{l9XwVy0THMn;$JwD%!ejUyT*LLtNv(BN9~OJ
z49=MN=hFw3Tb|tW{{C3@^^!<+6eFW<$j;%OoNDFx37GE>JGpL+c2AM4X#Q<QlY4M^
zd)==&!~A#^-Fe-h-l5b-Z+N^bgos2q=choXy=KTi`L|t9HHE$J_q&y7*~?Yfe*gj^
zfK*Oz+uAkqzqdWH>xO?k^vRVcEebNuN`rXN3L1`R_{C8NfC}e8Xm-H1dmtrpVy)Mu
zwoX<zQ4{nXSlQViO6!Po(+AMbt8yU%M-+U#<W9?(=QSM%6Iq;zYP0`~mclCDKpagM
z06?HdFBqq-Zojua?0W#K_pk*aBovF{%)6iOdy8K|u5ey;6x>YD3g!y8dsh+w(26E6
zeB|=u3a>hS(T|#I{J&K>({05^#Hu}bt!n0$y}j`dKCx@?y6f-y#3lexPW$Ym&LhVE
zbG^dsy*id!WK=n>L&g}4WQ`WiFj!6XOZyK1E*CE_&c#vKQwi3H_J)WDJZ?VvY!x0D
z5h0b=pnEtao!R$CIK%XabaCv@BJAx?eMCxXa4J+K$=6!zA5az8wWQfQqb8Mz;HIZ~
z)$ZZ!yN45nj}~;{snBV*w>Ns#?w)8XLk4qAT>M+oe+2d<TT(a&;VI?7p~CRb^&!vx
z2{sXjQn}^8*efj|VB^`6*<ju+))J6F?)lbv_@0Yb-@T%}_A_1=kC{o04*Orl%NL49
zxO*^p{&RbWZv6A3-B;cDTz}nF-+N3ljx%EakCh`}yF6O}01YIvix^|IQ;4*5hG;J^
z9gP{4t+jsixs<Zzj+K{YNI=HIme3w?C^Q?KPo+~OA~?nhm+dN6Q{Z(XDI}b}EOM|e
z<o&zfUD0g0WcU$MdCfhT%)RQL&mLM#RUDTi$7E*bNV^nJRHXzAq>y}ROL+Lw)0Td&
zrOy8cRbfVz`J`j6q9N3S!|BVmcaPlu?SH)Zq3^wQ{gRJg_q5;^HR=Q&#~Lu2O~1hP
zqj1y1WCRZkWmc;SuOF{k(e`X1ujfE(U08SS;~d&w0RZ4rRpiKkXl#z6{XG#t#P4kc
z02dK~W6tUk=}r~rJMPTtDtc?#D}wvC4X^1LN}gm$$(*@W&h>XG9O}cq?_P9#dsK><
z8*Czp=x_h(15L<_qv^sIk~z&&O)9cQqdAeufAHa*37TjBrmS|A9m)+Nk{4}giFGuF
zzwC9Z<7%NH$hWH}`d`l-$~Pi`K&txLv(gn(n9a8&5=fGg06@NK<D%cLXs`LK*QE?k
zdPf8>B*93!u<C{VgTHz4!0->QIIZJU02G`tI#T<8th9lV_IMJma7bnftpE_7R4N!2
z+$x{(9FD1gr80+rP|*jlyf-uvKx4$WLIC(?Kobyg*7l7#j4_7&>M<c7HoH<b7u9Df
zypR$uIC061ez&^0l8JRj|M7Hw<Drr41pwfd(sW#wA+V1G=Hgui0Fck4_Fj7G($BQj
z27c#K*cjY84pWRHsl2~8n*I92FLqzMdO_q30E+kV-SJAf{LkwfM&=$?#K|ZXvr+DB
z<;>9p-%W|NK!o9BZUq9?O+oX}>){;CT&Fq9RpJV_i<Kq}VjT<?G}ZW4F#tbQ6y8m^
z4+JnlM2R-wt4R2tXbnc!Xmq@zYg_m5TQoziqUw`FfrudDb3LFHwOf2{MW4xaZVdYx
zSGNaO?eB|ydN`S@9oL}?sdf)0Urt2p?SOW^*ke*vluE@+Y)J$`pG*4O%ENUb&z)86
zCK3oDNM?)6AA8}z8=vmVRILI=nt7e$3I{iu&&pI@D)tR}mHgG`ulUW<)|xK{Jj!r&
zT^SoJW(&H~8%<xet0(%~@7=QXbG!R<?f=|7Q1TbkAu^O>VRs#l7ytkUW2q*?5R<kV
za4GD?rOkl?kpwB`oUh^lAPBL_Xu&E~Nr0ZBko-v1fDo{tiT0=VrQCPEFgRlz$bH+Q
z7;tlx1myQ{g(HGXQQQFfF>f6TnlZGbCH(VbzPP@iiMlb$WX{}$csl?7KR>kZpI4o|
z<d77SNoMBXl=@H3Vj{LPHNg-9?mCd^J8whBr|*5PdoY$Remqmu6^E>hZNzI`njx`2
zmN_qxDIEWU+jd;Iye;z0{X^RJKi<DpRyyPVn8`C}0{vW3=YM^)+x6tW1mp?=Y62?r
zC=8Q%L*Dtq;EWujF~%UC&ika4-ts+|TEe9v1aJu;t*|hg*J}vKGX)yXaF0eXN0ird
z*ukuY{F>`RiA*t`N$0h|L_S0kEPA0Q?LTfoFku@%3q;cRri-TPl{k#D{S&VknPa{X
zfgax-#TWMtzH~5_eSM**bEq-_ga{xK@FW*Bg|B_vOP2P$QwRjBFX)j1FewQZG=|@O
z&vOS~+aJvwU)8BQb~mpXi=W!r^Y)H<&&N9I+{R2i!++{EnH{_0w2uCt-kJMT*zdkD
znJqFb>kVf9y`YOgG+lUgK^M1g8z^pfU=N(DV|e>Lgdva-<=%DSitB!I`?i7JSoTMm
zg3%-;RfiA`*^vY=lFF^f6tvI%?a|$vRxGId$^9=5+yWPVTjz|SRX8i{U~ICrQ;wm1
zE_FZ86?C5@;*(hn&M?*-b{S5)GZ&5zM+f^sV{IUtDH;XH6+L5&rlT^|kn)5_wv99Z
zK>SD?ArbIxT|;lpYsLzOI4jf#GH3t-Bj9z_MbncdQy=x<fmXG6V0&NYmVspceO&{|
zGo>X-_NX<moX=1qzr3?Cd?|pu*5PcVm47Pc&;YZS9@hYzvBAM;=F45t>~Mc9^ND!6
z&?N0=rgA25bZUrc1TqD^wreo)rr~(zjCi{EgEbv>e^F|`I7ErhgB1UHDMh*=p`go`
zMbr6va=H;E0_csFp|Tl~Qgbw4zIh~>3p2#B!R?dX!9E!fR^=Z6h~ap;-Y}#xi7g|a
zTlv?rW`7odz#3XVcZS!Rl^XE5^PI6sBR(N$L4PdoJ#ImeDn0`ZA*%M89PgCIQNKxG
z_Rh2W!cJz}tlUMwvp#vN{P?R+IysRyzLm;rt(CN&IYA1+<Kjld=lWeoL-?OW1j6*V
zHrptVRBm?mj4aj->1gj}fnG<Bmn~A7A=FGkf6c#k58b{e=(+p6wGHz~?=UZ0&WMOk
zT+(oNGF$w6wxGXe^i&^9K7fcIlh;o8*Y<<wUH$Uob_0N6Xhunn2p#$5b8tyS&H!@n
zd)b=5|LBe#gZZNI<KB_<Dnm#G?LmNmC05=^G9e)3iw0_X?j@OA@wf#Iq4RG3(rbSV
z0NXfY;&AN$Q3rsA5LQX!sCgalD6dK+YQHI~FbuP8FiQ?6?04rl!z=)JCix|wexwYb
zbtQ9J!;AX|FPUA=#gxH+g8?RJXa;49dVWED;4=fU>^-@n?kaaWrXmTFd|!Y3Blm0x
z+<xBbx-J{Y6M&rKL>vhKOb7`Voz$LxXjlA)3!7_R{MU}2kEe3lON+WNJsV)vD9|=)
zx*;)XXiJ{mJMhi=koOfg-hK6VauS~EeAnM%gFk-_pMn$p_aOlhAyd#pyADM+Sw(U%
zTbUnZGCETs_N(a)!%;>I3`0(-_=_=2)dgMB9>A+`K9`z@06^B{0UE<tRWW8lnlfkz
z!7X{7xSG}aypnIT?kkBxse;6S3e29>3T8{NaRYg)2O1I)eD|iU;-jxR>6EU)<fr?i
zndQ}UEde+~stdUP{*v{rpF3emeO?M_HxDT#>Q9rl06?=sEmzP>t({cPq}Q!ruBf9h
z$f3LDVpuc4TMrH<FKVj`J`Mnxc}RD7S7}71lz8g$`oSBXJa|Jor)^5+icQWjoA#f1
zO{kLatxs(2yX*Gn2Y170Ck5`n4RfJ)1OP$`35+v*|HZ2weRy}`if4Bn`fz_V^QOEe
zLbh>3Cv8l2&af#zk_a+Iqi$Dk{9Vy>;j*D*_G4$QU-<kN|McXJ-+%Bd1*S0e?pVkd
zLV!~BE&>1~vKkpePHJz%Ue_t{tPB7k9ECng!gA#_rQ{i$0g&Oii;#q~_ys}|$YUS<
zp_TuG;eZER3ftCP<NIfM{6HlGK$0MyDYozJjlS>N2YaO@ZxR4d&b`bN0+1AvZCV}?
z7aiAl|E4u9Z*H#fUt1IK#H;-V;#f9?0HCM~S8p_P!QQ^a9e;Un@0b4h$@6L$XaA9v
zAeL;RoJIMU6Yy_S&|&GAISRn0$SRCZf1Ue&rgl#;>A3d<Pi;sF=8Br)u!D919*0R5
zu3`{^vac_GK|#|!c4?c{=W^0KrVsso3vq>`-|M<$SI@|G2l^A2R<l<`1Pl=x!rt4?
zKcVyUCoHK?SS@*MPWQ;x8^_LeRv6P4G=y(m(^3Bwze^batb&BWZ1t4Wj5l@-#Mj;P
z;>bLro8Dyx4n**(Qy2Ur<aa;jY_?n}f<uYy3H$mJui}h(5UqgCsy{IJ-0Pkc5~Pr5
z7-H{7E?N8G)*AmOn<D;QrQU`m7lvcfQAQ08@u6g{wQDH-r^mMS{OV;Vv~S=_o1u}I
z)PWx}3FG$QYyg0Gwg~#vXiSOay>RAa1D!1&$Rui}V&vHj!znrQUsU=3j&%fRHV$XV
z0Dvw#r6XD&_I|TA<Q*aBphX%r*1C|czG&v+6)pY^iM$YmfWSb)+|CHDRA2xA>W$`x
z-uANf*DP<Zd2dtL`;^LAS!gjC9`71V!mgg^`|f<U`$v!LiXGP#GvHVE?x{LR{NMB%
zoso*OA^%4v_ehTJ{B*cgBUL(?&yg*wkVBj-bjcUA#-Ui+LpGf?3f{zoofyUzJ7Mog
z>WtsqyW_)|g5hSI(QhBvM~pLK9O3zc8F=PE3a)>m2Y!C{F8H6{Jq1Jrw>{g-t~jN2
z?T>EX@ts2>>1+FAnbl_M#MoPUT$~~S*Dp_A-u$7Hm(=c&RzF6&@|i0$J8gq8G#0ni
z+;rNiwrjq9%Z|@1Yz%#0;Vf$lg>!ik0b=QVLn^Po6#xR?y>)xZ7aohPj*elQV1sXM
z#e&+OdtE#_p6u#ELOfIWKrESRWur~o1ri8F%ncr@V73VW<aulA0RW7Tyz!SmUAw5^
z9c^{~-}pTW!C};iCo%=t+BLj!$D!D-e*e(pmkYtv@7=ns#FkdG|H`qlfz!ECdF|PV
zk7eN~%s91k4don<mSKp{(G-5JKbpCws2jyuV+9alX=~)veSL|qNFwjC$^fEC<4YkW
z2r1#wUGXh_(adiZjvp=Mf>bC_UNe?Gy`$%fHA{}$f+WqftI?^Ua*k!*>~3!!r@=LU
z+L8Ogbr1gax1YY?nU;EZ<L*Pz_Z2kJC`nFg37A<a1_e#L{(%>|m#<n-|JAV1eG5PP
zCx&?FefE(sn`@9UQycAd@dpsu=krCwAeg;2MYRFX8M=^PPvtZ=wu*F$dpmk^Ght?t
z-M`WTj8w%z9CMc;Bo;Np$-p?L?oLt&f)$@d{-I?4Q`g<UXZ3=H@V5!zz{9&@+NKq?
z+HniaNuaSz!Cfrs3TKG_c&hK*-jVFPdUy7BYKB}<)CAf`V$K~v1`zbQQ%w>7&%!?T
zwQNC;OCb?(_SP8zpuIk{Yh_2>XU|xE<$g{3Cp_Do{>?xl_x9mrepxkyN(69lDD|4=
z8sG0f@~yibE&I38abNS-6l|#!63$-T^w4l3f78~k;j1MPvl&-s>Omq?Xv`Glhj#X*
zKf0{hZ;+J%ibuyLlGEoB5q|Zi?T>C+)w+FGZ|bo_BgqfP(uLK+c`7juC`3fibOE{{
zPuS8m{9_?$eRGZfyT7=1cN)r%xl4ZTr~^QS^UBhH;mnhm0)si{ZD1;te)1z=nn4Q#
zU@UA3|M{bru6ghSKYwh--a|7a{QSRu4R8F;o8b>%z7v+d=6HBpjju}*eHH+YRRBoV
zm^gFf4iVe5vNqcnP2ZI(8n2C}@(Zw>T>$`qf+p0WAz%5_*1lWz4;P+!{{;(ea71$@
z^L7wX8N7V&4J+a6S1tvw%e8MXOJBSFNqVq9p8MfQGPhhaq}xsnE-NW?R1Mp?A;XE>
zNm2~|rm4pN>AQaMnx8Vxq!gvmivLz_g^0kVaHtD-`ybsg@JK;7;GQpj8{YNfFMvy#
z`JnxTWi7a)XJjaqD>jc}8mt;jRAnQkuiG<HkZ^cl&M`{c8kaaJ=|F_sWZmrm#+ko2
zVNh9Z!HT-{cMT@5>W`)`YK#OjyAQ>l?~i7lAR<WwN~WNno6Tz-nh@a!U+nTBV9=1#
zPQSB2r5O+~L~y&5jL*%VKWSy#H=DxV`%YQ%p{y{j0|X#8TSNf?1bu3zKIr=7h3jf}
z3voLG0=Bf?HN3wi@}*=>yRN8<sy8oaMosrn>Z|*2c+dHZ8$MDj`L)r|VZKrHI82&l
zL%AV@00LIySBol#AB24BrHQQ8RI1)DPoLuZ`(iI&*cklj%GTgcVV*W}Ik!2}Rf7;I
zB>eYfE0HsnIAdkQ?{9hP;N5!%Qr{U&<j%<#^t$ns(=0f`l8)sHdh7P?k^gC}3#_S&
zc;Cl3iwP;wo=<-)F@Rt@a2rT*26I~mcVxRbdEXtZHtywcWd)r~j|KqhkKTRejP+<S
z3;{oR_Zf%)@Y@fbjf^w+pWi)o>@!1Ds<;Fo5JDHMF6_TI<V*cIlh;06)TIlnurAS5
zZq1%U(bpZfq-7`PEF-0yb5vBR{0cV>n#wsdgdonqc<W1+J^WNx`r>DI_P%Q%mVH|y
zTU;Qd6Vr|_deL+~2nb&%AbECE(@$JAh-6ZU|30xjIsiNZ-SU~2f{>EK=Ps5=K(oP?
zUGPjFqBAX&a7w7o;gCv97}}iExLB%GB+3Ae+EfSzW;zX<6u=UIsw(M_-(%SNXDyLk
zNqN}d9s$7+5)C11yM~grIFwuf2(NHdV6#%3L(cfc<P=E~D4apSs}>c8k1c7fy>(Gb
z&2P?HQIiG$DU9pPB9~cYRtW%VL;f2#ENgoLfTNjB14ezsk1iL#uerwmMAu;Q%xWN5
z5<!0~v+UX3Ll+a#KdkCxdyLrp8Svzn0P`dWiHKc%%95^sJht~&*+TSxG(%JkJTyag
z?m85I-v#U6|3M-D$}9$O3bp>TWBV&YK@?H~M--Qy(s5|dVD>G~>>0UyTi4JvIn6jh
z)5W-r+MQRnsJr`Onb#1J@%sn&f8<T)EFKn8N~ew<D>;C+@k0QBx_}2<oc&wS6dWfo
z@fGu3r(a}V#{Lq)tT}O%SZe72|2`OebczfB$P3ps&?zh1|KRtiU2Gf)7m_fU)85iM
zoLFvs%A5`gpO7dqbIF(#lBAI6@wj^5d(q0ztzOvh-u8yjT`ET!+oqVR1{u**-a8b}
ze)GC#cfB7Gi7WiyWM8SA<a8K1t7w5{>!{Qgf0!I&k73hFmE~Bfq@T381eM?EsM%yW
z#o!>|bq_Svg>n@#%dBR)hh^33L%wKj(DS{{rtmFo^}%S+s{#Qa5ikN^l7N|68c<f0
zMnu3ELd55S&Zh7SO*OtBbT)=Q`07n--^dt#@2nM(q>z%VzG!Ig&#Wme0Q8Nd9%YP0
ziOgJTu5b`i;;T>V90+*Sp9a0I31o+w_VmTC2J^JVw!ibt-$bZyG?$GbnWJPSk!%XE
zOl?s8TW!#@ZH)3-B2?6c(jQH~>^Bd7?R<u1&7O|-WmnXblq7_ZxTL|WNuoclSy=z}
zRUP$@)doFd0EayZ2P#?#X7f4@#ImoCX7b;?{)w)-r}ibw-cmV6HnFVPjo+)XOwj;1
zM*EpLk4?r+oE}5f&VX!@O_$Dm4wq%QrBT)G9=imfy|5M&A-PK-0D#B$<+m(uiTvV)
z{li~0g!GQt6GVuo3-!6W;Cr6hm-xuZi^BzrcBW%Nt}IrZApx#wQ~>~Fe=K|3#?@`x
zp4=|qn#^jSN@R<!vIvkIc~*{hjb#d6K>Yk)9y>TpM1K`hf-nUA8#juc)tq0lz^Bd5
zP1TMO0Fw;J0;q?p@Qeru=f0Z(tFXT~rS%!rKRW=M{BXP?YXKP=%iC*z`TDaKe!e#;
z8y?$!Xvu;8SWs1z(1M2Gahf65NTLoPFiOdT2%s{)UrPC6Uo^F)#_!RNU)I`v?wYz?
zKe=OPaU_+8*Pqb|R{k#nfVB2xmfIo`6b)%?Sl|Og2nM4wU{Ml@fY)`e->Y7qD;lpg
zeTJ3HVKJb(At0UCPW$Zx`(AnDlih#1>A%nUce2Arm4kuNoMwguQcB5RbNZrf-@SS3
zO&*o4Dj0%cl@yL_(OA+S&0O9SEj-#5@f44dwPuc%cx1JGrK?WwK*re%yN9%QJ@R7L
zNve|lf6-LF&CU%R9o1NF{}BN)MGg96^!k9uJ<OZfrvRWR3?aF~j|m2lS&Aa<b_8sS
zRu_-%h9LkzG@Y$AgpAPmq@A^^FcS8;3e!h^)`psM+h-i8SmlqLnUNF%05fz+$GG_+
zi+w;VS}r}Uq7On#jsH*eK~MjfJ)u$0)jg27Y#^3if{2o^e9ovL1kl{K!m%B)o!baV
z=8?b$E?&XEe#_3T^EY&SZN-ARx7LTfFO~rY90h`v3{xzfuie`h`{+GehBo}<u6<=c
zVHl$Hb!=h#5>_w!e%68OXy>5AWp16!IRXIqye`2Qn^2Ds0CYhMcMlhp+3RjN;oI81
z|5PGklZiY6>I!3}9<28J;zVy*vRGxzt_=EI`v5?1kMN#v-MZt!ef_Chzxeu>|M4B?
zFZ=unOPb%hxTWUeC2cjYSkhYi@|6qfE?M4I`-+XL+ur%!i&lK?FOTf~dorip8qXA7
z?2Z+S@4I*<Y+4y+0HD~_*hs*c1m>vFo`n@0Ei|YC+9VO7kaB2IbIsp_K3CMq8QAln
ztx;d~y(8%hK5@mmkd%_FWA*uKa@SP~L<oTqfJ|wiAHTTv=U%t^*mzAkG($i^6K}kK
zOaIwl`TJI?491T}6}Dr~?Et-$3o3<V%Nt!g-*Lg`ez&+Ka_OSx$Q_(v5piss;NZ#u
z0U=i~U~gaS&D{gZE15Y%7)#*msG4~b_z@uhlXi1uZ$?#^n6&f0q=vl%S?(B0KIa}P
znJ&0AU2x_YJYMZb8N*#CEv-o&sWUVs0@*Ao699k#0HR<BI>twp+YZ;{Mje{|kyjiC
zr!A`;>THU9%cCmg`r&A(T+odA?&0hwz+}Va?(zl4qh-)6Lm<GM#b;LUBFLLRaw^-f
zFj9QurbX9Zczowo4I$6Xs=~$u3YA<A5x{UFd*bFj{a3fv`hthH9|9s1>>?hucbzom
z`FFd&17t`5fcl8PPhqS$8KNrc#?nJWY4vElsr}3&scfAtL<lEI0#p>X<+#OlX&7y!
zTPj?c5M*H$TW03DkuU^2@udskr|&%zi2B%;L%G~YDnFdbYlkv9tv8)3_V&keJtL{Y
zp}}}Iy5&%|@cj2)17CRk32@oTEwHf8V<UD+INDW_5R3}jHalhr%VfYgA%y?{-*WC!
zICEvwEnXLYKpm%WNl1c%F5bB9VEk-FRqTAw|Ge4WGw%bO>m&ex0$jqm>)U!dn!*ow
zTuQz=)rJsD<-@Ub;bj{Z)%Y1_rGOZ^kD5&s&Y+C26i_lS`^LgIZv0<(!&!@-zu?5q
ztCzGyzUx+ba%|)4vDhJ7&}&}UKls7BUl>@)xY?ENs0){5fRdScPj-!BFyJ0RL~SxF
zT-3!$dj_-KqvWim)7Mhfs^!r$`9jecR5<2mz%LOH%@*37vtEib_`CwD_iw|n-z~f@
zep^Gt`|y}lEPI-UB-S&OJom4Uc3sLDgJq-ns0;x3QF^Cp?_+;oYI(D9dEL&7PhRxF
zhOp=NRjXlXjDx{UUVmFEn?J2{{r7G0Wi<NE$07lEbncDQ-a8XeS!*9~yM_TEhmN45
zDuG1i1pqXctb_9?J(Vt?VMv^egd{1WO#xN_C^c=CmGj(>1ZN9La&@Z*E;*?Mt~|X1
zUVC~6TzPs2y#CAuaOK8!xNv<VtX~j-fL8@XR;`bpsIX<T+tU;$URMB+hzLy)S7BvG
z!=F{os&;)g`vMk}M5ge@zdW+PhMP>yd>-KBewe_@xpXe4Fa&N|)B1zDkoWoVP8A6N
z;_1S>b;2{s0{_mVYZEKpr7Xw&n#)##kP=rldj~)8^3`8n&=B~n&#ffO4N8@FC4!<M
zH*7l?z1sF|$5M2`08Hw&F6iA$1j7?f7P890ye>ZH@1Sr#3hZ=)1j*`;N?1CFABbo6
z0YLs()grF&vah4rTIX<|*>22T=?J8d>{S~(4%Gy`*M<Y0qNECIYk@*t$d<kRiPsi%
z5dgCrww%=?qY31y+L!YkoyToLN_O&+x}Ftnk!$LL?%%kTiN=SSM*;C{!I#JsKlZ1G
zKTr$67??RH43EOu<A2!Ybk-aYIFW?5y1+gFiaW@@ssW;q6d;ny5y8)CTvfp)jsHC&
z!f+zD8vyb`NM@y8Fv24Xmbk(}nLwAF`f<2A<g{yZ?Yy1O>~$${^5V#CEwzEC8QRvR
zylo^RKr~%=HB*%Hg_LZr``i9IU6nqonvDGK4LhQpjiK);9CKw+n&WfgnS!fpAaTtf
zAKdG41ZX8zdoI&}DxZHG88+EK0xpEifAHmNf7BfDej?yevJ=NC2)Uw;iA>>&KRmSm
zOnAdTkDXC);7Ai*))L+$q#T*Jfht0ws4wV`rj-eT$=S8|L=a6EF<%tb!GiVt9+gAT
zry8z#7dzK-MF1e1YW=r+RemGqmB;kve&~jPf^J-R{o`G)<eU*=={T8=ieOEVX#xOY
z2*Iy7ZBcAhd*l;!A<rLFK2GX{jT{M~KbG0#ar4zemMRU!Jkv0v5xdQZVPnL#yP%7a
z>TJ>I(P1)^Z|NRP91U;J?9_>9IuFIslmLg|Q;Aq`DGYSOh=G}!Kowu}25yq;wilAc
z);j;UJSs0Z1;VMa*@A{${qc|8zPaB&zhXz`1KxU82Sog?d)n(mo3R3LIFKOr$1<nZ
zg#71N7@Q*=)9rWclLldn&wpLA_E*idzK;dGYJsXTfsX8;R8Ctol+2wqeC<osV*vn4
z%Rg~2X-Vc*@qlKC*ksluThRQ`WR6uct#c({M4-qCTBi~bc-^XKjs%Q(eVjuLv2t&u
zkQmPt6HD7_Zw+|VSb1Wc`DG1ej||4LuXu1vUp;4xth`i$V~81Zur<znM<JzzS8iN9
zeBp@;KUEWO-@{l1n4{6kRzQHFA=G%L__l352@cjcJ;zY=`RfS4lm)1)Vef>kVb`$2
zAkT2TKo;yEM9md6Yjfw+mPsVwIZ2Qz=u$W2_>n9C_}r?|ToVw^N;D*3gywtC7cESD
zy?jFrG=_cK7B)q8S2|~>R178aD>Oq~zNIHSzi88S0R-}*b@iaSd<m7~=Yw7~L)EA!
z0MHB>e(}K2hjp_Jk72cTD`Apj#r`{nX4pPJ3MolNCa>DG><{gA!SA_Mo+lzeh83>P
zN_#Q{NP<^Bx?^zRG4WzZCYmJ#$TN_QK}9r0d{UX@bXU^_7>1~l@;+zC)zebzUqsHz
zwknrfReBb+)E=DI$+;FQvo*k*&gjHTPHMj?>~r6)a5O4?8xet&0MS(bHG2j!=Kug>
zoY~b^$c~-!aOY%zQeKml7>Fk>s_8p*)q=0p1ib^5zvX-dBI0}dVyDK^SryIV7iJlh
z`KGf+7<5=hxvYA#xh7EHRT@&3##0eMI;TOlpwD^h4@U*PyT<Qnq-xUU7*p)?xcZkZ
zXt3prTY3h^IAe0mTy3=;JZVYouuFw&!ai4mD%<0foV^U~I2e6LP0(92ziQLcMU>36
zYv-*B<Hog}FVuzoe^<td#0()Zo++&R(*t{7{;#eKklEL=1h()P=@AQ~MK7h46C~6A
zH-<dfvp019thpxen8JA(0cgI&XrahZBC~$$fq}-O4gj6!-AEvP)<EW#x?mS)I3a76
z%xN&3Dy7xQxvB&fh5$w35Y17*?NU1?tyUId$I3QG#5aG<3HpQ4Dq-e#^TN7CO+Rp}
zd}P#}jrRLM1jrT*Ol6Dj|KkIDBG&Io0FZK;Ng&QP0yETB0BxPIzZ|)WD20T+So)E+
z`r!A+09~hXB!HqO!;0!%59T0rva0dtD-LHQE-b4~r<VHAHbm5`?Ss9m1w*3W>vF-|
zXB;4yy~cv|Vc+t}h8;_y-OE}-2W|ZzM;}0>`80y5thdk2<f@LQ=R!XBpA?Q`0XF~`
zk|0yi*WUH?-dFdhspNNrF^{tk!~Sys;F1Uw&PHm3?(2eH*ZyiEj)+jujFyAL>9-+b
zyD96X;tN4DpqKXk9qB$+=D6A)FQkOEt-kKoTK{)FDnD42I<T&MQScQD8m~EK7(hqP
z6#@XZ)de3QfXrlTamvcp<wxtepWh#?F@#(-@n$hF4AIqD?;4ovo<j3M;An-g&;=yY
z+06@^!gnxcj`^xa?Dfa9rz>vvr8)p`F*DA12KRkD2V05;Lq4}`t@Zu2r8cl<q7UMn
z)ej_c7cj<%%+`S%=9|tQvCKhvjyVAUHH193NJ*&)_k*EpS$qA7k}A2N*$1o&=kWNB
z;d&uy^+Z3&5Fwi@ZeQByIb{DX0|3S9S5_I&&tK!~ur5S=VN2x3u;0^N6_5ZxzG&1A
zByyKza$2)Nqg^}Zf#4MFKY>9eXGUvV-R!OBF8x=?=X%hkGO8Zj0OSh#*^h7OI%~(U
zL5SroeU23PIe_Vr+Ltko&iZUdB=HSrFTQ<2W9R{evvGw45inmgt~lxd&=A6I|3cPG
zS0X7fmdf2XnbAZ9g)z1~p4IKP9dk<4_6{W@MNMBp6YQ2pRb+E*AP@6N_D91R+lqn6
zkX^NDvGjY?@6-m|Bc*??nD(IQLfO_e^ntry7>Za&2-AN6B2@U9@ZO(33d-^?f^`fy
z?a!Q`3bMo{U4CK<)P;kGJTCryx5~$@V3SgM2BPUT_U~HlqvsjGBe)!UQvySX2t7mb
zZK{Grhg?*Hcxq9X$0c$GpUVj?_PvH<xv*}C&hcd#z~@nAb8R36W&ud0JkAFXhan?p
zEw6!-SG7OtR`4#D!m9cPAqg`1;stj-v-f;HhjVKG!+d}Zf@sH(C>+0}`L}+zdT=~}
z%@lN>LGZqV1JRCqwvPaqDKgW-6+{HLKGO>y`qks`fnPojj5Dw>0183%z7C(R?2H#+
z+4+E00Ld5B`@iK;Sz;o69*Sp<sVZP*O+FwZ03?}ftMkX$giDXE;bi`z?%}LD*Im<z
z2o4P=St?r`zmkRk7{kFOZ8eWbQED4Ff6bv|vz5cn&3e^o9ebN<0zXzAisT$HPXL%J
z8Y@QPnRju;cSN9F;T(T<=dV@;z3R)p{FO^ydiyh7K8^)B3ruj?1d7v^gyf>8$mY6`
zckg&Zs-+~zWDEBEV>@`7rxj+v!H;R>z|=x(U2qVs)*o0UqX-ezXexKy!`lbxXq*%b
zM3df3t{ACmqAdO3j18>ptbe*1b3T`CR{m2!;GziAj$hXH7mthgj#+RkFiq!mPeBu}
zyZ^=BM#dRAngZiI3pla;=L)wAsu%#!szwgyu5Wy7K||<vg)<Gu@!12>^u|L&sf#2s
zj))S#Y?C#e=WwaK&F4}ss`0yC{=HjYY_}Ra&&I|vLIBB3&>Qrs2iqG$+p3M0v`K+n
zK|6LLek+NX>=!w0b*s_d5Wa=L>f}z<ac?4%TePb;W*>pkTnQHd05S}v^ZF=-RB?C4
z0ECbSPFmXV0GK&4^Vb|*4&zuV5|ToK-=h{h3f>&{ySJH<Mj3_z002!F{sRN?*Wd9%
z_wh&a4EJvxT97S@AL&BgmM$1K3FbPth`!lis!QqmgusaiFFC1gOHng!uHFC@K}ZSF
zbaAwufIV$w{+jV?(}sn`rkcR-0jf7}w+6Z6U@YC~;gZjFph!fJ&TDG6sN*=|0t^v|
z<o>lCwU5s4=#1DVltSXg>uVlt4ErAB48`aHUpd?uPUbG|i51QS0A`&cNSG%Rrg}*$
z#~Bb2H~=ke4Szl4bH%F$ItXA0;?Z>BBR(eE&@2U!({bHfdy=l9MDCNikhkYF@lVc_
zw{ymnN`hjh#0mhA3qUC(J9lk!)~)idG8~%=ZYy~zcWeMa(pY)U3;<wzjax4kwL9Dj
zH)tI0OcKFBEX#Hrj5{6uoLs|2a<w;vH<sPAeAEOeoIz`SaIoIbQvl<`*!gSLE@sWr
zwdYKekP_Z{-tukrA@B8mkBg{ERs{h-wxF-s-ZgaD?!FY8ZVmA6L$UU3UO!7pf`TSp
z>70Jaj-J$P=$y18KO`VB`w|trZe_^lR;nHYO9_zIg>4Mx2hY#q)e&JLmdZZjbt&UU
z{z4K&(*@R6>-Wqx_>)BLmb%c|>MdZ*OH~-uhkc2xCZh8@Ix~Uke{qn2DqGZ2`vbSi
z<2a57%N7kBPUNon!-IQ+)=$h+t%u<QD%M^iQGa`V@LGkVs1`;O01PK`OLq4pK6dAG
z1JlM7TMk5(!C2-ch9uN<0TS?nhj)%N&vc%w<MX6_(ggrwD3<LBc+@iU@APFwL&7mp
z1?)B}e^KY74Q|04?b6>=;~yyV%w@(4Ku8F_?3Bf;VVnfXnNr$Hy8Zdx!@j&`oQva4
znl}G;VtHH5Z)Wq^{?o4`temx5xm1o;Gfr4s|0kEio*2*Q=8FbT=k)hJzH@j}z@tv9
zGh42pF+&K|1X;v;h7u?2JTT0F04v{Vx>dqfB#BIdf&_{!YlsMoTWd$WZZ$p5(v%Ia
zC!7a>vx%~nw$==`)CQib+AkgvhyWO3O~zpB=Xf93qdTJkA>k4jcU%?$BBqzL)ZVtF
z-Zj6UM-E630N}?06hNYYM}2WobIq-Gia?d!$ANhE_<-Mg5tNULoWG`^ebyHT0FXk$
zx$By>&Zf{GT55wk#w&W)g;ctSk}vnV`P$oG9GRB??vG`tpy@8Vn`|<jKk@lJgSAdX
zHUpg@OaUP)*PTY&d9CVfEYyX3-J^}%X$%=~tQ`4B03bz{vv@B%p?%2fQhv!C{irHm
zK;hWfHI#nEY#B|%qU(`NAZQ2yL<IlZc1Rf*Nv|32rU(EbuPZtjPd`K?aUKC4ra%Lb
zi!+=7<rhw{WcN(kxUykyTV3!$kID;GvLpZzA(}2U#xwcX_`IrbjKQ74W~K`o6m>x+
z<3<oo<^79WBR*!%_$*U?ET0H9o%eJ)1q1+Stqbi(#6es6%hs16*i#f8Lqg`SS#tUt
zS2PU(z~4%J$SU^(0T^S=d;8)S%|<2d1dc{--ZS8erg9s`M=S_#l@BK}`9ITqtLGW<
zb=IE-0Lb&!HRwW!9|nDDV!U6{4e8n5HT<DH!+OxRKK}=DE{TYa2p~yQ)9kTs%mRTJ
z@Ok%nT<nKlHy2dRngT#xHx@p<yZ`g;wca!R9<}~({dXc;fJ|O5+4n@Y;K~*CKsBal
zhAO+6!{6bm&bqY6t!^8=Q90Y8a(2w+{18}e38awlotwAi0qCi)&z-UzSY?umE~IyV
zUu=D}fV0(t00Ds6O(pH`7q`{6#j{0qg5Q!IP2qp8=%_Cc9Nodv|E|g(aC1bf+3ex|
za;0&s_?`(N@s#EDUyt}b&yVfVi=~KZAf9>0fq~RJ3z|@!2+ghXNuOyXRfKHOaQJ(y
z%^P%v#Tu97$<7IaBRkrWP{DLIMSR0jQUU-g8$pA@88k+`w$jbZ=WKd>A{p{BJ>+$5
zsR?*u<S0FN0&~p}u6@1Hh5LuJ*|K9*_c-i#*F{q~W!zsxl#5#;54Sf23iEO@^Vgj_
z3v~Lrh5PG6{y(yDc1aLnGFM#w^o~QXIhc_Dx!}){myI)Em#kkFP(G_0PhT3AOIm6l
z33=U*jbr{n(*;ZB^vn15#vk%~)D4H*4mOd~A*&fBu0R)rZ4Ke`_eF~?2e8G-^z$$-
z;UE)eMYAUlK)of~UqPVN2HeL60JJN1Dy!$dbk%We{nD1_T`IqAe1=X!BCCZEp>{T+
zW~W#X0z(2EP8Ah}^VczkJ^-ppt@>T6Ar0}zjZ14aqH<<t{+c)jAR;VkMpH6ZWI)p|
z?x{8e-m1a^0Dzi+FT1EE@-xm@-pRx`@0KkXu4GpGXgaTXOTi%mc-@C*fvt8@^?Q5c
z7YRwucd~&W!ZeR{Du<$%ij?gChp9UN0Du5VL_t&vu*&KqL|};EajBwGZs#itXD6x!
z061}F`({;TcQa?&umhOuhJ=DH8xXN+ww}eAY9OV6=MQFBz~{L{NN|t$J3hCfc~t(h
z^VT*MXQD6K|K+PvPZ1ehyvhrVVKU%VZ>R})d&d9`3lkMIQMZ2}@oL7fAtU}nw*k?v
zfJT~uTt`LWpp5kcQVPisaYu8F|F?dRniwxNBbh1SlRJ7@JYB%UO$H|_Z}yG=ES}C^
z{@m_CAB=WiAf$0I#}UXCLbcM=(lihNaJ!UazUq&3Lkx#~uIB<Cm#CV|P6Te1uY7Uu
z@Tt=r*f3cLfq<n+2}lTR>%kc8>yK`dM5@iMR5FuIwf>vec0{^pw9H`s6nGLRNj5?Q
zfK;~7kVxkk0MINjyAlv^QKOe4e)XSi^}%gUrFFt|L@bNqY=D?6>SD6v33QZJAR=gn
zP`U<_t=$R90D!E*4ZHs(zzpa0lng+{sT~3UV@$6P`wx!VrXy}rI)6<RnoEi4ydJF$
zc%JrqRHM47B5=Eulb_kwf9@>UGF!!%!}-9Lz5V>)K;jYraM|Xm(#^Hh1s|^q`S-(o
zmH4b(8`{0F>G(y@s2p!rctxXPEXfe$ipCkYJ-Pd0Kl=|Q6D)UWwaOGQ!~IF^2_b~Q
zUb?;!PFcO+W|zXAVmMBp5dg>)4ba8RnCAXix~+RKp~7hEJNr-DAIva(83%MAKqTlr
z_5t8ID{A4a)h&N^sr+#^PD)s=s4sYSPu~eJJMA^>ZQ&Lm5Fx*BaV%YElu}mQ&=g#s
zThR<5{&L#NrV&ZR70yeM$ow@SSh2EM0SqarKb{jaebnJdmIWy!G(<vij_{d)N6nYg
z49e^gReW#81a3u{RMzaSoc0G8f`B1g)fY{ls&Ecm09xbM3_A~RCc$nhuenXfUhUar
zw4*6>uZ6g!J@<It$<La2^NmXaxM@wx^$KTCk1=IdkdVz6Yj$=IFPv#x<~Ut(2BU06
zXMJm5Jkuy0k&kqMPoG;Uk`zBVbxl*0=CIY)|HUhw%4gRtZu&>S<2qOt0MKYKn$ByX
zbV0v-b62csRtS%yO#_tX)dG@q1ne^l0HqBCFZlY?(GAO*f9-WC$w>qG>0sG%bN~fi
zdR8oGc&TPcL;_(|?Sf<_Uoc~OKa?tRBJkljp<qjGpzoM9_FUn>EN6K9;@aMYjiCo!
zoM*>w4nn>tm@Z($=B{M(OiN^GoQ#!2zz}Im*T`!Tu#KoPY%piaM*N-!Hm>b_R+tDt
z*u^9B*W_Ljt(g=E0AhBIIc&B|0)X7zo0J>Zb#AT;`~G25{8S-dW55#-qzjsy3^0`k
zET+~OLM+>MVDQAfBU<S-olbcs<z#Pfd>tYzuIjUiQk;yZB~htu03nPO>G^9$WX>+~
zS=}D)?Q9C)uPQuWCK{Hg6m<zMRXOvS{qeS0d=^9kj>zQRzT{PmvF0(YA}ryo33%>4
zWo7%nY7|LQN}O-@JbTw^i$k!gv+2H&PkmTbxNri8GN70)T>A8`;nV-qf;H#i3lRua
zG3_g#!E+H^9Zds506FlzGZ_-y+!XOWWi@@Cv8y)dvfKwb-!5=qFn-m6{usgN{A?Te
z&1iu2)V_GFF68M{#RGDVw6{0<?xP_<XtE(<>OW6}XRdAeVaV?~gyXdD)(r8|CwBC$
z?#r0@c+StCZmG4xOL+s*c8~%fcwiuYVo?`ftnlwd0Dg}e5BXhxcB#A@Mk(IIDr7hs
z-2RacsXo%*$I5T0N(eP`cBa%X>5ZEffU59ezf1XR$mfnt7FFA5$x)yzrxpR4KcC5K
zN-SMi?^1Y8$>@y|k8CFQFPhBO7()8X3l0bvBDFUJ4;+eSGZw*FHeq1i6;J(|qHU3c
zAQ^xpDRJY<<{#7qJbMwx3O^A*I;Wlg(3ak1+xv@Uwrsksf~q_V5`qK(@7~_%8M-0-
zSfPM~00FNn8TPwxbScUqLDL_JHGfV0T2hO&W>M46T?!kqb+$4H%@lN=E$Hw0>%)7)
zwtUP~;CMLpTgaKU&sdfl3!#Cy<^li`IP*uqNI-_!c1~V?%7WsemYQ$--HJYw@8eb#
zByxHU07FRVAIWqW%)6ju{&944<dL2L*|I=HaA+vuPUnlwSj8{ZhkgA4pX<<E1Hf3V
z-(!!x!HEEaFa#7xfZ#w<5{29>v8Xxx6P1r9>^oVSM5YiVg7XYR`YlJs90ENp=US;)
zkMEAtUmrd2GD)yeNI3=>s2m~Wb3J{*$;<wE%EF+NOq#N$OqM8!vulEx{oYeOUht@#
zIevH25k-?ZvC6y386JFIZDiGn9ey}-{lZ5=KG!`ih0|pF<NV7+3OlCt3ZiW^FMoPx
z?}ir-rX-G5nICENR|+(U2ns}SU?Abk<~4?ujs&6D(G>pOMJIGnHE=xFfyXgN*O@CZ
z)peMv#U^1IUpg?D9%*j~d`neWp_+atvqiT{RbD293|ODf9Qra7Hk*`nKa)2+*>j;L
z=tQG3zli`8hTwJcCr(}6aodJYAAr@Hb*4`<{vGm8g&{2cnM4p6&WJy42`Nu?G=}eE
zh!V%S=|DVl4tKdX8HPZo-JR~<7^1??(xf_qXRK<Pk@t1lN&D}TVCmleR4vSK?z|;$
zf)FH^qd1|#&!29o4ct1D{&gUplhsH8L9lr9p20WW{`{bm1*GW`C>1`nH2Er<*Fw=$
z-e1kecwOpEr>$%|z~<7d5>TS8J~-eooD&*PMFKKu{!DX?PcCk*`Ez?i_-Vwk!XpSE
zo-SN{-{$@8P;s1QovqWf9@=4W4Pwf^zQl>?oYsI=wUSd|T&mK)rlam#$xI<yf#)W5
z?l$P+i~&v*Xj3U^e@6GtDnvN!HBzkEL>8@HYGpEsAoEIqO%+UP(GXA&0su#qbQYMv
z(rq(&Ew*-1!_Pe~rROlB-wD@itnoHM*ia(Z&^3@;$vIc;D3FWkt-ozF9J8<)0N{?E
zY-df#d!{aA1&e@C8}MX{1?`EDcg*1m%~C6+iGnLLP?OTDF!W9Vde8brm2`wtFs(0S
zrBnjMDviCXfgCg^)8H#kY^4?LHMcKl2tSNfva|^RqN(g#UfAE;W;uw`y^drd3?Tpr
zfHOwV-1bKI?w;sbxq{w;V>6k-<5C9KEo!`WD48F&V=+r_kB|wZO1pE+gFidCm6I@D
zG&2>_v~Eeg9`vf;^}Chyct2Fog|fdl_K6p|(*ZlL*=c{L>!h(1*gC*$6*7PTsCy_w
zN4mDQpFl(;M{{~2@)k5hvG59;1r@hz4p9op&Ro-ytPgpAT^ICbs^x-D=jv$ivr3Ui
z1Sn{RZ&z>ZxQ39|3#0uBodsG)=*ofhCpA}7ZSZA=kgQCzi6G!n1(JMvU3(xohX9bk
zFk}u9(!vm6wG)`I!p8z`0Os68nGt+MW%R!Rj~XQ^cd|$TBk6pdl<<b_2L?+$)+_@f
zrfE2B@W$%iemz8!nNtAp%2LGa*eX?Fu&6n5_p<i7hfi4)EU`n*7B^D>o9K0MK8YQ%
zb#$hg8XKhI;RvFY=LpukMQN?|WWOt%?_6$X%>Tx-MabrLvw`+3IEhLo@bqP2c-iqS
z4>p86PjQBmhhbPfKd^!kz$)Al=!OJ9{L;bU)Iw$wsHG;%(*$5hXwJL@1a|A@0eW_4
z-|Bq9xUlj$83Hud2L7^YVe_zM@a;XvxdMOAy|-ir(m7*s$md>9<HY-RZs;7EtBQnB
zG&9r9(J^E@dqOicWHOnzY?%ZAl0vfOZM885^2<T5Iz&|t3gYQvO;MLuCbMbh@jN?H
z2M+-J=BImM*P-aig6JyE5M!R)rEpl(9JzJv!iGC8TwTK)_CSEyyqKoDsQ*r<$fe*E
z#|!AnFE#^}xm@9(N(rkw8+Wxegsva26+RK*a4NrYcVGOH2X`f;gB=#rWYrO@>ev!e
zasCsH1%th@EX|M|03tH+tF_otlH1VD-m9}<|4RjLNI+&f1i|r(Y9H{rm0zfqUpUg2
z>x<=O>240bKu8G)c;3!K(O3NUpSF=RxI2P2OMKMEqj>Sau&*zgc_Yy1|0k=^%<pkM
zyJ2bTGek6pNu&sn$`|_pV4vl%Xfhzf3=o!sKtg~ER|fGd7p>mh*%Z0f>sEAf7Fi+y
zN#fmuiFfp6`AW`MDY=qx`hXs-i21kgKO@u^O}{;w%5_vT1QEaMk@h<G=U2D-^ywNJ
z0KkG~a{B+Dz3+~Xt2ozwX69_))$XcyS?=A&fax7VNgzN-BRwSHl7#ef6H*B2A%WDJ
zMj(Y$2<6g4?*@Z$@3Lj}uI=rdnfLo+&siO<C3gt9-+6vNVIi$fnR(}(XWlx!3MDYG
zSHmi1&jK~ui5<wZE1H7Lv4glDBO+*=6>FeQ)G5-Ao8MT;&1T%M)waKQR{f8IUU%w9
z7@u4kV|(R+n7XxRI9@T5$ep>aE2)_q>*1XK1TQ9}D@Kr!fIE=L6;2&ZW;JFr8W9m9
z0dF@+dCTI~KuRj5O!D$?D^aA(O*0&DZyifzSF>r$q0(zd9qh(kHnshVO&qsre==_l
zt~5W%i2+r9hL4`VaCLKa#ZSC$EmN!|Y8w|KZC8Kn!_RHcEMm$e_b8R4I&=#UhZ2H^
z2FKHH9Z6&x%XJxnP!;k%-&*DUMx{^WQ5MERN1`k*@E?;)Y}S6AV6XcKjnNH~hntxz
z4GBeH$Fzr%9Q4O<z&kwMCG&=iZtWhuD&%uFbHSB)2tVY4!scPd3<I*oI8D;*6J+pk
z8a3?|nngB?Fk{H(_4?kCxF4@hSc*lqB(b7}T`EE>ozFE@1s@K0UENc;r4JSbV=B8-
z(HoP<<e~%PsSD3Ps-c=q7#z4d^h1OM08vz+ERmI`^pB>h)TC!;@VG=g;1Rbr)KqO^
zjJ<XQP~#kx0S}v3@0~aiv+3K}2u(m1Gj^<iWhj8_jHMbYL%*pCdY^P@lMm?xAd%Ip
zH|-w$Bmg%fC>(}a-dy-a^FR1$gX5WZPZjtY4f%Q(&Zzz7n~rJgIz(Z$S;6-Ez(hpy
zRZyYK%mOVHtu`%K`*m}g3&c(&rX7RwKvydP!cYoQsgiybf_tMC{srby^flT0Nf93z
zi><DU`fp-@oQ?z>D8fFE@_Ss9ih-7hvm#gTNb+NeR3>cJ>^!6xK-2ZEtnZoN)ISn$
zkEin=oM0bCrR{ZTs0w-i^3JmtY;#S)?qB0f^+~C5#khw5fQT@fyd~P6cz^4j+iiD-
zQc4p%w`4}`pX#H5=Uis-R{MQYS-o!4o}o{0jcdHP7WtuWhK7`21aQWgt{W$eCUPHC
zWm)z`^5-9!*H-hL^N()ri}*OVbhW)q&|zDO|FaB%6LY&C>8uQSqlbDhx4A<(0Fp}K
ztfOab@_Dp>Xwx(uQVPiF@~Ac2dft9SMwCp87||5*jI`IGNF9C>2qRYPTV6CWn&A}y
zGPweVV;RJ<>eZXiqHLEE0$0wh=I=Un=EIdC--8~PhUvPG9Vq5Eg?Yf^nrx*q2Tv*>
zmDMlZ)i-(+Gpjb!VKf3Y!vFw)Wew@u)jxg}fx`A@*91dl$h-Nn(-!@0PPJk<5>})G
z4P4X&M|Ie*hzJRB;LYc)_y%+2RD=|}=-qeksf~C)<<Z32^0_e#<r<8o-}dLHJKx6z
z$5{5j&zw^vConEJdt_Z--G+|AFD5d1_hcppg#2#VQXTwFdt>xz0O}BhrSxnbP3*HA
zBm?8Rq?X;I22=8gCP*ll*-_T>RQTPTv3)C+=IUU`ep)65L1n~#nxqe_2r>hJu~eQX
zvU<t$z!?j54JA*u7VB$4PKZdo^VB(o%AofTHDP~>=zzp#S_Hf<xHUe_y@1g~X5QM}
zqu=ODLRfM>-Fn(jrk^5KbU8n_eJnIEp8ei<DpzBtDvSdmzw7blYX42ij7pVhbt;jR
z*Xlr6BoU>!N4}p8OcEiLD`0JRdde{=PcAx?fNJY+Rk1ifoATg4(aP5vtNq3@mtvUI
z1c57h<{1jQ>~5^|e?Ag$Z!Gt^8B)1M;+e~S_2AYkW$~Aih#)jkdRO~<THh;O?lm^1
zF2loXdz-d&jeISZ$^}psOC9pMjk>7+7h%8mQR_frtxsYQ2NO$8L{QUNPA1Q^Fiavb
zm5kfp??`65?rVKblA~gB7R%{MW8AFsHO+dis_@NLKwzrp-<K?wS;MTbD3Q$@<5d;D
zH6fonKcy9l5i)r_Fq+D}|E|@&%fI{2bqCaOs(c&`DIxM^nX_UMYF}Gjc-G#NW+M}v
zxvqNSlF)QKp!~hd*QUL+8UV!7g+|7h-}?JZK_U|JHif)ggBj0%Aaide!c6yBO$duq
z2uu{?w?dbQNXhxl(Vx3DzMUOrMu<n#sKV!-Qe@0zufS+B_np5yy}N;PUh)7PMl}r%
z#@H>7Z2$OZB6pHhG-<WEHQwJ)>A$8WQp_+vY=v)o$o-BKXDnY632p&^3R#RNIl%M2
zvYoVxN@?>)@Wq=~ZCW|6{)<7cYiGHfCbN2Y_dx9G=eCc(xhBZH*5Wc`Q8|_i$y5u1
zi16691U<50D7d93{*zdyaANs;dfnWJ1Y9?@*M)wZ%;a-Qs>2~<^w63fa=TrgOkUsD
zc=A{(yL?0EkeDF762<|RVX;0l)grNK5?LpSMXlA<*hiQEV@$NvMUEvvGuw_ytH|;c
z&%vfM#VbI}X{wwf)x;S{D)Mypj~&g}S0POMgJJAl+UE#w2&AN9i`$|*gFe@fG{L7o
zmJzVOP9UaRZ`$BBU6Ok+mbq;8_JKEZ!NDgw0I-93FTpuD=L9g`JDPq+?`Z0Dr6xak
zB;f5hW?s|vm!8nRvoTnlAIlnRv*^4k$L)P=;sk)!^=9$Pu91i$YMXx3BBb)tQ(x)#
zPIDk&Q*5aF-}ue7W1q90!B{w#*K4P<e-n{5p2{@ZHf6P+O<yrKsZVpsf`8W1TD<GD
zrE8bZs`+Bj>)Jh;12UxZMtC%p|M24*M^5<iKeo^`9usCri5{~9U~Y=8zPGI&rQM^3
ze`{~z$D_&IiOjaU5fQvDQHX}zziNp1en?PR(*tPV93`fRrJhRXSaZ`@74+<L9vHM(
zI(OXTn}#_})XSK1e=%OgY;lVq6Pqkoi~vo7an7)3G(Xqxaa9pbacrL`uM{0k>pn}a
z_M`izvgVUbKGd=nKTcZIvPRSRrfG};=0P&MsB>Wal3Dcux2Z#veXZFK2=@20|G#pB
zo5_WwE4cg~SMziaP5|6)t)XWuf6jPra;+;vH(<v2Pxa^a{}<-rf_DrTuxBW?f{3E_
zj|6<4H^1DO3<z7Z_I(`^%CcbhtClS5cDYgFf)$OMW;a#-OB4LCgf5lS%>xvh*6K49
z%dF59eHeht>NW;|NpVnys-p52Ts^*d<k*o!?p@h}5wb>{irvyOi`r^`{(&>+K1)(r
zIT(js8im;hoU`d&hEmZ;TV2KdrtMENzxV-I2`mLWNy+ZG>7BUr__jxv&#L=e!0X;C
zPGTsQUc9|$^sgJclP756TH=!fwb>Mc!tGywetHj9&W-r@3?)9_JDPlBd1#);C8)Z>
zd-HiKTL1g(r47lG=12Fr(5E6^h9RM(gy0+xt{ud=E1DSr@5}1aH@&HcSod&zVRgt`
z%S;c=KHh1%^Eb1fHl|QO?CM`VuL=gehjZ@V7a1K-WKUVWW7LIW#}xYscCUL{%E-1*
zx!{~O>`CxMrtl612upf&sW6_-H$J}U6Z48CJ10uil$FF_^2nBoVasMjb2VZA6O|$F
zp6ML&W`4&$<Mx6+kKxjIdS8N(%;f9V?(F}}zt;60hskZaSQ(Mv)<$U96t@+s!vOL3
zXLet(c4z<RGkHB=-`p$wp2C9G>gyNJtba%qv%>x@^*`;+&n=Snn<x6Pw9$*LT_dN(
zlG!EGDoY{+ysnk2Hus)-^CLU#7870~d{j!3IG}g5>oG%0Yb;s8kk!ddK}==*W&jF?
zWP@WVzj?3e!Z9Wh3ffzz+17N5zv-CPhnLT+`Bc#7>X^hp01S?$mb|pB_e<|Pt?e=Z
zu6#jRmPdOqJ8vq+fikx$StdCEocf)YaPs_`&@-EQuI?F4o{`rlbQuBgx&#`l1Ao2n
z*co5G<oLGKS&OO}Fv*HvvJz`TOjmYd(!1{&&9HebRp)WW-1|5n@t&c?N!1nJ$gkI6
zy7k0u?|M-6R1Zcr0dTG+tv?QcDG>hr)Gm5?$Kdgdp>c9b=j3lv*}|z$ZRoFiYSTcu
zUOo_?IR{WGw9N?gug!6I-J0B&6Q<Z=QnBH9_VYPIoIlO@G25ZY9_u|3hiz67h<@lU
zdT!gOFPk@xrYX{zO>ZrvR5e?>M&2Cpz!l3A%E!%&B^2fRlPNQnr-^gMXhyZm-#3z&
zQPAb2w7J3Y^jUqw@v4_PMo{*=(;1=Ifv){B0~6T~!o9J)tO~hrsPMZ74o9dODueL4
zw8XvyR+o}*?i~Ku?;hRu7IE$`6c@$Z1v935y2JCN-wDC!ly5vmPiz{T_uQuLYeo~<
zC6f<#8fG>`Z=2Z|{WX9B+1A;iZUo#k<64ug03e>uR!F71Q<FKC%w{v26p@v&D7k_C
z?%MVS$5vJ?69UnI2LR-fIgO7jnOXad3cq`InMt0#(lO}y-&H$5{`}5QFXortKDogI
z?Nv&pv={{Ehy*>+zdpX>+MRvlA5LX;pQQ(>Nd8b&;rY!eiyFVMv^5$B5t(H?#DOew
z>%K4TN#n8~KZkez@DaFOV#N>dT7T{Kp0ST9BKN+0#i^`b_41B^PZfaZ-=6QHN4Lhn
zimum#QIM0yKyZ$yI?@o@8T97wJwi)oR$biDKmML#zvJl@_@2S|={bp6CK&}yVMQ{7
zlkJ<9bx-xNU$3I)xAwhn&tUuwMaw*uCQvF`w5e<Cv;ApA=YIZi_2TZdb)4EufgK?t
zXqL=V;@zjsTvZeC|H1Ed=~H@N4q)ZXYN|>Ge4ei92n`5O1dJtfCp@=#;D-;b?Oj>_
z;kyZZ(iqzN1XGzD!Ch=g=laL9otLcJJ^X`2MxSkWOGSg;Y-45c##~PSeRD`n+v(#-
z6d~P@+;bc0XWzPWE3WwIQ|g4-(YZglXVW)^VwsPp^M<W*{@xp1DqCpVyl3RIE={z3
z_s&fuILCNKSypR%C@`l3jj3VaLU8uge{97?-+ma6Z5%p2mDfK~kZRiFl%WU(Nk{$p
zsgBd`da(xv<0WD=#fHR)O&;I!u~v$}LGo=UwLh?|z54SZpKE)$E*l(Azww1_12;dp
zW%ShZzWb;>=4VnrX)4A&3zHDziB0|Z<p&mr9$wS&=}n!(pGoKRsC7;+ZVnukA<wVQ
zTsGqy3z{P-CDo+Ep`~+eMe)n^7zcpSQ~^7Ob9iji2tISmD^!rIX?I`jJ%p)s84*D#
zU?h>PTfM#Ssw-E(`=fie%U8Nmv~wr}z5HOxz|Go>uX>!Q;GE9<_H*jGe{LX62(Af!
z&X%s>k7V;kv}8D)99s{@7>48N#;&2n$Ezy5VNE-lUf3D44g%@`WRV1CYHMExZ}|Rm
z`1u1{v7|K|eP(0#+cJ55#{SfLB2)16kEOox=*E$gZhCYlUfdNgMeLdTg0LTs%<Nar
z2%0ATLYfd#JLRY9mG7NfvubnCXZpvI$J%|DtmO45exK7N>lsO$^rxpf-W_(s^N&^C
z#06Ju?~?x9PFixjr1nT~PCxwL&Gd&ycbuEa8OIxPvdeO4JRRuhAG_-4S=FBVUKvz7
zh6}{a)~sdLOiLaGu}S;Q0dWfZTI5`yA>@M3@9S%<3jTo~h8uW(TQ${0gFCR#gi3}?
z-l*B!HTI34edf%|0Jtx?VV$&*Wxfw1kC`_aq#wL%$<k*x_x*4vmbuXK{ty8^w}!^5
z&_g~=eEZZ(9vYQOO^PEv=(n>b(=zc|B@i?r)Qxwq|KeaQ`@K|7w*~qVIOAxmuUH=q
zdRD&ov{?mH_+Sd6I@tC?cJrqw`p#Ydcy7-HYdZ$No670)>>D=c4Am9>=VvuUZ#1O5
z^S!6fG*C*+CPtEIeP0&KS^@yTxAdjiz*uV5i`xdiH=fR4l*;PDK6yc}OIC-yzn$M4
zxdy=4`OBLvtEW|m37Xiud5vYoAqFNmS2y0Zjvu?Q_2@l=v5U8LkIXS7=IT-|%@q!e
zz_<iURaN+(Z><Zz(pDROJeEv9(_HB@R?Mo-Gd8vOTbq+r^N$gT3(lV2HpVvZ9@LJW
z*Vx`Ql)89(&&bj7bpE7F!DyU5(3ean1oZmK;PVUHs&8tFxbGfF<nqn65#y+qkQ~YZ
zRc>X)a0)KC%;?ab-q_1i**tHot(>*KWAJ@FBgre`nS6Aw-ibu;xV2bwb!gShhRT1{
zhJ7z=?i_r@<6_iW9pX(j6~^r2Zz#x}*VxItv^LyBkZoaYsDg8PeDm<!$6ncUYb>2V
z*8aJ=ux~WzaV_}N#Y;w%R1i}XkCukS_MvRe1FJiKx@#czj(rSasRd?b$h&4iYt1i|
zlz)2XsqNBSsO*^?2|8tN7>1Nw@RFOD(hB^+7rH~^>HL>=_Km$Wlha$tcFoEPU&qYG
z=#A|Sk>8y<H>#T!E-RxHuSHugt41NCB41h4854|)g}eF_7jN$wS&}UnExMsrX7c(0
z1zUUF8hkGCY+FO*l{rn75BW8$9*kwi&tBRfU7W!@Tv~?>No?xchD3zjqdM;!8uj^o
zzQ*p+%p10L4V@5A7nY`Sg(ak>U>Y>xQgLY<5x-~4{MPF0TWWlFtnV1hT(Gj8TB|$+
zKq;xXZPQnpn7+WcA&V9{#uzwbuD?CA>w;}P<JXTR^RtzCP@tlVQPoBRTiWX*H!f<f
z`1A7i%EFn~J*po2+F1aIDM^eA%SHpar7xxJ>K||1+8euJIF@}wDq9HI$Jp=DWOar2
ze^$({{@U3~n`6?D)-zigWXfLL(vX(>j&#Y`ZO`szUN>{iX|A3>6wA)*9Z8<nJCc|;
z63?6<37-9(xZExgs|@*`n%NN9*j!Wbyx+|`x9l0}$mI+b2>R7KPHIyCDQV`|a@$6J
zBwfG?MBn}UEA0H^W;Kq)^Ka-HPMp>|91k0cmS^+E3|-PhEPweLo)}Z|d9;Vy>!aJ+
z>%w;!M*gMotjb+>T#GSPR}V`T`^K|kd+(Sx;0w;|9!Xx%JrrL)6w6c@idJUwdU#sh
zF<BQ=H00gfSQ&g|UTgIqUf$ZX>8zt#v3y3<P*RoR?idwy@e><|G%0mYWh6ACdpLF0
z&c3mgBk@eVZm2VIy4?3-H1F?oYXd>Qwt7}W)w<b@kp~Ke+{VtqRPOZU?M6#Pz+4B!
zizSp;x&S=2p-1S36g8Eh<!g5hU$(1n?EHeE=1;ZNxHX=y40#_|&{q4z%!Y~=Vu|#&
zZQWzps*s;n&a0!=YM-v8g58p7wfQ_0`LOnBE;!x&V)v{~U1NXi8%-T;t51mLnu_~A
zbl%ecstK^3eO+s7e_E{HJ@T<<H}(8bsxm{$6p2uNkJdl8sq)r_%FqulSW!Q0|MZsr
zEaujE%*0HlzH|FJeEG6vY->;4w{B1T?OVG?{%b6mU8!uxV2@iQ=QLIQu`2Al{<7nm
zyDai&*Z+r108R2HO+vvmA?OFUul>SED);?N-k9`)oHPBz1x?>PX<^-W8^Ug7Q!v(p
zs6!!u6U0JFz$I>ZblcM2k@QXD>HLy{p)~6m7-R6c#Yj_C=+Vjw-;aV`_nvc(X-Wn-
z@-wgetJ?MJce+>Yj0bk~42LE7J30r(-<L=i<`<;0OhVwCp~CMOXs?g_thU1Qi*=pD
z@oV04l$9yOO>#%mDSK0}1QQH<hBMLIp4oZLa6J1KFjimCrOS|H)4XhNA(=v81ZQw-
z!XTm%MHrh=6I!!kUfowuo>$Yk58-C!DTxD+Yq}ENj{d}lcJz$CZ#<Q&6oS_qQu&J~
z@Tpy3G=)3VrSUP3ix1XCgPH2E|FLMm{fG09YTRnp2_(Te?poDZV?ey6cR2CBf$>x@
zZ%}|UG#3o1O(uU6FH1ylYXUA!qz$RY3%ZO4d>SGFk7}+7KhhBO{q&55wVgJ%gR%}J
zhbY^YTye4RXY!_BzVX{z_KbW!Z`P!;2w>RnUU2n=i$|1HG|eUnE;#?`y&JFWAItnC
zl`XiYj0U!^69)J_u2@4f@UO<2&<&ZK(Rt3%?b(J9%Uu1NXN{j<`Fjxk*7wl5L0`{s
zEEEp<&)%|U_(SnbVMR`tZdOv7!U+3atB#t}aLt8BH$D$QS5j$~rQWK&_i7VBn>9=$
z26}k?5W0qwKiJtfb_oL-3x@O<QkCjc5z*fDm)X?rcehIrfFY$+thuIQS9@LbbEhn-
z-x~FTm&!|-Aq`Uzz^As4v%%5i72CQ-K02Drg*Cxz45dQ4uJ&^*F0nue&Y@{MCK&o_
z!+}&&O~r#10oPwE{GN^zXGe7_#)%Q+bcX~}e}WMN0g4fUYeM_$(;dsZhEu;AjAa(+
zWxCAo5wWJK(EXJm-;DvEyX&kKP05fzp1t80ivIPn;0wFr(VgAHnoj&(yZXj1Pv!Il
z1zp-pETJOc8EL8t{j4VB{_&lwcPIXI_34_Wzii&;>13M~la4s<TQhL%?*6f_cMc_H
z5D*lM1prj#bvbns$P{{v2yh7ww@c^*8Z)HIlA?6j??zo^sINZi|JJVlv8O+M;S!>W
z!oc!K5<oM~jQ#kYjc1M}bHB_LWDNn4f?f;_U~C#;v=YV4@+bs?R(uzw6b-lq8=u)!
zwQ6x&<=2m&UD-Po0eoTGpx8B(e)opX;m_p?GAKB&RYd-RE}^JwPr%ge;mX336cM^T
z8c)t@s(fr=TlF=^wO3@U!x0yZKeBOT$?m@Kuk0R-Ey(NAFF3D}ih>10!W!f$Ppaib
z!>PCrd0e6(6%BGmsk&$|KchbKyG5;4f0$SA%@LrwM>1?}=XkRMyRdgS@z(C4M3p3}
z;EXlo4Jr1Dt+d@v;~XxHrxek+Zpf4-xN58nqAnVIaBfTG^($wF2bEHsCU}LKPIeDK
zXK`W>j4>l#VB+Qn)?e8_p80W3ml``r0!T#cFHd#M-_aL)ax{^vp?&P3){A>wTA?cF
z-8iSY>U;fTiDyn((w1A^8q8k!&HHor{&p3p>pIt_-gI!yj)8#R>z>mylK6Q4XzI*N
z!HAS9XmE~zNA%5YuKe2Z3tRp<|H|8A@}KXt);HPp|KSioVpM9X%u<rm*Z=tZ7o}9+
zqq2$gxI|`Id-VzLIeqpfixAjV{K0P4hLol91g4TTHV^8wq`|}Pc%i##IG+7-=RoYz
zY{95Aq$({I&KP;!Jk?ko-V+IU{uN7QU)nto?_JVfd!iybrEes8;#e~0Cy?8+e6ixD
ztOq^Xvnv-ge`juE`04f<Z(bRaz?j`UOT>y+wW$Qp8aKf?J+yx4te1E6|Gsx5+02<$
zT8gGR)J<Awj7_mVN){vrkH#tB)joCIyH@^m8sWCw0mTIZ&=cFok9%Qr-z_~Oi3O90
zmvz&9Z-bLJX3k)~j}Z-epMLk*3;(OF%DvsiIW~4Dnjc!T=Z2n<<V8xsqlhNgSF-li
zDXwdmC&o&a5_sIgXsxce?dmrzyxf*qHf^lcfylPacFGxmXLpR&J@QJ|y`#z8QP!i?
zNBpBrm4U_YKYQ+wl1lB10PY@6*WLMS$CpM@`FCY<dZk?+ryESp7`Zi`uCELZR)_tM
zXY=~HO?yVx)r5V)*4pr~*}QS~U@WyPp3b|af>)VwfF73ym&SL^ZmPUzR$buwGZr@t
z08pmoPuLbeJJb{MArL_8WX}KkR0n@~=kN=LqUA*8r<((ZwdZoP4w+z#j-J<e(@}Hk
zKe4FZQ&5ToSg9bnoJtlx{k!J`5k03At+X_fy%0H-hj`KSvd<WU;0%gLZ>kCn)K>=2
zf9FXpn=~%OzBVJ<fy~mSW(W!cV_e<&Vo%G!c<P$2!T80QysR_i1c?)zDNV3+OKqex
z67)PW5=*b@98B~dJ-6XJ&e)ZE2F6`ushn3za+8fX_`R+|$m@D$<@}~^2fVIlPM%j;
zkV>&B&Unl`%<dgDWk@i=F__S!e|u`jcX#!VUuhaDQJy)!&x$Xsm;qMaRH<mgWac^t
zT!OvqbBhzMeaA5dO<YcCM~(n;Q4}WcO^@vO`np|1pSP1NvKn^N^zT%%!C5Y0+fV%-
zSMJ#Pjo-iQxaMo8B5psucVjq})i?H!r<<*2Gnh^m_o;`=zGFbp>mEA)xY<{oJUe`c
zC94Kl^v8E^_+j@*;sXGfGAC}Tf|k?2>%_h^Nnx70*`@Ks(Q_JpaOrW)-}Sfze)F$w
z6(fo4s^NHM1`#|`77LzDPc5g=4CX$^m>Gl3pdU4>?z+>LG+p0T={9Jh8;ZdQQ+j3&
zL>4UL$j1I0jmER{pWWPb^Wb>q1ROL0RD39wbKk$F=kkvJ*fp7goS82a9gyXuzWjj8
zIm=dtyorXY(DNGSYd3cdKjRjhwbq8`aL&&hiKkB+j%WOaBEKw#Z9oVPk4wbrqk$*p
zG)AvK@93r%T@2FH|J-)?XQoaft=#%yU+^}INa>_%8J_?cRyhL3z&tK^D?{GKp;%_G
z(|dDusIwyA9k!IAjkq0ZJv(JPIWuBli|SnjfG$0*rO#0GslPw7`?2of)O%C8!nyHu
z-c3YM0L$cMXlL)(VgQS+i>>V#+&exCfZyYS-=n=)R~fwLnAx?zJ9bud5JXH#Ww9}4
z8xykm=hZ$`9)<nhoi5I|y994mir{u}xHSQfTgz&KXKkHmdI2)q$}ulvXd<Ejz$AgZ
zx*>zPyn(z?ELrDuYq?OsJ3g&IvWc;?o+s$_40_$7$M4n_WD5p}08QiYx&*v#t-!gN
z#J~^-&__g)F;+uFb*5y9f^LL$Ln3FMmO1B|$0aI;#&djjRnagto-wKkY=L1Y5916z
zk7k|vbDZ&vO=1R%UjzVt##l2I-AF`4?t*TFGI;|9LjeR$6Ks6H2de{wCIqA*xs7Ac
z{FbWWJ;RxwCU^ARR4}B*7z2S2%jD$&kSH~?HrPL!%75{pb^U90bdR1V6`hmF<SUEa
zK+H{}Ob;j`meXahb1)w49E_LBF$~AE7>;Kr9STg>0sLO~)_~WwYGG^D&F3y_cwQ4s
zN~IW6An;<Sg)Kf#y@nO{N^7Dv>}R!+z@FiFX1Q)CxHS%sOTg{YAOs(`#+@@ZtcZqf
z4OrVzEU!TTkS`R8bUJTXZea$Z{!})v81pJqVOjcRO0?93sG!ST<H`KWbY3^-pTXyL
z!R-=C)5N%0vo%G8E)Yq^Sv3*WGR6c^QSJ??kkbtqlE4{U4B$kh+%6YXv1sVN<kCur
zleQR7G8BPG0bu=7(I;<T)%8f{K>YG_PCqY}&iRxgNChqxg?9FiEn&bC>jp1x@1Ite
zc|F>OrmE0Cj+<L|)AAYNq2E8TTb(c`VqZH}aK4%@R@71;0$?L4LwH=m=Mn-rqZkD3
zbBm${%&lbvFZDWRjIlV8>I0bN9;#vtb@pZihLjQAkX%Ya&a@!9Tmn@gU-$9zYKsX!
zfD}hQiAhPNxZqrBf_H~}ZltmWC=`1l``sEuQCFqFSfN;xoCW|4amH(js1^(?UogUk
zRH9(WqJ9E{OECD{LXo0f)7`u>=+!kL)+V!s=2T9%T%Xq3FofWVQvFGyQGoRpWt@2d
z%wi0j2rO67qlQtGN2Q9_s0g?R6``}l2|1&j4y6gk1n0~!6a;7RdtGq5M9!s&Oi5@7
zjB^Ysg<(tEFfcC|T8YSIZGuGP%@+*6E+w*s;=Fk@5lZLsH9x$2le*@zr9u;&=XE6v
zNuHvoqZoVVatS3kAGZo50~jZwK2vNsfY|`t_WR@w19@G-t7$IASnWVO&u3J+rDCOe
zm500^nDQ?fftIxdKtx-&bPxU|kuEIA<c-L|%2i#|pDXC{mVa#=+_7o*@Y{0*<LqR%
zFfW^*XsA=Z2Ph&aNxq?2#y1qpTnyl10LTK^&^cVD{|Ro5Bjoptx&(iswJ!Y7yH08U
zi{LCxL|hq)z!|r9UMqEt0-%-L5M~V`L&{PBC1FzCr-|bCrt4Q4R&jf%4V{T*zt?s2
zROV^7CT==ze$$xcj**5WvjF2ES!%3+Yde_{Tuk&81dMl|cE#<3smGtX>)CB@i22;d
z8A`ovB$2Jv4K=MMwqCR<<TJuS@1uZva9&IGBRhIy8yRCTPs(IFu*$M=*<XoM#Tds&
zZ`rnIK{fljM-#VY3I=>`fk?oMnrI*v4*If0#E8IZBL3dl4UrM0h$XVRXU|}~Ih)r5
zj4_!vq<1uyj*iCCY&2Dj@2Cm;V<#<{@x*?NCsZQ!4D8V@vCgKD_u5Q0e@CuhKmqV*
z9F-vtYNMf4MZl8+fd&2Ucza!>&+QUM=U}{cG?A$VF&IjU!Lej@CR<=5sT|Td1&`qH
zyLrcnb0gbeTBnxJs@(p{j{eVa&e{Qnu-}D%$Bn8;AQAF=($+gyR`>?zHbw`DfU#88
z*EN)ADHzhjz*Js0JVRrts{Zj53WkEm#S3#<YM!^=?Fc3@l*Xw#z@^f&chyA#Z|NRR
zUPx0|ai$`H_MZvPvJCkBSKhwj_CG(h>#VN9_&mSYePS+eoHLrt`1AU~S6^{y43!~Y
z+UwTt(skpR6>}TcpS!ef)lVPTuBg;r5+qy#N|(xt`?TD^uSNWyjV~|&w0L$CDen5l
z<7;={UoaH+x&@*^52E3K9u4}&irkt}b=W`BQWF~H#ZYx@=p1e@=rT}r=`!>U$D+wh
z!ILQzH@a}pLlM8Y3qXM}2181<KTGgOE?lhE?udQ!+4XxKOXqb#AZVPUD(FR|!k4R#
z1jdPo`8}@G+@{JsZkJG<1F`DSM7CBc2t`T^ji<utOu;>p%)*ccT!N!68rXB(g4R9e
zfr}2n8ZD>t8WeRINg=rLhbMO3{=Z*7<FP+K-toraMCRCnq24?a&$L>)jHYo7G6sbG
z9wQp^{R^ZXncH0T>}PLT{o-w3s0RRi;Eb7g&tyGX(q{*-mI%Oz-y^Q~dtJ}xk_H%O
z2>M(IcwDH724fXLZ<fq}^pM|^Xs-+RdECNajA74UymC06saHhYWQa5}o{lEcId8#`
z7>Z}0BoOqu<Oz#fR-H7j(jZa_wAUf_uwmOzaz3y7w(77ioz4~9qzIZ2s0w-E^SY=y
z5*%~6gbqclGUOeY(^NUMdoWQu981@N0Rtn6NHUZ6#nX9{00KUjhUyAm;<QE8|D4Wk
z`>xYwWuMtP@|jeoaBn7WFi8ZzTSH~Yi%7_yj)r`RV(KJV(B~dq)K=X~M2MyHo*lj8
zvx$fkz<LH_Rq0$o8;VOPM(}$yv^PYz*M@y-?E6y4g@jk*zx24oi@G7<(gdnP9#mEY
z5|v?pnru~it0RHIw%X7LXRO$=F{^t!ddFK0CEb=aHdipbLu1KsB2!@5LKX}_*zeAt
zw6O7ke?Gel0MxwJ>Lla*)8naJGv`1!;DOJrp|ZlC3kQ8AqeVr)JwB(gvJVW<b;&pF
z8JTS;<pwe2b;C0-nh1}lvOHIi2>M)z__cqXFgubp4-jfP<Ke+N7`RK6tovQ#@~^*t
zXze=-hCI=*6;C&$+~+!BW~QJ^<HW$8{^F&}p1$vu{u9@A^dIGl1eOwT!FV#;oXYA4
z{jiJy(U3=0_`R$01^w>!`snt{Pnq%fckkSokV-PaH7Fvbxio8B*i`yovoT-m+-xU(
zR?MxMq-ToOJB<+(S5Qpj*xobRnlBj3Cyykdy24)&g00CHjJ&NP03D!`+2^9&hoOt4
z@hHFbg`uH8J+<rBC%<+6-G6+d;|~F^)>IMjotP^ar>8Rc*|~xtESFnF&^zpNYwMEP
z!X1pzee#mlJ;%Q3C*8OI?tS>wn-;*_SoR8@w|-_4+i3;Q{=C0<|6Kg+-i^-?=oyR2
zD<T+@pc|@`A)3wW7)fM7M93B-@`l{3n{kQ7vK98UID%A!Y(YPuF|(fai|04v507^|
zrwGs4C#Ps-P_Q<cd_f|X&cP)(vUz={^>e8dlroFH(uC`yd=RpEWC(a@(kX`shE&ka
zxDx`9%j+0TWI+Iu*#ZiN+^8GHch+U`UNn&?T*?+af`n$$0wE%L&zbWRe|Ws(cd49y
zZX%O+SNOey4OKyf1MJKY@fYQOUS~XINOku$Z#{}_AIvvCyQ#Z1mCb97RTXX3VgE7N
zg0WOLWOXu~Yn4(>a(%m9f`Wceuixw5Xc+RP-2-E*8>@o)nakU@|KJ}ljx3yEAW<NE
z;{3U0%o5+r@@rb`Uo+dE&?eS|CIq$*>Xg@wR|xb<iDj#z3>h}voh|4XPvuIAsG!Ro
zmc<mNY^t(WO;Q0T%!(Ya`+qb6Wb?*mMc7<seJ#m75h2ed##1?Hf+L$ZIts>wyi4<6
zdnkjo1{rf7$XM^M1s_SJaL<3A#sn~a%VRtL^u#yM{olVlyZiTyR83uF#p(HiaaJ;u
zZ_gJDml=kkDuTWtuUlJR(Di$yA-hgqGGo`#?V;{pKC~TMe|k|d5ZojqWgedcY6)m}
zz5f2O9UBR>(LNVSS?hX2m*sQ=v2+evj)O6VOkRg>sM0(#hAGFBh>$DjC>R8f06eeW
z!9_8~@YiSd3@Al6mn<HpZnoaZvJ_<V21XNEWb=B5Zb*PJNL74xDzm^UA`Fb@_9+KB
zUG7jsJM4R>GS`MKWyvLx*NgWhBBXOVbVF_?Gxku?-s{4U3Ofg4m|t5<WubekLA1K`
z^47hqIV-)Vj<Mv#TFezB3WnTdsEPGxNCl-Pw@gz+NM&_=^=-@W6(Z6Nr5lp&HFGaZ
z<5Jx7CJgntf`PGQwshTsA-Bs?q41)-P?@Y*2cIKh0GMhYiKla;GwZ|OPUrNSvqW_u
zN`*YMy>oXZQP=LhV;h~M<Bn}x9ox3OW1~AA?AT6+9osfKPRF)wzx#fk^WNtZod5P1
zyT+=jRhYHrysqCIA>z@YNB!qa9$V6cEckdx?J3awQQ{P}aW6Ji-E<<Q>`a`tuI}yH
zs~m+~nxU4iuh(VVa&8{N9+o$Sh4!@e)qr5Lq2B3i<^<8nV+Pc=QU7Jp<L%=b)t+R0
zOD>fmOg)my6F^I_UQ9kDj|*-@o`|ojLym08MHv~Ky!H|}x39bMG+rU$@t%FU!ILq*
zo;apBd0<yw`E;tMl>L5;ukq2V<=an-RCcfcM4UW6=N6I-{Jbk=fb$|T;>N=xmh-85
zhMdAlGDZHU|0Vu5GykQpSOipyR=z3lg5&oz&PC|R=?wSzOiC_LQ@py!o${*PWZY3*
zS+z5ycCh$Cd4>gxQ@=FdYoO#Bv^&>ap&7^L_%<EEVAAt=^Pu9ymboZpMjKlthmBkW
zFWGmxz+qcf%5u=Gc?`4_v;q8Fq5lZ`z1bOxFB>lhZ<YUn);ErBsre^rYt7iH1EJ-!
ze*_aIN1OlX6hE>(VOaplhl32EtWqeSpj>>}g0XiGF7TzW;~A?2g47aMH<wFsIT#9V
z&KZgP(P7Hn-<Bh%6&$%*t{tcw)@YOS{Vg6Z?G%tEpJ%58X_)m5e_p=5nM{{4`x>of
z-+>0$tF+0_6$*g>Jj5^x`jrYUOm(0x-Tk1$?~1P@Y#zK2kP?B$pa^lkC3d}&?&NnG
zhZHDH@;r4}*DaCKKw{`ajtsI(2PLP^@nXSmn{_-?5a8f&{1X0&mD^^0JyDGlUq@LA
z-!2!5fsHpFL!XFdll#6-_jVMy>vDJ^r0;4ErIyel6OKlk`LFj=5E4Gf;Vd#Dm0#)^
zUyi_Bt`V(JIs|p&#hSio00R9G*;xMlB0^E$<pngm((w`Fe)lYws>8%4DY5pJN?S@r
z&bfr)9?0+Yj?p8-w>oQt1}0%&YZHvR;TB?72?Rnk>)}QuDO<|t&YOt1&{P@Dp#$Jh
zKW-lZMP5^0qTqyP)G~@m0n=Va)4;#plfa=Kyu*_&E!d;z@AG;%7)xj$tub9!DHemW
zkY}ngZg5aR;P3hpAZ}P|zUoChpZfb$AsHtN`k{X3YuGR?rgJ|fp5?!v*P}>ZSW9Ky
z*U~xzbLYi=A}Kz0tt~dx3ws29-}EYl<TMkpn0_7c9-e&OLIC!ySCc*0!<TF1`Se8_
ztg-Uqk-E{8{@%N3f{E3=9PtvywsXNQ3Y9@nS<8_J{y~}=H>c)`Z?%Lua2=GBAxI%i
z)M&x6Y!_VRMarm8|IAJ1HXh6Kuw>phmL6qjZ3t<fAn(>ZD>2E&CEPje>Fw>8M{TGc
z?WY5?gWq|7f68Eqr9Y~8T$UGonov0l^oV|ucCHeaS=>Gjj7F+zle(RctvYS1D;?`5
z&!)Frpv_-$DEA3C38{UUpx2XG4rx_@5g(iiMj2FX^e-4~^@xg?+LdjX54+?jJPC+C
zlud);5mlYDbSST>k@b(8?cr3Q(Odg#i0@lBgW(5vAjS$c$XFNlZsm9B1k`GBH&1O(
zXoSTwAg&l-IQ3j}j7|-6)A#x0Hi=~|q7C@o)K6~)9jZ;6xQMu)ih2#)+Nv?pFs)yh
zC)3Jixfg~I5e?F}t@~)sFxNiKEZ>=`fQ~_mM0kw5*wIZ56WB{ITmkZ%Rx&%Dz8XF%
z?~ser8L_4Vj)-OQjsGO-+4iLd-`-bf&*&jQNEf&<S=c=e@CN&{XtBs5P=)}@ki8Zd
ztjjy^mmaq41XU<cBRK!iDZ9c!B~sVu*Cyb!IA%f*!<-pcD=hL^6HdBwkd1(PEgmv7
z_IsaF{!XT-=h?&IqKfdT-39{ZYEenl9Eh9xc>px5(0?)#{IH|s>itvuUAB69A!zOC
z+zEx>JboZsdZkxR`5xz4>g^VV>U%@xPB{0V{sP@F`i2QN`jS4RgfRJ!jTe&v+I~+9
zr(_T*9d=z^wTUu+8SEod&A4_~Kigm7&&X*kcY0H>{FDKR^c)fx3bkK+r5;D1zB=K!
z+{m4!Sf!G&Ky;p+sxS=FbP`#$NhiEH%zJYV9;IYIVyQCQ94JYucWrIJfOy^H%Y6NG
zfz)QgDkhvZbj7C~rL<=|8TkbT_X`1tfpPrZ-k`jbS@@Y<?fA2<QOUx0v=^)AqxK(l
zGPwg?1{IFUW6zULR;gtWg0_X3{aHPb^kLLqluK7*Yn>@3<s>q*FW@ptQ;^K}azhJi
zKRUaFEsC%A2G#JACe@7pOktepZ8J6aeYDrm1Lpe_ue0;sVgXg8(T0EZUoE{)J-bh7
z!B8&3DfxpUi%u&PnfP9>i+71TPWr1fB*j3TdAWjbyO&4!G6ofL`;dU%;IX_zk7g<-
z0TV}#U#4NEGe++OL$nHXeZJuo(<*tWhK%V;n`HaSduzepQe*shoUd{l4beQKi7z&9
zWr}2%H#X@da}4}<OFWPdIhc6-@0yvF*}e;8%Xago%mil5TZG<I$LBr9C*pGW?MK#s
ziJ>csWSm3YG`i>kzrIlkd7QVl!V^@^_@^h75lgdbkCrFibC(u&yY2R~Y%g<(H+^n;
zyk1so-GpQ<g}}vhr#aT!3`YTfmvyevq<LS(Jng3Zf7t8zuKzn$PjF};J48jrl_>?Q
zzZTXI-Yl<RTcy?AX|43dx8p2Ber!3oLUK`cw^ymqt}mM7j!z3AO&M_G0j-EM?#SCs
zFTwwFNdI$e9}`;|I3KHz-zV{)62K+T@0UwwT{G%G)hPfyyE}g(o<1IvC?`m|ADSE<
zC(AI4_R#fE=%_On{(D~T(3MqeSMS<<foEG@Z4QAXO)IUe0MTyW4k3c4b<j$W(v;y%
zv}3+e&^w?<NLCj-7yoBcyok_y$|ASgQV4blB=54X&6@_}2F5o^63dVfCx0VU0gxO2
z{qW$>gsIlyDR7)fpfb|sMovQ@L7f5qcZg1`9{aAmMJcGj%z3=!4j;Qs>epe_dJcsK
z=ONn&5s=AvoGqb5XniIfKS4QUcO>QQ>ZXUO-b2;Azxl&&=&O|!AMkVde@@2P+Bhs}
z1c#z+2<d6}Tjq6gGs#!ObylA>Wr2ESmSS7FxAuIK%ZnlfdTxl!tMMj`ERK={T1n72
zU!WzkEM(ppW95G?_5TjnfodW>icH)3JZN|_xee8A5Zf)d__HZ?9WFd~`+>87rSM-N
zS!MoYT-Xm2X(|+2@TMGvEUy`S|Nj6yUPPhzs*vhu3lOmNWh51R0i>{O1|5V!4eeei
z;vot@5ZM4RNK60sLZCW+*g3g6ai%o~NTn4YcN8l)%Adsv|M#9*6vk6-*wgtj88>)Y
zIe_{bLHhO)wfDbs`QQEizx-iLz+weYNJulr-=|Gv9yDrT4iLfmwZUoGcADf_3IQ<1
z%3VEuez+ER*i_I(Z$jmHZ6E6GkbB;g6%WzAey;S@&E=h*<X7{*5cz0&TjykLg^GJm
zN8QJdN%K=+^-y!C<5Qh7HCWvWl2aU6*dG5>pW?`tPdfNXwV?j&9`jH8@|1T>*etry
zzmli|S^oZQ;hhqWgZI7t+^M73m-zMmOc=~>i(+%ja}6jQTJ}M;izYgHsXJ{KpRa^t
z+eH2N3^_m!4i%@eF232r{aqUmmM$nRSc-p-tjI8jJ-8)20%8w9i`K)>V0^8wyEb-n
zMqJ%!nh{#?b%+UL@GovS*3Qg_E==<y8HA`RCLRtY`!VNWJZE5riC+}aT^m?TN~yhU
z)97b-_O*yp>WmbA&J7OowSE~*V4;zcilEF6PNc2UTh;HzSJCscbg0_T57m~KMX<GD
zB|wq#QUG@32m(Z=074<^X*r8>MpOt2uQ6kZx&}!Y3B(T7Smo|oNF#LND*r7eq36G*
zPrbc?nh8#rJYHzwpg<&aE{(Y6U4v5RrCs}C6Tk>x1vLzLwDpOGlm{TSsSEpptPB`=
zYYlq-EcTmGUugA5N3k^gg%#@NV)7{nMT*y(PdE0450!_9bIHgdIvZDX8kE-~;$dMe
zJvw1UahpC`%rd&>HtI)=DngtOGra@1z|O7!pAd$9aH~TY5;V)s1^{VaJ5{`{z{8sj
zP62^yqfNhY0yR6}lYvyXI}7}@c4y;(m0#3BJ>ynpTsNW*3d3s^_y8)7-w(Aq=eGnY
zSuAv}v{K5EsT(jMdB|MP*$%sf_U~h@&pETjYiZNaMDU`VpFh45&QkVLM`QGMo$5}h
zVj~BMhK$9|9k9Euu8=0`YhQ|pVPW0sCRw1;jfH3?nVmkT@d_c*e5{jUib(N=w*zk9
z!&PK97)>Ya*#;+Gbx(6_Oiim4@YB4$(;0$W77!U%MbxtG%caB{lJOQ&MFAZ)QW>6G
zcK5NFbTH5X0)RRYYaIro$ygiR=7_w3@+4;KM1s^o`wWG_ES#9Q4U_DtEUU;ap=Tw2
z-%t!h3v@s0?N{fA{ES(Mai+f(zuFT}Sr?J0VI5CcTS+>t7u7O&g=S}yQy7lC3tDPm
zG9`>#Gk9opZYvQ)Cmj!ZY5{yaG;xih3ictGiQE<zS{ztIqB41%$WUs*tEiL%8Y;ay
znW?S3_{Cz3{)*Q-R*Zd??IhgXjwo1DwhpA`{Ll*IV#Tga@C;_q?lqbst5ikf(UHp+
z->F8)zXoS>a8W&D0g^Z<bTXSAvXUAjbR;HT)J%;JUdwGzeL)!5+#<6uj+t|_?&0Bh
zcu@3${1Ah++}u=k9tK~@a#8!FJ}PRdY^2$Y(xP3J-DkqbahSV|Mt3;LJk_~k5NRfw
z-K_!E)l(Nv{bo(f`*l2v18+3^GR=X8I+q0s8DJta+G*TxWLn02W_z|#7P6_a2mj~#
z_-tdsK9gJs05CLI_B7e0_c``rz07r)@Z(w?Zs8M*nK=6T#h51^#!7j3vPxs!<!5~=
z=k_d4BWV$7L`7)CJBtaUS}!29RSx)jywzmDUbAaQkH(>I-I0hm>NGCtBq~(6!SU*D
zTl=Nq1aXrn){#(G??H&;=LcybsmtmcmO=%jok4+o6*Pft;S9KC(=(5yM-i6*fe~06
z-kO}R6;4ByY-;gh&fIA^me9CBQIKDu`%BYigjS}rXwIImzv`Pe{wQOmLI#i5652=#
z12`yX1BlK@a=Z6=!Ko*?zNt%a*}@m6Au1rsGn@BolrEji--1<^0<ItIsT^DTF)ADy
zug21SDLHs`-cVn1S-$4BT9c`%>S&T=92D9)+?*7xyYOA?7=3GLLA~S$GL?l83)I>G
znC*$A3<jdAGP{xUuomC0C80YAa@Ll9{i^cC>Zuxn3%Cw=dl=gXVf4UQtxyIkyNXjQ
z`mZ9g0IV2nB&e*mw9xD!@Zz3^;gqi*Hn)MP#6KUi>#-+r<mq*~#ucKs(e-I67mW{E
zkz9Ib9K77bNt*1x_HbA6U<AY>N&NU_c-*Tj06rnUukCP&Pxnr_{ri`^u!gZ=nbkt;
zbxpIWj+e^%N;{({>qqzs1-=ew$|QaKO1W^Q@bxEj%J?^VK%NMFQPb~%kV^iKV#bq=
zJO}wjUdP`(x6_zAzm|8o<8mW#7`z^){;^^RohRx~GtdjhcPBd|kTnS~!wd8>h1+ns
zk$=1U*4?M|;l}B4H+cPMBp4bc>_?un^e^^WZxfCZXBSr5W>6tpW#*&}?QbHWr>e^)
z`M1kf3CS}j`k#Kdu6HR3abkf+7eMrUfW({Q+PxSxW&|=z3Y4jizRP`j%W=Ull7;J5
z+2eYdsUED$Ae^QtMSTByto?AwraMC!y};AlE`LoUbxcOh7qI`5E8s2u^m=`@3%0jJ
z3&{V9^)wN!wDUru|3hOnthPfi(o1s~&B>eiFjo*wZngX>`)}-aBG=^3bqG7;=XEH5
z^m!-3KMi>O$JZ}cqjIVFRA+;mxfdetg)O)#jauY6v|AA<@7CogJL#*25I3-I)RO~n
zQyEcU0hkr~JK9Qh2ayxM16fz#vl#f6g;XxH4#r(LP(mIG9*9w}gJ75?nuG@-RpGV&
zc)@I*(rzsJ`~fzNfP!l|5}qNyg^&Yqgiw-2nHv`*Usq51GGj7v0`|Xkd+BC3*~4zQ
zxFY4LkBW`(6acIJ(b0@NA%eIe4?8g~1NsGMVDJ*ImR^MLBzOY7**m~>r|IZnu7{lg
z@B3-!{!lxgtA)gat<_7z-emQyknQwy*3E^BnQq5Zg4UYS;+>{MTu%Rs6u3oC7y+iY
zj~OBWgjS9)35eD2wAxNt?YRG{@)Bn91LlOF5ASX6b4y{-A7yx|d{9g1P0Lt^eIxLg
z?d^Fqd;0PP{c6aFOm%zk*JS#y`)sqcO~T%4;<b?2>~3_F<Ny%m?bCj!p7Jz`xnnn^
zXZOFy6n7#h;fFQ@e=FkohYvGLb$gf~r;lA60+2<X3+yC*+b?$Kb{+z9Ar6UmWzse0
zMOfrF8_ueiL5r5leaG>NYLO{z+|_^wTt}U{k{ePM$itAY73Z%a*g<jt4Cb!uUaT#k
z-JrH9`P(r#tG2Qhi}Rgxr|{YR53-lb`TlocX_BPJd{@ObDk0=iPp>(o*DVl;EV(&N
z$x2S%Oy%W2pj3l;eh`rnnMwZl6jln8R}~r-66`QTowA2F^4y+@Hj|~3KcwR@j|Z@Y
z=+;wjTpbQK1&k_}Wx%WB+>jmtYx5C6tafOyqQ|Nx0?B6HrD<CX&nbxq>t3@RW0iob
z#S*^v{)n@CqW4`Bm^`k%pHl@8<~~J{k%=gjb$_mWo4ZcOCA)Jpc8S$hml++QHQm;U
zkh`8<4-3KyN%r0Mp}^ofD-ALd1j7za9ECrUic&q&bidR!zxN!PGS2^|7V_A5+hh~z
z{Fh$gV>8?{pV=;T0HqwlPJG?I{n&y%CLftnN|!uw>!rypsD(_;Wc#tIr)2GoE)fZM
zzSq+V<kW~m0@Q?WRzezn8L11e+0kc%#Zy&b$o}3&+6y?CeLC`;)W2FUu>A3wMDbUZ
z?os?qE~V9It$V-hVta3Nxg$7kGrBDA3nDb?5WK{Jzm60)OUej)DfU5=x&x2dGiZ~c
z>=p#63@$+T@AqGyW!Aw!WwENm3NmcDhcXnzdZXBhV4w5<NMvOh!jr~e9ZDl7AL`jj
zhulf;sQpuU;!DD7^}#hqVY^%zj-k!Q+=*6A;#K1K$J=aFUYt*DTK<%F=U5=Y+idJr
z#iceWhkgr#f#9w#ZZqm3j3$E)%CmB1<#q~5YqBsbd^Jc_OhMb}NI5Y+4YFkjB1pK_
zCLBMKG0?*5^A#(-4|8%Eb3HD(HROj0@dA7h70(*Muf=m&x_pD5AzhtRXWY1riyymp
zgN={Acy|edhRz=a?c7<5ym4M=yC4jVpB)5}kcOx0=eWG+LGq#{GZI}-cCZ;5Bt&-Q
zDDRJw!(0|TR!`^stKz$*Zz~tS_mFeH(>qUU(iZuYhZs6Lo`fK9JI+`-)Nkk*I<Vou
znt^mS41z>4Xm8=L7Qt5J`%h3XzN`sF{H|f_>5@6sCh=XH)7wdV%zItzx4^mn)zrJ#
zk}eLO9I{FLGWrg%t7|q>ajo*nKEhsfa1MqBGNTOA^ZijTWnCunt`t7cn`P*wtKoB>
zNSs8k8MsBmG6YF1iqZWG&x{hL2Txp)d&yY<3Gxrq*%XK7i|}Vx{lP#*|NZZ=*wD{R
zXZ;@3hlmbQ48S4MWJeYd4c?w5hnAmT{)r|WJ_P8Tmly6<`cJvbcuhp2;I%b*pO}J0
zbBdQ=FQ}On)|1L9TwSx7P<t?syzY$~ITA3box;k+fba=j^OIF7J){a>HoE{}0}|Gg
zyyROBV;XUDgn>9WwySsN)m9Y$>y(hkf&#Eif86q#mUc4qEB8Lcl6S_C9bTxVv@uQm
z`eDT1bIdK!zWX}fZU{Z3wt&_Z{WP!Gl^G8dbx6v{iS)tU=%fOdC`?4;wJh`zNHjIl
zCrs^RrVTK4=r)Sn7m-!ZF5*mmzL+7eE&?qr>5PO5iTvfz7ke8Q@Xpn2)cVzyxjb>7
z_#a9u2<TD!mvRk8{5sfc$j^|AOdQ&|my{<F<N_t)z^C^_zl9GAZalzXiJaQTu?-RA
z)l{>1t5-J?b>^6RTPU3lVKV5#{>)oQ?mRhTXtd0i@SA(_DW1!_onyWDPHcFDcGY*N
z3sdQE1qVX56cdYk3O$jO2d<g7X7nvn-cT%62SjmuL?YL(d@>$4b^lDwg|?PFxfgKz
zyZ$hy>ltopq+Qr|M<_82b*qIL@aH>Mbi>(2l)oPsv%%<X&?F}b5bwLh4Xtd%xq4k&
zPq+W=BanViSqGzJm>Tb<h=x%k@7x4#^%HzVpd*wcE$sbVTEC-8d&-jUy}qJ4xtz&s
z6%yQ9-zFiH5-d~t;F^R9NF9O*fq+e4ysy&ly5gMlMh%j+L|N?7UO|F2zX|)=Q<?=K
ze$CJm2^?Sk7ob31m$tbfq%%1taE|3#jh(z_vZEVBhI8YC%GBE{9c_g);;oOEZckQ5
zVQ@4n5yT3HFp<Xe4ItHx@*x2@qC??fV)af<K<jd({(PMyFozLty+uiU23T+MfSius
zEJ-7#bzeGYhBLMRd5B>%V7rC>H=bV%iZf)#Y;dR$ZdaeQ$THOxS$fS4MNAzO+3YDE
z55r90yePxpgXqWvOlo=%0{pXnc6|n2ZkEcI0BLS>D~9CbKZtGLOYtPZ5uTNvs)X%m
zgQs4|?<K$KWWTh7wYtFREGxM7&ZabrID?`i<X8h|Bh`vaQ*c)KIM4&QIKtPvOhaob
z&thAOKylFW?F&<`hUY(HhI&`6&7+wJ&_>3hMY$>^5-FaUv?bDomtyySj5&QBD{bzU
z`mxE{uG3z-V-FZZ1C(t83EF(kh4q2<$I|4Bn$s5i@ix8C^$ZOD;9VM2y2m$i;&lBo
zkna+abwQ}fV)8xY)CzD80M5G_s7O94zl{q*?b0Xc_*bd=u3oKYh)kDP=4@)-`nfN4
zFHo}V%{<A46@s8yCk9<ESlNXrBbdL&SSYT|a|fI#ld~Jq-lWXA3@L|)b+qf+^;dIf
z{35<@e~!sOoU){i2u{WfERGz^5U4!(@cTD@QjO`kp2#@j-TpGHct#MJtIVP<sDq)4
zeAUwM0nUF;)Mu7oT(x<TU{s>D=zRH|rtq?N&}$?;su^vCNbm6&ojt>ylA*?;9!mKO
zcN;59CReKh#b^DejS`^*c#;^BxziSlG+ZYwi-3G3JME$DQsNQzeiZ46jS#si9B8=A
zXd;t#h)zpOtWQfVs!_-=&2m@E-Ugc5@zG7Czk|M?Rz*MYm-qDw_3CJ1v|^R&pBgjl
zj~$mqvU5=<F2+xs0?as<I}F(grx;LQ1yeuHmj#92&VC|YY=@L2fhEywLa+57>4h5a
zAGO(khi0EQJ<tYIzb0^0>qT*D!5OB{5q3KM&Dln&|Jp}Rctem4g)nx22nGtgs(=Oc
zRm9;In#I7MQsku7a>y|y<;e+}H%U0{CCu3>@2k&@8q8SQ?wi&p4HvyjGG?W3I~`2a
z>GRq%#Kq(gp5N^p%+-JjgWUp5LKTIzGt<#&s)!&Lq3*_)keqFa>AFHz)TcEj8Hx5=
z7_;%qdGnuB4%9E>4R`TmqhJ5-S?%A{h14RDAG$<IEL|>_x}-FlhXu2;S+MA_fbHY`
zl@a0S9lw=X`r6cse7)8nHE;qn_s!x|Z!FG}nWc-fIWxwMcnIZkr`Z#%$B$hDL{k@h
z6}8oQ3#@zUJv7zk%i3f=Z^@b;ID*dHnEdH`OO5}4rA$v{%P2Eu-xHPx`F+1DkwjX=
za2MX>70Tg35-%<kXGBhd&4m3kL8msVw%Aw^LW;<Fp`ECIQ3>UvVn0vt-_FJoc>pOR
z81BdxVoH^p!1Jn)C9sg5;?@8x{a{X4xy`{knBSPu+)V{srDqxqWDUVr34q)1%y+Pf
zKOXft2}wWto%^NCMZDiW3o?{^T?G4o?ktK%%yRx|_!<m~moV72=(BmgwWQ-92l=?o
z!B3_c>Clmx<oFzyI_#p;O_}MqubepeZ{o@AD>da_i3Xt~M{5KdP1(Tvaiz@7UJ5q4
zk7d}yA(DrP8iOr0$YbbwKA49U%e-!Xoyqm={HczB?#$Met+rkQ?w_o;na(S3vZlc5
z(6t*J)S!<>o3u$IO*C>bVq;H%RTxz7lct<&VYI<Ig?-qnl#FStnV>@=pq^crVz$vJ
zo(PigbNAhmyOV@PxF}T1Ru;gRCg%_ziu2RYa+T9Syp$e0eo82ptlIZ2$)YlselJ>&
zNq-p{V!BmW_a~fCV_nj`^+}ve`fm$d5&HJT^hb?OnVL<(8dTX6VMX(MrK2`T46-Aj
zCv&!r|3tbR!N&#C(_IbHfVN8_Wymbm0e1DB`EpsbFJ8D%mle-*wz?+T(r|tk8;iEa
zHZN5&-CZk((Ei&GD947ypV}Np+up8>>n*-dmA+x7)L<c`8e$uy6*4fB-pIBBF~@s!
zthhnl+@X*#XeVSh@M3+SZCLeejJq;iptI{VI}>c_;mYP-Id>i8{hQm`aXxqZRh%24
zcm5*2B4ZPMD;r8?&y7eZ(0IA2DjsZb!J?+D|GDcfu8{CMGKhLn9p}YDW90EZge(!f
zlpLbCeGNtz>E$KX6&CRIAMV+FkxKmjrBH0-kXT=!;NaKrq|?WIhpyI1RJYO3%kEDv
z-#R3?o+gyS<%OK5%mw?)c1wY5;V^jgtms@{`PbG%W+yKlIv7iWiIeIltgw$OAy@!f
zB}tzb8@C-;+FT85c3eM}Th6vOK_u9I5q+E4uQd4wDL!)*^G8x&|LL!3UcArYhmG35
zh;60;v9P&%0`t0GyltPySK7Y`*}wFgt==R<Ff$BaxZK(<X|pNuzhJd&J+(S`3bmhW
z`NHlx)r@?_FmVLV3tCmCxy*kn_k7SZ|4jOc;-}{%FDXuZ;EB{xWA=lI$8n*fa(%Z1
zyf_qMOrKYp+KYxQ#m*P41}kB1qArEa7NWuCLVzttQPiu^MX_{`pp)Iy6Ep14tQGc{
zY!g__S{?hMs6V}Mb>scl)&SH?DjA6+^!|pA2@1w>g31|_5$J5~e!0FEWVS%6YxTO?
z{e$Z3#BZi^+DP;s!T5z_aR|z>N_z$ijZ%a4PddLlfQScI1P)vY_*@))8vPZS9))A5
z$fPj9JNzv^=&+z*G%HhKk1wAJ)2peoYv*SZL3fg@f+`1zV32P*C2NxU58v(ph<}_3
zYlM5QHpb>}Oh+#MZ1Id<<FSW&w%@vWEH&M;Sfw<J&z7U?sNki+q8Gqc!*enz`SaYX
zr$CdrW#vkZ?_Ns0m!;IGt8v?T*zPjg_mD0@4V)D;ST)rJm>>S)0yNSZ@@9_JH*Aw+
zKv{t~*0iTXa3{NDo1&2KyP4k;o?nIhxARS@+-*d<^8s6L7Wc(=k-c#V8y;fpB+OT-
z&G!QzkYQ6wu6C?pAxwb}@xVhgS#hz_d=l5cN}Hge+?0_ZHg0=$ID-Kgl8&P%-p$tA
z*6i<(APS|LHA`dUu!QYw!Ef{%{x4aXKc9s@U)B(MN(%Q82S0{}8hEThU;rEabawd2
zZL*#g9Ex}_z-yg=fDY<WKVWP;5RyiF*BB4e{4bj&NR$BcFM)u)PQQL*6Sc3_qgp02
zlyl^J*G!(1v8KPeu$2uXYg6;?Kh4Kzh&-iFhsVF&ori;|B<&Yjtx6S_9}hj+2W;eG
z<k@zJuN;-*Bc&BqwHu#>=Nu0kNKPMv_sw0q(Ec3jhG{hCkdXKVqvKJx85ziwICb{{
zw(5C9f3y%Luh%A@GS<aQIZK#r^lLm8uNJ@Wwl-7Ox22aMQp|cY!{=&7c5DY7)&pE?
zEGS7tf{kE%8oj5czrzmA$+?$qdD>Z5AiiHd)2{|5PpXX(TwZo)rC7zxtr3t+IKgk+
zPigpq+}pzf`7=fq*oXXKLMP=9_X%<>6X;PF0{{9ZF|_qPFGrQ_pEKW`L65fkGr4GL
zrg9m|r){cM>YkGO9&kYysUzCvV<Q7<ngd^%+~<FO=fta%{s7w|d~qAp1AGpR>%`$>
zUjBU9^jjP&MI~z4(hh#yqxt^h44W%9z7?fwe21I4F|}Uhucc;+5NAmvtoU_bO1svh
z!}IR2)`s7y8vo|Q*yQ#l0U{%)UVr2ezqt}<kzyEsDd>-T;ooW{N3r!$txgqqt1X*j
z-tF5&>dk}4Q%!baghp-`alu`m2`^6|8!Qgf3HiR;8jF`eCmKYBM-K5=8Kp-r?72>)
z2S@U>Gs;q!RQ$2(g!d!#Gq5+C%-LCFh#^R(bZJM5S=aU(e)TX;W_VRCVwlOWOFjMW
zFEgD54wbql1O*XmNGJZ^_z=0wtI9YlX2)#N?5(XpXEj}gf5`OBEVg1vT?~>>5l9DH
zN3E6RJZ6uIsvxT9Q5=d(e8qs-(}fLH3}^2NZeI1~2#qN;Xk!&K(i!dqzb^H>G~m4I
z*W>r`%mcl~MXLC1T83gmfRdF3UHYiIjb>hsg~FY!Gg1z{rJC_<dunWiq<Nw0(V5&m
zw+3w2;ez%g0W~bCQJkN(z}$&Tgfy>8bL_d>KtH)Ji`|IMawH7-n_d4jX6$OCDn>AC
z&pD>u0oTt!!5r}+A?6f;CB~YAMm)g7J->%XQQglC%vk34zjefmUF^h_vmBrxlJ3rI
ztIszNkHx6<)zm55maekCZW#<wS;=F46V<&Hb>M9=FSKmH*JzF3#HmB~GD8+H_csr{
zqH@HzGW6P5%BtH>59nR)&nnV~AV#?rJ0W}oFUR|qmaGwJV`+0l@G`BlnQuoJXb>1Y
zZaVaEn4y{4y_;%x{HSy$;46^qYoy4n57_dEn2@tn-xSE8DO@4<<9m+Nyp|>x13KK*
zO84AS=YCfMtd-iEc(9eQ`?`W0IB5ZAMAVHYmS*!Un}>=xmCabO_R`N$4=Y<NN#H)4
z)H^{CZHIbJ!aB4IJ$MD+HKMe}i0}Mn?&;anQ7dpD!BQC9m@0|QY64}Z!6#fLJ%%pz
zaH!^MRHk3ce!*}}A;kN7$>fKfnu2wvoHI*J{7FrBiC;%W;FR7*P_*w*^OQr?i^lXR
z!S7L38lSM;OMh>xz;GAK<kb9|ok<JXAECj)P$9q-t~lEhLXeTY^i3!PSRAme_JvE|
zcN1(3d+m0fRlhyz2^I$s5enExmKF|iOyZcaP_?MIBofG(H`9|we<Lv;*E^v@?p|#C
zJGKu0D)<Na866vLn6pR~53JfFC=rU4lSh^1Mi|p((`#=JJX<94AN_opB{5Pj{<#$w
z4vc_Gn=&A}ArP%bN{MIOdqmFkOnv>z4eU>4r{nw=1~q<)i^a{7=(LtmQxWcGlqqXd
zz!5&lTLw!U+f-ItX3D;tqmLm9AMh{K?nSWyD-TLihjBo4ARlSVuIau~x)6v2fUVZs
z1E;*!y+0xkP)t%+F5Puz9*V*t9{qgq$sxGwWjmL+^m69+KOi??sBC4~Bzqn=-LYfR
zjk;m;+SfCBq}hH2Uo)_%q&knW8iT+x%d)sYq#z3V)!qWpSW8s;blRFM1$6G>!oHvM
z8B48lU&FK}OF#UkPkpM^=+h>TKTwE5b8|3e{HVNX=6{%Ry!(D^5g~P&fm_^PyLF}1
zo351()t-VRIhiNRa#j$`<KavPTUbl6RhPvq2iL+@{T&ynz$#9;z%J5JiU>-6`b>BP
zSOB!&1?1s)O%ys*O~B%Gq6sr5WmK7trwUHxaziWd2y@6fY^oB42yEYbHTZVCky*`0
zoJMtgG|%@2o3;b6HdLEG2h}oQP@d#}{PvGHLLt!~O`UM;5Zv<=WaLqMSe7wIWQ3?b
zM3iT;?yBZ29B+$;1Q7cd9gbx=GH4c}qk0V;HfwD9tRxH>eBS;}{E;xW+(5_!J>83u
zf(7e{15;IXHH6QiT-7d>*m{l8!-7Q<^9RD!+fMnhWdK$#wv7Wq$l|u{eRt`B)Q<^;
zz)F)h9n*#80G4fG=1bJt?b?GHy`S!VrP8?&$1RD1E&m|C8T^YNZz`2*F$B##I*LGG
z+^I$y?ljw?;`M8SQUukP8pcia3T9_`%4BWUdLN11rwF|()hr@2<WaXSZ}1wg!1y%9
zwWgocO`Fve8DO}^xw3cN!Zd+fh)xLav%Y%39Trd&=~{2JOlIMs#*$0D<vyq0_`$=B
z`<`|~U>~y9lafV|EVgL8qP?V9PrC&ib`!T@Y{U^l55A-hO$b-k**;g5NWa;%K$41w
zW`{QwF$uY%Z!glOrJU4W;Si`UddU^Lx`Xb8E>yJ_$j+@V<qybzHhqx4w;6U=ubSZ^
z*SDSjdjRUabSeGaqW+5S<OuFFI2C2Od{#D3V5P?k6*iOr<b&HCG#$iZ4PB_GvAkfe
z6VOhREeVk^v+)Qt(N%fydQ@hxm8Q%n7h?vucAf&m5Gb=yhlvAgHo_6;by?HWX3WCK
zZX?4l|CWzMnn&Yu`?wBII<)m3&J3LwfL{R~cQJ-5L~7fr&H4TYlN59M&fb?NIm_a;
z<E80Y+Y8Xl5|9^xS+##INMy@Yov!_f-oQ!{_`J~2l^%s6ONRuGW?nYvJWpmyX%Qkb
zI0srjQc(+eh@?gxoUrl&h+XCC5?RhJa5zMxdvny*2{I1)4ZnNbM@iiL`W-<2Ye%ux
zQ?Yiax|K^lAa|7phhQ&prKa{>zZ%<@TV*3gfs4B?RQGkfOpdL7q1v|ZqY}(?bZoz_
zBgNrRXy5xt^*toYmy4jS<Uw7dK=|qr%LEp^PeDNOSp7^#iO&=da6)$L(c_1L;=z%*
zS3eiR6A}7y_)2yZN$+fuwDu1=h}r4dWH9Rm3}}DMD9(t}pC&+o+NxX9blgPaqGb<&
ztRffR2Stw0v+F$4X{TpE7fF|9ATCxy#M=Ex?j*#>;JU>34wbCI3#LVl@Hb!bL>uZm
zwl}HLnXPjpD2=jyEmWh5Pye@0kq-gP$VY4Zr@nUlXdPHxb+sZxvwSzWcO$YdsICkW
zHU<SNq|i$OsL65fS}qPj{y;^!{YA}j&`A??*5XR148Z#QB7P%f<*MF&W(&93*z9#C
zU7vQ42glH{3Q?$UyeWo7jvj0><BhFczK;o}z)q{|Jm9d?pj8DlRK@?JT<!><GZO(X
zrvzM7PNN%fatI@Fwae$Tu<j1GdQf>M(Rp~#_qF)u*~4a#BW9f*xWpiOFz&^rCFPwh
z-Owl(GCe2`W(TY#r+r7D<0`OCz#f{5=z*&Srv2nru~JWF#IQav;c~-zMEGf`IKKKP
z<i=Vpdq|?<(kmIXM8#6bXxqLZ^P?QP3~-e)V@)c$*w#=uf<_)an08p(nW=b7{FOX-
z(6A3=vKZdg3S{}vN*(GpKKKVP4le?JuU-{gOHps&9f@?YD{93U3@A`@7>R`9%j=j;
zBAR^q+ipf?up=1Zmct6TPwtJ$Q)*Yh)_W7pPkXAZuCS};Bw=<j<Whl|VM^pRDD~Iz
z9QURuJ*T{kO9(3q>0*BMtCHr}x=@vvTj0-fZA>Cd<W8E{v7(lfw-#GQw4_JR`zkP}
zZ%|7;!?|XugcYY6=|%U2GtmT_)w!zpg4~2GXHuogIH_+V96lXPPYyOssezzbr98XN
zw1mSiA1v@Wq&*Le#Bnwnh&D;XZ!gY5!ua)@u9$>?YZOy3oiZ~58w5a=Jo)P+yz7zm
zxW1bC)z*13Kf~0?(&zkdnzmWHo%+E}`kK|wF=3|4{zay$7Jd;|VkE$0q3+;KIscX#
zXZJV?+}wAn3^9>@Wb7h-FS8zI*Ku<FU6mHEEel>UeT`qQRUb0Y%k8xzUgALfxC2$;
z%1^xLb0|^hD)HRwL07u}`kl|2AqdqFX5VTRIq4bY{50vA4w@7Tu*{i7*tniNV}dit
z>CX%*FO%NdPoewNDhtUGZii0wI91>)DJTeh>+ioFhl{mA3in%9K<imIogc$0-lwdB
zcbO|nojAD3OrpbuT{D{Kg~C^pEF3#<@L}V<mITC&;i@t=ZhxVB%Tb{>&aK@6e;;=>
zJDo|ZFWteyk&XvUsQrzB?!XMG2<QWi-nV}j>+HYg1{i~#ibvpzLKQs9L?s%FsqtG6
z!|xhl?q%9Vi3v_kMV4anKU`l3xf$KEFj$%=p-TTDwI!jQ<ByuF*P^iz1?6EbZKS7i
zh@$)*&MwZ2Z7|ie06UR9g?m5<@r)~%cL=am461F+yyJVw*vJ}niUg5yjDN;Z)@!MN
z100FatFL0Ki-OCH$(^j7)9hLJc;R2nYg>_}rRxJc1`E40XZiRBgI%f{4F!;+`&%x%
zm*x)}l-{zoH>KR9F!6De_M5>tpr&_1*GgvuDT#)TN>iY}#g*pktDL0t;hh?ZkISKV
zjQ{|v`wOUNfUCZi?eHIamb1)GDwsqE#5c39Ks~pvjQV&7YGui(9odvD<U;)di)rd4
zrlmv-?O@c@UEi)w@y{gl@wR4v_zjoV4^IhN4rzT8ee@5hml+dGW|RP@OuAntzqBcO
z7D&u8<oBU&8B#_~)unS90GiUwCzh6YIV>LtdGewi0`K7?9+aX=GKY&8Av#RF=gg43
za)4YB3Mes@%*$<7c|#4V$pqt<r|%BSZIO62RlZrif>rdYyx;P%fAbM8Ehyvb`yagh
zfhNVvtA$$XseqD-*dos7$iT%l`DQ*ISpXSCKb86GF;@u5in)rhuJFArkWqY(!<!tW
zAQ@Bq)+MxKk};I{kgFpZ#W=;K;MM&olVL|5fP2@*tDqe#nzx&~F5rbTjwuK#ZrV>J
zq6lXVJNWZeE(6Ls+^fgig=zdrFyr77)uK~vIc++)Dp(7NepQ^uUz8G@*=wGfr5nf_
zZ;g)(XSA-A_u>f9qu9cycV{sbp(}r7hj~sZX-4L7o~&#fs7y2AY;Kqw$Pg^mS?|A&
z{TjLTXx)lf?ctzV@^LEc{N@Jd0ai4Eb!oyR8R*qwjO8*F!S1M5ODgR728R{apRNWR
zHBD~?#~nhS$uYl4)*tMLS-FLqCS*x@^6vX99IXJ}>te|arO<2>p=)cd`*ys=YRN98
zNJik2WmLs=tAP%GDyRrah^j#@XM@f3XFN`({><ILynK+#ft9*N&zG|!qIPXM*H-}o
z0fNuhLHbPYcyg7!{%VI3$kkg20QgzA+ItOr0|&nBEZ2&dJ(IByi_i3PLp50eD`ev4
zs8uxF*9<7YpGxEdKf}dv%G-)1O$9I#%kb{e%9Hu^mYrypoM~?6_&r+m`IWFre`VM#
zvi4aih|^#8Bw@qnq$3p^Mv0hAtj2!KNF0PXbdDzC^+?v{2SeuZB)dUO70&(>B>a5x
z;xj6B$)noocDk;3(YXEkw)gJ|xP>kPhl;Apuf(;1)2>d-kge8gVQ4qu4u)IjGIHXP
zf*F>~t1VV%-vI|vI}!f5I=`E|NOkHUmoLWfv=zB!dp)XS=U$yysvBQ$duuY;HC-RD
zwoV8@-Jbr3aev2z$Y2#;Hv_R{`k=lJUB9LDHV@4`YZhVqIXb+9R;C5BShGiVdL<V(
z?d-S7u-gla_p?rvh}hd-d_(Hn0-OCFsIia3%4!D&$#*lhN{3D&9W;Hp<+VSHllx&R
zH4K;()(k#wy{Nj)=W^mN*^P0(5N5Gk3G+bi5o6(wCs?Ggg1EJQT_Ad`8(<yMg}F@{
z*Q#G&1qQZ#gJhvk&Bm9=!s~%iQ@Wfa++@${7>OIhx$)-vN)OddtZ>|klu4fj;rYK4
z<bTQh@whLDKn{Z&l>Z{V^-)~5_4rNHDHy)!%ve%HpdgncW!9<del*5XLBGa`J`YF4
zDM3=Yd=@zpw`XGpR2XQcuf~f^>#}p^Y$a#l#ds*c{3<xi{#<;JF_Os>&oP=Yj~JuW
zjE@PmOMq1}UVi6zNhGM1N_XOJM~8XW<$R3bz?`{lFX_p)8K_tPjqepJ0;QpezSEf^
z{k8hH*F%}QDC3*wRR0N4=!zGeCmTS8q0Y)fD&7`MDjjiHz=y(OEuJ%#)RP1y`7v#|
z%hb)Qs6bQHzohf~?gsi<JABHB%N*~7Bw~qrc+ueGG>aiai%Y9{t<s*y&Pzf~`Ra=;
z)1xATN@Cqx*Q~^Fd;#DC1tCoKc-~nFoO(V4(v<v_YNk@x_<30hHgZ1S{gf-`{YJgI
z_@5;$4-fPSw|DBTb%(aN9K^MoVf9sO0MV80=8d<+!ipb0fEEXKn4KC#0SxSTEz-07
zXmsgNr&&Q9NR>MF<>#$y6Q-$K7ucFlOQw8lBMXj{uv0in&DgSeRg;mN0VmNnArzdM
zo(dTT{V|M(e)_r2^~k`4#&8ve=wYqXTP}VgBqkyOeBAHi9fV{l7U@(S+LTJII9gS;
z%7n5P)ktMKR?O@L6W?{t{#>~RZc6)Uj&jKc$vN9rp*rX;>|72?bPD<g?%22t-QX)p
zglr5)ZHHpr*h}|m(*^FRW%_*jZJr2fZGO5co!?Ate$Vv5^o?F5!xjJh#B5?>pN%}m
z5^+?~o98wlxNK^7L<)0tI>&!sd)d&3w?rm|0$py@>U1>#pZd%-yQR64uG$(E{#q^g
zSuOsF?+@re0}Jz^r-MU}HSWwIwUNtJ0Y_^VoL;9g>^GMJ?hCPCIYYO*ojx@k4yr_N
z+!2vX_8ND>o2H!tOPiGh9MzZIMlC0`humbMWgU*K?VWp^)d9}Qq#<iLFh;TdJ=k5m
z)~lH1Gh(8j-u@+2=f@CvFYYoM0u;pF#tNsK6rGm!BDB`M%BN57GO$Tun!NY>IOZ5$
zxWdC#Dm8BspLItKeS{!XEf=lEQe4stm|Dx-P4^2m|Hj_HG7#5B8Xd!M4S(2}!PkYn
zTx+G{Cp?^>-65BTUvOHD6&Cwkh`{-hOi_)g)Zhge=EG?6DS@UZCx#$Gp6#^0!g<pP
zM%Hl~0GDUE8SD{TH9ZGPj95?~0*A81AZKCx_4ND)cd@cn0OY2%>H(9*lkv@K-IN9J
z%KFZ>;V-rxXffTCIg=W#&~t~)>hBhYauW~u7*?wu|MNh{&=55wvXo*wl_OY830p2a
zo|1nETZbsEv;w{R8_NUV9Ll*o1OJ`Q_E}^WYHUsr&z0b<(u*d&46#)=zZl5UN*)r=
zW1~H<akoKdHpJtrGWxmLeR?yWDe6?6t2&r`C1W_oVdA;)FLDyXXfMo@JHY%8=IP@Z
z6h`;zMYlDA`#A58A6?>V@1TXK@9f-roLq&y;9i9OcBI?(@KAX-^~Lvk*oqJ1&iiBt
zGrgeBn+Oz)nci1!qVt<J96J+lan_lodC2PQEXrS<7Ez5=4IT_Gn=G*84%n?I$auxe
zJA2OxeqSgsSEc=^QnOss{CXX*S$RD^Ag+KtGm48m!<p6MZ1;G?$i3G2aT*X?eqa2n
zc{w+kU$X})l$XSVKHC(un_rL_{o=gkYtn6{+ML!wRsa@or83(+ET#n(V9Y#^g04Me
zL<~5tX2GToMid+@1G6l9qyK0Mwzfv95r*L9=>|XE<UKdNhsWO>!3EmHv_I&OEToBx
zV6?=r(EZhlLF%5{?32Xpg3Lv_zXfyGM#2*wj$AARiyyIrf&)iQWfD?9SVd$7Q>w^?
zyzF=b*C%0<rh5D(iUw5Zw!w*K1n$tnU~c?Roydna@vm?}X1B2FazQa=5yy#mty|Re
zQ#<&JyhDTN`d0HCl$*(jcW7}l^fl+{tpcby4j(8>_Y$mnx67Zq{?_bO?3}{3yBi)O
z@Cmf@EAwy|fR}fkgg(qdn)Dg2*HCged--Qg-rf^ewq(@O*9}~*Xsgckl<)VqA)2kr
zQcND~M7xmz9BXCO9|h{#iGlBnzh4d;-jw@T5=J;4Vv0XsW2~A#H0XS$I+-8s!0I5q
z*kjk`w$M@;<5oW`jJ^OZPyc_Pt1|8olw{e9v7Juy4^cn1{-gXnfg4K5m9@Nuf7SOf
z(>Myl0iH_9U^G7ld2DStZ@EFbY#CoFEZepZRr&uV9OAts1TFfD%D%{4TJKOK!DEQ?
zm1|XR3ceAg4%IpF*S?czRyKW_vp`DIHd$NxTP-T@D23EJ(Mw_xaUl-jSKD6-o+MS<
zuQ|Nkvq>eI;f*r*79dPUmvB95;2^&kb2;^#E}&o8cJr9X*g1OK$7=~{>hVCL%N=%>
zbKUF9iBt`sIb^TknTZZ<kwGq^j22}Xw+c1$%uZ1%8m_JR%rq09>24TClY9V)0-P6p
z?9Yixw!LC5UQeD^1npkvehsr%Udw-+W`|z0Cy{sSVX^IbR`XjL_BU_T<Zo{fb@H0Y
z)L+K!>(ZG9a4Y4sR@}d$b#&OVq|o{CC$o!D?u<u5Nesj-CW8q;N}!TkB;=nsk5$K?
z&zYJWXQq)1shl_xGp^4h>KXf+sHNDG`l`9d#4?+PSPy~LEGej8+GC-PD~O4e_L<ja
zQ~dz?zQfYw&GODM8PZ!M^ef+D-D6LO)?|ODtL+VZ@`Cd*P;lQ7!;-cTC6N2dDm4Wh
z5(TVeCjR=unhH;VHv}^@iY3A$Ip+zu+++MMi`}X&`$@Xak3X+Wi~-UZJX&EJs(Q7&
zP9sas3|J<XkU)WKPc02ECI6GJR_{^lDe)4KDx-+n72eWuu%ZKM&)$OK*98$lV1>D2
z?rNhA$Hz%)Cv_QBsLLe%4`V}ZUSV8*hB`upR9QUL-Z=B-fx56ixg^^V?vF9W%(*&7
z+mLxbwv3JK9nG{QpbYPY(HsqEaQthWC|J3(Qz3fsZ+!-0@bBY?trmZEc~91XJJZ^#
zWWPGYfB8OxQ<2i3S%|#+b~+(AdUp~0L@kAtZqde(`LR;b841s8Al3W`@)t;+=yoyY
z2pIgec~w~gEZywgcSlI2YvgrvcBY8S=IMS&j`uo9Wg`1nJ=u{0NF}Cof4q-6XkoiO
zZ-kI}gQ+I^I5+R^V0K|k!kFxf%HGoF|AxPBk{^+@#F~eQx4w`fvY2?^{Eo7Cn?E16
z&vQB)-+1C_)mJ`+QkcfK1a33+90Lw<WDP#na1O_xm530<YE^=D|KV+M&P=O;@f)8z
zpw4Lf;s$f6K&S7YPZQKr)hl^6gtKv)q88DS`!Y}K%x4(K`rC;xr5!@Cu+4IBZ(vaQ
znO7rYA+pHy<$|6q9XR##>dA%*t6?A1iyulJ;wb%j<yA(NGHT;(K$J<3H;h-eL~gcd
zyc3S8E?MYsO57=nVUWy-2gV-c0kDw~#Odsh2^vv^ehU%yg`|wxG4odKR|CMJ6;k+}
zUs*I}Dfra&*ztXJO|tIlzV{Q%7qS&TuICp@1NASyX2%_|3JTf{8}WNt<_rFF%p`9&
z8XI|!tffiN@$-^{WWjJULx&tc5K(5fx_8q1^oA3bbrv#4PG!x;TdMNpz{q_NtYCB3
z;az&{%z!qSOO}B?na0oKC+Zf~(_3yL^Qwe`+l<4V(=dZ6N>h5Z4LsP;CAxX@oGve-
za=H;=&`8@9PI;ybDu6J4jT5pFFpOTR(~5#f;Kcn=+h)Dx-94O~Q-g{DzCBnDch-6f
z@0Ny82lBL4s{gD==XLt>xP+(<^ao5GEJx310Fe9RSPUzlSm`kTf#J|qVP#vXfH!B-
zobC^&srxsU_?f^4JDT@z0Acd2%+7*irH&5ZQ^D|L>HD9l%1&dIBCXI|9}YzgGUjIi
z0t_g8S=A=vvvuY#$jAB+yQPHmN0_p-mMO!$=J^&L0wy3FBtdq_&}ADjx)s0u%#;8}
zMYTIDcZ^TEXr=cXS%`=Rp=M8xcVXe~!2aX(j@}j<N>AAVLJkUmet_<)no0MeOSNA8
zF{~JulL%c;(RECL<@_crsa#~zvcEKM|6vbV5b8%Iw8_dQ4$g8pCib37@@rQN;cvQ!
zu54=wVM#OhA}^KafR_om&n0!7js;4u$p91sS>!=>xjFCm@>nMV8}Ga1b5e46_RVc!
zYoAKU>3`@H9yfhM{G7c_u9lR#EHsdp1d;w*5$?VnqxQWV3PF1N`jamUY7LFPXNhfv
zx(d#aGvB@t&D}*Zw6z;{*0QYMI4TVI5pxH=MpQm6>9AkNDv=BeIS(QGJtzRL9<cOW
zp22@?_{I#b8}X%f@Bet12TN;Ulw2lt90md<Xi2RqSiq79y=2k<udc5SYU_*s4DJPr
z7cJJ}9$bqQr?^ujNO8Afr4)CkXj|MJf;$u~?!jGyTiMs|?9TjlXZDZGo6I}+-aJWi
z-Z>u`FaFT+kQ4^<QwSONJIL(cbN`deKfnqrWwHrZLTIBK;(*;-$g%7i>a{^|{09*o
zM~hj1VG=xMbCqS1b?4+PCi^^-)jH}=Q24>h`uEEHD_KShNOfesdO<lk(vv%yANp2N
zl$<vD#_CnnyMwC4-lYsS?T;n@YXtU^Esc?~l(Y4d@^<0Dp+c%gYW0hg;%RSLwsrse
zh3M))!~CcL5xQN@m`f_w>7(FumP0fzTN8{L`BT#?Hyye%5>>v);+smEMn*C?3OgjB
z<lp44ebvnm<;`wXImO0?=2h9!cys}bfy$gVGnFJR)IR0X$eExMckDb2rw`1mN+9bG
zibbV>IBh9oQ=L_G45N#(SnPux3CF9k!Hvl;)zq!Ndhg@{?7SuU4S|Hp_b&6>Ltvdl
zOyKAG&>)bkC7LC9I!E6)hO#bPTKX#)PV$nd0~?*dO`EBV?*qOOhxA>%*L6zJ`3Mi)
z&|~en()t2N#{=$oX250Csdh+f4u+Q@Ms1Va{`x?|-CggU9H=QRJ#8!DV`#jSBxtSp
zOi4@--)WodnC2b4F&ns#B9}_V8(bWomP(jgE1USZm>BUGRVgXPN90K`b=oHRc2M6_
zoKMX^Q$MgHbdsmcRlY!5Y0In)kFrDL?TT)I?(OKNU@ICQ`z28R*LeuQm0yF@bB}n~
zD9_}fU<z#DFh{P-e$~|GFZ1jF89sWEL3>9B;r$fh)tq&yd@o9bR7oZ&pPx(q?NjMp
z<rb5o)uW57(H3>ans*K4!z=aXhO4O7&GY(}O2!|UFKmbo{Z>Lp$knCP!y`F+jwr#E
zf<A4G^LrL%!4c&v=F}HWMj7%Bf?N|9luY#iU}A~PTn?ja{l?lh=DW^AQGysk;3iD1
z-wf)))K?^K#&bYdiq1=e9Eqp98Q&ftl9dSag*hKVhhr^Gd<QTcEF3yhPO2Y^u?_<O
z?N$GB=Xk&@4+%T}wxQw=i-XCGdj{SYr5kF>h?e?R3VFQ+N?DKwhzuTkbcw<j-bewv
z?>s6fQM8<vbGY$|$a`T?8$zybs&jfP?0H#`N=T@)bMK_kLox4f5V4)8Q0=zbdDLIB
zi((a0GiB4jd<3I7uYSHQLHw_7^$aQ#0INf?h(@SLGB0wGzpg#b#A&qW=9etM&$(i{
z3c@_nkSDlp11C|MJLWUgT|LJH`M0QZG5+|CEcZL|>FBhUVQ7Q33FlWuuZvAYo3`oj
zDgxIsM}*r~>jlj<6$0b2`zDF9-i6x(S5AlY`&(@Lw;}P;4cqI=Jp%{?WO?jehoSvE
zZwhNsD%V)sEi8{I>k-Dnk{46BT3_^caXT>4df53|2PeSwyB6x%gV^Nmjg1@M>idlV
zrHxJw!_~0-;}7mT@N1a1mY1;if~*Or=E_M>FG(XVNrs)d$1f6++VPfz2$tf(<Cg%H
z;*bAK>rM&G$56j!9>Z$mGGR`OK7A;_$Ja4a@L>qyH~CV_wt45zJoXKfcVFe#>7^%v
zjmxQHY*+qO6Iv_K5m{0qSYYzu*8MCkn?X8~Hu3E8rI-)Ix^rVBQU0L&JODUmZx`uK
z)Cs;Uur?tk<mu}A)TsjEtQ-q*4XCp8eW!I+3v?_h;5Z_M!`8bWHx$J?JU8<>+_8D4
zK3OK0`aY>tmA|R9;oEc+^!CVOJbF&Hv^DKRSCi&3UYPZ)e%betv#VLZ@#l@#O5E<t
zz(uEzMc}PwRs8sw)@&jIyy-(gUrn7ENCDugrJmR@CpX_;r|)6XgdsgHPehO)l7VI`
zY~cgizm-kAV#PNZd4#9~>I34pkJmT$`@?~%NrEB?YMh6^mE9;&1BR=WNtJix0Kdx1
z6+z`c9Zg0*J~m3rtDDJ<&u3n(a6Rw{I;=gFQbKl^B#F$0$cSoBj(rFnqS00|JpL=c
z0IlLfg>V!j(&6nZ|HC37Q=&L5&l^)Eh!F@xNhmKZsX0KG!d%NAFHZQw(v|5?3FXfq
zO>OZd`rJJ#6}L4`Gn{mraHd)B*U?OiyF3|POTX)qbW;eZp08}YZUUZdwE%@$Px>{g
zw>PtZTNec8b^mR*@r5yFpfjp>QOw%hV4p0AUMCSfqr@o)xn~Dz{V{=DKqrp2aCpji
z8QhSBUgP>!E1sRSuz#{ScTK5bCd9D>%XfoUlADk#Y)+YyN4o@j@bCRXZJk_2i^ThU
z#^{@PQqRN-9s>%SA-Km-HPK@<i`9yF>1&^swM3jl!4>o<D@^KzS#seJE@P!t&Xsbg
zP*XkMjMYMKkMno;lrVs$%2LE;e;qw7`#2|ycoHLD_5q);=##PQ_MZY_J-kjzQ~^=O
zzEiHiQZo<!1w;JV%sq7ojyi|n&z*n&^2YrAuZsyq5ce%AIB5j<8^<=pt2G;Z<Zi;2
z>byXSzME8>zroD?qyv!#4}I@N+DJh_7ZzyfcAHL845JTj?%RDg7<_7Gu!-P60mVMT
zrRs`~rmpZ|p-5EaEswZ~dOu47L@o|<V2$e=3gTfBp}!dcwxPQPKU1#mWIn59jEc!*
z)xhZ=|8BurQJmMwGqJ|?`Mb!ehYBKX5GN1TtPUyW+>&nLq~F_r1#88Aja1iQ&QpOS
zMb*?BV3Wc!jdgr+-^dP!?B>{qs@zpuKA*_ojqX6VJx7U3pA-n$iWnC5fyAk_KP-ho
zonJaQJL_89GRl-b+h%D+{5lV?`O#TzgTj0Y+GbO(C=@uvhIy6-&Lbbdo7t%kT-lR~
zIc&m7dZL;6Hrowm#)2S*<d=^63nC)5)dyCYTz(r1fR$0+xtGH?AdgWPGO65<+=Y3c
zspSX%(o-o+vgsj3kNcJ<i%@tJ`LDrp%b%ASr3agqc|3#m(X(X?66(rK@FNjGUYF%s
zb)i4hYbLEkF7ZQWp=!t0w59m?%#W+9>Csx}Js2@Nt-1ut_^%V0tSgXQhfCTBxe=^b
zOEyW`&fj{>phTY77GZxL+9eL&4oLgGm+WW0@Z7xHm#uo=c9I_2^w7#!?YiCO%|=Ul
z7S=kzmK}>K+M_@e3?dCwGTw#$;(xxltE?x-B#G(eksJ%e?6~oK!)8NW<=Ak<*<e)a
zpDQ^DeZh|wQueHpVcu}s6Iy_A=y-JTDc)C>Ab`WpTs$0a@v2B3e5aY1^qW|ZOumon
zk8rv^eUoNHm)h;#9aaNaJEWcDvc=V7RRI~hk2}@zK;EC%fRk67@Nv==kEMdrR_WK<
zp<CAec{IaUCN`)$iydXQZ^@lmqJy;@WAJ}rsZom3;ogaMzx|oD{zuaz<?ikoPlq8W
zEa{dXs$Cf0(nc&~mB`xnUzM?|W4#R1DKU*{nE8((<%zQ`t)55r4prmr8XY+*(lt(>
zosk<2!qN4393<Y$=@>Lly!Q^7nD<Eww-R8FEOBm0=4_TL8!)MZ75csy+vjKp!Lbz#
zw3*OA^2o)6553=sVS<(U2R@CAW)^oIBbLp@nZ_OMuD2J}79Q(epWK&j69v&I9(MTU
z2o;!bml5Bv*_*HkbZ9tf!>Ngf0{*rF-Hp2J_1<lTAUUa3Y$U2TKisM-(&rBk<1jdr
zH+JXOGhA33GbfOFFHCBWa2uO#i*re-mIOKj9?pnox#H58Uk88hCzq{@4P_qtPUJZ-
zh%KXrZTLzZF#S2fy)(SAef6Et;yc^>Rng2xUHjg~1ZT@wIjz>TBnCo`7&M(ep~#-i
zef}86+K%N`t34K5Ph1|BuU+l4=<~M5O{JRP!lZeg#(aGsk-Gmq+v3(Doo4}Aft_!;
zOF+;ic(;U`#T|GF{&bWFUbPQ5$JL4p{{wW)4aF~T7ZGATb#v0kWEeeU(k^1Oz|pF#
zXgo${PtGIGrN8J_2fr#q{Eb$n!?RRi>@unqvOEK@2F4TAn?F-YMr;+N6xmXQeo*rg
z$udVtlz55obLerMZ~;=LDI~MaA?7?biltxryyfA%VqsgbjX+AFYEB<7nl$-hc{U%J
zAe_E3>rEuSG;LPwoD7n!l)q|)=5rhueXS4xftdqtZ5}dos??^#v}X6iojGVs87sGw
zHod@ggLT=SM^=%Fv~)K^nt=4;2bXbtr01Lgp&8dDxm58E`@K`qpZlTIdUG1)!qe!9
zBYIyomHFy3l3(|J{OCd&RF@qO2<ffvnTY&6K=W_3wLK>*+ImHyxbrD{W&SGp4fs;W
zMvC;>xf&DOpd33Kec=msp$l}DKO+OML9}%0D)(f;-{0>apxtBqQ*d9B3->m3{!WP_
z)>u%l6?)+G@v_o_D|@Z`IeW^Ez0&h`S;S$CGC<UgtGW`%-~|L_awz|Ss18Axkdp`$
zoWG-Dse9AC7WlunqMaso+nM$80d-Ykr;1KVyt-JNA0yp;BgzV`5B-JKJ%EFn1(r(%
ztrTE~SpQLW<rA%X)ub1$KduCAX(D^%Cvg0bMhMLPJp;I`!MUK@QUOs*tak^j<L_ws
z?$X3TS!vM<x5wVTIk--qcw}y(Oe#>QU373<wB*iV?CeY1+d9Dldv!^y`{lc^@wWS4
zed6SPL{#tePvnPv`gP++I1T*Z=h6j!huQ1zPTU>zbI8n<*c_s7XFahEiB%!6A2u7-
zv5Kx2W^tu58)X~2nl4dGpI5Ng<i?9_tMF}Fgwu~g!xUr6cYy~*_+;izNr#tchB3H9
zBz*=}<+FQs<=Hw<z%|UqwbZ0^K=^?SevJAWwfiWw0oNa1b1G^IF(a9S2hGsaf(Ex4
z2;*%xc!AE@j;A)VVMpIrmh{uuwIqV905lYU4D)tvTfU|J&XYygk{s`Q%$f=g&~5}c
zXFf<sVLNr2ur~s#otDT+5zD~s4D-7=$R>cfUU<-=%426Dz(MNBWD)jq7@(GiZG_E7
zqER4giS1xA;cRtCrRk9Kg<po$kC*&-+E%Nh>%$~)zK;mh9Q*j8LjVx9<I;{w4>n``
zohj5&!?EY#*Iv+llh7ag{whZPTwhAp0I98@wj#{Spgy1~kGpATGs7Z;S<ihrL)Z7H
zJLLg#awwOS_o-o{FL5y3A&%LcKzahmW8a!9gRsX)`_1;v69k`vz2D`rzg*<3Y-S9_
zAaM$RL<n4qil~akn4*3CY__!9Y*StPNl_m;h6|{*V){6}a}U{-9Ag{+@3;e+mUo^p
zZ(Mb73)Wa_Tb-Bkkn2rPJ!A;S57n{zCkKpuq?R6en`puNarNSbY)GB=vIMk=Y|~10
zzNSUkJ`bKZF&;%`7|P%L=!2B{%p*=*e4>3VNK)ItdY-Lfx+#8f5ooV^7C!w46YO|!
zVg_*<xPX{wM{Lpmwh0|AeT0PQQ<b`{S|9Gk1{|9g@;8V#G1(-mz6SlyE*6l{!z{o+
zkX$SE>`NNxBbU>-JGFqr1rXO<BUeN0`2sM!Ll97ne`hb{3=O8_+lU2J)pXC<J*^(U
zB+}c-5?p~Dp)s(0k(VFxh=6U}gyG5sXV+(*Bl+~hCj}dJ(=5sXAzX#jPaVX4#oXGs
zC{hIZF>gWE9<DQ|q*Xd?BC(1^C77qfQfJ4OEDgbnE5pO6$yN&)6+;Qhkzt%Ezu%P%
zHOlW`@+J&^c*JW^i_9!vAs9HNUeK<t^*qv`J6ECvF#{^x`8=g&DI(t?_6qbr9C`V^
za}n2!W7R&bS_aWwJRduV$zuNk`5cDxOaZXG)&}qfOaemjI(w)^_`YBKaisr*A9GC6
zb}EzF0F=JWn5=*oXlG0%o-*b~PqS9Ytyt(6b$s~8g$4Fx#_uf4*K@8k|3iU-m8Z1$
zz;R%-w03!H_CeWxB?FO2tPu^8J|u^jQpTyx=u7ZSa!khIWjCwiPJbFe121Q3zHwC%
z;_akV+%<UfRGyB!{T~dtHAT)cqx<DMKa6CD%F^5Txls)Q1JYb_%$1&S4dcaC&#Y8(
zp60a(EC(V9c9a~0Sv<va?*3_UlRZ?DOrHVnE1-Z*eHlYPK7NSG+cj1D>W6s1RK&M8
z$?DoVC>iXf)9NV&3H5JkI^zg2@0Uj{7_FN4@Rhft?zB^}YNJq*I*0lD+RD@khOiTt
zw<hDfPWO**H_eyk8Z^-=E}?e?#bBupIrCo%+X7|he>ep%l9o>2Nrx3;iBt~Dx0E@&
zw}1-?pF-tw8NsA6;Za%-H=&sMzoeee_MrU-YI4mX5kXgMKw04MPdNgr#~@e6=aOfR
zH&s!&^j)K*7iwfvgG9fqRBW&GOmU9-R)*T%tL)oG+(D%2>&D~ej^R7>5B*LezQvjn
zyn7#-zqB(M1m^v!>GR7-*`|maB(&uGCXueoUHmpPwd$n=D+v!mE;ku1h(38^*lMoA
zwj9gEgz`F{5eL;gtk9JW@Yt*HT}xg~Y+k#(j1l1cU>V9%e6}e(M^ERk8pQL>u1n&^
zX(KDZ+hy0l9pgr^g#a_KZsspR!7|r_ms1zh-o@>dpttg>8dc<1uvX|SzqB8qw`;`k
z5|UuxK)pEtj{e9SUonY8^fqT}XYtHT=4@dnjnaCh4}Qn`rb_y&vTuJWs7{THAUR^k
z;W%7N_v=i`NZ-+%FOmFf4v3v%-#}vX-NzB5{-m~ZCI8j??7FP!l*_a!kMaK=hqdQf
zNcEh8uu`;YlWTwAj+NmC%pGso)aWs}z4HNX=GDL5DJN(^3N-6_Vj`?I+DBfrP1pFx
z-P7SiPKPT$@~(nYMZ4#m$1TS*kIL<1R#7xEZKLR%H@xC+bqioA(jltDxy-J-1$0pX
zRl0-**rIjucK2e=Rn2seq*MQ?cQNj`-H{{bFMWp-LN)1nVINr?|I-*8-5GqeDFq{P
z1?M2N_lU(&oV6qYUe(uKwY!yKJ}OVCUyOA+R#x&cPMXhd_3S~l=*~=Ou*cO14xW|B
zOwl?Abl?6OE#M)Y;MhSGl}3Y84gZYYZ>`h{qhDb23l@)aD1uzsEjLmptj>k8en$H-
zfy?5J2VXAp*1ltm&+Oa$vh2c8x4kM>4PH}2cAT1z09xS=mVMZrXdoMd>9cZCC{XVO
zI4OL-j?v1(ip|2GuDBciB)xj|X4wm-{dl+UFw%Z6Y53f+l-IVKbGxbG>ojam*|c79
zb$bn)3k($t^)<QwJOefQNxjTN7~rVF;p0Uo3DbDP`iZOvD`iUo4+&uTob}^N`(gX_
zoNF7U>gxr5FLmVgZdaLzd0I^DMQQYBDN~wYsu?bxR0ND&pNk@q+bJ=8-qd57N4?Hw
z9|4%sFx8WQ{L_&{E1f1P7#KV>d)3a*YRYyMf1JtFM;bRoenl=M*AXyq(E@AFW!q@E
ztInct|6Ptu%+eM1or3hHby*HnC#@Y|HU1vXO1nWrv*j8+Bi9Bi!M>9%o?q`XR>ViP
z{=#!UJmBlp-^~o>&8`4gxfL$9^{;2r$)SnuOD20EYnfb#oA+YT-=<y_u{B#@11(Y5
z^Tlp;spn^Hc1PCgfjI;T+Qjcu;t556_fEBgvBuy~(@H-9NOT_^<jGv-SmF!M>TUQy
zGM*thsbba(TGfSl>ntR5*HqrS`hQ=K!#LCN`wd1tPe9T$xfu|Yw8m8b7<K+fQ|z*_
zPzc~uzCG_#0&F3lwq0!(0A8%?-Iu@Q9Rs4P`qbeZ*&lc`31(|HZCvbEMGdEfi7Uw{
zlzoEzVN0R?8YBQTmj=IUbCSFE6^o;wrNbtFtUc}vK*G^MMivhgRs5SNt>WdF#N|1x
zykO<Y+-vQSq$y{S8@#!o?8{EBy4#UFL>3hk<VthoDK6B)bbrMYlQFaKYQ0V14%N`>
zFYWSqUu~I$ojYf~q8%~6Q`cLp9l%^c^ts++cWrL6S&GOU+aZ;-W$TwWwTQRFC6!hg
zu1*9)?Li5~Sg#)J^N6n6Vb^%2Nb^Se@g|OH<>_a0?$T$Pc&0G8ay|RYFr5$)YH|C8
zs9j?4J+$)g70>;6KrKdpB(-G&e3YKoc_K(>M;K#@n?As0yuSdzK|JYl>*&$fv3f2)
z?u=837M|$T0vZu@@Bef~84SI$(|=#2%bAMb$$L!jY2l6P093p}%JMRF2Df#7s?PtW
zb3jAyDWbI^&8V1YX_BiX`c~*O)%bY7#=gB40`(_a1Ud><{b|B@vrMio(*8ujQ)|}k
z;VpMkX(EKmL!>8w+{rE2gM(_FnOD+%bBEh@UPL{ECZW!x?BtMn+1GM@bOS3ZKb3E^
zTNZUY`nJ+DM_(I~zkRWBW8HJ{ou$&=vFtPa5=Bc`3?5mC<tPTBTCFQ{$T1&9DfBft
zAQJsq%yAX-tDR7Bw{;s+<Q3clkYv3ZTT-!m8aByZP4vBCvR9Jt!*ROqWbsQCt!}um
zd>21nbA{5nXzlBFsbiXRc<q@v*}e|#s7&to48?sC?+AQ484|;6X2@Qh2x-g~scbf<
z`Fm}Iw3!o?4`;&;gq23u&wd4DfC;RY^+~Sx6%yJSYlW_{o)v8!ZbidRCc&4G?<T9O
zo<28faK+%w;%49Ia083OpMO=$4d>yP@~%IObe?w>+$N|yJz;H`NNQ6d{Ato9($h_A
z2j6_bc|ScVo-Ymyb35SMze8s)+2*T!Z|VGg7T++_8Wc&NKQt^xm;kQGk4t?6y!J%_
zu$&E^Pr(AGY0bBb1|&@*f9)m}&V+gsWPI28OR;XRD;?iyKtSXE&!ZLfM*FkTS;21{
z4=CE%v-tHT6IB=fYfF?ZI<jYcco?in(eUJWyd`mdZ1oXoU<-<&&M^@Z0OTqGNMNO)
z9epTfl$^iobN@{*cL+xm(Ctt-9_8R2!^7}dzk&LG{YWnVA%Sa0L;9iy#m|y8odCrs
zNCGdD?&akr;LgRL={Is@kk5`rGVR;653V(|u@geEe~-~6s1dm9iR&nCqp^Dij|glO
z+Vac+XAe^43?3w2lXQ`mbhrC=%(X!QzqnRS1;<hx`@50vsbT<&n&L?I<6%uMY2a6j
zJu)WboRtZ8d%5R%D;_~4sz%b5+umks0f_OosCqK`$7(_QKg{n$oyk6*<&W|3AHB+4
z(Iv{*o$XCn_ws=-TAj(4&h31OMHUGee6~H+H!mxxIBUEXyy3gyay=b^7-Os9ks1e`
z70}2^DsHDY6yQ#Qd*c>+b@`N+rYC+WThH{?HGE1WmM-P9zel2<jtH(pssBBU26yYW
z{YbhqHsY7<o9$cqx9R4Er<XNy2CwsMrK`h9Y+~E&>cbMu_|PDJ?ZohpSR98?8`hiH
z<)@iDI+zB-Z#ARS{ttdQ47}=<Z2??9C2v~J2N&9Xe@oD>5aKF7{H+;UQDE*u&bOVz
z;SE>R8Ds?^;)6XXa#j|y^emtWiwV{H>LuyLcFqZ#+QryLsISskLh*1UJ7VHsRtM0O
z?l`HJ+Um~Tb<1gQ7+ea3EhN&(h88ncG3W>m+ncRNzzjrb5~}tOd$XEjnS~hm<&ocZ
zcCim<qfF77yMs-w)dK{!G_I;Ad}W4Lk{g#@Jx9qJ4;3{Msr|Z1v8j-u56Z_g!MoJW
zX<Y3yIV>tVS`q#uQPB6CL@)L@RLGZ_{Q5?A8p8%nEq(4!<!ozZT7L4IkE9$-p55{W
zN%ML|p$FK8&0IFpn->;Dt7PvD8`K*W8!p99iT(sR%dNVl>?)&)Dd*;arx$cdeC$@T
zjsubvh<X_!bO2c~#_Ij(O&+n6v1R<T{c~M<0bvt2htu&49~3VmxZ-{rW3KNI^sMX0
z_TNGdE00mqrFp(QzIOKyzTM@&EDUWO@Pp;?p6|-1^8S|Ht&CmlJ55xgB-F2mXKP0K
z^7NDHiK-Df2CvVdB%BX=BN{70qEsZSNR-L0u2*XHUtpTC?~<?>F5vG%*|g%^$2<J@
zhDj#-;^TA0@v+c$aJNRAn|~O&)|aQDhs_{IdC}jhs4m{7?l)Q1fLR9`8HL)bbB{S;
zy^iD!<Apd=gFPN};4~>bLhgsPh+@@_y4~sZ_Y+M^Rm=5$3uaLWU51l37ld@lY={dR
z??H5->>iL0e9@Y;uWCLawil9{<;B!1_Ma}pC10=<ehbZv8P@r(?hn70D}2aIE4)QO
zDH|eW#okg5d=F5}-&-1vkVb#qV5ZeaL8YHc`4WnRZ*-}B-{UV^%IhnqX`a1E@%BVm
zL6xDq>s*~!N1Ns8W$<vI+$X3uV%5vf1^gz5@>fN8yoFXDIcL;^N|@Q3F>g8Jxa>sD
z=M}@NGzk!?E;w*xsK2FAwr5&4slZ7<DWlfEpDhgzWIEVSoJB%rzwO1On;&9<G-OJ)
zQOy;DUcaY9QJXlQ{<Neaju2Se)+{-PQ;yN@LVyCAN_o5ZGH|;?<XSCdsc27MC!2)~
z^-Ea3`g10KzP)F?BhR<C<@4!Sit%@2ZY3?knoi%>FlA{ws|>kWJ&6PLBs(gUhV^jQ
zN<dYJ6GjYrZ`BOA%T7$Lw+H|QVAzbeSQ}WU?sSy!=q)-TjRlUdjqVtoZ>(pf|HZ^Y
zaL1DW?O^E2EOm#+(ywgMP$(7peZ3=z(#@({SBcihmO=;snmC#2^%+iXiUa$+01$Ti
zcrrw;iKc>ADVsrI6L?sHIQ4uS{IelB5BCRFH)V?UG@Ldiyw}KqC>W@ct_&$l75B|H
zOYcdJ)a?jRPvNDz!l1O6!kp%`EX7X4*XSryVuEq}LV?md8HvzL+d*VvY33}pepHv)
z%`a~)TVC#}GngV;rs!0KX(nK~F#K%cfMKq+C#VT_xwTnAQIBp0=pT(CDa%TMT2+w5
z{Ss<uf>i+)F)F6^_O#2)<4a4m*{UAB9%HG_my*ciiG#!=XnaP(kI%U#0(Lzhr+KlC
z{^Sr{W?E$lV*AtRhg8*-htC7mspOmMWYB`ABPkOH&4AUfEM)IaUYA<$NwbD=DLW%S
z-50dGnQq>ssEW1T4phF1AVaPI{33`1;*=)7LK`lTd0fYm($CAoMK?1xZ6(f7U3LJx
z+U!$=rUFj#0cg?SP%hS1*$(eZzfZQ1sqTra-0N}&rGg*5h}^uzNsD)qs&l{flyS)$
z0G&)YagIdZBQ<kyXP0se%B515in3j_W7fpWoF0a(^zOc(?r(l?*dKk*mQY@A&AAU~
zsB=w`ys57UACSWRVB<ghexecf;q1#R<Y~2qgz>E-7kpeTEfciQ?LpA@YL-|jRdzS|
z3!4Y7sp?%7JVn-OMQI3c8PdPXdT98$9qtv}9{cBjV?kaCs@Vtf&l`IS?F2MZ`4<r;
zBvQ<lMg&`4E)#wXq9+N)u_h$l$r0&Y!<Ny;_&8_*pIQ!HnpG+^ne`R&glte_4-e%U
z*5=y2`U~F~I<9saWj6Zhfr<ymu)y5;Hiw)2G%&eeUMtH$=V`y_{ip-NdvW{uV$q<T
zp!gi7Q&9DQB=+aLxJ%;z{yM4fmwLxsw%7Lj-x(~%&#g~c1m8O*D4ruWheGcsWe3mc
zYh=*A?MEp}f7z5mfQ{jQfbF=g#ups7w^*7!R8u|(ndtK7UoB+2IbWwo>Rs+C&sRtb
zIU-09kScE*ccP=XEl(zN{`!D%%jZ+apx5X<5H$6a>4U9atWe)#d|s;=!TY+kU)sWV
zDQr>r3EB{be)dbK?q>cj^r0q29`qTC7XeVJiLd9D!R5!Q*-{s4LU=e|H~FB~Oy@n@
zNT>*FA&CakU2wytmZY%35z(w&RR0O@*RC%<dL=YxBxQ+#1x=^Jps<&(wf`<$ADYPl
zp`GKm)O##~R5@Am+Rr1);Xwl-x!)4ZzC3g<g|=k#Q0{B#PAasrXU{@&1@ZlA_67ls
z!d$P8Z2{kJjoGFxWY;Fp(Qjwl5Y96Y2xR0Yqw8kwXm1a3a0AKO*g<Gq9o=2bAwo3P
zZf;IO9321lIXbwqd)nCA*h4IAOxYb>tT_I?<M<aYq~K`n;A-;T(aF{Z*lIc0ctBiS
zZQQ&~>>RBe*_|A$UdWoO0}C<yTgbxH4I*Uj0s%H4ZcZ*<HZCqUPCiX8ULh_%AwC`!
zPEH|C&ST2g`v2Dedq)c!OYi^xfJ}+>9bka$zZF~^E!{j#T_7NNds8cjvZ<Agxh)Nt
z9n3Ds#!vI#lAJtjoM4&{T;Tt6gr1(B>^A?Rffo?{@A!hIYI}egN&n5LY3*)rCL|>7
zXzu>+Bnb(r0imuSuEOUQj{hg-zcqz~tRXg5)@~p!L9P@GI&mO|<bTZbzhiW4EZnR?
z`~u)1ouUmOgzSHWC<7<j&D#kA`tMV50OwG=MqB0onMYX@hyro(a`ABqf(7|_pquEw
zfDnTJF^`6g4+I3}WCtrsNx9^e9s;|0{>N@Db$O8Gze~u$!Qp1_#9`-XZfa-l=;Up}
k$HNJRKsY(gc+D)>xc*(lG!SVKFdQT=qas})_3_L90>k0aG5`Po

diff --git a/docs_old/logos/meta.png b/docs_old/logos/meta.png
deleted file mode 100644
index 33e9e5888434f24ec087347b62860814245030e4..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 20763
zcmXtf19+TW*L7^Ov8^_C(pZgc+l@J~?Z&ok+qRuFw*Tq#e*ZPOCK=@1_c<GDW9=|`
zS#d--TsROA5JX7{5k(LXaAM%!xiC<`X91`6P~a~Jv!B0yf`HV<z`q+n0$&pwNhtmT
z0r8*&0RaSpfV=|V0-S(=IDZELInxIL;Z6ks!M4w6Q{V-D0A(mGE&}rP?~&74oB#qM
z&nzkOQ`v3pBGcVXW$~eV+xfCn@*5<E1Q-l$?0^K6eUnOsaARfWgBgM5@hbN;7<R@n
zH{D)ErOLWx<<g^8Qj_LzEJ`FfARI;##U!4l^(pJ&jf@kVa`*Cg@dH2d1sS>7RsG{y
z|MB0>Wi-bq^sO^V>+x)qy*J`~`S6=>Aoqf>xiJ0!azK(GUJaDG-wE{dEoh&Y^3i^X
zWsC1f$hXl$v5VfZm2bQ*Jt@qAjyW&|Oy&YP$S7c{AP)L8Lhxlo{`7(eHt!;Q2a_bU
z4lE@G-+~RkiM<kQd<8>u+?z(|Qu*V9X(HWA)mdozaiDsG!+sDA3U8CscD1Tvr_JfY
z))CbZjk)>}O1hnC?y&7p1Y+OxSEm#IK<NJt`~i*klFeREX^fNh>sJWpc3}l$`2|0=
zEtpN2%XHntem*hD-iN6v!s`Mv@<UJ*gn|+L1J6w^C)XDx$;Ardza<dRi8qf2kbPxI
ze?YGI+-)<^6um$kfC@QsxFV-pT{dNn)^_^+27sS%N+rQ*0xZnkXcaWfqYVFTB-j9~
z;4Y(PhaN2~0tF|vs7PT4GIl+c{ckn#pY%R5AL0{=9zqM{a)>^?>j$_zcV~9;|NUDk
z$SUXXlmK=pX=dkg99B^&WM`rUS=KlXQ##r@_#vf!F>6&JDK9v8P?$R#J8MDo-dtHf
zhOKlEa^S+cr@?%rhgHzMq50@ac;GWrq5hOdZei`Ez7=Ty_s9dZsS!=0kI!nD(aN4=
z9Q8NLSTnB6fG<B^*!iZkti~<pLR(2sFVZm!K+2<`_REivpEA*WVZqZBwD8uU5<v(v
zULvo>dFFcel=?(gf%l{&pUyo13{E4v`Yxd8Zf}^&n#7xs_F7)5(c-twN}ivY+XJ4^
zhP|&5h8ghxn7~UIIKG_@>|K$*v-iKt00QFNz?BTK9h=3hQ%6WH0(`1l8XSG=MPB8V
znIA!V)&_ubgph#a{QM{%)mldQ`^UuQnoMG1Zv|NvV)C7bOYKCTG2#pH082jt!6#vh
zT5T5BH-hG3-@0Uk!0UjUlhZg^0vZV^?4;2Y8Yob7;$Fu-9#Q-!A9Px5V&n6?9HKrQ
z{PrCI{5xIAp)OI@PwBlIS9zaUQ5z+{v2vNPUk472q83QR@(S9_#~h|%3rgDTPv)Y-
zsT4r{s<W;7Wfc9g<0j<6BSjt}Q;&OSd`a+We*}X(DU8+($Vxr@IHc&??Qa#)v_7$K
zIiWA*)fByyc3{Jjte!bJ`tyfviL}(fUR0FH;^_zmqqqv|)k}JR0=$wfav|)x?87VD
zrz&fEjprVNjqOK$11jsOC+tTHG9eV7e<DQ1uTsYIplg(2-x?w`uCUU<(@v1^Gh*^j
z^6O3f6DCjem}Rsm0x{OxJVYXmNfh5GUnKjqG*UK+c2`@=!4QK2LX|4i+_l9!=n5FS
zS_~P$H@^fQ@tS`Yvwy_lM;O<~m<^SWB)&JiwfBrXHoxRq)SyOMyZl?k%WFtK(3iRJ
zgB6dz&I`1{<5SgSKYRj>=EQhi1tC7#xbUC$!=eo1KlspC-C=>|S6YBj3%e)n9~W<<
z0?gl`OWu)<MUa}LVUo1MhHUFRNi=?4g}x?CAkd_yl5QOe64eL(w?Z@i+g4VfNztIr
z7nHZWd+<J1Mg}|EAN()OJULLl-)$y=EEbUhqYQuSZ2wfAI0UV$C8rbk=bJCAFz=ZE
z|1~nce|_h@q9C6BV;m4Dc(v2=qE^7Ge3kf^AV>sCF8~F&RZ1StKE`dVZq!(N;?=8S
zXMy0Nds$8>m2^Y|+6)3JG1Y_sLbmt}f^>K)+JRxT)-|Vir(SIQ9@Q|T(4lNi)9W7?
zoawgb?N-g^??%Q(cenX_b!Xm_XqPnfmAEm+D&OpY%13qqF>(VRGjNPL$A@q4aU)H-
z^*iHMp+<a>a+ndiiuKMek<UCrVBC3aTqZ7vbYS^h#;rX&2BM0C1?U%2K^%r*iQb{2
z1kCsHdCEy1i)=2oxr%<I8VM+^O+zsL=yw$gtOyhMr3^oTp%E;vqO9niiql6BRHz~j
zwm*Re6bYr^hW%gYx1HbFVKI}MDBjT3!*}@n&DHacl+D@UW4}zVW$Tf3-(y4;r)tb}
zvJ=X~hojCdf+hHcLoSzu^}{v#{#j!v*-dQ;?ykE$&s5FJU)dO5S5jlC_hsVdf?m%m
z@IL_D9f}>q93=Jl<jTDF7BZHOIGb~0)wgao7V>P7t41?Wt@MA!;~1ddJ(tC7_sypE
z%K%k2e$@6{bOe-_+tOq%3peTuf6-HNl5v%%M?mm1D%dh@ASBe9lgC)uQ^*>c*JZ$r
zN#qgwwa^caD{ZvaH4US+SDV#S{(kQ#&tLm5al7u$m5MlgIB`{EjC(IOrT>mklAFpr
z@DX`SsOIjbN@RqF=wg~6K-H$}IrJj>Bp(!h5R>}J|BHC`{k3|@OvyQk&FODw2k6F6
z<$eh-qM2>LJA&ew9nhq@rCHw-Tg{hdh9g24L9MCCCEZA(Xw@OOfU2U6`vy)YiQNH)
zz?{H?L2_Z0eY6uiIbMjS)6CGP1j@LP0(Pv3LMX=SIK?p_o^sCQKq>m!ThP%zb8Wu^
zr|&-YnsOqny`*Y-nZ^g>c>8MXK4hw>t*GE{-|@f0m$(IbYaWG7nuO$UIAc#I%buFe
z37EQtA8|AFJv|_2xX`wwOuX@uad<6){J%V46<$Im<D_S=U8p(0E@;dX_@B43LoMwh
ztx04?y89+-UP_L2@{<=9{&TapgTwyW0AI!6UlBOG`M=4BnNB>8j@sFTJl|n+SOJ<{
zQD=lQ;Te;nbrbFu3K~+Zj!1u~i}%qyXA*Y@iNqH3x=fG-=+?US+D(Hmm6TOGsmA2~
zPx0c%<0=DCzG6;0^NIOuz8x?qX9*tKF7V*Zk^#^aC_(1|Uts0Tc`8YFT%Q3fZTCEt
zVeNk28_T~g^L*XJ^W5zqY*Wm8d09J{`a}guci3NOlnkj1jFHcDVuH}F9(&5m)Y5VO
zX%%{NO61-8XGnZ@@YQDYnG@QU4e$5&v?(V2MWfy~BbXp0!NcO5_M%#t*0_Ca`lGB$
zqPL~&%ncvfli6si2CM_Oyj^i6%aoGmo<*7HP9$EjoTvB$k31VD)JeOBTq-;W7#{2t
z6~Z}Q&m*GAZ#wxc4QFa_?P>q1G`cGf@o=Xv%ZZD{V;3DtE0&#`-R0)dn+_ULHiPcf
zx(nqKLU*u*))ka4MVTuI-CUiG3eiCeWI8`3Ger~LNL2LsE$rB*j8rWUHxo#iP~l=A
zKe6>o`B1ME^8SRx5#Wj(mc@~YAEi@p%elL|6B<fF13i=sfIMBVTMaZLOYEu@>mkN$
zE!1a&<sGa@ik?O;lf5F#Q#?aT>k}mq9#yqS<L$_gf3)@ral0mGPAHLz$7c%jmQmM>
zDlaZ&e(H@WNz~l!>OrB~-;jQQYIP=o&AOg;<Cv927`sbsfXu9<F7y91F+QbpYx}37
zFS{n_{dR>qU#Mqu=21mP2)x?w>1!)<cpLZ)A@RUsCj%7SXNzd-i-)A{PAxKt_ZY$2
z{yIF^phrWYn5X$BXpzFqxoai`Z}J-Epu4c9n8sZy9+XAtki^T5LXExkRga0(Pf~63
zWT1)yBRcKq4`3(}u{Ko)CBT|xu^xFGcB{`ABfxr(UW6zBF6Rn5cbw5}Ndal|<V(S8
zC;jfBFBekCp1a%GqKsAO&BQ5!W=+HecfSjeLLhVGh_M<8&asv!zP}68u?~KAF}#@}
z*9ZLH4w(msLCE7?!_*&yJ34kZ1paQ*(D<H(;@sa1&DG+?Q<4S-0JT*bBy`9(nsf4C
zBXS6tn3l*yMh<up^8;#SBDj(Pf;0kwdEg~}%He79yA-S%>?08JnQCD~K$ro&61fII
zVoy^s^FrI*&r>{7-|}XdrgHdK{}Qv!GY)kKNcp@`6nO$y;ftF{F&DiF*B+5z!|;3L
zIikW1i6d{{JwPf$#_0kEFSyR1cf0u@Fg?S%V}=c&v%&#9taxfc+2<Xeo*y*2`T{+j
zWRE)8>tuW_uFvSt%!KBW5s%SSm*iMiS0^v$C_WPd?;1GkWO;MZM4yn~sQ_~xB)jJ}
z-^P%XxmsEZERpL6IB>MtUDS54O=16GnL8MHnSoblf->ddZELn?2vKITYM0%4^Yvak
zSgBrWdGNC_j`>`DCzbpQCOX;I<g+enQla@CvAzXJ;fFskYPO*sgM(0zmTk!m6Z5W*
z0K$EoF=(Cr=S>e%sVDpk*;5oJWhDcjb35d~2j5R!X?LAH2leFp1q$@g;OwzKS*Q%D
z0gm5M25i@^{t9`x<cw5_7A6U>NPGP>kX-BqTp-{85l<3>IAr<)9_`mEUg(o+5t0Ba
zS;a>W`?$vXqxViy;<Cx4rg|s4hTI!GdcT^P&Q++Gu1G68Af7grfTuS&-}+}29TQWt
zae@;$0Y>lT(7*6p5{npeH$rlA@b0O_?3~Cragd0hF}{Nk8l+U9ticO%fSt!=6#uqg
zoGfp3RkRNszyYS-AbW&+VI-`mT60k+vK3|urG6G72PyWS(}+BPjYZ$u#EoN7RQw&M
zB4Ruo+3*SkrrBg)Vkyh|fbu>n>E5d*7xkcLfny<!xJEv5S_f7K&K>TQ3Bjd7hJ2Ug
z9CIggcX!$2_HPO*i^xgR6L>)BM;^eT2(W@uOZ@CaZ=iVGnc5v^Ub;obuA+<~0(YjV
z6z2t{YbzweIH0CH;xP7R3d5QZzV85$;*KeKtT&)R(cZqmL5Vre%pQlb)f=V&CJP(E
z|H3-5zRl0cge{_<8G6MfTiDk#O%FJBA3PyU`zOgCu#KGv%-kjq>h)N+ce!RUjPLOq
zLj~8_jzz^Xp&OHl&$&IiNYbasH+LX!mVa?~pQyz-8HC<h!pF+4tkZPfepaP5*ECoL
z_MN6}1rG(|a62^<-do}w?%VCdf5@Mny8GyfzUJIRc+kv6*R7!^=U4O^HPVhl+Q61h
zeASdTLCV(`#nB<qGhGqL9;5)GQF4j+JCz$!%B*+Y*VK|?X~%DjlS;aAjgx4{$q$-6
zE;_o+b-h_M?sp5{z{sbJiJv2q%c_50v2z$liErsO1fmPPqlOL85rZqK#vIa-FoEsL
zwhmao1~QIRLY7m{HYuT0BZYEmD`(0o+Y!@B|Jc~Q<Lx8X_YIL%0-RySdPW%g(K)0$
z*&~h==1>aaa~$r}g**1Hj|Vuah=Jy#R}c)n){GH+<(G^tdkOwrqC<EQvZO}AE-Yzq
zByhWe(X5urr2VpDxKtazKXqVfk2^)c<+5$JC$Hs@N)m#lr66}&F?;=)G!32*adMo7
zD$?~DLkhUs4bj*(D*!FTnr)T{{G4ZY9r}xqMb*~(<0@&#9Z_&5S9rO`@IgO|S})0Y
z%j0hd580<vPRKe|N|YNwkQ@QL`Tc0xE9qOF=l361lNtCbd9_w3+C-IUBLyQSt=#t5
zDawSEq))9MvAw7y9U8DVW}Jk>Ue_+R78wc&bQfxzZI)CwZUiCj0wND}M|q!0;hHJK
z8(mTS+jE?EJ=$l!L5GX$`{!GEk;O*C93#j^<v(??+}L-AvvqjQmONX#Vi{;F|9~pD
zd-!Pa`bHGq>FQa;8{B<UY`aT_wOJPzd4zm~Y=SL0iGld6%9E3xz83gnWw~VB48mpm
z$-1*5XjO(Qd{8WdccZ=JaeV`iB>M2gNOk!O4h6v9nPZpz%Y8yrCQ>`IK7Yb_ErRWa
z3y4UH{6%JN5XYmI=){*064FYW`Jxm_^C%8{$-nF@D#p`{6GRJDzMvkRR@^FY<h>T+
z{lJ&gKqM8@lT?t0m&s6~A*AL?EbT~M-&#_hLMRaJnu;z`RWvbGIOi7Eu$tS$?j&Q3
z_~e>dmohUI2+ZD}?H$I$iJtR@;obq0Tq=q9{3>pWykDQmu4f!PbCw1g57na1@#SMK
zo9e(+0|q|}td#Xf`5<%;pO(Oo!`(1IJA;|v_0*DjB%}i|tI9KN?%J8_BW<5oV>0nk
zvSeewCb3c-kB9J?AXur-4SoM4SCA=XUfIE7@aJUI!?ba?Tg7V_HzpAXvxB5?KkZNq
zayqnYw%_4lC)v$)FYT?B@{>q>Vk&3r-_Wt>U6H?MeRE?1p8U-m(`he`Z7Or(O&Ct=
z2+l#M7K$-H(+q2J4~@hr(LR67wY^hghLd}jEs5`#dtII6&6Q~?+gYCye3G%TZNAFy
z(`Zq1-knz=bD~a>K98u`?R&@bJ@K=juJ%)b$Q3srjjLV1_zf_r|D^VU2$3K?HoKe~
z`2eh+=#8z4EMlogzW~i#W5;9pkb28G5)XUS(kPmUNOf0gxUgehaplVAZe#MfO+j(4
zxTN4L(w>yr?n8FpO+xzS#^e@)OPTM*PQhZ4Zm<v&diS;fU|<c%t2^5Tiq@41&4<iB
zPDlY)tr<5iBtn(y2rEf769hVBuF_uz&j~4e9Xlxp-TEbGn8kTmUDnM_+3USg9PF2*
zuZvm8xf__O*M~$&R)1mq?^!SQ5t4Y2))5;LW(3LiiO7pAku2X#q56G8sc_YoVbgj7
zcx)yC)RnEu_`9iE>9I|t^B+d9*eR;A(KAM%eOG{9oLAa*x-Gp3sq#-Ax3_Se3KPU-
z1Ogqdf|80`S%pma48iNCY>trKo=wrh-&Wz$<u23OpmSti2sNTPY#4+lYwXpBiOt(;
zuQYBUAXZAMA#7F#n{Csd=M<D;m;KNFlDlRg!`L?u$}A0@Sj)@RkIGanMB}V&Dw2|_
zrtL)E9aFmUWSq1o1hoE`^4bxuiw;FaU?i#PU_=BR(J`au>Lquq5@ua>YAUNf*zb;<
z`7O<e)!}eC8v6iG7ZWzdy(FW6kxaj>SaJ6-FTTtZDsrRH4p<Suj7<Qm+SVj@u#olm
zY%q8zDPoUX4VNQ{Q;$gDBt(r^{+z@-#Egbti7?j+T3zLv{FP6nY@exRZa;9ImNW``
z*KdQp_lV$1TCnAQRI20DQdirKRsCpn0u!;p0k5YZA01mW#YkJuKc~4NffbZr?xIO>
zsROQ!pcu>cA<D@pu)$tm)8J99tX6af1`NF4(43qQX3taGHvKa$K%b&p5Z#zf4ynAJ
zjaNNosNI%>^AW%QxL3<y(iCojM}v_gSN^Eaz|QxOy+PtZ7^fUZ1j+=0v%hk)#rI#?
z?QUKcE+#x&H;7_U&GW3g;v^N{?T|siyQHMycC>Oh^3;w{-znW;pLEb93EPdZ8st#p
zHJJ>;JGZ5QRtoOQ*SSmoK@PsfHaph5o&T68p!@OL3>OWZKo<3JATn~jf`u@c21Z;t
ze#Et&L*^s1QM;Lg(nmy2Rm}7K_5SYYs&R85l^Pfy0ah({qzp6iMZq&QUVHg6)*psE
z>{t!6#e^UFpQKpuix{?g8n#*bPOVpwg!MKzmfd11Ra4jds+lAO-%Cpx&SL&0{A8Gz
zCmA4I9$Vi<e7>1aDpy^!z{l-x0Cj4_Ku`+~d#%}iW3Trc9!h8~)lUH@bs?gqAtZKT
z=*B-tnTVX72lUtJdv)QptNK+W`Q>yD8K*kCZOG}C>62XDnLw|A&in8cby9=R-p{@_
zUJeu$GHc9tZfOUL#mg*g4>oZZgzgEZk1Z=bsX@N7DcD&1XAX!ribJDY&kX||L%#Xo
zWsp){o-T^GmB}R(JYrf0k&V}~F$L2Tee*OPBlaEciSptz6H(^rz-Nvfdw9KEjXSS(
z*e@)x$ZvY`%8FcwEqEg=5i`Sg_ou*GjnO2=P;+=p{8&DwE@^;c9JzbOrfA>m-p&Or
znD>{g3_dOOPrb{wWFDx>I@=peR7d!cY>q3EIxfQ3FV0h8c=&b$KObaUvgC_Vlk+AP
zVQ67qj5L@QLf=O^S{zS7%JtHkiZf`^x)`HEb^{0R56io;C>HY&0;pz8`Y&YCrj^gL
zJlednd0-&K#A5XgPYVMxSy`u^pGdX}{)wO?^hqd|hh&>kFj}XE=Q}s2t3HLUVcA0j
zVr3-Sus?(^&DZZu(q@OFw~q>vuA#wOSXy8EH87tM(hi7$_+Sn#%+{%olk@G=*3}MV
zq%g<IeMn>P-}5KXlVltW{6aU^@IA7oQBhp+x|-^$e7y(;UYULIXXPTC+w?a<FmW~f
zYP`1E)5LJ+6G$U={si<b>80zH-vw4ckge-8VhPb}&YqbE8xB_;LJ=RyoinKp<yjo<
z1Jy9KL&id6;T6Whv8$hdQLYu2I88ZlMx*TK9`~Zd-VGtL)OZ#5$@9LGuHuzax!0$i
zCFa$Q)lZ-e9UG}@EFPYNvwhPu5nm!F)US`8AQuMUZ^VXBL6Zpy`Wm8i4X*j3V_@4z
zXJbH#Dn{=n9(_~-_dYN}*JB+*k@&sPCK8AH8y2INbL9?COvgdFERME9oPxp>RC3A_
z*0MrrU^fTH)O)e}o>-9|fy*~6se#Xk^NG9Fz%j%@iX`!LlE}Wfv)}TJKcuY-%Bi8T
zlT}+~>E9T@3iPf%NO~$tNvFTtoCsNcM%f2p`DyoKd6Q(+orL>WD{&p>Z!HmT_gpxj
zKG5T;0$4dFy-*6LY|Qf1*MyIa<8?^GuHE58ks>6XpIzt;$q?2ILV9tL<*>h5ES(s-
zqPMoUR9FhI0KX7ZSEdBfn1H{{&+$v~{<svK7^3YBl=Mln-p*1pdK>ay90Kazwf^T)
zs~J+r3g!~A9yZKw2<Wx}Y)EPx_@LUo^%Hn!R<p`#OP<X4k(&53Ls4z`B2Vn;D&O^u
ztMtT_l;fbiT>-H5KWuG3M1)0>#6yylvD4WcV}KonP0`j-`q|UPo#zFZ*Z{m+6ePH7
z6YM&t6M>3Jjy_;0Pc88I-hFqMBL~Mpk~q)lzFW2QU5}J@66VW6T?lrToyQTCxbhc^
zo{;;L!;em~KcS}ne1?Q6po}q*mEn@qmRZdP$wOq#;GSs|>=Awrv)lc>B*0y>dA!I5
z@$>qLwoBo$GYY=i)e@0@fUkdCcf7d^{q5U@!ftBAEZgqDn{P%I>7%(PXiYB(uJr4o
zpSX8uR(<j@7DK|Co#En?m4(6QX{j&_acb({nfqZRLI6#n(0R1K%EoZyyj{>zapn~W
ztcyw-RPW8TV&TxKedAz6iykJzuF0&9L?O--Kyq9SzS3$Lj6!S2d^WW<MO~@$9sVi~
zhU2&7T3kCG&%$@_J2g~!$;IJP|8u;+>VLPNOEUdbC3!X4*N)W;Yk@6!bscr&uw7v9
z^x#*fon*S`L3A)(<`@-<U)+a``3Pkve(iEUxc*Dz5&Wf)^qABmrK}6Y>kpPNo3^I9
zVr#Rz8q$6C;td|kN0kO?=gRZ4)h+-;-2cnyZ0Sn#w>4n&CP0vb(4P&>udCBQlw;`L
z_ylWlob$Aws?}<R)QTQhjbNd8Buc#EqBU^M;|7$_vl&5)lj5gigohx5zjD%vJom<4
zls{?Q%U-~rK7I}sCcV+^ImybmG!zz@M!B!6;b`je_sE@eo_*i<qOt&YR}CuWn@-=_
zke~0&P~IO7flp6S^CWpxP5Of7W*5CeM~iD0kH+UR-Rf|+IkH%loPEUH9!!~^Yw)mw
z&=pUs%)1KrxKlP(pdNs;iKvEum{|llSH3II@jwknxK$1z{#*)g4k%FtGa8vOGQq<l
zredssXhN#M^4H<tMo~M*Y8bqFU&=mThhC}*cvPpgT!n_+4AhDJyj48IiM`@FuaAG`
zzlN^=ZSQsexE>K9RaW)0Wy6H4t9lufXOBJgr`ioh>yk}MZdGIHCCOa_V(7NW0oH~X
z!5jNg*>!sjwx_j$aQ~6-b1FaI7M$JO&GA~_?b!uI?l36T&ZEz4@V7?pjW_(AvB@*y
zEZ^*G6so5$krt?xA&q797sRfrUoRCnw+UIs?gt}rFXi58QXpImZGut6a{iyit!gK~
zEG^sW>^-l}eHDNXRZfTCtE$H*{2lCD+$z|>J~Zf(0j8CaZ=;1Nc!jif%=ZE%0iGNI
zdR?$G9aUe+9er8L4^&so&q-Xw>yPd#Y?CT1tG7QOu!N@sZIWtKvS7j7^hIqo6|bR7
ztBn@*-<TLn-=l;Im4eXYf6vP}6^31u`{dcXg@@Y-wN?~bEEqAO3pKGXbb)8+@Oo?{
zol#7IO@VnpJZE8JU)A*YPug{FUJT?<UQJX$TIwiQ0pkMF6sB#$aZk(Bj%77dqc08g
zOXnXF7P!iXWPzTco-s`mnQ;qV29pHsLtpA&#~pjKNuE2SwQ${FxAU#u>~)27&@S{c
z;LrJ6cf2)|-07BGLkBBcp@i$F*)f4UE<%b;+;?_I7090I{h+h{Pxfhv@-@H5X>wYD
zx)`p#*G1>>GLze0D5^(G$Wc`su6@*ia@gv`>ym64{I6ITzJ=*E+9n$(2`*F{o?O%>
z9xx`h&0Bo5pFGqiEksfc_N)eA+K3Q&hPrF{n^)7};VH@}1MJ1+X3%sBtv5_mwX?47
z83|`LyH8D|q~o1OM~9G-DIekz+r%{bC;>9o*WhDx{O5?=cujN;t+b0f1&d}EFT%Py
zc`gI3-DdwJCS6f}T4=2aC2QPAh%VM65p^NXHj+(@^v-j)KceD$pT6fYo!0#8ffNPd
zRflIIW^M#s5IP`^yB}$64-|h$#rElkFr0}tmNnM8NC^uId(v)s8uUlHGekMR!QN{5
z2~J_U8v9f=T@S5_Y;*@LY(|+Yg+7K?uT>S6ULGA`Y}jbcU7K22DP07S=!wJg&dhes
z>i{a2_!LIr&7sMopQp9tUU04U)Br9_vkL&9Ql1OyjNveb<l;Q=W<96Q88;RO%cY>M
z*JCL8sh}XTa_FL4QRs`Beti*qGx8MAWuo36T_O&WqOkTdJ%juQ4{#St8_L^>NbKge
z6`d-4zC`*C-{(_wt+P@HFp^cTC`jK5%n9z9VXNC3Yi%95Avn}I&5rVphT8H)jeU1M
z?y-pXe3@!Ck!Qc`-)7`0+P%We->L+k$k-1=b6l$z7Ia3*3S9l9rkahauYLVoLYmk?
z<4LhN7Q3l;Ub~d^m?{z!U$S!DjF}npeR|8>sPWli)&!rhTK#{Cmdd!e(=Io#E`1bn
zP3FZmb|`)6A#b9?kdEzriWV9Tr#+eQ`^(BI%qnR?J+1Qq@>O-P@9V9;Fy|YSNO`ju
zo!0;$H(l5(AIL&-T52z7&F1{1^D7(bZ{BZ0oF<aG-Q&O-+L4!57FD_5GCopmyH<$g
zF4&LV%$T=>K&px_G(t|%{GsHZ*q1rZaw2-21-p}~cSY5T|B55A+0aMh_PYP%bZQwR
z8pU$#cJc&~K~;ir!}J|_!a97s><QAJU$z<RFche~=xO|XSyfFhEv2XPN2|^u;U_@_
zy>S)QaIxsgwtn)M8%QwI>aH(n()%f3hw0m8dymvh&HQTBhZ9e48UeR?{bn%-Qk!!!
z8MvB9Q>hGne9_p!sDtg?m{(LJU)&o$etP7c1Y+Ui#v;|EIROpfGL|+K?fjd+>Pj4r
z@CUw7qnQ@UfaN?BmHY`<7VL+#Zw=2VaYU{V)*}e$Y|~lph+}a-H*3!_K*9!GC`*jR
zh4`8=yX-bi_>|V@P~)E-C<KT+?qwF&5sLVH9jBU0Z`<Yx7qV6cxn<8J8r~FoY@Vfd
zi$zTqJ)vHJsZ)|C-TREiGD>U9YCMupz`sZYoTt6Y!_La+B2lLH*WKS&9BL{~M#@Ba
zu3_FE!f(?jV^<LAH`>dvufXGJ{2F7`YGhsHE6e*re&Tt_#2_YYClMm{-TA6$ub?U;
zq;&<0E8uU|Bf3^h;#HG9Obk+K-K^Q=uy38DX)AZmTBv>Ae2oEl;=w3eM%Olnl@qD@
z&C5QaPV{Tyd58mE7S`A)cv;IML~d$86uk_&y)lZA+i>6+>~09*ufU#mj!$DdTnHO5
zv@|PPD>r$Z=+5zArQwc#cv!TshsTZk(sX^9z^Y{dY{q|X$ip<9+zlq8)02>UY02tI
z#EdD%x)Mw!Hi5?*-Y!41!#?*o{NTY%Q6<71c*96iAwuHp45cIB=}dOIw!(-ZY{ooc
z0z)C9Lp%E_i@-^p4|cA-KJ5_-A+?Wiei=t*j+vkrQ&xEpKEVE&M0H}qP;d50F@Lc%
z$9jhvl$fD@)0IaMlnK5Ds-XO3rUTWY9TL#84t2E&-<KQ1Vb-)3nd^nz9(&n+`gc~U
z(K%;<+4u9_xkUG@4PUexpUbgezM_dtERsp^!Lj*@WZHPoAxIwx3!Es&dRyLc+l;nl
z(<}Sa<6Gq#B%Dad42FKn%#n}Ij^rp4t-lXjkA~7Teou4&9RnOnOX{l-AmekkNl-pT
z)9km^EZR9UfT$bs<f109QRFrTtSxquJuDqds7g=5cYe1X+A#>JNj6ba^N%S%T*qNc
zDr$LOr<|*XpjPi~J)n{R3w8L8i8*HVTD#&T>=Ik{b{hzZ$YyP|fAFk`z%4fzBJTmw
z1T<ty@{5AEM=%vIWWxzM?Ja-wdVciyFTkM0l=WQt!nnM{=}W!CBoc8k9@5$pJk#4Z
z*I%|+K+R_nO|G(Ix%Z#m8Qn;oOIUn=^1HL%0>eh8*<2kJ&)b{bZ$h)j^;b?Tt-PAB
z{_grhDN14V1TDt-nEkCQLju=S<We>pymMf;pUmw>c#?%B;9sT;qXNpc<xa%D$ZnR7
zc)YyG@WjhSIdaqcHg|%tAtc6&PONs8T(Y&(k{D#(k{_wQhXyuA-R>v)T$<6v$$>rA
z+0SB+b2gRVSCqXlzoyQivs)b<ae8Idp0Kamg`Hhe6V%^UEi>R95!pQZW9~}>g!ZM^
zNhWqlM-$iVQ#e+mD(i|)fNgjJgX`>rGu3Bo90)t}b&HV^TczHhBsKPZdTrpOy@y{j
zv9Ph~e<}52_k+G5MGDCl4#C&y#UI}7PPMCG_RlZ4+(|yHHQ-y`&d$%jrF+kT%YEEs
z4y$0x;;=eyxQu!o#LkV)5<14OTaK%o=Lh59*%N>+r(~_Gw2;09;&HU6^RF76%xrq3
zdYS@0dtM~tr};6oWnBIe+a|VPb0IJ99&!*1(t?fgw*b<i=|@ht(4xGN4`6@=z^T9)
zqOMzx`6i!lZd{_m1B4lCujlv#gqhxj!p0511WtecR`N#wzUc77eOF?ytiV$J!19+1
z%yo5LkCuvkz5B=zg#H5~oT#igah$j&-uaPPrsG?~&QlMqcz~7Pw%LX2gz$X20mb5A
zkNT5lN{Z%I>GW&Nu{7?MHRio*FKp~@El(0_g2>OAX2-T>kwz*~B?mlgHH9hB11l*V
zjO+wSl|7s|E3TJ-&(`G==H=afX=^wsqW2aVEcIQdAEob(5*E&ieJ<EBRqseUkIyk@
z6!6$xZA%?S{p5++E!ijN&1U%*OK*!TQ?K>YT}Qd6zqIc|AY#pW#2t@dPS+l<qk8J}
zAtZQ}Z&xquk9|cP-_Y^~iUGFAO(@=x=EnUz1W&df?6j^0Cl}DDv@N_(Ya68?r_)Jo
z6k?h)+|@3yd1|eX)Y%2(R2aOiTMjzgmKJ|rJntfZAtHAvfs&qBP1c{A8CQR}ZEV&L
zLP-2UpSdY-*21&;%A{l0dDNYRn`zfcnI{nxN`!Fv0%vlC97%kX+CdmDtRTZ!bt)Mr
zJf|dXXvxH6=hn^`D6MQw5}bZI=JfXAwH>aR3+-t9up9Dr?AwHwF5{p(RdvufBne+P
zP7cqApdi?w78X}S0;O@@fO;~iM>Nv?t9N;+Ymrnw@>{I#o(dE{VgBv+WR0%2-SCY?
z8WL#Umbt$o@z^@o>aRRpu|pr?zbysnEz0bj&#lf2QZvef`%fsXnE`BxEH>L@%-F@)
z%f^Y5!l(W3FZ0W=+JAkH7OmRSGio|Ic##J#3XKm-wO=h)2+#<t1cnCK80CPskiyDZ
zal3iAd*ij{!YcPm2~>~Yg-0K7&X0k7c(}oMndklhFS_9)q2&v!BMx04U}*hA-&Xb!
zBP<(yibc&gFu3}1K~8q_l_VB}0m%tJKuc~St*Kj(A5wEOO(|qmIZ>CDW9b95S&|yG
zNr&KUv+_hGoBUQ*ZE#gV-Y<1W&phkeMrV0_)d6*0>v+-Bdrr?UWdqi`R$JO4{EBJ_
zaz^^A@n6(Y-$VyWy2TE90hpkmz2GuCJXu5Sfq2TV(G^lwb$JZAxo73>98CVpqw9!i
z#z-?p!emd~0hBIo6$9S^f_8{C1|=c`)72by&%Y8TQ^`C5E>WkIxAJ$Rgb1z8#adt9
zbi80E><zI+M~1>H>T%gio$<GJoV8LA1Gn$5Wa3OY(r&af;|??|G*M0$8GAZf-BFHa
zXq2s&GbtkPe>QNa2Y2~U2rPv)UTaHgA5{{I39#q4pU8C9EJ6o`(Cyu=e`;L5?9v>8
zZV&a6^qgXAGm$VI`cnw&nkV-UuOmsUogq6g2@d#f(O*abMA(q25lT8X(214SV!$5Z
zbEvY$N#aj#uP`gxG2*>RIupW04kcu*K6GtE<ui}3Pk~rl1zvFiS(SiZc6JJ+OObKk
z)dz^jtYUF4mX>>JF9K5tBdZTPi-@_Q3%@O-h-bIij5Uo`{qZ0~mMw5qo*Ogj(@Pz?
z{)>w6H=~Oy2FnbtVHx0^RA~LcK&>W&tKJtdx3w7+cSfP*fRLYK5h43JTe92Ywxk#l
zrNI}nJ%ord?qmvWiR5M<2(G&~R@k55;`c}~I;KV8PC4YpfYU)d;HsE-2bza%ZSvIw
z2W<7k&DmdPM=G#1<Ox^OYmz$3>gC6y+S9?F+e|2MZ+z^#GL5xD5J?C(KNa;vt)r+-
z84y(5naPg7T+SoHJ-&*ph<0+H@%8n-9t$z}{N5l6h(-kz?x?<KH46?ux?h_P0t1?>
zS45Na+6eXz_g?ObtLs&-1`f_gVN#~mav6uoH-*glUBMz)Q*A%Q*Mr*`h2u2q8>B(~
zF`%&gISvk^LDKLAP~wEasZU+<UaqW<dMX8v84GwuIm*zv>P<vkcmjd$M5IRj!D7r`
zqdvLRv2ovUK)ibEKJY>(W<0XD4{tK6F-9Z8)kP&@9y0NDVuHrA>`Dd@YSunsLB)Z-
zrwcIeg9sJwBM#tR$_AZzw2+onXDjO@aY-3(_2H8a3uKZWODIB0AA}{^3bNOPqhCsB
zaV5LH`|BJ&A};&;bzlt|EW$=YS`KdQb#qO%<{hJ_aAvE%rQnQe@|W{=%d@Vs=F9H|
zhd#!e=|Sg(YHLlV_6JB2w~msAF@#J6tul;s%(+ShW2<G+KvqS&V3&}P@85F>VhcA|
zt!->m*?b^duJA);j|^YPEvCoyTw!Bz<EY;wm9%^l2W4;K)(6v?lMu~L$N#yTu$FGC
z!alZ0yr)GTmHi=6LHw(pk!-E4(y!`9r<9=%RMWsn)Ci!S`c&?DL_L&3&f7Bx>gyA~
zg4;~NM#zR;V9hWJkp+;sQa*$=M{9<i43^%BwMGHK$fQuu>7ZNV(w@x-jC|4}OXqF)
z0XslPQL%;6;XUOm&n?N9xrL{n5d}U>bFR0FWkV+RM=FRU$H%1yw1~R7stbc4c}`Ho
zI<??wFT<Z0N91g?R9BkMLwwo9gOH~VtCaxauNP~R-If5~-$J%2@O3M~8=K%a?_KN}
zg?zLl8l3z^<Wdb^P5VtC|F{WqqEwu~q!VkZD&1;w-5p9z`M%HFmYU3B92=eZl;h9b
z_RSrTqS1~`>>U8xOSf80nKDQPatNAu4>o)Kdn%df%#}pgW3Cl2&(FvPI%ncxRKOR3
zYo!PI60vUxg>^tN1+Ctk90B4=l$Lmyb);j;us*2Z@~fm+gCt<74OhL2HUIA3eoZ`g
zRpzfIAg7mJJ8-F@$+)2tj#B^-zZdEoJdXy{nIIQRSHA>?c;1cyo1FL{-MS3}Xco|T
zl4qZH-keE?A7+>~dN(?#AC>b#X(NXbxZBSInvD^Gi!K7p5m^u@DnloZ>74lM>IRg+
z^YdQ=MO8JeexCVGIuG&$`=ESjXxQ}@dHcer<Vr&DxZ5o4lNpq?$0*>xh9pr`<2F_C
zgn~~+RRG=fgd7VNV$!kA^Lc8sl($Jt4%7gR5nu%bGZn{C9FWtt^xrPL!Dwkc6O`O<
ze)~CqCN|&;C4jpZYc*?!jZlo@b+y0VNZ1_|Ed5+Pq&|}eNDxIvlAy`T0AW_BSR6z7
zq0jy9s8PepB_Q<zV7$S)?VAJ}i;qSeZU#2c3W8c~Cg>_ySO0@aOU<8|CI|8_Mjo2B
zdJ9hk(et~IM-B`z_(%d?y|k~J4bm^H-jQ!u6Mq_I;^<f!Kf$V-*(bgFH0#<~s{gHg
zy-9d!pgKYTYRE(U69#1j-gIcJZz!@Xgy;_x#I=dYR-x}fG#~{hJbMiO)REJSel%-?
zc<+A0cO|%f!tmp!@QK-hA=nWq*~y;eKIAA3lR2U(&<u=x(^R=xuK8vXc(bt3zoE(D
zo9jDPEOt2Cudy=*?I@o2kqEHou4H!=kJ)8ksHHDCIC=ep?7+zTQ#tH()5~-dEgrCT
zM~{ly)G8o3jL1=ZGc`#1i{3g~2zsTMGOB1q7uVKXL16!mmzpjEJX?{(4{rfZpAj-}
z0twtqH&io8D0Pw_jJf?`=B&~kLncH(ike>oaOXlZ%giNX7g@(T)(p+95qPTUn9SC-
z1%K>RCFqt~Z0vu4?r8&Hb_JPp&E`?*ReEFOApKjO(?Px_61%Q4`%{3OxcS<QQM;7L
zaj<EJ<^YqF85S2-Nr@46$3F7M!F4S5iBz-6rnD$R<n@=^$n?j4tYD5#%>frdW?cX=
z;ZJH5M3ZTxk;D}R>Nn{5;R8a-qiLxZo^6eHkZa3uDP$54WQEDU)$gPM{8wn%Uz^LH
zHDSK%3q*2sHk{)!SkqZIxYxx9C1-W=B_D8Bwd4qp?m)=}*Se3Z74r~>doTQE5`=&?
zh9)1luBcf#<i<ZJ0!V57)HOH!Ba;S~0Y~0dD}fYjne5|bYRVI1`-SMxODQ6rLU=J~
z%{i{^^)NjQ4D==mz!<ewW@;8R>0k%&iHLqgRt(=D!C<6E>A<6ifUknym<SGO(UlQ~
zJ%99IfhM`(V2{T)xPP?=FXTaZl6;i6Lnd%}-?*gl=(a^BQ(mxVm@r19pKU{5N-V#I
zU0;GQ`Ls7GJC0ag)x$-Qntn@X$*!`0a|=$51FnyX2)ui=vG<l^-FF%ND?rh^>nV|J
zH+?@^H-`Eg*I=xZ$DaEW!1PQ*+Z9Zth9DM!3QbZQ$}2wF`OdqwCO5ZN#5O)v!1tAV
z#|TI@cXFLdKqg4h=OLbgl#u5kJ~MlCZ=<m}r2d)Z%TGmN;MyGpBuNP;IFod0aU0rA
zy&r@kJdnC7C@*k;#_!UnvRUEeow^P0`BwgF^sU2^dKC|NnuH0S-Q<$ffsgeY+#k~D
zlMe~JxmFOm%nZtlY@FM~&1L&U@gv8F2s}>v=EH4s;h=W&k+I>$TVoY2mbLA06z_Y}
z?#F=Bv}2#<HPIz&1`6Ksd@=q$Qd(9pt|0sn`@42u2q_r+i0<#-zb|FOfp_o46LX`l
zg5NDECcX<7z-Q45W+F*veOct;yUj-eLmELI339+J`)^1ID=)O~>^XC5*XU*iFAJq^
zlf`>?P~-|7QpH~mAbrX^IS6+JSx9fo%y?7G*}=iVkHCf^yy$&cgFF#!43lM<_o61{
z%QOGo;I4rprWjKy+xSpW!H;p7Sh142o(BzUCUQ(5Uuq58_}6!LL1TF_P3pBSI{Tjh
ziY<u-ec843!^b_Go`r=d{`XK80dC5AF&T$%9fF$Q>-?9B)~HZ*Wdq6@Li!y`8$;{)
zd4RAooE8=J7n3*)@-hX7dx-u*8E{ez>MnhB0r;1uA){$0p9Yyfk4^lL|BJ@%PO|v|
z;e&2Qzsdc33C~9^<ijcs(QGv!#+8tU|B?KpGTCWn_u#zLYW)%hI%Hx_jJ>YK)tDq&
zIHkOeht$GJq*pT<=JLx8=UXs`1Ne9D7971P654X+AzoJ`Jtz+a4mFO^9HzbPY*dBX
zytQ_+(_*^qS;%jzEUi;nH-E}_v-)W1?}TpXDk(PNQl`_RMNPl@W`A2_4~)oKS{nAO
z%cdCDD1F%s+ZYVFPzWmf!$Ov5#ww&$<ky0u@crPt>5Y&<5Xf>MWG#92+3}^R=zZcm
zPe_>WE*{)ugb;#_C`m9W8DgST8HG(61#X4m?))qn{P$)w&laeug;9deW{mXFlJ1tH
zwqT3+!1j2@KoZE$BZxZiJy?l1WENqs6IGZeH0`XbZ(Y6~Vjl9!US3xs5(b2DhniE;
zPY^-6h3m)Iz?@aXrS>13L3A=aUk-GD?yh7mf%%n8EN{wQfBKmk8_m;MSBpcr=HM<<
zJnvmu>F!y2p$W9KIMNDCf!yqp+=#=l!%i}Y+_AWVx~uK3!z>rqKy`+O5bo*!+P|Cn
zbR}@1&z3f>1xiL-*H2rvXA~B)vnL)2%GhcbvDr}31&7yTq6xU6{D_{Fwyd;u-}gTt
zQZGT9g^m=09TN7ls=TWVAAH>WsLwtCil&@hnzU|3Ded6f%OTTMkd@3_qU^M$M}<<y
zm0DE;V~T%=l&_7I^Xk@G=x~%hdE2}5cH4H2^)SFQL7zFxylC+~JXFNPtjy%qGpyOc
zOGYKfZRGK8==jevVpYjW;h11En6dwaC|E=jGi@aUKi8wTzEBU^r&nGb^M2<;k%83>
zE1T;N{$&nFNHd1jAAH9s!L6QdhOlBwb)TFI8zA9|dhfp{_j+g7sFbN{`fl)lllKx(
zvB&c}Kg+t3sWhk+trg|w<`Ot$7YlpyksiNSG2C0_RBM6G?fO}Pc@%%2b$%-@Y-CzC
z=poBqC;z5vDs(biakH`>@=i2j+BI!DneamFySI>}Rg|PtWX@*Z)o=Um_iKJtdf0)w
zbLNFS;?v=V=b<`%5WJv>vgkTvUr`J@S>>ZAGBb^OkA0zHhOQs+KI4t_`Ecg;7L;q>
z3D_VMy_Z=YtkAPQ5Ao|v#pb5Eh$L9#MY|dHs|2HOM0}7F5vd<@@ScKF#Njiuy4`y!
zQZd&Ra0p+Yovp{{)Q|;ELkaN6%B~cZVW$@=)%YlSk2&MytxTy7k)Gpw>R-t;MIVp<
z2E~}?$r|BdP~)v?F<GsKC58-Y_HinMtzjuNkf<NX?5XOFrqRxt#h&IE(g`9WarBWF
z+ZX}7qEG31sJ=$Vp0Pjs4}F>@I6WqGH*NdGNf-hm1Z-~Je*(dTA@Mhyxf}*>%$xoA
zen_I~mFd!ypkH{~1L9(Cnsv75Z_9JPpi4D8kJlX_^Jn$a4(O>Ll;b3SCN9RM2zU7A
zc(@5}(qxZbP%+b7T{hG_Ca#Dm$H2BMq#+>bB{ULxt?#^XRz651cH&2ZWgc1rfkDNm
z`LLX(anOVtPJn-871sJ^TH+U}LV8Jk=D$=7d?}5h?mt$f29SP58a^M}%I`sNts^*j
zmR2BWTsiJ}65jl4$f_aQC=ysFH!GKykp-_BHFdpHdqEdn;SR0ib`jjm?S^;_CQT^{
zDx@0r4T9=&iXii#r;?)g)KSN{FTh|*CUtv{`>kr5w6=xcy?Xi}XM#i8H8~d?H!EF2
ze6+(CE>-SpJ2cgBnn+=*!!#KH?rb?oG7UBNONEhuyUkIesckLxwv1Np{%_9MISR3Y
zMi81dwVVm{?6aJb<*B)~-V8J>Klcct%tSfKdTV71>U|!@vf(#e&t&vJO6bS$+!v5-
z-TPgv!=~o_Bi4p(k3Wb3=Aj;UOGZQY91&)7dzxCCXLtnZr|f4$w!VZgq`1acfmF*T
z`Pu)h?eOhgmjtZ><TQs-P!u_<=n_V?!P_iQ05vCfc|I^WpNqv2iM|ce&7uC?_--{#
z_XQbEh3Wc9Z~YS`bR9x$%qcw&k*xk-$3+R%KQxq+oDhC9HC=UT!r0NjspT=Oa%>3G
z#BsbzF5^`~aPXt7rQmgVb%{slV(Wt1<Y)5BV4h3|6_~e{35wEQH<GT`z@x~)mp%Tr
z8~8KKFnZDkS3dzJ(L)Saa<Lo47X@9)7KVA6QUm^h+0^(cdRKq4U2xnB`8D|IIGi|U
zlIqcdLdD)(ctE(+@shD2waRva>avN*_0!}%yQ<t3Vg>!|Tf|sdh}o|%>hcc3U5=kA
z=iZ~^SMc;CyJm$YTT)J&O@5sh2hV;m)5U^bAdXc382POl578t<{ux=X;UjijZwdM$
zBm200G2H-efvx8F-v`17{UdHzK&lAdv#a3#Za{$aaO5atJ65uCc7>ZSC@p%}h*2aF
z(Nw`}i1)O6AeD;1{@p-B7fb!mnPo6qTR_Q|I&efqw`jc*$i%c)X=TPV@g))DzV!1t
z(021u*}E4Nnsz8ea{4S%+6(dLhrE~Y%S=DPX(Ki>TUv(_0K07@Xu>EsMd3=Q7ie95
zaj0-L`~o5P`_8uL4W-`y-CsCq9h|nTS^VU*FK!XMekDUuUC}uzbPLqW<=wJ7Z*aVN
zni_O&$rTC3)sp%k2b9e9H2vlSx|OM?&PC@)oN#m@-<x2s;}#_l8kQaO2?|1D*q5Im
z_WGHbVqnjgsw^NQ(L3S@J!(7rj0m7yvdFHC3}n}4uCrMEHO?_XC)1ZFJ!tl7`zx2R
zeny$)+HgTL;|<1q*572z68Sf`(PAOnFBlNM9$fQT?snf&SvMK|ub~PgmGV>dz2K?G
zWGHAWx;qthjK5y5UvV=Qsp%G6ghr}mbPC0QMS#ES+uJLQX0%I=VlPhS1AFgosP6aD
zd%S~S<lE^P_rmI#o?D#ckKD6O;RlRR%TdM}@0cP=DZPp<speae8QB_kM6a+XwN7+U
zo#gohT@!L7UnWy8#%~8Ne;&!#9-+qMz|fQ(D8+zdpJJ*vB0|bw;}T5rxw@p~KqR+#
z&_!F)j0I<JqepG6Z)vc6LeSLY!7j|p5CC?%{AFz4{>`q!^<LVC;cEfs43U9F3+4eC
zT_E{0sYjWA95NLP>;TL{EIv^|vAt~}5jFG)D#0kBMf8rN{4breH05xWxKKU7fd+$U
znuWv*T@1Pz^MhcvJwvWEGan9i(SGX|p39r%!^ZI~&iu_#*%fXgc}y$$3pjEcb5yj}
z4OR`IZcLFoVD*nkZdSKBHa>m_CtXen`wDh#(~5hG`7s|Ghlvvhsl9+~sAlcv*&EKn
z()nec<x9ocC0hfEs@OvN`Gl&X@-er{9?)F=O<7s<0eL6SsU)V^FKCj;ACpjz00Tfh
zzIVk;;likJv8I-0t_3-}q5_;;20I2RRlbXW4T=#OsGIUB3_9eUVI9@Uh3rweSN{=n
zr<yUPLB~x*LQw`OVua1&uu1wpAYAfz`R8+WBkNqxY|8lCYS^=fK@M4NT_*Yc;y=S?
zp4APIrtNBGMa`l?DV!IVDwsj`^!;!A;q=powiiZGG=I<p6DJef_#b%?)fT2>I{fc^
zQ}i;lonEap_gZ01!I?f2DT`l2nA_vz;oRn0Ld+iBtZfFo^rs>ZXa_<^qy*mysh2P0
z&aNc?nGNKp`SqCKFgV32vu@5i!2G~HU5caUG=k1=xr-FV1e~#TwEJz&*X!fTI3T}2
z_)Clb%^j2kYD+rwi`K0k#b)qHHI4`;Dw}#Bp_iJ6EOA#UvI6`1w$QQreFHHmJ;Sy=
zI&7D4^^EWzm9Ct#OQr;Wc24=J#CZt86v%y7IKNHZ&wYPM9!i5QGVYbvd}t7yd4a+>
z+-ef}<dQz_>|);3(N~vrcF?}}*ilQrNfiKg@(C`DArt8RR1{{%t)7z6Cis?+QC7=M
zTrQx~xLN-<^dYhZu83+(l1sac-C%>N<LgT`cxNFByQfoqo!{^_Q`g&_iS|ZSqZd9X
zew4Q%!Rq+`>NwLtsJ{P?Gj@?ZjBF!eguzgjjD63pu``xImWi@`Wspf2LW`7r%f1U)
zA`)faQVAjZGWKD@{H}-p2mkxzK05cFd(OR|^S<ZvdVikrx>`aC4?|N*y|1f(B>m&#
z;NT;XG5@j(7szkoPdZ?fuAtA<`bUf@3wr_fJ`h>_OI*?lmGzL1Us<>uzbEmNDAzPj
zm7r>U8|+?hs!Shf{=jW(a&2>4$HSVeb7G?2(_SpBRPsLWD9e~%09ZAj6)$FMm4!N^
z0(}-s^!~(a@0?rQzPglu7=js>J2;5d!n}WKx;2*Wwy@aWHWanfgO0__fA0#9dOBu?
z4OtenwDlB45FS1d>hAM@%XjZeTuK>>ZMP1vGg-0z9za<el<LeK?#ecn6z&O|wkefm
zW-{%L*5XDb8T-qy=|ks2WiqrjzZ;*AOijrwv>R_CfHWTwL^eWkHL``~0*L1H#hkHR
zcxZ4LK+c(9>u~q*R@uI~sWEc5Hh?nl!vKx<=F1%zfYjSDJc;~-DH=2$Wnd44LhI#8
z{Exp&d}ESMK|c`a?xSRrYu$Dlnc!ct4!%1{mXxlnl7UvL%I4}!Aa(Q%bPNK#ts6=j
z%FHLrHU8?;;i0B4=eU!=_Gyhne|D`pMUQCT0r9fk+TrdA`L|w};eTq(b7BF{)dA%w
zRs;=nzQyJ(Azj(h0q~L?g|jMPgqnkc-TUy1Fa#j|gaQ)A`m%}V{O(qNaeq(<`z})P
znB%&#qBfI<bABp0U81Fv*T`_^Y>Jf>(JiX-fZ9m$&)8frwNH#*RZAFWc|9T{vxkCy
z@eR*Oh+->_|C5}u$}W1Jtzwu*uFkbYa1R|JtNZSyf;ib~-aF}JSPI*8F2YSHPaO$)
zbQnV1`63B+|8n#Jkdv=z>d{h0dar?}zei!gRk<k=y`p|Dajbv`dk#^xaK;B*Bs@pT
z{j4<EMRa(IJ|=4O6u}Jt6an-{;R-}Ty1eD?tNpPGiRmaLSC~PraG~7m$&&}%0NuLP
z^wTuFn)e-Ky1?qB&qjj?QRs#916G%>wKv&<;k%=}M(baV2U`1?Mww0QPa{;bX*Acw
z)D<DpEl|7X`x+~b*O!wAf<u58J)Po&1Sun#S)bkZ4py|ng;<C2H|S1eh<&5;S(rWb
zAWT;LSZz(@n;N%osuEg$nxOSf(aVcZr-@Qm$cn1l^3;y(>wWbw#wv4~fTQ@p(+s(N
zuV6#{N^YHv&g-x)qzSg9j)uS3moyr{`^ZQ+*jn*Rfi!jBg>D|(o@6Ne_q(*Sx3jpD
z(Pfa-Ap?{{p^B<71h=e7fn^@igTsmY)-U6elTHR>uB$1=Ow*EM<n6@roz$ty-AnS?
z+y-r#M2KF%7Pk#B|2#sK+uUV!{5AZ2C};vB=U-8AXeG!)NmHeU@yFw5^d=DxgC0|}
zwGF3nu5p)Q$@ya+rryH?gTYm1kuvvry#r*l7^6nxo~2!O);GFvNC;Rs1=Zw?AYTM|
z_RV*>3zf$Ie)X1X@H#o@aU08QR`8VDi#A<yUpzW?@bB^`BxQ1Odz}88;9WHO->76a
zOwq*1ZKep~$SY#`F>{@c3Dj7$xDk@WBuBLE@v!bKvM;(wH&aD6&W<QtQJ9#FkE6kS
z6{iH5fk{$3WpD1q%!#OJZ;gdLM@x!M1ar19IQOR~;R}&g3}Jfd$vnFLrnlnbBAE=c
z-ashaB%7F}`C5H`4N7M_#YeDj%K=&2cMPgtX{+oei2tyy6iR?Ub%9>F(L71!n8&ok
zbOORInv@Y+*@HpQdl{-|2uWBW))8Ms_ZpSrP=6Ki8yiTYTL&C2dA8lVHJ9!P{CXaI
zX)Of(J^ysZu7vM^F)QB;6T3Q5w}#d@e`!YvVwDPIIk50a9kz{Mk2dbh@q4E)*Tiy0
zJ%?cRV$Of+(J5Y&86Vd852zT@3u?$iM-#-3Kx#0bW_H0nV+iN8h+};;e#Ml6{kw(Z
zriy}s&jP!$Ixb`oobmTnTDI^?YPWJ^s)0xoNXE9c=JJx5_WrD#<*kLrE}{ljTkl$&
zRV0o6(hYhrf|_kEvkR6uY;)l&i`(?LBtfV3@xCKK5}5+o+rq8V<mI1-@@D?;zppC`
z4p(^_9G_!5A@yHqU#m)?UTKw-24o|O_CeR6YV$kQ>WG2)#`u`4nsB+9-xIP;Ocfdj
z>MkC-AxneoLjMWAU_2kr`~*v^9+zu9ib>}_xxDktRz#u!zb*@m|4=dhM9LyMhv3XO
z(1hNIjlz^g&S`bf-B9#-WSw)8y-h?}`UxQ~wOskO#crFpBkvM)S<x7jTJVMw`{fZ_
z2z&F`TAM{bo}NuFKYx57hF5$4#%2|YhtI>HI|Fc}xeMZM4uCThW`}7p#N~Hyh@JzG
zmXBD4Rf8Yq75=psB-ajBUvHav15~5=nT*{~k>`61t>36E6kDF8Ku`us>{+Tnp32Ie
z#(nUHYR8eOG6i}KTU&*4Mi6bS3P5p}EH>eG*YYFk_MVDS56<*;z0)iJ9GP&pTe834
zM_5~Prv3g6mh7H=md;I5@=s7$KcQ0?uF7F^&zmtq%S-HmARV*|A4Xu}2$bt_z($n>
zhj4Xn9)aT0<71|jF;51c)E$#6zkb3JhvYD`ZH<67@Qhop<>xI+GO5uYp|S&Sm~8&I
z(`O7YJk{o}5OxoOlY1VZ+gDP_{8r}~1>m+xZcGl~pQDN-Z+nhs@?fvN%Ax$J`s2Nj
zayB4vHr`1rr%_@!MjfN7oPnu26)qQc_?b!p)IKP{<41uZ(^yrBN)pbn3CJVe7w0VZ
z^3OaO_6-%RZVF*?A=0##YxB(=7a|wN&!Xm5Y)ngD+;?5+srPJ4OIT<hByD*eW5f~E
zX=v2=z%|ZUpfnH0j%dtL#W!JiOIG06XK(7<;Pi<<gMh7xhSxO?_bP=36wgdiQGjmJ
z0pnIKbSj1Nb!m+DpP*<$1AOrGuq!sU!nO%MHBrIKns7$HB;m8PJz9&m329C^r9Fnz
zk-F;r;!oS(&S6zNqKn!Fv6}Ebh?6T)D12z7R8@U=N5BCpXxC&cUnS!(m|uy#?7;Z>
zdCi{b0wE-FP@OOmoGDZJy%VsYyJ`ITWfh*+rKYY`HC-U$P<vYUq1Jy2fHcV-iz<q6
zctg{{Wqy$Az1z9}8iuMO`iLCX^fI1nl8u)xw9bqMAlI_T#0#7(vgPtzq`|Wcf6tjP
z>DEGeY|Mg#VRgS+T;J~Dt{u<ZMG8HlOMk8SwxE32g5tafw_`SxR`k6F=tGd~q|MFt
zq8eweZ*My5uEIa*C1g&|;r-^uZ2lEjL{7hbYXihjwz9y48Jcx9{K~+8)gDJByAX$z
zJHb{ZM9s~y-!|cgE0QUR+(H!QwmpSAvO*oyZ{=(JoI3GHn8ayrPzTfeyUN+z!D`xe
z;{MP}3F2*S*SPF&pZl<BUL!CjPvaZ`jCWaUm}9sXpGI*CRj*Oh2{iMH-K3i~5H6U3
zV@pX?HRB%PK(o$XFBDY}u9^z0HX)U`E>RFm%FkY21kLT@_ce3PI?O>qYZekRYo?i;
z=Nw>Ztq)HyS5pm!KPIf!8Z!}kaRR{sGULr|BT#>E=v7E7RaS?-=*)R9=G+&Fb)a05
zY8!!FU!FlRd-3S9-UzQo>pWN-Fvv6Wr7q*>`*_Wl@id1X;p2T(p)M$RX6phFQbbON
zn`=2b8V#F?cG>=2-bPw@`Ip#^{I&^ckiiyp&dF7fwi13{VbRfTf)BN&u8ogF11nel
zmYUldzyoRDiqk}-9~i%^XKa4_c*hI4_tQ8#WWS65?>z;7aE%JI0}@f+t+*(9+1<ij
zOq#8~^B#}xpBuhjHopLpaL;Mix|{3Sw}bj^#(nB6c6LMHOFhD5I}$lnQD0AT7xlky
zw#>x~f|K@C0^NxH40z9}jv1h9DaRy7HUYY^Fr?qB{3eJTzz8250M6NWOJ`#><HdTH
zW~s1u9&Tx0dj!6JV??Hh(8jr>xYyNmzxI*q3^apv>AJuo^%+Nd+8JncJi7Y&fl*Hu
z%Onl1?ZF55DeiqLif65}Eo}c+6kRkO-Uw>Qs${J4%nXx5IT^_psv8ke7Bpns4<bv!
zhf7a*ZuxZfIbP?&{YthNwTvEJQ-~;dgX(`HLnXs<Vb7Tl{p+QZOxJjah2z@|L4h0M
zBjGDkgAUn8l8x#{_<NMsu((yW&U5wTD8nPRcLjlhU=oLUZNKe@E+wNc8mrWQdeu=+
zdn1E=*Qd_~S$`grF?%8I{9TdsPlo5&!=}Oc`%DruVMaN}^shZ|ntlXNOP(QzvjsCN
z<=_fS1Y6`%*Uum&E^zAwwa!kWiav`X(B3k|+{+uw-(iQ%Aj!_aR^?Xn!SX!w>_=Cz
zVIuZMpGLwn0}cMW(QP1-p!xX>;%d0O7GXUU_|g0VTtj%37`+$Ok1>1I@uZ7nQ%4A4
znyqq8;bvgzzf8I3;Xffzkw|!QpQ1+6jOMRcEiO3<sh*=FGR*oa*>~V8Du2Cp;RQcK
z7=rdlrw~-A*M#ef#p-HnsOcU^Sz$!%b~1TqV2Q@)&vCmnTQvQ>(3bMJ%`!up8vbn)
z)8kH%%jH!j>uTE3rVHI;?{kYY3jl9ae>kG_T&}{dtY&?NT%lwJ9jdi$#UskYcA`iR
zTVJFw#U@tpl%<XkHHb(LF_(;MYEIEVp?h!X$VSKU-14@G#ij$fCCl?YUvg)?wDsmU
z#I<QbPdYmfol%etl_#M7wPyReLM`7_P3`3<_(JbS+?9b^RL>7w#}nJu3Z_@J`d;vS
zEwgI+QDLD7R#H?bz5T-j_Uc7avseJ`eW%=xo%(Foz7uX_zp)<8quXfd<doe6$#z~Z
z5Mhmt@vnsGyF_V5zI39GxVcmA6V>>v_QjGoP@TXWmCAzeB<J?FNUaRzAZEUd!fp8c
ztre?anQ<%}$$2GAS`B7G`TBy<k8s)i3Fc~Clg+s~r1hSjeQx|WS1wy|tZM)y=LsOh
z9w=6RiYqI_`fZ)_81(5Mw)vUQ*tr<FNOF<#ogalFcWCZX_=yWL!2dz;x6|<}H^)SN
zPdHfCU$^(&CsQ^LF-Cv<ZubMICn)+*$L66s>Y<0STc8JUqL72h!KGmGQZP9y*_+Dp
tx0GdXO2S~uFc^1o>Hhy|;D>SdMu+|X4bWq3ra%J<1056XDovNi{{!6=kVpUk

diff --git a/docs_old/logos/mit.png b/docs_old/logos/mit.png
deleted file mode 100644
index 60c5528e8d7fa4a812511189a2ede58ad9d8b2ae..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 1483
zcmeAS@N?(olHy`uVBq!ia0y~yU}|7sVD#o-28tA3Ir|$(v7|ftIx;Y9?C1WI$O_~u
zBzpw;GB8xBF)%c=FfjZA3N^f7U???UV0e|lz+g3lfkC`r&aOZkpoB+&Plzj!UMz32
zSl)DroH3BGO2%@joGy?tSKS&&E|ND{qF@SS%$zy1v!|z{w|A+W=^}a4<;te*?d@IN
z-9V|X?w<eu|6g=U`3f{kv?RzcnBhcp?lvL8?Yw6vg`Z;$vQi3?72GZ!!!HF?z?kIi
z?jp{8ZAL$k!(QU)>&pI+gOgd+nw{%m7|=iQo-U3d6}R5rwG=w!z~g%GU|Rg%qW_D9
zCPz;3x$3Z=`;Wui)n_DgPkov-<*3<)9lnuwbmi|X=aKVTQnqe0lR^Un6UWdecCa0I
zthu<D<H6IYbcS^d0gVhS0uBs}xQPj(U|B|x-?>041Hj?H1kxn}3X=t36Iek(qk*Or
z)8uTnirwj7IPPwIZ7jv`hqd9c@nX%A<8%Byd51^v4w)3P&-9CADFf3F$pi1%hkmLa
z@L>OEc|e@6D5=&*q-YMX^ieHwjVMV;EJ?LWE=mPb3`Pcq=DG$Jx(1dZ2F6y#=2oVb
p+6D$z1_l?tX<S9okei>9nO2Eg!@|`QGJzTxJYD@<);T3K0RTnuA)EjJ

diff --git a/docs_old/logos/mount_sinai.png b/docs_old/logos/mount_sinai.png
deleted file mode 100644
index f18af0686149081a01f463017e2cd78223925260..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 23749
zcmcG0Wn5HI_wJxbBMd1e4Bbcz5&}ax(jhG=-7TpC(g+d?0x~qxDUFf?0wUc=OG-D~
zJ-+|<ez+g+@7@m{$C-2XS$pjj&wAG0MyRVQ65>(gK_C!9WhJ;K1cHHtKyKONV1aLt
zm?^Q~FI*QTeRl{1p9K8}<JWB9B?Q6%QHINC`%G;&2YoQ=3|PIOAi-e@vn_w$En=c(
zkr0mgLuiieRlwG!ZK=N~(ZoRY<F+@?j;pt}oGM=xkk8I;t?(Ayx%gY`j6`si&&346
zCGXmIcf}gmN0&L9{y~K``p~r#d<*-AaboI#{U%G1iHezd<9cWmx%T7FPLcb@$l1{D
zn(sjN(9;t!&w)ZM=YaX510Gl|CM<-Qgfjmx39&3r7BTdHE-*4iS)Bh|LS*4Mu>V}t
z!@;M27h*gpc<$dt7RDGt{GSUu9DMqB!3hlq<Ndq*Uk$q0{nF%?Y&dV-oR{C-$1+#m
z*U4$lb_Yf8%5gYY%5K4Mms3vyDv#PTn<VzWM=Y@gc@L#^KkFI@n(p4c-gge_%sNfP
zNQ97KMEhMg5q!q|?Aq5aI_f(PM0!V1{;*c`q>N2B@O-xYt=iGWg#yfFkEQj{dH1?6
z=u8>=miC<oS!YQs5FI=cAbo;c-p!ul+xs_@`xBoB9msC283&@eiLNIVj};jni$Odg
z!x_>%5F4MsXSW{wIh&P%SWS5Wg(=~k^(1)*9vRLXHauCYT66Xvj<`8#tkb7k(_tJr
z4DTzB^e_l$dH*baI9jnfl8gy?+qxf`KwN9wdNUg*y|<;)kQsE|!?js@)kdeL$E(n+
z7?O-(<OXJvbP+)enQ=Q%DKT_vDI{1RIif2F81Z?A%fu+F*@4aWrA(^c-{%d=+aZ>6
z_T}_83k3Tmya0zG|N1DK@Ut%%q5otn`!i$FPEHt4-HgB18}F0{jEu~`{g-<mwc2sY
zLR{caF8}W3PV<mwi@xUIc*SvRV%ECxc3U=Bk&C9v!_O^R8&zuP#b$<TAqM)~yJOSc
z`Y)s=D~Sm`t-Jf@fN(0AaXgR@TGN_?yKF;LOs`3@S;d#0KpwbmI@VG8rr?6Lz&zML
zNp1;X&XK?Fap!yLaMg1y3G>P#?bLRvKg&Z-?L&?a&sx0ct>&zNi7{tk)+}H8{B~{)
zI;h<|X??UX-l);1x%**|_<YLAj()bFu<elwTBdj9$)8fzV(CcBBo5wV-wJHD?-TZv
zD$a8l_X}JRg=ICCDbG0y;H;)HF|#`8xtXkH2K}Kdojw0J7;wgQb4iNfJw*Mt4%@kT
z+w1*$L1*LBdr)YCx`Tn6Nk+iopSZW>>rX$*uNgB%-bEV-g@d7+iOEA(TD~CXKhqek
zQZ^FS`BTXfna+16DA8WD$cP7Y^uF({sFCPZp&wxjQn11lZPG=BgWLu4C8-Uwis2MK
z<7kzH5a+V2tweMO{{6PNOYc**+MTvBpS62caOs+q@qAmIid+?WY$k;98tsTs(i(7c
zc%siiP|#(WkhIC}S-tpHdq#@xtD5)ZMFGbuXp8@P`hB@9u=j|c?`(3@*?@N1aEL5f
z>}Z~&hGOW!%g)th?~zumjT3jQT^1<tX=cujS6=c?4GRIzH%BT1vA+DO?ZSMSxRBA#
zie_Iqv%(|h2ea9=$%;(|klK^W=Cs3VQSqX`JMXgwv8$X(N$PP{Fn-8c<+Lo;N`G5K
zyk(SJRlpgEZMw78@^WA2TfGWbV`O`aH?V?n$d>wzv4S-nV@XFY+Sl)-!wZa$(Uz?<
z20^cQAk1!&@C%+z-QJGj4HNQ{Sl!i8g_F=oUj`T@46nVUZq}rGLw;u!GIGwd_L(G4
zS2x{)&DHLscP7wJ#Pg6Qja!g~MeO@+zx{3$nY)|r<^Ptx-0Hu&_DAI<8OwD$I5lq{
zbBZG*4=alq8+T6d?PyN$?lZyAFQ;NmREz`qeG_z;`L`o3=*7_o8rWa!w}DWLvLm`{
z<B94@#HRdk#$E<TfBRt8`>eG0q-cGI;Aps;GR-+HiC$j?bw#no8a`U3zb{Rr)t66n
zG0_4fuV#a?%W#4x{Ce4B?QRta_7kBEi5t3Kv)t6QVo0=9-1>FCm!Gd4Yi7U7-9tL?
z3`nao$B}^fFd&Wif-&)J7a|MtcQH5R_v@(rthjpUtw%&FPg^vf(@v)?%`yU`t5l{7
z=nB33x=T9!H<s{s<@Jn<gs6>CPaLjs1v5H6H6UTTLN9pkpRbcou){xFn1m6EFUbaO
z=>Y>-G-g5%!6bY3S)(fgpHKE(F+K)&z3>nXk2qR2T$fiTL~U&e2mAe-b&JLpG%+ot
z;9mlFSy@=XgTDsHzDNh$G)f<>+?v`}zo@?<s<k)tNZZT%oIJz%Vo?8M%TA=U?+5RA
zuu94to%7Ov=ko8LU9U1}E=E3*rgyTgr|rPzqdBjri}XCDd^><=e4H=R62=$N?z4!Z
zH)+_O8gyEjWp*vkND1!fueFzMJgT9bs{8a1HW<IzaemaMmKk(%);fLpo`)zm>(Y8a
z@_0V5V6NNH{|_;5-~xt3t*`sSm1M22pwwxP*6k|Z85a+UVis)3*Mmb<Om7hV4jU0c
z-$Y(@U_7^BsSr3&v!6dp%g^wOJ2{~VZ3}Jn*m)kbSH`KO%zN{1s@e0tc2#v-LEe44
z81=;N5G+gM({hR1trx$1)++}ys$K)pzXYDd1^t;}YxTbwcy86t%$p^wXuf2zdy~y}
z2``wIyUCmn+PNCPajvP)Np7`Aoh8kN$)EfX?i6kf{G#{Pk;@{hZw?rU#My~-i|MpZ
z(D~cqbg+^FgNlKhTHca}TL+W`7c)1P*Y3OX=kZs52|joK&HS8a<!8<M_k|SCZ?!CS
z3UJt7dv}VHH6d841lxc4X3+m+a(?*g5<lqTDy{RC7jn*X^5s$fZjia}O0`U8{PUoD
zK3*Int}m;2{q-_lofJFZ7#~l%>;4rwXEO-cnW)gJZMliduuwq{sc+)D%0|>=`p)&d
z-lDr>)Z2e`uLI{#P3G)|HL4_nH>U$z>LMj8?&POpS@;aFq4lc4%(M}uL|_cKo&V<2
zFMaci??WJ;GU1XM{qu9=&PLH-G3__!z&)<4SntK}QGd&BQlf6M2@yMks^ia*m-Ws(
zbCuiYSGzAt&>~Ph<q_jkE?+Ngyo{swJ6|iir{a3OS4Y`y;BoOJ^Tt~AbJbrX(;cyk
z2lw0guQme~9|q2yzh$-O%_5HfVQPme`IFV{Fyz>HTxFPPBk4>F3#neGSWY(bhBgX_
zM-5ROWDmzJ_2tuwNS~?&dJWVVN*{=gnE=lyWGi4<V^NJ7=twvq`HtlB4gyi)%jiWA
z>CVj8XeJ!pW46|Vk#5@|5NVZ#&484ydLA;|Vtmrq$Be20tGg;Km5NTN{*!C-;oefL
z;5V#~_(9TRY)EOXJ9HegM8ATim}RZAMxJsc;lTdq4Jw~#J<wmq38jM#S2lQ~D#KGR
ztTD9zo6xZA`tYi_9~g+{TOPzf6e5Z@8ol~)_dAm^8x7a8R|5YDN2r4SVQr;Qv8A>C
z^a6XX&j2J*X5ooi5R3lS&EI%yWVYAPe_e_wPo7C#U)-Y@3-C65T-TLw0L^9%4tOI)
zI+kz{wXWKD6VGZd0yH>cI-OuhYE1jct5<mU!jh}#N-gLTe>&6uyqgkUv3qiTCVGBl
z`;KLeJR2|gPHXFAB{1r%gJ)yUz0c^SyQa{u--BDBe0`VeDOr(Eex3E!Wt{$q53d|1
zDqrsnYuvwTQUMv2v}A2w@@kLI!J3B+EV>>NNDPn9IhZ1v+3OfqG|{KDst@39NN?vB
z%Pg?D-!?EeKM8K3jVIY|^}Vbr{sVMF()UH$Pzog7I@X4VUR#OtAJ3N*i*f&;y=(sL
zg5G!mxwU(B;QW|}Pl~N>*-qxT!q(~;A?K#7cJilr;kaKE=zw>N>25XRd2@@$wF2xD
z-GP5`_kdytOO%BlzH&_YxyZU=++us?O)oB{zedelf#4N(u<&sbC*Ltivq^v<q~feV
zM&kQq{QnAAcQ@~izTd4Klgn&34)X5qD{EdotUry(k1O`ux5(^!JJ}6VLCMYLRGs8K
z02R9V_%40A7boUu?WKY0bJVdP+=awB&PA)NeIh@@D0^?c_$>C*)Ya}XxW^6$#<MUJ
zr0vRqq+Aj*I@tdRQ70Py9Cds9GBaSkx6jz<U{0y}a_i3$q^S5DXeWV~FbVSX<a&oj
zG>|~Uzymz>qpA8uOvWsXyP*LAoyB=Y6Bq4$^9WigQ+E4a+<EFr1l*fDcAICPgMo~Z
zBp+vG!8%IbX2_S?^|=hc#9Pi_<!6j|knxs4)`0-e(Z<SG)E=&%jj<k#Hm^$8JGn2P
z%#cQe#jd0CX!%GMa10H3N~sthzV|?xHP)XaZg1ZTo;88SRLbmDERPu>-eVbi-qKCq
zj7l*Hd;aLgF+S<azxX5y&epWT04880FJ})1sm+eUi1GQk{$2ESWRVO#Nx#ZH1&@~W
zV81%!7;lFOVw5Bpt?36^@V!DH_@ZWrn?cMA-nZ8I^J_?`Xe@72@~7u-u9H1=wffC2
zBC4?D-60Q3hCdT+VJ^G*tY<&dYh9B{HYx~@1!WO#l(X|p+)5z_*H9-hVt}uq;I2*2
zxPhA#;kn>AS&4CZUpdCJ+Cdy1q<DHTDWypWK5wrfmmfVa>4)4;vw<zkdLXm1l0R*)
z%MOFA+JhXnd+@yJP=Jf@TmffGB8QHf!2V~AIL1kUInA7K?NF~Iy(#KU-h*6<1<V;q
z29rb#n0uJnpa^4l!}Q$(313d|r_jW%J$P|kT3jY?J%#)ArmhT6QX$_A<y*)byWM|a
z3xFyKxk2+@;E+6Vl&Xc174a8taS%7&KCs|TF}t0YtVYCxd?k*t`>9P#>;0>~Qd`tV
z1rJaUDlORbOe7qb+YM1ge<47QBk@NbXMFw3c)9QVVNF)Jdij$2P+>x~ZRYUpEwyaX
z^5v$dp(qa-fYVPS)E|Jr^h%zx68fCfIJ`0Qg6IMG-T&acq|1UAgIlZ>tC^&hl-QCw
z00NZbWQejsW8}7GS7{F(ar~H5<&3G*sSX|QAmKyCm#h}NFz$LCZfBn>-OEGzYY7MJ
zwq=&t;N(w1*q?_Z&{eFgyj&J=y>@PK!x*2GM{3R%*C_f$p81y#XJY5XUe(>4v*r)Y
zv<OaH(0|>GvJRV#L)|mRK{%0Dtrb5t3Fx-IzorET(J2~dvSQi)<@)^Obyg3J>*&2j
zTMvp&zj4X_9dnOYXRAksZ)58(>AMGNN&5R^S|i`OlKnS->54<zUqB4Xmm4(LTbS=I
z4OwSOpD?u^6*is<fuw-@gf1Bv?3GDFwFBxmVlhSs@m#`{mcHTQOV-IFn;$Ff`vgta
zEdG({Nm|dljW_3s`{{%HpddSYV4MCVcpQ~XyCZQGQz?BgYaQe~eRMJ3-yNYI9&RgS
zIh$~>7@)|8u0>8Vu9bbFO4Ypf0Kl54HP45&E)^dakw1%5iZ}_{=!^G+K)xi;pxDVE
z=)y)ZBjCFApb)@Hb4J-AJvS*2zjbHwO~$I_rit?P%gU&E7mt}DkVERT6<t@>c53G^
zAX|oR_NC)-CbT!`GB`<%5+GdTjqsj?&2hQ9(QqO=g2svJ^@jJnNObf+^TFS2p$WXa
zDZ3f1VcYI306O?r1XrU&!8%+n-q1}-@-Sdmf3n=vKIjc~J8vy_gLNr~f*tB?Z%+C~
z;q>;=4@k8YYS#VucZ&x5&kD-vdB0=lvqYd&DrVUM?)v-5Kb70HYAdR4JO=jLy2>eO
zC}z=N!<e#Zvvn$b^*+_34SSSth1F01Fu>E<gSj&~n&g={Rq0s1OyJ~e8G0sSj+>*I
zEtzUGAl3$s{06hh;;vhsO4Ie~zdb=(3%f}@t#1jRs8~NXJ{5gyo%k8+#mdOI>v)*e
z_Ml<qI)&6_U70a~sn}L5zylIO37h*PO!M<i|M2XR0z15ehES|-8M5`>(5-(bNTg(|
z=oXIU&_|JFV_5?a1f&SOs0Jg%AnS0ag8cH0gxO&_I-7o1@UYT3Rv%iIw5IQ}c$jx}
z27q*a9A7}Z8XXYS1R6P<U^x5`>rWQ8ZBJuJ`T7$#L}>|J$OUdqjq>!}Ss!`;oxNYr
zB3jlv2Px<4QS|^$qIPqcbTl5g5X*LYs1c6yM^o`>@~4xOBAK!Vo1R9@>UC9-Wi!Tx
z&+$e*;ZCYfDBm(ya*Qo`Qea;1)oxB-J3hPmL%+Y5O$=3#v-(=Me0&HZpt<(#2Wv<e
z(ZpxIQ=#Q%X!?64UQ>A8*0U~#<HdmbOd5SRlUSt)dZ`wVN13u<<s6~Cw+qCiwmvJ&
zOn)ZQxkp3j8;sQ~@ff~_vE>0p-HBc=0&@}myk&5_*jrd;anm4|)ptEpVUh0;a6>QI
zogXg?S4a^zFJJbMss@=GUk>;_VZGU#maYVRlA+1SWdf!Kv3izgG@8epLR<3bC+Qx(
zhm&iAvuFf;`cN~xXI);79HJFV^r#v-?)cMDbUBr<!AVF|2w6pivQ_Zt`&mbbc+)$_
zcSob3XWcNW?!o4*(D3MFoE5W&u(n&QF{eMVBqPI&vSr2{d$1KeYMJWi!goNKGrv0+
zz$5m)k_aI~mPn~e+s!q2xr`x|PR1x8J`%5vP|eqfE`sek{tVz2&$cOLLq1w(OxUbE
z{RK0+Y)d`%FJQ^n@@gw(S+lx6O9}ivRg!r|ORt6pMFk2a`XEZxvL6CX?=DuXzgBR!
zXKY|+s^{I`Bsx*>U`?KBtK7_Ez-zCHle%&>@BrWm=$T0puqGtIR+NOkDc-Zgm?nX?
zcVLyQhc9@;#S=E&;JqeMs?!NH+FsA-g3;pk%^RMJWa@$!@_K%)Sx$xu9H}^GpFWqm
zTpdK4NR!Xv!|Q&^mScyXVNLm4F`1W)EgL{0^bE*$n|JjAb5ynkd134Y$uN>?`rYtH
zZMLsFNfus7N!V$I(rFi2lw28{^-{6@;?3W{UJq*brFrqd-vG#%PXriWw?6k>KjrwY
zKPf7^Fh6@TlSbKn9A>igdvWS*>%R0Dy}N;H;L)zhr3#G|8XnR@d)u2?L@8@UcgTey
zr1M|*EZ8VLq8ebWdp)Ryiu-vQWY|;kO-GK?ne}<=TZ_*x&OiEe|MsN85c=)~iA_I<
z@~-E;=zK^x-dgS&Fi}YgfE$G;FzjB6&}tl0YzRZjo6`j=RqB2B=f?9_yKp5~+3ob2
zsKD*ZIx*XxGT*ZwJ3;vr-N(9MI1`Y~ja)S5lvmD9zgSOlJ_2qU0+%bSTV6mBTu-d_
zhyOwz2rV0XVOFdIIz|VV3bHWg_-u)eMg*?wrN->N9|*uoKQ{XX4IKmmaDZWs6?X9E
z^2fj2Ba{VOXz0cy>uU+BkZS=>5a{Is1uFUf4TaIRqtp6h<u53{T*y(LREp(RQe!jS
zZ>3(!TaK-t>BW6a>ALJyjfw?z!&5qa0E^(3Lq{NNd@%56iMat}KxD`21}Vyuzp`GN
zS@h4dV~{oTlsE@t%8o}*QGg@Tks1K%Zn0r^PyR$XO)>i}tSKs4e!L8}Yd*b7O@c~q
zk`@<9ab_NG7cHBQ^p0_?HgNR^@xBqPAH}JqR=CnjI6Kiv@{9~6Fi7W_Q~oBa*_T{|
zXE(m<4%<rB@8m^`!C1rSu{?)r>Nnb@TyE$c>U56*0{iTSuB`5QT8<_hB=mjj8d*f+
zp5xSG3;rKy809!=lp1f8@=9d9T&eV=VY*TQThA@fs%&pBk2ro=QO;G*t)(vzoBKu`
zF@Tt`9r_RS5xq(+lh2=h$Mt>6C>6}=c^X0##BQ)or=xYPVk1{$<;&Hj=tW!(SHHif
z=O$t;tdj-6RlWE0=qv7q7~%<E_P>oq08$NyHT7M!2I!ckA%+u%W2U0BW6Ew0Hg6kd
zV*$6^R0Zfle}tbC%siMPJquKN_Cuq$w^!0hA6nb}I<*f2Kj=^xFDQwJ2-^YBXYBpY
zecNnomVpZMD2yM2_!|pWK_1P^0?ZZt0)kb&+_Yr^XdN6H<yN0RRu6-Q>lNm4O;Nrd
z_0eR{do>s%3Yw)(sQegEnVy&cf~^~Pz<BC)B(Cw<c4QixA8iFbHTgwohAL^^QGDL~
z$btFlS!<QCpZK|9X9M7LQY(j%*<kYOiVZyVB*C^^kA|xwWewcEoDPpae;G9xS^BEC
zRFIS>hovskZ$L`oENk#G<z#9i#i=WtCf49fk&K`M0L~S(Xva@VutfP%y%=)rN(1C9
zB={fC5bO8MgYZZBM;m~SAZ`DXlP3%&)r@?gfoGz;_f9_&^$GsXApdf<kp~PZd6EP`
zQ-8pe`5~M#XA<mLB~u$u!qiR;<N<Q|%XfDY)%ACzYy2;Z>*?~%yr)`~cf?K`L2AT{
zp@jA|k^A4FN?)w#h3KwI%;{=NqUAN%?4L$Ifidnum6@c1S<=2+0|Wau$l36F{i`ry
zXJGi5=nP-R!@GtQ>6y(o!wG3LLF7bB1y_~XHy$2o<8!xp#POgJ9Msfj`flxiSt)C5
z@jXG3t##lnL*H&FcH}uz(?X{|xi=6YY+(!qtn^neXlVFDimY2UDBB?wUb|t6^PxZ8
z&p?KBm^5lWMyeooVymXuc<Kv!=hEJZ$ReEj3;FpuL&LweD>Kc>6#;u73OK~OoPIV?
zYmIU=x=-7$S6z?h`{E6N$+s>J3>u2*iozU2l8sue-vD)W45^dDOg<4Pp`T^rZL*m!
zqZRXviwOXr%1k26LQdM+m33;r71ipV8B=)z;2(F32M(q5-^Zu7SKWO4&LVH9D_GV-
z?I|_DV2ot!Df&v9wkY3hH0fIeq4>U4@Mu<_5+U;JC{rjXpQLsR^7l{G-6KG1^^{3F
zWN#s2qFAjVTkb~I-h$|vSay5qy)$$h!wNPA_;1Ka><1b*70<rAt;^joXElykZ&IQS
zSL}*i#ArgIQO^|M2LK*uwzWzjhY1Jhe(-!H3*S<d)9Elp>2tTbBL-;t9StUhi5T9=
z=L4{ZYc^EUYl1H6@iDsp=&A<Lc{@HBFMmgoE}FpmsO5yzMLO%Pdbj|OEC>vK%gHlM
zD>eh5?564Gh;kaRRLuM5r?aV6<DvkoeN3i<?A$l}Cf!5&s`vg+q~UR7@CN3Vhtb=5
z^jhjUbr@GN)GZeuu`X3Ehh3^Xf<FLL$1Go8#o}&s(mqB{#C~z@RbL?diT}dQoU{M-
z6>A+P`LB0EHbX$ArwUDY25!39Vx(zYfM@D#L(+Uu*G)3syat_-CW$IH%+|BcGC&9E
z>3T82Y9Lj-U{QM%J|!E$hd!z90)}%<ei&(JfjUcU##mS=rwaw){ql4=s;vr<1+P|A
zGPZvFVRcF#l!B#DZ{;x_vFtftJy4>B$&2<R2y$L%(LBcoyo?wli`3uAR$2qr^pk!x
zNZrIK+_sL0(BOb~G(pQ=CmZDq%`ij_Bm-;`EN@-iH4He5ZJEpyR?C`i#8CqtSskxr
zsWGb80pKl%#<eF*i{T}IQUjO_S>=fG{bMmiShMWkI~P3IhrA-GocbImZ{4JhB+(rA
zq9V=K3=cPB+94rDC@CtvcR~?vrB4G^=Ty(btTc!`eG3X1D~-2I0|@vk^L#&blal$)
z2%0_rI@q9YWv&e{<y`o)-(v|?&0F5Ipk%ix^@d1tf@mKnPq}gK$J4+LjDL_$GLlQ$
z+xw&ytZ?SLh93=&(HSvJXc5>!SMoarp|xKLRU~Lz(z%(F{yq-}X6`#Ksj~1lOD(MV
zdbWr02^t8;)s<iG-oNmjQ2kL}Q5tUg@BP2R8=1G0KaJY#6iI)OBL8mudp>ZHOyaC;
zXVTK*Ii>$KX^rCGLzuB{hT3)KjrW9t?(u`d6xs4L3cAqnV#kgzpIImkh1l*^_xT(c
z@*b6PQ<F<_l@92=?9RP@cC#g#j#{^x2qwDzJMz4mTpv@p*>WkB9}P;IhH!o*MDeF;
zF&f;~E#szU$sm6wCmcR&wbYErR^&i}9s)9)yVvDWt{=(2X!nKMHRWdl4FeL5mL}Md
z_`LqQMuZZ<ymzY)nq&S1WKM)V7h8(%ksqmAEK2HO%qzzK_Z+{z8%xxRS|dxHF5F!V
zZZ&iS;@RuGEPc|5Yw4T)jh>L|!OE2}#;AK$ygxzah$+SafnVS7*l_4^1CQ8qXiP8i
zK+i3Ux}%h(Zj}gd4%0ya?~f{ph&86;<*A%_8gN~nh=$sPxEi0ew;W`R6D!d4C-FSL
zY2o?e2;h#;P*kmaKD^qQ?7s~P<H&-V|1BRi%BCRFV=$qw{pztZizYB|`6wqPA!v?F
zF;NtJ_vX#^SctY9yfHpBNgS#*b_b+bWIRb0x1O6E&37DReM9bMnPlk^I{nui%&#)0
z{&XU)VSB4bqIny7XUo7t=P-xQ?VtB4;EgAxCmZ2Fn*k8;$>)t1`tM$VzJ{Y$M0r#y
zHp1a~@8uk#J*oals1Oy4cB7ARVMYH|nbM}n7tm0pxML{MXx1h|bCd!z^^0Pz;P6`%
z*f$ov4eBevUy<BV#60!)vg;pc^#h`)@-c;Runp4L5_%d?VssY}YgGs+9dnLMM*$wI
zLM}1XKE||t&+KF<p^6O&FhiTo_c|GZixypC7$FDJK17I_<!kyc$<B@fFtVmjawu1_
z>5wdqGGc(O5_A$|6-Wc)<%J!X?HF??;3IaH{5Ha~IA*}MHS&;IjKOrajvbD0QVgXK
z7D}PYa$=QF%YOAs&mT9^wOrrBDP)0elWWqLYs(SnQ%=0r2(M(2b<Js;1ng=m#h_UZ
zkJxa}cTFJ-h#ts^mwV?aa~_@PnRS(|lzhz&Hs&LnIg-$Alb|RHW2#uvY(QR>yiyO>
zR3)!oe#*hpn{W`p-BNWW*Ai4|gTnZ4G}il{z7N+$(|`J5H2ybUQezgYx%|^7!+G+Q
z-LfyyhWo@*NE?%<egav~o;?(Z;GL*m2CP72Swv>$W$Wd6`{U_>k^sQ=onBveJ!3*}
zn{Hoy0;!GKbnFhu?sZU)q3|#6E{-VunAM;97VHAhBLxZ(9x=#Q`*u6*jqt>BtD!U8
zKc4$fFMH#)m^`t$TQapmr;G7P7to7Hrg;bbZZYzh0l4vklcm{oLYid01)wAxhdfEM
z6#y&(pXvu1%HN=E#5OW4(#08-WP_@YQvkLK@IYZq8Q_du=3A$$*LQBVzm#i)qp4;s
z3Z90CmH#{~S2kp^eEGPCr`|Ve?VT55z*gwxJX<YY)9Q*~N%L9I+MhS_U3v0vy;T>A
z`ghMZ!$CXYq&r!@yBHsU`d|g(w%0QWRb%1S9SQDKF<$wpDUMqX$Dhb0VwM0p7(yBr
z1RkGrK-@W;ei&ODe?X&7tSn*{A!@f-oLk5Fm18o90X?|m1RRj^e97DmZck(CYskd;
zt&9uM*o8fU5sn!6q;2T-*D?slo|F(^488Jo)v?SEbAuJhvWiy5O77Z<pbO%FS!Ie!
z5<v5w_qyUc0rN2ih$O&mjE&GA1*)cMt#?32UAGzWG|Z8(@fe18frtSEA<BG~wPC?b
zQSw9j^g!C@QY$VCuK|_OdpmG<|LT)6Fe+M&S-O9w@ET_*AxbP&?gu(E*;ta!*1J+E
zGTf$Q4VJGfdwzuqmLG>aNLwsG7fM4o16F3qMv>VXvmwP+HWToZPXYJhe(SBvQt5ff
zb+3RNX~!5wCQfmpqil~BU*8T>g$82ahQ(MUSeh&Np6frUc{rW(c}y06UBZg^wgw3H
z%+1i^UY`2oPWR6&)?q``%I(|@U~dB@VU((0k$*IYM`UtG0q%)DB{fJ;Ft?aZ1sA}?
z=o?aoCV#53sV1P0+G&{?ECIn6OV6WAYh^e?mM!c*G1`rqFgF)UJK=YmM~q1ZS73Bz
z5Q#>iThG1OD_<W2)*xAWJEq=Qk%91^Bqh*lN0bi6Q*rJaC4D!km^02OH2l;B<|Rd8
zJh4S?)qOOZCvPp0KBvGI2Qn4An+AyG^?<IA&8|=A+|L)L=E2a*_=k&amP2<c)@7~B
z)ML(6y9Yn2M8)b;vco?{hW(ntDSg>COxq7*0eZKX2;BQ67_?h@<Y;{-1#}_vn-7O+
zxRd5REu>xzd=IM)XX57^=S33hQ)YebVlki?%)+Y!yt&PQrw}!W1^X@YjXSB{p9EKn
z|K4nFI)ArFD!Z!fKVSdKpII^bp<5_74UJW|PPT?;aJno}N&46CSNJeKTzAFLw=2LH
z`7RAR#T^W(s-^_vIQnW-q6R$2LR~#r??Cb5lN54r4BWcDt00_ogk~cM+{MC)OelPR
zJ3kbUtr%C0K=PM;b2QhUUg9aO;(Y*YyT^bjVM;+gU5@-u-Lg^gyq(ZSOdkDo6II%P
z7+2-@=V63CgS}zwGzS8j;&UYL<)5J=J0p#xtAQK8ygL<ptFsVYl6>92R)6Khhw12z
zRF^cwKUHfpY2JE}KBiEvsFZ$u4eAB>FjU_|ly>mQz)eGBwRD`fl6n-4RbEX~dVai@
zBx_j@RhD5Y7{~&6vO+%lp1`-xT6u3O5mUP}59=7QnKEM`N8hGSPwj4W4tOl-n!@2y
z3H(jd<hrtRpE8U8+X}iR<g8yOb-0HGD(D!oZo_xil%5>$$+$2hLpOv{D~q*oYTx6t
zBgS^4yMsnr1Oer+WFyNMv!_cuV;c$td>g8-(P4}lcs%d5jZ;qV4SGMr_&|{1!ji3?
zTfE4sORLwAHYNof8uO9J(>d|fm*a9Y$1qr<a?3AIoEpX7$y<Xe79A8-tu2O;_OH-Z
zgc}(KLY%|fKZUWY3bmIQJmPW)-pZ~AVKq?>h90HX7*hkeim={r;NOC8lRtItrx|#F
z%?$-K2b?pi5gC;dEDZ#86iYt}*g&d0JF&Dy{Q(2K-roc@F$h|<Rq&MGUQL}-i8kn5
z+DS+FFHW=<{79e)%agCW>d-giuJ564|3Cp4!cpP)Mg3qh<Y6Q+kC^)lTT}^W(4Dr#
zk}JZhb>1%9uayA`<}~dO^WY$2oA)$4kSs#$<cpscFHgNBN8NJrPn4;{mrLKA>4ZK-
zD8|gi-sWB`A<fujAQPsO1c|a|u6$kXB`*~UtP^90If|!;M{GJqw<x<lI&r>5;FBvB
zy5O@94A|bZUYvP0uITCLVBo=%Mn^I3$pJzElkHGU6LF^%N?+<N=Zt|tEt{TRO|_6x
zHDL2QLMg%alDT!}=h4G1{e9MXdHge=n|-%V3&{pb>@n`-`I^p$!gdiFRM1WfRKKAE
z=2H1`5k+6-+i96?Z4h;KE;>Y-EZW1+NyxTt9W-iPjIZOLM2r8D1&cm$`y(EL>&;3K
z$f%~EsfMgx9{ZFgAH}<EPg(QMJStu`(FlrUvA)YwUqeS_y&~=UTizWs&XoMdUhBVI
z$t<TT>9XW+@W5w>^Di{8sUn;_&kmI8Sk@Ytst7*Q%8lKcrZB$aR7GL=4avp06rDqy
zJDbq={gE07rcZVw>#Vv@GAL#@_EUZt1mbD9!P#Roink*T@9>C`BFmSJyexG~e9~TZ
zAXUEh8<+}L*gReF{&QbF+~|<kd<!*YgBkTxZ*SO5$oisFCxu9d6?B=q%JJ&7A{`Aq
zbiUTyvj)yWr>`0<t|?3AttJv0!$900Sz%kdY=^tilL$=cHHEQ4uNo{;Erev~ea%n&
zG#8({K>s<j30${6xYEXqVF`utbPwe^0*z2u6!m2m9P2+3<CGn7p(Wg|rX?iDkXZi;
z<<{d4y5H8+ZevdMcurR+M_k|__R+1Bbq`FyHHy&TUs4E@Pve31>Dq6vAH!Z}S^-y7
z(0Wpsf!pdt?CJ*4SIV+Yb#$VE2Y1lI{u&4T4+;2fNaVHn2?Kuer%hZ&omnF54kW_;
z%+Mvx6Xs4a%MOPVc1!L(!Fqwx;YpSz8c2weppJp@=a1I73#ea<&%7}H(EDpj{v!qQ
z!^=g_sFi#c%FmZfmZ-DYM=2EX+coZbfd~86jQn;lLfN+^_dZa6OHbyh7x?9fBL77e
z^P&xD|7|<bW_rat-4iBAVVul_)IRC3C|^c?Fx7C^yJi=^RT!m^)R|-urPPlF1WQ8;
zMjD|Z$N{IiM?3f(GpgDfvk25N6RVi<0yeIl`q!K@tX?WtKqgW~1GjIn*1oSpH~;Kw
z=(+I<rWvL)khB}#Xy_T)^0jVyNW=(P8~2#Q6OS`|pdeOP68G14yv$C+i28z2N0BnR
zQg6}I@NT1{;3EkbE5fz|L<ozcbRPOw*`2v@I#FC<HQ8P{p4-SDI8dH?)9!4Ry1ct$
z#tsi&=PvotfJV-qVmnLDBYVtH_TQ3wBs}#`72J6vn&MZug)CTk3_W<d>6USGT%LHd
zhOlEH299N2tCw#HEhay@Rah!V{W`@U$_Z=^!J9EKs^t+M*V@P*uJzK`2nF{#alBsA
z0k!Ryp9y_Nbkrlnc~eMta!CbRy2c%XoL*#~?0+T&B58F#%v^}_Y4qZU&=<o<cS(Fl
zH&=(U{*}(9b%LB1h2h%xP#!Un#cY$ex8QTU)>)Ia>qSa?+|||ISnTj?LZ$oF%ZpAO
z!%nD`cV-A)ShL$ST}kzzEW*<|ST?a4C~k*GtmPmvOPowSI#sDoe*E=?zKu@QN*oiv
z-6Br|sKe32b40H$Rxb;Ta>bc#GXig=T{6~T{xE{8L6AHHfaVwYPhwDirrYr9f`&#}
z)7Ia28vRH_i9BVy&od-xb{;>Y+Z<^fcaHqe1VP6Rs{TpJY}Q}SL-l0CxD(IEI?jA3
zLwi)2)~6DtEqt@Sf&xw$+z3+!&3wy@7wBvrPL;+N+4W$~%Ja#&5}pQVacwV+0u4qn
z5)M5zpxOH%erq&GN~?PPwPbrT5IfF0#vHDVra6H{3JR2Jc7_9HiqMSI33~we{ilTc
zRUcxbiH8ypPnj}SaVg|;d>nYq`HQ}9xA~G$r1$NW+V)YT+w;emH*rnNCc+T0yAm;~
zLX@2oMn$y2_mGjCGqK)4*VK>~bjd-i$`KmF#L#+&juKIZ>g9X=xy+nyS5Ejm4K>=o
zxbr@eBkyt7bMn+X2m>ufQy*~tavTz-i~1wfDYPU@_J+mUGwbU;WIeYK_<50Sln~Y_
z^zK8KiS>J0^q96NE5Xz+pg(Ra_rWeI)|@)rU?xI?lmou_xDoFp*(fUJbi@9&hoRdY
zNF!acbWO4mR(51@c;Oe0*$P9C2vX!d>UTh;ECy~ZCvD{V<zO9&Ir%-$kTzuI2A+D?
zdgHuLfER6GF=L^pd8Rn>-Zo~HgoNBG3-+|}L&HFWi=q*DAq$*eh{3jQ8R@fD2=EB2
zWmf{?v7F7LBAZTzyxKTX5^xU_9QeS=sl(hGMjKQowFVop0{II)V<%V`5T@8TF(<so
z^S1d>i8+8|aL_z54=)h3La{Ixl{*gQiWh~$HorTKgK=PR+ml0E6wBiM<@oE9fdozk
zke4cnbu7g<Ww&G#p=zynfdpXlQSNbb6`8t11Mo0=%G_CGn#*JbLKeO(kg&jAf4VE9
zoor+`{|KS1J#x!>sPo*Yv%ILZswQfzo0f0|Luh36^$X05z`J*!X8=a4&~HIAoEjSQ
z<!b}>wg%WaK>d9Rbg`V$)hy1Yi_AQeJzu&yIGaR#EpaaZZ5>ZB-jCkMJDN|xCVH5w
zD;^)le|Ove|J!m8HkSQA+Q*OHnEXG71OC5>n}Y>|%iw4%CxvGo;T90El4tScexe_J
zi}hIb<Uq5R^@}KsEi;og`1Mqz9SwFlHY77igOU+KM6Fx!xqFf$x4852K_Z0K#q*)A
zvw*CumNs3D>|hzbWGl^@4w9JVl|Zn;4=eVHUMNY>+hpb_4Qi%T?PJ|fV{HUqz+OOR
zK1T8I8Vh=j&|g-Nq-c3L1jHa?zDBPG7i^+Y$w|zG#ly+loTAK$Q;3szb~fEq6&OM+
zLp9P`B(qov9?IjyV?XHJPyi<^kWSi(t2W@U1UuZ8l2O<-7=7A;45@d|n~J>fgoagb
z6bHs^C@VxM!Wk(>ZD2ae6!OjkUqf>FB!3k17Nw8;E<>=$m;11g97AAVI3b=P#8$-t
z^hE=9ia0P1N=6PVnl^`?*+h%*u;koCLfuEQILbkAOD4`pp_RDG>pZ6LW%TKe#CHV+
z7^Y(1G~zRlo>?b>mu$h|B8<`O-4RWUbsg0bA;gu*)qJ-#<i19ak;;ZAY`fA}ov9d5
z^uv+dJx%AkP1LPSOyTp4DO>*WKk(xd-Ar&QLR|QwlNpI4DhGErIGjw~){AW>jb{bs
zcT-QejVAi>q{?dZ5-XEFyYCzBiLAK^L@0JNF6SnCVs{@$DyVug8(@u>jcD<8VUN6J
z@z9lVbg8_HGKY^iF0@;&@#4bVlKu8gU<OLlBx_InpKKVtW_CzdpVl6fN=aNFZ#N+t
zeRm$A+&ce<n@hm@7l%fS>o>Kh=a)|3KdWL(dVM%(rIkx2)+l4v_Gi2{^EAV1izWv6
zY1E^}vel-K#J4@5E@33Amq4=@?hGk{SnjT#`ORINKny3%eGyU@Wj<2F1p0I4A>)tA
zsK3{SZi?I3ht=<BB%#U)`q_J-OBGRa8=ifO6{{Fz<}LFwwRXRze+W6{TRv7Y#K{!M
zcc@SbBxN9R2CL*|x1CR4F!yEWdkBIv;lrCr?k*0<MJVwB^w=vR{pAqE1>)t-s(Q<&
ztKLpP`L4e5O9T9~D-IftpmT0)sgy%?6Wni2uR7}qOo@UIauR(xtTY_OASrEIR!hnd
zB0QKKzok5zJ&#1$)p*C$$!`aKRenZ&5@b%!rSfc}94Fq(pP%a*KXWG)Wwt%xSj$bc
zVo^e1$xoV1uy1P^ic%X!{@x$8;Fdd?!1jx!PyQ5#qoUw{e`(>ulSz$&mFw}7NeWq7
zf=Tx;df~;K&<IOyM(wwghbMPDx}a0kCzd9{FL>6c{f!&*9y$oTNvxb=tF*sI%jeMi
zrCTRVs(YYNAwb-+xrYl%O|MRr>SJLp=WUKqA4(;Y)xO%XB}yTh?0$n0Cuz?f65%EK
zYbHGZW##vG%D3_iVap7y-a#?pSgBDIrS`LbKWsCMW2gUV{?%DAve)Y(r2oz3d*mnz
z?{;ju+}{XRouez?!%S06TDR>t_Zwpd4t>YpKYMr<{GPlR&u)ezh|{)WC4^YU1he$c
zN4F9a8zB|-X}U)}cOIHZ!aQ9$@+|7)6n@u8#rLZU-<<%1aDn}g0zFCzHRJ~EQ>JG3
zL$ocYba!8EJ-D25Q<|=%E;qw_d9_O7QfSB_;Z>-HZPmErC;cS)waE=0PAL2BkZr;^
zd{{Ia&!PTZi3r9-B6jwe2Yi=B%7oar{XL)H#E*3qQY}@yY2lWbR1DdO&1lDci?2(9
zr(szOC;iP6&rr2?$yjJ`i%)~_8`h9xxZTeHZLyN+dxrY5FK!#86Lt^M1JW(FJZ3do
zxqbXSUVch$PVS+GhT*$O?ON`x&lmU93JR>#Sum@6wGU|zS=?K+;ILjy5$ckJ-`3qs
z;S3I1gs07Uxi-R2mxS62dIAD)KL%Bh|Ml}Vd5Nd75dm9K<6z{krXnOK=62;E_HX)j
z+|uiBG%GJh@q!+!d0eaMk_=f)d<^^b<ZRel>~AP$-XX5w!hE&fUA!S132CrxkXuqI
zH#k#yQI7TnWMXyyo>_=Zf?$#6(Iw_i+k+F@P<B#w=L^%GS?ybatQt#lcH-wg;r6z}
zsrWP;ZVEbw+$;>pH6MmhcEJ$xkO&z`!6VV#MmcO!n?7<mfA7f`GTcTq$BEhbD8yrS
z_%U^}I8x>8eKKag8TIZNK{~!GX$^@itdg7HO$a*={sN05h2{sUnD_pcR8T11Pyued
zb8n{95^oi3zOh_Y7W-tHN`eiXXzQUz=)Xgx#%q#fXO+iz3BFl_gJW~KyqpF{x~=Dv
z7dz^Vf{t&h;?O}Hu3lFjr-_#h3x0C}@}b6M!HMbjyvsrYC?o3%Gy0nxjiDT3#xktl
zX2Cj=6x%d(8bGy&dFm=eCf{a5-|-r!Lbl*p_^$-RSoDun(UGj>BDrcLf{vs>8uo9p
z)egRy%5h<-u-43uKl8M4Ody!6YLDHX9xmRbYIO(=nd=r(w>$th7d+RtjC#XiL-zij
z<T6lEY#cWbvRYAIWV>DgZ=jCw!pl;RRe-hlY9V19tk1b75PwAoCqA^xk5g`Kk)ZU!
z6QF=lVqr}o-Jz%Iw~c?iBr9Tn4G#ZpyzPBWEW+f?B%e=PR7}y)PJw^7V_uh8>&39F
zQf&x3{Wq!h<uHU@-gZIA`05d#c!PCGaQD4B;v@35JC``97UE8ay%8;+%=W~|RUgHl
z0Uu*gHRerY<RV;!$bQ6Q*spKrh$K!L*`I}NWQ->lMfKG0@dbw-2Y83|KPg(k|3${g
zbLkL^486j#{iOBL))05sp{-2qh?n3VGJb$*Vr1LugJGdc8q}xj5r@7RN7#jmf=oD;
zo{PxKKQdeOU@I3nW>9{C<?mCO6AeOQ!W!Acs(a<7V{H{#4{=`(9;BX#@`bQZuP4SI
z%EMQrS~Kq&!nBuOP_yLdhL4=-T96?bEZFchdkuo!9Yfb)G?`zBkp!l1)tD7$pf#R`
zcgpG8`eeh!=gDcHFilyl%9--eh)B&$TSMUunXaz9q#6p=O5vK7WH~%j!G#mn5MuH-
z?wca&frON<it2q5W%@yW-PbSLygL<G8H+6bdQ3f;Fo`hh%`E1|HjypQ9d7>=^iH9k
zO!KrIUN&kbD43)(j9EUtHs~(HC`J-ROxQSBFXdS{fXHx#k;sl<JS94tcnG0^RXT;*
zS^f^=iYg^=-?TBnfp*juwMppw^y07*T=o@#&ScPF|NUHP`ox$<FG5Xubddf~tnAK2
zg@Ce0`V(hNqx%f9Dw#HcBJv+~G~I8LRuy7@#eGceFk2Dl-)2D}6gyPzpOLq=J{w8V
ztFGRxG;DAD;Gu$8+jU>^=6jE?B@I#=H?Ma1f{nrtc)RrL7CGN+Xk#@yTV*y3ZyZb-
zB~0~mJIERzK160z?pwPTX>c=KrfnmN_J1tJeCs=Z5T%GS^nHOpdRw!%Vbzzskq~!m
zV5jff+>atF`bWswtMtUmB8{<6k%xH%afWm+BnvCv&xW0pTKc_gB1UFGji*gzgkKy;
zb~}9!T%733O;kC$B$M#`(IvPXko<CUw8&aHh9+6^Soi*KfnuVfr31+7g=AEb!~v}A
z+xK_Y+KoI69O2>I@H=BKpK&5LrVRxo%8HguzV5!zTYVBd=Z6R=E1=GNeO^D7&n9N`
zMSiqTdr_Kx@5t#iNg63`;&FL-7?3QhASqLqr(hhrF(ItLH{_779{DzwR9NA&pr)3r
zxj=frtu{KyeeETdxS~w&dxzm`+C;4+No%FYhVXJ+YH`Wqbj@CpZUPdW6l<o5qvFJb
zKWU7wWkcTYrgKLZ5qp-sN~|PUkbqneZ~HyKPJ)oPt3LTj&ctDb(RLrPR#SYOw6^sX
zKDHtdyiKut<W-|a0#&}pWY-?FVTG_hcgnP_^=;Y<d~ED4-8}!NWJIO?fOC;>u*ti)
z7cuQ2k@K9&kDZ<CiG8%vk`IdqmSYso+65mMSK+G<<8GQu=2kT-3ch!xEZ}qAUoSTT
z@$IWHfB40XA6$AkiqjnR?dosPZElGwjMSB4`vj_vnGtnncx5@clyMbB%gkAP`DT_O
z2kM8HM4nmsGVFIxeZE+yso4d#WVmmPpL)_v=u*2ABnaPnrhU@Dwa}a%5}FY!BdCIC
z{#Smyd*>4UzmFjQ-=9PNKZ~MYgD*{9FaK)vsQQ!87h*G3BJwPgrD9yM^k-^87YzGz
zajkM@95<vunPaobAjC5Y?7kqA91#!3jfvyKV6R|9=+%MU%oBF(y^v#?Ku4;_@q}1r
z=+k;6wOKv4KVUz&zp&W{+1Kn9DPsoT+lE%HF6L90X9b^|gr>YJcY-RwnV*&kZxQMx
zYp(jRMC6;SK$>XXL=l?35@oCrsYg`)kON30q~S*PYUYu8V|1()285?Z7~=;~730Y5
zxoItjJL$F~k0c*qF3E=9y7FJlGBu+IfW9fQva?T_9o|lJcuRa(6Y(>RO6acXldG94
z@4BgRl3=W00M&mV%xa!gMT&X6VD1D*jY)LUd%na{gCrt#OIvpcmD4$^3Yow#eM@Yv
zAE7BH|MU(-SeFJ=S8$Mg;vnQ%Hzx3VT^K(KV%`avaYg<v79$ZWOciDe>A{6<K<?8r
z%JxBCU^rm7WF%^Oh(gv1=Xd4SR+=AH2$Z%TN;5OU3|1$YvWQE-hA8pli66umQ*Q<?
zh^#oQ+8_Po%ylT7e>M{RnP)vm)HOEs9zUg<XjO4nN4`@H_ecp=8MEuJmVC2wuQ(jS
zmu%)3=_AY_0~Ih#_)%i#Cgu{a>1snw)OpRylHl5P1543j-@}!jhk*U!$|R>*@tR{l
zKSrBU(}j{?Sbf>XKRwzegOQlcB5^<$=%sajDKt0HA%{_e$OIZ&BF<rThYm4RomEhU
zM>@tM3yVzV3{&gRsG%Zfmx+>#E`so4xVTK5A=_{ezabFX=ji_zz~W<h76Ys2rrQ`#
z!o+8$LJK(7wHLDmGbgpZ9nSCjN1vyqyfB<C!yrO}(!}2z<N>0&|ABR`=jG2hrGsMI
z7kst6;OxD?6Xsv3TBKxoLQ>R!9ye19a#p1h%l3rhY9vqxB{}afe8Gc-L?;s`k?3X!
zi1YQ4$l{<Hh&c>ZMc*2Ilm9rv;KVZ^fIhe1Lnr@cFqZ4Ln@owCb_aKN2yp^Gtd}OR
z<x_A0@G7OpB%%R^?o(cTee29(j*r+4@3#?2v}U~5q|i@_F3YZKA5zt7U?h(?i3&4(
z8YdSnrLULeNt@37Ej|iFB$kkeIHuxZVS~Gu`x>ox<_~UJE-c}}o@`tbd`jjEfrwM_
zq(U~7iCH~oqn<oGXh>)F<eg>tvo<Ow{QmTpjge>6J8pTn@e5F@V6Sl9o^U3C9{Hv<
zN6qhg3_j<g)*_5~PBeEhODpuOQOEuf;iI+4sjk~5J7N%>@lBpWKcmQkc31TXhjvYU
zixR_-2aE>&(UtyBpeN#Qgd*>RAiC)(^1CTywdL^CowLtA6Pr^!X6e1KQ-0|A0qX%z
zh~6#rT82v3aUKB2qzIoOa5N%%b=BuuO-J_}f8eiMA~LI+HmM+YU~qmsemwS12)$-J
zyMez=Y>7QPT2z&|RQjr@1+kdh6MP{bo>AbtGRd;6vqTW~TDIAEn36Gve84U@)Cx(4
zI4H8od73=Y-<rgO3@BKm?z~UQ^e$-V9f$O+*V0o_Gf|KDj6j4IQ&BnG?UftC-q_dp
zIL;V<Vxc;takGP~SvS92qz85s*rb+TXqdCSRzJzHb)l<%KW)dmy3X?RNh^8^#SFPm
zNAlou3f005M2@lpH6;7~sjygMxV6Z$pB9v(fxXn@P4)pgqgRe^C3ekwHZuezPTYbm
z5^sNc#)9t|MI*X&p86P81dhZ+kg|+E?Ea<4&X329RkSv%roFNL6k~@C^XT5-bWYVU
zmO75lr{v$hH5bNJ)uTL4Gf!)oW1+%ybk+uw%6#}RYT0nNr&0luT{d3}i~AUs6F>Oz
ze71t5AUOK2iq$!+3{`4~Qb+wcZS2@Q<Gk<1iizAFBOLa7RNBFG<QYXwO2P);U{M(V
zSUMh_(0<_9>5qDIzjZ^frLx#TaCQJL*3o?TW!!5^FZyTk4N<B(9CE;nJn-U?pj6Kv
z`sg`MAv-c+>N3KeTfX!37z*2Vz_F{8`{_ql{f)Ol+@#Grq-&G_4%VDsOXV?D*HDE^
zAVWxjBDa$MdVTWkyk%TYu-|1@`RB#MKYtCT$dF<p?jf>dGnyY%)Z<^g#EjVWD_$`T
zPZT!WN0Do}I8xm5e`{|91ppt3%qfW!f1|pJWx)0Y_tHeI$2FQ^NJVg=;`LZ3!$$^I
z{agU_#WXZXv@)Y_KWtlHXRe~a>U=Y$D8u;nGgNPQMcz|LL>VVkNiKRrDOr9kCN`BU
zTaOa|$DIN#Ld-PU#d8S`IH(F3_&M41qFEJ<qLaBxsFq*GWRUzQuGJq$XaEHNB=6of
zqWqq)LiP28`0i(O5%O}W{%pY^88(5bCG&JHF!4Xc;IJl-(i|q%=YzZ;Ps~j<yKGF3
zm8L|2I8~ADH$E1h?3AyXu+zDxbwo%*;E19zrUXq95L5`$`BRxI3~`2(zVDCSwavZ4
z9C>J&u7UxJ4u47=>y%M>@UE#Mdx~8r14?QPqK}OV)5JcGCJA>V2MiuR8c8gx@HpHb
zlOry6p5FV(`~C2dbSo*T5Mr9*!QO}h%v4XvY?BKM3ts8GCVdiUXj=3Yo+m1<-I=1*
zEeo4siTnFy4wQPm@7Qa8Qk4~UGCTpPk8%Eya>@F#b^u3#W8||?P-#)i7lxCmIo>zh
zD-8CX!*Lzpz_lmNv97XnFWk#XO{|10kibPzSv^HbcQF8n{`oPiC6f`3!-NEf>zMWN
zM@E)*_7})&VhpDw-)FHt58+(5^dWjVn4&#0y%7<(HmbK$G(cV^0{cUOEHx&fD*9be
zl;KLA-ud_Y_YBY1*$f`6HBYf;I=O?X=d&0)?ki`uNWwx~DwRIG)Tq6+;O1iImMmXD
zr5-}@Q~4~rnZ&UtmzLcFm>6l`E+omA;&Hzhmu9qpGLLC>1LX5P)7f1A@Rt*-+l*w$
zWp%HBwv4SGWZ7GAITp-5wXenS@b8a%aPWt|5;OAhb5=!Ka#G{Tri1B(;;cVqbcr?S
zcaI%qCLL1$S&Z#zA_G?cf3<R5K}~J_J}6a069nlc^w2@N)DVg^DWNKz&_OX231T4$
zgx*4=6GOlV1Q4W4S5QzoNJm5gsnQknoSSpLx%c%x-21v`&zk?t-m}*FReirPs8YE%
zQvlBYyLfU}L_+-v4fr0he`05)UDmu-a%AI%Uf=#)M#G;IgFY7CMj^R#NDh(@NX7m3
z^h<EZw1sIOhI##KvUzW>C0N8$M2WU2i!cA4#cf4Sf<)z27BGz8Ta1H*sTYk;WQ|4-
z%hTWA%M{D5HP@%7K2M>rV5JyT_-}<v^jExeK_n0ZP!so8HEM*xGn}Nb=7!F2IRXNP
zmn+1GeAsFt+E`+r7G(spf=2R;9HT}=<#`W-gC4NzWnx~on$c}oOfwwcJAYhc7Zjn2
z1x|7=)zIc9=Y|OzBR&qx-GEe9*(Qov9BPM*O;-l*d2V)-;}xywrrCTS3UiQxlQRP!
zhhISuoQl}Zo!a1=4RrSI`L&8_Uz@aEN*hk@w}5EOf|T)rnxrBn_Pg%axx5P!EmCda
zYCag{B?V=6=-0`fq|2wc5!)_rhu94yTM%+pNkrs@Uu^skf?c0>q-5I0qcn2b$40hh
zhT?Su!=Uy?RD;qSmVfxXtMG|&0ajVk-zvKdM<-7Waqdl_Ik7PRGy&l(1`J+V6ASVZ
zj3mqypSQM8p3gHj;B5G=;uI6!hn?W~tgKbP+uA<Qza<ap7#Y`S1iWfRT!F$zqq(G)
z4t_zn`gd9wjl}7(xPe-7+Co|45c?%+u`-@+E6S-%A=7rK(m%k2{!FjNgiNZs)efnC
zu&V&}3`u?y3?P4_mTjecdF;IkKkw|A`eNwe9Keim%|7Mxe0(U#&2iiN1pkcJP%}yQ
z8|AuI<cwXvi_MnaPV3v$f>b%SsoPD@Z!vpRT2PADFM%=cHJYaB53E$r$whkG&g@CE
zf?d`>Xe1|Xw?|U(byNu<70)&CPdcyAHwdPuK3Fv1g6#kaugv7Zs)ZghK!#aL?d4q%
zh9Uv-j)-T>Ww%l)D&*yy7mwoVRyK6|x8(JNR=>?6#P7X&Yg$eVo4~)v8o;T!vj8ir
zJKLc)Wwzga%_s$F>e2pm)8hyAB^e(JUSF*2E85q6w14)e*1O5#jeTkvr%L)l+05FS
ze8~8d6W#i0)Zi4aurV7LcBN>qOF6FQ^K9y?AlrdiAYCDowY7I7XD{k{@=&v%j3@2r
z84(MqgN`12R#WYCv~K@M3dHoNNF&!@CoaKzOpKQe*29BaBTW5>k8ZF8iu&t(*Yu4d
zb^8o=w2{Vi7_UXrHmnPL1nAA(`y*#YA>!I0R#<V}qke_Dfc*=!zmF#6WPz-f#fAS{
z()m~1p(SqrPyI}MRco{rGq%3eAI3y=kGQq68cfmndKyR*+liN9A|_1^mMakJgQIf>
z36eVxbLj9oBxS+7QI^IqOhz+b1N-CFn}%vH8tZ^0*|Ojtf#UJnk6);i0QzZykt;|V
zprJ@s*SYlh@hOd%5`p6EEl7o03S2K4sG3kyb>4i%p95qdNw}<VZ8Et?OUwKE0t=IV
zf3*znqTN3uIGE9O8iJE}AP81{R^2OXG!_LACH9Bk|04*R#sA_P7HH&_*d=0rzBU^)
z%EviyX9uNlVDKD!`XT#dF{Qk_gVFvp(ppAj;6PF==nFcH3O}dF=5g68-=UCGi5)7Y
zC4Dd8BPT#xgM7}|p@~hG&yh!At=HI#Oy?2xcozdIj0@yyEqX^fjgD*T)k2#7;JA|p
z0zql>azSNza9vs0#0#qA-+#EgnErX*iWfbVO)T-vVqH@f38d4&sh4v_zF$sN5mD>8
z*}0Oh$olu5#4VjB2EgJ=V&vm{>|WQU122FZD^1Im%&CKb0g07rjF>TQf}{7D9SjS!
zL>oU!$JYh66e3cU9iHtzAHD%et_l<r_kuePdl>1|xK@U;6e&H+M-f<VAYd@&8wuPE
zc%c1#RUE5^@&9yu{i8qqr{n8?lkw_U?S7{)>#+XSkSmh{3>uy!GAOi6i(~zLgi^9l
zib)$wVI{-ik8bzy1uYDp_m^1iwjs?doFe?(*C8>zGSwAN)<5P<W!X-N0*|wEeaTbW
z!<xg<jo!^d3{^w<B~8(Cc48WDi~VFSr#2K!)K5_>*^W|PikgB4s(#`}ogHQ{EDe@N
zTbo3bN1+-jFnT>nD?Kc=#-bvGw@UzuV+bcGtP|Q8J^IFppU(K3=4!+Dj_*Yfq(@zb
zw7R0LUW71qJp{jzeT>RLaN~baL7XMXPtJt0Q0WQXuPVvKpXN&@aj?Xj&LN(Jitp4)
zAoK$r<sKw;vb`A=vf)X{zolBP?7%iKol?t>m>kan1&2fwx8Q0JR;q!@U#J1gZj2qn
z6)&5jizCKkybn;V+737>cdgZ`>*RY%Ne$pgo1m7H%{j(H$B6+R0FDjs1gs%ALE-8G
zeWT9E7vdNSEL_k?oP3R^)14C$JF7j+12`B^6FuIs=j^(Cac8n{xO@<g>M;IG&=xoL
zi>O`U$teq%zVuleg0T7rHqbvd#}~b-0-HuLgYyR}Y9j;@fXkO^pQ8p9Q0n%_+wqrU
z`A?)JG5&yN!}0iQh;#bcY<Ueas_DzDZt`@~OrP29g_3m(tdEce^}Xu*CmG%_2+Ded
zr;Vo(6%qwm>jRocx*bu1)>E7>#sJ#dn@FdjPM}#m;%|EhBE)HiZo(dJZ+Q1|=ga|e
z6xrdq>xeXMS)GV}6_e(!$UzM?EJwN@Qyv}M_r5Jv!(;FFXNEVjhCl}omkgi`Phx=)
zsq2CK0y1}f-r}J6BBnAvBfN$0H%j+>g^OUT5*eftaCll|Zop#FnXur$6xh{7{MB`G
z_7M5lW6Gj54}22DBGB2f>jPB%MfMVoUpUJWQqfQ-tGd)I8G+%;IJwu-=NEn*SkJAj
zNxIy6L{TF1_YK<*t148UDh7+tczM$7t{_6t%E9#5hasHtVC|WDi!j%|lhOK3A1T<~
zM>60+_j*^ykvyPRcnQ;dO~vv{8AZO|p_nSa5U2bqk6+F;#@fr08*64Kv+=rDC1o-w
zBik<sJyuN)quqjN!RFE2%(~Ijqy49hGhzAPY5*GkEu{|qKf+!uA`jc#2iH>|aJF<;
zL%%A6rvaRn44T7{T@=x6xatN<)kh$(l48ErBAxnKhX@p>*<|#07l^Tqi#M!Ozh$i&
zjw{Z`B5*oh=(@za)35I1nLqrT*1{4U%GrB#7~NS!S3iPcI`&8Sgyuj$bR_O1*fP-~
zlM{xv`hJa*iiF<d>5B}dPa&#lB&2TdgL^Ux!^f{bY?N{D$rbDe1h-x>ak;9+#F{Tu
zGSV(-vw-HvRNwlHiQo&D@}j4SPFxGc)T4{cfqe^Ep_;04tzPG9ib7sz^O|}3aWk&-
z=RPD%b5S4qTdp}|@?vps1lc&B>3K_67&}JMc0D%t?r!uVAE!XdFSVVQc|Atk=1oXN
z&7@T43nKapE2;DC=4Xf7uNijLsuxFvEAmTj)lILsT?sH4=B3waG*bPlDnnC-ua>q+
zd_eBbctJ=3&8mYZK4jX}g%wO7s#{Ie!6hTKa$9WLxB|FHJU+5IDZG4RsqwrZU9wz`
zDcrAb5d$Qtd=;PdgifSZ1*;ZYRt>#Y-2I_uEDn$;LbGE{leOYIKP>zyl(}G|JI5hH
zLML`Er8LW0u6KTQs{8R{>nhaAx!S-eDGWAfZX=gT)M9GY`FUdy<m*sMs;E<9itzW#
zgIssfFve?Jl;1I47%WMr%%i7k_P>|qP-5yN*hG*c-)WZ0E~_d5^sRHXu78TGamMFW
z7t0E)d}q4P-I6#H5>j^a6A5g?0;#!ufA`ql9MOU7zZkn(nvYk${5d#$>`IYb8Gj1{
z00QftUitW8%L_fj8jqta=SKawt0qTSdiDAWjaQC?H|G7i4kd^dYO{-p4AduHt3C*U
zAhHP(Sbw|{h(st=&hvptIR)ys$G!Sj!$6I2sHNRi0{4)gkb$(Ity15c@hEYaL~2l|
z{>^&qkKN-B01aV`za$)=ZzI&D_=WV(Tcak<H&-Bfsh;CQA?<s5JECSrQzA?LS+vAw
zq8P7oDxCdEQc~kBTD}=SrdG@mkI<DDGykDOI{Xr-*pQe}nS**msdk%P0NK#j3Kjj1
zYPhJNtaj(17#IteeExW{avoK>$KNBTJWU2>Hox=xY9M#1AVjjPiW$5L@;I_XO9eCy
zKwVb{abgy+4p2MHFXbKB6&D&jakAd^<EYhQyU76>Lql=Am0U`|k}Te+;Pdx^MI8Ee
z9AH|wy_~y&j5i8*6Hc+sh@|A$vG9oLBfv1$F1qkO0f=@QQPm{aTRXC>s-9G5Fo2WD
z+)e?Wv-kvkKu>vWWqffwtcWItZ@O$Og)Wqw#DE=`N9p^IX~DM}_-jI6+AIN<4KS2*
zNcPoaby1A}#rN#R;u>vMq8Ffkamn&bIsQs#aYp_h#S<K$gWn!VzcQ=0rSWWl!UQYm
z0E2m3jh_dqDfHLcFDK_K&(MIOY+C{u&z1zl;-C}V_P&Vh6{p=Nm0{#+Dc5rzLWhph
zN8Y!MTmpQv(#`!r(mc44NO65RH%;1y(6mi#W0Ef`d}`|eCw!G-*6n_#ED}L}-~YxD
zb8logt0SAI83GeEG+vAd(0uKd3cR)>G2!JO5Dc>%unekxWCHUup3lOMtxi0uzH9_G
zV~`-&joRc{`2Nzk^AkSsmx$Ji>7eL{lr?UPIrEbJV86gxY$>qp3Cz9=fIj`Gd-j3_
zoUc=G#ARVW+~6;49Kb2Ea8Du3qPI!@DjWSX9uppuFcEsJlyffp0oSf5w<VlgF`|t)
zFt-p1I|K~#CE3tiqKrb%?^>P17D8d*vCRdeg-Z3Azlk)a@mq)K6uYe#9ARh%z5w|e
z)<z4nuwSkKsq#oBx!{r*OQ@HB7Pyn)Nk6ovfT2lIA~vacLlfKL5-3Xo@W~ZUDHx<w
zFw+_aff?|2<l$lxol}^@-xl0(8|Qm(i!<H8NYzoGGY?zQt&@)v68B?^J)!tMy5+h)
zKLXn(kca<C>kf=22K=VbjX8fzPd8DWI`?Pmyg9GJ?}k$ro0P@qit=|ACkf#alnGmv
z%?h4grCW?8a!##L42pb=SSf5SjC-o|g7_swGGhDWN4jW7BEtbs)fICdx11^RUsq`u
zR#)Ngd}bR+{8}EZ(~6UGd-V5DJW9PsqV-Pka6q}<`^!J)(wGPIs?=i?Eq4Pco2B(C
z#tz#;<L}5xLtc@3*1J^B^O5>yS9e@lx?UkNL;hvtlF)he93&7|sr2DBD@Qqp!$Mzt
zLoiVFTf*;lQ1^LN(|J<<JCAl7#dpb#*FLcGv!t6L89!BGCa<pI*Ul`_`<9f83kJKc
zf^>bG@zaxt$s2`?cbSy4)M?J{V6%}JO{5uh*&E&Js{(KQ*|Ve}D?dzxQQZ~7Aw6P*
z@xa7@TWgtiY-?{a(uZ7z5?3!QoL(zKT2A@}>YlWXj&}!@)|1il4Z<gOiPq=49irtf
zX>?|lT5)1=g%A8)Mus`=n&d7PRIEo>@_EM4Cl6Yl^T(#9sEFVwF&Wp^Wq15mwDY-j
z-{}-g+56ekKQCY~?3W1oiEAvm<smY=vhgq<hA`sXj@~V*1~{l(Pc(OS{z@A>Ey3O~
z<(P?6R2-_c0-O^~k?Nz<mc~%xYoUy)i%bGK&Dk5;8wPa`Dr2Sx>X;eUM^92MFhET(
z?0k?Pr~J=J`=g<`!sS1~Y%z@(AcK7xQ>@4{w5{2dgcTnj>d<gawv&l5(AG?yX+t~X
z8{IQMx8At*{&Q0uTw@OH?71JrHUj8boi;5uJ7#wJZG~h=GgLUbN`A68)ZeWX4UpEF
z%G+q4R2d555x8Cr`n>2k%A>8;?xq^yI?Fq1$bgn~mIX*L=8<XJw8-tyPV|QE@JY;`
z<3d8#tc1#r(c4#s#2v#+mHl-+6W`^Lx)H5qcDj@Dzq^Q5a5@z-ep|4zVs<Uxb?nZ(
zg+P?x>eS4=6E{@#GV!dDo=6-BK9!X{Y|(2!ldVvjob$y7*!`KUBFt_~9sIZfDGJ_z
zzSmIey*H5CNHU(eXO<FvGGm>Jyd%EtFq>Zi*E$P0_b#FNBsJrcuUk%2h2T$;R!xO;
z!O#g3Jg}tFyx6)nq1OD@VvG&*eL?w6<T&wm19O`Dlt!q*LV{F{VcW^W7k;$}&(8L<
zDhVo8xw?QBykNGu$zx8Ct{*alzr4F!aFx$%oO;#fSM2Vy!1;Vop^5qPFw0Wli@HhI
zqzsxj`Bwz5EPwshjQa)_=MDSLuO3MGR@Y?sD5)vHw8O;6paFMW;<yI%!P>jFt>1$n
z-tM*M?K*276S1y>^tDE2dBgizlYv!1=~z@m%lio1DOYfu)_v>5kq0H-8);n8A>DG$
zRZN=#f|ZkYqpPEOpB&9UJ%X=D{PC^-3X&b)5lk1FtvRR8*4KBMQ~Igq4#WNXo9h-a
z@|3+MuXjpS<@-w^m)B#2`mzK3Cn97kHsPcEqC8W8i7!LC)U+k|+N8kW5dXrtxK_7?
z7|(;Zb5A1XzUIH!<LS(Mf)zY_f0%DSNqq1rPC%W~7C~=uXQHU6qR*uQ+<*QlK6!-O
zL2m+i3%e-x%(=n6^C0N>*G#jez({p`U}e1E_@ocRm-u2v60YI=czd`c@)bUH%xpxH
z0~K8T`ZVhG&C-QFK6L*_DP;+63y0?0&y{u$h)s4sp%uv}>-F#ORULu3(1j2e-B7_h
z7W|_vy9-vgbxfU6S~3j72fP98Ruk>zK+qQl+J3t8$lVcrGIwb+nuQrm)*F!X-!p?T
zNy_~Hw^W;+F8AEW9FEjapq^@<Rt<eObf&J(?Emh_A(u_F+?Yr9m-Wy}DdBWmkq?~Y
zWvp;<Q#1$YzungvqgZbk=Fo3wEh+?Xpxdc%JYz!FrJ*1W{<F+YE6C1pq7{7pW3T)l
xbp0=;{@;N0zcdX^CjLJ)X#a-;dhpjH!%ru#zbM@Y(|XAPH;k-c)liSP{{p%qW$OR{

diff --git a/docs_old/logos/nvidia.png b/docs_old/logos/nvidia.png
deleted file mode 100644
index 22f9e9edd013bd071fd3091bba4221a48ad1ea10..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 15146
zcmch8g;$hcv^Fus&?z9@NOz~SfJh^al+xV;3|-Ql3P{TUQbTu&h;$6y-5uZXyZ2A{
z)?F-Mt#>$wz0bSrdG<b0AKoisW4y*dKtRA&R+7_1KtN1@e@mmI!k;98S=ZseP(P_C
z${{?ze0^&xNk%}RK~R>H(e}<cg!%XzSzX?ojzN>%3pvC%_$D-~2RT_)K8Af#Nv;cG
z=Lil)GDta2|1nw=)BzD?VdYPzTZ=9zjWi(Y{m1^Hl$-{=6unz7jq9J~Y<j~R<0R<z
z^?({<ieT<<(9xo@>k?$9m1Xp|PhL@`$hG8%PbaBGEKmRc{SPKh0xz`RD_YCq0jA&g
zLYwZ*!_ikwq4QS@Ye)Sbw25Cnq&Jzm*Jxus3do>c*X<-w!fI_2GmGW1Hme1EisdP0
zNRx((p3+MY$<v>aq>ltwLyJI#2CmGzPrzPM3sq%yFkG=WQH*$S9T{u>e=XAPgCX*H
zTh8}86f!U=$u#P$>ZYW6R}xvK-nCBMLxW9E<S`0jyC-;PEHo6N1{=ynag>nPgUoCE
z`T8?{!ezg+l5t~&aU3x2m>~Hf`qA>@TOmmnqI|g6GIDs*CONd4??0r2=_sq6nCL9+
zfA?cMK&d6+?w2R;&-e#d(=g==d1DI^K}7igT5mdT`H2j*Csx12<{`7KJ7{p!d)I2k
zYQ?GTSu(rN5afN<w&sUM?~BKZ2BreUMpp^-XDqRUGlJ+c#h4G64TKeR;*jF{nSXR~
zps*TZelVG1m)yt)<N+L!2053UX0vJ_>ZhR0=)oGK2xR(Chow9r8H@?E2WMNSVorEY
zSu}i~H@#21bXeK7E@VCbtdj>xU5U4+7)yTX&GLW^yz!;u##HK?m$5)4Q=yUl-8y}-
z!TO-}K?I74lPc3nz$JDG^^oTK8%u}nFoetEiZQGRzV2O=us{`u|9NMt7}HUrl6yYm
zU|G3ur*!)WOXjz)8`OwYh)>MF_$B1!#y*Ts$n+1u;0x$fXhJ4i`*ss`EZoz&jad9i
z<>+$ushnN;7N`-Z2CXxfee}<SwYQL%EslW}(bi6BG7Rt&5?~aV3Q-TrE2vgm1eLo>
zB%h81S)DH=5sEL1;Xa%nR`q85`~~1e|LvsmUg?InQA=bYB+BL5=9>XOeFxH@$*lOT
z9lhlt8OC|gbsvn&f;E*f0nuK@?(T!B2bE;W8*S{qvac*gde;}eD+(+9w;W}-ekUR;
zP2PT1s+%<c-cHM3-nkJ>Hgv9u5K)e!CoInxTd^b5qrBxb@wXVRP;|a7&3Ya)r%=T8
z?k&oGoaDr9L=vECbc>W>HKI4!_$#xTa-mo5vCN_dASd@@W{lRb9b3~~Rqs~a6HG1#
zGue9&4+tc>WWTS>DbqQ!<VjCEgUZYFci>`H#i6sN=)grhV*-W{M4&BX+paD0L1<^T
z=^nYY>ck5-PxIo-Ngirm*S>E}$pL!i^hIXrQ)TZ>ZYLkB_?vQd^s_NQScj3${hC`<
zJ0eN&caAoUjg_YRPwe2{zVK?+Nh>ea#r6fZ-?C6rbNeEck_IJRa9}>`HRi*V`blG6
z)E3KER2-sBX*{AGCYfT!%UV`k-@K=Bk~Y8Jk6aG|`ZC1v^7WKng%tCDlX42_M1uFa
z`mx@mNwfa`;dNN!@ZAr_2-9o=Shesn$4zw}%WZNJ2T0w4uLjm05|(;7>3(abek^#q
zoib^6`IG(JzfG-Kke^F@cWTP&Sfg6xG(EGy#p_$tSRpSio0L;H9(i{;9;V~+msnGY
z2fEPPP8=)yJF<j@$#lkwfkW5lOEjrPVnMS(+Dgqzv^4ONw*Db3U^M5d-pc-7YHuOQ
zFNSLdv7zaEt35g!lP4O%uLIX|U5Lt`%S5)I+~G-kvF><Kla+?r!1D6beyL&!9Unyo
zke81~K~J!b!RZalvWa&3vz>_8I6wU}19#hZMgha`z>E)nuD-qzYC(?DkhN=8NM-s9
z8u`2EyW)v|JMLC7f81<^6IS{<dKJe~C7`5yCeDZZdBGL84<14sLQEg!|K3U$<T&8U
zz{Mei?;!&^zX8d&GU@z`YUNpVDt#zzHVC|`|3V3oUox>0(Uprk%t`VsZr7<&sWiRr
zaUD(Fux_nG)lc}M3=2QA2$dKauj1LX^rEqH_c_1GZE5W#X31wd+xY3QVw_sx@N8Un
z*lLIZ1Ggrh6u&K0x~`X?KCwzPPNJx99yVmzL7DaA#CQ?q-ikBC{*}{Z%7%^H23yx~
z`5=+@P3oPm2Xx@suVNN()<7b6dp2y`|J>PE0RA-9$XF7?C1mDCU-c0wYUrDUYN_UL
zCpF3@>vwx5GVR3XXR`?Y>XQo<WyR}-E==)=pY({^C9YCQo-@BF2c+;)Cj8^457fJE
z-n*uKD55wdBNTuih7v@1(nn6j=*L6z%8=E}WStb{NAgF0gd^!CnU%5Vocs!hz&axb
zYlDB(Xpfw_uFEL!MNu%Y9L+>lR$0K5#W!rZH1P}C%Qg#F56Ti9%`?(SA{=Cp@I+GU
zRq&=*{L&p;_sCSka#0>t?3upEJRcKAnlwdotIxxq%(O`^jibowmDLWJvB9YgT!$>^
z8ZLf450MVoh~E}=qDXG=&tvm*^%sRFWr*D)eka%PcpeX2eV3wkM$o%-1}fw&xk))h
zeg0;{>(QyLz;P|6Vd0V%w&uF;j6M_p`N!l>ju&Hx&d2fLwI!Hwj$|PAY)QY(n#o$C
zUA>>7b+y)ZK!1>xldP{Ue(WlpcQ#*IyBO>xy^?#tUG?*$eX1xptK~XNitHHH(>Q$)
z5|WhqzNAmB5XC!8;@N8x47g`xx)$&6y&6Tp7z5``X=guciOi3%tEqQQECd;kj4#oA
za--p?0T8^EHMC8;Gg5bVy17I*{a8P7Q!xx(bhoP=DKq#bc`Iouq8fT#F>DEznA1<x
zz1%V~ezGuM7^?faU|he)qAv0-u~7Q@`$NYsloUuD)!gZi1MTKQM|R<K<1_>`1=uvr
zCm;h;n)Ey8uCrXlcmuHma`bJvndk>X_4i|=`uO(@cAm<v+BWs7@|&c=!0nJA4MTr$
zA)vM_CjF4Ns=VdSQsX^J+>h8)BEZvu&hD<B`j`~8han03vd_g}eNO;qcV+3qYX*2+
z>m^W<;?NgeH8Xw4E@W4u-N9|x-C8>U-ZmE4TG+kLVs2Gzd&@DbnqUFOm+ufR0Q@t2
zOxk#F^47ZIglq+*(J(c_mgjueofWsKA;j;;69wxpc4bMIgX>`SpAOMKl!^s15ur&t
zfBf~|ly3`7Zw8WSJV=fjCcW2TeBjgGVDI*4Nv8#0=q@LTze^dT<0UVRYl1q4p{UhU
z&d4p|AYA?|8;d+stE90o{U3?(U#Go)9q<0g@XxI}tb@3ju*8IK|JH`y&Vp}?<9+-^
z{|*K2AI*&Dm{cFD_cY|j)4Tzih{w-w&VY398G@fG{pzoGU2`Ooz9{Eo-MIzVUT(o&
zHcmTrl$Uio^S*3;qWW?cWJ(KDOy%>L<bruL-fdk>Me#JhT~P?b<?9Tjt||weLi^_*
zN=}=vHJk7Y*jB3s2g>_*Bra8w67yG+DjYhnf5ffyDp|0G!@%}3f-y59jJG}sTzD|F
z3Oautp~CfH*Ten6x*{B0&%u!>IRSQp3Ch9ZJwql6x;4`XUdCF5lo9}>p~kU~FR^~y
zqU#?J?LKe$?JW)!Im-bK0~7LHG#_g^m&)O+2qTn;LB(CFuu=Om#Yk-+eI!x*byl(>
z+|eC6K{U*Kfd0vBTfGNH_&;@)8<Y2T+lR1#t4pE-mxHG-z9;q`Tdn>k#w{yR<O_<A
z;}3~pOi;5-D=u+&Siw(V!S(YU)stL=hjO)02ZXNciN8q(#dLv45|VpYta0C<Sr73T
z#<fy^DKywxZpyDG8yO2JwgACr%mqu@>`@sQX27EK@U0D~_nlv8RH8tE_%ZQAA<2}-
zgGj9)I#UWCM_K;*?xVyxx$L|EbDh_Qt(Y~dt0Z1u@3r9Ch!fxF5FuUUBhFV3w!9jK
zPpn`@yx`oaOO7F@(m`yw=6~`}yIqNxHf<+z+8i$xZ1NX1lPR?|;7Z}F_>36;RD&t5
zE9prGeke6~5JGOC`HLRDS(=<GBk37%<IJjzQf%hOJOIl5FFa-7q`8>0a{qIDc~p4A
zB>O}8fCe<>$^L=-lWYB9yf67QH65J>7OgJZgZM;Nbb0Ak;Ke&SE;(2lPc~l><o!N8
zVfwYR>&(&*djg&`euUu${ygf_D~Jp{tu^k~aVmX!^%kmI&{}-LGAM;VT6k9+dCerb
zfd16bK8XI;h9F?AtiE>3bY`lpy<GKiFWu~i@!q=?-o*}*X1v_oilOBr;iHz9B(d|R
z1C0(sW%S*iSXVXN9_h>H_tb?2jfe}1bMGgI=;5(E>P*%won`HsCDJwi?JuUfdU((s
zYQMjTBLt?JHSHkE1upAM!Tb;F(oLhQA&+Talyh<Y$*;XO_KA;+isy+P;?~<$#*W?C
zWY7Oh5GbjJTOyS?y$@8Cqt81=0Nb<_?Uh1i9u{u#WUE7R6lAz)3IY|ww&gZkOi~}*
zS5Mi_`q~}Y6z5q_Y_tRF@)`p^zVTv~0JY;^i_Ok>w|gCQXR=YbQd}GRdR8uX<LMW)
z0aa*t$XwlMGPfq53CHsZd_JOs@Gs-y>3m)k&LHiP4SpigV8C4kCj9D;-I?Xnn+fm;
zOj6VsUv-v)P;=7fkMWnZyU$S)(zju<D^wLfT2o74>u(ER2(ms=!mFYdvABEA7cfQy
zrP=e0o%=k<kO0M{)HYtuYDZjsSI}{8u+HY8aqzy%NzL@&4Q|lUzGwpr7h*n{MC&>T
zzA$^(`Khk~J^%PU(UZ<{(jDOSG39B&{DYJHB?Co1R%y5JL;04^&s|EReppWN{M9~w
zJD!JFe<z+Y&H1Ym#W>Jr-*V(0)avtl`3=IDy|X|MGX56t4)e>OyX110`%N8wnsOb-
zI=IVhdJCs*t`s?SQMmv|`_F2RTS*H{4tiS+%in7_1DU%&HbDyOz24rcdEOQcZnL-w
zP@1ATMmJMhpn{N~X19L|2`wIVzg4kgW0=}ketE7RNk|upUxvE<9=6h8kk&BKRJYl%
zJ#&NPJsrH8IlY%%?JdVQ#vQd&Q;+*a4b6({i&m(vG>ktoTDWw$c3F0~h}=^T5jTo!
z`2AO4hj5THcS@3VQ<Un01WHD!*Nm0%CA&H>U0VXdw{E0KB4B1y2lNeYZB^)>()T2Q
zm(Lkrl+*VUN=IMweVWAu2a4db=kRrK_=KJ}W$Q!x)q~$hHdFG|Vc!9gBxs*sb&n<r
zF^Kat8;-esBy|ZeJVT#+rZ3$G>JvAeEmw#Y%kvCT=f$FRa8!vU{+H?8jAtY|{c_Rt
z0@^++JV|c5_+`En1Boo`jPpNW{a8-`q!r>;Y6Vz0i>bR0z~LpjkEM%g7>xn#KU`RR
zt*6ZIu);wZsG&?c)hTy)38(shbhkOws6W7T9ub_q0jv2&WAjSetF<agg#_gtC|7Z{
zlon=@3<+U;dg4VVW+@J(+*3N0cU$g1qaR+)_Urr(syzE%RVMiC*633By;@O^B<9yh
z8LOfnHpgyrXvsBFiwG`f>+XFH(Pn%4iw(j)GBzS8Dolm*v|&d9kb;JV3cikgaibO;
z9Ae_f&yBxN)n$_~m253-Sa%$7DfBejp4Z+O(cCwjAYr>RG*bnvV(yJ(##nnc<!kLG
zj;?(L-G-6BnfZioLHZVGsk+_Ji03d4&25C6V0;sL5uI04<iyBpd7D%VA$Nw*k!_(8
zoESQHPZ1sL(X!(U_V_qUw8qGQs<Ns!#^_n{pLt17sB|_8bKyMMuX%wCV%w+KuP4Q#
zV}p~8J5dN?x;9O)ZjH1Yp45-lQjJs5q~Jx4n`n{OkVA6AzEQfZ#ws7@`W4g8*lW8p
z%))5gzkjpZ3rbkQJXQjh1nANpw5J>iEOuO!aN<Pr0_QX@Rl`ek2N!`g#g5F)Jyhbf
zBFiW&dnPN-NT6Upz$~jGBU@9y{k>u?iH`pcnXguT?l#jaSyr%-!2Yd}_N`;6TI7ez
zk3xwzHOaz3EG1Uy42sZblEQP6y5hH7x*2c`O84ED3%^R6;6~Bh`0~xnlti0kxgamc
z(7Bw4hW6>#xh&dLX@uqo2Yoqy#Vk}gDHro|9E+ZCvNqe6Wdw-G4*oX_u?pdFwMy?>
z)@7_;`rq3beZsd#U~(;rpTbx?Or>t2LWQ>9zKi&94qjq)U00V;$R4F1&wt_SC{jwf
zf7_3ho@f(_P4IiJv<Ew3N#y8LpYk>3{V-<w#kI~HyGgNtPb!TN;bA5j#CPX{MLq9#
z9mAMb;xQj!xuduiYsj=SbWf*=YXKHxlrkFrJm(8M7R@2PoY+loc)vKNR@pQ}ven2K
z!#R-%XI0d;Q#K&Jad%(;z4p(nuNI{4^VvtRG)Uf_&5bA{l08eW-rnQxCjQv$w4HLH
zY+lGuwfsa>Jt-4`Q1tCdT(M?NaQ^&9Y0Wc=*L$Tk<^VKT*Iv+`X6hU?^AJ~VO1hZQ
zLy}d>EVA86ZIi6FBXa(yX^f|Q+Ovp|OZ5FW_PBt9HLI3bHxe%05`ns6yF1^vpoJ>5
zTa;%m8Z^blbR@d`SXYQ5$gZ5~21oWr!j5$;;nB)ol=hkR2EWQIF=XG6={Gt#qu#=k
zSb~)34qj*E9`;m-bw^OSN#l=mf!78TCDF%^h)N2J*rs7_t^$XN#0T)>>R@>nN9zly
zFEf%A1S5Ep3$i1bM6tH7JjcffOZtYz2qU1Aq^?`G@@BUB9)_c2KUu*QftD{J$+fx9
zkqATJsH-GVPLE+C$pk&=D?>XEE%q*;_M}KpFGN_P!-7yzpsq-EhH0<Nbkmmkwju8h
zJML58dpXX0uYq0(1Wx-rad?Gr$8(}n&|z>bvn@F7xkxI1G8d!FUsv3YDf;>8=kFE1
zB0}`e^4tLs+D5)cpWa0Ur><qxp4G}$CB6)&(cheREs;8i3Je5c`+MmOUavXr(!b{z
zhYPmING4W{F5}6c_{2=mGk;Zvm2(S4-}E=}ZV(Q>0Q${B68YI)a0RYf#1hFr{qzo!
zo79je1p7j|#AzGWU!rA61K`5pztg@t@S?&Ye&hG)Ncs;?)!1>XcXL7kqGITySd_Ao
zVCgZu#I*~G^Cj_dE-R8_A$I$mx3n%~D@FX_1=J3oVFJ+>uIrj{6W7zFCCvO&n}LM#
zCwD(uj;5#j@`DEBTVcJ0RL!_bI5zijzdyiBK*dxdUODYGIrI1_izTeKyWOmhV^S5Y
zGtE1MG||^>nAyxNl3te8zQ*HOrGMh{m$O7ah#yN12PUl$IQ85WzM9^87SdZ-`qnN)
zyjjM3-G%ibP1^TArUnmI9oI>Wd6a|atGt^L9Efrf8^Q^$9pj!hV4V4h=dCLt!CRhm
zb2++epnfYc&Ly%G8n6yeChu`xB?Gdot8RMAjr)&F=+Y4??!9Yx;M<bJ4!Hog9g*Gf
zr<hMV<{_5qgB?He0$%A6?sW&>r2!+FA_bISdOyN?vaC1qW9htQ?Oef;ONjR5@(&!d
zH09fyw&x_vyb;~Srmasa5iMD^^N>pk7_NToXnY1oHQ=AhRp^6Z-j0vlb@;8ftlM2i
zwG4cxW6hjYKM28tA?o-No>z(~iS9@pWU6h)AM(NTG@xhCP*X}Xwc^e3{fUQq>6gk8
zr}}R@1!T9xa5y|6k=R4wXe|&D2`}B*D|hwdc5vx1xLEJL!YP|>)HC@#8B_Q1H)f5Y
zd)E|wud)0S*j(!=e?c-Qj{Hs7Cj=eS|L{01(Vr;K)Z8;Y6J6pM>rd3QN<j?tC!WmU
z)f8#y1bwTl5OM-z!l?c+WCVeY4&%9fYG0Ei9NSe*?S3EKX4P@(_ZNxFwD9H`9V^|c
z28o!a4#M1*_xM<j7DR76aFpG^SRPI}uHRTtcgrW^CO{c8vmXhPt4H+iad`eMZ3(pG
ztE%t=5f^L|jDJY$)H9|4WmgPwC7aFE6Be-lt}806GB6x^GGJlAi7IFLB<r?kuMSl=
z(<C}=lv}ye5j*zno|?!vCKH_Y@KA?1RgP?9&?V!yy;d9<9_@thj_E3JYt^YvhavTr
z;WPy9jiK0eqmglJ6mjZ)--NJ7l|7*5@M?#?D)aK;R~5YtJatSp76lY_@$_wq8#Eb)
zjlmr`C-AWg*9(f}|DrV<`&cC<Aj>f%fq06~J|gYP>d^Awdeau1L>vo><^stX#8MsA
zFI5+b!zdVUM$Zkk&_tv4c1?7wJgXWWwdTskM5HY3WmH@t0y<~gXJcL`yn`3TS#ie-
zb4u%{hw1{;b@{Pp<}2eY3%scV1WMFZS(gvwbgrBlhkuCM=q3cspR3JZO?%YE><u-E
zS-CK)e-s2H4JK9XRXt|!i5jIw`b5B?*RJFlNf>fLWCtS#uJ5o+mj=RSV#R*bHrxp$
z(fz~G*^IJ>KLIV-UfazZ;yc8$JUIJ555Vh>Qc?SF{ml<;1$WAeqYDQcZ@h5-tg?R$
z+kwJOju4W2b*55P_;O5E2Aa#7iKYk^MgzymJJ&)xgShfOO!sMvyb*7!8rJ=y%2)la
zQsn1897$%J)dDm_$9vA7krzocZgAs3m-5VFE9CoU9Ynq3%8>{MEH@o?pOr^FN%<Tb
znaUrK;ctyv9^W5MK>z*2>H}z^iM|2p#}6+WsRbOtU-W=z4e0X1MsLe@ft-Q%K4g)~
z##FVA9ZW<kmU<+Ynupj*rHnkxm^nR3LGt~z-2#N5X-U44s*ot)<8sael4y{6FJYdv
zyDwbM$%>Ngd@ss?BUv;Zp}JzoiOw;~l`CZ++tXo`Ka1YX(<5)=LP`JnLjLenL!g|t
z&FoWR*ARhHVpU#Vph?27NZa&g*Pb6g39sGo%WI0)`OueL@GBxB2Z^yc${G;8<CK^O
zi3{}{uAlN!J^pYsS<Ha?+t0c|<##CP9m;i5+R#Fpv#v0KL)^4qu&2+JBg-H4M*}2A
z0+w||`>>4Cr0>4hG_;X5F%_H?ViGxibiRMTI`A8X@^L%v+0&mn7-IsoI<9JWr<(3U
zU#qZ^1vi5sICQaxcQK9z)``nrpbc|tgba=+Rfm&t!TewC1(!7efy41dGBNiT-uK0x
z+g*vq9sUVeSp$pm6(CDTu3?|6u2Mh`<~_M78bj8n<aVNxg@cVoz57;v+JSKB2wUBY
z8>*R+-%y#FI+)g9;E{?}b`A1TnLx8N&<ezxf*{iuelEmi^H+~|fcs>@EC=-uc@N6K
zqw*usURyhC9JpR@UaHEy-7f`lXI+5bL;NLN9t6sNi?gmU<-fQR9OqiUsyn$>b=!-0
z>0{7EE}QyjIGhMp=1;=oQy)Za-<u;^;niaML`L|QHs|UteabeVSB;~rY705WdZW-_
z5?>^+XI2OWL<euZouD2Sef|wEk>XGMp=gn7(o85K3!61%nQ0q7055ut3olCNn9=|U
z9Bnxj3xJr)rDDG2s(FC<3!ktH6}Jp)wRYobYr@ke?XknTwKN)jp*WR&SmrNW_d`Cc
zDr{6;Xo8&*#!%wk!r=V~;VE-*Ed~`D+LxN^TkTZsEaRSLD`A$cUL6!;_HT~3W|%2B
z){FSuIAs8Eq$^Bj{t3X*io&jVF@w^&Fv9b+>}=z!&T*plcfHWj?tH@>_t^r|?);ft
z{oc_sE18+*%)-oX;5)~EJf2c!nmhgHog@aszq_|flP>^egLNf%OM9NAjpLuP=~T9x
z$8G$mVU}d4%Fbzls)le;FI}+DiW5yBzSXU$!UNRko`TbU$^3v@!G!g##P}lAK@CyA
z$#(nH)ZKWbJO|Eh?iuB`5~QNYoo8D7$L_rywmuatDGI<5o*Sqk<06rZYb`E*{Gx}r
z4d>x1U;m8J7S3l0K=UOSEDi*{V1f#&X#r)z=Iom#xiAfHQPesN=}A8=otY?RHh?a}
z!_jlh8h6Rhq5oQS<7{Jk<73ET=`A4ixN-Rhb;0RY?r+T5+x5lF%-Vr+b?5SF(x2eH
zQSiZ;GQH@G(`VUDS$u)h*qbRufj|C#rVup|`0I-HiJnonWep6KXi^=oJ`oTqmPoM{
z=ph_Om61#C2%Tc&kBoYMQBwt>JG^{4P|S(HsUl|zo`o|X_QYw(@G$+eVz25BeNc+t
zc1<E{r+GVR8s_OiRtNV+X!(bFu7YJ7^n5=FIQa>u2?spMlYNckoaov&PPydcho-i;
zs0N4D|4m2T9|{V3Kv+bE4cnJeG8yp{{|o0PDU9SPpz=Du+g8ReC2T?eJ%oFr(egxL
zkfKq4mHXUgvchCzvD6fsOG*o1NC73m$h^L_0}e4uGVbu^7b9@W|I`&rn*QykXVF?a
zijT@;#EW=I3nUmz{NLycn6BPw?;i6ytb2M<dXL9GfC{O1gynQ3t4?zEZ&Ej2I-fH_
z%pKL%4l`~Acc3*D3WRuXI>ilmL@f{wY%|(3Oj7&`M~@wNSRb)tTBM4_?Qs~8Fpp+E
zsHIjPab_6{m`qN>=H&$_1LNs#!bX!LaBX07ktXlnjG(0WA|Sk;c$o$05g`^u(|n<L
z6QG^;W^%t_KC*IB&x_>`UQU^O&ik_ev9v-_6~KUE7f{|Tg_0^eSk5xQZckbr8ndH&
zT6qzgOfv1x9?v5AizIp0LDf}oFc4Ys)nhltI~r^i=C9X;u#y;F68;O33!KvK_`dyi
zexdar>FY<TWiP>jG;sE01nkqBANxP<63Xj9c$FU9p&I6bFHnPc5e#ouhexhGD;7I+
zKM`H{%WYTJi&bP04*u8m6B`Fb)7Y?3V6Kkkd2JjQ+?js$xb<5<QYs6!+8=TYVxxy8
z%uD@ejtrkL4U+tB2VrssPj5STeC66D_cW>gy6N!>iV$an470Dm*m^!G(OizMUlX5N
z2T$T3h_Kjwy!`oD@dESNB=vbnnyLyT4YZ%yiMDn$!t_6!H`#@Ku>tDAD;;tquZkIv
zH8etg3(iv{oVYi$YDoyog}G=jC%}79o3y)+t)WzJh7&u-<!*j)<W?fJ*}6BHWWxy-
z-Nm*yI-rqIGO7UG)#|4ET9p3;E>eB9C_=kS@?gEviJt9)xVa8(j`O3lFI2x&@q(#|
zLljb@=<X_+&+NaiNZmhVk<8!wUFN>D#D@g}cvp#;R(IP=4{#6AC=l$)hlE!8QRO=n
zg-mVXK+bkC_l-`mfCzRfTM#{58C`v(s3^t+-sX!ci3SdLNC&YnXScO~3MBTW3*w6P
z#o^DZ@>53y6NKsDN<?F_hs>{fTR=1Rf}RsAv`}L(<;aX3vmnjg`dhClotfXhPjgLT
zTIDD|%gWk^iN7fhz|!g99W`afuY|b1dg_6mn{ad74<DH)2gfT+#zvgn{_)st_E8bp
z&dNM|jx5c6dfguwo_RNWQN@eLWj9^m8j^oP!kLvwoltUU^}mg5lMsmS1s<oW*<BZb
zQl{}s=R1Xuk7Kuk_D+Xx;ddDtqd3){Iwq2@CiRabbRz$zDiPbdR2ZsEydR?d;rjQz
zxJ10!;chNnI_gLchfBs9T*i=ZBJ+?cr}q-{-CsKe@$0p}4*$rt(9KJLGuS!sx;LgY
z_xsi|-Q8h1D|FRv;w&-8pR|4?d*FDHaS5gR3MtH(-6ca4ew-Rw#siS8#^OAJO(8}&
zf5HGGNo|1>8Ve^Fmryb;c`b;Yv(}2E&g1lX?kyv-hlnG{(ADX5Gl^3;`z(Lc@nw*K
z$54CX^!ko*NVYUB|J$U(LRL<!sZRD6WfQKaY6z4QSBeGbRIqqJ{NTiybJ-mvJ|OnO
zUglaXKcEla-l$s9{Nc@vwgeU-@#iK45?finw0v$cK2Vyw(Mxk`LiV*@$feH(eyh@)
zW%!BMG#ur=6B+nZUX~@2jhsg~sRbqPrxpe7wxFLar|^E`!K)8V^ep=2CbQkp&Ar<W
z?OFP@SBhmH;!Q7?;jPBCV6Lykb)VRn&Z;q_D7U<pDF&1yIsY3xV2qcwaumwJ(k_EJ
zWkl*Dm2qR*RC({fNmNR0IE5s-vTpU+RoGd#dm>(lza*`!WZ9s8qG}ZR7V9WLA9Im@
zSQBE9ZX(gPe%mCbU5>&r)aIY~OEmSD22(GLnvWm8q4a+g5BS$!_JmN(&ZTrS)@U@w
zghr+$yg$4A`Li-?tx(FLd}uiaE00mMptYzw7fHpk=WaTTxvg3BNHTS=R*cDUIrYQ_
z)3*Uv@ta_y`S3`<D(fPAlmrJjW%p(bHn=6uQC)<fGElX26yJBiJlFmHLii?Hk``Vo
zx<c+2J3d<vbgUg|ladIkQ%R~$j}i%!*S!*J1==(F+OvbP@Ls^K3`BmGuia#j^{*96
zpxow@&U(K@E%bX;20<j(NlrH*QXmSn!_IgLB;n<TJb%!~;7Oh<s$kZD;TA->cBB;s
z?{X5V&>#=~7thb9r(@1U3KsllsqJK_2&NRZz29o*U0^<~YaUWkYapG<n4<asy%kcK
z`;rGh%Ykpj+L+PK605WC)aU$DC(xOC3PoYA+|~cH=ii5h4Zv7BQ?P{`+c*b}UU|}6
zcaqSQ?O({<q&nsJ3NcFc<+{jDz-C0tuWyA4DyGtdX(}ey1=KI(sh`6_wcEJMefvgS
zh$Gn};P`gBQjMyFs^q&XZ?)DU=rzqtxMUhf`<53Ba^qE*(CVFlu=oofNw$U9>Qz;(
zYpjenwP@~g>bIZoB<_b(bCv%RopGV-jtj9F3MM;%AD`I^yM-|ElJ<a0dT1YR4{15z
z7}@KxqNLC_xY9y&$C(_H8#$3NLFOAg2{Wfy@Hz|;YCwsE#g8O28?Dl{rl6c4B5p$W
zSCW31w#b9lo$!GheSp}U6PhE$Y2lLQv@HhA2gI@EwGupzi}LdZ24;Pm5S`({cE4}B
zw~Vz*lSchsOvWXSl_L@OW$)dXg`^A}8KGMvYQ2xjzc-L~CCc2?0@xTW{V-;W0&Acp
z-u@i_btY!oan78z!zBf=2#GN7Q2;h-`ci6+K8v^FcV7l)ik&5w>!A!+xF`UBzdg3x
zPFyfG>b8vcc5t4I&$}WlZ+KSC;1<1Y2P|o@`U0G_zh~D`c3Xyv!UvaBCt@FxJCRaU
zNa}vzo}Q|ArE9IhR!u&>;xGBvUV^qdB7+`*tQEv#3+&18ss`|C1Dg}B(7<Tmk(VS}
zvXWoa9l@yICqTDFmDa~Jv__+Wa~9@OI?MXTqM@505|Kv8qR*n!qEoMSg!!wt9cX4X
znxdk705O8MGv*53oH+g<Z!Tp@c<)_)k93c8yukl)c8;u{XpA7zCq0kwChc0hJ;S(l
zxmv(td6qi=`o8-f<@m7^{LNzha98Yw1pfK6YQ9a{Jn37qwv}lZNZiU)hqA;Q-B!7q
zRPo@a<-0<qtv@cwEZT>hI}Gwnw5mv+XjGIE1<gxS8r5)8c6)W*^Iz~msP1uSiI~sY
z@!tKC9DZ&Qx3U?5uCsG#t${v#D&SMjzRsUELT~I1oxegKbm?x;?c{IOT{SdhtXlG(
z$uW}<8W87RxibeB1e;(jN0X7hSZa8+YKj6HE)3iI-c?@cBf@<g&yS_R1|9W&ZQho0
zzW<{U!2caK6ifUotQVv^c8(j@LT&x|1VhKEABOgSIZ$c+EvZSYI!X&nX<^%fxdUhh
zPqjU@yz)l$E>wLKMk3v0nT7?4M(f1Fz}Lc&=cFfVui6_ZovT9xsSr)L)_kTqd2r_0
zrVY#}yE+!gCE|rfW&?O6Rd2@Po{{rh{-$f9J5!8<o+PlvNi^&yB030KU06?RUfr0k
z2(H)xh8$<m15l;BX18$_{&uuK%xiQkFDQ~LirQKjBrZ?j#LZ)#vmii_o)H3&H-g3p
z7#b4g1E@J2X5R(I_!xb>GU+(Xki6Sm8`o`j>buY#0$QrXmjo%SI4>HF=WJCw<-7gO
z|AdXEE3$Jg)$s4n2UQx7#57_cpk4JuOdkhX0{l=H9{}|xLO<$P%f#*l{JTFTa^|U4
zi_Gt0G5cEh^uQx0b!Qh|C>iL&cBqfEdw<H{{D!-I@EAgbECRNVXt~xijps_=Cy!1h
z@j^q*9O|3_2NE(Q$pLwmkdE9Z^Lv1j2qWgjg~A=4L^h@Km`%|uUMB`-_N}loVViZw
zxEii}x(q4uNqFULp%^UZ%JoBywdD>^4NuiOEhGkT#uW^7CtTuQfy2fYx6)P=K9^hV
z%z3i-mYDKrD<Fe`R!~F`;r(s17<xF+sMLd~gzFy3PdL)Ti~8s*T7c5YFBs=md+#Cl
zN)T^@P-Cn^UpIl$MFYLQ$oXzS>z3LImEz$H!Q+%oAmK*Z!>|U$xg#_3H6kIr%#OV~
zr*bz(tv1YDU}saI*`)3C3gWvz%`&8{m!J~<t5L~O=7zT}5-d-|Bzlh!h{V*l7%v=)
z@}A2ro66boe-|awkiKK_bm~6@j363Ex{}(V1&_v?J1ok6DiA=_M?E0$QXD?O>i!wW
zJ_Su_$gq~K256e}fK4?QhcnV$8nJ|i0h$rnLXp^}sbhZVN6ry_utZ+g>ys&kFN=%c
z_258%Q;4joU~s$nb-!rN4$}wYO8#90DtEjAqGP2MOc$>LRY?R2B8&SFPhJ>-k~rtL
zCLNW=HEaEn=+9~Ay7;O@;T~)&Vv!bi+nQ^SW*q6&UerPn{veJJ_tRj;1aGXuriu=n
z6+}e`1s8G<Lv>q-aMe+r>a#Xt%-WG$ufljoA6xbvNmua-_sAY$C$ns8`zM_~7?}2;
zP+H<-UN_H+t?TZx+7?|m1e#Ntw@DK|0?m8pN$VU)ZMqz8=4fviyxNZ(Bny6w@>D#J
zKp$>qkE#ADJW6|$nYO18{5FlFQ<-)x2hAVF3)6zN+f0=Pf+Uo3(rgaUn-IGYU&4*`
zoQ0@iGb_<8&kvtFqP!CQ%mr-4y=e*2Cx<H0beJc!l{Kt<Sy?485ei;;f2<n3oX#@r
z>>t;aE!~Y}N{o)d_3wO{8)!+_<?Q4ZNovhzM5N`egHebf9Qv(kc?9RK9;#SFI1uPI
zxn%YD%ws=psR$Bn&p)JEErq#~XjSfRUqeM&5N7aljMvYu|CSsR+-1LJ=_9(voEw-n
zsR<{{p(bQna?k)3y3~jyaO|T#y{-Oa>lKW<%sJpdfpwtWdu}v}-*khwFxUqh*za^j
za|w@52m1gRUGqFh_0YrhQM0Q(AdUx@+m`^9lb#HDZUXMl`bcm7yk4;lHvIc@^y5VY
zf+ESOxE^o=z56TL^%foY;nlUW4~uu`%Fzve1dqB0X$3(n!XJ#&J0bLNJ-_?it-M)W
z9Y7cd8nz0}<FA|*)_~lei~*w4oB{Om@^BrsQ&}MpnTku4lU=3vr(F}YV#41)n@*^L
zY4f|*EW0V=A2F96B|Gv@q>j~jA|<`hMujvLa1qnRV!f0S$0~}<E7yS?F+Rl3=^^b{
zRuIBUURMtiy}04gO#;v3fv?%OVknLnWL^X^Yt%&S;FPUw0CNZQgFTtM9<W#5!-S+q
z&kHY>E!+7C!)TjjIx%jo*Fk+{77y~wY-qAldn&xhy~bLxmcy9v&mDQ$naGZ!kJ!gA
zV<iPztr;pt*wr&yEVm%O^G6T(z80&($fjKJP`uli0X91@lgB8`!Wl~#bv3wJTLaK+
z&UWE*ry=qpDbVico+dKePB`q4ujUzPI(@wwp-a=0p)3+fSYq3Kcz0yIFEvt~ylZ~1
z&jT0Pk+EwIgOkY6U}2M6x6BC3R6y&F_!s49vri0V-`s*mPE(Fj0=mi+Mr(EI_5*=F
zc>y+hcn#{q_HocN>ZdC|MVDyUX#n*}LC95HQB%@c_IkmcV}=~5=Yvo69dh-&Ob&>m
zH&Sc1jarIxO@s~ngLYvKS(wu@cJ?C@a9|4P+m}{y;_@rIKI>ok`Fn<k-SB;MroUee
zjZ<`4t+bNiYeyU}I%;K{Ac!i_{zvL>>wb*~<*POV2>wV>_U(=-e{sRFfHvJXVG8LD
zZ^4X={)Uo0MJ+%>?^W+b`XE^LC?>%1SDlKaE6lP6NI)AD$8c?ht958c&#2>#CeuIT
z-+{ART{}(sKZHIIDg~cx$1l&uZ+qg~3u0N2o<t?^e1ga3oXgM-4Dgd0mY3FpgQqDI
zUJ3@+G>e(-pkNF7aC*jPcj3sIkPuRrQbu9P2JQHN&$OM@6*V*&zsRaFW!b@@jjZ8<
z#>RcuHl%i<k1bmBIW0h2tS}z4WqYgYewo-+oKL;9(Yf1(=Oinq35V_JYr|wZ@PN1q
zo$Z`59dYK1|C<jWxZbPyYEL|!2E8$+Sfx>bG6EZ+^_VvwqaYlY9&OmOd=nhtP!dI?
zkBT=|p?K`ulv=&#p(ib4n4O2T78&h~6um547d`J(*0%AEoLjAfHE)01kag`1a=-uH
zWI@Rq$Cz-^?a0EK8dGs=PSp77F(h>P`vcm$z~rp83mbvuG$O4p%}L?-VJD3LoUl&k
zw}Bnh_QKAdgQPWZXp$G>vaQL#&g4VvW~1?@N3YmvK`c_^H}CYjX5>tT57F(~5WFYH
zr9R8MlCsKPBcd1O#L0`KrV$)))`))P2pyamQ^QjLcb80T(^07kvFF)?O(^8cHCs2X
z)$!MCKF{G{0wr%3ldsU54|z(<Xuz*%I2d#N`C$J+QtN#tvB=Zr5Yo+BK4;BAQX$cu
z0ox^zulXC)#Xs(FAhtkY77`*1f`$F2dpI6~ylr11+WvsX6Rmua$~$=m?54E8LO=0R
z1i5Hn-}0l*N>aS6$p#D@gip=5E>%}Rh=eX%RF^%w)EHr(qCIrGDI!1Qbw~9zC;!wt
zg=JN5H>Kw?CV<EKin|Nv{N-0xL!|=F%~zlaJin)*O<JMf#K)wa`K$FJXPjt+Ck9fH
zWH^3SqfT<C!51A12A?sR+kX)`){f>9=wKA|*P+@LiPlmD|9wZ=jd3s20`Q)aSM+tM
zd2J`Pf{x}g+)XgpdCj^SKN~uVYcY2Q1Ft$Sf_8|KrPrvs=c$AREMv8R?YQZ<dIMru
z8M(V~K!irpqo1upW1fC2i6KtR8@`*35JS;7yk7jzZBp|6O|yNfy<oNRVlGQKV?x)e
zZnr7%QNNgap;*vK1m?%xs~rJ^2$5)=gIxPXS$*@72EAqtfO0LIh#{j84ML;r-35_+
z>d{MeVmcv?IDOLIIA5da&O&OH>c1MXl2$D?L?RP6^4D)srFq5_f5jU<S33v5B<d!@
zU(}w>dcVm{BdJg6O?`C4PHzFUJaJBx7UbN9C7OK_#3FBx?k8O#@8ohZ^nLue=ujRr
zOuBN>B&MN2oX^ey#-dUud_p!8?)nEa=uStCX9f4RA+c~4f>nZ;!I&;mxIB1ihWRo#
zR-4E}yPdO5V&f#~jU7w@jYw$Wu^K8jt{BgXADFsRs-jm0S7m`ocTs@k3Jn(*Zp}&7
zr|88RhoE@GWguBe^Y-e)!ONABk)NH!v=)<y3*oWmJ$lEU_|C4>rVu9mW++D&-kRT=
zWzDH&LUw4C<uMvgTni!34gaG$4KP@?T=Jds=f$SHj$;57-K^X#G6lte1JKghC)Vst
zR5dCBMsd+YFxYareY4_v@4HCJ(*2?3es%{L5Ow~YE<cb)CMpf&_tZgtBP5ggq}*8}
z5&w+Hi4cHa>;`|i`G(*T3;PaAZ<_uqwdaSZVJTa0qOK3`NqrWz?RVlCx+taf-QIA5
zDb$=5{wl3729HOYA~;?6gM_(}Z+N>{A^C(kv=cku;)z!wp<K)#j(!xzP+r{9MLdek
z$^>?cS9#pD+5cIMyA%02c;)FlX~HPk{N}JS{Ss%sy_WhA)GgzeNzgu4-B0)s+j9jH
z=)$OLkEItLkq1g5<I?EwwVi6HI144y=(l;PjJbh=i+EY|kzCe4qJda1@h{746@~Xo
z7!%A${N4x(e{PWiZY(7@5C^$OE!*jj#2l<oQyG$X-@hL_UqZSv8zC`Z3&O=`NF4Bc
zIs(2knz9-3Mk{FwXD~rJifO|BD*W`pg)o8bw0`qJXg2*Mj4n`}yY6*FU)<V-j+qN-
zSEvXed425aZr&7KfiVuXT|PmH1fQZW_t59h5<u+Xew1Kvg=mK1$^oF~rxD$PGhudu
zxkB+y9RT0YnaAhoY%%`nj(9cRFo*`|?~G}BPvb;lx_EKYpaZZudx&Zb#98-yYDqB@
zCJ+qF>CuIINX}7VQ}U(}o8lo`3o>wf0YLcATyk>w(uq6gg!$qp^O6o2f)MCVWT!I3
zKes=Yt=l0Pw8FXLx%4Dr&;IfJmD6%&Du1sH+*r=Q%PG*&8E3cFqv=^nr7z9k81OUQ
z%_DlD(KBiZ-a!<wFsxCJwr}&5qKjnz7W0H(K4!;mSeH`p&^aM+OX3y6_Vt~w-dJFT
z$}W5up{rt1&$y#+^yl5nZqE`*-2IQ(il=bN@UDKlMS9)X78<(n)rp~I{L1V&2EDou
zRcpVZ>B5`A1oRl;656wbnvoCmi^coDp{xb)dOzlweKdu_Q|a(Id)L&~Dpu=1-elmr
zrqCLUrXCs7ZeO5RYMhWz?4{mP-bm@(?-@NpdkVk}lhbkyMc+bw-x1djKDaSMID=b)
zUG;(NY+8oJPIh4H75dJ+&dX5h+UTAVbSZ65@~*K$f~ME~S#PU>)r~p;Sr3Hg(Jf5w
zB_(J48eH!96`T8K-}WmEXZ1I^@EcW4sr%9;V&3-{gDErTfR=c2V3nj7gYd|vC#v;Y
zkeT=sA<mk1Hyngih1Zy!tN1=gn4$ZS1T)S<5cJMH%LqEQJ@g0{!Tx!T*Ka|2W=`rI
zItFi&ThIf;uP{g`cO#_4BJro7t6umy<oK6i-PiPS%L{f*&@8{Q!EFb)uLM?Z9}bt<
z`QH6SX?=Inb%JriD>JqJ7v=QbiCuS~=SBUh7s2XrUQ1AmY`-)7vR+K599qq^{njxR
zvp&tK#0tQSIar=BaQ8AN0=V_I9W@7b?j@wPbPa)DbHuoG366>WDmIqSFdO4a-(`~U
z^@omsFm@Q@&vxE^Qa<4y=;c`MCr6r06>w|MN$GDCxP(9UZi_H0G#BXlyG@v418w_1
gzVK<yegCWgI7it?QjdXOenn80e=k?{)-3q{0p_HAE&u=k

diff --git a/docs_old/logos/nyu.png b/docs_old/logos/nyu.png
deleted file mode 100644
index d93d0fcb3d48a8beef4b6a5c6a12cdfe08e81331..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 30682
zcmd?Rc|6qL`#(G=Dp^V;yONYb5<-?zBq4itWy><gHe}aEo26vQzVG`s_7RmW`xb*C
z`!IF}W6XVCQ=j|3ANS+?&*%H+?|v_gm-9O3x~_AbYk6MR@$QkjGQ%<UV=x$uLFN8k
zEf|bC00yIpJ3<Sdj2FN>z_&v-x7BXLV8x;I`%eyopLs3rYpKCtUKe4ofY&hC9(WWm
z4}&4E!C*LZ7)&}E24i)KtJb^)enI<GRrxNA4E>u_ml+P8(7CCo-J|<Mdx-u5vxk!d
z2?lc?SGjvz=lKYJ%){rj?s)UsN|j`AxA%%rGW{jqGet>RhaP<!r~ln9$XsM8bZDu%
z0%<Z<enpUZ30|zv{30wMDWI(ToC2<HZdcKRXH0iu`)gGqVR5TAL02qX3N7ZkQD<Y_
zDau#6fr8j<{buL!K5b8j8Ze@AHhnJ(d{A&X2Ls=jVN~EV><}15#SFe2QUG86%ZEn-
zz&pIqYyq$%;3G9J`0~Gb_+PI3Ka%~wS@(Y=`~Ns%DB}9Rx2d(@cDNxfZ0yS&XzLV8
zZu<ZE6mzgYY&TSX^an*F3RT<hYm(2!e~FnLI*sBHNW?xm1?k6XNqvypBc)&+6|H8+
zjvE&;>F!ZMbBl01*NMXJZ`ek0rS#vBfAja@dge|)c~u>~>k(|!Z*>Ytl^Ir>GZ#>i
zqboV(e?WR$y4SvX36Q`Gd)_5#LEK+PBl@a}y%-#nUqFi3pvrzDJN;Fl`#}>~?oZyq
zsiTw`rlX7LS`ycrdBRDID`|RM>{^#7v{9q6nW^fy@}WdgC3ZhUv-gLg%{9HYnk8)a
zyKGy|f4QYcHo`zA{v{PyGXN7@ZQkHVGFGyVVU&`H{rfPM>t<QFG~ro2R$kuif-5y3
zjXEGVxX#k)zI@^sD#EbLUSX0#TksywxM*4r+FxdA$MF;sMA_*47geUC_BDJ;Tb5SL
z5KT_8fO@^TgpQ_>3YGS1+#s6a9(a5RhFZ<saa-O}<Qr-3j$1x;lJfeW%Dz3N{qxP=
znk_HO@<lwQEZ#Se6MkYn7L67t)vIQdE)hF?7@8oH!@Jddpg^BbkSP89vHr1R5U=Wk
zwQ9ax9$uS+8+h+$bdEGM5c%OIVcuV~v`Cs5GcTo_J?@bgjg*;LxV(%jATDr3E+-w<
z_w<x0!VRvN%&FbHevv5m?0#leh#)mYTgt?eeNYxqo8qHU<$VM8=0@+2Q-=3cDZ4Uo
z^a*63EqGBcz(}eFHn3Lp(%|I!u0@U0N&sKY_kCbj!LPQPqMhk-O@=Otl4X$`mVz}1
z`(upIEbsaY>{;S{){-hoa({_dpiWQQLvEF^Ol!QuHjm#{l-7I>3H1J#;;AXuMZ!=_
zMM35)9H;FAJ$;K}99Bd2-g|X&y@q()QG<@6*SFHsGv{;mcgFWfYOq_CT=5hewdU4|
zR2-;KoqA0=Fr~rpU-f|)ROpqO2-}uk-t;>v8^|u+*PpIWFHHjx{WM?w)k71htwU^d
zzkZC)nJ3Ud8yFCMU8(xd;FJzpQ%#wZN#rGE#g7c~`Gt&}xHzdDphE?Q$m{cGDfGR8
z=v&@WzPb2n5YcaME~7tG4ulERQbOP?w4FBDnqq20V`2|%7cPg$NJwhXO>J<Dt$(*|
z^PhcaHTYoeE)B456P!`AT)Ai9@m1?e2v)iV#UKyuTt<e0G%J&M*Re<HHy~klUUPWJ
z!{y~8+01R$r?=IgWxz!seOl!RX7gY5KiHiQ_sUh9`x)m1F?TW-9`&>O8sSnikBvK8
zuOJscVNp7equ7K!)L2?56&KD36iDGuf7aKG#TB;nvX7^ldF=HPBMWl!*VuxqcKnAP
zEz75cOK&~<*-8iYo>!D=M#E!v-}-f1Yc&CZ2ioT1Fvv5$EW6hDA}#=0^3Tc_RoYx~
zn0kTKctt!%t~K1_7u_=-B>^*^Rh|RFZDpkHVo31Kqf|ggziW`2WN=`74K6ykE6X0?
zkw@>dT2x?%NnbO?xHKA&q)t%ADQ`$@Av?JEEPZ|=o5!<lT$_1Pu`7MqQCuZmi84NS
zJ%&C$gLK}h_#=CE8>#(W0Rwlu_f2^iBL~I??YLsye}CYU!1i?7At2=x+xZr5cH}th
z*xHpStb1JUT(axl5%Z1_HfgxL68QkWW9w@ob&?+Nds7k?g3ER-<jI-L+)6+DSt-FI
zjFBw?kCW&}==All+zDlqzmH_MTqbl$BDBq9nH-v0kAnB^s=oSJenG{?Oy({_O#cI%
z;U;Id9{A}y3)~sU>7`~g)VF5y`rv8v34!fDPY**xal!2SA8tI{IjTH0i*y?s<C`IP
zPx0kic-YW3xE6Bc^xyDr!EpJw>01q|&%LC)a>C1xq`b9yLiP(%p*<~E5!h`tzxfLT
z43iaUeR#N6#N7oya}EaSYWv5S46C&HM^V@qI<OXJvWZ?ecE42gLA9pBlvuUb-jVVy
zJ6yJVIllk5Gg8j~&xsTGskc3&w6tLK`N66=vijN4d^7tk?tWogJExx!LADc`CbxzP
zF6QGd^NDqcHj6DykK=v|m0)^JyviQ|wXAdk8y9eYlf8S}gG^e`=Zrap=kalfRIh#(
zCV;X2*k7G0HYien&C2L}-`G+j+MC?$%yAZM;SnP*B^*!v<&km?)h*eySD6@Wwz8Os
zI5?n88X`j#sw~Q!%|9fMYv)V%eAOW`gYycAY#eemH%d=CSi?kTzP_eeh`Hfts;te;
zoeET1gWwDf(MiAKG$HeOx0|Y!;Qsh(iQ&T2cOgT+9|BDH$fEzsNX?sSZ>ROKkH2oH
zdG1&afDPX89nC|MULPF%GF+(M34Ef7T><B(G5})j5nE!<3q@)C#+-U>gOTcYHa-()
zj5BbtDq5SarGr~1_lRA!drsmGdNPpxwX_j_3QjMtj)|9qKItdXbSrfwnaiZ^;<CTY
zV$3-eaj!2eu;1k#^bGP_0ntk}CPNu^qbb6y*Maq5e&3|L`MWStOo`pMUQ&rFRIUo8
zUh8D8wdD$&8&3QqIe)qT23okrs`8_V1mgGE(iP>i6y6>@$dOOWzQHI{AImPDJB5=V
zu#K8dA*3%}Cf3+eD^Yb-;)<d}IeFX;0ox9kBDqh5spps!{`4c)%Kylm>Y861*{)a^
zeay1V#qa;X#qX5(Gp#EH=)$3Z3U<%V+rZGvGCHKjl#5j+QQjAI(+LejNqIfwNw-O>
zF38{F80rlUjmrZ9RKZ}^Y7ZNix(60Sd-?owBpnb~!-0i%xeZxdqE_bZj6JRq(P_Nh
zKx=n$vxlp$QQvi~FT^O8@B^zC&uJUhBzaPQc1#f}6;@thoHZ$Lx$LZsHP)Ywadr0}
zD(l<%c;IFeQoU*SY17wKM16ycv321D!D^(o-jD&3W4wGXpZDuJcx-3z08nTBvW=H2
z%61K(Oy-?bw%DzI?t9#DFf?LSmlIru+xM0f#GtY@1fX^N(-hsckxPSDQ+zr7t>v3h
z8?VKPi{SshN%Y%6_FDh%D})6NkD2fLg#Ax)uP}YBz!Wx&ztl{l<sSagqnZVU9i55K
zFT|6s*oyjj&gI!NY0S4ct=z`XTLQJ1AoSaYWh<tv`a}Y;pX2kEE(zwU5${V}{`IK~
zF~4d?*Rc5|8t>h3B3^ub^t<}0aX>y>H77VPJ7&!v7g)-9oi<hKI0<wp%~I*=$xq=C
zpS~*DkUe0(-m1YI29lEt66DTL&fSUM6MxRg484XpIF%S8%kO1hQt@o4K~v;anpN7f
zUK)Yrp^={|Y_Pr*9+=qmz8jPEJvo|vY%gST5m%(yp%=?<D|_fRS(m3)ccp49<KI>Q
z&5!VZ34DZ-GWkXSB`mcBaGe!X{TbSP&1IrQ5dHz#uW}SW=i8omVi58;(0+;kmH1Bd
zRcTL={mVUd%CWAUx7mI2O6?Llj!?#u!DJbEG=@sPXj`_+?6+<epEK|7F+t+2f_htt
z%vr~j;AXx|_@q+ujqlEIhvE7sjTy;QU=>b*$c;%ZcjG^if%ap?p9(5TH^&b~ncC?k
z?^?;jr6mWE9>HI*E486+7HQkVY6}iP_DMiu(w5JcB4=^q=yG<6d@h3f0N4!<9<EfN
zwr=ZkhjL!UtPORYAJ=HgC`WEsSVS0Ls>BhA1Fsn&bE*{Y9|l4qQ4D(RYc%aI>=c1`
zW39f&uD3*(tCnd%mrC~94Dxh`8B33!UH`+r;Ls42`@fllVArXSFvDJ4(9iT^;6iFG
zVf$robfR8#9rQ6HSFkaR<3T1eZCi66%zA@dkMR*#w>eC_l|x35$b)%yKFHPH_mZ09
zOal%NoWgjo8`>W#*1o>>=q@T~lbYj|_TU-ZFS<9BxUM0z=c4<a$nHNSD(&=4K)<X8
z*{t+HZNrrsgf6^N19|buQ|c=mD6HhO$3xNNkVrho{Cmvv1q-u!qEGt8_=B>+!tpa}
z^__KoS%Ja$shA3)wvXLF@(BS*;-7&d+y1aff0D$wc(#vZ103ylqQ-Q3sPA*a%!*rk
zWmK8Hjpuiw^i)pM4O}f%K`xbQBUR_;J4XX{TCzUnuyoF|4q?9zWJcHxbaWSR&4><@
zD`P{P`*T>-S&l!JbtMORZLNu<l%TamWbRu0<XUPoykV~$%0lhLtOR>;lTD%`6nCTL
z<I3S>Nt~+_F?xwu*+;Yk9^P$<=Mu&wJ39)^S-$GgdyfWEpUMjE4nr0guvKA}T7COR
zYv8OOFb!NKvAEi!%Du(@<{N?WZL%9|z8jP8vz8G3o-}TQ;)=#5;nt(nw485@8WcD=
zMRPJi-Vzap9kX$AQ-+DzrQtL@f~WI4<%?3BUOComTEQ`8SN{02@x_xeli7U_U_(Rc
zQhWgcsn@8Fe1B{gV<sT5T9v;SY^Q`wnfTy%(@+Q_nPlPL9JM0S_oXy3_`!~<%GP4c
zf>gp0|IhtJw(-&Ge%vbYM#yI^t@OZ3&r4e`QQ3{ws^1Aswm?~H4tFItTP?Qq)Xn!}
zMTxD2H}E-M1x0e-KR5eY-iPe7ut5GkI(AE&ClQ<F7?=2Bj7Pxa#v_zGl2+gvP~Q}R
zRHYpC>gSIdLsJz|Em42$MsgB_t9BntZ=ruD4VWQ(l{Ln9UpRCt)uIE{?Co*dHd(;@
zC*<nly~H_@x4xG5im%7L(m~zMup8{0uvF5TWK-Y`43RHQ#7X9SU7dRCc3If}y~fb^
z*VJDlzX$|--3Y2NnI|D~-kCQAiS5gn2Q14R__wYX&%{YkKmYYxjx3eQ&~y&inzwE9
zckph(a{fttA^dz|LMeESD$AnkZO7Z02W<%^r`ynf2C}X$UL@U8>;9n~;y&%_(xa7G
z@2UjT)*?wy<{glC=WuN>0dXjdk%WlMEf90OGJItDp*ybU*9lZ{yd^Pmta|28)TW}Q
ziq6ZHH{qeZ#hIKFyVV2PD{&_pn$+jFW1~`Ya*hN*^z1O1FwDfcCRv2o^i^{`fk%P-
zIYL__4BJI2_Q$ODxzCvFyp-54ll|j6VE<T~WrZhysXqVnbhVQb?4jDmV!tZcUklDr
zc1(v(s_HiINMBh*II}=L!s&c<@WD!#@!7$OOf%<8s&%9`rhEJlVP(GU%Wnw-AO=$N
znz$=&rzc>Gi;A55C@<BV*kh(<I7sS}7u%x)YVmdeVXWGuvA7sLwQD|j=d^!~#55iV
zEBVtLW0InOF{GUJt8cRMa((94#@u%En<5fE#psA{vW2f&Wa$bh3Xq5R7;cd%a|Ltx
z7g-z>A+RMz8slrXTQKP$fNaYaXql6A%S!u$7{W!aiP&n`-n&Z_l}iP=gmJ3K+2}7w
zx5iz!HmjihuF|THu$Lg<IS&Q9Jpu&xUu(DIw(5$M@;WQmujvtYeth<nUl?2y)2XPr
zo<aiNXC7xOFb_P0ijk_>%S3+oct(9-;vQNA1C9;EUYu_SbMwQgwT{u5z3GmWFx2Aj
z{I5!JNZ$C%wrl#iNDRN&dDF2})25i1{caUp@0F^|-nHG6GStsOM(9u2yNc0nZDj;A
zt-9gI?$Gf}ee8*Edv8Y)BG+}36^Ow<=eU~k%~Y_y^;+5@-uF(=e0qZ(<l<NeR94pm
z5$)&tjEgU*oFEqr5C=37yS<dd3`G@tJQ95I-lP-P8C%uF$Oq{U8ly8Tcd~s4s+2d<
zR=N=KyPCWD^(k%ZxVPJV_6kO7KIS*tKaz+aR9<f#*SF4hGx2FElHXtPb{C~EdhNKh
zbi1f)72FV|SSWXJuLUCxq7oZp<l|n#%*!r4JKT&&u3+O_BQ4jME_A8@BE#Au@&FaZ
zKTSwLc1+s+&f5m5)^#Nqq+;v$zTk!ohPaobr#%0JFf~@gJ)bOVw&g{P+LMHhD%JZ6
ze8m5(di1N*A#KNtM<0+^V3m9aR7Hd&cj@i^I*s<SSQsS#M^N$$BBGVI7MEktN{Q&w
z8`Jnbu0hRLfk-u+-@fe+>oIA+z&1XgbLN|eX^`9Hikura1H$iGF#dSdpA(*Z1tIU3
z==<#mFF1{Nv_KsAGerKkX!AzDtgGVqPY^(jE(Tph$~z7}$}#%c@pk@9x@fP#(D9yk
ztc=r*J2Y$chi@%sSZsSO60q)3Cn>zWES_T0gUpcg)kJiS&b^JH*R^A-eS0~`=4q2<
z_}rju^}@@!CQ}iEy<Pl*>PE<broh&0%I^=)Ltr2_OjZnSA1_P{b_-Dkfxv}N8TpF~
zm128Re;!l6NJ^+DG=-sZbtM%uP2KPJZa*1belT|ZfIpw)Z2y~hO)m9{ql~Gd^7cf^
z@1W1+n5lK|5jVRu5Tur3<-2ZNyRRho0QPz=FPF*V7e@lmnNxOSZQY8$E=L`#lss-t
z$+@1i-)`2k{)6NEasTEM2ir}Za4Tmvsyu)b0CSlCk)4R<oGOT)abpJJ+RnFMIm7Pb
zPyy^9^AWqBX7co&HISnQh|)zX@^I3CEX%R7)pp$zZ_w=st^$BiO~3rjWbZ$EG79TD
zEYt+${9zTt^`4SkhrnL?k|)tl>pdZ!?Hf%m@8p}+2~FcW5BeVMYJc9~n@IHNVpxPP
zm?hFgk#-+*NmaA5e)ro;&bly*wwU&$TL_y>_7}xe*s@aQs*`ci(LgRQd>pvm$VFNl
zO|4>NACbCDT&dHq@5>WFAD7=}TlgFsI{7A)Iseiqb03KZ3WlpBYEi=|_|rqoAinng
zWFTW=V<Ph%sKYQuwTYc1gMY6W?vhY_&#a)A&)-ibWmWQE)<PUWAHspmLL^7{ukA4a
zDuIwl{V@w-<#PVp=XEw<6Q7Sl7-coXbuQOilOHXuxVp7Kq!{Mri*g*0P(yoVP!D`t
zC6d<_k-r`{nq`4(iLEDg&|4eeJd~_2>+6}pZOronCWxxOY70I9x5+nib~TFV1tN{`
zW$jFA)%91f+S#`HwI_RwvYLKw_WV|*bv^fbvZhP_rzIM4hclC+kbC}3oY<8IDdy4u
za4RnTsw|}mJA%{jYM$Rpb#&#b)%D)1KPAh(c4v!FD*$ErId<dY2i}Kf^#tF3!q>=8
zDXmNiY3i~u_sV#vp)+&Po%F<n>vY`UL8UI%Z%?rW6YtXcO=ZZVl)M%p=UIAb2`@u%
zC|F$PzKl)`ZFgF_oPT7|c#H%1UY2FHbZ)xxchAj07y~ksl?Uzjdpw7DCbKNOi0KF`
z_%lpdmt78om+4}K4(!Sks=Xb5V<1dJQttZ2G?h1NMv1isX*q2Q{o6J^O8RQDw@E>6
z@7oT}QXPWklU{rI0tCDT79p7>wT3*uj8Cn*^L5Ey%ie=GgF{H%qdrVMS+}VPk~~K*
zhdkWP`E(vscFyr<KP?iK#rI9N=O{r+N<~-xA%zQ<2dQbdQ=C6kL(_&xXM|#nPp<ds
zGQ<ut+P^eLmFQK=6UvK}J$ie!-I|&Nay?`Bj1Qf7j$8J(1#Ve$II0WbHq*~?!}2a6
zi+gI8A>Hj1ke}b;{!!&iLQMuytu-S4tm_iZ@#7kovk%f1P%r)c(sy}+$fLtMv7wq#
zrb3ml^Q)a7565h5N`mP?PG-OIt7pQZB*ia|Uml)W3S~58FNSBi08fy{MS}=ABc1+E
z>?27&F}3EnUIE)Z_`OM5S>KhQCkS}}ZRf3UBo8IVXEtZnb42*2XY3-!WK(ZUyqzyJ
z*4Aa~jtr(a%L!IFbkA;hh})>M@6F6>(^2vOY{~~P<L4BtU|X&m0a>bRa(Y~qu8y{%
zuZksPg8GQI)xC1=#Jz;^uf@EA#9YXFtZ?Lm6tsEzxnw@(97mZOC&at^^t>qmC*Gxo
zv>Gm$9sZy=zC0fuZqqShvrKxls0XLF$AI8&vAV`3dYS(rR>ZI{$k9vtN|fWrwgZq8
z<#TXL8Q@D}elA6;4lX54LHw73xW#QRB1mevGE(?4$NI(txU{&|(0J?GRijBs`Gs<u
zxi{tcSn`ILb?q8QwTVpo)~`Aq9G;CnMIvUfLe*kyXZ|b{&}1~5mRQShK~5%-%U?+z
zf?k}Pw|2c%6~||@!arMV7N*obU>aXl`b{-mc7;)waF8#xtGKy<`Qf!R=uu{|n2?1p
z1^z1LIrbTV$3l143Q5lN@nAyOWf0uDbh^*_c$(FtuBIhwsBPHaR}W2_M`bQ2<Zk6Q
z)ofvGoV?@RCKc2Vn6K_Qq$kvo6@Agqe02yi4h}rQHBTOoMj9I{!hoI!T;TGyGgBU?
zH7<7umw5f0h$2tM>wx^Ni!j`K*naEh#s(+^sFT>M?Qm<?<(gX2Jnv@<drFyn(teob
zhv5ABfWoc?H@Wq{tNSyOATIBat_@uDXA_VU@O8<wd}`jl_cqo4vs+P9M$x-B3cOj$
z2v30}BXrcLPaaOP;72Q(N&ZyC{4i0zw>DJot}hnEe<0sd%C#U^JHxd-4I(58=lGdH
zTAAk2vg)8LM-))mkY~g*A;1ctnlisD=S}+RxmR3yArBsUcA(FB<Cb(sTRMmMR8QCD
zu*{Eh8uDfGUX0IjffdQFK6(~uMx2U_E;|Z>Ut!nL1rn#HagAzwxzAc)SjB>DR{VAQ
zbnj|(h2v;CEB79llS8sFmEJY<Pa2s9D`Hy6%5$KzA!p1T9yYzf5zHn3IMew0%_k`A
z(0;>)KbTXcP_^(3ZnU(DFBCq>ntSTVA;l5viJozu;XPs*q0C1A-eXJ%dN;bV)(R(~
ziZ;rO;@-V+5#>eBH(2zl_8K_??ECd(IH~v<9~X%1!j|G{bw)fA^N%mg`TC^-U~k*a
zU+60#T+TQ5W*(eLR@exnsVyqxD?hLW10O<tQb2GfYK?9>yD%^6>vr&(@5`sJk*g9x
z#CaId@LT7F`0l;py!EtVK3De~$p@6gZjrVN%X{UQ_w^rR20=_iU%5q$<t6=TO2k3t
zf0F=kVQmOhPJJm!1w0#XE5pr{Qj5eWOnIvDqF$hc?!CenaM>}l#5$8aEBc`~<@l;0
zCf&L~@$iE<k$tan)8{+dP{RJpe+WR0H<o`|aBihrwg+nANA-RN@f-<IxGv-K+EqZ+
zw8F`xg=vy-tEGmjs9G>7b0wZ*umYg3-2muT$GO(6#wKo#((q)1?@teEV#*6LKOoL#
z)Cajc8FQT_fpWur7jb)=`G9YDE^i+oTaJ;#@i=`@B}l**qD=QCrjHh<Cnn0*IIU}5
zSCPhU;8r9Tvc2W|C0(PM_dS3K6W@FJLV7pl_Jek(*~I9XHyno`IAd^SmC@PAU63<*
z9Jkb)`=(b+OLkcQ^o(xI+PuvZxXgvjAPAsyn<Hdw$!zVv;#o8+Wet)waKGfmWS8kI
z$*BFooq^LN$_ks$0`}l@bRRa-q{>bEzC0En!d=?~G9p6VA^F~S#)W~cwJ{O9klncm
zAtmb;iHpH=v)Lsn(zED7=M8CTqJ|=J-g!8rFu1(81k~jMU~i{oceP14@B4J1c(Sm4
z9ZJ;G;L?`jYnEoZy0kYo-ib7gxNq*#^e3sfTqWq4obHaxtRHL-^-Ug2pAjYdmM;!;
zx3V(zf*4a!XpGBW9XR8AHSS;NA!Qa^42TR#IR1QVz2h9BuQQcum88#VxA@cK974aw
zlp*+LjlWi07d`rMcC(|mr0^AE%l`Yiol<mm0A?_jOqM|T16<;>aVQPu=H8B;5Iuey
zVW{-pu)==6$`53WTR1HGh&_fy<m|*Q|5^onB;9{pR-dQSfODf9PP#0|<@r=3-hKF7
z4)%9^YEE+f<HI2K2wec6R%$4Mgo6VO#;6C+Y<RDT4~3eVr3Bl1yskANDTyOT*+{p%
z4```x*R3txk|j?FJP~<^LIZq6+UG74sN3<1+naO^Ag}C2!T@CTsSNxEH-Z@!dFWt@
z7E4P+3=}5|5OKkfy&qFyWm;e2=z$JcDylX=Le-)s0S=NJF4s0Hc4s|*ja$l97K_}Z
zGM{h&QGZGUP~Y-zskQ@n-(I&;9Xl#(vC7Y|!YRJqtX6IskLxc?PSw~RPwVo2aAk8k
zo@X9Jl^+)H<9osOm*<uaAS9%~`zHc%#=l&87S##_IAIRp9O;9zlKJ%TLq(dUBMFO{
zk_oXqgYLlJce$qe)60=GnZ|cApxCbz+f(986D*Ag2%O%0WEX>VozuP*Sr1M}z$<|F
z{_1Jg2kyBOp4{O%R9S^mf_XTMrD|XWk9pRB$^w^%p;=ddvV%u@LhS}kos7lpUrL42
z?ud+V5%#nf%E(V^o*q?S-u%_P!Q|LdWta#468&zdZ|dwgJCzSW$aHksj(CEK^0rPq
z?bn2?ZzXRQ-3VKU1&><n){uIpk|dxLhMSt0VNSJDRmmRAA8iZbQGneo0SI0$0P;E+
z17s8vvnu@f5NhuG-+sM0hp_ng#kKQA0iuS%u*2t%sw}ZV8E8*xHNql|Cdh3NY!Kw6
zOK**PBo<CLG4aCAe-(%8cJ(Zh9~`q!fBEs@%B_uUQp%Pg%Fd_-gOo1{o*GIXi-N~`
zEFP%1mQ9q-(?y!njNoeDDQEfW^`+G;&-s?UWReHR%W^PJ&9|AyB8#efg>adFK!yqL
z4+5Dy{OG|fqGw;cdJ6Y-q{7ofhk!|e`^LMHPOv-OolC2or5U*GOI!%`*`WgAU+JuH
zbW24_tsrpMgzkAWUgmxQMBu@?GAPlpyXoi;RS@H&&de8Mu)0WHUOVEimWO>=8;rdw
ztPJbZMs>tor*VxDZNqZ_-@G6y`4Ki^{tOXd*}tj|+h5*@7^&yqJW+H`YK%)BHMNp`
zwQ#V>6JhK4?N1iIZ%&e3A|q5Cz5l4ImYL<KBboY;Lc{5enZwk{vO6mT;;lY;9(4VM
zJbcKl;;6-uYWRlZnBT&2A+y?Zc0>xyUP;<X^ZI6yNVTd6;of(9bG5O%eCNL@JmU+9
zI6aUsA8XzaC~nvX3i<jEmr3@yQHa6(DVt}MxHYj0apTVt4f5ydpemOOH}t@6=YSv~
z8JuJ0)`9|h$9D$&Hn4raH80=|&aAt?3hi*wm0|TO*6J&WVJn~O1?>w-Ak8*4peviE
zCW|E2>OMRH@ONJQTP&w>*R3OF#*YL{%H<+gMVF5TkYnl|gL9yyGfNPYON}dK>Sxox
zjw2{H8mQ3%bH)vDY#orLPlby0M@fT(82q=GcsP#iJWo{w+;2e@iMUs~vANUlky?^w
z1S$(|_;|j7xe0gU-~JPsr(?_QhA$oi72!`yT7BA<#zp<MUS|<ok&YtTz}=Z(Bry^z
zgYO%=%>vKPf6g{D^;>e4b;Sm04(o${3UN?(04WqB>z3(jv3Lk(seRs<S#9z9Du4l}
z;S~;-kVvqbr94gSY_3hx6<wo*WH7@<*3#l{ZN~1yRm%KZwSL4!?F2G(mM^4dR84&;
z)GkzN!8nW;X>-*`y27!iAcPDKl8F9%#1&=K;=6^F+=)ap<=Vu}k2=BV-Lni_!>{N)
z<@X=yG-A+T4|Zl=d$hXPZpHS07CmNr*I^@2OP`>&_OhGC+#RbI;<r0lMJM<E!H)fC
zN<qdgx6dO;c;mg=#BqP*xqGnfM&+ZhITck6&HU=*U}{HGbYD4M2q4r0?TY@7(I9*c
zf%oNx`X@(mldk?Dl`W3k?MsyTu*&+f?LDq&C>+#c1L$<Cp4HdVs@c8S+>p=zy)@#t
z>$kQSCV-AiGo^`VgPHleCnQwcIDRKCh!@3Y$oVR@%-ty=uGd+Ol$XZc0S97&tV}za
zqFLHKzPg{B^}U=_;&%Ia7rDI4WjyP5P~mCZTK0-kZ6d%uGhz><l|VLI<)pZ|d}Z_i
zl$;w1b-w6V%{~T5^x57n#8{=OOvSRZ75b`CrSLQyaC_eWeW_gIG4!qq6s>>X*az5c
zK*869sTdo*__XOvyy51|14t)ckXf@Bm8=fa3eYO3>^*rWqf9lFcZ3A*P?wDZVe|+O
zN|5|s9~j5C&*bM!<#pzaCHqfr=t}q06#B46ysLohyr?4^KfM5~;Bm<}BuP`3?NRFp
z{>h<+*$~^N&U|mT#T`(3P`2!fkN3qh0}BshwT%{R&)ouP-6@b5P?PO3872icI|0P^
zrnUpvLN^nChaFp^$07l4E-i6E0Hs!|`*Rv}19;yhN|D-{-2S|wsZ!v(hB7a9cf67c
z9ZDwzEDr!gYDNR}wy`O@_B!>My}G0motAZeXbphsyM13H%aZGgE*2k@Q|cGZJimUh
zQ2{HPS)Tz-u;XiFauUXWIai}S%&#7Vag7K+7hV56fHAdTZrB<Ooe&A|+21=^xAg~g
z0&fA;BS00y>50-jM#Im1h6x}Z`=XL+(WBzuVV6{ZXY0`=6+7xeCp6aUsQmH(7SXFH
zS;y8R=h7+l2uSkFhwH~E(O-ipX*xyoWgg8PrT{a))w<mAm!pO!_fEQZ7!GO7i>?go
zCfn+F=y)3J^c`bhAJm;Q@jt|Bc`DDiS`+KL&G!a$0XCRYPq+~BqI*CtXEn$OKR*FN
zimV-sc!BNIwoo!UVk{3?_nLd7NfMooyn+g>^Bo!&a_cBNab0fK>s=r#&+eCOGhm*I
zJQnp4t{uh!KxlNFqrmOU`E<xGCIyNKX`H8-fV47P@<Zczz3rkGfOcB}crp3(<I4zD
zN!$mIC@qlQ9$7L;l0G}=YV)plr6}YaVq?MRN5PS*!3Q#^H{7sAiL9?-Q)4B`I-z6i
z2iFnhILrB;%0IIlUp*B5e#vt(S>vxn4vyL^6yKGa=w>{J6}&oTnt_lroZP9vj~j67
zJ<2yJRJzlMqtP}UTJBxF=&i1@We>`wf#TmJ{jdP5-&-eHQS^$K`Ru@Oh|?YfRs-2a
z^Wog+gH6d01ZG>PrfH-=owP&OFqBBTnDVZo3sls+IQs)Xh`cwo+ysu8I{C1IM@NLm
zYO%pNL@9vt46EEyQ!(Y<HuI^rKW%h!)s>7xUB`0ZUbuuR04zLvtRL@W?Jrz=Y0t06
z%Xh1A<uZV+4&B*#Rqv8(KX@oItx02EI8lC6Qi=XxVX3Nz`o%Rz@?ghXBBs>8%U>g6
zX5a}NlZTWwab9#bcGI-X-~7HwV1tqag?WDk5L1naH)E1igtl%L!CNrC`#(>@F}0Xt
z{4pDb3HI{dDX))-^}2bUPvkVdMkb^jB?L2i3i2$CDHpQlJUd{g9Q1DAtrs<pJPd3-
zveWP!L7JGY?@=Le?VEf7e*XJ%6>aYNIAC{!>PWz5XM-xcaK)vZONG5+1i$;Bi<Or-
z0U>8R8M6xmxd>nTaf{8xj=4Ge4O$yvf7Z?;Ng^RA|3vQzKKN0K^ph%}j9U@Hh6$n<
zYBSx%u#QZp%y2@M;u@5*w%kL#04+9sCsMrvYfS)bV(PP_tz2<A?u3HH9~PMZRpB0v
zt_~)93@FUmjRrSWwxl=!uN<>A9YxvKPtF+5r_G7OB6+O5a?w@y?fu=xr-#0o&r@BC
zi+a+yHqAvob|vXSFF=q~T8!VA5>S{Xe>OVkkcVoFU)9pJY5>BlRU;MY(H=eY@`Rej
z$twVT3PZhXJ~Dwy+|PPrmUrA916f`wW%tQH?@L0>%o@u?zyxnYdhu2TM#^sA4JJDT
z2s6a?O;7|0u~Y51mSB6F)<T95#4u9S<O0R=J=3XGMJWydb=$-Adr<TtWiCDn?K-fw
z&E2m)wpY$*k3o;`g|xNjaP~zyWhh_9n6UL(?5%F`|HTDF>YdM`{46Gljj%lcl%_D>
zQ@WF@3j)H}+>qTbs&ql+V<3AIpVFMGW&Ug+wh61uEIV)A3_)2N{<|R6rsD}u2YmuF
zFW;Alj2mTWV1+5>?yazWx@2PPT#m&u01v7OOo&rX+w%2;Ah(p_g9VCFS7x5KzM@b|
z>O~0}l*&%yLH3ATe$7`Ios#F~{q?}B9YZ0fc&`bqfRVtv42ULrZcRRc>w`?qWaxBO
z$@w*ewzXx(n|(dRHFI6fGbBvI3tW=5{H`uQ66(zZ*d+^_-kh2dv4%2$RQ(bv;8ojo
z^>lJ4vJu9}U&Fpe+2lzskOXcKVjJrR*wlW$_TYH#aomxHoNMm<nB}-6Xe<Jyu7-Q3
ztPYcR+K_!WK*-wgX58yZ{R33Z;_VH3mryOc82LQ74Qc8Ml*%_^x|3}2VW{Eq)LS{t
z=C34{_%}MM9J2s#RRJl_7CaU=`|DgOJN2{><Pz58sVn+iuAt=&^m@E%WRxY$)fWKt
zjAbM+BOe!7+mV5kN%1;vR*QaU@Av!Ux;xm@K>!w1Jx_Ty6yh?5#9!4H>H~=yRn(Q8
zxuGX@hr;d)gvvZ-XgYf@JgtMn9asO()zab?YJB{Ly{^bKT%zE^vSCr-0scc6N>oW2
zbPHb(T)b$Q7qOhU&u&iwF)(kv;FX-M?_~rN<S#aHE#&RxPfrl!j-V@>jZKLcmUd!a
zv%rRw5Qm&5S+8u7{%oB?cua(K;`&qBSk+E-Jm1vzB^Az4c5aO@Lx;6LXqK<UMKf_n
zrhjX;ep5$M4fV<C!m9jwf9c5>wJA>>Y-dj_S%b>*<Bi$Ruc{uW%=e+*Ls)?}Wa}?<
zVl(UJa>QTq+l~e|XMuX(6S#f;W)uguyE?*X%5Sve<W1^IE5+>kT_;dH<GN8+3c58r
z-<;$J^`ZHC71yT{Xg<j8m_bN-*C;tTH>3&z?JlfD7oo7)hRGOVU#Vrp9Z6dtow@l3
zDW_mq-x(Q@1k^d0e}u}LPM_|!?hRnR`2V<4Ky0Ms*@8gShC4qV=yPp=me^+iV=O|M
zp<Hqu(Mwjkk<EzD^Pp41j{VQ22@sr|PpFh(K7t#&Huyqr{;e?7t4)95m^=Q`5Xc3J
zIz@mHkDHS3OIy0+G2k3p;~Z@WBChRiD5<V<_7khLAat8$>=Zw@xQpIc^cNb6{M;Lc
z;%G1R*VHuCOUH<KOPvelf?8I+%hS_zn6UeifJLQ_8)}t^vg>xQ<)zl<R)hzzQ#%rF
znb#6BiJ-QVg5BL##Rk@@0dRGom;-U3lmkFll*x5PAhkzOGw{tmo0U_G5!YtlsisN9
zv)lTN7lSsza+~@d)lk1PiLkSSN6dGqiYz<ag?1dQfZ#!=l<?b2uhAG0PlFF=Lp4z#
z^a+QqQAvH^HI{NLxUZ+w0CO#n;(LqGpzM4U4Z`y9wKv-rTS8BKg!%0&X||v$IRN-f
zrR+BO^sR!GXL*L{_WLu^=11=Uddw3K>h9dW$wDj83-tpQQo%vDbswQ~02G}*E;w5c
zKG6SAu>}A{fPIXV=AQortGqA`4@?}?KR6|Ky-61sE3Tnm{UGv7I}62fp>6ASYG{`X
zfM(9bSZrtRK+R8fJq5S#5_jiJ$h{a5p|=ag?l|rd*MPwXSdnA0VdS!>rFue`Q@s+P
zub1831tI+M8XiPWM$KJE>^tv44K|YiX2_21^fv|Rj(dmbP)=+z)LA6GW&gU4$vzzb
zyhez+9r&*hu^O}7FqFZmpZVWIv6qf{#wv!i>j6-}8202h2wlRvyatilg233o(8aSs
zLe-OTHhc*|!U>=xTL@;%-Tb(UJzjA;7d^gSVJ&we7W2)QX@NNkw$mD$OxM7tyfyuu
zsR<Zb+dm5$_A|Y~r)YqFETguY^Q(7XvH}pDdx+B&q+ZCaO_pgwkpHj|aRWaw$Gvk2
z*F2PFH5P`@6#;o}+8j*(>M-+qt!r@XkS5A6Uur?IY@&}xeBq-_nP-#wdl21Itvs&t
zDRsQHvRibKMg0qK1OvLeQ0u@6{9L-SeBYEd_QT#tKHYIgp587x>f1uQvzerWU5)YM
zB(-J5(3Lam9Br-3q_6bIMSihHKNZM{I=LU&_Y1!mx2JR0wf6_+iGW*Qx<5Df`d$qg
z59uJOsxZ4+(96D(oB1bv;!(z_UF+rbFoL}0T@Uq5l8%99J|t@Yuj_%-1GkS+Lqv{f
zY2}QHB{EzwzR;#Au-$u4JMbPS_1u}OgF>w6w6XLa1L4Bv+i-A60(5Kn75h^;O^tBv
z+5VUfqtqVD7e=GQedP0u%GFpjybTd`t3p3ORIK1~36ux3^=ngx5_W%zH~Wk(_yU8e
zc_OIcjah-#GFumv?4RJ#2JO051O~ik4G3g&KyF0buiE7yGq104cFtq{c6Wl$lNP20
zOvFzz0%GspyJo(%(@$_)`7Tx9xxE1OAHK(ghR*+BUYMIZClI$?rgzHm?N0O_*O@)|
zMAx`Ubywo<Y-;yh5{H|czWP#KR$kU1pC@8rdC@x8!Gr%W@O%pEdrs=f1+ojRMHert
zzXny~{?|mMH$ID7Xn>5WYIl07CxTW|3{b@m#@+M}U08$auB4Ki-!EGjRv?nB^B0}(
zkYB8@<_~RU?u*T`yoOY&pm;J9ALF*y=LG`mJEpD@Culi`?@7wLzX!zv<+8+e*$4;*
zqJ*%hnkS_rxN-S1Oamx6WvVTbpI3isEcDm%G|;_IT=ICVtUiqeHZ47K`D^aGO!?F6
z?*)N~1Lh6`ta7zm095x`%RphbnFkU2YGl4kBQ6JOt19EMPuJ%m-`Ul#W#5K>YYG7x
zwX*-G&dPrWJLk{2hsV|jorIAw0LormpCV^KW4{HY^8z(%;4#>o5i}8BcLq3Ub535C
z`)0BGa*we`?Q_t32S2%1U%xj*h;hPcKY4?F720}iH*#~_kwyeqy!*tqmWTA7r+V8f
z;N$r+h4JaYk9ZdH<pYXge(9j%50yjoXKxkkChkwnTfj$aLo#H&GoW($A72tMB?UZ0
z4d2~wgmZ9x5bC$^Io-n+?2f5asiLB6JD8Hy$i4Oon74`VHrqU9h&*Ns?^zB&XeBs%
zhxXR@#DSfxTWzrK&7fxh+EqPYv<kjPLKomNoSt!)v;DZP-!eLFH6UnOB+<yMnXiGp
zss%b4w%EJ%>PbG|n&u?;mhjAXf<PZ${+c|W2fe&mIzWj}_KNa&D9Ixi2F}}dY1^lR
z<$B9<O$P)JXQce5*%ev@j|M1E=NS~=L)6_g<M10Pa5ss~(i=P8%#|vt<V%av506Ba
zPxM8cC|oLX`)24%*nPltt&-$d=g^y{D_eK4bW2xvnf!o{-@XEBsBPc3b7>;j2<rf}
z@Eq47n?832fIIkSs&+xdTWxdag{i-Fg8RzVxxZcWts{v7qJA51c1M7T#FxZmS>xTB
zxon7e8Pd<5*?{OP0<^{4D=QS@cpwh)_7Qj6lv()V%9ahl%RzsQcDrJ<k&}n*o07dU
ztqln>dIglR?#3!UcNCK4(`B9JF30jkm`btE`Pbeds7I$7X!xbWW8Ix!fS3p-N}A#*
zGc2}zHSXhAFOiSUz9@vtF@25rqhtgCF68&D1Q5fT0Dv##nc3-?fxY>yI}t0{S_B%*
z67nPt>NUw{oekQ~*3--f=j))0?B|24IagO4Gmx1cBAxA_73U3_M?2!Rvxu%{+TY~S
zgM-0`0LprVx8UXNT%O1;^ZBgkffNya(1Nc9zy_yZ_jN#-4+_1dx5Ua<KSZ3+pwX%|
z&#(Vla|deE22cG7HaLkEoi&tw*%SFlFCuCyw_+i>*ca5jGC(<MA)XEU#?3g0dq@p*
zC~Rz2RRFh8x4kJ@?eT}RL+EiHm(ez8><7%htFRqY_uY0c2H;jQ2o48QOlSsL9=d)n
zMlKH<lNj+?{2P9~o$ya%C&%kOQRCjoWV@mK<dCsg@^bZesPHr;xlQJ0<S5J0me&Sh
z_d@DW<TDUHDoST)&T&E4zP#3J1wo^c#+aBs)7WOl8J}&wb`<3Q)IrOb)`xoIbSVwr
zeEyBWOOLo7h3jp7iA?z1J!RLO78xR4i~{M6pXs{r%Wn+|$VXV8$~#vg9Udj##(y#8
z+ma;L-NWobr^Nlu>YYXqH4XsSz-ysLL(wk*zyD4-2gFKUaf*;mZrRK(OvaA)gP@`p
zc+bmcEl=9;4$~yDeU0xDpqe-joh+yO-dxhPvEcSr>Ggvv*M4BYrYyX;jG!3$m0W-`
zn0)rgqval$ktF+dYMU*t^o@sl8i97n?ou;T{GKaR;PX8g*Ye%dD}IkQ;cL&l6@VQ9
z@q_VIpog-IcB!DEzX(coEbfag8n$n^LGsS?%kWeLiOc<tq{%I--~JiR!ys~{4|#rR
z7In2|(=J)Y^d=@^u2JKPg#EbkmCfCn_Ohj#^4D@uP+|bCN6gJ#dw$=0i~YBHIqtiM
z-#1U5{BEe(lVeCpj{Py(S;!>DBae}DMCJ45Z}`eQD6KVb?gWJ_CFibdI5lSyI$cit
z)fsNcAKcymMN~n)BFM}ccg{)LfzE4bf<v;fY;3M;;jCR>)tYIh$-BV!>w<Nw69Q16
zd<^8jD=Tj`Ws%EB#jS&1Z}#TKTrIqoK$Bba&yO;{Kar}CQ^KIlC;u4wK;;A4G+!5s
z_&j3(5f3ANPWY6=I2M!R+BJ&<?M8;~&$2)-)FoNIy{;1l0H1>5-AEy0*0uVNn#PW{
z>XnEujNFkCt9~P8ewluT^#(u{y4UY5me)EptrWN(hbDazl&jQ%%d#TU+6~tygK#_R
z44xhOF$Y_zcb6|v;@Q?LJr>8~;hWUvGFK?zjs3MTP+%WkxnBSr+6ZX<1YJ~WKw<0w
zJYz^AUCcMFHvl!Oo1#?TZ@@Ah(ePFo5I5fTfwZ}NqBH`{5G0rC#bBRvd&)2en-<A_
zqaYpFxop3wM6ySnl6VBTrN{O)I|m%gUu46!&Yh1TGIfRS4rBSihxD(NKPHi8i)hSu
z6CUX5)0NvRS3U6fVoL~diCDfgRKR65Qj83(`0~XLAe<w$$vQ@o_}<R^hxNyKUzi=R
zQ5|AdBQw5c`)Zp5+C{|e)9rtJK8Y6=^7yf4W0`T8e`Zdz3lz7~55C2Hka$*}S#}3N
zp}Jr6d)8p1uTLnzddh9|-)t|rtO)&Dmqo|eSqqSkHpKdTfx-TFmxS;9Fl>NbPyxN&
zd8z74_s;#X1O?P;Qc2Y1Lm*)Y_?}HcScg)XolA1d;?BeFpLL|7`LAy>gOmyo2!c}>
zwB7$+2O9goyWSi?vR1O5Gin2`huwhsHo<pG%54A7<p1M}|KEpq6_4@5nBkzcNe^th
z>`+0@dpn>yUm0n^$N;j$h0L1HYU{UiBf2{s8Zj>0rEb??LsMt21!Q4rll#N*BbDxS
z@syjLXdBg|fIUFO=PS_N0*FnOGceT$R{3-^6Vedm%?sNv03ELpjv^bTFiTLb2jdTZ
zM}Ce^u+soVwpj+5g~!lqJ58|p;yD<~;gW8W^!;Y&)drs7rDDk!RV2z<#G(jr6A_GT
z{o%?NP)h}+hcNHCavK=XE{Jg7!wR9nu*ke&^-Rj!;|``P`Iogpw;1KOmqc=VmwR0_
z%J?5~^+yJ$y9q+XEq%~Wpc$?8%X~?t8bo<b?p3&Q0?k!}n)pu2i=@ej6AGZD15H9=
z2;gR;8m$Zw4}ek*-fU`K?FK04Z8)G+fHVZ$`gB=qQ{{!;RDk`~Fg;6^S8(jvVup#^
zMapuapr$W~)n33!I2fa}C0aNE7Xh#$tVE5^sW4FR0Y5OGN1>0_eo{%IEb$%Ia)cMU
zSarbwTn1i;Rfa%gj=%tH8-=A7HULdR5M^n{0zjJ`^yv{z(uYK<JP>(rcl-hYHyC@)
zdBcu3b`bx?>KKKtBWep3vQV5kQ|wA)g?6OTJiiQ?DuYOkyhcaawAVK6f$;^mmen3o
z(Lhs7Ixo5g{QLaGWihtl765Fa=d=wcf!HY9%2fdR@IZAYV*~n`HxD-7Xw<-Oj}rSg
zWqYv53#Y&9oiFljchuE~t`q`_!mc`W+hHzxsL`tZ<>m<7eb_YsVhmiw4a1BdvSzzU
z`F!L5rDc-U0<g#^UE8^e>(UGm6X%Zw2(yFT0exMx5N)4a`iuiW5ebsYC8xe4=8&#J
z<4{QL?3Yv?dHEI2T>l=OIsO=qSqf=Qk}QBAU>LQ6PNkKwLc(9=c?N`ii>~EDH)VhL
z9iOvQ0?-U#Cz+)bR@p16d13z%-Q7?QCD>PLT4<@17KSd^Iye|$%DZB)NU%PjjCX-L
z3Atn3{g1&ETuuNmvw(LnEWq;5<{a!O7zH%JO5Km;`B0||Uf7E9uP_7!LSfhU=pgn)
zi;#R^2{7(+n>Af1tmGJFnePGNbJU|{jsjeVpi!Dk0UU80B@YH2Ghb0?9f{jNbFJ$y
zkH>W)^>f|4+5S!QjwO)>szH>DMYu6SBQB`It{ODW^FW`!3&W(H1fzo&EBZt!#FfJ4
z8(Nb*4=DH4pR&Ou|1p<+(^>E=cmaBc8QdCzHrP@^#z{OZ_$mz@L{qUg2&TN<b#Kl-
z2Yjcl%l^|-(V4U=*uUexb|iq~pLXcPZC-S0jlsM9<u4{x2{r`GA7v({07gRwOTI-I
zlm;qrQuF?2L8|U^T`<Ed@TdO-C!6_@whmG{;8f00=(h(wG`Y1`*S^{FA^yI;?|^K8
zo6r8$!J96P#sU2Qwuncd_UJE;2FiO~uofY6=+N`Rz#V%)MMFE)wQs0^e9#DI^KUgX
z*u&K$6NAC&;FM+XQRHrVQkLxoXt#lmMyI<YCKWWc{-Ngmx3rHz?C;hj)j^7aQt>XI
zlw#Yws067H74N~x-HX=%Ux4GVx3I^bD_L(#6pX3=y^+38Gg0t*-;|?lWGNth9`c5~
zM`_HT-uVqoTu#BUZYm1Rh<7pcVRK{(A-7M8Q~y)tFkXNV=>dEDPxYR>(le6$LB7ZQ
zcl$0pIzLNo`S<AEw}d@xa1Hnfu}!T&{e!4Bp}-5U%TP!Hu+J5aA6LWzj3mBXoc_T=
zv4|pnWZ9eXW{@!N7d@A!0c$`9@h)A%P4wY^vJ5jfl3dohW<JIGA0z%9glQx7Rvx2%
zW?;upaX}I|32*<oGe&Fu*LlP-!=^Sl%|Y#LsOTs(=T0i@Y^LN3$wDmK&>RDWoTH2Z
z{J=v%zhyic;9PyU_7kfDB%1~{iDuhFPE*o@Fj|Nc<=-bWXC|V-VFkjCI|+HNsZ~c>
zVAR3F0g5`hv^v#rrhmsnB`=ZYpEW<l4x<V?33Gb?7?^73G4)|yR~U5XKfsmR{G&3Y
z!_e9)tT2?eB_xge1^+SQH``{TDZ-#EKdf~U^KaWB>I|ryXixzZ6OsJ|n6{zLb#O1Y
zB*hRT2^mqK3^&#Dl&C<HEn8IjEW{8jnYQ6<-O*n_&|n#-tPYN+8|UvI{YQ-$eC%xO
z+@^%+;p3E-we!G!9y|MY=;MyVEn)%C@to*~RlbO#2#7b^A>^A;1E|u0cuw-iVTeir
z(Abg;Y=Y*P4JdbmvkLL1eg;oDs}vct;w4`IlLbxX)Z@nz4d{Y~FhS}9q@wWoFl0%g
zxv0%AJf=GfNt;tRM&t#=B(UG_kNusp<3vNSeTDWT3W3$200qi&@6}<`yYiGLK|=gB
zDJ6;j98Tp0nB&_{;{cAK6Mv5(7uB@~T~v-x-z>D0{@X!{n}VhmE=F3oQU7O?RPS0y
zRRV@a%2VT>Qnro5A*4|(fd3zF%O`?QC}adsL9?(97b8)ihIj)1>YobT%ZsRhzjwI)
z4Ez@CNae#cSE_%8!0LQFeb`^-g-g>fl*~*3Mj@$G!_FRv1=P4l+Iw97w>LG^=7&%L
zpzsN;B1zkD#E}YI5`?ti&3VtGuXX?G>l=XUq$}`(J6gbhr~REw;#<JLwi>AHR#@WQ
zfzb#2IhINN0=Afz;tq@jM_F(GTX+z5_U8xQbWA(QUqAn$f>iVhjrrN|+fjK>HkqJ(
zFy@3Qq7MTlc0Wc(nd&r<DY%hE5nBb0s0OUa>#`1T2}tWEWfyW~(@%p6r9P3g0VTvM
za>P%Kl!;?y2~nY&KV3Jh@Z<i=pc@C-^;(L!F9o~rZf)r<#}ffFP6pN1c(*m)Evd?Q
zkc#efp|SN=AXB$WHa%Fh0T&2U9>Iz{Y5-ZIgpL85e$cVSW}&00mkvru*Ku^PbB7vE
z*})p%AYQ<S8i8Q_^*b9F6%6771`!9NoCz2t(*OD)5F-rQP#Cmh;3GKoFh~gh^~3+;
zh-nqR*X#WS2nPJQrwC0*Hvj+l2mD1A@HemWKZFdi=ZcoTcQs31jI*$GKe(&XXY5IJ
zpU@@PRwu1T<oP7)o22nggMYX7H$iU}_cBF=;ClMz#*~ZYc9{l0!}=(X@G!m2X|)4y
z9r@<!8P-<8_6S+{mdR~rN1o2zN;2MCxAz%}8;v=*#Ad%~sXnv-{y5A8t?T=Xe&_7T
zv%`q65>#BYueYn$h^La(T;j@)Rhx7so6|FLx0z^KPs8}LE7lI1vG?H;*xh(WlS-;x
zUZ$0E!%gD@*3{CY;;kMMyx9*~t;_x{JH}Bo#c#1-#k)(LbYUwe&!iPDP}vyUzTw_x
zQ@HIwli$fU496vJmW69+-K+n^?!96x=II)#CY%xUVwxe~^U^&j1MRM4`bpYSdI9v+
zuLj)kgiqpx&>mGkFX=UxYV{RFp75r5-x-N~1G1yn+53C2OKcY0Y-7HLi0TssJ8dU#
zM5&BS_RMlU>Vf~bx9Re9=5m*f&x@exx~LFpiTRp|V$Z3Zd8+`>StS1QHnDeA-4i+g
z;HJs9-gk?60}Q6{#Nrs~ly;Ng`GxPh5;DHEmW=D8Q39#NW0C6nf!v^B4L!&&Bu)~o
zKIzzvs{HdrCDF6YbIZ5B4KqLU{cOB=d2*W3t>0z33mNA_DZ}T!Xb$%YC(BH_dA8B-
zo#GV8?hR+SF`J?}mdSHhqd8H0)=`B)*5iSJ$*t6D2v*tS%rJj)6X*|)E9fKkr8`xB
zy{COfwYo1i@v_9(=a)1*<-DZ*fF!sa(o2EFGp$e0)hja59BQajk*uG%f*z%%;><6}
zna#hxKJRuWgS_v!FxT6@CTe-#^!-QA`=8ey_eo^x4lm9>34m3~j~x2A8)V5~LNDO$
zx*Ku2K8Z`DHJoLHJy&Mu`WYUiYH#5+rSVvXZ!x*nR}*XBK^zeYLCVtS1iaa??32&M
zJ_z_fOB0YCsSUinVC^MU5whr~Eaq9<sfXwgPT1sx*imG7p>6OL<`{-kMVo596X3A<
zmHw-b&FJ#C2HDRzuRH^9t(Vmw$3e38#8<d2ChBn-x4LBN>PY@i9DArl7f;#tIqufC
z!EsBrL2L^;h2`yS5B;nRmYFuMDZq~Fmj<lKx3B9sRU7=M>S})$0;U(gnTgpHo1d9y
z?~@hjXps{Pt=oB$SFbLDTjhla;@h9H2-0&m@?#l0o=MN{T6ICA6KS1SXa6wB7P&Ap
zpM5;TjPEkWq9ccK?}lUGPg+P;|5sga9u8&u#t#o7^;D8XL<>`v82dh!Fd1uhLzcqW
zLN%5wgGY;mFpMqRlbu1fWEo3kA7mM_OSZ9Q-+8a;{l3S0{Qmg;<v4O*?(;gg>pVZ7
z^S)w#ebLdam695G*Yf<G4~8;M6}-g(C?|6;HINyfVQUvTVy-oAVeq3>SI`W1AVhsn
z#sGq)ygM}kYhh8%sq3Yz-*kkl`}24toNUZfmc0d*J?Kzw$KsrQ2&j(F@y6r2tF)&Y
z90Tp77+X)K5>|YiJY<sv^n6q}&%KW^nrWQPVjnt~pJ@$mZ@-SV*V{BJd-5YkYqp3a
zm=i)nYjgKpZt2xMu!LYkiG6J{YX$q<$-x!fClq*&crEo;3b#fjg6YNT_fL_X{G_8G
zsx3tilS>A{pHm8T3yH~!&z?bgy=EIh2Uk02@6?)aKA-P3*w&8iC);)oKS*(YPO$`f
z*r-<;m~R8#S<}1qH2<ngkx-e6Uv%vk?PUFec4dyhKdK&&w_RQBm(-X7`0JOp)T>7R
z>W`@+0=C<mV6xMne(=kPUMuHNt5<ydkON)xTg%Kbe<tvOL3Ym9v`urBwQ1A8Z_*<Q
zEEtX6-}%<VmQyj<>YQ=<L)`2J{qa4f)ima@><Ee8Gu$o%T{bsFQmNka%z+@>)t2OY
zAqSG|j9!}-YW}%c`%_8~v-#Go57FD<C!I}(OZuCP`|-CXoGBKndGyyxESp#DxH)=U
z&AoN$WXOb%ZF*Y1Eiu;>2NXp<T3g*F9<~U$m}*Bq@r#sP%ox(Tt%#snNiCmJwoU$(
zvcB+CprORth|r82UT(~;Sc}inOAiUO*4;VO#T`$VppADA>(+23RA<#L7k0{{_;bFt
z?G)&;bh2|_dptrQzhbs@ch!TPyiezjiv3k$GA4&v`+iG~?FK}yiC=bX-RH*D_Z38^
z88`x6clv|Lw^gmGc<$rEWVb!raEVddPV?B=f-f;t1-ySuSmDIuCn7Z)LT9@=3obBp
z${=ZqD>^lK?xXeK@2!r9-Uj$2Hn+Q|@bkGeck_`7GJ)fy_hnjSm^ORV=g?#Yv9R7v
z*^QWP%xVZYKI;Xp!eh$p>%h5tH>W`>wv1La0~0M3DB`|U>5NC-*)%CZUHIse`4Uji
zUP|sazSLQaO)5PYwp*Qu9?S8HDKQ1EUGP$6BH|$tLi2+%VWh{7ms~a%pt+MI4s(wT
z7v7kzaPasRI#gMqvvA>a-DzkRxa(_nJvfU+tI8M1_K#w*gsWX;UWmaN8GV&UD;78`
zcuNN%ZuPz0@?f%CUsr5}DiOc1j42Dd)A=hOsKG7zmW!2K(E(Q{RWf#sntQ!ykTqJv
zxe_Ip6YLhdsa9im$?+qgEd2TN2A@^{_NUS*?h2!VJOZ}IPoIqLDqWPa$Ouh+l3cQ5
zHM&}y*Vdor#80WzbpBQHzSe8@>!S2?>Z-vK#C;>X!Y}^<jh~Iv_B9Y&_50*B#Odu$
z6hrMlvl}Q+6|3&I80tQB@#^`WQ>6)Z6sFYTVJJ3^3l{9s<&0XrgtX^AT=&F)Q(35X
zB4nWT=Y43_i(Gxb@NRGB4k7@E1_xU?4mQtNy15|vCnbtmb#|?Km${#qyY2<|5yt~%
z$FfuIP;A_>LOnWVSM<yp$>6%Kv6FvqOKu+=!6GTC@d(LD3Hc-w?RXL`la;pvHGNEx
z1LW%*0<b8Z4TBE^viHab<I!$ou#jkHV}IkfS34vpQC_-ucvqoHJGlhGbm1y4eyS-z
zE3b49;k@+UZDR`-g}{~3TIV9oAjo`$(*3}85P7#=2lbHzl6Uc<zTbf>Sh?FaL#<r;
za|wq5Vzc)mv5Mo8IgaiAMF_B_jnOfDb@h&C+p03yRb!BQX=YmOfUwK5>mBsO_^9A=
zn_+4<2b{M})lfU}dWV65`)`3$j6vM!MdQ8Ag}-U!%2OQp3a#`vI0ofZ_7qrzuh=<+
z$4E42U4Mro?xwsRT%DyZNkxbInhq3w3DInDLbK@vs3MAAthmH`jK5u?(2b41??D-I
znh#kJ$<ni-L+Vwl&$ehzHQ0#;fa<D@pDsDOeHRvA-sv(?b6Cn$g|$yj?;eHb{TH%t
zuX>}&s@oV`A{P^mePOqgH<nA6JRR}%45yuT(Oi+#P#u0O;OTN0i>zhNg*dWo>H<rK
zptW^uO+*2>qwe;Hq_at;m}=c(9sWt)vUqKO?dHP{zlo|~!}}gp1<@lH{KBo)In!2E
zkj+{Xl4s+lyC$6hDB6}^utv|Faqj6ATi1q%xAI19mz^g!OsiI|R=-=iE*$^lI+N#j
z-95VBRb3ETo6?gKxTlfHaZ^8yuE8gdDFfT-rkQ?WZZdXX#ZC5SGIyo4q(N%S{tfT#
zDb9c?0q_DO(7K3#VEuJ`P)ughyF6UnixrIEMSl@JpQ7DEF%?Xt-J(zKu{HMtqcuKL
z(6oN<?k(&Qw5^`#3o>ZnM_-;bW|*l_Huzy{{pG^LDE!K9<)8arAAEBOvYo0~#pgIM
zDyi$i4=q6hbW-e^Rk#c4kE|d=5`WU}Yr@fo!D0I>O2+;)8MCl<9(dg?7j9gL)!`=-
z!@0YOMy<wRGHtCs5RaI)o3U2CImwGB<5_qMw~R8vlX7ptyYoU!i+PvLk#M-tT9$l=
zU^Ye)gwSEn1k^S=az%0Zq5w_K3yjaWuB|72r_P<S-ZPf7<u&{)<)@RcpdcnAM3ZZC
zkMtU2I*b2*SC+)uG$<-%_w_f&(&~y9$n3DTk=D)14*yNZ{&0%4Jk=Y#gUp}#FZZaG
z^c9)*loh;fDNMg0|MEugNPJOn<<ImgZLtxeXC5Ug8r8?O@ljoOdo#n{6BnXc+m&fL
z-aw|i2q~!#d=l|Km$d)rEI!Ea^4*j#UFLona~$ZUF<Uw9f)DfG*r}xq-!2;rFXc$e
zD!Vf6I=2AOcgAI=zRrx&^$>kV8TE;&cOmwhc`Zp~g4o#Y!wwT;5$wmzqSU=hf(O0H
z&=@p^3>wZ?`nG~_?7jFAbAz)oR8uH&UW4WJ+UnGxJ+_zOm$o|Yd!!9CYqG$Ny1CBo
zC-#8G_tq1H&SAhw2iwkpi$GHl3+ZlBsabCdw~LA9gX4osJ38Hpe0kZX@0=ovgv{ig
zahd)Dz7w)zh+7)+mA4NvG%_LA8XH7cn%hTS8^SKMt$~l7$2KDIWJV^|=WdNSq~3Ep
zkW8(`6ZAv(4H;w_KXeom{-FFqc^C8E(09VW`Ho$!jQ%28xRQ6XAVk5aQEahK9ar)Y
z3Y@6t5%0!#q~*H&gR?S?q^Y*+9BnQ<@In_O<GJR=cw@3~+7m*pht5e=x`?-|>1iOC
z(U7mPB2kW74dl`%->CTdP_WIV+t*S_&c!+2>;#v*vX??2d2;8Ee9cZgzw#>|^#5;Z
zZ2N0IdhLxF(dvr)4nfKjNNEZKgz|Jz(X`0!Q=avi=OX*>+7P+|pNxMIC^`$i95Xxm
z6;)+twB28pdn!3S-Ldx{v83N6=<}S3l0s-0`f$Cp?OAGGoa9cIKvzl&8DIm<i`^E9
z_Yq!c!x5SdXM&p_3OmK;6p5x0-8E`zb0dF5%*V&)rB`*RW|=3>!cN3ZX~1s$`;C6o
zCiQvxx@!Kb_O~$2{7~C}+m`&F8>0DDZs5}uWOq#efT|~lG_}F_yXMdX58E{b*9gg^
z=S~G6)BN)Fn{_#Qwg$F)hR0wT+opcm3cX4Wptblm#02CzVVdlFUYc#C290tAZOpLE
z(^_5_k@3N|IjzuPEz{dJ$G6%93-+Jr7M2cTX2JDxyaJ~<HX7b8jT{)9;WcmU<^QLU
zL<Gg!y#84B`ACmDpswlugt-4=7SpoOezoI{`xF7(JEoC*E{&u`^>{7*C`U?`lMl}F
zSC$mY<vRN~<=@Ru->e5I6%k~v)Vz?*S;Exloa_jgg}-!yt+lm(gEcRGV}2tVO{n@z
zVpvvR9}s(%0?OBiEZZSjuonfHSGe<<+&Bq!;o@93AYWyzJMM>*QoST!D}Mwf&pHgL
z=eH;yM_l{9G;%M%&zl+w1dqIs1reXFUe3VjyFk3Baq<Z7G%rSb<*!ON2t*STPI!MN
zDd@0I_0R-OyAJ2;uHAi9uW8Xsx!L6gkhVGG8HS||TR9z9ZT=_?&?>o$He1b=pi{&?
z`W*+<w~Cfp99JRaMtNL|r4;3<jfPAOC=d$j+kHPfgqs^@Qz!~E%)QYM%HG{p5^6Kt
z_`Vms-_dk*T;0t~{Ovb8X!=8YqpRgzid4bVcG@myh#M2zj!(5WK^CpzwSQUHkXJ#*
zEFz|SL{(?u^P`<O!R;+oesYpP#%*q=jnZX%d?iMT?kN@`L1C?I{aFc5p~<S%N9}!+
zF^<*%aD*wxkc9%Pb3|JQMec}cVu1^kmU=kT+m4kgX~Ky$V#Ud2SA!PmY?_4W`vg4W
zGApcBc>b4Jo>$+B{w!Ts+tNzfc@X2wY5m9BeRUSJ;t|}q=^YJ=6S*9b9lEZ^kW|JX
zoP_P(w2sQ2Zk!Qi$V$w4YUjXVsasr2nLtYUVCw2GXKhdslH@r%Pslsj2yOxWESna`
zWVBR~G=;S)xe=lO017AFp6{@5vR&j3IsLrf=dot~FtsF<3`HIpxYj#ZkS_YSfLUJ+
z6a^u!YU^1LGW=qQK<_gm>qqlK7^2xrm_qKbCz=vRQHbsR=%Kv~S=VTdM$$V@LJt$S
zvq5Av^y=^9KE=J2*#1_Y>m0=-;r5g^)ISYnG!Mi~C^Kv?T#tHB_GIFHrZR>1xMVz|
z`7pCH%hAy-am;0v?<o=$Yj0QfPPhNWDD>v0{JYR#oLmA?;D%(O{#pM#k`C85Nc65D
z4HT?X$kZIyNC{Kp=ObY|G`szBkz0a&B(A!2ACh2NSo6aEVBFZJxH8tDIl6RS;JKv_
znjik2(G^D!1zk)>n<!zg0$9&vTJbxTd2?i9pd&3jg?o-BYnk=W;%GIJ0PX=wjs1em
z9Pq7d{QLG)-zzyHm+b#n?;3r5Cg2}1@uJO4lS)3P^Ww9z9||~y#>;oND=3s6^t&ZD
zYs3aaV^U)`x$#e9EgXbX&D5?!c<mwn4!^c)umkVo%KN?@nQEzQ^nSAP9+);lJZ9&N
zIx+Csud*;j^-GjBa>ddp_ku6vl28ghiJ4*W@sKg`{CuB*(`Po!dtR;DpTij4w))|)
z{H0|!q5zDk<K%#OmOhTMLE#HCQXG_WU>&ySay<H868!b5V*li-O_?R0Mqkl8PSyk`
z)$vDweUNdpKDCK~SEI3h*1Ar-U%NZLCqBc`8u8pT$vJ3u&pZbbdj0zBx}f0U*<@a)
zu#gXlzhYLKM7grbGv$a||0`i%Y4Qe1#)B?tMExd6&L<rO>e7UDlwT1qOfTX`o>-?M
z6!k4HLXDBZKL&cF2A#Tkyt%YAg~adyeqi!?0!m>|5{0SE;dF`>7m!bzNsGEjuXQI*
z(@O#W_JdGed8`XR0OuzammEz}1u3@fOpFQgRA7fhiBL<hl-N6u+7QFPZP|hQb&_5W
zx@TSjQXhzTyIFbSyOm6OeVL@i$?O|n-wJz6PF=o#%V-a4sxgd!K4w2AC?*)k9-PAi
zq<K`&{j%;)o0pB^(=*anBFy&=$*^z$r|DvNr#NCntBh`;xA~d^MM0-=1M(&pAP0eM
zNfxzZntpcb+f0Ydh<hX}u{BZKEgJmWM!f5C1hvIb0rMS@Z%>t<?YWlvc67K(lm>RS
zr3Ogk$X$B6jpFeO&3x|6Os*FGIYd~eSpb+hd$z(1ecvI)LBo7zMo~&?mqcx6+9>Z|
ze4x5PMG^w~F<UOTfgUqAYitPEng&J^U#(X`I9%Nz%lq`XesCFS_m?jd)`qe(%<h!q
zcm7DNT(!%>_8IbXmvs{LsPbgkY~$Lj&0UDcgwb>y+`>JplKBLN^1-X@2}fPd<cRi<
z?Aytc@2h$PkoSzdbeSEitHh}_v@Say?s=IQ@#2G_#M(c|Nd3@HU{XoWwzI-M7LYe0
z)f2^q=Xbke%KQ5wVtpQ~4O)!!O}wpa&wm$+8!!m{bcprRU!lU+G$ITDCA?eFIBABo
zI*zlPa5u2J2F@)H-)n1|v+K1f&F}2bs#g9dr_RC$3~!^YIvM+Vz+CSnx)gCbJ8Px-
z!#JWAAOl!+v9eq>jlo5HF|6I*m=$OvA*5ldEl|8yi8!W-LFraHIrNUam9D3a*F?+X
zf5M8{?zW|<Tm?$_&08UHbjf8aC5I=yA~JkM6b}BV0?}N;o>;Ms5Gc`;KQgXc`z!4G
zzb7S3-lSKTNxIBL<f6ZmT79iH5VoHRY+l7`LbnBW$TKX14;X+5GJh>a4OkL)U1@oJ
z&&b;xU{tJ>jFnYt5M0dq8nqv=2QH@_1-dAh`OtlZF%1{p>v;GUtuN=ZqhMnR`;tug
zyzeKuvE5ctISRMC7{J4s`T6F98rz(D4$6tA@7qm`$B!c(#^@S%T0mL#L#-{tObi{^
zbkP(AAcNXpkWwXWl^1EDFiJX?<_~*M5h)agzkgw6q38mOw#$3Nn=zHYkkm(w!E|sp
zS#tSsd-Cr!C8HF9z9mDP<fhD&J-~*T3P~(R_f$APvDH-_wx89c0UhZLdi5BqO0n?(
zbQ|BDJAV84O{pn<;=Mpzxp1P3O*pr(nOvfD25_DfM~A+3ew^aE7el%CtVfDm0x%qD
z_fhW!!HpbV;l}Fz=NLK%TB-n@)FbSuu>?2Z_a!l}*&+3M&rA^l2$4mt2TKJsnm?b8
zsPbrLit&1S1*9*zJtp!LKU9icLDit)Tj!JI4byrN07_7veHUWf>tw}wK~ZJa$z^uE
zPkgTfMCJU=ryydlQo6_8vf!m+eoef-oNHn^PkWs+?96X>xm6rb8gD89Ny0Yy^S_K~
zy8G59R!-aLDJ6{~4vPxfPxE3^FGomL7ul@EUh?H)PC+0@qd`&i*h>(esGM4xYS@uB
zQjmBEZTvYY4*J;2^cLpg#de05zoEuHwRso^SliKQ_<KH$oKuYYtTZC`#LpIwIs4Zc
zPR~f{L?pGc`Mb47)m{cp_2837`#t2&i!Y(}136*5-%kD1ePns0?ajcI5LF-HcKksD
z68;ADnx$T>l!b^*UHntc+2;mA80M5k35+Wr{V*8QDac2%)P>3OouG_aAk0(oZ#aaH
zFW-V7gic-2snZ*%U0RMA)4uk<a84+h9Tv0ojIOQR@A`T>@%(C52-o)94^YjnOUn*2
zC<|&q!dIj<Fm>VYi2l_UiFXhl#0oeu;){O?sK*5<PE3;g|6e`tJNOLrT5C)npSeOL
z3?ykp-ccjn_(wRtyyaUb0nE_CF05`sn7DX^{RgZB3nI9ky5OLMn(HaJa7eql692w;
zBerE?<4c<OKxbo3Fw5<BG@G6|Z(HAnFDiqsug_~j-JORUoMPjT!--vVIw(|J(z=6h
ze`0veeaM!ji8<C+bS3bgv1<_TOAuRT+OX%><9n18uu~S6seP`)iNIAKne<HOK#9Ls
zUOigHQcq)wPDns@l|Ow_y1zD=^&}?yb67pFGQ`%dCH0*aZ&^d%L_#6f&LfgC1eR|)
z`PJ8)m?UWK|7c8eXkPK~VF8e;M7Ez*QZz@^E}he98W98)Yi4Sd{{K@3vKig``|Z)`
zIG8f?#hQ(dF9&7AY`l<Wed8O%*VD7CW_K%tY$+8V&rDb(KI3*rGodh&#Pbd0Jw0Kj
zATW*co6pfz%R5c@4TRl=8h%krs^BRuUyh&;SF4V$d~(%=HLv+2qiYsjJH2Rau1R!0
zc@crLf93~yI+wP&EUfn_W&WGYY)*P{R*fE8&A$-#(~+H=rmUNZhEUi4I3SbCbvB_j
zETI(jJI$iXsF6l;P8i9T)+YIXnSp0j$b}xV(jB<&`qYOdDRRZrfyGzM%Ey8@=6@K#
z33g(r(vvx8zvLl61L-ng-Nb((o+tmiz`bebKb02!5I)86DZq`U#bOwzE3M%hsR;?f
z>;?}k^8~G*7OMRB_ata1Ml_ypwF}m9*h8sV^uH3CMW$4KCpZ}bao{tPD>1?qm^zm<
zZ)V@IEgqi3_@#K0%%fT?8qVycqZG2fo23YchPu);$ePsb&lkSweO=-}Tc;@O{*?)*
z22w3xJuuKQZ78Gx707N$aMj)M&wd%|*(J5j7%y!@4<?5I<dx){IRySYo#a0uJ=JhX
zJDbG)zUUHh!eEMIgv1YWgh(7@{*_g(W{(ff#NZ!@#-BT-&yE<=dAZc}nsp=E8<=su
zmH{76G$0;9gKRTy;))aeIDtF{2=DufI@TihmrFlxs;(Q%cbc844tPaopzUA`x*%Hc
zbjlQTMR-WisbeQ5hWZ_4@rA*Wghtm%GtpW0oxmQXwHm}Dib`g7ETDjPL!(OweLJ}J
zhu7^2;bG(9b<ATMOtJk<Go5Dg4nv+F&IDbEnJs3AWjqECuSu)g(jl^2mzPC)=k>Ac
z3q2Y7MJ_*5=eDKHoPO1xNYng4URQkWQ*&9u7P<@^Au32Is#$nbSzOF0+UpYWDqMAV
z#-2UV5dLQdq<$j+9t)KsOdp(qXf#}JO~2yzAD2oZhSE8oXUJn4`>^hC1UB)MF-X_s
z6YtI$)CMoumdpk{+lyPRnjzEF6W~IG34B=wR?Y);tOsd)^!@s}78I)mh1>`T{x`OG
z8gDO$MW2OpR9OdaUSyKU0{>72E>h%jTWPOd4?4`z7)DA$ZYWaI3f^e`^=EdcU6^~(
zt&i7S-PW4-Y!r`lYUa)7d<z}F?ldQa<(`6kCp%a<y;b5v(C|d{c`n$vc)Vg8eysBc
z19CmdT#N!m9k5Xq^;mI7Q;Ps9P{BY(EXQo(q<E!Iwy4U4!ztx>C-bsjX|<aXmm$}V
z?oX^9^lNpNAR<8~Q<j5sCPfeRb;U`i9>rv)=(r4=BthoIsIdhNoHRmy0OtQbKxkqy
zG0V8F(4SdEajL;khK{7E+$mwnl6#-~PowQdvoSNo(vb=+;+AA_!A5kD$A@!bjuTr3
zegxx{dG5Y4ux2SRa7GoN%8A<=tA{(_JJxD`kk13YvFqeqf?WLFuqhuXVks=d&ZsC)
zg>2@>h=Ak>#6@3S%tP#90d&7umN~_MT-84VdCO&um^r=P?jLlr8uV_MHL|BB@X6m+
zK04>hW0BsSztt<mqY6|mg-S27Gd-yy*$xEPF5?up%xl6uW^<SFO#8movmBRZOPyn%
zc!I1T_6rLV_iL=5&p+ScHv=K*uuQ?_QxzZ}G-O))^R6Jqfr{x-CQqcPEGyop&CK(U
z`AD!L@g8I|DzPa>soX@GDqV!K=O0$&z)Kx1(X(B*PaQu2EL(Xsr&#Jv`>XB~`<djg
z$cc*(cy{`WUJ0JgY_NMQm8L4pgaLw;1JNj(QI?$=MZzl7_R&%Bjp1zeOKnz9(j<M=
z#XtzlHk)M4E3_SI2tjUKNh|do`qrt%#362Lx8466SK=PBRb7TQUq~H<%C{G|mLp)t
zJ%K>iql;nBE9tS@myxVM7at3HR|8GP1IFc4#fl?HU1@PSwV7?F03<)~8DUI5jf3}o
z>wG8yx_1}a^bjhKILwZLgJ_ud&3O8i0qO2nn~3QB?q1gS&->Lx>xo6I+{cmfXqv3M
zQ6!b|72|{5@ndnCA3SKG#7#C7oEjh1gbF=gq+z69-e>E`C8@0kdJOe|r2-78;$)du
z8~~eZVS@gRL<d^%#>_pzW-Wn_Kr?X4ar+>loL|=5MXo)G={|byL_DWuS%=u04=|>P
ztz{%EPSljQDgFwD3`|f7Q2OuTUu!KD>0301RY9ydrxoIHK4PIsPk2*YX^*Fp35Ojm
z9^G0P7uDz0<Kr%15Tl+@0n>C&xy$7^<7Z>7qzc(gptA8DNz?#T(-V1c6qK1i?Oajl
z5PocMRofb#ZxN9CA(iRc7<Tw(M(KvNnBs&#QWcj!60Vp+?Jj6-E+ql}OqK@@TBsXx
zRu*30%mwxI&Xh^(6p}mn$@`*r{DwEF_+#>mC|mO}Lt8*0^?|(wvYZ=^ER;atl!oLw
zPI;~#A3i)AUm@)ZEBHcii`b;&<x&P)dahVdQ`qB#d>CAf@Cpu25)qRR3>8gc&ixD{
zp`@y0^iN9wN;4a<dC_8C+F1gKHrvJBH_Rsi+CnQLVUUFH)bqWYlBzUtYAYhM+UOMI
zmUw}Tx0SBfaR@>#**W(^MfK2}v~(n?3l_UsVyCz>G|2{VEm#;C#sy{{9{q3(3TI$1
zQ1ObC)o#x2<5DV<N>Bos`cwn2>4StCf3-@dxm`j^UfT5qdWc$s`*Np>?W{rQ#QgEb
z$<UgLtppIbEOT$bF)P@d5@W(Kgf~HBjq|<i+_Ye6pq?u|8E<1JPEYh<)VTt)ir>Jp
zs*w%*BOC=@h|h)5PA>L-r?P~V?r^ya_g;H+XwgGJe+yh%)nkFc-MVndvis<n6Lo~C
z!FkAA5)~<aPfFdcYcY*_Ys7>67JlS2->6wyAFh&Vq=l~#1hQD;yfn@pm+FjZ7XHaH
z(psmtf2p_^z`Az#nY+@in-4&Gj{@{V<CSsBeM#qwVtR7GA^Wjm;r6?Wte#Nv$;?SB
zPc-UthSMJAW|O+}EcS2uw4Rjsn?BXhUTc}%yhL5-<Yk1{)M%B@%|et8L_3a&&?J3m
zr~StWNs~_li7rvJ6v_+MY@k_NJwmC+IJ~X~b#^F7@iSOtAJq@D0kB;q=^LeaqcI<Q
z_90i-8=!6sm^RQsY+HgOfcfn1@m(i406N)ixf4(l_M$Ww4`Op_(~&xG!(KV?j$eYH
z4(V^z&Ag-$Wz3473B2HKOLf(%Kl>4iE_(4sI>5R7sVJl?wUBs*w(tBR9Y43s3%ON(
z(M(}a6@}peIiU5_%Q`KnYsRIw)`w-0)cFrD-WtGkl?*kp<Y=L<fJgoS#Og~|tCb$~
z;++D40>L|K)hx6&2I8OPno?nj*YoHhuiQZ$ThEzjns@3kVVGh0YV_zn6mk!@x&Lw0
zRV59sPI91!L?dDKL9qU}P?zQJ!C5gGrKM~%7n2MV4l3-9)j(wVQwfrjnSX#4l+V>#
zJxKO*FZV(~Uhz74^<;(+HUTse#904;GxfAuzfgg04+>WT=m*Q{;;BnWbzsC>yZ4B8
z>vSgD%mV+;f3g6$)(fn@o6&^C9mtdtV7Qjo`DILcxv8pii5vPOKs|iaas}qaZ9xfd
zb!C~01+vB|9+rJKPLd5Ie#QyvQD~gwNC<zI$MDq-iuslNb(HCG35NE@aSMQqx{<yB
zy;9z_0_<R?HAOy2`X%!f+dCxTq)j7&mmcK%>af1av)9p7q~nHvv2fa#=3VUlu&qI6
zgI#e$G%V5I^ll(M&Bb~bYB^z@E&fyGuf+gX^ULrLJJsK|+lKQ1^q0i3XsrtQeq8{x
zQf}DoC&)Nw@Lu#49n{js!G)OeIA<GJcDHb<0+hS{E#xisRYM!S$!q}Cjovggv4n+_
zZ{p$%&mXe6*cYW*#>(?s)klHHoa0b|!QqNoATcL`l2p+3hxeDvCL_A~T#@Ev^9K_A
zMK6T_G1f=b!=4~E=iQts<i=wFhZU8nOf-rmsY>F;(SqVn25?AN(*YX`q{Q0FQt?@Y
zXrm#NK9>Qac8yl6FV5qdkdqu)!~CD4vzsu!q)q)>0kXiCT?cUds}@cIR2S_0@v{M-
zg-FEqJ*2%6Zz%i0LzHuSWAT--PXhWOfG*M4zhXR%^bG;(j;(vhrgL1=y}EeEiJAPT
zR!k7HwFHgnenIV?SbS~(u8bE<-&i}hqaF*;LEa`MhTXZ6h0kV+BGVatD?5ul%YAt@
z9uCN%@zZ2560lF;;>mjl^yxt|!?+GDO^p^1D()bMmDLq8+BGXpOvp>pWsdJ`$B%DA
zZq-r)2gBn>2c(`C;7dKsM;5n!Mnr~&?pk)5V;u+#te0oT0d6EHnC$Tch~jFZzmkGO
z8-O}ueQps{Ec9B>TTp55W6HzdeTf(nrVOWa3hZ}--gWLf4~R@0%V^T(yN$vgrhbr>
ziql1Q3IUttYnE6Lc!u@q=k&~+LZv9l8or(JD$GWHif=lIfI}g&J1O+>S+LnqD*h_K
zm*f#>Zu}_>k5w~(8b8Y%eA&F2WEtz{)d)}uT_rQx8@Hs&p7wg<iZX!d<;COl#*H3G
z?DeL1dmPWW{yLjK#khJk&7EozmRZ2x&d0q3dlx{n%_!cbI03IJkpy7H!cbF{Y1XNJ
zo!;NL_B_o|`Ur`G_(|_g!q-siszMt6&6^*y?n=(!IE=|hyq*~xkc$?@cRP+$des5B
zlJYRoxhtMLBj&7b|2V&&1MXh1Or*{fdSO)AY1H~_%l_<-NzKda8}O4MLAwbh{_EKQ
zpz7;8WaTRCNfev4e!QRH_~A}shwG2QQXAxoOl62tD6_H}3Je8m-M62KEJJkbgCoQC
z;$2cf>?R@8L2pTdXnaHR#T&`$pJtxva*u}45oitHa<I74%H<l(MSW>N6z_oo`i<uH
zEs!PFb{5t{U3hikC9vQK96KttC6JwP)elmX&{;YNU#$U>qj}k}1vnq(hFFV%LxpnL
z&zKGrME7*JZ)bt-MW((s(VUMTmSb^DbK4Qo>JZZA43Sl)K~8A+>NR)qgOM}59d2sd
zryZu%`NaSwiMc6D9>x32WcZGX4lvNMTs!t$#rSqW4;#}6l0`?gI7zK}HQr%ZrC2@Q
zqk}pIY0xY7X5T)tbl}>pA#UVFMmk)a*^r+9K&>@CpNX27mQPMLrZZ*P&c)L~Ab^i9
zHUUk3{Z``3h>FvwL@S)@+!VnET4pdHjO#INK2^`iU{EJuCKt`d3TF2qlaZiLvKn!x
zg$FBI!)#qYOD?3RSqo5Z<h@b7fQih!Znv3HfqvaRe2b?hx3;?UPv?w}JVl*4Y5f+*
zQs(Y%l@<G0>~mhUYf*Yq`_Js=gxjV0E`$a14D-@_p^T*ZQ_bN=j2PB^AGs3C`oj|g
z9@nx`PJ1k8#L}De3AaCO^*0^!O{*RaNrK^G4LLCUazw@Zvo3|M1y*7B-ZQRWGy=bY
zdZ?Ru*gf*Fm$Q9t4}L(zL_}qTM5Kg7B#cBw<;11rM5P2pMC3$7)>@&O|HlGXcRMGv
X&;NZv5I$uaEPx=@b=8XQ*#!ST?R0w1

diff --git a/docs_old/logos/safran.png b/docs_old/logos/safran.png
deleted file mode 100644
index ece23c8e5ec4bff96ac5aa7ac17545896d1c30f3..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 24435
zcmYJa16<v4`#%23_Oe$C%eJ+=YHiuJjgxJ=Wn<Z{wOq@#ZU4`n=lTBruTGshz0R9A
zuKT*~6ZTa>3I&k>5dZ)b=`Z5S002b>{=Ebq7W@;>Gh7(_1Z^rJF9HCyF-R}^FyOz*
z48JJL1Aqq=0DuAk;0gQ^=l}qmnE~KP4*+;m0RY!7vsH;7{0~?ISt)Vg{ohw^M{xoG
z=!8p)i+poiI?Zx*`!+i_xZW@^S^)Voz(DF-&CdnGI~hzetg?9$R)~}NDy}t45s6D>
z%uwwQlw#74AKKKuCQ^O&P^$QV&_PL=mR2`^F9j(VFP$+gZ4!U$nSH>#d*U&^c&_eQ
zDPh)U<WtuIhwRNQ3>POI!u)_v#fz8j87CTPQV-V$#CGdJ3BxgaAumDM@mQX~G=B(W
zWwwU8gq?ec=ppQ`_X~Y<aQjP|hU-EXOa|}&pGPhyIlG{KM)<N5GF?6s+r43t3Is{F
zFq++f>qB>hcQs(rHQ&&#dpYr{Yf$wj2-UrgOi#VHJ6f-?I7*tsyFW=P($5M5BZ3YM
zeqfMcM6;zt8KSLvuwv-;j;5e*YQair(lD2Gp5F(r|M)CE{x)BaAE?;#Bvzdwl<M)g
z*)`(y;on<D(8WRG@M!e3M^zt!WB2IrngQXEL4-&^^dqI8(&_8dz=Q_Xn;Ofm>yrAF
zzKMlUjGW?sCl!JIf*1F=+p0jaP(|*K6*+=m6H?o~@E+rFEW9urDQeK%^dY5lABW}R
zf43Y8nBz|;*LGMg4w9!IPyx99`qv4F3{1@a7(%Ll%9NU{k(;qc57}Ue>tu>5{&!B0
z1^5s9)ES<5oafF2c(UE(&P-=5=?iB1eo7KAf0f|g@9$GI3vEVveoyh8R(E;yVHmS|
z|G!g_A6<=@%PB-&87ByUSg=%sFN24!qQWU#z_Y-my$#1j)hjI7eQiBLw^&7eF~bUV
zWIvfPK|SnHFRJ~&`6<QEUW{JT#9zojmPb1Ma5XxCQQc<3#77O$f7P2X(L0PxUw6t&
zUpa@JaQ^Q+*}uDawZBnP`Jj)|>V5qJ8Bk7ts;Q#~P2qD?2?f|8uy=?T<fIB*1%6DM
z`>!?l!^e6wZzni+$g2oRb?5R1iK7TbNFqiOEH}CF2@~Qkh>vFAE#!$1A}pXB2#oHi
z^cNOM;(tlgvj#wOW@u&s!nMfZ`gUAqt5H8aFzsDoTm9^5x|QS^eJpT!Jk_peE=}6P
zwkZDJO&|zsYz{7zm`%Luy<lyW6yf4M<6ioTUL(riu;B^O6}=}}o~O@;5H*dJH5!}z
zdTKaN@59?;A#g%wCxzNjgH?c;>^!WX<9OQqVc@;|a{S!IprZ93QxrjmP##_g@<oQd
zqy(W^Li%lXg)&hMDx=Vo_s$FU-QAY|AjRMm#7s-8@Zr5`>s<P3ihhY-aVQ780wu9c
zv5y@S8=@%7r?%^ETehsT|GmmUPCDUfq<Lbw;C&e22P`df8DB0NNENvqPWk+lmPwb$
zeE2z&srtZT$tuPF5z#~oHrEp#a`gR`)!@xluGdzvO+`gM?drd0Et-KzW{5bC3T2NI
zz{?@32Yuyx0r?Os(!y!1`Cu>|S2^^Uo`$Kb;mk@a@P!Qp&Q9z18@Zfr{FX005!=R(
zw7dTeQ;th!PWrM0QMfllkT@sor{oNwy$Py<@ESmFnm_ZlbWvUoePE~AU-qHOP(YXT
zw|9bnBlm>hzYrNTPXIgb>;LxlomSX}@nKy;qkJB?DO>Gny@UUuX9a!XUuwnF$#ZjX
zYwc2>diV=pb^<B^DUuu@>ft`H{7TM07!`3wQlqK>@n3iRsfrMLjPsHyIjREDI8U%y
zCc`TD@XSL`9W`9rb~iqb&(=-X=M0L-(Vfn+b`pR%)FUDZd>{4rNOE;}x=)|9_P=oO
zQ3u5BI6ojx_=ySqx`Y>6GMD;B5U7*Z@69NE^+_ZztIlzZ8YA>XAuoZ$^ZJ$3T|y;#
zhtTzT82uh|W=wa*`_Af6`oBveH-PzF;$H3%cbGb{F=G+6@$C}`8wgHGTl4=(lEL=U
zoMUW8Fwv6G*%~PKYD<;5JHJ}O<VV5cz>mPPog9%Q{a=d;Wrtv8b-g{Txb)R?s-$AJ
zh_Kr=97GC6QS9W{3UrQQm`EFu{MJ_4A)V$8+`!sFvpMS*x+a~~zIwh*R{1tHhzJ~)
zPB5A%$6!2+O75V!H?8QTU#$u;F~!Y@{%;C+i1?>HG+az_(nuG-Ah;jPsDWrny;)~d
z)IRHztZ$c_Pvb?6If~u*Ko-Ov1hBP9ZQ~{Av$$a1wLLbK<=V8{BVgUTAJF~652Ap&
zd3{4&eowDSZ1<YI8buo5(f;o&@U{@~yw`0;D5J3G+NDrsHg#oB_1-WT>jLks2}0(-
zYpcC&XD&yNEI`#F19u-OChyxx&J|)(Y&Dp7Vjo%Sv2RASPD42xTT{8Qgdv|%<jB^E
zVBev?r+BBhZ+aXG{O{E1VSdNXcM?NQvICM?%U&Suid*mEisPJwH0jnVL!EntDW#>2
z3=T7Hh>-aoA**dN^ES&moCfw!wIfDqmecQH6Rj52wwNHL+fXs)i;%fJV<7>TMW$)T
z{~5j-LhR0u@2jvElxwc^@0Ef+`g+nckTfJe>}F^QE%}n!TpK9LyyuOmI)gwFgvvrD
zgURxILhDMu_?!<HX1Io9=j6Mj1UoZ9R>6XKM|iV%Z6D<aVJu17%%X16{~LGFaUdx#
z2W)MPws0Ye(iQfSm_Skec=x{M#f&AV-1+OP(^xX=u!AT-D@VOo^-zejul1wZJkGOf
zIoq?l5jWvrm}?jAeng(H<)k$b+5JY23>>%raa}u)(S0rczey(ZAa~L;cLUO5k-EQg
z^jjv;?<2JfUMI&4X4NuGvPgDzTrwN8ok7@9eD;<ikWnv%>w3StACkJ`J!N(MoW`0M
zwSVxu&Ii?<p+}@D@AmK<@^VzS*)EjRyk&dH@;|SXfc~vIlcMuSJZLl&k%k|nJm>We
zXQ=pYo4d}pu02nS7!4<aheX6JDMri4*q75~^JKQWHp3()&`Bw}dPDNA;7J^11(G~V
zA)S2@!+Yr2w?kgJ*!i?vF!bMS6KH}>mLmVWXNB$!Pga!dg}LrO8b=x}a9n)UMf`xU
z?BUL}jyNa=@WQHWuD0E6&F64cMV}GO>SFG55QnTjjz<AhP((i*>4<?n+tfQ$Eg{;5
zZyvJ$GgzG2R3=?k!^NNJEV`D&h88F@MrH~6N8Y}BmYvP<b&z2(ap%@N4JpM3bj;Z(
z-@Q&4X!|CFs+K>jHjD^kkjjtbOlaS-)}14(Tx(dfj{dQqEebZbNKIFq?@avxP#rS-
z&t%mI+Ee+DTbK+DtdH1Gkb$qWQ<?e}8$A;?LdT^?U2S(A^7hCcu-&PHh$O!JBut%}
zkDAhQqA!PZwgS6S_`H?}&O>ckax>RBcq!j+PfLj`Cj%sAFmqzFG5>o^{y+8oL1{$c
zI(OzIb{mJyRbIT~%ZpM!U_X$a+v0c2(E-h(7^F*oAAEL6LnESnzV?SSROC1f+;zP(
zwkcb_CeCT;zcO>w&tZROR1un+O*&8f|A!XAiM`H#zQO=vzDL->0G17TbX~32Ck?YO
zBL`^VXK*T1fM-dP*7LRG$)akz*{#^VHjUVvmlHvPE0mujUX7|WBf^PaSe<J-`wraU
zr~g@wG|X@J32%2l^P?Cg^PY<tvUR#a?1*!LCK-bs!t18`Yw_cG2udyq7*{K%LB;jg
z+Wp+S=)TV8_jY1dqcPbK%{^_=NEqb)AE6_!n&)$p5{GY^zmy{iQ~qZ-!OHuMC5|~S
z?fLVTomCv=;ip{4l^k9gOw*p$``>X%!Ctn~#;@T!`)y0iGU~ljC&UGTWg_Ci4Bw@L
z`9^GPZ(yl^4m|Q_Nc0=VRavaUkpB%#g+->*8&R|;CSNzbE@o+lav#00Mb|mb@7VP2
zD%JUXA4H}jBivO~w({hx32{*A9nF2LgkP3ycrRPJUbE+?ag0*bq8Q5RRYZct=b`ZH
zCiZ_p;TOQO8z#~8Am{ic=%NIE_z%@Y)O9Mw{F;Yq3wY$Nwp|N}BVqtSUX>+hzF)Ts
z_ow&sf$66i_YsvBjXxGM{lY58DA?qwbbCQ5k3kEYb!hP<|7RaKWYq8bGx-fM)M;-#
zf^q`YS;G`K#ySVr?`J}5tBKIl*MK;e@c`#*2h`|ANk{ZBtK6o@BVnl}!jZwuj(ky7
zjely<nKGh}>Q%u1Aif9^x^=p`hT}>n_3F}&t88O(R2ie7WJeQ^1lAZ7`s`oq<$8O*
zttAQH4t~dO9sHsiOma)P9NS#;^o-BF*Q{#Sg?8fv3>_5z9gkfDbmO^`k5M$2@@?l)
zFExFgZdE6_-9_fs`bWkD#f2Oosq&-D$(ASl{zSs*+bAKcxPIvK5(o3&U7+#YH-;I5
zV*P|B>;Fbm1Ko^!6?e>eZ21`lwEH7d6t+Gz-vvYsVc3(TyDiTOx|~)pd-SK?_=3mz
za}u{Ma%coR?@4l{x@qxB{x?39M0|d}d*;7bGh&Xa9=)j%$=FM?%;~h~cz6AL?NtFx
zE|<x?UUi`?)az&5PZTpb|3(kuDXmk6$kwJc;Rv1R$()icFsvYFxg2yc@k+E_@9L1u
z+<#5Ok}WX(k1>UELrgz+qCEwS6#~d6C;ml_zFBVWQF8P7YfK(lc=H<SrRU#Yg290;
z`Hf|BOASBo3p%;&3fd%A(z}LVXIEzMs2l^a3Kvv$mtd=*`CpdwD6yaR@ZE4?KXLI^
zl`xbTA#$NXe7X8_gIP(w63ABMy*SKs7;X%8PpWAir~WWAT(T2L|MK2s{E2V8MS3SS
zfT#EH#qhv(LR(49OY7NYs==!E0RYMDyVm-MBevw{h#zH0K{n%e)2gj{a)Z>TS>Wt=
zZI5;gpBK~ZsDxhuq8Ln~pAb+RP%)6rPfRT=1K+}Nh!lK;jM0UWttlvZwD|xXh@vF2
z3ZqFbn6ruoR?+|DU3Oe(AY`rKLd)QdNY8Ae%6UTjRqK{1`oYK$H)BYxPnf!#uEAoU
zdZE@D4fUL8S(-Ih3f`RUWC8yj0pti&NQm0&?nGWT&Yt19z1~ip${H|I<%#GCl6KdY
zH14uQF2MJi_P^jK`Uu%7j90V%q^pVv8;{fs<2KXfGWo0$I97D1sm{%0DsPBwOc{GH
zdLZ8~X50$R(!>6IwWr|Qc!|cI8648OLVS5@x`uKxay?Ke*N#$TO}N;uIbyA(iPOnr
z+%y_b#Vw^WGeJJtLQX7hnDOz&AUUO)hQ`fZ@%W@TQWJiK_M$}wc`J*|YOYA<Jj&%|
zC3)X8k8K<Aqabaf;Q|1XzZOb*Holq{^1Uq()@A1(+#Vwr1js)BK?P_Um!*K;@gq{u
zv2248n{`sc$Gd=s;g7aspnai;T}>ms!z`pVSUA50It3~eN~+)Env6M^&5XO=kJ5DT
zz5B(tg1ef`^(m#&erg3hUPbE%^TXp;4Mja>C>i`O4y|*ThQG2JO<GpZz9(oU!4LVd
zh%ow#z<<lLg#i2J2i*G_o-J>T^?j|~G-TC+1Z9E>6mEV&@P+CCnh;h{H8_zCK((>)
z%ZGvX-qZ2V$?D&@($dl_c8~Ii0B<qADF_r$Ufv%}6)ndkeUrw<5Y>J!E|8$##O$jk
zqOeTQ=4G0+!n_5uZ~tf=;@+6rToFlj4GH8V!%reHV)o7pR+VdE>no;Dp7~W!C4;AT
zq@-47G2y2iW1`Qf_5fg;W*e&|_%XCevI1(x4xqwIv)dc24>Cc+XYKqE&$d0BM?9zt
zjG5}egPP%WaEOq4SNm@y@@5)1c{nK8rqZU`D&<Y7I;+Fb7BdhM1P-o(W_UWwwV0t{
zJL7apu=qgPKr~z|a_=fxx<ZA}Ey(o~k9K*2vUc^SqSh)xR?5m^ryug2ueCHRMwd!S
zW*;L>tSe*`#_*<s+vWhfkD+H|g~?Pr+H~=mhXEL!e8jz|h3f0(>FvWtjt9}1vaT>s
z@pCVLg|vnljCQLF;P<t?*lME`E7>_#Kb&lZLJ0N(zPlFWYIKZ4VlDulH$lFy31{OO
zxG{59<N;BnJ?cWg@eChoWTqa;C@ORc{A^I%?*Dt^2=K1w_2nG;oo_cz#)t}V>BYjw
zer-iNKpGSZy(}}wBaF|iQUxD;{&3%Pbo^QAhnC1^dR`|Ywh1{oX;Q`xWZ(d8@`bY4
zQhICh=53nz)^M4f<1oCn({7bdLMp9$W3M8S;!m)R0H49k`f~n=moXE>m$;wdWDZgS
z5F250OYMG8Y}UlBHm%Fh0+5n&g!rnEcAPBZCZ3OM!{aw8iRN<YLcU8mmDg4B_v%``
zmsU)^#KM?M2RK|1WQASHI%d<Bf5QL;BWwQ511AR9%>fdGy^74tOwPlY)YNo)^3;i%
z8u)-v9ejSPQAMP^QMFGh=ztwB?Gx2-8jJmWvJD!OM&e-tW9ch`XjUp`Cozr@U$s>2
z_V0YpUYjQU`KtJy*BtfFh??+7@mcuPt_K?RA9AVvI4K0nuqJYs<;1Gx<)m|Fm$L7W
zNs{e}AbxO&DFuKJ4a4DkgL=5N$Fkxy%3LT0<oKt-atXdjUCb=}CY1KKrvuJ;gwcBr
zF_tNf@{~-4V)7XqAx_Zw@nnP9CE0#6MFF2bIvJFZK4EGwhi?O(OB>429p8t^f=27r
zTZi|sh>4U<OpW;d#G6gzH_YzgX+msV^^XK1;ord?2NMt+eL7D-V4ObIy6a?MuoZP`
z^tnMgxuHl8T7Og@+NV=Ydh~vn_7G}WCLa!k&=V0fl3v=Z(sAIBAlE>US*6D)@6=kA
zTxC`O@btIsvI1jA^oOMEC5FXrH;2neMf13I*4JyU&qBAVdB|y#nBX~M(5L9@wacYC
z?)eID%EXJ02P`uW;iRl6GPk42L>)D>Jj=|6PFcN|NaX0>NXKMtsh+r}4%6*-OD<0}
zJq=Oj?m3RVU`dGCY2@Rj<_p*$f~0Ay5TM{pmJ0X?XoT6DVe-D(7s8i*{kZBuxzUV;
z8ekH=*lv9PvQ7Gag@DiblSf*?eid~?>5Ef@$GJQ7iV?_1LB@HNFpSDwqO)x?CAUGn
z%1zEyM__tZsYo%S?J*oM_X!2DhYa2C#*!PNSyo21n;fL(r?C{IMV#u3FAffWK;l0a
zZf+8r&x^s5T3OL<&4+*tOl34%BV>8*xJ!u9=G!Pxi57RNUtQ?5zJHYx`emm29wz(r
z=K|q-LgH4dAHsyRB73+RGnLz<v3=f*vs}S9kBf~P8@<+_FyRW!{7xZq8I3S>pCAi%
z4@?O>z&^wRrD3>@EVde8a<>WtP<2W*Ce97J@6<8Kx)kLwYAb~RE#PWk6%cG(q^`-i
z-3F2A_YVwh>_AB6@AqVVRaSP{+Us(y<h`3>+RGNaJsxL`^v9;;W%CGN#y5C`PsU1O
zrJ#a@Up{oHpTyz&*g1CCUYa16#@V<2Xlq=VU}L7_5vxo}%`=oWnNHvw#rCSlTxI=p
zi%%udutKhqxB9LF??&Cg;wkE*utgib<=Jv35|eN3rtRy6Z{>&0(0sJsVJMG$Z_x_t
z`Eda;uD4ptRr?rkPIJ9I>M7RRprYPGCjvx%B;ocO$w@S<H!RvThiW2@P60M82DPBO
zT_WeU(R9v$z;zU~B?yzAjh@3*jj_Dwzr;YEhMk{xlbXoUr|?uA8K+j0g2KW@3DCFq
z?y7X_2(0HYlshcS6Y4K_uSlbzVNs&%Sg3vW)I_jSi3}z~jq%I#$!<-g&*zHQF*!H0
z_N+}oUupNp`|!CL<jU&syJY}UNE)LeEm()USBO+;XfBKuWP0ce4u~#bVb<UI?WYvA
zqd`GrQ{VO&5IY&~j869@19Sh^>t`=+tz|PCOFMKIFv<}hHD}O5sr6^SNk=QNUefpS
zN@B~G(zNoeq|OxiE5-H0hhC`G&%z+!^_$G?JOc5y{dB36{{DE!LcNdWXf*La*n*P7
zSR9(yOQis(M{jz_xXZ`~T!N8qzckLgJSmOb17%PO<O5MbCu~L+FZy%avOp>O0u&Y1
z+CuaxOUUPd7^ATvC?{5o9OPWC3M7v5G;RUqcwcSz-3OX(S4+Ha#yQwK7(7UtT!>rm
z7$%u1lg8IsKz8(wq}v7_92s&2nv*LySj-?I>a0<yl<QzPR1vq+n;$Ive7r7dpdB%*
zgcQf%8cE>p%KG6Pic+Og;`Aj{<QX+2x4wWV<87^|@x!TLVL<`?{iLVR0s+>xn~+Re
zLYL>xM;3P^lsvJfE+(jQ!Le_d!`-41Zo}i6>-KG{*+u+ryC{0jySkg5_gUkSZlt4F
z>{I4?t<c>;W@V=);!GcQOqik*FQdcn*M8?_>)W%H)%f$DyL8oSkBftZNkML=mX;~R
zYsUIo9k!tt#SQn~zaUmng*`!r055@XgiXH5mhUH2(tF7sO%*Q5txN)T8ir&)d;P7N
z<_Sbrktzd}Ej^l7^IvR+R@r>VpD@Vel+19DLUk9k@2+7=XFp>Km)H&HI#oc0kS=1u
zkuISuH=PaMReH9btf}&1_EW^TUF2KWvfQB&*D7KCSaTm_O*<E@j3B90Zo#kG8g&b>
zx)ByZ;?8efRUZI4%U&`vlao!DmTWSIb7jLGJ_E(D?g#Qa_A9uwhNn)vC};=c%w@`+
zOgv}6MNMWH0P?(YsOw6yF&hlxX66#&dQRIDL>rm9g*Q1erSXQvqeUiK9hHv>G8cp<
zqxg8PIn@XTk{pp1l;2|qF$Okpv9hXW@@t&ERD>cU=^+eap#&{qzF3d%IcrKZ>y(F%
zeLYy}fNFI*25eh!`5JKz_wTmG+gy@S^Ro*3qCF1sQIw|c(0=>j{#bW>Gf2A^orR%|
zXJ8yO>XnB%JDKk-TuoXo6s&&nG`gs%^_H6<5)r3>xG{g=8FSs)7wBSB9$XG!?E6|v
z%;1Fsiqy;+&!7s!7B{2&W)i5Y`Iv&X;n&kQnAyM7oYupYpH@KuttI4ku@HhwpW4U}
z|HQ<=Fs4%<(8rJn_I>}10CN(jL)RoWid|yYxuUt1IgNYJ@|SbfWFeS6jlQ$*wc&q1
zbC!i;-T7)i{v6X6OF2V0@LuUTm^nvx;wzy*&N$tOUoD)+wWWLCItUS3n?0|cSQk1h
z9pK7rwMx@2@?9P*ZoPw;>{+5iUi}eED7Zy|j<hHD+@g1OBXzyHIE4Oy4AO=9`h8+z
zjUkoqwGy82s;#d^;kkm$f6_#q*>tds-*wInC7%hTm|nVo$9dBbH}@mMbwLsWzOO$V
z6NB!{cx^xm3c((4ojfA?^o?hI5DPg0rb|=nr$Ed0y+8ACcLufVXyqxJp=Ib4A@^r~
zYp<vhF*;co*r#T?o%4=|+1iCn=Q=f>6ZY?qUWd|HI6_WaY!Pk4Jd<~m>Cc*UEwzQ7
z{^wnBf2rNi4%c_uiq~_k_sD^Dw2gB>cez|&{O?>4I6YecJ<&~zfCx_SEk=`!IK2pf
zxT@mos&`z7Y{VtyE{BIB{hE-LR*L7P2p>XITjK#<yBq3vf{1u!Ev_aWh%Ko155{$K
z(8{V-(5P)~@7`<g=6bcY??SUW^XV$$KfOiiJ&2}ZLhcU3&fr`{wL%VPPOnjD7-*jr
z7tgMuyh*j|_W}`g<pdAIs-J(iOmga($leoU)$l9liOs6PkB%><pJxm9_#NcqIq`v_
zd{CZUe|jFBW6TxgdaH&qlOoe)!3-dNgimT-wW;gQ62J-EtEtJ}bc#MFp!KeExGkat
zdWuMy!_477la#1A&tuVjxe)ie5Y3z6=sR@cU5W&zf}gmTP#ZsD>ud7}9wD|7e${(;
zE&l#fd7qOB6pD_88=$a?X*apKI<=cSZnn;{MppRrAk(oeklpdlw|P0ib(?cBu#!OS
zlSaQUejG%<<}<*MXx?Uy3zZ;)w4(mb5EXsgC-)9<;d~E$-SWkvFIT^c4S}0$VLc;o
zUoXEHP@|$svwzOTFxTFx_z)Pb(%WOg3qtw64%@yXAmpEH|BkHBketEAb?p6dw5i5F
zYcrI9dWNgOSvnm~TM!>yQ<z#>{HTPa+<&R^wH(|_rCJNdc34<_>X)<@Eu6E~;doUb
zQgZ2c4wXm=;Qgr*6CJ7;O5K-xMiOINqJE@qx3!&fdM30P5u=dSOlhsRcOLK8p_+A2
z7BGMtUxyhI!&ky_GyBX&o%~^R%%i}!>&#RtT~*4Q|AT+%OTC@KUHnV<f$?|2xF{~U
zZB8CK82$CWm`rVGpB0C~J_})S{x9}Ly3iNuB26>QRFFWN4-rnVz^7hq0~my~I{7{|
z&IO*min-}rUaYuB_Rh!{WqB)T^O68}0ex4c`QzZnz`yn3-zEl8Y1;e_n$7xMZDQ_L
zG0|5UQ90Wxvp8=hOj=9_s}7v`6x}^8xn6VLwmMa)>5s|wL%*V-8mF_tp~ys7Gobt#
z9sL{kZN2|z6&8dRb}#u|2#soZHng7lbDF&;r8OJ|botb7HWzTx@C^iti3#jv&J~4s
zA1B2gPiurMv3r2K=nPWDru#Bu7aROUxp*jJi4s++2}*TatHM)~H-a&DE7AT->e3V^
zNbYkeQ3fAbR(;&qP9sq>#Erfxh{Hm>+=D?1)n^i8FjgG5M=hOXO~`yXlB3}5GK}?F
z>}4ODNr&O)sIs8%jE}4(J~fyEhP~GXLm8yX&A9M)Er_q@i9V~iXV3HgJnBLi&6B-H
zgPt4MbfIM5l)qkFk-!Qz+z>zz;h(+6o{Vy*^`rz&{x&URfX%6|R~$f`{wB`COfh`u
zv3l4}k=NujPb*xcJ$cHOBi+>$0a+ql;n<)?V$tC9m)ONsm%HLKq~ZJVA9(eKpgR=u
zP4fI++kU$Ry?#lGGHv9@1+(rNd_AIaxoaQImfE+QFx@Jl)5mIu@znOb>P$9_U)(UZ
zw%HDLnD9ANJTt5Rs*3<!bfhT#SjQC%Kv~Q!{K^T4n83oBNM7y-JMP;)xq6ly_sRVd
z+f6{vI@iPV%gUiDJqgmvVh&^tD^t3fNP4oR0L`;kEZgZMSSAFzhf*z6gfl7XZ<RXq
zU0ZC*zvR}mt!yxShxj9QfPi*iiirYA3lsheY6{gZ5sLW(hFy*R7LIZ9tKD2d^<2*5
zN_?6r|MNha!&+bi^kb*`Eh<e=n4q*ntB<_G*l!|ss}tIfk`*U**EuB=5S9W~PLp$7
z<y~p}*>5Niq?&5hwf^v;vq;6W2uql(Y;3B{jY95fOG*F=>d*IV5g-R`z{-4D7>xdS
ziZLUNlh9Z+bAZS57Hz7ubk6!ZsGpL%l_QUQUl&%pc?9o{10}gLxU75oK37A{`oU|^
zx}CbXA;T~c_U8tm1Wj>foR&LgzqcNYzWclmmQMIO2WJXx1_eZYs+q*EMvd0n^QEm!
zq2|#hOi*Ffkx`IWB<APVzM5jk2&1q$Q|P`GS?5oCJ9cRkfT*@MI=zGWxAr9L8<TQb
zwN5X?1=!T7rDjqP#*U0t2l4VA^pjD~%AqZknz{Zg*lKPJ+Uc;oV-t^Bn`RNvKW<^p
zxY7m0q05U1WK;?aNuftP2aETXQJ>gUg5%7SUBIhSneDp)x|cYOajy<lVg?Adl?Oh5
z3&n8QU8Qr;V|>SUjrSHGgJKi3EMSb(EmUvE;lX?vHOelvY^D-Fb1MdhUNQzUgRwmU
zLoUfmh*zSgjw{g{ePri;6LBibA&)JKq>yR``nUfAKAg?O!1ojkViZsA##aK<rygi6
zLa+6yf+doIZ;Y8#8Xm+rbfMS_-Kps2KIAB@EaS}^xsQTH6V~BVopHrRD%OsO^qiBF
z`mW{ex-knSX;B?zXfhkW=+>Ay`uc=x6%A9Q467#sO`Kft@aD+J8+ZImWt(!ij1N~F
zb9AW;3O?bM$uWh@L!sUnUT}ZRF1Gft%$vCde{)c<4D7ah6qM&js=aFQSa^AYW&%3R
zv37TM1-k0aEQtjSFTbIKJs0>oIc7(4LW44qM74@Qwou^ImajLLCOqp`31%uAY7h)y
zYNIVXXkHC|$JT~45I{fzo-+KyR5Rmvd=y1WAN%vH6(C@w?etw=b>}}bzkHtjkc^w)
z*P$~w&Pv*NcB`-05^QpYXPp0Yisj)b@5FXam)Tl<+<j)iucM_lVc%JeXpco)!hsUP
zT>N)~d+n#i!zU^hj@MFg(?M(nQ_1#7JrE1SeH2VA`p!hhXy@>)cd4Gu93)WAU?tfe
z;@vc2hG<QU@=F8sa8!d?kEgINpM}@_=>UrP*XmiOwzK!ZsY`~O+uXEgs+tIWfJpHJ
zatAvL$9C^`3Sphrt=Q9Dl2<en*5yS)ob>rJYXua9HqDGrlwTLLZGHxZJg;$a<TtNG
zFnl7hTG9?)_4GDkB_=Zd(bSLk(l@}C{BHkNU#K$frZ3IJ(@`;W>SN9q*D&rTAKKGx
z-tUCwhwIV3oRC??gsH;%CSaC&JZk4sV8sZl<Lk|iF&JVO+Jv~u<IAzLq(~;)#00V{
zR}%?rL}gKb4Rn>|Zfwgu-G+c}v%4o);%pGnT`}~8>6S#wm`=;f9~_ZuJmU1L>VNTR
zehy-{u#b1Pj}groG}Oz@%GxZMhxxov(`J_VZy&3l-|+O)FPLMwuPp|vl$qYvl)`V9
z?+r|AL%O-?1qV5+>cXF{x9~@N_<jFy-D~qarqJ+=RT!FDEwTYVQQc+B5Se2AXmnB4
z->Vou%gEV}{Uh7fzBgGq(?oY*kO$6EPCBBz+B8LXazVV9>pT97r~UhyTD`4ymRGqV
z=Cx!5guP`qQIFTf6}G>B;4)?^k9S`ubB5X%gKuO2b}+F2!~Ce-Et2`o%+xTh(*(WY
zC^_@eGDn^z204nl)jT6-*|YXT&}t>q5S#|*rv7VktgwH<9w!Iv#%<by&Q=ep=H149
zp1{I)<tnRS-YW055^7bi%V=30waYX2#Sx~I>xx75o1=2$ip4PIweVTtR&;%KjIl<r
zwSig4<jnNAv=gVSa8C5Ic`zeD)=3d4)}FCH{?zzWcYJvG5>vCyTcm<aRMX^NQ2}IH
zUGm&%Ha9GXvQERp-?f`smq}`;{K?z?1z6S9NE^VIqgE>5or)}O+1(%^Sa?d#!3YF<
zrU?NMmz-dx=-;Mvt>&}0^J^oX!TUJGB$=yF6c8kDP?La^fZXQ&Lw9dswWt)X%Xu=G
zBjH(7{1+F?kc;u0$kg}#=tt-4bKc>|FSyQMts$8<@=`tRhjeMounbtq^YNZ|AQdpV
zicBx-(E9epr8S$FoqoH1jyHIB#hn|p*7K$lKzOQVlbpc>ksOQw&fAkFQ$_AgyAhmY
zBfX_#=H})uRDllTT`JtR8u>3XYdJi%f$)pToE5j~BE8O)lk4mc$Nhh6_d`?){ntnZ
zZ~X%puPr`&3q_COCOr4eBsel?>sik9M`#<Fd(8hkI;eB%j3o|fIJcm3)RU9pYIsY!
zd^w*iaFO0BCF=FfgE~aVcTtuu1vRtt1+hh*BMyFe%qIZk1hisvo_~cs`<C7X&WCSb
zWQHGXjX(C!U0;M8RH2_ymLsmt2$`RpBO*E8L9#C{CI2hoJSkC#Jz+y_YP5D}49&S4
z+RWNm2!tmUYg8q+dIoh(+K=4T2(OwO)I~a>H#;x1nCx5)UX{2Xg93MQd9C#(D66Hf
z0Wz7iP_|JhK|BmFh-$PtX+VnHQ*b+&A<%S9-YOPq`nQ7`_h%gn#y_07mZE=qEp?vn
zVj_x}+ECI~_OfEdK22n0iESK1xM|q4p+Q|%Vuzv6bjiv%3otvC9RL1eGr5s?Qx0*1
z_U18$<?o%h3$pzT(#GLrT9fVM+d28LggDPVgBCpP2rSSuTDX3&)xngMr$#?Rd>c}0
z?aq2NXYSIj(Ocq#>@qrSZN60#!P|IQIVldcM*$J*s||#Ek(c7zA`>LyEMFxd#F^vX
z$_+m}ock_^lGM=6nAsYB$Uxll8Ij};T&<MP{7cRZX@ya)<bp)SU$OQU&&m`o+wo9t
z#??yMzm;2c8RC41_w42Hf`*3M)qFA_bm#IgzoZ$%n6j)^^U>&OwEw4}&&NUs2%@Q0
zO<8J?y}#n-C765XNu_c<-^s{INZ5`l_->Z>Ec)<B+SYWJN_Z0UFCvGP9<qfVl8o56
z{Hwn&6F$eAHfm%mC^(;5{cU!1TYbQeP1qmTVf-<*A?*--IFc(Gs0U;A;4X3Ax*x0v
zzt*1^*#q1o`6$ULhOqm}Ka0;?w4i&P#XT*U!no(nR^aZLFv&TYWGZaH0MTedh5h;8
z>ZiEe-JW9Gn~L5sY9XP4UyMhOp~oFvzN#+iRiXLI_*61!U&qU$t@c+U$Jezi7qn4l
z0z;y3O|joE%{R`i2Nvo(@Q9Zs@_#hBWVN}3c5sn5@^6P`8mlj-Y;NxIK((V2$oI74
zkHDglcoHl-bFyVx7c3V0KQw@a92IGWAJI^arExVI@-p^Wa2uH$9C(Tv^)xfM%Ab(e
z9WNG#C>~CZZ!vzD!?@=*C<wypE~ycIYnJ#<CNVKG%f`*vc6ZjC`m3V}>V-_*l|+@0
z>AImNT`2IYmyVCC-iZI#KRH6af1V1y&1dMX=xv7in1X3Dm4CF^jKv%hShZ9z9aJYd
z_yNcu^nJcFO-QCHjwH_Ja$S;QxUnlT|J~WXZ>e*2GhgurfXu`wzCQ{lDd2vmZ!C`p
z64rO+k;5XH(+GL3w#5DH970?rY^k?grR?>5Z9@NUj^%RBU_0O7_|hSzMbz;)o$^D@
zZf}nOonQ<s*Ea6%niPND6RF``v1p!x>ZTxh0Q<7NygW&^kDJ;&C;Rb`3tV}MSSQ4o
zPi?5*k4o>@Pn~4?<5daK$BmM$KR1m`Ov?VrLh)IL6GMqfFR&MeW>l9D8}o18o}Gk@
z-f@|%cLtyd#5YJ0tW}$)Md<moV!cXTvGT~!T^CdA{kG!b?@vLgV8>vTds<DtLNZr=
z2W29~Jgm>13dCHdXHGyj(n|7WgqanJIEP1V;4VjA$hoQAI&XW4f+MjZw5|K1x>j_O
zWq8d^NYsM1R`w(X`JIm4y7Q0r8n4imGye}C9gE*}&7HJ_O!z?;E63*8wPcaT33HT%
z%|$x9qwsCNEyKjLsUf|LFP31`ZCt-6aO<!}ngS2>RyLUr{()@m^lWS@vO?c>kK(GL
z#eyTxIouW&1f|7>vg0z-Sw7MZ+`7fH8NU6Ql{gMNM1oy~&DFd{#U(*2zBHI77x+RO
z6=scTso=-wiPC+|aP~ZxjvILGSeIk)RqZ*5vZ!Ff!lo;h%5~47I37;;D(bvSv4M7!
z>>!Y4;4AP)4#t*)eDZI^>F0B<gxx}YL2_p)NJF)p3%j9g){}?Q4_Qk?x#vpS9xmqu
zLf{tEu2+-i@%T91HIMOXbb^6@wK(8*1zz|eC)yb-Ih3#{p?N1%V6Dr=Em)(Q$u?M4
zLBdr)KqIBtdIk_(s1;g5%r#V03nv9`?+gA$^5~HmFRoTvKlqCuOb2vE*MG*ka0BD-
z`obVLh?xsjew80d_q11ShYiT(#PzqgcK9Jirh~)Co!k#_wCsgfB(vhzq5|U^KWl7K
z49mgX=;y6pYw{$?E(DehKrUAASvxg>$kd>HK<0Mvht}e^H6mjT9P-0mI*;A6g+rc|
zksA^yD=740x&B^grE3#hw81Oo>V=ZZH$P~}q<IGL-8Wh?3F=<rc515xy3|I`c#<v|
z-*Qw}zNA#(_-~)8FY;aOn+W~h-3?*2!|i2`s0!+baQlmKu9*au`t)Ugm>hhaj}!k;
z?~4jm-LXBs^tS+4y?49D<ykZnZZFG$DwveuC<DHtv4a5>8!N^<{8UE1NV^9usAhqw
zp^l}J_E;q?_7*CF(0e{9f3V56GJ*bIHvnpXCPeSCzQ#ZT_GqDWX_f3p4G;k0yDiaX
z(zQRpUfL%#V&!B8G7S0?a9NBX8JfpR$-x|ku1?GnYh8#^Z{N~KKNcuo>!l#Hl$U5B
zuCvIaX%<E{^Sv@vQ$=kryugiYyASQ{#o%nC_3pEZ2DhpFv3aUh?Ec9SnZ-9TG5%7w
zc;?WoROSKxUHl?ho0HxSPluWMm>RIvY}djUa<lW`;z8;|`Zc9|7!<lu>%Vds(O1YC
zrZCR}z=m9q?W)qz=k>o749K7FT%=a=^0ofc4ZuV)FXUkN(rHW2G=8MXC6akE5eAdC
z<KJaZ8B~E$B>o{`e25x4$@WdL#OY@`{aFLg_m5qu<?!>#X#&VtB(S(?<tDHApn21+
zWey3ob_ZeugA;0!sr>6BSyOsTI*e?3a<HfE@0<9hretItfhuuB&!8Wk8*A1eFDpr$
z9=(3YUUsCv771+aosntb_L5)G4T#R<&u-&z2`MZmf_)YD7;!DVRjHNyg!W*)`Z<5|
z8@Vc29ACIiD{aU&lIQ?;Ug$z^$VQMBtt_Tjl0839XJzGsWC1JGz9W3>M^UZ1${Qr}
zMoQ<iZ>+jqPZw>YW}Zx><y@(YQUK9ps47+7aNtKv+~wK6Glx`gM!srgS^isw(6H2p
zI!|A1rpY4saKvS`s*X8dr}^B(=xx7v?VxKBUQDdgxU+8}|HoBPNQsV@xL+N>Zn!eZ
zu_5SWO2{;P;w*o<Tsjw~nQ}<Yt}U|xYbV|)dT|3>Hnrd_t$)D{<a>#On(FHtzI^#t
zzsm9R268eybh)$n)K(0Eis{e%zuUqr4nex!6*HRw!Xv%cH{EvOoSj5FM5fYbbtV(>
zuTP&k8auMx9vTFrROAK?VA=ijL&CTp&?TsZtqu*^3b9DOd#^U1T-^RQre3;vskK`W
zcWSmt-=0jLHU7g;`EtVrYepZ&hc(i3_`I#ahKCBd?M1~OGe`L69$na52P5u}N7X-V
z4&G3dyMODYMLk(?k%Fqci76DDGLno6gKSein{g)0=DuLst}L;q8Ht?obtXK!;g=6s
z7!W;Xc(U$8garQSip$td_xuL!IMKHh8<a4P=6Q~zg&0i2cb6ttO6y&5PJdgBQN8_R
z{<a~r;v@6nU0#sG3NGZ{Gc;s))mj_(_HDc>K}ro6u-j7B=)SE4e=46N|LJ`rF%t7+
zV}3r|B|h9ycK91i^gTM(2{HX1>!euE@@}3xa{K7<p*yCsqBQG^{w7%CWw@tvaP!_q
z7lmdWJHjtNPW{KN9_KiPVN+V!tn};3G6jAjh@^hHKNhU7VK%kpRviV)qgbIf2%9hO
z6=22_GopMr$z6oL!?hlp4j!QQxSjpg-m*CV<0&Z6Fz;X-x}j#8sbDxSx8Y&N`d(G)
zoty9QH{G!54V!uglmRJ#2Q40lR(sJSLLg40ukQ+ll%xS8Dc1j$O%hVG%biP5`-yrh
z&${?Vp5{?QJk<Hact#^H6BixZde$o1<ZI}m4y>Sr8*d3mazMTQxT36V9sRHI?&Inh
zSVJxNkcY>=+if_UD!yvo+c<PzQt2JWuCB1BQS^t-3B1hTR!-phScGJGmWQB1(L6+k
z+=_UP@FkYlAZz9fvSv&w$ew>Yn(i}~%Jw>CxCH5j&Tu^cvOJHr#D+8?wJjiNF;S2?
zFBix)^y+?Rsl!cD9vLU7U4nzNEWtr&Teg#he-`*yvgx{|>w-tgUok;yd_0cnbu~sB
z8r7Kdw^;kq)8YJAw%2_m>Ght8fcC~Xi5dZM3Xmrb>j1nEms%T4(q_0Wk<=eO5Z?{1
zK<AoH-cqD5Xhp=my|#{p>?SjUeW#~FSNk95c^Fco&&BpCm6A4u+gj{P6pc;vPtgk%
zb&vhESFr8bK11Z}+_p;Pms8iA%4rO<C?#@MKk<9g6jBYpNR<dj`};3Eaq@F3__o*A
z>)^Y`u~UANSZea0Dbf}MSdBmEF#$qz@tzQx2fh41r|LN1DtL^VQqSa$S>A1d+x~mQ
z+w-3_Rjpvfl-IDpJ0)DL<T`q(GsCf=f9V+<7qD@f>xOR5=R0ugc@yAcAk<JTuVvU@
z9U>ng=!h1<tsW4PG>R`=Qucod$7a#MJy7XkIStLJ+}5!>Fy`u03~!{HJ)9ps{z6P(
zz&O^CAT%-jc*9eYBmgn?2z}Ws&X_4v(XsNy2Ms8#J)G_AEMqM#6^waWK9RO<$xOg*
zCfQYxLQq3A)BoLG=ye8gwAKvSK=yhhIWKdHT*?C<^O*qF)SnB1Ssd*Z8&Ewv;Wh}y
zwZsd<Z5q*Nfr`C-yvHgNoDCX%sBkp4&u1NKzbpjC4}A~D_7z_phE2D}ebmn^$7FY-
zc8~vbg}|YaEHAR(NRVlXb{7KC@Yhre*w1Hhv2z7nB1f$Lcv3-jv{@@F3MB%VH=|%=
zP61S0N0CPo@^inoz#tXbMjwPtu;YhQFeWmjS;T2IXdt6(s8FR752T)sk9|hHM=S=b
z{1GmJS7brqE<?%V7UR53THJ;FJfF6^<g9iZr=c4?2O-_sE9JK&syj7cHwKN0EfJtp
zrVz0Ll_^#4!qwMv%uKhH3Lj;drOd0GzRrV8x`&WSBVS{JTQnHa?nWqnt=G0So}Vo9
z?%M8}{eFJNx%1JiEbG)AaXqRqB|e{5ZYnc|E0fwD5!^f=;XaC0@buAcdi=fXs@tr3
z=D69%AvQ#0t>T0rrj2+J1m8oxEXps)=g}@e=P9M4-39LFDhGJU0}W3^Mjn>>$pgP|
zp=h>U&%m_IFep+bC8o2$kHXBY<d*mv!<zKO&{Y$W?uOncNh^NKxp51(9^Cmd(iIPG
zC4mV1N^;^3Q6D(@3&PCQ9)DQ?br}O`J@eP{mAzf-9B_d3a?^(kj=R_NsigaUzftV&
z=bT!XK9>GEkLKUx0|MX_G>O-%prU33JClWnAS>XJ3Gu?cnemyWd!U0F$A<j_L!2eI
zEdqtYb42owETKQQnPZZ?)ilBjwAEYAHyE~Q0l34fj9w*?&u>T|tKu2+#8Sp${LvKF
znL=P3F~M<F|C$WnjG)dz*<z4GPmsL7uimoK(LZ$%l_c}cs$WwC`e#o<_IM0y1W8!E
zMu44b=ui#R!GerbLqf-jD;;QP>8O4@C*$=omV6s7HQ1H}G@~io_K|g{FR%7B97*av
z<Kpg?`i3GTKgVPn8)d1{w!w<IugW|0%3Xh62d@bBaBrAVXZzKo_w4CEWt3&7O-o=1
zxT$?F<n~~i#b;wtr4t7Nv*@2-_)ooRUpagqb~VkJ*xk*-`(i*wn0&riY3mKLOh3^Z
zz5io2ANJr6Z;dQCB9=uhp&W&!!U~&j;Mwx^*?YUH@c?OS;^k3g&NJM4A4{ct5t8&$
zdJj>G<hm<}>#pr`l~D0yi;X1^TX0HzHxhBkc>ozoaG?R$Wd^R0<I8;4f=yXXTLU7p
zP%U~i0-9xAyVZINuCLXC0`d0g#BXC3*8enx`th~pc4uWxb?M%V)%U9F$w5z^C=TLt
z7fRbKl^RaK*+oa%c)Qv}CWNYqQc#aUXACNRe^Wow8cB=;n+&-F&ispFn&WMJuJ!Kr
zk_R>YKhLqXnUmMgWGaYjD*pauDH8uq_whs@bPvlsu75x?hdn|%1~P$V>-h$ag{Vq2
zhA+S4I;EzI#GWsND0+opPKp~;&cq!&Z?oXO7Jq5VkZfsa<glBxolj)H$4u2($66*)
zXNDfrmyCYovj$d~i6N^Jm8^byjB4njUqYDwbe#W4iVk~5YiFwY1Tx>aotJW@346p8
z_KgUqp}2qnE7tnNQS>~<Cj)X%Vrf$-2Gs8P(Z5GLeTt3ee=*-eAoS~*4z>`Sz)n&q
z!+mQQR8MC!iM-Fh!CHHuc}<KMJJ3d$WY@uBn9mJg`6x%mj64GuG6A)FE&3?eb^67x
z$s7i;Z3Ils*39vKD>Z!;B=`I1clw!eXpq6P#5?g-u!TzUZYB{7b*giwtDkB6o3Vx6
z9r<rUuaaC*4tx2mMx~*($SD@OQL8UtFz7<jIsZWbR@g$wnm@JN=_2l8!3DR-`#yTp
z4rtcg1~8X~etgGBSo{EX9E$`fa5v~I7->Y1#PjH{BAWv7SCgk!(BBR#Jsvjx=SEQs
zD5G>{ejU7QhgfsSr5jB*E))B8egrhMcAIv^e^OV&V=gpfM(th`Luia^=JL|M;xD%C
z9dfb3dy6saQ84^*cdRRBrokV>z>pzCAb-uA^)@S<(~JmZJ8hP26@jNlTv4|Sd%!`8
zI4~y&UIwe+y|u4lVl&TB&1xp16eLBVOkc3yCIq|#iWgiU;F-b!1jo+K+tJmWXn|WW
zACS$NYkTv@?fa2N<=-JWno`zbdD6roIMp*4cL6dLF-vX7pSJqFeS?pI-^P@BXLWyI
z;i-<Gj$d8fCkrrt_>_6OVv1Wo_ZZL4zl{#d#E!DAI82LwA9x&9fq)Br?%#|bFaHbU
zgee5onHWJbe8q#n9gJf_(Svt(6((Q|(}N^3w<Y$A`>ry--Q@CrF=sb`DIM$$QuzZe
zzimrzkS;@kFu-9j^3sRr?S`wf_4XzrWyM9()7lXk;wLDy@tJ8_n(g;~kkT1Uk|#tC
zf*J8Qu>2-kC|IkM@fCrR8KlwZuby)8mCbN_Z`H1!JGg%Ek;Uw*27MYY8ul~YcG?3i
zTB}#ejvVPqx-9{+TK4Flzq|+5xPk#^B7s3@XjottKJe2jUz#g$ljJ<l-pa6&ISp>U
zF$9p4r_e;3!6>sP)fEh}E+v3Ba3E4=)9`g%E=B6>Jgz9HKId(#n+Tp!n=%5{4cC)D
z7zkT-TSAgBm5f3$-h<@p!WH4uyQwK-*I_g*+(lLmN^nL-p2j@Uja9rgcOyf8b@b*Q
zVPr%@nE8MU`dz)^a#al<D?!7KS>px#Hvd-hnC2F~)OT{+3Nq>ju;jr^NliUezDbCx
zm*Jd1svagbgbbe%ggYbf>kW~h*8i`l$l1++6&@5oj{$-Cv!eZa#dzf1ecYPvOyFKz
zR1TNZ$}si7@cXxucP*GASbo8;<~(1Ig6`6N5MW+$--(z{Y^*UFduccIv1Pw-l>bgX
z{baXb9n#czsVHYJ@TI@h*=tu0g<i&`RZvF8S$jHtX*~FxEFyx2Ba&Do=>1d9D<i<-
z5QLb{ZJpm%NEm3BKbdu%V$ezA)AWmqDm^iMAKd7%G;Iz3tdAEH6$wK=iGTJ7x-X|6
zcPdK>1qK@Z^nAb{&<?D7>4a$O?ASlaRK3sfy>vy(TpxEzO#udvkt?8x)$;4A&?)xT
z91fJxoLX>~O#RhEwErox0XI?*ppHTk;9xp^kdJ3JNiI-|uwD(05pp}gsQ5So%!iVL
zaT|uOVA|L3a5xWBV7)8qHEt$t4njCxD^8S33@lx6{{r_6y+duQBSmokMOxPcx3V@K
z;3%lLpF?lNVxWaVbN?BX<$^By*t>gEZ+DM`0WrNWJJW&debd*suif+%hlKWt6D=f)
zBsgzCj%Jk!i5*H1OY1CX?buI!55`kY*s%@_v>($0uB1vC2-}3(?h+f)YvmoH`TX$N
z2t9R}?FcX?G+SB^iotPs{GYNK-GBEFP9b=eCmEy9l<E2G`&ko#P#s_p@xk)2{zz5j
z>N(s+i|UZ+Mjonmc(zsP9ZwdQGvpKl>$E&`vYcx;c7e^+$|4VRCdvjA23cj~Y+B0O
z?Ti?%;9Zxmw$s@JT=i-0hr(WA!O*4$-OB}b{p4r8<ix%oM}8CAFwZxBFdYM6-L*4@
z7u>3a(xd5FU+JBg_}I)5x%;mR24Eu<_JR*N6zi^ADw?M*G-+2#tK>6SXzxn}VhhPb
zI3899-%S1`4Jdj~pA6K0iyqFCNE7U<k22DWa8?un3k+q)t;uFNDS0+1mn0NK!WCpF
zSmYzbH7Gz0&B_tglu>#(i($JZ!1eiU@_7?&!XPRmZ7d?y{=lbpe<o>h!5e_EqKTCk
zDTTscBqVg1e|N5Q7r5&}9lKJ_x^mF%mkN{tmtD3?`lB#*;#kD21f%)z{GW#ppTg4i
z(HHl?1gH;b^y$Rt=r=lFnkO_UR}HOiJp8*+IAo`?dA5>fi#q%07jiqJU&ay8fuFQt
zGqA5%-9o#)41t8#`-<v9ztaRWh0mdwD`*1n76XE#UOVl^>#@gXt469i_@YN-z%`EY
zc)|Zy$5)0$)qQWDp&V(H5EK|XrAxX)0U0`^OIkXGp%GBz0i>mo4k<xm2<ZV4gaM_c
zr3GO?_#Ym5e%Jfu{WRyg=Gy0+efHXGueI)b-8?!Kj~FfQ2|t5$e!1^$Q7n-)Qa|A8
zJHqq8)3I;+?R(n=<nSX2FOC4hh(IMoy~pIm+CSAm5H2t)IKM@yQ{^V4k%CkC7|-)p
zaOXm4%_{)$o*H2qN{Hn}i-oNRRzvE*YRKT6$K#%y6D0{39jLROZuxQy`)xi!5L#_e
zm&{SO4!&r0m`93758);KxHs+#_HmTLo}dXo{V|v8915g448J1wN91dD^}V0ml!-hn
zu6&hznwGYdAsg}J5dp&oJQB?iyUwtUOa`c>a$X>&?*#|-W77R@N(!=q`nNJq6=x7#
z(+^BC3pDsw0>=TI4yn}BBdp)`ql~tM*_vxjbYAX|br(y1+O|Sh+&%gSPMdaP`ReqD
zGq84DPA!CON%~o-(L%TnOZ4yn6(Z@s65;*WJa`i0p$UuL#(Q6y0A;uCz-`~y?=;XR
z9^LUck!;6a^y_9`w1ho8#$3ezs8#c^`;N29kJ5eOX4P`zAuouGO|m=~H&;{lHWjt$
zM|WqO_F6C(+dWCk+8o54;LJ7Zxk2t#kQ2q?H7kULT$G!y3vFUj3j&ZFeZDz27F6Qm
z$jnI@hDXC)({V7!E=FP<Y3s)E4WSCxN5;k(w8Kr&2QVdSM{VYpIp4NLZk}g-T>+2e
z<$lk{Ii)G$RI+i^vprZ}zN<Hv1!3=TA`Ai(Lf9N1llQ(RX{uRmU%;0u5mrdWAZ^Xq
zKSDCQqc{a}({ItyiFG=;X{wd}PF1BNCfRFj>)P0Fwd289Agb7UkBajWf2f;3MtYFG
zN$P!gr+3s;!Y^bLL|{cep{k!oFB`JXJK(e+yS*Op`?{GO5`;7&r3z{XdSm2%N9b~w
z+B3Rn+S%D{f|!B*iw%~zxIxu>?uT`_b>`gv5Y&;rTDzif7fdU2K@2JA&To;lNaHO8
zbA5a7@UG<zLLt08K9qoRc;JHLOK(jp`r<v)+;2X_TT05hQ7T1n=N4YszUJD=xa1y6
zPEZO#4!LDlk?o=P05Jft`cj|y(7#_)J3hsO)L;Wb-Q2&sOkBHmw6ur|)Na#>u+kW~
zB)>_ie))snuP?j70o)RCw^}Iv#8H#veyRwYhzm8dVCIK>rcsh0GzZU2HsoEv8Bd3d
zV`gyS3r)u)`wf7#{zd^U#5DaGpKM#K&mJ7%z{Uz3C6wa&r8ow8RVu!7B1J5RcQ{x(
z1iSulHrjiq686@pzWanq-mqQ>A$Bmc8wI8imV;M~bR<44%8m;PTa#QtZXS-9K`{Gx
z;~WS^ZMs|r(e*FMGxpS(U(H{IGhp}5C9ZW?%8IVBT=5lmL}w%;3cf+*Ve|+S*-u%)
zXSr2h9x`_@UmmUeseCR<&Y$p}-$A9YWRY|-RciQi?-wt{8y!wrBr9e)JuLhOA>0fQ
z-uQ}+*kA4DAOx10n)!$Cwkg)a6kk~e5fTn$lW5^Xtw9jmLcQdoFG5W{4QByPx?=|x
z4|pIa=s#hNMHZ4x79;V>(eo@3K5(1Q<B8&Io*o{pJvG5YxC~70J*e3geC~)!sJng0
zV7y9&BthCnQD^s;gr=p1`HK$@e28|2drO&Db!jH=^AF(vNaDD-#27=ls^?%qf4*<H
z0Fn&0V=OxdIes=bdB;rL$nkBzaAK3wBrhobU`Zg&Y!*@T(c5hrKWn+*qUPgQcD&S(
zSz}B)Omyq^jr_;3d=;|Zfk4K1UMZHw5aDB$@a!rAEwT<t!++sA)g7_A@58l1_Z)C*
zQ}jvYpz}_L-+$hOwc1*e?HlXlQ9Q^>Q#fi^x=AMGgUh=M5gYQzV`_&nWSNi8+o0Bf
z69t?YMCEwZg};jam;ERWsZTib7fI^ab5<nk)kgMM^8dCT=*km)G{Nn{Y7I~@a@zE5
z3e<KmbJb^c%EDW4N`H~J;UGe|cD25D&lHHhQ_alAu;L}TOJ%Wh4^NlJyfZX=4TfcV
z(X{uTGa=l#xtE{0jMrs+;gI6^>U|R-7%n5c|L8JU>VWC_Uh-%fPA+=M>K588xIBwf
zF7~zmxPfD)KMjOF4j=;niDH&{dxKRa#U~!f$<96Uz+8<(*@y~p&mbXF^sHlfPpZl8
zK-wQuu;e?+x+epi^9a_CIcvr8aQHcG_V!*p+pFOR3rLa?@0(J4d)g;y#K)WpN`Yg;
zC-LAb)_@0$u49^=0CMX-_v(9(9ZWT=1kK^|ygykvs4KFiv=0HL`Zv7dRl}=7F%G`{
zn)7UP(b<ZHBny^e2`R($$moDw$LknkNjX^9JG2P{c?SQ_Dthz9uVit?o*nj}JCXeT
zD}7cXqmPsi4>>Pz_ioQ_;qaBCz~^*~o7@Zr<k&ti=X30iW*L5=5kI)Xg^+2M{&Equ
zFS`3^nRN`H_N$|#r<t+?0QZoLQV$8syOw_xq(*Wrsoz1o+@aCvpSfvMCPQfI89%&9
zH<ull{_D@^H7cy)EsI+7n67?N39E^_+eP2HGiQxnJaQTI5so2t?z2LgtbV&pScQJv
zYEloXpG?SV##3&eCYx%fo1(!;g8?<jU>V_f`AVWiKzmX%J+a~7?ZJM!zo$o-$11(}
zgf}xHk0H5ZYNWS*aw5+x-yUMv_LU#cc%Eb20O|_{RznTiR1maX3Z7-}u$)5QF@Z}b
zZN^b}%)6G^q=zO$A5b|mwo*_x5d}ENc)<d9PFoduHb}|P<~ZTIYc>;V>Gp)xly2^4
zR^qSp_hVcayXh;eLI$3aO<*|A!nHTp|Jx}?Z;tz<hIXZ~v$FJZ9%P#oTDTr3(M50b
z;qAx^N;pS8kVfq8f?$hYp3T%T*?hECm&>h=O`pfpUoymxLDD@dsZo1Td>Y8aQ^bk%
zEUwggXzsiyEM~?bwL1u34i6B^T=$~wmi2(H><SZQ&(8F|<Z3)NY%5Z)fNAtxaajHD
zHP%(&19?b4QXtfUHL7*&QHx|NM>~3rG5`Kq54K=8w<vb>tWtOPPJXDRFPA>&rWpb&
z#1oxafK<h{7RFpimJ7e?>=J$VljoVe%o=iXZdes}@1`-VS^0xjVx?hDGDsX}Nd7Zu
z@IWvCu|7juW+kAa-?$KEC3w1lpJ>l9W`#?}Xzwm$o$^Vuu1`#|JsoN#!LIj14;maN
zu~>8!b(nlhqxeua1w4JV9sz#qt|QH0*W;MM2F)a6T8i%lZ?KbZ<lc4u5hXaqU`p3p
zSe&7rwoQ+TeQsB9ixdz4qse}7_~hs6#T83>ymOnLirbz2rIMKLr??}Y-itmTQDM!~
zJ3+q3N`#vV5o}{o=Zbd>i;P&-n(KImd`F?v>Q46U1hQUGVPma@5#w((X907qKj+h?
zU7FL#!Gu_jd<102uZ^@9dURlX2x3)1c9@$k@jq9U04!@=sBMPgYKX~J8reF(;U!WN
z{P6eb*6co}Jf_7i9sn|!yKExmMfdnPhpp`8{d3+_5xWs)24tNgsCBlm%L<gyrZQV9
zE)S$XET)E`_@qJwTUa~(0g~3RV{mKk?ZEwAHh*~8SE7bc!MPwOyp4Rlbug*EAvW<7
z>>>9v``^4-5|wK5isIQ0w6Y&qlC7UAjW)V|Qn{V6*n3}}lwoNk{;IrEXPE``3IQ54
zh_Dzs>a2f)H558_>rQvDNC(5eH+Nz_D`ENXRJcKcJ`*avpeTG{1afd1!GTS+y&SNB
z+ZQ+M!YN^ODph!VA3cf!?q<1}@tV@MlRu_^pH99VGSh8rFtkr*VWL+#vPz^W4n?^2
zZB2a@>Lfbv!HZtL52)xpsXYxfZhG3_0aAI7uzV+}K<voX(_)LECx`)I4+KR+=Zck(
zyo$d{O1*QywQ4?XHj31ke(NF9i@i?TMf0Ps>%sIR<A(n1)IYQcUv7FQFda$s3vLzm
zwg#~lcWDV?$e0ZE3Ax5?Z#9Q}Q9P2n-`-u^;1Km@>_su86q~Gxd~uvgW6R9DQyXUJ
z$2&NRqCrKui}AZwe#e`Bx)_3-)+vq7J+c65VLxC;sH1+`qdYHfzb?fsU`JcQ`J^<S
zmH49sRdsw8?Ou}W-8y%eTX$fZAm%xBUhI8ifKu;b))Q<$i_UhZCO0SSr3Nsth-Y<_
z%eW_8vTdU?LA)Vz@wWPu#mJ&za<8B6E?Q=pV!Oh0ur3n)RO>X1DnK;2m=ASoXk>M5
zZ7fB3u#~DfIM|Pq8Q5r41Q#)iL$<!vz4ryupcYG4jKdgxdrti}XbV?(3tr%pq$49B
zxN1wnmUENc)?*KTLhgdf5v}~nhD2pbJcnDwZb9%j@^r$JHJ5SN#|S=CY?Z;>Y(naV
z$4D&j5geND#`j9Ao~xr&ZhJn)^JamykL*-&&f7~tm-;v_8}8y|M^7_#Bn>@3A$y*5
z=DEG>na(9*`e5b(v2$lxz27~*?Izl|^3&jF;T80k;eC?WZm;G{my;$$>481Jm5#wq
z4D*L-Qzj-|nSp^OHstC9+J`xi;v{M@Vha7jH+$U}S|mQHKHqpO`vO6eM7g2zzgkQY
ztW(i?xT;myKTaugmpNtTfE?-Dhbq;nRrsQif6hve7lTlQR>-+IvnF(jO2kAbPnwsw
z8NJ!zK*HfmOIXIXXo5hZ#3r0Ndx2VvZ=7=ebeD;+TZ~fH4Ud?0)fH#6bKxz}0$C4>
zQZXDObY}W2f@p%EiF@%_&y{oM8}jG#8-5!!N!X|UYX4Uf1jQ>0fR;f3397#o?DgF{
z<>?CBYE<)$B!P?mxagtjb3|#0AStb@6`UuSR2`!`lfw+6N2@uc?Mxv{k^nqKaD%GH
zPqU|7G<jfMh~&i!(MG}L0%y2wqZnfjHlQxyRGn`MkZ!S58}k3FM*$6iXfw#q0c8=1
zXOzwzcXlSWzt}E@E=PT@tQbod?tVt9n_MY3nb7OfmcQ?3dolqWF5d!h0ju-5Hk#&w
zxx_@WzalA{jC&%))&6V<#tvUiJb$WxPun5mx9hcs#7r+W&w}~}>?Wwerl_I<5GFkh
zMJ!pL42J>-<*_6I8a>h1f+($k(Ov=y)(+9@X38ceq3lY(%R4*Sm7Fdrpl4XE3b83L
zu1mi`?ep8RRp7l{TI+=9NT_~fN&iu8f;n7Y+ze0YTIa_M0U)!+2q0A$oUC<c<k`)v
zX49fA<6A90IcP0ny+yltM-Z%yeHqZbBapV!#l=KdzW>5=QVF=P__fASn$I*YEc3WO
z?OOOG56tY*;~O~^x<uWN?f@%)2uG$!Qg)nzm)v|vY)~oyKqxz{hmeG&RBx%v8isoy
zE<+8XWek>+EZvA9`x3y6-q{cM2bEx!f@{S(p_hP>pWXvx0YvwubB`FM4R_Sx%l0pk
zsJNS$UFj6dIPKuBf;WZXAf2_WmS|yUf3RQFF*d<yx$~K|d3`x{5TsFt<3IU3p=<yG
z{9fOcyK{RX$*gc<tufJaB-PcKZfX3xrpVC?K-LFaOE(TJJFA~qR)Mh0iZW0RHeQ(>
zRxgSg4Zs>savHfk0~AGdW;+g+Tu=FB=hM$k`lIHN!3xWeEK^(RhvWs&sToc6{j~W}
z6Yg0D28YO6k@52wVXFaH33~x?Eg2FM3cOm2NU!N5`9z@5sOx{(d+yVSqP52kdW6VG
zmMbman~b>}$o=mLpC%XFj{mHfRY+qDNvj^fWH}S$gcvY0O|Q7Qbe0!2k5rpPq-uX=
zcV|d+1G>rZ#PNa*G?d&YmF)9oEm04(pTCXk5lO!C+Hq8q<m2!&eyDJh^S)uL@RAcX
z*|iSSof7!G&#DjNfCsW{2x!HYol`%2PMq0obYP{O=;bm_mLq+sKVFb?BaO5O$j=zS
zD`%mq5Z-_-nx4dG?gwl}(J-zl|4!lae>0h7;TTnQepr_!8TlRldm!d|8Ym`Nz+yd9
zr5i~FwCGX5@S_;nBN+wOTV{C+TiIs6;Vd!Tzm(TGo6{R6e#OeCvrRqfM8N<Ex?wHz
zw2E8l1C|@>pF5A!95Q?zP>wl!eC1G5V??{mJD8x((0{9eVgnxn1|60MY!BgVGF#Cd
z!VyGDI^s%7^L0LhdMjU0on*eA7cXL4%^hK{Xv@mbB?m1tjpb#NB=k;iA<0q^25xAh
zw)lu&xY2b9_WXxN6kQ6tO+i0zGgO>1_4rqHUU`iFJq`3i$Ol{8whP!r>{S^jdzU6^
zh04w-O=Bk0Suye1bl;dJ^tt|RU&d}kF<@2TO4emD{1uO`i=aD%SXbG=;KJK@tJK)@
zJoLVS9xl|0TZDFQi2snDMS$~Kacv0PP>p1DJXmAjqO`|Caa4==ICStId+%sp^lYlT
zYq>$Q$IT_sL)&lo;Uq5QkDPHRnKSl9xB{TOg~4~%Pom?8Wcmrz`8DXYpMTU7hP)qC
z!N?EOG$<%vJAkL|V2cI3z@80@(|{wkObPMx>`asTgYBO}xp~#5Db1kX^W|ic9LKxf
zwvqVoMXLQW;;kEkdk}|<rrjdTPz6Q_&zEL5KD`uf2!7&3yX&?g29rM~cug-!=S}v#
z)Wk;jS{kZ`3`W7o{9C(-&zvK+S^*);AwV&v8=XLBB(YL9`{j?`hEV&0dd_H2rc2rF
zx~gaF)x#_BXy|lhXcM7+k`!-YMb$Ps`too)sewdL9rf<UjH8|sf{&MjtEhZYbC%~n
ze-Qz0aRl0FjTrWcAVtoU82UPP-4N1f@1j$9;raB*M7>fv6~^Wc;+XpIQch9F2rfY@
z>o*baS7gAa(?~T|_ge6&n@iuer-3C^wPj$M2Az|mh}Zpgy0b=$^x}&T6O<|OsY%EE
z|5<<~7kCa`fICsg=UZUm`fKqD*o9@_6ni+G+2M)pPNO6=yKhBi1$W27Do~1eRpu7j
z<rY;yolfeM_ppv2fpy%_&t;8XY-7)}vS7LfDCugeRhxZ;;>T-oDn@w(Als#JNF1K;
z8}qrzF_^SNrtGamgtRWQpq#$rS6t)7>FoHLFPC=-6@-SIFN4CG3fufCDl$sBb1o|{
zWXgAVZh8L2d*7P*{tWuh!)=b@=~Iq>(pt`5@46*P(qCJdB4}mn8HsmykEp9@9uifO
z-h(KrzQl>b>CXKGlRTR3;NfxJrp<X4xF|b`p}-@q=JVNO#hZM(N}qcwq3x)qmg;68
zs!N)6G92iR-uhsaiE*uLo{~<q6fihEDp@>(?^sK@8jO6VTJidO?2s~QBXp4y9QSt_
zx^vZwRJRXt`LUp{%cPlbw0|{NT6870HIcXI=%Dncwnm;5SXZO#8a8Gh6<cqORI!9G
z^^<;jr+HpTkXp_^GtuOPGCfy{Ds7G`@GnX_Z66hJOjBsGCoyGw_jh_>q>EUEVO7=#
zRoWr}Ivfw7%p#6Zva*9yWdXh<QsJYmz|kkXpg^YJ+w*hXN-vtsJC(wHbZE^Y?3h>B
zOWlu?gG!&bCsz`oCt0pt85!uxa<?iYRlcRoYsG3-^9Ys2CK?Nt1v=v&TB7ZOYPW_*
zU#q!%dA#_~%^~sRZ$mm7D4xpZko%_+xy_o3c@3*;;#4V8A7<{qVE9i9lwB48(sR%O
zE?muVgy$w4D<w)#b%|ylK|1O}RZ(@{7BBdg<3GWcCM9LD%}<$)w_mg;YxrR9DAcch
zD8pi^`Bb26cNjZ={)uMFEiDMeCUA6p=gIrBoEzRMiyIG1rzpnw_2hqPTiiIJSt{K8
z9jv3+tk^5DlIa`r)z~BUi6khR5%rQMl;j1gXTIGllJz0DaD?mgkySw#;=Yslw=G_&
z>VelHbtot)AjVz^YGzfx)s3PlC#Qx)DihU`Sp@3q@xOn>f$8Vld#f=Xc<3g*GO9|A
zWA9o8*Y}uk_}NX7;T{Rj!bxPa-s{<RFr8k+%@eorA10Rp$7Rg|v)8XJ!XbO?-qW`G
z)&{%ypL|cQ9>q#&m^Z;K{FAm0BBdcU%VvT2mA4+dYyP6Y=BbIZr2|$Iq@<$ga@$r*
zS(kNDL9ZsikI6htnIx>@;rZ~YUtd`1#bYNMc<Jfh42Qw?n^K;?i&^gv(T3>kp3UEN
zVGl4)+=S~Xmf%4X#g(AzNT(b|Oe7&D$EpJUu#%tfB8L^Ta>UT*;<Ty%z;imL?sAv(
zI4@z+t6~6S$oHE+Jz+e9#HwmGqjUJkP{J$w4XMsyA5HtdQ%>uqO8U~ZH9}tl_xyKn
z(B2b~v|%6AinVg0x_jf^L@6Ca{_7l@{L9f8s3N;gY$M`nxHcNASR=M38)~aqMqbed
z<D3I4(pJSdz~>4Hz6A>%Uu!L2yPbh}*WvGaf?SgnMyTH=!Y7Dl3$l!1*e23m3UlM<
zaRZo6;RN&5H>LOgP8d{U2HJK*-Z>>pZSp9LRX&8PhjZWiD)TO5_x5Q>6!ww=8F`*|
zjBX4$W7&Gz;M7-%*gs}`cZPv9_`Xb(9pd*3qAI_Hh7PElUM)NoC>^JHKY!~Wdh*=H
zpxdvxL2T(?)ifBCQ8pvxHi)?>+kDjF@Rp5vjqkitVz~ihr$Els?gpkXMjW>hGs|)Y
z+T{Q@`-8{edCGn{XbWBFO}iTDy?~CB5+2p_NG0#fgq9}jKMr9z>lJDd*YPh5tB_+o
z=k!8ukaC4F(u5rD8v>EneK#7Nvo1?SCjSHEklG)i*<{bT>2(7|-6L-@=K`fxe@c{s
zaOz_xSk=qgFH;iuQR_6RAA^R0YIW<sMh)Vox5s%s8<=<;AKglPs1wU`lUbHMm!<IF
z@`ADOi@f&uCcO5jtQfdCp+_;9@Q_H`lBkqbw#-o8k&OPLE4*0~qFvRA%Xd`hMtA2V
zMcwp&w-t0h9*=u$x0FfFpi~;$Z*(C|ch;lyoUte5qn^ekiHyu^&e>1Fx0n&Y&8Ig@
z(XrkON02R5**>r1gqX;p;?-#zo$BY_Yt!Fs2l4;f5u>beth)YMFR!!-N4FY&X>5s#
z-WWt?A0FN9AC9Z4M+_yM<5d&`U%o{!Kx5xsHUe-yNEn_W8uvS_3miFDDlYw-OTe(0
zhp*ZwOhoQ^<O!_fGq$9WXP3n25|L>;;O21#!zfJQso3JUx9EehBDm&DG!PT$R}C@`
zDZ_=SoK{Ek_G?M{FMo=V(f-fAW2_e9ZA}U3NrYl}8S@D!;Pl!QVKzCqil+px{#?)>
z5s+5pE1ol1p57#xAcqA9(+3J4YTI3;8HKcV+npI_4g7ThptyOZ+G}+aT6zWBzPh}}
z+KOyk+H7QtIE=CLyt2?UCQCi&Ye!SXuU}QZZX4V<2iHM%QgB(qt^(`CWP7}P#{2Yn
zrG)U;<PEm}S-$m?i=|D?G3AX;XPRjFu}++O>Wa~j7~3U7EG4ixgB6|onT(L+4hwKB
zpryG0>TP4YlVt{k+c_h;%nSXyU9Ep(ltiBxvW}}?A7q>-A;sr46i=Bj#h^|-i<gq2
zmyM;Dt+=&^E%*Zn@CgXq=jXpK@IaT3Ph3bqoL`6wyx`-b=pzENq5o%qi<^yuo&W!T
UfK5ij88`q?Q_@tdlD7!^KP>T!p#T5?

diff --git a/docs_old/logos/salesforce.png b/docs_old/logos/salesforce.png
deleted file mode 100644
index 031f942b0f9533d811b31aa3c5d15e3fc9741f23..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 20712
zcmXV1by$?$(_T^<2?3>)Mp~pxq@@Hzx?F*m?rxQmSm_3(yIHzp=~cRxhNZh}zlY!T
zUHq}^v*(=W)Xdy-&rJA7H3dSv7kD5Lh)_vURucq52mVE~z`+JSs3jO)0w0)`(yG!R
zP!$CK&I}9q{t~RHsR{ylvw%Q>KR}=x;8EZf2;{~K0_~WBK%&VY5T#>!lZH6(1NIkX
z1z8a4@$YwQK^zFAj#849{^U8gkMQs{oJ>I-!yC2U=GZGA$i?E%P^A&|)&BJ%Rzshu
zR(fU2cB|~(o&U7uYaDa#J4I&91sqI_KP+cN`9jnxPBH8Ddevo;sm}02;^&E9ly%1b
zd&_^*X2R31xcBeBO_9sR(*+l}7g1PPSbrdx{2M29z4c{Gi$>4B?zd4YLwWS?9kh`G
z--9}-u+dC$(f+{iTJoNSvQ71M)Pn3x>%YvPt4SnsC0~7xoDCc?>fKx6!fOf(BLyiX
zTx-9&v*0Ps2(h{7t0YdA=!`~BC-;wyqLM_vB}s+^);oTwa9Qb*7l?gqzg;F*0$UXg
zH5>k;OVrIvcvP@m;1)#O+48fh!pt*#(%gX`-e9drLc$ChyQcN$i*})z%KX8{8b}+r
zvQ$X77iUTbJCq#7TC<+$XJGgM!d#w8b-F^N=(@b-m54e#id$jiB>u9%{3+T;JBo)P
z8`Ct<?WUw}89yE4Lfo_uWS3P(<wGy*!Dq7<K?Qa*4GUvK>z6z&c`rPg)aj2QtZ8>*
z<<2~H&7o$CR>L~+n(px>Z<fBLvI`&7S6Wc)8UAE#Cc|>!>3ay)Ggr6y^iEb5_gx6$
zuYPw|@fu&-VUUFJ@nH<4;~^yE3Y|M^rrh!I6^bT8t#mI9ozaerN>hphnR+cs8|ebh
z;KlqefjPcx6}P)DEo%dT{lA%~qmQw=g94Z~Zv~d9YE)OAl4zj+4cOERJxQ%FyTYvu
zMUN$&uiDgi&RF7eQAz8HVVZhjZzS|BkoJi?XE{4oJtp`=XS^w*lL{ON`h_aWvIQ9u
z5b7>ugzIA@pHm60R>!>iA`t{t9Zo0J{w)^|`r0O^tzb5C-r<iY->yq+sM=urieL5p
z%9i|?KGZ^764j>L%)?8V42nO!mSBjj!1g5Gb9P+pbz5Sta0@2I0FEkU<?$5u-E>E_
zdB?h%seIULI5M@_+!G4zN4uf)cn2${(IBFS1zO*L2#lesO})rl0zTBKcMs#^;ujj;
zgl2a@0#PcR8L_SqteP;{GpWY^R;Pk!acy)z)$oL9Ri@+7$XaM8xq7g;!GA5Ao#<zc
z;RJ<|20qky?vpeem;@@}$?MOD@S%E3s{&|03Sa^o<|JT(ew&cNoTp&^uYNvgC96Z~
zKYYoqUPc4A<OnvjEuZ=_PA0lLtUlvXDvNkZ#pDm%uMOfRmz-BpzgM+YnC+>Uu7m!Y
z{=7`Wdx{J3({+)OjR&1YU2Bg!4cm2iVliro;h)O99`aAqUfM|{#ybjb52-?7@9_uD
zdyvh@QMr)S43)SxJie+m&6xP<>x?IcxL4%A2rL_kg)Yh>v#1>nMvQwJ3BWucyot&z
z#~=mXyC6Mt;u7ClU}>3pE%xYw{$f)!xu#zeMocX<^e$vKY2zmO9rqJV)i7uT%nKI#
zQc-!{dx&BF@~&6em>p?y0tMiGhTpoK{U2~kTYnCo*zm7%J2{OUGcA<xZsG_S)%#)I
zeRCYL)0ch;2*;(U#YdD-B0akrMpx_A7nd=2lb~Q5B5WkGqsqG2xTNz)#v%dnM*$-n
z^PQxPzc!zw%ag=7@<_A|8Q)DBZcc+3N!aR6xNw_IP>ov~3xY%URAMu}weTY#>Opdo
z&l<#knDJnN-&yz7A^GZ8WBl*Qh3tlzCm#HGb+Xa5{Ptg!o*oWAZ`S|wlo^CE*(LA<
zobiO0K#Lc<mMytLOV8b){q6!wfS>OUZ;whqZcsKb>kV{IG+NLpSSB7sP&nFtW4kQ)
zQm6B2$CrIbvet1s`)yQ2iNdc+A+zCaynuEu{43Ff4F`5|V6iWCKk`$99prXzHzF+i
zvvjrGt+IIN4L<SX+XUbfX#Tx5q5=CWb+v{xJ|lS?{;M6*#0|Iaj;<^(5fP1*jP(g}
z@14bjhCiM3Zk3|a^7UUUq@QE^hm8hvyx^}nL5OXx(g*d(>+hlg+v2KF7q*IPYW4Z`
zgG`}zYtbGP9b^`i<@+ffzD@Z@Hd#i{>{pG+cJRo#C}u8o!B9pdt&dp+Yq*|`ZtaD0
zWjiMNPaAnd_QOA1$8OwoOTSHl8xeVMSJM7{!()pi%ynu1Q^Ls3@fv6{x&JGDkWVHZ
zt|FAmQ)UUj<aO0T;i!rCmT%wWU7KYF3OFB&B2iRi-4kJDJ>5J7K|ttE5AJM6F<kJy
zH+fpk8h->&O6?i1>iNBqT(ar-ip86=5!9$~iq1q$klwr;t-qt5<Mt3+`nKap)gozW
z%xzUq;E9br_8p$V=f)I8gh#_wNAmkuA3$T;3KHje?}}0lW7(oNI5*=_HF097l0X~s
z;?i?o7FW8O(0VrDC|)8K+N$DpZ#8+zCKtw@yx>!@bicmMx}J|lFa<z1NQ2pi^lrt&
zFSM?2a#^rs-YB_C;Rnp!gex3sB6<7@pV}N+`fDFsilTnj)fb=Av(;;C;@tfyxZZ8*
z#0&Wk^Q94I@dIHZs=y+VJpBWNgw6*KWM|`mW>Az&AnZVbli6|2xJ!-)=+%mhfJvRI
zO(KD++~|9(WQ~sCpS^+SL_y0=AJ754-tP8bbJ^F#gW+ZUg+yszd$LWQ?U2fX$nG;Q
zQ-=8xO(0<WN^gbS9vH@WN=3q;Q@4#y=F#7PozT3A8s-VIveqZrR)BpZPlk(^WP^c)
zGgD}1;P@(xyk@Xkwuzr!DCAXa8q5>{)tA$JBh!sGOHS369qTv$l!!7mdm05gn-#)+
zJWScKzvrCAq96YEELd)`!1THxRZDg()*Rm-6=W#v-TL^_*bQhxFw?igZ?rj^<S}p6
z&=;AeJ%ssTwd_=}L~l?e@5GN>fVsw^)f?Q7C;rWS_YdctCZ{1kVY(1nR;s4#8!%Ic
zZ?K5D^4LiYmluzsmNff=Cwm_|mKeFsht72~gdqWK9bu`~k2O)Tc6G6tsc5?1w=1bs
z(Pgk1@1sdDLNsi;#=@8OG_nNa9qzkW(!ia9z?{iQhFEE6ki&}vu;AH{L;J>e^lyKP
z8r&HYov5_XA~9YE#`s;O<#g!(eo=^C%;!#cwU}3SA;T@nn=d1S8z>Y}4DWP?q<)lr
z<3Qdasb+p1P6fU_yIg2;#%WuAMb%84=ojP{bZq6h4?*YPb$?#27Y#Pa;X(><)6Eut
zm&5I_L;wb6wF8fHZQ=(*?4P8ip(4&(BWH)m&ye05NKb>pa%wl9%IGq6E7cKpCAnvi
z>>d_(YJ)ax<efT*V#*NvbDWhdQ(WE3+<l6{u{31K%wHRaUSTD5oG<-7jMpFP%y&U(
zNY{75>$OqAH_qfrMVO1ZoFCQrBfBJM#afbN;E3ahfB3anz5xWCqMWC*2i~>umH@`@
zC?r|ZeZ)4kC?|7A@`x2MUeA*SVhJq4W=smVWto(@yK4$Lo6B~^{E(id8CdDKn9u6r
zXI}|*={k~|aRj6WhuJl-nCD3Y4J6q*==Cw;iywR0vcx9$^sZ4$p4LyjpeUEX26F_}
z1fmBI#(9_OMS@XXdncFAqHUEx6rWM7jv1+{vem<fMV#%nNCW`5v~Vv-P1S?*DNFt6
zV2vq1d?;Nw81NMHSwN2v=2H_kRhb1@3};-&oF0|STSrNZ%SL~bS?s&#(+|;&&(fYn
zt0;pgCWpJ`<gDM2kZk{9b7Y;m`zB(3$d1-ej$g8rof8%u7N+Q+Tk3}cF3+#N<6;JF
zBd+NQ#q+?QWo2ViaXKWfh9?bmp<jOcd9li*ZW)NG$;svt+kb0U>EgCDJO5%|_FU+1
zn{VpUeFaa@74}RcegZJ!QkdCZmgB09696^=$?^F~nw0}cj;!X(K!Kz<9N&&_sSbQ_
z(b+rm@!KIin@V39SZ-KY?q8h9*6^QoJbdleOSbp1+xIx8fc@|El7nT3g`LF-_dSqw
zNqxOIqKE?J9<2Gv!7_ozj8apE(NQhk**oC3=)drQZ7wTA6O)H!0RzgwG4C=f4?kE$
zTWO*H)uEW638~`Q@+2dPd3!2ZE|Jyv%)#;`NpVR5)F+`1O9p!U84%uNrkF5{u4B<^
zm%Z^n3ZCQ$!WA8rm1T<Z`4>QXKC(#-hO(n6h8WoNt<;pbeGLy&#CFZw<=v#QfcQzo
zNxFGUq8~XG6S0wy3=p^zO!;NXX}bVmMKYOnH;|MhhJdsy^_sJ=F_U>J371~@o(^r{
z%scSoHqlERSX@}x*WaPMe=72QR(_vRqv0g&RXpz7n;EN_eocD=DSg~`9=a~c(X_ze
zj^m0wCKku3e(@kr^W;Bk^0Knk9c+e3t4$qM9rR@@%y+nfz<5#8!kf6j%1bbR^VgTZ
zz=-^N%)r2)$3zUKH`SO7{{V7_A(sqb;Xu<rhNX>JST^vF=#?B(2PI%xV8l;EeXVI&
zlU5pg+^wV_aQ@EKQ<4}?iVjE0h~j8X^dr3wAYgCigR3^tHY+3)aY2<X%5wV;tPBho
z(yyX#dGW3D+0Yu?5I8S$!S7{dvBO9s{c77^W9O@*FV{R1rI3~V8u(mJ++Qj}5J|Xz
z2{I^O18_VxP4~moM*hiWoX2**F&`hX3>DQ*8Vd1}fpngM<Yd3TyX6K9>=R>j5VN%p
z3CZ)B;VfO<)}@XQAc8z=PvF6FR|n~)&NJe4%#Js@``io+U6?c!2c;1Hm-3)99AL=Y
z94+zhy^}fwk(i*9$z@<$#PnRa@TXO%ad6*7>l?_*u8-emyNI2Z+~j_+7i9)X*VO?|
zo25!8pVZJI=5KJe%)5O~28I(X6y#dlWN;L7x%?SgM`4tltZZAy@=?-PDLhAV9TJk5
zS~wcZmtja;nszRR4Pk^e=?ko<BqXzC!sQZSkCzaR6UavqR%8csbRU9%ABM%#;@iYB
z=+M~EfOgvpuRcu-4?r<r7P~(KzJA+4E<me!BS=BxrJo-k6gPjR3hzkJiwDh!))fPe
zCF>HnbMAOH2Bgaj0+vO$5qK-5_Xgc`XEGD`YTT^a*0m~d`_&)a^e+x@NfbT2)piXN
z2yr_s>hgfELb{`7X6ODX8K8zwCCtF2n!Ii!8%1l#K=j(y3cz~oi80bU?~48Sh0y+Y
z%M>B<Xf&sY1JuVkCf=#a!u)4AMO=^lnB^HHGb30^hgwKLV9vhsPt-rd!dwkNS%+eX
zYc$p}$u@z*LPDUWw=1iklE(#(<JjIfj}`{d>P+#K6ACLO1w>Wyc>cMt`@(DQ5EhX9
z<F!_V({`-x-1GCGC2P0g&iuY01GeE6rFsrX?Z(gf^qamq(4zpsM(H@-w?Lo*^3*5;
zuH=|ZB7^fBe7FO%!}@<?dUS*MfiZBBxVr!T?t2`=3I!insBJ2&NxB+8kb%F&rT8Qf
z%2!Z=C5N3E(<m7d;v2s%9KU(RhWFw{PGVw%va0g8NmV}m)z3ks+lEB1P}?1LDRCfI
zPpXgTynzIxPRRn?kY5g$^I9!65`B!+bL6F*S~RX#&U)Mh5F&o?e?)mA8N@1mTrI6!
zgQo@zociN~^Yhtw+5Oy&NVh~#a_!x4fPt3$Leb|nLK>-L{T}$`u50bjiywK*(dRve
zh<o$zD`xakDW4TLu6&*EC7Q`4pjA<1Su@3^Kle7Pm4xf#Q)JRS&LL|0{_FUClcsuq
zpwz-$ZRI9$Yy4M{UlIu0`1rCQqc1v{eJIWvPM=&ML&CnXj}4f|DbD|EzExnglY*|8
zSc}E%t@0lw$vkm6{qvn2=j`Bw%!MnU!dm;IoG5r-DJ2Qia6wm>u6V)QQO5t_YP07q
zw`B=T%{U=XidU+LR2yx<G_~&?>aecMP;_qq&{z!&$-IA0QSlGa9`@wfu!#g;4QXJs
zPe}8GJ41>?-!B^fYITjV#fn!uCmn$7F=Prq#J%A`b0iJW84CX@h+!D;(W#!rv57+R
zzp&$;d>n}Ff8f~*Hn;r)R+PVvDu*3z88*prA6->R`+32+Wv0LZEybsbj>@DLADQ_5
z8ab1`E4zrdy8GFDLm4gh?`w1Ig*BO6T{>a;=O=9*t+R|gP;Q4jPZ2`@>vZcFF{j4F
ze@$>_E9D@AQm%*Apk&gn*FsgB_wLZ9-u5=O76)^s-m#9+<|LbFXL{iV+BjgHRjM;#
z(26k(_jJ8WsxUy1BZmv1)!MYpiYDf3{J>I@4XQBy#=eUoaFv!zGKh_<bX=@)%kt{u
zVNkLV;XEY;fo$wGcf-FX8n&<ORjqKFc%Yg-FKS8z#7J5Y4gbaLAOtnN5BvIyOd&)<
z`&LUpDkk1FS1Z|4%H@6j<!b;D_g-B&ls3Kv$37!LTR!^J$!!9h8N^DyUGeogpZ~qE
zsrC&7#91st4D$^WNGds{91tH5b!|O6+-PDW)R4HPRKgYF6KGWc*_-f3CcAJ<WoA#2
z4ycGoc0YV?rbF{Ii#T|Y)9gzw>S*0g0deZOpleu?gqT^g#|+oOf4_W1oQE>5fUxT_
z6n#OkZwsLx)t=WtZ(ye73=tu>tdv+YIlS+bqE83tBg}o8%?&2Z4VsmWnun_Q{~;(O
zw{6g19}|%|oJ;WK-!f($uXs#xZVAeqlx2qVUd<gZB}DNF)>@!M_kV*6`!D(1I@B|l
zKGL}IV(YIa_7x5(L^(~pg@rfXoBV`|FTo|hf6Ef-1L?1!JR<EPno4I#z;?>w=2<pU
z#2zTn5VXME-oYxmyQd43CqiV_{-@*cFO`c`xPf3jv+g8$0bjlXY4a)-Rot%S`#)LL
z;!?qAp}*iy9bfL~&yglhHJB?Kd^<0r4baq{eznIR$#cYC-l5(<?KF!1wwF^Zo10#s
zuh2f4f4M|~`t=wG8gNNQn|7E@i3UTW*IG^tM`_45`^`iM@UL5#%{z+ws46BEyu57U
zeWsz+ZN#E;fnec=xBVL3?`inn_UC4Uz$bTz;(u*rh>hSmce;71m4no&(!%~tA;;B1
zUe;)W<2<uh`Q%R=o5GYP>RL1@H^QSx5+`a;q-8kekUZ&cUk^{eb{28gB{KI~R9~4>
z{5HLl-ST;9^nRD21rd?w;g7MjkQ3O|QI6quQuwQ2?V;sF+;IL{d$z99(M=fweV|R*
zU0bMXq~T`>Wef?CRQe@a(<ac#pT;QOSd`&2UB}`V-F1a7MXB1~Z8>be`bAF@0UgSV
z>3%Oc$~Cl+ydt8)P4qDn$vVAsEE&a@RJ<vyJ>n#38&>0LBc|?PUr<DeJM!xnq#_>a
zd>8gqfndn&W2MifCAB`;<1|a$_%dfE`gYwkV*fn}^$p?g`ypa48<rEYQ5hQgMT^U6
zD58!8Xfe7}VxX7Q^+VWC9;DKXtYE#jbUG&>y<;2?y>qz_;b5V3-p;&^96mq4xi`e;
znUKHoaia!wnwbv|_dBz!m5G0}w0%m9-C2g5v9HWl7VFCHe7(RG(@rKe4kS9NyG>th
zuN-#6nAoG}$z87wjIn(EzDP25wc6C{?94T~dy!pM*yQopzoEeZHUin~^NhEV?N|Ke
z-S1a@z|0!#R5?o0nmqW#-CUDHqo2oCyVy?QpD#f3drx<XsZ+Zz3&@=Zw|HAc)(zGY
znFY8En_I$(spI0-?{7N3v^DFP6;w19<R|b*8C(vavObY7!?f^4D#Vk1uifJ#jK&AO
zvb^alh&XXiz83|jONy#=C@l(r!8YH-iVdzFsKvGhKy!tL^h;Tq{iVlV2sL@hfG9s}
zmmes-ss0)aJGu`p%4ka|V+-w!Ai(j1ZBP*WQHkD=&XcB~H}1*|4$5sU05A8Oy^5xx
zSe?6%Xjr*tDL45yQ3k`zarV1I#=qxI{e`sRa#z1nfywhABS$tWIMS{mLhfff_Ke!T
z*yYKF{g8rZ7ds3mR-y5mj)(t5#H-l5<do}~&kwRW;qb2RCoU6}>zS|PexeW07bKVD
zKXaK+wI~PX_W}~$5yTu651qZGQ_y{Va=+kt^iqwQR&DTb<eYjl5{~S0x*j%}98Pf9
z+tPmjaI9#k^Gz&t<Go~ei{H<y#rQ5d<~Kw2uqMA!grg&$W%R&)Tyr;I;2-SZ=>B(~
zW$H<<f5DI9tb*@uje~#qwW7X82d&Bf5C*Q9Sh0S?+2p=0HCqB3K6Vt^V(5eCVq4FR
zuC5zbV!O13$HUTMP9UZFi^Pdh)F~bY5n!BqQeEVo2QslT{Wrfkd9?&=;sC4v3ot@R
z^?Zl@7<fqm0r<VNcTysi&N8wwF64L@BO-3<I$+L;@qLTeoMW_#rlsFr*we#wN%3u<
z7g_e;@wKnxCXT<y(MxsOk%?8)!8uN|R%XyrdNRD4U!JM#6!o@!+Y{vhr|Jx6?Dl)e
zAmnA(Tbm^db>3K%*uCRAs(*c$!$Nte(Q(2b;_{i^?SZ~Iz{sXv6m|dPiPvfE%AKjT
zhNdVAeycA>SCJ)}&fbfL9db3JUXEbDZxt#Z{!_Ne&8uVKjl%&$p2D`j$BK$irASMM
z?h>(GoXrifY!u|KEltvH@Us$u4O3arLWxIK60}12!HQ~G51C^$p)d%2mmb*J=Ug(B
zXkg_bC-ft^imCyC{qy4R#O)UO?LO{sCvs4OYy0d9GEV7<p7)p#(W<nyJHTO&C*`&G
z^l=lB$lEm-@p((sNqI|c9b!BugC=XGf*<=7x&2+!8PQ-DbvXz@CT^VEDRswQ%#_r`
z<N#;0S>!I)r_3auD~Z-%1?HCPybY@1m+)te!gCpl>}1b+cf2g>zzQhDo5PeTKuRuY
zw0Ep<bzcMayoU@%ocEShAk_CS--6dCCCH&XEBV#+bd^m1X@OugFkvIL|Ax=0`zUj{
z(R8N<c%k5BDC38#dI@~Z3ol=6d*T0n6^L+1xXt`cH(rW)7?Dw+b|nB_FgVS+UclsD
z7lR40J9GC-m5_umD<rbGEoxf+{6++!5&MCm{Ozj%j6lr}-En+};r%~_&dlWPDh@o+
z1?Mf(o16fHhB*WDN>4Zb`^NWnOP$gWHiuocRy7zOzAbp1J$2Sx_)i+<cuT^JX!GIg
zC<{&St&V3;peWn+CS6<Xl};xQalVd5u!7pfJfE7d!*`R`c2fN!t}f`Xa&ro88f%*J
zhSv5j7(WQRxeA+S?H+2EWhTNi1x1Is_&;cRr~CQ(wO?+%%G7Z*XBuL~eYXUeTm;C)
z|IRc_QMO?SRGQ(kSAM`moj8=QYphH|u#QDhS6L|<(kT*R_KMzz7_aj1_wND%4;h<p
ztb%o3H?6w{Wa5n?UB`?q5<>Qojkf=(RyJ};+c8PYOlZ=$_7hVLp5E_iP;FiC_oLD^
z`{^ZZ4=(Q!5#NbCXtn3c{+La7QJ-skek-dt9c!xG!#r(C1(>al*q7s>>z-;Ld~~7+
zA|A7wc0nKG@4h$5yb1&($DZ8X016kc$;;T;#N=|SU!%eojbEo?d4?ri$<+UpnDXs(
z*XpE^6#rrkZym&Qv2f`eFHUdm#M>3w&bXsFPHG7#*v$rW$G>K}GYWET?emP>?Gg>i
zLeRQXeIxIE?Th{poO2@;>)Yd6hEcBFS#UA*hdHofE<0Q(l+}@U{f5xXzdKo6s{8#Y
zH$7x=(@gEXTw7LeXp-x0Dp`GHtzg$+zA#`2C!me%PeiTLY2Gk}-eW(JfXqaXo*x5p
zD14bBS{)l#jJXVOkUiND)9Rd4V$l%gi!;Gb?h4xYOMZS&(&j}ELHF1fuZf)3D``gQ
zr~&x#$B6E#zRYU-Gf4ziiN1RZh{+aK(`hNF;OS%&zcHU<L~)6GO#q9+(GZb@xD2(Z
zJ{>%#ZWG|k%Hnd^nijlWd-WXvV(!T>!=Wrm+s$c?R8g)}en98=XyG;ZY6caEynK>~
z+32c?*6!CnFpfOB@%7rCa&J2y6I`7CcaGTzz^X?@P+Ij$?Rnj<(M=|T^iN;+PH?DD
z?Q!e0M`bgtSLbm2{A(Gjkeug2H_N;EIU~CHtKJ?DmCP(DHTs1zqA$lg7WOxV!P7hQ
z9wP54TnS8G(5;I3Y`XL%S0Zh?mVon~`<@mD6`zbKi@~gzkn?JP)dzJ~o~R=r7K<F*
zSNsZCmuoK}up-gv#r7g&<c)jMbxEq8=J=q&xrS&;qI3tJ%IlMw!{09zh~X<gV$7!M
zdYV<@CH-^P+|Eh_?5{U`qV68Pv+Rf}d9nan;a~0h*D$0{lYa)kU=n~+yiAR#^<Z<I
zS#(%C)Q<s}f_suk5e9=*_me_ilqht!4ldNT)bp&u?(p9s+klg>r7gFP3is78e+T=e
z!xuTBpCiBxNgJh>TNy1m6PwRjxYJOsh`6K?d>bkj&1M!EivE`4#TS*vfVhIo{RmOe
z;|6S#g?M}KhfHBwO0*_S@3t-t(CU(|0}9vk1xaxf6V)Xos<uZAwb$J(DVQl#JVbVy
zGD4!}?`#7R&G-d(1Q6I#N<17F077@=jXd_?&Yex;3`S-VA~%!gh>cx~_K)yY`_f%h
zix`B{+@i6*d0Jc5Y19W=ZV(KKrYBhiv>(-u6WbNcLBLbZ-9g;*X_D&BY|i%rA%&~+
zpDYh`ZDp`?mT%E3BjYV_Zx618BY#HJy&5=6M=4$GRhEIJr}Ez7JjvQ{LA`rPNH2R%
z*WG<Tsd^l3k{zUaNhZyyrp)_y9fZGN^+$APpyc+Q3Gvft?t%PK`sL&mBjTZQ2i2o{
zA&;{wkBR!`g_i1+B4XRsvjXVJ*=F=lAC4wHt8xMRMt?anb%$uj!8n82!_JYrmF|3F
zodcgLXa<<B<h<6xy&^jh*?ib`!~rQ`c>(5TmDHFo1Kx0c!&V^Pn6T+NtrPa=75!5B
z%()y&x%<75`$KuKLrZNL%p*PK$42M5UHX|U%&E##72m-=!r4PrxdPk<f9XJ0n0UD5
zrI7*X_DnGpF*TUx;V98wN5a=v>jxu!(YbHi-SkA4q~7P7?SDla#`B|mkjWFcvFpQM
zjgNyi$fd-}^!wq@ZI?IEl81LLz9CurNV`v80JnCHzGB}Utt7ziKk%l@d8WKp0U_SS
zK<t=5QaFo%kQEhX^u_PT2m5YkYiO^X12OB!{jR>1de_I_SakYJeZX5T3S%idT0C9|
zZ&vBkE&ffM*l{gMT{XP5&ncL`n66HJ#6un3w#N8|N4~8EW@koTmOk-=lsgiw?o(Ig
zVr8vQ+%kAu>$v9j@l7nV0xs~+X^#F{sE0=;MOf&pCK@NGJ+#4a^MTD-u-x%ZKl7iV
zS}TE1`ZQq-U#3YKgg<rTDv;8KMbXi}a&x;xeJ@&+EU;O*L+)q^5=i<OW)}Q>+}KB_
zx~R-@|E51gu-q3MiiF3zm8Ok*;J&jthj$J$;d?#bL5vCV7AqmDi)Xbos}6rZu$Mb+
zGdZ{jBz~$t`g^lzwER#7O$Zfoy{Ee+CUVz%Yn>a4Zz_7&&O0fCeyRa3h2>B_x-QRm
z_YY9&^n&NYRkt&Mv%`5T{Kom1&dcwO%{_3pkGA^OUAru!arvj#9!LTn3d3)w>hcjH
zKHaxbEwe*oYeFqcTiY{$^=wdAaK8z~JyS)mtlWGbdF>B%_x5@y?-po#8Ic6>9fV#)
z({46tCT87ul0TtUJNZ~Pf-dFN%*Wn-N_ptrEG7nxNO*4i`A4oxog5D%Xvn7uW7%t?
zG4r$O+g`(mOGJLIZ$VZIV%chY4zMJl%w9b<s#@twrT?kS-Hv1|??=098GNH}OK8Gv
z@p*<NPAdkuOm#~CP5%fsX`Md5{oxzJ7-FvH^ir8V62w`)NWZ5#yRhxmXd_a=2_MI7
zw5_Zz+Ql^eYQF<^n%k-@5I_NfbEFX~=;_ra*d<RaCQK*!c4ilB3+W0(>g^P*^V3qE
zl#>N!O0@|OtzAC&xY>*){w}wTJUzm8=(UMOa8I`{#n*D#0;U`0!wL2s5sOk<!m8po
zPGVftn2fE3yH&tobpU{@#_VA7!CX$PV|4O*+Rq%})%Y7whY&iUukV=9j0IT-FALhf
zj8!qVWK>>GFdfe3+pILsm5YVnV|!$RCgfX;769iZw)P%7*8IJ6$2d;k39SUcvtJYN
ziyKECs-L**`b>NKwDS<E8>hYP6-`7i4;?eU&zzbPg4xwyJXPEfxrNjX*i45M4$afa
z=QN300`y_!!#Xoa-9XCU1DE&ByjY8egrs`9A-U=Gc6Xz%v-F^HW4<Q1c6YB)>I1qD
zLYP$#t}S(8;WrI~U7-gQ)M2#HfdM+9K#gk9f%rvF;li2c-nv^rp!vLy;EuDY(%*3l
z-VvvvxU^UUqT9~%vdn{?OKKVw{juALO~s@k`Q|k$NXA~qBo{kkl7F3F<+2hf&1B|f
z;o>}hPDsoVplNhf4A;9Iz@52u_^jpktdFJ-bkvjY&VAnCU(dF6H2)pj>AaQ!Fq`30
zzB3Evof)6-ghN^*F!rQ``_Y)uF4zco*<gQ!mwErQZ1tT!8E~ARy1?Y1UMsZ=uPN^}
zw^L3Z`#60KE>`xUctSfU!6~sYb--fBat~{1G3nro96i^VY}uWbY-In|YY;?~6*R<3
z={S&i%~@o>adVj_ZuIwN@DnlaJFVfo{6<TstWBrz6@UWDrQ9fT@X(ti-*qM}aIiG1
z(Ak~yxfe^CVyWl4y!2$<J$F*5R^i6;^!|Fe7)@oTWIrvSYGe|pfOKQpuPSB7S(c59
zdt|q0Yyx^f$6I}Z){MCDUTF|EQW$V;NS2nG!;nx}T}?zELVzpdbJ<Rfg^**<-={Tw
z$diq##!UAVj@cA~NRA^HMXMVr{4`CPqY1C6I#%Hd&PMgpi!<HR_L>HV|5T6G^KoEN
zN0RY0js%pEL%TEBO|SZZ<5<u`OqJxh`w5^Q9YpIwl5*A9FWejTt8yILE`QSUsT%~S
zvlB5fqj^e&?<1*_cld(5&24->K+;;9HR4{*B1@B$2C>-)sCMRSn9(ZtTh&28Gy}YE
z*U^!Dp$2<8;dUr^T!InNd|xApp+HRRbawQ#UF_>qxNWqfx+*nFcm9JNJ<smR;lkm)
z!Z_Y-9k2T9$kVHP;5aMS3&;GgBjt=fy}I45mIuN~I~pmY19E7a`&z?*f2;?U7?cpO
zlX%TEspGLV<XwgSdw8uS^%>!@<h4J5E;6BSXw#C?i?j!5_-J07czOF)R|$09kV8Is
zx3bVmzAac{S$~F=!F~C;|2O7xTwEnmp}1T1-E)lLDEhbO-fP6|+iags3%0+{crLPK
zikq?3^lf;FN;lhNNlP1xxCsYj@HLo-qCUYvR4Hi>*Z-OQJ}nflXYLxIKleYGu9cR4
z@L6J56m!8m%n<p;amoGFv%z6i>*>+OR-O<Cs09i@oi-(qV6FEqNIoRaB~yUuSmZ8Q
z+@Z=~oJRfnn)l80K%k3Ynl)OoF$)NPDl?|6*T0F>g$pTZ04rvakh*v-Y~dLu7n)Rg
z+@f)J61bXcl>6x9p0Ia|OWh7(%e>zX$i|jn^44jAN)*MBU%tk3+<)jTmGEFvj(&Y^
zSUv4iy|VvL+2I3-c=UJ*I*=}WX*EJ8wjP_lPV+sgYTXagDeW*FzyEJyX@h7au*}0H
zy)<<gF?RaDU^UfcJ%Q=2{we?jC>G}u7nDSMr9)B4n*9KYZ~_nmJ_2c1Cj(f+Qnrrm
z!(E5!D+E_<BXrz*fH90L*a0{!=Oi!|O}CeMV~g93<AGSyTdSP+#;jYd^*Kmn<TSWZ
zWJ(D4T^#IEbxI^@c<$C19zlEm3-q-|gc_cWSJ5|dE{=wyfO^7a88}a><k-V#a5eH%
zKDHJAzZpw_<qB@`*~C$!n!WF00NiTlZt3s57?noO)mvXk-MA>lP2kKt4(i?-6d36u
zEJ+)dk;^{^uv#TQ=X)pru5T0gSAr~*hIp8jg|{{8)_aTpi*q7i;Z&N{%5Ban%A*%;
zQW3UH`=VFJybOh)6fVb5!8ZA3jRdn^)w%%2A+oQq-UzthTJ6zT$#ris_wuKcbjL|*
zZOV}H-zvM?m=TvcwIn$BhrFuf>jNlYz{mVAvXP0C9E<SUKUBQk%rUxq%TNk@;+x6S
zS4061%o8<WT;!=P-!B|^=(za|2td!Clb2Gf>E<<|5Y5{#fD5_48T0lQ^<JE24n%_w
zRlfIgwe@b*ifS5|CE!eHVIeaf%wWAwS3T958mqEqi?BTe!fIt^8Y3o(^810Dr}BRB
zaajw4ZQsA&-LCrr^rx~u-&0rzKxfFBb*#*JG?$O~bxZeVUNk+-kfkv!XiN^~P|u6Q
zMFl&;4C&iNDw8YX3#RS$V%|!r>j?A~9dyc;M3vz_x=ov{;~nO4Cn*x#QblBRmJoOj
z+ke8ylqUPb0Q-*di)w)2@}7k*O6vO4tIoW}NzU~jjnAZ{qpoY97i?34dHT&o)a}>y
z)c|=q|Kc!BSLRN6X|fOeFBcan%ts^UuSPlMMF9T;86PZSCOf0$HBo-&i(*(rknFo6
zo2<TnOW$xRsGU@GdUOEB=t{5a#Kdyv6mKLbPxU)(ABruw^_#fXqN{+}`X5<fNcoop
zffarL7-D!E+(n?=C0oo7TR&-S#RHrs#Jq1l=h9)(n(pGz`8B|pgC*9}{IvZ=m(~zY
z0V=ttq+&cj(OK-^vo?tWj_31)iaghQthY%%V%ZMf5`k4xTI($UT~ASAwX#>R%hu{5
zQ6N>9&|YqPFuBQueKmg8nL#9D+@g?K@CW4qRY*BLN~Cjpkhor8-+w7#4-;odpWWaf
zrJ$Ri=@T-U>+5e8blmDTXt+Rx7(3XXjr&E2+ijiB_S85ig91dakFzT*5{QX?nhou(
zRol%9&$$-mGvv>^H;s(v1;*jCwJS}#%}JmiA->efzx&-x47yJK-bCY{hXKP5gX<n0
zM8SGD6a$~=w8F*};xNVUW~-c)z+b$oR4}%G(9_2VLsC3~o25KBsUCXzU5I+-v`tmT
z&y-B39RPLhr|_ZU=j98{m1(%|rs1eHC^%Zlq4ab{LxuYwP0f}E@@#i1&kVh?h-<})
zsJ8)KN=CDV6K>-dyaHq<b*I)Xk_9r?U1Jn2IG8NeNwkV|nOE3b?nzLKAX7w*w=D+#
zp29W`1VxUjF;@I+=DjEQzT#DEXt9!G_fPqy0>{6WYEHO!-@%fPvcdP5taaaQd=N2j
zMCcC+PRhWp%w$TZYvy$JjuIE#JTSXm(3g+>baGd&6Dk)+2{BZ7RP7K?E3{v;KC`~j
zdqaekeNNrN3W}NH0Fn(CVmUVT4~*sq&jRcKpuao2>EtzWHnz-d8Ki#U>2!k(ftE(^
z?&LQVyyw04??z)5(F;g#)tnw3w%kpAII@5bXeijWc%U28p>fVH-lje`t|!S9WK*ox
zNqmM$b8+++O|Ru}E7{t?;q;<0nksSbv}?DDuZs5<NrL3~0ydE7w8;eoXP$LA7k!zk
zKPa?b*@9V1s7~Mtha5MvEo$yK&6GEZ#7tJXa%C!LN2{$a?r$H|HL-<CRrk6e*UiVN
zw{%|!{p+?#I^6Qx7}kM8eO~l&4>%?BF*Rn)0DfzR{n}mCrF+j!UUD=g&xE%aPn&*b
z+L)2JP(8DIv)JZ4Ny}Ts$pVx*5;stWnb(j{S(W1MbYs!N<gl2@Iumxh*O#1jUgi5G
zt{xvD#Iniv3y;~}#-nj>azE2ptbfh9vh^oM=9dO>Tt*Op8Exq5#UyQPg^Q}hd`j!1
zS!Zu2S;rZvTm7JXUjCyT%KImatU_M5wv>l46{FcgoFy&Hpwm(eUv(1Y3<aT!yVVsp
zuKh}K^&>u(bBR5fT?2_$d*7fem$;VJ|C)U4m3ziBcZ}0UY;czsr$ART-*;Ow7?C>b
z%;wizt}2pZezo;$>p^bYw_$ch%IQMlEFYy1^}Kn^-t_v1WkwLH)VQ;M8EC!98;hiY
zKkJZzDTQHv$KP}&6nfvZl!qd;^*&VOeSvM7qxfq#=D{udPFW!)v>(&m(_-Q`uAfgT
zz(tMl?dUcYJ!#obG|?OQHmjk`B59TgJ;Z9!hsC2xi1H<;;^5)k%EWv1fu2n$rh48n
z4ydE7q-CEAIQ0&<C*%ctE)p$&qC_F5zaQTEeH@Ez4oU*)LU2JBcyhZn@*6jv`x9oR
z;x?F+(aH}`Q>M-31)3X@%P$(lRsbc2kzMvsJvJX~4-imd(VR+YV|~#<mTEMEJ2eU?
zOScrvt!g(%mGJR$Zvi^wPfrds8fAT}c_7-Q?n%S_@UR%%s&)WSi&s;!4&9^mrPeGX
z5pzyH+7Zb0gE1wnJ4b7c50W|)b)34Bd)LX$*MB>O2hqG=KG;v=c63epfI``Iv(D>&
zihqkXQ|yPWb}M@x<eJgM_b~Ijz&tWOl{%>IImiAh2E@fVp&*LLWiM<mOVMSw*ka1Z
z5aKs{^c0>hTnj&FYF-2Qig@Jh!0~hcH)UbZ0GmCe<g4QJ57QxIwwcayL7jHp9NpB}
zW2EK4$ic<@gEP&6aCI(BX=0fsv>EdH{OlCK9v+861@~T6j62TVx2l;YsG9qqx}Lg9
z9|ZTVCw){il@BsaVqgGLJr>uI1I-NY;nvZ~gVG~F8#-AoE6Ay3dcL=zI%>~HjT92z
zv?jsxxT%tm+g@km&CkC?HYFl$n%nr_bVw*9I2$IiqV3}jsW)i|p8b6#@)#d^@N@oy
z#7$^k<9A%Id@4jQnoMXy*Q=D%B_Y|@_w1^1w@}kq)QbwhD;L&dt3k}t{%`Ac8Ki))
ztoS}YAH&|MYiO&lk9d*YZBKr`J^Z^7vj;IP|Bu-812^(DygPBXu8-<IG@ykme=>tN
zxeZh298Ef5wij38iCL2Avui$*R_=yDSpW|#N^1pRrlYiHi$fo?L`8?w3J8(exYw<8
zo%o_5aq_{9_hF=hzFy?Fgha8yi}!aXzQ<^b<5TdtRK1;pcF%$uICqE`;n7C5VVhy%
zkd2$hgE3+dAyAf?<+XW5@EEI`y!E8EZZ7e@BwKiZ<|mLz*wP!4yLw(V{CSLb5ptx8
zQ?6d=0yHHz9xu04E4~YGdxnG8oJIQ9#{K{V)inL5VMzWnXMhP-x3fkz&M77dyuiky
zz2GLQ)N%UN9fB?5^(lEw-Ay=2eGo}Pd`{UVSu@Y%c=L<qK=O6y&Ze-(r#eR+RcZ%^
z2=5pA9t6WLlHd-+uP@H_T$SqTmEsp!LGp2n&__^KuSR`fWwi(du}XzqFsEC_`?CR8
zYqak9ixa0~x@+Hn!}o9>Jb>3fk{Y7s-_(Z!m*=`<UeEbc5i%=9m&$a-j2Vdhg*}<B
zJ6_+!sj!XGq*n4>p&CAZ4v1zl;A8jwdI^Kz1iaBqrFXp2Viik!fS<@eZXFZEO%(_a
zEW82po?fu8M;Cw2oE;klxc#>te6hSt>tJiAKtNa02yljy$%nArO|WPJS`b!m5*=%9
zJg1F1*Nd=LgUvme*a(=ayQTCfk=-fI7gn~Jzb@n(-N_f8<Z{wq<t}p?0z}{h;8+Mp
zR&1U%W@c%X+yCU>Zi{M0NCM7yskjeQPP0gSYd-tvSlNDyF}=B;&1xKOFpxv>i-w-j
z#g71A1V|SMIALApTQi;{w_I(dZuUh8+)_AA)|SWpK3bY`v$?DU)U~mgn=V90s(}Is
z6BI@?Ow_Z6HT~KZTE8a5@(-03QA_In<%FMeF!$H|dLw0{altG1GU8-P@f)S}{5248
z-pR)V$h)^NBY?V;Lkrbt^e1{vc$9D=&O+{~JT*s_C~+{q&K`6RXAjO~`AK+ldn2hw
zjk;$KzROM{n*@>^DF3%~c{`>~fG@Y|H~ZR<(=GZ{QLX;9Y!ii=Z}!w_3uYiSF<p;&
z9A1;S1$xBv#kPRm;{nnwH8-kI6h<)p_fzk^q|A(is06+7;@s#TVu$TB34qI3zm5wz
z_hjO>7aNU&z-Jrw+3XYQT?|ggex0U_`s+aJnj;%*WHh1fYSew1J$pYD#OTpFJ{LEm
zR%s`$pCNpR#DIK_>+Ic0Z3iLOCJ%Qev+K?@F<oCzrTS1LApW;szkmaz6!fP_k?94y
zo~u9pNa#<kCr9VS(c;R&KMdcfDj*+jiA4i0<`7rJxYxoyw$4d>569gs@EjfEOjLzU
z_@$m%x=Gt~znP4Sa1J$@LAPW#aI2MkFfYc<2@yMY;{&}s4q(lgEFi1M+sz*sgld&9
z9SAg4DwzEqS1nnbqsTky^*#`<u2q20+~uzwe}ZjFONd!MK?<jfza9p-*8x&)T;uDm
z^Zgodz;2hW+UF#cUK}1q=FZ+_-tBZ9Ah6T?^9}~%T|n-@#D7`UK&nK6MB}7Z_;Q<b
zP$*_~dr5kkZKRX7(55P2X%;-pow$xBkvpF5$|%>4fSby2ufwkg7qx(<j%TeunSxFL
z^u2KTuFFbh>Sv4AkB3Dk;H-?Awd!K|{^sGbnSWmBe<`HUsB{xF#IJty^`SUzw$3eB
zB<g&9B4>F#mzp24DeLpx<f<QzCN^;tq@d<W%4}LIzg~$LMdSGG=GkRl{!oP4?Q>g-
z8zDuMPST=#8iv3g3QO+XcDO@U>h=I7Uc}Ymf@qF1+^H)5B3k0$A#zbQ7eB?*;Fw--
zH*^TVud{+3@#!au&~b&(+tN+t!RN*clO_QN^C|oj+IwI2dfW-x)qNcKA$BVJ!JAPx
zY%odV`7+WrJ2C!SS1Me^nKQ4D3w@*bd-}QIjQgjQ80k}T0*ycbKiL<>VsD%>cfL+-
zKt^+~%SG=BezTRE9axZOHeifa$pL_;S#^gE%OSA<*+c>$MP-&}!et{(tUnRfJ)gtP
zDsDWao+OaF9^K5a1YeP9m2mfT1ykQjG=#%$SNN3sErkUQF6b&RW=F{ay{0;yi<@o{
zw}pIt;p(;<J*?0ZUpt?y2kz;vL#h8(j})Mlbx(#O03oZ8(18c@Ow<%{oe#yWLcv@2
zspi`PV%`SLLpoU{NC1s`wMq)eWKnp2m7%`$^32qwRyrz6g7TdQsRD^3d2q4XfN}>g
zDpb(Jr(P1y0dmh}>BGf;biJMvJ6sqZK%41wr~4L(ZGp_;J)fS$%A>e?>KSC`Nap9#
zv#B6NvPUJ5$b$=+b{^$-^PQnm))NOv@{9SkysrVYIO0{!)Nw@k)<oSiloKO99`BTX
zMO^kH)2kX0#+D0uFfY{HXa8ek#)?U1<a>&X=WUXFSq^H$r@ku(2bK?j7BZO6ZqZvh
zUJ}}wZ#MjSYW4pKjlR#`P1LO~wMiCvus;T@s+Fz*p-4ZQLjq|*kKBB3iGf!^sPc+D
z^5&|q#}|MGNfSzWaQpD1-OE&Oan}QLZ53NfpFD4tI<&_OBdL6HM@6f{9Tz5Bg_DcZ
zjVMk;YI-he#}6;1UfrpnVwO~i*LNVqzZtqOCfxrlPORL{Xe9Y?RB;3VxsA>FQGmrc
z-)e9qNEdxQ?7_0(GJNz4IjLyeTF=<mYGa_+&=d@WxT<@UKPAGq7|;rD93rp7o6ZkT
zp96_np<436!cJe4t-hlpiqoIDaXCyVczx0Ok){K!EGN#=QU~awuKuS_2#DS^vUTJ#
zNMX^bA1QWeB4`e;fB^FBg|Nev0>HuTSba|!1=39c1FNpaT5_If2RU}6M`Wzu60$<@
zT!6qW5jSD=;j(G{&xp4$u0wlBwXIL22Nb5;REIeA$23;uJaZWY62c&OP9K$`3u4={
z4*C-vASZi??)G|<3%es;bW;BRCGl@B0S<qAsNm&pW225nAWzbU^VtziKJmb-^_3Re
z$VTz{ybM;Mx|-bl;Ha_zT50@i0w7Sk?HY1hoF9E!+Kxrr&ByqjhF?7N`-g9PLNoKm
zZ2P#%+Aisr)L%FeFU+m%#2joRSls9)fB)_-%IT)zt6fV+_<XE1P!{%(HA*}TtFrZp
z_ki{Sy2_`ZgzQZspALWXUMe6v!ISD}3MFEP5-?ZWt!hh+z;G;oXOHqI!fOJw0x*@g
z&qQ?@!rWXdQfMa!p+BE#3*;UFA(N9kqF^*2Y4DGLnU%Wk3ovaRzPl{K-z)xamYuNp
zUXC~%epE@E6(lVtwxHNGMp0ke!<9BqY8%cpGm)rx&A$Sqrzpvxv5;d@kvaouhmHO+
z{vDOA2p~zZy0sO|m^r<ZlPQ#T;vCw#Uy{Rqkj;2?s-S9k;(VhLe@ac2c4+xUI<QlO
z2n$W-FP|(TdaW6#_AKVxZ2WQdPMgM+DsH;!t>BjHL{Au#=xAt2^3YV+(9M+RApA*t
zP)wtL-{!S@W=%-8sRHHni^GOYfnKgOAmV=@ZutcJ^66_>Ytt>J>41;)+zB-28Hiyl
zSOO2j*K;b2+eGNH*N#&cQ5qP2Mma=LJS6`HzW3(#P;q7cgy<uZhryU;&k3UxNKbpl
z8S~(Fc_4~HDT8<wUaK*z_J%Vy8^5Sy+{w4?gNE&uht_+z-N0#4PL^z%3}er-I>!$}
zc<k(0(*_G8zWyHhF=%r13zTg2Ga<}%V|gqczUSH?MqH-D{2|8QU5H;b`xMjk-=_F)
zl`7Jh^5Nr3pL8s`{!G<RvTx$nIJZxjndVr=P-)O`#Z(R)Vc(JNeEEv0Ive&3Mbu3g
zKFn2b2HrU%LC!(4JqQ2_@n=y@0U04Extj!mdQ+988b=z(cu<&P0kBio|0cS)O_C}-
z=b-wsH!3t)Mtd8<M17yxj2WuMfWV4R+ouoZp`s-y!=*$DW?MPz$|ot>Z371ArX#$)
z%3qhCle^A1I9871zIzgn>2=fdhemg`><Z4aWb?^X?q+(#*klXFYj{lh-19-)kC%VQ
zMQ@%I**oVh2xXpm;WThc4M0(0S4E<@2=vUQ!x%!lmX9Sn0SxzBGR1>*=b4~);a^zZ
z934EsX~s4FTH0o2RXuB+w{GAvd-U3Y+~|Neo6{Ky5BK6vBGom>p6gm9BHe6;cFONy
zyJFQ=DAyR3Ps&|J3mNo|<oq!~mQxMu`}c0AgBA~0W^vzMkG1>57k=88b!wx(eADrm
zc?$kdO>@!aIF*9i%PZ}%fj$l<Y63y82{c$fcsTVox*11r_pFZ*;hBp2n=i$1KYSN`
zsxaktS#XD!-rwQ8v*lRf|B$h-!R=?>Y;hNn8)`xNZ|Q9PNdhR0fLF@X<;i0@fD>ca
z`fY0z*4sWqIs}zjXE6m3BMz~@ksIqfH+<_<LJA_;9s>wXAgAwrG#t>zFnyE+zTf&a
z?FmLmNTIS{^ANN!k2`ibW1sKyUKpuP1y8Io0imp)(AgR6@l78AAI}BI#i^G|yA1ov
zH00(BR)JWqynZZ+7-*SCxyhmnoo&+A;sMnGOs5AAsP7R9(Et7VTsTov<f44BEgYBx
zNzCxdmvj!^{ONILU|68eC0dH#XK#1FlktDW5h-S5<@XxZiz@%CE3mjCa^Z0QU)jZT
zVY=+<yTz|<fBrYh8I;g>N$ny{^j+&04hDwU^1XHq5epf5CXjpUx5o$3e(iei%zHEZ
zj;Dxb2!B=rRSI)OUG%Nl6^jd<$!<TZfCu{FYmTod)ok~gF+s)TmcZoRe!1eYNZ{EY
zGFdV8GKrxeAyFbryiL}=NNal#Ac|QI4J-+Z0IDIHiH~8g<1}{LcY(qaF=+|x$HEX3
zRjSd0A>>1rDoC^?ui>#>8fNloicw}Y?1Y{b#KV38)O+0b-buY4qWiGnu%`$zZ2r_q
z33MgCsQ;9B&pp{*$B5Q%<3|=&NCMRRXwF?ry=%Dhxj#-5BMnT^%I@H>#sOye>HPk}
zi$TN$oUe3df(|nEG%R}sESdE1spIHEBWZ=noeBDKs=@?Nq|e}Yz=(bjZuk+}Dkb&f
zdOZNBqt&g!?dGkE0yZG)tjQ)!MB(VXiu(38$9IP$u!<@Q9XOS<7-FY8g*+J~Y8LZ1
zBY&g9J|pk~*F!Ro=#{R2yW<-eeu8K~p&g#Bju=p^#1y`vir&dMzA+jAkRd?vU8LW~
zGG?F$!KTO@8O-sazsdFX>nOM6UWPR<umI!c?;%^bhT?6nXr&e?$pSyl=9o?uA%G=&
zd#^b-<QDI}2g?33%ju6Ja644uI%a~cC4g@B%Ah4P_ugm2{Pv;&A1#Ff0yvh7X@Pnb
z@e&vxtCrpaOy#PwSq)^7q!XlJWlBOq;;`7kbYb~fukS2l)E?FA=J3(yvE*f2kfD<?
zMZ({X#qYnG`G6{*NX9~hSI!Q&I44+TR6yx9ce4nY)ZKvsD6)#GT;~?(<lEWY(6*g2
zcMZg@K}>r~frQq;jUfSX9#Oye^5l)j@QKO;L;jEsXX%5`GO@MGW6hJb<FF9FRkjIS
z0R-ye3cxD<zdFu59?G_h<M$Zr*cps1Viei8vSi7UsgxzlR8tKig^-Yl8QG1sES0rK
zvSvvck8SL-_9VhX$kX(CjI2#)jQ1M<y??xa%^%l&?sLv{o$K7!eShZn9si-Apo3rC
zP1)mJ@Biex#Ar={AhoJLQA35nd_c_vZ5~=Xe{Asjz&lOq{Rhq^OqY4MEEljb*Yj8r
z<Y}Z8rsTK(5{3vI<FDNBWBOvi2GZu8Njbx?DXm^!1$B&f8~y`t_hB6r#HC54wxs22
z9WzNeT%C<SLv=NGHR2ix1OX)2vLdT)P0t4E_9-Z1`?iL$YS&Sfa6ymAmmXj`>&^8i
zaJFP$IqH)+-E^a<|NKyd2ujPH4Lx%(U5CC@;I!yh9X;nsmL}C(RIcolmdS4U0W4w6
zgc$}F>QlI|N8~wgnpy5G5x5D;WR)V}ye~YcWK_VC?pb%tBnQ^q4Czrt*yxB5o*$=*
zB(=e89Z}X+=-~)<s*A0!DT^^7^WGMM#H&q&mNl#^*?@#-KfHmns!8xUIcp&4JJ)nc
z+a~}Q0kU!S@M;Scu(OId1|MzH^1H)gSW0!YGY{cyoVCv?M|f!`7*?s3^f}bAa6Nrf
z(@W6b0kf+JgUGjng%8gyQ`lpjfW%S_(SkTh>QcmE>Dh6=#*|<Xmh2^EaE^E~))|zh
z8&{k}omx*|!O^~!ks)nA#JN9N{wE5Ef%~v+<YnQEdw5s(yySo>UXNd#bk0(?mrH94
zT22e~d^8=>8<`VK0sTc67RRI6_AJQEdCzzPSMwn$p`^?C%fXv;`~B)``p5_|>_CS~
zpJS@WU>yn4Mfvii|2;U@ChIGkYlPRe=O~pb>vXgZFG>O(wj!gR6{h`OINzMZeQT@T
z@mZ(r!q1C^*XlKFKRv<!p@fcztdgb#ZWuHF8t%$PO;~A9^yu?L2Jgh$oB5-x^D9>P
zyV>yt!DVGl{M*kg)PQ9B@_KW4U<X+$)*_^5>2YZOO`Jk-7WQ}1!Qg8kH7~b=An<uU
z&hyCHlTI~-Op6IqJLf5WM9jU?)zHmm@E6|V$yU6bJsAXqQ+z!e-y}AUh}|0HrW{p-
zDPG|@QJDJ8H1>*BNzE#A*z;e8l8Fe!>TeC7huNF04oe6xDaM88UVMik3i-#MJF9yU
z;4U8O=T?e65Y8yMhIjpXSBv#r5YLR{q4bH5M~M3Db8nBcD=4&1kOBof>64*ii1mS`
zoH?Urnpahj#)q;aLc&Y4)n?g-z?m*^X6PYDIWNQA^2!Sg9PS#;y%?Fsb{J^Ipds0I
zk?X#3F5I9C{)ZaH<1?W5>g<DD<I-df{WsclyRFDg9^nk){4FW>qd_~}A+Z)Z#-Na@
zy);GWA$U3cl!X2}>RoSbZuPcr#XWW@q;}Se#XO8>Ylk%?KS}79n9R!EUI$S$pN35R
zBFEbds4jRj{l^L1kX%NBT%Wb*gbFhAsv=Z)yz#4Fr}~&S^>)$u^GXWvnQ$@?R53I6
zz}UNhS!E1EGr+Ak!b>>gpRi;rniWf$U7SwLf+Bn!#vQPms`KXh`3ZbN=K@#C0nPw!
z4BNh8RYWzY{am~QQ^=hCp+Kv=K}DHxiYY;bxi@8d$0NjOQEtOd`Y#i12=mgVymIzW
z^kP}W2inMsXI*vlv-5iT<q0>)A*7unZwAFx0kMEm*)#roal>!E(wz9r)k<gRm`qz|
z)y4@i7DYg`F~Xjp$<QaYavN9;t0~+X-3VpWB{v1U2ss0so~iTxiLKiAt^E#*FMn^1
z9=4(|B(gl)os|J%;QFN&FFV|Krz#U3k5u-%_qY=<W|xx|q=z)3uCt&RD$<gZdGXRa
zBE5%T)BvN~yvm1#GMv0QH!UlAnZ&SOX9IAPs6PWe1s<_+rPuz(=RKrcfYXoYKjsmo
z+@6=^TP0kI`?ic9b?QpEG3)uufsvTpgvX`0vv3VljEV(=QZ`Etd*UE>5qsB%{C#zb
zBeWgLRDoWY4~~R-mw0GCdY~1IU)J{Gov7}ZK2}w&py}8(r1IA&<4ccE8YQ^!A{^*_
z(*kH&LltB@R>&Wltepp){=uLZ{op~e{C`Q4Bix%i4=;7r2$g92F!>m#h(hl%DU{Vj
z^y|g1>->m#Z+7#RX21nt6!VvrWt8}34m?DS3D9w~_7N|{yHe0JxdvjIGAzB|iqoEr
zT8^?%W}FE2rX7`Zb(m3cmlVX&8~G4AS~ZT3v_uE+N~HOYFM#w)a(F1yt#v<i-e#2)
z8<(k>?dwMAwpXC~aNM|V%4D@}+yYF5Ui0aRS*g-P0659?ar<dTcq7JrMp3+PHto{l
zm#Gz1ao1J5=wz)3G3EF5{_Si-7SV~Zd1iN#1-Km8<R2O$)&%C2QJF_vQ=@d6Xa4^2
z{9IL9|FmU?iJ;F2$7xUBSN^oC+j&Yr&2GU(1Q{TeVB?`4hOWf)hIs=U9Ut1}EVC(W
z!VUNH6h5`<E}s^h`{hz&RnOYV?3-)Nf)xR*w3*sbvrc*4e@k(s>#gZ!ed|Mi%g?Uj
z9s75LQa4s71Ia+7mH1U;--d8pqc%NE=DVQ=>bGexR=<&D+*teX1ub1M?j;)y0j{4*
zkv{=z!a|;`$rH-fYfG6<#P@oj_>uIl`5qK;yn5VuY%?M+G;lWbz0cM4*xPx4e}wi^
zD|;GUw+fkZSne)m)};^B8}8AU{c%q%9m<eLU6Pii*^U#mcOdMQn7Y7#%LMN1cd2G~
zcx#^;e!FtdMiX!^g32`rTZP9()=iqcJB?y3(3?wPcT3lw%E8_uVoC~HFXXqg=unyx
zPyDzdi_9@ee%&!q<yrupl+M&^_BE|_BBmq1>_z2FbzMPnvFK2Rq&jk&KRl4b((wa0
zK7u1~(jyj-iYus3jjOUc`en_yXziRM3NU*xMv4D0gtpCz>21wBjXC8Lt(cIHIuzT+
z+6QLyp7LdQOs4TZ*VBuEacqZQ2t#LV05Jw;<|2=N|L@w<WHC<QE_0+}Ad>%Nxj812
zO!4)~@^@4|xbgFwxHV_CpNP1vf<2Y2Z-S}<toagkN}VDEYfOjzI`XlrZO-(9^Nv0u
z21NHB9n@<I;a-|!1grZ}u7A*efmWEhPzlwq3XIf}y|lA=k$g*W%4<sCYy>>$vB^t5
z{U~{^GxAaDOQJ%-$I~FA?_fU7b!KOhxL=z{lS1x@YQ8S1cQ5V)FJeC=!~ofg{2&Tf
z07D@8rc&k{E=4wV%7didA9*t{*Zu4<(Tl^ft>4s~OxssEh*f%6C0w{K4{n9z+j(*`
za0da{PsKu@xGx!t*wYPq3pg57QPFQlMbq*|W)MoT3h-Pi#E0gs9OcZ0A~`&uxt;hR
zI}fV8iMYsL+yp*U*0*2oE57ojcs`7vC_TRlumtqjwCvT(48=Y6VaQ`BcyD{aN|O{_
zLO2EDyAd={&~4D?KJ^06$K5D;$M6cM1kewR)GA(6HneJAMQ3EGO)lqNWx`2yh<H{(
z>~DS!e#C=5#O|exy7W?Z=mcP7oG%svt5p9ygD@O6*<VQH6(d#EN_vdCvrb>Jz)DCk
z(R{qIj>;)-)6tLhcN4j(9nQEMYhh*d#~H##l$R?*GF}ojp^TF3bEl=~bqhFy?-o*^
z<un}}og(IaITRH?xeZwlX@uJ!IQuyG`-~43w|P~~jH45QnJ(xU#XppAH1v8DeVA1Q
zVbq#}gvEW;NkuVIg9t40ESV-T{lSJ+HghSQYPjF$Nqb`UMTD*9n}I8hfGk3^VA++d
z&en&Zncanyykh)TvLAL6eys0HRi`Ncg_%40yA9`bFg3MKbzpdS8Di~CnY;t<RApFW
zyQAl8xcVh^Bs`)Je9q>(L!#Kt>q9ExmIu*hnr45Zj!kUH;oH(&c+q33sK|cW#w#QO
z;Kk&Zmi>JOFu8K>!@lw*gsm-7!(hiw#>7@Kt7O$0Ya~z4G$99-%oEwZPQ9V{mmg~e
zf?@FFvEm`zD3x}l+M^hVUZ#-Wi56pr`AFPvX{t+06#$<B>qDR+VuqiRLj-#B*x8ZD
z<I$~>IcB<z?r%B|W~q6v+Dk}#q-%1D;s&SL`cO8Y-kF}<I>DfZ#AO0&QFYd4T;53V
zgdPtN$W6SotFB0C*J7l^?6J5c0J#9d->74url|KxCIVeN*d(28(q%ZEC!j!n!E|IM
zN%l6Lb<xKxrB#kwgT{Qc#CaJ+h}p;Red1dRFZMS(ld{Mn=twz$rAa_{sr%T7)N<ix
z;S|tW&+e;f&#<?^==`KjPG)x{tMyA^SQnNIw1f%Be5?IzAD@kj_inX3NwY`Z^A6-Y
z^=fgFx!s<h_@}J&GCShxbmL{V6^-CWR~f=Nd7L??xX44@9DWo<g{j4T_`>(xYbOS7
zV*wmc>7F(A-VL{Vo<<&bJ;4v8qot#xuC1f4b=*n&gpr=Uk)DpKmX?v0R?p{VssAzX
c^S^P+EA0OpG$g@GfdO>J`hwMK3w-o{0WP_VQUCw|

diff --git a/docs_old/logos/shopify.png b/docs_old/logos/shopify.png
deleted file mode 100644
index a4f9cb575fccaccb38534bd5abd64e7cced6d36c..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 8003
zcmZ{JWmFVS)b|2XBEpi=vUG@mNP`NC>;lq_uyi-b(j~EU!;;b^Aq~>q-5{N^bhi)x
z=bZP$^Wi;n&di;-F>&tvW^Rb0yfprE%I5$803QkgD+2&%peI@f2kR-H-9+F$70=8*
zefb0cR7Bz38(==wUmHP`zW@O4%m9FI0040N)a1Jh0DR*H0QU3&0AVBmKyIJ*TS@e(
z1ItiW8VvaNU&?GQhyws<wV~ios;&zM>27JN)5%w@JL+oXV|;uf?5pTF>^SU{QlBup
z#6ngX<@LVEea4hB%Qiht2}DBW_{Pbl!Z-Lor@wMY?g~LCqErG|k|6LZzrU{Jll?v+
ztG0CZ=wK_Uq^ow-P-lCAesy(ozB8%zEy-zA=n9^(BQ%OmFA;zP1c4ER|4U>iWNF_9
zUA6LNX(r5(b4&>hOh}HaqD5#c*;t`kij4G@NM6c5jRlfEBWU%6YGq5_yNn5-t7*Ml
zjIe<TZ8gnq2T2lxz!0$Cy5Eh_Xm$L>i!p%<k5Y5hl=P1pulK6Wf7D($2iEj6H(2*}
zyjw}(w@IOSsfHj7<|>g`S#8JMXf<Q|DI6HkGNmn+StH*3Ss-44;#~oGA!__p<#;?>
zo;PBGKzqQwYEj6pG($`5@YKHWX$njrA4XNN$Iq*e@pny$UAOXaH0Ik><lkmOx`^9M
z{0AQl2arM^o(FS+5sIMLxUqY@mLStxZ=wm&%Jd+-RLE}o9-hZSe^)AYTy&XiLzi&i
zBY}S1*-a!idHCI|ec%780GbXp+WUqwu3!iQ%)6TfA~Vh&@>h<M-<36$K;B@<n;X6^
zoeoEaE+uC=kSZ3DMpQIQrN`xc6Z^)<yP|YJr2j+MWXMXkGYbz0ibqtgH|#KfR_*f6
z4+&7Tw7s$+tt+5aQ8yM#laO#It-tMYGE-=_9U6S!h7QPM3r*@BMS4Cq)g%rx_6Ct)
zq8{z%o{S1%WwHEya6`MRk`_f3hPn`7{*sblee1RD??C(!Rs*B>-v6JOs-f<eR~V-$
z;=t9Qv+bswHR#Tw(!1>_DvFo7|81v9S58$geEgUWTTp80bO~>JUoIiBT7B~BUy0j#
z504Y=m&TzzjEKp*hX`lzza|T_FIl#>w~twS4lMVl-Dj<r>AfTSuU?w{Crb&IoA1!3
z<!~AAWY&tjy{j$rXn9G!<N3#OR;Ml();Z%oiAI|vJNZu_{T-Yp^yahjdg#+CdmOj#
zouF){up?5)voFTK)crHCKC*{M1_*|S>}CmV{N~7Oo$aD0yRAJpiZ}dl!g_r(g$G-T
z^Ooz1+!xENEt2s~J{$RP`y^!rJJQ}eG;7+JdWBJ>sK{=Nv=RQ^+wqQpI1(rLJs8m$
zkJtxU@hYx+-k;$cZF@f%iBTSfB1k**{G$Z>dg9uH)gAi{yH6%vyvqGvnI@zJ%^Tyz
zfU|r@#mDJcRAeL7)tlCvwX>+_!Te9s!VwvTw7UKi0@SYwQ5wU!^ftQf-f*;wu2wG-
z_2N-Nf`49M>mGjxBVs5j_^x&8)*N3_7phoL7>xhBv&Gf(W{RbUmi(n1g75%i{54l)
zRQBMQGhCc2S#*24o8o1o{)fAHcd|AkJH3Sp`f*pkh!)X`Bs@~p+Y{@`IPBo;_k4T2
zd6Y@Zr*pd^V=))dbt+4g(T_*(0T?`cZ0%Pt&Uh^$LUJP<bK_qwUy?ALT?()1NchxB
zfm&3gOO7k~7+N=N^=}#zSg86ATKCPe&q0Ah?GXe)I};imE>-&b)DeZc(tY<K{FDH=
zt{9D=O&4pCeL%Bd(fiZpQ$VL8)vuKq@jWLW{rqUGUFaK2EHAw@Mn7YW@x-B`vf<J*
z>H%=BuMG4G&7)IeEm~S(l`}lO7JCen75x%W)7IMt(yurxu6_%$<WwhTFAYnJ!b`mm
zI{fj$+nj`Jk}N6f;#fjFApKlY?~*6@x=V!i8e;!vvaPP;s(P9{qRQ9FSH8`xZLlo`
zuzXaGE`XL3u=5rFzGVp*qwa1^?8831ojkp9G|3!Y)0%Hm|K7r*6$&czpjnT*GhTC$
zm7q0OJI8pW5@a~I^YDeCf4Fd4T5D5j(~sJDV+k*w_nOY(7!McwC{tQhc^iYEB>oLz
z%#KEAQ#AVu_=#&R5B~sA`ALiEUEQ_mejM6xOGS_F@C*ycu>aSAy9C1Z@K=_rCf$h7
zFRYPv`YV0G$LF3Dnq6W1qt{Nm`i3xl^UU|`&ok|Cmiw}3Vr|S(z>2ew>34?r&=c9!
zV7Ka>AB!bmKBwgyX%u9WC@`fk?ovaM)kd{~-!FF*Txr6@l;it>HapG}xr)4?$E~3h
zMkB`Tz32v$Ktg5iWoa3o{rXiS%)kK@yFX<z=|An!sp(eJe%HqN!)6k*_Yw+@F+V+*
zj7LO*JpB9>-&j$-TL1KI@3o*QA-C8lfAjJ-f_1!NjjK9kVP@$cpZ?+_r3K3rZtyN+
z(n4@zkmr~sYmB_gM?-Ps-?fzFUznQX+X$S}5nF<szrBjhusqoXX}WFjxpyCSX^>jS
z2APqkm*bkKet2+=2Y-)PCKhY=e#!e<Q?t!VAk8A}=XVx&<n!zO4+d<(ltH*rcxqKM
zou7x)Z<4uk9ST;pXK{}j$dQ|bN~sb(nX|#$;V7;t*~J<#NfheTV)8jmxhO&gtP=?G
zL3FHPg=;wRC+xfEM-IL(*7tEuNDaNZsD(&qg|JfO_+;4@D#nv*j2Bf4=!#HFN?E)O
z&7n~g?_55OL0D5=Xw9BQighu=NqaPDWF&3q=l}(1Rf2@3Rd}vw8x{{#?eELLWaT%E
z$J#mvi16a%%95P%cFiKA80A~F^d4Y(@MMl5VM3nQuafOL8@YX>JMy1%zcy&pt&^p7
zo76vp`|lSt)CXAgO$rkO&6!G^r<k}K%^jk4WNjoT7aduQ89_FvNA^+VR3y`q&`n>+
zi+!O)&8{URjO0y@!j|H#>Px9C&B~J-e_8NIXU20;7i5tMA!YF*XkC0nNX34Ld6n)U
z62ZCE6>_809N&$|DKmCmu_LUx7u2<yvG$q#ZkY(fp4)8?o-<CA0n2t?wJdFD${`U{
z+Y*Z|UHzQEwb9S2)CX+=#NIMT1H@he=8MZ}`$a1tmRQjcR^aTzLYrVb28N$4^8AGf
z@_F4CACw;rnhe;Zk5+}RjlN2p8bVw_RMn{h%iOVBkf43x>>|D)dL}IGZuN!W*iU{j
z(s?<#bOhNH0nM41Ay4nNNmt6z>PmUkvjIzu^K#L-5UA0dRymc=KL!<S#P#1J#AC*E
zb&8I#F;=ifdcZP`fr}+rk3k{WDZ%rY{pS({B!4`#7M{;zBK@IgF>k~M#8>H8$<bmY
z@|;>_dDz$ug)lW-v5oA#hzieK%jwWpz6os`>bni>@aps0<M-eR6W5~q0K8xBpO0k}
z@Cf<+nW_#`>1iMzo^*mZRjqv9e)IUOhc%<^)er~f>@EJGh})NX^lNy#!_QPL{|paX
zLt~FRv2U)Sf7HV19I<<yML3S;Z-y3E@@o&BPELN<K!5!`t3SG9!@RSElZ>WZD#;2t
zrw5+kQ$+lGkv#9%6HZd*J?W>Db_Y9(mNi>z)-WVmD?OXPWS3LDsflFVB~A59_1h6k
zI}<fU36r}67SsN<Vu=DEgDPH!;UUHv#Tcf8%fmLtS}9#lY{vz@JL|z8<8CVfaRg9_
zy;dSUANopB$gBV=oM!2gHl+6)jp0_00teS-rE!257n2@nfxqPO%_2lGN_kaWPIMG0
z3h86^FohNswscvH;vvU;M`#Z}0Iuoo^RAm`V)pZI8R@AVky4iQ3h;WSb}K+F1~51m
zcKlJ|%f!Io$ohIvTr_E^X3Ap<slGfHltKJ&mCUpjE}J!GhV=}6?L>z2&sa^&&ZytK
z(Ry0VOCHB|Ej^aL5T)@#^f*PKo|t{I)AF=R9Afg#S@D1z|E7Tc<sEDmA!Cpf#W5#y
zxxC3oFL&(IHx>e*+aZdi0x@w7_D~vs7lvO3;t+IcF9J$gFT$Ak%yLq$+DX6L?YDvI
z8SCXibl;dNAVnyLixV;H<p(sB<?{O3LR4YtVz_*jWa-5(9kO_^akYa77!6v8a~)=c
z_Z<E0=Jgi|@Inl8&(3@#TSv%DcZG$nmTyaXsE7Cu+K<v`z$LtUm5hVG0sYuhdI)o<
zAA%L@p03@&gU6*;4okTKY^<debl$>!mA+S3X+Ql!eCdW(n~t-Eq*vGxyv2=mE|_$>
z&OT~MvDs>c6%8g}e+^t>o@xfh_JwJAhvo0yHDs=SR38={n+3LhaEN&ZYo9B79*FI0
z*(<*xO^F^S!N6K{t<yHzXDdm?Oub`bv8On;H8pv30_-tAsI>qo4RiG@y&^dNfz(~w
z7MKowoCq*$W)Iqo4_faReyl#n)~ef}TKCYZdcc2j4Bru?XVa-J03s#c_83fe=q<*F
zQ`ZrfUeob6-)XjO<bE0Rr#HnXc3TJ`AO?;+QJnTvzxBgs=UvPg=|z{&t)nnK8N!Si
zbvb7o>|bcp=VmDgxfYZ!Qw$P61LyV~BsrI^K1RGsWupjzSTqofE9uJj+nmkUD((}<
zS3ewhVj3`n_8xFODQ=12lpYe6*1csoHc8RuArd4f#yXY&fP92&g_LC@oZh~T{ZyuQ
zq;A}f)b~utH`-Shp|(I-u8K%p9BJ!QCO@Rz3zyhBAX;g!pSx<8aE(fUzXzPr>%Kv)
z7S<IBlEXbVlG=2%yggVt;e*e>s_W4{{H?X(xg;vXerD7ONY3KqG=y*x&EMigQcAaA
zS3n4()XHzOuQw`n!w<Zx+g_58X9Z6BC_z#kzYiilEAag<H*>)gQfH0S=yDf>Hx#BG
zy8hOfIFZsVKWhe(>#Zi$Q3Bj$I8iud<D{>4&Q*kiT=$c_<Y6C1{HVo%tFt3lL}{Dh
zYDOO~vpzW5GmjsR{ITx@HO-?Zp=m(lv#7z&5tA?>QVO|**za9^5A|1e#Ik|EX#jar
z2|_+cdOW`{5v;FF8nTVN#;{QAS=z_eWJH#>0|N(r0z(@Nu&{YInbz&{{*09DdF{BZ
zI|{Dc`2d<WDGSE<Q0W-*_8~H-Eh$j(H^z>Yych_P5h^Zmfz2hCuGOe|Y75>Pka*;k
zIwle<>>aUspCmNrcCHU%U%Zm?5xuU9?{d`G{HyUkmcIEdJ1Y>=pakm|D5eVCb^$mc
zjkD{m$C&Nm%i;)mHW)ZNpT5?3MpN_2-^zYel@%W%F4hU(>f_Kim*c&=a=L$P6rG;Q
z#4*-v>xyLuZb^d8o|CUMrZ)RvHv<z7`hy{cvqE?EY-wXVu^muwf2QHhc|V%K7fNAA
ziW`<Jk%93=rfbbi_X9FLjyTsxqqT-J-+%b_YDWp(%nTQc?l|z#UAR~{G*1D2KFFiv
zx5RYG;^)gHX<VT08^NnDa)rt+posyUU#O&rLkFacR+`rzk{;TB7%{Ce8WyaX)zCTq
zUQ9Xg#9Yo}r27m{l!qJrxCnQnW5eqUcfIs>e#VQW+t!W^e)=%#qu5}tU0GvNwdU)j
zTssW9kdn$!a3}?{XTF;FcI@V3GNhG6%PkjXHU&AQ&$jW{yy=QW@dR;vmMd_&QSxj4
z`{`A2YBXd|9F2IqH!_8xT?(li?&P&BD+8vZ57USZ@Z5bt{6J|SvV8Sgmk%duuzCP0
z38!t}r~KVMICGeU8EZK3soMOpvtZ$^(kL$!4C@wj{yp5<T2B-+;d)wpNy)a-u+Lid
zN<7_Lt-U^Ru76dNfp#_ljaUhzhn1!K^=diNW|%$dDt6%oEjfgojF+V?f&y{LYar6Z
zn2ib;veinGZWU!q@1R}K)p^}G0<AxixltNCWden;TrACRue5_Vc?gt-$*tz92N%zJ
zKzVY4YXxPr^Ss~KAt8v}c!~lwAm?F+M@x#GVRrUg0!>gQ4e1!ZSB%hSZ{+38D`PeY
z;X+W#>zbN_3n-W)n(AA#pt$Wh6kKXjcqsw?TdYMl#2l#Ob5B3Jm&9&$b+)j#NBS=+
z=vq%<ht<Sk#J4DIZfBxt`81b?(@K-6y8v4bJQ!Ud2a!?;cUlpvr_#ymKm7n!uQ4hd
zCI7qt=iu>{P=0ZIvwj66lCBA^g{U<@8<qvj;%wsi=@yK2zteaw_q=%`cglwhVsW*l
zY;&b>N^v&wlW5m<I4SG4bO`2PR|QbV9TB}0c2}ofF-1UtBHvd}R91z4>@0_d^eUm1
zL{Cg%d29Sp3D|LquuhbZz7YRm6!WACx=Q@A)<zAav&rN_)9=K@O)BBptWJ@;<V=zW
zEw+`97IAQpTT%cE8{Wv5)|YL+(bIY4uwG0hGc3X}!g`=e482A1h|KXd)so3#WN@Kg
zOfiU$wvt_c0=y08(=CL7ViQen=Qhq{nMzH>e{{+M(JZb9Jt%VJwsI6Efuu4fKcZpU
zu(vMR^!2@?!_K*4a8PI=L{CUh!_FhU)$^E`k_hH1T<$a<^I~_+u8KAJ1(L_QQSto<
zQCq8it$>whnh(o-UH11%G1msY{Ipr+nq^t9Z+LDEX>c!5X^c~4WiMZqJhCwKi1F9P
zmyMpXQf?K*4SDdQPGg8YjX48wUic_GlnUNFig&akXR~CMF+%kRg$}#I(+0^fIx9R6
z+ex3^kc!&r?mG-%jX_-E;%2_%^-ysiv5FQ^b*Eof37maEC(hiv*n&lRzFTw5Qq8eA
zd_@tjYo_^e7I*!gIi&AR@FX#g1hRw#1NwSe7k+dXo|O1U=IMNP?VJ}1W@MWu!*QAQ
zLkfAyIK~P)9DzFAH*s%!bozH>Ee`#3gRBeoFr=@NAd=o{I7M}qrNL&8iO2PB7W~Q2
zjk$1?MM68`gA?_Yf*cYEAz$@Vg&G18yL9!9Zs$F@<-=Q(@z1<DNWId_E)$b^c1{=T
z!~Q+=M$OH#I&rDXD5)K$V}IvYU*w<J^cWTQPi%+zzUxcZ<RoR{8<lke@%|wC+eMwy
z3r^y~Q!YO+G@i%Hys15gN=Q5?x<24Vdsuu(xNN{lZr3%OwO|V|e>;y6qoVdsz+%^}
z{o4crg8GPBI<orHErQA*qJt5kHq!-T+h@bWTtd^KgKH{g(@DyfKhWoC{p&MSw_;5x
zoAe43r^o*2e7&V63SkXpBWG;qlH;vceeo$VL?8NbqYBL~FT$~C#n#Coc8~`yYiEik
zm&!jZp~rwLD+TT?EsZ$}XQz8#W+UQiIwq{O8}9%2x)HmmaJ-<9mJ#?q)@|p>i^U!X
z$oIX|SynjXj(#1oOSVbM=PiA<l$Du_Y>_b3qa15;ZXD4hA92Me5Q!!GSjXdo`u<B)
zE;#nk^I~&lqG(F&n~kVeaP@?Zu2Jbuh72Q~l9WdJ$3e-IXqzemQ0#3y3B%x0Roc)A
zOu6zdWfhs+Ct3^_0&fN@!pHyo<)$&yC+dd&B3AL_hOkvdPuR5(#iS}-*}fI10ys<9
z79KubXtbo?k<J5h^7~E}@^qOQ=|Qi**s0Q8$%6x<5k$BmFdZ#BIl`ZO4`7Wf8)iRT
zN;x$O#+xXO*FVFBUulUoT65%0N#1a5av7Z;)oW~<cSM=n-bd1()i_iUqz6=NK;cS<
zL)k=7j-Q-(2t;BnA#kfhhfl!<B@a$yV#$6~>uKdAV>6j@h`ZVDlNM^3b>&3qOYABB
ziBRg*dpjvWboT^Z_irrXFrUnU{@!s5`N>_O`jflCEoeu7gI(^qVhFwgu*atWcf$B)
zA|&}Jv=QLOw#&j3w5wWqxDSzlyCL_Ji4yRMWUdN~PQPlH{=(IbIsqK*XT5pH8#4Pc
ztHD6*(4+}3;X5mpc(;#*Lpdgc+=ykwG%Q<!RBCWn#qD@qS7{=S2jgGDyQqf9;TpPM
z1&2<dMl}Dyf%QVpmD%kDf?qXP&904r6fGl(Bn8i>m(onGmUlWviVIPsi+aBZ&rRS;
z&*qUuF{?066%&lU>2L#+Z2C;s0dC;=nHjHa9B4Gv+#4e|z%4Gx)WN^+h!`5#5Hds_
zr!MlaY0enAoR45BgekpYet_oRIPz8gD#}BP7?p48K7r;|kJFB(9^JdKfZ`9|@@!Gh
zyMf>Jvn1r+=U-{Us^(_ho;I>NV()juwq}zXpg@K#S*G>;quw}(c)2|nn{D&C&`{^*
zP|ync3aN~%zKdSdwx4C&{)}~BoMNhCXpzWH!R6Np`mO}bT9kA9tDibDZ)DX*n~27v
z9{%(><|p-WO|)7a^JLvo!Ydq6um_+Lnl%_8wUxMXKmAYDcbiXteK;1gnu0Ye-IAtZ
zvad`k!)-jNZLhH93_W`PH1Ht$@aG%-SzPVgH}UDaq`bL)3B)m(zkf<J%<!Cjzv35_
z?XW!;AbwtHg++N+%`XAIXd^Y|FOgvc6)8af0n`v4Cy91_Vb7iE8>|{@NPzG4iToZk
z@PLr$ut;*wi}Qu+_#i;Z9V}TfT)p|vQMQ|1Ctb8{uk7*P5%|q&)rUKE{WuqcY@S%{
zJGKEhZ?bIPwG_TRG%-oo{739-mftRYXHjUE-G<Uz--y$jNr1;;5s|-7<vu=We{ZDv
zBuiDS!t|9bqg#jh`}Yu5?d-cOW1Zna{^qXzDDO3)4soL+pNb&2!K0He)~2Wy?x!~g
z{Y3BNxL^E0f*Y<ebHZLGH9KT1+kK3~7Q+<pvA54F_X&RVT#LzEE0ZWw$^ZxqzhW?c
z9=7e1k-^+j(hdI5XvziA)w(ovG`k7POQ;)*tXW*yoQIbiP0v`u`o{qL`>w(Nd3`U>
zD4$B!Xl<+~gjFUk@6oB`P=;>Gppdo_FpMlk`9&TWNaK}72mbmrhqNrm@C}YhVx4$p
zcZN!T^u%!sXgU|#W*}%;+Nw5qZO?$+s9621Aqj`+fZ!w}aSp$gkoM|b6Q;d@z{8Lx
z_aRo^wA=?4jc>c7(aYoxU$4fqOK+F%_$yqOh)b4QSdVCn?Z!Co>Ra@UjSnhh{#h;f
zd)g}cvkAh5nJwmk$b&V+y9T((_qr~(6ZO|oLa4BzGuMV_!YJwO`YS!!2O#E1$0k25
zVjlDsu0E+fG%^_+dAp@i=^689B({jd3!fSJE&i;5s;5++oY|&ZS^Bgn-_1PY8BI)H
z@zR<^Xm=kqAAC|&T2=fyU-997j(*8i@9DqsvA(fJxo*oT20@ym)4Q>&#e?M9@?s15
zmdmL9NtxxkGAn4|_i)Kd1@gc{1A9gmHZ$*e4V7$16y~xDuYa&h`zKDnKs>XSk!e$B
zBt2mOH|Y*}#qtH3A8>ZOCNO24Fd?%&gjD%bxnW)OfP7O!SEs2k?_y96Q#FhN5p3un
z$fTKgVW}qcxLaL-A^xU-fQ*;^0RYZ;@!w*X>hw0?$h`G<d-+<9X9?uD_C5A;g^^u-
zEiaAU;q{RC=lgi<S7e^4nMK!Q1)Wvu8X%ABqY=CtB1qRBJ%6K(S5(S2iUM_GLt~*I
ztVlJ(Wd;djbTva*lQ;Y9ZEHEIRVk+>@q2}f=ZyM{9=!dV*u!zKLn;YcUAzmxeD0uK
zpmtGvqAB8Y)bUswS!bdRF25()+qsJzASGR^n}LI%!Xkd+I&J<0Vt&~2MPnGTUn)0R
zBM-sOFh&o<%G!lIIGMZD-MRWF7j2<_fclGCeMC@<D@QApAirHm)yWz7(@{(~zq7mP
z-l9i`(BigIfCGj-UahXgcIIX$t{ArgueUO&yoWLDC+=^hvIP8s2(7XBXpZ-6<OEIP
zWj<Y|g*}v5?u9uyS5`vPCHAvV2t#`)Lo>ID$`QMkur#>65A-NmGZftipxZfUW4!;#
z2F}U$R%rU0)_~$pJo{&Hd-$lqtO}O7fXXa>ujwYtD+N+|2o_FcQqdZ-WofE1g@sM;
zbn{q2IkQcZrpd&_Y)*gDqt(zs^3s?LG-96Vb&%Z0XOkg)n_ih57i+bcD<5#m;?Jbe
zKNpkWkeA#^!gUx9`fTaqZpOgj&lRavQvs^UX3Zxf$)RmU?X{M@*fDGDan>(ht#|Sb
zG=TGvAvr`{sPeH^8UwAPElV?8^$+Kk)nh2Lswu-w5jH#$^DkntR?D;(`$iUWQJtjZ
zI`~pN;t@SlA{B_+$s(pRSQ=l5u^>4CTT^6z2@;Po?zV=1pDP=A3=$AIB9iMH3X%|j
z)Z{InWG{GC=KC#(xuk>!>VCDhiTJFku4+8|S5f)u&LVDkfnuallJYq*-FBja#hc2#
z@?bpLZpJ@@pBJJGT&DMQED0C+XdmwonvU?gje|lS1jn#qsi0eTaZuz*?t%`rksDur
zagIx8Rcnjsxb)h+@V@;EQ-NOmxB91yf!y`JL-R~YLeI`W%a$|KOH`Z1zy3+9BYQZ-
z9!zD}M)s01RtM`(a0s21!=w}hOfNy*QjfQ>Duicb_7qM0fAQP@OT2{L`xMYe%)n{!
z-aW;gnVh9GosA8gO@xgcO`Zw>FE=mW8}4^+czIR1xrKQIg!zOxo(OL4SUA+~|1j9t
b8C#gT{r?#@*iR;(7ywWyd2pG8zW@IK(!g>;

diff --git a/docs_old/logos/smithsonian.png b/docs_old/logos/smithsonian.png
deleted file mode 100644
index 0cd759abc48b8ea69b73b7d80ce4595c0b4ae1e1..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 35574
zcmdqJg<DnK7cG3~ZlqgMknTo7kS+=7ln&`Wbf-u+h;(;%NP~2DclWoE_jkYhH(Y&q
zd^oZ9+H1`<=NMy-Ww5-gI0_;GA_xRR`79x-2m(Qofj|(3@UXxuGt@eKz(3I6L}Wxj
zpz=th2R#_zcQOMBMHvvtl^O&B`+-1rz)RpA5XgZU1lrRDfp`-@AY7Z21_gfL4Oo3C
zaZ%9o%ioOV-#>s)f<B9iC_B#`E;>0AP1eJnTqkQesDJ$&=R`FyE*TeYSr6iDzFu-G
zt)Yv@i~(hK{~i6XqbrSo{sHOz1txqmo|46arRgR`v1ZQC;*!PNBT`mWSoOKo!?Sz0
zG`Do!!@(4u?ZLrYV_6JzjIp@G({BG0LaV$wa$4Uum`TVyY~dVG0SriytZ%+d>>aoW
z<+<^VFfK@V3_635!aTD9C(GRJiK|Ke?}xuojl1o$Z&lbPf`e)_I~d3)=05OOxzlzD
zp~6ywcv0KbKtEgIsr7%oX~U%50<psRkm^+qwe$RPV4$R+fF%<ZjSV}}N(qoogywJz
z(8mSwLEy!#<XeF*q=W_G2AM`PD{W4LWO1l5Frd&WC?weneDKo{<j;kJh~Ra_L8hNQ
zRfs_?kQ>Aacc{aKH_<RMqFo@rprEzQrEXJ+T4GQi2ABjo!vvgeOAM~nTjQ{YSz4dQ
zc)c~YB~u=X%p_lX)OrRte%s1>b{cn(c3?1W^zsv9MCqRXCK-i3=!2+e^w~~Kxx4`@
z_HW3(Zvhr(lp07%=>Z9m`hSAz2Z3R|JcnWTMT0cnAq|KXKO{AQTtr~11;h-MbOo7<
zQ_CV06odyx$58MxR_4>&yJg;Z&Y;x|P#Z^X|8eqrA>ur+*w7Ijg9SuRk)`8i->(L3
z5ifFKNcoJc5}bt<^$ql_b&w|Z`edM$)j1YsNL5H~7;rmsFe1zewX|mEu6yvmJ6I+f
zJO9y3QvG0q3L<2qq?U%)_Sxe|V0Fi+^bHDvfX5hXyS`cba1%}l!2<_Q#mB_;3>pN_
zIuPetRR_$9{JVw8O5&3xV$KSra4B5W0P<&eL18Pgz-KE)3yJ^UPbk08&09dx?Golr
zjZtR8Zu`SH-9}5={UJ!xhi>z@q5|{Zvq);AO>MQ~)V=YcgW9c%_}+x%zw%}I3F<Ka
zhQh)d$@VH}vZzT^bgX?=B`7G%$o0~;3=RjiY6S{ZGmsR1SYeFT#HvID*J?>_{8w5J
zhMFP-71fiHBBTm`QkIqk%BD|Pp!<;Z+ef1|Ptg(IJ$va=LF`-?#^F-UZb=zYY@P?m
z52DzgBmWYCy04*lZqR9|ji3BJ<J*yq&hQs?h4luLz$UZ6p{bJgguI3gZ8^Wfj|lw9
zYtS7hILq6;pmQRQd}Jw%-;M3vgt@`@!H~b3&V^?dMCsSYgFa~}0c~yO3EBe0axRr>
zW3LO)y__xeZmf7&*p{UR+%MqCb7jytI1>h?KCT{cOWb+t-Yu5Ea|a(bpoTY~gae<O
zk$q?<t;3)4ChgzyC_G5(xWY@X6N1+DZR@a44MZE@+|ceY@3-6|5E9R$O{;o2LH!Tq
z)4G&%M^@Z~5Hrjy=QQpddnPYqDw1-8?7AY(ohIrJkrOo}hRejlP|tjquv$+vs)4>4
z&%+6A<Nl<<&GXE0z_x>Ah)10Ku&4~RT>WKr$iZ#Be6rWT#mU|cY<UJlx!;>$bbEG0
z(X`w|k2?5>m-vY<jjk1&8u_7cH$yGX9aU7P=dF~rxK{U<1f0lrI#JR7@O>c0{r7^C
z?Ctk|fFdHhS0nn};I>gbVH?rzVDrZpx2+$|IxE8i*9#hi^u@4o5UR+*Tb9&h113<Q
zszm(0M?KC5v(V2|I`TJY{PM1gJ<nEbY-Ni+sw|}fB>Q}O^rK-)EY)@4xG)$BCo?di
zLA2Otf9pnz)4Sf?fWYxds9m0+>7Qm(Mr=MlcB`$A%pX`EVz!yWl<MkD3MTO1)Cc|k
z_0Hh52Z%7%a7Se$h7k2DulWX3xmxLLV!3)R(scC-GI0@J!LBef8Ybk|jU9b$bBLF-
z6A!8h?OknN^uMuN{57-p`*|g?=k48tW$svn_@9yvtfK+Fc;T*bW;*j>>!<fBP-76f
z>Q1NDPnP;ya@<tU)y|V4#2Gi3Z2Z|S-mhCdBn6*?uLIs4wvG5WhaHxPyL!vI&b|BR
z<I~lkkJqZs8(dKa_}%w239&ph#@i$_(yFCG+~2w?CH`b?eRDS4b!?J<O)LJ#t9y%?
zcO>8@QOPM{6eSz2#pAojMU04R+vPITXar$be0!3C#4@5QhvujIn=lv_8F8jFh+BBJ
zxPu|2fY-poYZo8#W<Eg$iv%Y$B{$=)H9It3-Z!2jrne)wSo=8?+3y}WJciriBSkT$
zvhXY`m`ZWo*ZSHChB(<8iiA&}N(@0X+59)4qv*X(y&ZE9D<(&9E<v5)tpj|<vw9v0
z?IRmejYC`{2f3YRX{F7tAX1HQ!=e4RN6F{6#?LhxSA(LoG19_1ot=|`r;rr?vM=_B
zlX80<&<K=-%ta344qilj@}1YNwyrgzd6B4&R*L@EBc0d-_uT2Q@v|VWfoT@@X9DC0
zq)2DfHGk2635zyp#h`x}+=NIbpdRz!K9MQ%c+}M4MKQEfVIOwL#?`u`%V6BWTfFin
zHpi9g{yHo6i)~Uwt!h5_mWf}pnU9I~H4>7qO)js%r>Ni}L&+<?p~lX?Pu9CISJVeW
zJ4#Z$2?@6oTToH7UJ(7&#l7jFN4w`-kYaIFG>YYh9c}w5#G$VfeVofgf{JBLCIXjo
z{cCd2?{_0*-tv*)PIA<>P@hS6-o=oU?cBR&-l9{g$NJWb5n7qz{c9&JiO*G8{oZ}^
z&d$5=+Pb2m$ihKEo?{(a3ZU|4m<(ohIXwk87x~uYC?25=54pjViEr1`-um-{^$I+t
z%Y3BW`M0H4_l=xvP^rd{sIv;#ZOltO>!C<4nHc|aHDp@eXK}OZR){d^tqGHfm4G7b
zIf9Q%PRv?Gt{QJCHMfyHA)I0#sOmsP+5Fy8;(?vkZuoM4bIyI<{wsePtY1WG1rc_J
z`#s;aw>*wia*5M`f>(s?z1JGD+*q;?<j_sz=iQeDsmDT1RPEPo^AMT%Ddu8k6JbQF
z#F7;}sK|QJ{FM4R=X~x`f0h%pL*qsh1f?06w}({-t==?oeVjWg-fE5^UlTxxoN`B7
zLDhB_t@^cwi%v23U7*RHRM?7$_P0{X9Qxm5So~sg8C4AvYd=0qvDQ|4nPT6G7V089
zZnABxL3}&izIO2+Lb%pk$I(PEziyCnRgXuIqF&4A=Uk@u{^TPNA`Ijtek}N*zDMt{
zt3i}TUu?KMSro80ti4AYRMa04W310t%PZ(lu7WCkk`DK#N!0ByzYY!NZ5@}z-boz(
zfrWbRLYCuU68#s#RaDM;R`0r;0gNGWSg|PG+e%N9P2r%fDVsmKgr_La!dgtdvn=~3
zL>ojQpQ3%jiW9bo`wh)=Oc8y0g)~f~d0Xtg^+Xus0{_KuUx`B&EjVdJ@2-}vi7fbE
zO&6YywZ+Q7dRsq`G^}@{yU3WJ<-p8HaAr8Dy0IkKg`f8+$0Z9>+3&2HvU8_R=#N(*
z&7zlZX7kv%3qADq<Px+F84d4k((T(8LI=E_9H#bx7AT4qH;MMk|8s{co693~O`LG8
z<=Q*sK`TOkw?M7P=Lmj;pQhCagkb5xJGyre5@kA4nskMyr+rwq-=A7Ld5`z^8G8y_
z`p|w$#LF&lUU0gz+^J#z=3MD88VkX?Kpa%7;9`>gH%IK~0t)M3@81}iwxq4b1>TkX
zwb>sh$~3N|2TAPEOKb#?2_Y4Ps#7bC{V(pdDq2wXoQ(46KN6^s41yg3Es2*(_~{QA
zWtiBBeF*oMBAm645-tT=2<`9~on!)0AT(<p2>v?oWnUAitbgWj;7`Y;{TKHHCJhd}
z;BgZYeZ6lmRt3*OAOWo~D`HnZK@tawPr1{ghY56Ry%FlmfmK5ZVN`e))_d1ZKPtBT
zQqq9T@ytOzVA9F=cDQRQ@w|g~TTY{<b2qK^14ca+2T)-isyB??Irp5yWOX1x&R&lt
zH}`V}Sz2sTx_8IhFp~fFFl>x?90?*B^}ff#QECY3{3q*&W339~)xfd96|`Y|g<FCO
zhb?}?`MD2spE9_)ud_Y9z8TrJxA??01}+8>S9rqjyQ@0juH5Y1`)WJ0;58bJbfIOV
zwHpoha>gl0kzpyBSqGa$Rl2xh8vi9~NBa3M&{3eT8HXV@GS!GnG!D75uVn4_pQl`h
zn0al)`*k>78cFCn-$!up^qk7JwjwxKAe7xQk%{vzb=Lc)qm?G2w%1qJcFb$xMK|5w
z(oc$Oq=sSarpY`xepw`4-DL>Y68PeQJ`ull=0o)_IrK9QIl_Ts)JV>HYsTsdR~MEa
z?nl;<UKX`cbjw^q;g6Zr{9*b%-FM4<u6w7Kw3Ys0>9|H+=+PLb%tC{Z<Xto8ft`op
z<!9$+m-L{G`LtP`)!zkBZp6RO#Rh7ebuUfJOY*zX>w|F_UZJP)MEntA`-U^o-Q8Vi
zdER=h`Oq&ZnwN!bnc2ewN<Zp7xBpdjT#Zz`Y>N6x>Clsh;a8b-h(BA%tY2B)dF)a@
zUp10Pwi~g?-_4qN2|g)b_m-X0&rb!wNtSj7YNU?uH@qiK3a~S?_RY6-5!D6*#r&+}
zzvnecB4FOI7MA8ceg4##7-45_y=d{#v9V}2fn$$ZbJLZ%sFWO{S}=ba#{YTdV9jfF
zQ1No0{B3>B<&Cf1r<vQ^-#_EC`$@IYetFml_NI)%UAZiA{nZMCA?{qi)>Ju%kX8~*
z|F`!Q#_0-p2tz)fN_YHzKfk)lL1TS&EL2+YIGLBt%xIAL%fQ|wY4ajny0ksBp7fI-
z=(f9tW~14gNJC2Pwl2efJ59~4X!Fp{;(da)BgYb<t!w%OwAZq>KOaGESN+<b8rrjz
z9;AeSK|}SU7YEsAB|uc-&rkZ2wu+K9K`*u!_{n!?qp51oq}W9|{^ap=q5{#Hd1KvN
zvS4KdTeJij#fGGf$cJWqs;yfu^%sICU3WMEipA)pGWl^20_0h}nv$fks;cooAHHKL
z@Pp<R97&>j#}-0hQ+-gJN>)#3IN0*^ihs9Q*K%X!&bHSVHPx>+im&aIS<JS*%H6Ax
z{GbtX)A_Z~;KQAJjwDxz5<!{Lm6P7D(p)>dG)sQ|jY9x-I2RotHVLdT#O6+2-rvs?
z=C0je6^Q1y%NouwZu?9B+l?{p_Hr0xxKxTUOQuUua>W9?J7C?dytg*{qw~?m_Q3aU
z4_6~q>zX&FWTr6jlUva7KLu_+{=(P7ORont8ha5s&KVc+uxylK!fknm&3=JfJ!ogT
zx}9AAhQvp2W3VU|{3Y^Dy#A-^AGqwVk!r864MWB@7UM7*QaH9kO7!;yv%V<_JRQn=
zd!LF}TQVO!SCXmstC+nSl?!LN`n0vU2>fI|Z#3M)I9u5<7pP<D%-Yw60y;Ywavpz#
zYtnMT3HbFFt~xv)cKB+6WJ)cGw%_N^o2HD{fT2ikslpB#scL*w-yi7|8_4yq8Uirk
zjq6{9y8N3`4^I@AY?tG*38%3rnsO->xi<7+&_y_e#8f0+^cI3S>2c{iio;^N2Q+ub
zso+)w59l{;6VW)!k&Xv%i^(jf2z^yrBz~xrvmVw9WzQ0&`i=sw5|vpBm4RD-vkXCQ
zX%G)8VRQv8M+mxCdbjVloZxO(l7qM3F?Q&>DXBq?KI4K1KPmhvD7(Ut*8QSu-NCjQ
zzc#t5^ifb-W}8{F`V1{tGkYSlM(DJIIt$`5&lm@N4179zmRaVY3V_57L!4hl9SbaQ
zWFt#NXEi{44F~kswH(FBG$7BSxA~OcXR#J@9$a{oS>Yb=>=c%%nxVA?TzWHbTkJC&
zXIl(^^B5*`tXkLNS95%ya>YS@tU*5Gb*l7ISA(@H`7%H8<TmWPL}QH!w=t24vV)II
z7UL;hPar^-iyu$!->7QLO))}f?z-YFhw~uw89tVMHh50^@=c7=17L9^jwh#f^4PK`
z<#nCSvn$!7OF`T8j2%*)BemG>ptg1M2&n*oo{WRepPjw`o5AtQquvd}&qvGCB)-?v
z$7U|~W4lr(w|VRTlJ&64Wml3xdDZ}jZpcM)4FUl1PhG6_ttOwi1&?J0?o0L^*U;!o
zB+~}bQzGHEjL$#N{7cHg9qHd0LE;sZs^kd=w@AP?i@VnQ2=cNSZcAZ-(~lI<$(|hS
zI#$u-vw@77l7WgMr!zm*@Ah|YS);(7BkY{Jetw?}t3#GQbvxiIaBiW!^9r^xnld<p
zG*c}4OAZM_%vGKQ4<3=_Gy>4XQ+ygFtvWP@+&&q8DiUQbHR$Sj_5ff4i}m0WH|OzT
z+4Vhu&B^7Uqsa93sS<MCtzMtM&b&kr1m?s4L6T3Wi^6isx4nL&cR}uKKZKw*r=E1t
z{dJxLTa2#oWbTB{Irl*kdnUPKX`8@{gi-rxyDmj8qL?aA41-Hb-a9MF!=)Wdp`wxv
zDTsF0)V{~FPEp0+chNC3uLm8X5&=N3vpJ%?C!Zh(tw{Y#HEMeZytLJNnJ3$Tdbf5z
z76@Ct=i>D#$L-{M_w9|%qb#kI7JhX5#wnd*yY{Tk+Eiw(M$&r=J4P)}VigD6uFVwG
z;YaxQ-bE)Z@|R(Du93I`7*uSsol>Ixhu#FZiuU;6ScwD1-YFk8MC}fB-)H|~wyn$C
z9$pZkpoS2_#B@p+ZZc|sj%8e=+s}Ow0RG9weZy{^ZNf#>I#XMScvhQz1JHv`fc2t&
z<67K_Q`A&DG7850xAB}46g=Kwa|JdM5<+r&^|2+W*hl^<5e(&90Me<kdzB}RRIeO{
zPz$~TWzBF(FKf7ugSUn9kQ<X#rQhb<l)>#yYBXai-r0KXDhe1^Got^EQ<-uwt-38T
zJO&jP*6+Z^(3gl<>YI3BnQ^`NVG`FXLT}}|FT0&I_O_Rc9aEIt%++)hOwJi~T#Ph-
z(+Yk&LBcbvq7ZvXE?U+P{)5k;@XGMeBL7;Tw5zNh=$jParib&R0+I*o!3I3`^qa-d
z7pPibAK3KhWNBdgt<x~h+|L6TJb%SqP91S;+mc_QCYsg5@v7-S$3nKsYIOG(2KCp=
zn)_gN%<@l|^fvzdrYqV&lhyyi@tV|KIKNxGC%1O<Sfo;$QE#sZj{<1X5A$cpaHU8C
z?==9|bKO>O_I+r+IgGHRDYJjgn??B~gu`1-KNrFZ!nzec&$aNRuU8gFbmm?BBDb8*
zH07SQqnBp9M#q?*(!!~w3x)}43|I_%OLS~~c8o!%0Rtq%7cmiAi7ieVTHe&*ynHv5
z+2zEJ@Oi$Jli{GKO?=;}18(23J<>9J%rrOPXBWV=S?vB|a2K!qjH<aWGQxae83WwX
zI<;2eZUn<&1Lbh|#m)^6teU(Vr^^qPE=~%Rs4SRhs&cw?SPHCbykw%WDbhR)DcT&W
zI0Dch=SCMf{+cW2rs*a^V7n(jP>VgxxQ*fhH+Ykh@C<7K?;#Topslp;&m`h}OWoZA
zPek_X=W_JVL2rqV+kaLh8Ci<N=dPs8&7-&F`rb(?HqTg(e#`r#=2wm%Dj!>cI9tOB
z5_Tg)p>GH+Ybu0)8EB-2W7`_4lSD5^R~OxF>1se$YkO)<*7WkAIVncs4|W?oFsb^S
zXsiV{3KIs^Z}Eaw*s|uIGHhX}{JT$3C589IIQGbBX35dJ9Cg++><8)R`J0rr(5$wA
z#8c;Wxog5TasT)jd|SFcdnTneh4WB<q_`mO#R3A-e;A!_dN+^^^^ync$%hnGmIa_+
zhLl*m&Qj^XALoRtXB2O4P-gZKv}sBYpe2Yaq6?0QAveNqI+Wza<$@zf?4es;MXdwD
z-q;W=V3aS~GVNS06A|)fQ*gIc4&9n~ZNL+RTE<Q4x$T1X_rhPL<gFHepSs-yjf(^^
zh?Azat(QyqkpL_($col}aHTL~^<>`1`PJ6)^pU%Er9NDTheC`zE>|P1<<2mr_8?b+
z>MiK45&SUmRG;E&o*ujC8&?9|iZ`KpX)bBJB)FU)p`yp4%k9<vYH@?Rq#xyZ*-g{S
z$~lakDFwb!rd5>-zALJPj@mbvpMEiQ>quFB$`mM{+tL{E!V;<pkJw<8Yl{uRRXY=n
zmF9*|4T^~(LLbR)ov~+@&1+#guwn-!4b3fgWTe8Y_S$T;UJkKq1NR<CJbkMN83$E5
ztGw1AF_N4>)pQ>rUeb1@qh%V+lhXq0X9+k`@E>mm2MWoC-`thppqSyelN_LhQ`w#`
zC2+RqqkL~$bacqzbgC_U6D`+)K@$x$gq~L2vjTvu{NR2Rvrp7=dNfa_m?*-LjTmdI
z_03tv6&XW&YzPK}$LrpCANijCQ(G%5i&aJlE<Ixx3jkC!^W<<D8Zpp{1*v%Xj^ij^
zQXnIc<aKcwC?pv@pBaN=u>E!SDw4|!8o`?%k~s{9g2#{qL1dE)iS@6jFH6imUQ}}U
zs~3r9>zY{{(&ZkY?b&!9hATw)8R2fd0<{or=n8|ZUpu7X;%kczjg=-QZ<SCMG`+}W
z3+wl}LdmTVrlBzu(w*KLLts#ao0?_R854txM5VqOeSN>BOdMptYsh1{isKNw{$Rwl
z+Ih*>y%$xSM?cbDr6ZmGNi??Y2F5QaM%>oHhys=rRF+eTT>Dq2jdjk)V}cW6`|naf
zfy;3>E1+{S#4_4dLI=^R&j8v<EW2orhMGICu~mNS3;qm!v=cN^M-eV;Kq{-ML~apA
zO@ibxLEO*}7a|2x|4owv1%E2lsg!0d4~Z~`)T?MM7PD6;dc;EjMxmKn2(|u19Q>#1
zYn&>k#`;L-H-J_A09r&w1DtB3EJQ8BvXF4wcxWqM;)wp?*ANT(!gpgzGaDn_Eul}|
z*WxJw7`r4m>X_<>BfYIjs2#vhPw#$hWc$i#5bU;rn$PPlS_@}TrtJKSU#;~KKxEse
zM?q}Nf(8%+@7sqgh$HV+(<8Q%QN?~ICgqg#{85h)jb->3tNGEJvT-@CjY#F|KJm84
z1#Ie82z$RHqPbh{JS|#RzVKIH?QEb3x-x^vvnX9turR26-Eg<b7N=6eJ`4RhWjuOU
zkAGD4hM^Z40hLL^7A4J`ADcId#CP$~*GBIPl2k$vCUGkmUE#Y#HPs>ElBzqL?%yV~
zlg7>Iwt20-iZCqym2!?=gTbcj6N_~{D0*N$s@S=t68{k-F<DbqnwAOa?#064o-!Vt
zl@ltqlZs<d5t3o0vVox%8DTG@2m1EZQV6xzQ5%0hIJP_|m=@m#ofm@&?jsW1hf;;s
zxf9oOL4f6w&Y-u|4f@Zp?7PM0IF|!D$U*EQv0~TASww~k8&<FCa-`Grq2MS}Ogn?+
z>+Ya4Lk`#h^P~d$C7>C@VO4r|bPe*%2%d{0e^wv(9^u1~jdC2ujWy&ga+7aBX<Quv
ztnQB!(*Z}QBB>Nsp!@XF$ijGQEK=bK(O+Jz3+#I7!4%)%L?Jyu#m4(lo?t{^0Fcxf
zDsvpi{v`j<zz1TGJdjl@=X&&2=D7#Wh(kv-sIypBB*x7}5!L>@F7@}~!W<LLUy@us
zMhsmgvus?jgxTZq9q2#A62KvfDyyRlXtpLtfrQV+4Tae0rRg`VZ8??<b3+$hLik^R
zA^pp`-^1zqrK%X!91(O2FZP~2X8V$Sf#%QKFAeJ2pNl9Q@ZTZEtycM{f-b`5)&DA#
zlat|xj(n}@M6=Asvi@Z+A}AI4x<*B|M#f!@NGf7`tOvY1^Wp71HTZVXdiG@&uy?+S
zkib*o8L;A5BTIgTOuQ*KU<amY{^pmVwwqDY?<<*arvEyTSwS#lF(G)KAUavo8FZ5i
zWXxq?P5=hR00#*I#2U^Bn%mhC-vuF!R$@Wx#31XhG#Txr6Sk|S4EX;oR>)tI!z4CE
z1<Hr-QTHSIynR|PEy5SdBoQL99s<0>0K?YzI6q~(GYYS4Z3JM47tNPKJX-&kTQds8
z0ic)24tc$q|MnOJY}YNS?#WWPdM8svPZLsKpv0XTpKCjpx^S68Eo9&PjQS@&Te^r!
ztKmUof>8kVGZ;{-dZ+NuZ#I1~KNThhlg_>6j#?B3*S`u=6u;S0!Zm<70Dy;}d1BS+
zvZuh6L2uD9%gj<&N^a2orE&|f?yJdD2s`7keZ-=jjh|lSf6KXsy^sr%{l>tR%Po(W
zcqV*4+@39U>`(L8f*aaUZZxd7x-7zHMM$re?r^gFV#3Cj<*uyvzbDM2$AQb=cg!l8
zt!JUVz0~(@H1rC!WZkkgvYYF}?Q5?sUO#c!r+kUF(yz7cW?uYL(CqmCO<s_KeUCvn
zSL+;etq1g0O=O#rc_i#;B`kFB4~s-s4X};XeGv5LRQqzEOAPcw6}skB)Z!(=M|C1;
zY-gtQ6sL#|k{;hv8xOamstaX|F7<tv(r+cHJyjoYyvekzs>UzS_oX*_h1l@0Wab9-
zn67#5-bC&Iodka3!=JDAPbudO2S4VH4Od9Z1K(vS{1?rJ4H9*()UQbs?`O5d27C_g
zV>nG>!er&1Ysigyi!CCjr*`JUz?5ZqPkQ{ubkP5{WVAtv2tTa1=;&6@kCzPz#Jxda
z5vh&*QEwmXoZ14$@HLZ0Z8z?FtW%kNXm_Aq&^Ks8ok$y@NG1Ol^p-|N{(E4);y7Vy
zLG4zk7L&xFc_}L+++q{{9ZIlPWovie1;Ii5IX=MQanh@$I|BEeA|@=pDCnOc!M~0<
z1w7S>m+)I9x&e7a`2m&B*kj~-gvtY=|3nL{>^*@+@-5brb)V9X3g42&5@BR152|&^
zJW?*T3k+6e8kLM_f1552kXYGE{<4kA_5&1|bpzpndrKFV)HaI3lW5l&?YSf+Y&ILM
z!`8)`>XswsYo{VrR!*H33&7Us09JWpHV4W<^_$?~-fpck+96Je;z>737S=vL3<MN5
znI4E;VxXROu$M%Ae%&hD(MeQbRJkR>68%=7DQG$7F~f1M5%<lf0N}*2d#$Uj@A;_Z
zvfd!ERO@2{`5^|PQ5Qo2vwG!6CE=8=+rP(25ONf1;#)(8B}FLV{snOy!4K8D)to9V
zXXn^Tk(JXfvLhPHkBtOz@Z-c56FZwNPIu&>)V6A!_#p9|!venSAx5IWe!V^<)HZJ0
zk-zyW&rsrwB97p7O5G<sWwRJQC0Lq&Y#9m!+RTU+g33AXJAt(blhg=Ft!V88h7n5N
z4+s|qMyeS#=Cp$9tsYr4PMlI<uPdvm>&WjKL7=a{ERw!H(^Ok7HN}7|S16V=*1a|Z
znG;0H6sHq(h|?5cq}7c;lx%=175e(cq#+?VZ)LGC>QQuB**{*yT&T@RZ}JBco;C7b
zc%98I2e)(odT^2`R(_(Ly5j!|a4%2x)JFrhnW>)@1lB5PzTf^Z8J&o5(_Xi7Qy-@S
zI*($?k(m+QYnI$us5Tb6M+fxsmEJM9^^ApQ)>hMOKtTnzkOO3{G<WfT6G8BEB)k6p
zkj<-0{H<(9RBZL2`J5jRSJ$YGf)MLyk14-C{#(+lZuab{neV@g+X&wdi_{c{-%2<0
zG9JGtT6OjcMf~#*&WyzkH6Fl$-bXJLqd@Q<wjldo_4x4_`DY9b>dsypR11-&7cG;}
zH)<*f3}+jucBIx&D9E}I6Il&|#a?e>Jg4m`+!B0YHA4UJ?L&$DlCkm1Q9vco_++^E
z+?Y)|4g9H^X*i7A_w+<_P`{(D*?flfLayvhzIH{h%`Ew;(&thPNCYt%>8v}izKegY
zU<JxTVH-z4bLz(KQ~<##$!=a;cez_`9h@Sk7u5g#I8e(K2UYr~RaH{rjq@XFUAG$q
zs`eQ-qF0Lldf{;|(34qF)tKo`isB(<LTLrze-ckAg}DV&$-NHRYu}YG#*Ggki++2q
zlUBLwa}5J!hYwsfHp5;l)CMe}{1WS|IBRkxa9-!f@4cm)0Kgxieuw^N%RRu<Azrsy
zNT|qwIO-e$=%gU=)WmV8mEUqo0rVi8`9&?sLp1(d1TBvpyyyYQz0!jZ`Nzg-1OR{b
z<~h`8pHPSN&rni|wm#3KoD7YLb(QW*iNrqDkQFy*^A1!=^|@C`^@e@ceR(-CI|AN3
zT_cN8roGFhbR7V19#4#bu>MT%jtHnv!rsmOXQy%YajUU=ji)m6eo-~^?}`ZuA8$>N
zxr;~|NPY|ieh^jPd!5udW94V!5VnrRa3ESPu^hn}*PIJ$s9Z-{9@N@qyy$=?cekM0
zV6UHCR6j$04tKz5LhU$^HhUcZ^QRVLQ==7E(^3=Kl2$F>+>kv;U@^Fbz(Kjy8{_f_
zkkV#co?6j`pBIY(_4sCBl#dw@t>YRS914jeOL!o>0K`4ecpni;KVf$2^)apEMcN)*
zxXw@^?e-C8sn=)yFZ*^=g7SGF3P$`JlRJq?SE^T)JkO60jd=lCgo;GlBV>~W3R28v
z?>n-LrSRthT;DJpd$V_QgUCe(j(VQGHte@{Zi`ol)17q=JBJM0%n!ER={Da!o*bSN
zSGM#5#si>Tk-ur+SV&A-yo_MYdF)?L{C4$xJLBU7Geb2h7VFk}+k#w~_|-HzEu)St
z)saa8W)J>ec>?PLRos8v#vc#{-%=U80hew4ethAwU%pRJKfe%WIL0SF<J!oOf&7;q
z0?=6seg<t*PFv*~bUf6S)pZhIaH@YIeCg~D(*{{JdWyv(z1{Y#I}MG8v6L7g@|n;o
zW-z6Y<RtQw746^cD8K<(1M@&+@JRi<2yh0$&?k)Al}>c5$T+S3mA8<rVW@+&j}yj?
z)0eq`{Fs==$6SBKw4>stJpKlZ^3C~)K=uSe_<3l>_g@z4UP6jqT1~8W()AQKT9Dmo
zYJcvsnq^b3z^LfWD7DpI75cbzG{!-*QbrZ2-<<g7lwGUl!OXqR^EdOgzMJ<isegM%
zUuI%iIH54B^rjoU`eGw#^|<djw3k%cw=wUxWSE9;EZ$H7SgL>1V&lGJF9}jwX9#?8
z(UB_QllAOnozrCE>XDhiSBv8^!~e7*nfF${eXX6d-At{H58ZAensVP<cPH@-%T#4@
zP=E$g^99iH@&wu}#5hep0FEb7A;57|(Tx-vS@)j&Ew&b`-EP!5r(W?I<RsEA>?uR7
zrm5f_FjRW2Y*$fDxh-G#1d?`-cRgt(fA&V(LZq#0`#@YxF;Lmsy1jaBC(-;g#er)A
ztyV7Gg$1V*pidUfgWsMFh;EX`DS&w<Mq?z)4+$cytRWOms<b;R;`>7L-a%c853rNY
zdyXRAlbQ7IH38pJ#owQw8-w&}=f*NkF&-)Y;@M=F9W|af;s#Ww1F&q(rWbC8=bkU@
zvrgCtx3}YAjP|C8fc&W|(re?s*ub&$wHCL<Vz%dwm$Z4xiJ>d@Y)sSrwcp-7@s>js
z<XSj6G%ykOR%-5Ei-+&4c2ONVP-2OrJrd1U0O5S?Dj+|PwDqGR?mmF!K4mszz->Np
z>EFBO?+O!XJnFK38z5!*wI-{5AsRdQ;tGmM-&uyR*H~=IwNqR0iOeQG{={(elBpJP
zn!y7qnoz2L_Hvk}CJ?pTUnF&)(`@b7-Xl-^<65@sj1RQ|ojH9C<~6IXS$-FT%3`Ln
zuG*YUw)NUg?_81taW0H8!^{u58LV^rMw@c<ld66=m7lG4u5_^uOT9+bzCOf?eq7SI
z+YWMc3eDWM@!Wi;ee$}JEME(2p(F}CPhWJztk{#eALn!8y4$zU9h>we3x;})FRr&B
z{h``h8@K-AJYX?TezDoOUp@kECM5H80A?)+9P}6#ZFsHw%l?a9%<helS#<XVhW<sN
zZKV{2lbFpzz-6V=v~4gCQ@RdhvDVxBt3)Bxge-yV6)KBV9yHL2{zN+-*D-O^8XsAa
zN1DosWW^Bg6=QDfa~o3XzkNl8$?=qg+&@@Ttp~&{dMK&IR+Y6sF>aGjp~@3Rs_~xn
zQyODxLM%t`_dCy9Dk{xlW8;~7bXrfYjnnYw!$ZcdF)7lH&m7l&mmC|b<K|J8=28(2
z?{DatZ2r-oSQF~_W+<}Ndy+*(cen*p`VUJ$Bf75qi9$N*`B-o|dxmBk$&g@bIzM??
z0sl(5S_lsD_w)k?#l&b@!t<SVL!rR5GhV?~b08p)*Hn!FAzD{{Q$h(}bSg0X{uzwR
z{SRPCvQr@9O<?i_^BH*sA|a)tx-<gSO7Axe(|SY0a;B8h-NWH(#K~NN_OtGw1u^MW
z{V*Ut#CyQ4w0!yI*IcT++Y=pe4b)3OPI*}cZj1hGU{L~^mP3)ABLH|Nb>dn)ZM%r}
z`cQemLm}!KQ&7Lx;yrbqeZ$;^?3xUex%7R;?ylQXxN1Zv01NglTz5!rhm=qZS|AC5
zZfh)wKr3fIRrP@kHgnI%a4nSlI8Qj_X3)`8`=T*q3P5*vb^e;BJn{+mv1X}XlhK}e
z4JL(PTIEu?t{g2@cP2pTL%F+t;o;*CUh4sOp<~5YD+Id{&$<$S=E}xz9rmX%5QIIK
z#D6D-<X3YAF1|4@O<}r6#-6%4xA{IKz-_kMN$x;Sm|hO@vKNd=xSZ{pe7F;N4Y9?)
ztJWV*j3}r~lf&voxW4lZDZfCT8QXUHgo&0f61|Clx2|mB4$z2hI4QKsp1Am?U%Jt!
z9KHV4t@gWJCslKvUe^WIUdlrrxN@Ze?{++NG~7u)<<cz7(WUE~b-VqqJE<5o5Jk4{
zGeZUL{Zs0v1qqnpUJ9{e`~C|adN4!U<;G&&Hh4~v>y$LNoR((*3}GBk{K`&0zb?ti
zY?mz$%5xMyDFR4LZI>{$1+Z3e0M5F_^XZJ<fE+0r61wI;XM=_cx*ct#&cv_%B}10_
z2YS%6v|rA{*e!fbf5QB>H7`4gYEcs;-^erWR&`{VeTuo(wf`dcsdw%zm*da78i$n4
z$BQX{bD-LxmGE*DU7$!6qwQW>vmb{S)E}=Mn@1jA*SaC!5p1J+3bO2#{`tXOyGE$p
z_quS2(26CXQGyZ5FiD+|{gC6sb1t1Ki%`q0X7BlCB@FP;xGgY@(CqUil+q!jcBOFf
zXQm+dY{IFnn&w5m>Sgt&yBI6`fTnP!sF&=uzCslV7&jHtTbqHtbeZb!NHC6Y$rj9i
zE$yM(PL&U~8Vgos`fCc;KPBRq2!||<VHpAS%TeP35|Gzx-HE){S}IKSydG~<Cq&f#
zDm=B4QG|K#B;UJ^uaH`}ETV!WelQNqw=6c%^bC2x0q{q!5^9aO#sSdet8S(=CX-nv
zwMPEu6UZKJI|Oa@u8-p42!;b@mdm@^7iY}W`Eu7X!EiwS>SBl-le0@@FM<k)Q&aW2
zQReBeajHD7r7yuuS7m;y(ORJFj(uPk*~#c~tBp*$<Pix~?Sj&*G2rSh+2CxcJvxFr
z{85ELw7`~&N%M8@6)}>GAz*Ej_dty=n$p`yof0>5W-zIH5?^UvFyJ`VbqWrTDl;81
z)KeWi@|W#ZExY@xTN~?%xQX@Me+7ii{h^|vr^kUf%krv;InZ^(5jh53I?!zmv9()n
z?<|zty)^H8?93&ft`VH7o`k;8O!>0ZjuiizEi2F82|}<NBMHD`Pe_%1swuYo;q`1i
z5P>Ij{eGcwZ?LS@74Q``r7_$?!Okal%#AEuVbp!!+-!cbGzI|fHLmn@F_3k>O74f0
zWCi`4aAiZ<W@^N^XSnONHhv&T+UqX<XdUSw<AJUe=+%Q1_wTMK(8ODtAeF8)`ENK)
zRsHsL`$G(xWe1-QnoknTa%&!|3$@U6(^Hr_=EMOT5Y6<0uOnOYbgM=p0)0aJe(B8>
znoBFt9S1Bq7)+&+Q^$U%cFP@)g|f?(%qgrl>VOPlS!8BHtX#V4|DRPUBvib}m+l9~
zW{Fx!?BtDk{}AEx{Y2bh*<lmc$*6I|otlI65+BfLlvBIk8OH{SGZ_FK=MJn}%1>Y@
z6D*BuWOpi)RjfTNNQuul+s!!GY{$+_XDr8%)MjQU4vShpLsLj(aw91Pa-Dy8wbDTK
zyKEi{+vu>Po$Bs7^i^^h04geAX=*1yO7Dyrt_UJAxE~~a=_26ifBl!;unLd60hD%p
z?o>H1?$2&%6`y;7Kl(6O%IQAYAR}S=RAkhcry9-lwWhn~T)TI%(3Ki)1|S%!Zv^lT
zJD<Yo(_YhpcY%Nn4|sNL=Ca(ydY!hn(fPLvT3G5m<BPu|OGf~d0Wcj^`rj0tWrDJ^
z^IAd8IL}z6H>mLM<P#S;febfdx{R1=Y!(&%7)<`_et)XiKv4`p^$%9J95WuP=P!VS
zUK1_*oq?@AYVHjJo#y!~rbW@jjog8Tq=7Pt+^{;g6kK4xDCM5F<3s&=CfbU$q)sl+
zujQ$(hP3HX=SgEB^!&wrw!(6B-mikBZ*Xd*(PGFN&GPUEpg49_6}T_)VgN`5>bis%
z3oTxocnbIT&<^OraM<;g>cTeAz1{j?qfO~lFE@-k^KDG%q-)L~JMIAMe7OxUS*g!C
zh6|y6b3W<<#C?qKHrSH#7?NdY17c|Rl~d_~d&7DxrSAYL^HjGcce5TMq*p4UBj$YU
zN}h|F-iZ_wRKxBFv_mYTsa5&?@h2NaY^^9AafkPo+R<sBHYzxAR_l8@ZBEaG09L*{
zezjZ!BZQ^G%%=NjVXAs!U;u3RPfdIpv>%8|$GyD&d`!?h<pwO3bC{=qCS_Sa2lWiM
zl58lk7cskRpxM(ARii$s1%4YZ+Y0~>HB>>6E+E><(?``NB_>R3<6n4A!&HnFw>QHT
z9y1b2a2%Ug-Fx-^&-rQkl0PQbK}E6`IV5rS=~14~2J<E_Erq77d8PvcINeS_seF7o
zYOHw&t&3#3Ec3LG;aNMU!z^#61x0CNF{1Y0i588t{LXj?S9oxZsO%P_dZo|P;rx)B
z{<IiVR*ktyR|Ki_bb)+>&SMeF3jCR#5vt$Z6zT*w;t%nn6@T%61-HgfDPY8J1)0oN
zquziG7v{3@x#l3qpnz0Vii|C*7tPq_w$>5{GK#DfoP)2`uRS0glooQnpmX-085df*
z<OUa7m0EvXu?v9lrOV6jf&Mav6p2V<BFQN`G#ggjPbRVfp9e`LL!F-smr@(o3H%R6
z+9v<mit+sxAXc74rBrH)tCyGU3yFVEJ)rDE&MJKV=&)TT*eC_mRKR*caOT;H{M|7>
z?>#ggQh}4f_uIv$wpR)G64NhM2e~~vI5r2Bt$2BvzwFN5eV&)m%v4??{r(xKpTA7M
z>sT+*<Q*a&$N<}q_jd2c%ICkT4{5kKuiDWNJDymyh9^nJ2$f_E&<qf4{8F8~TQ|hV
zvA4TZ82a<)kt#@jf(MzjWS%Z>nFIju-zTYl#+O%IX-Z@L_&-<S(}o<)oq%YHrv1EU
zK6%2GbRED2yKi{Gf&K^pi@0mP;-~QpPp)4l>N%25G0*@iP;;ZmJ>1-ec|CBG=cPpB
z&^hLbtc<C@Vxv<S-8$CfBt1Sltr}s|;N>yJiU3$n$B&QyD5~n4tqT2kdn564oy9?s
zi|M~`Xy<DwKyo=+CwFFD`s`x8Xa&GKpb^aS`J+}X7N9~kKe=huvK4vglpVkp?HDb$
zUd3lR=_=oiLzl3hF^oDR55GHJfq8YZ_kBR8fGA=R;#-26ShiR~0y;TB2@z9O7#ttd
z;9@9T7xFYM7D5YT^e6&ALPZ#76VDnEU~0!IEuke5NxfJ0ynQ{=gq>!pzqW8jC%=GE
zN(l%~&fgDNnxBT0gVlLm_YpZOE!{=09)b8Ls@gbyEQ!ztG~aY<2L@`Rq0c!h>p0SM
zLXl9z|1}Q-f&c)wiJpQ)#%Nt<4TvH@DP4YW%WP&bHr+Wi7pH6iKOBStg>~Nt_e80;
zPtZ#>>1!ZU4KHkM`~tzm`LRD~)HzG_npqGn6ZU^+E;#DfSm041${42ttxnK-(p5B{
zlD$Y%%PEabZ!TR)AB)lSyB{JXU$WoAmH-++NQ}MWDLh{QTwM4v3?L>1Lwy$%3BZ`h
z#C?x00IBqcbj$zscoS`sNZ&p903w5PD$o-!j<zPUm;qU<`{1Tk`_$d^n`x_7ZV})r
zu-*JKz+$!pSLZf4djj*F^L`!xvOK;zZ8F_n-7UU?kuslX3o;wnsL{Maae>aSiJ!W#
zsNvWnJz~_{B{CRF-(?1%4}h_&7cgEzf&iQ?&$>Mz*oq@ME;PW@{`s~P3Iuo)k%7XD
zb!m_rMA~RP(gisBm??&Q0U8Y*0=WJXmRX3bOXn*`sud}SJepA6UMi>X6-DDhyl#;h
z#W#Z|ny5SPgH~0I8li1Io@&*NUou07D48Ag2oa)1_LqBYyAuIi6B=~$7*c4f+R39K
zZ3^YexA(4OP1o*qvQIB1E?3d{QN=_U&>Ahb<N#?sHBb6H!)(co<rc95=sdbtOd^25
zi@1W30F0Z*lyv4$9_au$K}7}DE6qY?D@c|1fQ<6h_t*ib7zhlET|r1CZY%&e-U9-C
zZ+*rnHr^6unnlyCO!u~v+*j*=4IZkvG)8q?C|5#omA7kBWiJu5zivkpgaB+KR+Lco
zum@bDCI*IBL@F?~*dhYvA`8CwlcW&N$RVGhPT9ZJL@XQfE<x-N=B0(^zp+!x4-TRT
zojAe75bFd;v6^}dlxQxb3BC=4#QaTAF*r?k4pfgBahA&B!0n4hkwyDr5y&!0UZo36
zLW7#|%19aS8cfdswU`s~fe|H_1IV}67ec<Oe0`m#e&Co*r*7Xeh&b)wcg(v*5SUR0
zAD*fbI;4=fdQD74@%3X+0~gjxWH7iOR{^l5v4m#zhf7)eB(L`Fx4enOuL%~m1kOye
zY@Y*N8aC<=a&W#ZC{oyljarqRneV?28V`vMD>5K_%S(5I(rUOw?4g`d9p~9ov{Z%;
zUUzU63>-!En-jPaNSk+1(Lwv56L!e&NW)Cm|GUQ62+3Dg=m1U7$0xK9NMi}LBywlW
zr`UPhr?Q`esIOzK;sK7Xjm|n!D-Xj}h2Ue3XInDr3c?TF@crMX7HN$c65%HjAlv&v
z6^Z1{d*Bs^JVBrSH_x+W6aufuM<@n>Q=Go+^QX)<6xeQ%WtOWvn4lAK=#gxJ6ADI>
z@`#_W?KTXtk0WZ)24vqM!CJK$x|E<5CE>Kd#b?Z6lJcaV-v3&17~SL91Lqh!OK@#1
z=mOADnNSDFsvl2ugaw6hpTdy_ni=e0M;{BX3={d+3JUrP6sD!OGb~>hbHK0%;SLc!
zLUzV7On2{X!23F%3Ka=0&fLiJjPkPtqqd+PD7+Z5d~*;pT$=?*NEk{-cQ5D{1N-Y$
z*`o&Y96vY^{l!CtvP$@h>P_gwN}Gb=E=UWFBL5fiD-;Vq$PAr1{(}SR%SpqI|IUa4
z2VmOYh9c_9(FwOPfK9Q5qfjNNk#pMpgjZ(JwQqEV^|G)=Ica%y6U_D4raG3Jag;ul
z7+up*{J;K3@)PF^1jKkpKk<G1`_&Lptm?wpnnn*YIR!-s8XsU3f-Ml~ehW80c;pM?
znpbm!28x1WDTO$pvkMG*?a`{G$bM64j4m&fDX;x-<vm-C0{WQ9A^N(RI=-l>{amy(
zPleW%1Gzat^5KR2O#}Pt!C|~6NoHgc{-U~ht7S1!Cx1Gu#$OJ6p!`96VA@~Y4tfa*
zF~?s8#{DcQL}?&-Vc%gr2xkhr30lFo+MvJ2Hlor|lwN1ITH~mnUgt<VtRMbIQBihb
zyg^6`NlIhM5R?>(l%OD!wvbI+bbMt^!SRwU$}jj3Z76LhA}rYXpzoj@s1v9Y%bQke
zkf4IIOf;4Lj(#UQE>|w0)at2y*$wcNFJWr2VN<_|PDB?1^`#t2+F_@(QSo9a+zlP9
zN9aWX7KXrL)@vJsz!E$9CI2W1F8xcbbED$A1%fY=2#$Hk&DjB!$!C6q=r@i4#*M;&
zajbS|(@4D2mB?%h3$4+6q+BINmJa1CV{tW2sLH6`zD6Z!)!p5lxKGvGwZ$cl#QaIj
z5yD=`kLCPlkoJVKjvJd@d{7W=KU9?FH&cIzdPOl=*r+$#=ZeBDP$vZAmxAfN>jnx~
zD9AiiRJQ@ko-AN<UA-dsmt)#lH;9c$QyG~bkq*3Lh2GiZLz)I-sqKn_u!dyap;okT
zIJzfR+B-v_9Ph`#M<wsN-!;gRop-}GEhyI>*lo$5Ockq*!Rvl4skEpM;{QWLp7lP6
zTk~i!lN*!bEp`%<ut0`Rke*w4eu*o7Wl&5OMPOxl=cXW3NF%w8U)iiHs1o!Sab+<9
z)xVQH%91vKo*7Ztic(e{W}9Q$Ew?g{ju8mPv#$?a32oiuZ2cds1#5!ST5DPy1=$$7
z+@dBPJ@T{!ZeQ$J3RVicq6+IGpn2;pnniT}Sj%&rdj0MY$1OZ6Rw*6fiqy`)neJyi
z@JC-HJ$n>ge$#@Mx{`J_Eb#0y&TYv!OKzXHJWpn1=zBaP2IG){&a5hSxhXhp)po{5
zO=>r2>DvwkWlz7-mMyF;O?W0eWEUyrT4P!CG2t!RcYKKNMnHW#>)+_)H5b-ukGV~D
zMt!GCHS6Pe#Dm{$I0&DmyjefMydjAC=zDYfHBU3e5-s+wiNi~6dgYck0kDx1G!Jau
z?wKE-xd`wjgb;VTa=qgUO<Zd;w#Og&U>~31`QAqO@{PCx57ZWVP;hA?k|=xBCify=
z-XEcR1VK+zH#f3Pe4An}0$K<O+30)FoAYyBf=ft#>$%6b9w;2w{`Yh?TE5%+)TP_P
zv=Pu{yg9Iltzdbg4c>ua<rOeG*FfM}>0t9pK89MqCRvvEf-D=d^l>(l_3`zm6>w8l
z#g7D!*b4{=E8E+rEax`@=Jpp<WTVgKXRfjf%xdiV?kH)~7LL!HpcaJHclMPDbW@@m
zayZF<#Wzv$Ia7G>D3u5j5k%SLc|sNgKeDusPl#@Q-y)-O0Ik{7U!yDOpin@f(Qa%|
zTOUG=NuJETp;wuEG`)ws;qH^s%qx;MP*5t`;0cU`(d7tYM((T;vSaDH={(ZueHp9U
zk4_9d3jskd6C(t+=onT|KYa-;%n60hj$0lDF1wq#c93J%bdh@9(vch=vJa^*Pj8BG
zWvrM?d1Ju?U_91@(O|*yHIVv@z>FJY0wN?G?ty6K2~n5m#|76d?wY}BL(1cIwWOJp
zl)6x=Am~&YueyG;2!z;iii2a}M)lM0L#}Q4%*n>WeHF(Jq$>6(o(|Wf8EHW|+FQt|
z1;={WsLzkbpHy?8D;Ogj)(H3G7*)seMRG=lvU1fFSS(g8m<r4jLfb#fXpI22RagcD
z+qKkXE9(+tMny#!GMC{HTMB}|AZm<txkA+u%ivxF_UY*s=x+K)O)zkBo+V7!-cc<X
z3JG)9o~YYso?CO%X$$A4Tt6LMzg!++f*vhB5eCXkZ}3hEGbh?jbW2pd#OF8Qyo75=
z2?Tvo4w*}ik)GQ(?M8_97pF*Ep-K8w6H}Lh;7c4*E>EEYYwADs8$9$q?Okr*CvD~5
z^Nql%dcYlBxD~#($;|oe?y*SWsqlLPL70f~ngn!(Tu=|qk4#pwNan5`*(=Xqe@m|h
z7$)^|OG-*m+Lnt1EH=t<j${083<dM*-077DuR-{Oic}OK&A{UGL7F3#5csq*!_nnK
zq6?RxxS|y>QFjDGOV6izK!30T@{wXv%`(XbgJo+~k;cI^^@lq<MEXxN(vR{WT&W*H
zv{uR}0;Iwk5`+*V98+#fdRGeL&}6;=Dq+y}#~C#^3|3Zo#h?AK)N5ULF(IC+6NPZ0
zu$0smu<21yP>LyfR>{M*^8)DtX*#CS#1&i;_1Kn=>#cz7n&$3My&U={SOktdi_{Qb
zYnyqS78qjZUpA0L6b5h{nipSTp*c}#_Iqk!V0j%~T+&`aJo)4Tej65OP|)+PL9f#K
zSmVpNLDNKvs-^<sWF$ydOGY%7dE`ex(zISD<D-4A8Pvrm9gjm1!UjkzDu#P&!c$kp
z6BJLaBng^{nU&i=tKXR*na{XP5Aop?hbVrlnQ7M)A+*f`Pq$!(ltA;YEF1KZyg^dA
zg{|AUyuRgbUR1*TY4i^!+sZS%FppG35N$b7h&(}MKRUnE(C89D?Qny9$<$!6NN1t$
zI+8f5F9+!)8$gs!ePyee?1IoI)GIs!P%C%Ziu=a4N=-#N>do+ND8WfBMdX{q!fnqj
zpf?QfDYE>230fbDg9P<e<XPvdkm80~oM;J?^>46_{;)%hfOae`L5_6fdtx<qH&NIh
zg|Kz8?STK{B1oSA3_p1IaZXLfEh|<2ykJub;<hASs9t?Rv=vcM^QXV#=`&m#-WDtX
z54!n5-jWibVRTW^3BP}pUY|t*fq#L7T05^NzG5ld!^#C*kX8^vg2}h`o|*>%-!tg;
z6Vvn8jy38#f7uk69Z_TGm&=SZHOeD^JJ4&LXx2V+Of%h=eB)$LN8R~daA(TP+hS~s
zWl2x~$Ft4}d{|)vDypNSpvbC<fSf}Ij1fFn0c7W7dm3|Nbu8e0jP6Ay$cQNmq<L%K
zKY!Sb-@d$t;g1cr&%JH@u&4&M7niyo9{ltVHX_o2)NA#-dFlD&`%eG=*C1BlMsfB~
zp}F-=R*GR+=!0JX25l?|8oj8hEV9YUZCcrw>h-!f;DiKoBO1u&ck^G=`wjrj0w~`r
zQ0O~yIM+YEO&>KTq-N67(&6(0wZ~_N#7JL~)#1m$`}ae3ys7FM`X0JYJvqiB!0+?p
z17&q}+-%e8Usv(=mUwIP^c8{P+2yWk`sM#4>8j(R=-T$~(%m5)B8?!OF0jNy2uOE#
zr!))FA)s`Jba!`32uOEGNlQw9!~6Zc``ej4bIu)C+*4UsPuEn><0wC#+}<9!67=<J
z?qcOg@7J$iquC!#m^nCpvH_PES(cs|hPILjt~d-OX~4haoxf94FTh?v*?rSM)L2DS
zI8e$_B&4@<df(2g^tbginSDarIWAIAaASy;x$p#eV^sD?pvcLyU;*KtG_r$Gh$1ex
zTc?PclY-+Dd5$}}Hs)}cNy_uz+1V#lE%wW^Lqh|V+YhC?{_q-Zn0zji6v;0k1_bQn
zWyETVyeOjc3ao$4Tv*Q`L(^Q72a=U_CU4`r|M>q|!RwYaphFuUCjTf#jO>BprU}wz
z<s^lITR~kN1lY;s=$_!2@L`dZU6Rt>q4Bb$x0JH4;*U$&6sArdXX1{HM_z}2vEt@}
zAwog8DIm-g5Der7#!mL3gfl?cG~%-n|1~p@USlP({h>hhdiN5|=;5oFDa&QU;aKyo
zHNCKI$1qC&`hUkx%_9%!+VtIYD*DC9#H3)~#pY6vH{OG0*kjN!ZLSXtoCV_v(=cZS
zco==HX_K&Mle`B%L<n!vg;$w_rB7`VEX9~Js%D0!@DPVTsWew!Qb8SloBqDO_BHNV
zHeHY_WWSx&<d)ntBt(62^*ZTsAOS(miAG5Iz^R1@N)K;fNihdyho@!pcqfme<RBF2
zpD=k)m%b^4Uctr1*5j_amxi}XmUo&)I}0OJIO{1QT>^<A(%d-}prlBXnn?dddmP^e
zpqjcamLBeUh<(`)#9e+^a0OOj)to-g=tm&hy*DK{=bkD*NCo`_bK05O4$fqfYtHIm
zLu*45gjF#Wz?XKxtwJ3}Tna4{RI-ovX}lQoL)*xISSa&t^vZ+2fwqISe)vM#jYz$3
zmhP5rf9O%WM9DFtM^EPppmm3GM7Q?tRZtTD5Gu0hZcJ_7v%~8U`hLW}4O&3p?o-rs
zwXxBCO_o7^qD`xqMVmLzGdsUV;o^n$icb%7`48v=&aKe>`jn;kCH3`5{g8=!ht$K}
zc-7d{%!w+I09sroXrXtt2to<*Wr#~|o%PHuF#c0&<{h}<lz7a+!dSY23i`=;(?m5H
zE48i2J>c8vwOCtd7H}<?`Lq1P4*`w963EeImo45N|Ev!ie;fuzS#8@<MG@4w8RQeX
z;BiY8RD|&L?((ix%J!?!{&3Y8m+GSof&fDI$--nC!^%U9%;YA&4vuP14G0^8JCiH{
zCWg3ljc4DP)vjr}W}5FnGO*(7z=Y2ar(oTijVo47U}as<!!#p}vzYTxzze(c$4_Ez
z5`T&9JV!M5ygK$0KN2F~f`F~J4!Vrvyk4|o!nU?M^BssoaB5z02%%t{GzHtu>t^X<
z{qJ|k_@3SOQ_%(6k!Xp;@Hz^A=-%VzT}t(HefJFtU#kpylxRlf8wwt&V=jC$zOsqs
z*Cu1}8Xywx;lWYgV7k03pmza>=MVca69&i+XDGKH*q%`&VuEGAOm}=nePZhl#4B!k
zZyPkM-^sR2`Qv=|y{&8u_dubh_>~>BHeXBd0o{w*Vj~asWHNKHCG!@<`3kC%fIW@2
ze;7J!8grHPDIVtN^}l6}o|A-a3qWyBD!8m00`-(XzMG-0e$HkpgrL^xW;dd4gWbpi
ztZhmiJ|sPidtg(WkX%5!MX=)@y=@U*%Y(kY|8B=PZIOX6`x_iQZEu%Zx-y@U%g-RG
z6rpFSE!Z!hkuzj(Mg0^y3wmAEY@&&klAOC&5RPY&Q;l27&%-+-M*hQR(vhmr>%&b1
ziznaMV<Xz-KcIL?*EJfW{X)C*dv+25QNY7Bk)<5;y_99Prtug(#DD?nZMGcp?-$Q@
z`IS2_27SXsR+C}sJi;yme1Se=JqA5>>8RpXQ({dkOe3jyu_Fvk5T-zZ7?_`pA1?b+
zCQm@lv=GV1@EpzLitL@5aUQIYX@IK{%uMPc3J65Lsu||IDj!sDFnoQ&Zf_!IFqJl{
z?riRc5NnwDZA(cJD@&K)Zvj615R^f29hE)li!+t(d(}`y=f7%kZ%<g%-BlIoe7PI+
zrJ8J5UA^Q>Lq!1OtSEf1!=#{FpsFPIE+)7gv#j*vMcSNWQDF@ZO&CI}eZb94Fg6tE
z(I^hvlmdqfK!Z|uoh)>Fe6#buX;#O5B9;SMs$-wMgRoO)){%388wPTPx}U6fCykWA
zl_z!FkIuWarD8;y?ta2BTYLL|j&3tJ!)Dp@7dL@a&X;bkTsI-Zosc6eBUz^9Fnstn
zk>-(2+AbzoTIz2?8i?Zf!uyH*<jc99IST8xILLg;ZXD9Qws)mRecH<Dc~-sU(B?-X
z>264Lgl?n|UWD%f`l_qu5nU`2i|Q5=@djacLHe||W%G!h3@mGGtnSfNVXu<_*<Evs
z*!T*wW7=+$D^uXg3N(>GO-}_0;hEh;bW(YAdQUu|xn_TAqx_Yt$-KQqibB=Jr;GW!
z7q)+(r>B>(NG7|!i0f(?a)hzZhq`HR%Y?Q4Kx;0X8af72CK|Vl05$P@iE0kKA5KsC
zs$1_ZGP9tW(*~&(mN=@-2nWj=7)VtX9!e=cz7Jm4ugUfZT-&~wGa#9`U}tk{YqhJ@
zRZ55AAb)?Ia4d-mece>-TU)#qan?z9Wyzj1blcuQ9nzL>Kvh}S8h9QC>YyE47{uN{
zlTlgZu(lW~cqwxc=oSFLVaee`FzGi$A+)&vl-;9XH&U+7&dwIPG*@qU$K_v-O<4D=
z(1Y634_V$<Eb3kaB~qGUF@IXZ>d*}&F#W3m?yg^<`=sOAu~$l}j^qOr4m0oyIVk{{
z!HbLV&ymfU6!7~r%+9fHsPJf+he8diUnZ7qu5@jY9Jx8<)1gM-WDBK^{iyy|*+4}*
zFtLVNrEQ+5EtW@71nKz3c&;)?Gv_>VoU_xR+|@Hsd$qV!ZrrfVKUm^gZJ2BSPdb%v
zhuTrl$VfeGKQ9TmMPkak9EqSzhfbo%8>D~I_9-|7iY6)kh(}&5X*fnAMkk<Q58|NI
zzcy*v06xJ8X*?P2`$J82Ao|u(_Lh{$^=l;86#Ro7cTG%qYAN|Tf|R(qxsHMSjpCc_
zcd+M0Ms&USJ+@Ito8rt!Qg^S>(XwGF<U#B145n_<5Vex>OeUM}&!>3F5B7|18~7n!
zXoNK>SW2LlIhF07e`kICNup|3IEgKFp^T-&FcrIXEX-%abQ+<ZltZ=(+<3;^R$I*T
z22k85U?a<fEsy@^9lKCk+h$i0h2$i~)WaKEQuM}y|1%8-jpm{WG*qUQ`054;2u^M!
z>c7#Lr2GFRw9(XN1K7kj<%0wOUm@tu1f83%mnTNO^!`#^;#-xm8&`iv3928Ez85Ju
z<&Z(Ql#JQ#0;@Vbx|}`{^t`3HUZK0e%4?=Wj>+gGTTQ+r6S?W{=s9K+mVr_I;eV_4
zY`k!!$velczjmrx_T9{?+p!C&+nRPo^5)V@lzNDj+C@(8oWZC|iI5Sc<%eGQN)mJC
z+u7OK2Bf-x^mUJ~5`~wD47nO4o=Dy<uh*onUAufV;PkjBnicDMQGQ*ypD^7WW8h*~
z!wUi{-%rlHZ2hUPJVNS@?7~He6S)C}HZUKB;%CF2q?5kVZ(a_EGcVDAFFG4s^|dgB
zC!BgqJxkbf*rOl#hkH^sYu+c?Rzp$mwi`E8CnJHIJp*cvm(DrRNC$CR$ARqJFl_YE
z5@_vZbw0lOt%>ImHtN@rHaO&r3YZOvO5g4Qn1x7jnEU3L9PmXjeCRj6?<J4j>6o*+
z9m!Q6h&e(-*vA6<SK905I5Gs22gAJAh1HYCE&`*^tdZ{rc>gUZ2T*d{Wc~G<+7gq5
zSzKn49B3x~=nP$&4nhOS?&C-zRzHJ@S8t*zL2T&nv{rZ>ZaIN8yV(>#t#+1sR<a?I
zJ<q+SX?Wqr*@FTq*R9E`0n#yi;&2f;jvJ(px8*mCG<@$RWv(excM@sj_Wg70)6yI4
zqI-PAGj-n}9MI$_h15=w)Z<rvv+l}^xV>_LnN6L00>m3Tv44}i=pwHtKaa!hVcoLq
zn4~2kd<gl*8^u@zC2Sbn%=6nVBYAvDRpb_h5Z!@Vzm#JNv3`|pe|4B;ztF!0r_=vX
z&LjkEZ|xcbT!GU+rI&mkZTKOTq>vCaO+o2h+C0MHa6Z?G)w`R&RO`aEO|5)HvG<)2
zKmjR4RVf@!2nyp5S-KQ(n`sAeN(?>&F?c;NLifgvaiuogSh!y;KzS7gi@^Y8P?{jz
zT7`sOamxOg^IlKeaYs9cQ_QpP9>S9tWYLzu57V0-v46cpI#c#!oChjx(qb<!Xw*ZT
zU0fx{qcpzM{L~e7ot(@SUWPEg%`ei`Fhp76;^+yR_CxG{kcs<z(#3DrNFRjN&)0Y9
zWt-*Fy!tGS0?aCOLtS1{+j$dK=1Xqac${EnsKoHn51Z79CW0o!pjUwpTbs{Vb#_ex
zx00;#3SG|zM0LQR9YxfWdBV)O_VUK2lEQhA7HC5q6fowVNX<Uk1tN8z{Vx=vw40Rl
zg#vY2M>uliOXyvZq`6bZw|D(VjqkBH0BoHG1mQ#|>f!VcKN@5FSkkQvJWq90|Gf)u
zk^j_AM9XZXI#kp=Y9mEZr5xVBw?{h5L{Ex3(COX}O%W`8rv{&@ZrA8>JUVeo%${4O
z`xX-7pxH#JB+i?1!cz#rkg9<qM&+J~$CZ^z(hgcdo21PKIij5jD_Vv%W83m>jCx&A
z&*P0UUDE;F0_nbyX<pmlExu0E0^B&g&ENcr4ioeex>dQJL!MAPeg+Qn=Lumb$1=B~
zze&SvFz=%xMQlivbT60q6fxOKDzzV%4(h*Tjqf}5unM~sf%~rwkLZsZ*!B!f%mKJ4
zvAX)b;9O%n;&86tgWC|+yw+G@_PA@JesOmPob+)W`Rqf|l)FN;p~zhw?Y_vHi9ekm
zkpj^E7P>{}&t`wyb)!lwH|3xN_PF>}Ak83U1<F~)b@_8OVUWA>HW_DFu^a=UFwHwv
zdYpHP7moX>Q}2Xg&=aKt7x715?dIJgGfeD=O<?|W{Cue$3Wol?o<-?!a?b(MoQbpZ
z-*kJ6gGbQM+PM0jU(EB6W;Si^CZJ)iQ&c?(9BC1M@D(InkKF&*U5unafF8#D&Q9{)
z$p^L(QiFay;ohob<!~lK?Gkb6BMCeS?EP;lBNHE>RHzHB4iTHs)@t-F_35E*Oa=Jx
zYlH>cKFg-+2NC4&mLS!eNBHI6MNLiA?NreAj{92L>Euvk%k4m4SSp!-G~u2UO(xc&
zN5Jh{&A}DUzFC5OpSda_r~zK$r!k)AJ--gWVIV8b@?6M*Na(`<MBUd2P4BzNlp0Uk
zKJ9AmcCU$QV43QMw4l)V82t_kN_x8&&H-Cc0=cLbW8fj%4QfOX!nG(7AOkpWwhjFh
zOhri|WsHVxa}@*^xPPPljLQmJiR1jhZq!Y&zOT>s)*|M8^8Mc@sv~sI^l25U62a^L
z)#fpCO0)|?cn5k}Wk-;Z73!5Jq+6xNj{9n>3gRQDBtyh09|`{-wEyS^vfaB6W?gmZ
z5|es9H<1y_9%to%SEaF5|0DFFhj5cVib($t6S=jg#F5yZ5%U+7pYheU_gvQ06efRU
zi=M07YRv-gn+3OS99LyeMP}x;H>Pe2$&U_LM3bm~(io-^ojwZX$1r3Yg1_)e0!B@u
zmE!GKnmu0QS6P4Ns$y!%X<cGM6x%TSG+xEL)j?ulkQ_gmu~!?j2QTX9w!utNgR_U=
zb}7V9i&#C9j!U$jIFXu|>5uPYSd%YTGwW*Kjn{}d!6UOs0-1emeGMjtJD2j4AaPmg
zwf7oGmEi`n$J(gFyGm_>iyAd!?nxS`(NQ!hm{Fetuv9-?N2P>rys~?`*iv9}xaRdV
zklCL`SaOLPnTY7Vw6nZx&tmjpseINz<@nQqT|&)ifN<##RK*b?mvWWUe(k7KP}0wH
zh<5>sYfC65T9lYTSR=*MbLhASA;ejs(166mC(^HG46>rLS<Q4@C9eW-2X$|<8XSK;
zLx!Jw?lfzQfE#|rg(eo-+9VCCo17pq2zjVk)OEFlA>3|l_9<|m9q&T;@I)I*I<HfU
zQK+R(F4X=S#qBhxsg57#t!DOZvUJUjL?@uP&mLV-;MshpNm#b+obHh9?jFb!pd#+N
zi`aO8lht7&KD8J}?^IPFT>gL?ix|1v|38OVfALnllXop+9!P|nHSJAA2?>~3Zp%ME
zZ+t8>lrbTK18E3bWmpx};>a?XIM<yp@jd#iE1QU^WAwmGI|Bp5tBTdee6{TxP-DZ>
z8<vJi;(f9Iq#RG#b5x81!Cuf-J0)H|4sFYl2k+K3u}}jN8SK{D%#*SzFQF|kDu{k!
z$Nxj%?>rro$V8is!QH!{p>Gw>{{K%RDY80-AbJ|)4%zM657DXozKPeOU@odQ2?o$$
zBxQ#R@z|=z&<tblD{av7C7EaV-`8+*`Wof&(9j>ks9x+bF&6jjS!mA1*o{DP6a)XX
z1bvMI%bwm)#mlF6@V=!DTJ>|$-x)%<`#!*Lyc#-gFm;+<F3aPVd7?=pP`mB}H%_qM
z-zC{>-4_!_oUq@O>ZJsp^Ju=;)2-AAg5m(VIN$d+qx~<PtxKn-$^!$NECG0vrV?w)
zTpcAnt_F5w*`7aj=(XP?n4vsae8jO2+;T7*Hl*hzwvn9QWt?;*>02WxE$vGuw=os_
z5oRiTi%5D$4BtY;gkpx3*^$xABQ9pyp^}!E&d;;!5JdPqdv`qwW7y`+-nM6byu3hS
z|9T9TiNEJy;%IDslmT5Iu%CgpC&*j-LLbMLs-I?E;Z!EhB;3H0y{1*qq6x6%^Acse
zS~1J>DjQH%5jk?I(9h6S6GMvVO6rjwzZ$i$*UA}@SVm!d5c;Q_BS{!=PGlAnH8WCv
zsG<3m4|ACc-IKl06(zl7{f5jJbq#01P910)UFjR-EKFB_?f5qwBjqe@aSZMWh7Fk_
zCqnDyUGd;vIsHYR3`VC}lR#TdLMJl!65fMy2c1O?ktCDwQU57@x&E_9q?sV$q=oxl
zpDYGo;k@r~<WF5k%3xOv5+95QP*w4Imju0aqk)P_Ie*lH4<6SRU2J3n#I_(S#%iD>
zd!ts_*w}D%ceo)5-&2LW>z!Bjv472IVr5hO$cUK%vc<rR40pB>aZu`1=Y1c!bt5an
z2b7eEH1o)>e4#n~(saOB0)XtQm%Ng|NGIMCUR@sG%!U9RCIhIva^&f7BcCW}o!3Xy
zl)rqMINoyZC1m;{rOHHYiAJ?8ROT#De&i$_fgeEYnl{|P2GQbyi4g~Gz$Nsv>K-^>
zM_^&4$hOm_T7MR-0k#I~?$6^$GqM*3iUhBW-*}_xb5a^OW2fwf0yl>LH#?TRW>2Uo
zNNXg_-(oZTKv9MUr^Y>dBR6cgp%ZyxYKr3X+yzA3q4xR%vWnwn@xdQ)-cE)QRS^u1
zqhldW@fiqi(R8Z$!kA`Y4vBQxwRus$Tg~v`Aj2V|cjo}Qzh!4xJ<F2yB`=`}qLF(_
zn>I!weOh}Kfo9b;%sl?1VdkDXh?qp1FE(~QrnJNo?kF)v@sAt$GpO)s$%XH&B{aZ9
zDnq>)f7Fe>IQfreU0Qp|WOG^Iu5)I14X|db;e#X@Ev{`4=BIBGnb^Lj`9m<-dD(Do
zq45I;fT;vI`9+V|wMRZJZ@L8RGjS}xK_laNqxK%BE^Te^g7^r3yr&4*SV}P7hmcsr
z_i5ne*Fenu1anRzTbG-_VbQRx|JAWI*UCFKAs*pV0onU^VCxV$&;x|haoh(h<7wxX
zpT8oR=4WV7Xynu^HO^<}XCkkKHY2!Sur9KpBc68-(pLH+j?qIrP2?IB{)@eZuL?;g
za%$hPw=!qUVhuBDG!bvuaz)z%ecSVwrv@!2jGXt{vgWfJJ?2v3I7r*0!DwO`C-k^m
zgeRrv$+BI_$6jxPWZ2F2vTA<TiA&!URsW-Nm%M$R-Mc!XJWe&82l)BoAbIvT+~2!@
zjwI9`TzD5oZ1{p|TJcViR6E_9S7LF1>`02HFz|0`tI(d#yL3X=$C+)gD}3DFl#~P^
z$CGz?cinpRpg`UvzUph}GN`=HY4#+zw6vM+a(i?%Yt+NP*PuB0C1aU*!n%qL%xkV?
z*uy;l)<JnyOeJz4cx_4dQv;#yzdEcZ9Uw!{6Ox2f|M&>3Y^>)_4ERDv%j>WtWy9*P
zA>Nfdy3q7N-FIe6AL%k5!>XQ-6QwJ(hT&lH>&W$4V+O0pIRx!WEa#}X1n9xVl&qXa
z<g@zaWk2dbvzFyz=;C>Kf-zLR@E^deLoeSXF2G#<jz>YAb;?{GPC}XDN`q!zicy8$
z6Mb*YJ_UD4m@`?6jyg86yAUQCA}A;{Cl-UDIT~V;Z;4brA_o-SzPBycZKZ&eMTv`@
z0B9z8#UDl=`!OF8F=#e=he%vjQ=wm3vs!3YljwiEWy{1{>z$!#_a=Um02ZhV+6^@b
ziz(*y7K$#P5?8t^(ZM|yv2oDSsG#fmvs-3CT32^j2!MlxBE%!y%CtB`rQpd!QGi^f
zQp;{=zTMd@*P__&n6ov`p9i@0-*mhcu9Fkr?v^yc!6F(e0}FfxJ2+CNTPqU0GZDrH
z(Q}wZ<V&=<{*-KE0OE;%Pgno$N#iF#c6#y5{ii;Q!fu!R>q~Ub&}wKy@tTn1IR7u}
zXp#YXuCWxh;<bIJid&8$^-fXz7`fr;cai%9ie)wyL$rF{KSng(m<gHif9`+VVQ0_n
zYDZxUqi34xcVy5*>u7^Veu&m^!Q{r3=rb6kLB^=*<)M036dH)$mk(}lVBe#wuhh&e
zj6c6=p4Us?8Bni-_}{ZnRV=lR^zw24CY&iL*vsDk*Sso(`Cp1<MDV{YZ8d{0HpV*p
zcUh58DT?z7+8KKY-z^K3x!uxZl}@UNRP+WT7TUe>uNk$~m6?rH8&vFCA*TTol`p(T
z{wGW>>jWc=fTeI2@}mx7pVo<uv09DD6eHu}x3;7Ntech=68%=nXeQ0lzuF#8!OKHM
z)=iHh8Afq&oK9r;gP!GL*3{4Z-OvM^fJe3|R5tIX_~p#X?re#%VjMbNA0@4uZXn~*
z{_EY^Ey8Buh+UgE>PrIjH(1k{q97GJqN>ZZG=T|xqILqgU*q-NmMR@pOXeEGx37Nq
z$ejt!e`TAOndoQ3cQIlldze26?-;i!ziH1e5u)n;Rx?UhBkAm16#v%`I3lJ&Je&_d
zKKA%h#VWC1UJy(fJImElvz%%*(`@lM$o{@42*;lZ^*^EgNzCXW^J&<173CHe+x{3C
zPmqDwp1s=>vHMWDW#Ns`;9iBO=ttp+!#GzkGuj^^jrKIo#9{APgg(xzU&W&Lkiz$=
zl>;xJG3{klu2lGdX176ANB@s?pi~PtGBiNnh57V+SRV;$K@XmZQNoIl1ev^^z4OTP
z<iA}0hP*n<ok#NVqSjndtZGj|wI&~AICK^&(o6WrGr!XbnbV~Fq7e8>`XjHvK(7u_
zp3}hbBC8PNrM)s?*8?yw=JL>%<1&Kffhp=k{+sQl(jowF9fAqR`mUvszm$b9$7ybE
z{%7n)r9C?LHh7U4d@&we0{!EOMch@6?sNwTm-62wF&d?;{^!A*ri^xVoBWmcdpoCC
ziDE|MKq&Lg=fht`=h^#3RvJwpc}d%$l+Rm}RI;Ts`n1G@Y{<{O{4DVgt;iGeiUt!u
zS>7Ramy!}y=)-j{^iKBV<mBFM%{8B)d@b)Op+%u#k&V@TD%G-%#*Eaxj^qU88m`Ys
zV&@Hnf5fezouK;GXpET>AMKUB$0AHKQ$>x-W>8nyD5mn$xG$LERCh&W+G1%Ue58WI
z=z>w#4>yJO{c|+oj~FGL0fq1F7ha<L(E2Ve`$IJZoc`TGmd?o~eAy`XtW>)H++z@r
zYP%(*u<q`{cP0DnDyvdyGY4e@*YFPZxDF>&DZ17ipOqTRYH<k_>d9O_8;mU&Hj}d^
zzVZg;g&TImE|=RXL#jN<GC?qPTgi1f5BKJgQ^lVfIF{NA`*p#>I9vuz2D>g4Z+o%4
zjDw#<YCf@MGm_U%Xc`J@tG~%_l42qf<K0=<4K<(TvKjj)(H8pp`k0j4`>;T&?=Eq6
z7S%nYnr<Hlq!JJ{`*JL|D>_yz>S7l-bv0=2u5g7HH=o&-rm?XR3l}7wBnt4mKxX<$
zEh`d!8wj8A8J{@ErE(n$yryKM46&kYXZJ&}HE%P+U6bdPFWp4H_DOUF>^Gs!pI?wu
z>#S<PDbJ(AaDFGEhDXG8*&)HWQ(T{?SClV`-%=K!3xb1ukUS04s`ykN`qDLL+o{{4
z==wphfx%yg&v;;(nIHT?4vqsy+}^)O7+BNDL{z2hbrJBRp6clpyB^<1bUGQgFIJ2%
z5p_w+O!)p?r7fH@`cJi&$NR{_b$HpzZEdchveX)GR5QT%bV^xS8?cv@w96pbN$!6i
z58Z#OG}EiOWwl%lDxc+T?#E1#IhR8Oq21nEYVKToPu_E>YuZIJ?AZGXD6skqLzi>`
z<s9C$^=F0%ixjy^c|=+R`W{1SuCBN=vZN;f@29XyX>}!?Mnk^eSQv3^>PSx3j33ph
zZz=uy)fZ!jSx4I?p?+RZ*L!$gPi;s)74>&W90iO1VEM|{`6I!5NW5RUoyCo*xWpWN
z&t8(eq8q<K1U%9ucLCg+uzj$W06VTIM)Giu+6E|E|376nN2PJU!LYhmqr_ATF${S2
zvUi)6>cfwlIcRgM07Df}!?8oHa7K$g=<CBo-Mw2<?kfA#uk-qLJQ$DkCGpN9&(Ymp
z{{meaHe_`*trbpYdNM#*kk``)uIx9EXx<w>kFTYX_#tWM=Qb0D|4g?BV~?=1w0*tW
zZ>oKKGSg$K%4CrFL2-j%h4IKoOxc;3MGf-B4-`JG0TByEj#vOPwGjg5z6<>VQ^Z^)
zY6tV3Ez-F>>RzTY{4WUL*Ey>>z!JM<zwZ7!7Ev6&#QT-h@73PV<8xlm6BXVe8C&0Q
zmcpH8mPS*3x;(_mc-aUpLQN3fpWZk~qN|iy1n^h90olMhOrQ%++NNWZ16hMrKugH(
z>}<NW=etP_LK9X3ht6tRIBwEe{xQORcV%7vEACMn#~Lm5TT7EP-e7_~jDK6eN{J29
zuL%$5n*!uI=#v4k>qWY-$dprIJ#+4hjFWHh2<E>+@3|@}UOPf!B6KXuG=0c$rK`Y!
z2{pT_wa8T1$@l*ycH=-O#wd(I_*|Vr6J`DjL*i#T!JkCM=CwWEL}iBWYhwNSuq*`<
zQCYZnO$30TI&D1j>Q4Y|a2$TsQewm^&ph9M_wnj*8ojJtA)$6o^&=A108yKeo_SRn
zq6LBgrEl&S=rx&6yS&EzfvAjwmZGSMsU)I{=!!!nhB)jJLs4;doO4H(#^##{Cc*yk
zKD455Co3u2FqNQg1D({`S{Th4uX16ChBxT~mT+jMvQ<{D;MzQ*^r?Adx7gYreK=P(
z?<$N54IwU@cYb~hm)8SQEkY+ui0SUO0V5X0n=14zv+-Bs{UYtg=sgTRv*d2@DFGhA
zJs>OC=Qz?(PekvwH&-R7<*Sib(9mA}*aBr`s0OG+fw1XEen3=%4$Ve1v!hO2M9KNq
zG6;ITl}1*&x9}K9nY%f5R}*oeeF=kg&8{dvf^)(^R$ory%{Xa3HU2xs0O%E!!y(b;
z+yW4Bn++)^f%iBqnqT6Up&cu8`Ei1>xQv06gM-Q~VOZS*H<2qJV)FG0BcClV{4hSy
z1**x7@(K<cNaGznj0^0*jsHVe;|Fv!kpoJwOWhX<ltuRfc6SCTMWq_8<8I5!TeCJ6
z4El}>jv8~_RSgCq)=8O2G;LEqTVO&M+7$M<d(z1_-EIFt1@4YKA<DeA%ewJ*FVfJP
zWT4Ns7K<GkvNHqWuZHCf{jR&8#zr#JXue19_P~wG4ln-n(i(q2GdpJM#;6g<f($Kg
zW|RLZ@nK-#9Ck?~EQxpo&2ti?5mW|>JAXT6sc>mtnFjQECV0OQR5B>~gUV)=^$z5R
zG+t7yM!8lGZx1QF|L2P&qX_X*8Dw1XfmQj#sx)jSY_@E#kq7ACAp!)Gyj%iixw0F_
z?Q}O@cs|1=@kE2Jbgc2CGU|Q%H<cCTn}qU$Aw=}Rh1`Lo;!7~71bV{Ep$lhqPLlX?
z<0VQDl(Ie>f899wU%~tT<nI5C97fyMR#{l8AXz}m=#2?4USBSlIP`jF^gRao>=VqQ
zr?-=g+(TVt`=z#vw@gTzYcmo@CulM1uNFJgpk*&#BD~7U|NeS?)6Ec|$h)}gcIiO#
z7(sf;GH0EoHQ5uo+$Arlu7JIm32>kQsbjW;Q`ZhTe_WQTmP!u`6gJa!<sLQnuRW`-
zpVK33WN!+`$H_Ts`jMtL!-bp1q-BZ@jS8M(qje%pWN^ct*SUkkyu7%{=A-j)ornv(
zS*Z=ME^m18y5&`^)dfs^C+!Elp2A<Z?jnZoSmR^$QS<@VAZ#yeJ(%#t;%$o@*$h9Y
zaVJt32cjQK`Cf1+w`dXefBTj>fMWJSW3G-*_4^JMGj;Z%OP4j-xghLA;zp6r3I2c^
zFM=*F@D%vwXZ?p(%s^kO$;gw-2?h-N7DCe1;o-v)G{2NDS^<Vghlml9LfJqA+X;0-
zE{rk`<yz2*f=8eU>n5aCi^wO#%x8U;Oj=g|U~4XR*zDgxg5M4D(M0|UH7>KWIm?)D
zx4Hdjv71B=U-<S1@u6|%6?|aUM#@DCDLkBwJDpr)0>HK^0DX7&n3YHoZ0yyCqUk00
z8B~t+Fg0AXsQO1{H8|YZj~$209j)xGl!SPVPeML{A=m{^!n>v-yC)$6P$FEA0SJWU
z2M1$7P;70T)u-*>+^7De8CLeNte5QL@h-FbMJGe!x%>-@^v%nlJ=FRTKR{s&G}_k?
zQLVtY^z8wkip(CSMnXNfR|$Rw+BZVY^DJ6Z<}|ra8$Z3ZXef|Hqy`f_hXBld6UKY^
zA<zMc`T<eLjKd)p9PUOoi|dCqISR^T>fQ%*Zy#e~a0l9wvQjwTz5GwQA^_U)sGG05
z2!)#YZ%a_<eFd^F`2i5GC+x*1B7e+bv(~C_i9N(QHl)1J8A%O^y<ApS1H<r$XNXg1
zwKn67z~J7M0s^iPheHn-6|w8~sKACIUJZ;o7F%IIqz#M*%p(aWRDf^1gKZhGI;H&a
zZf{UjvA=CDRFK;!G7KER>axTtI8#ulLAuPQMW#ki{QUw6SUkmpt&SaS!Z(J^gd(IJ
zH@C2|FD2SPv08#@=_1p#R{<1c9m2N^gIa{Fu4st<81X>Qd^ITEBtNr4R6J`P^xk6i
zyOV&P?K9$jKkngSCZ^av30Ss@+t`jD(65zk3&v=#!rAK1)ND+MXy_s<uUN}SZqeU;
zLOz3~v@NxK|NeTrvJQW(m4V(xUj71~%aFIeX#x|<M2C#CW`he~4{O%WAJ6uuca7yk
zra{kdY`Sk)<(W{db^ROC8fZ;M5iA4K6(Tp`Vz^&v(Fd9a3Z^IGL4#jQt#-@74>|IG
z1IPxsq*~1<g4qP6%=goC`jWp?X&?V!>8dzX0MROL<p;^km1_?hFkW`uzvJU3X_oj2
zRadvc&y*h~T23-ap8$@VZc+p@w&mg?P)R-lVTc4!?7g62aaz2oz$O}1J5wqX=EQo1
zcfe<>S+0d<l(mD}pMZd0NxFLCh8ig$Mlu`##HceNno*LP{U5aR!UMu%f=fK0c<8vr
zlSWEg;Z^bX78>|9F8>*CrC#k<bvRRS4yDB#SK1cAOxBOI=cG9#=+y*p1tQTmN!GIb
z%SBpPs4(~9fxb?oj-HT5f(nX@CHutcvmH6L>`is3Mq)yO8EUnC@v`~y<2vV!)i==V
z0%21XrTdO#^2ETXZ#%HpmjkHJZ7mmX^2YTn3o%VNUN?n?hOT#(t5D#LzFm6}bwD0J
zC)z9aRoBxPcgOAZ%DQ+tI4?FvsC{86OuVcw%**@?|52;K?M?^BJtbPI{hu|0${rtH
zu3hIOO6cA(|LQ11bz6+)zn=`>Z#SjRsG_o&W;Ad}PANbNNPNdZ^iJFSjVF|&Fdi@4
z^~Oe{TphBi+RjP~+MIPAoX!|k{6l-F6*sQ^*+)r9N!GXES~jDkVBZ0W=56pJAL!5R
zbL(OaZs=C%Yr$mYLVC53hS}+e#v@R$Aru+%r(m?y+QjCF8cN)XV7*W9<)Lbv<5P&r
ztrBbk{@CA4NVFX0f17qwd##!h=;*(nPLugcrUQk>hVAG#|MH$7E4Jc(9bgkcz#ZxC
zf4*y{!g;xQ<mjEJn*=Q#?+7{U3FdNAyir(6DQy?(JH&B~*!h_aFYp`*1KUq_7U6ez
z=x+hpt5Oavdt7n1?h7PcbD*j<&&JjsjBiFu%zYf<M$M(ForZ=BN3G5|B~SLE$KM~H
zsM@JVd;^k2yVJq-i<Zis88@Q$`ihQL6Y{o3E-gG{KsXSsJOCl6Y2LC{@okJbT^f0k
zI)W8u3D_)cQz4rQH+%;ePR-+=j9V~ozKnzGBa7qH*B&zaq}im~W-XsLBWLnm5<<);
z=S-N)FjLnTY&MiT!2X|S>Ox*JB}f5l?3B@abqPp)Da|SgRHVHVhv~nz3yN3i>_SDi
z?bWtC2#)rLYjp)i1yDd=$xM)k*aLWjyFL5p-*{LP>027-2YZLVwtPeaEl8Hj#8;5~
zx^I3Tn3W0;{;vn;WhJffU<XG|U*@;Rvn#uOqvpprZY4)Nz66jf$6|8Du3MMtUVEvH
zP7zn19dU1Yv)e#FvgwQdl`S(4^{-N`k;y&L)*Y1uj<St_oYn78`d=CVN%NmNcjvt9
zFafsg<R9Lk&Rh(Kk+)Q0Y~rSIra@J%-mW}*P=3Y#>XiGhJ<atjOAA8`XxUfm>6yRt
z$Wu(SP9{wH$EFmhE6VF@OY3S!2<0ABg^Pl}QXT9)hZkzA^YJ5j73xEp2LF+1llM-P
z=>j!w*r$H&Rpq{O%viAGNwg+CA?W;^#YM^)#*`dkJzOx=l})N)fZTtD5j<`Y?Ayg_
zFRgb%U7A~@&TwvzL9T4;Jk7o3wfeyK<yU--<5<|!5f?&Zvsl&ZOG~B6`Fb#jeT|_k
zci&MlTD<U;Co<X`(deejDcHz2?)?o9k=aDJO)zWV%%mg!Ut5|Y_lBZS1KQtI@b94>
zQ}kc{VBLliw57+85_iIB3L2LEVR?Gb8V6OmvVaKxm-_;4U1jY1>{LLnShO(Pr+k+%
zS{5&02C#W3%34dwAYs9blfK`Yf5yLogR7}xS5~+YwCc$(a=UV4_v)$`t<ZzliuuQz
zSKecCNh*~|iv=hiw~hK!^VgtxRn~sa(@e?rFihg&@>{!xiRK#8UqgMfHSy0AMsty9
z$f+Rj5&zz=0+k#uzzZ`unV*?E>IDPXdR{6OfwaZ`PE>P;$_XDErEnIv&H{)sK?E2+
zb)RZ;gnXDfXF>1|sfcmp%CX>yWlMx!e!`o)5%&}iPqD67Yi@p1LGnbV+<-5Yl)`t6
ztqOKf?O&q=R<`_`>wB}&qdQ<KEzkzZEzf>8lx*K7`etCiDdg!$_4~lZM>SHU-1mW7
z7fAghDM*2@{tC4kL$ZaU_Ri)3o&NBbju(-bdkb|i%?#SdcdlUxr$p0@?XT`GMlOeL
z$O7s5wI_mbyknlFM*h(b+gFYP2Ok-YrfAp?4(j>x>{<JwaV*y4RKajHGnH*WSH{$h
z2qAmbbV;0$(9Gtm&II$d*r-OpKNzr`@K7i2FFZq#uW0CS>k^k_m44GJ3bnh=CH{~@
zlV$C}(@3P<yo*ksiX3vuk8HB6XL)-y^)ssHa~P4NGKE>o<ARzdWsNiD%id8Rp9ZH8
z^L;|G%ly>ee$u>B1Sv*6-#dhaQiL(c3z^3A^;zh6a_nz-8bl#)pnK)RcL#P1ikw0_
zV+8UOGHxnFg*i)?t-P2%Mhw3kB}e;1Mo5NZMMo&N$x;%(#9K-7=w1N>x}eGBqR+V7
zqh*G^U^F2R7i_1*x5#J6b4}poJYWr6(gYZkl~~wWyj5Y27>HqT;pdf~E=}rZ?O1&x
zWc{NUp`GvlZBRY6pqo)jQIqZ0{7Us0Le98EMa?o~w@()S)NT^m$wW-}*Zwlah883a
zJA@^<jU85Pvgc_6t}hwsEAJ0A)cAIy_ykShNmN6GXK;CidqEKMm%Q41liM)_+#>~z
z(mczdyu;bIK>(KRNYIt+2r`N!4)cAKYY!@%G?XaLVIpJ3+rOCOPf~T&-@&0Ne_Q?{
z3p<Yj`05UbaPSmJi223OfQd0B@hZ(31B#4AnT8YmfHX*|Ci1SB<}tt!3K_(qrJ$K6
zY$urZ@9mOnDnz0}QYMD|ZRk#*PfKREjZ|4`Y;-b>SY9@pj;kT&b9L|}9n8QgEJ&9s
zT@<C8v&LGem_Y)C4%!ql6Lj+vH?6YH?>yztBJ2?-eu#$sRt%gX=3`y1_E7e%6x=7p
zP#Ou!`1<uhwjlS#0R=Bw_Xv$(L?KFcB&ztQGLwOCCK#{?BqK4h2td=u5As~m5mw0Z
z0OW9#-y*-+tYITA=BRbFv~*q<q-fZMBwGIGrDtfwB;g-#Q>sE(w5?NOsp6INl9X|_
zo~GtQTm$=d{9CFgDG#EE=$nd<NM)v$T5vl^Q15LUX~{ebVy|Lf&uiCI_3b<fR5j=E
z9PLKwco|r)RO0z`JN%LSyt+=QX4J(6*j2p$d<5l=Nq@B^J9hv)g>WrPIe_`(gMPs}
z(SBQWr!o6L`ps=eeH+r+7BX4`{3vm<+#7V(C<U@Kss%=>KkD7TX!xcrjwr)cR;Pan
zmJYtFpw<lt-{1r4d>zF5#G+Cb-U)4gt;*5I!s5T|GfU}viUh`5B{<f%hpmi*hlQ1N
z_u4|L)15xS-vEQp`u2i?-;j!$0r4N-VL(`gAj|oduoX-7=jVpXkK#lT6Njm#-OI#{
z;}THASdU~FAnFT6FGTbby16B@sfv-DU0d!IH7m2zH1}OQyKWC&;JU<6X^M6K7!?)8
z4PN}&TUa0%(#M(~_xe3y0ER1d*r(-cP5{RC1w>?i$oO7AH%$Rtc$uQIR!8NQq<%i6
zh~1wu5x>vE|I{gQJr%%xhk?ld2M}rlvak0WPAz3pV>jTV%2MMPzZc~oMz_R`=py?U
z+SUTgFU%Io?z0QuR}p+Ex15n*Me@7qpRnSlpOAjtV`<Z;?ZNED00}J5Z+Nj=U`4rI
z*lA@{48k)=4@-9d+7h{2K&#BSXh5jrMYKXxTh-7zm2Z;16v2#6;pqIE-h{7v*24Zx
zj-lBR762*f0@~lH4or@EBEmKqLA{eYAgccc^xao1caC7}H7CJXb|w~b`xY_aV{mIs
zqCJB2(cxPVA#tnzv+%QWl_0DYJAh>S6aZ8lJImY@zGI63(rS~yH3GB#>-$1=-nx7}
zo+ojSXD+x7DI*huboYb6grKh-4}OxaUS}5s2j%u+2x!3MO`h18S=K4x%Ol*Y*w3_o
zn~7cLQ9gW5L^C23?mqtV2SyAiubklxJ;R~wPBuKhw#m%^0o7N|Pv7Xffeug357;S5
z`Na=ufFD)$JYy$phbU41*eS7NA6RU4h$~?;#ePRr8^loDOzBV@wHEZ)A+ZKI!-F#-
zdiT6>*SQ*VT53UtBxMPrA1xTdiLqW}B<NbD!gLGt!z8%rE3nGIgP1u9g!ZaJueH!b
zdwK`uzr#Izbr)#RAWzdOe-%MMU~yqrj?OxyUXuyF$j^$jD@nRN3WUzSph70`YcE=+
z?WRxum`3EI!YUN6+CYvba+S8nWl7|YAT=`^SHE%k8d>|SQ9s=YtZ5gU>}xOE3E1V@
z-P397r<MSSGEtzuJDPHSTvAp9dVTRT8p5_vcGH_npAYKlrKa|WK2RtA*0mEn&gzb;
z%IkZr&JQE5E!mnI0%g{G3P)4TrPop6Jlz>Y_YR(IB$uM-aohmx@QnQywQS@B5XzuH
zIUT<~pImtVj5L0=2!Pl_V0rL?P)Nq|QH2+m3)X#YzWPq8k%c(HMk4@JmujtW`!fT6
zvlPg45xUlKspe-lyTc1lUoH%>P)u@;&|`V6{cl~gb#E_beV&M;>UVJ}*w7FkxCib8
zsy%q!=63<Uj4rb=gYE_#x@F(*C}PwUZm<qeX5^r*zCTY?CN+nyPV9%Gz5p<z<rP6w
zcv~2P=H`uE=J~y(hnu$$+&cOg2inuFbS+gM$7<m9d$jr{4z<>9@xZGQV(oUqO9kHU
zDEr^&MC$6|3Ps7l)9Z;pV*lZK62%#{D)OY!4e|JhRt>&BV*;$O6JuQqa4}=SfBP-D
z1?i#A3QIW|6GFOahxsXH^q*vJI}693ziK_+{q5qR&dl1;TAvbn5Y7>^!RY{?$L33M
z`)5L<uCuRElGDh7ztCOYb?J5a`+Z6vyEwVD8xjR^pw=xZ$xi%CPe{$Yko1wiIIx=A
ztS@ipz=N6SBap*;K#|~W<T|=irdDAczh~4P8YC(6{OP4uLquHZ<kws|0&nO?Yijcw
zy!`9Q?O!4hw~H$^ij$jwwMKmYbnQi{u&yr#?k@Z&V{A*kaLV0ll64;a7p(xZpR&V$
zat_xd`iJvPVN{n18Y!XAexPM!=MJ$`;7P-RQL=9Oj;yTxc|}?0k>m9x?X(Keob>K;
zR`7j1HZ}FxW9&5uQ~=|G^(b32Wn5d3Rz|(9`p*kqPls4@gO}!X;obQcewZPMQzS!I
z<UxQ&l<_A0%hmXrRoC!{Yhe~@pC=f2_0&2yyRf5*Wt5`>A%X1#5TUWK+L_|EIIq9a
zu`rrry8JtLeYYodzBs?QrJE);9FKybzes14Mdlq|)p;F~eY*ZfutggVv(@URAieWt
za<9(;bmlhFJw%WW>do@(Hz_IcJ4|xpqfdFRtF6~o7B+xHCwwepB`mPo0#zQbN6cFL
z%O@e-bU(k9-_MqPx_{pUzkj|L3j7de{XieBw0Etic^eOTG>vl{0U0%xda2cS_JXPb
zdO;pGUZecV65WHu$Pnh6D}0MXSk+UReWyV@hGe1LX`BEoL4C0wE25bQtSMy63KVZJ
z)d`La=$%x;qBpnrEwF8@00N<d|AUFE0n}12S{z_xWL6uLbh8D?AWmzcO?iXfieIVn
zUJi-#xvl}~Z#Mmz&nP&oe5!W8XZz<^7N|uXx9LT8s>UFfSm8hC;J2oLAGQJXfak~5
zHun?lH&!81t2FH=zM|h$o+5so<xDS55ZGYPq&<j&AwFPXR{U=}#847pov%vgL|lpF
z!zeOQEQKB7gRbOQ#4m0K>rb$2_RV)h52}cUw$Om1Q$<y3@C84p{F2=f;?!hlkJH4%
zrcb2sxph{dLzYle0vLD}vmO#(u}3Qw?z46xW?qxK#FqET4JleiS96vp*sbfNF%vP~
zT70d@wx9EZV`<qF+sGm<kdFqQPG+V=Z_G2BH*MWMd_Q%k+htXL0QK<|0Yf0XQp6M4
z0plrvl#62>fd-t(4dP%U{1896oM-z>4qZ|WKliq2O8Z(O%vG|uV?ys}nCG=HR5!;{
z9nrA2dYtbVP!lRms;Va*SlM|I*wpyA?L*)z$MW;G{*Oma&8ygMMo7J0uU1AIFnf3a
zy6s79w5j%;y>gqj1)^Jfv7m7uM)WHE3e|q_7CYV}DAc~OFztQF{SN2bDmu++_aGJz
zPPH5Lp3Zo=ItV55SsF`L0}6yYV6|_L9|tBAe~)6d4GeuS%4Jy#rCpD(_RWq=N&}FX
zx$zFl&kJwJY)&jADTp5iiu?mt63uUv>XzbGju5<yeQdEd#8na`W~{FWbO3cU%6ARI
zSU%u@VI8oWsw{Hlx$t@yj>Zguy4;o3B;CGD=#tNLvjchiUilTa5fgky;Cl-g%5Il(
zp!?@G_f%SU@em{peuo)8N+%9xYG={n;LBJy^!{ARExPKwCUrS4Z>_d*?PlK<;Cd7I
zkekS0k#nyNjpXAYHtA5d<cAZ%efI#4k^D&7>wm(|0C-YQZ=SL70c0b@2nkb<)Gl+_
zwZ&0=Bhu*$Yqw0PCwUO{0B0%htlr*fb<y|#ZZ<%iatajR>0}TW<4*C1Pt)6ZZ|K1<
zLfZw$ruUG2+N8s;8|*56>pW-XuBBtk^;0}t=980!E^X@>&6_;|ghABBci*SXyP9j9
zf3#X<226B(7_{?D>i^xjvgpcw;G5laU5yJl90DD+(1u6N+8iD$vjrOo=1?KrIpu#^
zY!|C~Z#j_5vU^2-&TpWK-^5a_0P8%PozJi}N4dRC6bw{*7KXrlEM~kyw=xEiCp7&M
z9^FB7HKFm4yVwM{UZ;1%P+QA2EhMEgH*3Jwquv6VEV-j%_?fZT*&T4H0k3K2zZ{bw
z8FjbUkTj1cYaLT&dsZ}EQcI<qv^H^W%r%g6VnuBXVo`(a!;pdV4os`FVkh@cKo&Qi
z!1|C&m)p+r@Ik5wkce4InP%x<*>;8>&3w!8baN#XhHtWv5=l&IhB^&;z&3V_mEJ3$
zJ;_InaD@MkJOgEbG7jcx3WX9g;msTtmY<D^L6V>49dSQZ1IiRYBJ4;sTb4(h015c^
z;kRRF;lb{%n68~gAy%y|#RY%C-k@FRjFU<t9$aIfMA(f5{*g&evV;IoIH<Qs?abzR
z3xilyZHri(K!_+Hc-osu@WQ4H)V!~N@-pU7COm6aZSR%}&7_aKO_K7%4Ib`F%p{5q
z;A0Jz_<~E|<OXtTS<bNvq9M>q{|vGt`<OVDJ;Uk0uBUlnfSL9FZ|8ggm=rslzTV7~
zj%Oc<92iyK_wREBSR6E11iHh78l^Byh6ilos3(E)!N;`^8YRwvp?A8)FwCr-w50L3
zqmGHm%36kbTW(hfZ{6{eFu=2L<NYL@2%?sXem*=M-3q<lymsd3(u2B4TGN;IGKb4|
zum)S@GI^+6o;SK?HtnWcRP7mYnL~1v=*KEi{X{mxaxA6h(Pj4?JBYME!>`1uEbn*k
z%<G`AO(fbY-g)Td6zXZCV$LT-yHz6KLmgFBcWzgq+?4Sv-q`PmoamGHlNW=6jfAY~
z=CxI8++2g-Ac*5HrTO%J>Br$6P%mYEa?56Fa+q?fFsnu>VBTr+I9x!ahV|H5Sy|a%
zdt*yZ%!kP~l6Pq$2m_-#^i~alcaXAb0JH0SEQqi~b3Xpdaj5}*nHhPNtjRO|*kP)^
z)S?m4;kA|u;!l!;*_7gV0$3uyrFUf2h&hg{s-VY(H^L6esxs;F<}ckIG2}^dZk8W)
z76I*r@8d|KZE8n^Env}|->c|Y;IyMh^hSXXT2ZUQLm(ZT({sGFExEf>u&(5Jz@EkH
zV|#LDzC0@B<Ub)_O^W+~I{d1$N1NzHB=5Vc0>J>NG3%u|jS+t1lQMBBKwFn}ERoq`
z3oJCbdIq+Y{A4c5nOV~alcD6eU-<8ZI`;p%e!t?w-0#Hhu79Arc`Dq3F<q~oAv0>&
z(0a})yGjG6cO_*vF0FyCad}H$`OS9Q#y7ik`)=RYYt|Y$y#J^V4W(W)oLFnr@r7rv
zvO+C@*D3WB@PF3_Or>iXx_Z}*DgVeNvaL(y2?I1G9RPb>d4b|_Oi<<WR(1>iuX(b-
zTUwgcA9{>0Bom60*MI;0$kuH;mMIM+P;-WIPIn`dABJz46NxNuU-?3y>2YR&f#2>x
zJNftKN?zoDCTT@bCxEVQ_7bFdo7R1Ccwp;I=CanDVK1?ANQi1L%|)OTe!0fDj_87k
zwKe*UhD1C%LNzeI=}Rc&&!_W1c*LoF&@z6mPFY0wcxMZ((E=CH<Q7Lh;X#?WM=J0K
zRvV#9kG6AcdX9ovSXq0(_>xZUqd=5<3Yw~aJ<1)Ezb<#teu*8GIFVfn`G<}`!q?_}
zJY*w_El^ZWRT1}}KSdIfcAQSQSndSn)lL<X-KR|IgY4M5QG?(6%uNVxd=5;7JZd2f
zlGo_`$Oz|4Jas!I#T8V{K{35}Tqo+L!1EK2PGIcIK#IXvcyWB%()d|IZ-Vz>fnQcv
zf{ePbrNR2&>FSVD&=8)lDlW&QqvuP)7aO(!pFZa`r%%vzML|)NGpBwO0w@VAgkj?g
z8DQ+pTBJ{Gh(!)sq3sLptERSx-b`g_-}v)>P<({k#zH16$@C{)wwu)XKY)(GAyU8_
z$o;oIX0Hh8XavYM&VTA$VqB`<J+|G1>RRKueiFVyD<;qD>rePd#0(nKgTDLQB|b<A
zS_1Q6RHH}gKyAS#*UQ`YfRj4P^MXN+O_Uq5C=GEcqZtSF^TaR|jz9IiP>3m#x?f&~
z6<n=K#~Vw=zp8f35RvG=>_aw`BWZt8Uo@WGHv<KU1`C~e-n(H!Gw25H?Eg)#b&JE%
zbaN){>uF0v*e`FNo3Y>oXd8L>W)8C_w-VO%-82mVH9)dfUTWy!>G*pXc%02uo!%cm
z>x10gI-HJ9JHx=hpjzS@QIe8al4_M)lnSI6j0}v7bqx)54U9sJ46KZdtqe@G4GgRd
k43_QQz8OVBZhlH;S|x4`*Prq812r&sy85}Sb4q9e01S-1RR910

diff --git a/docs_old/logos/tinder.png b/docs_old/logos/tinder.png
deleted file mode 100644
index 28d0d0ddb561bca80a58c74ab15b79e93a6941db..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 28182
zcmYhicOX^&A3uJsSyWtEA#^Lr9+9%SsO;R!wa2x}PRPC}T`OcnvJ<Yovu|W%R<dUy
zD<gaRz4ZQkzrXM84>|5R=Q$tG`3lj}P&rS<LIpw4`TO^f4<U%G7=mDIXDGmx!8Y|H
z@Po`oUR@r7N+PKb&B?*<oY;F0)gj1>3xaUZA!r|5!p%dFn-ByonnBRr1PEesepjO{
z1^z%`p{jy}PEP-&)n~_mE0j;~t1D6dq9QvZeEl2o@@oikV7ZT!M?D)^9`}4^gx;fE
zTlKaseVT6A!FAAZRg3-sll98`Pmq;ZX}NtUKR@a-x%Q9Av@<`>!!0ikD+&E!Q0(hU
zPNyc%id7<$r$y4H-z@)n{Y#gpzyCD{owrX#zVPi#?D?f+(^x(ctzPwBk#_A(^6p{K
z8tw}Yrd7VMFOAZK&4``(wcAy=^FYcm@iOAS506+gJ+1ONnM<KM5G>7GZfob+T*Tbd
z{lqrAxiIAaKG!c=Q=K8yI&tfqg~-z=<^BqNNd8++@G<rCwxPbLg!I$a)2mXvn>;~D
zY=GR!gy@O*LMML`vp7Nn^%MtmEbO&8ugAj5D%!;c2cPKMe^*!n{bs{O&1QNWi|oCg
z=LnF@JltD0))c}bXT-L)w$Kg^4l1NMdi@6Q;p=Tx7?@cD{u*k_Z_sVhTsH&{K6OFl
ze1RI0u2OuA_CB|4omB2q=T{>gtFdQARghVX5z0FvtzD@8o6zI+&B{wqAz?xUP(k_3
zhVhgH?S&GY(pN$MJyD?r4{>RP7r({Lw8oeHs~!A}Cp4n>kS&a1E!*h+RZ<D^7skDR
z#nAPSH|M{36h|qY$^><n&kRma`_B9KW2irgouQ^W(@G3gsKj-6v3hGDGVwlwr~FII
z4H}b*$Neo(17%tUzH#3Z5;Y$5eO(~;?+HHO36NZ`^u{W*y|CB*)n27*6lWBew0an1
zUoH&SK1ODKyQ(y!@l>G0<aOagWMM&me%;61+}sF$g?|r_3ci`vV}5Dk@W<fzJ;9=)
zD$=yZwoTFBOXW8yg%XjmHgplnjl2kcZsWgH5E9)uFfdS=Q(5V*Cr<rO0)LBfO2*A1
zp4gZ{FJ6l+N2$+Jd2`7w5j}53!tY<t2zYF3n-i{_=}#NY3q6-EHN@suXtr3C^p?Ch
z6+b>SNaxI1C~Yp~uezLH!`@Y78OI$~6MxF$$=D=6e~vO$<Porn+t*^gNTm@}JfAYj
z!lPs8LOx4XG6Dkt%=N*}Z7m(^p-@s|V`GOhhOhCqI_4j65HwIn+1Gq1r&N^@Yvv2L
z%N+C`^OGcw<@IH!L=;PF`nmQRs)@k2`V)k)JW?igB)D|K{yktX#;U5F3Y9ECw{GJl
zwK1kUjhUI#o)C~Td<&;@=20?DdDZNot0B!&->5(A(S}E|%$DTHS6Yj=A>2vNq#T23
zIQ(9=h3PGwD;<yl!rQ@io(c-iaDf{0<GyGolqRQ?{|PE=%IP5A?=O>RznanQ^(*9t
z0J-zqNO#Xzyp)jR*hl^$uT%9PUwk5Abj#f}8$V&l!IR6iHd&_hFWC6J@xa2hHQCw{
zxe+IG&Bl{?1i~U?CFehlm;;R@qh#kzVml1sb~<r5lYqC2H{3=Zd5iu<mTUO>9tlB>
zlv9HwJ8#(99+|1qo!&LeL5(4UqDGxesi$Jb!Z7-q25aAYo1e!ohUZRxd-ksF3Ov1_
zU|9qjJ*W|6sQv&fK*BKQRD;}&7>Tr%c(?PPo?I$)*qt0(9U%n&;yK*eVwCpi*jAI(
z-V1H9pd|T7Ty@D=tQ;6o9m@E}+FlD_jn(sKkUAx{-f^@W7mC5z(wpxkrFgXF+LL97
zEc>CRp(fZ_8{KPkm%4SYpF&Ls2V-&0=ovZ=husWl$=t#3;3w`_zBkD$QzJ<{AwnJ{
zlksid4|+wTbx#p#$!Hk=k3@eAG2=X$Lp?)cDC#o_Zb2yTgwU7m7k^#>B`!cUG8c_k
z3qvBz#rgjU&<_ZZHK)?!*dx;%Q{}X??JDh*>I%0D^1brRAX~FODnfzYKIVodJ0HT7
z%FK3@8Ov4j;vbr<12i$vR2mWY`F12p&TV1kOM{YM;{_HXAUz#^#r|@JnL)kR)_th*
zjAbg}?hAgkQ%S^f@R?~>eTy%5I&|9++}HZtvEZ?FIFUqON2Vi8);qsvv)o33pfan<
z^fJ=IUW~rqzFboHPlP|HPo-h2Ix2qI__qk%_<VP1qe^PkV14u#zJW_Nw~C|jf&EJm
zo=l>l;IW<Ej*aNWQx4CwQDe#jGNcbMWp;hP_!k>`7`5s(0cQ_bSrqK~5GP?kVbG9B
zyRd;>TUpjW)+k<)^DPz@wCSmyST>!Xr1zoB%sYN>D(g*_6xP0T;hYx^QF=Qm@l@Ql
zQ$XA+Bd+`+acOvR?gm`NLiW><M{LY=dkKUUG=*?!>c6PbK?<m^?dXB)IbAJTF{LlT
z_m|bvQb}ite)v%~jQ6qI5y89M5Enfq`5O@+Ei3+<!O2M$50GnQgbBxmcD7_;D@NAH
z_DJJb;e0_tRc5OH@n@89Gi{5$KUs;(c*m{xwt~ave(*9H77bffj9jfx^gSy-GkDjv
zV`Vq(i1qtWg1O@N-vCOEkTd5dQ$7$*G`M00DKk2?FB?$29CJ=_`zLGCHejwC8W$Gx
z>1GW+cWhU-S8&H2Y)2KDKkZN1pB41K?^wTut-QX~%3Su6;Mdl6a6*q;BK^)5)G-w`
zt)ugwfe15NP}yD%$Xz)`3&>V$+-G!~T-#^T>W}{6syX_$;QkDCYOH)XU(P>1W?`p#
zdfe-wxcl?V^Ci<k;hZ!40}I)!pD3=c*D+YpUZuwTqs@~DXuHjAs`TMCL&@Fcz$pE+
zxnwV)q}cXa<V=vrnGNa6TmM&z<8BiHvkM4a#+u_sL&c^IxBVaY?R62o`;^#gr+Ud~
zFoT8FuvQ+QJV1Z$;K}yAL=|-B?#5=!P&slUlOXs;KD?;p|1k>3+6Wbs@kVYrIPa48
zjFl-N?RVr|Y=^$R38L3id>$6s-e+(9@4*3BSQrG22)T0O_{EP}l@&`&OjcI~yZ_c7
zH2R*28h!dZHqrhOV%6Dr=Q318Ch@sa%x-|s`hQ9VM*HJYSjFh9+j$;n!c)Y2u$5<U
zp(~fEv?N3->pHQ7-=OZ<UlwuU|I2Ce_l4d~2SpB<MCVKdqhFS2wowPgjdcpRrx3B^
zDEj~LzQJg-N^&iH23$(}^T@7^;NT^l#%|>sJoQLKG|8GWq-aO!ENhtX4@_i%b-TT6
z3ML;rdic4cg=@7y)(Rfhei4OSIb&u-!$lXOD=v6SgZ3$qKu}(dLqPWGl^uj`rTMDd
zj*^Y5xktu#>X*%fuw4l=^P_Nn)l((L%Ki)GoTEhtjayXG<3%*y=qX<~cVYh}o!*~t
zIn-U(xgO&SQ~S`1%^QPmsaBo;0XNNQSSc9{7Nn}9=&w;%ZOoB`gmK5ZHfnd0DPh_R
zC=}|Do=6Ehlhw@ZDr^}0kB3GK5n(Ik2#sUSUZmMKzMtFKz6a+en{rR~+y5rTf`5>?
zB)~Vx<cHc?jE|3hGyMM}Htghhrw5{Nd(}|mpDEFIRHyw}fjoSyNK})xmozZvmeF86
zDN%~!<yvHDXedX?_5aL1lAMo3q*Qaj;c(ruDS%;$LcaT|*YTr+ziFI2XV1TE)`8V6
zHT#;{h2JTtuJ&X|ivCCC37`_%o{D%GHQu;35_9&TYC`wIfe6*WttHRwm9FW)oZYuc
zNt1g0eXysyj~dZihk)2Dd-;FH)&#6LO_ye5vt`BfSphzdKMl6pX=A3Kes_8^TIw!7
zR4`FZCY_K?(!Oi%uQYS=`pZm#{hj}AH3xRup1^J@pbgKXIWXMy)&6rn?xmZqFGD0v
z#xGT_7#fm<7h@fRS<ViK=&2gLe@;l=N}`+CSL9<wk4x4fy!hemMeQV&NK31RY|CNw
zjHQhWUaGl`Z%>3qq-(;;j0cZc3`0-Z%vr+_c|0y}pUKZ#p?`9oXJ&hXoyQ`%K=gxO
zBqf#hqib{_wMsJ>sDOQ>pn|Z^Fz6;|{xA1}Gf+}vta=EEQQqb^s2imci+f8JXMyW2
zNEzTCT#dfhRW`s*W)?u{J|||kj4AZ|FX=SmGmE<u@XB<@zxwcpm!GG|PvXER9aJ^H
zDSBZRu83mEMJ>t>7*G<cU_M5H#`DmctKISyaPdy%nSW;OQi4ObJ{>9l1c@6H$9em9
z2I`I4b0d8HSLKDa<)v4{3!cr;#wqi}J*X`_*wErGzm2Pf-iLidhqhw}Mih+(kKB3U
z_Mh|1pNb6yGT68>v1K@SjkHw}UX!Iu56+uyK9Ae6qQv`Zv9j=ZuF=Nz)Ol`9xEb%L
zVY^HqN3!A?sPTg$k{xokMMO)8$2~ZTNW5{XI-X=99j%6ZF_n<CGOe<Pl(Ll<nvE0c
z+Eu+7y9Y8C#294B*WB3Uky-uUzpqo^E={*a)P5{2T}MLMA??u+#nuMDBx>ss6hZLP
z<cTwA<@<-jo}@s|YGkIE6Q$fx%<#^Wqe=oUwl0keH>Du$)It4ZdbJua8coS(P(F`^
zJV-v4H8eC3K)NJ^zHAdd%t}ll)QI9fLBVcE?K}`KbVx&#Mb22+EsGNFrs|OWbCJ3b
z@_dmjo8*cl>*Bdq!HnV;oR;2^e^!^15%DJ3ec1xhNh^LSDi@<YZi$Iiz!E06B2SG6
zPZDTn7MJ&Gjmg_bnu9uXr&Ums>yKf6klChRvX~V!4cDny0H`-v!1B~?Zid}qa(^;B
zaQda4i`3+hy9&$g4cq^|gW%y~EbfR(@!Ia84PQ7!yy5uMxx_ADVbx#cdq>GeP5Pw@
z!>IuPWcyi+L%&i;7rbI<m5o%&tz#%DZBUWfQ7Y<N(Al01_-AN9x6DAdSIB#|pJ;ce
zE;($u9DEY`sM#x1`&NbNe-Anawq(6oE*ME(tU0F#UN&1?;Dw#sQE}U+VmJTiS4iYu
z8}O`>C|4gBn>f=HwH#Ilzw{K3;O5Nme<C<dL5Vz9?^<;T!QXw~Q`%W2!`7t6H$XO=
z+q%gELhn>q|Igc=2eGH62yJj;YHvqrbYxZ?7S$!yE_On}4)-q9^xKdv@!n#spO8+R
zKTtfP*qk_;G+~JHsq@#M99CYfzo-y52uUQ3|L1%F-pRHv@JU;@ws*~3YDDMt^32fi
zFsN?rCQ8QIDcUJ2-a|K3F5Is1k5;|j_EM9ptJ%xYckE5sX5GyjrQTQmlX*v^9Aok3
zZeq6W8Goiv)fhcrEqm&IY)54lZ!iZ<i-51Dg{jGoIXn+j!cr~Y_+2Q}_}qWYd?S)S
zrb-{^9WiN6O!~3-O?uo6OFrvO$@u4b{e&ATW5LgJRHh0^Z!^hYm`T{Tff|||ga4D4
ztr>|cWjy=sO!?--IY_!@C!*N+j(BOxejGM!y}FQXGd{_&&fNxG33>aizLrhY+G3OT
ze-J+Yf8V^@fXk}r3%21(^^YMO!sWVyo$PB0($uIi2-(f`r;UGnb}bc_N4|EP_tI|W
zbQk-d9*^dhNaEQ=v4OfTP3ij|v6_slxul<4C4Sdw@Z!Vu+H|W&7BQI}<gZ4b_<sA|
zT0=RY>zitP{(s6l*(=3m;pT-!8!{0BUn>gHrlHh<%bmF{VstclpZ>h2&x*?4**1XZ
z`Iq<^_DW*?-EaS&_fFR;ja9ghI2T+vx)2p}`-rQQ|6~1&fnF9a^}#~Euz~UY<&Csb
zCl1%LL?dfE`dth7Gql`;i+1G5A{cS27<ifG`plOg*~7JFLPtl(r`80*)4LAg!WU6)
z?}WKX&0Y`;AdXPHft{>KYrby3Kk2h<P>{*0tE)S7U7J<BMweTdlf#@v5Q83zB&tQ=
znc>0%BR|K*)x+A`k_VM*#+Ww(gsq41@T>7+;^LK}F-U1vQERmpa6uWC1CQ2DRI`r7
zquhMGpA!l>&NE|02sLyN@_n<xfjKw5fdHc5r}M`+U)gUIpy|;(GCEpSGT`_qeE1=n
z7v)BaThw&ePbs3pcx`qCq1J@xB@2o;W(>@7dh97V*U0&#5B|VwbdQTPtG{#rJKZ$N
zf7K(ADew3(FWQ(M9xWIOyNa~D4K1Bq!d7bc^TX`Vn{}e|bk7&&=aWGgJ9D+Uh-C!H
z(T{>gLd};YP(nMi&G|zUur&_o<Y9kfSo>tSuwX@bdA^yd)=n6Gn~OWTzt{b8q-5S=
z59Cf!d=YmA@YmdG>|bH%@n>r{*tpG-D@NYVj+d>?(-b+O)qfkRyMidP6iJ+e)xCa@
z_1&I?suYykzk5>~!|-g7nI0tRmW-8jJ7O&<q*-04TRe8#Y497b958_7xMBjPCGw!U
zJc1MKnJRLEY%QM9Kp+TaJ<Qb{LeO73fBt-2XFR{89P$FJ_y%qq_6KHGRasfNsA(#@
zwO(D(ej7;0`uBaR>WjoE7z1;@Bjwa=8V9GldP}6z)2EZKsr{L<o_q`<IC+>eJf7UL
zx$7{3eX`M=H~az)_f&CkSbjzD!dGlh6;kH~eQORfMODx?`YjduzkJ$gRyIJGmN?Ps
z&sLgwgicOX<1)3di;KJ;!?hg!#j+J9agJyM1N_<UG@)SjSp-9vgwOcuJWp4l8{1q~
zA=V|m!d)K3QX0dMLms-2=gmLYU<5Cu-lYH3;oim$xzKyLxq<BSu;Q2aKi`g&-~sp?
z@H&z;5?o{Mq9_9cLs$~XpKLm#O@H+FJH%;ZN76^*3G4$SBVIT>>cXb4v?=kSob%@1
z8^SM@D7UNtm|B3&@b!7hu#~$XxJi~(J*Oz(uBZ;mKG8W#O}C_Rrl^5g?wv8{p4_Jt
zsC}uBeiglS4^lC(^Dw&YyhaNvuEPI0Kg9=^=VLq&#Cbw*d3bqG#(w-Le_e1XC+x`h
z`@*KMG1I}cNMe^b`$+v`2#az%hf8c_mUy-C0e{QTjw3H8=iaT`w~H)zU+eV`4(ho*
zeX3SGpbQ)l@6%%-`8mk<q~_Rvu+aYo{>=A<Tfl<U(W&&&O05Y<(AZ(M*Iy3}4i46`
zvs;otOPe5HJ;?4mAc^u}dfw7`|7-BGEaR<D4kYg;HrSVbXU>}&zBokOn%j3G-y(i{
z)W`Gmg>hg{<M-KERwsMYI^XE;pKB=2tn%KTIOeLY{*G_4xRJA{=r`^U_vDGb7jC6W
zH~Ul@z4!AahX$(u%fQGNM8Gaqmla2bQ?^?^dgu5{*0?XOHLSga;B^k?>&w<-fL0G_
z=ZFxdiGQ-6H!a~rV%!<A``5r^Yt2$*?^bVAz*0@vUuviU;v_M1CMx(a$i)!Ea~Ch>
zq8{@bCWrd?GCX#0hzM!FNw|B77*Bp27bay${JyW>{9}kRJ=I{>YpUf%Th}|*SLYYy
z-)F1UqSD`Z=W+Ktct<k)cz>;=!}m$Eady0`4O3W4p;rHD=_pO?Y~j+txlMJ@RdGW+
z`4~ybE~~~B<KNSKM}(0$EF;5yoO174;myBFXe%I6?fJv2L*Jla^4&wj>j)ZryOW`T
z0b{A;dgZ-qPojjasc<?0axz!E9R}0(SK@f}dh|;z@2d#=-=W6L7J!x#2iifap8}fQ
zq+dfrzjGGr`}_MNU88qpwitHOT_aQ*YreRRPsZEwRTWii^T87X+2f74Ih_gvvQ&D<
zl>22UmW^mgQVY+Yvwu5iA}@Wm;^&w!)zMGrNzCMO^s;p5N5K~<8VNf2``4vU!t-Wt
zd;I<Nd|sz!Adq=_`l#qjpW$*Yq0i^t%WCMsLn27vE;|(@e;*kcfzm#xX(V4EczqzG
zEJ*Z}5xlzRzh$n@UYwN@%V94h*mxul9#BY)J%(<Cwudyor19@{c{d5$Em*2?`g<=;
z0gptCvSNEEoI@+zPejz4X`$@U_FRCJyg;ta(8>d@7=iucTX|2)1P_g84%LMH^%WHr
zZHG&(+DH?!2+<GK>$i+xs*!K2YZczY{ya?k)goXZoA=;t;dvC=fYtf}y6-O~H&2+}
znT|@GwH7bKdd89`k8$s|m7+@CkS2WYJ&(u;Sn@htzeKp(Y*1^Om_1Dkug<Kf*iv~l
zZG_g*apk4f&(bk8^agNwXVO5j!VahJSP65!iBkn0ghJbFtROmG-n&M3;e|pQiG&`y
z5#s!upMTp3*+b&2J6NONi0&VKnEONIL-4*!;d!yvcY^>!0`ReiE4tHJGxJOWW2zDN
z9m>DfIDRn0t&j)FIB_Ustz0T6?R>C5sP#vfIL29J^^Ia}B4tQs6sF(q8*Py6V$ENR
zL}VO;Zj|@kSFVbNhNF@Mugkcjgt#qt5@$tVaCYBpH=jQ98ShIAQO=xZYuhWU*d~W3
zzFOxqabr@LGuGF<^D5>ETXu@d*)0`g<8fW^eGzQeC}Ce~aNmW`WXq8TzLzWfTlUrT
zVMAr*w%Ld0`v|&_aJ+J+tWay+*(sZ~_8<1WV*y#FADso2T3hxFe^Nc8d{u#k>X>7L
zR5kDHsG+58<f`Q>4}V`-H5q14ax{E><08Q%i#YeZYKsm&dyO&7_X5!awS|24C(?zz
zJQII7;qs0;zrs{C>tV0Z^%p?whIR!<GPtXW5hj<<PTgyBNuiHc1x;>}56?>hUk4%W
zgMAB|D0r^5@)KS1si$_!3$LbcktQtH6n2(>%?3_#{)?r^B&y#DVqU2@+Caiomjs4t
z^PW{W2!}H13R4i5<t2!lv5#Kn^s{P@-E?Lbc(5fq@QbPB>A9Yy5ghC4@zvS~h>SlL
zB|{kWQ3;_nA>v(xSgX{mL{GonK>ps(UoBySLf1Vn8~HO5dkZS+YHL<sY4z^`sSPu`
z-Uq-Mrgql%?X*NH<(faV_(}5qhHfUL1z;qlWZ>9ThFwPScyJ5(Nzg_lu@mKHShQmE
z<>Ya$N@`<va1=@LvNAdJR1n(Syj$8h(OHR*xOIjyQM_wRxj=@>Ol!N&xyIDAaxkNf
z=glLjo?1#gUi>O!*yW)Lzn&NfQ@7Yii*q{9a%+CCuy)9HDrjk$t*si3OksdC7VXd3
z+AWs>G6|;*>ppq8xxTxX+dS!see*uA`NH(}*i{MaJc!6UuQrf_eUZ6p{R0Ea28M>I
zl+I;u!t~@qs4)iKKVQ9#_SCYMo(hzAY<bc9_!|56hWv+^?DM*o>`gUoy8ETvv)x;j
z<vAsb6VFytO)abLBg=C^5ch3$c_)c>yNUbKb#LNNI)VCf+a4q3i=)1;UumDP%`HJD
zFy{*rac*f%kU;rB!!L=ROKK{o-E!|%i;{x~;oj0o!M7oVeM>$@ZwZ!&m6oTm{d<)a
z(-I!~cE-Dwm%qLC9eR9}&o371$wqwTsSo3>`qjjF&S7g=p$JSQzCW&J{o3s7X@efv
z;XZ1xbHN^^Jb~nq<tDzqzUW&lM~+-Edz3g)3>!E80VM0nHg{Z8Q*%X4W%AA|*L%Rb
z`k+Dzq1a1b@*JQs0DGxdquaLuf%67J`%HxW3_{8|6ILew@LaDtd>t(a8B=o#Qz6q5
zrs#e>NNuZcPxm#oLP4sST7fv}Q1&Y6kq?ugNa~JN8wtMRs-wbOQw4d)`E{Kf`H=0b
zuS)rcZTtFN2_20>aJU;I<syJx6S`<#KwM}HNC%q0@0SxpyHE4t7t<1Y8Lkidn62NZ
zEk+0n0&uvX1Gfj6TsnqNe4KfMfcGIyZgJyDet<r5yqkRNeTzjpWZ_IoVeRM%*_1VO
z^|10pNzm-kHFn)FS&}DOH^Z117kB^S%1jRz;}Gj{3ctCQmXz$XzuTJzV$cFs9uJVS
ze{wQku`om^J5dTT5gohk!Hc?4G^GOqAWV$E9G>3_^pE`6{8Es_`K;6?{VI+XHZ_3Z
zbC(@#rTSodYkwHb&9c#3%8;SR)s)gqycCcZ`>PfUL($8hz<)KmEky(~T#fIT@_4~I
zbp;wJLNi14!sk7`10?*CD&zKk>G|_Tt<qn+t-n{#-QS1_tIDbMGqKUg#a2B)THVFp
z$purYRm2m<GHsDjL%TVZOzjE%&6}^L-^F~fOd%(}SXGrnYxY=*sP0V#Oo4dR50LN%
zVqI^|QMoA9zb$WZ@^ly^erWe+s3ZsgX#wFFM&y=ld-#JSUZZBpBA|y)4aPaQT-D?v
zdSOjf(W*h>h)R3U&*VwWmNeDdonMZrFEL@dpF0v%hwACH?_9d=IL2tK>Ff<bCryT4
znM#%>e2kJ&xtK=AQGUpgJi9DD7XK%MK%Mz*9wh&hAP`DmyIc|L(hx%zURH(`I{fDH
z&bu{|L-zH;+M9dVN_W!agHkA*9ch2vGyni+t*#r*OKkwn+of9^?X+w}{?e5bxuJQI
znB!NaDZKSpJ?zbnS$-8d77=fY%GNyMtJO#FP?liq?a}D|hfifbMbu&66U5&f#N;4=
zU4vk(nQvMdP?z=)={GIWvxRL9H~-n}KZgRIXTpWhgr;Iv4j0NnSMz%Gy&0_!O&oya
zFRclUBodt5!VFostdfl#REx3naRZ%3fDW;boX^2Fr~HO`qbknHYEnYelW$NgXTP|f
z@v3fgQAytQ43Up%HTTe|&|u9vp#XMEJwYg)Z}5Fciv<?AJ|q$TGVn$>vy9K~7=}-`
z8MUhj;+H^c!X9E>5xQdixe;l}ggv(!yeY2(=BXT^_pXw-Q@VVj#(0gj80JL3gq_>@
zUM|E9wIB;p(X%bV-+~`Ci-{8Mlj-cgdjl29&nDft_?ur>+8C6|yxY}CdSvuHOHdx{
zgtQBO%DtMx@g8)JrBJ4(rf3ZpmCSF`Xfc?^qb(nu+(i-@>AnNDwmbaE8nTpTd}VNE
zw6TZcgsA8G%qO*%Rca|xp2grnghi{h;kT2%+FwcHNiysPCSVD^rHtpE_)PoIUrie#
z`T4YdkVQ!|p)$!pA0qx_L9J=Rd1Xs<N$(~b3&P2rY0Hv?e?NjuEyR7WM&8~BMV;4&
zfPdH9Nmt6-dzlX1i21@d(<fSg6I4C`S+C=~p>tSel7;gAJUv-q^QPwxiB!Eluln;%
z>hHp5><XGD0??963eb(;GG7HvgqsZB{mP>Fb}`#uUU6N|w#t)W6hs)asF$<p=E-Y(
zc2T<+1cO(UYxFqZ#Pj~z{f~e%F>Xo#d!<2bDLZ?`6ye)142+QbD5O_`F~PUiglb5@
zLY(B}l8dbbo8MiQh$ojPznj`E=Mbn-TWg$c`o+MN_V?wfsylRA@a&aL?Un@x15vM2
z4{`G$bH)TBRx5l{!BM=b%*1K1!70+O#b8Qlyh>qoJ!bg)MQz&87A3g@BaBaFTJVPp
z9{?U22eB9pHhq`15IxXU9}J>w6=FzXMskc*1ix#xkEMhU%Oe?;GrwgfM!fcYwQN&h
zOvbt9@t*NLF3W9_CFAibxz=ouMH?+DKhdZ|`D<|EC?G@n!`Dt2v&K%4;J_%4EnIab
z?@wq*`zi>ja_CivSo`3}Axw>HLz2vyKcetgKh5yxhT95AgWad)-vJU9R1AE8{(QEN
zO-@Iw-vbUGmdm~GSrv&dhlU#1e(~Rfbv`~2n?HVriqxdm>3^+$^~c3*j-p2gBsedP
z9bUQo1|N}hh!}m>b#dj->yY+H6dL3tk&sasLi%q4l`0Yel)&^pfvqj1>=^_1)o|fR
zQt2+sw7@XsOw-B%@qi`cxUyox#G=H76I#=2C73^QcFPx`%A0QzHvF`6E6QHMTlA>V
zLfq};Lh|+OqM11y?gp}UnBTh_j=FI#(}zlJOBiph&g$d(nm+Ti&%YVwzZk8lgxaE1
zQ=Br=*VALd-<WlmCOJMCdLN)e%BU5pUK9GMk(Ac@Zbu5tSqGH0Tsqp$2OiR59-U7l
ziR;RC+2{4*SbZh<^QodT7^pXQ7h)<1(@Jok3bGO}4{wguTE_Q7+;g^4P1G>Ndgg-<
zoKeBqeG8pO>_`x_^pSquzpt?ffh=$^u*Gcuhf6v^Hc~m$=yPLn175_)^i~l$UtW-3
zgqy3NuQ<Qj6((CnMfp^z%p2;X1ML1LnF{RW#n@a6{=o%58NPhSppzA?-UcwGm73x8
z^Baksmbj=i<GA)H2uVs8*{06NSgVcfu@TQ{i;ip;1o(E#I&l4_6FFam-?A;NXY()p
z$K(>PB`j66rNc+tD0bOt@3+DyayD7zKS~0Z9A;$)n(|-vH$OE34QDzNNfxZY#`G%{
zM07H2j3(`E_q`wff1izYrOKW#qODv4#Z{0qt+uy1thBlPnXc?TyLj3A8z)0`Wt+U$
z_VnG8mKT1GMPFBEMXRasr@hM2!DHiqRH*cYp2hOy_Au%8N(0l40TTL>R|6Zp`R00{
z>1>itXgggl*?V-Ft7u`nu5y=EDB#!m*&~z4sTs#8rr4~5_C<!6qLYAGm8g-=Dw<GX
z+_io`gld1I=+~_O{t`--DiaW7@@VUpP>xJ0K?cYwl$23o$DbbjIf^tLt;i|)U5!Tv
zl&262!Str?7AMR(H+B0qJ=OSyDLL`W--uFwKVNLr?H5fv+aZrZpJ;)iN=QeZRBVbQ
zURiKVm2<Mg&L#bMUlXd^>aAf`=D}#FMv6<Iv}3C%)!|v<CzDrh8>>Hmy06RU@9|qK
zyct4+O{X_Gi!&es)^cq3yS!V?S^Q!zmJtAK;*Co;?_zfPz{nkd-QFJpDJ-VQ`94)Z
zgmq(xd$?Svzu9H8OUJcC{u=uL)+Qw*#zrcesHhcw^L~E2ux@Xqj*vySA)NZx{GB=H
z4pPvrT4edo?3zwPG4>L+3r`BOH3MRL0RoP7Aca}27WXZiv5zi>LJU7v!y;k*<On8q
zYO~jRR20Ko%k}7)Y=fPGEBe#2f9nc16E01D<Cl46Ouj?%#LOJqU8#7zLlS%049;tp
zZ;tw~_XdkVtiKh#p_w8zlqacYXeeofT!CqQ`hZw$1Zh(!!K7R5t6c7)JQ|Iz%19MW
z!S5Rrzu%L}4;+RiGgj6YU@ot24XUNMJhNWktFn8!R~2eR-hX+6&Q2GgcsO?Umcr=^
z%=2^$*8gx|@+)qqRf!pd<Y)qh=1^PWjBMPI4a{LJ0v0Sevits&@%k<TINuEes9h-_
z44&~Mnq71Qzm?Pv+Ut=io*|$C*hwhGpn99+K4@5vxBY7A*s7#rwM+Rxr}1xcZw&Uq
zJVOXx>wo7#X$DC|@FQ=fZP-epeVK`z)<Nrs!!Bq(#w4PeR^dLH_}zucDV^|!+{XZO
z?~2EK$(;WnyYD9>v2Aq*qqAmCh4L*J*9b8hrF`0JM!f=Iz|N3JE{(Uh7DP=(vOzXL
zI(4`5g7%4<I*lJ0(I(+M4~`1jX|WJK4oG?Oy-~85N75QL^f*S|W4rU+Arli-b-jg0
z#Yd71H%LJrJIjI!cM~0@sl#Wr6?cL$M3J$oft}{&_Y1j;MdC);blU#;1{I53X>TZX
z;Ir996O6=F;zDk0r;<t!rGe#dE%cF>h7wF8U>ZEAA}^Djco6pEk)q=DT7Dq%A<mg=
zMSoNQcCipql~md%pq3Emy~^`_u1y#uUo4{XzWtGa!iEcGsv6X!)Xw1*rlT}zc(HR=
zDI>5|z=_#KOBx#dFu5KX>FMbp8Eg5?x!4G-UlwPVaF&M=ePqSQI2x0f*3h1d%i1an
z&h1dj5`dhkj1pTyr&<U{uj2a7gy7p<o9qS)BVxU7#Pb_&t3%$!ckuiF%$FIvLMfCm
zJamtO$chkQB32-tz;KyLGiT7!?aXCK74DtR5>h|hjKPxJ*{Z6lNV;gy>8}lGheHiy
zQwXS$F1ih>Yc4OmjX|&4+uEKHW{BMiyqaAw7~B(N_JfP`>4J>hAiIemaa*6k#guso
zs_JbLTt>aaM3S%ZyBi~sGuII!XNj3v`$w;)?J1|c?ayW$g9WY{3Zh9DnJy{j43Ru4
zpFr?x02CAy$&L*TCBW9qtKLk^KdL=@k;`6RU*9u|z)q3INW?(1HAd~`oo@sip}K>}
z(s@^yYMWY~R+#?vn~>$@dCHt2?f8;-n-Xn?sIeU@!K01^$uFxbo?+umaP3APlBr3=
zdS%7-RcsfPox@AK&-bk(a|9iURskUs0&RZ7rvg+xZfCsp7TeNI$<4Tv%;iR)e3CWo
zLC;dcQo^JHvU2_|2O;5uS*UVm8zhk(uCoSKB>54M{Xc$4^$!gVnarDhH2V}t19dI!
zdw7C9E*^vm3-*;R=<*rH>yYLwMgpKFJ)Jm9UCq7i+em8Hp4JE1n|@VLFH#W=)Qh#q
z=M&FJRpPt&_RN%kyD1W>?>)l=J0vG=Al7|AM`)5!z-ZT8c2X1UH+hr{tWcWCLX@O<
zoMx?|&14C-?ZW9&$Xkn(UjS9o2q`^W^D*L3Z{A`jVx0=-kFhg7-kpv#H9hS^R&u-o
zeL8%*JLFCm?TdEvhYxcQCm@NVw(Qd~<RQ{gpl`$c6=6agtbkS<duYU|f!;eA5eu<{
z?J=g%d&zh;m@`QS)?kjYHNlirBE;tU)`wLOvyf_hUlX(46t>&VHuk0OmufsY+gV01
zR`Oa;!Am_ZP=DXIU^!@7wV?#sr4w><sHi(jcjrKYcZ7C(O-P}da<zJQ5W2XyND7lO
z{$rX?c#~;L!^J~QHAN0>pen59zot|F?n*54f=K`DKy{R6XK|E+^(LmKviV>L?F^~p
zJXS-t29onn=H%xeis5==wcB|iBYwo-b+D`=xMBuI^`yI^I9U*?;jIY^P=hYT@>2`o
z!vbJOqC+&s>w7kcUJdNpE%NUeTQQ^@b|Kr>+Yq|XjhW!zyjDp+(h!d>-q0{&j%;%f
z4kP8~=NHV_&?quhMnu;~2;0CUXoz?b`!HedCoo>R5Vi|uBAP+!bvcA8P-2XYg1t?@
zbboyvoxTT9xADoZ>9$A?eQj;w`$!GQnGJWvp3Z6U?OllqwWcFxV$ta6sJ}3-)+J0J
z?tw3vnY*H5b3aHtvq?WP6J_KbuEF=Lz@_Pygx34VeSmE8o~4a$o>Q_f*)>-7rsFG;
zz`n0a^<IzvLB%v^PpKnWR5zEZNS&WY9xUs@q*gy2?LX(pMIos;Zz66-vWP9FyD?Q4
zF2#FA{iVXKf<oRT?u%su7lH}KtS=B@5q?Q0aMJUR=F9U(AoI*7RK0(goBg2rs&Pwz
zEXghwRBcgeT%yAU-tqV!0Un*6twSYP%8>S=PEHi{l$YJI73kh4ng-49=hsTpg^W3b
z-(ezJo%wxhdmJE_y9_SRGk$;F$on!;6`FY1UllHVLcSJ)&sWdxO9ca;bSUbJB>`zk
zk7I#f$|6XHwX+jyC~<Em{Nx8fxsgzjWpF0|SRXG|X(1GobXb=RP3`zC4}*0VX=vqU
z%PMF>oK7udj?EILYXnk}_h$*D#CpPA$wak=nf_$^N00PjQ*CZ;Znv}g7VMyR1{WeS
z2}=BCZsWcT(~koB=j-xjvQ6sp-k&b+;JjgDpq*KpSK-crD}tE2Kkya+yH$n$u&fE0
zGDN#x4EY?GM)pIjeyw$Xf8QI}cmw6s(wQMrM&!(a%4N47mru8xdPMvCWJrqTz>Ym5
zXc;oq4m@Fv%6xVAF-k?HZ3R(!0|(rb13DmWw`?R;Z8C*vwYVh$7PlV4E~qrMk@F!k
z*s{=D9*8GdOEnFiRCa?u3i?0^LrGkQl*#Low+o7jiaKUpHb~9G_uLaxnJy#NN4K}P
zJDK<}Yo@Y)oCohdc>_jbM;|jYv2XD{7n41db)uCJ(c?-pzo3^1Fz2neVs?I*(3h(8
zii1DKRY3<~JY8#E04>ce_cLN~etuU5=d@<O^^OVcb^)iN63Kp~c7@&VEBfQ1>R{s*
z3k!?FPd==*cE9W3Bwf=#+PlpU_$b8V*3N7vBT-!JsQ%UKF<<<5#h^jAIZ$3UhYP!Y
z-7Bw5BVfv`yDvb;`tzprV7VOCf3aihJ#BO~LZmE@b4JqG!^+s$)YRV>V51)Fkd_$#
zXC<_a+4Veb5f{9ZPDqJY<C=?CH2m-|w^8iYEw)OZz4ea-lkh1iIdf}ZtNq>G-A)_f
z{MCc7F^7VBNw8Z}hLZdSV$kp*+h+UiHhIe`b&7AS49s3$zR;+RZzN)c8f%qL8d`C(
zS}<HSsbV93&dT%w61M%CplmcKJ>1)p=SUjv4^~M@Fl;l*)Grhy=@;~TFV0rSf9HWj
zFNPrNO~drY#lTuib)mO0=+3M_0+huRDCdfb3OJ-fd8<wjIkWcUrimNtP}L2(kkw0A
z#JtZ#TUwNrNTTYq)_65mTrJwIZmGumV<D@c)!)a--F|DM;mNO0NaO^RB5?GNfo&|e
zHCW!Xxb&X#@@Ky|Jc+FsPE|J_<Kirm=!3Wy0H(o_U~V1>TT{?t?KLo5BN@<Exh(bo
z-OJtVI(^@6S-x;-A}X7gvDfzTV}pg76&i>wjSyMJ%Ew(Y&~w4oUcEJ;85noo?8rv|
zf7B7eFFz&G9Deb|WbMa+<V9L@G{cLMc>VI_%aR`py_H20jb^fX-<eQjYE0`M;)35%
zL&FDZCk%oexlkKMx?(kn-$GwH_J|iw-qnYAQoK}C_J&XYJNz@m1mazK#Wpwk{7mc_
z_|0dqy!!ehJ}9W;&)BY#x_UQK@Nla(%4pD6I#F#PtUVWs4_FsqiAW)v+JK!~{>i|r
z%Qm@18&Ns{HlUpBmY+e96rAsPuw5R{elxbn!h9+SgvYlB-E0Ig#Z#n$E0!RBMeOg|
z=6m~?-%+RT^m_z=lozZ?28P4n+CuMPheQL(2DalTDG#zHE=K3h8(B(dp(?(T^TDGR
ztlS;Do*M5MzQAw~=7E8R@4yHtbgRU>D;12A7^73k_yype?>~I_9h-LiGcz?+G*#`L
z#9t&-Mb7!o^@dtPF~4Dj*#xS;?D6Br+wR6Y@fOnftVS<>GyVSHKYBN1Zkjk7bG0sR
zV7zZyfc;>k{F(qRVX5Yghcrta-^W5pQ)J;81Ebc2K#eUhoP|T)abL{Wg>ln0AI`;3
zLx<9XZb0E;(DEN1N&$MRWESGa+&Aq^W#8PAlIXVf?9bCKk}khKyTZnnH?#=6JUrH=
zN*gP_&_8m(%+6zfHu9P@WAW3jaw_-_SA&GMDoQN>24?5`O`YJ9)$54Hhy5rCHkl!}
z@x58*lN{|#0hPHJ-@qRq$m>Zv4GaZ5O!;rOvMtm&R!;`JFI+OA<_zs|)6Y3UqJt_q
zw@nB0^<BYGuRg48x)KP<9`fJx<n9BDePV{U1ivHJ?d+CKz+H1y8`*P`#e`raEPT@k
z#_27)-1v)pFwG+0w~@Ltji;RBzzngTYp7djq(68XY^%A_N=izMk%8iAY$iFk3yX?4
zxDah|pVI$^5ZK4Re=m(4N1m>j(nkv;*6GW>$%0<71SDVM=yL-|srcDn3z=$>0~>Yg
zmQV&ASl5ONvth78`c3<8n4x|6oomtej1nGX$*f}`ARGJkckPeNP4IDcTEYVnRg>@W
zR+diN&Bz&19>k=J@lKvm5g9NB8faG)r@%AxYjJsb`37P)tg7k~Rs)QhHOumbW{mbG
zr#=ONJj+`A$qmFLwU&fke#MQhq}!{;mG+Z-`las`ks73HADjf)QG#7AUzJ$%DV)E3
z4!PJ?RbE~<KPTtKXBIMo4SNW+FW|TC{|zkJ{wd9DHxY^<0clvfCLb;_1o<mltwp*1
zsECAK^9r?EHu)i@J!f$NI5WQ#zn8d8&%1sAQKn?hq-EbcfI#k%&3_3Zc!Bi`uVDV9
zBqZY7;{s_o&UDp4lIZg8to@gFN94qS!lh##_DI7<%Ch?)GdC{h;2fQ^`I_7*m#Q-*
zRWHq)_h#GI7p_`i#8`Sg13K5EG}8%yrX6LU4-^iA?>Wge-Mk0mWejsDAe3-&P$Jfo
z!72b}FTUH(-o6VUiA*4Rmw&T_;XQdVhdNhGRVn)5M%)8wu)S$_mazC{IMi434gSy`
zXlmMc@O+n7Qqn|#m6~xX<@j)I=Xa*7666B1C8b02{WmVIU*JQm2VH<CzP@)2<b+;q
zQB`@5wJ0oq3S(PQ^k!e5(BYhMG(jo=<^(;(H;<ygo;Cu$cS)zj^CFkp5F;@Jc6N>w
zp4h5;nR(D{m1Hf7W5*cTfH5`geTlKEdWgOOz&k9c(x~}^hnc=)u0N^o2?GQONGq{k
z-4CQ+D!l7pKn{vsUVen>gKKm<!VF0a36f5gBaidyCSPO`<W1_1x<QyOi%#`!?tFlV
z{>TiiE{=seOZW)j<cxR;i{VJbU)jpXCS4@O-(Ma`zX$0rz4)Z{Q6IDGGXV5__)iWf
z;`gIpaX7WjEnJGu6tmqW23E8-tk&9NyAL@t8#V);Ak<gvS-2#C?ym}>?F50IVs7K8
zHG;M31s+K{vbhXt=znzn`BI8{KN~2C1n!AT970{AJ4?FzSgtAYjT+Z*&)z|orwY1p
zsm+%Ou5tCx_%ARvXf`k~FuCv;$2u7xc9<y)1v?usgTSdY!tNOo2U3=Gf~j+NUc=6M
zYr-A*GuPNI=>6U}A`l1-P#RC#@-cSM?WC!>@#ORXTwGk74#a06E`Djg4Qcrjdj49s
zKWr#t&uu1JW#SCcLsJUf-)sj8A}eag*KguMJ^N^7thzIZZLXD^Pk36&5Ne-LYA!qI
z!E{Ot*h{O+Y+t8k;zH6((P#ktwfJx)JFNGGJigf=X|?SgrQ7~g9%#cDHQ%Ip`}XY!
za%LuNtR+DZAtJ_^K9+F0vlRAQPl~+?u}+%mXW|#l|500py{NEIrV^xt=&#uEXH-`#
z`LbUK>$bja-X0dbOw}OBSR-BQfLI?P`gFe)(+OJp_&dWS_R_|!vai#WkBsH8NuYrI
z(aO~mSy&ORK}LK>l9QLmN;35vdS2RL;C;!k+-$B}3hvqAbkI#Qpq#>X_v!qX*>%tL
z1*NezM*FYHeQ^B@3o;<y>ej^&-`jPhVfZ?<3+>+Eze)DR>PL?rd6E(jx;|5YB;$Z|
zhzFNGzFOlbYADeNGGjVuRrsPD$Tj~$!TZt|(DgUt9`Iws7++Ovx5<nxD}tJ}$6op{
z6GYyjJFdU;1>WEWnTj{@PAl(29X3Sfw-tp0mqdgIaG!agRU!gG^=d9LFoL2;18O#K
zD46w<L1`hL7(h+77o$x}#i)4`)sFR%DSt<=IELTcsV&ykw`;^W(!lghi5)p?-#yc#
zraAAv$$<We&;Y;eNy!M;c0P=`IlnE;)I~DTeOf=;8}q~#E(BHhMJgg!PESx&+WyqH
zb6Wd3*DP!`VLz#n_0XTUDy=*C0!ixfWc2g~iQE7fkcIbO9o=!1>@Ikyf<#VSF0Sw4
zdkW6%oq;7V7Z;D@S2UU}J-h<)tosmwl(xfZ=xvL(v9Yf#WU79|6zu}`r$;l4M!T3d
z4D?&;e?rBSg&eZ32F$0yXCka(XQ#0B&RqC!6+yT*XECndYHxk^0LBb2E?$`#*F%xT
zkkU5y0pukYw{ih%%>`s@_+IzgjGCIetpmJ*tOLY{cb;Vm3unJ50ZEZBecnK^<W<DF
ztlhGI;)|_n2`RHHRto7~dxuI<_dsKr%=~d2VWMk*&#Py3`=<;?OFXY21@u(SM+lcJ
zF3l;W3$<r)6G_;x{Mnd@%U@Q)@$Dx~Ki8gEr!U25&bY#;b6qCtJc~-Ka-$geFhUf>
z0T~)>9N{>+K;MXPmq{+K*ZqLnzIy$;C#CLg%y_zVWJ6qE69$LPpNQ;dIj!ddoz1P)
zc*}}S(pHEM`wfTQqihF8?g2eN$a<JsEEdKa_aF<_wKMFv$`*ojw_b|-=0*ouD!4~>
zpU(h~Vz%qidlX)gbKn?zTEalglJ0+0Ap7dMD|8{}z=E{aFYW%SNU8PqYmv@8yHh<0
zSC80(@KRmzeEOzw>!UJY-_FIpT4Zq=lDk!|QYl~l!~k>Y^V<RyY(D!dH?mq$1UTB&
zHq9tUN%`})9Fz{dc&ev2O2?oEmW}SAk7N>7+bu;ROT<(Mc0L?d_QE87-+Y5Dx09-4
z&3+LlpRZpsG(U<VKlLSipk*T`c_zW5<bLi~ncb<Npr0qQj|J4S9&&)&L55Oy#w@Yd
z-CN4A(C6uRK$hg@a`WIEIqqHIEfQwlu!w+&egZfggG2^gK^#zWzsuIi{lROG=Gi6o
zB$W6?%A@b}SAJH9%u0H<JwfkJBv4(C>&R$OyT-wk@@vN<^Y9Ah(V&vw+Y@)g$1lO*
z<kn<aHU2mY&*%IPl<p3+?n(!1aepS!U8r9yI}-cpysIT{=qY3z^lSOaC1+mXMguUX
z)a8o_CBG{LAASOcfG4?j&%;ct>`pE-a)vYYWwG^4d$hl~YJ5ClzX10_jmuTYuuzH_
zOj60sP=!IZv*kjqchfNX_FbzDw%1PuOBMJR^exDW(Q-|DQ|(tHjXHF*6hK?3AKKpX
zWcCNq+TO-4{-c~(m!03}97OMwThw+#hP`fUAMTnShN+$cPh5P1kuF}%Xd(g4WO4;1
zrItb%l?{8U%sr)2S*UR``h+d1WE1Q{*vX7JfLSJ2lkF<=S*$oaz3Ps9`SzMuRX0Oa
z!$kP&X>agpWl6xe{_MxR&;i|0Zf2zAAk@;d5QNIdI!r>(qYFXL#L5ov7fSd5d+u94
z>a79BwrlV1xayXX2l0m&pL@9sP{(`ltUuFZW;-El_ke2iNsDsx&F4rYeL}#l0D%RM
z0o%-hry_USJ|o90nqMJ0^!5i$I+(w%Qce}ADHSJ%_rrZBM&4+0&73xi`36RRT?R+i
z=+6v{q>Tk^`Tg5njSZDY@0c;>boB;|C%%{x)e`07Aa$IrQ45bNVU!t|sO+|7_$T7e
zExE#yl-FYChTfw15KW&X<7%Z714TY6)3!wUoc|bR__@O>ZFNV|nc2?FLMgz4L7Nr`
z5=U^n7$uQ{wR;Hvdx4K<r}sh%sTGm<RJy*)PANaX!vciiE#v%QuK|+fu*a<b-h`K<
zFAf)09ganv1M#65b`Hyx^X{*}=XEovpnV5tQY~;1%Rh=yEf++b9t@6N1*P<JDEvaZ
z*`jA9H^TyEG*qASmoa{p2z|()4RSoSmkzMBv5Hv!(}Tyiuf_|S%CN|BL+Ry_Y<Jet
z;GUogcUJ3wGnt@W5-fE7)4=itJ`L(6k9?z`(31FP-e@b>soR+?;XeO$#A30*-S?%O
z@S!Sb*=9bnq<|OQk7$z>wdL-RzdB8{Em}HGjc6~W!rfAiLi@xVFdun)B0%Yfr+|GM
zxB>S5YE>qp3Rd;Rfcdlsf2L%S%H0NN25JtXeU7i|{%|ejHnIgYx}H($5-;H^nY@Sv
zpow;hp7!~Of`44>Fnwqx`*r^@qDitiC!vjX@aW2-6cdI>weLn^KltNnBYjCJN2TYa
zM`0B?Bdaufp0;t%sk0q|MO{A5^t$GO_ZHgypSnok5U%eLX-|&4VF|@1aBm|gY8j^+
zMXiKjfP-Kq1b!B*w4o=h02iRl`w?>bWvH|?2LOu9Q5Xg?%ioKPYia$pT>H)HlrnB8
znBv+i&7oKn@}0uHACJg%Xzi8xWx7_=Bs(Qi|HJIPNtvr0r68Ul8ZSn7$QhZjmIW8w
zZ;_fwAVJ8XRfMQvtACofHncT@=Ks85*QXr&M_1nb=+&W;2f<-2bZ~823XY$gpTVX!
zI1y-(a{CfcA1z$t$~lRk^1U2-ovheXNSCSVok%9XGvNNGX32jz862L0$TDy(Kw|A}
ztl<1+va%34Y;{$L`^UkrUT_-gN;Xk=U|m4Gf&cGED-UHP5~c-@3M8S%pl%U5mgF)t
zeA48Jm-L_Yzl3z(JjNBCb~as)W3BU3y;iG+;=uFe$bMbmgX?#Gc#n7>N{)m`)>UX1
zjpqXWzi;K37J;KWrtCAmrV`{AjGXL<qnSA+Jg3J<>1hc(@_3Na7!SWiA%Ru9S-^Ow
z&O!JN{|6@jBIiwDC$Xb8=rP~hp|^*@o8XkcQXY1L*DRV*;h|z2!?O>r=lE7*y$G?u
ztgu_E`6;|P3!Q&DfJV+BuhEkJ1(_(m;J(t29(r1s8H{(W<^yeqAt?uv#^*mqF4?V&
z8hVBppJF<alL|9n0FDUTBA=7<WoYo3ksHlj#t=@rfT^o$69tv>|8Sgd8D#>Vbx<cg
zBK-cBn7b8VBq~56o0e|2-Jla-Rq{C<EsFDO4}Hb5Q`a|R&Q?<}VEHpKO2=1jPj-bd
zA`rji#uSmyBcEjdIzJ!&2+N4i;oFx4#|91wgB0@}H=R%Y&pHAf&{)EH6~yH8pY_PE
zT(DcFt>j|(`|JR(&zVvBQ3HbXy=EzCedd!8y1GVhLQ3LQ@|9<m+_8WhdcwX7?JwGB
z2O!ta)60t10r-x91#!mXq#M*wH9h&qQ!J^)SP9n(gD~<{&c{klVfYm=#SB7cB01Pu
zTB62S%noNTKu;E<sKH=Y{vRg%wmCRA!jo!#`_x2A5uD^NE(3e#JB`jAN>!4@Pahpq
zj8~uAls>dn&<N=K`#7S6(PMH}=~3PB#sAU8f|o2tT6zOcRJrLb0KRlI?UU%<TmCLX
zlTqexKj$k~^p2zrxcvCM8@YsU6u=)h^T%ph6u-cM^>3j0d?HMZL@^GYT{>|Z`r1f}
zNji{gd`9XuOE1AV#k>?8qxbh#tq@>}4t};n&Zm}5@F9M`AUF5XbrO-I`2Vl2?+m2!
zjr+e34n>m8RMIIU*$T;aq#+}Yy+TrUve%L1WTvu;jDFU!lf4cy$}AyUWn|0Vp6}^-
z-uz$uUwL($`?{~|`d;7bvkt+=MIt94IwS+Pq?ga|Ct7=~4CZEjsJF$(4V+)h=K(8Q
zq$~jAs`T;ab4{#im~94a1+3zI367ga0;>&p(<^A9xWXrelS?D{pG``4LrvIW+1wzL
z88-Av0)WSIL0t_c*0c*8)9QkY-<_{P<lRD7>RX~=+BDGTt7XnvDGbk-si1aFrzep7
z{>ZhAz_(OCTze?Xz$)bgpOS1Z?&FTWQJ3%K=_<QOyTz}2smKeZh0tJ&wV|4g65sVm
z-Gky+s+O1k9R*rkSu@_45B`YKl{i@GtJa8x4Q0Wb+c&V){h4^~Th{q4Vda}D2o1|e
z&F)#2TMmo!<rrO*<vmnHp_{jCc!}mq;N}?7-;ERxDPf_oq-yo(VSNsU@H~C48dkQ)
zwH(zL^4o`P3Wf&9synNvubr7_nXi^zBF{3Nb<UD&j>fIu?LObYtc70r^*v6K;*RDI
ziQ~`h8sH5kN($3(g*EiQ#D46rS?Ctl7meJpH{tn53A^O)lqY^{j`55GR)-NCO{AVY
z%iR>mXY@Mc=le)GFM5UG&W0agZ=P%GJ)deSuP48!q#r9xQBbczx!n<(cxqIBc5&4q
z$2`ZEaGBKDr1s}5neAlOA#Pv8TU-edyxCY__4(ydq(<IdDdpH=Jto!E{toFSx1{@_
z7vuj9NZI4FO@l(wI}Uz{|DFLSop@9Y)yl3N9FU)I^Shq#mMrq`AoMygo}X5^xXtUx
zihOYOXdGfw53tPf9pc;y#o=lC?V$?35F`VbNNeD+P@==h@M{cK*>^+<EiaL$TCFkX
zQ%6?eoQm&iK)#J!L_F-Z+{vW-&M{lL4eoOc1%n`95nH;0x5%<|xf8W}84;ZWG4r~W
z_Hiu@z>+?j<=L?`GNA>L@%rxn>uqcbYq&Osm&M4QOm(d35lyM)LX|9bU^w1=?*5bg
zou}yX^Gx3KKIgJ8dw%a&)5%qWT0l2oq>fDRe+ej-5CNvJMGEXumf3ew1dYp<l#zo~
z?^8rC8A#gj`euu2)p-+)SnbLPs_lsh=8QuJ?Bzdt6<P6iGXS&YQS7*59j$7I1&lLu
z#*E@v6C<!TMunH3|LF$E^mDh~t_HKiDE0P;Ai|eHTAQc{DfE)nZA_YADBX=D3mBp)
zZRhC_JA4T3eSkh2V$zhDqIfBeUE@nsPx=gVt}!@AlXbalxQKGwxTJHM_8h(t{K})v
z{KwSKoX4bohyv7)V(B4XX~6SmeGpUs2OTO0g<n#UDrT7NkL_->DPPuuk(?zzl=;eQ
zv{ci~qW-Ar<oo6yCNN!Bw0bfEnKi(E5Sx25XtH8q78#*c;Hzv;lCO;syonvQFTrz%
z=9^Of@#PjqtR@<GnI2UOHVrgy=#KF`facmAZW9w&?7O4-G8w=-4wJt<pqwrgcHVHj
zoSl=A*1u|7wNBnBv#k>+E!*wQ2I%L)=ZN6JJ|cU%z(c7i&hdjmRu}d>;|oomzTV!2
zFRa|Dt{+$n#`}}x(n#Wz{4n9*ogHT`ldGb!JpHF8Gg)h)dd<WQ2NU1`e8OK0UW0<Q
z_sT1t+uKKsVrl1`k!j=;qi)ZErOWZJ(!R#*8h+hVb>0Cqt6M*yfJGEe60tiD-?nYt
zwBcVwz*X)W%3P6DPXuF96<sJT7>PboPYh}GTmbciSqVQQ%2&WV=oxO5&fH+pXfxj<
zC4oYZy3l$DyXukir19HkWo^M*SiO%}lCw02E#E9ib+`dEu;jdcT|wz^*33OSEeyrB
zKIx^7k05jDW*+jw9t@Xt`Ud!m!|N2~A>jC9jNalJqBqx9MvU8g)0Dozosfzz;T3=$
zDRwxFnD0o}vwX^~$t?<nqp!8c^zkbWIi0^pMmz4Zk*C0Jck^#26ku~3`iS4(3xie?
zzCU|*^D*n8R#SjVnMmL8PtP^Qy~RXbj}yG5GqP<=dO794d{k$V-xtM2S1TX?V%8ol
zt~<_}&F3eu<tYgQmC=yq-?}&>3XYLta;THt#O?!o`$T&4OX6Q$gZP(skRTTV!T6Id
zH@_oKMtHa>dVb<Ls~qexIKA{~s(e|?t2agq7u^^Nz2U#nem9g$Dw!H6wiULfn^Fal
zA}c8ffFJ;#H;@*pOku*+h@z85F_}ISI#8YDy-#=1JdqMqBTtrt!_M4&)sb}-D)|_X
zp%3JLN0C!}8<Bp({+6)I1>AOrE@vF@J*C8v_}`62W8@-KOOzH?I7?6M<RZf<X=kzs
z_Q)4Dm!zM?CtI{^jZ0eOBj$nd|H+++N<_2k{>FsoT+#5E{>xvjsjsp?cO3mnhkuO`
zzlEaDuV;5zBok91g2VD;(DfVp3Ohck?ONmlt-egh6Exi~JKe}j&wi^>i<n^-;$=LS
z&8?r(3QkeP$n31JA=UYjMgSy<B+viD<Bu;?X8OH#H`8M6QA06Cr>Ih#8dQg4*B$T9
zFkwg5dg<uuf(Jv-`u%<7;G>{r3RVocffJs8BX*M;*he9%rDM`(5(=h>xNN_U9L<!q
z+4Wnm3S%j3rvBCQmGWviq43f+%Tla?(7+~}HHQBx9Lp#~ai6lY5Y(B2cGER*Nmd5=
z3#T?ty>2Ls45v{$*>*J*5WRT&8u(9u%&k_l#|(AI)qAhX=91T?Ze_IW){g6^nxz$g
zR9U+>8X3Q5825-4ZYKNdx;cK4Cdb$6BFN);>$x&nO~-X&oE^7D+w9EoW1v6$q@+V=
z-h)7<xNy00RNX}+l4?mLW0X}7AP_#*&|z3?f3noLFtzYhKF9f3!ln-xW+IWCb#)Zi
zAWqR?F_+vAvWn{BhJ+Y)9WGR!p9xHir@-*5LSg_(`3S1)D)@Z@a)<vZb&PPhH7fk`
zwrithc{>$^OK6dgd*I};1hpQYQTY<E#$6MwvGU18vynDy$Psd|F3Hpcq1($ed0#ft
z9JG8N7PUq~>AnC%Z0l@ca`(K?h01d??{j%?zgazz4vl0uQSOkrJ*go_r-z0tp)c7?
z2ZdV=SUWW+q~iKp!i(1j@auHJCoMUjV=>0(KH^U`_kU(Dx{ixnY?$OFJ12^+1&Ji9
zzIgL9(2sV^%u^T{zr##U+aRyUc2V8JG98Y5|Hxs+H8_nqTcMRIpZ4PTne>J0dmUyO
z)M)EhO17_2{-MVC(pE8*o4x#OC61p8^zs0mRW{J<7PqF%KwK|~uSKLpg3nZ@=jA!D
zSw}A0E1(5#H1_e24=N_XX-Iyn0=fniD?h{MLikiwMbJ<UvHuyTr-y0wfC1)e%FtjH
z2*SQOCzw%y-h>t4clr7|Jr^lXjXHzU%U1?$TjttRssO};T0KeMc8cN$fAg8qR4_(#
zfaUX;_>C<q>TT+uX?n6;__u=!qyCL_f*BW_%mIRr4dd&0)zoyP67$6walZrmG;JuD
zOnVe?6Ec4se~zWY`^VfUQ?5v7GSz9K{Idrlha?Io#z&RtcV13p)SY208ZkSg5++Iq
z7kD>wTL!gD9K)^b;%3;IzR@=UgfO!byZ**Do(XMJY6XjX)Lgc=fOKb;0wMl;eU>sU
z&a&wfZpz;ouF8Xc^y{6yWwLgOL&EMyQuzoKV>F(kEGr?eM`b(~sHX}b%!Fc-0cMk8
z4K=X~^Q1Oj<ahk_w+z>Fk&<(AfBLbP^ZDaf2!ih_$zE4*`jij^&>~~p7qREzJUH7X
zFfp$zdd3mcX|}~r6-(o57)L$na1I_Nz}LU~V@7SOQ08V%v7%?#<1E(Z*D#<#?0>!N
z;?Gr?8KaZW)Sj+0YuoD%B&{f?bb+p_#IeiL_0;`TQR^&X2Q}$(_;SwdgLnC8nYQ^b
z(NMC#qr_tr#1jp%!e;xg36MsUEK_aIjjwEqIDM6HBHgEDkeOLL$4i3hR-oL!Iy-Am
z@8f;Zq9aPxaWBq9^9D7_mhyOWEbNyBZg&0M>sa;afnX*NETdK-5cl_x1)+xeh7@82
z*=zUydv^<63VD$=JNq$kq0Rj{r_opazYz_O`eSohqej6>-1WaDboAdQJsH}#Akrk0
zMGR&g)0?~Mu{4$)_uxx#mG9;a{&x~5ygBh|T^T_aGDwApBwnAzD}2b`GGtie0iop$
z{^=b?+8{sEHZ%6{Yn@b14@&F4I~ymwBHXMq+@V8Cjwt8r_Lza-S^mW9%0k$fy<oR|
ztDa$oa*{6V#npkrg9ara^vX%Y#?jHg!#P2^WVZ=MNdU*tE1-<}bk^FG-gy(HfdKs5
z;mc7_wZC@l-NCe3a@j)?8b4Q?HQspW-)(D~CA-G1?ZvJBwbzhdb@(QVj_BnB6aOfR
znw`x}b}!qTMq;~xK|X=>K7<e=7=5#vV(hD!dL0($Tj!iGNG=K)fm6e$o6Ajc2k0@1
z$Lg0;F(yjvmS3!$?HT(~Bx9(b<)o{cASx5!2>0t_U-9b0%qxd{Jwo(4x%ljTz?Yr*
zW}c6`V#uv+V=`PYHqFZv)5BbL%3{LHX!!A=akU3kUi6jkUYdT&Ijk?!cQxWS^UU5q
z0kCtzNu4#lKX(O>+s+@BArdmNd!rhyGU&6&eag+?-%_>2?CilWvsX|3lc>LigJ(gq
zsfxm!dbzas=tOc_+P$-QnpS~1;q>n5SXgiFv$M0U4!*^(G+jZ`o(e3WAY(a&LSg8Q
zLGX(Yk%}UzF^gycgGVmGqGBwh8M!?KP#4|WPtv3f)<2CtGs!f0>&hv~EXT4D8L;I&
zN3&ZcR6Oh<vLOkj@wYXQ4DVxFhas4!5iz!=z#&3v8QqQ`bk*oPb6KWGnA|yXgf`Yl
z`tR@d=@c8_G_Ku8MckY{iReZfv?*i9Lm*#358ie1DOR?~MvEk{-bvtmi){Cb7`cZM
zg+aoeC9{@UOZx%&PoR96g<v{2_)F!e3cX<E1t+-(7Vhx1voanlm**bIbWNXFxq&or
z(2k*Cxy$<PZbnghz9&P`k5Aq2g%E&)bfP~}U=(wv|FiQcxnBu{n@ZT4PJtXf`msyM
z;bv!3a88Wo!uw5o8aU}|S@@$7j^+H`fj-<CFXDDj+%p(F%+NO8<JgZt@=jl}>8gcu
zrWBgNE-p0NbL=(PPPk_0{ZyjA;?(e)9@8I^xay94?WC;L`XI00T}4IymJIvq*t2GH
zr@CgJp{iH!ATJisc#M70-Z#Ek?`(=YA-=+jIoyLAKA#Lf`1X~aPq?DDtZn!nCH;$V
zE{5FypRO~qRYux~IXN=2a7NRPtpiEBVJA;ECly<_zX^mo#7OkIgny3kKiVDUrSOP>
zfEgAeOqBi{*cOpft#)=+l8oh;M0K;DQXTTBHcF9B!?$F%g2jeZ<;}+|t%f^C^m<KP
zAXzWffH8gl;pWoM`-oJsnsz}kVOIxRGbPn=1{qqYSTNgVZy_uAaC&-u&s)Cd$rTQ~
zuaq{%IYu`BE`(4DPu4p^^+7%^56(P}(9IhL1?f9JH!H!1s|VGxdY|~sbKVV(3>KC=
zf2JPO^ftDr^6xYF5O`AANF6Bp!fUu%`mw=?-&G7_Cvd)l>_x`7K^3b4Vg%LF38b>5
zxOo0ySeDLdP2V*tM1Jh63N|g7ILa3FUMo2sjVkc&t3A;a7xr7qpJoZvt(lN)vwt_t
zQ#|UrfNG3<$!#*i53O1Wu4f)K$Bpq0ZK)wjQ<MI<CUIjD3>MgK#RyD;^;sNi<rkl%
ze6FjUXYdFtk>gQ=I%}p7n)-KHz1gCW!**T^^9Y0YV!p4q9?*9fi&>fKJfOxWtqA;Q
z%v^$ByR*wAyLmOku*Cui4AHInqpfxBVCT*fCMMWI@P^ix;KGnNy782gF$LLw6G|C}
zD0JN+*uoH7;Gii=Qo=c|CtZ5*jdNwP-8(#(6{Zo0dpVBqeMM}#$Vr2sF3*g1lu2OA
zS`xiUC#6Bp+9DHj#%-^yB5T?x*Rc4Z%|_D2u_;=XjZ=o>QXQAcmMjF>$(rdz?G8C7
zupyUDSaKdLH88LVf^6jd>#Gvvv`uPyj>mDu5<Hk<$m$?$7B6{t3b5L$pFb|V^x}OD
zE!#F*%eVf{42BA;^g|A_W_2X9#?Szk!d;eGQVD}cPU2|AfQ(0j;GDJ0w0$~PEyP$H
zM*N!8j%i&Ww74UuB0DalvJRd--Wv?b;uG#200mnFMS*ZpVH73mluhnl$I{ORJ?V%Q
zX6e}59;~FIyxE!n3PC35JD!EK_+cI)cxB`i%G}@9KGS7rdgqrj@SGs>J19?2PnY0)
z_?_)%>h8;_yb4*DCHJGq_{nl+WY@6jPcb2SZ<#K@G?18JQM_E1WxQ_hCOYyT@!;p`
zvnmEo4JyEWbujQ{Okm!fGvrGkJbHaItrwa?YJA3v9mBW@3wC@(o<D*#JIfu9($iV9
zM-CXHN^^=Gvu7$X-{9<%yVUd;{B_%}oTRFFK(df)QrSzm?#jo^L1F%p@SWZATK3F3
zu_BZ(mqV-!Y3Vq$|LPrsex=(y3sUpBA#@vKoMUDcb9B3e^P`(aw4Bg~&(P{-`&;@N
z@QY{M1e(Jn?0$6z^Z4*x@@>bh?g{$f9zP+I8FEAXzxMX7-?Q@f;%8hVxenh43m_x@
zjQB}k2_+l@jgwTDr`~HqPLj24Zcs5RC?;lWO*als((}p{`x#%^OTl4lI71uvUtd4}
zQYL}#ZEAZg%ctC@kNNSjlFW1!EQr^|Bgx**DW3+X6sk)W(P1oM#XGYHuhh2LA5?jw
zE7=zB;N(Q}ktAN|wVTE1u{!+ZxF6#7W$P!I2R!Nt9%$(#!y7suZLizTpN58?N?Xx@
zA;w}~%s|XL=*Dk-WYInyd0|_ZEHXIt6Hxb0N~V@*OEmMv##GbEp=Dt=!I6@Q7$Z=B
zB*wQ?GLi2XX(0W%kGP96VfUCoX$=Hn7BXTbk!j;f5p_CDwDv9!j6=Rq!L(|I{#CEE
zg8UcdIxcy#MsvTm)GG@KK0bfT^HYUx*eTw~k+RprAsghhAxe+u@CAQ1|Nd4LgJIl+
z@tDyp$-^-GCp!`=jd2?2gCDAmu3pfDM?Xkzzwr5-<V0h?uwgV4>LY!!dBN7ai`H0u
z=yn-06*}#gHo6gZDE*7Fq)eSqqr6mjfynPBP_ev^r^fKFjTvEu%y|gg<m^3qp`6=|
zwQ{v`fK7I)X*#N9`8hYb=%s!CC4j{HF(27>T6I52T@7NJcwC#DSDHrQgX66Ei*n=L
z$6z`|=+Na$QUQRz-?;@!A<DAgc&-`wKG%GvsM+|#{HalxM<<^gtg6iLYJV}~^xG%y
z=uBKh=6KM7a}AvfDo;(X-5D7$ZFQ?KD`Cucw8SyCy;OB53paUw0I+s>o08%w*?(?U
zO2+6n#ctAgy?8KboV@(-dmqwvmnTyn`e6uhh(A2@{KDGNr%9E|M_I10&4%h%v(kBa
zB!af9(0AQ|k5FjH=%d0}9uP{ut>fadQj6@8C$kP*Ng<-}t{;MMgUNQD+3EC|R6nwy
z>tk{8ZPb!Xk)q0D$gZA8sZsjA`CG64G^u)7g3sDDhW?2xTHOFor<8L-OZKh1=~fbP
z|G5F*A-V7u1evp-`+83ty$|o#@oTEenc#+>pOFy@iOlOnoi*|m=w|wp3USl^s4f>h
zlYIVe)Q@D10B!Y+l@%|e2U=H@Cx=>8AJgCoGED-^pPZwvpKrT#EQ$@Hmy;;VV&V1H
zPEJmiqCO#2t#+D#Fe(CrEN{V+dG&W%;$eCB2}c%bw1R3^2I)e^&Q>8(dNY~#OTz2{
zDeoLRWIj{a+~Zl!gAx&WpM%hH@pm$aw(<xv%{s@{l4dLNvmc($Ps}8E(6HKn@~{uC
zzjYrnjd>~bnN}}2eVO+2<0ln|C?>X*M`?FZoF}U&mK-XS!dS2FIAbsxHVzJCNxnMO
z5~@h?yZ>jaWp3MfHpJ(Xo#t6#Zj|w_iUr9y+L;ay=ZyN!cE2`-TBKzebDM5OgR!-_
zaJvNI-L9~)&XIYv_i9e;NwcB1_PY+LflSA4CXy-_Q&cbr`X$3fc~SZecE{-d7<*AC
zb{{r4mae3qP1H8&OKjKa$~cj~5GRl*JIWpw_$z$wTw1u_huFUA%IcVI4xRnbwR}wU
z?{N*u>ZS!u0o&dD%o(OO?8rP^8W`@v+qebFFjfjFw8ESIe~UP&xEIE925e(CC-Cpd
zN_{o-+9wasDXY)X`EfV5s+3rB{iDd@wWt!k#xvM&jGn7+4hpjmd8pd7-ELDtZ97FV
z#i23FV<hjR`1m%euOt|PUQRI+Qq^In$yo(QPQq-(N<8h1>(JyDste$3`uG`Xg$1Jn
z3+RMl0zDt>eB`TP^qSl5zSjX+R=24}LB7gKwd81xSu8us%INuWwxX6L>Q?-7(lk;K
z^MLOiU{uz8%x?~vA1+d)6v@cVwKoR9?^RZEyQtrf8wSVCP8)jlWF3o-a4q%m&e9Cr
ztt#D4;R}MA^w+|k#in|9YdgDN_H|@aQ<EEV(5yB-?Dy5?u`r+<ReT^RO(C+0GX;DN
z=$KBaa;a&VY=66vSL{|r9BH=Gj1WVF-by+DIvmO>xE(j0Y!{Z~Cb2q)lQg>6jOZWb
zdNC_@p)Z-=BqyXcB&zeXrR2u~TRwSktW+_PR@$)d<gLQ4yffg!<b$YJo^r|Fx_(0%
zNq}Xetfuk+UR^OPsY^(#=*jP@X9(A;XDj|f&lZ1lru#>|<5lCoySR29ZRbo1f7G!S
z*Z9}Ank-j!abx$ztXltyO;z|88u4j>JSn7QBmQ~SZ!jh<U+Q!9A93YnO}aj-0V+6d
z)_fU7lfH0DDH@*vUi*>Vh=_<zAX+@ZpjyA_oG}s9dceXVXLAZGL}?yaCiE(1gF~k=
zY>l!|N@8_t)L0<=ubbQ93BjoAyEky!ZCK8HE1bOigKz(JiZDUWBs~0mqL+m9;`rGE
zeK0&IVWK?ZvX+kaxjO3g6e9bd?Yc4^Bb$D=Z-*CwSL;CR^y`bi`%cH|$ei|CCzb&N
zxS2c^)qPF=E5t#L$Yp3@$YT1oXs1oYb63*eT2EmXd^|oUkg2h$zb-tlK09%oW%M)^
zi?7ng&7yLwBz%6#TB?0aOXF_T@mTips)761F(kZ=+I6dv7y+rCyy4;DC(;AAFBdF%
zBR|uM-m38w)$(!9Ia8j=Z~bqggc|pE-8V=f{Z_Z*%*0pQvMyeGaGPwIVIDDxN9pkb
zR5T7=M+yZU+^DsL%skq5{s%b-RxQX~ketW~yqQ+d3To-|$V4k~MmV?F^5y6ANxC;E
zEuI{ETD7#Y(f+NUH#=GwWixC=>7Up7ee02Djv}tj_n^pX4)yKvaC?2RP$t}Uez{|o
z5%DO=ujKoc6>MAvklE|v7*JzkC*Q3?<R9@c6!hEa^=y~o)l413t|;Kx#7^IvQbCwt
z6H9`|$uK>T`U};CxFr|8%N#wF=(wT>ef@iMo%yXF|4!CHqozOT&(+nf*6rGSePwMo
zoBtjQZICs^CWCks8w+SKRiV)sX9L4fe+bLP8f47Sx<NdLunQHm8N|uXZ!$VRg5WuY
zsD;}*Th?oQMW24`B`E&GKxlm<J#e{^N|4K7yW73Js3=l<ObR&|`SVup9Yuz*V|MMJ
zQP2=KX6N(o+Va>8nKc|7S36B{?@>@i$I(gGu48LHklDLF*IFJ-#^0jk@e%Ew3b==z
zTRQ4B3i`HbXih8{uN)%zSf4L=1G>8Xyx9)Z4_1dmqhIUsWCjE!CCWCpB>N)eM~1U^
z$=@23u)y|dt%O;F@a1eubH1$ho4h@I#?=X8TH1sP<CVh-CyPWydIvhi>9FIf@b*m+
z8k9cCjyZ|Y29d!@^M*-bPn^~FhFymVgoN-B&L$PV)#Ld0CVjFM<kg`@!}8?tBc~7N
zYowtRIG<xbr?6rzxqc(%TB~Kn=p@K8mK{TX>2&*;U?)v?WHekj##D4a%u=+9o!nbb
zAtaFC^fN2ZgJ^}@x0y*Bb@C6HTlo<)XgAitBeN)eI^aD#R0O~@)wnguv5!jf-Ws@!
zzVF|CdR>Qk(gSHfzgB+n!D>BpvcF%|Ao>Pky1F-f?TYUuDv{oi(MlZtmX*GT@PnEK
zxpyP4g~{V%jIUR`Ra?c4VI=1}^AlP+$zorSkbh!#5%G}JO(_Q{8~B0p(HTdm2ZqXB
zVj%*94r}F~9Gmgwm0m956kqHXX(i_ALKX;JNKq1>V76=gmTlNppcsruF`N|anEQIc
zsoMSv#ZboR&N-AHS5Kf@RC!-VOjCB>()V<qS$>P#N*Wr{X(MfmU}7?cN)eF7m6~h=
za@l^@;d3Y-7E*BAbyHt1%v435QR{{4_uLcLh_e;E*)twUXdI))&LPr0U{Wf<iK79c
zM-grvf%GAY7;-IfTfc1`s_2buO5-;wYKrc@|E3Ypx;-)zd^uBd$=4w3$;JOL)!CQK
zyvt>_yCXesY6l4Cu@b;{YkFVeL&aT?uou&@3XBb~LRQI4#e!h(fGo<GM%|wy^Adg*
z%sU&P8Ly>Wtl&5LZ1pH=w&FMRD(B-FHMWi^3>j92Le&nUWBY*DP+=45BXP69FCB24
zW;)1i95{jZqkPesb5NzTYV0!b$5pYHoEEozFhazSe-JKT@UFEK+^nfOm)b`v8Md1o
zsd|zj%AyVL<}wgI{Fpx1+ZX=}hoFdj(6^6@vtXGY!wx-xHvlBX5SK_lMrF-?-=8&V
zaoNFSro+{Q4L>W@!H5W<H6w=>4khn;;&Ay{nVDtp$qu@1#WbrgvMBPZ{WwlEiz^Ov
z)2cJTSAE=nE2kfLv-M%Mpr;m9cbetI!Z)A!&i{fMFY@=}C*E6$U11@`A_A7zdMjHE
zD;YxdfRL<6mrvxws1jN%$4+B?3rXzab*{sf`c<hTz(@QCMT6aP9U5koQ<Dl~r37QV
z47go<p;WC2m`1TZjenxPe5dL(DKTVoI9^%rjBQTaov-X(#8tL!_4L${=g}(Dmi9DD
zgv?#m=eCAV*`AK)h7y~A{35Zt$l{q=rC9PQ<G|b^9!ZPgpZ#izDP$)Vf}D;fdnS$O
zh1sO}{f@e^2;2lR=})%FoBkd$)z{bO?>;|Jd_T112;wa+?z!=fIC6Nb7m7<{o?p1Z
z*-mrRtTvT%==~)=yz1JETW+NVQ%X)k_&4DatG6Jnn=#gB=5FU;Ex)d#O4)CT2Q^WV
z>O7rLpB7P47dXNrn4<cTrf99r1D&3WdU?j4eJ{ChmF?3petlNAeg`p2AVUw+kGEwQ
zuh5WM#MEzakd*&<A%%W2D(!(N{o`>`K57>fDW`-m+A8DUZ#O-BJ-gNaAF)=ur}<8n
z&YKHo^DR)e;^Y0<`-jJl!vfQy&ORA7b_AWOzY%*WlBLjVJU3ORWZsBn`-N6H-NI!$
zk5ZPc8(;{dG6%#~Rvnv%6qw-~e9<J9-U0a~OfVtSoeYN1lOzEx-3<hFWl}oP4DF<U
zB8}|dzKYvRoUK?~A|(63Ff~vf#FUhuCn3@_n=PvQYq@SOYn!_6JaU}qRVz9fg~ilm
z*u4@4F*XgXF%HscWTC4;Xk;W^6#YaTnbUijsv|5@n&~}Sa&SE$IQ)<Arj32QKzY2o
z^NW0dt5s2LS-H942DSi>332((Kg9W7$1WsDIa}NU2LO6o_7=D%y6r)g?Z5I@j<}7%
z_xrU67O&&hsVSFT<Tp<C8s=$jU2vIfGoT-HNBktAOOXXh>olTHQ*a+?*u&z^pqx4(
zU2#f6Lc-dEO7rah-chMY=l<^peIf`6$;4*Vk7p+g9l515mgK$x4q#I*ucC=_WQoe%
zt2lu9Yky=XDfjs<7tP*Twna;Nbqc5BRKbsT)Yp&P@-LYmTz~m{oo`@$w$xV}9eJ|O
zlKt`r1CHPGGH>`Y0SZ4#(!;H(G4?m!uS`M}T*obD=H=DY&X)MX?~jb$|8y4fD2K>N
zIi|VkUY!$-5$iPEX--iN-p|T5SJk-#ogO&?goe@yk-~_fNH4^02B=J(p`Hq9F}9t5
zRps(-Hv*j_^R+>^zki_3N-IlSk&37+`SvFTTlmd8pJrT*Oz2vC0%VTP+r0eln!iLy
zdBpr2yAD1vx-oqunDUUPlcX7CS1R#2BoU934}qcy0)ZC%77p2JJlc-A;UO(jNTs)r
z`K~0=j=TX;LLuJv?90CJ6XNL>AZrdbux+N?l)-nRe#j}D&7)q@!<krx8Js#AoHXuw
z3QbR^k51lcyx1Kmccq>^{=cs06jzuP)+UP<i!i;DB{+fLmRG6U=aniB*|wS8``csd
zVRg*@7y!YcEm+I86u=V`{XNg%#)myJY>5MOO>t~Gt|@>a7rGJ%Hy!Jr9v;-y(Fs$X
zdSgK2ws*1TaV!IF%@gq#Z4$7twKdXN=#J!*nlg|&wO>UzMP+V=c&DsPKf`g3uIVD3
zlpb1G(~rG!rH>}Qu<2h<Pqsv5A9}oDfMdVn5yo<wx2upAJ`;P}DEX60&i;IBux3^A
zwFB!MA@ul-wu7FF#`r`&cM!!?LX>bPgwircoyNI(lj#k0<k{I$l^c3)d)}(p5y~ZS
zddM<vMG_za5-s%O26kK+<QKLG4ROZdADT7fdb|9bbW**<i{kEH3J)66N#gc1(Mfuo
z?+Fs!?r-YoS7hc5GW0r#6|salLPzCCH=Ul-SgWgu5sKY(Y<QE3v}4le3<1ghxjT-0
z%-d7si2HlDw5~u<J9afWKT<)Z@xI_{k%+Z_z(NX%6YmgP?7D68X~sPu=kybQ<-h41
zp56MsECs^J^}Dx+vbV4&QZYmzT(=?6fCldVXCV{%(Oy5X0{D-1aINR@jZw@(`ovwb
zqB*ztJqpmdXIT&RaixE%IA7QG8nQ%PCJwOD2I)fnkOnkxJFt)T;8u57{Q7rESk?mi
z1?K+|BMN#Wjs!QHLr8{|XbI+*!S4Pd;iBnES7YNpXU^P29x!g{W2Ld<fNMOmo!>!X
zfT1gH6Vr-$gpPj4xtV)5^QElV*<nS<Q5GFkdVH19&Jy>KN}A%yPvjcH{(VjRt$lkQ
zg)R+V13_c;6w|70cP0=+7X71&7nyUZ^^d-9HE*wtDm<wF_49zmN#Wlw=s?bQ+TVi^
zr6#+n3_mLP;WX~YD*BJD?moUJZTaXP{6WM-M5P2pMFd4I>4->4U%DbKE^$#rL|R1T
j`EKd*|MLNRN2>?cp8x+3Y^E@P+!H}=s6eGwGJo<vW@uqj

diff --git a/docs_old/logos/upenn.png b/docs_old/logos/upenn.png
deleted file mode 100644
index 36fb621841f4f89ab280f4f040738a312174b601..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 97063
zcmXV11ymKy+uuufH%Pa1cb9ZaNjFG$BVAHb3P^}j(jwgr0wPFCN(e|w*EhWX508h#
z-Fxru&dxmZt0zuVT@e$F3=M)H%x6k++7JX62L7|4Ac0qSmg5k?F9bVjHE9T{O-8@9
zc>J5%N=aJ{f&v&JC@d0!uE9%TI}qf}1wp&!5G0ZfLBwvkEm~sW0FtGOq8#+__^+V-
zQ#yDj^h{1#*Kc7z&)Z9PmS~_;l*ktw2k%wu<4;3_&`bA)sTiFlmV>H()B^-e1Uwn)
zgM*EWJMMHGWt5y@deZsMFUtXZ!(W0Ag||g-a@z&rWn5R^JZa29WK3f~c|{$&9t|TJ
zmBqS~&YZ;kCdD{O^&4YPMV}fRLRk+y8bS!*zY9RuKc3*>@DXYc7MYI(hnTfnl1$^~
zB6T4)m7mlwr8fw8ahy$XP?#>GY&#jHH9q)ok$n_GvWmi_v^}c9dlbo36lyYPiZ{|4
z`|;SNImro>2GJ|ym_b8Bc;O7LV$wJW(m3=GrX}1t5%k-GD96x}2j%}?JbTid>;LZV
z@Smy?5+k^>c!ygUJ1iZ19!nuV9WDACh6g^3APqN(v+B7F7jMnwKo6mU6TPB_*+U4F
z%81l(P_3Nu`G2?d3ZWpp3+_!?`hS1ng&{FR&;I*@{zwz`2rN{oR|wSqf8sbioG|Oh
zp?F2D1d(M!^oUB2mx+KXg%|etTZ>+k*8asSMaD@xSm`}#)nB2+CXH_EXW$d!gL5Nz
zcU!8AvMCV#KLto87;&f7mVfNdEQJAF+CLIu68fJGa48U<(BlXp7LajBH5*_&C~~ga
z+9~MMqr937#O>;tdmzN4EkY0FiOvjr2H_deLNoTACLH7EPr@`0x2@&DnPf@0n#9EB
z>e4Z!;Zi24V5Kc^)=egF3>(mRK575=4*0NPCE43(Li0@z9wbyLnZbAU6q=GOfxOt{
zb0|7PiQrbESGmm%m45u-r-sV#P*A0c`rdi^6zj)mv^L68QmNR2>*1mwf>#1(2u8d#
zL@HNzb|W!Np+@PMM8X<*q5@vWhn3orWnccrr~|_nS%^sybq;A0aDnrQA{4x>Z<hOS
z%#dO<nHlLG$3zw?_pNzz9u7v7gu>8}mIn^JRT7~f08JAsnhgxJg*`m7+C8}IXm+R@
zd;(ko-*29U|1QG&>Uk-Hd8XLoV~7^?(){l{gBEDpFzaFX$Itq~f!PE;+p_W+o&Oyj
z?C~`>ctwjdDp?M2C34S-nGqE!rNH<V6$FP`L`c~mftkD`G#w3|4u&R-$Y-A#n;KQB
zr{D$xZ<0a2x$oH%B$H5r@I-#0rXWAw`f5u38{`xOYKztXU0u_#<kcYt8fA`@<RUk}
zt&DMMM0j=F)BJZC@(3$rq$YGiD%zS1Psx&aCAzg4Zx4j!Ft7_18c{YDsb9N>3*{rO
z`)}cRT)Jg`C~(v#AhSd0*3OhiVv2=B?~=hH^C0tnJb4x6Fnl&SYOj1v7glXh)X-{J
zRD}>J<vPNgGVhQS>rwX*uHrr!ZJO&@|3rGz$0J^!Zuvd3oz(q^#<IiiVJlHBwh1`m
z&ra)64x~yXF8Q-cTW`Bo<_!I-X6>_f70viy5Du>x0{J{`xZP`nD$+g*%Kef*%na($
zG0YMZuIev7U?9`LiOQX_r}ptjrS!(+m|&1QAdSSWuQ99rkbx<wbb-a!t1zml0xptb
z0uC?wRDO`dN~k95W3T$#)I4<Km)cE?evbM?5{x!D*4D@ll$H^&RYrBG`hUsxY4@l>
z`nKlyh}0(L5!O{Ki3Ti@7})1i7){D)r?)qYM1<~3jS|1j&7J!cJ2^1FSNZeL8Rcw5
zx#D2ub|e?n2<0}E#eA}*B*{M_vFUpzJg9ieUi-659Gnt@@U38u`Wz7>`Blu%f1++K
zQJ6cpyxQ*Q7V^86kcb*s|Dg6l?)Wp8*cx}76g)#ZT^C)w#}PMsl7_;$f1?e=m7bs_
z$V>27=Wdn$ncj^Nr|Pexu}U_$B#I7yb_m>G@w02D1|{riG7bK#L;F%Due*}WCzL2P
z`Je7{UQ6WdRs`L$eO+Lt+A8LsptJPzb6Q6;!++i;=%YY8!Dq%<Qfu;R*D%#S1vin*
z5En6jCV_;~^lfJN*Bs<qwOpDBZ%l9ltQnrt{=Q5o@_22>*s@A;LF0$))-qli{z*g*
z;bdpD_D@m8c=u~u!_Dbht3$R~pAD?`*_ARU4qwf)CnyWK7*nx@`Ftz47>uV{HC<R!
zIZTv`-!vY#n&0xoi7QzRlW)}xZBRxJd$aaCwi8tSxfU-Bk;YOOCFdbbpDoq$`!%mY
z%avITTSCY)&Pke?WD#U|Fjb<T>dzd{jcuU2);8yGi~XWl(;uNEvK>G{iSKdh<cej`
z-Ygg@Qnw`$DQS~6bno%4rn3<`;%32g<;8B}caP$m?bGerE>zg$ar3SV&-)_0j;C{c
zuR{PAA^8;(_Vn?pywv^C>GX|?1f2HB!%k*cOM^`QDCppjrXyt9Y*YU`Lc6~9f<J-q
z%qHNzSR3K=e(igb5btNLk7Y=4Pu#iTAzZ^Jc(rclA)N50K@~+fNR0Psf}O361>HBf
z`hQY<qn9Y;4z`r0Y)DjzPmYHwDNM7QrTc<h@F;C4xIWn7dpYW%C++&lbUoBHUv^**
z7S>}QO%#$VsUyCg6X+8^jVDRl!PL;oT@Inrj*fVV^%;k;29`V+RUbj9t<7s;WfL8n
zO)u9&=9ITkXnaEMrFPsOoNyw%?rPn+HFtL5=hkdC_#X9A>{j?QHqx1W;`^=ro)@m6
zeunztSvG`jpadY<nC!dl%>}*j^c=gOL=2)N`bGjl26cZZI1FaxA`)>7MtRD@_k<lx
z=uGO$!vnVnJ^HdM(lHgzoWjRHI~BVBI$3?NTTgkLAB8SxYWb^b)bCsb<#87GOk?WR
z7fmwE5EDk$n!R8GYcFOi-_LUEHO6bL@rf4{pSf!S>Hhs$f-s%@d*)&v&T6&dgaYny
z8=tg=YZTTtIqN-H28%B95&?XV@|X`#&f9)?50e;-D;)1(LpjFD-Yg>}?A5V4Z%vwi
zdU3xV6zPy3?=0Ej4~542qzuxQhN>uzmyU%#Gj+^|2MBcCvrn9~m5~ktZZ>|)yXLn2
zL!dq0Hf;aIOli}Py2^6V{J|~hczKBQnb?oqjla%g+%?8l7=M2esO{-NI$>{w%KjJ{
zbX=KLE(RmYq#|I1ef&f9{pponI2@T*l2fLh`@%=8G1VluhiSO!^7YKq)|6O3)$Fh2
z&kx#=AR*-+ID@>%=t!~w0^Rpt@J|Rr{V|yX)tCAg^0JNvs3{;cwtEk{)Dg|%US|XZ
z`C_MdPBFP-KbvxcpUE4BVWT4!45Knhl}dagm;d_HcnghVNhzz0^#}Mz%W+tl5@()f
z3{dIZ6li?9UinA#kVTHYY}x<wv?*joaNABjO;?wozekrt0d1BfT;X->cFy+ga?cpa
zU8_Lf<np)47WWLe;Enl}?bmc)j|R)}m6V4e_yYH9rhjT7F!Y?#xtDXh8Ir6s<sVMm
zOUTp}be(w&i0W|zc3iK?z=lW<>7B4%WxpA~9A&+<wtjzl8o)2c<)Dj<aqyjfBF8JF
zkLjbu!=IZu?|($MaCG%0{m<WvClAVJ{rtjJXNLaI-fa7+hFN%&lM=?q+h$J1TasGs
z26c3)(FSPl&$Ak(*e)WcTv4gMqZ+Ss>am<jI*sGTpgn&!qfjL9ARw-W=Q{hwr=H=+
z150vpOlz`kd?U@mJ>O>3H#cSvJL1iVD4~ek`e~jKhV71LoS2rRW@af|CSp9CF)I2q
zsbhMoTO*lugGHq9kKg1bw2SGvEJ0Q4{G;>o;yS@qVNk$T6V(c-u_vqw)568J@|;(0
zu*Cd?D$xCUL<b5hGfAWzuKv!-USQDKpUHWr+3aJ37X~)0UxZ=35)gB!$0Faqze`fy
zupLukLK#z{tKfJNO1Pva#{(Y-^4~VoMbK$3N%j_~>=|9Y40_(+pshzY8*s~m<4bw7
zcD1J=<Zt+W_%-Py-e<Ds@}KniSFK%3YPUtauUKq3MPF$Y#@o<g8BC93$QCD3>*5V`
z8E||bK4guc>!O`?0UL1<nEp+^CmPu~WWlG0vUK=0hrh+?89!Of+*&xsNva)zidf$G
z4iQ>CZQ7Zsr4;#E6{Wms0zvmwbhQt^#h1>`ks-cul%CSdm}y~0VXdI2k5#U9i!fzC
zFP4-1DeqX`G=93Gx2}KB<+o=Va$+3IK~bSt%U$m-EJ_WVF&L8%B44vRmGaXJC-Xjk
zzcw3%?0HW7Zlu4IEm0w(lDR1^-TbKvr}l(-0`Z|{i8|pKB6YSQif~g}hdN$Z<ZHAp
z(gSZTEj5_m&lZ7ORg;bzUk|R>b>rY#i-oail!z9=0ZvIT=`Zmm34H=_!+&b3@9Jd=
z#1_J)I2z?Wp47_4i!&Wqo_07L1ykMkA3B$ix}|+&Ezam9Fd(I%vvxzUf4`eNK>EI6
zw{^j3g&_Hx7#QKGZDQiynPCN{1^CgrAw1=y?OlBPl_DgOujM)0E1&26NqMs<kVflv
z-dtk^r<rQoX{)wzdH3p7IIi4@vh>1|Gh~O3z<x1y)tnK<!ieD-Ln`SV@%f8?>e5Ev
zRbb8ENz7ZZ$|sF%q&n7(LwNDEmV5q{x7)OJk?YkVu6@cDH2AM3MySng%ni;fcGW&l
zEf+hq?A|G!Zo6uIsoG6<!3%3aMGVLBXI47fMwIwig4@s8B=~J<jN~(!p2CAm(qy$~
zgbaJJJwBWSwj}1ugk#x-)q#kwbN>H`s+K~MrvI_gpNypp@s3_xDuwfF%2W&to8TKf
zFtN)Qq{8QC`$JG8J9!1~y2i0EFBd<q`SDn|-<81SyU>e0>=)D-BV2Y@TA<oW|M-{e
z_z9jymoCpK5+{5&z4n@4oU78;Gx5nZ8d1QMZ#D_~(Jbd3*XgH&rmG|Ns$anopf;`R
z*t7NFu>i12oA00M=fj-L3rG?{fsx`f5|{DHpznz=x0)}PyRPTCO%Y6-l9rkm1KDCN
zbYwH<Y)4zpzow@!76hwbpMR^6rOhbx+4es|PwWbFNzxquhWm$9M(G2On=r+kLC*T0
zJ&mK5?S#uil)fUV0XMgS7AzNIUAet~P=m^GL=qdm2bYlTBE%L}_0qBUc}xI<<>(k2
zcasU@#OnK3kAN_}<?2&Y_L-+|Zu>PxvKi8PTn=&o<Bg`TCdFSl7O^Fhzg(^!)nX4@
zyLq!I^-@_s`MWngKtUz+%k-Wyeg{@vq2f);&i4)ugUtT8gk?~XiRrEd6unI9EbB7g
zUiIJim7$-@vZeMImD@H179tkp63_#(Bxga`UBBCC<8ADQq?nQ}dibgErN#RgsBQa>
zQo}Sv?#DIZ_sS6YhyeB=N<lj_?XQjApq8Fi-`n1-y?eanNK2wH53V_@6@m%XQ8=i|
zST{O~4?RzNrF*mvtp`B>=g8mSblWY-RFH$4u-oDo1uGdqKd~aqG~BH0__<46P~@n=
zAwwDNG#y?~gf9SefHE0l%p%&<yin}euWhf}Dh8mVR$y<aixGd-t3gh9+`;;Sjt|-H
z9yL9eg)dKUe>rt(*06P41qN?zF5+P1x8aIW5-(;YkDj_l(xQgse<g=HPFC>&`z;Fl
z;eLjsrB!7TykazxF<LgQ86qTV01r(G_qXPH_f^s4=VT)Ga^V~`;XM1`XRZaqn)1r6
zFT;egB})WZ1B5DqT*`E#lM8G8HmIV%!q)uAaj+heQ&G?oWEDO-k}nu$%7bmI$TI!-
z(q@&ov%Zc?zIS1B@`I_om#Z=InPU?%bCQPXEF_iMwGvp3f$TtG^}#)<+s>vY&8AzI
zL%(`-`0By6vpRN%Ude_hqQTo|S*A=EDbwckwu-l2;w!@oOg=imrWg`idMS0FEV`H6
z!&FaI<URHt7&|zW4|1{w#ZOP5u(QylzMJIKxwaMbT7atE?6HV2M7~tU`KJdE?F4Zj
zvot;MK^JXz$E$`=&`QkaZ>!s%tNH^oSO+D4wMOQNuo{Wn8os_NZfBryX18!~fz!c<
zIaiKR6FkGApp2OuLWq>7r)Flu;#@(Zqh=;!amWF)0lhk~!=UMaPq}rj$PP@$AI-1@
zzA_{8L-X^xqU2{RxnHYLpSyc>gdTFJPrl=mT%j{C2bAXkh08}Kvg<m*Q5Z63!7obV
zB7;xQFC+N;_C-tTw@(*wvCqaXM5ml!Ll13g3asg&8AD`HAD&@V_>1;90plI4w|LS(
zQ}LoW6AKF(>c~#o%7~U$?;kj25JfIbWXXTSVgW}^QLpzqK9{c!o-uhhEKwErUf2s?
zI|qX;tRs*7_q#4(UX{cOU$c#omsCrqA=u!N`rqfht9Vkb$CgBvF<PJ;rG4b(DK|fL
z?&jT~#UigdgI3aUZ3uu~e%o~{6DPd;T?Z#+`(@ZsshqR0+Y@6flNdWgrbMa*A}geZ
zF;6EM7o-d4SHVn&>uaajP4;>ltru2oS>{Sbf@C%bdw+fZxb7$N3V3&rm!^*J>!Zlw
z+l?i*panX#D2uCyz|qr+uuPO1MBs!qBfK|Y`=$eJVZFlT5i6|^4=*7V0s6!28Tlw&
z;&qbYx0ky5Q*0!)@yuU{JG$*x+5#w_V6^Ao4w>Q+4kaxAps^8pT&~*D2?O3ER%cYb
z5VOF0U&j~t#I9|T>Fo@!U30E8E>t<}-0#9UeV()@B?y|)^6rg>&krx7eIwVEJ4MRS
z>imdY{+t!_Df$oEKq9a?d%5W0=NZrRz4{Z3`z|FuwB2-d>5E;H&jqGLbRT&7bOtqB
z^wIyEUtQ@0gw_|uM&)p~VSVfG(7ABC-0uyNS9A_y?J2^mneAW^g2;AqPzZR%C>op)
zTzP#VD5keSubB05Hj&M;Bzw4E)qD|%F&U>762p2h4GlLVrj%nHR4UnAb;d5ZCFe%q
z$)Tu4ElU~xtu%_WtJdcl&J_J>LiUT?+Y6qioHAV1L`1L^rVj8&%fk8vf%HTzk-*1K
zXpT2Qz_VG4<&(+Rs~Zt%E&u7IHb$TyhtZ4_{R-JfM!0=5(X|9LW@F-Hg{VMR9Q^n3
z=~-=nT@>fSt|#Y_4cgY&BoGusi5`eP?ks#HTHyN}y@knIRQJg#ZsGg)aXd*3mn15r
zuaB_Shq68CiJ}Xw7%T20IPd1g<|+?GAA3JVZlf|E>fIJs3rS)-#k`OCGI}8bz-RBL
zm)(D!YPioe(y*df{j-G+;h80P_r#b^)f_(BVzOEVo-Ichhwx+$d1Q@?d3!7SM~6y6
zx*lFwq4LD~*D3T8J2aYf7JckAZ5|AD$iw3^iuEE5=f3k8E@ein*Xz{l4)`9oj;B1Z
zLR8atZ+eusU#IYx=vEk9CtTM&oKfJeY{{m`=n|goi>J9}r2F{}k4ZBh-rI;IO>y{;
z%vU2q=75Bo8xn<tr<ShV`6ga_lY^m%nAj*X%R*|Zu|u*nb?z?2Oh4*f{mH;_f1i+0
z%zvM~KrVeTyVm#9q11356R!+IG<jKJQ-Jr88kqTWp~u(aQX&6TlLDH7wfkk0mDV%P
z4q2L*WVP2X0YME1Wc{_v+r$myk6SGF9-iB^R8T=0SBrVNLKRD+NMjFs$**hgm)T#r
zh6(?f$EFW0aZ_E1ceW>4t_kX>!AiqLOR(}0s0#(#zxfWWqSTCbkTwc6Du8k2c>PnT
z)_Yg*Azv))>Xcgu2{NbQG~%2>T?|NxknT0g&*o{d$`9oYl)*moGr7EsvRu37wV&MZ
zZ3`fVgT0SoDHf&O3766IeyL;p_j<S=HEUEF2tG&aQ+O1MVu&-WnbAsY{b_33n#CK=
zxt;eipHJVCpmTYv+Cg3DKX^nQFzd`~esv)BApQirzx%IMr<K%Kg%xpi#`KK?e0kya
z_@3;;`#XGIUpBA5SXAj#<|IhG1HQe?n;#+&Rzr~-v6yCWuFg+IS)ZG}O5C4m!BU%i
z)Dx>$=X$)}Ad0glIVH}FKRMQImbCYKZ|mpz(!Ya}8ZYeXM~j2r=Bfc!>;Ah6UA>Zm
z%k0eP_qFIG(cLYb3PHRgjk)jWz#4{w9;Z4_eP5~0w<pzAZ2i}r3cI@eDV)eUNboKE
zJr$$z+cjX{(=?z6xT`@b4*3;5?KRWy`;IaGx~w(`J(l#>vf=4u8^v-?Q`q3f`dlVU
zsqwFNsloO;oY%jnRo%LZUNH--bhbNb!2<ujqe<vh1XEy1Oawc0g8bw!<=+H}zMAdd
zS8)w%!Gz87hX%S`95=Q&<=JXs7NsI=K>3_Rv<|=DeFHSHlh5Wa9pXJjL;C?6?l6u?
zSH|3&s4-RJZc@arf%4yG^y6h!39}gSg}+%_edGNC>Kf|!_Pt5{emh^ZjEMv(Wq1>m
z{hqvyPN!dZ)h%w#6sn}Za6<&2TiREDgu>uI9To#v*!1wPU3uH1?yI(Xezv9BTCoF<
z5M1t;DtPG9#>CFc(XhU2&~~=1G3&#2YYab-oYDonyWvO{Dp*Wkm`3X-l@l0blUQ<m
zUrKMdZJD%ae`mR+@e6`9zAe=*ltnnMSXm%a|0n)Ns_Uc><@T>_#7&+0`fXSPPW#n-
z`{MzPnV%R3mTPY0=#5>W^k<J09tw(Dr&zb}FAC$Ky+u5hS43QbyaQ>5@SQgo8R9Rr
zzP1!0)imeSE>Am-CjrQXRrNj=-}qDa_y9%sn!CNIHOtTaHj`D)ScWN+yz+yP--2uv
z)JaZ}fb3jaIZ|c3KEu}i@Zg^M;rESi^4N&<*pX(j2kHKcJA73&1TuW7Yf%QyY{-@b
z>FE|$_wdUf<}S90T;Z7N3)Aq@Kyv02Ku_E<4o8FsvLHYuTB|xb3@NX^m)WR^iQN&u
zovB6<CTY{mw10}hVuEkdsQqCpN3i*Mc$S;*{T&@iba=8EN?WkR?&ohe-VR~Uv$6_b
z{wPmTw#lx&#EI5e)GmNhfhrOeX1B?e73XzwWrKly@l6~C)rTAu*3BiyJx7r=HoSjU
z8<9u>L14<Vx;=4QfG>Ru9V|A(e5xs`ahA3=QwG=#G*P}yJT4t<<7wW_{k9k7F855>
z=G|)^!lln;*Me0xh?Y-%=F3GQDI~;3Mb@(1^U#A#KC`;mubAL+x4CQ+wm15-*pZuv
zb0OC={eybBgn|xe7BEZ`FZ&vTk^}#(dm6T!slTlKUW<t}#}8B%)<<O(wRh&)F4RFA
zsmuf)AbShB729$Rt|@}Ex1Cu%L$P|d65_I@KXa#XT2nYPyU?>$J3*CotNZ$j0^=3p
zdB69k>E1^;WGo)C0&mZxhMb&1$H0ysGcKVk4QW3VpyAp0GpW=3&Ftg<c6@Mqrea2d
zs=N3}AlQT$>KYzhMoChg3|wn>KBuNIk;ENP2ogKHiWG>*qNWx3^KNkILSL>^nH1|b
zDew4aA0hUbI8dZG1VkFU_oss0dp_ptBgeM!>3nb@n{KZ)de*A*gP@yNVFfoVBZI|K
zV81zjaioE}9EJJ>A~N9<5j7|i2>(Eh0~mO?PMAcO45Q4cSJbT4ZEhYA>9oSt)f|v?
z=>msy6MES5%)XQ(vGmH@XgmjskKE>l=-2W%pooJ9Kj(XrFFhBcG{M-^7D`mfruwLL
zeY?5ChDrNxe<ha$6qx-<!b_FG^+SKzM|tOw6$5CBrs9cj?Sar_VMtn$D@q+fH`!um
z0}JBP)>Zhk)RGzSO99p|d)grN2ESjHIjvx?w6k_&h&^NcWo&S)h&0!5S&rWY4IHGY
zkgJ661>85Sl<oraC;puuh?zO#BQ)u|Ct2QcUnRHW#M4iA+oyc0mV`l-C8Un;>EbTt
zZ!T`8msw|ieG`Y0c0MU_yZ>I2ji?Q+P|$PR7!ng&Fr=lYrDVV1>>K^s?T_~RSa#9!
zcCs;>yMug7btjx%Wngt6)ju|%k+k)BQVgK~GKh^$j(RWl@_FC7M(I{@lSE68DrC%v
zDEe^4ME2f{7xY?s?0=@fz5KECf<)4eNp9Ws;t7T91yoKGw#^}l7N?6uZpdiTiJx8E
zO#1Xco3g<|=DlmrWDxQV-2IdH85WLXYJ9psC4YIWijZ9xr5tLLmV~nisR-otqOX{9
zLA$omWx@GRGiPXQno6~o34v&$e<Ri#APC>fb@V`z^9g0@`L4e2U@vcIcYP(kCBF7|
zpCfItz$dz^8)}J&B)>Wt1UM-%p(Z<1)4lUrL{@m-@;AQUo-tOoK)MKI1lBo=x*P-5
z+@|(9=P;<|bmyfPN#pH1+7#Asg4?U&4Q(##6UJN5OGc>R`P5T4^d{Go|IEkeDoVW_
z1cXWgs|_l1xgIMZ{I#89%YXg;v~&JhD)P}y8zze4O`*y$Kps%o>#{@hYFmdU`g|73
z>@0ZbR^m&t(Q>CVAmNh@rif72)gCA2ZXVFG`#u++YR&&31ng&r#}U!=CZ^o$^=o#5
z?<a;75@~+6MrT3CLGq}zWEs=d)|M{ib9Rhus}hQoPX{OEToaZ35G0=V-hDc(4H(=!
z+{KgKm+?4+-DV<d(q>o;fAlP!z80itJtBc7#Wyj5rxp&ZTs%blFh`0XY<G0F3TC6Y
z;l6Je3YDPpfm1<g`{_N6rJzeJ3<F>{B$z&{o|ahA8+9Z*zPI_IZ&=oMGou(0C|L5s
z*?lit27!^jq_}r-6VO-F)eTRkXT%PMTuN}yX_Q7Y^(Bp_pOLzyjOnn4eE`0YmqrM9
zB<?F0<Zzm2%m%@PVJo}RgA>bflKOd4M$tIuQ(mg^?Ri$SuVg8Ky&M@{Y|5l*l#Ae0
zMoTDXBIB4JDl1KW<+(EkZ*3c=Wt5vJto8i6=vBr-T+|O?1;2u&8^KWJL)Xk8ov3Sm
z!sR}uYB)^oG0}PBwUO|7H_pFdO?|pv6de?YlJbN*24$+56&o`giwvydX)Yzx#9`o$
zV9d1Rdi}~Dl~B{)ocwLCnx~!QQt95LLP*P`>Fch?hM?ed)Aj11)6!HzEjVB0Ct^Se
z_`qVqzDk>mkOt2cDiOuvSo(7uL*CGRZR+z<Z7B+BZxmpENIp&<Qg56LjA@wh)22Mr
zsL07&s(hM)Z+)Axe53%7H^MeCq1Gr+J@;R=HDD(UtMZ8=3LADr>4L-5A3?=nNPl>{
zTGNz!SRuPBdNh5OHuc{KS@cfwNX4b5e3;v_V<aZt6$7@5C{g@ffX;vyt6P86;hME;
zh7Jv&2j9(C(!q7OQ05^g^I*lsX-K!ysCYWAg#<d=c!&w}D73bi*79h1TPWXUXzOP_
z+a@D?w!XKI%5D`#%Wz?x-&<($4!FYvdUmpm%-neC^@!$cBXQ!~r&6r52$@@qmlkeL
zNz27~9-PLO9?{aBPcNphjzsu~0S%!OpdtK>%sw#hG|~BM-h<>QL*}-{h7j0{4O@o4
zN1mBkG@@E_Qq@Cv@zg^M#_*&cp)qmkS;q?)CY1<6vZ_?j%shbw^1EP1G^DFcXiO7L
z>P$S0EsQ)f-o#}C>5G0vxvX>>?%QisAkksH4~&iYbVvsj1yEFL8u{eswa@LB3-AJb
zN@IZ5rz>J-n@OTLA=y)=vg~)pLBDMxiRpwsQKwE;FE1>pxFjM#l^0#z6|-vp?@1@o
zuf>_}`6rzr3?GFsBEo->udU`F`<WlaA#dd~v#TqN{)6~5(AmBRhkI;}vL#-u0F$Ch
zT_85`nGD|w+IX6-<0P4jXEbXVDwxiWWMSuvf2il*t$wrK4TM&5nyAxUF)5Q1%B(~k
zZYfJf+7$*A;`=%QO%tT*=>Mx2Qq#aC4Xz1Db22s>S%>ix|1f9*9-5HA;hPaGX$$wJ
zyE*Ix!F+iTIk^mWbp7XScF_vII;^sM`SLqWq<rwQEGgl;L?Cq4`pnuS0!_Z+fdmVk
z*D7hQx4i*A|F#uJhj_m?HI!<tCn?L|%N?!l-O}CYc;;Y#2ft*ocY66;z^X`oC#p<S
z7g@|p^7XPD(U1L0nxTi8WMH?8-7==lv_mcP7;datC{p%+UIH~t2Yvdu78m1!+EvZ;
zU7k<=lX^Bt2h0(aAWyo{U3E0{&PuANO}sLo)kEo$VS}yIY5F}}d%J|ymGBuONT@K6
z<CYO^4a^0eaNL^&^Gw`@0RfJkOjqXO+<UUqN%oEVl}MNQ_(?GkXP+PhQGQI`OSC<E
z+S9Xy0BR>FLa^#qSGU<ks)-eZ!ii)jt?FoGBGto$gTK2$-c>*gOMG9CpXGW3L6z)k
z+%2Ih4UXD~G;7Cmp}Ul(mq=FI-xnXp^x@zd6QEmuOJXM#9YEJ%{H(1OKzz8gnn7)F
zG(k*yBhA~~`A`dQY_Z>^e~CkoBtCd({_EiBC8W3{$eH}p0VaR_i@{UvXsC@MNOgp|
zZ|J<DTr%pHnLeoUf|PCY7!!3UflR&2@TD`9QvXY~AQbY<-#ie&X|^-0%H@*+QUgK2
zj3_-06!SbIt&6k-Byr5x9pfw24cPqIRKIul^8*h&2qO+KBsMuq78mh=7~|w+i}AW*
z67d3tEwzITI^~t%*r780n$trn?lz5j&Z?OTS@E%V)A%sy?*@&3D6sMZW4B#jO?oal
zPFMd4=Y}^fG+q>Ob8i^P-|&4D&FMeczUtMAtt_cW>R$dyyB(x6=+#q;p3LojPl<pu
zSa3IK)8y?X!RjaEG$zq`H%B;o*cm=6Ziex;3C79E8I`2Be6t3QTK(CVj#X3YKiKe1
z>Eh+AbA1}^8VCHW<LNPRyG|+y<x;FEqb@>`6xr{M`&is>y)0#A>CX<>1%<ApEF7?*
zDJ(azhAed00->_vVOfY(%!#6gD^+D$WKg!=1MumLlPY`t&LV|F4f;&;?zcJ+1oYL#
z<h!m-4=x*BMX>o^5XZgV^dJIenfvk{qPm%X{KfmC8i2$0Mg6hSL;fV>Y_wW{xdj9t
zdy`LM(Pu8#-}dzXxeescNLn2HHU(@I<Mr(g%7iKM+Wa21YL6+(WgEqW16ZQ_ru>xb
z<XbN$e1P&_YSs(?5}8xpKo{<DwN7wygxBJl(Fv@(kx&>2;(igGkhiyn3whg$KKNDj
z+eNl>Yl)j0-@V!}2^NR3elKCgZVD@1^3x_!E5DVkou7>Yj?dXEQg2h{b<e%D(4;D8
z_hm6A;j<N~lyRaCs)rD+7tKmVooww-%~4UMqL3%$IFO2?rEI%b3we%Do4Q}}g@lGy
zNyH`7a0@@Y80g2E#h&PYD;X0?b6*#wxgoMyYyDHIt^HmQ2zK$i8OlK1x4d5&TwC@M
z;aKuRjkc(99xs5y+$9vxtquK)a2fEq%T?0no1r$5$_Fl9{+`PBts66&B`XA>^Pp0_
znq6zI;o^-pGq+YB7FS!dw~G_62!#WT#ebMw^!3un#o=FPg2vHR5{G{hJVFWvhjT>4
zl}<7vZkk_PMDKr{k>9$0?xWHPEOznsO^TO<O0_!(Ao*rdwJmtQe;j3VnPKR`E@WZ=
z!u0vt9ugql@}nxl?5k{{AyIr{U=mTUo^lnr@+K`WMv-#@L{yUd;9^t_mv3qi2?%mf
zXZjk~&dL`WQ&Xe(eqQ8P6C1QV0;o`p-&Pk~9-UyuNVECoZ6S((110YAyGQ&iC;lsO
z(-VnUDEWmvwq50QFbQOS8p7Gf;3#1EMQ8NwBQuA(8amInjRh~GT)Chw!?#7Xdv+cO
z27i}eumN>oR9*%y<A@1=#R1z1i#80%;*56l;=1O=d1O%t)TCKpW5rqM+olNH{rfP=
zar3iPM_g#m(FOY_0PG|p-k$2Dov>eIYXGIYHR${$GvPo!WjyOkqC54#(^)hm<}k|z
zG8kYX<@#_NpeA(}C&B}}0%(Juk1v{?$Ibc5|DqNuy|U2xO3k~scR>cUy3lrlX85Oe
z@=m{7fa|h5iIy7l)W9%%eYacqkr*^Bt^YZ20@Kb<0257W0p382ENN>Kn8ZaWOxa+}
zU&2b~;O)z!A=M4C?!gaykogkzH};2Ud4__)xf!Ji&1uIwfCwm-xlk40&MwCIfhGN8
zfO+c2m|+OPcXt*BOIMDp2vBWckjv7=%;nXU7{~eNJjj#R3P9+=>IAHiGeP{&75G#G
zG(`upayxm)-z}z@dPm)JxWFO&^i@M~Z$K}8rb?b2!feM=N>z|fhX*;d!addsw?HT^
zjiwAoPYP6CV4`U*^b`GktM_B*?t=}=cdIc+xdG-(C%eSM=+MlhAF)h-s?Dwoomm1w
zrCZN>c*4y6POwUcGR%^+k$FJ{JaT6fAvCDw?nn(_XrXOi-}XEA<E^KV6vYI^&Hg*;
z4CAJy$TAZvlbR}2=+ZO53^?Ln=3W|Al#@f$y5`38+k$?u(q9uA{{<^A^|$Z5V-%Q&
z&Y`yzG?3ZbOc?a<kFFUV`Avy7DU!xQcD?W4(9ZM0mD6Z@=Ob>gzP9>{VbCCl*nn{L
zHiM-nQ=oi<qp}fEe#d>}-bJwcNt)oJgFSm+sHFal?H~VR)Ft)qEj$#G+l~*QE(|gf
zs7#EpyO$Y_&gg#CZN-cIDEFD+pLyahH<ls*fC~JDvFkG9xVZ*(eqTrmZi2#);+x!a
z@&H$@ptu5nT@a)RIw4gp1=y&rG2HlXl%r*X)X^%}EF`k#tmlU^?t7_4B7;B_Iv$)}
zCJS8#=A;#%(84{;v?675FY`Z|EZ7<nB?Yoe4X%xHyR9SrBWXngWiTT6@B7JuL8{B`
zBemE*yf4-D{@+P-Sq|d?&H?gEnrD^57jgHS%R`~ojpx9{qsg<1WPknHv#42bGcKpd
zdU={V&*R*X=yBC{^%+1!Z(kq@Si?U2rIdO*36zTlzn`V-)`w<=SajGRn)zgPma&A|
zM#!z9UhecdgU^HW%qd`Xp<DRTSRBins~^huKV9d3WD1vke2KR<Ppx_UYZux~F@4$c
zg>h^@DuAU7<PjNA@ZWee!m>ugO~%)+-cdisW-tTqQ9Gs1&i~}(kgd&am2q(st%5^i
zFGBVcRxb86iGjc=yRFYpgB%gzleB7AY4mE>zJmeJ&laez!J>l9wTIcGn*|CCQHmmO
zTq_DY#=0V;3h1=D-pD2bJz%OGF=OMqJYL$=8#v7`46DyaHdjoa>^bmF)%!;%XG=7C
zrb^U*DPei-l8+&gobU((w$gh09X;t_RG?VM?I(O5LYqq{wR?x_=Zo1x=^Dvp&KXq2
z)|-ne{-~)5-dorUS`QO9<LDIzluQ|%-DpBx*W%CJ&lgUso!wn#U3cb+kbaxH4OQXN
zt9G~hS&WGLZjOo+zNuG>3mNq;XMvNZjqiiEj}3(QKYd4e)G+9W-pn9%=;^#E3hPw!
zNqV7I6lTZ;EIPO_vAvC5IFO^V>~Eui!n|9nf4_-&st)`Vv%h)2fhMNajFCA`8A(i>
zRR9*|V|-<4?1^LG2M|n129+?}#_*QQ30Z~?zvHr<*5lI7!1yg~zTImOWkU}-kh590
z0UU?nUVTs37g_U23P(7-vx(OW+uR53V-ZnT1CK1QYNkOxG*4%uL7z~NCpO-i6IP*(
z#1Zcrr}|o^&A#QIu{%$&v)wIU13P5C`~Jplqucm%OXtafx={TFF#uv9W-_?BDB<|t
zX?sPos3?XPFZ~C%TTgWLIg!Y`v4wRf5frxl5s^47$3&h17g=-^@9Xn+viSDcRJaN_
ze{T3Bf^?FJy4U_1Y&P$sP_wN3dNbu%zg>L|Gm9iNL1*W1IEw=#*i+M=kHiS>49xm2
zH~?>GM_NTU7dt0aPxnYe8n?GnUeE>c)NxL_a|iXCn>D_<d%ij6m4u0|SooseF-6IS
z<ZiLias~TR96*Q!*b8zv2<v@t>A64KSfaaqFFgtQ@!H>0+#e6uBADunA7|dcb*}Pz
zH&^zUYL``-r^n1$?~8h4hgOP&qn`R$EtuW7ul~EsZ5fGroFiZ{0N7<MF+pw7BxF5{
zg8gk221WlHX=QBCJ-^EF6N4uROPc^cbWNWUD_)FGOw3JdIqX4?M%gsuE7zoq@h7+w
zyfNwRgg%v*H1AOv5yOAa;|OE;RFh!?Jq>8?HB-&_KJ$W#JhSvG8+ChwjzuTPqjist
zIjDSVHxg}gh$Qa^`>QfG2v>BywBs0J$CE>pVFm`+Hjk9ZZe9rLl8kY+yf_v$yOn5_
zsA2Vo`B%!&f3QP44IU(R?T{ITC1|#MCrG^dOT{+th4nzu5tkHLZ82TFs9Ew?6$qBP
zrQ5uCM<B{6Vds%e41%SVUI(#(Tk$4|KL&ji-m9OT5`{I(u_rtEtt+JMU0vN)zh7CH
zBEVN@K!!}}E93h)Uf>ofP%1?8JX*`KuhItidl)|nD6PFMNKAO*LiO+j5RG9j$H@;a
zh<@+QBGuWQULhV$SF#bc_zVVGeR8F=V%Paymhf7IWBLmf@P)~qY!~QnH}41q3Ge#6
z!=51OTs?eA>~YRW#1H(+P7Yn&Q2^hg)7wjawioZS5(A4qs&O>7<J&Rt%%HI0(EWZO
z@(6dH9_#!z^U5S>Q>cy5O;uoC)mdiP1WRhtVvgsrK3}}=0ApQ~Zd(vyi(>Q%vA{oo
zI&l*ZA*#P{rc3^)<$2RCGsq8jus}c+H-2J7r)!SRt3$VMc!&S<R`~YacG}k4M1tqf
zT9;OA4u(_O5E!eQ#lu$cx||*;UaKW2^iC}ya0jVWx*vp@(}@x?g-5mP7{M{Zlt?hX
zva!;|IiV3NIxUHwF@em+aAzRcFw_2=d;lwr<M4e{LZDp_5dL_C;heqgqo!T`;GQtV
z53L>T^yMP-mY>|J;VfXki1wwTt&4#vmGDF=Fg}t-H&+qNP}kOi)RJyjT#!BhTJ-9_
zj8Cf`!gDRw&T*Tyqscr~?~$iV=d1y-ij?c4qhYyANV-O}+33--*Q}rbsOtKmjq*4y
z{=)z7ffE$=iMgt4+ar@%?4$IWJ5GGZ((@hf3F0pz-}=AB9CpDH6iB5n#dq;C<#PM6
z-7mO>j#;9a#JbH=x7Rq-HJ2wuK#KPHqP|TAWf%SO>{&j1VlECxO>|itgsslm#k`q6
z=!z!gaLfto1fa-lPrX>CVuj1)=XtK_-cBd>cd&IDbl$K6`7>eDv@Wa5U+h2-znvE&
zB93mMcYBv-x@-#zx3|~*-E}#t6dE@e+a0k=TU$@W9r_GC^exqWRKe$>pLt}QWmv~e
z179<|8@K3b`_<&JeMa1A*06^0sImtxAExFo{45VVoN6J;&o_$%VLR&IM<D1d?h0d8
zlSQ;(1HCF%cFxZ$Jo?8NVp18#?rzPn;2cP09<d(3=y-R89gHk^93+l!o3H5|ecmSA
zXsiHB*K%>Hc4xk@N?)LIDZhZ|j<}~`io}oa9afBpWN+h^AJAX)8OTXdFM)|lw*QhH
z2M;&7P)CrCpZ%#_TyubaBpMtSVURu&pm{M9lD}_nkD6n4By1V9r+C(wpY<m0IFaB#
z;FR>6F)cS{+E^HkVZKSDr(>6mwfG@Wkq0e~55Pew-MB$_%Nh7&K&~n^yt;u|fjKJQ
zj6i{#7pb?%Sp*hZ>34T5qAJm()gz&ONpC9D5$&Y1KQdyW_>mx}8D_juwsX2+l-qIA
zpc&WGso02|VixvH3y4laO(C?<YI>@8Bw6`@%np?p4@`?hTR2`^c_n`NCLpt1w^Y7B
z-3njZ9qcVjO@|3!q>pJXM@woR^M6|!FEXyPyFtL!!=JFJRIeZN<vV@(?Vn#J+oVh2
zyc^}?z-DhGP@V`VY)TLB-PX$4vyiwi>VJ=S`jMv(#E4t<Uvs*NuL2}>ARhdX*Ngzq
zB_*ZFCxT%dU4gt+Am7`=kb6%+mhckU)qJ*gQCOu6)FF5CUy->e+ArxZOe-HaV4ptP
zt?hr=7TP|83sv8XK>Tq-w37>TOQAY<9+6~|a8QJ~ov(?kt>euZi3`(eR)vwTDltHQ
z+Rttst~$_lve7{=_cY-3_s%<b@RXE$NNx#yFIpX<Re$3pCb!0vaJ)Dgki;NAH|bU7
z9~m(;WbK%*6hkmKtfjq6dKgp!8-Le4&4YOsr+C=gmsvov5DGja9XAa|a($n?^6$5;
zbx7WO9k$f<+*wa>o=0WN%RvXdU35kfWOdDLEXt1$s|teCO0KpE;hWKJuK(9_75i$P
zT1^(DG^+1<C%>ruwQTnRn&3tL^~S*6*T=w{L`dNknhHo_%Gyx|Z?26{ngb222UJai
zs|k67Y|`T3)7_n=79x;6jpN~9*!ZRTJd>3-Ei~}zOD;1dzLl$0maHeJIRzU>n+6ZF
zN~)AIYpvTFqc|Z<a}igS@Gwwa#%^v%4)ShwO14u*4KG6z`;w4d-A|_Q#r%j;cqbn*
z#QmFP63GU&?;o2cA?-2@e%35$fZ)HkSujjj?jtb;s({#F+Y#o#$M#vbt48emuLpd&
z6qQG(H9m+UuC2B+HRb_kk1Os#@(d(mH6BnTp6dL$FGndp8t+5guI}tgF;(Mo-n9kw
zQ5RRiLVN$^(HyY-gq~@C7w#wCu5@~Ah{)z@douPdQLlyAQ>T9=A*Xt!te&3m^%P&<
z30sX{FR8N;pGatH=o*aewr6HB%q_R+H^}*zhSXm0NG2lO9yXS25a+L6MlpdR`zWeh
z`@W34e?d+p!tLurmfZ#?)$w#!cg{^`C$3rKh8hJ(JTBVuBuS7JQ{^vrQo0*~n2L0@
zaNF$Nf4pD}KrE!m=Mb}M>AV}*9W;0qjhrvdlNBxl7o!fnBS1=LeQd@~2&iR{yivmW
zxANi_r@QvLi}>e_ME#a4fT<8H6Bus!uJpc7C3f$s!d=dO(*u}$F+el+znYXg2FvjO
z>}ht{0UZl_8$^%h-)VppDZS#DtxkO%Ipb^R<h9_pz0-09_$#d-u@CU`=&A1uqgk29
zXt;N`g!967)VxZvIUqlX=LmN^Rp=bQf}()_3UYb&dFQ5^Hx=(l<3T#>=s*@h1k)@W
zgf~ZG(~ms&f59bmIxl>>-uNg_NdE~Ac|wA@Ci|`u@%N?W`4N@Y6x|g+07AV#BLpW~
zp6UI2u3!*a@pl`;h&2G><&j6hhzp-+)m#eshc!!e3FLk)n?*8ow2n8d>BEn~JW4V2
zF4qL?qL3sH`^5&<w}O?CteYy`AaY(}PNbI^2BYSZtGxynS*}L=zZN%d5`8W4x*aB~
zmpXUuvqf<0c;PO^&mZLhw=pK%mp%x2v%dtDR3>BsqPgt+!CwA5L7kgUC$_<hW<7lh
zhcW}Dvuj8OnU;jG6#t<G|6ec_*2=}2!Pgr!sB8Hr9H=!Gu2xuq#^?swIa1V@OSzG>
z+;x$7B}{!NK@!*>85JJpow6j!FH9rM%;}pdDZF_)J4sixGo#{aeQUneG6bnATY)~D
z6~eG`Z4jb$B2HwUToI@?!!2_#oD01+eSKN6dXOlQw^tlGt59n8LzdHcW#mfMefLC1
zuAgzE>kl&GNgrI-8FGi;-|E==jEHW~=(Ll2LPSBhP@BgGKT}V8yzqg*a$U!~Px)B(
zHkhQg{42S{Q`z)HjV4<|M7t_6JwwP=<D`RuW!Da2Bg27IOwhYG_K$HF7sj=>!x1L-
zoE<JNun|9ew4+gBzGt0dSVYZ8%gi|qRFkhIC$IRLrOsjZh`xay8UA@2#+)YnE{Cjp
zq=!HOZdtIs0+=d~2%Dv!@BPgkji?9*;twKb50#%o@tTI0_>WSI`K$Ph_3b%PgN-E^
zC|UN)hoMb%p#q?pAf(t+83|hQUyl*hxQ)LHk*EFUJpEcjwf9bLMw}1PuxEIP;8D+>
z1O6*ipk@TI*%%_G-<y9I-ed<OyU|?_4OyQAiDLpY0|cApOe}Z5eyQ=Grux}RDsbyc
z4+5NMKbw!59wX9tX$0@KA99z8e++oN%82?SNJsi0y3-jz{;Dg5L~f9QPM87S5<FLs
zeFu*fv^GV*s}Khvho0eOWzmJmuw?fLBdg<qdLgf^kZ?rvYOJ%P+jYFedvEf`KxF7=
zr?t;<PeWo`vB?d)BLa;+1a~>5YeQj$sLO6?rBr1`3{UZ9*(z3#6QnYd)t>1r)gKG8
zE^<K%{gN`}>D&@sxDiaQeQzbBtwwfU`v*+d2?uXF)OPA{WF<5tvL)qdu%*i6(&`!J
zo*-UbPyuVed_VCo@W;YICmQIJQ0cmo0yz_m=(=?|)uiS)K*VrKC?Wt_Ss%Hd2Sjyj
zw%$J6vMjlG?Y9vvYXAbAM4u(g{r$E~-2Cv*1&V<II5`EdF|YvKq28OdajeoGwkWiL
zf6dsg`A39f#EaT8j-W6V_3CHzrW;9<kWr{HR2K#W86Y$=2Y8o$nhJiTzZ=U=!Vm3k
zzZ}?1e90iqk$I$i7j=7+uVNn)58l)(?gHIW-8ASG@&x?RzITc;rH-Pg%5?c|^fmuT
zq0PpKy=f;N==D+l{o=yjj2JjZ{hll4Kn=6d>bmm;;>qgruw<>}BI09<3e{Q|*q6Bl
z?qTI9=$Tue0`%$kKaiHvCrTG57RZPr6Q59USF^g05U>!PxH5ets?k=BYmRz?_06-9
zBT}Kys(*>bOck&0M*$ZGycLW}ruKtD`tYbV4`^U&=(}0C%?z6WSTh`7P>MC9TOl)!
zYU5z@A0lE@iQFWZ)G~Tv^6uHV!I<O}i?pmnAV7o~)UF@2nNj;yk3Q<}R0|_WcLU~!
zQTC^Y_He9UKD%g5MY=z{tQInd=wR5418{Ca6kw0;o>PK4yKa93><f9IgCM+2m+~Y0
ze*3Y)G?m7el$lQYUQG<F%PsMFgPg~`He0Mb!l;{c97~^M`4cH<rF@t<q%BIv0Z6cQ
z`Ti*6443ZuOV>|$Q@`_TmF^y!O}956yOlE$d7DWv47k|PL7)UlF^r%J70=%mwCnUS
ze(2fD=WIK;vYgKh?cR3(4hWv=lSR~Ekr~dPD1#z1BR#+^1Ta}F1UD9oy_~pBXUm;C
z)PvK@Uh4@wld}}?2n0mFNKaT(c`5tf{GxdaOOSZRsU;glz_UUt8dslC{AFZNftg7p
zlgHvdomW2|wbs{dx)F9@q}#RoB+J~At*d9{diz#w=A@*?Y17Z7^8)G7deVP2w|Z?L
zO75w9;&l}*m0>loE(@YBrTK($+{f7FjDlk1xb#pQUMXVkU;np+o`e;7BGeM72K5_w
z`eRQ{`d4m+KClDm`3mho{c@${htjjrsi&i(n<u&2CVw2A#aqh89)5Cf<{7PHL$`@}
zo=Ay@h?@YY+*^hc7R<fpbC(MQW{hv)9{Qp2DKE&=pcg8^0k8$K)PP@6N<4(wyc?%s
z7Q+HY*V?(2WQiXPWK}iUpW<!G9lH@T5!mAg!cAb{202BdJ0A@X!9cSjO&Z*zE(?r5
z{->>L$qvDjpLFiv*d>WutTx?2NDcWw?6}u+g*pSlTma}!>sS2@B;3L1RZIR!^*_Jh
zj(_bQiX6lVvePiT8^cKwQ!Gfee5&|s_kfjZ??G#=){CtRj}MtGnlYePo1+i0g6y&5
zP~prKHc=PPDzs%gK*C$VFo0o?nrXB4z>St;Sj3sBS>IVLw#YC_D<4IXLuyEy+jfMI
zY`WF?PrSgGj<52sp*gmzse$3!SH-!xk5kq#NPW%Hl1&Q4yg(yC@7&H}AWlT@#t{o2
z5yT~M-|ee&7vy=OIrp)95)ZVL_n6nF+7>yqSUzam%k@0(JOn;b%a_lX(9UEL(lS!y
z_O2+2f$<g<w}U0y&{w)!P75v6(qI0j_seg051K;k4?-g;-qpvHQ3EV$joLpnh9^s(
zNiD8}5GNP^rh#o}(}TXmT57ybNp19%9IZSe_O{qWdb>4?-L#4~&@e#*Aq<KL<hwdp
zSzI(S2UO1Nw({i<N6@+Wq1Lqrh_Uz^oW((p*aYptWb@VN$afKTi2a9T=@+|i>ThA9
z4oHry?WP`FC<1l8@TVt!&RgL?!Y1Le5O>t&h>-kvF1<328E=B}vtsQv*u;+HlI}d4
z+a-w4$NCPRCR_D4qWjtE69EC?mckd=XR3((d)6@SYDpRGGtUd3a)zyOy!c~odIe!_
zw#%FS{97^!#u8k{nr`80;>)ub4u5+LB)t*h;O1w5`0ZT$XbF1w_<S$`Bq<%@E?S%G
zX|f4=P8InVKxR*F4Jl~z<6&&}hoFb?xmdH`&q6=5t64a`d@W}~^w4Jiy+AuZ9MeZ&
ztKX&789TcCYNf8(QV(+tj+_M5FI%GO)c2u7TR7`5^YAJ2V|T+bx;%jK=)XZS2attg
zJW4`V0~*d`ZDB1j&_Vv4p`ZA9W%*jFp)hD}?(K>r*KF5zV9Q>{4i+W;HRV<@pjTtN
zZL<`zk6fnv(Fe{7E2evDK0>p;oIE7^3%o^5x5qXI303H07pwd^yJ<stI2#rf*v3Ii
z&uisD3I8)(yfRwEXDrXsl<~9WY^F|K5z`2XLF8U8y@l{-r$0t4)U|4iupHC+In5e@
z2Qk%O7#~|HL2ih)bIIxnZQu`I6YiPeF>GK9dfZRC-gf}sf9$rb^=Dht?`e!LcE|Sv
z<4xu-@S&R3tPtT*yZ-)Z#GCH{tCqJyf!n?h=l?lmvINOymQ_x=w|B&W4kC||NY7B{
zcQJ7s6a7b+_&5vR{<}v9IftiddyOvIguwFwW7p)jr&l;32qL#jqE}2HMwJ4@G^0=F
zO)iGbN(Xm2hK5VJ`8UAM3Eu3!qnuZ{1_gt)ga#FKg&+;0D8NxJ5ZAm_OI5Q@LKNz}
z5lq~ly{*`;24eI0-&_8bH_!XyR6p+Q3o}gr(LuxoROGyinapO`WXa^?(Lo<mi9-}O
zSmQZ5uKeW2PM){I?VOZ(cmGG!dk1pazW@I(BP)dLJ+o&NviA-d5i(08v#e~gS0X7p
zS=nS{Q$%LALPqx9S>NO8et&+y|L$A7#&uoic^>2Wd>r>PEB5RSP+Uo=`oG!|ny<HM
zh5muEXejxxpzHH$%Z5#xd*?m42TMZ*f7~z5>t8dMkKZB9{ZxHNA5K*<xl*IQ%1#-b
zeIve0x^DDk%*wCq9y41%h-yqF^IU(SF%>-C9g~;nFM9K7L2cW~DRbo$lS|+Mdvxc@
zB%SfRk)xsM2FRA_KYuVs+glw(fgDEnnD$`_2T2}<u!~sr7N$RV84FB2-*bT1p<`-T
znq_7D0p^rxQ1fT|t?en}`ua*zhZMcx6knEGYAHdjHU`>;ojzvbAJiD@FnQ6gQnTOM
z9QEUte#LrLCoFUFXSnpRT@r~b{_Jzx)qlvlrlM(x`6apYGJO>fb`aeGZ(Ofnp!Xwe
z==>SwaAAbl?8m4d`7Gd~n)k|5cZ&e~a@$90Vc!3-NvAN|3SJ8_ZtQ^6@RbYW(N^vs
zLP&B~zj1=t=k6Ze^NU^&_zB*72*RCK2W}yL!;y`;K#PVH%kg(D(?xzKytL~)&g(O@
zkW7mjY}N!*e0XwXJ@0RUdmE4Jm<E=+(dvYUeb(6h(M-jspyQCW(V3y3e3Q^$_e3dA
zt;1JX_dU<bulz=XDs^?&1pbv(JDTxCUA$E^o|sv*-~$<YnTB8syz6hdd0ZBrM&1b~
z#pJ$xM}uY(w}{4YKy4eFk-b2N>`Dfh(}5PzqSD;9y#6E7VKYmQtBwEi6?5pofR~}|
zM7XU0oKfCUcbUH&iAAu@a<y{2B4<xbE;421JmTEod!Tq<QPBb)9Crh|vD!svJQ{y}
z_`&)mveRQKCO|X!`ojPJ<f$w41@6hFGt`xyy+1uoP9+_2dN+<PX&o4Vb=JJX%Kov_
z<4E)Dz|kVDnK!@FJUw~bH=@}1Oc8}Gew+Ekm=>5QK6I%!p-R^?m;-7*bD+fx29}_g
zn!614>XqD00ViN*%={wxAcjiL4E|+c38YY9b>SF1b?Kn0{PkYQ`T=L~ec3~l8YN#b
zced|E@(4v$VG?hWcXW(8B2slZ@Y19oMq4G1`hK(AfbXH;lRt#uY(Szf*5COCW51B4
zZ&f@8F@3M`s1oJl(*2pIYtlYfAbV*2p8lDlCe6`o3%;2^jmTF5yMHeIzZZmvL+s4Y
zEVg>|f=`cbo_>9Gwx<d&`Me23bZ3*L{<?DuH2;B*p`$~2*{T-vnv0;uU0kqh8quOb
zjTlq`L_DSPT%)3+V|^J<=eX_)TcJ`$3NuE|SXpzdHp4%o{6VtAMfbK>r0bi7(bwS^
zaZIw#;#aMl+Z<+U7|<^guoV)G(7QI0^uQWF@!GUTuUSf9ZyQE5E!ckop%nKadRdR>
zu=Zle=k(JNBedyUC3Jpt{9$l@w|;!f?EsAuB~OxGFi*!tSYV@=1#FYFH;G@GMyiNj
z<MRAE<v$mDe%1BbLHql7Iz2VnSEDEE;J{oN-ja!qaYgE$ZHx4?p)Ez1pH4$51F0;2
zh*=K?*<>?RjflThE3r=3$#i_KoyhbXeR1S$J1M6ee9^EjF(?#V3Q>aZgHp8)7uQRX
zqz@ArePq&pIL4hb1I|bEqa|)lu-xD4i8O<kjjtzZ6Z)GF)|e1N+cSn@-<x%t^yDQ-
zcPKo^D?kQ+D&c;!c*-Z0g_iK++c0th?geiXYYh^lQ9H-G_I3|lb4bI*be@@(huRu5
z`^f_i#6!qdxT2zh{(w&`ir{m}Y{~r-3+w4WArn(GE7n28K5wns(h6(5I&%Bb1>U}>
zl5pN{ro{~H_zm?U9K2{n=yL1+AiKDJW47&U7yb={2TspQ<<{*sprmuBh>O8<8>%Q!
zqsBumrR*r>EZv{!9#XE|oglunw5|V6)b~WfS{-`VnP_NB6qErwyna2$o@T1=jwPPN
zE1Wos(cU$lZ+LV|OhkSmMvFCRo~N%s&d!+Jn#<!lRJ4a{A6+6u|8|g~lB2jN7_P4I
zrNEERvO8e4vy(KS%|Za8Se`c`OP|Szv`l>?(ihj|<aEWMP$7S{VSwi>`ykTU#&YG_
za=KGnxlpl@0U`$a+VeAJc#_}p+s6Q|QKeo@Ivbnl<{8T9>CdoXe*atRafe1VM=0ve
z_7xCwbwykq^6HQ<z`}~4HAq$BL4?^+(MQJ{^1>p-%_<p7cR&$_qxjO&=ZXI0VKtOu
z0`)$edzRZjh}H0T-+1(+6F=ggW*JK}Nf(8u`10M3LNHz$kuy24Y%cx1r@S!T;K}d;
zd4Gc~04Ow1Px6`nXtWo@Ehv2D{^ejnI>YEJ;+Xzz8dE!z$gE;rq?4g8l0C<e6{u3(
zh|OXUQOwzh=%-&770ksbUaCJK3Xqa6R{Y^bR{Hp0X46Y%m)ZNaMGCoM7sU&*Nxqr<
zKK8A<RFIdWD-(#da42R3|0WL?7;&dWErd8*|LYj%8JT1fI>w))RSC7ZfZb0UT8m<D
z>jV1I<N(Q2kIwH+9*JXRW4DldJziaJDzq(#7X5gUjUP$_>iwQ!vpNRv!idJQ4rA?|
zd=K$KNX9W#5vt5)xj1ldtg??gH&6`<^N0Z#iQUiLne*<Sgj)8uM|)hT#$yyaD-mI_
zN$GP|MSrs3qp*+}bgk6f?eJVp0SZU}v=}m=?}f;*$j?U@;5eo@damE~+43Nme!41$
z4=KSyn*ieP_*3fQm!wsF`-bzL)=p8-Vqgri3g(Gv{meeU#l+vjMP|}z!ODMHE{y$H
zD}_fhJ@m3z)L83J7WteRP&Ii8QR&_9AS=HLE4LNv%^rC~_`vp+vcX!GDW5Qz^<Gy&
zXKQTy)nE9-+zaVyNYKu8gNk@S6(>@yIr3&p&2IY@5U4HSV?xk}h(e?2#tLL~ej%Mr
z#Ne2ujeoTkvyCJ@!&fP6YRvfex97nzoW7#=cW6p9)qc<R7}mITV?&bK0TxW1Y?n8i
zX}P*$6g+PwRTkH`=ZBT?-_@Xj$4)<ZU@30J(Dr14DTw-Mc+c5o?s86IVX(zBMi5Y_
z;`v6`#+V-T5GA609P^^*{*)F91ycywW@c@Jpp_EFhPNr>Vs~xc#ZzlIzCI6%b+{~(
za~&-Xc%&vT0ndW4UqEJvh632h5z~egEndy5B74~41qBao0B^{tx9)}%<*ixh--&v!
zQ$bxhOgeX85!|$Ue?RZ~FvQSX88Wfl!My3r>>Kg9aAc4_ldk_C1uK|P*Z5uVvCiu)
zAlz&i>B;0ZldE!4A#1?Csr32v)3+%&+8~pIo13MRx&x8h2eo5NM-W#sJXn_jGa4nz
z0r&=vgO?MtWuC_GWQm@V&t3AfdI+xM@K_~6QQzOVT4p<6XXrJz7WW$SmVSS9@AH8s
z&)$WnB+{H>)a;layGLy_0X#2g@^esPndi$?XNo*9h#_AN$V(&(5X(7{A;h933~Gb|
zrhH@y=$J#mrg$X5CrV5BZjBc)R*xjDu!(8)5m}yh&B4QcnWwu|gYRota7SLz&x=PY
zdgMykE^fl0BAp4;*94J3$8Xmw&TFk!@aT~4vL7m4k|YlN96Q|-?D?4~zH5eAwFN#@
z=_}lPw3Tn2I-S?!SmHrXHK|QAVguE|yhN&)*B@-4UE)AHEu=X&D*EiVAhNB4O&KK<
z{SgOW-2>g;C`op*5e~P42Z}AD?PQ?qV+LdPS?<Q%R;@DSNFk?h`UFjyO{&|@IHH$0
zj`b1sGK;r+|55-;`pYl<{KrfP0T0(9?UF8Qa_<i=UKGD(_YCq(!cz^Th=ESe1|KQQ
zsR%1X|BGtY;FAN3jBP<qc&d5?aHEhK255YoTI#OJSw#BeZv5`A4W%3_B7UH`6qAzp
z%k7UR(&&U9xT1Ab)-%`c$}5z~d*2z8jtL%2+~!t-_ioU3!XIm`*;PfYh$n^&z}-Nc
zk(4){UZzh!V*o=sx!Mu0i96QoCk?vKEzJic&RD$I8v@^$C}3S+iBvFavv8wJeXwzA
zDZpyA2tv;7&ZgM@oTt2}JMt7D4HKEB?qd4}!aner2|SV1#qm`)B8Ln}w&x>`S1l0y
z;`(oj&Z7-oesRr%rTq(eyr4|v`DJIX6SqW{;hcv!$SqtG_o9K4)}PaQxYk~5CmA0_
z1ENggjw$^o-dMn$!n&WJTk}Alj3J8Fd3l7}q;4M%Avk%-a7WSiqXR;zayBi!fv|OY
zvUd*F4>X_z%m+jzan4?vHNV4bGHaE?Sdd?+M8IyDbCPIP7Cy~tNg=9b)rh}Lj(K(<
zZxkI-lh1{IQiLLjCl+o4Z2q7*`aRO-zEyYlCsK_JfhqAW5J)KMImnnY-IZS^KDC|q
zUj+#tl+^=UXG_g{RC!GNsOeG+m79tQ2C{uj!Kp;PdKTVJ@~-wT{w>1kuI(c<8R#$n
zWtUVoa%OH&0GOeQ?XjNL8yw??aoTmXAqHrk2yp+UJ{T8i!M0LHN_`+P4NQ#|<MOU+
zZ5cri|3(!SqpsxPHfy*a)3d-Vtae4NVy{_F--EHS{SGO1UybN{m+0-WkF@kD-ptfV
z(EApFq7VUy6CoTKfc|Brg8@Ob-Rsc&aGAM7ro=9%-GN*0>gJ<=7^X2ZExS&A?P|4!
z`dlPetF4y>a{C1EZ(qWd@o6<niMel8k8q``dAXSS$Ix(9mp!(47-QW_R9s9{DTBI<
zbn>{OvZBuw!*Vpjz%0}+xKAfVEj9xx`{x?!EVv&$_GU2<6aX~l13Kmqh5b<BMu1(S
zS~kcaATUEj{16jdYBWpIjw;Plfb#3yQd6u`@9Hr{R$o|RmaLM1@dqD&8qZVwlFeK-
zMtHt4HZeOf@<Oxd^IlU$c)iI4%Q6t?Vm;EtK#VJp7-_-V9h{?q!>-D;zRkr1#n>&<
zFBaOfcY^WOpl?Wp_O%yB*u`G~52;SjT$H%-q`akS15-LgXCV8<;y3iMD7zaG#Y3$7
zqdYG8<$1_h_lv$RZGFlRajML`R5C5xRNTZ&rXa*IZ5d!(A!T)qA{%PVRC1ADyo{XD
zA@1pCTi~_E$NmikKD@`ocGjVkX>4FTg$xI!jk)^%A&$OyQzFmhXM=9S-*M!!qs&Qs
z6b^csU@y(YwhH^Z@jRuMIC!MlW`ToSOApiHy<f~%azP2!ZH@R|K{@JC%4J%`o<BS>
zj1Tc1ephP-Z|6&Ob3Td|gA_Edt?{m2RTSClpFtMl`ZbLp3|ZS-N*_tV%OQlBS=*TY
z&L8ViD0otMQIANJ5fmm|)6^eMA#ZxZkH1>WBA~=C@uBtD$d9jm&mo}-08FBP2I5S_
z47mGtuyY=#6ZtbN7L-UlFcg6uDjX7l{s1i)=kd1)_y^3)TwpoIv)}&S_4}iygAn{E
zIL(T;r*-TIn_Bd0P5=|gIE3;2JsQR2FrQiD!#hf{2PJ4J!AmX0zreT?rrAK^B#>3W
z22V(W2twLSX35={O?L1R&gDfwPi-$Q5m{P?{$atI>0ol`WQ2F{`hHOy6|C_g03C(T
zi?&KaiYbSD$bZ`r23{gg3M0Y*{fDo_lwIQdSnjX<dl%L92p8NP$glnT&40gxtWi?3
z$C+FK$d7|p{&6YcZ|PWhxB{YNC~3QR{pW?@mk>fFnAXt~jr)*tbtfezV(4{q7Vm)i
zrfgpG`JKao&Bc9iJSNx93y?dAsw*)e=Ow&oW4UJ&#n<mg*t+>Viw`f4ylbx~>*y2>
zwm8Cf8EhuqQE>T($V8ZxcgyxA5J=ZU)c2m9qD8^w!-lv!cGnS_DTz3me4-36UQeqw
zg#Y-qb-(64{zP97Y6XWM8mW5lnR$V03U@CKlK3V*@v!~RC4a7m{{0S=1t1EQIj;9X
zgU(te@aWn+A3I$9BUC;u=C=@)B`OeN%`L9Ar(Jc|7vESqRtCZ=XhJy!Zp<by;Wh-L
z?H|8@;>zm7>mNAG1COpi3kaH2---PS>bQFE`vv$)7s)4SYkXzhvu(Ytb?e_{_fi9m
zf@DYW*Zl6=(zg?ak_}B8*Ms<2E|iR4#6<x{_9lsL0^Hm{#>yb@zPnME(T;93aQ+N`
zmG%<)7q-%FoN(lRtNC}80j+ui{%^pu9S`&d83Vgg;LMF_d(9#x7#*dtL{RAv28(I8
z6pjD^4=6+oc=cO)oIK|Vj3@E<O%r8+LBS++Ej&tdX~Q=2o8rCy`B!9LF(RICF>_)7
z!B7n3<sV5mHLFz8)nh{K+1l;jHOG{FMG9Sf+W|i&0D6#i@1qB14U}IngOwz18$LWN
z2M#^lU;!1xbyOG-NknW~-(Ai`ybUkQ1ZC|X>WpsgGZ})gf53xK?A&)EIYD4#Y#BfW
zs&Y`Uay~zmTecw*4|=eR(q%&yoh;FpPEcwtNw{3n{ZXN|C-t9OltjXFKi&)vS-;_y
z%4L`uHa<8QGaw{EWJkBS+)8D1)s*7jr@=G#?_ouo0qYkOCNUH&^T{<E{9g$x?mPla
z&E89#U)!{#URN3atxO=|wm+WQ41tFzNRR7B*GczxlI+DAebCUnGb7Ke(&jaX!eW}{
zJZXg(Y$)kSaD3<dC$?vAn(I1kJ0yNsnm@yU2eO<)jI2I*hUuU#^f5<S!2L^E{T~VA
zaHt~1alS3x9Fp`32=;|X7#?L1XonoVg>nd(xUzbGeb%_vF-A36_9&gAK}9ZdpC<d+
zB&*w%e^^#$3+n5#w%PuFEk+1a53<U5A|Ls!hbnM=;h(GPS7)o7`^<Wkl4l-`6-Ta+
zIGl}A)H%qEC$pR<a+;j3^<pIT1BLn$(<>a~*8qo&?}*YKc=0U&_Ash9%zI~#il%bz
z979hCR_)yz-b4-7fsAUHP(;GaOk~Ynap!ATQ+Y5Ut!R5$c=;v|oI|mt5zLJ6l=bUF
zk*^;f$A*%^&#<^zb*Vn_&|?sWPH^j>w}o6|XA~mrdEkcg@1^t+Y0a>|BHs2xREjO|
zqbQHN4T>~vHHiSs1VCqIHmDYX0Y_FwyK)vAYxXu`mVco+iH;jClZBb7&jNR#x7xA*
zQ*McVQ%WI2@gn4il-fS7&u@12>fW(4FJ#t#DZ?z~uL8)G7Fqzt2WDclUJb!dpt(w9
zOkfh+naqljRwG#Bwn~O@ph0BiLmIphuA-#mjSzR)BwulKTn+5^g07Q1*n#8YUOidL
z=0MFrH_MH#fdt5md(Qv1OzpD<7s`FJ5|>L<uTld!><}e?akk3Z>1)u)W&a`~`SNeQ
z>LoaJv6+jkit94lW>F&4$-gnVl8)1}$24EcJ-|Ln6~rMlUmM}>g!gOt4mBmveR11_
zLWun_b%qep8jIh!l1|C=?_Z%fl{Dp~kS$u{XW1%6n!JquijO>x^@euegKw0>^8(vy
z=*|=w5rB)Z=;$;_d0lEH7N<#j5WaE3p<48t{u>65C1dk1=bgSOzzpTyVe#$kja4C{
zW_DNSsr&)t3=2eg|GU8j9?J=#&QhKSJW|26+%btusKc}ENY0nLc7cF>s-!K&aXf8X
z3St(<@<b6(6kPh4&T)u(*~qR&|3)JQch?;;<9*>cPUfcxB7ttDaLIk#r}iGr6(3nZ
z1dSLrC~=P<G3q(~40Jd|uc>&AQGgYTAuwxu@FwM9Iu_u-_6v8(5e2NRq^%&L8-<{5
zd5&3_5#@R*p#KvS@*v22qi~C4!1FZEf6w`}HDNaGVqQBp<rYF&;sa!;@fY1oJ1%!q
z-Zq0^HRf4To?+WuMWwY7Qtwc&e(nTjd*HR`uHDd7`zEtU%<bg7sUe(yFF;Bt%jpN(
z4R=x}i&QL^f67)wNI(~VH;^geCZenR;_Co1XXvI9_)-pHK^xvy%UB2>0Y%x7sapZL
z18UXbeuTf8i><2T<2kuLEvjqvD1e2+`B_Ja?dV5ppU1&}U-G_wkWFTeL5deVm6?Z$
zZr7vmEiR2oA-1eA()3hp%orgR5dHPQFz96{(Aj=Xfz&L11rp<U*3e&^R7r!8Fml}p
z`gBYCZFVLNCn-{*at?Rr&Csu@%D^p;JXE8X$zp}%81ZfZo651wGy$}mE6G|Oe%@pc
zqBgOha&GDK0O^PbR^3G3EN%f;M)Lys?a`DsY)U20_D+DE{JVuyl_7})A0+<i4}%{+
zV<TF@#~DU^$ZuuUw~lPg%S<cG2(HxEmE%M0_=q_!o3*q;u%qr0i_>8{1CWUB@{E$I
zzo!&_#Bz{ccbfsyX0*FUeCdPDa8WB2P=l*%5wTFDtXyltDLIpmWK&4$UNGNiPxJ8j
z!w;5{0{u_6y4C1rCdPpv;Ueqaw-yrn+A|UJt7Ef2R3%U3H?}Y_rX0MM8nq%d<<1Q8
zm-ds}I=16}#8Q~tSIH>K7@9%AlKV*$vAIAuziRLr2z17po}t5aHRe&bha+FQcPboW
zgh6aoQc65@BusmKbO$p)+x}Wa01)16Z+YMXnkr(!3o0`;uBTV_*0c@T*LEjkO`Y9-
zkb)LJ^3<58d-in*cB#jl(O-e^K%eU=_~@BP+#ddNp8t=l6-anVd7*=nxu5YREJCxF
zUNhpAZtsm6pEdWPqK)D^&sM<Nru8`0;PF(Kv|&0=d;xdxhO8_S_&5B%5lVeshv)u;
zm2@RT9Vw(EwGt>rcS&|{ZiFkc@GI4A(CpX5$k%wygcj~=#5KTp{BHIvSV}Le89mM*
z8rE~2hDfLhdYl=*j<8A1Hi@b7S#%}7MVU$8kb*x#jIj&R6?dox5Ssb7?*U()H6&h`
zv;#h}k#ZiF`R0AB08#_gV5ky$9EX_Dcb6&7o`tZi86-X4%zX+c#upb!=qiFzVW`;H
zfPd|#)pJ9zMm3EUR|9@B>fC|my)Kncef#Lk=zoPxuR!6T0~CoO*EJ=LX>?V=i10^N
z;ug?4^0JbL*Lk>q+sU<x`;$Gr#4d{EgCp<Aq0$l-$a^a>HQ>RvVf_mKL5d5NsgC#4
zhvcg+mN_nbcZKXTrp5<hrN1SQ<{Po6|GZUmSF$?z3FAYNCA9Y(aCj@=TK42J5@o5h
zPyZzAXxt(v^;~SXDD@up0Cwcw;0N&d$zBu_MDC|hJgHZh-nNftBmJSQ$$`muDFB|Y
zJ<K;@N&dLeE#*AErXHmVx)D)JKWH^q4EpU)_})t9jb#NUvPeJ<8qLQN$?qT9e|{tF
z`UChwc$@@ueTx~1?7k~7{UPb<B3XTQpYy<;2G`^6miqmM>M~Fb;ja(B7ghE~H3K#o
zc6v<Ui|%F8h(b6IUzr9iSysJSA(I%^k&w-u@ZqF0XB8R%5D4Jh>q{LfR7jdiYao;h
zwde+<q8dXzTJ``V5!E^xy!ERYw-{wvMukk!31lB$v3J^e=^k#9Zb7L*|LdFinrZ&K
z%PAw=Z*n$XI|d=mUWKo$1EG$2@-K8zgXhw)%@$j6<w*JmPeArI;QEuxq;|KieIrwT
zcLmC0kF}pMjF%{en@;izknsNHSAZIu2uhyR-$g8%0(o*}lDaWeu8YfELSk01CjgzI
zSYgv@En@`h;fb#)^{bz#?<lat+g`U7L=UpcW1^%S=Wb4+;?MgB^2f||{n>}CfpByZ
zy4Qh->i=6|ESnd@b9#zY<6!~OlS$*Z*#?Ou1#Vfi<$#>+cWIUbWqvM}g^K)(cmF+X
zgDH^gXd^0cIJt=6<Wj5mHHU6IsEqC6HW6-+u^=%LWkUd(Z$XJOF3Tuv2Di<vQRH$S
z+0{#5FRWON3GeQCE|&yTI8BHYJ}BMH@-1>Pc4h#^9P-KOoABRbS>niJ{indZGEbs&
zAU(``+-^ad9`f#!>yiln^PmB{X3Jn@HoG3l;lke*rJHR(dQ|N<*u+Gx(uMMO6e(}!
zWiE*c8T}4|;;M%0-Gb{duYjX=<a>A5d2Zc8T1}Tju#k-#Wu_m%31gqhC)t@2)`$=P
zBsfTY<?cRMaH85rl+`h&4-7_i0;Sk^HL^2_*lyl-F>syr#=lfVf*ZxC+&`EZQJ5K0
zF1x>PxzSXRKx^Lk>3zaaNw8(#96Pjhc!X!g(rEDO&cBqUs1?W$`M;MqjhMfP>Lfez
zcVKj~A%g|7Fpt#<Wj+%5uT&}Wg<@tj?v#vPc4Md@SvJ0lJ@Tz(aArC--oNDOAuEZx
zpzZ7n5_NnnQeL?Q(M)1A6@-7G@a^^rit!GI?roF{nKv?Lqe&EcZGUgyEjgI#&-ywL
zgP9TUpPP8d7T+ac-F0nE((^U-%SHlnwJRCnWimx`MoDLDJaLeQL$*VP4vd%Z$4CaK
z)AZz|-cb4S?_Ce9i?kXOhqagE@6Hhq+E~pYwFrkKv$tGcF}8M!_q&;Is{h7#Bx}f&
z`6<(t-^8%5xXXyw8}VPRY0NN8m;UO()}|Wk$XBw;zJW}hZcFHAn)mc}50&mMyybq6
zdZpqLeN~t5zi}SPa-OJ)*O`@1^W!)~{fZ<*AaYu6Z}`ggp8E8JB)qDr2Lg7g$Scd=
z-4&ldEY;pX({slaz0Q4Tz@kNNr01R?LUG1Q=0Bm62C5`a;m2~Xk>r&}B8<Shc;&lf
z$F=uz*AncPg=_~{623jP8?1s&9@Shv9+cOzEI`%eYi5d;oRX%#`)1pgJXbyrTU%Nc
z)Bio`K4!+7@c1mNPFA$MLODyg65?#<sF03hd8f*j#_Fvlh15y+&+u9r=T?qdVuEsq
zDq7!As4?TWWCp0NZC`hBqgPPs4iu2l-FQ({juHm^!=45c)twHae}{T@uaZ5r3&8|T
zFG#K5<i#$YTlD;#?-CdD@;xp4<Bxc}s2|$1Usxe}P(WUw7{z)e;|at>a(!im2-Z5)
z=Y^r8JpRF%4}SQ-xc9X4JUJW+asgH^I8Z-cp~qd~y+DSo(0*`2@7Jrve3vGRu5Ac@
zS%qBUZs+wWnpEGI)kA{nwXuZP0pdOv%I7}8zS1S5;WFg2_Gq?6A5a?IT3mhqcHvi|
zJ91vAiXa&w4jct`KpH4cwy1Ay9b>2ZB>^WBPURb&haQLFO})}>0Ql*5KNYXKPDI4U
zjuI(FP9AvF2psAJ(EH(#LB+jGkxd$MK~iiGfIuAi@|Nk_-qvy+XtX-SZEnjuMQupq
zw9V`|Fu2iG?NmGr>$iJ(hpZ)VB<Vv4NgV3sw-h8cQ2NS>CSzIdg=PfpVHmQ*inlcQ
z0E(=zkuf6{qXPCbk76J84XU4?C7vGV`zj1_rT^}|D(yzou+vS-6m0M^L!<eLy-o4R
zwxBcKua43%CVS+1QS)8JKtcFyN5J*$MnnuFlm`ZH*>WNYyqI~W7B=DYNYG|MnEFV2
zbGNfL)ay_KY-7rRmLX18Av3E?9NPOxd)omdZB@O*bZ=ko-Y(~Wb>BYCb@dWQmzvGz
zL;aR1rQc|;RBS4vvO^G44N_ZR7FTslcV3LYhqWsu{i1!QEC$<u9yj~Q;NBJE$E7FY
zRjHlHeLi6zfb>Ea&|*R@(Ec||hd$w>Ee8eW@Uln6Yk#!<HPh~z)pbWAvqV3_X&H^@
z5wi6^L~ik|=H{9nt00Nr`V1RRtlE35k2cn~nH`FHhKn(GYEy;&O-*6zoxEzvQ5cZN
zg5vGbWSrqHz#Np4SqG|*;gZmfW6+pk;v2TF)A#t7&RoB?=Po@Se4_7h!HLI<%`@1|
zd$~^I6L$abH{*3z1@1#sVB!x%vnQt>rfY~ouRTD!7LTn>1AQy8zh~BWmqn0G2WAL(
zokew5pV|v-3X7T$SMEV~Tg2Aks(%b%0j`F#seDt#)clA2dDdJSLgEj(2+CZb+iBu^
zz>h=3RUOu@8|3u_jQ)^3tvWCc(p(5Up7Q%u`XHz58&Vnptdwrs2zq&g`^v^eWR&Xp
zPpJjVe^;(j{z7n|d6CFRpL5;vGOE%sH+UR?zg7=UtxDNGAsSY|;d>pLz~<GJeuBT#
z0N*&HCxdP4;8=yAKKR?l^l&takgzZRe(H;@tHO6-p;XGF3B{v^Vn}({VTT7nsd^OI
zy*?}WbH>*Xy_@l`v-J&0A1<33CyU(TpDV6zvhW@I$n5Eq`<r_&%PX31S5TqsOKV#X
zfYd3|C1)_>MUX^1kaj2T;(Nbdt3D^HEn5{uesa`Vo?Nwu^QTz!;J&_N+yR(CEc@RR
zbaoEK4hBP$#B$1Tb0qFsZp4MU>YbsI8}onr*fbC5?*Z{=1-H3$Y)vZ?6V4*JNHYhb
zj6Cv8!b(gNp-yX;QwhhL1`bUsUpqrKPn-}CskKn8mb5Y`9>ArK6CS{~eDNA2t6ov2
zAwa2(6O4FM(?0Yr3Vi`@8O<8?kEqX^VuOt)1Ma%PVU<57DH8r)Vn7*GV>DUT=Kqqh
zuoXzG+jDEF;xwc$-=&1}WuGg7=C|Ry3EW7e%bLq4&1OZ)GTz+;(ls@){Az50$reWJ
zs~Y5N&<DB9!;k4ti-LBHK83&96?2du07}!kkC_3J?f=rGubrU{?3?Eg@4tT%m&Byd
z=Ulk`RXWPdk{*VUAav^!tzRrwq;FAr)(+16>nJ`$NzS;6FJUZD&4hT7I`9xL^ZTVd
zEj|(O2?|76R>X*-zvK{&uqhn|lg*pVC%bpTVXwAo4&E0}kmd)Af&&)sS8(PmFDwK>
z399DJWZCMNgOHNo7N;K5)R5_U&vz0(J~nn6sdyxmFusO?wts+@0=9?o7CaOYrN(k<
zL<Ca3q%Ho5(<ktPdZ|$8VX>4vgrt>?Uq8Z)5MDi@08mdt4L=H|m~6W~vjFOjbrka-
zd2LcuMqp@#amR^Y)c1pSb&GeDGXE|VG5!`VDsUJdlXJ5;4F4o=ahOsKPf@D!i^rba
z8@wR4{am~&>3%Xry{=U^{LL@&PizMtwk+aOHcX-a7o-zK-gpN$f@yq`xWIrTXJk8_
zyRI~y$iqnC-|PQ)oMI!5E+Z4?MPI+uHH2aa%5RPRfE%qL(|sN}Ge@D8mMKS+p8P^;
zZgdt-wW<$Ko_gppBR<k>zsV0b^X=6FFWGDTiN_1?#{`elIRja(NTDHD_K4W2)ayNO
z<$rJTYtrTvZC=<-pEH-#KY$e_(v{nPx00%olA-+d$hYfy98e@YXuGMC--8e3M}5m-
z5rAPOIZDPBtUvs{m-v&oXfJTs21abaVhNi{!r<YQ!t9V?O$xL~mh(nY2{?Ysf=JsT
zDE<U3LWTD!vnTAJ81gD8y9gw66*G^YIHn4BOh|qg!|AsnzwqgLnP-z>nPN%qLmY%H
z^~&*>`%jT*lJW$b%QpE7uufXeo=S>S92fN*H5&3<BsrIbROLL%#LS@m$Y7**^LIS%
z9_0uJDz0<hCFz7ol)9>*CjS}=e6WzigRcz~VC9dPJXUsbO_eXsJo8_&IcSqU$BX{*
zO8U^^cinYFqK_14u&TPqzMV+8Y%meNt1S`f?)%#Xes5;UJHe8LUxSp-w12Z_scnDQ
z*eL;qV(uCjW88lm@hZ*~x5BuEEUK&NgebEd_3>`+CHyvUBrwgucnN(T%^Ug&aI*Eg
zuGzg`I3SEa;?{*}Jjgc<aqs<wYdSfk1W3ezEDJW#G2?>gjqUSpKlgoOZ^cgk9@I?l
zQ$JEDb6Q%%P<gKl#Gm+U*&$@e65Ur@*<W_VXssyRQDI6rPC(YcM7Xs7&gSqKwynhj
z{KLTm0Km7wBnnp4@I3Urk?L#je9;eONNY+h`t%=<xq^<bxo4OV(sk{YUsN_~<1MY<
zF@Y)#o(GmY97;+baG<;x7N*u7nqm7KCq@Y%`MmN;0v-GkvhZ*0pR|I?h|}@2u0b+^
zh7!?t!?WJqd-`lb`VU>}ZR`J*B@V_>nT$KoEYM{E&K(3nS^ecOMyrArkS7YPymrt#
z(T{|%5~}hz;b9%o*p+^!6zqDxluFubQ=3|Ueow~FV%`paB<3AQGo<He-yV5WwrgAQ
z{M`qIy33aWmc(3MY-6}AKGuzfRyaDBMUMF7mFf1aNB2}tiGyF%?k~~RMH__=xw%N?
zt9?4`?P?Wf@4sz>mbb^))#54pZ#i)DJVz;VQ9jLLF66w4dx905z^jj#WRhj#BQbTX
zk>ut;{<SVDrU1QPnF+Nzy4T5hn2!gImd4G*IkYw@mPm+iD#={q`B7yW*FqirSIp|E
z)zBzO$=@n_U1sJ-YMf9P&;4lw1$_WZ1{1a=G4Tl|M(N0;aldvPHa*|>q`)gvz(Waz
z^_cy4C%YhGK>4D!b{xXGLJfD^0{r`eOnP5ZY~6xUQBXb4TDgR&69qwDpaBLs3$u85
zln&#-$1T)I(eKQWZT=}N?u#$=it`5=NRT8KX>qq6)8DYMg^Qtk(b-k^l64<#cuD-X
zc|dMbfDTsUD`S}D|FX)k+j`X!_o28%MD`Y6rVR5VVQd;U$5*l$P9*HkT<C0R#2ng1
zp{KvbOFMqfZ1pDqnh5>*Z~yJiShBA=c_@&<0}#mOx!Fh|+<Fwk5b%B!DwitXTYA{}
zZ*vG;fDZ&BeY;`RBG`q}U8RY~fy6j)5-K9<yp2m0#HjW$_Ri^jIK&{+JrYw1I0`&b
z+a3W;X;n;Yl<HrTue#S7ZvVT1w>|_KqUHg0cg~QlCne#X-wgS2j%lZ&*9)Qs-@DP=
z?zRlVvbL9zt_~#CW(}8_tWHC{Z<C>ydbX*Bh||lEw#N9d{=18*ho66-Fj373h5=j*
zW(}5)pb*3Oy29g41=?7t>VK6>&UgV?TP7sahAJFn+N@R_sv26M_cNFo9!FvU+_VED
z2W3t9qTUa-LDK-V6C4zMlXipr{B}*AX%!DjapKaB!266IMrsorxwop2Il!6Z1o^KJ
zmjCCeO{Uisps!X2_0f>Ha__mc*!l$w+*`tTA&*6y8Hl`MM>n03WXUjNB@#PIfhZ`T
z!T{vZOO$_$;pM=MmkB*0P(MevOwKk|OQSLrgVbi6^fv1Ow7&JHLFhG4n?yzlL@Jrw
zr9VSaNMX3$dieU{O~;I}>n<0M{9CSx9$Zj{gD=fDM(DjnA7<u0a%*7!dRJ>VrpR3`
znaN9IXmH|TW0TDq9QJ&vn;AYIo<-Uw>U}BCWk-?0TI<Wr$w#Em1)7B~!!U;IVgD}>
z91jsn-o7s^OG*g_9od8%volm}ZYGN0%kJPOMfH~{5>)j%n=UKbP<o)T=QWGLbL)Pj
zPG~c$^)$&jmF}bVKpPdx>2~caa}MYqBG;^{;3sL3)K%l4nD0tZ&qKPn-daCJY7E`h
zH)vp;KsD2D^J2Gd^D-)~B8tLi?=M=chPkK!K7BdQMB(QvrHXA`Fj4KR&*mZJ&Ackv
zv9*zG_ocHo>i1;Mbk<OX*5g_Xgcsphe|(u8yHMgntG7HXMpMbjL!y=W>db{8iClcJ
z)RV0Chb$mj$Xi>+gV+Wk&%v;pG(W!hE+Vo7&{ALqE3wX;<*J0W4_a#H<3qI=u~$4=
zHcA%%kkC@k%%5{uY!yjuPC1+#%OGzlWoY!f^Is*vjX-QH6rWPUix>wuxKgo&gWEkc
z2CY!ScOI3I@JD>t8LBjq3g%+t_cFn9g0LxJOpL;%9Cv+vII|(RncMD`F=^b#tENBB
zPXxl0-WM>1O5Pe$oP03cU!&!zx94P_ZUP}3!inEBOm~`V<S+i^9mJDaW2?tSD)QOS
z$b$$Zdf;N#-<-bMU7mi)yfDQsb<UNDfW|+g?UlJs)1N27vNHnB?*z$M(PyMZ8&FVz
z)L!0#t={9)(Iphoj;&-@&De5LJ_6JP|Nb2u!7skU^<W|t$hz%{zcGy2v!ajry6~8e
z5_WpTat(9PWW{K3quYS5Hr~kh75{MvSPz<2nn#aBVC-ql^GWOzt*N%eTT<fz*ov)d
zxRM|fWo+<$2nUf))5+$H6QzRmBsy4v6Z;0BAdI{wW$D{_rqEn%_5OqdjcWpovs&-v
zYp-ah{=5r%CYKduv!;?J+_-AGx`6nm$Gq>^*DH>=7aj*SEfUC%%|(xuTjK-!%7hh+
zXuyxynv1*P;p;3|%#aq^d+}!b&K6E4Aw<^KqSsItZwsBgPWZtj^CUh@8ig><VLXqM
zMIV0s=Q5H6rDy(SIs+#5Nc>cBnrGJZc}-1m-_9eC?Oef9RjbuotlKPm1nHOPP|lE}
ze)_&Qa~T3pA2()3+Mi)#&>&pvoz5mM34*B+4{k37xZgCe{Ce~Rx>vj-4<CUen{WMX
z$ZF_f<^!ARi%#7@4oavZ2b7HBA?z_Y=_B?h5c;0;Zw}0#X3+p}E=X+J@IGx=N56K~
zSgj%%&GxO<K4R}C#3cr+NB3e+%hmlK$l8Ik7CNGG%JCI=Xh$<f{>I*u6E_WUWJ*pN
zy+$p1fhFl`Nc*F+*3124=oR-7^YJxV(D3jLEZV_GN;AD1Oy(LY)12$jq|gGs-Ij&3
zimlD!=^3Mue1*-N%1B;>g^iti0dP{o=ybo;mephdzv`*|N6=ps8-DxOt+-@F?7ZQ?
zWcv#FC{e9sf-y3B_2IPW1sW<-n>C<1cBQI*@@5GksI36Fb#%1cVP>9Wv%T6HbhmN;
zd1jvt%%(Ku?wH#<xIc3mtuplzDE|o2ersv%K475}C(yR14Yt1c?Jw&wl<@H@<|c_$
zt?<zdAGGgAMhZUFV6Q^g+Og`0zeg$M5NUQvClweG;LtOhV0uU<Gu*h{+VeWX&o$X~
zNb6s3W6O-GSpscn65AFg71j;^LPMDyK7Tqz3qg6d-w8HSZr`f~clGRc?S;j6|0Ls+
zdXNFnZP$j}UnhXhUP!VfCD!E+hhKnGKQzH_Knv$k>5~>5+F!y{;7g58I{_&ORNk+)
zljNJ#4?o^Cy)V+>2Vz|t_NLbX>~%8h!LY%vy80wF;N~^OmdGCnmBwshkU|q4+FsyD
zy}#rtQe=o0+e_ZF^XfyH!1b?9pMBe6<tMbQ7+pY%GS*-x0kqRX=S^cB0dN_@r~?MA
zgOvTs^tw}0pN^6r$EPWXjSA=nsQzqu(|yeMqq}Rktt^;nVc7;qi{NR^8X3xCXC>SC
zX+VXm>aShj^aioJi=cVW9IX+d&c8ANhq=iuQ3@#bitnBx(~**5z6AeX&zIyVgBcS8
zns;XYq;NOH2Lx-n*3u%}_w2t_{rM+PfWlB6dr`c-P7caRxkEv2sNM~ZyL&AsU4dN>
zmS=yxFB=>t2FtS6Ot}m9p=V%5q_<7g-2Tma#{^gEveu6}GIM(eLO6&DyGSwWZd~7x
za0xN)JHM}ku#?_R?$f}?0P%;|(EE8bz!8+c@{$3<V)!{X)-%W8TkPjDApTZkS-z+M
zLJn_ahdB(x5cj-5?9U@*7ZJ^^+=p+kOlPjgvDs<9tO}T0WWYdqeJUw5n8^QcK2!0K
zU-@-mCkPz(0-NUX#|lwS3!h@Uly^N$+#K#)_RlhY!50r1l2)7HOiP8oTvT$zMd8x#
zDD{&ENgl5pEwC?&)W0GVXs{-!fpsR@<gFA#fQ<613UUQMgcVWW#bTwDcoSU)2(I6a
zZ*xasO4#bwO8<Y_9L?OdcqkNbyOa8y%%o`o`cyE-A2Tu+=hgu+ve2dDFkYk0>=cdK
z<wz|}X_e*95ArgkhNnbEXezDBMh-{#p7X;%QqhJp3Sd31oK`o_IXhebum;;Xjma&T
zumz+YF5pQ1{nE$nAvy=VNn^5arliPB9zTusY>K|Yyie>mc&UqiQ@zZ3sf5ULs1n{^
z|0L?OzzNC~81OVuSLT)Q`7v6j%UCudvGNMmcz(&B2j+6YF0U_DZ~xX<fMhEUF49TJ
z_6M@7|G!?ye*XC_gVV4AGPUKN#hEqpkf2remG(b=nh{M2lOKGp%RYhjT#c7GAq<-#
zlkTrL^GC+3ls>VHXEo&G1glWdI4?n$`vyct(AIdM+-@h?(JKvdwv_=K2{QXejEJS_
zD=h<f_XA$Jzr)gi$4iRXmzmiACd$OXifZiVchQI6)&i6GF-|TiWg&_XN|NmO-Tlo8
zB8<;B2S4A=(o86TfFqIKZ!evU7NBF>ymtl~OjDo7BD6Y~K@?I{RXeqgZBlLC{?NHl
z8S(FfTG;Ni9vF;|Dj7JrAb|u>V!AG>p!(GNRIYbt(<gW{HUljI7QK5LlQ(WB-AW1d
zc>bkc6CbRkVXebX1ja7nM9iEeQt{jh45tJaG&$AN?K@&YfK|=Uy`8J=1T4^#k@Kju
zOB&HSpY4o1=u$9Xic&+FLiMMrr#F`7yx~S$cFzhbx@;ibaoSRjTcUE@Vjxk6;Ol=N
z`r(dCNj=2=JV92-E&H2?8o%yu8LO_MP~R1L2wN-Y&>=36@HtamrQb|D^oGWwUN%k5
z;_c;}(UaE4MLmLkYL1}GhHc!(-uJ6HnIsFcFSe#piPLQ*{8)CS1oyz*lFy%j!c1yJ
zmH>DyM1=xv`l_ArEAMpYMd*gI%Z%@dgIq=#k_VJZ%iO#70WaaY#0JxQUY?QcF1K4~
zbA4Ja3!yq1ik;ufemMJ|{4LwSR`|9B)R$onL+X?txisOUgsW(Uw<o7WX|X;V#&q<X
zX%34ws{X>OsdOa&{?r7MaBYMVrjL%CY<j@dKE+SDB7i4{vQ<lh;RG4&$qhyc_D%;q
zzkM!}bunM%?JHQXx3*<?2wp*2AQ@tjT!C^ri15OVKXJFmks%iRz}wxtSf7@?c~k|R
zQTv(4>FJMp+taA|C3B*op}#$ORNN&f<Jjo>*h~To=IHScdBRYVu<Mh41LK?i7L+il
zshjcdCjgW8SNQzO5)-=&uIr55Fu4z}1_j1vi=0z;>@MZ!0EcPF#&Ll~$lEp;X#@+p
z2s}j1oL3xzxLL;(^~?bWt~~^;MacvV(Cx;!_Sj$LW4>`39<DX89pl9f*fIs8Rz{f(
zwM2eV_|Z_rwBDAVw3~kpSkyMao5MKU64y@%UJyzj4g%BL!SU-ltk<a)j=`=l@TlFT
zC@+!^q|Zk`<x$PHBCGKXXNNim?P6#Tvebp;kp1w@491i>@wL|#c5F|d0U@*d#dvq~
zZM8j}Pqa0NppyG#eQ=wX$Z+J@{Yzg)CQd&Sm)Etzb1di_{6Kx`DG(2Gt1q%#naAU<
z38MNNjh-UjoxukDH%69wwAFZv$NG2AP$++ts)DTE5dMcBq>GOA3T`HNu#HlV!FZ6x
zG4I)8-D#nd?So)SWGgJOpn4&RjY2RjtC$$CICWx7J@DUXm@!{LMpq(Ts^a$}DeK!E
zOHaLC)l*G;ego|qFjB+d$0x8>dW-C4x4wRxDU5$u#0ndu*PBL;X|$L&(#R#U@El5E
zXBKdM8C&+H@J@mA1jk-|`05bU9oy$#2cS^3Q91w4FDAx^V+klHaDKG2eJJ2VeZCRs
z-zTr&M)E}JxnccEqK%;UIPrJW?divtQts_K183!odI5jqGaz~jj71cxLKLrKnfC_M
zD|$u4<m@^4JQu2;3R5$$!wlxu!Jq9jjdM>NB0CazlT-vfc9a0sSTdF_$iv0Np1ztK
zu%c<;PS=onb|xi%lz9_b*;q18Io_yNHeD*MJmMbvhgNBKXpWkfXID?5cnE$~%LQqQ
z$UX7Yd!s{zcDR?2tYg2xduhJm0<6w|Zu6y?heES`W~MH@w!QqBFZTtn`$G<potQnY
z<+TAhP48Gt&q+hXK2<eC9WL;`xE4$Y3Uq7|WBW)V$`0SR26YO(OBLpT=LP=-oS&q)
z(y`|!Dn-|$wD~Fkpho(Ls?~DU_U2lx@%*UW=*PMiUn*W|gTY$w^YxD6BYUrD%Z5MA
z8_}g!U#=KS>OkW3Qr5Q(Hm6UTg=DMkZYzOjz!y?0GXbQFW$vK&O$!C=D&1*X$qP!*
zx}f3$dHyg)hr>F?kn1g1M!^w{0<U5I^XITtS~%JxEG8{8u8DX4%w?EC*T|-I)z|Hw
zV4au(CS&-o_xFYwaUg<zd?#J0uwmtPku=tLid|Hc`|);Fp?;21qR^KU&3}>(qRXS{
zoICcZ_4Cn?PA9*mvQoIef3W`KmIoMC1ja&>F{p%@k!uj%-`0m#Z)HRd=aj**6Ab)n
zktB_Oh#jGwY-1gP5ZT`~>PoyV@|55M1ft+rUzVlV!=k2+$F(9hZpt5zT|L65y2k>F
zkPx$|A|CufCbUHahA0EBJIteJ_Uri)2kiDqMcAaX4j!8iO3Sq+kLDI8%_a7A^juzc
zFHAlXfUSg*k*Av;d+0tE{rK&8@0H%G7tEZ=Y440r#j0U~xdrPC&Dzc1%YWWqJ<Qi^
zRi_DF^5Zo8s$(yi@CAVDY>VH_{?k{NsE{l)9bD;ARXE$l*)Je<nDFL3Nx%7OI1@p7
zwJ<HxqN5m^w81oR=XeWtJ8i&e-A(ElPsrl@KRNF6^US-Nv)USOJF3XkFR#Z^4}Z-W
zeQ6RgGJKfVx9$PRZH?q6cEc2|o>Gfy%PVd$JrG`H<(@v6ZJqUMXWIhD)q<dt)2fXt
zO^GwZ6jw<9KH^lgk^qJ9?TEtJsB4Q5-QijjQOQ&YskeVlQPmq+KavO6l#JKK2LH@N
zPW!G!NJolMvNavr_FtuPO(5wTtpV;ckbn!nk_>mW=g#Y$1JR)Zwi~Z4aZri+D^kj}
z_6;yXw?x3ad{vrW)b&0u+Rggg*J0celawIAVsQh<-WkHoM0$-l5xtEqmJFi!bwM)%
z#`ktUf?M}OwJFQzR6lCjcBid9`HFmpTJ(Hp${%1EQcC$;W~C+ypm}ybM4J^E1hKcG
zeS2X}(GvF)$B*Gs{1J@moPX;Ys0AGUd%^GBKu@o`uQk$28uZF0>=|vXQQkdIm@5RD
zn)clOF%qIyJRu>)7m`foM$w)wBC0?y1q$r@3s&<HEf}QkKVD0s`&uEzwoCy_p1px`
z@D5op(AzTLa3hKsax?^q?_<M*lBV-9{UX(V(C1gIHsuCzfB1>i%!1fA6vW$RS+{B{
z;|ih}M#xc69sAsuc8kmtCalB!4<&02-SA{4Z&^XkFQwO+t#R3?r>ozetLxO0DdK)o
z6~7ZmfJ*A252l*??6d7a#{NPo{@ZsF3N@!ke{VH<?2^^R{JrO(QwOtpz_Z;3Xr=5S
zj;9^vJR{t@q(aTlo{PuJ1u=mB*Z-Z(GmK^^Gk_sd`}hbHpGaT?l_i+*4FCT>E&S}&
zA8J3z!@s=%Wx@533D@g9kukh>GmvSz2?=?B`Pq^znkCuLLGq|$*?6B0qj|l<cS+of
z4HV>h;wD^)7H#w)9&h1SzH8`DPkRCn9OR&ADByg24p5Q=)p;i4+Ki7s0a&X*e|x{^
zjt2EG3>rE)qD$loGO0htGgUe-a>>1`_Z&ZZfd0x|Tq5fD43WdZM$e}Z_(zUxsGib;
zUmviefB<yZzPa3qNh!n-m|eBgxFPL>2zMSB{^lTpIVMUVmI4ou*xAQ|kK@j&+nVU?
zTaWxW2)y8)%<j@AetCi#^PDG#=4BT<(lvRvRuuO#$!I2L-im@AiCZG_gn7Xz`QE^W
z)Q5?YLydbQ8;rlAXMJvvzPe4M!N-935&qPu%<Se3CjHO#=3k<q6IawwjSP;kn_&i^
zn%|)#Ojhu7ymbWyL9z(EiC*Ira6ci{yzML{P-t5N<xryg%r@?5L@yND=b*L0#qfl6
zX0C%N;*|hk-p-0C)|;140LoLSlr<;Z11%7&6RPJpT%LWZ048yIG|dBB>8f&WHz@mt
zZzG@<!N_CVptG3iOTqSsXJCy*R0qVV7io*2o(Dn{EvUP$`>uQ#sY-e*@m7+qHaISb
zA7Ocg@rOx}^vMw8zcjT|(Z3dDdj+;$@Iix3&VjfVB+yD12eDswtRAI!-##TcxDM#2
zGHbq%QLkXQARX{${6o^hnjV8q{-EesjNP(P)MIBSyT!M|y`id!0BJtnTt(|Hzgwpo
zU*NdspTu#m%WT0O0W}VPb}a9!euX6t&qqAFOJ_BzxysvdXonv1FJv+!${$2Je~xb=
zlEk4)ig&K5-m!L1e(?f&#?Vt$|1-;@4No#;-`F<{{=SKN3j)`RI$S`gJ}EQV?45JR
zI;IO%WNYRZ&}H8c5jBZF6z1<cX&SvhQ{~6uA+bU6C`8W!n0ezq>jc5{UNYIVuv$9b
z#<9erBpw=(PhR<@y-NDyE0@sF{ie-G5b`x8_w<^$)O23Xu@8g{7)PX8bMLm9j~=JL
zasq;&ERlBw3{LUrjX%3JqK!6OfFClQAFYF1UUXJp*Sl~}O8jXPu+5--I6$!=YLwJt
zR7W>ahzli@g5lWIrp&wBHdM#V)H85KC^VLuSEDzBV<!Ghj+c|tSqi=FkM~@tnav^!
z6bcd7_WTMN?otuvy@^82^c%Ik<w>ubH)S3z((AosNcjClYILpam{IfM@1vxA2%zlZ
z&DcQ{2Q)<LpYG4+f3l>0UWDVglBymP<g;if#0L7XBnTM};{MVr(FXiQ**_zDekyOp
z--12$C2rXB_+$pOxL2=U1_S}~2=}_Zocq72cF5@N(noL0U?ckF;k#f~X{zjKzb{nv
zAzsZ!)T{9byElTqx5m>=>K)xj{DL=cV4(PBF%^QBj5k~KPkEH;mgt{hn>u#!ngZ0B
zOTw072%Yg_Bn=A{wKho$<%qoJ#ek}G(iO|J665i_K&}p2*w3_Ys9R`?8yrtvW}n&_
zU4_U@=2}AbZWtN(jl-Xe`=d#Pe4YIx-!?A)bfI^3O-Sd9(tpz~86Z){8Xtn<h-mfd
zP2nQL%+6xKx?hKR2h>rT5F_UNxLS=g-6=oeLr%s7^D`iytLoidTuDf8@RJQK9%xpA
zb%Pgd^40W;3@uE%fl0mvPh(WA#L-dEbb8tc5-*lp=zT<Y_xrwBHl)E2m!G0SS3u^B
z^=kw{jU$ryj|)bui+$fotl%UMDzyQ*0!&W6vzX5EIoR~|jjExElg0I2LH0z7?oQua
z;%LN1_HOHp7~hE&HL_OoafH*$Jui3nK@GGApGXe+@z21P{H5;kpFeh?G|GQ60j6Cg
zf-)fgfDG<i(R1a0xmz;%w6;CaJK`xx#aY&f9_b*aK6jGAYE5l-EN4T;ceaTWW%ycp
zO1<$Vn9Wn3v)GOd!@J?mwX|Eu5~<Z~;3?ay6Q+fwT>8hl%(UcH1;G>AH`oA-8tQbh
zfD)lS_F~2Zo14+;w*V9U_R*WY2ffsZVVWm<4|9HWwPO0^m7T6W?Yrqc*j3)lFjS62
zLMxbbo&+=3P#>qCDlW8nnh^4ByxA5j!p?sC0IF^*pYq;N-lGt8`ir*VNpX8a`XM&A
ziL}HOa7^d<EChh9wU{m?2udW6PJedaNSD`6K2H*Tx7T3?1m3+rp66*d^%khdKtg_~
z-%)X~OR9f{ga>wQfu<NJc0SKS6Nb_DyZ$hR-s*SM`8fLzqdylu_!c^UUsDKhvSCyt
z(eowQgJYm|wcn(QnVJ9>Ac9B<5gY?{d-aft0hT7#Y+`LJq(jut)UPrIK#snYKaF{B
zas231F;dc+l1BXOIC6ydv1YgvyIfrU=6A&gO)fdP+z+wS1YyqBdri>$dF^}o!SEYs
z#wq&J*{gq{ueP(>cqF7i?6E#$678*40^;ZOP{)<TBIuL}F7)|$?Q3`&oZ-CCarmYH
z!+oakLJngs%C5F3pXRe6s5&`LM)JE@DCi>{Kl!~=50y}a9z^GJ+LyYOO~D}Gek|DF
zSq7Nlo^MwTULj*nCR%D?H%~QAz=J^!Ej`)0GdfBY-fX!So6QttUT$xAGAYk|jYFX-
z-Qc+itr3~GABOd$o}6tIju#3qyAWRZ4Fw|i6YIm+tQCrekwm>rsYi_-$`2U#?0=PP
z)`3z0^0C{C=^$;Ae!wfK%QL&V(lQPt(x8cW>Q~NXL8F<q9P5^a;kC3V#fW&|A(BTf
zR&&I68v!Vb9lAiPJtu$u4GfZ9m;>GO_Ti_?UDe-t<hQ15?Ie*D7vo)AnYqC~z<r*G
z6DTHtN20?25m6M0g2=CTswwt9hgbByJM-P+FuMX4|0r{}<3kQ)-e3>QF3j28uDE$(
zrGk${R=IaRXqc8>3GI^{Ds-28U~XTvbAIOd>k;d<TnrgZ)(@+B!-&(h$HMU$N=D^D
z(%qskbJUTAw!P@Ys&6enWP28OZqC6JF>&`zD0pRdcao{7a!+&-Atl&pNeo^ZgAbpw
z(IT=eyI{iO-nm6|uVOEl$^dV*WNO{`CY*SLpE7gD`J92JY<-3x#DVg=GIYZ21d(s*
zDlyGx5zvGNb9YMC#y_aR^xTgpt~F-KFsdl4@2G0bOWPfb37F8}*U*W7?R_0)Z~W|?
zj;Q}yl&J$9NW&}N0~(4#baBbf-vLw*De;O|(DFdq{I}Y#v^=?RgcO^GJSGr4-V1Xd
zE}o*Gd4@`3scA*C<5Xyz6L{mVefh1IBV}g?D6`ZF+cE$v%$1(yl_NCfdP=A}=q7zn
zy1Tr_ZI=w#pT;+f(Fv8kuhOM89MD25+jKrn8o-nfpPdF0^4e*8bE4C(^A2Px61=Nf
zisp&G%d!|1rP^9+-#1XJuQ@hu>saag+>-il)cr`;(X`cH-bJ&^5p>+X+Q&6uIY(HM
zWWR{#&*IBDql|G8ld{COlJ}*>oYwwN_@_h6e`bsb+!QfrsNmiNdiDOp&0Yu{em~tL
zn7C1hMZ>+0=LW~r-J0J#*DlH^EW<yll(5b0I=M^=Lk0Wo|6}Q@!>a0@E+8PGv~)-}
zNOuazrAxZIJER-wxO5|pfOLs;r*t>cp@4w&xB0!_U*ZGLJ@=fm_sp8L*34d#!Q47F
zi~;rZ^8_AHGKGqe7a&o9Ls40VQyM?S44c)TrauqD7^l?OXfVL<eCKs3Oc2%uPBrdX
zn^P&;1|`vR{~Q?6*%h!l5=OizR~h^q$VqlGe;Bh<_fTP_$H*~!C{gG7C=%Mrp`hy%
z#49ATib0wk9&AC{IXi?1rd%uYeb!KZ9+2z>4$u1hQy5;S7t-9Kiw33+1%TEPB_eu6
z-E3Ph2Kfzf#~)3TE<xbxD5(;Z1g0u&&I)*J6R`*p3t)Bm@7Rj#w_MD!tm``qVwLaQ
zm28sZbZXCR8+kaB5lUE9^h;Dn-lDu)rwkY05O(G4VfO}5`nj(OXrE<1D4=4UIbkPb
z`dig+AFRDo?X+2n73Bv!Y}>#6Sq%G9{K>KKe7CFt_x4~z`#kh{hC4F-c~Hy+Sp7dS
zK+tTQ+tuW^dFw|!RoqL%@htv=X?U1em{m=`e@?C7u(5LYNatw}H39Qy>6#HYn1896
zk{&w|d{oATfpAzg1TXPL`rCb2%`RHf@UJTnf?XXnKJ2K10g;=J)7M^FnVjr2$RO^8
z`WsRWd-L`ze)&X^%Td?6;M0lvzSy=K9DzliI23iC0ceu8a{vD7HTg2VBp5Sr0x9q=
z5qJE9$9m0jW`W$-?6^(MvVIj<7yO37ix)2nBWXk{3oKw@Q45<DV0m+sGYj4Aoh!4R
z^(AHVWxR)@ghh0~z&d+@!BjFQiu2s#89HG1Pc+OZ?#xpcoDn99fNqhjh?yFXGLM)@
zo_OT2-uLdw;&(+){=yQ>W2PP<GPWdwnr%&MYb(OCjfaPae(NP+$m;59j9y_7#cFly
z-rb8$>_jZ(y=y_?>)KWq^Iai}anC&=2Wiey{b+ywnLMKy<9$n;qJMlV>$kO@^OEZJ
zd_%41fo1LvD51UITZ9hgiof0q-Wx0Ex)$GvM1_Y(KhmG1(%JR}d~>r}t!9!6^P*pC
zukuG)M$<Q&ol_O`JKL?=uRGj9buwRWmLO}|o^V@2*F=09BK0-w^Mm9Ka(u!Y(e#;F
zJ0^iOMCJUC!R*JnT)dE&zr<}hua|#5yn!1@6DBz!Of3<M*-~TF2yh+JlotBo!{@h_
zPV)Ji)9x@OXU3C5QIy)JL28Kc?5O?U+Ddv6WUG)Ki_^)k*IP-7f0!{-o%<(5Q|Dd}
zPgUq!S*tUJIPpJF@!a(<=c=ZX)&5>!j?HO(8kk*tB>m2>{b^M@u*_*Is2?t<n@-mo
zNB{SWc#I=1`&gl!1XfDM0QR!DBD>ef2d_Fqv$4x`Y5emu!_~`%45y<e{tR}=xGrhS
z3@29?+t#e$jAmw2+_*rODL4q^W44MCmCI)SA%^_<JF7z~eDP~j6&;)gM_Bi8F~aE|
zL9NQqkta)XLm+K(EZ~&s{*cnY8+un}X+>YAG?j5qT|uA3`jQ)b8l%N7E5#;dLH^m<
z>)nI$CPA;qFknS%alKe%CF$wNFq>|j>QFtAAe;ACk(%Spf&D}qLKogz&>?teW{RM+
zJO7GCwzWX?!MX3k(d>5;)eUlGt%&3ka;B+~AC<{1s8I<p@QK5!{FhLVGljmB6x?wq
z2r$Id?k&iJR~O5mfGqbq)6u8zo$9UY{=wm&E)m|o7ixYUk$APg#uL&bgimkVG}Ns5
z5Mrd>QVrp1c++ZSu-C%TT^tP>Sy0lqpwfAKBa|yggvh>twoYN_Zs$LQ2+B><aY!|q
z={^<LzMbL~KsOk$|1+$ZB${3y(3pbNR4zN1vB;@BASqc+xB3YVJ=Zg6quJ4;lvP%h
zhJLVU!&C)L@udYa_lxMvR-qvsMF9=;N2-qUiu)qT;#YRmjQ$<Sc7(|$Ct0Lry4g~L
zK~JzQ{q>6V=P%#c+&tp%pm)RP=Q2Uk3n@wh6CjIL>!rPR5c1o92qBghoLu=n$VU=Q
zM)oN=BTf2$mLA%eCO8NG!H%WzS0poquZM4iG7Y6r3w{2%B3r~U%xb@0ph=F(&3v4=
zzkMp~Ty#{;Mb_lkXSqTgH$0UDVW>q=qA@#LIhTTJ!n|`kv-i1&s%!dPXNc`vj{K&z
ziuoc$_|8{?NDHHb95_@A$Bs#@^4Z7MqhlL!ONg;^&#ASGitoZ3Y6KWH9LH#BeNrH}
z=O~PcTj8z<-(*qD2Eqp|r#tiBWz4Xo?{mAl)lbe&@6Pmnt`(g|3jRL3y$&HU6T9zE
zVloabt05owDH0!g#u#td^Fv>aF8yD2!h!5{&GrPBs00c*7mZ{34@rvWHE$K2H^4B3
zcL~kAY#XqjsnU-=*5aW0o-R{#ezqQa0sr{LO~jwptG;|Z@<7k*Xp&6y25Sz#XnL;~
z!9okaOs-N!Qj@ofdvajae@kKBs9f2G6RD7`S%5gci2i?{p9r(}0ctgQC~kQ9ua5}S
zy(v^XU(AcTE1Y;KM=tQr74Hr!vMC)%9wFds;)RNLwUo!185LsnOWm#ru%i~SJLPC7
zPcjgCWdB_Dc!ZgEj5<V!+o2;O(1J~Z^x(f~qom(I*xV@@G_jJZ3BrxA?`BFwj?@2f
zV<RJ8#KqtAbrgg*RlXoK8m?0HZ!p#kVW>Pg!`=Ueq_K#X*~0`=w6J`S|J?{IxE#}n
zHpg?mMqW6jlvSgljt>?!^&`ze8^<!{>fB#cfQP59OuL)&zq?P$C}3YNaqFqANKNh#
z$*}6t;b_dw8sD>+9{GklJN!oauT!eMu*v3kN*JU<<!1h&rSEc)<T@~Behl5cyBSmv
zz4zS|(+HnKrmM}*B$~bysj0ln<y%JAPejO&sjo0VEHQgAF$FCN7s=L6M!_b<?Ay`#
zX=Zc}@4PcOihRcwA<RNYmF1PayG4eN5`^6>@G9lMhuwi5IF6ZybD2V=z2>(sp3T7h
zYcz#zatvqCWIXgp(djbDL&T4e%=ey9jkP~E`Nibq9MtR_k3Nv>KB(npqk$<JM?yz-
zD7&EU9ov-R-x{zYHtsJ=U|_{{R$-n#2i8iMsbG1cYVPiBpBkcP%~`@f(Avs4L%s^x
ze9o&XVIH}#p!XM_sFrfN`bHG_EtYAoqx_CKMgFK;<fMG0TUtzg>{wykHsWVM_1(UF
z<8eED0EUW#zTE{_EQwQtxFBJWn?%?^F@)ND;Tq(kUzaJdJLG3hOwaaHc+*^wbWo36
z&?=S4U|xMoPVVT?Lbg?_ga4KzyyCybf=ZRnS*9B4X|hav%iF|5X1QVqi<vMfZOFPp
zVm>&P^qh{?@t*G5(-tftP_>8Q6TF3n(vXM*vHVPW^@*T54btta-C(9^oI<Hk@nZuK
zVqYN=EqwnLtc5~%i2ir4fw|0N2;8oDvF<t#*JN3u8UvTSH`A}sD4w@%SWBEeKlxNr
zN=JW+uEC{1gk10UV19~=Q|@!PzlUZnHu>R)A%3qFR1J?TI<NJAc=zhB*^VM&k_>{*
z<B9AY%^OoADzl7f6=%CAPMpYI)~ZCjlU?u)VU)!w`TtCjOM_2V^qw=r94sXCW1UUI
zZDA&N^FFvzs2xp+0AUJyH9F@Tcj(1#d_j<E0LUqPjc(Cd<{xO<%|yv3>+R{YazFh!
zZb)T{51|O$u7!m-JMnp6vibHqNFZS*p7*U9H8CkY@2DFWB%l4Vcgz{cbRL_xs$ly-
zsdfrPNnPv(TK}x{>7-a;(d%*I4V(-ZW+@ophK;611&fPC-4?ZM9}mmUdpId?v9A}G
z*vU&>+d=oY*L;x*?FabaKTE(6_m3IHGs&mtKIYR#iPAJoN)W}u9whr4(zPYlyg9D7
zg(>b2s})TY1)G?7RgfRzm&Wdo{+)m6SdS+uj4LuYzC6zNsfd%v#@tQu-lr7PmGnD`
zL#}9mhxnHR*un0Un;TZoS|n4I?2YBgv;{#Z6~B<jSqh3~rYe;|tOQR%L8T;prmpR(
zHzdW1qMANFQ8jjZFsVx#+(aB6a#7IkrHsqYm5Z*>&PI8aQC`(ZL88l~tT@JOFM`10
zk0Xiz89E7U-K<}5WB*HfFR}NJ^TC56p>um&!sc(N5dThPi?G)?*u8uWyW-@YV3q-M
zzB4>bKj4&R#-0#Lo@iL#kX&!229izA2<E?BatPKyV{~y|gJ*u6kk2mZ3mQ9dBn)X2
zM>j_(HAABIpiU=y$Ez52YpRXMF4nZ1f_dte?BwZI=m0|$(*#-kJOiUelj2()TvaW6
z_K1E_gU=5{Vane=k!s=RzB99Av{s`_+uzm|^w}45eM1-n)8)5vjQ9T=p(A)+;6<p^
zA(ACpZp#3FNgc#ms!YwX-%60*8E)GF02|bj?EazQT)EnJ3MD@$nG5L65*c@cI|Nw@
z#bS)*h-U;R2PCB=ryJL&()c?zXgHNu6CNH-RS2G=!#vjCPCX+PP9dzFpL*Wn%l$ip
zh8ZlAV*r4}gHQI3T}?8(B}iH1aB{!<#5Bq=+#L_$_WHh5Z?BmGtVDmt_kJ@gW)4uG
zYFyC>FLH?fW+UQtz7iG<5!J(sBdJA?lKV>e98LKML$5S9oTM$`^*&dLRnAeBku00P
zsj<;l#@(m=HJB2p8f#KBnghWNlZ#B~;v}IZ?pLDnkL@^rjCpn0;%146BDs~^;m<?7
z(~*q_g8`jEZ?f~)!KA)mjAssD@5l-8W{|V}CJUVz+perq(a^|bFuR~FE2C+Utv%`F
zx`z{g^s31)W8X~L<~`pTr-@@r#~DyENL7D5;{5Aa$mc%jDdTQVg`wHglAkZpP+7!p
z6LwXqLJ}#EN|R}Ycz2Kzva9i&7N!tSW1A3+2|JVgJzP#sN2s_8DvN$~k4O0X$}f)+
zMDy+Nazhd#*x8Y(Wu;6M1Y!zAb2?12AOStKk2vF*pGNjv@ErJGX_D6UkVhmEUh`1y
zU{XCNUv?P#8}mTk0+lkWc?YeEVAuSijKvTV#+MJiXoqBagB)h>wvx6rQT9g2<LJ|K
zW3_0oW0&J-G{#5$@7rPz)}y&!k5Jz>=fyi}U+T+=D&Yr2g<~~3Dp1FfrW-{3KwwG?
z<E4Ga;H0e*&LT#>iCDF|K3%NbLe-GOs-b=#+})fbA-nb4I!D^Uc#wXe28Yp-9Gx~@
zt*3V|z}!u?=6Zn{F1Y7u(A!qKgz9Mb9^fiYLb1$t*=C<3KGAB74DC<+EkgxTQtB{f
zrdw?c>*)RU3zW$Gp!kp)w#rGs3ay0BsI`6N&T9UHq3F=jVU8)>c$tVp!p>e64lq!R
zO|t3Fn#>w5yC-}UO^dzz*a%4EKN7Byh_O~yILnZ{*Qc?mJChWNQnrZf>{jTZMXM;V
z53gRkwm2DjL46myy9);OJdu9;20vPQ5)d3Z*&bOkB#knH;Xnq%b@Tc&mOPY=jVBI@
zh$ioKzp-@DAty(2|8AFV+O&bekS=hn%VmwjNaw5@AEt<hL|M)#Q^q@@y^bn$%k6{z
zOf!~mm{-*^>#I8X5S%DMM7Zy~-;q#zPMb(Gm@7a;OVSVJ3qEu3X9Xc}g;$UnGQ8f<
z%6h=z{sIQeHaS`3Ia((`ka7fiLS^dV;BtTX$nKGsV^CN{oV)Cdjt0i`Bs_|GH?`-W
z-OZybIxi6ur00rl`tb1{@P<Vnh|st_y|&S{q<@pO&K9`BH-_a~F@FcM^jA;ZGzn#+
zQHlhVGH=+eUN62>qB+6(b^O_C(gLgA_i9O2#-j8?pUMq*@}yBr`v(V(w^KbLQi5Rr
z9_xriFC|9}nn==&8Kn$`fS;=Y;-tvW_b<0erHT-;2~(+4*_0q7A|@1Djdl5Sy7l^x
z)Z5uklgti53R6(SNYSvt^NV~eWGzNl6ZXDD{y2==$?#!nN?wg6J5J;Dx96YUhbFuz
zfS3i;(ejOAw$AYrAa$lA#xelw91HqnhYzkxASXRz>jiFEpO0TkE#EMEu|&rFdyP#f
z>AP~0J3=W&!%X1-J5uP$`E70+BTK1lrv7)wE5)k95101(3z@Zj=l65x6LTuHzB{4-
zYty5nrE@9{QK%E&I!Hk8_x%tO5u5TX?g%4<6_CW9LzCwjd4RSwl!R2rVTe-ne~T45
ziW8&6Xa_rg-7Z#t(v(!XU$~=`OcfE;N}E7w6Sl5`rL8SPoN`s7m927K9e-n2s>BY8
zK(ql-5oV2j%)hneLrLV7P}jd{e9OkuZAgLpG35`!Eql*(o)k^XkHsxUoP3h-K^qSP
zcH=OBFs9~IobLu-fwk+;h)I`>!p(a4{wU0B=j`ppvT~)a*i|J_SdRTt^q_-oIaRrZ
zyW&($`LQCK|JY`nbs;AK&%kTvxRJseE=k<LQHS>d7erIb|74_N>{t9_%wpwmSWVLl
zi(g&3S*f(bA2nr57C`REBLQ;|cGCvVx`hg&nfmEC>zIBilxJ}#Yr#~-=7~{i{6wn2
z7b~xlzxP1vFm2bX9fBcUt<cOj$mRFH+t{V03v9wE=1zH=%<K0E``Q|ZRtI;QipvbF
z9Ou4DrxkyMgfKTz2vn1;fbprmoaMS>)Im|dydsGuBUA6gw-RLPiKXFaozLcZYBYZD
zkGo?^>i2J0Cnv1~-glqKTR6@RwlV!L3W~rQscnQ5DsiinacTY3^$gLwkl8<?9&4hD
zD<YUvH$N8g6BuLgA1QYG;xMq}JTP+4rm2&zJi#yE9k+hl3&$#bvZcm1#_?zW*AFV{
zNdJGMi2!_KVBsP=ev}~MrvBtnxYGIz3Men?ki04Y#xJCTWIw=(UplOKOMW(gllT`A
z-B7hN$TjRX>&LIGU3<6jw5wMK$_A${=_0Y>o^}4k@G@mz;XyLlJuAB@lofZa{hZ3$
zz&{$v%yuJJn$+Og$|imK@qc^%W)YtM&P4mZIHcO`@;>kHrz6=5ucE{i-98mh<X32W
zORK3?ySRI?C!M9jF+1%kHwDR}<;3ZHj^|e&c%o;NA8SoHw-#F@)9?>yZ#ADfm8_S?
zcaxU|U|7k^Y&JGGsIQBoD`18P3p23(b%y@!6Qs6ue~J(Q3-)UYbNsK}!;g#^=2()7
zL~9u}4;jP0Ocw*H6?q7_Af5l2$yPh-8?(%awb;WowziGmHwhN+E?c)^xh8Feg-4xn
zb6mbsJC*z|cG<y^uBfKr%)ZY*Zz)%#Gqh7k@Zp25=tg-JwcHKHSXBA1qCOOMF{g1h
z7p*k0`@il()q)DyAfjcr->LpIv9|3@<hU40TBDmB2n&;(YF_sFw&n?o!J@?3H+J@Q
z#rwZE5`_^sZh=!Qd4XKb=X1-##ehpw;@MBtG+(UDr=aJRRuM@FF>_s~8O5KJI#o;W
zh=od4Jm;&L3v@r3I~u85K|T($IyssMVQvO($s=J|Uf&x`(IkxeuGPop6`TJqQl2bY
z0QZ3^w6<8La>#RIhD3XY@jc?SE5us%OMJDRdV<h^{aF_)@9e3OWgOxt(Y)BHTzW@}
zGj#<U=R^#1Ns2K(!Qt7jQJ-2fNbbH#!AEQDYpw2tx@(vjacL4Kj|Il-l>c4|LJ(>B
zJEmF#6$b6l&=2O%!KZPIk%mtwx-kWcCeK*?Fu?^E-0gR}0}j#-E>absrN^v{9?g{i
zAzRv_bYLio9y{5gv?}Uis0Bcze|oSD+V4xq%NM-w)uYc}W`k?dTtol67IhF=n348z
zVMz7e23dv2!zrU}CasWAG{`9=8R~`uT{iHcAy4%qr)D~dvvuD`hI6Ckvar(d$l>>@
zXj$94C*Dfk)D0)lPYzgf27T+9-*DYv2*3yuZWgdj@&53?KsiMbh%Ox8zTk#zySnD1
z)P8nVptY(3v1iE+Bn9_Xdoke#7c8O-r%I=%H9z-Ji=*`=N@&IX(4ddZB}!;`boB02
z<sO88jW?S#PGNk2-Fu7KUmWRH@n77NyMFi$HsZ!<wbbdG{}$W`yr5&>OuwwSwb$<&
zB5|az8gxla)YTt*YL9tcV>Twj)6i8#4eD(4BVsPRsTalRDKAP?%bhWU-gmcn(K_Am
zaikS)4=vuLWfmz;d43pkHeTGtCQBCU2N%&ief{6=Btdd*FPAouBP;cJyj7Kpc};K7
z%|A4jj!!W0U763F^m`74jA1#V*~0f`R&DKbN)R-J4%KS9Sl@;aT^rEuUu&@wUL3E$
zv75b<nYxks>2BP$M9DTxHuY(RWOuOb?f)WL2PVZvmzCtL@nV1qEemfVr4FJDZjDUJ
z|7ult(iSmG%?hlvPt4Vxt17lSY#4T0H;D;1<WMT*4lgzzpWv3ujXRmV7A>!wT!U~>
z0z=2StTtJ8D+KZH{CM6s^T{Q8|Ch-8WU1dhDG#Y>A<?=1UGcR=(NuaH=LubYCU51f
zQw`@gk3zzz+fTi>wW$|stTFhPw4Ff<1eG|ltx_BQse_M6xuZ|u$JkLp0RgfXzN@`I
zvBlq+9;)M32;D2+B;o?{tbZ7q3W{@r^!1bhxFeLi!s`7g!kEEUfDVUoe#SHuX+d>H
zlh(rdK}WCCKSE5*No3(~`m_1RS_1HTBz7%2rl!z5c<yf>gXT|vJNX@A{%#fb9P@Ow
zCgq2GY7D`6mGL8VCsVOM$KqT6m|s1UQ@Y4wI#lxQg@)nUi0BlW09t-+U^^P^@qkK=
z+L|es=TRW4o4`2DTUbvx8+E4Mt{6&aR;rSe6irzC1QN<DVMGb&I;4)$bl&carjnd0
zpCx=GXov-n^3mYNrGbo6R%_Qm6ip=QM&MH`w;UZMQLrlylVGIz6VtD=D4rqfTGvV4
zEO$b@JzEyvZQrlfi*SH#&{cl~_c8Rc!|>28BMMvrOR3CN*q&4_;~cM2W!h-tTysYe
zsc;hj#0#G1qT!L{+;y0uj1w>5e%Jh<rP&6n#o1gzJyFr1ttfWYcJ0$$fs(W(C&g(i
zpd6fCM=YaXU79YA3{t(oLnkA2PagmPcK^__#?MdDHd9?r1@01HrIR~Wll6#Yw&7fi
z_l_RA46<7P+og}DOaeR530=6+%F5B^Y8Namj!dd@2>77Gg|Xd+JMSbE1nZbE0G%9H
z{9K`wSZ1cC#?78jT=L&pH`*L94&+}gw%XHu9Kn@V(b>dQu0L)_+r}dmmGf-UZ|1X2
zDq^~5A_ti4-nZG#4O3hd?~m`nZ7ku5`Jbz7|JsbvWDX9C#u(bFkLojoua}mzqlp&e
z{(D|$sOY730#N{FX##GnXf-jh;yx54gZhE)Cq{qqLl#DVc7H~|%<lfCPPNm5P-er(
zp=f8iELr)~G6${+2hJREzT7Jz#os@w_BeTWs76mQqAf*(Y`fx=BfQO>uToY6n1Y_C
z7^)AAuOQZpZ|S1@S9bwVfEqv-Xd0FIV6Qd&UkczcHqw!S7GjHr`LxtJFIfpL))gzN
zCkp$C%^dNjddIB-7R!8KSd8=ANAZ+~XgN4c@pnd#3g%UUFujFJOEuO?P+u!Rds!`*
z>D!`p(UC(=92?eYWwm!5q{d&+U3dV#K@u_31scBuq=tPGK9{(Z%MrsQ1(6|n!o#Df
zMkmj41Mi^aO!Q+x=V0PRc9s0<A|?ldbTSM~0(nr9Xqx$>1lZes-KwtehZgMTwq50$
z-=z!sUgEyhM4EiX^O>@?@IPSYK-+oqx;Ru(noj%f_A@@Qpj#^LrHqMfk}I{6f#PR(
zc?QSrHCg?Zw!Muq0RT3?rZB!8(UICE`r*5$7(9N#>;4D}lNeTSIw>KLG+fwHzW(DW
zgWui7*w@E5RN}=7G3y{SCE<U29v_a(D{*Z%BxI(a^=+@=?=v0Favc;^lO8Ibvho7S
zyW%&Wz)XQX3(~B%+cXdGx~lY<Af6+prjMoon@ZX(FFxOZke(g)=<mJ_`%7rjF2cr4
z(Pz$6ZnCF5B^bOfYIB`^kL^rW1pf=q5sCpW6<7=WZ|O!2*Uy&0#UlUGKs4ZzWe}K;
zQeh!m_YJfw&hDFC*d7^k;4t{c`+q}&Ft#GDG3sr2+jtl<jZDSFtLzzokfLpAtMq1>
zY=2_!rYsQa24nvl9GABkWVQdRSKBM$&FS35vFg!Yp>ZtuQ{2epQ9;yL4dVaOgaJ8w
zbXvoK&-aFW>@cPK9dhWAndu`=2!@<qZE9Cy69!3T@h5Uw0G|sS*N!3fE<^p>e>4A(
zSvMeVR47T!_KXdENNJ27wXjKKf+I+;iZ3@|gHS5KraObm4-mw^YB*)x>j2MExV@1q
zAt7aAX9gw}7F0K~C+;v`%wvM`3)G4W{^#5<K!LO9x!1?vr)6GvUx4{4*XSoSX21lO
zr^`J?S<Qm<BpO=8O!>d06QKkrfRoHQtooMSV$8PtnG{l=pjaHs0k0;d4Hp||OCY15
zIS{5(oD2`%v;1)N{cw>J_33`=H7FIL6YL09OPrt&5Bd2YT}c|l>|R#&GudLva((ms
z+>H~%WcS}2f(WK$qsWN65JzMc_alPHnDxxMv@X6>T(z2U6L$FVz{Wn)V`Syer;#^m
z&=nCizJR71XdsHzfADt^<W!KA<ea_zV9D!G>q3An00xPi4Nc=S-}B<0g2JV>#vn6P
z64Sj|R7S;C+7FgO52e8W5+Q*RLM1Ln5K(xEJ1@WHj5l=ed|v;PzATA00Pfa{-&RNX
ztrn^})2V~6)ry@kkg34ARcSy&qx9l}f=Buxe-g=@W&LM8wYO3^7kK<8>)*fNjHgl#
z*$I1`6+@a3emGtthT}r_L(8{mw3gV}JJeup3wl+`nV_{fR1$o7bcQ*DQr!Xb(mzr-
zf6PfwF;X>feEGWk@Ay?ht0K+p;3Q3O;S3+Vni9jCik#VO(vz>?nE+zO_T{&Y!9{OT
zxvAxkb>9KWNj;*w8`m{20;aWUIQ`X?!%exJjjVNZ5@9b^O#Q{py6PiUz9tKf)`R<l
z7sTBxkd;!z3@C@ye)wbscbtq-%Ovx+qd3sP=y>*m_{ZD@m2lD56iARW#F;9wYb|Sn
z4i$aa`eQYvz*{mUSKM$pgFDFXtA4@)S1%PW?DxoT`L*Shq?>7YIdvtZz81z;hm)B-
zsJ&36?vjt`mFHaP-5%59ErEW>9#!nANc>V*z8E7R1c9b;6bn&$)2<trsN1Af8`8RH
zkih2|m;C@qK8cAS#w>m0{%XZ@Uj~pvj%R<wpj1+qR->Z{c8STTE<IN+e{-uNK>7ki
zytKUI@~uQlquf{3<`^`i?+BBYk#tm{F&GO0bZ}GyHOn7^WC^?c)WSZ1<jiEp7@&Sq
zS!*uWaopyhL`iPEG&`lD@}-aqlnKR}VyKb0gOFHTf{EG5M<W-9^_z>O2cfD}-%lrJ
zq<GiYKOE0?AF+8bNiiQ9j?mzoq@=S%D}@6Gt~AaElWqvmTPZXxqT(%IzAM!cdd!|-
z8yY{#t#umkyp4H^1K$?p>)p<<&7mbyfX!xY7kwkzw|THgF(v!|9sqPmc3J*L-r2jx
z%A}Zi4f}4=><h_kP^!QVCo6kduqW$_dT~ZNDrNbynXhG-8SI2H|IGKlM<l?4)Ev)O
zz?RN|pK!As)fU@`6(fa#gEbD+esWgbwX(JLi}|=nb>AE!CJAxv5@kl*t(G163>BDA
z43+s!o4FG~pEjFM5=mHjay2V|Ej6+mV@5!&6`x<LQ8pv!brzO+ZQye8$Y{UjA9KNh
z{zctJxFdBQx)CA;MN42gFU&#_rP;A8P}AIJ!o%`~c^2SqTZAU3Wc;+p0P{Wl_1bvt
z4l!X|3H`3`!BV@VkjXiHJrZb}h;l2v{UqRUXS2kp>~ch-WJ(~gze0}5o2Iogp;{<~
z;LU2}ESveUsN!h;D9sCyx<%x_q@c);vI+_3S!Be)YGJNbst-VrJm1Tce#~0H8I1TC
zD~HVf$#<Bs$s_Cg-~F*&Jv^Up>)y#aNSLi}aLC~ko#M2S1ZF#QVgLSpHOy<f>woP*
z+LAz$r?hrFO0pa%#T8XzeH{`a<#@V;t)TC^3q(&~Rl2>Wg@qb+Ai96q_JF2-pxhB-
zPSnLoai2zzgR#_a!nu;YrRDpq^@9hj5zdpE_yN|nJP<Af8ejB0TD?Wp-EG=fv|$BY
zTX&B|4$P~4mCp#+$IzYm!_6Jp=~L8}r(P0KN_Z0|k2G3Yo!t_{*^lV*u68ekPUgfX
z1SuGneoa@s{_9q%$pGwzxOrYTK}oek$rLV6dlFKTFXeTc@47pV(DpQJB=lk6@j}E~
z3BF~*`K81CdwjTaeNkfXvh1|)6}Ir%XN_0quWN(9=Xy)qO=kq^IG|AjC@p^bwwn>t
zL5w>PL$SZj(f$%tW*_SAc8{lmJi$dyNxIbdKRDK!vB|sF0m|1EzglqIs@*pcr6PMe
z*fq?ncGPT&DfX^v0(caH-cvr8oJ)K{mg|M$wM##oo&X#77zK#N03~t=iPJN_(V>wX
znv^1AJ~2o9&fhr6Ag`A!L(=r7yr`(qdhvedt8R$(4vI>#rhR>bN>#kJTK!Uyc+?Q>
zZ(-LHl>LCeq7vmcXLASdzcL1hBb8k*yhfkr%Vvgu=VAL9$0_WIAS6uE`QNCJJ4ws<
zCWnWJ2I6Wc)x&4;nuK|sl^t498i?jnV^Vx093lYGeAPz00M}EuYBgy0G<z;<RbL^A
zBLQ~2>y+#LU@BudlS?|Mq@yML3tBtlkyBY1oO*7H`9pH>kD_`j+8xWZ_c0<MjbYYM
z$TPN1Kk1}2%b}1bj5J2pw2M^@mhPtjA^Ty0b&B9nx$5g{`J<$K<*_Z(ck<WeRUg|7
zPZHh~nDe>6hz@X}NHo&3uZmIlh~Ri1lucyLy1uT5)Ky5%oX%dj<1jh>#VCQ8&pYnq
zWfkCDNH8l^2G%+dyCASfY?1tL?~(?oGISfwsAKBM2CYp?KTn^-4l$TN(Z5t!`!Z5J
z3y=8O$h33-aCerg)fLb8)K&SCW#P3pW@c2J0yATsvDdt*RyiOc!W5e<ZgvxE7-!|b
zsPhb%EbC}(u@BLFw$m39I<W^p{3#?vvX`$o0~EJk#R#-#Sl?I5giiOy?plQR5ydf+
zagVj5=P{GxTCEU&F3Eu7b<Hr@tn?li0I*W}*|)KL4oi;(o4)kp`nB;<j%<YoKjZWM
z`ir=RM4kd`<K*!-H5Vs6M3Eu8PF7}241*ddrl_<FY^6GFI99hUQ9PA*OuL&Lic}C|
z!};%{WDc2MmoeDc&(tZ%HE-9ELG?@lcoHXmU-j|hFuS_QA?x`8*6^!`RF|v6?FGqU
z{esxT(7ka-twVN*MI>Al-fQ`ozXR~0QfgS@Fl@`MZsiLTHD=Tqv-F|uISLv6f3^Q=
z->*MnH+_FW_$A@%wvj*EN1V<j5_2JNRJt-O>P7U50JlJ2fsa4JvjbrJ>cUpw(?@Jf
z0+P(&cM{Nr(#de@W`Xt;GSYPTgcf;zFex0_^3}vhBh5&O8zcc~6I7Uw1Z}av0rf(u
z!4|f$5?^|5l%{sYOeNWa4DNv2=QZengUQo96Xaz&QM8X@8nV7_tV6BzV(ovZWDSr(
zVuSS<XsPWrgGPtidHMQg>ubyGOn^_VTZoH#Q=jK@#NcYFeb$uuJ`@ACKh$dl^>y4=
zG(jwT<vu7DK^7gtg!@D!A+RC{6-*)l5nrZ`zv!{lX$uddn{a@GY4MZX!{d9U<H$<H
zE0j=keijAvUnP*;fqG>IrUpFJLRgr~uefG+!-4oE(~6it@^|}y&*SitAvq&!k-gSr
z_U-VSpOjI0G#`0C9ij<IRc;3xypt&kI(w19W@LC_+VQKxO|U0#NXIx);0`IGv3V&s
zk(wou%6Vazi<>PC6f(2q{MwbT^MIBM*zJS$BD6O$yAPEPzd>PL5;;Tg5rImNypW}N
z+2O_$#c-GVAToK_`=|@M>3&C6(*>BKMBG;-B+DIF{6kyH(WHdEcis>o7bZ}-TZI@$
zW;yB8h-ogOC5Ld5Dd93&b^#4#D4p3PlG7K2|D|h#e(&|SuJsLRd|#PCApfKD-DL2d
zG$ix`n^z+?V^fYeQedd3R3ppjhveFu#94b&HBJyH%pV$g^?uiiS_fygr82pt_cHpq
za}#jjkH11fv#FRwLhf8)5^A%xnHp)_C2KG-)^jVJw(-bC>m=FSjkk=_4bnjE33{4A
zMnx5Yt+q09c$VaidCvZ%(}J_%H#c{pq$q(ztEh1C3b;IC&93A+<eJ5i6!g6Yn>_`@
zPkCE=!(hs&9cL`nRClZ?*+tah1m;ph7SQl5h7Ciurygg%=~78%T>U*ScnC1lZFztT
zK%9*_Tkl?TM&?V$**na}b0umI+hbu1B&{S$F$2m<ArG6JdMg0pR?{BH^QZ06TJcm?
zmHqFW^N7P~6D^$=Q@A*|()YH>beZQms!TB%0ag<S7J(;e{{tO#eKPuGSCto{Sk{qS
z2OIaQIGybBG^fqrXj38x^pAHnB6h*An@5DukUGZ`BjPK)&4#->hesA5bAz?>ey0|p
zRQbzWY80IZEHSc<KLKA<j~{Z|{W2f7KTKj(%$63=Y_?1*`rVqgN0{AVVGa8Rgr=lG
zJ-7Ma6?#A5SZNojaClDI?|u^wm!cd@=}TgMx++pH>IORuDi2<m>f>rpP{U{CaxSof
zEhY<NT^)3o7wvLe&hcBt7(ZQ;W4slYRuR!Xc}gr|No>ApFd2%b2^)EUd%uOHpw5sT
z=ky|DGS90<ttI47I9+VEe`oqg+`kPxIYUSMt}mD&A0#j_aHQ1gQLrgr)Mgfyh?o7C
zDKmZ_3Z0{`F)83?ns|VGl2F5hrF3-B{0E*5*6RV!HSb?3t1}BQ9WQMX-TFXoeP<l@
z0#Fy7Pa~zKJR<>(jZm<`^_#u6kI2sZY9vuL-~GzHb+y-`*6Qm`90t*f@}=C$uczH?
ze_QrOba5!nhIN~Q0t54fe67bgofR0_(Lhd9bMd|9$<Q({lVU6v6m$yj)z7Z7JBX>7
z5-9{f9Rh2YO*6=@0mEkJ7wbX#&lRS6t2okbrEqTuf&h44FgfBX_5E%~ZXlK-)1+3K
znX4w6wLgl*goaO=0bZrM8Fma9unfQIih~Rq#Y1c|)NcZecvOcAD%>q;f`!yTu9cqS
z0vgWmY@isW!k<b@?{V2*^{A-9;)v5-sFRW23}0rm0~RPJTPOZkRCe9r+wEQZmz}`y
zWpc+;BwyfcP?@uFurmOJ(iZnU1Uv`fZ4ev@$DZU{GjXrJNIw#n37>{wR5%H4v$-Cs
zKMc}+JSM#)A;%AC?J>I<O1j~nZNN#{bwQYV7XVx}i*4pe&!k$yobP?mKDQ6^hw-%+
zdA2ozyezVQrp2S{p)y70z~v>UU!Mu^Jg5`p2v>p<+K6cHik6RjVMtz9Ma!PS5o0eY
zXM46xe@OWu7X%8q_uO@(F2es_dT5k!$pip&!|p97Fq~Sg&@jRaECOu<w&5|F1H9S7
zm*sfW8ski9_(X=*wgF$-S-=&5hXVM$-?ESjXSJP&%Qoc{sPw^~Tpg)p_9Y5X^MzY4
z7Gq-um)K55sVc_TmXtnKVB|Z88MlR)TG7gI5wb#ukCR_M|A$IeVA6A+0TZ7%rF7e%
zwFB>ciT#`Jb9&Fedxh$Xt?=?&xPc@n<-P(x@Hgoc_XM)vvIW)kI)PfW?{DanZ)nsT
zPpD;EF@~lTOqu3%fWiE2Uw@%g0Va4;Sp1&+sL#KkTarEIcYY5T;gHWi1tkvpvA1>)
zhAWFB@mSXAr>@c&<3c>sA8YJBn9Mq%$SDtE2g!9?7I4x}xrB&4+Hclu*GXqjZULI9
z>@jOpyT*Xd#SN#%<sZl^5|)~#L2>?@Tg?2gwFs`WW{-G3l+I>Q>5UZY;TbR0)J8#>
zPT+CfNmSL2Va2}G;i$9WY3KEaQ|T2pDF#f85EV4P&iJ8a$(ccg`UKpz`KCQ*;<feI
z!@4AWVJCZj2Zf&3HAlYN;!nEaCvrpGr+g_S41?$uqdY^iUrJ4_u4TX<YqtnQ3_YDH
z%yFFY&Hrx0@%;KKCe!fyD;;XMAaS$TA9l+@w&|^Sa~JB@{_XkGZ*RUwdX}RjC(C3X
zH-2PjNW1@S|0TIq_4w1@&iwago>=jacess3>~DTwQn(9`=(`7K?zwVH_FEVeu=SAJ
z>)S1veED;QEbXjwlr`Q@P!GM$Z}+)CCf8a*4mBINW_xN_&QPkM8O-tD>`z>tT^glB
zW&{0oD>~Wx>>^;oLviB?)2qL3E$#jcp%>H+N8A+b8E7w3>Yjet)z>ExFLJW?<ufKq
zEA1JT6Q5k`p2DQTSVP_^n&5ElG(}r-{5A_v4`<<@)RwR!^1bBI)@D%qYREZ>>74Rd
zXlEvkLh^y?5EgixVlOYejFY0Vv~#iWiTeX0GMyIiwv!EwauQ-ty;SRr$+fQ7omQHk
zBJ~S+{%lyu@%?=+EUb*kP<F<0jvLoM-I?%MBH-Bz#ysp-E77<9F^{TZ@=Pa#cC;$H
zV?UUI=$Fk!NPb{x!&d7-*?;8_7U*}Vah%q4i(#^B`@-<k&oEHGkuTa~W`|XD8?Bp#
zxmNwmheYtKy-xq0T=YaTz?67a=Rp>99{h`PrJ|08ie~e(2204qa`l6c($C4G`s8Ou
z8?j{Ph2?y+Pd>{z02nxB9E1U{FT5u!<!4Pjr0Qkl6YWuXohH)(LE-6a?pHgFMU~nb
zx5}Ml#VOn*rL?{MAvH85pSr5C{V6muOc*6wSCsoL5j06MkVZgS3*Ho7PUJ3iCc3Xn
zly~FhKMIH+RO%naW_iLN3Cgv%^lN$Q&6@<L*-4{2{%^5x8w;o)83=S0&ZhlVHvu^E
z$3HIqxZNtQ^a9BMz+a`N@1umXryg{kR2SID$T5an-Pug*K3EOf!9cFVe%WLw@iUBG
zrDgZ;&Ieb*T6XRQMQSQ5ns0N{zYO`(Lk!iLy;X|}tI=Pf@dV+EbR=`MDo-rkr|SOf
zBl=ltCV-kOGgj?SSZx@jSG$ZOKGC?Y;!f7q^(QALO=OLG(r#gfq^DpnCwhN&(5`E#
zhYvPXlp!?pAv;M4rW`HP+~4o4-vs20ZJ;v`<Bs$sx2a+k-5|n;Ce&y;gkY%AS&aei
zUKPLy;0KgEjP3`~6L^|U&VDI$_-vq%UuEK)y75sl`5@-KXhse#K2WTrf)>OG%b3AZ
zRl=c=qrkQx{G!-3nB)BCQx40s(L!~mOZh1_KBqugu-|1Qv%|izuHKHG+zwt27^lHk
zj)m1bEY!UCEet;eYBsd2=yYt9Uu+f^`NB!`DLVRAkK(aeZ6APl-qHI4HUKb?b>{Bq
zD8C?&TZC^aji<WxMh#rj;%*jJ5HhZs`Sg;hSufsK5bf`drdgb1dcbCfVwesa_Wp1p
z*`NA4%qr_zrBX&e=_nYN<E_%_j2UZ3kWHZ_z4qgR)}kd4hACr)ZStVYUdZu_?R11R
zw70Gfm2T1U{vm!M+!)I^rGL!HM}{t-2MG1rpLT~$5U(vehsS6^bApyoxd6b9H2=gs
zh`%e`!HyV@(;14U^4|-TVau4E`vx=}w@b`#mIIZ!Z+oe>UM2UR`Hs4(1;Ba6B=HDH
zn9%+3^6f{{rmWUF1j3KV2r3YAI%FI~roi7Q)1n4E6a{5Qrl0`H$l4`%vpB}SZ_2?j
z3vtL$ure-;0}d;|(C!~t4<s{laB=@TeUAr-WmhnvZ(bRs`1VI~klr_r$b$S0k^Er4
zG<Tf!ot(UEm>2MuG)+4EuzmW%QBnMRiI@v|k_RB|IL&|Ecl|oGB`=#@4nTn!J4`a}
zxz7RBajJ32LhT(Ez|s`FoN&y{)IHwZriaz)prK_5^`}86V78(&c6#T8Qnzw`^O-D7
z*S(Ly(bCOr_mUmxVr;pXxZFmeB*B@Cv_DU&lIzwF5qL8S?fL^Io;#uisJ|Yr_EJX&
z<mk05E2H|r#QgN;WLb;mj&%CyYrU|_SZ`b&=0o|tlnBOZ&~{>cMcN@5gA&Sn!Wj#j
z$YY@0a${<?(ctJ({WzJCj)CA9r8VT24Xl5>%}(&oCE)4TU~mfqZ3T6M3C!<<Sn7c`
zr(obGQAVM@^2p1_S)~)n<>D33_x_v#PojKAQyo79m^uN2WVBpitW7Zf4Rwy>l+U_&
z8ZHk%4-0CUO-DUZkx1@!SSw16EQx?&7Fd<?L79#UbTJ6$OScmPE@FPn%&2$}XlCks
z><P;hvKfo#-oJ@L26PObC7YHgJXYNwo{vjW7FBIrATzG37(WCs!W}R`Z`qZlcsyIV
zkK;07P@5)|L-xlp!^s0olPNit)n5VNleCQGck-BHPdVp!*3SVWgVSjI@WbCv3+k4h
z8;21TBQVZkor3)s_(fEXW!RtAqn9ZW5?SBjd_*rb=1d_-f$~-tNoxDF`aaIzN!IKL
zf0BUX?F<3N@GDWTByqBscVb`SCoJVU9L|+PC?u%DVL~Xd{(K<!6-71B-?{n2*t9n`
zf=0ZtM^?|vE>pxgF>f^48w+fa4{~1{vd2N7!e3}$ku7#rNZ>AES?ihU^WJ*22Yj6O
z9Gd)}Gg=n@>AVZU5R5Em;tykyp`W%-@UrUe-@X?m521@`-99m;4**_rMdy!%DlnZV
zhgj?xgvy2EmbyZ|bXlw~_dMkOeF@##eV9*j5GEjp=RuR(1E*vH!6Xh;U|igW-Fdiw
zYs!<EN)(m@85i9DSVKwZ{dZ-rTNvYx=8xamuwc0bA_eAS+UxxE(b=3<`zXo`J1fEv
zYmjCEw0@P;#-So6t=`Cd1ZH}}?3nwOx-aDg4>~^`a`gRxzkM8d@mKth`^WC83Rc#9
z{tb6xB)rwb63>#D%IdURZWFQoQnNoH0mXTOC1|KmiYA(fh6HODYM#5qNcoIjHbw!4
z6=eA6{l)l*V4cDCK!1D>((mksTgA=JHn_V4T5JVGG1j<>HF)2F)7a&YTGq%U<(K$l
z$C?gSSrN>n!=kZP4928i2$J@p7`xKHFv5>zLfJn{g%gX1=l*D;z6tmgMb#oJzFmw7
zeJTBo&NFPnE9|H^Jl9F393@!2we9i*K?&G|Vn-0~*$p~=f)w@EcT&G}me8I!P2nKF
z1v;w_>aXp+6T`{d0ygy@h-Qnve-VLj^k7nljq~>CVCO?9ZldS*iQ>qu*+PSt<BA90
z2)fr<|1xV8`~{aiS^=~I28G-;E~^a>c!A(5fzI}C&EQ4n1MR7rgFtZS#yf_T;gN)~
zg003^9a0YpE_@($1B;?#x8Ez6&kE~i6Tg_7z50JWae{I>af!N2`x^|xd723@X#$!S
zl=;Snd9tZwGo++!J%N%(aZ_z}TysB<(&tF+;!uGZw0zC)kUtL#gMzBz^|xhRSmcqb
zKCww@ab+E7DP2f-Ngu)fky5cGDMRwOYzx+05)(A)SZn%;>tYLJ@NfN8FXDv#9>pd=
z7L@{IzOJk%YZv+JNQ7oP-}4VEpWIuCp((%M^VvmDh9!YI;1H>4YwUS@$lTU%yD>F}
zZ)4c5tEDAyO$JO_Z&?RB_<#Z?r&}G>pkM(MDbQ?2FwufG9oD4aiu>6CoZa{0dz|Jo
z5eq{OVVbtQiP_qjlx}(1{tWJ4&4f<UMKVyyV<Jn6F*BP5S|;<iVo-EV@_;TtQ2h5U
zNRt8M1%Q!T57#bmBHs_-!$NGt#VI(AHkAoGIx);Z2`m1yjJ)6BE}aT;%|&>cH@VU`
zeP#wVXs;yN&&b^6+t}GQz{}Ig{92K<$ShC54xC$JMJcLm6Z7Pqo_e)#h2#3Zqkstd
z^mhe2;4%BC+SS?uZ1S$6%%8wcg?UH)(7r6CB+dI#M{%!9)njyQ%H?R{Rjv8sQ^q@&
z8^5LqRCjHPKNzQIDeFg>Wb^^lARxGSA8rZFl&OSl>kRbzdH_y`&^1TaG$}G@ZueM+
zC;CTAt>3JiR>!n16=dr?XXuO_R2formJiCf--L*pn9sPtWw!m?4m>K!YP=8Vt@7oK
z^ef%?=bL{nTxN!d*M10&%4w_(EsW2Ql8_p3p#$XihX!YaQmwLg$ufHd5=M+Ih4Zwa
zBH;BuIru!QqbMZc^_Zh9QTP?<>DqrrTH1!wJ|}o1>+rf``A{1r>UcV&*HSb=$i@q(
z+mfpc-1P2<<nRFavx6uL$aiU|!~kycKM$1woJ)};2uTQr?W#+<avibfZzA|!cu%Ix
zxzKzBCY4{rfG>_t+w3G9XpVWM7??l8<H7x??%c#CR-}grVW1U*go-nKn?z!4->%B0
zE?@(mNZ<{Fy<j>R0+8KSOKvEE>qR;{enQtw{Hx*v6<hi6=*Zn*1EhA?t{)f3-P6nH
z^E_m|BlG%3RV(WsTD|^R-uy!#=ugqP%TRz@Nx?|v#7donemkEc-~UY8SS{~hZ0pEC
z$NK~LB;`ovB9UKYmb9=4f{qkmd3^5Q(r;3y+D`*=k?O<}F$C)9e>4KZyXJ3KZkAaU
z{`W$=F-M{f%NYPFO}bn^%fd26aC_J;?F7*~WcV^sIZw`;{i?w5!rS3O0bGOqcQGKT
zE{>p0Ad%yUE8Td{4*3<rxH~MFY=*rr`OV&FplTj#P4xTDWRGmZsM8s^(w{AW&kVVK
zliJEPXc_~je{SE;1OqXw#_p!GDu)&nPQfLoj`#O3P6b_e8V0<Bf=s4{4-TwNmVZ!O
z3@rL_^vP;|daej}RzHhAdwt(e*QW;au`^_VQ0^^f@O}QI;jXndLHR2$YKl0y@)8Zt
zr43Gjdr@~1qY*Qs4>Ik2(NQZ`^0J(V-*4=R7p4E4Ovy05dFtG=(bGCF&L(r(BGiH<
z@HF0#F3|Ws(tiGX$7Rv_zo#vkp&iit@CLMtg3cw+qz9T6cXy-CUU}V)Ym@-+Ho>32
za35nV)`Z^y>EwXZl=`RM4--<Z2ol1W{ovMuCj%Bg<S4XICFzfE2+Oj>>67P?5`No6
zCIX8g+}tkCP-<|HXI&zXhI$3_*kC5(k%z~$-y6_uCFp(9SEr#b=y?<79k<eaH^}wv
zm9`+cG!(HJeXCHa2if(W#n%Y4v5{jAQ`@58<5wu~i&f;#C&n@}6b&h~T}&iT#n9$e
z7WJ-suA=MBf-YK_%SMj&+iaDxJgXBZQ<~wmB=W@h7>&#JBjXY$!@$)WgF_isz4gh$
zJ2;>mh=cC6$Jae&29^67KUIHpY5shvDCjH=tykZ+UKc1s*_b*ud->|G7g;AzbS#yt
z{^E5!$I!7SEgT+#`cjPy(5(AxrUL`IBTqWt0EBtAKeGb=26NsL{*AWW?sN7BP0_$&
zArcKkD6|}LV}q7RAB@S$$|S>q9*+?O3<uR_^O9HD&GLqhnu3f5VApb1Xk|SsiTjOf
z29fCR6lxDcLNpO9k$Xj&`RxChoKAd~H>cP3W|kg<1s-Z6P=ly3WV1>tnsi*V_Y^N*
zbg|-xJ1OW37IA<)y`!9>{<6aWBN6x5c(t+4@_&H%MeB^&`9ummH<?uV43oA;2n5vw
zlSFlZQyPg4%VVIyzp^{E=Nl5rS6OIjaeUG&m-t2Bd+4}Q+!S{UosJ+nH-fXV>RVcn
z?Vd0buAiF$F9ws#n+zO5KNN_}5NXTaJsHJ9xC5UgsRr~0a}lj*-$L{+1A&BqkpdQt
zV+}l~e_--n#$QQqOhVkD*yLa!EggIKuY=d}++kFLm=iR&Xr)^Y@5?{e&3>n0!pjGo
ztC##^K+3l#$slE=a~Sw?^f&h(87Ldl2#KmqcJh7?+xFK!V1hfHc~7BKT2Sj-uGHi4
zrgM))fBr&)Oc}TH;c&`9>i+g+82hJT+~ezhO+wJvgFjPAs`2~B#%S^Yz+-*G>8n^&
z$Cu0FlQyGXp!jR{_d&aHHhcJ1%L@_(4eJs@=Tw=V3a5v*iqqL2ql?oh+eJ6F@d3<E
z&{^KkB|Fgw@jTREWDr<>BEG`MJKiM3BISGKh-Z(M;=e+oy>t~H|A-I`pp_i{LYC7*
zX0i17n&O+tFRCZN!A@;|980#4W%R@uui9V8HZlslHEDI{hAXJ%ATelhSE2&7TDs=q
znF=ALobS6~A$L+nlEhnS{C@utuE{<zoi5u8B?1vzSbE*kO$)&ay)Y{gANrBRobwrY
z<moN#awrbQ{a%7z#OvH<AZnQ?==luxW&SiS=-xi2kSXf^P&sEhslw3Pe%Vt9OiLv<
zcZUM@ls3dal_r+5j_yi)=RX5qbKiuOj#k4ZCydV1$!EQ&5Hp&xStDGFD&b<i?Q>pc
zvVf9QGSHtEIkKKHFclq4^x2=spaghsiii|DXD@m_yib}z^SC34P=^8YEO8frCu%$7
z)EBVkX@1*0{aP>F>XSitZye;kwo-w3_psteE1?(=-8u3#C_yOyGq%3!xj$U06`KWk
z_5i{+C{9yGwvmz$8r&Eu$K!inbz9*f;PjiYG`?p6;P!c_0f-orrREU^D($-#D@Py_
z1Gmw2PtzBJu1pWVRF@PYHCQDiF{2d{(kCYlI~E;TNWsFZd_vJhW6y<q2eR|M@s#`$
zV8FrzW#x0Ty{H`A!OnE+@XjwkHgP0(;`Z?czU6IwcIHtIK+k$F@zCcQFSmghH>%|p
zzc_L|p#9JtilOTCljo_bY~KiY|HvohzNm4}SaMQ}tWL#z{0I8)Bxaov%CuShYo{|F
zAb|UKI@{%l&2IbIj@`6PAR(GB4zc0qESGMj*_o<e0h1Y2n&+g&dkW~{2eLq0!@VQ(
zZy@+(dKgeyG)Dhu|LYK>4`=KP>*iK}u-?LLlRb`wMHom00Va(Y&u!a|%stHlOwNP$
zfP%eK&_Grcp<agbmITxODh8_vAY=BJ+@>CjtKk=%U$wG=i9S$CD<vpTR;;_nBG{k}
zn5Dm@Q)S`-X<@QnVaS0-+2?MYED9xf!V=yiZaiO8EQ*zL8RW{%`X`hD)?S=hB|02S
zPQw3V>8!)5T%WHGD$*fR25{&uNoheqLXd8d?vidn8WAO=yF{c@DUniAx;9<X-Sy7?
z{@#Djb?}0{pXa`3)~xjz7>)-Vgg-45<%By^t9k2j(U#Id*OoqF6=}ne-}vfctGt$A
zzw=dX-9abB_+ak}BUpm^os}T;(d~JU`hlgv1n15DwAvSWDvN7dp%Cp^F&E>Fc&cV(
zc5*k@XV&}T?Dog9e*!OJc$}XNj*JZ%4bG9g-<%eZG7QRAw+4_!dt0#3VvYO5G3J(t
z2l{`KUe1sx+JQZ#clQCF)2As2-s0?xgbnyUEcjRQ>&UzGpQtgel0g$lixsb!>kquT
z51LyiEi7zv^Sqg9g)4drC@mmR)c94ZP_vP-8u9|FcXQO5c_^FzYnVaL)q165?F9Dm
z3nhg4*ryGEG~5I#0U9ECjOFF5Vs)WG)9d7hW36DkpY;BUHSB8_`ezqtK_7Dv;?&eR
z5X*H5ymdk$>{{S7hSuP@N~tS_Zo|So+Oz%XKbpxe+KQ}N-eh?yEt>;dJYWW3LW%!a
z<-G9Yj}RBu>v%B48VLAbVXB5!n=fRpgd6a{e9ZhT|Nc!s*3CW0coagweOXQ?7NVh^
z&SrELH8`81Ua>V-#x=?M>})IG#4Sb409wPO&%m=MeyZPORs!{ldLbu;fuFGdR(o7l
zF=Q>IzT$?|?p$i0&pK&LWzaLyntSWZ=thE2^0!%DA9|>ZZgkskXCAXZj0~?tA&7oA
zZPf`<i*+p|k#J2zC<`e7dyUxsPHjoV_iNFszlecUXPMuW_FcE)6MdGg0_;Bx>gdS!
z>u(vE;T+h80b`7&Gs<zG>eF52?B0|&YAJW)<oDLF`mP%*J!Uy^0B&1uBvR}Gk5BGD
zD`<dU{~Ipe6uK9yI}C%8MRYz35I~!Kum8-Vf-AYBce?LG>)1UBu-%*%=)30Q*u-aZ
zB!Tl-;mn;sa+Ff4coZcFBe%S6qI8Z!ZvC6hw5OC>)sGDL{okxz4i_23foV?VTV^{4
zi!4cZ2@SvHlmpX2hwJO-rP9rLK9s_)oldjFHRN><ITCXnn;svoX+?r_qM-0?B6#w8
zx^ezv6@JpkIan@57jiwci){j|=yP_XlYx|3*LU8FTm9fgnW(s+*0gU9yFfhgrUwaI
zZj4X;{*v3lPDl<;#x}}NL^(OxKoQl0$fG@646I(pk0FcZ&5SKruN^I)?|jQUFTul(
zUZ@(XyW=qL7X=ohxw@SvQ~3JP7IYVn$;s`qv$G-IFHt<urYO4iJ7ype^36llFV&wU
z=I4k<k##TSs2-@$Ips!%;#X>JT7diFxpcef1FmjUiHr}`|Linb3`En1=MdEl;L7L*
zvp95q&*3?+jF0$R_QK#t)8adSPSn1QrnO1}5{h>3S@C_1M-)mkn#YRa2^Fi(&|k`5
zt+!iF)gjzVIYwnF!RM~<7;y_8(>-jAdC=2gF!T<Ns@Sov_t2{^^`}MgR(j`#airMK
zuq)-{jGzT&-rvSM_HU}r4k)1(T*L_=yUPd(w<OVQT&Q?nI^E>-*9IdGXgiNb5Wqg{
zChZXW1Wz2Ro_i48qv(6!d;S*F{5sGsLicS6!tYu1CmYY>NtWZB%3D&=&3$9|FElfC
zUt4;_?X5`p{u{2?R}vW$%Ncdrp1c`K;DbwM?{<p6vq(6b4~nDt)ai5W-Wtju9s4gC
z7-Kll>FnMNfY4~jYyo%F`o5%w-P86gZuiAeDjG=<!J01$a2Ti#XYf94(-}WJs=|y8
z`*6`<iHbeR!!DCIc*bZK%VOWCU$G}3SzwpvgK1N;nI-`O2pmkyf4XZC4JQ$?2^80}
zrN=aRkt^8=c5E>CPyMf538MuDcu94y86+^grbbOizHOu6q|}Uf?{zF{2Za9WZJ}?$
znuCX*eDX-w8eCLuJJ-oX@{iZAM-!Z3|IPKJNot1Y?%)wQ>a*FBe|ccC{=P6e{6#Aa
zP!p<IrZaVDb{!SI3OaR>&^p$++m0%2o$}9I_vqC`gT8m4m^|lHTQ`ZTL6hc3)z`pu
z`*j_0_-tDMkb!*Bk*`g588Mu>r|2R3|GqP<mWYH`)}&2w&4-5jMvhUECgd=}ZC0nR
zzdt?efN-LyUm<R`ISVh&NQBI=J!N`$R=_rCGvqjsL12#y(YGw#`MKloN?=HqDNUmg
zl*zY^S8$W8-d!PyhS~0wNpRWl!~x!HRvrI1@N{(8VQaEzehj|9#i-qem9%u9fT-hF
zHJI`Cv=EiHP~^f}oC)a>Ca)4ugrNNEjbLLiA)Vu-YO8AQ8rxCDsvY<w;X)m2qvruq
z=Tsj%&o|}Eq60%#cgpV-9<5w8+uz&Tp7ka@sT$!o%m<~nj8TlS4~5I>g?v7aRnnY|
zAXr<F_=Xu1eiZX4`_9CsmxDp$kZyWi-1>meR*W>E^~rIt#X=^D8dJ(mg2C(~2mgJw
z2Xbmv5nvv8b3LgFpXWUX^p}NdA9*V=Av-ZJI{K@7ijo#HPi0{`^gp$XO=CanEgS7-
z9Q`S^_I&ZaJBIzM&#2RM)y;wS;v_f-AcmzPQ9YEHkIymn$amUx8Jb6S?JN=qxv;~`
zBAEQkM+5tcnTNisFMHgcTr8$!t`=m!G5oC10i*&ddFe$5QH*-LS(HrIoD9;iIZCpw
z9<vLr??-BgsoIwTsmMYXZi9TUzp~Fw6&@sk(@Qx|0Z=Ka7wVliaQ=P1c2CvrK5;%e
zj-+IM`#=G=`TS<K3Z<s_czOq1WgY#qc;X!OI<u*lwpdF`o5z@O_QF)Kt`JUco`?4i
z)>()BrPV*)>CaTu40e4N6Y=kG5p6KBK3cDug(u_J`1Q*NQ_WTpQoHLLYn*=K+&^lI
zJ3#9lFS~lYS}K;M5F3sJ37jq3Emv{L(n9jaRna1;(bykBZ3$`y(x8!}e?(OKZyWN$
zRI0Td8V4w{sYI4u^EJIk4qNYP84@hD)hMELim7zs%Y61J44U-n4`PJy^}laz{<5<m
zg?s!DI7ilrvHUbqS(V4J7G2JvcRa7}qWmdX;QU<b#xT6jk<Cd}4^%{C&(nLY-Ig`v
z{(#?IXvg=KHfPIr=tdG$)ZiwUwfG``jhOg5Iw&{5`xh<(&>^~)ve+j47x+Nvd9yy)
zTIUbqk;!`5XfrDYoA>Bpm4i+?TRn%<`5;Znb_&wlDzI7q3v4EJSAT=yXolp^LC3|0
z6&GY0z`f7`We5b_o=<<}3&)y?)F=dJ1fAkVX_@DV)GwOskF?=X8F4$SO?jAfPIlqg
zuvFGUKW(OoA!UJ!kpC<D&^3*cT6;1)OSLwl>Nng2$T4<IUK)^mUq;kYk{1(p^Jfqw
zMVx#PQVMuT6EqaSB2heg8RLD7jd71ySK%J~zyAbntHVBMT<s<@aLk4on2g@+x<~!!
zi>~i?L936P`?zbWztR3kX>oGOZ+8;kzu9kUf+%(f20Fv=!yj2U5jWN^sVxeqV}hrT
z#{J3)g>>Mo*)MqC3t!4}5C|5|5no>6b9oN1gPnoilu+7KtqPvk+}{Rn;;D~&yAMf`
zzavz;On#zD9AZd(Se>)%XSXJ;@ndRK-h~^NAOg>qST{6aig4QN7`pSPQhld=?_3f7
zRCBruEST13E^dtXl7JRJdle;@N1{QD9_i#<+nFYrRFt#pv61rcI^%L@+hc4J<fD5>
z?|A8`pU4SpH&er<NbgcyAJ#*69>(6P5nkkOa9Opw(9x)wZ516XVKl<Etz+Yw$i%Ma
ze=eV{3Q;7du{UR<2ONhX@W-`51T?6%soS{ybF6Vxs}=Ab)tw-R=d_v?iGdxhu>XBW
z^bPUdv-<Xp+CW~Jh*;9J#jEWtA^-UeO4c9(<2C=zE3xZi@;tBhsR~selidwyw2U!c
zQQ_X<_K5<QAu5;m&TMyhEK#4Hr^4RdMKS>^Ze$0Zl4o(%kS>ssC<$;HKiRKIa9CPb
zrTiHY;$SrVmF((#Bm1#bFF<&OHEHlH^1o;fHo+T8Qji%vye@nXYQcc=zhcL3XP&_i
zm$c;O6u0^S-&%wr<Tk_3bCefYPRFGN?ZP#pMBB>5XvVYz$KYp+*L`Vr>Vh69salio
z9sWxrby_t2Y9sU$be!OC%4eI9PiK})s$ZX5{r=NblnT1m&(jG2a^nFA(E|L<p!bnY
z*J3L5vuFCR3Nj^oE9a?HI!@oi8h+f8Zo-cmbmvj9AoEUVZlA<9F}2A?ar5Pe+4BNh
zy}ywK7Au_x3I02e+RnlH;bCj3MJu(xd}<VRB5ARmU0TpuHDA0~VFhuzK1xn5y=yWl
zibt$9@MZkK(<M*%wP^!Mx}hv9G`t)B7|Zx%s6p=BZQq>q?X?i2L$+0bU|8%3u;MAy
zCzOp5a9-vbnN%#O+4|-YaKG@`r^}}c%#N&D&Wtnekg|rUTK2Oe<8i*C7tM;*YmlvV
zi|X$DZcWFaA_g}nw2@iGdAV?f+3F(_&^s+U<*~?#i?dxb=;ExNe1f3aKg8FHLW!)f
zI~BWB-?s6Ie-=i%kDVN2K98YgPWY2SpV2qIHOnW4A@KqSzG@fB^3Q^9m7bRiFA;oW
zL)-^zGc`I9^pnAfbBU*(pPJ72<RCn@;ByET+RtBX28Kqivz(i1MsG+H0Zr0eq~?8b
zr$c`4fLAmPM`I$x{Lr?b^j!4v=8{QsY=8XhVv_4VP3V%VZNWqSPBaOOM{{vH#q+bc
zGEA>b?&!K1%I#J17V5{y6s+v;<SLr}v3w7ZL6cSc0`aI%w&Q-=^(J!;G;oH)CG)9%
z|J4sX!(KS?a=C|1Y@74z477hir^G4u?Br3{9&gdGH8YKcV8=Oe6OqL5NZr%JpUwpt
z2o@(FMh#B-(93t7m{Tv1W=T+xk4q^@I0P&%8^}*jeVci<m#6xfKB9=V@*WqgnQ7k7
zQ+amccH`bU@q@vse;=FF0p5@I5>(|jN3YMvM<lX7Itn_*LqPEI@9U4CgR)jyz<*2R
zFS!jq)0;!nu{_&1;FIUvrbv@bkzQUPY5~s^GL%SJBFpqAY=-O87`o#08PDtWf><tJ
zLH5nmu;v575B3j&0Zm}|?-u+OZ;tw~g&Qy7E&m0$6tLYjMk=2^uZIg2Kcm?W^%bLo
z8`6dEth54WvQyWqDQTu06hZPjx%tVQ@v4nm5*XlGZeJWaNEFR}8F4Ny3fk=W_o?!)
z0*TbF(nCf+_bEkf@h(c*@@Bx6H%Ho@p?a(N?Z+FIU#Kk}UT41U`=|F@zmB(Q%LNl4
z%?_8m(q-UI5C}caXqj28#lVdEdp8g>T9moegwn~~hYj&~9V<8eZ7Ix@;7fW5cn=ky
zTSEh*#*Qam4Gm>sOp>}rsfv{N9EM&W43q}EOeQ=}&)ti0`~|@m*`w<lhKLX4v`rR)
zl#NNMVg@K_`{xVIJ&#9@Q(XIFBfu8|N=?4)kSwC%<{`l+i7+OKsZxyX4+Tr;w4yPz
z!P?rU$9ES3=?LYJ->w50NkVcbWU56?`29{24>U9213TWiMuHCd4E^OgSGahcy{n2S
zP8K#n)-Q4=w8yN6c!C55R1?yB5}#o%6(#Cc{Wj5BkH%c!x-1GOKO~Orb7;eC;ph=J
zqt4I3Etb2G0OkRd+TEXUyO$RVg*+|7BYZ%7HB_1f2(dVpbo40c-JSDceRFv4ZhC2l
zLCP@&e;x{J`XCcKx&G{j>K>FkJgr}gX55LzN%?p731TfciuVxhDNYB*f!J^o$Q{pN
z<KX2S{8a~kehc`IE6n+fhJJ|x6zedB_ks}IzsMM$>!!Ktmr+K0z)kFYsD?$GT(znC
z0NFirXS3o?*owTU*I3RNVh|M6*(<3v&DjIH&O7BS3dV1O9Ty27F}Vy&zV(nz@LkvK
z?;7pKu@RMsFdKBei!*JlaW21KcVARUaw(6M;uinKkwjQSOq+_h^ZKA<#3*^|$3ora
zFY`%y5NY7Da>O<;&}svlP-a0<FXS7o#{3_&KYDLA)l%diPyKI^v@NdP8%fjwkJjS<
z#e$P~a_QI;NV!c+r-+o8S%Y3ftb*v<)BT%!O@ySuALSOd1>%tjo6dy|tz-+qM@6jd
z=6)}B%O|+Y?b_egCCoK;I+Ty@_3-=qukW|Q%fV8cnr(LV#L0vf9K#PsSSfZcg7!~M
z2v_bkKI-)4vvH}fWBUilC$p{vky$&MCvsAiKk{N|rvMl5dS(CWPYnje3{MQmV@SRh
zRn|S=PLikmFcEM$-&Ufel)Sb=X)|pj6Y%WjATNUJ4%uoN<c!_^o2O2E?)HuR#5Z;_
z&!xJ$ni|z)`0I~ZWt1t#jQXZ-v{#kJ>nv|!Kni`4-=61lQy#Fc7J#{NQ|#XT{Zz!i
z*`WkVBUN_=J={hANzK(iNoRZ1aj}vBUBVtC&w5q->D!D~GICtK`da}IP{`x2=fO}q
z5SpCUV-EPI>fBDugt2>T2^HywyGXrDGY&kH-#wfD-`B#|x6@uS^6sgTScuI<!9tY}
ziYIF?p-U5&Zwgq-GdNhkKq*=XIqJg5lgq#7Nn}6`6Y%EEUr&)9yqx%x5o>{aI94Z!
zLqZrY{^L=xO66hmB2h6vAA6d(Tnvmqu<6J509kNtMQKv)aoTeq3=oVe7Cgt^95(~H
ztuS0#1I<~3IF!?FkHY6l5~yoo4``+q-gt7UhEypyMH+4*luIUP;oKm9^m3CfXjrOF
zQ0;=In<l?5?~ip2XoB#;A?i+;2Rf0>9mwgI&A&KZS^AVjnoiT;Xqef#{pWS^c8VR+
zHx3CquZhw3C!YOE*E61{f?)EZS3vdITZ!y{KeSfqDA8HhT);NddK}$!Y^aHnW2IL4
zJbQT`Os0SVo^QpkX}>R*licp<Xv|kBQ%jSUMS&bJRdq@{bl3f9b8^CTL&A0kv~ly!
zfQ@^ET%y%1d|;~@-sxW$?6ts2tP7W8D~QQ6HI(jip#`KT+|=LFr$uHQkeMyzrcAN7
z5>%6Ie%E;xPV8GM@Q6)90^;j3R_?iEt;>*;Ir~&7Bvp96D5pYZ!V$<T&YBcU!9^zp
zH8|Fq%rQ%_+|q?|==0{YoO^Ix_Dnv#`3s6I(f3f$fXaM^Fl)9OpO@nCKHcKEPU&l8
zTTkc^znNM?MG|RmsCdu)%nzYOqYybkia9^$906m7fg8P)7s=nvmzD`uc%)(*xme+=
zkz8!E@fGI4^O3XhaPsLr<@O)5K4XG5GhZ>D8mXi`a!-vuz#X={qe{9nKhKT7nQ$oo
zNsiElyuVE=jyBcFsey{aNbKjxd~)W?2du6aa$#%Wu*vWBX3usiCS*;0ov@%Q^t95<
zbSu@Lj6d@M@-GDxn+J^K(n3p&Cu2n3Z_VO>LZHesH?Ik0u`^8KdjeG57g*B(AIDWE
z<9ltTDI%pWFWNrE@DU2~O3#z_anf&=n4hhyk~-us{_~t;cps5>AV69ovfjE#>hL73
z`I<-edY4<KdinB*0WOndaK^xbj=7G2wY~k0^rp+OOV5fcNBdY>0T@Wvf6!wn@RWWy
z%J|_x6(7(!NEH=gT*#3G6os+5D7UKY^nm`r?g(``%1Wu|^Qq`-H$zo2?A7k{f7RtB
z(~lts_uDF+sr_6|Qzm7=H|6hlb>fbmiIA(AZe)t9Qj@a4(z6F<&zw#ziU3+fX%qPg
zgV8&8LasT(80e5NoOs<Eha(u9|IV~Of=_CKRP^X(t3h0?O3>|59n&bTi>D{geq_?<
z_}@$;M@NA?yEneI|91PuF<NS_G%;}_P1A<7N1wLiFDO`i2@Nj~SbfCN{P@MI>Rui=
zBU^rpMlop$D+Lz6fsZ|c`yM}NUtrXwtB;G3rh%i69onZkFpfHXF4l?1_IH0(bryiX
z*nQ`8$~WRu76r1K9Nl9D;6W#3aT?zKPd7WN>zCjDx9;bCk6o7)e_#N!b{sfeUC9$7
zWQayl(6*hHHXG}9z(pEf(kF;~az0;5O3-MPFPrjYvJn(5Ewf~8sx5ZBF&c4Aka*sI
z|L@-4wfe!m1N?uZH4lemsucIAx4_|6X|3Vjfot2*pdN};0FbY$$&wB|`}iL~SFhRW
z9u#P``|&~Vhm;$MQ<UByrRsk9i6vvL{*WpwExz`5e~f|O{B8VA1k=qnJ*NQQ_a*_F
zATnA+8pR(Ik-Zm=1pnOT&+m8OTYmnxSX~__z`CaNM|u$)Omh%Dqwkq*3B+`BYJTdv
zHt$YcsD<tO+EJjEq9NaIaeuhjTJrPzj-{I45+2WqRK@Ja*r!|q=Nlgy8y{{}3kg$4
zj8V+iyl2apAx`;`g9(Bzq_(f8Jm(`DN|9QIWk6eeoAcgo)`#iMI>cwc>lUUN!-<N-
z3QuK9IOB35kpVEgkaFi*=ktA<(cUYDKlb|2q#iz~-5aWwk7_fbB1&<sB6^gQtxnIt
zA(rdNtij5pNsUjgtCs!z9*f+>5)y)3+9YiF<LrIYA|D}fG1OAIgb$4rT=M09CUMlf
zjgJ|=m1(b@GVDK28=exHzgYK8toUJE3&fUS`n%MXlmxfT-qN8@;5!qny(BX}ZIQ!_
zyA*IXfBEb4WLoCim63I1Ldg!}m<u7;tfZh$hOh@Z|A!Z!1d|^|g}_vDpuu3xV=B3T
zLqx7dL9~6bcosweQ#lKX4Ev>#cK?ZHzj2GvfVU3JA2J9@uCJ=dK@pKZY$WPRNt3*Z
zdJQq1y_r{!PYmbVq*hJ;SVavbeTi<Nqzsd*)i(G+XoHJbDt2=I$4Zp#wOO;P)5%Tr
zl{dmHMX{}@B`t=e#cb%u?*1=ou!u)z1+YW?$!)*w#>#?c{(HE%&g1lwV&UL@;gF58
zA19c}7XYsXri<2x?Qp<;NI@@vwl6p%|B@VD<OSRgE|<wst>u1oaryZbEo|*G`6p1V
zy9Sck*c)G=n3UH%F}e1bs%yTPc6<*U^LAAfY#o1HRD|E|L$I_9uy6DS;I^)fXA^2a
z>1c=pGs#e~UA4f52Be2l+zc+akAQY@T+gJ7<UHiI+?29DIb_it`*2b6s$TAw`<&=0
z8dx0l)_vY6NDOVzLOkWb32afoVhvO&L}D?BiugZQf3P6ja;pNzKLqBA&%Rt*(K-^!
znNiKJ|HY{-_Rbz&Z%2t7C70bNlf^QNTsG=%_I)s!g9`AL)35uhAR~|ZekOp1Qw7`r
z%*MnRnd15^E6sb-)+s#!E%Q<+z|1?6rx37XquWYRd%gO##lygmq~JGeIwBk*u)i}u
z?h2#KP+lg0U7daB-D0ICN5)KinoNo6l-nqOcb7h>$Z&z~{~AewslelL&eT7WI{8|P
zVqpA*?)cpG>&yC%WwCO{=~E-WTqQQNjDzp0r<Ku1n*O<WIzpECzbOl+yB?0-PH<>4
z^XrPdqeGFSzFtR!4zpY2YZ0oPC;4N{V&%&zDz&bgzQX5all5$oXIv+Jax*2ku?Z!#
z3kx34>LwBhC92R6zE_fD!e-cfuF{nUyGI#s<eV!H!LEI;mPH;J1B6!*X0R=THy>h4
z&mAX^b-N(ZqB?BILkFfi<k_AtLIR0CGG`6)woVQVm<&LX7zG1UBuubk{mV@lT-txZ
zBmt!bL47ePyX>IeJrc8bDXj*c;ta7b)*>WghN1SEO%=K?0fr}@8!}F+pv={Ou8^4h
zH8K<m$Rj4h!ShEUXf2n#p^2gmt_`lss=!q3cw_?I_2qox**(b{D}sbBATk6(pTGFT
zp^dqCwut!VdwaQOzmyu2?%R2hN9~ng>4M=OyHB)M1jRcv)dI~X)Skr^ghYhUj6DMC
z1**d=U5}vy3<o~3d0(*vO8i3d7O<2h7gt_)Z?N5*1gJ={qe+M~UXJ_i>Lfq-0k<SF
z;|@&=w8zFL$y8Ne3_FO#|MQhXRa@hMNEh_4d2vzd$=)Pm0;%D>?3Wmt1qCs*QtNAL
zhg)NXZuP2oCX$AA9z#tx6&yq*FywzV<wgQT9ps_Y?M**aCcT&8mIftGctHTAz>#mK
zSEp}6bD`srrKU)F%#W?P>gOCm#SgE=v{okhRnXc#f4^|qr40f-*z<glWjjE~dHLSh
z&h@y_5BZJL@QlSXclsDl8@vo0%*fw&f^6a<i1)w7+uqs!AS9)wW`t>uVg$7;va10;
zRCp~xF6{$_AuceF>9KQ|LsN!C4(JpcjeUAFT$3gRVkn4J|G{+IztAzg$*UXL{OH;V
zMAfs+!A)k#69^t(E^ikhy;4@A_V*=X@7ERZAXT}>^@dGqBPzU_9uqS)NP=j%ZZi54
z&75kOyRIa!1mbOCan(|pTc22>|6%p<F=j3x<Mo=3Qek}4+5PC0Nhn&c^Q^FwpSVrS
z+ApSjahqmr$K2XuKm690(+-jnuyBo3)^`?*0M+!1$)khKi&M1uh56Y3Y-)6hTkQX0
z!5m-vNglU$<)we^)>i#n+CN-S|4bekb^x9BrrBrS9l!$|;0b`pi5DXgp-lgD!Si=}
zz*pF<@5D{7I6BG=u6GF*zg4ajBt-(dpMQ~x9U|5ksIG5ol^r`Kdn%I)SqSm(G)pUK
zrT{nmBlr!gS%|~8ss^=Od7}zY<N`Yg?yJWpHeL-IU$hiWq7WMQTac8JFUx6>8Vapu
zkQT6unVvIPse-XcX39?<Zq00%tAU0Rw-7HWTA<u}8r;0q{3(|kWFs!;n(~f$9PWgj
zi*<+nl`gpirLU$9Z@fGETEin9zyanWY_m9~;^32v&Y1<Nh(P#(Ph3|wIb&El78V0P
zXv7c_*ND)vXM@E1p|PafU4-21xi1?g&nc4PS}ZKW0BW86ise8yrs{LFm5ChRW2O4*
zZjZHd=Nd;y3WE~&<m5x&!uaR_PmwAX%=NJ~9Xk9F6x)J6yagO$a@9OfK!W>*@QT3d
zWk7cDfckh>RDciFAF`^foa#IPC?K5Q`>MXAt-o>-;eiUajzSj6#ydAZ2-F7Zo{ldp
z{G-Q5#7Rl~c+5R3qNE`NPR;qUD#KJ2SHKXvK`>;);WQvLMCytC_HlH?<Vh^+YB#Z1
znCYauN?qIe+%hoYOR#>AA%)$Un(AvY@>4YBDK|I7dF)5B0Kfa!i^Iz@n_5pXZu_Qr
zG_A~tqROg#uBwjHO0+PYjA_(TfBb$rlX!Mo>Rlq^IHVArCaoTWQ*Cb#JNsFJCm0=f
zY@EZOWj>fUOZl*R{l$Tm)c#DzcPCHo-J5%vZa{2v@V?Q^9Kawc5&E?d)rq1dFz|se
z*aXK5U<?2X;duBeTSNWo=1YZrX3?t`N2m|!sj7s&`*1|MLw5|##@zKNzs&nGVna*$
z7}`eL=`!1_kder=q;eMy_>heOXjZj(#PLa)`D+5lZ^#*`(8+azx$B4e^^T+`d_?ID
z@0(q{0s983EYj1z)SZSK>Z{pgg_bIk@0;3>{<CcD*a_cB0!*P~{A+ja1cUy*k)RQk
z;BX(24uV%l|J1YIV{TaoTQa4r(XpWwuq4A}aIT-xqqJ;fZ7U8Q%~;wED%GvD;fF7|
z#EX?HUl^45-AdP;vDDl0#t#01q6aX2*p7iz1OQm~z#-#j#v2{km@m5RA&na$3x2Ll
z(2KVI63!<Y1l9YC3I8K7=a4;dwW%oPHblng;j~9i6cQ8G`XO&ucg;oqBCBB{2^d8b
zV+j>@XX0PuSx!~!#X%GIn|caki4IT$P14y~umddV5RbbSw;i=9zKB^pi?73>wX|52
z@kxKU0_T(($IVeW$mpL0D93XE4x*{Ozq=oGasnQv!cn(#P5%v2f>`rntF#y~mw*V@
zKc|ODWIUeSa(N4(>eeeIg1><-&YCgG%icR)AoLP0VZ)ba$26C>C7-@IzLsfMOyG;^
z=Jsw!Wv)s51rwj9dJ#R~)q*GR*_E{v<cc9&8wtoY$Tk;D1uN>*c_Siv88-aXu19SU
zkI`G=b}pR0+qr&vE%fu7=?cFBGlXt(l*i6YS*9~Sag*PZcs<sj(%|zJdhGr7;&>FS
zJjSD&ncDSnJwoGBDEDS+H99I^U)kcN|Bx&amABMx=7E?y4Ir?AvgUYGE;P2&v;=^q
z47|sh5Xet)7%lo%Vx~N^^6$6kOKaQ=dk(3hVI<ZIB%|DQk^jGOW#PI8uyIh?<`fh~
zK{pTo)~hahQm_dH5%6q2p{(rbu+<q~E!xs=*n2-&Q@P-C1f}>wZtM7+#^5uwpw}8-
z4H2Ur;1L9Fo@IlH4UAsSl&p_yV6cfG(Z;|F{|P4iDCu|b@KFfw{X9#OL74K`dDzhV
z5NhjXoh{W#RY2NV;lX4Cr1^c7Zyfit_2Es#hpJ3S12<bcE%z*SeRac&`H7tRqjC8D
zCx3UVGru;{)s4wL6W<%J-3ZX%;QFexm_hL!1Rv+QC30BT@VngkWbv$IrMgRQNRZA-
zt5QI{Ye~uFXD~?mpDVH9Wn1_+;jPzrK%MT2#2VD}G7sjt5@W5(r&)?o-Kqb8cn#6H
zjPucROsbWv#bzzVrBtk?vR)RZyj2ub{4w88TvX-YkN>UfDga|wOB*Ik-yeIC<?A5P
zf`ijIvsB)t3d_!p4O~d_FkEduf8sm0War{q9e!D-m`wI4Gfxe<P{c4*$a@0y^gNUk
zs1kSb-iJ?1T~wvl0te(4MajhqH1&wraJ2wXDgwS77$)QZ@Q0|XPq7I!0zdbi@ox_H
z{{E3At(u+j<yCeXT<q^4Ek>|fB5Jf#uDjG98wD+oud^tj1gFCh6+P=XT>Df22!sNj
z5^S~2W`FK3j^5JvFO4(_XWh};?w4GomK!r;mL~?I;&^npW(Gem@K#BH+HuzyQ66$)
z`x~rR3-SLBfWlyTRSu2+Eo1l8Ta4P1+qVkiVI%;f9)5L>%OSig5*@i<>AX=!+BWK_
zl*QehtK%iQ)#+ky4yy-Hsf|>i2{c%l2FAmu4H0qBhf==@>qFO-2S<aG{|2(5c*!e$
zWOq68>Tu)4V8HV?bKh1X!AS>QP*^Ro|3WtR);U|G{NBgTL7dLKD<)Oe6aG@9Rtsdv
zNKMaH*fP*ruxetIelJS8tm~R6#a;jBD?@xcJ36pn0BP?2V?sp$W|h@+$wA*s_V6EJ
zvC4e{vWTKpbTH33njZYFi@OcYI8mwC2Ut@9j@x-~yPTp28yn{;oz#qc(JA>4337A4
zBmq9Pa#Ij5Y!8ZJ(TvrCA4i(#q6j3Yg7UGp6%#(n(&aOjc^I+CDxkgT)Jc{(w(7|&
zIcn1W%1F0&gl}|_Wh1a%-vblNMeP@ZDP$^nwtig+J@UX>00%pFdXdDY>A%aKcVOiE
z?sP&2MwK(_upSt}gAq*<a217G%!UzYQFMpx9`Mj2{0Uc`A5c-TjlIy}cA6&oFS<o7
zTaCkRP0~^zv%%X%qatRnbAPrE`;hEA2{Z}N^c)X4Q1QLC0>K}OBLaPLHXPhT<bZN3
z<ud~e@@7S@!6E_?6w(Asofz6cN{7>RXTPf%mi+7b6q*zn07ZnrFg>;TIEY)5LbWN#
zwSk#Dlt1b{%j#zW_8AZ9BRB4a^dtQu^r8Lx8Qx(Nx6&H-QLx7-^5r*})k<H5J&|$_
zcoXG}g5pIl`}~<&=b52jSD`$9`J*aA@G{cMMvP+gx?l<h#>rX^^Lf~Mr;t6B|Li6d
zFsk0Qf@I|v3c`ueWCH3v^3?#aPctW8<KA(B^u>c4_vw2JPCd5fb{5}W$?gOZc!h6$
zj9WMjMz<Yz`v5zN=NCzlnBaCSo!8*GA(mgwYkf;>O%=2+gMAez<=tH}<+eUK<&LTX
zMEbfq8<p(+2YoPP!ky}SbM>=oXGKNsx2{1F{r$yXtFBemj`crim6<HI#hh-8D?GOs
zdJ`uUH^n^BsqSC}KSzRKCbH;9x=G&;9YepSRkJ2gRFmGpTFC`6nCf?1KjdlAq&K1B
zN);)Ur*d%Yivm{l_iICR@=XMlQcvFG)BKl?&EI)zwg#*FvZwuC#z*@Y+Lw2GP@h<E
z<~yuT;%q64P57q10^8x-Tw(oVLD$fOu?&I-M`i@Euy(~fNdW^d3LJm?kv|#2!dEy{
zdHAbYNb_a4Q~*`ZG9u-*w!#fE(U}$DsU>EO)hmS#yP`<pe=3xJ2Dfovr>kt_=oDvZ
z!FwMZ1o0xPuZVX9x0D>jZry4H+3*wa*0}wy-Kyf`z#_Kax=7jVAvwY~1SUEpt^8Z!
zz)|EZ$P};2+ld}09Nkc#i_r1}0+z1)=zhkjk`2L4*zYfvUA$?JNrTLdP2Y|c6Ys@X
z`*&p@$smye(gV)OG^Am4hd)H?pm=QqUnm`lEDebz(L)(T4P8^ZUde}c+Ne707prax
znc(8HGQI>(bAd)^lCb+;2V<IO3q+9!cZ{o0<fkYI2hSl6zt_;TS+M+7jyc6me;6jr
z(x@KTwG9)^jg*G^MdDF|+rX|d<K~y|hah+vXMf_{0dfN9bWr2G_|pYRGCp=iGDG)!
z89Ez_6(|SjTmC!;28R|{)PcT#x_$!c5|T2S;2?q=4wvn|cw^cv4wF_UU15Icm*M6B
zi_T%>)|eO?a9dzlORP$BrGAY+b|SelW;D}{lJ@sm%xj{L6#v0uG@P}8x5zwrGBpO5
z6$?yy+eW51>bzFbYSZ+a9>m4Lgf;@$NkHWM8qe{fdLJ#eu{}Wz>3Yiz7JBMTS0b09
zb4w<QK-CfEp!gMrg(z+f_}TwH`(OA~$Ec`~nl-;A0H@Ps-pR_=ec(A^ax3V;y$=j|
z$DMK8eP0bGM&QaKC&a~eyl@+w_L5c$tA})a*Fdu2@>J_-{~<u|H|uJyU%WODtj50m
zWB9D3g-Wt~MvYDPEl6OeSuOdUzg?AqLAK#j<+wfkg}Q$3JGbzFNqd*QSYUoNGw})(
z7lxv44iGYL{kaS+kPP4&NtCt~5{!yoG3HvbS>_CDFv}DUKLktq$CWrSl#6N7+i!6I
zba(oJJKwWVLLLG(UHLi1?djE$gx@aDAa>z5gQp>sDjk!{ao0@+E(`vbHVLLTf2L3b
zHXBo4qwO}6lL~kp&+rqAr`&pqB<TVi<*GWiqQ<R9i_I!l=w>Z+pZZ8X{7<8RjH!L)
zW8Z0t7}^fFxJ}c*;dr1yo@=-&2xjkp^+$@TSRkLkOq+OB^Q%hX-Z5s|&{Q<oQsBh^
z$~Lfk4vVHF3VR_T2pEXJ1-d4;1Xy8@@w~CWA%&h5>L<Jx1N**6DDTLi(JXaV7p3^-
zsQKnYB<?w&#R!V}au}XOEcNfbT*f!nV}`6>=Q4`Vc^AvGRa_iAlRuqbG~U9-k*1={
zXYVhNQ7GVTUX}Th+pz)isYtadQ)(AXkRcLbd@(AW5=)?_krq796C$uSe}R>p+I0Cs
z&~?q|>&2~(hZ(>Y=mg%L-c{l`oAX^0*tzmyihN*D<pmaN^;FStvJX_N@IdCD2hTu>
z(%R&JUKR`x!ub_$8kDFQht3+^@sa&w+)ZpwBsqrU!pW-cZ6?ypD7D=sslDYWn12xT
zy1!hP$<6=%UEb)6n}X6AdPdYd=q#~EDOR>=YD?`^|M;Ye_ZnLhv(OgCu}1L(_V*DW
zwi3e1zBJsjBHCpKxd`l(F-3y}9eva3bF|->kRT=C3crF&87N;!Y7%e<kGkWV!NjYY
zcf|9xCh59x-(Bm!coKWOmmlMstDY4aO!~MuaCVzsjzuo2W*kKFXH-~kFu{zgvpq!#
z!n`-r?<vYTES<VkZVmlpz1n%M*H%U=3jvSA&`ys={MP_X7CUsfjDzRq!BuUhdl>CK
zw76KF2o^2Z7YNak(AJlt`ClJ#TTF5M!4`1cqsV8UxnFR}3ovk4wV~Ybk*<Y`OZw$)
zUwv6Bs+V^xhsqwq<WuE{y5GQ8p`AqapP&ye%wbo-6##nH-tN8`G>bTZ2%)<nm&sGH
zxs7VdDSR;!@}_EFY;5Qn0Nfqba!8WYoiu;gECX-RO}pC;0FW@wo@sx-(@)O#ec_D4
z$6)qUl_xndtQjd}!tSOomLu->y{@;2IBYS0f{aD)=SRg?@q93m`k%R&q2+}(7M0iS
z^zw+;@bw|(CH|co-}+<;3}jPsW*&<SIKoL6N)|awl|`@|lp<+@*xdPjn-O=OROQy+
zAwZUX%gkeqRw4tPwq65&1-3PQF>o;JpQJh@uidL(ycr@gpWUCmhy(FeU?T(Osesd7
z9L|%eqAYt4=H6L-ZyVp3|Aew!tn}MY-w&O>fY|SUA9V>MBDHFi)a-?`h95Entix=g
z<|=Ugzv?2QtT#!QklakJm$7hq-MTAKv;`}OXk~x>7T+nR1sKg+$kcuUr22Y(zAP}e
zhDvUp98=uXwD(R9DjW@NrG37m3z6-#=Ho}f!PBf%T%+LPk5=rJFVhy`VSU0b=y7Pv
zuu}T;@yT?UK@|9EBqI>WoUi4vl0t+aEM}ufOO6DpywmIjdPs%Of$Zqhw3eZ6{9p-4
zs>lOvO2VOD2UNnRBj_KcX)ti`khcU}gCK5&MQA{gN<m8Z7^sOOhrC>T;#9Jbk6xVF
zg&01yjrn~DiTQW&BH<*}YGME@W+ii|H}@!&_FZCn8f82!PO5&s9rG~+b+p`J9{_=z
z-z-`^cqMM0Q|Dwmfh$OA<HPQ*w(T>TVZzP7mPD1i<IF-_xyt{?2SsXU32sWC(PF|}
zkDS?Oy%gg|hk#O`Rw;na?>8&vFdA`NA+h;(7?a(@jed8i2F**ym7sW8P~nCHchIiw
z;a%6Fo8HXmp#9BiP0K$ai-F+ctVBh<^->a%T0-xE&>Al$B=m(O;r!m3r(yBP1<QEd
zqOh*Jp^x+4N~KYZsTJwQ(w77l3v&=^QV$E?#5{CpgFslBX=}L$uPJxFlFagxv!`!X
z(gk($G_Yi?1B$MiI3AX5#>p2F^z%N<I-)7~y93-;0Iv(NlcvL|AYacwXUX!Es!OdU
znP#yoZGyM-IW&l#CsbjP{ONB_6%?2X`J%c8GNrk!)tMDmaKs#ns5k|1!0-!h)x{&x
zw-7k-Tw9NVFanaY+aeb9_b&zylj)DXkD}x&kUzHn9PlmaB_K3zpB*Hs3K+T}K|!o&
z<zUl<?JbdDkGjx9@G?aTU*)?(SKu?C5I8TM%I^!y<Z#vvGLBz%-6%hk2Q3y($K-fj
zMZaWDAss>i%c{7_#Sx--+1N$(3ZJvb&(o2lxNOa?b3K`&VTd-j0f>e@U9`b5JSnts
z=HfmX?D2c1r9}j6plWH!=g)a!XtQ3uU?j#mJgj<Qkr@gAUg^7DiOCPMq49?TSbFV{
zi=Vq4BF#@L%zv_0$~FYvDnxo?);M3E6<9?f_t0H*n){EPC)Dr%t&>x)AHxaPoWL@6
z+Xo)*^P~OXx~)wK4^twJJ`RO@t?xQ@XLo<ON?(KoNiU&D1kv)gj9lvl;?=Att841k
zNl6vDtVGvdl{lDM?*SzY=~Dg87*BiWrKmJ&EeBPWFsf-+5*v*SU$jE?^MVY(I!b_q
z2EQa^bxu{gwZmS?n4Rf7SF9Pla`(KrQ&j=KUcHPt9ZMr}TX38uC%b(%JEYU5LM?KC
z2Ugh*Vcu1m*l%BjImqBp4gGDwF)9d45ycmPCj<QQ$edKaLF4k7jROeaj}p3>*dW}S
z>oR}<>wCrz^+=Usy#SAJ7XLO#cYhgm6O5t##O)oDY@d*j9#2|4dxxcdKjyW--&?OF
z6h!7qii`eEm5_L<GLmg*!Zt%vdA`Rc-RY4T(*L2$16%W1CE1u=$v(;{?(d26!d!w2
zWarRWL92{<KjZ%6hb^!~bKLXvE{tY77Na-<sEpbgEU@QR3swdPMkAClD$M{S#|0B*
zlQb3<n0&-=xNE#I?><#k1@^;pr_IaTHQP<U!H}z5II^iD613vlyxk$<yC#avkhU!~
z`g)mPYmTAU@>4QSIQsP1Pac*WZm0>hmz)7w+67#B4Pbp?h4rgZ*=!WY=!9>+0r?CY
z+FBfbp1;d2drzB9`FB(yHuDHZ3}`eT4KQneUuLM?T6-l<7@;bD3~*fm_AMn?wBfhb
z?nYjtQ)$*u7ff3&xw!s!`(S4ZUrWR5<kKU5HS4G<=aJVsyo|A%&gpgk6k6F27Jd!i
zuO~)iI`@zqo_mTc`ijE34|AVy%C<=&_j@<uv$dxO)tF+%l(G72F>oW7R^E*v#qg99
z1E-TuMG(KgXk^gFZO~*UnsF_f9M`M6R%782UXwyou|DsLYrz1E#)H9}7R~$S{!t<N
zE65M)N#aA%{l@0eyH6?Sc7lWL?1KZJNxaZUQ4O1*DAE7U2|?g|*<wPgCqgZx_`{v1
z^)`?DAnSQ`qj~S12ZpRp$dfm0rS5mT$geN{)1FMbW9@FM9L{z_-z1KlC_#>*`?G!m
z^nyq7w{8(48CS??ER^UGBvtQaKdFuHhnmgCNPy1@meSG45~A&!Mrt0S80-1G(WD?u
zNWE$fZd?O46aR8*?K-Ezdt8;C^OGbeITKZf_n~9;DDwZwVd%_nJN^qr*Wv%$n=~*v
zH8j;<1;!WH8hajjp#ZC;8!5-XO!?4=Q~ku_NAEn<G@f8e_#8-M9hLu9rSLVdX>c~r
zovGiASogXUXMimjN6NcMqLsQTh*ofJy>9DZ7R*9xr;tC{Ab|y~IjYX83S}A0Hc<Rw
z3o)o<ng05!j!d=e9n>7m&m;?oA(^^dK0j_BeVydA7w1Nrd8x5x&o-2x@}UA)+H#93
zzFJ0H`*{$-9OOdrc_oBpc5&F2od-9*w|%aI*Kwp}eJh_IFZzYhy-paWN*`Od#TgYZ
zT;enbgS$nJ?_sn~@w5MQ=2z-u41V}Y_6R=6W)OT4)+DaiTLiF@p>ioR2wEwCusnn?
z^M?8i*y|7vi@capvZ;vBl~7@&&S@#RlMovtE6G7mjwK)S>%m7zDad_az;s7HpCj%0
zt^EPw3h(9eWLey4gEH;fpLlXF^;v1?8)_DLK(Z|zhLElXbsql8XbdFa!kuq;`USZr
zJ8vI8kQpNR_~UA9g!=;D3B~s}u!RFqN_WHuZjm6KJ~>Dk{WC4&FbHTe*f<-LJ=&~<
ziG^7JeRb)yu-MXFi68=!I-zUMyk{Ftky9`ju?Yq9Bxf_}^rYI$qYwV)UETGT(_DfO
zhPmzdqV6fre35ht!cJjujcf}aH~QP%zm<}>&@jbi`#pv6RX%i43KWAeqqyB2)ms3a
zgt`vd)q#WNU~*C9KPhaHwD-bOWS5>o?d#E7S99SP)&TuPl(M}nuO)(|OT~DV3s3e|
zA@(LDV^06#?*9)6!esakt_>0T>(%)Ut(D(#w*J3l`b8%E_wg9SR*u6XuCphfkp=3;
zNc{LZU}>K-Pmz9RaP{S5H9@vKT)zCi>+g~@0IN>xJ%)Fc_2#Dki!1Chp&;m&%6E}k
zLObE_{HppCZQW1eg<U`$dOa}SU97c(VR4j?@+4bzv2Q`CuyNYr1H9CR1KF3x_n4@*
z1}t;cJ;n{=g!WIQj7FTI2(mR{1vKW@xO?63w+>gQM9p?q(OF@aE7dLA5ZBWKVuRAK
z#@gHCY}9LhLH~N-fPXDD<A*cc1+X{4kC#{*1i~-s<(#}M#QD%fFr}+lfyM*~FQNpF
z-C5Z*@CVJfM^Au<8MaKFT0IM;#O0ut9GICOV2%$UP|-(hDbX+0!N82N89a~E;1WOx
zaI&d$mw+7MR$j=h!=jVdhgpXiUqR$hbLyv$dMES8XYD_MH;pH_>lm*D4DnJw;{qan
z=>iRligC=Au$H2(P21~@X|MRJ7aplO7CyQBzk07X7;5Y@QxvBV96YSQ{>vkG(FYKe
zh68I0<xmUG&csIu7&UwMK(GKXxJ(il|3?@kk7X-@HBOZ~r)<X*^?#g6cl!Ma-_xeu
zf2c^)MqEKQBOkFq69g;pW<HEnYEry-<Q(_B;iVq%YyfMsS0CE2cij17{gta}oE#JF
z`NDdcj=`^Mls8p>-l~jgNDQIn3hX#Q#|frm>F&E`mMZn8*<B*^p%7Dpj=Sera%|U&
za(6A$mo5kbsl)yah&38_goEO7-vi=FYkaD|?mD7xAx^;XuMF_C-->YByBX}Z53zZr
z_FNpV`*bo<ux`f(JQ*1Hs?yglj0Ii{xsCJb1js)<VNcvsMMe`~(1YBXvJk97udWDy
z6NsFezyp(B?@{SIlF8B4oyP~FkK?u|bw}?D45PF;ERm$CTy9=&-v?xm&CNj#Bz03%
zh+lsJYt`!!V2SsU2<!Rjo%QkSg7MAfP#&QEzH(dFrF4dR8qI&g0@<|V(54GQU$NwU
zn-E1qCcUZ-Zat&gTNDIpo6MhI8gNyq+<!=sMG*5Zq-ZhUp*p_FvNQ=bAZrnX0S`ZW
zqlmjI-<$>I;R$?8>3~Klbn_Orxxij+DTrQ-HyIP475B!v7@X8L!4if~kgd9Ic07=Q
zHh@A}s^}O9p|N$|`tO{-#?_7K%Qytpu2nwgU{llqkH6Ew+#%(4$xYBp2gp8j;WZH1
z<DGr<MZIBV6(rJNBuWd;q@lcj(82tUmxUx->gox6bj<BE<Y^gRWjk$lsmF%%YegQ_
z^Eq!NB?xVtdfoSBKD#B&Vsw9KrEU8K{?GmOh3q>QnFe5riD1jm9jW(XLOkB5KmLy@
zh~o8aBJ}-67hlc8!{tyuO^0SqA>ZXs3kP~WP{a)VFw%yjI@Q>08@h}>;5#tc#ilT|
z>RiQGt^j&Lzj)Bg>)B7Wz^I4(p%mfi&9v>jsB6GxWP=4RbXVW>-|vQf39Io<Fc9)N
zHXt^>wTTG&%kyh`ZD6(L^Nvc^NEM-<Uw}n=^D$pAna$kgUd?c+2G0cP1LEQZaK(YE
z^zWcb3`9DSXr}!IN!V(2n}BXud14o_c0!IF&=em2X9|p{L+Rw!W#(?tMEA1^0z0Nj
z9ARaN#C3y)_S0IETToVZTug&u3n#TIX`imRP=6u>g5`vb(?uFK$O#J21|Y2)D}eq6
zt_ncvQJqkD3rA~6=?^{+Uu$qZ|2TMi1qjw0EH#NS48?tVu+;c!cdLJ^H&p05GcFUB
zC8n%0#aQv!a{Q)JZd8n(U^UT09qind{}v=RXl1iKvJxLBSF?1|O1^wiF`*O=&?8uf
zj?{mVB!u+`{h16O>u5hdekHYixs9Mbn943&_mII*HU>li%-)z=I;ZPSBKCWS%uB>E
zQ+e$1YMAOotx>fZ*b&;-m(><gU~W+dSX11^ka&VI8r#3&uUJ4}F#i09R(n;oi1%BC
zSZ^S2DAMrgD>U*!^-e2IHZpipw#x`UsI-MoJs=BsYgPB+R7QWUVvrzLe=6JGB-IIZ
z^zrTU2SWweDSQpp22e5*g%@h<i$L*_1lria1Pgv55E;g+ELZQuu-P9Q7(Tc!makdL
zJBd1}JcXapG(^pp_Y4?$$m~eqI~qdClY2JE0~zBO+dXte?7Y$7n%(WwJ!Wr6Gl30m
zeHlA{yc*7~0?S4&Co@Z=J?+PL5x|ptfWk~N-+ulo^gntW>u2pM;R#aEalj5yv-XwP
zJ-L6tD=cs>Vunj7si3ssGZ<%h$;26R6#S)W=eI)HN#*#HU(vZm$w$TAfe2!Em#x8J
zt8o7qu?G2wxYYkJ*^C@!{sB)c3-#4e+0az5|1*dS15qBRPbx5Rg_O2(mAGPsMU-p+
zEZhcY7GzuYPTL=qOnbH<G0{MWbZ%xqMcgB-f5k}d+1RfC`GQg*T~V7hzT?ryRHYtS
z7df6QpXspoa^CE}V_tvgm^Ru}5()^mJ16F@hXroLfA<h0ola)I|Lc&Ot#Nt^j04ir
zI6V~)2$LmFOqI>LC;5D@FouN-;67$l9*@{!L8sRpxJ*?|plm7~b1F$E_(kRjw9XOJ
zUQ`Y8B6$e`RHsioPb<t`X>u68(*CaC*A-tz4+#y-5B0md`^%GQ3AU4-^H;F>sKdIN
z-i?7n{O29SV~zc&>9Xg=`#Gz&rCu8|Ma34SA~j(n``yI-t#cGdR@%)96}2I^1Al34
zc!YovBNB8U$Ok*?k)JRujVrW!XPg}27bZjM;p9WF+(2eLa=De6mQO;(O|^Tvk+Pm0
znP1-~dn5naOD|O>W{tKPlO4K_8TsM1?ScVXdg+(%r?pVLcNzmCU`L?xGp`-;nw~Si
z%%Y;V*n!!7zQNs_n?0LJ+!p5Pp8dM(YgXBH&%^S#FZ}*Ktro6$gSFe4uX<`RhGsSK
zluiou?ul&(*E?S`>OScn#Oc9yYXA#s?x5C<4rcR@HyciwJ3ongH`YbID_L-WLvTy7
zXlq#;k686JBh_hI)LurnNN*Dp#<!3k0cKP8<tE={(xEKxYY^>v5@2{Ztp4*(Fh)sW
z%wrIrhz*-fX>;A-4?N4<4&v;@b>aQ!<KP=#DX$iHaO5ocY=vgjL2B!0>f_U(x!RiE
zG^dsu3*l(RfSXU^n}S4)CJ4*PFQVpATOl)P!eY$MosFRbi4BhEgenzfmO32oDaP|V
z*!(-$gTl_=>vmVL7Gca`66eqmUfmb<>g2mRe311^L5>qc{JUZ-eaPe&2LpVPe6Lp;
zDsKbdqN@|r5s}3I$Ht9s^}y5Pl3F=6J?W%mw<|m75_fm(KXVzyW&R#(PS5$0mfF_z
zQY_qPE<d(Ayd&y=M{23OT=xT&RAs)?94^FiJ1zEcKNcaCP}0N6%agcw8zs(|ec{qS
z;ilLwGl?+gFk6jLB~{=X`L#A*g*3WE3TpF<q9ByFO&{(amt^Rf6%0278edi);f&e}
zy+d3+xsjo{&89#$inkbY`dJe-_xXt5l3{z^LdG+v4;^RjU!!bJGW1C)l(g$wF<s-*
ziaJfSc1S37>`+afR-wIBtoEy1{Boi7LG9K(zPrKbk`M5))=__wl*TAFOWt~0Pqaj$
zg}BSssA;g^`N07tlIlKYS=09da{I@6YQ_4Ojf=v5dfPV|E*@zFZzBmdYQK?O6qBOj
zr99f8URlvuGya?>;KW3<;coCV?`LOb>3d~%H*+n)BK)R}Ib}O4#!qDg5pC5fBEN4e
z3E<Ict6=R$!N@l@7#ka?<ehS8{U%PwMkb*%STXm=8%6ykUaG=d7FM$p34I^DXnc(2
zi7z_z_5Po6@lgo=n$Jbh2M4z|i};X?3G+9HQ95qY-}*4kT7Qzf-r71ss9}Eh@1R=n
z=MXHikpH;rQM4z}!d~_;;{7;zpKHFv)JP|d`PLNcoFgutPpwc{*?aLp=MIBp7~7*u
z_AFl(%|qKAWs}t}uBzyTC@${hZCNjemKD|X=q;J9g&fJ9{U+XhLsVE=@DfY)WK3dn
z`~6c(Iy!W7D*S>eE*_9l?Yb2r{C*JfJPTf4$%))qk3x>m%bfP*{tj`X92iFBzuJ~P
zj=Eo&bLXck%Gm643s$4o_q6fjc<&k!wU`k8%q^haS6~`5dR)3a%8U|_EO%tGkpF)3
zZGcF!aSJ87(^3z)sVS+Ne8osE3*K#OjUr_^;Te+`v|&np5^`Q8Rn}Oo1n?IGq;HAr
zE=f_JM;S&ID9d8_Q$&j{j{cmOA=>ziQ>qcf5#1@Cq<OI5;QZ4JWY07eth+y@bZ9q&
z4{xJY4?2|B=b%2X{ArJ3<~k(hHJWMsh_fvORTV$(>pKEbM)ljTF#e0mqa_cD3Qa^j
z%l4OWrKYekcX<%49*~(C`{V2WT#Sk`dbyU$$ScI1TY5&%j-M$lUv@4zav!I<h&~l5
zXiWbg_)D?$ZTNiWDudU9gphe|zB%*X*f?7wk5e?RU*++CDC;aI6JDllNU<mlh(!yG
zW-E4C7v>c?(WEJr9YSSPer4Epci5+pd({t%{_9ahj7nvZwtZH^qmdM|<n=C2uEF5y
zB+<vV3a>59Ft};bu<i#`#z`pYl6wiRbMg*<8vAzh-h%6pnT`zozt0_58@JKd`!@K|
z(K{0YmghbN#oQns_c-0y_A0;<hKHfu{FF4iecVC95zl<ytgL>)B7AOgr8~TWPY_8J
zj`IHYb$U?1bYr*vs>)xJD$0GotJ#`f{j$Y=C*J*@Yckk98=x#^@8|0GPnr2Q#c6fZ
zfBQD39I`2qD3ANl@oxWDMq!Wr`RFwx0#zsa)+EZnEk=BC`mEn~o;}1yphuI2CDbqE
zCP-|`@X_UzHGjyrR*=?cS_w;f<T$gl8hwoIaW6J7^xoe8<LN7as(ik`FWueU4N8Y}
zBi-F8AR*mdlA^$uknV2jE(s;2y9Md)|1$4yUT2(f#+iHXv-|8h=M(3wu7D?GS9$-<
zjfZ(HtYq;^3~2e5`?cycz1(*l8t{$lzlZ<wSg{<@FK3`de*}(5#uN)z-*2j13roCL
zgQ)P?yE#rnl!zC*R|d|%ZGRvUb@HX8BYcqe`MuRofq)=;wQ7?V2^rWTup(l1YY-99
zl7*3V<#y&Cilul513^Ysz{E<99eRs>D#T2bkTpO<&5XokE5OM5Mw-o5(n%MsF;Ec0
zFb+lR3k$B>&%dKLEVw}q?U|;QIakW$O6%&X{Hz%&eX*07XG<zDq6)(eI0TX$KEG4u
zHXkYDl!!GoN@owKx9x=(c_L2?2>y$%s$r#L*+uVG(ZY1VQ~N{G^Q_TBDCl|U>y5r<
z?Uju`e`ZsdSSf%!+c*AM&jSJlJ+BM8uvYbUL-pCiTI5G>V(w2#V6%fWLIqlHc_qpR
zuRr|h-^dRQE*rUDA;?w)Q4}50Rir`Y?Ck94ikbCi0ym`)zMKGU;8HD9<ZprkEvw_Q
zf5bO}K4*@7tln&fkmMSt7}lWmIJjDQt^wLFgY*rhJZcvs7^+HINK7}+;n6O4TvlAq
zO(gbB+2ziBAo1|m9?5dM{kXM!%1TVe`0v^^F(nNl=zEb}VaN7W;TEhZSzp0$QrB1Y
zx+0WYp=JJt{ru6IVFhz}UXa#81u^nOD8G9(`@1nM8?mw&GSCToR@crIWI>Xc6nv;x
ztF0B>Q&BnS{FY%<y^Yd69oeBpFjmQbqzbfhy=$_)?a}XOyv|c*+O3%XcD0u>y;!a9
zl~RV}8*oIx2KJlHg2M|6O3%8a`ZL3S+WvL&Tsw)G22rka6r0-v?B6@CPcx(SPLh<q
z9_Fg>*3<^z*caCFX5x8cuW+Xn5(_?)CT#eRWPgf}eIITxls&@6#o)jI1%LG+{wOf^
zj#eg|N@LDq{LBFIMb2AdZ&9vEd(O0T7DfRzfV(X#%Q{SswI}FvAh-atYj)u0_0yc`
z9@k;$CI+M^;Xt$W7AB}x)M;2ok2pq?u;hE`B)mbqCbbahea@|~djb`#a~(%d1YW-C
zqaQ-mu|W{|k+4X{t-|~9@RM#6<mY7<eFO+q7-<f+nUnJ4C$PWV&~@mPf0_Qa5OR5r
z+Bii2C}M{9xQSR>{49rdN1$`_wtMD8u(Pbo;Ht$O`5n}*Wbq$)3!3pH^^_Gji2V<(
z$Dehz6_MEtSAO|sqy9zEP(vZ6fBfE&%gi_){IVdBFpd#bjXqloxh*d^?82^y%qV9;
z%EH4<VQjSD2;mak(<Jmd<)EdtF&kU1<oA5aH9TC>C#L>42HeS8knTEe9o0Zx&O&F(
zZf4%uK;<~*zzZrt#-1pQ%Aq}u{pkUez&C{YTX0?Fp<vbGVdnlOjKauYpu{Ar;&avE
zetI75P7c^wUoT-|jqsq$VSs!I*XZh-O7lO|ywFYA99%*CCm2A|hK`U|(k%Z885?;K
zrWUH$zTM0k=73A6$5o!m)#Jv{?ZaRFV>?jC5PPKG7&}{fpAv?={5psc*lL@*<q{HX
z!hCbIM=ijH0NbA(cSUkviTY_($KP!M!^NF7Jv=!HIZ|q%2$Vb$ZTN3^?s0l<@+TZm
zk%hdN*GQs+#{15WM0-aRyQZ#q?@e0$56=BnAk-uOp^AHW8OQM~`t5ULK5E(u!uctZ
zF5*WvjrITRtjwt@del>~Dpd_zi42|aT~1ej$rty_l^a9Altp?VGl%gA4h;>>ryuse
z4ViS)fE6D-?7_-`=<|KLW$i)Q5l0LzEEQ?};SdHMp{I8YazimBcQG396JAWLRkFAT
zYg?#dVcEkwp`OTrPe*>^668}EjE@?Q*=OEDUafi-bGal_Q^k1VOwM<^IEF1(ghore
zA!X6#aXg+=A!LZ{52z6J($qP!JB-e>$L#j%CkV>d<4$MilLBzu*ZzTg$G?=;J3Lz;
zbKP;%&vr@vK;Uv7y}i0S4ktNTTt9~ZFDU(AV!6}+orIA@iD=wvef@6#8YY8;b90_+
zbx!wC5*FDf=PmBb!&z9nGWqPQ#{mYL%^NfZ4(D^9$>(oP-rjcueh_I`UCGa;h~{>0
z^Ws87x0WJyMOw!`wETs9a5=ix_})y&T<WBjP|M3#qH$zN2p{*Ijr}R|_o<yKMN=Pj
z;f?}}c%mwIXlWmMfoSPu1(Ig*1`Ihn2NPRlZ7p|LM!t8;VH)vYG0OEVko;P%N&ZXv
z99n#mV}wFrOSYb{ahd#(TJJJ$pLi#X{V35{c&5~uywzYY%ylZQXvf8)MYk-{dYLc1
zo^7CP@jF^rYh-68ze6zoW7b&@pIjG>xNV=!M|KEhuBjnb+k<VKd)UAi97t<Nt74&~
zn6w@iPGn=>7>7{-l8`6?T_H88fG36^#&QlpD0#5e-$V3KhA>APrV9h_@Ja!Getx^H
zyB=LsJ(=JI2ri_1;WzG9<d1<bC+UH+ki?PK4HJF&fgyfloWprSi2RPE7AkP&@`PoU
zqE?&{KY?#%PU<#%jd?7urw%t$_Avs5uXoL#RV%8f6&V!lvGJA5Wh<Q(-=Uk`ABo<V
zE<l{W4Bp>7?80C3xfa|%&BU^y8PgZ<>@VpNQQzLa<dhnBpm^SFh{b+yES29JFAwH<
z12^`NBdtW6(mn5L?qjW-D+0AJICvjdRVz|MS}LV$CW4q4q51bi+2-D`Fo(4&$p^nd
zTjS4R1C8s5K#j#gP==dce%x#tvWA7z!+=@|mwOQc>C4&APBC(k-LE1i>AI&JA`b^e
z;gBJ?)fS(1Ta_>CT__|fxG|1h^Kk`>^qngKtAQii7e)PET}pRPvtQRbqGWh@e@Zsh
z6KBDZt9eSO#j8sHniXT+kwQsNWpup4f_Te+x0dk7<IsoYfG-(2tSU<A%X1}~?KvA+
zMe7rz@5s8EnObR%tgk2ODtG?K`@FZyh_X#5I$EWJ`I*U7j->jubHB_MT4@GibSTu}
zRTT$n$`gC1A*Q18nMsImAq_~74|a?e1K*(txoXgEvzfl~u@plX-iB|?hpWP__(n9K
z5;^=QD~4X9SqJmUH>4ZR3q11-Tm`%VJL_P`7YauIZ_>d-78AYG<Q?z4^K=+gQi=Y~
zmRidW6<gf3=Fj`z!uMnAQOOatB2Gf_m@v$nf1rT&!GxfHOw}%Zn7#m9$-zUHj`!`h
zUr{;p#z-d|#aPMN8w8a#@t4nsBk`FNCzv?S{|rb+whaW=@7ZinW8cfI_0{2W$$Ab_
zplpv<eor%kyr%bbs=%=3Pr?qc<od7*8c!Hi51en%&^zm;&1<E^kanpC1Fhmn8{9>e
zsMYvkx=A2p6Dw^qfk#q>36)>e@ubX=j+Ya9!Dbt_+!0bG<guJ)7Qo@^!Xz$^=-^p{
z_nZHK^2w`4^J8gY5qFqbYVbl!_>_RM1fq~3+vGwq6ex5;{>$JI*?IcG2@2-3rLLh`
zrgh{n>NLNJNT%$xppq3^o;%;gDvS7?ZM>@qO`V>OZ=<Sfd7EPv;OjO7PYJdOQhs*4
z0BF-3=mB@|cX+QgDLoMdJe#ex2NYG<vAG4bj9)oTs$UGk(X!+50mEBtYe@I~4h@Id
z2f>yf87~h36^}a4hGSQ32OF313pI_LkL#_h3L9LeE$pYqTyT(t()5WA8J9&pkkK{W
z@WgN=<o@==O(XldU(Vd&K9LfWhbo9Ce7c&&G3>$#?xk*-#+74a$E$sQals%$M~{EI
zp||L`Uv0##iA%5w%)@wO+4SGarBr0eGWFYXT{9x~C#LXP7wxh;deRo)bXzuHXHWPs
za^)_|9!cL9haCl_20idxOJ3*CG~SNYFNc`Z@FBPwiZ<7%d*;8kW4CdXtEs|a@o%^7
z&UaXc7#>~|Ul7C_Awn^3GSN6oj>VHILjo+CJyLx|9`z)AMY1C;3~ci1RYEZwEvD`0
zjR%KsTxRNOg~8iJ?^!T)LZY%+Y;pq~lp5j?J}=y4x5-ZwLzThA9)YRuV!o6hZRAHX
zQ+Dfn%8d>T(}SJgns$!MqjM-<F*qtI_+x^Yq#lQK(c~l;z|o-NV)xtknMa5`lpgIc
z|Mg)ff0I?ATW2?Rc*70c+Ak4RtbNL5q!BcM6C>U^Yd>|-`DJDDgLu1$R<&J6pBdZj
z<4y$5l8x4ZkK4<8*4Bi3n9ocLE=O5I7MM5<`1BFUdu3pf6?|uQ<KI8F)MQwDo*Nkf
z_IDl_E0Io1BTs6UAIm~p%do+rhUG)JkZ)3TB@NHp9Eo!QwWlkC+fbfu$M$t)+XW>u
zo6k!Eo&qf@oh6g`W6I1=s&ou;8N>xR2#HSY2$A5tPC2j7xveEO80j7E&)_(kJX3E+
zjX8>S*!9J?;`M-gv%-5)g|R|~6F}HebN$=6=~u};z}TL_g>sCwi%<2ud}8C*<om0u
zWNaP(><e{+g=h<{6)FtI<glXq=_#AVEy5gvV0co<Wd;H?6V!808AsJQD?CviZ-$!X
ziwCATFW9Y*wO{k&+_TE_4H&y<tGjV?Hk{_{@PhM;g^?ci_P@BR?8{0FopU9Mj$fE)
zIAoP21v8|=Tr@_@?^+9{><EMdZgD`QN`4^g!*sJ>LIg7qFww8#%3UUV>X#&V>4YPp
zoWBjy|ENloSfTH8M>07_4RU;ORGPYZ=92EBZ&RjCdA((Pe%}*I$4RvE&vZd~{=H2E
zyDz!UE=cD2$F*U>dX{{>6W)+?aHT}p(OltY=f4#q^4GI6Q#^Vp)Z1#@$|fxd8vpfS
zo-@kOXM|P#Aa1?qmAdgD-=A5ml))h%?VG%~LBAqzD?)%k!KH=hb_LqGyy!dT7w_%j
zzraCg9)9%Bt_}H3@VT6PHqd?CqiT<OU+PQ-UhLxk0;(svzF0B6Lwi9Zfa!yx{F=g!
zW-oM*h7)8$E{{&brUe^Pyx=V@fBYt;`A48xIDwI!%o{LKEc{}w4@&j)PKevBP7LL(
zFY4jSX6PT`)xr{8#I@kA_A0XPQhJ{WHgD+v^nc>mmE=_oIeVvHCe0ykuVu%a>nd)B
z#0zIQc)(Vi_&w!f`1iN;X!YQ*DR=4K$F;8HcX8lcxD!4F*s0>9(uXM`tn1Q0iut;q
ze`S5Cr7$6v6q9hlS2jc*sCA`k_IW5(RE_wW2RYJj&IxQPx~&QwDwZEH`8e_6aX0o~
zh%Kb9RP2Mv^k210;y9DkwB#322;p))(T^7nd}Qc=!lkGI?C<S<6bLGCTl%LOw1vAt
z>c~*p7ziXZtm(6DQ;1tTffNIIq8QLYVd@Ps-vcrEf&`!tA?zRcYxRuHHm=ljQ|EFE
zS>hcsz`a7^mCr@0w1b}cI$Qr5BO$61M2p0eP@Z;{d-yOUSf|I(>vz=|VdBE=qCWrK
zh12P_J+aO@ipGrn3sG?~{>X;5acmb3HoE}5M_wKojyiE0qgJ;a{S(x}^8U{H`fY(O
zYP~sSnxc3BBnJuNN1i}7pU)JKHS_IL14q@Yqw7VC^Yx-6sUSfJmktFb--{A9ES?qI
z<%hlCVBz-3;i03@kiQmk<6|$4Sr1`Ul<YJ)<qPvcdOS}g2g>=oc=$+xxk<fEX*lQ(
z>Bfca?Eo!AD@eJ}ZM@!<_YT$Qy<Q5o{9o*TIMy9_4hD!7`$ZNfqX5s43ANe#gzIDe
z8Z%<p#v}?A))dpznJa#;d*uZZq@R%fd$O|HZ+0Bp)ImSkNCaziZhqN_6nPX%^_o9C
z+v~R9Q>stdubmc^bj#Xg!xJvH!2#A0DT3<f2vhf(;P0*M%)P?RZ{rrxOofL7tlF+x
z;;*&flb)?W3104)CPetGEUBodHQIUPCpt1tL}O@Tuy!0VV%17~H<&Eze0oV{xzq0b
zmv`b%_sx@m#idxbyhfQHtlm4vEqstDpJ8&s>b~Cd@Nix6%6qIO;K%f5(6Wxs8<1>t
zN7y>o@>Nc*$tjPD_x%=CnBuzd&f!<t`X~sapzFx*3Vka4ed!n3#XqYl>(61*C(6e2
z598a!Dqjtman1G{0zd7K89{PzEm_T<e$38e=%U-oGde<bXKszI|FzaSGFoZAK-IuM
z>hFtc)<5z@<FRu4cob3988?yN;T!JcUR|91vgotxen)=l6&p3ab@T=+W{zs{1d1;j
z)WdeaIoT%&g;o)JH+?bZJB|NY+EZt7v~mUh^!$}up!-lS4W{k%1Eu}bqd8-=!iZrm
zQACcBD?u#9EV}Sw%iM>op5D!=w1fD32B-#+T;1WpyYcem(0bx2@!$;#tj`0?Wz5mK
zUH^PhPp0!^JlL=m0OXUu9?V397o)~QB=L{=Bf<}*3wO|zerMrkxX3JDx~yF0(xttF
zn=?1x<qY_9h$&MKZU3!P#Uzd$$1SDU{cD2aYTMPcr;r|y7&B?a2n5+XrSMBRzXjg<
zh4-mTz04s!`l-WlthN$q8F!%?SjW8{65YnXxOwIpg`;qO&qg5G^AU!_L@lWG##w`s
zq09jKo+(GHRuJlmT#TW;?wSKN*JzQuw?U2U{nGC5>_S_RJYQu`-g~3oCx%+5ECc*2
zc5f+!Sc(G!jbnvcWL}zQ%ow^rn4dY=xdPbexN{@oY3Pk=>bqob@uJC~a>|_~KN_C(
zA!xK}!WZHKsi-Fqp$aA_wXUDwEP$`M=ND2phWU!iVBUUnLm}j}4?^+r`3k{rjGI|C
zWv=9Ap68>Y#5))<IV7^+)SwQY2VR~Ma<eoqNJxF#UY?}F9P`Phhc0b|(i$aAg`N6o
z<F$vBD@Ggfrw=J|x}^MMv4NgOQtK-=Q#(TB-}>a|9V)^cB_z609Hm?VI=FonXn>;0
zrD5t(h4@VE1*Zjj&4^JvM~SeB$|ucm0~A<0c|r{a$Frf4*RoIceMefcE9kcY8y0n6
zX<Y@A^PY()n7_-Y(SyC{DajkUu#t5%9$w6?@lTDPM_?@Zx<+LKzrrl(;L9xO8=TOv
zSlQS$ddDDFXjRA9cKS-W_oH`2ue?r}C;@tD3SvQ$c_^58@R&`(g8P{WLX>1qx%v3^
zk5@h${EM8JwqE)H&H4QSv%woakO5|-1<N(}nuW&V+&4Dt!e!p1oZ6WhS%g$`_GZ(W
z=C4WLk|9YL2VY7jpL=cB>5e%<=R1egBfB^||1{+t3pb;EVv-AlTvrpYurSp;wegyD
zQcMtodO>s%qMFgqU}JujGCkjVQ`WHfHKv1UmX#c^i^>%BhhkYf6Op*7i0jmGS577m
z)P8aZ<Ue^+3X7@}w&t?oLA?jlGXA>B7MI`Ia*{%Vs!T0lc3ojIae}mDh8+G;fYQ7k
zVk(uLGJn3b&)lIUoS@u;A?k>a5H1wR)AjyD4c;G%4AX#xVcJ{{GsqgHI3N<Jb|W1|
zfbL+5@h}HU<2f*B+m`24mPTvGk*zOu6O&hIpE@^&b;b|c2-?7f(;@PETy!vDN0^;6
zu?YNB-%Q}&_>->aJkry124}`GhNU@`jX6aTq)CM(G0~8S_S6|@&za+^Q4*Nmz``Tr
zqfP$2=mKy5BU}1?DI#Q|?Hj@D8zN(e(VhL4B^k#xPb&mSioDmZ+<da`A%ZzTYfN~*
z5!f@;%Mc-GnXK$K5`33O%gs|ZHEW5K0OWqQ(o3NVq98dGHN5^j(1X;?oMS0yu2-ts
z4uP~bI>5fTe-rVt0D-b6IxxO2<FQ(dHIW_C4|^L!7b4wg9bP6n7L-`SloYZo*g^^l
z3Ks`$O;)M76P*oxysrWVsrzQC7*3b_>rij`4#!hko+X#jWBRRMM)}=#;}t)BiZD}W
zWI;0RxuuzXn?$k$3TFXrNs=I|HXMojP>Dgl!M<P7WD}K4)7ADkx~cpAXOZ^+oYe;+
zbtk&sKp^+;i0AL-yxi=Lw0Hb#S=-{Agxk>|_Q;C&1y>7f2>P5!Hz;uM>&X;Fxt;V8
z*LJ}h{`w<QsP(J^(N{yX0NGlOpa$~G{s|Xu-h}!NsR$0{_kxD0%TmXbC9xn~<|o_P
z@o9ab#W#R{Q@?a=Vihhi%u-H)l%GTv+8p#n-y#2Q;yG->fAFdSE6_L7aQ8f<7rTL^
z{oI}UU<3h~YX9FfWIAv^3}~f?mq|a{aQ{Pm4cgH0tNktZ7*<~^&k4^J$La!dE)h4c
zg?8Zu5r%YvY(=e3as7$p+thMc5G|39d0l?wJhk!Afxne5&@XSpV}dsTBg$fq!`00W
z>pJ8)1W3cO0}|be9ZqXcWAQUxOwwYGXJOI1;8e7dam?qB^6bz-Gxz3JMG3z}(AF(&
zqK_V`QX(RT+N;)`Q<3Z7X@I!pN#zV(QzT`oBC>I(psTH=C5`op43ZMbUKIq_Ir7c<
zJKs&9|2T!$eJ!ag`1S)ToB9NE&1K&XsgUd5?YcJ@ZF(C!Py{B0hFfSi^zg@N-4F!|
zeEz4<dy_}ov9UMZ!P7m6{60A1H(8@eX4GNruvQ6mVdS#0{9iKYch_%2fZ&3zJr%C%
z2D6SLc6UUei=m@%d9WQGV3giHShvbINv9_ON66#L6>q|rvA;jLO7<|TQmS5XTGy5E
zTz@+r`V9(Kl_&SuroK!seg;|!25CU9ow?-(-5j|Ux7LHD*)R}naU4l3ndE|GQsJ9q
zH(Sa=H=IkY8!F5CeGML^KHJO8$6oVlJXeJEVx$!oqyN;Sv+{Lxbmwk)M&yxi{Z}a!
zNoS8^>TZV>KAOV=dGmFUHRd({H5o?~YfzRYlPjro;og^koS-%cfc=;NuBsU>rb8{N
z!#k{HE^4W1TM9OMSOA_s2vh^6CDnII9yo=#<xN>V%gMgob+|_UB%42XiChezQKID?
z?fB!aVROWgrg1siLPa0M|4@8@$C^&VoI=V;huvmUm=DCPX9#Iy2ie-!D;CV3`9MwO
z1~sA%7X<~Y@x>J)5H-J|A}U)yLn>p1i~rvN`3fhF@M4=z>+#nv^o>CK1Wo5AMiM+3
zczym(2YeyJ%FfTz#M$3J;{f^O@$rF20EwzFX=*fS0tNG1jNB;(!k+F}&zWs4X(=r8
zrL~wJ(*%AH0VEAJZ*~{H!C*<IpxlY3!Oq;|kWRRZ_Avq8Y`a3~i=3_^`6T&aO8gg3
z4eeo%ZHrvFfAYb@QCBA8SM&^!u1|iq?0meD(8FQ{fvbBCqu3EPcy#7^DJ{O{dWCsM
zrweWSd}f)kjS4b%P!oT&`~LW&tbeq*SISxdLj3oZd&toEe}A^ehFQLLd_a5;Q%ImE
z@g{_!GpeNHZ279YN3AdQS_59~5;oxnC2l7qeXG*e2r{9nQ6#b&Ul7;C_3np9xTOEq
zyUEL$Si^og>)!HU_soN-F3<14#%cBUQTgKw7TA@M#!Q`?@Hq)`=_dM|m)csW{N|ov
zB(eC`kTXhQYSSqw%IE7fyog7e9o+6oPu{8ebC%`g=sIX1blj@ZZFkZGny3mv7F-|^
z*-S_nABpv;x-7pvc<2iOtHOAFHyvVFuwe@odc2(}#|)wv7yJih@eZhF7*#J{JYWEE
zIJTAb%Mk2qO5+kTsytA{hpyEPDNcO<^Vi$nnSr*6u|E74Rua9vR_}@Jz?ze72M{VZ
z7XFBCm-TY{4rF~1<;y;Y^q>)Mtwobmm>~F7p>)1sR(O7#uJ&cY3xw3zJ+Hr`dImpH
z*%-Mm{4n7Z#MtO)X#N?#kc1|j#U|&{GvZ%(RO=TLKU%nq{|k^GAL-l4P;q8rCnbQP
zprNNC!%kU*uA%FujlvaUjTAvpaH<P~hi=b^S|NjS97zq`LL;|;`^oAIPxp(kI|KR`
zlwG*z8(e8Pkt?@)dJ!eEjiqIx)#D!pKkQGm;pCanvpXUIb+O(X=WLs43|T)Tl%xHa
zsqn4QdYJutJ88sxKk6-*ly~yruRG4;%>9<DQa#lr^yV<Q9uk0-iZM$cW(g&>T<Olv
zGv)1++}$=l(fV@kQWUT@QJYMneDeKi0Y)TChTd#_yX%ZG@1u32LPkU?UkgJ!v2G|t
zZhS<xyEbEgdI2S+D4Q=_(o&K64Se{Px+#CvYCtpqb>~vvyb4%y;!YA0Az}(AfAo!%
z(Bt2+T(9>8o#iC?1a+f16<&J91y{j4?3yX0`!4h?t>BYarMOQeyagmF5Lp0B2!9O0
z22i`CoSBM)j)d=6#`N2ol~8O{-?iwVx#e1`4SPnX8{h1cCwk|v@6H^is|a%#ofOg?
zT?UE}eR&I*q!t2wWr)iR9;+-f2PT1r{MqSpH763M$+Z6r1;6`UKzJMGskfKcvkrhf
z{<D(r<$uX+k;i;VN=Jjdl5I8~zr0nF2I;MHKp_hXl>V<o{-%Z}7jydsNz~NLL`@PV
zl>5hA%c$PLHgCVJb?qZ2vWE#-(ft#~M})(SHs64GGnKj(Gt`jHNf?ZSI31Nc6UuoC
zDRhLW`FU!QCj@`1L`rX2>RswX|0VRXE9#OpKsE$Sg%tj0C`=9z$W9dvzCw-H6_rq&
zV*kc3u7hQc*YCpjPYD^{qdGa$_WN0Of9nlxDEO%9ksZ^x^T}4b6TO*fy*fCRjPBq~
zzU#$#3{U>1`Wj9UtHP(d-EZf#F_&bf|4i4$-#iD^A@0lpVR8`tjBSqE)+{e_)ORm;
zYhulGZC7@TXpIaks;QF$tT5Tt_yw;tBB?{{#s`a7wI>4@#thC*JnYXIRO47~Cj#Ka
z=ND-9e^Whxh9-cHg@xpm{XCzEl$NKxe1j4eLW#S+CJWd_P&m&BzeciJm?#uUT>|7y
zRAe1YU6^$!Tupu6uR%i=uiu8D!_wTAlpOnU-%{<6r`JARwCVdGAR(wh>5M!;0RivB
zBWJNFdS;6OLqgxm05*IpF^GAR=JNOl1~U)q9Y<^eKZ)Wyv72%3l1=wH&dls&st(KU
zLOJXe3DQpLTv2kDObx@A%1W83&hPCA_wD><EkKY6N!+8gB^!-U#&33g3{1X%C@k3P
zZCN1&{Dp+xd#Ic7ilEA5vrJ(Xi9|w->_L#giE}qNz#!DY{3x#^<bj>4@huJ}Z0+-?
znr#jgA*fT>*mVvDQ~?nr04=E2mqK<HLR1AKdc&cALrBk*g9Z&OuFElfPh40o;q<Qz
zB0weVY&+iWWJee#st{x_G>6|U1?8+YW<+Wt@*VD!g^I81^c`f5#FKgzCkVV7riACl
zmk0LN35VAUU|E3Kg8_u8bxZs69#afSe<ekdKS&DC@4}v5lq;vfF5|-b0qJiTb}_0S
zTPPqaXyTMGEPRL(&}`%~{ked>`W(gs8%0I;;jO<xKK1D#hKhyWHCEwA{?Sb&vApj5
z#Hb%Jdqp<<Wf2$@MT6SwJ@~NKs8%1DO2{Av5NlI$$o;)25v$@QNIb|o2weON>D>dX
zxbeL7B?`gBJOk(MvQ&vR6LVTGyE1a-cC+O#G=IO}ZoyK@lI80%3cL?`|6d3>c#CLV
zVYD$%CntvUmg}4V|C=h9Ymmtg4|)JldOpYG%E|p^<}2B8SZs3Z0SUV2Pfx|vRaU!P
z*xH5|BNLMt*eNBFg&Zr3hw*@Az}$jzeqv%6+Kb1XHL)ftT0!!*-b+e~ZtJYx3!JIm
z>7P<y>i__Thb%%DRJwUY(3w_H4+vR%RY>)U<mV)^#>in3!$o`!k+<d6A$>9^puo8D
z9>l~#3DON_PeC;m7iWV4G!9FfnQBlc%>^PAJKt=0Wir+Gk4mo#LEx(alu`*vrMsAa
ziLzK{&3N|`_X|M2`z3tDY-qxWy){zF$(Q~`mcvv}-*zy(IxaOVmNSq_!lL=`Gfohx
z`-!IU&KYfCxO8DOpeZLS9ca>=_EhfBpXe)+YoXsQq*I!%(6*e5`6C0{k142L_iyXC
zeE`3<FW|%T`(L_7X}CIAlg0G1i7C}mw-@Cf^n^3@M1x~1(ZGd4xqeaOB|{PWZwWi%
zUf=y7;`bIuO+9DWiXgrc5a&jlxx++Nx~G5A3VCM*fe431^euY%k-CO;r@o><B<S7>
z!wEtOQ4mMQ(+mD}ikeb5iJcM$SBS)FwJG)mB7sebH9)gU6<$vn@c#e1#Y#8`2g9IZ
z#Y&0bSRv*}-Kge>tNf1>E|UCL?%bv(Ylt4&5jC}(VU<=3xAj0{*<|y<Mv4J-Oh3J~
zY5j}U#}!XJU=sola_g;*;nlXJ-RiGwoap93&CJ`#ntNUnzc$8*f*v=9kR#4W-Ez8C
zjqn@clqJMQfIFzrO%<>r&Fe20zH1b1m?2~j)snO+IYtTupk}T59o04Wsw2Lxt)4Yl
z%dI8rKe|?50bU{WQI=9WBeBp0&LPEzTpS<I^Lz73Vj{h}*LTWN5joF%8WS%gx@2Bc
zYOsKsjZ0;fBt(~Ty|^}MFnZvt6m)v8`d9P`Ty{CQLR*L^9QBSSt;3tms#bk$_pAI%
zO7knR&u<9J5ah4KQ+-fNsOZ??P$V%^q1nIONE=J*L8TPkS3h5xI_4puXb9_O5OvRv
zoQA$tHQ}uGpVv&nqr|VDQctkQ$LO0m5oO($52yaX9~U9V3Z<Ij<nDEnLO7Anq|f6K
zRQI)4N0|DGByN0o?9D`8PCgb%O<bFd0Mt*lxf<fdpqfh&$p0{^$+j7*ZX*SdkpI&w
zDy<6n_VT3rms?3|;uFb(@n^mqR-f_dKgdvlD);{c;O)OY_F@B*1hohgPVMJ3Ir(7;
z33+?~zP^0<9^RkyG3|cHrNj9&6<Z|XJ<Jz<X53;dN)j<%F<0I5`<FCQv9D;oqflmb
zJ#sE0fq(l|!gl1?Fmvg}c&K)5Lni7(F_nToaD>@~sFNY@L&cJdT|x2;uATlI{{jd6
z>XOIIt^@`Iyi}sY6r^^Yz!YnriMh${CV4=$Z+_+6P+cRu3m*V8IfZ{4`Ue56+u3@K
zwB>rzeVZAS#_JtB$j~B}?tjcIbKs2F+%0#vL9GTuRQ>h%@bCk$lwtd~@eZDfuu<Jg
zz5iz){h(s5ck||g2$EEXt<is?v`PNsJ7l{g^prKy*o^6_bR@2_N}Oq+&Rw?F?Uz|v
z|DS_0dHxR9-?A~?o@=`%TzM)AudWAAiuV2zs7f>If`?zRB=St^Q?7|$4Czs~zHufd
zCMOikd@Qipp<$yT7`7FV8y)+GG*kKpRWh0=g2@VTxHt!{A`D+%mlpvyJnV}(=5}x%
zCai=ivo#Z&Z#uRN>`2Ovf~6dj?!ElLG<J}ta0@RLv})P4cHc~QU5L(uO}F|la6Ks}
zs+xCW6*S^x9;+54m>Hr4K?VlC5=hXyAmaCmrB!9H{zho~hb59gfCS*(OjCtq5MH0X
zsu&GN(t(iZLw;7)ppDA<a|`1eENUJa0wO&)7afj~u{S}P<E*%N|50D`SeGyc&W8|W
zqz0$>zeI*@SJ*W*Cf3u%J7s8@wft_CyI57ocMR_=!H7QO{@ulMle7c!fzg1UoU4R0
z8RTLudy73!W%TW}ss{yne`<??6{`kV&Sno&Grevpo)+9o!lTHFBt2e_nzy`Fx6W}F
zo7hw!8A#FCj5>`VJvyADD2#-eS(8t3UVKSUjQO`ry}gZ3%}|ZG&{F*p{pHC{&wF;(
z2?>zTAIi$bG`9Y99WVp!I0Cq+F5PBw%Z=zyFWqNytIMH;9o+Z-mi!rJCUg%5ha9{(
zm^M|*k)V9#tsvC05*=9Rf?XeHEB^G<EzxlL^(`T*hyYGKxKe8W4l`=aJ2>e-AZu<7
z_DE+}eBT*zX1SDB2M-?y!9*@%Gxs#vm7!pO2pq7-4y<=E{j?-w;}LrI!V-(xa9~LC
zYC0(1=%QlqLz!V2#_-sX;rB^+KmjRX%Bo1_U2~u{oLj?8+0~Nho13A7N}GagVCly6
z_U8dKrU;|o^Sm22IK`1b+Z>B^uf3a_{4iA`!nZF1DZET3WR)-$=ZIDaV`OEP2LF<m
zmu6&S6%R1`?B@Mkz60_IWTfIpwLh<dy!2Fh`2bz-4eM9wfkM1pEk6=CC=>d$>ejNa
zm$_pWgda*+x_D$jrE`&+TZIY$SyD1?PEi>wR5k(|8XhC#X>M5sq86TX2_uVyMu*~l
z2fFgTcBLzi^rz35b0IX&*9S1*E?3A~b9D&hoxYtM;&Yq6IP>J$-<pfG@U=a(Uf|Z$
zv0WPb2OTCH&oo+FR*BBS<-h--l)5^H5a{7iDkJ^%{|$THa-$c&S(Ei}%>AdZpXL2w
z>qxW7<Pur_{J>jL?kyAE9|Y&{!TSdnn#9-qn(5Q0`5F5GGpMmGhOXmBAXO=={nOt1
z`d`7@=OLw^f#+s{n}?p%6J|fFIa9h%9vXXpXjzm$wy`Rxb?u;nlqB~bTlNaY1^v|*
zUy29i)xn(M7rB8gfv%Ph%RxB7q@<0&cT4{UPIz!9XXsQAoAx3T_%@HTjnt0(JxQ-w
zxc_L?ysVWrG87;yu-m6mfKmL+M%p^|uw6wpCESyF3o~IA-+SpH$R%N|G*s(we8gRx
z7Fcs|CRC+uC`>(=kO5GH7Cc}CS}V}Hv((GF`L)hKUs8c!-wjG$tCS-V?6BMIG3J-!
zJCX#;{_OY^VqnTvE%Zm-*T%m?NYZ-ySLo3btmwud8E{UTS{KA5`Wu+dnc%Sd8qVGH
zn@J-b@ALp@28NwgV0Fz#tLkX=7A1?mx}^Scic}vfzeI?}bQLc-e;i~vV<^H6<v$w&
znE&l(!k49_FFg=i9Vg*Q$7PF2CE%)7)k*hc*%pATpX%M&St%1Oh|{CR>az9>bPs7g
zudqSYy<we}aW8X2Zlzu(U8BwX{hLi{F0tIwju5c@Kfiv{3H$GpEroj~04GNDrZm90
zmype%=^F^><DYL|Kju_l*aO^h=zg9Q@T@v@uksn#3bVGF)~5r#b5&k;5LVtj|D(i}
zCx$Q<<-lO>1%4&z0MYw)Qz~aoP5!SU1ksmtT5-u;@izFIXRnB7J@(i7g1@){cMqw{
zPHO;zTZt4G>}me>q!5bht3{!j8n?3AF8gmxOT%a<^T)RN&zZoTA2Nv2<`pXe=)$tq
zgQcT4Ewv+OTG${2eiC&GbqxIk5-3WBHka9=CGsV4{LsFD0p_<a#)VIR?aFhV{u|^R
zRW>YpX8Se$ix}d3bUm$q5;LdW1s}irVg1AN!nhJpp`V+s5y#S%6T#FO^0^bK65}qM
zNrE8aAO_s8yHRNTV{x0+1mxfq5ok#W$=yj&4pOGB?SEYVYnuNHN+<8AGLBHtK7U+2
z^XV)GHZ@~k^1WKnS8S|a@}cgg7A{;o=5ZTeFyb_=bT+^!qkH;~{P9|5s}}>gs0vtA
zhgHgk(+XpzW?&!2%>Ir9GNx^qJs|o}(HaUclI7n514mKZ$MqfSI0|d0!88^O=lc==
zetN4QGeW{YwIj5oq{*KiB^a|h6bAv<2?PpcV4Li{J4~oK?sDDVTtXC4$~3hr2;G}x
z0(}>+spa1%RuA&_;P}cFKnBv?I3dCM{S)xexEBC%+yX)8t%$gh7Lvh@vA9_+=Gx7F
zqJ8fJK;r+#S&kd^!}JdnIHT+9gSOc`12dTjYPWYy{)ECdC>?$m989dtRuEh$%pmF*
z>^ewD#blI7dUxEvYLa85i%L_03K}yy2Jt9Ot2C0sqGyr78s)X$ad6oRAAuPo|LIFI
z&}jPstNbc?or2b_B06k*ZMLG(?R6iI9gO)-4dzPDv5w-Aj}!qLd;kq;maj+_KgA4<
z<AD0ZP?sm0KJMKg;8WU)>FR0Ddi3rDD7vrwp_i|CM3cfX-kNZ-2@i<DXU4F$;CYhp
z;7p{5k>cNGZ(?sCuo&H5n#2=uFagaWQE#<(lza}OooGp|w)B*&Sz<77?pF_~nN2{V
zpt?RIcHe|Wo}yiD2}xkjNQKYmCEvI^6ar`ps=Z14mMW%Px!50}+Bv!3;GUn?qU7g(
z0TMbU75C$H7b@_i0-q4|y{G;8@OUoJ_JcKgZf%CVVb>CT9?Jj2y(N&ULDQT^mD=Dm
z_egTie_jg8fPU7FTJVpDR$@w>gq=qdBmFN)&uP7&3ywaHfdHI}_}L;FH)+DT>A@q5
zm%f@+Hg=l<^5%x@7&3%vg=xQbYapNzo$-FsZDao21*)L5qd!{7fELg^@>oNU2?OB^
z^@`{0p`8zkx#dnewbe@}`Fdx=j_yiwW?F3sc)pz7Minm)_Vtb>Cr{)k8&(GMeg&DZ
zpEUXN2T*Aqt2>dBQb@3-tb4m5UuQLkbGSD3)-Q!%IJ&yOKnwA=VFQ--1Tn`iZ5+6~
z%%08@IY$9Hrs5WoIlg8>yxm_|m9$Q4@9f{%*!>WZXHJE^bE%|>d1~h55QY{&aCQ@a
zak5e##JSe51ohv>ya#Xkm41kQkIJXjmtvJS&+1$*XG;+{HMTrBeML@Q29I!R@(<JV
zQojB7FP~G%1i<0_^|{LlMPv?c4f7Ke6+}YUltdBXt%X;rq<BQJ^>*2>+cn3>>yxvo
zZ2Buz(m!21Gn%3+|7t*P56I?29JRUW3n$r__db0Wpx;8!@y@4rb`dclYq00!gWf#?
zc|s59060BVfhoTvVOTrEv0GYhF;vJG0{v+{0PDHgPD8}>g5;fbX`zjk6v#w;31ok-
zNLI&ZxtwUIsRe-=Uy*3xEOc5Pd~z^mNDE-|fNvx%vb?DcEuFhAxTy`?={3gBLM0ns
z!#YqpYBpcDhkSk>lUrUHQ8LdBaJM)kwq*C7nb=Jc=Ue{d+~WNBG>cZ#^IkGeiG8DY
zjUJ)tW!T8bYailBJhOZ9k&j6;m$>_XOY#+DCzv4iRB#xT{lUPmmKQ0q-tO&{QmJwP
z9cb8Q2^Jg<!*9lnS`hvp7lhD<X=zPBNkxYOfMWUqbE;?y1L!Y#et%L5B;m80z&&SR
zPO!aam5`%i^r)o_r17}XI7VJyhh1-y2jVh5AxybAikLSJVj-70rwdQf+m$%Da`c*f
z_y%k1*kjfkv|4&qKvICVxmQP$ha;6^M0@uNf|{x(uIaL=$-F*Mt_{MC-JgAb9f*hH
zbj{+CkZU1pP~o`2?|slE6a=Vk#6ay`UM}}D<y&4?pSQXN)G-Q<##n?Arc!GdI*uGO
zbN#ne4jZ2Qn0%6}mNrq*O5uPjJQj@~xLQY?nVuGNed}ApD($CUq^Q?1=fzWk0o2{)
zt5dW(rgiK$tB6C5ODcs(O>4UoMSnpvuv+r<0{FG`U1?|1@;+}s_oSt<-K`&ay6^nP
z@>S%N6}8%*GCOWt+6>PNj}Nrx$kDmEMTR>8CB-`0M{n6A16W#t&#3L0I3Xf)FSkH5
zthqq3pDtO+{_sO-t^=1%gY9nECrQ7ZR)iT?a4zS;xD8$e!<5<*YzgB!*0wbHFoQ47
zlrbcCq#)fg()g$_2hq)VO^C(@DHc-V03F9|4v@Hqo<=$T?|c(&^WTAz<N6z~f`Nt>
zGrVZZWrfi9uIK$D66#jJFo=-v6%>d9lT)MOt%^W{L*P@E-8x1|)KQHESo87mP~`Dp
z^oC-2ZS>X+kX^Uj?qMype!n-cky!9Ds?7*Plvo0Tayxf@4~wmKH)eJoZk=9uU|0R=
zjXIs4aF2;^p+VY_U^OdGt}CIYM|Frql3d#8A5gs{?YQqa*<nk#=EbM?VdtFqdIr~i
z!$?e~-#St;_j5sl8IqigTPDO~XO@Qm0BJ$NUJ8<2Ev$~Cu(y+)F8TJbv_AQ7kD^re
z1~z3AZ-|11u-k{72$0y;cTKN{nY&cwvAk{sKnWNEeCX}MWJ=zsu4-dbjogE~s5N0b
zRn+mG8-5Uh5xK>0J%jM9c_FMg^23M_@L&qp*{CuAP;4wXKzWcMhn1vXy%l8uWU^&h
z6t{z=#*z6y;KhHnopu@#cM%lY+gL>z1uA`6edGh#7wEjA&Eru&X(3|c=;ta#T&UjT
zd2<E1N-cmlM0Mji_3zSd4lt#<e%7GBpV3U7sBI+SyFO&3u7%~3GzKH+gxSx(!z%4f
zsx}ZoB~+oXTTLnhd$Tc3*YuyJNndC*ueCrG<JHP&*N*SFEH~NxDSDK*La%N`yI<Vg
z0mZOe;Z4TpWxW^w4MS8b)56^qWGMkPYlxZ}2^t)7gY)kgumXYYV4l>O2u*NY^6+`I
z=E0c~z!p_C^`ARuscWTS^PlrQJ;f^4m_T2QCaewA_7^&!*ajKl&ZCX?kRj4dL%E@%
zB~a1WIP`~zURz9^Uf-GpVE~kO__!fQ%gTG)Nn8J~jrn%BV>H=7Mb(fxDec)mgZyZP
z!`D#fv9p%d^B_d)AveD;`b=_Bn-!8j0&Cia|J$VrOkQ>8u}Uj#yRJ0cS^UvgekFr(
zS~{DQ*8VC~T&>$+ejaGX!4EGw{;@w*B%U1N*$Ef0<O~f8C5CaF#AE{qaJ82bwSdwy
zrbjp<*H_RJbXN7RJgCk4ZsLQgjPl<n)Y+5Ppjjo-v++7C&i7O0AK^?914!ra_MCnq
zm+arV%`}zh>^!x+OC8=`67t`HN*O=K$f=0GsB4scb;SYG_-2>rkk%4hF39uoCWy%a
zA}+_|z!Rcv)dhx%P>AUmp#AosclY@Y;x+&ZH4Z0uz&=XvUus1taplhASZuldWPHnd
zPy0$pYrP)gTS7*2=-+DVq>2bjoPL~X>W_fyDZIaMdU(|<F8{3ubA&GDIJoY;)A%O>
zdRc)96mGOmRYzkmZXMBf<96=ySkoU$clGk5?bX_NECPmlHDy5{;5q|11_K=1yJk=C
zQ(Amrch*vVPZKaiyaWzd)e#`O*nM&^0@9|tw8!C~#JeXc!JOrouj*LWneA~zFH30l
zt}xuTx)fPx6&(JW#R=qs!2jLD4*TZX8goWsLII#n6=T|0z5-xIRZUDWB`vJ0OYX_;
z$;W&chrF%nPX{i9*cz7ksu{kEW?Ru}Q{$mNUKYb+BDZGO)}-$=@UMK>ZkgJB*?NM}
zG$w1ct}0sqFnCR6u@!vUU7&8|+!@^-ILrLx2YkpWRRXc|VW{6*{V4wGdDn3N8CuU;
zZFb(m3ON>zEKNiQEj?auqnkO2`G@0gwi`&lH{)s-u9w3UhhX`MB8U3&mUG?IcmiE&
z5OlP^zW<YFPS`f~wk5bFrHpq&#YHdaE+=b*$B6@KSH9c`k?pCdbKzo~DK)7OXj?WK
zu#&MPQ}}&8O+hIUBQ{RE-v~bI-VQ;L3R9e~a@ztnGc2<-;LK}jwtk|taS7ofu+VC^
zC;!mzujI<&G6dsyRajYc-k8t0F@0ZSpUSCoZGUCHLir3n&yX4;bHKL{>g(CX@pRK4
z)M3;d=vkei;Nn;+7TF)f8=sFv1Cp|B)a`hII2HD9S%Q)I6ckP9DmwPfZTh`^e~>{W
z{DU11R%O2Pg!oldT|szjWr(;c1425`IYIqPckk{14JUrJS}JXhMF0B*W9ix``HgEW
z-$0*hSISa0xe>w7Bry#wwf>`g1<?^~{)3@k^Pf&zFE9Sr@S(Lj-WSEs{f}G4df`wE
z#mSQd;GNtC%`nh(+DwMe<Qn(GaWkq5A46KQ7^7>fJys!9sXUEy6zocWvj4@7-CoNf
z#-!(?J|hbg<?Z}SS*C0aFGvoX^Wm7cWaz!%f6rP|+JeUO(Y~{0?GpUQ8r@vt6PB^M
zb4$oiA%=eUemNNc;biXLvtQL0voKvCLAS#(Qi=w3D1*k|*iEO_8jW=>Bmp{D2YK4d
z*@G^^Z_bc@MM2-19O$Ub9l6@!;DrL&Us_av6x`Xy<80c;T}tT%8jt@NjJG$f(swBU
z^<sa2I!E3iQBD{4zbH%Cpwe9BOj2GzYrYOe=)y<yKWWn<5v>~Azx7@!ER~TDH(&NY
z#SV~##=^-)Pn3f$iq_5qJEWM}M7FC`Yz`V)PCc3i;zDD-d<vwQClrm!^Vk?JT&s2b
ziwJ&dPE83e&>}cK-oxwiW<(f8LYKRfR%=wXE7|!Nhd&8i`7U5b0CozDHQ1BAva)wt
zP><=kp($^q@U_pssmKE={O4`o!%tops1<rbs3h945)&H(QO)hh4()4-P$nn=(Yqc0
z$ty@}M>MuTrTZ=pv$o+s;<CI58$0$t*qdWdf$my-w})qp*SL0`b>}YeSj#F6OoD`>
zPJ*#Q4B$S}D}MhS5n60~YngY_pkwHSmvfc_m?hF&`v<fl0Yh<C>x?fJpU`Jl1K0PJ
zgOX5*sQ_IP^p9~^JCmUPJAPWI3;|$j-jkDX&Ph~S^`|QstRMk+&;Wael*313tkGGr
zfmaLno$0q_tHtL9Zzv6d|AbQk@pigBb@Eyo==&>@sEBJjLgm5Cb<fBJK>hbU1?Jn}
zx58I;?#t7M3xp;?-y^C;UeBuJQgv_G2FVPHH3M=tuTPD6W1gVz`ozZJt6ou6Sy&Ah
zOBr|aZTc301jq57gWRv5;lR)bJQuZ7>#!+>w`;95x}C>SifH?e4kTtb=}212FQ0((
z_cFU#`CiXN7N#`!C8g~coUX}&J?e!f?+D7nja|D3cbUP0-rqJU{{-6Z89uU>A0y1?
zOpklvV!eG50#a#3x@dE6z@y`sU7Ds6vE2w_*Xa^M_2(SlanMck-%ZFgv4BCINfW)Z
zXKh$j3NawC&(>{8!KLJ3IbZfbqdK@}L(uf`%MXeQ7xksEh@tKmIZOb#=x~H9bBh!+
zc<a2Y%JZ)kx?6|aS|ZU9Ug=N9Cogt_hK;k-zpM&v?c@g0$bJn@f1}Mh4Aw!*SGKfQ
zdmP|%G9T2M@kxX6jXm)0cRCcbm7zZB{|d{GLkdBETQ%i`o_Sr({od`y<Cy<P<HyAV
z_xrdqMS~7D?M{*ldUr4YUObYZPt^qHeW^fAZ3P015n7r<zd2Hc!EdqhLG)7t!Lx5S
z{<rn8pcn3M$IhDjX*Uqw%NnYiiUa=Fp1pU=+~kn|LZ_(LfP;;M0%ltX3Ei-`HSUyO
z_^fNP!z<m%%E0=x>TUe1YSAjg^93!TZWLFMhR~<?nTP<mzp1Zyx3zDiNUV~AMb-5r
zgjhf1UU?p1{XC8VEQ`qjj=N?t-I91Fe89U~<H#X7C~fLIZtW5`4f=7}!Kq8@pE}vJ
zw1>_vig$MBj*N`K16@+0<>MxRX#_VMMP>phO_OJ`_O@fi!S%p^kbFi{?e1=zE|ja;
z21VAP2QJuo1`~lcs!tlp8LKGNfry+)l8_TLxbpIQGogZ}kDi}mD3ZQBLqPQq9ntw~
z0Xz~Pj%{yNXVk1z<vFg49exh8g#0M7OAfp^qMc-T6JZXT0YM`Kbb#zgS%xOAnjw+r
zYFbu)Ql4Hl9p@}1i1`g4KVmpx7y!9$b0Vz`sJLYwtMv)eCQXkeXIli4XWX1f@w|gB
zCx=I+YfZMhW;#0_M<S$ApD$j%VF{l{i$eD28j;AQ`z`L)WD7@+g=1wBD4zR)&uJSS
zTTMeNsq3w~<$!P>=2*t^MgF*&tBN40kiYD`fZp%(4<}PU!ee3~sNT47>bZ!~$RRFs
zbE+DH%{`sEcPAMQ?*U+OdC~?QH(uUbZ$F;4g>|=Kfe4E(k`SDnCIULtCkL9a(!J*c
zt_3Z@+jgCeEYm%if!KOu_)bX)5uo0<W-XjZ1K^e*XmtAPbA@~(wy_RNIYjVYFF8>l
zXk17))x#1|2GU^Tq95N|rc_*GH_ZgJcr&kTZNizpHhGLCm0?g^zGnQ_zPGV^pbkli
zo4rpSc~@G-*z0s8<>hXGn5Y}_SHZIn5QAXFLoh>5-f0SZA04slLCp^?ApsEdExHHz
zoxtxR<&(^^0P`_G%bQ$rr2(J&BkPkMGb>5~5T^&W6UkmV7l@TcP}j3Bsi4lNP0JE!
zi~<=r6vSIl?pxmXjy2nfvybO%u>Y-`vSJ=V-ZwxK@AUF<dR^4?DFnzX-UX<k7L2U9
z8~j}NXmMbK^)elH{9$csCx1LKvOGB!!pdCyyHcpZZW98P&5)VSh8K-ROs?@)v#UHm
zts=F$S0NqD(!f`KssbDX=pJ_7G-m{+j!;mJj=<swET!eSX292G`V)V&+d67U!BlVf
z>g(g|C(%ra8Lz)xMpzqAJWts5%&LIKl#6pGw7Y7LGCVUNXg1}==>r%p-}p5NGd0^x
zJO9X8$C}y_1fWAmhu3))0xbC&8R?B8gqj6XE$%TgAxUt=4Ma7hpbNLTO;>yP27?g=
z((<rgI98`;(I#e~f%ht(PpV6Tw+b-NXtGUXQu05$bgwN$|Kb*o+h7MF4O?lIJkBcj
z1BPN^yHcw2SvJ(tZ;+XqiHC&iWe-7Oq$~@3^oXB{I&+zN#%|R-4VU};)F_mEbO<vA
zNy5<j-h1IygU-!t3%Re;(e%~TAI;R5@Si^cKIVNLfU9427anw18^?y|w4LD90pF=a
zyzrB?4Nu=&bvVD+>30|4O*&8J?qk{ytt@vefW><MoeLhSFui&ISR7!!aDizBjBQBe
zuP{2^_%+QZ`tGIts?0l-%|An+nl=nKICKlAPY)s>f(9CHAiPmQXf=epKUF>W`)p*8
zrOo?5<Z=Edfe5Pa31C6EJekD>8{_A(_Yb&M3PeHxwtV#^Jux`ZfG74JsvZDN`!Y>+
z$H|7(`9My>+RsFCqpUHf6?T<c5U)#CV|u6tVEkL;t>*vSQ1=yjSLOIKh*~tlTqaex
zB#k}xCJbB_Mwk*cS2rA|2c4K;X8<s1U%T1pzouYW8S007p#tQ8{(X}S(DS`cxvLhT
z**WR)7~X`BuyRdImt3Er{o9BG#ptl25%~Qd=l!up1ODCg9Q{&q>)GO%LUgZ{3IV_g
z6s1_%uM#zDm`F4zvBiI~Fb+O~7G+1dCFU+M&B3-7Mw9*mZC?{`x-Cc0SJC~Ua$H*u
zckyr)G$z<ld+x1hSSsu^2;KwpH*i}!-;R+PqQrk0=FF+i4?aFx2?yAq`s#4XGC7a>
zh5G;M`tEot|M2hQIJR`MvO^gW*)t=1MD|g(%w#7sTgaAC8KGnyvSn|g%w#2-Ldhz-
zp6h&jp4apG{qggUzs~Ev&*yWW>vLW2b+d(goF$+DMOuJ=*JfZpQm*P-9m0sznLdre
z#-~XZd?9hCTwxV|Z3M;)Q-`a<h@b%u1$>}5%cAtPju8*lEmyMA;<RpD%iECOX}ezb
z+u^7AN^NH2Z=Kbuf)cNY7DLyUd9R|<;^J&a7XvgDE+}gr@6^;delD({6Far|50=|^
zKh;)>2$2MCO~PG}{L#_#(Wp`Hi{>SZ6Rn_&eAX6M_t}UU6~H2OEYdbhRcDj-Q<jN4
z9f1QnfggGrUTM<`b*gZ#@?@Uw5!=@9@>1PxqLfrcxUH1iOLwS*VSL6KwfZ#cQ*K&Y
z`qT#o!%6l{I*>2<qX%wXf`SfN0O$F1>OKAH1{1)H5bW4JkoZ;TPG6S}qs(W2wP_L0
zmM`4}?YC`hBbLGwN3CYQya)rnjtSu=AL@{4AF_+}Y~6l``PPOeXAR%$zj{UYmoy~`
zT=&D{2HoDw*M{Gi=mWBL^xb;v2T`Em+Uy|CM7!{v_uN+p{a-u!_GRx@)t-p2^!5D>
zJi!`$O2i;_R!YUl*B9b=#V_D~K#1TMMw`menXaffut?C5Yg`5o0somEU-1gDgd~d@
zSs&gbJszyAkWjTP)$u;c(o~O=QonMgSkC#*t(!au`bYEQN%=TsnxBzuL+pVW!$C(S
zHDm&G0ZLtL*jdX?_oywmrDD%Cx|<RRxKbl@cjI*#Nnd3Nn9U)?gHUI+zZ}vPt{m;M
zrce_Y%Qnh!QaE9_lKU`SG2c>zM!bLX!$FwVC+^@_oM4`H%`wu*B_W_W>n{~CU66vB
zl=q9Ta10Tt+Fp}oMY$ytXm@zLqI7WdXGwrp(Ts9Q*XbO#;!kYPg>2~P8U?!w+wc7t
zMS=EtVl7fx8EY;=oi6o<Jjqzk_T00l$#EIES8dXsWvutrIBS{26f1zU0tRWM&s@Kl
zt&M6CqKTO{y3_t=suP}+kJQ`m$~jI?NR@pIu^GA*jw`jhcgpMd<&qC!`Z+}9D9bpq
zd!^;;80v<@k%JvLGhg?B9W1+wcOCotFUz-*Y`87W=iNQzGYYTDiZ(zX+km?HL>att
zEB`BYGAuE21n9mHK)st}?z!8}A)hjCjQ2Si0<I5UB5kvl1d#s=ogxuNc1PdF2eTeH
ztWl0n6|@^YvZ6pMydC4i0T}azw1Uzah|EL;myYYC5BJB-@Z>+_ic;rk$WqM<QgzW=
z3m?bM$w>R<s7jx$XDRo>i#+OiDpSS_O)Q_c!rH`1{o$CuXz)NW_jl+YL1|)Tdb3d_
zLJ=oWg*cOtJmg=Qq}6pxLg>@Lt9coL1Ihr4&5aGl4RoNGjdw}nJ2f;mg`~4@AgnC!
z6|6)`i*CG(ugdeZvUld5_&rWb=t{yZ`yD%bZvt%aX@lQ+&mi#nvBmKE^sZG4)rrY^
zzX_g)-}abqT=-O^aZC(7POG_L%fha&UWh93b(g?%K=%@3h^nM~A9Ci~E$yWeNqDZv
z@}nNLrdteyKyMDbrT6cx`8C!TWL*fp6P{?Z*)z~iB2N+j{F0Y9TW~Y=B(ES#2{Iy&
zf?U&|l4b2Ob#%?poKN_!IE&RzM;6YuHWbeS%*c|A{qKgJovG6jruxg2nr_Q8&HHIz
zS228V=Guk51@2F`s<qVpc|y|WfA%FWjl&a_`-!A&z~_Ru`__p_u5_@F#oLf4f7{>4
zoe6iBxs~*Hv$LK@zYR`0HLClqt6mK+*4fcbt|Ovl1!j)QNp%%M{udFN5!PKCw*5D@
zhcd?S7tU=@dhgWUpBit*j0{gvLc~T%lIJg%OIrhp=c8iPOZAi!ld8l3TPBPYAChNu
z?d8EJAEp5Ow<#MCt)JcjTqQohNxZ@}CD>O8*tc)P{WXPgbsr7IRd*l0QL^Co4qL_j
zrh2)2MdwUW^W0=i1V`JJMUy+Jbl<xx#})^}xMzIP5r?fmBO#&%7!5_$3*yzztETx)
z553|_{oh!_OF(er`=rOH2X^WK;c8C<xHL|~{v<(&56+`py;U7_6`b0Xzj&Tk=dF{;
zd~WQVF+T6E{tVGST0s7NmijJ_hJ&09N41CnBxz*q*S_yh(jqiVU8lfr5byqOfwb|c
zvb8%u54R!`$Q8VB2Y}9#wHvcpujQcH1HYak*<$9zDc?~kba(`BKOte}Gb{@F(i`nH
zMXi{kSAC(9lCSCe^~=ftLJS_C8m!5>FyGQEcvBAnA(^e)#06L9dZ(Pu&S`GgfMl58
znITV0xWH!4Zjk7HwZU0@;!ELbZSDU(K|_6bB-U=J1ZBvlc)QmJ5o;>MYZN)Rzz{g~
zEr^8i^eQgf0@Rt1KBKT-8#lkX<NIFTkOl@SU{E^7`_a%>ys}?qla(YJvfu=z9xIz9
z*(q097$oF@483!t2Zako6CPK7=!uVmXg~=v5q7cMYG3&kAp7Z(r*dn8O}|Td0nX>9
zY~?aI3}ERw;~WFKL!@Ij#R!rxz>uS&G&)YJc;Cy|3vN69B$hw?Y?D(Gg5$pbIwXGB
z0gWKW1VNkr?g(AX&1?CA)#8vKCl`xoqB$~EK_G1d6Yw@Tfei!54lm`9EdFW5IfYzQ
z=W%tR(PygUDgmc@If_iuM;Lr35s%s{-ZH)?E*sAeczI|S?cRBslKRDh^d21`r_qzT
z_)8{F^Rr5|nef-B)$H11`bU!2iAfeVYnUu}P5CV2-<*BI_Imz_HtHUi!qeX=di46~
z(Yaz$H7>yvrKTe-`5Wiy)sP5qE^o?3cSc~+V(qWKpGj1*R?qj+m4>6dh`CD6(WDoo
zpU&WIimuIYdhc~K^sYa$et&EqDM+jJGmhh<e|qDb7=B=T&3wV=P{D%_D&R}@sN)4S
z6aSCZ{$)c0J?i<9Lkr!xJoD6qUBJQi7<SY5c3+T1TK1gHTA4Lxx5~Zux7i>Qz!@Ak
zNIGG*Gr@*E*?%h{5RMK#eBy7tP;Kr(L<grHQ!Cx$J6{#&R~_PIvzqUUR33P;Qfi`^
zXrpmC87r$);Dbx4*hAVL9!k}@D;}qn5rEa*yttsi9oeR?pvD7cXuNmr+xZ_8h&-<=
zB7T_PB5K|?B47e#eRIvZ+dfk-0f4@G8{Oix4JEhxN4eSz#lw*KV_hGHXDBa2j%y8D
zydPp^oyTl@ty&ksNde+=8<|Oi+z5wdY}}+5x6aYj>@@NIP+6hCP_BRa;V9}O0w*5?
zFRxL4t@D&8!BXHXXBAda6O%bD*w`;g6{&rUS;tI5vBs&~RvsGr+*E|*J;o%z<DVM%
znS*9QQIM#o;&FR0BhbB~YjLd^M7=Ms_D`P*0{qz7-R6dSCUz$-G-EW4LHRjw?fY=g
zXWg-}FJjpI^v-j#FY1X|Q_vkA`x>}L(k)=XI9>#$#7wUtr91o^VUvP>F{HReIKJh$
zPJF?YOJuh^>~UoW#{C;K7C7jrxn_jx^vl8Wjmetq2)$f7kXZY;)h<3RurHI92dz!N
zi+1?@Xtr-^TA9<NBMXa4QxT!Z{|(Vv5#u-Wq|=;WO8Y0dJK1TTL9Va=C~y$>AuvwM
z^P*DCoRw_s_dOr;I0BFhv)h*m5rK5`VIbd1UX{f<G=gQQbDhsVG5eRDi43zB3}Dx_
z4A+{8*RRrp-Waf2pyA>{;_kLGP6ZV1JM2r=`In9t84tv~Uu!Xp2hA22b-LRsN9=O%
zuJA6nJo@h-Kd@Wt4SS|{QGoIS^7R>c6FeP%|1;p5c`i~Na#NTK_&=p3!e6Zn*Nro9
zwHexekg8f${pE0Uiv&SIw)vmt4g__s8oBdC1%SI?EHRq=(>qBT1Qn<4ohN0b$LzI(
z1lY~e9X~|5-v<4MzYi~HM7X{t^{@8s?V*tiD8DiV6;aS(M-DeXyju9Nb==f$__(|U
zY>^W#s_<g1jJLb2>jP(=qeWaSLE}w+<+$R4%Gtf5`*J{0<K%CW-vrt3=;uej1dtcH
zp~dsk*}%fOOJAV`ifw+i`77YnhuInUooFZ{;jY54v(=g|Yb$Ols+zy91Z##TUfC>O
z1_~P*K7xZSE30LKt!ebI7~(lz;<K}>yS@(JY~t-$9zeIQEDV5yY}OIS)~3PcXPg$}
zEt|98a--dfk>K_p7s0&6(XHjhpTi$@uVM}&<YIcCJKI+rmw$J~<$$P)4HZ79Zv+{v
zpzi_&Onp!8nP@5!g!zdm&XvV2Y+7H|a!dvVD-ie#t)~-iJ9>0bs+WSQj*NUs84zh$
z$=JWY&B7vzSU}yF_yiM0baPQ4sNegkP-UmMcpzwo(LVe1E|Dtikf{Ue^TeVzWGVHo
zZzrGSMWo9(&y)vd-qi1XZ`ad+L6}|)?{dem5#hi?iQ-64IRa{u=L(@>;R4QxX;hp4
zpcCjL9sX#b{Te^#B~=j|eB!3s7rmHN4)_)F#Za*AWzct7+F?!jg5JZK(AyuBfH`Oe
zO(!uMv??vrKbq-xkQw;=3+tr?dK<RB#?RMZKGZyqYm<MNG`Mj!Fi_KwxdY3CW3Nnd
z|MSAY2ry@I6M_?QHU#yHj8^zuyXQge&ri4JWJZYYx^?icXz4bFmg}@L=gfvD+i_Q+
zTQJ9(x76hs#mm{UU5wnkYeEZ3XzpFvcqlx0P%{7yljUWa>lPU1`onIxc_nT0a~`|1
z5~oxO{R51~%n;eT4R*9c0uCH=4SN5h1_pp&53P%b#%Lh^pv(N-OFEcvRn&Y$z|+-?
z`9jDY6YR2JOC+uuv!}wp+Fr^ioo!@5jY9f-AaI6^8f0SF(yq1WB=U;mV6dNlq8(yJ
zM>|jpl(Op!ca^~gIlp<A<m`EEYL9?Z>!0l(B_LsxpUOYMdh3!<g=s|r>wM?H2zhqd
zc(?X%&Y=w7Ny2(*lPr9<7<BUBFV25Ewk`(QVma(vNH#5!2rk4~1no*`aar*PfKJcd
z6z3g<4h~33w?fyG_PZ<)ixLOy{<XlQVDM4+>|DbG;iJz<UO)xc{O&w-*%<#ScBl7#
zx)kFob>IWc`t(UJNS@fY@Bdw`g#vr*sPMs?z=wrAm!<j!wN#q->9!BKsNh%<)S>M!
zW~KA@oRGknNlZTtwtD+Mb!iu;INY*_tFh}I?Q^rU6w$AwcdzXFme*ianVoNV@M-<B
z7^@R_`(>*4M@s&)>jG?-A~$p?{uO4E$-R$}YVA}xET}C+6eFVKahW=Ya!6N8OVVU#
zx5h}{Ah=_CaCVD`XqNMC9lA>p^1G;fpA>dzp)UpnCY>m_xSIM<sT(6mXeW)wdBsFW
zwM0e7eeGPGsb@-^!#rtw2y;zmW&T~#%^)0p1IxS4c-cgqZP9w`<ZUFuUdkC;bdN+u
zhN-7-GYCBxS0M#Yj$5s82uxmAw0=pJfLaGUPK&mt&-Z@TC|?!g{MF3E;0lGeb4;yZ
zvP(zWP}M|c^i}pa^F!u1-cu<FQ||{4Xgy@!|JUTfoY3Zz$b#ykvCO1LPGCBQS%S@N
zGcNOxD$(}%alki>ZH}vW91+bA0SsaAP2rlk#d+_3{lMIS5prAonVMRk0tg86to63p
zISG;ki=a~nEwX$5X9#G+7A0x7K&Z*%?0CQM;^IX!JVpK=##s-Dj60MUAPWJW)3Bu~
zOIc$^@*xUrS3Fx61$Au-S~_@oG3zns3S7z98pg7*=^HO#{q(CVMD#?&M<U=pVmBN+
z<uPGZtmC>s6hZS1QOojxphHPOzetIG+eHo`3T$kWB?_%g-pbi6Wq!434H|imA6rx7
zkqq&1H!z;daJ!)EPleDatAj`qXhfhBv6PxYU8`W`zDM6N8QK<FIfg{hBG&t=$R1f4
zYk93u!l9R>F5`Un?yHOC0-yAumFX519Swn+1NYOeV^N}EJ0!uw352=hGhc97Dr(&3
z2iX{S0wTv^uf2Nxm0gDSEJ@!j5dMV}W)gqi(b1z$DJ;2cU&O=Pwd;FB$^3fX>DY(R
z$oUv^H}-_E|3DL_InuO?slj@h=pH6YV*%T|;mlCu5Ir(99e$ynO>|ypj)9q>dSPjP
z|N7k$Sd^@SH>HCBqv?!|7}-~g8iMM95Qc<r4r!+)d06984OXu#VB9@+V-hngVrQB_
zRZr8bZj@ya-h}BZfhZoFg<-imQ)ndp8CPAQ|Cxl|s1$0HBSd(pU1rqRG2V&Izr=zf
z$+^drlDf2Y&mZLIu!8@V5Dy%J!XgN(ScEl7VFNE#5Pv*4^l8qAPzO{SI(Vn|&vW@U
zDqdm~CG>9<m6b??6ckAIZ9jdY_T|wajjesnpo`Hrf@MJi?7Y-in1>}+ju7{y5*0OQ
zYtkqiSP}$hvkF82_<PyRad5pMPH7Pgr%8{}(c@79aX`L;AZd&SE48|gjxiD-_mfcT
zC=Tr2T+@m5G7a@b@m6Av{JaENd~!unAMz(WR#OnJ>xfdbBT*;HUt_&=jxrh1BEd4^
zuUEefX%)CJh$8-G3<Tq;rNp-D$!r-j2an+8OOY2+aF5{OGi&7HzdLaZkY-lW#7ia{
zW<n>EJvLQP6XPDhQ{&g1A_v2e?Pz<l0e6%ho34iH2MT^?Qlfw5Q=)|vmP%|C`NAPs
z^%EiwMIzEBa2HX6qGe!>Wk7H|lE>fsB=zcsgr3GBVr*_YR|t_`r<5)wFKenDk%OjQ
zdBvAY%_^B=Nuua+oel3g3pY(zVgSxXGC*XdlyeV{j<`J!-Gos%dxM`_?z}t&*D~zU
z|NmP@{UzdW<4e}Jsh>e05FGcFjqlr7-M5vx<6#T`5JCb%LVN<kd;*dNf|61qqEf<=
qyaEDJ0s>^2V$T2L1J15Cj&{EP`v?5m<452F2z6y`r7{J}(EkJXLX}(q

diff --git a/docs_old/logos/vllm.png b/docs_old/logos/vllm.png
deleted file mode 100644
index 1ead9972879c29e17160f1d3ead150482467b1e8..0000000000000000000000000000000000000000
GIT binary patch
literal 0
HcmV?d00001

literal 89976
zcmeFaWn5Hi_dg6HD4_y^q9QP&gpx}4C?e9`Eh^I8ISL|@2GU_rA|c(3prTSrcS%YO
z-SAv{XwUhb`~UKJ;kiHOb8q(SeZ}haUF+J%S58Lk7{OTrEG(>J;<s)nU||vFV__ZA
z!aEG^)UozI!NS6(Hxdz%GZ7IJu`so;R<zVNxO?r6#qGNaV%M+m^73F|F+4TU(^I&`
z#PC)Bgr45l8pcZm)(#4fANwoldDNo5z(2uj;a}m=VOm<fr-^$juyEMUo^&nxg2H>v
zI`*Kt{6!^UzJh|^bL1Q~yjF&V;j+G-UNu%`NR=c3Vw?)DdGQiOwGP&8a%@VxI~Ld<
z8qSZC>Twb;P-a8~!wuo9hk|QDaqh7r$NMFHGH?;5NhNUjNZ7O37w@oUN*=w6kN2kY
zb%*0gC@@MOvTfY2-Ic%ReAD;A1-7>eCk1Ao&D)asANBV?a~jciTz;LG_*{(tZA0SQ
zc%H<O7e1nT%U&RngGAS>--aUfk;e}m+Enrii#UJip{LNHic;Qh@uGL-3{Ho8+Mnn@
zgmvCkLN!)GxCc<~#$7e>d(zTam%%k277Y6Y)**0(4gQDaV}yl+xyHhx1OJbO1&f32
z+#$@5JG66sNDKPWSYt>HJc4Vaq-LuoEyZVGVaB2lQJ=-p%o2J4OTdv2T$<gr)rUKp
znVQ@1ISSHXp5O!5&}~*4IOY*s6G0j^X*sxvh4o!H7YiE;8;uYF91a(-zH^UH;l|DF
z*TG+cH1}<7E%{hk9UL539N1YbtPNQKt9e=3uCiXe$_$=hwsA7I)pukzx4E!2$j&%7
z?%Ehw8(G>KS(wA2arJLo*x3rw&_EOI{<F1CTcdk>Gnw0LZwqXY75ats3JV+SzrA+X
z(dhqp4f<v4HO#)YrW1e$<CC>Ex(gNzElcQ%0A_*%_x6@20FA;YXXJR-RPBb5*<EuR
z=pA-8HZ}p)fBorSUsd0Gm51xVoBMv*ep7%IBGf*@ZE*#27qCT$K!A0Z#X<x<eZLH`
zun<_{H?Ao;VvlzbW`{Jn!O=FyUz2=*NxyG@b>(BD|1mCcvFv-I?*ghc*p=kHE=F@j
zzioIsWtAh%ZkCwN;dbY$Swn(&Km?a$)nKV3&IPg?@`p)~^?EOED6w6W>xKv6Eq1mI
zKQfV@8gm`)oE2P{djGj_(Y42YZL`d6DsOqR%`jHs+e0jD7!DpW9O3!*;x7^YO2S_)
z_^SnfwcxK7{MCZLTJTp3{%XNrE%>Vif3@JR7W~zMzgqBD3;t@sUoH5n1%I{RuNM5(
zg1=huR}21X!Cx)-|CtsHNX12EIeJSTtY^L^S1WOz5C8QgAd=%=b=cI@#6<LI1l^0O
z!K{jw>AoKIzSi)*u`_+G6y=MCULFP&vmg4F!_f2D==6m}PYE6!eBD*5KCKC-`4_`E
zQ>(E)iX}~Q31>?e8cUmNz0(EhgIY=#sD<5UR%48F+FdfGCbVQzCoIQrOsq=8E10CD
z*Tlrcl;+C398Fgo`xP=J($+1Pm|mt)q8QT>xR_)2eR^ZK<RgQvda91)N8BFAhRAN)
zq*D)%K*iYs5tOT8X;r1{I|h7W_>X*4q21V*7^k6d!&n}x{wPZi#Z~8p)(-}iJ}yyK
zujaGLnQJJ@r%P9h(Q9*x8}nUb-qr@?Wew$mO=0b2Xt$nqouyLyx)z5`^3nxC<KdOE
zsm-(B=X+M?Jtk743Gc2>wi)-deD@NYph}%`XNz(f)OoLxZNGMBxzyz&@suyR*nU%$
zU#BRP(C2cctkIkFAj?vfQv0$c2^acRhNj7rBGDW6tCJ)m{BHBmGAf=An^YPnu1ub<
zSakcaNKNRo+CJ(*HA%Yqy*aR@)Tw5=)L<P~hJPZvTcbZGJygs3+auzRX}NBwG+l7Q
zaiPtyg^JI1BvbK0ns(Jn+Hl{PFFM_xa<XVBwwP;SeSLC(vpBu;+muPm!bZWX%@wVb
z9%1e|XeZz2^PIltl%<bfnrLRpTc4QMZZR2_iJ9_CR_T3F5SK1>KhL8(IV5?oezCY&
zcBz!?d(fTI(V~r(6~4QrqrxgCc^0>Md3&Oqi%%)^yb(@&F1q@?QD&}XG}dCJ&PUPB
zraHOO#6~Jr#w=#kLn=mK`J4E2I?ANaz1Oy-@3DH^!s}cN7x$IegGrIhnz_Zpzo<8t
z-bHyuhO<>%_hP)x7Icpej)a~=1;d7l&O03zG%BTd@dY2qz#SEMwUg>2m*UOPHZ0tQ
zC#Xc>_0b@2`pu>9*gfDC_z{>`5LF5K5cLx3t`8Bl5NCTk#an@^mn}QY^0X<hyRg0K
z!}r@ZRzI#HTS`RA?h1}Q{=<AFX~M^cX5H*m6_A);2Bc7~p;_3ZERY;tJcVb|?e!@{
z*)$oUs@l@i#MNeIl*o4h;fZ(%^CM^7NNHg*x5yA}0#dr&PLO>Q1zT+98S_3k4~Hn+
z)i{;%<%wI5f;O1xlM)D^@!ww5E0sKKQ7So<a0>igPljg`gWF(q96{-pu3P3l{#;hb
zNs>~Tj22Az4%R2lYcW1K853em%$KeimzY@Db7f3LackYg46i+LCq>UG^OEc=1zHnA
zANO*|_pow0`YT{QVz{LO%5_)_^U7@!w&6jOw^IefF&z~T<QE5*&vN}d@mNJ~v_8J9
z%w=)%{%j}@v`Wl4rMOI%b&E^BF$eZVLyA|l>U3T(^GYa0u-jy8J)e1gThoDwqhR)y
zbuIbn+C>ZFi80Q<S6oGzHK0fEEWVuSX=_`ixH80IxIH`;@;07Yua3Xf3*4O<G4tG@
zkz)yEwhX%SK3?1t3mJx6?QkWiv=fk<I1~{!TdWhAbv~BcqB~l@FUhK^EgwG=VG0Wp
zS{bR#HS0K~***g(x3yq~U{8uE*s%nq1Dzjm0;DyWIv4M^3`Uwfx=7<d@bDX^-oj|C
zR@YE0>5~wTn0gf@6{tLMy3#dres@l}4)d$hCcT6Br$Iy9?PcSvso9~M`<6X=R|ber
z(_#~Lis7qd9gG(AIB=NUPajh@I{|T~8K>Gooq}W<dJV-u;HcQb&>wiSp>Ok2e1_#*
zT83w1Yrx?QE_~wqYI+R`(pja`0TyaRjTv|NjwxL}IV;IG=e)hUt8~4>*=Hz-`}S4<
z$pG>dj;rgs9A7OY5eRuOo#+{HC(Lw~u%Y5UuS_TR&reQTv}GCAd3?KP`<ey;zlTH*
zMM{W?Rg&N43g+M81>y=0lO^jHp6?G1a-Rf`|8qg;zjDWumo@g6TmL8D`pyEIscb(N
zp-S0CoF`1^@vORyfe22MS{Ca-YoJM6BR&>E`Sn}*XEE(*5?E*)9A@yv--<O>kE!A{
z3H3xIYKtYbKyB#usSVq3d~%Ok&*n<{CZAnTshe=U8ly2Oc;5+@!`)@;;NX$$#BA)l
zyLf~%OovRiPN(i5f8xxM?UhGcw8e2TVj0YiJ*!=NSSlZSDp(B_6&l7e|D-aIlF`o2
zt}DbD4ud!+j_fLOC>z^Yo$_EO-VtYPm?$y&wIj(*`QrzsAb#!{{ng;CSpkp$xDFV;
z*nrgry!44queNIXagWPXNvOB?q@s1ygCF?B48f?i<=HYvKt<SHGQ-P*=*R3f>R?|O
zqUkW}Ay-ob1$Jq;bzhwv4HIh?yX6W^{NpH?gPkk~W1XFdfx_#*IP5!(?yHCo3=CvF
zYTd36X3v(=+uT@JG`1)-QupDb+2IMC?K|vLe$ds7p#x0uM6lz&YkKFhj1w0k@%i;r
z9P@X8Rd8l7(^V(%FLw?^t^^<e((6NWCHH^6HVHb%I};nx5TOG7BBl538SN&$mftgF
zl^y#G<nu6YF+<G9Ya}Ojgl>ylr=?J0jmPPJVJs1uILcM^?avp4&@)GA^=c_3=&6Ux
z)4s)Ln9S%7T+om|0hhah4ApTSJu*}m>tR4*^PF^NJVY_rO(p-rRr=ixf(zt1g4S?O
zpNDjJAl@Zuf(f|82{0U2#J16qIWozq@8ca$4czgMf^^>TOJFz+8uSklqV?|X5*N+-
zP#LhP?GdvjfbmxmAM@FjG|UCZwYa>zmJ(OebDS7Pw4Gw>x$EM3Lk!|$4Jq3&XWiAB
zm-Z9w${GdMgVYIU9+u)0KURcSUogze%iGk5Y9nVK1JRWA753*Rx3?1~gw6V-WjLN=
zXM0;y-qJ)7p2NQhFK(@5Mn+x+`?w{Kvg5}tU1q_K!2GET<apyW2kiqO0J_I*ro9q4
z5=g+_{pEW5e5f~?COYzhGO}+&6xiC;%mqE9i~ARew|lps-{R=>rXR*0f~fZR8p@Tt
zf}!duv<72fS{|&O6WCmvjbEKA+jJm&7RZ4|{1u6CRIC~8tKp3Oel6+x#a$lA<pIBC
z_dI#`#@0nWDITj<L8xqyE=Eaz&;arBi?b7fNR~_(HmgAWAmQ+j?(Pj=wuE01?MYw2
zn%EwrTFN%&iA=kT+j<C&Ok0I*5d}zR4{YO2<kU2&-~n>opp#2h8lNO@3zG~5k6s}N
zo4S3@ju&VcTm~@z!<Ti?G02<>?0(CWkPuU_i`NK8_L?xK*&#O}zYo6qEDAfYkQR?#
zbM2B1bAj-~*;19}QvnwdicnB+2~6^=FLLTxL9!?iKMve#C!^m%GClg_i)DC(vUR7U
zn{y<<4^#L8>vIh)LJO@1hSLl&2ws6Fp&|Y|!_N~s$``w{bME|okT=^#75*c`Q`Gbr
z>Af^$O(b_{bGq&(VK|`_0tb%_NQiGjeOsH8q5g?yc)LVFVuNi^V|_jkv2{nPfzB#p
z`%!1|UG0Z#1M(i&K}2#g^VTzQO|aRanpG_|Prlry_1{5QNFE&f$<Ad1A^|iC*Xj<V
z<jz<~H4sAt*$oUz@2n5QYNwNEk{l;WUMtA3Ap}7j=$!om@@z-}z~>nO=ZBw4t$uy^
z5Xog0!THt52N<JZHF&X8f`G@yn#X&U6yFXxjO1-moC@rqu7bgL?-oV~qR8<0+Y3Ww
z(GA2pb6NK<ZYzf*i1wtqT|=v-q3yiAq*p7c=wZ7W!|yOTp@05&*v|>kF~~eyqI*C3
zO%_d+I(&Ya=elbT`$>V7cQs%!*xBPQB@bSl7Dd{PHAm;%t$G<t^_hi3+7sag^E<`7
z5@}EwC>rJ@aLI?etS#w6Oy5h$z<`~wc>%Q$>E!`3vVW+n486F4M~^=VhBg~F3W$~P
z+&CC|<mzSMLPk&=7v6g;hk6t+j`vS;F5lBLeu#yv@2l{s)6KVh#a+9kzRT6e*HJaL
zWED)}2iDH;1d*d*$Ey3{;|oj>fAq56PDD=35a!97u6Jl9!d4|+XZ?#=N3OI-jlL|z
zkXw4oGlL$ky(r`jJLNqO63h*()Ues7Pk$t*c#X`E<~>o%i`cTY+(5m%Z%P$ITbHv%
zq7Oe(aB*1;tzem)6!#Nqow)#fQ6HbRjG^H$YoIlE#GVZgJ_BMcc6Kt@`7OhNkpe)F
z3c?o^21^PO7%@h=5{6C92)e(9-a|eGa&^-@q~#f-%6rGz+Gu{(DwDN-Is}{oMkKV<
zYw^fLoa+I}UL=Y<1%lIV*d6VCDsgZ!jJD`uVW+no_f&W*A}A;r4{Su&UGhbmfOlGW
z*(z`?;cjA~WBoxI9Vy*AO9TSP*h@q=x-u4hfXR4`jJ-OgI#&iEIfDlQCJO{cgl!2O
z5*zm=v*yAD)?#!y|6zeqIAHxF616h-^*Zym%)4O<GUTGD#3Hi07YDafGLFl=q7QE3
zHtmrNv-Z@J40*Xlx99F97T@-N!vw6j1keS&{2B-RA4-6hh!tSS=PLCYFE}m_hDs8i
zVp^(Ca(D|2J^?-PMv!pAn>W8~yd|Zs?^%jRBpE2T)9}Py_1N3!0m>mRz!tk4dIpai
zKbN2Ml&g^z0T+TS)aKlyV-&UitTtzMVjR3rEWr7U3uDh5SUpk;-%C5PvufD*6AnhQ
z#6fT?-q;9<08)SeSPdTIZ?}c42X2`Gt17Fa36%RbSyBy&wzjszr?8@^y%jpjch(&w
z{(+@#{dD>+Qq=F1cAjBf%(pk40U#b?c#K+@=wS6&>hq~%zQtF+t5tBhP{^NIU6MMe
zxb-KKlr9h2UifM_)npu^DdZ#5h=X?nD3>G`dZ_4*xiJXR;nKkAM69Gvt@eI=7!PcP
zTf)5H*i$TIF}_z<M^#r>moC9(Wu4|oznNUzIjUn^ulTlYLmV-PZOCt*rrk-y_ttYz
zP}RxRGW0%<XIM4`R;yDpOq1MFhr6~{bcC{KuUxp~^U7t<7$oDX^~Rhn3EZO<L>a_7
ztWVT(ab`X%88a6zDO=B=Das4O24CRJd`GP{uP|vaGi&9egKQ!_Q6PcwMfHF7+gun7
z)ioE_<=a!#U{PQm46T$t-8e|Bf5?qD*~stTE3F&rtMsR?@`{rH&xU$>LCoFq=WEdx
zuFl}Ydv6S=ff(!@Q4ENm_YnaufwX?1&K+hOuh7>V_v$wg+WQ|o$LwL2v)V8YW$=1%
zDTr9tuiSUS0k)q>aAo98S=UUU-7QZ93(iX>^%;7f<CNb;-#^%kT-Y?A@c89>`vWum
z%T7>OFiJ|w*zt)8N@3I`AE0V#aSNlzqoboY1#iQWcZHpJgfs)yMFiV#@IlNG>glrV
z8W4IPng$;3xy%wCsR1cnurLspjW6#nkyv!(rp}=&h&G)S&o%&W?oFarWItgs(VmmS
zwtVZYJJLAkcZ2qCL&fd04-9dh!eL~Va`FF*?ro>SKz1GLt=-uZJi#C;=7~tbF>ZN?
z9zDL9qP#x37v~~gLioVU$m4*6%2a$42x>a-z$hNhxGM+jqA0ZP^T60=eq`&RwNZo;
z03)EV^^LM+Qx<Mxn(<=PZgUq##b{tKoJ=^n)gWlSyg5}fi}=auJs!V>tio>*IM$xD
z?pdL|$A}-x$CZ?<r&pxa7#LJ**W%!<`5^L-M>RJ$FSKl0eR9})ssB_4N^o*I^k96r
zUxFaMUc&d1?Lj-6Ni3+3+Jl_jfZAhix=r0z>>M!V7tjOGWj9(Z1$iz+ch7?mZkvs8
z7ubg0vyv+$BnSC=S_#$fbIm9Ul2RlLKWvJCBTQjiHy)kLU!5!}J8f@4#cp~H7mt|X
zEn;as-AmRcza;pZ_&al?wPWK>o(hx_8&IFMH%?+NAl;hXF$WcP#h{pN&iF(fM3*>E
z;7nl0-@SnT0N}?l3a+U68VmQ?b@_{oAP~@@M|axPX%{%p|K={8^g6UhAR353{U1Yj
zEY8;U24vx;y}xvXJSymJL%c{*1o;zvSU>b$uqF^sbDy+PWL-4>Vp=;(t+a*OLjde)
zz}LD%iwX-T@pRcK6MOggm92p~W}WhR9+MaCO}+QP4nJf_90#6?(Fx90v#Z8hf6W7y
z6rSNQ)h+pwRoC4LCs|@I_(Ny|C2jc}&vuab`%J?&qwc36%oc|h78Y1r>}|}=t28bk
z;8ieSb7O_w*BoDpkWuW8fK&o8rmVMi;=wh-KX{TClKxOmQ8vf##JJEWu(S?*^dBA-
z&Q9`PtH62&@*j0l2oSgEDd7g5<Gt!M-T%2kEQd5;&meqF2F`?+l1G2<)<0Jc?jWpi
zoSmH|fR)+2L%!JWpDYE^W6P<M#SJI50U;%lVHyCtY306TEVl~1y0s6qGxp{s=Mvuw
z2RE@RwmW3;_G~fHd=>RZ2!tK>)1_{!2}2BbUY&pkBjCkP?<xRT)aQhf@*doYAOqrl
z;xn7%vmFJ%j6`q|gfxhUmmo-u7>K?rH$wpF*3^inxPjEdI>sg@MIhc2rMg={!Vm(P
z+;4&|3vv7Qt4EB@dg3VT*PpzUJBI`%2NoHwt6S=G2n2hF?X6y57BC<UXL2nW2OgDt
zklB3t1f1<E2rpk+0a}%DX4XsW!KIHCf#=|<)NcMyio9i#J0A2x<cHf~+(0Tpjo-j8
zxgHiF1+axQv0B!>8nM##2@CF5w9;hqsLf+fc-~oWhr`c65E*krn)0B{3V!BE&ZF?4
zg&E>XHlRz3f_e1zn$D1^6+2Gn=)YPey359K9(X-ZL~DtgDF7S|Dehla-Wy3Cj7{|q
zn*URAW@s!&e!QxGfJ?~qpr*>x_c;m=m&?4<@fWGAz!iJ1&nNf5B;r&cdOwesd^+ge
z5CX6!<D7G{xmPhMni@c4KLi#RLl4Cz!Ss=<%=&r5)C|V%Gc?4Jh$Xi)fWGxmX?E`+
zvGCv<fX)|<fA47X)-sVVaLUU$7iTWDjlF~zr$Mh*Y7(6j1N{ou2BxHNLw(9^s=FjU
z<McvZCRO4RkfU%E@D1r2fb3pRLsXvZN*tmS0DM6%Yj<V;l;?fo1pRd8WSFqWhFE!8
z9Vz?Sm<%xFE5wo@i-Wy=_m%SgC;Nm$2P}HT0|)Eis&{R|&IRHWu8w1$cReVM<;dQ<
z!gYlN&Ss4y*LAh}y}sBZe0PBJg}*ZJi64eoEVlWFN(Lq>_Fdn;*b9=tU@Ihg<q?3{
z43v_UirHTL2lg2oTMF!{JOB8GtGzqTDUd>@pTvv(g8mFZqZX=#s9m~XA+O;SGYH}N
zSNx}pLnJptT6nB>#GhnTS-aV8H4=-3yaQWZm?&a%Jy8)vpOOj{EVenv1LgHLop%rB
z_5YlKmt<soe61nFZ<WIOL>e1DF^?#m$$T^AjaX2O^(SD(cWHZ+9$=f6*K7aV?CII!
ze1FsMUd4|t0|d+U7lH(4e!d>0rG|-;<5@^K{Q_cs@1&sY@*WrO=K`UwIn_k9yLimS
zX>EXSoLG89PFg@~xsB4~o71DXduX>n8&HzG6$cU?iwai{=fLZyE+a>9PqFH(eOq5$
z6q~}*8vshtw2uB-@M1eWBwzv&3i61m@xD#%?rz_O+)-pDwMX%c#pZ%VmfzC=COk4-
zc=7qWt{~Ah`&_5=Xz%?9RxsqObk42+6lvsBJjW9IixS_9X^8EuHY+pkFJ!A8UwVXv
ztR@hc`yo;`w>Dd6$R&i_Bg)rI5DngOvi)@PKSu_Xk|B7nnKxNYAL}}^jsTHqF2eD`
z_tA-o36Bs=bhsmD@FF15h7P7T{uuG_2}@St;O}gemqR3+!f0e5_9AjY?uP%RVyE5$
z1YZPMU^DE_%M>%e8GkM?I`|h!YrfSl=AM=2ytSAw>;o9ngJInLOwUaz?lVd@dt}-Z
zqJ3mW=w;(HzgK7P3pb1@xR>4D0;*L3_+H#`{`f_C9v?oaGzR{^DdoXz3YrzDX>M+=
zN3NgCmm}hiKM4fp8x>X40hz|{V_%fnrDz;JAPG0N{I<{Ap;VkQOsCj9B-<TYHpu^G
zwIfzmK=}*{sRE*6jb!cjrBu!%uLyH3dfc5pWIj8FTqE=vb1ulrnhjhtq`|<iTPAIk
z0JxX5vhIWC8Vg_-b8~Y`6t@VY>@W%2C_wM(=UcXFJNM%O+QXa!&}n-s*1SwzH(;%@
zQGt~F-8t5S-?@pjzpL+PIiU9=Qgr|1|I|vbc7J>(-O@EP*3zZ^_%no(JUFjUCYAKP
zEh<`FXs3`9*vF##QoyW!S!;y0jN#}X_Ia1D%;Y>E3`++4mMO~5i_-DIISfJ?9;dNq
zw36<osOQI_n%edd|3?bGZ`$o=GtbLuO1S`!n0NQWS$X2eno#Z2W43F`o!cCzjT`dx
zk0^fbEG!NSu#jAjIDU5`%*8YwD95o{UOa7XzKxtfVc$1&!ESn!Z%>r&(dSM6@<9U{
z_X4xkc<ToU-6Fprt=o=s^#XDCee20`Z^E9~aL=p6Uq6%UMj;uzb|(o|#rDAeI`-8!
za%Uks``mp5bAe0F12A_<Rb4l`k6ETnC7_nH_#2qk4LO6<Bmg*GfC0E`>EXj8L|pYZ
zUi}6&lK6Os;$M{B-^H))k~#Hc2C5{d+j%bzI5^4ngGfG{do>X`se%1JPr8l9he`lo
z#e-bC{{&qy39`GpJBy!>&z<u^-yIr%O^cHG`*bi8oE`Qe{7sHz?|ciu97-D}e&FwO
zQ9IA=-|c5|$1%;Gl`2&&tIw1Dh7kf4-h||}c!XVblHCs?K%xGmV=wZE`@A=~&Ml1+
z%lYnw+&tw(>Bt4f0|*r}w7d9DsDQtJbYKZ6Uv*Z|!vUB)h|>gx?pPoksdo6*TjTJp
zvk?SvhzNj&k&Kl&Rx#(T<&lI2X7mCuLQ?W2)AB5yV(3+*5^lNOnm<S%WaH1?_`*j7
z%nFlN_9X*-{M{rOZU};sx;9UIsMfGc@qb=C#h-(iQ<W$a?aYx`JSLF`6sJP}!^GRd
z%K%nFh`>x*(xPYQ8e}zHY!s|!2EN6QHYD~?#w>8*P&nMBu=^yqk-3eR$q+2b=2CCf
zGY@Wh|GZzb5d~U8S6mjxgYNNteetj($3)c#WFuvv)~5^Ui31RhBHzMlYisMoo}=+j
z9E;7JF6wihsJV3^L&0ljFM9U@4(qdEg$dsNscO4-uEN>O+OumfDJCef6HD9^0l-cn
zGgF};e-<`@KLUwUx?E$SQv?A2EP@vbBFV+0(iWGsJ=W(>t#Q9i+Rj}7Kgxh3Uq4a0
zPwO$;^M+tHq4vCtH;kBuhY@nNo_Wb7Og1ih60ZnWCSX__)x%`$Uh~xigfsDjRiQo2
z{<+!LJW;){zJQB^fQ9@=SV?-q!`Tft-1240$aIIU?wY65$Dqo-h;rn9^6gok9TK0`
zL5<~9P?Ws;G&kBX1=YfI?W&)v6586^Z^)~;x(uZRIp2f|srKWo(*`LjsfBUjW<v$3
zHnqHowzD<dn^OYAaQhu0L7L((JrvLVBEoTSRy`i<#hD6xUQCcHO3c=CQlD&aYY8{6
z!N-rg-dPxLi#_X|K-e(AUMHYPKxnZk2Gx6jfo`<7fEp(#`%J#jNU}zqVZT>R*fQ)k
zDk>KhQ&AVP+Sh+HQEH26$ip~miAmT3%73_)KZQ6h2d1jyG-yaK9nsh>W7DkJ`gwix
zk?6C;VJ6eQ6JNbiY1&0B8o8!zK|?gHr*L@9JM-HoCpNQiwMTdR3g9$=z8B7#CLd%h
zC><H|n|zKa!Xz4SD&Qvt444%Gfj+UYS@nNfXE&@}lNdiPGPwDPR91=0v=#mF)ps6%
zS6xM>yU(Y@@&o6Q`5pJY-THh>k4Lk*;RJ)8vBGYl0u)JMi)7Pw|0{y6Aebx7{dA1G
z6V0!GZP0|s#r+8avg8#My)yiR!4?4Pl<SQT)#0-M+9TYg$L^gh1Xx@j#cShk<>{s4
z&_xyJFS5tQ!@wZ9$(^O&)sBC^fcwFIRR($JfAK$~TXox1-S^UXultvwnn=Stv{Y8a
z3=dGi3vOIwvSz)0S^8U_+muVso8c&%+UD)r?sg=U2(m<Xf=6JL|CHE$ci_}8&@|DI
z9_79=x{Sva&oEWGI>`@Ga|4CnI(aw*v=G!{Q&?1Vq&>A*&MA;19Aw{xgD*9OG+06>
znuQdyNnQ}f43dy3?idF-N5EA=;ZG;_z|$=;EoK`q#F*mSKzcv3eYd`dh`SF<>k0@M
z@xmv~)pqEz8>29ZZ_Q~*)8^N&ddV!TKF)MkERGX4)||RJuU%xHlaO_8&bmoiRZ!4r
zX24<qu5^A^bOz)ASI-@f2tKf3(0AcS-B0Nf4x&#u9grlcH_WOSRb!(;Vmll;zG*m8
zGW~@@SgS*{)xZ)H-5oE=7dadC^$e_nz^msKcQ&ZXf6KtVa2frP$Q><`Glqs=BC>Z4
z?mxblusG&;Mz*TDIQ|g=|AWE2BuNugt~H!30+G)x74dDF-YE|YE_T`o$4CtG(aI4c
z#i*PHL^^WK7;kX}BwhiwlN_r3RGzgEC)_7U0G@NxBNscc(*sTp^DO)qET`)suz<_2
z{C&cW9YHEjW*?I_(XS)JNgPf3H5swg#hVHeomt*v9!ljgf-d%EiwnP#IMU>Ij1e~M
z8Vcqhe0Xne%$5*+xW$ikm(IV0LuLU7dp4SSt2aQo_q893^V}20v{nj}jSJsgK+BP0
zjK{{ZL*U!@An+*G=<V5D@A3FOT}dtct4Zva+d{2F=`6Z2S&{m&s-7)7k%h#}y|+JU
zo429IU{WAVHdp1!{}iT!q(!`im52;#cZfRHM)BHPQNu<*`I5#@PENAAh;$O)k|8;b
zfUngmL2V+#Sc$I%A+zhjqvhc82>bL*p`euZzA{$p$FBcbBpT4{eJ07GPGwIf6fwng
zlY1pzGIV1EP>_YHOD9_<P7S7+XC9pkkX0MM5ak(D_az^Xr6o`jDP?maG~Sh2fN6MA
zc8U$XCHvF7{=v;VO7L+y*W19tQ`g*g496QQ>RC4j<*p8P5`Q?}03vR|_eh_^P}(`K
z_DtdmyM@q-`31)+zQ$b?8j%gcSuy^(69-Cb*f4o~kS!=xFhi054Mx3EKl>3{8BatY
zu2xYbk0U_6$D*#@mB{gxK9!aX#h4`!Dv)96EB+Y8fCoa`U;V(RZVt=e=L?mVi4iE6
z>s~BfoYXqZdVfcRke!e}n|vgNIv_;E1)$LJ)dTA-@Mp+dApClIG@An>&;X@!7E@pu
zzq~dBSyG#ecMX_r2EIP6e=X{_!PUGrC>=+IL_Z2H6SnHVA*f8Cz0j;_v3?`JmWt8#
z*|Xhf@2kHjIsVmumWmIo0vIObO`HVTSoaoOa;+b?NWIfno6_CwzCcBER@c`jF8=UD
zO?DM|SVAVg4MFPt&Uvz6_+|I{q{F5NrFg(dfVf>Yx%5vo-2-cmZxUN5<o{XFKPa>y
zm|7;m8-o$mMR1NaEc@oqKc(Waj3YZWxgk2{(tH8Ah6iO<iaD|Z=%pA`bvXMb3zM~n
z?_R$xQU#Rlx-T28|7qis@#hv6+LswP-6%0>cM2eI>a%7dNX+hN_X<uq)fv)BN4%J4
zQw*^9_yAYf0kk|!$#UWWvyg;>DsQ#1!kZOmj-?OkONX(&B0H78sb9t4e11nx5si>M
z39dc+L!SPiE@BARdCb=m^t7#3+^$Y<RHV<%EOMA)N!2QtA-Y|qS&DO$2L1IyP-iai
zGEd0@URV2{VIO)VTiL2=m}#e-H)g77wJS(?0EiFL3y%LEm`E6od4Auc9^SF{S+^w$
zCbjta`8_VU!m1aWwB>POxTo;SNDXs=O?O8szGut<v1f>R;b=m~?{{zapk^c;)C7QY
zL;6p`9^7A?4l*OSek^DHbwI}2()2G~yrSc*l={;Qeu7X(6i^M#{Ii22lFn>MA8Pld
z;j-E2siXFMg}?f0u1i+T+Zx*XCWcwIGM7c&UM$=RJMY{?QA_=c;8qZ-t+5sVM9^Li
zc|!i^@B=G<AV6nMg321j>77nIL|i^hzojoI8hG>1(i!;S-qG(i+17f-s_8i5@nHqz
zcx8l%cDA;zEQ`JzmyqrAHUKE0Hc_yN*sIh2zb4eor$(gfdv-nd^TvSqS^&^W>R0~O
zeL}12bmFuG)XF;5Q|vU``s({9+sSs5bZy<DK)Pz)pLYEEy6<*TDKf}+?^LS(Fn#+t
zKwJbemr0DfJ|{GI(f-N#St5QF53X@7vy|2IAyG_)h$YIT?3|n~^*I5t2;GT7UXbO)
zrfZ9dX^8Re$PwtmBeSIMk&-ssoqyjPTcOI9m%E2UCmXXCV3Jla)K-o2;v=FwuqmXH
z%4J;S%u(rEI2v^5YW6y<GwwumIqxg{9XQ-p<`V1pr;6#XQYD~wb^ZTB2T_ZlUCcOq
ztMi1R&~rI$<N}j0w8~pFavg3<Ly;C6Z5Qnf-jM=m?|SRUmFzI<myAevBsxp?c-bL`
zdEvzzr3erUADQTmsg=E+pPu!Sk<z(j+q;}0MaRDOKla#icOn0N0orC26QtB_KA%(_
zH`DR^Y+?jXkr=wJkLi|ua18(U7=?y`8hP=+$0O9P9cbEL3?ylxtfhm@7V6|;xtY6e
z`RhB7qp4G#vHD*`t$^v1_Zzncw6aWA+yd=5i)k6p=fm`$@Jh#7!e)o-Hm}Rd%J$7j
zMEv;H)sZ4Tma<nMj5`CtY9jI!2et4Y0u!u`lw>Km(*~l7pnF`Xqd2+l>nmA#rfoB&
zN4;opV9caE)&pM$K+$!ogn5kZ7iGz6*Obt?v{0gZ0#bdu8@K@n7%a|1Ceb~~fhGo0
z7_V@t(j4vq-DfBTj9W5b62UNz1;DVvB=)o$p9h#f$<M~POfE~be(vIX!1GA+<+V7F
z+Oq~x@_*@ni%QMwUa{$OHsW^?CfVIiGGYD|X*WEaeL}dCaUa;E2;SUrEAvl@EJfp)
zflzWSrggsjL5ga+&?0E?FS=UZ7=P{+3VG-V9n_d{h$KNs?>hwW2l-HTk{@%-+;}Km
zVBXS`dUkhaQP9m(!f@dd(ZRj`V<D$OAR@6gGv1bEL6PlA8lU0flAM~Fn(J32dwq6R
z&JhWu1nsam)uX!vG_aRBuXAR)JyF{~yV&~81>!HLc?aA3P{$+WN5JU|*kaVNFw6fH
z$A+V1*FMR1T7ASNznF4PJoN~bV=ptm{kWs*#Zd`%1tgj-&<bEuWt2uM>5{9U!fiA9
zE+wsS*pB|(UZw+TZ@80}K5o^U_Nl`CXn=~&a|btDjbRGv3Mbfl;&IY7!2&txdUfKj
zloguz1Xyi3X**T_BuGAux{IBl;_KtXg5^r8X<Exq(KiF}*r|C2Gh9~b%jc(CQHGRI
z&1}nG{&2?jU^6NypgOoyfYMD4Qv6iJKtwY%dZkrWw|t*t-pRfhmlxjEX<2bBHk`vq
zfjyE*OS3AJHMS6RCFPN`T)`*7gZ0%=7<6=W2owHtfelAH=!_PQrLFqVxp2%Eb`7+0
z?CMSMAy7#j&<pnZZ^sGjA_-m**;FjA&5&wDgT@t2C#V+;G_L4gzGt8LL|botWu99I
zAl?QB4kLA&36bV2Qu!>UjhmfY85L%`>cpURfbb3c@c$5``{4+H7?6TY*FxRHX?z4^
zoi24h--xkR8;swkV+#Yd&yxC|Kdn~vh7I_IK-RCI*1IwMVK92owIcG^clAJ7({17X
zk08iVpMlePD*_?}wM*tt&6tkr0Uc0@>fr^a4dM%A?~p;(;1-bLiP<-;-U;~GT6&<z
zb*d{vzryRKL)6dX9z%xiLot@rA6WMtGD5&_L*NCK>%NYjJz)b!1vU+*gir(TY9CJb
zbSa0bFkU>515@GXI8BcIg`-2SknJWs^&;v+Ky;l|P3Tl?{^vl+6P5P2`F#y2D(xAp
z7_Mu>Z;(EwalSz+N_PI;*`&D4ckHvm51%r7Wji=<^4nC&)>XXea_O<DUEQe7aIl*S
zXWyX*8QzORn#v55X>ymjk)J2E4HN22(n}Vq=f_&6qQf7MJu^W9we14dk+p`i-2U_8
zL<ctkVP{!k*DDG4Hd^64#_+h(T@85#P9C^q4vX3E+KPX__<;+PmX@vq-T&4)<hIe_
zEFg=g@D`mE{w8&e;9CwV^KMnQO(mtnxnL8oi6@(_l|uF+%MDDq`~$l^!`LuiV1@**
zD4vIU)c@TnZ)sH0O9W8(OXE6g;VaHTH62%Dxy;(>%WGR6wm-27M4ZI|!R=hp?GIs(
zl!^o@L0)!~nVDI4d<J=RS0@p1&tV(!-Qqkof6RoO!+Rn8fKRaP`&Kk>D}$O(voY6z
z`~spBcL6jrR?6^=z7<_oxnHyr!l-()Ma!mUu_Ms>B_b17B~_!zwc;4}x#)4Gl+Z4t
z#OMW-C-xzor0&Nzjnfg1vmvWt6Q56PE9V^zU;?FN)BFEe!%mpJZ9jsYkqWBHa^0WZ
zs>7~$G%K$g#`k=CPVYfm9G>Q(@>^bP1E)Na!@Ur+p{|`Qeg1^v?Ah1gz@(1a#&X{C
zh7i8S5I5D!VtZM_MW}5KM>eqWzo$jSncyDh8;NrX5?p8v(dL&ZbkQxeHL`7x=5}N5
zQvNEG4l`+fe?|Z((#zuJGJ`Jq7{`8g5uf93)|J0X_cDXvi$GPLb5L0C1(|zn*s+n|
z4)Ps=m$ttjunBp7F8(VZh2LZu@9(Ckt2To9l7@i>k-X+K?a2IM<EL^<PUPC)C~53S
ze^-Zze8iPq%X*a#$m<t6CEh<I2-SgGE>cvOM{TzSRNaw8&9i9ECZB~wEh!BJGV005
z9+177B~2!co}B=d8UM-rUxBO%j3#YE+EB~$dV*B=%;sR1zUF%6Ih*SG^@R?L?%b*M
zJs=SdRrl%b?hX9W$^8#HE0f?^kae5|k)4+EVaaJYDCEqSd1-gLB;?~H4@>p_6r*1E
z%lIm2s618JIg_aD73=U>TZ~811c_#E)=1NkeK5VXW8@9IrFaIC1~_z#@;_B39wjeE
z$+;Z&@OUOVD?r>49I*WjGWu@WT#|k=<R?5Y?{YypnW&}bpDrr>x^dJ0^an0ijn8JB
zG8O2(k=tnjb&JAqc7vpUpAvKkmZyD*ET8=1@qF+sqkP9(@#icAgg!;8uUXxQM<kE{
z1lpto$AxIdZg<}hnkt?ZqI9o4Sy->tvp3ge+FtTy3Gv^%HI;2qx%D35=r~Mu>$aiT
zZRlyaOV}+%JI~E9S@!$VZ<xorEqr6YW;y{P%Bc;IVMb9<FnA)4;g(0Tou!isL+f%%
zotGD(2IW@GK%V}`;l2_QdxbHac4D+;-m!Cu|4C8M1TzGSH@-{U&6exlnQtZhuG!Iu
za`-!Adhy$0PoC7}D*jcNtibdqUvNnAe$TjSdno}pP`0{^o;8zPea(16IVZN`<Tn-i
z-7WiqmQX<}MVj;fi<coNP}^Q+J#^&w`zCbq>BUz*RKHh2jVSizlRLEG{NP-lUm=JK
zwIYhj9pusA1lSuQw-FK5iT><;y&xwp0bbH1iYn~R|L+%agvgAl>EFOP>+1M@Xy6z+
zsj4--ni1R@@sLAWL&n+-R6B20NAo+BMZKWVytkzEB;7mKgu>tu?#1hIYD~?vG$i6X
zVq(nHOC8A>gH^GIodRGs05+%{oFJ%lR_NEqa`O7HvVA?<m^YD1_%Gf?PgDvmd`Jgw
zEWN!V!MGUUQ4VMT3JJY<_m(!&2|43dm8_Vkb8e&7HFe?=)cf@FQexE1*utkLdxqy<
zZ4sgz_dC&JbEn(0Bp!lRV<PO-%uL4JS*0%JiZ7{<+&JIUwaGBFfset`9Mog==P}NJ
z6Sus7A>I&(+I!=`@Z1mI2_Mr-g#k0&RaVfz<#eXj0T_)V)iHvr=%b;nuj&D!KU=52
zxG+y?tQ4*{#g^p%MrTLD3P5o{Fi$bvazE>{b>r6JtS8yy4N8ULm#RaE8tXK(K1{9;
zHXk|eYi{nrO0#SBcE%Yc;Q@GZSv@%BId@U(p_E10`oa*c<jZsN4__S<*Gq&!&Z-XN
zLYWOVPW(KUy{HjTL|GdyZc5eJnHp&db*}|DDeaw#01m0}MqF`*TleewD?Q=zTDGHI
z-YzO>td!YiaSs>$Uw@(cY12`<$o+@sLck?4VL`0bmou}ON{aC&mrnT&YIkmSuuO)$
zY|=Hv0T!Kq07S0*DePuht@j#}<U^01X46wXWumxunr{KB9B-UU!5n_s*(&8J2H_X3
zrsAVX)LN=qjz1~AU0xRJ7<YeoF5=T%Lr85=<OI2);2TJ^#=O<2c+_dkacP7ouGZA$
z@5$_#?Gk&Nm?X!?53<kCQ>En1TSHz(pHVeUAUpN+FrJc#9so#sG|SxGI3G!@odyxG
zU#-yUxCO0{JpQB<%B?y&nP(gHACN(neX`*mB-jCRAjw6g;eIFVh*j6&TchVTcbn-k
zzudXG_M{%W=!uZ+R6VP`UF|NwwL>}tF0RDM@e~p5iJTUlplI8AaYFt^2E&h<xu(8p
zr7OQBtyOZZN$H3uO}Jztxh5p>B~$K=M%gv1vsch8-sO};?L%7ap*F`ZoW=tscXeV0
zR#Ojs(a{Xu?!x(})VjKk=;XV0Q>H0t=6$-)SBdWDmq5GUQQDmgBXWkRej)IE3E9Mw
zz_i-Um9cbT0C6`Rv7hHzq0eAMQdksEOM=FO)}~P{%BiLlRSxI5k@Uv=k%*Ej$zSX!
zcF&W?T?c+r6;FYCzs0$gf|_ZO^jR}E?~Cz>Qg<Bh3mxn3h~Ycl#s8$-=2>V5DdVH<
zfM_>-2)>7W2tt$?0}!I*aJ2krDzNCxzY@V`_ghq3FmMBZTncX5{nG{jkiy+<1p-vy
zGYk!r?ciyhuLcvwyPWzO2$I9LaLJJUTaFWfQE<IjC|fJ-A1{lPcU&nQe*K0&GrqQH
zA&?4>Kg_JF%@Z8!`aKU$ImcvK_LV0@M>{~E-2P3$e_TTbcIJ-gfzNRCQQev@o~_OD
zb2-`s&|FyyGc(Flq(dCYCxo32J#N3CMt~=akzbA%*^Qlw233<-NrOq~eEs&A86{wq
z9VSZ=`{&>#R#UxvELtF)XCgh3OeZWW&e8f6Y0zULN)3n4flB&`Z{79fR~JEioUIS)
zXbqr{m}g#jx`#9=1o1)5mgud@(icp3=R5rTZsmY>a9|MAl_f$XIG3<kGUY~Lt>#=L
zsZ(Tc#sS8N%_%}6U?++37B5mjel(K6HpG`jXys>+5a<0Lak$|OH}~+2f>UTq?v{ph
z$9*K3xtNilbk?@(g7@aekg_*zi?E7-q0K9sro~Ar(bc2x%RDs4lv7lWIRc!0V=y@}
zMxzMy`%RTOtwjK=*3R0?OmOud%`w48ku&sd0o1}9tk0fGTW5hj_%BNty^faF>@U=t
zaX?Y+sUj6<-|tpWNnxM&idB?f%Jvl|^l-Q>`MlP@^<Lz~fgkrChzp)^f22_r()jHu
z`4$)Dl(EamltVi7_Ompp6p??vdvl0BZ=-UN?vzgCish&xMeexH@hM6bay!4*-TJC&
z4#EHdCJhB2!Y~Heq;AKn7D@Z1?)`0#+m4O!oL!?lDRdO=_B+3k{(*{{c#wVk;OFyG
z5OO_)$V91d>obAEc;~<dSSRv&+-Dv4#PsU}2zLt;<=ndkN9fVR`mkkVi%y*?-3++o
z-9q0<y*?Nbz>oj?)2oL^EPNqB(*)J~kV<azP9A$h?YZ%`+0tlA2Y*B=PI;}eUC!AK
z%t3fWoCa`sP0zTFL-ZI>hHIH{#{$9hc;2$op}7cB$TH$dNzNQ@opZ=q0xy>rVx7P%
zi|}^{nM{-?N@<;Tmuce_G$Bo)+Li!>3}neiKQZoOA#$9+>&Rol;S#wO58ztVJic|-
zFRHvxH%)cREk2P08t$5mtd`JyRUHJD&^H_Q^8y}x6G`N;R8iIi+;DxEbCdP)*adQ^
zg+w>K-ShDGctrjY4Q~B%oLHLWp>cr0_oehK*E)2+tRjE-de?bgg(4xkL}mGXLl4Nq
z1sX>e2A<V>Z*rvMFTCy`*vuW6|IT=Hq=2{mP0P#t!1Sa1xXbyIJha2RBW;;VWFx+p
z3sourWx9hDLLEa$V+5RL^1yk8?)2{!XIaEx>A;MC%<Fa@bFopfj*NdK8@lB0l4*8=
z7kv0)YlFd8LC&PerC+~?A^~p_2-DYhsmMwXsa+qH_U7y>lO!)u`CV!t6<=y$-cH=1
zEFw~|`ROqJek1}j;9I3$=yc6D017HbxY^aagFdd;%d^9q(sno5C8%%{WTIF6uD=Sd
ztnX2o)%T&yHTv|JguTbU9H~U$#eKOk{;HMuPRMo{YIpU%X15D4L8+TDzL)pJe0s48
z-(aW>L6wKCw*K!2@$yq?R7C~Joc52w@%cKxV<!-C-(iD6scAF8kG3jx%H)<MDSQRk
z5C!3hutSHBsx{Tu&#Xqh?3v62%_Vc|=vDD_dv4-eanVp800kdNIU738#4Klk4A=eq
zd*AMe<?APRp{-y>KsOuCt|$BG1}B6KdP_>`vQpYich=AL6s1|tR<MdHcU(2?`4V8|
z6$wrv=3Zs_>U2=FK)QbfM}qQks>I0Grx(h8g0Hl^^ABmhc8W#obeBz?V63dk#*9oF
z3Y4teGxSB)=%wlM9(=!R51P|%0-&Q)Ib=*;c2A<7od;yQI&|&S-YsxA>?Cl-x&`J9
zF6sH{OG^%hWWekd^t(5?78a@Y1jzSHnt6;{e4WnI&x<ooI5|eQBOkj?LCy%7k?%Ij
z3tB)7;#X@lJV@zf)><sflEj*SE67gZsHADNL6r&NbnV06&N=}ALHF3A0XHpU3U{Ve
z5D)S>4egF@MnETB8ce^1>?s+tlfbl7jqm>Vhu$uwf+c-Fd1XV6GdK25qtF}EMKtpx
zKMii1(~I}*-xGwBq~Y8QI6EH?ke8u%LxHDOmwpAc2(O_l$ZfNu9brW$pc7A<Q@@*1
z9PId1D;L4Z`;kUP*>$_LCn`)k8jdgr>#3_zV%({!LUP7Ike&PTQH|@Jv3cES4abF*
z%GYnW^GcA;p)d126irMwfdn~naNWW}w&Ud=QJ`>m1Jy$ib|B_^H)h5GZS`vbLQ%2d
zH8@gOTg5A4x(wysqTocn`3EEn59Glu^6B6^N9!U|YUVO|+1YCh<K0vq4P4wbZ44aM
zM%rAL<#*y@y+xeP@oquk)cyMZgBI4OUlZ3xND%6=+FoAOKowZqnh`Jg$jBklPX1DR
zV$YZSr4W$1X;F|52<&Wjhvab&VVp}$MN3C3&VH%~ZcM1*#}8uas9%o(wll8-j&rv0
z(#bh{<3P}aER#-g{`HdhqDV_>E<RB2+@J-h%x;hc1)ofoJvjk8$RK8}q(X#RYqw)d
zmgeR!3Ppl+Z9CmjQj^;)FI4G2udi4<cIL=lXW|uTTo+ybXl{4)D(2fo`<+YN0?Y+O
zrUvvNGJV}{_-y=WT7Lf%zX~_jG16y((!RAU#qNdDQ9RSCzA~9lzD2uC7KD1cd`L;*
zxg+Ld9h?8=cK<H9x@mf;gDv{d$M%!K_yRH2k2!bra=k+-`<slRXGO2ev_6z_IgJ}P
z4kawf7EZKf4f|{7I2;ISAqfl?Mb1R7gD*v`e`9J^`4$14;l~Cc8ejY~x8IDqFmPtz
z>ls<$b=x{u<&s%r3JQuU(7I_eydSF%hMNAet^~c?XG7ufklhWc)Vbuke5B>1hKHR?
zAN9-PnCk0~Kt<)(1_w7c>tjEw7~nmBkByBPbLLuMwus2V0b)mMb`GOqXqBy4^;aOj
z`JiUdin$mPoQ{a^(+T4w;4fbiQi)qD1Rd(SXDPy-djg%l2hMrF9DJ*Fceh_;KnW#G
zWy3f9eJ>_N#O1>JUP<M7OXlgIvm*V*Hq7itL+jF$%!JGWx=Tt=9~&)7P7R(Z5Y?~D
zTvA3`mGJ!In9QwkL_WR3Ibru=?M{Osv&qhaRUlbQtzm45_{Gw_Ksg0hkVmLaujP{R
z3j6?NrV((gVAJ}zr0QOaW-X+vZC;G~+Fg<D1TP#YG-jMft97d{y<h7K_<2&6<&s^x
z+0E%UL>9&Y?H?BB>wdYM_{TZsuL&rCXgD{l_H1snQo=IT-B#aSKIQjQSykK4C|`I;
zvAJE0?lRsBUDoIHvbip+k?D^b09(_5lPMed{8Sq`!|pY%l6ZB;u&v^*DKxq>KHYDn
zMzRsh*Q#eY<B7Cg4W~n<q*%2Uze>s*B=V<NR}6t<d{WlH;MXy$EB`!~l6e`F`Fb$m
zoyu5>k~ZYX&qFT#vBZ30>^LEldgg$3UYC6bf0nM1t9MtSo!`wMs)=DKfe^gyC1D|t
zLOC{4TxtIS;BOdjkdwyI|77)v?0VN+dGsRF<zvjW++ET4xQdM`WP&U_EJ4QQm%G{B
zg7b_{Zo}h>5%0GxgiJ{T0M+%cSGF80CFv6Ch!ER!??WOM`}fMFuR+d#0#Zu{cZ+W=
z#Zxjlpg6y3=d|eI@p|O#a~WrA00!!ycAGOj5nCVEU`p9LUx|f`z=Pbuo!<{XA1UQ5
zLH~$Kl<*;|&&?D}njg*Ogu)`8GAAEO`_`d+_@7Y`HY}R#)@SVyAoK~~-xfY~1M`S3
z$6o<rX=)$J13L@Cr)$aE+uQlZp~Ujltl*u*yKZi7bH1{!{m=V`r7pWu-4oNH-DP6V
z2~e|CnZaZ)+x7#V1xrnhcQX{yC>~8w1Rq#&es4Q4blq7}&=O%{I@9;+JL~O-IQy*`
z#-CclEz9xg*X}`};kEc^R*+}}`~=OstiN4}G=@a=zQ4F6+?{QtsPo+{s}@pUtJaeB
zk6aI*+0|cPyHn!>J{GfOEs`Ms&rXp}e}~OrUC%BP3u!o^$8o*!drEWB_1Qj863riX
zEeRxttX)b61`T&IARxeIqp@6dcWc`h$Tb450IX&okPq;d!6s5%hpNM=wR4P{&<mn|
z-`}8s|6Ihg@LI|eXfDl)0G+(TSuU%!cdk0UXL+uCY3GF8k88l9@#E1QFqixUAYNOa
za?eHkSx8+fc5ZTupElu|JpHtLYSGM}m2UjpoP;yu3*P;P8QNuBCQLtZ&;v9o>D}e8
zQL_MGTu?d9VyxkNra>jS-rNrJ)!*QQUzb}AoQqa~0A{R2gRiJfjHV)n_IAey9qF#D
z$=+vTJy8^pjh%90B`aNwKI7fOTADH!a&jySmpy+mA<3gI)k?I6fd9zJ{s&R%Nr(@x
zO9m6!7ZEe;EtkO;B`83a5*;blQdC=D1bt{ngCKt>N4Bi7G~>>VV-T7o3H~dB-@ein
zr~~wi)6mCz(uJN|B)mx?&b7x5aJT1O(~>yFA~0)ZY@HA7N^Z4(`nn4{$4TOLj!!D+
zMDGN^Tb>5(8DIH4{YB}_!!d$X@-CH?jb|-cYIn&vxy>FkgQHXCWl+5PXT0_`90diF
z;PX6XQOaE(tX0!>D4}#kctnlyrxS5L)LC0(J6bRB;;iN`m!8#b>AteD+lQW$2fyQ{
zqYaY1xl6@U`7X!yOEQR$0Ri_MfjeIj+HY7jJ4;`p;t_YwmwVl<t&Xy$!T&f%kEq3g
zaLhM10}9U@-gU_D1SqPzeN6=CL%vSAht(jh%jlv9jW8DB%oz{811R|yW&w!V8q_wf
zmsh^)%JD8jhbBMCKE>I)MVyp_a7<Dt8<?;Ede8;f$!H6^Z^c6TPedLab>f7=Z*gM0
zP3jif^F+LujUm&>EK9~DwB0a-M+Opt0tp7$4CJns_XA-YXCWxTl!d%`7IO9i8W_5m
z9Du@{>+}df;w2J+!qrHaL0~vB^(7ifSl_W@lLq&Ut~XuZ;)pF&d6Q`{+D#llkk|~c
zI<J@Mm)u7B`Tf#e4W0FyFR+RRsHZz=x>W!Y>2y8mDD7(M0pZNVhigT~<GWu>+>PF%
zVe$4~XfHFF9z|<`!`Q6y;<Rkx0Eu6-%O`3PMkb3VgHPMo$qDP1dr7_%Fum<#Vq~-&
zX}M{qc$?Ie|AW#-r+Lvu!(9r#gA!BDoPd4hB*bNa33RBYGL~!}a1u2gT}wj=&8-E%
zD72jXa1FL&lXeA$_aq*nqNG0Zw;`wxT3`NAnyCvQfDx$A$awX&=w0>RZ}hQiKPrfX
z1>jYjxp}$z6^}%r<9{BxAiv)=1jzT$w?cBxC}X+_w<WNM7QOPN{-Yu_Q|DspsJG9s
zfHL3P;JlCti{Fqw*c~M@qJHvk>hJ-y0l005(?@4BRN_or798*2h|U<jF=dZS`28)y
zso^P_542Q7X7K|gk^0eU`gBq7otdUMhCKr-e+OBlioLr$aPa_PVq(-<z5JP?R4@l!
z{T`$UT;wxV^>V8Ij~2RG=@5>++3uU`i@II+f}i7@xJXQe*G$VmnY<-*EF?haPN+m<
zqAq0D3A~nJOdy`489fj;4e~zn8ql@m$IGDXTSN?fE~kf7=g^Ufe%rL;L{tJiD`+3#
zwdYzVQ`PpC$Pavd>7mE_d-1=9BGi%uN1AcHtIFKW`UZ`NPRaIU(CW@ly>FkHoanl}
zk#m}=R!7MlQIxe1_|J0Nns_hj5K-|eNCPyN6EBpuhoeA^&Hcn=(|l_iZ)GMcvAxBn
z^ar(S_g521Y~$VT-$37*zRgp6HT>OT$k#MP^)v7BQ)xSa5+uem%*0eLIlJdg`+CZ*
z#G~6!E`|twNSHWQW#iq{DZKIAVs&nm{gNuf{senlr4-<%C*(|lK-<zs4kxsM51Oib
z05W@A<;G>43xWYuy%Xvh;adngV!b8`zC!qhy_V;fg^g73YJEi(p-F-D=4GsHi}?bC
zYy)=t%qK`G%WL{{h)UE}njP+uHEUU`j;5=;uTx&0xSU)3(4{k$O3+z=g#6b(QYudr
zHBR?z%wi{cW3kMGG$<FhoTa;rU~#H2hjP%g8EXIC$YR!X&A#WPBfK#kq}$NinlE5t
z^L_`k0%W0t4wLQATtJszJi?K!dHvhx_`3Cm>hcxc+DmH<&sr4KZ-&x$rMx0OHd(@J
z9`zZN!{?hl0Y*ehOKe|dLv&G|o#qTA8Y2XtY?&t4M2`{OB!5J{j73TB141Fy@b1ix
zDuBN>+H9;X58unX4i4NBPj(zHTxd7Zv=@k}oB|(grNp3}vuZ^VyVe44oT0aGGLU9p
z=^X)A1W=%O1$$Lq?%G^Jd3$WP*3!i55TS4HQt^fK5GG2hEh?WVhTW+s&j0DE;Bvf3
zg~7si+Okmd>L-!oyl>Ttj@Dn9X;Rf)@0bxRvAx4s)dkL3EqIuK7`LcHnywwH^EiO6
z57v9H!x+AQs+;hy_P)AfiIT}cbQID+XSAmoIQ=-~A{i^HDxosVO;P<O!lgnH*?LBe
z>}dhB@{OI1XRCfRHgd3oe15-zV+KTN;*_&YMt%vPSj@>gH&@evGg-E|flWxxA~0`8
zf9`7(BHyRO;df)QM|43spS;}SWY-&T<Z5`+_H{Ddu2IDK7wm6_k(la9W66T^=!oC&
zDczNB|MD855~Z`jx&eWzr53*)wcZ3=)jmye@gF2Z^qRrwqb$oU0Hm~#`|EhIaxYX5
z&MroN1{oiV2@KTbHab(%OCxeJ`IPG*gS*u2GPSOhc-G{_B_#S@(o&sTjne1cisiT#
zN*K>!vY%n!{x6YYK%?Bb7{*CPp^a1%g2nDxt^bd%_l~FffB(l{i83<EDp3>-L@9)D
zN~L6HCM!us_MV+el8lV3>`fUVdn6LF$=-W!$M(CP=O{kkcfbC6-P~S|=XpFIkLz(=
z_v?PYU%Ds5(Cuk=JEK~LpL(jP4A50`&hXFx#DI<_@^!k+eBtaWF2ISA0MP$H&I!8)
z;q!QXg_UcNj|(dnju<Rix2Rf?A}UXhS_S#UHZ{f#XqO=Ela4JVwW8#d6we0_9ze5R
zC_5&kg%+Lrck^`v?_#~<&K|}c*b3ks7mMjbE`_D3W6rB)p{xuK&c%Ly@p!PKQ`ix+
z{QC9l?K`)>e%YG`A}`vM*(>2z*lAIZjgjjw4-7qB<5$y5Q+fye2IB<YS&L4N3m1^m
zzP?s?JD;0PvqWeyg1jJOduv@d;6skSPH5m-PFGEk^WUf`CtWN|!}=zHZO_I_WJ@(v
z)g86)G$>fsSEQz?wq@tHG>ga`(0P`D+@s!WOB1{T|J-;S5J0Mvpqp#9%CY~WwTK3Y
zP7YpxqK}ZijS{R^o)YF|Z7VLbPCUGYK)`X6-Q9cwpZ}Z!{4Z>9sauXHXz$sCu;ZRj
znpF8X-zzdx)vRc8iAv`&s4z&7zW@;B3PUIOQLUNE`*5{(;Fo;o2$PxRp+1OwAfrCg
zkPenvLSAigGXBE_OBSEa0Uhv9Zvl8CU47J0E*snkpMr3dwxi5}mx8i=5y&NiCJ<U2
zD1+4G8~MlMqZEL(`OQksLi*+3HKVwLMhKGu%;*2Y4-|~E2hl)k$K<WIl}HuxWv0$o
zuHU0gZo3mJizl*f_-@VIQ<c|2%X+pjQ&wx-1?5N3xj%s`yAOHN@H6nF+?u?&jXau-
z0_$G?cXZt2KRmL2DY^v%34ar@=u<3Atg*}-;$;_y-e!Y=V%ygE56_d^x<;0IMT(NA
zCQN1Hxc+RqXT;fUU1vOfb{CS6`8wmZ#nGiYx0sZz{%N#6Qbshl{AOWFG8n3Zc%mYm
zqE_H|{9i{mToq5}qsb3{;EcB=x8@&!Zr6FFlbEvbz=>wU5n(xWTC}LHYo~qn_`O(&
zDKNUfVx%D2;kt+EbP!6dg^N{se|gc2Ub5)a)PnUFEHSkkf3Atga}(z;FcawVVtOco
zwZnuQrak_V8!^^#pqWHoU*iR`vP(Pl$AR#P<h<!u&{jR^rS;;MuUdD31Jo|GI<eth
z?2kc=6Sm+QnCBY*w`M~<A;{JK%C&nA03myTo(+%E?0Y*pAY71OJcK~>O*G!#r4tik
z_1&91KeScA6H3s(oB$^@-LtY_Ff0ULB{KuK10=vFY!@YVzOha>bTKhAlRF3+ZbdE)
zqC=@0;(qz|tMmJx)P_1($G#nZe9490^v>L9lxJbCPc`>neGa3S%|3x_9mlu3*gc9-
z5Gi3ctCH@W-PDuRT=h(56Hm`B;5W-sA9x)vDw0tBt)%nei4(ar9fwT4Q7;H81zxdS
z6->ZxTJ&2&h-<H8aCDmRMBtfYR+02IGLbB(<XcJbe>G}=s0p1HSFP+ILh3-_9vC%X
zDLTKd4duL1$>#6sRch(KtM?!JE5N5l9p#8N3L7~gG9@xftG!tmUu!tX0UN6mqeEqA
zGaB<$gZedhX)s>3#OE{`z}i>CS{e84{u1uD(aDFOu7|9uUVb8Pdjb0BkkCrWYo}6?
z&jZ{cifCG8-roD^Fb;Zeh9kqXpE^bGpF3c>c=_i_DV>5%+<t9!PLJD=)ZY6vG_Ln}
zbEg<hpqRZU!y0?nd-u<MIWMi&T4;HHtd4DlO=x~9!TI2{vXX&iOWvAUS7!#3WnFA+
ztj)hF1blvWW9VLVJ8aCAr%W1u|HvHxed9R9OyjDKwoYtkYWC)gipl_Q`%m$2&!`%p
zQS)+knC9rSqkOv+3J3;M_FPu@dZ(Z{l3>&R(wU1+cKeSPKu3{K!ffehDZg%&bHAD0
zJIULzSY>*Ggv;#hS4-e6G|uz#<L)u~lL?OcqLfA-93oNUoRk}&=YEj(=$84!=PXBF
zQ5K!pCS#{GleNV$^`GL`^JIPk{r*@843ZG;a_6|R+Yj<0L0UwYs^9K@x)(iIwu;%Q
zVxF3`x5QG17EPy(eh((8sJP%6NPjV$J26E+Z+_Nmsb$Es%21%Pjg{D1;Lh39@A_AV
z{kmNxxsj9_SFOHEo=`nocLjccnj(0(peKCL^=EJDcd|T3XsA7#x#9IjTtB!J8mHS_
z>>mshpV8c&i7)=7oL>0FG=}{AxjY0eFCr<*D(aRIncID*^9ZZ4`^m$C-5=>Or|QY;
z7Ehl2y47Np%*GzFCO!9&R7KDCjb_iOIwVy$2C<&N*adfctda8WZ^C|*mO-cXT1<du
zNm#vJ@^Wh%Xo@3|7@TyB{X#m6GcWVvN9b9SUslmLNK6vueJlhLJEs=(nHzs%FSLUd
zpoH#}EZBnq$cZvuD;_n&l$4M=+3RaI%ah-9Dtw}4@q=4z9X}|Tr}O~C)&<vpoQ{2X
zc`U}@^Co$9&`ZQJ$@AJ3GPjs*j!v`55SIt((gTP(xVwt8=%cf&W!~%bS|K}IfM@5s
zrtciw{bHtwVuS3Avg_`T^q8c(14q7%kT`929dllyt-Y{0r9Ywf-uY6;k<ktZ{S48R
z*3XNy<j1&5Co>7gcFWwIEgX)A#1RQiTB|&tawQ>X>eoYSZ!QL~aY{yt>BGDa9hRDl
zBvGu~QV~*NLiNfzAxnzfQaABjX$f0i<d&X2e|;x>pr`>NPr>`b^f#Cg4&G(FmNB~A
zTBRbq`CYjzvnaP<F}*BhBw@|G*|zuj@77n)O>jB^mlcX1zjOcH@@o}XqcuyZ;w3}O
zMh318z)idIQ&ea^?E7odB%2d=r8F9%?`|KN7FoVy8){Q)1Q&hW1*Ozt!}UJri*S_)
zfD`M=qi%J|@!fl?PHi7=UI!b?p7uJLhoEGC%$klGoBLW@wR7{f<vsH@vlPGFthVBc
zAd)M*Wt_(3Cj$l2hh2B=4|nq6pLwYBod0PKbT5J9@#bB%@hO84DZwSVO>o89PN@0X
zk~~3BiE;ZrJ>_q)d8guJ1k^M2NVCg-LtJ|okzwRR*t9wvk9RN|B;WDF8#+kM`I$9j
zHi^$!TevX!;${VlOphFm_xEVbvrY<?rMd_a+8_UDfm6f>U|Ew<kW$KjD*9we1?*ex
zY~HpDNlP&W$>90wz?e^@d8d6M{^nNPN7KrUZ!ZrDtGX7-Hi;T`2lz(vy=UFo6pZrl
zU~zi6%)3s<@T`4(GF?eYyeCa6RLgX>w}MKZkrz;rN=jOCxl*>K*T%-Di865*IJR;d
z+<)k`u)9V7><2$eLX4Jzb5|&5%0mFx#@>sb7(2FgJfwQOK~;XW+r)Z*3Cmb=Bv)HY
zON%IYy%b*DaC-L5O8d^z-}Tv%OV?NKhBoQDXzt_pX7cB|`}2*|RfK1zon!WQ2Bt(C
zTCC|!_s|JQ181OC3sCVpuBg>-dSbGEXRu#$G~@Y%YTVcr+()^>@Q-Y#g0!?tCN?Y=
zB8Vu@Y$$@*j!D;MNHDBL&2GvIWFzz+tXxbtPp<wHRns#xbbLdAUM7AOru*A8llI=u
z$@0<#Mvb<uE%h#Y03c-m?Kfk8v30KjTZJG?NR>Zd`jf%7ZBi)8tnqd0x03SV?j!$5
z+3J??ayrx_jCu<1LA%*PtEDm4?iDmjxrM+@JuW8i?WDFa7zMZyJGV-C{!;4`NO%f7
z-LPL{(^EO?IR!prR(~hNXz;@6>><4-u_wOW(s-7TG4e^&&8P5E)NKE!^0z0u^0_(v
zbkb{b<Ecm|vMzniMoQzMf5tP_Qxr7WPB{SZsy0}C;4c_YOmZ|T9~k^K3Y7&Wlw1FI
zMMXutWgx=;VMF*o0{Z^w_q%^SS`otQVok`e-5;6f0OZiL{Q*M$S{@}yRopc>1KqC9
zArZQ8V!ZEkytf7O$_$H~21JF|WQ?@qOYD#CwsUq0=+w(-xcx@}cX{VxX<3sdOw<WU
z#!K1@4;CYY8Rq0K5Fl{KclYB1ye;wC9?f)iAZ(?5-PNIGRI0#sa^O!Q1o(#!)Camj
z7m)E5JLf*$8|f{NOzV(r&sHZmx1`uJu9eW$o=#*|577(?exP-bIPc{q57M2dB6{WG
zI*pEgS{RKDA0d>0x_t~*z#a>r(f75h=>d`?E+Qf#dtdyfTA$jJE4mHQRV6(Zu6NFr
z-~IF;Qx63W7n&YKTS{wh4xi@92gAU;Ezg7Yd)kR87yN6TQxyIhyD3~74Iw$&_S(vg
zG0q}^sMM|gJh2{|XR2Q(s`V|ZH+f4xbjJa63eH2*O~wtU2Z<HM2XLu_Jb=+jr@lrY
z$2B|_FWH8BNUE$C!LAp<3s%kd?PL9bK~SudDSdfdahO~DTitaHXr(&AU@kqH>l%=s
zJ00$h?x*A`p!;*;@d?dk5g58ech@)q2(x{fdRgVP{z7x#C)_3+EmZo1X^KK)#+$wr
zpt5FNvqjc_7*?ob({!VrnwI|!B+Y+$nu<<?=I%=8s3HMzSZgp2Yvh>b4$9T&5txkL
z4xu9ri`VR#^RS}wgc{HlVZbLGCKSeO(%-8#xVuBT7!`wLDNj~$4=?`+2gM2gCgGwl
z=EJpyte-Yof=T>^F?tCITe{nj_zK{?bdR=Jj!uVd3vDQNR!3afy=eaKl;_3!4$>R9
zfDc5`7d%z%TsIg@Q)D$%>F4q!_a{EB70h*VV4aK%>phg@{{Xr->MrX^!AZQ4?|0Yk
zu@vJIWszlKg}*Kbu*YTOx+|W}RW`UdTIg$52&<`NPglgqoc6aTyaGKCns(@3S7C!)
zAao|4jzsxZ2bvR6{499R+hNhq(rep7bVUOGgtXnhmi+eG>;0DrTUH(^U9{8er~GX4
zBAD06qlB!#=(7o34!uyP(KOAZ)BCD^t8=zMvv!jNu7A&zaqC`|pE)9JCVN4&y97X4
zoiAA)nh-n9k(0O5WR-HEbG5Fj-s?)sX^aQS6^y5P?+IIn)B!>LJsTN!dPlj?vi?gS
zy*<bUCQ~OwL85V<W$ZV5wQ~9sh9Y3%?$`f-k~|0`Hy7b$kS4E~*InH1t3J1VwA;5K
zIXIQiA?MsK?+%p*5zmO$hjI68jL}+eE;TSIb}QQV&csrm$X#!1V9Ie~4cIzAWlT_z
zM;&>+AS<JdBcJHeCd|Mo%B^NRyBnSUgoZ>`3YuSr4B(_YlEnX%_+SU*|3DqRBH*fM
zGF`G+y5>rGR!z2yiBh1+i%BF>%Y$!N$5kRV*VP7JM%5_l4+6mjwvcwiv2xdUH+VN|
z-2+zA;CYR?Sfd*DYYr3lV@{I@t*m`U@vTiikNq8=oILdcnj4PCP6XJl)baSF@BS`s
z-=Gso=z$DMNyazmmFOA#bstj5URF_2af!JwlQE2FC{ALEw%3dHHwHPeEC8guXmkMb
zC0myDCPfZ{N7MZy4L9p?pNb+wxQu<#Y=Jwm6$NHG`eHHPS_B_t&mX5G=By1;zh<%U
z+rH^@r7Fo47lLUMC%WUu<4Vh}N`dB~>`}d-*A>Z+Sg`|;jgzey`j6n8<YmuN@UXNZ
z35wqMkrdMss0rPHcRlCBN*-L*S28_i0mO$wcK@;&=|xbNMK;2GIROFsWvsF6&U*-<
z0u76ganPUjgovF8d%n(k!Uw27c9bM3k`D3ajuXStPgqm*NG3b7nJj<KhA$%7c_z&)
z-=WOVeFJ-1o>0+b64<wQ2KNIsW7G)I!Rf8X8Rml>S=h;GA2LbR{`2>&!BcOh5L)D9
zV#@8R-qZ0iQL(S9MbxIeYZ}`U!aBx2^8TG^bmTO=;774HPVK%VODV={sbdb9bmtj+
zGd;~}YK*8;&Xa-Zr8tYP_y?BT=}AYMmwC1{P1f%r2yEErO8Nh(ZGL(b!?ybEew?>v
zdhpWyT<|tL$8Nf<CLYk3cOKgnS)q#Pnq=9mhE4{h`w79d?U<q?)PHs_EE7TouLvXw
z-2D+!b+&d-eWSgnzA^po#zdKw-&S9yMacN=sCSSGXZ}!Vt*e_|_-7u#O$~SLWydSP
zjt(03di@QY-q!q#XEM%{=EPnq2g}t-D?~3C_M?+6WC3J7Hg-ybTe30h%_n!VO(-p|
zkwD9Bq)l;mNY@$sP-TUq7)XNM;Bs&^rK-P$hg@IF^e0MwA)PkhEcH<mkIM$-W?+vf
zDi0`d?Gh|F2}Se+#)k`vnHd=mL&r}gdTlhs+>1*6Z~eIZQC5EuEH=`SZ)3)x-%~gW
zXTaCWXxnYdb2X^gkXWfIGx&MX&c0|v&<CDj8S1~UkdH<wHDjxU-alp0lE^pG;rIY;
zHli{k>1Ynw6KQt-e{>M>I~W%=0R#x{Y9`W`QZQuD2+XG(8ocy_Odo>PZ#YO-Ee(Q9
zrxPDa@eTQ|f}^2hY|>f_=kDU<BW9Q_fr@{%C#W+3k_Jvoaa;%ua(zTquZ?Gi>7rVi
zRx?Q7o_%0@ZgNYzDu&1Jf6)bIk?_B{B-fX_hBRPm`REEbD5Ws-em8%@1IVBe_?q#A
zR487YSX*1eW|L$X`Fqrp3VX>L_;6<y9NN99qOn8W);Vo2lo8x1V8!A@1B>I$6T-ko
zVZOn5?V|pA7qRP4*(2C>0b1e18OvHJeIzp9@qDhYW;@L48u*%ETAd7I?j7(T(v&za
zd;6Th^nZyr(vPsOO9pZcB~g!aG<QMkfru0I2twp>xr7zFr}s`g>}jHqqq@7%;HhVE
zV~T~a_Pq$f_mAe;>i`-0x&egE8dqfZbm2jAX1DbiOBS1g<V-)z2iJ7<BEn{nw>{`r
zrO&(_D>;SC>RbcZVx&>^>doZb!lmCT!dU<TK5WiUdW864NVOF>DDr#kK2572l$%yN
zUl5l)a#LID=`Q93oza*rC24YzD(}kc`O$g}1I-bC<i7r!WA6Ni-f8K?>ePx8nfa-f
zz-bNBefso{WhBv05R7iQr7l=w@T0xQsa|M$BAJg)0!QSHij!u)F0qkc?R}mDOXb$J
zlGC-S{CUFu{)InRSR0}~7BEXV^5;fGe}pWRNRXEO0z+W>7r&0{=*@MbH@Pk4LiYlw
zPrO-Ue@@m;ynD0nBsWbDG=6$9d4}50SGehgh&Ggm2|GDrVAO%Rc*BCYts^okDF;}I
zUuu}eo51T<h+#T`#z#nJeh)UPuy1?z&cv^Zzt=S*1}~Ub^k6|Y7v|7^kMM)mvwqyX
zDFe{)zp-pa6Fj5oen-vzBlk3J_^UgvIyp;-7+CdG<ecwt?&@%AsTdB_vxG#!$hy|9
z+e{?(t?Z%Q6ZHz|@TU^J$FlyhOAm$IwKR7A^rLx~tnBP4{pHbE;u46Z@_Cb%lH7_d
z@r2r~1P-Yq#Q@;W4`;XpGKq0tD31o#eE%?_zxZ0x%3Xh4=LDoEg)n61=rPLSMGkyW
z4ZBj7aICoJT@6Y44oYoElplWKR=mZw>n{nt1VB%AdI`94Lr$#Mm8FbRE|GKd@%?a7
z5vKO^+TrZ}uN!MEMUGfL$=|5-C)_*+*3;KAggAaCunj2C(3+cXNTtlNIKuGpgN6dH
zcBO9>e=ECB-;8Zb-dj-yi>6X>nz%h=1EM3NAHp9-?-bk_OtbBiKAvyC(qIyj*p^tE
zn=$98-2^Ab$Mb0Km7UPE`)b&Cz|u1}H(o_9QJ;{PF3*2Fa$4rRUcB2lUF2FkQ>j%_
z3W>UC-vmX`)J)x1p+q;j8+mKO^qkEtE;BvZ-;N(3bTDVTH56Q6(Pz9OV1LqiprfY9
zg#9C1m}mF9b&+E+6x+9ZhIouR*~vCtOu9F6IW{(jTbO3u7nnSSOQ}U3uGW4X7;@ZN
z9C1$(`P^~RZooiSCGT09fNPs&WsfRb9j0qry~(a9{rkP*u;e3z+vLlu)9*^It2P%o
zrZX9%8ehk6e9!(Y+_z@Z26@ruBb7@vEce2d7TZWeJ{}IUsmJN(rAeTdEt!<pZ4o4i
zIH5G%F6K_@noc&kBsb%T$<uK^dg;;q)|h-9@u$gBJ)m`ZiKs}(55DR!Bd!0|uMQ^a
zhs{Q6x>8<Na~(U;Ek3W<7pgazd@7J#O{V#Xadpy%43nJ%?j~{-j*h*@;R!1~D2D+7
zTWYletF>jTE0yX-Dp^Wfw(Djc4-B0QbdqSZgo{2`*YsD8SkmOBYjv#YPIP_^nD5D>
zGg=Q|C^9ta5%>xNVe!SB-J6e{E04<W&K;^NODURiIKf9FvsMG3<pV3rl?&y7Dw_?d
zfVVcvlHluRyKfh<Jg#QjG7o-^<AeVUdhXO%_UoyVah?VEgQyddbaJ{ARWFUDn%qHB
zxv*I)vx^Jy5ezA>HRLff+gme5Y#l?JQf6jmi+vzs>CYF4FRB92IQL`gf4ChBP6nUe
z-2Kj8f#YD}qXmVsTU7f}VlE;z3XQd?@^?OF!crn^)zA%`GM9lhpdsG|u(0CiO;+p#
zn0BaJ^?k~5lfS&4p-+R}IT4!q%pJEB!gB?tv$cv4eW)@z*|ffFeyEB(dv!S2#<`6Q
znwSpOTBWZaiszrwKSR+i-rr-v-N?J+k_AHAA}SVLo<ZDMXs3uzT6aua4;6Kabf*B}
z^w0OjFIejRP34kh1(@nm)_*@mrF?$&^$@w10{PL4y-WW>_nXv;f7ee1yfjYHX#_vE
z*-Xn<!~D@o?6-mU+dNYx;4?xO8B$jJk#AuH#E`l==9b@q5|J{R(I~mzGN};1hsRO@
z<WNE71*YTJSY8*4YCP!Vph9T<8C$aa%tu+zHf#qD#h?Q@<<I~<_dZ`roBA$NtNwcX
z8l#8b{}T7^xD*R?&v|SkO1;MS2BhivK-D8t1&32jkdjl2!0cPgpCH(@Q(*O;>Sr#8
z?yyX@e{QUsONA_ET0lpa=MVLaA|C0ia-(>eaX8y2RJMu2q|k~|UbYpDv;f|*!`2qh
z77MivqjVng<R!mDIPaBHEh#;V$yKQTXNoBj37ct-ay|Z;3#8S}jP`-pXZiCUgJ9{j
znvyZ3I%RJJW@ts~MM93J-StM~r{+rHG;Skyp_oL$<hk7xWl$OF)N8ec?Z{;aD653%
zHrx-te)%Pct2N{tA<TQnJDAwZO0z!rrGJH$Oefx#c(UVYQL&)!@?)_oA09Lr$tzys
zwbnVZBwg%1s0%s6y^ePkI6}C&d+{-yk8@D~sai>l92@6m?-s&JN44Xl-+nkUJ;(c+
zcb|PkWJqjAI#AzbS5!OBvd5WkMgR3y@REe2s*%s=3j%th=}BH4e_A#zGT9$G;E!Sz
zQ_p+yECX^5PnYApjy<1d-w`zC61YXx%4Mr`X^U&z36qd1I^mq1NHYbw(PxwEHD{&5
zH}X*PU%YYPLHs8GJGBLuuxGl0+PM1ObZ0bQ`fTn4KB&8@8Xt;qoAzXzXkD=wZ&H`n
z`gFzPaQA<+evrLKUs`%Q+?i$6b1q+Z(1^7?p@uMniNf)9M|>3N$sGd$aPm_!_6^HX
zz|Q-7d?$Q+;cPekw85tu>APYT0jP=Bzn_8?Xd_HcPAGvv24WrCm=Y3BUoHj1%&KE6
zzvg9o^vLz{&4~bYPf>xf80)o-@_q(?@|N9euAL1*_>Hdc@O<UHmSuwAvSVP+v_4Fu
zLF??|u3!rujjJx$YgQ-3;`c{@!V7$pBB}S_*!vL7UE@eJE}7wHmF6{k7_r_vAXHy(
z3Brv#MaABJ_3xAVTR7s1q^T!Apv<;O)28f}#w|qt!7cp{(T}Od(S`x@hBuCDmGw-4
zVOQ?c_6*(a2W3q6FEv3G>9FO-Ss3v4VIEfnq@TBumVv#ZN5r3r8Ax$$354PG8dZ86
z66Z=g>a<A2+g>I(*T{SwkDAu{k<HwF>KQqgz$5W)V)dy%fKv@XkD6~n$%4H2CG{bN
z*!D+hqbKfYpTlA0L^QBDX9w@#23R$rG(dK)P8>|%q%mktiWmc>AzfJS<WG4V9pIId
z1$rRWCX3V2aU=JkkMSN15`}|9-)ZkQJ{Se?PE0@1;gH8hLq5v*q~nTDkDC9AGXt&m
z<x>=fUM?7&l|~tKufldITg-w$yQF;~{%dZ-ceYo2+u8Z^KTHFp^PMN_*dQ({0LJ(g
zue(lO#VHEOn^8+Z?P%BOTCVn3UV5JwpL)@%4|v2UD^>d#*sWpj)JyPoEs`@Bd}x~m
zSzp*12(B<S2IwjFPZ1GJbYS@G40DVfw=x>VkM8DqVo*7$-Iahr0LYnQB{_6;IC3H~
zKI1E>6g@yeMRdq!IrHbtZ03Sg+0XBQ-!4ox#?^BED<3C}eh{92F$<LR=|j%BTNxKF
z!bz!hkzhKI!Qi)<J|v+4sD;0FQNs$cFo=aY=)ku5^xXSXmQ20w`|_T<3OwA}zR2<8
zj|30I2OF}!F3Jk0k*9_6@SPty%#Yh|oV~0PUP4G=?N(euU7t3=TL3=Hg!vNTIZhjR
z7;t&(+WPvM?0<>{!pD*twSpM?K?JA`$SGlcSO*tf)I>_rRxtH6k3|<^*C^VBNLN9w
z=}bvh+Y<@lR8<4BO#FR(w<!jCKz8Gq|2{}#FBLEr*eAMe`sAm+!x8N0VLxi?alaCE
zxv!!H+7jV*4wu@Kj5hKZ0YA#Vs-KzUg8GQEW>lvALs7hw2QjBzM|ul;eu0QJM34);
zM&X6pe(R1@vsK=bnhJ-3)9J?!>!k!kGd8B}Q?iyk15`0pF61X2oys4$?$rT#Ba(XA
z`|7w)JAl_0YaY=NTHaPxaA5g0v$O<y);OZCLc9K0Gkv4d80Aw#J$Ha<{b-Yq->+)%
zSd>4jh5c5zbH9^p)e!p9j-RmVG^rA3<dJ%&=D0dEiny(ch3a(|{zk$JL&TOymhgv`
zlyc9L>fHqe0WE9Ie`7WZTxl-dJL&BeNx$=fSJC@UZVYiV{ot53k*=Y<lU1BLq9$Rr
z@5_uWwzgu3hGT1~`Bwf2Xt`#k8w9=BB|ilrzubEP0J2@F*laAyCc;oDb^Sc!9MYjZ
z=~8{$woLR1=ObYJ6?NxY=r0pbXxy4{+Foxd0(-y0&o8TA?N*?S(wh6ot=JAF3Shr9
zSezJ}<<>dOrqTX>WT16~QJMiA=P!EJVBR2tzxDhHZR^%=g44A?p*1mj`0gLQyS5|N
zEs*_IK_kMzm-6ywU?IMn23HFbtr9LP{>g5kNyMD(=`iuc&FdqO`<|_;?c%SOqr{^=
zOqf!X`_?>~a%c5Xl~CwNt>TM?VNpc`%BTFclZ&$!8@~%W=vIOupb}l%66-Td5L5ah
zPdnQ6nSyjOlf|%80x%Rd5JyPmqt4GA6Z=0-8L%bp0D{Gt!_D5gDeNl78c`t*7lAPF
zC%b?H<{xbVmQI`2-w${OkO?l<T@a-=DgDK9{rApzTI`l!K-4xO4`1NGBvXR%aiQGI
z^<pidkj_KeU~jJj7Ap5?*xiLX+m_phM0$(&O+;losFu9^<@Vi)Ms#F=B|;2$Y%KQ`
zrAIp)7{WmdaM$XGQN=CuWgfGDdQ{4xT%cRV;PTIkQRofK2pM(fr`bd(K@;UfW1Q4>
zdVW)fLI?!t)koP*ba`%fR2;5iME412ZQto5`Lln~HHdrbG45yBO^3Qc5ahwYdm;LX
zoBOL8&kVxJ*Tu6VQ#l34?N^_+15cvd3FtqG_D|!z!6WKzS)?7Kk&X%0B#P@yudRO?
zkvy&rj!#7?W#{n$z@P|0!>7v`WI1dG*BZPk`Yq4tL90g{k>-$zE?}k*#~n}m!PAO^
zDW=rByY5cxbC<k|kF?6TfGbIqpXRn6t|i=<@PqC!3_3=;sI!JDq^8I|8aaF3b*jUI
zU>YiH`zz;HD|d!D?w}vY7Ykp*gK0t=ETE({S~@7%_D1s{rd#tY=JNdymM)E?ZYy?D
zL08697z_ix9YEaANR05>0=_l)?v8<kG!NR?rtQqQ18aGq&iq&F`96aUp2EE85+Erb
zIWybzWz>!6-d%no%A>9NW=&0I`VSRSNa?Q{Kl^77yA3o3)+!JSwmsWEr<(df-Y)O&
zOwb#scHac?D9pB=>#vN6$%u;(ak(8AV%csa8&sHa`Ha_6n82}Mj-j2<t$NOD%;}v)
zL~t^0yFDVP%*)0cweQrS3=S^^?z9&w`+b}KH}3{b20iAhR(Qzeq;L52>58`XbLF?B
zRC}Ws_P)Cl`rM9=`lAn`6Vw3^F<Apz?BmeHIGGKa5y~&hs#MN)wOwby0BhmX@0Rzu
zh*^)1@l4Y&Bg1t!&m}W~?mN3vO^(CDWuXUfw%vD<P}@;Xe0+n2=fU#{PN*=EK{q|a
z{CuLY!wegq0EF7N+r1u;G)VO*tCSNC-eSBK{i*8S=W(R94jQlQ?jfsbdEEFQ{!<i=
zI@Senk75{feOg+9nCs-?+#8r5hxjDhZ-61jcxX#bmx9$wQx(>_r~dyqkiifHxTB<g
zT>FV@{mRL~)!E&CarS6lwxXQ84!+*$*`NLW%3t{s+#PIhU%2(14iH+PzYHcUHZeDT
z1v+-~BbAX59CV{hVO&sE!k&mmVp03)-k_EfESFT7kYQ6mztK`O4@nRjsR_yJ0cMW9
zT<lxovR7e(OFm~La<A%JSj2|PMBP>u(kT9~<CpmXDO@RreKV@(0Q#rxAICYFG|6gs
zYH<b~{Q<Ub7tVfb#b;$q)llxDWvg@N;9^geDsq>ZF-)5X2_rIcY!A>p5VH$n{y7ik
zsBrZn$10$Ae>~1#$Lt?b_UKa3Ma>dz&Fj{&mL^VMb|z#FS_JZ+L(ztc<}2Hb{OgnH
zMPIv^u!t2l)B-z_0%{COuEPzukM*MgzngBYx-=)XF-I6w2}xLqh>xQb2gkHx!C`mb
z+i+!-&vbVKr`zeC*0ViMEq}J(C_mcMX=6Zr5=fgZq+ahsVLr}p5pV@r4t?i)AFfM7
z$b|j1V?w(UPHzYsYbgyDnp>^a?ZRHbDg#imK>y*+jL{=Zbw%qsQ~jZD0=@9jHAJ7<
zC<^RW+iNnLS%tBOz-_7{Yo;0@_dQlG7WA5|O(VuubZ;Nd<P+~)!j>f9u+-}eQ$ozt
z@{SK)F$Q0SdT`t0tf}Of9x7YYtl!$l7g8hqfq{p>wFc%BaBj>!=Okpl8D3Z@bU*Cq
zdkO{Jor8Lw$~axX+2@QlHul!3p>xUYHST$SD|bjPm;+bCJ0FU*Qm91FvYl>YmI#kX
zrkd(9x^M6R85TwIPX?kX3xyETS95rg_PLO`Dq2{@wgwv&uEqmjg1#y6>s|fJ0MhkP
zyOl-9$S>JT3U4mjRT|C5uN5D_5mj*M4>AZe%k3@&?(C<)0TsT?^!Q}a+NA=P%8#fs
z5Ks0A=3H5mJ>zy?t_**9riacViL-3UYUE(a@lfO%^+dIPE2sRqDDE$gEFic=g2eDu
zBi7$X_H1vEfA0CJhb7R3a_INHFR6^nDnX1;m6xq9*Jqg2-IXYx^tuNeTtyWfdS8*Y
zepboC>ZTXwgRdR=5M+*f8Q2H32A3u)P6FKZXoAkoB{vx384ak_$rm3w{4)~O*jOdz
zuFNhj;Rmq<Y(A6`+?V<vgcvV_eQwfuPaP{s>{WoKe{5?y^6;0&DPS#IjlZR4TfS#@
zE8=ySAMM1*X07Yl(3M}|*I&{Mc15^NV9fH|&=!BB;XiW$q4?u-_8*$@VsC(O5m#sM
zSGyF&{lMK|S7mTlN)x(8bT9J^-Vh>zU!3k*ugN|-LNo{j?08JH5IN=N+wg@3AA-%|
z2<Lx8*jo@S3G@Ko9xLFN7X7Xm&e27jnMZ;Dy4{jNLodW_KAWN2csq`JTWm<R@D>$F
zm~O6g;7y7}>@fjE(IU4yL3}Rmecz=-4(Y|&)}lKP)aGHnY73qe2U}>?M0yRyo5)a7
z=cjkVs?(?-<g3yBW5rO6S$LWeMkc^=3Hx;cb&xphlGg32_Dp>j^YFWmZir0xsHQ)W
zcr>3Ne+u#Rz<NHaeg47AfxZfA4J!Sdey@)^9RO3NUjN94;_0Vwca{w;rdADjgDLoU
zV-$qEoD-oBWpghwv{imlKu&5@_GbL|?-fav<mdGLqHulzJ7EYHJsOy1VK#=@a-Yk|
zLMDc`-yBsCL_~SP6PR4j|N6*nGR((y326Lq3bgR6e6E2$3+TqC1E1(W*zJjWA=2N=
zR};i=!*2wzYo%3pDXa%HVsJV?7KiKjiwaWu`@K5Caxwq>mW>vm3!L}l8oU$UK(R75
z;KOz*0h^lM`2y97w~nYGa=@hT0zE=ZZ(QFBzvM;FXOK&;3+FY8s<k7AWM(kK^n`Ar
z6{sTAALKS26nxEV@`==FI9pJ{?Cc(^U>n5pF6DBHA+qm(#-OwY6amhu3RE<2$5&IM
zb}+8~k)eH|2Wd?M&{U9+Fy@pL;y*{oEYHYJXF|@L;B%T2`PKLUc(Sr{7P!R5^gNbE
z>IiA1rp`n_o6}OWViYXSf{|9q5i|Y`k5;UlMNIq#xA-99V<2aJSv2DH`J;PWwa}Kt
z*4Z7B=QOdw1ngrhz!4{><3zdRC6df|G`L%yP<fbrv1%#d8HV7%V&@Uk$pV(-C{h>@
zsZRY5QxCZ1zo3*>@>XAvaI838=_f#Z<ug-_^VRJiPl1Q<>^BoA7=CTGdwnMpFhGik
zC%+wUKmK!zz-@%G!#Cab3|MA##i{p{k37{|o8N9~zLl1ADqSx90b%sr@ce=P7iIMU
zSBBVI#z`i7n&<?c0=LiUwTpL5E9!sv=n#3jsjh;L2uC-V1X`C8g6G3kWU|kek+@Q|
zl*@WGDdidQm1otIcTsp)P$pYFmwR8r9W*K!G^`I=o;?2}yg+|M%nW>76>!YK)-$tE
z-|hEYZ@{J|sDAv{9QGIH?QHvFS(J%LXNs73BEYIa>K*A3FNJ+4sY8DU`e2*4<~PB?
z;Pit>*c+=FC{|O!f=2yDVU32O&K7>3!5DV?o1vBOOESc#R7!K7O$xJruEw_)_mPHD
z_uZN_x*P_W4JTv0@I$`&Xw^|Cs#&vjPpZN{SpvoAyXVfALu49pK623y!xOgs9gS!)
zzEY3Y8nf0*IfiH36r;$W%^oAkt240<E-svcSOZ+k@~>s!ItG4Ro%`o*ARmXm25N&L
z%@55ux@xXc@5_@Ly|o9(Pjr?=MHj6Do%9ye&E;WdGri~IoDtmZOZ2c}bv}|;HqQ>b
zyCY>dc7NfzS(Bm9gS+xZ;<#~FR!s3pu##V{Lf}_?E{@sK{t_K7wa9tP`v7JlzCg}A
zc0}YUAIOezJys3AF1bs>`}<{EU5lY$P0K}cNddm0Ix;-XeXB*zb|zg`S3kZn^5;;s
z9qi08b2)P=knT)ct<d_-H~=7x^a5T)U?UhV8jZe!s~`l`xZ|38J$x*HbK6Spt^LL`
znltxF0}m6D-^jh7L>KZNtfd=RAhhw6E_VR>=H&S=O$$nc7B4Q$aSc%?S(*Iqdxi&n
zz|k$xvcBtCc97j7!vsxWO%B~|d7_cr!wLEj9fz3Db4uFm<Q%YQPvhXBrd}PvwP3!S
zkyf;79iD#axp@OjkI2!!B$IignZAmQn;f*QMNayYM?p{U>!0ild`oCfs26D?kT_}H
z)u~xaJV;;Hp4lvTPmwimR3T}HJxmKasjzTU?_)9eXGK9$yP5q%_h=}v3rlJ#^>g@{
z&jjEW3#K5bzx~3WK8}mGQxvIgb^>0nDB>fFwK!=dyJh41pT8ChL8-2cx@)>0*IGEi
zZ$6q1iBJQ(w3yj^W~Ja{le}V5nmZPd4LJgwNc`acS5Uy!>RjBKrd4#5N4B+0*(HnU
z8W<}TjMRp0r#<UA+cyL9{Vl|F?(UtKSiV=hx(`n{imItLG9TI<Q2HDdhag#0!8khs
z@J|5Gx#I^I)?*ChqKNUvXq8ewO;?Sj+#parpK7#CD&zUOb)<XYyGXEvYM@Kup6#)-
z+dywjlU0O(&j-J#k4-CHMFj5NUnb@eu500NkiKNsW@1B^2ow#AbC41zjV0e$-rfE4
zh$PN&L~RKBqh3<9pvUHXjHXEMDKb2KJrKt^E&{E@-1=GiMcW2~*H|W1$;<Mva$r*J
za8LhkduU%sG8#lm(*bu&2XRuQ4O(A7dhW%@zKgp192z3W9kT(3W{#F^qz6k`&o1m1
z;Fc)q-F+m&|LYI{(?h$<3f6nICh0=l$K-CkB`Aq2JF(JVNkz-Jl!zr4dNejQncw>d
z2#(6de<a+%p*j?Owgc#96LX-9wvn;rN>+^p<f^B|mKJky>p9}rt0dt3_#HjZKG%}q
z%Y|O=3?A>H0HSn|SN1N52m6eST6hvKg$-#j@2V3_8_v#_HjKWDEH@tu&Srhm2~wJ=
zW~=*?ZxAyqzJF;L|3h06`dXP70mb&_N*+V~Dp?NT2pF5K)e&K(d(PWo1?&I$Onf4D
zc6wrWGE-Igiy~VeU69GQTVMCRJpMK9f=addjY}F0y|29n&+yuKliwgB^5UfJr^sIY
zS$65bI}xM48&6N}yJ3+3h4kq0Nx#R+9Nw1AmcyYeR23hVVWdD{qoeH!;r5UP;hFXA
z?e4kI=0K~hh1FW8EelMKlau%P=tt8eF#2o)?!|lo>Iy1eq7-W?tErtT86v=&=cLo5
znpX(1;Sz=ukF~mC%sEsygA5dz?C{!XvY(j^HMb?BmaAz3#=5&{92w>s=iJ=agG1cr
ze5l_Dn8<rqGX|$CvQlGEUvkW$trQ*(DgM%h{0UVD!y04^s5jg|cgKVC6wqmDX-1*o
zu-c*L&Z+lsBTva3C|mi$Tig8hv?)a3>RB+iv0XqUqNDfI!^0zRdWy3eQYaSV=O}wm
z;rC%o2cvucxZHz(Lw<JfTWv2Kns7%gJbyN8n<?=&&o7oDDz0<k#lJ|%2Pm7hNR)ED
zM;I^iT%mq9JuzV?b@3k>=>O1sgr#;D*pBlUSbw9JJZ~`SU$_`EkY(7l7R~%$;nHWL
zf~(VvVW2k(S_rJQA6@|$ex4ZC^A!{UFHsv$0zLR%k>GkNSUJxDOee=@fY;$H6?fu`
zm7mCqb69!T^RYuY5{4bVf|-u>bCl$~kC?Rvl{K_GzX>kt)JNV@Coj7iN`juvV=$G8
zJ%#*scrUCZffL?21j8hN{ykgWBb}V40Z-vAGq4Tw9?{o03^PJbBE~1%b>ke<bM_%~
z`7?yk{_EeLJ$ojz@#G!t52MQX;X);XVP%brwPgvoP7P$9N$}n?#awat1ObIo_DZHx
ziksa&ng^~4l-b(%hUS}CSR&+z7A+<{X_lr*hUCU(W{QNV2CbI~`f0~4b&B5JWHlyd
zK5Xd&cV@T{5Es@u`_muJjK`Xtg3aW87@B`Sx;~lFzTrJ21y{qCwwZ9cwT}IAdYIr4
zXfpU>{rq}tCqCEaK(55E!ly$w@&Mq;;rjaendnQrSdZbIME)W#@SH}5S#YW4-~$vb
z<{@WFzbt#I?kQXCw~amzi1ANkF?JTfzs=3dCuu`on8)nN`gT4BW@O)&<Qw{t?UT1I
zh0g4P30mT`7z}<1o#oZ(?(SuTJWxP-{NnfhWnCR%6DM?E(@Kx;Mlv#r)Nh;r;ix(A
zi82|p=Bk~7<b)ZrjE-}fvmT$DPe0&`%lUFPHIlx&x4WZb9X%<3W`Li-zSK0tSQ1%j
zv^bc<YU}0T1TPB|#?`?GL+`7wQ5ey8+d+c8ZK<zQJ!}xRF?hDNtI%Qnt-|B&fs@-S
zA5$Er^hq9EQVbe-)AKy-wB!CC{*51_I``DR@v6lKFzugjViOqj4MJbB^64=;zJ)Ae
znNO#PqEQ+X9;2>vPFoZK@*3vGm`~+8WXl+g6YtS~i0_{lU&B{OQQPc>cVY8R`rZAr
zXU|4Ct@=ByCOuI+&MST1Ves@x8}$vV^f<B26%))dZK+X7fTRP=d843ss;|#RaD)C&
zONHtNJl75OooWt_!&lEw>+A7!s{!>nX|89a{fnnmSSQB$>+d=rR_e?<OE=zqGSR9W
zy0<6D2>2w!f8x!Nd{jHiI!}q3f_HK54HPNYX&waPqW5+53dX$xtjvopeWT*Kyj0EG
zV?PAbko<Wr_aP&6?y9ep&?<17d9c6REIlh^(sMs_-dE}|lOo@qL{B;xISYKQxMaim
zh*IWwXP4?}ot1$48=rF@`A`ImC7X3P)K_^2bbM?(eDrt_@tKbL`nlwlvRluBUdZC@
zc?~<)=P;Y`+E`qXg|)SH;rGF<4VXvCyao?Y-^9D$e&zegt#ya><)Lmaxn3xd)erZa
zB()?Wav#`hgfn(H^_!reJ`=8?;l+xcIz5rnSSBcc?^K{YF?+VzJ&TKBF$sw>cx+WZ
z-y(oUDRkt}|MD&1Z0^H@=x_fNyd`FgiluW-_D7xE98D>Am!#o|s!K>n@Pi#z9@hRY
zE&a+cXxbS*s}-!hRIXG`-rf6zuLul`$&b6@zMeBxU0H!wRq`keN#{Dsd|sq)xbs3y
zrTN&A_}ijn(`2!Tws}s&>v<<(Dd$@S)BEpk;DVth5HiX34`^a5)es<LR^J=<nb3>X
zljvO?3EO`C?RtU0jdn;fP{?qMO^ozI!EykthONXUUk7V7UAFvC<C>RM#4+7nc=jG2
zEwiF<7Ri^ozbdUO_h7F*si4DgB?eI%Ph;Kkgkb=Yghu10-w5wB<NYHyG**kJ<Ks?e
z`@_5`0{WE4LyyX+>&4M4+msko5Rria@@#jUVVALfxQ^qIhEw3}-@ji4<hu)FH>d<^
zPYrNPr?~h0YN)f?4{{VOI1H;lwHL%MEW~lD7hbcS_X*pu3go=G51)+Ujgwlor_FO0
zPF${s?k8S}iO<$gF-86iBL&sD&-_+a<=&T_1dg7A_`pX}%6sl!mc-54;00I7bH2q&
ze4YJ8Fz<@N`G(A2N6(#j23IR9et%6{%J|sW-CG>Gl8+ueIt8=Sgf~^(f6Y~?Z_RJF
z7CP9ScKJ1^G-oy|ar*RWfy;sU(R<R)0C`w3WgGe(%#Vg$6?1|ZKZsFUuOvYExar_Q
zmBbL`dk+dfQ!DXN8U8w9Rn>4lDYNPk5BBl*hFG)`V|S+`hQ<Hkyabd9h?%ex4X@xB
zsp#v|PjbJN4;1jz2hO|!iquWS>0f1DHD-MG+)ydPbMz+{RAQeEC4qNYwKMO-r>&5n
z7X*hHrPGVaze<D^>Kj$};gOuHQ%k%ldT5eNp%5Gy@?QMIq6uEPQn(zed<r;f`dgWs
zU$GwCAhp7DS&6EzRw}Bszo{%-YfdQ~J}bO>_w?(^#iIzdbDfss;fi#uo7Z0F8y9K_
z`l(m`^umiMj`D976|B>$BW3Il!cM;W-(Iwr>3rN}8qTFYddfOZxZ!qT`uAe>*=v17
z-K_6_5EvLrHV>19)z-85iS=&B+yDN|3XAtx2}*R2GP8oD{nh7@9j8{z&AH0$_3aw|
z2Ph~fY(_%j_KA?b-rlI)-tbs)R8u64a3OWAohgK@yW6_;`3w4^^z`pI3Kko9srGtX
zQk;u^PHRPyGx-)5fz2q-`6-O`3-NN*<3FK(qyFk?5oB~<lwZ=b=NBv4us*G8X~FP*
z`(K?s+~GmY7aPkqVFsA3<A#;blb^MzP>PFoAi55t0I6Hxs)h*_5E2p!V=^_F*7jGm
zTqrIs{_v10>aRT`d<fZjjhE|8%GY-EsMnC&ThwhO6t#(W|Ni|dCu^t85rOohcv?dh
z8WD5l!rNgQ|Ld&=J6?k7=<(z0Kfpz^;Ir%f!6%~#?Lg3EzpsZaaS-3kkK;xgVM`x_
z<IHSrZ6isK^K9G>)srPb?<eTrZkZb5rzqV~c%d}G5=pK#MiCcm1A>bo3GgV>$u~6A
z0U|)JGz{u(vZ(3abm8y6gtyI7*C@n5ikT)gMTUqL{^QwD%LIvSoJ>X3rCl9tbFT+z
zhqFq4$S;R&uZ9(5b(&ViEWNWvdUlI2Q7t_$)NH}L<`%8$;!<~ax81zleM$W}Gl4jg
zB=(&ng-XIVFgEhbu)&@V6d69v$GNql>g&hE7`?KKSB6LZT+cG0<w!m~ym-u6C?F*E
z{U0xY<{TKl^<$;%`<NW<DZc+bp*INW;r)dPtPc9b5z9<obK8h>T8qP!F$8staQzY&
z9r1KrjG76q;<I@-H97eXiFXq=pClyqTm$s%@G`%hD8!|W_)iYkl{wJRyAU>=qbQ=N
z&(SqAmvrwdA!bcAeafY@+S{wqr(Q>Viii<1_>I?I4f}f5u)=G?L0N#a{i_oHQsGxH
zUYp>7=wx%oTjFX3SRD$G^czl&)Udb%-XA)bvJE5YhD{|;XaWDEa}J#ByF$y`n|6D!
z&x^m|@noNz?86ZeP>hV>Zg&6u)Wq9|7L>{8FDJ);LC=w2EF;qQL~(O-bK9vef|!`1
zNajcRI*s|1A9QZfryw_@*3E1NHWK;~N%k<c-R6RVL+7_m5$F%wvlv-$M}=m)t#8wA
zZ>u*tx?fFRd=I_p&Bn&Et&8O=loz5#U>+>zn;!ALZl%kizCU*RC|+FXpsjDdUp~g{
zsBD<e&8Blq->ca$B8t_}!waNoHJ2L7*X5jhY)cv(vhiZNpuQw@GeUQ{l%cag{#5kf
z-oya+I#5;k{qE&<yfC%B{ewh6pKIaPQp!mah;d<R_lRKBB#?ZBS1YviU8~K0ftubz
zy|Z95&n8*s8U7P|doKTxKUQzOMX=v6;(rns_SD{=BimCSrA)z5RhY6hgFBao0NsZF
zqq1*$3jVc>$z2JK)PiH(d*bJxW6cL7#j3U!tT?a7#}d6DohUSfxT~+5b*(3x&CVP~
z$k&>ia5|qU0#<BFM}eKyId)7rV)i;xhPj#PS+Oa`w#&6Bb|T7dn+n{yUw$cvU&*yi
z6rsiN<?*3W8;@>bR&FlcWhVS}=AuIA$En#`kENw0)sz3ue#-FX4h&*b#I&-;C#Jca
z4rBL3dw`=P6?$P6vY#}>uHjePqqoMB2~rN@BFIQ8#kbN)bbGGJ70!LfI|V6m_Z2!9
z-`pg{t@<tjZEFrM|8Vy~5|8kW=O}JEhC^1yGvniHwZ*7CL(O+q7GnI(USK2IPt495
zpEA?YZ5SIH8-0D;(7MM!wNrR|wRU?6%*uRYz65zsKm$yjzR2cL?)jI;_u6*L5x4e+
zR%*ZB>6jNGv1^(|XUlNyvL6#nT<a7%ESMJ*YW~>!20CWbF0!zuH*V5U1(dWkRI!{f
zG(<4fl^~4yf>_xV>^2J5g$h$5iW^Htjur>D`7nWr%I7DEgZl^Ef0BPz&09X{UCX?A
ztn_%RRQuOw9ceC==H(s?r%$fcxg=k0N;@UE9DP<$S8Ai@>|21(_pI5QQ?O(tf0m5k
zfT#V|whE>~l35ys?%+{$N`J1+W<!RjCBkiC<RrRrDuUKxq?gsVxE9Qg=H!q?L!Sa{
zm|D|s2OH*Mgh6Gr8+~<k(g(OCc*-DyzQx;ZJ<!9z>V~l6orEr3GKm*)wl#oEx1|&#
zf}u?uwWpdmVJY<cud)hV3RdHi@48E|U!QRuNi+m#!&dFuQf?C92XRt3{W9oTaqz1j
zu6_KPo#dV0*_UIOr6mXUf7VH;7f5Cp+yap@5dZ$qUrk|aO>v<@+}xS{#jgg|E8LQc
zw$}Vlt|%QY5*iNBibQ<+)=vFS9_}YDH5yIzPs*;^$plfo{{au2;d$Q<zM(OVHxw1R
zotzUkbO=u-;pqYL88^pg6CRQBpSyDiGs+~|{K+TDLcf5;sldyBW<bpB#$Z2Im>3s#
zwvW9**daehaIN>CuyvjO1*q7zM`{GR4m<=SpPABN2TZgR^&Z2B=cp)1_l+{*al~fS
zH`wr;tuNTJ4AC<^_j4kreL|E7(NPEDB9oVU<Gy;7bB?~U`bK+0{Qo#Z<MzES%{l)n
zVdzu46@?7bH+pxJgEwH6<#eY_skjh4h%?fOV{7zQk+ne-{PmYZ=iW_t2leFdxpkP?
zG+x?$*?NL&5av*v3!7U%XhBnPOXe+O9SXfd;M%Zqb8-9tnQlBi@jFrpf~4{3X-CnE
zcTm#=tbSs%Arv{H@lnMqI%#%{7TJ8xkWJi$VBdUp##S*bI34z=S2PqO-fP2#Us0X$
z5xGyQgZ#}QeAy-b@+^X$4Px%;xFF7i?L@!;*lt?mW_iVeOO9kyl^gx6d!SPY4*A(C
zw81_*8-G86SWJCFO<%Lo)!9qxl#Iq+#l2H1iEd=JrM&njSRj7cH=Xv}{F;(7W9I45
zbb8OX{sYrDoKJ<ftb!=O@)`nFbHA*?(8uKx;K{4n4!HBu6e?FjL@}>m@y};bWn<n$
z55a!`xtASVW;M=L$dp}LXo_^^mx+zo>5H6b*Mm`%SreBpbAHbc>ugqY{jfYk^ns6y
z%k=*TBJ(BuWp3grU?MEMCBSklhSwb^5~2t(aC-=94O3em1DPj^Fbzexv^LO1<w}Sy
zVYFEBS+f{DKIXsI9z4#Al~`P6YICHnOi(H*;lpi`PDdwghp6%^YK8beUZ08jTC8tB
zN5ZfEX1gOS<sZ`2_`Bf`?)!$z5x;YHcNl=Xru#$XpzMbx;O<h%v-<gJH@BOECtf6i
zd}34tlG=h8dq*rOrAhyc1v%#p75}XV&oXgn$r%z~QT6%cGg6w?c!88}1fKaC=B2DP
z{9@tg`#_9@fQ6;S#e4xI1faje$)zC5cOOiR<Hy}kT&IEE78v7=b7VW`i`ozn(*_e6
zPIkHax1Uvdzf^JL*%%ue>v!$Hy4M5zAtA(aH*L8c0MiAfyPDt=@;#}?%L4#wMs7T}
z=}x+(*fYmnqh&R;ZCC?Z`pew5FZ??Y5fgZT1f@0v#BzPX8s~qhihZUWotknFUHCY%
zaeLy)6-r3OaeHzfo-PzMf}h!NZC^$&nSH9AmvfC941<cLzKX_8w{~Upk$&8d-m2Mz
zG{hl7j1@AbTpDLzVpIFr3FYfz^4R_)6;i?!yQP@J2U8(ehX#pSuj#<Z+7@Vr3-gQd
z#Zs>K_PcRzjl*!c=>#S@9L9_b3w*Tj2qPq)u3?|5P7fVsVWMvO<79@d+QMO&V6`$_
zNzxD$oZrBdnt1J)cr*Ze`_*R2xc+w}3&owZVg%J8X8>VbcXn7mw-{GuO-;?ysrF3H
zdt3+2DDS+4(ZrelY@{Si2XTwQXo84t3h{S9pjFCqDb`~G^NN0N34f_-44)Z%hEYff
zT@S1|=UuEsdc1SMb9l4-R3MCc@`28!?vYQL$VCVCHOW>956CKZPU=D8Qd0WmAF7?|
zR=rw<XJ$#zI)W3*18%gOg`2cJd#5cS`DL5x<dUKw_N_d9`<!E1LX-O`55aB6)C=<h
zRy-}q+RC0}5(Fbbp`lxn|HiC+Nswt}PQ3j5LG$b}EqZ7e@q=LIZ+h@$o-K6_?<DES
zTW1Wx<LQVtQ+<q!1YDeu9S{3U?<1wiHq+0r#S|Vm^!QCDG5c5-C8?!^%XmlHqq4Uy
zH?Hpyo9?DDd8mA}4cB!4R&2(|`J&&%A03w4Fl87N_v+INd2N3-89q?ip{SSR^c=+g
z2n1h_zsnq~mrg#(!VBb84sRNqyMJk8rF60C{`)t~vTPR>$gGTJCMP#kLfmBYUfpGh
z^^upIA4YI>v;vU(GG#j+B3t-Rtd_jl3%moTDr`Cxd?f^LF8D;~#EGZ;Ew^#@C)lv;
zeFQezd7A~?Nyj5ZPI?Q$x%;k=!AgHt=-_?k!NEUxo&3m3PH}79<GRSKEX)sVm8W;u
zm^_+><r4O9S^&QNDJ6UZo8SNmpx^x$E;}NTPCTMdaTVsFo#2NnzED;qo&wJqaG%>W
zk$2^oXzuUaiN*Plc!bi1`CvccKgg)fO~NF~)9?uCSEJTRq+^au(44!~JR1M4>)Qe_
z+bmW7@zNQ4@c6q%3>-MKdtdZ`K|`god}V?tGEkcHEoMPsoUes1fo9fBpt<VwpZX8A
z23NV%HeCfSyMBz6-{PEMgCk6{n|5)vT{9mxuHi2@jnAXIWUNGI@0wa|^h;(}+im5@
z?Nsc4?jdC9Pov>qHH;2g755O66=T3dM%9KQg+|z?0lB!n|En3K%w~nfmAGAamt?d*
zbdfkf_&j>fbu=h?n8%0*sVN*Lpo=y0>o)h7-EwIs?WbV8$cw(mS+LkFC@5$qaMwH;
zAH)u`^V#3Zw;oc*9Nmp)_;jb>Vi7q?idBu^LtOwD9r?EH*8z{iKZD`$Pqk;T@Lx8v
zNQg@VoFSq2IW!7#X$ky@kM#~XOeig{l*wXH<Hlylkd`BeKh6H0rDU=z$oFBi0gBFG
z3+ez7k#E&d9tUJHMrW@S=u*;LU2fH4Lt6c$L~)w7d!9y?wYRLxm*ZUqZr_LEi^UB2
zFu2vOL|to+rol@mnskj!v-?_0)2CEVz~9bV|7EM|`fI3ZH_!b$xE~<s=N*wpvP^Gi
ziyeBxIkZX)&iK=rb7qW%A=6+WevsIDU>;3ZxP!XCg)-gly`nmTm;(Upj7L(l5|x!@
z%zyvz9l73hD^Xg7#cnZk(gvZ`w-E{Pf*UDmMrFKTFtG(RtN%j^eZbgjs6yi96YvO3
zDY8PXx^0BT9~qjRIeU?xpH6V#SZy{(&Y@qOc)%W4d2T?wvk<6If)W+h`G*DV*j4xs
z{eft-zne>sG6tp5_@Po?*6THuEIYL2j{P#!(CAhDHv+}Q4hWP363O3g*BDdFh9#k(
zjIABuChuVmnJDhZ9VToL{pk*OdXBP~Gj<cA7fINamMV+HoIY?dUlc~`Wwu`yk`Z)t
zjJhf?gX{qfSJ(sU#E8LJ_!IQXNV;OB{_i8$Ew4-s)$DlXj%b`pMWgI3wYo4c8@|%2
zL|knqCnpz0THqaT@283Fd4PCP2zRzkA9s?SCKhS_{!&#{sPYGLB`uGj-w{r>A7>Dl
zGrsNZ_v92_0YhWGcXz?J9Po@XSO3Rhhp_jKq%<U-4Z%gO|EThEbjL&TE4u-^u<6K_
zQI_2qi6|L3<ecM`NjRN5GxNoa`Jk+`S2A@QS-d(pa<e*BY$S8zwLxF#H&k(M(Y0#M
z&6c-pTO#g))oQee#_+8k^T(%f#Q(Z$5bB28=&)`q4J^RWOkVPmFvUY?dpd}B#BBMA
zGyai*$AMh@?Zqfmv>PuE9Mawk`I&OS_N$>)vcYaYurJQdj(c`%u<<Ecg0eo-{Y0bn
z>XD;vyxA^Wix^B1FV+9)dtiSDzI#4lsnK$45$J~$2V+iH&_BQ?Qo4{xO+<Lw`u>ML
z1}{~Q(?9Od<QGPo*?aHfc3_P<Q%Gu5Bn?Ywh)Q`G4o<hqs&+??iTlW)>%SSFPK=0%
zFnXMEO_%ac`g1!*T{VUrWZR>npg%#!qKlg|gbE+RhBs0$qJANjte<1Ae`a!G!Yw&>
z{Qu+ZzvH?7zyEQ(9uX2rRw`7sj7TWs(Lj+AB4qT;mQ6N2(v}e+WMn3@kR27-p+ebv
zZ$jqx+#f@p)${#%{eI8C9+#eVzu#}Sv(LGnb6$2{uf1-BgOK6QIEu2oVeC$fYsMgg
z^0T+=c6x;>VKVWlb90NSC(U-yswrdcNVvulODvyhoP1EgAVIam_Fub1a3@Dc67}#+
zU=us}pf7^;$i$bw*R9DFR<3dsyOeI8k3l6`2t0o;2iIi*#rzn=k0=$k{|A+0`Mvn~
zt4f7uWSCdatrzCuzlD$d1DAv@`m=80S)5T%-EEy7C;aI_oo$n*u}%({{?b{0^fluA
z{CuPQoZ`Y)v5dU^C=bg-H%uI1-W+sYQB0WX9mMGi%(ES!-@n}9wbpDj%g!>584n_c
z{aO^*tgf1;=Q;@lsXKU;lhg`D)rO33B7&>D6BzBEluT~qdsrS`nH47|-G78i#N!*S
zqmAd6cQEIB6ooYTEQB1<w3Tfd^_SY@!=~=Zcvs}mWP~f)K`0gNy&Q{n3wOZG9MFF0
zrOAJKt(FGLQ51Ji=JIR58FQ_zdH;=Q$B{0ka~OKPVYq8fe<<=`=(Lk?6ii1UeH7aF
z?%kW!((N8GGdK5!?IP=^qqH=0<29^TQENIH#Sucd%fsv^gLUDu-bplAb>}aP3FVsr
z6~7Xg($A}hx<|b@e5OH_iE>g$@@o<k&;63!*kp{{ixvx;l26M1y*OHL6mpOsGu((k
zku<02Ec-d9rkglOjGHPgpKtrxTOggRy=h`8X^j$Snx8+#@KY_GCYIc~)Ldv0c}6q>
zDniE<)FQ(lo@ITkg?Tk&Adr)mmgYEOKx6Qi+MvH3hMX_AYXO2KTV2DWRRv2U$x)M$
z#z&j9f_{A=BN9a(7A9=ZsL=!$LM%}1;2jju`q*1B7n>cbfhrFnbbHUcw!FM76&&*<
z9&d`V${0^JmfcwILF0-|S>m41wm6DCqA*`DCtWLhq?7S@GzMoVuz#8J5pdTXO|v^V
zw`~sLu8wTee~jYUeIrWi<vlXuTS7}q2Wsx3wUZcmncEIrxM5(>=L<E*_vEEkFO%q)
zo=a{4n{H=E2M?r%T@=aT?LX2iPjJ8!S@wcWH)yL3qreP`4~jL<&&?Se*3@KD(FX)B
zzi&Z~2BmYo*RKm8rFtq|Qahn(y&unQT8F_b6`3#ANLe6ypxz0?EWse*(A@FS&%AVn
z^bw_^bcDq_iOSw-Z;5W5sh?LblHu=QbE@FOjt!53`FIJW^q;|wV%Ny2+Z=A(jpyZj
z(I<#!npn9bs$Sz?tHKY27zsId3;Su&3@BIiR9Eez+?NcQh(CI2xu0hmRZ;B)w<_@6
zyZnNJ&fR~{h_aw;gzh8Tg>*kRG+gw5ID2UTsY27Gcdxl!if7#9$!M{`3zOq&@g}W?
zTR_luLQ$v9n9Wuo9yMKlF2|zfie~udxws%|U^X>@z_)MT){*>W&I}@MB`_VyN&+Cn
z6Wq#`d7M9>w>Dk;N0L&}LHaVzO%oO|XCWi4bh|wcM}d;(;3`W!d076LPe7QKce#7j
zv`)q2{0{qUCE3I7uD7@zJ$lr3068kU9c5E4090?nLQqHzq@l{d!^5V8J5sbdBBeGQ
zQ}m^t%`OjfX-IKSRM@z}DQ=aG_y+hR7Ny%(JM{iMPQdyu0>g`^nXQ@Uk9H5G6iFHg
zuaiso?a=tx-r9;vFD~PscRRRNCuGX>Ns)j8`o;GZi6zi7@YVxbF|OAcJN!)M)=eKW
zRdQ{CpBJPzF<F)hNVU3IxHnEKkOI#$u~I|rpq1lg`Z-Ox&*wvt2t^<^GBVN;x#Pt=
zG)!D0-i~Fh3Zx6RL*ueL^28~aBT}e$j%Yn?`(dI!br#E=7d0W(M{cXF|7nLe45ZL3
zH0lnI5lhw<^V<#B-lMEc8^?d@4Weh)K#lKZKc)kRPk?qY3?l8<z6?}lGK$7Mx8)R|
zbF|*IIy^eMY{vQVkLBBN25DLZ6)ovn#g%y5DvOqulg)u&C;wRf)O+%ELAGIt%GtrM
zz5lH`s4^(<7*0OwO&~BA*81v%Cw(FDU3H!1H_p5MxtT@&<a4Ax@6U1Hm<DDkW%;<L
zF*|nJPx#TzkC6xIX+nxS7^_!nl%iFVa!CZGP8(x8zPE!-7tA=Pf{!mosPBuoO~4F-
zUSWrbvAAHeE3mw$xy<~Bbv663%@YvwXs3ECEko4t;PnDj{9}%>vCFmtQHBt|(o1*w
zMkz|!<*Pg~Yt-~Bc}d)@)MH^~wT@ni@iSJDDEYL>JHQJ)=c5v67IFGO#__C6QcF(5
z?cFWn1_bwTk4r&GYZE6JTmVjZ#QPK`Dk5IVMIt)!iIyCicy$LO4ShGe;g&HY{#kX>
zM?oL<wYj>wuD_zqd-`W%<4wTV_ptAF1#hA_yq|IA*>XRWxqG|1x>m7epkL}XOA+4J
zq#6DI$9Ye^z>Tmgp$@x`gief6FCw2*^s(1IDjDih6KDn<K1l@JCwe*BWhw_>3Cx3_
zsk;9)H+EMV<-x8nM^c(yyG5TNtJOL)!J+tYVBYC83f9myWJaxyHM2tmg|nnM=6w!4
zCw4`R4I6xb>j&ySB&f#t0MxN0mO4mgHN#&r9u`30K|^7b6mMd$@JZ1i-V^37Y$yFp
zCF6y?s4AG3r1BR18K3wV;GlA}>CA@pOCohvG{z!YU7zLoSEV2yvvxkVT$l-S2og5I
zD+Ouy$<vK+Dx&Ks%7e>g=K~Ua(2|9GNpO&V`E#P<7NhyfuBm4#oyO=9w?@y+`CIYH
zlPA}YdtCgaI-GHufDzb4NVudf7^>nrJ2f@+Zr2juQ4APkPLD6|c&Oz|a3u&FZ$LeC
z`@$8z<QVLG(H@}j>gu!xQUj)IgF0)p6-G=`kYc@PSABiG&)-8O(iIhOABcCsV6}N>
z&!0E>sc9igKyJBr|NcUu2G$FkQOgscVC&igxN&29`{Y5y-BOKpUp|mxdhO^Ilh4in
z<3;Q=D*grI0bz!YuD1r20j-Qu1-mzqjG!<2w6R64Gi~CBI;D-G^TE;N^vHbY?Ivfq
zolgGYugef*2V3%y?m|BHkD~8GP(W|f_(f;-Oke|<wqx8$S(qf1;b!glwNaOX&Bk%%
zeVqYkg$?Q+6{%z%rHx`Y^TTyd9$KcY2$Q~s5u+OAE_oe|m;uBEW95f9WreYV#R~<w
z$~{P!WGwj9B~8U+=}VNy0!%4bKdd3mh3i+q>Sl}@l78L5kFp4a1Jh>?iw55=NnN<a
zC;yMGosq15zp_)Oi?J~P6jUE$``0$vZ-jH><hugo`SRZ6<cR2;r{xR})rDR6*PTwS
z<cN&Wn6wAAP;Yh3(|Ug$KMIXy9N0^4c}JxmPSO@b;iZwJrn_ILC&w#<$2P~YF+0bp
z>Zgd)UO3`0Js$X1QVXGu*Sj!@%S2z{Tj%WPTN^LmKZTYjLVEbdlhsp~%;edH&eIXE
zw0-J%S&0qffUEouK|V$`i$Mjimnb0xxZG_a=o0s&q0sm<Q|>i+hx2!5LtFxhPqcD3
z>|QPK$~*#ov+Q3#frxa!ihhFKBCW=i>CpoR4%`!;csH;k1yG*S(~N3iC%qdsu?pu)
zkZx%mX9ryG+D^oodna9ewQ9k_FEYzrgOXfT<pYuWP6g@lZckHp9*=2bs}JnL&~%th
zdV9St#gGP@gF}Ytr8y4kGJnUXC@AO{_U_H=ha_Ne@;lfE&SKs3fJ0GY`hn7lJxk41
zoJ8XY_LaBG$g$mC5pM*vc(MLl<%pEkkHUe&*3l*7t=#6jR$v(@6E9MenLFe6^DWe5
z#{Hb@xBcSbUn;ZayncfB@+A$OAMY0@#UePIJW$B~<LYa(tF!yczHj~~zXfUMo|n}3
z>aGsZB(TNbB2`On{JiMis6hu9o+d85O@$ojX~d@9+jCz}e6FGH-iK){0H99>{23AX
zBr4=+q@=k6#5eb|D|=!6iyMf4k$zC^+%_Ub(eB$-$V#N=zs2RG`_)0seE2Mi&#=WI
zj?J+&x#zZ8*5$!3a+GO#ZYZMM4=l%)6}fs1!t{dOyAV}Rko8SJsA!%#$5tS2Bh5~l
zdMXU0{MT+=XL?K4;{Sc;>1cn{ye+NmcUt@i3vsOyerZ3tX<WLS#oq;jbJgOrc?wgJ
zBr&QtCMqqQcbMNSshUxvSvuASX>7;QPfc5{pP_mhi(XleaAmPOwg;=fz9T?`j<8C6
zpNPpLblatsFEC~PVL^^hxA+(*EJu5Ug|2?sPB<k_+_#fC+2ow^zl&#oEe?p#M^kF;
zGmuilETu=OBM~JX1CJ8ZO3A^uqr<~LJMTGtJF}UNgj@K^MOEBjwGJH_db)$dipdvQ
zCZ4P(zH(S(H}UBu+mv*_S<(%9R_fWKa-1-LhYy@T(qCz43$B#K%G(igJI!Y?jO+p@
zj?wN<E-qeo{SRTQIo}1VWtRqX16>UjImps9yY3(My$_Mwt(<1JTj<T4UULRW?A#@W
zUp{_L5lGxOw(po-(y898>rZtGV!D~Hd#|ju2~LkGWMvpzsZDf9nfFZZW4pntm1R<%
za*C3glw9q@K`a9OCVBvwL0Eiy2XcB0BeIGE+J#tHsT}D#3m`YsNl_-I8&!e`!o;`5
z$4j4XAo13QoD6`{|I}7^7AyH%QCvvLPulz9O*M~vp%d2@-lAw;mgW<y2Y-mFK@hb~
zi_$FoSDI7D&Z2&ht064~BZGF>Vv_;1wDBD#XH&pE9B>QJ8s$+3o^=@5UlQG1>#O##
zLva1##Ze6r3>|r{-}is4zu*<cFb#}XJ<%+UFAJb4!1grXWR1Iu%287DEV}Bomoo%_
z2&(V$U0hwQusK1T+StVK=x9DnZeD2`Aj=B#(gGvwFJqvmsg0r4^5pfK|56bq8KCYc
zwhQAY_19}5nW#~nuisxbnk^^r230@QD$Y0G4Y`R%z!t(Zg2JNUhSJTiBD`EVsswQl
z2Od9uyrg)e=^D}Y?b{(Cv;OHCr4!-rU`1Jh29hBUa_{2PzhOvuXV#~dN*ar>vzH=6
z-J`$URLn6l{$jr*izKz9=+gO>&Itek{w3><*4IT%OXOk`5g2ERo<6-DS8OZU)>{B5
z_(e~$YhykLLg*|sup1pqI6izcNh|4yP4|$uGZt`ZAf9k}v0%Dv+*xqNKIVOMag6xh
zV?!e&E~=31#4r*iBqk;c-zIa0f@#kiR~A3PX=Lp3=ObX*J4c{8KRAFu;#j%Bvd6v6
zn>~V{KMLPo;Ty~=4gT8H-hKd(VH07&7mV=66?Y^1Ss?koO~4Ib>NQZ+cUl}f5aC_t
zbnSR%hyxwbThGq9<k$X*`vdn}7CT1!Onkz2?)zV0eTan~;H~fMlzIhlfVvvj=v%w2
zVR@B@m5Ip^39+%VBqn%=TJj~Z@7$JxIJRU-MeO@%RRunVdm32s_FCiuz1@jZ0w6p+
z`yNDuAD1<5WZ+A=i5ibL%K~>NV+P1u@K+26C0o~*8p`;dSDM6pd3^!q;OiUv*i22K
z0Eg88oAGq_d&>E*uT$3bRdf@JTTxo|Zj@Mo$B!dCj%zoBabw_U4C}?p8eVq!x{$Pb
zqb|~<OrljaJhD`UMSv0NZYJa%$?I|}r|Z{f;&^Z^^c!^;?2kKC2$18UNS<c_RoU_9
z6|mlFA+5W+ds6tS$r)^#1~2$5oZ#sB6S{2tX@UJSU#ontnthWRG@JYmfMPGtzwx03
z82wI42FHg>;n>R$<Auyz<{ckwADSu5%aH%TU4fmKfxxj#E(}idq~-QBgTt-*SSdK*
z{&)<4e~Zp7%pXY&yYkD~=#z&;jazD>8P1V@P7KJExfzkvMW{LnE9`44$)jT6xhIl~
zV0zf-%eXhC9q^xj6MdIRMn=X`3`tZ}DOV+(`vEiH+uoQpzD!RzwD<fbK~+8nKGiR%
z?N3_JZpG%|XAMEU8@t6U-DVv)ZhV|5zcn%T@!O(z29U>mEes_dv=0i}bAlMp;8u!Z
zYM{Dlcw?9qYvSx^pey9?Q)`qR&C_AD2ZA%PPB@$=l}#?bkXIKm+%dtVHg0g}*O94y
zQ#k%9K<mn-L5q1E*VtZmrcc_JMoi<d#?{5*h{yA#s2+N+gneS^uQt^5rXMd9u;q@E
zNV%h#j(3Dh$+_yQP5FU^y#R)1Jxo=c&~-qpvOb((UqV6IacNTx<F93w9uT2xcX=5O
z?hzTZ%*%P5MogDbkpaPr;8S-;$AxuV#f28eB4?fcz8|p^{>Xxq&pi4bco}j)2}M*O
zS<s%ZY{8#zBQ-xkijyk&qMz|;F!(phEYJ}|p7l)*mv4~<9BL6KPy5P=a*h~{W1e(E
z2AcZUISbzvfs5=KZ2n|p;5W|i#u~E}a%9SL)QSEls`Ca#zD9%41tIm8`eggJG3Q5{
z+aWRjaK?iN30C|9;=fiJ%3u22Xy!2FTif_pjg2i-jcn(Qac{c0v0eRH{2g6u%1G?;
z%Mfnda2N!otewn`HpL0fLVefzfQwhvuV;%eH@?`!i<l$i5S{cl2I4>5`Z7q-F8N;I
zoQ28zk#6gs7`^V$0xU!fnx0}~dk;*L!%AipSAHleQnmK0@NtBh<B;?hFVDeU$ak2$
z0fZAGUtNPJ8{?VLUd0JomV0!Uv&}BqL)O1&p<Y(v*Bm~u^-TaZ?9j5x#M@9_r03(#
zH?e1CZt{j_D63Uewt?MFycPQcx`?oKON=@h#Nynhvj{67O@EHi^oLJTpUXzK-T$W!
zfi*^ru<p449UN;MuP!bo7~c{L3q$WF>QQ2tpUgjWBnuC_#6&{}@U64>)*4I-a_5af
z@{*H_16NH~-VU)<8XxjHM*fKMj^5}4J*p*-FmwNX1PR_m<U?dB(9O~g{j$>Bc?nuj
zSf_(klFpu&R5SiHM*TNChIgBUBxhqzzEN8i()F4P=M?yy!!a`XjIuVv0xF?_&sp#s
zvH7*PP>uSxx1fYcODVW+SbWRqP8GgQ$K40ZOIKRI4Lm+goBGX&97bkezq+v#;NR7<
z=piI9Ufrv#x4Ne_7D@;rgK1XQ#{H-DQ}%b{bTl^p7+91(C^BvLs^=lWBmJe4D!kEc
zNh?UFhclOPhr<!AEKZjJp{Soy0}}pQH+T?!)T2Rl?YnP^pUDjP5UA1F>c}P4j0~P{
zf0I4){@5T#P9L%Q&O4o3j~+WV_n=QG=oAz2%jqcbrg(NNXIn?#9Bs!Jonf?o?m)k@
zLu0TyCYP(s3+6tqqP1qt9dwa&|Ggg9&{i#p^q-qQ0U~Y+B;v;JZ0S1<o~jz(HCOJ;
zhk+0E)UFp;9MXzvTImZaL~FM_S$*+tYWqT&H~qT6&+CTJSh*Ec$bLLtvLY2!$VRU2
zCi*aOk3xGoe@L~@e!;+N6qJ-QNQ1|ufuQsGJr(IUJ)8R-E%0-PNo!6)Yy#OXJo6i9
zM7z~ktxDg>3~}}*mYHZ%8NWXLETP)J1VKn>xMpDH$89)X6z%1ffB6^j-f@_!vmpF3
zOx!o#p*nSSM^RsZ{99<TaGbnK>57k9aAbQx1`^It^Z=*G0parGuF|)LB7+~{*e*h@
zsShMELL3(Ip9XreIY*&v+jXJ6y}d(ikJ?`fIcOF!jFa6$Hq^8$+5D$!05Q>E2iPwS
zwr6*TH<=0+Ys6&js_Z$1(@iXXC@<OdyA%_{n%`7hRK!FB6&4hWd%9%ZWV`gCatK$x
z*y~g5G^I&^{BNcpnneLT#X@Esm*MzMMODSrnto(5#{^`%b|Ht@{CEG|a7rY|k3`^h
z(|fII+z#$A34=_dMQ0XU{tMGR45G8B2#cVh`xuFLC+PUDIN^%psdNCxY3#wGnlP-8
z)D2J;!R>rz@js7INm46XnW@%m`Tjk$a%C{kURG8%3x?u^)G*W-?;qUkpG0dytyV<7
zDHlgRG&gZo`AU;soZ!*(@zPxt%$5rB+%xB5OCILBx@vdLlC-q6So-QQv_ySe>3_0s
zE5G=F5f?4Cyp;*~<l0-k*y%`pa?rq|QJwTOg$(KuE5>b=D`_eH5}j^QKWmITgjJW_
z%55Y!Yl<+WwvuUp8&LqtS_RAg@@4_~*UabUioPVBmp#Nq0}&~4CiHhXTm2Owuod-y
zyvAOjc&3U+k47Z*oOp;j8&$LTm!j1>`8hdqZQ?`&>{RQ;n$MWe*~RY0N8<q2<CiAZ
zw(iH@%EdVE0^@hx;2BTO$?tNk>4gpf^<WM!Djgl2PEssFq92m~jS~LB$>CTSkrpp(
z_%XAY#oGpFDux3GR!3lHSf<3T?5`>xAnTixdihn4d3Nc>24<{PP=UgVf?Nk4+>8DP
z+vgRM;-UHv*;COV)U=HMr|H^;FRG!>;#3aSH@}k7YAfJ>B5w%#w)O95G!}nKOs=+V
zKhmXSbK<qB2OW}!xv(%VH)|}oZS#eX0OYX9<D@bE)O+a|#Fs=M4?49ZeE5fijon0n
zg}mi<3(JpJvZ0k&VvV!6c(zB`GWg2DKjoZwH{@9MNIHxpC4S6#FuOVk6Hz-HU`~nq
z@y<&xe@!|?=9_?+-@BxS{}cgZFyw<5f|vJv^i;{y&ktJ`q!rzTp@rZx)78aAk^$;i
z`jM#VeO&h6!V;f<L1dR@70G+)9o~z4Ir<t*ux0Eup+U{k-VawF^Mg5ALY+3BJ=W}^
zmfJ$gvK130l>P<cDcSB=hrfsD3{dkN`E&7SjipNaa;zTMuJ@E3teUUiPh|GJtE01X
zo!O2-1%Rxnv><HR9Y#nt17m;)n~ew$?jio$uwB$kddt~*)fU$<9Fo%yf)E54^QU(z
zTo`m3@yo`SpkZC^5O&jj<%WSyTJfcLZpqsNZnC4imgA}E+V_#u4@Lq<m`Wbx(O>?W
zDT3|(wOa<j=#AUA_W!XG0=W!#a9?I6kSweFlA!qo85p{@T$}sR-Dhtt=Vn5gq=9*<
z{ULk;|Jev2MD;D2SI(-rW9XYJ^YsRUgi6FMHV~MEq_yKmeyU4e7`(%=nQsQTyCSQv
zl$AEKWx%LyM?V-j<6s%4bYf{JHWp^T%`$(6PHbBaKp0O1Z){om-F@Oh4sA#tpnM{y
zT_f*nMAPjlA;ED-=|45KvYITv^;ZZ#)wxu3Tgg8fZDp$#NeUUCN}IqdF1NAh4u9t0
zJ9!MJ!f|2mF$8NV<IRXqELS{(7$8*p^<9oaQ~BXb6wF_5iXqLp6nFKOlhl)Q249%O
z$7z?i+Kya~T9m%>{{+7gQBsRSJk%iy@bLO;nrlWsXZjm5BY&h6f3l^p-9)EoCy;R7
zj<96Bt#f(;AL55Ht<8yVy09mbt>cNffQO#pvHe^zFHqIn=g&$U2-2+f6QN!;0K9!6
zxkuB*zvd)ZOfiePonAKOv%6R$WMc>X%zuXGCLHi#tP_Q1PYo82_A1@90qT7bINz`0
zKDdv9ua!UJ^BZA2kP)$S|02G0Shb-yCqFxCfK5(3=hgE{eM-=B|CRG$*@Hsdm(l8e
zTXx$B-b!#kWJ7dahNEiedKotJ$Y&(xWz>*X#M_i=h&!#soXyw^e{@s@C7qW!ltgn`
zlhVq1a&mJ03U^=#u7T$_2hNV}{>exs*kU^%WG8vlkfU_pk9D!Me$112wga1qO@uUK
zQxl-j4-diFE~+nSIDBNDy_|1hXJ4DnG);@~B0eD(Wb(aa!i!nF)Nuh8-L>OVnsM?l
z(a=8i-gU%2M~i`SskITt@8pE$9H?jXwLo)Xjk=_$D1*+8?}q8s*SAt{CUQOMz@0SR
zX}q{$wh<ZS=St1^n0ICww>EpM-@NA@YK6t&kk9DEj0_w`5}XdVb4cAszp5&=i$pm;
z`!y;3S0W4MyHxeh_%)?!FSPS}OnY<0^JKQX&KrvHmV{NEj#d^8xbhf|#4P@N;L3c-
zW_pcJ=~Bahk{(Mh*Z1D_?}tUCXe11m)94gUUe=JM+()oxr8>_#=pNcUL%c(!s29g=
z#jjyC66P3$-|;q=`mo)EwJ?;y{O-%6WXEKngSfWMODykbY_`Xfg`!FK)TpnI{-t7I
zd@B7_FLpt~&KDSQhBPiC293_P*2y>iSfA9}_}51%oGT6%VX;uQHuQ0x<bPYMO5g<|
zq8^dvZ`HH{ws;ZZybHYi@YSc6cwQc0K`@b&-KT}PsT|9>eU#Mno;A$Hx1%tT^TQ@_
zR#w&t&}Ds{40#C)1+stWDs3M&g;NXT+Rfi&OmB7we<gXMhY`Oh;u{g3hntWhWXr?G
zh{>LlZ_c(mM2(lWfwN)7Pb_I34V|V@J(k>G^2#9;-^`ncna308*N<pv$VTOmw&0r1
zZld2Hk8W~l3>;Ttex%r%ml!D&p-KbT#3cPBaB|mq{)&VB^RY(20p<xf)Fw0`AT=4+
z!5*QG%W+}e>Z7R(3dOl5!pV-7sMs@s@Al9gxjZ$FB%6HbUmW*1cNpskV7@{pP`u@N
zRxI@uV91I&9^Sv{7??4n@2)*N5Wm&D-T>oXE<VX0CNBN>9zUC=8+fm+_Imc00>qx4
zY)AD^xMTY_Sr>LvP)IOf3)NqtP#q8!?~l#U9lHpv6TkAcSU(v}P6VZB;)(?P1%BoS
zt2vkccqcwUxJ`8TUR_g}#KC2{3$CRR3$h;cmRs%B6Cs^dkOMvLg~@8%!>@xIrGqyX
zc>Qr<gd_ROZ4Gx($vYB&&!~JxREU;(%q!=WuC<{tiFX5SJ`rC`zEdL^(hIpznK<+A
zZyn-3RP_eYyB~gpPezXtt+ZK2WQP)*{yfM@OKi0GqN6W@>medHbF}vf)9g8N=G^hO
z!y#FKNH$cAX7r)Fb22w4C#N$w?=vfIbsym|o8@hTwa(=sStmV~VZ2LGi4;~=7>@~h
z0nYaobRl0nTz*XbEs@O9yo`H9kn+XWyZ7!n-T{xIo^@-x--wsPt~f~%WBAuBTmV8`
z>eHFTBgFzB`zEi_+4kFOAxMC>!T#6humA_hNCtvgGu86B+h_aLZ^EiVbbDnIGOMAe
zC^>nwHeRKYyk5+mk`tc>-uUi6f1E?07ky*3_(L0<gD7ApuR2Skw&#h}lvN-E=G6XA
z*(KkUoxGNz_|IYM6D|5*^2S02)#goM(Fp+4Y_K+j82ECZyO>1_by4A4p@>UFm^Zmx
z%a6SZyy9VVu9wt*)IJD&Ly+jFnQ;Gh>~?tQt+E>UR{Qd2;_v#Z_v%CP)7;I@-UVp{
z=d|wYQbP**mn1#LUX@Rf!AO^z=n-TIcYDA&Ec^g^vs{{q)c%!1{`EezIb6JDY7V}w
z1hW&Nba&aZ^T^408NUK!*ZEQ9Q?o?s&?-YR2~B_U{}CqrPEQ>qJzh4zio43_7r>mx
zuKk|5V)-u^uYZadXEc<O!q7s{y`|swr#Hg<VVw0%P3opSQ}`-N5|X1X6Fi5_+5tvB
ztQlpX`n&cE1(d3Uojm(h+<REg!$k#tjaqNZu}CX4T8m{kn2ypJhgf$_;@!-zSogQD
zCCD)}h0fd7w%6>%{eM`T3CkTmNMEAHX<&)d0ak=r0QFbkP2r=hDGkYuW|Pn^J!Ir#
z)iy?+!VvbTMJ}`XB6pA^(|V5QJf|!f)wv`9L38Qgw%Vc3K*f)-%2eR*Tz2j>B-mNM
z))ML~T8PeV4rug@96l7FB$?rpJ545;#dE@=ryCE3gnmr#;qmbzzW-6FTY^#_#SM`H
zC7^78d0t+$R4bh(<(+NSqLL@>?OdJbfE{{M2>|7{e+N1K;voFxDdzyoI(!ksci?fp
zMw|ALopb;*B&6>KQ>5Mq23znDFP((h6fO&Q5>k%n<lAc_Ib2?<7jN__ag@cx!QBMt
zZk@#<%aS1y(_*t=`t71v0H2X-Y?Q=`&nU$yfJ|qlR&>vT=k1kamBeJTNGkh~Pg#Cp
z9!lM#a`7thL+T?O{+D0~jUG%<%XgYV3VE0?c$wLTb;Gh2tbr*_H_h`4tqQ{TfV0$_
zfs|)!wYlM$!V<BqpHwW{cT3o`ej4lJ`O<`KV-5tA>&oPR6uN%XwistfPD%1ha^UN>
z(va1c*jIP6HAk(<@UW^TI~y|VMAQL*-Jbse5Fn@ug37BwX7e~ti{bGoocW+Y9i+C&
zIYfz-@9FikJ<@(dZ|=2NBzqwm<?Xs~)=THh^M{U*z%-Hlk4*aS-~1E#W+^Ew>JK6M
zA&B?ZSua~J4~*!KWIoaV2X#%@9%(Rhb4^lv$sQc{fNVbtRk2OE-SieeF}Go+iv1b4
z!6#rb1(T4k0T%k-XyMJ-aYW!CeyW|YPxl+ceF$bJf)$&P&R@U`^~v5y#NwF%OR|%_
zzP;TYsd?h0^!JMay6Q8bY%&kH<_&ZYs^Jxd6ciGx>a<=G>UYY(LqyZtRIT!hXI9Yn
zKM*6$VkY(iE0p(h!&}CQja2Y)?ytwgIiie$q2~&j9%r{_%FfkN7v|653Bm;FCj>B0
zaKvB8B7zK)h}Kgg?IOO2W1LiNTP^l9%!Y}0J<jCIxUTQ`wf!Pl(DVlw3YPSD1p99k
z*@~KllnYdYWC%>KazX!HZ<;&#!!TkwVcAcItu<&UI$!0ri1{6S7G{TeWeaF44UJ~K
z{%5D0iFO<mm`zALP4n^`awnWfmFRhh9wF%X_hmx!551#Z=93-xSRjB77<QP}X>4Ur
zD=cDfS@>2iak`5h#dD9{bGrZPIuirKV$IcE_zvDbzWZPLw_X_((!LFah)vKZj_$Qy
zKHDn)CFt?>&#d(g>JNe`#wZhxiHo~A0;M7PSJOTjfiqg@Ll9v-$Oy;IaiE6qLVip8
zlYhMdyr+DmOK)MR0-rjncnabn%zrs)pyI?f(BQ2JwL#S*;_arykZODOu;s)0p(XVs
zwHdK3#7>f&o4ShsbuFyC0<%l%pT)2ofV|OvJK_-#!U337C<kn1e`N6M*RPMI+QNLV
z7(7<Q;<#S{j=Oj#HiFm#1K7@thn`alz<x9EYMSG0i5)&({KalKjT<KeV*d5WowInM
zC`6jZCmAJYm1b3x)JJodpQqW)D;f*CI_E$|LwxVE1n9TqLG0`QO$6+FD3G3yt53EQ
z>VPY9`gJcItd<)c-^!iVcav|032n{KzEcg2KlQ!qy0AS^2SEk@k1FpjsE-Z-xovot
zlE{M%!0u^)o$WFCsR2HXom-5pWi`j~t47!!brsWjd-?d!B)#iylqF*qeA!v3)*mIa
z3#4*kMGP$u%3G_-0t*Z{rwqB_Q*NqwZMt#PB%}{JTmJA-ywHz=7gCph3zJ%w8M#k9
zm*b&&TM;h<dvpmlbo}Av<u$nrOA~pB>I_V?`>-nM2ZXPUJcGDSYCJvS&@7gPy`2tR
zmr1zr5?@IU9V;+l3f~~*uZp-L+TRow)INnbb&BQ@SCrK7+)cQp31b1*UH>n2^~<(`
zuK;r*I1}Fyh(`kMnMa0f^4r!dHkV7u1A<EPM0EC(K1f}ZO;Q-@Nc#VyHvfo}A1^Cy
ztTCsghUmWI6|`K44V~(&A$<~72JO2A<suqN*P9kU*1pBL-_R%S`H^|sh~1Xg#ckXK
zy+crXgwNcd0MWlusW;K^xheSr7T>E+6zrXB+jn4A)jcP(;3jW{+jSE`(P{usQy|P5
zuxDz(`@~qUNNO&L%Dr$^c(6ZI!*mWTra3x9V<${hLP3nJc)Q^5-=M;dbcgcZ$1g6x
ztdKqN6}1WBn@H7eZsmwE&apf$WT-@liL$fZrChTKc(>o51@qo>Zx&tItA`%O^OnzU
zfI3~<?}M#Rz%id*0twf2W^|c)^Pnnu--+mVwrm+$o=RVvHOU&Y91IK$jQ>}HKt;O4
zNfj40gPLK&^}tkydagX(#$@sW!HVyTdXZ)}6urGV2-@b`TMG&b)&*hx>?YXlzx2~4
zviKhaB`cJ5Sp`?nCwk3`ZDBMnxOLY^&uqv<3Pw9s4mH1-F!@;VvBn$06I_;4QmQM*
z0%s~>z=74gA}og7Ldg{ny#a3a$iEw06%K%cjs%}Qjbj3|E|RKStMc%Sv6Y@RK6*x-
zxJABkr~=deZq9L{?dOO8QGFypp1*@e-3yykfKyWP0>1XS4&$^*wia*u@!duFyRYp%
z9EFladV|Irm2G;<VXAl}Vn_pM?t69+M^&63FviaCXQ?Lz4@8EO5sT@64WUKIol61O
z4Q`um4E(-*ZQ8t+k~rlNU5_1QLl-qIdq>7ZcN!ecG(NMADVb0C&^4l#aeHE7B4_*G
z!JIgy5q4OgU&HQ-58ls#CIo)zz1jl9_!E7nzij=HxkzI9Qe65iWOfMivkQ8<f^vQ!
zcq>Gj(7Q(eCGgb+KpJEKF$nm#IAqXQQ1u}$+*kfm9XR4Zvq^aSuVxb}lbh(IEH@SU
zg%@A9RC(Rab(c3>J+*7`Zpl^%lfE>3UXvgZw;kf|rI=V?TDlLR%bybQ`=4kXgw!n0
zlUl+EXnaOGbT9{$x_#@wi2R_~qc3(yD>7|1afdoSGLkCaDYqHRiTz%KOVMGCrTxgX
zuv<G8^K|O(Xkd{R@EMTbI1c%Z&ybGBeiTd;Ml9cc-PL<`qS8ZSFwNr|^^Z?XJtmKd
zkp`_9q;W$M0R!zIYH4ld;W~F_2lk&vCxL-g=oiK!e&f!8<&gG1Nc-FM!E})0xrs(x
ztav%6QHT~_3Ml5&swMflZz$d!Z9t*7X5eZ+@<~FNXa`qH$pw;y`lvhP^d1t}Zre~O
zrm}vi=D_X=e1T-6Wbj)W@DzfO(6c$-{%Ra<U<7HFI&eekkK+W4MOO$azCUrKXHDMc
zr~A>jc{8_u`gDznlVdw_U|ev}<-H6H-{}6@;rfSMM&}`S*W5frxI0ba9UW!v>fu_u
z)RP<Ev`(bz>X;Yi_K>9<QQGc3F*KB$2E!e1{jFH@r-l#X2KuWyQxN(ep6YZN>$HVD
zn<mMW6=S4g$eUfOn9SZ9cZiuO`Kp&IV3bI_ugBf^?ZSU&T;MXa7SVMgxoWuzeCrKo
z3fj3$zVqU@o_)%T5!$hhuJa=LENA-E^~*s1-mI;yts})$e<qBXi%O*$`*>)T`fKFG
z2;k8UNPy2sAv-}>f7IfOwHq7`k_6K<_`Kur{B?+ICB=JrYD9!4X&|>rC3fUM)Kd@V
z*pGsWYed0w6A!lMC!D!6ei_zs9sVm)g>X{}bTEgivg9s4zyBK9zH&V!@CZ(}Fwe-J
zD7k((X{Y;-YE`m&@_6r_25E)8ZP8d99V8xa+<03AiO1kzZ(0r9B%9v_knkepsQwP1
zv@+k`n!&%#_MkGikD>#g<m`g&#S7-%qBCh0sS)+pzf&XdK=!sTUuIwZj|w;#QNYwj
zfrJZtj~k$6t!8;h&gVtn@`7SiLGkDNj~_jj#i*$H$lQB>pe+REwQ~9WNPe&dQx5sr
z83yWI$Uld~%9Tmm+!t~LN0abO>`~ASFv)|j<zX^z%q&u*Gqul#j;SjV$Y=|($U_jq
zR+HS2)d4kM<`^OuKgG9L&-u4eDt}znv9mzry8?`T>u($<kJ{=iT)jEj{(XVZG1o1H
zQ~I-t_0D+VmV_h=UarHu<V5EW^kLq=iV%IZ@1%pdyh(T#W&J$!bK-xZPmL`?W47q=
z&+Hf<8q3`p3M$_?1M<zfSR6)3*vcM8(s9WhVtd%uZ<}Z5%2%mb%jY=oX{>OfZJ9!U
zvPPnKcG2B9*Z7#-0Ab>6v<QmrWLXhU7VCL-B}w`Nqc3*{k@IF;sqM&-6chWId`HIr
z%0c4ylf^YwmNbGk=Mf(ookW!47$0U?3HBrUE2^nbMX9*&H-(L1M+6^ZLugVF*om2=
zSx7olE-CQnElHbkn}z4-#jKUD-z~K-7FVFE8yjop1_lS$IfeeOXj%ZWWE3AY`El9=
z{fW4cv+5_&UM8IH0k~cF&e_a`5dP6mLTgQd4ZBW$$hFIP{q-qBvb0E@PSu44SUad>
zfH>OGoT!d%TvZovYXh~&E>CI=bN*o!u-t)pOf_Y+aPC6jDZITjq&}x)!z^V#rWeFD
zSk&h-V?Vy?nv|$0Mr5gd*_(UsdiI_*qQ$6>HiZ`{cWV;dhRwqATWlqBhUHpyxe4o4
zLhy)*AsZ_m-KJ<`bu{VfwVAmOiGjsUht6KCBo3xm7a|@Q7?6Cn7iL!dm(TU<`TEg7
zhh0y!jd54D1v5|7v*;_|OQNx6=-SZlo$n;(GM*Gt9g_2?mHu*Cn&I}kPoLJ(Yrqqp
z;qvdZyDWn<6E58PMWbAaACXhAlmpH8EBQVr1XP5heSpE%?xGVecu$YiV8@u=&wdg?
zFBuV*38SITMa%K>{G@ctSHwo;>1Xd5@fou_Pj$Wf5{lHwVRF|3`r)x~TO`oO7@?X9
zj0Dp*$5r}JKNFqrbT}-abQeFsW=^lfJ7#U)C>~ABlEHy+S>>5>PaCBma^2y+q@l=L
ziJV1zM4!L3_C88ihOuI+mQI0AnckLags=RU3i$0xhrA%S=#Fk=#ou(a28tY3>IN2>
zC3B9mo)=uN814&>={xM#L4HI-Lug+_RMh$vqrXnM#S>tIWEc#%-ICU}2M7$(LVtf!
z)W&oLw&^u^_&k14BdS;6%)}q*x3)x19~Q!M(0cj37W@0QwzeokgoXd#<Y&l*uiwVe
ze30<<LYNTCr6eWk^J=3h>Nc^(Kih(GXhcqiI`Cw8j@YV2<Oej?EWLmKJ}-r(AEp*i
ztp!L2G{5S1yV9|(5-(?r#3_zGj0f6tIvh&bkUHiE`iubzPs&HXe@;czO_73G5e^w_
z`3+A0jon&Sw9MVsH2!{Ku2pqr1WEU{;>B*&YwKANmyh<-h-%eJ78R)^wK12!qIlo$
z_ON2p!l{jG`}dCuq+BEx#qxn5x^XCStCe`AlVz(TcF&h8h7F>OiGi6wzZgktH!qC0
zF5F2anfPy#+a98#bbTGd!(n%o0Y|YCYtS-`X~i7`4QD}CI&I#}5#qZ#`{ow;{<#|Y
zhJ~nNwJxIgrg$`U`x}4vTg%_Reap!oJLmBK_BFO(B$4z4$K_fAAGq*lOnuZ_@j*rO
zFin;XcB^JbeV3PTs5)qEU?~x{H{FoU)yULz__VseEebq9WiA7Ce6_C?Xwxbo{#V<!
z1+yE<cJ4$Rwf=bfZ@VKdF;}V0ex<|Kh*`Q@N$?|w65Hev^R<~zbf5{0wks$otZMy_
z%={cl(H?U-KoHmy^e_0<hOFp@XciQ=Qc5gcIo4>(Hr>DbvhL;T3pZR`mgEtk@6%(c
zT&MEE4Y0E31r-yfjP1A)I0PzYcPQtF-@910b3k&lE9D1)sfM&zJjTPDIVe-xzBruu
zV8t?FPwMHw=klRNCNaA-L^_(fIcD;^BbQM7VOd}C;Qvc$3J_sLeaHJ5Zx!(2Z)k<g
z>Ee@Ov7?9ZWJJqjMyiV9#miRvgevQDZeP4?hSB~oFhBjOi6oNk{7mZ9$0?Vj=d4Q6
zZHRSQSgA%LRm0sG9YKZ({rI}dzj+UXB6rFxC<){d#=Rr@DavLgzdFdPyhEzzyMz5L
z*Kbe7Z_KET^B#`O>|VI2pO+u@eF=0x`v0l}kQU7|{muCLHd@~b<y}7U^ZI_-eXhw8
zF4;{@G2zUv9&;ZpO1<_V2H?gUtIoISSR?s|NXMc_rP@Qvcke@H4p#AYE!@Z()FW&u
z!lXgUXcA9AM2Ej42L|(F7L^=H%*~-#SgdYn-xqFikNikziJGBMzUd7Kp;1@}=P3d<
zAG4c^)U)yZy)zryy&Y_F{;Vn%W-cRJ5+-uO&JaI0Vna!nbU_;)^Rw(of(4SGeUVnU
zo+GnJvyi*zk*p$HasBziZ?<Gd_*102J5N_@?=RnVcU{MHZyb*2{!I*=ZV@<(nd;A}
zqAF;{Z5jiECjv%v(1?2{K4;HTA6n9=+D7kJl%A;-OMjrFVxN?A{`{F(5!=H|PfBbZ
zgfo*vLqnB}|3@{QkB}$|wT+<D$8@5#iUy^8r?@68*gUu%TDtb!d2^aLNDZhT9Qw1T
zr{@gTWD@@V-%f!u4|ic+({g=c$ri!|xzgX)TLAvrR^CytKsTc|+hBZr-GjGsUDU@~
z6t&DQ5>qzplk&-Yc6RnH`{Veyzl<^3xPAN&fDKOXfAo%u@Cu@Qs&B<J6P+H@@9s%P
zur{2fukCv?Qu$?C;-bMlS4$3)if!hDbMtcXGohls=4Y5*PLR#1l<=2Hm_(j%;9jcI
zc>B7R?|eWD=uo4gyt(Ng%O^b(LPncMrhk<Gh`LzhvFfO}YNM6cp|sj!c{H`^#>Yvw
zXY;RSwcrn%QpP_%U0RKJ<@4i9!sqht8^V(=*Ik}pT6NJ(y)kZX4o|r8d8u37YSxUI
z;;2a0fq!y*dd<e?NAyxPm>sV%|2RI{eELF>pM}Ts!nOV9(yjKHOBCCSpKMK%N?Qor
z9Z1()o}cPhTU(oIGk!f&GV<e_@a({kw_DTPukhrn(?u4vE?JGmMlAhUjF9>wdAe0&
zu&_f`f}Phqx7gwlMRV&*J-_tzo$2InRz{j{tX)`6FCRaWEM-X>7~I%<<BeTU=>w~K
z&aP}eHq|uW!jfywZPf-p$lTm{b>Z4#j?<5Y3#&FeE)^GRYp!aMCQGf4n15-E()=-M
z(cJ8{)T?Zj?6K;xx^VVP4aK!qG2g6p>tw&-5r*)u#ailhiAUMLcB-dK`et>1TvRT1
zflZtRjS@NE$JNyrQr0hAOL1(DUol~`{*l+njo$ljGYkImbr69Ni;w;IsuGPNi?3d>
z&BXJC)m|R{wWvBDWm}t**C?1tl{yxd!#xXdCOtnREOc>hZq5oT3ei(2<GM&6$;EfO
zCTfIMt8c93w=2^6kx>nWmyC>zv_GF-Z_lATcEx4usLpT{Y71%K=@OR^Z&LyM2r_+p
z;zB*Kd^_VloM}ey0@FNlFd>1!dLO8P^CCOR+|zd@uY3G(?S9|iEN+*kmuj9viptSw
z7X0!wJbc;ZmrX0#LiGN@M2Fvjo0)vbL-cduzPp!3@nFMW0;Wu~mH7qDRl0t>+x0}3
zMWW1oroM`whN7pNp4~IoM65E=5p~bFDK_sxZOV2%%_~>Lqi8n-NPrZhl{&Ec?G$<x
zn^m|A4)s3n;!gyxV#7VWLp6nwJD6}0x1j{oD5n;KXWgLJhZoO!m~w=SIg}bFEI1S|
zh!F=TCx<@p^YdGG*<XI%D0&OtO2L^vzPH%YK7{K!+IgwOMMTXbf?$~mcLi0ru5Cf5
zFT}DE{^TM}@!F4lLECsH`5sCKunKV~ainv}NzM!vflA}eYyg$UZ}dy0?L5a)`0<Cw
z*v%9N>=dC7donsG>U8TiPJ;h|i6LD?$8!pCVY`*!O3_r8#pg@t-}{Z}6d#WMv0^so
zZmhc(gJJAZhA)3UmNQZ9@v_slns?VO!<#4A>SZw*nx%SdL$T=za9j0CxQjfFNp-Y1
zjdW#0P@m!$q3}>RU5LNox3{Ckh}@!8e4T^Ya2%~n4Mslw!hZ4j*~9*p&xAk9My9Sh
zsXv;T+0iEM$rYAjFfr{ez7k&1L9C}M6N<mIZFCIeIa?i*a4QG|thbJ~msC37gAf||
z5?7M4Fs1K!HS9_=6=ULO{gMzJ&2TI0{7hQk|MB=txC^u7f%b5XYrTwbOB(Jp)||Sa
zsOYgN=GDK6I=(_drH!2hCvXhjQ$OVJ2W$#ALn~D70YUMnflK;;XAg;bN06OgFh}k^
zsoKKH$6qR+KO59{Izw|g&B*iKH4l%qUH3%K>JS*glIlgH8Lu8mr}8^oSE#@(VE*0D
zj-A0b8x!~fs)@+hnr904y9qCJIjF4TIZA)#B)w}xlR%5p{Pj7JPln9(5|5q_3z4Aq
zT+1FB_$eQHSj=s4YR@KHwT*^UuP<e1#z`IQ?O>jhUY@-BK1~(Rx*Mhi{TvKC|5tlC
zLV>|rfko1$r)uu=D@Wg#<%f;UtoAm(+UK}VjD*}DU)M>xbNcrBtoh08BD#CXe`mZ#
z1maGkn^D)gn;Z`Jo0hm}MCJu^!Ll{rQU~$pO@+Rd*irPG72QKVkcgPNj1}C;F`swT
z6nC$!GfG$T#BQfj{Q0oh{e<)63!hde5)SLEF!#dsv9U2(5|Kj^1X{Q3Ax(EEVs2#$
z!nR*ATm=e}Dn&lW-?=l@RL3bMqjwDW7d$6-x_E4kOMoD{XiuW|!;Rs+_a@fz*0MU9
z)fc3|b8c>CIwtuURozXrkfQyqYmGzmC~lh$T$dLeX?jBu^GOr?Rpc8=N<b7gMSL3`
zT(mVxQkkpmi5wQ01+SlaS+Z6jxfXvBLZu}7*kU`I)yDm8hwC{7L~Phj9COW^{+zb3
zi%Hzcr5Qy}P*NXgU+m0rEyA7#g8|xYf1iTeVue?W2!Eh$zz^h~XsaSjv80Mm%+;V|
z@|30I>M^=@!Z|;<eERslw9##JCOALf=Ukpt*YjaQbA>;bKkI{qYeGTT`4lBD$S+UI
zxNrACzz+YCZUHyY$vZcNp7k1-vY_Gx{cj3uDtae}FZK?TL~}$bbbOB;IHn4_2|!<J
zg6(!cj&(TcNcDQo$v7?Eh5Ru_h5UANfnt`$>|XsM^F4A11l*Pr;*skrDb5kjusvqu
zCb2)NkkW(LfMzi6gU}C%dlpl}_mF6Kup1K<M0R;h|6{+%ALUmlQ~TYZ-&Az`Sr^ms
zTu)x5YscyEhd}GFtOSMX=#ey6ui~N{X{roZim+i~c3vzQG&oaA@Iw8WxAH%n5u%qN
z6ac|!zW>6p#yJ)z)|x}<@^wmxJ|^e~kA&F;p*wpnKYP2M_kyA1sU~_lx=*xx&SZEN
zy+Z}6>@@2kGVEv6gMn>E$!?yGg+Gxt@QbN8em^`I#w_BVoV%)RyE-FsRNJi3?tN-A
z6Z`n`(Ovl^4I^d=Wu>L1+q&a&S%s7dWYQ-^g@=a+9kfrzZ9WiC24O(wIpDhj391A#
zKXKsy-g%~uO)et;+LGq7Q*Zg2Z0^!~AGR}j)Ra#;J>&!8Y$4HBo}>5Uen~Q_n%q2H
z(!|!v%xr{Q;o%hmeRk@wxYwSXJH>$eh@s4zKho+-!Zsn?addow3^$;@@S29e*-2>8
z#{A<p*8PdIFJHZRbE8_X-*DMo_pzKr8f$*455+`aU|{DF;Xr(z6LW?9(H-qH5)*6e
zbw!ha=44_q+tlZa<5gu_V9_TlWYF7!`-@jljX@-Dso5Gr#f?@M<%OhjHLH6o9)WF?
z9cYRjp044j2#t@ACwFE@KT0)}iT*q|FO2P2cBi(Ev5?JZD)7xE0{LT+f;ajOz7K0Y
z8j7nf{bEpb10sM+JFD?6BP{hS{*gVB9zzp3!dDoZ+MSHcJ6&$fx}QvUz7WlqaYZM!
z(_+aEZt{!*IW2)KGY6F`%LN|cPBaHTM6-sj4_=KC(?emQYtIo0{GlX(Gh6Ja@dYI+
zn1;3%bjEYPqK)TqC9*q#MeIyq^&%3F?v;N_<ja?0NNGYy0N~$t2F!vexDr&W4%p&|
z7lh2e^1K9w0x>;A4z5H+CXWbSh)VcSHg0;Uud`<6bslbaHf<$?O`r|qN)ZPP7M2*<
zFw?no?XUS=y_O#zmVGNE=e!r+>IA9csWQdM@?fhLLolzH&k$_NiS$zw_=s=|q;pqF
zNVp|98d>nVhh-h*y}p`|e)mgO<frO$#qJ67XU+$NG%(0^5LxQv()bv!hg6Haw{wxO
zx+X%v50+43s(j0tun8U*O?G+mp=h%0A6c56$}Fzo6<bJRdGLpW2%5BXp<Tn9{+`Ff
zW%bt_Q9cvxE?tzE1E?JSl!;v1-Vd;7#;?t(krq`(Lz;MSNly7_qB_h%d4gtfJ6N~;
z%HmVzk2D2r{jcIVh(hN1R|5`&ZJQfsq~g4HbC+7x(YlRE-EAz^UEfxB-SPJ8R2Fzj
zZc`LKb+BWu@=N9Dh~Fw}hH96IgT-L{J|-mv1^1tKz4-`_n&Id=+qzQwU^k*{(34<`
zibNU@;YEQOGnE(Gtil#*Zi-KJr6L%U@mRtd9OcUx>o9de=dQGzT_@jV(#JGEb8-F_
zTkok$oWK*_@OjK-V9IyD*vy4Z{`(ZM+c}99F#!~EXG6ts#`TXb@bX_ggS?#v>7hEU
zl-@^hX;O*leBFD@5BBhD%l^ojYZx?>YUyOklB@zNJ2&q%TDaaWM5fo|B(O4BH4&di
zNX^-@>3`5mOZcFVJSSQSL{RdbS@6H#1;AE1D1}uEewiU8uf`4`@@HexpXkWN_$@xL
zu55uaMd*OfGbB}DS0f}ORQ&i?W(wlzsioGt#rtc#4k8ze_Jn&}{n7Xb3@~~p5|dpG
z?k9|MH#w+WG>VbhRmP!xdoGppd_dkkrAEt9Z4yQ%XG|;oYmr<an8dM}2XZ&j#;Ehg
z-B{2di!3}QM`~|V!8~@!D2o-G>yuGwSaWEI34s;y>>o{pyiz|Pe&dyFSk<zq;3Zim
zp;$im9-Mi8-Rr_93=QcnKW)2DmzZv_`)hD;aAe&M6+)r5!VA54D{<l4tibM%SW>yu
zW_0`r(&)wN$eZYQ#P0`hbe!Z|+%Jz;VJ1qD=%0Ufmf&aY;LTX~VP5s@e(|E1!9IW^
zDrxn@sfDEChTY)nMs?^WsY&kX7)$k0RF}bn9-dq&dGLnK6Lk)8BiN9XoIy<dA1S$^
z5HLln0gZh4^XlIw`mEZ<BVXQmwBq2sFNHmUiLnd2UMso#1cTYr8n~I5nCQvzt7gT*
zf$DZxFi^bv4=#Hgjojti1~Yua&o9PQW&|FQ7iy#?R3WnHM5pfhd7hMuNze8j(KOw!
zrRB6}|GL<i*X)Hp>b(8(>3OmVJ3G7e9Q_CKxNTfioEhlFfI%nBmq14B_vS)OIeYfk
zO9)~xgbW9>ByJ>_L7KBj@H$daRY$=ckyt0~IlB^n>PJG<#XXn9iuTtke$}EpDQ>K%
z1rcng;5kC|0QltU2R$~~4@YsMk?ZLpOxL?C-)XYcVK}^d3qc9seOqeC2<ygF1{2_f
zEb{uDqf*l*_kWtHU*WG}7SnNc<!@<_IZE6jhNS)7YVL)XJ@eGerXvlPc6-qob`HFC
zB3mAT<HwD>T*EW3{g+>cX#lT!)ublyB0)@NkRkbz#^v7#;=YaUelKUJ=uCnmMRe;b
zlIGrvmda<ii0KC!Ru~%_ugm|^DQI6J=rwWjy?YJUbsEi&&@jUcuckQt0GHJtY95Mz
z8!u5Irx_-e;8!clU{=%Dzg;^x$-f}9P+Qy5%H+-$He$@&!lZ+FHASCj)gDsxZWY!J
zNs8rbGUC&+$}(FyT7qqw`VnI^1IDOIWKi%=V}z(NU2ijjU02?MnJ(*fm(W?Qw3|NJ
ze0;|iwmWZ0+XRs~mQ~Y2ifOIXbe?Z9Mg0lha1(9k1O>hIpHXbH%UyOzK`<8i5_}+>
z^nrNXfjVB1fGXE+y!4|diRdQZu2*IhExzl>&Mk@$ObmE1t|5L1f{2EZk>p)rZQ-!Y
zL^<$K2;o_Iw4{3V-gYNNx8W)?64W5!q$)xi@4)}SP(u|=x2H0MpxvT_N>(rh9jz}5
z!(YQ!gpZ8SUFn*?Gv@UjAyY=z&d-PI)mX@K67K9ksLL{m>UCz;*ZpW*R9$_6fdp7)
zo)!G|o3&xhTZ5H<S6{~v^=0&ujez(bx}ZP5$Wgq{H)N&glbNBcUXpafwL0VS2b_r|
z2Jv+5kP;C3`QhF!sm=OX3M036*WfJn1kr=Q9Fc)$f25O@ksiis;m+3t0k>xra33O>
z9Di|o-#evZ@x2z;-|Xg$dxFXt*Y^4NUW7-`a@QRx$G4An#}br<zbr7tfQ|Yi?DSR_
zq&$-SVjeGEf@I&SFC}~r#yd3iYM>w!j$!WvYy1W@BxkNr%N94vKj6>}8!@zN;eLvw
zafD*OfB){cDUH*oT5`A8?UVOKtj)GUZ1#_@IDN87lSJDiH)mPJMo31$%`MeW>y4zy
zj1HC8L5sQ8i?(=Kh_JR?^;(YFR6giaW;sv2Z_QSe|1P#@Q-tXHCROu4I6)9_p}HS6
zLxk|UH+mc9PQirdQMnkU8_MyfL;ba)OA|bS3e~hvwDm1}<m|1D4)5vidgqlW2;eu)
zH)xfyIRhAC51V*PZLV-x-jJxPacQ<wp#VYE7XUK(Mx;6)2uvbqMZ)Q5kp0@fqW-NY
z2`YxDb?x<hlG?LMQBe;9SSO@aTR5XjZ1*MRXmAJB-P=`VPJBU64tD}H8PWSu1`e=_
zk%(2qyoPYKYHcTBq<y^)+KJbYUqj^P2wq{_**z8WVzYFu6VWNTO~hvit!fmEcUR74
zwjY1)fq_zglVB!9P|Mdh=_W<u0M==b>xzRPGZ03Ic7$_vXqw^Y`x1@>K8QLJcsL9o
zh&Jqf4k@#QO07+vNU=S$sM<~BD1xdbY7u+q4q2hPutrBh!V?@RMQdivdBL-Y=nQFr
z-!J@pJ3)_b(5#;hWV3CS-<6;tT2b;GOu1CDmBkhX?|n-eeO?XyY^*$>A;D^`&{Z<m
zZ_?RkL|5Jdo6kQP5pxj?QQu|YJpz&dT?eodiP*2$p&OfxN%S*>_v|_h31|5hSki*o
z!`z`-raU}6o7uhITcoa$-i1QgZ?EEo3vPnGy_*72mPFL~fNxVFU-FJ(SFsL7fSc$T
zto4Tn{sh^X{Q6U2Awi3|Qss^>-fMnO_rvqey_XGvgU0hJ3K0RMZErJzLx-UXM~`Ja
zOv&P6F2BnYEC;|ATJUP^%r>%<k|0tCYP^Nhw;g%Dae;W=F-6i`wQBMN7r1^O#>;V~
zG_gmtWhg6~L+{PIBGt^-rR5i?voqhVCWKoq8EIbV!@Dgssw^wQjQiFeie<k>&Vtzv
z26@}v0{lxbl*DG)TJaya-xH2=2N@>l|Am_X*+j1s^)$CTie;@uD!PakOEMK@Nq#?1
z`-+WPRIr#y!KYcvX#ZN_&*7=9Kko#e+Cg~CPh`W_x}Cvo<VE#GaG9=WcsCOa%W`tO
zV1$zr^=1g4Q$)7^+3JH%3Ss^?+RK%*idI(5FCW=dlP8CoDk&+gZ_3J}*NCq#R<G~h
zi*-Hh`9bk@D3jx#gVzplTouH0j0}emwmL&;bNaz@TB*&9?+tS)Loe3?wfILTwJglN
zmP6zo95%)AYP2${oBy+u)Ei|X?!Cw#eN!lL$q9e>heFzlia-+Adw5@MyjdGX(zy~?
z%*<4n;qbgHB2hQe;^NB}HK<<5cLqbglStNg|4&6kT-kP?d!NdhBo_$MCrtyPdw8Z9
z!GW3Kf&=O6rZ^waCnHz~!u81=hFddD(2614TE~vkY4_|GbVY2*jhW_q8tza;nHz}(
zWv?{3n_pyGE|z<{ThUXSKlw49X)!&dw<)ggDXWxW$38;e5wb25d>bK5%kHUW>z0eu
z_`&~APO-T1oZm~+M{wv749yv-`9aV9+>%W%XMp9kXnA<@N@m_`C6quVM;76Dyq=Lc
zQz9ZIdoA)2cJ*&Ql35*0B*11jLa^vLM10q{wh;a#!x3Zs@N7Wo`D)d+C>Q4&lLuaC
zI)N)@X5<S~E%iSA+HJr<v31{?%>U0XsE;V=7V%u=kKz3JrCsZM0H;e8*+fA@E*+84
zN0He0@71AnKqzPSUw*s(F(P)EYfoixl9dn6ibGkV;0CJ^en4;#`D-Gm4q2(qlSdjc
zIg?Elp^cS8pN$N24mFdWh2UJK`)8oTve#4yD$M}Yow6Wi>c>HB(t_b481*+&HvfM`
zTSGY|JyMHHd#*;!OpcEYR)!cm>drhZ2?oP%G3t?dGBYzX%1nKL;7jR$B?@4*i5Y*P
z5H^h%A2a~B3jUJ!1{r{S<}xMKE`o>?&9e&}m0VRC5iT!JA8&OYS-G$p5JhB3b(xYQ
z*SI|)A%Ust;YC6`$3Tmj3NbBYKFJ+u_zdB?3}axr$YC`?08C&QQmR}?vC}GTOrgX?
zlB#+3J<mK(vo6>2t#r{TOVc~XSohVtB92_*>v1QtcaKNC4G7Y-0Km&n1eElLCHnJE
z$`J2W)Q*dgm@cU&=KI9EvMlO*{QY^Ih9vJV4KlUWwa=C06iEQ+h{YVa9TyLIuff*#
zKRlkO>X3kL@Y!E*j(ShbNyU{qdE>R#T`SjzhG&(E#^&bNy!|Y9$$|+dk&=SV=U6-O
z9KDz`JwDa%SFEihaX!E)XSw_TYwz4+n!3UOZV)GP3uGA+){1UIhS>(JQwl=?2kK;O
z3_(F4B817K;IQDMl+ub*nM!$thXxAFDKrDy@+h=CYS9S<gKcbw1zNdlZ~_G?6|Z2s
z+wQnoYQUIff1mr;&3ALp`}dviob#P?f;V=?ewR==_&y%5Yz;}EMT?`9Omgc;1Cl|?
zf3(vS&x%2QCQY~K=n3wZrJOFA5#M*uQL$g#6f!%i@!%ki3W~UjT#KII3q0F;+i(du
z9veN8pOu=%B!fk8cR3Y}*$Uq9V{+onF^*rLDniLrR5A%1<XTAdQ4wt;-?M6zhiDGx
zb>P(_dok)l`ygYJ5*O)veZAtEH1Svf#yKu0R`htrrmqVYiapEu^f3x<il6Dr*`V1{
zz;2oW$MlSl1{p0)r+s2(>d@afIx>4+Fqlr1tW6k>VD#<(I3=E%dQaObo7Uy&V-9;3
z)ZW*B_(KC-h$J5solEtvV+(v+1!Y-{2fC>a)w|UZk#9=xa*&nGc{~F-JH)xfuST`|
zTAWN=0Sm|T>Zqfub>)nol=X6o=&o2t1*KNSpP;s*+#3{j6s7A=gd)D%PBzN)cW&dx
zK$$4Noyr4?^le!=F<37hXGrft97`f>1&LTXhrm3BQ+7m%Qb0ZfF;3glF#jhl`xt!`
zo}GHcZt6a<U|{uRg@dtjasfDA?!R=!8^dahwmj1}<MnjkTYF_wXsSA?c0Jyw{wf(w
zos>=~xV?FODV=3{G_>64*MSQcuhp=n7aIs_x;~pUMHW`mxvcU7hh1}6_ue8*O1Ef9
z&(NAltU)1v+?$$GsV0K+&2FL1R7^kiB#AnsaofgTcQUkZ_IYpO?DB`_jnR)TC`VPl
zzp!A=yEdYAX}YkI#d-Yh&j9(jt#eE%Z9Ml;7g_+A>d$e^EHe3(OjK{6Y^$hB#)cor
zWHRH*)wW$M6D@z)2xh6B3?K&Xk1+`C(Y_x?UhKvMllOw{-U@Ep9=xeI$-}#<StULF
zaM<HctVE?+A7Z4YkjXmQTtulw16XX-r4iWg0p{06Nj=L<Vc=RJYCzCrH=JR%DDSd?
zwdT2#<yS-kL8oi%_w9e%erXV|iY9bZt}ft^^FMw4^T=~mHmc*wA&){4Uti}^Fd7Or
zBJ{qOmwntVom~1@GVdNcFaQqh(qDLK-Ez5>u}Idq7HK^)%npdaBJ#?02}+c*zL-0g
zIW7@xp7}l_<MR&MvkXKpKSHq9cFjqKkDM|@6Pk-xnLCz$ZS+?b&Gt^Kgqb(aFNXh1
z&-lcO75Ma^9bfMKpwS_>F(~U8(2g6<)Yxfjulv;6QwAPB4U=?UUDV~~;<O~$EGY*4
znf0PL$HAtutfZ%>M}Uf)o$LeR5)&UfpV_^`?1}K&#5w_EZ1m-|i0pT4*usO3c)X!I
zR#PADo(qn|w!2-{l=dIeWv#IyV+A#fDYMqQ6}j|Jj}`jUb>PMaO#8LL`O_ogRVPBu
zhMT;0d9MmIk0@T%0L+bK=*r1=Rr5CN#lh|NajSUwW0H3d7U{$Lbyt-2&CCZ4M8U;F
zZ5Hir{&{)b2=~PO)f>#U#5hsY4SZ)ALU5<Ec=Di~Q|ol;Y-Br0eH#0O=eJHG0{#<)
zw(KO)i*cYOcjoGVlGktmKP_*9Py&&GTtEsS1&{(r0i*y@04abJKnfrQkOD{nqySO?
pDS#9}3LpiL0!YFCrl8h-!6cShZH@{^lCJ<y(9Vz@+<>@Ke*;LjI~f1~

diff --git a/docs_old/overrides/home.html b/docs_old/overrides/home.html
deleted file mode 100644
index de5b2258a..000000000
--- a/docs_old/overrides/home.html
+++ /dev/null
@@ -1,154 +0,0 @@
-{#-
-This file overrides the home page to use HTML tooling
-better.
--#}
-{% extends "main.html" %}
-{% block tabs %}
-{{ super() }}
-
-<style>
-    @media screen and (min-width:60em) {
-        .md-sidebar--secondary {
-            display: none;
-        }
-    }
-
-    @media screen and (min-width:76.25em) {
-        .md-sidebar--primary {
-            display: none;
-        }
-    }
-
-    .mdx-container {
-        display: flex;
-        justify-content: center;
-        align-items: center;
-        padding-left: 1rem;
-        padding-right: 1rem;
-    }
-
-    .mdx-hero {
-        text-align: center;
-        margin-top: 6rem;
-    }
-
-    .mdx-hero__image {
-        margin-bottom: 1rem;
-        margin-top: 1rem;
-        max-width: 80%;
-        margin: 0 auto;
-
-        @media screen and (max-width: 40em) {
-            max-width: 60%;
-        }
-    }
-
-    .mdx-hero__content {
-        max-width: 600px;
-        margin: 0 auto;
-
-        .subtitle {
-            font-size: 1rem;
-            letter-spacing: 0.025rem;
-        }
-
-        h2 {
-            max-width: 460px;
-        }
-
-        @media screen and (max-width: 40em) {
-            .subtitle {
-                font-size: 0.8rem;
-                max-width: 70%;
-                font-weight: 700;
-            }
-        }
-
-        a:hover{
-            color: #A7623A;
-        }
-    }
-
-    .md-buttons {
-        display: flex;
-        justify-content: center;
-        gap: 0.5rem;
-        flex-direction: row;
-        margin: 0 auto;
-        margin-top: 2rem;
-        flex-direction: row;
-        width: max-content;
-        font-weight: 700;
-    }
-
-    .md-team {
-        margin-top: 4rem;
-    }
-
-    /* New media query for smaller screens */
-    @media screen and (max-width: 40em) {
-        .md-buttons {
-            flex-direction: column;
-            /* Stack buttons vertically */
-            align-items: center;
-            /* Center-align the stacked buttons */
-        }
-
-        .md-buttons .md-button {
-            width: 100%;
-            /* Make buttons full-width */
-            max-width: 200px;
-            /* Limit maximum width for better appearance */
-        }
-    }
-
-    .md-button {
-        transition: background-color 0.3s ease-in-out; /* Smooth transition */
-        border-radius: 6px !important;
-        padding: 0.4rem 0.8rem !important;
-
-    }
-
-    .md-button:hover {
-        background-color: #A7623A !important; /* Desired hover background color */
-        border-color: #A7623A !important;
-    }
-</style>
-
-<section class="mdx-container">
-    <div class="md-grid md-typeset">
-        <div class="mdx-hero">
-            <div class="mdx-hero__image">
-                <img src="assets/images/logo.png" alt="Outlines Logo" width="600" draggable="false">
-            </div>
-            <div class="mdx-hero__content">
-                <h2 class="subtitle" style="font-weight: 500; padding-top: 1rem;">
-                    Structured text generation and robust prompting for language models
-                </h2>
-                <div class="md-buttons">
-                    <a href="{{ 'welcome/' | url }}" title="Get started" class="md-button md-button--primary">
-                        Get started
-                    </a>
-                    <a href="https://discord.gg/ZxBxyWmW5n" title="Join the Community" class="md-button">
-                        Join the community
-                    </a>
-                </div>
-                <div class="md-buttons">
-                    <p>
-                        Follow us on <a href="https://twitter.com/dottxtai" title="Follow us on X" target="_blank">
-                            X
-                        </a> and
-                        <a href="https://bsky.app/profile/dottxtai.bsky.social" title="Follow us on Bluesky" target="_blank">
-                            Bluesky
-                        </a>
-                    </p>
-                </div>
-
-                <p class="md-team">Made with ❤️ by the team at <a href="https://dottxt.co">.txt</a></p>
-            </div>
-        </div>
-    </div>
-</section>
-{% endblock %}
-{% block content %}{% endblock %}
-{% block footer %}{% endblock %}
diff --git a/docs_old/overrides/main.html b/docs_old/overrides/main.html
deleted file mode 100644
index 94d9808cc..000000000
--- a/docs_old/overrides/main.html
+++ /dev/null
@@ -1 +0,0 @@
-{% extends "base.html" %}
diff --git a/docs_old/quickstart.md b/docs_old/quickstart.md
deleted file mode 100644
index ec445dcba..000000000
--- a/docs_old/quickstart.md
+++ /dev/null
@@ -1,228 +0,0 @@
----
-title: Quickstart
----
-
-# Quickstart
-
-After [installing Outlines](installation.md), the fastest way to get to up to speed with the library is to get acquainted with its few core elements. We advise you to take a quick look at this page to see everything Outlines has to offer before diving in the [documentation](reference/index.md).
-
-
-## Core elements
-
-### Models
-
-The first step when writing a program with Outlines is to initialize a model. Weights will be loaded on the device at this step:
-
-```python
-import outlines
-
-model = outlines.models.transformers(
-    "microsoft/Phi-3-mini-4k-instruct",
-    device="cuda"  # optional device argument, default is cpu
-)
-```
-
-Outlines supports a wide variety of inference engines and model weight types. More details on different models can be found in the [Outlines Models](./reference/models/models.md) documentation page.
-
-### Generation
-
-Once the model is initialized you can build an `outlines.generate` generator. This generator can be called with a prompt directly.
-
-([Outlines Structured Generation Full Documentation](./reference/generation/generation.md))
-
-=== "Text"
-
-    ```python
-    generator = outlines.generate.text(model)
-
-    result = generator("Question: What's 2+2? Answer:", max_tokens=100)
-    print(result)
-    # The answer is 4
-
-	# Outlines also supports streaming output
-    stream = generator.stream("What's 2+2?", max_tokens=4)
-    for i in range(5):
-        token = next(stream)
-        print(repr(token))
-	# '2'
-	# '+'
-	# '2'
-	# ' equals'
-	# '4'
-    ```
-
-=== "Structured"
-
-    Along with typical language model generation behavior via, `outlines.generate.text`, Outlines supports structured generation, which guarantees the tokens generated by the model will follow a predefined structure. Structures can be defined by a regex pattern, JSON schema, python object type, or a Lark grammar defining a parsable language such as SQL or Python.
-
-	Example: using pydantic to enforce a JSON schema
-
-    ```python
-    from enum import Enum
-    from pydantic import BaseModel, constr, conint
-
-    class Character(BaseModel):
-        name: constr(max_length=10)
-        age: conint(gt=18, lt=99)
-        armor: (Enum('Armor', {'leather': 'leather', 'chainmail': 'chainmail', 'plate': 'plate'}))
-        strength: conint(gt=1, lt=100)
-
-    generator = outlines.generate.json(model, Character)
-
-    character = generator(
-        "Generate a new character for my awesome game: "
-        + "name, age (between 1 and 99), armor and strength. "
-        )
-    print(character)
-    # Character(name='Zara', age=25, armor=<Armor.leather: 'leather'>, strength=85)
-	```
-
-## [Deploy using vLLM and FastAPI](./reference/serve/vllm.md)
-
-Outlines can be deployed as a LLM service using [vLLM][vllm]{:target="_blank"} and [FastAPI][fastapi]{:target="_blank"}. The server supports asynchronous processing of incoming requests, and benefits from the performance of vLLM.
-
-First start the server:
-
-```python
-python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-Or you can start the server with Outlines' official Docker image:
-
-```bash
-docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used.
-
-
-You can then query the model in shell by passing a prompt and a [JSON Schema][jsonschema]{:target="_blank"} specification for the structure of the output:
-
-```bash
-curl http://127.0.0.1:8000/generate \
-    -d '{
-        "prompt": "Question: What is a language model? Answer:",
-        "schema": {"type": "string"}
-        }'
-```
-
-Or use the [requests][requests]{:target="_blank"} library from another python program. You can read the [vLLM documentation][vllm]{:target="_blank"} for more details.
-
-## Utilities
-
-### [Prompt templates](./reference/prompting.md)
-
-Prompting can lead to messy code. Outlines' prompt functions are python functions that contain a template for the prompt in their docstring. We use a powerful templating language to allow you to loop over lists, dictionaries, add conditionals, etc. directly from the prompt. When called, a prompt function returns the rendered template:
-
-```python
-from outlines import Template
-
-few_shots = Template.from_string(
-    """{{ instructions }}
-
-    Examples
-    --------
-
-    {% for example in examples %}
-    Q: {{ example.question }}
-    A: {{ example.answer }}
-
-    {% endfor %}
-    Question
-    --------
-
-    Q: {{ question }}
-    A:
-    """
-)
-
-instructions = "Please answer the following question following the examples"
-examples = [
-    {"question": "2+2=?", "answer":4},
-    {"question": "3+3=?", "answer":6}
-]
-question = "4+4 = ?"
-
-prompt = few_shots(instructions, examples, question)
-print(prompt)
-# Please answer the following question following the examples
-
-# Examples
-# --------
-
-# Q: 2+2=?
-# A: 4
-
-# Q: 3+3=?
-# A: 6
-
-# Question
-# --------
-
-# Q: 4+4 = ?
-# A:
-```
-
-### Outlines functions
-
-Once you are done experimenting with a prompt and an output structure, it is useful to be able to encapsulate all of these in a single function that can be called from other parts of the program. This is what `outlines.Function` allows you to do:
-
-=== "function.py"
-
-    ```python
-    from pydantic import BaseModel
-
-    import outlines
-
-
-    tell_a_joke = outlines.Template.from_string(
-        """Tell me a joke about {{ topic }}."""
-    )
-
-    class Joke(BaseModel):
-        setup: str
-        punchline: str
-
-    generate_joke = outlines.Function(
-        tell_a_joke,
-        Joke,
-        "microsoft/Phi-3-mini-4k-instruct"
-    )
-    ```
-
-=== "Call a function"
-
-    ```python
-    from .function import generate_joke
-
-    response = generate_joke("baseball")
-
-    # haha
-    # Joke(setup='Why was the baseball in a bad mood?', punchline='Because it got hit around a lot.')
-    ```
-
-=== "Call a function stored on GitHub"
-
-    You can load a function that is stored on a repository on GitHub directly from Outlines. Say `Someone` stores a function in `joke.py` at the root of the `TheirRepo` repository:
-
-    ```python
-    import outlines
-
-    joke = outlines.Function.from_github("Someone/TheirRepo/joke")
-    response = joke("baseball")
-    ```
-    It make it easier for the community to collaborate on the infinite number of use cases enabled by these models!
-
-
-## Going further
-
-If you need more inspiration you can take a look at the [cookbook](cookbook/index.md) or watch [Remi Louf's AI Engineer World’s Fair Presentation on Outlines](https://www.youtube.com/live/R0X7mPagRiE?t=775s). If you have any question, or requests for documentation please reach out to us on [GitHub](https://github.com/dottxt-ai/outlines/discussions), [Twitter](https://twitter.com/remilouf) or [Discord](https://discord.gg/UppQmhEpe8).
-
-
-[pydantic]: https://docs.pydantic.dev/latest
-[jsonschema]: https://json-schema.org/
-[fastapi]: https://fastapi.tiangolo.com/
-[cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
-[ebnf]: https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form
-[requests]: https://requests.readthedocs.io/en/latest/
-[vllm]: https://docs.vllm.ai/en/latest/index.html
diff --git a/docs_old/reference/applications.md b/docs_old/reference/applications.md
deleted file mode 100644
index 93124daa4..000000000
--- a/docs_old/reference/applications.md
+++ /dev/null
@@ -1,40 +0,0 @@
-# Applications
-
-Outlines offers a convenient way of encapsulating a prompt template and an output type in a single object called an `Application`. After instantiating an `Application`, it can be called to generate a response by providing a model and optional arguments used to create the prompt from the template.
-
-## Create an Application
-
-To create an Application, you need to provide 2 arguments:
-
-- A template: either an instance of `outlines.templates.Template` or a callable that takes arguments and returns a prompt
-- An output type: a type instance from `outlines.types` that is used to define the structure of the output
-
-```python
-from pydantic import BaseModel
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from outlines import models, Application
-from outlines.types import JsonType
-from outlines.templates import Template
-
-class OutputModel(BaseModel):
-    result: int
-
-model = models.from_transformers(
-    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-)
-
-template_string = "What is 2 times {{ num }}?"
-template = Template.from_string(template_string)
-
-application = Application(template, JsonType(OutputModel))
-```
-
-## Call an Application
-
-Once the Application is instantiated, it can be called. The first argument is the model to use to generate the response. The remaining arguments are passed to the template to create the prompt.
-
-```python
-result = application(model, num=3)
-print(result) # Expected output: { "result" : 6 }
-```
diff --git a/docs_old/reference/chat_templating.md b/docs_old/reference/chat_templating.md
deleted file mode 100644
index e3f33fbfd..000000000
--- a/docs_old/reference/chat_templating.md
+++ /dev/null
@@ -1,38 +0,0 @@
-# Chat templating
-
-Instruction-tuned language models use "special tokens" to indicate different parts of text, such as the system prompt, the user prompt, any images, and the assistant's response. A [chat template](https://huggingface.co/docs/transformers/main/en/chat_templating) is how different types of input are composited together into a single, machine-readable string.
-
-Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself -- if you do not apply chat templating on instruction-tuned models, you will often get nonsensical output from the model.
-
-Chat template tokens are not needed for base models.
-
-You can find the chat template tokens in the model's HuggingFace repo or documentation. As an example, the `SmolLM2-360M-Instruct` special tokens can be found [here](https://huggingface.co/HuggingFaceTB/SmolLM2-360M-Instruct/blob/main/special_tokens_map.json).
-
-However, it can be slow to manually look up a model's special tokens, and special tokens vary by models. If you change the model, your prompts may break if you have hard-coded special tokens.
-
-If you need a convenient tool to apply chat templating for you, you should use the `tokenizer` from the `transformers` library:
-
-```python
-from transformers import AutoTokenizer
-
-tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/SmolLM2-360M-Instruct")
-prompt = tokenizer.apply_chat_template(
-    [
-        {"role": "system", "content": "You extract information from text."},
-        {"role": "user", "content": "What food does the following text describe?"},
-    ],
-    tokenize=False,
-    add_bos=True,
-    add_generation_prompt=True,
-)
-```
-
-yields
-
-```
-<|im_start|>system
-You extract information from text.<|im_end|>
-<|im_start|>user
-What food does the following text describe?<|im_end|>
-<|im_start|>assistant
-```
diff --git a/docs_old/reference/generation/creating_grammars.md b/docs_old/reference/generation/creating_grammars.md
deleted file mode 100644
index 78e41282a..000000000
--- a/docs_old/reference/generation/creating_grammars.md
+++ /dev/null
@@ -1,99 +0,0 @@
-# Overview
-
-Outlines allows the use of [Lark](https://github.com/lark-parser/lark) grammars to guide generation. These grammars are used to construct parsers that filter out incompatible tokens during the generation process The result is a generation that adheres to the grammar's production rules.
-
-# Primer on Creating Grammars
-
-To create grammars for Outlines, a solid understanding of Lark grammars is necessary. Here's how you can get started:
-
-- Read Lark's grammars documentations [here](https://lark-parser.readthedocs.io/en/latest/grammar.html).
-- Review Outlines' existing grammars [here](/outlines/grammars).
-
-
-# Compatibility With Outlines
-
-It's important to note that not all Lark grammars work with Outlines. Changes may be necessary to ensure compatability.
-
-### LALR(1) Parser
-
-Outlines utilizes Larks LALR(1) parser, meaning the grammar must be unambiguous at least up to the next token (one token lookahead). Read Lark's official LALR(1) parser documentation [here](https://lark-parser.readthedocs.io/en/stable/parsers.html#lalr-1).
-
-If your grammar is ambiguous, you will recieve the following error at runtime:
-
-```
-GrammarError: Reduce/Reduce collision in Terminal('B') between the following rules:
-```
-
-### Regex Terminal Restrictions
-
-Outlines converts terminals to finite state machines using the [Interegular](https://github.com/MegaIng/interegular/) library. Not all regular expressions work with Interegular, mitigation is described in the subsections which follow.
-
-
-#### Avoid Lookarounds
-
-Examples of removing lookaround while maintaining the same functionality
-
-##### Example: Escaped String
-
-From Outlines' modified `ESCAPED_STRING` in [common.lark](/outlines/grammars/common.lark).
-
-Before:
-```
-_STRING_INNER: /.*?/
-_STRING_ESC_INNER: _STRING_INNER /(?<!\\)(\\\\)*?/
-
-ESCAPED_STRING : "\"" _STRING_ESC_INNER "\""
-```
-
-After:
-```
-_NON_CONTROL_CHAR: /([^"\\\x00-\x1F\x7F-\x9F])/
-_ESCAPED_CHAR: /\\/ (_NON_CONTROL_CHAR | /\\/ | /"/)
-ESCAPED_STRING_INNER: _NON_CONTROL_CHAR | _ESCAPED_CHAR
-ESCAPED_STRING: /"/ ESCAPED_STRING_INNER* /"/
-```
-
-#### Avoid Backreferences
-
-Backreferences, for example `([ab]^*)\1`, cannot be simulated by a finite state machine, and will result in an error if used.
-
-# Creating a Valid Grammar
-
-You can use Outlines' test suite to verify your grammar.
-
-### 1) Create Your Grammar
-
-Create your grammar file named `your_new_grammar.lark`, adhering to the guidelines provided above. Add it to `outlines/grammars/` (ensure attribution is included and license is compatible).
-
-Update `outlines/grammars.py` with a line including your grammar.
-
-### 2) Test Your Grammar
-
-Test grammar for false negatives, ensure sample grammars can be generated:
-- Add valid example outputs which are compliant with the grammar to `tests/benchmark/cfg_samples/your_new_grammar/`
-- Run the tests for your grammar via `pytest -s tests/fsm/test_cfg_guide.py::test_cfg_grammar_sample -k "your_new_grammar"`
-
-Test grammar for false positives, ensure invalid outputs aren't generated.
-
-Currently there isn't a builtin false positive testing utility. It is recommended you smoke test via
-```
-from outlines import models, generate, grammars
-model = models.transformers("mistralai/Mistral-7B-v0.1")
-generator = generate.cfg(model, grammars.your_new_grammar)
-result = generator(<your prompt to generate output for your grammar>)
-print(result)
-```
-
-# Converting
-There are a few tools available for converting from other grammars to lark. These tools serve as a starting point. However, you will typically need to make additional adjustments to ensure full compatibility and proper functioning within Outlines.
-
-Tools:
-- Larks built in "Nearley-to-Lark" converter https://lark-parser.readthedocs.io/en/latest/tools.html
-- Convert ANTLR4 to Lark (Note, most antlr4 grammars are not LALR(1) compatible, so will require additional tweaking) https://github.com/kaby76/Domemtech.Trash/blob/main/src/trconvert/readme.md
-- Extract EBNF from Yacc files https://www.bottlecaps.de/rr/ui
-
-Reference Grammars:
-- Github Lark Grammars https://github.com/search?q=path%3A*.lark&type=code
-- Github Nearley Grammars https://github.com/search?q=path%3A*.ne+%22-%3E%22&type=code
-- Antlr4 grammars https://github.com/antlr/grammars-v4/
-- Grammar zoo https://slebok.github.io/zoo/index.html#html
diff --git a/docs_old/reference/generation/custom_fsm_ops.md b/docs_old/reference/generation/custom_fsm_ops.md
deleted file mode 100644
index 0dc7713ed..000000000
--- a/docs_old/reference/generation/custom_fsm_ops.md
+++ /dev/null
@@ -1,41 +0,0 @@
-# Custom FSM Operations
-
-Outlines is fast because it compiles regular expressions into an index ahead of inference. To do so we use the equivalence between regular expressions and Finite State Machines (FSMs), and the library [interegular](https://github.com/MegaIng/interegular) to perform the translation.
-
-Alternatively, one can pass a FSM built using `integular` directly to structure the generation.
-
-## Example
-
-### Using the `difference` operation
-
-In the following example we build a fsm which recognizes only the strings valid to the first regular expression but not the second. In particular, it will prevent the words "pink" and "elephant" from being generated:
-
-```python
-import interegular
-from outlines import from_transformers, Generator
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-list_of_strings_pattern = """\["[^"\s]*"(?:,"[^"\s]*")*\]"""
-pink_elephant_pattern = """.*(pink|elephant).*"""
-
-list_of_strings_fsm = interegular.parse_pattern(list_of_strings_pattern).to_fsm()
-pink_elephant_fsm = interegular.parse_pattern(pink_elephant_pattern).to_fsm()
-
-difference_fsm = list_of_strings_fsm - pink_elephant_fsm
-
-difference_fsm_fsm.accepts('["a","pink","elephant"]')
-# False
-difference_fsm_fsm.accepts('["a","blue","donkey"]')
-# True
-
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
-)
-
-generator = Generator(model, difference_fsm)
-response = generator("Don't talk about pink elephants")
-```
-
-To see the other operations available, consult [interegular's documentation](https://github.com/MegaIng/interegular/blob/master/interegular/fsm.py).
diff --git a/docs_old/reference/generation/generation.md b/docs_old/reference/generation/generation.md
deleted file mode 100644
index ad030fffd..000000000
--- a/docs_old/reference/generation/generation.md
+++ /dev/null
@@ -1,195 +0,0 @@
----
-title: Generation
----
-
-# Generation
-
-Once an [Outlines model](../models/models.md) is constructed you can use a `Generator` to generate text. The second argument to the `Generator` constructor is the output type. Standard LLM generation is possible by not passing any output type. Otherwise, you can provide a variety of different output types to benefit from structured generation. (For a detailed technical explanation of how structured generation works, you may review the [Structured Generation Explanation](./structured_generation_explanation.md) page)
-
-Before generating text, you must construct an `outlines.model`. Example:
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-import outlines
-
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
-)
-```
-
-### Text generator
-
-```python
-generator = outlines.Generator(model)
-
-result = generator("Question: What's 2+2? Answer:", max_tokens=100)
-print(result)
-# The answer is 4
-```
-
-### [Multi-label classification]
-
-Outlines allows you to do multi-label classification by guiding the model so it can only output either of the specified choices:
-
-```python
-from typing import Literal
-import outlines
-
-generator = outlines.Generator(model, Literal["Blue", "Red", "Yellow"])
-
-color = generator("What is the closest color to Indigo? ")
-print(color)
-# Blue
-```
-
-### [JSON-structured generation](./json.md)
-
-Outlines can guide models so that they output valid JSON **100%** of the time. You can either specify the structure using [Pydantic][pydantic]{:target="_blank"} or a string that contains a [JSON Schema][jsonschema]{:target="_blank"}:
-
-=== "Pydantic"
-
-    ```python
-    from enum import Enum
-    from pydantic import BaseModel, constr, conint
-
-    import outlines
-
-    class Armor(str, Enum):
-        leather = "leather"
-        chainmail = "chainmail"
-        plate = "plate"
-
-
-    class Character(BaseModel):
-        name: constr(max_length=10)
-        age: conint(gt=18, lt=99)
-        armor: Armor
-        strength: conint(gt=1, lt=100)
-
-    generator = outlines.Generator(model, Character)
-
-    character = generator(
-        "Generate a new character for my awesome game: "
-        + "name, age (between 1 and 99), armor and strength. "
-        )
-    print(character)
-    # name='Orla' age=21 armor=<Armor.plate: 'plate'> strength=8
-    ```
-
-=== "JSON Schema"
-
-    ```python
-    import outlines
-    from outlines.types import JsonSchema
-
-    schema = """{
-        "$defs": {
-            "Armor": {
-                "enum": ["leather", "chainmail", "plate"],
-                "title": "Armor",
-                "type": "string"
-            }
-        },
-        "properties": {
-            "name": {"maxLength": 10, "title": "Name", "type": "string"},
-            "age": {"title": "Age", "type": "integer"},
-            "armor": {"$ref": "#/$defs/Armor"},
-            "strength": {"title": "Strength", "type": "integer"}\
-        },
-        "required": ["name", "age", "armor", "strength"],
-        "title": "Character",
-        "type": "object"
-    }"""
-
-    generator = outlines.Generator(model, JsonSchema(schema))
-    character = generator(
-        "Generate a new character for my awesome game: "
-        + "name, age (between 1 and 99), armor and strength. "
-        )
-    print(character)
-    # {'name': 'Yuki', 'age': 24, 'armor': 'plate', 'strength': 3}
-    ```
-
-!!! Note
-
-    We advise you to constrain the length of the strings fields when first testing your schema, especially with small models.
-
-### [Grammar-structured generation](./cfg.md)
-
-Outlines also allows to generate text that is valid to any [context-free grammar][cfg]{:target="_blank"} (CFG) in the [EBNF format][ebnf]{:target="_blank"}. Grammars can be intimidating, but they are a very powerful tool! Indeed, they determine the syntax of every programming language, valid chess moves, molecule structure, can help with procedural graphics generation, etc.
-
-Here we show a simple example of a grammar that defines arithmetic operations:
-
-```python
-import outlines
-from outlines.types import CFG
-
-arithmetic_grammar = """
-    ?start: sum
-
-    ?sum: product
-        | sum "+" product   -> add
-        | sum "-" product   -> sub
-
-    ?product: atom
-        | product "*" atom  -> mul
-        | product "/" atom  -> div
-
-    ?atom: NUMBER           -> number
-         | "-" atom         -> neg
-         | "(" sum ")"
-
-    %import common.NUMBER
-    %import common.WS_INLINE
-
-    %ignore WS_INLINE
-"""
-
-generator = outlines.Generator(model, CFG(arithmetic_grammar, max_tokens=100))
-
-result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
-print(result)
-# 5+5+5+5+5
-```
-
-
-EBNF grammars can be cumbersome to write. This is why Outlines provides grammar definitions in the `outlines.grammars.` module
-
-```python
-import outlines
-from outlines import grammars
-from outlines.types import CFG
-
-generator = outlines.Generator(model, CFG(grammars.arithmetic, max_tokens=100))
-
-result = generator("Question: How can you write 5*5 using addition?\nAnswer:")
-print(result)
-# 5+5+5+5+5
-```
-
-The available grammars are listed [here](https://github.com/dottxt-ai/outlines/tree/main/outlines/grammars).
-
-
-### [Regex-structured generation](./regex.md)
-
-Slightly simpler, but no less useful, Outlines can generate text that is in the language of a [regular expression](https://www.regular-expressions.info/tutorial.html). For instance to force the model to generate IP addresses:
-
-```python
-import outlines
-from outlines.types import Regex
-
-regex_str = r"((25[0-5]|2[0-4]\d|[01]?\d\d?)\.){3}(25[0-5]|2[0-4]\d|[01]?\d\d?)"
-generator = outlines.Generator(model, Regex(regex_str))
-
-result = generator("What is the IP address of localhost?\nIP: ")
-print(result)
-# 127.0.0.100
-```
-
-[jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
-[pydantic]: https://docs.pydantic.dev/latest
-[cfg]: https://en.wikipedia.org/wiki/Context-free_grammar
-[ebnf]: https://en.wikipedia.org/wiki/Extended_Backus%E2%80%93Naur_form
diff --git a/docs_old/reference/generation/structured_generation_explanation.md b/docs_old/reference/generation/structured_generation_explanation.md
deleted file mode 100644
index aa27a7a85..000000000
--- a/docs_old/reference/generation/structured_generation_explanation.md
+++ /dev/null
@@ -1,54 +0,0 @@
-# How does Outlines work?
-
-
-Language models generate text token by token, using the previous token sequence as input and sampled logits as output. This document explains the structured generation process, where only legal tokens are considered for the next step based on a predefined automata, e.g. a regex-defined [finite-state machine](https://en.wikipedia.org/wiki/Finite-state_machine) (FSM) or [Lark](https://lark-parser.readthedocs.io/en/stable/) grammar.`
-
-
-## Worked Example
-
-Let's consider a worked example with a pattern for whole and decimal numbers:
-
-`^\d*(\.\d+)?$`.
-
-### Creating Automata
-
-The pattern is first converted into an automata. Below is a brief explanation of the automata conversion and its representation.
-
-**Automata Diagram:**
-
-```mermaid
-graph LR
-    node0("1-9") --> node1("1-9")
-    node1 --> node1
-    node1 --> nodeEND{{END}}
-    node1 --> nodePeriod(".")
-    nodePeriod --> node2("1-9")
-    node2 --> node2
-    node2 --> nodeEND{{END}}
-```
-
-### Generating a Token
-
-Let's assume that we're in the middle of generation, and so far "748" has been generated. Here is the automata with the current state highlighted in green, with the legal next characters being another number (1-9), a dot (.), or end of sequence.
-
-```mermaid
-graph LR
-    node0("1-9") --> node1("1-9")
-    node1 --> node1
-    node1 --> nodeEND{{END}}
-    node1 --> nodePeriod(".")
-    nodePeriod --> node2("1-9")
-    node2 --> node2
-    node2 --> nodeEND{{END}}
-
-    style node1 fill:#090
-```
-
-Generating a token requires the following steps:
-
-- Feed the previous input sequence ("748") into the language model.
-- Language model runs a forward pass and produces token logits.
-- Outlines logits processor sets the probability of illegal tokens to 0%.
-- A token is sampled from the set of legal tokens.
-
-![Generation and Logits Processing Flow Chart](../../assets/images/logits_processing_diagram.svg)
diff --git a/docs_old/reference/generation/types.md b/docs_old/reference/generation/types.md
deleted file mode 100644
index 037db313b..000000000
--- a/docs_old/reference/generation/types.md
+++ /dev/null
@@ -1,69 +0,0 @@
-# Custom types
-
-Outlines provides custom Pydantic types so you can focus on your use case rather than on writing regular expressions:
-
-| Category | Type | Import | Description |
-|:--------:|:----:|:-------|:------------|
-| ISBN | 10 & 13 | `outlines.types.ISBN` | There is no guarantee that the [check digit][wiki-isbn] will be correct |
-| Airport | IATA | `outlines.types.airports.IATA` | Valid [airport IATA codes][wiki-airport-iata] |
-| Country | alpha-2 code | `outlines.types.airports.Alpha2` | Valid [country alpha-2 codes][wiki-country-alpha-2] |
-|  | alpha-3 code | `outlines.types.countries.Alpha3` | Valid [country alpha-3 codes][wiki-country-alpha-3] |
-|  | numeric code | `outlines.types.countries.Numeric` | Valid [country numeric codes][wiki-country-numeric] |
-|  | name | `outlines.types.countries.Name` | Valid country names |
-|  | flag | `outlines.types.countries.Flag` | Valid flag emojis |
-| | email | `outlines.types.Email` | Valid email address |
-
-Some types require localization. We currently only support US types, but please don't hesitate to create localized versions of the different types and open a Pull Request. Localized types are specified using `types.locale` in the following way:
-
-```python
-from outlines import types
-
-types.locale("us").ZipCode
-types.locale("us").PhoneNumber
-```
-
-Here are the localized types that are currently available:
-
-| Category | Locale | Import | Description |
-|:--------:|:----:|:-------|:------------|
-| Zip code | US | `ZipCode` | Generate US Zip(+4) codes |
-| Phone number  | US | `PhoneNumber` | Generate valid US phone numbers |
-
-
-You can use these types in Pydantic schemas for JSON-structured generation:
-
-```python
-from pydantic import BaseModel
-
-from outlines import models, types, Generator
-
-# Specify the locale for types
-locale = types.locale("us")
-
-class Client(BaseModel):
-    name: str
-    phone_number: locale.PhoneNumber
-    zip_code: locale.ZipCode
-
-
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
-)
-generator = Generator(model, Client)
-result = generator(
-    "Create a client profile with the fields name, phone_number and zip_code"
-)
-print(result)
-# name='Tommy' phone_number='129-896-5501' zip_code='50766'
-```
-
-We plan on adding many more custom types. If you have found yourself writing regular expressions to generate fields of a given type, or if you could benefit from more specific types don't hesite to [submit a PR](https://github.com/dottxt-ai/outlines/pulls) or [open an issue](https://github.com/dottxt-ai/outlines/issues/new/choose).
-
-
-[wiki-isbn]: https://en.wikipedia.org/wiki/ISBN#Check_digits
-[wiki-airport-iata]: https://en.wikipedia.org/wiki/IATA_airport_code
-[wiki-country-alpha-2]: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2
-[wiki-country-alpha-3]: https://en.wikipedia.org/wiki/ISO_3166-1_alpha-3
-[wiki-country-numeric]: https://en.wikipedia.org/wiki/ISO_3166-1_numeric
diff --git a/docs_old/reference/index.md b/docs_old/reference/index.md
deleted file mode 100644
index 0cd42ba88..000000000
--- a/docs_old/reference/index.md
+++ /dev/null
@@ -1,15 +0,0 @@
-# Reference
-
-## Structured generation
-
-While LLM capabilities are increasingly impressive, we can make their output more reliable by steering the generation. Outlines thus offers mechanisms to specify high level constraints on text completions by generative language models.
-
-Stopping sequence
-By default, language models stop generating tokens after and <EOS> token was generated, or after a set maximum number of tokens. Their output can be verbose, and for practical purposes it is often necessary to stop the generation after a given sequence has been found instead. You can use the stop_at keyword argument when calling the model with a prompt:
-
-```python
-import outlines.models as models
-
-complete = models.openai("gpt-4o-mini")
-expert = complete("Name an expert in quantum gravity.", stop_at=["\n", "."])
-```
diff --git a/docs_old/reference/models/anthropic.md b/docs_old/reference/models/anthropic.md
deleted file mode 100644
index d3db67ddb..000000000
--- a/docs_old/reference/models/anthropic.md
+++ /dev/null
@@ -1,44 +0,0 @@
-# Anthropic
-
-!!! Installation
-
-    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines: `pip install anthropic`.
-
-## Anthropic models
-
-Outlines supports models available via the Anthropic API, e.g. Claude 3.5 Haiku or Claude 3.5 Sonner. You can initialize the model by passing the model name to `outlines.models.Anthropic`:
-
-```python
-from anthropic import Anthropic
-import outlines
-
-model = outlines.from_anthropic(Anthropic("claude-3-5-haiku-latest"))
-model = outlines.from_anthropic(Anthropic("claude-3-5-sonnet-latest"))
-```
-
-Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-claude/models) for an up-to-date list of available models.
-
-## Text generation
-
-To generate text using an Anthropic model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
-
-```python
-from anthropic import Anthropic
-import outlines
-
-model = outlines.from_anthropic(Anthropic("claude-3-5-haiku-latest"))
-generator = Generator(model)
-result = generator("Prompt", max_tokens=1024)
-
-# Call the model directly
-result = model("Prompt", max_tokens=1024)
-```
-
-
-!!! Warning
-
-    You must set a value for `max_tokens` with Anthropic models.
-
-See the [Anthropic SDK documentation](https://github.com/anthropics/anthropic-sdk-python/blob/main/src/anthropic/resources/messages.py) for the list of available arguments.
-
-The Anthropic API currently does not support structured generation.
diff --git a/docs_old/reference/models/dottxt.md b/docs_old/reference/models/dottxt.md
deleted file mode 100644
index 2ac07de6c..000000000
--- a/docs_old/reference/models/dottxt.md
+++ /dev/null
@@ -1,51 +0,0 @@
-# Dottxt
-
-!!! Installation
-
-    To be able to use Dottxt in Outlines, you must install the `dottxt` python sdk.
-
-    ```bash
-    pip install dottxt
-    ```
-
-    You also need to have a Dottxt API key. This API key must either be set as an environment variable called `DOTTXT_API_KEY` or be provided to the `outlines.models.Dottxt` class when instantiating it.
-
-## Generate text
-
-Dottxt only supports constrained generation with the `Json` output type. The input of the generation must be a string. Batch generation is not supported.
-Thus, you must always provide an output type.
-
-You can either create a `Generator` object and call it afterward:
-
-```python
-from outlines.models import Dottxt
-from outlines import Generator
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-model = Dottxt()
-generator = Generator(model, Character)
-result = generator("Create a character")
-```
-
-or call the model directly with the output type:
-
-```python
-from outlines.models import Dottxt
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-model = Dottxt()
-result = model("Create a character", Character)
-```
-
-In any case, compilation for a given output type happens only once (the first time it is used to generate text).
-
-## Optional parameters
-
-You can provide the same optional parameters you would pass to the `dottxt` sdk's client both during the initialization of the `Dottxt` class and when generating text.
-Consult the [dottxt python sdk Github repository](https://github.com/dottxt-ai/dottxt-python) for the full list of parameters.
diff --git a/docs_old/reference/models/gemini.md b/docs_old/reference/models/gemini.md
deleted file mode 100644
index 6ce530fbb..000000000
--- a/docs_old/reference/models/gemini.md
+++ /dev/null
@@ -1,96 +0,0 @@
-# Gemini
-
-!!! Installation
-
-    You need to install the `google-generativeai` library to be able to use the Gemini API in Outlines: `pip install google-generativeai`
-
-
-## Gemini models
-
-Outlines supports models available via the Gemini API, e.g. Gemini 1.5. You can initialize the model by passing the model name to `outlines.models.Gemini`:
-
-```python
-import outlines
-import google.generativeai as genai
-
-model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
-model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-pro"))
-```
-
-Check the [Gemini documentation](https://ai.google.dev/gemini-api/docs/models/gemini) for an up-to-date list of available models.
-
-## Text generation
-
-To generate text using a Gemini model you need to build a `Generator` object, possibly with the desired output type. You can then call the model by calling the `Generator`. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
-
-```python
-import outlines
-import google.generativeai as genai
-
-model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
-generator = outlines.Generator(model)
-result = generator("Prompt", max_tokens=1024)
-
-# Call the model directly
-result = model("Prompt", max_tokens=1024)
-```
-
-### Structured generation
-
-Gemini provides support for structured outputs.
-
-#### Json Schema
-
-Outlines provides support for JSON Schema-based structured generation with the Gemini models:
-
-```python
-from collections import TypedDict
-
-import google.generativeai as genai
-
-import outlines
-from outlines import Generator
-
-
-model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
-
-class Person(TypedDict):
-    first_name: str
-    last_name: str
-    age: int
-
-generator = Generator(model, Person)
-generator("current indian prime minister on january 1st 2023")
-# Person(first_name='Narendra', last_name='Modi', age=72)
-```
-
-Because of the current limitations of the Gemini SDK only The following objects can be used to define the structure of the Json object:
-- A Pydantic model
-- A TypedDict
-
-#### Multiple choices
-
-Outlines provides support for multiple-choices structured generation. Enums and lists of choices are supported:
-
-```python
-from enum import Enum
-
-import google.generativeai as genai
-
-from outlines import Generator
-
-model = outlines.from_gemini(genai.GenerativeModel("gemini-1-5-flash"))
-
-class Foo(Enum):
-    foo = "Foo"
-    fizz = "Fizz"
-    fuzz = "Fuzz"
-
-generator = Generator(model, Foo)
-generator("current indian prime minister on january 1st 2023")
-# Person(first_name='Narendra', last_name='Modi', age=72)
-```
-
-The following objects can be used to define the choices:
-- An Enum object
-- A Python list
diff --git a/docs_old/reference/models/llamacpp.md b/docs_old/reference/models/llamacpp.md
deleted file mode 100644
index 313f58eb9..000000000
--- a/docs_old/reference/models/llamacpp.md
+++ /dev/null
@@ -1,88 +0,0 @@
-# Llama.cpp
-
-Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python library][llamacpp]. Llamacpp allows to run quantized models on machines with limited compute.
-
-!!! Note "Documentation"
-
-    To be able to use llama.cpp in Outlines, you must install the `llama-cpp-python` library, `pip install llama-cpp-python`
-
-    Consult the [`llama-cpp-python` documentation](https://llama-cpp-python.readthedocs.io/en/latest) for detailed informations about how to initialize model and the available options.
-
-
-## Load the model
-
-You can use `outlines.from_llamacpp` to load a `llama-cpp-python` model. Assuming [Phi2's weights](https://huggingface.co/TheBloke/phi-2-GGUF) are in the current directory:
-
-```python
-from llama_cpp import Llama
-import outlines
-
-llm = outlines.from_llamacpp(Llama("./phi-2.Q4_K_M.gguf"))
-```
-
-You can initialize the model by passing the name of the repository on the HuggingFace Hub, and the filenames (or glob pattern):
-
-
-```python
-from llama_cpp import Llama
-import outlines
-
-model = outlines.from_llamacpp(Llama.from_pretrained("TheBloke/phi-2-GGUF", "phi-2.Q4_K_M.gguf"))
-```
-
-This will download the model files to the hub cache folder and load the weights in memory. See the [llama-cpp-python documentation](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__) for the full list of parameters you can pass to the `Llama` class.
-
-
-### Load the model on GPU
-
-!!! Note
-
-    [Make sure](#cuda) that you installed `llama-cpp-python` with GPU support.
-
- To load the model on GPU, pass `n_gpu_layers=-1`:
-
-```python
-from llama_cpp import Llama
-import outlines
-
-model = outlines.from_llamacpp(Llama(
-    "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf",
-    n_gpu_layers=-1,  # to use GPU acceleration
-))
-```
-
-
-## Generate text
-
-
-To generate text you must first create a `Generator` object by passing the model instance and, possibley, the expected output type:
-
-```python
-import outlines
-
-
-model = outlines.from_llamacpp(Llama(
-    "TheBloke/phi-2-GGUF",
-    "phi-2.Q4_K_M.gguf",
-    n_gpu_layers=-1,  # to use GPU acceleration
-))
-generator = outlines.Generator(model)
-```
-
-You can pass to the generator the same keyword arguments you would pass in `llama-cpp-python`:
-
-```python
-answer = generator("A prompt", presence_penalty=0.8)
-```
-
-You can also stream the tokens:
-
-```python
-tokens = generator.stream("A prompt")
-```
-
-
-[llamacpp]: https://github.com/abetlen/llama-cpp-python
-[llama-cpp-python-call]: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__call__
-[llama-cpp-python-install]: https://github.com/abetlen/llama-cpp-python/tree/08b16afe11e7b42adec2fed0a781123383476045?tab=readme-ov-file#supported-backends
diff --git a/docs_old/reference/models/mlxlm.md b/docs_old/reference/models/mlxlm.md
deleted file mode 100644
index 33953c11f..000000000
--- a/docs_old/reference/models/mlxlm.md
+++ /dev/null
@@ -1,70 +0,0 @@
-# mlx-lm
-
-Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx-examples/tree/main/llms), allowing models to be run quickly on Apple Silicon via the [mlx](https://ml-explore.github.io/mlx/build/html/index.html) library.
-
-!!! Note "Installation"
-
-    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration: `pip install mlx mlx-lm`.
-
-    Consult the [`mlx-lm` documentation](https://github.com/ml-explore/mlx-examples/tree/main/llms) for detailed informations about how to initialize OpenAI clients and the available options.
-
-
-## Load the model
-
-You can initialize the model by passing the name of the repository on the HuggingFace Hub. The official repository for mlx-lm supported models is [mlx-community](https://huggingface.co/mlx-community).
-
-```python
-from mlx_lm import load
-import outlines
-
-model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
-```
-
-This will download the model files to the hub cache folder and load the weights in memory.
-
-## Generate text
-
-You may generate text using the parameters described in the [text generation documentation](../text.md).
-
-With the loaded model, you can generate text or perform structured generation, e.g.
-
-```python
-from mlx_lm import load
-import outlines
-
-model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
-answer = model("A prompt", temperature=2.0)
-```
-
-## Streaming
-
-You may creating a streaming iterable with minimal changes
-
-```python
-from mlx_lm import load
-import outlines
-
-model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
-generator = outlines.Generator(model)
-
-for token_str in generator.text("A prompt", temperature=2.0):
-    print(token_str)
-```
-
-## Structured
-
-You may perform structured generation with mlxlm to guarantee your output will match a regex pattern, json schema, or lark grammar.
-
-Example: Phone number generation with pattern `"\\+?[1-9][0-9]{7,14}"`:
-
-```python
-from mlx_lm import load
-import outlines
-
-model = outlines.from_mlxlm(*load("mlx-community/Meta-Llama-3.1-8B-Instruct-8bit"))
-
-phone_number_pattern = "\\+?[1-9][0-9]{7,14}"
-model_output = model("What's Jennys Number?\n", outlines.Regex(phone_number_pattern))
-print(model_output)
-# '8675309'
-```
diff --git a/docs_old/reference/models/models.md b/docs_old/reference/models/models.md
deleted file mode 100644
index d9d581e96..000000000
--- a/docs_old/reference/models/models.md
+++ /dev/null
@@ -1,50 +0,0 @@
----
-title: Models
----
-
-# Models
-
-This section provides detailed information about using structured generation with a number of inference engines.
-
-## Supported providers
-
-### Open Source models
-
-Open source models offer more flexibility for structured generation as we have control over the sampling loop:
-
-- [transformers](transformers.md) - Run open-source models locally.
-- [llama-cpp-python](llamacpp.md) - Python bindings for llama.cpp.
-- [mlx-lm](mlxlm.md) - Run open-source models on Metal hardware.
-- [vllm](vlm.md) - Run open-source models on the vLLM engine.
-
-Ollama only supports Json Schema:
-
-- [ollama](ollama.md) - Python client library for Ollama.
-
-### Cloud AI providers
-
-OpenAI has recently integrated [structured outputs][structured-outputs] in its API, and only JSON Schema-based structured generation is available. Google's Gemini API supports both Json Schema and multiple choices:
-
-- [OpenAI](openai.md) - GPT-4o, o1, o3-mini and other OpenAI models.
-- [Azure OpenAI](openai.md) - Microsoft's Azure-hosted OpenAI models.
-- [Gemini](gemini.md) - Run Google's Gemini model.
-
-
-## Structure generation coverage
-
-Integrations differ in their coverage of structured generation. Here is a summary:
-
-
-|                   | [Transformers](transformers.md) | [vLLM](vllm.md) | [llama.cpp](llamacpp.md) | [MLXLM](mlxlm.md) | [OpenAI](openai.md) | [Gemini](gemini.md) |
-|-------------------|--------------|------|-----------|-------|---------|-------|
-| **Supported HW**  |              |      |           |       |         |       |
-| CUDA              | ✅           | ✅  | ✅        | ❌    | N/A     | N/A   |
-| Apple Silicon     | ✅           | ❌  | ✅        | ✅    | N/A     | N/A   |
-| x86 / AMD64       | ✅           | ❌  | ✅        | ❌    | N/A     | N/A   |
-| **Structure**     |               |     |           |       |         |       |
-| JSON Schema       | ✅           | ✅  | ✅        | ✅    | ✅      | ✅   |
-| Regex             | ✅           | ✅  | ✅        | ✅    | ❌      | ❌   |
-| Grammar           | ✅           | ✅  | ✅        | ✅    | ❌      | ❌   |
-
-
-[structured-outputs]: https://platform.openai.com/docs/guides/structured-outputs
diff --git a/docs_old/reference/models/ollama.md b/docs_old/reference/models/ollama.md
deleted file mode 100644
index f6805f55c..000000000
--- a/docs_old/reference/models/ollama.md
+++ /dev/null
@@ -1,74 +0,0 @@
-# Ollama
-
-!!! Installation
-
-    To be able to use Ollama in Outlines, you must install both Ollama and the `ollama` python sdk.
-
-    - To download Ollama: https://ollama.com/download
-    - To install the ollama python sdk: `pip install ollama`
-
-    Consult the [`ollama` documentation](https://github.com/ollama/ollama-python) for detailed informations about how to initialize models and the available options.
-
-## Ollama models
-
-You must provide a model name when instantiating the `outlines.models.Ollama` class. This model must be available on your system.
-```python
-from outlines.models import Ollama
-
-model = Ollama("tinyllama")
-```
-
-To download a new model from the Ollama model hub, you can use the following command (it will return an `Ollama` instance):
-```python
-from outlines.models import Ollama
-
-Ollama.from_pretrained("llama3.1:8b")
-```
-
-You can find the list of available models on the [Ollama library](https://ollama.com/library).
-
-## Generate text
-
-As with other models, you can either first create a `Generator` object and then call it
-```python
-from outlines.models import Ollama
-from outlines import Generator
-
-model = Ollama("tinyllama")
-generator = Generator(model)
-answer = generator("Write a sentence about a cat.")
-```
-or directly call the model
-```python
-from outlines.models import Ollama
-
-model = Ollama("tinyllama")
-answer = model("Write a sentence about a cat.")
-```
-
-The input of the generation must be a string. Batch generation is not supported.
-The only output type supported is `Json`.
-```python
-from outlines.models import Ollama
-from outlines.types import Json
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-model = Ollama("tinyllama")
-answer = model("Create a character.", output_type=Json(Character))
-```
-
-You can also stream the tokens:
-```python
-from outlines.models import Ollama
-
-model = Ollama("tinyllama")
-tokens = model.stream("Write a sentence about a cat.")
-```
-
-## Optional parameters
-
-You can provide the same optional parameters you would pass to the `ollama` sdk's client both during the initialization of the `Ollama` class and when generating text.
-Consult the [ollama python sdk Github repository](https://github.com/ollama/ollama-python) for the full list of parameters.
diff --git a/docs_old/reference/models/openai.md b/docs_old/reference/models/openai.md
deleted file mode 100644
index 95c03b6ac..000000000
--- a/docs_old/reference/models/openai.md
+++ /dev/null
@@ -1,110 +0,0 @@
-# OpenAI and compatible APIs
-
-!!! Note "Documentation"
-
-    To be able to use OpenAI in Outlines, you must install the OpenAI Python SDK with `pip install openai`
-
-    Consult the [OpenAI SDK documentation](https://github.com/openai/openai-python) for detailed informations about how to initialize OpenAI clients and the available options.
-
-## OpenAI models
-
-Outlines supports models available via the OpenAI Chat API, e.g. GPT-4o, ChatGPT and GPT-4. You can initialize the model by passing the model name to `outlines.models.OpenAI`:
-
-```python
-from openai import OpenAI
-import outlines
-
-# OpenAI models
-client = OpenAI()
-model = outlines.from_openai(client, "gpt-4o-mini")
-model = outlines.from_openai(client, "gpt-4o")
-
-# OpenAI models deployed on Azure
-client = AzureOpenAI(
-    api_version="2024-07-18",
-    azure_endpoint="https://example-endpoint.openai.azure.com",
-)
-model = outlines.from_openai(client, "azure-deployment-name")
-```
-
-Check the [OpenAI documentation](https://platform.openai.com/docs/models/gpt-4o) for an up-to-date list of available models, and the [OpenAI SDK's code](https://github.com/openai/openai-python/blob/54a5911f5215148a0bdeb10e2bcfb84f635a75b9/src/openai/_client.py) for an up-to-date list of the initialization parameters.
-
-
-## Advanced configuration
-
-For more advanced configuration option, such as support proxy, please consult the [OpenAI SDK's documentation](https://github.com/openai/openai-python):
-
-
-```python
-from openai import AsyncOpenAI, DefaultHttpxClient
-import outlines
-
-client = AsyncOpenAI(
-    api_key="my key",
-    base_url="http://my.test.server.example.com:8083",
-    http_client=DefaultHttpxClient(
-        proxies="http://my.test.proxy.example.com",
-        transport=httpx.HTTPTransport(local_address="0.0.0.0"),
-    ),
-)
-model = outlines.from_openai(client, "model_name")
-```
-
-## Models that follow the OpenAI standard
-
-Outlines supports models that follow the OpenAI standard. You will need to initialize the OpenAI client properly configured and pass it to `outlines.models.OpenAI`
-
-```python
-import os
-from openai import AsyncOpenAI
-import outlines
-
-client = AsyncOpenAI(
-    api_key=os.environ.get("PROVIDER_KEY"),
-    base_url="http://other.provider.server.com",
-)
-model = outlines.from_openai(client, "model_name")
-```
-
-## Calling the model
-
-You can call the model directly. The method accepts every argument that you could pass to the `client.completions.create` function, as keyword arguments:
-
-```python
-from openai import OpenAI
-import outlines
-
-model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
-result = model("Prompt", seed=10)
-```
-
-See the [OpenAI SDK documentation](https://github.com/openai/openai-python/blob/6974a981aec1814b5abba429a8ea21be9ac58538/src/openai/types/completion_create_params.py#L13) for the list of available arguments.
-
-### Structured Generation Support
-
-Outlines provides support for [Structured Outputs](https://platform.openai.com/docs/guides/structured-outputs/json-mode). Currently only JSON-Schema is supported:
-
-```python
-from openai import OpenAI
-from pydantic import BaseModel
-
-import outlines
-
-model = outlines.from_openai(OpenAI(), "gpt-4o-mini")
-
-class Person(BaseModel):
-    first_name: str
-    last_name: str
-    age: int
-
-generator = Generator(model, Person)
-generator("current indian prime minister on january 1st 2023")
-# Person(first_name='Narendra', last_name='Modi', age=72)
-```
-
-The following objects can be used to define the structure of the Json object:
-- A string that represents a Json Schema
-- A dictionary that represents a Json Schema
-- A Pydantic model
-- A TypedDict
-- A dataclass
diff --git a/docs_old/reference/models/sglang.md b/docs_old/reference/models/sglang.md
deleted file mode 100644
index da3cfa89c..000000000
--- a/docs_old/reference/models/sglang.md
+++ /dev/null
@@ -1,51 +0,0 @@
-# SgLang
-
-## Prerequisites
-
-The Outlines `SgLang` model is inteded to be used along with a separate sglang server (running either locally or remotely). Make sure you have an sglang server running before using the `SgLang` model. As the sglang client relies on the `openai` python sdk, you need to have the `openai` package installed.
-
-## Initialize the model
-
-To load the model, you can use the `from_sglang` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Based on whether the `openai` instance is synchronous or asynchronous, you will receive a `SgLang` or `AsyncSgLang` model instance.
-
-```python
-import openai
-import outlines
-
-sync_openai_client = openai.OpenAI(base_url="...")
-async_openai_client = openai.AsyncOpenAI(base_url="...")
-
-sync_model = outlines.from_sglang(sync_openai_client, "qwen/qwen2.5-0.5b-instruct")
-print(type(sync_model)) # <class 'outlines.models.sglang.SgLang'>
-
-async_model = outlines.from_sglang(async_openai_client, "qwen/qwen2.5-0.5b-instruct")
-print(type(async_model)) # <class 'outlines.models.sglang.AsyncSgLang'>
-```
-
-## Generate text
-
-To generate text, you can call the model with a prompt as argument and optionally an output type to rely on structured generation:
-
-```python
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-answer = sync_model("Create a character.", output_type=Character)
-answer = await async_model("Create a character.", output_type=Character)
-```
-
-The `SgLang` model supports also supports streaming.
-
-```python
-for chunk in sync_model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk)
-```
-
-## Optional parameters
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `openai` client. An optional parameter of particular interest is `extra_body`, which is a dictionnary containing arguments that are specific to sglang and are not part of the standard `openai` interface (see the [sglang documentation][sglang-docs] on the OpenAI-compatible server for more information).
-
-[sglang-docs]: https://docs.sglang.ai/start/install.html
-[sglang-structured-output-quickstart]: https://docs.sglang.ai/backend/structured_outputs.html#JSON
diff --git a/docs_old/reference/models/tgi.md b/docs_old/reference/models/tgi.md
deleted file mode 100644
index ed1207449..000000000
--- a/docs_old/reference/models/tgi.md
+++ /dev/null
@@ -1,50 +0,0 @@
-# TGI
-
-## Prerequisites
-
-The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed.
-
-## Initialize the model
-
-    To load the model, you can use the `from_tgi` function. The argument of the function is an instance of `InferenceClient` or `AsyncInferenceClient` from the `huggingface_hub` package. You will respectively receive a `TGI` or `AsyncTGI` model instance.
-
-```python
-import outlines
-from huggingface_hub import InferenceClient, AsyncInferenceClient
-
-sync_model = outlines.from_tgi(InferenceClient("http://localhost:8080"))
-print(type(sync_model))  # <class 'outlines.models.tgi.TGI'>
-
-async_model = outlines.from_tgi(AsyncInferenceClient("http://localhost:8080"))
-print(type(async_model))  # <class 'outlines.models.tgi.AsyncTGI'>
-```
-
-## Generate text
-
-To generate text, you can call the model with a prompt as argument and optionally an output type to rely on structured generation:
-
-```python
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-answer = sync_model("Create a character.", output_type=Character)
-answer = await async_model("Create a character.", output_type=Character)
-```
-
-The `TGI` model also supports streaming:
-
-```python
-for chunk in sync_model.stream("Write a short story about a cat.", max_new_tokens=100):
-    print(chunk)
-```
-
-## Optional parameters
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. These parameters will be passed directly to the TGI client. Common parameters include `max_new_tokens`, `stream`, `temperature`, `top_k`, `top_p`, and others as specified in the [TGI inference client documentation][tgi-inference-client-docs].
-
-
-[tgi-docs]: https://huggingface.co/docs/text-generation-inference/en/index
-[tgi-inference-client-docs]: https://huggingface.co/docs/huggingface_hub/package_reference/inference_client
-[tgi-structured-output-docs]: https://huggingface.co/docs/text-generation-inference/en/basic_tutorials/using_guidance
diff --git a/docs_old/reference/models/transformers.md b/docs_old/reference/models/transformers.md
deleted file mode 100644
index 0c5600b03..000000000
--- a/docs_old/reference/models/transformers.md
+++ /dev/null
@@ -1,68 +0,0 @@
-# Transformers
-
-!!! Note "Documentation"
-
-    To be able to use Transformers models in Outlines, you must install the `transformers` library, `pip install transformers`
-
-    Consult the [`transformers` documentation](https://huggingface.co/docs/transformers/en/index) for detailed informations about how to initialize models and the available options.
-
-## Create a `Transformers` model
-
-You can use `outlines.from_transformers` to load a `transformers` model and tokenizer:
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from outlines import models
-
-model_name = "microsoft/Phi-3-mini-4k-instruct"
-model = models.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name)
-)
-```
-
-The model name must be a valid `transformers` model name. You can find a list of all in the HuggingFace library [here](https://huggingface.co/models). We currently support `CausalLM`, `Seq2Seq`, `Mamba`, vision and audio models.
-
-Be cautious with model selection though, some models such as `t5-base` don't include certain characters (`{`) and you may get an error when trying to perform structured generation.
-
-## Use the model to generate text
-
-Once you have created a `Transformers` model, you can use it to generate text by calling the instance of the model.
-
-```python
-model("Hello, how are you?")
-```
-
-You can also first create a `Generator` and then call it.
-```python
-from outlines import Generator
-
-generator = Generator(model)
-generator("Hello, how are you?")
-```
-
-`Transformers` models typically support batching and the generation of several samples at once.
-
-For instance:
-```python
-model(["Hello, how are you?", "Respond with one word. Not more."], num_return_sequences=2, num_beams=2)
-```
-
-This would generate two sequences for each prompt, for a total of four sequences (two lists of 2 elements each in a list).
-
-## Use the model to generate structured data
-
-`Transformers` models can generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
-
-You can use most common Python types along with the Outlines DSL types `JsonSchema`, `Regex` and `CFG`.
-
-For instance:
-```python
-from outlines.types import Json
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-model("Create a character with a name.", Character)
-```
diff --git a/docs_old/reference/models/transformers_multimodal.md b/docs_old/reference/models/transformers_multimodal.md
deleted file mode 100644
index 85ed8b8b2..000000000
--- a/docs_old/reference/models/transformers_multimodal.md
+++ /dev/null
@@ -1,203 +0,0 @@
-# Transformers Multimodal
-
-Outlines allows seamless use of [multimodal models](https://huggingface.co/learn/multimodal-models-course/en/unit1/introduction).
-
-`outlines.models.TransformersMultimodal` shares interfaces with, and is based on [outlines.models.Transformers](./transformers.md).
-
-## Create a `TransformersMultimodal` model
-
-`TransformersMultimodal` models inherit from `Transformers` and accept similar initialization parameters, with the exception that you should provide a processor instead of a tokenizer.
-
-You can use `outlines.from_transformers` to load a `transformers` model and processor:
-
-```python
-from transformers import CLIPModel, CLIPProcessor
-from outlines import models
-
-model_name = "openai/clip-vit-base-patch32"
-model = models.from_transformers(
-    CLIPModel.from_pretrained(model_name),
-    CLIPProcessor.from_pretrained(model_name)
-)
-```
-
-You must provider a processor adapted to your model. We currently offer official support for vision and audio models only, but you can use this class for any type of multimodal model (for instance video) as long as you provide the appropriate processor for your transformers model.
-
-## Use the model to generate text from prompts and assets
-
-When calling the model, the prompt argument you provide must be a dictionary using the following format:
-```
-{
-    "text": Union[str, List[str]],
-    "...": Any
-}
-```
-
-The `text` key is mandatory. Other keys are optional and depend on the processor you are using.
-The format and values of those keys must correspond to what you would provide to the `__call__` method of the processor if you were to call it directly. Thus, the text prompt must include the appropriate tags to indicate where the assets should be inserted (e.g. `<image>` or `<|AUDIO|>`).
-
-Example of a correctly formatted prompt for a vision model:
-```
-{
-    "text": "<image>Describe this image in one sentence:",
-    "images": PIL.Image.Image
-}
-```
-
-### Vision models
-
-For easier use, we recommend you to create a convenience function to load a `PIL.Image` from URL.
-```python
-from PIL import Image
-from io import BytesIO
-from urllib.request import urlopen
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-```
-
-You can then call the model with your prompts and images to generate text.
-```python
-from transformers import LlavaForConditionalGeneration, LlavaProcessor
-from outlines import models
-
-MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-
-model = models.from_transformers(
-    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
-    LlavaProcessor.from_pretrained(MODEL_NAME)
-)
-prompt = {
-    "text": "<image> detailed description:",
-    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
-}
-model(prompt)
-```
-
-You can include several images per prompt by adding more `<image>` tags to the prompt. Batching is also supported.
-```python
-from transformers import LlavaForConditionalGeneration
-from outlines import models
-
-MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-
-model = models.from_transformers(
-    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
-    LlavaProcessor.from_pretrained(MODEL_NAME)
-)
-prompt = {
-    "text": ["<image><image>detailed description:", "<image><image>. What animals are present?"],
-    "images": [
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"),
-        img_from_url("https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg"),
-    ]
-}
-model(prompt)
-```
-
-Here we have two prompts, each expecting two images. We correspondingly provide four images. This will generate two descriptions, one for each prompt.
-
-### Audio models
-
-For easier use, we recommend you to create a convenience function to turn a `.wav` file from a URL into an audio array.
-```python
-import soundfile
-from io import BytesIO
-from urllib.request import urlopen
-
-def audio_from_url(url):
-    audio_data = BytesIO(urlopen(url).read())
-    audio_array, _ = soundfile.read(audio_data)
-    return audio_array
-```
-
-You can then call the model with your prompt and audio files to generate text.
-```python
-from transformers import Qwen2AudioForConditionalGeneration, AutoProcessor
-from outlines import models
-
-MODEL_NAME = "Qwen/Qwen2-Audio-7B-Instruct"
-URL = "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2-Audio/audio/guess_age_gender.wav"
-
-tf_model = Qwen2AudioForConditionalGeneration.from_pretrained(MODEL_NAME)
-tf_processor = AutoProcessor.from_pretrained(MODEL_NAME)
-
-model = models.TransformersMultiModal(tf_model, tf_processor)
-
-text = """<|im_start|>user
-Audio 1: <|audio_bos|><|AUDIO|><|audio_eos|>
-<|im_end|>
-<|im_start|>user
-Write a sentence about what you hear in the audio.<|im_end|>
-<|im_start|>assistant
-"""
-prompt = {
-    "text": text,
-    "audios": audio_from_url(URL)
-}
-model(prompt)
-```
-
-You can include multiple audio files by providing a list for the `audios` argument and adding more of the appropriate tags to the prompt. Batching is also supported.
-
-## Use the model for structured generation
-
-You can use the model to generate structured data by providing a value for the parameter `output_type` (the second positional argument of the `generate` method, right after the prompt).
-
-You can use most common Python types along with the Outlines DSL types `JsonSchema`, `Regex` and `CFG`.
-
-For instance to do classification, you can use the `Regex` type:
-```python
-from outlines import models
-from outlines.types import Regex
-from transformers import LlavaForConditionalGeneration, LlavaProcessor
-
-MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-
-model = models.from_transformers(
-    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
-    LlavaProcessor.from_pretrained(MODEL_NAME)
-)
-pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
-prompt = {
-    "text": "<image>detailed description:",
-    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
-}
-model(prompt, Regex(pattern))
-```
-
-Another example could be to generated a structured description of an image using the `Json` type:
-```python
-from outlines import models
-from pydantic import BaseModel
-from transformers import LlavaForConditionalGeneration, LlavaProcessor
-from typing import List, Optional
-
-MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-
-class ImageData(BaseModel):
-    caption: str
-    tags_list: List[str]
-    object_list: List[str]
-    is_photo: bool
-
-model = models.from_transformers(
-    LlavaForConditionalGeneration.from_pretrained(MODEL_NAME),
-    LlavaProcessor.from_pretrained(MODEL_NAME)
-)
-pattern = "Mercury|Venus|Earth|Mars|Saturn|Jupiter|Neptune|Uranus|Pluto"
-prompt = {
-    "text": "<image>detailed description:",
-    "images": img_from_url("https://upload.wikimedia.org/wikipedia/commons/e/e3/Saturn_from_Cassini_Orbiter_%282004-10-06%29.jpg"),
-}
-model(prompt, Json(ImageData))
-```
-
-## Resources
-
-### Choosing a model
-- https://mmbench.opencompass.org.cn/leaderboard
-- https://huggingface.co/spaces/WildVision/vision-arena
diff --git a/docs_old/reference/models/vllm.md b/docs_old/reference/models/vllm.md
deleted file mode 100644
index fe4e55474..000000000
--- a/docs_old/reference/models/vllm.md
+++ /dev/null
@@ -1,51 +0,0 @@
-# vLLM
-
-## Prerequisites
-
-The Outlines `VLLM` model is inteded to be used along with a vllm instance running on a separate server (can be local or remote). Make sure you have a vllm server running before using the `VLLM` model. As the vllm client relies on the `openai` python sdk, you need to have an `openai` package installed. If you instead want to use the vllm offline inference mode, please refer to the [VLLMOffline model documentation](./vllm_offline.md).
-
-## Initialize the model
-
-To load the model, you can use the `from_vllm` function. The argument of the function is either an `OpenAI` or `AsyncOpenAI` instance from the `openai` library. Based on whether the `openai` instance is synchronous or asynchronous, you will receive a `VLLM` or `AsyncVLLM` model instance.
-
-```python
-import openai
-import outlines
-
-sync_openai_client = openai.OpenAI(base_url="...")
-async_openai_client = openai.AsyncOpenAI(base_url="...")
-
-sync_model = outlines.from_vllm(sync_openai_client)
-print(type(sync_model)) # <class 'outlines.models.vllm.VLLM'>
-
-async_model = outlines.from_vllm(async_openai_client)
-print(type(async_model)) # <class 'outlines.models.vllm.AsyncVLLM'>
-```
-
-## Generate text
-
-To generate text, you can call the model with a prompt as argument and optionally an output type to rely on structured generation:
-
-```python
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-answer = sync_model("Create a character.", output_type=Character)
-answer = await async_model("Create a character.", output_type=Character)
-```
-
-The `VLLM` model supports also supports streaming.
-
-```python
-for chunk in sync_model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk)
-```
-
-## Optional parameters
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `openai` client. An optional parameter of particular interest is `extra_body`, which is a dictionnary containing arguments that are specific to vLLM and are not part of the standard `openai` interface (see the [vLLM documentation][https://docs.vllm.ai/en/latest/serving/openai_compatible_server.html] on the OpenAI-compatible server for more information).
-
-[vllm-docs]: https://docs.vllm.ai/en/latest/
-[vllm-online-quickstart]: https://docs.vllm.ai/en/latest/getting_started/quickstart.html#quickstart-online
diff --git a/docs_old/reference/models/vllm_offline.md b/docs_old/reference/models/vllm_offline.md
deleted file mode 100644
index eac185b6d..000000000
--- a/docs_old/reference/models/vllm_offline.md
+++ /dev/null
@@ -1,68 +0,0 @@
-# vLLM Offline Inference mode
-
-
-!!! Note "Installation"
-
-    You need to install the `vllm` library to use the vLLM integration: `pip install vllm`. The default installation only works on machines with a GPU, follow the [installation section][vllm-install-cpu] for instructions to install vLLM for CPU or ROCm.
-
-    Consult the [vLLM documentation][vllm-docs] for detailed informations about how to initialize OpenAI clients and the available options.
-
-## Load the model
-
-Outlines supports models available via vLLM's offline batched inference interface. You can load a model using:
-
-```python
-import outlines
-from vllm import LLM
-
-model = outlines.from_vllm_offline(LLM("microsoft/Phi-3-mini-4k-instruct"))
-```
-
-Models are loaded from the [HuggingFace hub](https://huggingface.co/).
-
-
-!!! Warning "Device"
-
-    The default installation of vLLM only allows to load models on GPU. See the [installation instructions][vllm-install-cpu] to run models on CPU.
-
-
-## Generate text
-
-To generate text, you can just call the model with a prompt as argument:
-
-```python
-import outlines
-from vllm import LLM
-
-model = outlines.from_vllm_offline(LLM("microsoft/Phi-3-mini-4k-instruct"))
-answer = model("Write a short story about a cat.")
-```
-
-You can also use structured generation with the `VLLM` model by providing an output type after the prompt:
-
-```python
-import outlines
-from vllm import LLM
-from pydantic import BaseModel
-
-class Character(BaseModel):
-    name: str
-
-model = outlines.from_vllm_offline(LLM("microsoft/Phi-3-mini-4k-instruct"))
-answer = model("Create a character.", output_type=Character)
-```
-
-The VLLM model supports batch generation. To use it, you can pass a list of strings as prompt instead of a single string.
-
-## Optional parameters
-
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `LLM.generate` method of the `vllm` library. An optional parameter of particular interest is `sampling_params`, which is an instance of `SamplingParams`. You can find more information about it in the [vLLM documentation][https://docs.vllm.ai/en/latest/api/inference_params.html].
-
-!!! Warning
-
-    Streaming is not available for the offline vLLM integration.
-
-[vllm-docs]:https://docs.vllm.ai/en/latest/
-[vllm-install-cpu]: https://docs.vllm.ai/en/latest/getting_started/cpu-installation.html
-[vllm-install-rocm]: https://docs.vllm.ai/en/latest/getting_started/amd-installation.html
-[rocm-flash-attention]: https://github.com/ROCm/flash-attention/tree/flash_attention_for_rocm#amd-gpurocm-support
diff --git a/docs_old/reference/prompting.md b/docs_old/reference/prompting.md
deleted file mode 100644
index 986ca82fc..000000000
--- a/docs_old/reference/prompting.md
+++ /dev/null
@@ -1,395 +0,0 @@
-# Prompt templating
-
-Outlines provides a powerful domain-specific language to write and manage
-prompts, via what we call *prompt functions*.  Prompt functions are Python
-functions that contain a template for the prompt in their docstring, and their
-arguments correspond to the variables used in the prompt. When called, a prompt
-function returns the template rendered with the values of the arguments.
-
-The aim of prompt functions is to solve several recurrent problems with prompting:
-
-1. **Building complex prompts quickly leads to messy code.** This problem has
-   already been solved in the web development community by using templating, so
-   why not use it here?
-2. **Composing prompts is difficult.** Why not just compose functions?
-3. **Separating prompts from code.** Encapsulation in functions allows a clean
-   separation between prompts and code. Moreover, like any function, prompt
-   functions can be imported from other modules.
-
-Outlines uses the [Jinja templating
-engine](https://jinja.palletsprojects.com/en/3.1.x/) to render prompts, which
-allows to easily compose complex prompts.
-
-!!! warning "Prompt rendering"
-
-    Prompt functions are opinionated when it comes to prompt rendering. These opinions are meant to avoid common prompting errors, but can have unintended consequences if you are doing something unusual. We advise to always print the prompt before using it. You can also [read the
-    reference](#formatting-conventions) section if you want to know more.
-
-## Your first prompt
-
-The following snippet showcases a very simple prompt. The variables between
-curly brackets `{{  }}` are placeholders for the values of the arguments you
-will pass to the prompt function.
-
-=== "Code"
-
-    ```python title="greetings.py"
-    from outlines import Template
-
-    prompt = """Hello, {{ name }}!
-    {{ question }}"""
-
-    greetings = Template.from_string(prompt)
-    prompt = greetings("user", "How are you?")
-    print(prompt)
-    ```
-
-=== "Output"
-
-    ```text
-    Hello, user!
-    How are you?
-    ```
-
-If a variable is missing in the function's arguments, Jinja2 will throw an `UndefinedError` exception:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    prompt = """Hello, {{ surname }}!"""
-    greetings = Template.from_string(prompt)
-    prompt = greetings("user")
-    ```
-
-=== "Output"
-
-    ```text
-    Traceback (most recent call last):
-      File "<stdin>", line 9, in <module>
-      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 38, in __call__
-          return render(self.template, **bound_arguments.arguments)
-      File "/home/remi/projects/normal/outlines/outlines/templates.py", line 213, in render
-          return jinja_template.render(**values)
-      File "/home/remi/micromamba/envs/outlines/lib/python3.9/site-packages/jinja2/environment.py", line 1301, in render
-          self.environment.handle_exception()
-      File "/home/remi/micromamba/envs/outlines/lib/python3.9/site-packages/jinja2/environment.py", line 936, in handle_exception
-          raise rewrite_traceback_stack(source=source)
-      File "<template>", line 1, in top-level template code
-      jinja2.exceptions.UndefinedError: 'surname' is undefined
-    ```
-
-## Importing prompts from files
-
-Outlines allows you to read a prompt template from a text file. This way you can build "white space perfect" prompts, and version them independently from your code. We have found ourselves gravitating around this pattern a lot since Outlines came out:
-
-=== "prompt.txt"
-    ```text
-    """Hello, {{ name }}!
-    {{ question }}
-    """
-    ```
-
-=== "generate.py"
-
-    ```python
-    from outlines import Template
-
-    greetings = Template.from_file("prompt.txt")
-    prompt = greetings("John Doe", "How are you today?")
-    ```
-
-=== "Output"
-
-    ```text
-    Hello, John Doe!
-    How are you today?
-    ```
-
-## Few-shot prompting
-
-Few-shot prompting can lead to messy code. Prompt functions allow you to loop
-over lists or dictionaries from the template. In the following example we
-demonstrate how we can generate a prompt by passing a list of dictionaries with
-keys `question` and `answer` to the prompt function:
-
-=== "Code"
-
-    ```text title="prompt.txt"
-    {{ instructions }}
-
-    Examples
-    --------
-
-    {% for example in examples %}
-    Q: {{ example.question }}
-    A: {{ example.answer }}
-
-    {% endfor %}
-    Question
-    --------
-
-    Q: {{ question }}
-    A:
-    ```
-
-    ```python title="render.py"
-    from outlines import Template
-
-    instructions = "Please answer the following question following the examples"
-    examples = [
-        {"question": "2+2=?", "answer":4},
-        {"question": "3+3=?", "answer":6}
-    ]
-    question = "4+4 = ?"
-
-    few_shots = Template.from_file("prompt.txt")
-    prompt = few_shots(instructions, examples, question)
-    print(prompt)
-    ```
-
-=== "Output"
-
-    ```text
-    Please answer the following question following the examples
-
-    Examples
-    --------
-
-    Q: 2+2=?
-    A: 4
-
-    Q: 3+3=?
-    A: 6
-
-    Question
-    --------
-
-    Q: 4+4 = ?
-    A:
-    ```
-
-## Conditionals, filters, etc.
-
-Jinja2 has many features beyond looping that are not described here:
-conditionals, filtering, formatting, etc. Please refer to the [Jinja
-documentation](https://jinja.palletsprojects.com/en/3.1.x/>) for more
-information about the syntax of the templating language. The Jinja syntax is
-powerful, and we recommend you take some time to read their documentation if you
-are building complex prompts.
-
-
-## Tools
-
-Several projects (e.g.[Toolformer](https://arxiv.org/abs/2302.04761), [ViperGPT](https://viper.cs.columbia.edu/), [AutoGPT](https://github.com/Significant-Gravitas/Auto-GPT), etc.) have shown that we can "teach" language models to use external functions by describing what these functions do in the prompt. In these projects the same information is often repeated twice: the function implementation, name, docstring, or arguments are copy-pasted in the prompt. This is cumbersome and error prone; you can directly pull this information from within an Outlines prompt function:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    def my_tool(arg1: str, arg2: int):
-        """Tool description.
-
-        The rest of the docstring
-        """
-        pass
-
-    prompt = """{{ question }}
-
-    COMMANDS
-    1. {{ tool | name }}: {{ tool | description }}, args: {{ tool | args }}
-
-    {{ tool | source }}
-    """
-
-    tool_prompt = Template.from_string(prompt)
-    prompt = tool_prompt("Can you do something?", my_tool)
-    print(prompt)
-    ```
-
-=== "Output"
-
-    ```text
-    Can you do something?
-
-    COMMANDS
-    1. my_tool: Tool description., args: arg1: str, arg2: int
-
-    def my_tool(arg1: str, arg2: int):
-        """Tool description.
-
-        The rest of the docstring
-        """
-        pass
-    ```
-
-## JSON response format
-
-To build reliable chains with language models we often need to instruct them the
-format in which we would like them to return their response.
-
-Without prompt templating, the information is repeated twice between creating
-the parsing function (e.g. a Pydantic model), and writing the desired schema in
-the prompt. This can lead to errors that are hard to debug.
-
-Outlines allows you to directly pull the JSON schema of a pydantic model, or
-pretty print a dictionary from within an Outlines prompt function
-
-=== "Code"
-
-    ```python
-    from pydantic import BaseModel, Field
-
-   from outlines import Template
-
-    class MyResponse(BaseModel):
-        field1: int = Field(description="an int")
-        field2: str
-
-
-    my_prompt = Template.from_string("""{{ response_model | schema }}""")
-    prompt = my_prompt(MyResponse)
-    print(prompt)
-    # {
-    #   "field1": "an int",
-    #   "field2": "<field2>"
-    # }
-    ```
-
-=== "Output"
-
-    ```python
-    response = {
-        "field1": "<field1>",
-        "field2": "a string"
-    }
-
-    my_prompt(MyResponse)
-    # {
-    #   "field1": "<field1>",
-    #   "field2": "a string"
-    # }
-    ```
-
-## Formatting conventions
-
-Prompt templates are opinionated when it comes to rendering a template read from
-a string, and these opinions are meant to avoid prompting mistakes and help with
-formatting.
-
-### Whitespaces
-
-If you have experience working with strings between triple quotes you know that
-indenting has an influence on the string's formatting. Prompt functions adopt
-a few conventions so you don't have to think about indents when writing prompt.
-
-First, whether you start the prompt right after the triple quotes or on the line
-below does not matter for formatting:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-
-    prompt1 = Template.from_string(
-    """My prompt
-    """
-    )
-
-    prompt2 = Template.from_string(
-    """
-    My prompt
-    """
-    )
-
-    print(prompt1())
-    print(prompt2())
-    ```
-
-=== "Output"
-
-    ```text
-    My prompt
-    My prompt
-    ```
-
-Indentation is relative to the second line of the docstring, and leading spaces are removed:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    example1 = Template.from_string(
-    """First line
-    Second line
-    """
-    )
-
-    example2 = Template.from_string(
-    """
-        Second line
-        Third line
-    """
-    )
-
-    example3 = Template.from_string(
-    """
-        Second line
-          Third line
-    """
-    )
-
-    print(example1())
-    print(example2())
-    print(example3())
-    ```
-
-=== "Output"
-
-    ```text
-    First line
-    Second line
-
-    Second line
-    Third line
-
-    Second line
-      Third line
-    ```
-
-Trailing whitespaces are not removed, unless they follow a linebreak symbol `\` (see [linebreaks](#linebreaks)).
-
-### Linebreaks
-
-You can use the backslash `\` to break a long line of text. It will render as a single line:
-
-=== "Code"
-
-    ```python
-    from outlines import Template
-
-    example = Template.from_string(
-    """
-    Break in \
-    several lines \
-    But respect the indentation
-        on line breaks.
-    And after everything \
-    Goes back to normal
-    """
-    )
-
-    print(example())
-    ```
-
-=== "Output"
-
-    ```text
-    Break in several lines But respect the indentation
-        on line breaks.
-    And after everything Goes back to normal
-    ```
diff --git a/docs_old/reference/regex_dsl.md b/docs_old/reference/regex_dsl.md
deleted file mode 100644
index f42b6dd34..000000000
--- a/docs_old/reference/regex_dsl.md
+++ /dev/null
@@ -1,344 +0,0 @@
-# DSL to express constraints
-
-This library provides a Domain-Specific Language (DSL) to construct regular expressions in a more intuitive and modular way. It allows you to create complex regexes using simple building blocks that represent literal strings, patterns, and various quantifiers. Additionally, these custom regex types can be used directly as types in [Pydantic](https://pydantic-docs.helpmanual.io/) schemas to enforce pattern constraints during text generation.
-
----
-
-## Why Use This DSL?
-
-1. **Modularity & Readability**: Instead of writing cryptic regular expression strings, you compose a regex as a tree of objects.
-2. **Enhanced Debugging**: Each expression can be visualized as an ASCII tree, making it easier to understand and debug complex regexes.
-3. **Pydantic Integration**: Use your DSL-defined regex as types in Pydantic models. The DSL seamlessly converts to JSON Schema with proper pattern constraints.
-4. **Extensibility**: Easily add or modify quantifiers and other regex components by extending the provided classes.
-
----
-
-## Building Blocks
-
-
-Every regex component in this DSL is a **Term**. Here are two primary types:
-
-- **`String`**: Represents a literal string. It escapes the characters that have a special meaning in regular expressions.
-- **`Regex`**: Represents an existing regex pattern string.
-
-```python
-from outlines.types import String, Regex
-
-# A literal string "hello"
-literal = String("hello")   # Internally represents "hello"
-
-# A regex pattern to match one or more digits
-digit = Regex(r"[0-9]+")     # Internally represents the pattern [0-9]+
-
-# Converting to standard regex strings:
-from outlines.types.dsl import to_regex
-
-print(to_regex(literal))  # Output: hello
-print(to_regex(digit))    # Output: [0-9]+
-```
-
----
-
-## Early Introduction to Quantifiers & Combining Terms
-
-The DSL supports common regex quantifiers as methods on every `Term`. These methods allow you to specify how many times a pattern should be matched. They include:
-
-- **`exactly(count)`**: Matches the term exactly `count` times.
-- **`optional()`**: Matches the term zero or one time.
-- **`one_or_more()`**: Matches the term one or more times (Kleene Plus).
-- **`zero_or_more()`**: Matches the term zero or more times (Kleene Star).
-- **`between(min_count, max_count)`**: Matches the term between `min_count` and `max_count` times (inclusive).
-- **`at_least(count)`**: Matches the term at least `count` times.
-- **`at_most(count)`**: Matches the term up to `count` times.
-
-These quantifiers can also be used as functions that take the `Term` as an argument. If the term is a plain string, it will be automatically converted to a `String` object. Thus `String("foo").optional()` is equivalent to `optional("foo")`.
-
-Let's see these quantifiers side by side with examples.
-
-### Quantifiers in Action
-
-#### `exactly(count)`
-
-This method restricts the term to appear exactly `count` times.
-
-```python
-# Example: exactly 5 digits
-five_digits = Regex(r"\d").exactly(5)
-print(to_regex(five_digits))  # Output: (\d){5}
-```
-
-You can also use the `exactly` function:
-
-```python
-from outlines.types import exactly
-
-# Example: exactly 5 digits
-five_digits = exactly(Regex(r"\d"), 5)
-print(to_regex(five_digits))  # Output: (\d){5}
-```
-
-#### `optional()`
-
-This method makes a term optional, meaning it may occur zero or one time.
-
-```python
-# Example: an optional "s" at the end of a word
-maybe_s = String("s").optional()
-print(to_regex(maybe_s))  # Output: (s)?
-```
-
-You can also use the `optional` function:
-
-```python
-from outlines.types import optional
-
-# Example: an optional "s" at the end of a word
-maybe_s = optional("s")
-print(to_regex(maybe_s))  # Output: (s)?
-```
-
-#### `one_or_more()`
-
-This method indicates that the term must appear at least once.
-
-```python
-# Example: one or more alphabetic characters
-letters = Regex(r"[A-Za-z]").one_or_more()
-print(to_regex(letters))  # Output: ([A-Za-z])+
-```
-
-You can also use the `one_or_more` function:
-
-```python
-from outlines.types import one_or_more
-
-# Example: one or more alphabetic characters
-letters = one_or_more(Regex(r"[A-Za-z]"))
-print(to_regex(letters))  # Output: ([A-Za-z])+
-
-```
-
-#### `zero_or_more()`
-
-This method indicates that the term can occur zero or more times.
-
-```python
-# Example: zero or more spaces
-spaces = String(" ").zero_or_more()
-print(to_regex(spaces))  # Output: ( )*
-```
-
-You can also use the `zero_or_more` function:
-
-```python
-from outlines.types import zero_or_more
-
-# Example: zero or more spaces
-spaces = zero_or_more(" ")
-print(to_regex(spaces))  # Output: ( )*
-```
-
-#### `between(min_count, max_count)`
-
-This method indicates that the term can appear any number of times between `min_count` and `max_count` (inclusive).
-
-```python
-# Example: Between 2 and 4 word characters
-word_chars = Regex(r"\w").between(2, 4)
-print(to_regex(word_chars))  # Output: (\w){2,4}
-```
-
-You can also use the `between` function:
-
-```python
-from outlines.types import between
-
-# Example: Between 2 and 4 word characters
-word_chars = between(Regex(r"\w"), 2, 4)
-print(to_regex(word_chars))  # Output: (\w){2,4}
-```
-
-#### `at_least(count)`
-
-This method indicates that the term must appear at least `count` times.
-
-```python
-# Example: At least 3 digits
-at_least_three = Regex(r"\d").at_least(3)
-print(to_regex(at_least_three))  # Output: (\d){3,}
-```
-
-You can also use the `at_least` function:
-
-```python
-from outlines.types import at_least
-
-# Example: At least 3 digits
-at_least_three = at_least(Regex(r"\d"), 3)
-print(to_regex(at_least_three))  # Output: (\d){3,}
-```
-
-#### `at_most(count)`
-
-This method indicates that the term can appear at most `count` times.
-
-```python
-# Example: At most 3 digits
-up_to_three = Regex(r"\d").at_most(3)
-print(to_regex(up_to_three))  # Output: (\d){0,3}
-```
-
-You can also use the `at_most` function:
-
-```python
-from outlines.types import at_most
-
-# Example: At most 3 digits
-up_to_three = at_most(Regex(r"\d"), 3)
-print(to_regex(up_to_three))  # Output: (\d){0,3}
-```
-
----
-
-## Combining Terms
-
-The DSL allows you to combine basic terms into more complex patterns using concatenation and alternation.
-
-### Concatenation (`+`)
-
-The `+` operator (and its reflected variant) concatenates terms, meaning that the terms are matched in sequence.
-
-```python
-# Example: Match "hello world"
-pattern = String("hello") + " " + Regex(r"\w+")
-print(to_regex(pattern))  # Output: hello\ (\w+)
-```
-
-### Alternation (`either()`)
-
-The `either()` function creates alternatives, allowing a match for one of several patterns. You can provide as many terms as you want.
-
-```python
-# Example: Match either "cat" or "dog" or "mouse"
-animal = either(String("cat"), "dog", "mouse")
-print(to_regex(animal))  # Output: (cat|dog|mouse)
-```
-
-*Note:* When using `either()` with plain strings (such as `"dog"`), the DSL automatically wraps them in a `String` object that escapes the characters that have a special meaning in regular expressions, just like with quantifier functions.
-
----
-
-## Custom types
-
-The DSL comes "batteries included" with types that represent common text constructs:
-
-- `integer` represents an integer number as recognized by `int`
-- `boolean` represents a boolean, "True" or "False" as recognized by `bool`
-- `number` represents a floating-point number recognize by Python's `float`
-- `date` represents a date as understood by `datetime.date`
-- `time` represents a time as undestood by `datetime.time`
-- `datetime` represents a time as understoof by `datetime.datetime`
-- `digit` represents a single digit
-- `char` represents a single character
-- `newline` represents a new line character
-- `whitespace` represents a white space
-- `sentence` represents a sentence
-- `paragraph` reprensents a pagraph (one or more sentences separated by one or more ilne breaks)
-
-
-For instance you can describe the answers in the GSM8K dataset using the following pattern:
-
-```python
-from outlines.types import sentence, digit
-
-answer = "A: " + sentence.between(2,4) + " So the answer is: " + digit.between(1,4)
-```
-
----
-
-## Practical Examples
-
-### Example 1: Matching a Custom ID Format
-
-Suppose you want to create a regex that matches an ID format like "ID-12345", where:
-- The literal "ID-" must be at the start.
-- Followed by exactly 5 digits.
-
-```python
-id_pattern = "ID-" + Regex(r"\d").exactly(5)
-print(to_regex(id_pattern))  # Output: ID-(\d){5}
-```
-
-### Example 2: Email Validation with Pydantic
-
-You can define a regex for email validation and use it as a type in a Pydantic model.
-
-```python
-from pydantic import BaseModel, ValidationError
-
-# Define an email regex term (this is a simplified version)
-email_regex = Regex(r"[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+")
-
-class User(BaseModel):
-    name: str
-    email: email_regex  # Use our DSL regex as a field type
-
-# Valid input
-user = User(name="Alice", email="alice@example.com")
-print(user)
-
-# Invalid input (raises a ValidationError)
-try:
-    User(name="Bob", email="not-an-email")
-except ValidationError as e:
-    print(e)
-```
-
-When used in a Pydantic model, the email field is automatically validated against the regex pattern and its JSON Schema includes the `pattern` constraint.
-
-### Example 3: Building a Complex Pattern
-
-Consider a pattern to match a simple date format: `YYYY-MM-DD`.
-
-```python
-year = Regex(r"\d").exactly(4)         # Four digits for the year
-month = Regex(r"\d").exactly(2)        # Two digits for the month
-day = Regex(r"\d").exactly(2)          # Two digits for the day
-
-# Combine with literal hyphens
-date_pattern = year + "-" + month + "-" + day
-print(to_regex(date_pattern))
-# Output: (\d){4}\-(\d){2}\-(\d){2}
-```
-
----
-
-## Visualizing Your Pattern
-
-One of the unique features of this DSL is that each term can print its underlying structure as an ASCII tree. This visualization can be particularly helpful when dealing with complex expressions.
-
-```python
-# A composite pattern using concatenation and quantifiers
-pattern = "a" + String("b").one_or_more() + "c"
-print(pattern)
-```
-
-*Expected Output:*
-
-```
-└── Sequence
-    ├── String('a')
-    ├── KleenePlus(+)
-    │   └── String('b')
-    └── String('c')
-```
-
-This tree representation makes it easy to see the hierarchy and order of operations in your regular expression.
-
----
-
-## Final Words
-
-This DSL is designed to simplify the creation and management of regular expressions—whether you're validating inputs in a web API, constraining the output of an LLM, or just experimenting with regex patterns. With intuitive methods for common quantifiers and operators, clear visual feedback, and built-in integration with Pydantic, you can build robust and maintainable regex-based validations with ease.
-
-Feel free to explore the library further and adapt the examples to your use cases. Happy regexing!
diff --git a/docs_old/reference/serve/lmstudio.md b/docs_old/reference/serve/lmstudio.md
deleted file mode 100644
index db383186a..000000000
--- a/docs_old/reference/serve/lmstudio.md
+++ /dev/null
@@ -1,89 +0,0 @@
-# Serve with LM Studio
-
-!!! tip "Would rather not self-host?"
-
-    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
-
-[LM Studio](https://lmstudio.ai/) is an application that runs local LLMs. It flexibly mixes GPU and CPU compute in hardware-constrained environments.
-
-As of [LM Studio 0.3.4](https://lmstudio.ai/blog/lmstudio-v0.3.4), it natively supports Outlines for structured text generation, using an OpenAI-compatible endpoint.
-
-## Setup
-
-1. Install LM Studio by visiting their [downloads page](https://lmstudio.ai/download).
-2. Enable the LM Studio [server functionality](https://lmstudio.ai/docs/basics/server).
-3. Download [a model](https://lmstudio.ai/docs/basics#1-download-an-llm-to-your-computer).
-4. Install Python dependencies.
-```bash
-pip install pydantic openai
-```
-
-## Calling the server
-
-By default, LM Studio will serve from `http://localhost:1234`. If you are serving on a different port or host, make sure to change the `base_url` argument in `OpenAI` to the relevant location.
-
-```python
-class Testing(BaseModel):
-    """
-    A class representing a testing schema.
-    """
-    name: str
-    age: int
-
-openai_client = openai.OpenAI(
-    base_url="http://0.0.0.0:1234/v1",
-    api_key="dopeness"
-)
-
-# Make a request to the local LM Studio server
-response = openai_client.beta.chat.completions.parse(
-    model="hugging-quants/Llama-3.2-1B-Instruct-Q8_0-GGUF",
-    messages=[
-        {"role": "system", "content": "You are like so good at whatever you do."},
-        {"role": "user", "content": "My name is Cameron and I am 28 years old. What's my name and age?"}
-    ],
-    response_format=Testing
-)
-```
-
-You should receive a `ParsedChatCompletion[Testing]` object back:
-
-```python
-ParsedChatCompletion[Testing](
-    id='chatcmpl-3hykyf0fxus7jc90k6gwlw',
-    choices=[
-        ParsedChoice[Testing](
-            finish_reason='stop',
-            index=0,
-            logprobs=None,
-            message=ParsedChatCompletionMessage[Testing](
-                content='{ "age": 28, "name": "Cameron" }',
-                refusal=None,
-                role='assistant',
-                function_call=None,
-                tool_calls=[],
-                parsed=Testing(name='Cameron', age=28)
-            )
-        )
-    ],
-    created=1728595622,
-    model='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-Q4_K_M.gguf',
-    object='chat.completion',
-    service_tier=None,
-    system_fingerprint='lmstudio-community/Phi-3.1-mini-128k-instruct-GGUF/Phi-3.1-mini-128k-instruct-
-Q4_K_M.gguf',
-    usage=CompletionUsage(
-        completion_tokens=17,
-        prompt_tokens=47,
-        total_tokens=64,
-        completion_tokens_details=None,
-        prompt_tokens_details=None
-    )
-)
-```
-
-You can retrieve your `Testing` object with
-
-```python
-response.choices[0].message.parsed
-```
diff --git a/docs_old/reference/serve/vllm.md b/docs_old/reference/serve/vllm.md
deleted file mode 100644
index 14277a526..000000000
--- a/docs_old/reference/serve/vllm.md
+++ /dev/null
@@ -1,73 +0,0 @@
-# Serve with vLLM
-
-!!! tip "Would rather not self-host?"
-
-    If you want to get started quickly with JSON-structured generation you can call instead [.json](https://h1xbpbfsf0w.typeform.com/to/ZgBCvJHF), a [.txt](http://dottxt.co) API that guarantees valid JSON.
-
-Outlines can be deployed as an LLM service using the vLLM inference engine and a FastAPI server. vLLM is not installed by default so will need to install Outlines with:
-
-```bash
-pip install outlines[serve]
-```
-
-You can then start the server with:
-
-```bash
-python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-This will by default start a server at `http://127.0.0.1:8000` (check what the console says, though). Without the `--model` argument set, the OPT-125M model is used. The `--model` argument allows you to specify any model of your choosing.
-
-To run inference on multiple GPUs you must pass the `--tensor-parallel-size` argument when initializing the server. For instance, to run inference on 2 GPUs:
-
-
-```bash
-python -m outlines.serve.serve --model="microsoft/Phi-3-mini-4k-instruct" --tensor-parallel-size 2
-```
-
-
-### Alternative Method: Via Docker
-
-You can install and run the server with Outlines' official Docker image using the command
-
-```bash
-docker run -p 8000:8000 outlinesdev/outlines --model="microsoft/Phi-3-mini-4k-instruct"
-```
-
-## Querying Endpoint
-
-You can then query the model in shell by passing a prompt and either
-
-1. a [JSON Schema][jsonschema]{:target="_blank"} specification or
-2. a [Regex][regex]{:target="_blank"} pattern
-
-with the `schema` or `regex` parameters, respectively, to the `/generate` endpoint. If both are specified, the schema will be used. If neither is specified, the generated text will be unconstrained.
-
-For example, to generate a string that matches the schema `{"type": "string"}` (any string):
-
-```bash
-curl http://127.0.0.1:8000/generate \
-    -d '{
-        "prompt": "What is the capital of France?",
-        "schema": {"type": "string", "maxLength": 5}
-        }'
-```
-
-To generate a string that matches the regex `(-)?(0|[1-9][0-9]*)(\.[0-9]+)?([eE][+-][0-9]+)?` (a number):
-
-```bash
-curl http://127.0.0.1:8000/generate \
-    -d '{
-        "prompt": "What is Pi? Give me the first 15 digits: ",
-        "regex": "(-)?(0|[1-9][0-9]*)(\\.[0-9]+)?([eE][+-][0-9]+)?"
-        }'
-```
-
-Instead of `curl`, you can also use the [requests][requests]{:target="_blank"} library from another python program.
-
-Please consult the [vLLM documentation][vllm]{:target="_blank"} for details on additional request parameters. You can also [read the code](https://github.com/dottxt-ai/outlines/blob/main/outlines/serve/serve.py) in case you need to customize the solution to your needs.
-
-[requests]: https://requests.readthedocs.io/en/latest/
-[vllm]: https://docs.vllm.ai/en/latest/index.html
-[jsonschema]: https://json-schema.org/learn/getting-started-step-by-step
-[regex]: https://www.regular-expressions.info/tutorial.html
diff --git a/docs_old/stylesheets/extra.css b/docs_old/stylesheets/extra.css
deleted file mode 100644
index e71a6124f..000000000
--- a/docs_old/stylesheets/extra.css
+++ /dev/null
@@ -1,149 +0,0 @@
-@font-face {
-  font-family: 'Source Code Pro Custom', monospace;
-  src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
-}
-
-:root > * {
-  --md-default-bg-color: #FFFFFF;
-  --md-code-bg-color: #2E3440;
-  --md-code-fg-color: #FFFFFF;
-  --md-text-font-family: "Inter";
-  --md-code-font: "Source Code Pro Custom";
-  --md-typeset-a-color: #d16626; /*this is the brand color*/
-
-  /* don't inherit white fg color for mermaid diagrams from --md-code-fg-color */
-  --md-mermaid-label-fg-color: #000000;
-  --md-mermaid-edge-color: #000000;
-}
-
-.index-pre-code {
-  max-width: 700px;
-  left: 50%;
-}
-
-.index-pre-code pre>code {
-  text-align: left;
-}
-
-.md-clipboard::after {
-  color: #FFFFFF;
-  transition: color 0.3s ease-in-out;
-}
-
-.md-clipboard:hover::after {
-  color: #D8DEE9;
-}
-
-.md-source-file {
-  text-align: center;
-  padding: 24px 0;
-}
-
-.md-typeset pre>code {
-  border-radius: .2rem;
-  box-shadow: 10px 5px 5px #D8DEE9;
-}
-
-.md-typeset p > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset strong > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-content p > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset td > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset li > code {
-  background: #ECEFF4;
-  color: #000000;
-  font-weight: 500;
-}
-
-.md-typeset code {
-  font-weight: 500;
-}
-
-.md-typeset pre {
-  margin-left: .5rem;
-  margin-right: .5rem;
-  margin-top: 2rem;
-  margin-bottom: 2rem;
-}
-
-.language-python {
-  background: #FFFFFF ! important
-}
-
-.language-bash {
-  background: #FFFFFF ! important
-}
-
-.language-toml {
-  background: #FFFFFF ! important
-}
-
-.language-text {
-  background: #FFFFFF ! important
-}
-
-.language-json {
-  background: #FFFFFF ! important
-}
-
-h1.title {
-  color: #FFFFFF;
-  margin: 0px 0px 5px;
-}
-
-h2.subtitle {
-  margin: 5px 0px 25px;
-  font-size: 1rem;
-  max-width: 540px;
-  margin: 0 auto;
-}
-
-.md-typeset {
-  line-height: 24px;
-  font-weight: 400;
-}
-
-.md-typeset h1 {
-  font-weight: bold;
-  color: #000000;
-}
-
-.md-typeset h2 {
-  font-weight: bold;
-  color: #000000;
-}
-
-span.md-ellipsis {
-  color: black;
-}
-
-.md-nav__link--active {
-  background-color: #ECEFF4;
-}
-
-.md-typeset ol li {
-  margin-bottom: .1rem;
-}
-
-.md-typeset ul li {
-  margin-bottom: .1rem;
-}
diff --git a/docs_old/welcome.md b/docs_old/welcome.md
deleted file mode 100644
index bd7377f5f..000000000
--- a/docs_old/welcome.md
+++ /dev/null
@@ -1,183 +0,0 @@
----
-title: Welcome to Outlines!
----
-
-Outlines is a Python library that allows you to use Large Language Model in a simple and robust way (with structured generation). It is built by [.txt][.txt]{:target="_blank"}, and is already used in production by many companies.
-
-## What models do you support?
-
-We support [Openai](reference/models/openai.md), but the true power of Outlines is unleashed with Open Source models available via the [transformers](reference/models/transformers.md), [llama.cpp](reference/models/llamacpp.md), [mlx-lm](reference/models/mlxlm.md) and [vllm](reference/models/vllm.md) models. If you want to build and maintain an integration with another library, [get in touch][discord].
-
-## What are the main features?
-
-<div class="grid cards" markdown>
--   :material-code-json:{ .lg .middle } __Make LLMs generate valid JSON__
-
-    ---
-
-    No more invalid JSON outputs, 100% guaranteed
-
-    [:octicons-arrow-right-24: Generate JSON](reference/generation/json.md)
-
--   :material-keyboard-outline:{ .lg .middle } __JSON mode for vLLM__
-
-    ---
-
-    Deploy a LLM service using Outlines' JSON structured generation and vLLM
-
-    [:octicons-arrow-right-24: Deploy outlines](reference/serve/vllm.md)
-
-
--   :material-regex:{ .lg .middle } __Make LLMs follow a Regex__
-
-    ---
-
-    Generate text that parses correctly 100% of the time
-
-    [:octicons-arrow-right-24: Guide LLMs](reference/generation/regex.md)
-
--    :material-chat-processing-outline:{ .lg .middle } __Powerful Prompt Templating__
-
-     ---
-
-     Better manage your prompts' complexity with prompt templating
-
-    [:octicons-arrow-right-24: Learn more](reference/prompting.md)
-</div>
-
-## Why use Outlines?
-
-
-Outlines is built at [.txt][.txt] by engineers with decades of experience in software engineering, machine learning (Bayesian Statistics and NLP), and compilers. [.txt][.txt] is a VC-backed company fully focused on the topic of structured generation and is committed to make the community benefit from its experience.
-
-We are also open source veterans and have authored/maintained many libraries over the years: the [Aesara][aesara]{:target="_blank"} and [Pythological][pythological]{:target="_blank"} ecosystems, [Blackjax][blackjax]{:target="_blank"} and [Hy][hy]{:target="_blank"} among many others.
-.
-
-Outlines does not use unnecessary abstractions that tend to get in your way. We have a laser focus on reliable text generation with LLMs, a clear roadmap to push the state of the art in this area and a commitment to clean and robust code.
-
-And last but not least, unlike alternatives, Outlines' structured generation introduces **no overhead** during inference.
-
-
-## Who is using Outlines?
-
-Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) are using Outlines. Some of the prominent companies and organizations that are using Outlines include:
-
-<head>
-  <style>
-  .row {
-      display: inline-block;
-      width: 100%;
-      margin-bottom: 50px;
-      margin-top: 0px !important;
-      break-inside: avoid;
-  }
-
-  /* Create two equal columns that sits next to each other */
-  .column {
-      column-count: 3;
-      column-gap: 20px;
-      padding: 20px;
-  }
-
-  </style>
-</head>
-<body>
-
-<div class="column">
-  <div class="row"><img src="../logos/amazon.png" width="200"></div>
-  <div class="row"><img src="../logos/apple.png" width="200"></div>
-  <div class="row"><img src="../logos/best_buy.png" width="200"></div>
-  <div class="row"><img src="../logos/canoe.png" width="200"></div>
-  <div class="row"><img src="../logos/cisco.png" width="200"></div>
-  <div class="row"><img src="../logos/dassault_systems.png" width="200"></div>
-  <div class="row"><img src="../logos/databricks.png" width="200"></div>
-  <div class="row"><img src="../logos/datadog.png" width="200"></div>
-  <div class="row"><img src="../logos/dbt_labs.png" width="200"></div>
-  <div class="row"><img src="../assets/images/dottxt.png" width="200"></div>
-  <div class="row"><img src="../logos/gladia.jpg" width="200"></div>
-  <div class="row"><img src="../logos/harvard.png" width="200"></div>
-  <div class="row"><img src="../logos/hf.png" width="200"></div>
-  <div class="row"><img src="../logos/johns_hopkins.png" width="200"></div>
-  <div class="row"><img src="../logos/meta.png" width="200"></div>
-  <div class="row"><img src="../logos/mit.png" width="200"></div>
-  <div class="row"><img src="../logos/mount_sinai.png" width="200"></div>
-  <div class="row"><img src="../logos/nvidia.png" width="200"></div>
-  <div class="row"><img src="../logos/nyu.png" width="200"></div>
-  <div class="row"><img src="../logos/safran.png" width="200"></div>
-  <div class="row"><img src="../logos/salesforce.png" width="200"></div>
-  <div class="row"><img src="../logos/shopify.png" width="200"></div>
-  <div class="row"><img src="../logos/smithsonian.png" width="200"></div>
-  <div class="row"><img src="../logos/tinder.png" width="200"></div>
-  <div class="row"><img src="../logos/upenn.png" width="200"></div>
-</div>
-
-</body>
-
-Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
-
-Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
-
-## Philosophy
-
-**Outlines**  is a library for neural text generation. You can think of it as a
-more flexible replacement for the `generate` method in the
-[transformers](https://github.com/huggingface/transformers) library.
-
-**Outlines**  helps developers *structure text generation* to build robust
-interfaces with external systems. It provides generation methods that
-guarantee that the output will match a regular expressions, or follow
-a JSON schema.
-
-**Outlines**  provides *robust prompting primitives* that separate the prompting
-from the execution logic and lead to simple implementations of few-shot
-generations, ReAct, meta-prompting, agents, etc.
-
-**Outlines**  is designed as a *library* that is meant to be compatible the
-broader ecosystem, not to replace it. We use as few abstractions as possible,
-and generation can be interleaved with control flow, conditionals, custom Python
-functions and calls to other libraries.
-
-**Outlines**  is *compatible with every auto-regressive model*. It only interfaces with models
-via the next-token logits distribution.
-
-## Outlines people
-
-Outlines would not be what it is today without a community of dedicated developers:
-
-<a href="https://github.com/dottxt-ai/outlines/graphs/contributors">
-  <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
-</a>
-
-## Acknowledgements
-
-<div class="grid" markdown>
-
-
-<figure markdown>
-  <a href="http://www.dottxt.co">
-  ![Normal Computing logo](assets/images/dottxt.png){ width="150" }
-  </a>
-</figure>
-
-<figure markdown>
-  <a href="https://www.normalcomputing.ai">
-  ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
-  </a>
-</figure>
-
-</div>
-
-Outlines was originally developed at [@NormalComputing](https://twitter.com/NormalComputing) by [@remilouf](https://twitter.com/remilouf) and [@BrandonTWillard](https://twitter.com/BrandonTWillard). It is now maintained by [.txt](https://dottxt.co).
-
-
-[discord]: https://discord.gg/R9DSu34mGd
-[aesara]: https://github.com/aesara-devs
-[blackjax]: https://github.com/blackjax-devs/blackjax
-[pythological]: https://github.com/pythological
-[hy]: https://hylang.org/
-[.txt]: https://dottxt.co
-[vllm]: https://github.com/vllm-project/vllm
-[tgi]: https://github.com/huggingface/text-generation-inference
-[lorax]: https://github.com/predibase/lorax
-[xinference]: https://github.com/xorbitsai/inference
-[sglang]: https://github.com/sgl-project/sglang/
diff --git a/outlines/release_note.md b/outlines/release_note.md
new file mode 100644
index 000000000..35812623b
--- /dev/null
+++ b/outlines/release_note.md
@@ -0,0 +1,392 @@
+# Release Note
+
+### Why a new major version?
+
+The v1 intends on making Outlines more closely focused on constrained generation. To do so, we delegate a wider range of tasks to the users and inference libraries. On top of making Outlines leaner, this design provides more flexibility to the users and let them use interfaces they are already familiar with.
+
+Our approach is inspired by the unix best practices — each element does one thing well, and we compose those functional elements.
+
+As this new version deprecates some previously available features of Outlines, we have written a migration guide that gives detailed information on how to upgrade your v0 code to v1.
+
+### Deprecated
+
+All deprecated features listed below will be removed in version 1.1.0. Until then, a warning will be displayed with information on how to migrate your code to v1.
+
+- The model loader functions from the `models` module (`transformers`, `openai`, etc.) have been deprecated. They are replaced by equivalent functions prefixed with `from_` such as `from_transformers`, `from_openai`, etc. The new loader functions accept different arguments compared to the old ones. They now typically require an instance of an engine/client from the associated inference library. This change was made to avoid duplicating inference library logic and to give users more control over inference engine/client initialization.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+```python
+# v0
+from outlines import models
+from transformers import BertForSequenceClassification, BertTokenizer
+
+model = models.transformers(
+    model_name="prajjwal1/bert-tiny",
+    model_class=BertForSequenceClassification,
+    tokenizer_class=BertTokenizer,
+    model_kwargs={"use_cache": False},
+    tokenizer_kwargs={"model_max_length": 512},
+)
+
+# v1
+import outlines
+from transformers import BertForSequenceClassification, BertTokenizer
+
+hf_model = BertForSequenceClassification.from_pretrained("prajjwal1/bert-tiny", use_cache=False)
+hf_tokenizer = BertTokenizer.from_pretrained("prajjwal1/bert-tiny", model_max_length=512)
+model = outlines.from_transformers(hf_model, hf_tokenizer)
+```
+
+- The `generate` module and the associated functions (`json`, `choice`…) have been deprecated. They are replaced by the `Generator` constructor. While you had to select the right generate function for your output type, you can now provide any output type supported by Outlines to the unique `Generator` object.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/generator)
+
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import generate, models
+
+class Character(BaseModel):
+	name: str
+
+model = models.openai("gpt-4o")
+generator = generate.json(model, Character)
+
+# v1
+from openai import OpenAI
+from pydantic import BaseModel
+from outlines import Generator, from_openai
+
+class Character(BaseModel):
+	name: str
+
+model = from_openai(OpenAI())
+generator = Generator(model, Character)
+```
+
+- The `TransformersVision` model has been deprecated. It's replaced by `TransformersMultiModal`, which is more general as it supports additional input types beyond images, such as audio. When calling it, instead of providing the prompt and image assets separately, both should now be included in a single dictionary. The model is loaded with `from_transformers` just like the `Transformers` model, but the second argument must be a processor instead of a tokenizer.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models/transformers_multimodal)
+
+
+```python
+# v0
+from io import BytesIO
+from urllib.request import urlopen
+from PIL import Image
+from transformers import LlavaForConditionalGeneration
+from outlines import models, generate
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+model = models.transformers_vision(
+    model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
+    model_class=LlavaForConditionalGeneration,
+)
+generator = generate.text(model)
+result = generator(
+    "Describe the image <image>",
+    img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
+)
+
+# v1
+from io import BytesIO
+from urllib.request import urlopen
+from PIL import Image
+from transformers import LlavaForConditionalGeneration, AutoProcessor
+import outlines
+
+def img_from_url(url):
+    img_byte_stream = BytesIO(urlopen(url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+model = outlines.from_transformers(
+	LlavaForConditionalGeneration.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration"),
+	AutoProcessor.from_pretrained("trl-internal-testing/tiny-LlavaForConditionalGeneration")
+)
+image = img_from_url("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg")
+result = model({"text": "Describe the image <image>", "images": image})
+```
+
+- The `Exllamav2` model has been deprecated without replacement because its interface is not fully compatible with Outlines. We had to implement cumbersome patching to make it work, so we decided to remove it entirely.
+
+- The `function` module and the associated `Function` class have been deprecated. They are replaced by the `Application` class, which serves a similar purpose to `Function`. There are two notable differences: an `Application` is not initialized with a model (a model must be provided when calling the object), and template variables must be provided in a dictionary instead of as keyword arguments when calling the `Application`.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/utility/application)
+
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import Function, Template
+
+class Character(BaseModel):
+	name: str
+
+template = Template.from_string("Create a {{ gender }} character.")
+fn = Function(template, Character, "hf-internal-testing/tiny-random-GPTJForCausalLM")
+response = fn(gender="female")
+
+# v1
+from pydantic import BaseModel
+from outlines import Application, Template, from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+class Character(BaseModel):
+	name: str
+
+model = from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+template = Template.from_string("Create a {{ gender }} character.")
+app = Application(template, Character)
+response = app(model, {"gender": "female"})
+```
+
+- The `samplers` module and the associated objects (`multinomial`, `greedy`…) have been deprecated. You should now use the inference arguments specific to the inference library your  model is based on to control the sampling.
+
+```python
+# v0
+from outlines import generate, models, samplers
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model, samplers.beam_search(2))
+response = generator("Write a short story about a cat", max_tokens=10)
+
+# v1
+from outlines import Generator, from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+response = model("Write a short story about a cat", num_beams=2)
+```
+
+- The `load_lora` methods on the `VLLM` and `LlamaCpp` models have been deprecated. You should now load through the `Llama` instance provided when initializing the model in the case of the `LlamaCpp` model, and provide it as a keyword argument when calling the model in the case of the `VLLM` model.
+
+```python
+# v0
+from outlines import models
+from vllm import LLM
+
+model = models.vllm("erwanf/gpt2-mini")
+model.load_lora("path/to/lora/file")
+response = model("Write a short story about a cat.")
+
+#v1
+from outlines import from_vllm
+from vllm import LLM
+from vllm.lora.request import LoRARequest
+
+model = from_vllm(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+lora_request = LoRARequest("path/to/lora/file", 1, "path/to/lora/file")
+response = model("Write a short story about a cat.", lora_request=lora_request)
+```
+
+### Modified
+
+Some objects are maintained, but their interface or behavior has been modified.
+
+- The interface of `Model` classes (`Transformers`, `OpenAI`, etc.) has been significantly modified. Models can now be called directly with a prompt and an output type without having to create a generator first. Additionally, all models have a `stream` method that can be invoked directly by the user.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import generate, models
+
+class Character(BaseModel):
+		name: str
+
+model = models.openai("gpt-4o")
+generator = generate.json(model, Character)
+result = generator("Create a character")
+
+# v1
+from openai import OpenAI
+from pydantic import BaseModel
+from outlines import from_openai
+
+class Character(BaseModel):
+	name: str
+
+model = from_openai(OpenAI(), "gpt-4o")
+result = model("Create a character", Character)
+```
+
+- The interface of the `__init__` method of the `OpenAI` model class has been modified. While it previously accepted a client and an `OpenAIConfig` object instance, it now accepts a client and a model name. The inference arguments from the config object should now be specified when calling the model to more closely align with the OpenAI Python library's functionality. If you provide an `OpenAIConfig` instance when initializing the model, a deprecation warning will appear and your model will behave like a v0 model.
+We recommend using the `from_openai` function instead of initializing models directly.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models/openai)
+
+
+```python
+# v0
+from outlines.models.openai import OpenAI, OpenAIConfig
+from openai import OpenAI as OpenAIClient
+
+model = OpenAI(
+	OpenAIClient(),
+	OpenAIConfig(model="gpt-4o", stop=["."])
+)
+
+# v1
+import outlines
+from openai import OpenAI
+
+model = outlines.from_openai(OpenAIClient(), "gpt-4o")
+```
+
+- The return type of text generation is now consistently a string (or list/lists of strings for multiple samples or batching). In v0, Outlines automatically cast the inference result into the type provided by the user for constrained generation, but we have removed this behavior. This change was made to create more consistent behavior and to give users more freedom in deciding how to handle the generation result.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+```python
+# v0
+from pydantic import BaseModel
+from outlines import generate, models
+
+class Character(BaseModel):
+	name: str
+
+model = models.openai("gpt-4o")
+generator = generate.json(model, Character)
+result = generator("Create a character")
+print(result) # name='James'
+
+# v1
+import openai
+from pydantic import BaseModel
+from outlines import from_openai
+
+class Character(BaseModel):
+		name: str
+
+model = from_openai(OpenAI())
+result = model("Create a character", Character)
+print(result) # {"name": "James"}
+print(Character.model_validate_json(result)) # name='James'
+```
+
+- While Outlines was trying to standardize inference argument names across models in v0, we decided to stop doing so and to directly pass on the inference arguments provided by the user to the inference engine/client. Our objective is to let the user use all arguments they are accustomed to with their inference library instead of having to learn Outlines-defined arguments. The deprecation of the `samplers` mentioned above is a part of this change of approach.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+
+```python
+# v0
+from outlines import generate, models
+
+model = models.transformers("microsoft/Phi-3-mini-4k-instruct")
+generator = generate.text(model)
+result = generator("Create a character", max_tokens=256, stop_at=".")
+
+# v1
+from outlines import from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+	AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+	AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+result = model("Create a character", max_new_tokens=256, stop_strings=".")
+```
+
+### Added features
+
+- There are 8 additional models available. All of them are loaded with an associated `from_` function that accepts an inference engine/client instance.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+    - `Dottxt`
+    - `Anthopic`
+    - `Gemini`
+    - `Ollama`
+    - `SGLang`
+    - `TGI`
+    - `TransformersMultiModel`
+    - `VLLM`
+- Some server-based models now have an async version. To create an async model, just provide an async client instance when using the loader function. The async models are the following.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/models)
+    - `AsyncSGLang`
+    - `AsyncTGI`
+    - `AsyncVLLM`
+
+```python
+import outlines
+from huggingface_hub import AsyncInferenceClient
+
+async_model = outlines.from_tgi(AsyncInferenceClient("http://localhost:11434"))
+```
+
+- As explained previously, the `Generator` constructor has been added. It accepts a model and an output type as arguments and returns a generator object that can be used to generate text by providing a prompt and inference arguments. The interest of a generator is that it's reusable such that the user does not have to specify the output type they want each time and the output type compilation (when applicable) happens only once.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/generator)
+
+```python
+# direct model calling
+from typing import Literal
+from outlines import from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+		AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
+		AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
+)
+result = model("Pizza or burger", Literal["pizza", "burger"])
+
+# using a generator
+from outlines import Generator, from_transformers
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+		AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+		AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+generator = Generator(model, Literal["pizza", "burger"])
+result = generator("Pizza or burger")
+```
+
+- As explained previously, the `Application` class has been added. An `Application` is initialized with a prompt template and an output type. The application object returned can then be called with a model, a dictionary containing values for the template variables and inference arguments. The objective of this object is to let users easily switch from a model to another for a given set of prompt and output type.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/utility/application)
+
+```python
+from pydantic import BaseModel
+from outlines import Application, Template
+
+class Character(BaseModel):
+	name: str
+
+template = Template.from_string("Create a {{ gender }} character.")
+app = Application(template, Character)
+response = app(model, {"gender": "female"})
+```
+
+- The regex DSL and the associated `Term` classes and functions have been added. Terms (`Regex`, `String`…) can be used as output types to generate text with models or generators (they are turned into a regex). The term functions (`either`, `optional`, `at_least`…) are useful to build more complex regex patterns by combining terms. On top of the objects related to regex patterns, there are also 3 terms that are intended to be used by themselves as output types: `JsonSchema`, `CFG` and `FSM`.
+[Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/ouput_types)
+
+```python
+# term used directly as an output type
+from outlines import from_transformers
+from outlines.types import JsonSchema
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+		AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
+		AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
+)
+json_schema = '{"type": "object", "properties": {"answer": {"type": "number"}}}'
+result = model("What's 2 + 2? Respond in a json", JsonSchema(json_schema))
+
+# creating a complex regex pattern
+from outlines import from_transformers
+from outlines.types import at_least, either, integer, optional
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = from_transformers(
+	AutoModelForCausalLM.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct"),
+	AutoTokenizer.from_pretrained("fmicrosoft/Phi-3-mini-4k-instruct")
+)
+regex_term = "I have " + integer + either("dog", "cat") + optional("s")
+result = model("How many pets do you have", regex_term)
+```

From fc7ea9b7cab1882ff20192d4294dede46e0d4b13 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 6 Jun 2025 18:00:15 +0200
Subject: [PATCH 346/505] Restore deprecated regex DSL functions with a warning

---
 outlines/types/__init__.py |   3 ++
 outlines/types/dsl.py      | 104 +++++++++++++++++++++++++++++++++++++
 tests/types/test_dsl.py    |  46 ++++++++++++++++
 3 files changed, 153 insertions(+)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 8673b4e47..85af9adc0 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -27,6 +27,9 @@
     between,
     zero_or_more,
     one_or_more,
+    # deprecated
+    repeat,
+    times,
 )
 
 
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 14b0c8c60..1a0a0ca1c 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -15,6 +15,7 @@
 import json
 import re
 import sys
+import warnings
 from dataclasses import dataclass
 from enum import EnumMeta
 from types import FunctionType
@@ -200,6 +201,16 @@ def one_or_more(self) -> "KleenePlus":
     def zero_or_more(self) -> "KleeneStar":
         return zero_or_more(self)
 
+    # deprecated
+    def times(self, count: int) -> "QuantifyExact":
+        return times(self, count)
+
+    # deprecated
+    def repeat(self, min_count: int, max_count: int) -> Union[
+        "QuantifyMinimum", "QuantifyMaximum", "QuantifyBetween"
+    ]:
+        return repeat(self, min_count, max_count)
+
 
 @dataclass
 class String(Term):
@@ -785,6 +796,99 @@ def _handle_dict(args: tuple, recursion_depth: int) -> Sequence:
     )
 
 
+# deprecated
+def repeat(term: Term, min_count: int, max_count: int) -> Union[
+    QuantifyMinimum, QuantifyMaximum, QuantifyBetween
+]:
+    if min_count is None and max_count is None:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `between`, `at_least` or `at_most` instead.
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        raise ValueError(
+            "You must provide a value for at least `min_count` or `max_count`"
+        )
+    if max_count is None:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `at_least` instead.
+            For instance:
+            ```python
+            from outlines.types import Regex, at_least
+            digit = Regex(r'[0-9]')
+            # 2 ways of doing the same thing
+            at_least_5_digits = digit.at_least(5)
+            at_least_5_digits = at_least(digit, 5)
+            ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return at_least(min_count, term)
+    if min_count is None:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `at_most` instead.
+            For instance:
+            ```python
+            from outlines.types import Regex, at_most
+            digit = Regex(r'[0-9]')
+            # 2 ways of doing the same thing
+            at_most_5_digits = digit.at_most(5)
+            at_most_5_digits = at_most(digit, 5)
+            ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return at_most(max_count, term)
+    else:
+        warnings.warn("""
+            The `repeat` function/method is deprecated starting from v0.2.2.
+            Do not use it. Support for it will be removed in v1.5.0.
+            Use `between` instead.
+            For instance:
+            ```python
+            from outlines.types import Regex, between
+            digit = Regex(r'[0-9]')
+            # 2 ways of doing the same thing
+            five_to_eight_digits = digit.between(5, 8)
+            five_to_eight_digits = between(digit, 5, 8)
+            ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
+        return between(min_count, max_count, term)
+
+
+# deprecated
+def times(term: Term, count: int) -> QuantifyExact:
+    warnings.warn("""
+        The `times` function/method is deprecated starting from v0.2.2.
+        Do not use it. Support for it will be removed in v1.5.0.
+        Use `exactly` instead.
+        For instance:
+        ```python
+        from outlines.types import Regex, exactly
+        digit = Regex(r'[0-9]')
+        # 2 ways of doing the same thing
+        five_digits = digit.exactly(5)
+        five_digits = exactly(digit, 5)
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    return exactly(count, term)
+
+
 def to_regex(term: Term) -> str:
     """Convert a term to a regular expression.
 
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index d5d53997b..4e13b1d38 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -52,6 +52,9 @@
     exactly,
     regex,
     python_types_to_terms,
+    # deprecated
+    repeat,
+    times,
 )
 
 if sys.version_info >= (3, 12):
@@ -334,6 +337,49 @@ def test_dsl_aliases():
     assert isinstance(test, JsonSchema)
 
 
+def test_dsl_repeat():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `repeat` function/method is deprecated",
+    ):
+        a = String("a")
+
+        # error missing min_count or max_count
+        with pytest.raises(
+            ValueError,
+            match="You must provide a value for at least `min_count` or "
+            "`max_count`",
+        ):
+            repeat(a, None, None)
+
+        # at_least
+        assert repeat(a, 2, None) == at_least(2, a)
+        assert repeat("a", 2, None) == at_least(2, "a")
+        assert a.repeat(2, None) == a.at_least(2)
+
+        # at_most
+        assert repeat(a, None, 2) == at_most(2, a)
+        assert repeat("a", None, 2) == at_most(2, "a")
+        assert a.repeat(None, 2) == a.at_most(2)
+
+        # between
+        assert repeat(a, 1, 2) == between(1, 2, a)
+        assert repeat("a", 1, 2) == between(1, 2, "a")
+        assert a.repeat(1, 2) == a.between(1, 2)
+
+
+def test_dsl_times():
+    with pytest.warns(
+        DeprecationWarning,
+        match="The `times` function/method is deprecated",
+    ):
+        a = String("a")
+
+        assert times(a, 2) == exactly(2, a)
+        assert times("a", 2) == exactly(2, "a")
+        assert a.times(2) == a.exactly(2)
+
+
 def test_dsl_term_pydantic_simple():
     a = String("a")
 

From 74c187c8360d2e0c60c33d56555478b37207953b Mon Sep 17 00:00:00 2001
From: cpfiffer <cameron@pfiffer.org>
Date: Fri, 13 Jun 2025 08:41:02 -0700
Subject: [PATCH 347/505] Correct the docs logo aspect ratio

---
 docs/assets/images/logo-square.png | Bin 3323 -> 13864 bytes
 docs/assets/images/logo-square.svg | 192 +++++++++++++++--------------
 mkdocs.yml                         |   2 +-
 3 files changed, 100 insertions(+), 94 deletions(-)

diff --git a/docs/assets/images/logo-square.png b/docs/assets/images/logo-square.png
index 9bbb39243415b826a71ba65121b20ada5a7b9503..ebf8cb8cb9e1d7d59f427a2ddda99de5d12e5ba7 100644
GIT binary patch
literal 13864
zcmeHtbySqy+CE5#AT3D8&?22fcS$z_!Y~ZoGc+PCDGky}!w^!^B?uxRjdX)FGSUdY
z(Kn9oI_Ep*_pbGQ|DCDz?7i>%+WWrtb?^N=Yfprxx*{$%1vUx_3a+w}oHhyyYTS=6
z7ACSLpj~hP1%>jXkFFt18{|pn;^u5+>j0sHdAmU9AYQgsC@5Yt6B$MxRLuzvH$0IP
zGOuoj1$&gIcT)O0%F1ao%5R6C&gHPE4CpE_E*<vF%>{Dr+<ZR}yy?nGFyW=&NU&46
z|4asVLSP|j=Xz|&**6q%c~w&Bd9u2>u@|&Df!IF4VVbWzIX*{(PQe{pq7E=0tTqly
zJ}F-IcV8cq`Xu*+2)HniD)&ihVgCLupHy?Id%@$QOQvt`?GlUa6CILyX0}aUm^+_R
zCBI8r)(SBoyf#znuUG0%V^N;m<bI72iddXvxOaea@clewF_w{aja|I;yYz_Fqw&La
zHQF+_r{MA_QN;A)PVxFP4Hb3gN;7{ahpU+fipK|5c<X~|5`Ni1&9|E*Y|U;?vLQJF
zHIbx5rRtw@pF9E{G~h$g)lBc^7O*3@c;zG{lXHR$@J&EZR=2P28kcvO%wmc<*34EQ
zM0TaRmr`8Qgobc8KRthMZ7|>AGkN>_=e_Dme+#wEULN1WeKk_w!tyC@>$mR)#@ygF
zcdgDp#y%U+Xn#{L(>Vig*}`tjOspo9O&N1`>H>I3s>CRER_1(qW@!+Mkyl$rP{R6l
zQTCh7Gx|Wuu#-75?^f{e#`)cV$d@93=wm|^xDl;mwSfnxaTEXt(=L49&xIQ2GugbH
z(*1C92c68q4fg45V^f%)!x<HK7g7A#$IpcqVO28WXBv1Iw00xC1YlL3ULt#kzM?n_
z6*iVYi++jQiC}f$W`e!b`&3N}Wu{bZJKG68e4CzhyNZIz849SNf|FmftE~OJ`=V@Y
zVNtg4R$YDIWi<DS5<@KSw#}HDWM^!eRW@XDI>oeM6q~)G_UHk;;;n1l-XNanel3Et
zdH=+!uD7iH`&EbCNmaUN%4p2_pq*bj!XArVIHH;y6UuD(FdK4<YNpT1po(7DNL{eG
zl;J_H><88iHTEb_=bN&q*)|93IY%v{eY5GowZl%QeaX7a_$_~%#>Ksp<cY%3VPVqd
zJWZ8Wv60?my>xc9zJ34s?|$P=S<DSu!?v{Iv4eYjy~PHz7W-2$TNcHD0&2Uoni*<Y
z&th!%W|~mDDJKsmo$pqr^gI1t(qd$k6huQNJ$6yb7CY?no^>-gOMM2FaikY!Pj$V}
z^3=!lzlCqYqa<eE?3BeYiAY*+mwo=U;PcXe*8VV*cQmpu_a(fH26Zfq6I{kz>%TPy
zntkDD=2|hW(5>7Qno(vct8jD=+{M(Nt&{*?<7y4n<jY&DD+ftXJ;XcMX7$hON>)S(
zdbr^T<*L`pIJkayzq)t4%d_mh+iaRwUCnxK5_OUD7(TbH+>$v9cS}Q@beg7pl3MU1
zYQ3PXNX8}m=q&AU2H^so@wTUY6ZF0kdE*Us^HU!7gP}5~hFQ*`N@cLjM%JN}RN5ik
zE;}+lV2AbNQ7tp3^YG^h@AV&+?@APA_Q#T^euAu2#DpHiIoOJtG_Ng0X7WvByLw$Q
zIzr=%FkK@rjJsGHrv!<}_?QJv_Sd?#vt|R(c?MhIRpWB^yxNUpyJuC=Lhe2?GfgSB
zINN_Jir4tAHX{6SAS+j%?n*sNAtiD96d)oVJ|k?`%X3$YGhYd`%l*9Gbx6!_yP%;2
z)mKWd?z|dWQ8ge?5z@HcM$BQ{oYGaEvi#26ozW>XUtP|uZ3aVERjjD)OF&>PINN`H
zt<ifU8u5BepR}!3ks+1~Y|u^E`SOm+@e2y-AfTswCF?hTU59BbL(jp84N5Wr4PF_a
z@h@njv!JVV;TxrC^{GcTa8Iuuf*l)^BaleA+M}9#E!kqGn_u9!cT=!TwVqSIb5aN+
zH+-uwrG$mS{w*bf&IM|gXY5*{-B$@ep>Vm{?hPFb3~iAU-ngr3U-rJD3o!wa{WyeX
zy;s~wa5J}`ddq}(6_Y+%>Wb@LiFt2mO|1LFcC~K$8HY0i_?g-P(>l*#A}&@v%EG#r
zyO4Z}$V3kcIde*^V6&ho=_-_Nj(~+*K(ma<H4Z-RYr00Du+<``>ycb~MmIp52a~#t
z^oG48Cy0Q0w3w$i0yy6~1kN-*d6fC|t~r}R7)HVpmedj%r(5VST4yE3p?q^Oo=jU)
z%sFeb`%zDuA;#FI_LNddMfhXtbIe3e*S^6f5OwP(N5s1r-fcEl+pML-bY=*^|C=&z
zHxw)GZktsx2Ymt?l{S;oo@#Hz4!IyEK+>i$)c~edv8=fD82v1;@|6rA@iW=Sm?b_5
z6w7?843>1jzRuU^2td($pK0T`@>J9$6HG3=xXDLAv=N?hx)vN;E`5BIgc1fqHL-US
zmOG6m@ro<z-JY(TLoT?C017G;e^+UHBd_GaHC!B<_>RLHEozk(vKI&ZfJjv>rQueY
z)Iet`Jox*_JE_h#+@2ql7){YuikBm!CM^Bn)aGn_3FZX&1-vEJbP_S7=<n4Ay<;An
zNwjS=DqC?Li1jI}aJA+vp9lj(mRmVHh$w(wOVfCXsAOl+^RL*PQdmj?Y^ykxckL(@
zTRx=a5ugFV_iqPT0GCBzU~{oVLoVI~$kA>_=3W0q0iyLgG-mHyOy&c~HDiu@J`B5l
z9;9f#Rc)lX{C1f(D1jqFU{Ijt`qSfpfqq&xC0|P7ZG;qF+N|umD-M(n1H2sqgyPZU
z+Csk0?kGuOTJ0^elZY3c5~Gp?ckzQ}3nUEHBpgtqHxTp90*bEdGFs$AA-OxnnmzL)
zy(Mv0_%sidO0yGYLvCZ6#4#lWmU7;!mSL<+MqA4mXVjKmvGf}Ct10@71~s>Mh;l;~
zQi3soQB8)mLE!Evef;LJa8KENjF-BMsq~Pc@EgMoGT8%lX`qK3Ost&X`!ITb`!I&n
z=QlN-{MjbY*%@>d>+Mm(Jyc<nJ{~W<E2D*T>6ZG_-f~~~O=_Cb&e@VG>z4(8d6O|f
zje^-=uq8$G@i_?^@i0qE>Fem)ueXD!;MzM6A9isM2*eaKJP|>O>cykGW8d=qWk@5%
zmk=SVjC%W|a@=d;_q3JF6iQ`;{R?H?FIlevF+sFsY)#k|c|GECXywu0ZAKF|&db2H
zt9s`Aa~NTs7_x5+4e9&SITNo=siI#E$kp!VUo}bH7xOS97Y5u!(s7SStR<%@4esy`
zXgc0@mk%IeO25Nfhhl2x1ib1yCOg105N1$eG7ZelRit!yU0A0`*PPT)zVeZ_vJYw-
zDRW(WcLX#WLxn@K=R)d-9oAqS{y?oj7QssBCSR2pX~1ZJogo^xbH~!g7<Ttopph+c
zjLkho45?HLf7zZ?;9~~9Z-=jNb1mdgLpWy`H3OLEKnqDUY*#_DbfgcIvgG&_F|WSM
zm}8NL%U#Gie>Kmr)L&f@UFw4Q(8fPc=q4{E$Ic`6>L|+B+MZ5~Am&UvxqqJ|anr4E
z?rZbsxp4oL0%s+Fjo&9yc6yhEHiC2KCxZwDom4*!YA4lfHmbf2hUw=1PwyGXQ-ba2
z2)DQ_%!NgiT6fz>(0OD=8b5on15%<}Z(eaDq``}<yJ{?E@LdyrTK*Fw5aF#iC>l)i
znrJ@N!R;t$cA*3_ngAG3{-JYjmau9Fam2YHi41OK7oTZB)mgN&wz|A|x!|kYHm};E
z*7X*(BD5Bd9a#?_(#hLmc$t_FMZAqt$xLFR+8<w+<4H=SFaD53@|{ZRzL(nJLk?$>
z)qt<7V;>*kyi*H=>o%fot)Rn;H%&7y>b1T*gVQ~O7r-f^W39A0UM>o)y;p8>h~$?!
z`T4pA3z?|j!qVS`f4ntFG#z_K;URi0Us9L>=IYZxYSf1q1)<mnfez*)BPMuBI!QQz
z=t3ownm{pvw>i)gwuJg(pY${a7@0|Bk+b~Q%_VTsu=>Ntav9YUU2npd_jF$KM{LVW
z@!b!ls4(vP*bt?x1L+DE9JI$PCLBY*<s|xLf-Jb3?0ZOO`Eto>_2~nPL}5nfV#ytn
z4i#rzf4Iy0%Z@9;#~95_rmq{)*ON$#Pt&7D(ao|v7W^YUX(Y>!#$>csQ>jRffFBI*
z83GG-lw9K|t<+u$@Jf&;m|$_^H*UYj_-v8Ghu7f349PKcyLYzdg|ir1SZ;jWBrDXy
z`K$r!D<YCC6BHuztt<O#UOZ4Zo_h(Tegj=%k9N3!T2gm3Y0-yh#OH)y&1dOlMCo+q
zjlCvX-Ae9oRJS-=IfA;%hVkl9h6I;)$hm}JWCV<L7*V78Jj9CHD9WX!#ZJ$(eB{Pt
z8(Fp?ay+HkYYWE}Pi#^ToDnLS?KU0%U`k1G52v)hgX7Ft;q(PjY4NMqSnJePx1u8G
z5j_ng(8r7j3=GNqbpU;YR)@|;mhBX7o8<(t@Nmj9YJ|(Yw3<bK2iCo(w&8I7?JZ(E
zA)Jr-^J)~r!Ot34#bb$Q2nxmunl6^(Je}H$kCzB^WiOpV0+R<wU#_WP@esW3Ae0$?
z*L31?cncbZ@2&IVgU}t$xcFvTRQNUn6>L<9t>d^g_jL|V5Sc0O46aYb+Y(P4EThrH
z2LlWly(gq<BoW`F1M4qd91Y}j`zM{=#+??LPv=UaY+k^H@$V+<E#zQyHN0Jds5qDo
z*CG<`4&gIYVD~X;1O~Onw>q>H2e2)^&Z}tVXEjbV&=4>b7O^}fTE5I@fpV>m=%m*B
z8oK47guW*7=-`^8wM3_={=(u;zqBh}C&H+^cK7Vb$Qc&5@P}0K-Cb=s7QTKCGu;v_
zj<2U2{<Xj`@lCo3TSr+IFKAg-_*&3hATN<D9+l;GwjarW?*n0Y7z5c><p}dik`hxA
zn<Z->A3?5FqUD1w4>ZF-hPCA7ajNm>#9>njpD<peH*crtF?Bm(YY0quJlHPbZ@b$N
z8g^@NgeHyVz_2)uN{oICSf+fel7`bT0XKFhwV;@MjS8S`bCFI|W=E|HKC>jO(j|KP
zpy(ww@VT?5gwBv5H(m1+*5KDfc0_=B1KLn*{GBHsl=l?aFwnK6?cgJ&2UzCXWLmc$
zTC$826vv~|b1^dHyaJ5R^-WkB?>v1FTl^XoSY%aDtjwaxpp2E?PC7Ksz=X-Iri?mB
z&UaZYR#t$gF9Z#hV=8&{Il=%u)hA^;MS{IwsIO_?vG__(Wi7Lb;SQ^SfON1=Ud&<O
z)==D%e7;FzJS?clk#FDO$g#r9^on^b-+-4#@6GdE(#k3S>RWeF-bcgd`{eUwKlqRK
z9-ouU7mN<{$vj!W;OJyUq?AoEr0IdMQ=6F(2qvTQSj!VmD3wEquS+}j>1#oj2(7c?
zg0bSAu$du)p3(9iY9IX~>)Ir{fR7=X9a)bB9nh#lzrbheL&Q0WU7Z%g3EHyhL_K*M
zSUvC<`99qFIzKAi_b4Na45!rgWRCa&6E=>yPGJ>)cz(JLI#>^++A4~7h&nVJfxxT2
zrM9rWf}P1(>JVs;Ykll}d(ODUYt<qjrXN%EwvJmH(vytOyPq0YAFe0`mA&F%c=>?s
z*_{%*B{hW`qwdrIMKNgq?LIfd_it;*8wx2@UyBIQDaWay_sNG+U6Bvj^tGdYxFz>;
zpc%CWK64)g%4*SgA+w$+a>Hn`#~O%I(h(4p&T<EiLNlZEZV}6=G_6$IF8@p9*AU%<
z1+yhKd$Z&O(uMmIn*-eOTAE4@(v-vtJsTM_qr+_yY^7*=(nphp{m~sJN&QQbRD22q
ziGJu6rT3fjq9p8$RYkGtP||iDk9kj`6wEjYat3{wB=3eFhA4B0mmIurIAl*=$m+y#
zsKyPaCF8a}mz}q2E8USudb)q7|M|+(-opV6wgu0-pRSTto?3aOzcNV-RC>rO*NwmA
z)oB6XIyLw6GAc2b6w7_JL9>1zb!k8^-gxwohoHqh-!4T=FIh^zl=?VH;5^CM9<Fy3
zh4Iu?tM&H!r8usDvLTquhLI{RC@ypRCOB}gIbQEor$Pm7^u@)3%LYF5+C~>)@c1lB
zYw+~lwW|K=P1gGb1%dWE>~B6|oG6K>tZGh{<2Zz4R4hNl`@(e{6!?VNdJ0vE05`WF
zWUKB{f#O)DEc9L6JCAEt_w-MVHFYT<cLc=_!D;(o)&_nbB5Lm3)MvM|rC_)cvE$hY
zjlUCkA09K0!BuNFIf(ftf$Ubkb7-xCg*+b`>Aj^gf;q<erb2N?m)g#0b5`}w0@md_
zFv@7sh{k=({O|qB^n=gZF2r;kW?2_f5BQa*gK51+5rU?n2~7p>sWB^SP@2*lUNa{~
z^@2aFUT1vLqMw;7Om_c-tqT$|WV;jOZSEb6uiH+`)J|NrKv^04<r4{FYcyWsin)cN
z&Q)`{n7#Dmyv@D^AmMjnA3gK+*d_YdYp61cwgMz<BD05C5YZs;unnivl(m3nplyZi
z;)G4{1*4`i6UnQz6;?k4u(pHtowR^j>jq2^MPIGpbW~Kkv3d5|=yW@+YJ3s)%yh$j
zS?jv|(pf;V8-ujb5?5GS2Y#{Zc?#N?R4_d?`Qom*+!02{ttEJeXF2hV#*)*9lHo0f
zb@#L0=o8s(0ZHeQ3zp-iZ`t@6H<<*_qHPg5#4E8pjuM!4>b{u`Z;RurWn4Lub7-Er
zOSg8GefRnRa95vD2o$U`6;KZ;V9iIA%ZCc5S8NAlefi=U!|}FQ^T-ad5PEURzA?pQ
z_2SVbOoB^mpqfJ8>Bu}am4UnGYrMax0TIglJ^j|r)#>jd*T-9<JJ`&LsKgZ^&r1lU
zh?cTgA`ZXCaP*5+aj;FP^~ADSZW|PBG~3gz3E2fTR1Ckm5RD-@(C}(|?<w9C5&lm6
z^vU2>jbDqEgh_sJ>e*8`4OX?jZa_%+o`_UkCd?<ucULr=i9=5P<AMnJxAz5`9|y4g
zScgqN7aPEg15OVd%oivF>mE%t$=^J+4(6)xd(#QT{mvUqwcJX-uVmmTj);KK$Ebs(
z(3iKD2FJZgQZqhc=&V1*cAHBK$;112=g7~R$}w^3sF$t{Jm=S6DT#hAS>%LR<-TE#
z8?02i{-k}n4w2##9G%(TG$hW<R_P044uq<wp1jyI)OUO-n;Y!$*l}#%Bh+rziOoBl
zWt%opRBtu$)@F^X_5@fv!rPNku_5v7>3L%@Q6ts!B{!${ABJ@1<1?JVn#|AJr)lw=
z+i1+DQw9X&fs(P`>nu0?Ca%ZkZny;(1#T3$4tfQjv7AZvJ6@|FK1-Rs%35mgp$VWN
zaRcPv0`{m;SC!gRK&~F!<!bjDC)f1T6&9m?VjEfrK!LZ<ZduM?;b3QF;*@?sxH|)B
zi}yr~J+_B@QVWjR9_^{sWmCeizRl%Q&cI{W7av$5%%<<5atsJ%0G^elE`HqdB&j(s
zw``MZPH>*l%d<i~>3JnsGPw1jx{hR{>!qU)wP>>vW}Dea2C4XRWw2^T=B)!;pzF?<
zTV&BX{~oG~yTiC^LfN)32lbm-U3E-IOxs(H@Dd%FWJZ2<q3sFSXrl?!yEOGNwtQ>P
zHK*`J)1v*D8TD~n??TZn?|RV-XD#ufS6$>qjvhkUPCV`RM#Z#M6xa!A?^JSOh*93D
z4^Z`%OQL+Fv?pUTCEMD(FBy4yd)a>~_dJmTxkZ?X@ZaQjw~vE(?6kc_vtrfN9=IpT
z;(|q}8|onU;JDs)OD#UE(%syt3-%+*{KpbSERr)e=FinBszvC672BzEtFIkI=Tbkp
zzJvx3J;*X{pl{hWOE$WR8LHan`|<0<vsj=xVPmMh8DX^W)}z!3h4r#n#8M&r3EN{S
zo-f5tLOQIueDC+a5jG)g8`e^?>GOoE@XYT~%~-dqnWF90n<94WzxrmXS9_a|6n^$Q
zi65GfVDEo^QZ7v{VPurQ(aJtuJH#%0tF+_uOZn{Y^P20v<oiZ)E!u;d!`WI~r*LwA
zk-&4OupD)B|E7=w=Cd19#V5vk=N{VZEKGcd!HHJT7=D>kL3y)7?IVu73(>v`qDq#&
zF!z{-wIY!+Tw%Enz>$kd1h3>Gm)*_;!bc->upRG`WqVLksP+1VaK_@hY@_rd%dhB$
zw;!N-vf>>RJjk(o-EBUBAE;t^-^ZhGmZIrSfxEaWl)F77K~{UiSw~Ter1%ps2a=Vg
zuTj>Jz0z6<y}Mve-02zY!F<+ia4)W&Y$dBaY%FcV@1TU0nkG)XC|TWOY<iP2q3%;d
zozB>&Duy`mP4m>k&q>hM4nfQL1Bf>j;z<ahymz^53}Y6jNOi*ZgrznGBJzSHJ&HLl
zEgI=0YiH%?$l(LhD(31q!RFA{T8A8MhgpJyX2mtTAyoNeTuJB5q29!+KoQ3qx@)0G
z2OqlLX!#_cN<~_>*V>-4ijNuYaW-|h=>X^zjyQ3~R{=hmk${D#cnpYOVhJCz9dx5h
zi__jKQO?V#kLk~PA<wjo^3Wt0P6^$=B~#q$k#s*&`MNr(m-~XLl$uY>BWS?fWgoA~
z)4^nJI1O)jP|9!vU7AzCctyBZ%t<|^D%32`>V_%<NGfL-cOdp9r1viMyR7Eoz&D<c
z{c0pO^*<&@HL*Ko$#Sq`A9A$-LZ9fNY!MvK$lsoWU*y(oXYy8gs4sjJorey~!S<Q5
z$yNvtTfORC+U1DH;awH#33zlXes+gn3<2)C{&EB{Pwxc2m3WjcJ*o48Jd0U_E|Ug~
zT+D$2-OrxSh?;#V@?&CeA0NId;OqPP2xL~gBgS@J_~vvpT|?y36k?~mf+1ekWb)iJ
zeo7{K$~=thCEM-e(xGLYZSq+ZCvj@>$&VyM4{pLiVbic-#Odb?s>d4|MVl*d&9a*o
zd+;@HFroON%&-k8lWc0jU3*?7!^e`Tq&R|dXd1?QiS1sI_4U5^-Di>?CPv(M%MzJR
z;~qUhm8$l6@!o1cNMrl}@~vBbc(qFpAIEd=)l&qwLXJ9r9aOfmta*F-vAAK4@ofd7
zXu4CpFEN)eZl<Wx=aru@pWWE)3;FcSi;C9;>7=aO!U{a|y9lRcydI_}J{FAK6AAZm
z5YkOZd+X@n$F^<bGDt)*ZrE{?G+eCCTeN?CrX?1O-{)U+8I{6N5y>8;*q)6+_obrZ
zxw!t`<{B6Ov#qNG@9GgX@s9x(i*J&=T|!f12Ccz4es134UYg@yW-6ashqH)_6H@iP
zU_G>xv_s2l;uc*xm<aD8a}jCF&BC2~_yrWne}sWBla4jFa0Nsp6UR*LWxwey0`gVd
z+vlLYH!(4cQ_x$`djH6{mUL$ub=l^)ww(qqcsr)FcO#s4XEamzKKV|NNqPbm<du#}
z%`!8H-Dhq&S$S@;mG4<YbPx5n&gDpsn#Wm7g>NW*U`}4v?vuIOpWB1R{90FhJEJa-
zhhgJGp}Ss82OKjnw9qVQyR7`)WS;ld!u}|^ynN?mq8S3v*9(hL*NgIf5=HQiUEzs<
zl|WboiLr*gFc>AME|(z(_UayMBszi7i8B1nhp(7-8o1lcukm%G5TX(f2_}$7d?B{7
zvYN`WvVWZFA<yzMpT>(TbxBdZH`A#Da8U2kjnbRDSJORMDb{0^3Dsn;-8OgIdOn;_
zuk7l2&jbh0|Dis$Nv6BI3kiBBK7nVa@`$|3m6^5;;ZCQk<ifn~Tq|0LPnxa%6xpk-
zw`uCk-40IjcUy9A(}(bc6c0QE>nCV`5pLUGUfnUxw04&`-NY*KjzX1)_X;B~RBKTe
ztFxudJdI!|V}y5acIPph#+}9QMD{voMsebZ@mW}nLc0kqsy_FqP}U@r5NZ(h>Z%p_
z(eG4!^p2|0uc+P-KgS97e(D{u!=I;Qff^M@CnHTl$=aomsh@QZ$96^oc8K5PA7O&s
zDA@6VCl;IkLj~;>zQ3nE33E?tzWJ>${;rv^Ai6g={B|!8Pj`i5h&yiXnrv}AE5K8I
zmhi;hwu*tS1=qx+o*#nWFo%kneRn#^v`<5a^<Ka!-$fzS@H3om<vy(fr>++$mc{UK
zTo)It=zDCw<z3Cbv%Z(7sx|GjwzO_}I<In|$kSH}TjVLLp_-})*x8X2Wa(@H;q-EJ
zL7u{*pomF&xq!g-5Ez{W#M;(L9B|Ow4xqEO6bBdxsBx>g$U<yvm3-VFIzH;UU>|!h
z&=Mdifi31Gf&_4cz(90fjt)*x5ifDTPh1h?^^a;U0Nqa$n7uf_P)(Cg*4Yh0$Ir>n
z$;~0}W$VEUkie!BbF;J((Uw#A9RgVr2iU-1E+Sl9o}QkZo_w6nZq{5pKp>Edo0p51
zmjh|R0rhr*fxI}JpbS4Ce#4N1K*4UdE-+hXC%PY)APZ-Am^c7{Y^VEUe2y+^YJb8z
zL4Rif$p@Dg$c2lClbg%Yk?U6vC`{f13G(}Z{-p<07kOCCr44~PySsrQ@*WT;7{jj+
zmf%1AUEJLqeuiTS=7Kmt9Fe9_WUoAb=~7WyP4iEW9~4;II=cMyLX!O#OPH<IKgjya
z*nYJ94CmK@Al?7O{fqS<vHvtiTB)gt$T@@Ee+*AqP8{%~e-TS(u&t%Y&r2{bFSmuI
zkQE1nSBQ^;U(m{e1IW)O#9<*QY-J&AX(4FE5BUX3*$D~*Ie{TRppf94wn!X)D<M8^
zK??y6K^`kh4t^dWghQAg4Cde!6ymcIwtxr<S_=LGq2XqWR3*sa*HQg|vP43GEV%_O
zL6$rmAR%rG4t^mEK@JO`pfCqFzXcz+6>>nlV7{MFmS7PDXE#R>ayo4tLDmp17bokV
z4L=AMk=9fe2k>%o|D#6J0R*!`I*0>QZJpe`{?VXo>j=?-fqt;bBg79B<`o74h4_Sp
z1$cS?VWbangCdpq1CxiFlkYd&AJZa&3<gOo=!Z^`06#sDv53gJK|nBPH(h6E2XVlU
zA<_Nt{3&lbvEL>|$rg&V@cyCr-!-oTas93ITM9VX{%oS7`zc!y5cs!Fpdb&3<<E|g
ze!n$=Z9q=e5M+M;E~r21ZU4n&frWX2Ag~oL2d^bikb_@X0LWnh0`hTytUwR}9&Qkb
z-|Dx(|3rs6Tfsa*ZV+i}WTePwkOKM{4IS%GD%t)z7Ec?<51zPr1vt0`Ie7SVxrIge
z1VxaF;uaF&<_2*69x&IBto|cnF|L2&MC_-*FUtVZ@3%VS@`7Bgxc*$Oe&_55jsJ(Q
z-{<20um>dce>?eG{Qj4&|I+oh82DSp|5ew2>H1p?{4L}Es_TD^F6@6i@Iai9cR`-W
z$C**0(-+7`Axw)$igGA7Ki;oeixQDFcU+W=kk3z&&Og4W0c~(SWaVv`vYPzuElgZ|
zMkXGN_D3ivIBd#t(z;$Vdvj*t1rX|vuZz&3w0Qw{13rGoPsPibBSHE6;LuTQ+=v9g
zc-reAPDa2uD{D+HolG%Kgj6v0U{=k^flP4601-}#3V>(+WS~%>hBI)>$!Tn0C=(%J
zS}QaX^ND}+&}{<|*y`uCa6ZU)a5{gwq9)D3G31`D4CEv~kpLd)CC-QTKAL<u=s^<`
z?umOdV_{W2>-)-ba5-&g3L2?_|NVZ_%Fv15!AY)tM#hEL(ANHPPWHD|kJF{+kwdcs
z<EknpqCtyWpIDk(?Xho27t-CX-nzhL_ne$RW?42+D4+xcpwQ8wU}2%$x`q7wTDoB(
zmrg5~9?D@@2?8A?!Y3zK((N@;%BJHfW97O@LXGv_KWEkxs*t-Xvm9j8YmzMbc0J0I
zd3E|RCtTb8kz~#&$0dXC*R*@CipzE73f(Hp=}siG?V2Yc^(UR-S9`B%cQgH~X2|?r
zeQD9#N>|!ar2Jg|K-Bqbb8>M7hY{4L?)%Yk?X}XfyZHS6$-)}v>BqY0SB6l=8>syW
z@;Cy&)r7QuOM3achwKebxi-|%?x6FX{mwU`vjClMafuALLkAM>wU&b`yWh4kH{PzU
zwFlYv4<6a?Jd9mq<61ZITi6h}c|dD?p4Jac*O{-j_w{$WkXu`+nTPpO9u1<QprfOp
zqDo1j5D@&h{A=lYD&_j+%INMlGlrvzcM#D>o0AT@saZD1IUGh~k7rL@<W%pM_r>fO
z)K^LXcN^8p9GgWIO$@3FRb1@_s!o=MbHXQ?+vdW*jpPc%^cY-BbbCfl<`dV;zu+Q$
zvd~ZX_4J)Uk$m?w6OLPc&e~^v!Qhso1*I5g(kI630CkSX@d)tuy&ZX4M~8~K3YC$D
z%EOs-7Hb2TD#MD`NNnQDhc&+gT;pS!ruUMCFFobED?kUl#mq0vybfN2ne0h7MT&|{
zJRJ3293%hB++PK{5sRuBD=K_<C<t9cD7rdA$7UH)O^pSLA8#J8Uw&8~Rkeno#hQs~
zTvMLtEYDaX=gPVe(-!?14zR#~i(K4qOjlAU?3v@&93fu$%Y3SS<8!=i64GB4^;2Sh
z>I+%QFn4GRt>d3Hn(L@uIUp+d;Otb3$(39N^!(bwKzKXyq(mDrKby|NtTIVu3S4()
z0P+K$J!HP9eBh(qSJWyTY2WWXyH`GWWPnT%e2VvrU$&jinv1I5?%j63GA3KATRGsX
zQ#bSH*7wYm+_iti9+NyOup>H2JO4x=%-_+w)zvnc0-uCa)SBTm+oN6xh<7?)bxv_G
z>cA4aw$b<`>*+o|e(TaST2KF|!)9do*&yqQou@C@_23vA=YLklUv=`oQ2I@Ne<<;v
zN`i|0SUtVV(_07ngsJIrYB`hIbM3y@OQu~5)~+Wjr}-vN+bq>yoE@*VEl@Mrr#Pki
z80UYhOAugNm)zgy^m}JZKw!tLXp#wNf6r)liR0tI_&RbsGjnAR0w1kfiAmp@i0?n?
z8*%wWUi2KmfbwTTprZat!q3vu&1N${IDeYVuevn7Sv5DiwA0-6-sPIUSUTqUQ+fM~
zl2(4c<Nj(|R>@=vty&EV{wHok=phs$&Ihj-Gj8o29B0RdZz892NOe$7#o)eUl7;)J
z-vKG{VpGezTmP<SbK*17Niw7K9X|%QmJ!Bnwt}m!c?8bsUEiFpnz4g2fP`K(^X9e`
z|LC#2MvmT1^!w?p=aE)%aJ0RrI63YV=pjvgTSwt7y>xb$m*>d;PT3W)UaC0dE2!>^
z8^XY5={qK=Zk+R*0(kGv4*&i12dFI>8-f`Xj~kOljhMAG^zs_HBcEOjeuG;{N|;nL
zzc(*AYr!2&WJ}&l)bZ(5M$&In-F(=<=F<87c1~3IW|;p1jj$IZ;3WNQa{WIf(JxE(
zKPA}TX_Jbni#GS8%Fbxx*lZYPEQTkoG|nS#AMoBmeQo9Tf)a2<llZF5ntv}A3zeMs
z2?+I%O#_+He<t<M4w0odcl>bdqw|y6_13R+5?{U5C}T@em6j@gJ&$O;R7EA5MhhhT
zbJzJ-0RImBZ?%mZ{78-!lg%E57YNm$rc!OHw9JF2i;U<Y>B_+QbcO?-J9d5wr-~r~
z(RV7}{Jzxx%(Y+ChP*_6KBKpV3aGpJvucZh3C7Q&X?W35Ux&?H`go?_;(qWSDEV8X
z{$nQpiC84cxZ%%;cE?4vRjkv{-IJmIu0PbHjD;nTnCoi2*3!#Z{FQv3me3%*+w1>F
zs;i9aEP&ofY0Xu>i&}al%y{;}yajegu~#U@FC+MME&r!`g<quh_uDkV91LW%k7`m?
z3%=yo5smA;m~?CQHSa`D(oD&IfBeZyGy#dcONJE{C>j2L;Q6QY%GSCFzm@?t5qyyW
znnHU{2bel23h&t;{<??%(>nUgO8q~!eYSfH{#Mgc8z^9{AoswC+opC%>?%GAo*+=^
z`#tAHA(#AmL#K_HD*{4XX==J)RLy)0&!RkI7$kIs!H0m3>#}e8XMv=I&0+02HDmiU
zhKJU;c~!!mBXw*-G&G(1H#VNW)(#Uu-+#}12eS=%=OHfU_eo+QGeHm$H-+uF*VsLD
z%ZL0KTWQd1?>V}Fs+Iq+a%7mppW)aO5GHGO86r2&6LWTaVIa7x9=RchhexU2QrOUJ
sp3UIJ)ad!`?(na9`kNZKRc<hIAT+s3$>CAR`)Cwpd3Cu88S|k32Lf%CzW@LL

literal 3323
zcmc&%X;4#H8V#<vfo)fVB9L~likU${76E068zM47+wj;43J5eIgguf#5NSc1B~AAz
zvg<H(h8MyzAwh%$Xl1bt1SJ6_K%xyqLPFS*K*EEGT~^OHwk=a-)%>{is_w1(z5AW-
zob%m#;|n+E9b5Ib!eFo+E@w`C34>`IP>;WAZc@(>$KYu7XG_c(-*^}dfq6YNt}w}f
zdU8*KlTU&NCM*GaG42u!i^ZCL6^)7yxfpZF6cZO-JZ-NBgKfX=a_YF}<r2wA(DiX)
zVArfPej>`+eAUmyrKmLb{CB12HfHM?{V9U{^z0`Gzi9{!uN0hf+E4jogvp<;o;har
zjV9&Jt#3=6Dv(xJzd$v3?(?Mh&?eoi>9-4aoIvl|I|$jt8=8GLfDA3NUvJq8MIp^@
zH!sblC}Mi?a_lUtVH6+eVx3oOKZ+Zn4GI@T7{?a*Lu2PKYj1Bu3J8Bzagzypk6<sO
zA>!$<UVZ;1$*4%Eh$PiU%G%~`$P~n!k|TsM5*`B|BHvsIQc=fg9td-b`dD#yDuwnv
zp7z9h0c<_mLSGrkFU<Yfgg=6VAi;Te(^?Auh%FTp=3xe#IlDz?=gGP&yRUA1xbwj0
znHMz7H*CAU&*#|o_I1IUzb-|n5|@(W6UX^vk(}X3GMm6O)g{bY8d-WPo_CQ&W$hI@
z8T=s1!O_GTk(qN?jc-F14uv>+*)wf+>a${62jmYG-F0oDm@c0nE8F_Q03lg4EMPry
zGFZNupY9vDJPWR<GK4u(l3F&Ko)#54CUU}~XmV1sE&7$ipda+0$*%<+I-%@nZ#JX3
z!r}g8$7Xe6k_^jc3qJ8AvcXd|i@0A&na7scLgV@QJ=_XjM!6IWT=v&NPrfwJF4Y@G
zUqUUEfRSbdMlfSYjz+D{iMv^yq9^msmq%&{>Vr;k3p%m>N9-2XiM^&$7ont-@wBcW
zI;fYzmLXlS)wvS+!fb^usNLFFu&Fg|(0<pt-$d&zAaeo(38MiHN;m*@w%6~CoFVX@
zM$HE5PWYF@3k!o<L?Hmp!m2-9YMVB*+;?qq#5QIY)RuvbMAiX61~yjk()o4GZwK1X
zOi?4HhDC)vgb-A$Ksz&R=$Znk4Km=>Fav<q!?y8(z=)Ma9~nxnC3hk-2-8{Q9s)Lk
zsyjV{w6l7mGNx-G_3no<5@J^-yr`SAdXrxu1&JA!*&vP(dZ!VmoVg*h*<MEFJ!V6l
zpx%ywI!Q;m5VP7nDLfsfCTaj*`7AmcSrEP2t()2VJU5qHF-ULkGn0uLC0)^7M8j%F
zIxiS+-YOY;Ryo!aVj}D$L;>7w2yyi8GCdAxYG}v^O%-=1@v_(_Flf|*18#z8<ey4c
zG;_DPN1hZmAj7fvoyRu3MZLGce8b8RX4CySj_E|EG?6Can-^mcmKO6$QgkYAp~T5w
z*-pM?s=vE1Sw4JtHF@gQ@#aHPh$Lzou*qSul!-w@<W#1Y)SHp!h)@+9S%x=;KfvlY
zY9W_r#4iDDVy`4iLWo@?@RpTSIAz{DZ5}C}rwuhde--q&#??Crw;%|n3!7!GR2$PF
zaejM!F}A;qt`;40nW9{(>^RqGpQ6lG#p=|hsg*ok1-eVRd=0mxe?VftmFI_r`O$aX
zc4xGW3s|BjhyYJUtjOW=S5z-;gQKyc418*DrzjA?QNAp24#RQz298ZD>G_2{s=O3;
zn%v*Wk|Im9Ymh%==dc29Y#@U;?h`M_jx?HKu%LWM(6=Oa^f0)JymK^RrnX-s-zQaE
zg0w8>&YjWR@X7m~=S_mc?jTtCT+TMtakX({Y&JvJGawG3q+ckD;7ubetf*WqjG^6}
zO@=dsD(3bbLF45t!5;7Z?I1XaNKlslFq%Gj5OI(21K69xiqf}>3!=$y-OcbTMI;16
zD|h#RKN;=YdFq2c7*kLOS`%xR-DFC7l|oQS8(1CdyJ76uGAJzZbx=atVENv{NuZqU
zi|^`YMb*oS)FL6~uo{?)!*h>}39+FQXZQO60AnPOD=o$CNb2Jur2WUU{Bt3lYmwBW
z%Y9XMzl|~sO2$uxw0`!tA#rggb9L)@{?2zT`JKw&bM*ljI;|W({0Ij>^Hhgg*{_Q2
z9r#kktE4C&<rjMK_ErcNB=W+6&fZB%YV1OZ$$6e_(%+?B$LC$Mi=&el8*~k74i%G4
z6rNm?{O1G2um7tHyz|R{H;9kE^oEDC71AjN^ODHnf-!}wt5s?%Yp+t=F1i2!QduuZ
zkaF;CZa54P;F2AE)~2!MuK}*#D#J;dW-c?V``(wFTZO%+ujvIC9IamwaO_}aY&8}B
z56XK3;e>}-%#mld%()uhT7S~$`lNqG$QE9FI+bQGGxYeh8jNHx8ogVwzaY}s1&RRp
z+6wg`@JA3Fg;#kI-5Hz2|4<l{kT2ps_W@`raZq{{7KSWyKJniE|Iy3)PCyn3DTu49
z$Bn7{P_@N4?Po4h$Te%QY;lZ(|I4{L5af!}MbK!K;!0^lk%Je2DZ&w}lJ{@;k4wrs
zG2hU0dY2bKTOo3%hB>IVJik>j$0R^YZBVz|tN93-R@>wGGn&75`FUU5-@h`gbyv<3
zXIp@lwKqQ|oJ>KzF8EUvAt_lh+wUPYKdkmIup@&p#PGmq*!kQ4l6Ut1kT<}u2HEb6
zjpAyI8>4;vSRELdjD|`*q_7tIkFIa?S};7xTrmVj8m;G1mr(2$RZT}X>n_GY76tji
zi!_p?{n92?+d(0F^pJ%m`Tj*+_){dn&eNies%Bi&&cg<{xeqcP+{!yp>p;Ls*%}~d
zoyT9I4}`;cSrsdI_1O<soLrO8J5xj(nPn0QWr@U(SEQ?n<DFfMqFc$dg4mY~sw*qy
zi#@CJ%N!J}V9T122x-J_X(UG?R{lV}^WzGS$d=2aGm~Tm<%-7>qMs(g`ynC+0WkJx
ooF=m)A%l%qHm_eBt7g+qwHY~_RyAv@A1*K#C%02ICoWw57oI%f82|tP

diff --git a/docs/assets/images/logo-square.svg b/docs/assets/images/logo-square.svg
index efc68a642..2c1b9321e 100644
--- a/docs/assets/images/logo-square.svg
+++ b/docs/assets/images/logo-square.svg
@@ -1,16 +1,16 @@
 <?xml version="1.0" encoding="UTF-8" standalone="no"?>
 <svg
-   width="69.82"
-   height="94.089996"
-   viewBox="0 0 69.82 94.089996"
+   width="100"
+   height="100"
+   viewBox="0 0 100 100"
    fill="none"
    version="1.1"
    id="svg39"
-   sodipodi:docname="Outlines_white dottxt.ai.svg"
+   sodipodi:docname="logo-square.svg"
    inkscape:export-filename="../projects/outlines/docs/assets/images/logo-dark-mode.png"
    inkscape:export-xdpi="137.93103"
    inkscape:export-ydpi="137.93103"
-   inkscape:version="1.4 (e7c3feb1, 2024-10-09)"
+   inkscape:version="1.1.2 (0a00cf5339, 2022-02-04)"
    xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape"
    xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd"
    xmlns="http://www.w3.org/2000/svg"
@@ -24,98 +24,104 @@
      inkscape:pageopacity="0.0"
      inkscape:pagecheckerboard="0"
      inkscape:deskcolor="#d1d1d1"
-     inkscape:zoom="0.27368421"
-     inkscape:cx="171.73077"
-     inkscape:cy="45.673077"
-     inkscape:window-width="1416"
-     inkscape:window-height="820"
-     inkscape:window-x="30"
-     inkscape:window-y="38"
+     inkscape:zoom="3.3793722"
+     inkscape:cx="30.479034"
+     inkscape:cy="41.279857"
+     inkscape:window-width="1424"
+     inkscape:window-height="875"
+     inkscape:window-x="8"
+     inkscape:window-y="40"
      inkscape:window-maximized="0"
-     inkscape:current-layer="g39" />
+     inkscape:current-layer="g39"
+     inkscape:pageshadow="2"
+     showgrid="false" />
   <g
      clip-path="url(#clip0_56_2)"
      id="g39">
-    <path
-       d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
-       fill="#e35a26"
-       id="path12" />
-    <path
-       d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
-       fill="#e35a26"
-       id="path13" />
-    <path
-       d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path14" />
-    <path
-       d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path15" />
-    <path
-       d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path16" />
-    <path
-       d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path17" />
-    <path
-       d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path18" />
-    <path
-       d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
-       fill="#e35a26"
-       id="path19" />
-    <path
-       d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
-       fill="#e35a26"
-       id="path20" />
-    <path
-       d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
-       fill="#e35a26"
-       id="path21" />
-    <path
-       d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
-       fill="#e35a26"
-       id="path22" />
-    <path
-       d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path23" />
-    <path
-       d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path24" />
-    <path
-       d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path25" />
-    <path
-       d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
-       fill="#e35a26"
-       id="path26" />
-    <path
-       d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path27" />
-    <path
-       d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path28" />
-    <path
-       d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path29" />
-    <path
-       d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
-       fill="#e35a26"
-       id="path30" />
-    <path
-       d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
-       fill="#e35a26"
-       id="path31" />
+    <g
+       id="g863"
+       transform="translate(15.09,2.95)">
+      <path
+         d="M 15.18,39.46 V 42.5 H 54.64 V 39.46 Z"
+         fill="#e35a26"
+         id="path12" />
+      <path
+         d="m 39.46,30.3501 v -3.04 H 15.18 v 3.04 z"
+         fill="#e35a26"
+         id="path13" />
+      <path
+         d="m 48.57,27.3199 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path14" />
+      <path
+         d="m 51.6,24.28 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path15" />
+      <path
+         d="M 54.64,27.3199 H 51.6 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path16" />
+      <path
+         d="m 48.57,21.25 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path17" />
+      <path
+         d="M 54.64,21.25 H 51.6 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path18" />
+      <path
+         d="m 15.18,51.6 v 3.04 H 54.64 V 51.6 Z"
+         fill="#e35a26"
+         id="path19" />
+      <path
+         d="M 39.46,24.28 V 21.24 H 15.18 v 3.04 z"
+         fill="#e35a26"
+         id="path20" />
+      <path
+         d="m 15.18,63.74 v 3.04 h 39.46 v -3.04 z"
+         fill="#e35a26"
+         id="path21" />
+      <path
+         d="m 15.18,75.88 v 3.04 h 39.46 v -3.04 z"
+         fill="#e35a26"
+         id="path22" />
+      <path
+         d="m 15.18,6.07 h 6.07 V 3.03 H 18.21 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path23" />
+      <path
+         d="m 24.28,6.07 h 6.07 V 3.03 H 27.31 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path24" />
+      <path
+         d="m 33.39,6.07 h 3.04 v 3.04 h 9.11 V 6.07 h 3.04 V 3.03 H 45.54 V 0 h -6.07 v 3.04 h 3.04 V 6.08 H 39.47 V 3.04 H 36.43 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path25" />
+      <path
+         d="m 51.6,6.07 h -3.04 v 3.04 h 6.07 V 6.07 h 3.04 V 3.03 H 54.63 V 0 h -6.07 v 3.04 h 3.04 v 3.04 z"
+         fill="#e35a26"
+         id="path26" />
+      <path
+         d="m 60.71,0 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path27" />
+      <path
+         d="m 15.18,6.06995 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path28" />
+      <path
+         d="m 24.29,6.06995 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path29" />
+      <path
+         d="m 33.39,6.06995 h -3.04 v 3.04 h 3.04 z"
+         fill="#e35a26"
+         id="path30" />
+      <path
+         d="m 66.77,9.11003 v -3.04 h -3.04 v -3.04 h -3.04 v 3.04 h -3.04 v 3.04 h 6.07 V 84.99 h -3.04 v 3.04 H 9.11 V 84.99 H 6.07 V 9.11003 h 3.04 v -3.04 h 3.04 v -3.04 H 6.08 v 3.04 H 3.04 v 3.04 H 0 V 84.99 h 3.04 v 6.07 h 6.07 v 3.04 h 51.6 v -3.04 h 6.07 v -6.07 h 3.04 V 9.11003 h -3.04 z"
+         fill="#e35a26"
+         id="path31" />
+    </g>
   </g>
   <defs
      id="defs39">
diff --git a/mkdocs.yml b/mkdocs.yml
index 3ff548000..d09a263e4 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -23,7 +23,7 @@ theme:
           scheme: default
           primary: white
     logo: assets/images/logo-square.svg
-    favicon: assets/images/logo-simple.png
+    favicon: assets/images/logo-square.png
     icon:
         repo: fontawesome/brands/github
     features:

From f273c6e88bc75f423f8962590b6db71bb3b07952 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Mon, 16 Jun 2025 12:21:19 +0200
Subject: [PATCH 348/505] Add guide for OpenAI-compatible models

---
 docs/features/models/openai_compatible.md | 83 +++++++++++++++++++++++
 mkdocs.yml                                |  1 +
 2 files changed, 84 insertions(+)
 create mode 100644 docs/features/models/openai_compatible.md

diff --git a/docs/features/models/openai_compatible.md b/docs/features/models/openai_compatible.md
new file mode 100644
index 000000000..7c9307743
--- /dev/null
+++ b/docs/features/models/openai_compatible.md
@@ -0,0 +1,83 @@
+# OpenAI-Compatible APIs
+
+Many inference providers offer OpenAI-compatible APIs, allowing you to use the familiar OpenAI SDK while connecting to different backends. Outlines allows you can leverage various providers while maintaining consistent code.
+
+## What are OpenAI-Compatible APIs?
+
+OpenAI-compatible APIs implement the same REST endpoints and request/response formats as OpenAI's API, but serve different models or run on different infrastructure. This allows you to use the `openai` Python library with any compatible provider by simply changing the `base_url`.
+
+## General Usage Pattern
+
+The standard approach is to use the OpenAI SDK with a custom base URL:
+
+```python
+import openai
+import outlines
+
+# Point to your OpenAI-compatible endpoint
+client = openai.OpenAI(
+    base_url="https://your-provider.com/v1",  # Custom endpoint
+    api_key="your-api-key"
+)
+
+# Use with Outlines
+model = outlines.from_openai(client, "model-name")
+```
+
+## Important: Provider-Specific Parameters
+
+!!! Warning "API-Specific Parameters"
+
+    Some providers require additional parameters in the API request for structured generation to work properly. These are typically passed as extra arguments when calling the model.
+    
+    For example, some providers may need special parameters in the request body to enable guided generation or specify constraints. Always consult your provider's documentation for structured generation requirements.
+
+## Popular OpenAI-Compatible Providers
+
+Many providers offer OpenAI-compatible endpoints:
+
+- **Groq**
+- **Together AI**
+- **Anyscale**
+- **Fireworks AI**
+- **Perplexity**
+- **Local servers** (LocalAI, etc.)
+
+## Configuration Examples
+
+### Basic Setup
+```python
+import openai
+import outlines
+
+# Generic OpenAI-compatible setup
+client = openai.OpenAI(
+    base_url="https://api.your-provider.com/v1",
+    api_key="your-api-key"
+)
+
+model = outlines.from_openai(client, "provider-model-name")
+```
+
+### With Authentication Headers
+```python
+import openai
+import outlines
+
+# Some providers need custom headers
+client = openai.OpenAI(
+    base_url="https://api.your-provider.com/v1",
+    api_key="your-api-key",
+    default_headers={"Custom-Header": "value"}
+)
+
+model = outlines.from_openai(client, "provider-model-name")
+```
+
+## Related Documentation
+
+For specific implementations that use OpenAI-compatible APIs:
+
+- [SGLang](sglang.md): Local inference server with OpenAI-compatible endpoints
+- [vLLM](vllm.md): High-performance inference with OpenAI-compatible API
+- [OpenAI](openai.md): The original OpenAI API implementation
diff --git a/mkdocs.yml b/mkdocs.yml
index d09a263e4..ce075062b 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -176,6 +176,7 @@ nav:
                     - Mlx-lm: features/models/mlxlm.md
                     - Ollama: features/models/ollama.md
                     - OpenAI: features/models/openai.md
+                    - OpenAI compatible API: features/models/openai_compatible.md
                     - SGLang: features/models/sglang.md
                     - TGI: features/models/tgi.md
                     - Transformers: features/models/transformers.md

From f9abff48a3ada37d802e779b43224ccd9fe67c72 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 17 Jun 2025 11:27:00 +0200
Subject: [PATCH 349/505] Modify VLLMOffline to rely on vLLM's
 GuidedDecodingParams argument

---
 docs/features/advanced/logits_processors.md   |   3 +-
 docs/features/models/index.md                 |   2 +-
 docs/features/models/openai_compatible.md     |   2 +-
 docs/features/models/vllm_offline.md          |   8 +-
 outlines/models/__init__.py                   |   3 +-
 outlines/models/vllm_offline.py               | 114 ++++++------------
 tests/models/test_vllm_offline.py             |  40 ------
 .../models/test_vllm_offline_type_adapter.py  |  93 ++++++++++----
 8 files changed, 117 insertions(+), 148 deletions(-)

diff --git a/docs/features/advanced/logits_processors.md b/docs/features/advanced/logits_processors.md
index cbd7b82a0..4e4d9e457 100644
--- a/docs/features/advanced/logits_processors.md
+++ b/docs/features/advanced/logits_processors.md
@@ -14,12 +14,11 @@ Processors can be used to:
 
 Outlines uses logits processors with steerable models — models that run locally and allow fine-grained control over the generation process. When using such models in Outlines, the output type provided is turned into a logits processor that is then passed to the inference engine.
 
-There are four models that support logits processors:
+There are three models that support logits processors:
 
 - LlamaCpp
 - MLXLM
 - Transformers
-- VLLMOffline
 
 Instead of providing an output type that will be turned into a logits processor, it is possible to directly provide a logits processor. To do so, you must create a `Generator` instance using the `processor` keyword argument. You cannot directly call the model with a logits processor.
 
diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index 2c1b3323c..89c4fa124 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -57,7 +57,7 @@ print(result) # '200'
 | Async | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
-| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ |
+| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ |
 
 ## Model Types
 
diff --git a/docs/features/models/openai_compatible.md b/docs/features/models/openai_compatible.md
index 7c9307743..86f9960ea 100644
--- a/docs/features/models/openai_compatible.md
+++ b/docs/features/models/openai_compatible.md
@@ -29,7 +29,7 @@ model = outlines.from_openai(client, "model-name")
 !!! Warning "API-Specific Parameters"
 
     Some providers require additional parameters in the API request for structured generation to work properly. These are typically passed as extra arguments when calling the model.
-    
+
     For example, some providers may need special parameters in the request body to enable guided generation or specify constraints. Always consult your provider's documentation for structured generation requirements.
 
 ## Popular OpenAI-Compatible Providers
diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
index 178f4789c..9c59e18ab 100644
--- a/docs/features/models/vllm_offline.md
+++ b/docs/features/models/vllm_offline.md
@@ -30,6 +30,10 @@ model = outlines.from_vllm_offline(
 )
 ```
 
+!!! Note
+
+    When initializing the `vllm.LLM` object, you can specify a `guided_decoding_backend` to choose what library will be used by vLLM to constrain the generation. Consult the [vLLM documentation](https://docs.vllm.ai/en/v0.8.2/features/structured_outputs.html) on structured output for the list of possible values.
+
 ## Text Generation
 
 To generate text, you can simply call the model with a prompt.
@@ -52,7 +56,7 @@ print(result) # 'Riga'
 
 ## Structured Generation
 
-As a local model, `VLLMOffline` supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
+The `VLLMOffline` model supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
 
 ### Simple Type
 
@@ -165,6 +169,6 @@ print(result) # '23 + 48'
 
 ## Inference Arguments
 
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `generate` method of the `LLM` model instance. Some common inference arguments include `max_tokens`, `temperature`, `frequency_penalty` and `top_p`.
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `generate` method of the `LLM` model instance. An inference argument of particular interest is `sampling_params`. Its value must be a `SamplingParams` instance containing parameters such as `max_tokens` or `temperature`.
 
 See the [vLLM documentation](https://docs.vllm.ai/en/latest/api/vllm/sampling_params.html#vllm.sampling_params.SamplingParams) on sampling parameters for more information on inference parameters.
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 39638a2fb..d818c5db5 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -27,7 +27,7 @@
 from .vllm_offline import VLLMOffline, from_vllm_offline
 from .vllm import AsyncVLLM, VLLM, from_vllm
 
-SteerableModel = Union[LlamaCpp, MLXLM, Transformers, VLLMOffline]
+SteerableModel = Union[LlamaCpp, MLXLM, Transformers]
 BlackBoxModel = Union[
     Anthropic,
     Dottxt,
@@ -37,6 +37,7 @@
     SGLang,
     TGI,
     VLLM,
+    VLLMOffline,
 ]
 AsyncBlackBoxModel = Union[
     AsyncTGI,
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index 97164171a..f970b2f9a 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -1,11 +1,12 @@
 """Integration with the `vllm` library (offline mode)."""
 
+import json
 import warnings
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, List, Optional, Union
 
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.processors import OutlinesLogitsProcessor
+from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
 
 if TYPE_CHECKING:
     from transformers import PreTrainedTokenizerBase
@@ -28,7 +29,7 @@ def format_input(self, model_input):
 
         """
         raise NotImplementedError(
-            f"The input type {input} is not available."
+            f"The input type {input} is not available. "
             "Please use a string or a list of strings."
         )
 
@@ -40,26 +41,39 @@ def format_str_input(self, model_input: str) -> str:
     def format_list_input(self, model_input: List[str]) -> List[str]:
         return model_input
 
-    def format_output_type(
-        self,
-        output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> Union[List[OutlinesLogitsProcessor], List[None]]:
-        """Generate the logits processor argument to pass to the model.
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the structured output argument to pass to the model.
 
-        Argument
-        --------
+        For vLLM, the structured output definition is set in the
+        `GuidedDecodingParams` constructor that is provided as a value to the
+        `guided_decoding` parameter of the `SamplingParams` constructor, itself
+        provided as a value to the `sampling_params` parameter of the `generate`
+        method.
+
+        Parameters
+        ----------
         output_type
-            The logits processor provided.
+            The structured output type provided.
 
         Returns
         -------
-        Union[List[OutlinesLogitsProcessor], List[None]]
-            The logits processor to pass to the model.
+        dict
+            The arguments to provide to the `GuidedDecodingParams` constructor.
 
         """
-        if output_type is not None:
-            return [output_type]
-        return []
+        if output_type is None:
+            return {}
+
+        term = python_types_to_terms(output_type)
+        if isinstance(term, CFG):
+            return {"grammar": term.definition}
+        elif isinstance(term, JsonSchema):
+            guided_decoding_params = {"json": json.loads(term.schema)}
+            if term.whitespace_pattern:
+                guided_decoding_params["whitespace_pattern"] = term.whitespace_pattern
+            return guided_decoding_params
+        else:
+            return {"regex": to_regex(term)}
 
 
 class VLLMOffline(Model):
@@ -69,7 +83,6 @@ class VLLMOffline(Model):
     users at a higher level to arguments to the `vllm.LLM` model.
 
     """
-    tensor_library_name = "torch"
 
     def __init__(self, model: "LLM"):
         """Create a VLLM model instance.
@@ -81,29 +94,13 @@ def __init__(self, model: "LLM"):
 
         """
         self.model = model
-        self.tokenizer = self._get_tokenizer()
         self.type_adapter = VLLMOfflineTypeAdapter()
         self.lora_request = None # v0 legacy, to be removed
 
-    def _get_tokenizer(self):
-        if hasattr(self.model, "get_tokenizer"):
-            tokenizer = self.model.get_tokenizer()
-        elif hasattr(self.model, "tokenizer"):
-            if hasattr(self.model.tokenizer, "tokenizer"):
-                tokenizer = self.model.tokenizer.tokenizer
-            else:
-                tokenizer = self.model.tokenizer
-        else:
-            raise ValueError(
-                "The provided LLM instance neither has a "
-                "`tokenizer` attribute or a `get_tokenizer` method."
-            )
-        return adapt_tokenizer(tokenizer=tokenizer)
-
     def generate(
         self,
         model_input: Union[str, List[str]],
-        output_type: Optional[OutlinesLogitsProcessor] = None,
+        output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, List[str], List[List[str]]]:
         """Generate text using vLLM.
@@ -125,12 +122,16 @@ def generate(
             The text generated by the model.
 
         """
-        from vllm.sampling_params import SamplingParams
+        from vllm.sampling_params import GuidedDecodingParams, SamplingParams
 
         sampling_params = inference_kwargs.pop("sampling_params", None)
+
         if sampling_params is None:
             sampling_params = SamplingParams()
-        sampling_params.logits_processors = self.type_adapter.format_output_type(output_type)
+
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        if output_type_args:
+            sampling_params.guided_decoding = GuidedDecodingParams(**output_type_args)
 
         results = self.model.generate(
             self.type_adapter.format_input(model_input),
@@ -203,46 +204,3 @@ def from_vllm_offline(model: "LLM") -> VLLMOffline:
 
     """
     return VLLMOffline(model)
-
-
-def adapt_tokenizer(
-    tokenizer: "PreTrainedTokenizerBase",
-) -> "PreTrainedTokenizerBase":
-    """Adapt a tokenizer to use to compile the FSM.
-
-    The API of Outlines tokenizers is slightly different to that of
-    `transformers`. In addition we need to handle the missing spaces to Llama's
-    tokenizer to be able to compile FSMs for this model.
-
-    Parameters
-    ----------
-    tokenizer
-        The tokenizer of the model.
-
-    Returns
-    -------
-    PreTrainedTokenizerBase
-        The adapted tokenizer.
-
-    """
-    from transformers import SPIECE_UNDERLINE
-
-    tokenizer.vocabulary = tokenizer.get_vocab()
-    tokenizer.special_tokens = set(tokenizer.all_special_tokens)
-
-    def convert_token_to_string(token: Union[str, bytes]) -> str:
-        string = tokenizer.convert_tokens_to_string([token])
-
-        # A hack to handle missing spaces to HF's Llama tokenizers
-        if (
-            type(token) is str
-            and token.startswith(SPIECE_UNDERLINE)
-            or token == "<0x20>"
-        ):
-            return " " + string
-
-        return string
-
-    tokenizer.convert_token_to_string = convert_token_to_string
-
-    return tokenizer
diff --git a/tests/models/test_vllm_offline.py b/tests/models/test_vllm_offline.py
index 21cbe8041..01d5d6a59 100644
--- a/tests/models/test_vllm_offline.py
+++ b/tests/models/test_vllm_offline.py
@@ -31,9 +31,7 @@ def test_vllm_model_initialization():
     model = from_vllm_offline(LLM(TEST_MODEL))
     assert isinstance(model, VLLMOffline)
     assert isinstance(model.model, LLM)
-    assert hasattr(model, "tokenizer")
     assert isinstance(model.type_adapter, VLLMOfflineTypeAdapter)
-    assert model.tensor_library_name == "torch"
 
 
 @pytest.fixture(scope="session")
@@ -111,44 +109,6 @@ def test_vllm_batch_samples(model):
         assert len(item) == 2
 
 
-def test_vllm_batch_samples_constrained(model):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
-
-    result = model(
-        "Cat or dog?",
-        Foo,
-        sampling_params=SamplingParams(n=2)
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert result[0] in ["cat", "dog"]
-    assert result[1] in ["cat", "dog"]
-
-    result = model(
-        ["Cat or dog?", "Cat or dog?"],
-        Foo,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    assert result[0] in ["cat", "dog"]
-    assert result[1] in ["cat", "dog"]
-
-    result = model(
-        ["Cat or dog?", "Cat or dog?"],
-        Foo,
-        sampling_params=SamplingParams(n=2)
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
-        assert item[0] in ["cat", "dog"]
-        assert item[1] in ["cat", "dog"]
-
-
 def test_vllm_streaming(model):
     with pytest.raises(
         NotImplementedError,
diff --git a/tests/models/test_vllm_offline_type_adapter.py b/tests/models/test_vllm_offline_type_adapter.py
index c1480dbed..b8a18b6d1 100644
--- a/tests/models/test_vllm_offline_type_adapter.py
+++ b/tests/models/test_vllm_offline_type_adapter.py
@@ -1,37 +1,84 @@
-import pytest
+import io
+import json
+from dataclasses import dataclass
 
-import transformers
+import pytest
 
-import outlines
 from outlines.models.vllm_offline import VLLMOfflineTypeAdapter
-from outlines.processors.structured import RegexLogitsProcessor
+from outlines.types import CFG, JsonSchema, Regex
+
+
+CFG_STRING = """
+?start: expr
+?expr: NUMBER
+"""
+
+JSON_SCHEMA_STRING = """
+{
+    "type": "object",
+    "properties": {
+        "answer": {"type": "number"}
+    }
+}
+"""
 
 
 @pytest.fixture
 def type_adapter():
     return VLLMOfflineTypeAdapter()
 
+@pytest.fixture
+def cfg_instance():
+    return CFG(CFG_STRING)
+
+@pytest.fixture
+def json_schema_instance():
+    return JsonSchema(JSON_SCHEMA_STRING)
 
 @pytest.fixture
-def logits_processor():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=model.tokenizer,
-        tensor_library_name="torch",
-    )
-
-
-def test_vllm_offline_type_adapter_format_input(type_adapter):
-    assert type_adapter.format_input("foo") == "foo"
-    assert type_adapter.format_input(["foo", "bar"]) == ["foo", "bar"]
+def json_schema_whitespace_instance():
+    return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
+
+@pytest.fixture
+def regex_instance():
+    return Regex(r"[0-9]+")
+
+
+def test_vllm_type_adapter_input_text(type_adapter):
+    # string
+    message = "prompt"
+    result = type_adapter.format_input(message)
+    assert message == result
+
+    # list of strings
+    messages = ["foo", "bar"]
+    result = type_adapter.format_input(messages)
+    assert messages == result
+
+
+def test_vllm_type_adapter_input_invalid(type_adapter):
     with pytest.raises(NotImplementedError):
-        type_adapter.format_input({"foo": "bar"})
+        _ = type_adapter.format_input({"foo": "bar"})
 
 
-def test_vllm_offline_type_adapter_format_output_type(type_adapter, logits_processor):
-    assert type_adapter.format_output_type(None) == []
-    assert type_adapter.format_output_type(logits_processor) == [logits_processor]
+def test_vllm_type_adapter_output_type(
+    type_adapter,
+    cfg_instance,
+    json_schema_instance,
+    json_schema_whitespace_instance,
+    regex_instance,
+):
+    assert type_adapter.format_output_type(None) == {}
+    assert type_adapter.format_output_type(cfg_instance) == {
+        "grammar": CFG_STRING
+    }
+    assert type_adapter.format_output_type(json_schema_instance) == {
+        "json": json.loads(JSON_SCHEMA_STRING)
+    }
+    assert type_adapter.format_output_type(json_schema_whitespace_instance) == {
+        "json": json.loads(JSON_SCHEMA_STRING),
+        "whitespace_pattern": "\n"
+    }
+    assert type_adapter.format_output_type(regex_instance) == {
+        "regex": "([0-9]+)"
+    }

From 58473e8a5eafc95be3e1c36e118d3ff7e206ab8e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 17 Jun 2025 15:45:54 +0200
Subject: [PATCH 350/505] Create optional dependencies for the models

---
 docs/features/models/anthropic.md            |  2 +-
 docs/features/models/dottxt.md               |  2 +-
 docs/features/models/gemini.md               | 50 ++++++------
 docs/features/models/llamacpp.md             | 32 ++++----
 docs/features/models/mlxlm.md                | 22 +++---
 docs/features/models/ollama.md               |  4 +-
 docs/features/models/openai.md               |  6 +-
 docs/features/models/openai_compatible.md    |  4 +
 docs/features/models/sglang.md               |  4 +-
 docs/features/models/tgi.md                  |  2 +-
 docs/features/models/transformers.md         |  4 +-
 docs/features/models/vllm.md                 |  4 +-
 docs/features/models/vllm_offline.md         | 26 ++++---
 outlines/models/transformers.py              |  2 -
 outlines/processors/guide.py                 |  2 +-
 outlines/v0_legacy/base.py                   | 78 ++++++++++++-------
 outlines/v0_legacy/models/transformers.py    |  4 +-
 pyproject.toml                               | 12 +++
 tests/models/test_transformers.py            |  2 +-
 tests/models/test_transformers_multimodal.py | 80 +++++---------------
 20 files changed, 171 insertions(+), 171 deletions(-)

diff --git a/docs/features/models/anthropic.md b/docs/features/models/anthropic.md
index f9b648ccc..dd0794386 100644
--- a/docs/features/models/anthropic.md
+++ b/docs/features/models/anthropic.md
@@ -6,7 +6,7 @@ title: Anthropic
 
 !!! Installation
 
-    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines: `pip install anthropic`.
+    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines. Install all optional dependencies of the `Anthropic` model with: `pip install outlines[anthropic]`.
 
     You also need to have an Anthropic API key. This API key must either be set as an environment variable called `ANTHROPIC_API_KEY` or be provided to the `anthropic.Anthropic` class when instantiating it.
 
diff --git a/docs/features/models/dottxt.md b/docs/features/models/dottxt.md
index 09f06e5d5..306ef2c62 100644
--- a/docs/features/models/dottxt.md
+++ b/docs/features/models/dottxt.md
@@ -6,7 +6,7 @@ title: Dottxt
 
 !!! Installation
 
-    You need to install the `dottxt` python sdk to be able to use the Dottxt API in Outlines: `pip install dottxt`
+    You need to install the `dottxt` python sdk to be able to use the Dottxt API in Outlines. Install all optional dependencies of the `Dottxt` model with: `pip install outlines[dottxt]`.
 
     You also need to have a Dottxt API key. This API key must either be set as an environment variable called `DOTTXT_API_KEY` or be provided to the `dottxt.client.Dottxt` class when instantiating it.
 
diff --git a/docs/features/models/gemini.md b/docs/features/models/gemini.md
index 5c1b1b949..e6daccf26 100644
--- a/docs/features/models/gemini.md
+++ b/docs/features/models/gemini.md
@@ -2,30 +2,30 @@
 
 !!! Installation
 
-    You need to install the `google-generativeai` library to be able to use the Gemini API in Outlines: `pip install google-generativeai`
+    You need to install the `google.genai` libray to be able to use the Gemini API in Outlines. Install all optional dependencies of the `Gemini` model with: `pip install outlines[gemini]`.
 
-    You also need to have a Gemini API key. This API key must either be set as an environment variable called `GEMINI_API_KEY` or be provided to the `google.generativeai.GenerativeModel` class when instantiating it.
+    You also need to have a Gemini API key. This API key must either be set as an environment variable called `GEMINI_API_KEY` or be provided to the `google.genai.Client` class when instantiating it.
 
 ## Model Initialization
 
 To create a Gemini model instance, you can use the `from_gemini` function. It takes 2 arguments:
 
-- `client`: a `google.generativeai.GenerativeModel` instance
+- `client`: a `google.genai.Client` instance
 - `model_name`: the name of the model you want to use in subsequent model calls (optional)
 
 For instance:
 
 ```python
 import outlines
-import google.generativeai as genai
+from google import genai
 
 # Create the client
-client = genai.GenerativeModel()
+client = genai.Client()
 
 # Create the model
 model = outlines.from_gemini(
     client,
-    "gemini-1-5-flash"
+    "gemini-1.5-flash-latest"
 )
 ```
 
@@ -39,13 +39,13 @@ For instance:
 
 ```python
 import outlines
-import google.generativeai as genai
+from google import genai
 
 # Create the model
-model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
 
 # Call it to generate text
-result = model("What's the capital of Latvia?", max_tokens=20)
+result = model("What's the capital of Latvia?", max_output_tokens=20)
 print(result) # 'Riga'
 ```
 
@@ -53,13 +53,13 @@ Gemini also supports streaming. For instance:
 
 ```python
 import outlines
-import google.generativeai as genai
+from google import genai
 
 # Create the model
-model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
 
 # Stream text
-for chunk in model.stream("Write a short story about a cat.", max_tokens=20):
+for chunk in model.stream("Write a short story about a cat.", max_output_tokens=20):
     print(chunk) # 'In...'
 ```
 
@@ -71,7 +71,7 @@ Gemini provides supports for some forms of structured output: multiple choice, J
 
 ```python
 import outlines
-import google.generativeai as genai
+from google import genai
 from enum import Enum
 
 class PizzaOrBurger(Enum):
@@ -79,10 +79,10 @@ class PizzaOrBurger(Enum):
     burger = "burger"
 
 # Create the model
-model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
 
 # Call it with the ouput type to generate structured text
-result = model("Pizza or burger?", PizzaOrBurger, max_tokens=20)
+result = model("Pizza or burger?", PizzaOrBurger, max_output_tokens=20)
 print(result) # 'pizza'
 ```
 
@@ -95,9 +95,10 @@ Gemini supports only three types of objects used to define a JSON Schema:
 - TypedDicts
 
 ```python
-import google.generativeai as genai
-import outlines
+from typing import List
 from pydantic import BaseModel
+from google import genai
+import outlines
 
 class Character(BaseModel):
     name: str
@@ -105,7 +106,7 @@ class Character(BaseModel):
     skills: List[str]
 
 # Create the model
-model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
 
 # Call it with the ouput type to generate structured text
 result = model("Create a character", Character)
@@ -119,8 +120,7 @@ A specificity of Gemini is that, despite not supporting regex, it does support a
 
 ```python
 from dataclasses import dataclass
-from typing import List
-import google.generativeai as genai
+from google import genai
 import outlines
 
 @dataclass
@@ -130,15 +130,19 @@ class Character:
     skills: List[str]
 
 # Create the model
-model = outlines.from_gemini(genai.GenerativeModel(), "gemini-1-5-flash")
+model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
 
 # Call it with the ouput type to generate structured text
-result = model("Create a character", List[Character])
+result = model("Create a character", list[Character])
 print(result) # '[{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}, {["name":...'
 ```
 
+!!! Attention
+
+    The structured objects must be in a built-in `list`, not a `List` from the `typing` library
+
 ## Inference arguments
 
-You can provide the same optional parameters you would pass to the `google.generativeai.GenerativeModel` client both during the initialization of the Gemini model and when generating text. Some of the common inference arguments include `max_tokens`, `temperature`, and other generation parameters.
+You can provide the same optional parameters you would pass to the `google.genai.Client` client both during the initialization of the Gemini model and when generating text. Some of the common inference arguments include `max_output_tokens`, `temperature`, and other generation parameters.
 
 Consult the [Google Generative AI documentation](https://github.com/googleapis/python-genai) for the full list of parameters.
diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
index 1abef52a3..cf7dd3e0c 100644
--- a/docs/features/models/llamacpp.md
+++ b/docs/features/models/llamacpp.md
@@ -8,7 +8,7 @@ Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/l
 
 !!! Installation
 
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration: `pip install llama-cpp-python`
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration. Install all optional dependencies of the `LlamaCpp` model with: `pip install outlines[llamacpp]`.
 
     See the [llama-cpp-python Github page](https://github.com/abetlen/llama-cpp-python?tab=readme-ov-file#supported-backends) for instructions on installing with CUDA, Metal, ROCm and other backends.
 
@@ -25,8 +25,8 @@ from llama_cpp import Llama
 
 model = outlines.from_llamacpp(
     Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
     )
 )
 ```
@@ -44,8 +44,8 @@ from llama_cpp import Llama
 # Create the model
 model = outlines.from_llamacpp(
     Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
     )
 )
 
@@ -63,8 +63,8 @@ from llama_cpp import Llama
 # Create the model
 model = outlines.from_llamacpp(
     Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
     )
 )
 
@@ -87,8 +87,8 @@ output_type = int
 
 model = outlines.from_llamacpp(
     Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
     )
 )
 
@@ -111,12 +111,12 @@ class Character(BaseModel):
 
 model = outlines.from_llamacpp(
     Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
     )
 )
 
-result = model("Create a character.", output_type=Character)
+result = model("Create a character.", output_type=Character, max_tokens=200)
 print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
 print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
@@ -132,8 +132,8 @@ output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
 model = outlines.from_llamacpp(
     Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
     )
 )
 
@@ -152,8 +152,8 @@ output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
 model = outlines.from_llamacpp(
     Llama.from_pretrained(
-        repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
     )
 )
 
diff --git a/docs/features/models/mlxlm.md b/docs/features/models/mlxlm.md
index 67d5d0d5e..d2cba95fc 100644
--- a/docs/features/models/mlxlm.md
+++ b/docs/features/models/mlxlm.md
@@ -8,7 +8,9 @@ Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx
 
 !!! Note "Installation"
 
-    You need to install the `mlx` and `mlx-lm` libraries on a device which [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration. To download the required libraries, run `pip install mlx mlx-lm`.
+    You need a device that [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration.
+
+    You need to install the `mlx` and `mlx-lm` libraries to be able to use mlx in Outlines. Install all optional dependencies of the `MLXLM` model with: `pip install outlines[mlxlm]`.
 
 ## Model Initialization
 
@@ -27,7 +29,7 @@ import mlx_lm
 
 # Create the model
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 ```
 
@@ -43,7 +45,7 @@ import mlx_lm
 
 # Load the model
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 
 # Call it to generate text
@@ -59,7 +61,7 @@ import mlx_lm
 
 # Load the model
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 
 # Stream text
@@ -80,7 +82,7 @@ import mlx_lm
 output_type = int
 
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 
 result = model("How many countries are there in the world?", output_type)
@@ -101,7 +103,7 @@ class Character(BaseModel):
     skills: List[str]
 
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 
 result = model("Create a character.", output_type=Character)
@@ -119,7 +121,7 @@ import mlx_lm
 output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 
 result = model("What is the capital of France?", output_type)
@@ -136,7 +138,7 @@ import mlx_lm
 output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 
 result = model("Generate a fake social security number.", output_type)
@@ -146,7 +148,7 @@ print(result) # '782-32-3789'
 ### Context-Free Grammar
 
 ```python
-from outlines.text import CFG
+from outlines.types import CFG
 import outlines
 import mlx_lm
 
@@ -173,7 +175,7 @@ arithmetic_grammar = """
 output_type = CFG(arithmetic_grammar)
 
 model = outlines.from_mlxlm(
-    **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
 )
 
 result = model("Write an addition.", output_type, max_tokens=20)
diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
index 2f75a7621..eaad4c3e2 100644
--- a/docs/features/models/ollama.md
+++ b/docs/features/models/ollama.md
@@ -6,10 +6,10 @@ title: Ollama
 
 !!! Installation
 
-    To be able to use Ollama in Outlines, you must install both Ollama and the `ollama` python sdk.
+    To be able to use Ollama in Outlines, you must install both Ollama and the optional dependency libraries of the model.
 
     - To download Ollama: https://ollama.com/download
-    - To install the ollama python sdk: `pip install ollama`
+    - To install the ollama python sdk: `pip install outlines[ollama]`
 
     Consult the [`ollama` documentation](https://github.com/ollama/ollama-python) for detailed information on installation and client initialization.
 
diff --git a/docs/features/models/openai.md b/docs/features/models/openai.md
index c06acbc96..e4532d213 100644
--- a/docs/features/models/openai.md
+++ b/docs/features/models/openai.md
@@ -2,7 +2,7 @@
 
 !!! Installation
 
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines: `pip install openai`
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
 
     You also need to have an OpenAI API key. This API key must either be set as an environment variable called `OPENAI_API_KEY` or be provided to the `openai.OpenAI` class when instantiating it.
 
@@ -104,7 +104,7 @@ class Character(BaseModel):
 model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
 
 # Call it with the output type to generate structured text
-result = model("Create a character", Character, top_p=0.1)
+result = model("Create a character, use the json format.", Character, top_p=0.1)
 print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
 print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
@@ -121,7 +121,7 @@ import outlines
 model = outlines.from_openai(openai.OpenAI(), "gpt-4o")
 
 # Call it with the output type to generate structured text
-result = model("Create a character", dict, temperature=0.5)
+result = model("Create a character, use the json format.", dict, temperature=0.5)
 print(result) # '{"first_name": "Henri", "last_name": "Smith", "height": "170"}'
 ```
 
diff --git a/docs/features/models/openai_compatible.md b/docs/features/models/openai_compatible.md
index 86f9960ea..4925d53d4 100644
--- a/docs/features/models/openai_compatible.md
+++ b/docs/features/models/openai_compatible.md
@@ -6,6 +6,10 @@ Many inference providers offer OpenAI-compatible APIs, allowing you to use the f
 
 OpenAI-compatible APIs implement the same REST endpoints and request/response formats as OpenAI's API, but serve different models or run on different infrastructure. This allows you to use the `openai` Python library with any compatible provider by simply changing the `base_url`.
 
+!!! Installation
+
+    You need to install the `openai` library to be able to use the OpenAI-compatible APIs in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
+
 ## General Usage Pattern
 
 The standard approach is to use the OpenAI SDK with a custom base URL:
diff --git a/docs/features/models/sglang.md b/docs/features/models/sglang.md
index beb991b6e..95a6eb7b5 100644
--- a/docs/features/models/sglang.md
+++ b/docs/features/models/sglang.md
@@ -6,7 +6,7 @@ title: SGLang
 
 ## Prerequisites
 
-The Outlines `SGLang` model is intended to be used along with an SGLang instance running on a separate server (can be local or remote). Make sure you have a SGLang server running and accessible before using the `SGLang` model. As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed: `pip install openai`.
+The Outlines `SGLang` model is intended to be used along with an SGLang instance running on a separate server (can be local or remote). Make sure you have a SGLang server running and accessible before using the `SGLang` model. As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies of the `SGLang` model with: `pip install outlines[sglang]`.
 
 When launching your SGLang server, you can specify the backend engine to use for structured generation through the `grammar-backend` cli argument. Add `--grammar-backend outlines` to your command to use Outlines instead of the default engine.
 
@@ -147,7 +147,7 @@ SGLang supports grammars, but expects an EBNF format instead of the Lark format
 ```python
 import openai
 import outlines
-from outlines.text import CFG
+from outlines.types import CFG
 
 ebnf_grammar = """
 root ::= answer
diff --git a/docs/features/models/tgi.md b/docs/features/models/tgi.md
index 331b68931..8fb807580 100644
--- a/docs/features/models/tgi.md
+++ b/docs/features/models/tgi.md
@@ -6,7 +6,7 @@ title: TGI
 
 ## Prerequisites
 
-The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed: `pip install huggingface_hub`
+The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed. Install all optional dependencoes of the `TGI` model with: `pip install outlines[tgi]`
 
 ## Model Initialization
 
diff --git a/docs/features/models/transformers.md b/docs/features/models/transformers.md
index f03baa991..9dc1cab79 100644
--- a/docs/features/models/transformers.md
+++ b/docs/features/models/transformers.md
@@ -6,7 +6,7 @@ title: Transformers
 
 !!! Installation
 
-    You need to install the `transformers` library to be able to use the `Transformers` model in Outlines: `pip install transformers`
+    You need to install the `transformers` library to be able to use the Transformers in Outlines. Install all optional dependencies of the `Transformers` model with: `pip install outlines[transformers]`.
 
     See the [HuggingFace documentation](https://huggingface.co/docs/transformers/en/installation) for more information on installing `transformers` with CPU, GPU...
 
@@ -163,7 +163,7 @@ print(result) # '782-32-3789'
 ```python
 import outlines
 from transformers import AutoModelForCausalLM, AutoTokenizer
-from outlines.text import CFG
+from outlines.types import CFG
 
 arithmetic_grammar = """
 ?start: sum
diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index ad7fea0ba..21a048f9f 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -6,7 +6,7 @@ title: vLLM
 
 ## Prerequisites
 
-The Outlines `VLLM` model is intended to be used along with a vLLM instance running on a separate server (can be local or remote). Make sure you have a vLLM server running and accessible before using the `VLLM` model. As the vLLM client relies on the `openai` python sdk, you need to have an `openai` package installed: `pip install openai`.
+The Outlines `VLLM` model is intended to be used along with a vLLM instance running on a separate server (can be local or remote). Make sure you have a vLLM server running and accessible before using the `VLLM` model. As the vLLM client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies for the `VLLM` model with: `pip install openai`.
 
 If you want to use the vllm offline inference mode instead of the server mode, please refer to the [VLLMOffline](./vllm_offline.md) model documentation.
 
@@ -146,7 +146,7 @@ print(result) # '782-32-3789'
 ```python
 import openai
 import outlines
-from outlines.text import CFG
+from outlines.types import CFG
 
 arithmetic_grammar = """
 ?start: sum
diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
index 9c59e18ab..38c476e59 100644
--- a/docs/features/models/vllm_offline.md
+++ b/docs/features/models/vllm_offline.md
@@ -8,7 +8,9 @@ Outlines provides an integration with [vLLM](https://docs.vllm.ai/en/latest/) us
 
 !!! Note "Installation"
 
-    You need to install the `vllm` library to be able to use the `VLLMOffline` model: `pip install vllm`.
+    You need to install the `vllm` library to be able to use the `VLLMOffline` model: `pip install vllm`. Due to a library version conflict between outlines and vllm, you MUST install `vllm` before installing `outlines`.
+
+    When installing `outlines` (after having first installed `vllm`), you may encounter the following error: `ERROR: pip's dependency resolver does not currently take into account all the packages that are installed`. You can safely ignore it.
 
     See the [vLLM documentation](https://docs.vllm.ai/en/latest/getting_started/installation/index.html) for instructions on how to install vLLM for CPU, ROCm...
 
@@ -42,7 +44,7 @@ For instance:
 
 ```python
 import outlines
-from vllm import LLM
+from vllm import LLM, SamplingParams
 
 # Create the model
 model = outlines.from_vllm_offline(
@@ -50,7 +52,7 @@ model = outlines.from_vllm_offline(
 )
 
 # Call it to generate text
-result = model("What's the capital of Latvia?", max_tokens=20)
+result = model("What's the capital of Latvia?", sampling_params=SamplingParams(max_tokens=20))
 print(result) # 'Riga'
 ```
 
@@ -78,7 +80,7 @@ print(result) # '200'
 
 ```python
 import outlines
-from vllm import LLM
+from vllm import LLM, SamplingParams
 from typing import List
 from pydantic import BaseModel
 
@@ -91,7 +93,7 @@ model = outlines.from_vllm_offline(
     LLM("microsoft/Phi-3-mini-4k-instruct")
 )
 
-result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
+result = model("Create a character.", output_type=Character, sampling_params=SamplingParams(frequency_penalty=1.5, max_tokens=200))
 print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
 print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
@@ -101,7 +103,7 @@ print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['arc
 ```python
 from typing import Literal
 import outlines
-from vllm import LLM
+from vllm import LLM, SamplingParams
 
 output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
@@ -109,7 +111,7 @@ model = outlines.from_vllm_offline(
     LLM("microsoft/Phi-3-mini-4k-instruct")
 )
 
-result = model("What is the capital of France?", output_type, temperature=0)
+result = model("What is the capital of France?", output_type, sampling_params=SamplingParams(temperature=0))
 print(result) # 'Paris'
 ```
 
@@ -117,7 +119,7 @@ print(result) # 'Paris'
 
 ```python
 import outlines
-from vllm import LLM
+from vllm import LLM, SamplingParams
 from outlines.types import Regex
 
 output_type = Regex(r"\d{3}-\d{2}-\d{4}")
@@ -126,7 +128,7 @@ model = outlines.from_vllm_offline(
     LLM("microsoft/Phi-3-mini-4k-instruct")
 )
 
-result = model("Generate a fake social security number.", output_type, top_p=0.1)
+result = model("Generate a fake social security number.", output_type, sampling_params=SamplingParams(top_p=0.1))
 print(result) # '782-32-3789'
 ```
 
@@ -134,8 +136,8 @@ print(result) # '782-32-3789'
 
 ```python
 import outlines
-from vllm import LLM
-from outlines.text import CFG
+from vllm import LLM, SamplingParams
+from outlines.types import CFG
 
 arithmetic_grammar = """
 ?start: sum
@@ -169,6 +171,6 @@ print(result) # '23 + 48'
 
 ## Inference Arguments
 
-When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `generate` method of the `LLM` model instance. An inference argument of particular interest is `sampling_params`. Its value must be a `SamplingParams` instance containing parameters such as `max_tokens` or `temperature`.
+When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `generate` method of the `LLM` model instance. An argument of particular interest is `sampling_params`. It takes as a value a `vllm.SamplingParams` instance containing parameters such as max_tokens or temperature.
 
 See the [vLLM documentation](https://docs.vllm.ai/en/latest/api/vllm/sampling_params.html#vllm.sampling_params.SamplingParams) on sampling parameters for more information on inference parameters.
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 06ed7df7b..01c7b34a5 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -286,8 +286,6 @@ def generate(
         prompts, inputs = self._prepare_model_inputs(model_input, output_type)
         logits_processor = self.type_adapter.format_output_type(output_type)
 
-        print(inference_kwargs)
-
         generated_ids = self._generate_output_seq(
             prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )
diff --git a/outlines/processors/guide.py b/outlines/processors/guide.py
index 8139417e4..bebddd9cb 100644
--- a/outlines/processors/guide.py
+++ b/outlines/processors/guide.py
@@ -8,7 +8,6 @@
 import warnings
 from typing import TYPE_CHECKING, Any, Generator, ValuesView, Union
 
-import torch
 from lark.indenter import DedentError
 from lark.lexer import UnexpectedCharacters, UnexpectedToken
 from outlines_core.fsm.guide import Generate
@@ -227,6 +226,7 @@ def get_next_instruction(self, state: CFGState) -> Instruction:
             ids.
 
         """
+        import torch
 
         if state.parser_state is None:
             return Write(torch.tensor([self.eos_token_id]))
diff --git a/outlines/v0_legacy/base.py b/outlines/v0_legacy/base.py
index 60bf655a1..9a5396b33 100644
--- a/outlines/v0_legacy/base.py
+++ b/outlines/v0_legacy/base.py
@@ -33,45 +33,65 @@ class vectorize:
     """
 
     def __init__(self, func: Callable, signature: Optional[str] = None):
-        # imports are in the __init__ as numpy is an optional dependency
-        import numpy as np
-        self.np = np
-        # Import required functions based on NumPy version
-        np_major_version = int(np.__version__.split(".")[0])
-        if np_major_version >= 2:
-            from numpy.lib._function_base_impl import (
-                _calculate_shapes,
-                _parse_gufunc_signature,
-                _parse_input_dimensions,
-                _update_dim_sizes,
-            )
-        else: # pragma: no cover
-            from numpy.lib.function_base import (
-                _calculate_shapes,
-                _parse_gufunc_signature,
-                _parse_input_dimensions,
-                _update_dim_sizes,
-            )
-        self._calculate_shapes = _calculate_shapes
-        self._parse_gufunc_signature = _parse_gufunc_signature
-        self._parse_input_dimensions = _parse_input_dimensions
-        self._update_dim_sizes = _update_dim_sizes
-        #
+        self.setup_numpy(signature)
         self.func = func
         self.signature = signature
         self.is_coroutine_fn = inspect.iscoroutinefunction(func)
 
         functools.update_wrapper(self, func)
 
-        if signature is not None:
-            # Parse the signature string into a Python data structure.
-            # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
-            self._in_and_out_core_dimensions = self._parse_gufunc_signature(signature)
-        else:
+    def setup_numpy(self, signature: Optional[str] = None):
+        """Setup NumPy and import required functions based on NumPy version.
+
+        As numpy is an optional dependency only needed for the legacy OpenAI
+        model, we put the import in a try block to make sure no error is raised
+        if it's not installed.
+        """
+        try:
+            import numpy as np
+            self.np = np
+            # Import required functions based on NumPy version
+            np_major_version = int(np.__version__.split(".")[0])
+            if np_major_version >= 2:
+                from numpy.lib._function_base_impl import (
+                    _calculate_shapes,
+                    _parse_gufunc_signature,
+                    _parse_input_dimensions,
+                    _update_dim_sizes,
+                )
+            else: # pragma: no cover
+                from numpy.lib.function_base import (
+                    _calculate_shapes,
+                    _parse_gufunc_signature,
+                    _parse_input_dimensions,
+                    _update_dim_sizes,
+                )
+            self._calculate_shapes = _calculate_shapes
+            self._parse_gufunc_signature = _parse_gufunc_signature
+            self._parse_input_dimensions = _parse_input_dimensions
+            self._update_dim_sizes = _update_dim_sizes
+            if signature is not None:
+                # Parse the signature string into a Python data structure.
+                # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
+                self._in_and_out_core_dimensions = self._parse_gufunc_signature(signature)
+            else:
+                self._in_and_out_core_dimensions = None
+        except (ModuleNotFoundError, ImportError):
+            self.np = None
+            self._calculate_shapes = None
+            self._parse_gufunc_signature = None
+            self._parse_input_dimensions = None
+            self._update_dim_sizes = None
             self._in_and_out_core_dimensions = None
 
     def __call__(self, *args, **kwargs):
         """Call the vectorized function."""
+        # Raise an error ourselves if NumPy is not installed.
+        if not self.np:
+            raise ImportError(
+                "NumPy is required to use the legacy version of the OpenAI "
+                "model."
+            )
         if not args and not kwargs:
             return self.call_thunk()
         elif self.signature is not None:
diff --git a/outlines/v0_legacy/models/transformers.py b/outlines/v0_legacy/models/transformers.py
index d7cf135ce..eb936c491 100644
--- a/outlines/v0_legacy/models/transformers.py
+++ b/outlines/v0_legacy/models/transformers.py
@@ -4,8 +4,6 @@
 import warnings
 from typing import Optional
 
-from transformers import set_seed
-
 from outlines.models import Transformers
 
 
@@ -15,6 +13,8 @@ def transformers_params_adapter(
     """Adapt the parameters of the legacy generator for the `transformers` model
     to the new `Generator` API.
     """
+    from transformers import set_seed
+
     # generation_params
     if (
         generation_params.get("stop_at") is not None
diff --git a/pyproject.toml b/pyproject.toml
index ec23577ac..19e51dd06 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -41,10 +41,22 @@ dependencies = [
    "airportsdata",
    "outlines_core==0.1.26",
    "genson",
+   "torch",
 ]
 dynamic = ["version"]
 
 [project.optional-dependencies]
+anthropic = ["anthropic"]
+dottxt = ["dottxt"]
+gemini = ["google-genai"]
+llamacpp = ["huggingface-hub", "llama-cpp-python"]
+mlxlm = ["datasets", "mlx", "mlx-lm"]
+ollama = ["ollama"]
+openai = ["openai"]
+sglang = ["openai"]
+tgi = ["huggingface_hub"]
+transformers = ["accelerate", "datasets", "transformers"]
+vllm = ["openai"]
 test = [
     "pre-commit",
     "pytest",
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 96d8d10f2..cd6f60811 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -16,7 +16,7 @@
 
 TEST_MODEL = "erwanf/gpt2-mini"
 TEST_MODEL_MAMBA = "hf-internal-testing/tiny-random-MambaForCausalLM"
-TEST_MODEL_BART = "sshleifer/bart-tiny-random"
+TEST_MODEL_BART = "trl-internal-testing/tiny-BartModel"
 
 
 def test_transformers_instantiate_invalid():
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 7f64e9821..73b5b538d 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -9,7 +9,7 @@
 from PIL import Image
 from pydantic import BaseModel
 from transformers import (
-    LlavaForConditionalGeneration,
+    AutoModelForVision2Seq,
     AutoProcessor,
 )
 
@@ -21,7 +21,7 @@
 )
 from outlines.types import Regex
 
-TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+TEST_MODEL = "HuggingFaceTB/SmolVLM-256M-Instruct"
 IMAGE_URLS = [
     "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
     "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
@@ -40,14 +40,14 @@ def img_from_url(url):
 @pytest.fixture
 def model():
     return outlines.from_transformers(
-        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+        AutoModelForVision2Seq.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
 
 
 def test_transformers_vision_instantiate_simple():
     model = outlines.from_transformers(
-        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+        AutoModelForVision2Seq.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
     assert isinstance(model, TransformersMultiModal)
@@ -60,26 +60,27 @@ def test_transformers_vision_simple(model, images):
     result = model.generate(
         {"text": "<image>Describe this image in one sentence:", "images": images[0]},
         None,
+        max_new_tokens=2,
     )
     assert isinstance(result, str)
 
 
 def test_transformers_vision_call(model, images):
     result = model(
-        {"text": "<image>Describe this image in one sentence:", "images": images[0]}
+        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
+        max_new_tokens=2,
     )
     assert isinstance(result, str)
 
 
 def test_transformers_vision_wrong_number_images(model, images):
     with pytest.raises(ValueError):
-        a = model(
+        model(
             {
                 "text": "<image>Describe this image in one sentence:",
                 "images": [images[0], images[1]],
             }
         )
-        print(a)
 
 
 def test_transformers_vision_wrong_input_type(model):
@@ -90,7 +91,7 @@ def test_transformers_vision_wrong_input_type(model):
 def test_transformers_inference_kwargs(model, images):
     result = model(
         {"text": "<image>Describe this image in one sentence:", "images": images[0]},
-        max_new_tokens=100,
+        max_new_tokens=2,
     )
     assert isinstance(result, str)
 
@@ -111,7 +112,8 @@ def test_transformers_several_images(model, images):
         {
             "text": "<image><image>Describe this image in one sentence:",
             "images": [images[0], images[1]],
-        }
+        },
+        max_new_tokens=2,
     )
     assert isinstance(result, str)
 
@@ -123,13 +125,15 @@ class Foo(BaseModel):
     result = model(
         {"text": "<image>Give a name to this animal.", "images": images[0]},
         Foo,
+        max_new_tokens=10,
     )
     assert "name" in result
 
 
 def test_transformers_vision_regex(model, images):
     result = model(
-        {"text": "<image>How old is it?", "images": images[0]}, Regex(r"[0-9]")
+        {"text": "<image>How old is it?", "images": images[0]},
+        Regex(r"[0-9]")
     )
 
     assert isinstance(result, str)
@@ -142,7 +146,8 @@ class Foo(Enum):
         dog = "dog"
 
     result = model(
-        {"text": "<image>Is it a cat or a dog?", "images": images[0]}, Foo
+        {"text": "<image>Is it a cat or a dog?", "images": images[0]},
+        Foo,
     )
 
     assert isinstance(result, str)
@@ -154,53 +159,10 @@ def test_transformers_vision_batch_samples(model, images):
         {"text": "<image>Describe this image in one sentence.", "images": images[0]},
         num_return_sequences=2,
         num_beams=2,
+        max_new_tokens=2,
     )
     assert isinstance(result, list)
     assert len(result) == 2
-    result = model(
-        {
-            "text": [
-                "<image>Describe this image in one sentence.",
-                "<image>Describe this image in one sentence.",
-            ],
-            "images": [images[0], images[1]],
-        }
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    result = model(
-        {
-            "text": [
-                "<image>Describe this image in one sentence.",
-                "<image>Describe this image in one sentence.",
-            ],
-            "images": [images[0], images[1]],
-        },
-        num_return_sequences=2,
-        num_beams=2,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert isinstance(item, list)
-        assert len(item) == 2
-
-
-def test_transformers_vision_batch_samples_constrained(model, images):
-    class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
-
-    result = model(
-        {"text": "<image>Describe this image in one sentence.", "images": images[0]},
-        Foo,
-        num_return_sequences=2,
-        num_beams=2,
-    )
-    assert isinstance(result, list)
-    assert len(result) == 2
-    for item in result:
-        assert item in ["cat", "dog"]
     result = model(
         {
             "text": [
@@ -209,12 +171,10 @@ class Foo(Enum):
             ],
             "images": [images[0], images[1]],
         },
-        Foo,
+        max_new_tokens=2,
     )
     assert isinstance(result, list)
     assert len(result) == 2
-    for item in result:
-        assert item in ["cat", "dog"]
     result = model(
         {
             "text": [
@@ -223,14 +183,12 @@ class Foo(Enum):
             ],
             "images": [images[0], images[1]],
         },
-        Foo,
         num_return_sequences=2,
         num_beams=2,
+        max_new_tokens=2,
     )
     assert isinstance(result, list)
     assert len(result) == 2
     for item in result:
         assert isinstance(item, list)
         assert len(item) == 2
-        assert item[0] in ["cat", "dog"]
-        assert item[1] in ["cat", "dog"]

From 4959875380cb1fb30f77d8aa110c6f63284b25f9 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 17 Jun 2025 23:07:20 +0200
Subject: [PATCH 351/505] Fix failing tests transformers

---
 pyproject.toml                               | 2 +-
 tests/models/test_transformers_multimodal.py | 8 ++++----
 tests/v0_legacy/test_function_legacy.py      | 2 +-
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index 19e51dd06..9216216d2 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -76,7 +76,7 @@ test = [
     "datasets",
     "anthropic",
     "google-genai",
-    "transformers",
+    "transformers<=4.52.1",
     "pillow",
     "jax",
     "flax",
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 73b5b538d..448bec221 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -9,7 +9,7 @@
 from PIL import Image
 from pydantic import BaseModel
 from transformers import (
-    AutoModelForVision2Seq,
+    LlavaForConditionalGeneration,
     AutoProcessor,
 )
 
@@ -21,7 +21,7 @@
 )
 from outlines.types import Regex
 
-TEST_MODEL = "HuggingFaceTB/SmolVLM-256M-Instruct"
+TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
 IMAGE_URLS = [
     "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
     "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
@@ -40,14 +40,14 @@ def img_from_url(url):
 @pytest.fixture
 def model():
     return outlines.from_transformers(
-        AutoModelForVision2Seq.from_pretrained(TEST_MODEL),
+        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
 
 
 def test_transformers_vision_instantiate_simple():
     model = outlines.from_transformers(
-        AutoModelForVision2Seq.from_pretrained(TEST_MODEL),
+        LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
     assert isinstance(model, TransformersMultiModal)
diff --git a/tests/v0_legacy/test_function_legacy.py b/tests/v0_legacy/test_function_legacy.py
index 24799a880..1fd03ebb9 100644
--- a/tests/v0_legacy/test_function_legacy.py
+++ b/tests/v0_legacy/test_function_legacy.py
@@ -26,7 +26,7 @@ class Foo(BaseModel):
             fn = Function(
                 test_template,
                 Foo,
-                "hf-internal-testing/tiny-random-GPTJForCausalLM",
+                "erwanf/gpt2-mini",
             )
 
             assert fn.generator is None

From 73962b8bec2fe9d717e78e1acf9e92882fa43ccc Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 7 Jun 2025 10:00:19 -0400
Subject: [PATCH 352/505] Improve the documentation index

---
 docs/index.md | 131 ++++++++++++++++++++++++++++----------------------
 1 file changed, 73 insertions(+), 58 deletions(-)

diff --git a/docs/index.md b/docs/index.md
index ab0a36515..1ae1999bb 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -1,19 +1,62 @@
 ---
 title: Welcome to Outlines!
+hide:
+  - navigation
 ---
 
-# Welcome
+#
 
-![](/assets/images/logo-light-mode.svg#only-light)
-![](/assets/images/logo-dark-mode.svg#only-dark)
+<figure markdown>
+![](/assets/images/logo-light-mode.svg#only-light){ width="500" }
+![](/assets/images/logo-dark-mode.svg#only-dark){ width="500" }
+</figure>
+
+LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
+
+Outlines guarantees structured outputs during generation — directly from any LLM.
+
+- **Works with any model** - Same code runs across OpenAI, Ollama, vLLM, and more
+- **Simple integration** - Just pass your desired output type: `model(prompt, output_type)`
+- **Guaranteed valid structure** - No more parsing headaches or broken JSON
+- **Provider independence** - Switch models without changing code
+- **Rich structure definition** - Use Json Schema, regular expressions or context-free grammars
+
+<figure markdown>
+[Get Started](guide/getting_started){ .md-button .md-button--primary }
+[View Examples](examples/){ .md-button }
+[API Reference](api_reference/){ .md-button }
+[GitHub](https://github.com/dottxt-ai/outlines){ .md-button }
+</figure>
+
+## See it in action
 
-Outlines is a Python library from [.TXT](https://dottxt.co) that guarantees structured output from
-large language models. It ensures LLMs speak the language of your application by making them follow specific formats such as JSON, regular expressions, or context-free grammars.
+```python
+from pydantic import BaseModel
+from typing import Literal
+import outlines
+import openai
 
-[Get started](/getting_started){ .md-button .md-button--primary }
-[Features](/features){ .md-button }
-[Examples](/examples){ .md-button }
-[GitHub](https://github.com/yourusername/outlines){ .md-button }
+class Customer(BaseModel):
+    name: str
+    urgency: Literal["high", "medium", "low"]
+    issue: str
+
+client = openai.OpenAI()
+model = outlines.from_openai(client, "gpt-4o")
+
+customer = model(
+    "Alice needs help with login issues ASAP",
+    Customer
+)
+# ✓ Always returns valid Customer object
+# ✓ No parsing, no errors, no retries
+```
+
+## Quick install
+
+```shell
+pip install outlines
+```
 
 ## Features
 
@@ -21,33 +64,27 @@ large language models. It ensures LLMs speak the language of your application by
 
 - :material-shield-check: **Reliable** - Guaranteed schema compliance -- always valid JSON.
 - :material-puzzle: **Feature-rich** - Supports a large proportion of the JSON Schema spec, along with regex and context-free grammars.
-- :material-lightning-bolt: **Fast** - Outlines has negligible runtime overhead, and fast compilation times.
-- :material-cog: **Universal** - Outlines is a powered by Rust, and can be easily bound to other languages.
+- :material-lightning-bolt: **Fast** - Microseconds of overhead vs seconds of retries. Compilation happens once, not every request.
 - :material-lightbulb: **Simple** - Outlines is a low-abstraction library. Write code the way you normally do with LLMs. No agent frameworks needed.
-- :material-magnify: **Powerful** - Manage prompt complexity with prompt templating.
 
 </div>
 
-## Supported models
-
-> [!NOTE]
-> Provide full model list with links to docs about each model
-
-- vLLM
-- Transformers
-- OpenAI
-
-## About .txt
-
-Outlines is built with ❤️ by [.txt](https://dottxt.co).
+## Supported inference APIs, libraries & servers
 
-.txt solves the critical problem of reliable structured output generation for large language models. Our commercially-licensed libraries ensure 100% compliance with JSON Schema, regular expressions and context-free grammars while adding only microseconds of latency. Unlike open-source alternatives, we offer superior reliability, performance, and enterprise support.
+- [vLLM](reference/models/vllm.md)
+- [Transformers](reference/models/transformers.md)
+- [llama.cpp](reference/models/llamacpp.md)
+- [Ollama](reference/models/ollama.md)
+- [MLX-LM](reference/models/mlxlm.md)
+- [SgLang](reference/models/sglang.md)
+- [TGI](reference/models/tgi.md)
+- [OpenAI](reference/models/openai.md)
+- [Anthropic](reference/models/anthropic.md)
+- [Gemini](reference/models/gemini.md)
 
 ## Who is using Outlines?
 
-Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) use Outlines.
-
-Prominent companies and organizations that use Outlines include:
+Hundreds of organisations and the main LLM serving frameworks ([vLLM][vllm], [TGI][tgi], [LoRAX][lorax], [xinference][xinference], [SGLang][sglang]) use Outlines. Prominent companies and organizations that use Outlines include:
 
 <div class="grid cards" markdown>
   <div class="row"><img src="../logos/amazon.png" width="200"></div>
@@ -77,34 +114,10 @@ Prominent companies and organizations that use Outlines include:
   <div class="row"><img src="../logos/upenn.png" width="200"></div>
 </div>
 
-</body>
-
 Organizations are included either because they use Outlines as a dependency in a public repository, or because of direct communication between members of the Outlines team and employees at these organizations.
 
 Still not convinced, read [what people say about us](community/feedback.md). And make sure to take a look at what the [community is building](community/examples.md)!
 
-## Philosophy
-
-**Outlines** is a library for neural text generation. You can think of it as a
-more flexible replacement for the `generate` method in the
-[transformers](https://github.com/huggingface/transformers) library.
-
-**Outlines** helps developers _structure text generation_ to build robust
-interfaces with external systems. It provides generation methods that
-guarantee that the output will match a regular expressions, or follow
-a JSON schema.
-
-**Outlines** provides _robust prompting primitives_ that separate the prompting
-from the execution logic and lead to simple implementations of few-shot
-generations, ReAct, meta-prompting, agents, etc.
-
-**Outlines** is designed as a _library_ that is meant to be compatible the
-broader ecosystem, not to replace it. We use as few abstractions as possible,
-and generation can be interleaved with control flow, conditionals, custom Python
-functions and calls to other libraries.
-
-**Outlines** is _compatible with every auto-regressive model_. It only interfaces with models
-via the next-token logits distribution.
 
 ## Outlines people
 
@@ -114,16 +127,17 @@ Outlines would not be what it is today without a community of dedicated develope
   <img src="https://contrib.rocks/image?repo=dottxt-ai/outlines" />
 </a>
 
+## About .txt
+
+Outlines is built with ❤️ by [.txt](https://dottxt.co).
+
+.txt solves the critical problem of reliable structured output generation for large language models. Our [commercially-licensed libraries][dottxt-doc] ensure 100% compliance with JSON Schema, regular expressions and context-free grammars while adding only microseconds of latency. Unlike open-source alternatives, we offer superior reliability, performance, and enterprise support.
+
+
 ## Acknowledgements
 
 <div class="grid" markdown>
 
-<figure markdown>
-  <a href="http://www.dottxt.co">
-  ![Normal Computing logo](assets/images/dottxt.png){ width="150" }
-  </a>
-</figure>
-
 <figure markdown>
   <a href="https://www.normalcomputing.ai">
   ![Normal Computing logo](assets/images/normal_computing.jpg){ width="150" }
@@ -145,3 +159,4 @@ Outlines was originally developed at [@NormalComputing](https://twitter.com/Norm
 [lorax]: https://github.com/predibase/lorax
 [xinference]: https://github.com/xorbitsai/inference
 [sglang]: https://github.com/sgl-project/sglang/
+[dottxt-doc]: https://docs.dottxt.co

From 2aeed061becbbbbdb52df3094de31020a765a184 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 18 Jun 2025 11:26:35 +0200
Subject: [PATCH 353/505] Add missing plugins to the requirements-doc

---
 outlines/v0_legacy/base.py | 4 ++--
 requirements-doc.txt       | 3 +++
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/outlines/v0_legacy/base.py b/outlines/v0_legacy/base.py
index 9a5396b33..e545b5bcf 100644
--- a/outlines/v0_legacy/base.py
+++ b/outlines/v0_legacy/base.py
@@ -76,7 +76,7 @@ def setup_numpy(self, signature: Optional[str] = None):
                 self._in_and_out_core_dimensions = self._parse_gufunc_signature(signature)
             else:
                 self._in_and_out_core_dimensions = None
-        except (ModuleNotFoundError, ImportError):
+        except (ModuleNotFoundError, ImportError): # pragma: no cover
             self.np = None
             self._calculate_shapes = None
             self._parse_gufunc_signature = None
@@ -87,7 +87,7 @@ def setup_numpy(self, signature: Optional[str] = None):
     def __call__(self, *args, **kwargs):
         """Call the vectorized function."""
         # Raise an error ourselves if NumPy is not installed.
-        if not self.np:
+        if not self.np: # pragma: no cover
             raise ImportError(
                 "NumPy is required to use the legacy version of the OpenAI "
                 "model."
diff --git a/requirements-doc.txt b/requirements-doc.txt
index 05f9f36ec..91f621d6c 100644
--- a/requirements-doc.txt
+++ b/requirements-doc.txt
@@ -6,4 +6,7 @@ mkdocs-section-index
 mkdocstrings[python]
 mkdocs-git-committers-plugin-2
 mkdocs-git-revision-date-localized-plugin
+mkdocs-redirects
+mkdocs-gen-files
+mkdocs-literate-nav
 mike

From 810818e0b29765bfc7e8ce677174473176c7e2e3 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Thu, 19 Jun 2025 10:53:15 +0200
Subject: [PATCH 354/505] Fix broken links in the README (#1625)

On top of the missing links highlighted in #1622, there were a bunch of
broken links in the README. This PR makes sure all links lead to
actually existing page. In some cases, the links do not lead to the
exact section it should, but that's a problem to fix in the
documentation in a separate PR. In the meantime, links at least lead to
a related page from which the user should be able to find their way.
---
 README.md | 25 ++++++++++++-------------
 1 file changed, 12 insertions(+), 13 deletions(-)

diff --git a/README.md b/README.md
index 0d85aebf1..390d51409 100644
--- a/README.md
+++ b/README.md
@@ -30,7 +30,6 @@ Need a high-performance commercial solution for structured outputs? Email us at
 - [Real-World Examples](#real-world-examples)
   - [🙋‍♂️ Customer Support Triage](#customer-support-triage)
   - [📦 E-commerce Product Categorization](#e-commerce-product-categorization)
-  - [💊 Pull Medication Data from Unstructured Notes](#pull-medication-data-from-unstructured-notes)
   - [📊 Parse Event Details with Incomplete Data](#parse-event-details-with-incomplete-data)
   - [🗂️ Categorize Documents into Predefined Types](#categorize-documents-into-predefined-types)
   - [📅 Schedule a Meeting with Function Calling](#schedule-a-meeting-with-function-calling)
@@ -551,7 +550,7 @@ Analyze the sentiment of the following {{ content_type }}:
 Provide your analysis as either "Positive", "Negative", or "Neutral".
 <|im_end>
 <|im_start>assistant
-"""
+""")
 
 # 2. Generate prompts with different parameters
 review = "This restaurant exceeded all my expectations. Fantastic service!"
@@ -585,27 +584,27 @@ print(few_shot_prompt)
 
 | Model type | Description | Documentation |
 |---------|-------------|:-------------:|
-| **Server Support** | vLLM and Ollama | [Server Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
-| **Local Model Support** | transformers and llama.cpp | [Model Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
-| **API Support** | OpenAI and Gemini | [API Integrations →](https://dottxt-ai.github.io/outlines/latest/installation) |
+| **Server Support** | vLLM and Ollama | [Server Integrations →](https://dottxt-ai.github.io/outlines/latest/features/models/) |
+| **Local Model Support** | transformers and llama.cpp | [Model Integrations →](https://dottxt-ai.github.io/outlines/latest/features/models/) |
+| **API Support** | OpenAI and Gemini | [API Integrations →](https://dottxt-ai.github.io/outlines/latest/features/models/) |
 
 ## Core Features
 
 | Feature | Description | Documentation |
 |---------|-------------|:-------------:|
-| **Multiple Choices** | Constrain outputs to predefined options | [Multiple Choices →](https://dottxt-ai.github.io/outlines/latest/guides/choice_generation) |
-| **️Function Calls** | Infer structure from function signatures | [Function Guide →](https://dottxt-ai.github.io/outlines/latest/guides/function_generation) |
-| **JSON/Pydantic** | Generate outputs matching JSON schemas | [JSON Guide →](https://dottxt-ai.github.io/outlines/latest/guides/json_generation) |
-| **Regular Expressions** | Generate text following a regex pattern | [Regex Guide →](https://dottxt-ai.github.io/outlines/latest/guides/regex_generation) |
-| **Grammars** | Enforce complex output structures | [Grammar Guide →](https://dottxt-ai.github.io/outlines/latest/guides/cfg_generation) |
+| **Multiple Choices** | Constrain outputs to predefined options | [Multiple Choices Guide →](https://dottxt-ai.github.io/outlines/latest/features/core/output_types/#multiple-choices) |
+| **Function Calls** | Infer structure from function signatures | [Function Guide →](https://dottxt-ai.github.io/outlines/latest/features/core/output_types/#json-schemas) |
+| **JSON/Pydantic** | Generate outputs matching JSON schemas | [JSON Guide →](https://dottxt-ai.github.io/outlines/latest/features/core/output_types/#json-schemas) |
+| **Regular Expressions** | Generate text following a regex pattern | [Regex Guide →](https://dottxt-ai.github.io/outlines/latest/features/core/output_types/#regex-patterns) |
+| **Grammars** | Enforce complex output structures | [Grammar Guide →](https://dottxt-ai.github.io/outlines/latest/features/core/output_types/#context-free-grammars) |
 
 ## Other Features
 
 | Feature | Description | Documentation |
 |---------|-------------|:-------------:|
-| **Prompt templates** | Separate complex prompts from code | |
-| **Custome types** | Intuitive interface to build complex types | |
-| **Applications** | Encapsulate templates and types into functions | |
+| **Prompt templates** | Separate complex prompts from code | [Template Guide →](https://dottxt-ai.github.io/outlines/latest/features/utility/template/) |
+| **Custome types** | Intuitive interface to build complex types | [Python Types Guide →](https://dottxt-ai.github.io/outlines/latest/features/core/output_types/#basic-python-types) |
+| **Applications** | Encapsulate templates and types into functions | [Application Guide →](https://dottxt-ai.github.io/outlines/latest/features/utility/application/) |
 
 ## About .txt
 

From 8b2227ed82f8d3c234d99224e06309fa7acc9a57 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Thu, 19 Jun 2025 12:17:28 +0200
Subject: [PATCH 355/505] Fix the broken links in the home page of the
 documentation (#1628)

Addresses #1623

2 additional things:
- The Outlines logo at the home page is not found when deployed while
it's fine in local. As the other images with relative paths work fine, I
changed the link to a relative path
- I removed many outdated redirects in the `mkdocs.yml` that were adding
noise in the build process of the documentation
---
 docs/index.md | 26 ++++++++++++++------------
 mkdocs.yml    | 40 ----------------------------------------
 2 files changed, 14 insertions(+), 52 deletions(-)

diff --git a/docs/index.md b/docs/index.md
index 1ae1999bb..8e90f2daf 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -7,8 +7,8 @@ hide:
 #
 
 <figure markdown>
-![](/assets/images/logo-light-mode.svg#only-light){ width="500" }
-![](/assets/images/logo-dark-mode.svg#only-dark){ width="500" }
+![](assets/images/logo-light-mode.svg#only-light){ width="500" }
+![](assets/images/logo-dark-mode.svg#only-dark){ width="500" }
 </figure>
 
 LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
@@ -71,16 +71,18 @@ pip install outlines
 
 ## Supported inference APIs, libraries & servers
 
-- [vLLM](reference/models/vllm.md)
-- [Transformers](reference/models/transformers.md)
-- [llama.cpp](reference/models/llamacpp.md)
-- [Ollama](reference/models/ollama.md)
-- [MLX-LM](reference/models/mlxlm.md)
-- [SgLang](reference/models/sglang.md)
-- [TGI](reference/models/tgi.md)
-- [OpenAI](reference/models/openai.md)
-- [Anthropic](reference/models/anthropic.md)
-- [Gemini](reference/models/gemini.md)
+- [vLLM](features/models/vllm.md)
+- [vLLM offline](features/models/vllm_offline.md)
+- [Transformers](features/models/transformers.md)
+- [llama.cpp](features/models/llamacpp.md)
+- [Ollama](features/models/ollama.md)
+- [MLX-LM](features/models/mlxlm.md)
+- [SgLang](features/models/sglang.md)
+- [TGI](features/models/tgi.md)
+- [OpenAI](features/models/openai.md)
+- [Anthropic](features/models/anthropic.md)
+- [Gemini](features/models/gemini.md)
+- [Dottxt](features/models/dottxt.md)
 
 ## Who is using Outlines?
 
diff --git a/mkdocs.yml b/mkdocs.yml
index ce075062b..3a6e71633 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -98,46 +98,6 @@ plugins:
     - redirects:
           redirect_maps:
               "welcome.md": "index.md"
-              "reference/chat_templating.md": "guide/chat_templating.md"
-              "blog.md": "https://blog.dottxt.co"
-              "cookbook/index.md": "examples/index.md"
-              "cookbook/classification.md": "examples/classification.md"
-              "cookbook/extraction.md": "examples/extraction.md"
-              "cookbook/dating_profiles.md": "examples/dating_profiles.md"
-              "cookbook/chain_of_density.md": "examples/chain_of_density.md"
-              "cookbook/models_playing_chess.md": "examples/models_playing_chess.md"
-              "cookbook/simtom.md": "examples/simtom.md"
-              "cookbook/qa-with-citations.md": "examples/qa-with-citations.md"
-              "cookbook/knowledge_graph_extraction.md": "examples/knowledge_graph_extraction.md"
-              "cookbook/structured_generation_workflow.md": "examples/structured_generation_workflow.md"
-              "cookbook/chain_of_thought.md": "examples/chain_of_thought.md"
-              "cookbook/react_agent.md": "examples/react_agent.md"
-              "cookbook/atomic_caption.md": "examples/atomic_caption.md"
-              "cookbook/read-pdfs.md": "examples/read-pdfs.md"
-              "cookbook/earnings-reports.md": "examples/earnings-reports.md"
-              "cookbook/receipt-digitization.md": "examples/receipt-digitization.md"
-              "cookbook/extract_event_details.md": "examples/extract_event_details.md"
-              "cookbook/deploy-using-bentoml.md": "examples/deploy-using-bentoml.md"
-              "cookbook/deploy-using-cerebrium.md": "examples/deploy-using-cerebrium.md"
-              "cookbook/deploy-using-modal.md": "examples/deploy-using-modal.md"
-
-              # Root files
-              "quickstart.md": "getting_started/index.md"
-
-              # API to features redirects
-              "api/index.md": "features/index.md"
-              "api/json_schema.md": "features/types/json_schema.md"
-              "api/regex.md": "features/types/regex.md"
-              "api/applications.md": "features/utility/application.md"
-              "api/templates.md": "features/utility/template.md"
-              "api/guide.md": "features/advanced/guide.md"
-              "api/parsing.md": "features/advanced/parsing.md"
-              "api/models.md": "features/available_models.md"
-
-              # API subdirectories to features
-              "api/types/": "features/types/index.md"
-              "api/core/": "features/index.md"
-              "api/advanced/": "features/index.md"
 
     - git-committers:
         repository: dottxt-ai/outlines

From 111667c036383b4193c3d5e1ca750ee560b683ff Mon Sep 17 00:00:00 2001
From: sukrucildirr <sukrucildirr@gmail.com>
Date: Thu, 19 Jun 2025 18:52:44 +0300
Subject: [PATCH 356/505] Fix typos across the codebase (#1631)

Fixed typos across codebase.
---
 docs/community/versioning.md                    |  2 +-
 docs/features/core/output_types.md              |  8 ++++----
 docs/features/models/transformers_multimodal.md |  2 +-
 docs/features/utility/regex_dsl.md              |  2 +-
 docs/features/utility/template.md               |  2 +-
 docs/guide/getting_started.md                   | 10 +++++-----
 outlines/caching.py                             |  2 +-
 outlines/models/gemini.py                       |  2 +-
 outlines/models/transformers.py                 |  4 ++--
 tests/fsm/test_parsing.py                       |  2 +-
 10 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/docs/community/versioning.md b/docs/community/versioning.md
index a6e12f218..9e70a7346 100644
--- a/docs/community/versioning.md
+++ b/docs/community/versioning.md
@@ -10,7 +10,7 @@ The Outlines project follows a structured versioning scheme designed to provide
 Each part of the version number (`major.minor.patch`) conveys information about the nature and impact of the changes included in the release.
 
 - **Major Releases** includes compatibility-breaking changes to core interfaces, such as `LogitsProcessor`s and `Guides`.
-- **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatability with existing use of core interfaces.
+- **Minor Releases** introduce changes of substance to internal or unexposed functionality. These changes are well tested and intended to maintain compatibility with existing use of core interfaces.
 - **Patch Releases** address bug fixes and incorporate low-risk changes to improve stability and performance.
 
 !!! note "Breaking Changes"
diff --git a/docs/features/core/output_types.md b/docs/features/core/output_types.md
index 3a49c2623..fd0a55b96 100644
--- a/docs/features/core/output_types.md
+++ b/docs/features/core/output_types.md
@@ -49,7 +49,7 @@ With an Outlines model, you can generate text that respects the type hints above
 ```python
 model("How many minutes are there in one hour", int) # "60"
 model("Pizza or burger", Literal["pizza", "burger"]) # "pizza"
-model("Create a character", Character, max_new_tokens=100) # '{"name": "James", "birth_date": "1980-05-10)", "skills": ["archery", "negociation"]}'
+model("Create a character", Character, max_new_tokens=100) # '{"name": "James", "birth_date": "1980-05-10)", "skills": ["archery", "negotiation"]}'
 ```
 
 An important difference with function type hints though is that an Outlines generator always returns a string.
@@ -70,7 +70,7 @@ We can group possible output types in several categories based on the use case t
 
 ### Basic Python Types
 
-The most straightfoward form of structed generation is to return an answer that conforms to a given basic type such as an int or a python list. You can use the basic Python types and the types from the `typing` library. For instance:
+The most straightforward form of structured generation is to return an answer that conforms to a given basic type such as an int or a python list. You can use the basic Python types and the types from the `typing` library. For instance:
 
 ```python
 from typing import Dict
@@ -88,7 +88,7 @@ from typing import Dict, List, Optional, Tuple, Union
 output_type = Dict[str, Union[int, str, List[Tuple[str, Optional[float]]]]]
 ```
 
-Values created with this output type would be dictionnaries with string as keys and values made of either an integer, a string or a list of two elements tuples: a string and either a float or None. Example of a valid response for text generated with this output type (it would be contained in a string):
+Values created with this output type would be dictionaries with string as keys and values made of either an integer, a string or a list of two elements tuples: a string and either a float or None. Example of a valid response for text generated with this output type (it would be contained in a string):
 
 ```json
 {
@@ -145,7 +145,7 @@ output_type = character
 
 There are two other JSON schema formats that require Outlines-specific classes: JSON schema strings and dictionaries.
 
-As those are contained in regular Python strings or dictionnaries, the associated output format would be ambiguous if they were to be provided directly. As a result, Outlines requires them to be wrapped in a `outlines.types.JsonSchema` object. For instance:
+As those are contained in regular Python strings or dictionaries, the associated output format would be ambiguous if they were to be provided directly. As a result, Outlines requires them to be wrapped in a `outlines.types.JsonSchema` object. For instance:
 
 ```python
 from outlines.types import JsonSchema
diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index b2b846f6c..6a6db8981 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -31,7 +31,7 @@ model = outlines.from_transformers(hf_model, hf_processor)
 
 ## Model Input
 
-Instead of a string or a list of strings (for batch generation), you should provide a dictionnary or a list of dictionnaries as a prompt when calling the `TransformersMultiModal` model. The dictionnary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
+Instead of a string or a list of strings (for batch generation), you should provide a dictionary or a list of dictionaries as a prompt when calling the `TransformersMultiModal` model. The dictionary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
 
 ```python
 {
diff --git a/docs/features/utility/regex_dsl.md b/docs/features/utility/regex_dsl.md
index 740a0772b..d6c1a68c2 100644
--- a/docs/features/utility/regex_dsl.md
+++ b/docs/features/utility/regex_dsl.md
@@ -240,7 +240,7 @@ The DSL comes "batteries included" with types that represent common text constru
 - `boolean` represents a boolean, "True" or "False" as recognized by `bool`
 - `number` represents a floating-point number recognize by Python's `float`
 - `date` represents a date as understood by `datetime.date`
-- `time` represents a time as undestood by `datetime.time`
+- `time` represents a time as understood by `datetime.time`
 - `datetime` represents a time as understoof by `datetime.datetime`
 - `digit` represents a single digit
 - `char` represents a single character
diff --git a/docs/features/utility/template.md b/docs/features/utility/template.md
index 631114f95..8a152cdf9 100644
--- a/docs/features/utility/template.md
+++ b/docs/features/utility/template.md
@@ -65,7 +65,7 @@ print(prompt)
 
 ## Custom Filters
 
-You can add custom filters to your Outlines template to extend the templating functionality. To do so, provide as second argument a dictionnary with filter names as keys and filter functions as values. The filter can then be used in your jinja2 template following the regular syntax. When rendering a prompt, the function will be applied to the associated variable.
+You can add custom filters to your Outlines template to extend the templating functionality. To do so, provide as second argument a dictionary with filter names as keys and filter functions as values. The filter can then be used in your jinja2 template following the regular syntax. When rendering a prompt, the function will be applied to the associated variable.
 
 For instance:
 
diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
index 80a6c2456..ed7525980 100644
--- a/docs/guide/getting_started.md
+++ b/docs/guide/getting_started.md
@@ -129,7 +129,7 @@ For a quick start, you can find below an example of how to initialize all suppor
     import outlines
     from openai import OpenAI
 
-    # You must have a separete SgLang server running
+    # You must have a separate SgLang server running
     # Create an OpenAI client with the base URL of the SgLang server
     openai_client = OpenAI(base_url="http://localhost:11434/v1")
 
@@ -145,7 +145,7 @@ For a quick start, you can find below an example of how to initialize all suppor
     import outlines
     from huggingface_hub import InferenceClient
 
-    # You must have a separete TGI server running
+    # You must have a separate TGI server running
     # Create an InferenceClient client with the base URL of the TGI server
     tgi_client = InferenceClient("http://localhost:8080")
 
@@ -189,7 +189,7 @@ For a quick start, you can find below an example of how to initialize all suppor
     import outlines
     from openai import OpenAI
 
-    # You must have a separete vLLM server running
+    # You must have a separate vLLM server running
     # Create an OpenAI client with the base URL of the VLLM server
     openai_client = OpenAI(base_url="http://localhost:11434/v1")
 
@@ -231,7 +231,7 @@ Supported output types can be organized in 5 categories:
 - [Regex](../../features/core/output_types#regex-patterns): through the Outlines's `Regex` object
 - [Context-free Grammarq](../../features/core/output_types#context-free-grammars): through the Outlines's `CFG` object
 
-Consult the section on [Output Types](../../features/core/output_types.md) in the features documention for more detailed information on all supported types for each output type category.
+Consult the section on [Output Types](../../features/core/output_types.md) in the features documentation for more detailed information on all supported types for each output type category.
 
 In the meantime, you can find below examples of using each of the five output type categories:
 
@@ -280,7 +280,7 @@ In the meantime, you can find below examples of using each of the five output ty
     # Generate a character
     result = model("Create a character", Character)
     print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
-    print(Characte.model_validate_json(result);) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
+    print(Character.model_validate_json(result);) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
     ```
 
 === "Regex"
diff --git a/outlines/caching.py b/outlines/caching.py
index 7cb2ed169..f134f216c 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -57,7 +57,7 @@ def get_cache():
     xdg_cache_home = os.environ.get("XDG_CACHE_HOME")
     home_dir = os.path.normpath(os.path.expanduser("~"))
     if outlines_cache_dir:
-        # OUTLINES_CACHE_DIR takes precendence
+        # OUTLINES_CACHE_DIR takes precedence
         cache_dir = outlines_cache_dir
     elif xdg_cache_home:  # pragma: no cover
         cache_dir = os.path.join(xdg_cache_home, ".cache", "outlines")
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index c66c984af..57cd60302 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -172,7 +172,7 @@ def format_list_output_type(self, output_type: Optional[Any]) -> dict:
                 )
 
         raise TypeError(
-            f"Gemini only supports homogenous lists: "
+            f"Gemini only supports homogeneous lists: "
             "list[BaseModel], list[TypedDict] or list[dataclass]. "
             f"Got {output_type} instead."
         )
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 01c7b34a5..e2071c406 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -245,7 +245,7 @@ def _prepare_model_inputs(
         model_input: Union[str, List[str], dict],
         output_type: Optional[OutlinesLogitsProcessor] = None,
     ) -> Tuple[Union[str, List[str]], dict]:
-        """Turn the user input into arguements to pass to the model"""
+        """Turn the user input into arguments to pass to the model"""
         prompts = self.type_adapter.format_input(model_input)
         input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
@@ -447,7 +447,7 @@ def _prepare_model_inputs(
         model_input: Union[str, List[str], dict],
         output_type: Optional[OutlinesLogitsProcessor] = None,
     ) -> Tuple[Union[str, List[str]], dict]:
-        """Turn the user input into arguements to pass to the model"""
+        """Turn the user input into arguments to pass to the model"""
         model_input = self.type_adapter.format_input(model_input)
         inputs = self.processor(
             **model_input, padding=True, return_tensors="pt"
diff --git a/tests/fsm/test_parsing.py b/tests/fsm/test_parsing.py
index b7446fa0c..a47943f3c 100644
--- a/tests/fsm/test_parsing.py
+++ b/tests/fsm/test_parsing.py
@@ -248,7 +248,7 @@ def _reconstruct_fsms(fsm, fsms_to_trans_finals):
                 for symbol in fsm.alphabet.by_transition.get(trans, [])
             }
             if symbols:
-                # NOTE: THIS RECONSTRUCTOR DOESNT WORK FOR MORE THAN ONE TRANSITION PER SYMBOL
+                # NOTE: THIS RECONSTRUCTOR DOESN'T WORK FOR MORE THAN ONE TRANSITION PER SYMBOL
                 assert len(symbols) == 1
                 symbol = list(symbols)[0]
                 alphabet[symbol] = trans_id

From adf546d197c0bf37a469b855b1154c2d47ff95b9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 19 Jun 2025 22:09:52 +0200
Subject: [PATCH 357/505] Document vLLM async support

---
 docs/features/models/vllm.md | 96 +++++++++++++++++++++++++++++++++++-
 1 file changed, 95 insertions(+), 1 deletion(-)

diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index 21a048f9f..3187e37e5 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -69,9 +69,77 @@ for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
     print(chunk) # 'In ...'
 ```
 
+## Asynchronous Calls
+
+vLLM supports asynchronous operations by passing an `AsyncOpenAI` client instead of a regular `OpenAI` client. This returns an `AsyncVLLM` model instance that supports async/await patterns.
+
+### Basic Async Generation
+
+```python
+import asyncio
+import openai
+import outlines
+
+async def generate_text():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_model = outlines.from_vllm(async_client)
+
+    result = await async_model("Write a haiku about Python.", max_tokens=50)
+    print(result)
+
+asyncio.run(generate_text())
+```
+
+### Async Streaming
+
+The async model also supports streaming with async iteration:
+
+```python
+import asyncio
+import openai
+import outlines
+
+async def stream_text():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_model = outlines.from_vllm(async_client)
+
+    async for chunk in async_model.stream("Tell me a story about a robot.", max_tokens=100):
+        print(chunk, end="")
+
+asyncio.run(stream_text())
+```
+
+### Concurrent Async Requests
+
+One of the main benefits of async calls is the ability to make multiple concurrent requests:
+
+```python
+import asyncio
+import openai
+import outlines
+
+async def generate_multiple():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_model = outlines.from_vllm(async_client)
+
+    prompts = [
+        "Write a tagline for a coffee shop.",
+        "Write a tagline for a bookstore.",
+        "Write a tagline for a gym."
+    ]
+
+    tasks = [async_model(prompt, max_tokens=30) for prompt in prompts]
+    results = await asyncio.gather(*tasks)
+
+    for prompt, result in zip(prompts, results):
+        print(f"{prompt}\n{result}\n")
+
+asyncio.run(generate_multiple())
+```
+
 ## Structured Generation
 
-vLLM supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
+vLLM supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model. All structured generation features work with both synchronous and asynchronous models.
 
 ### Simple Type
 
@@ -177,6 +245,32 @@ result = model("Write an addition.", output_type, extra_body={"guided_decoding_b
 print(result) # '23 + 48'
 ```
 
+### Async Structured Generation
+
+All structured generation features work seamlessly with async models:
+
+```python
+import asyncio
+import openai
+import outlines
+from pydantic import BaseModel
+
+class User(BaseModel):
+    name: str
+    email: str
+    age: int
+
+async def generate_user():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_model = outlines.from_vllm(async_client)
+
+    result = await async_model("Generate a random user profile.", output_type=User)
+    user = User.model_validate_json(result)
+    print(f"Name: {user.name}, Email: {user.email}, Age: {user.age}")
+
+asyncio.run(generate_user())
+```
+
 ## Inference Arguments
 
 When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.

From 0b57415170cf0a8ea66ee0cb9b8bc5ea0a5f8025 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 19 Jun 2025 22:24:33 +0200
Subject: [PATCH 358/505] Document TGI async support

---
 docs/features/models/tgi.md | 103 +++++++++++++++++++++++++++++++++++-
 1 file changed, 101 insertions(+), 2 deletions(-)

diff --git a/docs/features/models/tgi.md b/docs/features/models/tgi.md
index 8fb807580..38b1a1ae5 100644
--- a/docs/features/models/tgi.md
+++ b/docs/features/models/tgi.md
@@ -50,7 +50,9 @@ model = outlines.from_tgi(client)
 # Call it to generate text
 result = model("Write a short story about a cat.", stop_sequences=["."])
 print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
-```The `TGI` model supports streaming. For instance:
+```
+
+The `TGI` model supports streaming. For instance:
 
 ```python
 import outlines
@@ -65,9 +67,79 @@ for chunk in model.stream("Write a short story about a cat.", stop_sequences=[".
     print(chunk) # 'In ...'
 ```
 
+## Asynchronous Calls
+
+TGI supports asynchronous operations by passing an `AsyncInferenceClient` instead of a regular `InferenceClient`. This returns an `AsyncTGI` model instance that supports async/await patterns.
+
+### Basic Async Generation
+
+```python
+import asyncio
+import outlines
+import huggingface_hub
+
+async def generate_text():
+    # Create an async model
+    async_client = huggingface_hub.AsyncInferenceClient("http://localhost:11434")
+    async_model = outlines.from_tgi(async_client)
+
+    result = await async_model("Write a haiku about Python.", max_new_tokens=50)
+    print(result)
+
+asyncio.run(generate_text())
+```
+
+### Async Streaming
+
+The async model also supports streaming with async iteration:
+
+```python
+import asyncio
+import outlines
+import huggingface_hub
+
+async def stream_text():
+    async_client = huggingface_hub.AsyncInferenceClient("http://localhost:11434")
+    async_model = outlines.from_tgi(async_client)
+
+    async for chunk in async_model.stream("Tell me a story about a robot.", max_new_tokens=100):
+        print(chunk, end="")
+
+asyncio.run(stream_text())
+```
+
+### Concurrent Async Requests
+
+One of the main benefits of async calls is the ability to make multiple concurrent requests:
+
+```python
+import asyncio
+import outlines
+import huggingface_hub
+
+async def generate_multiple():
+    async_client = huggingface_hub.AsyncInferenceClient("http://localhost:11434")
+    async_model = outlines.from_tgi(async_client)
+
+    # Define multiple prompts
+    prompts = [
+        "Write a tagline for a coffee shop.",
+        "Write a tagline for a bookstore.",
+        "Write a tagline for a gym."
+    ]
+
+    tasks = [async_model(prompt, max_new_tokens=30) for prompt in prompts]
+    results = await asyncio.gather(*tasks)
+
+    for prompt, result in zip(prompts, results):
+        print(f"{prompt}\n{result}\n")
+
+asyncio.run(generate_multiple())
+```
+
 ## Structured Generation
 
-TGI supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
+TGI supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model. All structured generation features work with both synchronous and asynchronous models.
 
 ### Simple Type
 
@@ -130,6 +202,33 @@ model = outlines.from_tgi(tgi_client)
 result = model("Generate a fake social security number.", output_type, top_p=0.1)
 print(result) # '782-32-3789'
 ```
+
+### Async Structured Generation
+
+All structured generation features work seamlessly with async models:
+
+```python
+import asyncio
+import outlines
+import huggingface_hub
+from pydantic import BaseModel
+
+class User(BaseModel):
+    name: str
+    email: str
+    age: int
+
+async def generate_user():
+    async_client = huggingface_hub.AsyncInferenceClient("http://localhost:11434")
+    async_model = outlines.from_tgi(async_client)
+
+    result = await async_model("Generate a random user profile.", output_type=User)
+    user = User.model_validate_json(result)
+    print(f"Name: {user.name}, Email: {user.email}, Age: {user.age}")
+
+asyncio.run(generate_user())
+```
+
 ## Inference parameters
 
 When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `text_generation` method of the TGI client.

From bdc5a97d3c0d1ea1607b87ba80e541d134dc304e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 19 Jun 2025 22:24:54 +0200
Subject: [PATCH 359/505] Document SGLang async support

---
 docs/features/models/sglang.md        |  97 +++++++-
 docs/guide/fastapi_vllm_deployment.md | 314 ++++++++++++++++++++++++++
 mkdocs.yml                            |   1 +
 3 files changed, 411 insertions(+), 1 deletion(-)
 create mode 100644 docs/guide/fastapi_vllm_deployment.md

diff --git a/docs/features/models/sglang.md b/docs/features/models/sglang.md
index 95a6eb7b5..0df6a54db 100644
--- a/docs/features/models/sglang.md
+++ b/docs/features/models/sglang.md
@@ -69,9 +69,77 @@ for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
     print(chunk) # 'In ...'
 ```
 
+## Asynchronous Calls
+
+SGLang supports asynchronous operations by passing an `AsyncOpenAI` client instead of a regular `OpenAI` client. This returns an `AsyncSGLang` model instance that supports async/await patterns.
+
+### Basic Async Generation
+
+```python
+import asyncio
+import openai
+import outlines
+
+async def generate_text():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
+    async_model = outlines.from_sglang(async_client)
+
+    result = await async_model("Write a haiku about Python.", max_tokens=50)
+    print(result)
+
+asyncio.run(generate_text())
+```
+
+### Async Streaming
+
+The async model also supports streaming with async iteration:
+
+```python
+import asyncio
+import openai
+import outlines
+
+async def stream_text():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
+    async_model = outlines.from_sglang(async_client)
+
+    async for chunk in async_model.stream("Tell me a story about a robot.", max_tokens=100):
+        print(chunk, end="")
+
+asyncio.run(stream_text())
+```
+
+### Concurrent Async Requests
+
+One of the main benefits of async calls is the ability to make multiple concurrent requests:
+
+```python
+import asyncio
+import openai
+import outlines
+
+async def generate_multiple():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
+    async_model = outlines.from_sglang(async_client)
+
+    prompts = [
+        "Write a tagline for a coffee shop.",
+        "Write a tagline for a bookstore.",
+        "Write a tagline for a gym."
+    ]
+
+    tasks = [async_model(prompt, max_tokens=30) for prompt in prompts]
+    results = await asyncio.gather(*tasks)
+
+    for prompt, result in zip(prompts, results):
+        print(f"{prompt}\n{result}\n")
+
+asyncio.run(generate_multiple())
+```
+
 ## Structured Generation
 
-SGLang supports all output types available in Outlines (context-free grammars with caveats though, see the subsection below for more details). Simply provide an `output_type` after the prompt when calling the model.
+SGLang supports all output types available in Outlines (context-free grammars with caveats though, see the subsection below for more details). Simply provide an `output_type` after the prompt when calling the model. All structured generation features work with both synchronous and asynchronous models.
 
 ### Simple Type
 
@@ -162,6 +230,33 @@ result = model("Is the weather good today?", output_type)
 print(result) # 'yes'
 ```
 
+### Async Structured Generation
+
+All structured generation features work seamlessly with async models:
+
+```python
+import asyncio
+import openai
+import outlines
+from typing import List
+from pydantic import BaseModel
+
+class User(BaseModel):
+    name: str
+    email: str
+    age: int
+
+async def generate_user():
+    async_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
+    async_model = outlines.from_sglang(async_client)
+
+    result = await async_model("Generate a random user profile.", output_type=User)
+    user = User.model_validate_json(result)
+    print(f"Name: {user.name}, Email: {user.email}, Age: {user.age}")
+
+asyncio.run(generate_user())
+```
+
 ## Inference Arguments
 
 When calling the model, you can provide optional parameters on top of the prompt and the output type. Those will be passed on to the `chat.completions.create` method of the OpenAI client.
diff --git a/docs/guide/fastapi_vllm_deployment.md b/docs/guide/fastapi_vllm_deployment.md
new file mode 100644
index 000000000..d3bb1b836
--- /dev/null
+++ b/docs/guide/fastapi_vllm_deployment.md
@@ -0,0 +1,314 @@
+---
+title: Deploying with FastAPI
+---
+
+# Deploying with FastAPI
+
+This guide demonstrates how to build a FastAPI application that leverages Outlines' async integration with vLLM. We'll create a customer support API that can intelligently categorize tickets and generate structured responses.
+
+## Prerequisites
+
+Before starting, ensure you have:
+
+1. A vLLM server running (locally or remotely)
+2. Python 3.8+ installed
+3. The following packages installed:
+
+```bash
+pip install fastapi uvicorn outlines openai pydantic
+```
+
+## Building the Application
+
+### Step 1: Define Data Models
+
+First, let's define our Pydantic models for structured outputs:
+
+```python
+# models.py
+from enum import Enum
+from typing import List
+from pydantic import BaseModel, Field
+
+class TicketCategory(str, Enum):
+    BILLING = "billing"
+    TECHNICAL = "technical"
+    ACCOUNT = "account"
+    PRODUCT = "product"
+    OTHER = "other"
+
+class TicketPriority(str, Enum):
+    LOW = "low"
+    MEDIUM = "medium"
+    HIGH = "high"
+    URGENT = "urgent"
+
+class TicketAnalysis(BaseModel):
+    category: TicketCategory
+    priority: TicketPriority
+    summary: str = Field(description="Brief summary of the issue")
+    customer_sentiment: str = Field(description="Customer emotional state")
+    key_issues: List[str] = Field(description="List of main problems")
+    requires_human: bool = Field(description="Whether this needs human intervention")
+
+class SupportResponse(BaseModel):
+    greeting: str
+    acknowledgment: str = Field(description="Acknowledge the customer's issue")
+    solution_steps: List[str] = Field(description="Steps to resolve the issue")
+    closing: str
+```
+
+### Step 2: Create the FastAPI Application
+
+Now let's create our FastAPI application with async vLLM integration:
+
+```python
+# main.py
+import asyncio
+from contextlib import asynccontextmanager
+from typing import Optional
+
+import openai
+import outlines
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+
+from models import TicketAnalysis, SupportResponse
+
+# Request model
+class TicketRequest(BaseModel):
+    customer_id: str
+    message: str
+
+# Global model instance
+async_model = None
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    """Initialize the async vLLM model on startup."""
+    global async_model
+
+    client = openai.AsyncOpenAI(
+        base_url="http://localhost:8000/v1",  # Adjust to your vLLM server URL
+        api_key="dummy"  # vLLM doesn't require a real API key
+    )
+    async_model = outlines.from_vllm(client)
+
+    yield
+
+    async_model = None  # Cleanup
+
+# Create FastAPI app
+app = FastAPI(
+    title="Customer Support Assistant API",
+    description="AI-powered customer support with structured outputs",
+    version="1.0.0",
+    lifespan=lifespan
+)
+
+
+@app.get("/health")
+async def health_check():
+    """Check if the service is running and vLLM is accessible."""
+    if async_model is None:
+        raise HTTPException(status_code=503, detail="Model not initialized")
+    return {"status": "healthy"}
+
+
+@app.post("/analyze-ticket", response_model=TicketAnalysis)
+async def analyze_ticket(request: TicketRequest):
+    """Analyze a customer support ticket and extract structured information."""
+    if async_model is None:
+        raise HTTPException(status_code=503, detail="Model not initialized")
+
+    prompt = f"""Analyze this customer support ticket:
+
+Customer ID: {request.customer_id}
+Message: {request.message}
+
+Extract the category, priority, and other relevant information."""
+
+    try:
+        # Use async model with structured output
+        result = await async_model(
+            prompt,
+            output_type=TicketAnalysis,
+            max_tokens=500,
+            temperature=0.1
+        )
+
+        # Parse and return the result
+        analysis = TicketAnalysis.model_validate_json(result)
+        return analysis
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Analysis failed: {str(e)}")
+
+
+@app.post("/generate-response", response_model=SupportResponse)
+async def generate_response(
+    request: TicketRequest,
+    analysis: TicketAnalysis
+):
+    """Generate a structured support response based on ticket analysis."""
+    if async_model is None:
+        raise HTTPException(status_code=503, detail="Model not initialized")
+
+    prompt = f"""Generate a professional customer support response.
+
+Customer Message: {request.message}
+Category: {analysis.category}
+Priority: {analysis.priority}
+Customer Sentiment: {analysis.customer_sentiment}
+
+Create a helpful, empathetic response that addresses their concerns."""
+
+    try:
+        # Generate structured response
+        result = await async_model(
+            prompt,
+            output_type=SupportResponse,
+            max_tokens=800,
+            temperature=0.7
+        )
+
+        response = SupportResponse.model_validate_json(result)
+        return response
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Response generation failed: {str(e)}")
+```
+
+## Running the Application
+
+### Step 1: Start your vLLM server
+
+```bash
+# Example: Start vLLM with a model
+python -m vllm.entrypoints.openai.api_server \
+    --model meta-llama/Llama-2-7b-chat-hf \
+    --port 8000
+```
+
+### Step 2: Run the FastAPI application
+
+```bash
+# Development
+uvicorn main:app --reload --host 0.0.0.0 --port 8080
+
+# Production with multiple workers
+uvicorn main:app --workers 4 --host 0.0.0.0 --port 8080
+```
+
+## Testing the API
+
+### Example 1: Analyze a support ticket
+
+```bash
+curl -X POST "http://localhost:8080/analyze-ticket" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "customer_id": "CUST123",
+    "message": "I have been charged twice for my subscription this month. This is unacceptable and I want a refund immediately!"
+  }'
+```
+
+Expected response:
+```json
+{
+  "category": "billing",
+  "priority": "high",
+  "summary": "Customer charged twice for subscription, requesting refund",
+  "customer_sentiment": "angry",
+  "key_issues": ["duplicate charge", "subscription billing", "refund request"],
+  "requires_human": false
+}
+```
+
+### Example 2: Generate a support response
+
+```bash
+# First, get the analysis
+ANALYSIS=$(curl -s -X POST "http://localhost:8080/analyze-ticket" \
+  -H "Content-Type: application/json" \
+  -d '{
+    "customer_id": "CUST456",
+    "message": "My app keeps crashing when I try to upload photos."
+  }')
+
+# Then generate a response
+curl -X POST "http://localhost:8080/generate-response" \
+  -H "Content-Type: application/json" \
+  -d "{
+    \"request\": {
+      \"customer_id\": \"CUST456\",
+      \"message\": \"My app keeps crashing when I try to upload photos.\"
+    },
+    \"analysis\": $ANALYSIS
+  }"
+```
+
+By combining FastAPI's async capabilities with Outlines' structured generation, you can build robust APIs that leverage large language models while maintaining high performance and reliability.
+
+## Using Alternative Backends: SGLang and TGI
+
+One of the key advantages of Outlines is its unified API across different inference backends. You can easily switch from vLLM to SGLang or TGI with minimal code changes - just modify the model initialization in the `lifespan` function.
+
+### Using SGLang Instead of vLLM
+
+To use SGLang, simply change the client initialization:
+
+```python
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    """Initialize the async SGLang model on startup."""
+    global async_model
+
+    client = openai.AsyncOpenAI(
+        base_url="http://localhost:30000/v1",  # SGLang server URL
+        api_key="dummy"
+    )
+    async_model = outlines.from_sglang(client)
+
+    yield
+
+    async_model = None
+```
+
+Start your SGLang server with:
+```bash
+python -m sglang.launch_server \
+    --model-path meta-llama/Llama-2-7b-chat-hf \
+    --port 30000
+```
+
+### Using TGI Instead of vLLM
+
+For TGI (Text Generation Inference), use the Hugging Face client:
+
+```python
+import huggingface_hub
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    """Initialize the async TGI model on startup."""
+    global async_model
+
+    client = huggingface_hub.AsyncInferenceClient(
+        "http://localhost:8080"  # TGI server URL
+    )
+    async_model = outlines.from_tgi(client)
+
+    yield
+
+    async_model = None
+```
+
+Start your TGI server with:
+```bash
+docker run --gpus all -p 8080:80 \
+    ghcr.io/huggingface/text-generation-inference:latest \
+    --model-id meta-llama/Llama-2-7b-chat-hf
+```
+
+The rest of your FastAPI application - all the endpoints, error handling, and business logic - remains completely unchanged. This flexibility allows you to test different inference engines without rewriting your application.
diff --git a/mkdocs.yml b/mkdocs.yml
index 3a6e71633..1f0cf527c 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -119,6 +119,7 @@ nav:
           - Getting Started: guide/getting_started.md
           - Installation: guide/installation.md
           - Migrating to v1: guide/migration.md
+          - Deploying with FastAPI: guide/fastapi_vllm_deployment.md
           - Licence: licence.md
           - Utility:
                 - Chat Templating for Instruct Models: guide/chat_templating.md

From cb894006a531b727760970499c0821553e3b77f7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 19 Jun 2025 22:58:01 +0200
Subject: [PATCH 360/505] Style the code blocks

---
 docs/stylesheets/extra.css | 70 ++++++++++++++++++++++++++++++++++++++
 mkdocs.yml                 |  1 +
 2 files changed, 71 insertions(+)

diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index 381f5aa11..e5fa4115a 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -20,3 +20,73 @@
     --md-code-font: "Source Code Pro Custom";
     /* --md-typeset-a-color: #d36237; /*this is the brand color */
 }
+
+/* Nord theme code block styling */
+.highlight pre,
+.md-typeset pre code,
+.md-typeset .highlight pre,
+.md-typeset .highlighttable pre {
+    background-color: #2E3440 !important; /* Nord's darkest blue (nord0) */
+    border-radius: 0.5rem !important; /* Rounded corners */
+    box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -1px rgba(0, 0, 0, 0.06) !important; /* Subtle shadow */
+}
+
+/* Ensure inline code also uses appropriate background */
+.md-typeset code {
+    background-color: #3B4252; /* Nord's slightly lighter shade (nord1) */
+    color: #D8DEE9; /* Nord's snow storm shade for better contrast */
+    border-radius: 0.2rem; /* Smaller radius for inline code */
+    padding: 0.1em 0.3em; /* Add some padding for better appearance */
+}
+
+/* Override code block container background */
+.md-typeset .highlight,
+.md-typeset .highlighttable {
+    background-color: #2E3440 !important;
+    border-radius: 0.5rem !important;
+    overflow: hidden; /* Ensure child elements respect rounded corners */
+}
+
+/* Ensure proper spacing and shadow for the entire code block */
+.md-typeset pre {
+    margin: 1em 0 !important;
+    box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -1px rgba(0, 0, 0, 0.06) !important;
+}
+
+/* Style the copy button with Nord snow grey */
+.md-clipboard {
+    color: #D8DEE9 !important; /* Nord snow storm */
+}
+
+.md-clipboard:hover {
+    color: #ECEFF4 !important; /* Brighter snow storm on hover */
+}
+
+.md-clipboard:after {
+    color: #D8DEE9 !important;
+}
+
+/* Style scrollbars with Nord colors */
+.md-typeset pre::-webkit-scrollbar {
+    height: 0.4rem;
+    width: 0.4rem;
+}
+
+.md-typeset pre::-webkit-scrollbar-track {
+    background-color: #3B4252; /* Nord1 */
+}
+
+.md-typeset pre::-webkit-scrollbar-thumb {
+    background-color: #4C566A; /* Nord3 */
+    border-radius: 0.2rem;
+}
+
+.md-typeset pre::-webkit-scrollbar-thumb:hover {
+    background-color: #D8DEE9; /* Nord snow storm on hover */
+}
+
+/* Firefox scrollbar styling */
+.md-typeset pre {
+    scrollbar-width: thin;
+    scrollbar-color: #4C566A #3B4252;
+}
diff --git a/mkdocs.yml b/mkdocs.yml
index 1f0cf527c..336715702 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -64,6 +64,7 @@ markdown_extensions:
           line_spans: __span
           pygments_lang_class: true
           noclasses: True
+          pygments_style: nord
     - pymdownx.superfences:
           custom_fences:
               - name: mermaid

From 1b90c21776b3311e95dfe1d3064a31eeb5140f60 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 19 Jun 2025 23:08:57 +0200
Subject: [PATCH 361/505] Change line heights and section styling

---
 docs/stylesheets/extra.css | 105 +++++++++++++++++++++++++++++++++++++
 1 file changed, 105 insertions(+)

diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index e5fa4115a..fb6123f04 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -90,3 +90,108 @@
     scrollbar-width: thin;
     scrollbar-color: #4C566A #3B4252;
 }
+
+/* Better visual hierarchy with bolder headings */
+.md-typeset h1 {
+    font-weight: 700;
+    font-size: 2.2rem;
+    line-height: 1.2;
+    margin-top: 0;
+    margin-bottom: 1.5rem;
+    color: #1a1a1a;
+}
+
+.md-typeset h2 {
+    font-weight: 600;
+    font-size: 1.75rem;
+    line-height: 1.3;
+    margin-top: 2.5rem;
+    margin-bottom: 1.25rem;
+    color: #2a2a2a;
+    border-bottom: 2px solid #e0e0e0;
+    padding-bottom: 0.4rem;
+}
+
+.md-typeset h3 {
+    font-weight: 600;
+    font-size: 1.35rem;
+    line-height: 1.3;
+    margin-top: 2rem;
+    margin-bottom: 0.75rem;
+    color: #333;
+}
+
+.md-typeset h4 {
+    font-weight: 500;
+    font-size: 1.15rem;
+    line-height: 1.3;
+    margin-top: 1.75rem;
+    margin-bottom: 0.6rem;
+    color: #444;
+}
+
+.md-typeset h5 {
+    font-weight: 500;
+    font-size: 1.05rem;
+    line-height: 1.4;
+    margin-top: 1.5rem;
+    margin-bottom: 0.5rem;
+    color: #555;
+}
+
+.md-typeset h6 {
+    font-weight: 500;
+    font-size: 0.95rem;
+    line-height: 1.4;
+    margin-top: 1.25rem;
+    margin-bottom: 0.5rem;
+    color: #666;
+    text-transform: uppercase;
+    letter-spacing: 0.05em;
+}
+
+/* Improve paragraph spacing for better readability */
+.md-typeset p {
+    line-height: 1.55;
+    margin-bottom: 1rem;
+}
+
+/* Make navigation items bolder */
+.md-nav__link {
+    font-weight: 400;
+}
+
+.md-nav__link--active,
+.md-nav__link:hover {
+    font-weight: 600;
+    color: #1a1a1a;
+}
+
+/* Improve list spacing */
+.md-typeset ul,
+.md-typeset ol {
+    margin-bottom: 1.5rem;
+}
+
+.md-typeset li {
+    line-height: 1.55;
+    margin-bottom: 0.4rem;
+}
+
+/* Make important elements stand out */
+.md-typeset strong {
+    font-weight: 600;
+    color: #1a1a1a;
+}
+
+/* Better spacing for code blocks in relation to text */
+.md-typeset pre {
+    margin: 1.5em 0 !important;
+}
+
+/* Improve table headers */
+.md-typeset table th {
+    font-weight: 600;
+    background-color: #f5f5f5;
+    color: #1a1a1a;
+}

From 8b023d865334852fdcf60688f91effc6592cdd8e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 20 Jun 2025 00:20:33 +0200
Subject: [PATCH 362/505] Add guide on deployment with FastAPI

---
 docs/guide/fastapi_vllm_deployment.md | 127 +++++++++++++-------------
 1 file changed, 64 insertions(+), 63 deletions(-)

diff --git a/docs/guide/fastapi_vllm_deployment.md b/docs/guide/fastapi_vllm_deployment.md
index d3bb1b836..d87886ae7 100644
--- a/docs/guide/fastapi_vllm_deployment.md
+++ b/docs/guide/fastapi_vllm_deployment.md
@@ -4,17 +4,13 @@ title: Deploying with FastAPI
 
 # Deploying with FastAPI
 
-This guide demonstrates how to build a FastAPI application that leverages Outlines' async integration with vLLM. We'll create a customer support API that can intelligently categorize tickets and generate structured responses.
+This guide demonstrates how to build a FastAPI application that leverages Outlines' async integration with vLLM. We create a customer support API that can intelligently categorize tickets and generate structured responses.
 
 ## Prerequisites
 
-Before starting, ensure you have:
+Before starting, ensure you have a vLLM server running (locally or remotely) and the following packages installed:
 
-1. A vLLM server running (locally or remotely)
-2. Python 3.8+ installed
-3. The following packages installed:
-
-```bash
+```shell
 pip install fastapi uvicorn outlines openai pydantic
 ```
 
@@ -58,7 +54,33 @@ class SupportResponse(BaseModel):
     closing: str
 ```
 
-### Step 2: Create the FastAPI Application
+### Step 2: Define the prompts
+
+Let us now write the prompts that we will be using in our application, using Jinja 2's templating language. We separate them from the application implementation so they are easier to modify and version.
+
+```ascii
+{# prompts/categorize.txt #}
+Analyze this customer support ticket:
+
+Customer ID: {{ customer_id }}
+Message: {{ message }}
+
+Extract the category, priority, and other relevant information.
+```
+
+```ascii
+{# prompts/respond.txt #}
+Generate a professional customer support response.
+
+Customer Message: {{ message }}
+Category: {{ category }}
+Priority: {{  priority }}
+Customer Sentiment: {{ customer_sentiment }}
+
+Create a helpful, empathetic response that addresses their concerns.
+```
+
+### Step 3: Create the FastAPI Application
 
 Now let's create our FastAPI application with async vLLM integration:
 
@@ -69,7 +91,7 @@ from contextlib import asynccontextmanager
 from typing import Optional
 
 import openai
-import outlines
+from outlines import models, Template
 from fastapi import FastAPI, HTTPException
 from pydantic import BaseModel
 
@@ -83,6 +105,8 @@ class TicketRequest(BaseModel):
 # Global model instance
 async_model = None
 
+# The lifespan function is a FastAPI construct
+# used to define startup and shutdown logic for the API.
 @asynccontextmanager
 async def lifespan(app: FastAPI):
     """Initialize the async vLLM model on startup."""
@@ -92,7 +116,7 @@ async def lifespan(app: FastAPI):
         base_url="http://localhost:8000/v1",  # Adjust to your vLLM server URL
         api_key="dummy"  # vLLM doesn't require a real API key
     )
-    async_model = outlines.from_vllm(client)
+    async_model = models.from_vllm(client)
 
     yield
 
@@ -107,38 +131,23 @@ app = FastAPI(
 )
 
 
-@app.get("/health")
-async def health_check():
-    """Check if the service is running and vLLM is accessible."""
-    if async_model is None:
-        raise HTTPException(status_code=503, detail="Model not initialized")
-    return {"status": "healthy"}
-
-
 @app.post("/analyze-ticket", response_model=TicketAnalysis)
 async def analyze_ticket(request: TicketRequest):
     """Analyze a customer support ticket and extract structured information."""
     if async_model is None:
         raise HTTPException(status_code=503, detail="Model not initialized")
 
-    prompt = f"""Analyze this customer support ticket:
-
-Customer ID: {request.customer_id}
-Message: {request.message}
-
-Extract the category, priority, and other relevant information."""
+    template = Template.from_file("prompts/categorize.txt")
+    prompt = template(
+        customer_id=request.customer_id,
+        message=request.message
+    )
 
     try:
-        # Use async model with structured output
-        result = await async_model(
-            prompt,
-            output_type=TicketAnalysis,
-            max_tokens=500,
-            temperature=0.1
-        )
-
-        # Parse and return the result
+        # Generate and parse a structured response
+        result = await async_model(prompt, TicketAnalysis, max_tokens=5000)
         analysis = TicketAnalysis.model_validate_json(result)
+
         return analysis
 
     except Exception as e:
@@ -154,25 +163,19 @@ async def generate_response(
     if async_model is None:
         raise HTTPException(status_code=503, detail="Model not initialized")
 
-    prompt = f"""Generate a professional customer support response.
-
-Customer Message: {request.message}
-Category: {analysis.category}
-Priority: {analysis.priority}
-Customer Sentiment: {analysis.customer_sentiment}
-
-Create a helpful, empathetic response that addresses their concerns."""
+    template = Template.from_file("prompts/respond.txt")
+    prompt = template(
+        message=request.message,
+        category=analysis.category,
+        priority=analysis.priority,
+        customer_sentiment=analysis.customer_sentiment
+    )
 
     try:
-        # Generate structured response
-        result = await async_model(
-            prompt,
-            output_type=SupportResponse,
-            max_tokens=800,
-            temperature=0.7
-        )
-
+        # Generate and parse a structured response
+        result = await async_model(prompt, SupportResponse, max_tokens=800)
         response = SupportResponse.model_validate_json(result)
+
         return response
 
     except Exception as e:
@@ -183,8 +186,7 @@ Create a helpful, empathetic response that addresses their concerns."""
 
 ### Step 1: Start your vLLM server
 
-```bash
-# Example: Start vLLM with a model
+```shell
 python -m vllm.entrypoints.openai.api_server \
     --model meta-llama/Llama-2-7b-chat-hf \
     --port 8000
@@ -192,19 +194,15 @@ python -m vllm.entrypoints.openai.api_server \
 
 ### Step 2: Run the FastAPI application
 
-```bash
-# Development
+```shell
 uvicorn main:app --reload --host 0.0.0.0 --port 8080
-
-# Production with multiple workers
-uvicorn main:app --workers 4 --host 0.0.0.0 --port 8080
 ```
 
 ## Testing the API
 
 ### Example 1: Analyze a support ticket
 
-```bash
+```shell
 curl -X POST "http://localhost:8080/analyze-ticket" \
   -H "Content-Type: application/json" \
   -d '{
@@ -214,6 +212,7 @@ curl -X POST "http://localhost:8080/analyze-ticket" \
 ```
 
 Expected response:
+
 ```json
 {
   "category": "billing",
@@ -227,7 +226,7 @@ Expected response:
 
 ### Example 2: Generate a support response
 
-```bash
+```shell
 # First, get the analysis
 ANALYSIS=$(curl -s -X POST "http://localhost:8080/analyze-ticket" \
   -H "Content-Type: application/json" \
@@ -248,7 +247,7 @@ curl -X POST "http://localhost:8080/generate-response" \
   }"
 ```
 
-By combining FastAPI's async capabilities with Outlines' structured generation, you can build robust APIs that leverage large language models while maintaining high performance and reliability.
+By combining FastAPI's async capabilities with Outlines' structured generation, you can build robust APIs that leverage large language models.
 
 ## Using Alternative Backends: SGLang and TGI
 
@@ -268,7 +267,7 @@ async def lifespan(app: FastAPI):
         base_url="http://localhost:30000/v1",  # SGLang server URL
         api_key="dummy"
     )
-    async_model = outlines.from_sglang(client)
+    async_model = models.from_sglang(client)
 
     yield
 
@@ -276,7 +275,8 @@ async def lifespan(app: FastAPI):
 ```
 
 Start your SGLang server with:
-```bash
+
+```shell
 python -m sglang.launch_server \
     --model-path meta-llama/Llama-2-7b-chat-hf \
     --port 30000
@@ -297,7 +297,7 @@ async def lifespan(app: FastAPI):
     client = huggingface_hub.AsyncInferenceClient(
         "http://localhost:8080"  # TGI server URL
     )
-    async_model = outlines.from_tgi(client)
+    async_model = models.from_tgi(client)
 
     yield
 
@@ -305,7 +305,8 @@ async def lifespan(app: FastAPI):
 ```
 
 Start your TGI server with:
-```bash
+
+```shell
 docker run --gpus all -p 8080:80 \
     ghcr.io/huggingface/text-generation-inference:latest \
     --model-id meta-llama/Llama-2-7b-chat-hf

From a2e7d5f507b7dc922d5c94c6457c5312d80bcd56 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 20 Jun 2025 00:20:47 +0200
Subject: [PATCH 363/505] Update the styling of the docs

---
 docs/stylesheets/extra.css | 361 ++++++++++++++++++++++++++++++-------
 1 file changed, 292 insertions(+), 69 deletions(-)

diff --git a/docs/stylesheets/extra.css b/docs/stylesheets/extra.css
index fb6123f04..befd88dbe 100644
--- a/docs/stylesheets/extra.css
+++ b/docs/stylesheets/extra.css
@@ -3,59 +3,79 @@
     src: url(https://fonts.googleapis.com/css2?family=Source+Code+Pro:ital,wght@0,200..900;1,200..900&display=swap);
 }
 
+/* Header/banner styling */
+.md-header {
+    background-color: #DFD1B6 !important;
+}
+
 :root > * {
-    /* Accent color shades */
-    /* --md-accent-fg-color: #d36237;
-    --md-accent-fg-color--transparent: #d36237;
-    --md-accent-bg-color: hsla(0, 0%, 100%, 1);
-    --md-accent-bg-color--light: hsla(0, 0%, 100%, 0.7); */
+    /* Notion-like color palette */
+    --md-default-fg-color: #37352f;
+    --md-default-fg-color--light: #73706c;
+    --md-default-fg-color--lighter: #9b9a97;
+    --md-default-bg-color: #ffffff;
+    --md-default-bg-color--light: #f7f6f3;
+    --md-default-bg-color--lighter: #edece9;
+
+    /* Typography */
+    --md-text-font-family: "Inter", -apple-system, BlinkMacSystemFont, "Segoe UI", Helvetica, Arial, sans-serif;
+    --md-code-font: "Source Code Pro", Consolas, "Liberation Mono", Menlo, monospace;
 
-    /* Primary color shades */
-    /* --md-accent-fg-color: hsla(#d36237, 1);
-    --md-accent-fg-color--transparent: hsla(#d36237, 0.1);
-    --md-accent-bg-color: #d36237
-    --md-accent-bg-color--light: hsla(#d36237, 0.7); */
+    /* Notion-like link colors */
+    --md-typeset-a-color: #37352f;
+    --md-accent-fg-color: #eb5757;
 
-    --md-text-font-family: "Inter";
-    --md-code-font: "Source Code Pro Custom";
-    /* --md-typeset-a-color: #d36237; /*this is the brand color */
+    /* Background colors */
+    --md-code-bg-color: #f7f6f3;
+    --md-code-fg-color: #eb5757;
 }
 
-/* Nord theme code block styling */
+/* Code block styling */
 .highlight pre,
 .md-typeset pre code,
 .md-typeset .highlight pre,
 .md-typeset .highlighttable pre {
     background-color: #2E3440 !important; /* Nord's darkest blue (nord0) */
-    border-radius: 0.5rem !important; /* Rounded corners */
-    box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -1px rgba(0, 0, 0, 0.06) !important; /* Subtle shadow */
+    border-radius: 4px !important; /* Subtle rounded corners like Notion */
+    box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1), 0 1px 3px rgba(0, 0, 0, 0.08) !important; /* Subtle shadow */
+    border: none !important; /* No border for cleaner look */
+}
+
+/* Adjust padding from code content */
+.md-code__content {
+    padding: 1em 1.5em !important; /* Increased top/bottom padding */
 }
 
-/* Ensure inline code also uses appropriate background */
-.md-typeset code {
-    background-color: #3B4252; /* Nord's slightly lighter shade (nord1) */
-    color: #D8DEE9; /* Nord's snow storm shade for better contrast */
-    border-radius: 0.2rem; /* Smaller radius for inline code */
-    padding: 0.1em 0.3em; /* Add some padding for better appearance */
+/* Style only inline code (not code blocks) */
+.md-typeset :not(pre) > code {
+    background-color: rgba(135, 131, 120, 0.15); /* Notion's exact inline code background */
+    color: #E35A26; /* Orange color for inline code */
+    border-radius: 3px; /* Subtle rounded corners */
+    padding: 0.2em 0.4em; /* Notion-like padding */
+    font-weight: 500; /* Medium weight */
+    font-size: 0.85em; /* Slightly smaller than body text */
+    border: none;
+    font-family: "SFMono-Regular", Consolas, "Liberation Mono", Menlo, monospace;
 }
 
 /* Override code block container background */
 .md-typeset .highlight,
 .md-typeset .highlighttable {
-    background-color: #2E3440 !important;
-    border-radius: 0.5rem !important;
-    overflow: hidden; /* Ensure child elements respect rounded corners */
+    background-color: #ffffff !important; /* Match page background */
+    border-radius: 4px !important; /* Match code block radius */
+    overflow: hidden; /* Ensure child elements respect borders */
 }
 
-/* Ensure proper spacing and shadow for the entire code block */
+/* Ensure proper spacing for the entire code block */
 .md-typeset pre {
-    margin: 1em 0 !important;
-    box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1), 0 2px 4px -1px rgba(0, 0, 0, 0.06) !important;
+    margin: 1.5em 0 !important;
 }
 
 /* Style the copy button with Nord snow grey */
 .md-clipboard {
     color: #D8DEE9 !important; /* Nord snow storm */
+    top: 0.75em !important; /* Lower the button more */
+    right: 0.5em !important; /* Add some spacing from right edge */
 }
 
 .md-clipboard:hover {
@@ -91,107 +111,310 @@
     scrollbar-color: #4C566A #3B4252;
 }
 
-/* Better visual hierarchy with bolder headings */
+/* Notion-like visual hierarchy */
 .md-typeset h1 {
-    font-weight: 700;
-    font-size: 2.2rem;
+    font-weight: 600;
+    font-size: 2.5rem;
     line-height: 1.2;
-    margin-top: 0;
-    margin-bottom: 1.5rem;
-    color: #1a1a1a;
+    margin-top: 2.5rem;
+    margin-bottom: 0.75rem;
+    color: #37352f;
+    letter-spacing: -0.01em;
 }
 
 .md-typeset h2 {
     font-weight: 600;
-    font-size: 1.75rem;
+    font-size: 1.875rem;
     line-height: 1.3;
     margin-top: 2.5rem;
-    margin-bottom: 1.25rem;
-    color: #2a2a2a;
-    border-bottom: 2px solid #e0e0e0;
-    padding-bottom: 0.4rem;
+    margin-bottom: 0.75rem;
+    color: #37352f;
+    letter-spacing: -0.005em;
+    border: none;
+    padding: 0;
 }
 
 .md-typeset h3 {
-    font-weight: 600;
-    font-size: 1.35rem;
+    font-weight: 500;
+    font-size: 1.5rem;
     line-height: 1.3;
     margin-top: 2rem;
     margin-bottom: 0.75rem;
-    color: #333;
+    color: #37352f;
+    letter-spacing: -0.003em;
 }
 
 .md-typeset h4 {
     font-weight: 500;
-    font-size: 1.15rem;
+    font-size: 1.25rem;
     line-height: 1.3;
     margin-top: 1.75rem;
-    margin-bottom: 0.6rem;
-    color: #444;
+    margin-bottom: 0.5rem;
+    color: #37352f;
 }
 
 .md-typeset h5 {
     font-weight: 500;
-    font-size: 1.05rem;
+    font-size: 1rem;
     line-height: 1.4;
     margin-top: 1.5rem;
     margin-bottom: 0.5rem;
-    color: #555;
+    color: #37352f;
 }
 
 .md-typeset h6 {
     font-weight: 500;
-    font-size: 0.95rem;
+    font-size: 0.875rem;
     line-height: 1.4;
     margin-top: 1.25rem;
     margin-bottom: 0.5rem;
-    color: #666;
-    text-transform: uppercase;
-    letter-spacing: 0.05em;
+    color: #37352f;
+    text-transform: none;
+    letter-spacing: normal;
 }
 
-/* Improve paragraph spacing for better readability */
+/* Notion-like paragraph styling */
 .md-typeset p {
-    line-height: 1.55;
-    margin-bottom: 1rem;
+    line-height: 1.5;
+    margin-bottom: 1em; /* More whitespace */
+    color: #37352f;
+    font-weight: 400;
 }
 
-/* Make navigation items bolder */
+/* Reduce font size for navigation */
 .md-nav__link {
-    font-weight: 400;
+    font-size: 0.6875rem; /* 11px */
 }
 
-.md-nav__link--active,
-.md-nav__link:hover {
-    font-weight: 600;
-    color: #1a1a1a;
+/* Reduce font size for TOC */
+.md-nav--secondary .md-nav__link {
+    font-size: 0.6875rem; /* 11px */
+}
+
+/* Reduce spacing between navigation items */
+.md-nav__item {
+    margin: 0;
 }
 
-/* Improve list spacing */
+.md-nav__link {
+    padding-top: 0;
+    padding-bottom: 0;
+    line-height: 1.2; /* Tighter line height */
+}
+
+/* Make navigation sections more compact */
+.md-nav__title {
+    line-height: 1.2;
+    padding: 0.2rem 0.5rem;
+    margin-bottom: 0.2rem;
+}
+
+/* Reduce TOC line height for compactness */
+.md-nav--secondary .md-nav__link {
+    line-height: 1.2;
+}
+
+/* Notion-like list styling */
 .md-typeset ul,
 .md-typeset ol {
-    margin-bottom: 1.5rem;
+    margin-top: 0.25em; /* Reduced top margin to sit closer to text */
+    margin-bottom: 1em; /* Match paragraph spacing */
+    color: #37352f;
 }
 
 .md-typeset li {
-    line-height: 1.55;
-    margin-bottom: 0.4rem;
+    line-height: 1.5;
+    margin-bottom: 0.15rem; /* Slightly more spacing between list items */
+    font-weight: 400;
+}
+
+/* Notion-style links */
+.md-typeset a {
+    color: #37352f;
+    text-decoration: underline;
+    text-decoration-color: rgba(55, 53, 47, 0.4);
+    text-underline-offset: 2px;
+    transition: text-decoration-color 0.1s ease;
+}
+
+.md-typeset a:hover {
+    text-decoration-color: rgba(55, 53, 47, 0.8);
+    background-color: rgba(55, 53, 47, 0.04);
 }
 
 /* Make important elements stand out */
 .md-typeset strong {
     font-weight: 600;
-    color: #1a1a1a;
+    color: #37352f;
 }
 
 /* Better spacing for code blocks in relation to text */
 .md-typeset pre {
-    margin: 1.5em 0 !important;
+    margin: 1.5em 0 !important; /* More whitespace around code blocks */
+}
+
+/* Notion-style tables */
+.md-typeset table {
+    border-collapse: collapse;
+    margin: 1rem 0;
 }
 
-/* Improve table headers */
 .md-typeset table th {
     font-weight: 600;
-    background-color: #f5f5f5;
-    color: #1a1a1a;
+    background-color: #f7f6f3;
+    color: #37352f;
+    border: 1px solid #e1e0dd;
+    padding: 0.5rem 0.75rem;
+}
+
+.md-typeset table td {
+    border: 1px solid #e1e0dd;
+    padding: 0.5rem 0.75rem;
+}
+
+/* Notion-style blockquotes */
+.md-typeset blockquote {
+    border-left: 3px solid #37352f;
+    padding-left: 1rem;
+    margin: 1rem 0;
+    color: #37352f;
+    background: transparent;
+}
+
+/* Page styling */
+.md-content {
+    background-color: #ffffff;
+}
+
+.md-sidebar {
+    background-color: #fbfbfa;
+}
+
+/* Remove shadows for cleaner look */
+.md-header,
+.md-tabs {
+    box-shadow: none;
+    border-bottom: 1px solid #e1e0dd;
+}
+
+/* Admonition styling with custom palette */
+.md-typeset .admonition,
+.md-typeset details {
+    border-radius: 4px;
+    border: none;
+    box-shadow: none;
+    font-size: 0.6875rem; /* Very small font size - 11px */
+    padding: 0.75rem;
+    margin: 1rem 0;
+}
+
+/* Note/Info - Blue */
+.md-typeset .admonition.note,
+.md-typeset details.note,
+.md-typeset .admonition.info,
+.md-typeset details.info {
+    background-color: rgba(127, 154, 207, 0.1) !important;
+    border-left: 4px solid #7F9ACF !important;
+}
+
+.md-typeset .note > .admonition-title,
+.md-typeset .note > summary,
+.md-typeset .info > .admonition-title,
+.md-typeset .info > summary {
+    background-color: rgba(127, 154, 207, 0.2) !important;
+    border-left: none !important;
+}
+
+/* Additional specificity for info type and custom types that should be blue */
+.md-typeset .admonition.admonition-info,
+.md-typeset details.details-info,
+.md-typeset .admonition.installation,
+.md-typeset .admonition.example,
+.md-typeset .admonition.abstract,
+.md-typeset .admonition.summary,
+.md-typeset .admonition.tldr {
+    background-color: rgba(127, 154, 207, 0.1) !important;
+    border-left: 4px solid #7F9ACF !important;
+}
+
+/* Titles for custom blue admonitions */
+.md-typeset .installation > .admonition-title,
+.md-typeset .example > .admonition-title,
+.md-typeset .abstract > .admonition-title,
+.md-typeset .summary > .admonition-title,
+.md-typeset .tldr > .admonition-title {
+    background-color: rgba(127, 154, 207, 0.2) !important;
+    border-left: none !important;
+}
+
+/* Warning/Caution - Yellow */
+.md-typeset .admonition.warning,
+.md-typeset details.warning,
+.md-typeset .admonition.caution,
+.md-typeset details.caution {
+    background-color: rgba(189, 147, 47, 0.1);
+    border-left: 4px solid #BD932F;
+}
+
+.md-typeset .warning > .admonition-title,
+.md-typeset .warning > summary,
+.md-typeset .caution > .admonition-title,
+.md-typeset .caution > summary {
+    background-color: rgba(189, 147, 47, 0.2);
+    border-left: none;
+}
+
+/* Danger/Error - Orange */
+.md-typeset .admonition.danger,
+.md-typeset details.danger,
+.md-typeset .admonition.error,
+.md-typeset details.error {
+    background-color: rgba(227, 90, 38, 0.1);
+    border-left: 4px solid #E35A26;
+}
+
+.md-typeset .danger > .admonition-title,
+.md-typeset .danger > summary,
+.md-typeset .error > .admonition-title,
+.md-typeset .error > summary {
+    background-color: rgba(227, 90, 38, 0.2);
+    border-left: none;
+}
+
+/* Success/Tip/Hint - Green */
+.md-typeset .admonition.success,
+.md-typeset details.success,
+.md-typeset .admonition.tip,
+.md-typeset details.tip,
+.md-typeset .admonition.hint,
+.md-typeset details.hint {
+    background-color: rgba(166, 180, 163, 0.1);
+    border-left: 4px solid #A6B4A3;
+}
+
+.md-typeset .success > .admonition-title,
+.md-typeset .success > summary,
+.md-typeset .tip > .admonition-title,
+.md-typeset .tip > summary,
+.md-typeset .hint > .admonition-title,
+.md-typeset .hint > summary {
+    background-color: rgba(166, 180, 163, 0.2);
+    border-left: none;
+}
+
+/* General admonition title styling */
+.md-typeset .admonition-title,
+.md-typeset summary {
+    font-weight: 600;
+    font-size: 0.6875rem; /* Very small - 11px */
+    padding: 0.5rem 0.75rem;
+    margin: -0.75rem -0.75rem 0.5rem -0.75rem;
+    border-radius: 4px 4px 0 0;
+}
+
+/* Ensure consistent icon styling */
+.md-typeset .admonition > .admonition-title::before,
+.md-typeset details > summary::before {
+    font-size: 1rem;
+    margin-right: 0.5rem;
 }

From bf86aec03d9b443360342da3297105ca49e71b8f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 20 Jun 2025 09:53:32 +0200
Subject: [PATCH 364/505] Delete LICENSE from the guide

---
 docs/licence.md | 34 ----------------------------------
 mkdocs.yml      |  4 +---
 2 files changed, 1 insertion(+), 37 deletions(-)
 delete mode 100644 docs/licence.md

diff --git a/docs/licence.md b/docs/licence.md
deleted file mode 100644
index a74661fa3..000000000
--- a/docs/licence.md
+++ /dev/null
@@ -1,34 +0,0 @@
----
-title: Licence
----
-
-# Licence and citations
-
-Outlines is licenced under the Apache 2.0 licence. To comply with the licence you need to add the following notice at the top every file that uses part of Outlines' code:
-
-```
-Copyright 2023- The Outlines developers
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-```
-
-If you use Outlines in your work you can use the following citation:
-
-```
-@article{willard2023efficient,
-  title={Efficient Guided Generation for LLMs},
-  author={Willard, Brandon T and Louf, R{\'e}mi},
-  journal={arXiv preprint arXiv:2307.09702},
-  year={2023}
-}
-```
diff --git a/mkdocs.yml b/mkdocs.yml
index 336715702..438a9d62c 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -121,9 +121,7 @@ nav:
           - Installation: guide/installation.md
           - Migrating to v1: guide/migration.md
           - Deploying with FastAPI: guide/fastapi_vllm_deployment.md
-          - Licence: licence.md
-          - Utility:
-                - Chat Templating for Instruct Models: guide/chat_templating.md
+          - Chat Templating for Instruct Models: guide/chat_templating.md
 
     - Features:
           - Overview: features/index.md

From 17ce00fd89cbaca919b14fe4359c36b09f2d8d6a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 20 Jun 2025 13:51:24 +0200
Subject: [PATCH 365/505] s/bash/shell in code blocks

---
 docs/examples/atomic_caption.md             |  2 +-
 docs/examples/chain_of_thought.md           |  4 ++--
 docs/examples/deploy-using-bentoml.md       | 12 ++++++------
 docs/examples/deploy-using-cerebrium.md     |  6 +++---
 docs/examples/deploy-using-modal.md         | 10 +++++-----
 docs/examples/earnings-reports.md           |  2 +-
 docs/examples/knowledge_graph_extraction.md |  4 ++--
 docs/examples/qa-with-citations.md          |  4 ++--
 docs/examples/react_agent.md                |  4 ++--
 docs/examples/read-pdfs.md                  |  2 +-
 docs/examples/receipt-digitization.md       |  2 +-
 docs/guide/chat_templating.md               |  2 +-
 docs/guide/getting_started.md               |  4 ++--
 docs/guide/installation.md                  |  4 ++--
 14 files changed, 31 insertions(+), 31 deletions(-)

diff --git a/docs/examples/atomic_caption.md b/docs/examples/atomic_caption.md
index 08782594f..38035f9aa 100644
--- a/docs/examples/atomic_caption.md
+++ b/docs/examples/atomic_caption.md
@@ -7,7 +7,7 @@ We will be using the Pixtral-12B model from Mistral to take advantage of some of
 ## Setup
 First, we need to install the necessary dependencies. In addition to Outlines, we'll need to install the transformers library and any specific requirements for the vision-language model we'll be using.
 
-```bash
+```shell
 pip install outlines transformers torch
 ```
 
diff --git a/docs/examples/chain_of_thought.md b/docs/examples/chain_of_thought.md
index 42d6c8550..207363764 100644
--- a/docs/examples/chain_of_thought.md
+++ b/docs/examples/chain_of_thought.md
@@ -7,7 +7,7 @@ In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to apply
 
 We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```bash
+```shell
 pip install llama-cpp-python
 ```
 
@@ -34,7 +34,7 @@ model = outlines.from_llamacpp(llm)
 
     By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-    ```bash
+    ```shell
     wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
     ```
 
diff --git a/docs/examples/deploy-using-bentoml.md b/docs/examples/deploy-using-bentoml.md
index 5d000cb52..c0ddf7e2d 100644
--- a/docs/examples/deploy-using-bentoml.md
+++ b/docs/examples/deploy-using-bentoml.md
@@ -8,7 +8,7 @@ In this guide, we will show you how to use BentoML to run programs written with
 
 First we need to download an LLM (Mistral-7B-v0.1 in this example and you can use any other LLM) and import the model into BentoML's [Model Store](https://docs.bentoml.com/en/latest/guides/model-store.html). Let's install BentoML and other dependencies from PyPi (preferably in a virtual environment):
 
-```bash
+```shell
 pip install -r requirements.txt
 ```
 
@@ -45,7 +45,7 @@ if __name__ == "__main__":
 
 You can verify the download is successful by running:
 
-```bash
+```shell
 $ bentoml models list
 
 Tag                                          Module  Size        Creation Time
@@ -167,7 +167,7 @@ Now you can save the above code to `service.py` (or use [this implementation](ht
 
 Then you can run a server locally by:
 
-```bash
+```shell
 bentoml serve .
 ```
 
@@ -177,7 +177,7 @@ The server is now active at <http://localhost:3000>. You can interact with it us
 
 <summary>CURL</summary>
 
-```bash
+```shell
 curl -X 'POST' \
   'http://localhost:3000/generate' \
   -H 'accept: application/json' \
@@ -207,7 +207,7 @@ with bentoml.SyncHTTPClient("http://localhost:3000") as client:
 
 Expected output:
 
-```bash
+```shell
 {
   "name": "Aura",
   "age": 15,
@@ -223,7 +223,7 @@ After the Service is ready, you can deploy it to [BentoCloud](https://docs.bento
 
 Make sure you have [logged in to BentoCloud](https://docs.bentoml.com/en/latest/bentocloud/how-tos/manage-access-token.html), then run the following command to deploy it.
 
-```bash
+```shell
 bentoml deploy .
 ```
 
diff --git a/docs/examples/deploy-using-cerebrium.md b/docs/examples/deploy-using-cerebrium.md
index bf010d583..caf1c96b0 100644
--- a/docs/examples/deploy-using-cerebrium.md
+++ b/docs/examples/deploy-using-cerebrium.md
@@ -8,14 +8,14 @@ In this guide we will show you how you can use Cerebrium to run programs written
 
 First, we install Cerebrium and login to get authenticated.
 
-```bash
+```shell
 pip install cerebrium
 cerebrium login
 ```
 
 Then let us create our first project
 
-```bash
+```shell
 cerebrium init outlines-project
 ```
 
@@ -118,7 +118,7 @@ def generate(
 
 ## Run on the cloud
 
-```bash
+```shell
 cerebrium deploy
 ```
 
diff --git a/docs/examples/deploy-using-modal.md b/docs/examples/deploy-using-modal.md
index 23a846a6d..b6d9c7582 100644
--- a/docs/examples/deploy-using-modal.md
+++ b/docs/examples/deploy-using-modal.md
@@ -8,14 +8,14 @@ In this guide we will show you how you can use Modal to run programs written wit
 
 We recommend installing `modal` and `outlines` in a virtual environment. You can create one with:
 
-```bash
+```shell
 python -m venv venv
 source venv/bin/activate
 ```
 
 Then install the required packages:
 
-```bash
+```shell
 pip install modal outlines
 ```
 
@@ -187,19 +187,19 @@ Here `@app.local_entrypoint()` decorator defines `main` as the function to start
 
 First install the Modal client from PyPi, if you have not already:
 
-```bash
+```shell
 pip install modal
 ```
 
 You then need to obtain a token from Modal. Run the following command:
 
-```bash
+```shell
 modal setup
 ```
 
 Once that is set you can run inference on the cloud using:
 
-```bash
+```shell
 modal run example.py
 ```
 
diff --git a/docs/examples/earnings-reports.md b/docs/examples/earnings-reports.md
index cca1e9513..645726b07 100644
--- a/docs/examples/earnings-reports.md
+++ b/docs/examples/earnings-reports.md
@@ -12,7 +12,7 @@ The example is a smaller subset of a full demo found [here](https://github.com/d
 
 Install outlines and required dependencies:
 
-```bash
+```shell
 # Later versions of torch can have difficulty with certain CUDA drivers.
 # We recommend using 2.4.0 for now, but you may wish to experiment with
 # other versions.
diff --git a/docs/examples/knowledge_graph_extraction.md b/docs/examples/knowledge_graph_extraction.md
index f4b21c84c..39b1c1d6b 100644
--- a/docs/examples/knowledge_graph_extraction.md
+++ b/docs/examples/knowledge_graph_extraction.md
@@ -4,7 +4,7 @@ In this guide, we use [outlines](https://dottxt-ai.github.io/outlines/) to extra
 
 We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```bash
+```shell
 pip install llama-cpp-python
 ```
 
@@ -31,7 +31,7 @@ model = outlines.from_llamacpp(llm)
 
     By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-    ```bash
+    ```shell
     wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
     ```
 
diff --git a/docs/examples/qa-with-citations.md b/docs/examples/qa-with-citations.md
index 213942d02..756988c3f 100644
--- a/docs/examples/qa-with-citations.md
+++ b/docs/examples/qa-with-citations.md
@@ -4,7 +4,7 @@ This tutorial is adapted from the [instructor-ollama notebook](https://github.co
 
 We will use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```bash
+```shell
 pip install llama-cpp-python
 ```
 
@@ -30,7 +30,7 @@ model = outlines.from_llamacpp(llm)
 
     By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-    ```bash
+    ```shell
     wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
     ```
 
diff --git a/docs/examples/react_agent.md b/docs/examples/react_agent.md
index bd3d51d6c..55f150eb0 100644
--- a/docs/examples/react_agent.md
+++ b/docs/examples/react_agent.md
@@ -8,7 +8,7 @@ Additionally, we give the LLM the possibility of using a scratchpad described in
 
 We use [llama.cpp](https://github.com/ggerganov/llama.cpp) using the [llama-cpp-python](https://github.com/abetlen/llama-cpp-python) library. Outlines supports llama-cpp-python, but we need to install it ourselves:
 
-```bash
+```shell
 pip install llama-cpp-python
 ```
 
@@ -34,7 +34,7 @@ model = outlines.from_llamacpp(llm)
 
     By default the model weights are downloaded to the hub cache but if we want so store the weights in a custom folder, we pull a quantized GGUF model [Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Theta-Llama-3-8B-GGUF) by [NousResearch](https://nousresearch.com/) from [HuggingFace](https://huggingface.co/):
 
-    ```bash
+    ```shell
     wget https://hf.co/NousResearch/Hermes-2-Pro-Llama-3-8B-GGUF/resolve/main/Hermes-2-Pro-Llama-3-8B-Q4_K_M.gguf
     ```
 
diff --git a/docs/examples/read-pdfs.md b/docs/examples/read-pdfs.md
index 1854aba42..81e71f467 100644
--- a/docs/examples/read-pdfs.md
+++ b/docs/examples/read-pdfs.md
@@ -17,7 +17,7 @@ This cookbook demonstrates how to
 
 You'll need to install these dependencies:
 
-```bash
+```shell
 pip install outlines pillow transformers torch==2.4.0 pdf2image
 
 # Optional, but makes the output look nicer
diff --git a/docs/examples/receipt-digitization.md b/docs/examples/receipt-digitization.md
index 8ecb07969..994024a9f 100644
--- a/docs/examples/receipt-digitization.md
+++ b/docs/examples/receipt-digitization.md
@@ -4,7 +4,7 @@
 
 You'll need to install the dependencies:
 
-```bash
+```shell
 pip install outlines torch==2.4.0 transformers accelerate pillow rich
 ```
 
diff --git a/docs/guide/chat_templating.md b/docs/guide/chat_templating.md
index e3f33fbfd..d5752b8fe 100644
--- a/docs/guide/chat_templating.md
+++ b/docs/guide/chat_templating.md
@@ -29,7 +29,7 @@ prompt = tokenizer.apply_chat_template(
 
 yields
 
-```
+```ascii
 <|im_start|>system
 You extract information from text.<|im_end|>
 <|im_start|>user
diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
index ed7525980..f3c637988 100644
--- a/docs/guide/getting_started.md
+++ b/docs/guide/getting_started.md
@@ -8,13 +8,13 @@ title: Getting Started
 
 We recommend using `uv` to install Outlines. You can find `uv` installation instructions [here](https://github.com/astral-sh/uv).
 
-```bash
+```shell
 uv pip install 'outlines[transformers]'
 ```
 
 or the classic `pip`:
 
-```bash
+```shell
 pip install 'outlines[transformers]'
 ```
 
diff --git a/docs/guide/installation.md b/docs/guide/installation.md
index c626841fd..ba8a2cd25 100644
--- a/docs/guide/installation.md
+++ b/docs/guide/installation.md
@@ -10,7 +10,7 @@ We recommend using modern Python packaging tools such as `uv` for managing pytho
 
 ### uv (Recommended)
 
-```bash
+```shell
 # Install uv
 curl -LsSf https://astral.sh/uv/install.sh | sh
 
@@ -22,7 +22,7 @@ uv pip install outlines
 
 or with basic pip:
 
-```bash
+```shell
 pip install outlines
 ```
 

From 62b2648b2312c032205b011e5914349efd9c2815 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Fri, 20 Jun 2025 14:04:38 +0200
Subject: [PATCH 366/505] Fix typos

---
 docs/features/core/output_types.md |   4 +-
 docs/features/index.md             |   2 +-
 docs/features/models/index.md      |   4 +-
 docs/features/utility/regex_dsl.md |   2 +-
 docs/guide/getting_started.md      | 124 ++++++++++++-----------------
 docs/guide/installation.md         |   2 +-
 docs/guide/migration.md            |   2 +-
 7 files changed, 59 insertions(+), 81 deletions(-)

diff --git a/docs/features/core/output_types.md b/docs/features/core/output_types.md
index fd0a55b96..edaa155b3 100644
--- a/docs/features/core/output_types.md
+++ b/docs/features/core/output_types.md
@@ -13,7 +13,7 @@ Outlines models accept a __prompt__ and an __output type__ when they are invoked
 Output types can be from the general Python ecosystem, including:
 - Most native Python types, such as `int` or `str`
 - Types from the `typing` module, such as `Literal`, `List`, `Dict`, `Enum`, etc
-- Types from popular theird party libraries such as Pydantic or GenSON.
+- Types from popular third party libraries such as Pydantic or GenSON.
 
 Outlines also provides special classes for certain output structures (more details below):
 - JSON schemas with `JsonSchema`
@@ -164,7 +164,7 @@ output_type = JsonSchema(schema_dict)
 
 ### Regex Patterns
 
-Outlines provides support for text generation constrained by regular expressions. Since regular expressions are expressed as simple raw string literals, regex strings must wrapped in an `outlines.types.Regex` object to clarify the expected return type.
+Outlines provides support for text generation constrained by regular expressions. Since regular expressions are expressed as simple raw string literals, regex strings must wrapped in an `outlines.types.Regex` object.
 
 ```python
 from outlines.types import Regex
diff --git a/docs/features/index.md b/docs/features/index.md
index 00df02ba3..941103ee2 100644
--- a/docs/features/index.md
+++ b/docs/features/index.md
@@ -8,7 +8,7 @@ This section presents in details the different features of Outlines.
 - [Output Types](./core/output_types.md)
 - [Generators](./core/generator.md)
 
-## Utility
+## Utilities
 
 - [Applications](./utility/application.md)
 - [Templates](./utility/templates.md)
diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index 89c4fa124..c597599de 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -45,6 +45,8 @@ print(result) # '200'
 
 ## Features Matrix
 
+In alphabetical order:
+
 | | [Anthropic](../../models/anthropic) | [Dottxt](../../models/dottxt) | [Gemini](../../models/gemini) | [LlamaCpp](../../models/llamacpp) | [MLXLM](../../models/mlxlm) | [Ollama](../../models/ollama) | [OpenAI](../../models/openai) | [SGLang](../../models/sglang) | [TGI](../../models/tgi) | [Transformers](../../models/transformers) | [Transformers MultiModal](../../models/transformers_multimodal) | [VLLM](../../models/vllm) | [VLLMOffline](../../models/vllm_offline) |
 |---|---|---|---|---|---|---|---|---|---|---|---|---|---|
 | **Output Types** | | | | | | | | | | | | | |
@@ -63,7 +65,7 @@ print(result) # '200'
 
 Models can be divided into two categories: local models and server-based models.
 
-In the case of local models, the text generation happens within the inference library object used to instantite the model. This gives Outlines direct access to the generation process (through a logits processor) and means all structured generation output types are available.
+In the case of local models, the text generation happens within the inference library object used to instantiate the model. This gives Outlines direct access to the generation process (through a logits processor) and means all structured generation output types are available.
 
 The local models available are the following:
 
diff --git a/docs/features/utility/regex_dsl.md b/docs/features/utility/regex_dsl.md
index d6c1a68c2..937827450 100644
--- a/docs/features/utility/regex_dsl.md
+++ b/docs/features/utility/regex_dsl.md
@@ -329,7 +329,7 @@ print(pattern)
 
 *Expected Output:*
 
-```
+```ascii
 └── Sequence
     ├── String('a')
     ├── KleenePlus(+)
diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
index f3c637988..77bc46ccf 100644
--- a/docs/guide/getting_started.md
+++ b/docs/guide/getting_started.md
@@ -28,45 +28,64 @@ The full list of available models along with detailed explanation on how to use
 
 For a quick start, you can find below an example of how to initialize all supported models in Outlines:
 
-=== "Anthropic"
+=== "vLLM"
 
     ```python
     import outlines
-    from anthropic import Anthropic
+    from openai import OpenAI
 
-    # Create an Anthropic client
-    anthropic_client = Anthropic()
+    # You must have a separate vLLM server running
+    # Create an OpenAI client with the base URL of the VLLM server
+    openai_client = OpenAI(base_url="http://localhost:11434/v1")
 
     # Create an Outlines model
-    model = outlines.from_anthropic(anthropic_client, "claude-3-haiku-20240307")
+    model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
     ```
 
-=== "Dottxt"
+=== "Ollama"
 
     ```python
     import outlines
-    from dottxt.client import Dottxt
+    from ollama import Client
 
-    # Create an Dottxt client
-    dottxt_client = Dottxt()
+    # Create an Ollama client
+    ollama_client = Client()
+
+    # Create an Outlines model, the model must be available on your system
+    model = outlines.from_ollama(ollama_client, "tinyllama")
+    ```
+
+=== "OpenAI"
+
+    ```python
+    import outlines
+    from openai import OpenAI
+
+    # Create an OpenAI client instance
+    openai_client = OpenAI()
 
     # Create an Outlines model
-    model = outlines.from_dottxt(dottxt_client)
+    model = outlines.from_openai(openai_client, "gpt-4o")
     ```
 
-=== "Gemini"
+=== "Transformers"
 
     ```python
     import outlines
-    from google.generativeai import GenerativeModel
+    from transformers
 
-    # Create a Gemini client
-    gemini_client = GenerativeModel()
+    # Define the model you want to use
+    model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
+
+    # Create a HuggingFace model and tokenizer
+    hf_model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
+    hf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
 
     # Create an Outlines model
-    model = outlines.from_gemini(gemini_client, "gemini-1-5-flash")
+    model = outlines.from_transformers(hf_model, hf_tokenizer)
     ```
 
+
 === "llama.cpp"
 
     ```python
@@ -84,43 +103,30 @@ For a quick start, you can find below an example of how to initialize all suppor
     model = outlines.from_llamacpp(llama_cpp_model)
     ```
 
-=== "mlx-lm"
-
-    ```python
-    import outlines
-    import mlx_lm
-
-    # Create an MLXLM model with the output of mlx_lm.load
-    # The model will be downloaded from the HuggingFace hub
-    model = outlines.from_mlxlm(
-        **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-    )
-    ```
-
-=== "Ollama"
+=== "Gemini"
 
     ```python
     import outlines
-    from ollama import Client
+    from google.generativeai import GenerativeModel
 
-    # Create an Ollama client
-    ollama_client = Client()
+    # Create a Gemini client
+    gemini_client = GenerativeModel()
 
-    # Create an Outlines model, the model must be available on your system
-    model = outlines.from_ollama(ollama_client, "tinyllama")
+    # Create an Outlines model
+    model = outlines.from_gemini(gemini_client, "gemini-1-5-flash")
     ```
 
-=== "OpenAI"
+=== "mlx-lm"
 
     ```python
     import outlines
-    from openai import OpenAI
-
-    # Create an OpenAI client instance
-    openai_client = OpenAI()
+    import mlx_lm
 
-    # Create an Outlines model
-    model = outlines.from_openai(openai_client, "gpt-4o")
+    # Create an MLXLM model with the output of mlx_lm.load
+    # The model will be downloaded from the HuggingFace hub
+    model = outlines.from_mlxlm(
+        **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    )
     ```
 
 === "SgLang"
@@ -153,23 +159,6 @@ For a quick start, you can find below an example of how to initialize all suppor
     model = outlines.from_tgi(tgi_client)
     ```
 
-=== "Transformers"
-
-    ```python
-    import outlines
-    from transformers
-
-    # Define the model you want to use
-    model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
-
-    # Create a HuggingFace model and tokenizer
-    hf_model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
-    hf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
-
-    # Create an Outlines model
-    model = outlines.from_transformers(hf_model, hf_tokenizer)
-    ```
-
 === "vLLM (offline)"
 
     ```python
@@ -183,19 +172,6 @@ For a quick start, you can find below an example of how to initialize all suppor
     model = outlines.from_vllm_offline(vllm_model)
     ```
 
-=== "vLLM (online)"
-
-    ```python
-    import outlines
-    from openai import OpenAI
-
-    # You must have a separate vLLM server running
-    # Create an OpenAI client with the base URL of the VLLM server
-    openai_client = OpenAI(base_url="http://localhost:11434/v1")
-
-    # Create an Outlines model
-    model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
-    ```
 
 ## Generating Text
 
@@ -221,7 +197,7 @@ for chunk in model.streaming("Write a short story about a cat.")
 
 ## Structured Generation
 
-Outlines follows a simple pattern that mirrors Python's own type system for structured output. Simply specify the desired output type as you would when using type hinting with a function, and Outlines will ensure your data matches that structure exactly.
+Outlines follows a simple pattern that mirrors Python's own type system for structured outputs. Simply specify the desired output type as you would when using type hinting with a function, and Outlines will ensure your data matches that structure exactly.
 
 Supported output types can be organized in 5 categories:
 
@@ -235,7 +211,7 @@ Consult the section on [Output Types](../../features/core/output_types.md) in th
 
 In the meantime, you can find below examples of using each of the five output type categories:
 
-=== "Basic Yypes"
+=== "Basic Types"
 
     ```python
     model = <your_model_as_defined_above>
@@ -332,7 +308,7 @@ In the meantime, you can find below examples of using each of the five output ty
     print(result) # '2 + 3'
     ```
 
-It's important to note that not all output types are available for all models due to limitations in the underline inference libraries. The [Models](../features/models/index.md) section of the features documentation includes a features matrix to easily visualize output type availabilities.
+It's important to note that not all output types are available for all models due to limitations in the underlying inference engines. The [Models](../features/models/index.md) section of the features documentation includes a features matrix that summarize the availability of output types.
 
 ## Generators
 
diff --git a/docs/guide/installation.md b/docs/guide/installation.md
index ba8a2cd25..65997f996 100644
--- a/docs/guide/installation.md
+++ b/docs/guide/installation.md
@@ -20,7 +20,7 @@ source .venv/bin/activate
 uv pip install outlines
 ```
 
-or with basic pip:
+or with pip:
 
 ```shell
 pip install outlines
diff --git a/docs/guide/migration.md b/docs/guide/migration.md
index df766d8c8..729cceac1 100644
--- a/docs/guide/migration.md
+++ b/docs/guide/migration.md
@@ -4,7 +4,7 @@ Outlines 1.0 introduces some breaking changes that affect the way you use the li
 
 This guide will help you migrate your code to the new version.
 
-All previous functionality will be supported until Outlines version 1.1.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
+All previous functionalities will be supported until Outlines version 1.1.0, but a warning message will be displayed to remind you to migrate your code and provide instructions to help you do so. Please migrate your code to the v1 as soon as possible.
 
 ## Removed or modified features
 - [Generate functions](#generate-functions)

From 2e06e7f13415f429da025655f2f52025bce261af Mon Sep 17 00:00:00 2001
From: ganglike <81566899+ganglike248@users.noreply.github.com>
Date: Sun, 22 Jun 2025 16:03:29 +0900
Subject: [PATCH 367/505] Fix typos in getting started guide (#1644)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

## Description
This PR fixes minor typos found in the getting started documentation to
improve readability and code examples.

## Changes Made
- **Line 74**: Fixed "Context-free Grammarq" to "Context-free Grammars"
in the output types section
- **Multiple Choice example**: Fixed missing closing quote in enum
definition: `pizza = "pizza"`, `burger = "burger"`

## Type of Change
- [x] Documentation update
- [ ] Bug fix
- [ ] New feature
- [ ] Breaking change

## Testing
- [x] Verified spelling corrections are accurate
- [x] Confirmed code examples are syntactically correct
- [x] Ran pre-commit hooks successfully

## Additional Context
These are minor documentation improvements that fix typos and ensure
code examples are properly formatted. The changes maintain the original
meaning while correcting syntax errors.
This is my first contribution to this project, but I wanted to make a
small contribution by fixing this typo to help improve the project. 😊
---
 docs/guide/getting_started.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
index 77bc46ccf..f510bc16f 100644
--- a/docs/guide/getting_started.md
+++ b/docs/guide/getting_started.md
@@ -205,7 +205,7 @@ Supported output types can be organized in 5 categories:
 - [Multiple Choices](../../features/core/output_types#multiple-choices): using `Literal` or `Enum`
 - [JSON Schemas](../../features/core/output_types#json-schemas): using a wide range of possible objects including Pydantic models and dataclasses
 - [Regex](../../features/core/output_types#regex-patterns): through the Outlines's `Regex` object
-- [Context-free Grammarq](../../features/core/output_types#context-free-grammars): through the Outlines's `CFG` object
+- [Context-free Grammars](../../features/core/output_types#context-free-grammars): through the Outlines's `CFG` object
 
 Consult the section on [Output Types](../../features/core/output_types.md) in the features documentation for more detailed information on all supported types for each output type category.
 
@@ -228,8 +228,8 @@ In the meantime, you can find below examples of using each of the five output ty
 
     # Define our multiple choice output type
     class PizzaOrBurger(Enum):
-        pizza = "pizza
-        burger = "burger
+        pizza = "pizza"
+        burger = "burger"
 
     model = <your_model_as_defined_above>
 

From ce1d4900627b1c8985bcdb735d086dd4dd92f1cd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 21 Jun 2025 15:56:48 +0200
Subject: [PATCH 368/505] Remove the CI Docker build action

---
 .github/workflows/release_docker.yml | 36 ----------------------------
 1 file changed, 36 deletions(-)
 delete mode 100644 .github/workflows/release_docker.yml

diff --git a/.github/workflows/release_docker.yml b/.github/workflows/release_docker.yml
deleted file mode 100644
index 7a0ca4dbb..000000000
--- a/.github/workflows/release_docker.yml
+++ /dev/null
@@ -1,36 +0,0 @@
-name: Release Docker
-
-on:
-  release:
-    types:
-      - created
-  workflow_dispatch:
-    inputs:
-      release_tag:
-        description: 'Release Tag (for manual dispatch)'
-        required: false
-        default: 'latest'
-jobs:
-  release-job:
-    name: Build and publish on Docker Hub
-    runs-on: ubuntu-latest
-    environment: release
-    steps:
-    - name: Checkout
-      uses: actions/checkout@v4
-    - name: Log in to Docker Hub
-      uses: docker/login-action@v3
-      with:
-        username: ${{ secrets.DOCKERHUB_USERNAME }}
-        password: ${{ secrets.DOCKERHUB_TOKEN }}
-    - name: Build and push Docker image
-      uses: docker/build-push-action@v5
-      with:
-        push: true
-        tags: |
-          outlinesdev/outlines:latest
-          outlinesdev/outlines:${{ github.event.release.tag_name || github.event.inputs.release_tag }}
-        build-args: |
-          BUILDKIT_CONTEXT_KEEP_GIT_DIR=true
-    - name: Clean docker cache
-      run: docker system prune --all --force

From 068e8922f596fe8b0dbe4e8892204510b8d07040 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Sat, 21 Jun 2025 16:38:23 +0200
Subject: [PATCH 369/505] Move the VLM guide to guides/

---
 .../atomic_caption.md => guide/vlm.md}        | 86 ++++++++++++-------
 mkdocs.yml                                    |  2 +-
 2 files changed, 54 insertions(+), 34 deletions(-)
 rename docs/{examples/atomic_caption.md => guide/vlm.md} (65%)

diff --git a/docs/examples/atomic_caption.md b/docs/guide/vlm.md
similarity index 65%
rename from docs/examples/atomic_caption.md
rename to docs/guide/vlm.md
index 38035f9aa..979e65924 100644
--- a/docs/examples/atomic_caption.md
+++ b/docs/guide/vlm.md
@@ -1,18 +1,18 @@
 # Vision-Language Models with Outlines
 
-This guide demonstrates how to use Outlines with vision-language models, leveraging the outlines `TransformersMultiModal` model. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
+This guide demonstrates how to use Outlines with vision-language models. Vision-language models can process both text and images, allowing for tasks like image captioning, visual question answering, and more.
 
 We will be using the Pixtral-12B model from Mistral to take advantage of some of its visual reasoning capabilities and a workflow to generate a multistage atomic caption.
 
 ## Setup
-First, we need to install the necessary dependencies. In addition to Outlines, we'll need to install the transformers library and any specific requirements for the vision-language model we'll be using.
+First, we need to install the necessary dependencies. In addition to Outlines, we"ll need to install the transformers library and any specific requirements for the vision-language model we"ll be using.
 
 ```shell
-pip install outlines transformers torch
+pip install outlines transformers torch pillow
 ```
 
 ### Initializing the Model
-We'll use the `outlines.from_transformers` function to initialize our vision-language model. For this function to return a vision multi-modal model we need to pass in a transformers model and a transformers processor that can handle both text and image inputs. Today we'll be using the Pixtral model with the AutoProcessor.
+We"ll use the `outlines.from_transformers` function to initialize our vision-language model. For this function to return a vision multi-modal model we need to pass in a transformers model and a transformers processor that can handle both text and image inputs. Today we"ll be using the Pixtral model with the AutoProcessor.
 
 ```python
 import outlines
@@ -45,7 +45,7 @@ model = get_vision_model(model_name, model_class, processor_class)
 ```
 
 ### Defining the Schema
-Next, we'll define a schema for the output we expect from our vision multi-modal model. This schema will help structure the model's responses. We use the `outlines.Generator` object to create a generator for our schema that will then be called with our prompt and images.
+Next, we will define a schema for the output we expect from our vision multi-modal model. This schema will help structure the model's responses. We use the `outlines.Generator` object to create a generator for our schema that will then be called with our prompt and images.
 
 ```python
 from enum import Enum
@@ -132,11 +132,12 @@ pixtral_instruction = """
 </Examples>
 </TagCategories>
 <ShortCaption note="The short caption should be a concise single sentence caption of the image content with a maximum length of 100 characters.">
-<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">\n[IMG]<image>[/INST]
+<DenseCaption note="The dense caption should be a descriptive but grounded narrative paragraph of the image content with high quality narrative prose. It should incorporate elements from each of the tag categories to provide a broad dense caption">
+[IMG]<image>[/INST]
 """.strip()
 ```
 
-This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required. It's essential to include the <image> tag in the prompt at the location where the image will be inserted.
+This prompt provides detailed instructions to the model on how to generate comprehensive tag lists, captions, and dense captions for image analysis. Because of the ordering of the instructions the original tag generation serves as a sort of visual grounding for the captioning task, reducing the amount of manual post processing required. It is essential to include the <image> tag in the prompt at the location where the image will be inserted.
 
 ### Generating Structured Output
 Now we can use our model to generate structured output based on an input image:
@@ -162,33 +163,52 @@ print(result)
 This code loads an image from a URL, passes it to our vision multi-modal model along with the instruction prompt, and generates a structured output based on the defined schema. We end up with an output like this, ready to be used for the next stage in your pipeline:
 
 ```json
-{'tags_list': [{'tag': 'astronaut',
-   'category': <TagType.ENTITY: 'Entity'>,
-   'confidence': 0.99},
-  {'tag': 'moon', 'category': <TagType.ENTITY: 'Entity'>, 'confidence': 0.98},
-  {'tag': 'space suit',
-   'category': <TagType.ATTRIBUTE: 'Attribute'>,
-   'confidence': 0.97},
-  {'tag': 'lunar module',
-   'category': <TagType.ENTITY: 'Entity'>,
-   'confidence': 0.95},
-  {'tag': 'shadow of astronaut',
-   'category': <TagType.COMPOSITION: 'Composition'>,
-   'confidence': 0.95},
-  {'tag': 'footprints in moon dust',
-   'category': <TagType.CONTEXTUAL: 'Contextual'>,
-   'confidence': 0.93},
-  {'tag': 'low angle shot',
-   'category': <TagType.TECHNICAL: 'Technical'>,
-   'confidence': 0.92},
-  {'tag': 'human first steps on the moon',
-   'category': <TagType.SEMANTIC: 'Semantic'>,
-   'confidence': 0.95}],
- 'short_caption': 'First man on the Moon',
- 'dense_caption': "The figure clad in a pristine white space suit, emblazoned with the American flag, stands powerfully on the moon's desolate and rocky surface. The lunar module, a workhorse of space engineering, looms in the background, its metallic legs sinking slightly into the dust where footprints and tracks from the mission's journey are clearly visible. The photograph captures the astronaut from a low angle, emphasizing his imposing presence against the desolate lunar backdrop. The stark contrast between the blacks and whiteslicks of lost light and shadow adds dramatic depth to this seminal moment in human achievement."}
+{"tags_list": [
+  {
+    "tag": "astronaut",
+    "category": <TagType.ENTITY: "Entity">,
+    "confidence": 0.99
+  },
+  {"tag": "moon", "category": <TagType.ENTITY: "Entity">, "confidence": 0.98},
+  {
+    "tag": "space suit",
+    "category": <TagType.ATTRIBUTE: "Attribute">,
+    "confidence": 0.97
+  },
+  {
+    "tag": "lunar module",
+    "category": <TagType.ENTITY: "Entity">,
+    "confidence": 0.95
+  },
+  {
+    "tag": "shadow of astronaut",
+    "category": <TagType.COMPOSITION: "Composition">,
+    "confidence": 0.95
+  },
+  {
+    "tag": "footprints in moon dust",
+    "category": <TagType.CONTEXTUAL: "Contextual">,
+    "confidence": 0.93
+  },
+  {
+    "tag": "low angle shot",
+    "category": <TagType.TECHNICAL: "Technical">,
+    "confidence": 0.92
+  },
+  {
+    "tag": "human first steps on the moon",
+    "category": <TagType.SEMANTIC: "Semantic">,
+    "confidence": 0.95
+  }],
+  "short_caption": "First man on the Moon",
+  "dense_caption": "The figure clad in a pristine white space suit, emblazoned with the American flag, stands powerfully on the moon's desolate and rocky surface. The lunar module, a workhorse of space engineering, looms in the background, its metallic legs sinking slightly into the dust where footprints and tracks from the mission's journey are clearly visible. The photograph captures the astronaut from a low angle, emphasizing his imposing presence against the desolate lunar backdrop. The stark contrast between the blacks and whiteslicks of lost light and shadow adds dramatic depth to this seminal moment in human achievement."
+}
 ```
 
 ## Conclusion
-The `TransformersMultiModal` model in Outlines provides a powerful way to work with vision-language models. It allows for structured generation of outputs that combine image analysis with natural language processing, opening up possibilities for complex tasks like detailed image captioning, visual question answering, and more.
 
-By leveraging the capabilities of models like Pixtral-12B and the structured output generation of Outlines, you can create sophisticated applications that understand and describe visual content in a highly structured and customizable manner.
+This guide demonstrated how Outlines enables structured output generation with vision-language models. With the techniques shown above, you can build:
+
+- **Content Management Systems**: Automatically tag and categorize visual content with structured metadata that can be directly stored in databases, enabling powerful search and filtering capabilities
+- **Accessibility Tools**: Generate rich, structured descriptions of images that can be adapted for different contexts - from brief alt-text to detailed scene descriptions for screen readers
+- **Quality Assurance Pipelines**: Validate visual content against specific criteria by extracting structured attributes and checking them against business rules
diff --git a/mkdocs.yml b/mkdocs.yml
index 438a9d62c..ac723ccc0 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -120,6 +120,7 @@ nav:
           - Getting Started: guide/getting_started.md
           - Installation: guide/installation.md
           - Migrating to v1: guide/migration.md
+          - Vision-Language Models: guide/vlm.md
           - Deploying with FastAPI: guide/fastapi_vllm_deployment.md
           - Chat Templating for Instruct Models: guide/chat_templating.md
 
@@ -175,7 +176,6 @@ nav:
           - Structured Generation Workflow: examples/structured_generation_workflow.md
           - Chain of Thought (CoT): examples/chain_of_thought.md
           - ReAct Agent: examples/react_agent.md
-          - Vision-Language Models: examples/atomic_caption.md
           - Structured Generation from PDFs: examples/read-pdfs.md
           - Earnings Reports to CSV: examples/earnings-reports.md
           - Receipt Digitization: examples/receipt-digitization.md

From a97696fb035ee3d6600f3ceaac295600b210cce9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 23 Jun 2025 11:42:23 +0200
Subject: [PATCH 370/505] Fix example typo in README

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 390d51409..bcbb97ee2 100644
--- a/README.md
+++ b/README.md
@@ -135,7 +135,7 @@ review = model(
 review = ProductReview.model_validate_json(review)
 print(f"Rating: {review.rating.name}")  # "Rating: good"
 print(f"Pros: {review.pros}")           # "Pros: ['great battery life', 'stunning display']"
-print(f"Summary: {review.summary}")     # "Summary: Good laptop with great display but thermal issues"t(result)
+print(f"Summary: {review.summary}")     # "Summary: Good laptop with great display but thermal issues"
 ```
 
 ## Real-world examples

From f1e3e1387a20f85aee2872b49d1068be1b0407c8 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Mon, 23 Jun 2025 12:10:19 +0200
Subject: [PATCH 371/505] Add testing and documentation for vision input for
 VLLM model (#1648)

Closes #1639
---
 docs/features/models/vllm.md | 52 +++++++++++++++++++++++++---------
 tests/models/test_vllm.py    | 55 +++++++++++++++++++++++++++++++++++-
 2 files changed, 93 insertions(+), 14 deletions(-)

diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index 3187e37e5..d26045eb8 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -23,8 +23,8 @@ import openai
 import outlines
 
 # Create the OpenAI client
-sync_openai_client = openai.OpenAI(base_url="http://localhost:8080")
-async_openai_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+sync_openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
+async_openai_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 
 # Create a sync model
 sync_model = outlines.from_vllm(sync_openai_client)
@@ -46,7 +46,7 @@ import openai
 import outlines
 
 # Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 model = outlines.from_vllm(openai_client)
 
 # Call it to generate text
@@ -61,7 +61,7 @@ import openai
 import outlines
 
 # Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 model = outlines.from_vllm(openai_client)
 
 # Stream text
@@ -69,6 +69,32 @@ for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
     print(chunk) # 'In ...'
 ```
 
+Additionnaly, you can use the VLLM model with vision input if your server is running a multimodal model such as Qwen2.5-VL. For instance:
+
+```python
+import io
+import requests
+import PIL
+import openai
+import outlines
+from outlines.templates import Vision
+
+# Create the model
+model = outlines.from_vllm(openai.OpenAI(base_url="http://0.0.0.0:8000/v1"), "Qwen/Qwen2.5-VL-3B-Instruct")
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt
+prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+
+# Generate text
+response = model(prompt)
+print(response) # The image shows a black puppy lying on a wooden surface, looking up at the camera with wide eyes.
+```
+
 ## Asynchronous Calls
 
 vLLM supports asynchronous operations by passing an `AsyncOpenAI` client instead of a regular `OpenAI` client. This returns an `AsyncVLLM` model instance that supports async/await patterns.
@@ -81,7 +107,7 @@ import openai
 import outlines
 
 async def generate_text():
-    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
     async_model = outlines.from_vllm(async_client)
 
     result = await async_model("Write a haiku about Python.", max_tokens=50)
@@ -100,7 +126,7 @@ import openai
 import outlines
 
 async def stream_text():
-    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
     async_model = outlines.from_vllm(async_client)
 
     async for chunk in async_model.stream("Tell me a story about a robot.", max_tokens=100):
@@ -119,7 +145,7 @@ import openai
 import outlines
 
 async def generate_multiple():
-    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
     async_model = outlines.from_vllm(async_client)
 
     prompts = [
@@ -149,7 +175,7 @@ import outlines
 
 output_type = int
 
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 model = outlines.from_vllm(openai_client)
 
 result = model("How many countries are there in the world?", output_type)
@@ -169,7 +195,7 @@ class Character(BaseModel):
     age: int
     skills: List[str]
 
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 model = outlines.from_vllm(openai_client)
 
 result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
@@ -186,7 +212,7 @@ import outlines
 
 output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 model = outlines.from_vllm(openai_client)
 
 result = model("What is the capital of France?", output_type, temperature=0)
@@ -202,7 +228,7 @@ from outlines.types import Regex
 
 output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 model = outlines.from_vllm(openai_client)
 
 result = model("Generate a fake social security number.", output_type, top_p=0.1)
@@ -238,7 +264,7 @@ arithmetic_grammar = """
 """
 output_type = CFG(arithmetic_grammar)
 
-openai_client = openai.OpenAI(base_url="http://localhost:8080")
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
 model = outlines.from_vllm(openai_client)
 
 result = model("Write an addition.", output_type, extra_body={"guided_decoding_backend": "outlines"})
@@ -261,7 +287,7 @@ class User(BaseModel):
     age: int
 
 async def generate_user():
-    async_client = openai.AsyncOpenAI(base_url="http://localhost:8080")
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
     async_model = outlines.from_vllm(async_client)
 
     result = await async_model("Generate a random user profile.", output_type=User)
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index 76a9da814..b907a2652 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -1,12 +1,16 @@
+import io
 import os
 import re
 import warnings
+import base64
 from typing import AsyncGenerator, Generator
 
 import pytest
+from PIL import Image
 from openai import AsyncOpenAI, OpenAI
 
 from outlines.models.vllm import VLLM, AsyncVLLM, from_vllm
+from outlines.templates import Vision
 from outlines.types.dsl import CFG, Regex, JsonSchema
 from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
 
@@ -22,7 +26,7 @@
 # Otherwise, use the mock server
 vllm_server_url = os.environ.get("VLLM_SERVER_URL")
 vllm_model_name = os.environ.get(
-    "VLLM_MODEL_NAME", "microsoft/Phi-3-mini-4k-instruct"
+    "VLLM_MODEL_NAME", "Qwen/Qwen2.5-VL-3B-Instruct"
 )
 if vllm_server_url:
     openai_client = OpenAI(base_url=vllm_server_url)
@@ -32,6 +36,17 @@
     openai_client = MockOpenAIClient()
     async_openai_client = MockAsyncOpenAIClient()
 
+# Create base64 encoded image for mock responses
+def get_base64_image():
+    width, height = 256, 256
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
+    return image_str
+
 mock_responses = [
     (
         {
@@ -52,6 +67,18 @@
         },
         ["foo", "bar"]
     ),
+    (
+        {
+            'messages': [
+                {'role': "user", 'content': [
+                    {'type': 'text', 'text': 'Describe the image.'},
+                    {'type': 'image_url', 'image_url': {'url': 'data:image/png;base64,' + get_base64_image()}}
+                ]}
+            ],
+            'model': vllm_model_name,
+        },
+        "This is a white square."
+    ),
     (
         {
             'messages': [
@@ -130,6 +157,21 @@ def async_model_no_model_name():
     return AsyncVLLM(async_openai_client)
 
 
+@pytest.fixture(scope="session")
+def image():
+    width, height = 256, 256
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
 def test_vllm_init():
     # We do not rely on the mock server here because we need an object
     # of type OpenAI and AsyncOpenAI to test the init function.
@@ -178,6 +220,11 @@ def test_vllm_sync_streaming(sync_model_no_model_name):
     assert isinstance(next(result), str)
 
 
+def test_vllm_sync_vision(sync_model, image):
+    result = sync_model(Vision("Describe the image.", image))
+    assert isinstance(result, str)
+
+
 def test_vllm_sync_multiple_samples(sync_model):
     result = sync_model("Respond with a single word.", n=2)
     assert isinstance(result, list)
@@ -223,6 +270,12 @@ async def test_vllm_async_streaming(async_model_no_model_name):
         break  # Just check the first chunk
 
 
+@pytest.mark.asyncio
+async def test_vllm_async_vision(async_model, image):
+    result = await async_model(Vision("Describe the image.", image))
+    assert isinstance(result, str)
+
+
 @pytest.mark.asyncio
 async def test_vllm_async_multiple_samples(async_model):
     result = await async_model("Respond with a single word.", n=2)

From e47df45ef381684bbcc9ed288c0e72cd26b9e0c9 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Mon, 23 Jun 2025 12:25:42 +0200
Subject: [PATCH 372/505] Add support for vision input for the Ollama model
 (#1647)

Closes #1638
---
 docs/features/models/index.md            |  2 +-
 docs/features/models/ollama.md           | 29 ++++++++++++++++++++
 outlines/models/ollama.py                | 35 ++++++++++++++++++------
 tests/models/test_ollama.py              | 28 +++++++++++++++++++
 tests/models/test_ollama_type_adapter.py | 16 +++++++++--
 tests/models/test_openai.py              |  1 -
 6 files changed, 98 insertions(+), 13 deletions(-)

diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index c597599de..b9f747320 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -58,7 +58,7 @@ In alphabetical order:
 | **Generation Features** | | | | | | | | | | | | | |
 | Async | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
-| Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
+| Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
 | Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ |
 
 ## Model Types
diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
index eaad4c3e2..60043e38d 100644
--- a/docs/features/models/ollama.md
+++ b/docs/features/models/ollama.md
@@ -70,6 +70,35 @@ for chunk in model.stream("Write a short story about a cat"):
     print(chunk) # 'In...'
 ```
 
+Additionally, you can use `Ollama` with the `Vision` input if you're running a vision model such as qwen2.5vl. For instance:
+
+```python
+import io
+import requests
+import PIL
+import ollama
+import outlines
+from outlines.templates import Vision
+
+# Create the model
+model = outlines.from_ollama(
+    ollama.Client(),
+    "qwen2.5vl:3b"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt
+prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+
+# Generate text
+response = model(prompt)
+print(response) # The image shows a black puppy with a curious and attentive expression.
+```
+
 ## Structured Generation
 
 Ollama only provides support for structured generation based on a JSON schema. To use it, call the model with a JSON schema object as an `output_type` on top of your prompt.
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 862c71f69..b41e28fe6 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,11 +1,12 @@
 """Integration with the `ollama` library."""
 
 import json
-from typing import TYPE_CHECKING, Any, Iterator, Optional
+from typing import TYPE_CHECKING, Any, Iterator, Optional, Union
 
 from pydantic import TypeAdapter
 
 from outlines.models.base import Model, ModelTypeAdapter
+from outlines.templates import Vision
 from outlines.types import CFG, JsonSchema, Regex
 from outlines.types.utils import (
     is_dataclass,
@@ -23,7 +24,7 @@
 class OllamaTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `Ollama` model."""
 
-    def format_input(self, model_input: str) -> str:
+    def format_input(self, model_input: Union[str, Vision]) -> dict:
         """Generate the prompt argument to pass to the model.
 
         Parameters
@@ -33,18 +34,34 @@ def format_input(self, model_input: str) -> str:
 
         Returns
         -------
-        str
-            The formatted input to be passed to the model.
+        dict
+            The model input to be passed to the client.
 
         """
         if isinstance(model_input, str):
-            return model_input
+            return self.format_str_model_input(model_input)
+        elif isinstance(model_input, Vision):
+            return self.format_vision_model_input(model_input)
         raise TypeError(
             f"The input type {model_input} is not available. "
-            "Ollama does not support batch inference."
+            "The only available types are `str` and `Vision`."
         )
 
-    def format_output_type(self, output_type: Optional[Any] = None) -> Optional[str]:
+    def format_str_model_input(self, model_input: str) -> dict:
+        """Format the string model input to pass to the client.
+
+        """
+        return {"prompt": model_input}
+
+    def format_vision_model_input(self, model_input: Vision) -> dict:
+        """Format the vision model input to pass to the client.
+
+        """
+        return {"prompt": model_input.prompt, "images": [model_input.image_str]}
+
+    def format_output_type(
+        self, output_type: Optional[Any] = None
+    ) -> Optional[str]:
         """Format the output type to pass to the client.
 
         TODO: `int`, `float` and other Python types could be supported via
@@ -147,7 +164,7 @@ def generate(self,
             kwargs["model"] = self.model_name
 
         response = self.client.generate(
-            prompt=self.type_adapter.format_input(model_input),
+            **self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             **kwargs,
         )
@@ -182,7 +199,7 @@ def generate_stream(
             kwargs["model"] = self.model_name
 
         response = self.client.generate(
-            prompt=self.type_adapter.format_input(model_input),
+            **self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             stream=True,
             **kwargs,
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index 408fa2ceb..189066d48 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -1,13 +1,16 @@
+import io
 import json
 from enum import Enum
 from typing import Annotated
 
 import pytest
+from PIL import Image
 from ollama import Client as OllamaClient
 from pydantic import BaseModel, Field
 
 import outlines
 from outlines.models import Ollama
+from outlines.templates import Vision
 
 
 MODEL_NAME = "tinyllama"
@@ -23,6 +26,21 @@ def model_no_model_name():
     return Ollama(OllamaClient())
 
 
+@pytest.fixture(scope="session")
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = Image.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = Image.open(buffer)
+
+    return image
+
+
 def test_init_from_client():
     client = OllamaClient()
 
@@ -62,6 +80,16 @@ def test_ollama_direct(model_no_model_name):
     assert isinstance(result, str)
 
 
+def test_ollama_simple_vision(image, model):
+    # This is not using a vision model, so it's not able to describe
+    # the image, but we're still checking the model input syntax
+    result = model.generate(
+        Vision("What does this logo represent?", image),
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
 def test_ollama_json(model):
     class Foo(BaseModel):
         foo: Annotated[str, Field(max_length=1)]
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
index 58b6c6aa9..feced7834 100644
--- a/tests/models/test_ollama_type_adapter.py
+++ b/tests/models/test_ollama_type_adapter.py
@@ -54,11 +54,23 @@ def adapter():
 def test_ollama_type_adapter_input_text(adapter):
     message = "prompt"
     result = adapter.format_input(message)
-    assert result == message
+    assert isinstance(result, dict)
+    assert result.get("prompt") == message
 
 
-def test_ollama_type_adapter_input_invalid(adapter, image):
+def test_ollama_type_adapter_input_vision(adapter, image):
     prompt = Vision("prompt", image)
+    result = adapter.format_input(prompt)
+    assert isinstance(result, dict)
+    assert result.get("prompt") == prompt.prompt
+    assert result.get("images") == [prompt.image_str]
+
+
+def test_ollama_type_adapter_input_invalid(adapter):
+    prompt = [
+        "This is a first test",
+        "This is a second test",
+    ]
     with pytest.raises(TypeError, match="The input type"):
         _ = adapter.format_input(prompt)
 
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 455087707..6e387e961 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -4,7 +4,6 @@
 from typing import Annotated, Generator
 
 import pytest
-import requests
 from PIL import Image
 from openai import OpenAI as OpenAIClient
 from pydantic import BaseModel, Field

From 4dd7802a8a5021f31dac75bc652f88d410a28675 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Mon, 23 Jun 2025 17:19:55 +0200
Subject: [PATCH 373/505] Add setup example for local server (#1646)

Closes #1645
---
 docs/features/models/sglang.md | 15 ++++++++++++++-
 docs/features/models/tgi.md    | 14 +++++++++++++-
 docs/features/models/vllm.md   | 14 +++++++++++++-
 3 files changed, 40 insertions(+), 3 deletions(-)

diff --git a/docs/features/models/sglang.md b/docs/features/models/sglang.md
index 0df6a54db..7087be9e2 100644
--- a/docs/features/models/sglang.md
+++ b/docs/features/models/sglang.md
@@ -6,7 +6,20 @@ title: SGLang
 
 ## Prerequisites
 
-The Outlines `SGLang` model is intended to be used along with an SGLang instance running on a separate server (can be local or remote). Make sure you have a SGLang server running and accessible before using the `SGLang` model. As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies of the `SGLang` model with: `pip install outlines[sglang]`.
+The Outlines `SGLang` model is intended to be used along with an SGLang instance running on a separate server (can be local or remote). Make sure you have a SGLang server running and accessible before using the `SGLang` model. For instance by running:
+
+```shell
+pip install "sglang[all]"
+
+python -m sglang.launch_server \
+  --model-path NousResearch/Meta-Llama-3-8B-Instruct \
+  --host 0.0.0.0 \
+  --port 30000
+```
+
+Follow the [Installation instructions](https://docs.sglang.ai/start/install.html) for more information on how to set up a SGLang server for your particular setup.
+
+As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies of the `SGLang` model with: `pip install outlines[sglang]`.
 
 When launching your SGLang server, you can specify the backend engine to use for structured generation through the `grammar-backend` cli argument. Add `--grammar-backend outlines` to your command to use Outlines instead of the default engine.
 
diff --git a/docs/features/models/tgi.md b/docs/features/models/tgi.md
index 38b1a1ae5..f71e8dad6 100644
--- a/docs/features/models/tgi.md
+++ b/docs/features/models/tgi.md
@@ -6,7 +6,19 @@ title: TGI
 
 ## Prerequisites
 
-The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. As the TGI client relies on the `huggingface_hub` python package, you need to have it installed. Install all optional dependencoes of the `TGI` model with: `pip install outlines[tgi]`
+The Outlines `TGI` model is intended to be used along with a HuggingFace `Text Generation Inference` server (running locally or remotely). Make sure you have a TGI server running before using the `TGI` model. For instance running:
+
+```shell
+docker run \
+  --gpus all \
+  --shm-size 1g \
+  -p 8080:80 \
+  ghcr.io/huggingface/text-generation-inference:3.3.4 \
+  --model-id NousResearch/Meta-Llama-3-8B-Instruct
+```
+
+Please consult the [installation guide](https://huggingface.co/docs/text-generation-inference/en/quicktour) for more information about how to run TGI with your particular setup.
+As the TGI client relies on the `huggingface_hub` python package, you need to have it installed. Install all optional dependencoes of the `TGI` model with: `pip install outlines[tgi]`
 
 ## Model Initialization
 
diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index d26045eb8..9396cd328 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -6,7 +6,19 @@ title: vLLM
 
 ## Prerequisites
 
-The Outlines `VLLM` model is intended to be used along with a vLLM instance running on a separate server (can be local or remote). Make sure you have a vLLM server running and accessible before using the `VLLM` model. As the vLLM client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies for the `VLLM` model with: `pip install openai`.
+The Outlines `VLLM` model is intended to be used along with a vLLM instance running on a separate server (can be local or remote). Make sure you have a vLLM server running and accessible before using the `VLLM` model. For instance by running:
+
+```shell
+pip install vllm
+
+vllm serve NousResearch/Meta-Llama-3-8B-Instruct \
+  --dtype auto \
+  --api-key token-abc123
+```
+
+Follow the [Installation instructions](https://docs.vllm.ai/en/latest/getting_started/installation/index.html) for more information on how to set up a vLLM server for your particular setup.
+
+As the vLLM client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies for the `VLLM` model with: `pip install openai`.
 
 If you want to use the vllm offline inference mode instead of the server mode, please refer to the [VLLMOffline](./vllm_offline.md) model documentation.
 

From eae004c773da8920ad49daabfb852db542ef9f78 Mon Sep 17 00:00:00 2001
From: Jacob Marks <jamarks13@gmail.com>
Date: Wed, 25 Jun 2025 22:45:57 -0700
Subject: [PATCH 374/505] Add IPv4, UUID4, and Hex String Types (#1655)

Adds these three new types to the regex DSL, along with test cases and
documentation.
---
 docs/features/utility/regex_dsl.md |  9 ++++++---
 outlines/types/__init__.py         | 12 +++++++++++
 tests/types/test_custom_types.py   | 32 ++++++++++++++++++++++++++++++
 3 files changed, 50 insertions(+), 3 deletions(-)

diff --git a/docs/features/utility/regex_dsl.md b/docs/features/utility/regex_dsl.md
index 937827450..7fb72dc9e 100644
--- a/docs/features/utility/regex_dsl.md
+++ b/docs/features/utility/regex_dsl.md
@@ -241,14 +241,16 @@ The DSL comes "batteries included" with types that represent common text constru
 - `number` represents a floating-point number recognize by Python's `float`
 - `date` represents a date as understood by `datetime.date`
 - `time` represents a time as understood by `datetime.time`
-- `datetime` represents a time as understoof by `datetime.datetime`
+- `datetime` represents a time as understood by `datetime.datetime`
 - `digit` represents a single digit
 - `char` represents a single character
 - `newline` represents a new line character
 - `whitespace` represents a white space
+- `hex_str` represents a hexadecimal string, optionally prefixed with "0x"
+- `uuid4` represents a UUID version 4 string in the format "xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx"
+- `ipv4` represents an IPv4 address in the format "xxx.xxx.xxx.xxx" where each octet is between 0 and 255
 - `sentence` represents a sentence
-- `paragraph` reprensents a pagraph (one or more sentences separated by one or more ilne breaks)
-
+- `paragraph` represents a paragraph (one or more sentences separated by one or more line breaks)
 
 For instance you can describe the answers in the GSM8K dataset using the following pattern:
 
@@ -265,6 +267,7 @@ answer = "A: " + sentence.between(2,4) + " So the answer is: " + digit.between(1
 ### Example 1: Matching a Custom ID Format
 
 Suppose you want to create a regex that matches an ID format like "ID-12345", where:
+
 - The literal "ID-" must be at the start.
 - Followed by exactly 5 digits.
 
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 85af9adc0..208f4f13b 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -47,6 +47,18 @@
 char = Regex(r"\w")
 newline = Regex(r"(\r\n|\r|\n)")  # Matched new lines on Linux, Windows & MacOS
 whitespace = Regex(r"\s")
+hex_str = Regex(r"(0x)?[a-fA-F0-9]+")
+uuid4 = Regex(
+    r"[a-fA-F0-9]{8}-"
+    r"[a-fA-F0-9]{4}-"
+    r"4[a-fA-F0-9]{3}-"
+    r"[89abAB][a-fA-F0-9]{3}-"
+    r"[a-fA-F0-9]{12}"
+)
+ipv4 = Regex(
+    r"((25[0-5]|2[0-4][0-9]|1?[0-9]{1,2})\.){3}"
+    r"(25[0-5]|2[0-4][0-9]|1?[0-9]{1,2})"
+)
 
 # Document-specific types
 sentence = Regex(r"[A-Z].*\s*[.!?]")
diff --git a/tests/types/test_custom_types.py b/tests/types/test_custom_types.py
index 691b681b8..3bf8af930 100644
--- a/tests/types/test_custom_types.py
+++ b/tests/types/test_custom_types.py
@@ -61,6 +61,38 @@
         (types.paragraph, "One sentence. Two sentences.\n\n", True),
         (types.paragraph, "One sentence. invalid sentence.", False),
         (types.paragraph, "One sentence. Invalid sentence\n", False),
+        (types.hex_str, "0x123", True),
+        (types.hex_str, "0xABC", True),
+        (types.hex_str, "0xabc", True),
+        (types.hex_str, "0x123ABC", True),
+        (types.hex_str, "123", True),
+        (types.hex_str, "ABC", True),
+        (types.hex_str, "abc", True),
+        (types.hex_str, "123ABC", True),
+        (types.hex_str, "0xg123", False),
+        (types.hex_str, "0x", False),
+        (types.hex_str, "0x123G", False),
+        (types.uuid4, "123e4567-e89b-42d3-a456-426614174000", True),
+        (types.uuid4, "00000000-0000-4000-8000-000000000000", True),
+        (types.uuid4, "123e4567-e89b-12d3-a456-426614174000", False),
+        (types.uuid4, "123e4567-e89b-12d3-a456-42661417400", False),
+        (types.uuid4, "123e4567-e89b-12d3-a456-4266141740000", False),
+        (types.uuid4, "123e4567-e89b-12d3-x456-426614174000", False),
+        (types.uuid4, "123e4567-e89b-12d3-a456-42661417400g", False),
+        (types.ipv4, "192.168.1.1", True),
+        (types.ipv4, "10.0.0.1", True),
+        (types.ipv4, "172.16.0.1", True),
+        (types.ipv4, "255.255.255.255", True),
+        (types.ipv4, "0.0.0.0", True),
+        (types.ipv4, "256.1.2.3", False),
+        (types.ipv4, "1.256.2.3", False),
+        (types.ipv4, "1.2.256.3", False),
+        (types.ipv4, "1.2.3.256", False),
+        (types.ipv4, "1.2.3", False),
+        (types.ipv4, "1.2.3.4.5", False),
+        (types.ipv4, "1.2.3.4.", False),
+        (types.ipv4, ".1.2.3.4", False),
+        (types.ipv4, "1..2.3.4", False),
     ],
 )
 def test_type_regex(custom_type, test_string, should_match):

From 9c3c9bf0c4a8699945e887f1f283b5affd4cd8b6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remi@thetypicalset.com>
Date: Thu, 26 Jun 2025 12:30:44 +0200
Subject: [PATCH 375/505] Fix the FastAPI example

---
 docs/guide/fastapi_vllm_deployment.md | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/docs/guide/fastapi_vllm_deployment.md b/docs/guide/fastapi_vllm_deployment.md
index d87886ae7..a5660aeb6 100644
--- a/docs/guide/fastapi_vllm_deployment.md
+++ b/docs/guide/fastapi_vllm_deployment.md
@@ -116,7 +116,7 @@ async def lifespan(app: FastAPI):
         base_url="http://localhost:8000/v1",  # Adjust to your vLLM server URL
         api_key="dummy"  # vLLM doesn't require a real API key
     )
-    async_model = models.from_vllm(client)
+    async_model = models.from_vllm(client, "Qwen/Qwen2.5-VL-7B-Instruct")
 
     yield
 
@@ -173,7 +173,7 @@ async def generate_response(
 
     try:
         # Generate and parse a structured response
-        result = await async_model(prompt, SupportResponse, max_tokens=800)
+        result = await async_model(prompt, SupportResponse, max_tokens=5000)
         response = SupportResponse.model_validate_json(result)
 
         return response
@@ -187,9 +187,7 @@ async def generate_response(
 ### Step 1: Start your vLLM server
 
 ```shell
-python -m vllm.entrypoints.openai.api_server \
-    --model meta-llama/Llama-2-7b-chat-hf \
-    --port 8000
+vllm serve Qwen/Qwen2.5-VL-7B-Instruct
 ```
 
 ### Step 2: Run the FastAPI application

From 6edb724ded5b33bc3d9e9e2eeb350a8bc91bd672 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Thu, 26 Jun 2025 11:19:37 +0200
Subject: [PATCH 376/505] Add llm.txt file

---
 llm.txt | 185 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 185 insertions(+)
 create mode 100644 llm.txt

diff --git a/llm.txt b/llm.txt
new file mode 100644
index 000000000..0bc07702f
--- /dev/null
+++ b/llm.txt
@@ -0,0 +1,185 @@
+# Outlines Codebase Reference
+
+## Overview
+
+Outlines is a library for structured generation for type-safe LLMs. It ensures outputs conform to specified formats (JSON schemas, regex patterns, grammars) by constraining the token generation process, or calling an API that uses this process.
+
+**Core insight**: Instead of generating text and hoping it matches a format, Outlines makes it impossible for the model to generate invalid outputs by masking invalid tokens during generation.
+
+**Note**: The codebase has undergone significant refactoring. Core FSM functionality has been extracted to the `outlines-core` package, and legacy implementations are preserved in `v0_legacy/` for backward compatibility.
+
+## Usage Examples
+
+For comprehensive usage examples, see:
+- **README.md**: Quick start examples for JSON generation, regex constraints, and choice selection
+- **docs/cookbook/**: Detailed examples including:
+  - `docs/cookbook/prompting.md`: Advanced prompting techniques
+  - `docs/cookbook/models.md`: Working with different model providers
+  - `docs/cookbook/humaneval.md`: Code generation examples
+  - `docs/cookbook/qa-with-citations.md`: Question answering with structured citations
+  - `docs/cookbook/deploy-to-servers.md`: Deployment examples with vLLM and TGI
+- **examples/**: Standalone example scripts
+  - `examples/lark_grammar.py`: Grammar-based generation
+  - `examples/math_generate_code.py`: Code generation with constraints
+  - `examples/multiple_sglang_backend.py`: Using multiple backend servers
+- **tests/**: Test files contain many practical usage patterns
+
+## Architecture
+
+### Layer Stack
+
+```
+User API (outlines.models)
+    ↓
+Generator Classes (SteerableGenerator, BlackBoxGenerator)
+    ↓
+Type System (types/dsl.py: Pydantic → JsonSchema → Regex)
+    ↓
+FSM Compilation (outlines-core: regex → FSM via interegular)
+    ↓
+Guide System (processors/guide.py: FSM state management)
+    ↓
+Logits Processing (processors/structured.py: token masking)
+    ↓
+Model Providers (transformers, OpenAI, etc.)
+```
+
+### Key Design Decisions
+
+1. **FSM-based constraints**: For local models, constraints compile to finite state machines that track valid next tokens
+2. **Provider abstraction**: Same constraint system works across local models (transformers) and APIs (OpenAI)
+3. **Lazy compilation**: FSMs are compiled on first use and cached persistently
+4. **Token-level control**: Constraints apply at the token level, not character level
+5. **Type-driven API**: Python types are the primary interface for specifying constraints
+
+## Core Components
+
+### Models (`outlines/models/`)
+Base classes and implementations for different model providers:
+- `SteerableModel`: For models where we control logits (transformers, llama.cpp)
+- `BlackBoxModel`: For API models with structured output support (OpenAI, Anthropic)
+- Each provider has an adapter class handling input and output format conversion
+
+Key files:
+- `base.py`: Abstract base classes defining the model interface
+- `transformers.py`: Integration with HuggingFace transformers
+- `openai.py`: OpenAI API integration
+- `gemini.py`: Gemini integration
+- `mlxlm.py`: MLX-LM integration
+- `vllm_offline.py`: vLLM integration
+- `llamacpp.py`: llama.cpp integration
+- `ollama.py`: Ollama integration
+- `vllm.py`: Integration with vLLM servers
+- `tgi.py`: Integration with text-generation-inferece servers
+- `sglang.py`: Integration with SGLang servers
+
+### Generation (`outlines/generator.py` and `outlines/v0_legacy/generate/`)
+Handles the generation process:
+- `generator.py`: Main `Generator` class implementations (root level)
+- `v0_legacy/generate/api.py`: Legacy generation API (backward compatibility)
+- Stream functionality is now integrated into generator classes
+
+Base classes and implementations for different model providers:
+- `BlackBoxGenerator`: For API models with structured outputs support
+- `SteerableGenerator`: For modesl where we control the logits
+
+### FSM System (`outlines/fsm/` and `outlines/processors/`)
+Core constraint enforcement:
+- `processors/guide.py`: Base `Guide` class and `RegexGuide` implementation
+- `fsm/parsing.py`: Lark-based CFG parsing with `PartialLark` parser
+- Regex to FSM compilation now uses `outlines_core.fsm` module
+
+Key concepts:
+- **Guide**: Manages FSM state during generation
+- **State transitions**: Precomputed mapping of (state, token) → next_state
+- **Token masking**: For each state, compute which tokens are valid
+
+### Type System (`outlines/types/`)
+Type conversion pipeline:
+- `dsl.py`: Term DSL defining constraint language (Sequence, Choice, etc.) and JSON schema to regex conversion
+- `__init__.py`: Common regex types and DSL functions
+- Python types → Term DSL → Regex → FSM
+
+### Logits Processors (`outlines/processors/`)
+Apply constraints during generation:
+- `structured.py`: Main `StructuredLogitsProcessor`
+- `base_logits_processor.py`: Abstract base class
+- Processors mask invalid tokens by setting their logits to -inf
+
+## Key Algorithms
+
+### FSM Compilation Pipeline
+1. **Pattern definition**: User provides Pydantic model, regex, or grammar
+2. **Schema to regex**: Convert complex types to regex patterns
+   - JSON schemas become regex matching valid JSON
+   - Pydantic models extract JSON schema then convert
+3. **Regex to FSM**: Use interegular library to build FSM
+4. **FSM to token map**: For each FSM state, compute valid tokens
+   - Handle multi-character tokens
+   - Account for token boundaries
+5. **Guide creation**: Wrap FSM with state tracking
+
+### Token Masking Process
+```python
+# Simplified logits processing
+def process_logits(logits, current_state, guide):
+    valid_tokens = guide.get_valid_tokens(current_state)
+    mask = torch.full_like(logits, -float('inf'))
+    mask[valid_tokens] = 0
+    return logits + mask
+```
+
+## File Organization
+
+```
+outlines/
+├── __init__.py              # Public API exports
+├── generator.py             # Main Generator classes
+├── models/                  # Model integrations
+│   ├── base.py             # Abstract base classes
+│   ├── transformers.py     # HuggingFace support
+│   └── [provider].py       # Other providers (openai, anthropic, etc.)
+├── fsm/                     # FSM engine
+│   ├── __init__.py
+│   └── parsing.py          # Grammar parsing
+├── types/                   # Type system
+│   ├── __init__.py         # Common regex types and DSL exports
+│   ├── dsl.py              # Term DSL and JSON schema conversion
+│   └── utils.py            # Type checking utilities
+├── processors/              # Logits processing and guides
+│   ├── guide.py            # Guide implementations
+│   ├── structured.py       # Main processor
+│   └── tensor_adapters/    # Framework-specific tensor handling
+├── caching.py               # Caching system
+├── grammars/                # Grammar files (.lark)
+└── v0_legacy/              # Backward compatibility
+    ├── generate/           # Legacy generation API
+    └── models/             # Legacy model integrations
+```
+
+## Extension Points
+
+### Adding a Model Provider
+1. Create model class inheriting from `SteerableModel` or `BlackBoxModel`
+2. Implement required methods: `generate()`, `generate_stream()`
+3. Add constructor function in `outlines/__init__.py`
+4. Handle provider-specific input and structured output formats with a `TypeAdapter`
+
+### Adding a Constraint Type
+1. Define new Term subclass in `types/dsl.py`
+2. Implement `to_regex()` conversion
+3. Register type handler for Python type conversion in `python_types_to_terms()`
+4. Add tests for FSM compilation
+
+### Custom Logits Processor
+1. Inherit from `OutlinesLogitsProcessor`
+2. Implement `process_logits()` method
+3. Handle batch processing and state management
+4. Register with generator
+
+## Common Patterns in Codebase
+
+1. **Factory functions**: `from_transformers()`, `from_openai()` hide complexity
+2. **Abstract base classes**: Define interfaces for models, processors, guides
+3. **Lazy imports**: Optional dependencies imported only when needed
+5. **Type adapters**: Convert between Outlines types and provider formats

From 4a03f51016558ad20f6c07ea0843848b3e9409b2 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 26 Jun 2025 14:02:43 +0200
Subject: [PATCH 377/505] Fix nightly tests (remove Slack alert causing
 troubles)

---
 .github/workflows/tests_nightly.yml | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/.github/workflows/tests_nightly.yml b/.github/workflows/tests_nightly.yml
index c34a707fa..183cce207 100644
--- a/.github/workflows/tests_nightly.yml
+++ b/.github/workflows/tests_nightly.yml
@@ -47,14 +47,6 @@ jobs:
         echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
-        uv run pytest -x --cov=outlines
+        uv run pytest -x --cov=outlines -m 'api_call' --ignore=tests/models/test_dottxt.py
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
-    - name: Slack notification on failure
-      if: failure()
-      uses: 8398a7/action-slack@v3
-      with:
-        status: ${{ job.status }}
-        fields: commit,author,ref
-      env:
-        SLACK_WEBHOOK_URL: ${{ secrets.SLACK_TESTS_ERRORS_WEBHOOK_URL }}

From 696d89c86e0f3e59721adb3d29172cfd6ae54789 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 30 Jun 2025 16:38:17 +0200
Subject: [PATCH 378/505] Fix error with additionalProperties for the OpenAI
 model

---
 outlines/models/openai.py  |  6 ++---
 outlines/models/utils.py   | 27 ++++++++++++++++++++
 pyproject.toml             |  2 ++
 tests/models/test_utils.py | 50 ++++++++++++++++++++++++++++++++++++++
 4 files changed, 82 insertions(+), 3 deletions(-)
 create mode 100644 outlines/models/utils.py
 create mode 100644 tests/models/test_utils.py

diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 2112e1974..2e4209cc9 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -20,6 +20,7 @@
 
 from outlines.caching import cache
 from outlines.models.base import Model, ModelTypeAdapter
+from outlines.models.utils import set_additional_properties_false_json_schema
 from outlines.templates import Vision
 from outlines.types import JsonSchema, Regex, CFG
 from outlines.types.utils import (
@@ -166,9 +167,8 @@ def format_json_output_type(self, schema: dict) -> dict:
         specified a `Json` output type.
 
         """
-        # OpenAI requires `additionalProperties` to be set
-        if "additionalProperties" not in schema:
-            schema["additionalProperties"] = False
+        # OpenAI requires `additionalProperties` to be set to False
+        schema = set_additional_properties_false_json_schema(schema)
 
         return {
             "response_format": {
diff --git a/outlines/models/utils.py b/outlines/models/utils.py
new file mode 100644
index 000000000..ee9beb62a
--- /dev/null
+++ b/outlines/models/utils.py
@@ -0,0 +1,27 @@
+import jsonpath_ng
+
+
+def set_additional_properties_false_json_schema(schema: dict) -> dict:
+    """Set additionalProperties to False to all objects in the schema using jsonpath.
+
+    Parameters
+    ----------
+    schema
+        The JSON schema to modify
+
+    Returns
+    -------
+    dict
+        The modified schema with additionalProperties set to False
+    """
+    # Get all nodes
+    jsonpath_expr = jsonpath_ng.parse('$..*')
+    matches = jsonpath_expr.find(schema)
+
+    # Go over all nodes and set additionalProperties to False if it's an object
+    for match in matches:
+        if match.value == 'object':
+            if 'additionalProperties' not in match.context.value:
+                match.context.value['additionalProperties'] = False
+
+    return schema
diff --git a/pyproject.toml b/pyproject.toml
index 9216216d2..77bb4ac21 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -42,6 +42,7 @@ dependencies = [
    "outlines_core==0.1.26",
    "genson",
    "torch",
+   "jsonpath_ng",
 ]
 dynamic = ["version"]
 
@@ -168,6 +169,7 @@ module = [
     "tf-keras",
     "tf-keras.*",
     "mkdocs_gen_files.*",
+    "jsonpath_ng.*",
 ]
 ignore_missing_imports = true
 
diff --git a/tests/models/test_utils.py b/tests/models/test_utils.py
new file mode 100644
index 000000000..4607d1afc
--- /dev/null
+++ b/tests/models/test_utils.py
@@ -0,0 +1,50 @@
+from outlines.models.utils import set_additional_properties_false_json_schema
+
+
+def test_set_additional_properties_false_json_schema():
+    # additionalProperties is not set
+    schema = {
+        "type": "object",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"},
+        },
+        "required": ["name"],
+    }
+    modified_schema = set_additional_properties_false_json_schema(schema)
+    target_schema = {
+        "type": "object",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"},
+        },
+        "required": ["name"],
+        "additionalProperties": False,
+    }
+    assert modified_schema == target_schema
+
+    # additionalProperties is set to False
+    schema = {
+        "type": "object",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"},
+        },
+        "required": ["name"],
+        "additionalProperties": False,
+    }
+    modified_schema = set_additional_properties_false_json_schema(schema)
+    assert modified_schema == schema
+
+    # additionalProperties is set to True
+    schema = {
+        "type": "object",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"},
+        },
+        "required": ["name"],
+        "additionalProperties": True,
+    }
+    modified_schema = set_additional_properties_false_json_schema(schema)
+    assert modified_schema == schema

From 61ac54c877593dde7b2a325e701d6e48310759be Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 1 Jul 2025 10:55:50 +0200
Subject: [PATCH 379/505] Create the AsyncOllama model

---
 outlines/models/__init__.py |   3 +-
 outlines/models/ollama.py   | 131 ++++++++++++++++++++++++++++++++----
 2 files changed, 120 insertions(+), 14 deletions(-)

diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index d818c5db5..e705d0555 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -14,7 +14,7 @@
 from .gemini import Gemini, from_gemini
 from .llamacpp import LlamaCpp, from_llamacpp
 from .mlxlm import MLXLM, from_mlxlm
-from .ollama import Ollama, from_ollama
+from .ollama import AsyncOllama, Ollama, from_ollama
 from .openai import OpenAI, from_openai
 from .sglang import AsyncSGLang, SGLang, from_sglang
 from .tgi import AsyncTGI, TGI, from_tgi
@@ -40,6 +40,7 @@
     VLLMOffline,
 ]
 AsyncBlackBoxModel = Union[
+    AsyncOllama,
     AsyncTGI,
     AsyncSGLang,
     AsyncVLLM,
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index b41e28fe6..2a35d6175 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,11 +1,11 @@
 """Integration with the `ollama` library."""
 
 import json
-from typing import TYPE_CHECKING, Any, Iterator, Optional, Union
+from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
 
 from pydantic import TypeAdapter
 
-from outlines.models.base import Model, ModelTypeAdapter
+from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.templates import Vision
 from outlines.types import CFG, JsonSchema, Regex
 from outlines.types.utils import (
@@ -16,7 +16,8 @@
 )
 
 if TYPE_CHECKING:
-    from ollama import Client as OllamaClient
+    from ollama import Client
+    from ollama import AsyncClient
 
 __all__ = ["Ollama", "from_ollama"]
 
@@ -120,9 +121,7 @@ class Ollama(Model):
 
     """
 
-    def __init__(
-        self,client: "OllamaClient", model_name: Optional[str] = None,
-    ):
+    def __init__(self, client: "Client", model_name: Optional[str] = None):
         """
         Parameters
         ----------
@@ -208,23 +207,129 @@ def generate_stream(
             yield chunk.response
 
 
+class AsyncOllama(AsyncModel):
+    """Thin wrapper around the `ollama.AsyncClient` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `ollama.AsyncClient` client.
+
+    """
+
+    def __init__(
+        self,client: "AsyncClient", model_name: Optional[str] = None,
+    ):
+        """
+        Parameters
+        ----------
+        client
+            The `ollama.Client` client.
+        model_name
+            The name of the model to use.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = OllamaTypeAdapter()
+
+    async def generate(self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **kwargs: Any,
+    ) -> str:
+        """Generate text using Ollama.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        str
+            The text generated by the model.
+
+        """
+        if "model" not in kwargs and self.model_name is not None:
+            kwargs["model"] = self.model_name
+
+        response = await self.client.generate(
+            **self.type_adapter.format_input(model_input),
+            format=self.type_adapter.format_output_type(output_type),
+            **kwargs,
+        )
+        return response.response
+
+    async def generate_stream( # type: ignore
+        self,
+        model_input: str,
+        output_type: Optional[Any] = None,
+        **kwargs: Any,
+    ) -> AsyncIterator[str]:
+        """Stream text using Ollama.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema.
+        **kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        if "model" not in kwargs and self.model_name is not None:
+            kwargs["model"] = self.model_name
+
+        stream = await self.client.generate(
+            **self.type_adapter.format_input(model_input),
+            format=self.type_adapter.format_output_type(output_type),
+            stream=True,
+            **kwargs,
+        )
+        async for chunk in stream:
+            yield chunk.response
+
+
 def from_ollama(
-    client: "OllamaClient", model_name: Optional[str] = None
-) -> Ollama:
+    client: Union["Client", "AsyncClient"], model_name: Optional[str] = None
+) -> Union[Ollama, AsyncOllama]:
     """Create an Outlines `Ollama` model instance from an `ollama.Client`
-    client.
+    or `ollama.AsyncClient` instance.
 
     Parameters
     ----------
     client
-        A `ollama.Client` client instance.
+        A `ollama.Client` or `ollama.AsyncClient` instance.
     model_name
         The name of the model to use.
 
     Returns
     -------
-    Ollama
-        An Outlines `Ollama` model instance.
+    Union[Ollama, AsyncOllama]
+        An Outlines `Ollama` or `AsyncOllama` model instance.
 
     """
-    return Ollama(client, model_name)
+    from ollama import AsyncClient, Client
+
+    if isinstance(client, Client):
+        return Ollama(client, model_name)
+    elif isinstance(client, AsyncClient):
+        return AsyncOllama(client, model_name)
+    else:
+        raise ValueError(
+            "Invalid client type, the client must be an instance of "
+            "`ollama.Client` or `ollama.AsyncClient`."
+        )

From 1f6e250681af79c57ccf9d6759400a819635171d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 1 Jul 2025 10:56:03 +0200
Subject: [PATCH 380/505] Add tests for AsyncOllama

---
 tests/models/test_ollama.py | 121 +++++++++++++++++++++++++++++++++---
 1 file changed, 114 insertions(+), 7 deletions(-)

diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index 189066d48..e03c46fa9 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -5,11 +5,11 @@
 
 import pytest
 from PIL import Image
-from ollama import Client as OllamaClient
+from ollama import AsyncClient, Client
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.models import Ollama
+from outlines.models import AsyncOllama, Ollama
 from outlines.templates import Vision
 
 
@@ -18,12 +18,22 @@
 
 @pytest.fixture
 def model():
-    return Ollama(OllamaClient(), MODEL_NAME)
+    return Ollama(Client(), MODEL_NAME)
 
 
 @pytest.fixture
 def model_no_model_name():
-    return Ollama(OllamaClient())
+    return Ollama(Client())
+
+
+@pytest.fixture
+def async_model():
+    return AsyncOllama(AsyncClient(), MODEL_NAME)
+
+
+@pytest.fixture
+def async_model_no_model_name():
+    return AsyncOllama(AsyncClient())
 
 
 @pytest.fixture(scope="session")
@@ -41,8 +51,8 @@ def image():
     return image
 
 
-def test_init_from_client():
-    client = OllamaClient()
+def test_ollama_init_from_client():
+    client = Client()
 
     # With model name
     model = outlines.from_ollama(client, MODEL_NAME)
@@ -57,7 +67,7 @@ def test_init_from_client():
     assert model.model_name is None
 
 
-def test_wrong_inference_parameters(model):
+def test_ollama_wrong_inference_parameters(model):
     with pytest.raises(TypeError, match="got an unexpected"):
         model.generate(
             "Respond with one word. Not more.", None, foo=10
@@ -127,3 +137,100 @@ class Foo(BaseModel):
     for text in generator:
         generated_text.append(text)
     assert "foo" in json.loads("".join(generated_text))
+
+
+def test_ollama_async_init_from_client():
+    client = AsyncClient()
+
+    # With model name
+    model = outlines.from_ollama(client, MODEL_NAME)
+    assert isinstance(model, AsyncOllama)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+    # Without model name
+    model = outlines.from_ollama(client)
+    assert isinstance(model, AsyncOllama)
+    assert model.client == client
+    assert model.model_name is None
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_wrong_inference_parameters(async_model):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        await async_model.generate(
+            "Respond with one word. Not more.", None, foo=10
+        )
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_simple(async_model):
+    result = await async_model.generate(
+        "Respond with one word. Not more.", None
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_direct(async_model_no_model_name):
+    result = await async_model_no_model_name(
+        "Respond with one word. Not more.",
+        None,
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_simple_vision(image, async_model):
+    # This is not using a vision model, so it's not able to describe
+    # the image, but we're still checking the model input syntax
+    result = await async_model.generate(
+        Vision("What does this logo represent?", image),
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_json(async_model):
+    class Foo(BaseModel):
+        foo: Annotated[str, Field(max_length=1)]
+
+    result = await async_model("Respond with one word. Not more.", Foo)
+    assert isinstance(result, str)
+    assert "foo" in json.loads(result)
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_wrong_output_type(async_model):
+    class Foo(Enum):
+        bar = "Bar"
+        foor = "Foo"
+
+    with pytest.raises(TypeError, match="is not supported"):
+        await async_model.generate("foo?", Foo)
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_wrong_input_type(async_model):
+    with pytest.raises(TypeError, match="is not available"):
+        await async_model.generate(["foo?", "bar?"], None)
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_stream(async_model):
+    async_generator = async_model.stream("Write a sentence about a cat.")
+    assert isinstance(await async_generator.__anext__(), str)
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_stream_json(async_model_no_model_name):
+    class Foo(BaseModel):
+        foo: Annotated[str, Field(max_length=2)]
+
+    async_generator = async_model_no_model_name.stream("Create a character.", Foo, model=MODEL_NAME)
+    generated_text = []
+    async for chunk in async_generator:
+        generated_text.append(chunk)
+    assert "foo" in json.loads("".join(generated_text))

From d2e4a5b03d125b7f7e46cf0407220118cecca18e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 1 Jul 2025 10:56:33 +0200
Subject: [PATCH 381/505] Add documentation for AsyncOllama

---
 docs/features/models/index.md  |  3 +-
 docs/features/models/ollama.md | 91 +++++++++++++++++++++++++++++++---
 2 files changed, 87 insertions(+), 7 deletions(-)

diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index b9f747320..b79d36581 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -56,7 +56,7 @@ In alphabetical order:
 | Regex | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
 | Grammar | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | 🟠 | ❌ | ✅ | ✅ | ✅ | ✅ |
 | **Generation Features** | | | | | | | | | | | | | |
-| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
 | Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ |
@@ -104,6 +104,7 @@ print(type(model)) # outlines.models.tgi.AsyncTGI
 
 The models that have an async version are the following:
 
+- Ollama
 - SgLang
 - TGI
 - VLLM
diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
index 60043e38d..25fb8e70d 100644
--- a/docs/features/models/ollama.md
+++ b/docs/features/models/ollama.md
@@ -17,22 +17,31 @@ title: Ollama
 
 To create an Ollama model instance, you can use the `from_ollama` function. It takes 2 arguments:
 
-- `client`: an `ollama.Client` instance
+- `client`: an `ollama.Client` or `ollama.AsyncClient` instance
 - `model_name`: the name of the model you want to use
 
+Based on whether the inference client instance is synchronous or asynchronous, you will receive an `Ollama` or an `AsyncOllama` model instance.
+
 For instance:
 
 ```python
 import ollama
 import outlines
 
-# Create the client
+# Create the client or async client
 client = ollama.Client()
+async_client = ollama.AsyncClient()
 
-# Create the model
+# Create a sync model
 model = outlines.from_ollama(
     client,
-    "tinyllama",
+    "qwen2.5vl:3b",
+)
+
+# Create an async model
+model = outlines.from_ollama(
+    async_client,
+    "qwen2.5vl:3b",
 )
 ```
 
@@ -49,7 +58,7 @@ import ollama
 import outlines
 
 # Create the model
-model = outlines.from_ollama(ollama.Client(), "tinyllama")
+model = outlines.from_ollama(ollama.Client(), "qwen2.5vl:3b")
 
 # Call it to generate text
 response = model("What's the capital of Latvia?")
@@ -63,7 +72,7 @@ import ollama
 import outlines
 
 # Create the model
-model = outlines.from_ollama(ollama.Client(), "tinyllama")
+model = outlines.from_ollama(ollama.Client(), "qwen2.5vl:3b")
 
 # Stream text
 for chunk in model.stream("Write a short story about a cat"):
@@ -99,6 +108,76 @@ response = model(prompt)
 print(response) # The image shows a black puppy with a curious and attentive expression.
 ```
 
+## Asynchronous Calls
+
+Ollama supports asynchronous operations by passing an `AsyncClient` instead of a regular `Client`. This returns an `AsyncOllama` model instance that supports async/await patterns.
+
+### Basic Async Generation
+
+```python
+import asyncio
+import outlines
+import ollama
+
+async def generate_text():
+    # Create an async model
+    async_client = ollama.AsyncClient()
+    async_model = outlines.from_ollama(async_client, "qwen2.5vl:3b")
+
+    result = await async_model("Write a haiku about Python.")
+    print(result)
+
+asyncio.run(generate_text())
+```
+
+### Async Streaming
+
+The async model also supports streaming with async iteration:
+
+```python
+import asyncio
+import outlines
+import ollama
+
+async def stream_text():
+    async_client = ollama.AsyncClient()
+    async_model = outlines.from_ollama(async_client, "qwen2.5vl:3b")
+
+    async for chunk in async_model.stream("Tell me a story about a robot."):
+        print(chunk, end="")
+
+asyncio.run(stream_text())
+```
+
+### Concurrent Async Requests
+
+One of the main benefits of async calls is the ability to make multiple concurrent requests:
+
+```python
+import asyncio
+import outlines
+import ollama
+
+async def generate_multiple():
+    async_client = ollama.AsyncClient()
+    async_model = outlines.from_ollama(async_client, "qwen2.5vl:3b")
+
+    # Define multiple prompts
+    prompts = [
+        "Write a tagline for a coffee shop.",
+        "Write a tagline for a bookstore.",
+        "Write a tagline for a gym."
+    ]
+
+    tasks = [async_model(prompt) for prompt in prompts]
+    results = await asyncio.gather(*tasks)
+
+    for prompt, result in zip(prompts, results):
+        print(f"{prompt}\n{result}\n")
+
+asyncio.run(generate_multiple())
+```
+
 ## Structured Generation
 
 Ollama only provides support for structured generation based on a JSON schema. To use it, call the model with a JSON schema object as an `output_type` on top of your prompt.

From ca461c6b483ba491d32f82583bf24bf3f1b61d31 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 1 Jul 2025 13:27:07 +0200
Subject: [PATCH 382/505] Add the output_type Choice

---
 outlines/models/gemini.py  |  6 +++++-
 outlines/types/__init__.py |  1 +
 outlines/types/dsl.py      | 22 ++++++++++++++++++++++
 outlines/types/utils.py    |  7 +++++++
 4 files changed, 35 insertions(+), 1 deletion(-)

diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 57cd60302..ed0847133 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -11,10 +11,11 @@
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.templates import Vision
-from outlines.types import CFG, JsonSchema, Regex
+from outlines.types import CFG, Choice, JsonSchema, Regex
 from outlines.types.utils import (
     is_dataclass,
     is_enum,
+    get_enum_from_choice,
     get_enum_from_literal,
     is_genson_schema_builder,
     is_literal,
@@ -128,6 +129,9 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         elif is_literal(output_type):
             enum = get_enum_from_literal(output_type)
             return self.format_enum_output_type(enum)
+        elif isinstance(output_type, Choice):
+            enum = get_enum_from_choice(output_type)
+            return self.format_enum_output_type(enum)
 
         else:
             type_name = getattr(output_type, "__name__", output_type)
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 208f4f13b..c9c66f0f9 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -14,6 +14,7 @@
     Regex,
     CFG,
     FSM,
+    Choice,
     JsonSchema,
     regex,
     cfg,
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 1a0a0ca1c..f492d1fd8 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -392,6 +392,25 @@ def from_file(cls, path: str) -> "JsonSchema":
         return cls(schema)
 
 
+@dataclass
+class Choice(Term):
+    """Class representing a choice between different items.
+
+    Parameters
+    ----------
+    items
+        The items to choose from.
+
+    """
+    items: List[Any]
+
+    def _display_node(self) -> str:
+        return f"Choice({repr(self.items)})"
+
+    def __repr__(self):
+        return f"Choice(items={repr(self.items)})"
+
+
 @dataclass
 class KleeneStar(Term):
     term: Term
@@ -913,6 +932,9 @@ def to_regex(term: Term) -> str:
         case JsonSchema():
             regex_str = build_regex_from_schema(term.schema, term.whitespace_pattern)
             return f"({regex_str})"
+        case Choice():
+            regexes = [to_regex(python_types_to_terms(item)) for item in term.items]
+            return f"({'|'.join(regexes)})"
         case KleeneStar():
             return f"({to_regex(term.term)})*"
         case KleenePlus():
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
index 64537643e..ffd6d80ab 100644
--- a/outlines/types/utils.py
+++ b/outlines/types/utils.py
@@ -163,6 +163,13 @@ def get_enum_from_literal(value) -> Enum:
     )
 
 
+def get_enum_from_choice(value) -> Enum:
+    return Enum(
+        'Choice',
+        {str(item): item for item in value.items}
+    )
+
+
 def get_schema_from_signature(fn: Callable) -> dict:
     """Turn a function signature into a JSON schema.
 

From 27c8291397df6d4dd32eff7912892341b63aab04 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 1 Jul 2025 14:20:43 +0200
Subject: [PATCH 383/505] Add tests for the Choice output_type

---
 tests/models/test_gemini.py     |  8 ++++++++
 tests/types/test_dsl.py         |  6 ++++++
 tests/types/test_to_regex.py    |  7 +++++++
 tests/types/test_types_utils.py | 14 ++++++++++++--
 4 files changed, 33 insertions(+), 2 deletions(-)

diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 5df7b2669..bf383c3b7 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -13,6 +13,7 @@
 import outlines
 from outlines.models.gemini import Gemini
 from outlines.templates import Vision
+from outlines.types import Choice
 
 if sys.version_info >= (3, 12):
     from typing import TypedDict
@@ -185,6 +186,13 @@ class Foo(Enum):
     assert result == "Foo" or result == "Bar"
 
 
+@pytest.mark.api_call
+def test_gemini_simple_choice_choice(model):
+    result = model.generate("foo?", Choice(["Foo", "Bar"]))
+    assert isinstance(result, str)
+    assert result == "Foo" or result == "Bar"
+
+
 @pytest.mark.api_call
 def test_gemini_sample_choice_literal(model):
     result = model.generate("foo?", Literal["Foo", "Bar"])
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index 4e13b1d38..e07af09c7 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -30,6 +30,7 @@
     QuantifyExact,
     QuantifyMaximum,
     QuantifyMinimum,
+    Choice,
     Regex,
     Sequence,
     String,
@@ -69,6 +70,11 @@ def test_dsl_init():
     assert repr(string) == "String(value='test')"
     assert string.display_ascii_tree() == "└── String('test')\n"
 
+    choice = Choice(["a", "b"])
+    assert choice.items == ["a", "b"]
+    assert repr(choice) == "Choice(items=['a', 'b'])"
+    assert choice.display_ascii_tree() == "└── Choice(['a', 'b'])\n"
+
     regex = Regex("[0-9]")
     assert regex.pattern == "[0-9]"
     assert repr(regex) == "Regex(pattern='[0-9]')"
diff --git a/tests/types/test_to_regex.py b/tests/types/test_to_regex.py
index 4b0403ac6..6f33fb6e3 100644
--- a/tests/types/test_to_regex.py
+++ b/tests/types/test_to_regex.py
@@ -2,6 +2,7 @@
 
 
 from outlines.types.dsl import (
+    Choice,
     String,
     Regex,
     JsonSchema,
@@ -36,6 +37,12 @@ def test_to_regex_simple():
     assert a.matches("1") is True
     assert a.matches("a") is False
 
+    a = Choice(["a", "b"])
+    assert to_regex(a) == "(a|b)"
+    assert a.matches("a") is True
+    assert a.matches("b") is True
+    assert a.matches("c") is False
+
     a = Optional(String("a"))
     assert to_regex(a) == "(a)?"
     assert a.matches("") is True
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
index 4d73312d6..57f41a099 100644
--- a/tests/types/test_types_utils.py
+++ b/tests/types/test_types_utils.py
@@ -18,10 +18,11 @@
 
 import interegular
 from genson import SchemaBuilder
-from pydantic import BaseModel, constr
+from pydantic import BaseModel
 
-from outlines.types.dsl import JsonSchema
+from outlines.types.dsl import Choice, JsonSchema
 from outlines.types.utils import (
+    get_enum_from_choice,
     get_enum_from_literal,
     get_schema_from_enum,
     get_schema_from_signature,
@@ -368,6 +369,15 @@ def test_is_interegular_fsm():
 # Type conversion
 
 
+def test_get_enum_from_choice(sample_enum):
+    choice = Choice(["a", "b", sample_enum.A])
+    enum = get_enum_from_choice(choice)
+    assert is_enum(enum)
+    assert enum.a.value == "a"
+    assert enum.b.value == "b"
+    assert getattr(enum, "SampleEnum.A").value == sample_enum.A
+
+
 def test_get_enum_from_literal(sample_enum):
     basic_enum = get_enum_from_literal(Literal["a", "b"])
     assert(is_enum(basic_enum))

From 991adef9773c94e06e18439f536cae839f5d4866 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 1 Jul 2025 14:20:52 +0200
Subject: [PATCH 384/505] Add documentation for the Choice output_type

---
 docs/features/core/output_types.md | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/docs/features/core/output_types.md b/docs/features/core/output_types.md
index edaa155b3..a224312d0 100644
--- a/docs/features/core/output_types.md
+++ b/docs/features/core/output_types.md
@@ -16,6 +16,7 @@ Output types can be from the general Python ecosystem, including:
 - Types from popular third party libraries such as Pydantic or GenSON.
 
 Outlines also provides special classes for certain output structures (more details below):
+- Multiple choices with `Choice`
 - JSON schemas with `JsonSchema`
 - Regular expressions with `Regex`
 - Context-free grammars with `CFG`
@@ -115,6 +116,20 @@ output_type = Literal["pizza", "burger"]
 output_type = PizzaOrBurger
 ```
 
+Additionally, you can use the Outlines-specific type `Choice` that takes a `list` as an argument. This type is useful in situations in which the list of choices is dynamic.
+
+For instance:
+
+```python
+from outlines.types import Choice
+
+def get_multiple_choices() -> list:
+    # we could have something complex here
+    return ["pizza", "burger"]
+
+output_type = Choice(get_multiple_choices())
+```
+
 ### JSON Schemas
 
 Multiple different common Python types are often used to store information equivalent to a JSON schema. The following can be used in Outlines to generate text that respects a JSON schema:

From 7a25481468cef81f086431159b2f3fb9641751cf Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 1 Jul 2025 14:48:16 +0200
Subject: [PATCH 385/505] Add the ensure_ascii parameter to JsonSchema

---
 docs/features/core/output_types.md |  4 ++++
 outlines/types/dsl.py              | 13 ++++++++-----
 2 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/docs/features/core/output_types.md b/docs/features/core/output_types.md
index a224312d0..e0eeef20b 100644
--- a/docs/features/core/output_types.md
+++ b/docs/features/core/output_types.md
@@ -177,6 +177,10 @@ schema_dict = {
 output_type = JsonSchema(schema_dict)
 ```
 
+`JsonSchema` accepts two optional parameters:
+- `whitespace_pattern` (defaults to `None`): specifies the pattern to use for JSON syntactic whitespace. If none is provided, the default permissive JSON whitespace rules are used.
+- `ensure_ascii` (defaults to `True`): defines the value to use for the argument `ensure_ascii` of the `json.dumps` method. If false, non-ASCII characters will be turned into unicodes.
+
 ### Regex Patterns
 
 Outlines provides support for text generation constrained by regular expressions. Since regular expressions are expressed as simple raw string literals, regex strings must wrapped in an `outlines.types.Regex` object.
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index f492d1fd8..0af69ab12 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -318,6 +318,7 @@ def __init__(
             dict, str, type[BaseModel], _TypedDictMeta, type, SchemaBuilder
         ],
         whitespace_pattern: OptionalType[str] = None,
+        ensure_ascii: bool = True,
     ):
         """
         Parameters
@@ -326,22 +327,24 @@ def __init__(
             The object containing the JSON schema.
         whitespace_pattern
             The pattern to use to match whitespace characters.
+        ensure_ascii
+            Whether to ensure the schema is ASCII-only.
 
         """
         schema_str: str
 
         if is_dict_instance(schema):
-            schema_str = json.dumps(schema)
+            schema_str = json.dumps(schema, ensure_ascii=ensure_ascii)
         elif is_str_instance(schema):
             schema_str = str(schema)
         elif is_pydantic_model(schema):
-            schema_str = json.dumps(schema.model_json_schema())  # type: ignore
+            schema_str = json.dumps(schema.model_json_schema(), ensure_ascii=ensure_ascii) # type: ignore
         elif is_typed_dict(schema):
-            schema_str = json.dumps(TypeAdapter(schema).json_schema())
+            schema_str = json.dumps(TypeAdapter(schema).json_schema(), ensure_ascii=ensure_ascii)
         elif is_dataclass(schema):
-            schema_str = json.dumps(TypeAdapter(schema).json_schema())
+            schema_str = json.dumps(TypeAdapter(schema).json_schema(), ensure_ascii=ensure_ascii)
         elif is_genson_schema_builder(schema):
-            schema_str = schema.to_json()  # type: ignore
+            schema_str = schema.to_json(ensure_ascii=ensure_ascii)  # type: ignore
         else:
             raise ValueError(
                 f"Cannot parse schema {schema}. The schema must be either "

From c350ddc9ae79a2088286c018c4f54617d0f86922 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 2 Jul 2025 16:40:32 +0200
Subject: [PATCH 386/505] Add a batch method to models

---
 outlines/generator.py           |  65 ++++++++++-
 outlines/models/anthropic.py    |  10 ++
 outlines/models/base.py         | 191 ++++++++++++++++++++++++++++----
 outlines/models/dottxt.py       |  10 ++
 outlines/models/gemini.py       |  10 ++
 outlines/models/llamacpp.py     |  12 +-
 outlines/models/mlxlm.py        |  14 ++-
 outlines/models/ollama.py       |  27 ++++-
 outlines/models/openai.py       |  10 ++
 outlines/models/sglang.py       |  24 +++-
 outlines/models/tgi.py          |  22 +++-
 outlines/models/transformers.py |  92 ++++++++++-----
 outlines/models/vllm.py         |  16 +++
 outlines/models/vllm_offline.py |  97 ++++++++++++----
 14 files changed, 512 insertions(+), 88 deletions(-)

diff --git a/outlines/generator.py b/outlines/generator.py
index 1623fe85e..f9c5af2ce 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -4,6 +4,7 @@
     Any,
     AsyncIterator,
     Iterator,
+    List,
     Optional,
     Union,
 )
@@ -73,6 +74,26 @@ def __call__(self, prompt: Any, **inference_kwargs) -> Any:
             prompt, self.output_type, **inference_kwargs
         )
 
+    def batch(self, prompts: List[Any], **inference_kwargs) -> List[Any]:
+        """Generate a batch of responses from the model.
+
+        Parameters
+        ----------
+        prompts
+            The list of prompts to use to generate a batch of responses.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        List[Any]
+            The list of responses generated by the model.
+
+        """
+        return self.model.generate_batch(
+            prompts, self.output_type, **inference_kwargs
+        )
+
     def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
         """Generate a stream of responses from the model.
 
@@ -142,9 +163,27 @@ async def __call__(self, prompt: Any, **inference_kwargs) -> Any:
             prompt, self.output_type, **inference_kwargs
         )
 
-    async def stream(
-        self, prompt: Any, **inference_kwargs
-    ) -> AsyncIterator[Any]:
+    async def batch(self, prompts: List[Any], **inference_kwargs) -> List[Any]:
+        """Generate a batch of responses from the model.
+
+        Parameters
+        ----------
+        prompts
+            The list of prompts to use to generate a batch of responses.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        List[Any]
+            The list of responses generated by the model.
+
+        """
+        return await self.model.generate_batch(
+            prompts, self.output_type, **inference_kwargs
+        )
+
+    async def stream(self, prompt: Any, **inference_kwargs) -> AsyncIterator[Any]:
         """Generate a stream of responses from the model.
 
         Parameters
@@ -268,6 +307,26 @@ def __call__(self, prompt: Any, **inference_kwargs) -> Any:
             prompt, self.logits_processor, **inference_kwargs
         )
 
+    def batch(self, prompts: List[Any], **inference_kwargs) -> List[Any]:
+        """Generate a batch of responses from the model.
+
+        Parameters
+        ----------
+        prompts
+            The list of prompts to use to generate a batch of responses.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        List[Any]
+            The list of responses generated by the model.
+
+        """
+        return self.model.generate_batch(
+            prompts, self.logits_processor, **inference_kwargs
+        )
+
     def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
         """Generate a stream of responses from the model.
 
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 4ed3828d9..782ba7bed 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -153,6 +153,16 @@ def generate(
         )
         return completion.content[0].text
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError(
+            "Anthropic does not support batch generation."
+        )
+
     def generate_stream(
         self,
         model_input: Union[str, Vision],
diff --git a/outlines/models/base.py b/outlines/models/base.py
index 08beece18..7092ed57f 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -1,7 +1,7 @@
 """Base classes for all models and model type adapters."""
 
 from abc import ABC, abstractmethod
-from typing import Any, AsyncIterator, Iterator, Optional
+from typing import Any, AsyncIterator, Iterator, List, Optional
 
 
 class ModelTypeAdapter(ABC):
@@ -42,9 +42,9 @@ def format_input(self, model_input: Any) -> Any:
     def format_output_type(self, output_type: Optional[Any] = None) -> Any:
         """Format the output type to the expected format of the model.
 
-        For API-based models, this typically means creating a `response_format`
-        argument. For local models, it means formatting the logits processor to
-        create the object type expected by the model.
+        For black-box models, this typically means creating a `response_format`
+        argument. For steerable models, it means formatting the logits processor
+        to create the object type expected by the model.
 
         Parameters
         ----------
@@ -62,13 +62,16 @@ def format_output_type(self, output_type: Optional[Any] = None) -> Any:
 class Model(ABC):
     """Base class for all synchronous models.
 
-    This class defines a shared `__call__` method that can be used to call the
-    model directly.
+    This class defines shared `__call__`, `batch` and `stream` methods that can
+    be used to call the model directly. The `generate`, `generate_batch`, and
+    `generate_stream` methods must be implemented by the subclasses.
     All models inheriting from this class must define a `type_adapter`
     attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
-    attribute are used in the `generate` method to format the input and output
-    types received by the model.
-    Additionally, local models must define a `tensor_library_name` attribute.
+    attribute are used in the `generate`, `generate_batch`, and
+    `generate_stream` methods to format the input and output types received by
+    the model.
+    Additionally, steerable models must define a `tensor_library_name`
+    attribute.
 
     """
     type_adapter: ModelTypeAdapter
@@ -113,6 +116,47 @@ def __call__(
 
         return Generator(self, output_type)(model_input, **inference_kwargs)
 
+    def batch(
+        self,
+        model_input: List[Any],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> List[Any]:
+        """Make a batch call to the model (several inputs at once).
+
+        Users can use the `batch` method from the model directly, in which
+        case we will create a generator instance with the output type provided
+        and then invoke its `batch` method.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        generator.batch(["prompt1", "prompt2"])
+        ```
+        and
+        ```python
+        model.batch(["prompt1", "prompt2"], Foo)
+        ```
+
+        Parameters
+        ----------
+        model_input
+            The list of inputs provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        List[Any]
+            The list of responses generated by the model.
+
+        """
+        from outlines import Generator
+
+        generator = Generator(self, output_type)
+        return generator.batch(model_input, **inference_kwargs) # type: ignore
+
     def stream(
         self,
         model_input: Any,
@@ -165,8 +209,8 @@ def generate(
     ) -> Any:
         """Generate a response from the model.
 
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
+        The output_type argument contains a logits processor for steerable
+        models while it contains a type (Json, Enum...) for black-box models.
         This method is not intended to be used directly by end users.
 
         Parameters
@@ -187,6 +231,35 @@ def generate(
         ...
 
     @abstractmethod
+    def generate_batch(
+        self,
+        model_input: List[Any],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> List[Any]:
+        """Generate a batch of responses from the model.
+
+        The output_type argument contains a logits processor for steerable
+        models while it contains a type (Json, Enum...) for black-box models.
+        This method is not intended to be used directly by end users.
+
+        Parameters
+        ----------
+        model_input
+            The list of inputs provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        List[Any]
+            The list of responses generated by the model.
+
+        """
+        ...
+    @abstractmethod
     def generate_stream(
         self,
         model_input: Any,
@@ -195,8 +268,8 @@ def generate_stream(
     ) -> Iterator[Any]:
         """Generate a stream of responses from the model.
 
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
+        The output_type argument contains a logits processor for steerable
+        models while it contains a type (Json, Enum...) for black-box models.
         This method is not intended to be used directly by end users.
 
         Parameters
@@ -219,13 +292,16 @@ def generate_stream(
 class AsyncModel(ABC):
     """Base class for all asynchronous models.
 
-    This class defines a shared `__call__` method that can be used to call the
-    model directly.
+    This class defines shared `__call__`, `batch` and `stream` methods that can
+    be used to call the model directly. The `generate`, `generate_batch`, and
+    `generate_stream` methods must be implemented by the subclasses.
     All models inheriting from this class must define a `type_adapter`
     attribute of type `ModelTypeAdapter`. The methods of the `type_adapter`
-    attribute are used in the `generate` method to format the input and output
-    types received by the model.
-    Additionally, local models must define a `tensor_library_name` attribute.
+    attribute are used in the `generate`, `generate_batch`, and
+    `generate_stream` methods to format the input and output types received by
+    the model.
+    Additionally, steerable models must define a `tensor_library_name`
+    attribute.
 
     """
     type_adapter: ModelTypeAdapter
@@ -271,6 +347,47 @@ async def __call__(
         generator = Generator(self, output_type)
         return await generator(model_input, **inference_kwargs)
 
+    async def batch(
+        self,
+        model_input: List[Any],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> List[Any]:
+        """Make a batch call to the model (several inputs at once).
+
+        Users can use the `batch` method from the model directly, in which
+        case we will create a generator instance with the output type provided
+        and then invoke its `batch` method.
+        Thus, those commands are equivalent:
+        ```python
+        generator = Generator(model, Foo)
+        await generator.batch(["prompt1", "prompt2"])
+        ```
+        and
+        ```python
+        await model.batch(["prompt1", "prompt2"], Foo)
+        ```
+
+        Parameters
+        ----------
+        model_input
+            The list of inputs provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        List[Any]
+            The list of responses generated by the model.
+
+        """
+        from outlines import Generator
+
+        generator = Generator(self, output_type)
+        return await generator.batch(model_input, **inference_kwargs) # type: ignore
+
     async def stream(
         self,
         model_input: Any,
@@ -325,8 +442,8 @@ async def generate(
     ) -> Any:
         """Generate a response from the model.
 
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
+        The output_type argument contains a logits processor for steerable
+        models while it contains a type (Json, Enum...) for black-box models.
         This method is not intended to be used directly by end users.
 
         Parameters
@@ -346,6 +463,36 @@ async def generate(
         """
         ...
 
+    @abstractmethod
+    async def generate_batch(
+        self,
+        model_input: List[Any],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any
+    ) -> List[Any]:
+        """Generate a batch of responses from the model.
+
+        The output_type argument contains a logits processor for steerable
+        models while it contains a type (Json, Enum...) for black-box models.
+        This method is not intended to be used directly by end users.
+
+        Parameters
+        ----------
+        model_input
+            The list of inputs provided by the user.
+        output_type
+            The output type provided by the user.
+        **inference_kwargs
+            Additional keyword arguments to pass to the model.
+
+        Returns
+        -------
+        List[Any]
+            The list of responses generated by the model.
+
+        """
+        ...
+
     @abstractmethod
     async def generate_stream(
         self,
@@ -355,8 +502,8 @@ async def generate_stream(
     ) -> AsyncIterator[Any]:
         """Generate a stream of responses from the model.
 
-        The output_type argument contains a logits processor for local models
-        while it contains a type (Json, Enum...) for the API-based models.
+        The output_type argument contains a logits processor for steerable
+        models while it contains a type (Json, Enum...) for black-box models.
         This method is not intended to be used directly by end users.
 
         Parameters
diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index 62c40a7c5..f88d3d5d6 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -176,6 +176,16 @@ def generate(
         )
         return completion.data
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError(
+            "Dottxt does not support batch generation."
+        )
+
     def generate_stream(
         self,
         model_input,
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index ed0847133..9ee7c61b2 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -241,6 +241,16 @@ def generate(
 
         return completion.text
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError(
+            "Gemini does not support batch generation."
+        )
+
     def generate_stream(
         self,
         model_input: Union[str, Vision],
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 92d5fe15f..f3dc50c8f 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -151,8 +151,8 @@ def format_input(self, model_input):
 
         """
         raise NotImplementedError(
-            f"The input type {input} is not available. "
-            "The `llama-cpp-python` library does not support batch inference. "
+            f"The input type {input} is not available with LlamaCpp. "
+            "The only available type is `str`."
         )
 
     @format_input.register(str)
@@ -243,6 +243,14 @@ def generate(
 
         return result
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError("LlamaCpp does not support batch generation.")
+
     def generate_stream(
         self,
         model_input: str,
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index bea4ad293..775a56222 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -33,8 +33,8 @@ def format_input(self, model_input):
 
         """
         raise NotImplementedError(
-            f"The input type {input} is not available. "
-            "The `mlx_lm` library does not support batch inference."
+            f"The input type {input} is not available with mlx-lm. "
+            "The only available type is `str`."
         )
 
     @format_input.register(str)
@@ -128,6 +128,16 @@ def generate(
             **kwargs,
         )
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **kwargs,
+    ):
+        raise NotImplementedError(
+            "The `mlx_lm` library does not support batch inference."
+        )
+
     def generate_stream(
         self,
         model_input: str,
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 2a35d6175..f4d6453bf 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -44,7 +44,7 @@ def format_input(self, model_input: Union[str, Vision]) -> dict:
         elif isinstance(model_input, Vision):
             return self.format_vision_model_input(model_input)
         raise TypeError(
-            f"The input type {model_input} is not available. "
+            f"The input type {model_input} is not available with Ollama. "
             "The only available types are `str` and `Vision`."
         )
 
@@ -58,7 +58,10 @@ def format_vision_model_input(self, model_input: Vision) -> dict:
         """Format the vision model input to pass to the client.
 
         """
-        return {"prompt": model_input.prompt, "images": [model_input.image_str]}
+        return {
+            "prompt": model_input.prompt,
+            "images": [model_input.image_str]
+        }
 
     def format_output_type(
         self, output_type: Optional[Any] = None
@@ -169,6 +172,16 @@ def generate(self,
         )
         return response.response
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **kwargs,
+    ):
+        raise NotImplementedError(
+            "The `ollama` library does not support batch inference."
+        )
+
     def generate_stream(
         self,
         model_input: str,
@@ -265,6 +278,16 @@ async def generate(self,
         )
         return response.response
 
+    async def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **kwargs,
+    ):
+        raise NotImplementedError(
+            "The `ollama` library does not support batch inference."
+        )
+
     async def generate_stream( # type: ignore
         self,
         model_input: str,
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 2e4209cc9..be86d69e6 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -316,6 +316,16 @@ def generate(
         else:
             return [message.content for message in messages]
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError(
+            "The `openai` library does not support batch inference."
+        )
+
     def generate_stream(
         self,
         model_input: Union[str, Vision],
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
index 0c2a51da1..1ac159bcd 100644
--- a/outlines/models/sglang.py
+++ b/outlines/models/sglang.py
@@ -2,7 +2,9 @@
 
 import json
 import warnings
-from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
+from typing import (
+    TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
+)
 
 from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.models.openai import OpenAITypeAdapter
@@ -147,6 +149,16 @@ def generate(
         else:
             return [message.content for message in messages]
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError(
+            "SGLang does not support batch inference."
+        )
+
     def generate_stream(
         self,
         model_input: Union[str, Vision],
@@ -279,6 +291,16 @@ async def generate(
         else:
             return [message.content for message in messages]
 
+    async def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError(
+            "SGLang does not support batch inference."
+        )
+
     async def generate_stream( # type: ignore
         self,
         model_input: Union[str, Vision],
diff --git a/outlines/models/tgi.py b/outlines/models/tgi.py
index d9f3ffd75..31e9f5c20 100644
--- a/outlines/models/tgi.py
+++ b/outlines/models/tgi.py
@@ -40,17 +40,13 @@ def format_input(self, model_input):
         """
         raise NotImplementedError(
             f"The input type {input} is not available with TGI. "
-            + "Please provide a string."
+            + "The only available type is `str`."
         )
 
     @format_input.register(str)
     def format_str_input(self, model_input: str) -> str:
         return model_input
 
-    @format_input.register(list)
-    def format_list_input(self, model_input: list):
-        raise NotImplementedError("TGI does not support batch inference.")
-
     def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the structured output argument to pass to the client.
 
@@ -143,6 +139,14 @@ def generate(
 
         return self.client.text_generation(**client_args)
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError("TGI does not support batch inference.")
+
     def generate_stream(
         self,
         model_input: str,
@@ -252,6 +256,14 @@ async def generate(
 
         return response
 
+    async def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError("TGI does not support batch inference.")
+
     async def generate_stream( # type: ignore
         self,
         model_input: str,
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index e2071c406..19350de52 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,5 +1,6 @@
 """Integration with the `transformers` library."""
 
+from collections import defaultdict
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union
 
@@ -148,17 +149,13 @@ def format_input(self, model_input):
         """
         raise NotImplementedError(
             f"The input type {input} is not available."
-            "Please use a string or a list of strings."
+            "The only available type is `str`."
         )
 
     @format_input.register(str)
     def format_str_input(self, model_input: str) -> str:
         return model_input
 
-    @format_input.register(list)
-    def format_list_input(self, model_input: List[str]) -> List[str]:
-        return model_input
-
     def format_output_type(
         self,
         output_type: Optional[OutlinesLogitsProcessor] = None,
@@ -241,12 +238,14 @@ def __init__(
             self.tensor_library_name = "torch"
 
     def _prepare_model_inputs(
-        self,
-        model_input: Union[str, List[str], dict],
-        output_type: Optional[OutlinesLogitsProcessor] = None,
+        self, model_input: Union[str, List[str], dict, List[dict]]
     ) -> Tuple[Union[str, List[str]], dict]:
         """Turn the user input into arguments to pass to the model"""
-        prompts = self.type_adapter.format_input(model_input)
+        # Format validation
+        if isinstance(model_input, list):
+            prompts = [self.type_adapter.format_input(item) for item in model_input]
+        else:
+            prompts = self.type_adapter.format_input(model_input)
         input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
             "input_ids": input_ids.to(self.model.device),
@@ -257,7 +256,7 @@ def _prepare_model_inputs(
 
     def generate(
         self,
-        model_input: Union[str, List[str], dict],
+        model_input: Union[str, dict],
         output_type: Optional[OutlinesLogitsProcessor] = None,
         **inference_kwargs: Any,
     ) -> Union[str, List[str]]:
@@ -283,19 +282,44 @@ def generate(
             The text generated by the model.
 
         """
-        prompts, inputs = self._prepare_model_inputs(model_input, output_type)
+        prompts, inputs = self._prepare_model_inputs(model_input)
         logits_processor = self.type_adapter.format_output_type(output_type)
 
         generated_ids = self._generate_output_seq(
             prompts, inputs, logits_processor=logits_processor, **inference_kwargs
         )
 
-        # if single str input, convert to a 1D outputt
-        if isinstance(prompts, str):
+        # required for multi-modal models that return a 2D tensor even when
+        # num_return_sequences is 1
+        num_samples = inference_kwargs.get("num_return_sequences", 1)
+        if num_samples == 1 and len(generated_ids.shape) == 2:
             generated_ids = generated_ids.squeeze(0)
 
         return self._decode_generation(generated_ids)
 
+    def generate_batch(
+        self,
+        model_input: List[Union[str, dict]],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
+        **inference_kwargs: Any,
+    ) -> List[Union[str, List[str]]]:
+        """"""
+        prompts, inputs = self._prepare_model_inputs(model_input) # type: ignore
+        logits_processor = self.type_adapter.format_output_type(output_type)
+
+        generated_ids = self._generate_output_seq(
+            prompts, inputs, logits_processor=logits_processor, **inference_kwargs
+        )
+
+        # if there are multiple samples per input, convert generated_id to 3D
+        num_samples = inference_kwargs.get("num_return_sequences", 1)
+        if num_samples > 1:
+            print("here", generated_ids)
+            generated_ids = generated_ids.view(len(model_input), num_samples, -1)
+            print("here2", generated_ids)
+
+        return self._decode_generation(generated_ids)
+
     def generate_stream(self, model_input, output_type, **inference_kwargs):
         """Not available for `transformers` models.
 
@@ -320,12 +344,6 @@ def _generate_output_seq(self, prompts, inputs, **inference_kwargs):
         else:
             generated_ids = output_ids[:, input_ids.shape[1] :]
 
-        # if batch list inputs AND multiple samples per input, convert generated_id to 3D view
-        num_samples = inference_kwargs.get("num_return_sequences", 1)
-        if num_samples > 1 and isinstance(prompts, list):
-            batch_size = input_ids.size(0)
-            generated_ids = generated_ids.view(batch_size, num_samples, -1)
-
         return generated_ids
 
     def _decode_generation(self, generated_ids: "torch.Tensor"):
@@ -364,10 +382,9 @@ def format_input(self, model_input):
         """
         raise NotImplementedError(
             f"The input type {input} is not available. Please provide a "
-            + "dictionary containing at least the 'text' key with a value "
-            + "of type Union[str, List[str]]. You should also include the "
-            + "other keys required by your processor (for instance, 'images' "
-            + "or 'audios')."
+            + "dictionary containing at least the 'text' key with a string "
+            + "value. You should also include the other keys required by "
+            + "your processor (for instance, 'images' or 'audio')."
             + "Make sure that the text is correctly formatted for the model "
             + "(e.g. include <image> or <|AUDIO|> tags) and that the number "
             + "of text tags match the number of additional assets provided."
@@ -444,16 +461,35 @@ def __init__(self, model: "PreTrainedModel", processor):
 
     def _prepare_model_inputs(
         self,
-        model_input: Union[str, List[str], dict],
-        output_type: Optional[OutlinesLogitsProcessor] = None,
+        model_input: Union[str, List[str], dict, List[dict]],
     ) -> Tuple[Union[str, List[str]], dict]:
         """Turn the user input into arguments to pass to the model"""
-        model_input = self.type_adapter.format_input(model_input)
+        # Format validation
+        if isinstance(model_input, list):
+            prompts = [self.type_adapter.format_input(item) for item in model_input]
+        else:
+            prompts = self.type_adapter.format_input(model_input)
+
+        # The expected format is a single dict
+        if isinstance(prompts, list):
+            merged_prompts = defaultdict(list)
+            for d in prompts:
+                for key, value in d.items():
+                    if key == "text":
+                        merged_prompts[key].append(value)
+                    else:
+                        if isinstance(value, list):
+                            merged_prompts[key].extend(value)
+                        else:
+                            merged_prompts[key].append(value)
+        else:
+            merged_prompts = prompts
+
         inputs = self.processor(
-            **model_input, padding=True, return_tensors="pt"
+            **merged_prompts, padding=True, return_tensors="pt"
         ).to(self.model.device)
 
-        return model_input["text"], inputs
+        return merged_prompts["text"], inputs
 
 
 def from_transformers(
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index f787e7072..2cd974d4e 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -136,6 +136,14 @@ def generate(
         else:
             return [message.content for message in messages]
 
+    def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError("VLLM does not support batch inference.")
+
     def generate_stream(
         self,
         model_input: Union[str, Vision],
@@ -267,6 +275,14 @@ async def generate(
         else:
             return [message.content for message in messages]
 
+    async def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError("VLLM does not support batch inference.")
+
     async def generate_stream( # type: ignore
         self,
         model_input: Union[str, Vision],
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index f970b2f9a..a51211f7e 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -9,8 +9,8 @@
 from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
 
 if TYPE_CHECKING:
-    from transformers import PreTrainedTokenizerBase
     from vllm import LLM
+    from vllm.sampling_params import SamplingParams
 
 __all__ = ["VLLMOffline", "from_vllm_offline"]
 
@@ -97,13 +97,33 @@ def __init__(self, model: "LLM"):
         self.type_adapter = VLLMOfflineTypeAdapter()
         self.lora_request = None # v0 legacy, to be removed
 
+    def _build_generation_args(
+        self,
+        inference_kwargs: dict,
+        output_type: Optional[Any] = None,
+    ) -> "SamplingParams":
+        """Create the `SamplingParams` object to pass to the `generate` method
+        of the `vllm.LLM` model."""
+        from vllm.sampling_params import GuidedDecodingParams, SamplingParams
+
+        sampling_params = inference_kwargs.pop("sampling_params", None)
+
+        if sampling_params is None:
+            sampling_params = SamplingParams()
+
+        output_type_args = self.type_adapter.format_output_type(output_type)
+        if output_type_args:
+            sampling_params.guided_decoding = GuidedDecodingParams(**output_type_args)
+
+        return sampling_params
+
     def generate(
         self,
-        model_input: Union[str, List[str]],
+        model_input: str,
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
-    ) -> Union[str, List[str], List[List[str]]]:
-        """Generate text using vLLM.
+    ) -> Union[str, List[str]]:
+        """Generate text using vLLM offline.
 
         Parameters
         ----------
@@ -118,20 +138,58 @@ def generate(
 
         Returns
         -------
-        Union[str, List[str], List[List[str]]]
+        Union[str, List[str]]
             The text generated by the model.
 
         """
-        from vllm.sampling_params import GuidedDecodingParams, SamplingParams
+        sampling_params = self._build_generation_args(
+            inference_kwargs,
+            output_type,
+        )
 
-        sampling_params = inference_kwargs.pop("sampling_params", None)
+        results = self.model.generate(
+            self.type_adapter.format_input(model_input),
+            sampling_params=sampling_params,
+            lora_request=self.lora_request, # v0 legacy, to be removed
+            **inference_kwargs,
+        )
+        results = [completion.text for completion in results[0].outputs]
 
-        if sampling_params is None:
-            sampling_params = SamplingParams()
+        if len(results) == 1:
+            return results[0]
+        else:
+            return results
 
-        output_type_args = self.type_adapter.format_output_type(output_type)
-        if output_type_args:
-            sampling_params.guided_decoding = GuidedDecodingParams(**output_type_args)
+    def generate_batch(
+        self,
+        model_input: List[str],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[List[str], List[List[str]]]:
+        """Generate a batch of completions using vLLM offline.
+
+        Parameters
+        ----------
+        prompt
+            The list of prompts based on which the model will generate a
+            response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        inference_kwargs
+            Additional keyword arguments to pass to the `generate` method
+            in the `vllm.LLM` model.
+
+        Returns
+        -------
+        Union[List[str], List[List[str]]]
+            The text generated by the model.
+
+        """
+        sampling_params = self._build_generation_args(
+            inference_kwargs,
+            output_type,
+        )
 
         results = self.model.generate(
             self.type_adapter.format_input(model_input),
@@ -141,17 +199,10 @@ def generate(
         )
         results = [[sample.text for sample in batch.outputs] for batch in results]
 
-        batch_size = len(results)
-        sample_size = len(results[0])
-
-        if batch_size == 1 and sample_size == 1:
-            return results[0][0]
-        elif batch_size == 1:
-            return results[0]
-        elif sample_size == 1:
+        if len(results[0]) == 1:
             return [batch[0] for batch in results]
-
-        return results
+        else:
+            return results
 
     def generate_stream(self, model_input, output_type, **inference_kwargs):
         """Not available for `vllm.LLM`.
@@ -160,7 +211,7 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
 
         """
         raise NotImplementedError(
-            "Streaming is not available for the vLLM integration."
+            "Streaming is not available for the vLLM offline integration."
         )
 
     def load_lora(self, adapter_path: Optional[str]) -> None:

From 0b3deed8f27f341ca75e017dce276efaf714cdff Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 2 Jul 2025 17:25:17 +0200
Subject: [PATCH 387/505] Add tests for the batch method of models

---
 .../v0_legacy/models/transformers_vision.py   |  2 +-
 tests/models/test_anthropic.py                |  8 ++++
 tests/models/test_dottxt.py                   |  8 ++++
 tests/models/test_gemini.py                   |  8 ++++
 tests/models/test_llamacpp.py                 |  7 +++
 tests/models/test_mlxlm.py                    |  8 ++++
 tests/models/test_ollama.py                   | 15 +++++++
 tests/models/test_openai.py                   |  7 +++
 tests/models/test_sglang.py                   | 15 +++++++
 tests/models/test_tgi.py                      | 15 +++++++
 tests/models/test_tgi_model_adapter.py        |  5 ---
 tests/models/test_transformers.py             | 24 +++++++---
 tests/models/test_transformers_multimodal.py  | 44 ++++++++++++-------
 .../models/test_transformers_type_adapter.py  |  5 ---
 tests/models/test_vllm.py                     | 15 +++++++
 tests/models/test_vllm_offline.py             |  8 ++--
 16 files changed, 157 insertions(+), 37 deletions(-)

diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
index 1c7657a19..fffccd448 100644
--- a/outlines/v0_legacy/models/transformers_vision.py
+++ b/outlines/v0_legacy/models/transformers_vision.py
@@ -21,7 +21,7 @@ def __init__(self, model, tokenizer, processor):
         self.type_adapter = TransformersMultiModalTypeAdapter()
 
 
-    def _prepare_model_inputs(self, model_input, output_type):
+    def _prepare_model_inputs(self, model_input):
         model_input = self.type_adapter.format_input(model_input)
         inputs = self.processor(
             **model_input, padding=True, return_tensors="pt"
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 66b3c74db..a9fa9652a 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -104,3 +104,11 @@ def test_anthopic_streaming(model):
     result = model.stream("Respond with one word. Not more.", max_tokens=1024)
     assert isinstance(result, Generator)
     assert isinstance(next(result), str)
+
+
+def test_anthropic_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+            max_tokens=1024,
+        )
diff --git a/tests/models/test_dottxt.py b/tests/models/test_dottxt.py
index 0bd1277af..df637cfe7 100644
--- a/tests/models/test_dottxt.py
+++ b/tests/models/test_dottxt.py
@@ -129,3 +129,11 @@ def test_dottxt_streaming(model):
         match="Dottxt does not support streaming"
     ):
         model.stream("Create a user", User)
+
+
+@pytest.mark.api_call
+def test_dottxt_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."]
+        )
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index bf383c3b7..8ce32a8c7 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -227,3 +227,11 @@ def test_gemini_streaming(model):
     result = model.stream("Respond with one word. Not more.")
     assert isinstance(result, Generator)
     assert isinstance(next(result), str)
+
+
+@pytest.mark.api_call
+def test_gemini_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 6a77528f6..4cb1293c5 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -132,3 +132,10 @@ def test_llamacpp_stream_text_stop(model):
     result = next(generator)
     assert isinstance(result, str)
     assert result != "a"
+
+
+def test_llamacpp_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index 8ec889230..c92236e6b 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -117,3 +117,11 @@ def test_mlxlm_stream_text_stop(model):
     )
     assert isinstance(generator, Generator)
     assert isinstance(next(generator), str)
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index e03c46fa9..a450c14ca 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -139,6 +139,13 @@ class Foo(BaseModel):
     assert "foo" in json.loads("".join(generated_text))
 
 
+def test_ollama_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
 def test_ollama_async_init_from_client():
     client = AsyncClient()
 
@@ -234,3 +241,11 @@ class Foo(BaseModel):
     async for chunk in async_generator:
         generated_text.append(chunk)
     assert "foo" in json.loads("".join(generated_text))
+
+
+@pytest.mark.asyncio
+async def test_ollama_async_batch(async_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        await async_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 6e387e961..2c780fb17 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -170,3 +170,10 @@ def test_openai_streaming(model):
     result = model.stream("Respond with one word. Not more.")
     assert isinstance(result, Generator)
     assert isinstance(next(result), str)
+
+
+def test_openai_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
index 60960f463..8099bc155 100644
--- a/tests/models/test_sglang.py
+++ b/tests/models/test_sglang.py
@@ -186,6 +186,13 @@ def test_sglang_sync_streaming(sync_model_no_model_name):
     assert isinstance(next(result), str)
 
 
+def test_sglang_sync_batch(sync_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        sync_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
 def test_sglang_sync_multiple_samples(sync_model):
     result = sync_model("Respond with a single word.", n=2)
     assert isinstance(result, list)
@@ -238,6 +245,14 @@ async def test_sglang_async_streaming(async_model_no_model_name):
         break  # Just check the first chunk
 
 
+@pytest.mark.asyncio
+async def test_sglang_async_batch(async_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        await async_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
 @pytest.mark.asyncio
 async def test_sglang_async_multiple_samples(async_model):
     result = await async_model("Respond with a single word.", n=2)
diff --git a/tests/models/test_tgi.py b/tests/models/test_tgi.py
index 52704f6c0..538f5616f 100644
--- a/tests/models/test_tgi.py
+++ b/tests/models/test_tgi.py
@@ -120,6 +120,13 @@ def test_tgi_sync_streaming(sync_model):
     assert isinstance(next(result), str)
 
 
+def test_tgi_sync_batch(sync_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        sync_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
 def test_tgi_sync_json(sync_model):
     json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
     result = sync_model("foo?", JsonSchema(json_string), max_new_tokens=10)
@@ -156,6 +163,14 @@ async def test_tgi_async_streaming(async_model):
         break  # Just check the first chunk
 
 
+@pytest.mark.asyncio
+async def test_tgi_async_batch(async_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        await async_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
 @pytest.mark.asyncio
 async def test_tgi_async_json(async_model):
     json_string = '{"type": "object", "properties": {"bar": {"type": "string"}}, "required": ["bar"]}'
diff --git a/tests/models/test_tgi_model_adapter.py b/tests/models/test_tgi_model_adapter.py
index 543ab75ff..42c91e3d6 100644
--- a/tests/models/test_tgi_model_adapter.py
+++ b/tests/models/test_tgi_model_adapter.py
@@ -43,11 +43,6 @@ def test_tgi_type_adapter_input_text(type_adapter):
 
 
 def test_tgi_type_adapter_input_invalid(type_adapter):
-    with pytest.raises(
-        NotImplementedError,
-        match="TGI does not support batch inference.",
-    ):
-        type_adapter.format_input(["foo", "bar"])
     with pytest.raises(
         NotImplementedError,
         match="is not available with TGI",
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index cd6f60811..1ed935dfd 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -143,7 +143,7 @@ class Foo(Enum):
     assert result in ["cat", "dog"]
 
 
-def test_transformers_batch_samples(model):
+def test_transformers_multiple_samples(model):
     result = model("Respond with one word. Not more.")
     assert isinstance(result, str)
     result = model(
@@ -151,12 +151,16 @@ def test_transformers_batch_samples(model):
     )
     assert isinstance(result, list)
     assert len(result) == 2
-    result = model(
+
+
+def test_transformers_batch(model):
+    result = model.batch(
         ["Respond with one word. Not more.", "Respond with one word. Not more."]
     )
     assert isinstance(result, list)
     assert len(result) == 2
-    result = model(
+
+    result = model.batch(
         ["Respond with one word. Not more.", "Respond with one word. Not more."],
         num_return_sequences=2,
         num_beams=2,
@@ -168,7 +172,7 @@ def test_transformers_batch_samples(model):
         assert len(item) == 2
 
 
-def test_transformers_batch_samples_constrained(model):
+def test_transformers_multiple_samples_constrained(model):
     class Foo(Enum):
         cat = "cat"
         dog = "dog"
@@ -178,7 +182,14 @@ class Foo(Enum):
     assert len(result) == 2
     assert result[0] in ["cat", "dog"]
     assert result[1] in ["cat", "dog"]
-    result = model(
+
+
+def test_transformers_batch_constrained(model):
+    class Foo(Enum):
+        cat = "cat"
+        dog = "dog"
+
+    result = model.batch(
         ["Cat or dog?", "Cat or dog?"],
         Foo,
     )
@@ -186,7 +197,8 @@ class Foo(Enum):
     assert len(result) == 2
     assert result[0] in ["cat", "dog"]
     assert result[1] in ["cat", "dog"]
-    result = model(
+
+    result = model.batch(
         ["Cat or dog?", "Cat or dog?"],
         Foo,
         num_return_sequences=2,
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 448bec221..2dd14a55e 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -154,7 +154,7 @@ class Foo(Enum):
     assert result in ["cat", "dog"]
 
 
-def test_transformers_vision_batch_samples(model, images):
+def test_transformers_vision_multiple_samples(model, images):
     result = model(
         {"text": "<image>Describe this image in one sentence.", "images": images[0]},
         num_return_sequences=2,
@@ -163,26 +163,36 @@ def test_transformers_vision_batch_samples(model, images):
     )
     assert isinstance(result, list)
     assert len(result) == 2
-    result = model(
-        {
-            "text": [
-                "<image>Describe this image in one sentence.",
-                "<image>Describe this image in one sentence.",
-            ],
-            "images": [images[0], images[1]],
-        },
+
+
+def test_transformers_vision_batch(model, images):
+    result = model.batch(
+        [
+            {
+                "text": "<image>Describe this image in one sentence.",
+                "images": images[0],
+            },
+            {
+                "text": "<image>Describe this image in one sentence.",
+                "images": [images[0]],
+            }
+        ],
         max_new_tokens=2,
     )
     assert isinstance(result, list)
     assert len(result) == 2
-    result = model(
-        {
-            "text": [
-                "<image>Describe this image in one sentence.",
-                "<image>Describe this image in one sentence.",
-            ],
-            "images": [images[0], images[1]],
-        },
+
+    result = model.batch(
+        [
+            {
+                "text": "<image>Describe this image in one sentence.<image>",
+                "images": [images[0], images[1]],
+            },
+            {
+                "text": "<image>Describe this image in one sentence.<image>",
+                "images": [images[0], images[1]],
+            }
+        ],
         num_return_sequences=2,
         num_beams=2,
         max_new_tokens=2,
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
index 44f6ec4e6..06d6adea8 100644
--- a/tests/models/test_transformers_type_adapter.py
+++ b/tests/models/test_transformers_type_adapter.py
@@ -32,11 +32,6 @@ def test_transformers_type_adapter_format_input(adapter):
 
     assert adapter.format_input("Hello, world!") == "Hello, world!"
 
-    assert adapter.format_input(["Hello, world!", "Hello, world!"]) == [
-        "Hello, world!",
-        "Hello, world!",
-    ]
-
 
 def test_transformers_type_adapter_format_output_type(
     adapter, logits_processor
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index b907a2652..81fd789ea 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -220,6 +220,13 @@ def test_vllm_sync_streaming(sync_model_no_model_name):
     assert isinstance(next(result), str)
 
 
+def test_vllm_sync_batch(sync_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        sync_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
 def test_vllm_sync_vision(sync_model, image):
     result = sync_model(Vision("Describe the image.", image))
     assert isinstance(result, str)
@@ -270,6 +277,14 @@ async def test_vllm_async_streaming(async_model_no_model_name):
         break  # Just check the first chunk
 
 
+@pytest.mark.asyncio
+async def test_vllm_async_batch(async_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        await async_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
 @pytest.mark.asyncio
 async def test_vllm_async_vision(async_model, image):
     result = await async_model(Vision("Describe the image.", image))
diff --git a/tests/models/test_vllm_offline.py b/tests/models/test_vllm_offline.py
index 01d5d6a59..190f13b5c 100644
--- a/tests/models/test_vllm_offline.py
+++ b/tests/models/test_vllm_offline.py
@@ -84,7 +84,7 @@ class Foo(Enum):
     assert result in ["cat", "dog"]
 
 
-def test_vllm_batch_samples(model):
+def test_vllm_multiple_samples(model):
     result = model(
         "Respond with one word. Not more.",
         sampling_params=SamplingParams(n=2)
@@ -92,13 +92,15 @@ def test_vllm_batch_samples(model):
     assert isinstance(result, list)
     assert len(result) == 2
 
-    result = model(
+
+def test_vllm_batch(model):
+    result = model.batch(
         ["Respond with one word. Not more.", "Respond with one word. Not more."]
     )
     assert isinstance(result, list)
     assert len(result) == 2
 
-    result = model(
+    result = model.batch(
         ["Respond with one word. Not more.", "Respond with one word. Not more."],
         sampling_params=SamplingParams(n=2)
     )

From bc334e630cabc549f743161e1d68e5f975d9fd4d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 3 Jul 2025 10:10:00 +0200
Subject: [PATCH 388/505] Add documentation for the batch method of models

---
 docs/features/core/generator.md               | 10 ++++
 docs/features/models/index.md                 | 40 ++++++++++++-
 docs/features/models/transformers.md          |  6 +-
 .../models/transformers_multimodal.md         | 57 ++++++++++++++++++-
 docs/features/models/vllm_offline.md          | 16 ++++++
 5 files changed, 123 insertions(+), 6 deletions(-)

diff --git a/docs/features/core/generator.md b/docs/features/core/generator.md
index 508ba399c..6c44dfdad 100644
--- a/docs/features/core/generator.md
+++ b/docs/features/core/generator.md
@@ -19,6 +19,16 @@ The `Generator` class is the core component of Outlines v1. `Generator` accepts
     - `generate.regex`
     - `generate.text`
 
+## Methods
+
+Generators implement the same methods as models:
+
+- `__call__`
+- `batch`
+- `stream`
+
+All of them take a single positional argument: the prompt or list of prompts from which to generate completions. Contrarily to the equivalent methods of models, you do not need to provide an output type as it has already been defined when initializing the generator.
+
 ## Basic Usage
 
 ```python
diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index b79d36581..7997e6e16 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -43,6 +43,44 @@ response = generator("How many countries are there in the world")
 print(result) # '200'
 ```
 
+Some models support streaming through a `stream` method. It takes the same argument as the `__call__` method, but returns an iterator instead of a string.
+
+For instance:
+
+```python
+from outlines import from_openai, Generator
+import openai
+
+# Create the model
+model = from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
+)
+
+# Stream the response
+for chunk in model.stream("Tell a short story about a cat.", max_tokens=50):
+    print(chunk) # 'This...'
+```
+
+Additionally, some models support batch processing through a `batch` method. It's similar to the `__call__` method, but takes a list of prompts instead of a single prompt and returns a list of strings.
+
+For instance:
+
+```python
+from outlines import from_transformers, Generator
+import transformers
+
+# Create a model
+model = from_transformers(
+    transformers.AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    transformers.AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+)
+
+# Call it directly
+response = model.batch(["What's the capital of Latvia?", "What's the capital of Estonia?"], max_new_tokens=20)
+print(response) # ['Riga', 'Tallinn']
+```
+
 ## Features Matrix
 
 In alphabetical order:
@@ -59,7 +97,7 @@ In alphabetical order:
 | Async | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
-| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ |
+| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ✅ |
 
 ## Model Types
 
diff --git a/docs/features/models/transformers.md b/docs/features/models/transformers.md
index 9dc1cab79..0ffc56a26 100644
--- a/docs/features/models/transformers.md
+++ b/docs/features/models/transformers.md
@@ -54,7 +54,9 @@ result = model("What's the capital of Latvia?", max_new_tokens=20)
 print(result) # 'Riga'
 ```
 
-The `Transformers` model also supports batch generation. To use it, provide a list of prompts instead of a single prompt. You will receive as result a list of completions. For instance:
+The `Transformers` model also supports batch generation. To use it, provide a list of prompts to the `batch` method. You will receive as a result a list of completions.
+
+For instance:
 
 ```python
 import outlines
@@ -74,7 +76,7 @@ prompts = [
 ]
 
 # Call it to generate text
-result = model(prompts, max_new_tokens=20)
+result = model.batch(prompts, max_new_tokens=20)
 print(result) # ['Vilnius', 'Riga', 'Tallinn']
 ```
 
diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index 6a6db8981..87e1b9667 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -31,16 +31,16 @@ model = outlines.from_transformers(hf_model, hf_processor)
 
 ## Model Input
 
-Instead of a string or a list of strings (for batch generation), you should provide a dictionary or a list of dictionaries as a prompt when calling the `TransformersMultiModal` model. The dictionary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
+Instead of a string, you should provide a dictionary as a prompt when calling the `TransformersMultiModal` model. The dictionary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
 
 ```python
 {
-    "text": Union[str, List[str]]
+    "text": str,
     "<other_keys_depending_on_your_processor>": Union[Any, List[Any]]
 }
 ```
 
-Some common keys to include for processors are `images` for vision models or `audios` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets or if you are using batch generation).
+Some common keys to include for processors are `images` for vision models or `audio` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets).
 
 Here's an example of using a vision multimodal model:
 
@@ -84,6 +84,57 @@ print(result) # '{"specie": "cat", "color": "white", "weight": 4}'
 print(Animal.model_validate_json(result)) # specie=cat, color=white, weight=4
 ```
 
+The `TransformersMultiModal` model supports batch generation. To use it, provide a list of the dictionaries described above to the `batch` method. You will receive as a result a list of completions.
+
+For instance:
+
+```python
+{
+    "text": str,
+    "<other_keys_depending_on_your_processor>": Union[Any, List[Any]]
+}
+```
+
+Some common keys to include for processors are `images` for vision models or `audio` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets).
+
+Here's an example of using a vision multimodal model:
+
+```python
+from io import BytesIO
+from urllib.request import urlopen
+
+from PIL import Image
+from transformers import (
+    LlavaForConditionalGeneration,
+    AutoProcessor,
+)
+
+import outlines
+
+TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+IMAGE_URL_2 ="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
+
+def get_image_from_url(image_url):
+    img_byte_stream = BytesIO(urlopen(image_url).read())
+    return Image.open(img_byte_stream).convert("RGB")
+
+# Create a model
+model = outlines.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+    AutoProcessor.from_pretrained(TEST_MODEL),
+)
+
+# Call the batch method with a list of model input dicts
+result = model(
+    [
+        {"text": "<image>Describe the image.", "images": get_image_from_url(IMAGE_URL)},
+        {"text": "<image>Describe the image.", "images": get_image_from_url(IMAGE_URL_2)},
+    ]
+)
+print(result) # ['The image shows a cat', 'The image shows an astronaut']
+```
+
 !!! Warning
 
     Make sure your prompt (the value of the `text` key) contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are images present in the value of the `images` key.
diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
index 38c476e59..77db06047 100644
--- a/docs/features/models/vllm_offline.md
+++ b/docs/features/models/vllm_offline.md
@@ -56,6 +56,22 @@ result = model("What's the capital of Latvia?", sampling_params=SamplingParams(m
 print(result) # 'Riga'
 ```
 
+The `VLLOffline` model also supports batch generation. To use it, provide a list of prompts to the `batch` method. You will receive as a result a list of completions. For instance:
+
+```python
+import outlines
+from vllm import LLM, SamplingParams
+
+# Create the model
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Call it to generate text
+result = model.batch(["What's the capital of Latvia?", "What's the capital of Estonia"], sampling_params=SamplingParams(max_tokens=20))
+print(result) # ['Riga', 'Tallinn']
+```
+
 ## Structured Generation
 
 The `VLLMOffline` model supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.

From 0a979b3543987b2894d21fab0b81a5d360e0f907 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 4 Jul 2025 20:08:53 +0200
Subject: [PATCH 389/505] Fix error with the tokonzer inferenace arg in
 Transformers

---
 outlines/models/transformers.py                           | 3 +--
 tests/v0_legacy/models/test_transformers_legacy.py        | 1 +
 tests/v0_legacy/models/test_transformers_vision_legacy.py | 1 +
 3 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 19350de52..a247c7e56 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -220,7 +220,6 @@ def __init__(
 
         tokenizer.padding_side = "left"
         self.model = model
-        self.transformer_tokenizer = tokenizer
         self.tokenizer = TransformerTokenizer(tokenizer)
         self.type_adapter = TransformersTypeAdapter()
 
@@ -332,9 +331,9 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
 
     def _generate_output_seq(self, prompts, inputs, **inference_kwargs):
         input_ids = inputs["input_ids"]
+
         output_ids = self.model.generate(
             **inputs,
-            tokenizer=self.transformer_tokenizer,
             **inference_kwargs,
         )
 
diff --git a/tests/v0_legacy/models/test_transformers_legacy.py b/tests/v0_legacy/models/test_transformers_legacy.py
index 2c34c2517..0388a4348 100644
--- a/tests/v0_legacy/models/test_transformers_legacy.py
+++ b/tests/v0_legacy/models/test_transformers_legacy.py
@@ -168,5 +168,6 @@ def test_transformers_legacy_call_generation():
         "foo",
         2,
         length_penalty=0.5,
+        tokenizer=model.tokenizer.tokenizer,
     )
     assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_vision_legacy.py b/tests/v0_legacy/models/test_transformers_vision_legacy.py
index bcaf4ea3e..cd7324555 100644
--- a/tests/v0_legacy/models/test_transformers_vision_legacy.py
+++ b/tests/v0_legacy/models/test_transformers_vision_legacy.py
@@ -106,5 +106,6 @@ def test_transformers_vision_legacy_call_generation():
         "foo",
         2,
         length_penalty=0.5,
+        tokenizer=model.tokenizer.tokenizer,
     )
     assert isinstance(result, str)

From 479a86100a50cd5e078d8ff472fa5dc401a26bd3 Mon Sep 17 00:00:00 2001
From: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
Date: Sat, 5 Jul 2025 15:14:15 +0200
Subject: [PATCH 390/505] Create the inputs module and modify the treatment of
 multimodal inputs (#1677)

Closes #1649
---
 docs/features/models/anthropic.md             |   7 +-
 docs/features/models/ollama.md                |   7 +-
 docs/features/models/openai.md                |   7 +-
 .../models/transformers_multimodal.md         |  55 +++-----
 docs/features/models/vllm.md                  |   7 +-
 outlines/__init__.py                          |   2 +
 outlines/inputs.py                            |  71 +++++++++++
 outlines/models/anthropic.py                  |  52 +++++---
 outlines/models/gemini.py                     |  53 +++++---
 outlines/models/ollama.py                     |  35 ++++--
 outlines/models/openai.py                     |  50 +++++---
 outlines/models/sglang.py                     |  15 ++-
 outlines/models/transformers.py               | 119 +++++++++++++-----
 outlines/models/vllm.py                       |  15 ++-
 outlines/templates.py                         |  51 ++++----
 outlines/v0_legacy/generate/api.py            |  18 ++-
 .../v0_legacy/models/transformers_vision.py   |   2 +-
 tests/models/test_anthropic.py                |  19 ++-
 tests/models/test_dottxt_type_adapter.py      |  10 +-
 tests/models/test_gemini.py                   |  18 ++-
 tests/models/test_gemini_type_adapter.py      |  20 +--
 tests/models/test_ollama.py                   |  25 ++--
 tests/models/test_ollama_type_adapter.py      |  19 ++-
 tests/models/test_openai.py                   |  17 +--
 tests/models/test_openai_type_adapter.py      |  15 +--
 tests/models/test_sglang_type_adapter.py      |  14 +--
 tests/models/test_transformers_multimodal.py  |  82 ++++++------
 ...st_transformers_multimodal_type_adapter.py |  36 +++++-
 .../models/test_transformers_type_adapter.py  |   2 +-
 tests/models/test_vllm.py                     |  14 +--
 tests/models/test_vllm_type_adapter.py        |  13 +-
 tests/test_templates.py                       |  25 ++--
 tests/v0_legacy/generate/test_api_legacy.py   |   3 +-
 .../models/test_transformers_vision_legacy.py |  27 ++--
 34 files changed, 586 insertions(+), 339 deletions(-)
 create mode 100644 outlines/inputs.py

diff --git a/docs/features/models/anthropic.md b/docs/features/models/anthropic.md
index dd0794386..c73e407aa 100644
--- a/docs/features/models/anthropic.md
+++ b/docs/features/models/anthropic.md
@@ -66,7 +66,7 @@ import requests
 import PIL
 from anthropic import Anthropic
 from outlines import from_anthropic
-from outlines.templates import Vision
+from outlines.inputs import Image
 
 # Create the model
 model = from_anthropic(
@@ -80,7 +80,10 @@ def get_image(url):
     return PIL.Image.open(io.BytesIO(r.content))
 
 # Create the prompt
-prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
 
 # Stream the response
 for chunk in model.stream(prompt, max_tokens=50):
diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
index 25fb8e70d..d610987ed 100644
--- a/docs/features/models/ollama.md
+++ b/docs/features/models/ollama.md
@@ -87,7 +87,7 @@ import requests
 import PIL
 import ollama
 import outlines
-from outlines.templates import Vision
+from outlines.inputs import Image
 
 # Create the model
 model = outlines.from_ollama(
@@ -101,7 +101,10 @@ def get_image(url):
     return PIL.Image.open(io.BytesIO(r.content))
 
 # Create the prompt
-prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
 
 # Generate text
 response = model(prompt)
diff --git a/docs/features/models/openai.md b/docs/features/models/openai.md
index e4532d213..9d31bba94 100644
--- a/docs/features/models/openai.md
+++ b/docs/features/models/openai.md
@@ -62,7 +62,7 @@ import requests
 import PIL
 import openai
 import outlines
-from outlines.templates import Vision
+from outlines.inputs import Image
 
 # Create the model
 model = outlines.from_openai(
@@ -76,7 +76,10 @@ def get_image(url):
     return PIL.Image.open(io.BytesIO(r.content))
 
 # Create the prompt
-prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
 
 # Stream the response
 for chunk in model.stream(prompt, max_tokens=50, stop="."):
diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index 87e1b9667..24d119de3 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -6,8 +6,6 @@ title: Transformers MultiModal
 
 The Outlines `TransformersMultiModal` model inherits from `Transformers` and shares most of its interface. Please start by reading the [Transformers documentation](./transformers.md) as this document only focuses on the specificities of `TransformersMultiModal` compared to `Transformers`.
 
-On top of the difference of arguments to provide to the `from_transformers` function as described below, the only other specificity of the `TransformersMultiModal` model is the format of the prompt to use when calling the model. For all other elements such as the output types and inference arguments, it works exactly the same way as `Transformers`.
-
 ## Model Initialization
 
 To load the model, you can use the `from_transformers` function. It takes 2 arguments:
@@ -31,16 +29,7 @@ model = outlines.from_transformers(hf_model, hf_processor)
 
 ## Model Input
 
-Instead of a string, you should provide a dictionary as a prompt when calling the `TransformersMultiModal` model. The dictionary should contain key-value pairs for all elements required by your processor. `text`, that contains the text prompt, is the only mandatory field. The format of that argument is:
-
-```python
-{
-    "text": str,
-    "<other_keys_depending_on_your_processor>": Union[Any, List[Any]]
-}
-```
-
-Some common keys to include for processors are `images` for vision models or `audio` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets).
+As with other multimodal models, you should provide a list containing a text prompt and assets (`Image`, `Audio` or `Video` instances) as the model input. The type of asset to provide depends on the capabilities of the `transformers` model you are running.
 
 Here's an example of using a vision multimodal model:
 
@@ -48,7 +37,7 @@ Here's an example of using a vision multimodal model:
 from io import BytesIO
 from urllib.request import urlopen
 
-from PIL import Image
+from PIL import Image as PILImage
 from pydantic import BaseModel
 from transformers import (
     LlavaForConditionalGeneration,
@@ -56,6 +45,7 @@ from transformers import (
 )
 
 import outlines
+from outlines.inputs import Image
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
 IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
@@ -65,9 +55,11 @@ class Animal(BaseModel):
     color: str
     weight: int
 
-def get_image_from_url():
-    img_byte_stream = BytesIO(urlopen(IMAGE_URL).read())
-    return Image.open(img_byte_stream).convert("RGB")
+def get_image_from_url(image_url):
+    img_byte_stream = BytesIO(urlopen(image_url).read())
+    image = PILImage.open(img_byte_stream).convert("RGB")
+    image.format = "PNG"
+    return image
 
 # Create a model
 model = outlines.from_transformers(
@@ -77,39 +69,30 @@ model = outlines.from_transformers(
 
 # Call it with a model input dict containing a text prompt and an image + an output type
 result = model(
-    {"text": "<image>Describe this animal.", "images": get_image_from_url(IMAGE_URL)},
+    ["<image>Describe this animal.", Image(get_image_from_url(IMAGE_URL))],
     Animal,
+    max_new_tokens=100
 )
 print(result) # '{"specie": "cat", "color": "white", "weight": 4}'
 print(Animal.model_validate_json(result)) # specie=cat, color=white, weight=4
 ```
 
-The `TransformersMultiModal` model supports batch generation. To use it, provide a list of the dictionaries described above to the `batch` method. You will receive as a result a list of completions.
+The `TransformersMultiModal` model supports batch generation. To use it, invoke the `batch` method with a list of lists. You will receive as a result a list of completions.
 
 For instance:
 
-```python
-{
-    "text": str,
-    "<other_keys_depending_on_your_processor>": Union[Any, List[Any]]
-}
-```
-
-Some common keys to include for processors are `images` for vision models or `audio` for audio models. The value for those keys would respectively be an image object and an audio file (or lists of those if there are several assets).
-
-Here's an example of using a vision multimodal model:
-
 ```python
 from io import BytesIO
 from urllib.request import urlopen
 
-from PIL import Image
+from PIL import Image as PILImage
 from transformers import (
     LlavaForConditionalGeneration,
     AutoProcessor,
 )
 
 import outlines
+from outlines.inputs import Image
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
 IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
@@ -117,7 +100,9 @@ IMAGE_URL_2 ="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_
 
 def get_image_from_url(image_url):
     img_byte_stream = BytesIO(urlopen(image_url).read())
-    return Image.open(img_byte_stream).convert("RGB")
+    image = PILImage.open(img_byte_stream).convert("RGB")
+    image.format = "PNG"
+    return image
 
 # Create a model
 model = outlines.from_transformers(
@@ -126,10 +111,10 @@ model = outlines.from_transformers(
 )
 
 # Call the batch method with a list of model input dicts
-result = model(
+result = model.batch(
     [
-        {"text": "<image>Describe the image.", "images": get_image_from_url(IMAGE_URL)},
-        {"text": "<image>Describe the image.", "images": get_image_from_url(IMAGE_URL_2)},
+        ["<image>Describe the image.", Image(get_image_from_url(IMAGE_URL))],
+        ["<image>Describe the image.", Image(get_image_from_url(IMAGE_URL_2))],
     ]
 )
 print(result) # ['The image shows a cat', 'The image shows an astronaut']
@@ -137,4 +122,4 @@ print(result) # ['The image shows a cat', 'The image shows an astronaut']
 
 !!! Warning
 
-    Make sure your prompt (the value of the `text` key) contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are images present in the value of the `images` key.
+    Make sure your prompt contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are image assets included in your model input.
diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index 9396cd328..64bf91f27 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -89,7 +89,7 @@ import requests
 import PIL
 import openai
 import outlines
-from outlines.templates import Vision
+from outlines.inputs import Image
 
 # Create the model
 model = outlines.from_vllm(openai.OpenAI(base_url="http://0.0.0.0:8000/v1"), "Qwen/Qwen2.5-VL-3B-Instruct")
@@ -100,7 +100,10 @@ def get_image(url):
     return PIL.Image.open(io.BytesIO(r.content))
 
 # Create the prompt
-prompt = Vision("Describe the image", get_image("https://picsum.photos/id/237/400/300"))
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
 
 # Generate text
 response = model(prompt)
diff --git a/outlines/__init__.py b/outlines/__init__.py
index 1eebb8bf0..e588d4a36 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -4,11 +4,13 @@
 from types import ModuleType
 
 import outlines.grammars
+import outlines.inputs
 import outlines.processors
 import outlines.types
 from outlines.applications import Application
 from outlines.caching import clear_cache, disable_cache, get_cache
 from outlines.generator import Generator
+from outlines.inputs import Audio, Image, Video
 from outlines.models import (
     from_dottxt,
     from_openai,
diff --git a/outlines/inputs.py b/outlines/inputs.py
new file mode 100644
index 000000000..794174718
--- /dev/null
+++ b/outlines/inputs.py
@@ -0,0 +1,71 @@
+"""Contain classes used to define the inputs of a model."""
+
+import base64
+from dataclasses import dataclass
+from io import BytesIO
+from typing import Any
+
+from PIL import Image as PILImage
+
+
+@dataclass
+class Image:
+    """Contains an image that can be passed to a multimodal model.
+
+    Provide one or several instances of this class along with a text prompt
+    in a list as the `model_input` argument to a model that supports vision.
+
+    Parameters
+    ----------
+    image
+        The image to use in the text generation.
+
+    """
+    image: PILImage.Image
+
+    def __post_init__(self):
+        image = self.image
+
+        if not image.format:
+            raise TypeError(
+                "Could not read the format of the image passed to the model."
+            )
+
+        buffer = BytesIO()
+        image.save(buffer, format=image.format)
+        self.image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
+        self.image_format = f"image/{image.format.lower()}"
+
+
+@dataclass
+class Video:
+    """Contains a video that can be passed to a multimodal model.
+
+    Provide one or several instances of this class along with a text prompt
+    in a list as the `model_input` argument to a model that supports video
+    processing.
+
+    Parameters
+    ----------
+    video
+        The video to use in the text generation.
+
+    """
+    video: Any
+
+
+@dataclass
+class Audio:
+    """Contains an audio that can be passed to a multimodal model.
+
+    Provide one or several instances of this class along with a text prompt
+    in a list as the `model_input` argument to a model that supports audio
+    processing.
+
+    Parameters
+    ----------
+    audio
+        The audio to use in the text generation.
+
+    """
+    audio: Any
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 782ba7bed..1d2bd0824 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -1,9 +1,10 @@
 """Integration with Anthropic's API."""
 
+from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, Iterator, Optional, Union
 
+from outlines.inputs import Image
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.templates import Vision
 
 if TYPE_CHECKING:
     from anthropic import Anthropic as AnthropicClient
@@ -22,7 +23,8 @@ class AnthropicTypeAdapter(ModelTypeAdapter):
 
     """
 
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
+    @singledispatchmethod
+    def format_input(self, model_input):
         """Generate the `messages` argument to pass to the client.
 
         Parameters
@@ -36,15 +38,13 @@ def format_input(self, model_input: Union[str, Vision]) -> dict:
             The `messages` argument to pass to the client.
 
         """
-        if isinstance(model_input, str):
-            return self.format_str_model_input(model_input)
-        elif isinstance(model_input, Vision):
-            return self.format_vision_model_input(model_input)
         raise TypeError(
-            f"The input type {input} is not available with Anthropic. "
-            "The only available types are `str` and `Vision`."
+            f"The input type {type(model_input)} is not available with "
+            "Anthropic. The only available types are `str` and `list` "
+            "(containing a prompt and images)."
         )
 
+    @format_input.register(str)
     def format_str_model_input(self, model_input: str) -> dict:
         return {
             "messages": [
@@ -55,21 +55,33 @@ def format_str_model_input(self, model_input: str) -> dict:
             ]
         }
 
-    def format_vision_model_input(self, model_input: Vision) -> dict:
+    @format_input.register(list)
+    def format_list_model_input(self, model_input: list) -> dict:
+        prompt = model_input[0]
+        images = model_input[1:]
+
+        if not all(isinstance(image, Image) for image in images):
+            raise ValueError("All assets provided must be of type Image")
+
+        image_content_messages = [
+            {
+                "type": "image",
+                "source": {
+                    "type": "base64",
+                    "media_type": image.image_format,
+                    "data": image.image_str,
+                },
+            }
+            for image in images
+        ]
+
         return {
             "messages": [
                 {
                     "role": "user",
                     "content": [
-                        {
-                            "type": "image",
-                            "source": {
-                                "type": "base64",
-                                "media_type": model_input.image_format,
-                                "data": model_input.image_str,
-                            },
-                        },
-                        {"type": "text", "text": model_input.prompt},
+                        *image_content_messages,
+                        {"type": "text", "text": prompt},
                     ],
                 }
             ]
@@ -111,7 +123,7 @@ def __init__(
 
     def generate(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> str:
@@ -165,7 +177,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Iterator[str]:
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 9ee7c61b2..d62b3920d 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -1,5 +1,6 @@
 """Integration with Gemini's API."""
 
+from functools import singledispatchmethod
 from typing import (
     TYPE_CHECKING,
     Any,
@@ -9,8 +10,8 @@
     get_args,
 )
 
+from outlines.inputs import Image
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.templates import Vision
 from outlines.types import CFG, Choice, JsonSchema, Regex
 from outlines.types.utils import (
     is_dataclass,
@@ -39,7 +40,8 @@ class GeminiTypeAdapter(ModelTypeAdapter):
 
     """
 
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
+    @singledispatchmethod
+    def format_input(self, model_input):
         """Generate the `contents` argument to pass to the client.
 
         Parameters
@@ -53,21 +55,36 @@ def format_input(self, model_input: Union[str, Vision]) -> dict:
             The `contents` argument to pass to the client.
 
         """
-        if isinstance(model_input, str):
-            return {"contents": [model_input]}
-        elif isinstance(model_input, Vision):
-            from google.genai import types
-
-            image_part = types.Part.from_bytes(
-                data=model_input.image_str,
-                mime_type=model_input.image_format
-            ),
-            return {"contents": [model_input.prompt, image_part]}
-        else:
-            raise TypeError(
-                f"The input type {input} is not available with Gemini. "
-                "The only available types are `str` and `Vision`."
+        raise TypeError(
+            f"The input type {type(model_input)} is not available with "
+            "Gemini. The only available types are `str` and `list` "
+            "(containing a prompt and images)."
+        )
+
+    @format_input.register(str)
+    def format_str_model_input(self, model_input: str) -> dict:
+        return {"contents": [model_input]}
+
+    @format_input.register(list)
+    def format_list_model_input(self, model_input: list) -> dict:
+        from google.genai import types
+
+        prompt = model_input[0]
+        images = model_input[1:]
+
+        if not all(isinstance(image, Image) for image in images):
+            raise ValueError("All assets provided must be of type Image")
+
+        image_parts = [
+            types.Part.from_bytes(
+                data=image.image_str,
+                mime_type=image.image_format
             )
+            for image in images
+        ]
+
+        return {"contents": [prompt, *image_parts]}
+
 
     def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the `generation_config` argument to pass to the client.
@@ -207,7 +224,7 @@ def __init__(self, client: "Client", model_name: Optional[str] = None):
 
     def generate(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs,
     ) -> str:
@@ -253,7 +270,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs,
     ) -> Iterator[str]:
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index f4d6453bf..59cd1ed03 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,12 +1,13 @@
 """Integration with the `ollama` library."""
 
 import json
+from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
 
 from pydantic import TypeAdapter
 
+from outlines.inputs import Image
 from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
-from outlines.templates import Vision
 from outlines.types import CFG, JsonSchema, Regex
 from outlines.types.utils import (
     is_dataclass,
@@ -25,7 +26,8 @@
 class OllamaTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `Ollama` model."""
 
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
+    @singledispatchmethod
+    def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
 
         Parameters
@@ -39,28 +41,35 @@ def format_input(self, model_input: Union[str, Vision]) -> dict:
             The model input to be passed to the client.
 
         """
-        if isinstance(model_input, str):
-            return self.format_str_model_input(model_input)
-        elif isinstance(model_input, Vision):
-            return self.format_vision_model_input(model_input)
         raise TypeError(
-            f"The input type {model_input} is not available with Ollama. "
-            "The only available types are `str` and `Vision`."
+            f"The input type {type(model_input)} is not available with "
+            "Ollama. The only available types are `str` and `list` "
+            "(containing a prompt and images)."
         )
 
+    @format_input.register(str)
     def format_str_model_input(self, model_input: str) -> dict:
-        """Format the string model input to pass to the client.
+        """Generate the `prompt` argument to pass to the client when the user
+        only passes a prompt.
 
         """
         return {"prompt": model_input}
 
-    def format_vision_model_input(self, model_input: Vision) -> dict:
-        """Format the vision model input to pass to the client.
+    @format_input.register(list)
+    def format_list_model_input(self, model_input: list) -> dict:
+        """Generate the `prompt` and `images` arguments to pass to the client
+        when the user passes a prompt and images.
 
         """
+        prompt = model_input[0]
+        images = model_input[1:]
+
+        if not all(isinstance(image, Image) for image in images):
+            raise ValueError("All assets provided must be of type Image")
+
         return {
-            "prompt": model_input.prompt,
-            "images": [model_input.image_str]
+            "prompt": prompt,
+            "images": [image.image_str for image in images]
         }
 
     def format_output_type(
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index be86d69e6..2b31192dd 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -15,13 +15,14 @@
     Union,
 )
 from dataclasses import asdict, dataclass, field, replace
+from functools import singledispatchmethod
 
 from pydantic import BaseModel, TypeAdapter
 
 from outlines.caching import cache
+from outlines.inputs import Image
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.utils import set_additional_properties_false_json_schema
-from outlines.templates import Vision
 from outlines.types import JsonSchema, Regex, CFG
 from outlines.types.utils import (
     is_dataclass,
@@ -47,7 +48,8 @@ class OpenAITypeAdapter(ModelTypeAdapter):
 
     """
 
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
+    @singledispatchmethod
+    def format_input(self, model_input):
         """Generate the `messages` argument to pass to the client.
 
         Parameters
@@ -61,15 +63,13 @@ def format_input(self, model_input: Union[str, Vision]) -> dict:
             The formatted input to be passed to the client.
 
         """
-        if isinstance(model_input, str):
-            return self.format_str_model_input(model_input)
-        elif isinstance(model_input, Vision):
-            return self.format_vision_model_input(model_input)
         raise TypeError(
-            f"The input type {input} is not available with OpenAI. "
-            "The only available types are `str` and `Vision`."
+            f"The input type {type(model_input)} is not available with "
+            "OpenAI. The only available types are `str` and `list` "
+            "(containing a prompt and images)."
         )
 
+    @format_input.register(str)
     def format_str_model_input(self, model_input: str) -> dict:
         """Generate the `messages` argument to pass to the client when the user
         only passes a prompt.
@@ -84,23 +84,35 @@ def format_str_model_input(self, model_input: str) -> dict:
             ]
         }
 
-    def format_vision_model_input(self, model_input: Vision) -> dict:
+    @format_input.register(list)
+    def format_list_model_input(self, model_input: list) -> dict:
         """Generate the `messages` argument to pass to the client when the user
-        passes a prompt and an image.
+        passes a prompt and images.
 
         """
+        prompt = model_input[0]
+        images = model_input[1:]
+
+        if not all(isinstance(image, Image) for image in images):
+            raise ValueError("All assets provided must be of type Image")
+
+        image_parts = [
+            {
+                "type": "image_url",
+                "image_url": {
+                    "url": f"data:{image.image_format};base64,{image.image_str}"  # noqa: E702
+                },
+            }
+            for image in images
+        ]
+
         return {
             "messages": [
                 {
                     "role": "user",
                     "content": [
-                        {"type": "text", "text": model_input.prompt},
-                        {
-                            "type": "image_url",
-                            "image_url": {
-                                "url": f"data:{model_input.image_format};base64,{model_input.image_str}"  # noqa: E702
-                            },
-                        },
+                        {"type": "text", "text": prompt},
+                        *image_parts,
                     ],
                 }
             ]
@@ -258,7 +270,7 @@ def __init__(
 
     def generate(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Union[type[BaseModel], str]] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -328,7 +340,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Union[type[BaseModel], str]] = None,
         **inference_kwargs,
     ) -> Iterator[str]:
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
index 1ac159bcd..2438214b1 100644
--- a/outlines/models/sglang.py
+++ b/outlines/models/sglang.py
@@ -14,7 +14,6 @@
     python_types_to_terms,
     to_regex,
 )
-from outlines.templates import Vision
 
 if TYPE_CHECKING:
     from openai import AsyncOpenAI, OpenAI
@@ -25,7 +24,7 @@
 class SGLangTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `SGLang` and `AsyncSGLang` models."""
 
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
+    def format_input(self, model_input: Union[str, list]) -> dict:
         """Generate the prompt argument to pass to the client.
 
         We rely on the OpenAITypeAdapter to format the input as the sglang
@@ -105,7 +104,7 @@ def __init__(self, client, model_name: Optional[str] = None):
 
     def generate(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -161,7 +160,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Iterator[str]:
@@ -198,7 +197,7 @@ def generate_stream(
 
     def _build_client_args(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
@@ -249,7 +248,7 @@ def __init__(self, client, model_name: Optional[str] = None):
 
     async def generate(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -303,7 +302,7 @@ async def generate_batch(
 
     async def generate_stream( # type: ignore
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> AsyncIterator[str]:
@@ -341,7 +340,7 @@ async def generate_stream( # type: ignore
 
     def _build_client_args(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index a247c7e56..25e4eb09b 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,9 +1,12 @@
 """Integration with the `transformers` library."""
 
+import warnings
+
 from collections import defaultdict
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union
 
+from outlines.inputs import Audio, Image, Video
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
 from outlines.processors import OutlinesLogitsProcessor
@@ -147,8 +150,8 @@ def format_input(self, model_input):
             The formatted input to be passed to the model.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available."
+        raise TypeError(
+            f"The input type {type(model_input)} is not available."
             "The only available type is `str`."
         )
 
@@ -237,12 +240,17 @@ def __init__(
             self.tensor_library_name = "torch"
 
     def _prepare_model_inputs(
-        self, model_input: Union[str, List[str], dict, List[dict]]
+        self,
+        model_input: Union[str, List[str], dict, List[dict]],
+        is_batch: bool = False,
     ) -> Tuple[Union[str, List[str]], dict]:
         """Turn the user input into arguments to pass to the model"""
         # Format validation
-        if isinstance(model_input, list):
-            prompts = [self.type_adapter.format_input(item) for item in model_input]
+        if is_batch:
+            prompts = [
+                self.type_adapter.format_input(item)
+                for item in model_input
+            ]
         else:
             prompts = self.type_adapter.format_input(model_input)
         input_ids, attention_mask = self.tokenizer.encode(prompts)
@@ -281,11 +289,14 @@ def generate(
             The text generated by the model.
 
         """
-        prompts, inputs = self._prepare_model_inputs(model_input)
+        prompts, inputs = self._prepare_model_inputs(model_input, False)
         logits_processor = self.type_adapter.format_output_type(output_type)
 
         generated_ids = self._generate_output_seq(
-            prompts, inputs, logits_processor=logits_processor, **inference_kwargs
+            prompts,
+            inputs,
+            logits_processor=logits_processor,
+            **inference_kwargs,
         )
 
         # required for multi-modal models that return a 2D tensor even when
@@ -303,7 +314,7 @@ def generate_batch(
         **inference_kwargs: Any,
     ) -> List[Union[str, List[str]]]:
         """"""
-        prompts, inputs = self._prepare_model_inputs(model_input) # type: ignore
+        prompts, inputs = self._prepare_model_inputs(model_input, True) # type: ignore
         logits_processor = self.type_adapter.format_output_type(output_type)
 
         generated_ids = self._generate_output_seq(
@@ -313,9 +324,7 @@ def generate_batch(
         # if there are multiple samples per input, convert generated_id to 3D
         num_samples = inference_kwargs.get("num_return_sequences", 1)
         if num_samples > 1:
-            print("here", generated_ids)
             generated_ids = generated_ids.view(len(model_input), num_samples, -1)
-            print("here2", generated_ids)
 
         return self._decode_generation(generated_ids)
 
@@ -357,7 +366,8 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
             ]
         else:  # pragma: no cover
             raise TypeError(
-                f"Generated outputs aren't 1D, 2D or 3D, but instead are {generated_ids.shape}"
+                "Generated outputs aren't 1D, 2D or 3D, but instead are "
+                f"{generated_ids.shape}"
             )
 
 
@@ -366,7 +376,7 @@ class TransformersMultiModalTypeAdapter(ModelTypeAdapter):
 
     @singledispatchmethod
     def format_input(self, model_input):
-        """Generate the prompt arguments to pass to the model.
+        """Fomat the prompt arguments to pass to the model.
 
         Argument
         --------
@@ -376,21 +386,36 @@ def format_input(self, model_input):
         Returns
         -------
         dict
-            The formatted input to be passed to the model.
+            The formatted input.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available. Please provide a "
-            + "dictionary containing at least the 'text' key with a string "
-            + "value. You should also include the other keys required by "
-            + "your processor (for instance, 'images' or 'audio')."
-            + "Make sure that the text is correctly formatted for the model "
-            + "(e.g. include <image> or <|AUDIO|> tags) and that the number "
-            + "of text tags match the number of additional assets provided."
+        raise TypeError(
+            f"The input type {type(model_input)} is not available. Please "
+            + "provide a list containing a text prompt and assets "
+            + "(`Image`, `Audio` or `Video` instances) supported by your "
+            + "model."
         )
 
     @format_input.register(dict)
-    def format_list_input(self, model_input: dict) -> dict:
+    def format_dict_input(self, model_input: dict) -> dict:
+        warnings.warn("""
+            Providing the input as a dict is deprecated. Support for this will
+            be removed in the v1.2.0 release of Outlines. Use a list containing
+            a text prompt and assets (`Image`, `Audio` or `Video` instances)
+            instead.
+            For instance:
+            ```python
+            from outlines import Image
+            model = from_transformers(mymodel, myprocessor)
+            response = model([
+                "A beautiful image of a cat",
+                Image(my_image),
+            ])
+            ```
+            """,
+            DeprecationWarning,
+            stacklevel=2,
+        )
         if "text" not in model_input:
             raise ValueError(
                 "The input must contain the 'text' key along with the other "
@@ -398,6 +423,37 @@ def format_list_input(self, model_input: dict) -> dict:
             )
         return model_input
 
+    @format_input.register(list)
+    def format_list_input(self, model_input: list) -> dict:
+        prompt = model_input[0]
+        assets = model_input[1:]
+
+        asset_types = set(type(asset) for asset in assets)
+        if len(asset_types) > 1:
+            raise ValueError(
+                "All assets must be of the same type. "
+                + f"Found types: {asset_types}"
+            )
+        asset_type = asset_types.pop()
+
+        if asset_type == Image:
+            return {
+                "text": prompt,
+                "images": [asset.image for asset in assets]
+            }
+        elif asset_type == Audio: # pragma: no cover
+            return {
+                "text": prompt,
+                "audio": [asset.audio for asset in assets]
+            }
+        elif asset_type == Video: # pragma: no cover
+            return {
+                "text": prompt,
+                "videos": [asset.video for asset in assets]
+            }
+        else:
+            raise ValueError(f"Unsupported asset type: {asset_type}")
+
     def format_output_type(
         self,
         output_type: Optional[OutlinesLogitsProcessor] = None,
@@ -436,7 +492,7 @@ def __init__(self, model: "PreTrainedModel", processor):
         """Create a TransformersMultiModal model instance
 
         We rely on the `__init__` method of the `Transformers` class to handle
-        most of the initialization and then add elements specific to vision
+        most of the initialization and then add elements specific to multimodal
         models.
 
         Parameters
@@ -461,28 +517,27 @@ def __init__(self, model: "PreTrainedModel", processor):
     def _prepare_model_inputs(
         self,
         model_input: Union[str, List[str], dict, List[dict]],
+        is_batch: bool = False,
     ) -> Tuple[Union[str, List[str]], dict]:
         """Turn the user input into arguments to pass to the model"""
-        # Format validation
-        if isinstance(model_input, list):
-            prompts = [self.type_adapter.format_input(item) for item in model_input]
+        if is_batch:
+            prompts = [
+                self.type_adapter.format_input(item) for item in model_input
+            ]
         else:
             prompts = self.type_adapter.format_input(model_input)
 
         # The expected format is a single dict
-        if isinstance(prompts, list):
+        if is_batch:
             merged_prompts = defaultdict(list)
             for d in prompts:
                 for key, value in d.items():
                     if key == "text":
                         merged_prompts[key].append(value)
                     else:
-                        if isinstance(value, list):
-                            merged_prompts[key].extend(value)
-                        else:
-                            merged_prompts[key].append(value)
+                        merged_prompts[key].extend(value)
         else:
-            merged_prompts = prompts
+            merged_prompts = prompts # type: ignore
 
         inputs = self.processor(
             **merged_prompts, padding=True, return_tensors="pt"
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 2cd974d4e..8bfaea482 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -6,7 +6,6 @@
 from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
 from outlines.models.openai import OpenAITypeAdapter
 from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
-from outlines.templates import Vision
 
 if TYPE_CHECKING:
     from openai import AsyncOpenAI, OpenAI
@@ -17,7 +16,7 @@
 class VLLMTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `VLLM` and `AsyncVLLM` models."""
 
-    def format_input(self, model_input: Union[str, Vision]) -> dict:
+    def format_input(self, model_input: Union[str, list]) -> dict:
         """Generate the prompt argument to pass to the client.
 
         We rely on the OpenAITypeAdapter to format the input as the vLLM server
@@ -92,7 +91,7 @@ def __init__(
 
     def generate(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -146,7 +145,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Iterator[str]:
@@ -183,7 +182,7 @@ def generate_stream(
 
     def _build_client_args(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
@@ -233,7 +232,7 @@ def __init__(
 
     async def generate(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -285,7 +284,7 @@ async def generate_batch(
 
     async def generate_stream( # type: ignore
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> AsyncIterator[str]:
@@ -322,7 +321,7 @@ async def generate_stream( # type: ignore
 
     def _build_client_args(
         self,
-        model_input: Union[str, Vision],
+        model_input: Union[str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
diff --git a/outlines/templates.py b/outlines/templates.py
index cd6a4fbf5..79620e721 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -1,6 +1,5 @@
 """Create templates to easily build prompts."""
 
-import base64
 import functools
 import inspect
 import json
@@ -8,22 +7,20 @@
 import re
 import textwrap
 from dataclasses import dataclass
-from io import BytesIO
 from pathlib import Path
 from typing import Any, Callable, Dict, Optional, Type, cast
 import warnings
 
 import jinja2
-from PIL import Image
 from pydantic import BaseModel
+from PIL import Image as PILImage
 
+from outlines.inputs import Image
 
-@dataclass
-class Vision:
-    """Contains the input for a vision model.
 
-    Provide an instance of this class as the `model_input` argument to a model
-    that supports vision.
+def Vision(prompt: str, image: PILImage.Image) -> list:
+    """This factory function replaces the deprecated `Vision` class until it is
+    fully removed in outlines v1.2.0.
 
     Parameters
     ----------
@@ -32,22 +29,30 @@ class Vision:
     image
         The image to use to generate the response.
 
+    Returns
+    -------
+    list
+        A list containing the prompt and Image instance.
     """
-    prompt: str
-    image: Image.Image
-
-    def __post_init__(self):
-        image = self.image
-
-        if not image.format:
-            raise TypeError(
-                "Could not read the format of the image passed to the model."
-            )
-
-        buffer = BytesIO()
-        image.save(buffer, format=image.format)
-        self.image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
-        self.image_format = f"image/{image.format.lower()}"
+    warnings.warn("""
+        The Vision function is deprecated and will be removed in outlines 1.2.0.
+        Instead of using Vision, please use a prompt along with an
+        outlines.inputs.Image instance.
+        For instance:
+        ```python
+        import openai
+        from outlines import Image, from_openai
+        model = from_openai("gpt-4o")
+        response = model(
+            ["A beautiful image of a cat", Image(my_image)],
+            max_tokens=100
+        )
+        ```
+        """,
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    return [prompt, Image(image)]
 
 
 @dataclass
diff --git a/outlines/v0_legacy/generate/api.py b/outlines/v0_legacy/generate/api.py
index 625bac105..6e824f0b0 100644
--- a/outlines/v0_legacy/generate/api.py
+++ b/outlines/v0_legacy/generate/api.py
@@ -88,12 +88,20 @@ def __call__(
         seed: Optional[int] = None,
         **model_specific_params,
     ):
-        result = self.generator(
-            prompts,
-            **self.create_inference_params(
-                max_tokens, stop_at, seed, **model_specific_params
+        if isinstance(prompts, list):
+            result = self.generator.batch(
+                prompts,
+                **self.create_inference_params(
+                    max_tokens, stop_at, seed, **model_specific_params
+                )
+            )
+        else:
+            result = self.generator(
+                prompts,
+                **self.create_inference_params(
+                    max_tokens, stop_at, seed, **model_specific_params
+                )
             )
-        )
         if isinstance(result, list):
             return [self.format_sequence(r) for r in result]
         else:
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
index fffccd448..33c78c8b7 100644
--- a/outlines/v0_legacy/models/transformers_vision.py
+++ b/outlines/v0_legacy/models/transformers_vision.py
@@ -21,7 +21,7 @@ def __init__(self, model, tokenizer, processor):
         self.type_adapter = TransformersMultiModalTypeAdapter()
 
 
-    def _prepare_model_inputs(self, model_input):
+    def _prepare_model_inputs(self, model_input, is_batch=False):
         model_input = self.type_adapter.format_input(model_input)
         inputs = self.processor(
             **model_input, padding=True, return_tensors="pt"
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index a9fa9652a..49485edd7 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -1,13 +1,13 @@
-import io
 from typing import Generator
 
 from anthropic import Anthropic as AnthropicClient
-from PIL import Image
+from PIL import Image as PILImage
 import pytest
 
 import outlines
+from outlines.inputs import Image, Video
 from outlines.models.anthropic import Anthropic
-from outlines.templates import Vision
+
 
 MODEL_NAME = "claude-3-haiku-20240307"
 
@@ -26,7 +26,7 @@ def model_no_model_name():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
     image.format = "PNG"
 
     return image
@@ -54,7 +54,7 @@ def test_anthropic_wrong_inference_parameters():
         model.generate("prompt", foo=10, max_tokens=1024)
 
 
-def test_anthropic_wrong_input_type():
+def test_anthropic_wrong_input_type(image):
     class Foo:
         def __init__(self, foo):
             self.foo = foo
@@ -63,6 +63,9 @@ def __init__(self, foo):
         model = Anthropic(AnthropicClient(), MODEL_NAME)
         model.generate(Foo("prompt"))
 
+    with pytest.raises(ValueError, match="All assets provided must be of type Image"):
+        model.generate(["foo?", Image(image), Video("")])
+
 
 def test_anthropic_wrong_output_type():
     class Foo:
@@ -94,7 +97,11 @@ def test_anthropic_direct_call(model_no_model_name):
 @pytest.mark.api_call
 def test_anthropic_simple_vision(model, image):
     result = model.generate(
-        Vision("What does this logo represent?", image), max_tokens=1024
+        [
+            "What does this logo represent?",
+            Image(image),
+        ],
+        max_tokens=1024,
     )
     assert isinstance(result, str)
 
diff --git a/tests/models/test_dottxt_type_adapter.py b/tests/models/test_dottxt_type_adapter.py
index 0253af1ea..a0356502c 100644
--- a/tests/models/test_dottxt_type_adapter.py
+++ b/tests/models/test_dottxt_type_adapter.py
@@ -4,12 +4,12 @@
 import sys
 from dataclasses import dataclass
 
-from PIL import Image
+from PIL import Image as PILImage
 from genson import SchemaBuilder
 from pydantic import BaseModel
 
+from outlines.inputs import Image
 from outlines.models.dottxt import DottxtTypeAdapter
-from outlines.templates import Vision
 from outlines.types import cfg, json_schema, regex
 
 if sys.version_info >= (3, 12):
@@ -35,13 +35,13 @@ def schema():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -58,7 +58,7 @@ def test_dottxt_type_adapter_input_text(adapter):
 
 
 def test_dottxt_type_adapter_input_invalid(adapter, image):
-    prompt = Vision("prompt", image)
+    prompt = ["prompt", image]
     with pytest.raises(TypeError, match="The input type"):
         _ = adapter.format_input(prompt)
 
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index 8ce32a8c7..d923278ce 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -6,13 +6,13 @@
 from typing import Generator, Literal
 
 import pytest
-from PIL import Image
+from PIL import Image as PILImage
 from google.genai import Client
 from pydantic import BaseModel, ValidationError
 
 import outlines
+from outlines.inputs import Image, Video
 from outlines.models.gemini import Gemini
-from outlines.templates import Vision
 from outlines.types import Choice
 
 if sys.version_info >= (3, 12):
@@ -37,13 +37,13 @@ def model_no_model_name():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -71,6 +71,12 @@ def test_gemini_wrong_inference_parameters(model):
         model.generate("prompt", foo=10)
 
 
+@pytest.mark.api_call
+def test_gemini_wrong_input_type(model, image):
+    with pytest.raises(ValueError, match="All assets provided must be of type Image"):
+        model.generate(["foo?", Image(image), Video("")])
+
+
 @pytest.mark.api_call
 def test_gemini_simple_call(model):
     result = model.generate("Respond with one word. Not more.")
@@ -88,7 +94,7 @@ def test_gemini_direct_call(model_no_model_name):
 
 @pytest.mark.api_call
 def test_gemini_simple_vision(model, image):
-    result = model.generate(Vision("What does this logo represent?", image))
+    result = model.generate(["What does this logo represent?", Image(image)])
     assert isinstance(result, str)
 
 
@@ -107,7 +113,7 @@ def test_gemini_simple_vision_pydantic(model, image):
     class Logo(BaseModel):
         name: int
 
-    result = model.generate(Vision("What does this logo represent?", image), Logo)
+    result = model.generate(["What does this logo represent?", Image(image)], Logo)
     assert isinstance(result, str)
     assert "name" in json.loads(result)
 
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
index 46834eae9..a467df750 100644
--- a/tests/models/test_gemini_type_adapter.py
+++ b/tests/models/test_gemini_type_adapter.py
@@ -5,14 +5,14 @@
 from enum import Enum, EnumMeta
 from typing import Literal
 
-from PIL import Image
+from PIL import Image as PILImage
 from genson import SchemaBuilder
 from google.genai import types
 from pydantic import BaseModel
 
 from outlines import cfg, json_schema, regex
+from outlines.inputs import Image
 from outlines.models.gemini import GeminiTypeAdapter
-from outlines.templates import Vision
 
 if sys.version_info >= (3, 12):
     from typing import TypedDict, is_typeddict
@@ -37,13 +37,13 @@ def schema():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -60,16 +60,16 @@ def test_gemini_type_adapter_input_text(adapter):
 
 
 def test_gemini_type_adapter_input_vision(adapter, image):
-    input_message = Vision("hello", image)
+    input_message = ["hello", Image(image)]
     result = adapter.format_input(input_message)
     image_part = types.Part.from_bytes(
-        data=input_message.image_str,
-        mime_type=input_message.image_format
-    ),
-    assert result == {"contents": [input_message.prompt, image_part]}
+        data=input_message[1].image_str,
+        mime_type=input_message[1].image_format
+    )
+    assert result == {"contents": [input_message[0], image_part]}
 
 
-def test_dottxt_type_adapter_input_invalid(adapter):
+def test_gemini_type_adapter_input_invalid(adapter):
     @dataclass
     class Audio:
         file: str
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index a450c14ca..69df85532 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -4,13 +4,13 @@
 from typing import Annotated
 
 import pytest
-from PIL import Image
+from PIL import Image as PILImage
 from ollama import AsyncClient, Client
 from pydantic import BaseModel, Field
 
 import outlines
+from outlines.inputs import Image, Video
 from outlines.models import AsyncOllama, Ollama
-from outlines.templates import Vision
 
 
 MODEL_NAME = "tinyllama"
@@ -40,13 +40,13 @@ def async_model_no_model_name():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -66,6 +66,10 @@ def test_ollama_init_from_client():
     assert model.client == client
     assert model.model_name is None
 
+    # With invalid client
+    with pytest.raises(ValueError, match="Invalid client type"):
+        outlines.from_ollama(object())
+
 
 def test_ollama_wrong_inference_parameters(model):
     with pytest.raises(TypeError, match="got an unexpected"):
@@ -94,7 +98,7 @@ def test_ollama_simple_vision(image, model):
     # This is not using a vision model, so it's not able to describe
     # the image, but we're still checking the model input syntax
     result = model.generate(
-        Vision("What does this logo represent?", image),
+        ["What does this logo represent?", Image(image)],
         model=MODEL_NAME,
     )
     assert isinstance(result, str)
@@ -118,9 +122,12 @@ class Foo(Enum):
         model.generate("foo?", Foo)
 
 
-def test_ollama_wrong_input_type(model):
+def test_ollama_wrong_input_type(model, image):
     with pytest.raises(TypeError, match="is not available"):
-        model.generate(["foo?", "bar?"], None)
+        model.generate({"foo?": "bar?"}, None)
+
+    with pytest.raises(ValueError, match="All assets provided must be of type Image"):
+        model.generate(["foo?", Image(image), Video("")], None)
 
 
 def test_ollama_stream(model):
@@ -193,7 +200,7 @@ async def test_ollama_async_simple_vision(image, async_model):
     # This is not using a vision model, so it's not able to describe
     # the image, but we're still checking the model input syntax
     result = await async_model.generate(
-        Vision("What does this logo represent?", image),
+        ["What does this logo represent?", Image(image)],
         model=MODEL_NAME,
     )
     assert isinstance(result, str)
@@ -222,7 +229,7 @@ class Foo(Enum):
 @pytest.mark.asyncio
 async def test_ollama_async_wrong_input_type(async_model):
     with pytest.raises(TypeError, match="is not available"):
-        await async_model.generate(["foo?", "bar?"], None)
+        await async_model.generate({"foo?": "bar?"}, None)
 
 
 @pytest.mark.asyncio
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
index feced7834..47c93ed67 100644
--- a/tests/models/test_ollama_type_adapter.py
+++ b/tests/models/test_ollama_type_adapter.py
@@ -5,11 +5,11 @@
 from dataclasses import dataclass
 
 from genson import SchemaBuilder
-from PIL import Image
+from PIL import Image as PILImage
 from pydantic import BaseModel
 
+from outlines.inputs import Image
 from outlines.models.ollama import OllamaTypeAdapter
-from outlines.templates import Vision
 from outlines.types import cfg, json_schema, regex
 
 if sys.version_info >= (3, 12):
@@ -35,13 +35,13 @@ def schema():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -59,18 +59,15 @@ def test_ollama_type_adapter_input_text(adapter):
 
 
 def test_ollama_type_adapter_input_vision(adapter, image):
-    prompt = Vision("prompt", image)
+    prompt = ["prompt", Image(image)]
     result = adapter.format_input(prompt)
     assert isinstance(result, dict)
-    assert result.get("prompt") == prompt.prompt
-    assert result.get("images") == [prompt.image_str]
+    assert result.get("prompt") == prompt[0]
+    assert result.get("images") == [prompt[1].image_str]
 
 
 def test_ollama_type_adapter_input_invalid(adapter):
-    prompt = [
-        "This is a first test",
-        "This is a second test",
-    ]
+    prompt = {"foo": "bar"}
     with pytest.raises(TypeError, match="The input type"):
         _ = adapter.format_input(prompt)
 
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 2c780fb17..38b77eaa8 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -4,13 +4,13 @@
 from typing import Annotated, Generator
 
 import pytest
-from PIL import Image
+from PIL import Image as PILImage
 from openai import OpenAI as OpenAIClient
 from pydantic import BaseModel, Field
 
 import outlines
+from outlines.inputs import Image, Video
 from outlines.models.openai import OpenAI
-from outlines.templates import Vision
 from outlines.types import json_schema
 
 MODEL_NAME = "gpt-4o-mini-2024-07-18"
@@ -34,13 +34,13 @@ def api_key():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -76,7 +76,7 @@ def test_openai_wrong_inference_parameters(model):
         model.generate("prompt", foo=10)
 
 
-def test_openai_wrong_input_type(model):
+def test_openai_wrong_input_type(model, image):
     class Foo:
         def __init__(self, foo):
             self.foo = foo
@@ -84,6 +84,9 @@ def __init__(self, foo):
     with pytest.raises(TypeError, match="is not available"):
         model.generate(Foo("prompt"))
 
+    with pytest.raises(ValueError, match="All assets provided must be of type Image"):
+        model.generate(["foo?", Image(image), Video("")])
+
 
 def test_openai_wrong_output_type(model):
     class Foo:
@@ -120,7 +123,7 @@ def test_openai_direct_call(model_no_model_name):
 
 @pytest.mark.api_call
 def test_openai_simple_vision(image, model):
-    result = model.generate(Vision("What does this logo represent?", image))
+    result = model.generate(["What does this logo represent?", Image(image)])
     assert isinstance(result, str)
 
 
@@ -148,7 +151,7 @@ def test_openai_simple_vision_pydantic(image, model):
     class Logo(BaseModel):
         name: int
 
-    result = model.generate(Vision("What does this logo represent?", image), Logo)
+    result = model.generate(["What does this logo represent?", Image(image)], Logo)
     assert isinstance(result, str)
     assert "name" in json.loads(result)
 
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
index fcafbb8de..295ad08c3 100644
--- a/tests/models/test_openai_type_adapter.py
+++ b/tests/models/test_openai_type_adapter.py
@@ -6,10 +6,11 @@
 from typing import Literal
 
 from genson import SchemaBuilder
-from PIL import Image
+from PIL import Image as PILImage
 from pydantic import BaseModel
 
-from outlines import cfg, json_schema, regex, Vision
+from outlines import cfg, json_schema, regex
+from outlines.inputs import Image
 from outlines.models.openai import OpenAITypeAdapter
 
 if sys.version_info >= (3, 12):
@@ -36,13 +37,13 @@ def schema():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -62,7 +63,7 @@ def test_openai_type_adapter_input_text(adapter):
 
 
 def test_openai_type_adapter_input_vision(adapter, image):
-    input_message = Vision("hello", image)
+    input_message = ["hello", Image(image)]
     result = adapter.format_input(input_message)
     assert isinstance(result, dict)
 
@@ -78,11 +79,11 @@ def test_openai_type_adapter_input_vision(adapter, image):
     assert message["content"][1]["type"] == "image_url"
     assert (
         message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message.image_str}"
+        == f"data:image/png;base64,{input_message[1].image_str}"
     )
 
 
-def test_dottxt_type_adapter_input_invalid(adapter):
+def test_openai_type_adapter_input_invalid(adapter):
     @dataclass
     class Audio:
         file: str
diff --git a/tests/models/test_sglang_type_adapter.py b/tests/models/test_sglang_type_adapter.py
index 7530174f9..6cb51f662 100644
--- a/tests/models/test_sglang_type_adapter.py
+++ b/tests/models/test_sglang_type_adapter.py
@@ -3,12 +3,12 @@
 import pytest
 from dataclasses import dataclass
 
-from PIL import Image
+from PIL import Image as PILImage
+
+from outlines.inputs import Image
 from outlines.models.sglang import SGLangTypeAdapter
 from outlines.types import CFG, JsonSchema
 
-from outlines import Vision
-
 
 CFG_STRING = """
 ?start: expr
@@ -46,13 +46,13 @@ def json_schema_whitespace_instance():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -67,7 +67,7 @@ def test_sglang_type_adapter_input_text(type_adapter):
 
 
 def test_sglang_type_adapter_input_vision(type_adapter, image):
-    input_message = Vision("hello", image)
+    input_message = ["hello", Image(image)]
     result = type_adapter.format_input(input_message)
     assert isinstance(result, dict)
 
@@ -83,7 +83,7 @@ def test_sglang_type_adapter_input_vision(type_adapter, image):
     assert message["content"][1]["type"] == "image_url"
     assert (
         message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message.image_str}"
+        == f"data:image/png;base64,{input_message[1].image_str}"
     )
 
 
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 2dd14a55e..7741fbdeb 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -6,7 +6,7 @@
 from urllib.request import urlopen
 
 import pytest
-from PIL import Image
+from PIL import Image as PILImage
 from pydantic import BaseModel
 from transformers import (
     LlavaForConditionalGeneration,
@@ -14,6 +14,7 @@
 )
 
 import outlines
+from outlines.inputs import Image
 from outlines.models.transformers import (
     TransformersMultiModal,
     TransformerTokenizer,
@@ -32,7 +33,9 @@
 def images():
     def img_from_url(url):
         img_byte_stream = BytesIO(urlopen(url).read())
-        return Image.open(img_byte_stream).convert("RGB")
+        image = PILImage.open(img_byte_stream).convert("RGB")
+        image.format = "PNG"
+        return image
 
     return [img_from_url(url) for url in IMAGE_URLS]
 
@@ -58,7 +61,7 @@ def test_transformers_vision_instantiate_simple():
 
 def test_transformers_vision_simple(model, images):
     result = model.generate(
-        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
+        ["<image>Describe this image in one sentence:", Image(images[0])],
         None,
         max_new_tokens=2,
     )
@@ -67,7 +70,7 @@ def test_transformers_vision_simple(model, images):
 
 def test_transformers_vision_call(model, images):
     result = model(
-        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
+        ["<image>Describe this image in one sentence:", Image(images[0])],
         max_new_tokens=2,
     )
     assert isinstance(result, str)
@@ -76,21 +79,22 @@ def test_transformers_vision_call(model, images):
 def test_transformers_vision_wrong_number_images(model, images):
     with pytest.raises(ValueError):
         model(
-            {
-                "text": "<image>Describe this image in one sentence:",
-                "images": [images[0], images[1]],
-            }
+            [
+                "<image>Describe this image in one sentence:",
+                Image(images[0]),
+                Image(images[1]),
+            ],
         )
 
 
 def test_transformers_vision_wrong_input_type(model):
-    with pytest.raises(NotImplementedError):
+    with pytest.raises(TypeError):
         model.generate("invalid input", None)
 
 
 def test_transformers_inference_kwargs(model, images):
     result = model(
-        {"text": "<image>Describe this image in one sentence:", "images": images[0]},
+        ["<image>Describe this image in one sentence:", Image(images[0])],
         max_new_tokens=2,
     )
     assert isinstance(result, str)
@@ -99,20 +103,21 @@ def test_transformers_inference_kwargs(model, images):
 def test_transformers_invalid_inference_kwargs(model, images):
     with pytest.raises(ValueError):
         model(
-            {
-                "text": "<image>Describe this image in one sentence:",
-                "images": images[0],
-            },
+            [
+                "<image>Describe this image in one sentence:",
+                Image(images[0]),
+            ],
             foo="bar",
         )
 
 
 def test_transformers_several_images(model, images):
     result = model(
-        {
-            "text": "<image><image>Describe this image in one sentence:",
-            "images": [images[0], images[1]],
-        },
+        [
+            "<image><image>Describe this image in one sentence:",
+            Image(images[0]),
+            Image(images[1]),
+        ],
         max_new_tokens=2,
     )
     assert isinstance(result, str)
@@ -123,7 +128,7 @@ class Foo(BaseModel):
         name: str
 
     result = model(
-        {"text": "<image>Give a name to this animal.", "images": images[0]},
+        ["<image>Give a name to this animal.", Image(images[0])],
         Foo,
         max_new_tokens=10,
     )
@@ -132,7 +137,7 @@ class Foo(BaseModel):
 
 def test_transformers_vision_regex(model, images):
     result = model(
-        {"text": "<image>How old is it?", "images": images[0]},
+        ["<image>How old is it?", Image(images[0])],
         Regex(r"[0-9]")
     )
 
@@ -146,7 +151,7 @@ class Foo(Enum):
         dog = "dog"
 
     result = model(
-        {"text": "<image>Is it a cat or a dog?", "images": images[0]},
+        ["<image>Is it a cat or a dog?", Image(images[0])],
         Foo,
     )
 
@@ -156,7 +161,7 @@ class Foo(Enum):
 
 def test_transformers_vision_multiple_samples(model, images):
     result = model(
-        {"text": "<image>Describe this image in one sentence.", "images": images[0]},
+        ["<image>Describe this image in one sentence.", Image(images[0])],
         num_return_sequences=2,
         num_beams=2,
         max_new_tokens=2,
@@ -168,14 +173,8 @@ def test_transformers_vision_multiple_samples(model, images):
 def test_transformers_vision_batch(model, images):
     result = model.batch(
         [
-            {
-                "text": "<image>Describe this image in one sentence.",
-                "images": images[0],
-            },
-            {
-                "text": "<image>Describe this image in one sentence.",
-                "images": [images[0]],
-            }
+            ["<image>Describe this image in one sentence.", Image(images[0])],
+            ["<image>Describe this image in one sentence.", Image(images[0])],
         ],
         max_new_tokens=2,
     )
@@ -184,14 +183,8 @@ def test_transformers_vision_batch(model, images):
 
     result = model.batch(
         [
-            {
-                "text": "<image>Describe this image in one sentence.<image>",
-                "images": [images[0], images[1]],
-            },
-            {
-                "text": "<image>Describe this image in one sentence.<image>",
-                "images": [images[0], images[1]],
-            }
+            ["<image>Describe this image in one sentence.<image>", Image(images[0]), Image(images[1])],
+            ["<image>Describe this image in one sentence.<image>", Image(images[0]), Image(images[1])],
         ],
         num_return_sequences=2,
         num_beams=2,
@@ -202,3 +195,16 @@ def test_transformers_vision_batch(model, images):
     for item in result:
         assert isinstance(item, list)
         assert len(item) == 2
+
+
+def test_transformers_vision_deprecated_input_type(model, images):
+    with pytest.warns(DeprecationWarning):
+        result = model.generate(
+            {
+                "text": "<image>Describe this image in one sentence:",
+                "images": images[0],
+            },
+            None,
+            max_new_tokens=2,
+        )
+        assert isinstance(result, str)
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index 28569d25f..84df7f084 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -1,9 +1,11 @@
 import pytest
 
-import outlines
 import transformers
+from PIL import Image as PILImage
 from transformers import LogitsProcessorList
 
+import outlines
+from outlines.inputs import Image, Video
 from outlines.models.transformers import TransformersMultiModalTypeAdapter
 from outlines.processors.structured import RegexLogitsProcessor
 
@@ -26,14 +28,38 @@ def logits_processor():
     )
 
 
-def test_transformers_multimodal_type_adapter_format_input(adapter):
-    with pytest.raises(NotImplementedError):
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+    image.format = "PNG"
+
+    return image
+
+
+def test_transformers_multimodal_type_adapter_format_input(adapter, image):
+    with pytest.raises(TypeError):
         adapter.format_input("hello")
 
     with pytest.raises(ValueError):
-        adapter.format_input({"foo": "bar"})
+        with pytest.deprecated_call():
+            adapter.format_input({"foo": "bar"})
+
+    with pytest.raises(ValueError, match="All assets must be of the same type"):
+        adapter.format_input(["foo", Image(image), Video("")])
+
+    class MockAsset:
+        pass
+
+    with pytest.raises(ValueError, match="Unsupported asset type"):
+        adapter.format_input(["foo", MockAsset()])
 
-    assert adapter.format_input({"text": "foo"}) == {"text": "foo"}
+    image_asset = Image(image)
+    assert adapter.format_input(["foo", image_asset]) == {
+        "text": "foo",
+        "images": [image_asset.image],
+    }
 
 
 def test_transformers_multimodal_type_adapter_format_output_type(
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
index 06d6adea8..da02573a3 100644
--- a/tests/models/test_transformers_type_adapter.py
+++ b/tests/models/test_transformers_type_adapter.py
@@ -27,7 +27,7 @@ def logits_processor():
 
 
 def test_transformers_type_adapter_format_input(adapter):
-    with pytest.raises(NotImplementedError):
+    with pytest.raises(TypeError):
         adapter.format_input(int)
 
     assert adapter.format_input("Hello, world!") == "Hello, world!"
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index 81fd789ea..74a8425f1 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -6,11 +6,11 @@
 from typing import AsyncGenerator, Generator
 
 import pytest
-from PIL import Image
+from PIL import Image as PILImage
 from openai import AsyncOpenAI, OpenAI
 
+from outlines.inputs import Image
 from outlines.models.vllm import VLLM, AsyncVLLM, from_vllm
-from outlines.templates import Vision
 from outlines.types.dsl import CFG, Regex, JsonSchema
 from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
 
@@ -40,7 +40,7 @@
 def get_base64_image():
     width, height = 256, 256
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
@@ -161,13 +161,13 @@ def async_model_no_model_name():
 def image():
     width, height = 256, 256
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -228,7 +228,7 @@ def test_vllm_sync_batch(sync_model):
 
 
 def test_vllm_sync_vision(sync_model, image):
-    result = sync_model(Vision("Describe the image.", image))
+    result = sync_model(["Describe the image.", Image(image)])
     assert isinstance(result, str)
 
 
@@ -287,7 +287,7 @@ async def test_vllm_async_batch(async_model):
 
 @pytest.mark.asyncio
 async def test_vllm_async_vision(async_model, image):
-    result = await async_model(Vision("Describe the image.", image))
+    result = await async_model(["Describe the image.", Image(image)])
     assert isinstance(result, str)
 
 
diff --git a/tests/models/test_vllm_type_adapter.py b/tests/models/test_vllm_type_adapter.py
index 46d021a8c..526693e4c 100644
--- a/tests/models/test_vllm_type_adapter.py
+++ b/tests/models/test_vllm_type_adapter.py
@@ -3,11 +3,12 @@
 import pytest
 from dataclasses import dataclass
 
-from PIL import Image
+from PIL import Image as PILImage
+
+from outlines.inputs import Image
 from outlines.models.vllm import VLLMTypeAdapter
 from outlines.types import CFG, JsonSchema
 
-from outlines import Vision
 
 
 CFG_STRING = """
@@ -46,13 +47,13 @@ def json_schema_whitespace_instance():
 def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
-    image = Image.new("RGB", (width, height), white_background)
+    image = PILImage.new("RGB", (width, height), white_background)
 
     # Save to an in-memory bytes buffer and read as png
     buffer = io.BytesIO()
     image.save(buffer, format="PNG")
     buffer.seek(0)
-    image = Image.open(buffer)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -67,7 +68,7 @@ def test_vllm_type_adapter_input_text(type_adapter):
 
 
 def test_vllm_type_adapter_input_vision(type_adapter, image):
-    input_message = Vision("hello", image)
+    input_message = ["hello", Image(image)]
     result = type_adapter.format_input(input_message)
     assert isinstance(result, dict)
 
@@ -83,7 +84,7 @@ def test_vllm_type_adapter_input_vision(type_adapter, image):
     assert message["content"][1]["type"] == "image_url"
     assert (
         message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message.image_str}"
+        == f"data:image/png;base64,{input_message[1].image_str}"
     )
 
 
diff --git a/tests/test_templates.py b/tests/test_templates.py
index cf9d0ed5f..225ad21ca 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -1,15 +1,15 @@
 import base64
-import json
 import os
 import tempfile
 from typing import Dict, List
 
 import pytest
-from PIL import Image
+from PIL import Image as PILImage
 from io import BytesIO
 from pydantic import BaseModel, Field
 
 import outlines
+from outlines.inputs import Image
 from outlines.templates import (
     Template,
     build_template_from_string,
@@ -44,32 +44,37 @@ class PydanticClass(BaseModel):
 
 def test_vision_initialization():
     # Create a simple image for testing
-    image = Image.new("RGB", (10, 10), color="red")
+    image = PILImage.new("RGB", (10, 10), color="red")
     image.format = "PNG"
 
     # Initialize the Vision object
-    vision = Vision(prompt="Test prompt", image=image)
+    with pytest.deprecated_call():
+        vision = Vision(prompt="Test prompt", image=image)
 
     # Check that the prompt is set correctly
-    assert vision.prompt == "Test prompt"
+    assert isinstance(vision, list)
+    assert len(vision) == 2
+    assert vision[0] == "Test prompt"
+    assert isinstance(vision[1], Image)
 
     # Check that the image is encoded correctly
     buffer = BytesIO()
     image.save(buffer, format=image.format)
     expected_image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
-    assert vision.image_str == expected_image_str
+    assert vision[1].image_str == expected_image_str
 
     # Check that the image format is set correctly
-    assert vision.image_format == "image/png"
+    assert vision[1].image_format == "image/png"
 
 
 def test_vision_invalid_image_format():
     # Create an image without a format
-    image = Image.new("RGB", (10, 10), color="blue")
+    image = PILImage.new("RGB", (10, 10), color="blue")
 
     # Expect a TypeError when the image format is not set
-    with pytest.raises(TypeError, match="Could not read the format"):
-        Vision(prompt="Test prompt", image=image)
+    with pytest.deprecated_call():
+        with pytest.raises(TypeError, match="Could not read the format"):
+            Vision(prompt="Test prompt", image=image)
 
 
 def render(content: str, **kwargs):
diff --git a/tests/v0_legacy/generate/test_api_legacy.py b/tests/v0_legacy/generate/test_api_legacy.py
index 75cf72ff0..c34a3b706 100644
--- a/tests/v0_legacy/generate/test_api_legacy.py
+++ b/tests/v0_legacy/generate/test_api_legacy.py
@@ -176,7 +176,8 @@ def test_generator_v0_adapter_vision_call_stream(transformers_vision_model):
     media = img_from_url(
         "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
     )
-    result = generator_v0_adapter("Hello, world!<image>", media, max_tokens=10)
+    with pytest.deprecated_call():
+        result = generator_v0_adapter("Hello, world!<image>", media, max_tokens=10)
     assert isinstance(result, str)
     with pytest.raises(
         NotImplementedError,
diff --git a/tests/v0_legacy/models/test_transformers_vision_legacy.py b/tests/v0_legacy/models/test_transformers_vision_legacy.py
index cd7324555..8a8632418 100644
--- a/tests/v0_legacy/models/test_transformers_vision_legacy.py
+++ b/tests/v0_legacy/models/test_transformers_vision_legacy.py
@@ -96,16 +96,17 @@ def test_transformers_vision_legacy_call_generation():
         )
     assert isinstance(generator, GeneratorV0Adapter)
 
-    result = generator(
-        "Hello, world!<image>",
-        img_from_url(
-            "https://upload.wikimedia.org/wikipedia/commons"
-            + "/2/25/Siam_lilacpoint.jpg"
-        ),
-        10,
-        "foo",
-        2,
-        length_penalty=0.5,
-        tokenizer=model.tokenizer.tokenizer,
-    )
-    assert isinstance(result, str)
+    with pytest.deprecated_call():
+      result = generator(
+          "Hello, world!<image>",
+          img_from_url(
+              "https://upload.wikimedia.org/wikipedia/commons"
+              + "/2/25/Siam_lilacpoint.jpg"
+          ),
+          10,
+          "foo",
+          2,
+          length_penalty=0.5,
+          tokenizer=model.tokenizer.tokenizer,
+      )
+      assert isinstance(result, str)

From 999a611378056ec6a4eb2353518b955b4d5ebbfc Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 9 Jul 2025 18:37:21 +0200
Subject: [PATCH 391/505] Create the Chat model input

---
 docs/features/core/inputs.md  | 204 ++++++++++++++++++++++++++++++++++
 docs/features/models/index.md |   6 +-
 docs/guide/chat_templating.md |  35 +-----
 mkdocs.yml                    |   1 +
 outlines/inputs.py            | 114 ++++++++++++++++++-
 tests/test_inputs.py          | 174 +++++++++++++++++++++++++++++
 6 files changed, 498 insertions(+), 36 deletions(-)
 create mode 100644 docs/features/core/inputs.md
 create mode 100644 tests/test_inputs.py

diff --git a/docs/features/core/inputs.md b/docs/features/core/inputs.md
new file mode 100644
index 000000000..47c4969c2
--- /dev/null
+++ b/docs/features/core/inputs.md
@@ -0,0 +1,204 @@
+---
+title: Model Inputs
+---
+
+# Model Inputs
+
+Outlines models accept various types of inputs to generate text. The input format depends on the capabilities of the underlying model and the type of task you want to perform. The most basic type of input is a single string prompt, it's accepted by all models.
+
+## Overview
+
+The model input is the first argument of the `__call__`, `stream` and `batch` methods of both models and generators.
+
+There are 3 types of model inputs:
+
+- **Text prompts** - Simple strings
+- **Multimodal inputs** - List containning a string prompt along with assets
+- **Chat inputs** - `Chat` instances containing messages
+
+## Text Prompts
+
+The simplest form of input is a plain text string. This works with all models and is suitable for standard text generation tasks.
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Create a model
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+)
+
+# Simple text prompt
+response = model("What's the capital of France?", max_new_tokens=20)
+print(response)  # 'Paris'
+```
+
+## Multimodal Inputs (Vision)
+
+For models that support them, you can provide a list containing a text prompt and one or more assets.
+
+There are 3 types of assets defined in Outlines:
+
+- `Image`: contains a PIL Image
+- `Video`: contains any object (you must choose a format that is supported by your model)
+- `Audio`: contains any object (you must choose a format that is supported by your model)
+
+Among those, `Image` is by far the most important as multiple models support vision inputs.
+
+For instance with vision input:
+
+```python
+import io
+import requests
+import PIL
+import outlines
+import openai
+from outlines.inputs import Image
+
+# Create the model
+model = outlines.from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt containing the text and the image
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+## Chat Inputs
+
+For conversational models, you can use the `Chat` class to provide a conversation history with multiple messages.
+
+A `Chat` instance is instantiated with an optional list of messages. Each message must be a dictionary containing two mandatory keys:
+- `role`: must be one of `system`, `assistant` or `user`
+- `content`: must be either a string or a multimodal input (if the model supports it)
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+from outlines.inputs import Chat, Image
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
+print(prompt)
+# {'role': 'system', 'content': 'You are a helpful assistant.'}
+# {'role': 'user', 'content': ['Describe the image', Image(image=<PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=400x300 at 0x7FFA526CCC40>)]}
+```
+
+After having created a `Chat` instance, you can add one or several messages thanks to the `append` and `extend` methods. You can also remove the last message of the Chat with the `pop` method.
+
+For instance:
+
+```python
+from outlines.inputs import Chat
+
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+])
+
+# Add a message
+prompt.append({"role": "user", "content": "How are you doing today?"})
+print(prompt)
+# {'role': 'system', 'content': 'You are a helpful assistant.'}
+# {'role': 'user', 'content': 'How are you doing today?'}
+
+# Remove the last messsage
+last_message = prompt.pop()
+print(last_message)
+# {'role': 'user', 'content': 'How are you doing today?'}
+print(prompt)
+# {'role': 'system', 'content': 'You are a helpful assistant.'}
+
+# RAdd several messages
+prompt.extend([
+    {"role": "user", "content": "How are you doing today?"},
+    {"role": "assistant", "content": "Excellent, thanks!"}
+])
+print(prompt)
+# {'role': 'system', 'content': 'You are a helpful assistant.'}
+# {'role': 'user', 'content': 'How are you doing today?'}
+# {'role': 'assistant', 'content': 'Excellent, thanks!'}
+```
+
+Finally, there are three convenience method to easily add a message:
+
+- add_system_message
+- add_user_message
+- add_assistant_message
+
+As the role is already set, you only need to provide the content.
+
+For instance:
+
+```python
+from outlines.inputs import Chat
+
+# Create the chat input
+prompt = Chat()
+
+prompt.add_system_message("You are a helpful assistant.")
+prompt.add_system_message("How are you doing today?")
+prompt.add_system_message("Excellent, thanks!")
+
+print(prompt)
+# {'role': 'system', 'content': 'You are a helpful assistant.'}
+# {'role': 'user', 'content': 'How are you doing today?'}
+# {'role': 'assistant', 'content': 'Excellent, thanks!'}
+```
+
+## Batching
+
+In the case of batching, for models that support it, you just have to provide several instances of the model inputs described above in a list.
+
+For instance:
+
+```python
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Create model
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Create a list of prompts that will be used in a single batch
+prompts = [
+    "What's the capital of Lithuania?",
+    "What's the capital of Latvia?",
+    "What's the capital of Estonia?"
+]
+
+# Call it to generate text
+result = model.batch(prompts, max_new_tokens=20)
+print(result) # ['Vilnius', 'Riga', 'Tallinn']
+```
diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index 7997e6e16..525f5c1b7 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -15,7 +15,11 @@ Outlines models are objects that wrap an inference client or engine. Models prov
 
 All Outlines model classes have an associated loader function to facilitate initializing a model instance. The name of this function is `from_` plus the name of the model in lower-case letters. For instance, Outlines has a `Transformers` model and an associated `from_transformers` loader function. The parameters to load a model are specific to each provider, please consult the documentation of the model you want to use for more information.
 
-After having created a model instance, you can either directly call it to generate text or first create a reusable generator that you would then call. In either case, you can provide an `output_type` to constrain the format of the generation output. See the [output types section](../core/output_types.md) for more information on constrained generation.
+After having created a model instance, you can either directly call it to generate text or first create a reusable generator that you would then call.
+
+The input you must provide to a model to generate text can be a simple text prompt or a vision or chat input for models that support them. See the [model inputs section](../core/inputs.md) for more information on model inputs formats.
+
+In all cases, you can provide an `output_type` to constrain the format of the generation output. See the [output types section](../core/output_types.md) for more information on constrained generation.
 
 For instance:
 
diff --git a/docs/guide/chat_templating.md b/docs/guide/chat_templating.md
index d5752b8fe..dfb24fa56 100644
--- a/docs/guide/chat_templating.md
+++ b/docs/guide/chat_templating.md
@@ -2,37 +2,4 @@
 
 Instruction-tuned language models use "special tokens" to indicate different parts of text, such as the system prompt, the user prompt, any images, and the assistant's response. A [chat template](https://huggingface.co/docs/transformers/main/en/chat_templating) is how different types of input are composited together into a single, machine-readable string.
 
-Outlines does not manage chat templating tokens when using instruct models. You must apply the chat template tokens to the prompt yourself -- if you do not apply chat templating on instruction-tuned models, you will often get nonsensical output from the model.
-
-Chat template tokens are not needed for base models.
-
-You can find the chat template tokens in the model's HuggingFace repo or documentation. As an example, the `SmolLM2-360M-Instruct` special tokens can be found [here](https://huggingface.co/HuggingFaceTB/SmolLM2-360M-Instruct/blob/main/special_tokens_map.json).
-
-However, it can be slow to manually look up a model's special tokens, and special tokens vary by models. If you change the model, your prompts may break if you have hard-coded special tokens.
-
-If you need a convenient tool to apply chat templating for you, you should use the `tokenizer` from the `transformers` library:
-
-```python
-from transformers import AutoTokenizer
-
-tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/SmolLM2-360M-Instruct")
-prompt = tokenizer.apply_chat_template(
-    [
-        {"role": "system", "content": "You extract information from text."},
-        {"role": "user", "content": "What food does the following text describe?"},
-    ],
-    tokenize=False,
-    add_bos=True,
-    add_generation_prompt=True,
-)
-```
-
-yields
-
-```ascii
-<|im_start|>system
-You extract information from text.<|im_end|>
-<|im_start|>user
-What food does the following text describe?<|im_end|>
-<|im_start|>assistant
-```
+Outlines supports chat templating throught the `Chat` model input class. It contains a list of messages similar in format to the chat history you would use with API models such as OpenAI or Anthropic and to the expected arguments of the `apply_chat_template` method of transformers tokenizers. You can find detailed information on the interface of this object in the [model inputs documentation](../features/core/inputs.md).
diff --git a/mkdocs.yml b/mkdocs.yml
index ac723ccc0..2bb854383 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -144,6 +144,7 @@ nav:
                     - TransformersMultiModal: features/models/transformers_multimodal.md
                     - vLLM (online server): features/models/vllm.md
                     - vLLM (offline): features/models/vllm_offline.md
+                - Model Inputs: features/core/inputs.md
                 - Output Types:
                     - Overview: features/core/output_types.md
                     - Basic Types: features/core/output_types#basic-python-types
diff --git a/outlines/inputs.py b/outlines/inputs.py
index 794174718..50a6a6741 100644
--- a/outlines/inputs.py
+++ b/outlines/inputs.py
@@ -3,7 +3,7 @@
 import base64
 from dataclasses import dataclass
 from io import BytesIO
-from typing import Any
+from typing import Any, Dict, List, Optional
 
 from PIL import Image as PILImage
 
@@ -69,3 +69,115 @@ class Audio:
 
     """
     audio: Any
+
+
+@dataclass
+class Chat:
+    """Contains the input for a chat model.
+
+    Provide an instance of this class as the `model_input` argument to a model
+    that supports chat.
+
+    Each message contained in the messages list must be a dict with 'role' and
+    'content' keys. The role can be 'user', 'assistant', or 'system'. The content
+    can be a string or a list containing a str and assets (images, videos,
+    audios, etc.) in the case of multimodal models.
+
+    Examples
+    --------
+    ```python
+    # Initialize the chat with a system message.
+    chat_prompt = Chat([
+        {"role": "system", "content": "You are a helpful assistant."},
+    ])
+
+    # Add a user message with an image and call the model (not shown here).
+    chat_prompt.add_user_message(["Describe the image below", Image(image)])
+
+    # Add as an assistant message the response from the model.
+    chat_prompt.add_assistant_message("The is a black cat sitting on a couch.")
+    ```
+
+    Parameters
+    ----------
+    messages
+        The list of messages that will be provided to the model.
+
+    """
+    messages: List[Dict[str, Any]] = None # type: ignore
+
+    def __post_init__(self):
+        if self.messages is None:
+            self.messages = []
+
+    def append(self, message: Dict[str, Any]):
+        """Add a message to the chat.
+
+        Parameters
+        ----------
+        message
+            The message to add to the chat.
+
+        """
+        self.messages.append(message)
+
+    def extend(self, messages: List[Dict[str, Any]]):
+        """Add a list of messages to the chat.
+
+        Parameters
+        ----------
+        messages
+            The list of messages to add to the chat.
+
+        """
+        self.messages.extend(messages)
+
+    def pop(self) -> Dict[str, Any]:
+        """Remove the last message from the chat.
+
+        Returns
+        -------
+        message
+            The removed message.
+
+        """
+        return self.messages.pop()
+
+    def add_system_message(self, content: str | List[Any]):
+        """Add a system message to the chat.
+
+        Parameters
+        ----------
+        content
+            The content of the system message.
+
+        """
+        self.messages.append({"role": "system", "content": content})
+
+    def add_user_message(self, content: str | List[Any]):
+        """Add a user message to the chat.
+
+        Parameters
+        ----------
+        content
+            The content of the user message.
+
+        """
+        self.messages.append({"role": "user", "content": content})
+
+    def add_assistant_message(self, content: str | List[Any]):
+        """Add an assistant message to the chat.
+
+        Parameters
+        ----------
+        content
+            The content of the assistant message.
+
+        """
+        self.messages.append({"role": "assistant", "content": content})
+
+    def __str__(self):
+        return "\n".join(str(message) for message in self.messages)
+
+    def __repr__(self):
+        return f"Chat(messages={self.messages})"
diff --git a/tests/test_inputs.py b/tests/test_inputs.py
new file mode 100644
index 000000000..d020a48c9
--- /dev/null
+++ b/tests/test_inputs.py
@@ -0,0 +1,174 @@
+"""Unit tests for the inputs module."""
+
+import base64
+import tempfile
+from io import BytesIO
+from typing import Dict, List, Any
+
+import pytest
+from PIL import Image as PILImage
+
+from outlines.inputs import Image, Video, Audio, Chat
+
+
+@pytest.fixture
+def image_input():
+    image = PILImage.new("RGB", (100, 100), color="red")
+    image.format = "PNG"
+    buffer = BytesIO()
+    image.save(buffer, format="PNG")
+    return Image(image=image)
+
+
+def test_image_initialization():
+    # png
+    image = PILImage.new("RGB", (100, 100), color="red")
+    image.format = "PNG"
+    buffer = BytesIO()
+    image.save(buffer, format="PNG")
+    image_input = Image(image=image)
+
+    assert image_input.image == image
+    assert image_input.image_format == "image/png"
+    assert image_input.image_str == base64.b64encode(buffer.getvalue()).decode("utf-8")
+
+    # jpeg
+    image = PILImage.new("RGB", (100, 100), color="blue")
+    image.format = "JPEG"
+    buffer = BytesIO()
+    image.save(buffer, format="JPEG")
+    image_input = Image(image=image)
+
+    assert image_input.image == image
+    assert image_input.image_format == "image/jpeg"
+    assert image_input.image_str == base64.b64encode(buffer.getvalue()).decode("utf-8")
+
+
+def test_image_initialization_invalid():
+    """Test that Image initialization fails when image has no format."""
+    # No format
+    image = PILImage.new("RGB", (100, 100), color="yellow")
+    with pytest.raises(TypeError, match="Could not read the format of the image"):
+        Image(image=image)
+
+    # Empty string format
+    image = PILImage.new("RGB", (100, 100), color="orange")
+    image.format = ""
+    with pytest.raises(TypeError, match="Could not read the format of the image"):
+        Image(image=image)
+
+
+def test_video_initialization():
+    video = "foo"
+    video_input = Video(video=video)
+    assert video_input.video == video
+
+
+def test_audio_initialization():
+    audio = "foo"
+    audio_input = Audio(audio=audio)
+    assert audio_input.audio == audio
+
+
+def test_chat_initialization():
+    # Empty
+    chat = Chat()
+    assert chat.messages == []
+    assert len(chat.messages) == 0
+    assert str(chat) == ""
+    assert repr(chat) == "Chat(messages=[])"
+
+    # With messages
+    messages = [
+        {"role": "system", "content": "You are a helpful assistant."},
+        {"role": "user", "content": "Hello!"},
+        {"role": "assistant", "content": "Hi there!"}
+    ]
+    chat = Chat(messages=messages)
+    assert chat.messages == messages
+    assert len(chat.messages) == 3
+    assert str(chat) == "{'role': 'system', 'content': 'You are a helpful assistant.'}\n{'role': 'user', 'content': 'Hello!'}\n{'role': 'assistant', 'content': 'Hi there!'}"
+    assert repr(chat) == "Chat(messages=[{'role': 'system', 'content': 'You are a helpful assistant.'}, {'role': 'user', 'content': 'Hello!'}, {'role': 'assistant', 'content': 'Hi there!'}])"
+
+
+def test_chat_append():
+    chat = Chat(messages=[])
+    message = {"role": "user", "content": "Hello"}
+    chat.append(message)
+    assert len(chat.messages) == 1
+    assert chat.messages[0] == message
+
+def test_chat_extend():
+    chat = Chat(messages=[])
+    messages = [
+        {"role": "user", "content": "Hello"},
+        {"role": "assistant", "content": "Hi"}
+    ]
+    chat.extend(messages)
+    assert len(chat.messages) == 2
+    assert chat.messages == messages
+
+def test_chat_pop():
+    # Pop from non-empty chat
+    messages = [
+        {"role": "user", "content": "Hello"},
+        {"role": "assistant", "content": "Hi"}
+    ]
+    chat = Chat(messages=messages.copy())
+    popped_message = chat.pop()
+    assert popped_message == {"role": "assistant", "content": "Hi"}
+    assert len(chat.messages) == 1
+    assert chat.messages[0] == {"role": "user", "content": "Hello"}
+
+    # Pop from empty chat
+    chat = Chat(messages=[])
+    with pytest.raises(IndexError):
+        chat.pop()
+
+
+def test_chat_add_system_message(image_input):
+    # Add a string
+    chat = Chat(messages=[])
+    chat.add_system_message("You are a helpful assistant.")
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "system"
+    assert chat.messages[0]["content"] == "You are a helpful assistant."
+
+    # Add a list
+    chat = Chat(messages=[])
+    chat.add_system_message(["prompt", image_input])
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "system"
+    assert chat.messages[0]["content"] == ["prompt", image_input]
+
+
+def test_add_user_message_string(image_input):
+    # Add a string
+    chat = Chat(messages=[])
+    chat.add_user_message("Hello, how are you?")
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "user"
+    assert chat.messages[0]["content"] == "Hello, how are you?"
+
+    # Add a list
+    chat = Chat(messages=[])
+    chat.add_user_message(["prompt", image_input])
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "user"
+    assert chat.messages[0]["content"] == ["prompt", image_input]
+
+
+def test_add_assistant_message_string(image_input):
+    # Add a string
+    chat = Chat(messages=[])
+    chat.add_assistant_message("I'm doing well, thank you!")
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "assistant"
+    assert chat.messages[0]["content"] == "I'm doing well, thank you!"
+
+    # Add a list
+    chat = Chat(messages=[])
+    chat.add_assistant_message(["prompt", image_input])
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "assistant"
+    assert chat.messages[0]["content"] == ["prompt", image_input]

From bb21050925221a7ad6eb7ad86c33c9bd05d3dd03 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 9 Jul 2025 18:37:38 +0200
Subject: [PATCH 392/505] Add the Chat input to Anthropic

---
 docs/features/models/anthropic.md          |  73 ++++++++++++-
 outlines/models/anthropic.py               |  93 ++++++++++------
 tests/models/test_anthopic_type_adapter.py | 118 +++++++++++++++++++++
 tests/models/test_anthropic.py             |  22 +++-
 4 files changed, 266 insertions(+), 40 deletions(-)
 create mode 100644 tests/models/test_anthopic_type_adapter.py

diff --git a/docs/features/models/anthropic.md b/docs/features/models/anthropic.md
index c73e407aa..403040ca4 100644
--- a/docs/features/models/anthropic.md
+++ b/docs/features/models/anthropic.md
@@ -37,7 +37,7 @@ Check the [Anthropic documentation](https://docs.anthropic.com/en/docs/about-cla
 
 ## Text Generation
 
-Once you've created your Outlines `Anthropic` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
+Once you've created your Outlines `Anthropic` model instance, you're all set to generate text with this provider. You can simply call the model with a text prompt.
 
 For instance:
 
@@ -56,7 +56,9 @@ response = model("What's the capital of Latvia?", max_tokens=20)
 print(response) # 'Riga'
 ```
 
-Anthropic does not support structured generation, so you cannot provide an output type. It does support streaming and vision inputs though.
+#### Vision
+
+Some Anthropic models support vision input. To use this feature, provide a list containing a text prompt and `Image` instances.
 
 For instance:
 
@@ -79,15 +81,76 @@ def get_image(url):
     r = requests.get(url)
     return PIL.Image.open(io.BytesIO(r.content))
 
-# Create the prompt
+# Create the prompt containing the text and the image
 prompt = [
     "Describe the image",
     Image(get_image("https://picsum.photos/id/237/400/300"))
 ]
 
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Chat
+
+You can also use chat inputs with the `Anthropic` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+from anthropic import Anthropic
+from outlines import from_anthropic
+from outlines.inputs import Chat, Image
+
+# Create the model
+model = from_anthropic(
+    Anthropic(),
+    "claude-3-5-sonnet-latest"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the chat input
+prompt = Chat([
+    {"role": "user", "content": "You are a helpful assistant that helps me described pictures."},
+    {"role": "assistant", "content": "I'd be happy to help you describe pictures! Please go ahead and share an image"},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Streaming
+
+Finally, the `Anthropic` model supports streaming through the `stream` method.
+
+For instance:
+
+```python
+from anthropic import Anthropic
+import outlines
+
+# Create the model
+model = outlines.from_anthropic(
+    Anthropic(),
+    "claude-3-5-sonnet-latest"
+)
+
 # Stream the response
-for chunk in model.stream(prompt, max_tokens=50):
-    print(chunk) # 'This...'
+for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
+    print(chunk) # 'Once...'
 ```
 
 ## Inference arguments
diff --git a/outlines/models/anthropic.py b/outlines/models/anthropic.py
index 1d2bd0824..4b5823952 100644
--- a/outlines/models/anthropic.py
+++ b/outlines/models/anthropic.py
@@ -3,7 +3,7 @@
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, Iterator, Optional, Union
 
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.base import Model, ModelTypeAdapter
 
 if TYPE_CHECKING:
@@ -40,52 +40,79 @@ def format_input(self, model_input):
         """
         raise TypeError(
             f"The input type {type(model_input)} is not available with "
-            "Anthropic. The only available types are `str` and `list` "
+            "Anthropic. The only available types are `str`, `list` and `Chat` "
             "(containing a prompt and images)."
         )
 
     @format_input.register(str)
     def format_str_model_input(self, model_input: str) -> dict:
         return {
-            "messages": [
-                {
-                    "role": "user",
-                    "content": model_input,
-                }
-            ]
+            "messages": [self._create_message("user", model_input)]
         }
 
     @format_input.register(list)
     def format_list_model_input(self, model_input: list) -> dict:
-        prompt = model_input[0]
-        images = model_input[1:]
-
-        if not all(isinstance(image, Image) for image in images):
-            raise ValueError("All assets provided must be of type Image")
-
-        image_content_messages = [
-            {
-                "type": "image",
-                "source": {
-                    "type": "base64",
-                    "media_type": image.image_format,
-                    "data": image.image_str,
-                },
-            }
-            for image in images
-        ]
+        return {
+            "messages": [
+                self._create_message("user", model_input)
+            ]
+        }
+
+    @format_input.register(Chat)
+    def format_chat_model_input(self, model_input: Chat) -> dict:
+        """Generate the `messages` argument to pass to the client when the user
+        passes a Chat instance.
 
+        """
         return {
             "messages": [
+                self._create_message(message["role"], message["content"])
+                for message in model_input.messages
+            ]
+        }
+
+    def _create_message(self, role: str, content: str | list) -> dict:
+        """Create a message."""
+
+        if isinstance(content, str):
+            return {
+                "role": role,
+                "content": content,
+            }
+
+        elif isinstance(content, list):
+            prompt = content[0]
+            images = content[1:]
+
+            if not all(isinstance(image, Image) for image in images):
+                raise ValueError("All assets provided must be of type Image")
+
+            image_content_messages = [
                 {
-                    "role": "user",
-                    "content": [
-                        *image_content_messages,
-                        {"type": "text", "text": prompt},
-                    ],
+                    "type": "image",
+                    "source": {
+                        "type": "base64",
+                        "media_type": image.image_format,
+                        "data": image.image_str,
+                    },
                 }
+                for image in images
             ]
-        }
+
+            return {
+                "role": role,
+                "content": [
+                    *image_content_messages,
+                    {"type": "text", "text": prompt},
+                ],
+            }
+
+        else:
+            raise ValueError(
+                f"Invalid content type: {type(content)}. "
+                "The content must be a string or a list containing a string "
+                "and a list of images."
+            )
 
     def format_output_type(self, output_type):
         """Not implemented for Anthropic."""
@@ -123,7 +150,7 @@ def __init__(
 
     def generate(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> str:
@@ -177,7 +204,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Iterator[str]:
diff --git a/tests/models/test_anthopic_type_adapter.py b/tests/models/test_anthopic_type_adapter.py
new file mode 100644
index 000000000..0be6e0ebf
--- /dev/null
+++ b/tests/models/test_anthopic_type_adapter.py
@@ -0,0 +1,118 @@
+import io
+import pytest
+from dataclasses import dataclass
+
+from PIL import Image as PILImage
+from outlines.inputs import Chat, Image
+from outlines.models.anthropic import AnthropicTypeAdapter
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
+
+    return image
+
+
+@pytest.fixture
+def adapter():
+    return AnthropicTypeAdapter()
+
+
+def test_anthropic_type_adapter_input_text(adapter):
+    message = "prompt"
+    result = adapter.format_input(message)
+    assert result == {"messages": [{"role": "user", "content": message}]}
+
+
+def test_anthropic_type_adapter_input_vision(adapter, image):
+    image_input = Image(image)
+    text_input = "hello"
+    result = adapter.format_input([text_input, image_input])
+    assert result == {
+        "messages": [
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "type": "image",
+                        "source": {
+                            "type": "base64",
+                            "media_type": "image/png",
+                            "data": image_input.image_str,
+                        },
+                    },
+                    {"type": "text", "text": text_input},
+                ],
+            },
+        ]
+    }
+
+
+def test_anthropic_type_adapter_input_chat(adapter, image):
+    image_input = Image(image)
+    model_input = Chat(messages=[
+        {"role": "system", "content": "prompt"},
+        {"role": "user", "content": [
+            "hello",
+            image_input,
+        ]},
+        {"role": "assistant", "content": "response"},
+    ])
+    result = adapter.format_input(model_input)
+    assert result == {
+        "messages": [
+            {"role": "system", "content": "prompt"},
+            {"role": "user", "content": [
+                {
+                    "type": "image",
+                    "source": {
+                        "type": "base64",
+                        "media_type": "image/png",
+                        "data": image_input.image_str,
+                    },
+                },
+                {"type": "text", "text": "hello"},
+            ]},
+            {"role": "assistant", "content": "response"},
+        ]
+    }
+
+
+def test_anthropic_type_adapter_input_invalid(adapter):
+    @dataclass
+    class Audio:
+        file: str
+
+    with pytest.raises(TypeError, match="is not available with Anthropic"):
+        _ = adapter.format_input(Audio("file"))
+
+    with pytest.raises(
+        ValueError,
+        match="All assets provided must be of type Image",
+    ):
+        _ = adapter.format_input(["prompt", Audio("file")])
+
+    with pytest.raises(
+        ValueError,
+        match="The content must be a string or a list",
+    ):
+        _ = adapter.format_input(
+            Chat(messages=[{"role": "user", "content": {"foo": "bar"}}])
+        )
+
+
+def test_anthropic_type_adapter_output(adapter):
+    with pytest.raises(
+        NotImplementedError,
+        match="is not available with Anthropic"
+    ):
+        adapter.format_output_type(str)
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
index 49485edd7..fe9b976cc 100644
--- a/tests/models/test_anthropic.py
+++ b/tests/models/test_anthropic.py
@@ -1,3 +1,4 @@
+import io
 from typing import Generator
 
 from anthropic import Anthropic as AnthropicClient
@@ -5,7 +6,7 @@
 import pytest
 
 import outlines
-from outlines.inputs import Image, Video
+from outlines.inputs import Chat, Image, Video
 from outlines.models.anthropic import Anthropic
 
 
@@ -27,7 +28,12 @@ def image():
     width, height = 1, 1
     white_background = (255, 255, 255)
     image = PILImage.new("RGB", (width, height), white_background)
-    image.format = "PNG"
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
 
     return image
 
@@ -106,6 +112,18 @@ def test_anthropic_simple_vision(model, image):
     assert isinstance(result, str)
 
 
+@pytest.mark.api_call
+def test_anthropic_chat(model, image):
+    result = model.generate(Chat(messages=[
+        {"role": "assistant", "content": "How can I help you today?"},
+        {
+            "role": "user",
+            "content": ["What does this logo represent?", Image(image)]
+        },
+    ]), max_tokens=10)
+    assert isinstance(result, str)
+
+
 @pytest.mark.api_call
 def test_anthopic_streaming(model):
     result = model.stream("Respond with one word. Not more.", max_tokens=1024)

From 1b7d51f0542ec4f49c77b7565a7f642649df436a Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 9 Jul 2025 18:37:50 +0200
Subject: [PATCH 393/505] Add the Chat input to Gemini

---
 docs/features/models/gemini.md           | 96 ++++++++++++++++++++++--
 outlines/models/gemini.py                | 90 ++++++++++++++++++----
 tests/models/test_gemini.py              | 14 +++-
 tests/models/test_gemini_type_adapter.py | 61 ++++++++++++---
 4 files changed, 229 insertions(+), 32 deletions(-)

diff --git a/docs/features/models/gemini.md b/docs/features/models/gemini.md
index e6daccf26..ee48041aa 100644
--- a/docs/features/models/gemini.md
+++ b/docs/features/models/gemini.md
@@ -39,24 +39,110 @@ For instance:
 
 ```python
 import outlines
-from google import genai
+from google.genai import Client
 
 # Create the model
-model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
+model = outlines.from_gemini(
+    Client(),
+    "gemini-1.5-flash-latest"
+)
 
 # Call it to generate text
 result = model("What's the capital of Latvia?", max_output_tokens=20)
 print(result) # 'Riga'
 ```
 
-Gemini also supports streaming. For instance:
+#### Vision
+
+Some Gemini models support vision input. To use this feature, provide a list containing a text prompt and `Image` instances.
+
+For instance:
 
 ```python
+import io
+import requests
+import PIL
 import outlines
-from google import genai
+from google.genai import Client
+from outlines.inputs import Image
 
 # Create the model
-model = outlines.from_gemini(genai.Client(), "gemini-1.5-flash-latest")
+model = outlines.from_gemini(
+    Client(),
+    "gemini-1.5-flash-latest"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt containing the text and the image
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
+
+# Call the model to generate a response
+response = model(prompt, max_output_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Chat
+
+You can also use chat inputs with the `Gemini` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+import outlines
+from google.genai import Client
+from outlines.inputs import Chat, Image
+
+# Create the model
+model = outlines.from_gemini(
+    Client(),
+    "gemini-1.5-flash-latest"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the chat input
+prompt = Chat([
+    {"role": "user", "content": "You are a helpful assistant that helps me described pictures."},
+    {"role": "assistant", "content": "I'd be happy to help you describe pictures! Please go ahead and share an image"},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
+
+# Call the model to generate a response
+response = model(prompt, max_output_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Streaming
+
+Finally, the `Gemini` model supports streaming through the `stream` method.
+
+For instance:
+
+```python
+import outlines
+from google.genai import Client
+
+# Create the model
+model = outlines.from_gemini(
+    Client(),
+    "gemini-1.5-flash-latest"
+)
 
 # Stream text
 for chunk in model.stream("Write a short story about a cat.", max_output_tokens=20):
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index d62b3920d..742dfa7a1 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -10,7 +10,7 @@
     get_args,
 )
 
-from outlines.inputs import Image
+from outlines.inputs import Image, Chat
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.types import CFG, Choice, JsonSchema, Regex
 from outlines.types.utils import (
@@ -57,35 +57,93 @@ def format_input(self, model_input):
         """
         raise TypeError(
             f"The input type {type(model_input)} is not available with "
-            "Gemini. The only available types are `str` and `list` "
+            "Gemini. The only available types are `str`, `list` and `Chat` "
             "(containing a prompt and images)."
         )
 
     @format_input.register(str)
     def format_str_model_input(self, model_input: str) -> dict:
-        return {"contents": [model_input]}
+        return {"contents": [self._create_text_part(model_input)]}
 
     @format_input.register(list)
     def format_list_model_input(self, model_input: list) -> dict:
-        from google.genai import types
+        return {
+            "contents": [
+                self._create_message("user", model_input)
+            ]
+        }
+
+    @format_input.register(Chat)
+    def format_chat_model_input(self, model_input: Chat) -> dict:
+        """Generate the `contents` argument to pass to the client when the user
+        passes a Chat instance.
+
+        """
+        return {
+            "contents": [
+                self._create_message(message["role"], message["content"])
+                for message in model_input.messages
+            ]
+        }
+
+    def _create_message(self, role: str, content: str | list) -> dict:
+        """Create a message."""
+
+        # Gemini uses "model" instead of "assistant"
+        if role == "assistant":
+            role = "model"
+
+        if isinstance(content, str):
+            return {
+                "role": role,
+                "parts": [self._create_text_part(content)],
+            }
 
-        prompt = model_input[0]
-        images = model_input[1:]
+        elif isinstance(content, list):
+            prompt = content[0]
+            images = content[1:]
 
-        if not all(isinstance(image, Image) for image in images):
-            raise ValueError("All assets provided must be of type Image")
+            if not all(isinstance(image, Image) for image in images):
+                raise ValueError("All assets provided must be of type Image")
 
-        image_parts = [
-            types.Part.from_bytes(
-                data=image.image_str,
-                mime_type=image.image_format
+            image_parts = [
+                self._create_img_part(image)
+                for image in images
+            ]
+
+            return {
+                "role": role,
+                "parts": [
+                    self._create_text_part(prompt),
+                    *image_parts,
+                ],
+            }
+
+        else:
+            raise ValueError(
+                f"Invalid content type: {type(content)}. "
+                "The content must be a string or a list containing a string "
+                "and a list of images."
             )
-            for image in images
-        ]
 
         return {"contents": [prompt, *image_parts]}
 
 
+    def _create_text_part(self, text: str) -> dict:
+        """Create a text input part for a message."""
+        return {
+            "text": text,
+        }
+
+    def _create_img_part(self, image: Image) -> dict:
+        """Create an image input part for a message."""
+        return {
+            "inline_data": {
+                "mime_type": image.image_format,
+                "data": image.image_str,
+            }
+        }
+
     def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the `generation_config` argument to pass to the client.
 
@@ -224,7 +282,7 @@ def __init__(self, client: "Client", model_name: Optional[str] = None):
 
     def generate(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Any] = None,
         **inference_kwargs,
     ) -> str:
@@ -270,7 +328,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Any] = None,
         **inference_kwargs,
     ) -> Iterator[str]:
diff --git a/tests/models/test_gemini.py b/tests/models/test_gemini.py
index d923278ce..3582afa07 100644
--- a/tests/models/test_gemini.py
+++ b/tests/models/test_gemini.py
@@ -11,7 +11,7 @@
 from pydantic import BaseModel, ValidationError
 
 import outlines
-from outlines.inputs import Image, Video
+from outlines.inputs import Chat, Image, Video
 from outlines.models.gemini import Gemini
 from outlines.types import Choice
 
@@ -98,6 +98,18 @@ def test_gemini_simple_vision(model, image):
     assert isinstance(result, str)
 
 
+@pytest.mark.api_call
+def test_gemini_chat(model, image):
+    result = model.generate(Chat(messages=[
+        {"role": "assistant", "content": "How can I help you today?"},
+        {
+            "role": "user",
+            "content": ["What does this logo represent?", Image(image)]
+        },
+    ]))
+    assert isinstance(result, str)
+
+
 @pytest.mark.api_call
 def test_gemini_simple_pydantic(model):
     class Foo(BaseModel):
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
index a467df750..0f1985991 100644
--- a/tests/models/test_gemini_type_adapter.py
+++ b/tests/models/test_gemini_type_adapter.py
@@ -11,13 +11,13 @@
 from pydantic import BaseModel
 
 from outlines import cfg, json_schema, regex
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.gemini import GeminiTypeAdapter
 
 if sys.version_info >= (3, 12):
-    from typing import TypedDict, is_typeddict
+    from typing import TypedDict
 else:
-    from typing_extensions import TypedDict, is_typeddict
+    from typing_extensions import TypedDict
 
 
 @pytest.fixture
@@ -56,17 +56,58 @@ def adapter():
 def test_gemini_type_adapter_input_text(adapter):
     message = "prompt"
     result = adapter.format_input(message)
-    assert result == {"contents": [message]}
+    assert result == {"contents": [{"text": message}]}
 
 
 def test_gemini_type_adapter_input_vision(adapter, image):
-    input_message = ["hello", Image(image)]
+    image_input = Image(image)
+    text_input = "hello"
+    result = adapter.format_input([text_input, image_input])
+    assert result == {
+        "contents": [
+            {
+                "role": "user",
+                "parts": [
+                    {"text": text_input},
+                    {
+                        "inline_data": {
+                            "mime_type": "image/png",
+                            "data": image_input.image_str,
+                        },
+                    },
+                ],
+            },
+        ]
+    }
+
+
+def test_gemini_type_adapter_input_chat(adapter, image):
+    image_input = Image(image)
+    input_message = Chat(messages=[
+        {"role": "assistant", "content": "How can I help you today?"},
+        {"role": "user", "content": [
+            "What does this logo represent?",
+            image_input,
+        ]},
+    ])
     result = adapter.format_input(input_message)
-    image_part = types.Part.from_bytes(
-        data=input_message[1].image_str,
-        mime_type=input_message[1].image_format
-    )
-    assert result == {"contents": [input_message[0], image_part]}
+    assert result == {
+        "contents": [
+            {"role": "model", "parts": [{"text": "How can I help you today?"}]},
+            {
+                "role": "user",
+                "parts": [
+                    {"text": "What does this logo represent?"},
+                    {
+                        "inline_data": {
+                            "mime_type": "image/png",
+                            "data": image_input.image_str,
+                        },
+                    },
+                ],
+            },
+        ]
+    }
 
 
 def test_gemini_type_adapter_input_invalid(adapter):

From f1a558b51adc262c7527fd905b72fcc8d9ee629a Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 9 Jul 2025 18:38:35 +0200
Subject: [PATCH 394/505] Add the Chat input to Llamacpp

---
 docs/features/models/llamacpp.md           | 36 ++++++++-
 outlines/models/llamacpp.py                | 88 +++++++++++++++-------
 tests/models/test_llamacpp.py              | 37 ++++++++-
 tests/models/test_llamacpp_type_adapter.py | 40 +++++++++-
 4 files changed, 167 insertions(+), 34 deletions(-)

diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
index cf7dd3e0c..b138bf967 100644
--- a/docs/features/models/llamacpp.md
+++ b/docs/features/models/llamacpp.md
@@ -54,7 +54,41 @@ result = model("What's the capital of Latvia?", max_tokens=20)
 print(result) # 'Riga'
 ```
 
-The `LlamaCpp` model also supports streaming. For instance:
+#### Chat
+
+You can also use chat inputs with the `LlamaCpp` model. To do so, call the model with a `Chat` instance.
+
+For instance:
+
+```python
+import outlines
+from llama_cpp import Llama
+from outlines.inputs import Chat
+
+# Create the model
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+# Create the prompt containing the text and the image
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "assistant", "content": "What's the capital of Latvia?"},
+])
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Streaming
+
+The `LlamaCpp` model also supports streaming.
+
+For instance:
 
 ```python
 import outlines
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index f3dc50c8f..84b15da89 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -14,6 +14,7 @@
     Union,
 )
 
+from outlines.inputs import Chat
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
 from outlines.processors import CFGLogitsProcessor, OutlinesLogitsProcessor
@@ -130,8 +131,7 @@ class LlamaCppTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `LlamaCpp` model.
 
     `LlamaCppTypeAdapter` is responsible for preparing the arguments to
-    `llama-cpp-python`'s `Llama.__call__` method: the input (a string prompt),
-    as well as the logits processor (an instance of `LogitsProcessorList`).
+    the `Llama` object text generation methods.
 
     """
 
@@ -151,14 +151,33 @@ def format_input(self, model_input):
 
         """
         raise NotImplementedError(
-            f"The input type {input} is not available with LlamaCpp. "
-            "The only available type is `str`."
+            f"The input type {type(model_input)} is not available with "
+            "LlamaCpp. The only available types are `str` and `Chat`."
         )
 
     @format_input.register(str)
     def format_str_input(self, model_input: str) -> str:
         return model_input
 
+    @format_input.register(Chat)
+    def format_chat_input(self, model_input: Chat) -> list:
+        if not all(
+            isinstance(message["content"], str)
+            for message in model_input.messages
+        ):
+            raise ValueError(
+                "LlamaCpp does not support multi-modal messages."
+                + "The content of each message must be a string."
+            )
+
+        return  [
+            {
+                "role": message["role"],
+                "content": message["content"],
+            }
+            for message in model_input.messages
+        ]
+
     def format_output_type(
         self, output_type: Optional[OutlinesLogitsProcessor] = None,
     ) -> "LogitsProcessorList":
@@ -203,7 +222,7 @@ def __init__(self, model: "Llama"):
 
     def generate(
         self,
-        model_input: str,
+        model_input: Union[Chat, str],
         output_type: Optional[OutlinesLogitsProcessor] = None,
         **inference_kwargs: Any,
     ) -> str:
@@ -232,12 +251,22 @@ def generate(
                 "the llama_cpp tokenizer"
             )
 
-        completion = self.model(
-            self.type_adapter.format_input(model_input),
-            logits_processor=self.type_adapter.format_output_type(output_type),
-            **inference_kwargs,
-        )
-        result = completion["choices"][0]["text"]
+        prompt = self.type_adapter.format_input(model_input)
+
+        if isinstance(prompt, str):
+            completion = self.model(
+                prompt,
+                logits_processor=self.type_adapter.format_output_type(output_type),
+                **inference_kwargs,
+            )
+            result = completion["choices"][0]["text"]
+        elif isinstance(prompt, list): # pragma: no cover
+            completion = self.model.create_chat_completion(
+                prompt,
+                logits_processor=self.type_adapter.format_output_type(output_type),
+                **inference_kwargs,
+            )
+            result = completion["choices"][0]["message"]["content"]
 
         self.model.reset()
 
@@ -253,7 +282,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: str,
+        model_input: Union[Chat, str],
         output_type: Optional[OutlinesLogitsProcessor] = None,
         **inference_kwargs: Any,
     ) -> Iterator[str]:
@@ -282,23 +311,28 @@ def generate_stream(
                 "the llama_cpp tokenizer"
             )
 
-        generator = self.model(
-            self.type_adapter.format_input(model_input),
-            logits_processor=self.type_adapter.format_output_type(output_type),
-            stream=True,
-            **inference_kwargs,
-        )
+        prompt = self.type_adapter.format_input(model_input)
 
-        def token_generator() -> Iterator[str]:
-            while True:
-                try:
-                    result = next(generator)
-                    yield result["choices"][0]["text"]
-                except StopIteration:
-                    self.model.reset()
-                    return
+        if isinstance(prompt, str):
+            generator = self.model(
+                prompt,
+                logits_processor=self.type_adapter.format_output_type(output_type),
+                stream=True,
+                **inference_kwargs,
+            )
+            for chunk in generator:
+                yield chunk["choices"][0]["text"]
+
+        elif isinstance(prompt, list): # pragma: no cover
+            generator = self.model.create_chat_completion(
+                prompt,
+                logits_processor=self.type_adapter.format_output_type(output_type),
+                stream=True,
+                **inference_kwargs,
+            )
+            for chunk in generator:
+                yield chunk["choices"][0]["delta"].get("content", "")
 
-        return token_generator()
 
     def load_lora(self, adapter_path: str) -> None:  # pragma: no cover
         """Load a LoRA adapter. Deprecated since v1.0.0."""
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 4cb1293c5..e93114bae 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -5,6 +5,7 @@
 from llama_cpp import Llama
 from pydantic import BaseModel
 
+from outlines.inputs import Chat
 from outlines.models.llamacpp import (
     LlamaCpp,
     LlamaCppTokenizer,
@@ -45,6 +46,19 @@ def test_llamacpp_simple(model):
     assert isinstance(result, str)
 
 
+def test_llamacpp_chat(model):
+    result = model.generate(
+        Chat(
+            messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "user", "content": "Respond with one word. Not more."}
+            ]
+        ),
+        max_tokens=10
+    )
+    assert isinstance(result, str)
+
+
 def test_llamacpp_regex(model):
     regex_str = Regex(r"[0-9]").pattern
     logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer, model.tensor_library_name)
@@ -56,7 +70,7 @@ def test_llamacpp_json(model):
     class Foo(BaseModel):
         bar: str
 
-    result = model("foo? Respond with one word.", Foo, max_tokens=1000)
+    result = model("foo? Respond with one word.", Foo, max_tokens=100)
     assert isinstance(result, str)
     assert "bar" in json.loads(result)
 
@@ -75,7 +89,7 @@ def test_llamacpp_cfg(model):
         NotImplementedError,
         match="CFG generation is not supported for LlamaCpp"
     ):
-        model("Respond with one word. Not more.", CFG('start: "a"'))
+        model("Respond with one word. Not more.", CFG('start: "a"'), max_tokens=10)
 
 
 def test_llamacpp_text_stop(model):
@@ -90,6 +104,20 @@ def test_llamacpp_stream_simple(model):
         assert isinstance(x, str)
 
 
+def test_llamacpp_stream_chat(model):
+    generator = model.stream(
+        Chat(
+            messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "user", "content": "Respond with one word. Not more."}
+            ]
+        ),
+        max_tokens=10
+    )
+    for x in generator:
+        assert isinstance(x, str)
+
+
 def test_llamacpp_stream_regex(model):
     generator = model.stream("Respond with one word. Not more.", Regex(r"[0-9]"))
 
@@ -112,7 +140,10 @@ def test_llamacpp_stream_cfg(model):
         NotImplementedError,
         match="CFG generation is not supported for LlamaCpp"
     ):
-        model.stream("Respond with one word. Not more.", CFG('start: "a"'))
+        for chunk in model.stream(
+            "Respond with one word. Not more.", CFG('start: "a"')
+        ):
+            assert isinstance(chunk, str)
 
 
 def test_llamacpp_stream_choice(model):
diff --git a/tests/models/test_llamacpp_type_adapter.py b/tests/models/test_llamacpp_type_adapter.py
index e49586d5d..a7649d523 100644
--- a/tests/models/test_llamacpp_type_adapter.py
+++ b/tests/models/test_llamacpp_type_adapter.py
@@ -1,8 +1,11 @@
 import pytest
+import io
 
 import llama_cpp
+from PIL import Image as PILImage
 from llama_cpp import LogitsProcessorList
 
+from outlines.inputs import Chat, Image
 from outlines.models.llamacpp import LlamaCppTypeAdapter, LlamaCppTokenizer
 from outlines.processors.structured import RegexLogitsProcessor
 
@@ -25,14 +28,45 @@ def logits_processor():
     )
 
 
-def test_llamacpp_type_adapter_format_input(adapter):
-    # anything else than a string
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
+
+    return image
+
+
+def test_llamacpp_type_adapter_format_input(adapter, image):
+    # Anything else than a string/Chat
     with pytest.raises(NotImplementedError):
         adapter.format_input(["Hello, world!"])
 
-    # string case
+    # string
     assert adapter.format_input("Hello, world!") == "Hello, world!"
 
+    # Chat
+    messages = [
+        {"role": "user", "content": "Hello, world!"},
+        {"role": "assistant", "content": "Hello, world!"},
+    ]
+    assert adapter.format_input(Chat(messages=messages)) == messages
+
+    # Multi-modal (invalid)
+    with pytest.raises(
+        ValueError,
+        match="LlamaCpp does not support multi-modal messages."
+    ):
+        adapter.format_input(Chat(messages=[
+            {"role": "user", "content": ["prompt", Image(image)]},
+        ]))
+
 
 def test_llamacpp_type_adapter_format_output_type(adapter, logits_processor):
     formatted = adapter.format_output_type(logits_processor)

From 84dab11a0248eea61d9584ef4f1e435e2e86aeec Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 9 Jul 2025 18:38:57 +0200
Subject: [PATCH 395/505] Add the Chat input to Ollama

---
 docs/features/models/ollama.md           |  71 ++++++++++++---
 outlines/models/ollama.py                | 109 +++++++++++++++--------
 tests/models/test_ollama.py              |  35 +++++++-
 tests/models/test_ollama_type_adapter.py |  50 ++++++++---
 4 files changed, 203 insertions(+), 62 deletions(-)

diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
index d610987ed..202927901 100644
--- a/docs/features/models/ollama.md
+++ b/docs/features/models/ollama.md
@@ -65,21 +65,9 @@ response = model("What's the capital of Latvia?")
 print(response) # 'Riga'
 ```
 
-Ollama also supports streaming. For instance:
+#### Vision
 
-```python
-import ollama
-import outlines
-
-# Create the model
-model = outlines.from_ollama(ollama.Client(), "qwen2.5vl:3b")
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat"):
-    print(chunk) # 'In...'
-```
-
-Additionally, you can use `Ollama` with the `Vision` input if you're running a vision model such as qwen2.5vl. For instance:
+Some Ollama models support vision input. To use this feature, provide a list containing a text prompt and `Image` instances.
 
 ```python
 import io
@@ -111,6 +99,61 @@ response = model(prompt)
 print(response) # The image shows a black puppy with a curious and attentive expression.
 ```
 
+#### Chat
+
+You can also use chat inputs with the `Ollama` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+import ollama
+import outlines
+from outlines.inputs import Chat, Image
+
+# Create the model
+model = outlines.from_ollama(
+    ollama.Client(),
+    "qwen2.5vl:3b"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
+
+# Call the model to generate a response
+response = model(prompt)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Streaming
+
+Finally, the `Anthropic` model supports streaming through the `stream` method.
+
+```python
+import ollama
+import outlines
+
+# Create the model
+model = outlines.from_ollama(ollama.Client(), "qwen2.5vl:3b")
+
+# Stream text
+for chunk in model.stream("Write a short story about a cat"):
+    print(chunk) # 'In...'
+```
+
 ## Asynchronous Calls
 
 Ollama supports asynchronous operations by passing an `AsyncClient` instead of a regular `Client`. This returns an `AsyncOllama` model instance that supports async/await patterns.
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 59cd1ed03..24ea77eac 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -6,7 +6,7 @@
 
 from pydantic import TypeAdapter
 
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.types import CFG, JsonSchema, Regex
 from outlines.types.utils import (
@@ -28,7 +28,7 @@ class OllamaTypeAdapter(ModelTypeAdapter):
 
     @singledispatchmethod
     def format_input(self, model_input):
-        """Generate the prompt argument to pass to the model.
+        """Generate the value of the `messages` argument to pass to the client.
 
         Parameters
         ----------
@@ -37,40 +37,75 @@ def format_input(self, model_input):
 
         Returns
         -------
-        dict
-            The model input to be passed to the client.
+        list
+            The formatted value of the `messages` argument to be passed to
+            the client.
 
         """
         raise TypeError(
             f"The input type {type(model_input)} is not available with "
-            "Ollama. The only available types are `str` and `list` "
-            "(containing a prompt and images)."
+            "Ollama. The only available types are `str`, `list` and `Chat`."
         )
 
     @format_input.register(str)
-    def format_str_model_input(self, model_input: str) -> dict:
-        """Generate the `prompt` argument to pass to the client when the user
-        only passes a prompt.
+    def format_str_model_input(self, model_input: str) -> list:
+        """Generate the value of the `messages` argument to pass to the
+        client when the user only passes a prompt.
 
         """
-        return {"prompt": model_input}
+        return [
+            self._create_message("user", model_input)
+        ]
 
     @format_input.register(list)
-    def format_list_model_input(self, model_input: list) -> dict:
-        """Generate the `prompt` and `images` arguments to pass to the client
-        when the user passes a prompt and images.
+    def format_list_model_input(self, model_input: list) -> list:
+        """Generate the value of the `messages` argument to pass to the
+        client when the user passes a prompt and images.
 
         """
-        prompt = model_input[0]
-        images = model_input[1:]
+        return [
+            self._create_message("user", model_input)
+        ]
 
-        if not all(isinstance(image, Image) for image in images):
-            raise ValueError("All assets provided must be of type Image")
+    @format_input.register(Chat)
+    def format_chat_model_input(self, model_input: Chat) -> list:
+        """Generate the value of the `messages` argument to pass to the
+        client when the user passes a Chat instance.
 
-        return {
-            "prompt": prompt,
-            "images": [image.image_str for image in images]
-        }
+        """
+        return [
+            self._create_message(message["role"], message["content"])
+            for message in model_input.messages
+        ]
+
+    def _create_message(self, role: str, content: str | list) -> dict:
+        """Create a message."""
+
+        if isinstance(content, str):
+            return {
+                "role": role,
+                "content": content,
+            }
+
+        elif isinstance(content, list):
+            prompt = content[0]
+            images = content[1:]
+
+            if not all(isinstance(image, Image) for image in images):
+                raise ValueError("All assets provided must be of type Image")
+
+            return {
+                "role": role,
+                "content": prompt,
+                "image": [image.image_str for image in images],
+            }
+
+        else:
+            raise ValueError(
+                f"Invalid content type: {type(content)}. "
+                "The content must be a string or a list containing a string "
+                "and a list of images."
+            )
 
     def format_output_type(
         self, output_type: Optional[Any] = None
@@ -148,7 +183,7 @@ def __init__(self, client: "Client", model_name: Optional[str] = None):
         self.type_adapter = OllamaTypeAdapter()
 
     def generate(self,
-        model_input: str,
+        model_input: Chat | str | list,
         output_type: Optional[Any] = None,
         **kwargs: Any,
     ) -> str:
@@ -174,12 +209,12 @@ def generate(self,
         if "model" not in kwargs and self.model_name is not None:
             kwargs["model"] = self.model_name
 
-        response = self.client.generate(
-            **self.type_adapter.format_input(model_input),
+        response = self.client.chat(
+            messages=self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             **kwargs,
         )
-        return response.response
+        return response.message.content
 
     def generate_batch(
         self,
@@ -193,7 +228,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: str,
+        model_input: Chat | str | list,
         output_type: Optional[Any] = None,
         **kwargs: Any,
     ) -> Iterator[str]:
@@ -219,14 +254,14 @@ def generate_stream(
         if "model" not in kwargs and self.model_name is not None:
             kwargs["model"] = self.model_name
 
-        response = self.client.generate(
-            **self.type_adapter.format_input(model_input),
+        response = self.client.chat(
+            messages=self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             stream=True,
             **kwargs,
         )
         for chunk in response:
-            yield chunk.response
+            yield chunk.message.content
 
 
 class AsyncOllama(AsyncModel):
@@ -254,7 +289,7 @@ def __init__(
         self.type_adapter = OllamaTypeAdapter()
 
     async def generate(self,
-        model_input: str,
+        model_input: Chat | str | list,
         output_type: Optional[Any] = None,
         **kwargs: Any,
     ) -> str:
@@ -280,12 +315,12 @@ async def generate(self,
         if "model" not in kwargs and self.model_name is not None:
             kwargs["model"] = self.model_name
 
-        response = await self.client.generate(
-            **self.type_adapter.format_input(model_input),
+        response = await self.client.chat(
+            messages=self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             **kwargs,
         )
-        return response.response
+        return response.message.content
 
     async def generate_batch(
         self,
@@ -299,7 +334,7 @@ async def generate_batch(
 
     async def generate_stream( # type: ignore
         self,
-        model_input: str,
+        model_input: Chat | str | list,
         output_type: Optional[Any] = None,
         **kwargs: Any,
     ) -> AsyncIterator[str]:
@@ -325,14 +360,14 @@ async def generate_stream( # type: ignore
         if "model" not in kwargs and self.model_name is not None:
             kwargs["model"] = self.model_name
 
-        stream = await self.client.generate(
-            **self.type_adapter.format_input(model_input),
+        stream = await self.client.chat(
+            messages=self.type_adapter.format_input(model_input),
             format=self.type_adapter.format_output_type(output_type),
             stream=True,
             **kwargs,
         )
         async for chunk in stream:
-            yield chunk.response
+            yield chunk.message.content
 
 
 def from_ollama(
diff --git a/tests/models/test_ollama.py b/tests/models/test_ollama.py
index 69df85532..c6e39605b 100644
--- a/tests/models/test_ollama.py
+++ b/tests/models/test_ollama.py
@@ -9,7 +9,7 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.inputs import Image, Video
+from outlines.inputs import Chat, Image, Video
 from outlines.models import AsyncOllama, Ollama
 
 
@@ -104,6 +104,22 @@ def test_ollama_simple_vision(image, model):
     assert isinstance(result, str)
 
 
+def test_ollama_chat(image, model):
+    result = model.generate(
+        Chat(
+            [
+                {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "user", "content": [
+                    "What does this logo represent?",
+                    Image(image)
+                ]},
+            ]
+        ),
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
 def test_ollama_json(model):
     class Foo(BaseModel):
         foo: Annotated[str, Field(max_length=1)]
@@ -206,6 +222,23 @@ async def test_ollama_async_simple_vision(image, async_model):
     assert isinstance(result, str)
 
 
+@pytest.mark.asyncio
+async def test_ollama_async_chat(image, async_model):
+    result = await async_model.generate(
+        Chat(
+            [
+                {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "user", "content": [
+                    "What does this logo represent?",
+                    Image(image)
+                ]},
+            ]
+        ),
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
 @pytest.mark.asyncio
 async def test_ollama_async_json(async_model):
     class Foo(BaseModel):
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
index 47c93ed67..2061831a6 100644
--- a/tests/models/test_ollama_type_adapter.py
+++ b/tests/models/test_ollama_type_adapter.py
@@ -8,7 +8,7 @@
 from PIL import Image as PILImage
 from pydantic import BaseModel
 
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.ollama import OllamaTypeAdapter
 from outlines.types import cfg, json_schema, regex
 
@@ -52,18 +52,42 @@ def adapter():
 
 
 def test_ollama_type_adapter_input_text(adapter):
-    message = "prompt"
-    result = adapter.format_input(message)
-    assert isinstance(result, dict)
-    assert result.get("prompt") == message
+    text_input = "prompt"
+    result = adapter.format_input(text_input)
+    assert isinstance(result, list)
+    assert len(result) == 1
+    assert result[0] == {"role": "user", "content": text_input}
 
 
 def test_ollama_type_adapter_input_vision(adapter, image):
-    prompt = ["prompt", Image(image)]
-    result = adapter.format_input(prompt)
-    assert isinstance(result, dict)
-    assert result.get("prompt") == prompt[0]
-    assert result.get("images") == [prompt[1].image_str]
+    image_input = Image(image)
+    text_input = "prompt"
+    result = adapter.format_input([text_input, image_input])
+    assert isinstance(result, list)
+    assert len(result) == 1
+    assert result[0] == {
+        "role": "user",
+        "content": text_input,
+        "image": [image_input.image_str],
+    }
+
+
+def test_ollama_type_adapter_input_chat(adapter, image):
+    image_input = Image(image)
+    chat_input = Chat(messages=[
+        {"role": "system", "content": "prompt"},
+        {"role": "user", "content": [
+            "hello",
+            image_input,
+        ]},
+        {"role": "assistant", "content": "response"},
+    ])
+    result = adapter.format_input(chat_input)
+    assert isinstance(result, list)
+    assert len(result) == 3
+    assert result[0] == {"role": "system", "content": "prompt"}
+    assert result[1] == {"role": "user", "content": "hello", "image": [image_input.image_str]}
+    assert result[2] == {"role": "assistant", "content": "response"}
 
 
 def test_ollama_type_adapter_input_invalid(adapter):
@@ -71,6 +95,12 @@ def test_ollama_type_adapter_input_invalid(adapter):
     with pytest.raises(TypeError, match="The input type"):
         _ = adapter.format_input(prompt)
 
+    prompt = Chat(messages=[
+        {"role": "user", "content": {"foo": "bar"}},
+    ])
+    with pytest.raises(ValueError, match="Invalid content type"):
+        _ = adapter.format_input(prompt)
+
 
 def test_ollama_type_adapter_output_invalid(adapter):
     with pytest.raises(TypeError, match="The type `str` is not supported"):

From 6516f3cea4d6b674fca7f74d444e12ef7cc4054d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 9 Jul 2025 18:40:06 +0200
Subject: [PATCH 396/505] Add the Chat input to OpenAI

---
 docs/features/models/openai.md           |  72 +++++++++++++--
 outlines/models/openai.py                | 110 ++++++++++++++---------
 tests/models/test_openai.py              |  14 ++-
 tests/models/test_openai_type_adapter.py |  92 +++++++++++++------
 4 files changed, 212 insertions(+), 76 deletions(-)

diff --git a/docs/features/models/openai.md b/docs/features/models/openai.md
index 9d31bba94..a52544a35 100644
--- a/docs/features/models/openai.md
+++ b/docs/features/models/openai.md
@@ -52,7 +52,9 @@ response = model("What's the capital of Latvia?", max_tokens=20)
 print(response) # 'Riga'
 ```
 
-OpenAI supports streaming and vision inputs.
+#### Vision
+
+Some OpenAI models support vision input. To use this feature, provide a list containing a text prompt and `Image` instances.
 
 For instance:
 
@@ -60,8 +62,8 @@ For instance:
 import io
 import requests
 import PIL
-import openai
 import outlines
+import openai
 from outlines.inputs import Image
 
 # Create the model
@@ -75,15 +77,75 @@ def get_image(url):
     r = requests.get(url)
     return PIL.Image.open(io.BytesIO(r.content))
 
-# Create the prompt
+# Create the prompt containing the text and the image
 prompt = [
     "Describe the image",
     Image(get_image("https://picsum.photos/id/237/400/300"))
 ]
 
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Chat
+
+You can also use chat inputs with the `OpenAI` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
+
+For instance:
+
+```python
+import io
+import requests
+import PIL
+import openai
+import outlines
+from outlines.inputs import Chat, Image
+
+# Create the model
+model = outlines.from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
+)
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Streaming
+
+Finally, the `OpenAI` model supports streaming through the `stream` method.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+model = outlines.from_openai(
+    openai.OpenAI(),
+    "gpt-4o"
+)
+
 # Stream the response
-for chunk in model.stream(prompt, max_tokens=50, stop="."):
-    print(chunk) # 'This...'
+for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
+    print(chunk) # 'Once...'
 ```
 
 ## Structured Generation
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 2b31192dd..70f67d086 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -20,7 +20,7 @@
 from pydantic import BaseModel, TypeAdapter
 
 from outlines.caching import cache
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.utils import set_additional_properties_false_json_schema
 from outlines.types import JsonSchema, Regex, CFG
@@ -65,57 +65,83 @@ def format_input(self, model_input):
         """
         raise TypeError(
             f"The input type {type(model_input)} is not available with "
-            "OpenAI. The only available types are `str` and `list` "
-            "(containing a prompt and images)."
+            "OpenAI. The only available types are `str`, `list` and `Chat`."
         )
 
     @format_input.register(str)
-    def format_str_model_input(self, model_input: str) -> dict:
-        """Generate the `messages` argument to pass to the client when the user
-        only passes a prompt.
+    def format_str_model_input(self, model_input: str) -> list:
+        """Generate the value of the `messages` argument to pass to the
+        client when the user only passes a prompt.
 
         """
-        return {
-            "messages": [
-                {
-                    "role": "user",
-                    "content": model_input,
-                }
-            ]
-        }
+        return [
+            self._create_message("user", model_input)
+        ]
 
     @format_input.register(list)
-    def format_list_model_input(self, model_input: list) -> dict:
-        """Generate the `messages` argument to pass to the client when the user
-        passes a prompt and images.
+    def format_list_model_input(self, model_input: list) -> list:
+        """Generate the value of the `messages` argument to pass to the
+        client when the user passes a prompt and images.
 
         """
-        prompt = model_input[0]
-        images = model_input[1:]
+        return [
+            self._create_message("user", model_input)
+        ]
 
-        if not all(isinstance(image, Image) for image in images):
-            raise ValueError("All assets provided must be of type Image")
+    @format_input.register(Chat)
+    def format_chat_model_input(self, model_input: Chat) -> list:
+        """Generate the value of the `messages` argument to pass to the
+        client when the user passes a Chat instance.
 
-        image_parts = [
-            {
-                "type": "image_url",
-                "image_url": {
-                    "url": f"data:{image.image_format};base64,{image.image_str}"  # noqa: E702
-                },
-            }
-            for image in images
+        """
+        return [
+            self._create_message(message["role"], message["content"])
+            for message in model_input.messages
         ]
 
-        return {
-            "messages": [
-                {
-                    "role": "user",
-                    "content": [
-                        {"type": "text", "text": prompt},
-                        *image_parts,
-                    ],
-                }
+    def _create_message(self, role: str, content: str | list) -> dict:
+        """Create a message."""
+
+        if isinstance(content, str):
+            return {
+                "role": role,
+                "content": content,
+            }
+
+        elif isinstance(content, list):
+            prompt = content[0]
+            images = content[1:]
+
+            if not all(isinstance(image, Image) for image in images):
+                raise ValueError("All assets provided must be of type Image")
+
+            image_parts = [
+                self._create_img_content(image)
+                for image in images
             ]
+
+            return {
+                "role": role,
+                "content": [
+                    {"type": "text", "text": prompt},
+                    *image_parts,
+                ],
+            }
+
+        else:
+            raise ValueError(
+                f"Invalid content type: {type(content)}. "
+                "The content must be a string or a list containing a string "
+                "and a list of images."
+            )
+
+    def _create_img_content(self, image: Image) -> dict:
+        """Create the content for an image input."""
+        return {
+            "type": "image_url",
+            "image_url": {
+                "url": f"data:{image.image_format};base64,{image.image_str}"  # noqa: E702
+            },
         }
 
     def format_output_type(self, output_type: Optional[Any] = None) -> dict:
@@ -270,7 +296,7 @@ def __init__(
 
     def generate(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Union[type[BaseModel], str]] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -303,7 +329,7 @@ def generate(
 
         try:
             result = self.client.chat.completions.create(
-                **messages,
+                messages=messages,
                 **response_format,
                 **inference_kwargs,
             )
@@ -340,7 +366,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Union[type[BaseModel], str]] = None,
         **inference_kwargs,
     ) -> Iterator[str]:
@@ -373,7 +399,7 @@ def generate_stream(
 
         stream = self.client.chat.completions.create(
             stream=True,
-            **messages,
+            messages=messages,
             **response_format,
             **inference_kwargs
         )
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 38b77eaa8..2f9ba4c03 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -9,7 +9,7 @@
 from pydantic import BaseModel, Field
 
 import outlines
-from outlines.inputs import Image, Video
+from outlines.inputs import Chat, Image, Video
 from outlines.models.openai import OpenAI
 from outlines.types import json_schema
 
@@ -127,6 +127,18 @@ def test_openai_simple_vision(image, model):
     assert isinstance(result, str)
 
 
+@pytest.mark.api_call
+def test_openai_chat(image, model):
+    result = model.generate(Chat(messages=[
+        {"role": "system", "content": "You are a helpful assistant."},
+        {
+            "role": "user",
+            "content": ["What does this logo represent?", Image(image)]
+        },
+    ]), max_tokens=10)
+    assert isinstance(result, str)
+
+
 @pytest.mark.api_call
 def test_openai_simple_pydantic(model):
     class Foo(BaseModel):
diff --git a/tests/models/test_openai_type_adapter.py b/tests/models/test_openai_type_adapter.py
index 295ad08c3..5f523b8ea 100644
--- a/tests/models/test_openai_type_adapter.py
+++ b/tests/models/test_openai_type_adapter.py
@@ -10,7 +10,7 @@
 from pydantic import BaseModel
 
 from outlines import cfg, json_schema, regex
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.openai import OpenAITypeAdapter
 
 if sys.version_info >= (3, 12):
@@ -56,31 +56,56 @@ def adapter():
 def test_openai_type_adapter_input_text(adapter):
     message = "prompt"
     result = adapter.format_input(message)
-    assert isinstance(result, dict)
-    assert len(result["messages"]) == 1
-    assert result["messages"][0]["role"] == "user"
-    assert result["messages"][0]["content"] == message
+    assert result == [{"role": "user", "content": message}]
 
 
 def test_openai_type_adapter_input_vision(adapter, image):
-    input_message = ["hello", Image(image)]
-    result = adapter.format_input(input_message)
-    assert isinstance(result, dict)
-
-    messages = result["messages"]
-    assert len(messages) == 1
-
-    message = messages[0]
-    assert message["role"] == "user"
-    assert len(message["content"]) == 2
-    assert message["content"][0]["type"] == "text"
-    assert message["content"][0]["text"] == "hello"
-
-    assert message["content"][1]["type"] == "image_url"
-    assert (
-        message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message[1].image_str}"
-    )
+    image_input = Image(image)
+    text_input = "hello"
+    result = adapter.format_input([text_input, image_input])
+    assert result == [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": text_input},
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/png;base64,{image_input.image_str}"
+                    },
+                },
+            ],
+        },
+    ]
+
+
+def test_openai_type_adapter_input_chat(adapter, image):
+    image_input = Image(image)
+    model_input = Chat(messages=[
+        {"role": "system", "content": "prompt"},
+        {"role": "user", "content": [
+            "hello",
+            image_input,
+        ]},
+        {"role": "assistant", "content": "response"},
+    ])
+    result = adapter.format_input(model_input)
+    assert result == [
+        {"role": "system", "content": "prompt"},
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "hello"},
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/png;base64,{image_input.image_str}"
+                    },
+                },
+            ]
+        },
+        {"role": "assistant", "content": "response"},
+    ]
 
 
 def test_openai_type_adapter_input_invalid(adapter):
@@ -88,11 +113,22 @@ def test_openai_type_adapter_input_invalid(adapter):
     class Audio:
         file: str
 
-    prompt = Audio(
-        "file",
-    )
-    with pytest.raises(TypeError, match="The input type"):
-        _ = adapter.format_input(prompt)
+    with pytest.raises(TypeError, match="is not available"):
+        _ = adapter.format_input(Audio("file"))
+
+    with pytest.raises(
+        ValueError,
+        match="All assets provided must be of type Image",
+    ):
+        _ = adapter.format_input(["prompt", Audio("file")])
+
+    with pytest.raises(
+        ValueError,
+        match="The content must be a string or a list",
+    ):
+        _ = adapter.format_input(
+            Chat(messages=[{"role": "user", "content": {"foo": "bar"}}])
+        )
 
 
 def test_openai_type_adapter_output_invalid(adapter):

From c104baeaa5256b26ff953e349790dfcf7b57ad6c Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 9 Jul 2025 18:40:18 +0200
Subject: [PATCH 397/505] Add the Chat input to SGLang

---
 docs/features/models/sglang.md           | 125 ++++++++++++-----------
 outlines/models/sglang.py                |  23 +++--
 tests/models/test_sglang.py              |  99 ++++++++++++++++++
 tests/models/test_sglang_type_adapter.py |  71 ++++++++-----
 4 files changed, 222 insertions(+), 96 deletions(-)

diff --git a/docs/features/models/sglang.md b/docs/features/models/sglang.md
index 7087be9e2..428b6f69c 100644
--- a/docs/features/models/sglang.md
+++ b/docs/features/models/sglang.md
@@ -59,95 +59,98 @@ import openai
 import outlines
 
 # Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
+model = outlines.from_openai(openai.OpenAI(base_url="http://localhost:11434"))
 
 # Call it to generate text
-result = model("Write a short story about a cat.", max_tokens=100)
-print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
 ```
 
-The `SGLang` model also supports streaming. For instance:
+#### Vision
+
+Some models you can run with SGLang support vision input. To use this feature, provide a list containing a text prompt and `Image` instances.
+
+For instance:
 
 ```python
-import openai
+import io
+import requests
+import PIL
 import outlines
+import openai
+from outlines.inputs import Image
 
 # Create the model
-openai_client = openai.OpenAI(base_url="http://localhost:11434")
-model = outlines.from_sglang(openai_client)
-
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk) # 'In ...'
+model = outlines.from_openai(openai.OpenAI(base_url="http://localhost:11434"))
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt containing the text and the image
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
 ```
 
-## Asynchronous Calls
+#### Chat
 
-SGLang supports asynchronous operations by passing an `AsyncOpenAI` client instead of a regular `OpenAI` client. This returns an `AsyncSGLang` model instance that supports async/await patterns.
+You can also use chat inputs with the `SGLang` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
 
-### Basic Async Generation
+For instance:
 
 ```python
-import asyncio
+import io
+import requests
+import PIL
 import openai
 import outlines
+from outlines.inputs import Chat, Image
 
-async def generate_text():
-    async_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
-    async_model = outlines.from_sglang(async_client)
-
-    result = await async_model("Write a haiku about Python.", max_tokens=50)
-    print(result)
-
-asyncio.run(generate_text())
+# Create the model
+model = outlines.from_openai(openai.OpenAI(base_url="http://localhost:11434"))
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
 ```
 
-### Async Streaming
+#### Streaming
 
-The async model also supports streaming with async iteration:
+Finally, the `SGLang` model supports streaming through the `stream` method.
 
-```python
-import asyncio
-import openai
-import outlines
-
-async def stream_text():
-    async_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
-    async_model = outlines.from_sglang(async_client)
-
-    async for chunk in async_model.stream("Tell me a story about a robot.", max_tokens=100):
-        print(chunk, end="")
-
-asyncio.run(stream_text())
-```
-
-### Concurrent Async Requests
-
-One of the main benefits of async calls is the ability to make multiple concurrent requests:
+For instance:
 
 ```python
-import asyncio
 import openai
 import outlines
 
-async def generate_multiple():
-    async_client = openai.AsyncOpenAI(base_url="http://localhost:11434")
-    async_model = outlines.from_sglang(async_client)
-
-    prompts = [
-        "Write a tagline for a coffee shop.",
-        "Write a tagline for a bookstore.",
-        "Write a tagline for a gym."
-    ]
-
-    tasks = [async_model(prompt, max_tokens=30) for prompt in prompts]
-    results = await asyncio.gather(*tasks)
-
-    for prompt, result in zip(prompts, results):
-        print(f"{prompt}\n{result}\n")
+# Create the model
+model = outlines.from_openai(openai.OpenAI(base_url="http://localhost:11434"))
 
-asyncio.run(generate_multiple())
+# Stream the response
+for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
+    print(chunk) # 'Once...'
 ```
 
 ## Structured Generation
diff --git a/outlines/models/sglang.py b/outlines/models/sglang.py
index 2438214b1..2a8366d79 100644
--- a/outlines/models/sglang.py
+++ b/outlines/models/sglang.py
@@ -6,6 +6,7 @@
     TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
 )
 
+from outlines.inputs import Chat
 from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.models.openai import OpenAITypeAdapter
 from outlines.types.dsl import (
@@ -24,8 +25,8 @@
 class SGLangTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `SGLang` and `AsyncSGLang` models."""
 
-    def format_input(self, model_input: Union[str, list]) -> dict:
-        """Generate the prompt argument to pass to the client.
+    def format_input(self, model_input: Union[Chat, list, str]) -> list:
+        """Generate the value of the messages argument to pass to the client.
 
         We rely on the OpenAITypeAdapter to format the input as the sglang
         server expects input in the same format as OpenAI.
@@ -37,7 +38,7 @@ def format_input(self, model_input: Union[str, list]) -> dict:
 
         Returns
         -------
-        dict
+        list
             The formatted input to be passed to the client.
 
         """
@@ -104,7 +105,7 @@ def __init__(self, client, model_name: Optional[str] = None):
 
     def generate(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -160,7 +161,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, list, str],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Iterator[str]:
@@ -197,7 +198,7 @@ def generate_stream(
 
     def _build_client_args(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
@@ -210,7 +211,7 @@ def _build_client_args(
             inference_kwargs["model"] = self.model_name
 
         client_args = {
-            **messages,
+            "messages": messages,
             **inference_kwargs,
         }
 
@@ -248,7 +249,7 @@ def __init__(self, client, model_name: Optional[str] = None):
 
     async def generate(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -302,7 +303,7 @@ async def generate_batch(
 
     async def generate_stream( # type: ignore
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> AsyncIterator[str]:
@@ -340,7 +341,7 @@ async def generate_stream( # type: ignore
 
     def _build_client_args(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
@@ -353,7 +354,7 @@ def _build_client_args(
             inference_kwargs["model"] = self.model_name
 
         client_args = {
-            **messages,
+            "messages": messages,
             **inference_kwargs,
         }
 
diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
index 8099bc155..5c8d9be1e 100644
--- a/tests/models/test_sglang.py
+++ b/tests/models/test_sglang.py
@@ -3,14 +3,17 @@
 # The outlines backend does not support the EBNF grammar. The xgrammar
 # backend is slow and buggy.
 
+import io
 import os
 import re
 import warnings
 from typing import AsyncGenerator, Generator
 
 import pytest
+from PIL import Image as PILImage
 from openai import AsyncOpenAI, OpenAI
 
+from outlines.inputs import Chat, Image
 from outlines.models.sglang import SGLang, AsyncSGLang, from_sglang
 from outlines.types.dsl import CFG, Regex, JsonSchema
 from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
@@ -21,6 +24,16 @@
 answer ::= "yes" | "no"
 """
 
+# Image for testing
+width, height = 1, 1
+white_background = (255, 255, 255)
+image = PILImage.new("RGB", (width, height), white_background)
+buffer = io.BytesIO()
+image.save(buffer, format="PNG")
+buffer.seek(0)
+image = PILImage.open(buffer)
+image_input = Image(image)
+
 
 # If the SGLANG_SERVER_URL environment variable is set, use the real SGLang server
 # Otherwise, use the mock server
@@ -108,6 +121,50 @@
         },
         "yes"
     ),
+    (
+        {
+            'messages': [
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": "hello"},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:image/png;base64,{image_input.image_str}"
+                            },
+                        },
+                    ]
+                }
+            ],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+        },
+        "foo"
+    ),
+    (
+        {
+            'messages': [
+                {"role": "system", "content": "prompt"},
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": "hello"},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:image/png;base64,{image_input.image_str}"
+                            },
+                        },
+                    ],
+                },
+                {"role": "assistant", "content": "response"},
+            ],
+            'model': sglang_model_name,
+            'max_tokens': 10,
+        },
+        "foo"
+    )
 ]
 
 
@@ -193,6 +250,26 @@ def test_sglang_sync_batch(sync_model):
         )
 
 
+def test_sglang_sync_vision(sync_model):
+    result = sync_model(["hello", image_input], max_tokens=10)
+    assert isinstance(result, str)
+
+
+def test_sglang_sync_vision_chat(sync_model):
+    result = sync_model(
+        Chat(messages=[
+            {"role": "system", "content": "prompt"},
+            {"role": "user", "content": [
+                "hello",
+                image_input,
+            ]},
+            {"role": "assistant", "content": "response"},
+        ]),
+        max_tokens=10,
+    )
+    assert isinstance(result, str)
+
+
 def test_sglang_sync_multiple_samples(sync_model):
     result = sync_model("Respond with a single word.", n=2)
     assert isinstance(result, list)
@@ -253,6 +330,28 @@ async def test_sglang_async_batch(async_model):
         )
 
 
+@pytest.mark.asyncio
+async def test_sglang_async_vision(async_model):
+    result = await async_model(["hello", image_input], max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_sglang_async_vision_chat(async_model):
+    result = await async_model(
+        Chat(messages=[
+            {"role": "system", "content": "prompt"},
+            {"role": "user", "content": [
+                "hello",
+                image_input,
+            ]},
+            {"role": "assistant", "content": "response"},
+        ]),
+        max_tokens=10,
+    )
+    assert isinstance(result, str)
+
+
 @pytest.mark.asyncio
 async def test_sglang_async_multiple_samples(async_model):
     result = await async_model("Respond with a single word.", n=2)
diff --git a/tests/models/test_sglang_type_adapter.py b/tests/models/test_sglang_type_adapter.py
index 6cb51f662..f07ccac49 100644
--- a/tests/models/test_sglang_type_adapter.py
+++ b/tests/models/test_sglang_type_adapter.py
@@ -5,7 +5,7 @@
 
 from PIL import Image as PILImage
 
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.sglang import SGLangTypeAdapter
 from outlines.types import CFG, JsonSchema
 
@@ -41,7 +41,6 @@ def json_schema_instance():
 def json_schema_whitespace_instance():
     return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
 
-
 @pytest.fixture
 def image():
     width, height = 1, 1
@@ -60,31 +59,55 @@ def image():
 def test_sglang_type_adapter_input_text(type_adapter):
     message = "prompt"
     result = type_adapter.format_input(message)
-    assert isinstance(result, dict)
-    assert len(result["messages"]) == 1
-    assert result["messages"][0]["role"] == "user"
-    assert result["messages"][0]["content"] == message
+    assert result == [{"role": "user", "content": message}]
 
 
 def test_sglang_type_adapter_input_vision(type_adapter, image):
-    input_message = ["hello", Image(image)]
-    result = type_adapter.format_input(input_message)
-    assert isinstance(result, dict)
-
-    messages = result["messages"]
-    assert len(messages) == 1
-
-    message = messages[0]
-    assert message["role"] == "user"
-    assert len(message["content"]) == 2
-    assert message["content"][0]["type"] == "text"
-    assert message["content"][0]["text"] == "hello"
-
-    assert message["content"][1]["type"] == "image_url"
-    assert (
-        message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message[1].image_str}"
-    )
+    image_input = Image(image)
+    result = type_adapter.format_input(["hello", image_input])
+    assert result == [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "hello"},
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/png;base64,{image_input.image_str}"
+                    },
+                },
+            ]
+        }
+    ]
+
+
+def test_sglang_type_adapter_input_chat(type_adapter, image):
+    image_input = Image(image)
+    model_input = Chat(messages=[
+        {"role": "system", "content": "prompt"},
+        {"role": "user", "content": [
+            "hello",
+            image_input,
+        ]},
+        {"role": "assistant", "content": "response"},
+    ])
+    result = type_adapter.format_input(model_input)
+    assert result == [
+        {"role": "system", "content": "prompt"},
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "hello"},
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/png;base64,{image_input.image_str}"
+                    },
+                },
+            ],
+        },
+        {"role": "assistant", "content": "response"},
+    ]
 
 
 def test_sglang_type_adapter_input_invalid(type_adapter):

From db9e9fd9d3bf8eb9ab009d6045665d4e26edef2c Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Jul 2025 09:59:04 +0200
Subject: [PATCH 398/505] Add the Chat input to Transformers

---
 docs/features/models/transformers.md          | 32 +++++++-
 outlines/models/transformers.py               | 58 +++++++++++---
 tests/models/test_transformers.py             | 29 +++++++
 tests/models/test_transformers_multimodal.py  | 75 +++++++++++++++----
 ...st_transformers_multimodal_type_adapter.py | 25 ++++++-
 .../models/test_transformers_type_adapter.py  | 46 ++++++++++--
 6 files changed, 232 insertions(+), 33 deletions(-)

diff --git a/docs/features/models/transformers.md b/docs/features/models/transformers.md
index 0ffc56a26..77c5d4241 100644
--- a/docs/features/models/transformers.md
+++ b/docs/features/models/transformers.md
@@ -54,7 +54,37 @@ result = model("What's the capital of Latvia?", max_new_tokens=20)
 print(result) # 'Riga'
 ```
 
-The `Transformers` model also supports batch generation. To use it, provide a list of prompts to the `batch` method. You will receive as a result a list of completions.
+#### Chat
+
+You can also use chat inputs with the `Transformers` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
+
+For instance:
+
+```python
+import outlines
+from outlines.inputs import Chat
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Create the model
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": "What's the capital of Latvia?"},
+])
+
+# Call the model to generate a response
+response = model(prompt, max_new_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Batching
+
+Finally, the `Transformers` model supports batching through the `batch` method. To use it, provide a list of prompts (using the formats described above) to the `batch` method. You will receive as a result a list of completions.
 
 For instance:
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 25e4eb09b..475635054 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -1,4 +1,4 @@
-"""Integration with the `transformers` library."""
+"""Integration with the `transformers` library. """
 
 import warnings
 
@@ -6,7 +6,7 @@
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, List, Optional, Tuple, Union
 
-from outlines.inputs import Audio, Image, Video
+from outlines.inputs import Audio, Chat, Image, Video
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
 from outlines.processors import OutlinesLogitsProcessor
@@ -135,6 +135,9 @@ def __setstate__(self, state):
 class TransformersTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `Transformers` model."""
 
+    def __init__(self, **kwargs):
+        self.tokenizer = kwargs.get("tokenizer")
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
@@ -152,13 +155,20 @@ def format_input(self, model_input):
         """
         raise TypeError(
             f"The input type {type(model_input)} is not available."
-            "The only available type is `str`."
+            "The only available types are `str` and `Chat`."
         )
 
     @format_input.register(str)
     def format_str_input(self, model_input: str) -> str:
         return model_input
 
+    @format_input.register(Chat)
+    def format_chat_input(self, model_input: Chat) -> str:
+        return self.tokenizer.apply_chat_template(
+            model_input.messages,
+            tokenize=False
+        )
+
     def format_output_type(
         self,
         output_type: Optional[OutlinesLogitsProcessor] = None,
@@ -224,7 +234,7 @@ def __init__(
         tokenizer.padding_side = "left"
         self.model = model
         self.tokenizer = TransformerTokenizer(tokenizer)
-        self.type_adapter = TransformersTypeAdapter()
+        self.type_adapter = TransformersTypeAdapter(tokenizer=tokenizer)
 
         if (
             FlaxPreTrainedModel is not None
@@ -241,7 +251,7 @@ def __init__(
 
     def _prepare_model_inputs(
         self,
-        model_input: Union[str, List[str], dict, List[dict]],
+        model_input,
         is_batch: bool = False,
     ) -> Tuple[Union[str, List[str]], dict]:
         """Turn the user input into arguments to pass to the model"""
@@ -263,7 +273,7 @@ def _prepare_model_inputs(
 
     def generate(
         self,
-        model_input: Union[str, dict],
+        model_input: Union[str, dict, Chat],
         output_type: Optional[OutlinesLogitsProcessor] = None,
         **inference_kwargs: Any,
     ) -> Union[str, List[str]]:
@@ -309,7 +319,7 @@ def generate(
 
     def generate_batch(
         self,
-        model_input: List[Union[str, dict]],
+        model_input: List[Union[str, dict, Chat]],
         output_type: Optional[OutlinesLogitsProcessor] = None,
         **inference_kwargs: Any,
     ) -> List[Union[str, List[str]]]:
@@ -374,6 +384,9 @@ def _decode_generation(self, generated_ids: "torch.Tensor"):
 class TransformersMultiModalTypeAdapter(ModelTypeAdapter):
     """Type adapter for `TransformersMultiModal` model."""
 
+    def __init__(self, **kwargs):
+        self.tokenizer = kwargs.get("tokenizer")
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Fomat the prompt arguments to pass to the model.
@@ -393,7 +406,7 @@ def format_input(self, model_input):
             f"The input type {type(model_input)} is not available. Please "
             + "provide a list containing a text prompt and assets "
             + "(`Image`, `Audio` or `Video` instances) supported by your "
-            + "model."
+            + "model or a `Chat` instance."
         )
 
     @format_input.register(dict)
@@ -423,6 +436,29 @@ def format_dict_input(self, model_input: dict) -> dict:
             )
         return model_input
 
+    @format_input.register(Chat)
+    def format_chat_input(self, model_input: Chat) -> dict:
+        # we need to separate the images from the messages
+        # to apply the chat template to the messages without images
+        messages = model_input.messages
+        images = []
+        messages_without_images = []
+        for message in messages:
+            if isinstance(message["content"], list):
+                images.extend(message["content"][1:])
+                messages_without_images.append({
+                    "role": message["role"],
+                    "content": message["content"][0],
+                })
+            else:
+                messages_without_images.append(message)
+        formatted_prompt = self.tokenizer.apply_chat_template(
+            messages_without_images,
+            tokenize=False
+        )
+        # use the formatted prompt and the images to format the input
+        return self.format_list_input([formatted_prompt, *images])
+
     @format_input.register(list)
     def format_list_input(self, model_input: list) -> dict:
         prompt = model_input[0]
@@ -512,11 +548,13 @@ def __init__(self, model: "PreTrainedModel", processor):
 
         super().__init__(model, tokenizer)
 
-        self.type_adapter = TransformersMultiModalTypeAdapter()
+        self.type_adapter = TransformersMultiModalTypeAdapter(
+            tokenizer=tokenizer
+        )
 
     def _prepare_model_inputs(
         self,
-        model_input: Union[str, List[str], dict, List[dict]],
+        model_input,
         is_batch: bool = False,
     ) -> Tuple[Union[str, List[str]], dict]:
         """Turn the user input into arguments to pass to the model"""
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 1ed935dfd..336d9e25d 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -6,6 +6,7 @@
 import transformers
 
 import outlines
+from outlines.inputs import Chat
 from outlines.models.transformers import (
     Transformers,
     TransformerTokenizer,
@@ -85,6 +86,9 @@ def model():
         transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
         transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
     )
+    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
+    model.type_adapter.tokenizer.chat_template = chat_template
+
     return model
 
 
@@ -110,6 +114,16 @@ def test_transformers_call(model, model_bart):
     assert isinstance(result, str)
 
 
+def test_transformers_chat(model):
+    result = model(
+        Chat(messages=[
+            {"role": "system", "content": "You are a helpful assistant."},
+            {"role": "user", "content": "What is the capital of France?"},
+        ])
+    )
+    assert isinstance(result, str)
+
+
 def test_transformers_inference_kwargs(model):
     result = model("Respond with one word. Not more.", max_new_tokens=100)
     assert isinstance(result, str)
@@ -171,6 +185,21 @@ def test_transformers_batch(model):
         assert isinstance(item, list)
         assert len(item) == 2
 
+    result = model.batch(
+        [
+            Chat(messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "user", "content": "What is the capital of France?"},
+            ]),
+            Chat(messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "user", "content": "What is the capital of France?"},
+            ]),
+        ],
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
 
 def test_transformers_multiple_samples_constrained(model):
     class Foo(Enum):
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 7741fbdeb..fb77f7c07 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -14,7 +14,7 @@
 )
 
 import outlines
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.transformers import (
     TransformersMultiModal,
     TransformerTokenizer,
@@ -42,13 +42,17 @@ def img_from_url(url):
 
 @pytest.fixture
 def model():
-    return outlines.from_transformers(
+    model = outlines.from_transformers(
         LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
+    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
+    model.type_adapter.tokenizer.chat_template = chat_template
+
+    return model
 
 
-def test_transformers_vision_instantiate_simple():
+def test_transformers_multimodal_instantiate_simple():
     model = outlines.from_transformers(
         LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
@@ -59,7 +63,7 @@ def test_transformers_vision_instantiate_simple():
     assert model.tensor_library_name == "torch"
 
 
-def test_transformers_vision_simple(model, images):
+def test_transformers_multimodal_simple(model, images):
     result = model.generate(
         ["<image>Describe this image in one sentence:", Image(images[0])],
         None,
@@ -68,7 +72,7 @@ def test_transformers_vision_simple(model, images):
     assert isinstance(result, str)
 
 
-def test_transformers_vision_call(model, images):
+def test_transformers_multimodal_call(model, images):
     result = model(
         ["<image>Describe this image in one sentence:", Image(images[0])],
         max_new_tokens=2,
@@ -76,7 +80,7 @@ def test_transformers_vision_call(model, images):
     assert isinstance(result, str)
 
 
-def test_transformers_vision_wrong_number_images(model, images):
+def test_transformers_multimodal_wrong_number_images(model, images):
     with pytest.raises(ValueError):
         model(
             [
@@ -87,11 +91,28 @@ def test_transformers_vision_wrong_number_images(model, images):
         )
 
 
-def test_transformers_vision_wrong_input_type(model):
+def test_transformers_multimodal_wrong_input_type(model):
     with pytest.raises(TypeError):
         model.generate("invalid input", None)
 
 
+def test_transformers_multimodal_chat(model, images):
+    result = model(
+        Chat(messages=[
+            {"role": "system", "content": "You are a helpful assistant."},
+            {
+                "role": "user",
+                "content": [
+                    "What's on this image?<image>",
+                    Image(images[0]),
+                ],
+            },
+        ]),
+        max_new_tokens=2,
+    )
+    assert isinstance(result, str)
+
+
 def test_transformers_inference_kwargs(model, images):
     result = model(
         ["<image>Describe this image in one sentence:", Image(images[0])],
@@ -123,7 +144,7 @@ def test_transformers_several_images(model, images):
     assert isinstance(result, str)
 
 
-def test_transformers_vision_json(model, images):
+def test_transformers_multimodal_json(model, images):
     class Foo(BaseModel):
         name: str
 
@@ -135,7 +156,7 @@ class Foo(BaseModel):
     assert "name" in result
 
 
-def test_transformers_vision_regex(model, images):
+def test_transformers_multimodal_regex(model, images):
     result = model(
         ["<image>How old is it?", Image(images[0])],
         Regex(r"[0-9]")
@@ -145,7 +166,7 @@ def test_transformers_vision_regex(model, images):
     assert re.match(r"[0-9]", result)
 
 
-def test_transformers_vision_choice(model, images):
+def test_transformers_multimodal_choice(model, images):
     class Foo(Enum):
         cat = "cat"
         dog = "dog"
@@ -159,7 +180,7 @@ class Foo(Enum):
     assert result in ["cat", "dog"]
 
 
-def test_transformers_vision_multiple_samples(model, images):
+def test_transformers_multimodal_multiple_samples(model, images):
     result = model(
         ["<image>Describe this image in one sentence.", Image(images[0])],
         num_return_sequences=2,
@@ -170,7 +191,7 @@ def test_transformers_vision_multiple_samples(model, images):
     assert len(result) == 2
 
 
-def test_transformers_vision_batch(model, images):
+def test_transformers_multimodal_batch(model, images):
     result = model.batch(
         [
             ["<image>Describe this image in one sentence.", Image(images[0])],
@@ -196,8 +217,36 @@ def test_transformers_vision_batch(model, images):
         assert isinstance(item, list)
         assert len(item) == 2
 
+    result = model.batch(
+        [
+            Chat(messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {
+                    "role": "user",
+                    "content": [
+                        "What's on this image?<image>",
+                        Image(images[0]),
+                    ],
+                },
+            ]),
+            Chat(messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {
+                    "role": "user",
+                    "content": [
+                        "What's on this image?<image>",
+                        Image(images[1]),
+                    ],
+                },
+            ]),
+        ],
+        max_new_tokens=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
 
-def test_transformers_vision_deprecated_input_type(model, images):
+def test_transformers_multimodal_deprecated_input_type(model, images):
     with pytest.warns(DeprecationWarning):
         result = model.generate(
             {
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index 84df7f084..58be8060f 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -2,17 +2,25 @@
 
 import transformers
 from PIL import Image as PILImage
-from transformers import LogitsProcessorList
+from transformers import AutoTokenizer, LogitsProcessorList
 
 import outlines
-from outlines.inputs import Image, Video
+from outlines.inputs import Chat, Image, Video
 from outlines.models.transformers import TransformersMultiModalTypeAdapter
 from outlines.processors.structured import RegexLogitsProcessor
 
 
+MODEL_NAME = "erwanf/gpt2-mini"
+
+
 @pytest.fixture
 def adapter():
-    return TransformersMultiModalTypeAdapter()
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+    type_adapter = TransformersMultiModalTypeAdapter(tokenizer=tokenizer)
+    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
+    type_adapter.tokenizer.chat_template = chat_template
+
+    return type_adapter
 
 
 @pytest.fixture
@@ -61,6 +69,17 @@ class MockAsset:
         "images": [image_asset.image],
     }
 
+    chat_prompt = Chat(messages=[
+        {"role": "system", "content": "foo"},
+        {"role": "user", "content": ["bar", image_asset]},
+    ])
+    result = adapter.format_input(chat_prompt)
+    assert isinstance(result, dict)
+    assert isinstance(result["text"], str)
+    assert isinstance(result["images"], list)
+    assert len(result["images"]) == 1
+    assert result["images"][0] == image_asset.image
+
 
 def test_transformers_multimodal_type_adapter_format_output_type(
     adapter, logits_processor
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
index da02573a3..e82a42e8e 100644
--- a/tests/models/test_transformers_type_adapter.py
+++ b/tests/models/test_transformers_type_adapter.py
@@ -1,23 +1,33 @@
+import io
 import pytest
 
 import transformers
+from PIL import Image as PILImage
 from transformers import LogitsProcessorList
 
 import outlines
+from outlines.inputs import Chat, Image
 from outlines.models.transformers import TransformersTypeAdapter
 from outlines.processors.structured import RegexLogitsProcessor
 
 
+MODEL_NAME = "erwanf/gpt2-mini"
+
+
 @pytest.fixture
 def adapter():
-    return TransformersTypeAdapter()
+    tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+    type_adapter = TransformersTypeAdapter(tokenizer=tokenizer)
+    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
+    type_adapter.tokenizer.chat_template = chat_template
 
+    return type_adapter
 
 @pytest.fixture
 def logits_processor():
     model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
+        transformers.AutoTokenizer.from_pretrained(MODEL_NAME),
     )
     return RegexLogitsProcessor(
         regex_string=r"[a-z]+",
@@ -25,13 +35,37 @@ def logits_processor():
         tensor_library_name="torch",
     )
 
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
 
-def test_transformers_type_adapter_format_input(adapter):
-    with pytest.raises(TypeError):
-        adapter.format_input(int)
+    return image
 
+
+
+def test_transformers_type_adapter_format_input(adapter, image):
+    # invalid input
+    with pytest.raises(TypeError, match="is not available."):
+        adapter.format_input(["prompt", Image(image)])
+
+    # string
     assert adapter.format_input("Hello, world!") == "Hello, world!"
 
+    # chat
+    assert isinstance(adapter.format_input(Chat(messages=[
+        {"role": "system", "content": "You are a helpful assistant."},
+        {"role": "user", "content": "Hello, world!"},
+        {"role": "assistant", "content": "Hello, world!"},
+    ])), str)
+
 
 def test_transformers_type_adapter_format_output_type(
     adapter, logits_processor

From df88a3ffc730951466601567aae6c889bb4875bb Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Jul 2025 09:59:26 +0200
Subject: [PATCH 399/505] Add the Chat input to VLLM

---
 docs/features/models/vllm.md           |  86 +++++++++++-----
 outlines/models/vllm.py                |  23 ++---
 tests/models/test_vllm.py              | 132 +++++++++++++++++--------
 tests/models/test_vllm_type_adapter.py |  71 ++++++++-----
 4 files changed, 213 insertions(+), 99 deletions(-)

diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index 64bf91f27..df537402a 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -58,30 +58,51 @@ import openai
 import outlines
 
 # Create the model
-openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-model = outlines.from_vllm(openai_client)
+model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
 
 # Call it to generate text
-result = model("Write a short story about a cat.", max_tokens=100)
-print(result) # 'In a quiet village where the cobblestones hummed softly beneath the morning mist...'
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
 ```
 
-The `VLLM` model also supports streaming. For instance:
+#### Vision
+
+Some models you can run with VLLM support vision input. To use this feature, provide a list containing a text prompt and `Image` instances.
+
+For instance:
 
 ```python
-import openai
+import io
+import requests
+import PIL
 import outlines
+import openai
+from outlines.inputs import Image
 
 # Create the model
-openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-model = outlines.from_vllm(openai_client)
+model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
+
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
+
+# Create the prompt containing the text and the image
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
 
-# Stream text
-for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
-    print(chunk) # 'In ...'
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
 ```
 
-Additionnaly, you can use the VLLM model with vision input if your server is running a multimodal model such as Qwen2.5-VL. For instance:
+#### Chat
+
+You can also use chat inputs with the `VLLM` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
+
+For instance:
 
 ```python
 import io
@@ -89,25 +110,46 @@ import requests
 import PIL
 import openai
 import outlines
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 
 # Create the model
-model = outlines.from_vllm(openai.OpenAI(base_url="http://0.0.0.0:8000/v1"), "Qwen/Qwen2.5-VL-3B-Instruct")
+model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
 
 # Function to get an image
 def get_image(url):
     r = requests.get(url)
     return PIL.Image.open(io.BytesIO(r.content))
 
-# Create the prompt
-prompt = [
-    "Describe the image",
-    Image(get_image("https://picsum.photos/id/237/400/300"))
-]
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
+```
+
+#### Streaming
+
+Finally, the `VLLM` model supports streaming through the `stream` method.
+
+For instance:
+
+```python
+import openai
+import outlines
+
+# Create the model
+model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
 
-# Generate text
-response = model(prompt)
-print(response) # The image shows a black puppy lying on a wooden surface, looking up at the camera with wide eyes.
+# Stream the response
+for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
+    print(chunk) # 'Once...'
 ```
 
 ## Asynchronous Calls
diff --git a/outlines/models/vllm.py b/outlines/models/vllm.py
index 8bfaea482..1284ab336 100644
--- a/outlines/models/vllm.py
+++ b/outlines/models/vllm.py
@@ -3,6 +3,7 @@
 import json
 from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
 
+from outlines.inputs import Chat
 from outlines.models.base import AsyncModel,Model, ModelTypeAdapter
 from outlines.models.openai import OpenAITypeAdapter
 from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
@@ -16,8 +17,8 @@
 class VLLMTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `VLLM` and `AsyncVLLM` models."""
 
-    def format_input(self, model_input: Union[str, list]) -> dict:
-        """Generate the prompt argument to pass to the client.
+    def format_input(self, model_input: Union[Chat, str, list]) -> list:
+        """Generate the value of the messages argument to pass to the client.
 
         We rely on the OpenAITypeAdapter to format the input as the vLLM server
         expects input in the same format as OpenAI.
@@ -29,7 +30,7 @@ def format_input(self, model_input: Union[str, list]) -> dict:
 
         Returns
         -------
-        dict
+        list
             The formatted input to be passed to the model.
 
         """
@@ -91,7 +92,7 @@ def __init__(
 
     def generate(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -145,7 +146,7 @@ def generate_batch(
 
     def generate_stream(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Iterator[str]:
@@ -182,7 +183,7 @@ def generate_stream(
 
     def _build_client_args(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
@@ -196,7 +197,7 @@ def _build_client_args(
             inference_kwargs["model"] = self.model_name
 
         client_args = {
-            **messages,
+            "messages": messages,
             **inference_kwargs,
         }
         if extra_body:
@@ -232,7 +233,7 @@ def __init__(
 
     async def generate(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, list[str]]:
@@ -284,7 +285,7 @@ async def generate_batch(
 
     async def generate_stream( # type: ignore
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> AsyncIterator[str]:
@@ -321,7 +322,7 @@ async def generate_stream( # type: ignore
 
     def _build_client_args(
         self,
-        model_input: Union[str, list],
+        model_input: Union[Chat, str, list],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> dict:
@@ -335,7 +336,7 @@ def _build_client_args(
             inference_kwargs["model"] = self.model_name
 
         client_args = {
-            **messages,
+            "messages": messages,
             **inference_kwargs,
         }
         if extra_body:
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index 74a8425f1..c82e44e21 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -9,7 +9,7 @@
 from PIL import Image as PILImage
 from openai import AsyncOpenAI, OpenAI
 
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.vllm import VLLM, AsyncVLLM, from_vllm
 from outlines.types.dsl import CFG, Regex, JsonSchema
 from tests.test_utils.mock_openai_client import MockOpenAIClient, MockAsyncOpenAIClient
@@ -21,6 +21,16 @@
 answer: "yes" | "no"
 """
 
+# Image for testing
+width, height = 1, 1
+white_background = (255, 255, 255)
+image = PILImage.new("RGB", (width, height), white_background)
+buffer = io.BytesIO()
+image.save(buffer, format="PNG")
+buffer.seek(0)
+image = PILImage.open(buffer)
+image_input = Image(image)
+
 
 # If the VLLM_SERVER_URL environment variable is set, use the real vLLM server
 # Otherwise, use the mock server
@@ -36,16 +46,6 @@
     openai_client = MockOpenAIClient()
     async_openai_client = MockAsyncOpenAIClient()
 
-# Create base64 encoded image for mock responses
-def get_base64_image():
-    width, height = 256, 256
-    white_background = (255, 255, 255)
-    image = PILImage.new("RGB", (width, height), white_background)
-
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    image_str = base64.b64encode(buffer.getvalue()).decode("utf-8")
-    return image_str
 
 mock_responses = [
     (
@@ -67,18 +67,6 @@ def get_base64_image():
         },
         ["foo", "bar"]
     ),
-    (
-        {
-            'messages': [
-                {'role': "user", 'content': [
-                    {'type': 'text', 'text': 'Describe the image.'},
-                    {'type': 'image_url', 'image_url': {'url': 'data:image/png;base64,' + get_base64_image()}}
-                ]}
-            ],
-            'model': vllm_model_name,
-        },
-        "This is a white square."
-    ),
     (
         {
             'messages': [
@@ -127,6 +115,50 @@ def get_base64_image():
         },
         "yes"
     ),
+    (
+        {
+            'messages': [
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": "hello"},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:image/png;base64,{image_input.image_str}"
+                            },
+                        },
+                    ]
+                }
+            ],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+        },
+        "foo"
+    ),
+    (
+        {
+            'messages': [
+                {"role": "system", "content": "prompt"},
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": "hello"},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:image/png;base64,{image_input.image_str}"
+                            },
+                        },
+                    ],
+                },
+                {"role": "assistant", "content": "response"},
+            ],
+            'model': vllm_model_name,
+            'max_tokens': 10,
+        },
+        "foo"
+    )
 ]
 
 
@@ -157,21 +189,6 @@ def async_model_no_model_name():
     return AsyncVLLM(async_openai_client)
 
 
-@pytest.fixture(scope="session")
-def image():
-    width, height = 256, 256
-    white_background = (255, 255, 255)
-    image = PILImage.new("RGB", (width, height), white_background)
-
-    # Save to an in-memory bytes buffer and read as png
-    buffer = io.BytesIO()
-    image.save(buffer, format="PNG")
-    buffer.seek(0)
-    image = PILImage.open(buffer)
-
-    return image
-
-
 def test_vllm_init():
     # We do not rely on the mock server here because we need an object
     # of type OpenAI and AsyncOpenAI to test the init function.
@@ -227,8 +244,23 @@ def test_vllm_sync_batch(sync_model):
         )
 
 
-def test_vllm_sync_vision(sync_model, image):
-    result = sync_model(["Describe the image.", Image(image)])
+def test_vllm_sync_vision(sync_model):
+    result = sync_model(["hello", image_input], max_tokens=10)
+    assert isinstance(result, str)
+
+
+def test_vllm_sync_vision_chat(sync_model):
+    result = sync_model(
+        Chat(messages=[
+            {"role": "system", "content": "prompt"},
+            {"role": "user", "content": [
+                "hello",
+                image_input,
+            ]},
+            {"role": "assistant", "content": "response"},
+        ]),
+        max_tokens=10,
+    )
     assert isinstance(result, str)
 
 
@@ -286,8 +318,24 @@ async def test_vllm_async_batch(async_model):
 
 
 @pytest.mark.asyncio
-async def test_vllm_async_vision(async_model, image):
-    result = await async_model(["Describe the image.", Image(image)])
+async def test_vllm_async_vision(async_model):
+    result = await async_model(["hello", image_input], max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+async def test_vllm_async_vision_chat(async_model):
+    result = await async_model(
+        Chat(messages=[
+            {"role": "system", "content": "prompt"},
+            {"role": "user", "content": [
+                "hello",
+                image_input,
+            ]},
+            {"role": "assistant", "content": "response"},
+        ]),
+        max_tokens=10,
+    )
     assert isinstance(result, str)
 
 
diff --git a/tests/models/test_vllm_type_adapter.py b/tests/models/test_vllm_type_adapter.py
index 526693e4c..8f208830b 100644
--- a/tests/models/test_vllm_type_adapter.py
+++ b/tests/models/test_vllm_type_adapter.py
@@ -5,7 +5,7 @@
 
 from PIL import Image as PILImage
 
-from outlines.inputs import Image
+from outlines.inputs import Chat, Image
 from outlines.models.vllm import VLLMTypeAdapter
 from outlines.types import CFG, JsonSchema
 
@@ -42,7 +42,6 @@ def json_schema_instance():
 def json_schema_whitespace_instance():
     return JsonSchema(JSON_SCHEMA_STRING, whitespace_pattern="\n")
 
-
 @pytest.fixture
 def image():
     width, height = 1, 1
@@ -61,31 +60,55 @@ def image():
 def test_vllm_type_adapter_input_text(type_adapter):
     message = "prompt"
     result = type_adapter.format_input(message)
-    assert isinstance(result, dict)
-    assert len(result["messages"]) == 1
-    assert result["messages"][0]["role"] == "user"
-    assert result["messages"][0]["content"] == message
+    assert result == [{"role": "user", "content": message}]
 
 
 def test_vllm_type_adapter_input_vision(type_adapter, image):
-    input_message = ["hello", Image(image)]
-    result = type_adapter.format_input(input_message)
-    assert isinstance(result, dict)
-
-    messages = result["messages"]
-    assert len(messages) == 1
-
-    message = messages[0]
-    assert message["role"] == "user"
-    assert len(message["content"]) == 2
-    assert message["content"][0]["type"] == "text"
-    assert message["content"][0]["text"] == "hello"
-
-    assert message["content"][1]["type"] == "image_url"
-    assert (
-        message["content"][1]["image_url"]["url"]
-        == f"data:image/png;base64,{input_message[1].image_str}"
-    )
+    image_input = Image(image)
+    result = type_adapter.format_input(["hello", image_input])
+    assert result == [
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "hello"},
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/png;base64,{image_input.image_str}"
+                    },
+                },
+            ],
+        }
+    ]
+
+
+def test_vllm_type_adapter_input_chat(type_adapter, image):
+    image_input = Image(image)
+    model_input = Chat(messages=[
+        {"role": "system", "content": "prompt"},
+        {"role": "user", "content": [
+            "hello",
+            image_input,
+        ]},
+        {"role": "assistant", "content": "response"},
+    ])
+    result = type_adapter.format_input(model_input)
+    assert result == [
+        {"role": "system", "content": "prompt"},
+        {
+            "role": "user",
+            "content": [
+                {"type": "text", "text": "hello"},
+                {
+                    "type": "image_url",
+                    "image_url": {
+                        "url": f"data:image/png;base64,{image_input.image_str}"
+                    },
+                },
+            ],
+        },
+        {"role": "assistant", "content": "response"},
+    ]
 
 
 def test_vllm_type_adapter_input_invalid(type_adapter):

From 8c343ba806930aa9e4846d5cdcc96f6208baa2ba Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Jul 2025 09:59:43 +0200
Subject: [PATCH 400/505] Add the Chat input to VLLMOffline

---
 docs/features/models/vllm_offline.md          | 74 +++++++++++++++++--
 outlines/models/vllm_offline.py               | 68 +++++++++++------
 tests/models/test_vllm_offline.py             | 47 +++++++++++-
 .../models/test_vllm_offline_type_adapter.py  | 55 +++++++++++---
 4 files changed, 199 insertions(+), 45 deletions(-)

diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
index 77db06047..91b454d68 100644
--- a/docs/features/models/vllm_offline.md
+++ b/docs/features/models/vllm_offline.md
@@ -38,13 +38,13 @@ model = outlines.from_vllm_offline(
 
 ## Text Generation
 
-To generate text, you can simply call the model with a prompt.
+Once you've created your Outlines `VLLMOffline` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
 
 For instance:
 
 ```python
 import outlines
-from vllm import LLM, SamplingParams
+from vllm import LLM
 
 # Create the model
 model = outlines.from_vllm_offline(
@@ -52,24 +52,82 @@ model = outlines.from_vllm_offline(
 )
 
 # Call it to generate text
-result = model("What's the capital of Latvia?", sampling_params=SamplingParams(max_tokens=20))
-print(result) # 'Riga'
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
 ```
 
-The `VLLOffline` model also supports batch generation. To use it, provide a list of prompts to the `batch` method. You will receive as a result a list of completions. For instance:
+#### Chat
+
+You can also use chat inputs with the `VLLMOffline` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
+
+For instance:
 
 ```python
 import outlines
-from vllm import LLM, SamplingParams
+from vllm import LLM
+from outlines.inputs import Chat
 
 # Create the model
 model = outlines.from_vllm_offline(
     LLM("microsoft/Phi-3-mini-4k-instruct")
 )
 
+# Create the chat prompt
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": "What's the capital of Latvia?"},
+])
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'Riga'
+```
+
+#### Streaming
+
+The `VLLMOffline` model supports streaming through the `stream` method.
+
+For instance:
+
+```python
+import outlines
+from vllm import LLM
+
+# Create the model
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Stream the response
+for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
+    print(chunk) # 'Once...'
+```
+
+#### Batching
+
+Finally, the `VLLMOffline` model also supports batching through the `batch` method. To use it, provide a list of prompts (using the formats described above) to the `batch` method. You will receive as a result a list of completions.
+
+For instance:
+
+```python
+import outlines
+from vllm import LLM
+
+# Create the model
+model = outlines.from_vllm_offline(
+    LLM("microsoft/Phi-3-mini-4k-instruct")
+)
+
+# Create a list of prompts that will be used in a single batch
+prompts = [
+    "What's the capital of Lithuania?",
+    "What's the capital of Latvia?",
+    "What's the capital of Estonia?"
+]
+
 # Call it to generate text
-result = model.batch(["What's the capital of Latvia?", "What's the capital of Estonia"], sampling_params=SamplingParams(max_tokens=20))
-print(result) # ['Riga', 'Tallinn']
+result = model.batch(prompts, max_new_tokens=20)
+print(result) # ['Vilnius', 'Riga', 'Tallinn']
 ```
 
 ## Structured Generation
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index a51211f7e..0a80a54f7 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -5,7 +5,9 @@
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, List, Optional, Union
 
+from outlines.inputs import Chat, Image
 from outlines.models.base import Model, ModelTypeAdapter
+from outlines.models.openai import OpenAITypeAdapter
 from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
 
 if TYPE_CHECKING:
@@ -28,19 +30,33 @@ def format_input(self, model_input):
             The input passed by the user.
 
         """
-        raise NotImplementedError(
-            f"The input type {input} is not available. "
-            "Please use a string or a list of strings."
+        raise TypeError(
+            f"The input type {type(model_input)} is not available with "
+            "VLLM offline. The only available types are `str` and "
+            "`Chat` (containing a prompt and images)."
         )
 
     @format_input.register(str)
-    def format_str_input(self, model_input: str) -> str:
-        return model_input
+    def format_input_str(self, model_input: str) -> str:
+        """Format a `str` input.
 
-    @format_input.register(list)
-    def format_list_input(self, model_input: List[str]) -> List[str]:
+        """
         return model_input
 
+    @format_input.register(Chat)
+    def format_input_chat(self, model_input: Chat) -> list:
+        """Format a `Chat` input.
+
+        """
+        for message in model_input.messages:
+            content = message["content"]
+            if isinstance(content, list):
+                raise ValueError(
+                    "Assets are not supported for vLLM offline."
+                    "Please only use text content in the `Chat` input."
+                )
+        return OpenAITypeAdapter().format_input(model_input)
+
     def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the structured output argument to pass to the model.
 
@@ -119,7 +135,7 @@ def _build_generation_args(
 
     def generate(
         self,
-        model_input: str,
+        model_input: Chat | str,
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[str, List[str]]:
@@ -147,12 +163,20 @@ def generate(
             output_type,
         )
 
-        results = self.model.generate(
-            self.type_adapter.format_input(model_input),
-            sampling_params=sampling_params,
-            lora_request=self.lora_request, # v0 legacy, to be removed
-            **inference_kwargs,
-        )
+        if isinstance(model_input, Chat):
+            results = self.model.chat(
+                messages=self.type_adapter.format_input(model_input),
+                sampling_params=sampling_params,
+                lora_request=self.lora_request, # v0 legacy, to be removed
+                **inference_kwargs,
+            )
+        else:
+            results = self.model.generate(
+                prompts=self.type_adapter.format_input(model_input),
+                sampling_params=sampling_params,
+                lora_request=self.lora_request, # v0 legacy, to be removed
+                **inference_kwargs,
+            )
         results = [completion.text for completion in results[0].outputs]
 
         if len(results) == 1:
@@ -162,7 +186,7 @@ def generate(
 
     def generate_batch(
         self,
-        model_input: List[str],
+        model_input: List[Chat | str],
         output_type: Optional[Any] = None,
         **inference_kwargs: Any,
     ) -> Union[List[str], List[List[str]]]:
@@ -191,18 +215,18 @@ def generate_batch(
             output_type,
         )
 
+        if any(isinstance(item, Chat) for item in model_input):
+            raise TypeError(
+                "Batch generation is not available for the `Chat` input type."
+            )
+
         results = self.model.generate(
-            self.type_adapter.format_input(model_input),
+            prompts=[self.type_adapter.format_input(item) for item in model_input],
             sampling_params=sampling_params,
             lora_request=self.lora_request, # v0 legacy, to be removed
             **inference_kwargs,
         )
-        results = [[sample.text for sample in batch.outputs] for batch in results]
-
-        if len(results[0]) == 1:
-            return [batch[0] for batch in results]
-        else:
-            return results
+        return [[sample.text for sample in batch.outputs] for batch in results]
 
     def generate_stream(self, model_input, output_type, **inference_kwargs):
         """Not available for `vllm.LLM`.
diff --git a/tests/models/test_vllm_offline.py b/tests/models/test_vllm_offline.py
index 190f13b5c..e3c43bd50 100644
--- a/tests/models/test_vllm_offline.py
+++ b/tests/models/test_vllm_offline.py
@@ -1,7 +1,9 @@
+import io
 import re
 from enum import Enum
 
 import pytest
+from PIL import Image as PILImage
 from pydantic import BaseModel
 
 try:
@@ -11,6 +13,7 @@
     HAS_VLLM = False
 
 import outlines
+from outlines.inputs import Chat
 from outlines.models.vllm_offline import (
     VLLMOffline,
     VLLMOfflineTypeAdapter,
@@ -19,13 +22,27 @@
 from outlines.types import Regex
 
 
-TEST_MODEL = "erwanf/gpt2-mini"
+TEST_MODEL = "microsoft/Phi-3-mini-4k-instruct"
 
 pytestmark = pytest.mark.skipif(
     not HAS_VLLM,
     reason="vLLM models can only be run on GPU."
 )
 
+@pytest.fixture(scope="session")
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
+
+    return image
+
 
 def test_vllm_model_initialization():
     model = from_vllm_offline(LLM(TEST_MODEL))
@@ -51,11 +68,27 @@ def test_vllm_call(model):
 
 
 def test_vllm_inference_kwargs(model):
-    result = model("Write a short story about a cat.", sampling_params=SamplingParams(max_tokens=2), use_tqdm=True)
+    result = model(
+        "Write a short story about a cat.",
+        sampling_params=SamplingParams(max_tokens=2),
+        use_tqdm=True
+    )
     assert isinstance(result, str)
     assert len(result) <= 20
 
 
+def test_vllm_chat(model):
+    result = model(
+        Chat(messages=[
+            {"role": "system", "content": "You are a helpful assistant."},
+            {"role": "user", "content": "What is the capital of France?"},
+            {"role": "assistant", "content": "Response: "},
+        ]),
+        sampling_params=SamplingParams(max_tokens=2),
+    )
+    assert isinstance(result, str)
+
+
 def test_vllm_invalid_inference_kwargs(model):
     with pytest.raises(TypeError):
         model("Respond with one word. Not more.", foo="bar")
@@ -110,10 +143,18 @@ def test_vllm_batch(model):
         assert isinstance(item, list)
         assert len(item) == 2
 
+    with pytest.raises(TypeError, match="Batch generation is not available"):
+        model.batch(
+            [
+                Chat(messages=[
+                    {"role": "user", "content": "What is the capital of France?"},
+                ]),
+            ]
+        )
 
 def test_vllm_streaming(model):
     with pytest.raises(
         NotImplementedError,
-        match="Streaming is not available for the vLLM integration."
+        match="Streaming is not available"
     ):
         model.stream("Respond with one word. Not more.")
diff --git a/tests/models/test_vllm_offline_type_adapter.py b/tests/models/test_vllm_offline_type_adapter.py
index b8a18b6d1..c431230e7 100644
--- a/tests/models/test_vllm_offline_type_adapter.py
+++ b/tests/models/test_vllm_offline_type_adapter.py
@@ -1,9 +1,10 @@
 import io
 import json
-from dataclasses import dataclass
 
 import pytest
+from PIL import Image as PILImage
 
+from outlines.inputs import Chat, Image
 from outlines.models.vllm_offline import VLLMOfflineTypeAdapter
 from outlines.types import CFG, JsonSchema, Regex
 
@@ -43,25 +44,55 @@ def json_schema_whitespace_instance():
 def regex_instance():
     return Regex(r"[0-9]+")
 
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
+
+    return image
 
-def test_vllm_type_adapter_input_text(type_adapter):
-    # string
+
+def test_vllm_offline_type_adapter_input_text(type_adapter):
     message = "prompt"
     result = type_adapter.format_input(message)
-    assert message == result
+    assert result == message
+
+
+def test_vllm_offline_type_adapter_input_chat(type_adapter):
+    model_input = Chat(messages=[
+        {"role": "system", "content": "prompt"},
+        {"role": "user", "content": "hello"},
+        {"role": "assistant", "content": "response"},
+    ])
+    result = type_adapter.format_input(model_input)
+    assert result == [
+        {"role": "system", "content": "prompt"},
+        {"role": "user", "content": "hello"},
+        {"role": "assistant", "content": "response"},
+    ]
 
-    # list of strings
-    messages = ["foo", "bar"]
-    result = type_adapter.format_input(messages)
-    assert messages == result
 
+def test_vllm_offline_type_adapter_input_invalid(type_adapter, image):
+    with pytest.raises(TypeError, match="is not available"):
+        _ = type_adapter.format_input(["Hello", Image(image)])
 
-def test_vllm_type_adapter_input_invalid(type_adapter):
-    with pytest.raises(NotImplementedError):
-        _ = type_adapter.format_input({"foo": "bar"})
+    with pytest.raises(ValueError, match="Assets are not supported"):
+        _ = type_adapter.format_input(Chat(messages=[
+            {"role": "user", "content": [
+                "Hello",
+                Image(image),
+            ]},
+        ]))
 
 
-def test_vllm_type_adapter_output_type(
+def test_vllm_offline_type_adapter_output_type(
     type_adapter,
     cfg_instance,
     json_schema_instance,

From 3ac645a97fd51c02f81758ff3b3536380ecfed2d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Jul 2025 16:58:22 +0200
Subject: [PATCH 401/505] Misc doc fixes

---
 docs/features/advanced/logits_processors.md | 2 ++
 docs/features/core/generator.md             | 4 ++--
 docs/features/core/output_types.md          | 6 +++---
 docs/features/index.md                      | 1 +
 docs/features/models/index.md               | 6 +++---
 docs/features/utility/application.md        | 6 +++---
 6 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/docs/features/advanced/logits_processors.md b/docs/features/advanced/logits_processors.md
index 4e4d9e457..4a05609e2 100644
--- a/docs/features/advanced/logits_processors.md
+++ b/docs/features/advanced/logits_processors.md
@@ -5,7 +5,9 @@ title: Logits Processors
 # Logits Processors
 
 Logits processors are objects that control text generation by modifying the probability distribution of possible next tokens. They do this by adjusting the logits (raw model outputs) at each generation step, effectively biasing the model's token selection.
+
 Processors can be used to:
+
 1. Generate structured output (e.g., JSON that follows a specific schema)
 2. Prevent the model from generating specific words or tokens
 3. Implement custom token sampling strategies
diff --git a/docs/features/core/generator.md b/docs/features/core/generator.md
index 6c44dfdad..e407ed37b 100644
--- a/docs/features/core/generator.md
+++ b/docs/features/core/generator.md
@@ -5,7 +5,7 @@ title: Generator API
 # Generator
 
 
-The `Generator` class is the core component of Outlines v1. `Generator` accepts a [model](../reference/available_models) and an optional [output type](/api/types). If no output type is provided, the `Generator` will return unstructured text.
+The `Generator` class is the core component of Outlines v1. `Generator` accepts a [model](../models/index.md) and an optional [output type](../core/output_types.md). If no output type is provided, the `Generator` will return unstructured text.
 
 !!! note
 
@@ -27,7 +27,7 @@ Generators implement the same methods as models:
 - `batch`
 - `stream`
 
-All of them take a single positional argument: the prompt or list of prompts from which to generate completions. Contrarily to the equivalent methods of models, you do not need to provide an output type as it has already been defined when initializing the generator.
+All of them take a single positional argument: the [model input](../core/inputs.md) from which text is generated. Contrarily to the equivalent methods of models, you do not need to provide an output type as it has already been defined when initializing the generator.
 
 ## Basic Usage
 
diff --git a/docs/features/core/output_types.md b/docs/features/core/output_types.md
index e0eeef20b..85451ae95 100644
--- a/docs/features/core/output_types.md
+++ b/docs/features/core/output_types.md
@@ -108,8 +108,8 @@ from enum import Enum
 from typing import Literal
 
 class PizzaOrBurger(Enum):
-    pizza = "pizza
-    burger = "burger
+    pizza = "pizza"
+    burger = "burger"
 
 # Equivalent multiple-choice output types
 output_type = Literal["pizza", "burger"]
@@ -150,7 +150,7 @@ class Character:
     name: str
     age: int
 
-ouput_type = Character
+output_type = Character
 
 def character(name: str, age: int):
     return None
diff --git a/docs/features/index.md b/docs/features/index.md
index 941103ee2..537b7a050 100644
--- a/docs/features/index.md
+++ b/docs/features/index.md
@@ -5,6 +5,7 @@ This section presents in details the different features of Outlines.
 ## Core Concepts
 
 - [Models](./models/index.md)
+- [Model Inputs](./core/inputs.md)
 - [Output Types](./core/output_types.md)
 - [Generators](./core/generator.md)
 
diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index 525f5c1b7..4a0b46c69 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -35,16 +35,16 @@ model = from_transformers(
 
 # Call it directly
 response = model("How many countries are there in the world", max_new_tokens=20)
-print(result) # 'There are 200 countries in the world.'
+print(response) # 'There are 200 countries in the world.'
 
 # Call it directly with an output_type
 response = model("How many countries are there in the world", int, max_new_tokens=20)
-print(result) # '200'
+print(response) # '200'
 
 # Create a generator first and then call it
 generator = Generator(model, int)
 response = generator("How many countries are there in the world")
-print(result) # '200'
+print(response) # '200'
 ```
 
 Some models support streaming through a `stream` method. It takes the same argument as the `__call__` method, but returns an iterator instead of a string.
diff --git a/docs/features/utility/application.md b/docs/features/utility/application.md
index 4faccb10d..beb3a2d9b 100644
--- a/docs/features/utility/application.md
+++ b/docs/features/utility/application.md
@@ -22,7 +22,7 @@ import transformers
 from outlines import Application, Template, from_transformers
 
 # Create a template
-template_str = "Is {{ name }} a boy or a girl name?"""
+template_str = "Is {{ name }} a boy or a girl name?"
 template = Template.from_string(template_str)
 
 # Create a model
@@ -48,7 +48,7 @@ import transformers
 from outlines import Application, from_transformers
 
 # Create a function that will be used as a template
-def template(name: str) -> str:
+def template_func(name: str) -> str:
     return f"Is {name} a boy or a girl name?"
 
 # Create a model
@@ -58,7 +58,7 @@ model = from_transformers(
 )
 
 # Create the application with the function template and call it to generate text
-application = Application(template, Literal["boy", "girl"])
+application = Application(template_func, Literal["boy", "girl"])
 response = application(model, {"name": "Alice"}, max_new_tokens=10)
 
 print(response) # "girl"

From ffcd6254d6e54081926bbc1c5f190718f65d5465 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Jul 2025 15:30:50 +0200
Subject: [PATCH 402/505] Remove deprecated features from v0

---
 outlines/__init__.py                          |  32 +-
 outlines/models/llamacpp.py                   |  18 -
 outlines/models/openai.py                     | 335 +--------------
 outlines/models/vllm_offline.py               |  31 +-
 outlines/templates.py                         |  98 +----
 outlines/types/__init__.py                    |   3 -
 outlines/types/dsl.py                         | 103 -----
 outlines/v0_legacy/base.py                    | 296 -------------
 outlines/v0_legacy/function.py                | 168 --------
 outlines/v0_legacy/generate/__init__.py       |  17 -
 outlines/v0_legacy/generate/api.py            | 183 --------
 outlines/v0_legacy/generate/cfg.py            |  80 ----
 outlines/v0_legacy/generate/choice.py         | 126 ------
 outlines/v0_legacy/generate/format.py         |  70 ---
 outlines/v0_legacy/generate/fsm.py            |  77 ----
 outlines/v0_legacy/generate/json.py           | 145 -------
 outlines/v0_legacy/generate/regex.py          |  67 ---
 outlines/v0_legacy/generate/text.py           |  64 ---
 outlines/v0_legacy/models/__init__.py         |  21 -
 outlines/v0_legacy/models/exllamav2.py        | 400 ------------------
 outlines/v0_legacy/models/llamacpp.py         | 147 -------
 outlines/v0_legacy/models/mlxlm.py            | 153 -------
 outlines/v0_legacy/models/openai.py           |  64 ---
 outlines/v0_legacy/models/transformers.py     | 196 ---------
 .../v0_legacy/models/transformers_vision.py   | 107 -----
 outlines/v0_legacy/models/vllm_offline.py     | 102 -----
 outlines/v0_legacy/samplers.py                | 118 ------
 27 files changed, 11 insertions(+), 3210 deletions(-)
 delete mode 100644 outlines/v0_legacy/base.py
 delete mode 100644 outlines/v0_legacy/function.py
 delete mode 100644 outlines/v0_legacy/generate/__init__.py
 delete mode 100644 outlines/v0_legacy/generate/api.py
 delete mode 100644 outlines/v0_legacy/generate/cfg.py
 delete mode 100644 outlines/v0_legacy/generate/choice.py
 delete mode 100644 outlines/v0_legacy/generate/format.py
 delete mode 100644 outlines/v0_legacy/generate/fsm.py
 delete mode 100644 outlines/v0_legacy/generate/json.py
 delete mode 100644 outlines/v0_legacy/generate/regex.py
 delete mode 100644 outlines/v0_legacy/generate/text.py
 delete mode 100644 outlines/v0_legacy/models/__init__.py
 delete mode 100644 outlines/v0_legacy/models/exllamav2.py
 delete mode 100644 outlines/v0_legacy/models/llamacpp.py
 delete mode 100644 outlines/v0_legacy/models/mlxlm.py
 delete mode 100644 outlines/v0_legacy/models/openai.py
 delete mode 100644 outlines/v0_legacy/models/transformers.py
 delete mode 100644 outlines/v0_legacy/models/transformers_vision.py
 delete mode 100644 outlines/v0_legacy/models/vllm_offline.py
 delete mode 100644 outlines/v0_legacy/samplers.py

diff --git a/outlines/__init__.py b/outlines/__init__.py
index e588d4a36..eaef0ef71 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -1,8 +1,5 @@
 """Outlines is a Generative Model Programming Framework."""
 
-import sys
-from types import ModuleType
-
 import outlines.grammars
 import outlines.inputs
 import outlines.processors
@@ -25,17 +22,10 @@
     from_vllm,
     from_vllm_offline,
 )
-from outlines.templates import Template, prompt
+from outlines.templates import Template
 from outlines.types import regex, json_schema, cfg
 from outlines.templates import Vision
 
-from .v0_legacy import (
-    generate,
-    samplers,
-    models as legacy_models,
-    function,
-)
-from .v0_legacy.function import Function
 
 model_list = [
     "from_anthropic",
@@ -63,23 +53,3 @@
     "Template",
     "grammars",
 ] + model_list
-
-
-# v0 legacy
-
-generate_module = ModuleType("generate")
-generate_module.__dict__.update(generate.__dict__)
-
-function_module = ModuleType("function")
-function_module.__dict__.update(function.__dict__)
-
-samplers_module = ModuleType("samplers")
-samplers_module.__dict__.update(samplers.__dict__)
-
-outlines.models.__dict__.update(legacy_models.__dict__)
-
-sys.modules["outlines.generate"] = generate_module
-sys.modules["outlines.samplers"] = samplers_module
-sys.modules["outlines.function"] = function_module
-
-__all__ += ["Function"]
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 84b15da89..ce69f96a3 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -334,24 +334,6 @@ def generate_stream(
                 yield chunk["choices"][0]["delta"].get("content", "")
 
 
-    def load_lora(self, adapter_path: str) -> None:  # pragma: no cover
-        """Load a LoRA adapter. Deprecated since v1.0.0."""
-        warnings.warn("""
-            The `load_lora` method is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.1.0.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        if self.model._model.apply_lora_from_file(
-            adapter_path,
-            1.0,
-        ):
-            raise RuntimeError(
-                f"Failed to apply LoRA from lora path: {adapter_path}"
-            )
-
-
 def from_llamacpp(model: "Llama"):
     """Create an Outlines `LlamaCpp` model instance from a
     `llama_cpp.Llama` instance.
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 70f67d086..b17b64209 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,25 +1,17 @@
 """Integration with OpenAI's API."""
 
-import copy
-import functools
 import json
-import warnings
 from typing import (
     TYPE_CHECKING,
     Any,
-    Callable,
-    Dict,
     Iterator,
-    List,
     Optional,
     Union,
 )
-from dataclasses import asdict, dataclass, field, replace
 from functools import singledispatchmethod
 
 from pydantic import BaseModel, TypeAdapter
 
-from outlines.caching import cache
 from outlines.inputs import Chat, Image
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.utils import set_additional_properties_false_json_schema
@@ -31,7 +23,6 @@
     is_genson_schema_builder,
     is_native_dict
 )
-from outlines.v0_legacy.base import vectorize
 
 if TYPE_CHECKING:
     from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
@@ -239,16 +230,9 @@ class OpenAI(Model):
     def __init__(
         self,
         client: Union["OpenAIClient", "AzureOpenAIClient"],
-        model_name: Optional[Union[str, "OpenAIConfig"]] = None,
-        **kwargs
+        model_name: Optional[str] = None,
     ):
-        """Initialize the OpenAI model.
-
-        To provide temporary backwards compatibility with Outlines v0,
-        the class can be instantiated with a `OpenAIConfig` instance as
-        a value for the `model_name` argument. This is deprecated and will
-        be removed in v1.1.0. Please provide a model name instead.
-
+        """
         Parameters
         ----------
         client
@@ -257,42 +241,9 @@ def __init__(
             The name of the model to use.
 
         """
-
-        # legacy mode
-        if isinstance(model_name, OpenAIConfig) or kwargs.get("config"):
-            warnings.warn("""
-                The `openai` function is deprecated starting from v1.0.0.
-                Do not use it. Support for it will be removed in v1.1.0.
-                Instead, you should instantiate a `OpenAI` model with the
-                `outlines.from_openai` function that takes an openai library
-                client and a model name as arguments. Similarly, you cannot
-                instantiate a `OpenAI` model directly with a `OpenAIConfig`
-                instance anymore, but must provide a client and a model name
-                instead.
-                For example:
-                ```python
-                from openai import OpenAI as OpenAIClient
-                from outlines import from_openai
-                client = OpenAIClient()
-                model = from_openai(client, "gpt-4o")
-                ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-            )
-            config = (
-                model_name
-                if isinstance(model_name, OpenAIConfig)
-                else kwargs.pop("config")
-            )
-            self.legacy_instance = OpenAILegacy(
-                client, config, kwargs.get("system_prompt")
-            )
-        # regular mode
-        else:
-            self.client = client
-            self.model_name = model_name
-            self.type_adapter = OpenAITypeAdapter()
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = OpenAITypeAdapter()
 
     def generate(
         self,
@@ -408,36 +359,6 @@ def generate_stream(
             if chunk.choices and chunk.choices[0].delta.content is not None:
                 yield chunk.choices[0].delta.content
 
-    ### Legacy !!!
-
-    def __call__(self, *args, **kwargs):
-        if hasattr(self, "legacy_instance"):
-            return self.legacy_instance(*args, **kwargs)
-        else:
-            return super().__call__(*args, **kwargs)
-
-    def stream(self, *args, **kwargs):
-        if hasattr(self, "legacy_instance"):
-            return self.legacy_instance.stream(*args, **kwargs)
-        else:
-            return super().stream(*args, **kwargs)
-
-    def new_with_replacements(self, **kwargs):
-        if hasattr(self, "legacy_instance"):
-            return self.legacy_instance.new_with_replacements(self, **kwargs)
-        raise NotImplementedError("This method is only available in legacy mode")
-
-    def __str__(self):
-        if hasattr(self, "legacy_instance"):
-            return str(self.legacy_instance)
-        else:
-            return super().__str__()
-
-    def __repr__(self):
-        if hasattr(self, "legacy_instance"):
-            return repr(self.legacy_instance)
-        else:
-            return super().__repr__()
 
 def from_openai(
     client: Union["OpenAIClient", "AzureOpenAIClient"],
@@ -460,249 +381,3 @@ def from_openai(
 
     """
     return OpenAI(client, model_name)
-
-
-### Legacy !!!
-
-
-@dataclass(frozen=True)
-class OpenAIConfig:
-    """Represents the parameters of the OpenAI API.
-
-    The information was last fetched on 2023/11/20. We document below the
-    properties that are specific to the OpenAI API. Not all these properties are
-    supported by Outlines.
-
-    Parameters
-    ----------
-    model
-        The name of the model. Available models can be found on OpenAI's website.
-    frequency_penalty
-        Number between 2.0 and -2.0. Positive values penalize new tokens based on
-        their existing frequency in the text,
-    logit_bias
-        Modifies the likelihood of specified tokens to appear in the completion.
-        Number between -100 (forbid) and +100 (only allows).
-    n
-        The number of completions to return for each prompt.
-    presence_penalty
-        Similar to frequency penalty.
-    response_format
-        Specifies the format the model must output. `{"type": "json_object"}`
-        enables JSON mode.
-    seed
-        Two completions with the same `seed` value should return the same
-        completion. This is however not guaranteed.
-    stop
-        Up to 4 words where the API will stop the completion.
-    temperature
-        Number between 0 and 2. Higher values make the output more random, while
-        lower values make it more deterministic.
-    top_p
-        Number between 0 and 1. Parameter for nucleus sampling.
-    user
-        A unique identifier for the end-user.
-    """
-
-    model: str = ""
-    frequency_penalty: float = 0
-    logit_bias: Dict[int, int] = field(default_factory=dict)
-    max_tokens: Optional[int] = None
-    n: int = 1
-    presence_penalty: float = 0
-    response_format: Optional[Dict[str, str]] = None
-    seed: Optional[int] = None
-    stop: Optional[Union[str, List[str]]] = None
-    temperature: float = 1.0
-    top_p: int = 1
-    user: str = field(default_factory=str)
-
-
-class OpenAILegacy():
-    """An object that represents the OpenAI API."""
-
-    def __init__(
-        self,
-        client,
-        config,
-        system_prompt: Optional[str] = None,
-    ):
-        """Create an `OpenAI` instance.
-
-        This class supports the standard OpenAI API, the Azure OpeanAI API as
-        well as compatible APIs that rely on the OpenAI client.
-
-        Parameters
-        ----------
-        client
-            An instance of the API's async client.
-        config
-            An instance of `OpenAIConfig`. Can be useful to specify some
-            parameters that cannot be set by calling this class' methods.
-        """
-
-        self.client = client
-        self.config = config
-        self.system_prompt = system_prompt
-
-        # We count the total number of prompt and generated tokens as returned
-        # by the OpenAI API, summed over all the requests performed with this
-        # model instance.
-        self.prompt_tokens = 0
-        self.completion_tokens = 0
-
-        self.format_sequence = lambda x: x
-
-    def __call__(
-        self,
-        prompt: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[List[str], str]] = None,
-        *,
-        system_prompt: Optional[str] = None,
-        temperature: Optional[float] = None,
-        samples: Optional[int] = None,
-    ):
-        """Call the OpenAI API to generate text.
-
-        Parameters
-        ----------
-        prompt
-            A string or list of strings that will be used to prompt the model
-        max_tokens
-            The maximum number of tokens to generate
-        stop_at
-            A string or array of strings which, such that the generation stops
-            when they are generated.
-        system_prompt
-            The content of the system message that precedes the user's prompt.
-        temperature
-            The value of the temperature used to sample tokens
-        samples
-            The number of completions to generate for each prompt
-        stop_at
-            Up to 4 words where the API will stop the completion.
-
-        """
-        if max_tokens is None:
-            max_tokens = self.config.max_tokens
-        if stop_at is None:
-            stop_at = self.config.stop
-        if temperature is None:
-            temperature = self.config.temperature
-        if samples is None:
-            samples = self.config.n
-
-        config = replace(
-            self.config,
-            max_tokens=max_tokens,
-            temperature=temperature,
-            n=samples,
-            stop=stop_at,
-        )  # type: ignore
-
-        response, prompt_tokens, completion_tokens = generate_chat(
-            prompt, system_prompt or self.system_prompt, self.client, config
-        )
-        self.prompt_tokens += prompt_tokens
-        self.completion_tokens += completion_tokens
-
-        return self.format_sequence(response)
-
-    def stream(self, *args, **kwargs):
-        raise NotImplementedError(
-            "Streaming is currently not supported for the OpenAI API"
-        )
-
-    def new_with_replacements(self, model, **kwargs):
-        new_instance = copy.copy(model)
-        new_instance.legacy_instance.config = replace(
-            new_instance.legacy_instance.config, **kwargs
-        )
-        return new_instance
-
-    def __str__(self):
-        return self.__class__.__name__ + " API"
-
-    def __repr__(self):
-        return str(self.config)
-
-
-@functools.partial(vectorize, signature="(),(),(),()->(s),(),()")
-async def generate_chat(
-    prompt: str,
-    system_prompt: Union[str, None],
-    client,
-    config: OpenAIConfig,
-):
-    """Call OpenAI's Chat Completion API.
-
-    Parameters
-    ----------
-    prompt
-        The prompt we use to start the generation. Passed to the model
-        with the "user" role.
-    system_prompt
-        The system prompt, passed to the model with the "system" role
-        before the prompt.
-    client
-        The API client
-    config
-        An `OpenAIConfig` instance.
-
-    Returns
-    -------
-    A tuple that contains the model's response(s) and usage statistics.
-
-    """
-    import numpy as np
-
-    @error_handler
-    @cache()
-    async def call_api(prompt, system_prompt, config):
-        responses = await client.chat.completions.create(
-            messages=system_message + user_message,
-            **asdict(config),  # type: ignore
-        )
-        return responses.model_dump()
-
-    system_message = (
-        [{"role": "system", "content": system_prompt}] if system_prompt else []
-    )
-    user_message = [{"role": "user", "content": prompt}]
-
-    responses = await call_api(prompt, system_prompt, config)
-
-    results = np.array(
-        [responses["choices"][i]["message"]["content"] for i in range(config.n)]
-    )
-    usage = responses["usage"]
-
-    return results, usage["prompt_tokens"], usage["completion_tokens"]
-
-
-def error_handler(api_call_fn: Callable) -> Callable:
-    """Handle OpenAI API errors and missing API key."""
-
-    def call(*args, **kwargs):
-        import openai
-
-        try:
-            return api_call_fn(*args, **kwargs)
-        except (
-            openai.APITimeoutError,
-            openai.InternalServerError,
-            openai.RateLimitError,
-        ) as e:
-            raise OSError(f"Could not connect to the OpenAI API: {e}")
-        except (
-            openai.AuthenticationError,
-            openai.BadRequestError,
-            openai.ConflictError,
-            openai.PermissionDeniedError,
-            openai.NotFoundError,
-            openai.UnprocessableEntityError,
-        ) as e:
-            raise e
-
-    return call
diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index 0a80a54f7..5e38e72f4 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -1,11 +1,10 @@
 """Integration with the `vllm` library (offline mode)."""
 
 import json
-import warnings
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Any, List, Optional, Union
 
-from outlines.inputs import Chat, Image
+from outlines.inputs import Chat
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.openai import OpenAITypeAdapter
 from outlines.types.dsl import CFG, JsonSchema, python_types_to_terms, to_regex
@@ -111,7 +110,6 @@ def __init__(self, model: "LLM"):
         """
         self.model = model
         self.type_adapter = VLLMOfflineTypeAdapter()
-        self.lora_request = None # v0 legacy, to be removed
 
     def _build_generation_args(
         self,
@@ -167,14 +165,12 @@ def generate(
             results = self.model.chat(
                 messages=self.type_adapter.format_input(model_input),
                 sampling_params=sampling_params,
-                lora_request=self.lora_request, # v0 legacy, to be removed
                 **inference_kwargs,
             )
         else:
             results = self.model.generate(
                 prompts=self.type_adapter.format_input(model_input),
                 sampling_params=sampling_params,
-                lora_request=self.lora_request, # v0 legacy, to be removed
                 **inference_kwargs,
             )
         results = [completion.text for completion in results[0].outputs]
@@ -223,7 +219,6 @@ def generate_batch(
         results = self.model.generate(
             prompts=[self.type_adapter.format_input(item) for item in model_input],
             sampling_params=sampling_params,
-            lora_request=self.lora_request, # v0 legacy, to be removed
             **inference_kwargs,
         )
         return [[sample.text for sample in batch.outputs] for batch in results]
@@ -238,30 +233,6 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
             "Streaming is not available for the vLLM offline integration."
         )
 
-    def load_lora(self, adapter_path: Optional[str]) -> None:
-        """Load a LoRA adapter. Deprecated since v1.0.0.
-
-        Use the `lora_request` argument when calling the model or generator
-        instead.
-
-        """
-        warnings.warn("""
-            The `load_lora` method is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.1.0.
-            Please use the v1 of the `outlines` library by using the
-            `outlines.from_vllm` function to create a `VLLM` model
-            instance.
-            In the v1, you must pass the `lora_request` argument as
-            a keyword argument when calling the model or generator.
-            """)
-
-        from vllm.lora.request import LoRARequest
-
-        if adapter_path is None:
-            self.lora_request = None
-        else:
-            self.lora_request = LoRARequest(adapter_path, 1, adapter_path)
-
 
 def from_vllm_offline(model: "LLM") -> VLLMOffline:
     """Create an Outlines `VLLMOffline` model instance from a `vllm.LLM`
diff --git a/outlines/templates.py b/outlines/templates.py
index 79620e721..5ab8bc47b 100644
--- a/outlines/templates.py
+++ b/outlines/templates.py
@@ -64,7 +64,6 @@ class Template:
 
     """
     template: jinja2.Template
-    signature: Optional[inspect.Signature]
 
     def __call__(self, *args, **kwargs) -> str:
         """Render and return the template.
@@ -75,12 +74,7 @@ def __call__(self, *args, **kwargs) -> str:
             The rendered template as a Python string.
 
         """
-        if self.signature is not None:
-            bound_arguments = self.signature.bind(*args, **kwargs)
-            bound_arguments.apply_defaults()
-            return self.template.render(**bound_arguments.arguments)
-        else:
-            return self.template.render(**kwargs)
+        return self.template.render(**kwargs)
 
     @classmethod
     def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
@@ -98,7 +92,7 @@ def from_string(cls, content: str, filters: Dict[str, Callable] = {}):
             An instance of the class with the provided content as a template.
 
         """
-        return cls(build_template_from_string(content, filters), None)
+        return cls(build_template_from_string(content, filters))
 
     @classmethod
     def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
@@ -125,7 +119,7 @@ def from_file(cls, path: Path, filters: Dict[str, Callable] = {}):
         # infer one from a Jinja2 environment that is
         # split across multiple files (since e.g. we support features like
         # Jinja2 includes and template inheritance)
-        return cls(build_template_from_file(path, filters), None)
+        return cls(build_template_from_file(path, filters))
 
 
 def build_template_from_string(
@@ -159,90 +153,6 @@ def build_template_from_file(
     return env.get_template(os.path.basename(path))
 
 
-def prompt(
-    fn: Optional[Callable] = None,
-    filters: Dict[str, Callable] = {},
-) -> Callable:
-    """Decorate a function that contains a prompt template.
-
-    This allows to define prompts in the docstring of a function and simplify their
-    manipulation by providing some degree of encapsulation. It uses the `render`
-    function internally to render templates.
-
-    ```pycon
-    >>> import outlines
-    >>>
-    >>> @outlines.prompt
-    >>> def build_prompt(question):
-    ...    "I have a ${question}"
-    ...
-    >>> prompt = build_prompt("How are you?")
-    ```
-
-    This API can also be helpful in an "agent" context where parts of the prompt
-    are set when the agent is initialized and never modified later. In this situation
-    we can partially apply the prompt function at initialization.
-
-    ```pycon
-    >>> import outlines
-    >>> import functools as ft
-    ...
-    >>> @outlines.prompt
-    ... def solve_task(name: str, objective: str, task: str):
-    ...     \"""Your name is {{name}}.
-    ...     Your overall objective is to {{objective}}.
-    ...     Please solve the following task: {{task}}
-    ...     \"""
-    ...
-    >>> hal = ft.partial(solve_task, "HAL", "Travel to Jupiter")
-    ```
-
-    Additional Jinja2 filters can be provided as keyword arguments to the decorator.
-
-    ```pycon
-    >>> def reverse(s: str) -> str:
-    ...     return s[::-1]
-    ...
-    >>> @outlines.prompt(filters={ 'reverse': reverse })
-    ... def reverse_prompt(text):
-    ...     \"""{{ text | reverse }}\"""
-    ...
-    >>> prompt = reverse_prompt("Hello")
-    >>> print(prompt)
-    ... "olleH"
-    ```
-
-    Returns
-    -------
-    A `Template` callable class which will render the template when called.
-
-    """
-    warnings.warn(
-        "The @prompt decorator is deprecated and will be removed in outlines 1.1.0. "
-        "Instead of using docstring templates, please use Template.from_file() to "
-        "load your prompts from separate template files, or a simple Python function "
-        "that returns text. This helps keep prompt content separate from code and is "
-        "more maintainable.",
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if fn is None:
-        return lambda fn: prompt(fn, cast(Dict[str, Callable], filters))
-
-    signature = inspect.signature(fn)
-
-    # The docstring contains the template that will be rendered to be used
-    # as a prompt to the language model.
-    docstring = fn.__doc__
-    if docstring is None:
-        raise TypeError("Could not find a template in the function's docstring.")
-
-    template = build_template_from_string(cast(str, docstring), filters)
-
-    return Template(template, signature)
-
-
 def create_jinja_env(
     loader: Optional[jinja2.BaseLoader], filters: Dict[str, Callable]
 ) -> jinja2.Environment:
@@ -400,7 +310,7 @@ def parse_pydantic_schema(raw_schema, definitions):
     for name, value in raw_schema["properties"].items():
         if "description" in value:
             simple_schema[name] = value["description"]
-        elif "$ref" in value:
+        elif "$ref" in value: # pragma: no cover
             refs = value["$ref"].split("/")
             simple_schema[name] = parse_pydantic_schema(
                 definitions[refs[2]], definitions
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index c9c66f0f9..4fe8ef902 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -28,9 +28,6 @@
     between,
     zero_or_more,
     one_or_more,
-    # deprecated
-    repeat,
-    times,
 )
 
 
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 0af69ab12..10d47d7c8 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -201,16 +201,6 @@ def one_or_more(self) -> "KleenePlus":
     def zero_or_more(self) -> "KleeneStar":
         return zero_or_more(self)
 
-    # deprecated
-    def times(self, count: int) -> "QuantifyExact":
-        return times(self, count)
-
-    # deprecated
-    def repeat(self, min_count: int, max_count: int) -> Union[
-        "QuantifyMinimum", "QuantifyMaximum", "QuantifyBetween"
-    ]:
-        return repeat(self, min_count, max_count)
-
 
 @dataclass
 class String(Term):
@@ -818,99 +808,6 @@ def _handle_dict(args: tuple, recursion_depth: int) -> Sequence:
     )
 
 
-# deprecated
-def repeat(term: Term, min_count: int, max_count: int) -> Union[
-    QuantifyMinimum, QuantifyMaximum, QuantifyBetween
-]:
-    if min_count is None and max_count is None:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `between`, `at_least` or `at_most` instead.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        raise ValueError(
-            "You must provide a value for at least `min_count` or `max_count`"
-        )
-    if max_count is None:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `at_least` instead.
-            For instance:
-            ```python
-            from outlines.types import Regex, at_least
-            digit = Regex(r'[0-9]')
-            # 2 ways of doing the same thing
-            at_least_5_digits = digit.at_least(5)
-            at_least_5_digits = at_least(digit, 5)
-            ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        return at_least(min_count, term)
-    if min_count is None:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `at_most` instead.
-            For instance:
-            ```python
-            from outlines.types import Regex, at_most
-            digit = Regex(r'[0-9]')
-            # 2 ways of doing the same thing
-            at_most_5_digits = digit.at_most(5)
-            at_most_5_digits = at_most(digit, 5)
-            ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        return at_most(max_count, term)
-    else:
-        warnings.warn("""
-            The `repeat` function/method is deprecated starting from v0.2.2.
-            Do not use it. Support for it will be removed in v1.5.0.
-            Use `between` instead.
-            For instance:
-            ```python
-            from outlines.types import Regex, between
-            digit = Regex(r'[0-9]')
-            # 2 ways of doing the same thing
-            five_to_eight_digits = digit.between(5, 8)
-            five_to_eight_digits = between(digit, 5, 8)
-            ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        return between(min_count, max_count, term)
-
-
-# deprecated
-def times(term: Term, count: int) -> QuantifyExact:
-    warnings.warn("""
-        The `times` function/method is deprecated starting from v0.2.2.
-        Do not use it. Support for it will be removed in v1.5.0.
-        Use `exactly` instead.
-        For instance:
-        ```python
-        from outlines.types import Regex, exactly
-        digit = Regex(r'[0-9]')
-        # 2 ways of doing the same thing
-        five_digits = digit.exactly(5)
-        five_digits = exactly(digit, 5)
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-    return exactly(count, term)
-
-
 def to_regex(term: Term) -> str:
     """Convert a term to a regular expression.
 
diff --git a/outlines/v0_legacy/base.py b/outlines/v0_legacy/base.py
deleted file mode 100644
index e545b5bcf..000000000
--- a/outlines/v0_legacy/base.py
+++ /dev/null
@@ -1,296 +0,0 @@
-import asyncio
-import builtins
-import functools
-import inspect
-from typing import Callable, Optional
-
-
-# Allow nested loops for running in notebook. We don't enable it globally as it
-# may interfere with other libraries that use asyncio.
-if hasattr(builtins, "__IPYTHON__"): # pragma: no cover
-    try:
-        import nest_asyncio
-
-        nest_asyncio.apply()
-    except ImportError:
-        print(
-            "Couldn't patch nest_asyncio because it's not installed. Running in the notebook might be have issues"
-        )
-
-
-class vectorize:
-    """Returns an object that acts like a function but takes arrays as an input.
-
-    The vectorized function evaluates `func` over successive tuples of the input
-    chararrays and returns a single NumPy chararrays or a tuple of NumPy chararrays.
-
-    Its behavior is similar to NumPy's `vectorize` for Python functions: the function
-    being vectorized is executed in a `for` loop. Coroutines, however, are executed
-    concurrently.
-
-    Part of the code was adapted from `numpy.lib.function_base`.
-
-    """
-
-    def __init__(self, func: Callable, signature: Optional[str] = None):
-        self.setup_numpy(signature)
-        self.func = func
-        self.signature = signature
-        self.is_coroutine_fn = inspect.iscoroutinefunction(func)
-
-        functools.update_wrapper(self, func)
-
-    def setup_numpy(self, signature: Optional[str] = None):
-        """Setup NumPy and import required functions based on NumPy version.
-
-        As numpy is an optional dependency only needed for the legacy OpenAI
-        model, we put the import in a try block to make sure no error is raised
-        if it's not installed.
-        """
-        try:
-            import numpy as np
-            self.np = np
-            # Import required functions based on NumPy version
-            np_major_version = int(np.__version__.split(".")[0])
-            if np_major_version >= 2:
-                from numpy.lib._function_base_impl import (
-                    _calculate_shapes,
-                    _parse_gufunc_signature,
-                    _parse_input_dimensions,
-                    _update_dim_sizes,
-                )
-            else: # pragma: no cover
-                from numpy.lib.function_base import (
-                    _calculate_shapes,
-                    _parse_gufunc_signature,
-                    _parse_input_dimensions,
-                    _update_dim_sizes,
-                )
-            self._calculate_shapes = _calculate_shapes
-            self._parse_gufunc_signature = _parse_gufunc_signature
-            self._parse_input_dimensions = _parse_input_dimensions
-            self._update_dim_sizes = _update_dim_sizes
-            if signature is not None:
-                # Parse the signature string into a Python data structure.
-                # For instance "(m),(s)->(s,m)" becomes `([(m,),(s,)],[(s,m)])`.
-                self._in_and_out_core_dimensions = self._parse_gufunc_signature(signature)
-            else:
-                self._in_and_out_core_dimensions = None
-        except (ModuleNotFoundError, ImportError): # pragma: no cover
-            self.np = None
-            self._calculate_shapes = None
-            self._parse_gufunc_signature = None
-            self._parse_input_dimensions = None
-            self._update_dim_sizes = None
-            self._in_and_out_core_dimensions = None
-
-    def __call__(self, *args, **kwargs):
-        """Call the vectorized function."""
-        # Raise an error ourselves if NumPy is not installed.
-        if not self.np: # pragma: no cover
-            raise ImportError(
-                "NumPy is required to use the legacy version of the OpenAI "
-                "model."
-            )
-        if not args and not kwargs:
-            return self.call_thunk()
-        elif self.signature is not None:
-            return self.call_with_signature(*args, **kwargs)
-        else:
-            return self.call_no_signature(*args, **kwargs)
-
-    def call_thunk(self):
-        """Call a vectorized thunk.
-
-        Thunks have no arguments and can thus be called directly.
-
-        """
-        if self.is_coroutine_fn:
-            loop = asyncio.new_event_loop()
-            try:
-                outputs = loop.run_until_complete(self.func())
-            finally:
-                loop.close()
-        else:
-            outputs = self.func()
-
-        return outputs
-
-    def call_no_signature(self, *args, **kwargs):
-        """Call functions and coroutines when no signature is specified.
-
-        When no signature is specified we assume that all of the function's
-        inputs and outputs are scalars (core dimension of zero). We first
-        broadcast the input arrays, then iteratively apply the function over the
-        elements of the broadcasted arrays and finally reshape the results to
-        match the input shape.
-
-        Functions are executed in a for loop, coroutines are executed
-        concurrently.
-
-        """
-        # Convert args and kwargs to arrays
-        args = [self.np.array(arg) for arg in args]
-        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
-
-        # Broadcast args and kwargs
-        broadcast_shape = self.np.broadcast(*args, *list(kwargs.values())).shape
-        args = [self.np.broadcast_to(arg, broadcast_shape) for arg in args]
-        kwargs = {
-            key: self.np.broadcast_to(value, broadcast_shape)
-            for key, value in kwargs.items()
-        }
-
-        # Execute functions in a loop, and coroutines concurrently
-        if self.is_coroutine_fn:
-            outputs = self.vectorize_call_coroutine(broadcast_shape, args, kwargs)
-        else:
-            outputs = self.vectorize_call(broadcast_shape, args, kwargs)
-
-        # `outputs` is a flat array or a tuple of flat arrays. We reshape the arrays
-        # to match the input shape.
-        outputs = [
-            results if isinstance(results, tuple) else (results,) for results in outputs
-        ]
-        outputs = tuple(
-            [self.np.asarray(x).reshape(broadcast_shape).squeeze() for x in zip(*outputs)]
-        )
-        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
-
-        n_results = len(list(outputs))
-
-        return outputs[0] if n_results == 1 else outputs
-
-    def call_with_signature(self, *args, **kwargs):
-        """Call functions and coroutines when a signature is specified."""
-        input_core_dims, output_core_dims = self._in_and_out_core_dimensions
-
-        # Make sure that the numbers of arguments passed is compatible with
-        # the signature.
-        num_args = len(args) + len(kwargs)
-        if num_args != len(input_core_dims):
-            raise TypeError(
-                "wrong number of positional arguments: "
-                "expected %r, got %r" % (len(input_core_dims), len(args))
-            )
-
-        # Convert args and kwargs to arrays
-        args = [self.np.asarray(arg) for arg in args]
-        kwargs = {key: self.np.array(value) for key, value in kwargs.items()}
-
-        # Find the arguments' broadcast shape, and map placeholder
-        # variables in the signature to the number of dimensions
-        # they correspond to given the arguments.
-        broadcast_shape, dim_sizes = self._parse_input_dimensions(
-            args + list(kwargs.values()), input_core_dims
-        )
-
-        # Calculate the shape to which each of the arguments should be broadcasted
-        # and reshape them accordingly.
-        input_shapes = self._calculate_shapes(broadcast_shape, dim_sizes, input_core_dims)
-        args = [
-            self.np.broadcast_to(arg, shape, subok=True)
-            for arg, shape in zip(args, input_shapes)
-        ]
-        kwargs = {
-            key: self.np.broadcast_to(value, broadcast_shape)
-            for key, value in kwargs.items()
-        }
-
-        n_out = len(output_core_dims)
-
-        if self.is_coroutine_fn:
-            outputs = self.vectorize_call_coroutine(broadcast_shape, args, kwargs)
-        else:
-            outputs = self.vectorize_call(broadcast_shape, args, kwargs)
-
-        outputs = [
-            results if isinstance(results, tuple) else (results,) for results in outputs
-        ]
-
-        flat_outputs = list(zip(*outputs))
-        n_results = len(flat_outputs)
-
-        if n_out != n_results:
-            raise ValueError(
-                f"wrong number of outputs from the function, expected {n_out}, got {n_results}"
-            )
-
-        # The number of dimensions of the outputs are not necessarily known in
-        # advance. The following iterates over the results and updates the
-        # number of dimensions of the outputs accordingly.
-        for results, core_dims in zip(flat_outputs, output_core_dims):
-            for result in results:
-                self._update_dim_sizes(dim_sizes, result, core_dims)
-
-        # Calculate the shape to which each of the outputs should be broadcasted
-        # and reshape them.
-        shapes = self._calculate_shapes(broadcast_shape, dim_sizes, output_core_dims)
-        outputs = tuple(
-            [
-                self.np.hstack(results).reshape(shape).squeeze()
-                for shape, results in zip(shapes, zip(*outputs))
-            ]
-        )
-        outputs = tuple([x.item() if self.np.ndim(x) == 0 else x for x in outputs])
-
-        return outputs[0] if n_results == 1 else outputs
-
-    def vectorize_call(self, broadcast_shape, args, kwargs):
-        """Run the function in a for loop.
-
-        A possible extension would be to parallelize the calls.
-
-        Parameters
-        ----------
-        broadcast_shape
-            The brodcast shape of the input arrays.
-        args
-            The function's broadcasted arguments.
-        kwargs
-            The function's broadcasted keyword arguments.
-
-        """
-        outputs = []
-        for index in self.np.ndindex(*broadcast_shape):
-            current_args = tuple(arg[index] for arg in args)
-            current_kwargs = {key: value[index] for key, value in kwargs.items()}
-            outputs.append(self.func(*current_args, **current_kwargs))
-
-        return outputs
-
-    def vectorize_call_coroutine(self, broadcast_shape, args, kwargs):
-        """Run coroutines concurrently.
-
-        Creates as many tasks as needed and executes them in a new event
-        loop.
-
-        Parameters
-        ----------
-        broadcast_shape
-            The brodcast shape of the input arrays.
-        args
-            The function's broadcasted arguments.
-        kwargs
-            The function's broadcasted keyword arguments.
-
-        """
-
-        async def create_and_gather_tasks():
-            tasks = []
-            for index in self.np.ndindex(*broadcast_shape):
-                current_args = tuple(arg[index] for arg in args)
-                current_kwargs = {key: value[index] for key, value in kwargs.items()}
-                tasks.append(self.func(*current_args, **current_kwargs))
-
-            outputs = await asyncio.gather(*tasks)
-
-            return outputs
-
-        loop = asyncio.new_event_loop()
-        try:
-            outputs = loop.run_until_complete(create_and_gather_tasks())
-        finally:
-            loop.close()
-
-        return outputs
diff --git a/outlines/v0_legacy/function.py b/outlines/v0_legacy/function.py
deleted file mode 100644
index 465df53f9..000000000
--- a/outlines/v0_legacy/function.py
+++ /dev/null
@@ -1,168 +0,0 @@
-import importlib.util
-import warnings
-from dataclasses import dataclass
-from typing import TYPE_CHECKING, Callable, Optional, Tuple, Union
-
-import requests
-
-from outlines.v0_legacy import generate
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-from outlines.v0_legacy.models import transformers
-
-if TYPE_CHECKING: # pragma: no cover
-    from outlines.templates import Template
-
-
-def function_warning():
-    warnings.warn("""
-        The `Function` class is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should use the `outlines.Application` class that
-        implements a similar functionality.
-        An `Application` is initialized with a prompt template and an
-        output type. It can then be called with a model and a prompt
-        to generate a response that follows the format specified by the
-        output type.
-        For example:
-        ```python
-        from pydantic import BaseModel
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-        from outlines import models, Application
-        from outlines.types import JsonType
-        from outlines.templates import Template
-
-        class OutputModel(BaseModel):
-            result: int
-
-        model = models.from_transformers(
-            AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
-            AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-        )
-
-        template_string = "What is 2 times {{ num }}?"
-        template = Template.from_string(template_string)
-
-        application = Application(template, JsonType(OutputModel))
-
-        result = application(model, num=3)
-        print(result)  # Expected output: { "result" : 6 }
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2
-    )
-
-
-@dataclass
-class Function:
-    """Represents an Outlines function.
-
-    Functions are a convenient way to encapsulate a prompt template, a language
-    model and a Pydantic model that define the output structure. Once defined,
-    the function can be called with arguments that will be used to render the
-    prompt template.
-
-    """
-    prompt_template: "Template"
-    schema: Union[str, Callable, object]
-    model_name: str
-    generator: Optional["GeneratorV0Adapter"] = None
-
-    def __post_init__(self):
-        function_warning()
-
-    @classmethod
-    def from_github(cls, program_path: str, function_name: str = "fn"):
-        """Load a function stored on GitHub"""
-        function_warning()
-        program_content = download_from_github(program_path)
-        function = extract_function_from_file(program_content, function_name)
-
-        return function
-
-    def init_generator(self):
-        """Load the model and initialize the generator."""
-        model = transformers(self.model_name)
-        self.generator = generate.json(model, self.schema)
-
-    def __call__(self, *args, **kwargs):
-        """Call the function.
-
-        .. warning::
-
-           This currently does not support batching.
-
-        Parameters
-        ----------
-        args
-            Values to pass to the prompt template as positional arguments.
-        kwargs
-            Values to pass to the prompt template as keyword arguments.
-
-        """
-        # We do not want to raise the deprecation warning about the `json`
-        # function or the transformers function here.
-        with warnings.catch_warnings():
-            warnings.filterwarnings(
-                "ignore",
-                category=DeprecationWarning
-            )
-            if self.generator is None:
-                self.init_generator()
-
-            prompt = self.prompt_template(*args, **kwargs)
-            return self.generator(prompt)
-
-
-def download_from_github(short_path: str):
-    """Download the file in which the function is stored on GitHub."""
-    GITHUB_BASE_URL = "https://raw.githubusercontent.com"
-    BRANCH = "main"
-
-    path = short_path.split("/")
-    if len(path) < 3:
-        raise ValueError(
-            "Please provide a valid path in the form {USERNAME}/{REPO_NAME}/{PATH_TO_FILE}."
-        )
-    elif short_path[-3:] == ".py":
-        raise ValueError("Do not append the `.py` extension to the program name.")
-
-    username = path[0]
-    repo = path[1]
-    path_to_file = path[2:]
-
-    url = "/".join([GITHUB_BASE_URL, username, repo, BRANCH] + path_to_file) + ".py"
-    result = requests.get(url)
-
-    if result.status_code == 200:
-        return result.text
-    elif result.status_code == 404:
-        raise ValueError(
-            f"Program could not be found at {url}. Please make sure you entered the GitHub username, repository name and path to the program correctly."
-        )
-    else:
-        result.raise_for_status()
-
-
-def extract_function_from_file(content: str, function_name: str) -> Tuple[Callable]:
-    """Extract a function object from a downloaded file."""
-    spec = importlib.util.spec_from_loader(
-        "outlines_function", loader=None, origin="github"
-    )
-    if spec is not None:  # pragma: no cover
-        module = importlib.util.module_from_spec(spec)
-        exec(content, module.__dict__)
-
-        try:
-            fn = getattr(module, function_name)
-        except AttributeError:
-            raise AttributeError(
-                "Could not find an `outlines.Function` instance in the remote file. Make sure that the path you specified is correct."
-            )
-
-        if not isinstance(fn, module.outlines.Function):
-            raise TypeError(
-                f"The `{function_name}` variable in the program must be an instance of `outlines.Function`"
-            )
-
-    return fn
diff --git a/outlines/v0_legacy/generate/__init__.py b/outlines/v0_legacy/generate/__init__.py
deleted file mode 100644
index 617f7b2e3..000000000
--- a/outlines/v0_legacy/generate/__init__.py
+++ /dev/null
@@ -1,17 +0,0 @@
-from .cfg import cfg
-from .choice import choice
-from .format import format
-from .fsm import fsm
-from .json import json
-from .regex import regex
-from .text import text
-
-__all__ = [
-    "cfg",
-    "choice",
-    "format",
-    "fsm",
-    "json",
-    "regex",
-    "text",
-]
diff --git a/outlines/v0_legacy/generate/api.py b/outlines/v0_legacy/generate/api.py
deleted file mode 100644
index 6e824f0b0..000000000
--- a/outlines/v0_legacy/generate/api.py
+++ /dev/null
@@ -1,183 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-from dataclasses import asdict
-from typing import Any, Dict, List, Optional, Union
-
-from outlines.generator import Generator
-from outlines.v0_legacy.models.exllamav2 import ExLlamaV2Model, exllamav2_params_adapter
-from outlines.v0_legacy.models.transformers import transformers_params_adapter
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
-from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
-from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
-from outlines.models import (
-    LlamaCpp,
-    MLXLM,
-    Transformers,
-    VLLMOffline,
-)
-
-V0GeneratorModel = Union[
-    ExLlamaV2Model,
-    LlamaCpp,
-    MLXLM,
-    Transformers,
-    TransformersVision,
-    VLLMOffline,
-]
-
-
-class GeneratorV0Adapter:
-    """Object used to bridge the gap between the legacy `generate` functions
-    and the current `Generator` object.
-
-    The legacy `generate` functions return an instance of this class. When
-    initialized, it stores an instance of the current `Generator` object
-    and calls it when it's called.
-    """
-
-    def __init__(self, model, output_type, sampler):
-        if not isinstance(model, V0GeneratorModel):  # type: ignore
-            raise ValueError(
-                "You can only use the v0 API with models that were already"
-                + f"available in v0. Got {type(model)}."
-            )
-        self.model = model
-        self.generator = Generator(model, output_type)
-        self.sampling_params = asdict(sampler.sampling_params)
-
-    def format_sequence(self, sequence: str):
-        return sequence
-
-    def create_inference_params(
-        self, max_tokens, stop_at, seed, **model_specific_params
-    ):
-        """Turn the arguments received by the generator into arguments
-        for the model associated to it.
-        """
-        generation_params = {
-            "max_tokens": max_tokens,
-            "stop_at": stop_at,
-            "seed": seed,
-        }
-
-        if isinstance(self.model, LlamaCpp):
-            adapter_function = llamacpp_params_adapter
-        elif isinstance(self.model, VLLMOffline): # pragma: no cover
-            adapter_function = vllm_offline_params_adapter
-        elif isinstance(self.model, Transformers):
-            adapter_function = transformers_params_adapter
-        elif isinstance(self.model, ExLlamaV2Model): # pragma: no cover
-            adapter_function = exllamav2_params_adapter
-        elif isinstance(self.model, MLXLM): # pragma: no cover
-            adapter_function = mlxlm_params_adapter
-
-        inference_params = adapter_function(
-            self.sampling_params,
-            generation_params,
-            model_specific_params
-        )
-        return inference_params
-
-    def __call__(
-        self,
-        prompts: Union[str, List[str], Dict[str, Any]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        if isinstance(prompts, list):
-            result = self.generator.batch(
-                prompts,
-                **self.create_inference_params(
-                    max_tokens, stop_at, seed, **model_specific_params
-                )
-            )
-        else:
-            result = self.generator(
-                prompts,
-                **self.create_inference_params(
-                    max_tokens, stop_at, seed, **model_specific_params
-                )
-            )
-        if isinstance(result, list):
-            return [self.format_sequence(r) for r in result]
-        else:
-            return self.format_sequence(result)
-
-    def stream(
-        self,
-        prompts: Union[str, List[str], Dict[str, Any]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        return self.generator.stream(
-            prompts,
-            **self.create_inference_params(
-                max_tokens, stop_at, seed, **model_specific_params
-            )
-        )
-
-
-class GeneratorVisionV0Adapter(GeneratorV0Adapter):
-    """This object does the same as `GeneratorV0Adapter` but for the
-    `TransformersVision` model. The only difference between the two is that
-    it receives a media argument in addition to the prompts.
-    """
-
-    @staticmethod
-    def merge_prompts_and_media(
-        prompts: Union[str, List[str]],
-        media: Union[Any, List[Any]],
-    ) -> Dict[str, Any]:
-        """Merge prompts and media into a single dictionary
-
-        This function is needed as the `TransformersMultiModal` model requires
-        the prompt and assets to be contained in a single argument while the
-        legacy `TransformersVision` model requires separate arguments for the
-        prompt and assets.
-        """
-        return {
-            "text": prompts,
-            "images": media,
-        }
-
-    def __call__(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        merged_input = self.merge_prompts_and_media(prompts, media)
-        return super().__call__(
-            merged_input,
-            max_tokens,
-            stop_at,
-            seed,
-            **model_specific_params
-        )
-
-    def stream(  # type: ignore
-        self,
-        prompts: Union[str, List[str]],
-        media: Union[str, List[str]],
-        max_tokens: Optional[int] = None,
-        stop_at: Optional[Union[str, List[str]]] = None,
-        seed: Optional[int] = None,
-        **model_specific_params,
-    ):
-        merged_input = self.merge_prompts_and_media(prompts, media)
-        return super().stream(
-            merged_input,
-            max_tokens,
-            stop_at,
-            seed,
-            **model_specific_params
-        )
diff --git a/outlines/v0_legacy/generate/cfg.py b/outlines/v0_legacy/generate/cfg.py
deleted file mode 100644
index 0f33b6efb..000000000
--- a/outlines/v0_legacy/generate/cfg.py
+++ /dev/null
@@ -1,80 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models.llamacpp import LlamaCpp
-from outlines.models.openai import OpenAI
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.types.dsl import CFG
-
-
-def cfg(
-    model, cfg_str: str, sampler: Sampler = multinomial()
-) -> GeneratorV0Adapter:
-    """Generate text in the language of a Context-Free Grammar
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    from outlines.types import CFG
-    cfg_str = "..."  # Context-Free Grammar as a string
-    generator = Generator(model, CFG(cfg_str))
-    ```
-
-    You can then call the generator created with a prompt to generate text
-    that matches the Context-Free Grammar.
-
-    Arguments
-    ---------
-    model:
-        The Outlines v0 model to use to generate text.
-    cfg_str:
-        The Context-Free Grammar to use to generate text.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-
-    warnings.warn("""
-        The `cfg` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import CFG
-        cfg_str = '...'  # Context-Free Grammar as a string
-        generator = Generator(model, CFG(cfg_str))
-        ```
-        You can then call the generator created with a prompt to generate
-        text in the language of a Context-Free Grammar.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use CFG-based structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API."
-        )
-    elif isinstance(model, LlamaCpp):
-        raise NotImplementedError(
-            "Not yet available due to bug in llama_cpp tokenizer"
-        )
-    elif isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, CFG(cfg_str), sampler)
-    return GeneratorV0Adapter(model, CFG(cfg_str), sampler)
diff --git a/outlines/v0_legacy/generate/choice.py b/outlines/v0_legacy/generate/choice.py
deleted file mode 100644
index eff2fcffd..000000000
--- a/outlines/v0_legacy/generate/choice.py
+++ /dev/null
@@ -1,126 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import re
-import json as pyjson
-import warnings
-from enum import Enum
-from typing import Callable, List, Union
-
-from outlines.models.openai import OpenAI
-from outlines.types.utils import get_schema_from_enum
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
-from outlines.v0_legacy.generate.json import json
-from outlines.v0_legacy.generate.regex import regex
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-
-def choice(
-    model,
-    choices: Union[List[str], type[Enum]],
-    sampler: Sampler = multinomial(),
-) -> Union[GeneratorV0Adapter, GeneratorVisionV0Adapter, Callable]:
-    """Generate a choice from a list of options.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from typing import Literal
-    from outlines import Generator
-    generator = Generator(model, Literal["foo", "bar"])
-    ```
-
-    You can then call the generator created with a prompt to generate a
-    choice from the list of options.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    choices:
-        A list of options to choose from.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate a choice from the list of options.
-
-    """
-    warnings.warn("""
-        The `choice` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from typing import Literal
-        generator = Generator(model, Literal['foo', 'bar'])
-        ```
-        You can then call the generator created with a prompt to generate
-        a choice from the list of options.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        if isinstance(choices, list):
-            return choice_openai(model, choices, sampler)
-        raise ValueError(
-            "The `choice` function with OpenAI only supports a list of "
-            + "strings as choices."
-        )
-
-    if isinstance(choices, type(Enum)):
-        regex_str = build_regex_from_schema(pyjson.dumps(get_schema_from_enum(choices)))
-    else:
-        choices = [re.escape(choice) for choice in choices]  # type: ignore
-        regex_str = r"(" + r"|".join(choices) + r")"
-
-    # We do not want to raise the deprecation warning through the regex function
-    with warnings.catch_warnings():
-        warnings.filterwarnings(
-            "ignore",
-            category=DeprecationWarning,
-        )
-        generator = regex(model, regex_str, sampler)
-
-    if isinstance(choices, type(Enum)):
-        generator.format_sequence = lambda x: pyjson.loads(x)
-    else:
-        generator.format_sequence = lambda x: x
-
-    return generator
-
-
-def choice_openai(
-    model,
-    choices: List[str],
-    sampler: Sampler = multinomial(),
-) -> Callable:
-
-    choices_schema = pyjson.dumps(
-        {
-            "type": "object",
-            "properties": {"result": {"type": "string", "enum": choices}},
-            "additionalProperties": False,
-            "required": ["result"],
-        }
-    )
-
-    # We do not want to raise the deprecation warning through the json function
-    with warnings.catch_warnings():
-        warnings.filterwarnings(
-            "ignore",
-            category=DeprecationWarning
-        )
-        generator = json(model, choices_schema, sampler)
-
-    def generate_choice(*args, **kwargs): # pragma: no cover
-        return generator(*args, **kwargs)["result"]
-
-    return generate_choice
diff --git a/outlines/v0_legacy/generate/format.py b/outlines/v0_legacy/generate/format.py
deleted file mode 100644
index cd574ac4e..000000000
--- a/outlines/v0_legacy/generate/format.py
+++ /dev/null
@@ -1,70 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models.openai import OpenAI
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.samplers import Sampler, multinomial
-
-
-def format(
-    model, python_type, sampler: Sampler = multinomial()
-) -> GeneratorV0Adapter:
-    """Generate structured data that can be parsed as a Python type.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead::
-
-    ```python
-    from outlines import Generator
-    generator = Generator(model, int)
-    ```
-
-    You can then call the generator created with a prompt to generate a
-    integer.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    python_type:
-        A Python type. The output of the generator must be parseable into
-        this type.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `SequenceGenerator` instance that generates text constrained by the Python type
-    and translates this text into the corresponding type.
-
-    """
-    warnings.warn("""
-        The `format` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        generator = Generator(model, int)
-        ```
-        You can then call the generator created with a prompt to generate
-        an integer.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use Python type-structured generation with an OpenAI model"
-            + " due to the limitations of the OpenAI API."
-        )
-    elif isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, python_type, sampler)
-    return GeneratorV0Adapter(model, python_type, sampler)
diff --git a/outlines/v0_legacy/generate/fsm.py b/outlines/v0_legacy/generate/fsm.py
deleted file mode 100644
index 23a5477fe..000000000
--- a/outlines/v0_legacy/generate/fsm.py
+++ /dev/null
@@ -1,77 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-import interegular
-
-from outlines.models.openai import OpenAI
-from outlines.types import FSM
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-
-
-def fsm(
-    model, fsm: interegular.fsm.FSM, sampler: Sampler = multinomial()
-) -> GeneratorV0Adapter:
-    """Generate text that follows a Finite State Machine.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    import interegular
-    from outlines import Generator
-    from outlines.types import FSM
-    fsm = FSM(interegular.parse_pattern(r"[a-z]+").to_fsm())
-    generator = Generator(model, fsm)
-    ```
-
-    You can then call the generator created with a prompt to generate text
-    that follows the Finite State Machine.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    fsm:
-        The interregular `FSM` to use to generate text.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text that follows the Finite State Machine.
-
-    """
-    warnings.warn("""
-        The `fsm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import FSM
-        fsm_str = '...'  # Finite State Machine as a string
-        generator = Generator(model, FSM(fsm_str))
-        ```
-        You can then call the generator created with a prompt to generate
-        text that follows the Finite State Machine.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use FSM-based generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API."
-        )
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, FSM(fsm), sampler)
-    return GeneratorV0Adapter(model, FSM(fsm), sampler)
diff --git a/outlines/v0_legacy/generate/json.py b/outlines/v0_legacy/generate/json.py
deleted file mode 100644
index 6260b6471..000000000
--- a/outlines/v0_legacy/generate/json.py
+++ /dev/null
@@ -1,145 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import json as pyjson
-import warnings
-from typing import Callable, Optional, Union
-
-from pydantic import BaseModel
-
-from outlines.models.openai import OpenAI
-from outlines.types.dsl import JsonSchema
-from outlines.types.utils import get_schema_from_signature, is_callable
-from outlines.v0_legacy.samplers import Sampler, multinomial
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-
-
-def json(
-    model,
-    schema_object: Union[str, object, Callable],
-    sampler: Sampler = multinomial(),
-    whitespace_pattern: Optional[str] = None,
-) -> GeneratorV0Adapter:
-    """Generate structured text that follows a JSON Schema.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    from outlines.types import JsonSchema
-    generator = Generator(model, JsonSchema(schema_object))
-    ```
-
-    You can then call the generator created with a prompt to generate text that
-    matches the JSON Schema.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    schema_object:
-        The JSON Schema to generate data for. Can be a JSON string, a Pydantic
-        model, or a callable that returns a JSON schema.
-    sampler:
-        The sampler defining the sampling parameters.
-    whitespace_pattern
-        Pattern to use for JSON syntactic whitespace (doesn't impact string
-        literals).
-        Example: allow only a single space or newline with
-        `whitespace_pattern=r"[\n ]?"`.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-    warnings.warn("""
-        The `json` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import JsonSchema
-        schema_str = '...'  # JSON Schema as a string
-        generator = Generator(model, JsonSchema(schema_str))
-        ```
-        You can then call the generator created with a prompt to generate
-        JSON data that matches the schema.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        return json_openai(model, schema_object, sampler)
-
-    if is_callable(schema_object):
-        json_schema = JsonSchema(get_schema_from_signature(schema_object))  # type: ignore
-    else:
-        json_schema = JsonSchema(schema_object, whitespace_pattern)
-
-    if isinstance(model, TransformersVision):
-        generator = GeneratorVisionV0Adapter(model, json_schema, sampler)
-    else:
-        generator = GeneratorV0Adapter(model, json_schema, sampler)  # type: ignore
-
-    if isinstance(schema_object, type(BaseModel)):
-        setattr(
-            generator,
-            'format_sequence',
-            lambda x: schema_object.model_validate_json(x)
-        )
-    else:
-        setattr(generator, 'format_sequence', lambda x: pyjson.loads(x))
-
-    return generator
-
-
-def json_openai(
-    model,
-    schema_object: Union[str, object, Callable],
-    sampler: Sampler = multinomial(),
-):
-    if not isinstance(sampler, multinomial):
-        raise NotImplementedError(
-            "The OpenAI API does not support any other sampling algorithm "
-            + "than the multinomial sampler."
-        )
-
-    if isinstance(schema_object, type(BaseModel)):
-        schema = schema_object.model_json_schema()
-        schema["additionalProperties"] = False
-        schema = pyjson.dumps(schema)
-        format_sequence = lambda x: schema_object.model_validate_json(x)
-    elif isinstance(schema_object, str):
-        schema = schema_object
-        format_sequence = lambda x: pyjson.loads(x)
-    else:
-        raise ValueError(
-            f"Cannot parse schema {schema_object}. The schema must be either "
-            + "a Pydantic object or a string that contains the JSON Schema "
-            + "specification"
-        )
-
-    # create copied, patched model with normalized json schema set
-    generator = model.new_with_replacements(
-        response_format={
-            "type": "json_schema",
-            "json_schema": {
-                "name": "default",
-                "strict": True,
-                "schema": pyjson.loads(schema),
-            },
-        }
-    )
-
-    generator.format_sequence = format_sequence
-
-    return generator
diff --git a/outlines/v0_legacy/generate/regex.py b/outlines/v0_legacy/generate/regex.py
deleted file mode 100644
index 261f72037..000000000
--- a/outlines/v0_legacy/generate/regex.py
+++ /dev/null
@@ -1,67 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models.openai import OpenAI
-from outlines.types import Regex
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter, GeneratorVisionV0Adapter
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.samplers import Sampler, multinomial
-
-
-def regex(model, regex_str: str | Regex, sampler: Sampler = multinomial()):
-    """Generate structured text in the language of a regular expression.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    from outlines.types import Regex
-    generator = Generator(model, Regex(r"^[a-z]+$"))
-    ```
-
-    You can then call the generator created with a prompt to generate text
-    that matches the regular expression.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    regex_str:
-        The regular expression that the output must follow.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-    warnings.warn("""
-        The `regex` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        from outlines.types import Regex
-        generator = Generator(model, Regex(r'regex_str'))
-        ```
-        You can then call the generator created with a prompt to generate
-        text that matches the regular expression.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        raise NotImplementedError(
-            "Cannot use regex-structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API."
-        )
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, regex_str, sampler)
-    return GeneratorV0Adapter(model, regex_str, sampler)
diff --git a/outlines/v0_legacy/generate/text.py b/outlines/v0_legacy/generate/text.py
deleted file mode 100644
index 148cf131d..000000000
--- a/outlines/v0_legacy/generate/text.py
+++ /dev/null
@@ -1,64 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import Union
-
-from outlines.models.openai import OpenAI
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.samplers import Sampler, multinomial
-
-
-def text(model, sampler: Sampler = multinomial()) -> Union[
-    GeneratorV0Adapter, GeneratorVisionV0Adapter, OpenAI
-]:
-    """Generate unconstrained text.
-
-    This function is deprecated starting from v1.0.0. Do not use it.
-    Support for it will be removed in v1.1.0.
-    Use the `Generator` object instead:
-
-    ```python
-    from outlines import Generator
-    generator = Generator(model)
-    ```
-
-    You can then call the generator created with a prompt to generate text.
-
-    Parameters
-    ----------
-    model:
-        The Outlines v0 model to use to generate text.
-    sampler:
-        The sampler defining the sampling parameters.
-
-    Returns
-    -------
-    A `GeneratorV0Adapter` instance that can be called with a prompt to
-    generate text.
-
-    """
-    warnings.warn("""
-        The `text` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Use the `Generator` object instead:
-        ```python
-        from outlines import Generator
-        generator = Generator(model)
-        ```
-        You can then call the generator created with a prompt to generate
-        text.
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if isinstance(model, OpenAI):
-        return model
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(model, None, sampler)
-    return GeneratorV0Adapter(model, None, sampler)
diff --git a/outlines/v0_legacy/models/__init__.py b/outlines/v0_legacy/models/__init__.py
deleted file mode 100644
index 1dc012c37..000000000
--- a/outlines/v0_legacy/models/__init__.py
+++ /dev/null
@@ -1,21 +0,0 @@
-from .exllamav2 import ExLlamaV2Model, exl2
-from .llamacpp import llamacpp
-from .mlxlm import mlxlm
-from .openai import azure_openai, openai
-from .transformers import mamba, transformers
-from .transformers_vision import TransformersVision, transformers_vision
-from .vllm_offline import vllm
-
-__all__ = [
-    "ExLlamaV2Model",
-    "TransformersVision",
-    "azure_openai",
-    "exl2",
-    "llamacpp",
-    "mamba",
-    "mlxlm",
-    "openai",
-    "transformers",
-    "transformers_vision",
-    "vllm"
-]
diff --git a/outlines/v0_legacy/models/exllamav2.py b/outlines/v0_legacy/models/exllamav2.py
deleted file mode 100644
index 683512002..000000000
--- a/outlines/v0_legacy/models/exllamav2.py
+++ /dev/null
@@ -1,400 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import (
-    TYPE_CHECKING,
-    Dict,
-    Iterator,
-    List,
-    Optional,
-    Tuple,
-    TypedDict,
-    Union
-)
-
-if TYPE_CHECKING:
-    import torch.LongTensor
-    from exllamav2.generator import ExLlamaV2DynamicGenerator, ExLlamaV2Sampler  # type: ignore
-
-
-def exllamav2_params_adapter(sampling_params: dict, generation_params: dict, kwargs: dict) -> dict:
-    """Argument names mapping is done in the model class for the `exllamav2` model."""
-    kwargs.update(**sampling_params, **generation_params)
-    return kwargs
-
-
-class ExllamaV2Params(TypedDict, total=False):
-    stop_conditions: Optional[List[Union[int, str]]]
-    seed: Optional[int]
-    gen_settings: "ExLlamaV2Sampler.Settings"
-    max_new_tokens: List[int]
-
-
-class OutlinesExLlamaV2Tokenizer:
-    def __init__(self, tokenizer):
-        self.exl2_tokenizer = tokenizer
-        self.vocabulary = self.exl2_tokenizer.get_piece_to_id_dict()
-        self.special_tokens = set(self.exl2_tokenizer.extended_piece_to_id)
-        self.eos_token_id = self.exl2_tokenizer.eos_token_id
-
-    def convert_token_to_string(self, token):
-        return token
-
-    def decode(self, token_ids: "torch.LongTensor") -> List[str]:
-        import torch
-
-        decoded = self.exl2_tokenizer.decode(
-            torch.tensor(token_ids),
-            decode_special_tokens=False,
-        )
-        if isinstance(decoded, str):
-            return [decoded]
-        return decoded
-
-
-class ExLlamaV2Model():
-    """Represents a `exl2` model."""
-
-    def __init__(
-        self,
-        generator: "ExLlamaV2DynamicGenerator",
-        tokenizer: "OutlinesExLlamaV2Tokenizer",
-        max_seq_len: int,
-    ):
-        warnings.warn("""
-            The `ExLlamaV2Model` model is deprecated starting from v1.0.0.
-            Support for it will be removed in v1.1.0.
-            As the `exllamav2` library is not compatible with some key features
-            of `outlines`, we decided to remove support for it.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        self.generator = generator
-        self.tokenizer = tokenizer
-        self.max_seq_len = max_seq_len
-
-    def __call__(self, *args, **kwargs):
-        raise NotImplementedError(
-            "As a v0 model, the `exllamav2` model does not support the "
-            + "`__call__` method."
-        )
-
-    def stream(self, *args, **kwargs):
-        raise NotImplementedError(
-            "As a v0 model, the `exllamav2` model does not support the "
-            + "`stream` method."
-        )
-
-    def prepare_generation_parameters(
-        self,
-        prompts: Union[str, List[str]],
-        logits_processor,
-        kwargs
-    ) -> Tuple[Dict, Union[str, List[str]]]:
-        """Prepare the generation parameters.
-
-        `exllamav2` uses different default values
-
-        """
-        from exllamav2.generator import ExLlamaV2Sampler
-
-        exllamav2_params = {}
-
-        if isinstance(prompts, str):
-            prompts = [prompts]
-
-        max_tokens = kwargs.get("max_tokens")
-        if max_tokens is None:
-            max_tokens = []
-            for prompt in prompts:
-                ids = self.generator.tokenizer.encode(
-                    prompt, encode_special_tokens=True
-                )
-                prompt_tokens = ids.shape[-1]
-                max_tokens.append(self.max_seq_len - prompt_tokens)
-            exllamav2_params["max_new_tokens"] = max_tokens
-        else:
-            exllamav2_params["max_new_tokens"] = [
-                max_tokens for _ in range(len(prompts))
-            ]
-
-        stop_at = kwargs.get("stop_at")
-        stop_conditions = [self.generator.tokenizer.eos_token_id]
-        if isinstance(stop_at, str):
-            stop_conditions.append(stop_at)
-        elif isinstance(stop_at, list):
-            for stop_at in stop_at:
-                stop_conditions.append(stop_at)
-        exllamav2_params["stop_conditions"] = stop_conditions
-
-        exllamav2_params["seed"] = kwargs.get("seed")
-
-        gen_settings = ExLlamaV2Sampler.Settings()
-
-        temperature = kwargs.get("temperature")
-        if temperature is not None:
-            gen_settings.temperature = temperature
-
-        top_p = kwargs.get("top_p")
-        if top_p is not None:
-            gen_settings.top_p = top_p
-
-        top_k = kwargs.get("top_k")
-        if top_k is not None:
-            gen_settings.top_k = top_k
-
-        gen_settings.logits_processor = logits_processor
-
-        exllamav2_params["gen_settings"] = gen_settings
-
-        num_samples = kwargs.get("num_samples")
-        if num_samples is not None and num_samples > 1:
-            prompts = prompts * num_samples
-            exllamav2_params["max_new_tokens"] = (
-                exllamav2_params["max_new_tokens"] * num_samples
-            )
-
-        if len(prompts) == 1:
-            prompts = prompts[0]
-
-        return exllamav2_params, prompts
-
-    def reformat_output(
-        self, output: Union[str, List[str]], sampling_parameters
-    ):
-        """
-        The purpose of this function is to reformat the output from exllamav2's output format to outline's output format.
-
-        For exllamav2, it mainly accepts only a list or a string(they also do cfg sampling with tuples but we will ignore this for now).
-        The exllamav2's logic is:
-
-        1. If the prompt is a string, return a string. This is the same as outlines
-        2. If a prompt is a list, return a list. This is not the same as outlines output in that if the list is only one element, the string is expected to be outputted.
-        3. There is no such thing as num_samples, so the prompts had to be duplicated by num_samples times. Then, we had the function output a list of lists
-        """
-        if isinstance(output, str):
-            return output
-        if len(output) == 1:
-            return output[0]
-        if sampling_parameters.num_samples > 1:
-            if len(output) == sampling_parameters.num_samples:
-                return output
-            assert len(output) % sampling_parameters.num_samples == 0
-            num_items_per_sample = len(output) // sampling_parameters.num_samples
-            new_output = []
-            for i in range(sampling_parameters.num_samples):
-                curr_sample = []
-                for j in range(num_items_per_sample):
-                    curr_sample.append(output[i * num_items_per_sample + j])
-                new_output.append(curr_sample)
-            return new_output
-        return output
-
-    def generate(
-        self,
-        prompts: Union[str, List[str]],
-        logits_processor,
-        **kwargs
-    ) -> Union[str, List[str]]:
-        exllamav2_params, prompts = self.prepare_generation_parameters(
-            prompts,
-            logits_processor,
-            kwargs
-        )
-        """
-        In exllamav2, it needs the max amount of new tokens generated.
-        The reason exllamav2_params["max_new_tokens"] is a list is because in prepare_generation_parameters
-        the max amount of tokens that can be generated by the model for each prompt(by encoding with tokenizer) is calculated.
-        The minimum is picked because otherwise it might be possible for one of the
-        prompts to exceed the max sequence length.
-        """
-        output = self.generator.generate(
-            prompt=prompts,
-            gen_settings=exllamav2_params["gen_settings"],
-            max_new_tokens=min(exllamav2_params["max_new_tokens"]),
-            completion_only=True,
-            encode_special_tokens=True,
-            stop_conditions=exllamav2_params["stop_conditions"],
-            add_bos=False,
-            seed=exllamav2_params["seed"],
-        )
-
-        return self.reformat_output(output, kwargs)
-
-    def generate_stream(
-        self,
-        prompts: Union[str, List[str]],
-        logits_processor,
-        **kwargs
-    ) -> Iterator[Union[str, List[str]]]:
-        from exllamav2.generator import ExLlamaV2DynamicJob
-
-        exllamav2_params, prompts = self.prepare_generation_parameters(
-            prompts,
-            logits_processor,
-            kwargs
-        )
-
-        order = {}
-        if isinstance(prompts, str):
-            prompts = [prompts]
-        batch_size = len(prompts)
-        seed = exllamav2_params["seed"]
-        for idx, p in enumerate(prompts):
-            input_ids = self.generator.tokenizer.encode(
-                p, encode_special_tokens=True, add_bos=False
-            )
-
-            job = ExLlamaV2DynamicJob(
-                input_ids=input_ids,
-                max_new_tokens=exllamav2_params["max_new_tokens"][idx],
-                min_new_tokens=0,
-                seed=seed,
-                stop_conditions=exllamav2_params["stop_conditions"],
-                gen_settings=exllamav2_params["gen_settings"],
-                token_healing=False,
-                decode_special_tokens=False,
-            )
-
-            if seed is not None:
-                seed += 1
-
-            serial = self.generator.enqueue(job)
-            order[serial] = idx
-
-        # Collect outputs until all jobs finish
-
-        next_text = [""] * batch_size
-
-        def token_generator() -> Iterator[str]:
-            while self.generator.num_remaining_jobs():
-                results = self.generator.iterate()
-                for r in results:
-                    idx = order[r["serial"]]
-                    if r["stage"] == "streaming":
-                        text = r.get("text", "")
-                        next_text[idx] = text
-                    if r["eos"]:
-                        next_text[idx] = ""
-                yield self.reformat_output(next_text, kwargs)
-            return
-
-        return token_generator()
-
-
-def exl2(
-    model_path: str,
-    draft_model_path: Optional[str] = None,
-    max_seq_len: Optional[int] = None,
-    cache_q4: bool = False,
-    paged: bool = True,
-    max_chunk_size: Optional[int] = None,
-) -> ExLlamaV2Model:
-    """
-    Load an ExLlamaV2 model.
-
-    Parameters
-    ----------
-    model_path (str)
-        Path to the model directory.
-    device (str)
-        Device to load the model on. Pass in 'cuda' for GPU or 'cpu' for CPU
-    max_seq_len (Optional[int], optional)
-        Maximum sequence length. Defaults to None.
-    scale_pos_emb (Optional[float], optional)
-        Scale factor for positional embeddings. Defaults to None.
-    scale_alpha_value (Optional[float], optional)
-        Scale alpha value. Defaults to None.
-    no_flash_attn (Optional[bool], optional)
-        Disable flash attention. Defaults to None.
-    num_experts_per_token (Optional[int], optional)
-        Number of experts per token. Defaults to None.
-    cache_q4 (bool, optional)
-        Use Q4 cache. Defaults to False.
-    tokenizer_kwargs (dict, optional)
-        Additional keyword arguments for the tokenizer. Defaults to {}.
-    gpu_split (str)
-        \"auto\", or VRAM allocation per GPU in GB. Auto will use exllama's autosplit feature
-    low_mem (bool, optional)
-        Enable VRAM optimizations, potentially trading off speed
-    verbose (bool, optional)
-        Enable if you want debugging statements
-
-    Returns
-    -------
-    An `ExLlamaV2Model` instance.
-
-    Raises
-    ------
-    `ImportError` if the `exllamav2` library is not installed.
-
-    """
-    try:
-        from exllamav2 import (  # type: ignore
-            ExLlamaV2,
-            ExLlamaV2Cache,
-            ExLlamaV2Cache_Q4,
-            ExLlamaV2Config,
-            ExLlamaV2Tokenizer,
-        )
-        from exllamav2.generator import ExLlamaV2DynamicGenerator
-
-    except ImportError:
-        raise ImportError(
-            "The `exllamav2`, `transformers` and `torch` libraries needs to be installed in order to use `exllamav2` models. "
-            "Please run `pip install transformers torch git+https://github.com/lapp0/exllamav2@sampler-logits-processor` "
-            "Documentation: https://dottxt-ai.github.io/outlines/latest/reference/models/exllamav2/"
-        )
-    config = ExLlamaV2Config(model_path)
-    if max_chunk_size is not None:
-        config.max_input_len = max_chunk_size
-        config.max_attention_size = max_chunk_size**2
-
-    config.arch_compat_overrides()
-    model = ExLlamaV2(config)
-    if max_seq_len is None:
-        max_seq_len = -1
-    if cache_q4:
-        cache = ExLlamaV2Cache_Q4(model, max_seq_len=max_seq_len, lazy=True)
-    else:
-        cache = ExLlamaV2Cache(model, max_seq_len=max_seq_len, lazy=True)
-    model.load_autosplit(cache, progress=True)
-
-    print("Loading tokenizer...")
-    tokenizer = ExLlamaV2Tokenizer(config)
-    max_batch_size = 4 if paged else 1
-
-    draft_model = None
-    draft_cache = None
-    if draft_model_path is not None:
-        draft_config = ExLlamaV2Config(draft_model_path)
-        draft_model = ExLlamaV2(draft_config)
-
-        if cache_q4:
-            draft_cache = ExLlamaV2Cache_Q4(
-                draft_model, max_seq_len=max_seq_len, lazy=True
-            )
-        else:
-            draft_cache = ExLlamaV2Cache(
-                draft_model, max_seq_len=max_seq_len, lazy=True
-            )
-
-    # Initialize the generator with all default parameters
-    generator = ExLlamaV2DynamicGenerator(
-        model=model,
-        cache=cache,
-        draft_model=draft_model,
-        draft_cache=draft_cache,
-        tokenizer=tokenizer,
-        max_batch_size=max_batch_size,
-        use_ngram_draft=False,
-        max_chunk_size=max_chunk_size,
-        paged=paged,
-    )
-    max_seq_len = cache.max_seq_len
-
-    outlines_tokenizer = OutlinesExLlamaV2Tokenizer(tokenizer)
-    outlines_exl2_model = ExLlamaV2Model(generator, outlines_tokenizer, max_seq_len)
-    return outlines_exl2_model
diff --git a/outlines/v0_legacy/models/llamacpp.py b/outlines/v0_legacy/models/llamacpp.py
deleted file mode 100644
index 6c29630d5..000000000
--- a/outlines/v0_legacy/models/llamacpp.py
+++ /dev/null
@@ -1,147 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import Optional
-
-from outlines.models import LlamaCpp
-
-
-def llamacpp_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `llamacpp` model
-    to the new `Generator` API.
-    """
-    # generation_params
-    if (
-        generation_params.get("stop_at") is not None
-        and kwargs.get("stop") is None
-    ):
-        kwargs["stop"] = generation_params["stop_at"]
-
-    if (
-        generation_params.get("seed") is not None
-        and kwargs.get("seed") is None
-    ):
-        kwargs["seed"] = generation_params["seed"]
-
-    if (
-        generation_params.get("max_tokens") is not None
-        and kwargs.get("max_tokens") is None
-    ):
-        kwargs["max_tokens"] = generation_params["max_tokens"]
-    else:
-        kwargs["max_tokens"] = -1
-
-    # sampling_params
-    if sampling_params.get("sampler") == "beam_search":
-        raise NotImplementedError(
-            "The `llama_cpp_python` library does not support Beam Search."
-        )
-
-    if sampling_params.get("num_samples") != 1:
-        raise NotImplementedError(
-            "The `llama_cpp_python` library does not allow to take several "
-            + "samples."
-        )
-
-    if (
-        sampling_params.get("top_p") is not None
-        and kwargs.get("top_p") is None
-    ):
-        kwargs["top_p"] = sampling_params["top_p"]
-    else:
-        kwargs["top_p"] = 1.0
-
-    if (
-        sampling_params.get("top_k") is not None
-        and kwargs.get("top_k") is None
-    ):
-        kwargs["top_k"] = sampling_params["top_k"]
-    else:
-        kwargs["top_k"] = -1
-
-    if (
-        sampling_params.get("temperature") is not None
-        and kwargs.get("temperature") is None
-    ):
-        kwargs["temperature"] = sampling_params["temperature"]
-    else:
-        kwargs["temperature"] = 1.0
-
-    # model specific params
-    if kwargs.get("min_p") is None:
-        kwargs["min_p"] = 0.0
-
-    if kwargs.get("repeat_penalty") is None:
-        kwargs["repeat_penalty"] = 1.0
-
-    # The choice to stream or not should happen via the high-level API
-    kwargs.pop("stream", None)
-
-    return kwargs
-
-
-def llamacpp(
-    repo_id: str, filename: Optional[str] = None, **llamacpp_model_params
-) -> LlamaCpp:
-    """Load a model from the `llama-cpp-python` library.
-
-    We use the `Llama.from_pretrained` classmethod that downloads models
-    directly from the HuggingFace hub, instead of asking users to specify
-    a path to the downloaded model. One can still load a local model
-    by initializing `llama_cpp.Llama` directly.
-
-    Parameters
-    ----------
-    repo_id
-        The name of the model repository.
-    filename:
-        A filename of glob pattern to match the model file in the repo.
-    llama_cpp_model_params
-        Llama-specific model parameters. See the `llama-cpp-python` documentation
-        for the full list: https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama.__init__
-
-    """
-    from llama_cpp import Llama
-
-    warnings.warn("""
-        The `llamacpp` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `LlamaCpp` model with the
-        `outlines.from_llamacpp` function that takes a `llama_cpp.Llama`
-        instance as argument. For example:
-        ```python
-        from llama_cpp import Llama
-        from outlines import from_llamacpp
-        model = from_llamacpp(Llama.from_pretrained(
-            repo_id="<repo_id>",
-            filename="<filename>",
-        ))
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    # Default to using the model's full context length
-    if "n_ctx" not in llamacpp_model_params:
-        llamacpp_model_params["n_ctx"] = 0
-
-    if "verbose" not in llamacpp_model_params:
-        llamacpp_model_params["verbose"] = False
-
-    # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613 is resolved
-    if "tokenizer" not in llamacpp_model_params:
-        warnings.warn(
-            "The pre-tokenizer in `llama.cpp` handles unicode improperly "
-            + "(https://github.com/ggerganov/llama.cpp/pull/5613)\n"
-            + "Outlines may raise a `RuntimeError` when building the regex index.\n"
-            + "To circumvent this error when using `models.llamacpp()` you may pass the argument"
-            + "`tokenizer=llama_cpp.llama_tokenizer.LlamaHFTokenizer.from_pretrained(<hf_repo_id>)`\n"
-        )
-
-    model = Llama.from_pretrained(repo_id, filename, **llamacpp_model_params)
-
-    return LlamaCpp(model)
diff --git a/outlines/v0_legacy/models/mlxlm.py b/outlines/v0_legacy/models/mlxlm.py
deleted file mode 100644
index a69594173..000000000
--- a/outlines/v0_legacy/models/mlxlm.py
+++ /dev/null
@@ -1,153 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-from typing import Optional
-import warnings
-
-from outlines.models import MLXLM
-
-
-def mlxlm_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `mlxlm` model
-    to the new `Generator` API.
-    """
-    # generation_params
-    if generation_params.get("stop_at") is not None:
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support `stop_at`."
-        )
-
-    if generation_params.get("seed") is not None:
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support `seed`."
-        )
-
-    if (
-        generation_params.get("max_tokens") is not None
-        and kwargs.get("max_tokens") is None
-    ):
-        kwargs["max_tokens"] = generation_params["max_tokens"]
-    else:
-        kwargs["max_tokens"] = int(1e9)
-
-    # sampling_params
-    if sampling_params.get("sampler") == "beam_search":
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support Beam Search."
-        )
-
-    if sampling_params.get("num_samples") != 1:
-        raise NotImplementedError(
-            "The `mlx-lm` library does not allow to take several samples."
-        )
-
-    if (
-        sampling_params.get("top_k") is not None
-        or kwargs.get("top_k") is not None
-    ):
-        raise NotImplementedError(
-            "The `mlx-lm` library does not support top_k."
-        )
-
-    if (
-        sampling_params.get("top_p") is not None
-        or kwargs.get("top_p") is not None
-    ):
-        kwargs.pop("top_p", None)
-        warnings.warn(
-            """
-            The `top_p` parameter is not available anymore.
-            The argument will be ignored. Sorry for the inconvenience.
-            Please migrate to the v1 of Outlines.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-
-    if (
-        sampling_params.get("temperature") is not None
-        or kwargs.get("temperature") is not None
-    ):
-        kwargs.pop("temperature", None)
-        warnings.warn(
-            """
-            The `temperature` parameter is not available anymore.
-            The argument will be ignored. Sorry for the inconvenience.
-            Please migrate to the v1 of Outlines.
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-
-    return kwargs
-
-
-def mlxlm(
-    model_name: str,
-    tokenizer_config: dict = {},
-    model_config: dict = {},
-    adapter_path: Optional[str] = None,
-    lazy: bool = False,
-):
-    """Instantiate a model from the `mlx_lm` library and its tokenizer.
-
-    Signature adapted from
-    https://github.com/ml-explore/mlx-examples/blob/4872727/llms/mlx_lm/utils.py#L422
-
-    Parameters
-    ----------
-    Args:
-        path_or_hf_repo (Path): The path or the huggingface repository to load the model from.
-        tokenizer_config (dict, optional): Configuration parameters specifically for the tokenizer.
-            Defaults to an empty dictionary.
-        model_config(dict, optional): Configuration parameters specifically for the model.
-            Defaults to an empty dictionary.
-        adapter_path (str, optional): Path to the LoRA adapters. If provided, applies LoRA layers
-            to the model. Default: ``None``.
-        lazy (bool): If False eval the model parameters to make sure they are
-            loaded in memory before returning, otherwise they will be loaded
-            when needed. Default: ``False``
-
-    Returns
-    -------
-    A `MLXLM` model instance.
-
-    """
-    warnings.warn("""
-        The `mlxlm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `MLXLM` model with the
-        `outlines.from_mlxlm` function that takes a mlxlm model
-        and a tokenizer as arguments. For example:
-        ```python
-        from mlx_lm import load
-        from outlines import from_mlxlm
-        model = from_mlxlm(
-            load("<model_name>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    try:
-        import mlx.core as mx
-        import mlx_lm
-    except ImportError:
-        raise ImportError(
-            "The `mlx_lm` library needs to be installed in order to use `mlx_lm` models."
-        )
-    if not mx.metal.is_available():
-        raise RuntimeError("You cannot use `mlx_lm` without Apple Silicon (Metal)")
-
-    model, tokenizer = mlx_lm.load(
-        model_name,
-        tokenizer_config=tokenizer_config,
-        model_config=model_config,
-        adapter_path=adapter_path,
-        lazy=lazy,
-    )
-    return MLXLM(model, tokenizer)
diff --git a/outlines/v0_legacy/models/openai.py b/outlines/v0_legacy/models/openai.py
deleted file mode 100644
index 46546b003..000000000
--- a/outlines/v0_legacy/models/openai.py
+++ /dev/null
@@ -1,64 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import functools
-from dataclasses import replace
-from typing import Optional
-
-from outlines.models.openai import OpenAI, OpenAIConfig
-
-
-@functools.singledispatch
-def openai(model_or_client, *args, **kwargs):
-    if len(args) == 1 and isinstance(args[0], OpenAIConfig):
-        return OpenAI(model_or_client, args[0], **kwargs)
-    elif kwargs.get("config"):
-        return OpenAI(model_or_client, kwargs.pop("config"), **kwargs)
-    else:
-        return OpenAI(model_or_client, OpenAIConfig(), **kwargs)
-
-
-@openai.register(str)
-def openai_model(
-    model_name: str,
-    config: Optional[OpenAIConfig] = None,
-    **openai_client_params,
-):
-    try:
-        from openai import AsyncOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' OpenAI integration."
-        )
-
-    if config is not None:
-        config = replace(config, model=model_name)  # type: ignore
-    else:
-        config = OpenAIConfig(model=model_name)
-
-    client = AsyncOpenAI(**openai_client_params)
-
-    return OpenAI(client, config)
-
-
-def azure_openai(
-    deployment_name: str,
-    model_name: Optional[str] = None,
-    config: Optional[OpenAIConfig] = None,
-    **azure_openai_client_params,
-):
-    try:
-        from openai import AsyncAzureOpenAI
-    except ImportError:
-        raise ImportError(
-            "The `openai` library needs to be installed in order to use Outlines' Azure OpenAI integration."
-        )
-
-    if config is not None:
-        config = replace(config, model=deployment_name)  # type: ignore
-    if config is None:
-        config = OpenAIConfig(model=deployment_name)
-
-    client = AsyncAzureOpenAI(**azure_openai_client_params)
-
-    return OpenAI(client, config)
diff --git a/outlines/v0_legacy/models/transformers.py b/outlines/v0_legacy/models/transformers.py
deleted file mode 100644
index eb936c491..000000000
--- a/outlines/v0_legacy/models/transformers.py
+++ /dev/null
@@ -1,196 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-from typing import Optional
-
-from outlines.models import Transformers
-
-
-def transformers_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `transformers` model
-    to the new `Generator` API.
-    """
-    from transformers import set_seed
-
-    # generation_params
-    if (
-        generation_params.get("stop_at") is not None
-        and kwargs.get("stop_strings") is None
-    ):
-        kwargs["stop_strings"] = generation_params["stop_at"]
-
-    if (
-        generation_params.get("max_tokens") is not None
-        and kwargs.get("max_new_tokens") is None
-    ):
-        kwargs["max_new_tokens"] = generation_params["max_tokens"]
-    else:
-        kwargs["max_new_tokens"] = int(2**30)
-
-    if (
-        generation_params.get("seed") is not None
-        or kwargs.get("seed") is not None
-    ):
-        seed = (
-            generation_params.get("seed")
-            if generation_params.get("seed") is not None
-            else kwargs.get("seed")
-        )
-        set_seed(seed) # global seed
-
-    # sampling_params
-    if sampling_params.get("sampler") == "multinomial":
-        kwargs["do_sample"] = True
-    elif sampling_params.get("sampler") == "beam_search":
-        kwargs["num_beams"] = sampling_params.get("num_samples", 1)
-
-    if (
-        sampling_params.get("num_samples") is not None
-        and kwargs.get("num_return_sequences") is None
-    ):
-        kwargs["num_return_sequences"] = sampling_params["num_samples"]
-    else:
-        kwargs["num_return_sequences"] = 1
-
-    if (
-        sampling_params.get("top_p") is not None
-        and kwargs.get("top_p") is None
-    ):
-        kwargs["top_p"] = sampling_params["top_p"]
-
-    if (
-        sampling_params.get("top_k") is not None
-        and kwargs.get("top_k") is None
-    ):
-        kwargs["top_k"] = sampling_params["top_k"]
-
-    if (
-        sampling_params.get("temperature") is not None
-        and kwargs.get("temperature") is None
-    ):
-        kwargs["temperature"] = sampling_params["temperature"]
-
-    return kwargs
-
-
-def transformers(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-    model_class=None,
-    tokenizer_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
-
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    tokenizer_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the tokenizer.
-
-    Returns
-    -------
-    A `TransformersModel` model instance.
-
-    """
-    warnings.warn("""
-        The `transformers` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `Transformers` model with the
-        `outlines.from_transformers` function that takes a transformers model
-        and a tokenizer as arguments. For example:
-        ```python
-        from transformers import AutoModelForCausalLM, AutoTokenizer
-        from outlines import from_transformers
-        model = from_transformers(
-            AutoModelForCausalLM.from_pretrained("<model_name>"),
-            AutoTokenizer.from_pretrained("<model_name>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if model_class is None or tokenizer_class is None:
-        try:
-            from transformers import AutoModelForCausalLM, AutoTokenizer
-        except ImportError:  # pragma: no cover
-            raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
-            )
-    if model_class is None:
-        model_class = AutoModelForCausalLM
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
-
-    if device is not None:
-        model_kwargs["device_map"] = device
-
-    model = model_class.from_pretrained(model_name, **model_kwargs)
-
-    tokenizer_kwargs.setdefault("padding_side", "left")
-    tokenizer = tokenizer_class.from_pretrained(model_name, **tokenizer_kwargs)
-
-    return Transformers(model, tokenizer)
-
-
-def mamba(
-    model_name: str,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    tokenizer_kwargs: dict = {},
-):
-    warnings.warn("""
-        The `mamba` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `Transformers` model with the
-        `outlines.from_transformers` function that takes a transformers model
-        and a tokenizer as arguments. For Mamba, you would use a mamba model
-        and a mamba tokenizer. For example:
-        ```python
-        from transformers import MambaForCausalLM, MambaTokenizer
-        from outlines import from_transformers
-        model = from_transformers(
-            MambaForCausalLM.from_pretrained("<repo_id>"),
-            MambaTokenizer.from_pretrained("<repo_id>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    try:
-        from transformers import MambaForCausalLM
-
-    except ImportError:  # pragma: no cover
-        raise ImportError(
-            "The `mamba_ssm`, `torch` and `transformer` libraries needs to be installed in order to use Mamba."
-        )
-
-    # We do not want to raise the deprecation warning through the transformers function
-    with warnings.catch_warnings():
-        warnings.filterwarnings(
-            "ignore",
-            category=DeprecationWarning,
-        )
-        return transformers(
-            model_name=model_name,
-            device=device,
-            model_kwargs=model_kwargs,
-            tokenizer_kwargs=tokenizer_kwargs,
-            model_class=MambaForCausalLM,
-    )
diff --git a/outlines/v0_legacy/models/transformers_vision.py b/outlines/v0_legacy/models/transformers_vision.py
deleted file mode 100644
index 33c78c8b7..000000000
--- a/outlines/v0_legacy/models/transformers_vision.py
+++ /dev/null
@@ -1,107 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-from typing import TYPE_CHECKING, Optional
-import warnings
-
-from outlines.models.transformers import (
-    Transformers,
-    TransformersMultiModalTypeAdapter,
-)
-
-
-class TransformersVision(Transformers):
-    def __init__(self, model, tokenizer, processor):
-        self.processor = processor
-        self.processor.padding_side = "left"
-        self.processor.pad_token = "[PAD]"
-
-        super().__init__(model, tokenizer)
-
-        self.type_adapter = TransformersMultiModalTypeAdapter()
-
-
-    def _prepare_model_inputs(self, model_input, is_batch=False):
-        model_input = self.type_adapter.format_input(model_input)
-        inputs = self.processor(
-            **model_input, padding=True, return_tensors="pt"
-        ).to(self.model.device)
-
-        return model_input["text"], inputs
-
-
-def transformers_vision(
-    model_name: str,
-    model_class,
-    device: Optional[str] = None,
-    model_kwargs: dict = {},
-    processor_kwargs: dict = {},
-    tokenizer_class=None,
-    processor_class=None,
-):
-    """Instantiate a model from the `transformers` library and its tokenizer.
-
-    Parameters
-    ----------
-    model_name
-        The name of the model as listed on Hugging Face's model page.
-    model_class
-        The `PreTrainedModel` class from transformers to use in initializing the vision model from `model_name`.
-        https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel
-    device
-        The device(s) on which the model should be loaded. This overrides
-        the `device_map` entry in `model_kwargs` when provided.
-    model_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the model.
-    processor_kwargs
-        A dictionary that contains the keyword arguments to pass to the
-        `from_pretrained` method when loading the processor.
-
-    Returns
-    -------
-    A `TransformersModel` model instance.
-
-    """
-    warnings.warn("""
-        The `transformers_vision` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `TransformersMultiModal` model with
-        the `outlines.from_transformers` function that takes a transformers
-        model and a processor as arguments. For example:
-        ```python
-        from transformers import AutoModelForCausalLM, AutoProcessor
-        from outlines import from_transformers
-        model = from_transformers(
-            AutoModelForCausalLM.from_pretrained("<model_name>"),
-            AutoProcessor.from_pretrained("<model_name>")
-        )
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    if processor_class is None or tokenizer_class is None:
-        try:
-            from transformers import AutoProcessor, AutoTokenizer
-        except ImportError:  # pragma: no cover
-            raise ImportError(
-                "The `transformers` library needs to be installed in order to use `transformers` models."
-            )
-    if processor_class is None:
-        processor_class = AutoProcessor
-    if tokenizer_class is None:
-        tokenizer_class = AutoTokenizer
-
-    if device is not None:
-        model_kwargs["device_map"] = device
-
-    model = model_class.from_pretrained(model_name, **model_kwargs)
-
-    processor_kwargs.setdefault("padding_side", "left")
-    processor_kwargs.setdefault("pad_token", "[PAD]")
-    processor = processor_class.from_pretrained(model_name, **processor_kwargs)
-    tokenizer = tokenizer_class.from_pretrained(model_name, **processor_kwargs)
-
-    return TransformersVision(model, tokenizer, processor)
diff --git a/outlines/v0_legacy/models/vllm_offline.py b/outlines/v0_legacy/models/vllm_offline.py
deleted file mode 100644
index 8efb7cfc0..000000000
--- a/outlines/v0_legacy/models/vllm_offline.py
+++ /dev/null
@@ -1,102 +0,0 @@
-### Legacy API: file exists for backward compatibility with v0
-### Do not use it as an example of how to write new code
-
-import warnings
-
-from outlines.models import VLLMOffline
-
-
-def vllm_offline_params_adapter(
-    sampling_params: dict, generation_params: dict, kwargs: dict
-) -> dict:
-    """Adapt the parameters of the legacy generator for the `vllm` model
-    to the new `Generator` API.
-    """
-    from vllm.sampling_params import SamplingParams
-
-    # vllm has a `sampling_params` argument that is a `SamplingParams` object
-    # this is unrelated to the `sampling_params` argument in the legacy
-    # `Generator` API
-    sampling_params_kwarg = kwargs.get("sampling_params", SamplingParams())
-
-    # generation_params
-    if generation_params.get("stop_at") is not None:
-        if isinstance(generation_params["stop_at"], str):
-            sampling_params_kwarg.stop_strings = [generation_params["stop_at"]]
-        else:
-            sampling_params_kwarg.stop_strings = generation_params["stop_at"]
-
-    if generation_params.get("seed") is not None:
-        sampling_params_kwarg.seed = generation_params["seed"]
-
-    if generation_params.get("max_tokens") is not None:
-        sampling_params_kwarg.max_tokens = generation_params["max_tokens"]
-
-    # sampling_params
-    if sampling_params.get("sampler") == "beam_search":
-        sampling_params_kwarg.use_beam_search = True
-
-    if (
-        sampling_params.get("num_samples") is not None
-        and sampling_params_kwarg.n == 1
-    ):
-        sampling_params_kwarg.n = sampling_params["num_samples"]
-        sampling_params_kwarg.best_of = sampling_params["num_samples"]
-
-    if (
-        sampling_params.get("top_p") is not None
-        and sampling_params_kwarg.top_p == 1.0
-    ):
-        sampling_params_kwarg.top_p = sampling_params["top_p"]
-
-    if (
-        sampling_params.get("top_k") is not None
-        and sampling_params_kwarg.top_k == -1
-    ):
-        sampling_params_kwarg.top_k = sampling_params["top_k"]
-        if sampling_params.get("top_k") == 1:
-            sampling_params_kwarg.repetition_penalty = 0
-
-    if (
-        sampling_params.get("temperature") is not None
-        and sampling_params_kwarg.temperature == 1.0
-    ):
-        sampling_params_kwarg.temperature = sampling_params["temperature"]
-
-    return {"sampling_params": sampling_params_kwarg, **kwargs}
-
-
-def vllm(model_name: str, **vllm_model_params):
-    """Load an offline vLLM model.
-
-    Parameters
-    ---------
-    model_name
-        The name of the model to load from the HuggingFace hub.
-    vllm_model_params
-        vLLM-specific model parameters. See the vLLM code for the full list:
-        https://github.com/vllm-project/vllm/blob/main/vllm/entrypoints/llm.py
-
-    """
-    warnings.warn("""
-        The `vllm` function is deprecated starting from v1.0.0.
-        Do not use it. Support for it will be removed in v1.1.0.
-        Instead, you should instantiate a `VLLMOffline` model with the
-        `outlines.from_vllm_offline` function that takes a vLLM model name as
-        argument.
-        For example:
-        ```python
-        from vllm import LLM
-        from outlines import from_vllm_offline
-        model = from_vllm_offline(LLM("<model_name>"))
-        ```
-        """,
-        DeprecationWarning,
-        stacklevel=2,
-    )
-
-    from vllm import LLM
-
-    model = LLM(model_name, **vllm_model_params)
-
-    return VLLMOffline(model)
diff --git a/outlines/v0_legacy/samplers.py b/outlines/v0_legacy/samplers.py
deleted file mode 100644
index 9f1ff5cc6..000000000
--- a/outlines/v0_legacy/samplers.py
+++ /dev/null
@@ -1,118 +0,0 @@
-from dataclasses import dataclass
-from typing import Optional, Protocol
-
-
-@dataclass(frozen=True)
-class SamplingParameters:
-    """Sampling parameters available in Outlines."""
-    sampler: str
-    num_samples: int = 1
-    top_p: Optional[float] = None
-    top_k: Optional[int] = None
-    temperature: Optional[float] = None
-
-
-class Sampler(Protocol):
-    """
-    Protocol for all samplers.
-
-    As outlines now delegates the handling of the generation loop
-    to the models, the samplers are only used to define the sampling
-    parameters that are passed to the model.
-    """
-
-    samples: int
-
-    def sampling_params(self) -> SamplingParameters:
-        ...
-
-
-class GreedySampler:
-    """Greedy Sampling algorithm.
-
-    Greedy sampling consists in choosing the token with the largest
-    likelihood at every step.
-
-    We don't allow more than one sample. We could attribute this a meaning, for
-    instance the k-th sample represents the k-th most likely token. In which
-    case it would be equivalent to beam search without the sequence weights.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(self):
-        self.samples = 1
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("greedy", self.samples, None, None, 0.0)
-
-
-greedy = GreedySampler
-
-
-class MultinomialSampler:
-    """Multinomial sampling algorithm.
-
-    Multinomial sampling consists in randomly sampling the next token assuming
-    its distribution is a Categorical distribution parametrized by the
-    next-token logits.
-
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence.
-
-    """
-
-    def __init__(
-        self,
-        samples: int = 1,
-        *,
-        top_k: Optional[int] = None,
-        top_p: Optional[float] = None,
-        temperature: Optional[float] = None,
-    ):
-        self.samples = samples
-        self.top_k = top_k
-        self.top_p = top_p
-        self.temperature = temperature
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters(
-            "multinomial",
-            self.samples,
-            self.top_p,
-            self.top_k,
-            self.temperature,
-        )
-
-
-multinomial = MultinomialSampler
-
-
-class BeamSearchSampler:
-    """Beam Search sampling algorithm.
-
-    Attributes
-    ----------
-    samples
-        The number of samples taken for each input sequence. Equivalent to the
-        number of beams.
-    """
-
-    def __init__(self, beams: int = 1):
-        self.samples = beams
-
-    @property
-    def sampling_params(self):
-        return SamplingParameters("beam_search", self.samples, None, None, 1.0)
-
-
-beam_search = BeamSearchSampler

From f3b3f3d9c4e15b3dcb2f64b460359e560fea5244 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Jul 2025 16:38:52 +0200
Subject: [PATCH 403/505] Remove the tests related to the v0 deprecated
 features

---
 tests/models/test_sglang.py                   |   8 +-
 tests/models/test_vllm.py                     |   8 +-
 tests/test_templates.py                       | 288 +++---------------
 tests/types/test_dsl.py                       |  95 +++---
 tests/v0_legacy/generate/test_api_legacy.py   | 187 ------------
 tests/v0_legacy/generate/test_cfg_legacy.py   | 111 -------
 .../v0_legacy/generate/test_choice_legacy.py  |  97 ------
 .../v0_legacy/generate/test_format_legacy.py  |  93 ------
 tests/v0_legacy/generate/test_fsm_legacy.py   | 104 -------
 tests/v0_legacy/generate/test_json_legacy.py  | 200 ------------
 tests/v0_legacy/generate/test_regex_legacy.py |  90 ------
 tests/v0_legacy/generate/test_text_legacy.py  |  83 -----
 .../v0_legacy/models/test_exllamav2_legacy.py |  76 -----
 .../v0_legacy/models/test_llamacpp_legacy.py  | 169 ----------
 tests/v0_legacy/models/test_mlxlm_legacy.py   | 193 ------------
 tests/v0_legacy/models/test_openai_legacy.py  | 159 ----------
 .../models/test_transformers_legacy.py        | 173 -----------
 .../models/test_transformers_vision_legacy.py | 112 -------
 .../models/test_vllm_offline_legacy.py        | 181 -----------
 tests/v0_legacy/test_base_legacy.py           | 275 -----------------
 tests/v0_legacy/test_function_legacy.py       | 192 ------------
 tests/v0_legacy/test_samplers_legacy.py       |  60 ----
 22 files changed, 103 insertions(+), 2851 deletions(-)
 delete mode 100644 tests/v0_legacy/generate/test_api_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_cfg_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_choice_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_format_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_fsm_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_json_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_regex_legacy.py
 delete mode 100644 tests/v0_legacy/generate/test_text_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_exllamav2_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_llamacpp_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_mlxlm_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_openai_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_transformers_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_transformers_vision_legacy.py
 delete mode 100644 tests/v0_legacy/models/test_vllm_offline_legacy.py
 delete mode 100644 tests/v0_legacy/test_base_legacy.py
 delete mode 100644 tests/v0_legacy/test_function_legacy.py
 delete mode 100644 tests/v0_legacy/test_samplers_legacy.py

diff --git a/tests/models/test_sglang.py b/tests/models/test_sglang.py
index 5c8d9be1e..741a013fe 100644
--- a/tests/models/test_sglang.py
+++ b/tests/models/test_sglang.py
@@ -42,8 +42,8 @@
     "SGLANG_MODEL_NAME", "qwen/qwen2.5-0.5b-instruct"
 )
 if sglang_server_url:
-    openai_client = OpenAI(base_url=sglang_server_url)
-    async_openai_client = AsyncOpenAI(base_url=sglang_server_url)
+    openai_client = OpenAI(base_url=sglang_server_url, api_key="foo")
+    async_openai_client = AsyncOpenAI(base_url=sglang_server_url, api_key="foo")
 else:
     warnings.warn("No SGLang server URL provided, using mock server")
     openai_client = MockOpenAIClient()
@@ -198,8 +198,8 @@ def async_model_no_model_name():
 def test_sglang_init():
     # We do not rely on the mock server here because we need an object
     # of type OpenAI and AsyncOpenAI to test the init function.
-    openai_client = OpenAI(base_url="http://localhost:11434")
-    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
+    openai_client = OpenAI(base_url="http://localhost:11434", api_key="foo")
+    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434", api_key="foo")
 
     # Sync with model name
     model = from_sglang(openai_client, sglang_model_name)
diff --git a/tests/models/test_vllm.py b/tests/models/test_vllm.py
index c82e44e21..cd967a3cb 100644
--- a/tests/models/test_vllm.py
+++ b/tests/models/test_vllm.py
@@ -39,8 +39,8 @@
     "VLLM_MODEL_NAME", "Qwen/Qwen2.5-VL-3B-Instruct"
 )
 if vllm_server_url:
-    openai_client = OpenAI(base_url=vllm_server_url)
-    async_openai_client = AsyncOpenAI(base_url=vllm_server_url)
+    openai_client = OpenAI(base_url=vllm_server_url, api_key="foo")
+    async_openai_client = AsyncOpenAI(base_url=vllm_server_url, api_key="foo")
 else:
     warnings.warn("No VLLM server URL provided, using mock server")
     openai_client = MockOpenAIClient()
@@ -192,8 +192,8 @@ def async_model_no_model_name():
 def test_vllm_init():
     # We do not rely on the mock server here because we need an object
     # of type OpenAI and AsyncOpenAI to test the init function.
-    openai_client = OpenAI(base_url="http://localhost:11434")
-    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434")
+    openai_client = OpenAI(base_url="http://localhost:11434", api_key="foo")
+    async_openai_client = AsyncOpenAI(base_url="http://localhost:11434", api_key="foo")
 
     # Sync with model name
     model = from_vllm(openai_client, vllm_model_name)
diff --git a/tests/test_templates.py b/tests/test_templates.py
index 225ad21ca..17787ee6b 100644
--- a/tests/test_templates.py
+++ b/tests/test_templates.py
@@ -1,14 +1,13 @@
 import base64
 import os
 import tempfile
-from typing import Dict, List
+from typing import Optional
 
 import pytest
 from PIL import Image as PILImage
 from io import BytesIO
 from pydantic import BaseModel, Field
 
-import outlines
 from outlines.inputs import Image
 from outlines.templates import (
     Template,
@@ -77,8 +76,8 @@ def test_vision_invalid_image_format():
             Vision(prompt="Test prompt", image=image)
 
 
-def render(content: str, **kwargs):
-    template = build_template_from_string(content)
+def render(content: str, filters: Optional[dict] = None, **kwargs):
+    template = build_template_from_string(content, filters or {})
     return template.render(kwargs)
 
 
@@ -180,251 +179,58 @@ def test_render_jinja():
     assert render(tpl, is_true=False) == "final"
 
 
-def test_prompt_basic():
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def test_tpl(variable):
-            """{{variable}} test"""
-
-        assert list(test_tpl.signature.parameters) == ["variable"]
-
-        with pytest.raises(TypeError):
-            test_tpl(v="test")
-
-        p = test_tpl("test")
-        assert p == "test test"
-
-        p = test_tpl(variable="test")
-        assert p == "test test"
-
-        @outlines.prompt
-        def test_single_quote_tpl(variable):
-            "${variable} test"
-
-        assert list(test_single_quote_tpl.signature.parameters) == ["variable"]
-
-        p = test_tpl("test")
-        assert p == "test test"
-
-
-def test_prompt_kwargs():
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def test_kwarg_tpl(var, other_var="other"):
-            """{{var}} and {{other_var}}"""
-
-        assert list(test_kwarg_tpl.signature.parameters) == ["var", "other_var"]
-
-        p = test_kwarg_tpl("test")
-        assert p == "test and other"
-
-        p = test_kwarg_tpl("test", other_var="kwarg")
-        assert p == "test and kwarg"
-
-        p = test_kwarg_tpl("test", "test")
-        assert p == "test and test"
-
-
-def test_no_prompt():
-    with pytest.deprecated_call():
-        with pytest.raises(TypeError, match="template"):
-
-            @outlines.prompt
-            def test_empty(variable):
-                pass
-
-        with pytest.raises(TypeError, match="template"):
-
-            @outlines.prompt
-            def test_only_code(variable):
-                return variable
-
-
-def test_prompt_function():
-    def empty_fn():
-        pass
-
-    def with_description():
-        """A description.
-
-        But this is ignored.
-        """
-        pass
-
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def name_description_ppt(fn):
-            """
-            {{fn|name}}: {{fn|description}}
-            """
-
-        rendered = name_description_ppt(empty_fn)
-        assert rendered == "empty_fn: "
-
-        rendered = name_description_ppt(with_description)
-        assert rendered == "with_description: A description."
-
-    def with_signature(one: int, two: List[str], three: float = 1.0):
-        pass
-
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def name_signature_ppt(fn):
-            """
-            {{fn|name}}: {{fn|signature}}
-            """
-
-        rendered = name_signature_ppt(with_signature)
-        assert (
-            rendered == "with_signature: one: int, two: List[str], three: float = 1.0"
-        )
-
-    def test_function_call(one, two=2):
-        return one + two
-
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def source_ppt(fn):
-            """
-            {{fn|source}}
-            """
-
-        rendered = source_ppt(test_function_call)
-        assert rendered == "def test_function_call(one, two=2):\n    return one + two\n"
-
-
-def test_prompt_pydantic_response():
-    class SimpleResponse(BaseModel):
-        one: str = Field(description="a description")
-        two: str
+def test_render_filters():
+    def foo(bar: str) -> str:
+        """This is a sample function."""
+        return bar
 
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def source_ppt(model):
-            "{{model | schema }}"
-
-        prompt = source_ppt(SimpleResponse)
-        assert prompt == '{\n  "one": "a description",\n  "two": "<two>"\n}'
-
-        class NestedResponse(BaseModel):
-            answer: str
-            thought: SimpleResponse
-
-        prompt = source_ppt(NestedResponse)
-        assert (
-            prompt
-            == '{\n  "answer": "<answer>",\n  "thought": {\n    "one": "a description",\n    "two": "<two>"\n  }\n}'
-        )
-
-        class ConvolutedResponse(BaseModel):
-            part_one: NestedResponse
-            part_two: SimpleResponse
-
-        prompt = source_ppt(ConvolutedResponse)
-        assert (
-            prompt
-            == '{\n  "part_one": {\n    "answer": "<answer>",\n    "thought": {\n      "one": "a description",\n      "two": "<two>"\n    }\n  },\n  "part_two": {\n    "one": "a description",\n    "two": "<two>"\n  }\n}'
-        )
-
-
-def test_prompt_dict_response():
-    response = {"one": "a description", "two": ""}
-
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def source_ppt(model):
-            "{{model | schema }}"
-
-        prompt = source_ppt(response)
-        assert prompt == '{\n  "one": "a description",\n  "two": ""\n}'
-
-
-def test_prompt_args():
-    def no_args():
-        pass
-
-    def with_args(x, y, z):
-        pass
-
-    def with_annotations(x: bool, y: str, z: Dict[int, List[str]]):
-        pass
-
-    def with_defaults(x=True, y="Hi", z={4: ["I", "love", "outlines"]}):
-        pass
-
-    def with_annotations_and_defaults(
-        x: bool = True,
-        y: str = "Hi",
-        z: Dict[int, List[str]] = {4: ["I", "love", "outlines"]},
-    ):
-        pass
-
-    def with_all(
-        x1,
-        y1,
-        z1,
-        x2: bool,
-        y2: str,
-        z2: Dict[int, List[str]],
-        x3=True,
-        y3="Hi",
-        z3={4: ["I", "love", "outlines"]},
-        x4: bool = True,
-        y4: str = "Hi",
-        z4: Dict[int, List[str]] = {4: ["I", "love", "outlines"]},
-    ):
-        pass
+    class PydanticClass(BaseModel):
+        foo: str = Field(description="bar")
 
-    with pytest.deprecated_call():
-
-        @outlines.prompt
-        def args_prompt(fn):
-            """args: {{ fn | args }}"""
-
-        assert args_prompt(no_args) == "args: "
-        assert args_prompt(with_args) == "args: x, y, z"
-        assert (
-            args_prompt(with_annotations)
-            == "args: x: bool, y: str, z: Dict[int, List[str]]"
-        )
-        assert (
-            args_prompt(with_defaults)
-            == "args: x=True, y='Hi', z={4: ['I', 'love', 'outlines']}"
-        )
-        assert (
-            args_prompt(with_annotations_and_defaults)
-            == "args: x: bool = True, y: str = 'Hi', z: Dict[int, List[str]] = {4: ['I', 'love', 'outlines']}"
-        )
-        assert (
-            args_prompt(with_all)
-            == "args: x1, y1, z1, x2: bool, y2: str, z2: Dict[int, List[str]], x3=True, y3='Hi', z3={4: ['I', 'love', 'outlines']}, x4: bool = True, y4: str = 'Hi', z4: Dict[int, List[str]] = {4: ['I', 'love', 'outlines']}"
-        )
+    def custom_filter(x: str) -> str:
+        return x.upper()
 
+    # name filter
+    tpl = """
+    {{ func | name }}
+    """
+    assert render(tpl, func=foo) == "foo"
 
-def test_prompt_with_additional_filters():
-    def reverse(s: str) -> str:
-        return s[::-1]
+    # description filter
+    tpl = """
+    {{ func | description }}
+    """
+    assert render(tpl, func=foo) == "This is a sample function."
 
-    with pytest.deprecated_call():
+    # source filter
+    tpl = """
+    {{ func | source }}
+    """
+    assert render(tpl, func=foo) == 'def foo(bar: str) -> str:\n    """This is a sample function."""\n    return bar\n'
 
-        @outlines.prompt(filters=dict(reverse=reverse))
-        def test_tpl(variable):
-            """{{ variable | reverse }} test"""
+    # signature filter
+    tpl = """
+    {{ func | signature }}
+    """
+    assert render(tpl, func=foo) == "bar: str"
 
-        assert list(test_tpl.signature.parameters) == ["variable"]
+    # args filter
+    tpl = """
+    {{ func | args }}
+    """
+    assert render(tpl, func=foo) == "bar: str"
 
-        p = test_tpl("test")
-        assert p == "tset test"
+    # schema filter
+    tpl = """
+    {{ schema | schema }}
+    """
+    assert render(tpl, schema=PydanticClass) == '{\n  "foo": "bar"\n}'
 
-        p = test_tpl(variable="example")
-        assert p == "elpmaxe test"
+    # custom filters
+    tpl = """
+    {{ name | custom_filter }}
+    """
+    assert render(tpl, {"custom_filter": custom_filter}, name="John") == "JOHN"
 
 
 @pytest.fixture
@@ -470,7 +276,6 @@ def temp_prompt_file():
 
 def test_prompt_from_file(temp_prompt_file):
     prompt = Template.from_file(temp_prompt_file)
-    assert prompt.signature is None
     examples = [
         {"question": "What is the capital of France?", "answer": "Paris"},
         {"question": "What is 2 + 2?", "answer": "4"},
@@ -497,7 +302,6 @@ def test_prompt_from_str():
     Hello, {{ name }}!
     """
     prompt = Template.from_string(content)
-    assert prompt.signature is None
     assert prompt(name="World") == "Hello, World!"
 
 
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index e07af09c7..0c4db0a44 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -53,9 +53,7 @@
     exactly,
     regex,
     python_types_to_terms,
-    # deprecated
-    repeat,
-    times,
+    to_regex,
 )
 
 if sys.version_info >= (3, 12):
@@ -343,49 +341,6 @@ def test_dsl_aliases():
     assert isinstance(test, JsonSchema)
 
 
-def test_dsl_repeat():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `repeat` function/method is deprecated",
-    ):
-        a = String("a")
-
-        # error missing min_count or max_count
-        with pytest.raises(
-            ValueError,
-            match="You must provide a value for at least `min_count` or "
-            "`max_count`",
-        ):
-            repeat(a, None, None)
-
-        # at_least
-        assert repeat(a, 2, None) == at_least(2, a)
-        assert repeat("a", 2, None) == at_least(2, "a")
-        assert a.repeat(2, None) == a.at_least(2)
-
-        # at_most
-        assert repeat(a, None, 2) == at_most(2, a)
-        assert repeat("a", None, 2) == at_most(2, "a")
-        assert a.repeat(None, 2) == a.at_most(2)
-
-        # between
-        assert repeat(a, 1, 2) == between(1, 2, a)
-        assert repeat("a", 1, 2) == between(1, 2, "a")
-        assert a.repeat(1, 2) == a.between(1, 2)
-
-
-def test_dsl_times():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `times` function/method is deprecated",
-    ):
-        a = String("a")
-
-        assert times(a, 2) == exactly(2, a)
-        assert times("a", 2) == exactly(2, "a")
-        assert a.times(2) == a.exactly(2)
-
-
 def test_dsl_term_pydantic_simple():
     a = String("a")
 
@@ -661,6 +616,10 @@ class SomeEnum(Enum):
     assert python_types_to_terms(tuple[int, str]) == _handle_tuple((int, str), recursion_depth=0)
     assert python_types_to_terms(dict[int, str]) == _handle_dict((int, str), recursion_depth=0)
 
+    # type not supported
+    with pytest.raises(TypeError, match="is currently not supported"):
+        python_types_to_terms(bytes)
+
 
 def test_dsl_handle_literal():
     literal = Literal["a", 1]
@@ -795,3 +754,47 @@ def test_dsl_handle_dict():
     assert result.terms[1].term.terms[2] == types.string
     assert result.terms[1].term.terms[3] == KleeneStar(Sequence([String(", "), types.integer, String(":"), types.string]))
     assert result.terms[2] == String("}")
+
+
+def test_to_regex():
+    string_term = String("hello")
+    assert to_regex(string_term) == r"hello"
+
+    regex_term = Regex("[0-9]+")
+    assert to_regex(regex_term) == r"([0-9]+)"
+
+    json_schema_term = JsonSchema({"type": "integer"})
+    assert to_regex(json_schema_term) == r"((-)?(0|[1-9][0-9]*))"
+
+    choice_term = Choice(["a", "b", "c"])
+    assert to_regex(choice_term) == r"(a|b|c)"
+
+    kleene_star = KleeneStar(String("a"))
+    assert to_regex(kleene_star) == r"(a)*"
+
+    kleene_plus = KleenePlus(String("a"))
+    assert to_regex(kleene_plus) == r"(a)+"
+
+    optional_term = Optional(String("a"))
+    assert to_regex(optional_term) == r"(a)?"
+
+    alt_term = Alternatives([String("a"), String("b")])
+    assert to_regex(alt_term) == r"(a|b)"
+
+    seq_term = Sequence([String("a"), String("b")])
+    assert to_regex(seq_term) == r"ab"
+
+    exact_term = QuantifyExact(String("a"), 3)
+    assert to_regex(exact_term) == r"(a){3}"
+
+    min_term = QuantifyMinimum(String("a"), 2)
+    assert to_regex(min_term) == r"(a){2,}"
+
+    max_term = QuantifyMaximum(String("a"), 5)
+    assert to_regex(max_term) == r"(a){,5}"
+
+    between_term = QuantifyBetween(String("a"), 1, 3)
+    assert to_regex(between_term) == r"(a){1,3}"
+
+    with pytest.raises(TypeError):
+        to_regex(Term())
diff --git a/tests/v0_legacy/generate/test_api_legacy.py b/tests/v0_legacy/generate/test_api_legacy.py
deleted file mode 100644
index c34a3b706..000000000
--- a/tests/v0_legacy/generate/test_api_legacy.py
+++ /dev/null
@@ -1,187 +0,0 @@
-from io import BytesIO
-from dataclasses import asdict
-from typing import Generator
-from urllib.request import urlopen
-
-import pytest
-import transformers
-from PIL import Image
-from anthropic import Anthropic as AnthropicClient
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator
-from outlines.v0_legacy.models.transformers_vision import TransformersVision
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers(
-            "erwanf/gpt2-mini"
-        )
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=transformers.LlavaForConditionalGeneration,
-            processor_class=transformers.AutoProcessor,
-        )
-
-
-@pytest.fixture(scope="session")
-def llama_cpp_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        return models.llamacpp(
-            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        )
-
-
-def create_generator_v0_adapter(model):
-    if isinstance(model, TransformersVision):
-        return GeneratorVisionV0Adapter(
-            model,
-            int,
-            samplers.greedy(),
-        )
-    else:
-        return GeneratorV0Adapter(
-            model,
-            int,
-            samplers.greedy(),
-        )
-
-
-def test_generator_v0_adapter_init(transformers_model):
-    # wrong model type
-    model = models.from_anthropic(AnthropicClient(), "claude-3-haiku-20240307")
-    with pytest.raises(
-        ValueError,
-        match="You can only use the v0 API with models that were already"
-        + "available in v0. Got <class 'outlines.models.anthropic.Anthropic'>.",
-    ):
-        GeneratorV0Adapter(model, None, samplers.greedy())
-
-    # unknown output type
-    with pytest.raises(
-        TypeError,
-        match="is currently not supported",
-    ):
-        GeneratorV0Adapter(
-            transformers_model, type, samplers.greedy()
-        )
-
-    # valid initialization
-    generator_v0_adapter = GeneratorV0Adapter(
-        transformers_model,
-        int,
-        samplers.greedy(),
-    )
-    assert isinstance(generator_v0_adapter, GeneratorV0Adapter)
-    assert generator_v0_adapter.model == transformers_model
-    assert generator_v0_adapter.sampling_params == asdict(samplers.greedy().sampling_params)
-    assert isinstance(generator_v0_adapter.generator, SteerableGenerator)
-    assert generator_v0_adapter.generator.model == transformers_model
-    assert isinstance(generator_v0_adapter.generator.logits_processor, RegexLogitsProcessor)
-
-
-def test_generator_v0_adapter_create_inference_params(
-    llama_cpp_model,
-    transformers_vision_model,
-    transformers_model,
-):
-    args = {
-        "max_tokens": 10,
-        "stop_at": "\n",
-        "seed": 42
-    }
-    kwargs = {
-        "foo": "bar",
-    }
-
-    result = (
-        create_generator_v0_adapter(llama_cpp_model)
-        .create_inference_params(**args, **kwargs)
-    )
-    assert isinstance(result, dict)
-
-    result = (
-        create_generator_v0_adapter(transformers_vision_model)
-        .create_inference_params(**args, **kwargs)
-    )
-    assert isinstance(result, dict)
-
-    result = (
-        create_generator_v0_adapter(transformers_model)
-        .create_inference_params(**args, **kwargs)
-    )
-    assert isinstance(result, dict)
-
-
-def test_generator_vision_v0_adapter_merge_prompts_and_media(transformers_vision_model):
-    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
-    prompts = ["Hello, world!", "Goodbye, world!"]
-    media = [
-        img_from_url(
-            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-        ),
-        img_from_url(
-            "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-        )
-    ]
-    result = generator_v0_adapter.merge_prompts_and_media(prompts, media)
-    assert isinstance(result, dict)
-    assert result["text"] == prompts
-    assert result["images"] == media
-
-
-def test_generator_v0_adapter_call_stream(llama_cpp_model, transformers_model):
-    generator_v0_adapter = create_generator_v0_adapter(transformers_model)
-    result = generator_v0_adapter("Hello, world!", max_tokens=10)
-    assert isinstance(result, str)
-    result = generator_v0_adapter(["Hello, world!", "Goodbye, world!"], max_tokens=10)
-    assert isinstance(result, list)
-    for r in result:
-        assert isinstance(r, str)
-
-    generator_v0_adapter = create_generator_v0_adapter(llama_cpp_model)
-    gen = generator_v0_adapter.stream("Hello, world!", max_tokens=10)
-    assert isinstance(gen, Generator)
-
-
-def test_generator_v0_adapter_vision_call_stream(transformers_vision_model):
-    generator_v0_adapter = create_generator_v0_adapter(transformers_vision_model)
-    media = img_from_url(
-        "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-    )
-    with pytest.deprecated_call():
-        result = generator_v0_adapter("Hello, world!<image>", media, max_tokens=10)
-    assert isinstance(result, str)
-    with pytest.raises(
-        NotImplementedError,
-        match="Streaming is not implemented for Transformers models",
-    ):
-        gen = generator_v0_adapter.stream("Hello, world!<image>", media, max_tokens=10)
-        assert isinstance(gen, Generator)
diff --git a/tests/v0_legacy/generate/test_cfg_legacy.py b/tests/v0_legacy/generate/test_cfg_legacy.py
deleted file mode 100644
index 2b8a2734e..000000000
--- a/tests/v0_legacy/generate/test_cfg_legacy.py
+++ /dev/null
@@ -1,111 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import grammars, models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import CFGLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.cfg import cfg
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def llama_cpp_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        return models.llamacpp(
-            "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            "TinyMistral-248M-v2-Instruct.Q4_K_M.gguf"
-        )
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_cfg_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use CFG-based structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API.",
-        ):
-            cfg(openai_model, grammars.arithmetic)
-
-
-def test_cfg_legacy_llama_cpp(llama_cpp_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Not yet available due to bug in llama_cpp tokenizer",
-        ):
-            cfg(llama_cpp_model, grammars.arithmetic)
-
-
-def test_cfg_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        generator = cfg(transformers_vision_model, grammars.arithmetic)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
-
-
-def test_cfg_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `cfg` function is deprecated",
-    ):
-        generator = cfg(transformers_model, grammars.arithmetic, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(generator.generator.logits_processor, CFGLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_choice_legacy.py b/tests/v0_legacy/generate/test_choice_legacy.py
deleted file mode 100644
index fbc4650e2..000000000
--- a/tests/v0_legacy/generate/test_choice_legacy.py
+++ /dev/null
@@ -1,97 +0,0 @@
-import json as pyjson
-from dataclasses import asdict
-from enum import Enum
-from typing import Callable
-
-import pytest
-from pydantic import BaseModel
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.choice import choice
-
-
-class EnumChoice(Enum):
-    foo = "foo"
-    bar = "bar"
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-def test_choice_legacy_standard_enum(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `choice` function is deprecated",
-    ):
-        generator = choice(transformers_model, EnumChoice, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_choice_legacy_standard_list_strings(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `choice` function is deprecated",
-    ):
-        generator = choice(transformers_model, ["foo", "bar"])
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert generator.format_sequence("foo") == "foo"
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_choice_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `choice` function is deprecated",
-    ):
-        with pytest.raises(
-            ValueError,
-            match="The `choice` function with OpenAI only supports a list of "
-            + "strings as choices.",
-        ):
-            choice(openai_model, EnumChoice)
-
-        generator = choice(openai_model, ["foo", "bar"])
-        assert isinstance(generator, Callable)
diff --git a/tests/v0_legacy/generate/test_format_legacy.py b/tests/v0_legacy/generate/test_format_legacy.py
deleted file mode 100644
index 81503fb8f..000000000
--- a/tests/v0_legacy/generate/test_format_legacy.py
+++ /dev/null
@@ -1,93 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.format import format
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_format_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `format` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use Python type-structured generation with an OpenAI model"
-            + " due to the limitations of the OpenAI API.",
-        ):
-            format(openai_model, int)
-
-
-def test_format_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `format` function is deprecated",
-    ):
-        generator = format(transformers_vision_model, int)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        RegexLogitsProcessor
-    )
-
-
-def test_format_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `format` function is deprecated",
-    ):
-        generator = format(transformers_model, int, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        RegexLogitsProcessor
-    )
diff --git a/tests/v0_legacy/generate/test_fsm_legacy.py b/tests/v0_legacy/generate/test_fsm_legacy.py
deleted file mode 100644
index fe6e5ba2e..000000000
--- a/tests/v0_legacy/generate/test_fsm_legacy.py
+++ /dev/null
@@ -1,104 +0,0 @@
-from dataclasses import asdict
-
-import interegular
-import pytest
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import GuideLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.fsm import fsm
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_fsm_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `fsm` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use FSM-based generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API.",
-        ):
-            fsm(
-                openai_model,
-                interegular.parse_pattern(r"[a-z]+").to_fsm()
-            )
-
-
-def test_fsm_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `fsm` function is deprecated",
-    ):
-        generator = fsm(
-            transformers_vision_model,
-            interegular.parse_pattern(r"[a-z]+").to_fsm()
-        )
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        GuideLogitsProcessor
-    )
-
-
-def test_fsm_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `fsm` function is deprecated",
-    ):
-        generator = fsm(
-            transformers_model,
-            interegular.parse_pattern(r"[a-z]+").to_fsm(),
-            samplers.greedy()
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(
-        generator.generator.logits_processor,
-        GuideLogitsProcessor
-    )
diff --git a/tests/v0_legacy/generate/test_json_legacy.py b/tests/v0_legacy/generate/test_json_legacy.py
deleted file mode 100644
index d09d0347b..000000000
--- a/tests/v0_legacy/generate/test_json_legacy.py
+++ /dev/null
@@ -1,200 +0,0 @@
-import json as pyjson
-import pytest
-from dataclasses import asdict
-
-from pydantic import BaseModel
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.json import json
-
-
-JSON_SCHEMA_STRING = """
-{
-    "type": "object",
-    "properties": {
-        "foo": {
-            "type": "string"
-        }
-    },
-    "required": ["foo"]
-}
-"""
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_json_legacy_standard_callable(transformers_model):
-    def foo(bar: str) -> str:
-        return bar
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_model, foo, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_standard_json_string(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_model, JSON_SCHEMA_STRING)
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_standard_pydantic(transformers_model):
-    class Foo(BaseModel):
-        foo: str
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_model, Foo)
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert (
-        generator.format_sequence('{"foo": "value"}')
-        == Foo.model_validate_json('{"foo": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(transformers_vision_model, JSON_SCHEMA_STRING)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-    assert isinstance(
-        generator.generator.logits_processor, RegexLogitsProcessor
-    )
-
-
-def test_json_legacy_openai_wrong_sampler(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="The OpenAI API does not support any other sampling "
-            + "algorithm than the multinomial sampler.",
-        ):
-            json(openai_model, JSON_SCHEMA_STRING, samplers.greedy())
-
-
-def test_json_legacy_openai_pydantic(openai_model):
-    class Foo(BaseModel):
-        foo: str
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(openai_model, Foo)
-    assert isinstance(generator, models.OpenAI)
-    assert (
-        generator.format_sequence('{"foo": "value"}')
-        == Foo.model_validate_json('{"foo": "value"}')
-    )
-
-
-def test_json_legacy_openai_json_string(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        generator = json(openai_model, JSON_SCHEMA_STRING)
-    assert isinstance(generator, models.OpenAI)
-    assert (
-        generator.format_sequence('{"key": "value"}')
-        == pyjson.loads('{"key": "value"}')
-    )
-
-
-def test_json_legacy_openai_not_supported(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `json` function is deprecated",
-    ):
-        with pytest.raises(ValueError, match="Cannot parse schema"):
-            json(openai_model, int)
diff --git a/tests/v0_legacy/generate/test_regex_legacy.py b/tests/v0_legacy/generate/test_regex_legacy.py
deleted file mode 100644
index 86da00c35..000000000
--- a/tests/v0_legacy/generate/test_regex_legacy.py
+++ /dev/null
@@ -1,90 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.processors import RegexLogitsProcessor
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.regex import regex
-
-
-REGEX_STRING = r"^[a-z]+$"
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_regex_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `regex` function is deprecated",
-    ):
-        with pytest.raises(
-            NotImplementedError,
-            match="Cannot use regex-structured generation with an OpenAI model"
-            + "due to the limitations of the OpenAI API.",
-        ):
-            regex(openai_model, REGEX_STRING)
-
-
-def test_regex_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `regex` function is deprecated",
-    ):
-        generator = regex(transformers_vision_model, REGEX_STRING)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
-
-
-def test_regex_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `regex` function is deprecated",
-    ):
-        generator = regex(transformers_model, REGEX_STRING, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert isinstance(generator.generator.logits_processor, RegexLogitsProcessor)
diff --git a/tests/v0_legacy/generate/test_text_legacy.py b/tests/v0_legacy/generate/test_text_legacy.py
deleted file mode 100644
index e7412b295..000000000
--- a/tests/v0_legacy/generate/test_text_legacy.py
+++ /dev/null
@@ -1,83 +0,0 @@
-import pytest
-from dataclasses import asdict
-
-from transformers import AutoModelForCausalLM
-
-from outlines import models, samplers
-from outlines.generator import SteerableGenerator, Generator
-from outlines.v0_legacy.generate.api import (
-    GeneratorV0Adapter,
-    GeneratorVisionV0Adapter
-)
-from outlines.v0_legacy.generate.text import text
-
-
-@pytest.fixture(scope="session")
-def openai_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        return models.openai("gpt-4o")
-
-
-@pytest.fixture(scope="session")
-def transformers_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        return models.transformers("erwanf/gpt2-mini")
-
-
-@pytest.fixture(scope="session")
-def transformers_vision_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        return models.transformers_vision(
-            "erwanf/gpt2-mini",
-            AutoModelForCausalLM
-        )
-
-
-def test_text_legacy_openai(openai_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = text(openai_model)
-    assert generator == openai_model
-    assert isinstance(generator, models.OpenAI)
-
-
-def test_text_legacy_transformers_vision(transformers_vision_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = text(transformers_vision_model)
-    assert isinstance(generator, GeneratorVisionV0Adapter)
-    assert generator.model == transformers_vision_model
-    assert generator.sampling_params == asdict(
-        samplers.multinomial().sampling_params
-    )
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_vision_model
-    assert generator.generator.logits_processor is None
-
-
-def test_text_legacy_standard_model(transformers_model):
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = text(transformers_model, samplers.greedy())
-    assert isinstance(generator, GeneratorV0Adapter)
-    assert generator.model == transformers_model
-    assert generator.sampling_params == asdict(
-        samplers.greedy().sampling_params)
-    assert isinstance(generator.generator, SteerableGenerator)
-    assert generator.generator.model == transformers_model
-    assert generator.generator.logits_processor is None
diff --git a/tests/v0_legacy/models/test_exllamav2_legacy.py b/tests/v0_legacy/models/test_exllamav2_legacy.py
deleted file mode 100644
index 06fa9e17e..000000000
--- a/tests/v0_legacy/models/test_exllamav2_legacy.py
+++ /dev/null
@@ -1,76 +0,0 @@
-import warnings
-
-import pytest
-
-try:
-    import exllamav2  # type: ignore
-    HAS_EXV2 = True
-except ImportError:
-    HAS_EXV2 = False
-    warnings.warn(
-        "Skipping exllamav2 legacy tests because exllamav2 is not available."
-    )
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-pytestmark = pytest.mark.skipif(
-    not HAS_EXV2,
-    reason="Exllamav2 is not available."
-)
-
-
-@pytest.fixture(scope="session")
-def model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `exllamav2` function is deprecated",
-    ):
-        model = models.exl2(
-            model_path="blockblockblock/TinyLlama-1.1B-Chat-v1.0-bpw4-exl2",
-            max_seq_len=1024,
-            cache_q4=False,
-        )
-    return model
-
-
-def test_exllamav2_legacy_init(model):
-    assert isinstance(model, models.ExllamaV2)
-    assert hasattr(model, "generator")
-    assert hasattr(model, "tokenizer")
-    assert model.tokenizer.convert_token_to_string(1) == 1
-    assert hasattr(model, "max_seq_len")
-    assert isinstance(model.max_seq_len, int)
-
-
-def test_exllamav2_legacy_call(model):
-    # minimal
-    generator = generate.text(model)
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator("Respond with a single word.")
-    assert isinstance(result, str)
-
-    # with args
-    generator = generate.text(
-        model,
-        samplers.multinomial(samples=2, temperature=10, top_p=0.5, top_k=2)
-    )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(["Write a short story.", "Write a poem."], 2, "e", 2)
-    assert isinstance(result, list)
-    for sequence in result:
-        assert isinstance(sequence, list)
-        for s in sequence:
-            assert isinstance(s, str)
-            assert len(s) < 20
-
-    # stream
-    generator = generate.text(model)
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator.stream("Respond with a single word.", 2)
-    assert isinstance(result, GeneratorV0Adapter)
-    for r in result:
-        assert isinstance(r, str)
diff --git a/tests/v0_legacy/models/test_llamacpp_legacy.py b/tests/v0_legacy/models/test_llamacpp_legacy.py
deleted file mode 100644
index aead5bd86..000000000
--- a/tests/v0_legacy/models/test_llamacpp_legacy.py
+++ /dev/null
@@ -1,169 +0,0 @@
-import warnings
-from dataclasses import asdict
-
-import pytest
-from llama_cpp import Llama
-from llama_cpp.llama_tokenizer import LlamaHFTokenizer, LlamaTokenizer # type: ignore
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.models.llamacpp import llamacpp_params_adapter
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-
-def test_llamacpp_legacy_init():
-    # without tokenizer
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        with pytest.warns(
-            match="The pre-tokenizer in `llama.cpp` handles unicode improperly"
-        ):
-            model = models.llamacpp(
-                repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-                filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-            )
-    assert isinstance(model, models.LlamaCpp)
-    assert isinstance(model.model, Llama)
-    assert not model.model.verbose
-
-    # with tokenizer
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        model = models.llamacpp(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-            tokenizer=LlamaHFTokenizer.from_pretrained("distilbert-base-uncased"),
-            n_ctx=0,
-            verbose=True,
-        )
-    assert isinstance(model, models.LlamaCpp)
-    assert isinstance(model.model, Llama)
-    assert model.model.verbose
-    assert isinstance(model.model.tokenizer(), LlamaTokenizer)
-
-
-def test_llamacpp_legacy_params_adapter_generation_params():
-    result = llamacpp_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert not result.get("stop")
-    assert not result.get("seed")
-    assert result.get("max_tokens") == -1
-
-    result = llamacpp_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
-        {}
-    )
-    assert result.get("stop") == "foo"
-    assert result.get("seed") == 2
-    assert result.get("max_tokens") == 10
-
-
-def test_llamacpp_legacy_params_adapter_sampling_params():
-    with pytest.raises(
-        NotImplementedError,
-        match="does not support Beam Search"
-    ):
-        result = llamacpp_params_adapter(
-            asdict(samplers.beam_search().sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    with pytest.raises(
-        NotImplementedError,
-        match="does not allow to take several samples"
-    ):
-        result = llamacpp_params_adapter(
-            asdict(samplers.multinomial(samples=2).sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    result = llamacpp_params_adapter(
-        {"sampler": "multinomial", "num_samples": 1, "top_p": None, "top_k": None, "temperature": None},
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("top_p") == 1.0
-    assert result.get("top_k") == -1
-    assert result.get("temperature") == 1.0
-
-    result = llamacpp_params_adapter(
-        {"sampler": "greedy", "num_samples": 1, "top_p": 2, "top_k": 0, "temperature": 0.5},
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("top_p") == 2
-    assert result.get("top_k") == 0
-    assert result.get("temperature") == 0.5
-
-
-def test_llamacpp_legacy_params_adapter_model_specific_params():
-    result = llamacpp_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("min_p") == 0.0
-    assert result.get("repeat_penalty") == 1.0
-
-    result = llamacpp_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
-        {"min_p": 0.5, "repeat_penalty": 1.5, "stream": True}
-    )
-    assert result.get("min_p") == 0.5
-    assert result.get("repeat_penalty") == 1.5
-    assert not hasattr(result, "stream")
-
-
-def test_llamacpp_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `llamacpp` function is deprecated",
-    ):
-        model = models.llamacpp(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                top_p=0.9,
-                top_k=10,
-                temperature=0.7
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!",
-        10,
-        "foo",
-        2,
-        min_p=0.5,
-        repeat_penalty=1.5,
-    )
-    assert isinstance(result, str)
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!",
-        10,
-        "foo",
-        2,
-        min_p=0.5,
-        repeat_penalty=1.5,
-    )
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_mlxlm_legacy.py b/tests/v0_legacy/models/test_mlxlm_legacy.py
deleted file mode 100644
index e70943ef2..000000000
--- a/tests/v0_legacy/models/test_mlxlm_legacy.py
+++ /dev/null
@@ -1,193 +0,0 @@
-import warnings
-from dataclasses import asdict
-
-import pytest
-
-try:
-    import mlx.core as mx
-    import mlx_lm  # noqa: F401
-
-    assert mx.metal.is_available()
-    HAS_MLX = True
-except ImportError:
-    HAS_MLX = False
-    warnings.warn("Skipping MLXLM legacy tests because MLX is not available.")
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-from outlines.v0_legacy.models.mlxlm import mlxlm_params_adapter
-
-pytestmark = pytest.mark.skipif(
-    not HAS_MLX,
-    reason="MLX tests require Apple Silicon."
-)
-
-
-def test_mlxlm_legacy_init():
-    # minimal
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mlxlm` function is deprecated",
-    ):
-        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
-    assert isinstance(model, models.MLXLM)
-    assert isinstance(model.model, mlx_lm.models.llama.Model)
-
-    # with optional arguments
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mlxlm` function is deprecated",
-    ):
-        model = models.mlxlm(
-            "mlx-community/SmolLM-135M-Instruct-4bit",
-            tokenizer_config={"max_length": 100},
-            model_config={"max_length": 100},
-            adapter_path=None,
-            lazy=False,
-        )
-    assert isinstance(model, models.MLXLM)
-    assert isinstance(model.model, mlx_lm.models.llama.Model)
-    assert isinstance(model.mlx_tokenizer, mlx_lm.tokenizer_utils.TokenizerWrapper)
-
-
-def test_mlxlm_legacy_params_adapter_generation_params():
-    # default
-    result = mlxlm_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result.get("stop") is None
-    assert result.get("seed") is None
-    assert result.get("max_tokens") == int(1e9)
-
-    # max_tokens
-    result = mlxlm_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": 2},
-        {}
-    )
-    assert result.get("max_tokens") == 2
-
-    # stop_at
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support `stop_at`."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial().sampling_params),
-            {"stop_at": "foo", "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # seed
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support `seed`."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial().sampling_params),
-            {"stop_at": None, "seed": 2, "max_tokens": None},
-            {}
-        )
-
-
-def test_mlxlm_legacy_params_adapter_sampling_params():
-    # beam search
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support Beam Search."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.beam_search().sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # num_samples
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not allow to take several samples."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial(samples=2).sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # top_k
-    with pytest.raises(
-        NotImplementedError,
-        match="The `mlx-lm` library does not support top_k."
-    ):
-        result = mlxlm_params_adapter(
-            asdict(samplers.multinomial(top_k=2).sampling_params),
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-
-    # temperature
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `temperature` parameter is not available",
-    ):
-        result = mlxlm_params_adapter(
-            {
-                "sampler": "multinomial",
-                "num_samples": 1,
-                "top_p": None,
-                "top_k": None,
-                "temperature": 0.5
-            },
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-        assert result.get("temperature") is None
-
-    # top_p
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `top_p` parameter is not available",
-    ):
-        result = mlxlm_params_adapter(
-            {
-                "sampler": "multinomial",
-                "num_samples": 1,
-                "top_p": 2,
-                "top_k": None,
-                "temperature": None
-            },
-            {"stop_at": None, "seed": None, "max_tokens": None},
-            {}
-        )
-        assert result.get("top_p") is None
-
-
-def test_mlxlm_legacy_params_adapter_model_specific_params():
-    result = mlxlm_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {"length_penalty": 0.5}
-    )
-    assert result.get("length_penalty") == 0.5
-
-
-def test_mlxlm_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mlxlm` function is deprecated",
-    ):
-        model = models.mlxlm("mlx-community/SmolLM-135M-Instruct-4bit")
-
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial()
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator("Hello, world!", 10)
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_openai_legacy.py b/tests/v0_legacy/models/test_openai_legacy.py
deleted file mode 100644
index c07443dab..000000000
--- a/tests/v0_legacy/models/test_openai_legacy.py
+++ /dev/null
@@ -1,159 +0,0 @@
-import os
-
-import pytest
-from openai import AsyncAzureOpenAI, AsyncOpenAI
-
-from outlines import models, generate
-from outlines.models.openai import OpenAI, OpenAIConfig, OpenAILegacy
-
-
-# Set the OPENAI_API_KEY environment variable to "MOCK_VALUE" if it is not set
-# not to raise an error when the OpenAI client is initialized
-api_key = os.getenv("OPENAI_API_KEY")
-if not api_key:
-    os.environ["OPENAI_API_KEY"] = "MOCK_VALUE"
-azure_api_key = os.getenv("AZURE_OPENAI_API_KEY")
-if not azure_api_key:
-    os.environ["AZURE_OPENAI_API_KEY"] = "MOCK_VALUE"
-
-
-def test_openai_legacy_init():
-    # through the openai function with a client and no config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        client = AsyncOpenAI()
-        model = models.openai(client, system_prompt="You are a helpful assistant.")
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.config == OpenAIConfig()
-    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # through the openai function with a client and config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            client,
-            OpenAIConfig(stop=["."]),
-            system_prompt="You are a helpful assistant.",
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.config == OpenAIConfig(stop=["."])
-    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # through the openai function with model name and no config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            "gpt-4o",
-            max_retries=1,
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.client.max_retries == 1
-    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o")
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # through the openai function with model name and config
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            "gpt-4o",
-            OpenAIConfig(stop=["."]),
-            max_retries=1,
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.client.max_retries == 1
-    assert model.legacy_instance.config == OpenAIConfig(
-        model="gpt-4o", stop=["."]
-    )
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-    # directly through the OpenAI class with client, config and system prompt
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.OpenAI(
-            AsyncOpenAI(),
-            OpenAIConfig(model="gpt-4o", n=1),
-            system_prompt="You are a helpful assistant.",
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncOpenAI)
-    assert model.legacy_instance.config == OpenAIConfig(model="gpt-4o", n=1)
-    assert model.legacy_instance.system_prompt == "You are a helpful assistant."
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-
-def test_azure_openai_legacy_init():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.azure_openai(
-            "foo",
-            max_retries=1,
-            base_url="https://foo.openai.azure.com/",
-            api_version="2023-03-15-preview",
-        )
-    assert isinstance(model, models.OpenAI)
-    assert isinstance(model.legacy_instance, OpenAILegacy)
-    assert isinstance(model.legacy_instance.client, AsyncAzureOpenAI)
-    assert model.legacy_instance.client.max_retries == 1
-    assert model.legacy_instance.config == OpenAIConfig(model="foo")
-    assert not hasattr(model, "client")
-    assert not hasattr(model, "model_name")
-    assert not hasattr(model, "type_adapter")
-
-
-@pytest.mark.api_call
-def test_openai_legacy_call_model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `openai` function is deprecated",
-    ):
-        model = models.openai(
-            "gpt-4o",
-            max_retries=1,
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(model)
-    result = generator(
-        "Hello, world!",
-        10,
-        ["."],
-        system_prompt="You are a helpful assistant.",
-        temperature=0.5,
-        samples=1
-    )
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_legacy.py b/tests/v0_legacy/models/test_transformers_legacy.py
deleted file mode 100644
index 0388a4348..000000000
--- a/tests/v0_legacy/models/test_transformers_legacy.py
+++ /dev/null
@@ -1,173 +0,0 @@
-from dataclasses import asdict
-
-import pytest
-from transformers import (
-    GPT2LMHeadModel,
-    GPT2TokenizerFast,
-    BertForSequenceClassification,
-    BertTokenizer,
-    MambaForCausalLM,
-)
-
-from outlines import models, samplers, generate
-from outlines.v0_legacy.models.transformers import transformers_params_adapter
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-
-def test_transformers_legacy_init():
-    # Test with default model and tokenizer classes
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="erwanf/gpt2-mini",
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, GPT2LMHeadModel)
-    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
-
-    # Test with non default model and tokenizer classes
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="bert-base-uncased",
-            model_class=BertForSequenceClassification,
-            tokenizer_class=BertTokenizer,
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, BertForSequenceClassification)
-    assert isinstance(model.tokenizer.tokenizer, BertTokenizer)
-
-    # Test with model and tokenizer kwargs and device
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="erwanf/gpt2-mini",
-            model_kwargs={"use_cache": False},
-            tokenizer_kwargs={"model_max_length": 512},
-            device="cpu",
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, GPT2LMHeadModel)
-    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
-    assert model.tokenizer.tokenizer.model_max_length == 512
-    assert not model.model.config.use_cache
-    assert model.model.device.type == "cpu"
-
-
-def test_mamba_legacy_init():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `mamba` function is deprecated",
-    ):
-        model = models.mamba(
-            model_name="erwanf/gpt2-mini",
-        )
-    assert isinstance(model, models.Transformers)
-    assert isinstance(model.model, MambaForCausalLM)
-    assert isinstance(model.tokenizer.tokenizer, GPT2TokenizerFast)
-
-
-def test_transformers_legacy_params_adapter_generation_params():
-    result = transformers_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert not result.get("stop")
-    assert not result.get("seed")
-    assert result.get("max_new_tokens") == 2**30
-
-    result = transformers_params_adapter(
-         asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": 2, "max_tokens": 10},
-        {}
-    )
-    assert result.get("stop_strings") == "foo"
-    assert not result.get("seed")
-    assert result.get("max_new_tokens") == 10
-
-
-def test_transformers_legacy_params_adapter_sampling_params():
-    result = transformers_params_adapter(
-        {
-            "sampler": "multinomial",
-            "num_samples": None,
-            "top_p": None,
-            "top_k": None,
-            "temperature": None,
-        },
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {},
-    )
-    assert result.get("do_sample")
-    assert result.get("num_return_sequences") == 1
-    assert result.get("top_p") is None
-    assert result.get("top_k") is None
-    assert result.get("temperature") is None
-
-    result = transformers_params_adapter(
-        {
-            "sampler": "beam_search",
-            "num_samples": 2,
-            "top_p": 2,
-            "top_k": 0,
-            "temperature": 0.5,
-        },
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {},
-    )
-    assert result.get("num_beams") == 2
-    assert result.get("num_return_sequences") == 2
-    assert result.get("top_p") == 2
-    assert result.get("top_k") == 0
-    assert result.get("temperature") == 0.5
-
-
-def test_transformers_legacy_params_adapter_model_specific_params():
-    result = transformers_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {"length_penalty": 1.0}
-    )
-    assert result.get("length_penalty") == 1
-    assert not result.get("foo")
-
-
-def test_transformers_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers` function is deprecated",
-    ):
-        model = models.transformers(
-            model_name="erwanf/gpt2-mini",
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                samples=1,
-                top_p=0.9,
-                top_k=10,
-                temperature=0.7
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!",
-        10,
-        "foo",
-        2,
-        length_penalty=0.5,
-        tokenizer=model.tokenizer.tokenizer,
-    )
-    assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_transformers_vision_legacy.py b/tests/v0_legacy/models/test_transformers_vision_legacy.py
deleted file mode 100644
index 8a8632418..000000000
--- a/tests/v0_legacy/models/test_transformers_vision_legacy.py
+++ /dev/null
@@ -1,112 +0,0 @@
-from io import BytesIO
-from urllib.request import urlopen
-
-import pytest
-from PIL import Image
-from transformers import (
-    AutoProcessor,
-    LlavaForConditionalGeneration,
-    LlamaTokenizer,
-    LlamaTokenizerFast,
-    LlavaProcessor,
-)
-
-from outlines import models, generate, samplers
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-
-
-def img_from_url(url):
-    img_byte_stream = BytesIO(urlopen(url).read())
-    return Image.open(img_byte_stream).convert("RGB")
-
-
-def test_transformers_vision_legacy_init():
-    # Test with only model name and model class
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-        )
-    assert isinstance(model, models.TransformersVision)
-    assert isinstance(model.model, LlavaForConditionalGeneration)
-    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
-    assert isinstance(model.processor, LlavaProcessor)
-
-    # Test with model, tokenizer, and processor classes
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-            tokenizer_class=LlamaTokenizer,
-            processor_class=AutoProcessor,
-        )
-    assert isinstance(model, models.TransformersVision)
-    assert isinstance(model.model, LlavaForConditionalGeneration)
-    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizer)
-    assert isinstance(model.processor, LlavaProcessor)
-
-    # Test with model, tokenizer, processor kwargs, and device
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-            model_kwargs={"output_attentions": True},
-            processor_kwargs={"num_additional_image_tokens": 1},
-            device="cpu",
-        )
-    assert isinstance(model, models.TransformersVision)
-    assert isinstance(model.model, LlavaForConditionalGeneration)
-    assert isinstance(model.tokenizer.tokenizer, LlamaTokenizerFast)
-    assert isinstance(model.processor, LlavaProcessor)
-    assert model.model.config.output_attentions
-    assert model.processor.num_additional_image_tokens == 1
-    assert model.model.device.type == "cpu"
-
-
-def test_transformers_vision_legacy_call_generation():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `transformers_vision` function is deprecated",
-    ):
-        model = models.transformers_vision(
-            model_name="trl-internal-testing/tiny-LlavaForConditionalGeneration",
-            model_class=LlavaForConditionalGeneration,
-        )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                samples=1,
-                top_p=0.9,
-                top_k=10,
-                temperature=0.7
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    with pytest.deprecated_call():
-      result = generator(
-          "Hello, world!<image>",
-          img_from_url(
-              "https://upload.wikimedia.org/wikipedia/commons"
-              + "/2/25/Siam_lilacpoint.jpg"
-          ),
-          10,
-          "foo",
-          2,
-          length_penalty=0.5,
-          tokenizer=model.tokenizer.tokenizer,
-      )
-      assert isinstance(result, str)
diff --git a/tests/v0_legacy/models/test_vllm_offline_legacy.py b/tests/v0_legacy/models/test_vllm_offline_legacy.py
deleted file mode 100644
index 318eeddb7..000000000
--- a/tests/v0_legacy/models/test_vllm_offline_legacy.py
+++ /dev/null
@@ -1,181 +0,0 @@
-import warnings
-from dataclasses import asdict
-
-import pytest
-
-try:
-    from vllm import LLM
-    HAS_VLLM = True
-except ImportError:
-    HAS_VLLM = False
-    warnings.warn("Skipping vLLM legacy tests because vLLM is not available.")
-
-from outlines import generate, models, samplers
-from outlines.v0_legacy.generate.api import GeneratorV0Adapter
-from outlines.v0_legacy.models.vllm_offline import vllm_offline_params_adapter
-
-pytestmark = pytest.mark.skipif(
-    not HAS_VLLM,
-    reason="vLLM models can only be run on GPU."
-)
-
-
-@pytest.fixture
-def model():
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `vllm` function is deprecated",
-    ):
-        model_instance = models.vllm(
-            "erwanf/gpt2-mini",
-            seed=1,
-        )
-    return model_instance
-
-
-def test_vllm_legacy_init(model):
-    assert isinstance(model, models.VLLM)
-    assert isinstance(model.model, LLM)
-
-
-def test_vllm_legacy_params_adapter_generation_params():
-    # default
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].stop_strings is None
-    assert result["sampling_params"].seed is None
-    assert result["sampling_params"].max_tokens is None
-
-    # max_tokens
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": 2},
-        {}
-    )
-    assert result["sampling_params"].max_tokens == 2
-
-    # seed
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": 2, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].seed == 2
-
-    # stop_at str
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": "foo", "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].stop_strings == ["foo"]
-
-    # stop_at list
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": ["foo", "bar"], "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].stop_strings == ["foo", "bar"]
-
-
-def test_vllm_legacy_params_adapter_sampling_params():
-    # default
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].n == 1
-    assert result["sampling_params"].top_p == 1.0
-    assert result["sampling_params"].top_k == -1
-    assert result["sampling_params"].temperature == 1.0
-
-    # beam search
-    result = vllm_offline_params_adapter(
-        asdict(samplers.beam_search().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].use_beam_search is True
-
-    # num_samples
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(samples=2).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].n == 2
-    assert result["sampling_params"].best_of == 2
-
-    # top_p
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(top_p=0.5).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].top_p == 0.5
-
-    # top_k
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(top_k=2).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].top_k == 2
-
-    # temperature
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial(temperature=0.5).sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {}
-    )
-    assert result["sampling_params"].temperature == 0.5
-
-
-def test_vllm_legacy_params_adapter_model_specific_params():
-    result = vllm_offline_params_adapter(
-        asdict(samplers.multinomial().sampling_params),
-        {"stop_at": None, "seed": None, "max_tokens": None},
-        {"use_tqdm": False}
-    )
-    assert not result.get("use_tqdm")
-
-
-def test_vllm_legacy_call_generation(model):
-    # default
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial()
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator("Hello, world!")
-    assert isinstance(result, str)
-
-    # with args
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `text` function is deprecated",
-    ):
-        generator = generate.text(
-            model,
-            samplers.multinomial(
-                samples=2, top_p=0.5, top_k=2, temperature=0.5
-            )
-        )
-    assert isinstance(generator, GeneratorV0Adapter)
-
-    result = generator(
-        "Hello, world!", 10, stop_at="foo", seed=1, use_tqdm=False
-    )
-    assert isinstance(result, list)
-    for completion in result:
-        assert isinstance(completion, str)
diff --git a/tests/v0_legacy/test_base_legacy.py b/tests/v0_legacy/test_base_legacy.py
deleted file mode 100644
index f445b4f5d..000000000
--- a/tests/v0_legacy/test_base_legacy.py
+++ /dev/null
@@ -1,275 +0,0 @@
-import numpy as np
-import pytest
-from numpy.testing import assert_array_equal
-
-from outlines.v0_legacy.base import vectorize
-
-
-def test_vectorize_docstring():
-    def test(x):
-        """This is a test docstring"""
-        return x
-
-    fn = vectorize(test)
-    assert fn.__doc__ == "This is a test docstring"
-    assert fn.__name__ == "test"
-
-    async def test(x):
-        """This is a test docstring"""
-        return x
-
-    fn = vectorize(test)
-    assert fn.__doc__ == "This is a test docstring"
-    assert fn.__name__ == "test"
-
-
-def test_vectorize_thunk():
-    def thunk():
-        """A thunk"""
-        return 1
-
-    fn = vectorize(thunk)
-    assert fn() == 1
-
-    async def thunk():
-        """A thunk"""
-        return 1
-
-    fn = vectorize(thunk)
-    assert fn() == 1
-
-
-def test_vectorize_scalar_simple():
-    def passthrough(x):
-        """A passthrough function."""
-        return x
-
-    fn = vectorize(passthrough)
-
-    out_vector = fn(["one", "two", "three"])
-    assert_array_equal(out_vector, ["one", "two", "three"])
-
-    out_array = fn([["one", "two"], ["three", "four"]])
-    assert_array_equal(out_array, [["one", "two"], ["three", "four"]])
-
-    async def passthrough(x):
-        """A passthrough function."""
-        return x
-
-    fn = vectorize(passthrough)
-    assert fn.__doc__ == "A passthrough function."
-
-    out_vector = fn(["one", "two", "three"])
-    assert_array_equal(out_vector, ["one", "two", "three"])
-
-    out_array = fn([["one", "two"], ["three", "four"]])
-    assert_array_equal(out_array, [["one", "two"], ["three", "four"]])
-
-
-def test_vectorize_scalar_multiple_outputs():
-    def passthrough_multiple_outputs(x):
-        return x, x
-
-    fn = vectorize(passthrough_multiple_outputs)
-    out = fn(["one", "two", "three"])
-    assert len(out) == 2
-    assert_array_equal(out[0], ["one", "two", "three"])
-    assert_array_equal(out[1], ["one", "two", "three"])
-
-    async def passthrough_multiple_outputs(x):
-        return x, x
-
-    fn = vectorize(passthrough_multiple_outputs)
-    out = fn(["one", "two", "three"])
-    assert len(out) == 2
-    assert_array_equal(out[0], ["one", "two", "three"])
-    assert_array_equal(out[1], ["one", "two", "three"])
-
-
-def test_vectorize_scalar_args():
-    def passthrough_args(*args):
-        """A passthrough function."""
-        result = ""
-        for arg in args:
-            result += arg
-        return result
-
-    fn = vectorize(passthrough_args)
-
-    out_array = fn(["one", "two"], ["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn([["one", "two"], ["three", "four"]], ["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-    async def passthrough_args(*args):
-        """A passthrough function."""
-        result = ""
-        for arg in args:
-            result += arg
-        return result
-
-    fn = vectorize(passthrough_args)
-
-    out_array = fn(["one", "two"], ["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn([["one", "two"], ["three", "four"]], ["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-
-def test_vectorize_scalar_kwargs():
-    def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs)
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-    async def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs)
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-
-def test_signature_invalid():
-    with pytest.raises(ValueError):
-        vectorize(lambda x: x, "(m,)->()")
-
-    with pytest.raises(TypeError, match="wrong number of positional"):
-        fn = vectorize(lambda x, y: x, "()->()")
-        fn(1, 2)
-
-    with pytest.raises(ValueError, match="wrong number of outputs"):
-
-        def test_multioutput(x, y):
-            return x, y
-
-        fn = vectorize(test_multioutput, "(),()->()")
-        fn(1, 2)
-
-
-def test_vectorize_simple():
-    def test(x):
-        return x
-
-    fn = vectorize(test, "(m)->(m)")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-
-def test_vectorize_kwargs():
-    def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs, "(),()->()")
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-    async def passthrough_kwargs(**kwargs):
-        """A passthrough function."""
-        result = ""
-        for _, value in kwargs.items():
-            result += value
-        return result
-
-    fn = vectorize(passthrough_kwargs, "(),()->()")
-
-    out_array = fn(first=["one", "two"], second=["1", "2"])
-    assert_array_equal(out_array, ["one1", "two2"])
-
-    # Broadcasting
-    out_array = fn(first=[["one", "two"], ["three", "four"]], second=["1", "2"])
-    assert_array_equal(out_array, [["one1", "two2"], ["three1", "four2"]])
-
-
-def test_vectorize_reduce():
-    def test(x):
-        return x[0]
-
-    fn = vectorize(test, "(m)->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, ["one", "four"])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["o", "t", "t"], ["f", "f", "s"]])
-
-
-def test_vectorize_expand():
-    def test(x):
-        return np.array([x for i in range(3)])
-
-    fn = vectorize(test, "()->(s)")
-    out = fn(["one", "two"])
-    assert_array_equal(out, [["one", "one", "one"], ["two", "two", "two"]])
-
-
-def test_vectorize_coroutine_simple():
-    async def test(x):
-        return x
-
-    fn = vectorize(test, "(m)->(m)")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["one", "two", "three"], ["four", "five", "six"]])
-
-
-def test_vectorize_coroutine_reduce():
-    async def test(x):
-        return x[0]
-
-    fn = vectorize(test, "(m)->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, ["one", "four"])
-
-    fn = vectorize(test, "()->()")
-    out = fn([["one", "two", "three"], ["four", "five", "six"]])
-    assert_array_equal(out, [["o", "t", "t"], ["f", "f", "s"]])
-
-
-def test_vectorize_coroutine_expand():
-    async def test(x):
-        return np.array([x for i in range(3)])
-
-    fn = vectorize(test, "()->(s)")
-    out = fn(["one", "two"])
-    assert_array_equal(out, [["one", "one", "one"], ["two", "two", "two"]])
diff --git a/tests/v0_legacy/test_function_legacy.py b/tests/v0_legacy/test_function_legacy.py
deleted file mode 100644
index 1fd03ebb9..000000000
--- a/tests/v0_legacy/test_function_legacy.py
+++ /dev/null
@@ -1,192 +0,0 @@
-import pytest
-import responses
-from pydantic import BaseModel
-from requests.exceptions import HTTPError
-
-import outlines
-from outlines.v0_legacy.function import (
-    Function,
-    download_from_github,
-    extract_function_from_file,
-)
-
-
-def test_function_basic():
-    with pytest.warns(DeprecationWarning, match="The @prompt decorator"):
-        with pytest.warns(
-            DeprecationWarning, match="The `Function` class is deprecated"
-        ):
-            @outlines.prompt
-            def test_template(text: str):
-                """{{ text }}"""
-
-            class Foo(BaseModel):
-                id: int
-
-            fn = Function(
-                test_template,
-                Foo,
-                "erwanf/gpt2-mini",
-            )
-
-            assert fn.generator is None
-
-            result = fn("test")
-            assert isinstance(result, BaseModel)
-            assert fn.generator is not None
-            fn("test")
-
-
-@responses.activate
-def test_function_from_github():
-    content = """
-import outlines
-from pydantic import BaseModel
-
-model = "gpt2"
-
-
-prompt = outlines.Template.from_string("{{ text }}")
-
-
-class User(BaseModel):
-    id: int
-    name: str
-
-
-fn = outlines.Function(
-    prompt,
-    User,
-    "gpt2",
-)
-"""
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
-        body=content,
-        status=200,
-    )
-    with pytest.warns(
-        DeprecationWarning,
-        match="The `Function` class is deprecated",
-    ):
-        fn = Function.from_github("dottxt-ai/outlines/program", "fn")
-        assert (
-            str(type(fn)) == "<class 'outlines.v0_legacy.function.Function'>"
-        )
-
-
-def test_download_from_github_invalid():
-    with pytest.raises(ValueError, match="Please provide"):
-        download_from_github("outlines/program")
-
-    with pytest.raises(ValueError, match="Do not append"):
-        download_from_github("dottxt-ai/outlines/program.py")
-
-
-@responses.activate
-def test_download_from_github_success():
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/program.py",
-        body="import outlines\n",
-        status=200,
-    )
-
-    file = download_from_github("dottxt-ai/outlines/program")
-    assert file == "import outlines\n"
-
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/dottxt-ai/outlines/main/foo/bar/program.py",
-        body="import outlines\n",
-        status=200,
-    )
-
-    file = download_from_github("dottxt-ai/outlines/foo/bar/program")
-    assert file == "import outlines\n"
-
-
-@responses.activate
-def test_download_from_github_error():
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/foo/bar/main/program.py",
-        json={"error": "not found"},
-        status=404,
-    )
-
-    with pytest.raises(ValueError, match="Program could not be found at"):
-        download_from_github("foo/bar/program")
-
-    responses.add(
-        responses.GET,
-        "https://raw.githubusercontent.com/foo/bar/main/program.py",
-        json={"error": "Internal Server Error"},
-        status=500,
-    )
-
-    with pytest.raises(HTTPError, match="500 Server Error"):
-        download_from_github("foo/bar/program")
-
-
-def test_extract_function_from_file():
-    content = """
-import outlines
-from pydantic import BaseModel
-
-model = "gpt2"
-
-
-@outlines.prompt
-def prompt():
-    '''Hello'''
-
-
-class User(BaseModel):
-    id: int
-    name: str
-
-
-function = outlines.Function(
-    prompt,
-    User,
-    "gpt2",
-)
-    """
-    with pytest.raises(
-        DeprecationWarning,
-        match="The @prompt decorator is deprecated",
-    ):
-        fn = extract_function_from_file(content, "function")
-        assert (
-            str(type(fn)) == "<class 'outlines.function.Function'>"
-        )  # because imported via `exec`
-
-
-def test_extract_function_from_file_no_function():
-    content = """
-import outlines
-from pydantic import BaseModel
-
-@outlines.prompt
-def prompt():
-    '''Hello'''
-
-
-class User(BaseModel):
-    id: int
-    name: str
-
-program = outlines.Function(
-    prompt,
-    User,
-    "gpt2",
-)
-    """
-    with pytest.raises(
-        DeprecationWarning,
-        match="The @prompt decorator is deprecated",
-    ):
-        with pytest.raises(AttributeError, match="Could not find"):
-            extract_function_from_file(content, "function")
diff --git a/tests/v0_legacy/test_samplers_legacy.py b/tests/v0_legacy/test_samplers_legacy.py
deleted file mode 100644
index a4d750116..000000000
--- a/tests/v0_legacy/test_samplers_legacy.py
+++ /dev/null
@@ -1,60 +0,0 @@
-from outlines.v0_legacy.samplers import (
-    BeamSearchSampler,
-    GreedySampler,
-    MultinomialSampler,
-    beam_search,
-    greedy,
-    multinomial,
-)
-
-
-def test_aliases():
-    assert greedy == GreedySampler
-    assert multinomial == MultinomialSampler
-    assert beam_search == BeamSearchSampler
-
-
-def test_greedy():
-    sampler = GreedySampler()
-    params = sampler.sampling_params
-    assert params.sampler == "greedy"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 0.0
-
-
-def test_multinomial():
-    sampler = MultinomialSampler()
-    params = sampler.sampling_params
-    assert params.sampler == "multinomial"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature is None
-
-    sampler = multinomial(samples=5, top_k=10, top_p=0.9, temperature=0.8)
-    params = sampler.sampling_params
-    assert params.sampler == "multinomial"
-    assert params.num_samples == 5
-    assert params.top_p == 0.9
-    assert params.top_k == 10
-    assert params.temperature == 0.8
-
-
-def test_beam_search():
-    sampler = BeamSearchSampler()
-    params = sampler.sampling_params
-    assert params.sampler == "beam_search"
-    assert params.num_samples == 1
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 1.0
-
-    sampler = beam_search(beams=3)
-    params = sampler.sampling_params
-    assert params.sampler == "beam_search"
-    assert params.num_samples == 3
-    assert params.top_p is None
-    assert params.top_k is None
-    assert params.temperature == 1.0

From 07ebb5630a56bb5d56d1225262a07b6ce3daf51e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 10 Jul 2025 16:39:01 +0200
Subject: [PATCH 404/505] Remove documentation and misc related to v0
 deprecated features

---
 docs/features/models/index.md | 5 -----
 llm.txt                       | 8 ++------
 pyproject.toml                | 7 -------
 scripts/gen_ref_pages.py      | 2 +-
 4 files changed, 3 insertions(+), 19 deletions(-)

diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index 4a0b46c69..66673dbf4 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -8,11 +8,6 @@ title: Models
 
 Outlines models are objects that wrap an inference client or engine. Models provide a standardized interface to generate structured text.
 
-!!! Warning
-
-    The model loading functions have been modified in v1. While they used to be called `<name_inference_library>`, they are now called `from_<name_inference_library>`. The model classes' names and `__init__` methods are left unchanged.
-
-
 All Outlines model classes have an associated loader function to facilitate initializing a model instance. The name of this function is `from_` plus the name of the model in lower-case letters. For instance, Outlines has a `Transformers` model and an associated `from_transformers` loader function. The parameters to load a model are specific to each provider, please consult the documentation of the model you want to use for more information.
 
 After having created a model instance, you can either directly call it to generate text or first create a reusable generator that you would then call.
diff --git a/llm.txt b/llm.txt
index 0bc07702f..c6fa3fb04 100644
--- a/llm.txt
+++ b/llm.txt
@@ -6,7 +6,7 @@ Outlines is a library for structured generation for type-safe LLMs. It ensures o
 
 **Core insight**: Instead of generating text and hoping it matches a format, Outlines makes it impossible for the model to generate invalid outputs by masking invalid tokens during generation.
 
-**Note**: The codebase has undergone significant refactoring. Core FSM functionality has been extracted to the `outlines-core` package, and legacy implementations are preserved in `v0_legacy/` for backward compatibility.
+**Note**: The codebase has undergone significant refactoring. Core FSM functionality has been extracted to the `outlines-core` package.
 
 ## Usage Examples
 
@@ -73,10 +73,9 @@ Key files:
 - `tgi.py`: Integration with text-generation-inferece servers
 - `sglang.py`: Integration with SGLang servers
 
-### Generation (`outlines/generator.py` and `outlines/v0_legacy/generate/`)
+### Generation (`outlines/generator.py`)
 Handles the generation process:
 - `generator.py`: Main `Generator` class implementations (root level)
-- `v0_legacy/generate/api.py`: Legacy generation API (backward compatibility)
 - Stream functionality is now integrated into generator classes
 
 Base classes and implementations for different model providers:
@@ -152,9 +151,6 @@ outlines/
 │   └── tensor_adapters/    # Framework-specific tensor handling
 ├── caching.py               # Caching system
 ├── grammars/                # Grammar files (.lark)
-└── v0_legacy/              # Backward compatibility
-    ├── generate/           # Legacy generation API
-    └── models/             # Legacy model integrations
 ```
 
 ## Extension Points
diff --git a/pyproject.toml b/pyproject.toml
index 77bb4ac21..d8cbcf11f 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -185,13 +185,6 @@ omit = [
     "outlines/models/openai.py",
     "outlines/models/vllm_offline.py",
     "outlines/processors/tensor_adapters/mlx.py",
-    "outlines/v0_legacy/models/anthropic.py",
-    "outlines/v0_legacy/models/dottxt.py",
-    "outlines/v0_legacy/models/exllamav2.py",
-    "outlines/v0_legacy/models/gemini.py",
-    "outlines/v0_legacy/models/mlxlm.py",
-    "outlines/v0_legacy/models/openai.py",
-    "outlines/v0_legacy/models/vllm_offline.py",
     "tests/*",
 ]
 branch = true
diff --git a/scripts/gen_ref_pages.py b/scripts/gen_ref_pages.py
index 48c226442..ee956d806 100644
--- a/scripts/gen_ref_pages.py
+++ b/scripts/gen_ref_pages.py
@@ -14,7 +14,7 @@
 
 CODEBASE_DIR_NAME = "outlines"
 OUTPUT_DIR_NAME = "api_reference"
-EXCLUDED_FILES = ["_version", "v0_legacy"]
+EXCLUDED_FILES = ["_version"]
 
 
 nav = mkdocs_gen_files.Nav()

From 1e8022e210dc7eb193d8e5808a617b1a9dc15644 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 11 Jul 2025 11:30:56 +0200
Subject: [PATCH 405/505] Set add_generation_template to True in chat
 formatting for Transformers

---
 outlines/models/transformers.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 475635054..03fce82bd 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -166,7 +166,8 @@ def format_str_input(self, model_input: str) -> str:
     def format_chat_input(self, model_input: Chat) -> str:
         return self.tokenizer.apply_chat_template(
             model_input.messages,
-            tokenize=False
+            tokenize=False,
+            add_generation_prompt=True,
         )
 
     def format_output_type(

From 00b2a27d0e9e13895ed0a370956dfde3b5ec3be2 Mon Sep 17 00:00:00 2001
From: Francesco Laiti <25352428+laitifranz@users.noreply.github.com>
Date: Fri, 11 Jul 2025 14:30:46 +0200
Subject: [PATCH 406/505] Update index.md

---
 docs/examples/index.md | 1 -
 1 file changed, 1 deletion(-)

diff --git a/docs/examples/index.md b/docs/examples/index.md
index 0181d137a..8e1f9bc0b 100644
--- a/docs/examples/index.md
+++ b/docs/examples/index.md
@@ -13,7 +13,6 @@ This part of the documentation provides a few cookbooks that you can browse to g
 - [Structured Generation Workflow](structured_generation_workflow.md):
 - [Chain Of Thought (CoT)](chain_of_thought.md): Generate a series of intermediate reasoning steps using regex-structured generation.
 - [ReAct Agent](react_agent.md): Build an agent with open weights models using regex-structured generation.
-- [Vision-Language Models](atomic_caption.md): Use Outlines with vision-language models for tasks like image captioning and visual reasoning.
 - [Structured Generation from PDFs](read-pdfs.md): Use Outlines with vision-language models to read PDFs and produce structured output.
 - [Earnings reports to CSV](earnings-reports.md): Extract data from earnings reports to CSV using regex-structured generation.
 - [Receipt Digitization](receipt-digitization.md): Extract information from a picture of a receipt using structured generation.

From 222ca17254ca0a4de1961e16872efe917b8e3498 Mon Sep 17 00:00:00 2001
From: Francesco Laiti <25352428+laitifranz@users.noreply.github.com>
Date: Mon, 14 Jul 2025 11:10:06 +0200
Subject: [PATCH 407/505] Update getting_started.md to fix class import from
 Transformer library

---
 docs/guide/getting_started.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
index f510bc16f..617e27e5c 100644
--- a/docs/guide/getting_started.md
+++ b/docs/guide/getting_started.md
@@ -72,14 +72,14 @@ For a quick start, you can find below an example of how to initialize all suppor
 
     ```python
     import outlines
-    from transformers
+    from transformers import AutoModelForCausalLM, AutoTokenizer
 
     # Define the model you want to use
     model_name = "HuggingFaceTB/SmolLM2-135M-Instruct"
 
     # Create a HuggingFace model and tokenizer
-    hf_model = transformers.AutoModelForCausalLM.from_pretrained(model_name)
-    hf_tokenizer = transformers.AutoTokenizer.from_pretrained(model_name)
+    hf_model = AutoModelForCausalLM.from_pretrained(model_name)
+    hf_tokenizer = AutoTokenizer.from_pretrained(model_name)
 
     # Create an Outlines model
     model = outlines.from_transformers(hf_model, hf_tokenizer)
@@ -256,7 +256,7 @@ In the meantime, you can find below examples of using each of the five output ty
     # Generate a character
     result = model("Create a character", Character)
     print(result) # '{"name": "Aurora", "birth_date": "1990-06-15", "skills": ["Stealth", "Diplomacy"]}'
-    print(Character.model_validate_json(result);) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
+    print(Character.model_validate_json(result)) # name=Aurora birth_date=datetime.date(1990, 6, 15) skills=['Stealth', 'Diplomacy']
     ```
 
 === "Regex"

From e12442f1d17445c87e73a28c0615a47c1fc7e624 Mon Sep 17 00:00:00 2001
From: Bright Liu <122313949+brightlikethelight@users.noreply.github.com>
Date: Sat, 19 Jul 2025 15:03:18 -0400
Subject: [PATCH 408/505] Fix: JSON schema union types (type arrays) fail with
 'type must be a string' error (#1675)

## Summary
- Fixes #1676 by adding a Python-side workaround for JSON schema union
types (type arrays)
- Converts type arrays like `["string", "null"]` to `anyOf` format that
outlines-core 0.1.26 supports
- Handles nested union types recursively

## Background

JSON schemas with union types specified as arrays (e.g., `{"type":
["string", "null"]}`) currently fail with a ValueError: "'type' must be
a string". This prevents the use of optional fields and other union type
patterns in dynamic JSON schema generation.

The root cause is that the current version of outlines-core (0.1.26)
does not support type arrays. This was fixed in outlines-core PR
dottxt-ai/outlines-core#138, but that fix is only available in
outlines-core v0.2+. Upgrading to v0.2+ requires significant changes due
to breaking API changes (see #1380).

## Solution

This PR implements a Python-side preprocessing step that:
1. Recursively traverses JSON schemas
2. Converts type arrays to the equivalent `anyOf` format
3. Preserves type-specific constraints (minLength, pattern, etc.)

This is a temporary workaround until outlines can be migrated to
outlines-core v0.2+.

## Test plan
- Added comprehensive unit tests in `test_json_schema_union_types.py`
- Tests cover simple optionals, nested optionals, arrays with optional
items, and constraint preservation
- All tests pass with the current implementation

## Example

Before preprocessing:
```json
{
  "type": "object",
  "properties": {
    "age": {"type": ["integer", "null"]}
  }
}
```

After preprocessing:
```json
{
  "type": "object",
  "properties": {
    "age": {"anyOf": [{"type": "integer"}, {"type": "null"}]}
  }
}
```

## Related Issues
- #1676 (this issue)
- dottxt-ai/outlines-core#138 (the fix in outlines-core)
- #1380 (tracking outlines-core v0.2 migration)

---------

Co-authored-by: Bright-L01 <122313949+Bright-L01@users.noreply.github.com>
Co-authored-by: Robin Picard <robin.picard@sciencespo.fr>
Co-authored-by: Robin Picard <83579270+RobinPicard@users.noreply.github.com>
---
 outlines/types/dsl.py                       |  73 ++++-----
 outlines/types/json_schema_utils.py         | 140 +++++++++++++++++
 scripts/reproduce_1383.py                   |  82 ++++++++++
 tests/types/test_json_schema_union_types.py | 158 ++++++++++++++++++++
 4 files changed, 417 insertions(+), 36 deletions(-)
 create mode 100644 outlines/types/json_schema_utils.py
 create mode 100644 scripts/reproduce_1383.py
 create mode 100644 tests/types/test_json_schema_union_types.py

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 10d47d7c8..548c837ff 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -41,6 +41,7 @@
 
 import outlines.types as types
 from outlines import grammars
+from outlines.types.json_schema_utils import preprocess_schema_for_union_types
 from outlines.types.utils import (
     get_schema_from_signature,
     is_int,
@@ -343,7 +344,8 @@ def __init__(
                 + "specification"
             )
 
-        self.schema = schema_str
+        # Preprocess the schema to handle union types
+        self.schema = preprocess_schema_for_union_types(schema_str, ensure_ascii=ensure_ascii)
         self.whitespace_pattern = whitespace_pattern
 
     def __post_init__(self):
@@ -824,38 +826,37 @@ def to_regex(term: Term) -> str:
         The regular expression as a string.
 
     """
-    match term:
-        case String():
-            return re.escape(term.value)
-        case Regex():
-            return f"({term.pattern})"
-        case JsonSchema():
-            regex_str = build_regex_from_schema(term.schema, term.whitespace_pattern)
-            return f"({regex_str})"
-        case Choice():
-            regexes = [to_regex(python_types_to_terms(item)) for item in term.items]
-            return f"({'|'.join(regexes)})"
-        case KleeneStar():
-            return f"({to_regex(term.term)})*"
-        case KleenePlus():
-            return f"({to_regex(term.term)})+"
-        case Optional():
-            return f"({to_regex(term.term)})?"
-        case Alternatives():
-            regexes = [to_regex(subterm) for subterm in term.terms]
-            return f"({'|'.join(regexes)})"
-        case Sequence():
-            regexes = [to_regex(subterm) for subterm in term.terms]
-            return f"{''.join(regexes)}"
-        case QuantifyExact():
-            return f"({to_regex(term.term)}){{{term.count}}}"
-        case QuantifyMinimum():
-            return f"({to_regex(term.term)}){{{term.min_count},}}"
-        case QuantifyMaximum():
-            return f"({to_regex(term.term)}){{,{term.max_count}}}"
-        case QuantifyBetween():
-            return f"({to_regex(term.term)}){{{term.min_count},{term.max_count}}}"
-        case _:
-            raise TypeError(
-                f"Cannot convert object {repr(term)} to a regular expression."
-            )
+    if isinstance(term, String):
+        return re.escape(term.value)
+    elif isinstance(term, Regex):
+        return f"({term.pattern})"
+    elif isinstance(term, JsonSchema):
+        regex_str = build_regex_from_schema(term.schema, term.whitespace_pattern)
+        return f"({regex_str})"
+    elif isinstance(term, Choice):
+        regexes = [to_regex(python_types_to_terms(item)) for item in term.items]
+        return f"({'|'.join(regexes)})"
+    elif isinstance(term, KleeneStar):
+        return f"({to_regex(term.term)})*"
+    elif isinstance(term, KleenePlus):
+        return f"({to_regex(term.term)})+"
+    elif isinstance(term, Optional):
+        return f"({to_regex(term.term)})?"
+    elif isinstance(term, Alternatives):
+        regexes = [to_regex(subterm) for subterm in term.terms]
+        return f"({'|'.join(regexes)})"
+    elif isinstance(term, Sequence):
+        regexes = [to_regex(subterm) for subterm in term.terms]
+        return f"{''.join(regexes)}"
+    elif isinstance(term, QuantifyExact):
+        return f"({to_regex(term.term)}){{{term.count}}}"
+    elif isinstance(term, QuantifyMinimum):
+        return f"({to_regex(term.term)}){{{term.min_count},}}"
+    elif isinstance(term, QuantifyMaximum):
+        return f"({to_regex(term.term)}){{,{term.max_count}}}"
+    elif isinstance(term, QuantifyBetween):
+        return f"({to_regex(term.term)}){{{term.min_count},{term.max_count}}}"
+    else:
+        raise TypeError(
+            f"Cannot convert object {repr(term)} to a regular expression."
+        )
diff --git a/outlines/types/json_schema_utils.py b/outlines/types/json_schema_utils.py
new file mode 100644
index 000000000..4c99e14a5
--- /dev/null
+++ b/outlines/types/json_schema_utils.py
@@ -0,0 +1,140 @@
+"""Utilities for handling JSON schema compatibility."""
+
+import json
+from typing import Any, Union
+
+
+def preprocess_schema_for_union_types(
+    schema: Union[str, dict], ensure_ascii: bool = True
+) -> str:
+    """
+    Preprocess a JSON schema to handle union types (array type specifications).
+
+    This is a temporary workaround for the limitation in outlines-core 0.1.26
+    which doesn't support JSON schema type arrays like ["string", "null"].
+    This function converts such arrays into the equivalent anyOf format.
+
+    Parameters
+    ----------
+    schema
+        The JSON schema as a string or dictionary
+    ensure_ascii
+        Whether to ensure the output JSON is ASCII-only
+
+    Returns
+    -------
+    str
+        The preprocessed JSON schema string
+
+    Examples
+    --------
+    >>> schema = {"type": ["string", "null"]}
+    >>> preprocess_schema_for_union_types(schema)
+    '{"anyOf":[{"type":"string"},{"type":"null"}]}'
+    """
+    # Convert to dict if string
+    if isinstance(schema, str):
+        original_str = schema
+        try:
+            schema_dict = json.loads(schema)
+        except (json.JSONDecodeError, ValueError):
+            # If JSON parsing fails, return original string unchanged
+            # This preserves original error handling behavior
+            return original_str
+    else:
+        original_str = None
+        schema_dict = schema
+
+    # Process the schema
+    preprocessed = _convert_type_arrays_to_anyof(schema_dict)
+
+    # If no changes were made, return the original string (if it was a string)
+    if preprocessed == schema_dict and original_str is not None: # pragma: no cover
+        return original_str
+
+    # Return as JSON string with proper formatting
+    return json.dumps(preprocessed, ensure_ascii=ensure_ascii, separators=(",", ":"))
+
+
+def _convert_type_arrays_to_anyof(obj: Any) -> Any:
+    """
+    Recursively convert type arrays to anyOf format.
+
+    Parameters
+    ----------
+    obj
+        The JSON schema object or sub-object to process
+
+    Returns
+    -------
+    Any
+        The processed object with type arrays converted to anyOf
+    """
+    if isinstance(obj, dict):
+        # Check if this object has a type array that needs conversion
+        if "type" in obj and isinstance(obj["type"], list):
+            # Convert type array to anyOf
+            types = obj["type"]
+            new_obj = {}
+
+            # Copy over non-type-specific properties
+            type_specific_keys = {
+                "properties", "required", "additionalProperties",  # object
+                "items", "minItems", "maxItems", "uniqueItems",  # array
+                "minLength", "maxLength", "pattern", "format",  # string
+                "minimum", "maximum", "exclusiveMinimum", "exclusiveMaximum", "multipleOf"  # number/integer
+            }
+
+            for k, v in obj.items():
+                if k != "type" and k not in type_specific_keys:
+                    new_obj[k] = v
+
+            # Create anyOf array
+            any_of = []
+            for t in types:
+                if t == "null":
+                    # null is a special case - just type
+                    any_of.append({"type": "null"})
+                else:
+                    # For other types, preserve any type-specific constraints
+                    type_schema = {"type": t}
+
+                    # Copy over type-specific constraints
+                    if t == "string":
+                        for k in ["minLength", "maxLength", "pattern", "format"]:
+                            if k in obj:
+                                type_schema[k] = obj[k]
+                    elif t == "number" or t == "integer":
+                        for k in ["minimum", "maximum", "exclusiveMinimum", "exclusiveMaximum", "multipleOf"]:
+                            if k in obj:
+                                type_schema[k] = obj[k]
+                    elif t == "array":
+                        for k in ["items", "minItems", "maxItems", "uniqueItems"]:
+                            if k in obj:
+                                # Recursively process items if present
+                                if k == "items":
+                                    type_schema[k] = _convert_type_arrays_to_anyof(obj[k])
+                                else:
+                                    type_schema[k] = obj[k]
+                    elif t == "object": # pragma: no cover
+                        for k in ["properties", "required", "additionalProperties"]:
+                            if k in obj:
+                                # Recursively process properties if present
+                                if k == "properties":
+                                    type_schema[k] = _convert_type_arrays_to_anyof(obj[k])
+                                else:
+                                    type_schema[k] = obj[k]
+
+                    any_of.append(type_schema)
+
+            new_obj["anyOf"] = any_of
+            return new_obj
+        else:
+            # Recursively process all values
+            return {k: _convert_type_arrays_to_anyof(v) for k, v in obj.items()}
+    elif isinstance(obj, list):
+        # Process each item in the list
+        return [_convert_type_arrays_to_anyof(item) for item in obj]
+    else:
+        # Return primitive values as-is
+        return obj
diff --git a/scripts/reproduce_1383.py b/scripts/reproduce_1383.py
new file mode 100644
index 000000000..8a1500a4e
--- /dev/null
+++ b/scripts/reproduce_1383.py
@@ -0,0 +1,82 @@
+#!/usr/bin/env python3
+"""
+Reproduction script for issue #1383: Dynamic schema creation fails with nested optionals
+
+This script demonstrates the issue where JSON schemas with union types specified as arrays
+(e.g., {"type": ["string", "null"]}) fail with outlines-core 0.1.26, and shows how the
+Python-side preprocessing resolves the issue.
+"""
+
+import json
+from typing import Optional
+from pydantic import BaseModel
+from outlines_core.fsm.outlines_core_rs import build_regex_from_schema
+from outlines.types import JsonSchema
+
+# Test case 1: Simple optional field (this should work)
+simple_schema = {
+    "type": "object",
+    "properties": {"name": {"type": "string"}, "age": {"type": ["integer", "null"]}},
+    "required": ["name"],
+}
+
+# Test case 2: Nested optional field (this should fail)
+nested_optional_schema = {
+    "type": "object",
+    "properties": {
+        "person": {
+            "type": ["object", "null"],
+            "properties": {
+                "name": {"type": "string"},
+                "age": {"type": ["integer", "null"]},
+            },
+        }
+    },
+}
+
+# Test case 3: Array with optional items
+array_optional_schema = {
+    "type": "object",
+    "properties": {"items": {"type": "array", "items": {"type": ["string", "null"]}}},
+}
+
+
+def test_schema(schema_dict, name):
+    print(f"\n=== Testing {name} ===")
+    print(f"Schema: {json.dumps(schema_dict, indent=2)}")
+
+    print("\n--- Testing direct build_regex_from_schema ---")
+    try:
+        # Try to create regex from schema
+        schema_str = json.dumps(schema_dict)
+        regex = build_regex_from_schema(schema_str)
+        print(f"✓ Success! Generated regex: {regex[:100]}...")
+    except Exception as e:
+        print(f"✗ Error: {type(e).__name__}: {e}")
+
+    print("\n--- Testing JsonSchema class (with workaround) ---")
+    try:
+        # Try to create JsonSchema object
+        json_schema = JsonSchema(schema_dict)
+        print("✓ JsonSchema object created successfully")
+        print(f"  Preprocessed schema: {json_schema.schema[:200]}...")
+
+        # Try to convert to regex
+        from outlines.types.dsl import to_regex
+
+        regex = to_regex(json_schema)
+        print(f"✓ Regex generated successfully: {regex[:100]}...")
+
+    except Exception as e:
+        print(f"✗ Error: {type(e).__name__}: {e}")
+        import traceback
+
+        traceback.print_exc()
+
+
+if __name__ == "__main__":
+    print("Testing issue #1383: Dynamic schema creation fails with nested optionals")
+
+    test_schema(simple_schema, "Simple optional field")
+    test_schema(nested_optional_schema, "Nested optional field")
+    test_schema(array_optional_schema, "Array with optional items")
diff --git a/tests/types/test_json_schema_union_types.py b/tests/types/test_json_schema_union_types.py
new file mode 100644
index 000000000..681b781f3
--- /dev/null
+++ b/tests/types/test_json_schema_union_types.py
@@ -0,0 +1,158 @@
+"""Test cases for JSON schemas with union types."""
+
+import json
+
+from outlines.types.json_schema_utils import preprocess_schema_for_union_types
+
+
+class TestJsonSchemaUnionTypes:
+    """Test cases for JSON schemas with union types."""
+
+    def test_invalid_schema(self):
+        """Test invalid schema."""
+        schema = 'foo'
+        result = preprocess_schema_for_union_types(schema)
+        assert result == schema
+
+    def test_simple_union_type(self):
+        """Test simple union type conversion."""
+        schema = {"type": ["string", "null"]}
+        result = json.loads(preprocess_schema_for_union_types(schema))
+
+        assert "anyOf" in result
+        assert len(result["anyOf"]) == 2
+        assert {"type": "string"} in result["anyOf"]
+        assert {"type": "null"} in result["anyOf"]
+
+    def test_union_with_constraints(self):
+        """Test union type with type-specific constraints."""
+        schema = {
+            "type": ["string", "null"],
+            "minLength": 5,
+            "maxLength": 10
+        }
+        result = json.loads(preprocess_schema_for_union_types(schema))
+
+        assert "anyOf" in result
+        assert len(result["anyOf"]) == 2
+        assert {"type": "string", "minLength": 5, "maxLength": 10} in result["anyOf"]
+        assert {"type": "null"} in result["anyOf"]
+
+    def test_nested_union_types(self):
+        """Test nested objects with union types."""
+        schema = {
+            "type": "object",
+            "properties": {
+                "name": {"type": ["string", "null"]},
+                "age": {"type": ["integer", "null"]}
+            }
+        }
+        result = json.loads(preprocess_schema_for_union_types(schema))
+
+        assert result["properties"]["name"]["anyOf"][0] == {"type": "string"}
+        assert result["properties"]["name"]["anyOf"][1] == {"type": "null"}
+        assert result["properties"]["age"]["anyOf"][0] == {"type": "integer"}
+        assert result["properties"]["age"]["anyOf"][1] == {"type": "null"}
+
+    def test_array_with_union_items(self):
+        """Test array with union type items."""
+        schema = {
+            "type": "array",
+            "items": {"type": ["string", "number", "null"]}
+        }
+        result = json.loads(preprocess_schema_for_union_types(schema))
+
+        assert "anyOf" in result["items"]
+        assert len(result["items"]["anyOf"]) == 3
+        assert {"type": "string"} in result["items"]["anyOf"]
+        assert {"type": "number"} in result["items"]["anyOf"]
+        assert {"type": "null"} in result["items"]["anyOf"]
+
+    def test_no_union_types(self):
+        """Test schema without union types remains unchanged."""
+        schema = {
+            "type": "object",
+            "properties": {
+                "name": {"type": "string"},
+                "age": {"type": "integer"}
+            }
+        }
+        result = json.loads(preprocess_schema_for_union_types(schema))
+        assert result == schema
+
+    def test_string_input(self):
+        """Test that string input is properly handled."""
+        schema_str = '{"type": ["string", "null"]}'
+        result = json.loads(preprocess_schema_for_union_types(schema_str))
+
+        assert "anyOf" in result
+        assert len(result["anyOf"]) == 2
+
+    def test_ensure_ascii_parameter(self):
+        """Test that ensure_ascii parameter is respected."""
+        schema = {"type": "string", "pattern": "café"}
+
+        # With ensure_ascii=True (default)
+        result_ascii = preprocess_schema_for_union_types(schema, ensure_ascii=True)
+        assert "caf\\u00e9" in result_ascii
+
+        # With ensure_ascii=False
+        result_unicode = preprocess_schema_for_union_types(schema, ensure_ascii=False)
+        assert "café" in result_unicode
+
+    def test_union_array_type(self):
+        """Test union array type."""
+        schema = {
+            "type": ["array", "null"],
+            "items": {"type": "string"},
+            "minItems": 1,
+        }
+        result = json.loads(preprocess_schema_for_union_types(schema))
+
+        assert "anyOf" in result
+        assert len(result["anyOf"]) == 2
+        assert {"type": "array", "items": {"type": "string"}, "minItems": 1} in result["anyOf"]
+        assert {"type": "null"} in result["anyOf"]
+
+    def test_complex_nested_structure(self):
+        """Test a more complex nested structure with multiple union types."""
+        schema = {
+            "type": "object",
+            "properties": {
+                "id": {"type": "integer"},
+                "data": {
+                    "type": "object",
+                    "properties": {
+                        "value": {"type": ["string", "number", "null"], "minimum": 0},
+                        "metadata": {
+                            "type": ["object", "null"],
+                            "description": "Some metadata",
+                            "additionalProperties": False,
+                            "properties": {
+                                "created": {"type": "string"},
+                                "tags": {
+                                    "type": "array",
+                                    "items": {"type": ["string", "null"]}
+                                }
+                            }
+                        }
+                    }
+                }
+            }
+        }
+        result = json.loads(preprocess_schema_for_union_types(schema))
+
+        # Check value was converted
+        assert "anyOf" in result["properties"]["data"]["properties"]["value"]
+        assert len(result["properties"]["data"]["properties"]["value"]["anyOf"]) == 3
+
+        # Check metadata was converted
+        assert "anyOf" in result["properties"]["data"]["properties"]["metadata"]
+        assert len(result["properties"]["data"]["properties"]["metadata"]["anyOf"]) == 2
+
+        # Check tags items were converted
+        metadata_object = next(
+            item for item in result["properties"]["data"]["properties"]["metadata"]["anyOf"]
+            if item["type"] == "object"
+        )
+        assert "anyOf" in metadata_object["properties"]["tags"]["items"]

From ad7aa7b0eb1beacb63c37b82862ce609e2965322 Mon Sep 17 00:00:00 2001
From: Yakitori <89288375+Oni-giri@users.noreply.github.com>
Date: Sat, 19 Jul 2025 23:04:43 +0300
Subject: [PATCH 409/505] Update docs with OpenRouter guide (#1691)

The goal of this small PR is to provide some awareness and context in
the documentation regarding OpenRouter, which is fairly seamless to use
with Outlines, given the compatibility with the OpenAI API. It was
mentionned in https://github.com/dottxt-ai/outlines/issues/1687

---------

Co-authored-by: Robin Picard <robin.picard@sciencespo.fr>
---
 docs/features/models/openrouter.md | 44 ++++++++++++++++++++++++++++++
 mkdocs.yml                         |  1 +
 2 files changed, 45 insertions(+)
 create mode 100644 docs/features/models/openrouter.md

diff --git a/docs/features/models/openrouter.md b/docs/features/models/openrouter.md
new file mode 100644
index 000000000..948cf9b8a
--- /dev/null
+++ b/docs/features/models/openrouter.md
@@ -0,0 +1,44 @@
+# Openrouter
+
+!!! Installation
+
+    [OpenRouter](https://openrouter.ai/docs/api-reference/overview) uses the same API as OpenAI, so both services are [interoperable](./openai_compatible.md) using the `openai` library. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
+
+    You also need to have an Openrouter API key. This API key must either be set as an environment variable called `OPENAI_API_KEY` or be provided to the `openai.OpenAI` class when instantiating it.
+
+## Model Initialization
+
+To create a model instance, you can use the `from_openai` function. It takes 2 arguments:
+
+- `client`: an `openai.OpenAI` instance
+- `model_name`: the name of the model you want to use, defined as `provider/model`
+
+For instance:
+
+```python
+import outlines
+import openai
+
+# Create the client
+client = openai.OpenAI(
+    base_url="https://openrouter.ai/api/v1",
+    api_key="OPENAI_API_KEY",
+)
+
+# Create the model
+model = outlines.from_openai(
+    client,
+    "x-ai/grok-4"
+)
+```
+
+Leaving an empty string in the model name field will lead OpenRouter to use your default model defined in [settings](https://openrouter.ai/settings/preferences).
+
+The [OpenRouter](https://openrouter.ai/models) website lists available models. Keep in mind that some models do not support `json_schema` response formats and may return a 400 error code as a result.
+
+## Related Documentation
+
+For specific implementations that use OpenAI-compatible APIs:
+
+- [OpenAI](./openai.md): The original OpenAI API implementation
+- [OpenAI compatible API](./openai_compatible.md): Details on how to use OpenAI-compatible APIs
diff --git a/mkdocs.yml b/mkdocs.yml
index 2bb854383..5225fdf68 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -138,6 +138,7 @@ nav:
                     - Ollama: features/models/ollama.md
                     - OpenAI: features/models/openai.md
                     - OpenAI compatible API: features/models/openai_compatible.md
+                    - OpenRouter: features/models/openrouter.md
                     - SGLang: features/models/sglang.md
                     - TGI: features/models/tgi.md
                     - Transformers: features/models/transformers.md

From 665f11ad54861ddaf3edd89b1de33d7ac834b1d9 Mon Sep 17 00:00:00 2001
From: Neil Mehta <neil@lmstudio.ai>
Date: Sat, 19 Jul 2025 16:37:11 -0400
Subject: [PATCH 410/505] Fix broken wheel builds caused by ambiguous packaging
 instructions (#1690)

---
 pyproject.toml | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index d8cbcf11f..db3b4f9cc 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -101,11 +101,11 @@ repository = "https://github.com/dottxt-ai/outlines"
 file="README.md"
 content-type = "text/markdown"
 
-[tool.setuptools]
-packages = ["outlines"]
+[tool.setuptools.packages.find]
+include = ["outlines*"]
 
 [tool.setuptools.package-data]
-"outlines" = ["py.typed"]
+"outlines" = ["py.typed", "grammars/*.lark"]
 
 [tool.setuptools_scm]
 write_to = "outlines/_version.py"

From ba0505db147638076da049913f044f6b05e8ca7d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 22 Jul 2025 14:44:30 +0200
Subject: [PATCH 411/505] Create the backends module and add backends for
 xgrammar and llguidance

---
 outlines/backends/__init__.py                | 160 ++++++++++
 outlines/backends/base.py                    |  88 ++++++
 outlines/backends/llguidance.py              | 298 +++++++++++++++++++
 outlines/backends/outlines_core.py           | 109 +++++++
 outlines/backends/xgrammar.py                | 110 +++++++
 outlines/generator.py                        |  69 +++--
 outlines/models/base.py                      |  42 ++-
 outlines/models/llamacpp.py                  |   9 +-
 outlines/models/transformers.py              |   1 +
 outlines/processors/base_logits_processor.py |   7 +
 pyproject.toml                               |   6 +
 11 files changed, 859 insertions(+), 40 deletions(-)
 create mode 100644 outlines/backends/__init__.py
 create mode 100644 outlines/backends/base.py
 create mode 100644 outlines/backends/llguidance.py
 create mode 100644 outlines/backends/outlines_core.py
 create mode 100644 outlines/backends/xgrammar.py

diff --git a/outlines/backends/__init__.py b/outlines/backends/__init__.py
new file mode 100644
index 000000000..771580181
--- /dev/null
+++ b/outlines/backends/__init__.py
@@ -0,0 +1,160 @@
+"""Module to define the backends in charge of creating logits processors."""
+
+import interegular
+
+from outlines.backends.base import (
+    BaseBackend,
+    LogitsProcessorType,
+)
+from outlines.backends.llguidance import LLGuidanceBackend
+from outlines.backends.outlines_core import OutlinesCoreBackend
+from outlines.backends.xgrammar import XGrammarBackend
+from outlines.models import SteerableModel
+
+
+CFG_DEFAULT_BACKEND = "llguidance"
+FSM_DEFAULT_BACKEND = "outlines_core"
+JSON_SCHEMA_DEFAULT_BACKEND = "outlines_core"
+REGEX_DEFAULT_BACKEND = "outlines_core"
+
+
+def _get_backend(backend_name: str, model: SteerableModel) -> BaseBackend:
+    """Create a Backend instance.
+
+    Parameters
+    ----------
+    backend_name: str
+        The name of the backend to get.
+    model: Model
+        The Outlines model of the user.
+
+    Returns
+    -------
+    backend: BaseBackend
+        The backend instance.
+
+    """
+    if backend_name == "outlines_core":
+        return OutlinesCoreBackend(model)
+    elif backend_name == "xgrammar":
+        return XGrammarBackend(model)
+    elif backend_name == "llguidance":
+        return LLGuidanceBackend(model)
+    else:
+        raise ValueError(f"Backend {backend_name} not supported")
+
+
+def get_json_schema_logits_processor(
+    backend_name: str | None,
+    model: SteerableModel,
+    json_schema: str,
+) -> LogitsProcessorType:
+    """Create a logits processor from a JSON schema.
+
+    Parameters
+    ----------
+    backend_name: str | None
+        The name of the backend to use.
+    model: Model
+        The Outlines model of the user.
+    json_schema: str
+        The JSON schema to create a logits processor from.
+
+    Returns
+    -------
+    LogitsProcessorType
+        The logits processor.
+
+    """
+    backend = _get_backend(
+        backend_name or JSON_SCHEMA_DEFAULT_BACKEND,
+        model,
+    )
+    return backend.get_json_schema_logits_processor(json_schema)
+
+
+def get_regex_logits_processor(
+    backend_name: str | None,
+    model: SteerableModel,
+    regex: str,
+) -> LogitsProcessorType:
+    """Create a logits processor from a regex.
+
+    Parameters
+    ----------
+    backend_name: str | None
+        The name of the backend to use.
+    model: Model
+        The Outlines model of the user.
+    regex: str
+        The regex to create a logits processor from.
+
+    Returns
+    -------
+    LogitsProcessorType
+        The logits processor.
+
+    """
+    backend = _get_backend(
+        backend_name or REGEX_DEFAULT_BACKEND,
+        model,
+    )
+    return backend.get_regex_logits_processor(regex)
+
+
+def get_cfg_logits_processor(
+    backend_name: str | None,
+    model: SteerableModel,
+    grammar: str,
+) -> LogitsProcessorType:
+    """Create a logits processor from a context-free grammar.
+
+    Parameters
+    ----------
+    backend_name: str | None
+        The name of the backend to use.
+    model: Model
+        The Outlines model of the user.
+    grammar: str
+        The context-free grammar to create a logits processor from.
+
+    Returns
+    -------
+    LogitsProcessorType
+        The logits processor.
+
+    """
+    backend = _get_backend(
+        backend_name or CFG_DEFAULT_BACKEND,
+        model,
+    )
+    return backend.get_cfg_logits_processor(grammar)
+
+
+def get_fsm_logits_processor(
+    backend_name: str | None,
+    model: SteerableModel,
+    fsm: interegular,
+) -> LogitsProcessorType:
+    """Create a logits processor from an interegular FSM.
+
+    Parameters
+    ----------
+    backend_name: str | None
+        The name of the backend to use.
+    model: Model
+        The Outlines model of the user.
+    fsm: interegular.fsm.FSM
+        The interegular FSM to create a logits processor from.
+
+    Returns
+    -------
+    LogitsProcessorType
+        The logits processor.
+
+    """
+    backend = _get_backend(
+        backend_name or FSM_DEFAULT_BACKEND,
+        model,
+    )
+    return backend.get_fsm_logits_processor(fsm)
diff --git a/outlines/backends/base.py b/outlines/backends/base.py
new file mode 100644
index 000000000..a8723226a
--- /dev/null
+++ b/outlines/backends/base.py
@@ -0,0 +1,88 @@
+"""Base class for all backends."""
+
+from abc import ABC, abstractmethod
+from typing import Any
+
+from interegular.fsm import FSM
+
+
+LogitsProcessorType = Any
+
+
+class BaseBackend(ABC):
+    """Base class for all backends.
+
+    The subclasses must implement methods that create a logits processor
+    from a JSON schema, regex, CFG or FSM.
+
+    """
+
+    @abstractmethod
+    def get_json_schema_logits_processor(
+        self, json_schema: str
+    ) -> LogitsProcessorType:
+        """Create a logits processor from a JSON schema.
+
+        Parameters
+        ----------
+        json_schema: str
+            The JSON schema to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessorType
+            The logits processor.
+
+        """
+        ...
+
+    @abstractmethod
+    def get_regex_logits_processor(self, regex: str) -> LogitsProcessorType:
+        """Create a logits processor from a regex.
+
+        Parameters
+        ----------
+        regex: str
+            The regex to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessorType
+            The logits processor.
+
+        """
+        ...
+
+    @abstractmethod
+    def get_cfg_logits_processor(self, grammar: str) -> LogitsProcessorType:
+        """Create a logits processor from a context-free grammar.
+
+        Parameters
+        ----------
+        grammar: str
+            The context-free grammar to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessorType
+            The logits processor.
+
+        """
+        ...
+
+    @abstractmethod
+    def get_fsm_logits_processor(self, fsm: FSM) -> LogitsProcessorType:
+        """Create a logits processor from an interegular FSM.
+
+        Parameters
+        ----------
+        fsm: interegular.fsm.FSM
+            The interegular FSM to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessorType
+            The logits processor.
+
+        """
+        ...
diff --git a/outlines/backends/llguidance.py b/outlines/backends/llguidance.py
new file mode 100644
index 000000000..cd9a8d9b8
--- /dev/null
+++ b/outlines/backends/llguidance.py
@@ -0,0 +1,298 @@
+"""Backend class for LLGuidance."""
+
+import warnings
+from typing import TYPE_CHECKING
+
+from llguidance import LLMatcher
+
+from outlines.backends.base import BaseBackend
+from outlines.models import LlamaCpp, MLXLM, SteerableModel, Transformers
+from outlines.processors.base_logits_processor import (
+    OutlinesLogitsProcessor,
+    TensorType
+)
+
+if TYPE_CHECKING:
+    from llguidance import LLGTokenizer
+
+
+SUPPORTED_TENSOR_LIBRARIES = ["numpy", "mlx", "torch"]
+
+
+class LLGuidanceLogitsProcessor(OutlinesLogitsProcessor):
+    """Logits Processor for the LLGuidance backend."""
+
+    def __init__(
+        self,
+        grammar: str,
+        llg_tokenizer,
+        tensor_library_name: str,
+    ) -> None:
+        """
+        Parameters
+        ----------
+        grammar: str
+            The grammar spec to use to create the LLMatcher
+        llg_tokenizer: LLTokenizer
+            The LLGuidance tokenizer
+        tensor_library_name: str
+            The name of the tensor library used by the model
+
+        """
+        if tensor_library_name not in SUPPORTED_TENSOR_LIBRARIES:
+            raise TypeError(f"Unsupported tensor library: {tensor_library_name}")
+
+        self.is_first_token = True
+        self.grammar = grammar
+        self.llg_tokenizer = llg_tokenizer
+        self.tensor_library_name = tensor_library_name
+        super().__init__(tensor_library_name)
+
+    def _setup(self, batch_size: int) -> None:
+        """Setup the LLMatchers, the bitmask and some functions used in the
+        `process_logits` method.
+
+        This method is called when the first token is generated instead of
+        at initialization because we need to know the batch size.
+
+        Parameters
+        ----------
+        batch_size: int
+            The batch size of the input
+
+        """
+        self.ll_matchers = [
+            LLMatcher(self.llg_tokenizer, self.grammar)
+            for _ in range(batch_size)
+        ]
+
+        # we must adapt the bitmask creation and the bias function to the
+        # tensor library used by the model
+        if self.tensor_library_name == "torch":
+            import llguidance.torch
+
+            self.bitmask = llguidance.torch.allocate_token_bitmask(batch_size, self.llg_tokenizer.vocab_size)
+            self._bias_logits = self._bias_logits_torch
+        elif self.tensor_library_name == "numpy":
+            import llguidance.numpy
+
+            self.bitmask = llguidance.numpy.allocate_token_bitmask(batch_size, self.llg_tokenizer.vocab_size)
+            self._bias_logits = self._bias_logits_numpy
+        elif self.tensor_library_name == "mlx": # pragma: no cover
+            import llguidance.numpy
+
+            self.bitmask = llguidance.numpy.allocate_token_bitmask(batch_size, self.llg_tokenizer.vocab_size)
+            self._bias_logits = self._bias_logits_mlx
+        else: # pragma: no cover
+            raise ValueError(f"Unsupported tensor library: {self.tensor_library_name}")
+
+    def _bias_logits_mlx( # pragma: no cover
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for the MLX backend."""
+        import llguidance.mlx
+        import llguidance.numpy
+
+        biased_logits_array = []
+        for i in range(self.tensor_adapter.shape(input_ids)[0]):
+            llguidance.numpy.fill_next_token_bitmask(self.ll_matchers[i], self.bitmask, i)
+            biased_logits = llguidance.mlx.apply_token_bitmask(
+                logits[i], self.bitmask[i] # type: ignore
+            )
+            biased_logits_array.append(biased_logits)
+
+        return self.tensor_adapter.concatenate(biased_logits_array)
+
+    def _bias_logits_torch(
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for the Torch backend."""
+        import llguidance.torch
+
+        for i in range(self.tensor_adapter.shape(input_ids)[0]):
+            llguidance.torch.fill_next_token_bitmask(self.ll_matchers[i], self.bitmask, i)
+            llguidance.torch.apply_token_bitmask_inplace(
+                logits[i], self.bitmask[i] # type: ignore
+            )
+
+        return logits
+
+    def _bias_logits_numpy(
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for the Numpy backend."""
+        import llguidance.numpy
+
+        for i in range(self.tensor_adapter.shape(input_ids)[0]):
+            llguidance.numpy.fill_next_token_bitmask(self.ll_matchers[i], self.bitmask, i)
+            llguidance.numpy.apply_token_bitmask_inplace(
+                logits[i], self.bitmask[i] # type: ignore
+            )
+
+        return logits
+
+    def process_logits(
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Use the instances of LLMatcher to bias the logits.
+
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences.
+        logits
+            The logits for the current generation step.
+
+        Returns
+        -------
+        TensorType
+            The biased logits.
+
+        """
+        if self.is_first_token:
+            self._setup(self.tensor_adapter.shape(input_ids)[0])
+            self.is_first_token = False
+
+        # we do not make the matchers consume the last token during the first
+        # generation step because no tokens have been generated yet
+        else:
+            for i in range(self.tensor_adapter.shape(input_ids)[0]):
+                sequence = input_ids[i] # type: ignore
+                last_token = sequence[-1].item()
+                self.ll_matchers[i].consume_token(last_token)
+                error = self.ll_matchers[i].get_error()
+                if error:
+                    warnings.warn(f"Error in LLMatcher: {error}")
+
+        return self._bias_logits(input_ids, logits)
+
+
+class LLGuidanceBackend(BaseBackend):
+    """Backend for LLGuidance."""
+
+    def __init__(self, model: SteerableModel):
+        """
+        Parameters
+        ----------
+        model
+            The Outlines model of the user.
+
+        """
+        import llguidance as llg
+
+        self.llg = llg
+        self.tensor_library_name = model.tensor_library_name
+        self.llg_tokenizer = self._create_llg_tokenizer(model)
+
+    def _create_llg_tokenizer(self, model: SteerableModel) -> "LLGTokenizer":
+        """Create an llg tokenizer from the Outlines model's tokenizer.
+
+        Parameters
+        ----------
+        model: Model
+            The Outlines model.
+
+        Returns
+        -------
+        LLGTokenizer
+            The llg tokenizer.
+
+        """
+        if isinstance(model, Transformers):
+            import llguidance.hf
+
+            return llguidance.hf.from_tokenizer(model.hf_tokenizer)
+
+        elif isinstance(model, LlamaCpp):
+            import llama_cpp
+            import llguidance.llamacpp
+
+            vocab = llama_cpp.llama_model_get_vocab(model.model.model)
+            return llguidance.llamacpp.lltokenizer_from_vocab(vocab)
+
+        elif isinstance(model, MLXLM): # pragma: no cover
+            import llguidance.hf
+
+            return llguidance.hf.from_tokenizer(
+                model.mlx_tokenizer._tokenizer
+            )
+
+        else: # pragma: no cover
+            raise ValueError(
+                f"Unsupported model type: {type(model)}. "
+                "Llguidance only supports LlamaCpp, MLXLM "
+                "and Transformers models."
+            )
+
+    def get_json_schema_logits_processor(
+        self, json_schema: str
+    ) -> LLGuidanceLogitsProcessor:
+        """Create a logits processor from a JSON schema.
+
+        Parameters
+        ----------
+        json_schema: str
+            The JSON schema to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        grammar_spec = self.llg.grammar_from("json_schema", json_schema)
+        return LLGuidanceLogitsProcessor(
+            grammar_spec, self.llg_tokenizer, self.tensor_library_name
+        )
+
+    def get_regex_logits_processor(
+        self, regex: str
+    ) -> LLGuidanceLogitsProcessor:
+        """Create a logits processor from a regex.
+
+        Parameters
+        ----------
+        regex: str
+            The regex to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        grammar_spec = self.llg.grammar_from("regex", regex)
+        return LLGuidanceLogitsProcessor(
+            grammar_spec, self.llg_tokenizer, self.tensor_library_name
+        )
+
+    def get_cfg_logits_processor(
+        self, grammar: str
+    ) -> LLGuidanceLogitsProcessor:
+        """Create a logits processor from a context-free grammar.
+
+        Parameters
+        ----------
+        grammar: str
+            The context-free grammar to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        # We try both lark and ebnf
+        try:
+            grammar_spec = self.llg.grammar_from("grammar", grammar)
+        except ValueError:
+            grammar_spec = self.llg.grammar_from("lark", grammar)
+        return LLGuidanceLogitsProcessor(
+            grammar_spec, self.llg_tokenizer, self.tensor_library_name
+        )
+
+    def get_fsm_logits_processor(self, fsm):
+        raise NotImplementedError(
+            "LLGuidanceBackend does not support FSM logits processors. "
+            "Use the outlines_core backend instead."
+        )
diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
new file mode 100644
index 000000000..84a8896fd
--- /dev/null
+++ b/outlines/backends/outlines_core.py
@@ -0,0 +1,109 @@
+"""Backend class for Outlines Core."""
+
+from interegular.fsm import FSM
+
+from outlines.backends.base import BaseBackend
+from outlines.models import SteerableModel
+from outlines.processors import (
+    CFGLogitsProcessor,
+    GuideLogitsProcessor,
+    RegexLogitsProcessor,
+)
+from outlines.processors.guide import RegexGuide
+from outlines_core.fsm.json_schema import build_regex_from_schema
+
+
+class OutlinesCoreBackend(BaseBackend):
+    """Backend for Outlines Core."""
+
+    def __init__(self, model: SteerableModel):
+        """
+        Parameters
+        ----------
+        model
+            The Outlines model of the user.
+
+        """
+        self.tokenizer = model.tokenizer
+        self.tensor_library_name = model.tensor_library_name
+
+    def get_json_schema_logits_processor(
+        self, json_schema: str
+    ) -> RegexLogitsProcessor:
+        """Create a logits processor from a JSON schema.
+
+        Parameters
+        ----------
+        json_schema: str
+            The JSON schema to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        regex = build_regex_from_schema(json_schema)
+        return self.get_regex_logits_processor(regex)
+
+    def get_regex_logits_processor(self, regex: str) -> RegexLogitsProcessor:
+        """Create a logits processor from a regex.
+
+        Parameters
+        ----------
+        regex: str
+            The regex to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        return RegexLogitsProcessor(
+            regex,
+            self.tokenizer,
+            self.tensor_library_name,
+        )
+
+    def get_cfg_logits_processor(self, grammar: str) -> CFGLogitsProcessor:
+        """Create a logits processor from a context-free grammar.
+
+        Parameters
+        ----------
+        grammar: str
+            The context-free grammar to create a logits processor from. For
+            Outlines Core, the grammar must use the Lark format.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        return CFGLogitsProcessor(
+            grammar,
+            self.tokenizer,
+            self.tensor_library_name,
+        )
+
+    def get_fsm_logits_processor(self, fsm: FSM) -> GuideLogitsProcessor:
+        """Create a logits processor from an interegular FSM.
+
+        Parameters
+        ----------
+        fsm: interegular.fsm.FSM
+            The interegular FSM to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        guide = RegexGuide.from_interegular_fsm(fsm, self.tokenizer)
+        return GuideLogitsProcessor(
+            self.tokenizer,
+            guide,
+            self.tensor_library_name,
+        )
diff --git a/outlines/backends/xgrammar.py b/outlines/backends/xgrammar.py
new file mode 100644
index 000000000..4e7a4071c
--- /dev/null
+++ b/outlines/backends/xgrammar.py
@@ -0,0 +1,110 @@
+"""Backend class for XGrammar."""
+
+from typing import TYPE_CHECKING
+
+from outlines.backends.base import BaseBackend
+from outlines.models import SteerableModel
+from outlines.models.transformers import Transformers
+
+if TYPE_CHECKING:
+    from xgrammar.contrib.hf import LogitsProcessor
+
+
+class XGrammarBackend(BaseBackend):
+    """Backend for XGRammar."""
+
+    def __init__(self, model: SteerableModel):
+        """
+        Parameters
+        ----------
+        model
+            The Outlines model of the user.
+
+        """
+        import xgrammar as xgr
+        from transformers import AutoConfig
+
+        if not isinstance(model, Transformers):
+            raise ValueError(
+                "The xgrammar backend only supports Transformers models"
+            )
+
+        self.xgr = xgr
+        vocab_size = AutoConfig.from_pretrained(
+            model.model.config._name_or_path
+        ).vocab_size
+        tokenizer_info = self.xgr.TokenizerInfo.from_huggingface(
+            model.hf_tokenizer,
+            vocab_size=vocab_size
+        )
+        self.grammar_compiler = self.xgr.GrammarCompiler(tokenizer_info)
+
+    def get_json_schema_logits_processor(
+        self, json_schema: str
+    ) -> "LogitsProcessor":
+        """Create a logits processor from a JSON schema.
+
+        Parameters
+        ----------
+        json_schema: str
+            The JSON schema to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        compiled_grammar = self.grammar_compiler.compile_json_schema(
+            json_schema
+        )
+        xgr_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
+            compiled_grammar
+        )
+        return xgr_logits_processor
+
+    def get_regex_logits_processor(self, regex: str) -> "LogitsProcessor":
+        """Create a logits processor from a regex.
+
+        Parameters
+        ----------
+        regex: str
+            The regex to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        compiled_grammar = self.grammar_compiler.compile_regex(regex)
+        xgr_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
+            compiled_grammar
+        )
+        return xgr_logits_processor
+
+    def get_cfg_logits_processor(self, grammar: str) -> "LogitsProcessor":
+        """Create a logits processor from a context-free grammar.
+
+        Parameters
+        ----------
+        grammar: str
+            The context-free grammar to create a logits processor from.
+
+        Returns
+        -------
+        LogitsProcessor
+            The logits processor to use to constrain the generation.
+
+        """
+        compiled_grammar = self.grammar_compiler.compile_grammar(grammar)
+        xgr_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
+            compiled_grammar
+        )
+        return xgr_logits_processor
+
+    def get_fsm_logits_processor(self, fsm):
+        raise NotImplementedError(
+            "XGrammarBackend does not support FSM logits processors. "
+            "Use the outlines_core backend instead."
+        )
diff --git a/outlines/generator.py b/outlines/generator.py
index f9c5af2ce..ab41562c2 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -15,14 +15,14 @@
     SteerableModel,
 )
 from outlines.models.base import AsyncModel, Model
-from outlines.processors import (
-    CFGLogitsProcessor,
-    GuideLogitsProcessor,
-    OutlinesLogitsProcessor,
-    RegexLogitsProcessor,
+from outlines.backends import (
+    get_cfg_logits_processor,
+    get_fsm_logits_processor,
+    get_json_schema_logits_processor,
+    get_regex_logits_processor,
 )
-from outlines.processors.guide import RegexGuide
-from outlines.types import CFG, FSM
+from outlines.backends.base import LogitsProcessorType
+from outlines.types import CFG, FSM, JsonSchema
 from outlines.types.dsl import python_types_to_terms, to_regex
 
 
@@ -222,9 +222,14 @@ class SteerableGenerator:
     The 2 parameters are mutually exclusive.
 
     """
-    logits_processor: Optional[OutlinesLogitsProcessor]
+    logits_processor: Optional[LogitsProcessorType]
 
-    def __init__(self, model: SteerableModel, output_type: Optional[Any]):
+    def __init__(
+        self,
+        model: SteerableModel,
+        output_type: Optional[Any],
+        backend_name: Optional[str] = None,
+    ):
         """
         Parameters
         ----------
@@ -232,6 +237,8 @@ def __init__(self, model: SteerableModel, output_type: Optional[Any]):
             An instance of an Outlines model.
         output_type
             The output type expressed as a Python type
+        backend_name
+            The name of the backend to use to create the logits processor.
 
         """
         self.model = model
@@ -241,30 +248,34 @@ def __init__(self, model: SteerableModel, output_type: Optional[Any]):
             term = python_types_to_terms(output_type)
             if isinstance(term, CFG):
                 cfg_string = term.definition
-                self.logits_processor = CFGLogitsProcessor(
+                self.logits_processor = get_cfg_logits_processor(
+                    backend_name,
+                    model,
                     cfg_string,
-                    self.model.tokenizer,
-                    self.model.tensor_library_name,
                 )
             elif isinstance(term, FSM):
-                guide = RegexGuide.from_interegular_fsm(
+                self.logits_processor = get_fsm_logits_processor(
+                    backend_name,
+                    model,
                     term.fsm,
-                    self.model.tokenizer,
                 )
-                self.logits_processor = GuideLogitsProcessor(
-                    self.model.tokenizer, guide, self.model.tensor_library_name
+            elif isinstance(term, JsonSchema):
+                self.logits_processor = get_json_schema_logits_processor(
+                    backend_name,
+                    model,
+                    term.schema,
                 )
             else:
                 regex_string = to_regex(term)
-                self.logits_processor = RegexLogitsProcessor(
+                self.logits_processor = get_regex_logits_processor(
+                    backend_name,
+                    model,
                     regex_string,
-                    self.model.tokenizer,
-                    self.model.tensor_library_name,
                 )
 
     @classmethod
     def from_processor(
-        cls, model: SteerableModel, processor: OutlinesLogitsProcessor
+        cls, model: SteerableModel, processor: LogitsProcessorType
     ):
         """Create a generator from a logits processor.
 
@@ -273,14 +284,9 @@ def from_processor(
         model
             An instance of an Outlines model.
         processor
-            An instance of an OutlinesLogitsProcessor.
+            An instance of a logits processor.
 
         """
-        if not isinstance(processor, OutlinesLogitsProcessor):
-            raise TypeError(
-                "The processor argument must be an instance of "
-                "OutlinesLogitsProcessor"
-            )
         instance = cls.__new__(cls)
         instance.model = model
         instance.logits_processor = processor
@@ -351,8 +357,9 @@ def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
 def Generator(
     model: Union[Model, AsyncModel],
     output_type: Optional[Any] = None,
+    backend: Optional[str] = None,
     *,
-    processor: Optional[OutlinesLogitsProcessor] = None,
+    processor: Optional[LogitsProcessorType] = None,
 ) -> Union[SteerableGenerator, BlackBoxGenerator, AsyncBlackBoxGenerator]:
     """Create a generator for the given model and output parameters.
 
@@ -367,8 +374,12 @@ def Generator(
     output_type
         The output type expressed as a Python type or a type defined in the
         outlines.types.dsl module.
+    backend
+        The name of the backend to use to create the logits processor. Only
+        used for steerable models if there is an output type and `processor` is
+        not provided.
     processor
-        An instance of an OutlinesLogitsProcessor.
+        An instance of a logits processor.
 
     Returns
     -------
@@ -389,7 +400,7 @@ def Generator(
         if processor is not None:
             return SteerableGenerator.from_processor(model, processor) # type: ignore
         else:
-            return SteerableGenerator(model, output_type) # type: ignore
+            return SteerableGenerator(model, output_type, backend) # type: ignore
     else:
         if processor is not None:
             raise NotImplementedError(
diff --git a/outlines/models/base.py b/outlines/models/base.py
index 7092ed57f..4b948a56d 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -81,6 +81,7 @@ def __call__(
         self,
         model_input: Any,
         output_type: Optional[Any] = None,
+        backend: Optional[str] = None,
         **inference_kwargs: Any
     ) -> Any:
         """Call the model.
@@ -103,6 +104,10 @@ def __call__(
             The input provided by the user.
         output_type
             The output type provided by the user.
+        backend
+            The name of the backend to use to create the logits processor that
+            will be used to generate the response. Only used for steerable
+            models if `output_type` is provided.
         **inference_kwargs
             Additional keyword arguments to pass to the model.
 
@@ -114,12 +119,13 @@ def __call__(
         """
         from outlines import Generator
 
-        return Generator(self, output_type)(model_input, **inference_kwargs)
+        return Generator(self, output_type, backend)(model_input, **inference_kwargs)
 
     def batch(
         self,
         model_input: List[Any],
         output_type: Optional[Any] = None,
+        backend: Optional[str] = None,
         **inference_kwargs: Any
     ) -> List[Any]:
         """Make a batch call to the model (several inputs at once).
@@ -143,6 +149,10 @@ def batch(
             The list of inputs provided by the user.
         output_type
             The output type provided by the user.
+        backend
+            The name of the backend to use to create the logits processor that
+            will be used to generate the response. Only used for steerable
+            models if `output_type` is provided.
         **inference_kwargs
             Additional keyword arguments to pass to the model.
 
@@ -154,13 +164,14 @@ def batch(
         """
         from outlines import Generator
 
-        generator = Generator(self, output_type)
+        generator = Generator(self, output_type, backend)
         return generator.batch(model_input, **inference_kwargs) # type: ignore
 
     def stream(
         self,
         model_input: Any,
         output_type: Optional[Any] = None,
+        backend: Optional[str] = None,
         **inference_kwargs: Any
     ) -> Iterator[Any]:
         """Stream a response from the model.
@@ -186,6 +197,10 @@ def stream(
             The input provided by the user.
         output_type
             The output type provided by the user.
+        backend
+            The name of the backend to use to create the logits processor that
+            will be used to generate the response. Only used for steerable
+            models if `output_type` is provided.
         **inference_kwargs
             Additional keyword arguments to pass to the model.
 
@@ -197,7 +212,7 @@ def stream(
         """
         from outlines import Generator
 
-        generator = Generator(self, output_type)
+        generator = Generator(self, output_type, backend)
         return generator.stream(model_input, **inference_kwargs) # type: ignore
 
     @abstractmethod
@@ -311,6 +326,7 @@ async def __call__(
         self,
         model_input: Any,
         output_type: Optional[Any] = None,
+        backend: Optional[str] = None,
         **inference_kwargs: Any
     ) -> Any:
         """Call the model.
@@ -333,6 +349,10 @@ async def __call__(
             The input provided by the user.
         output_type
             The output type provided by the user.
+        backend
+            The name of the backend to use to create the logits processor that
+            will be used to generate the response. Only used for steerable
+            models if `output_type` is provided.
         **inference_kwargs
             Additional keyword arguments to pass to the model.
 
@@ -344,13 +364,14 @@ async def __call__(
         """
         from outlines import Generator
 
-        generator = Generator(self, output_type)
+        generator = Generator(self, output_type, backend)
         return await generator(model_input, **inference_kwargs)
 
     async def batch(
         self,
         model_input: List[Any],
         output_type: Optional[Any] = None,
+        backend: Optional[str] = None,
         **inference_kwargs: Any
     ) -> List[Any]:
         """Make a batch call to the model (several inputs at once).
@@ -374,6 +395,10 @@ async def batch(
             The list of inputs provided by the user.
         output_type
             The output type provided by the user.
+        backend
+            The name of the backend to use to create the logits processor that
+            will be used to generate the response. Only used for steerable
+            models if `output_type` is provided.
         **inference_kwargs
             Additional keyword arguments to pass to the model.
 
@@ -385,13 +410,14 @@ async def batch(
         """
         from outlines import Generator
 
-        generator = Generator(self, output_type)
+        generator = Generator(self, output_type, backend)
         return await generator.batch(model_input, **inference_kwargs) # type: ignore
 
     async def stream(
         self,
         model_input: Any,
         output_type: Optional[Any] = None,
+        backend: Optional[str] = None,
         **inference_kwargs: Any
     ) -> AsyncIterator[Any]:
         """Stream a response from the model.
@@ -417,6 +443,10 @@ async def stream(
             The input provided by the user.
         output_type
             The output type provided by the user.
+        backend
+            The name of the backend to use to create the logits processor that
+            will be used to generate the response. Only used for steerable
+            models if `output_type` is provided.
         **inference_kwargs
             Additional keyword arguments to pass to the model.
 
@@ -428,7 +458,7 @@ async def stream(
         """
         from outlines import Generator
 
-        generator = Generator(self, output_type)
+        generator = Generator(self, output_type, backend)
 
         async for chunk in generator.stream(model_input, **inference_kwargs):  # type: ignore
             yield chunk
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index ce69f96a3..67754c9bb 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,6 +1,5 @@
 """Integration with the `llama-cpp-python` library."""
 
-import warnings
 from functools import singledispatchmethod
 from typing import (
     TYPE_CHECKING,
@@ -247,8 +246,8 @@ def generate(
         """
         if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp due to bug in "
-                "the llama_cpp tokenizer"
+                "CFG generation is not supported for LlamaCpp with the "
+                + "outlines-core backend"
             )
 
         prompt = self.type_adapter.format_input(model_input)
@@ -307,8 +306,8 @@ def generate_stream(
         """
         if isinstance(output_type, CFGLogitsProcessor):
             raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp due to bug in "
-                "the llama_cpp tokenizer"
+                "CFG generation is not supported for LlamaCpp with the "
+                + "outlines-core backend"
             )
 
         prompt = self.type_adapter.format_input(model_input)
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 03fce82bd..2b8fca13e 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -234,6 +234,7 @@ def __init__(
 
         tokenizer.padding_side = "left"
         self.model = model
+        self.hf_tokenizer = tokenizer
         self.tokenizer = TransformerTokenizer(tokenizer)
         self.type_adapter = TransformersTypeAdapter(tokenizer=tokenizer)
 
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index bde129b2c..8d00f0c7b 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -31,6 +31,13 @@ def __init__(self, tensor_library_name: str):
             values are "jax", "mlx", "numpy", "tensorflow" and "torch". You
             must choose the library that your model is using.
         """
+        # Temporary fix as torch raises a warning that can cause can an error
+        # with python 3.12.
+        if tensor_library_name == "torch":
+            import torch._dynamo
+
+            torch._dynamo.config.suppress_errors = True
+
         tensor_adapter_class = tensor_adapters.get(tensor_library_name)
         if tensor_adapter_class is None:
             raise NotImplementedError(
diff --git a/pyproject.toml b/pyproject.toml
index db3b4f9cc..0b3fe4ec4 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -58,6 +58,8 @@ sglang = ["openai"]
 tgi = ["huggingface_hub"]
 transformers = ["accelerate", "datasets", "transformers"]
 vllm = ["openai"]
+xgrammar = ["xgrammar"]
+llguidance = ["llguidance"]
 test = [
     "pre-commit",
     "pytest",
@@ -89,6 +91,8 @@ test = [
     "dottxt",
     "sentencepiece",
     "mkdocs_gen_files",
+    "llguidance",
+    "xgrammar",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
 
@@ -170,6 +174,8 @@ module = [
     "tf-keras.*",
     "mkdocs_gen_files.*",
     "jsonpath_ng.*",
+    "llguidance.*",
+    "xgrammar.*",
 ]
 ignore_missing_imports = true
 

From d01013de675843799406bd425f0412613e30b84c Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 22 Jul 2025 14:44:49 +0200
Subject: [PATCH 412/505] Add tests for the backends

---
 tests/backends/test_backends.py      | 142 +++++++++++++++++++
 tests/backends/test_llguidance.py    | 198 +++++++++++++++++++++++++++
 tests/backends/test_outlines_core.py | 139 +++++++++++++++++++
 tests/backends/test_xgrammar.py      | 106 ++++++++++++++
 tests/models/test_llamacpp.py        |  64 ++++++++-
 tests/test_generator.py              |   5 -
 6 files changed, 642 insertions(+), 12 deletions(-)
 create mode 100644 tests/backends/test_backends.py
 create mode 100644 tests/backends/test_llguidance.py
 create mode 100644 tests/backends/test_outlines_core.py
 create mode 100644 tests/backends/test_xgrammar.py

diff --git a/tests/backends/test_backends.py b/tests/backends/test_backends.py
new file mode 100644
index 000000000..e4b00d637
--- /dev/null
+++ b/tests/backends/test_backends.py
@@ -0,0 +1,142 @@
+import interegular
+import outlines
+import pytest
+import transformers
+from xgrammar.contrib.hf import LogitsProcessor as XGrammarLogitsProcessor
+
+from outlines.backends import (
+    _get_backend,
+    get_json_schema_logits_processor,
+    get_regex_logits_processor,
+    get_cfg_logits_processor,
+    get_fsm_logits_processor
+)
+from outlines.backends.outlines_core import OutlinesCoreBackend
+from outlines.backends.llguidance import (
+    LLGuidanceBackend,
+    LLGuidanceLogitsProcessor
+)
+from outlines.backends.xgrammar import XGrammarBackend
+from outlines.processors.structured import (
+    CFGLogitsProcessor,
+    GuideLogitsProcessor,
+    RegexLogitsProcessor,
+)
+
+
+@pytest.fixture
+def model():
+    return outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+
+@pytest.fixture
+def json_schema():
+    return (
+        '{"type": "object", "properties": {"name": {"type": "string"}, '
+        + '"age": {"type": "integer"}}, "required": ["name", "age"], '
+        + '"additionalProperties": false}'
+    )
+
+@pytest.fixture
+def regex():
+    return r"[0-9]{3}"
+
+@pytest.fixture
+def cfg_lark():
+    return """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+
+@pytest.fixture
+def cfg_ebnf():
+    return """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+
+@pytest.fixture
+def fsm():
+    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
+
+
+def test_get_backend(model):
+    backend = _get_backend("outlines_core", model)
+    assert isinstance(backend, OutlinesCoreBackend)
+
+    backend = _get_backend("xgrammar", model)
+    assert isinstance(backend, XGrammarBackend)
+
+    backend = _get_backend("llguidance", model)
+    assert isinstance(backend, LLGuidanceBackend)
+
+    with pytest.raises(ValueError, match="not supported"):
+        _get_backend("not_supported", model)
+
+
+def test_get_json_schema_logits_processor(model, json_schema):
+    processor = get_json_schema_logits_processor("outlines_core", model, json_schema)
+    assert isinstance(processor, RegexLogitsProcessor)
+
+    processor = get_json_schema_logits_processor("llguidance", model, json_schema)
+    assert isinstance(processor, LLGuidanceLogitsProcessor)
+
+    processor = get_json_schema_logits_processor("xgrammar", model, json_schema)
+    assert isinstance(processor, XGrammarLogitsProcessor)
+
+
+def test_get_regex_logits_processor(model, regex):
+    processor = get_regex_logits_processor("outlines_core", model, regex)
+    assert isinstance(processor, RegexLogitsProcessor)
+
+    processor = get_regex_logits_processor("llguidance", model, regex)
+    assert isinstance(processor, LLGuidanceLogitsProcessor)
+
+    processor = get_regex_logits_processor("xgrammar", model, regex)
+    assert isinstance(processor, XGrammarLogitsProcessor)
+
+
+def test_get_cfg_logits_processor(model, cfg_lark, cfg_ebnf):
+    processor = get_cfg_logits_processor("outlines_core", model, cfg_lark)
+    assert isinstance(processor, CFGLogitsProcessor)
+
+    processor = get_cfg_logits_processor("llguidance", model, cfg_lark)
+    assert isinstance(processor, LLGuidanceLogitsProcessor)
+
+    processor = get_cfg_logits_processor("xgrammar", model, cfg_ebnf)
+    assert isinstance(processor, XGrammarLogitsProcessor)
+
+
+def test_get_fsm_logits_processor(model, fsm):
+    processor = get_fsm_logits_processor("outlines_core", model, fsm)
+    assert isinstance(processor, GuideLogitsProcessor)
+
+    with pytest.raises(
+        NotImplementedError,
+        match="LLGuidanceBackend does not support FSM logits processors"
+    ):
+        get_fsm_logits_processor("llguidance", model, fsm)
+
+    with pytest.raises(
+        NotImplementedError,
+        match="XGrammarBackend does not support FSM logits processors",
+    ):
+        get_fsm_logits_processor("xgrammar", model, fsm)
diff --git a/tests/backends/test_llguidance.py b/tests/backends/test_llguidance.py
new file mode 100644
index 000000000..7e0926e12
--- /dev/null
+++ b/tests/backends/test_llguidance.py
@@ -0,0 +1,198 @@
+import pytest
+
+import interegular
+import llama_cpp
+import llguidance.hf
+import numpy as np
+import torch
+import transformers
+from llguidance import LLTokenizer
+
+import outlines
+from outlines.backends.llguidance import (
+    LLGuidanceBackend,
+    LLGuidanceLogitsProcessor
+)
+
+try:
+    import mlx.core as mx
+    import mlx_lm
+    HAS_MLX = True
+except ImportError:
+    HAS_MLX = False
+
+
+def model_transformers():
+    return outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+
+def model_llamacpp():
+    return outlines.from_llamacpp(
+        llama_cpp.Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+    )
+
+def model_mlxlm():
+    return outlines.from_mlxlm(
+        *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    )
+
+@pytest.fixture
+def llg_tokenizer():
+    return llguidance.hf.from_tokenizer(
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+
+@pytest.fixture
+def llg_grammar_spec():
+    return (
+        '{"grammars": [{ "json_schema": {"type": "object", "properties":'
+        + ' {"name": {"type": "string"}, "age": {"type": "integer"}}, "requ'
+        + 'ired": ["name", "age"], "additionalProperties": false} }] }'
+    )
+
+@pytest.fixture
+def json_schema():
+    return (
+        '{"type": "object", "properties": {"name": {"type": "string"}, '
+        + '"age": {"type": "integer"}}, "required": ["name", "age"], '
+        + '"additionalProperties": false}'
+    )
+
+@pytest.fixture
+def regex():
+    return r"[0-9]{3}"
+
+@pytest.fixture
+def cfg_lark():
+    return """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+
+@pytest.fixture
+def cfg_ebnf():
+    return """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+
+@pytest.fixture
+def fsm():
+    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
+
+
+def test_llguidance_processor_torch(llg_grammar_spec, llg_tokenizer):
+    processor = LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "torch")
+    logits = torch.randn(2, llg_tokenizer.vocab_size)
+    input_ids = torch.randint(0, llg_tokenizer.vocab_size, (2, 10))
+    output = processor(input_ids, logits)
+    assert output.shape == (2, llg_tokenizer.vocab_size)
+    processor(input_ids, logits)
+
+
+def test_llguidance_processor_numpy(llg_grammar_spec, llg_tokenizer):
+    processor = LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "numpy")
+    logits = np.random.randn(2, llg_tokenizer.vocab_size)
+    input_ids = np.random.randint(0, llg_tokenizer.vocab_size, (2, 10))
+    output = processor(input_ids, logits)
+    assert output.shape == (2, llg_tokenizer.vocab_size)
+    processor(input_ids, logits)
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_llguidance_processor_mlx(llg_grammar_spec, llg_tokenizer):
+    processor = LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "mlx")
+    logits = mx.random.normal((2, llg_tokenizer.vocab_size))
+    input_ids = mx.random.randint(0, llg_tokenizer.vocab_size, (2, 10))
+    output = processor(input_ids, logits)
+    assert output.shape == (2, llg_tokenizer.vocab_size)
+    processor(input_ids, logits)
+
+
+def test_llguidance_processor_tensorflow(llg_grammar_spec, llg_tokenizer):
+    with pytest.raises(TypeError):
+        LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "tensorflow")
+
+
+def test_llguidance_processor_jax(llg_grammar_spec, llg_tokenizer):
+    with pytest.raises(TypeError):
+        LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "jax")
+
+
+models = [
+    (model_transformers(), "torch"),
+    (model_llamacpp(), "numpy"),
+]
+if HAS_MLX:
+    models.append((model_mlxlm(), "mlx"))
+
+@pytest.mark.parametrize("model, tensor_library_name", models)
+def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg_lark, cfg_ebnf, fsm):
+    # initialization
+    backend = LLGuidanceBackend(model)
+    assert isinstance(backend.llg_tokenizer, LLTokenizer)
+    assert backend.tensor_library_name == tensor_library_name
+
+    # json schema
+    processor = backend.get_json_schema_logits_processor(json_schema)
+    assert isinstance(processor, LLGuidanceLogitsProcessor)
+    generator = outlines.Generator(model, backend="llguidance", processor=processor)
+    response = generator("Hello, how are you?")
+    assert response[0] == "{"
+    assert "name" in response
+
+    # regex
+    processor = backend.get_regex_logits_processor(regex)
+    assert isinstance(processor, LLGuidanceLogitsProcessor)
+    generator = outlines.Generator(model, backend="llguidance", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+
+    # cfg lark
+    processor = backend.get_cfg_logits_processor(cfg_lark)
+    assert isinstance(processor, LLGuidanceLogitsProcessor)
+    generator = outlines.Generator(model, backend="llguidance", processor=processor)
+    response = generator("Hello, how are you?")
+    assert (
+        "+" in response
+        or "-" in response
+        or "*" in response
+        or "/" in response
+        or float(response.strip())
+    )
+
+    # cfg ebnf
+    processor = backend.get_cfg_logits_processor(cfg_ebnf)
+    assert isinstance(processor, LLGuidanceLogitsProcessor)
+    generator = outlines.Generator(model, backend="llguidance", processor=processor)
+    response = generator("Hello, how are you?")
+    assert response == "yes" or response == "no"
+
+    # fsm
+    with pytest.raises(
+        NotImplementedError,
+        match="LLGuidanceBackend does not support FSM logits processors",
+    ):
+        backend.get_fsm_logits_processor(fsm)
diff --git a/tests/backends/test_outlines_core.py b/tests/backends/test_outlines_core.py
new file mode 100644
index 000000000..4d40c3fbf
--- /dev/null
+++ b/tests/backends/test_outlines_core.py
@@ -0,0 +1,139 @@
+import pytest
+
+import interegular
+import llama_cpp
+import transformers
+
+import outlines
+from outlines.backends.outlines_core import OutlinesCoreBackend
+from outlines.processors.structured import (
+    CFGLogitsProcessor,
+    RegexLogitsProcessor,
+    GuideLogitsProcessor,
+)
+
+try:
+    import mlx_lm
+    HAS_MLX = True
+except ImportError:
+    HAS_MLX = False
+
+
+def model_transformers():
+    return outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+
+def model_llamacpp():
+    return outlines.from_llamacpp(
+        llama_cpp.Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+    )
+
+def model_mlxlm():
+    return outlines.from_mlxlm(
+        *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    )
+
+@pytest.fixture
+def json_schema():
+    return (
+        '{"type": "object", "properties": {"name": {"type": "string"}, '
+        + '"age": {"type": "integer"}}, "required": ["name", "age"], '
+        + '"additionalProperties": false}'
+    )
+
+@pytest.fixture
+def regex():
+    return r"[0-9]{3}"
+
+@pytest.fixture
+def cfg():
+    return """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+
+@pytest.fixture
+def fsm():
+    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
+
+
+models = [
+    (model_transformers(), "torch"),
+    (model_llamacpp(), "numpy"),
+]
+if HAS_MLX:
+    models.append((model_mlxlm(), "mlx"))
+
+@pytest.mark.parametrize("model, tensor_library_name", models)
+def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg, fsm):
+    # initialization
+    backend = OutlinesCoreBackend(model)
+    assert backend.tokenizer == model.tokenizer
+    assert backend.tensor_library_name == tensor_library_name
+
+    # json schema
+    processor = backend.get_json_schema_logits_processor(json_schema)
+    assert isinstance(processor, RegexLogitsProcessor)
+    generator = outlines.Generator(model, backend="outlines_core", processor=processor)
+    response = generator("Hello, how are you?")
+    assert response[0] == "{"
+    assert "name" in response
+
+    # regex
+    processor = backend.get_regex_logits_processor(regex)
+    assert isinstance(processor, RegexLogitsProcessor)
+    generator = outlines.Generator(model, backend="outlines_core", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+
+    # cfg
+    if isinstance(model, outlines.models.LlamaCpp):
+        with pytest.raises(
+            NotImplementedError,
+            match="CFG generation is not supported for LlamaCpp with the outlines-core backend",
+        ):
+            processor = backend.get_cfg_logits_processor(cfg)
+            generator = outlines.Generator(model, backend="outlines_core", processor=processor)
+            response = generator("Hello, how are you?")
+    else:
+        processor = backend.get_cfg_logits_processor(cfg)
+        assert isinstance(processor, CFGLogitsProcessor)
+        generator = outlines.Generator(model, backend="outlines_core", processor=processor)
+        response = generator("Hello, how are you?")
+        assert (
+            "+" in response
+            or "-" in response
+            or "*" in response
+            or "/" in response
+            or float(response.strip())
+        )
+
+    # fsm
+    processor = backend.get_fsm_logits_processor(fsm)
+    assert isinstance(processor, GuideLogitsProcessor)
+    generator = outlines.Generator(model, backend="outlines_core", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
diff --git a/tests/backends/test_xgrammar.py b/tests/backends/test_xgrammar.py
new file mode 100644
index 000000000..42d80dbf8
--- /dev/null
+++ b/tests/backends/test_xgrammar.py
@@ -0,0 +1,106 @@
+import pytest
+
+import interegular
+import llama_cpp
+import transformers
+from xgrammar import GrammarCompiler
+from xgrammar.contrib.hf import LogitsProcessor
+
+import outlines
+from outlines.backends.xgrammar import XGrammarBackend
+
+try:
+    import mlx_lm
+    HAS_MLX = True
+except ImportError:
+    HAS_MLX = False
+
+
+@pytest.fixture
+def model_transformers():
+    return outlines.from_transformers(
+        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
+        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
+    )
+
+@pytest.fixture
+def model_llamacpp():
+    return outlines.from_llamacpp(
+        llama_cpp.Llama.from_pretrained(
+            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
+            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        )
+    )
+
+@pytest.fixture
+def model_mlxlm():
+    return outlines.from_mlxlm(
+        *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    )
+
+@pytest.fixture
+def json_schema():
+    return (
+        '{"type": "object", "properties": {"name": {"type": "string"}, '
+        + '"age": {"type": "integer"}}, "required": ["name", "age"], '
+        + '"additionalProperties": false}'
+    )
+
+@pytest.fixture
+def regex():
+    return r"[0-9]{3}"
+
+@pytest.fixture
+def cfg():
+    return """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+
+@pytest.fixture
+def fsm():
+    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
+
+
+def test_xgrammar_backend(model_transformers, json_schema, regex, cfg, fsm):
+    # initialization
+    backend = XGrammarBackend(model_transformers)
+    assert isinstance(backend.grammar_compiler, GrammarCompiler)
+
+    # json schema
+    processor = backend.get_json_schema_logits_processor(json_schema)
+    assert isinstance(processor, LogitsProcessor)
+    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
+    response = generator("Hello, how are you?")
+    assert response[0] == "{"
+    assert "name" in response
+
+    # regex
+    processor = backend.get_regex_logits_processor(regex)
+    assert isinstance(processor, LogitsProcessor)
+    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+
+    # cfg
+    processor = backend.get_cfg_logits_processor(cfg)
+    assert isinstance(processor, LogitsProcessor)
+    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
+    response = generator("Hello, how are you?")
+    assert response == "yes" or response == "no"
+
+    # fsm
+    with pytest.raises(
+        NotImplementedError,
+        match="XGrammarBackend does not support FSM logits processors",
+    ):
+        backend.get_fsm_logits_processor(fsm)
+
+
+def test_xgrammar_backend_invalid_model(model_llamacpp):
+    with pytest.raises(
+        ValueError,
+        match="The xgrammar backend only supports Transformers models",
+    ):
+        XGrammarBackend(model_llamacpp)
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index e93114bae..76f028d8f 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -40,6 +40,36 @@ def model(tmp_path_factory):
         )
     )
 
+@pytest.fixture
+def lark_grammar():
+    return """
+?start: sum
+
+?sum: product
+| sum "+" product   -> add
+| sum "-" product   -> sub
+
+?product: atom
+| product "*" atom  -> mul
+| product "/" atom  -> div
+
+?atom: NUMBER           -> number
+| "-" atom         -> neg
+| "(" sum ")"
+
+%import common.NUMBER
+%import common.WS_INLINE
+
+%ignore WS_INLINE
+"""
+
+@pytest.fixture
+def ebnf_grammar():
+    return """
+root ::= answer
+answer ::= "yes" | "no"
+"""
+
 
 def test_llamacpp_simple(model):
     result = model.generate("Respond with one word. Not more.", None)
@@ -84,12 +114,21 @@ class Foo(Enum):
     assert result == "Foo" or result == "Bar"
 
 
-def test_llamacpp_cfg(model):
+def test_llamacpp_cfg(model, ebnf_grammar):
+    response = model("Respond with one word. Not more.", CFG(ebnf_grammar))
+    assert response in ["yes", "no"]
+
+
+def test_llamacpp_cfg_outlines_core(model, lark_grammar):
     with pytest.raises(
         NotImplementedError,
-        match="CFG generation is not supported for LlamaCpp"
+        match="not supported for LlamaCpp with the outlines-core backend"
     ):
-        model("Respond with one word. Not more.", CFG('start: "a"'), max_tokens=10)
+        model(
+            "Respond with one word. Not more.",
+            CFG(lark_grammar),
+            backend="outlines_core"
+        )
 
 
 def test_llamacpp_text_stop(model):
@@ -135,15 +174,26 @@ class Foo(BaseModel):
     assert x == "{"
 
 
-def test_llamacpp_stream_cfg(model):
+def test_llamacpp_stream_cfg(model, ebnf_grammar):
+    response = ""
+    for chunk in model.stream(
+        "Respond with one word. Not more.", CFG(ebnf_grammar)
+    ):
+        response += chunk
+    assert response in ["yes", "no"]
+
+
+def test_llamacpp_stream_cfg_outlines_core(model, lark_grammar):
     with pytest.raises(
         NotImplementedError,
-        match="CFG generation is not supported for LlamaCpp"
+        match="not supported for LlamaCpp with the outlines-core backend"
     ):
         for chunk in model.stream(
-            "Respond with one word. Not more.", CFG('start: "a"')
+            "Respond with one word. Not more.",
+            CFG(lark_grammar),
+            backend="outlines_core"
         ):
-            assert isinstance(chunk, str)
+            pass
 
 
 def test_llamacpp_stream_choice(model):
diff --git a/tests/test_generator.py b/tests/test_generator.py
index 5ce54dc4b..d75d03b0d 100644
--- a/tests/test_generator.py
+++ b/tests/test_generator.py
@@ -100,11 +100,6 @@ def test_steerable_generator_init_valid_processor(steerable_model, sample_proces
     assert generator.model == steerable_model
 
 
-def test_steerable_generator_init_invalid_processor(steerable_model):
-    with pytest.raises(TypeError):
-        SteerableGenerator.from_processor(steerable_model, Literal["foo", "bar"])
-
-
 def test_steerable_generator_init_cfg_output_type(steerable_model):
     generator = SteerableGenerator(steerable_model, CFG('start: "a"'))
     assert generator.model == steerable_model

From df9bce9f1086cf07b249e9fc8532474df337536b Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 22 Jul 2025 14:45:05 +0200
Subject: [PATCH 413/505] Document the use of the backends

---
 docs/features/advanced/backends.md | 56 ++++++++++++++++++++++++++++++
 docs/features/models/llamacpp.md   | 25 ++++++++++++-
 mkdocs.yml                         |  1 +
 3 files changed, 81 insertions(+), 1 deletion(-)
 create mode 100644 docs/features/advanced/backends.md

diff --git a/docs/features/advanced/backends.md b/docs/features/advanced/backends.md
new file mode 100644
index 000000000..68555f0bd
--- /dev/null
+++ b/docs/features/advanced/backends.md
@@ -0,0 +1,56 @@
+---
+title: Structured Generation Backends
+---
+
+# Structured Generation Backends
+
+Outlines relies on a structured generation backend to control text generation for steerable models such thah they conform to the output type provided. One of those backends is of course `outlines-core`, but you also have access to two other libraries that fulfill the same purpose: `llguidance` and `xgrammar`.
+
+## Overview
+
+To select the backend to use for your generation, provide a value for the `backend` argument when calling a model or a generator.
+
+For instance:
+
+```python
+from typing import Literal
+import outlines
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+output_type = Literal["Paris", "London", "Rome", "Berlin"]
+
+model = outlines.from_transformers(
+    AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct"),
+    AutoTokenizer.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+)
+
+result = model("What is the capital of France?", output_type, backend="llguidance")
+print(result) # 'Paris'
+
+generator = outlines.Generaor(model, output_type)
+result = generator("What is the capital of France?", backend="xgrammar")
+print(result) # 'Paris'
+```
+
+If you do not provide a value for the `backend` argument, the default value will be used. The default value depends on the type of output type:
+
+- JSON schema: `outlines_core`
+- Regex: `outlines_core`
+- Context-free grammar: `llguidance`
+- Interegular FSM: `outlines_core`
+
+## Features matrix
+
+As mentioned previously, selecting the structured generation backend is only applicable to steerable models, so `Transformers`, `LlmaCpp` and `MLXLM`. Additionaly, some backends do not support some models within those or some output types.
+
+| | outlines_core | llguidance | xgrammar |
+|---|---|---|---|
+| **Models** | | | |
+| Transformers | ✅ | ✅ | ✅ |
+| LlamaCpp | ✅ | ✅ | ❌ |
+| MLXLM | ✅ | ✅ | ❌ |
+| **Output Types** | | | |
+| JSON Schema | ✅ | ✅ | ✅ |
+| Regex | ✅ | ✅ | ✅ |
+| Grammar | ✅ | ✅ | ✅ |
+| FSM | ✅ | ❌ | ❌ |
diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
index b138bf967..6f3c83173 100644
--- a/docs/features/models/llamacpp.md
+++ b/docs/features/models/llamacpp.md
@@ -109,7 +109,7 @@ for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
 
 ## Structured Generation
 
-The `LlamaCpp` model supports all output types available in Outlines except for context-free grammars. Simply provide an `output_type` after the prompt when calling the model.
+The `LlamaCpp` model supports all output types available in Outlines. Simply provide an `output_type` after the prompt when calling the model.
 
 ### Basic Type
 
@@ -195,6 +195,29 @@ result = model("Generate a fake social security number.", output_type)
 print(result) # '782-32-3789'
 ```
 
+### Context-free grammar
+
+```python
+from outlines.types import CFG
+import outlines
+from llama_cpp import Llama
+
+output_type = CFG("""
+root ::= answer
+answer ::= "yes" | "no"
+""")
+
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    )
+)
+
+result = model("Are you feeling good today?", output_type)
+print(result) # 'yes'
+```
+
 ## Inference Arguments
 
 When calling the model, you can provide optional inference parameters on top of the prompt and the output type. These parameters will be passed on to the `__call__` method of the `llama_cpp.Llama` model. Some common inference arguments include `max_tokens`, `temperature`, `frequency_penalty` and `top_p`.
diff --git a/mkdocs.yml b/mkdocs.yml
index 5225fdf68..84dd00fe0 100644
--- a/mkdocs.yml
+++ b/mkdocs.yml
@@ -162,6 +162,7 @@ nav:
 
           - Advanced:
                 - Logits Processors: features/advanced/logits_processors.md
+                - Structured Generation Backends: features/advanced/backends.md
 
     - API Reference: api_reference/
 

From c1a3ab8e223dd4957fba961d380f8d3f588e61bc Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 25 Jul 2025 13:22:12 +0200
Subject: [PATCH 414/505] Add the logits processor reset to handle multiple
 generations

---
 outlines/backends/llguidance.py              |  4 ++
 outlines/backends/xgrammar.py                | 60 +++++++++++++++-----
 outlines/generator.py                        |  6 ++
 outlines/processors/base_logits_processor.py |  9 +++
 tests/backends/test_backends.py              |  3 +-
 tests/backends/test_llguidance.py            | 10 ++++
 tests/backends/test_outlines_core.py         | 10 ++++
 tests/backends/test_xgrammar.py              | 19 +++++--
 8 files changed, 100 insertions(+), 21 deletions(-)

diff --git a/outlines/backends/llguidance.py b/outlines/backends/llguidance.py
index cd9a8d9b8..940602f98 100644
--- a/outlines/backends/llguidance.py
+++ b/outlines/backends/llguidance.py
@@ -48,6 +48,10 @@ def __init__(
         self.tensor_library_name = tensor_library_name
         super().__init__(tensor_library_name)
 
+    def reset(self):
+        """Ensure self._setup is called again for the next generation."""
+        self.is_first_token = True
+
     def _setup(self, batch_size: int) -> None:
         """Setup the LLMatchers, the bitmask and some functions used in the
         `process_logits` method.
diff --git a/outlines/backends/xgrammar.py b/outlines/backends/xgrammar.py
index 4e7a4071c..36be92bfa 100644
--- a/outlines/backends/xgrammar.py
+++ b/outlines/backends/xgrammar.py
@@ -5,11 +5,51 @@
 from outlines.backends.base import BaseBackend
 from outlines.models import SteerableModel
 from outlines.models.transformers import Transformers
+from outlines.processors.base_logits_processor import (
+    OutlinesLogitsProcessor,
+    TensorType
+)
 
 if TYPE_CHECKING:
     from xgrammar.contrib.hf import LogitsProcessor
 
 
+class XGrammarLogitsProcessor(OutlinesLogitsProcessor):
+    """Logits processor for XGrammar.
+
+    This class wraps the `xgr.contrib.hf.LogitsProcessor` class and adds
+    a `reset` method to reset the logits processor for a new generation.
+
+    """
+
+    def __init__(self, compiled_grammar: str):
+        """
+        Parameters
+        ----------
+        compiled_grammar: str
+            The compiled grammar to use to create the logits processor.
+
+        """
+        import xgrammar as xgr
+
+        self.xgr = xgr
+        self.compiled_grammar = compiled_grammar
+        self.xgrammar_logits_processor = None
+        super().__init__("torch")
+
+    def reset(self):
+        """Reset the logits processor for a new generation."""
+        self.xgrammar_logits_processor = None
+
+    def process_logits(self, input_ids: TensorType, logits: TensorType) -> TensorType:
+        """Bias the logits."""
+        if self.xgrammar_logits_processor is None:
+            self.xgrammar_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
+                self.compiled_grammar
+            )
+        return self.xgrammar_logits_processor(input_ids, logits) # type: ignore
+
+
 class XGrammarBackend(BaseBackend):
     """Backend for XGRammar."""
 
@@ -29,15 +69,14 @@ def __init__(self, model: SteerableModel):
                 "The xgrammar backend only supports Transformers models"
             )
 
-        self.xgr = xgr
         vocab_size = AutoConfig.from_pretrained(
             model.model.config._name_or_path
         ).vocab_size
-        tokenizer_info = self.xgr.TokenizerInfo.from_huggingface(
+        tokenizer_info = xgr.TokenizerInfo.from_huggingface(
             model.hf_tokenizer,
             vocab_size=vocab_size
         )
-        self.grammar_compiler = self.xgr.GrammarCompiler(tokenizer_info)
+        self.grammar_compiler = xgr.GrammarCompiler(tokenizer_info)
 
     def get_json_schema_logits_processor(
         self, json_schema: str
@@ -58,10 +97,7 @@ def get_json_schema_logits_processor(
         compiled_grammar = self.grammar_compiler.compile_json_schema(
             json_schema
         )
-        xgr_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
-            compiled_grammar
-        )
-        return xgr_logits_processor
+        return XGrammarLogitsProcessor(compiled_grammar)
 
     def get_regex_logits_processor(self, regex: str) -> "LogitsProcessor":
         """Create a logits processor from a regex.
@@ -78,10 +114,7 @@ def get_regex_logits_processor(self, regex: str) -> "LogitsProcessor":
 
         """
         compiled_grammar = self.grammar_compiler.compile_regex(regex)
-        xgr_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
-            compiled_grammar
-        )
-        return xgr_logits_processor
+        return XGrammarLogitsProcessor(compiled_grammar)
 
     def get_cfg_logits_processor(self, grammar: str) -> "LogitsProcessor":
         """Create a logits processor from a context-free grammar.
@@ -98,10 +131,7 @@ def get_cfg_logits_processor(self, grammar: str) -> "LogitsProcessor":
 
         """
         compiled_grammar = self.grammar_compiler.compile_grammar(grammar)
-        xgr_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
-            compiled_grammar
-        )
-        return xgr_logits_processor
+        return XGrammarLogitsProcessor(compiled_grammar)
 
     def get_fsm_logits_processor(self, fsm):
         raise NotImplementedError(
diff --git a/outlines/generator.py b/outlines/generator.py
index ab41562c2..a7340bd1b 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -309,6 +309,8 @@ def __call__(self, prompt: Any, **inference_kwargs) -> Any:
             The response generated by the model.
 
         """
+        if self.logits_processor is not None:
+            self.logits_processor.reset()
         return self.model.generate(
             prompt, self.logits_processor, **inference_kwargs
         )
@@ -329,6 +331,8 @@ def batch(self, prompts: List[Any], **inference_kwargs) -> List[Any]:
             The list of responses generated by the model.
 
         """
+        if self.logits_processor is not None:
+            self.logits_processor.reset()
         return self.model.generate_batch(
             prompts, self.logits_processor, **inference_kwargs
         )
@@ -349,6 +353,8 @@ def stream(self, prompt: Any, **inference_kwargs) -> Iterator[Any]:
             The response generated by the model.
 
         """
+        if self.logits_processor is not None:
+            self.logits_processor.reset()
         return self.model.generate_stream(
             prompt, self.logits_processor, **inference_kwargs
         )
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index 8d00f0c7b..9b4e4fe50 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -45,6 +45,15 @@ def __init__(self, tensor_library_name: str):
             )
         self.tensor_adapter = tensor_adapter_class()  # type: ignore
 
+    def reset(self):
+        """Reset the logits processor for a new generation
+
+        Only implement this method in subclasses if the logits processor
+        needs to be reset for a new generation.
+
+        """
+        pass
+
     @abstractmethod
     def process_logits(
         self, input_ids: TensorType, logits: TensorType
diff --git a/tests/backends/test_backends.py b/tests/backends/test_backends.py
index e4b00d637..56d8e56ee 100644
--- a/tests/backends/test_backends.py
+++ b/tests/backends/test_backends.py
@@ -2,7 +2,6 @@
 import outlines
 import pytest
 import transformers
-from xgrammar.contrib.hf import LogitsProcessor as XGrammarLogitsProcessor
 
 from outlines.backends import (
     _get_backend,
@@ -16,7 +15,7 @@
     LLGuidanceBackend,
     LLGuidanceLogitsProcessor
 )
-from outlines.backends.xgrammar import XGrammarBackend
+from outlines.backends.xgrammar import XGrammarBackend, XGrammarLogitsProcessor
 from outlines.processors.structured import (
     CFGLogitsProcessor,
     GuideLogitsProcessor,
diff --git a/tests/backends/test_llguidance.py b/tests/backends/test_llguidance.py
index 7e0926e12..a5de1ca5d 100644
--- a/tests/backends/test_llguidance.py
+++ b/tests/backends/test_llguidance.py
@@ -196,3 +196,13 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg_
         match="LLGuidanceBackend does not support FSM logits processors",
     ):
         backend.get_fsm_logits_processor(fsm)
+
+    # multiple generations
+    processor = backend.get_regex_logits_processor(regex)
+    generator = outlines.Generator(model, backend="llguidance", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
diff --git a/tests/backends/test_outlines_core.py b/tests/backends/test_outlines_core.py
index 4d40c3fbf..da2eda6fb 100644
--- a/tests/backends/test_outlines_core.py
+++ b/tests/backends/test_outlines_core.py
@@ -137,3 +137,13 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg,
     response = generator("Hello, how are you?")
     assert len(response) == 3
     assert int(response)
+
+    # multiple generations
+    processor = backend.get_regex_logits_processor(regex)
+    generator = outlines.Generator(model, backend="outlines_core", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
diff --git a/tests/backends/test_xgrammar.py b/tests/backends/test_xgrammar.py
index 42d80dbf8..89c9393be 100644
--- a/tests/backends/test_xgrammar.py
+++ b/tests/backends/test_xgrammar.py
@@ -4,7 +4,8 @@
 import llama_cpp
 import transformers
 from xgrammar import GrammarCompiler
-from xgrammar.contrib.hf import LogitsProcessor
+
+from outlines.backends.xgrammar import XGrammarLogitsProcessor
 
 import outlines
 from outlines.backends.xgrammar import XGrammarBackend
@@ -69,7 +70,7 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg, fsm):
 
     # json schema
     processor = backend.get_json_schema_logits_processor(json_schema)
-    assert isinstance(processor, LogitsProcessor)
+    assert isinstance(processor, XGrammarLogitsProcessor)
     generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
     response = generator("Hello, how are you?")
     assert response[0] == "{"
@@ -77,7 +78,7 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg, fsm):
 
     # regex
     processor = backend.get_regex_logits_processor(regex)
-    assert isinstance(processor, LogitsProcessor)
+    assert isinstance(processor, XGrammarLogitsProcessor)
     generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
     response = generator("Hello, how are you?")
     assert len(response) == 3
@@ -85,7 +86,7 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg, fsm):
 
     # cfg
     processor = backend.get_cfg_logits_processor(cfg)
-    assert isinstance(processor, LogitsProcessor)
+    assert isinstance(processor, XGrammarLogitsProcessor)
     generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
     response = generator("Hello, how are you?")
     assert response == "yes" or response == "no"
@@ -97,6 +98,16 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg, fsm):
     ):
         backend.get_fsm_logits_processor(fsm)
 
+    # multiple generations
+    processor = backend.get_regex_logits_processor(regex)
+    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+
 
 def test_xgrammar_backend_invalid_model(model_llamacpp):
     with pytest.raises(

From f062dc557a92adf1d83a7a5883c80fbbe8009c08 Mon Sep 17 00:00:00 2001
From: Oni-giri <onigiri@ik.me>
Date: Wed, 30 Jul 2025 18:03:08 +0300
Subject: [PATCH 415/505] import fix in base.py

---
 outlines/models/base.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/models/base.py b/outlines/models/base.py
index 4b948a56d..2ad0407f3 100644
--- a/outlines/models/base.py
+++ b/outlines/models/base.py
@@ -117,7 +117,7 @@ def __call__(
             The response generated by the model.
 
         """
-        from outlines import Generator
+        from outlines.generator import Generator
 
         return Generator(self, output_type, backend)(model_input, **inference_kwargs)
 

From 7bc20051492ef3b41062613dbb03d86c295912b7 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Tue, 29 Jul 2025 11:15:45 +0200
Subject: [PATCH 416/505] Delete CFGGuide and CFGLogitsProcessor

---
 benchmarks/bench_cfg_guide.py        |  64 ----
 docs/features/advanced/backends.md   |   2 +-
 examples/cfg.py                      |  97 ------
 outlines/backends/outlines_core.py   |  24 +-
 outlines/models/llamacpp.py          |  14 +-
 outlines/processors/__init__.py      |   1 -
 outlines/processors/guide.py         | 265 +---------------
 outlines/processors/structured.py    |  76 -----
 tests/backends/test_backends.py      |   8 +-
 tests/backends/test_outlines_core.py |  26 +-
 tests/processors/test_cfg_guide.py   | 456 ---------------------------
 tests/processors/test_guide.py       | 301 +-----------------
 tests/processors/test_structured.py  |  27 --
 tests/test_grammars.py               |  31 --
 14 files changed, 18 insertions(+), 1374 deletions(-)
 delete mode 100644 benchmarks/bench_cfg_guide.py
 delete mode 100644 examples/cfg.py
 delete mode 100644 tests/processors/test_cfg_guide.py
 delete mode 100644 tests/test_grammars.py

diff --git a/benchmarks/bench_cfg_guide.py b/benchmarks/bench_cfg_guide.py
deleted file mode 100644
index a4d288700..000000000
--- a/benchmarks/bench_cfg_guide.py
+++ /dev/null
@@ -1,64 +0,0 @@
-import random
-
-from transformers import AutoTokenizer
-
-import outlines.grammars
-from outlines.caching import cache_disabled
-from outlines.processors.guide import CFGGuide
-from outlines.models.transformers import TransformerTokenizer
-
-random.seed(42)
-
-
-def get_tiny_tokenizer():
-    """1000 tokens in vocabulary"""
-    return TransformerTokenizer(
-        AutoTokenizer.from_pretrained("hf-internal-testing/tiny-random-gpt2")
-    )
-
-
-benched_grammars = {
-    "json": outlines.grammars.json,
-    "arithmetic": outlines.grammars.arithmetic,
-}
-
-
-class CFGGuideBenchmark:
-    params = benched_grammars.keys()
-
-    def setup(self, grammar_name):
-        self.tokenizer = get_tiny_tokenizer()
-        self.prebuilt_cfg_guide = CFGGuide(
-            benched_grammars[grammar_name], self.tokenizer
-        )
-
-    @staticmethod
-    def _run_random_cfg(guide, rejection_sampling=True):
-        state = guide.initial_state
-        token_ids = list(guide.tokenizer.vocabulary.values())
-        for i in range(40):
-            # simulate ordering of logits top prob to lowest prob
-            random.shuffle(token_ids)
-            # simulate sampling and state update
-            if rejection_sampling:
-                next_token_id = next(guide.iter_valid_token_ids(state, token_ids))
-                state = guide.get_next_state(state, next_token_id)
-            else:
-                next_token_id = random.choice(guide.get_next_instruction(state).tokens)
-                state = guide.get_next_state(state, next_token_id)
-
-    @cache_disabled()
-    def time_cfg_guide_setup(self, grammar_name):
-        CFGGuide(benched_grammars[grammar_name], self.tokenizer)
-
-    @cache_disabled()
-    def time_cfg_guide_run_rejection_sampling(self, grammar):
-        self._run_random_cfg(self.prebuilt_cfg_guide, rejection_sampling=True)
-
-    @cache_disabled()
-    def time_cfg_guide_run(self, grammar):
-        self._run_random_cfg(self.prebuilt_cfg_guide, rejection_sampling=False)
-
-    @cache_disabled()
-    def peakmem_cfg_guide_run(self, grammar):
-        self._run_random_cfg(self.prebuilt_cfg_guide)
diff --git a/docs/features/advanced/backends.md b/docs/features/advanced/backends.md
index 68555f0bd..d7ff45886 100644
--- a/docs/features/advanced/backends.md
+++ b/docs/features/advanced/backends.md
@@ -52,5 +52,5 @@ As mentioned previously, selecting the structured generation backend is only app
 | **Output Types** | | | |
 | JSON Schema | ✅ | ✅ | ✅ |
 | Regex | ✅ | ✅ | ✅ |
-| Grammar | ✅ | ✅ | ✅ |
+| Grammar | ❌ | ✅ | ✅ |
 | FSM | ✅ | ❌ | ❌ |
diff --git a/examples/cfg.py b/examples/cfg.py
deleted file mode 100644
index bd3b58a2e..000000000
--- a/examples/cfg.py
+++ /dev/null
@@ -1,97 +0,0 @@
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-import outlines
-from outlines.types import CFG
-
-
-nlamb_grammar = r"""
-    start: sentence
-
-    sentence: noun verb noun        -> simple
-            | noun verb "like" noun -> comparative
-
-    noun: adj? NOUN
-    verb: VERB
-    adj: ADJ
-
-    NOUN: "flies" | "bananas" | "fruit"
-    VERB: "like" | "flies"
-    ADJ: "fruit"
-
-    %import common.WS
-    %ignore WS
-"""
-
-calc_grammar = r"""
-    ?start: sum
-          | NAME "=" sum    -> assign_var
-
-    ?sum: product
-        | sum "+" product   -> add
-        | sum "-" product   -> sub
-
-    ?product: atom
-        | product "*" atom  -> mul
-        | product "/" atom  -> div
-
-    ?atom: NUMBER           -> number
-         | "-" atom         -> neg
-         | NAME             -> var
-         | "(" sum ")"
-
-    %import common.LETTER -> NAME
-    %import common.INT -> NUMBER
-    %import common.WS_INLINE
-
-    %ignore WS_INLINE
-"""
-
-dyck_grammar = r"""
-    start: s
-    s: /a+/
-    | "(" s ")"
-    | "{" s "}"
-    | "[" s "]"
-"""
-
-json_grammar = r"""
-    ?start: value
-
-    ?value: object
-          | array
-          | string
-          | SIGNED_NUMBER      -> number
-          | "true"             -> true
-          | "false"            -> false
-          | "null"             -> null
-
-    array  : "[" [value ("," value)*] "]"
-    object : "{" [pair ("," pair)*] "}"
-    pair   : string ":" value
-
-    inner: /([^"]|\\\")+/ |
-    string : "\"" inner "\""
-
-    %import common.SIGNED_NUMBER
-    %import common.WS
-
-    %ignore WS
-"""
-
-model_name = "hf-internal-testing/tiny-random-gpt2"
-model = outlines.from_transformers(
-    AutoModelForCausalLM.from_pretrained(model_name),
-    AutoTokenizer.from_pretrained(model_name),
-)
-
-batch_size = 10
-for grammar in [nlamb_grammar, calc_grammar, dyck_grammar, json_grammar]:
-    generator = outlines.Generator(model, CFG(grammar))
-    sequences = generator([" "] * batch_size, max_tokens=model.model.config.n_positions)
-    for seq in sequences:
-        try:
-            parse = generator.logits_processor.guide.parser.parse(seq)
-            assert parse is not None
-            print("SUCCESS", seq)
-        except Exception:  # will also fail if goes over max_new_tokens / context window
-            print("FAILURE", seq)
diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index 84a8896fd..3fa5d9726 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -5,7 +5,6 @@
 from outlines.backends.base import BaseBackend
 from outlines.models import SteerableModel
 from outlines.processors import (
-    CFGLogitsProcessor,
     GuideLogitsProcessor,
     RegexLogitsProcessor,
 )
@@ -66,25 +65,10 @@ def get_regex_logits_processor(self, regex: str) -> RegexLogitsProcessor:
             self.tensor_library_name,
         )
 
-    def get_cfg_logits_processor(self, grammar: str) -> CFGLogitsProcessor:
-        """Create a logits processor from a context-free grammar.
-
-        Parameters
-        ----------
-        grammar: str
-            The context-free grammar to create a logits processor from. For
-            Outlines Core, the grammar must use the Lark format.
-
-        Returns
-        -------
-        LogitsProcessor
-            The logits processor to use to constrain the generation.
-
-        """
-        return CFGLogitsProcessor(
-            grammar,
-            self.tokenizer,
-            self.tensor_library_name,
+    def get_cfg_logits_processor(self, grammar):
+        raise NotImplementedError(
+            "Context-free grammar output type is not supported for the "
+            + "Outlines Core backend."
         )
 
     def get_fsm_logits_processor(self, fsm: FSM) -> GuideLogitsProcessor:
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 67754c9bb..9984c725d 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -16,7 +16,7 @@
 from outlines.inputs import Chat
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.tokenizer import Tokenizer
-from outlines.processors import CFGLogitsProcessor, OutlinesLogitsProcessor
+from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
     from llama_cpp import Llama, LogitsProcessorList
@@ -244,12 +244,6 @@ def generate(
             The text generated by the model.
 
         """
-        if isinstance(output_type, CFGLogitsProcessor):
-            raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp with the "
-                + "outlines-core backend"
-            )
-
         prompt = self.type_adapter.format_input(model_input)
 
         if isinstance(prompt, str):
@@ -304,12 +298,6 @@ def generate_stream(
             An iterator that yields the text generated by the model.
 
         """
-        if isinstance(output_type, CFGLogitsProcessor):
-            raise NotImplementedError(
-                "CFG generation is not supported for LlamaCpp with the "
-                + "outlines-core backend"
-            )
-
         prompt = self.type_adapter.format_input(model_input)
 
         if isinstance(prompt, str):
diff --git a/outlines/processors/__init__.py b/outlines/processors/__init__.py
index f07ce0cbb..b2c2208a1 100644
--- a/outlines/processors/__init__.py
+++ b/outlines/processors/__init__.py
@@ -1,7 +1,6 @@
 """Processors and guides to control generation in steerable models."""
 
 from .structured import (
-    CFGLogitsProcessor,
     GuideLogitsProcessor,
     JSONLogitsProcessor,
     OutlinesLogitsProcessor,
diff --git a/outlines/processors/guide.py b/outlines/processors/guide.py
index bebddd9cb..9c4a3067a 100644
--- a/outlines/processors/guide.py
+++ b/outlines/processors/guide.py
@@ -3,13 +3,8 @@
 Logits processors rely on guides to control the generation process.
 """
 
-import collections
-import copy
-import warnings
-from typing import TYPE_CHECKING, Any, Generator, ValuesView, Union
+from typing import TYPE_CHECKING, Any, Union
 
-from lark.indenter import DedentError
-from lark.lexer import UnexpectedCharacters, UnexpectedToken
 from outlines_core.fsm.guide import Generate
 from outlines_core.fsm.guide import Guide as CoreGuide
 from outlines_core.fsm.guide import RegexGuide as CoreRegexGuide
@@ -18,9 +13,7 @@
     create_states_mapping as uncached_create_states_mapping,
 )
 
-from outlines import grammars
 from outlines.caching import cache
-from outlines.fsm.parsing import PartialLark, PartialParserState
 
 if TYPE_CHECKING:
     from outlines.models.tokenizer import Tokenizer
@@ -164,259 +157,3 @@ def from_regex(
             _create_states_mapping=cached_create_states_mapping,
             **kwargs,
         )
-
-
-CFGState = collections.namedtuple("CFGState", ["parser_state", "prev_token"])
-
-
-class CFGGuide(Guide):
-    """Guide to generate text that is in the language of a context-free Lark
-    grammar.
-
-    """
-
-    def __init__(self, cfg_string: str, tokenizer: "Tokenizer"):
-        """
-        Parameters
-        ----------
-        cfg_string
-            The context-free grammar to generate text from.
-        tokenizer
-            The tokenizer to use to convert tokens to ids.
-
-        """
-        warnings.warn(
-            "Outlines' public *community-contributed* CFG structured generation "
-            "is experimental. Please review "
-            "https://dottxt-ai.github.io/outlines/latest/reference/generation/cfg#disclaimer"
-        )
-
-        self.cfg_string = cfg_string
-        self.tokenizer = tokenizer
-        self.eos_token_id = self.tokenizer.eos_token_id
-        self.parser = PartialLark(
-            cfg_string,
-            parser="lalr",
-            import_paths=[grammars.GRAMMAR_PATH],
-        )
-        self.initial_state = CFGState(
-            parser_state=self.parser.parse(""), prev_token=None
-        )
-
-    def get_next_instruction(self, state: CFGState) -> Instruction:
-        """Return the next instruction for guided generation.
-
-        Current lazy approach:
-        - For each token in the vocabulary
-          - create a copy of the parsers state
-          - add the tokens to the parsers input text
-          - if valid, add token to returned tokens
-
-        Further refinements are necessary for performant text processing.
-
-        Parameters
-        ----------
-        state
-            The guides current PartialParserState, or None if complete
-
-        Returns
-        -------
-        Instruction
-            A `Generate` instance that contains the model and the allowed token
-            ids.
-
-        """
-        import torch
-
-        if state.parser_state is None:
-            return Write(torch.tensor([self.eos_token_id]))
-
-        valid_tokens = list(
-            self.iter_valid_token_ids(
-                state, self.tokenizer.vocabulary.values()
-            )
-        )
-
-        if len(valid_tokens) == 1:
-            return Write(torch.tensor(valid_tokens))
-
-        return Generate(torch.tensor(valid_tokens))
-
-    def iter_valid_token_ids(
-        self, state: CFGState, candidate_token_ids: ValuesView[int]
-    ) -> Generator[int, None, None]:
-        """Iterate over the given token_ids and yield those that are valid for
-        the current parser state.
-
-        Parameters
-        ----------
-        parser_state
-            The current state of the parser, or None if complete.
-        token_ids
-            The list of token ids to check for validity.
-
-        Yields
-        ------
-        int
-            Valid token ids.
-
-        """
-        for token_id in candidate_token_ids:
-            if token_id == self.eos_token_id:
-                if self.can_terminate_state(state):
-                    yield token_id
-            else:
-                try:
-                    self._get_parser_state_token_applied(state, int(token_id))
-                    yield token_id
-                except (
-                    ValueError,
-                    EOFError,
-                    UnexpectedToken,
-                    UnexpectedCharacters,
-                    DedentError,
-                ):
-                    pass
-
-    def get_next_state(self, state: CFGState, token_id: int) -> CFGState:
-        """Update the state of the guide.
-
-        Decode the token_id, and calculate the new parser_state with the token
-        applied.
-
-        Parameters
-        ----------
-        state
-            The guides current PartialParserState, or None if complete
-        token_id
-            The id of the token that was just generated.
-
-        Returns
-        -------
-        CFGState
-            The guides new PartialParserState
-
-        """
-        if state.parser_state is None or token_id == self.eos_token_id:
-            parser_state = None
-        else:
-            parser_state = self._get_parser_state_token_applied(state, int(token_id))
-        return CFGState(parser_state=parser_state, prev_token=token_id)
-
-    def _get_parser_state_token_applied(
-        self, state: CFGState, token_id: int
-    ) -> PartialParserState:
-        """Apply the given token_id to the parser state.
-
-        Don't mutate `parser_state`, copy to protect
-
-        Get the token string
-          - if first token in generation: tokenizer.decode (no leading whitespace)
-          - else: normalized (with possibly leading whitespace)
-
-        Don't allow empty ("") tokens, raise ValueError
-
-        Parameters
-        ----------
-        state
-            The guide's current PartialParserState, or None if complete
-        token_id
-            The id of the token that was just generated.
-
-        Returns
-        -------
-        PartialParserState
-            The parser state with the token applied.
-
-        """
-        parser_state = copy.copy(state.parser_state)  # prevent side effects
-
-        # normalize
-        if state.prev_token is None:
-            new_token_str = self.tokenizer.decode([token_id])[0]
-        else:
-            prev_token_str = self.tokenizer.decode([[state.prev_token]])[0]
-            combined_token_str = self.tokenizer.decode([[state.prev_token, token_id]])[
-                0
-            ]
-            new_token_str = combined_token_str[len(prev_token_str) :]
-
-        if new_token_str == "":
-            raise ValueError("empty next token")
-
-        # update parser with new token
-        parser_state.lexer.state.text += new_token_str
-        self.parser.parse_from_state(parser_state, is_end=False)
-
-        return parser_state
-
-    def is_final_state(self, state: CFGState) -> bool:
-        """Return whether the given state is a final state.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether the given state is a final state.
-
-        """
-        # TODO: remove this method, use can_terminate_state and
-        # must_terminate_state here and in RegexGuide per
-        # https://github.com/dottxt-ai/outlines/issues/885
-        return self.can_terminate_state(state)
-
-    def can_terminate_state(self, state: CFGState) -> bool:
-        """Return whether generation is allowed to terminate.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether generation is allowed to terminate.
-
-        """
-        if state.parser_state is not None:
-            try:
-                copy.copy(state.parser_state).feed_eof()
-            except UnexpectedToken:
-                return False
-        return True
-
-    def must_terminate_state(self, state: CFGState) -> bool:
-        """Indicate whether generation must terminate as there are no legal
-        continuations.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether generation must terminate.
-
-        """
-        return (
-            state.parser_state is None or
-            set(state.parser_state.accepts()).issubset({"$END"})
-        )
-
-    def copy(self) -> "CFGGuide":
-        """Create a copy of the Guide.
-
-        Returns
-        -------
-        CFGGuide
-            A copy of the Guide.
-
-        """
-        return CFGGuide(self.cfg_string, self.tokenizer)
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
index f1a55b902..0059ec131 100644
--- a/outlines/processors/structured.py
+++ b/outlines/processors/structured.py
@@ -24,7 +24,6 @@
 limitations under the License.
 """
 
-import math
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
 
 from pydantic import BaseModel
@@ -35,7 +34,6 @@
     TensorType
 )
 from outlines.processors.guide import (
-    CFGGuide,
     Guide,
     RegexGuide
 )
@@ -209,77 +207,3 @@ def __init__(
             tokenizer=tokenizer,
             tensor_library_name=tensor_library_name
         )
-
-
-class CFGLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a context-free grammar."""
-    guide: CFGGuide
-
-    def __init__(
-        self, cfg_str: str, tokenizer: "Tokenizer", tensor_library_name: str
-    ):
-        """
-        Parameters
-        ----------
-        cfg_str
-            A string that represents a grammar.
-        tokenizer
-            The tokenizer used to convert tokens to ids.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
-
-        """
-        # Build a guide from the CFG string and then pass it to the
-        # GuideLogitsProcessor superclass.
-        cfg_guide = CFGGuide(cfg_string=cfg_str, tokenizer=tokenizer)
-        super().__init__(
-            tokenizer=tokenizer,
-            guide=cfg_guide,
-            tensor_library_name=tensor_library_name
-        )
-
-    def process_logits(
-        self, input_ids: TensorType, logits: TensorType
-    ) -> TensorType:
-        """Same behavior as GuideLogitsProcessor, but uses rejection
-        sampling.
-
-        Parameters
-        ----------
-        input_ids
-            The ids of the tokens of the existing sequences.
-        logits
-            The logits for the current generation step.
-
-        Returns
-        -------
-        TensorType
-            The biased logits.
-
-        """
-        if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0]) # type: ignore
-
-        sequence_states: List = []  # vector of states corresponding to `input_ids`
-
-        for seq_ids in input_ids: # type: ignore
-            gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
-
-            if curr_state_key not in self._guide_states: # pragma: no cover
-                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
-                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
-                self._guide_states[curr_state_key] = curr_state
-
-            sequence_states.append(self._guide_states[curr_state_key])
-
-        mask = self.tensor_adapter.full_like(logits, -math.inf)
-        for i, guide_state in enumerate(sequence_states):
-            first_legal_token = next(
-                self.guide.iter_valid_token_ids(
-                    guide_state, self.tensor_adapter.argsort_descending(logits[i]) # type: ignore
-                )
-            )
-            mask[i, [first_legal_token]] = logits[i, [first_legal_token]] # type: ignore
-
-        return mask
diff --git a/tests/backends/test_backends.py b/tests/backends/test_backends.py
index 56d8e56ee..68054922f 100644
--- a/tests/backends/test_backends.py
+++ b/tests/backends/test_backends.py
@@ -17,7 +17,6 @@
 )
 from outlines.backends.xgrammar import XGrammarBackend, XGrammarLogitsProcessor
 from outlines.processors.structured import (
-    CFGLogitsProcessor,
     GuideLogitsProcessor,
     RegexLogitsProcessor,
 )
@@ -114,8 +113,11 @@ def test_get_regex_logits_processor(model, regex):
 
 
 def test_get_cfg_logits_processor(model, cfg_lark, cfg_ebnf):
-    processor = get_cfg_logits_processor("outlines_core", model, cfg_lark)
-    assert isinstance(processor, CFGLogitsProcessor)
+    with pytest.raises(
+        NotImplementedError,
+        match="Context-free grammar output type is not supported"
+    ):
+        get_cfg_logits_processor("outlines_core", model, cfg_lark)
 
     processor = get_cfg_logits_processor("llguidance", model, cfg_lark)
     assert isinstance(processor, LLGuidanceLogitsProcessor)
diff --git a/tests/backends/test_outlines_core.py b/tests/backends/test_outlines_core.py
index da2eda6fb..087db561e 100644
--- a/tests/backends/test_outlines_core.py
+++ b/tests/backends/test_outlines_core.py
@@ -7,7 +7,6 @@
 import outlines
 from outlines.backends.outlines_core import OutlinesCoreBackend
 from outlines.processors.structured import (
-    CFGLogitsProcessor,
     RegexLogitsProcessor,
     GuideLogitsProcessor,
 )
@@ -109,26 +108,11 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg,
     assert int(response)
 
     # cfg
-    if isinstance(model, outlines.models.LlamaCpp):
-        with pytest.raises(
-            NotImplementedError,
-            match="CFG generation is not supported for LlamaCpp with the outlines-core backend",
-        ):
-            processor = backend.get_cfg_logits_processor(cfg)
-            generator = outlines.Generator(model, backend="outlines_core", processor=processor)
-            response = generator("Hello, how are you?")
-    else:
-        processor = backend.get_cfg_logits_processor(cfg)
-        assert isinstance(processor, CFGLogitsProcessor)
-        generator = outlines.Generator(model, backend="outlines_core", processor=processor)
-        response = generator("Hello, how are you?")
-        assert (
-            "+" in response
-            or "-" in response
-            or "*" in response
-            or "/" in response
-            or float(response.strip())
-        )
+    with pytest.raises(
+        NotImplementedError,
+        match="Context-free grammar output type is not supported",
+    ):
+        backend.get_cfg_logits_processor(cfg)
 
     # fsm
     processor = backend.get_fsm_logits_processor(fsm)
diff --git a/tests/processors/test_cfg_guide.py b/tests/processors/test_cfg_guide.py
deleted file mode 100644
index 6091fd4c2..000000000
--- a/tests/processors/test_cfg_guide.py
+++ /dev/null
@@ -1,456 +0,0 @@
-from collections import namedtuple
-from pathlib import Path
-
-import pytest
-from transformers import AutoTokenizer
-
-from outlines import grammars
-from outlines.models.transformers import TransformerTokenizer
-from outlines.processors.guide import CFGGuide
-
-
-@pytest.fixture
-def cleanup_lark_import():
-    import importlib
-
-    import lark.lark
-
-    yield
-    # Clean up lark.lark.LarkOptions._defaults
-    importlib.reload(lark.lark)
-
-
-TestInputs = namedtuple(
-    "TestInputs",
-    [
-        "grammar",  # the lark grammar to validate against
-        "vocabulary",  # the token strings which can be concatenated for a generation
-        "generated",  # the tokens which have been generated so far
-        "legal_next_tokens",  # the subset of the vocabulary which can legally be next in `generated`
-    ],
-)
-
-
-cfg_test_inputs = {
-    "Next Token Doesn't Complete Terminal": TestInputs(
-        grammar=r'?start: "a" "bc"',
-        vocabulary=["a", "ab", "b", "c"],
-        generated=["a"],
-        legal_next_tokens=["b"],
-    ),
-    "Ambiguous Terminal Completion": TestInputs(
-        grammar=r'?start: "ab" | "abc"',
-        vocabulary=["a", "ab", "abc", "abcd", "b", "c"],
-        generated=["a"],
-        legal_next_tokens=["b"],
-    ),
-    "Token is Substring of Another Token": TestInputs(
-        grammar=r'?start: "abc" | "abcd"',
-        vocabulary=["a", "b", "bc", "bcd", "bcde"],
-        generated=["a"],
-        legal_next_tokens=["b", "bc", "bcd"],
-    ),
-    "Multiple Valid Continuations": TestInputs(
-        grammar=r'?start: ("a" "b") | ("a" "c")',
-        vocabulary=["a", "b", "bc", "c"],
-        generated=["a"],
-        legal_next_tokens=["b", "c"],
-    ),
-    "Prefix Matches Multiple Terminals": TestInputs(
-        grammar=r'?start: "abcd" | "abef"',
-        vocabulary=["a", "b", "be", "bcd", "bef", "bed"],
-        generated=["a"],
-        legal_next_tokens=["b", "be", "bcd", "bef"],
-    ),
-    "Token Matches Multiple Paths in Grammar": TestInputs(
-        grammar=r'?start: ("a" "b" "c") | ("a" "b" "d")',
-        vocabulary=["a", "b", "c", "d"],
-        generated=["a", "b"],
-        legal_next_tokens=["c", "d"],
-    ),
-    "Incomplete Terminal at End of Prefix": TestInputs(
-        grammar=r'?start: "abc"',
-        vocabulary=["a", "ab", "c", "abc", "abcd"],
-        generated=["ab"],
-        legal_next_tokens=["c"],
-    ),
-    "Complex Grammar Rules": TestInputs(
-        grammar=r'?start: "a" "b" ["c"]',
-        vocabulary=["a", "b", "c"],
-        generated=["a", "b"],
-        legal_next_tokens=["c", None],  # Allowing the document to end after "a" "b"
-    ),
-    "Empty Prefix String": TestInputs(
-        grammar=r'?start: "a" | "b"',
-        vocabulary=["a", "b", "c", "d"],
-        generated=[],
-        legal_next_tokens=["a", "b"],
-    ),
-    "Ambiguous Pattern Completion": TestInputs(
-        grammar=r'?start: /a+/ "b" /c?d/',
-        vocabulary=["a", "aa", "b", "cd", "d"],
-        generated=["a", "a", "b"],
-        legal_next_tokens=["cd", "d"],
-    ),
-    "Optional Patterns with Overlapping Tokens": TestInputs(
-        grammar=r'?start: "a" "b"? "c"',
-        vocabulary=["a", "b", "bc", "c"],
-        generated=["a"],
-        legal_next_tokens=["b", "bc", "c"],
-    ),
-    "Greedy vs. Non-Greedy Matching": TestInputs(
-        grammar=r'?start: /a+?/ "b" /c/',
-        vocabulary=["a", "aa", "aaa", "b", "c"],
-        generated=["a", "a", "b"],
-        legal_next_tokens=["c"],
-    ),
-    "Nested Optional Elements": TestInputs(
-        grammar=r'?start: "a" ["b" ["c"]]',
-        vocabulary=["a", "b", "bc", "c"],
-        generated=["a"],
-        legal_next_tokens=[
-            "b",
-            "bc",
-            None,
-        ],  # Allowing the document to end after "a" "b"
-    ),
-    "Recursive Patterns": TestInputs(
-        grammar=r'?start: /a(bc)*/ "d"',
-        vocabulary=["a", "bc", "bcbcbc", "d"],
-        generated=["a", "bc", "d"],
-        legal_next_tokens=[None],  # Allowing the document to end after "a" "bc" "d"
-    ),
-    "Overlapping Character Classes": TestInputs(
-        grammar=r'?start: /[ab]+/ "d"',
-        vocabulary=["a", "b", "c", "aa", "bb", "cc", "d"],
-        generated=["a", "b"],
-        legal_next_tokens=["d", "a", "b", "aa", "bb"],
-    ),
-    "Conditional Patterns": TestInputs(
-        grammar=r'?start: "a" /b/ "c" (/d/)?',
-        vocabulary=["a", "b", "c", "d"],
-        generated=["a", "b", "c"],
-        legal_next_tokens=["d", None],  # Allowing the document to end after "a" "b" "c"
-    ),
-    "Unicode and Special Characters": TestInputs(
-        grammar=r'?start: /[a-zA-Z]/ "é" /[0-9]+/',
-        vocabulary=["a", "b", "é", "1", "2", "12"],
-        generated=["a", "é"],
-        legal_next_tokens=["1", "2", "12"],
-    ),
-    "Unicode and Special Characters Are Choices": TestInputs(
-        grammar=r'?start: /[a-zA-Z]/ "é" /[0-9]+/',
-        vocabulary=["a", "b", "é", "é9", "2", "12"],
-        generated=["a"],
-        legal_next_tokens=["é", "é9"],
-    ),
-    "Whitespace and Ignored Characters": TestInputs(
-        grammar=r'?start: "a" / *\s*b/ "c"',
-        vocabulary=["a", " b", " c", "c"],
-        generated=["a", " b"],
-        legal_next_tokens=["c"],
-    ),
-    "Token Overlaps Multiple Terminals": TestInputs(
-        grammar=r'?start: "a" "b" "c" "ab"',
-        vocabulary=["a", "b", "bc", "cab", "abc"],
-        generated=["a"],
-        legal_next_tokens=["b", "bc"],
-    ),
-    "Interleaved Sequences": TestInputs(
-        grammar=r'?start: ("a" "b") | ("a" "c")',
-        vocabulary=["a", "b", "c", "ab", "ac"],
-        generated=["a"],
-        legal_next_tokens=["b", "c"],
-    ),
-    "Repeated and Nested Patterns": TestInputs(
-        grammar=r'?start: "a" ("b" "c")* "d"',
-        vocabulary=["a", "b", "c", "bc", "bcc", "cbc", "bcbc", "cbccccd", "d", "bcbcd"],
-        generated=["a", "b", "c"],
-        legal_next_tokens=["b", "bc", "bcbc", "d", "bcbcd"],
-    ),
-    "Ambiguous Ending Patterns": TestInputs(
-        grammar=r'?start: "a" (/b/)? (/c/)*',
-        vocabulary=["a", "b", "c"],
-        generated=["a", "b"],
-        legal_next_tokens=["c", None],  # Allowing the document to end after "a" "b"
-    ),
-    "Whitespace Handling in Patterns": TestInputs(
-        grammar=r'?start: "a" / *\s*b/ /c /',
-        vocabulary=["a", " b", "c"],
-        generated=["a", " b"],
-        legal_next_tokens=["c"],
-    ),
-    "Token with Escape Characters": TestInputs(
-        grammar=r'?start: "a\n" ("\t")? "b"',
-        vocabulary=["a\nb", "a", "b", "\n", "\tb", "\t"],
-        generated=["a", "\n"],
-        legal_next_tokens=["b", "\t", "\tb"],
-    ),
-    "Complex Nesting": TestInputs(
-        grammar=r'?start: "a" ("b" ("c" "d"))',
-        vocabulary=["a", "b", "c", "d"],
-        generated=["a", "b", "c"],
-        legal_next_tokens=["d"],
-    ),
-    "Repeated Optional Patterns": TestInputs(
-        grammar=r'?start: ("a" ["b"])*',
-        vocabulary=["a", "b"],
-        generated=["a", "b", "a"],
-        legal_next_tokens=["a", "b", None],
-    ),
-    "Multiple Non-Terminal Symbols": TestInputs(
-        grammar=r"""
-        ?start: A B
-        A: "a"
-        B: "b"
-        """,
-        vocabulary=["a", "b"],
-        generated=["a"],
-        legal_next_tokens=["b"],
-    ),
-    "Recursive Definitions": TestInputs(
-        grammar=r"""
-        ?start: term_a
-        term_a: "a" term_a | "b"
-        """,
-        vocabulary=["a", "b"],
-        generated=["a", "a"],
-        legal_next_tokens=["a", "b"],
-    ),
-    "Ignored Patterns": TestInputs(
-        grammar=r"""
-        ?start: "a" "b" "c"
-        %ignore /\s+/
-        """,
-        vocabulary=["a", "b", "c", " "],
-        generated=["a", " ", "b"],
-        legal_next_tokens=["c", " "],
-    ),
-    "Cross-References": TestInputs(
-        grammar=r"""
-        ?start: term_a
-        term_a : /a/ term_b
-        term_b : /b/ term_a | /c/
-        """,
-        vocabulary=["a", "b", "c", "bac"],
-        generated=["a"],
-        legal_next_tokens=["b", "bac", "c"],
-    ),
-    "Multiple Complex Non-Terminal Rules": TestInputs(
-        grammar=r"""
-            ?start: S1 S2 S3
-            S1: "a" | "b"
-            S2: "c" | "d"
-            S3: "e" "f" | "g"
-        """,
-        vocabulary=["a", "b", "c", "d", "e", "f", "g"],
-        generated=["a", "c"],
-        legal_next_tokens=["e", "g"],
-    ),
-    #
-    #
-    # TODO: fix
-    #       parser incorrectly requires consumption of a or b in this case for unknown reasons
-    # "Nested Patterns with Repetition": TestInputs(
-    #     grammar=r'?start: "a" ("b" | "c")* "d"',
-    #     vocabulary=["a", "b", "bc", "bcc", "cbc", "bcbc", "cbccccd", "c", "d" "bcdcb"],
-    #     generated=["a"],
-    #     legal_next_tokens=["b", "bc", "bcc", "cbc", "bcbc", "cbccccd", "c", "d"],
-    # ),
-    #
-    #
-    # TODO: fix
-    #       adjacent terminals with ambiguous starts and ends not handled properly
-    #       ensure parser isn't greedy incorrectly
-    # "Ambiguous Overlapping Patterns": TestInputs(
-    #     grammar=r"?start: /ab*/ /bc?/",
-    #     vocabulary=["a", "ab", "abc", "b", "bc", "c"],
-    #     generated=["a", "b", "b"],
-    #     legal_next_tokens=["b", "c", "bc"],
-    # ),
-    # "Ambiguous Overlapping Patterns In Generation": TestInputs(
-    #     grammar=r"?start: /ab*/ /bc?/",
-    #     vocabulary=["a", "ab", "abc", "b", "bc", "c", "abbbc"],
-    #     generated=["a", "b", "b"],
-    #     legal_next_tokens=["b", "c", "bc"],
-    # ),
-    #
-    #
-    # SKIP:
-    # Awaiting negative lookarounds in interegular
-    # "Lookahead and Lookbehind with Nested Conditions": TestInputs(
-    #    grammar=r'?start: /(?<=a)b(?=c)/ "d"',
-    #    vocabulary=["a", "b", "c", "d"],
-    #    generated=["a", "b", "c"],
-    #    legal_next_tokens=["d"]
-    # ),
-    # "Lookbehind Patterns": TestInputs(
-    #    grammar=r'?start: /(?<=a)b/ "c"',
-    #    vocabulary=["a", "b", "c"],
-    #    generated=["a", "b"],
-    #    legal_next_tokens=["c"]
-    # ),
-}
-
-
-@pytest.mark.parametrize("name", cfg_test_inputs.keys())
-def test_cfg_next_token(name, cleanup_lark_import):
-    inputs = cfg_test_inputs[name]
-
-    class MockTokenizer:
-        vocabulary = {token: i + 1 for i, token in enumerate(inputs.vocabulary)}
-        vocabulary["<eos>"] = 0
-        reverse_vocab = {i: tok for tok, i in vocabulary.items()}
-        special_tokens = {"<eos>"}
-        eos_token_id = 0
-
-        def convert_token_to_string(self, token):
-            return token
-
-        def decode(self, token_ids):
-            if isinstance(token_ids[0], list):
-                return [
-                    "".join(map(self.reverse_vocab.get, token_ids_sublist))
-                    for token_ids_sublist in token_ids
-                ]
-            return [self.reverse_vocab[token_id] for token_id in token_ids]
-
-    # create a guide and the appropriate state advanced
-    # per the inputs generated tokens
-    tokenizer = MockTokenizer()
-    guide = CFGGuide(inputs.grammar, tokenizer)
-    state = guide.initial_state
-    for token in inputs.generated:
-        state = guide.get_next_state(state, tokenizer.vocabulary[token])
-    instruction = guide.get_next_instruction(state)
-
-    # normalize expectations and returned tokens for simple comparison
-    returned_next_tokens = sorted(
-        {tokenizer.reverse_vocab[int(t)] for t in instruction.tokens}
-    )
-    expected_next_tokens = sorted(
-        {
-            t
-            if t is not None
-            else tokenizer.reverse_vocab[tokenizer.eos_token_id]  # None -> "<eos>"
-            for t in inputs.legal_next_tokens
-        }
-    )
-
-    assert returned_next_tokens == expected_next_tokens
-
-
-@pytest.fixture(scope="session")
-def tokenizer_sentencepiece_gpt2():
-    return TransformerTokenizer(AutoTokenizer.from_pretrained("gpt2"))
-
-
-@pytest.fixture(scope="session")
-def tokenizer_sentencepiece_llama1():
-    return TransformerTokenizer(
-        AutoTokenizer.from_pretrained(
-            "trl-internal-testing/tiny-random-LlamaForCausalLM"
-        )
-    )
-
-
-@pytest.fixture(scope="session")
-def tokenizer_tiktoken_llama3():
-    return TransformerTokenizer(
-        AutoTokenizer.from_pretrained("yujiepan/llama-3-tiny-random")
-    )
-
-
-@pytest.fixture(scope="session")
-def tokenizer_character_level_byt5():
-    return TransformerTokenizer(AutoTokenizer.from_pretrained("google/byt5-small"))
-
-
-# Collects all samples within cfg_samples/ and makes adding
-# a test case as easy as adding a valid sample to cfg_samples/
-all_samples = {}
-examples_path = Path(__file__).parent.parent / "cfg_samples"
-for sample_collection_path in examples_path.iterdir():
-    grammar_name = sample_collection_path.name
-    grammar = getattr(grammars, grammar_name)
-    for sample_path in sample_collection_path.iterdir():
-        test_name = f"{grammar_name}_{sample_path.name}"
-        with open(sample_path) as f:
-            all_samples[test_name] = (grammar_name, grammar, f.read().rstrip("\n"))
-
-
-@pytest.mark.parametrize("sample_name", all_samples.keys())
-def test_cfg_test_sample_valid_with_lark(sample_name):
-    """assert the provided sample is valid (testing the test itself)"""
-    from lark import Lark, UnexpectedToken
-
-    grammar_name, grammar_str, sample = all_samples[sample_name]
-    try:
-        parser = Lark(grammar_str, parser="lalr", import_paths=[grammars.GRAMMAR_PATH])
-        parser = parser.parse_interactive(sample)
-        token = parser.exhaust_lexer()[-1]
-        parser.feed_eof(token)
-    except UnexpectedToken as e:
-        raise Exception(
-            f"Invalid test, sample '{sample_name}' isn't a legal generation of '{grammar_name}':\n{e}"
-        )
-
-
-@pytest.mark.parametrize("sample_name", all_samples.keys())
-@pytest.mark.parametrize(
-    "tokenizer_name",
-    [
-        "tokenizer_sentencepiece_gpt2",
-        "tokenizer_sentencepiece_llama1",
-        "tokenizer_tiktoken_llama3",
-        "tokenizer_character_level_byt5",
-    ],
-)
-def test_cfg_grammar_sample(request, sample_name, tokenizer_name, cleanup_lark_import):
-    """Test whether CFG can generate the exact token sequence as tokenizer.encode(sample) produces"""
-
-    # TODO: enable these tests once improvements are made
-    if (
-        tokenizer_name != "tokenizer_character_level_byt5"
-        or sample_name == "json_outlines.generate.samplers.mypy.json.test"
-    ):
-        pytest.skip("CFG is too slow, skipping tests for this tokenizer")
-    elif sample_name == "arithmetic_lots_of_ops.arithmetic.test":
-        pytest.skip("CFG incorrectly handles this valid sample, skipping until bugfix")
-
-    tokenizer = request.getfixturevalue(tokenizer_name)
-
-    grammar_name, grammar_str, sample = all_samples[sample_name]
-    cfg_guide = CFGGuide(grammar_str, tokenizer)
-
-    sample_token_ids = tokenizer.tokenizer.encode(
-        sample, add_special_tokens=False, return_tensors="pt"
-    )[0]
-    assert (
-        len(sample_token_ids.shape) == 1
-    )  # ensure we're encoding in the desired shape for this test
-
-    state = cfg_guide.initial_state
-    for i, token_id in enumerate(sample_token_ids):
-        if tokenizer.decode([token_id])[0] == "":
-            continue
-        next_instruction = cfg_guide.get_next_instruction(state)
-        if token_id not in next_instruction.tokens:
-            processed_str = tokenizer.decode([sample_token_ids[:i]])[0]
-            remaining_str = tokenizer.decode([sample_token_ids[i:]])[0]
-            if next_instruction.tokens == [tokenizer.eos_token_id]:
-                error_label = "CFGGuide required EOS early"
-            else:
-                expected = tokenizer.decode(next_instruction.tokens)
-                error_label = (
-                    f"Mismatched expectations, Guide expected {sorted(expected)}"
-                )
-            raise Exception(
-                f"{error_label}\n"
-                f"processed:\n```{processed_str}```\n"
-                f"remaining:\n```{remaining_str}```"
-            )
-            next_instruction.tokens
-        state = cfg_guide.get_next_state(state, token_id)
-    final_instruction = cfg_guide.get_next_instruction(state)
-    assert tokenizer.eos_token_id in final_instruction.tokens
diff --git a/tests/processors/test_guide.py b/tests/processors/test_guide.py
index 0238e37d3..ee7f568bb 100644
--- a/tests/processors/test_guide.py
+++ b/tests/processors/test_guide.py
@@ -4,7 +4,7 @@
 import transformers
 
 import outlines
-from outlines.processors.guide import CFGGuide, Generate, RegexGuide, StopAtEOSGuide, Write
+from outlines.processors.guide import Generate, RegexGuide, StopAtEOSGuide, Write
 from outlines import caching
 
 try:
@@ -238,302 +238,3 @@ def test_regex_guide_caching():
         expected_misses = i + 1
         expected_hits = i + 1
         assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
-
-
-def test_cfg():
-    class MockTokenizer:
-        vocabulary = {"{": 1, "}": 2, "[": 3, "]": 4, "eos": 5}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 5
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            if isinstance(token_ids[0], list):
-                return [
-                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
-                    for token_ids_sublist in token_ids
-                ]
-            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
-
-    cfg_str = """
-        start: expr
-        expr: "{" expr "}" | "[" expr "]" |
-    """
-    tokenizer = MockTokenizer()
-
-    guide = CFGGuide(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(
-        guide.get_next_instruction(guide.initial_state).tokens, [1, 3, 5]
-    )
-    state = guide.get_next_state(guide.initial_state, token_id=1)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
-    state = guide.get_next_state(state, token_id=3)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 3, 4])
-    state = guide.get_next_state(state, token_id=4)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [2])
-    state = guide.get_next_state(state, token_id=2)
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [5])
-    state = guide.get_next_state(state, token_id=5)
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert isinstance(guide.copy(), CFGGuide)
-
-
-def test_cfg_early_termination():
-    class MockTokenizer:
-        vocabulary = {"(": 1, ")": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            if isinstance(token_ids[0], list):
-                return [
-                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
-                    for token_ids_sublist in token_ids
-                ]
-            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
-
-    cfg_str = """
-        start: expr+
-        expr: "(" subexpr ")"
-        subexpr: expr |
-    """
-    tokenizer = MockTokenizer()
-
-    guide = CFGGuide(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(
-        guide.get_next_instruction(guide.initial_state).tokens, [1]
-    )
-    state = guide.get_next_state(guide.initial_state, token_id=1)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2])
-    state = guide.get_next_state(state, token_id=2)
-    assert not guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    # possible to continue or terminate
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 3])
-    state = guide.get_next_state(state, token_id=3)  # feed eos
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    # once eos generated, can only terminate
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [3])
-
-
-def test_cfg_ignore_directive():
-    class MockTokenizer:
-        vocabulary = {"a": 1, " ": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            if isinstance(token_ids[0], list):
-                return [
-                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
-                    for token_ids_sublist in token_ids
-                ]
-            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
-
-    cfg_str = """
-        start: LETTER+
-        LETTER: "a"
-        WS: " "
-        %ignore WS
-    """
-    tokenizer = MockTokenizer()
-
-    guide = CFGGuide(cfg_str, tokenizer)
-
-    state = guide.initial_state
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2])
-    state = guide.get_next_state(state, token_id=2)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2])
-    state = guide.get_next_state(state, token_id=1)
-    assert not guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
-    state = guide.get_next_state(state, token_id=2)
-    assert not guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
-    state = guide.get_next_state(state, token_id=2)
-    assert not guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
-    state = guide.get_next_state(state, token_id=1)
-    assert not guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 2, 3])
-    state = guide.get_next_state(state, token_id=3)
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    # once eos generated, can only terminate
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [3])
-
-
-def test_cfg_multitoken_terminal():
-    class MockTokenizer:
-        vocabulary = {"a": 1, "b": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            if isinstance(token_ids[0], list):
-                return [
-                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
-                    for token_ids_sublist in token_ids
-                ]
-            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
-
-    cfg_str = """
-        start: S
-        S: "aa" | "bb"
-    """
-    tokenizer = MockTokenizer()
-
-    guide = CFGGuide(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(
-        guide.get_next_instruction(guide.initial_state).tokens, [1, 2]
-    )
-    state = guide.get_next_state(guide.initial_state, token_id=1)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1])
-    state = guide.get_next_state(state, token_id=1)
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [3])
-    state = guide.get_next_state(state, token_id=3)
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-
-def test_cfg_allow_both_extend_and_shift_terminal():
-    class MockTokenizer:
-        vocabulary = {"(": 1, ")": 2, "a": 3, "eos": 4}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 4
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            if isinstance(token_ids[0], list):
-                return [
-                    "".join(map(self.inverse_vocabulary.get, token_ids_sublist))
-                    for token_ids_sublist in token_ids
-                ]
-            return [self.inverse_vocabulary[token_id] for token_id in token_ids]
-
-    cfg_str = """
-        start: s
-        s: "(" s ")" | /a+/
-    """
-    tokenizer = MockTokenizer()
-
-    guide = CFGGuide(cfg_str, tokenizer)
-
-    assert_expected_tensor_ids(
-        guide.get_next_instruction(guide.initial_state).tokens, [1, 3]
-    )
-    state = guide.get_next_state(guide.initial_state, token_id=1)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [1, 3])
-    state = guide.get_next_state(state, token_id=3)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [2, 3])
-    state = guide.get_next_state(state, token_id=3)
-    assert not guide.must_terminate_state(state)
-    assert not guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [2, 3])
-    state = guide.get_next_state(state, token_id=2)
-
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-
-    assert_expected_tensor_ids(guide.get_next_instruction(state).tokens, [4])
-    state = guide.get_next_state(state, token_id=4)
-    assert guide.must_terminate_state(state)
-    assert guide.can_terminate_state(state)
-    assert guide.is_final_state(state)
-
-    cfg_str = """
-        start: CHAR
-        CHAR: /./
-    """
-
-    guide = CFGGuide(cfg_str, tokenizer)
-
-    state = guide.get_next_state(guide.initial_state, token_id=1)
-    assert guide.get_next_state(state, token_id=4).parser_state is None
diff --git a/tests/processors/test_structured.py b/tests/processors/test_structured.py
index 54afae2b2..261328922 100644
--- a/tests/processors/test_structured.py
+++ b/tests/processors/test_structured.py
@@ -10,7 +10,6 @@
     GuideLogitsProcessor,
     RegexLogitsProcessor,
     JSONLogitsProcessor,
-    CFGLogitsProcessor,
 )
 
 
@@ -107,29 +106,3 @@ def test_structured_init_regex_logits_processor(tokenizer):
     assert processor.tokenizer is tokenizer
     assert processor.guide is not None
     assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_init_cfg_logits_processor(tokenizer):
-    processor = CFGLogitsProcessor(
-        cfg_str=arithmetic_grammar,
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, CFGLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is not None
-    assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_cfg_logits_processor_call(tokenizer):
-    processor = CFGLogitsProcessor(
-        cfg_str=arithmetic_grammar,
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    vocab_token_ids = list(tokenizer.vocabulary.values())
-    logits = torch.randn(1, len(vocab_token_ids))
-    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
-    output = processor(input_ids, logits)
-    assert output.shape == (1, len(vocab_token_ids))
-    processor(input_ids, logits)
diff --git a/tests/test_grammars.py b/tests/test_grammars.py
deleted file mode 100644
index 69486684b..000000000
--- a/tests/test_grammars.py
+++ /dev/null
@@ -1,31 +0,0 @@
-import pytest
-
-import outlines.grammars as grammars
-from outlines.processors.guide import CFGGuide
-
-
-@pytest.mark.parametrize("grammar", [grammars.json, grammars.arithmetic])
-def test_grammar_module(grammar):
-    class MockTokenizer:
-        vocabulary = {"(": 1, ")": 2, "a": 3, "eos": 4}
-        special_tokens = {"eos"}
-        eos_token = "eos"
-        eos_token_id = 4
-
-        def convert_token_to_string(self, token):
-            return token
-
-        @property
-        def inverse_vocabulary(self):
-            return {v: k for k, v in self.vocabulary.items()}
-
-        def decode(self, token_ids):
-            return [self.inverse_vocabulary[t] for t in token_ids]
-
-    cfg_str = """
-        start: s
-        s: "(" s ")" | /a+/
-    """
-    tokenizer = MockTokenizer()
-    fsm = CFGGuide(cfg_str, tokenizer)
-    assert isinstance(fsm, CFGGuide)

From 4e66c52c60a309ee58cdc2761e56a85101c28980 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Tue, 29 Jul 2025 11:16:25 +0200
Subject: [PATCH 417/505] Delete benchmarks

---
 benchmarks/__init__.py          |   0
 benchmarks/asv.conf.json        |  20 -----
 benchmarks/bench_json_schema.py |  77 -------------------
 benchmarks/bench_processors.py  | 131 --------------------------------
 benchmarks/bench_regex_guide.py |  40 ----------
 benchmarks/common.py            |   8 --
 6 files changed, 276 deletions(-)
 delete mode 100644 benchmarks/__init__.py
 delete mode 100644 benchmarks/asv.conf.json
 delete mode 100644 benchmarks/bench_json_schema.py
 delete mode 100644 benchmarks/bench_processors.py
 delete mode 100644 benchmarks/bench_regex_guide.py
 delete mode 100644 benchmarks/common.py

diff --git a/benchmarks/__init__.py b/benchmarks/__init__.py
deleted file mode 100644
index e69de29bb..000000000
diff --git a/benchmarks/asv.conf.json b/benchmarks/asv.conf.json
deleted file mode 100644
index 92808ce83..000000000
--- a/benchmarks/asv.conf.json
+++ /dev/null
@@ -1,20 +0,0 @@
-{
-    "version": 1,
-    "project": "Outlines",
-    "project_url": "https://dottxt-ai.github.io/outlines/",
-    "repo": "..",
-    "branches": [
-	"HEAD"
-    ],
-    "build_command": [
-        "python -mpip install .[test]",
-        "PIP_NO_BUILD_ISOLATION=false python -mpip wheel --no-deps --no-index -w {build_cache_dir} {build_dir}",
-    ],
-    "environment_type": "virtualenv",
-    "show_commit_url": "https://github.com/dottxt-ai/outlines/commit/",
-    "benchmark_dir": ".",
-    "env_dir": "env",
-    "results_dir": "results",
-    "html_dir": "html",
-    "build_cache_size": 8
-}
diff --git a/benchmarks/bench_json_schema.py b/benchmarks/bench_json_schema.py
deleted file mode 100644
index 6ed223ef4..000000000
--- a/benchmarks/bench_json_schema.py
+++ /dev/null
@@ -1,77 +0,0 @@
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-from outlines.caching import cache_disabled
-from outlines.processors.guide import RegexGuide
-
-from .common import setup_tokenizer  # noqa: E402
-
-simple_schema = """{
-        "$defs": {
-            "Armor": {
-                "enum": ["leather", "chainmail", "plate"],
-                "title": "Armor",
-                "type": "string"
-            }
-        },
-        "properties": {
-            "name": {"maxLength": 10, "title": "Name", "type": "string"},
-            "age": {"title": "Age", "type": "integer"},
-            "armor": {"$ref": "#/$defs/Armor"},
-            "strength": {"title": "Strength", "type": "integer"}\
-        },
-        "required": ["name", "age", "armor", "strength"],
-        "title": "Character",
-        "type": "object"
-    }"""
-
-
-complex_schema = """{
-  "$schema": "http://json-schema.org/draft-04/schema#",
-  "title": "Schema for a recording",
-  "type": "object",
-  "definitions": {
-    "artist": {
-      "type": "object",
-      "properties": {
-        "id": {"type": "number"},
-        "name": {"type": "string"},
-        "functions": {
-          "type": "array",
-          "items": {"type": "string"}
-        }
-      },
-      "required": ["id", "name", "functions"]
-    }
-  },
-  "properties": {
-    "id": {"type": "number"},
-    "work": {
-      "type": "object",
-      "properties": {
-        "id": {"type": "number"},
-        "name": {"type": "string"},
-        "composer": {"$ref": "#/definitions/artist"}
-      }
-    },
-    "recording_artists": {
-      "type": "array",
-      "items": {"$ref": "#/definitions/artist"}
-    }
-  },
-  "required": ["id", "work", "recording_artists"]
-}"""
-
-schemas = dict(simple_schema=simple_schema, complex_schema=complex_schema)
-
-
-class JsonSchemaBenchmark:
-    params = schemas.keys()
-
-    def setup(self, schema_name):
-        self.tokenizer = setup_tokenizer()
-        self.schema = schemas[schema_name]
-
-    @cache_disabled()
-    def time_json_schema_to_fsm(self, schema_name):
-        regex = build_regex_from_schema(self.schema)
-        RegexGuide.from_regex(regex, self.tokenizer)
diff --git a/benchmarks/bench_processors.py b/benchmarks/bench_processors.py
deleted file mode 100644
index 02ea52b79..000000000
--- a/benchmarks/bench_processors.py
+++ /dev/null
@@ -1,131 +0,0 @@
-import numpy as np
-import torch
-
-import outlines.models as models
-from outlines.processors import OutlinesLogitsProcessor, RegexLogitsProcessor
-
-try:
-    import mlx.core as mx
-except ImportError:
-    pass
-
-try:
-    import jax
-    import jax.numpy as jnp
-except ImportError:
-    pass
-
-
-def is_mlx_lm_allowed():
-    try:
-        import mlx.core as mx
-    except ImportError:
-        return False
-    return mx.metal.is_available()
-
-
-def is_jax_allowed():
-    try:
-        import jax  # noqa: F401
-    except ImportError:
-        return False
-    return True
-
-
-def get_mock_processor_inputs(array_library, num_tokens=30000):
-    """
-    logits: (4, 30,000 ) dtype=float
-    input_ids shape: (4, 2048) dtype=int
-    """
-    if array_library.startswith("torch"):
-        device = array_library.split("_")[1] if "_" in array_library else "cpu"
-
-        logits = torch.rand((4, num_tokens), dtype=torch.float, device=device)
-        input_ids = torch.randint(
-            low=0, high=num_tokens, size=(4, 2048), dtype=torch.int, device=device
-        )
-    elif array_library == "numpy":
-        logits = np.random.rand(4, num_tokens).astype(np.float32)
-        input_ids = np.random.randint(low=0, high=num_tokens, size=(4, 2048))
-    elif array_library == "mlx":
-        logits = mx.random.uniform(
-            low=-1e9, high=1e9, shape=(4, num_tokens), dtype=mx.float32
-        )
-        input_ids = mx.random.randint(
-            low=0, high=num_tokens, shape=(4, 2048), dtype=mx.int32
-        )
-    elif array_library == "jax":
-        logits = jnp.random.uniform(
-            key=jax.random.PRNGKey(0), shape=(4, num_tokens), dtype=jnp.float32
-        )
-        input_ids = jnp.random.randint(
-            key=jax.random.PRNGKey(0), low=0, high=num_tokens, shape=(4, 2048)
-        )
-    else:
-        raise ValueError
-
-    return logits, input_ids
-
-
-class HalvingLogitsProcessor(OutlinesLogitsProcessor):
-    """Simply halve the passed logits"""
-
-    def process_logits(self, input_ids, logits):
-        return logits / 2
-
-
-class LogitsProcessorPassthroughBenchmark:
-    """
-    Benchmark the time it takes to convert between array frameworks
-    This should be on the order of microseconds
-    """
-
-    params = ["torch", "numpy"]
-    if is_mlx_lm_allowed():
-        params += ["mlx"]
-    if torch.cuda.is_available():
-        params += ["torch_cuda"]
-    if torch.mps.is_available():
-        params += ["torch_mps"]
-    if is_jax_allowed():
-        params += ["jax"]
-
-    def setup(self, array_library):
-        self.logits_processor = HalvingLogitsProcessor()
-
-        self.logits, self.input_ids = get_mock_processor_inputs(array_library)
-
-    def time_passthrough(self, *params):
-        self.logits_processor(self.input_ids, self.logits)
-
-
-class LogitsProcessorStructuredBenchmark:
-    """
-    Benchmark structured generation mask application for single decoder pass
-    """
-
-    array_libraries = ["torch", "numpy"]
-    if is_mlx_lm_allowed():
-        array_libraries += ["mlx"]
-    if torch.cuda.is_available():
-        array_libraries += ["torch_cuda"]
-    if torch.mps.is_available():
-        array_libraries += ["torch_mps"]
-
-    # accept very many or very few tokens, respectively
-    patterns = [r"[^Z]*", "Z*"]
-
-    params = [array_libraries, patterns]
-    param_names = ["array_library, pattern"]
-
-    def setup(self, array_library, pattern):
-        tokenizer = models.transformers("facebook/opt-125m", device="cpu").tokenizer
-
-        self.logits_processor = RegexLogitsProcessor(pattern, tokenizer)
-
-        self.logits, self.input_ids = get_mock_processor_inputs(
-            array_library, len(tokenizer.vocabulary)
-        )
-
-    def time_structured_generation(self, array_library, pattern):
-        self.logits_processor(self.input_ids, self.logits)
diff --git a/benchmarks/bench_regex_guide.py b/benchmarks/bench_regex_guide.py
deleted file mode 100644
index 14b15da89..000000000
--- a/benchmarks/bench_regex_guide.py
+++ /dev/null
@@ -1,40 +0,0 @@
-from outlines.caching import cache_disabled
-from outlines.processors.guide import RegexGuide
-
-from .common import setup_tokenizer
-
-regex_samples = {
-    "email": r"[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*@(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?",
-    "complex_phone": "\\+?\\d{1,4}?[-.\\s]?\\(?\\d{1,3}?\\)?[-.\\s]?\\d{1,4}[-.\\s]?\\d{1,4}[-.\\s]?\\d{1,9}",
-    "simple_phone": "\\+?[1-9][0-9]{7,14}",
-    "date": r"([1-9]|0[1-9]|1[0-9]|2[0-9]|3[0-1])(\.|-|/)([1-9]|0[1-9]|1[0-2])(\.|-|/)([0-9][0-9]|19[0-9][0-9]|20[0-9][0-9])|([0-9][0-9]|19[0-9][0-9]|20[0-9][0-9])(\.|-|/)([1-9]|0[1-9]|1[0-2])(\.|-|/)([1-9]|0[1-9]|1[0-9]|2[0-9]|3[0-1])",
-    "time": r"(0?[1-9]|1[0-2]):[0-5]\d\s?(am|pm)?",
-    "ip": r"(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)",
-    "url": r"(https?:\/\/)?([\da-z\.-]+)\.([a-z\.]{2,6})([\/\w \.-]*)*\/?",
-    "ssn": r"\d{3}-\d{2}-\d{4}",
-    "complex_span_constrained_relation_extraction": "(['\"\\ ,]?((?:of|resulting|case|which|cultures|a|core|extreme|selflessness|spiritual|various|However|both|vary|in|other|secular|the|religious|among|moral|and|It|object|worldviews|altruism|traditional|material|aspect|or|life|beings|virtue|is|however|opposite|concern|an|practice|it|for|s|quality|religions|In|Altruism|animals|happiness|many|become|principle|human|selfishness|may|synonym)['\"\\ ,]?)+['\"\\ ,]?\\s\\|\\s([^|\\(\\)\n]{1,})\\s\\|\\s['\"\\ ,]?((?:of|resulting|case|which|cultures|a|core|extreme|selflessness|spiritual|various|However|both|vary|in|other|secular|the|religious|among|moral|and|It|object|worldviews|altruism|traditional|material|aspect|or|life|beings|virtue|is|however|opposite|concern|an|practice|it|for|s|quality|religions|In|Altruism|animals|happiness|many|become|principle|human|selfishness|may|synonym)['\"\\ ,]?)+['\"\\ ,]?(\\s\\|\\s\\(([^|\\(\\)\n]{1,})\\s\\|\\s([^|\\(\\)\n]{1,})\\))*\\n)*",
-}
-
-
-class RegexGuideBenchmark:
-    params = regex_samples.keys()
-
-    def setup(self, pattern_name):
-        self.tokenizer = setup_tokenizer()
-        self.pattern = regex_samples[pattern_name]
-
-    @cache_disabled()
-    def time_regex_to_guide(self, pattern_name):
-        RegexGuide.from_regex(self.pattern, self.tokenizer)
-
-
-class MemoryRegexGuideBenchmark:
-    params = ["simple_phone", "complex_span_constrained_relation_extraction"]
-
-    def setup(self, pattern_name):
-        self.tokenizer = setup_tokenizer()
-        self.pattern = regex_samples[pattern_name]
-
-    @cache_disabled()
-    def peakmem_regex_to_guide(self, pattern_name):
-        RegexGuide.from_regex(self.pattern, self.tokenizer)
diff --git a/benchmarks/common.py b/benchmarks/common.py
deleted file mode 100644
index e920888f5..000000000
--- a/benchmarks/common.py
+++ /dev/null
@@ -1,8 +0,0 @@
-from transformers import AutoTokenizer
-
-from outlines.models.transformers import TransformerTokenizer
-
-
-def setup_tokenizer():
-    tokenizer = AutoTokenizer.from_pretrained("gpt2")
-    return TransformerTokenizer(tokenizer)

From 16853c6144650b9be7e3a2baee48f61de36bd34f Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Tue, 29 Jul 2025 11:29:57 +0200
Subject: [PATCH 418/505] Delete the fsm module and the FSM output type

---
 docs/features/advanced/backends.md   |    2 -
 examples/parsing.py                  |    1 -
 outlines/backends/__init__.py        |   32 -
 outlines/backends/base.py            |   21 +-
 outlines/backends/llguidance.py      |    6 -
 outlines/backends/outlines_core.py   |   23 -
 outlines/backends/xgrammar.py        |    6 -
 outlines/fsm/__init__.py             |    1 -
 outlines/fsm/parsing.py              | 1129 --------------------------
 outlines/generator.py                |   19 +-
 outlines/release_note.md             |    2 +-
 outlines/types/__init__.py           |    2 -
 outlines/types/dsl.py                |   25 -
 outlines/types/utils.py              |    5 -
 pyproject.toml                       |    2 -
 tests/backends/test_backends.py      |   23 -
 tests/backends/test_llguidance.py    |   24 +-
 tests/backends/test_outlines_core.py |   25 +-
 tests/backends/test_xgrammar.py      |   24 +-
 tests/fsm/partial_python.lark        |  314 -------
 tests/fsm/test_parsing.py            |  307 -------
 tests/test_generator.py              |   26 +-
 tests/types/test_dsl.py              |   14 -
 tests/types/test_types_utils.py      |    8 -
 24 files changed, 9 insertions(+), 2032 deletions(-)
 delete mode 100644 outlines/fsm/__init__.py
 delete mode 100644 outlines/fsm/parsing.py
 delete mode 100644 tests/fsm/partial_python.lark
 delete mode 100644 tests/fsm/test_parsing.py

diff --git a/docs/features/advanced/backends.md b/docs/features/advanced/backends.md
index d7ff45886..84e222e10 100644
--- a/docs/features/advanced/backends.md
+++ b/docs/features/advanced/backends.md
@@ -37,7 +37,6 @@ If you do not provide a value for the `backend` argument, the default value will
 - JSON schema: `outlines_core`
 - Regex: `outlines_core`
 - Context-free grammar: `llguidance`
-- Interegular FSM: `outlines_core`
 
 ## Features matrix
 
@@ -53,4 +52,3 @@ As mentioned previously, selecting the structured generation backend is only app
 | JSON Schema | ✅ | ✅ | ✅ |
 | Regex | ✅ | ✅ | ✅ |
 | Grammar | ❌ | ✅ | ✅ |
-| FSM | ✅ | ❌ | ❌ |
diff --git a/examples/parsing.py b/examples/parsing.py
index c8aac5a2e..a579b71dd 100644
--- a/examples/parsing.py
+++ b/examples/parsing.py
@@ -15,7 +15,6 @@
     set_seed,
 )
 
-from outlines.fsm.parsing import PartialLark, PartialPythonIndenter
 
 revision = None
 checkpoint = "Salesforce/codegen-350M-mono"
diff --git a/outlines/backends/__init__.py b/outlines/backends/__init__.py
index 771580181..680e54959 100644
--- a/outlines/backends/__init__.py
+++ b/outlines/backends/__init__.py
@@ -1,7 +1,5 @@
 """Module to define the backends in charge of creating logits processors."""
 
-import interegular
-
 from outlines.backends.base import (
     BaseBackend,
     LogitsProcessorType,
@@ -13,7 +11,6 @@
 
 
 CFG_DEFAULT_BACKEND = "llguidance"
-FSM_DEFAULT_BACKEND = "outlines_core"
 JSON_SCHEMA_DEFAULT_BACKEND = "outlines_core"
 REGEX_DEFAULT_BACKEND = "outlines_core"
 
@@ -129,32 +126,3 @@ def get_cfg_logits_processor(
         model,
     )
     return backend.get_cfg_logits_processor(grammar)
-
-
-def get_fsm_logits_processor(
-    backend_name: str | None,
-    model: SteerableModel,
-    fsm: interegular,
-) -> LogitsProcessorType:
-    """Create a logits processor from an interegular FSM.
-
-    Parameters
-    ----------
-    backend_name: str | None
-        The name of the backend to use.
-    model: Model
-        The Outlines model of the user.
-    fsm: interegular.fsm.FSM
-        The interegular FSM to create a logits processor from.
-
-    Returns
-    -------
-    LogitsProcessorType
-        The logits processor.
-
-    """
-    backend = _get_backend(
-        backend_name or FSM_DEFAULT_BACKEND,
-        model,
-    )
-    return backend.get_fsm_logits_processor(fsm)
diff --git a/outlines/backends/base.py b/outlines/backends/base.py
index a8723226a..1de7f21f0 100644
--- a/outlines/backends/base.py
+++ b/outlines/backends/base.py
@@ -3,8 +3,6 @@
 from abc import ABC, abstractmethod
 from typing import Any
 
-from interegular.fsm import FSM
-
 
 LogitsProcessorType = Any
 
@@ -13,7 +11,7 @@ class BaseBackend(ABC):
     """Base class for all backends.
 
     The subclasses must implement methods that create a logits processor
-    from a JSON schema, regex, CFG or FSM.
+    from a JSON schema, regex or CFG.
 
     """
 
@@ -69,20 +67,3 @@ def get_cfg_logits_processor(self, grammar: str) -> LogitsProcessorType:
 
         """
         ...
-
-    @abstractmethod
-    def get_fsm_logits_processor(self, fsm: FSM) -> LogitsProcessorType:
-        """Create a logits processor from an interegular FSM.
-
-        Parameters
-        ----------
-        fsm: interegular.fsm.FSM
-            The interegular FSM to create a logits processor from.
-
-        Returns
-        -------
-        LogitsProcessorType
-            The logits processor.
-
-        """
-        ...
diff --git a/outlines/backends/llguidance.py b/outlines/backends/llguidance.py
index 940602f98..c5a738616 100644
--- a/outlines/backends/llguidance.py
+++ b/outlines/backends/llguidance.py
@@ -294,9 +294,3 @@ def get_cfg_logits_processor(
         return LLGuidanceLogitsProcessor(
             grammar_spec, self.llg_tokenizer, self.tensor_library_name
         )
-
-    def get_fsm_logits_processor(self, fsm):
-        raise NotImplementedError(
-            "LLGuidanceBackend does not support FSM logits processors. "
-            "Use the outlines_core backend instead."
-        )
diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index 3fa5d9726..172991709 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -1,7 +1,5 @@
 """Backend class for Outlines Core."""
 
-from interegular.fsm import FSM
-
 from outlines.backends.base import BaseBackend
 from outlines.models import SteerableModel
 from outlines.processors import (
@@ -70,24 +68,3 @@ def get_cfg_logits_processor(self, grammar):
             "Context-free grammar output type is not supported for the "
             + "Outlines Core backend."
         )
-
-    def get_fsm_logits_processor(self, fsm: FSM) -> GuideLogitsProcessor:
-        """Create a logits processor from an interegular FSM.
-
-        Parameters
-        ----------
-        fsm: interegular.fsm.FSM
-            The interegular FSM to create a logits processor from.
-
-        Returns
-        -------
-        LogitsProcessor
-            The logits processor to use to constrain the generation.
-
-        """
-        guide = RegexGuide.from_interegular_fsm(fsm, self.tokenizer)
-        return GuideLogitsProcessor(
-            self.tokenizer,
-            guide,
-            self.tensor_library_name,
-        )
diff --git a/outlines/backends/xgrammar.py b/outlines/backends/xgrammar.py
index 36be92bfa..fcb1672da 100644
--- a/outlines/backends/xgrammar.py
+++ b/outlines/backends/xgrammar.py
@@ -132,9 +132,3 @@ def get_cfg_logits_processor(self, grammar: str) -> "LogitsProcessor":
         """
         compiled_grammar = self.grammar_compiler.compile_grammar(grammar)
         return XGrammarLogitsProcessor(compiled_grammar)
-
-    def get_fsm_logits_processor(self, fsm):
-        raise NotImplementedError(
-            "XGrammarBackend does not support FSM logits processors. "
-            "Use the outlines_core backend instead."
-        )
diff --git a/outlines/fsm/__init__.py b/outlines/fsm/__init__.py
deleted file mode 100644
index 4e38718ef..000000000
--- a/outlines/fsm/__init__.py
+++ /dev/null
@@ -1 +0,0 @@
-"""Finite state machines builder for CFG generation."""
diff --git a/outlines/fsm/parsing.py b/outlines/fsm/parsing.py
deleted file mode 100644
index 3831c57c1..000000000
--- a/outlines/fsm/parsing.py
+++ /dev/null
@@ -1,1129 +0,0 @@
-"""Build a parser from a grammar to create a finite state machine."""
-
-from copy import copy, deepcopy
-from dataclasses import dataclass
-from functools import lru_cache
-from typing import (
-    Any,
-    Dict,
-    FrozenSet,
-    Generator,
-    Iterator,
-    List,
-    Optional,
-    Sequence,
-    Set,
-    Tuple,
-    Union,
-)
-
-import interegular
-from interegular.fsm import FSM, Alphabet, OblivionError
-from interegular.patterns import Unsupported
-from lark import Lark, Token
-from lark.common import LexerConf, ParserConf
-from lark.exceptions import LexError, UnexpectedInput
-from lark.indenter import Indenter
-from lark.lexer import (
-    BasicLexer,
-    ContextualLexer,
-    LexerState,
-    LexerThread,
-    Scanner,
-    UnexpectedCharacters,
-    UnexpectedToken,
-    _create_unless,
-)
-from lark.parser_frontends import (
-    ParsingFrontend,
-    PostLexConnector,
-    _validate_frontend_args,
-)
-from lark.parsers.lalr_analysis import (
-    Action,
-    IntParseTable,
-    LALR_Analyzer,
-    ParseTable,
-    Shift,
-)
-from lark.parsers.lalr_interactive_parser import InteractiveParser
-from lark.parsers.lalr_parser import LALR_Parser, ParseConf, ParserState, _Parser
-from outlines_core.fsm.regex import (
-    BetterFSM,
-    get_token_transition_keys,
-    make_deterministic_fsm,
-)
-
-PartialParseState = Tuple[str, int]
-ParseStateType = Union[int, FrozenSet]
-
-
-@dataclass
-class PartialTerminalInfo:
-    priority: int
-    terminal_name: str
-    can_transition: bool
-    is_final: bool
-
-
-@dataclass
-class PartialTokensInfo:
-    fsm_state_seq: Tuple[int, ...]
-    is_not_finished: bool
-    terminals_and_info: Tuple[PartialTerminalInfo, ...]
-    final_terminals_and_info: Tuple[PartialTerminalInfo, ...]
-
-
-class PartialParserConf(ParserConf):
-    __serialize_fields__ = (
-        "rules",
-        "start",
-        "parser_type",
-        "deterministic",
-        "use_value_stack",
-    )
-
-    def __init__(self, rules, callbacks, start, deterministic, use_value_stack):
-        super().__init__(rules, callbacks, start)
-        self.deterministic = deterministic
-        self.use_value_stack = use_value_stack
-
-
-class PartialLark(Lark):
-    __serialize_fields__ = (
-        "parser",
-        "rules",
-        "options",
-        "deterministic",
-        "use_value_stack",
-    )
-
-    def __init__(self, grammar, **options):
-        # TODO: Could've extended `LarkOptions`, but all these extensions are
-        # already way too much (and brittle).  This library really needs a
-        # complete refactoring.
-        self.deterministic = options.pop("deterministic", False)
-        self.use_value_stack = options.pop("use_value_stack", False)
-        options["regex"] = True
-        super().__init__(grammar, **options)
-        assert self.options.parser == "lalr"
-
-    def _build_lexer(self, dont_ignore: bool = False) -> "PartialBasicLexer":
-        lexer_conf = self.lexer_conf
-        if dont_ignore:
-            from copy import copy
-
-            lexer_conf = copy(lexer_conf)
-            lexer_conf.ignore = ()
-
-        return PartialBasicLexer(lexer_conf)
-
-    def _build_parser(self) -> "PartialParsingFrontend":
-        self._prepare_callbacks()
-        _validate_frontend_args(self.options.parser, self.options.lexer)
-        parser_conf = PartialParserConf(
-            self.rules,
-            self._callbacks,
-            self.options.start,
-            self.deterministic,
-            self.use_value_stack,
-        )
-
-        # This is `_construct_parsing_frontend` expanded/inlined
-        parser_type = self.options.parser
-        lexer_type = self.options.lexer
-        lexer_conf = self.lexer_conf
-
-        assert isinstance(lexer_conf, LexerConf)
-        assert isinstance(parser_conf, ParserConf)
-        parser_conf.parser_type = parser_type
-        self.lexer_conf.lexer_type = lexer_type
-        return PartialParsingFrontend(lexer_conf, parser_conf, self.options)
-
-    def __repr__(self):
-        return "{}(open({!r}), parser={!r}, lexer={!r}, ...)".format(
-            type(self).__name__,
-            self.source_path,
-            self.options.parser,
-            self.options.lexer,
-        )
-
-    def parse_from_state(self, parse_state: "PartialParseState", is_end=False):
-        return self.parser.parser.parser.parse_from_state(parse_state, is_end=is_end)
-
-
-class PartialLexerThread(LexerThread):
-    def __copy__(self):
-        return type(self)(copy(self.lexer), copy(self.state))
-
-    def __repr__(self):
-        return f"{type(self).__name__}(lexer={self.lexer!r}, state={self.state!r})"
-
-
-class PartialPostLexConnector(PostLexConnector):
-    def __copy__(self):
-        return type(self)(self.lexer, copy(self.postlexer))
-
-    def __repr__(self):
-        return (
-            f"{type(self).__name__}(lexer={self.lexer!r}, postlexer={self.postlexer!r})"
-        )
-
-
-class PartialParsingFrontend(ParsingFrontend):
-    def __init__(self, lexer_conf, parser_conf, options, parser=None):
-        assert parser_conf.parser_type == "lalr"
-
-        options._plugins["LALR_Parser"] = PartialLALRParser
-        options._plugins["BasicLexer"] = PartialBasicLexer
-        options._plugins["ContextualLexer"] = PartialContextualLexer
-        options._plugins["LexerThread"] = PartialLexerThread
-
-        super().__init__(lexer_conf, parser_conf, options, parser=parser)
-
-        if lexer_conf.postlex:
-            self.lexer = PartialPostLexConnector(self.lexer.lexer, lexer_conf.postlex)
-
-        self._termset_fsm_info = None
-        self._symbols_to_states: Optional[
-            Dict[str, Set[Tuple[ParseStateType, Action]]]
-        ] = None
-        self._reverse_shifts: Optional[
-            Dict[ParseStateType, Dict[str, Set[ParseStateType]]]
-        ] = None
-        # self._state_transition_map: Optional[
-        #     Dict[Tuple[ParseStateType, str], Set[ParseStateType]]
-        # ] = None
-
-    def _compute_maps(
-        self,
-    ):
-        """Compute state transition and symbols-to-states maps."""
-        self._reverse_shifts = {}
-        self._symbols_to_states = {}
-
-        parse_table = self.parser.parser.parse_table
-
-        for from_state, symbols_to_ops in parse_table.states.items():
-            for symbol, op in symbols_to_ops.items():
-                if op[0] == Shift:
-                    symbols_to_from_states = self._reverse_shifts.setdefault(op[1], {})
-                    symbols_to_from_states.setdefault(symbol, set()).add(from_state)
-                self._symbols_to_states.setdefault(symbol, set()).add((from_state, op))
-
-        # # TODO: This approach is very wasteful.
-        # context_lexer = get_contextual_lexer(self)
-        # self._state_transition_map = {}
-        #
-        # for from_state, transitions in parse_table.states.items():
-        #     for symbol, action in transitions.items():
-        #         # TODO: Filter non-terminals
-        #         if symbol not in context_lexer.root_lexer.terminals_by_name:
-        #             continue
-        #
-        #         if action[0] is Shift:
-        #             self._state_transition_map.setdefault(
-        #                 (from_state, symbol), set()
-        #             ).add(action[1])
-        #             continue
-        #
-        #         antecedent_state_seqs = parse_to_terminal(self, [(from_state,)], symbol)
-        #
-        #         for antecedent_state_seq in antecedent_state_seqs:
-        #             antecedent_state = antecedent_state_seq[-1]
-        #             self._state_transition_map.setdefault(
-        #                 (from_state, symbol), set()
-        #             ).add(antecedent_state)
-
-    def _compute_termset_fsm_info(self):
-        """Collect and return information about terminal symbol sets and their FSMs.
-
-        Terminal symbol sets (or "termsets") are ordered sequences of terminal
-        symbols that are used by each parser state.  Associated with each is a
-        collection of FSMs for each terminal and a single parse state FSM that is
-        the union of each terminal's FSM.
-
-        This constructs a list of tuples containing the termset, the set of
-        parse states that use the termsets, parse state FSMs, and information
-        mapping the components of the parse state FSMs to their terminal symbol
-        FSMs.
-
-        """
-        context_lexer = get_contextual_lexer(self)
-        termsets_to_fsms = {}
-        termsets_to_parse_states: Dict[Tuple[str, ...], Set[ParseStateType]] = {}
-        for parse_state, lexer in context_lexer.lexers.items():
-            scanner = lexer.scanner
-            key = tuple(term.name for term in scanner.terminals)
-            termsets_to_fsms[key] = (scanner.fsm, scanner.fsms_to_trans_finals)
-            termsets_to_parse_states.setdefault(key, set()).add(parse_state)
-
-        self._termset_fsm_info = [
-            (
-                termset,
-                frozenset(termsets_to_parse_states[termset]),
-                fsm,
-                fsms_to_trans_finals,
-            )
-            for termset, (fsm, fsms_to_trans_finals) in termsets_to_fsms.items()
-        ]
-
-    @property
-    def termset_fsm_info(self):
-        if self._termset_fsm_info is None:
-            self._compute_termset_fsm_info()
-        return self._termset_fsm_info
-
-    @property
-    def symbols_to_states(self):
-        if self._symbols_to_states is None:
-            self._compute_maps()
-        return self._symbols_to_states
-
-    @property
-    def reverse_shifts(self):
-        if self._reverse_shifts is None:
-            self._compute_maps()
-        return self._reverse_shifts
-
-    # @property
-    # def state_transition_map(self):
-    #     if self._state_transition_map is None:
-    #         self._compute_maps()
-    #     return self._state_transition_map
-
-
-class PartialLALRParser(LALR_Parser):
-    def __init__(self, parser_conf, debug=False, strict=False):
-        analysis = LALR_Analyzer(
-            parser_conf, debug=debug if not parser_conf.deterministic else True
-        )
-        analysis.compute_lalr()
-        callbacks = parser_conf.callbacks
-
-        self.parser_conf = parser_conf
-        self._parse_table = analysis.parse_table
-
-        if parser_conf.deterministic:
-            old_to_new = {}
-
-            def to_tuple(v):
-                new = old_to_new.get(v)
-                if new is None:
-                    new = tuple(sorted(v, key=lambda y: str(y)))
-                    old_to_new[v] = new
-                return new
-
-            enum = sorted(
-                self._parse_table.states.keys(),
-                key=lambda x: str(sorted(x, key=lambda y: str(y))),
-            )
-
-            new_states = {}
-            for s in enum:
-                transitions = {
-                    term: op if op[0] is not Shift else (op[0], to_tuple(op[1]))
-                    for term, op in self._parse_table.states[s].items()
-                }
-                new_states[to_tuple(s)] = transitions
-
-            self._parse_table = type(self._parse_table)(
-                new_states,
-                {k: to_tuple(v) for k, v in self._parse_table.start_states.items()},
-                {k: to_tuple(v) for k, v in self._parse_table.end_states.items()},
-            )
-
-            if not debug:
-                self._parse_table = IntParseTable.from_ParseTable(self._parse_table)
-                self.states_to_rulesets = dict(
-                    zip(self._parse_table.states.keys(), new_states.keys())
-                )
-
-        self.parser = PartialParser(
-            self._parse_table,
-            callbacks,
-            debug,
-            use_value_stack=parser_conf.use_value_stack,
-        )
-
-    @classmethod
-    def deserialize(cls, data, memo, callbacks, debug=False):
-        inst = cls.__new__(cls)
-        inst._parse_table = ParseTable.deserialize(data, memo)
-        inst.parser = PartialParser(inst._parse_table, callbacks, debug)
-        return inst
-
-
-class PartialParserState(ParserState):
-    __slots__ = "use_value_stack"
-
-    def __init__(
-        self,
-        parse_conf,
-        lexer,
-        state_stack=None,
-        value_stack=None,
-        use_value_stack=False,
-    ):
-        super().__init__(
-            parse_conf, lexer, state_stack=state_stack, value_stack=value_stack
-        )
-        self.use_value_stack = use_value_stack
-
-    def feed_token(self, token, is_end=False):
-        if token.type == "partial":
-            # If none of the potential terminals can transition, we need to know now
-            current_state = self.state_stack[-1]
-            current_lexer = get_contextual_lexer(self.lexer).lexers[current_state]
-
-            # We have to feed the token and determine whether or not at least
-            # one terminal is consistent with the stack; otherwise, we'll miss
-            # invalid REDUCE cases.
-            # TODO: We should track separate parses conditional on possible
-            # token/symbol types, then we can coherently reuse the following
-            # results instead of recomputing it later.
-            can_transition = False
-            for terminal_info in token.value.terminals_and_info:
-                if terminal_info.terminal_name not in current_lexer.ignore_types:
-                    test_token = Token.new_borrow_pos(
-                        terminal_info.terminal_name, "", token
-                    )
-
-                    stack = copy(self.state_stack)
-                    try:
-                        self.feed_token_no_stack(test_token, is_end=is_end)
-                        can_transition = True
-                        break
-                    except UnexpectedToken:
-                        continue
-                    finally:
-                        self.state_stack = stack
-                else:
-                    can_transition = True
-
-            if not can_transition:
-                expected = {
-                    s
-                    for s in self.parse_conf.states[current_state].keys()
-                    if s.isupper()
-                }
-                raise UnexpectedToken(
-                    token, expected, state=self, interactive_parser=None
-                )
-
-        elif self.use_value_stack:
-            super().feed_token(token, is_end=is_end)
-        else:
-            self.feed_token_no_stack(token, is_end=is_end)
-
-    def feed_token_no_stack(self, token, is_end=False):
-        """
-        This is a copy of `ParserState.feed_token` with all the value stack
-        steps removed.  Since we're not exactly parsing in order to obtain a
-        CST or anything similar, we can avoid the growing expense of tracking
-        the parse tree.
-        """
-        state_stack = self.state_stack
-        states = self.parse_conf.states
-        end_state = self.parse_conf.end_state
-
-        while True:
-            state = state_stack[-1]
-            try:
-                action, arg = states[state][token.type]
-            except KeyError:
-                expected = {s for s in states[state].keys() if s.isupper()}
-                raise UnexpectedToken(
-                    token, expected, state=self, interactive_parser=None
-                )
-
-            assert arg != end_state
-
-            if action is Shift:
-                # shift once and return
-                assert not is_end
-                state_stack.append(arg)
-                return
-            else:
-                # reduce+shift as many times as necessary
-                rule = arg
-                size = len(rule.expansion)
-                if size:
-                    del state_stack[-size:]
-
-                _action, new_state = states[state_stack[-1]][rule.origin.name]
-                assert _action is Shift
-                state_stack.append(new_state)
-
-                if is_end and state_stack[-1] == end_state:
-                    return
-
-    def feed_eof(self):
-        last_token = self.lexer.state.last_token
-
-        if last_token is None:
-            eof_token = self.lexer._Token("$END", "", 0, 1, 1)
-        else:
-            eof_token = Token.new_borrow_pos("$END", "", last_token)
-
-        new_token_is_legal = (
-            last_token is None
-            or last_token.type != "partial"
-            or any(ti.is_final for ti in last_token.value.terminals_and_info)
-        )
-        if new_token_is_legal:
-            self.feed_token(eof_token, is_end=True)
-        else:
-            raise UnexpectedToken(eof_token, [], state=self, interactive_parser=None)
-
-    def choices(self):
-        return self.parse_conf.parse_table.states[self.position]
-
-    def accepts(self):
-        """
-        Adapted from https://github.com/lark-parser/lark/blob/be542c2ff6d968817df019b8bf03f37b3111c08c/lark/parsers/lalr_interactive_parser.py#L95
-        Returns the set of possible tokens that will advance the parser into a new valid state.
-        """
-        accepts = set()
-        conf_no_callbacks = copy(self.parse_conf)
-        # We don't want to call callbacks here since those might have arbitrary side effects
-        # and are unnecessarily slow.
-        conf_no_callbacks.callbacks = {}
-        for t in self.choices():
-            if t.isupper():  # is terminal?
-                new_state = copy(self)
-                new_state.parse_conf = conf_no_callbacks
-                try:
-                    new_state.feed_token(new_state.lexer._Token(t, ""))
-                except UnexpectedToken:
-                    pass
-                else:
-                    accepts.add(t)
-        return accepts
-
-    def __copy__(self):
-        return type(self)(
-            self.parse_conf,
-            copy(self.lexer),
-            copy(self.state_stack),
-            deepcopy(self.value_stack),
-            use_value_stack=self.use_value_stack,
-        )
-
-    def __repr__(self):
-        return f"{type(self).__name__}(lexer={self.lexer!r}, state_stack={self.state_stack!r})"
-
-
-class PartialParser(_Parser):
-    def __init__(self, parse_table, callbacks, debug=False, use_value_stack=False):
-        super().__init__(parse_table, callbacks, debug=debug)
-        self.use_value_stack = use_value_stack
-
-    def parse(
-        self, lexer, start, value_stack=None, state_stack=None, start_interactive=False
-    ):
-        parse_conf = ParseConf(self.parse_table, self.callbacks, start)
-        parser_state = PartialParserState(
-            parse_conf, copy(lexer), state_stack, value_stack, self.use_value_stack
-        )
-        if start_interactive:
-            return InteractiveParser(self, parser_state, parser_state.lexer)
-        return self.parse_from_state(parser_state)
-
-    def parse_from_state(self, state, last_token=None, is_end=False):
-        try:
-            token = last_token
-            for token in state.lexer.lex(state):
-                state.feed_token(token)
-
-            if is_end and (not token or token.type != "partial"):
-                state.feed_eof()
-
-            return state
-        except UnexpectedInput as e:
-            try:
-                e.interactive_parser = InteractiveParser(self, state, state.lexer)
-            except NameError:
-                pass
-            raise e
-        except Exception:
-            if self.debug:
-                print("")
-                print("STATE STACK DUMP")
-                print("----------------")
-                for i, s in enumerate(state.state_stack):
-                    print("%d)" % i, s)
-                print("")
-
-            raise
-
-
-class PartialScanner(Scanner):
-    @classmethod
-    @lru_cache
-    def construct_terminal_fsm(cls, terminal):
-        # TODO: This should really be done at the lexer/parser level so that
-        # the lifetime of these objects is tied to the parser itself.
-        regex_str = terminal.pattern.to_regexp()
-        pattern = interegular.parse_pattern(regex_str)
-        fsm, _ = make_deterministic_fsm(pattern.to_fsm().reduce())
-        return fsm, pattern.prefix_postfix
-
-    def __init__(self, terminals, g_regex_flags, re_, use_bytes, match_whole=False):
-        self.terminals = terminals
-        self.g_regex_flags = g_regex_flags
-        self.use_bytes = use_bytes
-        self.match_whole = match_whole
-        self.allowed_types = {t.name for t in self.terminals}
-        self._mres = None
-
-        fsms = []
-        for t in self.terminals:
-            fsm, prefix_postfix = self.construct_terminal_fsm(t)
-
-            # TODO FIXME: We don't support this right now.
-            assert prefix_postfix == (0, 0)
-
-            fsms.append(fsm)
-
-        self.fsm, self.fsms_to_trans_finals = fsm_union(fsms)
-
-    def get_terminals_info(
-        self, fsm_state_seq
-    ) -> Tuple[Tuple[PartialTerminalInfo, ...], Tuple[PartialTerminalInfo, ...]]:
-        """Get the possible terminal symbols for an FSM state sequence."""
-        terminals_and_info: Tuple[PartialTerminalInfo, ...] = ()
-        final_terminals_and_info: Tuple[PartialTerminalInfo, ...] = ()
-        for i, (fsm_id, fsm_reads_more, in_final) in enumerate(
-            get_sub_fsms_from_seq(fsm_state_seq, self.fsms_to_trans_finals)
-        ):
-            terminal_name = self.terminals[fsm_id].name
-            info = PartialTerminalInfo(i, terminal_name, fsm_reads_more, in_final)
-            terminals_and_info += (info,)
-            if in_final:
-                final_terminals_and_info += (info,)
-
-        return terminals_and_info, final_terminals_and_info
-
-    def match(self, text, pos, last_fsm_state_seq: Optional[Tuple[int, ...]] = None):
-        """Determine an FSM match over `text` starting at `pos` and continuing `last_fsm_state_seq`."""
-
-        start_pos = pos
-
-        if last_fsm_state_seq:
-            assert len(last_fsm_state_seq) > 1
-            start_pos += len(last_fsm_state_seq) - 1
-            start_state = last_fsm_state_seq[-1]
-        else:
-            start_state = self.fsm.initial
-
-        text_part = text[start_pos:]
-
-        text_transitions = get_token_transition_keys(
-            self.fsm.fsm_info.alphabet_symbol_mapping,
-            self.fsm.fsm_info.alphabet_anything_value,
-            text_part,
-        )
-
-        state_seq = walk_fsm(
-            self.fsm,
-            text_transitions,
-            start_state,
-            full_match=self.match_whole,
-        )
-
-        if not state_seq:
-            return None
-
-        if last_fsm_state_seq:
-            res = last_fsm_state_seq + tuple(state_seq)
-        else:
-            res = (start_state,) + tuple(state_seq)
-
-        return res
-
-
-class PartialContextualLexer(ContextualLexer):
-    def __init__(self, conf: "LexerConf", states, always_accept=()):
-        terminals = list(conf.terminals)
-        terminals_by_name = conf.terminals_by_name
-
-        trad_conf = copy(conf)
-        trad_conf.terminals = terminals
-
-        lexer_by_symbols: Dict = {}
-        self.lexers = {}
-        for state, accepts in states.items():
-            key = frozenset(accepts)
-            try:
-                lexer = lexer_by_symbols[key]
-            except KeyError:
-                accepts = set(accepts) | set(conf.ignore) | set(always_accept)
-                lexer_conf = copy(trad_conf)
-                lexer_conf.terminals = [
-                    terminals_by_name[n] for n in accepts if n in terminals_by_name
-                ]
-                if not lexer_conf.terminals:
-                    continue
-                lexer = PartialBasicLexer(lexer_conf)
-                lexer_by_symbols[key] = lexer
-
-            self.lexers[state] = lexer
-
-        assert trad_conf.terminals is terminals
-        self.root_lexer = PartialBasicLexer(trad_conf)
-
-    def lex(self, lexer_state: LexerState, parser_state: Any) -> Iterator[Token]:
-        try:
-            while True:
-                lexer = self.lexers[parser_state.position]
-                next_tok = lexer.next_token(lexer_state, parser_state)
-                yield next_tok
-        except EOFError:
-            pass
-        except KeyError:
-            if len(lexer_state.text) > lexer_state.line_ctr.char_pos:
-                raise UnexpectedCharacters(
-                    lexer_state.text,
-                    lexer_state.line_ctr.char_pos,
-                    lexer_state.line_ctr.line,
-                    lexer_state.line_ctr.column,
-                    allowed=False,
-                    token_history=lexer_state.last_token and [lexer_state.last_token],
-                    state=parser_state,
-                    terminals_by_name=self.root_lexer.terminals,
-                )
-
-
-class PartialBasicLexer(BasicLexer):
-    def __init__(self, conf: "LexerConf"):
-        super().__init__(conf)
-        # Eagerly construct the scanner
-        self._build_scanner()
-
-    def _build_scanner(self):
-        # This seems incredibly convoluted: `lark` creates callback-triggered
-        # nested scanners for regex-defined terminals that overlap with
-        # string-defined terminals when both types of terminals have the same
-        # priority.  Unless I'm missing something important, why not simply
-        # reorder the terminals so that the string-defined ones come before the
-        # regex-defined ones?
-        terminals, self.callback = _create_unless(
-            self.terminals, self.g_regex_flags, self.re, self.use_bytes
-        )
-
-        # We can't let people arbitrarily mess with the scanning process.
-        assert not self.user_callbacks
-        # for type_, f in self.user_callbacks.items():
-        #     if type_ in self.callback:
-        #         # Already a callback there, probably UnlessCallback
-        #         self.callback[type_] = CallChain(
-        #             self.callback[type_], f, lambda t: t.type == type_
-        #         )
-        #     else:
-        #         self.callback[type_] = f
-
-        # We used the "callback" results to reorder the terminals (see the
-        # comments above).
-        for terminal_name, callback in self.callback.items():
-            terminal = self.terminals_by_name[terminal_name]
-            for sub_terminal in callback.scanner.terminals:
-                self.terminals.remove(sub_terminal)
-                idx = self.terminals.index(terminal)
-                self.terminals.insert(idx, sub_terminal)
-
-        self._scanner = PartialScanner(
-            self.terminals, self.g_regex_flags, self.re, self.use_bytes
-        )
-
-    def match(self, text, pos, last_fsm_state_seq=None):
-        return self.scanner.match(text, pos, last_fsm_state_seq)
-
-    def next_token(self, lex_state: LexerState, parser_state: Any = None) -> Token:
-        last_token = lex_state.last_token
-
-        last_fsm_state_seq = None
-        if last_token and last_token.type == "partial":
-            # Continue from last partial lexer state
-            last_fsm_state_seq = last_token.value.fsm_state_seq
-
-        line_ctr = lex_state.line_ctr
-        end_pos = line_ctr.char_pos + (
-            len(last_fsm_state_seq) - 1 if last_fsm_state_seq else 0
-        )
-        while end_pos < len(lex_state.text):
-            res = self.match(lex_state.text, line_ctr.char_pos, last_fsm_state_seq)
-
-            if not res:
-                if (
-                    not last_fsm_state_seq
-                    or last_fsm_state_seq[-1] not in self.scanner.fsm.finals
-                ):
-                    allowed = self.scanner.allowed_types - self.ignore_types
-                    if not allowed:
-                        allowed = {"<END-OF-FILE>"}
-                    raise UnexpectedCharacters(
-                        lex_state.text,
-                        line_ctr.char_pos,
-                        line_ctr.line,
-                        line_ctr.column,
-                        allowed=allowed,
-                        token_history=lex_state.last_token and [lex_state.last_token],
-                        state=parser_state,
-                        terminals_by_name=self.terminals_by_name,
-                    )
-
-                # The partial match might be complete now
-                fsm_state_seq = last_token.value.fsm_state_seq
-                terminals_and_info = last_token.value.terminals_and_info
-                final_terminals_and_info = last_token.value.final_terminals_and_info
-            else:
-                fsm_state_seq = res
-                (
-                    terminals_and_info,
-                    final_terminals_and_info,
-                ) = self.scanner.get_terminals_info(fsm_state_seq)
-
-            priority_terminal_info = (
-                final_terminals_and_info[0]
-                if final_terminals_and_info
-                else terminals_and_info[0]
-            )
-
-            is_not_finished = (
-                not priority_terminal_info.is_final
-                or priority_terminal_info.can_transition
-                or len(terminals_and_info) > 1
-            )
-
-            start_pos = line_ctr.char_pos
-            end_pos = start_pos + len(fsm_state_seq) - 1
-
-            if end_pos >= len(lex_state.text) and is_not_finished:
-                type_name = "partial"
-                token_value = PartialTokensInfo(
-                    fsm_state_seq,
-                    is_not_finished,
-                    terminals_and_info,
-                    final_terminals_and_info,
-                )
-                # Don't update the line counter states until we've finished
-                value = ""
-            else:
-                type_name = priority_terminal_info.terminal_name
-                # The token value should contain all partial scan parts in this
-                # case
-                value = token_value = lex_state.text[start_pos:end_pos]
-
-            assert isinstance(self.callback, Dict)
-
-            if type_name not in self.ignore_types:
-                t = Token(
-                    type_name,
-                    token_value,
-                    line_ctr.char_pos,
-                    line_ctr.line,
-                    line_ctr.column,
-                )
-
-                line_ctr.feed(value, type_name in self.newline_types)
-
-                t.end_line = line_ctr.line
-                t.end_column = line_ctr.column
-                t.end_pos = line_ctr.char_pos
-                if t.type in self.callback:
-                    t = self.callback[t.type](t)
-                    if not isinstance(t, Token):
-                        raise LexError(
-                            "Callbacks must return a token (returned %r)" % t
-                        )
-                lex_state.last_token = t
-                return t
-
-            if type_name in self.callback:
-                t2 = Token(
-                    type_name, value, line_ctr.char_pos, line_ctr.line, line_ctr.column
-                )
-                self.callback[type_name](t2)
-
-            line_ctr.feed(value, type_name in self.newline_types)
-
-            last_fsm_state_seq = None
-
-        raise EOFError(self)
-
-
-class PartialIndenter(Indenter):
-    """An `Indenter` that doesn't reset its state every time `process` is called."""
-
-    def process(self, stream):
-        return self._process(stream)
-
-    def _process(self, stream):
-        for token in stream:
-            # These were previously *after* the `yield`, but that makes the
-            # state tracking unnecessarily convoluted.
-            if token.type in self.OPEN_PAREN_types:
-                self.paren_level += 1
-            elif token.type in self.CLOSE_PAREN_types:
-                self.paren_level -= 1
-                if self.paren_level < 0:
-                    raise UnexpectedToken(token, [])
-
-            if token.type == self.NL_type:
-                yield from self.handle_NL(token)
-            else:
-                yield token
-
-        # TODO: What do we want to do here?
-        # while len(self.indent_level) > 1:
-        #     self.indent_level.pop()
-        #     yield Token(self.DEDENT_type, "")
-
-    def accepts_token_type(self, token_type):
-        if token_type in self.CLOSE_PAREN_types and self.paren_level - 1 < 0:
-            return False
-
-        # TODO:
-        # if token_type == self.NL_type and self.paren_level == 0:
-        #     ...
-        #     return False
-
-        return True
-
-    def __copy__(self):
-        res = type(self)()
-        res.paren_level = self.paren_level
-        res.indent_level = copy(self.indent_level)
-        return res
-
-    def __repr__(self):
-        return f"{type(self).__name__}(paren_level={self.paren_level!r}, indent_level={self.indent_level!r})"
-
-
-class PartialPythonIndenter(PartialIndenter):
-    NL_type = "_NEWLINE"
-    OPEN_PAREN_types = ["LPAR", "LSQB", "LBRACE"]
-    CLOSE_PAREN_types = ["RPAR", "RSQB", "RBRACE"]
-    INDENT_type = "_INDENT"
-    DEDENT_type = "_DEDENT"
-    tab_len = 8
-
-
-def get_contextual_lexer(x: Union[PartialLexerThread, PartialParsingFrontend]):
-    if isinstance(x.lexer, ContextualLexer):
-        return x.lexer
-    else:
-        return x.lexer.lexer
-
-
-def terminals_to_fsms(lp: PartialLark) -> Dict[str, FSM]:
-    """Construct a ``dict`` mapping terminal symbol names to their finite state machines."""
-
-    symbol_names_and_fsms = {}
-    for terminal in lp.terminals:
-        pattern = interegular.parse_pattern(terminal.pattern.to_regexp())
-        # TODO: Use `pyparser.terminals[0].pattern.flags`?
-        try:
-            fsm, _ = make_deterministic_fsm(pattern.to_fsm().reduce())
-        except Unsupported:
-            fsm = None
-
-        symbol_names_and_fsms[terminal.name] = fsm
-
-    return symbol_names_and_fsms
-
-
-def fsm_union(
-    fsms: Sequence[FSM],
-) -> Tuple[FSM, Dict[int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]]]:
-    """Construct an FSM representing the union of the FSMs in `fsms`.
-
-    This is an updated version of `interegular.fsm.FSM.union` made to return an
-    extra map of component FSMs to the sets of state transitions that
-    correspond to them in the new FSM.
-
-    """
-
-    alphabet, new_to_old = Alphabet.union(*[fsm.alphabet for fsm in fsms])
-
-    indexed_fsms = tuple(enumerate(fsms))
-
-    initial = {i: fsm.initial for (i, fsm) in indexed_fsms}
-
-    # Dedicated function accepting a "superset" and returning the next
-    # "superset" obtained by following this transition in the new FSM
-    def follow(current_state, new_transition: int):
-        next = {}
-        for i, f in indexed_fsms:
-            old_transition = new_to_old[i][new_transition]
-            if (
-                i in current_state
-                and current_state[i] in f.map
-                and old_transition in f.map[current_state[i]]
-            ):
-                next[i] = f.map[current_state[i]][old_transition]
-        if not next:
-            raise OblivionError
-        return next
-
-    states = [initial]
-    finals: Set[int] = set()
-    map: Dict[int, Dict[int, int]] = {}
-
-    # Map component FSMs to their new state-to-state transitions, finals, and a
-    # map translating component FSM states to aggregate FSM states
-    fsms_to_trans_finals: Dict[
-        int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]
-    ] = {}
-
-    i = 0
-    while i < len(states):
-        state = states[i]
-
-        # Add to the finals of the aggregate FSM whenever we hit a final in a
-        # component FSM
-        if any(state.get(j, -1) in fsm.finals for (j, fsm) in indexed_fsms):
-            finals.add(i)
-
-        # Compute the map for this state
-        map[i] = {}
-        for transition in alphabet.by_transition:
-            try:
-                next = follow(state, transition)
-            except OblivionError:
-                # Reached an oblivion state; don't list it
-                continue
-            else:
-                try:
-                    # TODO: Seems like this could--and should--be avoided
-                    j = states.index(next)
-                except ValueError:
-                    j = len(states)
-                    states.append(next)
-
-                map[i][transition] = j
-
-                for fsm_id, fsm_state in next.items():
-                    (
-                        fsm_transitions,
-                        fsm_finals,
-                        fsm_old_to_new,
-                    ) = fsms_to_trans_finals.setdefault(fsm_id, (set(), set(), {}))
-                    old_from = state[fsm_id]
-                    old_to = fsm_state
-                    fsm_old_to_new.setdefault(old_from, set()).add(i)
-                    fsm_old_to_new.setdefault(old_to, set()).add(j)
-                    fsm_transitions.add((i, j))
-                    if fsm_state in fsms[fsm_id].finals:
-                        fsm_finals.add(j)
-
-        i += 1
-
-    fsm = FSM(
-        alphabet=alphabet,
-        states=range(len(states)),
-        initial=0,
-        finals=finals,
-        map=map,
-        __no_validation__=True,
-    )
-
-    fsm, old_to_new_states = make_deterministic_fsm(fsm)
-    _fsms_to_trans_finals = {
-        fsm_id: (
-            {(old_to_new_states[s1], old_to_new_states[s2]) for s1, s2 in transitions},
-            {old_to_new_states[s] for s in finals},
-            {
-                old_state: {old_to_new_states[new_state] for new_state in new_states}
-                for old_state, new_states in old_to_new.items()
-            },
-        )
-        for fsm_id, (transitions, finals, old_to_new) in sorted(
-            fsms_to_trans_finals.items(), key=lambda x: x[0]
-        )
-    }
-
-    return (
-        fsm,
-        _fsms_to_trans_finals,
-    )
-
-
-def get_sub_fsms_from_seq(
-    state_seq: Sequence[int],
-    fsms_to_trans_finals: Dict[
-        int, Tuple[Set[Tuple[int, int]], Set[int], Dict[int, Set[int]]]
-    ],
-) -> Generator[Tuple[int, bool, bool], None, None]:
-    """Get the indices of the sub-FSMs in `fsm` that could have matched the state sequence `state_seq`.
-
-    Parameters
-    ----------
-    state_seq
-        A state sequence.
-    fsms_to_trans_finals
-        A map from FSM indices to tuples containing sets of their state transitions
-        and sets of the final/accept states.
-
-    Returns
-    -------
-    A generator returning tuples containing each sub-FSM index (in the order
-    they were union-ed to construct `fsm`) and booleans indicating whether or
-    not there is another valid transition from the last state in the sequence
-    for the associated sub-FSM (i.e. if the FSM can continue
-    accepting/matching) and whether or not the sequence ends in a final state
-    of the sub-FSM.
-    """
-    state_seq_transitions = set(zip(state_seq[:-1], state_seq[1:]))
-    last_fsm_state = state_seq[-1]
-    yield from (
-        (
-            # The sub-FMS index
-            fsm_idx,
-            # Is there another possible transition in this sub-FSM?
-            any(last_fsm_state == from_s for (from_s, to_s) in transitions),
-            # Is this sub-FSM in a final state?
-            state_seq[-1] in finals,
-        )
-        for fsm_idx, (transitions, finals, _) in fsms_to_trans_finals.items()
-        if state_seq_transitions.issubset(transitions)
-    )
-
-
-def walk_fsm(
-    fsm: BetterFSM,
-    token_transition_keys: Sequence[int],
-    start_state: int,
-    full_match: bool = True,
-) -> List[int]:
-    fsm_finals = fsm.finals
-
-    state = start_state
-    accepted_states: List[int] = []
-    last_final_idx: int = 0
-
-    fsm_transitions = fsm.flat_transition_map
-
-    # Iterate over token transition key sequence. The transition key
-    # sequence represents the FSM traversal rules of the tokens symbols.
-    for i, trans_key in enumerate(token_transition_keys):
-        new_state = fsm_transitions.get((state, trans_key))
-
-        if new_state is None:
-            if not full_match and last_final_idx > 0:
-                return accepted_states[:last_final_idx]
-
-            return []
-
-        state = new_state
-
-        if state in fsm_finals:
-            last_final_idx = i + 1
-
-        accepted_states.append(state)
-
-    if full_match and last_final_idx - 1 != i:
-        return []
-
-    return accepted_states
diff --git a/outlines/generator.py b/outlines/generator.py
index a7340bd1b..f2e669d8f 100644
--- a/outlines/generator.py
+++ b/outlines/generator.py
@@ -17,12 +17,11 @@
 from outlines.models.base import AsyncModel, Model
 from outlines.backends import (
     get_cfg_logits_processor,
-    get_fsm_logits_processor,
     get_json_schema_logits_processor,
     get_regex_logits_processor,
 )
 from outlines.backends.base import LogitsProcessorType
-from outlines.types import CFG, FSM, JsonSchema
+from outlines.types import CFG, JsonSchema
 from outlines.types.dsl import python_types_to_terms, to_regex
 
 
@@ -49,11 +48,6 @@ def __init__(self, model: BlackBoxModel, output_type: Optional[Any]):
         self.model = model
         self.output_type = output_type
 
-        if isinstance(self.output_type, FSM):
-            raise NotImplementedError(
-                "FSM generation is not supported for API-based models"
-            )
-
     def __call__(self, prompt: Any, **inference_kwargs) -> Any:
         """Generate a response from the model.
 
@@ -138,11 +132,6 @@ def __init__(self, model: AsyncBlackBoxModel, output_type: Optional[Any]):
         self.model = model
         self.output_type = output_type
 
-        if isinstance(self.output_type, FSM):
-            raise NotImplementedError(
-                "FSM generation is not supported for API-based models"
-            )
-
     async def __call__(self, prompt: Any, **inference_kwargs) -> Any:
         """Generate a response from the model.
 
@@ -253,12 +242,6 @@ def __init__(
                     model,
                     cfg_string,
                 )
-            elif isinstance(term, FSM):
-                self.logits_processor = get_fsm_logits_processor(
-                    backend_name,
-                    model,
-                    term.fsm,
-                )
             elif isinstance(term, JsonSchema):
                 self.logits_processor = get_json_schema_logits_processor(
                     backend_name,
diff --git a/outlines/release_note.md b/outlines/release_note.md
index 35812623b..5c25adf51 100644
--- a/outlines/release_note.md
+++ b/outlines/release_note.md
@@ -362,7 +362,7 @@ app = Application(template, Character)
 response = app(model, {"gender": "female"})
 ```
 
-- The regex DSL and the associated `Term` classes and functions have been added. Terms (`Regex`, `String`…) can be used as output types to generate text with models or generators (they are turned into a regex). The term functions (`either`, `optional`, `at_least`…) are useful to build more complex regex patterns by combining terms. On top of the objects related to regex patterns, there are also 3 terms that are intended to be used by themselves as output types: `JsonSchema`, `CFG` and `FSM`.
+- The regex DSL and the associated `Term` classes and functions have been added. Terms (`Regex`, `String`…) can be used as output types to generate text with models or generators (they are turned into a regex). The term functions (`either`, `optional`, `at_least`…) are useful to build more complex regex patterns by combining terms. On top of the objects related to regex patterns, there are also 2 terms that are intended to be used by themselves as output types: `JsonSchema` and `CFG`.
 [Documentation](https://dottxt-ai.github.io/outlines/latest/features/core/ouput_types)
 
 ```python
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 4fe8ef902..18e7c60d3 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -13,12 +13,10 @@
 from outlines.types.dsl import (
     Regex,
     CFG,
-    FSM,
     Choice,
     JsonSchema,
     regex,
     cfg,
-    fsm,
     json_schema,
     optional,
     either,
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 548c837ff..7b73aea07 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -26,7 +26,6 @@
     Union,
     get_args,
 )
-import interegular
 import jsonschema
 from genson import SchemaBuilder
 from outlines_core.fsm.json_schema import build_regex_from_schema
@@ -275,26 +274,6 @@ def from_file(cls, path: str) -> "CFG":
         return cls(definition)
 
 
-@dataclass
-class FSM(Term):
-    """Class representing a finite state machine.
-
-    Parameters
-    ----------
-    fsm
-        The finite state machine to store. This object must be an instance of
-        `interegular.fsm.FSM`.
-
-    """
-    fsm: interegular.fsm.FSM
-
-    def _display_node(self) -> str:
-        return f"FSM({self.fsm.__repr__()})"
-
-    def __repr__(self):
-        return f"FSM(fsm={self.fsm.__repr__()})"
-
-
 class JsonSchema(Term):
     """Class representing a JSON schema.
 
@@ -561,10 +540,6 @@ def cfg(definition: str):
     return CFG(definition)
 
 
-def fsm(fsm: interegular.fsm.FSM):
-    return FSM(fsm)
-
-
 def json_schema(schema: Union[str, dict, type[BaseModel]]):
     return JsonSchema(schema)
 
diff --git a/outlines/types/utils.py b/outlines/types/utils.py
index ffd6d80ab..d86961970 100644
--- a/outlines/types/utils.py
+++ b/outlines/types/utils.py
@@ -20,7 +20,6 @@
     get_origin,
 )
 
-import interegular
 from genson import SchemaBuilder
 from pydantic import BaseModel, create_model
 
@@ -149,10 +148,6 @@ def is_callable(value: Any) -> bool:
     return callable(value) and not isinstance(value, type)
 
 
-def is_interegular_fsm(value: Any) -> bool:
-    return isinstance(value, interegular.fsm.FSM)
-
-
 # Type conversion
 
 
diff --git a/pyproject.toml b/pyproject.toml
index 0b3fe4ec4..74c207e2a 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -24,7 +24,6 @@ classifiers = [
     "Topic :: Scientific/Engineering :: Artificial Intelligence",
 ]
 dependencies = [
-   "interegular",
    "jinja2",
    "lark",
    "nest_asyncio",
@@ -156,7 +155,6 @@ module = [
     "llama_cpp",
     "huggingface_hub",
     "lark.*",
-    "interegular.*",
     "datasets.*",
     "openai.*",
     "requests.*",
diff --git a/tests/backends/test_backends.py b/tests/backends/test_backends.py
index 68054922f..07246c7d3 100644
--- a/tests/backends/test_backends.py
+++ b/tests/backends/test_backends.py
@@ -1,4 +1,3 @@
-import interegular
 import outlines
 import pytest
 import transformers
@@ -8,7 +7,6 @@
     get_json_schema_logits_processor,
     get_regex_logits_processor,
     get_cfg_logits_processor,
-    get_fsm_logits_processor
 )
 from outlines.backends.outlines_core import OutlinesCoreBackend
 from outlines.backends.llguidance import (
@@ -71,10 +69,6 @@ def cfg_ebnf():
 answer ::= "yes" | "no"
 """
 
-@pytest.fixture
-def fsm():
-    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
-
 
 def test_get_backend(model):
     backend = _get_backend("outlines_core", model)
@@ -124,20 +118,3 @@ def test_get_cfg_logits_processor(model, cfg_lark, cfg_ebnf):
 
     processor = get_cfg_logits_processor("xgrammar", model, cfg_ebnf)
     assert isinstance(processor, XGrammarLogitsProcessor)
-
-
-def test_get_fsm_logits_processor(model, fsm):
-    processor = get_fsm_logits_processor("outlines_core", model, fsm)
-    assert isinstance(processor, GuideLogitsProcessor)
-
-    with pytest.raises(
-        NotImplementedError,
-        match="LLGuidanceBackend does not support FSM logits processors"
-    ):
-        get_fsm_logits_processor("llguidance", model, fsm)
-
-    with pytest.raises(
-        NotImplementedError,
-        match="XGrammarBackend does not support FSM logits processors",
-    ):
-        get_fsm_logits_processor("xgrammar", model, fsm)
diff --git a/tests/backends/test_llguidance.py b/tests/backends/test_llguidance.py
index a5de1ca5d..a005e1854 100644
--- a/tests/backends/test_llguidance.py
+++ b/tests/backends/test_llguidance.py
@@ -1,6 +1,5 @@
 import pytest
 
-import interegular
 import llama_cpp
 import llguidance.hf
 import numpy as np
@@ -97,10 +96,6 @@ def cfg_ebnf():
 answer ::= "yes" | "no"
 """
 
-@pytest.fixture
-def fsm():
-    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
-
 
 def test_llguidance_processor_torch(llg_grammar_spec, llg_tokenizer):
     processor = LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "torch")
@@ -148,7 +143,7 @@ def test_llguidance_processor_jax(llg_grammar_spec, llg_tokenizer):
     models.append((model_mlxlm(), "mlx"))
 
 @pytest.mark.parametrize("model, tensor_library_name", models)
-def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg_lark, cfg_ebnf, fsm):
+def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg_lark, cfg_ebnf):
     # initialization
     backend = LLGuidanceBackend(model)
     assert isinstance(backend.llg_tokenizer, LLTokenizer)
@@ -189,20 +184,3 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg_
     generator = outlines.Generator(model, backend="llguidance", processor=processor)
     response = generator("Hello, how are you?")
     assert response == "yes" or response == "no"
-
-    # fsm
-    with pytest.raises(
-        NotImplementedError,
-        match="LLGuidanceBackend does not support FSM logits processors",
-    ):
-        backend.get_fsm_logits_processor(fsm)
-
-    # multiple generations
-    processor = backend.get_regex_logits_processor(regex)
-    generator = outlines.Generator(model, backend="llguidance", processor=processor)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
diff --git a/tests/backends/test_outlines_core.py b/tests/backends/test_outlines_core.py
index 087db561e..8b6cff771 100644
--- a/tests/backends/test_outlines_core.py
+++ b/tests/backends/test_outlines_core.py
@@ -1,6 +1,5 @@
 import pytest
 
-import interegular
 import llama_cpp
 import transformers
 
@@ -72,10 +71,6 @@ def cfg():
 %ignore WS_INLINE
 """
 
-@pytest.fixture
-def fsm():
-    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
-
 
 models = [
     (model_transformers(), "torch"),
@@ -85,7 +80,7 @@ def fsm():
     models.append((model_mlxlm(), "mlx"))
 
 @pytest.mark.parametrize("model, tensor_library_name", models)
-def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg, fsm):
+def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg):
     # initialization
     backend = OutlinesCoreBackend(model)
     assert backend.tokenizer == model.tokenizer
@@ -113,21 +108,3 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg,
         match="Context-free grammar output type is not supported",
     ):
         backend.get_cfg_logits_processor(cfg)
-
-    # fsm
-    processor = backend.get_fsm_logits_processor(fsm)
-    assert isinstance(processor, GuideLogitsProcessor)
-    generator = outlines.Generator(model, backend="outlines_core", processor=processor)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
-
-    # multiple generations
-    processor = backend.get_regex_logits_processor(regex)
-    generator = outlines.Generator(model, backend="outlines_core", processor=processor)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
diff --git a/tests/backends/test_xgrammar.py b/tests/backends/test_xgrammar.py
index 89c9393be..1133b9f0f 100644
--- a/tests/backends/test_xgrammar.py
+++ b/tests/backends/test_xgrammar.py
@@ -1,6 +1,5 @@
 import pytest
 
-import interegular
 import llama_cpp
 import transformers
 from xgrammar import GrammarCompiler
@@ -58,12 +57,8 @@ def cfg():
 answer ::= "yes" | "no"
 """
 
-@pytest.fixture
-def fsm():
-    return interegular.parse_pattern(r"[0-9]{3}").to_fsm()
-
 
-def test_xgrammar_backend(model_transformers, json_schema, regex, cfg, fsm):
+def test_xgrammar_backend(model_transformers, json_schema, regex, cfg):
     # initialization
     backend = XGrammarBackend(model_transformers)
     assert isinstance(backend.grammar_compiler, GrammarCompiler)
@@ -91,23 +86,6 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg, fsm):
     response = generator("Hello, how are you?")
     assert response == "yes" or response == "no"
 
-    # fsm
-    with pytest.raises(
-        NotImplementedError,
-        match="XGrammarBackend does not support FSM logits processors",
-    ):
-        backend.get_fsm_logits_processor(fsm)
-
-    # multiple generations
-    processor = backend.get_regex_logits_processor(regex)
-    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
-
 
 def test_xgrammar_backend_invalid_model(model_llamacpp):
     with pytest.raises(
diff --git a/tests/fsm/partial_python.lark b/tests/fsm/partial_python.lark
deleted file mode 100644
index 973e59634..000000000
--- a/tests/fsm/partial_python.lark
+++ /dev/null
@@ -1,314 +0,0 @@
-// Python 3 grammar for Lark
-//
-// This grammar should parse all python 3.x code successfully.
-//
-// Adapted from: https://docs.python.org/3/reference/grammar.html
-//
-// This version is actually a subset of Lark's Python grammar without the
-// regex look-arounds in the string terminals.
-//
-// Start symbols for the grammar:
-//       single_input is a single interactive statement;
-//       file_input is a module or sequence of commands read from an input file;
-//       eval_input is the input for the eval() functions.
-// NB: compound_stmt in single_input is followed by extra NEWLINE!
-//
-
-single_input: _NEWLINE | simple_stmt | compound_stmt _NEWLINE
-file_input: (_NEWLINE | stmt)*
-eval_input: testlist _NEWLINE*
-
-decorator: "@" dotted_name [ "(" [arguments] ")" ] _NEWLINE
-decorators: decorator+
-decorated: decorators (classdef | funcdef | async_funcdef)
-
-async_funcdef: "async" funcdef
-funcdef: "def" name "(" [parameters] ")" ["->" test] ":" suite
-
-parameters: paramvalue ("," paramvalue)* ["," SLASH ("," paramvalue)*] ["," [starparams | kwparams]]
-          | starparams
-          | kwparams
-
-SLASH: "/" // Otherwise the it will completely disappear and it will be undisguisable in the result
-starparams: (starparam | starguard) poststarparams
-starparam: "*" typedparam
-starguard: "*"
-poststarparams: ("," paramvalue)* ["," kwparams]
-kwparams: "**" typedparam ","?
-
-?paramvalue: typedparam ("=" test)?
-?typedparam: name (":" test)?
-
-
-lambdef: "lambda" [lambda_params] ":" test
-lambdef_nocond: "lambda" [lambda_params] ":" test_nocond
-lambda_params: lambda_paramvalue ("," lambda_paramvalue)* ["," [lambda_starparams | lambda_kwparams]]
-          | lambda_starparams
-          | lambda_kwparams
-?lambda_paramvalue: name ("=" test)?
-lambda_starparams: "*" [name]  ("," lambda_paramvalue)* ["," [lambda_kwparams]]
-lambda_kwparams: "**" name ","?
-
-
-?stmt: simple_stmt | compound_stmt
-?simple_stmt: small_stmt (";" small_stmt)* [";"] _NEWLINE
-?small_stmt: (expr_stmt | assign_stmt | del_stmt | pass_stmt | flow_stmt | import_stmt | global_stmt | nonlocal_stmt | assert_stmt)
-expr_stmt: testlist_star_expr
-assign_stmt: annassign | augassign | assign
-
-annassign: testlist_star_expr ":" test ["=" test]
-assign: testlist_star_expr ("=" (yield_expr|testlist_star_expr))+
-augassign: testlist_star_expr augassign_op (yield_expr|testlist)
-!augassign_op: "+=" | "-=" | "*=" | "@=" | "/=" | "%=" | "&=" | "|=" | "^=" | "<<=" | ">>=" | "**=" | "//="
-?testlist_star_expr: test_or_star_expr
-                   | test_or_star_expr ("," test_or_star_expr)+ ","?  -> tuple
-                   | test_or_star_expr ","  -> tuple
-
-// For normal and annotated assignments, additional restrictions enforced by the interpreter
-del_stmt: "del" exprlist
-pass_stmt: "pass"
-?flow_stmt: break_stmt | continue_stmt | return_stmt | raise_stmt | yield_stmt
-break_stmt: "break"
-continue_stmt: "continue"
-return_stmt: "return" [testlist]
-yield_stmt: yield_expr
-raise_stmt: "raise" [test ["from" test]]
-import_stmt: import_name | import_from
-import_name: "import" dotted_as_names
-// note below: the ("." | "...") is necessary because "..." is tokenized as ELLIPSIS
-import_from: "from" (dots? dotted_name | dots) "import" ("*" | "(" import_as_names ")" | import_as_names)
-!dots: "."+
-import_as_name: name ["as" name]
-dotted_as_name: dotted_name ["as" name]
-import_as_names: import_as_name ("," import_as_name)* [","]
-dotted_as_names: dotted_as_name ("," dotted_as_name)*
-dotted_name: name ("." name)*
-global_stmt: "global" name ("," name)*
-nonlocal_stmt: "nonlocal" name ("," name)*
-assert_stmt: "assert" test ["," test]
-
-?compound_stmt: if_stmt | while_stmt | for_stmt | try_stmt | match_stmt
-              | with_stmt | funcdef | classdef | decorated | async_stmt
-async_stmt: "async" (funcdef | with_stmt | for_stmt)
-if_stmt: "if" test ":" suite elifs ["else" ":" suite]
-elifs: elif_*
-elif_: "elif" test ":" suite
-while_stmt: "while" test ":" suite ["else" ":" suite]
-for_stmt: "for" exprlist "in" testlist ":" suite ["else" ":" suite]
-try_stmt: "try" ":" suite except_clauses ["else" ":" suite] [finally]
-        | "try" ":" suite finally   -> try_finally
-finally: "finally" ":" suite
-except_clauses: except_clause+
-except_clause: "except" [test ["as" name]] ":" suite
-// NB compile.c makes sure that the default except clause is last
-
-
-with_stmt: "with" with_items ":" suite
-with_items: with_item ("," with_item)*
-with_item: test ["as" name]
-
-match_stmt: "match" test ":" _NEWLINE _INDENT case+ _DEDENT
-
-case: "case" pattern ["if" test] ":" suite
-
-?pattern: sequence_item_pattern "," _sequence_pattern -> sequence_pattern
-        | as_pattern
-?as_pattern: or_pattern ("as" NAME)?
-?or_pattern: closed_pattern ("|" closed_pattern)*
-?closed_pattern: literal_pattern
-               | NAME -> capture_pattern
-               | "_" -> any_pattern
-               | attr_pattern
-               | "(" as_pattern ")"
-               | "[" _sequence_pattern "]" -> sequence_pattern
-               | "(" (sequence_item_pattern "," _sequence_pattern)? ")" -> sequence_pattern
-               | "{" (mapping_item_pattern ("," mapping_item_pattern)* ","?)?"}" -> mapping_pattern
-               | "{" (mapping_item_pattern ("," mapping_item_pattern)* ",")? "**" NAME ","? "}" -> mapping_star_pattern
-               | class_pattern
-
-literal_pattern: inner_literal_pattern
-
-?inner_literal_pattern: "None" -> const_none
-                      | "True" -> const_true
-                      | "False" -> const_false
-                      | STRING -> string
-                      | number
-
-attr_pattern: NAME ("." NAME)+ -> value
-
-name_or_attr_pattern: NAME ("." NAME)* -> value
-
-mapping_item_pattern: (literal_pattern|attr_pattern) ":" as_pattern
-
-_sequence_pattern: (sequence_item_pattern ("," sequence_item_pattern)* ","?)?
-?sequence_item_pattern: as_pattern
-                      | "*" NAME -> star_pattern
-
-class_pattern: name_or_attr_pattern "(" [arguments_pattern ","?] ")"
-arguments_pattern: pos_arg_pattern ["," keyws_arg_pattern]
-                 | keyws_arg_pattern -> no_pos_arguments
-
-pos_arg_pattern: as_pattern ("," as_pattern)*
-keyws_arg_pattern: keyw_arg_pattern ("," keyw_arg_pattern)*
-keyw_arg_pattern: NAME "=" as_pattern
-
-
-
-suite: simple_stmt | _NEWLINE _INDENT stmt+ _DEDENT
-
-?test: or_test ("if" or_test "else" test)?
-     | lambdef
-     | assign_expr
-
-assign_expr: name ":=" test
-
-?test_nocond: or_test | lambdef_nocond
-
-?or_test: and_test ("or" and_test)*
-?and_test: not_test_ ("and" not_test_)*
-?not_test_: "not" not_test_ -> not_test
-         | comparison
-?comparison: expr (comp_op expr)*
-star_expr: "*" expr
-
-?expr: or_expr
-?or_expr: xor_expr ("|" xor_expr)*
-?xor_expr: and_expr ("^" and_expr)*
-?and_expr: shift_expr ("&" shift_expr)*
-?shift_expr: arith_expr (_shift_op arith_expr)*
-?arith_expr: term (_add_op term)*
-?term: factor (_mul_op factor)*
-?factor: _unary_op factor | power
-
-!_unary_op: "+"|"-"|"~"
-!_add_op: "+"|"-"
-!_shift_op: "<<"|">>"
-!_mul_op: "*"|"@"|"/"|"%"|"//"
-// <> isn't actually a valid comparison operator in Python. It's here for the
-// sake of a __future__ import described in PEP 401 (which really works :-)
-!comp_op: "<"|">"|"=="|">="|"<="|"<>"|"!="|"in"|"not" "in"|"is"|"is" "not"
-
-?power: await_expr ("**" factor)?
-?await_expr: AWAIT? atom_expr
-AWAIT: "await"
-
-?atom_expr: atom_expr "(" [arguments] ")"      -> funccall
-          | atom_expr "[" subscriptlist "]"  -> getitem
-          | atom_expr "." name               -> getattr
-          | atom
-
-?atom: "(" yield_expr ")"
-     | "(" _tuple_inner? ")" -> tuple
-     | "(" comprehension{test_or_star_expr} ")" -> tuple_comprehension
-     | "[" _exprlist? "]"  -> list
-     | "[" comprehension{test_or_star_expr} "]"  -> list_comprehension
-     | "{" _dict_exprlist? "}" -> dict
-     | "{" comprehension{key_value} "}" -> dict_comprehension
-     | "{" _exprlist "}" -> set
-     | "{" comprehension{test} "}" -> set_comprehension
-     | name -> var
-     | number
-     | string_concat
-     | "(" test ")"
-     | "..." -> ellipsis
-     | "None"    -> const_none
-     | "True"    -> const_true
-     | "False"   -> const_false
-
-
-?string_concat: string+
-
-_tuple_inner: test_or_star_expr (("," test_or_star_expr)+ [","] | ",")
-
-?test_or_star_expr: test
-                 | star_expr
-
-?subscriptlist: subscript
-              | subscript (("," subscript)+ [","] | ",") -> subscript_tuple
-?subscript: test | ([test] ":" [test] [sliceop]) -> slice
-sliceop: ":" [test]
-?exprlist: (expr|star_expr)
-         | (expr|star_expr) (("," (expr|star_expr))+ [","]|",")
-?testlist: test | testlist_tuple
-testlist_tuple: test (("," test)+ [","] | ",")
-_dict_exprlist: (key_value | "**" expr) ("," (key_value | "**" expr))* [","]
-
-key_value: test ":"  test
-
-_exprlist: test_or_star_expr (","  test_or_star_expr)* [","]
-
-classdef: "class" name ["(" [arguments] ")"] ":" suite
-
-
-
-arguments: argvalue ("," argvalue)*  ("," [ starargs | kwargs])?
-         | starargs
-         | kwargs
-         | comprehension{test}
-
-starargs: stararg ("," stararg)* ("," argvalue)* ["," kwargs]
-stararg: "*" test
-kwargs: "**" test ("," argvalue)*
-
-?argvalue: test ("=" test)?
-
-
-comprehension{comp_result}: comp_result comp_fors [comp_if]
-comp_fors: comp_for+
-comp_for: [ASYNC] "for" exprlist "in" or_test
-ASYNC: "async"
-?comp_if: "if" test_nocond
-
-// not used in grammar, but may appear in "node" passed from Parser to Compiler
-encoding_decl: name
-
-yield_expr: "yield" [testlist]
-          | "yield" "from" test -> yield_from
-
-number: DEC_NUMBER | HEX_NUMBER | BIN_NUMBER | OCT_NUMBER | FLOAT_NUMBER | IMAG_NUMBER
-string: STRING // | LONG_STRING
-
-// Other terminals
-
-_NEWLINE: ( /\r?\n[\t ]*/ | COMMENT )+
-
-%ignore /[\t \f]+/  // WS
-%ignore /\\[\t \f]*\r?\n/   // LINE_CONT
-%ignore COMMENT
-%declare _INDENT _DEDENT
-
-
-// Python terminals
-
-!name: NAME | "match" | "case"
-NAME: /[^\W\d]\w*/
-COMMENT: /#[^\n]*/
-
-// We only need a usable approximation for something like this until we fully
-// implement look-arounds, or use a regex/FSM library that supports them.
-// STRING: /([ubf]?r?|r[ubf])("(?!"").*?(?<!\\)(\\\\)*?"|'(?!'').*?(?<!\\)(\\\\)*?')/i
-// Also, we're not dealing with this case now.
-// LONG_STRING: /([ubf]?r?|r[ubf])(""".*?(?<!\\)(\\\\)*?"""|'''.*?(?<!\\)(\\\\)*?''')/is
-
-STRING: SINGLE_QUOTE_STRING | DOUBLE_QUOTE_STRING
-DOUBLE_QUOTE_STRING: /([ubf]?r?|r[ubf])("([^\\"]|.)*?")/i
-SINGLE_QUOTE_STRING: /([ubf]?r?|r[ubf])('([^\\']|.)*?')/i
-
-
-_SPECIAL_DEC: "0".."9"        ("_"?  "0".."9"                       )*
-DEC_NUMBER:   "1".."9"        ("_"?  "0".."9"                       )*
-// Again, no look-arounds yet
-//           |   "0"             ("_"?  "0"                            )* /(?![1-9])/
-HEX_NUMBER.2: "0" ("x" | "X") ("_"? ("0".."9" | "a".."f" | "A".."F"))+
-OCT_NUMBER.2: "0" ("o" | "O") ("_"?  "0".."7"                       )+
-BIN_NUMBER.2: "0" ("b" | "B") ("_"?  "0".."1"                       )+
-
-_EXP: ("e"|"E") ["+" | "-"] _SPECIAL_DEC
-DECIMAL: "." _SPECIAL_DEC | _SPECIAL_DEC "." _SPECIAL_DEC?
-FLOAT_NUMBER.2: _SPECIAL_DEC _EXP | DECIMAL _EXP?
-IMAG_NUMBER.2: (_SPECIAL_DEC      | FLOAT_NUMBER) ("J" | "j")
-
-
-// Comma-separated list (with an optional trailing comma)
-cs_list{item}: item ("," item)* ","?
-_cs_list{item}: item ("," item)* ","?
diff --git a/tests/fsm/test_parsing.py b/tests/fsm/test_parsing.py
deleted file mode 100644
index a47943f3c..000000000
--- a/tests/fsm/test_parsing.py
+++ /dev/null
@@ -1,307 +0,0 @@
-import importlib
-from copy import copy
-
-import lark.lark
-import pytest
-from lark.indenter import DedentError
-from lark.lexer import UnexpectedCharacters, UnexpectedToken
-
-from outlines.fsm.parsing import PartialLark, PartialPythonIndenter
-
-
-@pytest.fixture
-def cleanup_lark_import():
-    yield
-    # Clean up lark.lark.LarkOptions._defaults
-    importlib.reload(lark.lark)
-
-
-def test_partial_parsing(cleanup_lark_import):
-    lp = PartialLark.open_from_package(
-        "tests",
-        "partial_python.lark",
-        ["fsm"],
-        parser="lalr",
-        postlex=PartialPythonIndenter(),
-        start="file_input",
-        deterministic=True,
-    )
-
-    # End with a potentially unfinished NAME
-    parser_state = lp.parse("x")
-    assert parser_state.state_stack == [0]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-    assert last_token.value.fsm_state_seq == (0, 15)
-    assert last_token.value.is_not_finished is True
-    assert not parser_state.value_stack
-
-    # End with an ignored token
-    parser_state = lp.parse("x ")
-    assert parser_state.state_stack == [0, 692]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-    assert last_token.value.fsm_state_seq == (0, 1)
-    assert last_token.value.is_not_finished is True
-    assert not parser_state.value_stack
-
-    # Could be a complete `=` or the start of a `==`
-    parser_state = lp.parse("x =")
-    assert parser_state.state_stack == [0, 692]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-    assert any(
-        term_info.terminal_name == "EQUAL"
-        for term_info in last_token.value.terminals_and_info
-    )
-    assert not parser_state.value_stack
-
-    parser_state = lp.parse("x = '")
-    assert parser_state.state_stack == [0, 58, 59]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-    assert last_token.value.fsm_state_seq == (0, 6)
-    assert last_token.value.is_not_finished is True
-    assert not parser_state.value_stack
-
-    parser_state = lp.parse("x = 'hi")
-    assert parser_state.state_stack == [0, 58, 59]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-    assert last_token.value.fsm_state_seq == (0, 6, 6, 6)
-    assert last_token.value.is_not_finished is True
-    assert not parser_state.value_stack
-
-    parser_state = lp.parse("x = ('hi")
-    assert parser_state.state_stack == [0, 58, 59, 254]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-    assert last_token.value.fsm_state_seq == (0, 6, 6, 6)
-    assert last_token.value.is_not_finished is True
-    assert not parser_state.value_stack
-
-    parser_state = lp.parse("def")
-    assert parser_state.state_stack == [0]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-    assert last_token.value.fsm_state_seq == (0, 26, 99, 100)
-    assert last_token.value.is_not_finished is True
-    assert not parser_state.value_stack
-
-    # Now, try something incremental
-    last_lexer_state = parser_state.lexer.state
-    last_lexer_state.text += " blah()"
-    lp.parse_from_state(parser_state, is_end=False)
-    last_token = parser_state.lexer.state.last_token
-    assert not parser_state.value_stack
-
-    last_lexer_state = parser_state.lexer.state
-    last_valid_token = last_lexer_state.last_token
-    assert last_valid_token.type == "RPAR"
-    assert not parser_state.value_stack
-
-    # Something incremental and a little more complicated
-    parser_state = lp.parse("x = 1\ndef foo(x):\n  ")
-    assert parser_state.state_stack == [0, 94, 600, 601, 602, 607, 608, 269]
-    last_lexer_state = parser_state.lexer.state
-    last_lexer_state.text += "  return x"
-
-    lp.parse_from_state(parser_state, is_end=False)
-    assert parser_state.state_stack == [
-        0,
-        94,
-        600,
-        601,
-        602,
-        607,
-        608,
-        269,
-        764,
-        95,
-        305,
-    ]
-    last_token = parser_state.lexer.state.last_token
-    assert last_token.type == "partial"
-
-    with pytest.raises(UnexpectedToken):
-        lp.parse("def \n")
-
-    with pytest.raises(UnexpectedToken):
-        lp.parse("def hot no")
-
-    lp = PartialLark.open_from_package(
-        "tests",
-        "partial_python.lark",
-        ["fsm"],
-        parser="lalr",
-        postlex=PartialPythonIndenter(),
-        start="file_input",
-        use_value_stack=True,
-    )
-    parser_state = lp.parse("x = ('hi")
-    lp.parse_from_state(parser_state, is_end=False)
-    assert len(parser_state.state_stack) == 4
-    assert parser_state.value_stack[-1].type == "LPAR"
-
-
-def test_sequential_parse_example(cleanup_lark_import):
-    input_tokens = [
-        "x ",
-        "= ",
-        "1",
-        "\nde",
-        "f ",
-        "foo(",
-        "x)",
-        ":\n",
-        "  ",
-        "  return x",
-        " + 1",
-        "\n",
-        "z ",
-        "= ",
-        "foo(",
-        '"hi',
-        '")\n',
-    ]
-    vocab = sorted(set(input_tokens))
-
-    lp = PartialLark.open_from_package(
-        "tests",
-        "partial_python.lark",
-        ["fsm"],
-        parser="lalr",
-        postlex=PartialPythonIndenter(),
-        start="file_input",
-        deterministic=True,
-    )
-    parser_state = lp.parse("")
-
-    token_seq = ""
-    for i, token in enumerate(input_tokens):
-        token_seq += token
-
-        lex_state = parser_state.lexer.state
-        lex_state.text = token_seq
-
-        lp.parse_from_state(parser_state, is_end=False)
-
-        next_vocab = set()
-        for test_token in vocab:
-            ps = copy(parser_state)
-            ls = ps.lexer.state
-            ls.text = token_seq + test_token
-
-            if i + 1 < len(input_tokens) and test_token == input_tokens[i + 1]:
-                lp.parse_from_state(ps, is_end=False)
-                next_vocab.add(test_token)
-            else:
-                try:
-                    lp.parse_from_state(ps, is_end=False)
-                    next_vocab.add(test_token)
-                except (EOFError, UnexpectedToken, UnexpectedCharacters, DedentError):
-                    pass
-
-        if i + 1 == len(input_tokens):
-            assert all(tk in next_vocab for tk in ["\n", "\nde", "  ", " + 1"])
-
-
-# TODO: Remove once fsm_union and walk_fsm are implemented in Outlines-Core
-import interegular  # noqa
-
-from outlines.fsm.parsing import fsm_union, walk_fsm  # noqa
-
-
-def test_outlines_interegular_union_consistency():
-    fsm0 = interegular.parse_pattern(r"abc").to_fsm()
-    fsm1 = interegular.parse_pattern(r"WXYZ").to_fsm()
-    fsm2 = interegular.parse_pattern(r"12345").to_fsm()
-
-    interegular_unioned_fsm = fsm0 | fsm1 | fsm2
-    outlines_unioned_fsm, _ = fsm_union([fsm0, fsm1, fsm2])
-
-    assert list(outlines_unioned_fsm.strings()) == list(
-        interegular_unioned_fsm.strings()
-    )
-
-
-def _reconstruct_fsms(fsm, fsms_to_trans_finals):
-    """Reconstruct the original fsms for testing purposes"""
-    reconstructed_fsms = []
-    for transitions, finals, state_map in fsms_to_trans_finals.values():
-        inv_state_map = {new: orig for orig, news in state_map.items() for new in news}
-        states = set(inv_state_map.values())
-        initial = inv_state_map.get(fsm.initial) or next(
-            (orig for orig, news in state_map.items() if fsm.initial in news), None
-        )
-        finals = {inv_state_map[s] for s in finals}
-
-        transition_map = {}
-        alphabet = {}
-        for trans_id, (from_state, to_state) in enumerate(transitions):
-            orig_from, orig_to = inv_state_map[from_state], inv_state_map[to_state]
-            # Collect symbols associated with the transition
-            symbols = {
-                symbol
-                for trans, dest in fsm.map.get(from_state, {}).items()
-                if dest == to_state
-                for symbol in fsm.alphabet.by_transition.get(trans, [])
-            }
-            if symbols:
-                # NOTE: THIS RECONSTRUCTOR DOESN'T WORK FOR MORE THAN ONE TRANSITION PER SYMBOL
-                assert len(symbols) == 1
-                symbol = list(symbols)[0]
-                alphabet[symbol] = trans_id
-                transition_map.setdefault(orig_from, {})[trans_id] = orig_to
-
-        reconstructed_fsms.append(
-            interegular.fsm.FSM(
-                alphabet=interegular.fsm.Alphabet(alphabet),
-                states=frozenset(states),
-                initial=initial,
-                finals=frozenset(finals),
-                map=transition_map,
-                __no_validation__=True,
-            )
-        )
-    return reconstructed_fsms
-
-
-def test_fsm_to_trans_finals_reconstruction():
-    """Assert that _fsms_to_trans_finals is correct by reconstructing original fsms"""
-    fsm0 = interegular.parse_pattern(r"abc").to_fsm()
-    fsm1 = interegular.parse_pattern(r"XYZ").to_fsm()
-    fsm2 = interegular.parse_pattern(r"12345").to_fsm()
-
-    fsm, _fsms_to_trans_finals = fsm_union([fsm0, fsm1, fsm2])
-
-    reconstructed = _reconstruct_fsms(fsm, _fsms_to_trans_finals)
-
-    # assert reconstruction equivalent
-    assert list(fsm0.strings()) == list(reconstructed[0].strings())
-    assert list(fsm1.strings()) == list(reconstructed[1].strings())
-    assert list(fsm2.strings()) == list(reconstructed[2].strings())
-
-
-def test_walk_fsm():
-    fsm = interegular.parse_pattern(r"abc*d").to_fsm()
-    # convert to BetterFSM
-    fsm = fsm_union([fsm])[0]
-
-    # if match, produce equivalent number of states, assert state can terminate
-    transitions = [fsm.alphabet[letter] for letter in "abcccd"]
-    accepted_states = walk_fsm(fsm, transitions, fsm.initial, full_match=True)
-    assert len(accepted_states) == len(transitions)
-    assert accepted_states[-1] in fsm.finals
-
-    # if no match, assert empty
-    accepted_states = walk_fsm(
-        fsm, [fsm.alphabet[letter] for letter in "b"], fsm.initial, full_match=True
-    )
-    assert accepted_states == []
-
-    # if full_match, but last state not present, assert empty
-    accepted_states = walk_fsm(
-        fsm, [fsm.alphabet[letter] for letter in "abc"], fsm.initial, full_match=True
-    )
-    assert accepted_states == []
diff --git a/tests/test_generator.py b/tests/test_generator.py
index d75d03b0d..2c26e98d0 100644
--- a/tests/test_generator.py
+++ b/tests/test_generator.py
@@ -1,7 +1,6 @@
 import pytest
 from typing import AsyncGenerator, Generator as TypingGenerator, Literal
 
-import interegular
 import transformers
 
 import outlines
@@ -16,7 +15,7 @@
     OutlinesLogitsProcessor,
     RegexLogitsProcessor,
 )
-from outlines.types import CFG, FSM
+from outlines.types import CFG
 from tests.test_utils.mock_openai_client import (
     MockAsyncOpenAIClient,
     MockOpenAIClient,
@@ -106,15 +105,6 @@ def test_steerable_generator_init_cfg_output_type(steerable_model):
     assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
 
 
-def test_steerable_generator_init_fsm_output_type(steerable_model):
-    generator = SteerableGenerator(
-        steerable_model,
-        FSM(interegular.parse_pattern(r"abc").to_fsm())
-    )
-    assert generator.model == steerable_model
-    assert isinstance(generator.logits_processor, OutlinesLogitsProcessor)
-
-
 def test_steerable_generator_init_other_output_type(steerable_model):
     generator = SteerableGenerator(steerable_model, Literal["foo", "bar"])
     assert generator.model == steerable_model
@@ -143,17 +133,12 @@ def test_steerable_generator_stream(steerable_model):
 # BlackBoxGenerator
 
 
-def test_black_box_generator_init_valid(black_box_sync_model):
+def test_black_box_generator_init(black_box_sync_model):
     generator = BlackBoxGenerator(black_box_sync_model, Literal["foo", "bar"])
     assert generator.model == black_box_sync_model
     assert generator.output_type == Literal["foo", "bar"]
 
 
-def test_black_box_generator_init_invalid(black_box_sync_model):
-    with pytest.raises(NotImplementedError):
-        BlackBoxGenerator(black_box_sync_model, FSM("foo"))
-
-
 def test_black_box_generator_call(black_box_sync_model):
     generator = BlackBoxGenerator(black_box_sync_model, str)
     result = generator("Write a very short sentence", max_tokens=10)
@@ -171,17 +156,12 @@ def test_black_box_generator_stream(black_box_sync_model):
 # AsyncBlackBoxGenerator
 
 
-def test_async_black_box_generator_init_valid(black_box_async_model):
+def test_async_black_box_generator_init(black_box_async_model):
     generator = AsyncBlackBoxGenerator(black_box_async_model, Literal["foo", "bar"])
     assert generator.model == black_box_async_model
     assert generator.output_type == Literal["foo", "bar"]
 
 
-def test_async_black_box_generator_init_invalid(black_box_async_model):
-    with pytest.raises(NotImplementedError):
-        AsyncBlackBoxGenerator(black_box_async_model, FSM("foo"))
-
-
 @pytest.mark.asyncio
 async def test_async_black_box_generator_call(black_box_async_model):
     generator = AsyncBlackBoxGenerator(black_box_async_model, str)
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index 0c4db0a44..e4175e221 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -14,7 +14,6 @@
     Optional as PyOptional
 )
 
-import interegular
 import pytest
 from genson import SchemaBuilder
 from pydantic import BaseModel
@@ -37,7 +36,6 @@
     Term,
     either,
     CFG,
-    FSM,
     _handle_dict,
     _handle_list,
     _handle_literal,
@@ -402,15 +400,6 @@ def test_cfg():
     assert not cfg == "a"
 
 
-def test_fsm():
-    interegular_fsm = interegular.parse_pattern(r"a").to_fsm()
-    fsm = types.fsm(interegular_fsm)
-    assert isinstance(fsm, FSM)
-    assert fsm.fsm == interegular_fsm
-    assert fsm._display_node() == f"FSM({fsm.fsm.__repr__()})"
-    assert fsm.__repr__() == f"FSM(fsm={fsm.fsm.__repr__()})"
-
-
 def test_json_schema():
     # init dict
     schema = types.json_schema({"type": "string"})
@@ -568,9 +557,6 @@ class PydanticModel(BaseModel):
         }
     )
 
-    interegular_fsm = interegular.parse_pattern(r"abc").to_fsm()
-    assert python_types_to_terms(types.fsm(interegular_fsm)).fsm is interegular_fsm
-
     def func(a: int, b: str):
         return (a, b)
 
diff --git a/tests/types/test_types_utils.py b/tests/types/test_types_utils.py
index 57f41a099..60296dc08 100644
--- a/tests/types/test_types_utils.py
+++ b/tests/types/test_types_utils.py
@@ -16,7 +16,6 @@
     Union
 )
 
-import interegular
 from genson import SchemaBuilder
 from pydantic import BaseModel
 
@@ -37,7 +36,6 @@
     is_genson_schema_builder,
     is_int,
     is_int_instance,
-    is_interegular_fsm,
     is_literal,
     is_native_dict,
     is_pydantic_model,
@@ -360,12 +358,6 @@ def test_is_callable(sample_function, sample_class, sample_dataclass, sample_typ
     assert not is_callable(sample_pydantic_model)
 
 
-def test_is_interegular_fsm():
-    fsm = interegular.parse_pattern(r"[0-9]+").to_fsm()
-    assert is_interegular_fsm(fsm)
-    assert not is_interegular_fsm(dict)
-
-
 # Type conversion
 
 

From 947c904608821ef6e33493181f87653e25b08862 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Tue, 29 Jul 2025 21:55:46 +0200
Subject: [PATCH 419/505] Upgrade outlines_core to 0.2.11 and update its logits
 processor/backend

---
 .github/workflows/tests.yml                   |   7 +-
 examples/parsing.py                           | 105 --------
 outlines/backends/llguidance.py               |   4 +-
 outlines/backends/outlines_core.py            | 235 +++++++++++++++--
 outlines/processors/__init__.py               |   9 +-
 outlines/processors/guide.py                  | 159 ------------
 outlines/processors/structured.py             | 209 ---------------
 outlines/types/dsl.py                         |   5 +-
 pyproject.toml                                |  11 +-
 scripts/reproduce_1383.py                     |   5 +-
 tests/backends/test_backends.py               |  15 +-
 tests/backends/test_outlines_core.py          |  30 ++-
 tests/models/test_llamacpp.py                 |  11 +-
 tests/models/test_llamacpp_type_adapter.py    |  24 +-
 tests/models/test_mlxlm.py                    |   2 +-
 tests/models/test_transformers.py             |   8 +-
 ...st_transformers_multimodal_type_adapter.py |  20 +-
 .../models/test_transformers_type_adapter.py  |  22 +-
 tests/processors/test_guide.py                | 240 ------------------
 tests/processors/test_structured.py           | 108 --------
 tests/test_cache.py                           |   2 -
 tests/test_generator.py                       |  18 +-
 22 files changed, 301 insertions(+), 948 deletions(-)
 delete mode 100644 examples/parsing.py
 delete mode 100644 outlines/processors/guide.py
 delete mode 100644 outlines/processors/structured.py
 delete mode 100644 tests/processors/test_guide.py
 delete mode 100644 tests/processors/test_structured.py

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index c6e37d684..ce96e223d 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -51,9 +51,9 @@ jobs:
         echo "::set-output name=id::$MATRIX_ID"
     - name: Run tests
       run: |
-        uv run pytest -x --cov=outlines -m 'not api_call'
-      env:
-        COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}
+        rm -f .coverage*
+        uv run coverage erase
+        uv run python -m coverage run --branch --parallel-mode -m pytest -x -m 'not api_call'
     - name: Upload coverage data
       uses: actions/upload-artifact@v4
       with:
@@ -61,7 +61,6 @@ jobs:
         path: .coverage.*
         if-no-files-found: ignore
         include-hidden-files: true
-        # TODO FIXME: This is only using the last run
         overwrite: true
 
   coverage:
diff --git a/examples/parsing.py b/examples/parsing.py
deleted file mode 100644
index a579b71dd..000000000
--- a/examples/parsing.py
+++ /dev/null
@@ -1,105 +0,0 @@
-"""An example illustrating parser-based masking."""
-
-import math
-import time
-from copy import copy
-
-import torch
-from lark.indenter import DedentError
-from lark.lexer import UnexpectedCharacters, UnexpectedToken
-from transformers import (
-    AutoModelForCausalLM,
-    AutoTokenizer,
-    LogitsProcessor,
-    LogitsProcessorList,
-    set_seed,
-)
-
-
-revision = None
-checkpoint = "Salesforce/codegen-350M-mono"
-device = "cuda"
-
-tokenizer = AutoTokenizer.from_pretrained(checkpoint)
-
-model = AutoModelForCausalLM.from_pretrained(
-    checkpoint, trust_remote_code=True, revision=revision
-).to(device)
-
-parser = PartialLark.open_from_package(
-    "tests",
-    "partial_python.lark",
-    ["text"],
-    parser="lalr",
-    postlex=PartialPythonIndenter(),
-    start="file_input",
-)
-
-class ParserLogitsProcessor(LogitsProcessor):
-    """Bias invalid token scores according to a running parse state."""
-
-    def __init__(self, parser):
-        self.parser = parser
-        self.parser_state = parser.parse("")
-        self.states_stack = [self.parser_state]
-        self.token_seq = None
-        self.token_idx = 0
-
-    def __call__(
-        self, input_ids: torch.LongTensor, scores: torch.FloatTensor
-    ) -> torch.FloatTensor:
-        if self.token_seq is None:
-            self.token_seq = tokenizer.decode(input_ids[0])
-            self.token_idx = len(input_ids[0]) - 1
-        else:
-            self.token_idx += 1
-            self.token_seq += tokenizer.decode(input_ids[0][self.token_idx])
-
-        # Process the last sampled token
-        lex_state = self.parser_state.lexer.state
-        lex_state.text = self.token_seq
-
-        self.parser.parse_from_state(self.parser_state, is_end=False)
-
-        print(f'parsed:"{self.token_seq}"')
-
-        mask = torch.full_like(scores, -math.inf)
-
-        # Determine which tokens in the vocabulary are valid next tokens
-        # given the parser state.
-        #
-        # TODO: This is a very naive and slow approach.  It could be done in
-        # parallel, easily memoized/cached, etc., but there are a few other
-        # approaches to try first that will dramatically reduce the
-        # amount of work needed here.
-        t0 = time.perf_counter()
-        for test_token, token_id in tokenizer.vocab.items():
-            ps = copy(self.parser_state)
-            ls = ps.lexer.state
-            ls.text = self.token_seq + tokenizer.convert_tokens_to_string([test_token])
-
-            try:
-                self.parser.parse_from_state(ps, is_end=False)
-                mask[0][token_id] = 0
-            except (EOFError, UnexpectedToken, UnexpectedCharacters, DedentError):
-                pass
-
-        print(f"next token masking duration: {time.perf_counter() - t0}")
-
-        return scores + mask
-
-
-set_seed(20399)
-
-input_text = "def "
-inputs = tokenizer.encode(input_text, return_tensors="pt").to(device)
-
-outputs = model.generate(
-    inputs,
-    max_length=100,
-    temperature=0.1,
-    logits_processor=LogitsProcessorList([ParserLogitsProcessor(parser)]),
-    renormalize_logits=True,
-)
-
-print(tokenizer.decode(outputs[0]))
diff --git a/outlines/backends/llguidance.py b/outlines/backends/llguidance.py
index c5a738616..2f9ea59f5 100644
--- a/outlines/backends/llguidance.py
+++ b/outlines/backends/llguidance.py
@@ -3,8 +3,6 @@
 import warnings
 from typing import TYPE_CHECKING
 
-from llguidance import LLMatcher
-
 from outlines.backends.base import BaseBackend
 from outlines.models import LlamaCpp, MLXLM, SteerableModel, Transformers
 from outlines.processors.base_logits_processor import (
@@ -65,6 +63,8 @@ def _setup(self, batch_size: int) -> None:
             The batch size of the input
 
         """
+        from llguidance import LLMatcher
+
         self.ll_matchers = [
             LLMatcher(self.llg_tokenizer, self.grammar)
             for _ in range(batch_size)
diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index 172991709..c87549263 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -1,13 +1,175 @@
 """Backend class for Outlines Core."""
 
+from typing import Dict
+
+from outlines_core import Guide, Index, Vocabulary
+# TODO: change this once the import issue is fixed in outlines_core
+from outlines_core import outlines_core
+
 from outlines.backends.base import BaseBackend
 from outlines.models import SteerableModel
-from outlines.processors import (
-    GuideLogitsProcessor,
-    RegexLogitsProcessor,
+from outlines.models.llamacpp import LlamaCpp
+from outlines.models.mlxlm import MLXLM
+from outlines.models.transformers import Transformers
+from outlines.processors.base_logits_processor import (
+    OutlinesLogitsProcessor,
+    TensorType
 )
-from outlines.processors.guide import RegexGuide
-from outlines_core.fsm.json_schema import build_regex_from_schema
+
+
+class OutlinesCoreLogitsProcessor(OutlinesLogitsProcessor):
+    """Logits processor for Outlines Core."""
+
+    def __init__(
+        self, index: Index, tensor_library_name: str
+    ):
+        """
+        Parameters
+        ----------
+        index: Index
+            The Outlines Core `Index` instance to use to create the Outlines
+            Core `Guide` instances that will be used to bias the logits
+        tensor_library_name: str
+            The tensor library name to use for the logits processor.
+
+        """
+        self.index = index
+        self.tensor_library_name = tensor_library_name
+        self.is_first_token = True
+        super().__init__(tensor_library_name)
+
+    def reset(self) -> None:
+        """Reset the logits processor."""
+        self.is_first_token = True
+
+    def _setup(self, batch_size: int, vocab_size: int) -> None:
+        """Set the guides, bitmasks and some functions used in the
+        `process_logits` method.
+
+        This method is called when the first token is generated instead of
+        at initialization because we need to know the batch size.
+
+        """
+        if self.tensor_library_name == "torch":
+            from outlines_core.kernels.torch import allocate_token_bitmask
+
+            self.allocate_token_bitmask = allocate_token_bitmask
+            self.bias_logits = self._bias_logits_torch
+
+        elif self.tensor_library_name == "numpy":
+            from outlines_core.kernels.numpy import allocate_token_bitmask
+
+            self.allocate_token_bitmask = allocate_token_bitmask
+            self.bias_logits = self._bias_logits_numpy
+
+        elif self.tensor_library_name == "mlx":
+            from outlines_core.kernels.mlx import (
+                allocate_token_bitmask
+            )
+
+            self.allocate_token_bitmask = allocate_token_bitmask
+            self.bias_logits = self._bias_logits_mlx
+
+        else:
+            raise ValueError(
+                f"Unsupported tensor library: {self.tensor_library_name}"
+            )
+
+        self._guides = [Guide(self.index) for _ in range(batch_size)]
+        self._bitmasks = [
+            self.allocate_token_bitmask(vocab_size)
+            for _ in range(batch_size)
+        ]
+
+    def _bias_logits_mlx( # pragma: no cover
+        self, batch_size: int, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for MLX tensors."""
+        from outlines_core.kernels.mlx import (
+            apply_token_bitmask,
+            fill_next_token_bitmask
+        )
+
+        biased_logits_array = []
+        for i in range(batch_size):
+            fill_next_token_bitmask(self._guides[i], self._bitmasks[i])
+            biased_logits = apply_token_bitmask(
+                self.tensor_adapter.unsqueeze(logits[i]), self._bitmasks[i] # type: ignore
+            )
+            biased_logits_array.append(biased_logits)
+
+        return self.tensor_adapter.concatenate(biased_logits_array)
+
+    def _bias_logits_torch(
+        self, batch_size: int, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for Torch tensors."""
+        from outlines_core.kernels.torch import (
+            apply_token_bitmask_inplace,
+            fill_next_token_bitmask
+        )
+
+        for i in range(batch_size):
+            fill_next_token_bitmask(self._guides[i], self._bitmasks[i])
+            apply_token_bitmask_inplace(
+                self.tensor_adapter.unsqueeze(logits[i]), # type: ignore
+                self._bitmasks[i]
+            )
+
+        return logits
+
+    def _bias_logits_numpy(
+        self, batch_size: int, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for Numpy tensors."""
+        from outlines_core.kernels.numpy import (
+            apply_token_bitmask_inplace,
+            fill_next_token_bitmask
+        )
+
+        for i in range(batch_size):
+            fill_next_token_bitmask(self._guides[i], self._bitmasks[i])
+            apply_token_bitmask_inplace(
+                self.tensor_adapter.unsqueeze(logits[i]), # type: ignore
+                self._bitmasks[i]
+            )
+
+        return logits
+
+    def process_logits(
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Use the guides to bias the logits.
+
+        Parameters
+        ----------
+        input_ids
+            The ids of the tokens of the existing sequences.
+        logits
+            The logits for the current generation step.
+
+        Returns
+        -------
+        TensorType
+            The biased logits.
+
+        """
+        batch_size = self.tensor_adapter.shape(input_ids)[0]
+        vocab_size = self.tensor_adapter.shape(logits)[1]
+
+        if self.is_first_token:
+            self._setup(batch_size, vocab_size)
+            self.is_first_token = False
+        else:
+            for i in range(batch_size):
+                last_token_id = self.tensor_adapter.to_scalar(input_ids[i][-1]) # type: ignore
+                if not self._guides[i].is_finished():
+                    self._guides[i].advance(
+                        token_id=last_token_id,
+                        return_tokens=False
+                    )
+
+        return self.bias_logits(batch_size, logits)
 
 
 class OutlinesCoreBackend(BaseBackend):
@@ -21,12 +183,30 @@ def __init__(self, model: SteerableModel):
             The Outlines model of the user.
 
         """
-        self.tokenizer = model.tokenizer
+        if isinstance(model, Transformers):
+            vocabulary = model.hf_tokenizer.get_vocab()
+            eos_token_id = model.hf_tokenizer.eos_token_id
+            eos_token = model.hf_tokenizer.eos_token
+        elif isinstance(model, LlamaCpp):
+            vocabulary = model.tokenizer.vocabulary
+            eos_token_id = model.tokenizer.eos_token_id
+            eos_token = model.tokenizer.eos_token
+        elif isinstance(model, MLXLM):
+            vocabulary = model.mlx_tokenizer._tokenizer.get_vocab()
+            eos_token_id = model.mlx_tokenizer._tokenizer.eos_token_id
+            eos_token = model.mlx_tokenizer._tokenizer.eos_token
+        else:
+            raise ValueError(f"Unsupported model type: {type(model)}")
+
+        self.eos_token_id = eos_token_id
+        self.vocabulary = self.create_outlines_core_vocabulary(
+            vocabulary, eos_token_id, eos_token
+        )
         self.tensor_library_name = model.tensor_library_name
 
     def get_json_schema_logits_processor(
         self, json_schema: str
-    ) -> RegexLogitsProcessor:
+    ):
         """Create a logits processor from a JSON schema.
 
         Parameters
@@ -40,10 +220,10 @@ def get_json_schema_logits_processor(
             The logits processor to use to constrain the generation.
 
         """
-        regex = build_regex_from_schema(json_schema)
+        regex = outlines_core.json_schema.build_regex_from_schema(json_schema)
         return self.get_regex_logits_processor(regex)
 
-    def get_regex_logits_processor(self, regex: str) -> RegexLogitsProcessor:
+    def get_regex_logits_processor(self, regex: str):
         """Create a logits processor from a regex.
 
         Parameters
@@ -57,14 +237,37 @@ def get_regex_logits_processor(self, regex: str) -> RegexLogitsProcessor:
             The logits processor to use to constrain the generation.
 
         """
-        return RegexLogitsProcessor(
-            regex,
-            self.tokenizer,
-            self.tensor_library_name,
-        )
+        index = Index(regex, self.vocabulary)
+        return OutlinesCoreLogitsProcessor(index, self.tensor_library_name)
 
     def get_cfg_logits_processor(self, grammar):
         raise NotImplementedError(
-            "Context-free grammar output type is not supported for the "
-            + "Outlines Core backend."
+            "Outlines Core does not support context-free grammar."
         )
+
+    @staticmethod
+    def create_outlines_core_vocabulary(
+        vocab: Dict[str, int], eos_token_id: int, eos_token: str
+    ) -> Vocabulary:
+        """Create an Outlines Core Vocabulary instance.
+
+        Parameters
+        ----------
+        vocab: Dict[str, int]
+            The vocabulary to create an Outlines Core vocabulary from.
+        eos_token_id: int
+            The EOS token ID.
+        eos_token: str
+            The EOS token.
+
+        Returns
+        -------
+        Vocabulary
+            The Outlines Core Vocabulary instance.
+
+        """
+        formatted_vocab = {}
+        for token, token_id in vocab.items():
+            formatted_vocab[token] = [token_id]
+        formatted_vocab.pop(eos_token)
+        return Vocabulary(eos_token_id, formatted_vocab)
diff --git a/outlines/processors/__init__.py b/outlines/processors/__init__.py
index b2c2208a1..d1997335c 100644
--- a/outlines/processors/__init__.py
+++ b/outlines/processors/__init__.py
@@ -1,8 +1,3 @@
-"""Processors and guides to control generation in steerable models."""
+"""Processors to control generation in steerable models."""
 
-from .structured import (
-    GuideLogitsProcessor,
-    JSONLogitsProcessor,
-    OutlinesLogitsProcessor,
-    RegexLogitsProcessor,
-)
+from .base_logits_processor import OutlinesLogitsProcessor
diff --git a/outlines/processors/guide.py b/outlines/processors/guide.py
deleted file mode 100644
index 9c4a3067a..000000000
--- a/outlines/processors/guide.py
+++ /dev/null
@@ -1,159 +0,0 @@
-"""Guides to control generation in steerable models.
-
-Logits processors rely on guides to control the generation process.
-"""
-
-from typing import TYPE_CHECKING, Any, Union
-
-from outlines_core.fsm.guide import Generate
-from outlines_core.fsm.guide import Guide as CoreGuide
-from outlines_core.fsm.guide import RegexGuide as CoreRegexGuide
-from outlines_core.fsm.guide import Write
-from outlines_core.fsm.guide import (
-    create_states_mapping as uncached_create_states_mapping,
-)
-
-from outlines.caching import cache
-
-if TYPE_CHECKING:
-    from outlines.models.tokenizer import Tokenizer
-
-
-Instruction = Union[Write, Generate]
-
-
-class Guide(CoreGuide):
-    """Base definition of a generation guide.
-
-    A generation guide defines the behavior of a finite-state machine that
-    guides a text generation procedure. Unlike the DFAs built from regular
-    expressions guides, it can also emit a `Write` instructions which tells
-    the model that it can append a sequence of tokens (or token word) instead
-    of generating it.
-
-    """
-    initial_state: Any
-
-
-class StopAtEOSGuide(Guide):
-    """Guide to generate tokens until the EOS token has been generated."""
-    final_state = 1
-    initial_state = 0
-
-    def __init__(self, tokenizer: "Tokenizer"):
-        """
-        Parameters
-        ----------
-        tokenizer
-            The tokenizer used to convert tokens to ids.
-
-        """
-        self.eos_token_id = tokenizer.eos_token_id
-        self.vocabulary = tokenizer.vocabulary.values()
-
-    def get_next_instruction(self, state: int) -> Instruction:
-        """Return the next instruction.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        Instruction
-            An `Instruction` instance.
-
-        """
-        if self.is_final_state(state):
-            return Write([self.eos_token_id])
-        return Generate(None)
-
-    def get_next_state(self, state: int, token_id: int) -> int:
-        """Return the next state.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-        token_id
-            The id of the token that was just generated.
-
-        Returns
-        -------
-        int
-            The next state.
-
-        """
-        if token_id == self.eos_token_id or state == self.final_state:
-            return self.final_state
-
-        return self.initial_state
-
-    def is_final_state(self, state: int) -> bool:
-        """Return whether the given state is a final state.
-
-        Parameters
-        ----------
-        state
-            The guide's current state.
-
-        Returns
-        -------
-        bool
-            Whether the given state is a final state.
-
-        """
-        return state == self.final_state
-
-    def copy(self) -> "StopAtEOSGuide":
-        """Return itself as there is no need to copy."""
-        return self
-
-
-@cache()
-def cached_create_states_mapping(regex_string, tokenizer, *args, **kwargs):
-    """Wrap the uncached create_states_mapping function in a cache."""
-    return uncached_create_states_mapping(
-        regex_string, tokenizer, *args, **kwargs
-    )
-
-
-class RegexGuide(CoreRegexGuide):
-    """Guide to generate text in the language of a regular expression.
-
-    This class is a wrapper around the CoreRegexGuide class that adds a cache
-    to the create_states_mapping function.
-
-    """
-
-    @classmethod
-    def from_regex(
-        cls,
-        regex_string: str,
-        tokenizer,
-        **kwargs,
-    ):
-        """Create a RegexGuide from a regular expression.
-
-        Parameters
-        ----------
-        regex_string
-            The regular expression to generate text from.
-        tokenizer
-            The tokenizer to use to convert tokens to ids.
-        kwargs
-            Additional keyword arguments to pass to the CoreRegexGuide constructor.
-
-        Returns
-        -------
-        RegexGuide
-            A RegexGuide instance.
-
-        """
-        return super().from_regex(
-            regex_string,
-            tokenizer,
-            _create_states_mapping=cached_create_states_mapping,
-            **kwargs,
-        )
diff --git a/outlines/processors/structured.py b/outlines/processors/structured.py
deleted file mode 100644
index 0059ec131..000000000
--- a/outlines/processors/structured.py
+++ /dev/null
@@ -1,209 +0,0 @@
-"""Logits processors for structured generation.
- _______________________________
-/ Don't want to self-host?       \
-\\ Try .json at http://dottxt.co /
- -------------------------------
-       \\   ^__^
-        \\  (oo)\\_______
-            (__)\\       )\\/\
-                ||----w |
-                ||     ||
-
-Copyright 2024- the Outlines developers
-
-Licensed under the Apache License, Version 2.0 (the "License");
-you may not use this file except in compliance with the License.
-You may obtain a copy of the License at
-
-    http://www.apache.org/licenses/LICENSE-2.0
-
-Unless required by applicable law or agreed to in writing, software
-distributed under the License is distributed on an "AS IS" BASIS,
-WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-See the License for the specific language governing permissions and
-limitations under the License.
-"""
-
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Type, Union
-
-from pydantic import BaseModel
-from outlines_core.fsm.json_schema import build_regex_from_schema
-
-from outlines.processors.base_logits_processor import (
-    OutlinesLogitsProcessor,
-    TensorType
-)
-from outlines.processors.guide import (
-    Guide,
-    RegexGuide
-)
-from outlines.types import JsonSchema
-
-if TYPE_CHECKING:
-    from outlines.models.tokenizer import Tokenizer
-
-
-class GuideLogitsProcessor(OutlinesLogitsProcessor):
-    """Bias generation using a guide.
-
-    Attributes
-    ----------
-    tokenizer
-        The outlines tokenizer used to convert tokens to ids.
-    guide
-        The outlines guide used to bias the logits.
-    """
-    tokenizer: "Tokenizer"
-    guide: Guide
-    _guide_states: Dict[int, Any]
-    _seq_start_idx: Optional[int]
-
-    def __init__(
-        self, tokenizer: "Tokenizer", guide: Guide, tensor_library_name: str
-    ):
-        """
-        Parameters
-        ----------
-        tokenizer
-            The tokenizer used to convert tokens to ids.
-        guide
-            The `outlines.processors.guide.Guide` that is used to bias the
-            logits.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
-
-        """
-        super().__init__(tensor_library_name=tensor_library_name)
-        self.tokenizer = tokenizer
-        self.guide = guide
-        self._guide_states = {hash(tuple([])): self.guide.initial_state}
-        self._seq_start_idx = None
-
-    def process_logits(
-        self, input_ids: TensorType, logits: TensorType
-    ) -> TensorType:
-        """Use the Guide to bias the logits before sampling the next token.
-
-        Parameters
-        ----------
-        input_ids
-            The ids of the tokens of the existing sequences.
-        logits
-            The logits for the current generation step.
-
-        Returns
-        -------
-        TensorType
-            The biased logits.
-
-        """
-        if self._seq_start_idx is None:
-            self._seq_start_idx = len(input_ids[0]) # type: ignore
-
-        sequence_states: List[int] = []  # vector of states corresponding to `input_ids`
-
-        for seq_ids in input_ids: # type: ignore
-            gen_ids = seq_ids[self._seq_start_idx :]
-            curr_state_key = hash(tuple(self.tensor_adapter.to_list(gen_ids)))
-
-            if curr_state_key not in self._guide_states:
-                prev_state = self._guide_states[hash(tuple(self.tensor_adapter.to_list(gen_ids[:-1])))]
-                curr_state = self.guide.get_next_state(prev_state, self.tensor_adapter.to_scalar(gen_ids[-1]))
-                self._guide_states[curr_state_key] = curr_state
-
-            sequence_states.append(self._guide_states[curr_state_key])
-
-        allowed_tokens_batch = []
-        batch_indices = []
-        for i, guide_state in enumerate(sequence_states):
-            allowed_tokens = self.guide.get_next_instruction(guide_state).tokens
-            allowed_tokens_batch.append(allowed_tokens)
-            batch_indices.append(
-                self.tensor_adapter.full_like(allowed_tokens, i)
-            )  # Store batch index for each allowed token
-
-        device = self.tensor_adapter.get_device(logits)
-        allowed_tokens_concat = self.tensor_adapter.to_device(
-            self.tensor_adapter.concatenate(allowed_tokens_batch),
-            device
-        )
-        batch_indices_concat = self.tensor_adapter.to_device(
-            self.tensor_adapter.concatenate(batch_indices),
-            device
-        )
-
-        mask = self.tensor_adapter.boolean_ones_like(logits)
-        mask[batch_indices_concat, allowed_tokens_concat] = False
-        logits = self.tensor_adapter.apply_mask(logits, mask, float("-inf"))
-
-        return logits
-
-    def copy(self) -> "GuideLogitsProcessor":
-        """Return a copy of the logits processor."""
-        return GuideLogitsProcessor(
-            tokenizer=self.tokenizer,
-            guide=self.guide.copy(),
-            tensor_library_name=self.tensor_adapter.library_name
-        )
-
-
-class RegexLogitsProcessor(GuideLogitsProcessor):
-    """Bias generation based on a regular expression."""
-    guide: RegexGuide
-
-    def __init__(
-        self,
-        regex_string: str,
-        tokenizer: "Tokenizer",
-        tensor_library_name: str,
-    ):
-        """
-        Parameters
-        ----------
-        regex_string
-            A string that represents a regular expression.
-        tokenizer
-            An Outlines tokenizer.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
-
-        """
-        # Build a guide from the regex string and then pass it to the
-        # GuideLogitsProcessor superclass.
-        guide = RegexGuide.from_regex(regex_string, tokenizer)
-        super().__init__(tokenizer=tokenizer, guide=guide, tensor_library_name=tensor_library_name)
-
-
-class JSONLogitsProcessor(RegexLogitsProcessor):
-    """Bias generation based on a JSON schema."""
-    def __init__(
-        self,
-        schema: Union[dict, Type[BaseModel], str],
-        tokenizer: "Tokenizer",
-        tensor_library_name: str,
-        whitespace_pattern: Optional[str] = None,
-    ):
-        """
-        Parameters
-        ----------
-        schema
-            A JSON schema that encodes the structure we want the model to generate.
-        tokenizer
-            The tokenizer used to convert tokens to ids.
-        tensor_library_name
-            The name of the library to use to manipulate the tensors.
-        whitespace_pattern
-            Pattern to use for JSON syntactic whitespace (doesn't impact string
-            literals). For example, to allow only a single space or newline with
-            `whitespace_pattern=r"[\n ]?"`.
-
-        """
-        # Convert the JSON schema into a regex string and then pass it to the
-        # RegexLogitsProcessor superclass.
-        schema_str = JsonSchema(schema).schema
-        regex_string = build_regex_from_schema(schema_str, whitespace_pattern)
-        super().__init__(
-            regex_string=regex_string,
-            tokenizer=tokenizer,
-            tensor_library_name=tensor_library_name
-        )
diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 7b73aea07..5a12a9b68 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -28,7 +28,8 @@
 )
 import jsonschema
 from genson import SchemaBuilder
-from outlines_core.fsm.json_schema import build_regex_from_schema
+# TODO: change this once the import issue is fixed in outlines_core
+from outlines_core import outlines_core
 from pydantic import (
     BaseModel,
     GetCoreSchemaHandler,
@@ -806,7 +807,7 @@ def to_regex(term: Term) -> str:
     elif isinstance(term, Regex):
         return f"({term.pattern})"
     elif isinstance(term, JsonSchema):
-        regex_str = build_regex_from_schema(term.schema, term.whitespace_pattern)
+        regex_str = outlines_core.json_schema.build_regex_from_schema(term.schema, term.whitespace_pattern)
         return f"({regex_str})"
     elif isinstance(term, Choice):
         regexes = [to_regex(python_types_to_terms(item)) for item in term.items]
diff --git a/pyproject.toml b/pyproject.toml
index 74c207e2a..af55d5984 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -25,8 +25,6 @@ classifiers = [
 ]
 dependencies = [
    "jinja2",
-   "lark",
-   "nest_asyncio",
    "cloudpickle",
    "diskcache",
    "pydantic>=2.0",
@@ -34,13 +32,11 @@ dependencies = [
    "jsonschema",
    "requests",
    "pillow",
-   "tqdm",
    "typing_extensions",
    "iso3166",
    "airportsdata",
-   "outlines_core==0.1.26",
+   "outlines_core==0.2.11",
    "genson",
-   "torch",
    "jsonpath_ng",
 ]
 dynamic = ["version"]
@@ -49,7 +45,7 @@ dynamic = ["version"]
 anthropic = ["anthropic"]
 dottxt = ["dottxt"]
 gemini = ["google-genai"]
-llamacpp = ["huggingface-hub", "llama-cpp-python"]
+llamacpp = ["huggingface-hub", "llama-cpp-python", "numba"]
 mlxlm = ["datasets", "mlx", "mlx-lm"]
 ollama = ["ollama"]
 openai = ["openai"]
@@ -83,6 +79,7 @@ test = [
     "jax",
     "flax",
     "numpy>=2.0.0,<2.2.0",
+    "numba",
     "torch",
     "tensorflow",
     "tf-keras",
@@ -142,7 +139,6 @@ module = [
     "mamba_ssm.*",
     "mlx_lm.*",
     "mlx.*",
-    "nest_asyncio",
     "numpy.*",
     "cloudpickle.*",
     "diskcache.*",
@@ -154,7 +150,6 @@ module = [
     "transformers.*",
     "llama_cpp",
     "huggingface_hub",
-    "lark.*",
     "datasets.*",
     "openai.*",
     "requests.*",
diff --git a/scripts/reproduce_1383.py b/scripts/reproduce_1383.py
index 8a1500a4e..af4e36dcf 100644
--- a/scripts/reproduce_1383.py
+++ b/scripts/reproduce_1383.py
@@ -10,7 +10,8 @@
 import json
 from typing import Optional
 from pydantic import BaseModel
-from outlines_core.fsm.outlines_core_rs import build_regex_from_schema
+# TODO: change this once the import issue is fixed in outlines_core
+from outlines_core import outlines_core
 from outlines.types import JsonSchema
 
 # Test case 1: Simple optional field (this should work)
@@ -49,7 +50,7 @@ def test_schema(schema_dict, name):
     try:
         # Try to create regex from schema
         schema_str = json.dumps(schema_dict)
-        regex = build_regex_from_schema(schema_str)
+        regex = outlines_core.json_schema.build_regex_from_schema(schema_str)
         print(f"✓ Success! Generated regex: {regex[:100]}...")
     except Exception as e:
         print(f"✗ Error: {type(e).__name__}: {e}")
diff --git a/tests/backends/test_backends.py b/tests/backends/test_backends.py
index 07246c7d3..73173b0d1 100644
--- a/tests/backends/test_backends.py
+++ b/tests/backends/test_backends.py
@@ -8,16 +8,15 @@
     get_regex_logits_processor,
     get_cfg_logits_processor,
 )
-from outlines.backends.outlines_core import OutlinesCoreBackend
+from outlines.backends.outlines_core import (
+    OutlinesCoreBackend,
+    OutlinesCoreLogitsProcessor,
+)
 from outlines.backends.llguidance import (
     LLGuidanceBackend,
     LLGuidanceLogitsProcessor
 )
 from outlines.backends.xgrammar import XGrammarBackend, XGrammarLogitsProcessor
-from outlines.processors.structured import (
-    GuideLogitsProcessor,
-    RegexLogitsProcessor,
-)
 
 
 @pytest.fixture
@@ -86,7 +85,7 @@ def test_get_backend(model):
 
 def test_get_json_schema_logits_processor(model, json_schema):
     processor = get_json_schema_logits_processor("outlines_core", model, json_schema)
-    assert isinstance(processor, RegexLogitsProcessor)
+    assert isinstance(processor, OutlinesCoreLogitsProcessor)
 
     processor = get_json_schema_logits_processor("llguidance", model, json_schema)
     assert isinstance(processor, LLGuidanceLogitsProcessor)
@@ -97,7 +96,7 @@ def test_get_json_schema_logits_processor(model, json_schema):
 
 def test_get_regex_logits_processor(model, regex):
     processor = get_regex_logits_processor("outlines_core", model, regex)
-    assert isinstance(processor, RegexLogitsProcessor)
+    assert isinstance(processor, OutlinesCoreLogitsProcessor)
 
     processor = get_regex_logits_processor("llguidance", model, regex)
     assert isinstance(processor, LLGuidanceLogitsProcessor)
@@ -109,7 +108,7 @@ def test_get_regex_logits_processor(model, regex):
 def test_get_cfg_logits_processor(model, cfg_lark, cfg_ebnf):
     with pytest.raises(
         NotImplementedError,
-        match="Context-free grammar output type is not supported"
+        match="Outlines Core does not support context-free grammar."
     ):
         get_cfg_logits_processor("outlines_core", model, cfg_lark)
 
diff --git a/tests/backends/test_outlines_core.py b/tests/backends/test_outlines_core.py
index 8b6cff771..e62951022 100644
--- a/tests/backends/test_outlines_core.py
+++ b/tests/backends/test_outlines_core.py
@@ -1,15 +1,16 @@
 import pytest
 
 import llama_cpp
+from outlines_core import Vocabulary
 import transformers
 
 import outlines
-from outlines.backends.outlines_core import OutlinesCoreBackend
-from outlines.processors.structured import (
-    RegexLogitsProcessor,
-    GuideLogitsProcessor,
+from outlines.backends.outlines_core import (
+    OutlinesCoreBackend,
+    OutlinesCoreLogitsProcessor
 )
 
+
 try:
     import mlx_lm
     HAS_MLX = True
@@ -80,23 +81,22 @@ def cfg():
     models.append((model_mlxlm(), "mlx"))
 
 @pytest.mark.parametrize("model, tensor_library_name", models)
-def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg):
+def test_outlines_core_backend(model, tensor_library_name, json_schema, regex, cfg):
     # initialization
     backend = OutlinesCoreBackend(model)
-    assert backend.tokenizer == model.tokenizer
+    assert isinstance(backend.vocabulary, Vocabulary)
     assert backend.tensor_library_name == tensor_library_name
 
     # json schema
     processor = backend.get_json_schema_logits_processor(json_schema)
-    assert isinstance(processor, RegexLogitsProcessor)
+    assert isinstance(processor, OutlinesCoreLogitsProcessor)
     generator = outlines.Generator(model, backend="outlines_core", processor=processor)
     response = generator("Hello, how are you?")
-    assert response[0] == "{"
     assert "name" in response
 
     # regex
     processor = backend.get_regex_logits_processor(regex)
-    assert isinstance(processor, RegexLogitsProcessor)
+    assert isinstance(processor, OutlinesCoreLogitsProcessor)
     generator = outlines.Generator(model, backend="outlines_core", processor=processor)
     response = generator("Hello, how are you?")
     assert len(response) == 3
@@ -105,6 +105,16 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg)
     # cfg
     with pytest.raises(
         NotImplementedError,
-        match="Context-free grammar output type is not supported",
+        match="Outlines Core does not support context-free grammar.",
     ):
         backend.get_cfg_logits_processor(cfg)
+
+    # multiple generations
+    processor = backend.get_regex_logits_processor(regex)
+    generator = outlines.Generator(model, backend="outlines_core", processor=processor)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
+    response = generator("Hello, how are you?")
+    assert len(response) == 3
+    assert int(response)
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 76f028d8f..9b881b7de 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -12,7 +12,6 @@
     LlamaCppTypeAdapter,
     from_llamacpp
 )
-from outlines.processors import RegexLogitsProcessor
 from outlines.types.dsl import Regex, CFG
 
 
@@ -90,10 +89,10 @@ def test_llamacpp_chat(model):
 
 
 def test_llamacpp_regex(model):
-    regex_str = Regex(r"[0-9]").pattern
-    logits_processor = RegexLogitsProcessor(regex_str, model.tokenizer, model.tensor_library_name)
-    result = model.generate("Respond with one word. Not more.", logits_processor)
+    result = model("Respond with one word. Not more.", Regex(r"[0-9]"))
     assert isinstance(result, str)
+    assert int(result)
+    assert len(result) == 1
 
 
 def test_llamacpp_json(model):
@@ -122,7 +121,7 @@ def test_llamacpp_cfg(model, ebnf_grammar):
 def test_llamacpp_cfg_outlines_core(model, lark_grammar):
     with pytest.raises(
         NotImplementedError,
-        match="not supported for LlamaCpp with the outlines-core backend"
+        match="Outlines Core does not support context-free grammar."
     ):
         model(
             "Respond with one word. Not more.",
@@ -186,7 +185,7 @@ def test_llamacpp_stream_cfg(model, ebnf_grammar):
 def test_llamacpp_stream_cfg_outlines_core(model, lark_grammar):
     with pytest.raises(
         NotImplementedError,
-        match="not supported for LlamaCpp with the outlines-core backend"
+        match="Outlines Core does not support context-free grammar."
     ):
         for chunk in model.stream(
             "Respond with one word. Not more.",
diff --git a/tests/models/test_llamacpp_type_adapter.py b/tests/models/test_llamacpp_type_adapter.py
index a7649d523..403b3589c 100644
--- a/tests/models/test_llamacpp_type_adapter.py
+++ b/tests/models/test_llamacpp_type_adapter.py
@@ -1,13 +1,13 @@
 import pytest
 import io
 
-import llama_cpp
-from PIL import Image as PILImage
 from llama_cpp import LogitsProcessorList
+from PIL import Image as PILImage
+from outlines_core import Index, Vocabulary
 
+from outlines.backends.outlines_core import OutlinesCoreLogitsProcessor
 from outlines.inputs import Chat, Image
-from outlines.models.llamacpp import LlamaCppTypeAdapter, LlamaCppTokenizer
-from outlines.processors.structured import RegexLogitsProcessor
+from outlines.models.llamacpp import LlamaCppTypeAdapter
 
 
 @pytest.fixture
@@ -17,15 +17,9 @@ def adapter():
 
 @pytest.fixture
 def logits_processor():
-    model = llama_cpp.Llama.from_pretrained(
-        "M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-        filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-    )
-    return RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=LlamaCppTokenizer(model),
-        tensor_library_name="numpy",
-    )
+    vocabulary = Vocabulary.from_pretrained("openai-community/gpt2")
+    index = Index(r"[0-9]{3}", vocabulary)
+    return OutlinesCoreLogitsProcessor(index, "numpy")
 
 
 @pytest.fixture
@@ -71,5 +65,5 @@ def test_llamacpp_type_adapter_format_input(adapter, image):
 def test_llamacpp_type_adapter_format_output_type(adapter, logits_processor):
     formatted = adapter.format_output_type(logits_processor)
     assert isinstance(formatted, LogitsProcessorList)
-    assert len(formatted) == 1
-    assert isinstance(formatted[0], RegexLogitsProcessor)
+    assert formatted[0].index == logits_processor.index
+    assert formatted[0].tensor_library_name == logits_processor.tensor_library_name
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index c92236e6b..fc2ec6f39 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -92,7 +92,7 @@ def test_mlxlm_regex(model):
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_mlxlm_JsonType(model):
+def test_mlxlm_json_schema(model):
     class Character(BaseModel):
         name: str
 
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 336d9e25d..a7b0996c1 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -161,7 +161,7 @@ def test_transformers_multiple_samples(model):
     result = model("Respond with one word. Not more.")
     assert isinstance(result, str)
     result = model(
-        "Respond with one word. Not more.", num_return_sequences=2, num_beams=2
+        "Respond with one word. Not more.", num_return_sequences=2, do_sample=True
     )
     assert isinstance(result, list)
     assert len(result) == 2
@@ -177,7 +177,7 @@ def test_transformers_batch(model):
     result = model.batch(
         ["Respond with one word. Not more.", "Respond with one word. Not more."],
         num_return_sequences=2,
-        num_beams=2,
+        do_sample=True,
     )
     assert isinstance(result, list)
     assert len(result) == 2
@@ -206,7 +206,7 @@ class Foo(Enum):
         cat = "cat"
         dog = "dog"
 
-    result = model("Cat or dog?", Foo, num_return_sequences=2, num_beams=2)
+    result = model("Cat or dog?", Foo, num_return_sequences=2, do_sample=True)
     assert isinstance(result, list)
     assert len(result) == 2
     assert result[0] in ["cat", "dog"]
@@ -231,7 +231,7 @@ class Foo(Enum):
         ["Cat or dog?", "Cat or dog?"],
         Foo,
         num_return_sequences=2,
-        num_beams=2,
+        do_sample=True,
     )
     assert isinstance(result, list)
     assert len(result) == 2
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index 58be8060f..10df1b0e9 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -1,13 +1,12 @@
 import pytest
 
-import transformers
 from PIL import Image as PILImage
+from outlines_core import Index, Vocabulary
 from transformers import AutoTokenizer, LogitsProcessorList
 
-import outlines
 from outlines.inputs import Chat, Image, Video
 from outlines.models.transformers import TransformersMultiModalTypeAdapter
-from outlines.processors.structured import RegexLogitsProcessor
+from outlines.backends.outlines_core import OutlinesCoreLogitsProcessor
 
 
 MODEL_NAME = "erwanf/gpt2-mini"
@@ -25,15 +24,9 @@ def adapter():
 
 @pytest.fixture
 def logits_processor():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=model.tokenizer,
-        tensor_library_name="torch",
-    )
+    vocabulary = Vocabulary.from_pretrained("openai-community/gpt2")
+    index = Index(r"[0-9]{3}", vocabulary)
+    return OutlinesCoreLogitsProcessor(index, "torch")
 
 
 @pytest.fixture
@@ -87,7 +80,8 @@ def test_transformers_multimodal_type_adapter_format_output_type(
     formatted = adapter.format_output_type(logits_processor)
     assert isinstance(formatted, LogitsProcessorList)
     assert len(formatted) == 1
-    assert isinstance(formatted[0], RegexLogitsProcessor)
+    assert formatted[0].index == logits_processor.index
+    assert formatted[0].tensor_library_name == logits_processor.tensor_library_name
 
     formatted = adapter.format_output_type(None)
     assert formatted is None
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
index e82a42e8e..da7bbfcd2 100644
--- a/tests/models/test_transformers_type_adapter.py
+++ b/tests/models/test_transformers_type_adapter.py
@@ -2,13 +2,13 @@
 import pytest
 
 import transformers
-from PIL import Image as PILImage
 from transformers import LogitsProcessorList
+from outlines_core import Index, Vocabulary
+from PIL import Image as PILImage
 
-import outlines
+from outlines.backends.outlines_core import OutlinesCoreLogitsProcessor
 from outlines.inputs import Chat, Image
 from outlines.models.transformers import TransformersTypeAdapter
-from outlines.processors.structured import RegexLogitsProcessor
 
 
 MODEL_NAME = "erwanf/gpt2-mini"
@@ -25,15 +25,9 @@ def adapter():
 
 @pytest.fixture
 def logits_processor():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained(MODEL_NAME),
-        transformers.AutoTokenizer.from_pretrained(MODEL_NAME),
-    )
-    return RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=model.tokenizer,
-        tensor_library_name="torch",
-    )
+    vocabulary = Vocabulary.from_pretrained("openai-community/gpt2")
+    index = Index(r"[0-9]{3}", vocabulary)
+    return OutlinesCoreLogitsProcessor(index, "torch")
 
 @pytest.fixture
 def image():
@@ -72,8 +66,8 @@ def test_transformers_type_adapter_format_output_type(
 ):
     formatted = adapter.format_output_type(logits_processor)
     assert isinstance(formatted, LogitsProcessorList)
-    assert len(formatted) == 1
-    assert isinstance(formatted[0], RegexLogitsProcessor)
+    assert formatted[0].index == logits_processor.index
+    assert formatted[0].tensor_library_name == logits_processor.tensor_library_name
 
     formatted = adapter.format_output_type(None)
     assert formatted is None
diff --git a/tests/processors/test_guide.py b/tests/processors/test_guide.py
deleted file mode 100644
index ee7f568bb..000000000
--- a/tests/processors/test_guide.py
+++ /dev/null
@@ -1,240 +0,0 @@
-import pytest
-
-import llama_cpp
-import transformers
-
-import outlines
-from outlines.processors.guide import Generate, RegexGuide, StopAtEOSGuide, Write
-from outlines import caching
-
-try:
-    import mlx_lm
-    HAS_MLX = True
-except ImportError:
-    HAS_MLX = False
-
-try:
-    import vllm
-    HAS_VLLM = True
-except ImportError:
-    HAS_VLLM = False
-
-
-def assert_expected_tensor_ids(tensor, ids):
-    assert len(tensor) == len(ids)
-    norm_tensor = sorted(map(int, tensor))
-    norm_ids = sorted(map(int, tensor))
-    assert norm_tensor == norm_ids, (norm_tensor, norm_ids)
-
-
-def test_stop_at_eos():
-    class MockTokenizer:
-        vocabulary = {"a": 1, "eos": 2}
-        eos_token_id = 2
-
-    fsm = StopAtEOSGuide(MockTokenizer())
-
-    instruction = fsm.get_next_instruction(fsm.initial_state)
-    assert isinstance(instruction, Generate)
-    assert instruction.tokens is None
-
-    instruction = fsm.get_next_instruction(fsm.final_state)
-    assert isinstance(instruction, Write)
-    assert instruction.tokens == [2]
-
-    assert fsm.get_next_state(fsm.initial_state, 2) == fsm.final_state
-    assert fsm.get_next_state(fsm.initial_state, 1) == fsm.initial_state
-    assert fsm.is_final_state(fsm.initial_state) is False
-    assert fsm.is_final_state(fsm.final_state) is True
-
-    assert isinstance(fsm.copy(), StopAtEOSGuide)
-
-
-def test_regex_vocabulary_error():
-    class MockTokenizer:
-        vocabulary = {"a": 1}
-        special_tokens = {"eos"}
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-    regex_str = "[1-9]"
-
-    with pytest.raises(ValueError, match="The vocabulary"):
-        RegexGuide.from_regex(regex_str, MockTokenizer())
-
-
-def test_regex():
-    class MockTokenizer:
-        vocabulary = {"1": 1, "a": 2, "eos": 3}
-        special_tokens = {"eos"}
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            return token
-
-    regex_str = "[1-9]"
-    tokenizer = MockTokenizer()
-    fsm = RegexGuide.from_regex(regex_str, tokenizer)
-
-    assert fsm.states_to_token_maps.get_transitions() == {0: {1: 1}}
-
-    instruction = fsm.get_next_instruction(0)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [1])
-
-    assert fsm.get_next_state(state=0, token_id=1) == 1
-    assert fsm.get_next_state(state=0, token_id=tokenizer.eos_token_id) == -1
-
-    assert fsm.is_final_state(0) is False
-
-
-def test_regex_multi_byte_llama_like():
-    class MockTokenizer:
-        vocabulary = {
-            "1": 1,
-            "a": 2,
-            "eos": 3,
-            "😍": 4,
-            "<0xF0>": 5,
-            "<0x9F>": 6,
-            "<0x98>": 7,
-            "<0x88>": 8,  # 😈
-            "\ufffd": 9,
-            "\ufffd\ufffd": 10,
-        }
-        special_tokens = {"eos"}
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            if token[0] == "<":
-                return "\ufffd"
-            return token
-
-    regex_str = "[😁-😎]"
-    tokenizer = MockTokenizer()
-    fsm = RegexGuide.from_regex(regex_str, tokenizer)
-
-    assert fsm.states_to_token_maps.get_transitions() == {
-        0: {5: 1, 4: 2},
-        1: {6: 3},
-        3: {7: 4},
-        4: {8: 2},
-    }
-
-    instruction = fsm.get_next_instruction(0)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [5, 4])
-
-    assert fsm.get_next_state(state=0, token_id=5) == 1
-    assert fsm.get_next_state(state=0, token_id=tokenizer.eos_token_id) == -1
-
-    assert fsm.is_final_state(0) is False
-
-
-def test_regex_multi_byte_gpt2_like():
-    class MockTokenizer:
-        vocabulary = {
-            "1": 1,
-            "a": 2,
-            "eos": 3,
-            "😍": 4,
-            " ": 5,
-            "\ufffd": 6,
-            "\ufffd\ufffd": 7,
-            "ðŁĺ": 8,
-            "Ī": 9,  # '😈'
-            "Ġð": 10,
-            "ŁĺĪ": 11,  # ' 😈'
-        }
-        special_tokens = {"eos"}
-        eos_token_id = 3
-
-        def convert_token_to_string(self, token):
-            if self.vocabulary[token] >= 8:
-                return "\ufffd"
-            return token
-
-    regex_str = " [😁-😎]"
-    tokenizer = MockTokenizer()
-    fsm = RegexGuide.from_regex(regex_str, tokenizer)
-
-    assert fsm.states_to_token_maps.get_transitions() == {
-        0: {5: 1, 10: 2},
-        1: {8: 5, 4: 3},
-        2: {11: 3},
-        5: {9: 3},
-    }
-
-    instruction = fsm.get_next_instruction(0)
-    assert isinstance(instruction, Generate)
-    assert_expected_tensor_ids(instruction.tokens, [5, 10])
-
-    assert fsm.get_next_state(state=0, token_id=5) == 1
-    assert fsm.get_next_state(state=0, token_id=tokenizer.eos_token_id) == -1
-
-    assert fsm.is_final_state(0) is False
-
-
-def test_regex_final_state():
-    """Make sure that the FSM stays in the final state as we keep generating"""
-
-    class MockTokenizer:
-        vocabulary = {"`": 101, ".": 102, "\n": 103, "eos": 104}
-        special_tokens = {"eos"}
-        eos_token_id = 104
-
-        def convert_token_to_string(self, token):
-            return token
-
-    regex_str = r"`\n(\.\n)?`\n"
-    tokenizer = MockTokenizer()
-    fsm = RegexGuide.from_regex(regex_str, tokenizer)
-
-    state = fsm.get_next_state(state=4, token_id=103)
-    assert state == 5
-    assert fsm.is_final_state(state)
-
-    state = fsm.get_next_state(state=5, token_id=103)
-    assert fsm.is_final_state(state)
-
-
-def test_regex_guide_caching():
-    assert caching._caching_enabled
-
-    cache = caching.get_cache()
-    cache.clear()
-    _, _ = cache.stats(enable=True, reset=True) # (hits, misses)
-
-    regex = r"[0-9]{3}"
-
-    models = [
-        outlines.from_transformers(
-            transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-            transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini")
-        ),
-        outlines.from_llamacpp(llama_cpp.Llama.from_pretrained(
-            repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
-            filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
-        ))
-    ]
-    if HAS_MLX:
-        models.append(outlines.from_mlxlm(
-            *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
-        ))
-    if HAS_VLLM:
-        models.append(outlines.from_vllm_offline(vllm.LLM("TinyLlama/TinyLlama-1.1B-Chat-v1.0")))
-
-    for i, model in enumerate(models):
-        # First call for each model should be a miss
-        RegexGuide.from_regex(regex, model.tokenizer)
-        expected_misses = i + 1
-        expected_hits = i
-        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
-
-        # Second call for each model
-        RegexGuide.from_regex(regex, model.tokenizer)
-        expected_misses = i + 1
-        expected_hits = i + 1
-        assert cache.stats(enable=True, reset=False) == (expected_hits, expected_misses)
diff --git a/tests/processors/test_structured.py b/tests/processors/test_structured.py
deleted file mode 100644
index 261328922..000000000
--- a/tests/processors/test_structured.py
+++ /dev/null
@@ -1,108 +0,0 @@
-import pytest
-from typing import Dict
-
-import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM
-
-import outlines
-from outlines.processors.guide import RegexGuide
-from outlines.processors.structured import (
-    GuideLogitsProcessor,
-    RegexLogitsProcessor,
-    JSONLogitsProcessor,
-)
-
-
-arithmetic_grammar = """
-?start: sum
-
-?sum: product
-| sum "+" product   -> add
-| sum "-" product   -> sub
-
-?product: atom
-| product "*" atom  -> mul
-| product "/" atom  -> div
-
-?atom: NUMBER           -> number
-| "-" atom         -> neg
-| "(" sum ")"
-
-%import common.NUMBER
-%import common.WS_INLINE
-
-%ignore WS_INLINE
-"""
-
-@pytest.fixture
-def tokenizer():
-    model = outlines.from_transformers(
-        AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    return model.tokenizer
-
-
-def test_structured_init_guide_logits_processor(tokenizer):
-    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
-    processor = GuideLogitsProcessor(
-        tokenizer=tokenizer,
-        guide=guide,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, GuideLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is guide
-    assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_init_guide_logits_processor_copy(tokenizer):
-    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
-    processor = GuideLogitsProcessor(
-        tokenizer=tokenizer,
-        guide=guide,
-        tensor_library_name="torch",
-    )
-    processor_copy = processor.copy()
-    assert processor_copy is not processor
-    assert processor_copy.tokenizer is tokenizer
-    assert processor_copy.guide is guide
-    assert processor_copy.tensor_adapter.library_name == "torch"
-
-
-def test_structured_guide_logits_processor_call(tokenizer):
-    guide = RegexGuide.from_regex(r"[a-z]+", tokenizer)
-    processor = GuideLogitsProcessor(
-        tokenizer=tokenizer,
-        guide=guide,
-        tensor_library_name="torch",
-    )
-    vocab_token_ids = list(tokenizer.vocabulary.values())
-    logits = torch.randn(1, len(vocab_token_ids))
-    input_ids = torch.randint(0, len(vocab_token_ids), (1, 10))
-    output = processor(input_ids, logits)
-    assert output.shape == (1, len(vocab_token_ids))
-
-
-def test_structured_init_json_logits_processor(tokenizer):
-    processor = JSONLogitsProcessor(
-        schema={"type": "object", "properties": {"name": {"type": "string"}}},
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, JSONLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is not None
-    assert processor.tensor_adapter.library_name == "torch"
-
-
-def test_structured_init_regex_logits_processor(tokenizer):
-    processor = RegexLogitsProcessor(
-        regex_string=r"[a-z]+",
-        tokenizer=tokenizer,
-        tensor_library_name="torch",
-    )
-    assert isinstance(processor, RegexLogitsProcessor)
-    assert processor.tokenizer is tokenizer
-    assert processor.guide is not None
-    assert processor.tensor_adapter.library_name == "torch"
diff --git a/tests/test_cache.py b/tests/test_cache.py
index 149d856c0..b282fdf2f 100644
--- a/tests/test_cache.py
+++ b/tests/test_cache.py
@@ -206,13 +206,11 @@ def temp_cache_dir():
     import tempfile
 
     import outlines.caching
-    import outlines.processors.guide
 
     with tempfile.TemporaryDirectory() as tempdir:
         os.environ["OUTLINES_CACHE_DIR"] = tempdir
         outlines.caching.get_cache.cache_clear()
         reload(outlines)
-        reload(outlines.processors.guide)
         cache_status = outlines.caching._caching_enabled
         try:
             outlines.caching._caching_enabled = True
diff --git a/tests/test_generator.py b/tests/test_generator.py
index 2c26e98d0..e14ca5381 100644
--- a/tests/test_generator.py
+++ b/tests/test_generator.py
@@ -2,8 +2,10 @@
 from typing import AsyncGenerator, Generator as TypingGenerator, Literal
 
 import transformers
+from outlines_core import Index, Vocabulary
 
 import outlines
+from outlines.backends.outlines_core import OutlinesCoreLogitsProcessor
 from outlines.generator import (
     BlackBoxGenerator,
     SteerableGenerator,
@@ -13,7 +15,6 @@
 from outlines.models import AsyncVLLM, VLLM
 from outlines.processors import (
     OutlinesLogitsProcessor,
-    RegexLogitsProcessor,
 )
 from outlines.types import CFG
 from tests.test_utils.mock_openai_client import (
@@ -68,16 +69,9 @@ def steerable_model():
 
 @pytest.fixture(scope="session")
 def sample_processor():
-    model = outlines.from_transformers(
-        transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-    processor = RegexLogitsProcessor(
-        regex_string="[0-9]{3}",
-        tokenizer=model.tokenizer,
-        tensor_library_name=model.tensor_library_name,
-    )
-    return processor
+    vocabulary = Vocabulary.from_pretrained("openai-community/gpt2")
+    index = Index(r"[0-9]{3}", vocabulary)
+    return OutlinesCoreLogitsProcessor(index, "torch")
 
 
 @pytest.fixture(scope="module")
@@ -138,7 +132,6 @@ def test_black_box_generator_init(black_box_sync_model):
     assert generator.model == black_box_sync_model
     assert generator.output_type == Literal["foo", "bar"]
 
-
 def test_black_box_generator_call(black_box_sync_model):
     generator = BlackBoxGenerator(black_box_sync_model, str)
     result = generator("Write a very short sentence", max_tokens=10)
@@ -152,7 +145,6 @@ def test_black_box_generator_stream(black_box_sync_model):
     assert isinstance(next(result), str)
 
 
-
 # AsyncBlackBoxGenerator
 
 

From b81c91860b6ea4e3ceee07844ad0219257af5415 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Fri, 1 Aug 2025 14:54:32 +0200
Subject: [PATCH 420/505] Create our own logits processor for Xgrammar, add
 support for mlx

---
 docs/features/advanced/backends.md |   2 +-
 outlines/backends/xgrammar.py      | 143 +++++++++++++++++++++--------
 tests/backends/test_xgrammar.py    |   2 +-
 3 files changed, 107 insertions(+), 40 deletions(-)

diff --git a/docs/features/advanced/backends.md b/docs/features/advanced/backends.md
index 84e222e10..c8b3fa665 100644
--- a/docs/features/advanced/backends.md
+++ b/docs/features/advanced/backends.md
@@ -47,7 +47,7 @@ As mentioned previously, selecting the structured generation backend is only app
 | **Models** | | | |
 | Transformers | ✅ | ✅ | ✅ |
 | LlamaCpp | ✅ | ✅ | ❌ |
-| MLXLM | ✅ | ✅ | ❌ |
+| MLXLM | ✅ | ✅ | ✅ |
 | **Output Types** | | | |
 | JSON Schema | ✅ | ✅ | ✅ |
 | Regex | ✅ | ✅ | ✅ |
diff --git a/outlines/backends/xgrammar.py b/outlines/backends/xgrammar.py
index fcb1672da..c9c935fd7 100644
--- a/outlines/backends/xgrammar.py
+++ b/outlines/backends/xgrammar.py
@@ -1,57 +1,109 @@
 """Backend class for XGrammar."""
 
-from typing import TYPE_CHECKING
-
 from outlines.backends.base import BaseBackend
 from outlines.models import SteerableModel
+from outlines.models.mlxlm import MLXLM
 from outlines.models.transformers import Transformers
 from outlines.processors.base_logits_processor import (
     OutlinesLogitsProcessor,
     TensorType
 )
 
-if TYPE_CHECKING:
-    from xgrammar.contrib.hf import LogitsProcessor
-
 
 class XGrammarLogitsProcessor(OutlinesLogitsProcessor):
-    """Logits processor for XGrammar.
-
-    This class wraps the `xgr.contrib.hf.LogitsProcessor` class and adds
-    a `reset` method to reset the logits processor for a new generation.
-
-    """
+    """Logits processor for XGrammar."""
 
-    def __init__(self, compiled_grammar: str):
+    def __init__(self, compiled_grammar: str, tensor_library_name: str,):
         """
         Parameters
         ----------
         compiled_grammar: str
             The compiled grammar to use to create the logits processor.
+        tensor_library_name: str
+            The name of the tensor library used by the model
 
         """
         import xgrammar as xgr
 
         self.xgr = xgr
+        self.is_first_token = True
         self.compiled_grammar = compiled_grammar
-        self.xgrammar_logits_processor = None
-        super().__init__("torch")
+        self.tensor_library_name = tensor_library_name
+        super().__init__(tensor_library_name)
 
     def reset(self):
-        """Reset the logits processor for a new generation."""
-        self.xgrammar_logits_processor = None
-
-    def process_logits(self, input_ids: TensorType, logits: TensorType) -> TensorType:
-        """Bias the logits."""
-        if self.xgrammar_logits_processor is None:
-            self.xgrammar_logits_processor = self.xgr.contrib.hf.LogitsProcessor(
-                self.compiled_grammar
+        """Ensure self._setup is called again for the next generation."""
+        self.is_first_token = True
+
+    def _setup(self, batch_size: int, vocab_size: int) -> None:
+        """Setup the logits processor for a new generation."""
+        if self.tensor_library_name == "torch":
+            self._bias_logits = self._bias_logits_torch
+        elif self.tensor_library_name == "mlx":
+            self._bias_logits = self._bias_logits_mlx
+        else: # pragma: no cover
+            raise ValueError(
+                f"Unsupported tensor library: {self.tensor_library_name}"
             )
-        return self.xgrammar_logits_processor(input_ids, logits) # type: ignore
+
+        self._matchers = [
+            self.xgr.GrammarMatcher(self.compiled_grammar)
+            for _ in range(batch_size)
+        ]
+        self._bitmask = self.xgr.allocate_token_bitmask(batch_size, vocab_size)
+
+    def _bias_logits_torch(
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for Torch tensors."""
+        for i in range(self.tensor_adapter.shape(input_ids)[0]):
+            if not self._matchers[i].is_terminated():
+                self._matchers[i].fill_next_token_bitmask(self._bitmask, i)
+
+        self.xgr.apply_token_bitmask_inplace(logits, self._bitmask)
+
+        return logits
+
+    def _bias_logits_mlx( # pragma: no cover
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Bias the logits for MLX tensors."""
+        import mlx.core as mx
+        from xgrammar.kernels.apply_token_bitmask_mlx import apply_token_bitmask_mlx
+
+        for i in range(self.tensor_adapter.shape(input_ids)[0]):
+            if not self._matchers[i].is_terminated():
+                self._matchers[i].fill_next_token_bitmask(self._bitmask, i)
+
+        biased_logits = apply_token_bitmask_mlx(
+            mx.array(self._bitmask.numpy()), logits, self.tensor_adapter.shape(logits)[1]
+        )
+
+        return biased_logits
+
+    def process_logits(
+        self, input_ids: TensorType, logits: TensorType
+    ) -> TensorType:
+        """Use the XGrammar matchers to bias the logits."""
+        batch_size = self.tensor_adapter.shape(input_ids)[0]
+        vocab_size = self.tensor_adapter.shape(logits)[1]
+
+        if self.is_first_token:
+            self._setup(batch_size, vocab_size)
+            self.is_first_token = False
+        else:
+            for i in range(batch_size):
+                if not self._matchers[i].is_terminated():
+                    last_token_id = self.tensor_adapter.to_scalar(
+                        input_ids[i][-1] # type: ignore
+                    )
+                    assert self._matchers[i].accept_token(last_token_id)
+
+        return self._bias_logits(input_ids, logits)
 
 
 class XGrammarBackend(BaseBackend):
-    """Backend for XGRammar."""
+    """Backend for XGrammar."""
 
     def __init__(self, model: SteerableModel):
         """
@@ -62,25 +114,27 @@ def __init__(self, model: SteerableModel):
 
         """
         import xgrammar as xgr
-        from transformers import AutoConfig
 
-        if not isinstance(model, Transformers):
+        if isinstance(model, Transformers):
+            tokenizer = model.hf_tokenizer
+        elif isinstance(model, MLXLM):
+            tokenizer = model.mlx_tokenizer._tokenizer
+        else: # pragma: no cover
             raise ValueError(
-                "The xgrammar backend only supports Transformers models"
+                "The xgrammar backend only supports Transformers and "
+                + "MLXLM models"
             )
 
-        vocab_size = AutoConfig.from_pretrained(
-            model.model.config._name_or_path
-        ).vocab_size
         tokenizer_info = xgr.TokenizerInfo.from_huggingface(
-            model.hf_tokenizer,
-            vocab_size=vocab_size
+            tokenizer,
+            vocab_size=len(tokenizer.get_vocab())
         )
         self.grammar_compiler = xgr.GrammarCompiler(tokenizer_info)
+        self.tensor_library_name = model.tensor_library_name
 
     def get_json_schema_logits_processor(
         self, json_schema: str
-    ) -> "LogitsProcessor":
+    ) -> XGrammarLogitsProcessor:
         """Create a logits processor from a JSON schema.
 
         Parameters
@@ -97,9 +151,14 @@ def get_json_schema_logits_processor(
         compiled_grammar = self.grammar_compiler.compile_json_schema(
             json_schema
         )
-        return XGrammarLogitsProcessor(compiled_grammar)
+        return XGrammarLogitsProcessor(
+            compiled_grammar,
+            self.tensor_library_name
+        )
 
-    def get_regex_logits_processor(self, regex: str) -> "LogitsProcessor":
+    def get_regex_logits_processor(
+        self, regex: str
+    ) -> XGrammarLogitsProcessor:
         """Create a logits processor from a regex.
 
         Parameters
@@ -114,9 +173,14 @@ def get_regex_logits_processor(self, regex: str) -> "LogitsProcessor":
 
         """
         compiled_grammar = self.grammar_compiler.compile_regex(regex)
-        return XGrammarLogitsProcessor(compiled_grammar)
+        return XGrammarLogitsProcessor(
+            compiled_grammar,
+            self.tensor_library_name
+        )
 
-    def get_cfg_logits_processor(self, grammar: str) -> "LogitsProcessor":
+    def get_cfg_logits_processor(
+        self, grammar: str
+    ) -> XGrammarLogitsProcessor:
         """Create a logits processor from a context-free grammar.
 
         Parameters
@@ -131,4 +195,7 @@ def get_cfg_logits_processor(self, grammar: str) -> "LogitsProcessor":
 
         """
         compiled_grammar = self.grammar_compiler.compile_grammar(grammar)
-        return XGrammarLogitsProcessor(compiled_grammar)
+        return XGrammarLogitsProcessor(
+            compiled_grammar,
+            self.tensor_library_name
+        )
diff --git a/tests/backends/test_xgrammar.py b/tests/backends/test_xgrammar.py
index 1133b9f0f..382c508d7 100644
--- a/tests/backends/test_xgrammar.py
+++ b/tests/backends/test_xgrammar.py
@@ -90,6 +90,6 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg):
 def test_xgrammar_backend_invalid_model(model_llamacpp):
     with pytest.raises(
         ValueError,
-        match="The xgrammar backend only supports Transformers models",
+        match="The xgrammar backend only supports Transformers and MLXLM models",
     ):
         XGrammarBackend(model_llamacpp)

From a04b941821a84103472ba462e9fa05bf0b58bec2 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 4 Aug 2025 10:55:10 +0200
Subject: [PATCH 421/505] Turn tokens into strings before building the
 Outlines-Core Vocabulary

---
 outlines/backends/outlines_core.py | 42 +++++++++++++++++++++---------
 1 file changed, 29 insertions(+), 13 deletions(-)

diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index c87549263..4da664897 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -1,6 +1,6 @@
 """Backend class for Outlines Core."""
 
-from typing import Dict
+from typing import Callable, Dict
 
 from outlines_core import Guide, Index, Vocabulary
 # TODO: change this once the import issue is fixed in outlines_core
@@ -184,23 +184,29 @@ def __init__(self, model: SteerableModel):
 
         """
         if isinstance(model, Transformers):
-            vocabulary = model.hf_tokenizer.get_vocab()
-            eos_token_id = model.hf_tokenizer.eos_token_id
-            eos_token = model.hf_tokenizer.eos_token
+            tokenizer = model.hf_tokenizer
+            vocabulary = tokenizer.get_vocab()
+            eos_token_id = tokenizer.eos_token_id
+            eos_token = tokenizer.eos_token
+            token_to_str = lambda token: tokenizer.convert_tokens_to_string([token])
         elif isinstance(model, LlamaCpp):
-            vocabulary = model.tokenizer.vocabulary
-            eos_token_id = model.tokenizer.eos_token_id
-            eos_token = model.tokenizer.eos_token
+            tokenizer = model.tokenizer
+            vocabulary = tokenizer.vocabulary
+            eos_token_id = tokenizer.eos_token_id
+            eos_token = tokenizer.eos_token
+            token_to_str = tokenizer.convert_token_to_string
         elif isinstance(model, MLXLM):
-            vocabulary = model.mlx_tokenizer._tokenizer.get_vocab()
-            eos_token_id = model.mlx_tokenizer._tokenizer.eos_token_id
-            eos_token = model.mlx_tokenizer._tokenizer.eos_token
+            tokenizer = model.mlx_tokenizer
+            vocabulary = tokenizer.get_vocab()
+            eos_token_id = tokenizer.eos_token_id
+            eos_token = tokenizer.eos_token
+            token_to_str = lambda token: tokenizer.convert_tokens_to_string([token])
         else:
             raise ValueError(f"Unsupported model type: {type(model)}")
 
         self.eos_token_id = eos_token_id
         self.vocabulary = self.create_outlines_core_vocabulary(
-            vocabulary, eos_token_id, eos_token
+            vocabulary, eos_token_id, eos_token, token_to_str
         )
         self.tensor_library_name = model.tensor_library_name
 
@@ -247,7 +253,10 @@ def get_cfg_logits_processor(self, grammar):
 
     @staticmethod
     def create_outlines_core_vocabulary(
-        vocab: Dict[str, int], eos_token_id: int, eos_token: str
+        vocab: Dict[str, int],
+        eos_token_id: int,
+        eos_token: str,
+        token_to_str: Callable[[str], str]
     ) -> Vocabulary:
         """Create an Outlines Core Vocabulary instance.
 
@@ -259,6 +268,8 @@ def create_outlines_core_vocabulary(
             The EOS token ID.
         eos_token: str
             The EOS token.
+        token_to_str: Callable[[str], str]
+            The function to convert a token to a string.
 
         Returns
         -------
@@ -268,6 +279,11 @@ def create_outlines_core_vocabulary(
         """
         formatted_vocab = {}
         for token, token_id in vocab.items():
-            formatted_vocab[token] = [token_id]
+            # This step is necessary to transform special tokens into their
+            # string representation, in particular for spacing. We need those
+            # string representations as outlines core first builds an FSM from
+            # the regex provided that only contains regular strings.
+            token_as_str = token_to_str(token)
+            formatted_vocab[token_as_str] = [token_id]
         formatted_vocab.pop(eos_token)
         return Vocabulary(eos_token_id, formatted_vocab)

From a8464ac154d89208a7bd76c5c02b79be1a22bc9b Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 8 Aug 2025 12:01:24 +0200
Subject: [PATCH 422/505] Fix bug in outlines_core for transformers about
 token-str conversion

---
 outlines/backends/outlines_core.py | 10 +++++-----
 outlines/models/transformers.py    |  1 +
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index 4da664897..279cf1391 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -184,23 +184,23 @@ def __init__(self, model: SteerableModel):
 
         """
         if isinstance(model, Transformers):
-            tokenizer = model.hf_tokenizer
+            tokenizer = model.tokenizer
             vocabulary = tokenizer.get_vocab()
             eos_token_id = tokenizer.eos_token_id
             eos_token = tokenizer.eos_token
-            token_to_str = lambda token: tokenizer.convert_tokens_to_string([token])
+            token_to_str = tokenizer.convert_token_to_string
         elif isinstance(model, LlamaCpp):
-            tokenizer = model.tokenizer
+            tokenizer = model.tokenizer # type: ignore
             vocabulary = tokenizer.vocabulary
             eos_token_id = tokenizer.eos_token_id
             eos_token = tokenizer.eos_token
             token_to_str = tokenizer.convert_token_to_string
         elif isinstance(model, MLXLM):
-            tokenizer = model.mlx_tokenizer
+            tokenizer = model.mlx_tokenizer # type: ignore
             vocabulary = tokenizer.get_vocab()
             eos_token_id = tokenizer.eos_token_id
             eos_token = tokenizer.eos_token
-            token_to_str = lambda token: tokenizer.convert_tokens_to_string([token])
+            token_to_str = lambda token: tokenizer.convert_tokens_to_string([token]) # type: ignore
         else:
             raise ValueError(f"Unsupported model type: {type(model)}")
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 2b8fca13e..c27f55aee 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -72,6 +72,7 @@ def __init__(self, tokenizer: "PreTrainedTokenizer", **kwargs):
         self.tokenizer = tokenizer
         self.eos_token_id = self.tokenizer.eos_token_id
         self.eos_token = self.tokenizer.eos_token
+        self.get_vocab = self.tokenizer.get_vocab
 
         if self.tokenizer.pad_token_id is None:
             self.tokenizer.pad_token_id = self.tokenizer.eos_token_id

From 446700e03ac07e4931deacefdebf32b5d69a4239 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 7 Aug 2025 17:06:19 +0200
Subject: [PATCH 423/505] Fix a bug regarding the eos_token in the LlamaCpp
 tokenizer

---
 outlines/models/llamacpp.py | 46 ++++++++++++++++++++++++++-----------
 1 file changed, 32 insertions(+), 14 deletions(-)

diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 9984c725d..c6a3bf8b0 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -1,5 +1,6 @@
 """Integration with the `llama-cpp-python` library."""
 
+import ctypes
 from functools import singledispatchmethod
 from typing import (
     TYPE_CHECKING,
@@ -26,34 +27,51 @@
 
 class LlamaCppTokenizer(Tokenizer):
     def __init__(self, model: "Llama"):
-        self.eos_token_id = model.token_eos()
-        self.eos_token = model.tokenizer().decode([self.eos_token_id])
-        self.pad_token_id = self.eos_token_id
+        self.tokenizer = model.tokenizer()
         self.special_tokens: Set[str] = set()
-
         self.vocabulary: Dict[str, int] = dict()
 
-        self.tokenizer = model.tokenizer()
-
         # TODO: Remove when https://github.com/ggerganov/llama.cpp/pull/5613
         # is resolved
         self._hf_tokenizer = None
-        try:
-            self.vocabulary = model.tokenizer_.hf_tokenizer.get_vocab()
+        if (
+            hasattr(model, "tokenizer_")
+            and hasattr(model.tokenizer_, "hf_tokenizer")
+        ):
             self._hf_tokenizer = model.tokenizer_.hf_tokenizer
-        except AttributeError:
-            # ###
-            for t in range(model.n_vocab()):
-                token_piece = model.tokenizer().decode([t])
-                self.vocabulary[token_piece] = t
+            self.eos_token_id = self._hf_tokenizer.eos_token_id
+            self.eos_token = self._hf_tokenizer.eos_token
+            self.vocabulary = self._hf_tokenizer.get_vocab()
+        else:
+            from llama_cpp import (
+                llama_model_get_vocab,
+                llama_token_to_piece,
+            )
+
+            self.eos_token_id = model.token_eos()
+            size = 32
+            buffer = (ctypes.c_char * size)()
+            for i in range(model.n_vocab()):
+                n = llama_token_to_piece(
+                    llama_model_get_vocab(model.model),
+                    i,
+                    buffer,
+                    size,
+                    0,
+                    True
+                )
+                token_piece = buffer[:n].decode("utf-8", errors="replace") # type: ignore
+                self.vocabulary[token_piece] = i
+                if i == self.eos_token_id:
+                    self.eos_token = token_piece
 
+        self.pad_token_id = self.eos_token_id
         # ensure stable ordering of vocabulary
         self.vocabulary = {
             tok: tok_id
             for tok, tok_id
             in sorted(self.vocabulary.items(), key=lambda x: x[1])
         }
-
         self._hash = None
 
     def decode(self, token_ids: List[int]) -> List[str]:

From 8cecc55b7f61590392aa5ca1ca81aadbb639d1dd Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 7 Aug 2025 22:43:04 +0200
Subject: [PATCH 424/505] Fix bug related to the device of the bitmasks and
 logits for torch

---
 outlines/backends/llguidance.py    | 11 ++++++++++-
 outlines/backends/outlines_core.py | 18 +++++++++++++++++-
 outlines/backends/xgrammar.py      |  8 ++++++++
 3 files changed, 35 insertions(+), 2 deletions(-)

diff --git a/outlines/backends/llguidance.py b/outlines/backends/llguidance.py
index 2f9ea59f5..362031adf 100644
--- a/outlines/backends/llguidance.py
+++ b/outlines/backends/llguidance.py
@@ -115,8 +115,17 @@ def _bias_logits_torch(
 
         for i in range(self.tensor_adapter.shape(input_ids)[0]):
             llguidance.torch.fill_next_token_bitmask(self.ll_matchers[i], self.bitmask, i)
+            self.bitmask = self.tensor_adapter.to_device(
+                self.bitmask,
+                self.tensor_adapter.get_device(logits)
+            )
             llguidance.torch.apply_token_bitmask_inplace(
-                logits[i], self.bitmask[i] # type: ignore
+                logits[i], # type: ignore
+                self.bitmask[i]
+            )
+            self.bitmask = self.tensor_adapter.to_device(
+                self.bitmask,
+                "cpu"
             )
 
         return logits
diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index 279cf1391..9929bff9f 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -47,7 +47,15 @@ def _setup(self, batch_size: int, vocab_size: int) -> None:
         `process_logits` method.
 
         This method is called when the first token is generated instead of
-        at initialization because we need to know the batch size.
+        at initialization because we need to know the batch size and the device
+        of the logits.
+
+        Parameters
+        ----------
+        batch_size: int
+            The batch size.
+        vocab_size: int
+            The vocabulary size.
 
         """
         if self.tensor_library_name == "torch":
@@ -111,10 +119,18 @@ def _bias_logits_torch(
 
         for i in range(batch_size):
             fill_next_token_bitmask(self._guides[i], self._bitmasks[i])
+            self._bitmasks[i] = self.tensor_adapter.to_device(
+                self._bitmasks[i],
+                self.tensor_adapter.get_device(logits)
+            )
             apply_token_bitmask_inplace(
                 self.tensor_adapter.unsqueeze(logits[i]), # type: ignore
                 self._bitmasks[i]
             )
+            self._bitmasks[i] = self.tensor_adapter.to_device(
+                self._bitmasks[i],
+                "cpu"
+            )
 
         return logits
 
diff --git a/outlines/backends/xgrammar.py b/outlines/backends/xgrammar.py
index c9c935fd7..c316477d6 100644
--- a/outlines/backends/xgrammar.py
+++ b/outlines/backends/xgrammar.py
@@ -60,7 +60,15 @@ def _bias_logits_torch(
             if not self._matchers[i].is_terminated():
                 self._matchers[i].fill_next_token_bitmask(self._bitmask, i)
 
+        self._bitmask = self.tensor_adapter.to_device(
+            self._bitmask,
+            self.tensor_adapter.get_device(logits)
+        )
         self.xgr.apply_token_bitmask_inplace(logits, self._bitmask)
+        self._bitmask = self.tensor_adapter.to_device(
+            self._bitmask,
+            "cpu"
+        )
 
         return logits
 

From b3fba1369736cc2f7e8f5d6295fe90d6b01da970 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 11 Aug 2025 14:20:52 +0200
Subject: [PATCH 425/505] Create AsyncOpenAI model

---
 docs/features/models/index.md  |   3 +-
 docs/features/models/openai.md |  56 ++++++++-
 outlines/models/__init__.py    |   3 +-
 outlines/models/openai.py      | 213 ++++++++++++++++++++++++++++++---
 tests/models/test_openai.py    | 173 +++++++++++++++++++++++++-
 5 files changed, 423 insertions(+), 25 deletions(-)

diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index 66673dbf4..8049107e7 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -93,7 +93,7 @@ In alphabetical order:
 | Regex | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
 | Grammar | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | 🟠 | ❌ | ✅ | ✅ | ✅ | ✅ |
 | **Generation Features** | | | | | | | | | | | | | |
-| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
 | Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ✅ |
@@ -142,6 +142,7 @@ print(type(model)) # outlines.models.tgi.AsyncTGI
 The models that have an async version are the following:
 
 - Ollama
+- OpenAI
 - SgLang
 - TGI
 - VLLM
diff --git a/docs/features/models/openai.md b/docs/features/models/openai.md
index a52544a35..07a7fa08f 100644
--- a/docs/features/models/openai.md
+++ b/docs/features/models/openai.md
@@ -10,23 +10,32 @@
 
 To create an OpenAI model instance, you can use the `from_openai` function. It takes 2 arguments:
 
-- `client`: an `openai.OpenAI` or `openai.AzureOpenAI` instance
+- `client`: an `openai.OpenAI`, `openai.AzureOpenAI`, `openai.AsyncOpenAI` or `openai.AsyncAzureOpenAI` instance
 - `model_name`: the name of the model you want to use
 
+Based on whether the inference client instance is synchronous or asynchronous, you will receive an `OpenAI` or an `AsyncOpenAI` model instance.
+
 For instance:
 
 ```python
 import outlines
 import openai
 
-# Create the client
+# Create the client or async client
 client = openai.OpenAI()
+async_client = openai.AsyncOpenAI()
 
-# Create the model
+# Create a sync model
 model = outlines.from_openai(
     client,
     "gpt-4o"
 )
+
+# Create aa async model
+model = outlines.from_openai(
+    async_client,
+    "gpt-4o"
+)
 ```
 
 Check the [OpenAI documentation](https://platform.openai.com/docs/models) for an up-to-date list of available models. As shown above, you can use Azure OpenAI in Outlines the same way you would use OpenAI, just provide an `openai.AzureOpenAI` instance to the Outlines model class.
@@ -190,6 +199,47 @@ result = model("Create a character, use the json format.", dict, temperature=0.5
 print(result) # '{"first_name": "Henri", "last_name": "Smith", "height": "170"}'
 ```
 
+## Asynchronous Calls
+
+All features presented above for the sync model are also available for the async model.
+
+For instance:
+
+```python
+import asyncio
+import openai
+import outlines
+from pydantic import BaseModel
+from typing import List
+
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
+
+# Create the model
+model = outlines.from_openai(
+    openai.AsyncOpenAI(),
+    "gpt-4o"
+)
+
+async def text_generation():
+    # Regular generation
+    response = await model("What's the capital of Latvia?", max_tokens=20)
+    print(response) # 'Riga'
+
+    # Streaming
+    async for chunk in  model.stream("Tell me a short story about a cat.", max_tokens=50):
+        print(chunk, end="") # 'Once...'
+
+    # Structured generation
+    result = await model("Create a character, use the json format.", Character, top_p=0.1)
+    print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+    print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
+
+asyncio.run(text_generation())
+```
+
 ## Inference arguments
 
 When calling the model, you can provide keyword arguments that will be passed down to the `chat.completions.create` method of the OpenAI client. Some of the most common arguments include `max_tokens`, `temperature`, `stop` and `top_p`.
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index e705d0555..1bd7aa127 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -15,7 +15,7 @@
 from .llamacpp import LlamaCpp, from_llamacpp
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import AsyncOllama, Ollama, from_ollama
-from .openai import OpenAI, from_openai
+from .openai import AsyncOpenAI, OpenAI, from_openai
 from .sglang import AsyncSGLang, SGLang, from_sglang
 from .tgi import AsyncTGI, TGI, from_tgi
 from .transformers import (
@@ -41,6 +41,7 @@
 ]
 AsyncBlackBoxModel = Union[
     AsyncOllama,
+    AsyncOpenAI,
     AsyncTGI,
     AsyncSGLang,
     AsyncVLLM,
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index b17b64209..dad987b93 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -4,6 +4,7 @@
 from typing import (
     TYPE_CHECKING,
     Any,
+    AsyncIterator,
     Iterator,
     Optional,
     Union,
@@ -13,7 +14,7 @@
 from pydantic import BaseModel, TypeAdapter
 
 from outlines.inputs import Chat, Image
-from outlines.models.base import Model, ModelTypeAdapter
+from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.models.utils import set_additional_properties_false_json_schema
 from outlines.types import JsonSchema, Regex, CFG
 from outlines.types.utils import (
@@ -25,9 +26,14 @@
 )
 
 if TYPE_CHECKING:
-    from openai import OpenAI as OpenAIClient, AzureOpenAI as AzureOpenAIClient
+    from openai import (
+        OpenAI as OpenAIClient,
+        AsyncOpenAI as AsyncOpenAIClient,
+        AzureOpenAI as AzureOpenAIClient,
+        AsyncAzureOpenAI as AsyncAzureOpenAIClient,
+    )
 
-__all__ = ["OpenAI", "from_openai"]
+__all__ = ["AsyncOpenAI", "OpenAI", "from_openai"]
 
 
 class OpenAITypeAdapter(ModelTypeAdapter):
@@ -348,36 +354,211 @@ def generate_stream(
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
-        stream = self.client.chat.completions.create(
-            stream=True,
-            messages=messages,
-            **response_format,
-            **inference_kwargs
-        )
+        try:
+            stream = self.client.chat.completions.create(
+                stream=True,
+                messages=messages,
+                **response_format,
+                **inference_kwargs
+            )
+        except openai.BadRequestError as e:
+            if e.body["message"].startswith("Invalid schema"):
+                raise TypeError(
+                    f"OpenAI does not support your schema: {e.body['message']}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise e
 
         for chunk in stream:
             if chunk.choices and chunk.choices[0].delta.content is not None:
                 yield chunk.choices[0].delta.content
 
 
+class AsyncOpenAI(AsyncModel):
+    """Thin wrapper around the `openai.AsyncOpenAI` client.
+
+    This wrapper is used to convert the input and output types specified by the
+    users at a higher level to arguments to the `openai.AsyncOpenAI` client.
+
+    """
+
+    def __init__(
+        self,
+        client: Union["AsyncOpenAIClient", "AsyncAzureOpenAIClient"],
+        model_name: Optional[str] = None,
+    ):
+        """
+        Parameters
+        ----------
+        client
+            The `openai.AsyncOpenAI` or `openai.AsyncAzureOpenAI` client.
+        model_name
+            The name of the model to use.
+
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = OpenAITypeAdapter()
+
+    async def generate(
+        self,
+        model_input: Union[Chat, list, str],
+        output_type: Optional[Union[type[BaseModel], str]] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate text using OpenAI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema or an empty dictionary.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The text generated by the model.
+
+        """
+        import openai
+
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        try:
+            result = await self.client.chat.completions.create(
+                messages=messages,
+                **response_format,
+                **inference_kwargs,
+            )
+        except openai.BadRequestError as e:
+            if e.body["message"].startswith("Invalid schema"):
+                raise TypeError(
+                    f"OpenAI does not support your schema: {e.body['message']}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise e
+
+        messages = [choice.message for choice in result.choices]
+        for message in messages:
+            if message.refusal is not None:
+                raise ValueError(
+                    f"OpenAI refused to answer the request: {message.refusal}"
+                )
+
+        if len(messages) == 1:
+            return messages[0].content
+        else:
+            return [message.content for message in messages]
+
+    async def generate_batch(
+        self,
+        model_input,
+        output_type = None,
+        **inference_kwargs,
+    ):
+        raise NotImplementedError(
+            "The `openai` library does not support batch inference."
+        )
+
+    async def generate_stream( # type: ignore
+        self,
+        model_input: Union[Chat, list, str],
+        output_type: Optional[Union[type[BaseModel], str]] = None,
+        **inference_kwargs,
+    ) -> AsyncIterator[str]:
+        """Stream text using OpenAI.
+
+        Parameters
+        ----------
+        model_input
+            The prompt based on which the model will generate a response.
+        output_type
+            The desired format of the response generated by the model. The
+            output type must be of a type that can be converted to a JSON
+            schema or an empty dictionary.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text generated by the model.
+
+        """
+        import openai
+
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        try:
+            stream = await self.client.chat.completions.create(
+                stream=True,
+                messages=messages,
+                **response_format,
+                **inference_kwargs
+            )
+        except openai.BadRequestError as e:
+            if e.body["message"].startswith("Invalid schema"):
+                raise TypeError(
+                    f"OpenAI does not support your schema: {e.body['message']}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise e
+
+        async for chunk in stream:
+            if chunk.choices and chunk.choices[0].delta.content is not None:
+                yield chunk.choices[0].delta.content
+
+
 def from_openai(
-    client: Union["OpenAIClient", "AzureOpenAIClient"],
+    client: Union[
+        "OpenAIClient",
+        "AsyncOpenAIClient",
+        "AzureOpenAIClient",
+        "AsyncAzureOpenAIClient",
+    ],
     model_name: Optional[str] = None,
-) -> OpenAI:
-    """Create an Outlines `OpenAI` model instance from an `openai.OpenAI`
-    client.
+) -> Union[OpenAI, AsyncOpenAI]:
+    """Create an Outlines `OpenAI` or `AsyncOpenAI` model instance from an
+    `openai.OpenAI` or `openai.AsyncOpenAI` client.
 
     Parameters
     ----------
     client
-        An `openai.OpenAI` client instance.
+        An `openai.OpenAI`, `openai.AsyncOpenAI`, `openai.AzureOpenAI` or
+        `openai.AsyncAzureOpenAI` client instance.
     model_name
         The name of the model to use.
 
     Returns
     -------
     OpenAI
-        An Outlines `OpenAI` model instance.
+        An Outlines `OpenAI` or `AsyncOpenAI` model instance.
 
     """
-    return OpenAI(client, model_name)
+    import openai
+
+    if isinstance(client, openai.OpenAI):
+        return OpenAI(client, model_name)
+    elif isinstance(client, openai.AsyncOpenAI):
+        return AsyncOpenAI(client, model_name)
+    else:
+        raise ValueError(
+            "Invalid client type. The client must be an instance of "
+            "+ `openai.OpenAI` or `openai.AsyncOpenAI`."
+        )
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
index 2f9ba4c03..249785d30 100644
--- a/tests/models/test_openai.py
+++ b/tests/models/test_openai.py
@@ -1,16 +1,16 @@
 import io
 import json
 import os
-from typing import Annotated, Generator
+from typing import Annotated, Generator, AsyncGenerator
 
 import pytest
 from PIL import Image as PILImage
-from openai import OpenAI as OpenAIClient
+from openai import AsyncOpenAI as AsyncOpenAIClient, OpenAI as OpenAIClient
 from pydantic import BaseModel, Field
 
 import outlines
 from outlines.inputs import Chat, Image, Video
-from outlines.models.openai import OpenAI
+from outlines.models.openai import AsyncOpenAI, OpenAI
 from outlines.types import json_schema
 
 MODEL_NAME = "gpt-4o-mini-2024-07-18"
@@ -50,12 +50,22 @@ def model(api_key):
     return OpenAI(OpenAIClient(api_key=api_key), MODEL_NAME)
 
 
+@pytest.fixture(scope="session")
+def async_model(api_key):
+    return AsyncOpenAI(AsyncOpenAIClient(api_key=api_key), MODEL_NAME)
+
+
 @pytest.fixture(scope="session")
 def model_no_model_name(api_key):
     return OpenAI(OpenAIClient(api_key=api_key))
 
 
-def test_init_from_client(api_key):
+@pytest.fixture(scope="session")
+def async_model_no_model_name(api_key):
+    return AsyncOpenAI(AsyncOpenAIClient(api_key=api_key))
+
+
+def test_openai_init_from_client(api_key):
     client = OpenAIClient(api_key=api_key)
 
     # With model name
@@ -192,3 +202,158 @@ def test_openai_batch(model):
         model.batch(
             ["Respond with one word.", "Respond with one word."],
         )
+
+
+def test_openai_async_init_from_client(api_key):
+    client = AsyncOpenAIClient(api_key=api_key)
+
+    # With model name
+    model = outlines.from_openai(client, "gpt-4o")
+    assert isinstance(model, AsyncOpenAI)
+    assert model.client == client
+    assert model.model_name == "gpt-4o"
+
+    # Without model name
+    model = outlines.from_openai(client)
+    assert isinstance(model, AsyncOpenAI)
+    assert model.client == client
+    assert model.model_name is None
+
+
+@pytest.mark.asyncio
+async def test_openai_async_wrong_inference_parameters(async_model):
+    with pytest.raises(TypeError, match="got an unexpected"):
+        await async_model.generate("prompt", foo=10)
+
+
+@pytest.mark.asyncio
+async def test_openai_async_wrong_input_type(async_model, image):
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(TypeError, match="is not available"):
+        await async_model.generate(Foo("prompt"))
+
+    with pytest.raises(ValueError, match="All assets provided must be of type Image"):
+        await async_model.generate(["foo?", Image(image), Video("")])
+
+
+@pytest.mark.asyncio
+async def test_openai_async_wrong_output_type(async_model):
+    class Foo:
+        def __init__(self, foo):
+            self.foo = foo
+
+    with pytest.raises(TypeError, match="is not available"):
+        await async_model.generate("prompt", Foo(1))
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_simple_call(async_model):
+    result = await async_model.generate("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_simple_call_multiple_samples(async_model):
+    result = await async_model.generate("Respond with one word. Not more.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_direct_call(async_model_no_model_name):
+    result = await async_model_no_model_name(
+        "Respond with one word. Not more.",
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_simple_vision(image, async_model):
+    result = await async_model.generate(["What does this logo represent?", Image(image)])
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_chat(image, async_model):
+    result = await async_model.generate(Chat(messages=[
+        {"role": "system", "content": "You are a helpful assistant."},
+        {
+            "role": "user",
+            "content": ["What does this logo represent?", Image(image)]
+        },
+    ]), max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_simple_pydantic(async_model):
+    class Foo(BaseModel):
+        bar: int
+
+    result = await async_model.generate("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_simple_pydantic_refusal(async_model):
+    class Foo(BaseModel):
+        bar: Annotated[str, Field(int, pattern=r"^\d+$")]
+
+    with pytest.raises(TypeError, match="OpenAI does not support your schema"):
+        _ = await async_model.generate("foo?", Foo)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_simple_vision_pydantic(image, async_model):
+    class Logo(BaseModel):
+        name: int
+
+    result = await async_model.generate(["What does this logo represent?", Image(image)], Logo)
+    assert isinstance(result, str)
+    assert "name" in json.loads(result)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_simple_json_schema(async_model):
+    class Foo(BaseModel):
+        bar: int
+
+    schema = json.dumps(Foo.model_json_schema())
+
+    result = await async_model.generate("foo?", json_schema(schema))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_openai_async_streaming(async_model):
+    result = async_model.stream("Respond with a single word.")
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+@pytest.mark.asyncio
+async def test_openai_async_batch(async_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        await async_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )

From 647223ea369916e4754f522b93ecd78d750987f8 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 14 Aug 2025 15:37:39 +0200
Subject: [PATCH 426/505] Fix a bug in OutlinesCoreLogitsProcessor about
 advancing when in final state

---
 outlines/backends/outlines_core.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index 9929bff9f..ff3fe8cb2 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -179,7 +179,13 @@ def process_logits(
         else:
             for i in range(batch_size):
                 last_token_id = self.tensor_adapter.to_scalar(input_ids[i][-1]) # type: ignore
-                if not self._guides[i].is_finished():
+                # This circumvents issue #227 in outlines_core
+                # Ideally, we would be able to advance all the times as the final
+                # state would accept the eos token leading to itself
+                if (
+                    not self._guides[i].is_finished()
+                    or self._guides[i].accepts_tokens([last_token_id])
+                ):
                     self._guides[i].advance(
                         token_id=last_token_id,
                         return_tokens=False

From c0aeb4ad954275a376e46e57a2d3db8aa3eb57ee Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 14 Aug 2025 11:43:12 +0200
Subject: [PATCH 427/505] Update coverage check in GH actions, require 100% for
 all files

---
 .github/workflows/tests.yml | 23 ++---------------------
 1 file changed, 2 insertions(+), 21 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index ce96e223d..309a754d8 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -53,7 +53,7 @@ jobs:
       run: |
         rm -f .coverage*
         uv run coverage erase
-        uv run python -m coverage run --branch --parallel-mode -m pytest -x -m 'not api_call'
+        uv run python -m coverage run --branch --source=outlines --parallel-mode -m pytest -x -m 'not api_call'
     - name: Upload coverage data
       uses: actions/upload-artifact@v4
       with:
@@ -86,37 +86,18 @@ jobs:
         with:
           name: coverage-data
 
-      - name: Determine base branch for comparison
-        id: base-branch
-        run: |
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            # For PRs, use the remote tracking branch
-            COMPARE_BRANCH="origin/${{ github.base_ref }}"
-            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
-          else
-            # For push events, compare against the parent commit
-            COMPARE_BRANCH="${{ github.event.before }}"
-            echo "COMPARE_BRANCH=$COMPARE_BRANCH" >> $GITHUB_ENV
-          fi
-          echo "Using $COMPARE_BRANCH for coverage comparison"
-
-      - name: Fetch base branch for coverage diff
-        run: |
-          git fetch --no-tags --prune origin ${COMPARE_BRANCH#origin/}
-
       - name: Combine coverage & fail if it's <100%.
         run: |
           python -m coverage combine
           python -m coverage html --skip-covered --skip-empty
           python -m coverage xml
-          diff-cover coverage.xml --compare-branch=$COMPARE_BRANCH --markdown-report=coverage.md --fail-under=100 || (cat coverage.md >> $GITHUB_STEP_SUMMARY && exit 1)
+          python -m coverage report --fail-under=100 || (python -m coverage report && exit 1)
 
       - name: Upload HTML report if check failed.
         uses: actions/upload-artifact@v4
         with:
           name: html-report
           path: htmlcov
-          # TODO FIXME: This is only using the last run
           overwrite: true
         if: ${{ failure() }}
 

From 6ae9c28189ecd1fe93841be8d49cf7dcb35dce8c Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 14 Aug 2025 11:47:03 +0200
Subject: [PATCH 428/505] Deprecate jax and tensorflow, remove unused files

---
 outlines/models/transformers.py               | 18 +++++++
 outlines/processors/base_logits_processor.py  |  6 +--
 .../processors/tensor_adapters/__init__.py    |  6 ---
 outlines/processors/tensor_adapters/jax.py    | 50 -------------------
 .../processors/tensor_adapters/tensorflow.py  | 50 -------------------
 tests/models/test_transformers.py             | 34 +++++++------
 tests/processors/test_base_processor.py       | 12 +----
 tests/processors/test_tensor_adapters.py      | 23 +--------
 8 files changed, 42 insertions(+), 157 deletions(-)
 delete mode 100644 outlines/processors/tensor_adapters/jax.py
 delete mode 100644 outlines/processors/tensor_adapters/tensorflow.py

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index c27f55aee..cf677cdde 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -244,11 +244,29 @@ def __init__(
             and isinstance(model, FlaxPreTrainedModel)
         ):
             self.tensor_library_name = "jax"
+            warnings.warn("""
+                Support for `jax` has been deprecated and will be removed in
+                version 1.4.0 of Outlines. Please use `torch` instead.
+                Transformers models using `jax` do not support structured
+                generation.
+                """,
+                DeprecationWarning,
+                stacklevel=2,
+            )
         elif (
             TFPreTrainedModel is not None
             and isinstance(model, TFPreTrainedModel)
         ):
             self.tensor_library_name = "tensorflow"
+            warnings.warn("""
+                Support for `tensorflow` has been deprecated and will be removed in
+                version 1.4.0 of Outlines. Please use `torch` instead.
+                Transformers models using `tensorflow` do not support structured
+                generation.
+                """,
+                DeprecationWarning,
+                stacklevel=2,
+            )
         else:
             self.tensor_library_name = "torch"
 
diff --git a/outlines/processors/base_logits_processor.py b/outlines/processors/base_logits_processor.py
index 9b4e4fe50..42fcbeb51 100644
--- a/outlines/processors/base_logits_processor.py
+++ b/outlines/processors/base_logits_processor.py
@@ -28,8 +28,8 @@ def __init__(self, tensor_library_name: str):
         ----------
         tensor_library_name
             The name of the library to use to manipulate tensors. Possible
-            values are "jax", "mlx", "numpy", "tensorflow" and "torch". You
-            must choose the library that your model is using.
+            values are "mlx", "numpy" and "torch". You must choose the library
+            that your model is using.
         """
         # Temporary fix as torch raises a warning that can cause can an error
         # with python 3.12.
@@ -52,7 +52,7 @@ def reset(self):
         needs to be reset for a new generation.
 
         """
-        pass
+        pass # pragma: no cover
 
     @abstractmethod
     def process_logits(
diff --git a/outlines/processors/tensor_adapters/__init__.py b/outlines/processors/tensor_adapters/__init__.py
index 8e3afe7f6..152166fc3 100644
--- a/outlines/processors/tensor_adapters/__init__.py
+++ b/outlines/processors/tensor_adapters/__init__.py
@@ -2,25 +2,19 @@
 
 from typing import Union
 
-from .jax import JAXTensorAdapter
 from .mlx import MLXTensorAdapter
 from .numpy import NumpyTensorAdapter
-from .tensorflow import TensorFlowTensorAdapter
 from .torch import TorchTensorAdapter
 
 
 tensor_adapters = {
-    "jax": JAXTensorAdapter,
     "mlx": MLXTensorAdapter,
     "numpy": NumpyTensorAdapter,
-    "tensorflow": TensorFlowTensorAdapter,
     "torch": TorchTensorAdapter,
 }
 
 TensorAdapterImplementation = Union[
-    JAXTensorAdapter,
     MLXTensorAdapter,
     NumpyTensorAdapter,
-    TensorFlowTensorAdapter,
     TorchTensorAdapter,
 ]
diff --git a/outlines/processors/tensor_adapters/jax.py b/outlines/processors/tensor_adapters/jax.py
deleted file mode 100644
index 30cd2ecc9..000000000
--- a/outlines/processors/tensor_adapters/jax.py
+++ /dev/null
@@ -1,50 +0,0 @@
-"""Tensor adapter for the `jax` library."""
-
-from outlines.processors.tensor_adapters.base import TensorAdapter
-
-
-class JAXTensorAdapter(TensorAdapter):
-    library_name = "jax"
-
-    def __init__(self):
-        import jax
-
-        self.jax = jax
-
-    def shape(self, tensor):
-        return tensor.shape
-
-    def unsqueeze(self, tensor):
-        return self.jax.numpy.expand_dims(tensor, axis=0)
-
-    def squeeze(self, tensor):
-        return self.jax.numpy.squeeze(tensor, axis=0)
-
-    def to_list(self, tensor):
-        return tensor.tolist()
-
-    def to_scalar(self, tensor):
-        return tensor.item()
-
-    def full_like(self, tensor, fill_value):
-        return self.jax.numpy.full_like(tensor, fill_value)
-
-    def concatenate(self, tensors):
-        return self.jax.numpy.concatenate(tensors, axis=0)
-
-    def to_device(self, tensor, device):
-        return tensor
-
-    def get_device(self, tensor):
-        return None
-
-    def boolean_ones_like(self, tensor):
-        return self.jax.numpy.ones_like(tensor, dtype=bool)
-
-    def apply_mask(self, tensor, mask, value):
-        result = tensor.copy()
-        result = result.at[mask].set(value)
-        return result
-
-    def argsort_descending(self, tensor):
-        return self.jax.numpy.argsort(-tensor)
diff --git a/outlines/processors/tensor_adapters/tensorflow.py b/outlines/processors/tensor_adapters/tensorflow.py
deleted file mode 100644
index 382f847b6..000000000
--- a/outlines/processors/tensor_adapters/tensorflow.py
+++ /dev/null
@@ -1,50 +0,0 @@
-"""Tensor adapter for the `tensorflow` library."""
-
-from outlines.processors.tensor_adapters.base import TensorAdapter
-
-
-class TensorFlowTensorAdapter(TensorAdapter):
-    library_name = "tensorflow"
-
-    def __init__(self):
-        import tensorflow as tf
-
-        self.tf = tf
-
-    def shape(self, tensor):
-        return tensor.shape
-
-    def unsqueeze(self, tensor):
-        return self.tf.expand_dims(tensor, axis=0)
-
-    def squeeze(self, tensor):
-        return self.tf.squeeze(tensor, axis=0)
-
-    def to_list(self, tensor):
-        return tensor.numpy().tolist()
-
-    def to_scalar(self, tensor):
-        return tensor.numpy().item()
-
-    def full_like(self, tensor, fill_value):
-        return self.tf.fill(self.tf.shape(tensor), fill_value)
-
-    def concatenate(self, tensors):
-        return self.tf.concat(tensors, axis=0)
-
-    def get_device(self, tensor):
-        return tensor.device
-
-    def to_device(self, tensor, device):
-        # TensorFlow handles device placement differently
-        with self.tf.device(device):
-            return self.tf.identity(tensor)
-
-    def boolean_ones_like(self, tensor):
-        return self.tf.ones_like(tensor, dtype=self.tf.bool)
-
-    def apply_mask(self, tensor, mask, value):
-        return self.tf.where(mask, self.tf.constant(value, dtype=tensor.dtype), tensor)
-
-    def argsort_descending(self, tensor):
-        return self.tf.argsort(tensor, direction='DESCENDING')
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index a7b0996c1..d98176c22 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -40,25 +40,27 @@ def test_transformers_instantiate_simple():
 
 
 def test_transformers_instantiate_flax_model():
-    model = outlines.from_transformers(
-        transformers.FlaxAutoModelForCausalLM.from_pretrained(TEST_MODEL),
-        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-    )
-    assert isinstance(model, Transformers)
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert isinstance(model.type_adapter, TransformersTypeAdapter)
-    assert model.tensor_library_name == "jax"
+    with pytest.warns(DeprecationWarning):
+        model = outlines.from_transformers(
+            transformers.FlaxAutoModelForCausalLM.from_pretrained(TEST_MODEL),
+            transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+        )
+        assert isinstance(model, Transformers)
+        assert isinstance(model.tokenizer, TransformerTokenizer)
+        assert isinstance(model.type_adapter, TransformersTypeAdapter)
+        assert model.tensor_library_name == "jax"
 
 
 def test_transformers_instantiate_tensorflow_model():
-    model = outlines.from_transformers(
-        transformers.TFAutoModelForCausalLM.from_pretrained(TEST_MODEL),
-        transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-    )
-    assert isinstance(model, Transformers)
-    assert isinstance(model.tokenizer, TransformerTokenizer)
-    assert isinstance(model.type_adapter, TransformersTypeAdapter)
-    assert model.tensor_library_name == "tensorflow"
+    with pytest.warns(DeprecationWarning):
+        model = outlines.from_transformers(
+            transformers.TFAutoModelForCausalLM.from_pretrained(TEST_MODEL),
+            transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
+        )
+        assert isinstance(model, Transformers)
+        assert isinstance(model.tokenizer, TransformerTokenizer)
+        assert isinstance(model.type_adapter, TransformersTypeAdapter)
+        assert model.tensor_library_name == "tensorflow"
 
 
 def test_transformers_instantiate_mamba():
diff --git a/tests/processors/test_base_processor.py b/tests/processors/test_base_processor.py
index 3b729410a..bcdd33308 100644
--- a/tests/processors/test_base_processor.py
+++ b/tests/processors/test_base_processor.py
@@ -1,6 +1,5 @@
 from typing import List
 
-import jax.numpy as jnp
 import numpy as np
 import pytest
 import torch
@@ -14,7 +13,7 @@
     HAS_MLX = False
 
 
-libraries = ["numpy", "torch", "jax"]
+libraries = ["numpy", "torch"]
 if HAS_MLX:
     libraries.append("mlx")
 
@@ -43,14 +42,6 @@
         (torch.tensor([1, 2], dtype=torch.float32), torch.tensor([[1, 2], [3, 4]], dtype=torch.int32), AssertionError),
         (torch.tensor([[[1, 2]]], dtype=torch.float32), torch.tensor([[[1, 2]]], dtype=torch.int32), ValueError),
     ],
-    "jax": [
-        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), None),
-        (jnp.array([[1, 2], [3, 4]], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), None),
-        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2]], dtype=jnp.int32), None),
-        (jnp.array([[1, 2]], dtype=jnp.float32), jnp.array([1, 2], dtype=jnp.int32), AssertionError),
-        (jnp.array([1, 2], dtype=jnp.float32), jnp.array([[1, 2], [3, 4]], dtype=jnp.int32), AssertionError),
-        (jnp.array([[[1, 2]]], dtype=jnp.float32), jnp.array([[[1, 2]]], dtype=jnp.int32), ValueError),
-    ],
 }
 if HAS_MLX:
     arrays["mlx"] = [
@@ -76,6 +67,7 @@ def test_base_logits_processor_init(library):
     assert processor.tensor_adapter is not None
     with pytest.raises(NotImplementedError):
         processor = MockLogitsProcessor("foo")
+        processor.reset()
 
 
 @pytest.mark.parametrize("library", libraries)
diff --git a/tests/processors/test_tensor_adapters.py b/tests/processors/test_tensor_adapters.py
index 4a5bcc1e1..064afe85d 100644
--- a/tests/processors/test_tensor_adapters.py
+++ b/tests/processors/test_tensor_adapters.py
@@ -1,18 +1,13 @@
 import pytest
 from pytest import mark
 
-import jax
-import jax.numpy as jnp
 import numpy as np
-import tensorflow as tf
 import torch
 
 from outlines.processors.tensor_adapters import (
     NumpyTensorAdapter,
     TorchTensorAdapter,
     MLXTensorAdapter,
-    JAXTensorAdapter,
-    TensorFlowTensorAdapter,
 )
 
 try:
@@ -25,26 +20,19 @@
 
 
 adapters = {
-    "jax": JAXTensorAdapter(),
     "numpy": NumpyTensorAdapter(),
-    "tensorflow": TensorFlowTensorAdapter(),
     "torch": TorchTensorAdapter(),
 }
 if HAS_MLX:
     adapters["mlx"] = MLXTensorAdapter()
 
-frameworks = ["jax", "numpy", "tensorflow", "torch", "mlx"]
+frameworks = ["numpy", "torch", "mlx"]
 
 def create_tensor(framework, shape, dtype=None):
     if framework == "torch":
         return torch.randn(*shape)
     elif framework == "numpy":
         return np.random.randn(*shape)
-    elif framework == "jax":
-        key = jax.random.PRNGKey(0)
-        return jax.random.normal(key, shape=shape)
-    elif framework == "tensorflow":
-        return tf.random.normal(shape)
     elif framework == "mlx":
         if not HAS_MLX:
             pytest.skip("MLX not available")
@@ -55,10 +43,6 @@ def compare_tensors(framework, tensor1, tensor2):
         return torch.allclose(tensor1, tensor2)
     elif framework == "numpy":
         return np.array_equal(tensor1, tensor2)
-    elif framework == "jax":
-        return jax.numpy.array_equal(tensor1, tensor2)
-    elif framework == "tensorflow":
-        return tf.reduce_all(tf.equal(tensor1, tensor2))
     elif framework == "mlx":
         if not HAS_MLX:
             pytest.skip("MLX not available")
@@ -243,11 +227,6 @@ def test_tensor_adapter_apply_mask(framework):
         mask = torch.randn(2, 3) > 0
     elif framework == "numpy":
         mask = np.random.randn(2, 3) > 0
-    elif framework == "jax":
-        key = jax.random.PRNGKey(0)
-        mask = jax.random.normal(key, shape=(2, 3)) > 0
-    elif framework == "tensorflow":
-        mask = tf.random.normal((2, 3)) > 0
     elif framework == "mlx":
         if not HAS_MLX:
             pytest.skip("MLX not available")

From 778cd105d7fe76365438a6ed5ba5fce1d2398e76 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 14 Aug 2025 11:47:25 +0200
Subject: [PATCH 429/505] Fix coverage issues, improve testing logits
 processors

---
 outlines/backends/llguidance.py       |   3 -
 outlines/backends/outlines_core.py    |   8 +-
 outlines/backends/xgrammar.py         |   6 +-
 outlines/caching.py                   |   2 +-
 tests/backends/test_backends_utils.py |  70 +++++++++++++++
 tests/backends/test_llguidance.py     | 120 +++++++++++++++-----------
 tests/backends/test_outlines_core.py  |  84 +++++++++++++++---
 tests/backends/test_xgrammar.py       | 106 +++++++++++++++++++----
 8 files changed, 308 insertions(+), 91 deletions(-)
 create mode 100644 tests/backends/test_backends_utils.py

diff --git a/outlines/backends/llguidance.py b/outlines/backends/llguidance.py
index 362031adf..5b6d168b1 100644
--- a/outlines/backends/llguidance.py
+++ b/outlines/backends/llguidance.py
@@ -37,9 +37,6 @@ def __init__(
             The name of the tensor library used by the model
 
         """
-        if tensor_library_name not in SUPPORTED_TENSOR_LIBRARIES:
-            raise TypeError(f"Unsupported tensor library: {tensor_library_name}")
-
         self.is_first_token = True
         self.grammar = grammar
         self.llg_tokenizer = llg_tokenizer
diff --git a/outlines/backends/outlines_core.py b/outlines/backends/outlines_core.py
index ff3fe8cb2..23457fd5a 100644
--- a/outlines/backends/outlines_core.py
+++ b/outlines/backends/outlines_core.py
@@ -70,7 +70,7 @@ def _setup(self, batch_size: int, vocab_size: int) -> None:
             self.allocate_token_bitmask = allocate_token_bitmask
             self.bias_logits = self._bias_logits_numpy
 
-        elif self.tensor_library_name == "mlx":
+        elif self.tensor_library_name == "mlx": # pragma: no cover
             from outlines_core.kernels.mlx import (
                 allocate_token_bitmask
             )
@@ -78,7 +78,7 @@ def _setup(self, batch_size: int, vocab_size: int) -> None:
             self.allocate_token_bitmask = allocate_token_bitmask
             self.bias_logits = self._bias_logits_mlx
 
-        else:
+        else: # pragma: no cover
             raise ValueError(
                 f"Unsupported tensor library: {self.tensor_library_name}"
             )
@@ -217,13 +217,13 @@ def __init__(self, model: SteerableModel):
             eos_token_id = tokenizer.eos_token_id
             eos_token = tokenizer.eos_token
             token_to_str = tokenizer.convert_token_to_string
-        elif isinstance(model, MLXLM):
+        elif isinstance(model, MLXLM): # pragma: no cover
             tokenizer = model.mlx_tokenizer # type: ignore
             vocabulary = tokenizer.get_vocab()
             eos_token_id = tokenizer.eos_token_id
             eos_token = tokenizer.eos_token
             token_to_str = lambda token: tokenizer.convert_tokens_to_string([token]) # type: ignore
-        else:
+        else: # pragma: no cover
             raise ValueError(f"Unsupported model type: {type(model)}")
 
         self.eos_token_id = eos_token_id
diff --git a/outlines/backends/xgrammar.py b/outlines/backends/xgrammar.py
index c316477d6..4b6a59742 100644
--- a/outlines/backends/xgrammar.py
+++ b/outlines/backends/xgrammar.py
@@ -39,7 +39,7 @@ def _setup(self, batch_size: int, vocab_size: int) -> None:
         """Setup the logits processor for a new generation."""
         if self.tensor_library_name == "torch":
             self._bias_logits = self._bias_logits_torch
-        elif self.tensor_library_name == "mlx":
+        elif self.tensor_library_name == "mlx": # pragma: no cover
             self._bias_logits = self._bias_logits_mlx
         else: # pragma: no cover
             raise ValueError(
@@ -101,7 +101,7 @@ def process_logits(
             self.is_first_token = False
         else:
             for i in range(batch_size):
-                if not self._matchers[i].is_terminated():
+                if not self._matchers[i].is_terminated(): # pragma: no cover
                     last_token_id = self.tensor_adapter.to_scalar(
                         input_ids[i][-1] # type: ignore
                     )
@@ -125,7 +125,7 @@ def __init__(self, model: SteerableModel):
 
         if isinstance(model, Transformers):
             tokenizer = model.hf_tokenizer
-        elif isinstance(model, MLXLM):
+        elif isinstance(model, MLXLM): # pragma: no cover
             tokenizer = model.mlx_tokenizer._tokenizer
         else: # pragma: no cover
             raise ValueError(
diff --git a/outlines/caching.py b/outlines/caching.py
index f134f216c..47d744e4b 100644
--- a/outlines/caching.py
+++ b/outlines/caching.py
@@ -61,7 +61,7 @@ def get_cache():
         cache_dir = outlines_cache_dir
     elif xdg_cache_home:  # pragma: no cover
         cache_dir = os.path.join(xdg_cache_home, ".cache", "outlines")
-    elif home_dir != "/":
+    elif home_dir != "/": # pragma: no cover
         cache_dir = os.path.join(home_dir, ".cache", "outlines")
     else:  # pragma: no cover
         # home_dir may be / inside a docker container without existing user
diff --git a/tests/backends/test_backends_utils.py b/tests/backends/test_backends_utils.py
new file mode 100644
index 000000000..69d4710bf
--- /dev/null
+++ b/tests/backends/test_backends_utils.py
@@ -0,0 +1,70 @@
+import torch
+import numpy as np
+
+
+def simulate_model_calling_processor(processor, tensor_library_name, vocabulary_size, eos_token_id, batch_size):
+    if tensor_library_name == "torch":
+        tensor_adapter = TorchTensorAdapter()
+    elif tensor_library_name == "numpy":
+        tensor_adapter = NumpyTensorAdapter()
+    elif tensor_library_name == "mlx":
+        tensor_adapter = MLXTensorAdapter()
+
+    processor.reset()
+    i = 0
+    input_ids = tensor_adapter.randint(0, vocabulary_size, (batch_size, 10))
+    while True:
+        i += 1
+        logits = tensor_adapter.randn((batch_size, vocabulary_size))
+        output = processor(input_ids, logits)
+        assert output.shape == (batch_size, vocabulary_size)
+        if all(input_ids[:, -1] == eos_token_id):
+            break
+        input_ids = tensor_adapter.add_token_inputs_ids(input_ids, output)
+        print(input_ids)
+        if i > 20:
+            break
+    return input_ids[:, 10:]
+
+class TorchTensorAdapter():
+    def randn(self, shape):
+        return torch.randn(*shape)
+
+    def randint(self, low, high, size):
+        return torch.randint(low, high, size)
+
+    def add_token_inputs_ids(self, input_ids, logits):
+        next_token_ids = torch.argmax(logits, dim=-1)
+        input_ids = torch.cat([input_ids, next_token_ids.unsqueeze(-1)], dim=-1)
+        return input_ids
+
+
+class NumpyTensorAdapter():
+    def randn(self, shape):
+        return np.random.randn(*shape)
+
+    def randint(self, low, high, size):
+        return np.random.randint(low, high, size)
+
+    def add_token_inputs_ids(self, input_ids, logits):
+        next_token_ids = np.argmax(logits, axis=-1)
+        print("next_token_ids",next_token_ids)
+        input_ids = np.concatenate([input_ids, next_token_ids[..., None]], axis=-1)
+        return input_ids
+
+
+class MLXTensorAdapter():
+    def __init__(self):
+        import mlx
+        self.mlx = mlx
+
+    def randn(self, shape):
+        return self.mlx.random.randn(*shape)
+
+    def randint(self, low, high, size):
+        return self.mlx.random.randint(low, high, size)
+
+    def add_token_inputs_ids(self, input_ids, logits):
+        next_token_ids = self.mlx.argmax(logits, axis=-1)
+        input_ids = self.mlx.concatenate([input_ids, next_token_ids[..., None]], axis=-1)
+        return input_ids
diff --git a/tests/backends/test_llguidance.py b/tests/backends/test_llguidance.py
index a005e1854..fb5faea91 100644
--- a/tests/backends/test_llguidance.py
+++ b/tests/backends/test_llguidance.py
@@ -1,9 +1,8 @@
-import pytest
+import re
 
 import llama_cpp
-import llguidance.hf
-import numpy as np
-import torch
+import llguidance
+import pytest
 import transformers
 from llguidance import LLTokenizer
 
@@ -12,9 +11,9 @@
     LLGuidanceBackend,
     LLGuidanceLogitsProcessor
 )
+from tests.backends.test_backends_utils import simulate_model_calling_processor
 
 try:
-    import mlx.core as mx
     import mlx_lm
     HAS_MLX = True
 except ImportError:
@@ -40,20 +39,6 @@ def model_mlxlm():
         *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
     )
 
-@pytest.fixture
-def llg_tokenizer():
-    return llguidance.hf.from_tokenizer(
-        transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
-    )
-
-@pytest.fixture
-def llg_grammar_spec():
-    return (
-        '{"grammars": [{ "json_schema": {"type": "object", "properties":'
-        + ' {"name": {"type": "string"}, "age": {"type": "integer"}}, "requ'
-        + 'ired": ["name", "age"], "additionalProperties": false} }] }'
-    )
-
 @pytest.fixture
 def json_schema():
     return (
@@ -97,42 +82,61 @@ def cfg_ebnf():
 """
 
 
-def test_llguidance_processor_torch(llg_grammar_spec, llg_tokenizer):
-    processor = LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "torch")
-    logits = torch.randn(2, llg_tokenizer.vocab_size)
-    input_ids = torch.randint(0, llg_tokenizer.vocab_size, (2, 10))
-    output = processor(input_ids, logits)
-    assert output.shape == (2, llg_tokenizer.vocab_size)
-    processor(input_ids, logits)
-
+def test_llguidance_processor_torch(regex):
+    model = model_transformers()
+    tokenizer = model.tokenizer
+    hf_tokenizer = model.hf_tokenizer
+    llg_tokenizer = LLGuidanceBackend(model).llg_tokenizer
+    grammar_spec = llguidance.grammar_from("regex", regex)
+    processor = LLGuidanceLogitsProcessor(grammar_spec, llg_tokenizer, "torch")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "torch",
+            len(tokenizer.get_vocab()),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, hf_tokenizer.decode(input_ids[0]))
+        assert re.match(regex, hf_tokenizer.decode(input_ids[1]))
+
+
+def test_llguidance_processor_numpy(regex):
+    model = model_llamacpp()
+    tokenizer = model.tokenizer
+    llg_tokenizer = LLGuidanceBackend(model).llg_tokenizer
+    grammar_spec = llguidance.grammar_from("regex", regex)
+    processor = LLGuidanceLogitsProcessor(grammar_spec, llg_tokenizer, "numpy")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "numpy",
+            len(tokenizer.vocabulary),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, tokenizer.decode(input_ids[0])[0])
+        assert re.match(regex, tokenizer.decode(input_ids[1])[0])
 
-def test_llguidance_processor_numpy(llg_grammar_spec, llg_tokenizer):
-    processor = LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "numpy")
-    logits = np.random.randn(2, llg_tokenizer.vocab_size)
-    input_ids = np.random.randint(0, llg_tokenizer.vocab_size, (2, 10))
-    output = processor(input_ids, logits)
-    assert output.shape == (2, llg_tokenizer.vocab_size)
-    processor(input_ids, logits)
 
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
-def test_llguidance_processor_mlx(llg_grammar_spec, llg_tokenizer):
-    processor = LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "mlx")
-    logits = mx.random.normal((2, llg_tokenizer.vocab_size))
-    input_ids = mx.random.randint(0, llg_tokenizer.vocab_size, (2, 10))
-    output = processor(input_ids, logits)
-    assert output.shape == (2, llg_tokenizer.vocab_size)
-    processor(input_ids, logits)
-
-
-def test_llguidance_processor_tensorflow(llg_grammar_spec, llg_tokenizer):
-    with pytest.raises(TypeError):
-        LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "tensorflow")
-
-
-def test_llguidance_processor_jax(llg_grammar_spec, llg_tokenizer):
-    with pytest.raises(TypeError):
-        LLGuidanceLogitsProcessor(llg_grammar_spec, llg_tokenizer, "jax")
+def test_llguidance_processor_mlx(regex):
+    model = model_mlxlm()
+    tokenizer = model.mlx_tokenizer
+    llg_tokenizer = LLGuidanceBackend(model).llg_tokenizer
+    grammar_spec = llguidance.grammar_from("regex", regex)
+    processor = LLGuidanceLogitsProcessor(grammar_spec, llg_tokenizer, "mlx")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "mlx",
+            len(tokenizer.vocabulary),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, tokenizer.decode(input_ids[0]))
+        assert re.match(regex, tokenizer.decode(input_ids[1]))
 
 
 models = [
@@ -155,7 +159,6 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg_
     generator = outlines.Generator(model, backend="llguidance", processor=processor)
     response = generator("Hello, how are you?")
     assert response[0] == "{"
-    assert "name" in response
 
     # regex
     processor = backend.get_regex_logits_processor(regex)
@@ -184,3 +187,16 @@ def test_llguidance_backend(model, tensor_library_name, json_schema, regex, cfg_
     generator = outlines.Generator(model, backend="llguidance", processor=processor)
     response = generator("Hello, how are you?")
     assert response == "yes" or response == "no"
+
+    # batch + multiple generations
+    processor = backend.get_json_schema_logits_processor(json_schema)
+    generator = outlines.Generator(model, backend="llguidance", processor=processor)
+    for _ in range(2):
+        if tensor_library_name == "torch":
+            response = generator.batch(["Create a character", "Hello, how are you?"], max_new_tokens=200)
+            assert len(response) == 2
+            for r in response:
+                assert r[0] == "{"
+        else:
+            response = generator("Create a character", max_tokens=20)
+            assert response[0] == "{"
diff --git a/tests/backends/test_outlines_core.py b/tests/backends/test_outlines_core.py
index e62951022..ce3242457 100644
--- a/tests/backends/test_outlines_core.py
+++ b/tests/backends/test_outlines_core.py
@@ -1,15 +1,16 @@
-import pytest
+import re
 
 import llama_cpp
-from outlines_core import Vocabulary
+import pytest
 import transformers
+from outlines_core import Index, Vocabulary
 
 import outlines
 from outlines.backends.outlines_core import (
     OutlinesCoreBackend,
     OutlinesCoreLogitsProcessor
 )
-
+from tests.backends.test_backends_utils import simulate_model_calling_processor
 
 try:
     import mlx_lm
@@ -73,6 +74,62 @@ def cfg():
 """
 
 
+def test_outlines_core_processor_torch(regex):
+    model = model_transformers()
+    tokenizer = model.tokenizer
+    hf_tokenizer = model.hf_tokenizer
+    backend = OutlinesCoreBackend(model)
+    index = Index(regex, backend.vocabulary)
+    processor = OutlinesCoreLogitsProcessor(index, "torch")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "torch",
+            len(tokenizer.get_vocab()),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, hf_tokenizer.decode(input_ids[0]))
+        assert re.match(regex, hf_tokenizer.decode(input_ids[1]))
+
+
+def test_outlines_core_processor_numpy(regex):
+    model = model_llamacpp()
+    tokenizer = model.tokenizer
+    backend = OutlinesCoreBackend(model)
+    index = Index(regex, backend.vocabulary)
+    processor = OutlinesCoreLogitsProcessor(index, "numpy")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "numpy",
+            len(tokenizer.vocabulary),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, tokenizer.decode(input_ids[0])[0])
+        assert re.match(regex, tokenizer.decode(input_ids[1])[0])
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_outlines_core_processor_mlx():
+    model = model_mlxlm()
+    tokenizer = model.mlx_tokenizer
+    backend = OutlinesCoreBackend(model)
+    index = Index(r"[0-9]{3}", backend.vocabulary)
+    processor = OutlinesCoreLogitsProcessor(index, "mlx")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "mlx",
+            len(tokenizer.vocabulary),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, tokenizer.decode(input_ids[0]))
+        assert re.match(regex, tokenizer.decode(input_ids[1]))
+
+
 models = [
     (model_transformers(), "torch"),
     (model_llamacpp(), "numpy"),
@@ -109,12 +166,17 @@ def test_outlines_core_backend(model, tensor_library_name, json_schema, regex, c
     ):
         backend.get_cfg_logits_processor(cfg)
 
-    # multiple generations
-    processor = backend.get_regex_logits_processor(regex)
+    # batch + multiple generations
+    processor = backend.get_json_schema_logits_processor(json_schema)
     generator = outlines.Generator(model, backend="outlines_core", processor=processor)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
-    response = generator("Hello, how are you?")
-    assert len(response) == 3
-    assert int(response)
+    for _ in range(2):
+        if tensor_library_name == "torch":
+            response = generator.batch(["Create a character", "Hello, how are you?"], max_new_tokens=200)
+            assert len(response) == 2
+            for r in response:
+                assert r[0] == "{"
+                assert "name" in r
+        else:
+            response = generator("Create a character", max_tokens=20)
+            assert response[0] == "{"
+            assert "name" in response
diff --git a/tests/backends/test_xgrammar.py b/tests/backends/test_xgrammar.py
index 382c508d7..e25f66508 100644
--- a/tests/backends/test_xgrammar.py
+++ b/tests/backends/test_xgrammar.py
@@ -1,13 +1,13 @@
-import pytest
+import re
 
 import llama_cpp
+import outlines
+import pytest
 import transformers
-from xgrammar import GrammarCompiler
+from xgrammar import GrammarCompiler, TokenizerInfo
 
-from outlines.backends.xgrammar import XGrammarLogitsProcessor
-
-import outlines
-from outlines.backends.xgrammar import XGrammarBackend
+from outlines.backends.xgrammar import XGrammarBackend, XGrammarLogitsProcessor
+from tests.backends.test_backends_utils import simulate_model_calling_processor
 
 try:
     import mlx_lm
@@ -16,14 +16,12 @@
     HAS_MLX = False
 
 
-@pytest.fixture
 def model_transformers():
     return outlines.from_transformers(
         transformers.AutoModelForCausalLM.from_pretrained("erwanf/gpt2-mini"),
         transformers.AutoTokenizer.from_pretrained("erwanf/gpt2-mini"),
     )
 
-@pytest.fixture
 def model_llamacpp():
     return outlines.from_llamacpp(
         llama_cpp.Llama.from_pretrained(
@@ -32,12 +30,20 @@ def model_llamacpp():
         )
     )
 
-@pytest.fixture
 def model_mlxlm():
     return outlines.from_mlxlm(
         *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
     )
 
+@pytest.fixture
+def tokenizer_info():
+    tokenizer = model_transformers().hf_tokenizer
+    tokenizer_info = TokenizerInfo.from_huggingface(
+        tokenizer,
+        vocab_size=len(tokenizer.get_vocab())
+    )
+    return tokenizer_info
+
 @pytest.fixture
 def json_schema():
     return (
@@ -58,15 +64,66 @@ def cfg():
 """
 
 
-def test_xgrammar_backend(model_transformers, json_schema, regex, cfg):
+def test_xgr_processor_torch(regex):
+    model = model_transformers()
+    tokenizer = model.tokenizer
+    hf_tokenizer = model.hf_tokenizer
+    tokenizer_info = TokenizerInfo.from_huggingface(
+        hf_tokenizer,
+        vocab_size=len(hf_tokenizer.get_vocab())
+    )
+    grammar_compiler = GrammarCompiler(tokenizer_info)
+    compiled_grammar = grammar_compiler.compile_regex(regex)
+    processor = XGrammarLogitsProcessor(compiled_grammar, "torch")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "torch",
+            len(tokenizer.get_vocab()),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, hf_tokenizer.decode(input_ids[0]))
+        assert re.match(regex, hf_tokenizer.decode(input_ids[1]))
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_xgr_processor_mlx(tokenizer_info):
+    model = model_mlxlm()
+    tokenizer = model.mlx_tokenizer
+    tokenizer_info = TokenizerInfo.from_huggingface(
+        tokenizer,
+        vocab_size=len(tokenizer.get_vocab())
+    )
+    grammar_compiler = GrammarCompiler(tokenizer_info)
+    compiled_grammar = grammar_compiler.compile_regex(regex)
+    processor = XGrammarLogitsProcessor(compiled_grammar, "mlx")
+    for _ in range(2):
+        input_ids = simulate_model_calling_processor(
+            processor,
+            "mlx",
+            len(tokenizer.get_vocab()),
+            tokenizer.eos_token_id,
+            2
+        )
+        assert re.match(regex, tokenizer.decode(input_ids[0]))
+        assert re.match(regex, tokenizer.decode(input_ids[1]))
+
+
+models = [(model_transformers(), "torch")]
+if HAS_MLX:
+    models.append((model_mlxlm(), "mlx"))
+
+@pytest.mark.parametrize("model, tensor_library_name", models)
+def test_xgrammar_backend(model, tensor_library_name, json_schema, regex, cfg):
     # initialization
-    backend = XGrammarBackend(model_transformers)
+    backend = XGrammarBackend(model)
     assert isinstance(backend.grammar_compiler, GrammarCompiler)
 
     # json schema
     processor = backend.get_json_schema_logits_processor(json_schema)
     assert isinstance(processor, XGrammarLogitsProcessor)
-    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
+    generator = outlines.Generator(model, backend="xgrammar", processor=processor)
     response = generator("Hello, how are you?")
     assert response[0] == "{"
     assert "name" in response
@@ -74,7 +131,7 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg):
     # regex
     processor = backend.get_regex_logits_processor(regex)
     assert isinstance(processor, XGrammarLogitsProcessor)
-    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
+    generator = outlines.Generator(model, backend="xgrammar", processor=processor)
     response = generator("Hello, how are you?")
     assert len(response) == 3
     assert int(response)
@@ -82,14 +139,29 @@ def test_xgrammar_backend(model_transformers, json_schema, regex, cfg):
     # cfg
     processor = backend.get_cfg_logits_processor(cfg)
     assert isinstance(processor, XGrammarLogitsProcessor)
-    generator = outlines.Generator(model_transformers, backend="xgrammar", processor=processor)
+    generator = outlines.Generator(model, backend="xgrammar", processor=processor)
     response = generator("Hello, how are you?")
     assert response == "yes" or response == "no"
 
-
-def test_xgrammar_backend_invalid_model(model_llamacpp):
+    # batch + multiple generations
+    processor = backend.get_json_schema_logits_processor(json_schema)
+    generator = outlines.Generator(model, backend="xgrammar", processor=processor)
+    for _ in range(2):
+        if tensor_library_name == "torch":
+            response = generator.batch(["Create a character", "Hello, how are you?"], max_new_tokens=200)
+            assert len(response) == 2
+            for r in response:
+                assert r[0] == "{"
+                assert "name" in r
+        else:
+            response = generator("Create a character", max_tokens=20)
+            assert response[0] == "{"
+            assert "name" in response
+
+
+def test_xgrammar_backend_invalid_model():
     with pytest.raises(
         ValueError,
         match="The xgrammar backend only supports Transformers and MLXLM models",
     ):
-        XGrammarBackend(model_llamacpp)
+        XGrammarBackend(model_llamacpp())

From 5f2649fb4ac51a2d598a2fe46014d805f8f37190 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 22 Aug 2025 16:49:47 +0200
Subject: [PATCH 430/505] Update workflow to test API models, use manual
 trigger

---
 ...tests_nightly.yml => tests_api_models.yml} | 20 ++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)
 rename .github/workflows/{tests_nightly.yml => tests_api_models.yml} (83%)

diff --git a/.github/workflows/tests_nightly.yml b/.github/workflows/tests_api_models.yml
similarity index 83%
rename from .github/workflows/tests_nightly.yml
rename to .github/workflows/tests_api_models.yml
index 183cce207..8f899b706 100644
--- a/.github/workflows/tests_nightly.yml
+++ b/.github/workflows/tests_api_models.yml
@@ -1,13 +1,16 @@
-name: Nightly Tests
+name: API Model Tests
 
-# Run every day at 6:00 AM UTC
 on:
-  schedule:
-    - cron: "0 6 * * *"
+  workflow_dispatch:
+    inputs:
+      branch_name:
+        description: 'Name of the branch to test'
+        required: true
+        type: string
 
 jobs:
   tests:
-    name: Run the tests
+    name: Run API Model Tests
     runs-on: ubuntu-latest
     env:
       OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
@@ -21,21 +24,27 @@ jobs:
 
     steps:
     - uses: actions/checkout@v3
+      with:
+        ref: ${{ github.event.inputs.branch_name }}
+
     - name: Set up Python ${{ matrix.python-version }}
       uses: actions/setup-python@v4
       with:
         python-version: ${{ matrix.python-version }}
+
     - name: Install Ollama
       run: |
         curl -fsSL https://ollama.com/install.sh | sh
         ollama --version
         ollama pull tinyllama
+
     - name: Set up test environment
       run: |
         python -m pip install --upgrade pip
         pip install uv
         uv venv
         uv pip install -e .[test]
+
     - name: Create matrix id
       id: matrix-id
       env:
@@ -45,6 +54,7 @@ jobs:
         export MATRIX_ID=`echo $MATRIX_CONTEXT | md5sum | cut -c 1-32`
         echo $MATRIX_ID
         echo "::set-output name=id::$MATRIX_ID"
+
     - name: Run tests
       run: |
         uv run pytest -x --cov=outlines -m 'api_call' --ignore=tests/models/test_dottxt.py

From 6469218d8ad78942f4f9e7f3d668c61f0d55b51b Mon Sep 17 00:00:00 2001
From: the-vampiire <25523682+the-vampiire@users.noreply.github.com>
Date: Wed, 20 Aug 2025 17:35:07 -0500
Subject: [PATCH 431/505] fix typings: add explicit package and root level
 exports

---
 outlines/__init__.py                          | 60 ++++++------------
 outlines/backends/__init__.py                 | 14 +++++
 outlines/models/__init__.py                   | 45 ++++++++++++-
 outlines/processors/__init__.py               |  4 ++
 .../processors/tensor_adapters/__init__.py    |  9 ++-
 outlines/types/__init__.py                    | 63 ++++++++++++++++---
 outlines/types/locale/__init__.py             |  4 ++
 7 files changed, 146 insertions(+), 53 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index eaef0ef71..e9753f7a5 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -1,31 +1,25 @@
 """Outlines is a Generative Model Programming Framework."""
 
-import outlines.grammars
-import outlines.inputs
-import outlines.processors
-import outlines.types
-from outlines.applications import Application
-from outlines.caching import clear_cache, disable_cache, get_cache
-from outlines.generator import Generator
-from outlines.inputs import Audio, Image, Video
-from outlines.models import (
-    from_dottxt,
-    from_openai,
-    from_transformers,
-    from_gemini,
-    from_anthropic,
-    from_ollama,
-    from_llamacpp,
-    from_mlxlm,
-    from_sglang,
-    from_tgi,
-    from_vllm,
-    from_vllm_offline,
-)
-from outlines.templates import Template
-from outlines.types import regex, json_schema, cfg
-from outlines.templates import Vision
-
+# re-export on top-level namespace
+from outlines import grammars as grammars
+from outlines import inputs as inputs
+from outlines import models as models
+from outlines import processors as processors
+from outlines import types as types
+from outlines.applications import Application as Application
+from outlines.caching import clear_cache as clear_cache
+from outlines.caching import disable_cache as disable_cache
+from outlines.caching import get_cache as get_cache
+from outlines.generator import Generator as Generator
+from outlines.inputs import Audio as Audio
+from outlines.inputs import Image as Image
+from outlines.inputs import Video as Video
+from outlines.models import *  # noqa: F403
+from outlines.templates import Template as Template
+from outlines.templates import Vision as Vision
+from outlines.types import cfg as cfg
+from outlines.types import json_schema as json_schema
+from outlines.types import regex as regex
 
 model_list = [
     "from_anthropic",
@@ -39,17 +33,3 @@
     "from_transformers",
     "from_vllm",
 ]
-
-__all__ = [
-    "clear_cache",
-    "disable_cache",
-    "get_cache",
-    "Application",
-    "Generator",
-    "regex",
-    "json_schema",
-    "cfg",
-    "prompt",
-    "Template",
-    "grammars",
-] + model_list
diff --git a/outlines/backends/__init__.py b/outlines/backends/__init__.py
index 680e54959..a7911fc29 100644
--- a/outlines/backends/__init__.py
+++ b/outlines/backends/__init__.py
@@ -9,6 +9,20 @@
 from outlines.backends.xgrammar import XGrammarBackend
 from outlines.models import SteerableModel
 
+__all__ = [
+    "BaseBackend",
+    "LogitsProcessorType",
+    "LLGuidanceBackend",
+    "OutlinesCoreBackend",
+    "XGrammarBackend",
+    "SteerableModel",
+    "CFG_DEFAULT_BACKEND",
+    "JSON_SCHEMA_DEFAULT_BACKEND",
+    "REGEX_DEFAULT_BACKEND",
+    "get_json_schema_logits_processor",
+    "get_regex_logits_processor",
+    "get_cfg_logits_processor",
+]
 
 CFG_DEFAULT_BACKEND = "llguidance"
 JSON_SCHEMA_DEFAULT_BACKEND = "outlines_core"
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index 1bd7aa127..d4b9cd842 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -17,15 +17,15 @@
 from .ollama import AsyncOllama, Ollama, from_ollama
 from .openai import AsyncOpenAI, OpenAI, from_openai
 from .sglang import AsyncSGLang, SGLang, from_sglang
-from .tgi import AsyncTGI, TGI, from_tgi
+from .tgi import TGI, AsyncTGI, from_tgi
 from .transformers import (
     Transformers,
-    TransformerTokenizer,
     TransformersMultiModal,
+    TransformerTokenizer,
     from_transformers,
 )
+from .vllm import VLLM, AsyncVLLM, from_vllm
 from .vllm_offline import VLLMOffline, from_vllm_offline
-from .vllm import AsyncVLLM, VLLM, from_vllm
 
 SteerableModel = Union[LlamaCpp, MLXLM, Transformers]
 BlackBoxModel = Union[
@@ -46,3 +46,42 @@
     AsyncSGLang,
     AsyncVLLM,
 ]
+
+__all__ = [
+    "Anthropic",
+    "from_anthropic",
+    "Model",
+    "ModelTypeAdapter",
+    "Dottxt",
+    "from_dottxt",
+    "Gemini",
+    "from_gemini",
+    "LlamaCpp",
+    "from_llamacpp",
+    "MLXLM",
+    "from_mlxlm",
+    "AsyncOllama",
+    "Ollama",
+    "from_ollama",
+    "AsyncOpenAI",
+    "OpenAI",
+    "from_openai",
+    "AsyncSGLang",
+    "SGLang",
+    "from_sglang",
+    "AsyncTGI",
+    "TGI",
+    "from_tgi",
+    "Transformers",
+    "TransformerTokenizer",
+    "TransformersMultiModal",
+    "from_transformers",
+    "VLLMOffline",
+    "from_vllm_offline",
+    "AsyncVLLM",
+    "VLLM",
+    "from_vllm",
+    "SteerableModel",
+    "BlackBoxModel",
+    "AsyncBlackBoxModel",
+]
diff --git a/outlines/processors/__init__.py b/outlines/processors/__init__.py
index d1997335c..bb61047be 100644
--- a/outlines/processors/__init__.py
+++ b/outlines/processors/__init__.py
@@ -1,3 +1,7 @@
 """Processors to control generation in steerable models."""
 
 from .base_logits_processor import OutlinesLogitsProcessor
+
+__all__ = [
+    "OutlinesLogitsProcessor",
+]
diff --git a/outlines/processors/tensor_adapters/__init__.py b/outlines/processors/tensor_adapters/__init__.py
index 152166fc3..a38b077c2 100644
--- a/outlines/processors/tensor_adapters/__init__.py
+++ b/outlines/processors/tensor_adapters/__init__.py
@@ -6,7 +6,6 @@
 from .numpy import NumpyTensorAdapter
 from .torch import TorchTensorAdapter
 
-
 tensor_adapters = {
     "mlx": MLXTensorAdapter,
     "numpy": NumpyTensorAdapter,
@@ -18,3 +17,11 @@
     NumpyTensorAdapter,
     TorchTensorAdapter,
 ]
+
+__all__ = [
+    "MLXTensorAdapter",
+    "NumpyTensorAdapter",
+    "TorchTensorAdapter",
+    "tensor_adapters",
+    "TensorAdapterImplementation",
+]
diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 18e7c60d3..c9e5a2795 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -9,25 +9,70 @@
 from pydantic import BaseModel, TypeAdapter
 from typing_extensions import _TypedDictMeta  # type: ignore
 
-from . import airports, countries, locale
 from outlines.types.dsl import (
-    Regex,
     CFG,
     Choice,
     JsonSchema,
-    regex,
-    cfg,
-    json_schema,
-    optional,
-    either,
-    exactly,
+    Regex,
     at_least,
     at_most,
     between,
-    zero_or_more,
+    cfg,
+    either,
+    exactly,
+    json_schema,
     one_or_more,
+    optional,
+    regex,
+    zero_or_more,
 )
 
+from . import airports, countries, locale
+
+__all__ = [
+    # Submodules
+    "airports",
+    "countries",
+    "locale",
+    # DSL functions and classes
+    "Regex",
+    "CFG",
+    "Choice",
+    "JsonSchema",
+    "regex",
+    "cfg",
+    "json_schema",
+    "optional",
+    "either",
+    "exactly",
+    "at_least",
+    "at_most",
+    "between",
+    "zero_or_more",
+    "one_or_more",
+    # Python types
+    "string",
+    "integer",
+    "boolean",
+    "number",
+    "date",
+    "time",
+    "datetime",
+    # Basic regex types
+    "digit",
+    "char",
+    "newline",
+    "whitespace",
+    "hex_str",
+    "uuid4",
+    "ipv4",
+    # Document-specific types
+    "sentence",
+    "paragraph",
+    "email",
+    "isbn",
+]
+
 
 # Python types
 string = Regex(r'"[^"]*"')
diff --git a/outlines/types/locale/__init__.py b/outlines/types/locale/__init__.py
index 97c9abff0..3f387cfa4 100644
--- a/outlines/types/locale/__init__.py
+++ b/outlines/types/locale/__init__.py
@@ -1,3 +1,7 @@
 """Locale-specific regex patterns."""
 
 from . import us
+
+__all__ = [
+    "us",
+]

From 7b6334cf1963e90fdbc415c1fff6d67fd3e9acc2 Mon Sep 17 00:00:00 2001
From: the-vampiire <25523682+the-vampiire@users.noreply.github.com>
Date: Fri, 22 Aug 2025 11:54:13 -0500
Subject: [PATCH 432/505] fix (types): remove unused imports

---
 outlines/types/__init__.py | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index c9e5a2795..515f642c0 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -1,14 +1,5 @@
 """Output types for structured generation and regex DSL."""
 
-from dataclasses import dataclass, is_dataclass
-from enum import Enum, EnumMeta
-from typing import Union
-
-from jsonschema import Draft202012Validator as Validator
-from jsonschema.exceptions import SchemaError
-from pydantic import BaseModel, TypeAdapter
-from typing_extensions import _TypedDictMeta  # type: ignore
-
 from outlines.types.dsl import (
     CFG,
     Choice,

From e8444e8c0f9434ec80fb9a4f1dab5be7e00ef7e8 Mon Sep 17 00:00:00 2001
From: the-vampiire <25523682+the-vampiire@users.noreply.github.com>
Date: Fri, 22 Aug 2025 12:23:50 -0500
Subject: [PATCH 433/505] fix (docs): docs building due to export changes

---
 scripts/gen_ref_pages.py | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/scripts/gen_ref_pages.py b/scripts/gen_ref_pages.py
index ee956d806..f31b9a2af 100644
--- a/scripts/gen_ref_pages.py
+++ b/scripts/gen_ref_pages.py
@@ -11,7 +11,6 @@
 
 import mkdocs_gen_files
 
-
 CODEBASE_DIR_NAME = "outlines"
 OUTPUT_DIR_NAME = "api_reference"
 EXCLUDED_FILES = ["_version"]
@@ -45,7 +44,11 @@
 
     with mkdocs_gen_files.open(full_doc_path, "w") as fd:
         ident = ".".join(parts)
-        fd.write(f"::: {CODEBASE_DIR_NAME}.{ident}")
+        if len(parts) == 1 and parts[0] == CODEBASE_DIR_NAME:
+            # For root module, just use the package name
+            fd.write(f"::: {CODEBASE_DIR_NAME}")
+        else:
+            fd.write(f"::: {CODEBASE_DIR_NAME}.{ident}")
 
     mkdocs_gen_files.set_edit_path(full_doc_path, path.relative_to(root))
 

From bebca54d3ecfb71e99011d9428862bc82dfd0199 Mon Sep 17 00:00:00 2001
From: the-vampiire <25523682+the-vampiire@users.noreply.github.com>
Date: Sat, 23 Aug 2025 15:54:04 -0500
Subject: [PATCH 434/505] fix (typings): remove model_list export

---
 outlines/__init__.py | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/outlines/__init__.py b/outlines/__init__.py
index e9753f7a5..007c376da 100644
--- a/outlines/__init__.py
+++ b/outlines/__init__.py
@@ -20,16 +20,3 @@
 from outlines.types import cfg as cfg
 from outlines.types import json_schema as json_schema
 from outlines.types import regex as regex
-
-model_list = [
-    "from_anthropic",
-    "from_dottxt",
-    "from_gemini",
-    "from_llamacpp",
-    "from_mlxlm",
-    "from_ollama",
-    "from_openai",
-    "from_tgi",
-    "from_transformers",
-    "from_vllm",
-]

From bac26f22f989cfbd7005d36a5af4e80379a9d4d5 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 25 Aug 2025 15:23:06 +0200
Subject: [PATCH 435/505] Remove redundant parameter api models tests workflows

---
 .github/workflows/tests_api_models.yml | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/.github/workflows/tests_api_models.yml b/.github/workflows/tests_api_models.yml
index 8f899b706..4d8633ac9 100644
--- a/.github/workflows/tests_api_models.yml
+++ b/.github/workflows/tests_api_models.yml
@@ -1,16 +1,11 @@
-name: API Model Tests
+name: API Models Tests
 
 on:
   workflow_dispatch:
-    inputs:
-      branch_name:
-        description: 'Name of the branch to test'
-        required: true
-        type: string
 
 jobs:
   tests:
-    name: Run API Model Tests
+    name: Run API Models Tests
     runs-on: ubuntu-latest
     env:
       OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
@@ -25,7 +20,7 @@ jobs:
     steps:
     - uses: actions/checkout@v3
       with:
-        ref: ${{ github.event.inputs.branch_name }}
+        ref: ${{ github.ref }}
 
     - name: Set up Python ${{ matrix.python-version }}
       uses: actions/setup-python@v4
@@ -57,6 +52,6 @@ jobs:
 
     - name: Run tests
       run: |
-        uv run pytest -x --cov=outlines -m 'api_call' --ignore=tests/models/test_dottxt.py
+        uv run pytest -m 'api_call' --ignore=tests/models/test_dottxt.py
       env:
         COVERAGE_FILE: .coverage.${{ steps.matrix-id.outputs.id }}

From 1804c67d63d4a560c045d2fd30f34e1760ccec9e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 25 Aug 2025 18:45:48 +0200
Subject: [PATCH 436/505] Remove the benchmarks workflow

---
 .github/workflows/asv_benchmark_pr.yml | 68 --------------------------
 1 file changed, 68 deletions(-)
 delete mode 100644 .github/workflows/asv_benchmark_pr.yml

diff --git a/.github/workflows/asv_benchmark_pr.yml b/.github/workflows/asv_benchmark_pr.yml
deleted file mode 100644
index d7546f854..000000000
--- a/.github/workflows/asv_benchmark_pr.yml
+++ /dev/null
@@ -1,68 +0,0 @@
-name: Benchmark PR
-
-on:
-  push:
-  pull_request:
-    types: [synchronize, labeled]
-  workflow_dispatch:
-env:
-  PYTHON_VERSION: "3.10"
-  WORKING_DIR: ${{ github.workspace }}/benchmarks
-  BENCHMARKS_OUTPUT: ${{ github.workspace }}/benchmarks_output
-
-permissions:
-  contents: read
-
-# Cancels all previous workflow runs for pull requests that have not completed.
-concurrency:
-  # The concurrency group contains the workflow name and the branch name for pull requests
-  # or the commit hash for any other events.
-  group: ${{ github.workflow }}-${{ github.event_name == 'pull_request' && github.head_ref || github.sha }}
-  cancel-in-progress: true
-
-jobs:
-  benchmark-pr:
-    runs-on: ubuntu-latest
-    if: ${{ contains(github.event.pull_request.labels.*.name, 'run-benchmarks') || github.ref == 'refs/heads/main' }}
-
-    defaults:
-      run:
-        working-directory: ${{ env.WORKING_DIR }}
-
-    steps:
-
-    - name: Checkout repository
-      uses: actions/checkout@v3
-      with:
-        fetch-depth: 0
-
-    - name: Set up Python
-      uses: actions/setup-python@v4
-      with:
-        python-version: ${{ env.PYTHON_VERSION }}
-
-    - name: Install dependencies
-      run: |
-        python -m pip install --upgrade pip
-        pip install asv virtualenv lf-asv-formatter
-
-    - name: Create ASV machine config file
-      run: asv machine --machine gh-runner --yes
-
-    - name: Run Benchmarks - `PR HEAD` vs `main`
-      run: |
-        # prepare main branch for comparison
-        git remote add upstream https://github.com/${{ github.repository }}.git
-        git fetch upstream main
-
-        # Run benchmarks, allow errors, they will be caught in the next step
-        asv continuous upstream/main HEAD \
-            --no-stats --interleave-rounds -a repeat=3 || true
-
-    - name: BENCHMARK RESULTS
-      run: |
-        asv compare --factor=1.1 --no-stats --split upstream/main HEAD | tee ${{ env.BENCHMARKS_OUTPUT }}
-        if grep -q "Benchmarks that have got worse" "${{ env.BENCHMARKS_OUTPUT }}"; then
-          echo "Performance degradation detected!"
-          exit 1
-        fi

From 12491811b95cf3cd97f480340bc75396a5cdc70f Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 30 Aug 2025 18:18:04 +0200
Subject: [PATCH 437/505] Update vLLM model references and API key usage in
 documentation

---
 docs/features/models/vllm.md | 72 ++++++++++++++++++++----------------
 1 file changed, 41 insertions(+), 31 deletions(-)

diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index df537402a..1769a2f47 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -11,7 +11,7 @@ The Outlines `VLLM` model is intended to be used along with a vLLM instance runn
 ```shell
 pip install vllm
 
-vllm serve NousResearch/Meta-Llama-3-8B-Instruct \
+vllm serve microsoft/Phi-3-mini-4k-instruct \
   --dtype auto \
   --api-key token-abc123
 ```
@@ -35,15 +35,15 @@ import openai
 import outlines
 
 # Create the OpenAI client
-sync_openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-async_openai_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
+sync_openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+async_openai_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
 
 # Create a sync model
-sync_model = outlines.from_vllm(sync_openai_client)
+sync_model = outlines.from_vllm(sync_openai_client, "microsoft/Phi-3-mini-4k-instruct")
 print(type(sync_model)) # <class 'outlines.models.vllm.VLLM'>
 
 # Create an async model
-async_model = outlines.from_vllm(async_openai_client)
+async_model = outlines.from_vllm(async_openai_client, "microsoft/Phi-3-mini-4k-instruct")
 print(type(async_model)) # <class 'outlines.models.vllm.AsyncVLLM'>
 ```
 
@@ -58,11 +58,11 @@ import openai
 import outlines
 
 # Create the model
-model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
+model = outlines.from_vllm(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"), "microsoft/Phi-3-mini-4k-instruct")
 
 # Call it to generate text
 response = model("What's the capital of Latvia?", max_tokens=20)
-print(response) # 'Riga'
+print(response) # 'The capital of Latvia is Riga.'
 ```
 
 #### Vision
@@ -80,7 +80,10 @@ import openai
 from outlines.inputs import Image
 
 # Create the model
-model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
+model = outlines.from_vllm(
+    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"), 
+    "Qwen/Qwen2.5-VL-3B-Instruct"
+)
 
 # Function to get an image
 def get_image(url):
@@ -95,7 +98,7 @@ prompt = [
 
 # Call the model to generate a response
 response = model(prompt, max_tokens=50)
-print(response) # 'This is a picture of a black dog.'
+print(response) # 'The image shows a black puppy lying on a wooden surface...'
 ```
 
 #### Chat
@@ -113,7 +116,10 @@ import outlines
 from outlines.inputs import Chat, Image
 
 # Create the model
-model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
+model = outlines.from_vllm(
+    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"), 
+    "Qwen/Qwen2.5-VL-3B-Instruct"
+)
 
 # Function to get an image
 def get_image(url):
@@ -131,7 +137,7 @@ prompt = Chat([
 
 # Call the model to generate a response
 response = model(prompt, max_tokens=50)
-print(response) # 'This is a picture of a black dog.'
+print(response) # 'The image shows a black puppy lying on a wooden surface...'
 ```
 
 #### Streaming
@@ -145,11 +151,15 @@ import openai
 import outlines
 
 # Create the model
-model = outlines.from_openai(openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct"))
+model = outlines.from_vllm(
+    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"), 
+    "microsoft/Phi-3-mini-4k-instruct"
+)
 
 # Stream the response
 for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
-    print(chunk) # 'Once...'
+    print(chunk, end="") # 'Once upon a time...'
+print()
 ```
 
 ## Asynchronous Calls
@@ -164,8 +174,8 @@ import openai
 import outlines
 
 async def generate_text():
-    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-    async_model = outlines.from_vllm(async_client)
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+    async_model = outlines.from_vllm(async_client, "microsoft/Phi-3-mini-4k-instruct")
 
     result = await async_model("Write a haiku about Python.", max_tokens=50)
     print(result)
@@ -183,8 +193,8 @@ import openai
 import outlines
 
 async def stream_text():
-    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-    async_model = outlines.from_vllm(async_client)
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+    async_model = outlines.from_vllm(async_client, "microsoft/Phi-3-mini-4k-instruct")
 
     async for chunk in async_model.stream("Tell me a story about a robot.", max_tokens=100):
         print(chunk, end="")
@@ -202,8 +212,8 @@ import openai
 import outlines
 
 async def generate_multiple():
-    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-    async_model = outlines.from_vllm(async_client)
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+    async_model = outlines.from_vllm(async_client, "microsoft/Phi-3-mini-4k-instruct")
 
     prompts = [
         "Write a tagline for a coffee shop.",
@@ -232,8 +242,8 @@ import outlines
 
 output_type = int
 
-openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-model = outlines.from_vllm(openai_client)
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
 
 result = model("How many countries are there in the world?", output_type)
 print(result) # '200'
@@ -252,8 +262,8 @@ class Character(BaseModel):
     age: int
     skills: List[str]
 
-openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-model = outlines.from_vllm(openai_client)
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
 
 result = model("Create a character.", output_type=Character, frequency_penalty=1.5)
 print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
@@ -269,8 +279,8 @@ import outlines
 
 output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
-openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-model = outlines.from_vllm(openai_client)
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
 
 result = model("What is the capital of France?", output_type, temperature=0)
 print(result) # 'Paris'
@@ -285,8 +295,8 @@ from outlines.types import Regex
 
 output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
-openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-model = outlines.from_vllm(openai_client)
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
 
 result = model("Generate a fake social security number.", output_type, top_p=0.1)
 print(result) # '782-32-3789'
@@ -321,8 +331,8 @@ arithmetic_grammar = """
 """
 output_type = CFG(arithmetic_grammar)
 
-openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-model = outlines.from_vllm(openai_client)
+openai_client = openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+model = outlines.from_vllm(openai_client, "microsoft/Phi-3-mini-4k-instruct")
 
 result = model("Write an addition.", output_type, extra_body={"guided_decoding_backend": "outlines"})
 print(result) # '23 + 48'
@@ -344,8 +354,8 @@ class User(BaseModel):
     age: int
 
 async def generate_user():
-    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", "microsoft/Phi-3-mini-4k-instruct")
-    async_model = outlines.from_vllm(async_client)
+    async_client = openai.AsyncOpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123")
+    async_model = outlines.from_vllm(async_client, "microsoft/Phi-3-mini-4k-instruct")
 
     result = await async_model("Generate a random user profile.", output_type=User)
     user = User.model_validate_json(result)

From 92eb8e07756f2fffdc9f0b7f0d434153d3e0bf27 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 30 Aug 2025 18:26:37 +0200
Subject: [PATCH 438/505] Fix trailing-whitespace

---
 docs/features/models/vllm.md | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/docs/features/models/vllm.md b/docs/features/models/vllm.md
index 1769a2f47..671354b3d 100644
--- a/docs/features/models/vllm.md
+++ b/docs/features/models/vllm.md
@@ -81,7 +81,7 @@ from outlines.inputs import Image
 
 # Create the model
 model = outlines.from_vllm(
-    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"), 
+    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"),
     "Qwen/Qwen2.5-VL-3B-Instruct"
 )
 
@@ -117,7 +117,7 @@ from outlines.inputs import Chat, Image
 
 # Create the model
 model = outlines.from_vllm(
-    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"), 
+    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"),
     "Qwen/Qwen2.5-VL-3B-Instruct"
 )
 
@@ -152,7 +152,7 @@ import outlines
 
 # Create the model
 model = outlines.from_vllm(
-    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"), 
+    openai.OpenAI(base_url="http://0.0.0.0:8000/v1", api_key="token-abc123"),
     "microsoft/Phi-3-mini-4k-instruct"
 )
 

From 039f2505b035cc06565576c77a9ffb6fc8ef899e Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 1 Sep 2025 10:06:28 +0200
Subject: [PATCH 439/505] Fix the image download issue in the tests by creating
 an image ourselves

---
 tests/models/test_transformers_multimodal.py | 101 +++++++++----------
 1 file changed, 49 insertions(+), 52 deletions(-)

diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index fb77f7c07..41ff058e9 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -1,9 +1,8 @@
 # we only test vision models here as audio models are too heavy to run on CI
 
+import io
 import re
 from enum import Enum
-from io import BytesIO
-from urllib.request import urlopen
 
 import pytest
 from PIL import Image as PILImage
@@ -23,21 +22,19 @@
 from outlines.types import Regex
 
 TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-IMAGE_URLS = [
-    "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg",
-    "https://upload.wikimedia.org/wikipedia/commons/7/71/2010-kodiak-bear-1.jpg",
-]
 
 
 @pytest.fixture
-def images():
-    def img_from_url(url):
-        img_byte_stream = BytesIO(urlopen(url).read())
-        image = PILImage.open(img_byte_stream).convert("RGB")
-        image.format = "PNG"
-        return image
+def image():
+    width, height = 256, 256
+    blue_background = (0, 0, 255)
+    image = PILImage.new("RGB", (width, height), blue_background)
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
 
-    return [img_from_url(url) for url in IMAGE_URLS]
+    return image
 
 
 @pytest.fixture
@@ -63,30 +60,30 @@ def test_transformers_multimodal_instantiate_simple():
     assert model.tensor_library_name == "torch"
 
 
-def test_transformers_multimodal_simple(model, images):
+def test_transformers_multimodal_simple(model, image):
     result = model.generate(
-        ["<image>Describe this image in one sentence:", Image(images[0])],
+        ["<image>Describe this image in one sentence:", Image(image)],
         None,
         max_new_tokens=2,
     )
     assert isinstance(result, str)
 
 
-def test_transformers_multimodal_call(model, images):
+def test_transformers_multimodal_call(model, image):
     result = model(
-        ["<image>Describe this image in one sentence:", Image(images[0])],
+        ["<image>Describe this image in one sentence:", Image(image)],
         max_new_tokens=2,
     )
     assert isinstance(result, str)
 
 
-def test_transformers_multimodal_wrong_number_images(model, images):
+def test_transformers_multimodal_wrong_number_image(model, image):
     with pytest.raises(ValueError):
         model(
             [
                 "<image>Describe this image in one sentence:",
-                Image(images[0]),
-                Image(images[1]),
+                Image(image),
+                Image(image),
             ],
         )
 
@@ -96,15 +93,15 @@ def test_transformers_multimodal_wrong_input_type(model):
         model.generate("invalid input", None)
 
 
-def test_transformers_multimodal_chat(model, images):
+def test_transformers_multimodal_chat(model, image):
     result = model(
         Chat(messages=[
             {"role": "system", "content": "You are a helpful assistant."},
             {
                 "role": "user",
                 "content": [
-                    "What's on this image?<image>",
-                    Image(images[0]),
+                    "Describe this image in one sentence:<image>",
+                    Image(image),
                 ],
             },
         ]),
@@ -113,52 +110,52 @@ def test_transformers_multimodal_chat(model, images):
     assert isinstance(result, str)
 
 
-def test_transformers_inference_kwargs(model, images):
+def test_transformers_inference_kwargs(model, image):
     result = model(
-        ["<image>Describe this image in one sentence:", Image(images[0])],
+        ["<image>Describe this image in one sentence:", Image(image)],
         max_new_tokens=2,
     )
     assert isinstance(result, str)
 
 
-def test_transformers_invalid_inference_kwargs(model, images):
+def test_transformers_invalid_inference_kwargs(model, image):
     with pytest.raises(ValueError):
         model(
             [
                 "<image>Describe this image in one sentence:",
-                Image(images[0]),
+                Image(image),
             ],
             foo="bar",
         )
 
 
-def test_transformers_several_images(model, images):
+def test_transformers_several_image(model, image):
     result = model(
         [
             "<image><image>Describe this image in one sentence:",
-            Image(images[0]),
-            Image(images[1]),
+            Image(image),
+            Image(image),
         ],
         max_new_tokens=2,
     )
     assert isinstance(result, str)
 
 
-def test_transformers_multimodal_json(model, images):
+def test_transformers_multimodal_json(model, image):
     class Foo(BaseModel):
         name: str
 
     result = model(
-        ["<image>Give a name to this animal.", Image(images[0])],
+        ["<image>Give the name of the color.", Image(image)],
         Foo,
         max_new_tokens=10,
     )
     assert "name" in result
 
 
-def test_transformers_multimodal_regex(model, images):
+def test_transformers_multimodal_regex(model, image):
     result = model(
-        ["<image>How old is it?", Image(images[0])],
+        ["<image>How warn is the color from 0 to 9?", Image(image)],
         Regex(r"[0-9]")
     )
 
@@ -166,23 +163,23 @@ def test_transformers_multimodal_regex(model, images):
     assert re.match(r"[0-9]", result)
 
 
-def test_transformers_multimodal_choice(model, images):
+def test_transformers_multimodal_choice(model, image):
     class Foo(Enum):
-        cat = "cat"
-        dog = "dog"
+        white = "white"
+        blue = "blue"
 
     result = model(
-        ["<image>Is it a cat or a dog?", Image(images[0])],
+        ["<image>Is it a white or a blue?", Image(image)],
         Foo,
     )
 
     assert isinstance(result, str)
-    assert result in ["cat", "dog"]
+    assert result in ["white", "blue"]
 
 
-def test_transformers_multimodal_multiple_samples(model, images):
+def test_transformers_multimodal_multiple_samples(model, image):
     result = model(
-        ["<image>Describe this image in one sentence.", Image(images[0])],
+        ["<image>Describe this image in one sentence.", Image(image)],
         num_return_sequences=2,
         num_beams=2,
         max_new_tokens=2,
@@ -191,11 +188,11 @@ def test_transformers_multimodal_multiple_samples(model, images):
     assert len(result) == 2
 
 
-def test_transformers_multimodal_batch(model, images):
+def test_transformers_multimodal_batch(model, image):
     result = model.batch(
         [
-            ["<image>Describe this image in one sentence.", Image(images[0])],
-            ["<image>Describe this image in one sentence.", Image(images[0])],
+            ["<image>Describe this image in one sentence.", Image(image)],
+            ["<image>Describe this image in one sentence.", Image(image)],
         ],
         max_new_tokens=2,
     )
@@ -204,8 +201,8 @@ def test_transformers_multimodal_batch(model, images):
 
     result = model.batch(
         [
-            ["<image>Describe this image in one sentence.<image>", Image(images[0]), Image(images[1])],
-            ["<image>Describe this image in one sentence.<image>", Image(images[0]), Image(images[1])],
+            ["<image>Describe this image in one sentence.<image>", Image(image), Image(image)],
+            ["<image>Describe this image in one sentence.<image>", Image(image), Image(image)],
         ],
         num_return_sequences=2,
         num_beams=2,
@@ -224,8 +221,8 @@ def test_transformers_multimodal_batch(model, images):
                 {
                     "role": "user",
                     "content": [
-                        "What's on this image?<image>",
-                        Image(images[0]),
+                        "Describe this image in one sentence:<image>",
+                        Image(image),
                     ],
                 },
             ]),
@@ -234,8 +231,8 @@ def test_transformers_multimodal_batch(model, images):
                 {
                     "role": "user",
                     "content": [
-                        "What's on this image?<image>",
-                        Image(images[1]),
+                        "Describe this image in one sentence:<image>",
+                        Image(image),
                     ],
                 },
             ]),
@@ -246,12 +243,12 @@ def test_transformers_multimodal_batch(model, images):
     assert len(result) == 2
 
 
-def test_transformers_multimodal_deprecated_input_type(model, images):
+def test_transformers_multimodal_deprecated_input_type(model, image):
     with pytest.warns(DeprecationWarning):
         result = model.generate(
             {
                 "text": "<image>Describe this image in one sentence:",
-                "images": images[0],
+                "image": image,
             },
             None,
             max_new_tokens=2,

From ef0a2d4a218a6e182f6bcc4a1cb0577efd1fd276 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 1 Sep 2025 14:02:13 +0200
Subject: [PATCH 440/505] Reduce the number of mandatory dependencies + update
 uv.lock

---
 outlines/types/__init__.py |   28 +-
 pyproject.toml             |   13 +-
 uv.lock                    | 6388 +++++++++++++++++++++---------------
 3 files changed, 3736 insertions(+), 2693 deletions(-)

diff --git a/outlines/types/__init__.py b/outlines/types/__init__.py
index 515f642c0..7d9145b31 100644
--- a/outlines/types/__init__.py
+++ b/outlines/types/__init__.py
@@ -18,7 +18,33 @@
     zero_or_more,
 )
 
-from . import airports, countries, locale
+from . import locale
+
+try:
+    from . import airports
+except ImportError:  # pragma: no cover
+    class AirportImportError:
+        """Dummy module that raises an error when accessed."""
+        def __getattr__(self, name):
+            raise ImportError(
+                "The 'airportsdata' package is required to use airport types. "
+                "Install it with: pip install 'outlines[airports]'"
+            )
+
+    airports = AirportImportError()  # type: ignore
+
+try:
+    from . import countries
+except ImportError:  # pragma: no cover
+    class CountryImportError:
+        """Dummy module that raises an error when accessed."""
+        def __getattr__(self, name):
+            raise ImportError(
+                "The 'iso3166' package is required to use country types. "
+                "Install it with: pip install 'outlines[countries]'"
+            )
+
+    countries = CountryImportError()  # type: ignore
 
 __all__ = [
     # Submodules
diff --git a/pyproject.toml b/pyproject.toml
index af55d5984..8fecdaa7c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -28,13 +28,9 @@ dependencies = [
    "cloudpickle",
    "diskcache",
    "pydantic>=2.0",
-   "referencing",
    "jsonschema",
-   "requests",
    "pillow",
    "typing_extensions",
-   "iso3166",
-   "airportsdata",
    "outlines_core==0.2.11",
    "genson",
    "jsonpath_ng",
@@ -55,6 +51,8 @@ transformers = ["accelerate", "datasets", "transformers"]
 vllm = ["openai"]
 xgrammar = ["xgrammar"]
 llguidance = ["llguidance"]
+airports = ["airportsdata"]
+countries = ["iso3166"]
 test = [
     "pre-commit",
     "pytest",
@@ -74,13 +72,13 @@ test = [
     "datasets",
     "anthropic",
     "google-genai",
-    "transformers<=4.52.1",
+    "transformers",
     "pillow",
     "jax",
     "flax",
     "numpy>=2.0.0,<2.2.0",
     "numba",
-    "torch",
+    "torch>2.3.0",
     "tensorflow",
     "tf-keras",
     "ollama",
@@ -89,6 +87,9 @@ test = [
     "mkdocs_gen_files",
     "llguidance",
     "xgrammar",
+    "airportsdata",
+    "iso3166",
+    "requests",
 ]
 test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
 
diff --git a/uv.lock b/uv.lock
index 2d927f136..cdd505bd0 100644
--- a/uv.lock
+++ b/uv.lock
@@ -1,44 +1,62 @@
 version = 1
-requires-python = ">=3.9"
+revision = 2
+requires-python = ">=3.9, <3.13"
 resolution-markers = [
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.12.*'",
+    "python_full_version >= '3.12'",
     "python_full_version == '3.11.*'",
     "python_full_version == '3.10.*'",
-    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
-    "python_full_version < '3.9.17'",
+    "python_full_version < '3.10'",
+]
+
+[[package]]
+name = "absl-py"
+version = "2.3.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/10/2a/c93173ffa1b39c1d0395b7e842bbdc62e556ca9d8d3b5572926f3e4ca752/absl_py-2.3.1.tar.gz", hash = "sha256:a97820526f7fbfd2ec1bce83f3f25e3a14840dac0d8e02a0b71cd75db3f77fc9", size = 116588, upload-time = "2025-07-03T09:31:44.05Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8f/aa/ba0014cc4659328dc818a28827be78e6d97312ab0cb98105a770924dc11e/absl_py-2.3.1-py3-none-any.whl", hash = "sha256:eeecf07f0c2a93ace0772c92e596ace6d3d3996c042b2128459aaae2a76de11d", size = 135811, upload-time = "2025-07-03T09:31:42.253Z" },
 ]
 
 [[package]]
 name = "accelerate"
-version = "1.3.0"
+version = "1.10.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "huggingface-hub" },
-    { name = "numpy" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "packaging" },
     { name = "psutil" },
     { name = "pyyaml" },
     { name = "safetensors" },
     { name = "torch" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/85/15/0fab0260ab4069e5224e637d2e400538bb27b0dfc36f17daf68db9770d78/accelerate-1.3.0.tar.gz", hash = "sha256:518631c0adb80bd3d42fb29e7e2dc2256bcd7c786b0ba9119bbaa08611b36d9c", size = 342758 }
+sdist = { url = "https://files.pythonhosted.org/packages/b1/72/ff3961c19ee395c3d30ac630ee77bfb0e1b46b87edc504d4f83bb4a89705/accelerate-1.10.1.tar.gz", hash = "sha256:3dea89e433420e4bfac0369cae7e36dcd6a56adfcfd38cdda145c6225eab5df8", size = 392446, upload-time = "2025-08-25T13:57:06.21Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/5f/a0/d9ef19f780f319c21ee90ecfef4431cbeeca95bec7f14071785c17b6029b/accelerate-1.10.1-py3-none-any.whl", hash = "sha256:3621cff60b9a27ce798857ece05e2b9f56fcc71631cfb31ccf71f0359c311f11", size = 374909, upload-time = "2025-08-25T13:57:04.55Z" },
+]
+
+[[package]]
+name = "aiofiles"
+version = "24.1.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0b/03/a88171e277e8caa88a4c77808c20ebb04ba74cc4681bf1e9416c862de237/aiofiles-24.1.0.tar.gz", hash = "sha256:22a075c9e5a3810f0c2e48f3008c94d68c65d763b9b03857924c99e57355166c", size = 30247, upload-time = "2024-06-24T11:02:03.584Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/73/de/64508cb91af013aaba214752309c0967568a4219d50a4ea30e822af3c976/accelerate-1.3.0-py3-none-any.whl", hash = "sha256:5788d9e6a7a9f80fed665cf09681c4dddd9dc056bea656db4140ffc285ce423e", size = 336647 },
+    { url = "https://files.pythonhosted.org/packages/a5/45/30bb92d442636f570cb5651bc661f52b610e2eec3f891a5dc3a4c3667db0/aiofiles-24.1.0-py3-none-any.whl", hash = "sha256:b4ec55f4195e3eb5d7abd1bf7e061763e864dd4954231fb8539a0ef8bb8260e5", size = 15896, upload-time = "2024-06-24T11:02:01.529Z" },
 ]
 
 [[package]]
 name = "aiohappyeyeballs"
-version = "2.4.4"
+version = "2.6.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/7f/55/e4373e888fdacb15563ef6fa9fa8c8252476ea071e96fb46defac9f18bf2/aiohappyeyeballs-2.4.4.tar.gz", hash = "sha256:5fdd7d87889c63183afc18ce9271f9b0a7d32c2303e394468dd45d514a757745", size = 21977 }
+sdist = { url = "https://files.pythonhosted.org/packages/26/30/f84a107a9c4331c14b2b586036f40965c128aa4fee4dda5d3d51cb14ad54/aiohappyeyeballs-2.6.1.tar.gz", hash = "sha256:c3f9d0113123803ccadfdf3f0faa505bc78e6a72d1cc4806cbd719826e943558", size = 22760, upload-time = "2025-03-12T01:42:48.764Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b9/74/fbb6559de3607b3300b9be3cc64e97548d55678e44623db17820dbd20002/aiohappyeyeballs-2.4.4-py3-none-any.whl", hash = "sha256:a980909d50efcd44795c4afeca523296716d50cd756ddca6af8c65b996e27de8", size = 14756 },
+    { url = "https://files.pythonhosted.org/packages/0f/15/5bf3b99495fb160b63f95972b81750f18f7f4e02ad051373b669d17d44f2/aiohappyeyeballs-2.6.1-py3-none-any.whl", hash = "sha256:f349ba8f4b75cb25c99c5c2d84e997e485204d2902a9597802b0371f09331fb8", size = 15265, upload-time = "2025-03-12T01:42:47.083Z" },
 ]
 
 [[package]]
 name = "aiohttp"
-version = "3.11.11"
+version = "3.12.15"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "aiohappyeyeballs" },
@@ -50,381 +68,411 @@ dependencies = [
     { name = "propcache" },
     { name = "yarl" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fe/ed/f26db39d29cd3cb2f5a3374304c713fe5ab5a0e4c8ee25a0c45cc6adf844/aiohttp-3.11.11.tar.gz", hash = "sha256:bb49c7f1e6ebf3821a42d81d494f538107610c3a705987f53068546b0e90303e", size = 7669618 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/75/7d/ff2e314b8f9e0b1df833e2d4778eaf23eae6b8cc8f922495d110ddcbf9e1/aiohttp-3.11.11-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:a60804bff28662cbcf340a4d61598891f12eea3a66af48ecfdc975ceec21e3c8", size = 708550 },
-    { url = "https://files.pythonhosted.org/packages/09/b8/aeb4975d5bba233d6f246941f5957a5ad4e3def8b0855a72742e391925f2/aiohttp-3.11.11-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:4b4fa1cb5f270fb3eab079536b764ad740bb749ce69a94d4ec30ceee1b5940d5", size = 468430 },
-    { url = "https://files.pythonhosted.org/packages/9c/5b/5b620279b3df46e597008b09fa1e10027a39467387c2332657288e25811a/aiohttp-3.11.11-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:731468f555656767cda219ab42e033355fe48c85fbe3ba83a349631541715ba2", size = 455593 },
-    { url = "https://files.pythonhosted.org/packages/d8/75/0cdf014b816867d86c0bc26f3d3e3f194198dbf33037890beed629cd4f8f/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cb23d8bb86282b342481cad4370ea0853a39e4a32a0042bb52ca6bdde132df43", size = 1584635 },
-    { url = "https://files.pythonhosted.org/packages/df/2f/95b8f4e4dfeb57c1d9ad9fa911ede35a0249d75aa339edd2c2270dc539da/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f047569d655f81cb70ea5be942ee5d4421b6219c3f05d131f64088c73bb0917f", size = 1632363 },
-    { url = "https://files.pythonhosted.org/packages/39/cb/70cf69ea7c50f5b0021a84f4c59c3622b2b3b81695f48a2f0e42ef7eba6e/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:dd7659baae9ccf94ae5fe8bfaa2c7bc2e94d24611528395ce88d009107e00c6d", size = 1668315 },
-    { url = "https://files.pythonhosted.org/packages/2f/cc/3a3fc7a290eabc59839a7e15289cd48f33dd9337d06e301064e1e7fb26c5/aiohttp-3.11.11-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:af01e42ad87ae24932138f154105e88da13ce7d202a6de93fafdafb2883a00ef", size = 1589546 },
-    { url = "https://files.pythonhosted.org/packages/15/b4/0f7b0ed41ac6000e283e7332f0f608d734b675a8509763ca78e93714cfb0/aiohttp-3.11.11-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5854be2f3e5a729800bac57a8d76af464e160f19676ab6aea74bde18ad19d438", size = 1544581 },
-    { url = "https://files.pythonhosted.org/packages/58/b9/4d06470fd85c687b6b0e31935ef73dde6e31767c9576d617309a2206556f/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:6526e5fb4e14f4bbf30411216780c9967c20c5a55f2f51d3abd6de68320cc2f3", size = 1529256 },
-    { url = "https://files.pythonhosted.org/packages/61/a2/6958b1b880fc017fd35f5dfb2c26a9a50c755b75fd9ae001dc2236a4fb79/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:85992ee30a31835fc482468637b3e5bd085fa8fe9392ba0bdcbdc1ef5e9e3c55", size = 1536592 },
-    { url = "https://files.pythonhosted.org/packages/0f/dd/b974012a9551fd654f5bb95a6dd3f03d6e6472a17e1a8216dd42e9638d6c/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:88a12ad8ccf325a8a5ed80e6d7c3bdc247d66175afedbe104ee2aaca72960d8e", size = 1607446 },
-    { url = "https://files.pythonhosted.org/packages/e0/d3/6c98fd87e638e51f074a3f2061e81fcb92123bcaf1439ac1b4a896446e40/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:0a6d3fbf2232e3a08c41eca81ae4f1dff3d8f1a30bae415ebe0af2d2458b8a33", size = 1628809 },
-    { url = "https://files.pythonhosted.org/packages/a8/2e/86e6f85cbca02be042c268c3d93e7f35977a0e127de56e319bdd1569eaa8/aiohttp-3.11.11-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:84a585799c58b795573c7fa9b84c455adf3e1d72f19a2bf498b54a95ae0d194c", size = 1564291 },
-    { url = "https://files.pythonhosted.org/packages/0b/8d/1f4ef3503b767717f65e1f5178b0173ab03cba1a19997ebf7b052161189f/aiohttp-3.11.11-cp310-cp310-win32.whl", hash = "sha256:bfde76a8f430cf5c5584553adf9926534352251d379dcb266ad2b93c54a29745", size = 416601 },
-    { url = "https://files.pythonhosted.org/packages/ad/86/81cb83691b5ace3d9aa148dc42bacc3450d749fc88c5ec1973573c1c1779/aiohttp-3.11.11-cp310-cp310-win_amd64.whl", hash = "sha256:0fd82b8e9c383af11d2b26f27a478640b6b83d669440c0a71481f7c865a51da9", size = 442007 },
-    { url = "https://files.pythonhosted.org/packages/34/ae/e8806a9f054e15f1d18b04db75c23ec38ec954a10c0a68d3bd275d7e8be3/aiohttp-3.11.11-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:ba74ec819177af1ef7f59063c6d35a214a8fde6f987f7661f4f0eecc468a8f76", size = 708624 },
-    { url = "https://files.pythonhosted.org/packages/c7/e0/313ef1a333fb4d58d0c55a6acb3cd772f5d7756604b455181049e222c020/aiohttp-3.11.11-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4af57160800b7a815f3fe0eba9b46bf28aafc195555f1824555fa2cfab6c1538", size = 468507 },
-    { url = "https://files.pythonhosted.org/packages/a9/60/03455476bf1f467e5b4a32a465c450548b2ce724eec39d69f737191f936a/aiohttp-3.11.11-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ffa336210cf9cd8ed117011085817d00abe4c08f99968deef0013ea283547204", size = 455571 },
-    { url = "https://files.pythonhosted.org/packages/be/f9/469588603bd75bf02c8ffb8c8a0d4b217eed446b49d4a767684685aa33fd/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81b8fe282183e4a3c7a1b72f5ade1094ed1c6345a8f153506d114af5bf8accd9", size = 1685694 },
-    { url = "https://files.pythonhosted.org/packages/88/b9/1b7fa43faf6c8616fa94c568dc1309ffee2b6b68b04ac268e5d64b738688/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3af41686ccec6a0f2bdc66686dc0f403c41ac2089f80e2214a0f82d001052c03", size = 1743660 },
-    { url = "https://files.pythonhosted.org/packages/2a/8b/0248d19dbb16b67222e75f6aecedd014656225733157e5afaf6a6a07e2e8/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:70d1f9dde0e5dd9e292a6d4d00058737052b01f3532f69c0c65818dac26dc287", size = 1785421 },
-    { url = "https://files.pythonhosted.org/packages/c4/11/f478e071815a46ca0a5ae974651ff0c7a35898c55063305a896e58aa1247/aiohttp-3.11.11-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:249cc6912405917344192b9f9ea5cd5b139d49e0d2f5c7f70bdfaf6b4dbf3a2e", size = 1675145 },
-    { url = "https://files.pythonhosted.org/packages/26/5d/284d182fecbb5075ae10153ff7374f57314c93a8681666600e3a9e09c505/aiohttp-3.11.11-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0eb98d90b6690827dcc84c246811feeb4e1eea683c0eac6caed7549be9c84665", size = 1619804 },
-    { url = "https://files.pythonhosted.org/packages/1b/78/980064c2ad685c64ce0e8aeeb7ef1e53f43c5b005edcd7d32e60809c4992/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ec82bf1fda6cecce7f7b915f9196601a1bd1a3079796b76d16ae4cce6d0ef89b", size = 1654007 },
-    { url = "https://files.pythonhosted.org/packages/21/8d/9e658d63b1438ad42b96f94da227f2e2c1d5c6001c9e8ffcc0bfb22e9105/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:9fd46ce0845cfe28f108888b3ab17abff84ff695e01e73657eec3f96d72eef34", size = 1650022 },
-    { url = "https://files.pythonhosted.org/packages/85/fd/a032bf7f2755c2df4f87f9effa34ccc1ef5cea465377dbaeef93bb56bbd6/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:bd176afcf8f5d2aed50c3647d4925d0db0579d96f75a31e77cbaf67d8a87742d", size = 1732899 },
-    { url = "https://files.pythonhosted.org/packages/c5/0c/c2b85fde167dd440c7ba50af2aac20b5a5666392b174df54c00f888c5a75/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:ec2aa89305006fba9ffb98970db6c8221541be7bee4c1d027421d6f6df7d1ce2", size = 1755142 },
-    { url = "https://files.pythonhosted.org/packages/bc/78/91ae1a3b3b3bed8b893c5d69c07023e151b1c95d79544ad04cf68f596c2f/aiohttp-3.11.11-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:92cde43018a2e17d48bb09c79e4d4cb0e236de5063ce897a5e40ac7cb4878773", size = 1692736 },
-    { url = "https://files.pythonhosted.org/packages/77/89/a7ef9c4b4cdb546fcc650ca7f7395aaffbd267f0e1f648a436bec33c9b95/aiohttp-3.11.11-cp311-cp311-win32.whl", hash = "sha256:aba807f9569455cba566882c8938f1a549f205ee43c27b126e5450dc9f83cc62", size = 416418 },
-    { url = "https://files.pythonhosted.org/packages/fc/db/2192489a8a51b52e06627506f8ac8df69ee221de88ab9bdea77aa793aa6a/aiohttp-3.11.11-cp311-cp311-win_amd64.whl", hash = "sha256:ae545f31489548c87b0cced5755cfe5a5308d00407000e72c4fa30b19c3220ac", size = 442509 },
-    { url = "https://files.pythonhosted.org/packages/69/cf/4bda538c502f9738d6b95ada11603c05ec260807246e15e869fc3ec5de97/aiohttp-3.11.11-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:e595c591a48bbc295ebf47cb91aebf9bd32f3ff76749ecf282ea7f9f6bb73886", size = 704666 },
-    { url = "https://files.pythonhosted.org/packages/46/7b/87fcef2cad2fad420ca77bef981e815df6904047d0a1bd6aeded1b0d1d66/aiohttp-3.11.11-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:3ea1b59dc06396b0b424740a10a0a63974c725b1c64736ff788a3689d36c02d2", size = 464057 },
-    { url = "https://files.pythonhosted.org/packages/5a/a6/789e1f17a1b6f4a38939fbc39d29e1d960d5f89f73d0629a939410171bc0/aiohttp-3.11.11-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8811f3f098a78ffa16e0ea36dffd577eb031aea797cbdba81be039a4169e242c", size = 455996 },
-    { url = "https://files.pythonhosted.org/packages/b7/dd/485061fbfef33165ce7320db36e530cd7116ee1098e9c3774d15a732b3fd/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bd7227b87a355ce1f4bf83bfae4399b1f5bb42e0259cb9405824bd03d2f4336a", size = 1682367 },
-    { url = "https://files.pythonhosted.org/packages/e9/d7/9ec5b3ea9ae215c311d88b2093e8da17e67b8856673e4166c994e117ee3e/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d40f9da8cabbf295d3a9dae1295c69975b86d941bc20f0a087f0477fa0a66231", size = 1736989 },
-    { url = "https://files.pythonhosted.org/packages/d6/fb/ea94927f7bfe1d86178c9d3e0a8c54f651a0a655214cce930b3c679b8f64/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ffb3dc385f6bb1568aa974fe65da84723210e5d9707e360e9ecb51f59406cd2e", size = 1793265 },
-    { url = "https://files.pythonhosted.org/packages/40/7f/6de218084f9b653026bd7063cd8045123a7ba90c25176465f266976d8c82/aiohttp-3.11.11-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a8f5f7515f3552d899c61202d99dcb17d6e3b0de777900405611cd747cecd1b8", size = 1691841 },
-    { url = "https://files.pythonhosted.org/packages/77/e2/992f43d87831cbddb6b09c57ab55499332f60ad6fdbf438ff4419c2925fc/aiohttp-3.11.11-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3499c7ffbfd9c6a3d8d6a2b01c26639da7e43d47c7b4f788016226b1e711caa8", size = 1619317 },
-    { url = "https://files.pythonhosted.org/packages/96/74/879b23cdd816db4133325a201287c95bef4ce669acde37f8f1b8669e1755/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:8e2bf8029dbf0810c7bfbc3e594b51c4cc9101fbffb583a3923aea184724203c", size = 1641416 },
-    { url = "https://files.pythonhosted.org/packages/30/98/b123f6b15d87c54e58fd7ae3558ff594f898d7f30a90899718f3215ad328/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:b6212a60e5c482ef90f2d788835387070a88d52cf6241d3916733c9176d39eab", size = 1646514 },
-    { url = "https://files.pythonhosted.org/packages/d7/38/257fda3dc99d6978ab943141d5165ec74fd4b4164baa15e9c66fa21da86b/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:d119fafe7b634dbfa25a8c597718e69a930e4847f0b88e172744be24515140da", size = 1702095 },
-    { url = "https://files.pythonhosted.org/packages/0c/f4/ddab089053f9fb96654df5505c0a69bde093214b3c3454f6bfdb1845f558/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:6fba278063559acc730abf49845d0e9a9e1ba74f85f0ee6efd5803f08b285853", size = 1734611 },
-    { url = "https://files.pythonhosted.org/packages/c3/d6/f30b2bc520c38c8aa4657ed953186e535ae84abe55c08d0f70acd72ff577/aiohttp-3.11.11-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:92fc484e34b733704ad77210c7957679c5c3877bd1e6b6d74b185e9320cc716e", size = 1694576 },
-    { url = "https://files.pythonhosted.org/packages/bc/97/b0a88c3f4c6d0020b34045ee6d954058abc870814f6e310c4c9b74254116/aiohttp-3.11.11-cp312-cp312-win32.whl", hash = "sha256:9f5b3c1ed63c8fa937a920b6c1bec78b74ee09593b3f5b979ab2ae5ef60d7600", size = 411363 },
-    { url = "https://files.pythonhosted.org/packages/7f/23/cc36d9c398980acaeeb443100f0216f50a7cfe20c67a9fd0a2f1a5a846de/aiohttp-3.11.11-cp312-cp312-win_amd64.whl", hash = "sha256:1e69966ea6ef0c14ee53ef7a3d68b564cc408121ea56c0caa2dc918c1b2f553d", size = 437666 },
-    { url = "https://files.pythonhosted.org/packages/49/d1/d8af164f400bad432b63e1ac857d74a09311a8334b0481f2f64b158b50eb/aiohttp-3.11.11-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:541d823548ab69d13d23730a06f97460f4238ad2e5ed966aaf850d7c369782d9", size = 697982 },
-    { url = "https://files.pythonhosted.org/packages/92/d1/faad3bf9fa4bfd26b95c69fc2e98937d52b1ff44f7e28131855a98d23a17/aiohttp-3.11.11-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:929f3ed33743a49ab127c58c3e0a827de0664bfcda566108989a14068f820194", size = 460662 },
-    { url = "https://files.pythonhosted.org/packages/db/61/0d71cc66d63909dabc4590f74eba71f91873a77ea52424401c2498d47536/aiohttp-3.11.11-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:0882c2820fd0132240edbb4a51eb8ceb6eef8181db9ad5291ab3332e0d71df5f", size = 452950 },
-    { url = "https://files.pythonhosted.org/packages/07/db/6d04bc7fd92784900704e16b745484ef45b77bd04e25f58f6febaadf7983/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b63de12e44935d5aca7ed7ed98a255a11e5cb47f83a9fded7a5e41c40277d104", size = 1665178 },
-    { url = "https://files.pythonhosted.org/packages/54/5c/e95ade9ae29f375411884d9fd98e50535bf9fe316c9feb0f30cd2ac8f508/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:aa54f8ef31d23c506910c21163f22b124facb573bff73930735cf9fe38bf7dff", size = 1717939 },
-    { url = "https://files.pythonhosted.org/packages/6f/1c/1e7d5c5daea9e409ed70f7986001b8c9e3a49a50b28404498d30860edab6/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a344d5dc18074e3872777b62f5f7d584ae4344cd6006c17ba12103759d407af3", size = 1775125 },
-    { url = "https://files.pythonhosted.org/packages/5d/66/890987e44f7d2f33a130e37e01a164168e6aff06fce15217b6eaf14df4f6/aiohttp-3.11.11-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0b7fb429ab1aafa1f48578eb315ca45bd46e9c37de11fe45c7f5f4138091e2f1", size = 1677176 },
-    { url = "https://files.pythonhosted.org/packages/8f/dc/e2ba57d7a52df6cdf1072fd5fa9c6301a68e1cd67415f189805d3eeb031d/aiohttp-3.11.11-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c341c7d868750e31961d6d8e60ff040fb9d3d3a46d77fd85e1ab8e76c3e9a5c4", size = 1603192 },
-    { url = "https://files.pythonhosted.org/packages/6c/9e/8d08a57de79ca3a358da449405555e668f2c8871a7777ecd2f0e3912c272/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ed9ee95614a71e87f1a70bc81603f6c6760128b140bc4030abe6abaa988f1c3d", size = 1618296 },
-    { url = "https://files.pythonhosted.org/packages/56/51/89822e3ec72db352c32e7fc1c690370e24e231837d9abd056490f3a49886/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:de8d38f1c2810fa2a4f1d995a2e9c70bb8737b18da04ac2afbf3971f65781d87", size = 1616524 },
-    { url = "https://files.pythonhosted.org/packages/2c/fa/e2e6d9398f462ffaa095e84717c1732916a57f1814502929ed67dd7568ef/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:a9b7371665d4f00deb8f32208c7c5e652059b0fda41cf6dbcac6114a041f1cc2", size = 1685471 },
-    { url = "https://files.pythonhosted.org/packages/ae/5f/6bb976e619ca28a052e2c0ca7b0251ccd893f93d7c24a96abea38e332bf6/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:620598717fce1b3bd14dd09947ea53e1ad510317c85dda2c9c65b622edc96b12", size = 1715312 },
-    { url = "https://files.pythonhosted.org/packages/79/c1/756a7e65aa087c7fac724d6c4c038f2faaa2a42fe56dbc1dd62a33ca7213/aiohttp-3.11.11-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:bf8d9bfee991d8acc72d060d53860f356e07a50f0e0d09a8dfedea1c554dd0d5", size = 1672783 },
-    { url = "https://files.pythonhosted.org/packages/73/ba/a6190ebb02176c7f75e6308da31f5d49f6477b651a3dcfaaaca865a298e2/aiohttp-3.11.11-cp313-cp313-win32.whl", hash = "sha256:9d73ee3725b7a737ad86c2eac5c57a4a97793d9f442599bea5ec67ac9f4bdc3d", size = 410229 },
-    { url = "https://files.pythonhosted.org/packages/b8/62/c9fa5bafe03186a0e4699150a7fed9b1e73240996d0d2f0e5f70f3fdf471/aiohttp-3.11.11-cp313-cp313-win_amd64.whl", hash = "sha256:c7a06301c2fb096bdb0bd25fe2011531c1453b9f2c163c8031600ec73af1cc99", size = 436081 },
-    { url = "https://files.pythonhosted.org/packages/9f/37/326ee86b7640be6ca4493c8121cb9a4386e07cf1e5757ce6b7fa854d0a5f/aiohttp-3.11.11-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:3e23419d832d969f659c208557de4a123e30a10d26e1e14b73431d3c13444c2e", size = 709424 },
-    { url = "https://files.pythonhosted.org/packages/9c/c5/a88ec2160b06c22e57e483a1f78f99f005fcd4e7d6855a2d3d6510881b65/aiohttp-3.11.11-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:21fef42317cf02e05d3b09c028712e1d73a9606f02467fd803f7c1f39cc59add", size = 468907 },
-    { url = "https://files.pythonhosted.org/packages/b2/f0/02f03f818e91996161cce200241b631bb2b4a87e61acddb5b974e254a288/aiohttp-3.11.11-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1f21bb8d0235fc10c09ce1d11ffbd40fc50d3f08a89e4cf3a0c503dc2562247a", size = 455981 },
-    { url = "https://files.pythonhosted.org/packages/0e/17/c8be12436ec19915f67b1ab8240d4105aba0f7e0894a1f0d8939c3e79c70/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1642eceeaa5ab6c9b6dfeaaa626ae314d808188ab23ae196a34c9d97efb68350", size = 1587395 },
-    { url = "https://files.pythonhosted.org/packages/43/c0/f4db1ac30ebe855b2fefd6fa98767862d88ac54ab08a6ad07d619146270c/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:2170816e34e10f2fd120f603e951630f8a112e1be3b60963a1f159f5699059a6", size = 1636243 },
-    { url = "https://files.pythonhosted.org/packages/ea/a7/9acf20e9a09b0d38b5b55691410500d051a9f4194692cac22b0d0fc92ad9/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8be8508d110d93061197fd2d6a74f7401f73b6d12f8822bbcd6d74f2b55d71b1", size = 1672323 },
-    { url = "https://files.pythonhosted.org/packages/f7/5b/a27e8fe1a3b0e245ca80863eefd83fc00136752d27d2cf1afa0130a76f34/aiohttp-3.11.11-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4eed954b161e6b9b65f6be446ed448ed3921763cc432053ceb606f89d793927e", size = 1589521 },
-    { url = "https://files.pythonhosted.org/packages/25/50/8bccd08004e15906791b46f0a908a8e7f5e0c5882b17da96d1933bd34ac0/aiohttp-3.11.11-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d6c9af134da4bc9b3bd3e6a70072509f295d10ee60c697826225b60b9959acdd", size = 1544059 },
-    { url = "https://files.pythonhosted.org/packages/84/5a/42250b37b06ee0cb7a03dd1630243b1d739ca3edb5abd8b18f479a539900/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:44167fc6a763d534a6908bdb2592269b4bf30a03239bcb1654781adf5e49caf1", size = 1530217 },
-    { url = "https://files.pythonhosted.org/packages/18/08/eb334da86cd2cdbd0621bb7039255b19ca74ce8b05e8fb61850e2589938c/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:479b8c6ebd12aedfe64563b85920525d05d394b85f166b7873c8bde6da612f9c", size = 1536081 },
-    { url = "https://files.pythonhosted.org/packages/1a/a9/9d59958084d5bad7e77a44841013bd59768cda94f9f744769461b66038fc/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:10b4ff0ad793d98605958089fabfa350e8e62bd5d40aa65cdc69d6785859f94e", size = 1606918 },
-    { url = "https://files.pythonhosted.org/packages/4f/e7/27feb1cff17dcddb7a5b703199106196718d622a3aa70f80a386d15361d7/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:b540bd67cfb54e6f0865ceccd9979687210d7ed1a1cc8c01f8e67e2f1e883d28", size = 1629101 },
-    { url = "https://files.pythonhosted.org/packages/e8/29/49debcd858b997c655fca274c5247fcfe29bf31a4ddb1ce3f088539b14e4/aiohttp-3.11.11-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1dac54e8ce2ed83b1f6b1a54005c87dfed139cf3f777fdc8afc76e7841101226", size = 1567338 },
-    { url = "https://files.pythonhosted.org/packages/3b/34/33af1e97aba1862e1812e2e2b96a1e050c5a6e9cecd5a5370591122fb07b/aiohttp-3.11.11-cp39-cp39-win32.whl", hash = "sha256:568c1236b2fde93b7720f95a890741854c1200fba4a3471ff48b2934d2d93fd3", size = 416914 },
-    { url = "https://files.pythonhosted.org/packages/2d/47/28b3fbd97026963af2774423c64341e0d4ec180ea3b79a2762a3c18d5d94/aiohttp-3.11.11-cp39-cp39-win_amd64.whl", hash = "sha256:943a8b052e54dfd6439fd7989f67fc6a7f2138d0a2cf0a7de5f18aa4fe7eb3b1", size = 442225 },
+sdist = { url = "https://files.pythonhosted.org/packages/9b/e7/d92a237d8802ca88483906c388f7c201bbe96cd80a165ffd0ac2f6a8d59f/aiohttp-3.12.15.tar.gz", hash = "sha256:4fc61385e9c98d72fcdf47e6dd81833f47b2f77c114c29cd64a361be57a763a2", size = 7823716, upload-time = "2025-07-29T05:52:32.215Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/47/dc/ef9394bde9080128ad401ac7ede185267ed637df03b51f05d14d1c99ad67/aiohttp-3.12.15-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:b6fc902bff74d9b1879ad55f5404153e2b33a82e72a95c89cec5eb6cc9e92fbc", size = 703921, upload-time = "2025-07-29T05:49:43.584Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/42/63fccfc3a7ed97eb6e1a71722396f409c46b60a0552d8a56d7aad74e0df5/aiohttp-3.12.15-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:098e92835b8119b54c693f2f88a1dec690e20798ca5f5fe5f0520245253ee0af", size = 480288, upload-time = "2025-07-29T05:49:47.851Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/a2/7b8a020549f66ea2a68129db6960a762d2393248f1994499f8ba9728bbed/aiohttp-3.12.15-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:40b3fee496a47c3b4a39a731954c06f0bd9bd3e8258c059a4beb76ac23f8e421", size = 468063, upload-time = "2025-07-29T05:49:49.789Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/f5/d11e088da9176e2ad8220338ae0000ed5429a15f3c9dfd983f39105399cd/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2ce13fcfb0bb2f259fb42106cdc63fa5515fb85b7e87177267d89a771a660b79", size = 1650122, upload-time = "2025-07-29T05:49:51.874Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/6b/b60ce2757e2faed3d70ed45dafee48cee7bfb878785a9423f7e883f0639c/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3beb14f053222b391bf9cf92ae82e0171067cc9c8f52453a0f1ec7c37df12a77", size = 1624176, upload-time = "2025-07-29T05:49:53.805Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/de/8c9fde2072a1b72c4fadecf4f7d4be7a85b1d9a4ab333d8245694057b4c6/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4c39e87afe48aa3e814cac5f535bc6199180a53e38d3f51c5e2530f5aa4ec58c", size = 1696583, upload-time = "2025-07-29T05:49:55.338Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/ad/07f863ca3d895a1ad958a54006c6dafb4f9310f8c2fdb5f961b8529029d3/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:d5f1b4ce5bc528a6ee38dbf5f39bbf11dd127048726323b72b8e85769319ffc4", size = 1738896, upload-time = "2025-07-29T05:49:57.045Z" },
+    { url = "https://files.pythonhosted.org/packages/20/43/2bd482ebe2b126533e8755a49b128ec4e58f1a3af56879a3abdb7b42c54f/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1004e67962efabbaf3f03b11b4c43b834081c9e3f9b32b16a7d97d4708a9abe6", size = 1643561, upload-time = "2025-07-29T05:49:58.762Z" },
+    { url = "https://files.pythonhosted.org/packages/23/40/2fa9f514c4cf4cbae8d7911927f81a1901838baf5e09a8b2c299de1acfe5/aiohttp-3.12.15-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:8faa08fcc2e411f7ab91d1541d9d597d3a90e9004180edb2072238c085eac8c2", size = 1583685, upload-time = "2025-07-29T05:50:00.375Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/c3/94dc7357bc421f4fb978ca72a201a6c604ee90148f1181790c129396ceeb/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:fe086edf38b2222328cdf89af0dde2439ee173b8ad7cb659b4e4c6f385b2be3d", size = 1627533, upload-time = "2025-07-29T05:50:02.306Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/3f/1f8911fe1844a07001e26593b5c255a685318943864b27b4e0267e840f95/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:79b26fe467219add81d5e47b4a4ba0f2394e8b7c7c3198ed36609f9ba161aecb", size = 1638319, upload-time = "2025-07-29T05:50:04.282Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/46/27bf57a99168c4e145ffee6b63d0458b9c66e58bb70687c23ad3d2f0bd17/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:b761bac1192ef24e16706d761aefcb581438b34b13a2f069a6d343ec8fb693a5", size = 1613776, upload-time = "2025-07-29T05:50:05.863Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/7e/1d2d9061a574584bb4ad3dbdba0da90a27fdc795bc227def3a46186a8bc1/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:e153e8adacfe2af562861b72f8bc47f8a5c08e010ac94eebbe33dc21d677cd5b", size = 1693359, upload-time = "2025-07-29T05:50:07.563Z" },
+    { url = "https://files.pythonhosted.org/packages/08/98/bee429b52233c4a391980a5b3b196b060872a13eadd41c3a34be9b1469ed/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:fc49c4de44977aa8601a00edbf157e9a421f227aa7eb477d9e3df48343311065", size = 1716598, upload-time = "2025-07-29T05:50:09.33Z" },
+    { url = "https://files.pythonhosted.org/packages/57/39/b0314c1ea774df3392751b686104a3938c63ece2b7ce0ba1ed7c0b4a934f/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:2776c7ec89c54a47029940177e75c8c07c29c66f73464784971d6a81904ce9d1", size = 1644940, upload-time = "2025-07-29T05:50:11.334Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/83/3dacb8d3f8f512c8ca43e3fa8a68b20583bd25636ffa4e56ee841ffd79ae/aiohttp-3.12.15-cp310-cp310-win32.whl", hash = "sha256:2c7d81a277fa78b2203ab626ced1487420e8c11a8e373707ab72d189fcdad20a", size = 429239, upload-time = "2025-07-29T05:50:12.803Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/f9/470b5daba04d558c9673ca2034f28d067f3202a40e17804425f0c331c89f/aiohttp-3.12.15-cp310-cp310-win_amd64.whl", hash = "sha256:83603f881e11f0f710f8e2327817c82e79431ec976448839f3cd05d7afe8f830", size = 452297, upload-time = "2025-07-29T05:50:14.266Z" },
+    { url = "https://files.pythonhosted.org/packages/20/19/9e86722ec8e835959bd97ce8c1efa78cf361fa4531fca372551abcc9cdd6/aiohttp-3.12.15-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:d3ce17ce0220383a0f9ea07175eeaa6aa13ae5a41f30bc61d84df17f0e9b1117", size = 711246, upload-time = "2025-07-29T05:50:15.937Z" },
+    { url = "https://files.pythonhosted.org/packages/71/f9/0a31fcb1a7d4629ac9d8f01f1cb9242e2f9943f47f5d03215af91c3c1a26/aiohttp-3.12.15-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:010cc9bbd06db80fe234d9003f67e97a10fe003bfbedb40da7d71c1008eda0fe", size = 483515, upload-time = "2025-07-29T05:50:17.442Z" },
+    { url = "https://files.pythonhosted.org/packages/62/6c/94846f576f1d11df0c2e41d3001000527c0fdf63fce7e69b3927a731325d/aiohttp-3.12.15-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3f9d7c55b41ed687b9d7165b17672340187f87a773c98236c987f08c858145a9", size = 471776, upload-time = "2025-07-29T05:50:19.568Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/6c/f766d0aaafcee0447fad0328da780d344489c042e25cd58fde566bf40aed/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bc4fbc61bb3548d3b482f9ac7ddd0f18c67e4225aaa4e8552b9f1ac7e6bda9e5", size = 1741977, upload-time = "2025-07-29T05:50:21.665Z" },
+    { url = "https://files.pythonhosted.org/packages/17/e5/fb779a05ba6ff44d7bc1e9d24c644e876bfff5abe5454f7b854cace1b9cc/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:7fbc8a7c410bb3ad5d595bb7118147dfbb6449d862cc1125cf8867cb337e8728", size = 1690645, upload-time = "2025-07-29T05:50:23.333Z" },
+    { url = "https://files.pythonhosted.org/packages/37/4e/a22e799c2035f5d6a4ad2cf8e7c1d1bd0923192871dd6e367dafb158b14c/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:74dad41b3458dbb0511e760fb355bb0b6689e0630de8a22b1b62a98777136e16", size = 1789437, upload-time = "2025-07-29T05:50:25.007Z" },
+    { url = "https://files.pythonhosted.org/packages/28/e5/55a33b991f6433569babb56018b2fb8fb9146424f8b3a0c8ecca80556762/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3b6f0af863cf17e6222b1735a756d664159e58855da99cfe965134a3ff63b0b0", size = 1828482, upload-time = "2025-07-29T05:50:26.693Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/82/1ddf0ea4f2f3afe79dffed5e8a246737cff6cbe781887a6a170299e33204/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b5b7fe4972d48a4da367043b8e023fb70a04d1490aa7d68800e465d1b97e493b", size = 1730944, upload-time = "2025-07-29T05:50:28.382Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/96/784c785674117b4cb3877522a177ba1b5e4db9ce0fd519430b5de76eec90/aiohttp-3.12.15-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6443cca89553b7a5485331bc9bedb2342b08d073fa10b8c7d1c60579c4a7b9bd", size = 1668020, upload-time = "2025-07-29T05:50:30.032Z" },
+    { url = "https://files.pythonhosted.org/packages/12/8a/8b75f203ea7e5c21c0920d84dd24a5c0e971fe1e9b9ebbf29ae7e8e39790/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:6c5f40ec615e5264f44b4282ee27628cea221fcad52f27405b80abb346d9f3f8", size = 1716292, upload-time = "2025-07-29T05:50:31.983Z" },
+    { url = "https://files.pythonhosted.org/packages/47/0b/a1451543475bb6b86a5cfc27861e52b14085ae232896a2654ff1231c0992/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:2abbb216a1d3a2fe86dbd2edce20cdc5e9ad0be6378455b05ec7f77361b3ab50", size = 1711451, upload-time = "2025-07-29T05:50:33.989Z" },
+    { url = "https://files.pythonhosted.org/packages/55/fd/793a23a197cc2f0d29188805cfc93aa613407f07e5f9da5cd1366afd9d7c/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:db71ce547012a5420a39c1b744d485cfb823564d01d5d20805977f5ea1345676", size = 1691634, upload-time = "2025-07-29T05:50:35.846Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/bf/23a335a6670b5f5dfc6d268328e55a22651b440fca341a64fccf1eada0c6/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:ced339d7c9b5030abad5854aa5413a77565e5b6e6248ff927d3e174baf3badf7", size = 1785238, upload-time = "2025-07-29T05:50:37.597Z" },
+    { url = "https://files.pythonhosted.org/packages/57/4f/ed60a591839a9d85d40694aba5cef86dde9ee51ce6cca0bb30d6eb1581e7/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:7c7dd29c7b5bda137464dc9bfc738d7ceea46ff70309859ffde8c022e9b08ba7", size = 1805701, upload-time = "2025-07-29T05:50:39.591Z" },
+    { url = "https://files.pythonhosted.org/packages/85/e0/444747a9455c5de188c0f4a0173ee701e2e325d4b2550e9af84abb20cdba/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:421da6fd326460517873274875c6c5a18ff225b40da2616083c5a34a7570b685", size = 1718758, upload-time = "2025-07-29T05:50:41.292Z" },
+    { url = "https://files.pythonhosted.org/packages/36/ab/1006278d1ffd13a698e5dd4bfa01e5878f6bddefc296c8b62649753ff249/aiohttp-3.12.15-cp311-cp311-win32.whl", hash = "sha256:4420cf9d179ec8dfe4be10e7d0fe47d6d606485512ea2265b0d8c5113372771b", size = 428868, upload-time = "2025-07-29T05:50:43.063Z" },
+    { url = "https://files.pythonhosted.org/packages/10/97/ad2b18700708452400278039272032170246a1bf8ec5d832772372c71f1a/aiohttp-3.12.15-cp311-cp311-win_amd64.whl", hash = "sha256:edd533a07da85baa4b423ee8839e3e91681c7bfa19b04260a469ee94b778bf6d", size = 453273, upload-time = "2025-07-29T05:50:44.613Z" },
+    { url = "https://files.pythonhosted.org/packages/63/97/77cb2450d9b35f517d6cf506256bf4f5bda3f93a66b4ad64ba7fc917899c/aiohttp-3.12.15-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:802d3868f5776e28f7bf69d349c26fc0efadb81676d0afa88ed00d98a26340b7", size = 702333, upload-time = "2025-07-29T05:50:46.507Z" },
+    { url = "https://files.pythonhosted.org/packages/83/6d/0544e6b08b748682c30b9f65640d006e51f90763b41d7c546693bc22900d/aiohttp-3.12.15-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f2800614cd560287be05e33a679638e586a2d7401f4ddf99e304d98878c29444", size = 476948, upload-time = "2025-07-29T05:50:48.067Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/1d/c8c40e611e5094330284b1aea8a4b02ca0858f8458614fa35754cab42b9c/aiohttp-3.12.15-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8466151554b593909d30a0a125d638b4e5f3836e5aecde85b66b80ded1cb5b0d", size = 469787, upload-time = "2025-07-29T05:50:49.669Z" },
+    { url = "https://files.pythonhosted.org/packages/38/7d/b76438e70319796bfff717f325d97ce2e9310f752a267bfdf5192ac6082b/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2e5a495cb1be69dae4b08f35a6c4579c539e9b5706f606632102c0f855bcba7c", size = 1716590, upload-time = "2025-07-29T05:50:51.368Z" },
+    { url = "https://files.pythonhosted.org/packages/79/b1/60370d70cdf8b269ee1444b390cbd72ce514f0d1cd1a715821c784d272c9/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:6404dfc8cdde35c69aaa489bb3542fb86ef215fc70277c892be8af540e5e21c0", size = 1699241, upload-time = "2025-07-29T05:50:53.628Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/2b/4968a7b8792437ebc12186db31523f541943e99bda8f30335c482bea6879/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3ead1c00f8521a5c9070fcb88f02967b1d8a0544e6d85c253f6968b785e1a2ab", size = 1754335, upload-time = "2025-07-29T05:50:55.394Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/c1/49524ed553f9a0bec1a11fac09e790f49ff669bcd14164f9fab608831c4d/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6990ef617f14450bc6b34941dba4f12d5613cbf4e33805932f853fbd1cf18bfb", size = 1800491, upload-time = "2025-07-29T05:50:57.202Z" },
+    { url = "https://files.pythonhosted.org/packages/de/5e/3bf5acea47a96a28c121b167f5ef659cf71208b19e52a88cdfa5c37f1fcc/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd736ed420f4db2b8148b52b46b88ed038d0354255f9a73196b7bbce3ea97545", size = 1719929, upload-time = "2025-07-29T05:50:59.192Z" },
+    { url = "https://files.pythonhosted.org/packages/39/94/8ae30b806835bcd1cba799ba35347dee6961a11bd507db634516210e91d8/aiohttp-3.12.15-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3c5092ce14361a73086b90c6efb3948ffa5be2f5b6fbcf52e8d8c8b8848bb97c", size = 1635733, upload-time = "2025-07-29T05:51:01.394Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/46/06cdef71dd03acd9da7f51ab3a9107318aee12ad38d273f654e4f981583a/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:aaa2234bb60c4dbf82893e934d8ee8dea30446f0647e024074237a56a08c01bd", size = 1696790, upload-time = "2025-07-29T05:51:03.657Z" },
+    { url = "https://files.pythonhosted.org/packages/02/90/6b4cfaaf92ed98d0ec4d173e78b99b4b1a7551250be8937d9d67ecb356b4/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:6d86a2fbdd14192e2f234a92d3b494dd4457e683ba07e5905a0b3ee25389ac9f", size = 1718245, upload-time = "2025-07-29T05:51:05.911Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/e6/2593751670fa06f080a846f37f112cbe6f873ba510d070136a6ed46117c6/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:a041e7e2612041a6ddf1c6a33b883be6a421247c7afd47e885969ee4cc58bd8d", size = 1658899, upload-time = "2025-07-29T05:51:07.753Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/28/c15bacbdb8b8eb5bf39b10680d129ea7410b859e379b03190f02fa104ffd/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:5015082477abeafad7203757ae44299a610e89ee82a1503e3d4184e6bafdd519", size = 1738459, upload-time = "2025-07-29T05:51:09.56Z" },
+    { url = "https://files.pythonhosted.org/packages/00/de/c269cbc4faa01fb10f143b1670633a8ddd5b2e1ffd0548f7aa49cb5c70e2/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:56822ff5ddfd1b745534e658faba944012346184fbfe732e0d6134b744516eea", size = 1766434, upload-time = "2025-07-29T05:51:11.423Z" },
+    { url = "https://files.pythonhosted.org/packages/52/b0/4ff3abd81aa7d929b27d2e1403722a65fc87b763e3a97b3a2a494bfc63bc/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b2acbbfff69019d9014508c4ba0401822e8bae5a5fdc3b6814285b71231b60f3", size = 1726045, upload-time = "2025-07-29T05:51:13.689Z" },
+    { url = "https://files.pythonhosted.org/packages/71/16/949225a6a2dd6efcbd855fbd90cf476052e648fb011aa538e3b15b89a57a/aiohttp-3.12.15-cp312-cp312-win32.whl", hash = "sha256:d849b0901b50f2185874b9a232f38e26b9b3d4810095a7572eacea939132d4e1", size = 423591, upload-time = "2025-07-29T05:51:15.452Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/d8/fa65d2a349fe938b76d309db1a56a75c4fb8cc7b17a398b698488a939903/aiohttp-3.12.15-cp312-cp312-win_amd64.whl", hash = "sha256:b390ef5f62bb508a9d67cb3bba9b8356e23b3996da7062f1a57ce1a79d2b3d34", size = 450266, upload-time = "2025-07-29T05:51:17.239Z" },
+    { url = "https://files.pythonhosted.org/packages/18/8d/da08099af8db234d1cd43163e6ffc8e9313d0e988cee1901610f2fa5c764/aiohttp-3.12.15-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:691d203c2bdf4f4637792efbbcdcd157ae11e55eaeb5e9c360c1206fb03d4d98", size = 706829, upload-time = "2025-07-29T05:51:54.434Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/94/8eed385cfb60cf4fdb5b8a165f6148f3bebeb365f08663d83c35a5f273ef/aiohttp-3.12.15-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:8e995e1abc4ed2a454c731385bf4082be06f875822adc4c6d9eaadf96e20d406", size = 481806, upload-time = "2025-07-29T05:51:56.355Z" },
+    { url = "https://files.pythonhosted.org/packages/38/68/b13e1a34584fbf263151b3a72a084e89f2102afe38df1dce5a05a15b83e9/aiohttp-3.12.15-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:bd44d5936ab3193c617bfd6c9a7d8d1085a8dc8c3f44d5f1dcf554d17d04cf7d", size = 469205, upload-time = "2025-07-29T05:51:58.277Z" },
+    { url = "https://files.pythonhosted.org/packages/38/14/3d7348bf53aa4af54416bc64cbef3a2ac5e8b9bfa97cc45f1cf9a94d9c8d/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46749be6e89cd78d6068cdf7da51dbcfa4321147ab8e4116ee6678d9a056a0cf", size = 1644174, upload-time = "2025-07-29T05:52:00.23Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/ed/fd9b5b22b0f6ca1a85c33bb4868cbcc6ae5eae070a0f4c9c5cad003c89d7/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:0c643f4d75adea39e92c0f01b3fb83d57abdec8c9279b3078b68a3a52b3933b6", size = 1618672, upload-time = "2025-07-29T05:52:02.272Z" },
+    { url = "https://files.pythonhosted.org/packages/39/f7/f6530ab5f8c8c409e44a63fcad35e839c87aabecdfe5b8e96d671ed12f64/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0a23918fedc05806966a2438489dcffccbdf83e921a1170773b6178d04ade142", size = 1692295, upload-time = "2025-07-29T05:52:04.546Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/dc/3cf483bb0106566dc97ebaa2bb097f5e44d4bc4ab650a6f107151cd7b193/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:74bdd8c864b36c3673741023343565d95bfbd778ffe1eb4d412c135a28a8dc89", size = 1731609, upload-time = "2025-07-29T05:52:06.552Z" },
+    { url = "https://files.pythonhosted.org/packages/de/a4/fd04bf807851197077d9cac9381d58f86d91c95c06cbaf9d3a776ac4467a/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0a146708808c9b7a988a4af3821379e379e0f0e5e466ca31a73dbdd0325b0263", size = 1637852, upload-time = "2025-07-29T05:52:08.975Z" },
+    { url = "https://files.pythonhosted.org/packages/98/03/29d626ca3bcdcafbd74b45d77ca42645a5c94d396f2ee3446880ad2405fb/aiohttp-3.12.15-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b7011a70b56facde58d6d26da4fec3280cc8e2a78c714c96b7a01a87930a9530", size = 1572852, upload-time = "2025-07-29T05:52:11.508Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/cd/b4777a9e204f4e01091091027e5d1e2fa86decd0fee5067bc168e4fa1e76/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:3bdd6e17e16e1dbd3db74d7f989e8af29c4d2e025f9828e6ef45fbdee158ec75", size = 1620813, upload-time = "2025-07-29T05:52:13.891Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/26/1a44a6e8417e84057beaf8c462529b9e05d4b53b8605784f1eb571f0ff68/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:57d16590a351dfc914670bd72530fd78344b885a00b250e992faea565b7fdc05", size = 1630951, upload-time = "2025-07-29T05:52:15.955Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/7f/10c605dbd01c40e2b27df7ef9004bec75d156f0705141e11047ecdfe264d/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:bc9a0f6569ff990e0bbd75506c8d8fe7214c8f6579cca32f0546e54372a3bb54", size = 1607595, upload-time = "2025-07-29T05:52:18.089Z" },
+    { url = "https://files.pythonhosted.org/packages/66/f6/2560dcb01731c1d7df1d34b64de95bc4b3ed02bb78830fd82299c1eb314e/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:536ad7234747a37e50e7b6794ea868833d5220b49c92806ae2d7e8a9d6b5de02", size = 1695194, upload-time = "2025-07-29T05:52:20.255Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/02/ee105ae82dc2b981039fd25b0cf6eaa52b493731960f9bc861375a72b463/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:f0adb4177fa748072546fb650d9bd7398caaf0e15b370ed3317280b13f4083b0", size = 1710872, upload-time = "2025-07-29T05:52:22.769Z" },
+    { url = "https://files.pythonhosted.org/packages/88/16/70c4e42ed6a04f78fb58d1a46500a6ce560741d13afde2a5f33840746a5f/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:14954a2988feae3987f1eb49c706bff39947605f4b6fa4027c1d75743723eb09", size = 1640539, upload-time = "2025-07-29T05:52:25.733Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/1d/a7eb5fa8a6967117c5c0ad5ab4b1dec0d21e178c89aa08bc442a0b836392/aiohttp-3.12.15-cp39-cp39-win32.whl", hash = "sha256:b784d6ed757f27574dca1c336f968f4e81130b27595e458e69457e6878251f5d", size = 430164, upload-time = "2025-07-29T05:52:27.905Z" },
+    { url = "https://files.pythonhosted.org/packages/14/25/e0cf8793aedc41c6d7f2aad646a27e27bdacafe3b402bb373d7651c94d73/aiohttp-3.12.15-cp39-cp39-win_amd64.whl", hash = "sha256:86ceded4e78a992f835209e236617bffae649371c4a50d5e5a3987f237db84b8", size = 453370, upload-time = "2025-07-29T05:52:29.936Z" },
 ]
 
 [[package]]
 name = "aiosignal"
-version = "1.3.2"
+version = "1.4.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "frozenlist" },
+    { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ba/b5/6d55e80f6d8a08ce22b982eafa278d823b541c925f11ee774b0b9c43473d/aiosignal-1.3.2.tar.gz", hash = "sha256:a8c255c66fafb1e499c9351d0bf32ff2d8a0321595ebac3b93713656d2436f54", size = 19424 }
+sdist = { url = "https://files.pythonhosted.org/packages/61/62/06741b579156360248d1ec624842ad0edf697050bbaf7c3e46394e106ad1/aiosignal-1.4.0.tar.gz", hash = "sha256:f47eecd9468083c2029cc99945502cb7708b082c232f9aca65da147157b251c7", size = 25007, upload-time = "2025-07-03T22:54:43.528Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ec/6a/bc7e17a3e87a2985d3e8f4da4cd0f481060eb78fb08596c42be62c90a4d9/aiosignal-1.3.2-py2.py3-none-any.whl", hash = "sha256:45cde58e409a301715980c2b01d0c28bdde3770d8290b5eb2173759d9acb31a5", size = 7597 },
+    { url = "https://files.pythonhosted.org/packages/fb/76/641ae371508676492379f16e2fa48f4e2c11741bd63c48be4b12a6b09cba/aiosignal-1.4.0-py3-none-any.whl", hash = "sha256:053243f8b92b990551949e63930a839ff0cf0b0ebbe0597b0f3fb19e1a0fe82e", size = 7490, upload-time = "2025-07-03T22:54:42.156Z" },
 ]
 
 [[package]]
 name = "airportsdata"
-version = "20241001"
+version = "20250811"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/ff/75/0af9b73babfec14df1395ca60e2a174e887c5d515e823c39a60f27bbbb30/airportsdata-20241001.tar.gz", hash = "sha256:fa0bd143b4f4be3557cb892fa0612ef210fd91a92bd720b4d8221de576a4fa00", size = 903075 }
+sdist = { url = "https://files.pythonhosted.org/packages/6b/61/3751adeb92d3197268ad5db00d6fb7ba9c2db72757f0d47462338c25dd0a/airportsdata-20250811.tar.gz", hash = "sha256:844b663c33e85c8f3b932c14c798ef314bec53bda3b14f5694eddc24c7ab4b25", size = 905387, upload-time = "2025-08-11T18:38:24.684Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/af/57/57635b0af4a23723b2b43109aa4eae7395d346d1eed8a00691cf3acfc5bd/airportsdata-20241001-py3-none-any.whl", hash = "sha256:67d71cf2c5378cc17ff66b62b1e11aa2444043949c894543ac8fd8dafce192fd", size = 912677 },
+    { url = "https://files.pythonhosted.org/packages/15/52/f229acde71b1b9c05f06e504d3dcfcc91bc0614946a35d68cbc36b746dfa/airportsdata-20250811-py3-none-any.whl", hash = "sha256:96f37fcf3f7732350b6dd606d2a8faac7e3b610d0c0b2c1637a3d2fab6b5bf39", size = 912696, upload-time = "2025-08-11T18:38:22.426Z" },
 ]
 
 [[package]]
 name = "annotated-types"
 version = "0.7.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/ee/67/531ea369ba64dcff5ec9c3402f9f51bf748cec26dde048a2f973a4eea7f5/annotated_types-0.7.0.tar.gz", hash = "sha256:aff07c09a53a08bc8cfccb9c85b05f1aa9a2a6f23728d790723543408344ce89", size = 16081 }
+sdist = { url = "https://files.pythonhosted.org/packages/ee/67/531ea369ba64dcff5ec9c3402f9f51bf748cec26dde048a2f973a4eea7f5/annotated_types-0.7.0.tar.gz", hash = "sha256:aff07c09a53a08bc8cfccb9c85b05f1aa9a2a6f23728d790723543408344ce89", size = 16081, upload-time = "2024-05-20T21:33:25.928Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/78/b6/6307fbef88d9b5ee7421e68d78a9f162e0da4900bc5f5793f6d3d0e34fb8/annotated_types-0.7.0-py3-none-any.whl", hash = "sha256:1f02e8b43a8fbbc3f3e0d4f0f4bfc8131bcb4eebe8849b8e5c773f3a1c582a53", size = 13643, upload-time = "2024-05-20T21:33:24.1Z" },
+]
+
+[[package]]
+name = "anthropic"
+version = "0.64.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+    { name = "distro" },
+    { name = "httpx" },
+    { name = "jiter" },
+    { name = "pydantic" },
+    { name = "sniffio" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/d8/4f/f2b880cba1a76f3acc7d5eb2ae217632eac1b8cef5ed3027493545c59eba/anthropic-0.64.0.tar.gz", hash = "sha256:3d496c91a63dff64f451b3e8e4b238a9640bf87b0c11d0b74ddc372ba5a3fe58", size = 427893, upload-time = "2025-08-13T17:09:49.915Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/78/b6/6307fbef88d9b5ee7421e68d78a9f162e0da4900bc5f5793f6d3d0e34fb8/annotated_types-0.7.0-py3-none-any.whl", hash = "sha256:1f02e8b43a8fbbc3f3e0d4f0f4bfc8131bcb4eebe8849b8e5c773f3a1c582a53", size = 13643 },
+    { url = "https://files.pythonhosted.org/packages/a9/b2/2d268bcd5d6441df9dc0ebebc67107657edb8b0150d3fda1a5b81d1bec45/anthropic-0.64.0-py3-none-any.whl", hash = "sha256:6f5f7d913a6a95eb7f8e1bda4e75f76670e8acd8d4cd965e02e2a256b0429dd1", size = 297244, upload-time = "2025-08-13T17:09:47.908Z" },
 ]
 
 [[package]]
 name = "anyio"
-version = "4.8.0"
+version = "4.10.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "exceptiongroup", marker = "python_full_version < '3.11'" },
     { name = "idna" },
     { name = "sniffio" },
-    { name = "typing-extensions", marker = "python_full_version < '3.13'" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f1/b4/636b3b65173d3ce9a38ef5f0522789614e590dab6a8d505340a4efe4c567/anyio-4.10.0.tar.gz", hash = "sha256:3f3fae35c96039744587aa5b8371e7e8e603c0702999535961dd336026973ba6", size = 213252, upload-time = "2025-08-04T08:54:26.451Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6f/12/e5e0282d673bb9746bacfb6e2dba8719989d3660cdb2ea79aee9a9651afb/anyio-4.10.0-py3-none-any.whl", hash = "sha256:60e474ac86736bbfd6f210f7a61218939c318f43f9972497381f1c5e930ed3d1", size = 107213, upload-time = "2025-08-04T08:54:24.882Z" },
+]
+
+[[package]]
+name = "astunparse"
+version = "1.6.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "six" },
+    { name = "wheel" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a3/73/199a98fc2dae33535d6b8e8e6ec01f8c1d76c9adb096c6b7d64823038cde/anyio-4.8.0.tar.gz", hash = "sha256:1d9fe889df5212298c0c0723fa20479d1b94883a2df44bd3897aa91083316f7a", size = 181126 }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/af/4182184d3c338792894f34a62672919db7ca008c89abee9b564dd34d8029/astunparse-1.6.3.tar.gz", hash = "sha256:5ad93a8456f0d084c3456d059fd9a92cce667963232cbf763eac3bc5b7940872", size = 18290, upload-time = "2019-12-22T18:12:13.129Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/46/eb/e7f063ad1fec6b3178a3cd82d1a3c4de82cccf283fc42746168188e1cdd5/anyio-4.8.0-py3-none-any.whl", hash = "sha256:b5011f270ab5eb0abf13385f851315585cc37ef330dd88e27ec3d34d651fd47a", size = 96041 },
+    { url = "https://files.pythonhosted.org/packages/2b/03/13dde6512ad7b4557eb792fbcf0c653af6076b81e5941d36ec61f7ce6028/astunparse-1.6.3-py2.py3-none-any.whl", hash = "sha256:c2652417f2c8b5bb325c885ae329bdf3f86424075c4fd1a128674bc6fba4b8e8", size = 12732, upload-time = "2019-12-22T18:12:11.297Z" },
 ]
 
 [[package]]
 name = "async-timeout"
 version = "5.0.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/a5/ae/136395dfbfe00dfc94da3f3e136d0b13f394cba8f4841120e34226265780/async_timeout-5.0.1.tar.gz", hash = "sha256:d9321a7a3d5a6a5e187e824d2fa0793ce379a202935782d555d6e9d2735677d3", size = 9274 }
+sdist = { url = "https://files.pythonhosted.org/packages/a5/ae/136395dfbfe00dfc94da3f3e136d0b13f394cba8f4841120e34226265780/async_timeout-5.0.1.tar.gz", hash = "sha256:d9321a7a3d5a6a5e187e824d2fa0793ce379a202935782d555d6e9d2735677d3", size = 9274, upload-time = "2024-11-06T16:41:39.6Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/fe/ba/e2081de779ca30d473f21f5b30e0e737c438205440784c7dfc81efc2b029/async_timeout-5.0.1-py3-none-any.whl", hash = "sha256:39e3809566ff85354557ec2398b55e096c8364bacac9405a7a1fa429e77fe76c", size = 6233 },
+    { url = "https://files.pythonhosted.org/packages/fe/ba/e2081de779ca30d473f21f5b30e0e737c438205440784c7dfc81efc2b029/async_timeout-5.0.1-py3-none-any.whl", hash = "sha256:39e3809566ff85354557ec2398b55e096c8364bacac9405a7a1fa429e77fe76c", size = 6233, upload-time = "2024-11-06T16:41:37.9Z" },
 ]
 
 [[package]]
 name = "attrs"
-version = "24.3.0"
+version = "25.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/5a/b0/1367933a8532ee6ff8d63537de4f1177af4bff9f3e829baf7331f595bb24/attrs-25.3.0.tar.gz", hash = "sha256:75d7cefc7fb576747b2c81b4442d4d4a1ce0900973527c011d1030fd3bf4af1b", size = 812032, upload-time = "2025-03-13T11:10:22.779Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/77/06/bb80f5f86020c4551da315d78b3ab75e8228f89f0162f2c3a819e407941a/attrs-25.3.0-py3-none-any.whl", hash = "sha256:427318ce031701fea540783410126f03899a97ffc6f61596ad581ac2e40e3bc3", size = 63815, upload-time = "2025-03-13T11:10:21.14Z" },
+]
+
+[[package]]
+name = "backoff"
+version = "2.2.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/47/d7/5bbeb12c44d7c4f2fb5b56abce497eb5ed9f34d85701de869acedd602619/backoff-2.2.1.tar.gz", hash = "sha256:03f829f5bb1923180821643f8753b0502c3b682293992485b0eef2807afa5cba", size = 17001, upload-time = "2022-10-05T19:19:32.061Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/df/73/b6e24bd22e6720ca8ee9a85a0c4a2971af8497d8f3193fa05390cbd46e09/backoff-2.2.1-py3-none-any.whl", hash = "sha256:63579f9a0628e06278f7e47b7d7d5b6ce20dc65c5e96a6f3ca99a6adca0396e8", size = 15148, upload-time = "2022-10-05T19:19:30.546Z" },
+]
+
+[[package]]
+name = "backports-asyncio-runner"
+version = "1.2.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/48/c8/6260f8ccc11f0917360fc0da435c5c9c7504e3db174d5a12a1494887b045/attrs-24.3.0.tar.gz", hash = "sha256:8f5c07333d543103541ba7be0e2ce16eeee8130cb0b3f9238ab904ce1e85baff", size = 805984 }
+sdist = { url = "https://files.pythonhosted.org/packages/8e/ff/70dca7d7cb1cbc0edb2c6cc0c38b65cba36cccc491eca64cabd5fe7f8670/backports_asyncio_runner-1.2.0.tar.gz", hash = "sha256:a5aa7b2b7d8f8bfcaa2b57313f70792df84e32a2a746f585213373f900b42162", size = 69893, upload-time = "2025-07-02T02:27:15.685Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/89/aa/ab0f7891a01eeb2d2e338ae8fecbe57fcebea1a24dbb64d45801bfab481d/attrs-24.3.0-py3-none-any.whl", hash = "sha256:ac96cd038792094f438ad1f6ff80837353805ac950cd2aa0e0625ef19850c308", size = 63397 },
+    { url = "https://files.pythonhosted.org/packages/a0/59/76ab57e3fe74484f48a53f8e337171b4a2349e506eabe136d7e01d059086/backports_asyncio_runner-1.2.0-py3-none-any.whl", hash = "sha256:0da0a936a8aeb554eccb426dc55af3ba63bcdc69fa1a600b5bb305413a4477b5", size = 12313, upload-time = "2025-07-02T02:27:14.263Z" },
 ]
 
 [[package]]
 name = "beartype"
 version = "0.15.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/d2/6c/fecfe9c7d7b903a52b732b8f33968a067cad8209848ae8037c4d0ed53055/beartype-0.15.0.tar.gz", hash = "sha256:2af6a8d8a7267ccf7d271e1a3bd908afbc025d2a09aa51123567d7d7b37438df", size = 1002709 }
+sdist = { url = "https://files.pythonhosted.org/packages/d2/6c/fecfe9c7d7b903a52b732b8f33968a067cad8209848ae8037c4d0ed53055/beartype-0.15.0.tar.gz", hash = "sha256:2af6a8d8a7267ccf7d271e1a3bd908afbc025d2a09aa51123567d7d7b37438df", size = 1002709, upload-time = "2023-07-22T05:25:44.074Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6f/e7/9220de010e015fdfb7ce22e7d9a846bfcd6e5d686c4fc555fa76a22846ba/beartype-0.15.0-py3-none-any.whl", hash = "sha256:52cd2edea72fdd84e4e7f8011a9e3007bf0125c3d6d7219e937b9d8868169177", size = 777565, upload-time = "2023-07-22T05:25:41.659Z" },
+]
+
+[[package]]
+name = "cachetools"
+version = "5.5.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/6c/81/3747dad6b14fa2cf53fcf10548cf5aea6913e96fab41a3c198676f8948a5/cachetools-5.5.2.tar.gz", hash = "sha256:1a661caa9175d26759571b2e19580f9d6393969e5dfca11fdb1f947a23e640d4", size = 28380, upload-time = "2025-02-20T21:01:19.524Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6f/e7/9220de010e015fdfb7ce22e7d9a846bfcd6e5d686c4fc555fa76a22846ba/beartype-0.15.0-py3-none-any.whl", hash = "sha256:52cd2edea72fdd84e4e7f8011a9e3007bf0125c3d6d7219e937b9d8868169177", size = 777565 },
+    { url = "https://files.pythonhosted.org/packages/72/76/20fa66124dbe6be5cafeb312ece67de6b61dd91a0247d1ea13db4ebb33c2/cachetools-5.5.2-py3-none-any.whl", hash = "sha256:d26a22bcc62eb95c3beabd9f1ee5e820d3d2704fe2967cbe350e20c8ffcd3f0a", size = 10080, upload-time = "2025-02-20T21:01:16.647Z" },
 ]
 
 [[package]]
 name = "certifi"
-version = "2024.12.14"
+version = "2025.8.3"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/0f/bd/1d41ee578ce09523c81a15426705dd20969f5abf006d1afe8aeff0dd776a/certifi-2024.12.14.tar.gz", hash = "sha256:b650d30f370c2b724812bee08008be0c4163b163ddaec3f2546c1caf65f191db", size = 166010 }
+sdist = { url = "https://files.pythonhosted.org/packages/dc/67/960ebe6bf230a96cda2e0abcf73af550ec4f090005363542f0765df162e0/certifi-2025.8.3.tar.gz", hash = "sha256:e564105f78ded564e3ae7c923924435e1daa7463faeab5bb932bc53ffae63407", size = 162386, upload-time = "2025-08-03T03:07:47.08Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a5/32/8f6669fc4798494966bf446c8c4a162e0b5d893dff088afddf76414f70e1/certifi-2024.12.14-py3-none-any.whl", hash = "sha256:1275f7a45be9464efc1173084eaa30f866fe2e47d389406136d332ed4967ec56", size = 164927 },
+    { url = "https://files.pythonhosted.org/packages/e5/48/1549795ba7742c948d2ad169c1c8cdbae65bc450d6cd753d124b17c8cd32/certifi-2025.8.3-py3-none-any.whl", hash = "sha256:f6c12493cfb1b06ba2ff328595af9350c65d6644968e5d3a2ffd78699af217a5", size = 161216, upload-time = "2025-08-03T03:07:45.777Z" },
 ]
 
 [[package]]
 name = "cfgv"
 version = "3.4.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/11/74/539e56497d9bd1d484fd863dd69cbbfa653cd2aa27abfe35653494d85e94/cfgv-3.4.0.tar.gz", hash = "sha256:e52591d4c5f5dead8e0f673fb16db7949d2cfb3f7da4582893288f0ded8fe560", size = 7114 }
+sdist = { url = "https://files.pythonhosted.org/packages/11/74/539e56497d9bd1d484fd863dd69cbbfa653cd2aa27abfe35653494d85e94/cfgv-3.4.0.tar.gz", hash = "sha256:e52591d4c5f5dead8e0f673fb16db7949d2cfb3f7da4582893288f0ded8fe560", size = 7114, upload-time = "2023-08-12T20:38:17.776Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c5/55/51844dd50c4fc7a33b653bfaba4c2456f06955289ca770a5dbd5fd267374/cfgv-3.4.0-py2.py3-none-any.whl", hash = "sha256:b7265b1f29fd3316bfcd2b330d63d024f2bfd8bcb8b0272f8e19a504856c48f9", size = 7249 },
+    { url = "https://files.pythonhosted.org/packages/c5/55/51844dd50c4fc7a33b653bfaba4c2456f06955289ca770a5dbd5fd267374/cfgv-3.4.0-py2.py3-none-any.whl", hash = "sha256:b7265b1f29fd3316bfcd2b330d63d024f2bfd8bcb8b0272f8e19a504856c48f9", size = 7249, upload-time = "2023-08-12T20:38:16.269Z" },
 ]
 
 [[package]]
 name = "chardet"
 version = "5.2.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f3/0d/f7b6ab21ec75897ed80c17d79b15951a719226b9fababf1e40ea74d69079/chardet-5.2.0.tar.gz", hash = "sha256:1b3b6ff479a8c414bc3fa2c0852995695c4a026dcd6d0633b2dd092ca39c1cf7", size = 2069618 }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/0d/f7b6ab21ec75897ed80c17d79b15951a719226b9fababf1e40ea74d69079/chardet-5.2.0.tar.gz", hash = "sha256:1b3b6ff479a8c414bc3fa2c0852995695c4a026dcd6d0633b2dd092ca39c1cf7", size = 2069618, upload-time = "2023-08-01T19:23:02.662Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/38/6f/f5fbc992a329ee4e0f288c1fe0e2ad9485ed064cac731ed2fe47dcc38cbf/chardet-5.2.0-py3-none-any.whl", hash = "sha256:e1cf59446890a00105fe7b7912492ea04b6e6f06d4b742b2c788469e34c82970", size = 199385 },
+    { url = "https://files.pythonhosted.org/packages/38/6f/f5fbc992a329ee4e0f288c1fe0e2ad9485ed064cac731ed2fe47dcc38cbf/chardet-5.2.0-py3-none-any.whl", hash = "sha256:e1cf59446890a00105fe7b7912492ea04b6e6f06d4b742b2c788469e34c82970", size = 199385, upload-time = "2023-08-01T19:23:00.661Z" },
 ]
 
 [[package]]
 name = "charset-normalizer"
-version = "3.4.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/16/b0/572805e227f01586461c80e0fd25d65a2115599cc9dad142fee4b747c357/charset_normalizer-3.4.1.tar.gz", hash = "sha256:44251f18cd68a75b56585dd00dae26183e102cd5e0f9f1466e6df5da2ed64ea3", size = 123188 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/0d/58/5580c1716040bc89206c77d8f74418caf82ce519aae06450393ca73475d1/charset_normalizer-3.4.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:91b36a978b5ae0ee86c394f5a54d6ef44db1de0815eb43de826d41d21e4af3de", size = 198013 },
-    { url = "https://files.pythonhosted.org/packages/d0/11/00341177ae71c6f5159a08168bcb98c6e6d196d372c94511f9f6c9afe0c6/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7461baadb4dc00fd9e0acbe254e3d7d2112e7f92ced2adc96e54ef6501c5f176", size = 141285 },
-    { url = "https://files.pythonhosted.org/packages/01/09/11d684ea5819e5a8f5100fb0b38cf8d02b514746607934134d31233e02c8/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e218488cd232553829be0664c2292d3af2eeeb94b32bea483cf79ac6a694e037", size = 151449 },
-    { url = "https://files.pythonhosted.org/packages/08/06/9f5a12939db324d905dc1f70591ae7d7898d030d7662f0d426e2286f68c9/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:80ed5e856eb7f30115aaf94e4a08114ccc8813e6ed1b5efa74f9f82e8509858f", size = 143892 },
-    { url = "https://files.pythonhosted.org/packages/93/62/5e89cdfe04584cb7f4d36003ffa2936681b03ecc0754f8e969c2becb7e24/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b010a7a4fd316c3c484d482922d13044979e78d1861f0e0650423144c616a46a", size = 146123 },
-    { url = "https://files.pythonhosted.org/packages/a9/ac/ab729a15c516da2ab70a05f8722ecfccc3f04ed7a18e45c75bbbaa347d61/charset_normalizer-3.4.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4532bff1b8421fd0a320463030c7520f56a79c9024a4e88f01c537316019005a", size = 147943 },
-    { url = "https://files.pythonhosted.org/packages/03/d2/3f392f23f042615689456e9a274640c1d2e5dd1d52de36ab8f7955f8f050/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d973f03c0cb71c5ed99037b870f2be986c3c05e63622c017ea9816881d2dd247", size = 142063 },
-    { url = "https://files.pythonhosted.org/packages/f2/e3/e20aae5e1039a2cd9b08d9205f52142329f887f8cf70da3650326670bddf/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:3a3bd0dcd373514dcec91c411ddb9632c0d7d92aed7093b8c3bbb6d69ca74408", size = 150578 },
-    { url = "https://files.pythonhosted.org/packages/8d/af/779ad72a4da0aed925e1139d458adc486e61076d7ecdcc09e610ea8678db/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:d9c3cdf5390dcd29aa8056d13e8e99526cda0305acc038b96b30352aff5ff2bb", size = 153629 },
-    { url = "https://files.pythonhosted.org/packages/c2/b6/7aa450b278e7aa92cf7732140bfd8be21f5f29d5bf334ae987c945276639/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:2bdfe3ac2e1bbe5b59a1a63721eb3b95fc9b6817ae4a46debbb4e11f6232428d", size = 150778 },
-    { url = "https://files.pythonhosted.org/packages/39/f4/d9f4f712d0951dcbfd42920d3db81b00dd23b6ab520419626f4023334056/charset_normalizer-3.4.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:eab677309cdb30d047996b36d34caeda1dc91149e4fdca0b1a039b3f79d9a807", size = 146453 },
-    { url = "https://files.pythonhosted.org/packages/49/2b/999d0314e4ee0cff3cb83e6bc9aeddd397eeed693edb4facb901eb8fbb69/charset_normalizer-3.4.1-cp310-cp310-win32.whl", hash = "sha256:c0429126cf75e16c4f0ad00ee0eae4242dc652290f940152ca8c75c3a4b6ee8f", size = 95479 },
-    { url = "https://files.pythonhosted.org/packages/2d/ce/3cbed41cff67e455a386fb5e5dd8906cdda2ed92fbc6297921f2e4419309/charset_normalizer-3.4.1-cp310-cp310-win_amd64.whl", hash = "sha256:9f0b8b1c6d84c8034a44893aba5e767bf9c7a211e313a9605d9c617d7083829f", size = 102790 },
-    { url = "https://files.pythonhosted.org/packages/72/80/41ef5d5a7935d2d3a773e3eaebf0a9350542f2cab4eac59a7a4741fbbbbe/charset_normalizer-3.4.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:8bfa33f4f2672964266e940dd22a195989ba31669bd84629f05fab3ef4e2d125", size = 194995 },
-    { url = "https://files.pythonhosted.org/packages/7a/28/0b9fefa7b8b080ec492110af6d88aa3dea91c464b17d53474b6e9ba5d2c5/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:28bf57629c75e810b6ae989f03c0828d64d6b26a5e205535585f96093e405ed1", size = 139471 },
-    { url = "https://files.pythonhosted.org/packages/71/64/d24ab1a997efb06402e3fc07317e94da358e2585165930d9d59ad45fcae2/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f08ff5e948271dc7e18a35641d2f11a4cd8dfd5634f55228b691e62b37125eb3", size = 149831 },
-    { url = "https://files.pythonhosted.org/packages/37/ed/be39e5258e198655240db5e19e0b11379163ad7070962d6b0c87ed2c4d39/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:234ac59ea147c59ee4da87a0c0f098e9c8d169f4dc2a159ef720f1a61bbe27cd", size = 142335 },
-    { url = "https://files.pythonhosted.org/packages/88/83/489e9504711fa05d8dde1574996408026bdbdbd938f23be67deebb5eca92/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd4ec41f914fa74ad1b8304bbc634b3de73d2a0889bd32076342a573e0779e00", size = 143862 },
-    { url = "https://files.pythonhosted.org/packages/c6/c7/32da20821cf387b759ad24627a9aca289d2822de929b8a41b6241767b461/charset_normalizer-3.4.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:eea6ee1db730b3483adf394ea72f808b6e18cf3cb6454b4d86e04fa8c4327a12", size = 145673 },
-    { url = "https://files.pythonhosted.org/packages/68/85/f4288e96039abdd5aeb5c546fa20a37b50da71b5cf01e75e87f16cd43304/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:c96836c97b1238e9c9e3fe90844c947d5afbf4f4c92762679acfe19927d81d77", size = 140211 },
-    { url = "https://files.pythonhosted.org/packages/28/a3/a42e70d03cbdabc18997baf4f0227c73591a08041c149e710045c281f97b/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:4d86f7aff21ee58f26dcf5ae81a9addbd914115cdebcbb2217e4f0ed8982e146", size = 148039 },
-    { url = "https://files.pythonhosted.org/packages/85/e4/65699e8ab3014ecbe6f5c71d1a55d810fb716bbfd74f6283d5c2aa87febf/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:09b5e6733cbd160dcc09589227187e242a30a49ca5cefa5a7edd3f9d19ed53fd", size = 151939 },
-    { url = "https://files.pythonhosted.org/packages/b1/82/8e9fe624cc5374193de6860aba3ea8070f584c8565ee77c168ec13274bd2/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:5777ee0881f9499ed0f71cc82cf873d9a0ca8af166dfa0af8ec4e675b7df48e6", size = 149075 },
-    { url = "https://files.pythonhosted.org/packages/3d/7b/82865ba54c765560c8433f65e8acb9217cb839a9e32b42af4aa8e945870f/charset_normalizer-3.4.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:237bdbe6159cff53b4f24f397d43c6336c6b0b42affbe857970cefbb620911c8", size = 144340 },
-    { url = "https://files.pythonhosted.org/packages/b5/b6/9674a4b7d4d99a0d2df9b215da766ee682718f88055751e1e5e753c82db0/charset_normalizer-3.4.1-cp311-cp311-win32.whl", hash = "sha256:8417cb1f36cc0bc7eaba8ccb0e04d55f0ee52df06df3ad55259b9a323555fc8b", size = 95205 },
-    { url = "https://files.pythonhosted.org/packages/1e/ab/45b180e175de4402dcf7547e4fb617283bae54ce35c27930a6f35b6bef15/charset_normalizer-3.4.1-cp311-cp311-win_amd64.whl", hash = "sha256:d7f50a1f8c450f3925cb367d011448c39239bb3eb4117c36a6d354794de4ce76", size = 102441 },
-    { url = "https://files.pythonhosted.org/packages/0a/9a/dd1e1cdceb841925b7798369a09279bd1cf183cef0f9ddf15a3a6502ee45/charset_normalizer-3.4.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:73d94b58ec7fecbc7366247d3b0b10a21681004153238750bb67bd9012414545", size = 196105 },
-    { url = "https://files.pythonhosted.org/packages/d3/8c/90bfabf8c4809ecb648f39794cf2a84ff2e7d2a6cf159fe68d9a26160467/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dad3e487649f498dd991eeb901125411559b22e8d7ab25d3aeb1af367df5efd7", size = 140404 },
-    { url = "https://files.pythonhosted.org/packages/ad/8f/e410d57c721945ea3b4f1a04b74f70ce8fa800d393d72899f0a40526401f/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c30197aa96e8eed02200a83fba2657b4c3acd0f0aa4bdc9f6c1af8e8962e0757", size = 150423 },
-    { url = "https://files.pythonhosted.org/packages/f0/b8/e6825e25deb691ff98cf5c9072ee0605dc2acfca98af70c2d1b1bc75190d/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2369eea1ee4a7610a860d88f268eb39b95cb588acd7235e02fd5a5601773d4fa", size = 143184 },
-    { url = "https://files.pythonhosted.org/packages/3e/a2/513f6cbe752421f16d969e32f3583762bfd583848b763913ddab8d9bfd4f/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc2722592d8998c870fa4e290c2eec2c1569b87fe58618e67d38b4665dfa680d", size = 145268 },
-    { url = "https://files.pythonhosted.org/packages/74/94/8a5277664f27c3c438546f3eb53b33f5b19568eb7424736bdc440a88a31f/charset_normalizer-3.4.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ffc9202a29ab3920fa812879e95a9e78b2465fd10be7fcbd042899695d75e616", size = 147601 },
-    { url = "https://files.pythonhosted.org/packages/7c/5f/6d352c51ee763623a98e31194823518e09bfa48be2a7e8383cf691bbb3d0/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:804a4d582ba6e5b747c625bf1255e6b1507465494a40a2130978bda7b932c90b", size = 141098 },
-    { url = "https://files.pythonhosted.org/packages/78/d4/f5704cb629ba5ab16d1d3d741396aec6dc3ca2b67757c45b0599bb010478/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:0f55e69f030f7163dffe9fd0752b32f070566451afe180f99dbeeb81f511ad8d", size = 149520 },
-    { url = "https://files.pythonhosted.org/packages/c5/96/64120b1d02b81785f222b976c0fb79a35875457fa9bb40827678e54d1bc8/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:c4c3e6da02df6fa1410a7680bd3f63d4f710232d3139089536310d027950696a", size = 152852 },
-    { url = "https://files.pythonhosted.org/packages/84/c9/98e3732278a99f47d487fd3468bc60b882920cef29d1fa6ca460a1fdf4e6/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:5df196eb874dae23dcfb968c83d4f8fdccb333330fe1fc278ac5ceeb101003a9", size = 150488 },
-    { url = "https://files.pythonhosted.org/packages/13/0e/9c8d4cb99c98c1007cc11eda969ebfe837bbbd0acdb4736d228ccaabcd22/charset_normalizer-3.4.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e358e64305fe12299a08e08978f51fc21fac060dcfcddd95453eabe5b93ed0e1", size = 146192 },
-    { url = "https://files.pythonhosted.org/packages/b2/21/2b6b5b860781a0b49427309cb8670785aa543fb2178de875b87b9cc97746/charset_normalizer-3.4.1-cp312-cp312-win32.whl", hash = "sha256:9b23ca7ef998bc739bf6ffc077c2116917eabcc901f88da1b9856b210ef63f35", size = 95550 },
-    { url = "https://files.pythonhosted.org/packages/21/5b/1b390b03b1d16c7e382b561c5329f83cc06623916aab983e8ab9239c7d5c/charset_normalizer-3.4.1-cp312-cp312-win_amd64.whl", hash = "sha256:6ff8a4a60c227ad87030d76e99cd1698345d4491638dfa6673027c48b3cd395f", size = 102785 },
-    { url = "https://files.pythonhosted.org/packages/38/94/ce8e6f63d18049672c76d07d119304e1e2d7c6098f0841b51c666e9f44a0/charset_normalizer-3.4.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:aabfa34badd18f1da5ec1bc2715cadc8dca465868a4e73a0173466b688f29dda", size = 195698 },
-    { url = "https://files.pythonhosted.org/packages/24/2e/dfdd9770664aae179a96561cc6952ff08f9a8cd09a908f259a9dfa063568/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:22e14b5d70560b8dd51ec22863f370d1e595ac3d024cb8ad7d308b4cd95f8313", size = 140162 },
-    { url = "https://files.pythonhosted.org/packages/24/4e/f646b9093cff8fc86f2d60af2de4dc17c759de9d554f130b140ea4738ca6/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8436c508b408b82d87dc5f62496973a1805cd46727c34440b0d29d8a2f50a6c9", size = 150263 },
-    { url = "https://files.pythonhosted.org/packages/5e/67/2937f8d548c3ef6e2f9aab0f6e21001056f692d43282b165e7c56023e6dd/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2d074908e1aecee37a7635990b2c6d504cd4766c7bc9fc86d63f9c09af3fa11b", size = 142966 },
-    { url = "https://files.pythonhosted.org/packages/52/ed/b7f4f07de100bdb95c1756d3a4d17b90c1a3c53715c1a476f8738058e0fa/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:955f8851919303c92343d2f66165294848d57e9bba6cf6e3625485a70a038d11", size = 144992 },
-    { url = "https://files.pythonhosted.org/packages/96/2c/d49710a6dbcd3776265f4c923bb73ebe83933dfbaa841c5da850fe0fd20b/charset_normalizer-3.4.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:44ecbf16649486d4aebafeaa7ec4c9fed8b88101f4dd612dcaf65d5e815f837f", size = 147162 },
-    { url = "https://files.pythonhosted.org/packages/b4/41/35ff1f9a6bd380303dea55e44c4933b4cc3c4850988927d4082ada230273/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:0924e81d3d5e70f8126529951dac65c1010cdf117bb75eb02dd12339b57749dd", size = 140972 },
-    { url = "https://files.pythonhosted.org/packages/fb/43/c6a0b685fe6910d08ba971f62cd9c3e862a85770395ba5d9cad4fede33ab/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:2967f74ad52c3b98de4c3b32e1a44e32975e008a9cd2a8cc8966d6a5218c5cb2", size = 149095 },
-    { url = "https://files.pythonhosted.org/packages/4c/ff/a9a504662452e2d2878512115638966e75633519ec11f25fca3d2049a94a/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:c75cb2a3e389853835e84a2d8fb2b81a10645b503eca9bcb98df6b5a43eb8886", size = 152668 },
-    { url = "https://files.pythonhosted.org/packages/6c/71/189996b6d9a4b932564701628af5cee6716733e9165af1d5e1b285c530ed/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:09b26ae6b1abf0d27570633b2b078a2a20419c99d66fb2823173d73f188ce601", size = 150073 },
-    { url = "https://files.pythonhosted.org/packages/e4/93/946a86ce20790e11312c87c75ba68d5f6ad2208cfb52b2d6a2c32840d922/charset_normalizer-3.4.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:fa88b843d6e211393a37219e6a1c1df99d35e8fd90446f1118f4216e307e48cd", size = 145732 },
-    { url = "https://files.pythonhosted.org/packages/cd/e5/131d2fb1b0dddafc37be4f3a2fa79aa4c037368be9423061dccadfd90091/charset_normalizer-3.4.1-cp313-cp313-win32.whl", hash = "sha256:eb8178fe3dba6450a3e024e95ac49ed3400e506fd4e9e5c32d30adda88cbd407", size = 95391 },
-    { url = "https://files.pythonhosted.org/packages/27/f2/4f9a69cc7712b9b5ad8fdb87039fd89abba997ad5cbe690d1835d40405b0/charset_normalizer-3.4.1-cp313-cp313-win_amd64.whl", hash = "sha256:b1ac5992a838106edb89654e0aebfc24f5848ae2547d22c2c3f66454daa11971", size = 102702 },
-    { url = "https://files.pythonhosted.org/packages/7f/c0/b913f8f02836ed9ab32ea643c6fe4d3325c3d8627cf6e78098671cafff86/charset_normalizer-3.4.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:b97e690a2118911e39b4042088092771b4ae3fc3aa86518f84b8cf6888dbdb41", size = 197867 },
-    { url = "https://files.pythonhosted.org/packages/0f/6c/2bee440303d705b6fb1e2ec789543edec83d32d258299b16eed28aad48e0/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:78baa6d91634dfb69ec52a463534bc0df05dbd546209b79a3880a34487f4b84f", size = 141385 },
-    { url = "https://files.pythonhosted.org/packages/3d/04/cb42585f07f6f9fd3219ffb6f37d5a39b4fd2db2355b23683060029c35f7/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1a2bc9f351a75ef49d664206d51f8e5ede9da246602dc2d2726837620ea034b2", size = 151367 },
-    { url = "https://files.pythonhosted.org/packages/54/54/2412a5b093acb17f0222de007cc129ec0e0df198b5ad2ce5699355269dfe/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:75832c08354f595c760a804588b9357d34ec00ba1c940c15e31e96d902093770", size = 143928 },
-    { url = "https://files.pythonhosted.org/packages/5a/6d/e2773862b043dcf8a221342954f375392bb2ce6487bcd9f2c1b34e1d6781/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0af291f4fe114be0280cdd29d533696a77b5b49cfde5467176ecab32353395c4", size = 146203 },
-    { url = "https://files.pythonhosted.org/packages/b9/f8/ca440ef60d8f8916022859885f231abb07ada3c347c03d63f283bec32ef5/charset_normalizer-3.4.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0167ddc8ab6508fe81860a57dd472b2ef4060e8d378f0cc555707126830f2537", size = 148082 },
-    { url = "https://files.pythonhosted.org/packages/04/d2/42fd330901aaa4b805a1097856c2edf5095e260a597f65def493f4b8c833/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:2a75d49014d118e4198bcee5ee0a6f25856b29b12dbf7cd012791f8a6cc5c496", size = 142053 },
-    { url = "https://files.pythonhosted.org/packages/9e/af/3a97a4fa3c53586f1910dadfc916e9c4f35eeada36de4108f5096cb7215f/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:363e2f92b0f0174b2f8238240a1a30142e3db7b957a5dd5689b0e75fb717cc78", size = 150625 },
-    { url = "https://files.pythonhosted.org/packages/26/ae/23d6041322a3556e4da139663d02fb1b3c59a23ab2e2b56432bd2ad63ded/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:ab36c8eb7e454e34e60eb55ca5d241a5d18b2c6244f6827a30e451c42410b5f7", size = 153549 },
-    { url = "https://files.pythonhosted.org/packages/94/22/b8f2081c6a77cb20d97e57e0b385b481887aa08019d2459dc2858ed64871/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:4c0907b1928a36d5a998d72d64d8eaa7244989f7aaaf947500d3a800c83a3fd6", size = 150945 },
-    { url = "https://files.pythonhosted.org/packages/c7/0b/c5ec5092747f801b8b093cdf5610e732b809d6cb11f4c51e35fc28d1d389/charset_normalizer-3.4.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:04432ad9479fa40ec0f387795ddad4437a2b50417c69fa275e212933519ff294", size = 146595 },
-    { url = "https://files.pythonhosted.org/packages/0c/5a/0b59704c38470df6768aa154cc87b1ac7c9bb687990a1559dc8765e8627e/charset_normalizer-3.4.1-cp39-cp39-win32.whl", hash = "sha256:3bed14e9c89dcb10e8f3a29f9ccac4955aebe93c71ae803af79265c9ca5644c5", size = 95453 },
-    { url = "https://files.pythonhosted.org/packages/85/2d/a9790237cb4d01a6d57afadc8573c8b73c609ade20b80f4cda30802009ee/charset_normalizer-3.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:49402233c892a461407c512a19435d1ce275543138294f7ef013f0b63d5d3765", size = 102811 },
-    { url = "https://files.pythonhosted.org/packages/0e/f6/65ecc6878a89bb1c23a086ea335ad4bf21a588990c3f535a227b9eea9108/charset_normalizer-3.4.1-py3-none-any.whl", hash = "sha256:d98b1668f06378c6dbefec3b92299716b931cd4e6061f3c875a71ced1780ab85", size = 49767 },
+version = "3.4.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/83/2d/5fd176ceb9b2fc619e63405525573493ca23441330fcdaee6bef9460e924/charset_normalizer-3.4.3.tar.gz", hash = "sha256:6fce4b8500244f6fcb71465d4a4930d132ba9ab8e71a7859e6a5d59851068d14", size = 122371, upload-time = "2025-08-09T07:57:28.46Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d6/98/f3b8013223728a99b908c9344da3aa04ee6e3fa235f19409033eda92fb78/charset_normalizer-3.4.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:fb7f67a1bfa6e40b438170ebdc8158b78dc465a5a67b6dde178a46987b244a72", size = 207695, upload-time = "2025-08-09T07:55:36.452Z" },
+    { url = "https://files.pythonhosted.org/packages/21/40/5188be1e3118c82dcb7c2a5ba101b783822cfb413a0268ed3be0468532de/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cc9370a2da1ac13f0153780040f465839e6cccb4a1e44810124b4e22483c93fe", size = 147153, upload-time = "2025-08-09T07:55:38.467Z" },
+    { url = "https://files.pythonhosted.org/packages/37/60/5d0d74bc1e1380f0b72c327948d9c2aca14b46a9efd87604e724260f384c/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:07a0eae9e2787b586e129fdcbe1af6997f8d0e5abaa0bc98c0e20e124d67e601", size = 160428, upload-time = "2025-08-09T07:55:40.072Z" },
+    { url = "https://files.pythonhosted.org/packages/85/9a/d891f63722d9158688de58d050c59dc3da560ea7f04f4c53e769de5140f5/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:74d77e25adda8581ffc1c720f1c81ca082921329452eba58b16233ab1842141c", size = 157627, upload-time = "2025-08-09T07:55:41.706Z" },
+    { url = "https://files.pythonhosted.org/packages/65/1a/7425c952944a6521a9cfa7e675343f83fd82085b8af2b1373a2409c683dc/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d0e909868420b7049dafd3a31d45125b31143eec59235311fc4c57ea26a4acd2", size = 152388, upload-time = "2025-08-09T07:55:43.262Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/c9/a2c9c2a355a8594ce2446085e2ec97fd44d323c684ff32042e2a6b718e1d/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:c6f162aabe9a91a309510d74eeb6507fab5fff92337a15acbe77753d88d9dcf0", size = 150077, upload-time = "2025-08-09T07:55:44.903Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/38/20a1f44e4851aa1c9105d6e7110c9d020e093dfa5836d712a5f074a12bf7/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:4ca4c094de7771a98d7fbd67d9e5dbf1eb73efa4f744a730437d8a3a5cf994f0", size = 161631, upload-time = "2025-08-09T07:55:46.346Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/fa/384d2c0f57edad03d7bec3ebefb462090d8905b4ff5a2d2525f3bb711fac/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:02425242e96bcf29a49711b0ca9f37e451da7c70562bc10e8ed992a5a7a25cc0", size = 159210, upload-time = "2025-08-09T07:55:47.539Z" },
+    { url = "https://files.pythonhosted.org/packages/33/9e/eca49d35867ca2db336b6ca27617deed4653b97ebf45dfc21311ce473c37/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:78deba4d8f9590fe4dae384aeff04082510a709957e968753ff3c48399f6f92a", size = 153739, upload-time = "2025-08-09T07:55:48.744Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/91/26c3036e62dfe8de8061182d33be5025e2424002125c9500faff74a6735e/charset_normalizer-3.4.3-cp310-cp310-win32.whl", hash = "sha256:d79c198e27580c8e958906f803e63cddb77653731be08851c7df0b1a14a8fc0f", size = 99825, upload-time = "2025-08-09T07:55:50.305Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/c6/f05db471f81af1fa01839d44ae2a8bfeec8d2a8b4590f16c4e7393afd323/charset_normalizer-3.4.3-cp310-cp310-win_amd64.whl", hash = "sha256:c6e490913a46fa054e03699c70019ab869e990270597018cef1d8562132c2669", size = 107452, upload-time = "2025-08-09T07:55:51.461Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/b5/991245018615474a60965a7c9cd2b4efbaabd16d582a5547c47ee1c7730b/charset_normalizer-3.4.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:b256ee2e749283ef3ddcff51a675ff43798d92d746d1a6e4631bf8c707d22d0b", size = 204483, upload-time = "2025-08-09T07:55:53.12Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/2a/ae245c41c06299ec18262825c1569c5d3298fc920e4ddf56ab011b417efd/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:13faeacfe61784e2559e690fc53fa4c5ae97c6fcedb8eb6fb8d0a15b475d2c64", size = 145520, upload-time = "2025-08-09T07:55:54.712Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/a4/b3b6c76e7a635748c4421d2b92c7b8f90a432f98bda5082049af37ffc8e3/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:00237675befef519d9af72169d8604a067d92755e84fe76492fef5441db05b91", size = 158876, upload-time = "2025-08-09T07:55:56.024Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/e6/63bb0e10f90a8243c5def74b5b105b3bbbfb3e7bb753915fe333fb0c11ea/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:585f3b2a80fbd26b048a0be90c5aae8f06605d3c92615911c3a2b03a8a3b796f", size = 156083, upload-time = "2025-08-09T07:55:57.582Z" },
+    { url = "https://files.pythonhosted.org/packages/87/df/b7737ff046c974b183ea9aa111b74185ac8c3a326c6262d413bd5a1b8c69/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0e78314bdc32fa80696f72fa16dc61168fda4d6a0c014e0380f9d02f0e5d8a07", size = 150295, upload-time = "2025-08-09T07:55:59.147Z" },
+    { url = "https://files.pythonhosted.org/packages/61/f1/190d9977e0084d3f1dc169acd060d479bbbc71b90bf3e7bf7b9927dec3eb/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:96b2b3d1a83ad55310de8c7b4a2d04d9277d5591f40761274856635acc5fcb30", size = 148379, upload-time = "2025-08-09T07:56:00.364Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/92/27dbe365d34c68cfe0ca76f1edd70e8705d82b378cb54ebbaeabc2e3029d/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:939578d9d8fd4299220161fdd76e86c6a251987476f5243e8864a7844476ba14", size = 160018, upload-time = "2025-08-09T07:56:01.678Z" },
+    { url = "https://files.pythonhosted.org/packages/99/04/baae2a1ea1893a01635d475b9261c889a18fd48393634b6270827869fa34/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:fd10de089bcdcd1be95a2f73dbe6254798ec1bda9f450d5828c96f93e2536b9c", size = 157430, upload-time = "2025-08-09T07:56:02.87Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/36/77da9c6a328c54d17b960c89eccacfab8271fdaaa228305330915b88afa9/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:1e8ac75d72fa3775e0b7cb7e4629cec13b7514d928d15ef8ea06bca03ef01cae", size = 151600, upload-time = "2025-08-09T07:56:04.089Z" },
+    { url = "https://files.pythonhosted.org/packages/64/d4/9eb4ff2c167edbbf08cdd28e19078bf195762e9bd63371689cab5ecd3d0d/charset_normalizer-3.4.3-cp311-cp311-win32.whl", hash = "sha256:6cf8fd4c04756b6b60146d98cd8a77d0cdae0e1ca20329da2ac85eed779b6849", size = 99616, upload-time = "2025-08-09T07:56:05.658Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/9c/996a4a028222e7761a96634d1820de8a744ff4327a00ada9c8942033089b/charset_normalizer-3.4.3-cp311-cp311-win_amd64.whl", hash = "sha256:31a9a6f775f9bcd865d88ee350f0ffb0e25936a7f930ca98995c05abf1faf21c", size = 107108, upload-time = "2025-08-09T07:56:07.176Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/5e/14c94999e418d9b87682734589404a25854d5f5d0408df68bc15b6ff54bb/charset_normalizer-3.4.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:e28e334d3ff134e88989d90ba04b47d84382a828c061d0d1027b1b12a62b39b1", size = 205655, upload-time = "2025-08-09T07:56:08.475Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/a8/c6ec5d389672521f644505a257f50544c074cf5fc292d5390331cd6fc9c3/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0cacf8f7297b0c4fcb74227692ca46b4a5852f8f4f24b3c766dd94a1075c4884", size = 146223, upload-time = "2025-08-09T07:56:09.708Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/eb/a2ffb08547f4e1e5415fb69eb7db25932c52a52bed371429648db4d84fb1/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c6fd51128a41297f5409deab284fecbe5305ebd7e5a1f959bee1c054622b7018", size = 159366, upload-time = "2025-08-09T07:56:11.326Z" },
+    { url = "https://files.pythonhosted.org/packages/82/10/0fd19f20c624b278dddaf83b8464dcddc2456cb4b02bb902a6da126b87a1/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:3cfb2aad70f2c6debfbcb717f23b7eb55febc0bb23dcffc0f076009da10c6392", size = 157104, upload-time = "2025-08-09T07:56:13.014Z" },
+    { url = "https://files.pythonhosted.org/packages/16/ab/0233c3231af734f5dfcf0844aa9582d5a1466c985bbed6cedab85af9bfe3/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1606f4a55c0fd363d754049cdf400175ee96c992b1f8018b993941f221221c5f", size = 151830, upload-time = "2025-08-09T07:56:14.428Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/02/e29e22b4e02839a0e4a06557b1999d0a47db3567e82989b5bb21f3fbbd9f/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:027b776c26d38b7f15b26a5da1044f376455fb3766df8fc38563b4efbc515154", size = 148854, upload-time = "2025-08-09T07:56:16.051Z" },
+    { url = "https://files.pythonhosted.org/packages/05/6b/e2539a0a4be302b481e8cafb5af8792da8093b486885a1ae4d15d452bcec/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:42e5088973e56e31e4fa58eb6bd709e42fc03799c11c42929592889a2e54c491", size = 160670, upload-time = "2025-08-09T07:56:17.314Z" },
+    { url = "https://files.pythonhosted.org/packages/31/e7/883ee5676a2ef217a40ce0bffcc3d0dfbf9e64cbcfbdf822c52981c3304b/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:cc34f233c9e71701040d772aa7490318673aa7164a0efe3172b2981218c26d93", size = 158501, upload-time = "2025-08-09T07:56:18.641Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/35/6525b21aa0db614cf8b5792d232021dca3df7f90a1944db934efa5d20bb1/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:320e8e66157cc4e247d9ddca8e21f427efc7a04bbd0ac8a9faf56583fa543f9f", size = 153173, upload-time = "2025-08-09T07:56:20.289Z" },
+    { url = "https://files.pythonhosted.org/packages/50/ee/f4704bad8201de513fdc8aac1cabc87e38c5818c93857140e06e772b5892/charset_normalizer-3.4.3-cp312-cp312-win32.whl", hash = "sha256:fb6fecfd65564f208cbf0fba07f107fb661bcd1a7c389edbced3f7a493f70e37", size = 99822, upload-time = "2025-08-09T07:56:21.551Z" },
+    { url = "https://files.pythonhosted.org/packages/39/f5/3b3836ca6064d0992c58c7561c6b6eee1b3892e9665d650c803bd5614522/charset_normalizer-3.4.3-cp312-cp312-win_amd64.whl", hash = "sha256:86df271bf921c2ee3818f0522e9a5b8092ca2ad8b065ece5d7d9d0e9f4849bcc", size = 107543, upload-time = "2025-08-09T07:56:23.115Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/ca/9a0983dd5c8e9733565cf3db4df2b0a2e9a82659fd8aa2a868ac6e4a991f/charset_normalizer-3.4.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:70bfc5f2c318afece2f5838ea5e4c3febada0be750fcf4775641052bbba14d05", size = 207520, upload-time = "2025-08-09T07:57:11.026Z" },
+    { url = "https://files.pythonhosted.org/packages/39/c6/99271dc37243a4f925b09090493fb96c9333d7992c6187f5cfe5312008d2/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:23b6b24d74478dc833444cbd927c338349d6ae852ba53a0d02a2de1fce45b96e", size = 147307, upload-time = "2025-08-09T07:57:12.4Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/69/132eab043356bba06eb333cc2cc60c6340857d0a2e4ca6dc2b51312886b3/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:34a7f768e3f985abdb42841e20e17b330ad3aaf4bb7e7aeeb73db2e70f077b99", size = 160448, upload-time = "2025-08-09T07:57:13.712Z" },
+    { url = "https://files.pythonhosted.org/packages/04/9a/914d294daa4809c57667b77470533e65def9c0be1ef8b4c1183a99170e9d/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:fb731e5deb0c7ef82d698b0f4c5bb724633ee2a489401594c5c88b02e6cb15f7", size = 157758, upload-time = "2025-08-09T07:57:14.979Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/a8/6f5bcf1bcf63cb45625f7c5cadca026121ff8a6c8a3256d8d8cd59302663/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:257f26fed7d7ff59921b78244f3cd93ed2af1800ff048c33f624c87475819dd7", size = 152487, upload-time = "2025-08-09T07:57:16.332Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/72/d3d0e9592f4e504f9dea08b8db270821c909558c353dc3b457ed2509f2fb/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1ef99f0456d3d46a50945c98de1774da86f8e992ab5c77865ea8b8195341fc19", size = 150054, upload-time = "2025-08-09T07:57:17.576Z" },
+    { url = "https://files.pythonhosted.org/packages/20/30/5f64fe3981677fe63fa987b80e6c01042eb5ff653ff7cec1b7bd9268e54e/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:2c322db9c8c89009a990ef07c3bcc9f011a3269bc06782f916cd3d9eed7c9312", size = 161703, upload-time = "2025-08-09T07:57:20.012Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/ef/dd08b2cac9284fd59e70f7d97382c33a3d0a926e45b15fc21b3308324ffd/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:511729f456829ef86ac41ca78c63a5cb55240ed23b4b737faca0eb1abb1c41bc", size = 159096, upload-time = "2025-08-09T07:57:21.329Z" },
+    { url = "https://files.pythonhosted.org/packages/45/8c/dcef87cfc2b3f002a6478f38906f9040302c68aebe21468090e39cde1445/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:88ab34806dea0671532d3f82d82b85e8fc23d7b2dd12fa837978dad9bb392a34", size = 153852, upload-time = "2025-08-09T07:57:22.608Z" },
+    { url = "https://files.pythonhosted.org/packages/63/86/9cbd533bd37883d467fcd1bd491b3547a3532d0fbb46de2b99feeebf185e/charset_normalizer-3.4.3-cp39-cp39-win32.whl", hash = "sha256:16a8770207946ac75703458e2c743631c79c59c5890c80011d536248f8eaa432", size = 99840, upload-time = "2025-08-09T07:57:23.883Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/d6/7e805c8e5c46ff9729c49950acc4ee0aeb55efb8b3a56687658ad10c3216/charset_normalizer-3.4.3-cp39-cp39-win_amd64.whl", hash = "sha256:d22dbedd33326a4a5190dd4fe9e9e693ef12160c77382d9e87919bce54f3d4ca", size = 107438, upload-time = "2025-08-09T07:57:25.287Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/1f/f041989e93b001bc4e44bb1669ccdcf54d3f00e628229a85b08d330615c5/charset_normalizer-3.4.3-py3-none-any.whl", hash = "sha256:ce571ab16d890d23b5c278547ba694193a45011ff86a9162a71307ed9f86759a", size = 53175, upload-time = "2025-08-09T07:57:26.864Z" },
+]
+
+[[package]]
+name = "chex"
+version = "0.1.90"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "absl-py" },
+    { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "jaxlib", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "setuptools", marker = "python_full_version >= '3.12'" },
+    { name = "toolz" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/77/70/53c7d404ce9e2a94009aea7f77ef6e392f6740e071c62683a506647c520f/chex-0.1.90.tar.gz", hash = "sha256:d3c375aeb6154b08f1cccd2bee4ed83659ee2198a6acf1160d2fe2e4a6c87b5c", size = 92363, upload-time = "2025-07-23T19:50:47.945Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6f/3d/46bb04776c465cea2dd8aa2d4b61ab610b707f798f47838ef7e6105b025c/chex-0.1.90-py3-none-any.whl", hash = "sha256:fce3de82588f72d4796e545e574a433aa29229cbdcf792555e41bead24b704ae", size = 101047, upload-time = "2025-07-23T19:50:46.603Z" },
 ]
 
 [[package]]
 name = "click"
 version = "8.1.8"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "colorama", marker = "python_full_version < '3.10' and sys_platform == 'win32'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b9/2e/0090cbf739cee7d23781ad4b89a9894a41538e4fcf4c31dcdd705b78eb8b/click-8.1.8.tar.gz", hash = "sha256:ed53c9d8990d83c2a27deae68e4ee337473f6330c040a31d4225c9574d16096a", size = 226593 }
+sdist = { url = "https://files.pythonhosted.org/packages/b9/2e/0090cbf739cee7d23781ad4b89a9894a41538e4fcf4c31dcdd705b78eb8b/click-8.1.8.tar.gz", hash = "sha256:ed53c9d8990d83c2a27deae68e4ee337473f6330c040a31d4225c9574d16096a", size = 226593, upload-time = "2024-12-21T18:38:44.339Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/7e/d4/7ebdbd03970677812aac39c869717059dbb71a4cfc033ca6e5221787892c/click-8.1.8-py3-none-any.whl", hash = "sha256:63c132bbbed01578a06712a2d1f497bb62d9c1c0d329b7903a866228027263b2", size = 98188 },
+    { url = "https://files.pythonhosted.org/packages/7e/d4/7ebdbd03970677812aac39c869717059dbb71a4cfc033ca6e5221787892c/click-8.1.8-py3-none-any.whl", hash = "sha256:63c132bbbed01578a06712a2d1f497bb62d9c1c0d329b7903a866228027263b2", size = 98188, upload-time = "2024-12-21T18:38:41.666Z" },
 ]
 
 [[package]]
-name = "cloudpickle"
-version = "3.1.1"
+name = "click"
+version = "8.2.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/52/39/069100b84d7418bc358d81669d5748efb14b9cceacd2f9c75f550424132f/cloudpickle-3.1.1.tar.gz", hash = "sha256:b216fa8ae4019d5482a8ac3c95d8f6346115d8835911fd4aefd1a445e4242c64", size = 22113 }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "colorama", marker = "python_full_version >= '3.10' and sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/60/6c/8ca2efa64cf75a977a0d7fac081354553ebe483345c734fb6b6515d96bbc/click-8.2.1.tar.gz", hash = "sha256:27c491cc05d968d271d5a1db13e3b5a184636d9d930f148c50b038f0d0646202", size = 286342, upload-time = "2025-05-20T23:19:49.832Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/7e/e8/64c37fadfc2816a7701fa8a6ed8d87327c7d54eacfbfb6edab14a2f2be75/cloudpickle-3.1.1-py3-none-any.whl", hash = "sha256:c8c5a44295039331ee9dad40ba100a9c7297b6f988e50e87ccdf3765a668350e", size = 20992 },
+    { url = "https://files.pythonhosted.org/packages/85/32/10bb5764d90a8eee674e9dc6f4db6a0ab47c8c4d0d83c27f7c39ac415a4d/click-8.2.1-py3-none-any.whl", hash = "sha256:61a3265b914e850b85317d0b3109c7f8cd35a670f963866005d6ef1d5175a12b", size = 102215, upload-time = "2025-05-20T23:19:47.796Z" },
 ]
 
 [[package]]
-name = "cmake"
-version = "3.31.4"
+name = "cloudpickle"
+version = "3.1.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/50/cb/3a327fa784a5dbaf838b135cb1729f43535c52d83bbf02191fb8a0cb118e/cmake-3.31.4.tar.gz", hash = "sha256:a6ac2242e0b16ad7d94c9f8572d6f232e6169747be50e5cdf497f206c4819ce1", size = 34278 }
+sdist = { url = "https://files.pythonhosted.org/packages/52/39/069100b84d7418bc358d81669d5748efb14b9cceacd2f9c75f550424132f/cloudpickle-3.1.1.tar.gz", hash = "sha256:b216fa8ae4019d5482a8ac3c95d8f6346115d8835911fd4aefd1a445e4242c64", size = 22113, upload-time = "2025-01-14T17:02:05.085Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d9/db/50efa1d3e29cb2a6e8e143e522e52698b3fc08f4b56100fb35f97a70af79/cmake-3.31.4-py3-none-macosx_10_10_universal2.whl", hash = "sha256:fc048b4b70facd16699a43c737f6782b4eff56e8e6093090db5979532d9db0f6", size = 47198138 },
-    { url = "https://files.pythonhosted.org/packages/c7/76/ccb8764761c739ef16bd8957a16ecbda01b03c2d7d241c376bfca6bf2822/cmake-3.31.4-py3-none-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:2a37be93534df04513f0845492d71bc80899c3f87b77e3b01c95aff1a7fc9bde", size = 27556485 },
-    { url = "https://files.pythonhosted.org/packages/ad/8e/888e2944655d7fa1ea5af46b60883a0e7847bbf9fb7ecc321c8e5f0a1394/cmake-3.31.4-py3-none-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:c9f5f8289c5e7bd2ed654cbac164021fa7723064fee0443a2f0068bc08413d81", size = 26808834 },
-    { url = "https://files.pythonhosted.org/packages/59/f4/0b2b1430a441c3c09ee102bf8c5d9ec1dc11d002ff4affef15c656f37ce9/cmake-3.31.4-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:926d91cae2ba7d2f3df857d0fc066bdac4f3904bf5c95e99b60435e85aabedb4", size = 27140820 },
-    { url = "https://files.pythonhosted.org/packages/d1/f9/a274b4e36e457d8e99db1038cc31a6c391bf3bc26230c2dc9caf37499753/cmake-3.31.4-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:929a8d8d289d69e43784661748ddd08933ce1ec5db8f9bcfce6ee817a48f8787", size = 28868269 },
-    { url = "https://files.pythonhosted.org/packages/9b/35/8da1ffa00a3f3853881aa5025cdf11c744303013df70c8716155b83825d3/cmake-3.31.4-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b463efdf5b92f3b290235aa9f8da092b3dac19b7636c563fd156022dab580649", size = 30732267 },
-    { url = "https://files.pythonhosted.org/packages/79/48/bb8485687f5a64d52ac68cfcb02e9b8e46a9e107f380c54d484b6632c87e/cmake-3.31.4-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:225d9a643b0b60ffce0399ff0cabd7a4820e0dbcb794e97d3aacfcf7c0589ae6", size = 26908885 },
-    { url = "https://files.pythonhosted.org/packages/e5/9e/2594d7fa8b263296497bf044469b4ab4797c51675ea629f9672011cdfe09/cmake-3.31.4-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:89143a5e2a5916061f2cfc5012e9fe6281aaf7c0dae7930bdc68d105d22ddc39", size = 27784555 },
-    { url = "https://files.pythonhosted.org/packages/95/16/5b1989f1d2287b05cd68792c0a48b721c060f728506d719fcf0e3b80ceb2/cmake-3.31.4-py3-none-manylinux_2_31_armv7l.whl", hash = "sha256:f96127bf663168accd29d5a50ee68ea80f26bcd37f96c7a14ef2378781f19936", size = 24965366 },
-    { url = "https://files.pythonhosted.org/packages/5a/4c/289fb0986c6ff63583383eca0c9479147f362330938856a9b5201c84cee8/cmake-3.31.4-py3-none-musllinux_1_1_aarch64.whl", hash = "sha256:25c5094394f0cee21130b5678e5b4552f72470e266df6d6fb1d5c505100f0eaa", size = 27824887 },
-    { url = "https://files.pythonhosted.org/packages/3c/f3/d45ba2b5bb54f4ef615a6a24cf6258600eec790a9d5017c9584107b445b9/cmake-3.31.4-py3-none-musllinux_1_1_i686.whl", hash = "sha256:466c9295af440bb4a47cc5e1af10576cf2227620528afd0fd0b3effa1d513b49", size = 31368421 },
-    { url = "https://files.pythonhosted.org/packages/34/3d/f6b712241ede5fb8e32c13e119c06e142f3f12ead1656721b1f67756106b/cmake-3.31.4-py3-none-musllinux_1_1_ppc64le.whl", hash = "sha256:f6af3b83a1b1fc1d990d18b6a566ee9c95c0393f986c6df15f2505dda8ad1bcc", size = 32074545 },
-    { url = "https://files.pythonhosted.org/packages/f0/23/48cd0404d7238d703a4cd4d7434eeaf12e8fbe68160d52f1489f55f582df/cmake-3.31.4-py3-none-musllinux_1_1_s390x.whl", hash = "sha256:23781e17563693a68b0cef85749746894b8a61488e56e96fc6649b73652e8236", size = 27946950 },
-    { url = "https://files.pythonhosted.org/packages/21/03/014d9710bccf5a7e04c6f6ee27bfaba1220e79ee145d7b95f84e7843729b/cmake-3.31.4-py3-none-musllinux_1_1_x86_64.whl", hash = "sha256:838a388b559137f3654d8cf30f62bbdec10f8d1c3624f0d289614d33cdf4fba1", size = 29473412 },
-    { url = "https://files.pythonhosted.org/packages/23/de/5a8142732f0a52dedac2887e0c105c9bbb449e517ade500e56bf2af520d1/cmake-3.31.4-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:a6a3b0b9557f41c955a6b25c94205f2ca9c3a46edca809ad87507c5ef6bc4274", size = 32971081 },
-    { url = "https://files.pythonhosted.org/packages/a5/a1/50c11f0b110986c753592f025970094030b25748df126abe8e38265be722/cmake-3.31.4-py3-none-win32.whl", hash = "sha256:d378c9e58eac906bddafd673c7571262dcd5a9946bb1e8f9e3902572a8fa95ca", size = 33351393 },
-    { url = "https://files.pythonhosted.org/packages/0c/7f/331d181b6b1b8942ec5fad23e98fff85218485f29f62f6bc60663d424df8/cmake-3.31.4-py3-none-win_amd64.whl", hash = "sha256:20be7cdb41903edf85e8a498c4beff8d6854acbb087abfb07c362c738bdf0018", size = 36496715 },
-    { url = "https://files.pythonhosted.org/packages/65/26/11a78723364716004928b7bea7d96cf2c72dc3abfaa7c163159110fcb649/cmake-3.31.4-py3-none-win_arm64.whl", hash = "sha256:9479a9255197c49e135df039d8484c69aa63158a06ae9c2d0eb939da2f0f7dff", size = 35559239 },
+    { url = "https://files.pythonhosted.org/packages/7e/e8/64c37fadfc2816a7701fa8a6ed8d87327c7d54eacfbfb6edab14a2f2be75/cloudpickle-3.1.1-py3-none-any.whl", hash = "sha256:c8c5a44295039331ee9dad40ba100a9c7297b6f988e50e87ccdf3765a668350e", size = 20992, upload-time = "2025-01-14T17:02:02.417Z" },
 ]
 
 [[package]]
 name = "colorama"
 version = "0.4.6"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/d8/53/6f443c9a4a8358a93a6792e2acffb9d9d5cb0a5cfd8802644b7b1c9a02e4/colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44", size = 27697 }
+sdist = { url = "https://files.pythonhosted.org/packages/d8/53/6f443c9a4a8358a93a6792e2acffb9d9d5cb0a5cfd8802644b7b1c9a02e4/colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44", size = 27697, upload-time = "2022-10-25T02:36:22.414Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6", size = 25335 },
+    { url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6", size = 25335, upload-time = "2022-10-25T02:36:20.889Z" },
 ]
 
 [[package]]
 name = "coverage"
-version = "7.6.10"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/84/ba/ac14d281f80aab516275012e8875991bb06203957aa1e19950139238d658/coverage-7.6.10.tar.gz", hash = "sha256:7fb105327c8f8f0682e29843e2ff96af9dcbe5bab8eeb4b398c6a33a16d80a23", size = 803868 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/c5/12/2a2a923edf4ddabdffed7ad6da50d96a5c126dae7b80a33df7310e329a1e/coverage-7.6.10-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:5c912978f7fbf47ef99cec50c4401340436d200d41d714c7a4766f377c5b7b78", size = 207982 },
-    { url = "https://files.pythonhosted.org/packages/ca/49/6985dbca9c7be3f3cb62a2e6e492a0c88b65bf40579e16c71ae9c33c6b23/coverage-7.6.10-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a01ec4af7dfeb96ff0078ad9a48810bb0cc8abcb0115180c6013a6b26237626c", size = 208414 },
-    { url = "https://files.pythonhosted.org/packages/35/93/287e8f1d1ed2646f4e0b2605d14616c9a8a2697d0d1b453815eb5c6cebdb/coverage-7.6.10-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a3b204c11e2b2d883946fe1d97f89403aa1811df28ce0447439178cc7463448a", size = 236860 },
-    { url = "https://files.pythonhosted.org/packages/de/e1/cfdb5627a03567a10031acc629b75d45a4ca1616e54f7133ca1fa366050a/coverage-7.6.10-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:32ee6d8491fcfc82652a37109f69dee9a830e9379166cb73c16d8dc5c2915165", size = 234758 },
-    { url = "https://files.pythonhosted.org/packages/6d/85/fc0de2bcda3f97c2ee9fe8568f7d48f7279e91068958e5b2cc19e0e5f600/coverage-7.6.10-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:675cefc4c06e3b4c876b85bfb7c59c5e2218167bbd4da5075cbe3b5790a28988", size = 235920 },
-    { url = "https://files.pythonhosted.org/packages/79/73/ef4ea0105531506a6f4cf4ba571a214b14a884630b567ed65b3d9c1975e1/coverage-7.6.10-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:f4f620668dbc6f5e909a0946a877310fb3d57aea8198bde792aae369ee1c23b5", size = 234986 },
-    { url = "https://files.pythonhosted.org/packages/c6/4d/75afcfe4432e2ad0405c6f27adeb109ff8976c5e636af8604f94f29fa3fc/coverage-7.6.10-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:4eea95ef275de7abaef630c9b2c002ffbc01918b726a39f5a4353916ec72d2f3", size = 233446 },
-    { url = "https://files.pythonhosted.org/packages/86/5b/efee56a89c16171288cafff022e8af44f8f94075c2d8da563c3935212871/coverage-7.6.10-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:e2f0280519e42b0a17550072861e0bc8a80a0870de260f9796157d3fca2733c5", size = 234566 },
-    { url = "https://files.pythonhosted.org/packages/f2/db/67770cceb4a64d3198bf2aa49946f411b85ec6b0a9b489e61c8467a4253b/coverage-7.6.10-cp310-cp310-win32.whl", hash = "sha256:bc67deb76bc3717f22e765ab3e07ee9c7a5e26b9019ca19a3b063d9f4b874244", size = 210675 },
-    { url = "https://files.pythonhosted.org/packages/8d/27/e8bfc43f5345ec2c27bc8a1fa77cdc5ce9dcf954445e11f14bb70b889d14/coverage-7.6.10-cp310-cp310-win_amd64.whl", hash = "sha256:0f460286cb94036455e703c66988851d970fdfd8acc2a1122ab7f4f904e4029e", size = 211518 },
-    { url = "https://files.pythonhosted.org/packages/85/d2/5e175fcf6766cf7501a8541d81778fd2f52f4870100e791f5327fd23270b/coverage-7.6.10-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:ea3c8f04b3e4af80e17bab607c386a830ffc2fb88a5484e1df756478cf70d1d3", size = 208088 },
-    { url = "https://files.pythonhosted.org/packages/4b/6f/06db4dc8fca33c13b673986e20e466fd936235a6ec1f0045c3853ac1b593/coverage-7.6.10-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:507a20fc863cae1d5720797761b42d2d87a04b3e5aeb682ef3b7332e90598f43", size = 208536 },
-    { url = "https://files.pythonhosted.org/packages/0d/62/c6a0cf80318c1c1af376d52df444da3608eafc913b82c84a4600d8349472/coverage-7.6.10-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d37a84878285b903c0fe21ac8794c6dab58150e9359f1aaebbeddd6412d53132", size = 240474 },
-    { url = "https://files.pythonhosted.org/packages/a3/59/750adafc2e57786d2e8739a46b680d4fb0fbc2d57fbcb161290a9f1ecf23/coverage-7.6.10-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a534738b47b0de1995f85f582d983d94031dffb48ab86c95bdf88dc62212142f", size = 237880 },
-    { url = "https://files.pythonhosted.org/packages/2c/f8/ef009b3b98e9f7033c19deb40d629354aab1d8b2d7f9cfec284dbedf5096/coverage-7.6.10-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0d7a2bf79378d8fb8afaa994f91bfd8215134f8631d27eba3e0e2c13546ce994", size = 239750 },
-    { url = "https://files.pythonhosted.org/packages/a6/e2/6622f3b70f5f5b59f705e680dae6db64421af05a5d1e389afd24dae62e5b/coverage-7.6.10-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:6713ba4b4ebc330f3def51df1d5d38fad60b66720948112f114968feb52d3f99", size = 238642 },
-    { url = "https://files.pythonhosted.org/packages/2d/10/57ac3f191a3c95c67844099514ff44e6e19b2915cd1c22269fb27f9b17b6/coverage-7.6.10-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:ab32947f481f7e8c763fa2c92fd9f44eeb143e7610c4ca9ecd6a36adab4081bd", size = 237266 },
-    { url = "https://files.pythonhosted.org/packages/ee/2d/7016f4ad9d553cabcb7333ed78ff9d27248ec4eba8dd21fa488254dff894/coverage-7.6.10-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:7bbd8c8f1b115b892e34ba66a097b915d3871db7ce0e6b9901f462ff3a975377", size = 238045 },
-    { url = "https://files.pythonhosted.org/packages/a7/fe/45af5c82389a71e0cae4546413266d2195c3744849669b0bab4b5f2c75da/coverage-7.6.10-cp311-cp311-win32.whl", hash = "sha256:299e91b274c5c9cdb64cbdf1b3e4a8fe538a7a86acdd08fae52301b28ba297f8", size = 210647 },
-    { url = "https://files.pythonhosted.org/packages/db/11/3f8e803a43b79bc534c6a506674da9d614e990e37118b4506faf70d46ed6/coverage-7.6.10-cp311-cp311-win_amd64.whl", hash = "sha256:489a01f94aa581dbd961f306e37d75d4ba16104bbfa2b0edb21d29b73be83609", size = 211508 },
-    { url = "https://files.pythonhosted.org/packages/86/77/19d09ea06f92fdf0487499283b1b7af06bc422ea94534c8fe3a4cd023641/coverage-7.6.10-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:27c6e64726b307782fa5cbe531e7647aee385a29b2107cd87ba7c0105a5d3853", size = 208281 },
-    { url = "https://files.pythonhosted.org/packages/b6/67/5479b9f2f99fcfb49c0d5cf61912a5255ef80b6e80a3cddba39c38146cf4/coverage-7.6.10-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:c56e097019e72c373bae32d946ecf9858fda841e48d82df7e81c63ac25554078", size = 208514 },
-    { url = "https://files.pythonhosted.org/packages/15/d1/febf59030ce1c83b7331c3546d7317e5120c5966471727aa7ac157729c4b/coverage-7.6.10-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c7827a5bc7bdb197b9e066cdf650b2887597ad124dd99777332776f7b7c7d0d0", size = 241537 },
-    { url = "https://files.pythonhosted.org/packages/4b/7e/5ac4c90192130e7cf8b63153fe620c8bfd9068f89a6d9b5f26f1550f7a26/coverage-7.6.10-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:204a8238afe787323a8b47d8be4df89772d5c1e4651b9ffa808552bdf20e1d50", size = 238572 },
-    { url = "https://files.pythonhosted.org/packages/dc/03/0334a79b26ecf59958f2fe9dd1f5ab3e2f88db876f5071933de39af09647/coverage-7.6.10-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e67926f51821b8e9deb6426ff3164870976fe414d033ad90ea75e7ed0c2e5022", size = 240639 },
-    { url = "https://files.pythonhosted.org/packages/d7/45/8a707f23c202208d7b286d78ad6233f50dcf929319b664b6cc18a03c1aae/coverage-7.6.10-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e78b270eadb5702938c3dbe9367f878249b5ef9a2fcc5360ac7bff694310d17b", size = 240072 },
-    { url = "https://files.pythonhosted.org/packages/66/02/603ce0ac2d02bc7b393279ef618940b4a0535b0868ee791140bda9ecfa40/coverage-7.6.10-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:714f942b9c15c3a7a5fe6876ce30af831c2ad4ce902410b7466b662358c852c0", size = 238386 },
-    { url = "https://files.pythonhosted.org/packages/04/62/4e6887e9be060f5d18f1dd58c2838b2d9646faf353232dec4e2d4b1c8644/coverage-7.6.10-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:abb02e2f5a3187b2ac4cd46b8ced85a0858230b577ccb2c62c81482ca7d18852", size = 240054 },
-    { url = "https://files.pythonhosted.org/packages/5c/74/83ae4151c170d8bd071924f212add22a0e62a7fe2b149edf016aeecad17c/coverage-7.6.10-cp312-cp312-win32.whl", hash = "sha256:55b201b97286cf61f5e76063f9e2a1d8d2972fc2fcfd2c1272530172fd28c359", size = 210904 },
-    { url = "https://files.pythonhosted.org/packages/c3/54/de0893186a221478f5880283119fc40483bc460b27c4c71d1b8bba3474b9/coverage-7.6.10-cp312-cp312-win_amd64.whl", hash = "sha256:e4ae5ac5e0d1e4edfc9b4b57b4cbecd5bc266a6915c500f358817a8496739247", size = 211692 },
-    { url = "https://files.pythonhosted.org/packages/25/6d/31883d78865529257bf847df5789e2ae80e99de8a460c3453dbfbe0db069/coverage-7.6.10-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:05fca8ba6a87aabdd2d30d0b6c838b50510b56cdcfc604d40760dae7153b73d9", size = 208308 },
-    { url = "https://files.pythonhosted.org/packages/70/22/3f2b129cc08de00c83b0ad6252e034320946abfc3e4235c009e57cfeee05/coverage-7.6.10-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:9e80eba8801c386f72e0712a0453431259c45c3249f0009aff537a517b52942b", size = 208565 },
-    { url = "https://files.pythonhosted.org/packages/97/0a/d89bc2d1cc61d3a8dfe9e9d75217b2be85f6c73ebf1b9e3c2f4e797f4531/coverage-7.6.10-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a372c89c939d57abe09e08c0578c1d212e7a678135d53aa16eec4430adc5e690", size = 241083 },
-    { url = "https://files.pythonhosted.org/packages/4c/81/6d64b88a00c7a7aaed3a657b8eaa0931f37a6395fcef61e53ff742b49c97/coverage-7.6.10-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ec22b5e7fe7a0fa8509181c4aac1db48f3dd4d3a566131b313d1efc102892c18", size = 238235 },
-    { url = "https://files.pythonhosted.org/packages/9a/0b/7797d4193f5adb4b837207ed87fecf5fc38f7cc612b369a8e8e12d9fa114/coverage-7.6.10-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:26bcf5c4df41cad1b19c84af71c22cbc9ea9a547fc973f1f2cc9a290002c8b3c", size = 240220 },
-    { url = "https://files.pythonhosted.org/packages/65/4d/6f83ca1bddcf8e51bf8ff71572f39a1c73c34cf50e752a952c34f24d0a60/coverage-7.6.10-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4e4630c26b6084c9b3cb53b15bd488f30ceb50b73c35c5ad7871b869cb7365fd", size = 239847 },
-    { url = "https://files.pythonhosted.org/packages/30/9d/2470df6aa146aff4c65fee0f87f58d2164a67533c771c9cc12ffcdb865d5/coverage-7.6.10-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:2396e8116db77789f819d2bc8a7e200232b7a282c66e0ae2d2cd84581a89757e", size = 237922 },
-    { url = "https://files.pythonhosted.org/packages/08/dd/723fef5d901e6a89f2507094db66c091449c8ba03272861eaefa773ad95c/coverage-7.6.10-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:79109c70cc0882e4d2d002fe69a24aa504dec0cc17169b3c7f41a1d341a73694", size = 239783 },
-    { url = "https://files.pythonhosted.org/packages/3d/f7/64d3298b2baf261cb35466000628706ce20a82d42faf9b771af447cd2b76/coverage-7.6.10-cp313-cp313-win32.whl", hash = "sha256:9e1747bab246d6ff2c4f28b4d186b205adced9f7bd9dc362051cc37c4a0c7bd6", size = 210965 },
-    { url = "https://files.pythonhosted.org/packages/d5/58/ec43499a7fc681212fe7742fe90b2bc361cdb72e3181ace1604247a5b24d/coverage-7.6.10-cp313-cp313-win_amd64.whl", hash = "sha256:254f1a3b1eef5f7ed23ef265eaa89c65c8c5b6b257327c149db1ca9d4a35f25e", size = 211719 },
-    { url = "https://files.pythonhosted.org/packages/ab/c9/f2857a135bcff4330c1e90e7d03446b036b2363d4ad37eb5e3a47bbac8a6/coverage-7.6.10-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:2ccf240eb719789cedbb9fd1338055de2761088202a9a0b73032857e53f612fe", size = 209050 },
-    { url = "https://files.pythonhosted.org/packages/aa/b3/f840e5bd777d8433caa9e4a1eb20503495709f697341ac1a8ee6a3c906ad/coverage-7.6.10-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:0c807ca74d5a5e64427c8805de15b9ca140bba13572d6d74e262f46f50b13273", size = 209321 },
-    { url = "https://files.pythonhosted.org/packages/85/7d/125a5362180fcc1c03d91850fc020f3831d5cda09319522bcfa6b2b70be7/coverage-7.6.10-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2bcfa46d7709b5a7ffe089075799b902020b62e7ee56ebaed2f4bdac04c508d8", size = 252039 },
-    { url = "https://files.pythonhosted.org/packages/a9/9c/4358bf3c74baf1f9bddd2baf3756b54c07f2cfd2535f0a47f1e7757e54b3/coverage-7.6.10-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4e0de1e902669dccbf80b0415fb6b43d27edca2fbd48c74da378923b05316098", size = 247758 },
-    { url = "https://files.pythonhosted.org/packages/cf/c7/de3eb6fc5263b26fab5cda3de7a0f80e317597a4bad4781859f72885f300/coverage-7.6.10-cp313-cp313t-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3f7b444c42bbc533aaae6b5a2166fd1a797cdb5eb58ee51a92bee1eb94a1e1cb", size = 250119 },
-    { url = "https://files.pythonhosted.org/packages/3e/e6/43de91f8ba2ec9140c6a4af1102141712949903dc732cf739167cfa7a3bc/coverage-7.6.10-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:b330368cb99ef72fcd2dc3ed260adf67b31499584dc8a20225e85bfe6f6cfed0", size = 249597 },
-    { url = "https://files.pythonhosted.org/packages/08/40/61158b5499aa2adf9e37bc6d0117e8f6788625b283d51e7e0c53cf340530/coverage-7.6.10-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:9a7cfb50515f87f7ed30bc882f68812fd98bc2852957df69f3003d22a2aa0abf", size = 247473 },
-    { url = "https://files.pythonhosted.org/packages/50/69/b3f2416725621e9f112e74e8470793d5b5995f146f596f133678a633b77e/coverage-7.6.10-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:6f93531882a5f68c28090f901b1d135de61b56331bba82028489bc51bdd818d2", size = 248737 },
-    { url = "https://files.pythonhosted.org/packages/3c/6e/fe899fb937657db6df31cc3e61c6968cb56d36d7326361847440a430152e/coverage-7.6.10-cp313-cp313t-win32.whl", hash = "sha256:89d76815a26197c858f53c7f6a656686ec392b25991f9e409bcef020cd532312", size = 211611 },
-    { url = "https://files.pythonhosted.org/packages/1c/55/52f5e66142a9d7bc93a15192eba7a78513d2abf6b3558d77b4ca32f5f424/coverage-7.6.10-cp313-cp313t-win_amd64.whl", hash = "sha256:54a5f0f43950a36312155dae55c505a76cd7f2b12d26abeebbe7a0b36dbc868d", size = 212781 },
-    { url = "https://files.pythonhosted.org/packages/40/41/473617aadf9a1c15bc2d56be65d90d7c29bfa50a957a67ef96462f7ebf8e/coverage-7.6.10-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:656c82b8a0ead8bba147de9a89bda95064874c91a3ed43a00e687f23cc19d53a", size = 207978 },
-    { url = "https://files.pythonhosted.org/packages/10/f6/480586607768b39a30e6910a3c4522139094ac0f1677028e1f4823688957/coverage-7.6.10-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:ccc2b70a7ed475c68ceb548bf69cec1e27305c1c2606a5eb7c3afff56a1b3b27", size = 208415 },
-    { url = "https://files.pythonhosted.org/packages/f1/af/439bb760f817deff6f4d38fe7da08d9dd7874a560241f1945bc3b4446550/coverage-7.6.10-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a5e37dc41d57ceba70956fa2fc5b63c26dba863c946ace9705f8eca99daecdc4", size = 236452 },
-    { url = "https://files.pythonhosted.org/packages/d0/13/481f4ceffcabe29ee2332e60efb52e4694f54a402f3ada2bcec10bb32e43/coverage-7.6.10-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0aa9692b4fdd83a4647eeb7db46410ea1322b5ed94cd1715ef09d1d5922ba87f", size = 234374 },
-    { url = "https://files.pythonhosted.org/packages/c5/59/4607ea9d6b1b73e905c7656da08d0b00cdf6e59f2293ec259e8914160025/coverage-7.6.10-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:aa744da1820678b475e4ba3dfd994c321c5b13381d1041fe9c608620e6676e25", size = 235505 },
-    { url = "https://files.pythonhosted.org/packages/85/60/d66365723b9b7f29464b11d024248ed3523ce5aab958e4ad8c43f3f4148b/coverage-7.6.10-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:c0b1818063dc9e9d838c09e3a473c1422f517889436dd980f5d721899e66f315", size = 234616 },
-    { url = "https://files.pythonhosted.org/packages/74/f8/2cf7a38e7d81b266f47dfcf137fecd8fa66c7bdbd4228d611628d8ca3437/coverage-7.6.10-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:59af35558ba08b758aec4d56182b222976330ef8d2feacbb93964f576a7e7a90", size = 233099 },
-    { url = "https://files.pythonhosted.org/packages/50/2b/bff6c1c6b63c4396ea7ecdbf8db1788b46046c681b8fcc6ec77db9f4ea49/coverage-7.6.10-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:7ed2f37cfce1ce101e6dffdfd1c99e729dd2ffc291d02d3e2d0af8b53d13840d", size = 234089 },
-    { url = "https://files.pythonhosted.org/packages/bf/b5/baace1c754d546a67779358341aa8d2f7118baf58cac235db457e1001d1b/coverage-7.6.10-cp39-cp39-win32.whl", hash = "sha256:4bcc276261505d82f0ad426870c3b12cb177752834a633e737ec5ee79bbdff18", size = 210701 },
-    { url = "https://files.pythonhosted.org/packages/b1/bf/9e1e95b8b20817398ecc5a1e8d3e05ff404e1b9fb2185cd71561698fe2a2/coverage-7.6.10-cp39-cp39-win_amd64.whl", hash = "sha256:457574f4599d2b00f7f637a0700a6422243b3565509457b2dbd3f50703e11f59", size = 211482 },
-    { url = "https://files.pythonhosted.org/packages/a1/70/de81bfec9ed38a64fc44a77c7665e20ca507fc3265597c28b0d989e4082e/coverage-7.6.10-pp39.pp310-none-any.whl", hash = "sha256:fd34e7b3405f0cc7ab03d54a334c17a9e802897580d964bd8c2001f4b9fd488f", size = 200223 },
+version = "7.10.5"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/61/83/153f54356c7c200013a752ce1ed5448573dca546ce125801afca9e1ac1a4/coverage-7.10.5.tar.gz", hash = "sha256:f2e57716a78bc3ae80b2207be0709a3b2b63b9f2dcf9740ee6ac03588a2015b6", size = 821662, upload-time = "2025-08-23T14:42:44.78Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/af/70/e77b0061a6c7157bfce645c6b9a715a08d4c86b3360a7b3252818080b817/coverage-7.10.5-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:c6a5c3414bfc7451b879141ce772c546985163cf553f08e0f135f0699a911801", size = 216774, upload-time = "2025-08-23T14:40:26.301Z" },
+    { url = "https://files.pythonhosted.org/packages/91/08/2a79de5ecf37ee40f2d898012306f11c161548753391cec763f92647837b/coverage-7.10.5-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:bc8e4d99ce82f1710cc3c125adc30fd1487d3cf6c2cd4994d78d68a47b16989a", size = 217175, upload-time = "2025-08-23T14:40:29.142Z" },
+    { url = "https://files.pythonhosted.org/packages/64/57/0171d69a699690149a6ba6a4eb702814448c8d617cf62dbafa7ce6bfdf63/coverage-7.10.5-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:02252dc1216e512a9311f596b3169fad54abcb13827a8d76d5630c798a50a754", size = 243931, upload-time = "2025-08-23T14:40:30.735Z" },
+    { url = "https://files.pythonhosted.org/packages/15/06/3a67662c55656702bd398a727a7f35df598eb11104fcb34f1ecbb070291a/coverage-7.10.5-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:73269df37883e02d460bee0cc16be90509faea1e3bd105d77360b512d5bb9c33", size = 245740, upload-time = "2025-08-23T14:40:32.302Z" },
+    { url = "https://files.pythonhosted.org/packages/00/f4/f8763aabf4dc30ef0d0012522d312f0b7f9fede6246a1f27dbcc4a1e523c/coverage-7.10.5-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1f8a81b0614642f91c9effd53eec284f965577591f51f547a1cbeb32035b4c2f", size = 247600, upload-time = "2025-08-23T14:40:33.66Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/31/6632219a9065e1b83f77eda116fed4c76fb64908a6a9feae41816dab8237/coverage-7.10.5-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:6a29f8e0adb7f8c2b95fa2d4566a1d6e6722e0a637634c6563cb1ab844427dd9", size = 245640, upload-time = "2025-08-23T14:40:35.248Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/e2/3dba9b86037b81649b11d192bb1df11dde9a81013e434af3520222707bc8/coverage-7.10.5-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fcf6ab569436b4a647d4e91accba12509ad9f2554bc93d3aee23cc596e7f99c3", size = 243659, upload-time = "2025-08-23T14:40:36.815Z" },
+    { url = "https://files.pythonhosted.org/packages/02/b9/57170bd9f3e333837fc24ecc88bc70fbc2eb7ccfd0876854b0c0407078c3/coverage-7.10.5-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:90dc3d6fb222b194a5de60af8d190bedeeddcbc7add317e4a3cd333ee6b7c879", size = 244537, upload-time = "2025-08-23T14:40:38.737Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/1c/93ac36ef1e8b06b8d5777393a3a40cb356f9f3dab980be40a6941e443588/coverage-7.10.5-cp310-cp310-win32.whl", hash = "sha256:414a568cd545f9dc75f0686a0049393de8098414b58ea071e03395505b73d7a8", size = 219285, upload-time = "2025-08-23T14:40:40.342Z" },
+    { url = "https://files.pythonhosted.org/packages/30/95/23252277e6e5fe649d6cd3ed3f35d2307e5166de4e75e66aa7f432abc46d/coverage-7.10.5-cp310-cp310-win_amd64.whl", hash = "sha256:e551f9d03347196271935fd3c0c165f0e8c049220280c1120de0084d65e9c7ff", size = 220185, upload-time = "2025-08-23T14:40:42.026Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/f2/336d34d2fc1291ca7c18eeb46f64985e6cef5a1a7ef6d9c23720c6527289/coverage-7.10.5-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c177e6ffe2ebc7c410785307758ee21258aa8e8092b44d09a2da767834f075f2", size = 216890, upload-time = "2025-08-23T14:40:43.627Z" },
+    { url = "https://files.pythonhosted.org/packages/39/ea/92448b07cc1cf2b429d0ce635f59cf0c626a5d8de21358f11e92174ff2a6/coverage-7.10.5-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:14d6071c51ad0f703d6440827eaa46386169b5fdced42631d5a5ac419616046f", size = 217287, upload-time = "2025-08-23T14:40:45.214Z" },
+    { url = "https://files.pythonhosted.org/packages/96/ba/ad5b36537c5179c808d0ecdf6e4aa7630b311b3c12747ad624dcd43a9b6b/coverage-7.10.5-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:61f78c7c3bc272a410c5ae3fde7792b4ffb4acc03d35a7df73ca8978826bb7ab", size = 247683, upload-time = "2025-08-23T14:40:46.791Z" },
+    { url = "https://files.pythonhosted.org/packages/28/e5/fe3bbc8d097029d284b5fb305b38bb3404895da48495f05bff025df62770/coverage-7.10.5-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f39071caa126f69d63f99b324fb08c7b1da2ec28cbb1fe7b5b1799926492f65c", size = 249614, upload-time = "2025-08-23T14:40:48.082Z" },
+    { url = "https://files.pythonhosted.org/packages/69/9c/a1c89a8c8712799efccb32cd0a1ee88e452f0c13a006b65bb2271f1ac767/coverage-7.10.5-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:343a023193f04d46edc46b2616cdbee68c94dd10208ecd3adc56fcc54ef2baa1", size = 251719, upload-time = "2025-08-23T14:40:49.349Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/be/5576b5625865aa95b5633315f8f4142b003a70c3d96e76f04487c3b5cc95/coverage-7.10.5-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:585ffe93ae5894d1ebdee69fc0b0d4b7c75d8007983692fb300ac98eed146f78", size = 249411, upload-time = "2025-08-23T14:40:50.624Z" },
+    { url = "https://files.pythonhosted.org/packages/94/0a/e39a113d4209da0dbbc9385608cdb1b0726a4d25f78672dc51c97cfea80f/coverage-7.10.5-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:b0ef4e66f006ed181df29b59921bd8fc7ed7cd6a9289295cd8b2824b49b570df", size = 247466, upload-time = "2025-08-23T14:40:52.362Z" },
+    { url = "https://files.pythonhosted.org/packages/40/cb/aebb2d8c9e3533ee340bea19b71c5b76605a0268aa49808e26fe96ec0a07/coverage-7.10.5-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:eb7b0bbf7cc1d0453b843eca7b5fa017874735bef9bfdfa4121373d2cc885ed6", size = 248104, upload-time = "2025-08-23T14:40:54.064Z" },
+    { url = "https://files.pythonhosted.org/packages/08/e6/26570d6ccce8ff5de912cbfd268e7f475f00597cb58da9991fa919c5e539/coverage-7.10.5-cp311-cp311-win32.whl", hash = "sha256:1d043a8a06987cc0c98516e57c4d3fc2c1591364831e9deb59c9e1b4937e8caf", size = 219327, upload-time = "2025-08-23T14:40:55.424Z" },
+    { url = "https://files.pythonhosted.org/packages/79/79/5f48525e366e518b36e66167e3b6e5db6fd54f63982500c6a5abb9d3dfbd/coverage-7.10.5-cp311-cp311-win_amd64.whl", hash = "sha256:fefafcca09c3ac56372ef64a40f5fe17c5592fab906e0fdffd09543f3012ba50", size = 220213, upload-time = "2025-08-23T14:40:56.724Z" },
+    { url = "https://files.pythonhosted.org/packages/40/3c/9058128b7b0bf333130c320b1eb1ae485623014a21ee196d68f7737f8610/coverage-7.10.5-cp311-cp311-win_arm64.whl", hash = "sha256:7e78b767da8b5fc5b2faa69bb001edafcd6f3995b42a331c53ef9572c55ceb82", size = 218893, upload-time = "2025-08-23T14:40:58.011Z" },
+    { url = "https://files.pythonhosted.org/packages/27/8e/40d75c7128f871ea0fd829d3e7e4a14460cad7c3826e3b472e6471ad05bd/coverage-7.10.5-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c2d05c7e73c60a4cecc7d9b60dbfd603b4ebc0adafaef371445b47d0f805c8a9", size = 217077, upload-time = "2025-08-23T14:40:59.329Z" },
+    { url = "https://files.pythonhosted.org/packages/18/a8/f333f4cf3fb5477a7f727b4d603a2eb5c3c5611c7fe01329c2e13b23b678/coverage-7.10.5-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:32ddaa3b2c509778ed5373b177eb2bf5662405493baeff52278a0b4f9415188b", size = 217310, upload-time = "2025-08-23T14:41:00.628Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/2c/fbecd8381e0a07d1547922be819b4543a901402f63930313a519b937c668/coverage-7.10.5-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:dd382410039fe062097aa0292ab6335a3f1e7af7bba2ef8d27dcda484918f20c", size = 248802, upload-time = "2025-08-23T14:41:02.012Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/bc/1011da599b414fb6c9c0f34086736126f9ff71f841755786a6b87601b088/coverage-7.10.5-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7fa22800f3908df31cea6fb230f20ac49e343515d968cc3a42b30d5c3ebf9b5a", size = 251550, upload-time = "2025-08-23T14:41:03.438Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/6f/b5c03c0c721c067d21bc697accc3642f3cef9f087dac429c918c37a37437/coverage-7.10.5-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f366a57ac81f5e12797136552f5b7502fa053c861a009b91b80ed51f2ce651c6", size = 252684, upload-time = "2025-08-23T14:41:04.85Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/50/d474bc300ebcb6a38a1047d5c465a227605d6473e49b4e0d793102312bc5/coverage-7.10.5-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5f1dc8f1980a272ad4a6c84cba7981792344dad33bf5869361576b7aef42733a", size = 250602, upload-time = "2025-08-23T14:41:06.719Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/2d/548c8e04249cbba3aba6bd799efdd11eee3941b70253733f5d355d689559/coverage-7.10.5-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:2285c04ee8676f7938b02b4936d9b9b672064daab3187c20f73a55f3d70e6b4a", size = 248724, upload-time = "2025-08-23T14:41:08.429Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/96/a7c3c0562266ac39dcad271d0eec8fc20ab576e3e2f64130a845ad2a557b/coverage-7.10.5-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:c2492e4dd9daab63f5f56286f8a04c51323d237631eb98505d87e4c4ff19ec34", size = 250158, upload-time = "2025-08-23T14:41:09.749Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/75/74d4be58c70c42ef0b352d597b022baf12dbe2b43e7cb1525f56a0fb1d4b/coverage-7.10.5-cp312-cp312-win32.whl", hash = "sha256:38a9109c4ee8135d5df5505384fc2f20287a47ccbe0b3f04c53c9a1989c2bbaf", size = 219493, upload-time = "2025-08-23T14:41:11.095Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/08/364e6012d1d4d09d1e27437382967efed971d7613f94bca9add25f0c1f2b/coverage-7.10.5-cp312-cp312-win_amd64.whl", hash = "sha256:6b87f1ad60b30bc3c43c66afa7db6b22a3109902e28c5094957626a0143a001f", size = 220302, upload-time = "2025-08-23T14:41:12.449Z" },
+    { url = "https://files.pythonhosted.org/packages/db/d5/7c8a365e1f7355c58af4fe5faf3f90cc8e587590f5854808d17ccb4e7077/coverage-7.10.5-cp312-cp312-win_arm64.whl", hash = "sha256:672a6c1da5aea6c629819a0e1461e89d244f78d7b60c424ecf4f1f2556c041d8", size = 218936, upload-time = "2025-08-23T14:41:13.872Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/21/05248e8bc74683488cb7477e6b6b878decadd15af0ec96f56381d3d7ff2d/coverage-7.10.5-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:62835c1b00c4a4ace24c1a88561a5a59b612fbb83a525d1c70ff5720c97c0610", size = 216763, upload-time = "2025-08-23T14:42:26.75Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/7f/161a0ad40cb1c7e19dc1aae106d3430cc88dac3d651796d6cf3f3730c800/coverage-7.10.5-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:5255b3bbcc1d32a4069d6403820ac8e6dbcc1d68cb28a60a1ebf17e47028e898", size = 217154, upload-time = "2025-08-23T14:42:28.238Z" },
+    { url = "https://files.pythonhosted.org/packages/de/31/41929ee53af829ea5a88e71d335ea09d0bb587a3da1c5e58e59b48473ed8/coverage-7.10.5-cp39-cp39-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:3876385722e335d6e991c430302c24251ef9c2a9701b2b390f5473199b1b8ebf", size = 243588, upload-time = "2025-08-23T14:42:29.798Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/4e/2649344e33eeb3567041e8255a1942173cae81817fe06b60f3fafaafe111/coverage-7.10.5-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:8048ce4b149c93447a55d279078c8ae98b08a6951a3c4d2d7e87f4efc7bfe100", size = 245412, upload-time = "2025-08-23T14:42:31.296Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/b1/b21e1e69986ad89b051dd42c3ef06d9326e03ac3c0c844fc33385d1d9e35/coverage-7.10.5-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4028e7558e268dd8bcf4d9484aad393cafa654c24b4885f6f9474bf53183a82a", size = 247182, upload-time = "2025-08-23T14:42:33.155Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/b5/80837be411ae092e03fcc2a7877bd9a659c531eff50453e463057a9eee44/coverage-7.10.5-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:03f47dc870eec0367fcdd603ca6a01517d2504e83dc18dbfafae37faec66129a", size = 245066, upload-time = "2025-08-23T14:42:34.754Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/ed/fcb0838ddf149d68d09f89af57397b0dd9d26b100cc729daf1b0caf0b2d3/coverage-7.10.5-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:2d488d7d42b6ded7ea0704884f89dcabd2619505457de8fc9a6011c62106f6e5", size = 243138, upload-time = "2025-08-23T14:42:36.311Z" },
+    { url = "https://files.pythonhosted.org/packages/75/0f/505c6af24a9ae5d8919d209b9c31b7092815f468fa43bec3b1118232c62a/coverage-7.10.5-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:b3dcf2ead47fa8be14224ee817dfc1df98043af568fe120a22f81c0eb3c34ad2", size = 244095, upload-time = "2025-08-23T14:42:38.227Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/7e/c82a8bede46217c1d944bd19b65e7106633b998640f00ab49c5f747a5844/coverage-7.10.5-cp39-cp39-win32.whl", hash = "sha256:02650a11324b80057b8c9c29487020073d5e98a498f1857f37e3f9b6ea1b2426", size = 219289, upload-time = "2025-08-23T14:42:39.827Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/ac/46645ef6be543f2e7de08cc2601a0b67e130c816be3b749ab741be689fb9/coverage-7.10.5-cp39-cp39-win_amd64.whl", hash = "sha256:b45264dd450a10f9e03237b41a9a24e85cbb1e278e5a32adb1a303f58f0017f3", size = 220199, upload-time = "2025-08-23T14:42:41.363Z" },
+    { url = "https://files.pythonhosted.org/packages/08/b6/fff6609354deba9aeec466e4bcaeb9d1ed3e5d60b14b57df2a36fb2273f2/coverage-7.10.5-py3-none-any.whl", hash = "sha256:0be24d35e4db1d23d0db5c0f6a74a962e2ec83c426b5cac09f4234aadef38e4a", size = 208736, upload-time = "2025-08-23T14:42:43.145Z" },
 ]
 
 [package.optional-dependencies]
@@ -432,101 +480,18 @@ toml = [
     { name = "tomli", marker = "python_full_version <= '3.11'" },
 ]
 
-[[package]]
-name = "cramjam"
-version = "2.9.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/c9/68/09b6b5603d21a0c7d4362d513217a5079c47b1b7a88967c52dbef13db183/cramjam-2.9.1.tar.gz", hash = "sha256:336cc591d86cbd225d256813779f46624f857bc9c779db126271eff9ddc524ae", size = 47892 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/27/5d/0b03115fa6a95a6dd9be344cd186879b763f1a6fab57ae55ffe2777aa0a7/cramjam-2.9.1-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:8e82464d1e00fbbb12958999b8471ba5e9f3d9711954505a0a7b378762332e6f", size = 2136622 },
-    { url = "https://files.pythonhosted.org/packages/6f/ac/a17644e182ede7e8e24fb3af038bc2c1cf3dd0447c935cb10409f21d099b/cramjam-2.9.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6d2df8a6511cc08ef1fccd2e0c65e2ebc9f57574ec8376052a76851af5398810", size = 1927947 },
-    { url = "https://files.pythonhosted.org/packages/9e/1e/e6c4f9695e4ba7b9c63160dcbfa76428bd3221930eedeb8f16364ab6f642/cramjam-2.9.1-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:21ea784e6c3f1843d3523ae0f03651dd06058b39eeb64beb82ee3b100fa83662", size = 2268766 },
-    { url = "https://files.pythonhosted.org/packages/ab/37/4c81e5d039bdfc75a695abd426e6cdd9ab18a87f65d57837d78936cfa226/cramjam-2.9.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e0c5d98a4e791f0bbd0ffcb7dae879baeb2dcc357348a8dc2be0a8c10403a2a", size = 2108762 },
-    { url = "https://files.pythonhosted.org/packages/b9/bb/3bf3a8877b9a4105b625d710410bd2bc83ef38d4a7fe4eaeb3895d997b2d/cramjam-2.9.1-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e076fd87089197cb61117c63dbe7712ad5eccb93968860eb3bae09b767bac813", size = 2086694 },
-    { url = "https://files.pythonhosted.org/packages/c3/78/317b7ab6a9b0f24c45d56305a8288cdb6408f855034dc80530ed16a5cc6c/cramjam-2.9.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6d86b44933aea0151e4a2e1e6935448499849045c38167d288ca4c59d5b8cd4e", size = 2441698 },
-    { url = "https://files.pythonhosted.org/packages/c5/2d/bc98992c29eb8647196b3bda814fd7ecfba6aff85177d44180be2aa320e8/cramjam-2.9.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7eb032549dec897b942ddcf80c1cdccbcb40629f15fc902731dbe6362da49326", size = 2759280 },
-    { url = "https://files.pythonhosted.org/packages/dd/64/a4e54d74110c22477e467586935167d61fc7bae5284d393e76779b214a3e/cramjam-2.9.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cf29b4def86ec503e329fe138842a9b79a997e3beb6c7809b05665a0d291edff", size = 2385128 },
-    { url = "https://files.pythonhosted.org/packages/b0/1a/6ee093bf8a41cf31980175310abbbcdd1a39dadadbe96843112f42cef0fe/cramjam-2.9.1-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:a36adf7d13b7accfa206e1c917f08924eb905b45aa8e62176509afa7b14db71e", size = 2373494 },
-    { url = "https://files.pythonhosted.org/packages/9d/a6/1ae1f1a8ef559c2fab9d6d7f09b19995684e6727e617bf1b73967ee1c6be/cramjam-2.9.1-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:cf4ea758d98b6fad1b4b2d808d0de690d3162ac56c26968aea0af6524e3eb736", size = 2386900 },
-    { url = "https://files.pythonhosted.org/packages/d9/e6/cf18deeaa0a96e7fc87f0eacde3c97e2893b573ac148ec746655570c18fc/cramjam-2.9.1-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:4826d6d81ea490fa7a3ae7a4b9729866a945ffac1f77fe57b71e49d6e1b21efd", size = 2400609 },
-    { url = "https://files.pythonhosted.org/packages/90/97/98a8fa24249dc72a936a9a51a81407a399070ba4ceb528d0af291c760eff/cramjam-2.9.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:335103317475bf992953c58838152a4761fc3c87354000edbfc4d7e57cf05909", size = 2553159 },
-    { url = "https://files.pythonhosted.org/packages/ae/6b/4f71f72bc3405f221ec8bd2ba869e324d5f87ddd58c14bf59f7937ea37ab/cramjam-2.9.1-cp310-cp310-win32.whl", hash = "sha256:258120cb1e3afc3443f756f9de161ed63eed56a2c31f6093e81c571c0f2dc9f6", size = 1817873 },
-    { url = "https://files.pythonhosted.org/packages/8e/f4/32639916897d59e94d286b5b22263ce8c2903ecc93a868ebe9443ece8f12/cramjam-2.9.1-cp310-cp310-win_amd64.whl", hash = "sha256:c60e5996aa02547d12bc2740d44e90e006b0f93100f53206f7abe6732ad56e69", size = 2092168 },
-    { url = "https://files.pythonhosted.org/packages/6c/28/dd2b62be30ffe1fa8df10c99ba7b46abfbfb2fc6ace6acbbf9264a1a6b48/cramjam-2.9.1-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:b9db1debe48060e41a5b91af9193c524e473c57f6105462c5524a41f5aabdb88", size = 2136699 },
-    { url = "https://files.pythonhosted.org/packages/03/c9/fcebeb6f06879af4226337715fbc42ffe543158bcba8c244bba144767897/cramjam-2.9.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:f6f18f0242212d3409d26ce3874937b5b979cebd61f08b633a6ea893c32fc7b6", size = 1927934 },
-    { url = "https://files.pythonhosted.org/packages/e8/f3/77032e4f5db4dfcc2b0365f92655b7d6f3fc1527ea5b637f9fb9f8156a65/cramjam-2.9.1-cp311-cp311-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:b5b1cd7d39242b2b903cf09cd4696b3a6e04dc537ffa9f3ac8668edae76eecb6", size = 2268584 },
-    { url = "https://files.pythonhosted.org/packages/38/16/52175e94390f57196382783a3386c122ace7656b57339abaacdc9433b609/cramjam-2.9.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a47de0a68f5f4d9951250ef5af31f2a7228132caa9ed60994234f7eb98090d33", size = 2108599 },
-    { url = "https://files.pythonhosted.org/packages/99/25/5f7476d127a8d18cd19a2f3fd25c0fe09ef7848069d23aac70bc96385eb6/cramjam-2.9.1-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e13c9a697881e5e38148958612dc6856967f5ff8cd7bba5ff751f2d6ac020aa4", size = 2086632 },
-    { url = "https://files.pythonhosted.org/packages/7b/97/76ff3e1209add6acb7e2aa7997be48dc1f92ad66ee3e8fa1179eb2bb9b44/cramjam-2.9.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ba560244bc1335b420b74e91e35f9d4e7f307a3be3a4603ce0f0d7e15a0acdf0", size = 2441757 },
-    { url = "https://files.pythonhosted.org/packages/69/c4/228e74c30576556d11e54d86f356955cd86ff5e11bbfec74b66ed0dd237d/cramjam-2.9.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8d47fd41ce260cf4f0ff0e788de961fab9e9c6844a05ce55d06ce31e06107bdc", size = 2758144 },
-    { url = "https://files.pythonhosted.org/packages/4b/e7/0fd22e12c6a2879abc501979779d4b8cfe8fe692c708c2c0d1664e88fd79/cramjam-2.9.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:84d154fbadece82935396eb6bcb502085d944d2fd13b07a94348364344370c2c", size = 2385062 },
-    { url = "https://files.pythonhosted.org/packages/dd/9c/845592ddf9eb7130ae8bc5958a01d469304a43f8071effe164e2d239e3fa/cramjam-2.9.1-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:038df668ffb94d64d67b6ecc59cbd206745a425ffc0402897dde12d89fa6a870", size = 2373473 },
-    { url = "https://files.pythonhosted.org/packages/10/c2/287cc94b7f8e87e3b0c21819d3a5deead99ebfdcb2b2d85cd04011b37292/cramjam-2.9.1-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:4125d8cd86fa08495d310e80926c2f0563f157b76862e7479f9b2cf94823ea0c", size = 2386816 },
-    { url = "https://files.pythonhosted.org/packages/7c/22/869a1eeea53db4d9fbde6693a2465909762bffeab1a671e193c95b26f99f/cramjam-2.9.1-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:4206ebdd1d1ef0f3f86c8c2f7c426aa4af6094f4f41e274601fd4c4569f37454", size = 2400713 },
-    { url = "https://files.pythonhosted.org/packages/3f/89/ff988bd6427f01041ccb1a9104c05b6373ae476682d317b6844f4b40af92/cramjam-2.9.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:ab687bef5c493732b9a4ab870542ee43f5eae0025f9c684c7cb399c3a85cb380", size = 2553081 },
-    { url = "https://files.pythonhosted.org/packages/2e/68/13fa8561335de609f3cd40b132c1a3abbaf26d3c277e8b8a7446de34ef2c/cramjam-2.9.1-cp311-cp311-win32.whl", hash = "sha256:dda7698b6d7caeae1047adafebc4b43b2a82478234f6c2b45bc3edad854e0600", size = 1817782 },
-    { url = "https://files.pythonhosted.org/packages/94/75/f3506ee802460e3b86a91e53bba1f67cf457fa04e4316fe7d5823ba5d28b/cramjam-2.9.1-cp311-cp311-win_amd64.whl", hash = "sha256:872b00ff83e84bcbdc7e951af291ebe65eed20b09c47e7c4af21c312f90b796f", size = 2092227 },
-    { url = "https://files.pythonhosted.org/packages/56/66/69a1c17331e38b02c78c923262fc315272de7c2618ef7eac8b3358969d90/cramjam-2.9.1-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:79417957972553502b217a0093532e48893c8b4ca30ccc941cefe9c72379df7c", size = 2132273 },
-    { url = "https://files.pythonhosted.org/packages/3d/17/23d0b1d3301480e924545cdd27f2b949c50438949f64c74e800a09c12c37/cramjam-2.9.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce2b94117f373defc876f88e74e44049a9969223dbca3240415b71752d0422fb", size = 1926919 },
-    { url = "https://files.pythonhosted.org/packages/8e/da/e9565f4abbbaa14645ccd7ce83f9631e90955454b87dc3ef9208aebc72e6/cramjam-2.9.1-cp312-cp312-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:67040e0fd84404885ec716a806bee6110f9960c3647e0ef1670aab3b7375a70a", size = 2271776 },
-    { url = "https://files.pythonhosted.org/packages/88/ac/e6e0794ac01deb52e7a6a3e59720699abdee08d9b9c63a8d8874201d8155/cramjam-2.9.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0bedb84e068b53c944bd08dcb501fd00d67daa8a917922356dd559b484ce7eab", size = 2109248 },
-    { url = "https://files.pythonhosted.org/packages/22/0f/c3724b2dcdfbe7e07917803cf7a6db4a874818a6f8d2b95ca1ceaf177170/cramjam-2.9.1-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:06e3f97a379386d97debf08638a78b3d3850fdf6124755eb270b54905a169930", size = 2088611 },
-    { url = "https://files.pythonhosted.org/packages/ce/16/929a5ae899ad6298f58e66622dc223476fe8e1d4e8dae608f4e1a34bfd09/cramjam-2.9.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:11118675e9c7952ececabc62f023290ee4f8ecf0bee0d2c7eb8d1c402ee9769d", size = 2438373 },
-    { url = "https://files.pythonhosted.org/packages/2a/2a/ad473f1ca65d3285e8c1d99fc0289f5856224c0d452dabcf856fd4dcdd77/cramjam-2.9.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6b7de6b61b11545570e4d6033713f3599525efc615ee353a822be8f6b0c65b77", size = 2836669 },
-    { url = "https://files.pythonhosted.org/packages/9b/5a/e9b4868ee27099a2a21646cf5ea5cf08c660eae90b55a395ada974dcf3fb/cramjam-2.9.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:57ca8f3775324a9de3ee6f05ca172687ba258c0dea79f7e3a6b4112834982f2a", size = 2343995 },
-    { url = "https://files.pythonhosted.org/packages/5f/c4/870a9b4524107bf85a207b82a42613318881238b20f2d237e62815af646a/cramjam-2.9.1-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:9847dd6f288f1c56359f52acb48ff2df848ff3e3bff34d23855bbcf7016427cc", size = 2374270 },
-    { url = "https://files.pythonhosted.org/packages/70/4b/b69e8e3951b7cec5e7da2539b7573bb396bed66af07d760b1878b00fd120/cramjam-2.9.1-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:8d1248dfa7f151e893ce819670f00879e4b7650b8d4c01279ce4f12140d68dd2", size = 2388789 },
-    { url = "https://files.pythonhosted.org/packages/05/1a/af02f6192060413314735c0db61259d7279b0d8d99eee29eff2af09c5892/cramjam-2.9.1-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:9da6d970281083bae91b914362de325414aa03c01fc806f6bb2cc006322ec834", size = 2402459 },
-    { url = "https://files.pythonhosted.org/packages/20/9a/a4ab3e90d72eb4f2c1b983fa32b4050ba676f533ba15bd78158f0632295a/cramjam-2.9.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:1c33bc095db5733c841a102b8693062be5db8cdac17b9782ebc00577c6a94480", size = 2518440 },
-    { url = "https://files.pythonhosted.org/packages/35/3b/e632dd7e2c5c8a2af2d83144b00d6840f1afcf9c6959ed59ec5b0f925288/cramjam-2.9.1-cp312-cp312-win32.whl", hash = "sha256:9e9193cd4bb57e7acd3af24891526299244bfed88168945efdaa09af4e50720f", size = 1822630 },
-    { url = "https://files.pythonhosted.org/packages/0e/a2/d1c46618b81b83578d58a62f3709046c4f3b4ddba10df4b9797cfe096b98/cramjam-2.9.1-cp312-cp312-win_amd64.whl", hash = "sha256:15955dd75e80f66c1ea271167a5347661d9bdc365f894a57698c383c9b7d465c", size = 2094684 },
-    { url = "https://files.pythonhosted.org/packages/85/45/f1d1e6ffdceb3b0c18511df2f8e779e03972459fb71d7c1ab0f6a5c063a3/cramjam-2.9.1-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:5a7797a2fff994fc5e323f7a967a35a3e37e3006ed21d64dcded086502f482af", size = 2131814 },
-    { url = "https://files.pythonhosted.org/packages/3a/96/36bbd431fbf0fa2ff51fd2db4c3bead66e9e373693a8455d411d45125a68/cramjam-2.9.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d51b9b140b1df39a44bff7896d98a10da345b7d5f5ce92368d328c1c2c829167", size = 1926380 },
-    { url = "https://files.pythonhosted.org/packages/67/c4/99b6507ec697d5f56d32c9c04614775004b05b7fa870725a492dc6b639eb/cramjam-2.9.1-cp313-cp313-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:07ac76b7f992556e7aa910244be11ece578cdf84f4d5d5297461f9a895e18312", size = 2271581 },
-    { url = "https://files.pythonhosted.org/packages/cb/1b/6d55dff244fb22c0b686dd5a96a754c0638f8a94056beb27c457c6035cc5/cramjam-2.9.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d90a72608c7550cd7eba914668f6277bfb0b24f074d1f1bd9d061fcb6f2adbd6", size = 2109255 },
-    { url = "https://files.pythonhosted.org/packages/ca/fb/b9fcf492a21a8d978c6f999025fce2c6656399448c017ed2fc859425f37f/cramjam-2.9.1-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:56495975401b1821dbe1f29cf222e23556232209a2fdb809fe8156d120ca9c7f", size = 2088323 },
-    { url = "https://files.pythonhosted.org/packages/88/1f/69b523395aeaa201dbd53d203453288205a0c651e7c910161892d694eb4d/cramjam-2.9.1-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3b695259e71fde6d5be66b77a4474523ced9ffe9fe8a34cb9b520ec1241a14d3", size = 2437930 },
-    { url = "https://files.pythonhosted.org/packages/b0/2c/d07e802f1786c4082e8286db1087563e4fab31cd6534ed31523f1f9584d1/cramjam-2.9.1-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ab1e69dc4831bbb79b6d547077aae89074c83e8ad94eba1a3d80e94d2424fd02", size = 2836655 },
-    { url = "https://files.pythonhosted.org/packages/1f/f5/6b425e82395c078bc95a7437b685e6bdba39d28c2b2986d79374fc1681aa/cramjam-2.9.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:440b489902bfb7a26d3fec1ca888007615336ff763d2a32a2fc40586548a0dbf", size = 2387107 },
-    { url = "https://files.pythonhosted.org/packages/33/65/7bf97d89ba7607aaea5464af6f249e3d94c291acf73d72768367a3e361c0/cramjam-2.9.1-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:217fe22b41f8c3dce03852f828b059abfad11d1344a1df2f43d3eb8634b18d75", size = 2374006 },
-    { url = "https://files.pythonhosted.org/packages/29/11/8b6c82eda6d0affbc15d7ab4dc758856eb4308e8ddae73300c1648f5aa0f/cramjam-2.9.1-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:95f3646ddc98af25af25d5692ae65966488a283813336ea9cf41b22e542e7c0d", size = 2388731 },
-    { url = "https://files.pythonhosted.org/packages/48/25/6cdd57c0b1a83c98aec9029310d09a6c1a31e9e9fb8efd9001bd0cbea992/cramjam-2.9.1-cp313-cp313-musllinux_1_1_i686.whl", hash = "sha256:6b19fc60ead1cae9795a5b359599da3a1c95d38f869bdfb51c441fd76b04e926", size = 2402131 },
-    { url = "https://files.pythonhosted.org/packages/b4/e7/cbf80c9647fa582432aa833c4bdd20cf437917c8066ce653e3b78deff658/cramjam-2.9.1-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:8dc5207567459d049696f62a1fdfb220f3fe6aa0d722285d44753e12504dac6c", size = 2555296 },
-    { url = "https://files.pythonhosted.org/packages/18/a6/fabe1959a980f5d2783a6c138311509dd168bd76e62018624a91cd1cbb41/cramjam-2.9.1-cp313-cp313-win32.whl", hash = "sha256:fbfe35929a61b914de9e5dbacde0cfbba86cbf5122f9285a24c14ed0b645490b", size = 1822484 },
-    { url = "https://files.pythonhosted.org/packages/55/d5/24e4562771711711c466768c92097640ed97b0283abe9043ffb6c6d4cf04/cramjam-2.9.1-cp313-cp313-win_amd64.whl", hash = "sha256:06068bd191a82ad4fc1ac23d6f8627fb5e37ec4be0431711b9a2dbacaccfeddb", size = 2094445 },
-    { url = "https://files.pythonhosted.org/packages/c7/5a/50523fd478390acb6ca8e57239f7cf79f7260dc0d16be89137d47823e50a/cramjam-2.9.1-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:af39006faddfc6253beb93ca821d544931cfee7f0177b99ff106dfd8fd6a2cd8", size = 2137158 },
-    { url = "https://files.pythonhosted.org/packages/df/83/54eca302e431d51149074d8aad6ec588870c5797060e2142dfe6ca3599a8/cramjam-2.9.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:b3291be0d3f73d5774d69013be4ab33978c777363b5312d14f62f77817c2f75a", size = 1927910 },
-    { url = "https://files.pythonhosted.org/packages/6d/e9/5d38ffa5376c5bffcbd16545707d9dac6beffccd00410f0cc19d83d85ef7/cramjam-2.9.1-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:1539fd758f0e57fad7913cebff8baaee871bb561ddf6fa710a427b74da6b6778", size = 2269458 },
-    { url = "https://files.pythonhosted.org/packages/15/f3/99fedc4210db1967256e602fdcb60947585421fd659f8baeeeb4ea16e4c7/cramjam-2.9.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ff362f68bd68ac0eccb445209238d589bba728fb6d7f2e9dc199e0ec3a61d6e0", size = 2109406 },
-    { url = "https://files.pythonhosted.org/packages/f2/e9/f380e0c1bd03046c522da4fd6d43ea897ba0b832c78fc4ea5708d8c35c21/cramjam-2.9.1-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:23b9786d1d17686fb8d600ade2a19374c7188d4b8867efa9af0d8274a220aec7", size = 2086677 },
-    { url = "https://files.pythonhosted.org/packages/13/a7/3ae887753f6d41f6e4af8e25654d103c56e13dda2f4b4d13acac570c65c1/cramjam-2.9.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8bc9c2c748aaf91863d89c4583f529c1c709485c94f8dfeb3ee48662d88e3258", size = 2442136 },
-    { url = "https://files.pythonhosted.org/packages/de/a2/763fd98340936057e44ea0b870c9cdb87ad5f90d49e492e8a11cf74e7b29/cramjam-2.9.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fd0fa9a0e7f18224b6d2d1d69dbdc3aecec80ef1393c59244159b131604a4395", size = 2754985 },
-    { url = "https://files.pythonhosted.org/packages/33/31/7c8cdf6b16fcd46bad4a307c8203a58b7a2fddf6cb3aad9dc441c050f62f/cramjam-2.9.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2ceef6e09ee22457997370882aa3c69de01e6dd0aaa2f953e1e87ad11641d042", size = 2385597 },
-    { url = "https://files.pythonhosted.org/packages/dd/ba/ec0f3b5a3a90721bdb42f4f4989b60adf823d137f40365e83df0cd299378/cramjam-2.9.1-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:1376f6fdbf0b30712413a0b4e51663a4938ae2f6b449f8e4635dbb3694db83cf", size = 2374339 },
-    { url = "https://files.pythonhosted.org/packages/ff/0a/f5bccdc8d12821aed4473a427e9eb8282a38c9337a30e02ed102b18941bf/cramjam-2.9.1-cp39-cp39-musllinux_1_1_armv7l.whl", hash = "sha256:342fb946f8d3e9e35b837288b03ab23cfbe0bb5a30e582ed805ef79706823a96", size = 2386933 },
-    { url = "https://files.pythonhosted.org/packages/a0/6e/ce3ffad2b3b8cb73156a19345e27a2e27fb5be79b64f2c81b0c6d6e16c57/cramjam-2.9.1-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:a237064a6e2c2256c9a1cf2beb7c971382190c0f1eb2e810e02e971881756132", size = 2400860 },
-    { url = "https://files.pythonhosted.org/packages/32/a9/e4509e5dfc8f41d9e7f9fdddbf567967937303621d410197c86b11d6a3e4/cramjam-2.9.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:53145fc9f2319c1245d4329e1da8cfacd6e35e27090c07c0b9d453ae2bbdac3e", size = 2553681 },
-    { url = "https://files.pythonhosted.org/packages/0a/83/52401c5c654ddff2850d890b0f1cfc355ff6887c6def420d0c8d8178ff97/cramjam-2.9.1-cp39-cp39-win32.whl", hash = "sha256:8a9f52c27292c21457f43c4ce124939302a9acfb62295e7cda8667310563a5a3", size = 1818130 },
-    { url = "https://files.pythonhosted.org/packages/93/b3/1645986d8b915fd0426a7224cd00c2c17c32b4d69bc5faad3fb3f5fd5081/cramjam-2.9.1-cp39-cp39-win_amd64.whl", hash = "sha256:8097ee39b61c86848a443c0b25b2df1de6b331fd512b20836a4f5cfde51ab255", size = 2092440 },
-    { url = "https://files.pythonhosted.org/packages/bc/91/3f7884172573072a4280bc8bc19b7562b2cd66d2a65576b11e72115cd5fe/cramjam-2.9.1-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:86824c695688fcd06c5ac9bbd3fea9bdfb4cca194b1e706fbf11a629df48d2b4", size = 2159537 },
-    { url = "https://files.pythonhosted.org/packages/ef/49/a0a89e9c45413e89a1e408d4ab416c0f88f19f6db7571fd5c517e429e276/cramjam-2.9.1-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:27571bfa5a5d618604696747d0dc1d2a99b5906c967c8dee53c13a7107edfde6", size = 1936244 },
-    { url = "https://files.pythonhosted.org/packages/26/f7/6422b9e4d148f1a351c0358a95d59023f25cab76609b180804f6a3ed17e9/cramjam-2.9.1-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fb01f6e38719818778144d3165a89ea1ad9dc58c6342b7f20aa194c70f34cbd1", size = 2119487 },
-    { url = "https://files.pythonhosted.org/packages/b5/59/6fc930217f7ae085eca6d22d3477cd0145a105cdc39e63b834cb0c1b25e3/cramjam-2.9.1-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b5cef5cf40725fe64592af9ec163e7389855077700678a1d94bec549403a74d", size = 2400910 },
-    { url = "https://files.pythonhosted.org/packages/2d/36/7e53cf5aaed4b446490e298f7571e69ce15d0dfb148feabe8bf02e58827f/cramjam-2.9.1-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:ac48b978aa0675f62b642750e798c394a64d25ce852e4e541f69bef9a564c2f0", size = 2100860 },
-]
-
 [[package]]
 name = "datasets"
-version = "3.2.0"
+version = "4.0.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "aiohttp" },
     { name = "dill" },
     { name = "filelock" },
     { name = "fsspec", extra = ["http"] },
     { name = "huggingface-hub" },
     { name = "multiprocess" },
-    { name = "numpy" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "packaging" },
     { name = "pandas" },
     { name = "pyarrow" },
@@ -535,265 +500,352 @@ dependencies = [
     { name = "tqdm" },
     { name = "xxhash" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fc/48/744286c044e2b942d4fa67f92816126522ad1f0675def0ea3264e6242005/datasets-3.2.0.tar.gz", hash = "sha256:9a6e1a356052866b5dbdd9c9eedb000bf3fc43d986e3584d9b028f4976937229", size = 558366 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/d7/84/0df6c5981f5fc722381662ff8cfbdf8aad64bec875f75d80b55bfef394ce/datasets-3.2.0-py3-none-any.whl", hash = "sha256:f3d2ba2698b7284a4518019658596a6a8bc79f31e51516524249d6c59cf0fe2a", size = 480647 },
-]
-
-[[package]]
-name = "diff-cover"
-version = "9.2.0"
-source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "python_full_version < '3.9.17'",
-]
-dependencies = [
-    { name = "chardet", marker = "python_full_version < '3.9.17'" },
-    { name = "jinja2", marker = "python_full_version < '3.9.17'" },
-    { name = "pluggy", marker = "python_full_version < '3.9.17'" },
-    { name = "pygments", marker = "python_full_version < '3.9.17'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/44/3a/e49ccba052a4dda264fbad4f467739ecc63498f7223bfc03d4bfac23ea95/diff_cover-9.2.0.tar.gz", hash = "sha256:85a0b353ebbb678f9e87ea303f75b545bd0baca38f563219bb72f2ae862bba36", size = 94857 }
+sdist = { url = "https://files.pythonhosted.org/packages/e3/9d/348ed92110ba5f9b70b51ca1078d4809767a835aa2b7ce7e74ad2b98323d/datasets-4.0.0.tar.gz", hash = "sha256:9657e7140a9050db13443ba21cb5de185af8af944479b00e7ff1e00a61c8dbf1", size = 569566, upload-time = "2025-07-09T14:35:52.431Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b4/8a/bddb8e4aea550066559144e72d3566e9ae2f757b8ac154e769c563f48177/diff_cover-9.2.0-py3-none-any.whl", hash = "sha256:1e24edc51c39e810c47dd9986e76c333ed95859655c091f572e590c39cabbdbe", size = 52561 },
+    { url = "https://files.pythonhosted.org/packages/eb/62/eb8157afb21bd229c864521c1ab4fa8e9b4f1b06bafdd8c4668a7a31b5dd/datasets-4.0.0-py3-none-any.whl", hash = "sha256:7ef95e62025fd122882dbce6cb904c8cd3fbc829de6669a5eb939c77d50e203d", size = 494825, upload-time = "2025-07-09T14:35:50.658Z" },
 ]
 
 [[package]]
 name = "diff-cover"
-version = "9.2.1"
+version = "9.6.0"
 source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.12.*'",
-    "python_full_version == '3.11.*'",
-    "python_full_version == '3.10.*'",
-    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
-]
 dependencies = [
-    { name = "chardet", marker = "python_full_version >= '3.9.17'" },
-    { name = "jinja2", marker = "python_full_version >= '3.9.17'" },
-    { name = "pluggy", marker = "python_full_version >= '3.9.17'" },
-    { name = "pygments", marker = "python_full_version >= '3.9.17'" },
+    { name = "chardet" },
+    { name = "jinja2" },
+    { name = "pluggy" },
+    { name = "pygments" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/36/b0/f3ccf97926f6e5cc76d5ece42f3c685d75673d1886fcec62886b8b00c51a/diff_cover-9.2.1.tar.gz", hash = "sha256:5fa5b2d71ccf5d16cd222a71c2ca069d9bf5fa3d657f6fac9b4d9c23379323bf", size = 94964 }
+sdist = { url = "https://files.pythonhosted.org/packages/4c/81/0496938bb72606231b2518b6fd404a5adfde3c1b60a97d0dc39a92b25570/diff_cover-9.6.0.tar.gz", hash = "sha256:75e5bc056dcaa68c6c87c9fb4e07c9e60daef15b6e8d034d56d2da9e2c84a872", size = 99756, upload-time = "2025-07-17T02:18:44.286Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e2/67/5785b653e392a17bec8a340fc5f4f9278ff3588098cc96b1b3b8f9c3f167/diff_cover-9.2.1-py3-none-any.whl", hash = "sha256:97393d4acaa7545eae86f66e287ad96ffbeef70c188cca5a7f433983363e3849", size = 52560 },
+    { url = "https://files.pythonhosted.org/packages/30/c5/5f3b69c74328c6a3a3a52a084afbd0c2d15c4b95e7784b752072c133f01d/diff_cover-9.6.0-py3-none-any.whl", hash = "sha256:29fbeb52d77a0b8c811e5580d5dbf41801a838da2ed54319a599da8f7233c547", size = 55671, upload-time = "2025-07-17T02:18:42.572Z" },
 ]
 
 [[package]]
 name = "dill"
 version = "0.3.8"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/17/4d/ac7ffa80c69ea1df30a8aa11b3578692a5118e7cd1aa157e3ef73b092d15/dill-0.3.8.tar.gz", hash = "sha256:3ebe3c479ad625c4553aca177444d89b486b1d84982eeacded644afc0cf797ca", size = 184847 }
+sdist = { url = "https://files.pythonhosted.org/packages/17/4d/ac7ffa80c69ea1df30a8aa11b3578692a5118e7cd1aa157e3ef73b092d15/dill-0.3.8.tar.gz", hash = "sha256:3ebe3c479ad625c4553aca177444d89b486b1d84982eeacded644afc0cf797ca", size = 184847, upload-time = "2024-01-27T23:42:16.145Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c9/7a/cef76fd8438a42f96db64ddaa85280485a9c395e7df3db8158cfec1eee34/dill-0.3.8-py3-none-any.whl", hash = "sha256:c36ca9ffb54365bdd2f8eb3eff7d2a21237f8452b57ace88b1ac615b7e815bd7", size = 116252 },
+    { url = "https://files.pythonhosted.org/packages/c9/7a/cef76fd8438a42f96db64ddaa85280485a9c395e7df3db8158cfec1eee34/dill-0.3.8-py3-none-any.whl", hash = "sha256:c36ca9ffb54365bdd2f8eb3eff7d2a21237f8452b57ace88b1ac615b7e815bd7", size = 116252, upload-time = "2024-01-27T23:42:14.239Z" },
 ]
 
 [[package]]
 name = "diskcache"
 version = "5.6.3"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/3f/21/1c1ffc1a039ddcc459db43cc108658f32c57d271d7289a2794e401d0fdb6/diskcache-5.6.3.tar.gz", hash = "sha256:2c3a3fa2743d8535d832ec61c2054a1641f41775aa7c556758a109941e33e4fc", size = 67916 }
+sdist = { url = "https://files.pythonhosted.org/packages/3f/21/1c1ffc1a039ddcc459db43cc108658f32c57d271d7289a2794e401d0fdb6/diskcache-5.6.3.tar.gz", hash = "sha256:2c3a3fa2743d8535d832ec61c2054a1641f41775aa7c556758a109941e33e4fc", size = 67916, upload-time = "2023-08-31T06:12:00.316Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/3f/27/4570e78fc0bf5ea0ca45eb1de3818a23787af9b390c0b0a0033a1b8236f9/diskcache-5.6.3-py3-none-any.whl", hash = "sha256:5e31b2d5fbad117cc363ebaf6b689474db18a1f6438bc82358b024abd4c2ca19", size = 45550 },
+    { url = "https://files.pythonhosted.org/packages/3f/27/4570e78fc0bf5ea0ca45eb1de3818a23787af9b390c0b0a0033a1b8236f9/diskcache-5.6.3-py3-none-any.whl", hash = "sha256:5e31b2d5fbad117cc363ebaf6b689474db18a1f6438bc82358b024abd4c2ca19", size = 45550, upload-time = "2023-08-31T06:11:58.822Z" },
 ]
 
 [[package]]
 name = "distlib"
-version = "0.3.9"
+version = "0.4.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/0d/dd/1bec4c5ddb504ca60fc29472f3d27e8d4da1257a854e1d96742f15c1d02d/distlib-0.3.9.tar.gz", hash = "sha256:a60f20dea646b8a33f3e7772f74dc0b2d0772d2837ee1342a00645c81edf9403", size = 613923 }
+sdist = { url = "https://files.pythonhosted.org/packages/96/8e/709914eb2b5749865801041647dc7f4e6d00b549cfe88b65ca192995f07c/distlib-0.4.0.tar.gz", hash = "sha256:feec40075be03a04501a973d81f633735b4b69f98b05450592310c0f401a4e0d", size = 614605, upload-time = "2025-07-17T16:52:00.465Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/91/a1/cf2472db20f7ce4a6be1253a81cfdf85ad9c7885ffbed7047fb72c24cf87/distlib-0.3.9-py2.py3-none-any.whl", hash = "sha256:47f8c22fd27c27e25a65601af709b38e4f0a45ea4fc2e710f65755fa8caaaf87", size = 468973 },
+    { url = "https://files.pythonhosted.org/packages/33/6b/e0547afaf41bf2c42e52430072fa5658766e3d65bd4b03a563d1b6336f57/distlib-0.4.0-py2.py3-none-any.whl", hash = "sha256:9659f7d87e46584a30b5780e43ac7a2143098441670ff0a49d5f9034c54a6c16", size = 469047, upload-time = "2025-07-17T16:51:58.613Z" },
 ]
 
 [[package]]
 name = "distro"
 version = "1.9.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/fc/f8/98eea607f65de6527f8a2e8885fc8015d3e6f5775df186e443e0964a11c3/distro-1.9.0.tar.gz", hash = "sha256:2fa77c6fd8940f116ee1d6b94a2f90b13b5ea8d019b98bc8bafdcabcdd9bdbed", size = 60722 }
+sdist = { url = "https://files.pythonhosted.org/packages/fc/f8/98eea607f65de6527f8a2e8885fc8015d3e6f5775df186e443e0964a11c3/distro-1.9.0.tar.gz", hash = "sha256:2fa77c6fd8940f116ee1d6b94a2f90b13b5ea8d019b98bc8bafdcabcdd9bdbed", size = 60722, upload-time = "2023-12-24T09:54:32.31Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/12/b3/231ffd4ab1fc9d679809f356cebee130ac7daa00d6d6f3206dd4fd137e9e/distro-1.9.0-py3-none-any.whl", hash = "sha256:7bffd925d65168f85027d8da9af6bddab658135b840670a223589bc0c8ef02b2", size = 20277, upload-time = "2023-12-24T09:54:30.421Z" },
+]
+
+[[package]]
+name = "dottxt"
+version = "0.1.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "backoff" },
+    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pydantic" },
+    { name = "pydantic-settings" },
+    { name = "python-dateutil" },
+    { name = "typing-extensions" },
+    { name = "urllib3" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/46/0c/c0507e1641185890ef173bce883d9a02d60b446441fd674db352d9cf52e5/dottxt-0.1.5-py3-none-any.whl", hash = "sha256:645af5f9c43721893b7be91c3a18d1a45bee0f7544caf370f54587df0e127dd2", size = 65293, upload-time = "2025-02-14T17:10:12.353Z" },
+]
+
+[[package]]
+name = "etils"
+version = "1.5.2"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ad/fa/8637c95271dd9eed00e258184295dd00ba163bb8924ba3be978ec89f093f/etils-1.5.2.tar.gz", hash = "sha256:ba6a3e1aff95c769130776aa176c11540637f5dd881f3b79172a5149b6b1c446", size = 87021, upload-time = "2023-10-24T12:22:58.996Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0f/6a/d2aaebacf73d5da7126c632ec0d9dc2df99cc4bbd259bad48904a034fc1b/etils-1.5.2-py3-none-any.whl", hash = "sha256:6dc882d355e1e98a5d1a148d6323679dc47c9a5792939b9de72615aa4737eb0b", size = 140603, upload-time = "2023-10-24T12:22:57.101Z" },
+]
+
+[package.optional-dependencies]
+epath = [
+    { name = "fsspec", marker = "python_full_version < '3.10'" },
+    { name = "importlib-resources", marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+    { name = "zipp", marker = "python_full_version < '3.10'" },
+]
+epy = [
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+]
+
+[[package]]
+name = "etils"
+version = "1.13.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9b/a0/522bbff0f3cdd37968f90dd7f26c7aa801ed87f5ba335f156de7f2b88a48/etils-1.13.0.tar.gz", hash = "sha256:a5b60c71f95bcd2d43d4e9fb3dc3879120c1f60472bb5ce19f7a860b1d44f607", size = 106368, upload-time = "2025-07-15T10:29:10.563Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/12/b3/231ffd4ab1fc9d679809f356cebee130ac7daa00d6d6f3206dd4fd137e9e/distro-1.9.0-py3-none-any.whl", hash = "sha256:7bffd925d65168f85027d8da9af6bddab658135b840670a223589bc0c8ef02b2", size = 20277 },
+    { url = "https://files.pythonhosted.org/packages/e7/98/87b5946356095738cb90a6df7b35ff69ac5750f6e783d5fbcc5cb3b6cbd7/etils-1.13.0-py3-none-any.whl", hash = "sha256:d9cd4f40fbe77ad6613b7348a18132cc511237b6c076dbb89105c0b520a4c6bb", size = 170603, upload-time = "2025-07-15T10:29:09.076Z" },
+]
+
+[package.optional-dependencies]
+epath = [
+    { name = "fsspec", marker = "python_full_version >= '3.10'" },
+    { name = "importlib-resources", marker = "python_full_version >= '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
+    { name = "zipp", marker = "python_full_version >= '3.10'" },
+]
+epy = [
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
 
 [[package]]
 name = "exceptiongroup"
-version = "1.2.2"
+version = "1.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/09/35/2495c4ac46b980e4ca1f6ad6db102322ef3ad2410b79fdde159a4b0f3b92/exceptiongroup-1.2.2.tar.gz", hash = "sha256:47c2edf7c6738fafb49fd34290706d1a1a2f4d1c6df275526b62cbb4aa5393cc", size = 28883 }
+dependencies = [
+    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/0b/9f/a65090624ecf468cdca03533906e7c69ed7588582240cfe7cc9e770b50eb/exceptiongroup-1.3.0.tar.gz", hash = "sha256:b241f5885f560bc56a59ee63ca4c6a8bfa46ae4ad651af316d4e81817bb9fd88", size = 29749, upload-time = "2025-05-10T17:42:51.123Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/02/cc/b7e31358aac6ed1ef2bb790a9746ac2c69bcb3c8588b41616914eb106eaf/exceptiongroup-1.2.2-py3-none-any.whl", hash = "sha256:3111b9d131c238bec2f8f516e123e14ba243563fb135d3fe885990585aa7795b", size = 16453 },
+    { url = "https://files.pythonhosted.org/packages/36/f4/c6e662dade71f56cd2f3735141b265c3c79293c109549c1e6933b0651ffc/exceptiongroup-1.3.0-py3-none-any.whl", hash = "sha256:4d111e6e0c13d0644cad6ddaa7ed0261a0b36971f6d23e7ec9b4b9097da78a10", size = 16674, upload-time = "2025-05-10T17:42:49.33Z" },
 ]
 
 [[package]]
 name = "fastapi"
-version = "0.115.6"
+version = "0.116.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "pydantic" },
     { name = "starlette" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/93/72/d83b98cd106541e8f5e5bfab8ef2974ab45a62e8a6c5b5e6940f26d2ed4b/fastapi-0.115.6.tar.gz", hash = "sha256:9ec46f7addc14ea472958a96aae5b5de65f39721a46aaf5705c480d9a8b76654", size = 301336 }
+sdist = { url = "https://files.pythonhosted.org/packages/78/d7/6c8b3bfe33eeffa208183ec037fee0cce9f7f024089ab1c5d12ef04bd27c/fastapi-0.116.1.tar.gz", hash = "sha256:ed52cbf946abfd70c5a0dccb24673f0670deeb517a88b3544d03c2a6bf283143", size = 296485, upload-time = "2025-07-11T16:22:32.057Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e5/47/d63c60f59a59467fda0f93f46335c9d18526d7071f025cb5b89d5353ea42/fastapi-0.116.1-py3-none-any.whl", hash = "sha256:c46ac7c312df840f0c9e220f7964bada936781bc4e2e6eb71f1c4d7553786565", size = 95631, upload-time = "2025-07-11T16:22:30.485Z" },
+]
+
+[[package]]
+name = "filelock"
+version = "3.19.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/40/bb/0ab3e58d22305b6f5440629d20683af28959bf793d98d11950e305c1c326/filelock-3.19.1.tar.gz", hash = "sha256:66eda1888b0171c998b35be2bcc0f6d75c388a7ce20c3f3f37aa8e96c2dddf58", size = 17687, upload-time = "2025-08-14T16:56:03.016Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/42/14/42b2651a2f46b022ccd948bca9f2d5af0fd8929c4eec235b8d6d844fbe67/filelock-3.19.1-py3-none-any.whl", hash = "sha256:d38e30481def20772f5baf097c122c3babc4fcdb7e14e57049eb9d88c6dc017d", size = 15988, upload-time = "2025-08-14T16:56:01.633Z" },
+]
+
+[[package]]
+name = "flatbuffers"
+version = "25.2.10"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e4/30/eb5dce7994fc71a2f685d98ec33cc660c0a5887db5610137e60d8cbc4489/flatbuffers-25.2.10.tar.gz", hash = "sha256:97e451377a41262f8d9bd4295cc836133415cc03d8cb966410a4af92eb00d26e", size = 22170, upload-time = "2025-02-11T04:26:46.257Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/52/b3/7e4df40e585df024fac2f80d1a2d579c854ac37109675db2b0cc22c0bb9e/fastapi-0.115.6-py3-none-any.whl", hash = "sha256:e9240b29e36fa8f4bb7290316988e90c381e5092e0cbe84e7818cc3713bcf305", size = 94843 },
+    { url = "https://files.pythonhosted.org/packages/b8/25/155f9f080d5e4bc0082edfda032ea2bc2b8fab3f4d25d46c1e9dd22a1a89/flatbuffers-25.2.10-py2.py3-none-any.whl", hash = "sha256:ebba5f4d5ea615af3f7fd70fc310636fbb2bbd1f566ac0a23d98dd412de50051", size = 30953, upload-time = "2025-02-11T04:26:44.484Z" },
 ]
 
 [[package]]
-name = "fastparquet"
-version = "2024.11.0"
+name = "flax"
+version = "0.8.5"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "cramjam" },
-    { name = "fsspec" },
-    { name = "numpy" },
-    { name = "packaging" },
-    { name = "pandas" },
+    { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "msgpack", marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "optax", version = "0.2.4", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "orbax-checkpoint", version = "0.6.4", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pyyaml", marker = "python_full_version < '3.10'" },
+    { name = "rich", marker = "python_full_version < '3.10'" },
+    { name = "tensorstore", version = "0.1.69", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b4/66/862da14f5fde4eff2cedc0f51a8dc34ba145088e5041b45b2d57ac54f922/fastparquet-2024.11.0.tar.gz", hash = "sha256:e3b1fc73fd3e1b70b0de254bae7feb890436cb67e99458b88cb9bd3cc44db419", size = 467192 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/3d/56/476f5b83476a256489879b78513bee737691a80905e246a2daa30ebcc362/fastparquet-2024.11.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:60ccf587410f0979105e17036df61bb60e1c2b81880dc91895cdb4ee65b71e7f", size = 910272 },
-    { url = "https://files.pythonhosted.org/packages/3b/ad/4ce73440df874479f7205fe5445090f71ed4e9bd77fdb3b740253ce82703/fastparquet-2024.11.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a5ad5fc14b0567e700bea3cd528a0bd45a6f9371370b49de8889fb3d10a6574a", size = 684095 },
-    { url = "https://files.pythonhosted.org/packages/20/37/c3164261d6183d529a59afef2749821b262c8581d837faa91043837c6f76/fastparquet-2024.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0b74333914f454344458dab9d1432fda9b70d62e28dc7acb1512d937ef1424ee", size = 1700355 },
-    { url = "https://files.pythonhosted.org/packages/e6/95/cf4b175c22160ec21e4664830763bfaa80b2cf05133ef854c3f436d01c16/fastparquet-2024.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41d1610130b5cb1ce36467766191c5418cba8631e2bfe3affffaf13f9be4e7a8", size = 1714663 },
-    { url = "https://files.pythonhosted.org/packages/2c/31/b6c8cdb6d5df964a192e4e8c8ecd979718afb9ca7e2dc9243a4368b370e9/fastparquet-2024.11.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d281edd625c33628ba028d3221180283d6161bc5ceb55eae1f0ca1678f864f26", size = 1666729 },
-    { url = "https://files.pythonhosted.org/packages/31/e5/8a0575c46a7973849f8f2a88af16618b9c7efe98f249f03e3e3de69c2b86/fastparquet-2024.11.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fa56b19a29008c34cfe8831e810f770080debcbffc69aabd1df4d47572181f9c", size = 1741669 },
-    { url = "https://files.pythonhosted.org/packages/bb/6a/669f8c9cf2fc6e30c9353832f870e5a2e170b458d12c5080837f742d963d/fastparquet-2024.11.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5914ecfa766b7763201b9f49d832a5e89c2dccad470ca4f9c9b228d9a8349756", size = 1782359 },
-    { url = "https://files.pythonhosted.org/packages/70/c0/1374cb43924739f4542e39d972481c1f4c7dd96808a1947450808e4e7df7/fastparquet-2024.11.0-cp310-cp310-win_amd64.whl", hash = "sha256:561202e8f0e859ccc1aa77c4aaad1d7901b2d50fd6f624ca018bae4c3c7a62ce", size = 670700 },
-    { url = "https://files.pythonhosted.org/packages/7c/51/e0d6e702523ac923ede6c05e240f4a02533ccf2cea9fec7a43491078e920/fastparquet-2024.11.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:374cdfa745aa7d5188430528d5841cf823eb9ad16df72ad6dadd898ccccce3be", size = 909934 },
-    { url = "https://files.pythonhosted.org/packages/0a/c8/5c0fb644c19a8d80b2ae4d8aa7d90c2d85d0bd4a948c5c700bea5c2802ea/fastparquet-2024.11.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4c8401bfd86cccaf0ab7c0ade58c91ae19317ff6092e1d4ad96c2178197d8124", size = 683844 },
-    { url = "https://files.pythonhosted.org/packages/33/4a/1e532fd1a0d4d8af7ffc7e3a8106c0bcd13ed914a93a61e299b3832dd3d2/fastparquet-2024.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f9cca4c6b5969df5561c13786f9d116300db1ec22c7941e237cfca4ce602f59b", size = 1791698 },
-    { url = "https://files.pythonhosted.org/packages/8d/e8/e1ede861bea68394a755d8be1aa2e2d60a3b9f6b551bfd56aeca74987e2e/fastparquet-2024.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9a9387e77ac608d8978774caaf1e19de67eaa1386806e514dcb19f741b19cfe5", size = 1804289 },
-    { url = "https://files.pythonhosted.org/packages/4f/1e/957090cccaede805583ca3f3e46e2762d0f9bf8860ecbce65197e47d84c1/fastparquet-2024.11.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6595d3771b3d587a31137e985f751b4d599d5c8e9af9c4858e373fdf5c3f8720", size = 1753638 },
-    { url = "https://files.pythonhosted.org/packages/85/72/344787c685fd1531f07ae712a855a7c34d13deaa26c3fd4a9231bea7dbab/fastparquet-2024.11.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:053695c2f730b78a2d3925df7cd5c6444d6c1560076af907993361cc7accf3e2", size = 1814407 },
-    { url = "https://files.pythonhosted.org/packages/6c/ec/ab9d5685f776a1965797eb68c4364c72edf57cd35beed2df49b34425d1df/fastparquet-2024.11.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0a52eecc6270ae15f0d51347c3f762703dd667ca486f127dc0a21e7e59856ae5", size = 1874462 },
-    { url = "https://files.pythonhosted.org/packages/90/4f/7a4ea9a7ddf0a3409873f0787f355806f9e0b73f42f2acecacdd9a8eff0a/fastparquet-2024.11.0-cp311-cp311-win_amd64.whl", hash = "sha256:e29ff7a367fafa57c6896fb6abc84126e2466811aefd3e4ad4070b9e18820e54", size = 671023 },
-    { url = "https://files.pythonhosted.org/packages/08/76/068ac7ec9b4fc783be21a75a6a90b8c0654da4d46934d969e524ce287787/fastparquet-2024.11.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:dbad4b014782bd38b58b8e9f514fe958cfa7a6c4e187859232d29fd5c5ddd849", size = 915968 },
-    { url = "https://files.pythonhosted.org/packages/c7/9e/6d3b4188ad64ed51173263c07109a5f18f9c84a44fa39ab524fca7420cda/fastparquet-2024.11.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:403d31109d398b6be7ce84fa3483fc277c6a23f0b321348c0a505eb098a041cb", size = 685399 },
-    { url = "https://files.pythonhosted.org/packages/8f/6c/809220bc9fbe83d107df2d664c3fb62fb81867be8f5218ac66c2e6b6a358/fastparquet-2024.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cbbb9057a26acf0abad7adf58781ee357258b7708ee44a289e3bee97e2f55d42", size = 1758557 },
-    { url = "https://files.pythonhosted.org/packages/e0/2c/b3b3e6ca2e531484289024138cd4709c22512b3fe68066d7f9849da4a76c/fastparquet-2024.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:63e0e416e25c15daa174aad8ba991c2e9e5b0dc347e5aed5562124261400f87b", size = 1781052 },
-    { url = "https://files.pythonhosted.org/packages/21/fe/97ed45092d0311c013996dae633122b7a51c5d9fe8dcbc2c840dc491201e/fastparquet-2024.11.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0e2d7f02f57231e6c86d26e9ea71953737202f20e948790e5d4db6d6a1a150dc", size = 1715797 },
-    { url = "https://files.pythonhosted.org/packages/24/df/02fa6aee6c0d53d1563b5bc22097076c609c4c5baa47056b0b4bed456fcf/fastparquet-2024.11.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:fbe4468146b633d8f09d7b196fea0547f213cb5ce5f76e9d1beb29eaa9593a93", size = 1795682 },
-    { url = "https://files.pythonhosted.org/packages/b0/25/f4f87557589e1923ee0e3bebbc84f08b7c56962bf90f51b116ddc54f2c9f/fastparquet-2024.11.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:29d5c718817bcd765fc519b17f759cad4945974421ecc1931d3bdc3e05e57fa9", size = 1857842 },
-    { url = "https://files.pythonhosted.org/packages/b1/f9/98cd0c39115879be1044d59c9b76e8292776e99bb93565bf990078fd11c4/fastparquet-2024.11.0-cp312-cp312-win_amd64.whl", hash = "sha256:74a0b3c40ab373442c0fda96b75a36e88745d8b138fcc3a6143e04682cbbb8ca", size = 673269 },
-    { url = "https://files.pythonhosted.org/packages/47/e3/e7db38704be5db787270d43dde895eaa1a825ab25dc245e71df70860ec12/fastparquet-2024.11.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:59e5c5b51083d5b82572cdb7aed0346e3181e3ac9d2e45759da2e804bdafa7ee", size = 912523 },
-    { url = "https://files.pythonhosted.org/packages/d3/66/e3387c99293dae441634e7724acaa425b27de19a00ee3d546775dace54a9/fastparquet-2024.11.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:bdadf7b6bad789125b823bfc5b0a719ba5c4a2ef965f973702d3ea89cff057f6", size = 683779 },
-    { url = "https://files.pythonhosted.org/packages/0a/21/d112d0573d086b578bf04302a502e9a7605ea8f1244a7b8577cd945eec78/fastparquet-2024.11.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46b2db02fc2a1507939d35441c8ab211d53afd75d82eec9767d1c3656402859b", size = 1751113 },
-    { url = "https://files.pythonhosted.org/packages/6b/a7/040507cee3a7798954e8fdbca21d2dbc532774b02b882d902b8a4a6849ef/fastparquet-2024.11.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a3afdef2895c9f459135a00a7ed3ceafebfbce918a9e7b5d550e4fae39c1b64d", size = 1780496 },
-    { url = "https://files.pythonhosted.org/packages/bc/75/d0d9f7533d780ec167eede16ad88073ee71696150511126c31940e7f73aa/fastparquet-2024.11.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:36b5c9bd2ffaaa26ff45d59a6cefe58503dd748e0c7fad80dd905749da0f2b9e", size = 1713608 },
-    { url = "https://files.pythonhosted.org/packages/30/fa/1d95bc86e45e80669c4f374b2ca26a9e5895a1011bb05d6341b4a7414693/fastparquet-2024.11.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:6b7df5d3b61a19d76e209fe8d3133759af1c139e04ebc6d43f3cc2d8045ef338", size = 1792779 },
-    { url = "https://files.pythonhosted.org/packages/13/3d/c076beeb926c79593374c04662a9422a76650eef17cd1c8e10951340764a/fastparquet-2024.11.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:8b35823ac7a194134e5f82fa4a9659e42e8f9ad1f2d22a55fbb7b9e4053aabbb", size = 1851322 },
-    { url = "https://files.pythonhosted.org/packages/09/5a/1d0d47e64816002824d4a876644e8c65540fa23f91b701f0daa726931545/fastparquet-2024.11.0-cp313-cp313-win_amd64.whl", hash = "sha256:d20632964e65530374ff7cddd42cc06aa0a1388934903693d6d22592a5ba827b", size = 673266 },
-    { url = "https://files.pythonhosted.org/packages/a2/3a/607301b22fd50065503a8be7ea0068ae61d9b7f63be44a4bad68a286f712/fastparquet-2024.11.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:1ae953c0e3832ae3936b6d92fde493ac7d8b775d7d59d02f7f46f67e1c21ed24", size = 911743 },
-    { url = "https://files.pythonhosted.org/packages/e7/a7/a9ff37efc9cf5164e372a813630488c0b3600e35c6bc1a7add9b9d17296f/fastparquet-2024.11.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:869e167a4067116b4a27eb7adbe597130b2e2e9cfc0f3e84f60e2e182a933f23", size = 684662 },
-    { url = "https://files.pythonhosted.org/packages/2c/76/dd9456425c312e8629e969c2099c4bc538fbf433542752395acb22c65559/fastparquet-2024.11.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:eb3356862fba2f9b2ea8e679d66901f466c92be8e023439fe854bc392fbf40a6", size = 1701403 },
-    { url = "https://files.pythonhosted.org/packages/b3/9d/17a296e458b62e7b9af301c2e7e77a21af37bcf29030a20f56d5d6757ae3/fastparquet-2024.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dc475993232c6a64f350aeb928013a807eb93f78675810fd019cbcff39f6baf3", size = 1715213 },
-    { url = "https://files.pythonhosted.org/packages/d2/05/2b5722eb6e24249706886423fbab2c42a5d60b0fc0f1a27e955ff2b829e8/fastparquet-2024.11.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d24c923a2d9d22a5e7564245f856e6462d524d57982ac8f7479cde991ff73362", size = 1668310 },
-    { url = "https://files.pythonhosted.org/packages/4c/9a/8a53dfa0c564f0683b693aa84878074d82ce2516e0f599d2a19ff3a0b073/fastparquet-2024.11.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6b936dcf40ca5fff9e70383d48811b1482b871ff74af857cb4db5f4d072f01ab", size = 1743186 },
-    { url = "https://files.pythonhosted.org/packages/cd/f6/61986beb1a894b40e7b9e9f5d0228ca068e005e66576af0b4fa94481ffbc/fastparquet-2024.11.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:4abd3426607335e5ad09be29ef4eeccdf097710e44420deac16893cee64ea0d8", size = 1784121 },
-    { url = "https://files.pythonhosted.org/packages/91/71/2cf345d0ccf1239108190bcd838a19b289fa1bac5676e8ae701ae24b602f/fastparquet-2024.11.0-cp39-cp39-win_amd64.whl", hash = "sha256:6ec7b398a86432993441d0a08dfae59e29649c803ed64ec4b1d7c3e0855b14cb", size = 671248 },
+sdist = { url = "https://files.pythonhosted.org/packages/29/fb/b259e2e33e5740c62fc05677c395e96809e83c0467677d3e714b125ec24c/flax-0.8.5.tar.gz", hash = "sha256:4a9cb7950ece54b0addaa73d77eba24e46138dbe783d01987be79d20ccb2b09b", size = 2862557, upload-time = "2024-06-26T09:28:06.803Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1c/a9/6978d2547b1d8ca0ce75b534c0ba5c60e8e7b918c5c1800225aa0169cb7f/flax-0.8.5-py3-none-any.whl", hash = "sha256:c96e46d1c48a300d010ebf5c4846f163bdd7acc6efff5ff2bfb1cb5b08aa65d8", size = 731342, upload-time = "2024-06-26T09:28:04.239Z" },
 ]
 
 [[package]]
-name = "filelock"
-version = "3.16.1"
+name = "flax"
+version = "0.10.7"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "msgpack", marker = "python_full_version == '3.10.*'" },
+    { name = "optax", version = "0.2.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "orbax-checkpoint", version = "0.11.23", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "pyyaml", marker = "python_full_version == '3.10.*'" },
+    { name = "rich", marker = "python_full_version == '3.10.*'" },
+    { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "treescope", marker = "python_full_version == '3.10.*'" },
+    { name = "typing-extensions", marker = "python_full_version == '3.10.*'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/e6/76/4ea55a60a47e98fcff591238ee26ed4624cb4fdc4893aa3ebf78d0d021f4/flax-0.10.7.tar.gz", hash = "sha256:2930d6671e23076f6db3b96afacf45c5060898f5c189ecab6dda7e05d26c2085", size = 5136099, upload-time = "2025-07-02T06:10:07.819Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/11/f6/560d338687d40182c8429cf35c64cc022e0d57ba3e52191c4a78ed239b4e/flax-0.10.7-py3-none-any.whl", hash = "sha256:4033223a9a9969ba0b252e085e9714d0a1e9124ac300aaf48e92c40769c420f6", size = 456944, upload-time = "2025-07-02T06:10:05.807Z" },
+]
+
+[[package]]
+name = "flax"
+version = "0.11.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/9d/db/3ef5bb276dae18d6ec2124224403d1d67bccdbefc17af4cc8f553e341ab1/filelock-3.16.1.tar.gz", hash = "sha256:c249fbfcd5db47e5e2d6d62198e565475ee65e4831e2561c8e313fa7eb961435", size = 18037 }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+]
+dependencies = [
+    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "msgpack", marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "optax", version = "0.2.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "orbax-checkpoint", version = "0.11.23", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "pyyaml", marker = "python_full_version >= '3.11'" },
+    { name = "rich", marker = "python_full_version >= '3.11'" },
+    { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "treescope", marker = "python_full_version >= '3.11'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ac/b9/3d2a5f94eb4aa9ff2e5981ac73ea91cfedef5f9de35d903e56cdcf539f31/flax-0.11.2.tar.gz", hash = "sha256:55452529b70c704128075a17f7a54d94f1f90b0f2d9498bdc9578cae3646e460", size = 5201170, upload-time = "2025-08-28T17:56:34.658Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b9/f8/feced7779d755758a52d1f6635d990b8d98dc0a29fa568bbe0625f18fdf3/filelock-3.16.1-py3-none-any.whl", hash = "sha256:2082e5703d51fbf98ea75855d9d5527e33d8ff23099bec374a134febee6946b0", size = 16163 },
+    { url = "https://files.pythonhosted.org/packages/bb/b2/5f520253823c84ca87ca0fe5a7b9d96417b1e549b98a38f5069555590d07/flax-0.11.2-py3-none-any.whl", hash = "sha256:61a5f60c27ab4e6420fda7f593f93b5ec1adba629fd33b2f0e52740ab7c04da9", size = 458093, upload-time = "2025-08-28T17:56:32.588Z" },
 ]
 
 [[package]]
 name = "frozenlist"
-version = "1.5.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/8f/ed/0f4cec13a93c02c47ec32d81d11c0c1efbadf4a471e3f3ce7cad366cbbd3/frozenlist-1.5.0.tar.gz", hash = "sha256:81d5af29e61b9c8348e876d442253723928dce6433e0e76cd925cd83f1b4b817", size = 39930 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/54/79/29d44c4af36b2b240725dce566b20f63f9b36ef267aaaa64ee7466f4f2f8/frozenlist-1.5.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:5b6a66c18b5b9dd261ca98dffcb826a525334b2f29e7caa54e182255c5f6a65a", size = 94451 },
-    { url = "https://files.pythonhosted.org/packages/47/47/0c999aeace6ead8a44441b4f4173e2261b18219e4ad1fe9a479871ca02fc/frozenlist-1.5.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:d1b3eb7b05ea246510b43a7e53ed1653e55c2121019a97e60cad7efb881a97bb", size = 54301 },
-    { url = "https://files.pythonhosted.org/packages/8d/60/107a38c1e54176d12e06e9d4b5d755b677d71d1219217cee063911b1384f/frozenlist-1.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:15538c0cbf0e4fa11d1e3a71f823524b0c46299aed6e10ebb4c2089abd8c3bec", size = 52213 },
-    { url = "https://files.pythonhosted.org/packages/17/62/594a6829ac5679c25755362a9dc93486a8a45241394564309641425d3ff6/frozenlist-1.5.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e79225373c317ff1e35f210dd5f1344ff31066ba8067c307ab60254cd3a78ad5", size = 240946 },
-    { url = "https://files.pythonhosted.org/packages/7e/75/6c8419d8f92c80dd0ee3f63bdde2702ce6398b0ac8410ff459f9b6f2f9cb/frozenlist-1.5.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9272fa73ca71266702c4c3e2d4a28553ea03418e591e377a03b8e3659d94fa76", size = 264608 },
-    { url = "https://files.pythonhosted.org/packages/88/3e/82a6f0b84bc6fb7e0be240e52863c6d4ab6098cd62e4f5b972cd31e002e8/frozenlist-1.5.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:498524025a5b8ba81695761d78c8dd7382ac0b052f34e66939c42df860b8ff17", size = 261361 },
-    { url = "https://files.pythonhosted.org/packages/fd/85/14e5f9ccac1b64ff2f10c927b3ffdf88772aea875882406f9ba0cec8ad84/frozenlist-1.5.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:92b5278ed9d50fe610185ecd23c55d8b307d75ca18e94c0e7de328089ac5dcba", size = 231649 },
-    { url = "https://files.pythonhosted.org/packages/ee/59/928322800306f6529d1852323014ee9008551e9bb027cc38d276cbc0b0e7/frozenlist-1.5.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7f3c8c1dacd037df16e85227bac13cca58c30da836c6f936ba1df0c05d046d8d", size = 241853 },
-    { url = "https://files.pythonhosted.org/packages/7d/bd/e01fa4f146a6f6c18c5d34cab8abdc4013774a26c4ff851128cd1bd3008e/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:f2ac49a9bedb996086057b75bf93538240538c6d9b38e57c82d51f75a73409d2", size = 243652 },
-    { url = "https://files.pythonhosted.org/packages/a5/bd/e4771fd18a8ec6757033f0fa903e447aecc3fbba54e3630397b61596acf0/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e66cc454f97053b79c2ab09c17fbe3c825ea6b4de20baf1be28919460dd7877f", size = 241734 },
-    { url = "https://files.pythonhosted.org/packages/21/13/c83821fa5544af4f60c5d3a65d054af3213c26b14d3f5f48e43e5fb48556/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:5a3ba5f9a0dfed20337d3e966dc359784c9f96503674c2faf015f7fe8e96798c", size = 260959 },
-    { url = "https://files.pythonhosted.org/packages/71/f3/1f91c9a9bf7ed0e8edcf52698d23f3c211d8d00291a53c9f115ceb977ab1/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:6321899477db90bdeb9299ac3627a6a53c7399c8cd58d25da094007402b039ab", size = 262706 },
-    { url = "https://files.pythonhosted.org/packages/4c/22/4a256fdf5d9bcb3ae32622c796ee5ff9451b3a13a68cfe3f68e2c95588ce/frozenlist-1.5.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:76e4753701248476e6286f2ef492af900ea67d9706a0155335a40ea21bf3b2f5", size = 250401 },
-    { url = "https://files.pythonhosted.org/packages/af/89/c48ebe1f7991bd2be6d5f4ed202d94960c01b3017a03d6954dd5fa9ea1e8/frozenlist-1.5.0-cp310-cp310-win32.whl", hash = "sha256:977701c081c0241d0955c9586ffdd9ce44f7a7795df39b9151cd9a6fd0ce4cfb", size = 45498 },
-    { url = "https://files.pythonhosted.org/packages/28/2f/cc27d5f43e023d21fe5c19538e08894db3d7e081cbf582ad5ed366c24446/frozenlist-1.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:189f03b53e64144f90990d29a27ec4f7997d91ed3d01b51fa39d2dbe77540fd4", size = 51622 },
-    { url = "https://files.pythonhosted.org/packages/79/43/0bed28bf5eb1c9e4301003b74453b8e7aa85fb293b31dde352aac528dafc/frozenlist-1.5.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:fd74520371c3c4175142d02a976aee0b4cb4a7cc912a60586ffd8d5929979b30", size = 94987 },
-    { url = "https://files.pythonhosted.org/packages/bb/bf/b74e38f09a246e8abbe1e90eb65787ed745ccab6eaa58b9c9308e052323d/frozenlist-1.5.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:2f3f7a0fbc219fb4455264cae4d9f01ad41ae6ee8524500f381de64ffaa077d5", size = 54584 },
-    { url = "https://files.pythonhosted.org/packages/2c/31/ab01375682f14f7613a1ade30149f684c84f9b8823a4391ed950c8285656/frozenlist-1.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:f47c9c9028f55a04ac254346e92977bf0f166c483c74b4232bee19a6697e4778", size = 52499 },
-    { url = "https://files.pythonhosted.org/packages/98/a8/d0ac0b9276e1404f58fec3ab6e90a4f76b778a49373ccaf6a563f100dfbc/frozenlist-1.5.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0996c66760924da6e88922756d99b47512a71cfd45215f3570bf1e0b694c206a", size = 276357 },
-    { url = "https://files.pythonhosted.org/packages/ad/c9/c7761084fa822f07dac38ac29f841d4587570dd211e2262544aa0b791d21/frozenlist-1.5.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a2fe128eb4edeabe11896cb6af88fca5346059f6c8d807e3b910069f39157869", size = 287516 },
-    { url = "https://files.pythonhosted.org/packages/a1/ff/cd7479e703c39df7bdab431798cef89dc75010d8aa0ca2514c5b9321db27/frozenlist-1.5.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1a8ea951bbb6cacd492e3948b8da8c502a3f814f5d20935aae74b5df2b19cf3d", size = 283131 },
-    { url = "https://files.pythonhosted.org/packages/59/a0/370941beb47d237eca4fbf27e4e91389fd68699e6f4b0ebcc95da463835b/frozenlist-1.5.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:de537c11e4aa01d37db0d403b57bd6f0546e71a82347a97c6a9f0dcc532b3a45", size = 261320 },
-    { url = "https://files.pythonhosted.org/packages/b8/5f/c10123e8d64867bc9b4f2f510a32042a306ff5fcd7e2e09e5ae5100ee333/frozenlist-1.5.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9c2623347b933fcb9095841f1cc5d4ff0b278addd743e0e966cb3d460278840d", size = 274877 },
-    { url = "https://files.pythonhosted.org/packages/fa/79/38c505601ae29d4348f21706c5d89755ceded02a745016ba2f58bd5f1ea6/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:cee6798eaf8b1416ef6909b06f7dc04b60755206bddc599f52232606e18179d3", size = 269592 },
-    { url = "https://files.pythonhosted.org/packages/19/e2/39f3a53191b8204ba9f0bb574b926b73dd2efba2a2b9d2d730517e8f7622/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:f5f9da7f5dbc00a604fe74aa02ae7c98bcede8a3b8b9666f9f86fc13993bc71a", size = 265934 },
-    { url = "https://files.pythonhosted.org/packages/d5/c9/3075eb7f7f3a91f1a6b00284af4de0a65a9ae47084930916f5528144c9dd/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:90646abbc7a5d5c7c19461d2e3eeb76eb0b204919e6ece342feb6032c9325ae9", size = 283859 },
-    { url = "https://files.pythonhosted.org/packages/05/f5/549f44d314c29408b962fa2b0e69a1a67c59379fb143b92a0a065ffd1f0f/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:bdac3c7d9b705d253b2ce370fde941836a5f8b3c5c2b8fd70940a3ea3af7f4f2", size = 287560 },
-    { url = "https://files.pythonhosted.org/packages/9d/f8/cb09b3c24a3eac02c4c07a9558e11e9e244fb02bf62c85ac2106d1eb0c0b/frozenlist-1.5.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:03d33c2ddbc1816237a67f66336616416e2bbb6beb306e5f890f2eb22b959cdf", size = 277150 },
-    { url = "https://files.pythonhosted.org/packages/37/48/38c2db3f54d1501e692d6fe058f45b6ad1b358d82cd19436efab80cfc965/frozenlist-1.5.0-cp311-cp311-win32.whl", hash = "sha256:237f6b23ee0f44066219dae14c70ae38a63f0440ce6750f868ee08775073f942", size = 45244 },
-    { url = "https://files.pythonhosted.org/packages/ca/8c/2ddffeb8b60a4bce3b196c32fcc30d8830d4615e7b492ec2071da801b8ad/frozenlist-1.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:0cc974cc93d32c42e7b0f6cf242a6bd941c57c61b618e78b6c0a96cb72788c1d", size = 51634 },
-    { url = "https://files.pythonhosted.org/packages/79/73/fa6d1a96ab7fd6e6d1c3500700963eab46813847f01ef0ccbaa726181dd5/frozenlist-1.5.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:31115ba75889723431aa9a4e77d5f398f5cf976eea3bdf61749731f62d4a4a21", size = 94026 },
-    { url = "https://files.pythonhosted.org/packages/ab/04/ea8bf62c8868b8eada363f20ff1b647cf2e93377a7b284d36062d21d81d1/frozenlist-1.5.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:7437601c4d89d070eac8323f121fcf25f88674627505334654fd027b091db09d", size = 54150 },
-    { url = "https://files.pythonhosted.org/packages/d0/9a/8e479b482a6f2070b26bda572c5e6889bb3ba48977e81beea35b5ae13ece/frozenlist-1.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:7948140d9f8ece1745be806f2bfdf390127cf1a763b925c4a805c603df5e697e", size = 51927 },
-    { url = "https://files.pythonhosted.org/packages/e3/12/2aad87deb08a4e7ccfb33600871bbe8f0e08cb6d8224371387f3303654d7/frozenlist-1.5.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:feeb64bc9bcc6b45c6311c9e9b99406660a9c05ca8a5b30d14a78555088b0b3a", size = 282647 },
-    { url = "https://files.pythonhosted.org/packages/77/f2/07f06b05d8a427ea0060a9cef6e63405ea9e0d761846b95ef3fb3be57111/frozenlist-1.5.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:683173d371daad49cffb8309779e886e59c2f369430ad28fe715f66d08d4ab1a", size = 289052 },
-    { url = "https://files.pythonhosted.org/packages/bd/9f/8bf45a2f1cd4aa401acd271b077989c9267ae8463e7c8b1eb0d3f561b65e/frozenlist-1.5.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7d57d8f702221405a9d9b40f9da8ac2e4a1a8b5285aac6100f3393675f0a85ee", size = 291719 },
-    { url = "https://files.pythonhosted.org/packages/41/d1/1f20fd05a6c42d3868709b7604c9f15538a29e4f734c694c6bcfc3d3b935/frozenlist-1.5.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:30c72000fbcc35b129cb09956836c7d7abf78ab5416595e4857d1cae8d6251a6", size = 267433 },
-    { url = "https://files.pythonhosted.org/packages/af/f2/64b73a9bb86f5a89fb55450e97cd5c1f84a862d4ff90d9fd1a73ab0f64a5/frozenlist-1.5.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:000a77d6034fbad9b6bb880f7ec073027908f1b40254b5d6f26210d2dab1240e", size = 283591 },
-    { url = "https://files.pythonhosted.org/packages/29/e2/ffbb1fae55a791fd6c2938dd9ea779509c977435ba3940b9f2e8dc9d5316/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5d7f5a50342475962eb18b740f3beecc685a15b52c91f7d975257e13e029eca9", size = 273249 },
-    { url = "https://files.pythonhosted.org/packages/2e/6e/008136a30798bb63618a114b9321b5971172a5abddff44a100c7edc5ad4f/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:87f724d055eb4785d9be84e9ebf0f24e392ddfad00b3fe036e43f489fafc9039", size = 271075 },
-    { url = "https://files.pythonhosted.org/packages/ae/f0/4e71e54a026b06724cec9b6c54f0b13a4e9e298cc8db0f82ec70e151f5ce/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:6e9080bb2fb195a046e5177f10d9d82b8a204c0736a97a153c2466127de87784", size = 285398 },
-    { url = "https://files.pythonhosted.org/packages/4d/36/70ec246851478b1c0b59f11ef8ade9c482ff447c1363c2bd5fad45098b12/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:9b93d7aaa36c966fa42efcaf716e6b3900438632a626fb09c049f6a2f09fc631", size = 294445 },
-    { url = "https://files.pythonhosted.org/packages/37/e0/47f87544055b3349b633a03c4d94b405956cf2437f4ab46d0928b74b7526/frozenlist-1.5.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:52ef692a4bc60a6dd57f507429636c2af8b6046db8b31b18dac02cbc8f507f7f", size = 280569 },
-    { url = "https://files.pythonhosted.org/packages/f9/7c/490133c160fb6b84ed374c266f42800e33b50c3bbab1652764e6e1fc498a/frozenlist-1.5.0-cp312-cp312-win32.whl", hash = "sha256:29d94c256679247b33a3dc96cce0f93cbc69c23bf75ff715919332fdbb6a32b8", size = 44721 },
-    { url = "https://files.pythonhosted.org/packages/b1/56/4e45136ffc6bdbfa68c29ca56ef53783ef4c2fd395f7cbf99a2624aa9aaa/frozenlist-1.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:8969190d709e7c48ea386db202d708eb94bdb29207a1f269bab1196ce0dcca1f", size = 51329 },
-    { url = "https://files.pythonhosted.org/packages/da/3b/915f0bca8a7ea04483622e84a9bd90033bab54bdf485479556c74fd5eaf5/frozenlist-1.5.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:7a1a048f9215c90973402e26c01d1cff8a209e1f1b53f72b95c13db61b00f953", size = 91538 },
-    { url = "https://files.pythonhosted.org/packages/c7/d1/a7c98aad7e44afe5306a2b068434a5830f1470675f0e715abb86eb15f15b/frozenlist-1.5.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:dd47a5181ce5fcb463b5d9e17ecfdb02b678cca31280639255ce9d0e5aa67af0", size = 52849 },
-    { url = "https://files.pythonhosted.org/packages/3a/c8/76f23bf9ab15d5f760eb48701909645f686f9c64fbb8982674c241fbef14/frozenlist-1.5.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:1431d60b36d15cda188ea222033eec8e0eab488f39a272461f2e6d9e1a8e63c2", size = 50583 },
-    { url = "https://files.pythonhosted.org/packages/1f/22/462a3dd093d11df623179d7754a3b3269de3b42de2808cddef50ee0f4f48/frozenlist-1.5.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6482a5851f5d72767fbd0e507e80737f9c8646ae7fd303def99bfe813f76cf7f", size = 265636 },
-    { url = "https://files.pythonhosted.org/packages/80/cf/e075e407fc2ae7328155a1cd7e22f932773c8073c1fc78016607d19cc3e5/frozenlist-1.5.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:44c49271a937625619e862baacbd037a7ef86dd1ee215afc298a417ff3270608", size = 270214 },
-    { url = "https://files.pythonhosted.org/packages/a1/58/0642d061d5de779f39c50cbb00df49682832923f3d2ebfb0fedf02d05f7f/frozenlist-1.5.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:12f78f98c2f1c2429d42e6a485f433722b0061d5c0b0139efa64f396efb5886b", size = 273905 },
-    { url = "https://files.pythonhosted.org/packages/ab/66/3fe0f5f8f2add5b4ab7aa4e199f767fd3b55da26e3ca4ce2cc36698e50c4/frozenlist-1.5.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ce3aa154c452d2467487765e3adc730a8c153af77ad84096bc19ce19a2400840", size = 250542 },
-    { url = "https://files.pythonhosted.org/packages/f6/b8/260791bde9198c87a465224e0e2bb62c4e716f5d198fc3a1dacc4895dbd1/frozenlist-1.5.0-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9b7dc0c4338e6b8b091e8faf0db3168a37101943e687f373dce00959583f7439", size = 267026 },
-    { url = "https://files.pythonhosted.org/packages/2e/a4/3d24f88c527f08f8d44ade24eaee83b2627793fa62fa07cbb7ff7a2f7d42/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:45e0896250900b5aa25180f9aec243e84e92ac84bd4a74d9ad4138ef3f5c97de", size = 257690 },
-    { url = "https://files.pythonhosted.org/packages/de/9a/d311d660420b2beeff3459b6626f2ab4fb236d07afbdac034a4371fe696e/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:561eb1c9579d495fddb6da8959fd2a1fca2c6d060d4113f5844b433fc02f2641", size = 253893 },
-    { url = "https://files.pythonhosted.org/packages/c6/23/e491aadc25b56eabd0f18c53bb19f3cdc6de30b2129ee0bc39cd387cd560/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:df6e2f325bfee1f49f81aaac97d2aa757c7646534a06f8f577ce184afe2f0a9e", size = 267006 },
-    { url = "https://files.pythonhosted.org/packages/08/c4/ab918ce636a35fb974d13d666dcbe03969592aeca6c3ab3835acff01f79c/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:140228863501b44b809fb39ec56b5d4071f4d0aa6d216c19cbb08b8c5a7eadb9", size = 276157 },
-    { url = "https://files.pythonhosted.org/packages/c0/29/3b7a0bbbbe5a34833ba26f686aabfe982924adbdcafdc294a7a129c31688/frozenlist-1.5.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:7707a25d6a77f5d27ea7dc7d1fc608aa0a478193823f88511ef5e6b8a48f9d03", size = 264642 },
-    { url = "https://files.pythonhosted.org/packages/ab/42/0595b3dbffc2e82d7fe658c12d5a5bafcd7516c6bf2d1d1feb5387caa9c1/frozenlist-1.5.0-cp313-cp313-win32.whl", hash = "sha256:31a9ac2b38ab9b5a8933b693db4939764ad3f299fcaa931a3e605bc3460e693c", size = 44914 },
-    { url = "https://files.pythonhosted.org/packages/17/c4/b7db1206a3fea44bf3b838ca61deb6f74424a8a5db1dd53ecb21da669be6/frozenlist-1.5.0-cp313-cp313-win_amd64.whl", hash = "sha256:11aabdd62b8b9c4b84081a3c246506d1cddd2dd93ff0ad53ede5defec7886b28", size = 51167 },
-    { url = "https://files.pythonhosted.org/packages/da/4d/d94ff0fb0f5313902c132817c62d19cdc5bdcd0c195d392006ef4b779fc6/frozenlist-1.5.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:9bbcdfaf4af7ce002694a4e10a0159d5a8d20056a12b05b45cea944a4953f972", size = 95319 },
-    { url = "https://files.pythonhosted.org/packages/8c/1b/d90e554ca2b483d31cb2296e393f72c25bdc38d64526579e95576bfda587/frozenlist-1.5.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1893f948bf6681733aaccf36c5232c231e3b5166d607c5fa77773611df6dc336", size = 54749 },
-    { url = "https://files.pythonhosted.org/packages/f8/66/7fdecc9ef49f8db2aa4d9da916e4ecf357d867d87aea292efc11e1b2e932/frozenlist-1.5.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2b5e23253bb709ef57a8e95e6ae48daa9ac5f265637529e4ce6b003a37b2621f", size = 52718 },
-    { url = "https://files.pythonhosted.org/packages/08/04/e2fddc92135276e07addbc1cf413acffa0c2d848b3e54cacf684e146df49/frozenlist-1.5.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0f253985bb515ecd89629db13cb58d702035ecd8cfbca7d7a7e29a0e6d39af5f", size = 241756 },
-    { url = "https://files.pythonhosted.org/packages/c6/52/be5ff200815d8a341aee5b16b6b707355e0ca3652953852238eb92b120c2/frozenlist-1.5.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:04a5c6babd5e8fb7d3c871dc8b321166b80e41b637c31a995ed844a6139942b6", size = 267718 },
-    { url = "https://files.pythonhosted.org/packages/88/be/4bd93a58be57a3722fc544c36debdf9dcc6758f761092e894d78f18b8f20/frozenlist-1.5.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a9fe0f1c29ba24ba6ff6abf688cb0b7cf1efab6b6aa6adc55441773c252f7411", size = 263494 },
-    { url = "https://files.pythonhosted.org/packages/32/ba/58348b90193caa096ce9e9befea6ae67f38dabfd3aacb47e46137a6250a8/frozenlist-1.5.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:226d72559fa19babe2ccd920273e767c96a49b9d3d38badd7c91a0fdeda8ea08", size = 232838 },
-    { url = "https://files.pythonhosted.org/packages/f6/33/9f152105227630246135188901373c4f322cc026565ca6215b063f4c82f4/frozenlist-1.5.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:15b731db116ab3aedec558573c1a5eec78822b32292fe4f2f0345b7f697745c2", size = 242912 },
-    { url = "https://files.pythonhosted.org/packages/a0/10/3db38fb3ccbafadd80a1b0d6800c987b0e3fe3ef2d117c6ced0246eea17a/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:366d8f93e3edfe5a918c874702f78faac300209a4d5bf38352b2c1bdc07a766d", size = 244763 },
-    { url = "https://files.pythonhosted.org/packages/e2/cd/1df468fdce2f66a4608dffe44c40cdc35eeaa67ef7fd1d813f99a9a37842/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:1b96af8c582b94d381a1c1f51ffaedeb77c821c690ea5f01da3d70a487dd0a9b", size = 242841 },
-    { url = "https://files.pythonhosted.org/packages/ee/5f/16097a5ca0bb6b6779c02cc9379c72fe98d56115d4c54d059fb233168fb6/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:c03eff4a41bd4e38415cbed054bbaff4a075b093e2394b6915dca34a40d1e38b", size = 263407 },
-    { url = "https://files.pythonhosted.org/packages/0f/f7/58cd220ee1c2248ee65a32f5b4b93689e3fe1764d85537eee9fc392543bc/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:50cf5e7ee9b98f22bdecbabf3800ae78ddcc26e4a435515fc72d97903e8488e0", size = 265083 },
-    { url = "https://files.pythonhosted.org/packages/62/b8/49768980caabf81ac4a2d156008f7cbd0107e6b36d08a313bb31035d9201/frozenlist-1.5.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1e76bfbc72353269c44e0bc2cfe171900fbf7f722ad74c9a7b638052afe6a00c", size = 251564 },
-    { url = "https://files.pythonhosted.org/packages/cb/83/619327da3b86ef957ee7a0cbf3c166a09ed1e87a3f7f1ff487d7d0284683/frozenlist-1.5.0-cp39-cp39-win32.whl", hash = "sha256:666534d15ba8f0fda3f53969117383d5dc021266b3c1a42c9ec4855e4b58b9d3", size = 45691 },
-    { url = "https://files.pythonhosted.org/packages/8b/28/407bc34a745151ed2322c690b6e7d83d7101472e81ed76e1ebdac0b70a78/frozenlist-1.5.0-cp39-cp39-win_amd64.whl", hash = "sha256:5c28f4b5dbef8a0d8aad0d4de24d1e9e981728628afaf4ea0792f5d0939372f0", size = 51767 },
-    { url = "https://files.pythonhosted.org/packages/c6/c8/a5be5b7550c10858fcf9b0ea054baccab474da77d37f1e828ce043a3a5d4/frozenlist-1.5.0-py3-none-any.whl", hash = "sha256:d994863bba198a4a518b467bb971c56e1db3f180a25c6cf7bb1949c267f748c3", size = 11901 },
+version = "1.7.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/79/b1/b64018016eeb087db503b038296fd782586432b9c077fc5c7839e9cb6ef6/frozenlist-1.7.0.tar.gz", hash = "sha256:2e310d81923c2437ea8670467121cc3e9b0f76d3043cc1d2331d56c7fb7a3a8f", size = 45078, upload-time = "2025-06-09T23:02:35.538Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/af/36/0da0a49409f6b47cc2d060dc8c9040b897b5902a8a4e37d9bc1deb11f680/frozenlist-1.7.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:cc4df77d638aa2ed703b878dd093725b72a824c3c546c076e8fdf276f78ee84a", size = 81304, upload-time = "2025-06-09T22:59:46.226Z" },
+    { url = "https://files.pythonhosted.org/packages/77/f0/77c11d13d39513b298e267b22eb6cb559c103d56f155aa9a49097221f0b6/frozenlist-1.7.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:716a9973a2cc963160394f701964fe25012600f3d311f60c790400b00e568b61", size = 47735, upload-time = "2025-06-09T22:59:48.133Z" },
+    { url = "https://files.pythonhosted.org/packages/37/12/9d07fa18971a44150593de56b2f2947c46604819976784bcf6ea0d5db43b/frozenlist-1.7.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a0fd1bad056a3600047fb9462cff4c5322cebc59ebf5d0a3725e0ee78955001d", size = 46775, upload-time = "2025-06-09T22:59:49.564Z" },
+    { url = "https://files.pythonhosted.org/packages/70/34/f73539227e06288fcd1f8a76853e755b2b48bca6747e99e283111c18bcd4/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3789ebc19cb811163e70fe2bd354cea097254ce6e707ae42e56f45e31e96cb8e", size = 224644, upload-time = "2025-06-09T22:59:51.35Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/68/c1d9c2f4a6e438e14613bad0f2973567586610cc22dcb1e1241da71de9d3/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:af369aa35ee34f132fcfad5be45fbfcde0e3a5f6a1ec0712857f286b7d20cca9", size = 222125, upload-time = "2025-06-09T22:59:52.884Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/d0/98e8f9a515228d708344d7c6986752be3e3192d1795f748c24bcf154ad99/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ac64b6478722eeb7a3313d494f8342ef3478dff539d17002f849101b212ef97c", size = 233455, upload-time = "2025-06-09T22:59:54.74Z" },
+    { url = "https://files.pythonhosted.org/packages/79/df/8a11bcec5600557f40338407d3e5bea80376ed1c01a6c0910fcfdc4b8993/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f89f65d85774f1797239693cef07ad4c97fdd0639544bad9ac4b869782eb1981", size = 227339, upload-time = "2025-06-09T22:59:56.187Z" },
+    { url = "https://files.pythonhosted.org/packages/50/82/41cb97d9c9a5ff94438c63cc343eb7980dac4187eb625a51bdfdb7707314/frozenlist-1.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1073557c941395fdfcfac13eb2456cb8aad89f9de27bae29fabca8e563b12615", size = 212969, upload-time = "2025-06-09T22:59:57.604Z" },
+    { url = "https://files.pythonhosted.org/packages/13/47/f9179ee5ee4f55629e4f28c660b3fdf2775c8bfde8f9c53f2de2d93f52a9/frozenlist-1.7.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1ed8d2fa095aae4bdc7fdd80351009a48d286635edffee66bf865e37a9125c50", size = 222862, upload-time = "2025-06-09T22:59:59.498Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/52/df81e41ec6b953902c8b7e3a83bee48b195cb0e5ec2eabae5d8330c78038/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:24c34bea555fe42d9f928ba0a740c553088500377448febecaa82cc3e88aa1fa", size = 222492, upload-time = "2025-06-09T23:00:01.026Z" },
+    { url = "https://files.pythonhosted.org/packages/84/17/30d6ea87fa95a9408245a948604b82c1a4b8b3e153cea596421a2aef2754/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:69cac419ac6a6baad202c85aaf467b65ac860ac2e7f2ac1686dc40dbb52f6577", size = 238250, upload-time = "2025-06-09T23:00:03.401Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/00/ecbeb51669e3c3df76cf2ddd66ae3e48345ec213a55e3887d216eb4fbab3/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:960d67d0611f4c87da7e2ae2eacf7ea81a5be967861e0c63cf205215afbfac59", size = 218720, upload-time = "2025-06-09T23:00:05.282Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/c0/c224ce0e0eb31cc57f67742071bb470ba8246623c1823a7530be0e76164c/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:41be2964bd4b15bf575e5daee5a5ce7ed3115320fb3c2b71fca05582ffa4dc9e", size = 232585, upload-time = "2025-06-09T23:00:07.962Z" },
+    { url = "https://files.pythonhosted.org/packages/55/3c/34cb694abf532f31f365106deebdeac9e45c19304d83cf7d51ebbb4ca4d1/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:46d84d49e00c9429238a7ce02dc0be8f6d7cd0cd405abd1bebdc991bf27c15bd", size = 234248, upload-time = "2025-06-09T23:00:09.428Z" },
+    { url = "https://files.pythonhosted.org/packages/98/c0/2052d8b6cecda2e70bd81299e3512fa332abb6dcd2969b9c80dfcdddbf75/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:15900082e886edb37480335d9d518cec978afc69ccbc30bd18610b7c1b22a718", size = 221621, upload-time = "2025-06-09T23:00:11.32Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/bf/7dcebae315436903b1d98ffb791a09d674c88480c158aa171958a3ac07f0/frozenlist-1.7.0-cp310-cp310-win32.whl", hash = "sha256:400ddd24ab4e55014bba442d917203c73b2846391dd42ca5e38ff52bb18c3c5e", size = 39578, upload-time = "2025-06-09T23:00:13.526Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/5f/f69818f017fa9a3d24d1ae39763e29b7f60a59e46d5f91b9c6b21622f4cd/frozenlist-1.7.0-cp310-cp310-win_amd64.whl", hash = "sha256:6eb93efb8101ef39d32d50bce242c84bcbddb4f7e9febfa7b524532a239b4464", size = 43830, upload-time = "2025-06-09T23:00:14.98Z" },
+    { url = "https://files.pythonhosted.org/packages/34/7e/803dde33760128acd393a27eb002f2020ddb8d99d30a44bfbaab31c5f08a/frozenlist-1.7.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:aa51e147a66b2d74de1e6e2cf5921890de6b0f4820b257465101d7f37b49fb5a", size = 82251, upload-time = "2025-06-09T23:00:16.279Z" },
+    { url = "https://files.pythonhosted.org/packages/75/a9/9c2c5760b6ba45eae11334db454c189d43d34a4c0b489feb2175e5e64277/frozenlist-1.7.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:9b35db7ce1cd71d36ba24f80f0c9e7cff73a28d7a74e91fe83e23d27c7828750", size = 48183, upload-time = "2025-06-09T23:00:17.698Z" },
+    { url = "https://files.pythonhosted.org/packages/47/be/4038e2d869f8a2da165f35a6befb9158c259819be22eeaf9c9a8f6a87771/frozenlist-1.7.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:34a69a85e34ff37791e94542065c8416c1afbf820b68f720452f636d5fb990cd", size = 47107, upload-time = "2025-06-09T23:00:18.952Z" },
+    { url = "https://files.pythonhosted.org/packages/79/26/85314b8a83187c76a37183ceed886381a5f992975786f883472fcb6dc5f2/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4a646531fa8d82c87fe4bb2e596f23173caec9185bfbca5d583b4ccfb95183e2", size = 237333, upload-time = "2025-06-09T23:00:20.275Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/fd/e5b64f7d2c92a41639ffb2ad44a6a82f347787abc0c7df5f49057cf11770/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:79b2ffbba483f4ed36a0f236ccb85fbb16e670c9238313709638167670ba235f", size = 231724, upload-time = "2025-06-09T23:00:21.705Z" },
+    { url = "https://files.pythonhosted.org/packages/20/fb/03395c0a43a5976af4bf7534759d214405fbbb4c114683f434dfdd3128ef/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a26f205c9ca5829cbf82bb2a84b5c36f7184c4316617d7ef1b271a56720d6b30", size = 245842, upload-time = "2025-06-09T23:00:23.148Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/15/c01c8e1dffdac5d9803507d824f27aed2ba76b6ed0026fab4d9866e82f1f/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bcacfad3185a623fa11ea0e0634aac7b691aa925d50a440f39b458e41c561d98", size = 239767, upload-time = "2025-06-09T23:00:25.103Z" },
+    { url = "https://files.pythonhosted.org/packages/14/99/3f4c6fe882c1f5514b6848aa0a69b20cb5e5d8e8f51a339d48c0e9305ed0/frozenlist-1.7.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:72c1b0fe8fe451b34f12dce46445ddf14bd2a5bcad7e324987194dc8e3a74c86", size = 224130, upload-time = "2025-06-09T23:00:27.061Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/83/220a374bd7b2aeba9d0725130665afe11de347d95c3620b9b82cc2fcab97/frozenlist-1.7.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:61d1a5baeaac6c0798ff6edfaeaa00e0e412d49946c53fae8d4b8e8b3566c4ae", size = 235301, upload-time = "2025-06-09T23:00:29.02Z" },
+    { url = "https://files.pythonhosted.org/packages/03/3c/3e3390d75334a063181625343e8daab61b77e1b8214802cc4e8a1bb678fc/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:7edf5c043c062462f09b6820de9854bf28cc6cc5b6714b383149745e287181a8", size = 234606, upload-time = "2025-06-09T23:00:30.514Z" },
+    { url = "https://files.pythonhosted.org/packages/23/1e/58232c19608b7a549d72d9903005e2d82488f12554a32de2d5fb59b9b1ba/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:d50ac7627b3a1bd2dcef6f9da89a772694ec04d9a61b66cf87f7d9446b4a0c31", size = 248372, upload-time = "2025-06-09T23:00:31.966Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/a4/e4a567e01702a88a74ce8a324691e62a629bf47d4f8607f24bf1c7216e7f/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:ce48b2fece5aeb45265bb7a58259f45027db0abff478e3077e12b05b17fb9da7", size = 229860, upload-time = "2025-06-09T23:00:33.375Z" },
+    { url = "https://files.pythonhosted.org/packages/73/a6/63b3374f7d22268b41a9db73d68a8233afa30ed164c46107b33c4d18ecdd/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:fe2365ae915a1fafd982c146754e1de6ab3478def8a59c86e1f7242d794f97d5", size = 245893, upload-time = "2025-06-09T23:00:35.002Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/eb/d18b3f6e64799a79673c4ba0b45e4cfbe49c240edfd03a68be20002eaeaa/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:45a6f2fdbd10e074e8814eb98b05292f27bad7d1883afbe009d96abdcf3bc898", size = 246323, upload-time = "2025-06-09T23:00:36.468Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/f5/720f3812e3d06cd89a1d5db9ff6450088b8f5c449dae8ffb2971a44da506/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:21884e23cffabb157a9dd7e353779077bf5b8f9a58e9b262c6caad2ef5f80a56", size = 233149, upload-time = "2025-06-09T23:00:37.963Z" },
+    { url = "https://files.pythonhosted.org/packages/69/68/03efbf545e217d5db8446acfd4c447c15b7c8cf4dbd4a58403111df9322d/frozenlist-1.7.0-cp311-cp311-win32.whl", hash = "sha256:284d233a8953d7b24f9159b8a3496fc1ddc00f4db99c324bd5fb5f22d8698ea7", size = 39565, upload-time = "2025-06-09T23:00:39.753Z" },
+    { url = "https://files.pythonhosted.org/packages/58/17/fe61124c5c333ae87f09bb67186d65038834a47d974fc10a5fadb4cc5ae1/frozenlist-1.7.0-cp311-cp311-win_amd64.whl", hash = "sha256:387cbfdcde2f2353f19c2f66bbb52406d06ed77519ac7ee21be0232147c2592d", size = 44019, upload-time = "2025-06-09T23:00:40.988Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/a2/c8131383f1e66adad5f6ecfcce383d584ca94055a34d683bbb24ac5f2f1c/frozenlist-1.7.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:3dbf9952c4bb0e90e98aec1bd992b3318685005702656bc6f67c1a32b76787f2", size = 81424, upload-time = "2025-06-09T23:00:42.24Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/9d/02754159955088cb52567337d1113f945b9e444c4960771ea90eb73de8db/frozenlist-1.7.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:1f5906d3359300b8a9bb194239491122e6cf1444c2efb88865426f170c262cdb", size = 47952, upload-time = "2025-06-09T23:00:43.481Z" },
+    { url = "https://files.pythonhosted.org/packages/01/7a/0046ef1bd6699b40acd2067ed6d6670b4db2f425c56980fa21c982c2a9db/frozenlist-1.7.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3dabd5a8f84573c8d10d8859a50ea2dec01eea372031929871368c09fa103478", size = 46688, upload-time = "2025-06-09T23:00:44.793Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/a2/a910bafe29c86997363fb4c02069df4ff0b5bc39d33c5198b4e9dd42d8f8/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:aa57daa5917f1738064f302bf2626281a1cb01920c32f711fbc7bc36111058a8", size = 243084, upload-time = "2025-06-09T23:00:46.125Z" },
+    { url = "https://files.pythonhosted.org/packages/64/3e/5036af9d5031374c64c387469bfcc3af537fc0f5b1187d83a1cf6fab1639/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:c193dda2b6d49f4c4398962810fa7d7c78f032bf45572b3e04dd5249dff27e08", size = 233524, upload-time = "2025-06-09T23:00:47.73Z" },
+    { url = "https://files.pythonhosted.org/packages/06/39/6a17b7c107a2887e781a48ecf20ad20f1c39d94b2a548c83615b5b879f28/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bfe2b675cf0aaa6d61bf8fbffd3c274b3c9b7b1623beb3809df8a81399a4a9c4", size = 248493, upload-time = "2025-06-09T23:00:49.742Z" },
+    { url = "https://files.pythonhosted.org/packages/be/00/711d1337c7327d88c44d91dd0f556a1c47fb99afc060ae0ef66b4d24793d/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8fc5d5cda37f62b262405cf9652cf0856839c4be8ee41be0afe8858f17f4c94b", size = 244116, upload-time = "2025-06-09T23:00:51.352Z" },
+    { url = "https://files.pythonhosted.org/packages/24/fe/74e6ec0639c115df13d5850e75722750adabdc7de24e37e05a40527ca539/frozenlist-1.7.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b0d5ce521d1dd7d620198829b87ea002956e4319002ef0bc8d3e6d045cb4646e", size = 224557, upload-time = "2025-06-09T23:00:52.855Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/db/48421f62a6f77c553575201e89048e97198046b793f4a089c79a6e3268bd/frozenlist-1.7.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:488d0a7d6a0008ca0db273c542098a0fa9e7dfaa7e57f70acef43f32b3f69dca", size = 241820, upload-time = "2025-06-09T23:00:54.43Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/fa/cb4a76bea23047c8462976ea7b7a2bf53997a0ca171302deae9d6dd12096/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:15a7eaba63983d22c54d255b854e8108e7e5f3e89f647fc854bd77a237e767df", size = 236542, upload-time = "2025-06-09T23:00:56.409Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/32/476a4b5cfaa0ec94d3f808f193301debff2ea42288a099afe60757ef6282/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:1eaa7e9c6d15df825bf255649e05bd8a74b04a4d2baa1ae46d9c2d00b2ca2cb5", size = 249350, upload-time = "2025-06-09T23:00:58.468Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/ba/9a28042f84a6bf8ea5dbc81cfff8eaef18d78b2a1ad9d51c7bc5b029ad16/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:e4389e06714cfa9d47ab87f784a7c5be91d3934cd6e9a7b85beef808297cc025", size = 225093, upload-time = "2025-06-09T23:01:00.015Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/29/3a32959e68f9cf000b04e79ba574527c17e8842e38c91d68214a37455786/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:73bd45e1488c40b63fe5a7df892baf9e2a4d4bb6409a2b3b78ac1c6236178e01", size = 245482, upload-time = "2025-06-09T23:01:01.474Z" },
+    { url = "https://files.pythonhosted.org/packages/80/e8/edf2f9e00da553f07f5fa165325cfc302dead715cab6ac8336a5f3d0adc2/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:99886d98e1643269760e5fe0df31e5ae7050788dd288947f7f007209b8c33f08", size = 249590, upload-time = "2025-06-09T23:01:02.961Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/80/9a0eb48b944050f94cc51ee1c413eb14a39543cc4f760ed12657a5a3c45a/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:290a172aae5a4c278c6da8a96222e6337744cd9c77313efe33d5670b9f65fc43", size = 237785, upload-time = "2025-06-09T23:01:05.095Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/74/87601e0fb0369b7a2baf404ea921769c53b7ae00dee7dcfe5162c8c6dbf0/frozenlist-1.7.0-cp312-cp312-win32.whl", hash = "sha256:426c7bc70e07cfebc178bc4c2bf2d861d720c4fff172181eeb4a4c41d4ca2ad3", size = 39487, upload-time = "2025-06-09T23:01:06.54Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/15/c026e9a9fc17585a9d461f65d8593d281fedf55fbf7eb53f16c6df2392f9/frozenlist-1.7.0-cp312-cp312-win_amd64.whl", hash = "sha256:563b72efe5da92e02eb68c59cb37205457c977aa7a449ed1b37e6939e5c47c6a", size = 43874, upload-time = "2025-06-09T23:01:07.752Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/b1/ee59496f51cd244039330015d60f13ce5a54a0f2bd8d79e4a4a375ab7469/frozenlist-1.7.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:cea3dbd15aea1341ea2de490574a4a37ca080b2ae24e4b4f4b51b9057b4c3630", size = 82434, upload-time = "2025-06-09T23:02:05.195Z" },
+    { url = "https://files.pythonhosted.org/packages/75/e1/d518391ce36a6279b3fa5bc14327dde80bcb646bb50d059c6ca0756b8d05/frozenlist-1.7.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7d536ee086b23fecc36c2073c371572374ff50ef4db515e4e503925361c24f71", size = 48232, upload-time = "2025-06-09T23:02:07.728Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/8d/a0d04f28b6e821a9685c22e67b5fb798a5a7b68752f104bfbc2dccf080c4/frozenlist-1.7.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:dfcebf56f703cb2e346315431699f00db126d158455e513bd14089d992101e44", size = 47186, upload-time = "2025-06-09T23:02:09.243Z" },
+    { url = "https://files.pythonhosted.org/packages/93/3a/a5334c0535c8b7c78eeabda1579179e44fe3d644e07118e59a2276dedaf1/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:974c5336e61d6e7eb1ea5b929cb645e882aadab0095c5a6974a111e6479f8878", size = 226617, upload-time = "2025-06-09T23:02:10.949Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/67/8258d971f519dc3f278c55069a775096cda6610a267b53f6248152b72b2f/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:c70db4a0ab5ab20878432c40563573229a7ed9241506181bba12f6b7d0dc41cb", size = 224179, upload-time = "2025-06-09T23:02:12.603Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/89/8225905bf889b97c6d935dd3aeb45668461e59d415cb019619383a8a7c3b/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1137b78384eebaf70560a36b7b229f752fb64d463d38d1304939984d5cb887b6", size = 235783, upload-time = "2025-06-09T23:02:14.678Z" },
+    { url = "https://files.pythonhosted.org/packages/54/6e/ef52375aa93d4bc510d061df06205fa6dcfd94cd631dd22956b09128f0d4/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e793a9f01b3e8b5c0bc646fb59140ce0efcc580d22a3468d70766091beb81b35", size = 229210, upload-time = "2025-06-09T23:02:16.313Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/55/62c87d1a6547bfbcd645df10432c129100c5bd0fd92a384de6e3378b07c1/frozenlist-1.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:74739ba8e4e38221d2c5c03d90a7e542cb8ad681915f4ca8f68d04f810ee0a87", size = 215994, upload-time = "2025-06-09T23:02:17.9Z" },
+    { url = "https://files.pythonhosted.org/packages/45/d2/263fea1f658b8ad648c7d94d18a87bca7e8c67bd6a1bbf5445b1bd5b158c/frozenlist-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1e63344c4e929b1a01e29bc184bbb5fd82954869033765bfe8d65d09e336a677", size = 225122, upload-time = "2025-06-09T23:02:19.479Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/22/7145e35d12fb368d92124f679bea87309495e2e9ddf14c6533990cb69218/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:2ea2a7369eb76de2217a842f22087913cdf75f63cf1307b9024ab82dfb525938", size = 224019, upload-time = "2025-06-09T23:02:20.969Z" },
+    { url = "https://files.pythonhosted.org/packages/44/1e/7dae8c54301beb87bcafc6144b9a103bfd2c8f38078c7902984c9a0c4e5b/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:836b42f472a0e006e02499cef9352ce8097f33df43baaba3e0a28a964c26c7d2", size = 239925, upload-time = "2025-06-09T23:02:22.466Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/1e/99c93e54aa382e949a98976a73b9b20c3aae6d9d893f31bbe4991f64e3a8/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:e22b9a99741294b2571667c07d9f8cceec07cb92aae5ccda39ea1b6052ed4319", size = 220881, upload-time = "2025-06-09T23:02:24.521Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/9c/ca5105fa7fb5abdfa8837581be790447ae051da75d32f25c8f81082ffc45/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:9a19e85cc503d958abe5218953df722748d87172f71b73cf3c9257a91b999890", size = 234046, upload-time = "2025-06-09T23:02:26.206Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/4d/e99014756093b4ddbb67fb8f0df11fe7a415760d69ace98e2ac6d5d43402/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:f22dac33bb3ee8fe3e013aa7b91dc12f60d61d05b7fe32191ffa84c3aafe77bd", size = 235756, upload-time = "2025-06-09T23:02:27.79Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/72/a19a40bcdaa28a51add2aaa3a1a294ec357f36f27bd836a012e070c5e8a5/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:9ccec739a99e4ccf664ea0775149f2749b8a6418eb5b8384b4dc0a7d15d304cb", size = 222894, upload-time = "2025-06-09T23:02:29.848Z" },
+    { url = "https://files.pythonhosted.org/packages/08/49/0042469993e023a758af81db68c76907cd29e847d772334d4d201cbe9a42/frozenlist-1.7.0-cp39-cp39-win32.whl", hash = "sha256:b3950f11058310008a87757f3eee16a8e1ca97979833239439586857bc25482e", size = 39848, upload-time = "2025-06-09T23:02:31.413Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/45/827d86ee475c877f5f766fbc23fb6acb6fada9e52f1c9720e2ba3eae32da/frozenlist-1.7.0-cp39-cp39-win_amd64.whl", hash = "sha256:43a82fce6769c70f2f5a06248b614a7d268080a9d20f7457ef10ecee5af82b63", size = 44102, upload-time = "2025-06-09T23:02:32.808Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/45/b82e3c16be2182bff01179db177fe144d58b5dc787a7d4492c6ed8b9317f/frozenlist-1.7.0-py3-none-any.whl", hash = "sha256:9a5af342e34f7e97caf8c995864c7a396418ae2859cc6fdf1b1073020d516a7e", size = 13106, upload-time = "2025-06-09T23:02:34.204Z" },
+]
+
+[[package]]
+name = "fschat"
+version = "0.2.36"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "aiohttp" },
+    { name = "fastapi" },
+    { name = "httpx" },
+    { name = "markdown2", extra = ["all"] },
+    { name = "nh3" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "prompt-toolkit" },
+    { name = "pydantic" },
+    { name = "requests" },
+    { name = "rich" },
+    { name = "shortuuid" },
+    { name = "tiktoken" },
+    { name = "uvicorn" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/22/ad/8336e2dc6323944d59e888d41ff8833993ebc9c79bdd8d07997c54d92d55/fschat-0.2.36.tar.gz", hash = "sha256:ac724458bb105cb5b3b9ec852150345dfa7f61fd7c275d6635d66fa1ed904d90", size = 195280, upload-time = "2024-02-11T15:42:27.02Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/cd/3651634ab25f74e64eaf6a6e08e359320d6b26e637afe622762fe99d127b/fschat-0.2.36-py3-none-any.whl", hash = "sha256:a66ecf1e90f6db1d015096f6aa8a305d20d0a13e067951285aecdb56a33e9cb1", size = 256902, upload-time = "2024-02-11T15:42:24.819Z" },
 ]
 
 [[package]]
 name = "fsspec"
-version = "2024.9.0"
+version = "2025.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/62/7c/12b0943011daaaa9c35c2a2e22e5eb929ac90002f08f1259d69aedad84de/fsspec-2024.9.0.tar.gz", hash = "sha256:4b0afb90c2f21832df142f292649035d80b421f60a9e1c027802e5a0da2b04e8", size = 286206 }
+sdist = { url = "https://files.pythonhosted.org/packages/34/f4/5721faf47b8c499e776bc34c6a8fc17efdf7fdef0b00f398128bc5dcb4ac/fsspec-2025.3.0.tar.gz", hash = "sha256:a935fd1ea872591f2b5148907d103488fc523295e6c64b835cfad8c3eca44972", size = 298491, upload-time = "2025-03-07T21:47:56.461Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/1d/a0/6aaea0c2fbea2f89bfd5db25fb1e3481896a423002ebe4e55288907a97a3/fsspec-2024.9.0-py3-none-any.whl", hash = "sha256:a0947d552d8a6efa72cc2c730b12c41d043509156966cca4fb157b0f2a0c574b", size = 179253 },
+    { url = "https://files.pythonhosted.org/packages/56/53/eb690efa8513166adef3e0669afd31e95ffde69fb3c52ec2ac7223ed6018/fsspec-2025.3.0-py3-none-any.whl", hash = "sha256:efb87af3efa9103f94ca91a7f8cb7a4df91af9f74fc106c9c7ea0efd7277c1b3", size = 193615, upload-time = "2025-03-07T21:47:54.809Z" },
 ]
 
 [package.optional-dependencies]
@@ -801,78 +853,196 @@ http = [
     { name = "aiohttp" },
 ]
 
+[[package]]
+name = "gast"
+version = "0.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3c/14/c566f5ca00c115db7725263408ff952b8ae6d6a4e792ef9c84e77d9af7a1/gast-0.6.0.tar.gz", hash = "sha256:88fc5300d32c7ac6ca7b515310862f71e6fdf2c029bbec7c66c0f5dd47b6b1fb", size = 27708, upload-time = "2024-06-27T20:31:49.527Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a3/61/8001b38461d751cd1a0c3a6ae84346796a5758123f3ed97a1b121dfbf4f3/gast-0.6.0-py3-none-any.whl", hash = "sha256:52b182313f7330389f72b069ba00f174cfe2a06411099547288839c6cbafbd54", size = 21173, upload-time = "2024-07-09T13:15:15.615Z" },
+]
+
 [[package]]
 name = "genson"
 version = "1.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/c5/cf/2303c8ad276dcf5ee2ad6cf69c4338fd86ef0f471a5207b069adf7a393cf/genson-1.3.0.tar.gz", hash = "sha256:e02db9ac2e3fd29e65b5286f7135762e2cd8a986537c075b06fc5f1517308e37", size = 34919 }
+sdist = { url = "https://files.pythonhosted.org/packages/c5/cf/2303c8ad276dcf5ee2ad6cf69c4338fd86ef0f471a5207b069adf7a393cf/genson-1.3.0.tar.gz", hash = "sha256:e02db9ac2e3fd29e65b5286f7135762e2cd8a986537c075b06fc5f1517308e37", size = 34919, upload-time = "2024-05-15T22:08:49.123Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f8/5c/e226de133afd8bb267ec27eead9ae3d784b95b39a287ed404caab39a5f50/genson-1.3.0-py3-none-any.whl", hash = "sha256:468feccd00274cc7e4c09e84b08704270ba8d95232aa280f65b986139cec67f7", size = 21470, upload-time = "2024-05-15T22:08:47.056Z" },
+]
+
+[[package]]
+name = "ghp-import"
+version = "2.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "python-dateutil" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/d9/29/d40217cbe2f6b1359e00c6c307bb3fc876ba74068cbab3dde77f03ca0dc4/ghp-import-2.1.0.tar.gz", hash = "sha256:9c535c4c61193c2df8871222567d7fd7e5014d835f97dc7b7439069e2413d343", size = 10943, upload-time = "2022-05-02T15:47:16.11Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f7/ec/67fbef5d497f86283db54c22eec6f6140243aae73265799baaaa19cd17fb/ghp_import-2.1.0-py3-none-any.whl", hash = "sha256:8337dd7b50877f163d4c0289bc1f1c7f127550241988d568c1db512c4324a619", size = 11034, upload-time = "2022-05-02T15:47:14.552Z" },
+]
+
+[[package]]
+name = "google-auth"
+version = "2.40.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cachetools" },
+    { name = "pyasn1-modules" },
+    { name = "rsa" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9e/9b/e92ef23b84fa10a64ce4831390b7a4c2e53c0132568d99d4ae61d04c8855/google_auth-2.40.3.tar.gz", hash = "sha256:500c3a29adedeb36ea9cf24b8d10858e152f2412e3ca37829b3fa18e33d63b77", size = 281029, upload-time = "2025-06-04T18:04:57.577Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/17/63/b19553b658a1692443c62bd07e5868adaa0ad746a0751ba62c59568cd45b/google_auth-2.40.3-py2.py3-none-any.whl", hash = "sha256:1370d4593e86213563547f97a92752fc658456fe4514c809544f330fed45a7ca", size = 216137, upload-time = "2025-06-04T18:04:55.573Z" },
+]
+
+[[package]]
+name = "google-genai"
+version = "1.32.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+    { name = "google-auth" },
+    { name = "httpx" },
+    { name = "pydantic" },
+    { name = "requests" },
+    { name = "tenacity" },
+    { name = "typing-extensions" },
+    { name = "websockets" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/03/ab/e6cdd8fa957c647ef00c4da7c59d0e734354bd49ed8d98c860732d8e1944/google_genai-1.32.0.tar.gz", hash = "sha256:349da3f5ff0e981066bd508585fcdd308d28fc4646f318c8f6d1aa6041f4c7e3", size = 240802, upload-time = "2025-08-27T22:16:32.781Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f8/5c/e226de133afd8bb267ec27eead9ae3d784b95b39a287ed404caab39a5f50/genson-1.3.0-py3-none-any.whl", hash = "sha256:468feccd00274cc7e4c09e84b08704270ba8d95232aa280f65b986139cec67f7", size = 21470 },
+    { url = "https://files.pythonhosted.org/packages/59/55/be09472f7a656af1208196d2ef9a3d2710f3cbcf695f51acbcbe28b9472b/google_genai-1.32.0-py3-none-any.whl", hash = "sha256:c0c4b1d45adf3aa99501050dd73da2f0dea09374002231052d81a6765d15e7f6", size = 241680, upload-time = "2025-08-27T22:16:31.409Z" },
+]
+
+[[package]]
+name = "google-pasta"
+version = "0.2.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "six" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/35/4a/0bd53b36ff0323d10d5f24ebd67af2de10a1117f5cf4d7add90df92756f1/google-pasta-0.2.0.tar.gz", hash = "sha256:c9f2c8dfc8f96d0d5808299920721be30c9eec37f2389f28904f454565c8a16e", size = 40430, upload-time = "2020-03-13T18:57:50.34Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a3/de/c648ef6835192e6e2cc03f40b19eeda4382c49b5bafb43d88b931c4c74ac/google_pasta-0.2.0-py3-none-any.whl", hash = "sha256:b32482794a366b5366a32c92a9a9201b107821889935a02b3e51f6b432ea84ed", size = 57471, upload-time = "2020-03-13T18:57:48.872Z" },
+]
+
+[[package]]
+name = "grpcio"
+version = "1.74.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/38/b4/35feb8f7cab7239c5b94bd2db71abb3d6adb5f335ad8f131abb6060840b6/grpcio-1.74.0.tar.gz", hash = "sha256:80d1f4fbb35b0742d3e3d3bb654b7381cd5f015f8497279a1e9c21ba623e01b1", size = 12756048, upload-time = "2025-07-24T18:54:23.039Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/66/54/68e51a90797ad7afc5b0a7881426c337f6a9168ebab73c3210b76aa7c90d/grpcio-1.74.0-cp310-cp310-linux_armv7l.whl", hash = "sha256:85bd5cdf4ed7b2d6438871adf6afff9af7096486fcf51818a81b77ef4dd30907", size = 5481935, upload-time = "2025-07-24T18:52:43.756Z" },
+    { url = "https://files.pythonhosted.org/packages/32/2a/af817c7e9843929e93e54d09c9aee2555c2e8d81b93102a9426b36e91833/grpcio-1.74.0-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:68c8ebcca945efff9d86d8d6d7bfb0841cf0071024417e2d7f45c5e46b5b08eb", size = 10986796, upload-time = "2025-07-24T18:52:47.219Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/94/d67756638d7bb07750b07d0826c68e414124574b53840ba1ff777abcd388/grpcio-1.74.0-cp310-cp310-manylinux_2_17_aarch64.whl", hash = "sha256:e154d230dc1bbbd78ad2fdc3039fa50ad7ffcf438e4eb2fa30bce223a70c7486", size = 5983663, upload-time = "2025-07-24T18:52:49.463Z" },
+    { url = "https://files.pythonhosted.org/packages/35/f5/c5e4853bf42148fea8532d49e919426585b73eafcf379a712934652a8de9/grpcio-1.74.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e8978003816c7b9eabe217f88c78bc26adc8f9304bf6a594b02e5a49b2ef9c11", size = 6653765, upload-time = "2025-07-24T18:52:51.094Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/75/a1991dd64b331d199935e096cc9daa3415ee5ccbe9f909aa48eded7bba34/grpcio-1.74.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3d7bd6e3929fd2ea7fbc3f562e4987229ead70c9ae5f01501a46701e08f1ad9", size = 6215172, upload-time = "2025-07-24T18:52:53.282Z" },
+    { url = "https://files.pythonhosted.org/packages/01/a4/7cef3dbb3b073d0ce34fd507efc44ac4c9442a0ef9fba4fb3f5c551efef5/grpcio-1.74.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:136b53c91ac1d02c8c24201bfdeb56f8b3ac3278668cbb8e0ba49c88069e1bdc", size = 6329142, upload-time = "2025-07-24T18:52:54.927Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/d3/587920f882b46e835ad96014087054655312400e2f1f1446419e5179a383/grpcio-1.74.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:fe0f540750a13fd8e5da4b3eaba91a785eea8dca5ccd2bc2ffe978caa403090e", size = 7018632, upload-time = "2025-07-24T18:52:56.523Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/95/c70a3b15a0bc83334b507e3d2ae20ee8fa38d419b8758a4d838f5c2a7d32/grpcio-1.74.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:4e4181bfc24413d1e3a37a0b7889bea68d973d4b45dd2bc68bb766c140718f82", size = 6509641, upload-time = "2025-07-24T18:52:58.495Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/06/2e7042d06247d668ae69ea6998eca33f475fd4e2855f94dcb2aa5daef334/grpcio-1.74.0-cp310-cp310-win32.whl", hash = "sha256:1733969040989f7acc3d94c22f55b4a9501a30f6aaacdbccfaba0a3ffb255ab7", size = 3817478, upload-time = "2025-07-24T18:53:00.128Z" },
+    { url = "https://files.pythonhosted.org/packages/93/20/e02b9dcca3ee91124060b65bbf5b8e1af80b3b76a30f694b44b964ab4d71/grpcio-1.74.0-cp310-cp310-win_amd64.whl", hash = "sha256:9e912d3c993a29df6c627459af58975b2e5c897d93287939b9d5065f000249b5", size = 4493971, upload-time = "2025-07-24T18:53:02.068Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/77/b2f06db9f240a5abeddd23a0e49eae2b6ac54d85f0e5267784ce02269c3b/grpcio-1.74.0-cp311-cp311-linux_armv7l.whl", hash = "sha256:69e1a8180868a2576f02356565f16635b99088da7df3d45aaa7e24e73a054e31", size = 5487368, upload-time = "2025-07-24T18:53:03.548Z" },
+    { url = "https://files.pythonhosted.org/packages/48/99/0ac8678a819c28d9a370a663007581744a9f2a844e32f0fa95e1ddda5b9e/grpcio-1.74.0-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:8efe72fde5500f47aca1ef59495cb59c885afe04ac89dd11d810f2de87d935d4", size = 10999804, upload-time = "2025-07-24T18:53:05.095Z" },
+    { url = "https://files.pythonhosted.org/packages/45/c6/a2d586300d9e14ad72e8dc211c7aecb45fe9846a51e558c5bca0c9102c7f/grpcio-1.74.0-cp311-cp311-manylinux_2_17_aarch64.whl", hash = "sha256:a8f0302f9ac4e9923f98d8e243939a6fb627cd048f5cd38595c97e38020dffce", size = 5987667, upload-time = "2025-07-24T18:53:07.157Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/57/5f338bf56a7f22584e68d669632e521f0de460bb3749d54533fc3d0fca4f/grpcio-1.74.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2f609a39f62a6f6f05c7512746798282546358a37ea93c1fcbadf8b2fed162e3", size = 6655612, upload-time = "2025-07-24T18:53:09.244Z" },
+    { url = "https://files.pythonhosted.org/packages/82/ea/a4820c4c44c8b35b1903a6c72a5bdccec92d0840cf5c858c498c66786ba5/grpcio-1.74.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c98e0b7434a7fa4e3e63f250456eaef52499fba5ae661c58cc5b5477d11e7182", size = 6219544, upload-time = "2025-07-24T18:53:11.221Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/17/0537630a921365928f5abb6d14c79ba4dcb3e662e0dbeede8af4138d9dcf/grpcio-1.74.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:662456c4513e298db6d7bd9c3b8df6f75f8752f0ba01fb653e252ed4a59b5a5d", size = 6334863, upload-time = "2025-07-24T18:53:12.925Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/a6/85ca6cb9af3f13e1320d0a806658dca432ff88149d5972df1f7b51e87127/grpcio-1.74.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:3d14e3c4d65e19d8430a4e28ceb71ace4728776fd6c3ce34016947474479683f", size = 7019320, upload-time = "2025-07-24T18:53:15.002Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/a7/fe2beab970a1e25d2eff108b3cf4f7d9a53c185106377a3d1989216eba45/grpcio-1.74.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:1bf949792cee20d2078323a9b02bacbbae002b9e3b9e2433f2741c15bdeba1c4", size = 6514228, upload-time = "2025-07-24T18:53:16.999Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/c2/2f9c945c8a248cebc3ccda1b7a1bf1775b9d7d59e444dbb18c0014e23da6/grpcio-1.74.0-cp311-cp311-win32.whl", hash = "sha256:55b453812fa7c7ce2f5c88be3018fb4a490519b6ce80788d5913f3f9d7da8c7b", size = 3817216, upload-time = "2025-07-24T18:53:20.564Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/d1/a9cf9c94b55becda2199299a12b9feef0c79946b0d9d34c989de6d12d05d/grpcio-1.74.0-cp311-cp311-win_amd64.whl", hash = "sha256:86ad489db097141a907c559988c29718719aa3e13370d40e20506f11b4de0d11", size = 4495380, upload-time = "2025-07-24T18:53:22.058Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/5d/e504d5d5c4469823504f65687d6c8fb97b7f7bf0b34873b7598f1df24630/grpcio-1.74.0-cp312-cp312-linux_armv7l.whl", hash = "sha256:8533e6e9c5bd630ca98062e3a1326249e6ada07d05acf191a77bc33f8948f3d8", size = 5445551, upload-time = "2025-07-24T18:53:23.641Z" },
+    { url = "https://files.pythonhosted.org/packages/43/01/730e37056f96f2f6ce9f17999af1556df62ee8dab7fa48bceeaab5fd3008/grpcio-1.74.0-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:2918948864fec2a11721d91568effffbe0a02b23ecd57f281391d986847982f6", size = 10979810, upload-time = "2025-07-24T18:53:25.349Z" },
+    { url = "https://files.pythonhosted.org/packages/79/3d/09fd100473ea5c47083889ca47ffd356576173ec134312f6aa0e13111dee/grpcio-1.74.0-cp312-cp312-manylinux_2_17_aarch64.whl", hash = "sha256:60d2d48b0580e70d2e1954d0d19fa3c2e60dd7cbed826aca104fff518310d1c5", size = 5941946, upload-time = "2025-07-24T18:53:27.387Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/99/12d2cca0a63c874c6d3d195629dcd85cdf5d6f98a30d8db44271f8a97b93/grpcio-1.74.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3601274bc0523f6dc07666c0e01682c94472402ac2fd1226fd96e079863bfa49", size = 6621763, upload-time = "2025-07-24T18:53:29.193Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/2c/930b0e7a2f1029bbc193443c7bc4dc2a46fedb0203c8793dcd97081f1520/grpcio-1.74.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:176d60a5168d7948539def20b2a3adcce67d72454d9ae05969a2e73f3a0feee7", size = 6180664, upload-time = "2025-07-24T18:53:30.823Z" },
+    { url = "https://files.pythonhosted.org/packages/db/d5/ff8a2442180ad0867717e670f5ec42bfd8d38b92158ad6bcd864e6d4b1ed/grpcio-1.74.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:e759f9e8bc908aaae0412642afe5416c9f983a80499448fcc7fab8692ae044c3", size = 6301083, upload-time = "2025-07-24T18:53:32.454Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/ba/b361d390451a37ca118e4ec7dccec690422e05bc85fba2ec72b06cefec9f/grpcio-1.74.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:9e7c4389771855a92934b2846bd807fc25a3dfa820fd912fe6bd8136026b2707", size = 6994132, upload-time = "2025-07-24T18:53:34.506Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/0c/3a5fa47d2437a44ced74141795ac0251bbddeae74bf81df3447edd767d27/grpcio-1.74.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:cce634b10aeab37010449124814b05a62fb5f18928ca878f1bf4750d1f0c815b", size = 6489616, upload-time = "2025-07-24T18:53:36.217Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/95/ab64703b436d99dc5217228babc76047d60e9ad14df129e307b5fec81fd0/grpcio-1.74.0-cp312-cp312-win32.whl", hash = "sha256:885912559974df35d92219e2dc98f51a16a48395f37b92865ad45186f294096c", size = 3807083, upload-time = "2025-07-24T18:53:37.911Z" },
+    { url = "https://files.pythonhosted.org/packages/84/59/900aa2445891fc47a33f7d2f76e00ca5d6ae6584b20d19af9c06fa09bf9a/grpcio-1.74.0-cp312-cp312-win_amd64.whl", hash = "sha256:42f8fee287427b94be63d916c90399ed310ed10aadbf9e2e5538b3e497d269bc", size = 4490123, upload-time = "2025-07-24T18:53:39.528Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/de/dd7db504703c3b669f1b83265e2fbb5d79c8d3da86ea52cbd9202b9a8b05/grpcio-1.74.0-cp39-cp39-linux_armv7l.whl", hash = "sha256:4bc5fca10aaf74779081e16c2bcc3d5ec643ffd528d9e7b1c9039000ead73bae", size = 5480998, upload-time = "2025-07-24T18:54:01.868Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/57/6537ace3af4c97f2b013ceff1f2e789c52b8448334ca3a0c36e7421cf6ed/grpcio-1.74.0-cp39-cp39-macosx_11_0_universal2.whl", hash = "sha256:6bab67d15ad617aff094c382c882e0177637da73cbc5532d52c07b4ee887a87b", size = 10990945, upload-time = "2025-07-24T18:54:03.854Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/f2/579410017de16cd27f35326df6fecde81bff6e9b43c871d28263fa8a77a4/grpcio-1.74.0-cp39-cp39-manylinux_2_17_aarch64.whl", hash = "sha256:655726919b75ab3c34cdad39da5c530ac6fa32696fb23119e36b64adcfca174a", size = 5983968, upload-time = "2025-07-24T18:54:06.434Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/6a/0f3571003663d991f4ea953b82dc518fed094c182decc48c9b0242bec7e3/grpcio-1.74.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1a2b06afe2e50ebfd46247ac3ba60cac523f54ec7792ae9ba6073c12daf26f0a", size = 6654768, upload-time = "2025-07-24T18:54:08.632Z" },
+    { url = "https://files.pythonhosted.org/packages/24/e3/1d42cb00e0390bacab3c9ee79e37416140d907c8c7c7a92654c535805963/grpcio-1.74.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5f251c355167b2360537cf17bea2cf0197995e551ab9da6a0a59b3da5e8704f9", size = 6215627, upload-time = "2025-07-24T18:54:10.584Z" },
+    { url = "https://files.pythonhosted.org/packages/77/84/4f8312bc4430eda1cdbc4e8689f54daa807b5d304d4ea53e9d27c448889b/grpcio-1.74.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:8f7b5882fb50632ab1e48cb3122d6df55b9afabc265582808036b6e51b9fd6b7", size = 6330938, upload-time = "2025-07-24T18:54:12.557Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/c0/422d2b40110716a4775212256a56ac71586be2403a7b7055818bfd0fc203/grpcio-1.74.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:834988b6c34515545b3edd13e902c1acdd9f2465d386ea5143fb558f153a7176", size = 7019216, upload-time = "2025-07-24T18:54:14.475Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/84/668ab6df27fb35886dfa1242f2d302d0cd319c72e3dd3845a322ecabf61b/grpcio-1.74.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:22b834cef33429ca6cc28303c9c327ba9a3fafecbf62fae17e9a7b7163cc43ac", size = 6510719, upload-time = "2025-07-24T18:54:16.775Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/6a/981150f20dd435b9c46cd504038b4fbae2171b43fe70019914d80159e156/grpcio-1.74.0-cp39-cp39-win32.whl", hash = "sha256:7d95d71ff35291bab3f1c52f52f474c632db26ea12700c2ff0ea0532cb0b5854", size = 3819185, upload-time = "2025-07-24T18:54:18.673Z" },
+    { url = "https://files.pythonhosted.org/packages/75/5f/d64b9745bb9def186e1be11b42d4d310570799d6170ac75829ef1c67c176/grpcio-1.74.0-cp39-cp39-win_amd64.whl", hash = "sha256:ecde9ab49f58433abe02f9ed076c7b5be839cf0153883a6d23995937a82392fa", size = 4495789, upload-time = "2025-07-24T18:54:20.582Z" },
 ]
 
 [[package]]
 name = "h11"
-version = "0.14.0"
+version = "0.16.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f5/38/3af3d3633a34a3316095b39c8e8fb4853a28a536e55d347bd8d8e9a14b03/h11-0.14.0.tar.gz", hash = "sha256:8f19fbbe99e72420ff35c00b27a34cb9937e902a8b810e2c88300c6f0a3b699d", size = 100418 }
+sdist = { url = "https://files.pythonhosted.org/packages/01/ee/02a2c011bdab74c6fb3c75474d40b3052059d95df7e73351460c8588d963/h11-0.16.0.tar.gz", hash = "sha256:4e35b956cf45792e4caa5885e69fba00bdbc6ffafbfa020300e549b208ee5ff1", size = 101250, upload-time = "2025-04-24T03:35:25.427Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/95/04/ff642e65ad6b90db43e668d70ffb6736436c7ce41fcc549f4e9472234127/h11-0.14.0-py3-none-any.whl", hash = "sha256:e3fe4ac4b851c468cc8363d500db52c2ead036020723024a109d37346efaa761", size = 58259 },
+    { url = "https://files.pythonhosted.org/packages/04/4b/29cac41a4d98d144bf5f6d33995617b185d14b22401f75ca86f384e87ff1/h11-0.16.0-py3-none-any.whl", hash = "sha256:63cf8bbe7522de3bf65932fda1d9c2772064ffb3dae62d55932da54b31cb6c86", size = 37515, upload-time = "2025-04-24T03:35:24.344Z" },
 ]
 
 [[package]]
-name = "httpcore"
-version = "1.0.7"
+name = "h5py"
+version = "3.14.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "certifi" },
-    { name = "h11" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/6a/41/d7d0a89eb493922c37d343b607bc1b5da7f5be7e383740b4753ad8943e90/httpcore-1.0.7.tar.gz", hash = "sha256:8551cb62a169ec7162ac7be8d4817d561f60e08eaa485234898414bb5a8a0b4c", size = 85196 }
+sdist = { url = "https://files.pythonhosted.org/packages/5d/57/dfb3c5c3f1bf5f5ef2e59a22dec4ff1f3d7408b55bfcefcfb0ea69ef21c6/h5py-3.14.0.tar.gz", hash = "sha256:2372116b2e0d5d3e5e705b7f663f7c8d96fa79a4052d250484ef91d24d6a08f4", size = 424323, upload-time = "2025-06-06T14:06:15.01Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/87/f5/72347bc88306acb359581ac4d52f23c0ef445b57157adedb9aee0cd689d2/httpcore-1.0.7-py3-none-any.whl", hash = "sha256:a3fff8f43dc260d5bd363d9f9cf1830fa3a458b332856f34282de498ed420edd", size = 78551 },
+    { url = "https://files.pythonhosted.org/packages/52/89/06cbb421e01dea2e338b3154326523c05d9698f89a01f9d9b65e1ec3fb18/h5py-3.14.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:24df6b2622f426857bda88683b16630014588a0e4155cba44e872eb011c4eaed", size = 3332522, upload-time = "2025-06-06T14:04:13.775Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/e7/6c860b002329e408348735bfd0459e7b12f712c83d357abeef3ef404eaa9/h5py-3.14.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6ff2389961ee5872de697054dd5a033b04284afc3fb52dc51d94561ece2c10c6", size = 2831051, upload-time = "2025-06-06T14:04:18.206Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/cd/3dd38cdb7cc9266dc4d85f27f0261680cb62f553f1523167ad7454e32b11/h5py-3.14.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:016e89d3be4c44f8d5e115fab60548e518ecd9efe9fa5c5324505a90773e6f03", size = 4324677, upload-time = "2025-06-06T14:04:23.438Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/45/e1a754dc7cd465ba35e438e28557119221ac89b20aaebef48282654e3dc7/h5py-3.14.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1223b902ef0b5d90bcc8a4778218d6d6cd0f5561861611eda59fa6c52b922f4d", size = 4557272, upload-time = "2025-06-06T14:04:28.863Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/06/f9506c1531645829d302c420851b78bb717af808dde11212c113585fae42/h5py-3.14.0-cp310-cp310-win_amd64.whl", hash = "sha256:852b81f71df4bb9e27d407b43071d1da330d6a7094a588efa50ef02553fa7ce4", size = 2866734, upload-time = "2025-06-06T14:04:33.5Z" },
+    { url = "https://files.pythonhosted.org/packages/61/1b/ad24a8ce846cf0519695c10491e99969d9d203b9632c4fcd5004b1641c2e/h5py-3.14.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:f30dbc58f2a0efeec6c8836c97f6c94afd769023f44e2bb0ed7b17a16ec46088", size = 3352382, upload-time = "2025-06-06T14:04:37.95Z" },
+    { url = "https://files.pythonhosted.org/packages/36/5b/a066e459ca48b47cc73a5c668e9924d9619da9e3c500d9fb9c29c03858ec/h5py-3.14.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:543877d7f3d8f8a9828ed5df6a0b78ca3d8846244b9702e99ed0d53610b583a8", size = 2852492, upload-time = "2025-06-06T14:04:42.092Z" },
+    { url = "https://files.pythonhosted.org/packages/08/0c/5e6aaf221557314bc15ba0e0da92e40b24af97ab162076c8ae009320a42b/h5py-3.14.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8c497600c0496548810047257e36360ff551df8b59156d3a4181072eed47d8ad", size = 4298002, upload-time = "2025-06-06T14:04:47.106Z" },
+    { url = "https://files.pythonhosted.org/packages/21/d4/d461649cafd5137088fb7f8e78fdc6621bb0c4ff2c090a389f68e8edc136/h5py-3.14.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:723a40ee6505bd354bfd26385f2dae7bbfa87655f4e61bab175a49d72ebfc06b", size = 4516618, upload-time = "2025-06-06T14:04:52.467Z" },
+    { url = "https://files.pythonhosted.org/packages/db/0c/6c3f879a0f8e891625817637fad902da6e764e36919ed091dc77529004ac/h5py-3.14.0-cp311-cp311-win_amd64.whl", hash = "sha256:d2744b520440a996f2dae97f901caa8a953afc055db4673a993f2d87d7f38713", size = 2874888, upload-time = "2025-06-06T14:04:56.95Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/77/8f651053c1843391e38a189ccf50df7e261ef8cd8bfd8baba0cbe694f7c3/h5py-3.14.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:e0045115d83272090b0717c555a31398c2c089b87d212ceba800d3dc5d952e23", size = 3312740, upload-time = "2025-06-06T14:05:01.193Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/10/20436a6cf419b31124e59fefc78d74cb061ccb22213226a583928a65d715/h5py-3.14.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6da62509b7e1d71a7d110478aa25d245dd32c8d9a1daee9d2a42dba8717b047a", size = 2829207, upload-time = "2025-06-06T14:05:05.061Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/19/c8bfe8543bfdd7ccfafd46d8cfd96fce53d6c33e9c7921f375530ee1d39a/h5py-3.14.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:554ef0ced3571366d4d383427c00c966c360e178b5fb5ee5bb31a435c424db0c", size = 4708455, upload-time = "2025-06-06T14:05:11.528Z" },
+    { url = "https://files.pythonhosted.org/packages/86/f9/f00de11c82c88bfc1ef22633557bfba9e271e0cb3189ad704183fc4a2644/h5py-3.14.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0cbd41f4e3761f150aa5b662df991868ca533872c95467216f2bec5fcad84882", size = 4929422, upload-time = "2025-06-06T14:05:18.399Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/6d/6426d5d456f593c94b96fa942a9b3988ce4d65ebaf57d7273e452a7222e8/h5py-3.14.0-cp312-cp312-win_amd64.whl", hash = "sha256:bf4897d67e613ecf5bdfbdab39a1158a64df105827da70ea1d90243d796d367f", size = 2862845, upload-time = "2025-06-06T14:05:23.699Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/ac/9ea82488c8790ee5b6ad1a807cd7dc3b9dadfece1cd0e0e369f68a7a8937/h5py-3.14.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f5cc1601e78027cedfec6dd50efb4802f018551754191aeb58d948bd3ec3bd7a", size = 3345097, upload-time = "2025-06-06T14:05:51.984Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/bc/a172ecaaf287e3af2f837f23b470b0a2229c79555a0da9ac8b5cc5bed078/h5py-3.14.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:5e59d2136a8b302afd25acdf7a89b634e0eb7c66b1a211ef2d0457853768a2ef", size = 2843320, upload-time = "2025-06-06T14:05:55.754Z" },
+    { url = "https://files.pythonhosted.org/packages/66/40/b423b57696514e05aa7bb06150ef96667d0e0006cc6de7ab52c71734ab51/h5py-3.14.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:573c33ad056ac7c1ab6d567b6db9df3ffc401045e3f605736218f96c1e0490c6", size = 4326368, upload-time = "2025-06-06T14:06:00.782Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/07/e088f89f04fdbe57ddf9de377f857158d3daa38cf5d0fb20ef9bd489e313/h5py-3.14.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ccbe17dc187c0c64178f1a10aa274ed3a57d055117588942b8a08793cc448216", size = 4559686, upload-time = "2025-06-06T14:06:07.416Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/e4/fb8032d0e5480b1db9b419b5b50737b61bb3c7187c49d809975d62129fb0/h5py-3.14.0-cp39-cp39-win_amd64.whl", hash = "sha256:4f025cf30ae738c4c4e38c7439a761a71ccfcce04c2b87b2a2ac64e8c5171d43", size = 2877166, upload-time = "2025-06-06T14:06:13.05Z" },
 ]
 
 [[package]]
-name = "httptools"
-version = "0.6.4"
+name = "hf-xet"
+version = "1.1.9"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/23/0f/5b60fc28ee7f8cc17a5114a584fd6b86e11c3e0a6e142a7f97a161e9640a/hf_xet-1.1.9.tar.gz", hash = "sha256:c99073ce404462e909f1d5839b2d14a3827b8fe75ed8aed551ba6609c026c803", size = 484242, upload-time = "2025-08-27T23:05:19.441Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/de/12/56e1abb9a44cdef59a411fe8a8673313195711b5ecce27880eb9c8fa90bd/hf_xet-1.1.9-cp37-abi3-macosx_10_12_x86_64.whl", hash = "sha256:a3b6215f88638dd7a6ff82cb4e738dcbf3d863bf667997c093a3c990337d1160", size = 2762553, upload-time = "2025-08-27T23:05:15.153Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/e6/2d0d16890c5f21b862f5df3146519c182e7f0ae49b4b4bf2bd8a40d0b05e/hf_xet-1.1.9-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:9b486de7a64a66f9a172f4b3e0dfe79c9f0a93257c501296a2521a13495a698a", size = 2623216, upload-time = "2025-08-27T23:05:13.778Z" },
+    { url = "https://files.pythonhosted.org/packages/81/42/7e6955cf0621e87491a1fb8cad755d5c2517803cea174229b0ec00ff0166/hf_xet-1.1.9-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a4c5a840c2c4e6ec875ed13703a60e3523bc7f48031dfd750923b2a4d1a5fc3c", size = 3186789, upload-time = "2025-08-27T23:05:12.368Z" },
+    { url = "https://files.pythonhosted.org/packages/df/8b/759233bce05457f5f7ec062d63bbfd2d0c740b816279eaaa54be92aa452a/hf_xet-1.1.9-cp37-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:96a6139c9e44dad1c52c52520db0fffe948f6bce487cfb9d69c125f254bb3790", size = 3088747, upload-time = "2025-08-27T23:05:10.439Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/3c/28cc4db153a7601a996985bcb564f7b8f5b9e1a706c7537aad4b4809f358/hf_xet-1.1.9-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:ad1022e9a998e784c97b2173965d07fe33ee26e4594770b7785a8cc8f922cd95", size = 3251429, upload-time = "2025-08-27T23:05:16.471Z" },
+    { url = "https://files.pythonhosted.org/packages/84/17/7caf27a1d101bfcb05be85850d4aa0a265b2e1acc2d4d52a48026ef1d299/hf_xet-1.1.9-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:86754c2d6d5afb11b0a435e6e18911a4199262fe77553f8c50d75e21242193ea", size = 3354643, upload-time = "2025-08-27T23:05:17.828Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/50/0c39c9eed3411deadcc98749a6699d871b822473f55fe472fad7c01ec588/hf_xet-1.1.9-cp37-abi3-win_amd64.whl", hash = "sha256:5aad3933de6b725d61d51034e04174ed1dce7a57c63d530df0014dea15a40127", size = 2804797, upload-time = "2025-08-27T23:05:20.77Z" },
+]
+
+[[package]]
+name = "httpcore"
+version = "1.0.9"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/a7/9a/ce5e1f7e131522e6d3426e8e7a490b3a01f39a6696602e1c4f33f9e94277/httptools-0.6.4.tar.gz", hash = "sha256:4e93eee4add6493b59a5c514da98c939b244fce4a0d8879cd3f466562f4b7d5c", size = 240639 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/3b/6f/972f8eb0ea7d98a1c6be436e2142d51ad2a64ee18e02b0e7ff1f62171ab1/httptools-0.6.4-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:3c73ce323711a6ffb0d247dcd5a550b8babf0f757e86a52558fe5b86d6fefcc0", size = 198780 },
-    { url = "https://files.pythonhosted.org/packages/6a/b0/17c672b4bc5c7ba7f201eada4e96c71d0a59fbc185e60e42580093a86f21/httptools-0.6.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:345c288418f0944a6fe67be8e6afa9262b18c7626c3ef3c28adc5eabc06a68da", size = 103297 },
-    { url = "https://files.pythonhosted.org/packages/92/5e/b4a826fe91971a0b68e8c2bd4e7db3e7519882f5a8ccdb1194be2b3ab98f/httptools-0.6.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:deee0e3343f98ee8047e9f4c5bc7cedbf69f5734454a94c38ee829fb2d5fa3c1", size = 443130 },
-    { url = "https://files.pythonhosted.org/packages/b0/51/ce61e531e40289a681a463e1258fa1e05e0be54540e40d91d065a264cd8f/httptools-0.6.4-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ca80b7485c76f768a3bc83ea58373f8db7b015551117375e4918e2aa77ea9b50", size = 442148 },
-    { url = "https://files.pythonhosted.org/packages/ea/9e/270b7d767849b0c96f275c695d27ca76c30671f8eb8cc1bab6ced5c5e1d0/httptools-0.6.4-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:90d96a385fa941283ebd231464045187a31ad932ebfa541be8edf5b3c2328959", size = 415949 },
-    { url = "https://files.pythonhosted.org/packages/81/86/ced96e3179c48c6f656354e106934e65c8963d48b69be78f355797f0e1b3/httptools-0.6.4-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:59e724f8b332319e2875efd360e61ac07f33b492889284a3e05e6d13746876f4", size = 417591 },
-    { url = "https://files.pythonhosted.org/packages/75/73/187a3f620ed3175364ddb56847d7a608a6fc42d551e133197098c0143eca/httptools-0.6.4-cp310-cp310-win_amd64.whl", hash = "sha256:c26f313951f6e26147833fc923f78f95604bbec812a43e5ee37f26dc9e5a686c", size = 88344 },
-    { url = "https://files.pythonhosted.org/packages/7b/26/bb526d4d14c2774fe07113ca1db7255737ffbb119315839af2065abfdac3/httptools-0.6.4-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:f47f8ed67cc0ff862b84a1189831d1d33c963fb3ce1ee0c65d3b0cbe7b711069", size = 199029 },
-    { url = "https://files.pythonhosted.org/packages/a6/17/3e0d3e9b901c732987a45f4f94d4e2c62b89a041d93db89eafb262afd8d5/httptools-0.6.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:0614154d5454c21b6410fdf5262b4a3ddb0f53f1e1721cfd59d55f32138c578a", size = 103492 },
-    { url = "https://files.pythonhosted.org/packages/b7/24/0fe235d7b69c42423c7698d086d4db96475f9b50b6ad26a718ef27a0bce6/httptools-0.6.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f8787367fbdfccae38e35abf7641dafc5310310a5987b689f4c32cc8cc3ee975", size = 462891 },
-    { url = "https://files.pythonhosted.org/packages/b1/2f/205d1f2a190b72da6ffb5f41a3736c26d6fa7871101212b15e9b5cd8f61d/httptools-0.6.4-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:40b0f7fe4fd38e6a507bdb751db0379df1e99120c65fbdc8ee6c1d044897a636", size = 459788 },
-    { url = "https://files.pythonhosted.org/packages/6e/4c/d09ce0eff09057a206a74575ae8f1e1e2f0364d20e2442224f9e6612c8b9/httptools-0.6.4-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:40a5ec98d3f49904b9fe36827dcf1aadfef3b89e2bd05b0e35e94f97c2b14721", size = 433214 },
-    { url = "https://files.pythonhosted.org/packages/3e/d2/84c9e23edbccc4a4c6f96a1b8d99dfd2350289e94f00e9ccc7aadde26fb5/httptools-0.6.4-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:dacdd3d10ea1b4ca9df97a0a303cbacafc04b5cd375fa98732678151643d4988", size = 434120 },
-    { url = "https://files.pythonhosted.org/packages/d0/46/4d8e7ba9581416de1c425b8264e2cadd201eb709ec1584c381f3e98f51c1/httptools-0.6.4-cp311-cp311-win_amd64.whl", hash = "sha256:288cd628406cc53f9a541cfaf06041b4c71d751856bab45e3702191f931ccd17", size = 88565 },
-    { url = "https://files.pythonhosted.org/packages/bb/0e/d0b71465c66b9185f90a091ab36389a7352985fe857e352801c39d6127c8/httptools-0.6.4-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:df017d6c780287d5c80601dafa31f17bddb170232d85c066604d8558683711a2", size = 200683 },
-    { url = "https://files.pythonhosted.org/packages/e2/b8/412a9bb28d0a8988de3296e01efa0bd62068b33856cdda47fe1b5e890954/httptools-0.6.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:85071a1e8c2d051b507161f6c3e26155b5c790e4e28d7f236422dbacc2a9cc44", size = 104337 },
-    { url = "https://files.pythonhosted.org/packages/9b/01/6fb20be3196ffdc8eeec4e653bc2a275eca7f36634c86302242c4fbb2760/httptools-0.6.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:69422b7f458c5af875922cdb5bd586cc1f1033295aa9ff63ee196a87519ac8e1", size = 508796 },
-    { url = "https://files.pythonhosted.org/packages/f7/d8/b644c44acc1368938317d76ac991c9bba1166311880bcc0ac297cb9d6bd7/httptools-0.6.4-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:16e603a3bff50db08cd578d54f07032ca1631450ceb972c2f834c2b860c28ea2", size = 510837 },
-    { url = "https://files.pythonhosted.org/packages/52/d8/254d16a31d543073a0e57f1c329ca7378d8924e7e292eda72d0064987486/httptools-0.6.4-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:ec4f178901fa1834d4a060320d2f3abc5c9e39766953d038f1458cb885f47e81", size = 485289 },
-    { url = "https://files.pythonhosted.org/packages/5f/3c/4aee161b4b7a971660b8be71a92c24d6c64372c1ab3ae7f366b3680df20f/httptools-0.6.4-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:f9eb89ecf8b290f2e293325c646a211ff1c2493222798bb80a530c5e7502494f", size = 489779 },
-    { url = "https://files.pythonhosted.org/packages/12/b7/5cae71a8868e555f3f67a50ee7f673ce36eac970f029c0c5e9d584352961/httptools-0.6.4-cp312-cp312-win_amd64.whl", hash = "sha256:db78cb9ca56b59b016e64b6031eda5653be0589dba2b1b43453f6e8b405a0970", size = 88634 },
-    { url = "https://files.pythonhosted.org/packages/94/a3/9fe9ad23fd35f7de6b91eeb60848986058bd8b5a5c1e256f5860a160cc3e/httptools-0.6.4-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ade273d7e767d5fae13fa637f4d53b6e961fb7fd93c7797562663f0171c26660", size = 197214 },
-    { url = "https://files.pythonhosted.org/packages/ea/d9/82d5e68bab783b632023f2fa31db20bebb4e89dfc4d2293945fd68484ee4/httptools-0.6.4-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:856f4bc0478ae143bad54a4242fccb1f3f86a6e1be5548fecfd4102061b3a083", size = 102431 },
-    { url = "https://files.pythonhosted.org/packages/96/c1/cb499655cbdbfb57b577734fde02f6fa0bbc3fe9fb4d87b742b512908dff/httptools-0.6.4-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:322d20ea9cdd1fa98bd6a74b77e2ec5b818abdc3d36695ab402a0de8ef2865a3", size = 473121 },
-    { url = "https://files.pythonhosted.org/packages/af/71/ee32fd358f8a3bb199b03261f10921716990808a675d8160b5383487a317/httptools-0.6.4-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4d87b29bd4486c0093fc64dea80231f7c7f7eb4dc70ae394d70a495ab8436071", size = 473805 },
-    { url = "https://files.pythonhosted.org/packages/8a/0a/0d4df132bfca1507114198b766f1737d57580c9ad1cf93c1ff673e3387be/httptools-0.6.4-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:342dd6946aa6bda4b8f18c734576106b8a31f2fe31492881a9a160ec84ff4bd5", size = 448858 },
-    { url = "https://files.pythonhosted.org/packages/1e/6a/787004fdef2cabea27bad1073bf6a33f2437b4dbd3b6fb4a9d71172b1c7c/httptools-0.6.4-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:4b36913ba52008249223042dca46e69967985fb4051951f94357ea681e1f5dc0", size = 452042 },
-    { url = "https://files.pythonhosted.org/packages/4d/dc/7decab5c404d1d2cdc1bb330b1bf70e83d6af0396fd4fc76fc60c0d522bf/httptools-0.6.4-cp313-cp313-win_amd64.whl", hash = "sha256:28908df1b9bb8187393d5b5db91435ccc9c8e891657f9cbb42a2541b44c82fc8", size = 87682 },
-    { url = "https://files.pythonhosted.org/packages/51/b1/4fc6f52afdf93b7c4304e21f6add9e981e4f857c2fa622a55dfe21b6059e/httptools-0.6.4-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:85797e37e8eeaa5439d33e556662cc370e474445d5fab24dcadc65a8ffb04003", size = 201123 },
-    { url = "https://files.pythonhosted.org/packages/c2/01/e6ecb40ac8fdfb76607c7d3b74a41b464458d5c8710534d8f163b0c15f29/httptools-0.6.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:db353d22843cf1028f43c3651581e4bb49374d85692a85f95f7b9a130e1b2cab", size = 104507 },
-    { url = "https://files.pythonhosted.org/packages/dc/24/c70c34119d209bf08199d938dc9c69164f585ed3029237b4bdb90f673cb9/httptools-0.6.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d1ffd262a73d7c28424252381a5b854c19d9de5f56f075445d33919a637e3547", size = 449615 },
-    { url = "https://files.pythonhosted.org/packages/2b/62/e7f317fed3703bd81053840cacba4e40bcf424b870e4197f94bd1cf9fe7a/httptools-0.6.4-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:703c346571fa50d2e9856a37d7cd9435a25e7fd15e236c397bf224afaa355fe9", size = 448819 },
-    { url = "https://files.pythonhosted.org/packages/2a/13/68337d3be6b023260139434c49d7aa466aaa98f9aee7ed29270ac7dde6a2/httptools-0.6.4-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:aafe0f1918ed07b67c1e838f950b1c1fabc683030477e60b335649b8020e1076", size = 422093 },
-    { url = "https://files.pythonhosted.org/packages/fc/b3/3a1bc45be03dda7a60c7858e55b6cd0489a81613c1908fb81cf21d34ae50/httptools-0.6.4-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:0e563e54979e97b6d13f1bbc05a96109923e76b901f786a5eae36e99c01237bd", size = 423898 },
-    { url = "https://files.pythonhosted.org/packages/05/72/2ddc2ae5f7ace986f7e68a326215b2e7c32e32fd40e6428fa8f1d8065c7e/httptools-0.6.4-cp39-cp39-win_amd64.whl", hash = "sha256:b799de31416ecc589ad79dd85a0b2657a8fe39327944998dea368c1d4c9e55e6", size = 89552 },
+dependencies = [
+    { name = "certifi" },
+    { name = "h11" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/06/94/82699a10bca87a5556c9c59b5963f2d039dbd239f25bc2a63907a05a14cb/httpcore-1.0.9.tar.gz", hash = "sha256:6e34463af53fd2ab5d807f399a9b45ea31c3dfa2276f15a2c3f00afff6e176e8", size = 85484, upload-time = "2025-04-24T22:06:22.219Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7e/f5/f66802a942d491edb555dd61e3a9961140fd64c90bce1eafd741609d334d/httpcore-1.0.9-py3-none-any.whl", hash = "sha256:2d400746a40668fc9dec9810239072b40b4484b640a8c38fd654a024c7a1bf55", size = 78784, upload-time = "2025-04-24T22:06:20.566Z" },
 ]
 
 [[package]]
@@ -885,84 +1055,97 @@ dependencies = [
     { name = "httpcore" },
     { name = "idna" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b1/df/48c586a5fe32a0f01324ee087459e112ebb7224f646c0b5023f5e79e9956/httpx-0.28.1.tar.gz", hash = "sha256:75e98c5f16b0f35b567856f597f06ff2270a374470a5c2392242528e3e3e42fc", size = 141406 }
+sdist = { url = "https://files.pythonhosted.org/packages/b1/df/48c586a5fe32a0f01324ee087459e112ebb7224f646c0b5023f5e79e9956/httpx-0.28.1.tar.gz", hash = "sha256:75e98c5f16b0f35b567856f597f06ff2270a374470a5c2392242528e3e3e42fc", size = 141406, upload-time = "2024-12-06T15:37:23.222Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/2a/39/e50c7c3a983047577ee07d2a9e53faf5a69493943ec3f6a384bdc792deb2/httpx-0.28.1-py3-none-any.whl", hash = "sha256:d909fcccc110f8c7faf814ca82a9a4d816bc5a6dbfea25d6591d6985b8ba59ad", size = 73517 },
+    { url = "https://files.pythonhosted.org/packages/2a/39/e50c7c3a983047577ee07d2a9e53faf5a69493943ec3f6a384bdc792deb2/httpx-0.28.1-py3-none-any.whl", hash = "sha256:d909fcccc110f8c7faf814ca82a9a4d816bc5a6dbfea25d6591d6985b8ba59ad", size = 73517, upload-time = "2024-12-06T15:37:21.509Z" },
 ]
 
 [[package]]
 name = "huggingface-hub"
-version = "0.27.1"
+version = "0.34.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "filelock" },
     { name = "fsspec" },
+    { name = "hf-xet", marker = "platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'arm64' or platform_machine == 'x86_64'" },
     { name = "packaging" },
     { name = "pyyaml" },
     { name = "requests" },
     { name = "tqdm" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/e1/d2/d6976de7542792fc077b498d64af64882b6d8bb40679284ec0bff77d5929/huggingface_hub-0.27.1.tar.gz", hash = "sha256:c004463ca870283909d715d20f066ebd6968c2207dae9393fdffb3c1d4d8f98b", size = 379407 }
+sdist = { url = "https://files.pythonhosted.org/packages/45/c9/bdbe19339f76d12985bc03572f330a01a93c04dffecaaea3061bdd7fb892/huggingface_hub-0.34.4.tar.gz", hash = "sha256:a4228daa6fb001be3f4f4bdaf9a0db00e1739235702848df00885c9b5742c85c", size = 459768, upload-time = "2025-08-08T09:14:52.365Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/39/7b/bb06b061991107cd8783f300adff3e7b7f284e330fd82f507f2a1417b11d/huggingface_hub-0.34.4-py3-none-any.whl", hash = "sha256:9b365d781739c93ff90c359844221beef048403f1bc1f1c123c191257c3c890a", size = 561452, upload-time = "2025-08-08T09:14:50.159Z" },
+]
+
+[[package]]
+name = "humanize"
+version = "4.13.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/98/1d/3062fcc89ee05a715c0b9bfe6490c00c576314f27ffee3a704122c6fd259/humanize-4.13.0.tar.gz", hash = "sha256:78f79e68f76f0b04d711c4e55d32bebef5be387148862cb1ef83d2b58e7935a0", size = 81884, upload-time = "2025-08-25T09:39:20.04Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6c/3f/50f6b25fafdcfb1c089187a328c95081abf882309afd86f4053951507cd1/huggingface_hub-0.27.1-py3-none-any.whl", hash = "sha256:1c5155ca7d60b60c2e2fc38cbb3ffb7f7c3adf48f824015b219af9061771daec", size = 450658 },
+    { url = "https://files.pythonhosted.org/packages/1e/c7/316e7ca04d26695ef0635dc81683d628350810eb8e9b2299fc08ba49f366/humanize-4.13.0-py3-none-any.whl", hash = "sha256:b810820b31891813b1673e8fec7f1ed3312061eab2f26e3fa192c393d11ed25f", size = 128869, upload-time = "2025-08-25T09:39:18.54Z" },
 ]
 
 [[package]]
 name = "identify"
-version = "2.6.5"
+version = "2.6.13"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/cf/92/69934b9ef3c31ca2470980423fda3d00f0460ddefdf30a67adf7f17e2e00/identify-2.6.5.tar.gz", hash = "sha256:c10b33f250e5bba374fae86fb57f3adcebf1161bce7cdf92031915fd480c13bc", size = 99213 }
+sdist = { url = "https://files.pythonhosted.org/packages/82/ca/ffbabe3635bb839aa36b3a893c91a9b0d368cb4d8073e03a12896970af82/identify-2.6.13.tar.gz", hash = "sha256:da8d6c828e773620e13bfa86ea601c5a5310ba4bcd65edf378198b56a1f9fb32", size = 99243, upload-time = "2025-08-09T19:35:00.6Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ec/fa/dce098f4cdf7621aa8f7b4f919ce545891f489482f0bfa5102f3eca8608b/identify-2.6.5-py2.py3-none-any.whl", hash = "sha256:14181a47091eb75b337af4c23078c9d09225cd4c48929f521f3bf16b09d02566", size = 99078 },
+    { url = "https://files.pythonhosted.org/packages/e7/ce/461b60a3ee109518c055953729bf9ed089a04db895d47e95444071dcdef2/identify-2.6.13-py2.py3-none-any.whl", hash = "sha256:60381139b3ae39447482ecc406944190f690d4a2997f2584062089848361b33b", size = 99153, upload-time = "2025-08-09T19:34:59.1Z" },
 ]
 
 [[package]]
 name = "idna"
 version = "3.10"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f1/70/7703c29685631f5a7590aa73f1f1d3fa9a380e654b86af429e0934a32f7d/idna-3.10.tar.gz", hash = "sha256:12f65c9b470abda6dc35cf8e63cc574b1c52b11df2c86030af0ac09b01b13ea9", size = 190490 }
+sdist = { url = "https://files.pythonhosted.org/packages/f1/70/7703c29685631f5a7590aa73f1f1d3fa9a380e654b86af429e0934a32f7d/idna-3.10.tar.gz", hash = "sha256:12f65c9b470abda6dc35cf8e63cc574b1c52b11df2c86030af0ac09b01b13ea9", size = 190490, upload-time = "2024-09-15T18:07:39.745Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/76/c6/c88e154df9c4e1a2a66ccf0005a88dfb2650c1dffb6f5ce603dfbd452ce3/idna-3.10-py3-none-any.whl", hash = "sha256:946d195a0d259cbba61165e88e65941f16e9b36ea6ddb97f00452bae8b1287d3", size = 70442 },
+    { url = "https://files.pythonhosted.org/packages/76/c6/c88e154df9c4e1a2a66ccf0005a88dfb2650c1dffb6f5ce603dfbd452ce3/idna-3.10-py3-none-any.whl", hash = "sha256:946d195a0d259cbba61165e88e65941f16e9b36ea6ddb97f00452bae8b1287d3", size = 70442, upload-time = "2024-09-15T18:07:37.964Z" },
 ]
 
 [[package]]
 name = "importlib-metadata"
-version = "8.5.0"
+version = "8.7.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "zipp", marker = "python_full_version < '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/cd/12/33e59336dca5be0c398a7482335911a33aa0e20776128f038019f1a95f1b/importlib_metadata-8.5.0.tar.gz", hash = "sha256:71522656f0abace1d072b9e5481a48f07c138e00f079c38c8f883823f9c26bd7", size = 55304 }
+sdist = { url = "https://files.pythonhosted.org/packages/76/66/650a33bd90f786193e4de4b3ad86ea60b53c89b669a5c7be931fac31cdb0/importlib_metadata-8.7.0.tar.gz", hash = "sha256:d13b81ad223b890aa16c5471f2ac3056cf76c5f10f82d6f9292f0b415f389000", size = 56641, upload-time = "2025-04-27T15:29:01.736Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a0/d9/a1e041c5e7caa9a05c925f4bdbdfb7f006d1f74996af53467bc394c97be7/importlib_metadata-8.5.0-py3-none-any.whl", hash = "sha256:45e54197d28b7a7f1559e60b95e7c567032b602131fbd588f1497f47880aa68b", size = 26514 },
+    { url = "https://files.pythonhosted.org/packages/20/b0/36bd937216ec521246249be3bf9855081de4c5e06a0c9b4219dbeda50373/importlib_metadata-8.7.0-py3-none-any.whl", hash = "sha256:e5dd1551894c77868a30651cef00984d50e1002d06942a7101d34870c5f02afd", size = 27656, upload-time = "2025-04-27T15:29:00.214Z" },
 ]
 
 [[package]]
-name = "iniconfig"
-version = "2.0.0"
+name = "importlib-resources"
+version = "6.5.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/d7/4b/cbd8e699e64a6f16ca3a8220661b5f83792b3017d0f79807cb8708d33913/iniconfig-2.0.0.tar.gz", hash = "sha256:2d91e135bf72d31a410b17c16da610a82cb55f6b0477d1a902134b24a455b8b3", size = 4646 }
+dependencies = [
+    { name = "zipp", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/cf/8c/f834fbf984f691b4f7ff60f50b514cc3de5cc08abfc3295564dd89c5e2e7/importlib_resources-6.5.2.tar.gz", hash = "sha256:185f87adef5bcc288449d98fb4fba07cea78bc036455dd44c5fc4a2fe78fed2c", size = 44693, upload-time = "2025-01-03T18:51:56.698Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ef/a6/62565a6e1cf69e10f5727360368e451d4b7f58beeac6173dc9db836a5b46/iniconfig-2.0.0-py3-none-any.whl", hash = "sha256:b6a85871a79d2e3b22d2d1b94ac2824226a63c6b741c88f7ae975f18b6778374", size = 5892 },
+    { url = "https://files.pythonhosted.org/packages/a4/ed/1f1afb2e9e7f38a545d628f864d562a5ae64fe6f7a10e28ffb9b185b4e89/importlib_resources-6.5.2-py3-none-any.whl", hash = "sha256:789cfdc3ed28c78b67a06acb8126751ced69a3d5f79c095a98298cd8a760ccec", size = 37461, upload-time = "2025-01-03T18:51:54.306Z" },
 ]
 
 [[package]]
-name = "intel-openmp"
-version = "2021.4.0"
+name = "iniconfig"
+version = "2.1.0"
 source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f2/97/ebf4da567aa6827c909642694d71c9fcf53e5b504f2d96afea02718862f3/iniconfig-2.1.0.tar.gz", hash = "sha256:3abbd2e30b36733fee78f9c7f7308f2d0050e88f0087fd25c2645f63c773e1c7", size = 4793, upload-time = "2025-03-19T20:09:59.721Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/45/18/527f247d673ff84c38e0b353b6901539b99e83066cd505be42ad341ab16d/intel_openmp-2021.4.0-py2.py3-none-win32.whl", hash = "sha256:6e863d8fd3d7e8ef389d52cf97a50fe2afe1a19247e8c0d168ce021546f96fc9", size = 1860605 },
-    { url = "https://files.pythonhosted.org/packages/6f/21/b590c0cc3888b24f2ac9898c41d852d7454a1695fbad34bee85dba6dc408/intel_openmp-2021.4.0-py2.py3-none-win_amd64.whl", hash = "sha256:eef4c8bcc8acefd7f5cd3b9384dbf73d59e2c99fc56545712ded913f43c4a94f", size = 3516906 },
+    { url = "https://files.pythonhosted.org/packages/2c/e1/e6716421ea10d38022b952c159d5161ca1193197fb744506875fbb87ea7b/iniconfig-2.1.0-py3-none-any.whl", hash = "sha256:9deba5723312380e77435581c6bf4935c94cbfab9b1ed33ef8d238ea168eb760", size = 6050, upload-time = "2025-03-19T20:10:01.071Z" },
 ]
 
 [[package]]
-name = "interegular"
-version = "0.3.3"
+name = "iso3166"
+version = "2.1.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/dc/9d/8b6dde58a028a3962ce17e84d5fe73758df61378e00ef8ac3d85da34b0ff/interegular-0.3.3.tar.gz", hash = "sha256:d9b697b21b34884711399ba0f0376914b81899ce670032486d0d048344a76600", size = 24705 }
+sdist = { url = "https://files.pythonhosted.org/packages/5c/11/b5023c736a185a88ebd0d38646af6f4d1b4c9b91f2ca84e08e5d2bc7ac3c/iso3166-2.1.1.tar.gz", hash = "sha256:fcd551b8dda66b44e9f9e6d6bbbee3a1145a22447c0a556e5d0fb1ad1e491719", size = 12807, upload-time = "2022-07-12T04:07:57.294Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c4/01/72d6472f80651673716d1deda2a5bbb633e563ecf94f4479da5519d69d25/interegular-0.3.3-py37-none-any.whl", hash = "sha256:b0c07007d48c89d6d19f7204972d369b2a77222722e126b6aa63aa721dc3b19c", size = 23635 },
+    { url = "https://files.pythonhosted.org/packages/08/d0/bf18725b8d47f37858ff801f8e4d40c6982730a899725bdb6ded62199954/iso3166-2.1.1-py3-none-any.whl", hash = "sha256:263660b36f8471c42acd1ff673d28a3715edbce7d24b1550d0cf010f6816c47f", size = 9829, upload-time = "2022-07-12T04:07:55.54Z" },
 ]
 
 [[package]]
@@ -970,202 +1153,229 @@ name = "jax"
 version = "0.4.30"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
-    "python_full_version < '3.9.17'",
+    "python_full_version < '3.10'",
 ]
 dependencies = [
     { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
     { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", marker = "python_full_version < '3.10'" },
+    { name = "ml-dtypes", marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "opt-einsum", marker = "python_full_version < '3.10'" },
     { name = "scipy", version = "1.13.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/15/41/d6dbafc31d6bd93eeec2e1c709adfa454266e83714ebeeed9de52a6ad881/jax-0.4.30.tar.gz", hash = "sha256:94d74b5b2db0d80672b61d83f1f63ebf99d2ab7398ec12b2ca0c9d1e97afe577", size = 1715462 }
+sdist = { url = "https://files.pythonhosted.org/packages/15/41/d6dbafc31d6bd93eeec2e1c709adfa454266e83714ebeeed9de52a6ad881/jax-0.4.30.tar.gz", hash = "sha256:94d74b5b2db0d80672b61d83f1f63ebf99d2ab7398ec12b2ca0c9d1e97afe577", size = 1715462, upload-time = "2024-06-18T14:47:17.125Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/fd/f2/9dbb75de3058acfd1600cf0839bcce7ea391148c9d2b4fa5f5666e66f09e/jax-0.4.30-py3-none-any.whl", hash = "sha256:289b30ae03b52f7f4baf6ef082a9f4e3e29c1080e22d13512c5ecf02d5f1a55b", size = 2009197 },
+    { url = "https://files.pythonhosted.org/packages/fd/f2/9dbb75de3058acfd1600cf0839bcce7ea391148c9d2b4fa5f5666e66f09e/jax-0.4.30-py3-none-any.whl", hash = "sha256:289b30ae03b52f7f4baf6ef082a9f4e3e29c1080e22d13512c5ecf02d5f1a55b", size = 2009197, upload-time = "2024-06-18T14:47:10.026Z" },
 ]
 
 [[package]]
 name = "jax"
-version = "0.5.0"
+version = "0.6.2"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.12.*'",
-    "python_full_version == '3.11.*'",
     "python_full_version == '3.10.*'",
 ]
 dependencies = [
-    { name = "jaxlib", version = "0.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.13' or (python_full_version >= '3.10' and python_full_version < '3.12')" },
-    { name = "ml-dtypes", version = "0.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.12.*'" },
-    { name = "numpy", marker = "python_full_version >= '3.10'" },
-    { name = "opt-einsum", marker = "python_full_version >= '3.10'" },
-    { name = "scipy", version = "1.15.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "jaxlib", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "ml-dtypes", marker = "python_full_version == '3.10.*'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "opt-einsum", marker = "python_full_version == '3.10.*'" },
+    { name = "scipy", version = "1.15.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/4a/cb/22d62b26284f08e62d6eb64603d3b010004cfdb7a97ce6cca5c6cf86edab/jax-0.5.0.tar.gz", hash = "sha256:49df70bf293a345a7fb519f71193506d37a024c4f850b358042eb32d502c81c8", size = 1959707 }
+sdist = { url = "https://files.pythonhosted.org/packages/cf/1e/267f59c8fb7f143c3f778c76cb7ef1389db3fd7e4540f04b9f42ca90764d/jax-0.6.2.tar.gz", hash = "sha256:a437d29038cbc8300334119692744704ca7941490867b9665406b7f90665cd96", size = 2334091, upload-time = "2025-06-17T23:10:27.186Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f4/58/cc0721a1030fcbab0984beea0bf3c4610ec103f738423cdfa9c4ceb40598/jax-0.5.0-py3-none-any.whl", hash = "sha256:b3907aa87ae2c340b39cdbf80c07a74550369cafcaf7398fb60ba58d167345ab", size = 2270365 },
+    { url = "https://files.pythonhosted.org/packages/31/a8/97ef0cbb7a17143ace2643d600a7b80d6705b2266fc31078229e406bdef2/jax-0.6.2-py3-none-any.whl", hash = "sha256:bb24a82dc60ccf704dcaf6dbd07d04957f68a6c686db19630dd75260d1fb788c", size = 2722396, upload-time = "2025-06-17T23:10:25.293Z" },
 ]
 
 [[package]]
-name = "jaxlib"
-version = "0.4.30"
+name = "jax"
+version = "0.7.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
-    "python_full_version < '3.9.17'",
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", marker = "python_full_version < '3.10'" },
-    { name = "scipy", version = "1.13.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "ml-dtypes", marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "opt-einsum", marker = "python_full_version >= '3.11'" },
+    { name = "scipy", version = "1.16.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
+sdist = { url = "https://files.pythonhosted.org/packages/bc/e8/b393ee314d3b042bd66b986d38e52f4e6046590399d916381265c20467d3/jax-0.7.1.tar.gz", hash = "sha256:118f56338c503361d2791f069d24339d8d44a8db442ed851d2e591222fb7a56d", size = 2428411, upload-time = "2025-08-20T15:55:46.098Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f3/18/ff7f2f6d6195853ed55c5b5d835f5c8c3c8b190c7221cb04a0cb81f5db10/jaxlib-0.4.30-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:c40856e28f300938c6824ab1a615166193d6997dec946578823f6d402ad454e5", size = 83542097 },
-    { url = "https://files.pythonhosted.org/packages/d4/c0/ff65503ecfed3aee11e4abe4c4e9e8a3513f072e0b595f8247b9989d1510/jaxlib-0.4.30-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:4bdfda6a3c7a2b0cc0a7131009eb279e98ca4a6f25679fabb5302dd135a5e349", size = 66694495 },
-    { url = "https://files.pythonhosted.org/packages/b9/d7/82df748a31a1cfbd531a12979ea846d6b676d4adfa1e91114b848665b2aa/jaxlib-0.4.30-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:28e032c9b394ab7624d89b0d9d3bbcf4d1d71694fe8b3e09d3fe64122eda7b0c", size = 67781242 },
-    { url = "https://files.pythonhosted.org/packages/4a/ca/561aabed63007bb2621a62f0d816aa2f68cfe947859c8b4e61519940344b/jaxlib-0.4.30-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:d83f36ef42a403bbf7c7f2da526b34ba286988e170f4df5e58b3bb735417868c", size = 79640266 },
-    { url = "https://files.pythonhosted.org/packages/b0/90/8e5347eda95d3cb695cd5ebb82f850fa7866078a6a7a0568549e34125a82/jaxlib-0.4.30-cp310-cp310-win_amd64.whl", hash = "sha256:a56678b28f96b524ded6da8ef4b38e72a532356d139cfd434da804abf4234e14", size = 51945307 },
-    { url = "https://files.pythonhosted.org/packages/33/2d/b6078f5d173d3087d32b1b49e5f65d406985fb3894ff1d21905972b9c89d/jaxlib-0.4.30-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:bfb5d85b69c29c3c6e8051a0ea715ac1e532d6e54494c8d9c3813dcc00deac30", size = 83539315 },
-    { url = "https://files.pythonhosted.org/packages/12/95/399da9204c3b13696baefb93468402f3389416b0caecfd9126aa94742bf2/jaxlib-0.4.30-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:974998cd8a78550402e6c09935c1f8d850cad9cc19ccd7488bde45b6f7f99c12", size = 66690971 },
-    { url = "https://files.pythonhosted.org/packages/a4/f8/b85a46cb0cc4bc228cea4366b0d15caf42656c6d43cf8c91d90f7399aa4d/jaxlib-0.4.30-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:e93eb0646b41ba213252b51b0b69096b9cd1d81a35ea85c9d06663b5d11efe45", size = 67780747 },
-    { url = "https://files.pythonhosted.org/packages/a6/a3/951da3d1487b2f8995a2a14cc7e9496c9a7c93aa1f1d0b33e833e24dee92/jaxlib-0.4.30-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:16b2ab18ea90d2e15941bcf45de37afc2f289a029129c88c8d7aba0404dd0043", size = 79640352 },
-    { url = "https://files.pythonhosted.org/packages/bb/1a/8f45ea28a5ca67e4d23ebd70fc78ea94be6fa20323f983c7607c32c6f9a5/jaxlib-0.4.30-cp311-cp311-win_amd64.whl", hash = "sha256:3a2e2c11c179f8851a72249ba1ae40ae817dfaee9877d23b3b8f7c6b7a012f76", size = 51943960 },
-    { url = "https://files.pythonhosted.org/packages/19/40/ae943d3c1fc8b50947aebbaa3bad2842759e43bc9fc91e1758c1c20a81ab/jaxlib-0.4.30-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:7704db5962b32a2be3cc07185433cbbcc94ed90ee50c84021a3f8a1ecfd66ee3", size = 83587124 },
-    { url = "https://files.pythonhosted.org/packages/c6/e3/97f8edff6f64245a500415be021869522b235e8b38cd930d358b91243583/jaxlib-0.4.30-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:57090d33477fd0f0c99dc686274882ea75c44c7d712ae42dd2460b10f896131d", size = 66724768 },
-    { url = "https://files.pythonhosted.org/packages/4c/c7/ee1f48f8daa409d0ed039e0d8b5ae1a447e53db3acb2ff06239828ad96d5/jaxlib-0.4.30-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:0a3850e76278038e21685975a62b622bcf3708485f13125757a0561ee4512940", size = 67800348 },
-    { url = "https://files.pythonhosted.org/packages/f2/fa/a2dddea0d6965b8e433bb99aeedbe5c8a9b47110c1c4f197a7b6239daf44/jaxlib-0.4.30-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:c58a8071c4e00898282118169f6a5a97eb15a79c2897858f3a732b17891c99ab", size = 79674030 },
-    { url = "https://files.pythonhosted.org/packages/db/31/3500633d61b20b882a0fbcf8100013195c31b51f71249b0b38737851fc9a/jaxlib-0.4.30-cp312-cp312-win_amd64.whl", hash = "sha256:b7079a5b1ab6864a7d4f2afaa963841451186d22c90f39719a3ff85735ce3915", size = 51965689 },
-    { url = "https://files.pythonhosted.org/packages/46/12/9de601dbae3c66666eeaaf5a28683d947909c046880baef390b7cd1d4b1d/jaxlib-0.4.30-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:ea3a00005faafbe3c18b178d3b534208b3b4027b2be6230227e7b87ce399fc29", size = 83544602 },
-    { url = "https://files.pythonhosted.org/packages/f3/1d/2d417a1445d5e696bb44d564c7519d4a6761db4d3e31712620c510ed0127/jaxlib-0.4.30-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3d31e01191ce8052bd611aaf16ff967d8d0ec0b63f1ea4b199020cecb248d667", size = 66695975 },
-    { url = "https://files.pythonhosted.org/packages/e4/f9/e29370046f4648bd464df7eceaebbbaefd091cc88c77da4a6e3a5f1a00d7/jaxlib-0.4.30-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:11602d5556e8baa2f16314c36518e9be4dfae0c2c256a361403fb29dc9dc79a4", size = 67784388 },
-    { url = "https://files.pythonhosted.org/packages/07/3b/a596036325666624ca084df554636fb3777e78e9386b52476d96fa14394e/jaxlib-0.4.30-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:f74a6b0e09df4b5e2ee399ebb9f0e01190e26e84ccb0a758fadb516415c07f18", size = 79643370 },
-    { url = "https://files.pythonhosted.org/packages/8a/a3/7342ceb02e49803af9a42ab3ad9b6c272cf7b2a83163e3a06859360012d5/jaxlib-0.4.30-cp39-cp39-win_amd64.whl", hash = "sha256:54987e97a22db70f3829b437b9329e4799d653634bacc8b398554d3b90c76b2a", size = 51946140 },
+    { url = "https://files.pythonhosted.org/packages/83/81/793d78c91b0546b3b1f08e55fdd97437174171cd7d70e46098f1a4d94b7b/jax-0.7.1-py3-none-any.whl", hash = "sha256:056e576e0e58465506125699f48111ac8891cce4c9ebf034704c42b219dfd4a6", size = 2827341, upload-time = "2025-08-20T15:55:44.576Z" },
 ]
 
 [[package]]
 name = "jaxlib"
-version = "0.5.0"
+version = "0.4.30"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "ml-dtypes", marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "scipy", version = "1.13.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f3/18/ff7f2f6d6195853ed55c5b5d835f5c8c3c8b190c7221cb04a0cb81f5db10/jaxlib-0.4.30-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:c40856e28f300938c6824ab1a615166193d6997dec946578823f6d402ad454e5", size = 83542097, upload-time = "2024-06-18T14:47:34.758Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/c0/ff65503ecfed3aee11e4abe4c4e9e8a3513f072e0b595f8247b9989d1510/jaxlib-0.4.30-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:4bdfda6a3c7a2b0cc0a7131009eb279e98ca4a6f25679fabb5302dd135a5e349", size = 66694495, upload-time = "2024-06-18T14:47:55.156Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/d7/82df748a31a1cfbd531a12979ea846d6b676d4adfa1e91114b848665b2aa/jaxlib-0.4.30-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:28e032c9b394ab7624d89b0d9d3bbcf4d1d71694fe8b3e09d3fe64122eda7b0c", size = 67781242, upload-time = "2024-06-18T14:48:08.546Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/ca/561aabed63007bb2621a62f0d816aa2f68cfe947859c8b4e61519940344b/jaxlib-0.4.30-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:d83f36ef42a403bbf7c7f2da526b34ba286988e170f4df5e58b3bb735417868c", size = 79640266, upload-time = "2024-06-18T14:48:26.78Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/90/8e5347eda95d3cb695cd5ebb82f850fa7866078a6a7a0568549e34125a82/jaxlib-0.4.30-cp310-cp310-win_amd64.whl", hash = "sha256:a56678b28f96b524ded6da8ef4b38e72a532356d139cfd434da804abf4234e14", size = 51945307, upload-time = "2024-06-18T14:48:46.909Z" },
+    { url = "https://files.pythonhosted.org/packages/33/2d/b6078f5d173d3087d32b1b49e5f65d406985fb3894ff1d21905972b9c89d/jaxlib-0.4.30-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:bfb5d85b69c29c3c6e8051a0ea715ac1e532d6e54494c8d9c3813dcc00deac30", size = 83539315, upload-time = "2024-06-18T14:49:01.814Z" },
+    { url = "https://files.pythonhosted.org/packages/12/95/399da9204c3b13696baefb93468402f3389416b0caecfd9126aa94742bf2/jaxlib-0.4.30-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:974998cd8a78550402e6c09935c1f8d850cad9cc19ccd7488bde45b6f7f99c12", size = 66690971, upload-time = "2024-06-18T14:49:20.825Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/f8/b85a46cb0cc4bc228cea4366b0d15caf42656c6d43cf8c91d90f7399aa4d/jaxlib-0.4.30-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:e93eb0646b41ba213252b51b0b69096b9cd1d81a35ea85c9d06663b5d11efe45", size = 67780747, upload-time = "2024-06-18T14:49:35.024Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/a3/951da3d1487b2f8995a2a14cc7e9496c9a7c93aa1f1d0b33e833e24dee92/jaxlib-0.4.30-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:16b2ab18ea90d2e15941bcf45de37afc2f289a029129c88c8d7aba0404dd0043", size = 79640352, upload-time = "2024-06-18T14:49:47.36Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/1a/8f45ea28a5ca67e4d23ebd70fc78ea94be6fa20323f983c7607c32c6f9a5/jaxlib-0.4.30-cp311-cp311-win_amd64.whl", hash = "sha256:3a2e2c11c179f8851a72249ba1ae40ae817dfaee9877d23b3b8f7c6b7a012f76", size = 51943960, upload-time = "2024-06-18T14:50:02.285Z" },
+    { url = "https://files.pythonhosted.org/packages/19/40/ae943d3c1fc8b50947aebbaa3bad2842759e43bc9fc91e1758c1c20a81ab/jaxlib-0.4.30-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:7704db5962b32a2be3cc07185433cbbcc94ed90ee50c84021a3f8a1ecfd66ee3", size = 83587124, upload-time = "2024-06-18T14:50:13.699Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/e3/97f8edff6f64245a500415be021869522b235e8b38cd930d358b91243583/jaxlib-0.4.30-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:57090d33477fd0f0c99dc686274882ea75c44c7d712ae42dd2460b10f896131d", size = 66724768, upload-time = "2024-06-18T14:50:28.951Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/c7/ee1f48f8daa409d0ed039e0d8b5ae1a447e53db3acb2ff06239828ad96d5/jaxlib-0.4.30-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:0a3850e76278038e21685975a62b622bcf3708485f13125757a0561ee4512940", size = 67800348, upload-time = "2024-06-18T14:50:39.89Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/fa/a2dddea0d6965b8e433bb99aeedbe5c8a9b47110c1c4f197a7b6239daf44/jaxlib-0.4.30-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:c58a8071c4e00898282118169f6a5a97eb15a79c2897858f3a732b17891c99ab", size = 79674030, upload-time = "2024-06-18T14:51:05.293Z" },
+    { url = "https://files.pythonhosted.org/packages/db/31/3500633d61b20b882a0fbcf8100013195c31b51f71249b0b38737851fc9a/jaxlib-0.4.30-cp312-cp312-win_amd64.whl", hash = "sha256:b7079a5b1ab6864a7d4f2afaa963841451186d22c90f39719a3ff85735ce3915", size = 51965689, upload-time = "2024-06-18T14:51:19.206Z" },
+    { url = "https://files.pythonhosted.org/packages/46/12/9de601dbae3c66666eeaaf5a28683d947909c046880baef390b7cd1d4b1d/jaxlib-0.4.30-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:ea3a00005faafbe3c18b178d3b534208b3b4027b2be6230227e7b87ce399fc29", size = 83544602, upload-time = "2024-06-18T14:51:32.669Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/1d/2d417a1445d5e696bb44d564c7519d4a6761db4d3e31712620c510ed0127/jaxlib-0.4.30-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3d31e01191ce8052bd611aaf16ff967d8d0ec0b63f1ea4b199020cecb248d667", size = 66695975, upload-time = "2024-06-18T14:51:48.912Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/f9/e29370046f4648bd464df7eceaebbbaefd091cc88c77da4a6e3a5f1a00d7/jaxlib-0.4.30-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:11602d5556e8baa2f16314c36518e9be4dfae0c2c256a361403fb29dc9dc79a4", size = 67784388, upload-time = "2024-06-18T14:52:03.082Z" },
+    { url = "https://files.pythonhosted.org/packages/07/3b/a596036325666624ca084df554636fb3777e78e9386b52476d96fa14394e/jaxlib-0.4.30-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:f74a6b0e09df4b5e2ee399ebb9f0e01190e26e84ccb0a758fadb516415c07f18", size = 79643370, upload-time = "2024-06-18T14:52:16.711Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/a3/7342ceb02e49803af9a42ab3ad9b6c272cf7b2a83163e3a06859360012d5/jaxlib-0.4.30-cp39-cp39-win_amd64.whl", hash = "sha256:54987e97a22db70f3829b437b9329e4799d653634bacc8b398554d3b90c76b2a", size = 51946140, upload-time = "2024-06-18T14:52:31.677Z" },
+]
+
+[[package]]
+name = "jaxlib"
+version = "0.6.2"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.12.*'",
-    "python_full_version == '3.11.*'",
     "python_full_version == '3.10.*'",
 ]
 dependencies = [
-    { name = "ml-dtypes", version = "0.4.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.13' or (python_full_version >= '3.10' and python_full_version < '3.12')" },
-    { name = "ml-dtypes", version = "0.5.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.12.*'" },
-    { name = "numpy", marker = "python_full_version >= '3.10'" },
-    { name = "scipy", version = "1.15.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "ml-dtypes", marker = "python_full_version == '3.10.*'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "scipy", version = "1.15.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/15/c5/41598634c99cbebba46e6777286fb76abc449d33d50aeae5d36128ca8803/jaxlib-0.6.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:da4601b2b5dc8c23d6afb293eacfb9aec4e1d1871cb2f29c5a151d103e73b0f8", size = 54298019, upload-time = "2025-06-17T23:10:36.916Z" },
+    { url = "https://files.pythonhosted.org/packages/81/af/db07d746cd5867d5967528e7811da53374e94f64e80a890d6a5a4b95b130/jaxlib-0.6.2-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:4205d098ce8efb5f7fe2fe5098bae6036094dc8d8829f5e0e0d7a9b155326336", size = 79440052, upload-time = "2025-06-17T23:10:41.282Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/d8/b7ae9e819c62c1854dbc2c70540a5c041173fbc8bec5e78ab7fd615a4aee/jaxlib-0.6.2-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:c087a0eb6fb7f6f8f54d56f4730328dfde5040dd3b5ddfa810e7c28ea7102b42", size = 89917034, upload-time = "2025-06-17T23:10:45.897Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/e5/87e91bc70569ac5c3e3449eefcaf47986e892f10cfe1d5e5720dceae3068/jaxlib-0.6.2-cp310-cp310-win_amd64.whl", hash = "sha256:153eaa51f778b60851720729d4f461a91edd9ba3932f6f3bc598d4413870038b", size = 57896337, upload-time = "2025-06-17T23:10:50.179Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/ee/6899b0aed36a4acc51319465ddd83c7c300a062a9e236cceee00984ffe0b/jaxlib-0.6.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a208ff61c58128d306bb4e5ad0858bd2b0960f2c1c10ad42c548f74a60c0020e", size = 54300346, upload-time = "2025-06-17T23:10:54.591Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/03/34bb6b346609079a71942cfbf507892e3c877a06a430a0df8429c455cebc/jaxlib-0.6.2-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:11eae7e05bc5a79875da36324afb9eddd4baeaef2a0386caf6d4f3720b9aef28", size = 79438425, upload-time = "2025-06-17T23:10:58.356Z" },
+    { url = "https://files.pythonhosted.org/packages/80/02/49b05cbab519ffd3cb79586336451fbbf8b6523f67128a794acc9f179000/jaxlib-0.6.2-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:335d7e3515ce78b52a410136f46aa4a7ea14d0e7d640f34e1e137409554ad0ac", size = 89920354, upload-time = "2025-06-17T23:11:03.086Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/7a/93b28d9452b46c15fc28dd65405672fc8a158b35d46beabaa0fe9631afb0/jaxlib-0.6.2-cp311-cp311-win_amd64.whl", hash = "sha256:c6815509997d6b05e5c9daa7994b9ad473ce3e8c8a17bdbbcacc3c744f76f7a0", size = 57895707, upload-time = "2025-06-17T23:11:07.074Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/db/05e702d2534e87abf606b1067b46a273b120e6adc7d459696e3ce7399317/jaxlib-0.6.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:34d8a684a8be949dd87dd4acc97101b4106a0dc9ad151ec891da072319a57b99", size = 54301644, upload-time = "2025-06-17T23:11:10.977Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/8a/b0a96887b97a25d45ae2c30e4acecd2f95acd074c18ec737dda8c5cc7016/jaxlib-0.6.2-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:87ec2dc9c3ed9ab936eec8535160c5fbd2c849948559f1c5daa75f63fabe5942", size = 79439161, upload-time = "2025-06-17T23:11:14.822Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/e8/71c2555431edb5dd115cf86a7b599aa7e1be26728d89ae59aa11251d299c/jaxlib-0.6.2-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:f1dd09b481a93c1d4c750013f467f74194493ba7bd29fcd4d1cec16e3a214f65", size = 89942952, upload-time = "2025-06-17T23:11:19.181Z" },
+    { url = "https://files.pythonhosted.org/packages/de/3a/06849113c844b86d20174df54735c84202ccf82cbd36d805f478c834418b/jaxlib-0.6.2-cp312-cp312-win_amd64.whl", hash = "sha256:921dbd4db214eba19a29ba9f2450d880e08b2b2c7b968f28cc89da3e62366af4", size = 57919603, upload-time = "2025-06-17T23:11:23.207Z" },
+]
+
+[[package]]
+name = "jaxlib"
+version = "0.7.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+]
+dependencies = [
+    { name = "ml-dtypes", marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "scipy", version = "1.16.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c8/41/3e4ac64df72c4da126df3fd66a2214025a46b6263f7be266728e7b8e473e/jaxlib-0.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:1b8a6c4345f137f387650de2dbc488c20251b7412b55dd648e1a4f13bcf507fb", size = 79248968 },
-    { url = "https://files.pythonhosted.org/packages/1e/5f/2a16e61f1d54ae5f55fbf3cb3e22ef5bb01bf9d7d6474e0d34fedba19c4d/jaxlib-0.5.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:5b2efe3dfebf18a84c451d3803ac884ee242021c1113b279c13f4bbc378c3dc0", size = 93181077 },
-    { url = "https://files.pythonhosted.org/packages/08/c3/573e2f01b99f1247e8fbe1aa46b95a0faa68ef208f9a8e8ef775d607b3e6/jaxlib-0.5.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:74440b632107336400d4f97a16481d767f13ea914c53ba14e544c6fda54819b3", size = 101969119 },
-    { url = "https://files.pythonhosted.org/packages/6e/38/512f61ea13da41ca47f2411d7c05af0cf74a37f225e16725ed0e6fb58893/jaxlib-0.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:53478a28eee6c2ef01759b05a9491702daef9268c3ed013d6f8e2e5f5cae0887", size = 63883394 },
-    { url = "https://files.pythonhosted.org/packages/92/4b/8875870ff52ad3fbea876c905228f691f05c8dc8556b226cbfaf0fba7f62/jaxlib-0.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6cd762ed1623132499fa701c4203446102e0a9c82ca23194b87288f746d12a29", size = 79242870 },
-    { url = "https://files.pythonhosted.org/packages/a0/0f/00cdfa411d7218e4696c10c5867f7d3c396219adbcaeb02e95108ca802de/jaxlib-0.5.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:63088dbfaa85bb56cd521a925a3472fd7328b18ec93c2d8ffa85af331095c995", size = 93181807 },
-    { url = "https://files.pythonhosted.org/packages/58/8e/a5c29db03d5a93b0326e297b556d0e0a9805e9c9c1ae5f82f69557273faa/jaxlib-0.5.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:09113ef1582ba34d7cbc440fedb318f4855b59b776711a8aba2473c9727d3025", size = 101969212 },
-    { url = "https://files.pythonhosted.org/packages/70/86/ceae20e4f37fa07f1cc95551cc0f49170d0db46d2e82fdf511d26bffd801/jaxlib-0.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:78289fc3ddc1e4e9510de2536a6375df9fe1c50de0ac60826c286b7a5c5090fe", size = 63881994 },
-    { url = "https://files.pythonhosted.org/packages/57/d6/d971b40cb156e0637aa3c1522a1e803b641142e9a8f3ade6a574711bb073/jaxlib-0.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:73e335715760c56e635109d61426435a5d7f46f3363a115daea09427d5cd0efd", size = 79246087 },
-    { url = "https://files.pythonhosted.org/packages/41/2e/ba9770330077c3e4082cd0353e6a61419f79bff3e2197f904ce70167b9ad/jaxlib-0.5.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:4b4b01afb0ddec96c08356bff2bb685ddbe97fdffe4ed6e2d834b30aba972f22", size = 93179593 },
-    { url = "https://files.pythonhosted.org/packages/66/e9/211ba3e46ec22c722c4d61a739cfccf79b0618006d6f5fa53eb4eb93ed6d/jaxlib-0.5.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:f980c733e98c998a8da87c9a8cc61b6726d0be667a58bd664c1d717b4b4eae75", size = 101984785 },
-    { url = "https://files.pythonhosted.org/packages/2d/cb/11bb92324afb6ba678f388e10b78d6b02196bc8887eb5aa0d85ce398edf9/jaxlib-0.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:5baedbeeb60fa493c7528783254f04c6e986a2826266b198ed37e9336af2ef8c", size = 63899871 },
-    { url = "https://files.pythonhosted.org/packages/22/ac/e400473e6a2f405fd6e4dc40a713bb9a3868a3f76a8ffc5eb66f6e686002/jaxlib-0.5.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:ed18ea7161d03aa8fd4d1b55494882f21420efdfea68e5f298c4aebcf2ac3f34", size = 79245359 },
-    { url = "https://files.pythonhosted.org/packages/44/2d/c210abf4a9b2ce2e0858fcd3567c8773a739114e37d751af6c228901af57/jaxlib-0.5.0-cp313-cp313-manylinux2014_aarch64.whl", hash = "sha256:7d9b17a7ea19355d45ecdb2ff0db5d707a86f0c5a862d94b89b4568d6c45311a", size = 93180025 },
-    { url = "https://files.pythonhosted.org/packages/30/f8/316f7b4797c5eb50c6d70e461724a7cbe08b4505ca4da1bfd260c135895a/jaxlib-0.5.0-cp313-cp313-manylinux2014_x86_64.whl", hash = "sha256:11eef01d37c0f1c5306265b76f207f1002d13480ded2e31fd63ec76912c93ca2", size = 101982281 },
-    { url = "https://files.pythonhosted.org/packages/4c/f2/cfa012a0417c9b13b44c8e1d3ebf5fd04e8bb738b5c93e20c9fc97919880/jaxlib-0.5.0-cp313-cp313-win_amd64.whl", hash = "sha256:61b4d26cd6a0c49ba0b1e4340c7d29198913ee2dc70b65ee90752717d22305bb", size = 63900219 },
+    { url = "https://files.pythonhosted.org/packages/8b/af/5058d545e95f99a54289648f5430cc3c23263dd70a1391e7491f24ed328d/jaxlib-0.7.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3f32c3e4c167b7327c342e82d3df84079714ea0b43718be871d039999670b3c9", size = 57686934, upload-time = "2025-08-20T15:55:58.989Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/77/ef7f6cd03e699da7d9755f88741c29b3015654473fc9d5f906da19edcb47/jaxlib-0.7.1-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:9fb189c3b39470c4394ffcb18b71e47cffc5bf85e8fcb1e33692686b0c3e04dd", size = 85134885, upload-time = "2025-08-20T15:56:03.484Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/72/304018d46703f337787f010735f70d17212f86778fcba8bb5cf678f8e460/jaxlib-0.7.1-cp311-cp311-manylinux_2_27_x86_64.whl", hash = "sha256:eaf5f68f53bf4dcb93b6512538547667625588e4f3ccaeef048788fd18d8c0d5", size = 81147868, upload-time = "2025-08-20T15:56:07.214Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/b7/0f0df407518691099d659ba6e19db01320dfb58e49d80594eaddd57d77c1/jaxlib-0.7.1-cp311-cp311-win_amd64.whl", hash = "sha256:ab4510fbaeafac6c794ab335f23e71200d824c48f6a0ab20553db8deab8805c5", size = 61185342, upload-time = "2025-08-20T15:56:10.452Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/1f/10543d7a3f7e76dd4bbdc77134890ac2f41bc8570c565961464f6320009b/jaxlib-0.7.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:127c07c727703e5d59f84f655169bec849f4422e52f8546349cecc30a8a13e1d", size = 57682851, upload-time = "2025-08-20T15:56:13.395Z" },
+    { url = "https://files.pythonhosted.org/packages/de/4d/76ee71959311fe3da9951aa6f55af8f98eb3572bb322f5a7c89faf7ab933/jaxlib-0.7.1-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:f0f1f52956b8c2518ab000a4d3d8c21be777e1d47f926ba03640e391061a41ee", size = 85133707, upload-time = "2025-08-20T15:56:16.908Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/50/e37d02e250f5feb755112ec95b1c012a36d48a99209277267037d100f630/jaxlib-0.7.1-cp312-cp312-manylinux_2_27_x86_64.whl", hash = "sha256:74abd3135797f82440dd3711a35cba16c430d1bba65474b85bb70e41733a52e9", size = 81156916, upload-time = "2025-08-20T15:56:20.41Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/97/c6c28dfe57cccffd85512615416024b52dd327d78270204caba9311e71f1/jaxlib-0.7.1-cp312-cp312-win_amd64.whl", hash = "sha256:c4023863b14f280516f24ecb7539b4300a3236ea81ed69ad82595beceed1ba1f", size = 61212445, upload-time = "2025-08-20T15:56:23.929Z" },
 ]
 
 [[package]]
 name = "jinja2"
-version = "3.1.5"
+version = "3.1.6"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "markupsafe" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/af/92/b3130cbbf5591acf9ade8708c365f3238046ac7cb8ccba6e81abccb0ccff/jinja2-3.1.5.tar.gz", hash = "sha256:8fefff8dc3034e27bb80d67c671eb8a9bc424c0ef4c0826edbff304cceff43bb", size = 244674 }
+sdist = { url = "https://files.pythonhosted.org/packages/df/bf/f7da0350254c0ed7c72f3e33cef02e048281fec7ecec5f032d4aac52226b/jinja2-3.1.6.tar.gz", hash = "sha256:0137fb05990d35f1275a587e9aee6d56da821fc83491a0fb838183be43f66d6d", size = 245115, upload-time = "2025-03-05T20:05:02.478Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/bd/0f/2ba5fbcd631e3e88689309dbe978c5769e883e4b84ebfe7da30b43275c5a/jinja2-3.1.5-py3-none-any.whl", hash = "sha256:aba0f4dc9ed8013c424088f68a5c226f7d6097ed89b246d7749c2ec4175c6adb", size = 134596 },
+    { url = "https://files.pythonhosted.org/packages/62/a1/3d680cbfd5f4b8f15abc1d571870c5fc3e594bb582bc3b64ea099db13e56/jinja2-3.1.6-py3-none-any.whl", hash = "sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67", size = 134899, upload-time = "2025-03-05T20:05:00.369Z" },
 ]
 
 [[package]]
 name = "jiter"
-version = "0.8.2"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f8/70/90bc7bd3932e651486861df5c8ffea4ca7c77d28e8532ddefe2abc561a53/jiter-0.8.2.tar.gz", hash = "sha256:cd73d3e740666d0e639f678adb176fad25c1bcbdae88d8d7b857e1783bb4212d", size = 163007 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/f2/f3/8c11e0e87bd5934c414f9b1cfae3cbfd4a938d4669d57cb427e1c4d11a7f/jiter-0.8.2-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:ca8577f6a413abe29b079bc30f907894d7eb07a865c4df69475e868d73e71c7b", size = 303381 },
-    { url = "https://files.pythonhosted.org/packages/ea/28/4cd3f0bcbf40e946bc6a62a82c951afc386a25673d3d8d5ee461f1559bbe/jiter-0.8.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:b25bd626bde7fb51534190c7e3cb97cee89ee76b76d7585580e22f34f5e3f393", size = 311718 },
-    { url = "https://files.pythonhosted.org/packages/0d/17/57acab00507e60bd954eaec0837d9d7b119b4117ff49b8a62f2b646f32ed/jiter-0.8.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d5c826a221851a8dc028eb6d7d6429ba03184fa3c7e83ae01cd6d3bd1d4bd17d", size = 335465 },
-    { url = "https://files.pythonhosted.org/packages/74/b9/1a3ddd2bc95ae17c815b021521020f40c60b32137730126bada962ef32b4/jiter-0.8.2-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:d35c864c2dff13dfd79fb070fc4fc6235d7b9b359efe340e1261deb21b9fcb66", size = 355570 },
-    { url = "https://files.pythonhosted.org/packages/78/69/6d29e2296a934199a7d0dde673ecccf98c9c8db44caf0248b3f2b65483cb/jiter-0.8.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f557c55bc2b7676e74d39d19bcb8775ca295c7a028246175d6a8b431e70835e5", size = 381383 },
-    { url = "https://files.pythonhosted.org/packages/22/d7/fbc4c3fb1bf65f9be22a32759b539f88e897aeb13fe84ab0266e4423487a/jiter-0.8.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:580ccf358539153db147e40751a0b41688a5ceb275e6f3e93d91c9467f42b2e3", size = 390454 },
-    { url = "https://files.pythonhosted.org/packages/4d/a0/3993cda2e267fe679b45d0bcc2cef0b4504b0aa810659cdae9737d6bace9/jiter-0.8.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:af102d3372e917cffce49b521e4c32c497515119dc7bd8a75665e90a718bbf08", size = 345039 },
-    { url = "https://files.pythonhosted.org/packages/b9/ef/69c18562b4c09ce88fab5df1dcaf643f6b1a8b970b65216e7221169b81c4/jiter-0.8.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:cadcc978f82397d515bb2683fc0d50103acff2a180552654bb92d6045dec2c49", size = 376200 },
-    { url = "https://files.pythonhosted.org/packages/4d/17/0b5a8de46a6ab4d836f70934036278b49b8530c292b29dde3483326d4555/jiter-0.8.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:ba5bdf56969cad2019d4e8ffd3f879b5fdc792624129741d3d83fc832fef8c7d", size = 511158 },
-    { url = "https://files.pythonhosted.org/packages/6c/b2/c401a0a2554b36c9e6d6e4876b43790d75139cf3936f0222e675cbc23451/jiter-0.8.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:3b94a33a241bee9e34b8481cdcaa3d5c2116f575e0226e421bed3f7a6ea71cff", size = 503956 },
-    { url = "https://files.pythonhosted.org/packages/d4/02/a0291ed7d72c0ac130f172354ee3cf0b2556b69584de391463a8ee534f40/jiter-0.8.2-cp310-cp310-win32.whl", hash = "sha256:6e5337bf454abddd91bd048ce0dca5134056fc99ca0205258766db35d0a2ea43", size = 202846 },
-    { url = "https://files.pythonhosted.org/packages/ad/20/8c988831ae4bf437e29f1671e198fc99ba8fe49f2895f23789acad1d1811/jiter-0.8.2-cp310-cp310-win_amd64.whl", hash = "sha256:4a9220497ca0cb1fe94e3f334f65b9b5102a0b8147646118f020d8ce1de70105", size = 204414 },
-    { url = "https://files.pythonhosted.org/packages/cb/b0/c1a7caa7f9dc5f1f6cfa08722867790fe2d3645d6e7170ca280e6e52d163/jiter-0.8.2-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:2dd61c5afc88a4fda7d8b2cf03ae5947c6ac7516d32b7a15bf4b49569a5c076b", size = 303666 },
-    { url = "https://files.pythonhosted.org/packages/f5/97/0468bc9eeae43079aaa5feb9267964e496bf13133d469cfdc135498f8dd0/jiter-0.8.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a6c710d657c8d1d2adbbb5c0b0c6bfcec28fd35bd6b5f016395f9ac43e878a15", size = 311934 },
-    { url = "https://files.pythonhosted.org/packages/e5/69/64058e18263d9a5f1e10f90c436853616d5f047d997c37c7b2df11b085ec/jiter-0.8.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a9584de0cd306072635fe4b89742bf26feae858a0683b399ad0c2509011b9dc0", size = 335506 },
-    { url = "https://files.pythonhosted.org/packages/9d/14/b747f9a77b8c0542141d77ca1e2a7523e854754af2c339ac89a8b66527d6/jiter-0.8.2-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5a90a923338531b7970abb063cfc087eebae6ef8ec8139762007188f6bc69a9f", size = 355849 },
-    { url = "https://files.pythonhosted.org/packages/53/e2/98a08161db7cc9d0e39bc385415890928ff09709034982f48eccfca40733/jiter-0.8.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d21974d246ed0181558087cd9f76e84e8321091ebfb3a93d4c341479a736f099", size = 381700 },
-    { url = "https://files.pythonhosted.org/packages/7a/38/1674672954d35bce3b1c9af99d5849f9256ac8f5b672e020ac7821581206/jiter-0.8.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:32475a42b2ea7b344069dc1e81445cfc00b9d0e3ca837f0523072432332e9f74", size = 389710 },
-    { url = "https://files.pythonhosted.org/packages/f8/9b/92f9da9a9e107d019bcf883cd9125fa1690079f323f5a9d5c6986eeec3c0/jiter-0.8.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8b9931fd36ee513c26b5bf08c940b0ac875de175341cbdd4fa3be109f0492586", size = 345553 },
-    { url = "https://files.pythonhosted.org/packages/44/a6/6d030003394e9659cd0d7136bbeabd82e869849ceccddc34d40abbbbb269/jiter-0.8.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:ce0820f4a3a59ddced7fce696d86a096d5cc48d32a4183483a17671a61edfddc", size = 376388 },
-    { url = "https://files.pythonhosted.org/packages/ad/8d/87b09e648e4aca5f9af89e3ab3cfb93db2d1e633b2f2931ede8dabd9b19a/jiter-0.8.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:8ffc86ae5e3e6a93765d49d1ab47b6075a9c978a2b3b80f0f32628f39caa0c88", size = 511226 },
-    { url = "https://files.pythonhosted.org/packages/77/95/8008ebe4cdc82eac1c97864a8042ca7e383ed67e0ec17bfd03797045c727/jiter-0.8.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:5127dc1abd809431172bc3fbe8168d6b90556a30bb10acd5ded41c3cfd6f43b6", size = 504134 },
-    { url = "https://files.pythonhosted.org/packages/26/0d/3056a74de13e8b2562e4d526de6dac2f65d91ace63a8234deb9284a1d24d/jiter-0.8.2-cp311-cp311-win32.whl", hash = "sha256:66227a2c7b575720c1871c8800d3a0122bb8ee94edb43a5685aa9aceb2782d44", size = 203103 },
-    { url = "https://files.pythonhosted.org/packages/4e/1e/7f96b798f356e531ffc0f53dd2f37185fac60fae4d6c612bbbd4639b90aa/jiter-0.8.2-cp311-cp311-win_amd64.whl", hash = "sha256:cde031d8413842a1e7501e9129b8e676e62a657f8ec8166e18a70d94d4682855", size = 206717 },
-    { url = "https://files.pythonhosted.org/packages/a1/17/c8747af8ea4e045f57d6cfd6fc180752cab9bc3de0e8a0c9ca4e8af333b1/jiter-0.8.2-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:e6ec2be506e7d6f9527dae9ff4b7f54e68ea44a0ef6b098256ddf895218a2f8f", size = 302027 },
-    { url = "https://files.pythonhosted.org/packages/3c/c1/6da849640cd35a41e91085723b76acc818d4b7d92b0b6e5111736ce1dd10/jiter-0.8.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:76e324da7b5da060287c54f2fabd3db5f76468006c811831f051942bf68c9d44", size = 310326 },
-    { url = "https://files.pythonhosted.org/packages/06/99/a2bf660d8ccffee9ad7ed46b4f860d2108a148d0ea36043fd16f4dc37e94/jiter-0.8.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:180a8aea058f7535d1c84183c0362c710f4750bef66630c05f40c93c2b152a0f", size = 334242 },
-    { url = "https://files.pythonhosted.org/packages/a7/5f/cea1c17864828731f11427b9d1ab7f24764dbd9aaf4648a7f851164d2718/jiter-0.8.2-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:025337859077b41548bdcbabe38698bcd93cfe10b06ff66617a48ff92c9aec60", size = 356654 },
-    { url = "https://files.pythonhosted.org/packages/e9/13/62774b7e5e7f5d5043efe1d0f94ead66e6d0f894ae010adb56b3f788de71/jiter-0.8.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ecff0dc14f409599bbcafa7e470c00b80f17abc14d1405d38ab02e4b42e55b57", size = 379967 },
-    { url = "https://files.pythonhosted.org/packages/ec/fb/096b34c553bb0bd3f2289d5013dcad6074948b8d55212aa13a10d44c5326/jiter-0.8.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ffd9fee7d0775ebaba131f7ca2e2d83839a62ad65e8e02fe2bd8fc975cedeb9e", size = 389252 },
-    { url = "https://files.pythonhosted.org/packages/17/61/beea645c0bf398ced8b199e377b61eb999d8e46e053bb285c91c3d3eaab0/jiter-0.8.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:14601dcac4889e0a1c75ccf6a0e4baf70dbc75041e51bcf8d0e9274519df6887", size = 345490 },
-    { url = "https://files.pythonhosted.org/packages/d5/df/834aa17ad5dcc3cf0118821da0a0cf1589ea7db9832589278553640366bc/jiter-0.8.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:92249669925bc1c54fcd2ec73f70f2c1d6a817928480ee1c65af5f6b81cdf12d", size = 376991 },
-    { url = "https://files.pythonhosted.org/packages/67/80/87d140399d382fb4ea5b3d56e7ecaa4efdca17cd7411ff904c1517855314/jiter-0.8.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:e725edd0929fa79f8349ab4ec7f81c714df51dc4e991539a578e5018fa4a7152", size = 510822 },
-    { url = "https://files.pythonhosted.org/packages/5c/37/3394bb47bac1ad2cb0465601f86828a0518d07828a650722e55268cdb7e6/jiter-0.8.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:bf55846c7b7a680eebaf9c3c48d630e1bf51bdf76c68a5f654b8524335b0ad29", size = 503730 },
-    { url = "https://files.pythonhosted.org/packages/f9/e2/253fc1fa59103bb4e3aa0665d6ceb1818df1cd7bf3eb492c4dad229b1cd4/jiter-0.8.2-cp312-cp312-win32.whl", hash = "sha256:7efe4853ecd3d6110301665a5178b9856be7e2a9485f49d91aa4d737ad2ae49e", size = 203375 },
-    { url = "https://files.pythonhosted.org/packages/41/69/6d4bbe66b3b3b4507e47aa1dd5d075919ad242b4b1115b3f80eecd443687/jiter-0.8.2-cp312-cp312-win_amd64.whl", hash = "sha256:83c0efd80b29695058d0fd2fa8a556490dbce9804eac3e281f373bbc99045f6c", size = 204740 },
-    { url = "https://files.pythonhosted.org/packages/6c/b0/bfa1f6f2c956b948802ef5a021281978bf53b7a6ca54bb126fd88a5d014e/jiter-0.8.2-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:ca1f08b8e43dc3bd0594c992fb1fd2f7ce87f7bf0d44358198d6da8034afdf84", size = 301190 },
-    { url = "https://files.pythonhosted.org/packages/a4/8f/396ddb4e292b5ea57e45ade5dc48229556b9044bad29a3b4b2dddeaedd52/jiter-0.8.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:5672a86d55416ccd214c778efccf3266b84f87b89063b582167d803246354be4", size = 309334 },
-    { url = "https://files.pythonhosted.org/packages/7f/68/805978f2f446fa6362ba0cc2e4489b945695940656edd844e110a61c98f8/jiter-0.8.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:58dc9bc9767a1101f4e5e22db1b652161a225874d66f0e5cb8e2c7d1c438b587", size = 333918 },
-    { url = "https://files.pythonhosted.org/packages/b3/99/0f71f7be667c33403fa9706e5b50583ae5106d96fab997fa7e2f38ee8347/jiter-0.8.2-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:37b2998606d6dadbb5ccda959a33d6a5e853252d921fec1792fc902351bb4e2c", size = 356057 },
-    { url = "https://files.pythonhosted.org/packages/8d/50/a82796e421a22b699ee4d2ce527e5bcb29471a2351cbdc931819d941a167/jiter-0.8.2-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4ab9a87f3784eb0e098f84a32670cfe4a79cb6512fd8f42ae3d0709f06405d18", size = 379790 },
-    { url = "https://files.pythonhosted.org/packages/3c/31/10fb012b00f6d83342ca9e2c9618869ab449f1aa78c8f1b2193a6b49647c/jiter-0.8.2-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:79aec8172b9e3c6d05fd4b219d5de1ac616bd8da934107325a6c0d0e866a21b6", size = 388285 },
-    { url = "https://files.pythonhosted.org/packages/c8/81/f15ebf7de57be488aa22944bf4274962aca8092e4f7817f92ffa50d3ee46/jiter-0.8.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:711e408732d4e9a0208008e5892c2966b485c783cd2d9a681f3eb147cf36c7ef", size = 344764 },
-    { url = "https://files.pythonhosted.org/packages/b3/e8/0cae550d72b48829ba653eb348cdc25f3f06f8a62363723702ec18e7be9c/jiter-0.8.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:653cf462db4e8c41995e33d865965e79641ef45369d8a11f54cd30888b7e6ff1", size = 376620 },
-    { url = "https://files.pythonhosted.org/packages/b8/50/e5478ff9d82534a944c03b63bc217c5f37019d4a34d288db0f079b13c10b/jiter-0.8.2-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:9c63eaef32b7bebac8ebebf4dabebdbc6769a09c127294db6babee38e9f405b9", size = 510402 },
-    { url = "https://files.pythonhosted.org/packages/8e/1e/3de48bbebbc8f7025bd454cedc8c62378c0e32dd483dece5f4a814a5cb55/jiter-0.8.2-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:eb21aaa9a200d0a80dacc7a81038d2e476ffe473ffdd9c91eb745d623561de05", size = 503018 },
-    { url = "https://files.pythonhosted.org/packages/d5/cd/d5a5501d72a11fe3e5fd65c78c884e5164eefe80077680533919be22d3a3/jiter-0.8.2-cp313-cp313-win32.whl", hash = "sha256:789361ed945d8d42850f919342a8665d2dc79e7e44ca1c97cc786966a21f627a", size = 203190 },
-    { url = "https://files.pythonhosted.org/packages/51/bf/e5ca301245ba951447e3ad677a02a64a8845b185de2603dabd83e1e4b9c6/jiter-0.8.2-cp313-cp313-win_amd64.whl", hash = "sha256:ab7f43235d71e03b941c1630f4b6e3055d46b6cb8728a17663eaac9d8e83a865", size = 203551 },
-    { url = "https://files.pythonhosted.org/packages/2f/3c/71a491952c37b87d127790dd7a0b1ebea0514c6b6ad30085b16bbe00aee6/jiter-0.8.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:b426f72cd77da3fec300ed3bc990895e2dd6b49e3bfe6c438592a3ba660e41ca", size = 308347 },
-    { url = "https://files.pythonhosted.org/packages/a0/4c/c02408042e6a7605ec063daed138e07b982fdb98467deaaf1c90950cf2c6/jiter-0.8.2-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b2dd880785088ff2ad21ffee205e58a8c1ddabc63612444ae41e5e4b321b39c0", size = 342875 },
-    { url = "https://files.pythonhosted.org/packages/91/61/c80ef80ed8a0a21158e289ef70dac01e351d929a1c30cb0f49be60772547/jiter-0.8.2-cp313-cp313t-win_amd64.whl", hash = "sha256:3ac9f578c46f22405ff7f8b1f5848fb753cc4b8377fbec8470a7dc3997ca7566", size = 202374 },
-    { url = "https://files.pythonhosted.org/packages/c9/b2/ed7fbabd21c3cf556d6ea849cee35c74f13a509e668baad8323091e2867e/jiter-0.8.2-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:e41e75344acef3fc59ba4765df29f107f309ca9e8eace5baacabd9217e52a5ee", size = 304502 },
-    { url = "https://files.pythonhosted.org/packages/75/6e/1386857ac9165c1e9c71031566e7884d8a4f63724ce29ad1ace5bfe1351c/jiter-0.8.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:7f22b16b35d5c1df9dfd58843ab2cd25e6bf15191f5a236bed177afade507bfc", size = 300982 },
-    { url = "https://files.pythonhosted.org/packages/56/4c/b413977c20bbb359b4d6c91d04f7f36fc525af0b7778119815477fc97242/jiter-0.8.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f7200b8f7619d36aa51c803fd52020a2dfbea36ffec1b5e22cab11fd34d95a6d", size = 335344 },
-    { url = "https://files.pythonhosted.org/packages/b0/59/51b080519938192edd33b4e8d48adb7e9bf9e0d699ec8b91119b9269fc75/jiter-0.8.2-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:70bf4c43652cc294040dbb62256c83c8718370c8b93dd93d934b9a7bf6c4f53c", size = 356298 },
-    { url = "https://files.pythonhosted.org/packages/72/bb/828db5ea406916d7b2232be31393f782b0f71bcb0b128750c4a028157565/jiter-0.8.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f9d471356dc16f84ed48768b8ee79f29514295c7295cb41e1133ec0b2b8d637d", size = 381703 },
-    { url = "https://files.pythonhosted.org/packages/c0/88/45d33a8728733e161e9783c54d8ecca0fc4c1aa74b1cebea1d97917eddc3/jiter-0.8.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:859e8eb3507894093d01929e12e267f83b1d5f6221099d3ec976f0c995cb6bd9", size = 391281 },
-    { url = "https://files.pythonhosted.org/packages/45/3e/142712e0f45c28ad8a678dc8732a78294ce5a36fc694141f772bb827a8f2/jiter-0.8.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:eaa58399c01db555346647a907b4ef6d4f584b123943be6ed5588c3f2359c9f4", size = 345553 },
-    { url = "https://files.pythonhosted.org/packages/36/42/9b463b59fd22687b6da1afcad6c9adc870464a808208651de73f1dbeda09/jiter-0.8.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:8f2d5ed877f089862f4c7aacf3a542627c1496f972a34d0474ce85ee7d939c27", size = 377063 },
-    { url = "https://files.pythonhosted.org/packages/83/b3/44b1f5cd2e4eb15757eec341b25399da4c90515bb881ef6636b50a8c08a5/jiter-0.8.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:03c9df035d4f8d647f8c210ddc2ae0728387275340668fb30d2421e17d9a0841", size = 512543 },
-    { url = "https://files.pythonhosted.org/packages/46/4e/c695c803aa2b668c057b2dea1cdd7a884d1a819ce610cec0be9666210bfd/jiter-0.8.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:8bd2a824d08d8977bb2794ea2682f898ad3d8837932e3a74937e93d62ecbb637", size = 505141 },
-    { url = "https://files.pythonhosted.org/packages/8e/51/e805b837db056f872db0b7a7a3610b7d764392be696dbe47afa0bea05bf2/jiter-0.8.2-cp39-cp39-win32.whl", hash = "sha256:ca29b6371ebc40e496995c94b988a101b9fbbed48a51190a4461fcb0a68b4a36", size = 203529 },
-    { url = "https://files.pythonhosted.org/packages/32/b7/a3cde72c644fd1caf9da07fb38cf2c130f43484d8f91011940b7c4f42c8f/jiter-0.8.2-cp39-cp39-win_amd64.whl", hash = "sha256:1c0dfbd1be3cbefc7510102370d86e35d1d53e5a93d48519688b1bf0f761160a", size = 207527 },
+version = "0.10.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ee/9d/ae7ddb4b8ab3fb1b51faf4deb36cb48a4fbbd7cb36bad6a5fca4741306f7/jiter-0.10.0.tar.gz", hash = "sha256:07a7142c38aacc85194391108dc91b5b57093c978a9932bd86a36862759d9500", size = 162759, upload-time = "2025-05-18T19:04:59.73Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/be/7e/4011b5c77bec97cb2b572f566220364e3e21b51c48c5bd9c4a9c26b41b67/jiter-0.10.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:cd2fb72b02478f06a900a5782de2ef47e0396b3e1f7d5aba30daeb1fce66f303", size = 317215, upload-time = "2025-05-18T19:03:04.303Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/4f/144c1b57c39692efc7ea7d8e247acf28e47d0912800b34d0ad815f6b2824/jiter-0.10.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:32bb468e3af278f095d3fa5b90314728a6916d89ba3d0ffb726dd9bf7367285e", size = 322814, upload-time = "2025-05-18T19:03:06.433Z" },
+    { url = "https://files.pythonhosted.org/packages/63/1f/db977336d332a9406c0b1f0b82be6f71f72526a806cbb2281baf201d38e3/jiter-0.10.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:aa8b3e0068c26ddedc7abc6fac37da2d0af16b921e288a5a613f4b86f050354f", size = 345237, upload-time = "2025-05-18T19:03:07.833Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/1c/aa30a4a775e8a672ad7f21532bdbfb269f0706b39c6ff14e1f86bdd9e5ff/jiter-0.10.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:286299b74cc49e25cd42eea19b72aa82c515d2f2ee12d11392c56d8701f52224", size = 370999, upload-time = "2025-05-18T19:03:09.338Z" },
+    { url = "https://files.pythonhosted.org/packages/35/df/f8257abc4207830cb18880781b5f5b716bad5b2a22fb4330cfd357407c5b/jiter-0.10.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6ed5649ceeaeffc28d87fb012d25a4cd356dcd53eff5acff1f0466b831dda2a7", size = 491109, upload-time = "2025-05-18T19:03:11.13Z" },
+    { url = "https://files.pythonhosted.org/packages/06/76/9e1516fd7b4278aa13a2cc7f159e56befbea9aa65c71586305e7afa8b0b3/jiter-0.10.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b2ab0051160cb758a70716448908ef14ad476c3774bd03ddce075f3c1f90a3d6", size = 388608, upload-time = "2025-05-18T19:03:12.911Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/64/67750672b4354ca20ca18d3d1ccf2c62a072e8a2d452ac3cf8ced73571ef/jiter-0.10.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:03997d2f37f6b67d2f5c475da4412be584e1cec273c1cfc03d642c46db43f8cf", size = 352454, upload-time = "2025-05-18T19:03:14.741Z" },
+    { url = "https://files.pythonhosted.org/packages/96/4d/5c4e36d48f169a54b53a305114be3efa2bbffd33b648cd1478a688f639c1/jiter-0.10.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c404a99352d839fed80d6afd6c1d66071f3bacaaa5c4268983fc10f769112e90", size = 391833, upload-time = "2025-05-18T19:03:16.426Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/de/ce4a6166a78810bd83763d2fa13f85f73cbd3743a325469a4a9289af6dae/jiter-0.10.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:66e989410b6666d3ddb27a74c7e50d0829704ede652fd4c858e91f8d64b403d0", size = 523646, upload-time = "2025-05-18T19:03:17.704Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/a6/3bc9acce53466972964cf4ad85efecb94f9244539ab6da1107f7aed82934/jiter-0.10.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:b532d3af9ef4f6374609a3bcb5e05a1951d3bf6190dc6b176fdb277c9bbf15ee", size = 514735, upload-time = "2025-05-18T19:03:19.44Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/d8/243c2ab8426a2a4dea85ba2a2ba43df379ccece2145320dfd4799b9633c5/jiter-0.10.0-cp310-cp310-win32.whl", hash = "sha256:da9be20b333970e28b72edc4dff63d4fec3398e05770fb3205f7fb460eb48dd4", size = 210747, upload-time = "2025-05-18T19:03:21.184Z" },
+    { url = "https://files.pythonhosted.org/packages/37/7a/8021bd615ef7788b98fc76ff533eaac846322c170e93cbffa01979197a45/jiter-0.10.0-cp310-cp310-win_amd64.whl", hash = "sha256:f59e533afed0c5b0ac3eba20d2548c4a550336d8282ee69eb07b37ea526ee4e5", size = 207484, upload-time = "2025-05-18T19:03:23.046Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/dd/6cefc6bd68b1c3c979cecfa7029ab582b57690a31cd2f346c4d0ce7951b6/jiter-0.10.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:3bebe0c558e19902c96e99217e0b8e8b17d570906e72ed8a87170bc290b1e978", size = 317473, upload-time = "2025-05-18T19:03:25.942Z" },
+    { url = "https://files.pythonhosted.org/packages/be/cf/fc33f5159ce132be1d8dd57251a1ec7a631c7df4bd11e1cd198308c6ae32/jiter-0.10.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:558cc7e44fd8e507a236bee6a02fa17199ba752874400a0ca6cd6e2196cdb7dc", size = 321971, upload-time = "2025-05-18T19:03:27.255Z" },
+    { url = "https://files.pythonhosted.org/packages/68/a4/da3f150cf1d51f6c472616fb7650429c7ce053e0c962b41b68557fdf6379/jiter-0.10.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4d613e4b379a07d7c8453c5712ce7014e86c6ac93d990a0b8e7377e18505e98d", size = 345574, upload-time = "2025-05-18T19:03:28.63Z" },
+    { url = "https://files.pythonhosted.org/packages/84/34/6e8d412e60ff06b186040e77da5f83bc158e9735759fcae65b37d681f28b/jiter-0.10.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:f62cf8ba0618eda841b9bf61797f21c5ebd15a7a1e19daab76e4e4b498d515b2", size = 371028, upload-time = "2025-05-18T19:03:30.292Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/d9/9ee86173aae4576c35a2f50ae930d2ccb4c4c236f6cb9353267aa1d626b7/jiter-0.10.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:919d139cdfa8ae8945112398511cb7fca58a77382617d279556b344867a37e61", size = 491083, upload-time = "2025-05-18T19:03:31.654Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/2c/f955de55e74771493ac9e188b0f731524c6a995dffdcb8c255b89c6fb74b/jiter-0.10.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:13ddbc6ae311175a3b03bd8994881bc4635c923754932918e18da841632349db", size = 388821, upload-time = "2025-05-18T19:03:33.184Z" },
+    { url = "https://files.pythonhosted.org/packages/81/5a/0e73541b6edd3f4aada586c24e50626c7815c561a7ba337d6a7eb0a915b4/jiter-0.10.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4c440ea003ad10927a30521a9062ce10b5479592e8a70da27f21eeb457b4a9c5", size = 352174, upload-time = "2025-05-18T19:03:34.965Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/c0/61eeec33b8c75b31cae42be14d44f9e6fe3ac15a4e58010256ac3abf3638/jiter-0.10.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:dc347c87944983481e138dea467c0551080c86b9d21de6ea9306efb12ca8f606", size = 391869, upload-time = "2025-05-18T19:03:36.436Z" },
+    { url = "https://files.pythonhosted.org/packages/41/22/5beb5ee4ad4ef7d86f5ea5b4509f680a20706c4a7659e74344777efb7739/jiter-0.10.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:13252b58c1f4d8c5b63ab103c03d909e8e1e7842d302473f482915d95fefd605", size = 523741, upload-time = "2025-05-18T19:03:38.168Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/10/768e8818538e5817c637b0df52e54366ec4cebc3346108a4457ea7a98f32/jiter-0.10.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:7d1bbf3c465de4a24ab12fb7766a0003f6f9bce48b8b6a886158c4d569452dc5", size = 514527, upload-time = "2025-05-18T19:03:39.577Z" },
+    { url = "https://files.pythonhosted.org/packages/73/6d/29b7c2dc76ce93cbedabfd842fc9096d01a0550c52692dfc33d3cc889815/jiter-0.10.0-cp311-cp311-win32.whl", hash = "sha256:db16e4848b7e826edca4ccdd5b145939758dadf0dc06e7007ad0e9cfb5928ae7", size = 210765, upload-time = "2025-05-18T19:03:41.271Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/c9/d394706deb4c660137caf13e33d05a031d734eb99c051142e039d8ceb794/jiter-0.10.0-cp311-cp311-win_amd64.whl", hash = "sha256:9c9c1d5f10e18909e993f9641f12fe1c77b3e9b533ee94ffa970acc14ded3812", size = 209234, upload-time = "2025-05-18T19:03:42.918Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/b5/348b3313c58f5fbfb2194eb4d07e46a35748ba6e5b3b3046143f3040bafa/jiter-0.10.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:1e274728e4a5345a6dde2d343c8da018b9d4bd4350f5a472fa91f66fda44911b", size = 312262, upload-time = "2025-05-18T19:03:44.637Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/4a/6a2397096162b21645162825f058d1709a02965606e537e3304b02742e9b/jiter-0.10.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:7202ae396446c988cb2a5feb33a543ab2165b786ac97f53b59aafb803fef0744", size = 320124, upload-time = "2025-05-18T19:03:46.341Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/85/1ce02cade7516b726dd88f59a4ee46914bf79d1676d1228ef2002ed2f1c9/jiter-0.10.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:23ba7722d6748b6920ed02a8f1726fb4b33e0fd2f3f621816a8b486c66410ab2", size = 345330, upload-time = "2025-05-18T19:03:47.596Z" },
+    { url = "https://files.pythonhosted.org/packages/75/d0/bb6b4f209a77190ce10ea8d7e50bf3725fc16d3372d0a9f11985a2b23eff/jiter-0.10.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:371eab43c0a288537d30e1f0b193bc4eca90439fc08a022dd83e5e07500ed026", size = 369670, upload-time = "2025-05-18T19:03:49.334Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/f5/a61787da9b8847a601e6827fbc42ecb12be2c925ced3252c8ffcb56afcaf/jiter-0.10.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6c675736059020365cebc845a820214765162728b51ab1e03a1b7b3abb70f74c", size = 489057, upload-time = "2025-05-18T19:03:50.66Z" },
+    { url = "https://files.pythonhosted.org/packages/12/e4/6f906272810a7b21406c760a53aadbe52e99ee070fc5c0cb191e316de30b/jiter-0.10.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0c5867d40ab716e4684858e4887489685968a47e3ba222e44cde6e4a2154f959", size = 389372, upload-time = "2025-05-18T19:03:51.98Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/ba/77013b0b8ba904bf3762f11e0129b8928bff7f978a81838dfcc958ad5728/jiter-0.10.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:395bb9a26111b60141757d874d27fdea01b17e8fac958b91c20128ba8f4acc8a", size = 352038, upload-time = "2025-05-18T19:03:53.703Z" },
+    { url = "https://files.pythonhosted.org/packages/67/27/c62568e3ccb03368dbcc44a1ef3a423cb86778a4389e995125d3d1aaa0a4/jiter-0.10.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:6842184aed5cdb07e0c7e20e5bdcfafe33515ee1741a6835353bb45fe5d1bd95", size = 391538, upload-time = "2025-05-18T19:03:55.046Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/72/0d6b7e31fc17a8fdce76164884edef0698ba556b8eb0af9546ae1a06b91d/jiter-0.10.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:62755d1bcea9876770d4df713d82606c8c1a3dca88ff39046b85a048566d56ea", size = 523557, upload-time = "2025-05-18T19:03:56.386Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/09/bc1661fbbcbeb6244bd2904ff3a06f340aa77a2b94e5a7373fd165960ea3/jiter-0.10.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:533efbce2cacec78d5ba73a41756beff8431dfa1694b6346ce7af3a12c42202b", size = 514202, upload-time = "2025-05-18T19:03:57.675Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/84/5a5d5400e9d4d54b8004c9673bbe4403928a00d28529ff35b19e9d176b19/jiter-0.10.0-cp312-cp312-win32.whl", hash = "sha256:8be921f0cadd245e981b964dfbcd6fd4bc4e254cdc069490416dd7a2632ecc01", size = 211781, upload-time = "2025-05-18T19:03:59.025Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/52/7ec47455e26f2d6e5f2ea4951a0652c06e5b995c291f723973ae9e724a65/jiter-0.10.0-cp312-cp312-win_amd64.whl", hash = "sha256:a7c7d785ae9dda68c2678532a5a1581347e9c15362ae9f6e68f3fdbfb64f2e49", size = 206176, upload-time = "2025-05-18T19:04:00.305Z" },
+    { url = "https://files.pythonhosted.org/packages/98/fd/aced428e2bd3c6c1132f67c5a708f9e7fd161d0ca8f8c5862b17b93cdf0a/jiter-0.10.0-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:bd6292a43c0fc09ce7c154ec0fa646a536b877d1e8f2f96c19707f65355b5a4d", size = 317665, upload-time = "2025-05-18T19:04:43.417Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/2e/47d42f15d53ed382aef8212a737101ae2720e3697a954f9b95af06d34e89/jiter-0.10.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:39de429dcaeb6808d75ffe9effefe96a4903c6a4b376b2f6d08d77c1aaee2f18", size = 312152, upload-time = "2025-05-18T19:04:44.797Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/02/aae834228ef4834fc18718724017995ace8da5f70aa1ec225b9bc2b2d7aa/jiter-0.10.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:52ce124f13a7a616fad3bb723f2bfb537d78239d1f7f219566dc52b6f2a9e48d", size = 346708, upload-time = "2025-05-18T19:04:46.127Z" },
+    { url = "https://files.pythonhosted.org/packages/35/d4/6ff39dee2d0a9abd69d8a3832ce48a3aa644eed75e8515b5ff86c526ca9a/jiter-0.10.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:166f3606f11920f9a1746b2eea84fa2c0a5d50fd313c38bdea4edc072000b0af", size = 371360, upload-time = "2025-05-18T19:04:47.448Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/67/c749d962b4eb62445867ae4e64a543cbb5d63cc7d78ada274ac515500a7f/jiter-0.10.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:28dcecbb4ba402916034fc14eba7709f250c4d24b0c43fc94d187ee0580af181", size = 492105, upload-time = "2025-05-18T19:04:48.792Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/d3/8fe1b1bae5161f27b1891c256668f598fa4c30c0a7dacd668046a6215fca/jiter-0.10.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:86c5aa6910f9bebcc7bc4f8bc461aff68504388b43bfe5e5c0bd21efa33b52f4", size = 389577, upload-time = "2025-05-18T19:04:50.13Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/28/ecb19d789b4777898a4252bfaac35e3f8caf16c93becd58dcbaac0dc24ad/jiter-0.10.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ceeb52d242b315d7f1f74b441b6a167f78cea801ad7c11c36da77ff2d42e8a28", size = 353849, upload-time = "2025-05-18T19:04:51.443Z" },
+    { url = "https://files.pythonhosted.org/packages/77/69/261f798f84790da6482ebd8c87ec976192b8c846e79444d0a2e0d33ebed8/jiter-0.10.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:ff76d8887c8c8ee1e772274fcf8cc1071c2c58590d13e33bd12d02dc9a560397", size = 392029, upload-time = "2025-05-18T19:04:52.792Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/08/b8d15140d4d91f16faa2f5d416c1a71ab1bbe2b66c57197b692d04c0335f/jiter-0.10.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:a9be4d0fa2b79f7222a88aa488bd89e2ae0a0a5b189462a12def6ece2faa45f1", size = 524386, upload-time = "2025-05-18T19:04:54.203Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/1d/23c41765cc95c0e23ac492a88450d34bf0fd87a37218d1b97000bffe0f53/jiter-0.10.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:9ab7fd8738094139b6c1ab1822d6f2000ebe41515c537235fd45dabe13ec9324", size = 515234, upload-time = "2025-05-18T19:04:55.838Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/14/381d8b151132e79790579819c3775be32820569f23806769658535fe467f/jiter-0.10.0-cp39-cp39-win32.whl", hash = "sha256:5f51e048540dd27f204ff4a87f5d79294ea0aa3aa552aca34934588cf27023cf", size = 211436, upload-time = "2025-05-18T19:04:57.183Z" },
+    { url = "https://files.pythonhosted.org/packages/59/66/f23ae51dea8ee8ce429027b60008ca895d0fa0704f0c7fe5f09014a6cffb/jiter-0.10.0-cp39-cp39-win_amd64.whl", hash = "sha256:1b28302349dc65703a9e4ead16f163b1c339efffbe1049c30a44b001a2a4fff9", size = 208777, upload-time = "2025-05-18T19:04:58.454Z" },
+]
+
+[[package]]
+name = "jsonpath-ng"
+version = "1.7.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "ply" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6d/86/08646239a313f895186ff0a4573452038eed8c86f54380b3ebac34d32fb2/jsonpath-ng-1.7.0.tar.gz", hash = "sha256:f6f5f7fd4e5ff79c785f1573b394043b39849fb2bb47bcead935d12b00beab3c", size = 37838, upload-time = "2024-10-11T15:41:42.404Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/35/5a/73ecb3d82f8615f32ccdadeb9356726d6cae3a4bbc840b437ceb95708063/jsonpath_ng-1.7.0-py3-none-any.whl", hash = "sha256:f3d7f9e848cba1b6da28c55b1c26ff915dc9e0b1ba7e752a53d6da8d5cbd00b6", size = 30105, upload-time = "2024-11-20T17:58:30.418Z" },
 ]
 
 [[package]]
 name = "jsonschema"
-version = "4.23.0"
+version = "4.25.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "attrs" },
@@ -1173,423 +1383,584 @@ dependencies = [
     { name = "referencing" },
     { name = "rpds-py" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/38/2e/03362ee4034a4c917f697890ccd4aec0800ccf9ded7f511971c75451deec/jsonschema-4.23.0.tar.gz", hash = "sha256:d71497fef26351a33265337fa77ffeb82423f3ea21283cd9467bb03999266bc4", size = 325778 }
+sdist = { url = "https://files.pythonhosted.org/packages/74/69/f7185de793a29082a9f3c7728268ffb31cb5095131a9c139a74078e27336/jsonschema-4.25.1.tar.gz", hash = "sha256:e4a9655ce0da0c0b67a085847e00a3a51449e1157f4f75e9fb5aa545e122eb85", size = 357342, upload-time = "2025-08-18T17:03:50.038Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/69/4a/4f9dbeb84e8850557c02365a0eee0649abe5eb1d84af92a25731c6c0f922/jsonschema-4.23.0-py3-none-any.whl", hash = "sha256:fbadb6f8b144a8f8cf9f0b89ba94501d143e50411a1278633f56a7acf7fd5566", size = 88462 },
+    { url = "https://files.pythonhosted.org/packages/bf/9c/8c95d856233c1f82500c2450b8c68576b4cf1c871db3afac5c34ff84e6fd/jsonschema-4.25.1-py3-none-any.whl", hash = "sha256:3fba0169e345c7175110351d456342c364814cfcf3b964ba4587f22915230a63", size = 90040, upload-time = "2025-08-18T17:03:48.373Z" },
 ]
 
 [[package]]
 name = "jsonschema-specifications"
-version = "2024.10.1"
+version = "2025.4.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "referencing" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/10/db/58f950c996c793472e336ff3655b13fbcf1e3b359dcf52dcf3ed3b52c352/jsonschema_specifications-2024.10.1.tar.gz", hash = "sha256:0f38b83639958ce1152d02a7f062902c41c8fd20d558b0c34344292d417ae272", size = 15561 }
+sdist = { url = "https://files.pythonhosted.org/packages/bf/ce/46fbd9c8119cfc3581ee5643ea49464d168028cfb5caff5fc0596d0cf914/jsonschema_specifications-2025.4.1.tar.gz", hash = "sha256:630159c9f4dbea161a6a2205c3011cc4f18ff381b189fff48bb39b9bf26ae608", size = 15513, upload-time = "2025-04-23T12:34:07.418Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/01/0e/b27cdbaccf30b890c40ed1da9fd4a3593a5cf94dae54fb34f8a4b74fcd3f/jsonschema_specifications-2025.4.1-py3-none-any.whl", hash = "sha256:4653bffbd6584f7de83a67e0d620ef16900b390ddc7939d56684d6c81e33f1af", size = 18437, upload-time = "2025-04-23T12:34:05.422Z" },
+]
+
+[[package]]
+name = "keras"
+version = "3.10.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "absl-py", marker = "python_full_version < '3.10'" },
+    { name = "h5py", marker = "python_full_version < '3.10'" },
+    { name = "ml-dtypes", marker = "python_full_version < '3.10'" },
+    { name = "namex", marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "optree", marker = "python_full_version < '3.10'" },
+    { name = "packaging", marker = "python_full_version < '3.10'" },
+    { name = "rich", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f3/fe/2946daf8477ae38a4b480c8889c72ede4f36eb28f9e1a27fc355cd633c3d/keras-3.10.0.tar.gz", hash = "sha256:6e9100bf66eaf6de4b7f288d34ef9bb8b5dcdd62f42c64cfd910226bb34ad2d2", size = 1040781, upload-time = "2025-05-19T22:58:30.833Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/95/e6/4179c461a5fc43e3736880f64dbdc9b1a5349649f0ae32ded927c0e3a227/keras-3.10.0-py3-none-any.whl", hash = "sha256:c095a6bf90cd50defadf73d4859ff794fad76b775357ef7bd1dbf96388dae7d3", size = 1380082, upload-time = "2025-05-19T22:58:28.938Z" },
+]
+
+[[package]]
+name = "keras"
+version = "3.11.3"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "absl-py", marker = "python_full_version >= '3.10'" },
+    { name = "h5py", marker = "python_full_version >= '3.10'" },
+    { name = "ml-dtypes", marker = "python_full_version >= '3.10'" },
+    { name = "namex", marker = "python_full_version >= '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "optree", marker = "python_full_version >= '3.10'" },
+    { name = "packaging", marker = "python_full_version >= '3.10'" },
+    { name = "rich", marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6a/89/646425fe9a46f9053430e1271f817c36041c6f33469950a3caafc3d2591e/keras-3.11.3.tar.gz", hash = "sha256:efda616835c31b7d916d72303ef9adec1257320bc9fd4b2b0138840fc65fb5b7", size = 1065906, upload-time = "2025-08-21T22:08:57.643Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/94/5b/4c778cc921ce4b864b238f63f8e3ff6e954ab19b80c9fa680593ad8093d4/keras-3.11.3-py3-none-any.whl", hash = "sha256:f484f050e05ee400455b05ec8c36ed35edc34de94256b6073f56cfe68f65491f", size = 1408438, upload-time = "2025-08-21T22:08:55.858Z" },
+]
+
+[[package]]
+name = "latex2mathml"
+version = "3.78.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/69/33/ad2c3929494ad160f5130ea132ca298627a6c81c70be6bedd1bc806b5b01/latex2mathml-3.78.0.tar.gz", hash = "sha256:712193aa4c6ade1a8e0145dac7bc1f9aafbd54f93046a2356a7e1c05fa0f8b31", size = 73737, upload-time = "2025-05-03T16:51:53.563Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d1/0f/8910b19ac0670a0f80ce1008e5e751c4a57e14d2c4c13a482aa6079fa9d6/jsonschema_specifications-2024.10.1-py3-none-any.whl", hash = "sha256:a09a0680616357d9a0ecf05c12ad234479f549239d0f5b55f3deea67475da9bf", size = 18459 },
+    { url = "https://files.pythonhosted.org/packages/1e/fd/aba08bb9e527168efad57985d7db9a853eb2384b1efa5ca5f3a3794c9cef/latex2mathml-3.78.0-py3-none-any.whl", hash = "sha256:1aeca3dc027b3006ad7b301b7f4a15ffbb4c1451e3dc8c3389e97b37b497e1d6", size = 73673, upload-time = "2025-05-03T16:51:51.991Z" },
 ]
 
 [[package]]
-name = "lark"
-version = "1.2.2"
+name = "libclang"
+version = "18.1.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/af/60/bc7622aefb2aee1c0b4ba23c1446d3e30225c8770b38d7aedbfb65ca9d5a/lark-1.2.2.tar.gz", hash = "sha256:ca807d0162cd16cef15a8feecb862d7319e7a09bdb13aef927968e45040fed80", size = 252132 }
+sdist = { url = "https://files.pythonhosted.org/packages/6e/5c/ca35e19a4f142adffa27e3d652196b7362fa612243e2b916845d801454fc/libclang-18.1.1.tar.gz", hash = "sha256:a1214966d08d73d971287fc3ead8dfaf82eb07fb197680d8b3859dbbbbf78250", size = 39612, upload-time = "2024-03-17T16:04:37.434Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/2d/00/d90b10b962b4277f5e64a78b6609968859ff86889f5b898c1a778c06ec00/lark-1.2.2-py3-none-any.whl", hash = "sha256:c2276486b02f0f1b90be155f2c8ba4a8e194d42775786db622faccd652d8e80c", size = 111036 },
+    { url = "https://files.pythonhosted.org/packages/4b/49/f5e3e7e1419872b69f6f5e82ba56e33955a74bd537d8a1f5f1eff2f3668a/libclang-18.1.1-1-py2.py3-none-macosx_11_0_arm64.whl", hash = "sha256:0b2e143f0fac830156feb56f9231ff8338c20aecfe72b4ffe96f19e5a1dbb69a", size = 25836045, upload-time = "2024-06-30T17:40:31.646Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/e5/fc61bbded91a8830ccce94c5294ecd6e88e496cc85f6704bf350c0634b70/libclang-18.1.1-py2.py3-none-macosx_10_9_x86_64.whl", hash = "sha256:6f14c3f194704e5d09769108f03185fce7acaf1d1ae4bbb2f30a72c2400cb7c5", size = 26502641, upload-time = "2024-03-18T15:52:26.722Z" },
+    { url = "https://files.pythonhosted.org/packages/db/ed/1df62b44db2583375f6a8a5e2ca5432bbdc3edb477942b9b7c848c720055/libclang-18.1.1-py2.py3-none-macosx_11_0_arm64.whl", hash = "sha256:83ce5045d101b669ac38e6da8e58765f12da2d3aafb3b9b98d88b286a60964d8", size = 26420207, upload-time = "2024-03-17T15:00:26.63Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/fc/716c1e62e512ef1c160e7984a73a5fc7df45166f2ff3f254e71c58076f7c/libclang-18.1.1-py2.py3-none-manylinux2010_x86_64.whl", hash = "sha256:c533091d8a3bbf7460a00cb6c1a71da93bffe148f172c7d03b1c31fbf8aa2a0b", size = 24515943, upload-time = "2024-03-17T16:03:45.942Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/3d/f0ac1150280d8d20d059608cf2d5ff61b7c3b7f7bcf9c0f425ab92df769a/libclang-18.1.1-py2.py3-none-manylinux2014_aarch64.whl", hash = "sha256:54dda940a4a0491a9d1532bf071ea3ef26e6dbaf03b5000ed94dd7174e8f9592", size = 23784972, upload-time = "2024-03-17T16:12:47.677Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/2f/d920822c2b1ce9326a4c78c0c2b4aa3fde610c7ee9f631b600acb5376c26/libclang-18.1.1-py2.py3-none-manylinux2014_armv7l.whl", hash = "sha256:cf4a99b05376513717ab5d82a0db832c56ccea4fd61a69dbb7bccf2dfb207dbe", size = 20259606, upload-time = "2024-03-17T16:17:42.437Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/c2/de1db8c6d413597076a4259cea409b83459b2db997c003578affdd32bf66/libclang-18.1.1-py2.py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:69f8eb8f65c279e765ffd28aaa7e9e364c776c17618af8bff22a8df58677ff4f", size = 24921494, upload-time = "2024-03-17T16:14:20.132Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/2d/3f480b1e1d31eb3d6de5e3ef641954e5c67430d5ac93b7fa7e07589576c7/libclang-18.1.1-py2.py3-none-win_amd64.whl", hash = "sha256:4dd2d3b82fab35e2bf9ca717d7b63ac990a3519c7e312f19fa8e86dcc712f7fb", size = 26415083, upload-time = "2024-03-17T16:42:21.703Z" },
+    { url = "https://files.pythonhosted.org/packages/71/cf/e01dc4cc79779cd82d77888a88ae2fa424d93b445ad4f6c02bfc18335b70/libclang-18.1.1-py2.py3-none-win_arm64.whl", hash = "sha256:3f0e1f49f04d3cd198985fea0511576b0aee16f9ff0e0f0cad7f9c57ec3c20e8", size = 22361112, upload-time = "2024-03-17T16:42:59.565Z" },
 ]
 
 [[package]]
 name = "llama-cpp-python"
-version = "0.3.6"
+version = "0.3.16"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "diskcache" },
     { name = "jinja2" },
-    { name = "numpy" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/72/cd/efae0a18621ba66baf19d19507bb17f1aa2dd5e7f0d5dd4aa008c9bc7091/llama_cpp_python-0.3.6.tar.gz", hash = "sha256:86e35a8888274466958e24201b856cd71c8def0ea72e14312be13da96c15c7a4", size = 66893177 }
+sdist = { url = "https://files.pythonhosted.org/packages/e4/b4/c8cd17629ced0b9644a71d399a91145aedef109c0333443bef015e45b704/llama_cpp_python-0.3.16.tar.gz", hash = "sha256:34ed0f9bd9431af045bb63d9324ae620ad0536653740e9bb163a2e1fcb973be6", size = 50688636, upload-time = "2025-08-15T04:58:29.212Z" }
 
 [[package]]
-name = "lm-format-enforcer"
-version = "0.10.1"
+name = "llguidance"
+version = "1.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/14/d2/e7f7176a04a4fe17277fc0a467477b6f41117f9217c83d8bc84065d452b5/llguidance-1.2.0.tar.gz", hash = "sha256:124aaf546381e76189deb6d0b324cf55163983a84dfa016032ba5b3d5b243290", size = 1069002, upload-time = "2025-08-12T20:07:15.939Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6c/cb/1f9672f9ed0474781c52334c756137799a219144dd2517956904de51cd4e/llguidance-1.2.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:f6831a232afe79303bf41e4e73f9e715e8fe201bed00c896ddd8e20b3ff245e3", size = 3227696, upload-time = "2025-08-12T20:07:14.424Z" },
+    { url = "https://files.pythonhosted.org/packages/30/e4/f0b3c16d936448efe4334fe0f35c35d914b6a1f25dd1c1e1e1f7a053ead6/llguidance-1.2.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:475b376b52681e96432ba026aa94b736efdd4428618248a041a910c23037edf7", size = 3095669, upload-time = "2025-08-12T20:07:12.553Z" },
+    { url = "https://files.pythonhosted.org/packages/93/a4/af9e3f09b754f0e0ed87746e8409d32f08ffc2aa6a70b7fa04f9393098be/llguidance-1.2.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:52102a6e65d7fcd08214d6392653b813f8c4e15736da46738c304dc3a5ea5ab4", size = 2815374, upload-time = "2025-08-12T20:07:06.984Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/04/34c2766bcda7df428b63e9936bf7b685a99d022884807f9f339204ffc682/llguidance-1.2.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a06356cf3315562fc395c4b15d41ebae0ef6833703db3a4b3c7125447160d05b", size = 3299527, upload-time = "2025-08-12T20:07:09.091Z" },
+    { url = "https://files.pythonhosted.org/packages/00/06/e42e64e0fc1f33877e41a55556322fee17e73efb7b1b58f60d1f40d6b31e/llguidance-1.2.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c144da83cc1927b5742ac9301b6e39c5fa9e04aefbd543fa4b9fc88c28bbb78c", size = 3044367, upload-time = "2025-08-12T20:07:10.891Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/3b/fa95e6ba555b5c652f19e5bf632a90fcbd43865d8722277642873fe63bae/llguidance-1.2.0-cp39-abi3-win32.whl", hash = "sha256:d5da82f54d98e155f8cb24425cc29bba7ee4e0750daacb84f5ea845390c0d5d9", size = 2607705, upload-time = "2025-08-12T20:07:19.676Z" },
+    { url = "https://files.pythonhosted.org/packages/08/69/d41416280d457e787aa77b446269539f11ff902298375fbfc95751c0ee9d/llguidance-1.2.0-cp39-abi3-win_amd64.whl", hash = "sha256:0a985c600f7f0c6b483fdbcf6586bfd45d8412a0115755d37d3089dc4726abf5", size = 2806088, upload-time = "2025-08-12T20:07:18.011Z" },
+]
+
+[[package]]
+name = "llvmlite"
+version = "0.43.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9f/3d/f513755f285db51ab363a53e898b85562e950f79a2e6767a364530c2f645/llvmlite-0.43.0.tar.gz", hash = "sha256:ae2b5b5c3ef67354824fb75517c8db5fbe93bc02cd9671f3c62271626bc041d5", size = 157069, upload-time = "2024-06-13T18:09:32.641Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/23/ff/6ca7e98998b573b4bd6566f15c35e5c8bea829663a6df0c7aa55ab559da9/llvmlite-0.43.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:a289af9a1687c6cf463478f0fa8e8aa3b6fb813317b0d70bf1ed0759eab6f761", size = 31064408, upload-time = "2024-06-13T18:08:13.462Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/5c/a27f9257f86f0cda3f764ff21d9f4217b9f6a0d45e7a39ecfa7905f524ce/llvmlite-0.43.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6d4fd101f571a31acb1559ae1af30f30b1dc4b3186669f92ad780e17c81e91bc", size = 28793153, upload-time = "2024-06-13T18:08:17.336Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/3c/4410f670ad0a911227ea2ecfcba9f672a77cf1924df5280c4562032ec32d/llvmlite-0.43.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7d434ec7e2ce3cc8f452d1cd9a28591745de022f931d67be688a737320dfcead", size = 42857276, upload-time = "2024-06-13T18:08:21.071Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/21/2ffbab5714e72f2483207b4a1de79b2eecd9debbf666ff4e7067bcc5c134/llvmlite-0.43.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6912a87782acdff6eb8bf01675ed01d60ca1f2551f8176a300a886f09e836a6a", size = 43871781, upload-time = "2024-06-13T18:08:26.32Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/26/b5478037c453554a61625ef1125f7e12bb1429ae11c6376f47beba9b0179/llvmlite-0.43.0-cp310-cp310-win_amd64.whl", hash = "sha256:14f0e4bf2fd2d9a75a3534111e8ebeb08eda2f33e9bdd6dfa13282afacdde0ed", size = 28123487, upload-time = "2024-06-13T18:08:30.348Z" },
+    { url = "https://files.pythonhosted.org/packages/95/8c/de3276d773ab6ce3ad676df5fab5aac19696b2956319d65d7dd88fb10f19/llvmlite-0.43.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:3e8d0618cb9bfe40ac38a9633f2493d4d4e9fcc2f438d39a4e854f39cc0f5f98", size = 31064409, upload-time = "2024-06-13T18:08:34.006Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/e1/38deed89ced4cf378c61e232265cfe933ccde56ae83c901aa68b477d14b1/llvmlite-0.43.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:e0a9a1a39d4bf3517f2af9d23d479b4175ead205c592ceeb8b89af48a327ea57", size = 28793149, upload-time = "2024-06-13T18:08:37.42Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/b2/4429433eb2dc8379e2cb582502dca074c23837f8fd009907f78a24de4c25/llvmlite-0.43.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c1da416ab53e4f7f3bc8d4eeba36d801cc1894b9fbfbf2022b29b6bad34a7df2", size = 42857277, upload-time = "2024-06-13T18:08:40.822Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/99/5d00a7d671b1ba1751fc9f19d3b36f3300774c6eebe2bcdb5f6191763eb4/llvmlite-0.43.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:977525a1e5f4059316b183fb4fd34fa858c9eade31f165427a3977c95e3ee749", size = 43871781, upload-time = "2024-06-13T18:08:46.41Z" },
+    { url = "https://files.pythonhosted.org/packages/20/ab/ed5ed3688c6ba4f0b8d789da19fd8e30a9cf7fc5852effe311bc5aefe73e/llvmlite-0.43.0-cp311-cp311-win_amd64.whl", hash = "sha256:d5bd550001d26450bd90777736c69d68c487d17bf371438f975229b2b8241a91", size = 28107433, upload-time = "2024-06-13T18:08:50.834Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/67/9443509e5d2b6d8587bae3ede5598fa8bd586b1c7701696663ea8af15b5b/llvmlite-0.43.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:f99b600aa7f65235a5a05d0b9a9f31150c390f31261f2a0ba678e26823ec38f7", size = 31064409, upload-time = "2024-06-13T18:08:54.375Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/9c/24139d3712d2d352e300c39c0e00d167472c08b3bd350c3c33d72c88ff8d/llvmlite-0.43.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:35d80d61d0cda2d767f72de99450766250560399edc309da16937b93d3b676e7", size = 28793145, upload-time = "2024-06-13T18:08:57.953Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/f1/4c205a48488e574ee9f6505d50e84370a978c90f08dab41a42d8f2c576b6/llvmlite-0.43.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:eccce86bba940bae0d8d48ed925f21dbb813519169246e2ab292b5092aba121f", size = 42857276, upload-time = "2024-06-13T18:09:02.067Z" },
+    { url = "https://files.pythonhosted.org/packages/00/5f/323c4d56e8401c50185fd0e875fcf06b71bf825a863699be1eb10aa2a9cb/llvmlite-0.43.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:df6509e1507ca0760787a199d19439cc887bfd82226f5af746d6977bd9f66844", size = 43871781, upload-time = "2024-06-13T18:09:06.667Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/94/dea10e263655ce78d777e78d904903faae39d1fc440762be4a9dc46bed49/llvmlite-0.43.0-cp312-cp312-win_amd64.whl", hash = "sha256:7a2872ee80dcf6b5dbdc838763d26554c2a18aa833d31a2635bff16aafefb9c9", size = 28107442, upload-time = "2024-06-13T18:09:10.709Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/73/12925b1bbb3c2beb6d96f892ef5b4d742c34f00ddb9f4a125e9e87b22f52/llvmlite-0.43.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:9cd2a7376f7b3367019b664c21f0c61766219faa3b03731113ead75107f3b66c", size = 31064410, upload-time = "2024-06-13T18:09:14.091Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/61/58c70aa0808a8cba825a7d98cc65bef4801b99328fba80837bfcb5fc767f/llvmlite-0.43.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:18e9953c748b105668487b7c81a3e97b046d8abf95c4ddc0cd3c94f4e4651ae8", size = 28793145, upload-time = "2024-06-13T18:09:17.531Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/c6/9324eb5de2ba9d99cbed853d85ba7a318652a48e077797bec27cf40f911d/llvmlite-0.43.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:74937acd22dc11b33946b67dca7680e6d103d6e90eeaaaf932603bec6fe7b03a", size = 42857276, upload-time = "2024-06-13T18:09:21.377Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/d0/889e9705107db7b1ec0767b03f15d7b95b4c4f9fdf91928ab1c7e9ffacf6/llvmlite-0.43.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc9efc739cc6ed760f795806f67889923f7274276f0eb45092a1473e40d9b867", size = 43871777, upload-time = "2024-06-13T18:09:25.76Z" },
+    { url = "https://files.pythonhosted.org/packages/df/41/73cc26a2634b538cfe813f618c91e7e9960b8c163f8f0c94a2b0f008b9da/llvmlite-0.43.0-cp39-cp39-win_amd64.whl", hash = "sha256:47e147cdda9037f94b399bf03bfd8a6b6b1f2f90be94a454e3386f006455a9b4", size = 28123489, upload-time = "2024-06-13T18:09:29.78Z" },
+]
+
+[[package]]
+name = "llvmlite"
+version = "0.44.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/89/6a/95a3d3610d5c75293d5dbbb2a76480d5d4eeba641557b69fe90af6c5b84e/llvmlite-0.44.0.tar.gz", hash = "sha256:07667d66a5d150abed9157ab6c0b9393c9356f229784a4385c02f99e94fc94d4", size = 171880, upload-time = "2025-01-20T11:14:41.342Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/41/75/d4863ddfd8ab5f6e70f4504cf8cc37f4e986ec6910f4ef8502bb7d3c1c71/llvmlite-0.44.0-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:9fbadbfba8422123bab5535b293da1cf72f9f478a65645ecd73e781f962ca614", size = 28132306, upload-time = "2025-01-20T11:12:18.634Z" },
+    { url = "https://files.pythonhosted.org/packages/37/d9/6e8943e1515d2f1003e8278819ec03e4e653e2eeb71e4d00de6cfe59424e/llvmlite-0.44.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:cccf8eb28f24840f2689fb1a45f9c0f7e582dd24e088dcf96e424834af11f791", size = 26201096, upload-time = "2025-01-20T11:12:24.544Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/46/8ffbc114def88cc698906bf5acab54ca9fdf9214fe04aed0e71731fb3688/llvmlite-0.44.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7202b678cdf904823c764ee0fe2dfe38a76981f4c1e51715b4cb5abb6cf1d9e8", size = 42361859, upload-time = "2025-01-20T11:12:31.839Z" },
+    { url = "https://files.pythonhosted.org/packages/30/1c/9366b29ab050a726af13ebaae8d0dff00c3c58562261c79c635ad4f5eb71/llvmlite-0.44.0-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:40526fb5e313d7b96bda4cbb2c85cd5374e04d80732dd36a282d72a560bb6408", size = 41184199, upload-time = "2025-01-20T11:12:40.049Z" },
+    { url = "https://files.pythonhosted.org/packages/69/07/35e7c594b021ecb1938540f5bce543ddd8713cff97f71d81f021221edc1b/llvmlite-0.44.0-cp310-cp310-win_amd64.whl", hash = "sha256:41e3839150db4330e1b2716c0be3b5c4672525b4c9005e17c7597f835f351ce2", size = 30332381, upload-time = "2025-01-20T11:12:47.054Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/e2/86b245397052386595ad726f9742e5223d7aea999b18c518a50e96c3aca4/llvmlite-0.44.0-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:eed7d5f29136bda63b6d7804c279e2b72e08c952b7c5df61f45db408e0ee52f3", size = 28132305, upload-time = "2025-01-20T11:12:53.936Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/ec/506902dc6870249fbe2466d9cf66d531265d0f3a1157213c8f986250c033/llvmlite-0.44.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ace564d9fa44bb91eb6e6d8e7754977783c68e90a471ea7ce913bff30bd62427", size = 26201090, upload-time = "2025-01-20T11:12:59.847Z" },
+    { url = "https://files.pythonhosted.org/packages/99/fe/d030f1849ebb1f394bb3f7adad5e729b634fb100515594aca25c354ffc62/llvmlite-0.44.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c5d22c3bfc842668168a786af4205ec8e3ad29fb1bc03fd11fd48460d0df64c1", size = 42361858, upload-time = "2025-01-20T11:13:07.623Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/7a/ce6174664b9077fc673d172e4c888cb0b128e707e306bc33fff8c2035f0d/llvmlite-0.44.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f01a394e9c9b7b1d4e63c327b096d10f6f0ed149ef53d38a09b3749dcf8c9610", size = 41184200, upload-time = "2025-01-20T11:13:20.058Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/c6/258801143975a6d09a373f2641237992496e15567b907a4d401839d671b8/llvmlite-0.44.0-cp311-cp311-win_amd64.whl", hash = "sha256:d8489634d43c20cd0ad71330dde1d5bc7b9966937a263ff1ec1cebb90dc50955", size = 30331193, upload-time = "2025-01-20T11:13:26.976Z" },
+    { url = "https://files.pythonhosted.org/packages/15/86/e3c3195b92e6e492458f16d233e58a1a812aa2bfbef9bdd0fbafcec85c60/llvmlite-0.44.0-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:1d671a56acf725bf1b531d5ef76b86660a5ab8ef19bb6a46064a705c6ca80aad", size = 28132297, upload-time = "2025-01-20T11:13:32.57Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/53/373b6b8be67b9221d12b24125fd0ec56b1078b660eeae266ec388a6ac9a0/llvmlite-0.44.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:5f79a728e0435493611c9f405168682bb75ffd1fbe6fc360733b850c80a026db", size = 26201105, upload-time = "2025-01-20T11:13:38.744Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/da/8341fd3056419441286c8e26bf436923021005ece0bff5f41906476ae514/llvmlite-0.44.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0143a5ef336da14deaa8ec26c5449ad5b6a2b564df82fcef4be040b9cacfea9", size = 42361901, upload-time = "2025-01-20T11:13:46.711Z" },
+    { url = "https://files.pythonhosted.org/packages/53/ad/d79349dc07b8a395a99153d7ce8b01d6fcdc9f8231355a5df55ded649b61/llvmlite-0.44.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d752f89e31b66db6f8da06df8b39f9b91e78c5feea1bf9e8c1fba1d1c24c065d", size = 41184247, upload-time = "2025-01-20T11:13:56.159Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/3b/a9a17366af80127bd09decbe2a54d8974b6d8b274b39bf47fbaedeec6307/llvmlite-0.44.0-cp312-cp312-win_amd64.whl", hash = "sha256:eae7e2d4ca8f88f89d315b48c6b741dcb925d6a1042da694aa16ab3dd4cbd3a1", size = 30332380, upload-time = "2025-01-20T11:14:02.442Z" },
+]
+
+[[package]]
+name = "markdown"
+version = "3.8.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "interegular" },
-    { name = "packaging" },
-    { name = "pydantic" },
-    { name = "pyyaml" },
+    { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/bb/ac/11fb977637125ed92adb1de4138b0e365918465e3e4e39a437fdb8156746/lm_format_enforcer-0.10.1.tar.gz", hash = "sha256:23e65a4199714fca348063e8c906838622619f905a673c4d6d428eee7e7d2095", size = 38781 }
+sdist = { url = "https://files.pythonhosted.org/packages/d7/c2/4ab49206c17f75cb08d6311171f2d65798988db4360c4d1485bd0eedd67c/markdown-3.8.2.tar.gz", hash = "sha256:247b9a70dd12e27f67431ce62523e675b866d254f900c4fe75ce3dda62237c45", size = 362071, upload-time = "2025-06-19T17:12:44.483Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/2d/fd/e21d1f0608da7696746c5d4cdf2f944d714daf63a97def416e0f3c154f98/lm_format_enforcer-0.10.1-py3-none-any.whl", hash = "sha256:5520004af248d787930327ead052aeff75e21fad595f388e5eade9f062ffddda", size = 42929 },
+    { url = "https://files.pythonhosted.org/packages/96/2b/34cc11786bc00d0f04d0f5fdc3a2b1ae0b6239eef72d3d345805f9ad92a1/markdown-3.8.2-py3-none-any.whl", hash = "sha256:5c83764dbd4e00bdd94d85a19b8d55ccca20fe35b2e678a1422b380324dd5f24", size = 106827, upload-time = "2025-06-19T17:12:42.994Z" },
 ]
 
 [[package]]
 name = "markdown-it-py"
 version = "3.0.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "mdurl", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/38/71/3b932df36c1a044d397a1f92d1cf91ee0a503d91e470cbd670aa66b07ed0/markdown-it-py-3.0.0.tar.gz", hash = "sha256:e3f60a94fa066dc52ec76661e37c851cb232d92f9886b15cb560aaada2df8feb", size = 74596, upload-time = "2023-06-03T06:41:14.443Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/42/d7/1ec15b46af6af88f19b8e5ffea08fa375d433c998b8a7639e76935c14f1f/markdown_it_py-3.0.0-py3-none-any.whl", hash = "sha256:355216845c60bd96232cd8d8c40e8f9765cc86f46880e43a8fd22dc1a1a8cab1", size = 87528, upload-time = "2023-06-03T06:41:11.019Z" },
+]
+
+[[package]]
+name = "markdown-it-py"
+version = "4.0.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
 dependencies = [
-    { name = "mdurl" },
+    { name = "mdurl", marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/5b/f5/4ec618ed16cc4f8fb3b701563655a69816155e79e24a17b651541804721d/markdown_it_py-4.0.0.tar.gz", hash = "sha256:cb0a2b4aa34f932c007117b194e945bd74e0ec24133ceb5bac59009cda1cb9f3", size = 73070, upload-time = "2025-08-11T12:57:52.854Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/94/54/e7d793b573f298e1c9013b8c4dade17d481164aa517d1d7148619c2cedbf/markdown_it_py-4.0.0-py3-none-any.whl", hash = "sha256:87327c59b172c5011896038353a81343b6754500a08cd7a4973bb48c6d578147", size = 87321, upload-time = "2025-08-11T12:57:51.923Z" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/38/71/3b932df36c1a044d397a1f92d1cf91ee0a503d91e470cbd670aa66b07ed0/markdown-it-py-3.0.0.tar.gz", hash = "sha256:e3f60a94fa066dc52ec76661e37c851cb232d92f9886b15cb560aaada2df8feb", size = 74596 }
+
+[[package]]
+name = "markdown2"
+version = "2.5.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/42/f8/b2ae8bf5f28f9b510ae097415e6e4cb63226bb28d7ee01aec03a755ba03b/markdown2-2.5.4.tar.gz", hash = "sha256:a09873f0b3c23dbfae589b0080587df52ad75bb09a5fa6559147554736676889", size = 145652, upload-time = "2025-07-27T16:16:24.307Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/42/d7/1ec15b46af6af88f19b8e5ffea08fa375d433c998b8a7639e76935c14f1f/markdown_it_py-3.0.0-py3-none-any.whl", hash = "sha256:355216845c60bd96232cd8d8c40e8f9765cc86f46880e43a8fd22dc1a1a8cab1", size = 87528 },
+    { url = "https://files.pythonhosted.org/packages/b8/06/2697b5043c3ecb720ce0d243fc7cf5024c0b5b1e450506e9b21939019963/markdown2-2.5.4-py3-none-any.whl", hash = "sha256:3c4b2934e677be7fec0e6f2de4410e116681f4ad50ec8e5ba7557be506d3f439", size = 49954, upload-time = "2025-07-27T16:16:23.026Z" },
+]
+
+[package.optional-dependencies]
+all = [
+    { name = "latex2mathml" },
+    { name = "pygments" },
+    { name = "wavedrom" },
 ]
 
 [[package]]
 name = "markupsafe"
 version = "3.0.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/b2/97/5d42485e71dfc078108a86d6de8fa46db44a1a9295e89c5d6d4a06e23a62/markupsafe-3.0.2.tar.gz", hash = "sha256:ee55d3edf80167e48ea11a923c7386f4669df67d7994554387f84e7d8b0a2bf0", size = 20537 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/04/90/d08277ce111dd22f77149fd1a5d4653eeb3b3eaacbdfcbae5afb2600eebd/MarkupSafe-3.0.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7e94c425039cde14257288fd61dcfb01963e658efbc0ff54f5306b06054700f8", size = 14357 },
-    { url = "https://files.pythonhosted.org/packages/04/e1/6e2194baeae0bca1fae6629dc0cbbb968d4d941469cbab11a3872edff374/MarkupSafe-3.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9e2d922824181480953426608b81967de705c3cef4d1af983af849d7bd619158", size = 12393 },
-    { url = "https://files.pythonhosted.org/packages/1d/69/35fa85a8ece0a437493dc61ce0bb6d459dcba482c34197e3efc829aa357f/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:38a9ef736c01fccdd6600705b09dc574584b89bea478200c5fbf112a6b0d5579", size = 21732 },
-    { url = "https://files.pythonhosted.org/packages/22/35/137da042dfb4720b638d2937c38a9c2df83fe32d20e8c8f3185dbfef05f7/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bbcb445fa71794da8f178f0f6d66789a28d7319071af7a496d4d507ed566270d", size = 20866 },
-    { url = "https://files.pythonhosted.org/packages/29/28/6d029a903727a1b62edb51863232152fd335d602def598dade38996887f0/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:57cb5a3cf367aeb1d316576250f65edec5bb3be939e9247ae594b4bcbc317dfb", size = 20964 },
-    { url = "https://files.pythonhosted.org/packages/cc/cd/07438f95f83e8bc028279909d9c9bd39e24149b0d60053a97b2bc4f8aa51/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:3809ede931876f5b2ec92eef964286840ed3540dadf803dd570c3b7e13141a3b", size = 21977 },
-    { url = "https://files.pythonhosted.org/packages/29/01/84b57395b4cc062f9c4c55ce0df7d3108ca32397299d9df00fedd9117d3d/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e07c3764494e3776c602c1e78e298937c3315ccc9043ead7e685b7f2b8d47b3c", size = 21366 },
-    { url = "https://files.pythonhosted.org/packages/bd/6e/61ebf08d8940553afff20d1fb1ba7294b6f8d279df9fd0c0db911b4bbcfd/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b424c77b206d63d500bcb69fa55ed8d0e6a3774056bdc4839fc9298a7edca171", size = 21091 },
-    { url = "https://files.pythonhosted.org/packages/11/23/ffbf53694e8c94ebd1e7e491de185124277964344733c45481f32ede2499/MarkupSafe-3.0.2-cp310-cp310-win32.whl", hash = "sha256:fcabf5ff6eea076f859677f5f0b6b5c1a51e70a376b0579e0eadef8db48c6b50", size = 15065 },
-    { url = "https://files.pythonhosted.org/packages/44/06/e7175d06dd6e9172d4a69a72592cb3f7a996a9c396eee29082826449bbc3/MarkupSafe-3.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:6af100e168aa82a50e186c82875a5893c5597a0c1ccdb0d8b40240b1f28b969a", size = 15514 },
-    { url = "https://files.pythonhosted.org/packages/6b/28/bbf83e3f76936960b850435576dd5e67034e200469571be53f69174a2dfd/MarkupSafe-3.0.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:9025b4018f3a1314059769c7bf15441064b2207cb3f065e6ea1e7359cb46db9d", size = 14353 },
-    { url = "https://files.pythonhosted.org/packages/6c/30/316d194b093cde57d448a4c3209f22e3046c5bb2fb0820b118292b334be7/MarkupSafe-3.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:93335ca3812df2f366e80509ae119189886b0f3c2b81325d39efdb84a1e2ae93", size = 12392 },
-    { url = "https://files.pythonhosted.org/packages/f2/96/9cdafba8445d3a53cae530aaf83c38ec64c4d5427d975c974084af5bc5d2/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2cb8438c3cbb25e220c2ab33bb226559e7afb3baec11c4f218ffa7308603c832", size = 23984 },
-    { url = "https://files.pythonhosted.org/packages/f1/a4/aefb044a2cd8d7334c8a47d3fb2c9f328ac48cb349468cc31c20b539305f/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a123e330ef0853c6e822384873bef7507557d8e4a082961e1defa947aa59ba84", size = 23120 },
-    { url = "https://files.pythonhosted.org/packages/8d/21/5e4851379f88f3fad1de30361db501300d4f07bcad047d3cb0449fc51f8c/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1e084f686b92e5b83186b07e8a17fc09e38fff551f3602b249881fec658d3eca", size = 23032 },
-    { url = "https://files.pythonhosted.org/packages/00/7b/e92c64e079b2d0d7ddf69899c98842f3f9a60a1ae72657c89ce2655c999d/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d8213e09c917a951de9d09ecee036d5c7d36cb6cb7dbaece4c71a60d79fb9798", size = 24057 },
-    { url = "https://files.pythonhosted.org/packages/f9/ac/46f960ca323037caa0a10662ef97d0a4728e890334fc156b9f9e52bcc4ca/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:5b02fb34468b6aaa40dfc198d813a641e3a63b98c2b05a16b9f80b7ec314185e", size = 23359 },
-    { url = "https://files.pythonhosted.org/packages/69/84/83439e16197337b8b14b6a5b9c2105fff81d42c2a7c5b58ac7b62ee2c3b1/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0bff5e0ae4ef2e1ae4fdf2dfd5b76c75e5c2fa4132d05fc1b0dabcd20c7e28c4", size = 23306 },
-    { url = "https://files.pythonhosted.org/packages/9a/34/a15aa69f01e2181ed8d2b685c0d2f6655d5cca2c4db0ddea775e631918cd/MarkupSafe-3.0.2-cp311-cp311-win32.whl", hash = "sha256:6c89876f41da747c8d3677a2b540fb32ef5715f97b66eeb0c6b66f5e3ef6f59d", size = 15094 },
-    { url = "https://files.pythonhosted.org/packages/da/b8/3a3bd761922d416f3dc5d00bfbed11f66b1ab89a0c2b6e887240a30b0f6b/MarkupSafe-3.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:70a87b411535ccad5ef2f1df5136506a10775d267e197e4cf531ced10537bd6b", size = 15521 },
-    { url = "https://files.pythonhosted.org/packages/22/09/d1f21434c97fc42f09d290cbb6350d44eb12f09cc62c9476effdb33a18aa/MarkupSafe-3.0.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:9778bd8ab0a994ebf6f84c2b949e65736d5575320a17ae8984a77fab08db94cf", size = 14274 },
-    { url = "https://files.pythonhosted.org/packages/6b/b0/18f76bba336fa5aecf79d45dcd6c806c280ec44538b3c13671d49099fdd0/MarkupSafe-3.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:846ade7b71e3536c4e56b386c2a47adf5741d2d8b94ec9dc3e92e5e1ee1e2225", size = 12348 },
-    { url = "https://files.pythonhosted.org/packages/e0/25/dd5c0f6ac1311e9b40f4af06c78efde0f3b5cbf02502f8ef9501294c425b/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1c99d261bd2d5f6b59325c92c73df481e05e57f19837bdca8413b9eac4bd8028", size = 24149 },
-    { url = "https://files.pythonhosted.org/packages/f3/f0/89e7aadfb3749d0f52234a0c8c7867877876e0a20b60e2188e9850794c17/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e17c96c14e19278594aa4841ec148115f9c7615a47382ecb6b82bd8fea3ab0c8", size = 23118 },
-    { url = "https://files.pythonhosted.org/packages/d5/da/f2eeb64c723f5e3777bc081da884b414671982008c47dcc1873d81f625b6/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:88416bd1e65dcea10bc7569faacb2c20ce071dd1f87539ca2ab364bf6231393c", size = 22993 },
-    { url = "https://files.pythonhosted.org/packages/da/0e/1f32af846df486dce7c227fe0f2398dc7e2e51d4a370508281f3c1c5cddc/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2181e67807fc2fa785d0592dc2d6206c019b9502410671cc905d132a92866557", size = 24178 },
-    { url = "https://files.pythonhosted.org/packages/c4/f6/bb3ca0532de8086cbff5f06d137064c8410d10779c4c127e0e47d17c0b71/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:52305740fe773d09cffb16f8ed0427942901f00adedac82ec8b67752f58a1b22", size = 23319 },
-    { url = "https://files.pythonhosted.org/packages/a2/82/8be4c96ffee03c5b4a034e60a31294daf481e12c7c43ab8e34a1453ee48b/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ad10d3ded218f1039f11a75f8091880239651b52e9bb592ca27de44eed242a48", size = 23352 },
-    { url = "https://files.pythonhosted.org/packages/51/ae/97827349d3fcffee7e184bdf7f41cd6b88d9919c80f0263ba7acd1bbcb18/MarkupSafe-3.0.2-cp312-cp312-win32.whl", hash = "sha256:0f4ca02bea9a23221c0182836703cbf8930c5e9454bacce27e767509fa286a30", size = 15097 },
-    { url = "https://files.pythonhosted.org/packages/c1/80/a61f99dc3a936413c3ee4e1eecac96c0da5ed07ad56fd975f1a9da5bc630/MarkupSafe-3.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:8e06879fc22a25ca47312fbe7c8264eb0b662f6db27cb2d3bbbc74b1df4b9b87", size = 15601 },
-    { url = "https://files.pythonhosted.org/packages/83/0e/67eb10a7ecc77a0c2bbe2b0235765b98d164d81600746914bebada795e97/MarkupSafe-3.0.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ba9527cdd4c926ed0760bc301f6728ef34d841f405abf9d4f959c478421e4efd", size = 14274 },
-    { url = "https://files.pythonhosted.org/packages/2b/6d/9409f3684d3335375d04e5f05744dfe7e9f120062c9857df4ab490a1031a/MarkupSafe-3.0.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f8b3d067f2e40fe93e1ccdd6b2e1d16c43140e76f02fb1319a05cf2b79d99430", size = 12352 },
-    { url = "https://files.pythonhosted.org/packages/d2/f5/6eadfcd3885ea85fe2a7c128315cc1bb7241e1987443d78c8fe712d03091/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:569511d3b58c8791ab4c2e1285575265991e6d8f8700c7be0e88f86cb0672094", size = 24122 },
-    { url = "https://files.pythonhosted.org/packages/0c/91/96cf928db8236f1bfab6ce15ad070dfdd02ed88261c2afafd4b43575e9e9/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:15ab75ef81add55874e7ab7055e9c397312385bd9ced94920f2802310c930396", size = 23085 },
-    { url = "https://files.pythonhosted.org/packages/c2/cf/c9d56af24d56ea04daae7ac0940232d31d5a8354f2b457c6d856b2057d69/MarkupSafe-3.0.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f3818cb119498c0678015754eba762e0d61e5b52d34c8b13d770f0719f7b1d79", size = 22978 },
-    { url = "https://files.pythonhosted.org/packages/2a/9f/8619835cd6a711d6272d62abb78c033bda638fdc54c4e7f4272cf1c0962b/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:cdb82a876c47801bb54a690c5ae105a46b392ac6099881cdfb9f6e95e4014c6a", size = 24208 },
-    { url = "https://files.pythonhosted.org/packages/f9/bf/176950a1792b2cd2102b8ffeb5133e1ed984547b75db47c25a67d3359f77/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:cabc348d87e913db6ab4aa100f01b08f481097838bdddf7c7a84b7575b7309ca", size = 23357 },
-    { url = "https://files.pythonhosted.org/packages/ce/4f/9a02c1d335caabe5c4efb90e1b6e8ee944aa245c1aaaab8e8a618987d816/MarkupSafe-3.0.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:444dcda765c8a838eaae23112db52f1efaf750daddb2d9ca300bcae1039adc5c", size = 23344 },
-    { url = "https://files.pythonhosted.org/packages/ee/55/c271b57db36f748f0e04a759ace9f8f759ccf22b4960c270c78a394f58be/MarkupSafe-3.0.2-cp313-cp313-win32.whl", hash = "sha256:bcf3e58998965654fdaff38e58584d8937aa3096ab5354d493c77d1fdd66d7a1", size = 15101 },
-    { url = "https://files.pythonhosted.org/packages/29/88/07df22d2dd4df40aba9f3e402e6dc1b8ee86297dddbad4872bd5e7b0094f/MarkupSafe-3.0.2-cp313-cp313-win_amd64.whl", hash = "sha256:e6a2a455bd412959b57a172ce6328d2dd1f01cb2135efda2e4576e8a23fa3b0f", size = 15603 },
-    { url = "https://files.pythonhosted.org/packages/62/6a/8b89d24db2d32d433dffcd6a8779159da109842434f1dd2f6e71f32f738c/MarkupSafe-3.0.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:b5a6b3ada725cea8a5e634536b1b01c30bcdcd7f9c6fff4151548d5bf6b3a36c", size = 14510 },
-    { url = "https://files.pythonhosted.org/packages/7a/06/a10f955f70a2e5a9bf78d11a161029d278eeacbd35ef806c3fd17b13060d/MarkupSafe-3.0.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:a904af0a6162c73e3edcb969eeeb53a63ceeb5d8cf642fade7d39e7963a22ddb", size = 12486 },
-    { url = "https://files.pythonhosted.org/packages/34/cf/65d4a571869a1a9078198ca28f39fba5fbb910f952f9dbc5220afff9f5e6/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4aa4e5faecf353ed117801a068ebab7b7e09ffb6e1d5e412dc852e0da018126c", size = 25480 },
-    { url = "https://files.pythonhosted.org/packages/0c/e3/90e9651924c430b885468b56b3d597cabf6d72be4b24a0acd1fa0e12af67/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0ef13eaeee5b615fb07c9a7dadb38eac06a0608b41570d8ade51c56539e509d", size = 23914 },
-    { url = "https://files.pythonhosted.org/packages/66/8c/6c7cf61f95d63bb866db39085150df1f2a5bd3335298f14a66b48e92659c/MarkupSafe-3.0.2-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d16a81a06776313e817c951135cf7340a3e91e8c1ff2fac444cfd75fffa04afe", size = 23796 },
-    { url = "https://files.pythonhosted.org/packages/bb/35/cbe9238ec3f47ac9a7c8b3df7a808e7cb50fe149dc7039f5f454b3fba218/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:6381026f158fdb7c72a168278597a5e3a5222e83ea18f543112b2662a9b699c5", size = 25473 },
-    { url = "https://files.pythonhosted.org/packages/e6/32/7621a4382488aa283cc05e8984a9c219abad3bca087be9ec77e89939ded9/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:3d79d162e7be8f996986c064d1c7c817f6df3a77fe3d6859f6f9e7be4b8c213a", size = 24114 },
-    { url = "https://files.pythonhosted.org/packages/0d/80/0985960e4b89922cb5a0bac0ed39c5b96cbc1a536a99f30e8c220a996ed9/MarkupSafe-3.0.2-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:131a3c7689c85f5ad20f9f6fb1b866f402c445b220c19fe4308c0b147ccd2ad9", size = 24098 },
-    { url = "https://files.pythonhosted.org/packages/82/78/fedb03c7d5380df2427038ec8d973587e90561b2d90cd472ce9254cf348b/MarkupSafe-3.0.2-cp313-cp313t-win32.whl", hash = "sha256:ba8062ed2cf21c07a9e295d5b8a2a5ce678b913b45fdf68c32d95d6c1291e0b6", size = 15208 },
-    { url = "https://files.pythonhosted.org/packages/4f/65/6079a46068dfceaeabb5dcad6d674f5f5c61a6fa5673746f42a9f4c233b3/MarkupSafe-3.0.2-cp313-cp313t-win_amd64.whl", hash = "sha256:e444a31f8db13eb18ada366ab3cf45fd4b31e4db1236a4448f68778c1d1a5a2f", size = 15739 },
-    { url = "https://files.pythonhosted.org/packages/a7/ea/9b1530c3fdeeca613faeb0fb5cbcf2389d816072fab72a71b45749ef6062/MarkupSafe-3.0.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:eaa0a10b7f72326f1372a713e73c3f739b524b3af41feb43e4921cb529f5929a", size = 14344 },
-    { url = "https://files.pythonhosted.org/packages/4b/c2/fbdbfe48848e7112ab05e627e718e854d20192b674952d9042ebd8c9e5de/MarkupSafe-3.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:48032821bbdf20f5799ff537c7ac3d1fba0ba032cfc06194faffa8cda8b560ff", size = 12389 },
-    { url = "https://files.pythonhosted.org/packages/f0/25/7a7c6e4dbd4f867d95d94ca15449e91e52856f6ed1905d58ef1de5e211d0/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1a9d3f5f0901fdec14d8d2f66ef7d035f2157240a433441719ac9a3fba440b13", size = 21607 },
-    { url = "https://files.pythonhosted.org/packages/53/8f/f339c98a178f3c1e545622206b40986a4c3307fe39f70ccd3d9df9a9e425/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88b49a3b9ff31e19998750c38e030fc7bb937398b1f78cfa599aaef92d693144", size = 20728 },
-    { url = "https://files.pythonhosted.org/packages/1a/03/8496a1a78308456dbd50b23a385c69b41f2e9661c67ea1329849a598a8f9/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:cfad01eed2c2e0c01fd0ecd2ef42c492f7f93902e39a42fc9ee1692961443a29", size = 20826 },
-    { url = "https://files.pythonhosted.org/packages/e6/cf/0a490a4bd363048c3022f2f475c8c05582179bb179defcee4766fb3dcc18/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1225beacc926f536dc82e45f8a4d68502949dc67eea90eab715dea3a21c1b5f0", size = 21843 },
-    { url = "https://files.pythonhosted.org/packages/19/a3/34187a78613920dfd3cdf68ef6ce5e99c4f3417f035694074beb8848cd77/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:3169b1eefae027567d1ce6ee7cae382c57fe26e82775f460f0b2778beaad66c0", size = 21219 },
-    { url = "https://files.pythonhosted.org/packages/17/d8/5811082f85bb88410ad7e452263af048d685669bbbfb7b595e8689152498/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:eb7972a85c54febfb25b5c4b4f3af4dcc731994c7da0d8a0b4a6eb0640e1d178", size = 20946 },
-    { url = "https://files.pythonhosted.org/packages/7c/31/bd635fb5989440d9365c5e3c47556cfea121c7803f5034ac843e8f37c2f2/MarkupSafe-3.0.2-cp39-cp39-win32.whl", hash = "sha256:8c4e8c3ce11e1f92f6536ff07154f9d49677ebaaafc32db9db4620bc11ed480f", size = 15063 },
-    { url = "https://files.pythonhosted.org/packages/b3/73/085399401383ce949f727afec55ec3abd76648d04b9f22e1c0e99cb4bec3/MarkupSafe-3.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:6e296a513ca3d94054c2c881cc913116e90fd030ad1c656b3869762b754f5f8a", size = 15506 },
+sdist = { url = "https://files.pythonhosted.org/packages/b2/97/5d42485e71dfc078108a86d6de8fa46db44a1a9295e89c5d6d4a06e23a62/markupsafe-3.0.2.tar.gz", hash = "sha256:ee55d3edf80167e48ea11a923c7386f4669df67d7994554387f84e7d8b0a2bf0", size = 20537, upload-time = "2024-10-18T15:21:54.129Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/04/90/d08277ce111dd22f77149fd1a5d4653eeb3b3eaacbdfcbae5afb2600eebd/MarkupSafe-3.0.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7e94c425039cde14257288fd61dcfb01963e658efbc0ff54f5306b06054700f8", size = 14357, upload-time = "2024-10-18T15:20:51.44Z" },
+    { url = "https://files.pythonhosted.org/packages/04/e1/6e2194baeae0bca1fae6629dc0cbbb968d4d941469cbab11a3872edff374/MarkupSafe-3.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9e2d922824181480953426608b81967de705c3cef4d1af983af849d7bd619158", size = 12393, upload-time = "2024-10-18T15:20:52.426Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/69/35fa85a8ece0a437493dc61ce0bb6d459dcba482c34197e3efc829aa357f/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:38a9ef736c01fccdd6600705b09dc574584b89bea478200c5fbf112a6b0d5579", size = 21732, upload-time = "2024-10-18T15:20:53.578Z" },
+    { url = "https://files.pythonhosted.org/packages/22/35/137da042dfb4720b638d2937c38a9c2df83fe32d20e8c8f3185dbfef05f7/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bbcb445fa71794da8f178f0f6d66789a28d7319071af7a496d4d507ed566270d", size = 20866, upload-time = "2024-10-18T15:20:55.06Z" },
+    { url = "https://files.pythonhosted.org/packages/29/28/6d029a903727a1b62edb51863232152fd335d602def598dade38996887f0/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:57cb5a3cf367aeb1d316576250f65edec5bb3be939e9247ae594b4bcbc317dfb", size = 20964, upload-time = "2024-10-18T15:20:55.906Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/cd/07438f95f83e8bc028279909d9c9bd39e24149b0d60053a97b2bc4f8aa51/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:3809ede931876f5b2ec92eef964286840ed3540dadf803dd570c3b7e13141a3b", size = 21977, upload-time = "2024-10-18T15:20:57.189Z" },
+    { url = "https://files.pythonhosted.org/packages/29/01/84b57395b4cc062f9c4c55ce0df7d3108ca32397299d9df00fedd9117d3d/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e07c3764494e3776c602c1e78e298937c3315ccc9043ead7e685b7f2b8d47b3c", size = 21366, upload-time = "2024-10-18T15:20:58.235Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/6e/61ebf08d8940553afff20d1fb1ba7294b6f8d279df9fd0c0db911b4bbcfd/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b424c77b206d63d500bcb69fa55ed8d0e6a3774056bdc4839fc9298a7edca171", size = 21091, upload-time = "2024-10-18T15:20:59.235Z" },
+    { url = "https://files.pythonhosted.org/packages/11/23/ffbf53694e8c94ebd1e7e491de185124277964344733c45481f32ede2499/MarkupSafe-3.0.2-cp310-cp310-win32.whl", hash = "sha256:fcabf5ff6eea076f859677f5f0b6b5c1a51e70a376b0579e0eadef8db48c6b50", size = 15065, upload-time = "2024-10-18T15:21:00.307Z" },
+    { url = "https://files.pythonhosted.org/packages/44/06/e7175d06dd6e9172d4a69a72592cb3f7a996a9c396eee29082826449bbc3/MarkupSafe-3.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:6af100e168aa82a50e186c82875a5893c5597a0c1ccdb0d8b40240b1f28b969a", size = 15514, upload-time = "2024-10-18T15:21:01.122Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/28/bbf83e3f76936960b850435576dd5e67034e200469571be53f69174a2dfd/MarkupSafe-3.0.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:9025b4018f3a1314059769c7bf15441064b2207cb3f065e6ea1e7359cb46db9d", size = 14353, upload-time = "2024-10-18T15:21:02.187Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/30/316d194b093cde57d448a4c3209f22e3046c5bb2fb0820b118292b334be7/MarkupSafe-3.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:93335ca3812df2f366e80509ae119189886b0f3c2b81325d39efdb84a1e2ae93", size = 12392, upload-time = "2024-10-18T15:21:02.941Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/96/9cdafba8445d3a53cae530aaf83c38ec64c4d5427d975c974084af5bc5d2/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2cb8438c3cbb25e220c2ab33bb226559e7afb3baec11c4f218ffa7308603c832", size = 23984, upload-time = "2024-10-18T15:21:03.953Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/a4/aefb044a2cd8d7334c8a47d3fb2c9f328ac48cb349468cc31c20b539305f/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a123e330ef0853c6e822384873bef7507557d8e4a082961e1defa947aa59ba84", size = 23120, upload-time = "2024-10-18T15:21:06.495Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/21/5e4851379f88f3fad1de30361db501300d4f07bcad047d3cb0449fc51f8c/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1e084f686b92e5b83186b07e8a17fc09e38fff551f3602b249881fec658d3eca", size = 23032, upload-time = "2024-10-18T15:21:07.295Z" },
+    { url = "https://files.pythonhosted.org/packages/00/7b/e92c64e079b2d0d7ddf69899c98842f3f9a60a1ae72657c89ce2655c999d/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d8213e09c917a951de9d09ecee036d5c7d36cb6cb7dbaece4c71a60d79fb9798", size = 24057, upload-time = "2024-10-18T15:21:08.073Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/ac/46f960ca323037caa0a10662ef97d0a4728e890334fc156b9f9e52bcc4ca/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:5b02fb34468b6aaa40dfc198d813a641e3a63b98c2b05a16b9f80b7ec314185e", size = 23359, upload-time = "2024-10-18T15:21:09.318Z" },
+    { url = "https://files.pythonhosted.org/packages/69/84/83439e16197337b8b14b6a5b9c2105fff81d42c2a7c5b58ac7b62ee2c3b1/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0bff5e0ae4ef2e1ae4fdf2dfd5b76c75e5c2fa4132d05fc1b0dabcd20c7e28c4", size = 23306, upload-time = "2024-10-18T15:21:10.185Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/34/a15aa69f01e2181ed8d2b685c0d2f6655d5cca2c4db0ddea775e631918cd/MarkupSafe-3.0.2-cp311-cp311-win32.whl", hash = "sha256:6c89876f41da747c8d3677a2b540fb32ef5715f97b66eeb0c6b66f5e3ef6f59d", size = 15094, upload-time = "2024-10-18T15:21:11.005Z" },
+    { url = "https://files.pythonhosted.org/packages/da/b8/3a3bd761922d416f3dc5d00bfbed11f66b1ab89a0c2b6e887240a30b0f6b/MarkupSafe-3.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:70a87b411535ccad5ef2f1df5136506a10775d267e197e4cf531ced10537bd6b", size = 15521, upload-time = "2024-10-18T15:21:12.911Z" },
+    { url = "https://files.pythonhosted.org/packages/22/09/d1f21434c97fc42f09d290cbb6350d44eb12f09cc62c9476effdb33a18aa/MarkupSafe-3.0.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:9778bd8ab0a994ebf6f84c2b949e65736d5575320a17ae8984a77fab08db94cf", size = 14274, upload-time = "2024-10-18T15:21:13.777Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/b0/18f76bba336fa5aecf79d45dcd6c806c280ec44538b3c13671d49099fdd0/MarkupSafe-3.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:846ade7b71e3536c4e56b386c2a47adf5741d2d8b94ec9dc3e92e5e1ee1e2225", size = 12348, upload-time = "2024-10-18T15:21:14.822Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/25/dd5c0f6ac1311e9b40f4af06c78efde0f3b5cbf02502f8ef9501294c425b/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1c99d261bd2d5f6b59325c92c73df481e05e57f19837bdca8413b9eac4bd8028", size = 24149, upload-time = "2024-10-18T15:21:15.642Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/f0/89e7aadfb3749d0f52234a0c8c7867877876e0a20b60e2188e9850794c17/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e17c96c14e19278594aa4841ec148115f9c7615a47382ecb6b82bd8fea3ab0c8", size = 23118, upload-time = "2024-10-18T15:21:17.133Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/da/f2eeb64c723f5e3777bc081da884b414671982008c47dcc1873d81f625b6/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:88416bd1e65dcea10bc7569faacb2c20ce071dd1f87539ca2ab364bf6231393c", size = 22993, upload-time = "2024-10-18T15:21:18.064Z" },
+    { url = "https://files.pythonhosted.org/packages/da/0e/1f32af846df486dce7c227fe0f2398dc7e2e51d4a370508281f3c1c5cddc/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2181e67807fc2fa785d0592dc2d6206c019b9502410671cc905d132a92866557", size = 24178, upload-time = "2024-10-18T15:21:18.859Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/f6/bb3ca0532de8086cbff5f06d137064c8410d10779c4c127e0e47d17c0b71/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:52305740fe773d09cffb16f8ed0427942901f00adedac82ec8b67752f58a1b22", size = 23319, upload-time = "2024-10-18T15:21:19.671Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/82/8be4c96ffee03c5b4a034e60a31294daf481e12c7c43ab8e34a1453ee48b/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ad10d3ded218f1039f11a75f8091880239651b52e9bb592ca27de44eed242a48", size = 23352, upload-time = "2024-10-18T15:21:20.971Z" },
+    { url = "https://files.pythonhosted.org/packages/51/ae/97827349d3fcffee7e184bdf7f41cd6b88d9919c80f0263ba7acd1bbcb18/MarkupSafe-3.0.2-cp312-cp312-win32.whl", hash = "sha256:0f4ca02bea9a23221c0182836703cbf8930c5e9454bacce27e767509fa286a30", size = 15097, upload-time = "2024-10-18T15:21:22.646Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/80/a61f99dc3a936413c3ee4e1eecac96c0da5ed07ad56fd975f1a9da5bc630/MarkupSafe-3.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:8e06879fc22a25ca47312fbe7c8264eb0b662f6db27cb2d3bbbc74b1df4b9b87", size = 15601, upload-time = "2024-10-18T15:21:23.499Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/ea/9b1530c3fdeeca613faeb0fb5cbcf2389d816072fab72a71b45749ef6062/MarkupSafe-3.0.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:eaa0a10b7f72326f1372a713e73c3f739b524b3af41feb43e4921cb529f5929a", size = 14344, upload-time = "2024-10-18T15:21:43.721Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/c2/fbdbfe48848e7112ab05e627e718e854d20192b674952d9042ebd8c9e5de/MarkupSafe-3.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:48032821bbdf20f5799ff537c7ac3d1fba0ba032cfc06194faffa8cda8b560ff", size = 12389, upload-time = "2024-10-18T15:21:44.666Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/25/7a7c6e4dbd4f867d95d94ca15449e91e52856f6ed1905d58ef1de5e211d0/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1a9d3f5f0901fdec14d8d2f66ef7d035f2157240a433441719ac9a3fba440b13", size = 21607, upload-time = "2024-10-18T15:21:45.452Z" },
+    { url = "https://files.pythonhosted.org/packages/53/8f/f339c98a178f3c1e545622206b40986a4c3307fe39f70ccd3d9df9a9e425/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88b49a3b9ff31e19998750c38e030fc7bb937398b1f78cfa599aaef92d693144", size = 20728, upload-time = "2024-10-18T15:21:46.295Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/03/8496a1a78308456dbd50b23a385c69b41f2e9661c67ea1329849a598a8f9/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:cfad01eed2c2e0c01fd0ecd2ef42c492f7f93902e39a42fc9ee1692961443a29", size = 20826, upload-time = "2024-10-18T15:21:47.134Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/cf/0a490a4bd363048c3022f2f475c8c05582179bb179defcee4766fb3dcc18/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1225beacc926f536dc82e45f8a4d68502949dc67eea90eab715dea3a21c1b5f0", size = 21843, upload-time = "2024-10-18T15:21:48.334Z" },
+    { url = "https://files.pythonhosted.org/packages/19/a3/34187a78613920dfd3cdf68ef6ce5e99c4f3417f035694074beb8848cd77/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:3169b1eefae027567d1ce6ee7cae382c57fe26e82775f460f0b2778beaad66c0", size = 21219, upload-time = "2024-10-18T15:21:49.587Z" },
+    { url = "https://files.pythonhosted.org/packages/17/d8/5811082f85bb88410ad7e452263af048d685669bbbfb7b595e8689152498/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:eb7972a85c54febfb25b5c4b4f3af4dcc731994c7da0d8a0b4a6eb0640e1d178", size = 20946, upload-time = "2024-10-18T15:21:50.441Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/31/bd635fb5989440d9365c5e3c47556cfea121c7803f5034ac843e8f37c2f2/MarkupSafe-3.0.2-cp39-cp39-win32.whl", hash = "sha256:8c4e8c3ce11e1f92f6536ff07154f9d49677ebaaafc32db9db4620bc11ed480f", size = 15063, upload-time = "2024-10-18T15:21:51.385Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/73/085399401383ce949f727afec55ec3abd76648d04b9f22e1c0e99cb4bec3/MarkupSafe-3.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:6e296a513ca3d94054c2c881cc913116e90fd030ad1c656b3869762b754f5f8a", size = 15506, upload-time = "2024-10-18T15:21:52.974Z" },
 ]
 
 [[package]]
 name = "mdurl"
 version = "0.1.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/d6/54/cfe61301667036ec958cb99bd3efefba235e65cdeb9c84d24a8293ba1d90/mdurl-0.1.2.tar.gz", hash = "sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba", size = 8729 }
+sdist = { url = "https://files.pythonhosted.org/packages/d6/54/cfe61301667036ec958cb99bd3efefba235e65cdeb9c84d24a8293ba1d90/mdurl-0.1.2.tar.gz", hash = "sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba", size = 8729, upload-time = "2022-08-14T12:40:10.846Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b3/38/89ba8ad64ae25be8de66a6d463314cf1eb366222074cfda9ee839c56a4b4/mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8", size = 9979, upload-time = "2022-08-14T12:40:09.779Z" },
+]
+
+[[package]]
+name = "mergedeep"
+version = "1.3.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/3a/41/580bb4006e3ed0361b8151a01d324fb03f420815446c7def45d02f74c270/mergedeep-1.3.4.tar.gz", hash = "sha256:0096d52e9dad9939c3d975a774666af186eda617e6ca84df4c94dec30004f2a8", size = 4661, upload-time = "2021-02-05T18:55:30.623Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b3/38/89ba8ad64ae25be8de66a6d463314cf1eb366222074cfda9ee839c56a4b4/mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8", size = 9979 },
+    { url = "https://files.pythonhosted.org/packages/2c/19/04f9b178c2d8a15b076c8b5140708fa6ffc5601fb6f1e975537072df5b2a/mergedeep-1.3.4-py3-none-any.whl", hash = "sha256:70775750742b25c0d8f36c55aed03d24c3384d17c951b3175d898bd778ef0307", size = 6354, upload-time = "2021-02-05T18:55:29.583Z" },
 ]
 
 [[package]]
-name = "mkl"
-version = "2021.4.0"
+name = "mkdocs"
+version = "1.6.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "intel-openmp" },
-    { name = "tbb" },
+    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "ghp-import" },
+    { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
+    { name = "jinja2" },
+    { name = "markdown" },
+    { name = "markupsafe" },
+    { name = "mergedeep" },
+    { name = "mkdocs-get-deps" },
+    { name = "packaging" },
+    { name = "pathspec" },
+    { name = "pyyaml" },
+    { name = "pyyaml-env-tag" },
+    { name = "watchdog" },
 ]
+sdist = { url = "https://files.pythonhosted.org/packages/bc/c6/bbd4f061bd16b378247f12953ffcb04786a618ce5e904b8c5a01a0309061/mkdocs-1.6.1.tar.gz", hash = "sha256:7b432f01d928c084353ab39c57282f29f92136665bdd6abf7c1ec8d822ef86f2", size = 3889159, upload-time = "2024-08-30T12:24:06.899Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ce/c6/892fe3bc91e811b78e4f85653864f2d92541d5e5c306b0cb3c2311e9ca64/mkl-2021.4.0-py2.py3-none-win32.whl", hash = "sha256:439c640b269a5668134e3dcbcea4350459c4a8bc46469669b2d67e07e3d330e8", size = 129048357 },
-    { url = "https://files.pythonhosted.org/packages/fe/1c/5f6dbf18e8b73e0a5472466f0ea8d48ce9efae39bd2ff38cebf8dce61259/mkl-2021.4.0-py2.py3-none-win_amd64.whl", hash = "sha256:ceef3cafce4c009dd25f65d7ad0d833a0fbadc3d8903991ec92351fe5de1e718", size = 228499609 },
+    { url = "https://files.pythonhosted.org/packages/22/5b/dbc6a8cddc9cfa9c4971d59fb12bb8d42e161b7e7f8cc89e49137c5b279c/mkdocs-1.6.1-py3-none-any.whl", hash = "sha256:db91759624d1647f3f34aa0c3f327dd2601beae39a366d6e064c03468d35c20e", size = 3864451, upload-time = "2024-08-30T12:24:05.054Z" },
 ]
 
 [[package]]
-name = "ml-dtypes"
-version = "0.4.1"
+name = "mkdocs-gen-files"
+version = "0.5.0"
 source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.11.*'",
-    "python_full_version == '3.10.*'",
-    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
-    "python_full_version < '3.9.17'",
+dependencies = [
+    { name = "mkdocs" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/48/85/2d634462fd59136197d3126ca431ffb666f412e3db38fd5ce3a60566303e/mkdocs_gen_files-0.5.0.tar.gz", hash = "sha256:4c7cf256b5d67062a788f6b1d035e157fc1a9498c2399be9af5257d4ff4d19bc", size = 7539, upload-time = "2023-04-27T19:48:04.894Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e7/0f/1e55b3fd490ad2cecb6e7b31892d27cb9fc4218ec1dab780440ba8579e74/mkdocs_gen_files-0.5.0-py3-none-any.whl", hash = "sha256:7ac060096f3f40bd19039e7277dd3050be9a453c8ac578645844d4d91d7978ea", size = 8380, upload-time = "2023-04-27T19:48:07.059Z" },
 ]
+
+[[package]]
+name = "mkdocs-get-deps"
+version = "0.2.0"
+source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "numpy", marker = "python_full_version != '3.12.*'" },
+    { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
+    { name = "mergedeep" },
+    { name = "platformdirs" },
+    { name = "pyyaml" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fd/15/76f86faa0902836cc133939732f7611ace68cf54148487a99c539c272dc8/ml_dtypes-0.4.1.tar.gz", hash = "sha256:fad5f2de464fd09127e49b7fd1252b9006fb43d2edc1ff112d390c324af5ca7a", size = 692594 }
+sdist = { url = "https://files.pythonhosted.org/packages/98/f5/ed29cd50067784976f25ed0ed6fcd3c2ce9eb90650aa3b2796ddf7b6870b/mkdocs_get_deps-0.2.0.tar.gz", hash = "sha256:162b3d129c7fad9b19abfdcb9c1458a651628e4b1dea628ac68790fb3061c60c", size = 10239, upload-time = "2023-11-20T17:51:09.981Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/56/9e/76b84f77c7afee3b116dc8407903a2d5004ba3059a8f3dcdcfa6ebf33fff/ml_dtypes-0.4.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:1fe8b5b5e70cd67211db94b05cfd58dace592f24489b038dc6f9fe347d2e07d5", size = 397975 },
-    { url = "https://files.pythonhosted.org/packages/03/7b/32650e1b2a2713a5923a0af2a8503d0d4a8fc99d1e1e0a1c40e996634460/ml_dtypes-0.4.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8c09a6d11d8475c2a9fd2bc0695628aec105f97cab3b3a3fb7c9660348ff7d24", size = 2182570 },
-    { url = "https://files.pythonhosted.org/packages/16/86/a9f7569e7e4f5395f927de38a13b92efa73f809285d04f2923b291783dd2/ml_dtypes-0.4.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9f5e8f75fa371020dd30f9196e7d73babae2abd51cf59bdd56cb4f8de7e13354", size = 2160365 },
-    { url = "https://files.pythonhosted.org/packages/04/1b/9a3afb437702503514f3934ec8d7904270edf013d28074f3e700e5dfbb0f/ml_dtypes-0.4.1-cp310-cp310-win_amd64.whl", hash = "sha256:15fdd922fea57e493844e5abb930b9c0bd0af217d9edd3724479fc3d7ce70e3f", size = 126633 },
-    { url = "https://files.pythonhosted.org/packages/d1/76/9835c8609c29f2214359e88f29255fc4aad4ea0f613fb48aa8815ceda1b6/ml_dtypes-0.4.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:2d55b588116a7085d6e074cf0cdb1d6fa3875c059dddc4d2c94a4cc81c23e975", size = 397973 },
-    { url = "https://files.pythonhosted.org/packages/7e/99/e68c56fac5de973007a10254b6e17a0362393724f40f66d5e4033f4962c2/ml_dtypes-0.4.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e138a9b7a48079c900ea969341a5754019a1ad17ae27ee330f7ebf43f23877f9", size = 2185134 },
-    { url = "https://files.pythonhosted.org/packages/28/bc/6a2344338ea7b61cd7b46fb24ec459360a5a0903b57c55b156c1e46c644a/ml_dtypes-0.4.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:74c6cfb5cf78535b103fde9ea3ded8e9f16f75bc07789054edc7776abfb3d752", size = 2163661 },
-    { url = "https://files.pythonhosted.org/packages/e8/d3/ddfd9878b223b3aa9a930c6100a99afca5cfab7ea703662e00323acb7568/ml_dtypes-0.4.1-cp311-cp311-win_amd64.whl", hash = "sha256:274cc7193dd73b35fb26bef6c5d40ae3eb258359ee71cd82f6e96a8c948bdaa6", size = 126727 },
-    { url = "https://files.pythonhosted.org/packages/ba/1a/99e924f12e4b62139fbac87419698c65f956d58de0dbfa7c028fa5b096aa/ml_dtypes-0.4.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:827d3ca2097085cf0355f8fdf092b888890bb1b1455f52801a2d7756f056f54b", size = 405077 },
-    { url = "https://files.pythonhosted.org/packages/8f/8c/7b610bd500617854c8cc6ed7c8cfb9d48d6a5c21a1437a36a4b9bc8a3598/ml_dtypes-0.4.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:772426b08a6172a891274d581ce58ea2789cc8abc1c002a27223f314aaf894e7", size = 2181554 },
-    { url = "https://files.pythonhosted.org/packages/c7/c6/f89620cecc0581dc1839e218c4315171312e46c62a62da6ace204bda91c0/ml_dtypes-0.4.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:126e7d679b8676d1a958f2651949fbfa182832c3cd08020d8facd94e4114f3e9", size = 2160488 },
-    { url = "https://files.pythonhosted.org/packages/ae/11/a742d3c31b2cc8557a48efdde53427fd5f9caa2fa3c9c27d826e78a66f51/ml_dtypes-0.4.1-cp312-cp312-win_amd64.whl", hash = "sha256:df0fb650d5c582a9e72bb5bd96cfebb2cdb889d89daff621c8fbc60295eba66c", size = 127462 },
-    { url = "https://files.pythonhosted.org/packages/8f/d7/6e1372052fe95c0cacfdb9718dba04726203885ffddb0cfddd8f8aa89a3b/ml_dtypes-0.4.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:e35e486e97aee577d0890bc3bd9e9f9eece50c08c163304008587ec8cfe7575b", size = 396578 },
-    { url = "https://files.pythonhosted.org/packages/1a/f6/ad0bd2735b9570ebf9c113f024b4f2b34f2331f16197c60babdc168b22d5/ml_dtypes-0.4.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:560be16dc1e3bdf7c087eb727e2cf9c0e6a3d87e9f415079d2491cc419b3ebf5", size = 2181057 },
-    { url = "https://files.pythonhosted.org/packages/3e/55/b9711de47135d4d8766ff7907fe54c8bffff545fd646817c352de37b0ad5/ml_dtypes-0.4.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ad0b757d445a20df39035c4cdeed457ec8b60d236020d2560dbc25887533cf50", size = 2156131 },
-    { url = "https://files.pythonhosted.org/packages/4b/f3/e5ff8dd27f66c8b80f97f0f89bb0b74e4a7005e5ff5f8f4237126c827911/ml_dtypes-0.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:ef0d7e3fece227b49b544fa69e50e607ac20948f0043e9f76b44f35f229ea450", size = 126727 },
+    { url = "https://files.pythonhosted.org/packages/9f/d4/029f984e8d3f3b6b726bd33cafc473b75e9e44c0f7e80a5b29abc466bdea/mkdocs_get_deps-0.2.0-py3-none-any.whl", hash = "sha256:2bf11d0b133e77a0dd036abeeb06dec8775e46efa526dc70667d8863eefc6134", size = 9521, upload-time = "2023-11-20T17:51:08.587Z" },
 ]
 
 [[package]]
 name = "ml-dtypes"
-version = "0.5.1"
+version = "0.5.3"
 source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "python_full_version == '3.12.*'",
-]
-dependencies = [
-    { name = "numpy", marker = "python_full_version == '3.12.*'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/32/49/6e67c334872d2c114df3020e579f3718c333198f8312290e09ec0216703a/ml_dtypes-0.5.1.tar.gz", hash = "sha256:ac5b58559bb84a95848ed6984eb8013249f90b6bab62aa5acbad876e256002c9", size = 698772 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/f4/88/11ebdbc75445eeb5b6869b708a0d787d1ed812ff86c2170bbfb95febdce1/ml_dtypes-0.5.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:bd73f51957949069573ff783563486339a9285d72e2f36c18e0c1aa9ca7eb190", size = 671450 },
-    { url = "https://files.pythonhosted.org/packages/a4/a4/9321cae435d6140f9b0e7af8334456a854b60e3a9c6101280a16e3594965/ml_dtypes-0.5.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:810512e2eccdfc3b41eefa3a27402371a3411453a1efc7e9c000318196140fed", size = 4621075 },
-    { url = "https://files.pythonhosted.org/packages/16/d8/4502e12c6a10d42e13a552e8d97f20198e3cf82a0d1411ad50be56a5077c/ml_dtypes-0.5.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:141b2ea2f20bb10802ddca55d91fe21231ef49715cfc971998e8f2a9838f3dbe", size = 4738414 },
-    { url = "https://files.pythonhosted.org/packages/6b/7e/bc54ae885e4d702e60a4bf50aa9066ff35e9c66b5213d11091f6bffb3036/ml_dtypes-0.5.1-cp310-cp310-win_amd64.whl", hash = "sha256:26ebcc69d7b779c8f129393e99732961b5cc33fcff84090451f448c89b0e01b4", size = 209718 },
-    { url = "https://files.pythonhosted.org/packages/c9/fd/691335926126bb9beeb030b61a28f462773dcf16b8e8a2253b599013a303/ml_dtypes-0.5.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:023ce2f502efd4d6c1e0472cc58ce3640d051d40e71e27386bed33901e201327", size = 671448 },
-    { url = "https://files.pythonhosted.org/packages/ff/a6/63832d91f2feb250d865d069ba1a5d0c686b1f308d1c74ce9764472c5e22/ml_dtypes-0.5.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7000b6e4d8ef07542c05044ec5d8bbae1df083b3f56822c3da63993a113e716f", size = 4625792 },
-    { url = "https://files.pythonhosted.org/packages/cc/2a/5421fd3dbe6eef9b844cc9d05f568b9fb568503a2e51cb1eb4443d9fc56b/ml_dtypes-0.5.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c09526488c3a9e8b7a23a388d4974b670a9a3dd40c5c8a61db5593ce9b725bab", size = 4743893 },
-    { url = "https://files.pythonhosted.org/packages/60/30/d3f0fc9499a22801219679a7f3f8d59f1429943c6261f445fb4bfce20718/ml_dtypes-0.5.1-cp311-cp311-win_amd64.whl", hash = "sha256:15ad0f3b0323ce96c24637a88a6f44f6713c64032f27277b069f285c3cf66478", size = 209712 },
-    { url = "https://files.pythonhosted.org/packages/47/56/1bb21218e1e692506c220ffabd456af9733fba7aa1b14f73899979f4cc20/ml_dtypes-0.5.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:6f462f5eca22fb66d7ff9c4744a3db4463af06c49816c4b6ac89b16bfcdc592e", size = 670372 },
-    { url = "https://files.pythonhosted.org/packages/20/95/d8bd96a3b60e00bf31bd78ca4bdd2d6bbaf5acb09b42844432d719d34061/ml_dtypes-0.5.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6f76232163b5b9c34291b54621ee60417601e2e4802a188a0ea7157cd9b323f4", size = 4635946 },
-    { url = "https://files.pythonhosted.org/packages/08/57/5d58fad4124192b1be42f68bd0c0ddaa26e44a730ff8c9337adade2f5632/ml_dtypes-0.5.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ad4953c5eb9c25a56d11a913c2011d7e580a435ef5145f804d98efa14477d390", size = 4694804 },
-    { url = "https://files.pythonhosted.org/packages/38/bc/c4260e4a6c6bf684d0313308de1c860467275221d5e7daf69b3fcddfdd0b/ml_dtypes-0.5.1-cp312-cp312-win_amd64.whl", hash = "sha256:9626d0bca1fb387d5791ca36bacbba298c5ef554747b7ebeafefb4564fc83566", size = 210853 },
-    { url = "https://files.pythonhosted.org/packages/0f/92/bb6a3d18e16fddd18ce6d5f480e1919b33338c70e18cba831c6ae59812ee/ml_dtypes-0.5.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:12651420130ee7cc13059fc56dac6ad300c3af3848b802d475148c9defd27c23", size = 667696 },
-    { url = "https://files.pythonhosted.org/packages/6d/29/cfc89d842767e9a51146043b0fa18332c2b38f8831447e6cb1160e3c6102/ml_dtypes-0.5.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c9945669d3dadf8acb40ec2e57d38c985d8c285ea73af57fc5b09872c516106d", size = 4638365 },
-    { url = "https://files.pythonhosted.org/packages/be/26/adc36e3ea09603d9f6d114894e1c1b7b8e8a9ef6d0b031cc270c6624a37c/ml_dtypes-0.5.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bf9975bda82a99dc935f2ae4c83846d86df8fd6ba179614acac8e686910851da", size = 4702722 },
-    { url = "https://files.pythonhosted.org/packages/da/8a/a2b9375c94077e5a488a624a195621407846f504068ce22ccf805c674156/ml_dtypes-0.5.1-cp313-cp313-win_amd64.whl", hash = "sha256:fd918d4e6a4e0c110e2e05be7a7814d10dc1b95872accbf6512b80a109b71ae1", size = 210850 },
-    { url = "https://files.pythonhosted.org/packages/52/38/703169100fdde27957f061d4d0ea3e00525775a09acaccf7e655d9609d55/ml_dtypes-0.5.1-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:05f23447a1c20ddf4dc7c2c661aa9ed93fcb2658f1017c204d1e758714dc28a8", size = 693043 },
-    { url = "https://files.pythonhosted.org/packages/28/ff/4e234c9c23e0d456f5da5a326c103bf890c746d93351524d987e41f438b3/ml_dtypes-0.5.1-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1b7fbe5571fdf28fd3aaab3ef4aafc847de9ebf263be959958c1ca58ec8eadf5", size = 4903946 },
-    { url = "https://files.pythonhosted.org/packages/b7/45/c1a1ccfdd02bc4173ca0f4a2d327683a27df85797b885eb1da1ca325b85c/ml_dtypes-0.5.1-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d13755f8e8445b3870114e5b6240facaa7cb0c3361e54beba3e07fa912a6e12b", size = 5052731 },
-    { url = "https://files.pythonhosted.org/packages/4c/17/b8f22639eb0cf52aee86c38c6e4cf1586e99445ef8e87bc1d30237bd5a1a/ml_dtypes-0.5.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:b8a9d46b4df5ae2135a8e8e72b465448ebbc1559997f4f9304a9ecc3413efb5b", size = 667609 },
-    { url = "https://files.pythonhosted.org/packages/95/87/4a1c91ea325ec4708293ae270b375fbdc53917031a64be9d5e3908ac0185/ml_dtypes-0.5.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:afb2009ac98da274e893e03162f6269398b2b00d947e7057ee2469a921d58135", size = 4615607 },
-    { url = "https://files.pythonhosted.org/packages/e7/6d/f53c438cbdf753164497958f4ff12a6844ae66a00edcabebcde1dc575ab7/ml_dtypes-0.5.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:aefedc579ece2f8fb38f876aa7698204ee4c372d0e54f1c1ffa8ca580b54cc60", size = 4729131 },
-    { url = "https://files.pythonhosted.org/packages/aa/10/03107a1ff3b03d6515d7365d9e356bc612b0750edbd5f4327a1db44404bc/ml_dtypes-0.5.1-cp39-cp39-win_amd64.whl", hash = "sha256:8f2c028954f16ede77902b223a8da2d9cbb3892375b85809a5c3cfb1587960c4", size = 209426 },
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/78/a7/aad060393123cfb383956dca68402aff3db1e1caffd5764887ed5153f41b/ml_dtypes-0.5.3.tar.gz", hash = "sha256:95ce33057ba4d05df50b1f3cfefab22e351868a843b3b15a46c65836283670c9", size = 692316, upload-time = "2025-07-29T18:39:19.454Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ac/bb/1f32124ab6d3a279ea39202fe098aea95b2d81ef0ce1d48612b6bf715e82/ml_dtypes-0.5.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:0a1d68a7cb53e3f640b2b6a34d12c0542da3dd935e560fdf463c0c77f339fc20", size = 667409, upload-time = "2025-07-29T18:38:17.321Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/ac/e002d12ae19136e25bb41c7d14d7e1a1b08f3c0e99a44455ff6339796507/ml_dtypes-0.5.3-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0cd5a6c711b5350f3cbc2ac28def81cd1c580075ccb7955e61e9d8f4bfd40d24", size = 4960702, upload-time = "2025-07-29T18:38:19.616Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/12/79e9954e6b3255a4b1becb191a922d6e2e94d03d16a06341ae9261963ae8/ml_dtypes-0.5.3-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:bdcf26c2dbc926b8a35ec8cbfad7eff1a8bd8239e12478caca83a1fc2c400dc2", size = 4933471, upload-time = "2025-07-29T18:38:21.809Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/aa/d1eff619e83cd1ddf6b561d8240063d978e5d887d1861ba09ef01778ec3a/ml_dtypes-0.5.3-cp310-cp310-win_amd64.whl", hash = "sha256:aecbd7c5272c82e54d5b99d8435fd10915d1bc704b7df15e4d9ca8dc3902be61", size = 206330, upload-time = "2025-07-29T18:38:23.663Z" },
+    { url = "https://files.pythonhosted.org/packages/af/f1/720cb1409b5d0c05cff9040c0e9fba73fa4c67897d33babf905d5d46a070/ml_dtypes-0.5.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:4a177b882667c69422402df6ed5c3428ce07ac2c1f844d8a1314944651439458", size = 667412, upload-time = "2025-07-29T18:38:25.275Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/d5/05861ede5d299f6599f86e6bc1291714e2116d96df003cfe23cc54bcc568/ml_dtypes-0.5.3-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9849ce7267444c0a717c80c6900997de4f36e2815ce34ac560a3edb2d9a64cd2", size = 4964606, upload-time = "2025-07-29T18:38:27.045Z" },
+    { url = "https://files.pythonhosted.org/packages/db/dc/72992b68de367741bfab8df3b3fe7c29f982b7279d341aa5bf3e7ef737ea/ml_dtypes-0.5.3-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c3f5ae0309d9f888fd825c2e9d0241102fadaca81d888f26f845bc8c13c1e4ee", size = 4938435, upload-time = "2025-07-29T18:38:29.193Z" },
+    { url = "https://files.pythonhosted.org/packages/81/1c/d27a930bca31fb07d975a2d7eaf3404f9388114463b9f15032813c98f893/ml_dtypes-0.5.3-cp311-cp311-win_amd64.whl", hash = "sha256:58e39349d820b5702bb6f94ea0cb2dc8ec62ee81c0267d9622067d8333596a46", size = 206334, upload-time = "2025-07-29T18:38:30.687Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/d8/6922499effa616012cb8dc445280f66d100a7ff39b35c864cfca019b3f89/ml_dtypes-0.5.3-cp311-cp311-win_arm64.whl", hash = "sha256:66c2756ae6cfd7f5224e355c893cfd617fa2f747b8bbd8996152cbdebad9a184", size = 157584, upload-time = "2025-07-29T18:38:32.187Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/eb/bc07c88a6ab002b4635e44585d80fa0b350603f11a2097c9d1bfacc03357/ml_dtypes-0.5.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:156418abeeda48ea4797db6776db3c5bdab9ac7be197c1233771e0880c304057", size = 663864, upload-time = "2025-07-29T18:38:33.777Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/89/11af9b0f21b99e6386b6581ab40fb38d03225f9de5f55cf52097047e2826/ml_dtypes-0.5.3-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1db60c154989af253f6c4a34e8a540c2c9dce4d770784d426945e09908fbb177", size = 4951313, upload-time = "2025-07-29T18:38:36.45Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/a9/b98b86426c24900b0c754aad006dce2863df7ce0bb2bcc2c02f9cc7e8489/ml_dtypes-0.5.3-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1b255acada256d1fa8c35ed07b5f6d18bc21d1556f842fbc2d5718aea2cd9e55", size = 4928805, upload-time = "2025-07-29T18:38:38.29Z" },
+    { url = "https://files.pythonhosted.org/packages/50/c1/85e6be4fc09c6175f36fb05a45917837f30af9a5146a5151cb3a3f0f9e09/ml_dtypes-0.5.3-cp312-cp312-win_amd64.whl", hash = "sha256:da65e5fd3eea434ccb8984c3624bc234ddcc0d9f4c81864af611aaebcc08a50e", size = 208182, upload-time = "2025-07-29T18:38:39.72Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/17/cf5326d6867be057f232d0610de1458f70a8ce7b6290e4b4a277ea62b4cd/ml_dtypes-0.5.3-cp312-cp312-win_arm64.whl", hash = "sha256:8bb9cd1ce63096567f5f42851f5843b5a0ea11511e50039a7649619abfb4ba6d", size = 161560, upload-time = "2025-07-29T18:38:41.072Z" },
+    { url = "https://files.pythonhosted.org/packages/19/2d/c61af51173083bbf2a3b0f1a1a01d50ef1830436880027433d1b75271083/ml_dtypes-0.5.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:5ee72568d46b9533ad54f78b1e1f3067c0534c5065120ea8ecc6f210d22748b3", size = 663552, upload-time = "2025-07-29T18:39:13.102Z" },
+    { url = "https://files.pythonhosted.org/packages/61/0e/a628f2aefd719745e8a13492375a55cedea77c0cfc917b1ce11bde435c68/ml_dtypes-0.5.3-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:01de48de4537dc3c46e684b969a40ec36594e7eeb7c69e9a093e7239f030a28a", size = 4952704, upload-time = "2025-07-29T18:39:14.829Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/2e/5ba92f1f99d1f5f62bffec614a5b8161e55c3961257c902fa26dbe909baa/ml_dtypes-0.5.3-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8b1a6e231b0770f2894910f1dce6d2f31d65884dbf7668f9b08d73623cdca909", size = 4923538, upload-time = "2025-07-29T18:39:16.581Z" },
+    { url = "https://files.pythonhosted.org/packages/70/3b/f801c69027866ea6e387224551185fedef62ad8e2e71181ec0d9dda905f7/ml_dtypes-0.5.3-cp39-cp39-win_amd64.whl", hash = "sha256:a4f39b9bf6555fab9bfb536cf5fdd1c1c727e8d22312078702e9ff005354b37f", size = 206567, upload-time = "2025-07-29T18:39:18.047Z" },
 ]
 
 [[package]]
 name = "mlx"
-version = "0.22.0"
+version = "0.28.0"
 source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "mlx-metal", marker = "sys_platform == 'darwin'" },
+]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/41/3b/5160be3949dbed395c73646948f45d792a3de0dd6d8de213f26f2ea8d428/mlx-0.22.0-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:8082581e05b09d576cdf241184992fe0f96dcae1a857de38a562b2439d4e3333", size = 27110425 },
-    { url = "https://files.pythonhosted.org/packages/ec/e1/5d52b4144a946f2d66261df576b1e5f34845d781a2eba06d28ea0e554c51/mlx-0.22.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:b45716a948df0727e4e3525ca22e3f590323842020b7505e74491120a3a06d47", size = 26853548 },
-    { url = "https://files.pythonhosted.org/packages/b4/e7/2b4b551cff66370e0ce5d37dfb107122688d2b981eced38438e595ef98da/mlx-0.22.0-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:564ba1ca7711fc7ca5a1a17f43fc1859aef164229ee830050c66ef7453a9109d", size = 27110977 },
-    { url = "https://files.pythonhosted.org/packages/69/33/16e3a1451f6b9d9d4252d48453470d907c20f13a39915fd3e18d7a049570/mlx-0.22.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:03326dde450b9a8d75933514d3d4dd86163d2296bf70ff5d37befed3d6ba4144", size = 26854060 },
-    { url = "https://files.pythonhosted.org/packages/bc/33/dfab1ea36fc9b11e310922dceaeee210fc589efd277c409b0dc51bd38b79/mlx-0.22.0-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:56dec470a1a75010d1888a2e8f9cf24780980ff163eb6b03e9871bf440642a5b", size = 27107090 },
-    { url = "https://files.pythonhosted.org/packages/32/2b/756b04394e41dacd04ccb6fde0fede837590ba752229d2c27e6c442929ce/mlx-0.22.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:1dface82280483f14f2557da2afe8c965f7b09caa4bdae962cfa78897d0bed73", size = 26850177 },
-    { url = "https://files.pythonhosted.org/packages/62/2b/427896261bc8d940eff561e6199d1aee9dbdc7caa117486654a44d7d793c/mlx-0.22.0-cp313-cp313-macosx_13_0_arm64.whl", hash = "sha256:50d0d76826cfe939025791ce2c014e743ec7aff7aa67194ffaef40c40e574ef4", size = 27106970 },
-    { url = "https://files.pythonhosted.org/packages/6e/5a/314451ba1600a57619643bfc61c65e8ce34fed0ad011488ad3bc0fe3e739/mlx-0.22.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:bff0785c1be9ebc323d733c72f8a0683e757210221a9b8d53db3ac40ac7b75cf", size = 26850056 },
-    { url = "https://files.pythonhosted.org/packages/57/51/19deaecfe6b34c6ca30b4e966c8224145664747330143463f76b4c218a55/mlx-0.22.0-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:121d23087ea706b216c9f43f684c8068a6803d4aec1ddce1e3f6a66e99e797fa", size = 27110566 },
-    { url = "https://files.pythonhosted.org/packages/e4/88/56d767e752fea62e44704c9960fdf45a164b35866e6bc94f4a245122d8cc/mlx-0.22.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:d287de53d69af9129d8b59e13071c9c0711d729ddfc749f3187b0c894bf85d6b", size = 26853719 },
+    { url = "https://files.pythonhosted.org/packages/1e/ac/3d78ad9a4625b2684ad4356958d5353c33e06eeb54cf6c625bdb167044b9/mlx-0.28.0-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:5167b4c8d4aae9361afb5d770c75b653d66aa0ebbb8b0099ddd6da5d55d8e6b6", size = 565787, upload-time = "2025-08-07T07:50:42.798Z" },
+    { url = "https://files.pythonhosted.org/packages/61/16/fd564926a3e2b96ee339933a46d3e47db9177c038a028916bb43cc0fefba/mlx-0.28.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:168e945bd0e6de19fd0d73ecd0123ba99108342347be2c5aacd387cf1c889310", size = 540079, upload-time = "2025-08-07T07:49:07.185Z" },
+    { url = "https://files.pythonhosted.org/packages/15/f6/9cfb7e021992f7c8ece0bff6684b585acae0513a11c70fb14b0271bec724/mlx-0.28.0-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:74e7069c12b3f6aef0f75dc8ec5c76248c4776b86b6ee64a0f28d699b2412a1a", size = 540078, upload-time = "2025-08-07T07:50:47.728Z" },
+    { url = "https://files.pythonhosted.org/packages/42/dd/3b7adf81affa0b3336a8b28088bb69d7f2459ba9f66d1db864a6d204f3e6/mlx-0.28.0-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:e1e4734ac951ddc64bdc635417ea5ab98feeac2a2791e88759a20e7bf107e30e", size = 636545, upload-time = "2025-08-07T07:52:30.117Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/e1/c4a654b58af8af5ed7605a47a513de21716210adafac956bfff84b9e4df1/mlx-0.28.0-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:f969f50da4e709f21240765347cf28b4fcb721d21eed5801d895c694eb315e9f", size = 565785, upload-time = "2025-08-07T07:50:30.893Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/6f/82c0a988deb13a303cc8c8b6c2c34c9ab21c634a4b02b405a3a6731ffe98/mlx-0.28.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:a19046afe989565ce383d82d9b87bb539f355be75765d00b51ff8a4c7c038104", size = 540283, upload-time = "2025-08-07T07:48:57.715Z" },
+    { url = "https://files.pythonhosted.org/packages/22/c6/bbc28fbd10ddad7727d239a29bfc0a0c9f9258ae6762cec8ee50b6d7e939/mlx-0.28.0-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:be3cbc4bf21debed9b6f7a3755b41b007333d289636a3fd4ea9040b0eb20e22c", size = 540282, upload-time = "2025-08-07T07:50:48.883Z" },
+    { url = "https://files.pythonhosted.org/packages/86/24/58b6e434dac2bc384d24a7e44bda86c8f0e1308421e137e4b0583ec40f28/mlx-0.28.0-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:bd37c1ee9fbf91879e02bb68c874061badc893509567c3392ee7be91414a6dba", size = 636553, upload-time = "2025-08-07T07:52:29.75Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/c9/d12ed6a8393450e28eb1f552b50200f83f138b1268b5f4e8074a76d745a2/mlx-0.28.0-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:97866d5e454e8f2d7bc42aadcbfd7565d40f4755564785e4fb964812fbad604b", size = 564160, upload-time = "2025-08-07T07:50:34.652Z" },
+    { url = "https://files.pythonhosted.org/packages/71/4f/3951766a5edb75c0d2d860381f592d271b4c3b7241e730e78dd63926f5b4/mlx-0.28.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:5204ebf399439e5da374295f6c1b6961355824604eed7026c18edfe4c83e9243", size = 540098, upload-time = "2025-08-07T07:50:52.67Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/52/cb8eb03544eace055a500bd4a3b776a3ce48198d7b7b398e21a5a3256e89/mlx-0.28.0-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:34776bd3fe97bca7c6c76d77f6104e0d6b05b3626bb3cf9ed48d3a9bbd46c180", size = 540100, upload-time = "2025-08-07T07:50:49.095Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/23/cb2703724f011d9aabc5a5a8ddf20481fdbdf251227468de89885b71f0ed/mlx-0.28.0-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:ead9a6c13b704239b5ca50d68f5effc505c8f15a6017f35d8b4d0e25832e29dd", size = 632943, upload-time = "2025-08-07T07:52:35.695Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/e5/f1edc60f2176f8600e4a040c9c92ba564030298199830ae0982c95c1fb8b/mlx-0.28.0-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:9a954e5f8f0c3c92d76e679162c8eb8e4618f8d3662cbd41e9b5cbdc087a298b", size = 566000, upload-time = "2025-08-07T07:53:04.304Z" },
+    { url = "https://files.pythonhosted.org/packages/56/8d/35dee5df087f6e313754c9ef15eadf48de21bff6030139b1af662f5ae763/mlx-0.28.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:56eaa39961c3ef23d4915466380d80407221134d12c0eef8d2391e55bcfe8802", size = 540561, upload-time = "2025-08-07T07:52:58.193Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/dd/54d8e9e82656537938fc8bc903352d731f5f1d025e8601354e0aed1ba692/mlx-0.28.0-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:3bc6dbe0d6b42c5bfcd783632b2217483fccb9994f7ce3b864d2b9ad2dfea7aa", size = 540561, upload-time = "2025-08-07T07:53:02.354Z" },
+    { url = "https://files.pythonhosted.org/packages/77/b2/254057fb1dce8070e789f14a724a514679646b9045b18f6c83435bc33f61/mlx-0.28.0-cp39-cp39-manylinux_2_35_x86_64.whl", hash = "sha256:f7903324346ac69c60f7e316624779c2ddf1ad41dd0c9670f978126b88ac5c6a", size = 636768, upload-time = "2025-08-07T07:55:33.109Z" },
 ]
 
 [[package]]
 name = "mlx-lm"
-version = "0.21.1"
+version = "0.26.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jinja2" },
     { name = "mlx" },
-    { name = "numpy" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "protobuf" },
     { name = "pyyaml" },
-    { name = "transformers", extra = ["sentencepiece"] },
+    { name = "transformers" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/06/3a/125adb6e677795129ee932d189bac031253ebfcebaa1df559bdd4f0671b6/mlx_lm-0.21.1.tar.gz", hash = "sha256:701d4d1de92785cc24e651a28d46894df567187711cc994c88cb7247f3d1cde0", size = 106147 }
+sdist = { url = "https://files.pythonhosted.org/packages/88/20/f3af9d99a5ad6ac42419a3d381290a28bf6d9899ed517a7ccc9fea08546e/mlx_lm-0.26.4.tar.gz", hash = "sha256:1bf21ede1d2d7b660ae312868790df9d73a8553dc50655cf7ae867a36ebcc08c", size = 176384, upload-time = "2025-08-25T15:57:41.723Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/9c/8f/41cd35d3ef6117a6831a02b27633b7a7dd8798c6d1a435a864a9db6ecb84/mlx_lm-0.21.1-py3-none-any.whl", hash = "sha256:fcc4a37c08cd3d6855f1fc5197cf4c8ebf790e1c778bce19f5c2c742096708c9", size = 140948 },
+    { url = "https://files.pythonhosted.org/packages/de/6a/4d20d1b20cd690a3eeaf609c7cb9058f2d52c6d1081394f0d91bd12d08f7/mlx_lm-0.26.4-py3-none-any.whl", hash = "sha256:79bf3afb399ae3bb6073bf0fa6c04f33d70c831ccc6bbbc206c10567d4eef162", size = 242038, upload-time = "2025-08-25T15:57:40.181Z" },
+]
+
+[[package]]
+name = "mlx-metal"
+version = "0.28.0"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f1/71/879284c71dfb12ded986a6532a4ab7df5c2794385ccf2766c1b40aee74cb/mlx_metal-0.28.0-py3-none-macosx_13_0_arm64.whl", hash = "sha256:ce08d40f1fad4f0b3bc87bfff5d603c7fe7dd141c082ba9ce9328b41e8f8d46b", size = 33840007, upload-time = "2025-08-07T07:53:07.437Z" },
+    { url = "https://files.pythonhosted.org/packages/06/90/44a261ccb9f6052c93c9da4faa4fc6d4f914938c51ecbbb68c546ab521b9/mlx_metal-0.28.0-py3-none-macosx_14_0_arm64.whl", hash = "sha256:424142ab843e2ac0b14edb58cf88d96723823c565291f46ddeeaa072abcc991e", size = 33196759, upload-time = "2025-08-07T07:52:59.436Z" },
+    { url = "https://files.pythonhosted.org/packages/72/59/8e4dee2893a56fc68a27eec7ec7ed9559c7ea01099313a9b8196373bf3cf/mlx_metal-0.28.0-py3-none-macosx_15_0_arm64.whl", hash = "sha256:214ece3781d44f57eb9686561594b28915ec5568df4a5a73da59c66880b204ed", size = 33167706, upload-time = "2025-08-07T07:53:03.852Z" },
 ]
 
 [[package]]
 name = "mpmath"
 version = "1.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f", size = 508106 }
+sdist = { url = "https://files.pythonhosted.org/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f", size = 508106, upload-time = "2023-03-07T16:47:11.061Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", size = 536198 },
+    { url = "https://files.pythonhosted.org/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c", size = 536198, upload-time = "2023-03-07T16:47:09.197Z" },
 ]
 
 [[package]]
 name = "msgpack"
-version = "1.1.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/cb/d0/7555686ae7ff5731205df1012ede15dd9d927f6227ea151e901c7406af4f/msgpack-1.1.0.tar.gz", hash = "sha256:dd432ccc2c72b914e4cb77afce64aab761c1137cc698be3984eee260bcb2896e", size = 167260 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/4b/f9/a892a6038c861fa849b11a2bb0502c07bc698ab6ea53359e5771397d883b/msgpack-1.1.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7ad442d527a7e358a469faf43fda45aaf4ac3249c8310a82f0ccff9164e5dccd", size = 150428 },
-    { url = "https://files.pythonhosted.org/packages/df/7a/d174cc6a3b6bb85556e6a046d3193294a92f9a8e583cdbd46dc8a1d7e7f4/msgpack-1.1.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:74bed8f63f8f14d75eec75cf3d04ad581da6b914001b474a5d3cd3372c8cc27d", size = 84131 },
-    { url = "https://files.pythonhosted.org/packages/08/52/bf4fbf72f897a23a56b822997a72c16de07d8d56d7bf273242f884055682/msgpack-1.1.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:914571a2a5b4e7606997e169f64ce53a8b1e06f2cf2c3a7273aa106236d43dd5", size = 81215 },
-    { url = "https://files.pythonhosted.org/packages/02/95/dc0044b439b518236aaf012da4677c1b8183ce388411ad1b1e63c32d8979/msgpack-1.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c921af52214dcbb75e6bdf6a661b23c3e6417f00c603dd2070bccb5c3ef499f5", size = 371229 },
-    { url = "https://files.pythonhosted.org/packages/ff/75/09081792db60470bef19d9c2be89f024d366b1e1973c197bb59e6aabc647/msgpack-1.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d8ce0b22b890be5d252de90d0e0d119f363012027cf256185fc3d474c44b1b9e", size = 378034 },
-    { url = "https://files.pythonhosted.org/packages/32/d3/c152e0c55fead87dd948d4b29879b0f14feeeec92ef1fd2ec21b107c3f49/msgpack-1.1.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:73322a6cc57fcee3c0c57c4463d828e9428275fb85a27aa2aa1a92fdc42afd7b", size = 363070 },
-    { url = "https://files.pythonhosted.org/packages/d9/2c/82e73506dd55f9e43ac8aa007c9dd088c6f0de2aa19e8f7330e6a65879fc/msgpack-1.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:e1f3c3d21f7cf67bcf2da8e494d30a75e4cf60041d98b3f79875afb5b96f3a3f", size = 359863 },
-    { url = "https://files.pythonhosted.org/packages/cb/a0/3d093b248837094220e1edc9ec4337de3443b1cfeeb6e0896af8ccc4cc7a/msgpack-1.1.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:64fc9068d701233effd61b19efb1485587560b66fe57b3e50d29c5d78e7fef68", size = 368166 },
-    { url = "https://files.pythonhosted.org/packages/e4/13/7646f14f06838b406cf5a6ddbb7e8dc78b4996d891ab3b93c33d1ccc8678/msgpack-1.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:42f754515e0f683f9c79210a5d1cad631ec3d06cea5172214d2176a42e67e19b", size = 370105 },
-    { url = "https://files.pythonhosted.org/packages/67/fa/dbbd2443e4578e165192dabbc6a22c0812cda2649261b1264ff515f19f15/msgpack-1.1.0-cp310-cp310-win32.whl", hash = "sha256:3df7e6b05571b3814361e8464f9304c42d2196808e0119f55d0d3e62cd5ea044", size = 68513 },
-    { url = "https://files.pythonhosted.org/packages/24/ce/c2c8fbf0ded750cb63cbcbb61bc1f2dfd69e16dca30a8af8ba80ec182dcd/msgpack-1.1.0-cp310-cp310-win_amd64.whl", hash = "sha256:685ec345eefc757a7c8af44a3032734a739f8c45d1b0ac45efc5d8977aa4720f", size = 74687 },
-    { url = "https://files.pythonhosted.org/packages/b7/5e/a4c7154ba65d93be91f2f1e55f90e76c5f91ccadc7efc4341e6f04c8647f/msgpack-1.1.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:3d364a55082fb2a7416f6c63ae383fbd903adb5a6cf78c5b96cc6316dc1cedc7", size = 150803 },
-    { url = "https://files.pythonhosted.org/packages/60/c2/687684164698f1d51c41778c838d854965dd284a4b9d3a44beba9265c931/msgpack-1.1.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:79ec007767b9b56860e0372085f8504db5d06bd6a327a335449508bbee9648fa", size = 84343 },
-    { url = "https://files.pythonhosted.org/packages/42/ae/d3adea9bb4a1342763556078b5765e666f8fdf242e00f3f6657380920972/msgpack-1.1.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6ad622bf7756d5a497d5b6836e7fc3752e2dd6f4c648e24b1803f6048596f701", size = 81408 },
-    { url = "https://files.pythonhosted.org/packages/dc/17/6313325a6ff40ce9c3207293aee3ba50104aed6c2c1559d20d09e5c1ff54/msgpack-1.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e59bca908d9ca0de3dc8684f21ebf9a690fe47b6be93236eb40b99af28b6ea6", size = 396096 },
-    { url = "https://files.pythonhosted.org/packages/a8/a1/ad7b84b91ab5a324e707f4c9761633e357820b011a01e34ce658c1dda7cc/msgpack-1.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5e1da8f11a3dd397f0a32c76165cf0c4eb95b31013a94f6ecc0b280c05c91b59", size = 403671 },
-    { url = "https://files.pythonhosted.org/packages/bb/0b/fd5b7c0b308bbf1831df0ca04ec76fe2f5bf6319833646b0a4bd5e9dc76d/msgpack-1.1.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:452aff037287acb1d70a804ffd022b21fa2bb7c46bee884dbc864cc9024128a0", size = 387414 },
-    { url = "https://files.pythonhosted.org/packages/f0/03/ff8233b7c6e9929a1f5da3c7860eccd847e2523ca2de0d8ef4878d354cfa/msgpack-1.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8da4bf6d54ceed70e8861f833f83ce0814a2b72102e890cbdfe4b34764cdd66e", size = 383759 },
-    { url = "https://files.pythonhosted.org/packages/1f/1b/eb82e1fed5a16dddd9bc75f0854b6e2fe86c0259c4353666d7fab37d39f4/msgpack-1.1.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:41c991beebf175faf352fb940bf2af9ad1fb77fd25f38d9142053914947cdbf6", size = 394405 },
-    { url = "https://files.pythonhosted.org/packages/90/2e/962c6004e373d54ecf33d695fb1402f99b51832631e37c49273cc564ffc5/msgpack-1.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a52a1f3a5af7ba1c9ace055b659189f6c669cf3657095b50f9602af3a3ba0fe5", size = 396041 },
-    { url = "https://files.pythonhosted.org/packages/f8/20/6e03342f629474414860c48aeffcc2f7f50ddaf351d95f20c3f1c67399a8/msgpack-1.1.0-cp311-cp311-win32.whl", hash = "sha256:58638690ebd0a06427c5fe1a227bb6b8b9fdc2bd07701bec13c2335c82131a88", size = 68538 },
-    { url = "https://files.pythonhosted.org/packages/aa/c4/5a582fc9a87991a3e6f6800e9bb2f3c82972912235eb9539954f3e9997c7/msgpack-1.1.0-cp311-cp311-win_amd64.whl", hash = "sha256:fd2906780f25c8ed5d7b323379f6138524ba793428db5d0e9d226d3fa6aa1788", size = 74871 },
-    { url = "https://files.pythonhosted.org/packages/e1/d6/716b7ca1dbde63290d2973d22bbef1b5032ca634c3ff4384a958ec3f093a/msgpack-1.1.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:d46cf9e3705ea9485687aa4001a76e44748b609d260af21c4ceea7f2212a501d", size = 152421 },
-    { url = "https://files.pythonhosted.org/packages/70/da/5312b067f6773429cec2f8f08b021c06af416bba340c912c2ec778539ed6/msgpack-1.1.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:5dbad74103df937e1325cc4bfeaf57713be0b4f15e1c2da43ccdd836393e2ea2", size = 85277 },
-    { url = "https://files.pythonhosted.org/packages/28/51/da7f3ae4462e8bb98af0d5bdf2707f1b8c65a0d4f496e46b6afb06cbc286/msgpack-1.1.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:58dfc47f8b102da61e8949708b3eafc3504509a5728f8b4ddef84bd9e16ad420", size = 82222 },
-    { url = "https://files.pythonhosted.org/packages/33/af/dc95c4b2a49cff17ce47611ca9ba218198806cad7796c0b01d1e332c86bb/msgpack-1.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4676e5be1b472909b2ee6356ff425ebedf5142427842aa06b4dfd5117d1ca8a2", size = 392971 },
-    { url = "https://files.pythonhosted.org/packages/f1/54/65af8de681fa8255402c80eda2a501ba467921d5a7a028c9c22a2c2eedb5/msgpack-1.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:17fb65dd0bec285907f68b15734a993ad3fc94332b5bb21b0435846228de1f39", size = 401403 },
-    { url = "https://files.pythonhosted.org/packages/97/8c/e333690777bd33919ab7024269dc3c41c76ef5137b211d776fbb404bfead/msgpack-1.1.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a51abd48c6d8ac89e0cfd4fe177c61481aca2d5e7ba42044fd218cfd8ea9899f", size = 385356 },
-    { url = "https://files.pythonhosted.org/packages/57/52/406795ba478dc1c890559dd4e89280fa86506608a28ccf3a72fbf45df9f5/msgpack-1.1.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2137773500afa5494a61b1208619e3871f75f27b03bcfca7b3a7023284140247", size = 383028 },
-    { url = "https://files.pythonhosted.org/packages/e7/69/053b6549bf90a3acadcd8232eae03e2fefc87f066a5b9fbb37e2e608859f/msgpack-1.1.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:398b713459fea610861c8a7b62a6fec1882759f308ae0795b5413ff6a160cf3c", size = 391100 },
-    { url = "https://files.pythonhosted.org/packages/23/f0/d4101d4da054f04274995ddc4086c2715d9b93111eb9ed49686c0f7ccc8a/msgpack-1.1.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:06f5fd2f6bb2a7914922d935d3b8bb4a7fff3a9a91cfce6d06c13bc42bec975b", size = 394254 },
-    { url = "https://files.pythonhosted.org/packages/1c/12/cf07458f35d0d775ff3a2dc5559fa2e1fcd06c46f1ef510e594ebefdca01/msgpack-1.1.0-cp312-cp312-win32.whl", hash = "sha256:ad33e8400e4ec17ba782f7b9cf868977d867ed784a1f5f2ab46e7ba53b6e1e1b", size = 69085 },
-    { url = "https://files.pythonhosted.org/packages/73/80/2708a4641f7d553a63bc934a3eb7214806b5b39d200133ca7f7afb0a53e8/msgpack-1.1.0-cp312-cp312-win_amd64.whl", hash = "sha256:115a7af8ee9e8cddc10f87636767857e7e3717b7a2e97379dc2054712693e90f", size = 75347 },
-    { url = "https://files.pythonhosted.org/packages/c8/b0/380f5f639543a4ac413e969109978feb1f3c66e931068f91ab6ab0f8be00/msgpack-1.1.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:071603e2f0771c45ad9bc65719291c568d4edf120b44eb36324dcb02a13bfddf", size = 151142 },
-    { url = "https://files.pythonhosted.org/packages/c8/ee/be57e9702400a6cb2606883d55b05784fada898dfc7fd12608ab1fdb054e/msgpack-1.1.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:0f92a83b84e7c0749e3f12821949d79485971f087604178026085f60ce109330", size = 84523 },
-    { url = "https://files.pythonhosted.org/packages/7e/3a/2919f63acca3c119565449681ad08a2f84b2171ddfcff1dba6959db2cceb/msgpack-1.1.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:4a1964df7b81285d00a84da4e70cb1383f2e665e0f1f2a7027e683956d04b734", size = 81556 },
-    { url = "https://files.pythonhosted.org/packages/7c/43/a11113d9e5c1498c145a8925768ea2d5fce7cbab15c99cda655aa09947ed/msgpack-1.1.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:59caf6a4ed0d164055ccff8fe31eddc0ebc07cf7326a2aaa0dbf7a4001cd823e", size = 392105 },
-    { url = "https://files.pythonhosted.org/packages/2d/7b/2c1d74ca6c94f70a1add74a8393a0138172207dc5de6fc6269483519d048/msgpack-1.1.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0907e1a7119b337971a689153665764adc34e89175f9a34793307d9def08e6ca", size = 399979 },
-    { url = "https://files.pythonhosted.org/packages/82/8c/cf64ae518c7b8efc763ca1f1348a96f0e37150061e777a8ea5430b413a74/msgpack-1.1.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:65553c9b6da8166e819a6aa90ad15288599b340f91d18f60b2061f402b9a4915", size = 383816 },
-    { url = "https://files.pythonhosted.org/packages/69/86/a847ef7a0f5ef3fa94ae20f52a4cacf596a4e4a010197fbcc27744eb9a83/msgpack-1.1.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:7a946a8992941fea80ed4beae6bff74ffd7ee129a90b4dd5cf9c476a30e9708d", size = 380973 },
-    { url = "https://files.pythonhosted.org/packages/aa/90/c74cf6e1126faa93185d3b830ee97246ecc4fe12cf9d2d31318ee4246994/msgpack-1.1.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:4b51405e36e075193bc051315dbf29168d6141ae2500ba8cd80a522964e31434", size = 387435 },
-    { url = "https://files.pythonhosted.org/packages/7a/40/631c238f1f338eb09f4acb0f34ab5862c4e9d7eda11c1b685471a4c5ea37/msgpack-1.1.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:b4c01941fd2ff87c2a934ee6055bda4ed353a7846b8d4f341c428109e9fcde8c", size = 399082 },
-    { url = "https://files.pythonhosted.org/packages/e9/1b/fa8a952be252a1555ed39f97c06778e3aeb9123aa4cccc0fd2acd0b4e315/msgpack-1.1.0-cp313-cp313-win32.whl", hash = "sha256:7c9a35ce2c2573bada929e0b7b3576de647b0defbd25f5139dcdaba0ae35a4cc", size = 69037 },
-    { url = "https://files.pythonhosted.org/packages/b6/bc/8bd826dd03e022153bfa1766dcdec4976d6c818865ed54223d71f07862b3/msgpack-1.1.0-cp313-cp313-win_amd64.whl", hash = "sha256:bce7d9e614a04d0883af0b3d4d501171fbfca038f12c77fa838d9f198147a23f", size = 75140 },
-    { url = "https://files.pythonhosted.org/packages/f7/3b/544a5c5886042b80e1f4847a4757af3430f60d106d8d43bb7be72c9e9650/msgpack-1.1.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:53258eeb7a80fc46f62fd59c876957a2d0e15e6449a9e71842b6d24419d88ca1", size = 150713 },
-    { url = "https://files.pythonhosted.org/packages/93/af/d63f25bcccd3d6f06fd518ba4a321f34a4370c67b579ca5c70b4a37721b4/msgpack-1.1.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7e7b853bbc44fb03fbdba34feb4bd414322180135e2cb5164f20ce1c9795ee48", size = 84277 },
-    { url = "https://files.pythonhosted.org/packages/92/9b/5c0dfb0009b9f96328664fecb9f8e4e9c8a1ae919e6d53986c1b813cb493/msgpack-1.1.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:f3e9b4936df53b970513eac1758f3882c88658a220b58dcc1e39606dccaaf01c", size = 81357 },
-    { url = "https://files.pythonhosted.org/packages/d1/7c/3a9ee6ec9fc3e47681ad39b4d344ee04ff20a776b594fba92d88d8b68356/msgpack-1.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46c34e99110762a76e3911fc923222472c9d681f1094096ac4102c18319e6468", size = 371256 },
-    { url = "https://files.pythonhosted.org/packages/f7/0a/8a213cecea7b731c540f25212ba5f9a818f358237ac51a44d448bd753690/msgpack-1.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a706d1e74dd3dea05cb54580d9bd8b2880e9264856ce5068027eed09680aa74", size = 377868 },
-    { url = "https://files.pythonhosted.org/packages/1b/94/a82b0db0981e9586ed5af77d6cfb343da05d7437dceaae3b35d346498110/msgpack-1.1.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:534480ee5690ab3cbed89d4c8971a5c631b69a8c0883ecfea96c19118510c846", size = 363370 },
-    { url = "https://files.pythonhosted.org/packages/93/fc/6c7f0dcc1c913e14861e16eaf494c07fc1dde454ec726ff8cebcf348ae53/msgpack-1.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:8cf9e8c3a2153934a23ac160cc4cba0ec035f6867c8013cc6077a79823370346", size = 358970 },
-    { url = "https://files.pythonhosted.org/packages/1f/c6/e4a04c0089deace870dabcdef5c9f12798f958e2e81d5012501edaff342f/msgpack-1.1.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:3180065ec2abbe13a4ad37688b61b99d7f9e012a535b930e0e683ad6bc30155b", size = 366358 },
-    { url = "https://files.pythonhosted.org/packages/b6/54/7d8317dac590cf16b3e08e3fb74d2081e5af44eb396f0effa13f17777f30/msgpack-1.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c5a91481a3cc573ac8c0d9aace09345d989dc4a0202b7fcb312c88c26d4e71a8", size = 370336 },
-    { url = "https://files.pythonhosted.org/packages/dc/6f/a5a1f43b6566831e9630e5bc5d86034a8884386297302be128402555dde1/msgpack-1.1.0-cp39-cp39-win32.whl", hash = "sha256:f80bc7d47f76089633763f952e67f8214cb7b3ee6bfa489b3cb6a84cfac114cd", size = 68683 },
-    { url = "https://files.pythonhosted.org/packages/5f/e8/2162621e18dbc36e2bc8492fd0e97b3975f5d89fe0472ae6d5f7fbdd8cf7/msgpack-1.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:4d1b7ff2d6146e16e8bd665ac726a89c74163ef8cd39fa8c1087d4e52d3a2325", size = 74787 },
+version = "1.1.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/45/b1/ea4f68038a18c77c9467400d166d74c4ffa536f34761f7983a104357e614/msgpack-1.1.1.tar.gz", hash = "sha256:77b79ce34a2bdab2594f490c8e80dd62a02d650b91a75159a63ec413b8d104cd", size = 173555, upload-time = "2025-06-13T06:52:51.324Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/33/52/f30da112c1dc92cf64f57d08a273ac771e7b29dea10b4b30369b2d7e8546/msgpack-1.1.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:353b6fc0c36fde68b661a12949d7d49f8f51ff5fa019c1e47c87c4ff34b080ed", size = 81799, upload-time = "2025-06-13T06:51:37.228Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/35/7bfc0def2f04ab4145f7f108e3563f9b4abae4ab0ed78a61f350518cc4d2/msgpack-1.1.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:79c408fcf76a958491b4e3b103d1c417044544b68e96d06432a189b43d1215c8", size = 78278, upload-time = "2025-06-13T06:51:38.534Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/c5/df5d6c1c39856bc55f800bf82778fd4c11370667f9b9e9d51b2f5da88f20/msgpack-1.1.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:78426096939c2c7482bf31ef15ca219a9e24460289c00dd0b94411040bb73ad2", size = 402805, upload-time = "2025-06-13T06:51:39.538Z" },
+    { url = "https://files.pythonhosted.org/packages/20/8e/0bb8c977efecfe6ea7116e2ed73a78a8d32a947f94d272586cf02a9757db/msgpack-1.1.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8b17ba27727a36cb73aabacaa44b13090feb88a01d012c0f4be70c00f75048b4", size = 408642, upload-time = "2025-06-13T06:51:41.092Z" },
+    { url = "https://files.pythonhosted.org/packages/59/a1/731d52c1aeec52006be6d1f8027c49fdc2cfc3ab7cbe7c28335b2910d7b6/msgpack-1.1.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:7a17ac1ea6ec3c7687d70201cfda3b1e8061466f28f686c24f627cae4ea8efd0", size = 395143, upload-time = "2025-06-13T06:51:42.575Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/92/b42911c52cda2ba67a6418ffa7d08969edf2e760b09015593c8a8a27a97d/msgpack-1.1.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:88d1e966c9235c1d4e2afac21ca83933ba59537e2e2727a999bf3f515ca2af26", size = 395986, upload-time = "2025-06-13T06:51:43.807Z" },
+    { url = "https://files.pythonhosted.org/packages/61/dc/8ae165337e70118d4dab651b8b562dd5066dd1e6dd57b038f32ebc3e2f07/msgpack-1.1.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:f6d58656842e1b2ddbe07f43f56b10a60f2ba5826164910968f5933e5178af75", size = 402682, upload-time = "2025-06-13T06:51:45.534Z" },
+    { url = "https://files.pythonhosted.org/packages/58/27/555851cb98dcbd6ce041df1eacb25ac30646575e9cd125681aa2f4b1b6f1/msgpack-1.1.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:96decdfc4adcbc087f5ea7ebdcfd3dee9a13358cae6e81d54be962efc38f6338", size = 406368, upload-time = "2025-06-13T06:51:46.97Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/64/39a26add4ce16f24e99eabb9005e44c663db00e3fce17d4ae1ae9d61df99/msgpack-1.1.1-cp310-cp310-win32.whl", hash = "sha256:6640fd979ca9a212e4bcdf6eb74051ade2c690b862b679bfcb60ae46e6dc4bfd", size = 65004, upload-time = "2025-06-13T06:51:48.582Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/18/73dfa3e9d5d7450d39debde5b0d848139f7de23bd637a4506e36c9800fd6/msgpack-1.1.1-cp310-cp310-win_amd64.whl", hash = "sha256:8b65b53204fe1bd037c40c4148d00ef918eb2108d24c9aaa20bc31f9810ce0a8", size = 71548, upload-time = "2025-06-13T06:51:49.558Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/83/97f24bf9848af23fe2ba04380388216defc49a8af6da0c28cc636d722502/msgpack-1.1.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:71ef05c1726884e44f8b1d1773604ab5d4d17729d8491403a705e649116c9558", size = 82728, upload-time = "2025-06-13T06:51:50.68Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/7f/2eaa388267a78401f6e182662b08a588ef4f3de6f0eab1ec09736a7aaa2b/msgpack-1.1.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:36043272c6aede309d29d56851f8841ba907a1a3d04435e43e8a19928e243c1d", size = 79279, upload-time = "2025-06-13T06:51:51.72Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/46/31eb60f4452c96161e4dfd26dbca562b4ec68c72e4ad07d9566d7ea35e8a/msgpack-1.1.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a32747b1b39c3ac27d0670122b57e6e57f28eefb725e0b625618d1b59bf9d1e0", size = 423859, upload-time = "2025-06-13T06:51:52.749Z" },
+    { url = "https://files.pythonhosted.org/packages/45/16/a20fa8c32825cc7ae8457fab45670c7a8996d7746ce80ce41cc51e3b2bd7/msgpack-1.1.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a8b10fdb84a43e50d38057b06901ec9da52baac6983d3f709d8507f3889d43f", size = 429975, upload-time = "2025-06-13T06:51:53.97Z" },
+    { url = "https://files.pythonhosted.org/packages/86/ea/6c958e07692367feeb1a1594d35e22b62f7f476f3c568b002a5ea09d443d/msgpack-1.1.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ba0c325c3f485dc54ec298d8b024e134acf07c10d494ffa24373bea729acf704", size = 413528, upload-time = "2025-06-13T06:51:55.507Z" },
+    { url = "https://files.pythonhosted.org/packages/75/05/ac84063c5dae79722bda9f68b878dc31fc3059adb8633c79f1e82c2cd946/msgpack-1.1.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:88daaf7d146e48ec71212ce21109b66e06a98e5e44dca47d853cbfe171d6c8d2", size = 413338, upload-time = "2025-06-13T06:51:57.023Z" },
+    { url = "https://files.pythonhosted.org/packages/69/e8/fe86b082c781d3e1c09ca0f4dacd457ede60a13119b6ce939efe2ea77b76/msgpack-1.1.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:d8b55ea20dc59b181d3f47103f113e6f28a5e1c89fd5b67b9140edb442ab67f2", size = 422658, upload-time = "2025-06-13T06:51:58.419Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/2b/bafc9924df52d8f3bb7c00d24e57be477f4d0f967c0a31ef5e2225e035c7/msgpack-1.1.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4a28e8072ae9779f20427af07f53bbb8b4aa81151054e882aee333b158da8752", size = 427124, upload-time = "2025-06-13T06:51:59.969Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/3b/1f717e17e53e0ed0b68fa59e9188f3f610c79d7151f0e52ff3cd8eb6b2dc/msgpack-1.1.1-cp311-cp311-win32.whl", hash = "sha256:7da8831f9a0fdb526621ba09a281fadc58ea12701bc709e7b8cbc362feabc295", size = 65016, upload-time = "2025-06-13T06:52:01.294Z" },
+    { url = "https://files.pythonhosted.org/packages/48/45/9d1780768d3b249accecc5a38c725eb1e203d44a191f7b7ff1941f7df60c/msgpack-1.1.1-cp311-cp311-win_amd64.whl", hash = "sha256:5fd1b58e1431008a57247d6e7cc4faa41c3607e8e7d4aaf81f7c29ea013cb458", size = 72267, upload-time = "2025-06-13T06:52:02.568Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/26/389b9c593eda2b8551b2e7126ad3a06af6f9b44274eb3a4f054d48ff7e47/msgpack-1.1.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:ae497b11f4c21558d95de9f64fff7053544f4d1a17731c866143ed6bb4591238", size = 82359, upload-time = "2025-06-13T06:52:03.909Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/65/7d1de38c8a22cf8b1551469159d4b6cf49be2126adc2482de50976084d78/msgpack-1.1.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:33be9ab121df9b6b461ff91baac6f2731f83d9b27ed948c5b9d1978ae28bf157", size = 79172, upload-time = "2025-06-13T06:52:05.246Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/bd/cacf208b64d9577a62c74b677e1ada005caa9b69a05a599889d6fc2ab20a/msgpack-1.1.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6f64ae8fe7ffba251fecb8408540c34ee9df1c26674c50c4544d72dbf792e5ce", size = 425013, upload-time = "2025-06-13T06:52:06.341Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/ec/fd869e2567cc9c01278a736cfd1697941ba0d4b81a43e0aa2e8d71dab208/msgpack-1.1.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a494554874691720ba5891c9b0b39474ba43ffb1aaf32a5dac874effb1619e1a", size = 426905, upload-time = "2025-06-13T06:52:07.501Z" },
+    { url = "https://files.pythonhosted.org/packages/55/2a/35860f33229075bce803a5593d046d8b489d7ba2fc85701e714fc1aaf898/msgpack-1.1.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:cb643284ab0ed26f6957d969fe0dd8bb17beb567beb8998140b5e38a90974f6c", size = 407336, upload-time = "2025-06-13T06:52:09.047Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/16/69ed8f3ada150bf92745fb4921bd621fd2cdf5a42e25eb50bcc57a5328f0/msgpack-1.1.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d275a9e3c81b1093c060c3837e580c37f47c51eca031f7b5fb76f7b8470f5f9b", size = 409485, upload-time = "2025-06-13T06:52:10.382Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/b6/0c398039e4c6d0b2e37c61d7e0e9d13439f91f780686deb8ee64ecf1ae71/msgpack-1.1.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:4fd6b577e4541676e0cc9ddc1709d25014d3ad9a66caa19962c4f5de30fc09ef", size = 412182, upload-time = "2025-06-13T06:52:11.644Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/d0/0cf4a6ecb9bc960d624c93effaeaae75cbf00b3bc4a54f35c8507273cda1/msgpack-1.1.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:bb29aaa613c0a1c40d1af111abf025f1732cab333f96f285d6a93b934738a68a", size = 419883, upload-time = "2025-06-13T06:52:12.806Z" },
+    { url = "https://files.pythonhosted.org/packages/62/83/9697c211720fa71a2dfb632cad6196a8af3abea56eece220fde4674dc44b/msgpack-1.1.1-cp312-cp312-win32.whl", hash = "sha256:870b9a626280c86cff9c576ec0d9cbcc54a1e5ebda9cd26dab12baf41fee218c", size = 65406, upload-time = "2025-06-13T06:52:14.271Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/23/0abb886e80eab08f5e8c485d6f13924028602829f63b8f5fa25a06636628/msgpack-1.1.1-cp312-cp312-win_amd64.whl", hash = "sha256:5692095123007180dca3e788bb4c399cc26626da51629a31d40207cb262e67f4", size = 72558, upload-time = "2025-06-13T06:52:15.252Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/bd/0792be119d7fe7dc2148689ef65c90507d82d20a204aab3b98c74a1f8684/msgpack-1.1.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f5be6b6bc52fad84d010cb45433720327ce886009d862f46b26d4d154001994b", size = 81882, upload-time = "2025-06-13T06:52:39.316Z" },
+    { url = "https://files.pythonhosted.org/packages/75/77/ce06c8e26a816ae8730a8e030d263c5289adcaff9f0476f9b270bdd7c5c2/msgpack-1.1.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3a89cd8c087ea67e64844287ea52888239cbd2940884eafd2dcd25754fb72232", size = 78414, upload-time = "2025-06-13T06:52:40.341Z" },
+    { url = "https://files.pythonhosted.org/packages/73/27/190576c497677fb4a0d05d896b24aea6cdccd910f206aaa7b511901befed/msgpack-1.1.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1d75f3807a9900a7d575d8d6674a3a47e9f227e8716256f35bc6f03fc597ffbf", size = 400927, upload-time = "2025-06-13T06:52:41.399Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/af/6a0aa5a06762e70726ec3c10fb966600d84a7220b52635cb0ab2dc64d32f/msgpack-1.1.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d182dac0221eb8faef2e6f44701812b467c02674a322c739355c39e94730cdbf", size = 405903, upload-time = "2025-06-13T06:52:42.699Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/80/3f3da358cecbbe8eb12360814bd1277d59d2608485934742a074d99894a9/msgpack-1.1.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1b13fe0fb4aac1aa5320cd693b297fe6fdef0e7bea5518cbc2dd5299f873ae90", size = 393192, upload-time = "2025-06-13T06:52:43.986Z" },
+    { url = "https://files.pythonhosted.org/packages/98/c6/3a0ec7fdebbb4f3f8f254696cd91d491c29c501dbebd86286c17e8f68cd7/msgpack-1.1.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:435807eeb1bc791ceb3247d13c79868deb22184e1fc4224808750f0d7d1affc1", size = 393851, upload-time = "2025-06-13T06:52:45.177Z" },
+    { url = "https://files.pythonhosted.org/packages/39/37/df50d5f8e68514b60fbe70f6e8337ea2b32ae2be030871bcd9d1cf7d4b62/msgpack-1.1.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:4835d17af722609a45e16037bb1d4d78b7bdf19d6c0128116d178956618c4e88", size = 400292, upload-time = "2025-06-13T06:52:46.381Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/ec/1e067292e02d2ceb4c8cb5ba222c4f7bb28730eef5676740609dc2627e0f/msgpack-1.1.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:a8ef6e342c137888ebbfb233e02b8fbd689bb5b5fcc59b34711ac47ebd504478", size = 401873, upload-time = "2025-06-13T06:52:47.957Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/31/e8c9c6b5b58d64c9efa99c8d181fcc25f38ead357b0360379fbc8a4234ad/msgpack-1.1.1-cp39-cp39-win32.whl", hash = "sha256:61abccf9de335d9efd149e2fff97ed5974f2481b3353772e8e2dd3402ba2bd57", size = 65028, upload-time = "2025-06-13T06:52:49.166Z" },
+    { url = "https://files.pythonhosted.org/packages/20/d6/cd62cded572e5e25892747a5d27850170bcd03c855e9c69c538e024de6f9/msgpack-1.1.1-cp39-cp39-win_amd64.whl", hash = "sha256:40eae974c873b2992fd36424a5d9407f93e97656d999f43fca9d29f820899084", size = 71700, upload-time = "2025-06-13T06:52:50.244Z" },
 ]
 
 [[package]]
 name = "multidict"
-version = "6.1.0"
+version = "6.6.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/d6/be/504b89a5e9ca731cd47487e91c469064f8ae5af93b7259758dcfc2b9c848/multidict-6.1.0.tar.gz", hash = "sha256:22ae2ebf9b0c69d206c003e2f6a914ea33f0a932d4aa16f236afc049d9958f4a", size = 64002 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/29/68/259dee7fd14cf56a17c554125e534f6274c2860159692a414d0b402b9a6d/multidict-6.1.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:3380252550e372e8511d49481bd836264c009adb826b23fefcc5dd3c69692f60", size = 48628 },
-    { url = "https://files.pythonhosted.org/packages/50/79/53ba256069fe5386a4a9e80d4e12857ced9de295baf3e20c68cdda746e04/multidict-6.1.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:99f826cbf970077383d7de805c0681799491cb939c25450b9b5b3ced03ca99f1", size = 29327 },
-    { url = "https://files.pythonhosted.org/packages/ff/10/71f1379b05b196dae749b5ac062e87273e3f11634f447ebac12a571d90ae/multidict-6.1.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a114d03b938376557927ab23f1e950827c3b893ccb94b62fd95d430fd0e5cf53", size = 29689 },
-    { url = "https://files.pythonhosted.org/packages/71/45/70bac4f87438ded36ad4793793c0095de6572d433d98575a5752629ef549/multidict-6.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b1c416351ee6271b2f49b56ad7f308072f6f44b37118d69c2cad94f3fa8a40d5", size = 126639 },
-    { url = "https://files.pythonhosted.org/packages/80/cf/17f35b3b9509b4959303c05379c4bfb0d7dd05c3306039fc79cf035bbac0/multidict-6.1.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6b5d83030255983181005e6cfbac1617ce9746b219bc2aad52201ad121226581", size = 134315 },
-    { url = "https://files.pythonhosted.org/packages/ef/1f/652d70ab5effb33c031510a3503d4d6efc5ec93153562f1ee0acdc895a57/multidict-6.1.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3e97b5e938051226dc025ec80980c285b053ffb1e25a3db2a3aa3bc046bf7f56", size = 129471 },
-    { url = "https://files.pythonhosted.org/packages/a6/64/2dd6c4c681688c0165dea3975a6a4eab4944ea30f35000f8b8af1df3148c/multidict-6.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d618649d4e70ac6efcbba75be98b26ef5078faad23592f9b51ca492953012429", size = 124585 },
-    { url = "https://files.pythonhosted.org/packages/87/56/e6ee5459894c7e554b57ba88f7257dc3c3d2d379cb15baaa1e265b8c6165/multidict-6.1.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:10524ebd769727ac77ef2278390fb0068d83f3acb7773792a5080f2b0abf7748", size = 116957 },
-    { url = "https://files.pythonhosted.org/packages/36/9e/616ce5e8d375c24b84f14fc263c7ef1d8d5e8ef529dbc0f1df8ce71bb5b8/multidict-6.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ff3827aef427c89a25cc96ded1759271a93603aba9fb977a6d264648ebf989db", size = 128609 },
-    { url = "https://files.pythonhosted.org/packages/8c/4f/4783e48a38495d000f2124020dc96bacc806a4340345211b1ab6175a6cb4/multidict-6.1.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:06809f4f0f7ab7ea2cabf9caca7d79c22c0758b58a71f9d32943ae13c7ace056", size = 123016 },
-    { url = "https://files.pythonhosted.org/packages/3e/b3/4950551ab8fc39862ba5e9907dc821f896aa829b4524b4deefd3e12945ab/multidict-6.1.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:f179dee3b863ab1c59580ff60f9d99f632f34ccb38bf67a33ec6b3ecadd0fd76", size = 133542 },
-    { url = "https://files.pythonhosted.org/packages/96/4d/f0ce6ac9914168a2a71df117935bb1f1781916acdecbb43285e225b484b8/multidict-6.1.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:aaed8b0562be4a0876ee3b6946f6869b7bcdb571a5d1496683505944e268b160", size = 130163 },
-    { url = "https://files.pythonhosted.org/packages/be/72/17c9f67e7542a49dd252c5ae50248607dfb780bcc03035907dafefb067e3/multidict-6.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:3c8b88a2ccf5493b6c8da9076fb151ba106960a2df90c2633f342f120751a9e7", size = 126832 },
-    { url = "https://files.pythonhosted.org/packages/71/9f/72d719e248cbd755c8736c6d14780533a1606ffb3fbb0fbd77da9f0372da/multidict-6.1.0-cp310-cp310-win32.whl", hash = "sha256:4a9cb68166a34117d6646c0023c7b759bf197bee5ad4272f420a0141d7eb03a0", size = 26402 },
-    { url = "https://files.pythonhosted.org/packages/04/5a/d88cd5d00a184e1ddffc82aa2e6e915164a6d2641ed3606e766b5d2f275a/multidict-6.1.0-cp310-cp310-win_amd64.whl", hash = "sha256:20b9b5fbe0b88d0bdef2012ef7dee867f874b72528cf1d08f1d59b0e3850129d", size = 28800 },
-    { url = "https://files.pythonhosted.org/packages/93/13/df3505a46d0cd08428e4c8169a196131d1b0c4b515c3649829258843dde6/multidict-6.1.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:3efe2c2cb5763f2f1b275ad2bf7a287d3f7ebbef35648a9726e3b69284a4f3d6", size = 48570 },
-    { url = "https://files.pythonhosted.org/packages/f0/e1/a215908bfae1343cdb72f805366592bdd60487b4232d039c437fe8f5013d/multidict-6.1.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c7053d3b0353a8b9de430a4f4b4268ac9a4fb3481af37dfe49825bf45ca24156", size = 29316 },
-    { url = "https://files.pythonhosted.org/packages/70/0f/6dc70ddf5d442702ed74f298d69977f904960b82368532c88e854b79f72b/multidict-6.1.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:27e5fc84ccef8dfaabb09d82b7d179c7cf1a3fbc8a966f8274fcb4ab2eb4cadb", size = 29640 },
-    { url = "https://files.pythonhosted.org/packages/d8/6d/9c87b73a13d1cdea30b321ef4b3824449866bd7f7127eceed066ccb9b9ff/multidict-6.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0e2b90b43e696f25c62656389d32236e049568b39320e2735d51f08fd362761b", size = 131067 },
-    { url = "https://files.pythonhosted.org/packages/cc/1e/1b34154fef373371fd6c65125b3d42ff5f56c7ccc6bfff91b9b3c60ae9e0/multidict-6.1.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d83a047959d38a7ff552ff94be767b7fd79b831ad1cd9920662db05fec24fe72", size = 138507 },
-    { url = "https://files.pythonhosted.org/packages/fb/e0/0bc6b2bac6e461822b5f575eae85da6aae76d0e2a79b6665d6206b8e2e48/multidict-6.1.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:d1a9dd711d0877a1ece3d2e4fea11a8e75741ca21954c919406b44e7cf971304", size = 133905 },
-    { url = "https://files.pythonhosted.org/packages/ba/af/73d13b918071ff9b2205fcf773d316e0f8fefb4ec65354bbcf0b10908cc6/multidict-6.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ec2abea24d98246b94913b76a125e855eb5c434f7c46546046372fe60f666351", size = 129004 },
-    { url = "https://files.pythonhosted.org/packages/74/21/23960627b00ed39643302d81bcda44c9444ebcdc04ee5bedd0757513f259/multidict-6.1.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4867cafcbc6585e4b678876c489b9273b13e9fff9f6d6d66add5e15d11d926cb", size = 121308 },
-    { url = "https://files.pythonhosted.org/packages/8b/5c/cf282263ffce4a596ed0bb2aa1a1dddfe1996d6a62d08842a8d4b33dca13/multidict-6.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:5b48204e8d955c47c55b72779802b219a39acc3ee3d0116d5080c388970b76e3", size = 132608 },
-    { url = "https://files.pythonhosted.org/packages/d7/3e/97e778c041c72063f42b290888daff008d3ab1427f5b09b714f5a8eff294/multidict-6.1.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:d8fff389528cad1618fb4b26b95550327495462cd745d879a8c7c2115248e399", size = 127029 },
-    { url = "https://files.pythonhosted.org/packages/47/ac/3efb7bfe2f3aefcf8d103e9a7162572f01936155ab2f7ebcc7c255a23212/multidict-6.1.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a7a9541cd308eed5e30318430a9c74d2132e9a8cb46b901326272d780bf2d423", size = 137594 },
-    { url = "https://files.pythonhosted.org/packages/42/9b/6c6e9e8dc4f915fc90a9b7798c44a30773dea2995fdcb619870e705afe2b/multidict-6.1.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:da1758c76f50c39a2efd5e9859ce7d776317eb1dd34317c8152ac9251fc574a3", size = 134556 },
-    { url = "https://files.pythonhosted.org/packages/1d/10/8e881743b26aaf718379a14ac58572a240e8293a1c9d68e1418fb11c0f90/multidict-6.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:c943a53e9186688b45b323602298ab727d8865d8c9ee0b17f8d62d14b56f0753", size = 130993 },
-    { url = "https://files.pythonhosted.org/packages/45/84/3eb91b4b557442802d058a7579e864b329968c8d0ea57d907e7023c677f2/multidict-6.1.0-cp311-cp311-win32.whl", hash = "sha256:90f8717cb649eea3504091e640a1b8568faad18bd4b9fcd692853a04475a4b80", size = 26405 },
-    { url = "https://files.pythonhosted.org/packages/9f/0b/ad879847ecbf6d27e90a6eabb7eff6b62c129eefe617ea45eae7c1f0aead/multidict-6.1.0-cp311-cp311-win_amd64.whl", hash = "sha256:82176036e65644a6cc5bd619f65f6f19781e8ec2e5330f51aa9ada7504cc1926", size = 28795 },
-    { url = "https://files.pythonhosted.org/packages/fd/16/92057c74ba3b96d5e211b553895cd6dc7cc4d1e43d9ab8fafc727681ef71/multidict-6.1.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:b04772ed465fa3cc947db808fa306d79b43e896beb677a56fb2347ca1a49c1fa", size = 48713 },
-    { url = "https://files.pythonhosted.org/packages/94/3d/37d1b8893ae79716179540b89fc6a0ee56b4a65fcc0d63535c6f5d96f217/multidict-6.1.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:6180c0ae073bddeb5a97a38c03f30c233e0a4d39cd86166251617d1bbd0af436", size = 29516 },
-    { url = "https://files.pythonhosted.org/packages/a2/12/adb6b3200c363062f805275b4c1e656be2b3681aada66c80129932ff0bae/multidict-6.1.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:071120490b47aa997cca00666923a83f02c7fbb44f71cf7f136df753f7fa8761", size = 29557 },
-    { url = "https://files.pythonhosted.org/packages/47/e9/604bb05e6e5bce1e6a5cf80a474e0f072e80d8ac105f1b994a53e0b28c42/multidict-6.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:50b3a2710631848991d0bf7de077502e8994c804bb805aeb2925a981de58ec2e", size = 130170 },
-    { url = "https://files.pythonhosted.org/packages/7e/13/9efa50801785eccbf7086b3c83b71a4fb501a4d43549c2f2f80b8787d69f/multidict-6.1.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b58c621844d55e71c1b7f7c498ce5aa6985d743a1a59034c57a905b3f153c1ef", size = 134836 },
-    { url = "https://files.pythonhosted.org/packages/bf/0f/93808b765192780d117814a6dfcc2e75de6dcc610009ad408b8814dca3ba/multidict-6.1.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:55b6d90641869892caa9ca42ff913f7ff1c5ece06474fbd32fb2cf6834726c95", size = 133475 },
-    { url = "https://files.pythonhosted.org/packages/d3/c8/529101d7176fe7dfe1d99604e48d69c5dfdcadb4f06561f465c8ef12b4df/multidict-6.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b820514bfc0b98a30e3d85462084779900347e4d49267f747ff54060cc33925", size = 131049 },
-    { url = "https://files.pythonhosted.org/packages/ca/0c/fc85b439014d5a58063e19c3a158a889deec399d47b5269a0f3b6a2e28bc/multidict-6.1.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:10a9b09aba0c5b48c53761b7c720aaaf7cf236d5fe394cd399c7ba662d5f9966", size = 120370 },
-    { url = "https://files.pythonhosted.org/packages/db/46/d4416eb20176492d2258fbd47b4abe729ff3b6e9c829ea4236f93c865089/multidict-6.1.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:1e16bf3e5fc9f44632affb159d30a437bfe286ce9e02754759be5536b169b305", size = 125178 },
-    { url = "https://files.pythonhosted.org/packages/5b/46/73697ad7ec521df7de5531a32780bbfd908ded0643cbe457f981a701457c/multidict-6.1.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:76f364861c3bfc98cbbcbd402d83454ed9e01a5224bb3a28bf70002a230f73e2", size = 119567 },
-    { url = "https://files.pythonhosted.org/packages/cd/ed/51f060e2cb0e7635329fa6ff930aa5cffa17f4c7f5c6c3ddc3500708e2f2/multidict-6.1.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:820c661588bd01a0aa62a1283f20d2be4281b086f80dad9e955e690c75fb54a2", size = 129822 },
-    { url = "https://files.pythonhosted.org/packages/df/9e/ee7d1954b1331da3eddea0c4e08d9142da5f14b1321c7301f5014f49d492/multidict-6.1.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:0e5f362e895bc5b9e67fe6e4ded2492d8124bdf817827f33c5b46c2fe3ffaca6", size = 128656 },
-    { url = "https://files.pythonhosted.org/packages/77/00/8538f11e3356b5d95fa4b024aa566cde7a38aa7a5f08f4912b32a037c5dc/multidict-6.1.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:3ec660d19bbc671e3a6443325f07263be452c453ac9e512f5eb935e7d4ac28b3", size = 125360 },
-    { url = "https://files.pythonhosted.org/packages/be/05/5d334c1f2462d43fec2363cd00b1c44c93a78c3925d952e9a71caf662e96/multidict-6.1.0-cp312-cp312-win32.whl", hash = "sha256:58130ecf8f7b8112cdb841486404f1282b9c86ccb30d3519faf301b2e5659133", size = 26382 },
-    { url = "https://files.pythonhosted.org/packages/a3/bf/f332a13486b1ed0496d624bcc7e8357bb8053823e8cd4b9a18edc1d97e73/multidict-6.1.0-cp312-cp312-win_amd64.whl", hash = "sha256:188215fc0aafb8e03341995e7c4797860181562380f81ed0a87ff455b70bf1f1", size = 28529 },
-    { url = "https://files.pythonhosted.org/packages/22/67/1c7c0f39fe069aa4e5d794f323be24bf4d33d62d2a348acdb7991f8f30db/multidict-6.1.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d569388c381b24671589335a3be6e1d45546c2988c2ebe30fdcada8457a31008", size = 48771 },
-    { url = "https://files.pythonhosted.org/packages/3c/25/c186ee7b212bdf0df2519eacfb1981a017bda34392c67542c274651daf23/multidict-6.1.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:052e10d2d37810b99cc170b785945421141bf7bb7d2f8799d431e7db229c385f", size = 29533 },
-    { url = "https://files.pythonhosted.org/packages/67/5e/04575fd837e0958e324ca035b339cea174554f6f641d3fb2b4f2e7ff44a2/multidict-6.1.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f90c822a402cb865e396a504f9fc8173ef34212a342d92e362ca498cad308e28", size = 29595 },
-    { url = "https://files.pythonhosted.org/packages/d3/b2/e56388f86663810c07cfe4a3c3d87227f3811eeb2d08450b9e5d19d78876/multidict-6.1.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b225d95519a5bf73860323e633a664b0d85ad3d5bede6d30d95b35d4dfe8805b", size = 130094 },
-    { url = "https://files.pythonhosted.org/packages/6c/ee/30ae9b4186a644d284543d55d491fbd4239b015d36b23fea43b4c94f7052/multidict-6.1.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:23bfd518810af7de1116313ebd9092cb9aa629beb12f6ed631ad53356ed6b86c", size = 134876 },
-    { url = "https://files.pythonhosted.org/packages/84/c7/70461c13ba8ce3c779503c70ec9d0345ae84de04521c1f45a04d5f48943d/multidict-6.1.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5c09fcfdccdd0b57867577b719c69e347a436b86cd83747f179dbf0cc0d4c1f3", size = 133500 },
-    { url = "https://files.pythonhosted.org/packages/4a/9f/002af221253f10f99959561123fae676148dd730e2daa2cd053846a58507/multidict-6.1.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bf6bea52ec97e95560af5ae576bdac3aa3aae0b6758c6efa115236d9e07dae44", size = 131099 },
-    { url = "https://files.pythonhosted.org/packages/82/42/d1c7a7301d52af79d88548a97e297f9d99c961ad76bbe6f67442bb77f097/multidict-6.1.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:57feec87371dbb3520da6192213c7d6fc892d5589a93db548331954de8248fd2", size = 120403 },
-    { url = "https://files.pythonhosted.org/packages/68/f3/471985c2c7ac707547553e8f37cff5158030d36bdec4414cb825fbaa5327/multidict-6.1.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:0c3f390dc53279cbc8ba976e5f8035eab997829066756d811616b652b00a23a3", size = 125348 },
-    { url = "https://files.pythonhosted.org/packages/67/2c/e6df05c77e0e433c214ec1d21ddd203d9a4770a1f2866a8ca40a545869a0/multidict-6.1.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:59bfeae4b25ec05b34f1956eaa1cb38032282cd4dfabc5056d0a1ec4d696d3aa", size = 119673 },
-    { url = "https://files.pythonhosted.org/packages/c5/cd/bc8608fff06239c9fb333f9db7743a1b2eafe98c2666c9a196e867a3a0a4/multidict-6.1.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:b2f59caeaf7632cc633b5cf6fc449372b83bbdf0da4ae04d5be36118e46cc0aa", size = 129927 },
-    { url = "https://files.pythonhosted.org/packages/44/8e/281b69b7bc84fc963a44dc6e0bbcc7150e517b91df368a27834299a526ac/multidict-6.1.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:37bb93b2178e02b7b618893990941900fd25b6b9ac0fa49931a40aecdf083fe4", size = 128711 },
-    { url = "https://files.pythonhosted.org/packages/12/a4/63e7cd38ed29dd9f1881d5119f272c898ca92536cdb53ffe0843197f6c85/multidict-6.1.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:4e9f48f58c2c523d5a06faea47866cd35b32655c46b443f163d08c6d0ddb17d6", size = 125519 },
-    { url = "https://files.pythonhosted.org/packages/38/e0/4f5855037a72cd8a7a2f60a3952d9aa45feedb37ae7831642102604e8a37/multidict-6.1.0-cp313-cp313-win32.whl", hash = "sha256:3a37ffb35399029b45c6cc33640a92bef403c9fd388acce75cdc88f58bd19a81", size = 26426 },
-    { url = "https://files.pythonhosted.org/packages/7e/a5/17ee3a4db1e310b7405f5d25834460073a8ccd86198ce044dfaf69eac073/multidict-6.1.0-cp313-cp313-win_amd64.whl", hash = "sha256:e9aa71e15d9d9beaad2c6b9319edcdc0a49a43ef5c0a4c8265ca9ee7d6c67774", size = 28531 },
-    { url = "https://files.pythonhosted.org/packages/e7/c9/9e153a6572b38ac5ff4434113af38acf8d5e9957897cdb1f513b3d6614ed/multidict-6.1.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:4e18b656c5e844539d506a0a06432274d7bd52a7487e6828c63a63d69185626c", size = 48550 },
-    { url = "https://files.pythonhosted.org/packages/76/f5/79565ddb629eba6c7f704f09a09df085c8dc04643b12506f10f718cee37a/multidict-6.1.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a185f876e69897a6f3325c3f19f26a297fa058c5e456bfcff8015e9a27e83ae1", size = 29298 },
-    { url = "https://files.pythonhosted.org/packages/60/1b/9851878b704bc98e641a3e0bce49382ae9e05743dac6d97748feb5b7baba/multidict-6.1.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:ab7c4ceb38d91570a650dba194e1ca87c2b543488fe9309b4212694174fd539c", size = 29641 },
-    { url = "https://files.pythonhosted.org/packages/89/87/d451d45aab9e422cb0fb2f7720c31a4c1d3012c740483c37f642eba568fb/multidict-6.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e617fb6b0b6953fffd762669610c1c4ffd05632c138d61ac7e14ad187870669c", size = 126202 },
-    { url = "https://files.pythonhosted.org/packages/fa/b4/27cbe9f3e2e469359887653f2e45470272eef7295139916cc21107c6b48c/multidict-6.1.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:16e5f4bf4e603eb1fdd5d8180f1a25f30056f22e55ce51fb3d6ad4ab29f7d96f", size = 133925 },
-    { url = "https://files.pythonhosted.org/packages/4d/a3/afc841899face8adfd004235ce759a37619f6ec99eafd959650c5ce4df57/multidict-6.1.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f4c035da3f544b1882bac24115f3e2e8760f10a0107614fc9839fd232200b875", size = 129039 },
-    { url = "https://files.pythonhosted.org/packages/5e/41/0d0fb18c1ad574f807196f5f3d99164edf9de3e169a58c6dc2d6ed5742b9/multidict-6.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:957cf8e4b6e123a9eea554fa7ebc85674674b713551de587eb318a2df3e00255", size = 124072 },
-    { url = "https://files.pythonhosted.org/packages/00/22/defd7a2e71a44e6e5b9a5428f972e5b572e7fe28e404dfa6519bbf057c93/multidict-6.1.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:483a6aea59cb89904e1ceabd2b47368b5600fb7de78a6e4a2c2987b2d256cf30", size = 116532 },
-    { url = "https://files.pythonhosted.org/packages/91/25/f7545102def0b1d456ab6449388eed2dfd822debba1d65af60194904a23a/multidict-6.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:87701f25a2352e5bf7454caa64757642734da9f6b11384c1f9d1a8e699758057", size = 128173 },
-    { url = "https://files.pythonhosted.org/packages/45/79/3dbe8d35fc99f5ea610813a72ab55f426cb9cf482f860fa8496e5409be11/multidict-6.1.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:682b987361e5fd7a139ed565e30d81fd81e9629acc7d925a205366877d8c8657", size = 122654 },
-    { url = "https://files.pythonhosted.org/packages/97/cb/209e735eeab96e1b160825b5d0b36c56d3862abff828fc43999bb957dcad/multidict-6.1.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:ce2186a7df133a9c895dea3331ddc5ddad42cdd0d1ea2f0a51e5d161e4762f28", size = 133197 },
-    { url = "https://files.pythonhosted.org/packages/e4/3a/a13808a7ada62808afccea67837a79d00ad6581440015ef00f726d064c2d/multidict-6.1.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:9f636b730f7e8cb19feb87094949ba54ee5357440b9658b2a32a5ce4bce53972", size = 129754 },
-    { url = "https://files.pythonhosted.org/packages/77/dd/8540e139eafb240079242da8f8ffdf9d3f4b4ad1aac5a786cd4050923783/multidict-6.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:73eae06aa53af2ea5270cc066dcaf02cc60d2994bbb2c4ef5764949257d10f43", size = 126402 },
-    { url = "https://files.pythonhosted.org/packages/86/99/e82e1a275d8b1ea16d3a251474262258dbbe41c05cce0c01bceda1fc8ea5/multidict-6.1.0-cp39-cp39-win32.whl", hash = "sha256:1ca0083e80e791cffc6efce7660ad24af66c8d4079d2a750b29001b53ff59ada", size = 26421 },
-    { url = "https://files.pythonhosted.org/packages/86/1c/9fa630272355af7e4446a2c7550c259f11ee422ab2d30ff90a0a71cf3d9e/multidict-6.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:aa466da5b15ccea564bdab9c89175c762bc12825f4659c11227f515cee76fa4a", size = 28791 },
-    { url = "https://files.pythonhosted.org/packages/99/b7/b9e70fde2c0f0c9af4cc5277782a89b66d35948ea3369ec9f598358c3ac5/multidict-6.1.0-py3-none-any.whl", hash = "sha256:48e171e52d1c4d33888e529b999e5900356b9ae588c2f09a52dcefb158b27506", size = 10051 },
+sdist = { url = "https://files.pythonhosted.org/packages/69/7f/0652e6ed47ab288e3756ea9c0df8b14950781184d4bd7883f4d87dd41245/multidict-6.6.4.tar.gz", hash = "sha256:d2d4e4787672911b48350df02ed3fa3fffdc2f2e8ca06dd6afdf34189b76a9dd", size = 101843, upload-time = "2025-08-11T12:08:48.217Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/eb/6b/86f353088c1358e76fd30b0146947fddecee812703b604ee901e85cd2a80/multidict-6.6.4-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:b8aa6f0bd8125ddd04a6593437bad6a7e70f300ff4180a531654aa2ab3f6d58f", size = 77054, upload-time = "2025-08-11T12:06:02.99Z" },
+    { url = "https://files.pythonhosted.org/packages/19/5d/c01dc3d3788bb877bd7f5753ea6eb23c1beeca8044902a8f5bfb54430f63/multidict-6.6.4-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b9e5853bbd7264baca42ffc53391b490d65fe62849bf2c690fa3f6273dbcd0cb", size = 44914, upload-time = "2025-08-11T12:06:05.264Z" },
+    { url = "https://files.pythonhosted.org/packages/46/44/964dae19ea42f7d3e166474d8205f14bb811020e28bc423d46123ddda763/multidict-6.6.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0af5f9dee472371e36d6ae38bde009bd8ce65ac7335f55dcc240379d7bed1495", size = 44601, upload-time = "2025-08-11T12:06:06.627Z" },
+    { url = "https://files.pythonhosted.org/packages/31/20/0616348a1dfb36cb2ab33fc9521de1f27235a397bf3f59338e583afadd17/multidict-6.6.4-cp310-cp310-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:d24f351e4d759f5054b641c81e8291e5d122af0fca5c72454ff77f7cbe492de8", size = 224821, upload-time = "2025-08-11T12:06:08.06Z" },
+    { url = "https://files.pythonhosted.org/packages/14/26/5d8923c69c110ff51861af05bd27ca6783011b96725d59ccae6d9daeb627/multidict-6.6.4-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:db6a3810eec08280a172a6cd541ff4a5f6a97b161d93ec94e6c4018917deb6b7", size = 242608, upload-time = "2025-08-11T12:06:09.697Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/cc/e2ad3ba9459aa34fa65cf1f82a5c4a820a2ce615aacfb5143b8817f76504/multidict-6.6.4-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:a1b20a9d56b2d81e2ff52ecc0670d583eaabaa55f402e8d16dd062373dbbe796", size = 222324, upload-time = "2025-08-11T12:06:10.905Z" },
+    { url = "https://files.pythonhosted.org/packages/19/db/4ed0f65701afbc2cb0c140d2d02928bb0fe38dd044af76e58ad7c54fd21f/multidict-6.6.4-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8c9854df0eaa610a23494c32a6f44a3a550fb398b6b51a56e8c6b9b3689578db", size = 253234, upload-time = "2025-08-11T12:06:12.658Z" },
+    { url = "https://files.pythonhosted.org/packages/94/c1/5160c9813269e39ae14b73debb907bfaaa1beee1762da8c4fb95df4764ed/multidict-6.6.4-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4bb7627fd7a968f41905a4d6343b0d63244a0623f006e9ed989fa2b78f4438a0", size = 251613, upload-time = "2025-08-11T12:06:13.97Z" },
+    { url = "https://files.pythonhosted.org/packages/05/a9/48d1bd111fc2f8fb98b2ed7f9a115c55a9355358432a19f53c0b74d8425d/multidict-6.6.4-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:caebafea30ed049c57c673d0b36238b1748683be2593965614d7b0e99125c877", size = 241649, upload-time = "2025-08-11T12:06:15.204Z" },
+    { url = "https://files.pythonhosted.org/packages/85/2a/f7d743df0019408768af8a70d2037546a2be7b81fbb65f040d76caafd4c5/multidict-6.6.4-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ad887a8250eb47d3ab083d2f98db7f48098d13d42eb7a3b67d8a5c795f224ace", size = 239238, upload-time = "2025-08-11T12:06:16.467Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/b8/4f4bb13323c2d647323f7919201493cf48ebe7ded971717bfb0f1a79b6bf/multidict-6.6.4-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:ed8358ae7d94ffb7c397cecb62cbac9578a83ecefc1eba27b9090ee910e2efb6", size = 233517, upload-time = "2025-08-11T12:06:18.107Z" },
+    { url = "https://files.pythonhosted.org/packages/33/29/4293c26029ebfbba4f574febd2ed01b6f619cfa0d2e344217d53eef34192/multidict-6.6.4-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:ecab51ad2462197a4c000b6d5701fc8585b80eecb90583635d7e327b7b6923eb", size = 243122, upload-time = "2025-08-11T12:06:19.361Z" },
+    { url = "https://files.pythonhosted.org/packages/20/60/a1c53628168aa22447bfde3a8730096ac28086704a0d8c590f3b63388d0c/multidict-6.6.4-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:c5c97aa666cf70e667dfa5af945424ba1329af5dd988a437efeb3a09430389fb", size = 248992, upload-time = "2025-08-11T12:06:20.661Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/3b/55443a0c372f33cae5d9ec37a6a973802884fa0ab3586659b197cf8cc5e9/multidict-6.6.4-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:9a950b7cf54099c1209f455ac5970b1ea81410f2af60ed9eb3c3f14f0bfcf987", size = 243708, upload-time = "2025-08-11T12:06:21.891Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/60/a18c6900086769312560b2626b18e8cca22d9e85b1186ba77f4755b11266/multidict-6.6.4-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:163c7ea522ea9365a8a57832dea7618e6cbdc3cd75f8c627663587459a4e328f", size = 237498, upload-time = "2025-08-11T12:06:23.206Z" },
+    { url = "https://files.pythonhosted.org/packages/11/3d/8bdd8bcaff2951ce2affccca107a404925a2beafedd5aef0b5e4a71120a6/multidict-6.6.4-cp310-cp310-win32.whl", hash = "sha256:17d2cbbfa6ff20821396b25890f155f40c986f9cfbce5667759696d83504954f", size = 41415, upload-time = "2025-08-11T12:06:24.77Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/53/cab1ad80356a4cd1b685a254b680167059b433b573e53872fab245e9fc95/multidict-6.6.4-cp310-cp310-win_amd64.whl", hash = "sha256:ce9a40fbe52e57e7edf20113a4eaddfacac0561a0879734e636aa6d4bb5e3fb0", size = 46046, upload-time = "2025-08-11T12:06:25.893Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/9a/874212b6f5c1c2d870d0a7adc5bb4cfe9b0624fa15cdf5cf757c0f5087ae/multidict-6.6.4-cp310-cp310-win_arm64.whl", hash = "sha256:01d0959807a451fe9fdd4da3e139cb5b77f7328baf2140feeaf233e1d777b729", size = 43147, upload-time = "2025-08-11T12:06:27.534Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/7f/90a7f01e2d005d6653c689039977f6856718c75c5579445effb7e60923d1/multidict-6.6.4-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:c7a0e9b561e6460484318a7612e725df1145d46b0ef57c6b9866441bf6e27e0c", size = 76472, upload-time = "2025-08-11T12:06:29.006Z" },
+    { url = "https://files.pythonhosted.org/packages/54/a3/bed07bc9e2bb302ce752f1dabc69e884cd6a676da44fb0e501b246031fdd/multidict-6.6.4-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:6bf2f10f70acc7a2446965ffbc726e5fc0b272c97a90b485857e5c70022213eb", size = 44634, upload-time = "2025-08-11T12:06:30.374Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/4b/ceeb4f8f33cf81277da464307afeaf164fb0297947642585884f5cad4f28/multidict-6.6.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:66247d72ed62d5dd29752ffc1d3b88f135c6a8de8b5f63b7c14e973ef5bda19e", size = 44282, upload-time = "2025-08-11T12:06:31.958Z" },
+    { url = "https://files.pythonhosted.org/packages/03/35/436a5da8702b06866189b69f655ffdb8f70796252a8772a77815f1812679/multidict-6.6.4-cp311-cp311-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:105245cc6b76f51e408451a844a54e6823bbd5a490ebfe5bdfc79798511ceded", size = 229696, upload-time = "2025-08-11T12:06:33.087Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/0e/915160be8fecf1fca35f790c08fb74ca684d752fcba62c11daaf3d92c216/multidict-6.6.4-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cbbc54e58b34c3bae389ef00046be0961f30fef7cb0dd9c7756aee376a4f7683", size = 246665, upload-time = "2025-08-11T12:06:34.448Z" },
+    { url = "https://files.pythonhosted.org/packages/08/ee/2f464330acd83f77dcc346f0b1a0eaae10230291450887f96b204b8ac4d3/multidict-6.6.4-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:56c6b3652f945c9bc3ac6c8178cd93132b8d82dd581fcbc3a00676c51302bc1a", size = 225485, upload-time = "2025-08-11T12:06:35.672Z" },
+    { url = "https://files.pythonhosted.org/packages/71/cc/9a117f828b4d7fbaec6adeed2204f211e9caf0a012692a1ee32169f846ae/multidict-6.6.4-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b95494daf857602eccf4c18ca33337dd2be705bccdb6dddbfc9d513e6addb9d9", size = 257318, upload-time = "2025-08-11T12:06:36.98Z" },
+    { url = "https://files.pythonhosted.org/packages/25/77/62752d3dbd70e27fdd68e86626c1ae6bccfebe2bb1f84ae226363e112f5a/multidict-6.6.4-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:e5b1413361cef15340ab9dc61523e653d25723e82d488ef7d60a12878227ed50", size = 254689, upload-time = "2025-08-11T12:06:38.233Z" },
+    { url = "https://files.pythonhosted.org/packages/00/6e/fac58b1072a6fc59af5e7acb245e8754d3e1f97f4f808a6559951f72a0d4/multidict-6.6.4-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e167bf899c3d724f9662ef00b4f7fef87a19c22b2fead198a6f68b263618df52", size = 246709, upload-time = "2025-08-11T12:06:39.517Z" },
+    { url = "https://files.pythonhosted.org/packages/01/ef/4698d6842ef5e797c6db7744b0081e36fb5de3d00002cc4c58071097fac3/multidict-6.6.4-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:aaea28ba20a9026dfa77f4b80369e51cb767c61e33a2d4043399c67bd95fb7c6", size = 243185, upload-time = "2025-08-11T12:06:40.796Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/c9/d82e95ae1d6e4ef396934e9b0e942dfc428775f9554acf04393cce66b157/multidict-6.6.4-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:8c91cdb30809a96d9ecf442ec9bc45e8cfaa0f7f8bdf534e082c2443a196727e", size = 237838, upload-time = "2025-08-11T12:06:42.595Z" },
+    { url = "https://files.pythonhosted.org/packages/57/cf/f94af5c36baaa75d44fab9f02e2a6bcfa0cd90acb44d4976a80960759dbc/multidict-6.6.4-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:1a0ccbfe93ca114c5d65a2471d52d8829e56d467c97b0e341cf5ee45410033b3", size = 246368, upload-time = "2025-08-11T12:06:44.304Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/fe/29f23460c3d995f6a4b678cb2e9730e7277231b981f0b234702f0177818a/multidict-6.6.4-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:55624b3f321d84c403cb7d8e6e982f41ae233d85f85db54ba6286f7295dc8a9c", size = 253339, upload-time = "2025-08-11T12:06:45.597Z" },
+    { url = "https://files.pythonhosted.org/packages/29/b6/fd59449204426187b82bf8a75f629310f68c6adc9559dc922d5abe34797b/multidict-6.6.4-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:4a1fb393a2c9d202cb766c76208bd7945bc194eba8ac920ce98c6e458f0b524b", size = 246933, upload-time = "2025-08-11T12:06:46.841Z" },
+    { url = "https://files.pythonhosted.org/packages/19/52/d5d6b344f176a5ac3606f7a61fb44dc746e04550e1a13834dff722b8d7d6/multidict-6.6.4-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:43868297a5759a845fa3a483fb4392973a95fb1de891605a3728130c52b8f40f", size = 242225, upload-time = "2025-08-11T12:06:48.588Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/d3/5b2281ed89ff4d5318d82478a2a2450fcdfc3300da48ff15c1778280ad26/multidict-6.6.4-cp311-cp311-win32.whl", hash = "sha256:ed3b94c5e362a8a84d69642dbeac615452e8af9b8eb825b7bc9f31a53a1051e2", size = 41306, upload-time = "2025-08-11T12:06:49.95Z" },
+    { url = "https://files.pythonhosted.org/packages/74/7d/36b045c23a1ab98507aefd44fd8b264ee1dd5e5010543c6fccf82141ccef/multidict-6.6.4-cp311-cp311-win_amd64.whl", hash = "sha256:d8c112f7a90d8ca5d20213aa41eac690bb50a76da153e3afb3886418e61cb22e", size = 46029, upload-time = "2025-08-11T12:06:51.082Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/5e/553d67d24432c5cd52b49047f2d248821843743ee6d29a704594f656d182/multidict-6.6.4-cp311-cp311-win_arm64.whl", hash = "sha256:3bb0eae408fa1996d87247ca0d6a57b7fc1dcf83e8a5c47ab82c558c250d4adf", size = 43017, upload-time = "2025-08-11T12:06:52.243Z" },
+    { url = "https://files.pythonhosted.org/packages/05/f6/512ffd8fd8b37fb2680e5ac35d788f1d71bbaf37789d21a820bdc441e565/multidict-6.6.4-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:0ffb87be160942d56d7b87b0fdf098e81ed565add09eaa1294268c7f3caac4c8", size = 76516, upload-time = "2025-08-11T12:06:53.393Z" },
+    { url = "https://files.pythonhosted.org/packages/99/58/45c3e75deb8855c36bd66cc1658007589662ba584dbf423d01df478dd1c5/multidict-6.6.4-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:d191de6cbab2aff5de6c5723101705fd044b3e4c7cfd587a1929b5028b9714b3", size = 45394, upload-time = "2025-08-11T12:06:54.555Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/ca/e8c4472a93a26e4507c0b8e1f0762c0d8a32de1328ef72fd704ef9cc5447/multidict-6.6.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:38a0956dd92d918ad5feff3db8fcb4a5eb7dba114da917e1a88475619781b57b", size = 43591, upload-time = "2025-08-11T12:06:55.672Z" },
+    { url = "https://files.pythonhosted.org/packages/05/51/edf414f4df058574a7265034d04c935aa84a89e79ce90fcf4df211f47b16/multidict-6.6.4-cp312-cp312-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:6865f6d3b7900ae020b495d599fcf3765653bc927951c1abb959017f81ae8287", size = 237215, upload-time = "2025-08-11T12:06:57.213Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/45/8b3d6dbad8cf3252553cc41abea09ad527b33ce47a5e199072620b296902/multidict-6.6.4-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a2088c126b6f72db6c9212ad827d0ba088c01d951cee25e758c450da732c138", size = 258299, upload-time = "2025-08-11T12:06:58.946Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/e8/8ca2e9a9f5a435fc6db40438a55730a4bf4956b554e487fa1b9ae920f825/multidict-6.6.4-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:0f37bed7319b848097085d7d48116f545985db988e2256b2e6f00563a3416ee6", size = 242357, upload-time = "2025-08-11T12:07:00.301Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/84/80c77c99df05a75c28490b2af8f7cba2a12621186e0a8b0865d8e745c104/multidict-6.6.4-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:01368e3c94032ba6ca0b78e7ccb099643466cf24f8dc8eefcfdc0571d56e58f9", size = 268369, upload-time = "2025-08-11T12:07:01.638Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/e9/920bfa46c27b05fb3e1ad85121fd49f441492dca2449c5bcfe42e4565d8a/multidict-6.6.4-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:8fe323540c255db0bffee79ad7f048c909f2ab0edb87a597e1c17da6a54e493c", size = 269341, upload-time = "2025-08-11T12:07:02.943Z" },
+    { url = "https://files.pythonhosted.org/packages/af/65/753a2d8b05daf496f4a9c367fe844e90a1b2cac78e2be2c844200d10cc4c/multidict-6.6.4-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b8eb3025f17b0a4c3cd08cda49acf312a19ad6e8a4edd9dbd591e6506d999402", size = 256100, upload-time = "2025-08-11T12:07:04.564Z" },
+    { url = "https://files.pythonhosted.org/packages/09/54/655be13ae324212bf0bc15d665a4e34844f34c206f78801be42f7a0a8aaa/multidict-6.6.4-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:bbc14f0365534d35a06970d6a83478b249752e922d662dc24d489af1aa0d1be7", size = 253584, upload-time = "2025-08-11T12:07:05.914Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/74/ab2039ecc05264b5cec73eb018ce417af3ebb384ae9c0e9ed42cb33f8151/multidict-6.6.4-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:75aa52fba2d96bf972e85451b99d8e19cc37ce26fd016f6d4aa60da9ab2b005f", size = 251018, upload-time = "2025-08-11T12:07:08.301Z" },
+    { url = "https://files.pythonhosted.org/packages/af/0a/ccbb244ac848e56c6427f2392741c06302bbfba49c0042f1eb3c5b606497/multidict-6.6.4-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:4fefd4a815e362d4f011919d97d7b4a1e566f1dde83dc4ad8cfb5b41de1df68d", size = 251477, upload-time = "2025-08-11T12:07:10.248Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/b0/0ed49bba775b135937f52fe13922bc64a7eaf0a3ead84a36e8e4e446e096/multidict-6.6.4-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:db9801fe021f59a5b375ab778973127ca0ac52429a26e2fd86aa9508f4d26eb7", size = 263575, upload-time = "2025-08-11T12:07:11.928Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/d9/7fb85a85e14de2e44dfb6a24f03c41e2af8697a6df83daddb0e9b7569f73/multidict-6.6.4-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:a650629970fa21ac1fb06ba25dabfc5b8a2054fcbf6ae97c758aa956b8dba802", size = 259649, upload-time = "2025-08-11T12:07:13.244Z" },
+    { url = "https://files.pythonhosted.org/packages/03/9e/b3a459bcf9b6e74fa461a5222a10ff9b544cb1cd52fd482fb1b75ecda2a2/multidict-6.6.4-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:452ff5da78d4720d7516a3a2abd804957532dd69296cb77319c193e3ffb87e24", size = 251505, upload-time = "2025-08-11T12:07:14.57Z" },
+    { url = "https://files.pythonhosted.org/packages/86/a2/8022f78f041dfe6d71e364001a5cf987c30edfc83c8a5fb7a3f0974cff39/multidict-6.6.4-cp312-cp312-win32.whl", hash = "sha256:8c2fcb12136530ed19572bbba61b407f655e3953ba669b96a35036a11a485793", size = 41888, upload-time = "2025-08-11T12:07:15.904Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/eb/d88b1780d43a56db2cba24289fa744a9d216c1a8546a0dc3956563fd53ea/multidict-6.6.4-cp312-cp312-win_amd64.whl", hash = "sha256:047d9425860a8c9544fed1b9584f0c8bcd31bcde9568b047c5e567a1025ecd6e", size = 46072, upload-time = "2025-08-11T12:07:17.045Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/16/b929320bf5750e2d9d4931835a4c638a19d2494a5b519caaaa7492ebe105/multidict-6.6.4-cp312-cp312-win_arm64.whl", hash = "sha256:14754eb72feaa1e8ae528468f24250dd997b8e2188c3d2f593f9eba259e4b364", size = 43222, upload-time = "2025-08-11T12:07:18.328Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/d3/f04c5db316caee9b5b2cbba66270b358c922a959855995bedde87134287c/multidict-6.6.4-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:af7618b591bae552b40dbb6f93f5518328a949dac626ee75927bba1ecdeea9f4", size = 76977, upload-time = "2025-08-11T12:08:16.667Z" },
+    { url = "https://files.pythonhosted.org/packages/70/39/a6200417d883e510728ab3caec02d3b66ff09e1c85e0aab2ba311abfdf06/multidict-6.6.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:b6819f83aef06f560cb15482d619d0e623ce9bf155115150a85ab11b8342a665", size = 44878, upload-time = "2025-08-11T12:08:18.157Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/7e/815be31ed35571b137d65232816f61513fcd97b2717d6a9d7800b5a0c6e0/multidict-6.6.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:4d09384e75788861e046330308e7af54dd306aaf20eb760eb1d0de26b2bea2cb", size = 44546, upload-time = "2025-08-11T12:08:19.694Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/f1/21b5bff6a8c3e2aff56956c241941ace6b8820e1abe6b12d3c52868a773d/multidict-6.6.4-cp39-cp39-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:a59c63061f1a07b861c004e53869eb1211ffd1a4acbca330e3322efa6dd02978", size = 223020, upload-time = "2025-08-11T12:08:21.554Z" },
+    { url = "https://files.pythonhosted.org/packages/15/59/37083f1dd3439979a0ffeb1906818d978d88b4cc7f4600a9f89b1cb6713c/multidict-6.6.4-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:350f6b0fe1ced61e778037fdc7613f4051c8baf64b1ee19371b42a3acdb016a0", size = 240528, upload-time = "2025-08-11T12:08:23.45Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/f0/f054d123c87784307a27324c829eb55bcfd2e261eb785fcabbd832c8dc4a/multidict-6.6.4-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:0c5cbac6b55ad69cb6aa17ee9343dfbba903118fd530348c330211dc7aa756d1", size = 219540, upload-time = "2025-08-11T12:08:24.965Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/26/8f78ce17b7118149c17f238f28fba2a850b660b860f9b024a34d0191030f/multidict-6.6.4-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:630f70c32b8066ddfd920350bc236225814ad94dfa493fe1910ee17fe4365cbb", size = 251182, upload-time = "2025-08-11T12:08:26.511Z" },
+    { url = "https://files.pythonhosted.org/packages/00/c3/a21466322d69f6594fe22d9379200f99194d21c12a5bbf8c2a39a46b83b6/multidict-6.6.4-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f8d4916a81697faec6cb724a273bd5457e4c6c43d82b29f9dc02c5542fd21fc9", size = 249371, upload-time = "2025-08-11T12:08:28.075Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/8e/2e673124eb05cf8dc82e9265eccde01a36bcbd3193e27799b8377123c976/multidict-6.6.4-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8e42332cf8276bb7645d310cdecca93a16920256a5b01bebf747365f86a1675b", size = 239235, upload-time = "2025-08-11T12:08:29.937Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/2d/bdd9f05e7c89e30a4b0e4faf0681a30748f8d1310f68cfdc0e3571e75bd5/multidict-6.6.4-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:f3be27440f7644ab9a13a6fc86f09cdd90b347c3c5e30c6d6d860de822d7cb53", size = 237410, upload-time = "2025-08-11T12:08:31.872Z" },
+    { url = "https://files.pythonhosted.org/packages/46/4c/3237b83f8ca9a2673bb08fc340c15da005a80f5cc49748b587c8ae83823b/multidict-6.6.4-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:21f216669109e02ef3e2415ede07f4f8987f00de8cdfa0cc0b3440d42534f9f0", size = 232979, upload-time = "2025-08-11T12:08:33.399Z" },
+    { url = "https://files.pythonhosted.org/packages/55/a6/a765decff625ae9bc581aed303cd1837955177dafc558859a69f56f56ba8/multidict-6.6.4-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:d9890d68c45d1aeac5178ded1d1cccf3bc8d7accf1f976f79bf63099fb16e4bd", size = 240979, upload-time = "2025-08-11T12:08:35.02Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/2d/9c75975cb0c66ea33cae1443bb265b2b3cd689bffcbc68872565f401da23/multidict-6.6.4-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:edfdcae97cdc5d1a89477c436b61f472c4d40971774ac4729c613b4b133163cb", size = 246849, upload-time = "2025-08-11T12:08:37.038Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/71/d21ac0843c1d8751fb5dcf8a1f436625d39d4577bc27829799d09b419af7/multidict-6.6.4-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:0b2e886624be5773e69cf32bcb8534aecdeb38943520b240fed3d5596a430f2f", size = 241798, upload-time = "2025-08-11T12:08:38.669Z" },
+    { url = "https://files.pythonhosted.org/packages/94/3d/1d8911e53092837bd11b1c99d71de3e2a9a26f8911f864554677663242aa/multidict-6.6.4-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:be5bf4b3224948032a845d12ab0f69f208293742df96dc14c4ff9b09e508fc17", size = 235315, upload-time = "2025-08-11T12:08:40.266Z" },
+    { url = "https://files.pythonhosted.org/packages/86/c5/4b758df96376f73e936b1942c6c2dfc17e37ed9d5ff3b01a811496966ca0/multidict-6.6.4-cp39-cp39-win32.whl", hash = "sha256:10a68a9191f284fe9d501fef4efe93226e74df92ce7a24e301371293bd4918ae", size = 41434, upload-time = "2025-08-11T12:08:41.965Z" },
+    { url = "https://files.pythonhosted.org/packages/58/16/f1dfa2a0f25f2717a5e9e5fe8fd30613f7fe95e3530cec8d11f5de0b709c/multidict-6.6.4-cp39-cp39-win_amd64.whl", hash = "sha256:ee25f82f53262f9ac93bd7e58e47ea1bdcc3393cef815847e397cba17e284210", size = 46186, upload-time = "2025-08-11T12:08:43.367Z" },
+    { url = "https://files.pythonhosted.org/packages/88/7d/a0568bac65438c494cb6950b29f394d875a796a237536ac724879cf710c9/multidict-6.6.4-cp39-cp39-win_arm64.whl", hash = "sha256:f9867e55590e0855bcec60d4f9a092b69476db64573c9fe17e92b0c50614c16a", size = 43115, upload-time = "2025-08-11T12:08:45.126Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/69/b547032297c7e63ba2af494edba695d781af8a0c6e89e4d06cf848b21d80/multidict-6.6.4-py3-none-any.whl", hash = "sha256:27d8f8e125c07cb954e54d75d04905a9bba8a439c1d84aca94949d4d03d8601c", size = 12313, upload-time = "2025-08-11T12:08:46.891Z" },
 ]
 
 [[package]]
@@ -1599,26 +1970,35 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "dill" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b5/ae/04f39c5d0d0def03247c2893d6f2b83c136bf3320a2154d7b8858f2ba72d/multiprocess-0.70.16.tar.gz", hash = "sha256:161af703d4652a0e1410be6abccecde4a7ddffd19341be0a7011b94aeb171ac1", size = 1772603 }
+sdist = { url = "https://files.pythonhosted.org/packages/b5/ae/04f39c5d0d0def03247c2893d6f2b83c136bf3320a2154d7b8858f2ba72d/multiprocess-0.70.16.tar.gz", hash = "sha256:161af703d4652a0e1410be6abccecde4a7ddffd19341be0a7011b94aeb171ac1", size = 1772603, upload-time = "2024-01-28T18:52:34.85Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ef/76/6e712a2623d146d314f17598df5de7224c85c0060ef63fd95cc15a25b3fa/multiprocess-0.70.16-pp310-pypy310_pp73-macosx_10_13_x86_64.whl", hash = "sha256:476887be10e2f59ff183c006af746cb6f1fd0eadcfd4ef49e605cbe2659920ee", size = 134980, upload-time = "2024-01-28T18:52:15.731Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/ab/1e6e8009e380e22254ff539ebe117861e5bdb3bff1fc977920972237c6c7/multiprocess-0.70.16-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:d951bed82c8f73929ac82c61f01a7b5ce8f3e5ef40f5b52553b4f547ce2b08ec", size = 134982, upload-time = "2024-01-28T18:52:17.783Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/94/8638a89f93c80df329116e6781a060506c7e91e1f4370dc831e9d17a041d/multiprocess-0.70.16-pp39-pypy39_pp73-macosx_10_13_x86_64.whl", hash = "sha256:0dfd078c306e08d46d7a8d06fb120313d87aa43af60d66da43ffff40b44d2f41", size = 133497, upload-time = "2024-01-28T18:52:22.644Z" },
+    { url = "https://files.pythonhosted.org/packages/89/21/222066f6bb8d8af287923ae3bd26cf4699a9ce020228ac273caca1de8250/multiprocess-0.70.16-pp39-pypy39_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:e7b9d0f307cd9bd50851afaac0dba2cb6c44449efff697df7c7645f7d3f2be3a", size = 133498, upload-time = "2024-01-28T18:52:24.576Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/f7/7ec7fddc92e50714ea3745631f79bd9c96424cb2702632521028e57d3a36/multiprocess-0.70.16-py310-none-any.whl", hash = "sha256:c4a9944c67bd49f823687463660a2d6daae94c289adff97e0f9d696ba6371d02", size = 134824, upload-time = "2024-01-28T18:52:26.062Z" },
+    { url = "https://files.pythonhosted.org/packages/50/15/b56e50e8debaf439f44befec5b2af11db85f6e0f344c3113ae0be0593a91/multiprocess-0.70.16-py311-none-any.whl", hash = "sha256:af4cabb0dac72abfb1e794fa7855c325fd2b55a10a44628a3c1ad3311c04127a", size = 143519, upload-time = "2024-01-28T18:52:28.115Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/7d/a988f258104dcd2ccf1ed40fdc97e26c4ac351eeaf81d76e266c52d84e2f/multiprocess-0.70.16-py312-none-any.whl", hash = "sha256:fc0544c531920dde3b00c29863377f87e1632601092ea2daca74e4beb40faa2e", size = 146741, upload-time = "2024-01-28T18:52:29.395Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/89/38df130f2c799090c978b366cfdf5b96d08de5b29a4a293df7f7429fa50b/multiprocess-0.70.16-py38-none-any.whl", hash = "sha256:a71d82033454891091a226dfc319d0cfa8019a4e888ef9ca910372a446de4435", size = 132628, upload-time = "2024-01-28T18:52:30.853Z" },
+    { url = "https://files.pythonhosted.org/packages/da/d9/f7f9379981e39b8c2511c9e0326d212accacb82f12fbfdc1aa2ce2a7b2b6/multiprocess-0.70.16-py39-none-any.whl", hash = "sha256:a0bafd3ae1b732eac64be2e72038231c1ba97724b60b09400d68f229fcc2fbf3", size = 133351, upload-time = "2024-01-28T18:52:31.981Z" },
+]
+
+[[package]]
+name = "namex"
+version = "0.1.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0c/c0/ee95b28f029c73f8d49d8f52edaed02a1d4a9acb8b69355737fdb1faa191/namex-0.1.0.tar.gz", hash = "sha256:117f03ccd302cc48e3f5c58a296838f6b89c83455ab8683a1e85f2a430aa4306", size = 6649, upload-time = "2025-05-26T23:17:38.918Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ef/76/6e712a2623d146d314f17598df5de7224c85c0060ef63fd95cc15a25b3fa/multiprocess-0.70.16-pp310-pypy310_pp73-macosx_10_13_x86_64.whl", hash = "sha256:476887be10e2f59ff183c006af746cb6f1fd0eadcfd4ef49e605cbe2659920ee", size = 134980 },
-    { url = "https://files.pythonhosted.org/packages/0f/ab/1e6e8009e380e22254ff539ebe117861e5bdb3bff1fc977920972237c6c7/multiprocess-0.70.16-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:d951bed82c8f73929ac82c61f01a7b5ce8f3e5ef40f5b52553b4f547ce2b08ec", size = 134982 },
-    { url = "https://files.pythonhosted.org/packages/d8/94/8638a89f93c80df329116e6781a060506c7e91e1f4370dc831e9d17a041d/multiprocess-0.70.16-pp39-pypy39_pp73-macosx_10_13_x86_64.whl", hash = "sha256:0dfd078c306e08d46d7a8d06fb120313d87aa43af60d66da43ffff40b44d2f41", size = 133497 },
-    { url = "https://files.pythonhosted.org/packages/89/21/222066f6bb8d8af287923ae3bd26cf4699a9ce020228ac273caca1de8250/multiprocess-0.70.16-pp39-pypy39_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:e7b9d0f307cd9bd50851afaac0dba2cb6c44449efff697df7c7645f7d3f2be3a", size = 133498 },
-    { url = "https://files.pythonhosted.org/packages/bc/f7/7ec7fddc92e50714ea3745631f79bd9c96424cb2702632521028e57d3a36/multiprocess-0.70.16-py310-none-any.whl", hash = "sha256:c4a9944c67bd49f823687463660a2d6daae94c289adff97e0f9d696ba6371d02", size = 134824 },
-    { url = "https://files.pythonhosted.org/packages/50/15/b56e50e8debaf439f44befec5b2af11db85f6e0f344c3113ae0be0593a91/multiprocess-0.70.16-py311-none-any.whl", hash = "sha256:af4cabb0dac72abfb1e794fa7855c325fd2b55a10a44628a3c1ad3311c04127a", size = 143519 },
-    { url = "https://files.pythonhosted.org/packages/0a/7d/a988f258104dcd2ccf1ed40fdc97e26c4ac351eeaf81d76e266c52d84e2f/multiprocess-0.70.16-py312-none-any.whl", hash = "sha256:fc0544c531920dde3b00c29863377f87e1632601092ea2daca74e4beb40faa2e", size = 146741 },
-    { url = "https://files.pythonhosted.org/packages/ea/89/38df130f2c799090c978b366cfdf5b96d08de5b29a4a293df7f7429fa50b/multiprocess-0.70.16-py38-none-any.whl", hash = "sha256:a71d82033454891091a226dfc319d0cfa8019a4e888ef9ca910372a446de4435", size = 132628 },
-    { url = "https://files.pythonhosted.org/packages/da/d9/f7f9379981e39b8c2511c9e0326d212accacb82f12fbfdc1aa2ce2a7b2b6/multiprocess-0.70.16-py39-none-any.whl", hash = "sha256:a0bafd3ae1b732eac64be2e72038231c1ba97724b60b09400d68f229fcc2fbf3", size = 133351 },
+    { url = "https://files.pythonhosted.org/packages/b2/bc/465daf1de06409cdd4532082806770ee0d8d7df434da79c76564d0f69741/namex-0.1.0-py3-none-any.whl", hash = "sha256:e2012a474502f1e2251267062aae3114611f07df4224b6e06334c57b0f2ce87c", size = 5905, upload-time = "2025-05-26T23:17:37.695Z" },
 ]
 
 [[package]]
 name = "nest-asyncio"
 version = "1.6.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/83/f8/51569ac65d696c8ecbee95938f89d4abf00f47d58d48f6fbabfe8f0baefe/nest_asyncio-1.6.0.tar.gz", hash = "sha256:6f172d5449aca15afd6c646851f4e31e02c598d553a667e38cafa997cfec55fe", size = 7418 }
+sdist = { url = "https://files.pythonhosted.org/packages/83/f8/51569ac65d696c8ecbee95938f89d4abf00f47d58d48f6fbabfe8f0baefe/nest_asyncio-1.6.0.tar.gz", hash = "sha256:6f172d5449aca15afd6c646851f4e31e02c598d553a667e38cafa997cfec55fe", size = 7418, upload-time = "2024-01-21T14:25:19.227Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a0/c4/c2971a3ba4c6103a3d10c4b0f24f461ddc027f0f09763220cf35ca1401b3/nest_asyncio-1.6.0-py3-none-any.whl", hash = "sha256:87af6efd6b5e897c81050477ef65c62e2b2f35d51703cae01aff2905b1852e1c", size = 5195 },
+    { url = "https://files.pythonhosted.org/packages/a0/c4/c2971a3ba4c6103a3d10c4b0f24f461ddc027f0f09763220cf35ca1401b3/nest_asyncio-1.6.0-py3-none-any.whl", hash = "sha256:87af6efd6b5e897c81050477ef65c62e2b2f35d51703cae01aff2905b1852e1c", size = 5195, upload-time = "2024-01-21T14:25:17.223Z" },
 ]
 
 [[package]]
@@ -1626,12 +2006,11 @@ name = "networkx"
 version = "3.2.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
-    "python_full_version < '3.9.17'",
+    "python_full_version < '3.10'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/c4/80/a84676339aaae2f1cfdf9f418701dd634aef9cc76f708ef55c36ff39c3ca/networkx-3.2.1.tar.gz", hash = "sha256:9f1bb5cf3409bf324e0a722c20bdb4c20ee39bf1c30ce8ae499c8502b0b5e0c6", size = 2073928 }
+sdist = { url = "https://files.pythonhosted.org/packages/c4/80/a84676339aaae2f1cfdf9f418701dd634aef9cc76f708ef55c36ff39c3ca/networkx-3.2.1.tar.gz", hash = "sha256:9f1bb5cf3409bf324e0a722c20bdb4c20ee39bf1c30ce8ae499c8502b0b5e0c6", size = 2073928, upload-time = "2023-10-28T08:41:39.364Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d5/f0/8fbc882ca80cf077f1b246c0e3c3465f7f415439bdea6b899f6b19f61f70/networkx-3.2.1-py3-none-any.whl", hash = "sha256:f18c69adc97877c42332c170849c96cefa91881c99a7cb3e95b7c659ebdc1ec2", size = 1647772 },
+    { url = "https://files.pythonhosted.org/packages/d5/f0/8fbc882ca80cf077f1b246c0e3c3465f7f415439bdea6b899f6b19f61f70/networkx-3.2.1-py3-none-any.whl", hash = "sha256:f18c69adc97877c42332c170849c96cefa91881c99a7cb3e95b7c659ebdc1ec2", size = 1647772, upload-time = "2023-10-28T08:41:36.945Z" },
 ]
 
 [[package]]
@@ -1639,154 +2018,322 @@ name = "networkx"
 version = "3.4.2"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.12.*'",
-    "python_full_version == '3.11.*'",
     "python_full_version == '3.10.*'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fd/1d/06475e1cd5264c0b870ea2cc6fdb3e37177c1e565c43f56ff17a10e3937f/networkx-3.4.2.tar.gz", hash = "sha256:307c3669428c5362aab27c8a1260aa8f47c4e91d3891f48be0141738d8d053e1", size = 2151368 }
+sdist = { url = "https://files.pythonhosted.org/packages/fd/1d/06475e1cd5264c0b870ea2cc6fdb3e37177c1e565c43f56ff17a10e3937f/networkx-3.4.2.tar.gz", hash = "sha256:307c3669428c5362aab27c8a1260aa8f47c4e91d3891f48be0141738d8d053e1", size = 2151368, upload-time = "2024-10-21T12:39:38.695Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b9/54/dd730b32ea14ea797530a4479b2ed46a6fb250f682a9cfb997e968bf0261/networkx-3.4.2-py3-none-any.whl", hash = "sha256:df5d4365b724cf81b8c6a7312509d0c22386097011ad1abe274afd5e9d3bbc5f", size = 1723263 },
+    { url = "https://files.pythonhosted.org/packages/b9/54/dd730b32ea14ea797530a4479b2ed46a6fb250f682a9cfb997e968bf0261/networkx-3.4.2-py3-none-any.whl", hash = "sha256:df5d4365b724cf81b8c6a7312509d0c22386097011ad1abe274afd5e9d3bbc5f", size = 1723263, upload-time = "2024-10-21T12:39:36.247Z" },
 ]
 
 [[package]]
-name = "ninja"
-version = "1.11.1.3"
+name = "networkx"
+version = "3.5"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6c/4f/ccdb8ad3a38e583f214547fd2f7ff1fc160c43a75af88e6aec213404b96a/networkx-3.5.tar.gz", hash = "sha256:d4c6f9cf81f52d69230866796b82afbccdec3db7ae4fbd1b65ea750feed50037", size = 2471065, upload-time = "2025-05-29T11:35:07.804Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/eb/8d/776adee7bbf76365fdd7f2552710282c79a4ead5d2a46408c9043a2b70ba/networkx-3.5-py3-none-any.whl", hash = "sha256:0030d386a9a06dee3565298b4a734b68589749a544acbb6c412dc9e2489ec6ec", size = 2034406, upload-time = "2025-05-29T11:35:04.961Z" },
+]
+
+[[package]]
+name = "nh3"
+version = "0.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/bd/8f/21a2701f95b7d0d5137736561b3427ece0c4a1e085d4a223b92d16ab7d8b/ninja-1.11.1.3.tar.gz", hash = "sha256:edfa0d2e9d7ead1635b03e40a32ad56cc8f56798b6e2e9848d8300b174897076", size = 129532 }
+sdist = { url = "https://files.pythonhosted.org/packages/c3/a4/96cff0977357f60f06ec4368c4c7a7a26cccfe7c9fcd54f5378bf0428fd3/nh3-0.3.0.tar.gz", hash = "sha256:d8ba24cb31525492ea71b6aac11a4adac91d828aadeff7c4586541bf5dc34d2f", size = 19655, upload-time = "2025-07-17T14:43:37.05Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ea/ba/0069cd4a83d68f7b0308be70e219b15d675e50c8ea28763a3f0373c45bfc/ninja-1.11.1.3-py3-none-macosx_10_9_universal2.whl", hash = "sha256:2b4879ea3f1169f3d855182c57dcc84d1b5048628c8b7be0d702b81882a37237", size = 279132 },
-    { url = "https://files.pythonhosted.org/packages/72/6b/3805be87df8417a0c7b21078c8045f2a1e59b34f371bfe4cb4fb0d6df7f2/ninja-1.11.1.3-py3-none-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:bc3ebc8b2e47716149f3541742b5cd8e0b08f51013b825c05baca3e34854370d", size = 472101 },
-    { url = "https://files.pythonhosted.org/packages/6b/35/a8e38d54768e67324e365e2a41162be298f51ec93e6bd4b18d237d7250d8/ninja-1.11.1.3-py3-none-manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:a27e78ca71316c8654965ee94b286a98c83877bfebe2607db96897bbfe458af0", size = 422884 },
-    { url = "https://files.pythonhosted.org/packages/2f/99/7996457319e139c02697fb2aa28e42fe32bb0752cef492edc69d56a3552e/ninja-1.11.1.3-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2883ea46b3c5079074f56820f9989c6261fcc6fd873d914ee49010ecf283c3b2", size = 157046 },
-    { url = "https://files.pythonhosted.org/packages/6d/8b/93f38e5cddf76ccfdab70946515b554f25d2b4c95ef9b2f9cfbc43fa7cc1/ninja-1.11.1.3-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8c4bdb9fd2d0c06501ae15abfd23407660e95659e384acd36e013b6dd7d8a8e4", size = 180014 },
-    { url = "https://files.pythonhosted.org/packages/7d/1d/713884d0fa3c972164f69d552e0701d30e2bf25eba9ef160bfb3dc69926a/ninja-1.11.1.3-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:114ed5c61c8474df6a69ab89097a20749b769e2c219a452cb2fadc49b0d581b0", size = 157098 },
-    { url = "https://files.pythonhosted.org/packages/c7/22/ecb0f70e77c9e22ee250aa717a608a142756833a34d43943d7d658ee0e56/ninja-1.11.1.3-py3-none-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:7fa2247fce98f683bc712562d82b22b8a0a5c000738a13147ca2d1b68c122298", size = 130089 },
-    { url = "https://files.pythonhosted.org/packages/ec/a6/3ee846c20ab6ad95b90c5c8703c76cb1f39cc8ce2d1ae468956e3b1b2581/ninja-1.11.1.3-py3-none-musllinux_1_1_aarch64.whl", hash = "sha256:a38c6c6c8032bed68b70c3b065d944c35e9f903342875d3a3218c1607987077c", size = 372508 },
-    { url = "https://files.pythonhosted.org/packages/95/0d/aa44abe4141f29148ce671ac8c92045878906b18691c6f87a29711c2ff1c/ninja-1.11.1.3-py3-none-musllinux_1_1_i686.whl", hash = "sha256:56ada5d33b8741d298836644042faddebc83ee669782d661e21563034beb5aba", size = 419369 },
-    { url = "https://files.pythonhosted.org/packages/f7/ec/48bf5105568ac9bd2016b701777bdd5000cc09a14ac837fef9f15e8d634e/ninja-1.11.1.3-py3-none-musllinux_1_1_ppc64le.whl", hash = "sha256:53409151da081f3c198bb0bfc220a7f4e821e022c5b7d29719adda892ddb31bb", size = 420304 },
-    { url = "https://files.pythonhosted.org/packages/18/e5/69df63976cf971a03379899f8520a036c9dbab26330b37197512aed5b3df/ninja-1.11.1.3-py3-none-musllinux_1_1_s390x.whl", hash = "sha256:1ad2112c2b0159ed7c4ae3731595191b1546ba62316fc40808edecd0306fefa3", size = 416056 },
-    { url = "https://files.pythonhosted.org/packages/6f/4f/bdb401af7ed0e24a3fef058e13a149f2de1ce4b176699076993615d55610/ninja-1.11.1.3-py3-none-musllinux_1_1_x86_64.whl", hash = "sha256:28aea3c1c280cba95b8608d50797169f3a34280e3e9a6379b6e340f0c9eaeeb0", size = 379725 },
-    { url = "https://files.pythonhosted.org/packages/bd/68/05e7863bf13128c61652eeb3ec7096c3d3a602f32f31752dbfb034e3fa07/ninja-1.11.1.3-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:b6966f83064a88a51693073eea3decd47e08c3965241e09578ef7aa3a7738329", size = 434881 },
-    { url = "https://files.pythonhosted.org/packages/bd/ad/edc0d1efe77f29f45bbca2e1dab07ef597f61a88de6e4bccffc0aec2256c/ninja-1.11.1.3-py3-none-win32.whl", hash = "sha256:a4a3b71490557e18c010cbb26bd1ea9a0c32ee67e8f105e9731515b6e0af792e", size = 255988 },
-    { url = "https://files.pythonhosted.org/packages/03/93/09a9f7672b4f97438aca6217ac54212a63273f1cd3b46b731d0bb22c53e7/ninja-1.11.1.3-py3-none-win_amd64.whl", hash = "sha256:04d48d14ea7ba11951c156599ab526bdda575450797ff57c6fdf99b2554d09c7", size = 296502 },
-    { url = "https://files.pythonhosted.org/packages/d9/9d/0cc1e82849070ff3cbee69f326cb48a839407bcd15d8844443c30a5e7509/ninja-1.11.1.3-py3-none-win_arm64.whl", hash = "sha256:17978ad611d8ead578d83637f5ae80c2261b033db0b493a7ce94f88623f29e1b", size = 270571 },
+    { url = "https://files.pythonhosted.org/packages/5c/86/a96b1453c107b815f9ab8fac5412407c33cc5c7580a4daf57aabeb41b774/nh3-0.3.0-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ce5e7185599f89b0e391e2f29cc12dc2e206167380cea49b33beda4891be2fe1", size = 823791, upload-time = "2025-07-17T14:43:19.721Z" },
+    { url = "https://files.pythonhosted.org/packages/97/33/11e7273b663839626f714cb68f6eb49899da5a0d9b6bc47b41fe870259c2/nh3-0.3.0-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:389d93d59b8214d51c400fb5b07866c2a4f79e4e14b071ad66c92184fec3a392", size = 811143, upload-time = "2025-07-17T14:43:20.779Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/1b/b15bd1ce201a1a610aeb44afd478d55ac018b4475920a3118ffd806e2483/nh3-0.3.0-cp38-abi3-manylinux_2_17_ppc64.manylinux2014_ppc64.whl", hash = "sha256:e9e6a7e4d38f7e8dda9edd1433af5170c597336c1a74b4693c5cb75ab2b30f2a", size = 1064661, upload-time = "2025-07-17T14:43:21.839Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/14/079670fb2e848c4ba2476c5a7a2d1319826053f4f0368f61fca9bb4227ae/nh3-0.3.0-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7852f038a054e0096dac12b8141191e02e93e0b4608c4b993ec7d4ffafea4e49", size = 997061, upload-time = "2025-07-17T14:43:23.179Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/e5/ac7fc565f5d8bce7f979d1afd68e8cb415020d62fa6507133281c7d49f91/nh3-0.3.0-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:af5aa8127f62bbf03d68f67a956627b1bd0469703a35b3dad28d0c1195e6c7fb", size = 924761, upload-time = "2025-07-17T14:43:24.23Z" },
+    { url = "https://files.pythonhosted.org/packages/39/2c/6394301428b2017a9d5644af25f487fa557d06bc8a491769accec7524d9a/nh3-0.3.0-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f416c35efee3e6a6c9ab7716d9e57aa0a49981be915963a82697952cba1353e1", size = 803959, upload-time = "2025-07-17T14:43:26.377Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/9a/344b9f9c4bd1c2413a397f38ee6a3d5db30f1a507d4976e046226f12b297/nh3-0.3.0-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:37d3003d98dedca6cd762bf88f2e70b67f05100f6b949ffe540e189cc06887f9", size = 844073, upload-time = "2025-07-17T14:43:27.375Z" },
+    { url = "https://files.pythonhosted.org/packages/66/3f/cd37f76c8ca277b02a84aa20d7bd60fbac85b4e2cbdae77cb759b22de58b/nh3-0.3.0-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:634e34e6162e0408e14fb61d5e69dbaea32f59e847cfcfa41b66100a6b796f62", size = 1000680, upload-time = "2025-07-17T14:43:28.452Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/db/7aa11b44bae4e7474feb1201d8dee04fabe5651c7cb51409ebda94a4ed67/nh3-0.3.0-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:b0612ccf5de8a480cf08f047b08f9d3fecc12e63d2ee91769cb19d7290614c23", size = 1076613, upload-time = "2025-07-17T14:43:30.031Z" },
+    { url = "https://files.pythonhosted.org/packages/97/03/03f79f7e5178eb1ad5083af84faff471e866801beb980cc72943a4397368/nh3-0.3.0-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:c7a32a7f0d89f7d30cb8f4a84bdbd56d1eb88b78a2434534f62c71dac538c450", size = 1001418, upload-time = "2025-07-17T14:43:31.429Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/55/1974bcc16884a397ee699cebd3914e1f59be64ab305533347ca2d983756f/nh3-0.3.0-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:3f1b4f8a264a0c86ea01da0d0c390fe295ea0bcacc52c2103aca286f6884f518", size = 986499, upload-time = "2025-07-17T14:43:32.459Z" },
+]
+
+[[package]]
+name = "ninja"
+version = "1.13.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/43/73/79a0b22fc731989c708068427579e840a6cf4e937fe7ae5c5d0b7356ac22/ninja-1.13.0.tar.gz", hash = "sha256:4a40ce995ded54d9dc24f8ea37ff3bf62ad192b547f6c7126e7e25045e76f978", size = 242558, upload-time = "2025-08-11T15:10:19.421Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3c/74/d02409ed2aa865e051b7edda22ad416a39d81a84980f544f8de717cab133/ninja-1.13.0-py3-none-macosx_10_9_universal2.whl", hash = "sha256:fa2a8bfc62e31b08f83127d1613d10821775a0eb334197154c4d6067b7068ff1", size = 310125, upload-time = "2025-08-11T15:09:50.971Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/de/6e1cd6b84b412ac1ef327b76f0641aeb5dcc01e9d3f9eee0286d0c34fd93/ninja-1.13.0-py3-none-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:3d00c692fb717fd511abeb44b8c5d00340c36938c12d6538ba989fe764e79630", size = 177467, upload-time = "2025-08-11T15:09:52.767Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/83/49320fb6e58ae3c079381e333575fdbcf1cca3506ee160a2dcce775046fa/ninja-1.13.0-py3-none-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:be7f478ff9f96a128b599a964fc60a6a87b9fa332ee1bd44fa243ac88d50291c", size = 187834, upload-time = "2025-08-11T15:09:54.115Z" },
+    { url = "https://files.pythonhosted.org/packages/56/c7/ba22748fb59f7f896b609cd3e568d28a0a367a6d953c24c461fe04fc4433/ninja-1.13.0-py3-none-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:60056592cf495e9a6a4bea3cd178903056ecb0943e4de45a2ea825edb6dc8d3e", size = 202736, upload-time = "2025-08-11T15:09:55.745Z" },
+    { url = "https://files.pythonhosted.org/packages/79/22/d1de07632b78ac8e6b785f41fa9aad7a978ec8c0a1bf15772def36d77aac/ninja-1.13.0-py3-none-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:1c97223cdda0417f414bf864cfb73b72d8777e57ebb279c5f6de368de0062988", size = 179034, upload-time = "2025-08-11T15:09:57.394Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/de/0e6edf44d6a04dabd0318a519125ed0415ce437ad5a1ec9b9be03d9048cf/ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fb46acf6b93b8dd0322adc3a4945452a4e774b75b91293bafcc7b7f8e6517dfa", size = 180716, upload-time = "2025-08-11T15:09:58.696Z" },
+    { url = "https://files.pythonhosted.org/packages/54/28/938b562f9057aaa4d6bfbeaa05e81899a47aebb3ba6751e36c027a7f5ff7/ninja-1.13.0-py3-none-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:4be9c1b082d244b1ad7ef41eb8ab088aae8c109a9f3f0b3e56a252d3e00f42c1", size = 146843, upload-time = "2025-08-11T15:10:00.046Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/fb/d06a3838de4f8ab866e44ee52a797b5491df823901c54943b2adb0389fbb/ninja-1.13.0-py3-none-manylinux_2_31_riscv64.whl", hash = "sha256:6739d3352073341ad284246f81339a384eec091d9851a886dfa5b00a6d48b3e2", size = 154402, upload-time = "2025-08-11T15:10:01.657Z" },
+    { url = "https://files.pythonhosted.org/packages/31/bf/0d7808af695ceddc763cf251b84a9892cd7f51622dc8b4c89d5012779f06/ninja-1.13.0-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:11be2d22027bde06f14c343f01d31446747dbb51e72d00decca2eb99be911e2f", size = 552388, upload-time = "2025-08-11T15:10:03.349Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/70/c99d0c2c809f992752453cce312848abb3b1607e56d4cd1b6cded317351a/ninja-1.13.0-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:aa45b4037b313c2f698bc13306239b8b93b4680eb47e287773156ac9e9304714", size = 472501, upload-time = "2025-08-11T15:10:04.735Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/43/c217b1153f0e499652f5e0766da8523ce3480f0a951039c7af115e224d55/ninja-1.13.0-py3-none-musllinux_1_2_i686.whl", hash = "sha256:5f8e1e8a1a30835eeb51db05cf5a67151ad37542f5a4af2a438e9490915e5b72", size = 638280, upload-time = "2025-08-11T15:10:06.512Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/45/9151bba2c8d0ae2b6260f71696330590de5850e5574b7b5694dce6023e20/ninja-1.13.0-py3-none-musllinux_1_2_ppc64le.whl", hash = "sha256:3d7d7779d12cb20c6d054c61b702139fd23a7a964ec8f2c823f1ab1b084150db", size = 642420, upload-time = "2025-08-11T15:10:08.35Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/fb/95752eb635bb8ad27d101d71bef15bc63049de23f299e312878fc21cb2da/ninja-1.13.0-py3-none-musllinux_1_2_riscv64.whl", hash = "sha256:d741a5e6754e0bda767e3274a0f0deeef4807f1fec6c0d7921a0244018926ae5", size = 585106, upload-time = "2025-08-11T15:10:09.818Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/31/aa56a1a286703800c0cbe39fb4e82811c277772dc8cd084f442dd8e2938a/ninja-1.13.0-py3-none-musllinux_1_2_s390x.whl", hash = "sha256:e8bad11f8a00b64137e9b315b137d8bb6cbf3086fbdc43bf1f90fd33324d2e96", size = 707138, upload-time = "2025-08-11T15:10:11.366Z" },
+    { url = "https://files.pythonhosted.org/packages/34/6f/5f5a54a1041af945130abdb2b8529cbef0cdcbbf9bcf3f4195378319d29a/ninja-1.13.0-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:b4f2a072db3c0f944c32793e91532d8948d20d9ab83da9c0c7c15b5768072200", size = 581758, upload-time = "2025-08-11T15:10:13.295Z" },
+    { url = "https://files.pythonhosted.org/packages/95/97/51359c77527d45943fe7a94d00a3843b81162e6c4244b3579fe8fc54cb9c/ninja-1.13.0-py3-none-win32.whl", hash = "sha256:8cfbb80b4a53456ae8a39f90ae3d7a2129f45ea164f43fadfa15dc38c4aef1c9", size = 267201, upload-time = "2025-08-11T15:10:15.158Z" },
+    { url = "https://files.pythonhosted.org/packages/29/45/c0adfbfb0b5895aa18cec400c535b4f7ff3e52536e0403602fc1a23f7de9/ninja-1.13.0-py3-none-win_amd64.whl", hash = "sha256:fb8ee8719f8af47fed145cced4a85f0755dd55d45b2bddaf7431fa89803c5f3e", size = 309975, upload-time = "2025-08-11T15:10:16.697Z" },
+    { url = "https://files.pythonhosted.org/packages/df/93/a7b983643d1253bb223234b5b226e69de6cda02b76cdca7770f684b795f5/ninja-1.13.0-py3-none-win_arm64.whl", hash = "sha256:3c0b40b1f0bba764644385319028650087b4c1b18cdfa6f45cb39a3669b81aa9", size = 290806, upload-time = "2025-08-11T15:10:18.018Z" },
 ]
 
 [[package]]
 name = "nodeenv"
 version = "1.9.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/43/16/fc88b08840de0e0a72a2f9d8c6bae36be573e475a6326ae854bcc549fc45/nodeenv-1.9.1.tar.gz", hash = "sha256:6ec12890a2dab7946721edbfbcd91f3319c6ccc9aec47be7c7e6b7011ee6645f", size = 47437 }
+sdist = { url = "https://files.pythonhosted.org/packages/43/16/fc88b08840de0e0a72a2f9d8c6bae36be573e475a6326ae854bcc549fc45/nodeenv-1.9.1.tar.gz", hash = "sha256:6ec12890a2dab7946721edbfbcd91f3319c6ccc9aec47be7c7e6b7011ee6645f", size = 47437, upload-time = "2024-06-04T18:44:11.171Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d2/1d/1b658dbd2b9fa9c4c9f32accbfc0205d532c8c6194dc0f2a4c0428e7128a/nodeenv-1.9.1-py2.py3-none-any.whl", hash = "sha256:ba11c9782d29c27c70ffbdda2d7415098754709be8a7056d79a737cd901155c9", size = 22314, upload-time = "2024-06-04T18:44:08.352Z" },
+]
+
+[[package]]
+name = "numba"
+version = "0.60.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "llvmlite", version = "0.43.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/3c/93/2849300a9184775ba274aba6f82f303343669b0592b7bb0849ea713dabb0/numba-0.60.0.tar.gz", hash = "sha256:5df6158e5584eece5fc83294b949fd30b9f1125df7708862205217e068aabf16", size = 2702171, upload-time = "2024-06-13T18:11:19.869Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f7/cf/baa13a7e3556d73d9e38021e6d6aa4aeb30d8b94545aa8b70d0f24a1ccc4/numba-0.60.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:5d761de835cd38fb400d2c26bb103a2726f548dc30368853121d66201672e651", size = 2647627, upload-time = "2024-06-13T18:10:29.857Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/ba/4b57fa498564457c3cc9fc9e570a6b08e6086c74220f24baaf04e54b995f/numba-0.60.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:159e618ef213fba758837f9837fb402bbe65326e60ba0633dbe6c7f274d42c1b", size = 2650322, upload-time = "2024-06-13T18:10:32.849Z" },
+    { url = "https://files.pythonhosted.org/packages/28/98/7ea97ee75870a54f938a8c70f7e0be4495ba5349c5f9db09d467c4a5d5b7/numba-0.60.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:1527dc578b95c7c4ff248792ec33d097ba6bef9eda466c948b68dfc995c25781", size = 3407390, upload-time = "2024-06-13T18:10:34.741Z" },
+    { url = "https://files.pythonhosted.org/packages/79/58/cb4ac5b8f7ec64200460aef1fed88258fb872ceef504ab1f989d2ff0f684/numba-0.60.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fe0b28abb8d70f8160798f4de9d486143200f34458d34c4a214114e445d7124e", size = 3699694, upload-time = "2024-06-13T18:10:37.295Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/b0/c61a93ca947d12233ff45de506ddbf52af3f752066a0b8be4d27426e16da/numba-0.60.0-cp310-cp310-win_amd64.whl", hash = "sha256:19407ced081d7e2e4b8d8c36aa57b7452e0283871c296e12d798852bc7d7f198", size = 2687030, upload-time = "2024-06-13T18:10:39.47Z" },
+    { url = "https://files.pythonhosted.org/packages/98/ad/df18d492a8f00d29a30db307904b9b296e37507034eedb523876f3a2e13e/numba-0.60.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:a17b70fc9e380ee29c42717e8cc0bfaa5556c416d94f9aa96ba13acb41bdece8", size = 2647254, upload-time = "2024-06-13T18:10:41.69Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/51/a4dc2c01ce7a850b8e56ff6d5381d047a5daea83d12bad08aa071d34b2ee/numba-0.60.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3fb02b344a2a80efa6f677aa5c40cd5dd452e1b35f8d1c2af0dfd9ada9978e4b", size = 2649970, upload-time = "2024-06-13T18:10:44.682Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/4c/8889ac94c0b33dca80bed11564b8c6d9ea14d7f094e674c58e5c5b05859b/numba-0.60.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5f4fde652ea604ea3c86508a3fb31556a6157b2c76c8b51b1d45eb40c8598703", size = 3412492, upload-time = "2024-06-13T18:10:47.1Z" },
+    { url = "https://files.pythonhosted.org/packages/57/03/2b4245b05b71c0cee667e6a0b51606dfa7f4157c9093d71c6b208385a611/numba-0.60.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4142d7ac0210cc86432b818338a2bc368dc773a2f5cf1e32ff7c5b378bd63ee8", size = 3705018, upload-time = "2024-06-13T18:10:49.539Z" },
+    { url = "https://files.pythonhosted.org/packages/79/89/2d924ca60dbf949f18a6fec223a2445f5f428d9a5f97a6b29c2122319015/numba-0.60.0-cp311-cp311-win_amd64.whl", hash = "sha256:cac02c041e9b5bc8cf8f2034ff6f0dbafccd1ae9590dc146b3a02a45e53af4e2", size = 2686920, upload-time = "2024-06-13T18:10:51.937Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/5c/b5ec752c475e78a6c3676b67c514220dbde2725896bbb0b6ec6ea54b2738/numba-0.60.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:d7da4098db31182fc5ffe4bc42c6f24cd7d1cb8a14b59fd755bfee32e34b8404", size = 2647866, upload-time = "2024-06-13T18:10:54.453Z" },
+    { url = "https://files.pythonhosted.org/packages/65/42/39559664b2e7c15689a638c2a38b3b74c6e69a04e2b3019b9f7742479188/numba-0.60.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:38d6ea4c1f56417076ecf8fc327c831ae793282e0ff51080c5094cb726507b1c", size = 2650208, upload-time = "2024-06-13T18:10:56.779Z" },
+    { url = "https://files.pythonhosted.org/packages/67/88/c4459ccc05674ef02119abf2888ccd3e2fed12a323f52255f4982fc95876/numba-0.60.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:62908d29fb6a3229c242e981ca27e32a6e606cc253fc9e8faeb0e48760de241e", size = 3466946, upload-time = "2024-06-13T18:10:58.961Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/41/ac11cf33524def12aa5bd698226ae196a1185831c05ed29dc0c56eaa308b/numba-0.60.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:0ebaa91538e996f708f1ab30ef4d3ddc344b64b5227b67a57aa74f401bb68b9d", size = 3761463, upload-time = "2024-06-13T18:11:01.657Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/bd/0fe29fcd1b6a8de479a4ed25c6e56470e467e3611c079d55869ceef2b6d1/numba-0.60.0-cp312-cp312-win_amd64.whl", hash = "sha256:f75262e8fe7fa96db1dca93d53a194a38c46da28b112b8a4aca168f0df860347", size = 2707588, upload-time = "2024-06-13T18:11:04.261Z" },
+    { url = "https://files.pythonhosted.org/packages/68/1a/87c53f836cdf557083248c3f47212271f220280ff766538795e77c8c6bbf/numba-0.60.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:01ef4cd7d83abe087d644eaa3d95831b777aa21d441a23703d649e06b8e06b74", size = 2647186, upload-time = "2024-06-13T18:11:06.753Z" },
+    { url = "https://files.pythonhosted.org/packages/28/14/a5baa1f2edea7b49afa4dc1bb1b126645198cf1075186853b5b497be826e/numba-0.60.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:819a3dfd4630d95fd574036f99e47212a1af41cbcb019bf8afac63ff56834449", size = 2650038, upload-time = "2024-06-13T18:11:10.869Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/bd/f1985719ff34e37e07bb18f9d3acd17e5a21da255f550c8eae031e2ddf5f/numba-0.60.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:0b983bd6ad82fe868493012487f34eae8bf7dd94654951404114f23c3466d34b", size = 3403010, upload-time = "2024-06-13T18:11:13.057Z" },
+    { url = "https://files.pythonhosted.org/packages/54/9b/cd73d3f6617ddc8398a63ef97d8dc9139a9879b9ca8a7ca4b8789056ea46/numba-0.60.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:c151748cd269ddeab66334bd754817ffc0cabd9433acb0f551697e5151917d25", size = 3695086, upload-time = "2024-06-13T18:11:15.497Z" },
+    { url = "https://files.pythonhosted.org/packages/01/01/8b7b670c77c5ea0e47e283d82332969bf672ab6410d0b2610cac5b7a3ded/numba-0.60.0-cp39-cp39-win_amd64.whl", hash = "sha256:3031547a015710140e8c87226b4cfe927cac199835e5bf7d4fe5cb64e814e3ab", size = 2686978, upload-time = "2024-06-13T18:11:17.765Z" },
+]
+
+[[package]]
+name = "numba"
+version = "0.61.2"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "llvmlite", version = "0.44.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/1c/a0/e21f57604304aa03ebb8e098429222722ad99176a4f979d34af1d1ee80da/numba-0.61.2.tar.gz", hash = "sha256:8750ee147940a6637b80ecf7f95062185ad8726c8c28a2295b8ec1160a196f7d", size = 2820615, upload-time = "2025-04-09T02:58:07.659Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d2/1d/1b658dbd2b9fa9c4c9f32accbfc0205d532c8c6194dc0f2a4c0428e7128a/nodeenv-1.9.1-py2.py3-none-any.whl", hash = "sha256:ba11c9782d29c27c70ffbdda2d7415098754709be8a7056d79a737cd901155c9", size = 22314 },
+    { url = "https://files.pythonhosted.org/packages/eb/ca/f470be59552ccbf9531d2d383b67ae0b9b524d435fb4a0d229fef135116e/numba-0.61.2-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:cf9f9fc00d6eca0c23fc840817ce9f439b9f03c8f03d6246c0e7f0cb15b7162a", size = 2775663, upload-time = "2025-04-09T02:57:34.143Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/13/3bdf52609c80d460a3b4acfb9fdb3817e392875c0d6270cf3fd9546f138b/numba-0.61.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:ea0247617edcb5dd61f6106a56255baab031acc4257bddaeddb3a1003b4ca3fd", size = 2778344, upload-time = "2025-04-09T02:57:36.609Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/7d/bfb2805bcfbd479f04f835241ecf28519f6e3609912e3a985aed45e21370/numba-0.61.2-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ae8c7a522c26215d5f62ebec436e3d341f7f590079245a2f1008dfd498cc1642", size = 3824054, upload-time = "2025-04-09T02:57:38.162Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/27/797b2004745c92955470c73c82f0e300cf033c791f45bdecb4b33b12bdea/numba-0.61.2-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:bd1e74609855aa43661edffca37346e4e8462f6903889917e9f41db40907daa2", size = 3518531, upload-time = "2025-04-09T02:57:39.709Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/c6/c2fb11e50482cb310afae87a997707f6c7d8a48967b9696271347441f650/numba-0.61.2-cp310-cp310-win_amd64.whl", hash = "sha256:ae45830b129c6137294093b269ef0a22998ccc27bf7cf096ab8dcf7bca8946f9", size = 2831612, upload-time = "2025-04-09T02:57:41.559Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/97/c99d1056aed767503c228f7099dc11c402906b42a4757fec2819329abb98/numba-0.61.2-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:efd3db391df53aaa5cfbee189b6c910a5b471488749fd6606c3f33fc984c2ae2", size = 2775825, upload-time = "2025-04-09T02:57:43.442Z" },
+    { url = "https://files.pythonhosted.org/packages/95/9e/63c549f37136e892f006260c3e2613d09d5120672378191f2dc387ba65a2/numba-0.61.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:49c980e4171948ffebf6b9a2520ea81feed113c1f4890747ba7f59e74be84b1b", size = 2778695, upload-time = "2025-04-09T02:57:44.968Z" },
+    { url = "https://files.pythonhosted.org/packages/97/c8/8740616c8436c86c1b9a62e72cb891177d2c34c2d24ddcde4c390371bf4c/numba-0.61.2-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3945615cd73c2c7eba2a85ccc9c1730c21cd3958bfcf5a44302abae0fb07bb60", size = 3829227, upload-time = "2025-04-09T02:57:46.63Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/06/66e99ae06507c31d15ff3ecd1f108f2f59e18b6e08662cd5f8a5853fbd18/numba-0.61.2-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:bbfdf4eca202cebade0b7d43896978e146f39398909a42941c9303f82f403a18", size = 3523422, upload-time = "2025-04-09T02:57:48.222Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/a4/2b309a6a9f6d4d8cfba583401c7c2f9ff887adb5d54d8e2e130274c0973f/numba-0.61.2-cp311-cp311-win_amd64.whl", hash = "sha256:76bcec9f46259cedf888041b9886e257ae101c6268261b19fda8cfbc52bec9d1", size = 2831505, upload-time = "2025-04-09T02:57:50.108Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/a0/c6b7b9c615cfa3b98c4c63f4316e3f6b3bbe2387740277006551784218cd/numba-0.61.2-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:34fba9406078bac7ab052efbf0d13939426c753ad72946baaa5bf9ae0ebb8dd2", size = 2776626, upload-time = "2025-04-09T02:57:51.857Z" },
+    { url = "https://files.pythonhosted.org/packages/92/4a/fe4e3c2ecad72d88f5f8cd04e7f7cff49e718398a2fac02d2947480a00ca/numba-0.61.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:4ddce10009bc097b080fc96876d14c051cc0c7679e99de3e0af59014dab7dfe8", size = 2779287, upload-time = "2025-04-09T02:57:53.658Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/2d/e518df036feab381c23a624dac47f8445ac55686ec7f11083655eb707da3/numba-0.61.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5b1bb509d01f23d70325d3a5a0e237cbc9544dd50e50588bc581ba860c213546", size = 3885928, upload-time = "2025-04-09T02:57:55.206Z" },
+    { url = "https://files.pythonhosted.org/packages/10/0f/23cced68ead67b75d77cfcca3df4991d1855c897ee0ff3fe25a56ed82108/numba-0.61.2-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:48a53a3de8f8793526cbe330f2a39fe9a6638efcbf11bd63f3d2f9757ae345cd", size = 3577115, upload-time = "2025-04-09T02:57:56.818Z" },
+    { url = "https://files.pythonhosted.org/packages/68/1d/ddb3e704c5a8fb90142bf9dc195c27db02a08a99f037395503bfbc1d14b3/numba-0.61.2-cp312-cp312-win_amd64.whl", hash = "sha256:97cf4f12c728cf77c9c1d7c23707e4d8fb4632b46275f8f3397de33e5877af18", size = 2831929, upload-time = "2025-04-09T02:57:58.45Z" },
 ]
 
 [[package]]
 name = "numpy"
-version = "1.26.4"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/65/6e/09db70a523a96d25e115e71cc56a6f9031e7b8cd166c1ac8438307c14058/numpy-1.26.4.tar.gz", hash = "sha256:2a02aba9ed12e4ac4eb3ea9421c420301a0c6460d9830d74a9df87efa4912010", size = 15786129 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/a7/94/ace0fdea5241a27d13543ee117cbc65868e82213fb31a8eb7fe9ff23f313/numpy-1.26.4-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:9ff0f4f29c51e2803569d7a51c2304de5554655a60c5d776e35b4a41413830d0", size = 20631468 },
-    { url = "https://files.pythonhosted.org/packages/20/f7/b24208eba89f9d1b58c1668bc6c8c4fd472b20c45573cb767f59d49fb0f6/numpy-1.26.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:2e4ee3380d6de9c9ec04745830fd9e2eccb3e6cf790d39d7b98ffd19b0dd754a", size = 13966411 },
-    { url = "https://files.pythonhosted.org/packages/fc/a5/4beee6488160798683eed5bdb7eead455892c3b4e1f78d79d8d3f3b084ac/numpy-1.26.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d209d8969599b27ad20994c8e41936ee0964e6da07478d6c35016bc386b66ad4", size = 14219016 },
-    { url = "https://files.pythonhosted.org/packages/4b/d7/ecf66c1cd12dc28b4040b15ab4d17b773b87fa9d29ca16125de01adb36cd/numpy-1.26.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ffa75af20b44f8dba823498024771d5ac50620e6915abac414251bd971b4529f", size = 18240889 },
-    { url = "https://files.pythonhosted.org/packages/24/03/6f229fe3187546435c4f6f89f6d26c129d4f5bed40552899fcf1f0bf9e50/numpy-1.26.4-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:62b8e4b1e28009ef2846b4c7852046736bab361f7aeadeb6a5b89ebec3c7055a", size = 13876746 },
-    { url = "https://files.pythonhosted.org/packages/39/fe/39ada9b094f01f5a35486577c848fe274e374bbf8d8f472e1423a0bbd26d/numpy-1.26.4-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:a4abb4f9001ad2858e7ac189089c42178fcce737e4169dc61321660f1a96c7d2", size = 18078620 },
-    { url = "https://files.pythonhosted.org/packages/d5/ef/6ad11d51197aad206a9ad2286dc1aac6a378059e06e8cf22cd08ed4f20dc/numpy-1.26.4-cp310-cp310-win32.whl", hash = "sha256:bfe25acf8b437eb2a8b2d49d443800a5f18508cd811fea3181723922a8a82b07", size = 5972659 },
-    { url = "https://files.pythonhosted.org/packages/19/77/538f202862b9183f54108557bfda67e17603fc560c384559e769321c9d92/numpy-1.26.4-cp310-cp310-win_amd64.whl", hash = "sha256:b97fe8060236edf3662adfc2c633f56a08ae30560c56310562cb4f95500022d5", size = 15808905 },
-    { url = "https://files.pythonhosted.org/packages/11/57/baae43d14fe163fa0e4c47f307b6b2511ab8d7d30177c491960504252053/numpy-1.26.4-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4c66707fabe114439db9068ee468c26bbdf909cac0fb58686a42a24de1760c71", size = 20630554 },
-    { url = "https://files.pythonhosted.org/packages/1a/2e/151484f49fd03944c4a3ad9c418ed193cfd02724e138ac8a9505d056c582/numpy-1.26.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:edd8b5fe47dab091176d21bb6de568acdd906d1887a4584a15a9a96a1dca06ef", size = 13997127 },
-    { url = "https://files.pythonhosted.org/packages/79/ae/7e5b85136806f9dadf4878bf73cf223fe5c2636818ba3ab1c585d0403164/numpy-1.26.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7ab55401287bfec946ced39700c053796e7cc0e3acbef09993a9ad2adba6ca6e", size = 14222994 },
-    { url = "https://files.pythonhosted.org/packages/3a/d0/edc009c27b406c4f9cbc79274d6e46d634d139075492ad055e3d68445925/numpy-1.26.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:666dbfb6ec68962c033a450943ded891bed2d54e6755e35e5835d63f4f6931d5", size = 18252005 },
-    { url = "https://files.pythonhosted.org/packages/09/bf/2b1aaf8f525f2923ff6cfcf134ae5e750e279ac65ebf386c75a0cf6da06a/numpy-1.26.4-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:96ff0b2ad353d8f990b63294c8986f1ec3cb19d749234014f4e7eb0112ceba5a", size = 13885297 },
-    { url = "https://files.pythonhosted.org/packages/df/a0/4e0f14d847cfc2a633a1c8621d00724f3206cfeddeb66d35698c4e2cf3d2/numpy-1.26.4-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:60dedbb91afcbfdc9bc0b1f3f402804070deed7392c23eb7a7f07fa857868e8a", size = 18093567 },
-    { url = "https://files.pythonhosted.org/packages/d2/b7/a734c733286e10a7f1a8ad1ae8c90f2d33bf604a96548e0a4a3a6739b468/numpy-1.26.4-cp311-cp311-win32.whl", hash = "sha256:1af303d6b2210eb850fcf03064d364652b7120803a0b872f5211f5234b399f20", size = 5968812 },
-    { url = "https://files.pythonhosted.org/packages/3f/6b/5610004206cf7f8e7ad91c5a85a8c71b2f2f8051a0c0c4d5916b76d6cbb2/numpy-1.26.4-cp311-cp311-win_amd64.whl", hash = "sha256:cd25bcecc4974d09257ffcd1f098ee778f7834c3ad767fe5db785be9a4aa9cb2", size = 15811913 },
-    { url = "https://files.pythonhosted.org/packages/95/12/8f2020a8e8b8383ac0177dc9570aad031a3beb12e38847f7129bacd96228/numpy-1.26.4-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:b3ce300f3644fb06443ee2222c2201dd3a89ea6040541412b8fa189341847218", size = 20335901 },
-    { url = "https://files.pythonhosted.org/packages/75/5b/ca6c8bd14007e5ca171c7c03102d17b4f4e0ceb53957e8c44343a9546dcc/numpy-1.26.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:03a8c78d01d9781b28a6989f6fa1bb2c4f2d51201cf99d3dd875df6fbd96b23b", size = 13685868 },
-    { url = "https://files.pythonhosted.org/packages/79/f8/97f10e6755e2a7d027ca783f63044d5b1bc1ae7acb12afe6a9b4286eac17/numpy-1.26.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9fad7dcb1aac3c7f0584a5a8133e3a43eeb2fe127f47e3632d43d677c66c102b", size = 13925109 },
-    { url = "https://files.pythonhosted.org/packages/0f/50/de23fde84e45f5c4fda2488c759b69990fd4512387a8632860f3ac9cd225/numpy-1.26.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:675d61ffbfa78604709862923189bad94014bef562cc35cf61d3a07bba02a7ed", size = 17950613 },
-    { url = "https://files.pythonhosted.org/packages/4c/0c/9c603826b6465e82591e05ca230dfc13376da512b25ccd0894709b054ed0/numpy-1.26.4-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:ab47dbe5cc8210f55aa58e4805fe224dac469cde56b9f731a4c098b91917159a", size = 13572172 },
-    { url = "https://files.pythonhosted.org/packages/76/8c/2ba3902e1a0fc1c74962ea9bb33a534bb05984ad7ff9515bf8d07527cadd/numpy-1.26.4-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:1dda2e7b4ec9dd512f84935c5f126c8bd8b9f2fc001e9f54af255e8c5f16b0e0", size = 17786643 },
-    { url = "https://files.pythonhosted.org/packages/28/4a/46d9e65106879492374999e76eb85f87b15328e06bd1550668f79f7b18c6/numpy-1.26.4-cp312-cp312-win32.whl", hash = "sha256:50193e430acfc1346175fcbdaa28ffec49947a06918b7b92130744e81e640110", size = 5677803 },
-    { url = "https://files.pythonhosted.org/packages/16/2e/86f24451c2d530c88daf997cb8d6ac622c1d40d19f5a031ed68a4b73a374/numpy-1.26.4-cp312-cp312-win_amd64.whl", hash = "sha256:08beddf13648eb95f8d867350f6a018a4be2e5ad54c8d8caed89ebca558b2818", size = 15517754 },
-    { url = "https://files.pythonhosted.org/packages/7d/24/ce71dc08f06534269f66e73c04f5709ee024a1afe92a7b6e1d73f158e1f8/numpy-1.26.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7349ab0fa0c429c82442a27a9673fc802ffdb7c7775fad780226cb234965e53c", size = 20636301 },
-    { url = "https://files.pythonhosted.org/packages/ae/8c/ab03a7c25741f9ebc92684a20125fbc9fc1b8e1e700beb9197d750fdff88/numpy-1.26.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:52b8b60467cd7dd1e9ed082188b4e6bb35aa5cdd01777621a1658910745b90be", size = 13971216 },
-    { url = "https://files.pythonhosted.org/packages/6d/64/c3bcdf822269421d85fe0d64ba972003f9bb4aa9a419da64b86856c9961f/numpy-1.26.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d5241e0a80d808d70546c697135da2c613f30e28251ff8307eb72ba696945764", size = 14226281 },
-    { url = "https://files.pythonhosted.org/packages/54/30/c2a907b9443cf42b90c17ad10c1e8fa801975f01cb9764f3f8eb8aea638b/numpy-1.26.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f870204a840a60da0b12273ef34f7051e98c3b5961b61b0c2c1be6dfd64fbcd3", size = 18249516 },
-    { url = "https://files.pythonhosted.org/packages/43/12/01a563fc44c07095996d0129b8899daf89e4742146f7044cdbdb3101c57f/numpy-1.26.4-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:679b0076f67ecc0138fd2ede3a8fd196dddc2ad3254069bcb9faf9a79b1cebcd", size = 13882132 },
-    { url = "https://files.pythonhosted.org/packages/16/ee/9df80b06680aaa23fc6c31211387e0db349e0e36d6a63ba3bd78c5acdf11/numpy-1.26.4-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:47711010ad8555514b434df65f7d7b076bb8261df1ca9bb78f53d3b2db02e95c", size = 18084181 },
-    { url = "https://files.pythonhosted.org/packages/28/7d/4b92e2fe20b214ffca36107f1a3e75ef4c488430e64de2d9af5db3a4637d/numpy-1.26.4-cp39-cp39-win32.whl", hash = "sha256:a354325ee03388678242a4d7ebcd08b5c727033fcff3b2f536aea978e15ee9e6", size = 5976360 },
-    { url = "https://files.pythonhosted.org/packages/b5/42/054082bd8220bbf6f297f982f0a8f5479fcbc55c8b511d928df07b965869/numpy-1.26.4-cp39-cp39-win_amd64.whl", hash = "sha256:3373d5d70a5fe74a2c1bb6d2cfd9609ecf686d47a2d7b1d37a8f3b6bf6003aea", size = 15814633 },
-    { url = "https://files.pythonhosted.org/packages/3f/72/3df6c1c06fc83d9cfe381cccb4be2532bbd38bf93fbc9fad087b6687f1c0/numpy-1.26.4-pp39-pypy39_pp73-macosx_10_9_x86_64.whl", hash = "sha256:afedb719a9dcfc7eaf2287b839d8198e06dcd4cb5d276a3df279231138e83d30", size = 20455961 },
-    { url = "https://files.pythonhosted.org/packages/8e/02/570545bac308b58ffb21adda0f4e220ba716fb658a63c151daecc3293350/numpy-1.26.4-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:95a7476c59002f2f6c590b9b7b998306fba6a5aa646b1e22ddfeaf8f78c3a29c", size = 18061071 },
-    { url = "https://files.pythonhosted.org/packages/f4/5f/fafd8c51235f60d49f7a88e2275e13971e90555b67da52dd6416caec32fe/numpy-1.26.4-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:7e50d0a0cc3189f9cb0aeb3a6a6af18c16f59f004b866cd2be1c14b36134a4a0", size = 15709730 },
+version = "2.0.2"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a9/75/10dd1f8116a8b796cb2c737b674e02d02e80454bda953fa7e65d8c12b016/numpy-2.0.2.tar.gz", hash = "sha256:883c987dee1880e2a864ab0dc9892292582510604156762362d9326444636e78", size = 18902015, upload-time = "2024-08-26T20:19:40.945Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/21/91/3495b3237510f79f5d81f2508f9f13fea78ebfdf07538fc7444badda173d/numpy-2.0.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:51129a29dbe56f9ca83438b706e2e69a39892b5eda6cedcb6b0c9fdc9b0d3ece", size = 21165245, upload-time = "2024-08-26T20:04:14.625Z" },
+    { url = "https://files.pythonhosted.org/packages/05/33/26178c7d437a87082d11019292dce6d3fe6f0e9026b7b2309cbf3e489b1d/numpy-2.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:f15975dfec0cf2239224d80e32c3170b1d168335eaedee69da84fbe9f1f9cd04", size = 13738540, upload-time = "2024-08-26T20:04:36.784Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/31/cc46e13bf07644efc7a4bf68df2df5fb2a1a88d0cd0da9ddc84dc0033e51/numpy-2.0.2-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:8c5713284ce4e282544c68d1c3b2c7161d38c256d2eefc93c1d683cf47683e66", size = 5300623, upload-time = "2024-08-26T20:04:46.491Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/16/7bfcebf27bb4f9d7ec67332ffebee4d1bf085c84246552d52dbb548600e7/numpy-2.0.2-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:becfae3ddd30736fe1889a37f1f580e245ba79a5855bff5f2a29cb3ccc22dd7b", size = 6901774, upload-time = "2024-08-26T20:04:58.173Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/a3/561c531c0e8bf082c5bef509d00d56f82e0ea7e1e3e3a7fc8fa78742a6e5/numpy-2.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2da5960c3cf0df7eafefd806d4e612c5e19358de82cb3c343631188991566ccd", size = 13907081, upload-time = "2024-08-26T20:05:19.098Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/66/f7177ab331876200ac7563a580140643d1179c8b4b6a6b0fc9838de2a9b8/numpy-2.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:496f71341824ed9f3d2fd36cf3ac57ae2e0165c143b55c3a035ee219413f3318", size = 19523451, upload-time = "2024-08-26T20:05:47.479Z" },
+    { url = "https://files.pythonhosted.org/packages/25/7f/0b209498009ad6453e4efc2c65bcdf0ae08a182b2b7877d7ab38a92dc542/numpy-2.0.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:a61ec659f68ae254e4d237816e33171497e978140353c0c2038d46e63282d0c8", size = 19927572, upload-time = "2024-08-26T20:06:17.137Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/df/2619393b1e1b565cd2d4c4403bdd979621e2c4dea1f8532754b2598ed63b/numpy-2.0.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d731a1c6116ba289c1e9ee714b08a8ff882944d4ad631fd411106a30f083c326", size = 14400722, upload-time = "2024-08-26T20:06:39.16Z" },
+    { url = "https://files.pythonhosted.org/packages/22/ad/77e921b9f256d5da36424ffb711ae79ca3f451ff8489eeca544d0701d74a/numpy-2.0.2-cp310-cp310-win32.whl", hash = "sha256:984d96121c9f9616cd33fbd0618b7f08e0cfc9600a7ee1d6fd9b239186d19d97", size = 6472170, upload-time = "2024-08-26T20:06:50.361Z" },
+    { url = "https://files.pythonhosted.org/packages/10/05/3442317535028bc29cf0c0dd4c191a4481e8376e9f0db6bcf29703cadae6/numpy-2.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:c7b0be4ef08607dd04da4092faee0b86607f111d5ae68036f16cc787e250a131", size = 15905558, upload-time = "2024-08-26T20:07:13.881Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/cf/034500fb83041aa0286e0fb16e7c76e5c8b67c0711bb6e9e9737a717d5fe/numpy-2.0.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:49ca4decb342d66018b01932139c0961a8f9ddc7589611158cb3c27cbcf76448", size = 21169137, upload-time = "2024-08-26T20:07:45.345Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/d9/32de45561811a4b87fbdee23b5797394e3d1504b4a7cf40c10199848893e/numpy-2.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:11a76c372d1d37437857280aa142086476136a8c0f373b2e648ab2c8f18fb195", size = 13703552, upload-time = "2024-08-26T20:08:06.666Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/ca/2f384720020c7b244d22508cb7ab23d95f179fcfff33c31a6eeba8d6c512/numpy-2.0.2-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:807ec44583fd708a21d4a11d94aedf2f4f3c3719035c76a2bbe1fe8e217bdc57", size = 5298957, upload-time = "2024-08-26T20:08:15.83Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/78/a3e4f9fb6aa4e6fdca0c5428e8ba039408514388cf62d89651aade838269/numpy-2.0.2-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:8cafab480740e22f8d833acefed5cc87ce276f4ece12fdaa2e8903db2f82897a", size = 6905573, upload-time = "2024-08-26T20:08:27.185Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/72/cfc3a1beb2caf4efc9d0b38a15fe34025230da27e1c08cc2eb9bfb1c7231/numpy-2.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a15f476a45e6e5a3a79d8a14e62161d27ad897381fecfa4a09ed5322f2085669", size = 13914330, upload-time = "2024-08-26T20:08:48.058Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/a8/c17acf65a931ce551fee11b72e8de63bf7e8a6f0e21add4c937c83563538/numpy-2.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:13e689d772146140a252c3a28501da66dfecd77490b498b168b501835041f951", size = 19534895, upload-time = "2024-08-26T20:09:16.536Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/86/8767f3d54f6ae0165749f84648da9dcc8cd78ab65d415494962c86fac80f/numpy-2.0.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:9ea91dfb7c3d1c56a0e55657c0afb38cf1eeae4544c208dc465c3c9f3a7c09f9", size = 19937253, upload-time = "2024-08-26T20:09:46.263Z" },
+    { url = "https://files.pythonhosted.org/packages/df/87/f76450e6e1c14e5bb1eae6836478b1028e096fd02e85c1c37674606ab752/numpy-2.0.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:c1c9307701fec8f3f7a1e6711f9089c06e6284b3afbbcd259f7791282d660a15", size = 14414074, upload-time = "2024-08-26T20:10:08.483Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/ca/0f0f328e1e59f73754f06e1adfb909de43726d4f24c6a3f8805f34f2b0fa/numpy-2.0.2-cp311-cp311-win32.whl", hash = "sha256:a392a68bd329eafac5817e5aefeb39038c48b671afd242710b451e76090e81f4", size = 6470640, upload-time = "2024-08-26T20:10:19.732Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/57/3a3f14d3a759dcf9bf6e9eda905794726b758819df4663f217d658a58695/numpy-2.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:286cd40ce2b7d652a6f22efdfc6d1edf879440e53e76a75955bc0c826c7e64dc", size = 15910230, upload-time = "2024-08-26T20:10:43.413Z" },
+    { url = "https://files.pythonhosted.org/packages/45/40/2e117be60ec50d98fa08c2f8c48e09b3edea93cfcabd5a9ff6925d54b1c2/numpy-2.0.2-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:df55d490dea7934f330006d0f81e8551ba6010a5bf035a249ef61a94f21c500b", size = 20895803, upload-time = "2024-08-26T20:11:13.916Z" },
+    { url = "https://files.pythonhosted.org/packages/46/92/1b8b8dee833f53cef3e0a3f69b2374467789e0bb7399689582314df02651/numpy-2.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8df823f570d9adf0978347d1f926b2a867d5608f434a7cff7f7908c6570dcf5e", size = 13471835, upload-time = "2024-08-26T20:11:34.779Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/19/e2793bde475f1edaea6945be141aef6c8b4c669b90c90a300a8954d08f0a/numpy-2.0.2-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:9a92ae5c14811e390f3767053ff54eaee3bf84576d99a2456391401323f4ec2c", size = 5038499, upload-time = "2024-08-26T20:11:43.902Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/ff/ddf6dac2ff0dd50a7327bcdba45cb0264d0e96bb44d33324853f781a8f3c/numpy-2.0.2-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:a842d573724391493a97a62ebbb8e731f8a5dcc5d285dfc99141ca15a3302d0c", size = 6633497, upload-time = "2024-08-26T20:11:55.09Z" },
+    { url = "https://files.pythonhosted.org/packages/72/21/67f36eac8e2d2cd652a2e69595a54128297cdcb1ff3931cfc87838874bd4/numpy-2.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c05e238064fc0610c840d1cf6a13bf63d7e391717d247f1bf0318172e759e692", size = 13621158, upload-time = "2024-08-26T20:12:14.95Z" },
+    { url = "https://files.pythonhosted.org/packages/39/68/e9f1126d757653496dbc096cb429014347a36b228f5a991dae2c6b6cfd40/numpy-2.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0123ffdaa88fa4ab64835dcbde75dcdf89c453c922f18dced6e27c90d1d0ec5a", size = 19236173, upload-time = "2024-08-26T20:12:44.049Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/e9/1f5333281e4ebf483ba1c888b1d61ba7e78d7e910fdd8e6499667041cc35/numpy-2.0.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:96a55f64139912d61de9137f11bf39a55ec8faec288c75a54f93dfd39f7eb40c", size = 19634174, upload-time = "2024-08-26T20:13:13.634Z" },
+    { url = "https://files.pythonhosted.org/packages/71/af/a469674070c8d8408384e3012e064299f7a2de540738a8e414dcfd639996/numpy-2.0.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:ec9852fb39354b5a45a80bdab5ac02dd02b15f44b3804e9f00c556bf24b4bded", size = 14099701, upload-time = "2024-08-26T20:13:34.851Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/3d/08ea9f239d0e0e939b6ca52ad403c84a2bce1bde301a8eb4888c1c1543f1/numpy-2.0.2-cp312-cp312-win32.whl", hash = "sha256:671bec6496f83202ed2d3c8fdc486a8fc86942f2e69ff0e986140339a63bcbe5", size = 6174313, upload-time = "2024-08-26T20:13:45.653Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/b5/4ac39baebf1fdb2e72585c8352c56d063b6126be9fc95bd2bb5ef5770c20/numpy-2.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:cfd41e13fdc257aa5778496b8caa5e856dc4896d4ccf01841daee1d96465467a", size = 15606179, upload-time = "2024-08-26T20:14:08.786Z" },
+    { url = "https://files.pythonhosted.org/packages/43/c1/41c8f6df3162b0c6ffd4437d729115704bd43363de0090c7f913cfbc2d89/numpy-2.0.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:9059e10581ce4093f735ed23f3b9d283b9d517ff46009ddd485f1747eb22653c", size = 21169942, upload-time = "2024-08-26T20:14:40.108Z" },
+    { url = "https://files.pythonhosted.org/packages/39/bc/fd298f308dcd232b56a4031fd6ddf11c43f9917fbc937e53762f7b5a3bb1/numpy-2.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:423e89b23490805d2a5a96fe40ec507407b8ee786d66f7328be214f9679df6dd", size = 13711512, upload-time = "2024-08-26T20:15:00.985Z" },
+    { url = "https://files.pythonhosted.org/packages/96/ff/06d1aa3eeb1c614eda245c1ba4fb88c483bee6520d361641331872ac4b82/numpy-2.0.2-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:2b2955fa6f11907cf7a70dab0d0755159bca87755e831e47932367fc8f2f2d0b", size = 5306976, upload-time = "2024-08-26T20:15:10.876Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/98/121996dcfb10a6087a05e54453e28e58694a7db62c5a5a29cee14c6e047b/numpy-2.0.2-cp39-cp39-macosx_14_0_x86_64.whl", hash = "sha256:97032a27bd9d8988b9a97a8c4d2c9f2c15a81f61e2f21404d7e8ef00cb5be729", size = 6906494, upload-time = "2024-08-26T20:15:22.055Z" },
+    { url = "https://files.pythonhosted.org/packages/15/31/9dffc70da6b9bbf7968f6551967fc21156207366272c2a40b4ed6008dc9b/numpy-2.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1e795a8be3ddbac43274f18588329c72939870a16cae810c2b73461c40718ab1", size = 13912596, upload-time = "2024-08-26T20:15:42.452Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/14/78635daab4b07c0930c919d451b8bf8c164774e6a3413aed04a6d95758ce/numpy-2.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f26b258c385842546006213344c50655ff1555a9338e2e5e02a0756dc3e803dd", size = 19526099, upload-time = "2024-08-26T20:16:11.048Z" },
+    { url = "https://files.pythonhosted.org/packages/26/4c/0eeca4614003077f68bfe7aac8b7496f04221865b3a5e7cb230c9d055afd/numpy-2.0.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:5fec9451a7789926bcf7c2b8d187292c9f93ea30284802a0ab3f5be8ab36865d", size = 19932823, upload-time = "2024-08-26T20:16:40.171Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/46/ea25b98b13dccaebddf1a803f8c748680d972e00507cd9bc6dcdb5aa2ac1/numpy-2.0.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:9189427407d88ff25ecf8f12469d4d39d35bee1db5d39fc5c168c6f088a6956d", size = 14404424, upload-time = "2024-08-26T20:17:02.604Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/a6/177dd88d95ecf07e722d21008b1b40e681a929eb9e329684d449c36586b2/numpy-2.0.2-cp39-cp39-win32.whl", hash = "sha256:905d16e0c60200656500c95b6b8dca5d109e23cb24abc701d41c02d74c6b3afa", size = 6476809, upload-time = "2024-08-26T20:17:13.553Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/2b/7fc9f4e7ae5b507c1a3a21f0f15ed03e794c1242ea8a242ac158beb56034/numpy-2.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:a3f4ab0caa7f053f6797fcd4e1e25caee367db3112ef2b6ef82d749530768c73", size = 15911314, upload-time = "2024-08-26T20:17:36.72Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/3b/df5a870ac6a3be3a86856ce195ef42eec7ae50d2a202be1f5a4b3b340e14/numpy-2.0.2-pp39-pypy39_pp73-macosx_10_9_x86_64.whl", hash = "sha256:7f0a0c6f12e07fa94133c8a67404322845220c06a9e80e85999afe727f7438b8", size = 21025288, upload-time = "2024-08-26T20:18:07.732Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/97/51af92f18d6f6f2d9ad8b482a99fb74e142d71372da5d834b3a2747a446e/numpy-2.0.2-pp39-pypy39_pp73-macosx_14_0_x86_64.whl", hash = "sha256:312950fdd060354350ed123c0e25a71327d3711584beaef30cdaa93320c392d4", size = 6762793, upload-time = "2024-08-26T20:18:19.125Z" },
+    { url = "https://files.pythonhosted.org/packages/12/46/de1fbd0c1b5ccaa7f9a005b66761533e2f6a3e560096682683a223631fe9/numpy-2.0.2-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:26df23238872200f63518dd2aa984cfca675d82469535dc7162dc2ee52d9dd5c", size = 19334885, upload-time = "2024-08-26T20:18:47.237Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/dc/d330a6faefd92b446ec0f0dfea4c3207bb1fef3c4771d19cf4543efd2c78/numpy-2.0.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:a46288ec55ebbd58947d31d72be2c63cbf839f0a63b49cb755022310792a3385", size = 15828784, upload-time = "2024-08-26T20:19:11.19Z" },
+]
+
+[[package]]
+name = "numpy"
+version = "2.1.3"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/25/ca/1166b75c21abd1da445b97bf1fa2f14f423c6cfb4fc7c4ef31dccf9f6a94/numpy-2.1.3.tar.gz", hash = "sha256:aa08e04e08aaf974d4458def539dece0d28146d866a39da5639596f4921fd761", size = 20166090, upload-time = "2024-11-02T17:48:55.832Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f1/80/d572a4737626372915bca41c3afbfec9d173561a39a0a61bacbbfd1dafd4/numpy-2.1.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:c894b4305373b9c5576d7a12b473702afdf48ce5369c074ba304cc5ad8730dff", size = 21152472, upload-time = "2024-11-02T17:30:37.354Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/bb/7bfba10c791ae3bb6716da77ad85a82d5fac07fc96fb0023ef0571df9d20/numpy-2.1.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:b47fbb433d3260adcd51eb54f92a2ffbc90a4595f8970ee00e064c644ac788f5", size = 13747967, upload-time = "2024-11-02T17:30:59.602Z" },
+    { url = "https://files.pythonhosted.org/packages/da/d6/2df7bde35f0478455f0be5934877b3e5a505f587b00230f54a519a6b55a5/numpy-2.1.3-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:825656d0743699c529c5943554d223c021ff0494ff1442152ce887ef4f7561a1", size = 5354921, upload-time = "2024-11-02T17:31:09.428Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/bb/75b945874f931494891eac6ca06a1764d0e8208791f3addadb2963b83527/numpy-2.1.3-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:6a4825252fcc430a182ac4dee5a505053d262c807f8a924603d411f6718b88fd", size = 6888603, upload-time = "2024-11-02T17:31:20.835Z" },
+    { url = "https://files.pythonhosted.org/packages/68/a7/fde73636f6498dbfa6d82fc336164635fe592f1ad0d13285fcb6267fdc1c/numpy-2.1.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e711e02f49e176a01d0349d82cb5f05ba4db7d5e7e0defd026328e5cfb3226d3", size = 13889862, upload-time = "2024-11-02T17:31:41.486Z" },
+    { url = "https://files.pythonhosted.org/packages/05/db/5d9c91b2e1e2e72be1369278f696356d44975befcae830daf2e667dcb54f/numpy-2.1.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:78574ac2d1a4a02421f25da9559850d59457bac82f2b8d7a44fe83a64f770098", size = 16328151, upload-time = "2024-11-02T17:32:08.262Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/6a/7eb732109b53ae64a29e25d7e68eb9d6611037f6354875497008a49e74d3/numpy-2.1.3-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:c7662f0e3673fe4e832fe07b65c50342ea27d989f92c80355658c7f888fcc83c", size = 16704107, upload-time = "2024-11-02T17:32:34.361Z" },
+    { url = "https://files.pythonhosted.org/packages/88/cc/278113b66a1141053cbda6f80e4200c6da06b3079c2d27bda1fde41f2c1f/numpy-2.1.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:fa2d1337dc61c8dc417fbccf20f6d1e139896a30721b7f1e832b2bb6ef4eb6c4", size = 14385789, upload-time = "2024-11-02T17:32:57.152Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/69/eb20f5e1bfa07449bc67574d2f0f7c1e6b335fb41672e43861a7727d85f2/numpy-2.1.3-cp310-cp310-win32.whl", hash = "sha256:72dcc4a35a8515d83e76b58fdf8113a5c969ccd505c8a946759b24e3182d1f23", size = 6536706, upload-time = "2024-11-02T17:33:09.12Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/8b/1c131ab5a94c1086c289c6e1da1d843de9dbd95fe5f5ee6e61904c9518e2/numpy-2.1.3-cp310-cp310-win_amd64.whl", hash = "sha256:ecc76a9ba2911d8d37ac01de72834d8849e55473457558e12995f4cd53e778e0", size = 12864165, upload-time = "2024-11-02T17:33:28.974Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/81/c8167192eba5247593cd9d305ac236847c2912ff39e11402e72ae28a4985/numpy-2.1.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4d1167c53b93f1f5d8a139a742b3c6f4d429b54e74e6b57d0eff40045187b15d", size = 21156252, upload-time = "2024-11-02T17:34:01.372Z" },
+    { url = "https://files.pythonhosted.org/packages/da/74/5a60003fc3d8a718d830b08b654d0eea2d2db0806bab8f3c2aca7e18e010/numpy-2.1.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:c80e4a09b3d95b4e1cac08643f1152fa71a0a821a2d4277334c88d54b2219a41", size = 13784119, upload-time = "2024-11-02T17:34:23.809Z" },
+    { url = "https://files.pythonhosted.org/packages/47/7c/864cb966b96fce5e63fcf25e1e4d957fe5725a635e5f11fe03f39dd9d6b5/numpy-2.1.3-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:576a1c1d25e9e02ed7fa5477f30a127fe56debd53b8d2c89d5578f9857d03ca9", size = 5352978, upload-time = "2024-11-02T17:34:34.001Z" },
+    { url = "https://files.pythonhosted.org/packages/09/ac/61d07930a4993dd9691a6432de16d93bbe6aa4b1c12a5e573d468eefc1ca/numpy-2.1.3-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:973faafebaae4c0aaa1a1ca1ce02434554d67e628b8d805e61f874b84e136b09", size = 6892570, upload-time = "2024-11-02T17:34:45.401Z" },
+    { url = "https://files.pythonhosted.org/packages/27/2f/21b94664f23af2bb52030653697c685022119e0dc93d6097c3cb45bce5f9/numpy-2.1.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:762479be47a4863e261a840e8e01608d124ee1361e48b96916f38b119cfda04a", size = 13896715, upload-time = "2024-11-02T17:35:06.564Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/f0/80811e836484262b236c684a75dfc4ba0424bc670e765afaa911468d9f39/numpy-2.1.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc6f24b3d1ecc1eebfbf5d6051faa49af40b03be1aaa781ebdadcbc090b4539b", size = 16339644, upload-time = "2024-11-02T17:35:30.888Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/81/ce213159a1ed8eb7d88a2a6ef4fbdb9e4ffd0c76b866c350eb4e3c37e640/numpy-2.1.3-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:17ee83a1f4fef3c94d16dc1802b998668b5419362c8a4f4e8a491de1b41cc3ee", size = 16712217, upload-time = "2024-11-02T17:35:56.703Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/84/4de0b87d5a72f45556b2a8ee9fc8801e8518ec867fc68260c1f5dcb3903f/numpy-2.1.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:15cb89f39fa6d0bdfb600ea24b250e5f1a3df23f901f51c8debaa6a5d122b2f0", size = 14399053, upload-time = "2024-11-02T17:36:22.3Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/1c/e5fabb9ad849f9d798b44458fd12a318d27592d4bc1448e269dec070ff04/numpy-2.1.3-cp311-cp311-win32.whl", hash = "sha256:d9beb777a78c331580705326d2367488d5bc473b49a9bc3036c154832520aca9", size = 6534741, upload-time = "2024-11-02T17:36:33.552Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/48/a9a4b538e28f854bfb62e1dea3c8fea12e90216a276c7777ae5345ff29a7/numpy-2.1.3-cp311-cp311-win_amd64.whl", hash = "sha256:d89dd2b6da69c4fff5e39c28a382199ddedc3a5be5390115608345dec660b9e2", size = 12869487, upload-time = "2024-11-02T17:36:52.909Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/f0/385eb9970309643cbca4fc6eebc8bb16e560de129c91258dfaa18498da8b/numpy-2.1.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f55ba01150f52b1027829b50d70ef1dafd9821ea82905b63936668403c3b471e", size = 20849658, upload-time = "2024-11-02T17:37:23.919Z" },
+    { url = "https://files.pythonhosted.org/packages/54/4a/765b4607f0fecbb239638d610d04ec0a0ded9b4951c56dc68cef79026abf/numpy-2.1.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:13138eadd4f4da03074851a698ffa7e405f41a0845a6b1ad135b81596e4e9958", size = 13492258, upload-time = "2024-11-02T17:37:45.252Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/a7/2332679479c70b68dccbf4a8eb9c9b5ee383164b161bee9284ac141fbd33/numpy-2.1.3-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:a6b46587b14b888e95e4a24d7b13ae91fa22386c199ee7b418f449032b2fa3b8", size = 5090249, upload-time = "2024-11-02T17:37:54.252Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/67/4aa00316b3b981a822c7a239d3a8135be2a6945d1fd11d0efb25d361711a/numpy-2.1.3-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:0fa14563cc46422e99daef53d725d0c326e99e468a9320a240affffe87852564", size = 6621704, upload-time = "2024-11-02T17:38:05.127Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/da/1a429ae58b3b6c364eeec93bf044c532f2ff7b48a52e41050896cf15d5b1/numpy-2.1.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8637dcd2caa676e475503d1f8fdb327bc495554e10838019651b76d17b98e512", size = 13606089, upload-time = "2024-11-02T17:38:25.997Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/3e/3757f304c704f2f0294a6b8340fcf2be244038be07da4cccf390fa678a9f/numpy-2.1.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2312b2aa89e1f43ecea6da6ea9a810d06aae08321609d8dc0d0eda6d946a541b", size = 16043185, upload-time = "2024-11-02T17:38:51.07Z" },
+    { url = "https://files.pythonhosted.org/packages/43/97/75329c28fea3113d00c8d2daf9bc5828d58d78ed661d8e05e234f86f0f6d/numpy-2.1.3-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:a38c19106902bb19351b83802531fea19dee18e5b37b36454f27f11ff956f7fc", size = 16410751, upload-time = "2024-11-02T17:39:15.801Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/7a/442965e98b34e0ae9da319f075b387bcb9a1e0658276cc63adb8c9686f7b/numpy-2.1.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:02135ade8b8a84011cbb67dc44e07c58f28575cf9ecf8ab304e51c05528c19f0", size = 14082705, upload-time = "2024-11-02T17:39:38.274Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/b6/26108cf2cfa5c7e03fb969b595c93131eab4a399762b51ce9ebec2332e80/numpy-2.1.3-cp312-cp312-win32.whl", hash = "sha256:e6988e90fcf617da2b5c78902fe8e668361b43b4fe26dbf2d7b0f8034d4cafb9", size = 6239077, upload-time = "2024-11-02T17:39:49.299Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/84/fa11dad3404b7634aaab50733581ce11e5350383311ea7a7010f464c0170/numpy-2.1.3-cp312-cp312-win_amd64.whl", hash = "sha256:0d30c543f02e84e92c4b1f415b7c6b5326cbe45ee7882b6b77db7195fb971e3a", size = 12566858, upload-time = "2024-11-02T17:40:08.851Z" },
+    { url = "https://files.pythonhosted.org/packages/00/e7/8d8bb791b62586cc432ecbb70632b4f23b7b7c88df41878de7528264f6d7/numpy-2.1.3-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:4f2015dfe437dfebbfce7c85c7b53d81ba49e71ba7eadbf1df40c915af75979f", size = 20983893, upload-time = "2024-11-02T17:47:09.365Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/f3/cb8118a044b5007586245a650360c9f5915b2f4232dd7658bb7a63dd1d02/numpy-2.1.3-pp310-pypy310_pp73-macosx_14_0_x86_64.whl", hash = "sha256:3522b0dfe983a575e6a9ab3a4a4dfe156c3e428468ff08ce582b9bb6bd1d71d4", size = 6752501, upload-time = "2024-11-02T17:47:21.52Z" },
+    { url = "https://files.pythonhosted.org/packages/53/f5/365b46439b518d2ec6ebb880cc0edf90f225145dfd4db7958334f7164530/numpy-2.1.3-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c006b607a865b07cd981ccb218a04fc86b600411d83d6fc261357f1c0966755d", size = 16142601, upload-time = "2024-11-02T17:47:45.575Z" },
+    { url = "https://files.pythonhosted.org/packages/03/c2/d1fee6ba999aa7cd41ca6856937f2baaf604c3eec1565eae63451ec31e5e/numpy-2.1.3-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:e14e26956e6f1696070788252dcdff11b4aca4c3e8bd166e0df1bb8f315a67cb", size = 12771397, upload-time = "2024-11-02T17:48:05.988Z" },
 ]
 
 [[package]]
 name = "nvidia-cublas-cu12"
-version = "12.1.3.1"
+version = "12.8.4.1"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/37/6d/121efd7382d5b0284239f4ab1fc1590d86d34ed4a4a2fdb13b30ca8e5740/nvidia_cublas_cu12-12.1.3.1-py3-none-manylinux1_x86_64.whl", hash = "sha256:ee53ccca76a6fc08fb9701aa95b6ceb242cdaab118c3bb152af4e579af792728", size = 410594774 },
+    { url = "https://files.pythonhosted.org/packages/dc/61/e24b560ab2e2eaeb3c839129175fb330dfcfc29e5203196e5541a4c44682/nvidia_cublas_cu12-12.8.4.1-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:8ac4e771d5a348c551b2a426eda6193c19aa630236b418086020df5ba9667142", size = 594346921, upload-time = "2025-03-07T01:44:31.254Z" },
 ]
 
 [[package]]
 name = "nvidia-cuda-cupti-cu12"
-version = "12.1.105"
+version = "12.8.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/7e/00/6b218edd739ecfc60524e585ba8e6b00554dd908de2c9c66c1af3e44e18d/nvidia_cuda_cupti_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:e54fde3983165c624cb79254ae9818a456eb6e87a7fd4d56a2352c24ee542d7e", size = 14109015 },
+    { url = "https://files.pythonhosted.org/packages/f8/02/2adcaa145158bf1a8295d83591d22e4103dbfd821bcaf6f3f53151ca4ffa/nvidia_cuda_cupti_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ea0cb07ebda26bb9b29ba82cda34849e73c166c18162d3913575b0c9db9a6182", size = 10248621, upload-time = "2025-03-07T01:40:21.213Z" },
 ]
 
 [[package]]
 name = "nvidia-cuda-nvrtc-cu12"
-version = "12.1.105"
+version = "12.8.93"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b6/9f/c64c03f49d6fbc56196664d05dba14e3a561038a81a638eeb47f4d4cfd48/nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:339b385f50c309763ca65456ec75e17bbefcbbf2893f462cb8b90584cd27a1c2", size = 23671734 },
+    { url = "https://files.pythonhosted.org/packages/05/6b/32f747947df2da6994e999492ab306a903659555dddc0fbdeb9d71f75e52/nvidia_cuda_nvrtc_cu12-12.8.93-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:a7756528852ef889772a84c6cd89d41dfa74667e24cca16bb31f8f061e3e9994", size = 88040029, upload-time = "2025-03-07T01:42:13.562Z" },
 ]
 
 [[package]]
 name = "nvidia-cuda-runtime-cu12"
-version = "12.1.105"
+version = "12.8.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/eb/d5/c68b1d2cdfcc59e72e8a5949a37ddb22ae6cade80cd4a57a84d4c8b55472/nvidia_cuda_runtime_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:6e258468ddf5796e25f1dc591a31029fa317d97a0a94ed93468fc86301d61e40", size = 823596 },
+    { url = "https://files.pythonhosted.org/packages/0d/9b/a997b638fcd068ad6e4d53b8551a7d30fe8b404d6f1804abf1df69838932/nvidia_cuda_runtime_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:adade8dcbd0edf427b7204d480d6066d33902cab2a4707dcfc48a2d0fd44ab90", size = 954765, upload-time = "2025-03-07T01:40:01.615Z" },
 ]
 
 [[package]]
 name = "nvidia-cudnn-cu12"
-version = "8.9.2.26"
+version = "9.10.2.21"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "nvidia-cublas-cu12" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ff/74/a2e2be7fb83aaedec84f391f082cf765dfb635e7caa9b49065f73e4835d8/nvidia_cudnn_cu12-8.9.2.26-py3-none-manylinux1_x86_64.whl", hash = "sha256:5ccb288774fdfb07a7e7025ffec286971c06d8d7b4fb162525334616d7629ff9", size = 731725872 },
+    { url = "https://files.pythonhosted.org/packages/ba/51/e123d997aa098c61d029f76663dedbfb9bc8dcf8c60cbd6adbe42f76d049/nvidia_cudnn_cu12-9.10.2.21-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:949452be657fa16687d0930933f032835951ef0892b37d2d53824d1a84dc97a8", size = 706758467, upload-time = "2025-06-06T21:54:08.597Z" },
 ]
 
 [[package]]
 name = "nvidia-cufft-cu12"
-version = "11.0.2.54"
+version = "11.3.3.83"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "nvidia-nvjitlink-cu12" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1f/13/ee4e00f30e676b66ae65b4f08cb5bcbb8392c03f54f2d5413ea99a5d1c80/nvidia_cufft_cu12-11.3.3.83-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4d2dd21ec0b88cf61b62e6b43564355e5222e4a3fb394cac0db101f2dd0d4f74", size = 193118695, upload-time = "2025-03-07T01:45:27.821Z" },
+]
+
+[[package]]
+name = "nvidia-cufile-cu12"
+version = "1.13.1.3"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/86/94/eb540db023ce1d162e7bea9f8f5aa781d57c65aed513c33ee9a5123ead4d/nvidia_cufft_cu12-11.0.2.54-py3-none-manylinux1_x86_64.whl", hash = "sha256:794e3948a1aa71fd817c3775866943936774d1c14e7628c74f6f7417224cdf56", size = 121635161 },
+    { url = "https://files.pythonhosted.org/packages/bb/fe/1bcba1dfbfb8d01be8d93f07bfc502c93fa23afa6fd5ab3fc7c1df71038a/nvidia_cufile_cu12-1.13.1.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:1d069003be650e131b21c932ec3d8969c1715379251f8d23a1860554b1cb24fc", size = 1197834, upload-time = "2025-03-07T01:45:50.723Z" },
 ]
 
 [[package]]
 name = "nvidia-curand-cu12"
-version = "10.3.2.106"
+version = "10.3.9.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/44/31/4890b1c9abc496303412947fc7dcea3d14861720642b49e8ceed89636705/nvidia_curand_cu12-10.3.2.106-py3-none-manylinux1_x86_64.whl", hash = "sha256:9d264c5036dde4e64f1de8c50ae753237c12e0b1348738169cd0f8a536c0e1e0", size = 56467784 },
+    { url = "https://files.pythonhosted.org/packages/fb/aa/6584b56dc84ebe9cf93226a5cde4d99080c8e90ab40f0c27bda7a0f29aa1/nvidia_curand_cu12-10.3.9.90-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:b32331d4f4df5d6eefa0554c565b626c7216f87a06a4f56fab27c3b68a830ec9", size = 63619976, upload-time = "2025-03-07T01:46:23.323Z" },
 ]
 
 [[package]]
 name = "nvidia-cusolver-cu12"
-version = "11.4.5.107"
+version = "11.7.3.90"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "nvidia-cublas-cu12" },
@@ -1794,56 +2341,68 @@ dependencies = [
     { name = "nvidia-nvjitlink-cu12" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/bc/1d/8de1e5c67099015c834315e333911273a8c6aaba78923dd1d1e25fc5f217/nvidia_cusolver_cu12-11.4.5.107-py3-none-manylinux1_x86_64.whl", hash = "sha256:8a7ec542f0412294b15072fa7dab71d31334014a69f953004ea7a118206fe0dd", size = 124161928 },
+    { url = "https://files.pythonhosted.org/packages/85/48/9a13d2975803e8cf2777d5ed57b87a0b6ca2cc795f9a4f59796a910bfb80/nvidia_cusolver_cu12-11.7.3.90-py3-none-manylinux_2_27_x86_64.whl", hash = "sha256:4376c11ad263152bd50ea295c05370360776f8c3427b30991df774f9fb26c450", size = 267506905, upload-time = "2025-03-07T01:47:16.273Z" },
 ]
 
 [[package]]
 name = "nvidia-cusparse-cu12"
-version = "12.1.0.106"
+version = "12.5.8.93"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "nvidia-nvjitlink-cu12" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/65/5b/cfaeebf25cd9fdec14338ccb16f6b2c4c7fa9163aefcf057d86b9cc248bb/nvidia_cusparse_cu12-12.1.0.106-py3-none-manylinux1_x86_64.whl", hash = "sha256:f3b50f42cf363f86ab21f720998517a659a48131e8d538dc02f8768237bd884c", size = 195958278 },
+    { url = "https://files.pythonhosted.org/packages/c2/f5/e1854cb2f2bcd4280c44736c93550cc300ff4b8c95ebe370d0aa7d2b473d/nvidia_cusparse_cu12-12.5.8.93-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:1ec05d76bbbd8b61b06a80e1eaf8cf4959c3d4ce8e711b65ebd0443bb0ebb13b", size = 288216466, upload-time = "2025-03-07T01:48:13.779Z" },
 ]
 
 [[package]]
-name = "nvidia-ml-py"
-version = "12.560.30"
+name = "nvidia-cusparselt-cu12"
+version = "0.7.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/53/10/5f34de4a71db8b2b7ec4269f4a33287f24c23e2857ea3187c977b7bc3604/nvidia-ml-py-12.560.30.tar.gz", hash = "sha256:f0254dc7400647680a072ee02509bfd46102b60bdfeca321576d4d4817e7fe97", size = 39194 }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b7/f3/a69ce0b1a1e12fbf6b2ad9f4c14c9999fdbdf15f2478d210f0fd501ddc98/nvidia_ml_py-12.560.30-py3-none-any.whl", hash = "sha256:fea371c94d63e38a611c17bbb85fe400e9c8ddb9e8684a9cd0e47786a4bc3c73", size = 40526 },
+    { url = "https://files.pythonhosted.org/packages/56/79/12978b96bd44274fe38b5dde5cfb660b1d114f70a65ef962bcbbed99b549/nvidia_cusparselt_cu12-0.7.1-py3-none-manylinux2014_x86_64.whl", hash = "sha256:f1bb701d6b930d5a7cea44c19ceb973311500847f81b634d802b7b539dc55623", size = 287193691, upload-time = "2025-02-26T00:15:44.104Z" },
 ]
 
 [[package]]
 name = "nvidia-nccl-cu12"
-version = "2.20.5"
+version = "2.27.3"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/4b/2a/0a131f572aa09f741c30ccd45a8e56316e8be8dfc7bc19bf0ab7cfef7b19/nvidia_nccl_cu12-2.20.5-py3-none-manylinux2014_x86_64.whl", hash = "sha256:057f6bf9685f75215d0c53bf3ac4a10b3e6578351de307abad9e18a99182af56", size = 176249402 },
+    { url = "https://files.pythonhosted.org/packages/5c/5b/4e4fff7bad39adf89f735f2bc87248c81db71205b62bcc0d5ca5b606b3c3/nvidia_nccl_cu12-2.27.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:adf27ccf4238253e0b826bce3ff5fa532d65fc42322c8bfdfaf28024c0fbe039", size = 322364134, upload-time = "2025-06-03T21:58:04.013Z" },
 ]
 
 [[package]]
 name = "nvidia-nvjitlink-cu12"
-version = "12.6.85"
+version = "12.8.93"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/9d/d7/c5383e47c7e9bf1c99d5bd2a8c935af2b6d705ad831a7ec5c97db4d82f4f/nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:eedc36df9e88b682efe4309aa16b5b4e78c2407eac59e8c10a6a47535164369a", size = 19744971 },
+    { url = "https://files.pythonhosted.org/packages/f6/74/86a07f1d0f42998ca31312f998bd3b9a7eff7f52378f4f270c8679c77fb9/nvidia_nvjitlink_cu12-12.8.93-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:81ff63371a7ebd6e6451970684f916be2eab07321b73c9d244dc2b4da7f73b88", size = 39254836, upload-time = "2025-03-07T01:49:55.661Z" },
 ]
 
 [[package]]
 name = "nvidia-nvtx-cu12"
-version = "12.1.105"
+version = "12.8.90"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/da/d3/8057f0587683ed2fcd4dbfbdfdfa807b9160b809976099d36b8f60d08f03/nvidia_nvtx_cu12-12.1.105-py3-none-manylinux1_x86_64.whl", hash = "sha256:dc21cf308ca5691e7c04d962e213f8a4aa9bbfa23d95412f452254c2caeb09e5", size = 99138 },
+    { url = "https://files.pythonhosted.org/packages/a2/eb/86626c1bbc2edb86323022371c39aa48df6fd8b0a1647bc274577f72e90b/nvidia_nvtx_cu12-12.8.90-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5b17e2001cc0d751a5bc2c6ec6d26ad95913324a4adb86788c944f8ce9ba441f", size = 89954, upload-time = "2025-03-07T01:42:44.131Z" },
+]
+
+[[package]]
+name = "ollama"
+version = "0.5.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "httpx" },
+    { name = "pydantic" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/91/6d/ae96027416dcc2e98c944c050c492789502d7d7c0b95a740f0bb39268632/ollama-0.5.3.tar.gz", hash = "sha256:40b6dff729df3b24e56d4042fd9d37e231cee8e528677e0d085413a1d6692394", size = 43331, upload-time = "2025-08-07T21:44:10.422Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/be/f6/2091e50b8b6c3e6901f6eab283d5efd66fb71c86ddb1b4d68766c3eeba0f/ollama-0.5.3-py3-none-any.whl", hash = "sha256:a8303b413d99a9043dbf77ebf11ced672396b59bec27e6d5db67c88f01b279d2", size = 13490, upload-time = "2025-08-07T21:44:09.353Z" },
 ]
 
 [[package]]
 name = "openai"
-version = "1.59.8"
+version = "1.102.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
@@ -1855,359 +2414,608 @@ dependencies = [
     { name = "tqdm" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/e0/c4/b4482784de63c7158f6c0afcb07fd66450ea6c912d6bddf9d7599f2eda25/openai-1.59.8.tar.gz", hash = "sha256:ac4bda5fa9819fdc6127e8ea8a63501f425c587244bc653c7c11a8ad84f953e1", size = 346775 }
+sdist = { url = "https://files.pythonhosted.org/packages/07/55/da5598ed5c6bdd9939633854049cddc5cbac0da938dfcfcb3c6b119c16c0/openai-1.102.0.tar.gz", hash = "sha256:2e0153bcd64a6523071e90211cbfca1f2bbc5ceedd0993ba932a5869f93b7fc9", size = 519027, upload-time = "2025-08-26T20:50:29.397Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/8c/cf/5b235e12ead3cd2098f9792776c966994c1bc558cba5799e12f3045227df/openai-1.59.8-py3-none-any.whl", hash = "sha256:a8b8ee35c4083b88e6da45406d883cf6bd91a98ab7dd79178b8bc24c8bfb09d9", size = 455567 },
+    { url = "https://files.pythonhosted.org/packages/bd/0d/c9e7016d82c53c5b5e23e2bad36daebb8921ed44f69c0a985c6529a35106/openai-1.102.0-py3-none-any.whl", hash = "sha256:d751a7e95e222b5325306362ad02a7aa96e1fab3ed05b5888ce1c7ca63451345", size = 812015, upload-time = "2025-08-26T20:50:27.219Z" },
 ]
 
 [[package]]
 name = "opt-einsum"
 version = "3.4.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/8c/b9/2ac072041e899a52f20cf9510850ff58295003aa75525e58343591b0cbfb/opt_einsum-3.4.0.tar.gz", hash = "sha256:96ca72f1b886d148241348783498194c577fa30a8faac108586b14f1ba4473ac", size = 63004 }
+sdist = { url = "https://files.pythonhosted.org/packages/8c/b9/2ac072041e899a52f20cf9510850ff58295003aa75525e58343591b0cbfb/opt_einsum-3.4.0.tar.gz", hash = "sha256:96ca72f1b886d148241348783498194c577fa30a8faac108586b14f1ba4473ac", size = 63004, upload-time = "2024-09-26T14:33:24.483Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/23/cd/066e86230ae37ed0be70aae89aabf03ca8d9f39c8aea0dec8029455b5540/opt_einsum-3.4.0-py3-none-any.whl", hash = "sha256:69bb92469f86a1565195ece4ac0323943e83477171b91d24c35afe028a90d7cd", size = 71932, upload-time = "2024-09-26T14:33:23.039Z" },
+]
+
+[[package]]
+name = "optax"
+version = "0.2.4"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "absl-py", marker = "python_full_version < '3.10'" },
+    { name = "chex", marker = "python_full_version < '3.10'" },
+    { name = "etils", version = "1.5.2", source = { registry = "https://pypi.org/simple" }, extra = ["epy"], marker = "python_full_version < '3.10'" },
+    { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/af/b5/f88a0d851547b2e6b2c7e7e6509ad66236b3e7019f1f095bb03dbaa61fa1/optax-0.2.4.tar.gz", hash = "sha256:4e05d3d5307e6dde4c319187ae36e6cd3a0c035d4ed25e9e992449a304f47336", size = 229717, upload-time = "2024-11-12T21:52:17.439Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/5c/24/28d0bb21600a78e46754947333ec9a297044af884d360092eb8561575fe9/optax-0.2.4-py3-none-any.whl", hash = "sha256:db35c04e50b52596662efb002334de08c2a0a74971e4da33f467e84fac08886a", size = 319212, upload-time = "2024-11-12T21:52:15.446Z" },
+]
+
+[[package]]
+name = "optax"
+version = "0.2.5"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "absl-py", marker = "python_full_version >= '3.10'" },
+    { name = "chex", marker = "python_full_version >= '3.10'" },
+    { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/c0/75/1e011953c48be502d4d84fa8458e91be7c6f983002511669bddd7b1a065f/optax-0.2.5.tar.gz", hash = "sha256:b2e38c7aea376186deae758ba7a258e6ef760c6f6131e9e11bc561c65386d594", size = 258548, upload-time = "2025-06-10T17:00:47.544Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b9/33/f86091c706db1a5459f501830241afff2ecab3532725c188ea57be6e54de/optax-0.2.5-py3-none-any.whl", hash = "sha256:966deae936207f268ac8f564d8ed228d645ac1aaddefbbf194096d2299b24ba8", size = 354324, upload-time = "2025-06-10T17:00:46.062Z" },
+]
+
+[[package]]
+name = "optree"
+version = "0.17.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/56/c7/0853e0c59b135dff770615d2713b547b6b3b5cde7c10995b4a5825244612/optree-0.17.0.tar.gz", hash = "sha256:5335a5ec44479920620d72324c66563bd705ab2a698605dd4b6ee67dbcad7ecd", size = 163111, upload-time = "2025-07-25T11:26:11.586Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/52/a0/d5795ac13390b04822f1c61699f684cde682b57bf0a2d6b406019e1762ae/optree-0.17.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:85ec183b8eec6efc9a5572c2a84c62214c949555efbc69ca2381aca6048d08df", size = 622371, upload-time = "2025-07-25T11:24:23.345Z" },
+    { url = "https://files.pythonhosted.org/packages/53/8b/ae8ddb511e680eb9d61edd2f5245be88ce050456658fb165550144f9a509/optree-0.17.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6e77b6e0b7bb3ecfeb9a92ba605ef21b39bff38829b745af993e2e2b474322e2", size = 337260, upload-time = "2025-07-25T11:24:25.291Z" },
+    { url = "https://files.pythonhosted.org/packages/91/f9/6ca076fd4c6f16be031afdc711a2676c1ff15bd1717ee2e699179b1a29bc/optree-0.17.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:98990201f352dba253af1a995c1453818db5f08de4cae7355d85aa6023676a52", size = 350398, upload-time = "2025-07-25T11:24:26.672Z" },
+    { url = "https://files.pythonhosted.org/packages/95/4c/81344cbdcf8ea8525a21c9d65892d7529010ee2146c53423b2e9a84441ba/optree-0.17.0-cp310-cp310-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:e1a40adf6bb78a6a4b4f480879de2cb6b57d46d680a4d9834aa824f41e69c0d9", size = 404834, upload-time = "2025-07-25T11:24:28.988Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/c4/ac1880372a89f5c21514a7965dfa23b1afb2ad683fb9804d366727de9ecf/optree-0.17.0-cp310-cp310-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:78a113436a0a440f900b2799584f3cc2b2eea1b245d81c3583af42ac003e333c", size = 402116, upload-time = "2025-07-25T11:24:30.396Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/72/ad6be4d6a03805cf3921b492494cb3371ca28060d5ad19d5a36e10c4d67d/optree-0.17.0-cp310-cp310-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0e45c16018f4283f028cf839b707b7ac734e8056a31b7198a1577161fcbe146d", size = 398491, upload-time = "2025-07-25T11:24:31.725Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/c1/6827fb504351f9a3935699b0eb31c8a6af59d775ee78289a25e0ba54f732/optree-0.17.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b698613d821d80cc216a2444ebc3145c8bf671b55a2223058a6574c1483a65f6", size = 387957, upload-time = "2025-07-25T11:24:32.759Z" },
+    { url = "https://files.pythonhosted.org/packages/21/3d/44b3cbe4c9245a13b2677e30db2aafadf00bda976a551d64a31dc92f4977/optree-0.17.0-cp310-cp310-win32.whl", hash = "sha256:d07bfd8ce803dbc005502a89fda5f5e078e237342eaa36fb0c46cfbdf750bc76", size = 280064, upload-time = "2025-07-25T11:24:33.875Z" },
+    { url = "https://files.pythonhosted.org/packages/74/fa/83d4cd387043483ee23617b048829a1289bf54afe2f6cb98ec7b27133369/optree-0.17.0-cp310-cp310-win_amd64.whl", hash = "sha256:d009d368ef06b8757891b772cad24d4f84122bd1877f7674fb8227d6e15340b4", size = 304398, upload-time = "2025-07-25T11:24:34.844Z" },
+    { url = "https://files.pythonhosted.org/packages/21/4f/752522f318683efa7bba1895667c9841165d0284f6dfadf601769f6398ce/optree-0.17.0-cp310-cp310-win_arm64.whl", hash = "sha256:3571085ed9a5f39ff78ef57def0e9607c6b3f0099b6910524a0b42f5d58e481e", size = 308260, upload-time = "2025-07-25T11:24:36.144Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/eb/389a7dae8b113064f53909707aea9d72372fdc2eb918c48783c443cb3438/optree-0.17.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:09fbc0e5e42b20cab11851dffb7abe2fdf289c45d29e5be2b50b4ea93d069a9f", size = 640773, upload-time = "2025-07-25T11:24:37.25Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/bb/2d78b524989cabb5720e85ea366addc8589b4bbd0ce3f5ea58e370e5636a/optree-0.17.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:90a5864689268eda75d90abded5d474ae0a7ae2608d510626724fb78a1955948", size = 346402, upload-time = "2025-07-25T11:24:38.25Z" },
+    { url = "https://files.pythonhosted.org/packages/73/5c/13a2a864b0c0b39c3c193be534a195a3ab2463c7d0443d4a76e749e3ff83/optree-0.17.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3080c564c9760711aa72d1b4d700ce1417f99ad087136f415c4eb8221169e2a3", size = 362797, upload-time = "2025-07-25T11:24:39.509Z" },
+    { url = "https://files.pythonhosted.org/packages/da/f5/ff7dcb5a0108ee89c2be09aed2ebd26a7e1333d8122031aa9d9322b24ee6/optree-0.17.0-cp311-cp311-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:834a8fb358b608240b3a38706a09b43974675624485fad64c8ee641dae2eb57d", size = 419450, upload-time = "2025-07-25T11:24:40.555Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/e6/48a97aefd18770b55e5ed456d8183891f325cdb6d90592e5f072ed6951f8/optree-0.17.0-cp311-cp311-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:1a2bd263e6b5621d000d0f94de1f245414fd5dbce365a24b7b89b1ed0ef56cf9", size = 417557, upload-time = "2025-07-25T11:24:42.396Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/b1/4e280edab8a86be47ec1f9bd9ed4b685d2e15f0950ae62b613b26d12a1da/optree-0.17.0-cp311-cp311-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:9b37daca4ad89339b1f5320cc61ac600dcf976adbb060769d36d5542d6ebfedf", size = 414174, upload-time = "2025-07-25T11:24:43.51Z" },
+    { url = "https://files.pythonhosted.org/packages/db/3b/49a9a1986215dd342525974deeb17c260a83fee8fad147276fd710ac8718/optree-0.17.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a146a6917f3e28cfdc268ff1770aa696c346482dd3da681c3ff92153d94450ea", size = 402000, upload-time = "2025-07-25T11:24:44.819Z" },
+    { url = "https://files.pythonhosted.org/packages/00/8d/13b79d3394b83f4b1c93daac336f0eca5cb1cd5f58e10618f2c2db779cb7/optree-0.17.0-cp311-cp311-win32.whl", hash = "sha256:6b0446803d08f6aaae84f82f03c51527f36dfa15850873fc0183792247bc0071", size = 285777, upload-time = "2025-07-25T11:24:45.976Z" },
+    { url = "https://files.pythonhosted.org/packages/90/32/da5191a347e33a78c2804a0cbfaed8eecb758818efda4b4d70bfd9b9b38d/optree-0.17.0-cp311-cp311-win_amd64.whl", hash = "sha256:e39f4f00b2967116badd9617ad6aa9845d8327fe13b6dbf5bc36d8c7b4a5ea03", size = 313761, upload-time = "2025-07-25T11:24:47.047Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/ea/7cae17a37a8ef67a33c354fce6f136d5f253d5afa40f68701252b1b2c2a0/optree-0.17.0-cp311-cp311-win_arm64.whl", hash = "sha256:50d4dbcbca3e379cc6b374f9b5a5626ff7ea41df8373e26c3af41d89d8a4b3d5", size = 318242, upload-time = "2025-07-25T11:24:48.708Z" },
+    { url = "https://files.pythonhosted.org/packages/79/ce/471ff57336630f2434238a8cb8401e0d714ee7d54a6117823fd85de5f656/optree-0.17.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:09156e2ea62cde66dcbd9a450a5517ad6bad07d4ffc98fab0982c1e4f538341a", size = 654627, upload-time = "2025-07-25T11:24:49.754Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/ef/3143b7840dd2daedf1257643119c0f3addd23cf90cc9d2efc88f8166931e/optree-0.17.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:750f24304d1d437c8b235d4bc9e4afda17d85950706c34a875c16049f707eeb4", size = 351124, upload-time = "2025-07-25T11:24:50.813Z" },
+    { url = "https://files.pythonhosted.org/packages/41/90/e12dea2cb5d8a5e17bbe3011ed4e972b89c027272a816db4897589751cad/optree-0.17.0-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e13ae51a63d69db445f269a3a4fd1d6edb064a705188d007ea47c9f034788fc5", size = 365869, upload-time = "2025-07-25T11:24:51.807Z" },
+    { url = "https://files.pythonhosted.org/packages/76/ee/21af214663960a479863cd6c03d7a0abc8123ea22a6ea34689c2eed88ccd/optree-0.17.0-cp312-cp312-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:5958f58423cc7870cb011c8c8f92687397380886e8c9d33adac752147e7bbc3f", size = 424465, upload-time = "2025-07-25T11:24:53.124Z" },
+    { url = "https://files.pythonhosted.org/packages/54/a3/64b184a79373753f4f46a5cd301ea581f71d6dc1a5c103bd2394f0925d40/optree-0.17.0-cp312-cp312-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:970ae4e47727b4c5526fc583b87d29190e576f6a2b6c19e8671589b73d256250", size = 420686, upload-time = "2025-07-25T11:24:54.212Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/6d/b6051b0b1ef9a49df96a66e9e62fc02620d2115d1ba659888c94e67fcfc9/optree-0.17.0-cp312-cp312-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:54177fd3e6e05c08b66329e26d7d44b85f24125f25c6b74c921499a1b31b8f70", size = 421225, upload-time = "2025-07-25T11:24:55.213Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/f1/940bc959aaef9eede8bb1b1127833b0929c6ffa9268ec0f6cb19877e2027/optree-0.17.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e1959cfbc38c228c8195354967cda64887b96219924b7b3759e5ee355582c1ec", size = 408819, upload-time = "2025-07-25T11:24:56.315Z" },
+    { url = "https://files.pythonhosted.org/packages/56/52/ce527556e27dbf77266c1b1bb313ca446c94bc6edd6d7a882dbded028197/optree-0.17.0-cp312-cp312-win32.whl", hash = "sha256:039ea98c0cd94a64040d6f6d21dbe5cd9731bb380d7893f78d6898672080a232", size = 289107, upload-time = "2025-07-25T11:24:57.357Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/f1/aecb0199d269ad8ea41a86182474f98378a72681facbd6a06e94c23a2d02/optree-0.17.0-cp312-cp312-win_amd64.whl", hash = "sha256:c3a21109f635ce353d116ed1d77a7dfd77b898bcdaccef3bf74881ce7d6d54d8", size = 314074, upload-time = "2025-07-25T11:24:58.499Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/20/615ad64d24318709a236163dd8620fa7879a7720bfd0c755604d3dceeb76/optree-0.17.0-cp312-cp312-win_arm64.whl", hash = "sha256:1a39f957299426d2d4aa36cbc1acd71edb198ff0f28ddb43029bf58efe34a9a1", size = 316409, upload-time = "2025-07-25T11:24:59.855Z" },
+    { url = "https://files.pythonhosted.org/packages/74/b7/6efb4955f7f08425d67b38ffd001a28896f670109bdbf1b1745f632a2306/optree-0.17.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:1535fb8725178715315af0f2862668fb49030a5737d9f6c68bcb4747b029b20b", size = 622899, upload-time = "2025-07-25T11:25:49.771Z" },
+    { url = "https://files.pythonhosted.org/packages/df/41/0d6e4d066aa25c59170055a4236d4b52abc6ac09bd0821bb85a58db7f0e0/optree-0.17.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:4aec2d138baed1357ca1ded81e40140bafbfdfd09b73d3d9d96c6c3cc527bcd9", size = 337376, upload-time = "2025-07-25T11:25:50.864Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/29/3bb53de2de3b36a51e46b6d9ada7ee1a3a312ac461cd54292a023adc807c/optree-0.17.0-cp39-cp39-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:537498cf7bf7a4fe71f7ffd815e72b8672aea0fac82e1513f6b6e35e8569f5aa", size = 350302, upload-time = "2025-07-25T11:25:52.016Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/3b/d17a31447ed7ef6f10bd0caf40742b016fcdeaa3abb7568307b04a0f50cf/optree-0.17.0-cp39-cp39-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:3b3bb2326b550ddb048e3454fad40183b7fed74dda4351b016d20362809180af", size = 405358, upload-time = "2025-07-25T11:25:53.085Z" },
+    { url = "https://files.pythonhosted.org/packages/db/f3/b9f0a8c98fd0c7f53fa9d9a46d75bb1182aeecd7ecde6f353d3e69ec9618/optree-0.17.0-cp39-cp39-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c0d3d702044e5acbec2cf8349789f6b096057bd00dc8e1e1c97b990347279fda", size = 402694, upload-time = "2025-07-25T11:25:54.537Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/dd/0d9d7426fd6b5d90ad40e4d93717a955d4257d06574dfe7a1da0d24cb06c/optree-0.17.0-cp39-cp39-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a9155e82717be1dda1f3c1244e9cb5b3733d5dd3ba47702730c7816be083a5cb", size = 398857, upload-time = "2025-07-25T11:25:55.921Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/57/dacec3f8c70f4685bb07fce19cf3361037fde2b596f6f7228e1a4b39677b/optree-0.17.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a8e825501f55360e8381718623b094579dedc485e57010e01593d72a43b43e68", size = 387849, upload-time = "2025-07-25T11:25:57.046Z" },
+    { url = "https://files.pythonhosted.org/packages/71/66/659501896bf169ad5044bec00ac9dd4fe7144bd5448c5dae06977b596166/optree-0.17.0-cp39-cp39-win32.whl", hash = "sha256:80865cf4287ed86e65af9bacd98d5395f424ffc08dc0d784590763fc1a1576b9", size = 280368, upload-time = "2025-07-25T11:25:58.209Z" },
+    { url = "https://files.pythonhosted.org/packages/56/d4/0ff0b8f762fd79249df3ed48f82eb3e6144a219c02278e39e135c582224b/optree-0.17.0-cp39-cp39-win_amd64.whl", hash = "sha256:c361ee45a97d69a427d949db5f0d6a8d9ad5f703ac7cef57a206f7f3df13d6f9", size = 310964, upload-time = "2025-07-25T11:25:59.597Z" },
+    { url = "https://files.pythonhosted.org/packages/16/62/3720d2ac320c327f304c4ed50cf6aa6059fc4f85172c69616b6ddc3b7e7d/optree-0.17.0-cp39-cp39-win_arm64.whl", hash = "sha256:4ad585248f82896ac85681b9f36b33a791d4ebf8588f3126b4dbbe5c31edbefa", size = 306151, upload-time = "2025-07-25T11:26:00.632Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/52/350c58dce327257afd77b92258e43d0bfe00416fc167b0c256ec86dcf9e7/optree-0.17.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:f365328450c1072e7a707dce67eaa6db3f63671907c866e3751e317b27ea187e", size = 342845, upload-time = "2025-07-25T11:26:01.651Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/d7/3036d15c028c447b1bd65dcf8f66cfd775bfa4e52daa74b82fb1d3c88faf/optree-0.17.0-pp310-pypy310_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:adde1427e0982cfc5f56939c26b4ebbd833091a176734c79fb95c78bdf833dff", size = 350952, upload-time = "2025-07-25T11:26:02.692Z" },
+    { url = "https://files.pythonhosted.org/packages/71/45/e710024ef77324e745de48efd64f6270d8c209f14107a48ffef4049ac57a/optree-0.17.0-pp310-pypy310_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a80b7e5de5dd09b9c8b62d501e29a3850b047565c336c9d004b07ee1c01f4ae1", size = 389568, upload-time = "2025-07-25T11:26:04.094Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/63/b5cd1309f76f53e8a3cfbc88642647e58b1d3dd39f7cb0daf60ec516a252/optree-0.17.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:3c2c79652c45d82f23cbe08349456b1067ea513234a086b9a6bf1bcf128962a9", size = 306686, upload-time = "2025-07-25T11:26:05.511Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/40/afec131d9dd7a18d129190d407d97c95994f42b70c3d8ab897092d4de1d9/optree-0.17.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:bd92011cd0f2de40d28a95842819e778c476ab25c12731bfef1d1a0225554f83", size = 353955, upload-time = "2025-07-25T11:26:06.75Z" },
+    { url = "https://files.pythonhosted.org/packages/69/c4/94a187ed3ca71194b9da6a276790e1703c7544c8f695ac915214ae8ce934/optree-0.17.0-pp311-pypy311_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f87f6f39015fc82d7adeee19900d246b89911319726e93cb2dbd4d1a809899bd", size = 363728, upload-time = "2025-07-25T11:26:07.959Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/99/23b7a484da8dfb814107b20ef2c93ef27c04f36aeb83bd976964a5b69e06/optree-0.17.0-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:58b0a83a967d2ef0f343db7182f0ad074eb1166bcaea909ae33909462013f151", size = 404649, upload-time = "2025-07-25T11:26:09.463Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/1f/7eca6da47eadb9ff2183bc9169eadde3dda0518e9a0187b99d5926fb2994/optree-0.17.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:e1ae8cbbcfaa45c57f5e51c544afa554cefbbb9fe9586c108aaf2aebfadf5899", size = 316368, upload-time = "2025-07-25T11:26:10.572Z" },
+]
+
+[[package]]
+name = "orbax-checkpoint"
+version = "0.6.4"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "absl-py", marker = "python_full_version < '3.10'" },
+    { name = "etils", version = "1.5.2", source = { registry = "https://pypi.org/simple" }, extra = ["epath", "epy"], marker = "python_full_version < '3.10'" },
+    { name = "humanize", marker = "python_full_version < '3.10'" },
+    { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "msgpack", marker = "python_full_version < '3.10'" },
+    { name = "nest-asyncio", marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "protobuf", marker = "python_full_version < '3.10'" },
+    { name = "pyyaml", marker = "python_full_version < '3.10'" },
+    { name = "tensorstore", version = "0.1.69", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/32/9f/6ca67c33b6a53d2484ebeb3c132a06331fbf085ce1619e3f919d36bb6a9c/orbax_checkpoint-0.6.4.tar.gz", hash = "sha256:366b4d528a7322e1b3d9ddcaed45c8515add0d2fc69c8975c30d98638543240f", size = 194820, upload-time = "2024-09-17T17:37:10.151Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8d/79/0a84020557a5c5bd987f74892e58e71cf74886dc9b264ad7630a1f288543/orbax_checkpoint-0.6.4-py3-none-any.whl", hash = "sha256:b4f2608ee4d1da67f7619fc35ff9c928ecdf4ccf7546eeb43ecf38c2608b6dea", size = 270502, upload-time = "2024-09-17T17:37:07.477Z" },
+]
+
+[[package]]
+name = "orbax-checkpoint"
+version = "0.11.23"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "absl-py", marker = "python_full_version >= '3.10'" },
+    { name = "aiofiles", marker = "python_full_version >= '3.10'" },
+    { name = "etils", version = "1.13.0", source = { registry = "https://pypi.org/simple" }, extra = ["epath", "epy"], marker = "python_full_version >= '3.10'" },
+    { name = "humanize", marker = "python_full_version >= '3.10'" },
+    { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "msgpack", marker = "python_full_version >= '3.10'" },
+    { name = "nest-asyncio", marker = "python_full_version >= '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "protobuf", marker = "python_full_version >= '3.10'" },
+    { name = "pyyaml", marker = "python_full_version >= '3.10'" },
+    { name = "simplejson", marker = "python_full_version >= '3.10'" },
+    { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/25/07/f533bb7a47de27c8666b00ebaeba597b1a725d6734c6f0e22d2e074a5d47/orbax_checkpoint-0.11.23.tar.gz", hash = "sha256:f59491f7c8c671cf255df7f467389e5317baba637ff9d57b2b4af83fe8e8526f", size = 366861, upload-time = "2025-08-18T20:13:19.555Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/23/cd/066e86230ae37ed0be70aae89aabf03ca8d9f39c8aea0dec8029455b5540/opt_einsum-3.4.0-py3-none-any.whl", hash = "sha256:69bb92469f86a1565195ece4ac0323943e83477171b91d24c35afe028a90d7cd", size = 71932 },
+    { url = "https://files.pythonhosted.org/packages/0e/fb/1f909c74aca659eb001f7416cc8a856b633d23ba84b2a88aa0890c14983b/orbax_checkpoint-0.11.23-py3-none-any.whl", hash = "sha256:db91516f574428cd6734042bbf380cf095fc297cefe00ec55601c68a3ecb60b5", size = 523142, upload-time = "2025-08-18T20:13:18.062Z" },
 ]
 
 [[package]]
 name = "outlines"
 source = { editable = "." }
 dependencies = [
-    { name = "airportsdata" },
     { name = "cloudpickle" },
     { name = "diskcache" },
     { name = "genson" },
-    { name = "interegular" },
     { name = "jinja2" },
+    { name = "jsonpath-ng" },
     { name = "jsonschema" },
-    { name = "lark" },
-    { name = "nest-asyncio" },
-    { name = "numpy" },
     { name = "outlines-core" },
-    { name = "pre-commit" },
-    { name = "pycountry" },
+    { name = "pillow" },
     { name = "pydantic" },
-    { name = "referencing" },
-    { name = "requests" },
-    { name = "torch" },
-    { name = "tqdm" },
     { name = "typing-extensions" },
 ]
 
 [package.optional-dependencies]
+airports = [
+    { name = "airportsdata" },
+]
+anthropic = [
+    { name = "anthropic" },
+]
+countries = [
+    { name = "iso3166" },
+]
+dottxt = [
+    { name = "dottxt" },
+]
+gemini = [
+    { name = "google-genai" },
+]
 llamacpp = [
-    { name = "datasets" },
+    { name = "huggingface-hub" },
     { name = "llama-cpp-python" },
-    { name = "numpy" },
-    { name = "transformers" },
+    { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+llguidance = [
+    { name = "llguidance" },
 ]
 mlxlm = [
     { name = "datasets" },
+    { name = "mlx" },
     { name = "mlx-lm" },
 ]
+ollama = [
+    { name = "ollama" },
+]
 openai = [
     { name = "openai" },
 ]
-serve = [
-    { name = "fastapi" },
-    { name = "pydantic" },
-    { name = "uvicorn" },
-    { name = "vllm" },
+sglang = [
+    { name = "openai" },
+]
+test = [
+    { name = "accelerate" },
+    { name = "airportsdata" },
+    { name = "anthropic" },
+    { name = "beartype" },
+    { name = "coverage", extra = ["toml"] },
+    { name = "datasets" },
+    { name = "diff-cover" },
+    { name = "dottxt" },
+    { name = "flax", version = "0.8.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "flax", version = "0.10.7", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "flax", version = "0.11.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "google-genai" },
+    { name = "huggingface-hub" },
+    { name = "iso3166" },
+    { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "llama-cpp-python" },
+    { name = "llguidance" },
+    { name = "mkdocs-gen-files" },
+    { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
+    { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "ollama" },
+    { name = "openai" },
+    { name = "pillow" },
+    { name = "pre-commit" },
+    { name = "pytest" },
+    { name = "pytest-asyncio" },
+    { name = "pytest-benchmark" },
+    { name = "pytest-cov" },
+    { name = "pytest-mock" },
+    { name = "requests" },
+    { name = "responses" },
+    { name = "sentencepiece" },
+    { name = "tensorflow" },
+    { name = "tf-keras" },
+    { name = "torch" },
+    { name = "transformers" },
+    { name = "xgrammar" },
 ]
-test = [
+test-gpu = [
     { name = "accelerate" },
+    { name = "airportsdata" },
+    { name = "anthropic" },
     { name = "beartype" },
     { name = "coverage", extra = ["toml"] },
     { name = "datasets" },
-    { name = "diff-cover", version = "9.2.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.9.17'" },
-    { name = "diff-cover", version = "9.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.9.17'" },
+    { name = "diff-cover" },
+    { name = "dottxt" },
+    { name = "flax", version = "0.8.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "flax", version = "0.10.7", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "flax", version = "0.11.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "google-genai" },
     { name = "huggingface-hub" },
+    { name = "iso3166" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "jax", version = "0.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "llama-cpp-python" },
+    { name = "llguidance" },
+    { name = "mkdocs-gen-files" },
     { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
+    { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "ollama" },
     { name = "openai" },
     { name = "pillow" },
     { name = "pre-commit" },
     { name = "pytest" },
+    { name = "pytest-asyncio" },
     { name = "pytest-benchmark" },
     { name = "pytest-cov" },
     { name = "pytest-mock" },
+    { name = "requests" },
     { name = "responses" },
+    { name = "sentencepiece" },
+    { name = "tensorflow" },
+    { name = "tf-keras" },
+    { name = "torch" },
     { name = "transformers" },
     { name = "vllm", marker = "sys_platform == 'linux'" },
+    { name = "xgrammar" },
+]
+tgi = [
+    { name = "huggingface-hub" },
 ]
 transformers = [
     { name = "accelerate" },
     { name = "datasets" },
-    { name = "numpy" },
     { name = "transformers" },
 ]
 vllm = [
-    { name = "numpy" },
-    { name = "transformers" },
-    { name = "vllm" },
+    { name = "openai" },
+]
+xgrammar = [
+    { name = "xgrammar" },
 ]
 
 [package.metadata]
 requires-dist = [
     { name = "accelerate", marker = "extra == 'test'" },
     { name = "accelerate", marker = "extra == 'transformers'" },
-    { name = "airportsdata" },
+    { name = "airportsdata", marker = "extra == 'airports'" },
+    { name = "airportsdata", marker = "extra == 'test'" },
+    { name = "anthropic", marker = "extra == 'anthropic'" },
+    { name = "anthropic", marker = "extra == 'test'" },
     { name = "beartype", marker = "extra == 'test'", specifier = "<0.16.0" },
     { name = "cloudpickle" },
     { name = "coverage", extras = ["toml"], marker = "extra == 'test'", specifier = ">=5.1" },
-    { name = "datasets", marker = "extra == 'llamacpp'" },
     { name = "datasets", marker = "extra == 'mlxlm'" },
     { name = "datasets", marker = "extra == 'test'" },
     { name = "datasets", marker = "extra == 'transformers'" },
     { name = "diff-cover", marker = "extra == 'test'" },
     { name = "diskcache" },
-    { name = "fastapi", marker = "extra == 'serve'" },
+    { name = "dottxt", marker = "extra == 'dottxt'" },
+    { name = "dottxt", marker = "extra == 'test'" },
+    { name = "flax", marker = "extra == 'test'" },
     { name = "genson" },
+    { name = "google-genai", marker = "extra == 'gemini'" },
+    { name = "google-genai", marker = "extra == 'test'" },
+    { name = "huggingface-hub", marker = "extra == 'llamacpp'" },
     { name = "huggingface-hub", marker = "extra == 'test'" },
-    { name = "interegular" },
+    { name = "huggingface-hub", marker = "extra == 'tgi'" },
+    { name = "iso3166", marker = "extra == 'countries'" },
+    { name = "iso3166", marker = "extra == 'test'" },
     { name = "jax", marker = "extra == 'test'" },
     { name = "jinja2" },
+    { name = "jsonpath-ng" },
     { name = "jsonschema" },
-    { name = "lark" },
     { name = "llama-cpp-python", marker = "extra == 'llamacpp'" },
     { name = "llama-cpp-python", marker = "extra == 'test'" },
+    { name = "llguidance", marker = "extra == 'llguidance'" },
+    { name = "llguidance", marker = "extra == 'test'" },
+    { name = "mkdocs-gen-files", marker = "extra == 'test'" },
+    { name = "mlx", marker = "extra == 'mlxlm'" },
     { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin' and extra == 'test'", specifier = ">=0.19.2" },
     { name = "mlx-lm", marker = "extra == 'mlxlm'" },
-    { name = "nest-asyncio" },
-    { name = "numpy" },
-    { name = "numpy", marker = "extra == 'llamacpp'", specifier = "<2" },
-    { name = "numpy", marker = "extra == 'transformers'", specifier = "<2" },
-    { name = "numpy", marker = "extra == 'vllm'", specifier = "<2" },
+    { name = "numba", marker = "extra == 'llamacpp'" },
+    { name = "numba", marker = "extra == 'test'" },
+    { name = "numpy", marker = "extra == 'test'", specifier = ">=2.0.0,<2.2.0" },
+    { name = "ollama", marker = "extra == 'ollama'" },
+    { name = "ollama", marker = "extra == 'test'" },
     { name = "openai", marker = "extra == 'openai'" },
+    { name = "openai", marker = "extra == 'sglang'" },
     { name = "openai", marker = "extra == 'test'", specifier = ">=1.0.0" },
-    { name = "outlines-core", specifier = "==0.1.26" },
+    { name = "openai", marker = "extra == 'vllm'" },
+    { name = "outlines", extras = ["test"], marker = "extra == 'test-gpu'" },
+    { name = "outlines-core", specifier = "==0.2.11" },
+    { name = "pillow" },
     { name = "pillow", marker = "extra == 'test'" },
-    { name = "pre-commit", specifier = ">=4.0.1" },
     { name = "pre-commit", marker = "extra == 'test'" },
-    { name = "pycountry" },
     { name = "pydantic", specifier = ">=2.0" },
-    { name = "pydantic", marker = "extra == 'serve'", specifier = ">=2.0" },
     { name = "pytest", marker = "extra == 'test'" },
+    { name = "pytest-asyncio", marker = "extra == 'test'" },
     { name = "pytest-benchmark", marker = "extra == 'test'" },
     { name = "pytest-cov", marker = "extra == 'test'" },
     { name = "pytest-mock", marker = "extra == 'test'" },
-    { name = "referencing" },
-    { name = "requests" },
+    { name = "requests", marker = "extra == 'test'" },
     { name = "responses", marker = "extra == 'test'" },
-    { name = "torch" },
-    { name = "tqdm" },
-    { name = "transformers", marker = "extra == 'llamacpp'" },
+    { name = "sentencepiece", marker = "extra == 'test'" },
+    { name = "tensorflow", marker = "extra == 'test'" },
+    { name = "tf-keras", marker = "extra == 'test'" },
+    { name = "torch", marker = "extra == 'test'", specifier = ">2.3.0" },
     { name = "transformers", marker = "extra == 'test'" },
     { name = "transformers", marker = "extra == 'transformers'" },
-    { name = "transformers", marker = "extra == 'vllm'" },
     { name = "typing-extensions" },
-    { name = "uvicorn", marker = "extra == 'serve'" },
-    { name = "vllm", marker = "sys_platform == 'linux' and extra == 'test'" },
-    { name = "vllm", marker = "extra == 'serve'", specifier = ">=0.3.0" },
-    { name = "vllm", marker = "extra == 'vllm'" },
+    { name = "vllm", marker = "sys_platform == 'linux' and extra == 'test-gpu'" },
+    { name = "xgrammar", marker = "extra == 'test'" },
+    { name = "xgrammar", marker = "extra == 'xgrammar'" },
 ]
+provides-extras = ["anthropic", "dottxt", "gemini", "llamacpp", "mlxlm", "ollama", "openai", "sglang", "tgi", "transformers", "vllm", "xgrammar", "llguidance", "airports", "countries", "test", "test-gpu"]
 
 [[package]]
 name = "outlines-core"
-version = "0.1.26"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "interegular" },
-    { name = "jsonschema" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/d3/f3/274d07f4702728b43581235a77e545ec602b25f9b0098b288a0f3052521d/outlines_core-0.1.26.tar.gz", hash = "sha256:481c4301341e77cc8f1832d616784adb4d461b4fec65878e7c0d2cba7163a189", size = 75139 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/e2/df/e9ff00f1dcf671cb8c4c20abcfd53406328b344cafa689a2832e8059c0b4/outlines_core-0.1.26-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:6a962a7452e7ac170fa04d405342cadae2d28fafa5b1830cef7aa610257ed32f", size = 322602 },
-    { url = "https://files.pythonhosted.org/packages/3c/f1/e9064f18c462a61f4abbe73b24f25e36d8abef19c593416fa69dce6a83c0/outlines_core-0.1.26-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:15a3684fa29564da2db03934cf0097bef3e871f70d3af0ef2b52fdb886da2e09", size = 301929 },
-    { url = "https://files.pythonhosted.org/packages/76/c3/6bc82db40b4818421e573237f43d4026c40a3305fa2558eb0aa1a7aa08f7/outlines_core-0.1.26-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:64e01c0cfa9ba371634d7c3f6ea1862397cef98e4509fe98e3f57faa721a72d6", size = 321355 },
-    { url = "https://files.pythonhosted.org/packages/c9/c2/1d85bfeaee3a83327e0d162bee4bdc7d7889bea5998e44fcc66c924dc1fd/outlines_core-0.1.26-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a3c4196148e47f455f1ace78e329d5b97e531cbc406456d681592952adae7e17", size = 343552 },
-    { url = "https://files.pythonhosted.org/packages/45/da/1e61d3d997ba1858fb8e71c3127f24a95c30575559da012ea5b45b147ad3/outlines_core-0.1.26-cp310-cp310-win32.whl", hash = "sha256:f38d290a7f6e5e12cbfcaee03269dfc0dbda49b360024b4279d1aba251fdc346", size = 234750 },
-    { url = "https://files.pythonhosted.org/packages/1c/04/6d7968019a81df235ad6bc7405eefe32be8da4c4153792655e7490d06c8d/outlines_core-0.1.26-cp310-cp310-win_amd64.whl", hash = "sha256:11ff56af56cb54c563b7f25d86cd9ee77f3fed825f1d4dccd9449bb1e4e89538", size = 243713 },
-    { url = "https://files.pythonhosted.org/packages/17/94/19d5c50c303ba71f3465c81620ca9b5af4db07fd8922dfe59ae5a9ae61d1/outlines_core-0.1.26-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:b6787b07b7c673fc3087d2b537719ecac8e03b10a47d032dd1926985c32885b0", size = 322344 },
-    { url = "https://files.pythonhosted.org/packages/f2/ea/f44beea7f610f2737ebb908c8dfa37d8324e92ca529468a56b00a77af199/outlines_core-0.1.26-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1e0ea28a76da31d25b6f53242bf13e1b59a0241badf82353c88f55e1cf81b128", size = 301670 },
-    { url = "https://files.pythonhosted.org/packages/6a/a6/ceac3760e1feb898b4047aeb54e0a3de975b59e87a17d6ba0a04dec5eaed/outlines_core-0.1.26-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a8932044a3d9329be53a226118850638f85b4d7842f9b863d0a123f23de220cd", size = 321067 },
-    { url = "https://files.pythonhosted.org/packages/92/f0/ad0074d6726fed86bb0bba1b9307cbbd67a2af5debd3540d66c69298a001/outlines_core-0.1.26-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a84b7cd2fb6268bf990dd3d479ffb4fa0bace6f571cb85b15b6cdb44b84f5b69", size = 343264 },
-    { url = "https://files.pythonhosted.org/packages/e6/bd/198c9a73d5f36e2ecad558a26359af3f0dbe4f5ba11c4629e46fccdfe2d6/outlines_core-0.1.26-cp311-cp311-win32.whl", hash = "sha256:f19765c151abfc970996368080aeea6d2a19e927817fe4e2af6726e639be3de4", size = 234529 },
-    { url = "https://files.pythonhosted.org/packages/b9/27/354b484045e6368c92f688d954124064ec2ce961681e56711852904e1ec2/outlines_core-0.1.26-cp311-cp311-win_amd64.whl", hash = "sha256:3f59aeccea21ed6ff3cf52102fd163f26d279821c20e5127ddd18d4ea4d0c8d2", size = 243457 },
-    { url = "https://files.pythonhosted.org/packages/c6/86/0fb40746e579db38d89f127122a3900d9e0350f76aae8cb61adeaff44cc2/outlines_core-0.1.26-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f54633bca50055d42ea4d94ae06dcbe52d3d76a9b621b75723b1177d0d952953", size = 321874 },
-    { url = "https://files.pythonhosted.org/packages/ab/0c/b91f7bc03843796c1d643ee030b6cd8fd5a8ba2cd4856c855f140c878976/outlines_core-0.1.26-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:9525321b48700dcaaabf60bcdc951e45f9357ba3fb3e1bfc81b662d7d4170e7c", size = 301995 },
-    { url = "https://files.pythonhosted.org/packages/ad/db/fa91a2d54288b900de82d86eda3adb2417b3b5b2db6256854a5e8bc85c32/outlines_core-0.1.26-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:00f409f72c11f6ffadb57066950dd384d5388015028c1a1a615c9a64988dae3e", size = 321050 },
-    { url = "https://files.pythonhosted.org/packages/e2/1d/a36292b6198986bd9c3ff8c24355deb82ed5475403379ee40b5b5473e2e3/outlines_core-0.1.26-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e86a1bb46adc5cbf6dfd7a7fe4105e0e2a4c6e041732a053126b41c521a1f223", size = 343201 },
-    { url = "https://files.pythonhosted.org/packages/08/63/5dd2b5a364412f674b6edcb59b0c21513bdb07cdcc7613b064c1a0660d01/outlines_core-0.1.26-cp312-cp312-win32.whl", hash = "sha256:19f462f6b00935708677ad27cb4df55e0e17f6ffe713ab750f5f2683b090f95d", size = 233970 },
-    { url = "https://files.pythonhosted.org/packages/a5/56/8adf0b7446d1e975c2314454813c59eb7b195889908a2932ed34148c113c/outlines_core-0.1.26-cp312-cp312-win_amd64.whl", hash = "sha256:9b36bff12779e58883747116893a17b3551bbd10865878b951b03a44d112229a", size = 243578 },
-    { url = "https://files.pythonhosted.org/packages/5b/b8/d38f13e417f2af82f6cd588df5a26ff8ffeb9a3a15b8a1a883d5922048ac/outlines_core-0.1.26-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:7b7849cf40028319ebb9d8ba0fe4c590ef5888eebe524a81b3af30aaa06ea21c", size = 323047 },
-    { url = "https://files.pythonhosted.org/packages/3b/59/43f5c7b8ad0e98a663db8f24cbff0421a58e946934e3f3751a17a06ae84d/outlines_core-0.1.26-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2f8641aab4a6bd84516907492ce82099503129da01b3c29c1dc9ad50320bae77", size = 302547 },
-    { url = "https://files.pythonhosted.org/packages/23/30/d9eb8202cbc994536a1fefa60e87c582a714ad1b46367c43af299f3889ba/outlines_core-0.1.26-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bba56604efdbc5932c7a8a88c2b8b0d0c740ab883b0012fb5464a9736796802b", size = 321817 },
-    { url = "https://files.pythonhosted.org/packages/80/c7/bd718f6329f5c65b936edff532e40614b66032b153ab9ac89e6cb6d2fcff/outlines_core-0.1.26-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8cc8c87d89bd267356f8149c9066cbb98970425ec162997fbf195c3f1feb7009", size = 344102 },
-    { url = "https://files.pythonhosted.org/packages/cb/3e/69ca7394fc89c98c763f2d475c08206fa56f2ffeec210833c1db0195ba51/outlines_core-0.1.26-cp39-cp39-win32.whl", hash = "sha256:9d792a43ed9d8a4e1b38f4d83fe99db442d57aad4404c2edf98b710892eda47e", size = 235081 },
-    { url = "https://files.pythonhosted.org/packages/56/7f/38decd8d63f6b2d513c372186c68337455efba3fb45d0ff722ac268f05d2/outlines_core-0.1.26-cp39-cp39-win_amd64.whl", hash = "sha256:ad8564ecd7b64bcb840596c5049ff1c1a96346de494302ffcc0f2b188c15675e", size = 244007 },
+version = "0.2.11"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/1a/d3/e04e9145f8f806723dec9b9e5227ad695a3efcd3ced7794cf7c22b15df5e/outlines_core-0.2.11.tar.gz", hash = "sha256:dfce56f717ff5083e54cbcfdb66cad243365437fccbb5509adaa7e31e030f1d8", size = 197263, upload-time = "2025-05-19T10:12:51.719Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cf/8f/83c83e2afd142067c7f3cf2e152809195eee72d6a9b6c8745f13b827273d/outlines_core-0.2.11-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:89d79d8454b321f60047541a896d410ca9db631d241960266c4fe839cf5cd1b1", size = 1961650, upload-time = "2025-05-19T10:11:53.12Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/e9/c6b99b4364b7026b71badc06b9809a2fc4154d6b0c475bc03ab4471f81e5/outlines_core-0.2.11-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:44d581893f8644da02db7be11887229a40d26077cbdd22072ad1ed1db0ad0b2d", size = 2133920, upload-time = "2025-05-19T10:11:55.15Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/b8/cfa2bd8e1260eb1870c42a1a34389e9673a12335d09004ea6f1c82266a5e/outlines_core-0.2.11-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:e88b7f717915d91136d915adb65c2603d2aa6457ec3fc336884bdb0b28d3188a", size = 1960688, upload-time = "2025-05-19T10:11:56.773Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/02/4cffd04e360e315b060692bf1a80f84bac1671ef90f12daf765db6d68791/outlines_core-0.2.11-cp310-cp310-macosx_15_0_x86_64.whl", hash = "sha256:8c7ecdba2162e9b30b837251387c26b1a23f80f58d01d02e7600e4b1962c5333", size = 2130263, upload-time = "2025-05-19T10:11:58.1Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/85/69a450a486824026eca181a8d573aae3ecfdb25f0c2af852065dde17a372/outlines_core-0.2.11-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dd5fcefd221c10c95ce74838869450c6fdbbe2f581f0ba27e57a95232bd88c3a", size = 2289453, upload-time = "2025-05-19T10:11:59.919Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/3c/d7cb3eac6870a68b9034854fbfa07e67abfa1fa0d92198b9fee83fe6d044/outlines_core-0.2.11-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:a3c7774b112106f3afe931c65637fb3e0725d43707ceff1d34d6899cf0fa8200", size = 2115289, upload-time = "2025-05-19T10:12:01.527Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/5f/4cef22e2cf1ec286bd78c0052a0fa7ecf8519144477e7d4e276cbd70c625/outlines_core-0.2.11-cp310-cp310-win32.whl", hash = "sha256:1cfbb4cdcf34be5c6b08d279928b2b1050ed4c5e96e6e8405e3e624305c6799e", size = 1768059, upload-time = "2025-05-19T10:12:03.058Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/3a/ce6aceb6545bb1e13cf05c1f34468c5c14c8c8be92cdabcf777b4bb067ef/outlines_core-0.2.11-cp310-cp310-win_amd64.whl", hash = "sha256:670c1c1fca26fb5c7f00dbb11d1f81cca4204863c3dfdeee82017a6846397bf9", size = 2062413, upload-time = "2025-05-19T10:12:05.097Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/ca/d5e92e197b40f62deb46dcc55567a51c8bf37943df7bc6658d93f30740f1/outlines_core-0.2.11-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:e96b8d0b56afcd3b86f4efca466c578f3725da1148ef62423249c92993841762", size = 1961746, upload-time = "2025-05-19T10:12:06.723Z" },
+    { url = "https://files.pythonhosted.org/packages/02/b2/f3d6e7e37ebe1de3c345b53d8dc01e9b5c5f05b20e494fe94bf8972db4b0/outlines_core-0.2.11-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:d108ee8cd5e2fe71c2b0720b949d004901fec8bdb64bcd0c01b8abe38ab7ae1c", size = 2133815, upload-time = "2025-05-19T10:12:07.934Z" },
+    { url = "https://files.pythonhosted.org/packages/07/21/62a680da6941b53d765160d22bdcf35849c22b7a987f4e9e8b7db7885c9f/outlines_core-0.2.11-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:ebf42ab5b7ae38235d3c3333b5cacd6e91449b87b8a48a85094ea28ad9de9878", size = 1960539, upload-time = "2025-05-19T10:12:09.23Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/57/20cfb402aee1a7be0e08d861349570255ad2d17ba7fe7f8fd5706326588c/outlines_core-0.2.11-cp311-cp311-macosx_15_0_x86_64.whl", hash = "sha256:fd4305ff8418d14059d95dc3276ca96ba1b5aa499908e1af8bb3c7207aa7ac68", size = 2129894, upload-time = "2025-05-19T10:12:10.534Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/db/32c6e1170f139420e948fdd18a09a6175244bc0760dcf4dc2470e18411b9/outlines_core-0.2.11-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:132605b8dd1e3d1369da6a851992dd357f6376068292f6bd47caa7a28b794d19", size = 2289078, upload-time = "2025-05-19T10:12:12.118Z" },
+    { url = "https://files.pythonhosted.org/packages/25/c3/b6e6f4e08fa84d2424f82705a6dc47fee33cb91989010fa678736957dcf6/outlines_core-0.2.11-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:b31d5fc83b78aad282dd667b8d6e684614481fe08a7609ce0ce45dee64cd2991", size = 2115075, upload-time = "2025-05-19T10:12:13.761Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/9b/b84c4933e4f35b34e9b23fadd63a365ad8563cc7561d8528b33de4ee8102/outlines_core-0.2.11-cp311-cp311-win32.whl", hash = "sha256:3e316a79f3ecfa12c17746edebcbd66538ee22a43986982f6b96166fb94ee6b1", size = 1768254, upload-time = "2025-05-19T10:12:15.02Z" },
+    { url = "https://files.pythonhosted.org/packages/99/5b/380c933c65ca9744c163fe4a3702ad7f3e9ca02e09ac84a09b6837cff9b6/outlines_core-0.2.11-cp311-cp311-win_amd64.whl", hash = "sha256:c260a042b5854ff69291649cfd112066e6bab0dad0bb9cec8a6c3705ef3a59cd", size = 2062167, upload-time = "2025-05-19T10:12:16.443Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/2c/c7636823244c70e2960060bf9bd978248dffb55c5e7c91c46d18354b2a24/outlines_core-0.2.11-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:4a9db4872bae083631d720994f4cee603bce0536b33d5a988814576863b657cf", size = 1957668, upload-time = "2025-05-19T10:12:18.29Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/09/5c62047da139d722317a444a4d01cd5f11943a8c2eaecce784341dd0844a/outlines_core-0.2.11-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:8359a45c59f6a8f2eb717245806501a59044c75f6ea8bd08faaa131cc8cdec45", size = 2130493, upload-time = "2025-05-19T10:12:19.537Z" },
+    { url = "https://files.pythonhosted.org/packages/89/7a/d6a2810f90e37d550168e0c0a9a915086ea721444727e3ca2c630898d1ef/outlines_core-0.2.11-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:5d26a46591377340e0b870b8a96ea8341058341a62ee0bded9098e0c88dd24f4", size = 1956804, upload-time = "2025-05-19T10:12:20.755Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/ea/339e6c273b5581128c3b7ca27d428d8993c3085912af1a467aa32ef0e9d1/outlines_core-0.2.11-cp312-cp312-macosx_15_0_x86_64.whl", hash = "sha256:ae460a34675fb11d92a5c605a480fbae4cd6c1b2d11b3698da64a7fcaba64dcf", size = 2127085, upload-time = "2025-05-19T10:12:22.02Z" },
+    { url = "https://files.pythonhosted.org/packages/92/c7/a65d1fddf49830ebc41422294eacde35286d9f68994a8aa905cb14f5aade/outlines_core-0.2.11-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86df9740368866295077346440d911df4972da2b3f1f54b8125e6f329e8a8891", size = 2287677, upload-time = "2025-05-19T10:12:24.24Z" },
+    { url = "https://files.pythonhosted.org/packages/23/79/8795aed8be9b77dd69d78e7cfbfcf28c179e6b08da6e56bbbf48a09fe55f/outlines_core-0.2.11-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:96ce4dd78f106799be4a0a5795cefd1352806162973756a4b6fce4bb6eddd7e4", size = 2113000, upload-time = "2025-05-19T10:12:25.446Z" },
+    { url = "https://files.pythonhosted.org/packages/59/e3/cbe9294b06d92ee1892dbb6f2125d833d68e8629d45d080d6daba54eec2d/outlines_core-0.2.11-cp312-cp312-win32.whl", hash = "sha256:358db161cce3650ba822e118dcf0a1efa571c7deb4864ab9d64ca2c9cca7425d", size = 1765703, upload-time = "2025-05-19T10:12:26.693Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/c9/ed3cf362515fac16e313368b9b2f2497051f4ded88679205830b6f889f54/outlines_core-0.2.11-cp312-cp312-win_amd64.whl", hash = "sha256:231f9d20d2630c70665345821780d7808b29539620a75c99f65113b518c51032", size = 2060945, upload-time = "2025-05-19T10:12:28.294Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/52/7427d91791496f596abac5f387fe30a4215b3e464ba460fc97da8ba09a1f/outlines_core-0.2.11-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:defe30707d2c7718e6572b222028de1973c150ce3ec29ecf3f16dc5309a313ee", size = 1962300, upload-time = "2025-05-19T10:12:40.568Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/e1/6f68b0ce742ca922b80574c9c0ba4e7205651e421406dfa90e872b1d6cd6/outlines_core-0.2.11-cp39-cp39-macosx_14_0_x86_64.whl", hash = "sha256:576fefbf50ff09ad3b42e3d5bd344d8668fc650188fcc06b9a0356fdc6a89b84", size = 2134788, upload-time = "2025-05-19T10:12:41.857Z" },
+    { url = "https://files.pythonhosted.org/packages/04/d1/2a7fd4b0b5be3b726e9907a49c4bb9143d5b3dd863ffaaca92b31ccfca19/outlines_core-0.2.11-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:63a2f1d54929421ac8af715921a67b6da1f52cfe7c3ca6cddb194268bbc99140", size = 1961635, upload-time = "2025-05-19T10:12:43.117Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/98/63bfea48425cc176ed4edd3f031d2ed52dc8e050fad43660f0b097a20f80/outlines_core-0.2.11-cp39-cp39-macosx_15_0_x86_64.whl", hash = "sha256:90f43cc83a109bfe72f4862d34b1d29e28c76477bbdf58b091ec34aa7f795ff1", size = 2129094, upload-time = "2025-05-19T10:12:44.358Z" },
+    { url = "https://files.pythonhosted.org/packages/42/51/7ad61482a08fc3cc4b8dd03f68263ee226c2c60e98ed48e32e1053e0d3e5/outlines_core-0.2.11-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dae17b09f6f08d01fa0c228ab282197379ea10aa46b27f40b80c2014331af217", size = 2289932, upload-time = "2025-05-19T10:12:45.903Z" },
+    { url = "https://files.pythonhosted.org/packages/91/85/c70d258ef04e10e2d8086c362005a2079c2fb23caa16d1cf7ead1f4a9b16/outlines_core-0.2.11-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:3a9db6831346ec4e683022c05b45403ec1c5f4a3fe52a2a7ebcc1d7d9dc3a5fb", size = 2116012, upload-time = "2025-05-19T10:12:47.496Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/8f/662ad51b3874a12e98109adfbb580e511191249c251edff4ec4a7ecdbfcb/outlines_core-0.2.11-cp39-cp39-win32.whl", hash = "sha256:a41c2d518367a4628bca3e4f509b268642c2cdec70b631c64f07d5158d029e0d", size = 1768535, upload-time = "2025-05-19T10:12:48.724Z" },
+    { url = "https://files.pythonhosted.org/packages/54/b4/945fcfe5054f2737159a866a3bfcfd5fab50dbb6bf48aa38093412f68fa9/outlines_core-0.2.11-cp39-cp39-win_amd64.whl", hash = "sha256:bc173be0f5c089c23fdb1df0dc4b9075140be2f4928748fefc58ea46a2bd36bd", size = 2062756, upload-time = "2025-05-19T10:12:50.032Z" },
 ]
 
 [[package]]
 name = "packaging"
-version = "24.2"
+version = "25.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/d0/63/68dbb6eb2de9cb10ee4c9c14a0148804425e13c4fb20d61cce69f53106da/packaging-24.2.tar.gz", hash = "sha256:c228a6dc5e932d346bc5739379109d49e8853dd8223571c7c5b55260edc0b97f", size = 163950 }
+sdist = { url = "https://files.pythonhosted.org/packages/a1/d4/1fc4078c65507b51b96ca8f8c3ba19e6a61c8253c72794544580a7b6c24d/packaging-25.0.tar.gz", hash = "sha256:d443872c98d677bf60f6a1f2f8c1cb748e8fe762d2bf9d3148b5599295b0fc4f", size = 165727, upload-time = "2025-04-19T11:48:59.673Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/88/ef/eb23f262cca3c0c4eb7ab1933c3b1f03d021f2c48f54763065b6f0e321be/packaging-24.2-py3-none-any.whl", hash = "sha256:09abb1bccd265c01f4a3aa3f7a7db064b36514d2cba19a2f694fe6150451a759", size = 65451 },
+    { url = "https://files.pythonhosted.org/packages/20/12/38679034af332785aac8774540895e234f4d07f7545804097de4b666afd8/packaging-25.0-py3-none-any.whl", hash = "sha256:29572ef2b1f17581046b3a2227d5c611fb25ec70ca1ba8554b24b0e69331a484", size = 66469, upload-time = "2025-04-19T11:48:57.875Z" },
 ]
 
 [[package]]
 name = "pandas"
-version = "2.2.3"
+version = "2.3.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "numpy" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "python-dateutil" },
     { name = "pytz" },
     { name = "tzdata" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/9c/d6/9f8431bacc2e19dca897724cd097b1bb224a6ad5433784a44b587c7c13af/pandas-2.2.3.tar.gz", hash = "sha256:4f18ba62b61d7e192368b84517265a99b4d7ee8912f8708660fb4a366cc82667", size = 4399213 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/aa/70/c853aec59839bceed032d52010ff5f1b8d87dc3114b762e4ba2727661a3b/pandas-2.2.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:1948ddde24197a0f7add2bdc4ca83bf2b1ef84a1bc8ccffd95eda17fd836ecb5", size = 12580827 },
-    { url = "https://files.pythonhosted.org/packages/99/f2/c4527768739ffa4469b2b4fff05aa3768a478aed89a2f271a79a40eee984/pandas-2.2.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:381175499d3802cde0eabbaf6324cce0c4f5d52ca6f8c377c29ad442f50f6348", size = 11303897 },
-    { url = "https://files.pythonhosted.org/packages/ed/12/86c1747ea27989d7a4064f806ce2bae2c6d575b950be087837bdfcabacc9/pandas-2.2.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d9c45366def9a3dd85a6454c0e7908f2b3b8e9c138f5dc38fed7ce720d8453ed", size = 66480908 },
-    { url = "https://files.pythonhosted.org/packages/44/50/7db2cd5e6373ae796f0ddad3675268c8d59fb6076e66f0c339d61cea886b/pandas-2.2.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86976a1c5b25ae3f8ccae3a5306e443569ee3c3faf444dfd0f41cda24667ad57", size = 13064210 },
-    { url = "https://files.pythonhosted.org/packages/61/61/a89015a6d5536cb0d6c3ba02cebed51a95538cf83472975275e28ebf7d0c/pandas-2.2.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:b8661b0238a69d7aafe156b7fa86c44b881387509653fdf857bebc5e4008ad42", size = 16754292 },
-    { url = "https://files.pythonhosted.org/packages/ce/0d/4cc7b69ce37fac07645a94e1d4b0880b15999494372c1523508511b09e40/pandas-2.2.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:37e0aced3e8f539eccf2e099f65cdb9c8aa85109b0be6e93e2baff94264bdc6f", size = 14416379 },
-    { url = "https://files.pythonhosted.org/packages/31/9e/6ebb433de864a6cd45716af52a4d7a8c3c9aaf3a98368e61db9e69e69a9c/pandas-2.2.3-cp310-cp310-win_amd64.whl", hash = "sha256:56534ce0746a58afaf7942ba4863e0ef81c9c50d3f0ae93e9497d6a41a057645", size = 11598471 },
-    { url = "https://files.pythonhosted.org/packages/a8/44/d9502bf0ed197ba9bf1103c9867d5904ddcaf869e52329787fc54ed70cc8/pandas-2.2.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:66108071e1b935240e74525006034333f98bcdb87ea116de573a6a0dccb6c039", size = 12602222 },
-    { url = "https://files.pythonhosted.org/packages/52/11/9eac327a38834f162b8250aab32a6781339c69afe7574368fffe46387edf/pandas-2.2.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7c2875855b0ff77b2a64a0365e24455d9990730d6431b9e0ee18ad8acee13dbd", size = 11321274 },
-    { url = "https://files.pythonhosted.org/packages/45/fb/c4beeb084718598ba19aa9f5abbc8aed8b42f90930da861fcb1acdb54c3a/pandas-2.2.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:cd8d0c3be0515c12fed0bdbae072551c8b54b7192c7b1fda0ba56059a0179698", size = 15579836 },
-    { url = "https://files.pythonhosted.org/packages/cd/5f/4dba1d39bb9c38d574a9a22548c540177f78ea47b32f99c0ff2ec499fac5/pandas-2.2.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c124333816c3a9b03fbeef3a9f230ba9a737e9e5bb4060aa2107a86cc0a497fc", size = 13058505 },
-    { url = "https://files.pythonhosted.org/packages/b9/57/708135b90391995361636634df1f1130d03ba456e95bcf576fada459115a/pandas-2.2.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:63cc132e40a2e084cf01adf0775b15ac515ba905d7dcca47e9a251819c575ef3", size = 16744420 },
-    { url = "https://files.pythonhosted.org/packages/86/4a/03ed6b7ee323cf30404265c284cee9c65c56a212e0a08d9ee06984ba2240/pandas-2.2.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:29401dbfa9ad77319367d36940cd8a0b3a11aba16063e39632d98b0e931ddf32", size = 14440457 },
-    { url = "https://files.pythonhosted.org/packages/ed/8c/87ddf1fcb55d11f9f847e3c69bb1c6f8e46e2f40ab1a2d2abadb2401b007/pandas-2.2.3-cp311-cp311-win_amd64.whl", hash = "sha256:3fc6873a41186404dad67245896a6e440baacc92f5b716ccd1bc9ed2995ab2c5", size = 11617166 },
-    { url = "https://files.pythonhosted.org/packages/17/a3/fb2734118db0af37ea7433f57f722c0a56687e14b14690edff0cdb4b7e58/pandas-2.2.3-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:b1d432e8d08679a40e2a6d8b2f9770a5c21793a6f9f47fdd52c5ce1948a5a8a9", size = 12529893 },
-    { url = "https://files.pythonhosted.org/packages/e1/0c/ad295fd74bfac85358fd579e271cded3ac969de81f62dd0142c426b9da91/pandas-2.2.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a5a1595fe639f5988ba6a8e5bc9649af3baf26df3998a0abe56c02609392e0a4", size = 11363475 },
-    { url = "https://files.pythonhosted.org/packages/c6/2a/4bba3f03f7d07207481fed47f5b35f556c7441acddc368ec43d6643c5777/pandas-2.2.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5de54125a92bb4d1c051c0659e6fcb75256bf799a732a87184e5ea503965bce3", size = 15188645 },
-    { url = "https://files.pythonhosted.org/packages/38/f8/d8fddee9ed0d0c0f4a2132c1dfcf0e3e53265055da8df952a53e7eaf178c/pandas-2.2.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fffb8ae78d8af97f849404f21411c95062db1496aeb3e56f146f0355c9989319", size = 12739445 },
-    { url = "https://files.pythonhosted.org/packages/20/e8/45a05d9c39d2cea61ab175dbe6a2de1d05b679e8de2011da4ee190d7e748/pandas-2.2.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:6dfcb5ee8d4d50c06a51c2fffa6cff6272098ad6540aed1a76d15fb9318194d8", size = 16359235 },
-    { url = "https://files.pythonhosted.org/packages/1d/99/617d07a6a5e429ff90c90da64d428516605a1ec7d7bea494235e1c3882de/pandas-2.2.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:062309c1b9ea12a50e8ce661145c6aab431b1e99530d3cd60640e255778bd43a", size = 14056756 },
-    { url = "https://files.pythonhosted.org/packages/29/d4/1244ab8edf173a10fd601f7e13b9566c1b525c4f365d6bee918e68381889/pandas-2.2.3-cp312-cp312-win_amd64.whl", hash = "sha256:59ef3764d0fe818125a5097d2ae867ca3fa64df032331b7e0917cf5d7bf66b13", size = 11504248 },
-    { url = "https://files.pythonhosted.org/packages/64/22/3b8f4e0ed70644e85cfdcd57454686b9057c6c38d2f74fe4b8bc2527214a/pandas-2.2.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:f00d1345d84d8c86a63e476bb4955e46458b304b9575dcf71102b5c705320015", size = 12477643 },
-    { url = "https://files.pythonhosted.org/packages/e4/93/b3f5d1838500e22c8d793625da672f3eec046b1a99257666c94446969282/pandas-2.2.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:3508d914817e153ad359d7e069d752cdd736a247c322d932eb89e6bc84217f28", size = 11281573 },
-    { url = "https://files.pythonhosted.org/packages/f5/94/6c79b07f0e5aab1dcfa35a75f4817f5c4f677931d4234afcd75f0e6a66ca/pandas-2.2.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:22a9d949bfc9a502d320aa04e5d02feab689d61da4e7764b62c30b991c42c5f0", size = 15196085 },
-    { url = "https://files.pythonhosted.org/packages/e8/31/aa8da88ca0eadbabd0a639788a6da13bb2ff6edbbb9f29aa786450a30a91/pandas-2.2.3-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3a255b2c19987fbbe62a9dfd6cff7ff2aa9ccab3fc75218fd4b7530f01efa24", size = 12711809 },
-    { url = "https://files.pythonhosted.org/packages/ee/7c/c6dbdb0cb2a4344cacfb8de1c5808ca885b2e4dcfde8008266608f9372af/pandas-2.2.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:800250ecdadb6d9c78eae4990da62743b857b470883fa27f652db8bdde7f6659", size = 16356316 },
-    { url = "https://files.pythonhosted.org/packages/57/b7/8b757e7d92023b832869fa8881a992696a0bfe2e26f72c9ae9f255988d42/pandas-2.2.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:6374c452ff3ec675a8f46fd9ab25c4ad0ba590b71cf0656f8b6daa5202bca3fb", size = 14022055 },
-    { url = "https://files.pythonhosted.org/packages/3b/bc/4b18e2b8c002572c5a441a64826252ce5da2aa738855747247a971988043/pandas-2.2.3-cp313-cp313-win_amd64.whl", hash = "sha256:61c5ad4043f791b61dd4752191d9f07f0ae412515d59ba8f005832a532f8736d", size = 11481175 },
-    { url = "https://files.pythonhosted.org/packages/76/a3/a5d88146815e972d40d19247b2c162e88213ef51c7c25993942c39dbf41d/pandas-2.2.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:3b71f27954685ee685317063bf13c7709a7ba74fc996b84fc6821c59b0f06468", size = 12615650 },
-    { url = "https://files.pythonhosted.org/packages/9c/8c/f0fd18f6140ddafc0c24122c8a964e48294acc579d47def376fef12bcb4a/pandas-2.2.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:38cf8125c40dae9d5acc10fa66af8ea6fdf760b2714ee482ca691fc66e6fcb18", size = 11290177 },
-    { url = "https://files.pythonhosted.org/packages/ed/f9/e995754eab9c0f14c6777401f7eece0943840b7a9fc932221c19d1abee9f/pandas-2.2.3-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:ba96630bc17c875161df3818780af30e43be9b166ce51c9a18c1feae342906c2", size = 14651526 },
-    { url = "https://files.pythonhosted.org/packages/25/b0/98d6ae2e1abac4f35230aa756005e8654649d305df9a28b16b9ae4353bff/pandas-2.2.3-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1db71525a1538b30142094edb9adc10be3f3e176748cd7acc2240c2f2e5aa3a4", size = 11871013 },
-    { url = "https://files.pythonhosted.org/packages/cc/57/0f72a10f9db6a4628744c8e8f0df4e6e21de01212c7c981d31e50ffc8328/pandas-2.2.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:15c0e1e02e93116177d29ff83e8b1619c93ddc9c49083f237d4312337a61165d", size = 15711620 },
-    { url = "https://files.pythonhosted.org/packages/ab/5f/b38085618b950b79d2d9164a711c52b10aefc0ae6833b96f626b7021b2ed/pandas-2.2.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:ad5b65698ab28ed8d7f18790a0dc58005c7629f227be9ecc1072aa74c0c1d43a", size = 13098436 },
-    { url = "https://files.pythonhosted.org/packages/ca/8c/8848a4c9b8fdf5a534fe2077af948bf53cd713d77ffbcd7bd15710348fd7/pandas-2.2.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bc6b93f9b966093cb0fd62ff1a7e4c09e6d546ad7c1de191767baffc57628f39", size = 12595535 },
-    { url = "https://files.pythonhosted.org/packages/9c/b9/5cead4f63b6d31bdefeb21a679bc5a7f4aaf262ca7e07e2bc1c341b68470/pandas-2.2.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:5dbca4c1acd72e8eeef4753eeca07de9b1db4f398669d5994086f788a5d7cc30", size = 11319822 },
-    { url = "https://files.pythonhosted.org/packages/31/af/89e35619fb573366fa68dc26dad6ad2c08c17b8004aad6d98f1a31ce4bb3/pandas-2.2.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:8cd6d7cc958a3910f934ea8dbdf17b2364827bb4dafc38ce6eef6bb3d65ff09c", size = 15625439 },
-    { url = "https://files.pythonhosted.org/packages/3d/dd/bed19c2974296661493d7acc4407b1d2db4e2a482197df100f8f965b6225/pandas-2.2.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:99df71520d25fade9db7c1076ac94eb994f4d2673ef2aa2e86ee039b6746d20c", size = 13068928 },
-    { url = "https://files.pythonhosted.org/packages/31/a3/18508e10a31ea108d746c848b5a05c0711e0278fa0d6f1c52a8ec52b80a5/pandas-2.2.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:31d0ced62d4ea3e231a9f228366919a5ea0b07440d9d4dac345376fd8e1477ea", size = 16783266 },
-    { url = "https://files.pythonhosted.org/packages/c4/a5/3429bd13d82bebc78f4d78c3945efedef63a7cd0c15c17b2eeb838d1121f/pandas-2.2.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:7eee9e7cea6adf3e3d24e304ac6b8300646e2a5d1cd3a3c2abed9101b0846761", size = 14450871 },
-    { url = "https://files.pythonhosted.org/packages/2f/49/5c30646e96c684570925b772eac4eb0a8cb0ca590fa978f56c5d3ae73ea1/pandas-2.2.3-cp39-cp39-win_amd64.whl", hash = "sha256:4850ba03528b6dd51d6c5d273c46f183f39a9baf3f0143e566b89450965b105e", size = 11618011 },
+sdist = { url = "https://files.pythonhosted.org/packages/79/8e/0e90233ac205ad182bd6b422532695d2b9414944a280488105d598c70023/pandas-2.3.2.tar.gz", hash = "sha256:ab7b58f8f82706890924ccdfb5f48002b83d2b5a3845976a9fb705d36c34dcdb", size = 4488684, upload-time = "2025-08-21T10:28:29.257Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2e/16/a8eeb70aad84ccbf14076793f90e0031eded63c1899aeae9fdfbf37881f4/pandas-2.3.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:52bc29a946304c360561974c6542d1dd628ddafa69134a7131fdfd6a5d7a1a35", size = 11539648, upload-time = "2025-08-21T10:26:36.236Z" },
+    { url = "https://files.pythonhosted.org/packages/47/f1/c5bdaea13bf3708554d93e948b7ea74121ce6e0d59537ca4c4f77731072b/pandas-2.3.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:220cc5c35ffaa764dd5bb17cf42df283b5cb7fdf49e10a7b053a06c9cb48ee2b", size = 10786923, upload-time = "2025-08-21T10:26:40.518Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/10/811fa01476d29ffed692e735825516ad0e56d925961819e6126b4ba32147/pandas-2.3.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:42c05e15111221384019897df20c6fe893b2f697d03c811ee67ec9e0bb5a3424", size = 11726241, upload-time = "2025-08-21T10:26:43.175Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/6a/40b043b06e08df1ea1b6d20f0e0c2f2c4ec8c4f07d1c92948273d943a50b/pandas-2.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cc03acc273c5515ab69f898df99d9d4f12c4d70dbfc24c3acc6203751d0804cf", size = 12349533, upload-time = "2025-08-21T10:26:46.611Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/ea/2e081a2302e41a9bca7056659fdd2b85ef94923723e41665b42d65afd347/pandas-2.3.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d25c20a03e8870f6339bcf67281b946bd20b86f1a544ebbebb87e66a8d642cba", size = 13202407, upload-time = "2025-08-21T10:26:49.068Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/12/7ff9f6a79e2ee8869dcf70741ef998b97ea20050fe25f83dc759764c1e32/pandas-2.3.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:21bb612d148bb5860b7eb2c10faacf1a810799245afd342cf297d7551513fbb6", size = 13837212, upload-time = "2025-08-21T10:26:51.832Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/df/5ab92fcd76455a632b3db34a746e1074d432c0cdbbd28d7cd1daba46a75d/pandas-2.3.2-cp310-cp310-win_amd64.whl", hash = "sha256:b62d586eb25cb8cb70a5746a378fc3194cb7f11ea77170d59f889f5dfe3cec7a", size = 11338099, upload-time = "2025-08-21T10:26:54.382Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/59/f3e010879f118c2d400902d2d871c2226cef29b08c09fb8dc41111730400/pandas-2.3.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:1333e9c299adcbb68ee89a9bb568fc3f20f9cbb419f1dd5225071e6cddb2a743", size = 11563308, upload-time = "2025-08-21T10:26:56.656Z" },
+    { url = "https://files.pythonhosted.org/packages/38/18/48f10f1cc5c397af59571d638d211f494dba481f449c19adbd282aa8f4ca/pandas-2.3.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:76972bcbd7de8e91ad5f0ca884a9f2c477a2125354af624e022c49e5bd0dfff4", size = 10820319, upload-time = "2025-08-21T10:26:59.162Z" },
+    { url = "https://files.pythonhosted.org/packages/95/3b/1e9b69632898b048e223834cd9702052bcf06b15e1ae716eda3196fb972e/pandas-2.3.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b98bdd7c456a05eef7cd21fd6b29e3ca243591fe531c62be94a2cc987efb5ac2", size = 11790097, upload-time = "2025-08-21T10:27:02.204Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/ef/0e2ffb30b1f7fbc9a588bd01e3c14a0d96854d09a887e15e30cc19961227/pandas-2.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1d81573b3f7db40d020983f78721e9bfc425f411e616ef019a10ebf597aedb2e", size = 12397958, upload-time = "2025-08-21T10:27:05.409Z" },
+    { url = "https://files.pythonhosted.org/packages/23/82/e6b85f0d92e9afb0e7f705a51d1399b79c7380c19687bfbf3d2837743249/pandas-2.3.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:e190b738675a73b581736cc8ec71ae113d6c3768d0bd18bffa5b9a0927b0b6ea", size = 13225600, upload-time = "2025-08-21T10:27:07.791Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/f1/f682015893d9ed51611948bd83683670842286a8edd4f68c2c1c3b231eef/pandas-2.3.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:c253828cb08f47488d60f43c5fc95114c771bbfff085da54bfc79cb4f9e3a372", size = 13879433, upload-time = "2025-08-21T10:27:10.347Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/e7/ae86261695b6c8a36d6a4c8d5f9b9ede8248510d689a2f379a18354b37d7/pandas-2.3.2-cp311-cp311-win_amd64.whl", hash = "sha256:9467697b8083f9667b212633ad6aa4ab32436dcbaf4cd57325debb0ddef2012f", size = 11336557, upload-time = "2025-08-21T10:27:12.983Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/db/614c20fb7a85a14828edd23f1c02db58a30abf3ce76f38806155d160313c/pandas-2.3.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:3fbb977f802156e7a3f829e9d1d5398f6192375a3e2d1a9ee0803e35fe70a2b9", size = 11587652, upload-time = "2025-08-21T10:27:15.888Z" },
+    { url = "https://files.pythonhosted.org/packages/99/b0/756e52f6582cade5e746f19bad0517ff27ba9c73404607c0306585c201b3/pandas-2.3.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1b9b52693123dd234b7c985c68b709b0b009f4521000d0525f2b95c22f15944b", size = 10717686, upload-time = "2025-08-21T10:27:18.486Z" },
+    { url = "https://files.pythonhosted.org/packages/37/4c/dd5ccc1e357abfeee8353123282de17997f90ff67855f86154e5a13b81e5/pandas-2.3.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0bd281310d4f412733f319a5bc552f86d62cddc5f51d2e392c8787335c994175", size = 11278722, upload-time = "2025-08-21T10:27:21.149Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/a4/f7edcfa47e0a88cda0be8b068a5bae710bf264f867edfdf7b71584ace362/pandas-2.3.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:96d31a6b4354e3b9b8a2c848af75d31da390657e3ac6f30c05c82068b9ed79b9", size = 11987803, upload-time = "2025-08-21T10:27:23.767Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/61/1bce4129f93ab66f1c68b7ed1c12bac6a70b1b56c5dab359c6bbcd480b52/pandas-2.3.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:df4df0b9d02bb873a106971bb85d448378ef14b86ba96f035f50bbd3688456b4", size = 12766345, upload-time = "2025-08-21T10:27:26.6Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/46/80d53de70fee835531da3a1dae827a1e76e77a43ad22a8cd0f8142b61587/pandas-2.3.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:213a5adf93d020b74327cb2c1b842884dbdd37f895f42dcc2f09d451d949f811", size = 13439314, upload-time = "2025-08-21T10:27:29.213Z" },
+    { url = "https://files.pythonhosted.org/packages/28/30/8114832daff7489f179971dbc1d854109b7f4365a546e3ea75b6516cea95/pandas-2.3.2-cp312-cp312-win_amd64.whl", hash = "sha256:8c13b81a9347eb8c7548f53fd9a4f08d4dfe996836543f805c987bafa03317ae", size = 10983326, upload-time = "2025-08-21T10:27:31.901Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/c3/b37e090d0aceda9b4dd85c8dbd1bea65b1de9e7a4f690d6bd3a40bd16390/pandas-2.3.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:88080a0ff8a55eac9c84e3ff3c7665b3b5476c6fbc484775ca1910ce1c3e0b87", size = 11551511, upload-time = "2025-08-21T10:28:11.111Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/47/381fb1e7adcfcf4230fa6dc3a741acbac6c6fe072f19f4e7a46bddf3e5f6/pandas-2.3.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:d4a558c7620340a0931828d8065688b3cc5b4c8eb674bcaf33d18ff4a6870b4a", size = 10797930, upload-time = "2025-08-21T10:28:13.436Z" },
+    { url = "https://files.pythonhosted.org/packages/36/ca/d42467829080b92fc46d451288af8068f129fbcfb6578d573f45120de5cf/pandas-2.3.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:45178cf09d1858a1509dc73ec261bf5b25a625a389b65be2e47b559905f0ab6a", size = 11738470, upload-time = "2025-08-21T10:28:16.065Z" },
+    { url = "https://files.pythonhosted.org/packages/60/76/7d0f0a0deed7867c51163982d7b79c0a089096cd7ad50e1b87c2c82220e9/pandas-2.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:77cefe00e1b210f9c76c697fedd8fdb8d3dd86563e9c8adc9fa72b90f5e9e4c2", size = 12366640, upload-time = "2025-08-21T10:28:18.557Z" },
+    { url = "https://files.pythonhosted.org/packages/21/31/56784743e421cf51e34358fe7e5954345e5942168897bf8eb5707b71eedb/pandas-2.3.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:13bd629c653856f00c53dc495191baa59bcafbbf54860a46ecc50d3a88421a96", size = 13211567, upload-time = "2025-08-21T10:28:20.998Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/4e/50a399dc7d9dd4aa09a03b163751d428026cf0f16c419b4010f6aca26ebd/pandas-2.3.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:36d627906fd44b5fd63c943264e11e96e923f8de77d6016dc2f667b9ad193438", size = 13854073, upload-time = "2025-08-21T10:28:24.056Z" },
+    { url = "https://files.pythonhosted.org/packages/29/72/8978a84861a5124e56ce1048376569545412501fcb9a83f035393d6d85bc/pandas-2.3.2-cp39-cp39-win_amd64.whl", hash = "sha256:a9d7ec92d71a420185dec44909c32e9a362248c4ae2238234b76d5be37f208cc", size = 11346452, upload-time = "2025-08-21T10:28:26.691Z" },
+]
+
+[[package]]
+name = "pathspec"
+version = "0.12.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ca/bc/f35b8446f4531a7cb215605d100cd88b7ac6f44ab3fc94870c120ab3adbf/pathspec-0.12.1.tar.gz", hash = "sha256:a482d51503a1ab33b1c67a6c3813a26953dbdc71c31dacaef9a838c4e29f5712", size = 51043, upload-time = "2023-12-10T22:30:45Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cc/20/ff623b09d963f88bfde16306a54e12ee5ea43e9b597108672ff3a408aad6/pathspec-0.12.1-py3-none-any.whl", hash = "sha256:a0d503e138a4c123b27490a4f7beda6a01c6f288df0e4a8b79c7eb0dc7b4cc08", size = 31191, upload-time = "2023-12-10T22:30:43.14Z" },
 ]
 
 [[package]]
 name = "pillow"
-version = "11.1.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f3/af/c097e544e7bd278333db77933e535098c259609c4eb3b85381109602fb5b/pillow-11.1.0.tar.gz", hash = "sha256:368da70808b36d73b4b390a8ffac11069f8a5c85f29eff1f1b01bcf3ef5b2a20", size = 46742715 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/50/1c/2dcea34ac3d7bc96a1fd1bd0a6e06a57c67167fec2cff8d95d88229a8817/pillow-11.1.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:e1abe69aca89514737465752b4bcaf8016de61b3be1397a8fc260ba33321b3a8", size = 3229983 },
-    { url = "https://files.pythonhosted.org/packages/14/ca/6bec3df25e4c88432681de94a3531cc738bd85dea6c7aa6ab6f81ad8bd11/pillow-11.1.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c640e5a06869c75994624551f45e5506e4256562ead981cce820d5ab39ae2192", size = 3101831 },
-    { url = "https://files.pythonhosted.org/packages/d4/2c/668e18e5521e46eb9667b09e501d8e07049eb5bfe39d56be0724a43117e6/pillow-11.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a07dba04c5e22824816b2615ad7a7484432d7f540e6fa86af60d2de57b0fcee2", size = 4314074 },
-    { url = "https://files.pythonhosted.org/packages/02/80/79f99b714f0fc25f6a8499ecfd1f810df12aec170ea1e32a4f75746051ce/pillow-11.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e267b0ed063341f3e60acd25c05200df4193e15a4a5807075cd71225a2386e26", size = 4394933 },
-    { url = "https://files.pythonhosted.org/packages/81/aa/8d4ad25dc11fd10a2001d5b8a80fdc0e564ac33b293bdfe04ed387e0fd95/pillow-11.1.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:bd165131fd51697e22421d0e467997ad31621b74bfc0b75956608cb2906dda07", size = 4353349 },
-    { url = "https://files.pythonhosted.org/packages/84/7a/cd0c3eaf4a28cb2a74bdd19129f7726277a7f30c4f8424cd27a62987d864/pillow-11.1.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:abc56501c3fd148d60659aae0af6ddc149660469082859fa7b066a298bde9482", size = 4476532 },
-    { url = "https://files.pythonhosted.org/packages/8f/8b/a907fdd3ae8f01c7670dfb1499c53c28e217c338b47a813af8d815e7ce97/pillow-11.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:54ce1c9a16a9561b6d6d8cb30089ab1e5eb66918cb47d457bd996ef34182922e", size = 4279789 },
-    { url = "https://files.pythonhosted.org/packages/6f/9a/9f139d9e8cccd661c3efbf6898967a9a337eb2e9be2b454ba0a09533100d/pillow-11.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:73ddde795ee9b06257dac5ad42fcb07f3b9b813f8c1f7f870f402f4dc54b5269", size = 4413131 },
-    { url = "https://files.pythonhosted.org/packages/a8/68/0d8d461f42a3f37432203c8e6df94da10ac8081b6d35af1c203bf3111088/pillow-11.1.0-cp310-cp310-win32.whl", hash = "sha256:3a5fe20a7b66e8135d7fd617b13272626a28278d0e578c98720d9ba4b2439d49", size = 2291213 },
-    { url = "https://files.pythonhosted.org/packages/14/81/d0dff759a74ba87715509af9f6cb21fa21d93b02b3316ed43bda83664db9/pillow-11.1.0-cp310-cp310-win_amd64.whl", hash = "sha256:b6123aa4a59d75f06e9dd3dac5bf8bc9aa383121bb3dd9a7a612e05eabc9961a", size = 2625725 },
-    { url = "https://files.pythonhosted.org/packages/ce/1f/8d50c096a1d58ef0584ddc37e6f602828515219e9d2428e14ce50f5ecad1/pillow-11.1.0-cp310-cp310-win_arm64.whl", hash = "sha256:a76da0a31da6fcae4210aa94fd779c65c75786bc9af06289cd1c184451ef7a65", size = 2375213 },
-    { url = "https://files.pythonhosted.org/packages/dd/d6/2000bfd8d5414fb70cbbe52c8332f2283ff30ed66a9cde42716c8ecbe22c/pillow-11.1.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:e06695e0326d05b06833b40b7ef477e475d0b1ba3a6d27da1bb48c23209bf457", size = 3229968 },
-    { url = "https://files.pythonhosted.org/packages/d9/45/3fe487010dd9ce0a06adf9b8ff4f273cc0a44536e234b0fad3532a42c15b/pillow-11.1.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:96f82000e12f23e4f29346e42702b6ed9a2f2fea34a740dd5ffffcc8c539eb35", size = 3101806 },
-    { url = "https://files.pythonhosted.org/packages/e3/72/776b3629c47d9d5f1c160113158a7a7ad177688d3a1159cd3b62ded5a33a/pillow-11.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a3cd561ded2cf2bbae44d4605837221b987c216cff94f49dfeed63488bb228d2", size = 4322283 },
-    { url = "https://files.pythonhosted.org/packages/e4/c2/e25199e7e4e71d64eeb869f5b72c7ddec70e0a87926398785ab944d92375/pillow-11.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f189805c8be5ca5add39e6f899e6ce2ed824e65fb45f3c28cb2841911da19070", size = 4402945 },
-    { url = "https://files.pythonhosted.org/packages/c1/ed/51d6136c9d5911f78632b1b86c45241c712c5a80ed7fa7f9120a5dff1eba/pillow-11.1.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:dd0052e9db3474df30433f83a71b9b23bd9e4ef1de13d92df21a52c0303b8ab6", size = 4361228 },
-    { url = "https://files.pythonhosted.org/packages/48/a4/fbfe9d5581d7b111b28f1d8c2762dee92e9821bb209af9fa83c940e507a0/pillow-11.1.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:837060a8599b8f5d402e97197d4924f05a2e0d68756998345c829c33186217b1", size = 4484021 },
-    { url = "https://files.pythonhosted.org/packages/39/db/0b3c1a5018117f3c1d4df671fb8e47d08937f27519e8614bbe86153b65a5/pillow-11.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:aa8dd43daa836b9a8128dbe7d923423e5ad86f50a7a14dc688194b7be5c0dea2", size = 4287449 },
-    { url = "https://files.pythonhosted.org/packages/d9/58/bc128da7fea8c89fc85e09f773c4901e95b5936000e6f303222490c052f3/pillow-11.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0a2f91f8a8b367e7a57c6e91cd25af510168091fb89ec5146003e424e1558a96", size = 4419972 },
-    { url = "https://files.pythonhosted.org/packages/5f/bb/58f34379bde9fe197f51841c5bbe8830c28bbb6d3801f16a83b8f2ad37df/pillow-11.1.0-cp311-cp311-win32.whl", hash = "sha256:c12fc111ef090845de2bb15009372175d76ac99969bdf31e2ce9b42e4b8cd88f", size = 2291201 },
-    { url = "https://files.pythonhosted.org/packages/3a/c6/fce9255272bcf0c39e15abd2f8fd8429a954cf344469eaceb9d0d1366913/pillow-11.1.0-cp311-cp311-win_amd64.whl", hash = "sha256:fbd43429d0d7ed6533b25fc993861b8fd512c42d04514a0dd6337fb3ccf22761", size = 2625686 },
-    { url = "https://files.pythonhosted.org/packages/c8/52/8ba066d569d932365509054859f74f2a9abee273edcef5cd75e4bc3e831e/pillow-11.1.0-cp311-cp311-win_arm64.whl", hash = "sha256:f7955ecf5609dee9442cbface754f2c6e541d9e6eda87fad7f7a989b0bdb9d71", size = 2375194 },
-    { url = "https://files.pythonhosted.org/packages/95/20/9ce6ed62c91c073fcaa23d216e68289e19d95fb8188b9fb7a63d36771db8/pillow-11.1.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:2062ffb1d36544d42fcaa277b069c88b01bb7298f4efa06731a7fd6cc290b81a", size = 3226818 },
-    { url = "https://files.pythonhosted.org/packages/b9/d8/f6004d98579a2596c098d1e30d10b248798cceff82d2b77aa914875bfea1/pillow-11.1.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a85b653980faad27e88b141348707ceeef8a1186f75ecc600c395dcac19f385b", size = 3101662 },
-    { url = "https://files.pythonhosted.org/packages/08/d9/892e705f90051c7a2574d9f24579c9e100c828700d78a63239676f960b74/pillow-11.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9409c080586d1f683df3f184f20e36fb647f2e0bc3988094d4fd8c9f4eb1b3b3", size = 4329317 },
-    { url = "https://files.pythonhosted.org/packages/8c/aa/7f29711f26680eab0bcd3ecdd6d23ed6bce180d82e3f6380fb7ae35fcf3b/pillow-11.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7fdadc077553621911f27ce206ffcbec7d3f8d7b50e0da39f10997e8e2bb7f6a", size = 4412999 },
-    { url = "https://files.pythonhosted.org/packages/c8/c4/8f0fe3b9e0f7196f6d0bbb151f9fba323d72a41da068610c4c960b16632a/pillow-11.1.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:93a18841d09bcdd774dcdc308e4537e1f867b3dec059c131fde0327899734aa1", size = 4368819 },
-    { url = "https://files.pythonhosted.org/packages/38/0d/84200ed6a871ce386ddc82904bfadc0c6b28b0c0ec78176871a4679e40b3/pillow-11.1.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:9aa9aeddeed452b2f616ff5507459e7bab436916ccb10961c4a382cd3e03f47f", size = 4496081 },
-    { url = "https://files.pythonhosted.org/packages/84/9c/9bcd66f714d7e25b64118e3952d52841a4babc6d97b6d28e2261c52045d4/pillow-11.1.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3cdcdb0b896e981678eee140d882b70092dac83ac1cdf6b3a60e2216a73f2b91", size = 4296513 },
-    { url = "https://files.pythonhosted.org/packages/db/61/ada2a226e22da011b45f7104c95ebda1b63dcbb0c378ad0f7c2a710f8fd2/pillow-11.1.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:36ba10b9cb413e7c7dfa3e189aba252deee0602c86c309799da5a74009ac7a1c", size = 4431298 },
-    { url = "https://files.pythonhosted.org/packages/e7/c4/fc6e86750523f367923522014b821c11ebc5ad402e659d8c9d09b3c9d70c/pillow-11.1.0-cp312-cp312-win32.whl", hash = "sha256:cfd5cd998c2e36a862d0e27b2df63237e67273f2fc78f47445b14e73a810e7e6", size = 2291630 },
-    { url = "https://files.pythonhosted.org/packages/08/5c/2104299949b9d504baf3f4d35f73dbd14ef31bbd1ddc2c1b66a5b7dfda44/pillow-11.1.0-cp312-cp312-win_amd64.whl", hash = "sha256:a697cd8ba0383bba3d2d3ada02b34ed268cb548b369943cd349007730c92bddf", size = 2626369 },
-    { url = "https://files.pythonhosted.org/packages/37/f3/9b18362206b244167c958984b57c7f70a0289bfb59a530dd8af5f699b910/pillow-11.1.0-cp312-cp312-win_arm64.whl", hash = "sha256:4dd43a78897793f60766563969442020e90eb7847463eca901e41ba186a7d4a5", size = 2375240 },
-    { url = "https://files.pythonhosted.org/packages/b3/31/9ca79cafdce364fd5c980cd3416c20ce1bebd235b470d262f9d24d810184/pillow-11.1.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ae98e14432d458fc3de11a77ccb3ae65ddce70f730e7c76140653048c71bfcbc", size = 3226640 },
-    { url = "https://files.pythonhosted.org/packages/ac/0f/ff07ad45a1f172a497aa393b13a9d81a32e1477ef0e869d030e3c1532521/pillow-11.1.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:cc1331b6d5a6e144aeb5e626f4375f5b7ae9934ba620c0ac6b3e43d5e683a0f0", size = 3101437 },
-    { url = "https://files.pythonhosted.org/packages/08/2f/9906fca87a68d29ec4530be1f893149e0cb64a86d1f9f70a7cfcdfe8ae44/pillow-11.1.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:758e9d4ef15d3560214cddbc97b8ef3ef86ce04d62ddac17ad39ba87e89bd3b1", size = 4326605 },
-    { url = "https://files.pythonhosted.org/packages/b0/0f/f3547ee15b145bc5c8b336401b2d4c9d9da67da9dcb572d7c0d4103d2c69/pillow-11.1.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b523466b1a31d0dcef7c5be1f20b942919b62fd6e9a9be199d035509cbefc0ec", size = 4411173 },
-    { url = "https://files.pythonhosted.org/packages/b1/df/bf8176aa5db515c5de584c5e00df9bab0713548fd780c82a86cba2c2fedb/pillow-11.1.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:9044b5e4f7083f209c4e35aa5dd54b1dd5b112b108648f5c902ad586d4f945c5", size = 4369145 },
-    { url = "https://files.pythonhosted.org/packages/de/7c/7433122d1cfadc740f577cb55526fdc39129a648ac65ce64db2eb7209277/pillow-11.1.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:3764d53e09cdedd91bee65c2527815d315c6b90d7b8b79759cc48d7bf5d4f114", size = 4496340 },
-    { url = "https://files.pythonhosted.org/packages/25/46/dd94b93ca6bd555588835f2504bd90c00d5438fe131cf01cfa0c5131a19d/pillow-11.1.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:31eba6bbdd27dde97b0174ddf0297d7a9c3a507a8a1480e1e60ef914fe23d352", size = 4296906 },
-    { url = "https://files.pythonhosted.org/packages/a8/28/2f9d32014dfc7753e586db9add35b8a41b7a3b46540e965cb6d6bc607bd2/pillow-11.1.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:b5d658fbd9f0d6eea113aea286b21d3cd4d3fd978157cbf2447a6035916506d3", size = 4431759 },
-    { url = "https://files.pythonhosted.org/packages/33/48/19c2cbe7403870fbe8b7737d19eb013f46299cdfe4501573367f6396c775/pillow-11.1.0-cp313-cp313-win32.whl", hash = "sha256:f86d3a7a9af5d826744fabf4afd15b9dfef44fe69a98541f666f66fbb8d3fef9", size = 2291657 },
-    { url = "https://files.pythonhosted.org/packages/3b/ad/285c556747d34c399f332ba7c1a595ba245796ef3e22eae190f5364bb62b/pillow-11.1.0-cp313-cp313-win_amd64.whl", hash = "sha256:593c5fd6be85da83656b93ffcccc2312d2d149d251e98588b14fbc288fd8909c", size = 2626304 },
-    { url = "https://files.pythonhosted.org/packages/e5/7b/ef35a71163bf36db06e9c8729608f78dedf032fc8313d19bd4be5c2588f3/pillow-11.1.0-cp313-cp313-win_arm64.whl", hash = "sha256:11633d58b6ee5733bde153a8dafd25e505ea3d32e261accd388827ee987baf65", size = 2375117 },
-    { url = "https://files.pythonhosted.org/packages/79/30/77f54228401e84d6791354888549b45824ab0ffde659bafa67956303a09f/pillow-11.1.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:70ca5ef3b3b1c4a0812b5c63c57c23b63e53bc38e758b37a951e5bc466449861", size = 3230060 },
-    { url = "https://files.pythonhosted.org/packages/ce/b1/56723b74b07dd64c1010fee011951ea9c35a43d8020acd03111f14298225/pillow-11.1.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:8000376f139d4d38d6851eb149b321a52bb8893a88dae8ee7d95840431977081", size = 3106192 },
-    { url = "https://files.pythonhosted.org/packages/e1/cd/7bf7180e08f80a4dcc6b4c3a0aa9e0b0ae57168562726a05dc8aa8fa66b0/pillow-11.1.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9ee85f0696a17dd28fbcfceb59f9510aa71934b483d1f5601d1030c3c8304f3c", size = 4446805 },
-    { url = "https://files.pythonhosted.org/packages/97/42/87c856ea30c8ed97e8efbe672b58c8304dee0573f8c7cab62ae9e31db6ae/pillow-11.1.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:dd0e081319328928531df7a0e63621caf67652c8464303fd102141b785ef9547", size = 4530623 },
-    { url = "https://files.pythonhosted.org/packages/ff/41/026879e90c84a88e33fb00cc6bd915ac2743c67e87a18f80270dfe3c2041/pillow-11.1.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:e63e4e5081de46517099dc30abe418122f54531a6ae2ebc8680bcd7096860eab", size = 4465191 },
-    { url = "https://files.pythonhosted.org/packages/e5/fb/a7960e838bc5df57a2ce23183bfd2290d97c33028b96bde332a9057834d3/pillow-11.1.0-cp313-cp313t-win32.whl", hash = "sha256:dda60aa465b861324e65a78c9f5cf0f4bc713e4309f83bc387be158b077963d9", size = 2295494 },
-    { url = "https://files.pythonhosted.org/packages/d7/6c/6ec83ee2f6f0fda8d4cf89045c6be4b0373ebfc363ba8538f8c999f63fcd/pillow-11.1.0-cp313-cp313t-win_amd64.whl", hash = "sha256:ad5db5781c774ab9a9b2c4302bbf0c1014960a0a7be63278d13ae6fdf88126fe", size = 2631595 },
-    { url = "https://files.pythonhosted.org/packages/cf/6c/41c21c6c8af92b9fea313aa47c75de49e2f9a467964ee33eb0135d47eb64/pillow-11.1.0-cp313-cp313t-win_arm64.whl", hash = "sha256:67cd427c68926108778a9005f2a04adbd5e67c442ed21d95389fe1d595458756", size = 2377651 },
-    { url = "https://files.pythonhosted.org/packages/9a/1f/9df5ac77491fddd2e36c352d16976dc11fbe6ab842f5df85fd7e31b847b9/pillow-11.1.0-cp39-cp39-macosx_10_10_x86_64.whl", hash = "sha256:bf902d7413c82a1bfa08b06a070876132a5ae6b2388e2712aab3a7cbc02205c6", size = 3229995 },
-    { url = "https://files.pythonhosted.org/packages/a6/62/c7b359e924dca274173b04922ac06aa63614f7e934d132f2fe1d852509aa/pillow-11.1.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:c1eec9d950b6fe688edee07138993e54ee4ae634c51443cfb7c1e7613322718e", size = 3101890 },
-    { url = "https://files.pythonhosted.org/packages/7b/63/136f21340a434de895b62bcf2c386005a8aa24066c4facd619f5e0e9f283/pillow-11.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e275ee4cb11c262bd108ab2081f750db2a1c0b8c12c1897f27b160c8bd57bbc", size = 4310366 },
-    { url = "https://files.pythonhosted.org/packages/f6/46/0bd0ca03d9d1164a7fa33d285ef6d1c438e963d0c8770e4c5b3737ef5abe/pillow-11.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4db853948ce4e718f2fc775b75c37ba2efb6aaea41a1a5fc57f0af59eee774b2", size = 4391582 },
-    { url = "https://files.pythonhosted.org/packages/0c/55/f182db572b28bd833b8e806f933f782ceb2df64c40e4d8bd3d4226a46eca/pillow-11.1.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:ab8a209b8485d3db694fa97a896d96dd6533d63c22829043fd9de627060beade", size = 4350278 },
-    { url = "https://files.pythonhosted.org/packages/75/fb/e330fdbbcbc4744214b5f53b84d9d8a9f4ffbebc2e9c2ac10475386e3296/pillow-11.1.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:54251ef02a2309b5eec99d151ebf5c9904b77976c8abdcbce7891ed22df53884", size = 4471768 },
-    { url = "https://files.pythonhosted.org/packages/eb/51/20ee6c4da4448d7a67ffb720a5fcdb965115a78e211a1f58f9845ae15f86/pillow-11.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:5bb94705aea800051a743aa4874bb1397d4695fb0583ba5e425ee0328757f196", size = 4276549 },
-    { url = "https://files.pythonhosted.org/packages/37/f2/a25c0bdaa6d6fd5cc3d4a6f65b5a7ea46e7af58bee00a98efe0a5af79c58/pillow-11.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:89dbdb3e6e9594d512780a5a1c42801879628b38e3efc7038094430844e271d8", size = 4409350 },
-    { url = "https://files.pythonhosted.org/packages/12/a7/06687947604cd3e47abeea1b78b65d34ffce7feab03cfe0dd985f115dca3/pillow-11.1.0-cp39-cp39-win32.whl", hash = "sha256:e5449ca63da169a2e6068dd0e2fcc8d91f9558aba89ff6d02121ca8ab11e79e5", size = 2291271 },
-    { url = "https://files.pythonhosted.org/packages/21/a6/f51d47675940b5c63b08ff0575b3518428b4acb891f88526fa4ee1edab6f/pillow-11.1.0-cp39-cp39-win_amd64.whl", hash = "sha256:3362c6ca227e65c54bf71a5f88b3d4565ff1bcbc63ae72c34b07bbb1cc59a43f", size = 2625783 },
-    { url = "https://files.pythonhosted.org/packages/95/56/97750bd33e68648fa432dfadcb8ede7624bd905822d42262d34bcebdd9d7/pillow-11.1.0-cp39-cp39-win_arm64.whl", hash = "sha256:b20be51b37a75cc54c2c55def3fa2c65bb94ba859dde241cd0a4fd302de5ae0a", size = 2375193 },
-    { url = "https://files.pythonhosted.org/packages/fa/c5/389961578fb677b8b3244fcd934f720ed25a148b9a5cc81c91bdf59d8588/pillow-11.1.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:8c730dc3a83e5ac137fbc92dfcfe1511ce3b2b5d7578315b63dbbb76f7f51d90", size = 3198345 },
-    { url = "https://files.pythonhosted.org/packages/c4/fa/803c0e50ffee74d4b965229e816af55276eac1d5806712de86f9371858fd/pillow-11.1.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:7d33d2fae0e8b170b6a6c57400e077412240f6f5bb2a342cf1ee512a787942bb", size = 3072938 },
-    { url = "https://files.pythonhosted.org/packages/dc/67/2a3a5f8012b5d8c63fe53958ba906c1b1d0482ebed5618057ef4d22f8076/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a8d65b38173085f24bc07f8b6c505cbb7418009fa1a1fcb111b1f4961814a442", size = 3400049 },
-    { url = "https://files.pythonhosted.org/packages/e5/a0/514f0d317446c98c478d1872497eb92e7cde67003fed74f696441e647446/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:015c6e863faa4779251436db398ae75051469f7c903b043a48f078e437656f83", size = 3422431 },
-    { url = "https://files.pythonhosted.org/packages/cd/00/20f40a935514037b7d3f87adfc87d2c538430ea625b63b3af8c3f5578e72/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:d44ff19eea13ae4acdaaab0179fa68c0c6f2f45d66a4d8ec1eda7d6cecbcc15f", size = 3446208 },
-    { url = "https://files.pythonhosted.org/packages/28/3c/7de681727963043e093c72e6c3348411b0185eab3263100d4490234ba2f6/pillow-11.1.0-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:d3d8da4a631471dfaf94c10c85f5277b1f8e42ac42bade1ac67da4b4a7359b73", size = 3509746 },
-    { url = "https://files.pythonhosted.org/packages/41/67/936f9814bdd74b2dfd4822f1f7725ab5d8ff4103919a1664eb4874c58b2f/pillow-11.1.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:4637b88343166249fe8aa94e7c4a62a180c4b3898283bb5d3d2fd5fe10d8e4e0", size = 2626353 },
+version = "11.3.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/0d/d0d6dea55cd152ce3d6767bb38a8fc10e33796ba4ba210cbab9354b6d238/pillow-11.3.0.tar.gz", hash = "sha256:3828ee7586cd0b2091b6209e5ad53e20d0649bbe87164a459d0676e035e8f523", size = 47113069, upload-time = "2025-07-01T09:16:30.666Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/4c/5d/45a3553a253ac8763f3561371432a90bdbe6000fbdcf1397ffe502aa206c/pillow-11.3.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:1b9c17fd4ace828b3003dfd1e30bff24863e0eb59b535e8f80194d9cc7ecf860", size = 5316554, upload-time = "2025-07-01T09:13:39.342Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/c8/67c12ab069ef586a25a4a79ced553586748fad100c77c0ce59bb4983ac98/pillow-11.3.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:65dc69160114cdd0ca0f35cb434633c75e8e7fad4cf855177a05bf38678f73ad", size = 4686548, upload-time = "2025-07-01T09:13:41.835Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/bd/6741ebd56263390b382ae4c5de02979af7f8bd9807346d068700dd6d5cf9/pillow-11.3.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:7107195ddc914f656c7fc8e4a5e1c25f32e9236ea3ea860f257b0436011fddd0", size = 5859742, upload-time = "2025-07-03T13:09:47.439Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/0b/c412a9e27e1e6a829e6ab6c2dca52dd563efbedf4c9c6aa453d9a9b77359/pillow-11.3.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:cc3e831b563b3114baac7ec2ee86819eb03caa1a2cef0b481a5675b59c4fe23b", size = 7633087, upload-time = "2025-07-03T13:09:51.796Z" },
+    { url = "https://files.pythonhosted.org/packages/59/9d/9b7076aaf30f5dd17e5e5589b2d2f5a5d7e30ff67a171eb686e4eecc2adf/pillow-11.3.0-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f1f182ebd2303acf8c380a54f615ec883322593320a9b00438eb842c1f37ae50", size = 5963350, upload-time = "2025-07-01T09:13:43.865Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/16/1a6bf01fb622fb9cf5c91683823f073f053005c849b1f52ed613afcf8dae/pillow-11.3.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4445fa62e15936a028672fd48c4c11a66d641d2c05726c7ec1f8ba6a572036ae", size = 6631840, upload-time = "2025-07-01T09:13:46.161Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/e6/6ff7077077eb47fde78739e7d570bdcd7c10495666b6afcd23ab56b19a43/pillow-11.3.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:71f511f6b3b91dd543282477be45a033e4845a40278fa8dcdbfdb07109bf18f9", size = 6074005, upload-time = "2025-07-01T09:13:47.829Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/3a/b13f36832ea6d279a697231658199e0a03cd87ef12048016bdcc84131601/pillow-11.3.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:040a5b691b0713e1f6cbe222e0f4f74cd233421e105850ae3b3c0ceda520f42e", size = 6708372, upload-time = "2025-07-01T09:13:52.145Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/e4/61b2e1a7528740efbc70b3d581f33937e38e98ef3d50b05007267a55bcb2/pillow-11.3.0-cp310-cp310-win32.whl", hash = "sha256:89bd777bc6624fe4115e9fac3352c79ed60f3bb18651420635f26e643e3dd1f6", size = 6277090, upload-time = "2025-07-01T09:13:53.915Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/d3/60c781c83a785d6afbd6a326ed4d759d141de43aa7365725cbcd65ce5e54/pillow-11.3.0-cp310-cp310-win_amd64.whl", hash = "sha256:19d2ff547c75b8e3ff46f4d9ef969a06c30ab2d4263a9e287733aa8b2429ce8f", size = 6985988, upload-time = "2025-07-01T09:13:55.699Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/28/4f4a0203165eefb3763939c6789ba31013a2e90adffb456610f30f613850/pillow-11.3.0-cp310-cp310-win_arm64.whl", hash = "sha256:819931d25e57b513242859ce1876c58c59dc31587847bf74cfe06b2e0cb22d2f", size = 2422899, upload-time = "2025-07-01T09:13:57.497Z" },
+    { url = "https://files.pythonhosted.org/packages/db/26/77f8ed17ca4ffd60e1dcd220a6ec6d71210ba398cfa33a13a1cd614c5613/pillow-11.3.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:1cd110edf822773368b396281a2293aeb91c90a2db00d78ea43e7e861631b722", size = 5316531, upload-time = "2025-07-01T09:13:59.203Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/39/ee475903197ce709322a17a866892efb560f57900d9af2e55f86db51b0a5/pillow-11.3.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:9c412fddd1b77a75aa904615ebaa6001f169b26fd467b4be93aded278266b288", size = 4686560, upload-time = "2025-07-01T09:14:01.101Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/90/442068a160fd179938ba55ec8c97050a612426fae5ec0a764e345839f76d/pillow-11.3.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:7d1aa4de119a0ecac0a34a9c8bde33f34022e2e8f99104e47a3ca392fd60e37d", size = 5870978, upload-time = "2025-07-03T13:09:55.638Z" },
+    { url = "https://files.pythonhosted.org/packages/13/92/dcdd147ab02daf405387f0218dcf792dc6dd5b14d2573d40b4caeef01059/pillow-11.3.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:91da1d88226663594e3f6b4b8c3c8d85bd504117d043740a8e0ec449087cc494", size = 7641168, upload-time = "2025-07-03T13:10:00.37Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/db/839d6ba7fd38b51af641aa904e2960e7a5644d60ec754c046b7d2aee00e5/pillow-11.3.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:643f189248837533073c405ec2f0bb250ba54598cf80e8c1e043381a60632f58", size = 5973053, upload-time = "2025-07-01T09:14:04.491Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/2f/d7675ecae6c43e9f12aa8d58b6012683b20b6edfbdac7abcb4e6af7a3784/pillow-11.3.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:106064daa23a745510dabce1d84f29137a37224831d88eb4ce94bb187b1d7e5f", size = 6640273, upload-time = "2025-07-01T09:14:06.235Z" },
+    { url = "https://files.pythonhosted.org/packages/45/ad/931694675ede172e15b2ff03c8144a0ddaea1d87adb72bb07655eaffb654/pillow-11.3.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:cd8ff254faf15591e724dc7c4ddb6bf4793efcbe13802a4ae3e863cd300b493e", size = 6082043, upload-time = "2025-07-01T09:14:07.978Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/04/ba8f2b11fc80d2dd462d7abec16351b45ec99cbbaea4387648a44190351a/pillow-11.3.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:932c754c2d51ad2b2271fd01c3d121daaa35e27efae2a616f77bf164bc0b3e94", size = 6715516, upload-time = "2025-07-01T09:14:10.233Z" },
+    { url = "https://files.pythonhosted.org/packages/48/59/8cd06d7f3944cc7d892e8533c56b0acb68399f640786313275faec1e3b6f/pillow-11.3.0-cp311-cp311-win32.whl", hash = "sha256:b4b8f3efc8d530a1544e5962bd6b403d5f7fe8b9e08227c6b255f98ad82b4ba0", size = 6274768, upload-time = "2025-07-01T09:14:11.921Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/cc/29c0f5d64ab8eae20f3232da8f8571660aa0ab4b8f1331da5c2f5f9a938e/pillow-11.3.0-cp311-cp311-win_amd64.whl", hash = "sha256:1a992e86b0dd7aeb1f053cd506508c0999d710a8f07b4c791c63843fc6a807ac", size = 6986055, upload-time = "2025-07-01T09:14:13.623Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/df/90bd886fabd544c25addd63e5ca6932c86f2b701d5da6c7839387a076b4a/pillow-11.3.0-cp311-cp311-win_arm64.whl", hash = "sha256:30807c931ff7c095620fe04448e2c2fc673fcbb1ffe2a7da3fb39613489b1ddd", size = 2423079, upload-time = "2025-07-01T09:14:15.268Z" },
+    { url = "https://files.pythonhosted.org/packages/40/fe/1bc9b3ee13f68487a99ac9529968035cca2f0a51ec36892060edcc51d06a/pillow-11.3.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:fdae223722da47b024b867c1ea0be64e0df702c5e0a60e27daad39bf960dd1e4", size = 5278800, upload-time = "2025-07-01T09:14:17.648Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/32/7e2ac19b5713657384cec55f89065fb306b06af008cfd87e572035b27119/pillow-11.3.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:921bd305b10e82b4d1f5e802b6850677f965d8394203d182f078873851dada69", size = 4686296, upload-time = "2025-07-01T09:14:19.828Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/1e/b9e12bbe6e4c2220effebc09ea0923a07a6da1e1f1bfbc8d7d29a01ce32b/pillow-11.3.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:eb76541cba2f958032d79d143b98a3a6b3ea87f0959bbe256c0b5e416599fd5d", size = 5871726, upload-time = "2025-07-03T13:10:04.448Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/33/e9200d2bd7ba00dc3ddb78df1198a6e80d7669cce6c2bdbeb2530a74ec58/pillow-11.3.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:67172f2944ebba3d4a7b54f2e95c786a3a50c21b88456329314caaa28cda70f6", size = 7644652, upload-time = "2025-07-03T13:10:10.391Z" },
+    { url = "https://files.pythonhosted.org/packages/41/f1/6f2427a26fc683e00d985bc391bdd76d8dd4e92fac33d841127eb8fb2313/pillow-11.3.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:97f07ed9f56a3b9b5f49d3661dc9607484e85c67e27f3e8be2c7d28ca032fec7", size = 5977787, upload-time = "2025-07-01T09:14:21.63Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/c9/06dd4a38974e24f932ff5f98ea3c546ce3f8c995d3f0985f8e5ba48bba19/pillow-11.3.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:676b2815362456b5b3216b4fd5bd89d362100dc6f4945154ff172e206a22c024", size = 6645236, upload-time = "2025-07-01T09:14:23.321Z" },
+    { url = "https://files.pythonhosted.org/packages/40/e7/848f69fb79843b3d91241bad658e9c14f39a32f71a301bcd1d139416d1be/pillow-11.3.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3e184b2f26ff146363dd07bde8b711833d7b0202e27d13540bfe2e35a323a809", size = 6086950, upload-time = "2025-07-01T09:14:25.237Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/1a/7cff92e695a2a29ac1958c2a0fe4c0b2393b60aac13b04a4fe2735cad52d/pillow-11.3.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:6be31e3fc9a621e071bc17bb7de63b85cbe0bfae91bb0363c893cbe67247780d", size = 6723358, upload-time = "2025-07-01T09:14:27.053Z" },
+    { url = "https://files.pythonhosted.org/packages/26/7d/73699ad77895f69edff76b0f332acc3d497f22f5d75e5360f78cbcaff248/pillow-11.3.0-cp312-cp312-win32.whl", hash = "sha256:7b161756381f0918e05e7cb8a371fff367e807770f8fe92ecb20d905d0e1c149", size = 6275079, upload-time = "2025-07-01T09:14:30.104Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/ce/e7dfc873bdd9828f3b6e5c2bbb74e47a98ec23cc5c74fc4e54462f0d9204/pillow-11.3.0-cp312-cp312-win_amd64.whl", hash = "sha256:a6444696fce635783440b7f7a9fc24b3ad10a9ea3f0ab66c5905be1c19ccf17d", size = 6986324, upload-time = "2025-07-01T09:14:31.899Z" },
+    { url = "https://files.pythonhosted.org/packages/16/8f/b13447d1bf0b1f7467ce7d86f6e6edf66c0ad7cf44cf5c87a37f9bed9936/pillow-11.3.0-cp312-cp312-win_arm64.whl", hash = "sha256:2aceea54f957dd4448264f9bf40875da0415c83eb85f55069d89c0ed436e3542", size = 2423067, upload-time = "2025-07-01T09:14:33.709Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/8e/9c089f01677d1264ab8648352dcb7773f37da6ad002542760c80107da816/pillow-11.3.0-cp39-cp39-macosx_10_10_x86_64.whl", hash = "sha256:48d254f8a4c776de343051023eb61ffe818299eeac478da55227d96e241de53f", size = 5316478, upload-time = "2025-07-01T09:15:52.209Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/a9/5749930caf674695867eb56a581e78eb5f524b7583ff10b01b6e5048acb3/pillow-11.3.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:7aee118e30a4cf54fdd873bd3a29de51e29105ab11f9aad8c32123f58c8f8081", size = 4686522, upload-time = "2025-07-01T09:15:54.162Z" },
+    { url = "https://files.pythonhosted.org/packages/43/46/0b85b763eb292b691030795f9f6bb6fcaf8948c39413c81696a01c3577f7/pillow-11.3.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:23cff760a9049c502721bdb743a7cb3e03365fafcdfc2ef9784610714166e5a4", size = 5853376, upload-time = "2025-07-03T13:11:01.066Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/c6/1a230ec0067243cbd60bc2dad5dc3ab46a8a41e21c15f5c9b52b26873069/pillow-11.3.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:6359a3bc43f57d5b375d1ad54a0074318a0844d11b76abccf478c37c986d3cfc", size = 7626020, upload-time = "2025-07-03T13:11:06.479Z" },
+    { url = "https://files.pythonhosted.org/packages/63/dd/f296c27ffba447bfad76c6a0c44c1ea97a90cb9472b9304c94a732e8dbfb/pillow-11.3.0-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:092c80c76635f5ecb10f3f83d76716165c96f5229addbd1ec2bdbbda7d496e06", size = 5956732, upload-time = "2025-07-01T09:15:56.111Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/a0/98a3630f0b57f77bae67716562513d3032ae70414fcaf02750279c389a9e/pillow-11.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:cadc9e0ea0a2431124cde7e1697106471fc4c1da01530e679b2391c37d3fbb3a", size = 6624404, upload-time = "2025-07-01T09:15:58.245Z" },
+    { url = "https://files.pythonhosted.org/packages/de/e6/83dfba5646a290edd9a21964da07674409e410579c341fc5b8f7abd81620/pillow-11.3.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:6a418691000f2a418c9135a7cf0d797c1bb7d9a485e61fe8e7722845b95ef978", size = 6067760, upload-time = "2025-07-01T09:16:00.003Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/41/15ab268fe6ee9a2bc7391e2bbb20a98d3974304ab1a406a992dcb297a370/pillow-11.3.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:97afb3a00b65cc0804d1c7abddbf090a81eaac02768af58cbdcaaa0a931e0b6d", size = 6700534, upload-time = "2025-07-01T09:16:02.29Z" },
+    { url = "https://files.pythonhosted.org/packages/64/79/6d4f638b288300bed727ff29f2a3cb63db054b33518a95f27724915e3fbc/pillow-11.3.0-cp39-cp39-win32.whl", hash = "sha256:ea944117a7974ae78059fcc1800e5d3295172bb97035c0c1d9345fca1419da71", size = 6277091, upload-time = "2025-07-01T09:16:04.4Z" },
+    { url = "https://files.pythonhosted.org/packages/46/05/4106422f45a05716fd34ed21763f8ec182e8ea00af6e9cb05b93a247361a/pillow-11.3.0-cp39-cp39-win_amd64.whl", hash = "sha256:e5c5858ad8ec655450a7c7df532e9842cf8df7cc349df7225c60d5d348c8aada", size = 6986091, upload-time = "2025-07-01T09:16:06.342Z" },
+    { url = "https://files.pythonhosted.org/packages/63/c6/287fd55c2c12761d0591549d48885187579b7c257bef0c6660755b0b59ae/pillow-11.3.0-cp39-cp39-win_arm64.whl", hash = "sha256:6abdbfd3aea42be05702a8dd98832329c167ee84400a1d1f61ab11437f1717eb", size = 2422632, upload-time = "2025-07-01T09:16:08.142Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/8b/209bd6b62ce8367f47e68a218bffac88888fdf2c9fcf1ecadc6c3ec1ebc7/pillow-11.3.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:3cee80663f29e3843b68199b9d6f4f54bd1d4a6b59bdd91bceefc51238bcb967", size = 5270556, upload-time = "2025-07-01T09:16:09.961Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/e6/231a0b76070c2cfd9e260a7a5b504fb72da0a95279410fa7afd99d9751d6/pillow-11.3.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:b5f56c3f344f2ccaf0dd875d3e180f631dc60a51b314295a3e681fe8cf851fbe", size = 4654625, upload-time = "2025-07-01T09:16:11.913Z" },
+    { url = "https://files.pythonhosted.org/packages/13/f4/10cf94fda33cb12765f2397fc285fa6d8eb9c29de7f3185165b702fc7386/pillow-11.3.0-pp310-pypy310_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:e67d793d180c9df62f1f40aee3accca4829d3794c95098887edc18af4b8b780c", size = 4874207, upload-time = "2025-07-03T13:11:10.201Z" },
+    { url = "https://files.pythonhosted.org/packages/72/c9/583821097dc691880c92892e8e2d41fe0a5a3d6021f4963371d2f6d57250/pillow-11.3.0-pp310-pypy310_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d000f46e2917c705e9fb93a3606ee4a819d1e3aa7a9b442f6444f07e77cf5e25", size = 6583939, upload-time = "2025-07-03T13:11:15.68Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/8e/5c9d410f9217b12320efc7c413e72693f48468979a013ad17fd690397b9a/pillow-11.3.0-pp310-pypy310_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:527b37216b6ac3a12d7838dc3bd75208ec57c1c6d11ef01902266a5a0c14fc27", size = 4957166, upload-time = "2025-07-01T09:16:13.74Z" },
+    { url = "https://files.pythonhosted.org/packages/62/bb/78347dbe13219991877ffb3a91bf09da8317fbfcd4b5f9140aeae020ad71/pillow-11.3.0-pp310-pypy310_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:be5463ac478b623b9dd3937afd7fb7ab3d79dd290a28e2b6df292dc75063eb8a", size = 5581482, upload-time = "2025-07-01T09:16:16.107Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/28/1000353d5e61498aaeaaf7f1e4b49ddb05f2c6575f9d4f9f914a3538b6e1/pillow-11.3.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:8dc70ca24c110503e16918a658b869019126ecfe03109b754c402daff12b3d9f", size = 6984596, upload-time = "2025-07-01T09:16:18.07Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/e3/6fa84033758276fb31da12e5fb66ad747ae83b93c67af17f8c6ff4cc8f34/pillow-11.3.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:7c8ec7a017ad1bd562f93dbd8505763e688d388cde6e4a010ae1486916e713e6", size = 5270566, upload-time = "2025-07-01T09:16:19.801Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/ee/e8d2e1ab4892970b561e1ba96cbd59c0d28cf66737fc44abb2aec3795a4e/pillow-11.3.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:9ab6ae226de48019caa8074894544af5b53a117ccb9d3b3dcb2871464c829438", size = 4654618, upload-time = "2025-07-01T09:16:21.818Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/6d/17f80f4e1f0761f02160fc433abd4109fa1548dcfdca46cfdadaf9efa565/pillow-11.3.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:fe27fb049cdcca11f11a7bfda64043c37b30e6b91f10cb5bab275806c32f6ab3", size = 4874248, upload-time = "2025-07-03T13:11:20.738Z" },
+    { url = "https://files.pythonhosted.org/packages/de/5f/c22340acd61cef960130585bbe2120e2fd8434c214802f07e8c03596b17e/pillow-11.3.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:465b9e8844e3c3519a983d58b80be3f668e2a7a5db97f2784e7079fbc9f9822c", size = 6583963, upload-time = "2025-07-03T13:11:26.283Z" },
+    { url = "https://files.pythonhosted.org/packages/31/5e/03966aedfbfcbb4d5f8aa042452d3361f325b963ebbadddac05b122e47dd/pillow-11.3.0-pp311-pypy311_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5418b53c0d59b3824d05e029669efa023bbef0f3e92e75ec8428f3799487f361", size = 4957170, upload-time = "2025-07-01T09:16:23.762Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/2d/e082982aacc927fc2cab48e1e731bdb1643a1406acace8bed0900a61464e/pillow-11.3.0-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:504b6f59505f08ae014f724b6207ff6222662aab5cc9542577fb084ed0676ac7", size = 5581505, upload-time = "2025-07-01T09:16:25.593Z" },
+    { url = "https://files.pythonhosted.org/packages/34/e7/ae39f538fd6844e982063c3a5e4598b8ced43b9633baa3a85ef33af8c05c/pillow-11.3.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:c84d689db21a1c397d001aa08241044aa2069e7587b398c8cc63020390b1c1b8", size = 6984598, upload-time = "2025-07-01T09:16:27.732Z" },
 ]
 
 [[package]]
 name = "platformdirs"
-version = "4.3.6"
+version = "4.4.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/13/fc/128cc9cb8f03208bdbf93d3aa862e16d376844a14f9a0ce5cf4507372de4/platformdirs-4.3.6.tar.gz", hash = "sha256:357fb2acbc885b0419afd3ce3ed34564c13c9b95c89360cd9563f73aa5e2b907", size = 21302 }
+sdist = { url = "https://files.pythonhosted.org/packages/23/e8/21db9c9987b0e728855bd57bff6984f67952bea55d6f75e055c46b5383e8/platformdirs-4.4.0.tar.gz", hash = "sha256:ca753cf4d81dc309bc67b0ea38fd15dc97bc30ce419a7f58d13eb3bf14c4febf", size = 21634, upload-time = "2025-08-26T14:32:04.268Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/3c/a6/bc1012356d8ece4d66dd75c4b9fc6c1f6650ddd5991e421177d9f8f671be/platformdirs-4.3.6-py3-none-any.whl", hash = "sha256:73e575e1408ab8103900836b97580d5307456908a03e92031bab39e4554cc3fb", size = 18439 },
+    { url = "https://files.pythonhosted.org/packages/40/4b/2028861e724d3bd36227adfa20d3fd24c3fc6d52032f4a93c133be5d17ce/platformdirs-4.4.0-py3-none-any.whl", hash = "sha256:abd01743f24e5287cd7a5db3752faf1a2d65353f38ec26d98e25a6db65958c85", size = 18654, upload-time = "2025-08-26T14:32:02.735Z" },
 ]
 
 [[package]]
 name = "pluggy"
-version = "1.5.0"
+version = "1.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f9/e2/3e91f31a7d2b083fe6ef3fa267035b518369d9511ffab804f839851d2779/pluggy-1.6.0.tar.gz", hash = "sha256:7dcc130b76258d33b90f61b658791dede3486c3e6bfb003ee5c9bfb396dd22f3", size = 69412, upload-time = "2025-05-15T12:30:07.975Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/54/20/4d324d65cc6d9205fabedc306948156824eb9f0ee1633355a8f7ec5c66bf/pluggy-1.6.0-py3-none-any.whl", hash = "sha256:e920276dd6813095e9377c0bc5566d94c932c33b27a3e3945d8389c374dd4746", size = 20538, upload-time = "2025-05-15T12:30:06.134Z" },
+]
+
+[[package]]
+name = "ply"
+version = "3.11"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/96/2d/02d4312c973c6050a18b314a5ad0b3210edb65a906f868e31c111dede4a6/pluggy-1.5.0.tar.gz", hash = "sha256:2cffa88e94fdc978c4c574f15f9e59b7f4201d439195c3715ca9e2486f1d0cf1", size = 67955 }
+sdist = { url = "https://files.pythonhosted.org/packages/e5/69/882ee5c9d017149285cab114ebeab373308ef0f874fcdac9beb90e0ac4da/ply-3.11.tar.gz", hash = "sha256:00c7c1aaa88358b9c765b6d3000c6eec0ba42abca5351b095321aef446081da3", size = 159130, upload-time = "2018-02-15T19:01:31.097Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/88/5f/e351af9a41f866ac3f1fac4ca0613908d9a41741cfcf2228f4ad853b697d/pluggy-1.5.0-py3-none-any.whl", hash = "sha256:44e1ad92c8ca002de6377e165f3e0f1be63266ab4d554740532335b9d75ea669", size = 20556 },
+    { url = "https://files.pythonhosted.org/packages/a3/58/35da89ee790598a0700ea49b2a66594140f44dec458c07e8e3d4979137fc/ply-3.11-py2.py3-none-any.whl", hash = "sha256:096f9b8350b65ebd2fd1346b12452efe5b9607f7482813ffca50c22722a807ce", size = 49567, upload-time = "2018-02-15T19:01:27.172Z" },
 ]
 
 [[package]]
 name = "pre-commit"
-version = "4.0.1"
+version = "4.3.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "cfgv" },
@@ -2216,343 +3024,326 @@ dependencies = [
     { name = "pyyaml" },
     { name = "virtualenv" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/2e/c8/e22c292035f1bac8b9f5237a2622305bc0304e776080b246f3df57c4ff9f/pre_commit-4.0.1.tar.gz", hash = "sha256:80905ac375958c0444c65e9cebebd948b3cdb518f335a091a670a89d652139d2", size = 191678 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/16/8f/496e10d51edd6671ebe0432e33ff800aa86775d2d147ce7d43389324a525/pre_commit-4.0.1-py2.py3-none-any.whl", hash = "sha256:efde913840816312445dc98787724647c65473daefe420785f885e8ed9a06878", size = 218713 },
-]
-
-[[package]]
-name = "prometheus-client"
-version = "0.21.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/62/14/7d0f567991f3a9af8d1cd4f619040c93b68f09a02b6d0b6ab1b2d1ded5fe/prometheus_client-0.21.1.tar.gz", hash = "sha256:252505a722ac04b0456be05c05f75f45d760c2911ffc45f2a06bcaed9f3ae3fb", size = 78551 }
+sdist = { url = "https://files.pythonhosted.org/packages/ff/29/7cf5bbc236333876e4b41f56e06857a87937ce4bf91e117a6991a2dbb02a/pre_commit-4.3.0.tar.gz", hash = "sha256:499fe450cc9d42e9d58e606262795ecb64dd05438943c62b66f6a8673da30b16", size = 193792, upload-time = "2025-08-09T18:56:14.651Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ff/c2/ab7d37426c179ceb9aeb109a85cda8948bb269b7561a0be870cc656eefe4/prometheus_client-0.21.1-py3-none-any.whl", hash = "sha256:594b45c410d6f4f8888940fe80b5cc2521b305a1fafe1c58609ef715a001f301", size = 54682 },
+    { url = "https://files.pythonhosted.org/packages/5b/a5/987a405322d78a73b66e39e4a90e4ef156fd7141bf71df987e50717c321b/pre_commit-4.3.0-py2.py3-none-any.whl", hash = "sha256:2b0747ad7e6e967169136edffee14c16e148a778a54e4f967921aa1ebf2308d8", size = 220965, upload-time = "2025-08-09T18:56:13.192Z" },
 ]
 
 [[package]]
-name = "prometheus-fastapi-instrumentator"
-version = "7.0.2"
+name = "prompt-toolkit"
+version = "3.0.52"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "prometheus-client" },
-    { name = "starlette" },
+    { name = "wcwidth" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/c4/01/e3ccb464ba9bf6c001ad85652e6fc7e63098c2685275a682cabc8e7871b8/prometheus_fastapi_instrumentator-7.0.2.tar.gz", hash = "sha256:8a4d8fb13dbe19d2882ac6af9ce236e4e1f98dc48e3fa44fe88d8e23ac3c953f", size = 19844 }
+sdist = { url = "https://files.pythonhosted.org/packages/a1/96/06e01a7b38dce6fe1db213e061a4602dd6032a8a97ef6c1a862537732421/prompt_toolkit-3.0.52.tar.gz", hash = "sha256:28cde192929c8e7321de85de1ddbe736f1375148b02f2e17edd840042b1be855", size = 434198, upload-time = "2025-08-27T15:24:02.057Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f1/f7/a67e804853d05b3f1f5def0ebd662b9f48dbcfe0b452f0aa5d4c183a6f86/prometheus_fastapi_instrumentator-7.0.2-py3-none-any.whl", hash = "sha256:975e39992acb7a112758ff13ba95317e6c54d1bbf605f9156f31ac9f2800c32d", size = 18998 },
+    { url = "https://files.pythonhosted.org/packages/84/03/0d3ce49e2505ae70cf43bc5bb3033955d2fc9f932163e84dc0779cc47f48/prompt_toolkit-3.0.52-py3-none-any.whl", hash = "sha256:9aac639a3bbd33284347de5ad8d68ecc044b91a762dc39b7c21095fcd6a19955", size = 391431, upload-time = "2025-08-27T15:23:59.498Z" },
 ]
 
 [[package]]
 name = "propcache"
-version = "0.2.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/20/c8/2a13f78d82211490855b2fb303b6721348d0787fdd9a12ac46d99d3acde1/propcache-0.2.1.tar.gz", hash = "sha256:3f77ce728b19cb537714499928fe800c3dda29e8d9428778fc7c186da4c09a64", size = 41735 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/a7/a5/0ea64c9426959ef145a938e38c832fc551843481d356713ececa9a8a64e8/propcache-0.2.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:6b3f39a85d671436ee3d12c017f8fdea38509e4f25b28eb25877293c98c243f6", size = 79296 },
-    { url = "https://files.pythonhosted.org/packages/76/5a/916db1aba735f55e5eca4733eea4d1973845cf77dfe67c2381a2ca3ce52d/propcache-0.2.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:39d51fbe4285d5db5d92a929e3e21536ea3dd43732c5b177c7ef03f918dff9f2", size = 45622 },
-    { url = "https://files.pythonhosted.org/packages/2d/62/685d3cf268b8401ec12b250b925b21d152b9d193b7bffa5fdc4815c392c2/propcache-0.2.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6445804cf4ec763dc70de65a3b0d9954e868609e83850a47ca4f0cb64bd79fea", size = 45133 },
-    { url = "https://files.pythonhosted.org/packages/4d/3d/31c9c29ee7192defc05aa4d01624fd85a41cf98e5922aaed206017329944/propcache-0.2.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f9479aa06a793c5aeba49ce5c5692ffb51fcd9a7016e017d555d5e2b0045d212", size = 204809 },
-    { url = "https://files.pythonhosted.org/packages/10/a1/e4050776f4797fc86140ac9a480d5dc069fbfa9d499fe5c5d2fa1ae71f07/propcache-0.2.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d9631c5e8b5b3a0fda99cb0d29c18133bca1e18aea9effe55adb3da1adef80d3", size = 219109 },
-    { url = "https://files.pythonhosted.org/packages/c9/c0/e7ae0df76343d5e107d81e59acc085cea5fd36a48aa53ef09add7503e888/propcache-0.2.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3156628250f46a0895f1f36e1d4fbe062a1af8718ec3ebeb746f1d23f0c5dc4d", size = 217368 },
-    { url = "https://files.pythonhosted.org/packages/fc/e1/e0a2ed6394b5772508868a977d3238f4afb2eebaf9976f0b44a8d347ad63/propcache-0.2.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b6fb63ae352e13748289f04f37868099e69dba4c2b3e271c46061e82c745634", size = 205124 },
-    { url = "https://files.pythonhosted.org/packages/50/c1/e388c232d15ca10f233c778bbdc1034ba53ede14c207a72008de45b2db2e/propcache-0.2.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:887d9b0a65404929641a9fabb6452b07fe4572b269d901d622d8a34a4e9043b2", size = 195463 },
-    { url = "https://files.pythonhosted.org/packages/0a/fd/71b349b9def426cc73813dbd0f33e266de77305e337c8c12bfb0a2a82bfb/propcache-0.2.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:a96dc1fa45bd8c407a0af03b2d5218392729e1822b0c32e62c5bf7eeb5fb3958", size = 198358 },
-    { url = "https://files.pythonhosted.org/packages/02/f2/d7c497cd148ebfc5b0ae32808e6c1af5922215fe38c7a06e4e722fe937c8/propcache-0.2.1-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:a7e65eb5c003a303b94aa2c3852ef130230ec79e349632d030e9571b87c4698c", size = 195560 },
-    { url = "https://files.pythonhosted.org/packages/bb/57/f37041bbe5e0dfed80a3f6be2612a3a75b9cfe2652abf2c99bef3455bbad/propcache-0.2.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:999779addc413181912e984b942fbcc951be1f5b3663cd80b2687758f434c583", size = 196895 },
-    { url = "https://files.pythonhosted.org/packages/83/36/ae3cc3e4f310bff2f064e3d2ed5558935cc7778d6f827dce74dcfa125304/propcache-0.2.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:19a0f89a7bb9d8048d9c4370c9c543c396e894c76be5525f5e1ad287f1750ddf", size = 207124 },
-    { url = "https://files.pythonhosted.org/packages/8c/c4/811b9f311f10ce9d31a32ff14ce58500458443627e4df4ae9c264defba7f/propcache-0.2.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:1ac2f5fe02fa75f56e1ad473f1175e11f475606ec9bd0be2e78e4734ad575034", size = 210442 },
-    { url = "https://files.pythonhosted.org/packages/18/dd/a1670d483a61ecac0d7fc4305d91caaac7a8fc1b200ea3965a01cf03bced/propcache-0.2.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:574faa3b79e8ebac7cb1d7930f51184ba1ccf69adfdec53a12f319a06030a68b", size = 203219 },
-    { url = "https://files.pythonhosted.org/packages/f9/2d/30ced5afde41b099b2dc0c6573b66b45d16d73090e85655f1a30c5a24e07/propcache-0.2.1-cp310-cp310-win32.whl", hash = "sha256:03ff9d3f665769b2a85e6157ac8b439644f2d7fd17615a82fa55739bc97863f4", size = 40313 },
-    { url = "https://files.pythonhosted.org/packages/23/84/bd9b207ac80da237af77aa6e153b08ffa83264b1c7882495984fcbfcf85c/propcache-0.2.1-cp310-cp310-win_amd64.whl", hash = "sha256:2d3af2e79991102678f53e0dbf4c35de99b6b8b58f29a27ca0325816364caaba", size = 44428 },
-    { url = "https://files.pythonhosted.org/packages/bc/0f/2913b6791ebefb2b25b4efd4bb2299c985e09786b9f5b19184a88e5778dd/propcache-0.2.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:1ffc3cca89bb438fb9c95c13fc874012f7b9466b89328c3c8b1aa93cdcfadd16", size = 79297 },
-    { url = "https://files.pythonhosted.org/packages/cf/73/af2053aeccd40b05d6e19058419ac77674daecdd32478088b79375b9ab54/propcache-0.2.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:f174bbd484294ed9fdf09437f889f95807e5f229d5d93588d34e92106fbf6717", size = 45611 },
-    { url = "https://files.pythonhosted.org/packages/3c/09/8386115ba7775ea3b9537730e8cf718d83bbf95bffe30757ccf37ec4e5da/propcache-0.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:70693319e0b8fd35dd863e3e29513875eb15c51945bf32519ef52927ca883bc3", size = 45146 },
-    { url = "https://files.pythonhosted.org/packages/03/7a/793aa12f0537b2e520bf09f4c6833706b63170a211ad042ca71cbf79d9cb/propcache-0.2.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b480c6a4e1138e1aa137c0079b9b6305ec6dcc1098a8ca5196283e8a49df95a9", size = 232136 },
-    { url = "https://files.pythonhosted.org/packages/f1/38/b921b3168d72111769f648314100558c2ea1d52eb3d1ba7ea5c4aa6f9848/propcache-0.2.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d27b84d5880f6d8aa9ae3edb253c59d9f6642ffbb2c889b78b60361eed449787", size = 239706 },
-    { url = "https://files.pythonhosted.org/packages/14/29/4636f500c69b5edea7786db3c34eb6166f3384b905665ce312a6e42c720c/propcache-0.2.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:857112b22acd417c40fa4595db2fe28ab900c8c5fe4670c7989b1c0230955465", size = 238531 },
-    { url = "https://files.pythonhosted.org/packages/85/14/01fe53580a8e1734ebb704a3482b7829a0ef4ea68d356141cf0994d9659b/propcache-0.2.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cf6c4150f8c0e32d241436526f3c3f9cbd34429492abddbada2ffcff506c51af", size = 231063 },
-    { url = "https://files.pythonhosted.org/packages/33/5c/1d961299f3c3b8438301ccfbff0143b69afcc30c05fa28673cface692305/propcache-0.2.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:66d4cfda1d8ed687daa4bc0274fcfd5267873db9a5bc0418c2da19273040eeb7", size = 220134 },
-    { url = "https://files.pythonhosted.org/packages/00/d0/ed735e76db279ba67a7d3b45ba4c654e7b02bc2f8050671ec365d8665e21/propcache-0.2.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:c2f992c07c0fca81655066705beae35fc95a2fa7366467366db627d9f2ee097f", size = 220009 },
-    { url = "https://files.pythonhosted.org/packages/75/90/ee8fab7304ad6533872fee982cfff5a53b63d095d78140827d93de22e2d4/propcache-0.2.1-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:4a571d97dbe66ef38e472703067021b1467025ec85707d57e78711c085984e54", size = 212199 },
-    { url = "https://files.pythonhosted.org/packages/eb/ec/977ffaf1664f82e90737275873461695d4c9407d52abc2f3c3e24716da13/propcache-0.2.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:bb6178c241278d5fe853b3de743087be7f5f4c6f7d6d22a3b524d323eecec505", size = 214827 },
-    { url = "https://files.pythonhosted.org/packages/57/48/031fb87ab6081764054821a71b71942161619549396224cbb242922525e8/propcache-0.2.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:ad1af54a62ffe39cf34db1aa6ed1a1873bd548f6401db39d8e7cd060b9211f82", size = 228009 },
-    { url = "https://files.pythonhosted.org/packages/1a/06/ef1390f2524850838f2390421b23a8b298f6ce3396a7cc6d39dedd4047b0/propcache-0.2.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:e7048abd75fe40712005bcfc06bb44b9dfcd8e101dda2ecf2f5aa46115ad07ca", size = 231638 },
-    { url = "https://files.pythonhosted.org/packages/38/2a/101e6386d5a93358395da1d41642b79c1ee0f3b12e31727932b069282b1d/propcache-0.2.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:160291c60081f23ee43d44b08a7e5fb76681221a8e10b3139618c5a9a291b84e", size = 222788 },
-    { url = "https://files.pythonhosted.org/packages/db/81/786f687951d0979007e05ad9346cd357e50e3d0b0f1a1d6074df334b1bbb/propcache-0.2.1-cp311-cp311-win32.whl", hash = "sha256:819ce3b883b7576ca28da3861c7e1a88afd08cc8c96908e08a3f4dd64a228034", size = 40170 },
-    { url = "https://files.pythonhosted.org/packages/cf/59/7cc7037b295d5772eceb426358bb1b86e6cab4616d971bd74275395d100d/propcache-0.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:edc9fc7051e3350643ad929df55c451899bb9ae6d24998a949d2e4c87fb596d3", size = 44404 },
-    { url = "https://files.pythonhosted.org/packages/4c/28/1d205fe49be8b1b4df4c50024e62480a442b1a7b818e734308bb0d17e7fb/propcache-0.2.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:081a430aa8d5e8876c6909b67bd2d937bfd531b0382d3fdedb82612c618bc41a", size = 79588 },
-    { url = "https://files.pythonhosted.org/packages/21/ee/fc4d893f8d81cd4971affef2a6cb542b36617cd1d8ce56b406112cb80bf7/propcache-0.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:d2ccec9ac47cf4e04897619c0e0c1a48c54a71bdf045117d3a26f80d38ab1fb0", size = 45825 },
-    { url = "https://files.pythonhosted.org/packages/4a/de/bbe712f94d088da1d237c35d735f675e494a816fd6f54e9db2f61ef4d03f/propcache-0.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:14d86fe14b7e04fa306e0c43cdbeebe6b2c2156a0c9ce56b815faacc193e320d", size = 45357 },
-    { url = "https://files.pythonhosted.org/packages/7f/14/7ae06a6cf2a2f1cb382586d5a99efe66b0b3d0c6f9ac2f759e6f7af9d7cf/propcache-0.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:049324ee97bb67285b49632132db351b41e77833678432be52bdd0289c0e05e4", size = 241869 },
-    { url = "https://files.pythonhosted.org/packages/cc/59/227a78be960b54a41124e639e2c39e8807ac0c751c735a900e21315f8c2b/propcache-0.2.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1cd9a1d071158de1cc1c71a26014dcdfa7dd3d5f4f88c298c7f90ad6f27bb46d", size = 247884 },
-    { url = "https://files.pythonhosted.org/packages/84/58/f62b4ffaedf88dc1b17f04d57d8536601e4e030feb26617228ef930c3279/propcache-0.2.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:98110aa363f1bb4c073e8dcfaefd3a5cea0f0834c2aab23dda657e4dab2f53b5", size = 248486 },
-    { url = "https://files.pythonhosted.org/packages/1c/07/ebe102777a830bca91bbb93e3479cd34c2ca5d0361b83be9dbd93104865e/propcache-0.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:647894f5ae99c4cf6bb82a1bb3a796f6e06af3caa3d32e26d2350d0e3e3faf24", size = 243649 },
-    { url = "https://files.pythonhosted.org/packages/ed/bc/4f7aba7f08f520376c4bb6a20b9a981a581b7f2e385fa0ec9f789bb2d362/propcache-0.2.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bfd3223c15bebe26518d58ccf9a39b93948d3dcb3e57a20480dfdd315356baff", size = 229103 },
-    { url = "https://files.pythonhosted.org/packages/fe/d5/04ac9cd4e51a57a96f78795e03c5a0ddb8f23ec098b86f92de028d7f2a6b/propcache-0.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d71264a80f3fcf512eb4f18f59423fe82d6e346ee97b90625f283df56aee103f", size = 226607 },
-    { url = "https://files.pythonhosted.org/packages/e3/f0/24060d959ea41d7a7cc7fdbf68b31852331aabda914a0c63bdb0e22e96d6/propcache-0.2.1-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:e73091191e4280403bde6c9a52a6999d69cdfde498f1fdf629105247599b57ec", size = 221153 },
-    { url = "https://files.pythonhosted.org/packages/77/a7/3ac76045a077b3e4de4859a0753010765e45749bdf53bd02bc4d372da1a0/propcache-0.2.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:3935bfa5fede35fb202c4b569bb9c042f337ca4ff7bd540a0aa5e37131659348", size = 222151 },
-    { url = "https://files.pythonhosted.org/packages/e7/af/5e29da6f80cebab3f5a4dcd2a3240e7f56f2c4abf51cbfcc99be34e17f0b/propcache-0.2.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:f508b0491767bb1f2b87fdfacaba5f7eddc2f867740ec69ece6d1946d29029a6", size = 233812 },
-    { url = "https://files.pythonhosted.org/packages/8c/89/ebe3ad52642cc5509eaa453e9f4b94b374d81bae3265c59d5c2d98efa1b4/propcache-0.2.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:1672137af7c46662a1c2be1e8dc78cb6d224319aaa40271c9257d886be4363a6", size = 238829 },
-    { url = "https://files.pythonhosted.org/packages/e9/2f/6b32f273fa02e978b7577159eae7471b3cfb88b48563b1c2578b2d7ca0bb/propcache-0.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b74c261802d3d2b85c9df2dfb2fa81b6f90deeef63c2db9f0e029a3cac50b518", size = 230704 },
-    { url = "https://files.pythonhosted.org/packages/5c/2e/f40ae6ff5624a5f77edd7b8359b208b5455ea113f68309e2b00a2e1426b6/propcache-0.2.1-cp312-cp312-win32.whl", hash = "sha256:d09c333d36c1409d56a9d29b3a1b800a42c76a57a5a8907eacdbce3f18768246", size = 40050 },
-    { url = "https://files.pythonhosted.org/packages/3b/77/a92c3ef994e47180862b9d7d11e37624fb1c00a16d61faf55115d970628b/propcache-0.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:c214999039d4f2a5b2073ac506bba279945233da8c786e490d411dfc30f855c1", size = 44117 },
-    { url = "https://files.pythonhosted.org/packages/0f/2a/329e0547cf2def8857157f9477669043e75524cc3e6251cef332b3ff256f/propcache-0.2.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:aca405706e0b0a44cc6bfd41fbe89919a6a56999157f6de7e182a990c36e37bc", size = 77002 },
-    { url = "https://files.pythonhosted.org/packages/12/2d/c4df5415e2382f840dc2ecbca0eeb2293024bc28e57a80392f2012b4708c/propcache-0.2.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:12d1083f001ace206fe34b6bdc2cb94be66d57a850866f0b908972f90996b3e9", size = 44639 },
-    { url = "https://files.pythonhosted.org/packages/d0/5a/21aaa4ea2f326edaa4e240959ac8b8386ea31dedfdaa636a3544d9e7a408/propcache-0.2.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d93f3307ad32a27bda2e88ec81134b823c240aa3abb55821a8da553eed8d9439", size = 44049 },
-    { url = "https://files.pythonhosted.org/packages/4e/3e/021b6cd86c0acc90d74784ccbb66808b0bd36067a1bf3e2deb0f3845f618/propcache-0.2.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ba278acf14471d36316159c94a802933d10b6a1e117b8554fe0d0d9b75c9d536", size = 224819 },
-    { url = "https://files.pythonhosted.org/packages/3c/57/c2fdeed1b3b8918b1770a133ba5c43ad3d78e18285b0c06364861ef5cc38/propcache-0.2.1-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4e6281aedfca15301c41f74d7005e6e3f4ca143584ba696ac69df4f02f40d629", size = 229625 },
-    { url = "https://files.pythonhosted.org/packages/9d/81/70d4ff57bf2877b5780b466471bebf5892f851a7e2ca0ae7ffd728220281/propcache-0.2.1-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5b750a8e5a1262434fb1517ddf64b5de58327f1adc3524a5e44c2ca43305eb0b", size = 232934 },
-    { url = "https://files.pythonhosted.org/packages/3c/b9/bb51ea95d73b3fb4100cb95adbd4e1acaf2cbb1fd1083f5468eeb4a099a8/propcache-0.2.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bf72af5e0fb40e9babf594308911436c8efde3cb5e75b6f206c34ad18be5c052", size = 227361 },
-    { url = "https://files.pythonhosted.org/packages/f1/20/3c6d696cd6fd70b29445960cc803b1851a1131e7a2e4ee261ee48e002bcd/propcache-0.2.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b2d0a12018b04f4cb820781ec0dffb5f7c7c1d2a5cd22bff7fb055a2cb19ebce", size = 213904 },
-    { url = "https://files.pythonhosted.org/packages/a1/cb/1593bfc5ac6d40c010fa823f128056d6bc25b667f5393781e37d62f12005/propcache-0.2.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:e800776a79a5aabdb17dcc2346a7d66d0777e942e4cd251defeb084762ecd17d", size = 212632 },
-    { url = "https://files.pythonhosted.org/packages/6d/5c/e95617e222be14a34c709442a0ec179f3207f8a2b900273720501a70ec5e/propcache-0.2.1-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:4160d9283bd382fa6c0c2b5e017acc95bc183570cd70968b9202ad6d8fc48dce", size = 207897 },
-    { url = "https://files.pythonhosted.org/packages/8e/3b/56c5ab3dc00f6375fbcdeefdede5adf9bee94f1fab04adc8db118f0f9e25/propcache-0.2.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:30b43e74f1359353341a7adb783c8f1b1c676367b011709f466f42fda2045e95", size = 208118 },
-    { url = "https://files.pythonhosted.org/packages/86/25/d7ef738323fbc6ebcbce33eb2a19c5e07a89a3df2fded206065bd5e868a9/propcache-0.2.1-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:58791550b27d5488b1bb52bc96328456095d96206a250d28d874fafe11b3dfaf", size = 217851 },
-    { url = "https://files.pythonhosted.org/packages/b3/77/763e6cef1852cf1ba740590364ec50309b89d1c818e3256d3929eb92fabf/propcache-0.2.1-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:0f022d381747f0dfe27e99d928e31bc51a18b65bb9e481ae0af1380a6725dd1f", size = 222630 },
-    { url = "https://files.pythonhosted.org/packages/4f/e9/0f86be33602089c701696fbed8d8c4c07b6ee9605c5b7536fd27ed540c5b/propcache-0.2.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:297878dc9d0a334358f9b608b56d02e72899f3b8499fc6044133f0d319e2ec30", size = 216269 },
-    { url = "https://files.pythonhosted.org/packages/cc/02/5ac83217d522394b6a2e81a2e888167e7ca629ef6569a3f09852d6dcb01a/propcache-0.2.1-cp313-cp313-win32.whl", hash = "sha256:ddfab44e4489bd79bda09d84c430677fc7f0a4939a73d2bba3073036f487a0a6", size = 39472 },
-    { url = "https://files.pythonhosted.org/packages/f4/33/d6f5420252a36034bc8a3a01171bc55b4bff5df50d1c63d9caa50693662f/propcache-0.2.1-cp313-cp313-win_amd64.whl", hash = "sha256:556fc6c10989f19a179e4321e5d678db8eb2924131e64652a51fe83e4c3db0e1", size = 43363 },
-    { url = "https://files.pythonhosted.org/packages/0a/08/6ab7f65240a16fa01023125e65258acf7e4884f483f267cdd6fcc48f37db/propcache-0.2.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:6a9a8c34fb7bb609419a211e59da8887eeca40d300b5ea8e56af98f6fbbb1541", size = 80403 },
-    { url = "https://files.pythonhosted.org/packages/34/fe/e7180285e21b4e6dff7d311fdf22490c9146a09a02834b5232d6248c6004/propcache-0.2.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ae1aa1cd222c6d205853b3013c69cd04515f9d6ab6de4b0603e2e1c33221303e", size = 46152 },
-    { url = "https://files.pythonhosted.org/packages/9c/36/aa74d884af826030ba9cee2ac109b0664beb7e9449c315c9c44db99efbb3/propcache-0.2.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:accb6150ce61c9c4b7738d45550806aa2b71c7668c6942f17b0ac182b6142fd4", size = 45674 },
-    { url = "https://files.pythonhosted.org/packages/22/59/6fe80a3fe7720f715f2c0f6df250dacbd7cad42832410dbd84c719c52f78/propcache-0.2.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5eee736daafa7af6d0a2dc15cc75e05c64f37fc37bafef2e00d77c14171c2097", size = 207792 },
-    { url = "https://files.pythonhosted.org/packages/4a/68/584cd51dd8f4d0f5fff5b128ce0cdb257cde903898eecfb92156bbc2c780/propcache-0.2.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f7a31fc1e1bd362874863fdeed71aed92d348f5336fd84f2197ba40c59f061bd", size = 223280 },
-    { url = "https://files.pythonhosted.org/packages/85/cb/4c3528460c41e61b06ec3f970c0f89f87fa21f63acac8642ed81a886c164/propcache-0.2.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cba4cfa1052819d16699e1d55d18c92b6e094d4517c41dd231a8b9f87b6fa681", size = 221293 },
-    { url = "https://files.pythonhosted.org/packages/69/c0/560e050aa6d31eeece3490d1174da508f05ab27536dfc8474af88b97160a/propcache-0.2.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f089118d584e859c62b3da0892b88a83d611c2033ac410e929cb6754eec0ed16", size = 208259 },
-    { url = "https://files.pythonhosted.org/packages/0c/87/d6c86a77632eb1ba86a328e3313159f246e7564cb5951e05ed77555826a0/propcache-0.2.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:781e65134efaf88feb447e8c97a51772aa75e48b794352f94cb7ea717dedda0d", size = 198632 },
-    { url = "https://files.pythonhosted.org/packages/3a/2b/3690ea7b662dc762ab7af5f3ef0e2d7513c823d193d7b2a1b4cda472c2be/propcache-0.2.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:31f5af773530fd3c658b32b6bdc2d0838543de70eb9a2156c03e410f7b0d3aae", size = 203516 },
-    { url = "https://files.pythonhosted.org/packages/4d/b5/afe716c16c23c77657185c257a41918b83e03993b6ccdfa748e5e7d328e9/propcache-0.2.1-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:a7a078f5d37bee6690959c813977da5291b24286e7b962e62a94cec31aa5188b", size = 199402 },
-    { url = "https://files.pythonhosted.org/packages/a4/c0/2d2df3aa7f8660d0d4cc4f1e00490c48d5958da57082e70dea7af366f876/propcache-0.2.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:cea7daf9fc7ae6687cf1e2c049752f19f146fdc37c2cc376e7d0032cf4f25347", size = 200528 },
-    { url = "https://files.pythonhosted.org/packages/21/c8/65ac9142f5e40c8497f7176e71d18826b09e06dd4eb401c9a4ee41aa9c74/propcache-0.2.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:8b3489ff1ed1e8315674d0775dc7d2195fb13ca17b3808721b54dbe9fd020faf", size = 211254 },
-    { url = "https://files.pythonhosted.org/packages/09/e4/edb70b447a1d8142df51ec7511e84aa64d7f6ce0a0fdf5eb55363cdd0935/propcache-0.2.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:9403db39be1393618dd80c746cb22ccda168efce239c73af13c3763ef56ffc04", size = 214589 },
-    { url = "https://files.pythonhosted.org/packages/cb/02/817f309ec8d8883287781d6d9390f80b14db6e6de08bc659dfe798a825c2/propcache-0.2.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:5d97151bc92d2b2578ff7ce779cdb9174337390a535953cbb9452fb65164c587", size = 207283 },
-    { url = "https://files.pythonhosted.org/packages/d7/fe/2d18612096ed2212cfef821b6fccdba5d52efc1d64511c206c5c16be28fd/propcache-0.2.1-cp39-cp39-win32.whl", hash = "sha256:9caac6b54914bdf41bcc91e7eb9147d331d29235a7c967c150ef5df6464fd1bb", size = 40866 },
-    { url = "https://files.pythonhosted.org/packages/24/2e/b5134802e7b57c403c7b73c7a39374e7a6b7f128d1968b4a4b4c0b700250/propcache-0.2.1-cp39-cp39-win_amd64.whl", hash = "sha256:92fc4500fcb33899b05ba73276dfb684a20d31caa567b7cb5252d48f896a91b1", size = 44975 },
-    { url = "https://files.pythonhosted.org/packages/41/b6/c5319caea262f4821995dca2107483b94a3345d4607ad797c76cb9c36bcc/propcache-0.2.1-py3-none-any.whl", hash = "sha256:52277518d6aae65536e9cea52d4e7fd2f7a66f4aa2d30ed3f2fcea620ace3c54", size = 11818 },
+version = "0.3.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/a6/16/43264e4a779dd8588c21a70f0709665ee8f611211bdd2c87d952cfa7c776/propcache-0.3.2.tar.gz", hash = "sha256:20d7d62e4e7ef05f221e0db2856b979540686342e7dd9973b815599c7057e168", size = 44139, upload-time = "2025-06-09T22:56:06.081Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ab/14/510deed325e262afeb8b360043c5d7c960da7d3ecd6d6f9496c9c56dc7f4/propcache-0.3.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:22d9962a358aedbb7a2e36187ff273adeaab9743373a272976d2e348d08c7770", size = 73178, upload-time = "2025-06-09T22:53:40.126Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/4e/ad52a7925ff01c1325653a730c7ec3175a23f948f08626a534133427dcff/propcache-0.3.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0d0fda578d1dc3f77b6b5a5dce3b9ad69a8250a891760a548df850a5e8da87f3", size = 43133, upload-time = "2025-06-09T22:53:41.965Z" },
+    { url = "https://files.pythonhosted.org/packages/63/7c/e9399ba5da7780871db4eac178e9c2e204c23dd3e7d32df202092a1ed400/propcache-0.3.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:3def3da3ac3ce41562d85db655d18ebac740cb3fa4367f11a52b3da9d03a5cc3", size = 43039, upload-time = "2025-06-09T22:53:43.268Z" },
+    { url = "https://files.pythonhosted.org/packages/22/e1/58da211eb8fdc6fc854002387d38f415a6ca5f5c67c1315b204a5d3e9d7a/propcache-0.3.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9bec58347a5a6cebf239daba9bda37dffec5b8d2ce004d9fe4edef3d2815137e", size = 201903, upload-time = "2025-06-09T22:53:44.872Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/0a/550ea0f52aac455cb90111c8bab995208443e46d925e51e2f6ebdf869525/propcache-0.3.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:55ffda449a507e9fbd4aca1a7d9aa6753b07d6166140e5a18d2ac9bc49eac220", size = 213362, upload-time = "2025-06-09T22:53:46.707Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/af/9893b7d878deda9bb69fcf54600b247fba7317761b7db11fede6e0f28bd0/propcache-0.3.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:64a67fb39229a8a8491dd42f864e5e263155e729c2e7ff723d6e25f596b1e8cb", size = 210525, upload-time = "2025-06-09T22:53:48.547Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/bb/38fd08b278ca85cde36d848091ad2b45954bc5f15cce494bb300b9285831/propcache-0.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9da1cf97b92b51253d5b68cf5a2b9e0dafca095e36b7f2da335e27dc6172a614", size = 198283, upload-time = "2025-06-09T22:53:50.067Z" },
+    { url = "https://files.pythonhosted.org/packages/78/8c/9fe55bd01d362bafb413dfe508c48753111a1e269737fa143ba85693592c/propcache-0.3.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5f559e127134b07425134b4065be45b166183fdcb433cb6c24c8e4149056ad50", size = 191872, upload-time = "2025-06-09T22:53:51.438Z" },
+    { url = "https://files.pythonhosted.org/packages/54/14/4701c33852937a22584e08abb531d654c8bcf7948a8f87ad0a4822394147/propcache-0.3.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:aff2e4e06435d61f11a428360a932138d0ec288b0a31dd9bd78d200bd4a2b339", size = 199452, upload-time = "2025-06-09T22:53:53.229Z" },
+    { url = "https://files.pythonhosted.org/packages/16/44/447f2253d859602095356007657ee535e0093215ea0b3d1d6a41d16e5201/propcache-0.3.2-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:4927842833830942a5d0a56e6f4839bc484785b8e1ce8d287359794818633ba0", size = 191567, upload-time = "2025-06-09T22:53:54.541Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/b3/e4756258749bb2d3b46defcff606a2f47410bab82be5824a67e84015b267/propcache-0.3.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:6107ddd08b02654a30fb8ad7a132021759d750a82578b94cd55ee2772b6ebea2", size = 193015, upload-time = "2025-06-09T22:53:56.44Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/df/e6d3c7574233164b6330b9fd697beeac402afd367280e6dc377bb99b43d9/propcache-0.3.2-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:70bd8b9cd6b519e12859c99f3fc9a93f375ebd22a50296c3a295028bea73b9e7", size = 204660, upload-time = "2025-06-09T22:53:57.839Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/53/e4d31dd5170b4a0e2e6b730f2385a96410633b4833dc25fe5dffd1f73294/propcache-0.3.2-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:2183111651d710d3097338dd1893fcf09c9f54e27ff1a8795495a16a469cc90b", size = 206105, upload-time = "2025-06-09T22:53:59.638Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/fe/74d54cf9fbe2a20ff786e5f7afcfde446588f0cf15fb2daacfbc267b866c/propcache-0.3.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:fb075ad271405dcad8e2a7ffc9a750a3bf70e533bd86e89f0603e607b93aa64c", size = 196980, upload-time = "2025-06-09T22:54:01.071Z" },
+    { url = "https://files.pythonhosted.org/packages/22/ec/c469c9d59dada8a7679625e0440b544fe72e99311a4679c279562051f6fc/propcache-0.3.2-cp310-cp310-win32.whl", hash = "sha256:404d70768080d3d3bdb41d0771037da19d8340d50b08e104ca0e7f9ce55fce70", size = 37679, upload-time = "2025-06-09T22:54:03.003Z" },
+    { url = "https://files.pythonhosted.org/packages/38/35/07a471371ac89d418f8d0b699c75ea6dca2041fbda360823de21f6a9ce0a/propcache-0.3.2-cp310-cp310-win_amd64.whl", hash = "sha256:7435d766f978b4ede777002e6b3b6641dd229cd1da8d3d3106a45770365f9ad9", size = 41459, upload-time = "2025-06-09T22:54:04.134Z" },
+    { url = "https://files.pythonhosted.org/packages/80/8d/e8b436717ab9c2cfc23b116d2c297305aa4cd8339172a456d61ebf5669b8/propcache-0.3.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:0b8d2f607bd8f80ddc04088bc2a037fdd17884a6fcadc47a96e334d72f3717be", size = 74207, upload-time = "2025-06-09T22:54:05.399Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/29/1e34000e9766d112171764b9fa3226fa0153ab565d0c242c70e9945318a7/propcache-0.3.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:06766d8f34733416e2e34f46fea488ad5d60726bb9481d3cddf89a6fa2d9603f", size = 43648, upload-time = "2025-06-09T22:54:08.023Z" },
+    { url = "https://files.pythonhosted.org/packages/46/92/1ad5af0df781e76988897da39b5f086c2bf0f028b7f9bd1f409bb05b6874/propcache-0.3.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a2dc1f4a1df4fecf4e6f68013575ff4af84ef6f478fe5344317a65d38a8e6dc9", size = 43496, upload-time = "2025-06-09T22:54:09.228Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/ce/e96392460f9fb68461fabab3e095cb00c8ddf901205be4eae5ce246e5b7e/propcache-0.3.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:be29c4f4810c5789cf10ddf6af80b041c724e629fa51e308a7a0fb19ed1ef7bf", size = 217288, upload-time = "2025-06-09T22:54:10.466Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/2a/866726ea345299f7ceefc861a5e782b045545ae6940851930a6adaf1fca6/propcache-0.3.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:59d61f6970ecbd8ff2e9360304d5c8876a6abd4530cb752c06586849ac8a9dc9", size = 227456, upload-time = "2025-06-09T22:54:11.828Z" },
+    { url = "https://files.pythonhosted.org/packages/de/03/07d992ccb6d930398689187e1b3c718339a1c06b8b145a8d9650e4726166/propcache-0.3.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:62180e0b8dbb6b004baec00a7983e4cc52f5ada9cd11f48c3528d8cfa7b96a66", size = 225429, upload-time = "2025-06-09T22:54:13.823Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/e6/116ba39448753b1330f48ab8ba927dcd6cf0baea8a0ccbc512dfb49ba670/propcache-0.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c144ca294a204c470f18cf4c9d78887810d04a3e2fbb30eea903575a779159df", size = 213472, upload-time = "2025-06-09T22:54:15.232Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/85/f01f5d97e54e428885a5497ccf7f54404cbb4f906688a1690cd51bf597dc/propcache-0.3.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c5c2a784234c28854878d68978265617aa6dc0780e53d44b4d67f3651a17a9a2", size = 204480, upload-time = "2025-06-09T22:54:17.104Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/79/7bf5ab9033b8b8194cc3f7cf1aaa0e9c3256320726f64a3e1f113a812dce/propcache-0.3.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:5745bc7acdafa978ca1642891b82c19238eadc78ba2aaa293c6863b304e552d7", size = 214530, upload-time = "2025-06-09T22:54:18.512Z" },
+    { url = "https://files.pythonhosted.org/packages/31/0b/bd3e0c00509b609317df4a18e6b05a450ef2d9a963e1d8bc9c9415d86f30/propcache-0.3.2-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:c0075bf773d66fa8c9d41f66cc132ecc75e5bb9dd7cce3cfd14adc5ca184cb95", size = 205230, upload-time = "2025-06-09T22:54:19.947Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/23/fae0ff9b54b0de4e819bbe559508da132d5683c32d84d0dc2ccce3563ed4/propcache-0.3.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:5f57aa0847730daceff0497f417c9de353c575d8da3579162cc74ac294c5369e", size = 206754, upload-time = "2025-06-09T22:54:21.716Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/7f/ad6a3c22630aaa5f618b4dc3c3598974a72abb4c18e45a50b3cdd091eb2f/propcache-0.3.2-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:eef914c014bf72d18efb55619447e0aecd5fb7c2e3fa7441e2e5d6099bddff7e", size = 218430, upload-time = "2025-06-09T22:54:23.17Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/2c/ba4f1c0e8a4b4c75910742f0d333759d441f65a1c7f34683b4a74c0ee015/propcache-0.3.2-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:2a4092e8549031e82facf3decdbc0883755d5bbcc62d3aea9d9e185549936dcf", size = 223884, upload-time = "2025-06-09T22:54:25.539Z" },
+    { url = "https://files.pythonhosted.org/packages/88/e4/ebe30fc399e98572019eee82ad0caf512401661985cbd3da5e3140ffa1b0/propcache-0.3.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:85871b050f174bc0bfb437efbdb68aaf860611953ed12418e4361bc9c392749e", size = 211480, upload-time = "2025-06-09T22:54:26.892Z" },
+    { url = "https://files.pythonhosted.org/packages/96/0a/7d5260b914e01d1d0906f7f38af101f8d8ed0dc47426219eeaf05e8ea7c2/propcache-0.3.2-cp311-cp311-win32.whl", hash = "sha256:36c8d9b673ec57900c3554264e630d45980fd302458e4ac801802a7fd2ef7897", size = 37757, upload-time = "2025-06-09T22:54:28.241Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/2d/89fe4489a884bc0da0c3278c552bd4ffe06a1ace559db5ef02ef24ab446b/propcache-0.3.2-cp311-cp311-win_amd64.whl", hash = "sha256:e53af8cb6a781b02d2ea079b5b853ba9430fcbe18a8e3ce647d5982a3ff69f39", size = 41500, upload-time = "2025-06-09T22:54:29.4Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/42/9ca01b0a6f48e81615dca4765a8f1dd2c057e0540f6116a27dc5ee01dfb6/propcache-0.3.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:8de106b6c84506b31c27168582cd3cb3000a6412c16df14a8628e5871ff83c10", size = 73674, upload-time = "2025-06-09T22:54:30.551Z" },
+    { url = "https://files.pythonhosted.org/packages/af/6e/21293133beb550f9c901bbece755d582bfaf2176bee4774000bd4dd41884/propcache-0.3.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:28710b0d3975117239c76600ea351934ac7b5ff56e60953474342608dbbb6154", size = 43570, upload-time = "2025-06-09T22:54:32.296Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/c8/0393a0a3a2b8760eb3bde3c147f62b20044f0ddac81e9d6ed7318ec0d852/propcache-0.3.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce26862344bdf836650ed2487c3d724b00fbfec4233a1013f597b78c1cb73615", size = 43094, upload-time = "2025-06-09T22:54:33.929Z" },
+    { url = "https://files.pythonhosted.org/packages/37/2c/489afe311a690399d04a3e03b069225670c1d489eb7b044a566511c1c498/propcache-0.3.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bca54bd347a253af2cf4544bbec232ab982f4868de0dd684246b67a51bc6b1db", size = 226958, upload-time = "2025-06-09T22:54:35.186Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/ca/63b520d2f3d418c968bf596839ae26cf7f87bead026b6192d4da6a08c467/propcache-0.3.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:55780d5e9a2ddc59711d727226bb1ba83a22dd32f64ee15594b9392b1f544eb1", size = 234894, upload-time = "2025-06-09T22:54:36.708Z" },
+    { url = "https://files.pythonhosted.org/packages/11/60/1d0ed6fff455a028d678df30cc28dcee7af77fa2b0e6962ce1df95c9a2a9/propcache-0.3.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:035e631be25d6975ed87ab23153db6a73426a48db688070d925aa27e996fe93c", size = 233672, upload-time = "2025-06-09T22:54:38.062Z" },
+    { url = "https://files.pythonhosted.org/packages/37/7c/54fd5301ef38505ab235d98827207176a5c9b2aa61939b10a460ca53e123/propcache-0.3.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ee6f22b6eaa39297c751d0e80c0d3a454f112f5c6481214fcf4c092074cecd67", size = 224395, upload-time = "2025-06-09T22:54:39.634Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/1a/89a40e0846f5de05fdc6779883bf46ba980e6df4d2ff8fb02643de126592/propcache-0.3.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:7ca3aee1aa955438c4dba34fc20a9f390e4c79967257d830f137bd5a8a32ed3b", size = 212510, upload-time = "2025-06-09T22:54:41.565Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/33/ca98368586c9566a6b8d5ef66e30484f8da84c0aac3f2d9aec6d31a11bd5/propcache-0.3.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:7a4f30862869fa2b68380d677cc1c5fcf1e0f2b9ea0cf665812895c75d0ca3b8", size = 222949, upload-time = "2025-06-09T22:54:43.038Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/11/ace870d0aafe443b33b2f0b7efdb872b7c3abd505bfb4890716ad7865e9d/propcache-0.3.2-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:b77ec3c257d7816d9f3700013639db7491a434644c906a2578a11daf13176251", size = 217258, upload-time = "2025-06-09T22:54:44.376Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/d2/86fd6f7adffcfc74b42c10a6b7db721d1d9ca1055c45d39a1a8f2a740a21/propcache-0.3.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:cab90ac9d3f14b2d5050928483d3d3b8fb6b4018893fc75710e6aa361ecb2474", size = 213036, upload-time = "2025-06-09T22:54:46.243Z" },
+    { url = "https://files.pythonhosted.org/packages/07/94/2d7d1e328f45ff34a0a284cf5a2847013701e24c2a53117e7c280a4316b3/propcache-0.3.2-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:0b504d29f3c47cf6b9e936c1852246c83d450e8e063d50562115a6be6d3a2535", size = 227684, upload-time = "2025-06-09T22:54:47.63Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/05/37ae63a0087677e90b1d14710e532ff104d44bc1efa3b3970fff99b891dc/propcache-0.3.2-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:ce2ac2675a6aa41ddb2a0c9cbff53780a617ac3d43e620f8fd77ba1c84dcfc06", size = 234562, upload-time = "2025-06-09T22:54:48.982Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/7c/3f539fcae630408d0bd8bf3208b9a647ccad10976eda62402a80adf8fc34/propcache-0.3.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:62b4239611205294cc433845b914131b2a1f03500ff3c1ed093ed216b82621e1", size = 222142, upload-time = "2025-06-09T22:54:50.424Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/d2/34b9eac8c35f79f8a962546b3e97e9d4b990c420ee66ac8255d5d9611648/propcache-0.3.2-cp312-cp312-win32.whl", hash = "sha256:df4a81b9b53449ebc90cc4deefb052c1dd934ba85012aa912c7ea7b7e38b60c1", size = 37711, upload-time = "2025-06-09T22:54:52.072Z" },
+    { url = "https://files.pythonhosted.org/packages/19/61/d582be5d226cf79071681d1b46b848d6cb03d7b70af7063e33a2787eaa03/propcache-0.3.2-cp312-cp312-win_amd64.whl", hash = "sha256:7046e79b989d7fe457bb755844019e10f693752d169076138abf17f31380800c", size = 41479, upload-time = "2025-06-09T22:54:53.234Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/39/8ea9bcfaaff16fd0b0fc901ee522e24c9ec44b4ca0229cfffb8066a06959/propcache-0.3.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:a7fad897f14d92086d6b03fdd2eb844777b0c4d7ec5e3bac0fbae2ab0602bbe5", size = 74678, upload-time = "2025-06-09T22:55:41.227Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/85/cab84c86966e1d354cf90cdc4ba52f32f99a5bca92a1529d666d957d7686/propcache-0.3.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1f43837d4ca000243fd7fd6301947d7cb93360d03cd08369969450cc6b2ce3b4", size = 43829, upload-time = "2025-06-09T22:55:42.417Z" },
+    { url = "https://files.pythonhosted.org/packages/23/f7/9cb719749152d8b26d63801b3220ce2d3931312b2744d2b3a088b0ee9947/propcache-0.3.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:261df2e9474a5949c46e962065d88eb9b96ce0f2bd30e9d3136bcde84befd8f2", size = 43729, upload-time = "2025-06-09T22:55:43.651Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/a2/0b2b5a210ff311260002a315f6f9531b65a36064dfb804655432b2f7d3e3/propcache-0.3.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e514326b79e51f0a177daab1052bc164d9d9e54133797a3a58d24c9c87a3fe6d", size = 204483, upload-time = "2025-06-09T22:55:45.327Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/e0/7aff5de0c535f783b0c8be5bdb750c305c1961d69fbb136939926e155d98/propcache-0.3.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d4a996adb6904f85894570301939afeee65f072b4fd265ed7e569e8d9058e4ec", size = 217425, upload-time = "2025-06-09T22:55:46.729Z" },
+    { url = "https://files.pythonhosted.org/packages/92/1d/65fa889eb3b2a7d6e4ed3c2b568a9cb8817547a1450b572de7bf24872800/propcache-0.3.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:76cace5d6b2a54e55b137669b30f31aa15977eeed390c7cbfb1dafa8dfe9a701", size = 214723, upload-time = "2025-06-09T22:55:48.342Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/e2/eecf6989870988dfd731de408a6fa366e853d361a06c2133b5878ce821ad/propcache-0.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:31248e44b81d59d6addbb182c4720f90b44e1efdc19f58112a3c3a1615fb47ef", size = 200166, upload-time = "2025-06-09T22:55:49.775Z" },
+    { url = "https://files.pythonhosted.org/packages/12/06/c32be4950967f18f77489268488c7cdc78cbfc65a8ba8101b15e526b83dc/propcache-0.3.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:abb7fa19dbf88d3857363e0493b999b8011eea856b846305d8c0512dfdf8fbb1", size = 194004, upload-time = "2025-06-09T22:55:51.335Z" },
+    { url = "https://files.pythonhosted.org/packages/46/6c/17b521a6b3b7cbe277a4064ff0aa9129dd8c89f425a5a9b6b4dd51cc3ff4/propcache-0.3.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:d81ac3ae39d38588ad0549e321e6f773a4e7cc68e7751524a22885d5bbadf886", size = 203075, upload-time = "2025-06-09T22:55:52.681Z" },
+    { url = "https://files.pythonhosted.org/packages/62/cb/3bdba2b736b3e45bc0e40f4370f745b3e711d439ffbffe3ae416393eece9/propcache-0.3.2-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:cc2782eb0f7a16462285b6f8394bbbd0e1ee5f928034e941ffc444012224171b", size = 195407, upload-time = "2025-06-09T22:55:54.048Z" },
+    { url = "https://files.pythonhosted.org/packages/29/bd/760c5c6a60a4a2c55a421bc34a25ba3919d49dee411ddb9d1493bb51d46e/propcache-0.3.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:db429c19a6c7e8a1c320e6a13c99799450f411b02251fb1b75e6217cf4a14fcb", size = 196045, upload-time = "2025-06-09T22:55:55.485Z" },
+    { url = "https://files.pythonhosted.org/packages/76/58/ced2757a46f55b8c84358d6ab8de4faf57cba831c51e823654da7144b13a/propcache-0.3.2-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:21d8759141a9e00a681d35a1f160892a36fb6caa715ba0b832f7747da48fb6ea", size = 208432, upload-time = "2025-06-09T22:55:56.884Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/ec/d98ea8d5a4d8fe0e372033f5254eddf3254344c0c5dc6c49ab84349e4733/propcache-0.3.2-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:2ca6d378f09adb13837614ad2754fa8afaee330254f404299611bce41a8438cb", size = 210100, upload-time = "2025-06-09T22:55:58.498Z" },
+    { url = "https://files.pythonhosted.org/packages/56/84/b6d8a7ecf3f62d7dd09d9d10bbf89fad6837970ef868b35b5ffa0d24d9de/propcache-0.3.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:34a624af06c048946709f4278b4176470073deda88d91342665d95f7c6270fbe", size = 200712, upload-time = "2025-06-09T22:55:59.906Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/32/889f4903ddfe4a9dc61da71ee58b763758cf2d608fe1decede06e6467f8d/propcache-0.3.2-cp39-cp39-win32.whl", hash = "sha256:4ba3fef1c30f306b1c274ce0b8baaa2c3cdd91f645c48f06394068f37d3837a1", size = 38187, upload-time = "2025-06-09T22:56:01.212Z" },
+    { url = "https://files.pythonhosted.org/packages/67/74/d666795fb9ba1dc139d30de64f3b6fd1ff9c9d3d96ccfdb992cd715ce5d2/propcache-0.3.2-cp39-cp39-win_amd64.whl", hash = "sha256:7a2368eed65fc69a7a7a40b27f22e85e7627b74216f0846b04ba5c116e191ec9", size = 42025, upload-time = "2025-06-09T22:56:02.875Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/35/cc0aaecf278bb4575b8555f2b137de5ab821595ddae9da9d3cd1da4072c7/propcache-0.3.2-py3-none-any.whl", hash = "sha256:98f1ec44fb675f5052cccc8e609c46ed23a35a1cfd18545ad4e29002d858a43f", size = 12663, upload-time = "2025-06-09T22:56:04.484Z" },
 ]
 
 [[package]]
 name = "protobuf"
-version = "5.29.3"
+version = "5.29.5"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f7/d1/e0a911544ca9993e0f17ce6d3cc0932752356c1b0a834397f28e63479344/protobuf-5.29.3.tar.gz", hash = "sha256:5da0f41edaf117bde316404bad1a486cb4ededf8e4a54891296f648e8e076620", size = 424945 }
+sdist = { url = "https://files.pythonhosted.org/packages/43/29/d09e70352e4e88c9c7a198d5645d7277811448d76c23b00345670f7c8a38/protobuf-5.29.5.tar.gz", hash = "sha256:bc1463bafd4b0929216c35f437a8e28731a2b7fe3d98bb77a600efced5a15c84", size = 425226, upload-time = "2025-05-28T23:51:59.82Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/dc/7a/1e38f3cafa022f477ca0f57a1f49962f21ad25850c3ca0acd3b9d0091518/protobuf-5.29.3-cp310-abi3-win32.whl", hash = "sha256:3ea51771449e1035f26069c4c7fd51fba990d07bc55ba80701c78f886bf9c888", size = 422708 },
-    { url = "https://files.pythonhosted.org/packages/61/fa/aae8e10512b83de633f2646506a6d835b151edf4b30d18d73afd01447253/protobuf-5.29.3-cp310-abi3-win_amd64.whl", hash = "sha256:a4fa6f80816a9a0678429e84973f2f98cbc218cca434abe8db2ad0bffc98503a", size = 434508 },
-    { url = "https://files.pythonhosted.org/packages/dd/04/3eaedc2ba17a088961d0e3bd396eac764450f431621b58a04ce898acd126/protobuf-5.29.3-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:a8434404bbf139aa9e1300dbf989667a83d42ddda9153d8ab76e0d5dcaca484e", size = 417825 },
-    { url = "https://files.pythonhosted.org/packages/4f/06/7c467744d23c3979ce250397e26d8ad8eeb2bea7b18ca12ad58313c1b8d5/protobuf-5.29.3-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:daaf63f70f25e8689c072cfad4334ca0ac1d1e05a92fc15c54eb9cf23c3efd84", size = 319573 },
-    { url = "https://files.pythonhosted.org/packages/a8/45/2ebbde52ad2be18d3675b6bee50e68cd73c9e0654de77d595540b5129df8/protobuf-5.29.3-cp38-abi3-manylinux2014_x86_64.whl", hash = "sha256:c027e08a08be10b67c06bf2370b99c811c466398c357e615ca88c91c07f0910f", size = 319672 },
-    { url = "https://files.pythonhosted.org/packages/85/a6/bf65a38f8be5ab8c3b575822acfd338702fdf7ac9abd8c81630cc7c9f4bd/protobuf-5.29.3-cp39-cp39-win32.whl", hash = "sha256:0eb32bfa5219fc8d4111803e9a690658aa2e6366384fd0851064b963b6d1f2a7", size = 422676 },
-    { url = "https://files.pythonhosted.org/packages/ac/e2/48d46adc86369ff092eaece3e537f76b3baaab45ca3dde257838cde831d2/protobuf-5.29.3-cp39-cp39-win_amd64.whl", hash = "sha256:6ce8cc3389a20693bfde6c6562e03474c40851b44975c9b2bf6df7d8c4f864da", size = 434593 },
-    { url = "https://files.pythonhosted.org/packages/fd/b2/ab07b09e0f6d143dfb839693aa05765257bceaa13d03bf1a696b78323e7a/protobuf-5.29.3-py3-none-any.whl", hash = "sha256:0a18ed4a24198528f2333802eb075e59dea9d679ab7a6c5efb017a59004d849f", size = 172550 },
+    { url = "https://files.pythonhosted.org/packages/5f/11/6e40e9fc5bba02988a214c07cf324595789ca7820160bfd1f8be96e48539/protobuf-5.29.5-cp310-abi3-win32.whl", hash = "sha256:3f1c6468a2cfd102ff4703976138844f78ebd1fb45f49011afc5139e9e283079", size = 422963, upload-time = "2025-05-28T23:51:41.204Z" },
+    { url = "https://files.pythonhosted.org/packages/81/7f/73cefb093e1a2a7c3ffd839e6f9fcafb7a427d300c7f8aef9c64405d8ac6/protobuf-5.29.5-cp310-abi3-win_amd64.whl", hash = "sha256:3f76e3a3675b4a4d867b52e4a5f5b78a2ef9565549d4037e06cf7b0942b1d3fc", size = 434818, upload-time = "2025-05-28T23:51:44.297Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/73/10e1661c21f139f2c6ad9b23040ff36fee624310dc28fba20d33fdae124c/protobuf-5.29.5-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:e38c5add5a311f2a6eb0340716ef9b039c1dfa428b28f25a7838ac329204a671", size = 418091, upload-time = "2025-05-28T23:51:45.907Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/04/98f6f8cf5b07ab1294c13f34b4e69b3722bb609c5b701d6c169828f9f8aa/protobuf-5.29.5-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:fa18533a299d7ab6c55a238bf8629311439995f2e7eca5caaff08663606e9015", size = 319824, upload-time = "2025-05-28T23:51:47.545Z" },
+    { url = "https://files.pythonhosted.org/packages/85/e4/07c80521879c2d15f321465ac24c70efe2381378c00bf5e56a0f4fbac8cd/protobuf-5.29.5-cp38-abi3-manylinux2014_x86_64.whl", hash = "sha256:63848923da3325e1bf7e9003d680ce6e14b07e55d0473253a690c3a8b8fd6e61", size = 319942, upload-time = "2025-05-28T23:51:49.11Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/59/ca89678bb0352f094fc92f2b358daa40e3acc91a93aa8f922b24762bf841/protobuf-5.29.5-cp39-cp39-win32.whl", hash = "sha256:6f642dc9a61782fa72b90878af134c5afe1917c89a568cd3476d758d3c3a0736", size = 423025, upload-time = "2025-05-28T23:51:54.003Z" },
+    { url = "https://files.pythonhosted.org/packages/96/8b/2c62731fe3e92ddbbeca0174f78f0f8739197cdeb7c75ceb5aad3706963b/protobuf-5.29.5-cp39-cp39-win_amd64.whl", hash = "sha256:470f3af547ef17847a28e1f47200a1cbf0ba3ff57b7de50d22776607cd2ea353", size = 434906, upload-time = "2025-05-28T23:51:55.782Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/cc/7e77861000a0691aeea8f4566e5d3aa716f2b1dece4a24439437e41d3d25/protobuf-5.29.5-py3-none-any.whl", hash = "sha256:6cf42630262c59b2d8de33954443d94b746c952b01434fc58a417fdbd2e84bd5", size = 172823, upload-time = "2025-05-28T23:51:58.157Z" },
 ]
 
 [[package]]
 name = "psutil"
-version = "6.1.1"
+version = "7.0.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/1f/5a/07871137bb752428aa4b659f910b399ba6f291156bdea939be3e96cae7cb/psutil-6.1.1.tar.gz", hash = "sha256:cf8496728c18f2d0b45198f06895be52f36611711746b7f30c464b422b50e2f5", size = 508502 }
+sdist = { url = "https://files.pythonhosted.org/packages/2a/80/336820c1ad9286a4ded7e845b2eccfcb27851ab8ac6abece774a6ff4d3de/psutil-7.0.0.tar.gz", hash = "sha256:7be9c3eba38beccb6495ea33afd982a44074b78f28c434a1f51cc07fd315c456", size = 497003, upload-time = "2025-02-13T21:54:07.946Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/61/99/ca79d302be46f7bdd8321089762dd4476ee725fce16fc2b2e1dbba8cac17/psutil-6.1.1-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:fc0ed7fe2231a444fc219b9c42d0376e0a9a1a72f16c5cfa0f68d19f1a0663e8", size = 247511 },
-    { url = "https://files.pythonhosted.org/packages/0b/6b/73dbde0dd38f3782905d4587049b9be64d76671042fdcaf60e2430c6796d/psutil-6.1.1-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:0bdd4eab935276290ad3cb718e9809412895ca6b5b334f5a9111ee6d9aff9377", size = 248985 },
-    { url = "https://files.pythonhosted.org/packages/17/38/c319d31a1d3f88c5b79c68b3116c129e5133f1822157dd6da34043e32ed6/psutil-6.1.1-cp36-abi3-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b6e06c20c05fe95a3d7302d74e7097756d4ba1247975ad6905441ae1b5b66003", size = 284488 },
-    { url = "https://files.pythonhosted.org/packages/9c/39/0f88a830a1c8a3aba27fededc642da37613c57cbff143412e3536f89784f/psutil-6.1.1-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:97f7cb9921fbec4904f522d972f0c0e1f4fabbdd4e0287813b21215074a0f160", size = 287477 },
-    { url = "https://files.pythonhosted.org/packages/47/da/99f4345d4ddf2845cb5b5bd0d93d554e84542d116934fde07a0c50bd4e9f/psutil-6.1.1-cp36-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:33431e84fee02bc84ea36d9e2c4a6d395d479c9dd9bba2376c1f6ee8f3a4e0b3", size = 289017 },
-    { url = "https://files.pythonhosted.org/packages/38/53/bd755c2896f4461fd4f36fa6a6dcb66a88a9e4b9fd4e5b66a77cf9d4a584/psutil-6.1.1-cp37-abi3-win32.whl", hash = "sha256:eaa912e0b11848c4d9279a93d7e2783df352b082f40111e078388701fd479e53", size = 250602 },
-    { url = "https://files.pythonhosted.org/packages/7b/d7/7831438e6c3ebbfa6e01a927127a6cb42ad3ab844247f3c5b96bea25d73d/psutil-6.1.1-cp37-abi3-win_amd64.whl", hash = "sha256:f35cfccb065fff93529d2afb4a2e89e363fe63ca1e4a5da22b603a85833c2649", size = 254444 },
+    { url = "https://files.pythonhosted.org/packages/ed/e6/2d26234410f8b8abdbf891c9da62bee396583f713fb9f3325a4760875d22/psutil-7.0.0-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:101d71dc322e3cffd7cea0650b09b3d08b8e7c4109dd6809fe452dfd00e58b25", size = 238051, upload-time = "2025-02-13T21:54:12.36Z" },
+    { url = "https://files.pythonhosted.org/packages/04/8b/30f930733afe425e3cbfc0e1468a30a18942350c1a8816acfade80c005c4/psutil-7.0.0-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:39db632f6bb862eeccf56660871433e111b6ea58f2caea825571951d4b6aa3da", size = 239535, upload-time = "2025-02-13T21:54:16.07Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/ed/d362e84620dd22876b55389248e522338ed1bf134a5edd3b8231d7207f6d/psutil-7.0.0-cp36-abi3-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1fcee592b4c6f146991ca55919ea3d1f8926497a713ed7faaf8225e174581e91", size = 275004, upload-time = "2025-02-13T21:54:18.662Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/b9/b0eb3f3cbcb734d930fdf839431606844a825b23eaf9a6ab371edac8162c/psutil-7.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b1388a4f6875d7e2aff5c4ca1cc16c545ed41dd8bb596cefea80111db353a34", size = 277986, upload-time = "2025-02-13T21:54:21.811Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/a2/709e0fe2f093556c17fbafda93ac032257242cabcc7ff3369e2cb76a97aa/psutil-7.0.0-cp36-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a5f098451abc2828f7dc6b58d44b532b22f2088f4999a937557b603ce72b1993", size = 279544, upload-time = "2025-02-13T21:54:24.68Z" },
+    { url = "https://files.pythonhosted.org/packages/50/e6/eecf58810b9d12e6427369784efe814a1eec0f492084ce8eb8f4d89d6d61/psutil-7.0.0-cp37-abi3-win32.whl", hash = "sha256:ba3fcef7523064a6c9da440fc4d6bd07da93ac726b5733c29027d7dc95b39d99", size = 241053, upload-time = "2025-02-13T21:54:34.31Z" },
+    { url = "https://files.pythonhosted.org/packages/50/1b/6921afe68c74868b4c9fa424dad3be35b095e16687989ebbb50ce4fceb7c/psutil-7.0.0-cp37-abi3-win_amd64.whl", hash = "sha256:4cf3d4eb1aa9b348dec30105c55cd9b7d4629285735a102beb4441e38db90553", size = 244885, upload-time = "2025-02-13T21:54:37.486Z" },
 ]
 
 [[package]]
 name = "py-cpuinfo"
 version = "9.0.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/37/a8/d832f7293ebb21690860d2e01d8115e5ff6f2ae8bbdc953f0eb0fa4bd2c7/py-cpuinfo-9.0.0.tar.gz", hash = "sha256:3cdbbf3fac90dc6f118bfd64384f309edeadd902d7c8fb17f02ffa1fc3f49690", size = 104716 }
+sdist = { url = "https://files.pythonhosted.org/packages/37/a8/d832f7293ebb21690860d2e01d8115e5ff6f2ae8bbdc953f0eb0fa4bd2c7/py-cpuinfo-9.0.0.tar.gz", hash = "sha256:3cdbbf3fac90dc6f118bfd64384f309edeadd902d7c8fb17f02ffa1fc3f49690", size = 104716, upload-time = "2022-10-25T20:38:06.303Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e0/a9/023730ba63db1e494a271cb018dcd361bd2c917ba7004c3e49d5daf795a2/py_cpuinfo-9.0.0-py3-none-any.whl", hash = "sha256:859625bc251f64e21f077d099d4162689c762b5d6a4c3c97553d56241c9674d5", size = 22335 },
+    { url = "https://files.pythonhosted.org/packages/e0/a9/023730ba63db1e494a271cb018dcd361bd2c917ba7004c3e49d5daf795a2/py_cpuinfo-9.0.0-py3-none-any.whl", hash = "sha256:859625bc251f64e21f077d099d4162689c762b5d6a4c3c97553d56241c9674d5", size = 22335, upload-time = "2022-10-25T20:38:27.636Z" },
 ]
 
 [[package]]
 name = "pyarrow"
-version = "19.0.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/7b/01/fe1fd04744c2aa038e5a11c7a4adb3d62bce09798695e54f7274b5977134/pyarrow-19.0.0.tar.gz", hash = "sha256:8d47c691765cf497aaeed4954d226568563f1b3b74ff61139f2d77876717084b", size = 1129096 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/1c/02/1ad80ffd3c558916858a49c83b6e494a9d93009bbebc603cf0cb8263bea7/pyarrow-19.0.0-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:c318eda14f6627966997a7d8c374a87d084a94e4e38e9abbe97395c215830e0c", size = 30686262 },
-    { url = "https://files.pythonhosted.org/packages/1b/f0/adab5f142eb8203db8bfbd3a816816e37a85423ae684567e7f3555658315/pyarrow-19.0.0-cp310-cp310-macosx_12_0_x86_64.whl", hash = "sha256:62ef8360ff256e960f57ce0299090fb86423afed5e46f18f1225f960e05aae3d", size = 32100005 },
-    { url = "https://files.pythonhosted.org/packages/94/8b/e674083610e5efc48d2f205c568d842cdfdf683d12f9ff0d546e38757722/pyarrow-19.0.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2795064647add0f16563e57e3d294dbfc067b723f0fd82ecd80af56dad15f503", size = 41144815 },
-    { url = "https://files.pythonhosted.org/packages/d5/fb/2726241a792b7f8a58789e5a63d1be9a5a4059206318fd0ff9485a578952/pyarrow-19.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a218670b26fb1bc74796458d97bcab072765f9b524f95b2fccad70158feb8b17", size = 42180380 },
-    { url = "https://files.pythonhosted.org/packages/7d/09/7aef12446d8e7002dfc07bb7bc71f594c1d5844ca78b364a49f07efb65b1/pyarrow-19.0.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:66732e39eaa2247996a6b04c8aa33e3503d351831424cdf8d2e9a0582ac54b34", size = 40515021 },
-    { url = "https://files.pythonhosted.org/packages/31/55/f05fc5608cc96060c2b24de505324d641888bd62d4eed2fa1dacd872a1e1/pyarrow-19.0.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:e675a3ad4732b92d72e4d24009707e923cab76b0d088e5054914f11a797ebe44", size = 42067488 },
-    { url = "https://files.pythonhosted.org/packages/f0/01/097653cec7a944c16313cb748a326771133c142034b252076bd84743b98d/pyarrow-19.0.0-cp310-cp310-win_amd64.whl", hash = "sha256:f094742275586cdd6b1a03655ccff3b24b2610c3af76f810356c4c71d24a2a6c", size = 25276726 },
-    { url = "https://files.pythonhosted.org/packages/82/42/fba3a35bef5833bf88ed35e6a810dc1781236e1d4f808d2df824a7d21819/pyarrow-19.0.0-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:8e3a839bf36ec03b4315dc924d36dcde5444a50066f1c10f8290293c0427b46a", size = 30711936 },
-    { url = "https://files.pythonhosted.org/packages/88/7a/0da93a3eaaf251a30e32f3221e874263cdcd366c2cd6b7c05293aad91152/pyarrow-19.0.0-cp311-cp311-macosx_12_0_x86_64.whl", hash = "sha256:ce42275097512d9e4e4a39aade58ef2b3798a93aa3026566b7892177c266f735", size = 32133182 },
-    { url = "https://files.pythonhosted.org/packages/2f/df/fe43b1c50d3100d0de53f988344118bc20362d0de005f8a407454fa565f8/pyarrow-19.0.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9348a0137568c45601b031a8d118275069435f151cbb77e6a08a27e8125f59d4", size = 41145489 },
-    { url = "https://files.pythonhosted.org/packages/45/bb/6f73b41b342a0342f2516a02db4aa97a4f9569cc35482a5c288090140cd4/pyarrow-19.0.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2a0144a712d990d60f7f42b7a31f0acaccf4c1e43e957f7b1ad58150d6f639c1", size = 42177823 },
-    { url = "https://files.pythonhosted.org/packages/23/7b/f038a96f421e453a71bd7a0f78d62b1b2ae9bcac06ed51179ca532e6a0a2/pyarrow-19.0.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:2a1a109dfda558eb011e5f6385837daffd920d54ca00669f7a11132d0b1e6042", size = 40530609 },
-    { url = "https://files.pythonhosted.org/packages/b8/39/a2a6714b471c000e6dd6af4495dce00d7d1332351b8e3170dfb9f91dad1f/pyarrow-19.0.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:be686bf625aa7b9bada18defb3a3ea3981c1099697239788ff111d87f04cd263", size = 42081534 },
-    { url = "https://files.pythonhosted.org/packages/6c/a3/8396fb06ca05d807e89980c177be26617aad15211ece3184e0caa730b8a6/pyarrow-19.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:239ca66d9a05844bdf5af128861af525e14df3c9591bcc05bac25918e650d3a2", size = 25281090 },
-    { url = "https://files.pythonhosted.org/packages/bc/2e/152885f5ef421e80dae68b9c133ab261934f93a6d5e16b61d79c0ed597fb/pyarrow-19.0.0-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:a7bbe7109ab6198688b7079cbad5a8c22de4d47c4880d8e4847520a83b0d1b68", size = 30667964 },
-    { url = "https://files.pythonhosted.org/packages/80/c2/08bbee9a8610a47c9a1466845f405baf53a639ddd947c5133d8ba13544b6/pyarrow-19.0.0-cp312-cp312-macosx_12_0_x86_64.whl", hash = "sha256:4624c89d6f777c580e8732c27bb8e77fd1433b89707f17c04af7635dd9638351", size = 32125039 },
-    { url = "https://files.pythonhosted.org/packages/d2/56/06994df823212f5688d3c8bf4294928b12c9be36681872853655724d28c6/pyarrow-19.0.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2b6d3ce4288793350dc2d08d1e184fd70631ea22a4ff9ea5c4ff182130249d9b", size = 41140729 },
-    { url = "https://files.pythonhosted.org/packages/94/65/38ad577c98140a9db71e9e1e594b6adb58a7478a5afec6456a8ca2df7f70/pyarrow-19.0.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:450a7d27e840e4d9a384b5c77199d489b401529e75a3b7a3799d4cd7957f2f9c", size = 42202267 },
-    { url = "https://files.pythonhosted.org/packages/b6/1f/966b722251a7354114ccbb71cf1a83922023e69efd8945ebf628a851ec4c/pyarrow-19.0.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:a08e2a8a039a3f72afb67a6668180f09fddaa38fe0d21f13212b4aba4b5d2451", size = 40505858 },
-    { url = "https://files.pythonhosted.org/packages/3b/5e/6bc81aa7fc9affc7d1c03b912fbcc984ca56c2a18513684da267715dab7b/pyarrow-19.0.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:f43f5aef2a13d4d56adadae5720d1fed4c1356c993eda8b59dace4b5983843c1", size = 42084973 },
-    { url = "https://files.pythonhosted.org/packages/53/c3/2f56da818b6a4758cbd514957c67bd0f078ebffa5390ee2e2bf0f9e8defc/pyarrow-19.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:2f672f5364b2d7829ef7c94be199bb88bf5661dd485e21d2d37de12ccb78a136", size = 25241976 },
-    { url = "https://files.pythonhosted.org/packages/f5/b9/ba07ed3dd6b6e4f379b78e9c47c50c8886e07862ab7fa6339ac38622d755/pyarrow-19.0.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:cf3bf0ce511b833f7bc5f5bb3127ba731e97222023a444b7359f3a22e2a3b463", size = 30651291 },
-    { url = "https://files.pythonhosted.org/packages/ad/10/0d304243c8277035298a68a70807efb76199c6c929bb3363c92ac9be6a0d/pyarrow-19.0.0-cp313-cp313-macosx_12_0_x86_64.whl", hash = "sha256:4d8b0c0de0a73df1f1bf439af1b60f273d719d70648e898bc077547649bb8352", size = 32100461 },
-    { url = "https://files.pythonhosted.org/packages/8a/61/bcfc5182e11831bca3f849945b9b106e09fd10ded773dff466658e972a45/pyarrow-19.0.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a92aff08e23d281c69835e4a47b80569242a504095ef6a6223c1f6bb8883431d", size = 41132491 },
-    { url = "https://files.pythonhosted.org/packages/8e/87/2915a29049ec352dc69a967fbcbd76b0180319233de0daf8bd368df37099/pyarrow-19.0.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3b78eff5968a1889a0f3bc81ca57e1e19b75f664d9c61a42a604bf9d8402aae", size = 42192529 },
-    { url = "https://files.pythonhosted.org/packages/48/18/44e5542b2707a8afaf78b5b88c608f261871ae77787eac07b7c679ca6f0f/pyarrow-19.0.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:b34d3bde38eba66190b215bae441646330f8e9da05c29e4b5dd3e41bde701098", size = 40495363 },
-    { url = "https://files.pythonhosted.org/packages/ba/d6/5096deb7599bbd20bc2768058fe23bc725b88eb41bee58303293583a2935/pyarrow-19.0.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:5418d4d0fab3a0ed497bad21d17a7973aad336d66ad4932a3f5f7480d4ca0c04", size = 42074075 },
-    { url = "https://files.pythonhosted.org/packages/2c/df/e3c839c04c284c9ec3d62b02a8c452b795d9b07b04079ab91ce33484d4c5/pyarrow-19.0.0-cp313-cp313-win_amd64.whl", hash = "sha256:e82c3d5e44e969c217827b780ed8faf7ac4c53f934ae9238872e749fa531f7c9", size = 25239803 },
-    { url = "https://files.pythonhosted.org/packages/6a/d3/a6d4088e906c7b5d47792256212606d2ae679046dc750eee0ae167338e5c/pyarrow-19.0.0-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:f208c3b58a6df3b239e0bb130e13bc7487ed14f39a9ff357b6415e3f6339b560", size = 30695401 },
-    { url = "https://files.pythonhosted.org/packages/94/25/70040fd0e397dd1b937f459eaeeec942a76027357491dca0ada09d1322af/pyarrow-19.0.0-cp313-cp313t-macosx_12_0_x86_64.whl", hash = "sha256:c751c1c93955b7a84c06794df46f1cec93e18610dcd5ab7d08e89a81df70a849", size = 32104680 },
-    { url = "https://files.pythonhosted.org/packages/4e/f9/92783290cc0d80ca16d34b0c126305bfacca4b87dd889c8f16c6ef2a8fd7/pyarrow-19.0.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b903afaa5df66d50fc38672ad095806443b05f202c792694f3a604ead7c6ea6e", size = 41076754 },
-    { url = "https://files.pythonhosted.org/packages/05/46/2c9870f50a495c72e2b8982ae29a9b1680707ea936edc0de444cec48f875/pyarrow-19.0.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a22a4bc0937856263df8b94f2f2781b33dd7f876f787ed746608e06902d691a5", size = 42163133 },
-    { url = "https://files.pythonhosted.org/packages/7b/2f/437922b902549228fb15814e8a26105bff2787ece466a8d886eb6699efad/pyarrow-19.0.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:5e8a28b918e2e878c918f6d89137386c06fe577cd08d73a6be8dafb317dc2d73", size = 40452210 },
-    { url = "https://files.pythonhosted.org/packages/36/ef/1d7975053af9d106da973bac142d0d4da71b7550a3576cc3e0b3f444d21a/pyarrow-19.0.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:29cd86c8001a94f768f79440bf83fee23963af5e7bc68ce3a7e5f120e17edf89", size = 42077618 },
-    { url = "https://files.pythonhosted.org/packages/59/13/e39417005ee632e131d0246cf5c1149618a55554ccdf2a4d887065e672a7/pyarrow-19.0.0-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:c0423393e4a07ff6fea08feb44153302dd261d0551cc3b538ea7a5dc853af43a", size = 30698254 },
-    { url = "https://files.pythonhosted.org/packages/06/87/1f9d7df296dd5c065e52ae3e9070dfe611f6bd97e90f28b6a45c410dcb67/pyarrow-19.0.0-cp39-cp39-macosx_12_0_x86_64.whl", hash = "sha256:718947fb6d82409013a74b176bf93e0f49ef952d8a2ecd068fecd192a97885b7", size = 32114467 },
-    { url = "https://files.pythonhosted.org/packages/b2/b1/9e7babf5d469bd35f1d062a04721ead72456101f6d851a2e8a43bb07a580/pyarrow-19.0.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3c1c162c4660e0978411a4761f91113dde8da3433683efa473501254563dcbe8", size = 41157781 },
-    { url = "https://files.pythonhosted.org/packages/9f/25/fa8e882a6c06e6d8dd640d3acce3912d7c39358940eb0c7b3c8b962457d0/pyarrow-19.0.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c73268cf557e688efb60f1ccbc7376f7e18cd8e2acae9e663e98b194c40c1a2d", size = 42190773 },
-    { url = "https://files.pythonhosted.org/packages/b8/d6/7fd60aa79cada815306d9804403386b06893ef63e73876174717a62002c4/pyarrow-19.0.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:edfe6d3916e915ada9acc4e48f6dafca7efdbad2e6283db6fd9385a1b23055f1", size = 40528310 },
-    { url = "https://files.pythonhosted.org/packages/0e/f1/c1ec7620a5768b8c7f9083572fda7b05b77ea083c3400fbd9e4ae40e63bd/pyarrow-19.0.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:da410b70a7ab8eb524112f037a7a35da7128b33d484f7671a264a4c224ac131d", size = 42080774 },
-    { url = "https://files.pythonhosted.org/packages/c4/28/c51c9af2703b5a592d1b66546611b24de8ca01e04c3f5da769c3318bca6c/pyarrow-19.0.0-cp39-cp39-win_amd64.whl", hash = "sha256:597360ffc71fc8cceea1aec1fb60cb510571a744fffc87db33d551d5de919bec", size = 25464978 },
-]
-
-[[package]]
-name = "pycountry"
-version = "24.6.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/76/57/c389fa68c50590881a75b7883eeb3dc15e9e73a0fdc001cdd45c13290c92/pycountry-24.6.1.tar.gz", hash = "sha256:b61b3faccea67f87d10c1f2b0fc0be714409e8fcdcc1315613174f6466c10221", size = 6043910 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/b1/ec/1fb891d8a2660716aadb2143235481d15ed1cbfe3ad669194690b0604492/pycountry-24.6.1-py3-none-any.whl", hash = "sha256:f1a4fb391cd7214f8eefd39556d740adcc233c778a27f8942c8dca351d6ce06f", size = 6335189 },
+version = "21.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ef/c2/ea068b8f00905c06329a3dfcd40d0fcc2b7d0f2e355bdb25b65e0a0e4cd4/pyarrow-21.0.0.tar.gz", hash = "sha256:5051f2dccf0e283ff56335760cbc8622cf52264d67e359d5569541ac11b6d5bc", size = 1133487, upload-time = "2025-07-18T00:57:31.761Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/17/d9/110de31880016e2afc52d8580b397dbe47615defbf09ca8cf55f56c62165/pyarrow-21.0.0-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:e563271e2c5ff4d4a4cbeb2c83d5cf0d4938b891518e676025f7268c6fe5fe26", size = 31196837, upload-time = "2025-07-18T00:54:34.755Z" },
+    { url = "https://files.pythonhosted.org/packages/df/5f/c1c1997613abf24fceb087e79432d24c19bc6f7259cab57c2c8e5e545fab/pyarrow-21.0.0-cp310-cp310-macosx_12_0_x86_64.whl", hash = "sha256:fee33b0ca46f4c85443d6c450357101e47d53e6c3f008d658c27a2d020d44c79", size = 32659470, upload-time = "2025-07-18T00:54:38.329Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/ed/b1589a777816ee33ba123ba1e4f8f02243a844fed0deec97bde9fb21a5cf/pyarrow-21.0.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:7be45519b830f7c24b21d630a31d48bcebfd5d4d7f9d3bdb49da9cdf6d764edb", size = 41055619, upload-time = "2025-07-18T00:54:42.172Z" },
+    { url = "https://files.pythonhosted.org/packages/44/28/b6672962639e85dc0ac36f71ab3a8f5f38e01b51343d7aa372a6b56fa3f3/pyarrow-21.0.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:26bfd95f6bff443ceae63c65dc7e048670b7e98bc892210acba7e4995d3d4b51", size = 42733488, upload-time = "2025-07-18T00:54:47.132Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/cc/de02c3614874b9089c94eac093f90ca5dfa6d5afe45de3ba847fd950fdf1/pyarrow-21.0.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:bd04ec08f7f8bd113c55868bd3fc442a9db67c27af098c5f814a3091e71cc61a", size = 43329159, upload-time = "2025-07-18T00:54:51.686Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/3e/99473332ac40278f196e105ce30b79ab8affab12f6194802f2593d6b0be2/pyarrow-21.0.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:9b0b14b49ac10654332a805aedfc0147fb3469cbf8ea951b3d040dab12372594", size = 45050567, upload-time = "2025-07-18T00:54:56.679Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/f5/c372ef60593d713e8bfbb7e0c743501605f0ad00719146dc075faf11172b/pyarrow-21.0.0-cp310-cp310-win_amd64.whl", hash = "sha256:9d9f8bcb4c3be7738add259738abdeddc363de1b80e3310e04067aa1ca596634", size = 26217959, upload-time = "2025-07-18T00:55:00.482Z" },
+    { url = "https://files.pythonhosted.org/packages/94/dc/80564a3071a57c20b7c32575e4a0120e8a330ef487c319b122942d665960/pyarrow-21.0.0-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:c077f48aab61738c237802836fc3844f85409a46015635198761b0d6a688f87b", size = 31243234, upload-time = "2025-07-18T00:55:03.812Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/cc/3b51cb2db26fe535d14f74cab4c79b191ed9a8cd4cbba45e2379b5ca2746/pyarrow-21.0.0-cp311-cp311-macosx_12_0_x86_64.whl", hash = "sha256:689f448066781856237eca8d1975b98cace19b8dd2ab6145bf49475478bcaa10", size = 32714370, upload-time = "2025-07-18T00:55:07.495Z" },
+    { url = "https://files.pythonhosted.org/packages/24/11/a4431f36d5ad7d83b87146f515c063e4d07ef0b7240876ddb885e6b44f2e/pyarrow-21.0.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:479ee41399fcddc46159a551705b89c05f11e8b8cb8e968f7fec64f62d91985e", size = 41135424, upload-time = "2025-07-18T00:55:11.461Z" },
+    { url = "https://files.pythonhosted.org/packages/74/dc/035d54638fc5d2971cbf1e987ccd45f1091c83bcf747281cf6cc25e72c88/pyarrow-21.0.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:40ebfcb54a4f11bcde86bc586cbd0272bac0d516cfa539c799c2453768477569", size = 42823810, upload-time = "2025-07-18T00:55:16.301Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/3b/89fced102448a9e3e0d4dded1f37fa3ce4700f02cdb8665457fcc8015f5b/pyarrow-21.0.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8d58d8497814274d3d20214fbb24abcad2f7e351474357d552a8d53bce70c70e", size = 43391538, upload-time = "2025-07-18T00:55:23.82Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/bb/ea7f1bd08978d39debd3b23611c293f64a642557e8141c80635d501e6d53/pyarrow-21.0.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:585e7224f21124dd57836b1530ac8f2df2afc43c861d7bf3d58a4870c42ae36c", size = 45120056, upload-time = "2025-07-18T00:55:28.231Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/0b/77ea0600009842b30ceebc3337639a7380cd946061b620ac1a2f3cb541e2/pyarrow-21.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:555ca6935b2cbca2c0e932bedd853e9bc523098c39636de9ad4693b5b1df86d6", size = 26220568, upload-time = "2025-07-18T00:55:32.122Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/d4/d4f817b21aacc30195cf6a46ba041dd1be827efa4a623cc8bf39a1c2a0c0/pyarrow-21.0.0-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:3a302f0e0963db37e0a24a70c56cf91a4faa0bca51c23812279ca2e23481fccd", size = 31160305, upload-time = "2025-07-18T00:55:35.373Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/9c/dcd38ce6e4b4d9a19e1d36914cb8e2b1da4e6003dd075474c4cfcdfe0601/pyarrow-21.0.0-cp312-cp312-macosx_12_0_x86_64.whl", hash = "sha256:b6b27cf01e243871390474a211a7922bfbe3bda21e39bc9160daf0da3fe48876", size = 32684264, upload-time = "2025-07-18T00:55:39.303Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/74/2a2d9f8d7a59b639523454bec12dba35ae3d0a07d8ab529dc0809f74b23c/pyarrow-21.0.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:e72a8ec6b868e258a2cd2672d91f2860ad532d590ce94cdf7d5e7ec674ccf03d", size = 41108099, upload-time = "2025-07-18T00:55:42.889Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/90/2660332eeb31303c13b653ea566a9918484b6e4d6b9d2d46879a33ab0622/pyarrow-21.0.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:b7ae0bbdc8c6674259b25bef5d2a1d6af5d39d7200c819cf99e07f7dfef1c51e", size = 42829529, upload-time = "2025-07-18T00:55:47.069Z" },
+    { url = "https://files.pythonhosted.org/packages/33/27/1a93a25c92717f6aa0fca06eb4700860577d016cd3ae51aad0e0488ac899/pyarrow-21.0.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:58c30a1729f82d201627c173d91bd431db88ea74dcaa3885855bc6203e433b82", size = 43367883, upload-time = "2025-07-18T00:55:53.069Z" },
+    { url = "https://files.pythonhosted.org/packages/05/d9/4d09d919f35d599bc05c6950095e358c3e15148ead26292dfca1fb659b0c/pyarrow-21.0.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:072116f65604b822a7f22945a7a6e581cfa28e3454fdcc6939d4ff6090126623", size = 45133802, upload-time = "2025-07-18T00:55:57.714Z" },
+    { url = "https://files.pythonhosted.org/packages/71/30/f3795b6e192c3ab881325ffe172e526499eb3780e306a15103a2764916a2/pyarrow-21.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:cf56ec8b0a5c8c9d7021d6fd754e688104f9ebebf1bf4449613c9531f5346a18", size = 26203175, upload-time = "2025-07-18T00:56:01.364Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/cc/ce4939f4b316457a083dc5718b3982801e8c33f921b3c98e7a93b7c7491f/pyarrow-21.0.0-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:a7f6524e3747e35f80744537c78e7302cd41deee8baa668d56d55f77d9c464b3", size = 31211248, upload-time = "2025-07-18T00:56:59.7Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/c2/7a860931420d73985e2f340f06516b21740c15b28d24a0e99a900bb27d2b/pyarrow-21.0.0-cp39-cp39-macosx_12_0_x86_64.whl", hash = "sha256:203003786c9fd253ebcafa44b03c06983c9c8d06c3145e37f1b76a1f317aeae1", size = 32676896, upload-time = "2025-07-18T00:57:03.884Z" },
+    { url = "https://files.pythonhosted.org/packages/68/a8/197f989b9a75e59b4ca0db6a13c56f19a0ad8a298c68da9cc28145e0bb97/pyarrow-21.0.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:3b4d97e297741796fead24867a8dabf86c87e4584ccc03167e4a811f50fdf74d", size = 41067862, upload-time = "2025-07-18T00:57:07.587Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/82/6ecfa89487b35aa21accb014b64e0a6b814cc860d5e3170287bf5135c7d8/pyarrow-21.0.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:898afce396b80fdda05e3086b4256f8677c671f7b1d27a6976fa011d3fd0a86e", size = 42747508, upload-time = "2025-07-18T00:57:13.917Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/b7/ba252f399bbf3addc731e8643c05532cf32e74cebb5e32f8f7409bc243cf/pyarrow-21.0.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:067c66ca29aaedae08218569a114e413b26e742171f526e828e1064fcdec13f4", size = 43345293, upload-time = "2025-07-18T00:57:19.828Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/0a/a20819795bd702b9486f536a8eeb70a6aa64046fce32071c19ec8230dbaa/pyarrow-21.0.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:0c4e75d13eb76295a49e0ea056eb18dbd87d81450bfeb8afa19a7e5a75ae2ad7", size = 45060670, upload-time = "2025-07-18T00:57:24.477Z" },
+    { url = "https://files.pythonhosted.org/packages/10/15/6b30e77872012bbfe8265d42a01d5b3c17ef0ac0f2fae531ad91b6a6c02e/pyarrow-21.0.0-cp39-cp39-win_amd64.whl", hash = "sha256:cdc4c17afda4dab2a9c0b79148a43a7f4e1094916b3e18d8975bfd6d6d52241f", size = 26227521, upload-time = "2025-07-18T00:57:29.119Z" },
+]
+
+[[package]]
+name = "pyasn1"
+version = "0.6.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ba/e9/01f1a64245b89f039897cb0130016d79f77d52669aae6ee7b159a6c4c018/pyasn1-0.6.1.tar.gz", hash = "sha256:6f580d2bdd84365380830acf45550f2511469f673cb4a5ae3857a3170128b034", size = 145322, upload-time = "2024-09-10T22:41:42.55Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c8/f1/d6a797abb14f6283c0ddff96bbdd46937f64122b8c925cab503dd37f8214/pyasn1-0.6.1-py3-none-any.whl", hash = "sha256:0d632f46f2ba09143da3a8afe9e33fb6f92fa2320ab7e886e2d0f7672af84629", size = 83135, upload-time = "2024-09-11T16:00:36.122Z" },
+]
+
+[[package]]
+name = "pyasn1-modules"
+version = "0.4.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pyasn1" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/e9/e6/78ebbb10a8c8e4b61a59249394a4a594c1a7af95593dc933a349c8d00964/pyasn1_modules-0.4.2.tar.gz", hash = "sha256:677091de870a80aae844b1ca6134f54652fa2c8c5a52aa396440ac3106e941e6", size = 307892, upload-time = "2025-03-28T02:41:22.17Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/47/8d/d529b5d697919ba8c11ad626e835d4039be708a35b0d22de83a269a6682c/pyasn1_modules-0.4.2-py3-none-any.whl", hash = "sha256:29253a9207ce32b64c3ac6600edc75368f98473906e8fd1043bd6b5b1de2c14a", size = 181259, upload-time = "2025-03-28T02:41:19.028Z" },
 ]
 
 [[package]]
 name = "pydantic"
-version = "2.10.5"
+version = "2.11.7"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "annotated-types" },
     { name = "pydantic-core" },
     { name = "typing-extensions" },
+    { name = "typing-inspection" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/6a/c7/ca334c2ef6f2e046b1144fe4bb2a5da8a4c574e7f2ebf7e16b34a6a2fa92/pydantic-2.10.5.tar.gz", hash = "sha256:278b38dbbaec562011d659ee05f63346951b3a248a6f3642e1bc68894ea2b4ff", size = 761287 }
+sdist = { url = "https://files.pythonhosted.org/packages/00/dd/4325abf92c39ba8623b5af936ddb36ffcfe0beae70405d456ab1fb2f5b8c/pydantic-2.11.7.tar.gz", hash = "sha256:d989c3c6cb79469287b1569f7447a17848c998458d49ebe294e975b9baf0f0db", size = 788350, upload-time = "2025-06-14T08:33:17.137Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/58/26/82663c79010b28eddf29dcdd0ea723439535fa917fce5905885c0e9ba562/pydantic-2.10.5-py3-none-any.whl", hash = "sha256:4dd4e322dbe55472cb7ca7e73f4b63574eecccf2835ffa2af9021ce113c83c53", size = 431426 },
+    { url = "https://files.pythonhosted.org/packages/6a/c0/ec2b1c8712ca690e5d61979dee872603e92b8a32f94cc1b72d53beab008a/pydantic-2.11.7-py3-none-any.whl", hash = "sha256:dde5df002701f6de26248661f6835bbe296a47bf73990135c7d07ce741b9623b", size = 444782, upload-time = "2025-06-14T08:33:14.905Z" },
 ]
 
 [[package]]
 name = "pydantic-core"
-version = "2.27.2"
+version = "2.33.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fc/01/f3e5ac5e7c25833db5eb555f7b7ab24cd6f8c322d3a3ad2d67a952dc0abc/pydantic_core-2.27.2.tar.gz", hash = "sha256:eb026e5a4c1fee05726072337ff51d1efb6f59090b7da90d30ea58625b1ffb39", size = 413443 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/3a/bc/fed5f74b5d802cf9a03e83f60f18864e90e3aed7223adaca5ffb7a8d8d64/pydantic_core-2.27.2-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:2d367ca20b2f14095a8f4fa1210f5a7b78b8a20009ecced6b12818f455b1e9fa", size = 1895938 },
-    { url = "https://files.pythonhosted.org/packages/71/2a/185aff24ce844e39abb8dd680f4e959f0006944f4a8a0ea372d9f9ae2e53/pydantic_core-2.27.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:491a2b73db93fab69731eaee494f320faa4e093dbed776be1a829c2eb222c34c", size = 1815684 },
-    { url = "https://files.pythonhosted.org/packages/c3/43/fafabd3d94d159d4f1ed62e383e264f146a17dd4d48453319fd782e7979e/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7969e133a6f183be60e9f6f56bfae753585680f3b7307a8e555a948d443cc05a", size = 1829169 },
-    { url = "https://files.pythonhosted.org/packages/a2/d1/f2dfe1a2a637ce6800b799aa086d079998959f6f1215eb4497966efd2274/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:3de9961f2a346257caf0aa508a4da705467f53778e9ef6fe744c038119737ef5", size = 1867227 },
-    { url = "https://files.pythonhosted.org/packages/7d/39/e06fcbcc1c785daa3160ccf6c1c38fea31f5754b756e34b65f74e99780b5/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e2bb4d3e5873c37bb3dd58714d4cd0b0e6238cebc4177ac8fe878f8b3aa8e74c", size = 2037695 },
-    { url = "https://files.pythonhosted.org/packages/7a/67/61291ee98e07f0650eb756d44998214231f50751ba7e13f4f325d95249ab/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:280d219beebb0752699480fe8f1dc61ab6615c2046d76b7ab7ee38858de0a4e7", size = 2741662 },
-    { url = "https://files.pythonhosted.org/packages/32/90/3b15e31b88ca39e9e626630b4c4a1f5a0dfd09076366f4219429e6786076/pydantic_core-2.27.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:47956ae78b6422cbd46f772f1746799cbb862de838fd8d1fbd34a82e05b0983a", size = 1993370 },
-    { url = "https://files.pythonhosted.org/packages/ff/83/c06d333ee3a67e2e13e07794995c1535565132940715931c1c43bfc85b11/pydantic_core-2.27.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:14d4a5c49d2f009d62a2a7140d3064f686d17a5d1a268bc641954ba181880236", size = 1996813 },
-    { url = "https://files.pythonhosted.org/packages/7c/f7/89be1c8deb6e22618a74f0ca0d933fdcb8baa254753b26b25ad3acff8f74/pydantic_core-2.27.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:337b443af21d488716f8d0b6164de833e788aa6bd7e3a39c005febc1284f4962", size = 2005287 },
-    { url = "https://files.pythonhosted.org/packages/b7/7d/8eb3e23206c00ef7feee17b83a4ffa0a623eb1a9d382e56e4aa46fd15ff2/pydantic_core-2.27.2-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:03d0f86ea3184a12f41a2d23f7ccb79cdb5a18e06993f8a45baa8dfec746f0e9", size = 2128414 },
-    { url = "https://files.pythonhosted.org/packages/4e/99/fe80f3ff8dd71a3ea15763878d464476e6cb0a2db95ff1c5c554133b6b83/pydantic_core-2.27.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:7041c36f5680c6e0f08d922aed302e98b3745d97fe1589db0a3eebf6624523af", size = 2155301 },
-    { url = "https://files.pythonhosted.org/packages/2b/a3/e50460b9a5789ca1451b70d4f52546fa9e2b420ba3bfa6100105c0559238/pydantic_core-2.27.2-cp310-cp310-win32.whl", hash = "sha256:50a68f3e3819077be2c98110c1f9dcb3817e93f267ba80a2c05bb4f8799e2ff4", size = 1816685 },
-    { url = "https://files.pythonhosted.org/packages/57/4c/a8838731cb0f2c2a39d3535376466de6049034d7b239c0202a64aaa05533/pydantic_core-2.27.2-cp310-cp310-win_amd64.whl", hash = "sha256:e0fd26b16394ead34a424eecf8a31a1f5137094cabe84a1bcb10fa6ba39d3d31", size = 1982876 },
-    { url = "https://files.pythonhosted.org/packages/c2/89/f3450af9d09d44eea1f2c369f49e8f181d742f28220f88cc4dfaae91ea6e/pydantic_core-2.27.2-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:8e10c99ef58cfdf2a66fc15d66b16c4a04f62bca39db589ae8cba08bc55331bc", size = 1893421 },
-    { url = "https://files.pythonhosted.org/packages/9e/e3/71fe85af2021f3f386da42d291412e5baf6ce7716bd7101ea49c810eda90/pydantic_core-2.27.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:26f32e0adf166a84d0cb63be85c562ca8a6fa8de28e5f0d92250c6b7e9e2aff7", size = 1814998 },
-    { url = "https://files.pythonhosted.org/packages/a6/3c/724039e0d848fd69dbf5806894e26479577316c6f0f112bacaf67aa889ac/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8c19d1ea0673cd13cc2f872f6c9ab42acc4e4f492a7ca9d3795ce2b112dd7e15", size = 1826167 },
-    { url = "https://files.pythonhosted.org/packages/2b/5b/1b29e8c1fb5f3199a9a57c1452004ff39f494bbe9bdbe9a81e18172e40d3/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5e68c4446fe0810e959cdff46ab0a41ce2f2c86d227d96dc3847af0ba7def306", size = 1865071 },
-    { url = "https://files.pythonhosted.org/packages/89/6c/3985203863d76bb7d7266e36970d7e3b6385148c18a68cc8915fd8c84d57/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d9640b0059ff4f14d1f37321b94061c6db164fbe49b334b31643e0528d100d99", size = 2036244 },
-    { url = "https://files.pythonhosted.org/packages/0e/41/f15316858a246b5d723f7d7f599f79e37493b2e84bfc789e58d88c209f8a/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:40d02e7d45c9f8af700f3452f329ead92da4c5f4317ca9b896de7ce7199ea459", size = 2737470 },
-    { url = "https://files.pythonhosted.org/packages/a8/7c/b860618c25678bbd6d1d99dbdfdf0510ccb50790099b963ff78a124b754f/pydantic_core-2.27.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1c1fd185014191700554795c99b347d64f2bb637966c4cfc16998a0ca700d048", size = 1992291 },
-    { url = "https://files.pythonhosted.org/packages/bf/73/42c3742a391eccbeab39f15213ecda3104ae8682ba3c0c28069fbcb8c10d/pydantic_core-2.27.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:d81d2068e1c1228a565af076598f9e7451712700b673de8f502f0334f281387d", size = 1994613 },
-    { url = "https://files.pythonhosted.org/packages/94/7a/941e89096d1175d56f59340f3a8ebaf20762fef222c298ea96d36a6328c5/pydantic_core-2.27.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:1a4207639fb02ec2dbb76227d7c751a20b1a6b4bc52850568e52260cae64ca3b", size = 2002355 },
-    { url = "https://files.pythonhosted.org/packages/6e/95/2359937a73d49e336a5a19848713555605d4d8d6940c3ec6c6c0ca4dcf25/pydantic_core-2.27.2-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:3de3ce3c9ddc8bbd88f6e0e304dea0e66d843ec9de1b0042b0911c1663ffd474", size = 2126661 },
-    { url = "https://files.pythonhosted.org/packages/2b/4c/ca02b7bdb6012a1adef21a50625b14f43ed4d11f1fc237f9d7490aa5078c/pydantic_core-2.27.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:30c5f68ded0c36466acede341551106821043e9afaad516adfb6e8fa80a4e6a6", size = 2153261 },
-    { url = "https://files.pythonhosted.org/packages/72/9d/a241db83f973049a1092a079272ffe2e3e82e98561ef6214ab53fe53b1c7/pydantic_core-2.27.2-cp311-cp311-win32.whl", hash = "sha256:c70c26d2c99f78b125a3459f8afe1aed4d9687c24fd677c6a4436bc042e50d6c", size = 1812361 },
-    { url = "https://files.pythonhosted.org/packages/e8/ef/013f07248041b74abd48a385e2110aa3a9bbfef0fbd97d4e6d07d2f5b89a/pydantic_core-2.27.2-cp311-cp311-win_amd64.whl", hash = "sha256:08e125dbdc505fa69ca7d9c499639ab6407cfa909214d500897d02afb816e7cc", size = 1982484 },
-    { url = "https://files.pythonhosted.org/packages/10/1c/16b3a3e3398fd29dca77cea0a1d998d6bde3902fa2706985191e2313cc76/pydantic_core-2.27.2-cp311-cp311-win_arm64.whl", hash = "sha256:26f0d68d4b235a2bae0c3fc585c585b4ecc51382db0e3ba402a22cbc440915e4", size = 1867102 },
-    { url = "https://files.pythonhosted.org/packages/d6/74/51c8a5482ca447871c93e142d9d4a92ead74de6c8dc5e66733e22c9bba89/pydantic_core-2.27.2-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:9e0c8cfefa0ef83b4da9588448b6d8d2a2bf1a53c3f1ae5fca39eb3061e2f0b0", size = 1893127 },
-    { url = "https://files.pythonhosted.org/packages/d3/f3/c97e80721735868313c58b89d2de85fa80fe8dfeeed84dc51598b92a135e/pydantic_core-2.27.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:83097677b8e3bd7eaa6775720ec8e0405f1575015a463285a92bfdfe254529ef", size = 1811340 },
-    { url = "https://files.pythonhosted.org/packages/9e/91/840ec1375e686dbae1bd80a9e46c26a1e0083e1186abc610efa3d9a36180/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:172fce187655fece0c90d90a678424b013f8fbb0ca8b036ac266749c09438cb7", size = 1822900 },
-    { url = "https://files.pythonhosted.org/packages/f6/31/4240bc96025035500c18adc149aa6ffdf1a0062a4b525c932065ceb4d868/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:519f29f5213271eeeeb3093f662ba2fd512b91c5f188f3bb7b27bc5973816934", size = 1869177 },
-    { url = "https://files.pythonhosted.org/packages/fa/20/02fbaadb7808be578317015c462655c317a77a7c8f0ef274bc016a784c54/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:05e3a55d124407fffba0dd6b0c0cd056d10e983ceb4e5dbd10dda135c31071d6", size = 2038046 },
-    { url = "https://files.pythonhosted.org/packages/06/86/7f306b904e6c9eccf0668248b3f272090e49c275bc488a7b88b0823444a4/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9c3ed807c7b91de05e63930188f19e921d1fe90de6b4f5cd43ee7fcc3525cb8c", size = 2685386 },
-    { url = "https://files.pythonhosted.org/packages/8d/f0/49129b27c43396581a635d8710dae54a791b17dfc50c70164866bbf865e3/pydantic_core-2.27.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6fb4aadc0b9a0c063206846d603b92030eb6f03069151a625667f982887153e2", size = 1997060 },
-    { url = "https://files.pythonhosted.org/packages/0d/0f/943b4af7cd416c477fd40b187036c4f89b416a33d3cc0ab7b82708a667aa/pydantic_core-2.27.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:28ccb213807e037460326424ceb8b5245acb88f32f3d2777427476e1b32c48c4", size = 2004870 },
-    { url = "https://files.pythonhosted.org/packages/35/40/aea70b5b1a63911c53a4c8117c0a828d6790483f858041f47bab0b779f44/pydantic_core-2.27.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:de3cd1899e2c279b140adde9357c4495ed9d47131b4a4eaff9052f23398076b3", size = 1999822 },
-    { url = "https://files.pythonhosted.org/packages/f2/b3/807b94fd337d58effc5498fd1a7a4d9d59af4133e83e32ae39a96fddec9d/pydantic_core-2.27.2-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:220f892729375e2d736b97d0e51466252ad84c51857d4d15f5e9692f9ef12be4", size = 2130364 },
-    { url = "https://files.pythonhosted.org/packages/fc/df/791c827cd4ee6efd59248dca9369fb35e80a9484462c33c6649a8d02b565/pydantic_core-2.27.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:a0fcd29cd6b4e74fe8ddd2c90330fd8edf2e30cb52acda47f06dd615ae72da57", size = 2158303 },
-    { url = "https://files.pythonhosted.org/packages/9b/67/4e197c300976af185b7cef4c02203e175fb127e414125916bf1128b639a9/pydantic_core-2.27.2-cp312-cp312-win32.whl", hash = "sha256:1e2cb691ed9834cd6a8be61228471d0a503731abfb42f82458ff27be7b2186fc", size = 1834064 },
-    { url = "https://files.pythonhosted.org/packages/1f/ea/cd7209a889163b8dcca139fe32b9687dd05249161a3edda62860430457a5/pydantic_core-2.27.2-cp312-cp312-win_amd64.whl", hash = "sha256:cc3f1a99a4f4f9dd1de4fe0312c114e740b5ddead65bb4102884b384c15d8bc9", size = 1989046 },
-    { url = "https://files.pythonhosted.org/packages/bc/49/c54baab2f4658c26ac633d798dab66b4c3a9bbf47cff5284e9c182f4137a/pydantic_core-2.27.2-cp312-cp312-win_arm64.whl", hash = "sha256:3911ac9284cd8a1792d3cb26a2da18f3ca26c6908cc434a18f730dc0db7bfa3b", size = 1885092 },
-    { url = "https://files.pythonhosted.org/packages/41/b1/9bc383f48f8002f99104e3acff6cba1231b29ef76cfa45d1506a5cad1f84/pydantic_core-2.27.2-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:7d14bd329640e63852364c306f4d23eb744e0f8193148d4044dd3dacdaacbd8b", size = 1892709 },
-    { url = "https://files.pythonhosted.org/packages/10/6c/e62b8657b834f3eb2961b49ec8e301eb99946245e70bf42c8817350cbefc/pydantic_core-2.27.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:82f91663004eb8ed30ff478d77c4d1179b3563df6cdb15c0817cd1cdaf34d154", size = 1811273 },
-    { url = "https://files.pythonhosted.org/packages/ba/15/52cfe49c8c986e081b863b102d6b859d9defc63446b642ccbbb3742bf371/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:71b24c7d61131bb83df10cc7e687433609963a944ccf45190cfc21e0887b08c9", size = 1823027 },
-    { url = "https://files.pythonhosted.org/packages/b1/1c/b6f402cfc18ec0024120602bdbcebc7bdd5b856528c013bd4d13865ca473/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:fa8e459d4954f608fa26116118bb67f56b93b209c39b008277ace29937453dc9", size = 1868888 },
-    { url = "https://files.pythonhosted.org/packages/bd/7b/8cb75b66ac37bc2975a3b7de99f3c6f355fcc4d89820b61dffa8f1e81677/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ce8918cbebc8da707ba805b7fd0b382816858728ae7fe19a942080c24e5b7cd1", size = 2037738 },
-    { url = "https://files.pythonhosted.org/packages/c8/f1/786d8fe78970a06f61df22cba58e365ce304bf9b9f46cc71c8c424e0c334/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eda3f5c2a021bbc5d976107bb302e0131351c2ba54343f8a496dc8783d3d3a6a", size = 2685138 },
-    { url = "https://files.pythonhosted.org/packages/a6/74/d12b2cd841d8724dc8ffb13fc5cef86566a53ed358103150209ecd5d1999/pydantic_core-2.27.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bd8086fa684c4775c27f03f062cbb9eaa6e17f064307e86b21b9e0abc9c0f02e", size = 1997025 },
-    { url = "https://files.pythonhosted.org/packages/a0/6e/940bcd631bc4d9a06c9539b51f070b66e8f370ed0933f392db6ff350d873/pydantic_core-2.27.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:8d9b3388db186ba0c099a6d20f0604a44eabdeef1777ddd94786cdae158729e4", size = 2004633 },
-    { url = "https://files.pythonhosted.org/packages/50/cc/a46b34f1708d82498c227d5d80ce615b2dd502ddcfd8376fc14a36655af1/pydantic_core-2.27.2-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:7a66efda2387de898c8f38c0cf7f14fca0b51a8ef0b24bfea5849f1b3c95af27", size = 1999404 },
-    { url = "https://files.pythonhosted.org/packages/ca/2d/c365cfa930ed23bc58c41463bae347d1005537dc8db79e998af8ba28d35e/pydantic_core-2.27.2-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:18a101c168e4e092ab40dbc2503bdc0f62010e95d292b27827871dc85450d7ee", size = 2130130 },
-    { url = "https://files.pythonhosted.org/packages/f4/d7/eb64d015c350b7cdb371145b54d96c919d4db516817f31cd1c650cae3b21/pydantic_core-2.27.2-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:ba5dd002f88b78a4215ed2f8ddbdf85e8513382820ba15ad5ad8955ce0ca19a1", size = 2157946 },
-    { url = "https://files.pythonhosted.org/packages/a4/99/bddde3ddde76c03b65dfd5a66ab436c4e58ffc42927d4ff1198ffbf96f5f/pydantic_core-2.27.2-cp313-cp313-win32.whl", hash = "sha256:1ebaf1d0481914d004a573394f4be3a7616334be70261007e47c2a6fe7e50130", size = 1834387 },
-    { url = "https://files.pythonhosted.org/packages/71/47/82b5e846e01b26ac6f1893d3c5f9f3a2eb6ba79be26eef0b759b4fe72946/pydantic_core-2.27.2-cp313-cp313-win_amd64.whl", hash = "sha256:953101387ecf2f5652883208769a79e48db18c6df442568a0b5ccd8c2723abee", size = 1990453 },
-    { url = "https://files.pythonhosted.org/packages/51/b2/b2b50d5ecf21acf870190ae5d093602d95f66c9c31f9d5de6062eb329ad1/pydantic_core-2.27.2-cp313-cp313-win_arm64.whl", hash = "sha256:ac4dbfd1691affb8f48c2c13241a2e3b60ff23247cbcf981759c768b6633cf8b", size = 1885186 },
-    { url = "https://files.pythonhosted.org/packages/27/97/3aef1ddb65c5ccd6eda9050036c956ff6ecbfe66cb7eb40f280f121a5bb0/pydantic_core-2.27.2-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:c10eb4f1659290b523af58fa7cffb452a61ad6ae5613404519aee4bfbf1df993", size = 1896475 },
-    { url = "https://files.pythonhosted.org/packages/ad/d3/5668da70e373c9904ed2f372cb52c0b996426f302e0dee2e65634c92007d/pydantic_core-2.27.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:ef592d4bad47296fb11f96cd7dc898b92e795032b4894dfb4076cfccd43a9308", size = 1772279 },
-    { url = "https://files.pythonhosted.org/packages/8a/9e/e44b8cb0edf04a2f0a1f6425a65ee089c1d6f9c4c2dcab0209127b6fdfc2/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c61709a844acc6bf0b7dce7daae75195a10aac96a596ea1b776996414791ede4", size = 1829112 },
-    { url = "https://files.pythonhosted.org/packages/1c/90/1160d7ac700102effe11616e8119e268770f2a2aa5afb935f3ee6832987d/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:42c5f762659e47fdb7b16956c71598292f60a03aa92f8b6351504359dbdba6cf", size = 1866780 },
-    { url = "https://files.pythonhosted.org/packages/ee/33/13983426df09a36d22c15980008f8d9c77674fc319351813b5a2739b70f3/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4c9775e339e42e79ec99c441d9730fccf07414af63eac2f0e48e08fd38a64d76", size = 2037943 },
-    { url = "https://files.pythonhosted.org/packages/01/d7/ced164e376f6747e9158c89988c293cd524ab8d215ae4e185e9929655d5c/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:57762139821c31847cfb2df63c12f725788bd9f04bc2fb392790959b8f70f118", size = 2740492 },
-    { url = "https://files.pythonhosted.org/packages/8b/1f/3dc6e769d5b7461040778816aab2b00422427bcaa4b56cc89e9c653b2605/pydantic_core-2.27.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0d1e85068e818c73e048fe28cfc769040bb1f475524f4745a5dc621f75ac7630", size = 1995714 },
-    { url = "https://files.pythonhosted.org/packages/07/d7/a0bd09bc39283530b3f7c27033a814ef254ba3bd0b5cfd040b7abf1fe5da/pydantic_core-2.27.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:097830ed52fd9e427942ff3b9bc17fab52913b2f50f2880dc4a5611446606a54", size = 1997163 },
-    { url = "https://files.pythonhosted.org/packages/2d/bb/2db4ad1762e1c5699d9b857eeb41959191980de6feb054e70f93085e1bcd/pydantic_core-2.27.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:044a50963a614ecfae59bb1eaf7ea7efc4bc62f49ed594e18fa1e5d953c40e9f", size = 2005217 },
-    { url = "https://files.pythonhosted.org/packages/53/5f/23a5a3e7b8403f8dd8fc8a6f8b49f6b55c7d715b77dcf1f8ae919eeb5628/pydantic_core-2.27.2-cp39-cp39-musllinux_1_1_armv7l.whl", hash = "sha256:4e0b4220ba5b40d727c7f879eac379b822eee5d8fff418e9d3381ee45b3b0362", size = 2127899 },
-    { url = "https://files.pythonhosted.org/packages/c2/ae/aa38bb8dd3d89c2f1d8362dd890ee8f3b967330821d03bbe08fa01ce3766/pydantic_core-2.27.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:5e4f4bb20d75e9325cc9696c6802657b58bc1dbbe3022f32cc2b2b632c3fbb96", size = 2155726 },
-    { url = "https://files.pythonhosted.org/packages/98/61/4f784608cc9e98f70839187117ce840480f768fed5d386f924074bf6213c/pydantic_core-2.27.2-cp39-cp39-win32.whl", hash = "sha256:cca63613e90d001b9f2f9a9ceb276c308bfa2a43fafb75c8031c4f66039e8c6e", size = 1817219 },
-    { url = "https://files.pythonhosted.org/packages/57/82/bb16a68e4a1a858bb3768c2c8f1ff8d8978014e16598f001ea29a25bf1d1/pydantic_core-2.27.2-cp39-cp39-win_amd64.whl", hash = "sha256:77d1bca19b0f7021b3a982e6f903dcd5b2b06076def36a652e3907f596e29f67", size = 1985382 },
-    { url = "https://files.pythonhosted.org/packages/46/72/af70981a341500419e67d5cb45abe552a7c74b66326ac8877588488da1ac/pydantic_core-2.27.2-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:2bf14caea37e91198329b828eae1618c068dfb8ef17bb33287a7ad4b61ac314e", size = 1891159 },
-    { url = "https://files.pythonhosted.org/packages/ad/3d/c5913cccdef93e0a6a95c2d057d2c2cba347815c845cda79ddd3c0f5e17d/pydantic_core-2.27.2-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:b0cb791f5b45307caae8810c2023a184c74605ec3bcbb67d13846c28ff731ff8", size = 1768331 },
-    { url = "https://files.pythonhosted.org/packages/f6/f0/a3ae8fbee269e4934f14e2e0e00928f9346c5943174f2811193113e58252/pydantic_core-2.27.2-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:688d3fd9fcb71f41c4c015c023d12a79d1c4c0732ec9eb35d96e3388a120dcf3", size = 1822467 },
-    { url = "https://files.pythonhosted.org/packages/d7/7a/7bbf241a04e9f9ea24cd5874354a83526d639b02674648af3f350554276c/pydantic_core-2.27.2-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3d591580c34f4d731592f0e9fe40f9cc1b430d297eecc70b962e93c5c668f15f", size = 1979797 },
-    { url = "https://files.pythonhosted.org/packages/4f/5f/4784c6107731f89e0005a92ecb8a2efeafdb55eb992b8e9d0a2be5199335/pydantic_core-2.27.2-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:82f986faf4e644ffc189a7f1aafc86e46ef70372bb153e7001e8afccc6e54133", size = 1987839 },
-    { url = "https://files.pythonhosted.org/packages/6d/a7/61246562b651dff00de86a5f01b6e4befb518df314c54dec187a78d81c84/pydantic_core-2.27.2-pp310-pypy310_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:bec317a27290e2537f922639cafd54990551725fc844249e64c523301d0822fc", size = 1998861 },
-    { url = "https://files.pythonhosted.org/packages/86/aa/837821ecf0c022bbb74ca132e117c358321e72e7f9702d1b6a03758545e2/pydantic_core-2.27.2-pp310-pypy310_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:0296abcb83a797db256b773f45773da397da75a08f5fcaef41f2044adec05f50", size = 2116582 },
-    { url = "https://files.pythonhosted.org/packages/81/b0/5e74656e95623cbaa0a6278d16cf15e10a51f6002e3ec126541e95c29ea3/pydantic_core-2.27.2-pp310-pypy310_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:0d75070718e369e452075a6017fbf187f788e17ed67a3abd47fa934d001863d9", size = 2151985 },
-    { url = "https://files.pythonhosted.org/packages/63/37/3e32eeb2a451fddaa3898e2163746b0cffbbdbb4740d38372db0490d67f3/pydantic_core-2.27.2-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:7e17b560be3c98a8e3aa66ce828bdebb9e9ac6ad5466fba92eb74c4c95cb1151", size = 2004715 },
-    { url = "https://files.pythonhosted.org/packages/29/0e/dcaea00c9dbd0348b723cae82b0e0c122e0fa2b43fa933e1622fd237a3ee/pydantic_core-2.27.2-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:c33939a82924da9ed65dab5a65d427205a73181d8098e79b6b426bdf8ad4e656", size = 1891733 },
-    { url = "https://files.pythonhosted.org/packages/86/d3/e797bba8860ce650272bda6383a9d8cad1d1c9a75a640c9d0e848076f85e/pydantic_core-2.27.2-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:00bad2484fa6bda1e216e7345a798bd37c68fb2d97558edd584942aa41b7d278", size = 1768375 },
-    { url = "https://files.pythonhosted.org/packages/41/f7/f847b15fb14978ca2b30262548f5fc4872b2724e90f116393eb69008299d/pydantic_core-2.27.2-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c817e2b40aba42bac6f457498dacabc568c3b7a986fc9ba7c8d9d260b71485fb", size = 1822307 },
-    { url = "https://files.pythonhosted.org/packages/9c/63/ed80ec8255b587b2f108e514dc03eed1546cd00f0af281e699797f373f38/pydantic_core-2.27.2-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:251136cdad0cb722e93732cb45ca5299fb56e1344a833640bf93b2803f8d1bfd", size = 1979971 },
-    { url = "https://files.pythonhosted.org/packages/a9/6d/6d18308a45454a0de0e975d70171cadaf454bc7a0bf86b9c7688e313f0bb/pydantic_core-2.27.2-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:d2088237af596f0a524d3afc39ab3b036e8adb054ee57cbb1dcf8e09da5b29cc", size = 1987616 },
-    { url = "https://files.pythonhosted.org/packages/82/8a/05f8780f2c1081b800a7ca54c1971e291c2d07d1a50fb23c7e4aef4ed403/pydantic_core-2.27.2-pp39-pypy39_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:d4041c0b966a84b4ae7a09832eb691a35aec90910cd2dbe7a208de59be77965b", size = 1998943 },
-    { url = "https://files.pythonhosted.org/packages/5e/3e/fe5b6613d9e4c0038434396b46c5303f5ade871166900b357ada4766c5b7/pydantic_core-2.27.2-pp39-pypy39_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:8083d4e875ebe0b864ffef72a4304827015cff328a1be6e22cc850753bfb122b", size = 2116654 },
-    { url = "https://files.pythonhosted.org/packages/db/ad/28869f58938fad8cc84739c4e592989730bfb69b7c90a8fff138dff18e1e/pydantic_core-2.27.2-pp39-pypy39_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:f141ee28a0ad2123b6611b6ceff018039df17f32ada8b534e6aa039545a3efb2", size = 2152292 },
-    { url = "https://files.pythonhosted.org/packages/a1/0c/c5c5cd3689c32ed1fe8c5d234b079c12c281c051759770c05b8bed6412b5/pydantic_core-2.27.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:7d0c8399fcc1848491f00e0314bd59fb34a9c008761bcb422a057670c3f65e35", size = 2004961 },
+sdist = { url = "https://files.pythonhosted.org/packages/ad/88/5f2260bdfae97aabf98f1778d43f69574390ad787afb646292a638c923d4/pydantic_core-2.33.2.tar.gz", hash = "sha256:7cb8bc3605c29176e1b105350d2e6474142d7c1bd1d9327c4a9bdb46bf827acc", size = 435195, upload-time = "2025-04-23T18:33:52.104Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e5/92/b31726561b5dae176c2d2c2dc43a9c5bfba5d32f96f8b4c0a600dd492447/pydantic_core-2.33.2-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:2b3d326aaef0c0399d9afffeb6367d5e26ddc24d351dbc9c636840ac355dc5d8", size = 2028817, upload-time = "2025-04-23T18:30:43.919Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/44/3f0b95fafdaca04a483c4e685fe437c6891001bf3ce8b2fded82b9ea3aa1/pydantic_core-2.33.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0e5b2671f05ba48b94cb90ce55d8bdcaaedb8ba00cc5359f6810fc918713983d", size = 1861357, upload-time = "2025-04-23T18:30:46.372Z" },
+    { url = "https://files.pythonhosted.org/packages/30/97/e8f13b55766234caae05372826e8e4b3b96e7b248be3157f53237682e43c/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0069c9acc3f3981b9ff4cdfaf088e98d83440a4c7ea1bc07460af3d4dc22e72d", size = 1898011, upload-time = "2025-04-23T18:30:47.591Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/a3/99c48cf7bafc991cc3ee66fd544c0aae8dc907b752f1dad2d79b1b5a471f/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:d53b22f2032c42eaaf025f7c40c2e3b94568ae077a606f006d206a463bc69572", size = 1982730, upload-time = "2025-04-23T18:30:49.328Z" },
+    { url = "https://files.pythonhosted.org/packages/de/8e/a5b882ec4307010a840fb8b58bd9bf65d1840c92eae7534c7441709bf54b/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0405262705a123b7ce9f0b92f123334d67b70fd1f20a9372b907ce1080c7ba02", size = 2136178, upload-time = "2025-04-23T18:30:50.907Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/bb/71e35fc3ed05af6834e890edb75968e2802fe98778971ab5cba20a162315/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:4b25d91e288e2c4e0662b8038a28c6a07eaac3e196cfc4ff69de4ea3db992a1b", size = 2736462, upload-time = "2025-04-23T18:30:52.083Z" },
+    { url = "https://files.pythonhosted.org/packages/31/0d/c8f7593e6bc7066289bbc366f2235701dcbebcd1ff0ef8e64f6f239fb47d/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6bdfe4b3789761f3bcb4b1ddf33355a71079858958e3a552f16d5af19768fef2", size = 2005652, upload-time = "2025-04-23T18:30:53.389Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/7a/996d8bd75f3eda405e3dd219ff5ff0a283cd8e34add39d8ef9157e722867/pydantic_core-2.33.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:efec8db3266b76ef9607c2c4c419bdb06bf335ae433b80816089ea7585816f6a", size = 2113306, upload-time = "2025-04-23T18:30:54.661Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/84/daf2a6fb2db40ffda6578a7e8c5a6e9c8affb251a05c233ae37098118788/pydantic_core-2.33.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:031c57d67ca86902726e0fae2214ce6770bbe2f710dc33063187a68744a5ecac", size = 2073720, upload-time = "2025-04-23T18:30:56.11Z" },
+    { url = "https://files.pythonhosted.org/packages/77/fb/2258da019f4825128445ae79456a5499c032b55849dbd5bed78c95ccf163/pydantic_core-2.33.2-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:f8de619080e944347f5f20de29a975c2d815d9ddd8be9b9b7268e2e3ef68605a", size = 2244915, upload-time = "2025-04-23T18:30:57.501Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/7a/925ff73756031289468326e355b6fa8316960d0d65f8b5d6b3a3e7866de7/pydantic_core-2.33.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:73662edf539e72a9440129f231ed3757faab89630d291b784ca99237fb94db2b", size = 2241884, upload-time = "2025-04-23T18:30:58.867Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/b0/249ee6d2646f1cdadcb813805fe76265745c4010cf20a8eba7b0e639d9b2/pydantic_core-2.33.2-cp310-cp310-win32.whl", hash = "sha256:0a39979dcbb70998b0e505fb1556a1d550a0781463ce84ebf915ba293ccb7e22", size = 1910496, upload-time = "2025-04-23T18:31:00.078Z" },
+    { url = "https://files.pythonhosted.org/packages/66/ff/172ba8f12a42d4b552917aa65d1f2328990d3ccfc01d5b7c943ec084299f/pydantic_core-2.33.2-cp310-cp310-win_amd64.whl", hash = "sha256:b0379a2b24882fef529ec3b4987cb5d003b9cda32256024e6fe1586ac45fc640", size = 1955019, upload-time = "2025-04-23T18:31:01.335Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/8d/71db63483d518cbbf290261a1fc2839d17ff89fce7089e08cad07ccfce67/pydantic_core-2.33.2-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:4c5b0a576fb381edd6d27f0a85915c6daf2f8138dc5c267a57c08a62900758c7", size = 2028584, upload-time = "2025-04-23T18:31:03.106Z" },
+    { url = "https://files.pythonhosted.org/packages/24/2f/3cfa7244ae292dd850989f328722d2aef313f74ffc471184dc509e1e4e5a/pydantic_core-2.33.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:e799c050df38a639db758c617ec771fd8fb7a5f8eaaa4b27b101f266b216a246", size = 1855071, upload-time = "2025-04-23T18:31:04.621Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/d3/4ae42d33f5e3f50dd467761304be2fa0a9417fbf09735bc2cce003480f2a/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dc46a01bf8d62f227d5ecee74178ffc448ff4e5197c756331f71efcc66dc980f", size = 1897823, upload-time = "2025-04-23T18:31:06.377Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/f3/aa5976e8352b7695ff808599794b1fba2a9ae2ee954a3426855935799488/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:a144d4f717285c6d9234a66778059f33a89096dfb9b39117663fd8413d582dcc", size = 1983792, upload-time = "2025-04-23T18:31:07.93Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/7a/cda9b5a23c552037717f2b2a5257e9b2bfe45e687386df9591eff7b46d28/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:73cf6373c21bc80b2e0dc88444f41ae60b2f070ed02095754eb5a01df12256de", size = 2136338, upload-time = "2025-04-23T18:31:09.283Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/9f/b8f9ec8dd1417eb9da784e91e1667d58a2a4a7b7b34cf4af765ef663a7e5/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3dc625f4aa79713512d1976fe9f0bc99f706a9dee21dfd1810b4bbbf228d0e8a", size = 2730998, upload-time = "2025-04-23T18:31:11.7Z" },
+    { url = "https://files.pythonhosted.org/packages/47/bc/cd720e078576bdb8255d5032c5d63ee5c0bf4b7173dd955185a1d658c456/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:881b21b5549499972441da4758d662aeea93f1923f953e9cbaff14b8b9565aef", size = 2003200, upload-time = "2025-04-23T18:31:13.536Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/22/3602b895ee2cd29d11a2b349372446ae9727c32e78a94b3d588a40fdf187/pydantic_core-2.33.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:bdc25f3681f7b78572699569514036afe3c243bc3059d3942624e936ec93450e", size = 2113890, upload-time = "2025-04-23T18:31:15.011Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/e6/e3c5908c03cf00d629eb38393a98fccc38ee0ce8ecce32f69fc7d7b558a7/pydantic_core-2.33.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:fe5b32187cbc0c862ee201ad66c30cf218e5ed468ec8dc1cf49dec66e160cc4d", size = 2073359, upload-time = "2025-04-23T18:31:16.393Z" },
+    { url = "https://files.pythonhosted.org/packages/12/e7/6a36a07c59ebefc8777d1ffdaf5ae71b06b21952582e4b07eba88a421c79/pydantic_core-2.33.2-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:bc7aee6f634a6f4a95676fcb5d6559a2c2a390330098dba5e5a5f28a2e4ada30", size = 2245883, upload-time = "2025-04-23T18:31:17.892Z" },
+    { url = "https://files.pythonhosted.org/packages/16/3f/59b3187aaa6cc0c1e6616e8045b284de2b6a87b027cce2ffcea073adf1d2/pydantic_core-2.33.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:235f45e5dbcccf6bd99f9f472858849f73d11120d76ea8707115415f8e5ebebf", size = 2241074, upload-time = "2025-04-23T18:31:19.205Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/ed/55532bb88f674d5d8f67ab121a2a13c385df382de2a1677f30ad385f7438/pydantic_core-2.33.2-cp311-cp311-win32.whl", hash = "sha256:6368900c2d3ef09b69cb0b913f9f8263b03786e5b2a387706c5afb66800efd51", size = 1910538, upload-time = "2025-04-23T18:31:20.541Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/1b/25b7cccd4519c0b23c2dd636ad39d381abf113085ce4f7bec2b0dc755eb1/pydantic_core-2.33.2-cp311-cp311-win_amd64.whl", hash = "sha256:1e063337ef9e9820c77acc768546325ebe04ee38b08703244c1309cccc4f1bab", size = 1952909, upload-time = "2025-04-23T18:31:22.371Z" },
+    { url = "https://files.pythonhosted.org/packages/49/a9/d809358e49126438055884c4366a1f6227f0f84f635a9014e2deb9b9de54/pydantic_core-2.33.2-cp311-cp311-win_arm64.whl", hash = "sha256:6b99022f1d19bc32a4c2a0d544fc9a76e3be90f0b3f4af413f87d38749300e65", size = 1897786, upload-time = "2025-04-23T18:31:24.161Z" },
+    { url = "https://files.pythonhosted.org/packages/18/8a/2b41c97f554ec8c71f2a8a5f85cb56a8b0956addfe8b0efb5b3d77e8bdc3/pydantic_core-2.33.2-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:a7ec89dc587667f22b6a0b6579c249fca9026ce7c333fc142ba42411fa243cdc", size = 2009000, upload-time = "2025-04-23T18:31:25.863Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/02/6224312aacb3c8ecbaa959897af57181fb6cf3a3d7917fd44d0f2917e6f2/pydantic_core-2.33.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3c6db6e52c6d70aa0d00d45cdb9b40f0433b96380071ea80b09277dba021ddf7", size = 1847996, upload-time = "2025-04-23T18:31:27.341Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/46/6dcdf084a523dbe0a0be59d054734b86a981726f221f4562aed313dbcb49/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e61206137cbc65e6d5256e1166f88331d3b6238e082d9f74613b9b765fb9025", size = 1880957, upload-time = "2025-04-23T18:31:28.956Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/6b/1ec2c03837ac00886ba8160ce041ce4e325b41d06a034adbef11339ae422/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:eb8c529b2819c37140eb51b914153063d27ed88e3bdc31b71198a198e921e011", size = 1964199, upload-time = "2025-04-23T18:31:31.025Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/1d/6bf34d6adb9debd9136bd197ca72642203ce9aaaa85cfcbfcf20f9696e83/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c52b02ad8b4e2cf14ca7b3d918f3eb0ee91e63b3167c32591e57c4317e134f8f", size = 2120296, upload-time = "2025-04-23T18:31:32.514Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/94/2bd0aaf5a591e974b32a9f7123f16637776c304471a0ab33cf263cf5591a/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:96081f1605125ba0855dfda83f6f3df5ec90c61195421ba72223de35ccfb2f88", size = 2676109, upload-time = "2025-04-23T18:31:33.958Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/41/4b043778cf9c4285d59742281a769eac371b9e47e35f98ad321349cc5d61/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8f57a69461af2a5fa6e6bbd7a5f60d3b7e6cebb687f55106933188e79ad155c1", size = 2002028, upload-time = "2025-04-23T18:31:39.095Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/d5/7bb781bf2748ce3d03af04d5c969fa1308880e1dca35a9bd94e1a96a922e/pydantic_core-2.33.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:572c7e6c8bb4774d2ac88929e3d1f12bc45714ae5ee6d9a788a9fb35e60bb04b", size = 2100044, upload-time = "2025-04-23T18:31:41.034Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/36/def5e53e1eb0ad896785702a5bbfd25eed546cdcf4087ad285021a90ed53/pydantic_core-2.33.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:db4b41f9bd95fbe5acd76d89920336ba96f03e149097365afe1cb092fceb89a1", size = 2058881, upload-time = "2025-04-23T18:31:42.757Z" },
+    { url = "https://files.pythonhosted.org/packages/01/6c/57f8d70b2ee57fc3dc8b9610315949837fa8c11d86927b9bb044f8705419/pydantic_core-2.33.2-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:fa854f5cf7e33842a892e5c73f45327760bc7bc516339fda888c75ae60edaeb6", size = 2227034, upload-time = "2025-04-23T18:31:44.304Z" },
+    { url = "https://files.pythonhosted.org/packages/27/b9/9c17f0396a82b3d5cbea4c24d742083422639e7bb1d5bf600e12cb176a13/pydantic_core-2.33.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:5f483cfb75ff703095c59e365360cb73e00185e01aaea067cd19acffd2ab20ea", size = 2234187, upload-time = "2025-04-23T18:31:45.891Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/6a/adf5734ffd52bf86d865093ad70b2ce543415e0e356f6cacabbc0d9ad910/pydantic_core-2.33.2-cp312-cp312-win32.whl", hash = "sha256:9cb1da0f5a471435a7bc7e439b8a728e8b61e59784b2af70d7c169f8dd8ae290", size = 1892628, upload-time = "2025-04-23T18:31:47.819Z" },
+    { url = "https://files.pythonhosted.org/packages/43/e4/5479fecb3606c1368d496a825d8411e126133c41224c1e7238be58b87d7e/pydantic_core-2.33.2-cp312-cp312-win_amd64.whl", hash = "sha256:f941635f2a3d96b2973e867144fde513665c87f13fe0e193c158ac51bfaaa7b2", size = 1955866, upload-time = "2025-04-23T18:31:49.635Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/24/8b11e8b3e2be9dd82df4b11408a67c61bb4dc4f8e11b5b0fc888b38118b5/pydantic_core-2.33.2-cp312-cp312-win_arm64.whl", hash = "sha256:cca3868ddfaccfbc4bfb1d608e2ccaaebe0ae628e1416aeb9c4d88c001bb45ab", size = 1888894, upload-time = "2025-04-23T18:31:51.609Z" },
+    { url = "https://files.pythonhosted.org/packages/53/ea/bbe9095cdd771987d13c82d104a9c8559ae9aec1e29f139e286fd2e9256e/pydantic_core-2.33.2-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:a2b911a5b90e0374d03813674bf0a5fbbb7741570dcd4b4e85a2e48d17def29d", size = 2028677, upload-time = "2025-04-23T18:32:27.227Z" },
+    { url = "https://files.pythonhosted.org/packages/49/1d/4ac5ed228078737d457a609013e8f7edc64adc37b91d619ea965758369e5/pydantic_core-2.33.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:6fa6dfc3e4d1f734a34710f391ae822e0a8eb8559a85c6979e14e65ee6ba2954", size = 1864735, upload-time = "2025-04-23T18:32:29.019Z" },
+    { url = "https://files.pythonhosted.org/packages/23/9a/2e70d6388d7cda488ae38f57bc2f7b03ee442fbcf0d75d848304ac7e405b/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c54c939ee22dc8e2d545da79fc5381f1c020d6d3141d3bd747eab59164dc89fb", size = 1898467, upload-time = "2025-04-23T18:32:31.119Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/2e/1568934feb43370c1ffb78a77f0baaa5a8b6897513e7a91051af707ffdc4/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:53a57d2ed685940a504248187d5685e49eb5eef0f696853647bf37c418c538f7", size = 1983041, upload-time = "2025-04-23T18:32:33.655Z" },
+    { url = "https://files.pythonhosted.org/packages/01/1a/1a1118f38ab64eac2f6269eb8c120ab915be30e387bb561e3af904b12499/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:09fb9dd6571aacd023fe6aaca316bd01cf60ab27240d7eb39ebd66a3a15293b4", size = 2136503, upload-time = "2025-04-23T18:32:35.519Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/da/44754d1d7ae0f22d6d3ce6c6b1486fc07ac2c524ed8f6eca636e2e1ee49b/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0e6116757f7959a712db11f3e9c0a99ade00a5bbedae83cb801985aa154f071b", size = 2736079, upload-time = "2025-04-23T18:32:37.659Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/98/f43cd89172220ec5aa86654967b22d862146bc4d736b1350b4c41e7c9c03/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8d55ab81c57b8ff8548c3e4947f119551253f4e3787a7bbc0b6b3ca47498a9d3", size = 2006508, upload-time = "2025-04-23T18:32:39.637Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/cc/f77e8e242171d2158309f830f7d5d07e0531b756106f36bc18712dc439df/pydantic_core-2.33.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c20c462aa4434b33a2661701b861604913f912254e441ab8d78d30485736115a", size = 2113693, upload-time = "2025-04-23T18:32:41.818Z" },
+    { url = "https://files.pythonhosted.org/packages/54/7a/7be6a7bd43e0a47c147ba7fbf124fe8aaf1200bc587da925509641113b2d/pydantic_core-2.33.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:44857c3227d3fb5e753d5fe4a3420d6376fa594b07b621e220cd93703fe21782", size = 2074224, upload-time = "2025-04-23T18:32:44.033Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/07/31cf8fadffbb03be1cb520850e00a8490c0927ec456e8293cafda0726184/pydantic_core-2.33.2-cp39-cp39-musllinux_1_1_armv7l.whl", hash = "sha256:eb9b459ca4df0e5c87deb59d37377461a538852765293f9e6ee834f0435a93b9", size = 2245403, upload-time = "2025-04-23T18:32:45.836Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/8d/bbaf4c6721b668d44f01861f297eb01c9b35f612f6b8e14173cb204e6240/pydantic_core-2.33.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:9fcd347d2cc5c23b06de6d3b7b8275be558a0c90549495c699e379a80bf8379e", size = 2242331, upload-time = "2025-04-23T18:32:47.618Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/93/3cc157026bca8f5006250e74515119fcaa6d6858aceee8f67ab6dc548c16/pydantic_core-2.33.2-cp39-cp39-win32.whl", hash = "sha256:83aa99b1285bc8f038941ddf598501a86f1536789740991d7d8756e34f1e74d9", size = 1910571, upload-time = "2025-04-23T18:32:49.401Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/90/7edc3b2a0d9f0dda8806c04e511a67b0b7a41d2187e2003673a996fb4310/pydantic_core-2.33.2-cp39-cp39-win_amd64.whl", hash = "sha256:f481959862f57f29601ccced557cc2e817bce7533ab8e01a797a48b49c9692b3", size = 1956504, upload-time = "2025-04-23T18:32:51.287Z" },
+    { url = "https://files.pythonhosted.org/packages/30/68/373d55e58b7e83ce371691f6eaa7175e3a24b956c44628eb25d7da007917/pydantic_core-2.33.2-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:5c4aa4e82353f65e548c476b37e64189783aa5384903bfea4f41580f255fddfa", size = 2023982, upload-time = "2025-04-23T18:32:53.14Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/16/145f54ac08c96a63d8ed6442f9dec17b2773d19920b627b18d4f10a061ea/pydantic_core-2.33.2-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:d946c8bf0d5c24bf4fe333af284c59a19358aa3ec18cb3dc4370080da1e8ad29", size = 1858412, upload-time = "2025-04-23T18:32:55.52Z" },
+    { url = "https://files.pythonhosted.org/packages/41/b1/c6dc6c3e2de4516c0bb2c46f6a373b91b5660312342a0cf5826e38ad82fa/pydantic_core-2.33.2-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:87b31b6846e361ef83fedb187bb5b4372d0da3f7e28d85415efa92d6125d6e6d", size = 1892749, upload-time = "2025-04-23T18:32:57.546Z" },
+    { url = "https://files.pythonhosted.org/packages/12/73/8cd57e20afba760b21b742106f9dbdfa6697f1570b189c7457a1af4cd8a0/pydantic_core-2.33.2-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:aa9d91b338f2df0508606f7009fde642391425189bba6d8c653afd80fd6bb64e", size = 2067527, upload-time = "2025-04-23T18:32:59.771Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/d5/0bb5d988cc019b3cba4a78f2d4b3854427fc47ee8ec8e9eaabf787da239c/pydantic_core-2.33.2-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2058a32994f1fde4ca0480ab9d1e75a0e8c87c22b53a3ae66554f9af78f2fe8c", size = 2108225, upload-time = "2025-04-23T18:33:04.51Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/c5/00c02d1571913d496aabf146106ad8239dc132485ee22efe08085084ff7c/pydantic_core-2.33.2-pp310-pypy310_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:0e03262ab796d986f978f79c943fc5f620381be7287148b8010b4097f79a39ec", size = 2069490, upload-time = "2025-04-23T18:33:06.391Z" },
+    { url = "https://files.pythonhosted.org/packages/22/a8/dccc38768274d3ed3a59b5d06f59ccb845778687652daa71df0cab4040d7/pydantic_core-2.33.2-pp310-pypy310_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:1a8695a8d00c73e50bff9dfda4d540b7dee29ff9b8053e38380426a85ef10052", size = 2237525, upload-time = "2025-04-23T18:33:08.44Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/e7/4f98c0b125dda7cf7ccd14ba936218397b44f50a56dd8c16a3091df116c3/pydantic_core-2.33.2-pp310-pypy310_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:fa754d1850735a0b0e03bcffd9d4b4343eb417e47196e4485d9cca326073a42c", size = 2238446, upload-time = "2025-04-23T18:33:10.313Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/91/2ec36480fdb0b783cd9ef6795753c1dea13882f2e68e73bce76ae8c21e6a/pydantic_core-2.33.2-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:a11c8d26a50bfab49002947d3d237abe4d9e4b5bdc8846a63537b6488e197808", size = 2066678, upload-time = "2025-04-23T18:33:12.224Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/27/d4ae6487d73948d6f20dddcd94be4ea43e74349b56eba82e9bdee2d7494c/pydantic_core-2.33.2-pp311-pypy311_pp73-macosx_10_12_x86_64.whl", hash = "sha256:dd14041875d09cc0f9308e37a6f8b65f5585cf2598a53aa0123df8b129d481f8", size = 2025200, upload-time = "2025-04-23T18:33:14.199Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/b8/b3cb95375f05d33801024079b9392a5ab45267a63400bf1866e7ce0f0de4/pydantic_core-2.33.2-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:d87c561733f66531dced0da6e864f44ebf89a8fba55f31407b00c2f7f9449593", size = 1859123, upload-time = "2025-04-23T18:33:16.555Z" },
+    { url = "https://files.pythonhosted.org/packages/05/bc/0d0b5adeda59a261cd30a1235a445bf55c7e46ae44aea28f7bd6ed46e091/pydantic_core-2.33.2-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2f82865531efd18d6e07a04a17331af02cb7a651583c418df8266f17a63c6612", size = 1892852, upload-time = "2025-04-23T18:33:18.513Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/11/d37bdebbda2e449cb3f519f6ce950927b56d62f0b84fd9cb9e372a26a3d5/pydantic_core-2.33.2-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2bfb5112df54209d820d7bf9317c7a6c9025ea52e49f46b6a2060104bba37de7", size = 2067484, upload-time = "2025-04-23T18:33:20.475Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/55/1f95f0a05ce72ecb02a8a8a1c3be0579bbc29b1d5ab68f1378b7bebc5057/pydantic_core-2.33.2-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:64632ff9d614e5eecfb495796ad51b0ed98c453e447a76bcbeeb69615079fc7e", size = 2108896, upload-time = "2025-04-23T18:33:22.501Z" },
+    { url = "https://files.pythonhosted.org/packages/53/89/2b2de6c81fa131f423246a9109d7b2a375e83968ad0800d6e57d0574629b/pydantic_core-2.33.2-pp311-pypy311_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:f889f7a40498cc077332c7ab6b4608d296d852182211787d4f3ee377aaae66e8", size = 2069475, upload-time = "2025-04-23T18:33:24.528Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/e9/1f7efbe20d0b2b10f6718944b5d8ece9152390904f29a78e68d4e7961159/pydantic_core-2.33.2-pp311-pypy311_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:de4b83bb311557e439b9e186f733f6c645b9417c84e2eb8203f3f820a4b988bf", size = 2239013, upload-time = "2025-04-23T18:33:26.621Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/b2/5309c905a93811524a49b4e031e9851a6b00ff0fb668794472ea7746b448/pydantic_core-2.33.2-pp311-pypy311_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:82f68293f055f51b51ea42fafc74b6aad03e70e191799430b90c13d643059ebb", size = 2238715, upload-time = "2025-04-23T18:33:28.656Z" },
+    { url = "https://files.pythonhosted.org/packages/32/56/8a7ca5d2cd2cda1d245d34b1c9a942920a718082ae8e54e5f3e5a58b7add/pydantic_core-2.33.2-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:329467cecfb529c925cf2bbd4d60d2c509bc2fb52a20c1045bf09bb70971a9c1", size = 2066757, upload-time = "2025-04-23T18:33:30.645Z" },
+    { url = "https://files.pythonhosted.org/packages/08/98/dbf3fdfabaf81cda5622154fda78ea9965ac467e3239078e0dcd6df159e7/pydantic_core-2.33.2-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:87acbfcf8e90ca885206e98359d7dca4bcbb35abdc0ff66672a293e1d7a19101", size = 2024034, upload-time = "2025-04-23T18:33:32.843Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/99/7810aa9256e7f2ccd492590f86b79d370df1e9292f1f80b000b6a75bd2fb/pydantic_core-2.33.2-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:7f92c15cd1e97d4b12acd1cc9004fa092578acfa57b67ad5e43a197175d01a64", size = 1858578, upload-time = "2025-04-23T18:33:34.912Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/60/bc06fa9027c7006cc6dd21e48dbf39076dc39d9abbaf718a1604973a9670/pydantic_core-2.33.2-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d3f26877a748dc4251cfcfda9dfb5f13fcb034f5308388066bcfe9031b63ae7d", size = 1892858, upload-time = "2025-04-23T18:33:36.933Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/40/9d03997d9518816c68b4dfccb88969756b9146031b61cd37f781c74c9b6a/pydantic_core-2.33.2-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dac89aea9af8cd672fa7b510e7b8c33b0bba9a43186680550ccf23020f32d535", size = 2068498, upload-time = "2025-04-23T18:33:38.997Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/62/d490198d05d2d86672dc269f52579cad7261ced64c2df213d5c16e0aecb1/pydantic_core-2.33.2-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:970919794d126ba8645f3837ab6046fb4e72bbc057b3709144066204c19a455d", size = 2108428, upload-time = "2025-04-23T18:33:41.18Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/ec/4cd215534fd10b8549015f12ea650a1a973da20ce46430b68fc3185573e8/pydantic_core-2.33.2-pp39-pypy39_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:3eb3fe62804e8f859c49ed20a8451342de53ed764150cb14ca71357c765dc2a6", size = 2069854, upload-time = "2025-04-23T18:33:43.446Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/1a/abbd63d47e1d9b0d632fee6bb15785d0889c8a6e0a6c3b5a8e28ac1ec5d2/pydantic_core-2.33.2-pp39-pypy39_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:3abcd9392a36025e3bd55f9bd38d908bd17962cc49bc6da8e7e96285336e2bca", size = 2237859, upload-time = "2025-04-23T18:33:45.56Z" },
+    { url = "https://files.pythonhosted.org/packages/80/1c/fa883643429908b1c90598fd2642af8839efd1d835b65af1f75fba4d94fe/pydantic_core-2.33.2-pp39-pypy39_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:3a1c81334778f9e3af2f8aeb7a960736e5cab1dfebfb26aabca09afd2906c039", size = 2239059, upload-time = "2025-04-23T18:33:47.735Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/29/3cade8a924a61f60ccfa10842f75eb12787e1440e2b8660ceffeb26685e7/pydantic_core-2.33.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:2807668ba86cb38c6817ad9bc66215ab8584d1d304030ce4f0887336f28a5e27", size = 2066661, upload-time = "2025-04-23T18:33:49.995Z" },
+]
+
+[[package]]
+name = "pydantic-settings"
+version = "2.10.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+    { name = "python-dotenv" },
+    { name = "typing-inspection" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/68/85/1ea668bbab3c50071ca613c6ab30047fb36ab0da1b92fa8f17bbc38fd36c/pydantic_settings-2.10.1.tar.gz", hash = "sha256:06f0062169818d0f5524420a360d632d5857b83cffd4d42fe29597807a1614ee", size = 172583, upload-time = "2025-06-24T13:26:46.841Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/58/f0/427018098906416f580e3cf1366d3b1abfb408a0652e9f31600c24a1903c/pydantic_settings-2.10.1-py3-none-any.whl", hash = "sha256:a60952460b99cf661dc25c29c0ef171721f98bfcb52ef8d9ea4c943d7c8cc796", size = 45235, upload-time = "2025-06-24T13:26:45.485Z" },
 ]
 
 [[package]]
 name = "pygments"
-version = "2.19.1"
+version = "2.19.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/7c/2d/c3338d48ea6cc0feb8446d8e6937e1408088a72a39937982cc6111d17f84/pygments-2.19.1.tar.gz", hash = "sha256:61c16d2a8576dc0649d9f39e089b5f02bcd27fba10d8fb4dcc28173f7a45151f", size = 4968581 }
+sdist = { url = "https://files.pythonhosted.org/packages/b0/77/a5b8c569bf593b0140bde72ea885a803b82086995367bf2037de0159d924/pygments-2.19.2.tar.gz", hash = "sha256:636cb2477cec7f8952536970bc533bc43743542f70392ae026374600add5b887", size = 4968631, upload-time = "2025-06-21T13:39:12.283Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/8a/0b/9fcc47d19c48b59121088dd6da2488a49d5f72dacf8262e2790a1d2c7d15/pygments-2.19.1-py3-none-any.whl", hash = "sha256:9ea1544ad55cecf4b8242fab6dd35a93bbce657034b0611ee383099054ab6d8c", size = 1225293 },
+    { url = "https://files.pythonhosted.org/packages/c7/21/705964c7812476f378728bdf590ca4b771ec72385c533964653c68e86bdc/pygments-2.19.2-py3-none-any.whl", hash = "sha256:86540386c03d588bb81d44bc3928634ff26449851e99741617ecb9037ee5ec0b", size = 1225217, upload-time = "2025-06-21T13:39:07.939Z" },
 ]
 
 [[package]]
 name = "pytest"
-version = "8.3.4"
+version = "8.4.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "colorama", marker = "sys_platform == 'win32'" },
@@ -2560,11 +3351,26 @@ dependencies = [
     { name = "iniconfig" },
     { name = "packaging" },
     { name = "pluggy" },
+    { name = "pygments" },
     { name = "tomli", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/05/35/30e0d83068951d90a01852cb1cef56e5d8a09d20c7f511634cc2f7e0372a/pytest-8.3.4.tar.gz", hash = "sha256:965370d062bce11e73868e0335abac31b4d3de0e82f4007408d242b4f8610761", size = 1445919 }
+sdist = { url = "https://files.pythonhosted.org/packages/08/ba/45911d754e8eba3d5a841a5ce61a65a685ff1798421ac054f85aa8747dfb/pytest-8.4.1.tar.gz", hash = "sha256:7c67fd69174877359ed9371ec3af8a3d2b04741818c51e5e99cc1742251fa93c", size = 1517714, upload-time = "2025-06-18T05:48:06.109Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/29/16/c8a903f4c4dffe7a12843191437d7cd8e32751d5de349d45d3fe69544e87/pytest-8.4.1-py3-none-any.whl", hash = "sha256:539c70ba6fcead8e78eebbf1115e8b589e7565830d7d006a8723f19ac8a0afb7", size = 365474, upload-time = "2025-06-18T05:48:03.955Z" },
+]
+
+[[package]]
+name = "pytest-asyncio"
+version = "1.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "backports-asyncio-runner", marker = "python_full_version < '3.11'" },
+    { name = "pytest" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4e/51/f8794af39eeb870e87a8c8068642fc07bce0c854d6865d7dd0f2a9d338c2/pytest_asyncio-1.1.0.tar.gz", hash = "sha256:796aa822981e01b68c12e4827b8697108f7205020f24b5793b3c41555dab68ea", size = 46652, upload-time = "2025-07-16T04:29:26.393Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/11/92/76a1c94d3afee238333bc0a42b82935dd8f9cf8ce9e336ff87ee14d9e1cf/pytest-8.3.4-py3-none-any.whl", hash = "sha256:50e16d954148559c9a74109af1eaf0c945ba2d8f30f0a3d3335edde19788b6f6", size = 343083 },
+    { url = "https://files.pythonhosted.org/packages/c7/9d/bf86eddabf8c6c9cb1ea9a869d6873b46f105a5d292d3a6f7071f5b07935/pytest_asyncio-1.1.0-py3-none-any.whl", hash = "sha256:5fe2d69607b0bd75c656d1211f969cadba035030156745ee09e7d71740e58ecf", size = 15157, upload-time = "2025-07-16T04:29:24.929Z" },
 ]
 
 [[package]]
@@ -2575,34 +3381,35 @@ dependencies = [
     { name = "py-cpuinfo" },
     { name = "pytest" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/39/d0/a8bd08d641b393db3be3819b03e2d9bb8760ca8479080a26a5f6e540e99c/pytest-benchmark-5.1.0.tar.gz", hash = "sha256:9ea661cdc292e8231f7cd4c10b0319e56a2118e2c09d9f50e1b3d150d2aca105", size = 337810 }
+sdist = { url = "https://files.pythonhosted.org/packages/39/d0/a8bd08d641b393db3be3819b03e2d9bb8760ca8479080a26a5f6e540e99c/pytest-benchmark-5.1.0.tar.gz", hash = "sha256:9ea661cdc292e8231f7cd4c10b0319e56a2118e2c09d9f50e1b3d150d2aca105", size = 337810, upload-time = "2024-10-30T11:51:48.521Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/9e/d6/b41653199ea09d5969d4e385df9bbfd9a100f28ca7e824ce7c0a016e3053/pytest_benchmark-5.1.0-py3-none-any.whl", hash = "sha256:922de2dfa3033c227c96da942d1878191afa135a29485fb942e85dff1c592c89", size = 44259 },
+    { url = "https://files.pythonhosted.org/packages/9e/d6/b41653199ea09d5969d4e385df9bbfd9a100f28ca7e824ce7c0a016e3053/pytest_benchmark-5.1.0-py3-none-any.whl", hash = "sha256:922de2dfa3033c227c96da942d1878191afa135a29485fb942e85dff1c592c89", size = 44259, upload-time = "2024-10-30T11:51:45.94Z" },
 ]
 
 [[package]]
 name = "pytest-cov"
-version = "6.0.0"
+version = "6.2.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "coverage", extra = ["toml"] },
+    { name = "pluggy" },
     { name = "pytest" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/be/45/9b538de8cef30e17c7b45ef42f538a94889ed6a16f2387a6c89e73220651/pytest-cov-6.0.0.tar.gz", hash = "sha256:fde0b595ca248bb8e2d76f020b465f3b107c9632e6a1d1705f17834c89dcadc0", size = 66945 }
+sdist = { url = "https://files.pythonhosted.org/packages/18/99/668cade231f434aaa59bbfbf49469068d2ddd945000621d3d165d2e7dd7b/pytest_cov-6.2.1.tar.gz", hash = "sha256:25cc6cc0a5358204b8108ecedc51a9b57b34cc6b8c967cc2c01a4e00d8a67da2", size = 69432, upload-time = "2025-06-12T10:47:47.684Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/36/3b/48e79f2cd6a61dbbd4807b4ed46cb564b4fd50a76166b1c4ea5c1d9e2371/pytest_cov-6.0.0-py3-none-any.whl", hash = "sha256:eee6f1b9e61008bd34975a4d5bab25801eb31898b032dd55addc93e96fcaaa35", size = 22949 },
+    { url = "https://files.pythonhosted.org/packages/bc/16/4ea354101abb1287856baa4af2732be351c7bee728065aed451b678153fd/pytest_cov-6.2.1-py3-none-any.whl", hash = "sha256:f5bc4c23f42f1cdd23c70b1dab1bbaef4fc505ba950d53e0081d0730dd7e86d5", size = 24644, upload-time = "2025-06-12T10:47:45.932Z" },
 ]
 
 [[package]]
 name = "pytest-mock"
-version = "3.14.0"
+version = "3.14.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "pytest" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/c6/90/a955c3ab35ccd41ad4de556596fa86685bf4fc5ffcc62d22d856cfd4e29a/pytest-mock-3.14.0.tar.gz", hash = "sha256:2719255a1efeceadbc056d6bf3df3d1c5015530fb40cf347c0f9afac88410bd0", size = 32814 }
+sdist = { url = "https://files.pythonhosted.org/packages/71/28/67172c96ba684058a4d24ffe144d64783d2a270d0af0d9e792737bddc75c/pytest_mock-3.14.1.tar.gz", hash = "sha256:159e9edac4c451ce77a5cdb9fc5d1100708d2dd4ba3c3df572f14097351af80e", size = 33241, upload-time = "2025-05-26T13:58:45.167Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f2/3b/b26f90f74e2986a82df6e7ac7e319b8ea7ccece1caec9f8ab6104dc70603/pytest_mock-3.14.0-py3-none-any.whl", hash = "sha256:0b72c38033392a5f4621342fe11e9219ac11ec9d375f8e2a0c164539e0d70f6f", size = 9863 },
+    { url = "https://files.pythonhosted.org/packages/b2/05/77b60e520511c53d1c1ca75f1930c7dd8e971d0c4379b7f4b3f9644685ba/pytest_mock-3.14.1-py3-none-any.whl", hash = "sha256:178aefcd11307d874b4cd3100344e7e2d888d9791a6a1d9bfe90fbc1b74fd1d0", size = 9923, upload-time = "2025-05-26T13:58:43.487Z" },
 ]
 
 [[package]]
@@ -2612,91 +3419,93 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "six" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/66/c0/0c8b6ad9f17a802ee498c46e004a0eb49bc148f2fd230864601a86dcf6db/python-dateutil-2.9.0.post0.tar.gz", hash = "sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3", size = 342432 }
+sdist = { url = "https://files.pythonhosted.org/packages/66/c0/0c8b6ad9f17a802ee498c46e004a0eb49bc148f2fd230864601a86dcf6db/python-dateutil-2.9.0.post0.tar.gz", hash = "sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3", size = 342432, upload-time = "2024-03-01T18:36:20.211Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892 },
+    { url = "https://files.pythonhosted.org/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427", size = 229892, upload-time = "2024-03-01T18:36:18.57Z" },
 ]
 
 [[package]]
 name = "python-dotenv"
-version = "1.0.1"
+version = "1.1.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/bc/57/e84d88dfe0aec03b7a2d4327012c1627ab5f03652216c63d49846d7a6c58/python-dotenv-1.0.1.tar.gz", hash = "sha256:e324ee90a023d808f1959c46bcbc04446a10ced277783dc6ee09987c37ec10ca", size = 39115 }
+sdist = { url = "https://files.pythonhosted.org/packages/f6/b0/4bc07ccd3572a2f9df7e6782f52b0c6c90dcbb803ac4a167702d7d0dfe1e/python_dotenv-1.1.1.tar.gz", hash = "sha256:a8a6399716257f45be6a007360200409fce5cda2661e3dec71d23dc15f6189ab", size = 41978, upload-time = "2025-06-24T04:21:07.341Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6a/3e/b68c118422ec867fa7ab88444e1274aa40681c606d59ac27de5a5588f082/python_dotenv-1.0.1-py3-none-any.whl", hash = "sha256:f7b63ef50f1b690dddf550d03497b66d609393b40b564ed0d674909a68ebf16a", size = 19863 },
+    { url = "https://files.pythonhosted.org/packages/5f/ed/539768cf28c661b5b068d66d96a2f155c4971a5d55684a514c1a0e0dec2f/python_dotenv-1.1.1-py3-none-any.whl", hash = "sha256:31f23644fe2602f88ff55e1f5c79ba497e01224ee7737937930c448e4d0e24dc", size = 20556, upload-time = "2025-06-24T04:21:06.073Z" },
 ]
 
 [[package]]
 name = "pytz"
-version = "2024.2"
+version = "2025.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/3a/31/3c70bf7603cc2dca0f19bdc53b4537a797747a58875b552c8c413d963a3f/pytz-2024.2.tar.gz", hash = "sha256:2aa355083c50a0f93fa581709deac0c9ad65cca8a9e9beac660adcbd493c798a", size = 319692 }
+sdist = { url = "https://files.pythonhosted.org/packages/f8/bf/abbd3cdfb8fbc7fb3d4d38d320f2441b1e7cbe29be4f23797b4a2b5d8aac/pytz-2025.2.tar.gz", hash = "sha256:360b9e3dbb49a209c21ad61809c7fb453643e048b38924c765813546746e81c3", size = 320884, upload-time = "2025-03-25T02:25:00.538Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/11/c3/005fcca25ce078d2cc29fd559379817424e94885510568bc1bc53d7d5846/pytz-2024.2-py2.py3-none-any.whl", hash = "sha256:31c7c1817eb7fae7ca4b8c7ee50c72f93aa2dd863de768e1ef4245d426aa0725", size = 508002 },
+    { url = "https://files.pythonhosted.org/packages/81/c4/34e93fe5f5429d7570ec1fa436f1986fb1f00c3e0f43a589fe2bbcd22c3f/pytz-2025.2-py2.py3-none-any.whl", hash = "sha256:5ddf76296dd8c44c26eb8f4b6f35488f3ccbf6fbbd7adee0b7262d43f0ec2f00", size = 509225, upload-time = "2025-03-25T02:24:58.468Z" },
 ]
 
 [[package]]
 name = "pyyaml"
 version = "6.0.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/54/ed/79a089b6be93607fa5cdaedf301d7dfb23af5f25c398d5ead2525b063e17/pyyaml-6.0.2.tar.gz", hash = "sha256:d584d9ec91ad65861cc08d42e834324ef890a082e591037abe114850ff7bbc3e", size = 130631 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/9b/95/a3fac87cb7158e231b5a6012e438c647e1a87f09f8e0d123acec8ab8bf71/PyYAML-6.0.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0a9a2848a5b7feac301353437eb7d5957887edbf81d56e903999a75a3d743086", size = 184199 },
-    { url = "https://files.pythonhosted.org/packages/c7/7a/68bd47624dab8fd4afbfd3c48e3b79efe09098ae941de5b58abcbadff5cb/PyYAML-6.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:29717114e51c84ddfba879543fb232a6ed60086602313ca38cce623c1d62cfbf", size = 171758 },
-    { url = "https://files.pythonhosted.org/packages/49/ee/14c54df452143b9ee9f0f29074d7ca5516a36edb0b4cc40c3f280131656f/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8824b5a04a04a047e72eea5cec3bc266db09e35de6bdfe34c9436ac5ee27d237", size = 718463 },
-    { url = "https://files.pythonhosted.org/packages/4d/61/de363a97476e766574650d742205be468921a7b532aa2499fcd886b62530/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7c36280e6fb8385e520936c3cb3b8042851904eba0e58d277dca80a5cfed590b", size = 719280 },
-    { url = "https://files.pythonhosted.org/packages/6b/4e/1523cb902fd98355e2e9ea5e5eb237cbc5f3ad5f3075fa65087aa0ecb669/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ec031d5d2feb36d1d1a24380e4db6d43695f3748343d99434e6f5f9156aaa2ed", size = 751239 },
-    { url = "https://files.pythonhosted.org/packages/b7/33/5504b3a9a4464893c32f118a9cc045190a91637b119a9c881da1cf6b7a72/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:936d68689298c36b53b29f23c6dbb74de12b4ac12ca6cfe0e047bedceea56180", size = 695802 },
-    { url = "https://files.pythonhosted.org/packages/5c/20/8347dcabd41ef3a3cdc4f7b7a2aff3d06598c8779faa189cdbf878b626a4/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:23502f431948090f597378482b4812b0caae32c22213aecf3b55325e049a6c68", size = 720527 },
-    { url = "https://files.pythonhosted.org/packages/be/aa/5afe99233fb360d0ff37377145a949ae258aaab831bde4792b32650a4378/PyYAML-6.0.2-cp310-cp310-win32.whl", hash = "sha256:2e99c6826ffa974fe6e27cdb5ed0021786b03fc98e5ee3c5bfe1fd5015f42b99", size = 144052 },
-    { url = "https://files.pythonhosted.org/packages/b5/84/0fa4b06f6d6c958d207620fc60005e241ecedceee58931bb20138e1e5776/PyYAML-6.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:a4d3091415f010369ae4ed1fc6b79def9416358877534caf6a0fdd2146c87a3e", size = 161774 },
-    { url = "https://files.pythonhosted.org/packages/f8/aa/7af4e81f7acba21a4c6be026da38fd2b872ca46226673c89a758ebdc4fd2/PyYAML-6.0.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:cc1c1159b3d456576af7a3e4d1ba7e6924cb39de8f67111c735f6fc832082774", size = 184612 },
-    { url = "https://files.pythonhosted.org/packages/8b/62/b9faa998fd185f65c1371643678e4d58254add437edb764a08c5a98fb986/PyYAML-6.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1e2120ef853f59c7419231f3bf4e7021f1b936f6ebd222406c3b60212205d2ee", size = 172040 },
-    { url = "https://files.pythonhosted.org/packages/ad/0c/c804f5f922a9a6563bab712d8dcc70251e8af811fce4524d57c2c0fd49a4/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5d225db5a45f21e78dd9358e58a98702a0302f2659a3c6cd320564b75b86f47c", size = 736829 },
-    { url = "https://files.pythonhosted.org/packages/51/16/6af8d6a6b210c8e54f1406a6b9481febf9c64a3109c541567e35a49aa2e7/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5ac9328ec4831237bec75defaf839f7d4564be1e6b25ac710bd1a96321cc8317", size = 764167 },
-    { url = "https://files.pythonhosted.org/packages/75/e4/2c27590dfc9992f73aabbeb9241ae20220bd9452df27483b6e56d3975cc5/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3ad2a3decf9aaba3d29c8f537ac4b243e36bef957511b4766cb0057d32b0be85", size = 762952 },
-    { url = "https://files.pythonhosted.org/packages/9b/97/ecc1abf4a823f5ac61941a9c00fe501b02ac3ab0e373c3857f7d4b83e2b6/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:ff3824dc5261f50c9b0dfb3be22b4567a6f938ccce4587b38952d85fd9e9afe4", size = 735301 },
-    { url = "https://files.pythonhosted.org/packages/45/73/0f49dacd6e82c9430e46f4a027baa4ca205e8b0a9dce1397f44edc23559d/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:797b4f722ffa07cc8d62053e4cff1486fa6dc094105d13fea7b1de7d8bf71c9e", size = 756638 },
-    { url = "https://files.pythonhosted.org/packages/22/5f/956f0f9fc65223a58fbc14459bf34b4cc48dec52e00535c79b8db361aabd/PyYAML-6.0.2-cp311-cp311-win32.whl", hash = "sha256:11d8f3dd2b9c1207dcaf2ee0bbbfd5991f571186ec9cc78427ba5bd32afae4b5", size = 143850 },
-    { url = "https://files.pythonhosted.org/packages/ed/23/8da0bbe2ab9dcdd11f4f4557ccaf95c10b9811b13ecced089d43ce59c3c8/PyYAML-6.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:e10ce637b18caea04431ce14fabcf5c64a1c61ec9c56b071a4b7ca131ca52d44", size = 161980 },
-    { url = "https://files.pythonhosted.org/packages/86/0c/c581167fc46d6d6d7ddcfb8c843a4de25bdd27e4466938109ca68492292c/PyYAML-6.0.2-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:c70c95198c015b85feafc136515252a261a84561b7b1d51e3384e0655ddf25ab", size = 183873 },
-    { url = "https://files.pythonhosted.org/packages/a8/0c/38374f5bb272c051e2a69281d71cba6fdb983413e6758b84482905e29a5d/PyYAML-6.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce826d6ef20b1bc864f0a68340c8b3287705cae2f8b4b1d932177dcc76721725", size = 173302 },
-    { url = "https://files.pythonhosted.org/packages/c3/93/9916574aa8c00aa06bbac729972eb1071d002b8e158bd0e83a3b9a20a1f7/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1f71ea527786de97d1a0cc0eacd1defc0985dcf6b3f17bb77dcfc8c34bec4dc5", size = 739154 },
-    { url = "https://files.pythonhosted.org/packages/95/0f/b8938f1cbd09739c6da569d172531567dbcc9789e0029aa070856f123984/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9b22676e8097e9e22e36d6b7bda33190d0d400f345f23d4065d48f4ca7ae0425", size = 766223 },
-    { url = "https://files.pythonhosted.org/packages/b9/2b/614b4752f2e127db5cc206abc23a8c19678e92b23c3db30fc86ab731d3bd/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:80bab7bfc629882493af4aa31a4cfa43a4c57c83813253626916b8c7ada83476", size = 767542 },
-    { url = "https://files.pythonhosted.org/packages/d4/00/dd137d5bcc7efea1836d6264f049359861cf548469d18da90cd8216cf05f/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:0833f8694549e586547b576dcfaba4a6b55b9e96098b36cdc7ebefe667dfed48", size = 731164 },
-    { url = "https://files.pythonhosted.org/packages/c9/1f/4f998c900485e5c0ef43838363ba4a9723ac0ad73a9dc42068b12aaba4e4/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:8b9c7197f7cb2738065c481a0461e50ad02f18c78cd75775628afb4d7137fb3b", size = 756611 },
-    { url = "https://files.pythonhosted.org/packages/df/d1/f5a275fdb252768b7a11ec63585bc38d0e87c9e05668a139fea92b80634c/PyYAML-6.0.2-cp312-cp312-win32.whl", hash = "sha256:ef6107725bd54b262d6dedcc2af448a266975032bc85ef0172c5f059da6325b4", size = 140591 },
-    { url = "https://files.pythonhosted.org/packages/0c/e8/4f648c598b17c3d06e8753d7d13d57542b30d56e6c2dedf9c331ae56312e/PyYAML-6.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:7e7401d0de89a9a855c839bc697c079a4af81cf878373abd7dc625847d25cbd8", size = 156338 },
-    { url = "https://files.pythonhosted.org/packages/ef/e3/3af305b830494fa85d95f6d95ef7fa73f2ee1cc8ef5b495c7c3269fb835f/PyYAML-6.0.2-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:efdca5630322a10774e8e98e1af481aad470dd62c3170801852d752aa7a783ba", size = 181309 },
-    { url = "https://files.pythonhosted.org/packages/45/9f/3b1c20a0b7a3200524eb0076cc027a970d320bd3a6592873c85c92a08731/PyYAML-6.0.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:50187695423ffe49e2deacb8cd10510bc361faac997de9efef88badc3bb9e2d1", size = 171679 },
-    { url = "https://files.pythonhosted.org/packages/7c/9a/337322f27005c33bcb656c655fa78325b730324c78620e8328ae28b64d0c/PyYAML-6.0.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0ffe8360bab4910ef1b9e87fb812d8bc0a308b0d0eef8c8f44e0254ab3b07133", size = 733428 },
-    { url = "https://files.pythonhosted.org/packages/a3/69/864fbe19e6c18ea3cc196cbe5d392175b4cf3d5d0ac1403ec3f2d237ebb5/PyYAML-6.0.2-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:17e311b6c678207928d649faa7cb0d7b4c26a0ba73d41e99c4fff6b6c3276484", size = 763361 },
-    { url = "https://files.pythonhosted.org/packages/04/24/b7721e4845c2f162d26f50521b825fb061bc0a5afcf9a386840f23ea19fa/PyYAML-6.0.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:70b189594dbe54f75ab3a1acec5f1e3faa7e8cf2f1e08d9b561cb41b845f69d5", size = 759523 },
-    { url = "https://files.pythonhosted.org/packages/2b/b2/e3234f59ba06559c6ff63c4e10baea10e5e7df868092bf9ab40e5b9c56b6/PyYAML-6.0.2-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:41e4e3953a79407c794916fa277a82531dd93aad34e29c2a514c2c0c5fe971cc", size = 726660 },
-    { url = "https://files.pythonhosted.org/packages/fe/0f/25911a9f080464c59fab9027482f822b86bf0608957a5fcc6eaac85aa515/PyYAML-6.0.2-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:68ccc6023a3400877818152ad9a1033e3db8625d899c72eacb5a668902e4d652", size = 751597 },
-    { url = "https://files.pythonhosted.org/packages/14/0d/e2c3b43bbce3cf6bd97c840b46088a3031085179e596d4929729d8d68270/PyYAML-6.0.2-cp313-cp313-win32.whl", hash = "sha256:bc2fa7c6b47d6bc618dd7fb02ef6fdedb1090ec036abab80d4681424b84c1183", size = 140527 },
-    { url = "https://files.pythonhosted.org/packages/fa/de/02b54f42487e3d3c6efb3f89428677074ca7bf43aae402517bc7cca949f3/PyYAML-6.0.2-cp313-cp313-win_amd64.whl", hash = "sha256:8388ee1976c416731879ac16da0aff3f63b286ffdd57cdeb95f3f2e085687563", size = 156446 },
-    { url = "https://files.pythonhosted.org/packages/65/d8/b7a1db13636d7fb7d4ff431593c510c8b8fca920ade06ca8ef20015493c5/PyYAML-6.0.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:688ba32a1cffef67fd2e9398a2efebaea461578b0923624778664cc1c914db5d", size = 184777 },
-    { url = "https://files.pythonhosted.org/packages/0a/02/6ec546cd45143fdf9840b2c6be8d875116a64076218b61d68e12548e5839/PyYAML-6.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a8786accb172bd8afb8be14490a16625cbc387036876ab6ba70912730faf8e1f", size = 172318 },
-    { url = "https://files.pythonhosted.org/packages/0e/9a/8cc68be846c972bda34f6c2a93abb644fb2476f4dcc924d52175786932c9/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d8e03406cac8513435335dbab54c0d385e4a49e4945d2909a581c83647ca0290", size = 720891 },
-    { url = "https://files.pythonhosted.org/packages/e9/6c/6e1b7f40181bc4805e2e07f4abc10a88ce4648e7e95ff1abe4ae4014a9b2/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f753120cb8181e736c57ef7636e83f31b9c0d1722c516f7e86cf15b7aa57ff12", size = 722614 },
-    { url = "https://files.pythonhosted.org/packages/3d/32/e7bd8535d22ea2874cef6a81021ba019474ace0d13a4819c2a4bce79bd6a/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3b1fdb9dc17f5a7677423d508ab4f243a726dea51fa5e70992e59a7411c89d19", size = 737360 },
-    { url = "https://files.pythonhosted.org/packages/d7/12/7322c1e30b9be969670b672573d45479edef72c9a0deac3bb2868f5d7469/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:0b69e4ce7a131fe56b7e4d770c67429700908fc0752af059838b1cfb41960e4e", size = 699006 },
-    { url = "https://files.pythonhosted.org/packages/82/72/04fcad41ca56491995076630c3ec1e834be241664c0c09a64c9a2589b507/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a9f8c2e67970f13b16084e04f134610fd1d374bf477b17ec1599185cf611d725", size = 723577 },
-    { url = "https://files.pythonhosted.org/packages/ed/5e/46168b1f2757f1fcd442bc3029cd8767d88a98c9c05770d8b420948743bb/PyYAML-6.0.2-cp39-cp39-win32.whl", hash = "sha256:6395c297d42274772abc367baaa79683958044e5d3835486c16da75d2a694631", size = 144593 },
-    { url = "https://files.pythonhosted.org/packages/19/87/5124b1c1f2412bb95c59ec481eaf936cd32f0fe2a7b16b97b81c4c017a6a/PyYAML-6.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:39693e1f8320ae4f43943590b49779ffb98acb81f788220ea932a6b6c51004d8", size = 162312 },
+sdist = { url = "https://files.pythonhosted.org/packages/54/ed/79a089b6be93607fa5cdaedf301d7dfb23af5f25c398d5ead2525b063e17/pyyaml-6.0.2.tar.gz", hash = "sha256:d584d9ec91ad65861cc08d42e834324ef890a082e591037abe114850ff7bbc3e", size = 130631, upload-time = "2024-08-06T20:33:50.674Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9b/95/a3fac87cb7158e231b5a6012e438c647e1a87f09f8e0d123acec8ab8bf71/PyYAML-6.0.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0a9a2848a5b7feac301353437eb7d5957887edbf81d56e903999a75a3d743086", size = 184199, upload-time = "2024-08-06T20:31:40.178Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/7a/68bd47624dab8fd4afbfd3c48e3b79efe09098ae941de5b58abcbadff5cb/PyYAML-6.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:29717114e51c84ddfba879543fb232a6ed60086602313ca38cce623c1d62cfbf", size = 171758, upload-time = "2024-08-06T20:31:42.173Z" },
+    { url = "https://files.pythonhosted.org/packages/49/ee/14c54df452143b9ee9f0f29074d7ca5516a36edb0b4cc40c3f280131656f/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8824b5a04a04a047e72eea5cec3bc266db09e35de6bdfe34c9436ac5ee27d237", size = 718463, upload-time = "2024-08-06T20:31:44.263Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/61/de363a97476e766574650d742205be468921a7b532aa2499fcd886b62530/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7c36280e6fb8385e520936c3cb3b8042851904eba0e58d277dca80a5cfed590b", size = 719280, upload-time = "2024-08-06T20:31:50.199Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/4e/1523cb902fd98355e2e9ea5e5eb237cbc5f3ad5f3075fa65087aa0ecb669/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ec031d5d2feb36d1d1a24380e4db6d43695f3748343d99434e6f5f9156aaa2ed", size = 751239, upload-time = "2024-08-06T20:31:52.292Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/33/5504b3a9a4464893c32f118a9cc045190a91637b119a9c881da1cf6b7a72/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:936d68689298c36b53b29f23c6dbb74de12b4ac12ca6cfe0e047bedceea56180", size = 695802, upload-time = "2024-08-06T20:31:53.836Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/20/8347dcabd41ef3a3cdc4f7b7a2aff3d06598c8779faa189cdbf878b626a4/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:23502f431948090f597378482b4812b0caae32c22213aecf3b55325e049a6c68", size = 720527, upload-time = "2024-08-06T20:31:55.565Z" },
+    { url = "https://files.pythonhosted.org/packages/be/aa/5afe99233fb360d0ff37377145a949ae258aaab831bde4792b32650a4378/PyYAML-6.0.2-cp310-cp310-win32.whl", hash = "sha256:2e99c6826ffa974fe6e27cdb5ed0021786b03fc98e5ee3c5bfe1fd5015f42b99", size = 144052, upload-time = "2024-08-06T20:31:56.914Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/84/0fa4b06f6d6c958d207620fc60005e241ecedceee58931bb20138e1e5776/PyYAML-6.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:a4d3091415f010369ae4ed1fc6b79def9416358877534caf6a0fdd2146c87a3e", size = 161774, upload-time = "2024-08-06T20:31:58.304Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/aa/7af4e81f7acba21a4c6be026da38fd2b872ca46226673c89a758ebdc4fd2/PyYAML-6.0.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:cc1c1159b3d456576af7a3e4d1ba7e6924cb39de8f67111c735f6fc832082774", size = 184612, upload-time = "2024-08-06T20:32:03.408Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/62/b9faa998fd185f65c1371643678e4d58254add437edb764a08c5a98fb986/PyYAML-6.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1e2120ef853f59c7419231f3bf4e7021f1b936f6ebd222406c3b60212205d2ee", size = 172040, upload-time = "2024-08-06T20:32:04.926Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/0c/c804f5f922a9a6563bab712d8dcc70251e8af811fce4524d57c2c0fd49a4/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5d225db5a45f21e78dd9358e58a98702a0302f2659a3c6cd320564b75b86f47c", size = 736829, upload-time = "2024-08-06T20:32:06.459Z" },
+    { url = "https://files.pythonhosted.org/packages/51/16/6af8d6a6b210c8e54f1406a6b9481febf9c64a3109c541567e35a49aa2e7/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5ac9328ec4831237bec75defaf839f7d4564be1e6b25ac710bd1a96321cc8317", size = 764167, upload-time = "2024-08-06T20:32:08.338Z" },
+    { url = "https://files.pythonhosted.org/packages/75/e4/2c27590dfc9992f73aabbeb9241ae20220bd9452df27483b6e56d3975cc5/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3ad2a3decf9aaba3d29c8f537ac4b243e36bef957511b4766cb0057d32b0be85", size = 762952, upload-time = "2024-08-06T20:32:14.124Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/97/ecc1abf4a823f5ac61941a9c00fe501b02ac3ab0e373c3857f7d4b83e2b6/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:ff3824dc5261f50c9b0dfb3be22b4567a6f938ccce4587b38952d85fd9e9afe4", size = 735301, upload-time = "2024-08-06T20:32:16.17Z" },
+    { url = "https://files.pythonhosted.org/packages/45/73/0f49dacd6e82c9430e46f4a027baa4ca205e8b0a9dce1397f44edc23559d/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:797b4f722ffa07cc8d62053e4cff1486fa6dc094105d13fea7b1de7d8bf71c9e", size = 756638, upload-time = "2024-08-06T20:32:18.555Z" },
+    { url = "https://files.pythonhosted.org/packages/22/5f/956f0f9fc65223a58fbc14459bf34b4cc48dec52e00535c79b8db361aabd/PyYAML-6.0.2-cp311-cp311-win32.whl", hash = "sha256:11d8f3dd2b9c1207dcaf2ee0bbbfd5991f571186ec9cc78427ba5bd32afae4b5", size = 143850, upload-time = "2024-08-06T20:32:19.889Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/23/8da0bbe2ab9dcdd11f4f4557ccaf95c10b9811b13ecced089d43ce59c3c8/PyYAML-6.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:e10ce637b18caea04431ce14fabcf5c64a1c61ec9c56b071a4b7ca131ca52d44", size = 161980, upload-time = "2024-08-06T20:32:21.273Z" },
+    { url = "https://files.pythonhosted.org/packages/86/0c/c581167fc46d6d6d7ddcfb8c843a4de25bdd27e4466938109ca68492292c/PyYAML-6.0.2-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:c70c95198c015b85feafc136515252a261a84561b7b1d51e3384e0655ddf25ab", size = 183873, upload-time = "2024-08-06T20:32:25.131Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/0c/38374f5bb272c051e2a69281d71cba6fdb983413e6758b84482905e29a5d/PyYAML-6.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce826d6ef20b1bc864f0a68340c8b3287705cae2f8b4b1d932177dcc76721725", size = 173302, upload-time = "2024-08-06T20:32:26.511Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/93/9916574aa8c00aa06bbac729972eb1071d002b8e158bd0e83a3b9a20a1f7/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1f71ea527786de97d1a0cc0eacd1defc0985dcf6b3f17bb77dcfc8c34bec4dc5", size = 739154, upload-time = "2024-08-06T20:32:28.363Z" },
+    { url = "https://files.pythonhosted.org/packages/95/0f/b8938f1cbd09739c6da569d172531567dbcc9789e0029aa070856f123984/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9b22676e8097e9e22e36d6b7bda33190d0d400f345f23d4065d48f4ca7ae0425", size = 766223, upload-time = "2024-08-06T20:32:30.058Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/2b/614b4752f2e127db5cc206abc23a8c19678e92b23c3db30fc86ab731d3bd/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:80bab7bfc629882493af4aa31a4cfa43a4c57c83813253626916b8c7ada83476", size = 767542, upload-time = "2024-08-06T20:32:31.881Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/00/dd137d5bcc7efea1836d6264f049359861cf548469d18da90cd8216cf05f/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:0833f8694549e586547b576dcfaba4a6b55b9e96098b36cdc7ebefe667dfed48", size = 731164, upload-time = "2024-08-06T20:32:37.083Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/1f/4f998c900485e5c0ef43838363ba4a9723ac0ad73a9dc42068b12aaba4e4/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:8b9c7197f7cb2738065c481a0461e50ad02f18c78cd75775628afb4d7137fb3b", size = 756611, upload-time = "2024-08-06T20:32:38.898Z" },
+    { url = "https://files.pythonhosted.org/packages/df/d1/f5a275fdb252768b7a11ec63585bc38d0e87c9e05668a139fea92b80634c/PyYAML-6.0.2-cp312-cp312-win32.whl", hash = "sha256:ef6107725bd54b262d6dedcc2af448a266975032bc85ef0172c5f059da6325b4", size = 140591, upload-time = "2024-08-06T20:32:40.241Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/e8/4f648c598b17c3d06e8753d7d13d57542b30d56e6c2dedf9c331ae56312e/PyYAML-6.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:7e7401d0de89a9a855c839bc697c079a4af81cf878373abd7dc625847d25cbd8", size = 156338, upload-time = "2024-08-06T20:32:41.93Z" },
+    { url = "https://files.pythonhosted.org/packages/65/d8/b7a1db13636d7fb7d4ff431593c510c8b8fca920ade06ca8ef20015493c5/PyYAML-6.0.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:688ba32a1cffef67fd2e9398a2efebaea461578b0923624778664cc1c914db5d", size = 184777, upload-time = "2024-08-06T20:33:25.896Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/02/6ec546cd45143fdf9840b2c6be8d875116a64076218b61d68e12548e5839/PyYAML-6.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a8786accb172bd8afb8be14490a16625cbc387036876ab6ba70912730faf8e1f", size = 172318, upload-time = "2024-08-06T20:33:27.212Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/9a/8cc68be846c972bda34f6c2a93abb644fb2476f4dcc924d52175786932c9/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d8e03406cac8513435335dbab54c0d385e4a49e4945d2909a581c83647ca0290", size = 720891, upload-time = "2024-08-06T20:33:28.974Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/6c/6e1b7f40181bc4805e2e07f4abc10a88ce4648e7e95ff1abe4ae4014a9b2/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f753120cb8181e736c57ef7636e83f31b9c0d1722c516f7e86cf15b7aa57ff12", size = 722614, upload-time = "2024-08-06T20:33:34.157Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/32/e7bd8535d22ea2874cef6a81021ba019474ace0d13a4819c2a4bce79bd6a/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3b1fdb9dc17f5a7677423d508ab4f243a726dea51fa5e70992e59a7411c89d19", size = 737360, upload-time = "2024-08-06T20:33:35.84Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/12/7322c1e30b9be969670b672573d45479edef72c9a0deac3bb2868f5d7469/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:0b69e4ce7a131fe56b7e4d770c67429700908fc0752af059838b1cfb41960e4e", size = 699006, upload-time = "2024-08-06T20:33:37.501Z" },
+    { url = "https://files.pythonhosted.org/packages/82/72/04fcad41ca56491995076630c3ec1e834be241664c0c09a64c9a2589b507/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a9f8c2e67970f13b16084e04f134610fd1d374bf477b17ec1599185cf611d725", size = 723577, upload-time = "2024-08-06T20:33:39.389Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/5e/46168b1f2757f1fcd442bc3029cd8767d88a98c9c05770d8b420948743bb/PyYAML-6.0.2-cp39-cp39-win32.whl", hash = "sha256:6395c297d42274772abc367baaa79683958044e5d3835486c16da75d2a694631", size = 144593, upload-time = "2024-08-06T20:33:46.63Z" },
+    { url = "https://files.pythonhosted.org/packages/19/87/5124b1c1f2412bb95c59ec481eaf936cd32f0fe2a7b16b97b81c4c017a6a/PyYAML-6.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:39693e1f8320ae4f43943590b49779ffb98acb81f788220ea932a6b6c51004d8", size = 162312, upload-time = "2024-08-06T20:33:49.073Z" },
+]
+
+[[package]]
+name = "pyyaml-env-tag"
+version = "1.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pyyaml" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/eb/2e/79c822141bfd05a853236b504869ebc6b70159afc570e1d5a20641782eaa/pyyaml_env_tag-1.1.tar.gz", hash = "sha256:2eb38b75a2d21ee0475d6d97ec19c63287a7e140231e4214969d0eac923cd7ff", size = 5737, upload-time = "2025-05-13T15:24:01.64Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/04/11/432f32f8097b03e3cd5fe57e88efb685d964e2e5178a48ed61e841f7fdce/pyyaml_env_tag-1.1-py3-none-any.whl", hash = "sha256:17109e1a528561e32f026364712fee1264bc2ea6715120891174ed1b980d2e04", size = 4722, upload-time = "2025-05-13T15:23:59.629Z" },
 ]
 
 [[package]]
 name = "ray"
-version = "2.40.0"
+version = "2.49.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "aiosignal" },
-    { name = "click" },
+    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "filelock" },
-    { name = "frozenlist" },
     { name = "jsonschema" },
     { name = "msgpack" },
     { name = "packaging" },
@@ -2705,130 +3514,99 @@ dependencies = [
     { name = "requests" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ca/42/492dc35c112c5adbcf258066d739f897f484b6e2aff29b28dd9ebc9832d4/ray-2.40.0-cp310-cp310-macosx_10_15_x86_64.whl", hash = "sha256:064af8bc52cc988c82470b8e76e5df417737fa7c1d87f597a892c69eb4ec3caa", size = 67061821 },
-    { url = "https://files.pythonhosted.org/packages/07/c2/0847df9d81524ceafe0fafcddce9e4e4799501807687759d2c48f1b34f43/ray-2.40.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:45beb4019cd20b6cb10572d8012c771bccd623f544a669da6797ccf993c4bb33", size = 64422550 },
-    { url = "https://files.pythonhosted.org/packages/62/e7/cf468bfb109d904cc5b7650d890f7c4284842d72acf65b1ae05a5c02c2e5/ray-2.40.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:6cede5fbf7de4fae22cebe2c6977aaf3c85fde6f7de2aa10c46992cf24ea8bda", size = 65916634 },
-    { url = "https://files.pythonhosted.org/packages/31/73/6763d7f87756816698fa5f841ba590a869e821a46c0a98d848055ea831bb/ray-2.40.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:f6eab11dc8490f88e78e06aa645905b259cde1fa03b15e8426155c4782ba0bbe", size = 66845129 },
-    { url = "https://files.pythonhosted.org/packages/1f/cd/0107c782d8511ae100a49da95dc1951c1da67a467e204671ae730f6a5401/ray-2.40.0-cp310-cp310-win_amd64.whl", hash = "sha256:f83cda1ecceb7abe021cd377f0c503596f26d2d66cdff13c1089a06c8b780c23", size = 25295665 },
-    { url = "https://files.pythonhosted.org/packages/a9/fc/811e88c982ea755c6351bd5f2c410cd8526695069a708d9aa3c9410a8dc3/ray-2.40.0-cp311-cp311-macosx_10_15_x86_64.whl", hash = "sha256:dac89bb2cb889c19549a4ac0383492e7550f3e63b78b629a3118e8b91e4e82f3", size = 67000351 },
-    { url = "https://files.pythonhosted.org/packages/72/fb/f048a8580d97429ab3b60b13844500132e756ceaa5bb643087a0d46d9c9b/ray-2.40.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3e4efdf8aebff6e71391c2d5dd66bb45835f2d6d629ac03a3e21e2d4283e2311", size = 64359351 },
-    { url = "https://files.pythonhosted.org/packages/99/34/e16767825432cf7553badba349c80344f4a86b227e5ff2ef4ae20ea40b6c/ray-2.40.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:c776f131e5d0a169a98ab8021c5796f52bf48fcfc6c44ffbd2a9d090fe10748a", size = 66058000 },
-    { url = "https://files.pythonhosted.org/packages/55/4e/a46f514a574c33f699f729e26866475bd725287cf9ba3adf55883d6f809d/ray-2.40.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:71711cbf2c156213fd49b0f9cc93180a7ba424110070a34bdea3dc09527f31df", size = 66967160 },
-    { url = "https://files.pythonhosted.org/packages/9b/fe/01adb0316658feab19ee48badb3dd232ae681633f47b2d6c191c4df14ebe/ray-2.40.0-cp311-cp311-win_amd64.whl", hash = "sha256:532321132618983366e39aeb4cc7867cf7241b0b1e49ee44b01d2aee9923e422", size = 25236232 },
-    { url = "https://files.pythonhosted.org/packages/86/01/994daffa3516e582cd8eacbe958938d2128b79284f2387ff6ebc53b970e2/ray-2.40.0-cp312-cp312-macosx_10_15_x86_64.whl", hash = "sha256:6992922fe91a90b5cc97d9f05ca51b64d72cd644db7ad55caa936be9a6098cce", size = 66981507 },
-    { url = "https://files.pythonhosted.org/packages/46/ce/bcf8416ce4137ce4bc7e0ebdedb2200a598f7beeca1a1004bc45478bb4b0/ray-2.40.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:28329e7a7471610a475d3bb09a4c1b31abcf3596cee25c4254f8d01ad161ba84", size = 64345984 },
-    { url = "https://files.pythonhosted.org/packages/8a/96/dbcb31dd8ff74b81dbfe9332f49941c716daffaffdc3efb1c30ca5a1b4cd/ray-2.40.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:8ea05221fa48e32c652c29498d320e90134b3a012421006af98965097dd1cc3b", size = 66071071 },
-    { url = "https://files.pythonhosted.org/packages/82/ad/2eaf308c64e39704629863720c59b82e1ceed814fca319b8585deb92bb74/ray-2.40.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:674755814f5692306c554cadbc24015af823dc0516e34bdef24ccac9d7a656e3", size = 67018887 },
-    { url = "https://files.pythonhosted.org/packages/21/c4/23616495341e01c97d5b72d07555ab8908e87190cf2f53b4f029cb65bcf0/ray-2.40.0-cp312-cp312-win_amd64.whl", hash = "sha256:bbc01d773cbc43e3efa462ec28ee4c0cacc50f098078332fb45b1ab38eaf9b5d", size = 25223820 },
-    { url = "https://files.pythonhosted.org/packages/44/6f/ecaf4363063c1a1f7e5d611083e459919c33815b32008c25eeb853b8da1f/ray-2.40.0-cp39-cp39-macosx_10_15_x86_64.whl", hash = "sha256:27292bf8921dd69757e7581644afcd3ccae13d6f10f3841f5523ae82b6612f4b", size = 67076882 },
-    { url = "https://files.pythonhosted.org/packages/5c/7f/d6da83e7e2357770f482f9f521d15b46ea13116aeb0ba0569e3986d11658/ray-2.40.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2b74ca43d0c4ccdcaefbf1e7d26aabb1c0d20f825688a9fd7134ba918bda8442", size = 64436021 },
-    { url = "https://files.pythonhosted.org/packages/87/79/95634cfe3fd0457a662e24967391597a5074b5aecfab8c453221481c5dfa/ray-2.40.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:5eb7a203f58defedff0dc53f78a4e1431d040b2b8458548704979c0113f3b892", size = 65938206 },
-    { url = "https://files.pythonhosted.org/packages/1f/c9/09de884f8e62efe0251f8915380f322d92e49f7f7e471c2f15471c40b831/ray-2.40.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:a36a20a3b936b36d14fab031222f92e3c5e731d7db6bb183ca4fba6d0ce3f52a", size = 66864869 },
-    { url = "https://files.pythonhosted.org/packages/ae/57/6742621756a7bce8ac99f9d30e2e7461e1d58716d5fd274a0a88657abd60/ray-2.40.0-cp39-cp39-win_amd64.whl", hash = "sha256:fbe9cd3e076dea676afd57caf19b2897a67ecdf14a542c03864800966cf2aec9", size = 25303127 },
+    { url = "https://files.pythonhosted.org/packages/ca/2d/67ac2d5700dbe51c5a3162ec8937299ba100ea016b540997d52776de08e6/ray-2.49.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:8c55a7a462a0cdeb1ed58fa3610f52e3a284ce4e0c55566359bd1158a9683f44", size = 69124127, upload-time = "2025-08-26T19:07:26.774Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/56/b954e0a81a94fad760e57185bc18f0955a3574ad4bf63ea2756e8d588462/ray-2.49.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:6c540a6dfa8e582e0725057c137245ccd3fa0d697b47648533176f5a8ab30a30", size = 69933848, upload-time = "2025-08-26T19:07:32.234Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/60/36b5773a14dda16c2ff997cbc6b3ee2700dca42550c026c7108c7bc7a2b4/ray-2.49.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:2a75d2fd8bcab1a85091ea30e5932c5e2897c7122ae23f921f4fafeec22c3828", size = 69264900, upload-time = "2025-08-26T19:07:52.476Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/47/5bc38d9d86b6fcea65dda30ea6d128c7e18bd45687fca9c5038ac8276d1a/ray-2.49.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:25ae4c208dc78b4a653c311cb26b93c74401e6762a193ced5fabaa910a817d81", size = 70068495, upload-time = "2025-08-26T19:07:57.491Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/81/d14e0d58358ca145f173b7b0604434f368cd90243172955d6cbf4104ce59/ray-2.49.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:363567f56112f3a1518ecd8d0e3a0ce4f053705cc8f4d43e933caf69fb35971c", size = 69285849, upload-time = "2025-08-26T19:08:23.849Z" },
+    { url = "https://files.pythonhosted.org/packages/de/8b/1de4d89347857d78f3ffd10f1b1eef767bfc6ec22bf7721ff61719a6b56e/ray-2.49.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:95285d23a303b174d070a6b3ea8723e9dabf07a4147a01cb9331d5eadf8ca0b2", size = 70113049, upload-time = "2025-08-26T19:08:29.453Z" },
+    { url = "https://files.pythonhosted.org/packages/07/89/857a879351dcc8b299084eba05e5d2d0e47c973b399ef769c80f45c62503/ray-2.49.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:d3a3f04d8975db0c1e9e59ff4169cae4e6f1026ad27dafb9fb9baf0121c32427", size = 69124048, upload-time = "2025-08-26T19:09:19.197Z" },
+    { url = "https://files.pythonhosted.org/packages/de/cc/9ace3df475404f45199c7bf6587d4760c0add58c7d32002f23d6f81da988/ray-2.49.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:0ff4a0c4fb5f11fe9dcb8266bb98e2d43b436c1e654fdfabd8a22892b4ae48e1", size = 69931493, upload-time = "2025-08-26T19:09:24.642Z" },
 ]
 
 [[package]]
 name = "referencing"
-version = "0.36.1"
+version = "0.36.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "attrs" },
     { name = "rpds-py" },
-    { name = "typing-extensions", marker = "python_full_version < '3.13'" },
+    { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/27/32/fd98246df7a0f309b58cae68b10b6b219ef2eb66747f00dfb34422687087/referencing-0.36.1.tar.gz", hash = "sha256:ca2e6492769e3602957e9b831b94211599d2aade9477f5d44110d2530cf9aade", size = 74661 }
+sdist = { url = "https://files.pythonhosted.org/packages/2f/db/98b5c277be99dd18bfd91dd04e1b759cad18d1a338188c936e92f921c7e2/referencing-0.36.2.tar.gz", hash = "sha256:df2e89862cd09deabbdba16944cc3f10feb6b3e6f18e902f7cc25609a34775aa", size = 74744, upload-time = "2025-01-25T08:48:16.138Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/cc/fa/9f193ef0c9074b659009f06d7cbacc6f25b072044815bcf799b76533dbb8/referencing-0.36.1-py3-none-any.whl", hash = "sha256:363d9c65f080d0d70bc41c721dce3c7f3e77fc09f269cd5c8813da18069a6794", size = 26777 },
+    { url = "https://files.pythonhosted.org/packages/c1/b1/3baf80dc6d2b7bc27a95a67752d0208e410351e3feb4eb78de5f77454d8d/referencing-0.36.2-py3-none-any.whl", hash = "sha256:e8699adbbf8b5c7de96d8ffa0eb5c158b3beafce084968e2ea8bb08c6794dcd0", size = 26775, upload-time = "2025-01-25T08:48:14.241Z" },
 ]
 
 [[package]]
 name = "regex"
-version = "2024.11.6"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/8e/5f/bd69653fbfb76cf8604468d3b4ec4c403197144c7bfe0e6a5fc9e02a07cb/regex-2024.11.6.tar.gz", hash = "sha256:7ab159b063c52a0333c884e4679f8d7a85112ee3078fe3d9004b2dd875585519", size = 399494 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/95/3c/4651f6b130c6842a8f3df82461a8950f923925db8b6961063e82744bddcc/regex-2024.11.6-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:ff590880083d60acc0433f9c3f713c51f7ac6ebb9adf889c79a261ecf541aa91", size = 482674 },
-    { url = "https://files.pythonhosted.org/packages/15/51/9f35d12da8434b489c7b7bffc205c474a0a9432a889457026e9bc06a297a/regex-2024.11.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:658f90550f38270639e83ce492f27d2c8d2cd63805c65a13a14d36ca126753f0", size = 287684 },
-    { url = "https://files.pythonhosted.org/packages/bd/18/b731f5510d1b8fb63c6b6d3484bfa9a59b84cc578ac8b5172970e05ae07c/regex-2024.11.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:164d8b7b3b4bcb2068b97428060b2a53be050085ef94eca7f240e7947f1b080e", size = 284589 },
-    { url = "https://files.pythonhosted.org/packages/78/a2/6dd36e16341ab95e4c6073426561b9bfdeb1a9c9b63ab1b579c2e96cb105/regex-2024.11.6-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d3660c82f209655a06b587d55e723f0b813d3a7db2e32e5e7dc64ac2a9e86fde", size = 782511 },
-    { url = "https://files.pythonhosted.org/packages/1b/2b/323e72d5d2fd8de0d9baa443e1ed70363ed7e7b2fb526f5950c5cb99c364/regex-2024.11.6-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d22326fcdef5e08c154280b71163ced384b428343ae16a5ab2b3354aed12436e", size = 821149 },
-    { url = "https://files.pythonhosted.org/packages/90/30/63373b9ea468fbef8a907fd273e5c329b8c9535fee36fc8dba5fecac475d/regex-2024.11.6-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f1ac758ef6aebfc8943560194e9fd0fa18bcb34d89fd8bd2af18183afd8da3a2", size = 809707 },
-    { url = "https://files.pythonhosted.org/packages/f2/98/26d3830875b53071f1f0ae6d547f1d98e964dd29ad35cbf94439120bb67a/regex-2024.11.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:997d6a487ff00807ba810e0f8332c18b4eb8d29463cfb7c820dc4b6e7562d0cf", size = 781702 },
-    { url = "https://files.pythonhosted.org/packages/87/55/eb2a068334274db86208ab9d5599ffa63631b9f0f67ed70ea7c82a69bbc8/regex-2024.11.6-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:02a02d2bb04fec86ad61f3ea7f49c015a0681bf76abb9857f945d26159d2968c", size = 771976 },
-    { url = "https://files.pythonhosted.org/packages/74/c0/be707bcfe98254d8f9d2cff55d216e946f4ea48ad2fd8cf1428f8c5332ba/regex-2024.11.6-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:f02f93b92358ee3f78660e43b4b0091229260c5d5c408d17d60bf26b6c900e86", size = 697397 },
-    { url = "https://files.pythonhosted.org/packages/49/dc/bb45572ceb49e0f6509f7596e4ba7031f6819ecb26bc7610979af5a77f45/regex-2024.11.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:06eb1be98df10e81ebaded73fcd51989dcf534e3c753466e4b60c4697a003b67", size = 768726 },
-    { url = "https://files.pythonhosted.org/packages/5a/db/f43fd75dc4c0c2d96d0881967897926942e935d700863666f3c844a72ce6/regex-2024.11.6-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:040df6fe1a5504eb0f04f048e6d09cd7c7110fef851d7c567a6b6e09942feb7d", size = 775098 },
-    { url = "https://files.pythonhosted.org/packages/99/d7/f94154db29ab5a89d69ff893159b19ada89e76b915c1293e98603d39838c/regex-2024.11.6-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:fdabbfc59f2c6edba2a6622c647b716e34e8e3867e0ab975412c5c2f79b82da2", size = 839325 },
-    { url = "https://files.pythonhosted.org/packages/f7/17/3cbfab1f23356fbbf07708220ab438a7efa1e0f34195bf857433f79f1788/regex-2024.11.6-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:8447d2d39b5abe381419319f942de20b7ecd60ce86f16a23b0698f22e1b70008", size = 843277 },
-    { url = "https://files.pythonhosted.org/packages/7e/f2/48b393b51900456155de3ad001900f94298965e1cad1c772b87f9cfea011/regex-2024.11.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:da8f5fc57d1933de22a9e23eec290a0d8a5927a5370d24bda9a6abe50683fe62", size = 773197 },
-    { url = "https://files.pythonhosted.org/packages/45/3f/ef9589aba93e084cd3f8471fded352826dcae8489b650d0b9b27bc5bba8a/regex-2024.11.6-cp310-cp310-win32.whl", hash = "sha256:b489578720afb782f6ccf2840920f3a32e31ba28a4b162e13900c3e6bd3f930e", size = 261714 },
-    { url = "https://files.pythonhosted.org/packages/42/7e/5f1b92c8468290c465fd50c5318da64319133231415a8aa6ea5ab995a815/regex-2024.11.6-cp310-cp310-win_amd64.whl", hash = "sha256:5071b2093e793357c9d8b2929dfc13ac5f0a6c650559503bb81189d0a3814519", size = 274042 },
-    { url = "https://files.pythonhosted.org/packages/58/58/7e4d9493a66c88a7da6d205768119f51af0f684fe7be7bac8328e217a52c/regex-2024.11.6-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:5478c6962ad548b54a591778e93cd7c456a7a29f8eca9c49e4f9a806dcc5d638", size = 482669 },
-    { url = "https://files.pythonhosted.org/packages/34/4c/8f8e631fcdc2ff978609eaeef1d6994bf2f028b59d9ac67640ed051f1218/regex-2024.11.6-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:2c89a8cc122b25ce6945f0423dc1352cb9593c68abd19223eebbd4e56612c5b7", size = 287684 },
-    { url = "https://files.pythonhosted.org/packages/c5/1b/f0e4d13e6adf866ce9b069e191f303a30ab1277e037037a365c3aad5cc9c/regex-2024.11.6-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:94d87b689cdd831934fa3ce16cc15cd65748e6d689f5d2b8f4f4df2065c9fa20", size = 284589 },
-    { url = "https://files.pythonhosted.org/packages/25/4d/ab21047f446693887f25510887e6820b93f791992994f6498b0318904d4a/regex-2024.11.6-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1062b39a0a2b75a9c694f7a08e7183a80c63c0d62b301418ffd9c35f55aaa114", size = 792121 },
-    { url = "https://files.pythonhosted.org/packages/45/ee/c867e15cd894985cb32b731d89576c41a4642a57850c162490ea34b78c3b/regex-2024.11.6-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:167ed4852351d8a750da48712c3930b031f6efdaa0f22fa1933716bfcd6bf4a3", size = 831275 },
-    { url = "https://files.pythonhosted.org/packages/b3/12/b0f480726cf1c60f6536fa5e1c95275a77624f3ac8fdccf79e6727499e28/regex-2024.11.6-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:2d548dafee61f06ebdb584080621f3e0c23fff312f0de1afc776e2a2ba99a74f", size = 818257 },
-    { url = "https://files.pythonhosted.org/packages/bf/ce/0d0e61429f603bac433910d99ef1a02ce45a8967ffbe3cbee48599e62d88/regex-2024.11.6-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f2a19f302cd1ce5dd01a9099aaa19cae6173306d1302a43b627f62e21cf18ac0", size = 792727 },
-    { url = "https://files.pythonhosted.org/packages/e4/c1/243c83c53d4a419c1556f43777ccb552bccdf79d08fda3980e4e77dd9137/regex-2024.11.6-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bec9931dfb61ddd8ef2ebc05646293812cb6b16b60cf7c9511a832b6f1854b55", size = 780667 },
-    { url = "https://files.pythonhosted.org/packages/c5/f4/75eb0dd4ce4b37f04928987f1d22547ddaf6c4bae697623c1b05da67a8aa/regex-2024.11.6-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:9714398225f299aa85267fd222f7142fcb5c769e73d7733344efc46f2ef5cf89", size = 776963 },
-    { url = "https://files.pythonhosted.org/packages/16/5d/95c568574e630e141a69ff8a254c2f188b4398e813c40d49228c9bbd9875/regex-2024.11.6-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:202eb32e89f60fc147a41e55cb086db2a3f8cb82f9a9a88440dcfc5d37faae8d", size = 784700 },
-    { url = "https://files.pythonhosted.org/packages/8e/b5/f8495c7917f15cc6fee1e7f395e324ec3e00ab3c665a7dc9d27562fd5290/regex-2024.11.6-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:4181b814e56078e9b00427ca358ec44333765f5ca1b45597ec7446d3a1ef6e34", size = 848592 },
-    { url = "https://files.pythonhosted.org/packages/1c/80/6dd7118e8cb212c3c60b191b932dc57db93fb2e36fb9e0e92f72a5909af9/regex-2024.11.6-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:068376da5a7e4da51968ce4c122a7cd31afaaec4fccc7856c92f63876e57b51d", size = 852929 },
-    { url = "https://files.pythonhosted.org/packages/11/9b/5a05d2040297d2d254baf95eeeb6df83554e5e1df03bc1a6687fc4ba1f66/regex-2024.11.6-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ac10f2c4184420d881a3475fb2c6f4d95d53a8d50209a2500723d831036f7c45", size = 781213 },
-    { url = "https://files.pythonhosted.org/packages/26/b7/b14e2440156ab39e0177506c08c18accaf2b8932e39fb092074de733d868/regex-2024.11.6-cp311-cp311-win32.whl", hash = "sha256:c36f9b6f5f8649bb251a5f3f66564438977b7ef8386a52460ae77e6070d309d9", size = 261734 },
-    { url = "https://files.pythonhosted.org/packages/80/32/763a6cc01d21fb3819227a1cc3f60fd251c13c37c27a73b8ff4315433a8e/regex-2024.11.6-cp311-cp311-win_amd64.whl", hash = "sha256:02e28184be537f0e75c1f9b2f8847dc51e08e6e171c6bde130b2687e0c33cf60", size = 274052 },
-    { url = "https://files.pythonhosted.org/packages/ba/30/9a87ce8336b172cc232a0db89a3af97929d06c11ceaa19d97d84fa90a8f8/regex-2024.11.6-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:52fb28f528778f184f870b7cf8f225f5eef0a8f6e3778529bdd40c7b3920796a", size = 483781 },
-    { url = "https://files.pythonhosted.org/packages/01/e8/00008ad4ff4be8b1844786ba6636035f7ef926db5686e4c0f98093612add/regex-2024.11.6-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:fdd6028445d2460f33136c55eeb1f601ab06d74cb3347132e1c24250187500d9", size = 288455 },
-    { url = "https://files.pythonhosted.org/packages/60/85/cebcc0aff603ea0a201667b203f13ba75d9fc8668fab917ac5b2de3967bc/regex-2024.11.6-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:805e6b60c54bf766b251e94526ebad60b7de0c70f70a4e6210ee2891acb70bf2", size = 284759 },
-    { url = "https://files.pythonhosted.org/packages/94/2b/701a4b0585cb05472a4da28ee28fdfe155f3638f5e1ec92306d924e5faf0/regex-2024.11.6-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b85c2530be953a890eaffde05485238f07029600e8f098cdf1848d414a8b45e4", size = 794976 },
-    { url = "https://files.pythonhosted.org/packages/4b/bf/fa87e563bf5fee75db8915f7352e1887b1249126a1be4813837f5dbec965/regex-2024.11.6-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bb26437975da7dc36b7efad18aa9dd4ea569d2357ae6b783bf1118dabd9ea577", size = 833077 },
-    { url = "https://files.pythonhosted.org/packages/a1/56/7295e6bad94b047f4d0834e4779491b81216583c00c288252ef625c01d23/regex-2024.11.6-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:abfa5080c374a76a251ba60683242bc17eeb2c9818d0d30117b4486be10c59d3", size = 823160 },
-    { url = "https://files.pythonhosted.org/packages/fb/13/e3b075031a738c9598c51cfbc4c7879e26729c53aa9cca59211c44235314/regex-2024.11.6-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:70b7fa6606c2881c1db9479b0eaa11ed5dfa11c8d60a474ff0e095099f39d98e", size = 796896 },
-    { url = "https://files.pythonhosted.org/packages/24/56/0b3f1b66d592be6efec23a795b37732682520b47c53da5a32c33ed7d84e3/regex-2024.11.6-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0c32f75920cf99fe6b6c539c399a4a128452eaf1af27f39bce8909c9a3fd8cbe", size = 783997 },
-    { url = "https://files.pythonhosted.org/packages/f9/a1/eb378dada8b91c0e4c5f08ffb56f25fcae47bf52ad18f9b2f33b83e6d498/regex-2024.11.6-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:982e6d21414e78e1f51cf595d7f321dcd14de1f2881c5dc6a6e23bbbbd68435e", size = 781725 },
-    { url = "https://files.pythonhosted.org/packages/83/f2/033e7dec0cfd6dda93390089864732a3409246ffe8b042e9554afa9bff4e/regex-2024.11.6-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:a7c2155f790e2fb448faed6dd241386719802296ec588a8b9051c1f5c481bc29", size = 789481 },
-    { url = "https://files.pythonhosted.org/packages/83/23/15d4552ea28990a74e7696780c438aadd73a20318c47e527b47a4a5a596d/regex-2024.11.6-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:149f5008d286636e48cd0b1dd65018548944e495b0265b45e1bffecce1ef7f39", size = 852896 },
-    { url = "https://files.pythonhosted.org/packages/e3/39/ed4416bc90deedbfdada2568b2cb0bc1fdb98efe11f5378d9892b2a88f8f/regex-2024.11.6-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:e5364a4502efca094731680e80009632ad6624084aff9a23ce8c8c6820de3e51", size = 860138 },
-    { url = "https://files.pythonhosted.org/packages/93/2d/dd56bb76bd8e95bbce684326302f287455b56242a4f9c61f1bc76e28360e/regex-2024.11.6-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:0a86e7eeca091c09e021db8eb72d54751e527fa47b8d5787caf96d9831bd02ad", size = 787692 },
-    { url = "https://files.pythonhosted.org/packages/0b/55/31877a249ab7a5156758246b9c59539abbeba22461b7d8adc9e8475ff73e/regex-2024.11.6-cp312-cp312-win32.whl", hash = "sha256:32f9a4c643baad4efa81d549c2aadefaeba12249b2adc5af541759237eee1c54", size = 262135 },
-    { url = "https://files.pythonhosted.org/packages/38/ec/ad2d7de49a600cdb8dd78434a1aeffe28b9d6fc42eb36afab4a27ad23384/regex-2024.11.6-cp312-cp312-win_amd64.whl", hash = "sha256:a93c194e2df18f7d264092dc8539b8ffb86b45b899ab976aa15d48214138e81b", size = 273567 },
-    { url = "https://files.pythonhosted.org/packages/90/73/bcb0e36614601016552fa9344544a3a2ae1809dc1401b100eab02e772e1f/regex-2024.11.6-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a6ba92c0bcdf96cbf43a12c717eae4bc98325ca3730f6b130ffa2e3c3c723d84", size = 483525 },
-    { url = "https://files.pythonhosted.org/packages/0f/3f/f1a082a46b31e25291d830b369b6b0c5576a6f7fb89d3053a354c24b8a83/regex-2024.11.6-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:525eab0b789891ac3be914d36893bdf972d483fe66551f79d3e27146191a37d4", size = 288324 },
-    { url = "https://files.pythonhosted.org/packages/09/c9/4e68181a4a652fb3ef5099e077faf4fd2a694ea6e0f806a7737aff9e758a/regex-2024.11.6-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:086a27a0b4ca227941700e0b31425e7a28ef1ae8e5e05a33826e17e47fbfdba0", size = 284617 },
-    { url = "https://files.pythonhosted.org/packages/fc/fd/37868b75eaf63843165f1d2122ca6cb94bfc0271e4428cf58c0616786dce/regex-2024.11.6-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bde01f35767c4a7899b7eb6e823b125a64de314a8ee9791367c9a34d56af18d0", size = 795023 },
-    { url = "https://files.pythonhosted.org/packages/c4/7c/d4cd9c528502a3dedb5c13c146e7a7a539a3853dc20209c8e75d9ba9d1b2/regex-2024.11.6-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b583904576650166b3d920d2bcce13971f6f9e9a396c673187f49811b2769dc7", size = 833072 },
-    { url = "https://files.pythonhosted.org/packages/4f/db/46f563a08f969159c5a0f0e722260568425363bea43bb7ae370becb66a67/regex-2024.11.6-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1c4de13f06a0d54fa0d5ab1b7138bfa0d883220965a29616e3ea61b35d5f5fc7", size = 823130 },
-    { url = "https://files.pythonhosted.org/packages/db/60/1eeca2074f5b87df394fccaa432ae3fc06c9c9bfa97c5051aed70e6e00c2/regex-2024.11.6-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3cde6e9f2580eb1665965ce9bf17ff4952f34f5b126beb509fee8f4e994f143c", size = 796857 },
-    { url = "https://files.pythonhosted.org/packages/10/db/ac718a08fcee981554d2f7bb8402f1faa7e868c1345c16ab1ebec54b0d7b/regex-2024.11.6-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0d7f453dca13f40a02b79636a339c5b62b670141e63efd511d3f8f73fba162b3", size = 784006 },
-    { url = "https://files.pythonhosted.org/packages/c2/41/7da3fe70216cea93144bf12da2b87367590bcf07db97604edeea55dac9ad/regex-2024.11.6-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:59dfe1ed21aea057a65c6b586afd2a945de04fc7db3de0a6e3ed5397ad491b07", size = 781650 },
-    { url = "https://files.pythonhosted.org/packages/a7/d5/880921ee4eec393a4752e6ab9f0fe28009435417c3102fc413f3fe81c4e5/regex-2024.11.6-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b97c1e0bd37c5cd7902e65f410779d39eeda155800b65fc4d04cc432efa9bc6e", size = 789545 },
-    { url = "https://files.pythonhosted.org/packages/dc/96/53770115e507081122beca8899ab7f5ae28ae790bfcc82b5e38976df6a77/regex-2024.11.6-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:f9d1e379028e0fc2ae3654bac3cbbef81bf3fd571272a42d56c24007979bafb6", size = 853045 },
-    { url = "https://files.pythonhosted.org/packages/31/d3/1372add5251cc2d44b451bd94f43b2ec78e15a6e82bff6a290ef9fd8f00a/regex-2024.11.6-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:13291b39131e2d002a7940fb176e120bec5145f3aeb7621be6534e46251912c4", size = 860182 },
-    { url = "https://files.pythonhosted.org/packages/ed/e3/c446a64984ea9f69982ba1a69d4658d5014bc7a0ea468a07e1a1265db6e2/regex-2024.11.6-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:4f51f88c126370dcec4908576c5a627220da6c09d0bff31cfa89f2523843316d", size = 787733 },
-    { url = "https://files.pythonhosted.org/packages/2b/f1/e40c8373e3480e4f29f2692bd21b3e05f296d3afebc7e5dcf21b9756ca1c/regex-2024.11.6-cp313-cp313-win32.whl", hash = "sha256:63b13cfd72e9601125027202cad74995ab26921d8cd935c25f09c630436348ff", size = 262122 },
-    { url = "https://files.pythonhosted.org/packages/45/94/bc295babb3062a731f52621cdc992d123111282e291abaf23faa413443ea/regex-2024.11.6-cp313-cp313-win_amd64.whl", hash = "sha256:2b3361af3198667e99927da8b84c1b010752fa4b1115ee30beaa332cabc3ef1a", size = 273545 },
-    { url = "https://files.pythonhosted.org/packages/89/23/c4a86df398e57e26f93b13ae63acce58771e04bdde86092502496fa57f9c/regex-2024.11.6-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:5704e174f8ccab2026bd2f1ab6c510345ae8eac818b613d7d73e785f1310f839", size = 482682 },
-    { url = "https://files.pythonhosted.org/packages/3c/8b/45c24ab7a51a1658441b961b86209c43e6bb9d39caf1e63f46ce6ea03bc7/regex-2024.11.6-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:220902c3c5cc6af55d4fe19ead504de80eb91f786dc102fbd74894b1551f095e", size = 287679 },
-    { url = "https://files.pythonhosted.org/packages/7a/d1/598de10b17fdafc452d11f7dada11c3be4e379a8671393e4e3da3c4070df/regex-2024.11.6-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:5e7e351589da0850c125f1600a4c4ba3c722efefe16b297de54300f08d734fbf", size = 284578 },
-    { url = "https://files.pythonhosted.org/packages/49/70/c7eaa219efa67a215846766fde18d92d54cb590b6a04ffe43cef30057622/regex-2024.11.6-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5056b185ca113c88e18223183aa1a50e66507769c9640a6ff75859619d73957b", size = 782012 },
-    { url = "https://files.pythonhosted.org/packages/89/e5/ef52c7eb117dd20ff1697968219971d052138965a4d3d9b95e92e549f505/regex-2024.11.6-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:2e34b51b650b23ed3354b5a07aab37034d9f923db2a40519139af34f485f77d0", size = 820580 },
-    { url = "https://files.pythonhosted.org/packages/5f/3f/9f5da81aff1d4167ac52711acf789df13e789fe6ac9545552e49138e3282/regex-2024.11.6-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5670bce7b200273eee1840ef307bfa07cda90b38ae56e9a6ebcc9f50da9c469b", size = 809110 },
-    { url = "https://files.pythonhosted.org/packages/86/44/2101cc0890c3621b90365c9ee8d7291a597c0722ad66eccd6ffa7f1bcc09/regex-2024.11.6-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:08986dce1339bc932923e7d1232ce9881499a0e02925f7402fb7c982515419ef", size = 780919 },
-    { url = "https://files.pythonhosted.org/packages/ce/2e/3e0668d8d1c7c3c0d397bf54d92fc182575b3a26939aed5000d3cc78760f/regex-2024.11.6-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:93c0b12d3d3bc25af4ebbf38f9ee780a487e8bf6954c115b9f015822d3bb8e48", size = 771515 },
-    { url = "https://files.pythonhosted.org/packages/a6/49/1bc4584254355e3dba930a3a2fd7ad26ccba3ebbab7d9100db0aff2eedb0/regex-2024.11.6-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl", hash = "sha256:764e71f22ab3b305e7f4c21f1a97e1526a25ebdd22513e251cf376760213da13", size = 696957 },
-    { url = "https://files.pythonhosted.org/packages/c8/dd/42879c1fc8a37a887cd08e358af3d3ba9e23038cd77c7fe044a86d9450ba/regex-2024.11.6-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:f056bf21105c2515c32372bbc057f43eb02aae2fda61052e2f7622c801f0b4e2", size = 768088 },
-    { url = "https://files.pythonhosted.org/packages/89/96/c05a0fe173cd2acd29d5e13c1adad8b706bcaa71b169e1ee57dcf2e74584/regex-2024.11.6-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:69ab78f848845569401469da20df3e081e6b5a11cb086de3eed1d48f5ed57c95", size = 774752 },
-    { url = "https://files.pythonhosted.org/packages/b5/f3/a757748066255f97f14506483436c5f6aded7af9e37bca04ec30c90ca683/regex-2024.11.6-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:86fddba590aad9208e2fa8b43b4c098bb0ec74f15718bb6a704e3c63e2cef3e9", size = 838862 },
-    { url = "https://files.pythonhosted.org/packages/5c/93/c6d2092fd479dcaeea40fc8fa673822829181ded77d294a7f950f1dda6e2/regex-2024.11.6-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:684d7a212682996d21ca12ef3c17353c021fe9de6049e19ac8481ec35574a70f", size = 842622 },
-    { url = "https://files.pythonhosted.org/packages/ff/9c/daa99532c72f25051a90ef90e1413a8d54413a9e64614d9095b0c1c154d0/regex-2024.11.6-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:a03e02f48cd1abbd9f3b7e3586d97c8f7a9721c436f51a5245b3b9483044480b", size = 772713 },
-    { url = "https://files.pythonhosted.org/packages/13/5d/61a533ccb8c231b474ac8e3a7d70155b00dfc61af6cafdccd1947df6d735/regex-2024.11.6-cp39-cp39-win32.whl", hash = "sha256:41758407fc32d5c3c5de163888068cfee69cb4c2be844e7ac517a52770f9af57", size = 261756 },
-    { url = "https://files.pythonhosted.org/packages/dc/7b/e59b7f7c91ae110d154370c24133f947262525b5d6406df65f23422acc17/regex-2024.11.6-cp39-cp39-win_amd64.whl", hash = "sha256:b2837718570f95dd41675328e111345f9b7095d821bac435aac173ac80b19983", size = 274110 },
+version = "2025.7.34"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0b/de/e13fa6dc61d78b30ba47481f99933a3b49a57779d625c392d8036770a60d/regex-2025.7.34.tar.gz", hash = "sha256:9ead9765217afd04a86822dfcd4ed2747dfe426e887da413b15ff0ac2457e21a", size = 400714, upload-time = "2025-07-31T00:21:16.262Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/50/d2/0a44a9d92370e5e105f16669acf801b215107efea9dea4317fe96e9aad67/regex-2025.7.34-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d856164d25e2b3b07b779bfed813eb4b6b6ce73c2fd818d46f47c1eb5cd79bd6", size = 484591, upload-time = "2025-07-31T00:18:46.675Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/b1/00c4f83aa902f1048495de9f2f33638ce970ce1cf9447b477d272a0e22bb/regex-2025.7.34-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2d15a9da5fad793e35fb7be74eec450d968e05d2e294f3e0e77ab03fa7234a83", size = 289293, upload-time = "2025-07-31T00:18:53.069Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/b0/5bc5c8ddc418e8be5530b43ae1f7c9303f43aeff5f40185c4287cf6732f2/regex-2025.7.34-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:95b4639c77d414efa93c8de14ce3f7965a94d007e068a94f9d4997bb9bd9c81f", size = 285932, upload-time = "2025-07-31T00:18:54.673Z" },
+    { url = "https://files.pythonhosted.org/packages/46/c7/a1a28d050b23665a5e1eeb4d7f13b83ea86f0bc018da7b8f89f86ff7f094/regex-2025.7.34-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5d7de1ceed5a5f84f342ba4a9f4ae589524adf9744b2ee61b5da884b5b659834", size = 780361, upload-time = "2025-07-31T00:18:56.13Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/0d/82e7afe7b2c9fe3d488a6ab6145d1d97e55f822dfb9b4569aba2497e3d09/regex-2025.7.34-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:02e5860a250cd350c4933cf376c3bc9cb28948e2c96a8bc042aee7b985cfa26f", size = 849176, upload-time = "2025-07-31T00:18:57.483Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/16/3036e16903d8194f1490af457a7e33b06d9e9edd9576b1fe6c7ac660e9ed/regex-2025.7.34-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0a5966220b9a1a88691282b7e4350e9599cf65780ca60d914a798cb791aa1177", size = 897222, upload-time = "2025-07-31T00:18:58.721Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/c2/010e089ae00d31418e7d2c6601760eea1957cde12be719730c7133b8c165/regex-2025.7.34-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:48fb045bbd4aab2418dc1ba2088a5e32de4bfe64e1457b948bb328a8dc2f1c2e", size = 789831, upload-time = "2025-07-31T00:19:00.436Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/86/b312b7bf5c46d21dbd9a3fdc4a80fde56ea93c9c0b89cf401879635e094d/regex-2025.7.34-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:20ff8433fa45e131f7316594efe24d4679c5449c0ca69d91c2f9d21846fdf064", size = 780665, upload-time = "2025-07-31T00:19:01.828Z" },
+    { url = "https://files.pythonhosted.org/packages/40/e5/674b82bfff112c820b09e3c86a423d4a568143ede7f8440fdcbce259e895/regex-2025.7.34-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:c436fd1e95c04c19039668cfb548450a37c13f051e8659f40aed426e36b3765f", size = 773511, upload-time = "2025-07-31T00:19:03.654Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/18/39e7c578eb6cf1454db2b64e4733d7e4f179714867a75d84492ec44fa9b2/regex-2025.7.34-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:0b85241d3cfb9f8a13cefdfbd58a2843f208f2ed2c88181bf84e22e0c7fc066d", size = 843990, upload-time = "2025-07-31T00:19:05.61Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/d9/522a6715aefe2f463dc60c68924abeeb8ab6893f01adf5720359d94ede8c/regex-2025.7.34-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:075641c94126b064c65ab86e7e71fc3d63e7ff1bea1fb794f0773c97cdad3a03", size = 834676, upload-time = "2025-07-31T00:19:07.023Z" },
+    { url = "https://files.pythonhosted.org/packages/59/53/c4d5284cb40543566542e24f1badc9f72af68d01db21e89e36e02292eee0/regex-2025.7.34-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:70645cad3407d103d1dbcb4841839d2946f7d36cf38acbd40120fee1682151e5", size = 778420, upload-time = "2025-07-31T00:19:08.511Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/4a/b779a7707d4a44a7e6ee9d0d98e40b2a4de74d622966080e9c95e25e2d24/regex-2025.7.34-cp310-cp310-win32.whl", hash = "sha256:3b836eb4a95526b263c2a3359308600bd95ce7848ebd3c29af0c37c4f9627cd3", size = 263999, upload-time = "2025-07-31T00:19:10.072Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/6e/33c7583f5427aa039c28bff7f4103c2de5b6aa5b9edc330c61ec576b1960/regex-2025.7.34-cp310-cp310-win_amd64.whl", hash = "sha256:cbfaa401d77334613cf434f723c7e8ba585df162be76474bccc53ae4e5520b3a", size = 276023, upload-time = "2025-07-31T00:19:11.34Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/fc/00b32e0ac14213d76d806d952826402b49fd06d42bfabacdf5d5d016bc47/regex-2025.7.34-cp310-cp310-win_arm64.whl", hash = "sha256:bca11d3c38a47c621769433c47f364b44e8043e0de8e482c5968b20ab90a3986", size = 268357, upload-time = "2025-07-31T00:19:12.729Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/85/f497b91577169472f7c1dc262a5ecc65e39e146fc3a52c571e5daaae4b7d/regex-2025.7.34-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:da304313761b8500b8e175eb2040c4394a875837d5635f6256d6fa0377ad32c8", size = 484594, upload-time = "2025-07-31T00:19:13.927Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/c5/ad2a5c11ce9e6257fcbfd6cd965d07502f6054aaa19d50a3d7fd991ec5d1/regex-2025.7.34-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:35e43ebf5b18cd751ea81455b19acfdec402e82fe0dc6143edfae4c5c4b3909a", size = 289294, upload-time = "2025-07-31T00:19:15.395Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/01/83ffd9641fcf5e018f9b51aa922c3e538ac9439424fda3df540b643ecf4f/regex-2025.7.34-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:96bbae4c616726f4661fe7bcad5952e10d25d3c51ddc388189d8864fbc1b3c68", size = 285933, upload-time = "2025-07-31T00:19:16.704Z" },
+    { url = "https://files.pythonhosted.org/packages/77/20/5edab2e5766f0259bc1da7381b07ce6eb4401b17b2254d02f492cd8a81a8/regex-2025.7.34-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9feab78a1ffa4f2b1e27b1bcdaad36f48c2fed4870264ce32f52a393db093c78", size = 792335, upload-time = "2025-07-31T00:19:18.561Z" },
+    { url = "https://files.pythonhosted.org/packages/30/bd/744d3ed8777dce8487b2606b94925e207e7c5931d5870f47f5b643a4580a/regex-2025.7.34-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f14b36e6d4d07f1a5060f28ef3b3561c5d95eb0651741474ce4c0a4c56ba8719", size = 858605, upload-time = "2025-07-31T00:19:20.204Z" },
+    { url = "https://files.pythonhosted.org/packages/99/3d/93754176289718d7578c31d151047e7b8acc7a8c20e7706716f23c49e45e/regex-2025.7.34-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:85c3a958ef8b3d5079c763477e1f09e89d13ad22198a37e9d7b26b4b17438b33", size = 905780, upload-time = "2025-07-31T00:19:21.876Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/2e/c689f274a92deffa03999a430505ff2aeace408fd681a90eafa92fdd6930/regex-2025.7.34-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:37555e4ae0b93358fa7c2d240a4291d4a4227cc7c607d8f85596cdb08ec0a083", size = 798868, upload-time = "2025-07-31T00:19:23.222Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/9e/39673688805d139b33b4a24851a71b9978d61915c4d72b5ffda324d0668a/regex-2025.7.34-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ee38926f31f1aa61b0232a3a11b83461f7807661c062df9eb88769d86e6195c3", size = 781784, upload-time = "2025-07-31T00:19:24.59Z" },
+    { url = "https://files.pythonhosted.org/packages/18/bd/4c1cab12cfabe14beaa076523056b8ab0c882a8feaf0a6f48b0a75dab9ed/regex-2025.7.34-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a664291c31cae9c4a30589bd8bc2ebb56ef880c9c6264cb7643633831e606a4d", size = 852837, upload-time = "2025-07-31T00:19:25.911Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/21/663d983cbb3bba537fc213a579abbd0f263fb28271c514123f3c547ab917/regex-2025.7.34-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:f3e5c1e0925e77ec46ddc736b756a6da50d4df4ee3f69536ffb2373460e2dafd", size = 844240, upload-time = "2025-07-31T00:19:27.688Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/2d/9beeeb913bc5d32faa913cf8c47e968da936af61ec20af5d269d0f84a100/regex-2025.7.34-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:d428fc7731dcbb4e2ffe43aeb8f90775ad155e7db4347a639768bc6cd2df881a", size = 787139, upload-time = "2025-07-31T00:19:29.475Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/f5/9b9384415fdc533551be2ba805dd8c4621873e5df69c958f403bfd3b2b6e/regex-2025.7.34-cp311-cp311-win32.whl", hash = "sha256:e154a7ee7fa18333ad90b20e16ef84daaeac61877c8ef942ec8dfa50dc38b7a1", size = 264019, upload-time = "2025-07-31T00:19:31.129Z" },
+    { url = "https://files.pythonhosted.org/packages/18/9d/e069ed94debcf4cc9626d652a48040b079ce34c7e4fb174f16874958d485/regex-2025.7.34-cp311-cp311-win_amd64.whl", hash = "sha256:24257953d5c1d6d3c129ab03414c07fc1a47833c9165d49b954190b2b7f21a1a", size = 276047, upload-time = "2025-07-31T00:19:32.497Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/cf/3bafbe9d1fd1db77355e7fbbbf0d0cfb34501a8b8e334deca14f94c7b315/regex-2025.7.34-cp311-cp311-win_arm64.whl", hash = "sha256:3157aa512b9e606586900888cd469a444f9b898ecb7f8931996cb715f77477f0", size = 268362, upload-time = "2025-07-31T00:19:34.094Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/f0/31d62596c75a33f979317658e8d261574785c6cd8672c06741ce2e2e2070/regex-2025.7.34-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:7f7211a746aced993bef487de69307a38c5ddd79257d7be83f7b202cb59ddb50", size = 485492, upload-time = "2025-07-31T00:19:35.57Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/16/b818d223f1c9758c3434be89aa1a01aae798e0e0df36c1f143d1963dd1ee/regex-2025.7.34-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:fb31080f2bd0681484b275461b202b5ad182f52c9ec606052020fe13eb13a72f", size = 290000, upload-time = "2025-07-31T00:19:37.175Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/70/69506d53397b4bd6954061bae75677ad34deb7f6ca3ba199660d6f728ff5/regex-2025.7.34-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:0200a5150c4cf61e407038f4b4d5cdad13e86345dac29ff9dab3d75d905cf130", size = 286072, upload-time = "2025-07-31T00:19:38.612Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/73/536a216d5f66084fb577bb0543b5cb7de3272eb70a157f0c3a542f1c2551/regex-2025.7.34-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:739a74970e736df0773788377969c9fea3876c2fc13d0563f98e5503e5185f46", size = 797341, upload-time = "2025-07-31T00:19:40.119Z" },
+    { url = "https://files.pythonhosted.org/packages/26/af/733f8168449e56e8f404bb807ea7189f59507cbea1b67a7bbcd92f8bf844/regex-2025.7.34-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4fef81b2f7ea6a2029161ed6dea9ae13834c28eb5a95b8771828194a026621e4", size = 862556, upload-time = "2025-07-31T00:19:41.556Z" },
+    { url = "https://files.pythonhosted.org/packages/19/dd/59c464d58c06c4f7d87de4ab1f590e430821345a40c5d345d449a636d15f/regex-2025.7.34-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:ea74cf81fe61a7e9d77989050d0089a927ab758c29dac4e8e1b6c06fccf3ebf0", size = 910762, upload-time = "2025-07-31T00:19:43Z" },
+    { url = "https://files.pythonhosted.org/packages/37/a8/b05ccf33ceca0815a1e253693b2c86544932ebcc0049c16b0fbdf18b688b/regex-2025.7.34-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e4636a7f3b65a5f340ed9ddf53585c42e3ff37101d383ed321bfe5660481744b", size = 801892, upload-time = "2025-07-31T00:19:44.645Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/9a/b993cb2e634cc22810afd1652dba0cae156c40d4864285ff486c73cd1996/regex-2025.7.34-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:6cef962d7834437fe8d3da6f9bfc6f93f20f218266dcefec0560ed7765f5fe01", size = 786551, upload-time = "2025-07-31T00:19:46.127Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/79/7849d67910a0de4e26834b5bb816e028e35473f3d7ae563552ea04f58ca2/regex-2025.7.34-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:cbe1698e5b80298dbce8df4d8d1182279fbdaf1044e864cbc9d53c20e4a2be77", size = 856457, upload-time = "2025-07-31T00:19:47.562Z" },
+    { url = "https://files.pythonhosted.org/packages/91/c6/de516bc082524b27e45cb4f54e28bd800c01efb26d15646a65b87b13a91e/regex-2025.7.34-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:32b9f9bcf0f605eb094b08e8da72e44badabb63dde6b83bd530580b488d1c6da", size = 848902, upload-time = "2025-07-31T00:19:49.312Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/22/519ff8ba15f732db099b126f039586bd372da6cd4efb810d5d66a5daeda1/regex-2025.7.34-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:524c868ba527eab4e8744a9287809579f54ae8c62fbf07d62aacd89f6026b282", size = 788038, upload-time = "2025-07-31T00:19:50.794Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/7d/aabb467d8f57d8149895d133c88eb809a1a6a0fe262c1d508eb9dfabb6f9/regex-2025.7.34-cp312-cp312-win32.whl", hash = "sha256:d600e58ee6d036081c89696d2bdd55d507498a7180df2e19945c6642fac59588", size = 264417, upload-time = "2025-07-31T00:19:52.292Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/39/bd922b55a4fc5ad5c13753274e5b536f5b06ec8eb9747675668491c7ab7a/regex-2025.7.34-cp312-cp312-win_amd64.whl", hash = "sha256:9a9ab52a466a9b4b91564437b36417b76033e8778e5af8f36be835d8cb370d62", size = 275387, upload-time = "2025-07-31T00:19:53.593Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/3c/c61d2fdcecb754a40475a3d1ef9a000911d3e3fc75c096acf44b0dfb786a/regex-2025.7.34-cp312-cp312-win_arm64.whl", hash = "sha256:c83aec91af9c6fbf7c743274fd952272403ad9a9db05fe9bfc9df8d12b45f176", size = 268482, upload-time = "2025-07-31T00:19:55.183Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/7f/8333b894499c1172c0378bb45a80146c420621e5c7b27a1d8fc5456f7038/regex-2025.7.34-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:fd5edc3f453de727af267c7909d083e19f6426fc9dd149e332b6034f2a5611e6", size = 484602, upload-time = "2025-07-31T00:20:48.184Z" },
+    { url = "https://files.pythonhosted.org/packages/14/47/58aac4758b659df3835e73bda070f78ec6620a028484a1fcb81daf7443ec/regex-2025.7.34-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:fa1cdfb8db96ef20137de5587954c812821966c3e8b48ffc871e22d7ec0a4938", size = 289289, upload-time = "2025-07-31T00:20:49.79Z" },
+    { url = "https://files.pythonhosted.org/packages/46/cc/5c9ebdc23b34458a41b559e0ae1b759196b2212920164b9d8aae4b25aa26/regex-2025.7.34-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:89c9504fc96268e8e74b0283e548f53a80c421182a2007e3365805b74ceef936", size = 285931, upload-time = "2025-07-31T00:20:51.362Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/da/467a851615b040d3be478ef60fd2d54e7e2f44eeda65dc02866ad4e404df/regex-2025.7.34-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:33be70d75fa05a904ee0dc43b650844e067d14c849df7e82ad673541cd465b5f", size = 779782, upload-time = "2025-07-31T00:20:52.997Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/47/6eab7100b7ded84e94312c6791ab72581950b7adaa5ad48cdd3dfa329ab8/regex-2025.7.34-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:57d25b6732ea93eeb1d090e8399b6235ca84a651b52d52d272ed37d3d2efa0f1", size = 848838, upload-time = "2025-07-31T00:20:54.991Z" },
+    { url = "https://files.pythonhosted.org/packages/17/86/3b07305698e7ff21cc472efae816a56e77c5d45c6b7fe250a56dd67a114e/regex-2025.7.34-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:baf2fe122a3db1c0b9f161aa44463d8f7e33eeeda47bb0309923deb743a18276", size = 896648, upload-time = "2025-07-31T00:20:56.655Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/9a/c8f4f0535bf953e34e068c9a30c946e7affa06a48c48c1eda6d3a7562c49/regex-2025.7.34-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1a764a83128af9c1a54be81485b34dca488cbcacefe1e1d543ef11fbace191e1", size = 789367, upload-time = "2025-07-31T00:20:58.359Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/4e/1892685a0e053d376fbcb8aa618e38afc5882bd69d94e9712171b9f2a412/regex-2025.7.34-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:c7f663ccc4093877f55b51477522abd7299a14c5bb7626c5238599db6a0cb95d", size = 780029, upload-time = "2025-07-31T00:21:00.383Z" },
+    { url = "https://files.pythonhosted.org/packages/98/12/af86906b9342d37b051b076a3ccc925c4f33ff2a96328b3009e7b93dfc53/regex-2025.7.34-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:4913f52fbc7a744aaebf53acd8d3dc1b519e46ba481d4d7596de3c862e011ada", size = 773039, upload-time = "2025-07-31T00:21:02.093Z" },
+    { url = "https://files.pythonhosted.org/packages/97/d1/03c21fb12daf73819f39927b533d09f162e8e452bd415993607242c1cd68/regex-2025.7.34-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:efac4db9e044d47fd3b6b0d40b6708f4dfa2d8131a5ac1d604064147c0f552fd", size = 843438, upload-time = "2025-07-31T00:21:04.248Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/7f/53569415d23dc47122c9f669db5d1e7aa2bd8954723e5c1050548cb7622e/regex-2025.7.34-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:7373afae7cfb716e3b8e15d0184510d518f9d21471f2d62918dbece85f2c588f", size = 834053, upload-time = "2025-07-31T00:21:06.298Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/7a/9b6b75778f7af6306ad9dcd9860be3f9c4123385cc856b6e9d099a6403b2/regex-2025.7.34-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:9960d162f3fecf6af252534a1ae337e9c2e20d74469fed782903b24e2cc9d3d7", size = 777909, upload-time = "2025-07-31T00:21:08.302Z" },
+    { url = "https://files.pythonhosted.org/packages/54/34/ebdf85bef946c63dc7995e95710364de0e3e2791bc28afc1a9642373d6c1/regex-2025.7.34-cp39-cp39-win32.whl", hash = "sha256:95d538b10eb4621350a54bf14600cc80b514211d91a019dc74b8e23d2159ace5", size = 264039, upload-time = "2025-07-31T00:21:10.346Z" },
+    { url = "https://files.pythonhosted.org/packages/82/0b/fba6f0dee661b838c09c85bf598a43a915d310648d62f704ece237aa3d73/regex-2025.7.34-cp39-cp39-win_amd64.whl", hash = "sha256:f7f3071b5faa605b0ea51ec4bb3ea7257277446b053f4fd3ad02b1dcb4e64353", size = 276120, upload-time = "2025-07-31T00:21:12.321Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/6d/183f0cf19bd8ac7628f4c3b2ca99033a5ad417ad010f86c61d11d27b4968/regex-2025.7.34-cp39-cp39-win_arm64.whl", hash = "sha256:716a47515ba1d03f8e8a61c5013041c8c90f2e21f055203498105d7571b44531", size = 268390, upload-time = "2025-07-31T00:21:14.293Z" },
 ]
 
 [[package]]
 name = "requests"
-version = "2.32.3"
+version = "2.32.5"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "certifi" },
@@ -2836,169 +3614,175 @@ dependencies = [
     { name = "idna" },
     { name = "urllib3" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/63/70/2bf7780ad2d390a8d301ad0b550f1581eadbd9a20f896afe06353c2a2913/requests-2.32.3.tar.gz", hash = "sha256:55365417734eb18255590a9ff9eb97e9e1da868d4ccd6402399eaf68af20a760", size = 131218 }
+sdist = { url = "https://files.pythonhosted.org/packages/c9/74/b3ff8e6c8446842c3f5c837e9c3dfcfe2018ea6ecef224c710c85ef728f4/requests-2.32.5.tar.gz", hash = "sha256:dbba0bac56e100853db0ea71b82b4dfd5fe2bf6d3754a8893c3af500cec7d7cf", size = 134517, upload-time = "2025-08-18T20:46:02.573Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f9/9b/335f9764261e915ed497fcdeb11df5dfd6f7bf257d4a6a2a686d80da4d54/requests-2.32.3-py3-none-any.whl", hash = "sha256:70761cfe03c773ceb22aa2f671b4757976145175cdfca038c02654d061d6dcc6", size = 64928 },
+    { url = "https://files.pythonhosted.org/packages/1e/db/4254e3eabe8020b458f1a747140d32277ec7a271daf1d235b70dc0b4e6e3/requests-2.32.5-py3-none-any.whl", hash = "sha256:2462f94637a34fd532264295e186976db0f5d453d1cdd31473c85a6a161affb6", size = 64738, upload-time = "2025-08-18T20:46:00.542Z" },
 ]
 
 [[package]]
 name = "responses"
-version = "0.25.6"
+version = "0.25.8"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "pyyaml" },
     { name = "requests" },
     { name = "urllib3" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/97/63/759996eea0f17e8dc4c9ea9c60765292d28a7750bdbee073ad55d83caa57/responses-0.25.6.tar.gz", hash = "sha256:eae7ce61a9603004e76c05691e7c389e59652d91e94b419623c12bbfb8e331d8", size = 79145 }
+sdist = { url = "https://files.pythonhosted.org/packages/0e/95/89c054ad70bfef6da605338b009b2e283485835351a9935c7bfbfaca7ffc/responses-0.25.8.tar.gz", hash = "sha256:9374d047a575c8f781b94454db5cab590b6029505f488d12899ddb10a4af1cf4", size = 79320, upload-time = "2025-08-08T19:01:46.709Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/92/c4/8d23584b3a3471ea6f5a18cfb035e11eeb9fa9b3112d901477c6ad10cc4e/responses-0.25.6-py3-none-any.whl", hash = "sha256:9cac8f21e1193bb150ec557875377e41ed56248aed94e4567ed644db564bacf1", size = 34730 },
+    { url = "https://files.pythonhosted.org/packages/1c/4c/cc276ce57e572c102d9542d383b2cfd551276581dc60004cb94fe8774c11/responses-0.25.8-py3-none-any.whl", hash = "sha256:0c710af92def29c8352ceadff0c3fe340ace27cf5af1bbe46fb71275bcd2831c", size = 34769, upload-time = "2025-08-08T19:01:45.018Z" },
 ]
 
 [[package]]
 name = "rich"
-version = "13.9.4"
+version = "14.1.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "markdown-it-py" },
+    { name = "markdown-it-py", version = "3.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "markdown-it-py", version = "4.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pygments" },
-    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ab/3a/0316b28d0761c6734d6bc14e770d85506c986c85ffb239e688eeaab2c2bc/rich-13.9.4.tar.gz", hash = "sha256:439594978a49a09530cff7ebc4b5c7103ef57baf48d5ea3184f21d9a2befa098", size = 223149 }
+sdist = { url = "https://files.pythonhosted.org/packages/fe/75/af448d8e52bf1d8fa6a9d089ca6c07ff4453d86c65c145d0a300bb073b9b/rich-14.1.0.tar.gz", hash = "sha256:e497a48b844b0320d45007cdebfeaeed8db2a4f4bcf49f15e455cfc4af11eaa8", size = 224441, upload-time = "2025-07-25T07:32:58.125Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/19/71/39c7c0d87f8d4e6c020a393182060eaefeeae6c01dab6a84ec346f2567df/rich-13.9.4-py3-none-any.whl", hash = "sha256:6049d5e6ec054bf2779ab3358186963bac2ea89175919d699e378b99738c2a90", size = 242424 },
+    { url = "https://files.pythonhosted.org/packages/e3/30/3c4d035596d3cf444529e0b2953ad0466f6049528a879d27534700580395/rich-14.1.0-py3-none-any.whl", hash = "sha256:536f5f1785986d6dbdea3c75205c473f970777b4a0d6c6dd1b696aa05a3fa04f", size = 243368, upload-time = "2025-07-25T07:32:56.73Z" },
 ]
 
 [[package]]
 name = "rpds-py"
-version = "0.22.3"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/01/80/cce854d0921ff2f0a9fa831ba3ad3c65cee3a46711addf39a2af52df2cfd/rpds_py-0.22.3.tar.gz", hash = "sha256:e32fee8ab45d3c2db6da19a5323bc3362237c8b653c70194414b892fd06a080d", size = 26771 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/42/2a/ead1d09e57449b99dcc190d8d2323e3a167421d8f8fdf0f217c6f6befe47/rpds_py-0.22.3-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:6c7b99ca52c2c1752b544e310101b98a659b720b21db00e65edca34483259967", size = 359514 },
-    { url = "https://files.pythonhosted.org/packages/8f/7e/1254f406b7793b586c68e217a6a24ec79040f85e030fff7e9049069284f4/rpds_py-0.22.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:be2eb3f2495ba669d2a985f9b426c1797b7d48d6963899276d22f23e33d47e37", size = 349031 },
-    { url = "https://files.pythonhosted.org/packages/aa/da/17c6a2c73730d426df53675ff9cc6653ac7a60b6438d03c18e1c822a576a/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:70eb60b3ae9245ddea20f8a4190bd79c705a22f8028aaf8bbdebe4716c3fab24", size = 381485 },
-    { url = "https://files.pythonhosted.org/packages/aa/13/2dbacd820466aa2a3c4b747afb18d71209523d353cf865bf8f4796c969ea/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:4041711832360a9b75cfb11b25a6a97c8fb49c07b8bd43d0d02b45d0b499a4ff", size = 386794 },
-    { url = "https://files.pythonhosted.org/packages/6d/62/96905d0a35ad4e4bc3c098b2f34b2e7266e211d08635baa690643d2227be/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:64607d4cbf1b7e3c3c8a14948b99345eda0e161b852e122c6bb71aab6d1d798c", size = 423523 },
-    { url = "https://files.pythonhosted.org/packages/eb/1b/d12770f2b6a9fc2c3ec0d810d7d440f6d465ccd8b7f16ae5385952c28b89/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:81e69b0a0e2537f26d73b4e43ad7bc8c8efb39621639b4434b76a3de50c6966e", size = 446695 },
-    { url = "https://files.pythonhosted.org/packages/4d/cf/96f1fd75512a017f8e07408b6d5dbeb492d9ed46bfe0555544294f3681b3/rpds_py-0.22.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc27863442d388870c1809a87507727b799c8460573cfbb6dc0eeaef5a11b5ec", size = 381959 },
-    { url = "https://files.pythonhosted.org/packages/ab/f0/d1c5b501c8aea85aeb938b555bfdf7612110a2f8cdc21ae0482c93dd0c24/rpds_py-0.22.3-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:e79dd39f1e8c3504be0607e5fc6e86bb60fe3584bec8b782578c3b0fde8d932c", size = 410420 },
-    { url = "https://files.pythonhosted.org/packages/33/3b/45b6c58fb6aad5a569ae40fb890fc494c6b02203505a5008ee6dc68e65f7/rpds_py-0.22.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:e0fa2d4ec53dc51cf7d3bb22e0aa0143966119f42a0c3e4998293a3dd2856b09", size = 557620 },
-    { url = "https://files.pythonhosted.org/packages/83/62/3fdd2d3d47bf0bb9b931c4c73036b4ab3ec77b25e016ae26fab0f02be2af/rpds_py-0.22.3-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fda7cb070f442bf80b642cd56483b5548e43d366fe3f39b98e67cce780cded00", size = 584202 },
-    { url = "https://files.pythonhosted.org/packages/04/f2/5dced98b64874b84ca824292f9cee2e3f30f3bcf231d15a903126684f74d/rpds_py-0.22.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:cff63a0272fcd259dcc3be1657b07c929c466b067ceb1c20060e8d10af56f5bf", size = 552787 },
-    { url = "https://files.pythonhosted.org/packages/67/13/2273dea1204eda0aea0ef55145da96a9aa28b3f88bb5c70e994f69eda7c3/rpds_py-0.22.3-cp310-cp310-win32.whl", hash = "sha256:9bd7228827ec7bb817089e2eb301d907c0d9827a9e558f22f762bb690b131652", size = 220088 },
-    { url = "https://files.pythonhosted.org/packages/4e/80/8c8176b67ad7f4a894967a7a4014ba039626d96f1d4874d53e409b58d69f/rpds_py-0.22.3-cp310-cp310-win_amd64.whl", hash = "sha256:9beeb01d8c190d7581a4d59522cd3d4b6887040dcfc744af99aa59fef3e041a8", size = 231737 },
-    { url = "https://files.pythonhosted.org/packages/15/ad/8d1ddf78f2805a71253fcd388017e7b4a0615c22c762b6d35301fef20106/rpds_py-0.22.3-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:d20cfb4e099748ea39e6f7b16c91ab057989712d31761d3300d43134e26e165f", size = 359773 },
-    { url = "https://files.pythonhosted.org/packages/c8/75/68c15732293a8485d79fe4ebe9045525502a067865fa4278f178851b2d87/rpds_py-0.22.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:68049202f67380ff9aa52f12e92b1c30115f32e6895cd7198fa2a7961621fc5a", size = 349214 },
-    { url = "https://files.pythonhosted.org/packages/3c/4c/7ce50f3070083c2e1b2bbd0fb7046f3da55f510d19e283222f8f33d7d5f4/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fb4f868f712b2dd4bcc538b0a0c1f63a2b1d584c925e69a224d759e7070a12d5", size = 380477 },
-    { url = "https://files.pythonhosted.org/packages/9a/e9/835196a69cb229d5c31c13b8ae603bd2da9a6695f35fe4270d398e1db44c/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:bc51abd01f08117283c5ebf64844a35144a0843ff7b2983e0648e4d3d9f10dbb", size = 386171 },
-    { url = "https://files.pythonhosted.org/packages/f9/8e/33fc4eba6683db71e91e6d594a2cf3a8fbceb5316629f0477f7ece5e3f75/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0f3cec041684de9a4684b1572fe28c7267410e02450f4561700ca5a3bc6695a2", size = 422676 },
-    { url = "https://files.pythonhosted.org/packages/37/47/2e82d58f8046a98bb9497a8319604c92b827b94d558df30877c4b3c6ccb3/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7ef9d9da710be50ff6809fed8f1963fecdfecc8b86656cadfca3bc24289414b0", size = 446152 },
-    { url = "https://files.pythonhosted.org/packages/e1/78/79c128c3e71abbc8e9739ac27af11dc0f91840a86fce67ff83c65d1ba195/rpds_py-0.22.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:59f4a79c19232a5774aee369a0c296712ad0e77f24e62cad53160312b1c1eaa1", size = 381300 },
-    { url = "https://files.pythonhosted.org/packages/c9/5b/2e193be0e8b228c1207f31fa3ea79de64dadb4f6a4833111af8145a6bc33/rpds_py-0.22.3-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1a60bce91f81ddaac922a40bbb571a12c1070cb20ebd6d49c48e0b101d87300d", size = 409636 },
-    { url = "https://files.pythonhosted.org/packages/c2/3f/687c7100b762d62186a1c1100ffdf99825f6fa5ea94556844bbbd2d0f3a9/rpds_py-0.22.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:e89391e6d60251560f0a8f4bd32137b077a80d9b7dbe6d5cab1cd80d2746f648", size = 556708 },
-    { url = "https://files.pythonhosted.org/packages/8c/a2/c00cbc4b857e8b3d5e7f7fc4c81e23afd8c138b930f4f3ccf9a41a23e9e4/rpds_py-0.22.3-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:e3fb866d9932a3d7d0c82da76d816996d1667c44891bd861a0f97ba27e84fc74", size = 583554 },
-    { url = "https://files.pythonhosted.org/packages/d0/08/696c9872cf56effdad9ed617ac072f6774a898d46b8b8964eab39ec562d2/rpds_py-0.22.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:1352ae4f7c717ae8cba93421a63373e582d19d55d2ee2cbb184344c82d2ae55a", size = 552105 },
-    { url = "https://files.pythonhosted.org/packages/18/1f/4df560be1e994f5adf56cabd6c117e02de7c88ee238bb4ce03ed50da9d56/rpds_py-0.22.3-cp311-cp311-win32.whl", hash = "sha256:b0b4136a252cadfa1adb705bb81524eee47d9f6aab4f2ee4fa1e9d3cd4581f64", size = 220199 },
-    { url = "https://files.pythonhosted.org/packages/b8/1b/c29b570bc5db8237553002788dc734d6bd71443a2ceac2a58202ec06ef12/rpds_py-0.22.3-cp311-cp311-win_amd64.whl", hash = "sha256:8bd7c8cfc0b8247c8799080fbff54e0b9619e17cdfeb0478ba7295d43f635d7c", size = 231775 },
-    { url = "https://files.pythonhosted.org/packages/75/47/3383ee3bd787a2a5e65a9b9edc37ccf8505c0a00170e3a5e6ea5fbcd97f7/rpds_py-0.22.3-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:27e98004595899949bd7a7b34e91fa7c44d7a97c40fcaf1d874168bb652ec67e", size = 352334 },
-    { url = "https://files.pythonhosted.org/packages/40/14/aa6400fa8158b90a5a250a77f2077c0d0cd8a76fce31d9f2b289f04c6dec/rpds_py-0.22.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1978d0021e943aae58b9b0b196fb4895a25cc53d3956b8e35e0b7682eefb6d56", size = 342111 },
-    { url = "https://files.pythonhosted.org/packages/7d/06/395a13bfaa8a28b302fb433fb285a67ce0ea2004959a027aea8f9c52bad4/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:655ca44a831ecb238d124e0402d98f6212ac527a0ba6c55ca26f616604e60a45", size = 384286 },
-    { url = "https://files.pythonhosted.org/packages/43/52/d8eeaffab047e6b7b7ef7f00d5ead074a07973968ffa2d5820fa131d7852/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:feea821ee2a9273771bae61194004ee2fc33f8ec7db08117ef9147d4bbcbca8e", size = 391739 },
-    { url = "https://files.pythonhosted.org/packages/83/31/52dc4bde85c60b63719610ed6f6d61877effdb5113a72007679b786377b8/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:22bebe05a9ffc70ebfa127efbc429bc26ec9e9b4ee4d15a740033efda515cf3d", size = 427306 },
-    { url = "https://files.pythonhosted.org/packages/70/d5/1bab8e389c2261dba1764e9e793ed6830a63f830fdbec581a242c7c46bda/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3af6e48651c4e0d2d166dc1b033b7042ea3f871504b6805ba5f4fe31581d8d38", size = 442717 },
-    { url = "https://files.pythonhosted.org/packages/82/a1/a45f3e30835b553379b3a56ea6c4eb622cf11e72008229af840e4596a8ea/rpds_py-0.22.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e67ba3c290821343c192f7eae1d8fd5999ca2dc99994114643e2f2d3e6138b15", size = 385721 },
-    { url = "https://files.pythonhosted.org/packages/a6/27/780c942de3120bdd4d0e69583f9c96e179dfff082f6ecbb46b8d6488841f/rpds_py-0.22.3-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:02fbb9c288ae08bcb34fb41d516d5eeb0455ac35b5512d03181d755d80810059", size = 415824 },
-    { url = "https://files.pythonhosted.org/packages/94/0b/aa0542ca88ad20ea719b06520f925bae348ea5c1fdf201b7e7202d20871d/rpds_py-0.22.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:f56a6b404f74ab372da986d240e2e002769a7d7102cc73eb238a4f72eec5284e", size = 561227 },
-    { url = "https://files.pythonhosted.org/packages/0d/92/3ed77d215f82c8f844d7f98929d56cc321bb0bcfaf8f166559b8ec56e5f1/rpds_py-0.22.3-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:0a0461200769ab3b9ab7e513f6013b7a97fdeee41c29b9db343f3c5a8e2b9e61", size = 587424 },
-    { url = "https://files.pythonhosted.org/packages/09/42/cacaeb047a22cab6241f107644f230e2935d4efecf6488859a7dd82fc47d/rpds_py-0.22.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:8633e471c6207a039eff6aa116e35f69f3156b3989ea3e2d755f7bc41754a4a7", size = 555953 },
-    { url = "https://files.pythonhosted.org/packages/e6/52/c921dc6d5f5d45b212a456c1f5b17df1a471127e8037eb0972379e39dff4/rpds_py-0.22.3-cp312-cp312-win32.whl", hash = "sha256:593eba61ba0c3baae5bc9be2f5232430453fb4432048de28399ca7376de9c627", size = 221339 },
-    { url = "https://files.pythonhosted.org/packages/f2/c7/f82b5be1e8456600395366f86104d1bd8d0faed3802ad511ef6d60c30d98/rpds_py-0.22.3-cp312-cp312-win_amd64.whl", hash = "sha256:d115bffdd417c6d806ea9069237a4ae02f513b778e3789a359bc5856e0404cc4", size = 235786 },
-    { url = "https://files.pythonhosted.org/packages/d0/bf/36d5cc1f2c609ae6e8bf0fc35949355ca9d8790eceb66e6385680c951e60/rpds_py-0.22.3-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:ea7433ce7e4bfc3a85654aeb6747babe3f66eaf9a1d0c1e7a4435bbdf27fea84", size = 351657 },
-    { url = "https://files.pythonhosted.org/packages/24/2a/f1e0fa124e300c26ea9382e59b2d582cba71cedd340f32d1447f4f29fa4e/rpds_py-0.22.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:6dd9412824c4ce1aca56c47b0991e65bebb7ac3f4edccfd3f156150c96a7bf25", size = 341829 },
-    { url = "https://files.pythonhosted.org/packages/cf/c2/0da1231dd16953845bed60d1a586fcd6b15ceaeb965f4d35cdc71f70f606/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:20070c65396f7373f5df4005862fa162db5d25d56150bddd0b3e8214e8ef45b4", size = 384220 },
-    { url = "https://files.pythonhosted.org/packages/c7/73/a4407f4e3a00a9d4b68c532bf2d873d6b562854a8eaff8faa6133b3588ec/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:0b09865a9abc0ddff4e50b5ef65467cd94176bf1e0004184eb915cbc10fc05c5", size = 391009 },
-    { url = "https://files.pythonhosted.org/packages/a9/c3/04b7353477ab360fe2563f5f0b176d2105982f97cd9ae80a9c5a18f1ae0f/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3453e8d41fe5f17d1f8e9c383a7473cd46a63661628ec58e07777c2fff7196dc", size = 426989 },
-    { url = "https://files.pythonhosted.org/packages/8d/e6/e4b85b722bcf11398e17d59c0f6049d19cd606d35363221951e6d625fcb0/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f5d36399a1b96e1a5fdc91e0522544580dbebeb1f77f27b2b0ab25559e103b8b", size = 441544 },
-    { url = "https://files.pythonhosted.org/packages/27/fc/403e65e56f65fff25f2973216974976d3f0a5c3f30e53758589b6dc9b79b/rpds_py-0.22.3-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:009de23c9c9ee54bf11303a966edf4d9087cd43a6003672e6aa7def643d06518", size = 385179 },
-    { url = "https://files.pythonhosted.org/packages/57/9b/2be9ff9700d664d51fd96b33d6595791c496d2778cb0b2a634f048437a55/rpds_py-0.22.3-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1aef18820ef3e4587ebe8b3bc9ba6e55892a6d7b93bac6d29d9f631a3b4befbd", size = 415103 },
-    { url = "https://files.pythonhosted.org/packages/bb/a5/03c2ad8ca10994fcf22dd2150dd1d653bc974fa82d9a590494c84c10c641/rpds_py-0.22.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f60bd8423be1d9d833f230fdbccf8f57af322d96bcad6599e5a771b151398eb2", size = 560916 },
-    { url = "https://files.pythonhosted.org/packages/ba/2e/be4fdfc8b5b576e588782b56978c5b702c5a2307024120d8aeec1ab818f0/rpds_py-0.22.3-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:62d9cfcf4948683a18a9aff0ab7e1474d407b7bab2ca03116109f8464698ab16", size = 587062 },
-    { url = "https://files.pythonhosted.org/packages/67/e0/2034c221937709bf9c542603d25ad43a68b4b0a9a0c0b06a742f2756eb66/rpds_py-0.22.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:9253fc214112405f0afa7db88739294295f0e08466987f1d70e29930262b4c8f", size = 555734 },
-    { url = "https://files.pythonhosted.org/packages/ea/ce/240bae07b5401a22482b58e18cfbabaa392409b2797da60223cca10d7367/rpds_py-0.22.3-cp313-cp313-win32.whl", hash = "sha256:fb0ba113b4983beac1a2eb16faffd76cb41e176bf58c4afe3e14b9c681f702de", size = 220663 },
-    { url = "https://files.pythonhosted.org/packages/cb/f0/d330d08f51126330467edae2fa4efa5cec8923c87551a79299380fdea30d/rpds_py-0.22.3-cp313-cp313-win_amd64.whl", hash = "sha256:c58e2339def52ef6b71b8f36d13c3688ea23fa093353f3a4fee2556e62086ec9", size = 235503 },
-    { url = "https://files.pythonhosted.org/packages/f7/c4/dbe1cc03df013bf2feb5ad00615038050e7859f381e96fb5b7b4572cd814/rpds_py-0.22.3-cp313-cp313t-macosx_10_12_x86_64.whl", hash = "sha256:f82a116a1d03628a8ace4859556fb39fd1424c933341a08ea3ed6de1edb0283b", size = 347698 },
-    { url = "https://files.pythonhosted.org/packages/a4/3a/684f66dd6b0f37499cad24cd1c0e523541fd768576fa5ce2d0a8799c3cba/rpds_py-0.22.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:3dfcbc95bd7992b16f3f7ba05af8a64ca694331bd24f9157b49dadeeb287493b", size = 337330 },
-    { url = "https://files.pythonhosted.org/packages/82/eb/e022c08c2ce2e8f7683baa313476492c0e2c1ca97227fe8a75d9f0181e95/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:59259dc58e57b10e7e18ce02c311804c10c5a793e6568f8af4dead03264584d1", size = 380022 },
-    { url = "https://files.pythonhosted.org/packages/e4/21/5a80e653e4c86aeb28eb4fea4add1f72e1787a3299687a9187105c3ee966/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5725dd9cc02068996d4438d397e255dcb1df776b7ceea3b9cb972bdb11260a83", size = 390754 },
-    { url = "https://files.pythonhosted.org/packages/37/a4/d320a04ae90f72d080b3d74597074e62be0a8ecad7d7321312dfe2dc5a6a/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:99b37292234e61325e7a5bb9689e55e48c3f5f603af88b1642666277a81f1fbd", size = 423840 },
-    { url = "https://files.pythonhosted.org/packages/87/70/674dc47d93db30a6624279284e5631be4c3a12a0340e8e4f349153546728/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:27b1d3b3915a99208fee9ab092b8184c420f2905b7d7feb4aeb5e4a9c509b8a1", size = 438970 },
-    { url = "https://files.pythonhosted.org/packages/3f/64/9500f4d66601d55cadd21e90784cfd5d5f4560e129d72e4339823129171c/rpds_py-0.22.3-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f612463ac081803f243ff13cccc648578e2279295048f2a8d5eb430af2bae6e3", size = 383146 },
-    { url = "https://files.pythonhosted.org/packages/4d/45/630327addb1d17173adcf4af01336fd0ee030c04798027dfcb50106001e0/rpds_py-0.22.3-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f73d3fef726b3243a811121de45193c0ca75f6407fe66f3f4e183c983573e130", size = 408294 },
-    { url = "https://files.pythonhosted.org/packages/5f/ef/8efb3373cee54ea9d9980b772e5690a0c9e9214045a4e7fa35046e399fee/rpds_py-0.22.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:3f21f0495edea7fdbaaa87e633a8689cd285f8f4af5c869f27bc8074638ad69c", size = 556345 },
-    { url = "https://files.pythonhosted.org/packages/54/01/151d3b9ef4925fc8f15bfb131086c12ec3c3d6dd4a4f7589c335bf8e85ba/rpds_py-0.22.3-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:1e9663daaf7a63ceccbbb8e3808fe90415b0757e2abddbfc2e06c857bf8c5e2b", size = 582292 },
-    { url = "https://files.pythonhosted.org/packages/30/89/35fc7a6cdf3477d441c7aca5e9bbf5a14e0f25152aed7f63f4e0b141045d/rpds_py-0.22.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:a76e42402542b1fae59798fab64432b2d015ab9d0c8c47ba7addddbaf7952333", size = 553855 },
-    { url = "https://files.pythonhosted.org/packages/8f/e0/830c02b2457c4bd20a8c5bb394d31d81f57fbefce2dbdd2e31feff4f7003/rpds_py-0.22.3-cp313-cp313t-win32.whl", hash = "sha256:69803198097467ee7282750acb507fba35ca22cc3b85f16cf45fb01cb9097730", size = 219100 },
-    { url = "https://files.pythonhosted.org/packages/f8/30/7ac943f69855c2db77407ae363484b915d861702dbba1aa82d68d57f42be/rpds_py-0.22.3-cp313-cp313t-win_amd64.whl", hash = "sha256:f5cf2a0c2bdadf3791b5c205d55a37a54025c6e18a71c71f82bb536cf9a454bf", size = 233794 },
-    { url = "https://files.pythonhosted.org/packages/db/0f/a8ad17ddac7c880f48d5da50733dd25bfc35ba2be1bec9f23453e8c7a123/rpds_py-0.22.3-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:378753b4a4de2a7b34063d6f95ae81bfa7b15f2c1a04a9518e8644e81807ebea", size = 359735 },
-    { url = "https://files.pythonhosted.org/packages/0c/41/430903669397ea3ee76865e0b53ea236e8dc0ffbecde47b2c4c783ad6759/rpds_py-0.22.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3445e07bf2e8ecfeef6ef67ac83de670358abf2996916039b16a218e3d95e97e", size = 348724 },
-    { url = "https://files.pythonhosted.org/packages/c9/5c/3496f4f0ee818297544f2d5f641c49dde8ae156392e6834b79c0609ba006/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7b2513ba235829860b13faa931f3b6846548021846ac808455301c23a101689d", size = 381782 },
-    { url = "https://files.pythonhosted.org/packages/b6/dc/db0523ce0cd16ce579185cc9aa9141992de956d0a9c469ecfd1fb5d54ddc/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:eaf16ae9ae519a0e237a0f528fd9f0197b9bb70f40263ee57ae53c2b8d48aeb3", size = 387036 },
-    { url = "https://files.pythonhosted.org/packages/85/2a/9525c2427d2c257f877348918136a5d4e1b945c205a256e53bec61e54551/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:583f6a1993ca3369e0f80ba99d796d8e6b1a3a2a442dd4e1a79e652116413091", size = 424566 },
-    { url = "https://files.pythonhosted.org/packages/b9/1c/f8c012a39794b84069635709f559c0309103d5d74b3f5013916e6ca4f174/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:4617e1915a539a0d9a9567795023de41a87106522ff83fbfaf1f6baf8e85437e", size = 447203 },
-    { url = "https://files.pythonhosted.org/packages/93/f5/c1c772364570d35b98ba64f36ec90c3c6d0b932bc4d8b9b4efef6dc64b07/rpds_py-0.22.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0c150c7a61ed4a4f4955a96626574e9baf1adf772c2fb61ef6a5027e52803543", size = 382283 },
-    { url = "https://files.pythonhosted.org/packages/10/06/f94f61313f94fc75c3c3aa74563f80bbd990e5b25a7c1a38cee7d5d0309b/rpds_py-0.22.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2fa4331c200c2521512595253f5bb70858b90f750d39b8cbfd67465f8d1b596d", size = 410022 },
-    { url = "https://files.pythonhosted.org/packages/3f/b0/37ab416a9528419920dfb64886c220f58fcbd66b978e0a91b66e9ee9a993/rpds_py-0.22.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:214b7a953d73b5e87f0ebece4a32a5bd83c60a3ecc9d4ec8f1dca968a2d91e99", size = 557817 },
-    { url = "https://files.pythonhosted.org/packages/2c/5d/9daa18adcd676dd3b2817c8a7cec3f3ebeeb0ce0d05a1b63bf994fc5114f/rpds_py-0.22.3-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:f47ad3d5f3258bd7058d2d506852217865afefe6153a36eb4b6928758041d831", size = 585099 },
-    { url = "https://files.pythonhosted.org/packages/41/3f/ad4e58035d3f848410aa3d59857b5f238bafab81c8b4a844281f80445d62/rpds_py-0.22.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:f276b245347e6e36526cbd4a266a417796fc531ddf391e43574cf6466c492520", size = 552818 },
-    { url = "https://files.pythonhosted.org/packages/b8/19/123acae8f4cab3c9463097c3ced3cc87c46f405056e249c874940e045309/rpds_py-0.22.3-cp39-cp39-win32.whl", hash = "sha256:bbb232860e3d03d544bc03ac57855cd82ddf19c7a07651a7c0fdb95e9efea8b9", size = 220246 },
-    { url = "https://files.pythonhosted.org/packages/8b/8d/9db93e48d96ace1f6713c71ce72e2d94b71d82156c37b6a54e0930486f00/rpds_py-0.22.3-cp39-cp39-win_amd64.whl", hash = "sha256:cfbc454a2880389dbb9b5b398e50d439e2e58669160f27b60e5eca11f68ae17c", size = 231932 },
-    { url = "https://files.pythonhosted.org/packages/8b/63/e29f8ee14fcf383574f73b6bbdcbec0fbc2e5fc36b4de44d1ac389b1de62/rpds_py-0.22.3-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:d48424e39c2611ee1b84ad0f44fb3b2b53d473e65de061e3f460fc0be5f1939d", size = 360786 },
-    { url = "https://files.pythonhosted.org/packages/d3/e0/771ee28b02a24e81c8c0e645796a371350a2bb6672753144f36ae2d2afc9/rpds_py-0.22.3-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:24e8abb5878e250f2eb0d7859a8e561846f98910326d06c0d51381fed59357bd", size = 350589 },
-    { url = "https://files.pythonhosted.org/packages/cf/49/abad4c4a1e6f3adf04785a99c247bfabe55ed868133e2d1881200aa5d381/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4b232061ca880db21fa14defe219840ad9b74b6158adb52ddf0e87bead9e8493", size = 381848 },
-    { url = "https://files.pythonhosted.org/packages/3a/7d/f4bc6d6fbe6af7a0d2b5f2ee77079efef7c8528712745659ec0026888998/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ac0a03221cdb5058ce0167ecc92a8c89e8d0decdc9e99a2ec23380793c4dcb96", size = 387879 },
-    { url = "https://files.pythonhosted.org/packages/13/b0/575c797377fdcd26cedbb00a3324232e4cb2c5d121f6e4b0dbf8468b12ef/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:eb0c341fa71df5a4595f9501df4ac5abfb5a09580081dffbd1ddd4654e6e9123", size = 423916 },
-    { url = "https://files.pythonhosted.org/packages/54/78/87157fa39d58f32a68d3326f8a81ad8fb99f49fe2aa7ad9a1b7d544f9478/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bf9db5488121b596dbfc6718c76092fda77b703c1f7533a226a5a9f65248f8ad", size = 448410 },
-    { url = "https://files.pythonhosted.org/packages/59/69/860f89996065a88be1b6ff2d60e96a02b920a262d8aadab99e7903986597/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0b8db6b5b2d4491ad5b6bdc2bc7c017eec108acbf4e6785f42a9eb0ba234f4c9", size = 382841 },
-    { url = "https://files.pythonhosted.org/packages/bd/d7/bc144e10d27e3cb350f98df2492a319edd3caaf52ddfe1293f37a9afbfd7/rpds_py-0.22.3-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:b3d504047aba448d70cf6fa22e06cb09f7cbd761939fdd47604f5e007675c24e", size = 409662 },
-    { url = "https://files.pythonhosted.org/packages/14/2a/6bed0b05233c291a94c7e89bc76ffa1c619d4e1979fbfe5d96024020c1fb/rpds_py-0.22.3-pp310-pypy310_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:e61b02c3f7a1e0b75e20c3978f7135fd13cb6cf551bf4a6d29b999a88830a338", size = 558221 },
-    { url = "https://files.pythonhosted.org/packages/11/23/cd8f566de444a137bc1ee5795e47069a947e60810ba4152886fe5308e1b7/rpds_py-0.22.3-pp310-pypy310_pp73-musllinux_1_2_i686.whl", hash = "sha256:e35ba67d65d49080e8e5a1dd40101fccdd9798adb9b050ff670b7d74fa41c566", size = 583780 },
-    { url = "https://files.pythonhosted.org/packages/8d/63/79c3602afd14d501f751e615a74a59040328da5ef29ed5754ae80d236b84/rpds_py-0.22.3-pp310-pypy310_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:26fd7cac7dd51011a245f29a2cc6489c4608b5a8ce8d75661bb4a1066c52dfbe", size = 553619 },
-    { url = "https://files.pythonhosted.org/packages/9f/2e/c5c1689e80298d4e94c75b70faada4c25445739d91b94c211244a3ed7ed1/rpds_py-0.22.3-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:177c7c0fce2855833819c98e43c262007f42ce86651ffbb84f37883308cb0e7d", size = 233338 },
-    { url = "https://files.pythonhosted.org/packages/bc/b7/d2c205723e3b4d75b03215694f0297a1b4b395bf834cb5896ad9bbb90f90/rpds_py-0.22.3-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:bb47271f60660803ad11f4c61b42242b8c1312a31c98c578f79ef9387bbde21c", size = 360594 },
-    { url = "https://files.pythonhosted.org/packages/d8/8f/c3515f5234cf6055046d4cfe9c80a3742a20acfa7d0b1b290f0d7f56a8db/rpds_py-0.22.3-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:70fb28128acbfd264eda9bf47015537ba3fe86e40d046eb2963d75024be4d055", size = 349594 },
-    { url = "https://files.pythonhosted.org/packages/6b/98/5b487cb06afc484befe350c87fda37f4ce11333f04f3380aba43dcf5bce2/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:44d61b4b7d0c2c9ac019c314e52d7cbda0ae31078aabd0f22e583af3e0d79723", size = 381138 },
-    { url = "https://files.pythonhosted.org/packages/5e/3a/12308d2c51b3fdfc173619943b7dc5ba41b4850c47112eeda38d9c54ed12/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5f0e260eaf54380380ac3808aa4ebe2d8ca28b9087cf411649f96bad6900c728", size = 387828 },
-    { url = "https://files.pythonhosted.org/packages/17/b2/c242241ab5a2a206e093f24ccbfa519c4bbf10a762ac90bffe1766c225e0/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b25bc607423935079e05619d7de556c91fb6adeae9d5f80868dde3468657994b", size = 424634 },
-    { url = "https://files.pythonhosted.org/packages/d5/c7/52a1b15012139f3ba740f291f1d03c6b632938ba61bc605f24c101952493/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fb6116dfb8d1925cbdb52595560584db42a7f664617a1f7d7f6e32f138cdf37d", size = 447862 },
-    { url = "https://files.pythonhosted.org/packages/55/3e/4d3ed8fd01bad77e8ed101116fe63b03f1011940d9596a8f4d82ac80cacd/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a63cbdd98acef6570c62b92a1e43266f9e8b21e699c363c0fef13bd530799c11", size = 382506 },
-    { url = "https://files.pythonhosted.org/packages/30/78/df59d6f92470a84369a3757abeae1cfd7f7239c8beb6d948949bf78317d2/rpds_py-0.22.3-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2b8f60e1b739a74bab7e01fcbe3dddd4657ec685caa04681df9d562ef15b625f", size = 410534 },
-    { url = "https://files.pythonhosted.org/packages/38/97/ea45d1edd9b753b20084b52dd5db6ee5e1ac3e036a27149972398a413858/rpds_py-0.22.3-pp39-pypy39_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:2e8b55d8517a2fda8d95cb45d62a5a8bbf9dd0ad39c5b25c8833efea07b880ca", size = 557453 },
-    { url = "https://files.pythonhosted.org/packages/08/cd/3a1b35eb9da27ffbb981cfffd32a01c7655c4431ccb278cb3064f8887462/rpds_py-0.22.3-pp39-pypy39_pp73-musllinux_1_2_i686.whl", hash = "sha256:2de29005e11637e7a2361fa151f780ff8eb2543a0da1413bb951e9f14b699ef3", size = 584412 },
-    { url = "https://files.pythonhosted.org/packages/87/91/31d1c5aeb1606f71188259e0ba6ed6f5c21a3c72f58b51db6a8bd0aa2b5d/rpds_py-0.22.3-pp39-pypy39_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:666ecce376999bf619756a24ce15bb14c5bfaf04bf00abc7e663ce17c3f34fe7", size = 553446 },
-    { url = "https://files.pythonhosted.org/packages/e7/ad/03b5ccd1ab492c9dece85b3bf1c96453ab8c47983936fae6880f688f60b3/rpds_py-0.22.3-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:5246b14ca64a8675e0a7161f7af68fe3e910e6b90542b4bfb5439ba752191df6", size = 233013 },
+version = "0.27.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e9/dd/2c0cbe774744272b0ae725f44032c77bdcab6e8bcf544bffa3b6e70c8dba/rpds_py-0.27.1.tar.gz", hash = "sha256:26a1c73171d10b7acccbded82bf6a586ab8203601e565badc74bbbf8bc5a10f8", size = 27479, upload-time = "2025-08-27T12:16:36.024Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a5/ed/3aef893e2dd30e77e35d20d4ddb45ca459db59cead748cad9796ad479411/rpds_py-0.27.1-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:68afeec26d42ab3b47e541b272166a0b4400313946871cba3ed3a4fc0cab1cef", size = 371606, upload-time = "2025-08-27T12:12:25.189Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/82/9818b443e5d3eb4c83c3994561387f116aae9833b35c484474769c4a8faf/rpds_py-0.27.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:74e5b2f7bb6fa38b1b10546d27acbacf2a022a8b5543efb06cfebc72a59c85be", size = 353452, upload-time = "2025-08-27T12:12:27.433Z" },
+    { url = "https://files.pythonhosted.org/packages/99/c7/d2a110ffaaa397fc6793a83c7bd3545d9ab22658b7cdff05a24a4535cc45/rpds_py-0.27.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9024de74731df54546fab0bfbcdb49fae19159ecaecfc8f37c18d2c7e2c0bd61", size = 381519, upload-time = "2025-08-27T12:12:28.719Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/bc/e89581d1f9d1be7d0247eaef602566869fdc0d084008ba139e27e775366c/rpds_py-0.27.1-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:31d3ebadefcd73b73928ed0b2fd696f7fefda8629229f81929ac9c1854d0cffb", size = 394424, upload-time = "2025-08-27T12:12:30.207Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/2e/36a6861f797530e74bb6ed53495f8741f1ef95939eed01d761e73d559067/rpds_py-0.27.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b2e7f8f169d775dd9092a1743768d771f1d1300453ddfe6325ae3ab5332b4657", size = 523467, upload-time = "2025-08-27T12:12:31.808Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/59/c1bc2be32564fa499f988f0a5c6505c2f4746ef96e58e4d7de5cf923d77e/rpds_py-0.27.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3d905d16f77eb6ab2e324e09bfa277b4c8e5e6b8a78a3e7ff8f3cdf773b4c013", size = 402660, upload-time = "2025-08-27T12:12:33.444Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/ec/ef8bf895f0628dd0a59e54d81caed6891663cb9c54a0f4bb7da918cb88cf/rpds_py-0.27.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:50c946f048209e6362e22576baea09193809f87687a95a8db24e5fbdb307b93a", size = 384062, upload-time = "2025-08-27T12:12:34.857Z" },
+    { url = "https://files.pythonhosted.org/packages/69/f7/f47ff154be8d9a5e691c083a920bba89cef88d5247c241c10b9898f595a1/rpds_py-0.27.1-cp310-cp310-manylinux_2_31_riscv64.whl", hash = "sha256:3deab27804d65cd8289eb814c2c0e807c4b9d9916c9225e363cb0cf875eb67c1", size = 401289, upload-time = "2025-08-27T12:12:36.085Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/d9/ca410363efd0615814ae579f6829cafb39225cd63e5ea5ed1404cb345293/rpds_py-0.27.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:8b61097f7488de4be8244c89915da8ed212832ccf1e7c7753a25a394bf9b1f10", size = 417718, upload-time = "2025-08-27T12:12:37.401Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/a0/8cb5c2ff38340f221cc067cc093d1270e10658ba4e8d263df923daa18e86/rpds_py-0.27.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:8a3f29aba6e2d7d90528d3c792555a93497fe6538aa65eb675b44505be747808", size = 558333, upload-time = "2025-08-27T12:12:38.672Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/8c/1b0de79177c5d5103843774ce12b84caa7164dfc6cd66378768d37db11bf/rpds_py-0.27.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:dd6cd0485b7d347304067153a6dc1d73f7d4fd995a396ef32a24d24b8ac63ac8", size = 589127, upload-time = "2025-08-27T12:12:41.48Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/5e/26abb098d5e01266b0f3a2488d299d19ccc26849735d9d2b95c39397e945/rpds_py-0.27.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:6f4461bf931108c9fa226ffb0e257c1b18dc2d44cd72b125bec50ee0ab1248a9", size = 554899, upload-time = "2025-08-27T12:12:42.925Z" },
+    { url = "https://files.pythonhosted.org/packages/de/41/905cc90ced13550db017f8f20c6d8e8470066c5738ba480d7ba63e3d136b/rpds_py-0.27.1-cp310-cp310-win32.whl", hash = "sha256:ee5422d7fb21f6a00c1901bf6559c49fee13a5159d0288320737bbf6585bd3e4", size = 217450, upload-time = "2025-08-27T12:12:44.813Z" },
+    { url = "https://files.pythonhosted.org/packages/75/3d/6bef47b0e253616ccdf67c283e25f2d16e18ccddd38f92af81d5a3420206/rpds_py-0.27.1-cp310-cp310-win_amd64.whl", hash = "sha256:3e039aabf6d5f83c745d5f9a0a381d031e9ed871967c0a5c38d201aca41f3ba1", size = 228447, upload-time = "2025-08-27T12:12:46.204Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/c1/7907329fbef97cbd49db6f7303893bd1dd5a4a3eae415839ffdfb0762cae/rpds_py-0.27.1-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:be898f271f851f68b318872ce6ebebbc62f303b654e43bf72683dbdc25b7c881", size = 371063, upload-time = "2025-08-27T12:12:47.856Z" },
+    { url = "https://files.pythonhosted.org/packages/11/94/2aab4bc86228bcf7c48760990273653a4900de89c7537ffe1b0d6097ed39/rpds_py-0.27.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:62ac3d4e3e07b58ee0ddecd71d6ce3b1637de2d373501412df395a0ec5f9beb5", size = 353210, upload-time = "2025-08-27T12:12:49.187Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/57/f5eb3ecf434342f4f1a46009530e93fd201a0b5b83379034ebdb1d7c1a58/rpds_py-0.27.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4708c5c0ceb2d034f9991623631d3d23cb16e65c83736ea020cdbe28d57c0a0e", size = 381636, upload-time = "2025-08-27T12:12:50.492Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/f4/ef95c5945e2ceb5119571b184dd5a1cc4b8541bbdf67461998cfeac9cb1e/rpds_py-0.27.1-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:abfa1171a9952d2e0002aba2ad3780820b00cc3d9c98c6630f2e93271501f66c", size = 394341, upload-time = "2025-08-27T12:12:52.024Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/7e/4bd610754bf492d398b61725eb9598ddd5eb86b07d7d9483dbcd810e20bc/rpds_py-0.27.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4b507d19f817ebaca79574b16eb2ae412e5c0835542c93fe9983f1e432aca195", size = 523428, upload-time = "2025-08-27T12:12:53.779Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/e5/059b9f65a8c9149361a8b75094864ab83b94718344db511fd6117936ed2a/rpds_py-0.27.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:168b025f8fd8d8d10957405f3fdcef3dc20f5982d398f90851f4abc58c566c52", size = 402923, upload-time = "2025-08-27T12:12:55.15Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/48/64cabb7daced2968dd08e8a1b7988bf358d7bd5bcd5dc89a652f4668543c/rpds_py-0.27.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cb56c6210ef77caa58e16e8c17d35c63fe3f5b60fd9ba9d424470c3400bcf9ed", size = 384094, upload-time = "2025-08-27T12:12:57.194Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/e1/dc9094d6ff566bff87add8a510c89b9e158ad2ecd97ee26e677da29a9e1b/rpds_py-0.27.1-cp311-cp311-manylinux_2_31_riscv64.whl", hash = "sha256:d252f2d8ca0195faa707f8eb9368955760880b2b42a8ee16d382bf5dd807f89a", size = 401093, upload-time = "2025-08-27T12:12:58.985Z" },
+    { url = "https://files.pythonhosted.org/packages/37/8e/ac8577e3ecdd5593e283d46907d7011618994e1d7ab992711ae0f78b9937/rpds_py-0.27.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:6e5e54da1e74b91dbc7996b56640f79b195d5925c2b78efaa8c5d53e1d88edde", size = 417969, upload-time = "2025-08-27T12:13:00.367Z" },
+    { url = "https://files.pythonhosted.org/packages/66/6d/87507430a8f74a93556fe55c6485ba9c259949a853ce407b1e23fea5ba31/rpds_py-0.27.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ffce0481cc6e95e5b3f0a47ee17ffbd234399e6d532f394c8dce320c3b089c21", size = 558302, upload-time = "2025-08-27T12:13:01.737Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/bb/1db4781ce1dda3eecc735e3152659a27b90a02ca62bfeea17aee45cc0fbc/rpds_py-0.27.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:a205fdfe55c90c2cd8e540ca9ceba65cbe6629b443bc05db1f590a3db8189ff9", size = 589259, upload-time = "2025-08-27T12:13:03.127Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/0e/ae1c8943d11a814d01b482e1f8da903f88047a962dff9bbdadf3bd6e6fd1/rpds_py-0.27.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:689fb5200a749db0415b092972e8eba85847c23885c8543a8b0f5c009b1a5948", size = 554983, upload-time = "2025-08-27T12:13:04.516Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/d5/0b2a55415931db4f112bdab072443ff76131b5ac4f4dc98d10d2d357eb03/rpds_py-0.27.1-cp311-cp311-win32.whl", hash = "sha256:3182af66048c00a075010bc7f4860f33913528a4b6fc09094a6e7598e462fe39", size = 217154, upload-time = "2025-08-27T12:13:06.278Z" },
+    { url = "https://files.pythonhosted.org/packages/24/75/3b7ffe0d50dc86a6a964af0d1cc3a4a2cdf437cb7b099a4747bbb96d1819/rpds_py-0.27.1-cp311-cp311-win_amd64.whl", hash = "sha256:b4938466c6b257b2f5c4ff98acd8128ec36b5059e5c8f8372d79316b1c36bb15", size = 228627, upload-time = "2025-08-27T12:13:07.625Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/3f/4fd04c32abc02c710f09a72a30c9a55ea3cc154ef8099078fd50a0596f8e/rpds_py-0.27.1-cp311-cp311-win_arm64.whl", hash = "sha256:2f57af9b4d0793e53266ee4325535a31ba48e2f875da81a9177c9926dfa60746", size = 220998, upload-time = "2025-08-27T12:13:08.972Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/fe/38de28dee5df58b8198c743fe2bea0c785c6d40941b9950bac4cdb71a014/rpds_py-0.27.1-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:ae2775c1973e3c30316892737b91f9283f9908e3cc7625b9331271eaaed7dc90", size = 361887, upload-time = "2025-08-27T12:13:10.233Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/9a/4b6c7eedc7dd90986bf0fab6ea2a091ec11c01b15f8ba0a14d3f80450468/rpds_py-0.27.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:2643400120f55c8a96f7c9d858f7be0c88d383cd4653ae2cf0d0c88f668073e5", size = 345795, upload-time = "2025-08-27T12:13:11.65Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/0e/e650e1b81922847a09cca820237b0edee69416a01268b7754d506ade11ad/rpds_py-0.27.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:16323f674c089b0360674a4abd28d5042947d54ba620f72514d69be4ff64845e", size = 385121, upload-time = "2025-08-27T12:13:13.008Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/ea/b306067a712988e2bff00dcc7c8f31d26c29b6d5931b461aa4b60a013e33/rpds_py-0.27.1-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:9a1f4814b65eacac94a00fc9a526e3fdafd78e439469644032032d0d63de4881", size = 398976, upload-time = "2025-08-27T12:13:14.368Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/0a/26dc43c8840cb8fe239fe12dbc8d8de40f2365e838f3d395835dde72f0e5/rpds_py-0.27.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7ba32c16b064267b22f1850a34051121d423b6f7338a12b9459550eb2096e7ec", size = 525953, upload-time = "2025-08-27T12:13:15.774Z" },
+    { url = "https://files.pythonhosted.org/packages/22/14/c85e8127b573aaf3a0cbd7fbb8c9c99e735a4a02180c84da2a463b766e9e/rpds_py-0.27.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e5c20f33fd10485b80f65e800bbe5f6785af510b9f4056c5a3c612ebc83ba6cb", size = 407915, upload-time = "2025-08-27T12:13:17.379Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/7b/8f4fee9ba1fb5ec856eb22d725a4efa3deb47f769597c809e03578b0f9d9/rpds_py-0.27.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:466bfe65bd932da36ff279ddd92de56b042f2266d752719beb97b08526268ec5", size = 386883, upload-time = "2025-08-27T12:13:18.704Z" },
+    { url = "https://files.pythonhosted.org/packages/86/47/28fa6d60f8b74fcdceba81b272f8d9836ac0340570f68f5df6b41838547b/rpds_py-0.27.1-cp312-cp312-manylinux_2_31_riscv64.whl", hash = "sha256:41e532bbdcb57c92ba3be62c42e9f096431b4cf478da9bc3bc6ce5c38ab7ba7a", size = 405699, upload-time = "2025-08-27T12:13:20.089Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/fd/c5987b5e054548df56953a21fe2ebed51fc1ec7c8f24fd41c067b68c4a0a/rpds_py-0.27.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f149826d742b406579466283769a8ea448eed82a789af0ed17b0cd5770433444", size = 423713, upload-time = "2025-08-27T12:13:21.436Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/ba/3c4978b54a73ed19a7d74531be37a8bcc542d917c770e14d372b8daea186/rpds_py-0.27.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:80c60cfb5310677bd67cb1e85a1e8eb52e12529545441b43e6f14d90b878775a", size = 562324, upload-time = "2025-08-27T12:13:22.789Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/6c/6943a91768fec16db09a42b08644b960cff540c66aab89b74be6d4a144ba/rpds_py-0.27.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:7ee6521b9baf06085f62ba9c7a3e5becffbc32480d2f1b351559c001c38ce4c1", size = 593646, upload-time = "2025-08-27T12:13:24.122Z" },
+    { url = "https://files.pythonhosted.org/packages/11/73/9d7a8f4be5f4396f011a6bb7a19fe26303a0dac9064462f5651ced2f572f/rpds_py-0.27.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:a512c8263249a9d68cac08b05dd59d2b3f2061d99b322813cbcc14c3c7421998", size = 558137, upload-time = "2025-08-27T12:13:25.557Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/96/6772cbfa0e2485bcceef8071de7821f81aeac8bb45fbfd5542a3e8108165/rpds_py-0.27.1-cp312-cp312-win32.whl", hash = "sha256:819064fa048ba01b6dadc5116f3ac48610435ac9a0058bbde98e569f9e785c39", size = 221343, upload-time = "2025-08-27T12:13:26.967Z" },
+    { url = "https://files.pythonhosted.org/packages/67/b6/c82f0faa9af1c6a64669f73a17ee0eeef25aff30bb9a1c318509efe45d84/rpds_py-0.27.1-cp312-cp312-win_amd64.whl", hash = "sha256:d9199717881f13c32c4046a15f024971a3b78ad4ea029e8da6b86e5aa9cf4594", size = 232497, upload-time = "2025-08-27T12:13:28.326Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/96/2817b44bd2ed11aebacc9251da03689d56109b9aba5e311297b6902136e2/rpds_py-0.27.1-cp312-cp312-win_arm64.whl", hash = "sha256:33aa65b97826a0e885ef6e278fbd934e98cdcfed80b63946025f01e2f5b29502", size = 222790, upload-time = "2025-08-27T12:13:29.71Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/6c/252e83e1ce7583c81f26d1d884b2074d40a13977e1b6c9c50bbf9a7f1f5a/rpds_py-0.27.1-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:c918c65ec2e42c2a78d19f18c553d77319119bf43aa9e2edf7fb78d624355527", size = 372140, upload-time = "2025-08-27T12:15:05.441Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/71/949c195d927c5aeb0d0629d329a20de43a64c423a6aa53836290609ef7ec/rpds_py-0.27.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1fea2b1a922c47c51fd07d656324531adc787e415c8b116530a1d29c0516c62d", size = 354086, upload-time = "2025-08-27T12:15:07.404Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/02/e43e332ad8ce4f6c4342d151a471a7f2900ed1d76901da62eb3762663a71/rpds_py-0.27.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bbf94c58e8e0cd6b6f38d8de67acae41b3a515c26169366ab58bdca4a6883bb8", size = 382117, upload-time = "2025-08-27T12:15:09.275Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/05/b0fdeb5b577197ad72812bbdfb72f9a08fa1e64539cc3940b1b781cd3596/rpds_py-0.27.1-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:c2a8fed130ce946d5c585eddc7c8eeef0051f58ac80a8ee43bd17835c144c2cc", size = 394520, upload-time = "2025-08-27T12:15:10.727Z" },
+    { url = "https://files.pythonhosted.org/packages/67/1f/4cfef98b2349a7585181e99294fa2a13f0af06902048a5d70f431a66d0b9/rpds_py-0.27.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:037a2361db72ee98d829bc2c5b7cc55598ae0a5e0ec1823a56ea99374cfd73c1", size = 522657, upload-time = "2025-08-27T12:15:12.613Z" },
+    { url = "https://files.pythonhosted.org/packages/44/55/ccf37ddc4c6dce7437b335088b5ca18da864b334890e2fe9aa6ddc3f79a9/rpds_py-0.27.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5281ed1cc1d49882f9997981c88df1a22e140ab41df19071222f7e5fc4e72125", size = 402967, upload-time = "2025-08-27T12:15:14.113Z" },
+    { url = "https://files.pythonhosted.org/packages/74/e5/5903f92e41e293b07707d5bf00ef39a0eb2af7190aff4beaf581a6591510/rpds_py-0.27.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2fd50659a069c15eef8aa3d64bbef0d69fd27bb4a50c9ab4f17f83a16cbf8905", size = 384372, upload-time = "2025-08-27T12:15:15.842Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/e3/fbb409e18aeefc01e49f5922ac63d2d914328430e295c12183ce56ebf76b/rpds_py-0.27.1-cp39-cp39-manylinux_2_31_riscv64.whl", hash = "sha256:c4b676c4ae3921649a15d28ed10025548e9b561ded473aa413af749503c6737e", size = 401264, upload-time = "2025-08-27T12:15:17.388Z" },
+    { url = "https://files.pythonhosted.org/packages/55/79/529ad07794e05cb0f38e2f965fc5bb20853d523976719400acecc447ec9d/rpds_py-0.27.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:079bc583a26db831a985c5257797b2b5d3affb0386e7ff886256762f82113b5e", size = 418691, upload-time = "2025-08-27T12:15:19.144Z" },
+    { url = "https://files.pythonhosted.org/packages/33/39/6554a7fd6d9906fda2521c6d52f5d723dca123529fb719a5b5e074c15e01/rpds_py-0.27.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:4e44099bd522cba71a2c6b97f68e19f40e7d85399de899d66cdb67b32d7cb786", size = 558989, upload-time = "2025-08-27T12:15:21.087Z" },
+    { url = "https://files.pythonhosted.org/packages/19/b2/76fa15173b6f9f445e5ef15120871b945fb8dd9044b6b8c7abe87e938416/rpds_py-0.27.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:e202e6d4188e53c6661af813b46c37ca2c45e497fc558bacc1a7630ec2695aec", size = 589835, upload-time = "2025-08-27T12:15:22.696Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/9e/5560a4b39bab780405bed8a88ee85b30178061d189558a86003548dea045/rpds_py-0.27.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:f41f814b8eaa48768d1bb551591f6ba45f87ac76899453e8ccd41dba1289b04b", size = 555227, upload-time = "2025-08-27T12:15:24.278Z" },
+    { url = "https://files.pythonhosted.org/packages/52/d7/cd9c36215111aa65724c132bf709c6f35175973e90b32115dedc4ced09cb/rpds_py-0.27.1-cp39-cp39-win32.whl", hash = "sha256:9e71f5a087ead99563c11fdaceee83ee982fd39cf67601f4fd66cb386336ee52", size = 217899, upload-time = "2025-08-27T12:15:25.926Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/e0/d75ab7b4dd8ba777f6b365adbdfc7614bbfe7c5f05703031dfa4b61c3d6c/rpds_py-0.27.1-cp39-cp39-win_amd64.whl", hash = "sha256:71108900c9c3c8590697244b9519017a400d9ba26a36c48381b3f64743a44aab", size = 228725, upload-time = "2025-08-27T12:15:27.398Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/63/b7cc415c345625d5e62f694ea356c58fb964861409008118f1245f8c3347/rpds_py-0.27.1-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:7ba22cb9693df986033b91ae1d7a979bc399237d45fccf875b76f62bb9e52ddf", size = 371360, upload-time = "2025-08-27T12:15:29.218Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/8c/12e1b24b560cf378b8ffbdb9dc73abd529e1adcfcf82727dfd29c4a7b88d/rpds_py-0.27.1-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:5b640501be9288c77738b5492b3fd3abc4ba95c50c2e41273c8a1459f08298d3", size = 353933, upload-time = "2025-08-27T12:15:30.837Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/85/1bb2210c1f7a1b99e91fea486b9f0f894aa5da3a5ec7097cbad7dec6d40f/rpds_py-0.27.1-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fb08b65b93e0c6dd70aac7f7890a9c0938d5ec71d5cb32d45cf844fb8ae47636", size = 382962, upload-time = "2025-08-27T12:15:32.348Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/c9/a839b9f219cf80ed65f27a7f5ddbb2809c1b85c966020ae2dff490e0b18e/rpds_py-0.27.1-pp310-pypy310_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:d7ff07d696a7a38152ebdb8212ca9e5baab56656749f3d6004b34ab726b550b8", size = 394412, upload-time = "2025-08-27T12:15:33.839Z" },
+    { url = "https://files.pythonhosted.org/packages/02/2d/b1d7f928b0b1f4fc2e0133e8051d199b01d7384875adc63b6ddadf3de7e5/rpds_py-0.27.1-pp310-pypy310_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:fb7c72262deae25366e3b6c0c0ba46007967aea15d1eea746e44ddba8ec58dcc", size = 523972, upload-time = "2025-08-27T12:15:35.377Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/af/2cbf56edd2d07716df1aec8a726b3159deb47cb5c27e1e42b71d705a7c2f/rpds_py-0.27.1-pp310-pypy310_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7b002cab05d6339716b03a4a3a2ce26737f6231d7b523f339fa061d53368c9d8", size = 403273, upload-time = "2025-08-27T12:15:37.051Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/93/425e32200158d44ff01da5d9612c3b6711fe69f606f06e3895511f17473b/rpds_py-0.27.1-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:23f6b69d1c26c4704fec01311963a41d7de3ee0570a84ebde4d544e5a1859ffc", size = 385278, upload-time = "2025-08-27T12:15:38.571Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/1a/1a04a915ecd0551bfa9e77b7672d1937b4b72a0fc204a17deef76001cfb2/rpds_py-0.27.1-pp310-pypy310_pp73-manylinux_2_31_riscv64.whl", hash = "sha256:530064db9146b247351f2a0250b8f00b289accea4596a033e94be2389977de71", size = 402084, upload-time = "2025-08-27T12:15:40.529Z" },
+    { url = "https://files.pythonhosted.org/packages/51/f7/66585c0fe5714368b62951d2513b684e5215beaceab2c6629549ddb15036/rpds_py-0.27.1-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:7b90b0496570bd6b0321724a330d8b545827c4df2034b6ddfc5f5275f55da2ad", size = 419041, upload-time = "2025-08-27T12:15:42.191Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/7e/83a508f6b8e219bba2d4af077c35ba0e0cdd35a751a3be6a7cba5a55ad71/rpds_py-0.27.1-pp310-pypy310_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:879b0e14a2da6a1102a3fc8af580fc1ead37e6d6692a781bd8c83da37429b5ab", size = 560084, upload-time = "2025-08-27T12:15:43.839Z" },
+    { url = "https://files.pythonhosted.org/packages/66/66/bb945683b958a1b19eb0fe715594630d0f36396ebdef4d9b89c2fa09aa56/rpds_py-0.27.1-pp310-pypy310_pp73-musllinux_1_2_i686.whl", hash = "sha256:0d807710df3b5faa66c731afa162ea29717ab3be17bdc15f90f2d9f183da4059", size = 590115, upload-time = "2025-08-27T12:15:46.647Z" },
+    { url = "https://files.pythonhosted.org/packages/12/00/ccfaafaf7db7e7adace915e5c2f2c2410e16402561801e9c7f96683002d3/rpds_py-0.27.1-pp310-pypy310_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:3adc388fc3afb6540aec081fa59e6e0d3908722771aa1e37ffe22b220a436f0b", size = 556561, upload-time = "2025-08-27T12:15:48.219Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/b7/92b6ed9aad103bfe1c45df98453dfae40969eef2cb6c6239c58d7e96f1b3/rpds_py-0.27.1-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:c796c0c1cc68cb08b0284db4229f5af76168172670c74908fdbd4b7d7f515819", size = 229125, upload-time = "2025-08-27T12:15:49.956Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/ed/e1fba02de17f4f76318b834425257c8ea297e415e12c68b4361f63e8ae92/rpds_py-0.27.1-pp311-pypy311_pp73-macosx_10_12_x86_64.whl", hash = "sha256:cdfe4bb2f9fe7458b7453ad3c33e726d6d1c7c0a72960bcc23800d77384e42df", size = 371402, upload-time = "2025-08-27T12:15:51.561Z" },
+    { url = "https://files.pythonhosted.org/packages/af/7c/e16b959b316048b55585a697e94add55a4ae0d984434d279ea83442e460d/rpds_py-0.27.1-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:8fabb8fd848a5f75a2324e4a84501ee3a5e3c78d8603f83475441866e60b94a3", size = 354084, upload-time = "2025-08-27T12:15:53.219Z" },
+    { url = "https://files.pythonhosted.org/packages/de/c1/ade645f55de76799fdd08682d51ae6724cb46f318573f18be49b1e040428/rpds_py-0.27.1-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:eda8719d598f2f7f3e0f885cba8646644b55a187762bec091fa14a2b819746a9", size = 383090, upload-time = "2025-08-27T12:15:55.158Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/27/89070ca9b856e52960da1472efcb6c20ba27cfe902f4f23ed095b9cfc61d/rpds_py-0.27.1-pp311-pypy311_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:3c64d07e95606ec402a0a1c511fe003873fa6af630bda59bac77fac8b4318ebc", size = 394519, upload-time = "2025-08-27T12:15:57.238Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/28/be120586874ef906aa5aeeae95ae8df4184bc757e5b6bd1c729ccff45ed5/rpds_py-0.27.1-pp311-pypy311_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:93a2ed40de81bcff59aabebb626562d48332f3d028ca2036f1d23cbb52750be4", size = 523817, upload-time = "2025-08-27T12:15:59.237Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/ef/70cc197bc11cfcde02a86f36ac1eed15c56667c2ebddbdb76a47e90306da/rpds_py-0.27.1-pp311-pypy311_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:387ce8c44ae94e0ec50532d9cb0edce17311024c9794eb196b90e1058aadeb66", size = 403240, upload-time = "2025-08-27T12:16:00.923Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/35/46936cca449f7f518f2f4996e0e8344db4b57e2081e752441154089d2a5f/rpds_py-0.27.1-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:aaf94f812c95b5e60ebaf8bfb1898a7d7cb9c1af5744d4a67fa47796e0465d4e", size = 385194, upload-time = "2025-08-27T12:16:02.802Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/62/29c0d3e5125c3270b51415af7cbff1ec587379c84f55a5761cc9efa8cd06/rpds_py-0.27.1-pp311-pypy311_pp73-manylinux_2_31_riscv64.whl", hash = "sha256:4848ca84d6ded9b58e474dfdbad4b8bfb450344c0551ddc8d958bf4b36aa837c", size = 402086, upload-time = "2025-08-27T12:16:04.806Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/66/03e1087679227785474466fdd04157fb793b3b76e3fcf01cbf4c693c1949/rpds_py-0.27.1-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2bde09cbcf2248b73c7c323be49b280180ff39fadcfe04e7b6f54a678d02a7cf", size = 419272, upload-time = "2025-08-27T12:16:06.471Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/24/e3e72d265121e00b063aef3e3501e5b2473cf1b23511d56e529531acf01e/rpds_py-0.27.1-pp311-pypy311_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:94c44ee01fd21c9058f124d2d4f0c9dc7634bec93cd4b38eefc385dabe71acbf", size = 560003, upload-time = "2025-08-27T12:16:08.06Z" },
+    { url = "https://files.pythonhosted.org/packages/26/ca/f5a344c534214cc2d41118c0699fffbdc2c1bc7046f2a2b9609765ab9c92/rpds_py-0.27.1-pp311-pypy311_pp73-musllinux_1_2_i686.whl", hash = "sha256:df8b74962e35c9249425d90144e721eed198e6555a0e22a563d29fe4486b51f6", size = 590482, upload-time = "2025-08-27T12:16:10.137Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/08/4349bdd5c64d9d193c360aa9db89adeee6f6682ab8825dca0a3f535f434f/rpds_py-0.27.1-pp311-pypy311_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:dc23e6820e3b40847e2f4a7726462ba0cf53089512abe9ee16318c366494c17a", size = 556523, upload-time = "2025-08-27T12:16:12.188Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/ea/5463cd5048a7a2fcdae308b6e96432802132c141bfb9420260142632a0f1/rpds_py-0.27.1-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:aa8933159edc50be265ed22b401125c9eebff3171f570258854dbce3ecd55475", size = 371778, upload-time = "2025-08-27T12:16:13.851Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/c8/f38c099db07f5114029c1467649d308543906933eebbc226d4527a5f4693/rpds_py-0.27.1-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:a50431bf02583e21bf273c71b89d710e7a710ad5e39c725b14e685610555926f", size = 354394, upload-time = "2025-08-27T12:16:15.609Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/79/b76f97704d9dd8ddbd76fed4c4048153a847c5d6003afe20a6b5c3339065/rpds_py-0.27.1-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:78af06ddc7fe5cc0e967085a9115accee665fb912c22a3f54bad70cc65b05fe6", size = 382348, upload-time = "2025-08-27T12:16:17.251Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/3f/ef23d3c1be1b837b648a3016d5bbe7cfe711422ad110b4081c0a90ef5a53/rpds_py-0.27.1-pp39-pypy39_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:70d0738ef8fee13c003b100c2fbd667ec4f133468109b3472d249231108283a3", size = 394159, upload-time = "2025-08-27T12:16:19.251Z" },
+    { url = "https://files.pythonhosted.org/packages/74/8a/9e62693af1a34fd28b1a190d463d12407bd7cf561748cb4745845d9548d3/rpds_py-0.27.1-pp39-pypy39_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e2f6fd8a1cea5bbe599b6e78a6e5ee08db434fc8ffea51ff201c8765679698b3", size = 522775, upload-time = "2025-08-27T12:16:20.929Z" },
+    { url = "https://files.pythonhosted.org/packages/36/0d/8d5bb122bf7a60976b54c5c99a739a3819f49f02d69df3ea2ca2aff47d5c/rpds_py-0.27.1-pp39-pypy39_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8177002868d1426305bb5de1e138161c2ec9eb2d939be38291d7c431c4712df8", size = 402633, upload-time = "2025-08-27T12:16:22.548Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/0e/237948c1f425e23e0cf5a566d702652a6e55c6f8fbd332a1792eb7043daf/rpds_py-0.27.1-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:008b839781d6c9bf3b6a8984d1d8e56f0ec46dc56df61fd669c49b58ae800400", size = 384867, upload-time = "2025-08-27T12:16:24.29Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/0a/da0813efcd998d260cbe876d97f55b0f469ada8ba9cbc47490a132554540/rpds_py-0.27.1-pp39-pypy39_pp73-manylinux_2_31_riscv64.whl", hash = "sha256:a55b9132bb1ade6c734ddd2759c8dc132aa63687d259e725221f106b83a0e485", size = 401791, upload-time = "2025-08-27T12:16:25.954Z" },
+    { url = "https://files.pythonhosted.org/packages/51/78/c6c9e8a8aaca416a6f0d1b6b4a6ee35b88fe2c5401d02235d0a056eceed2/rpds_py-0.27.1-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:a46fdec0083a26415f11d5f236b79fa1291c32aaa4a17684d82f7017a1f818b1", size = 419525, upload-time = "2025-08-27T12:16:27.659Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/69/5af37e1d71487cf6d56dd1420dc7e0c2732c1b6ff612aa7a88374061c0a8/rpds_py-0.27.1-pp39-pypy39_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:8a63b640a7845f2bdd232eb0d0a4a2dd939bcdd6c57e6bb134526487f3160ec5", size = 559255, upload-time = "2025-08-27T12:16:29.343Z" },
+    { url = "https://files.pythonhosted.org/packages/40/7f/8b7b136069ef7ac3960eda25d832639bdb163018a34c960ed042dd1707c8/rpds_py-0.27.1-pp39-pypy39_pp73-musllinux_1_2_i686.whl", hash = "sha256:7e32721e5d4922deaaf963469d795d5bde6093207c52fec719bd22e5d1bedbc4", size = 590384, upload-time = "2025-08-27T12:16:31.005Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/06/c316d3f6ff03f43ccb0eba7de61376f8ec4ea850067dddfafe98274ae13c/rpds_py-0.27.1-pp39-pypy39_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:2c426b99a068601b5f4623573df7a7c3d72e87533a2dd2253353a03e7502566c", size = 555959, upload-time = "2025-08-27T12:16:32.73Z" },
+    { url = "https://files.pythonhosted.org/packages/60/94/384cf54c430b9dac742bbd2ec26c23feb78ded0d43d6d78563a281aec017/rpds_py-0.27.1-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:4fc9b7fe29478824361ead6e14e4f5aed570d477e06088826537e202d25fe859", size = 228784, upload-time = "2025-08-27T12:16:34.428Z" },
+]
+
+[[package]]
+name = "rsa"
+version = "4.9.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pyasn1" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/da/8a/22b7beea3ee0d44b1916c0c1cb0ee3af23b700b6da9f04991899d0c555d4/rsa-4.9.1.tar.gz", hash = "sha256:e7bdbfdb5497da4c07dfd35530e1a902659db6ff241e39d9953cad06ebd0ae75", size = 29034, upload-time = "2025-04-16T09:51:18.218Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/64/8d/0133e4eb4beed9e425d9a98ed6e081a55d195481b7632472be1af08d2f6b/rsa-4.9.1-py3-none-any.whl", hash = "sha256:68635866661c6836b8d39430f97a996acbd61bfa49406748ea243539fe239762", size = 34696, upload-time = "2025-04-16T09:51:17.142Z" },
 ]
 
 [[package]]
 name = "safetensors"
-version = "0.5.2"
+version = "0.6.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f4/4f/2ef9ef1766f8c194b01b67a63a444d2e557c8fe1d82faf3ebd85f370a917/safetensors-0.5.2.tar.gz", hash = "sha256:cb4a8d98ba12fa016f4241932b1fc5e702e5143f5374bba0bbcf7ddc1c4cf2b8", size = 66957 }
+sdist = { url = "https://files.pythonhosted.org/packages/ac/cc/738f3011628920e027a11754d9cae9abec1aed00f7ae860abbf843755233/safetensors-0.6.2.tar.gz", hash = "sha256:43ff2aa0e6fa2dc3ea5524ac7ad93a9839256b8703761e76e2d0b2a3fa4f15d9", size = 197968, upload-time = "2025-08-08T13:13:58.654Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/96/d1/017e31e75e274492a11a456a9e7c171f8f7911fe50735b4ec6ff37221220/safetensors-0.5.2-cp38-abi3-macosx_10_12_x86_64.whl", hash = "sha256:45b6092997ceb8aa3801693781a71a99909ab9cc776fbc3fa9322d29b1d3bef2", size = 427067 },
-    { url = "https://files.pythonhosted.org/packages/24/84/e9d3ff57ae50dd0028f301c9ee064e5087fe8b00e55696677a0413c377a7/safetensors-0.5.2-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:6d0d6a8ee2215a440e1296b843edf44fd377b055ba350eaba74655a2fe2c4bae", size = 408856 },
-    { url = "https://files.pythonhosted.org/packages/f1/1d/fe95f5dd73db16757b11915e8a5106337663182d0381811c81993e0014a9/safetensors-0.5.2-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:86016d40bcaa3bcc9a56cd74d97e654b5f4f4abe42b038c71e4f00a089c4526c", size = 450088 },
-    { url = "https://files.pythonhosted.org/packages/cf/21/e527961b12d5ab528c6e47b92d5f57f33563c28a972750b238b871924e49/safetensors-0.5.2-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:990833f70a5f9c7d3fc82c94507f03179930ff7d00941c287f73b6fcbf67f19e", size = 458966 },
-    { url = "https://files.pythonhosted.org/packages/a5/8b/1a037d7a57f86837c0b41905040369aea7d8ca1ec4b2a77592372b2ec380/safetensors-0.5.2-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3dfa7c2f3fe55db34eba90c29df94bcdac4821043fc391cb5d082d9922013869", size = 509915 },
-    { url = "https://files.pythonhosted.org/packages/61/3d/03dd5cfd33839df0ee3f4581a20bd09c40246d169c0e4518f20b21d5f077/safetensors-0.5.2-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:46ff2116150ae70a4e9c490d2ab6b6e1b1b93f25e520e540abe1b81b48560c3a", size = 527664 },
-    { url = "https://files.pythonhosted.org/packages/c5/dc/8952caafa9a10a3c0f40fa86bacf3190ae7f55fa5eef87415b97b29cb97f/safetensors-0.5.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3ab696dfdc060caffb61dbe4066b86419107a24c804a4e373ba59be699ebd8d5", size = 461978 },
-    { url = "https://files.pythonhosted.org/packages/60/da/82de1fcf1194e3dbefd4faa92dc98b33c06bed5d67890e0962dd98e18287/safetensors-0.5.2-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:03c937100f38c9ff4c1507abea9928a6a9b02c9c1c9c3609ed4fb2bf413d4975", size = 491253 },
-    { url = "https://files.pythonhosted.org/packages/5a/9a/d90e273c25f90c3ba1b0196a972003786f04c39e302fbd6649325b1272bb/safetensors-0.5.2-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:a00e737948791b94dad83cf0eafc09a02c4d8c2171a239e8c8572fe04e25960e", size = 628644 },
-    { url = "https://files.pythonhosted.org/packages/70/3c/acb23e05aa34b4f5edd2e7f393f8e6480fbccd10601ab42cd03a57d4ab5f/safetensors-0.5.2-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:d3a06fae62418ec8e5c635b61a8086032c9e281f16c63c3af46a6efbab33156f", size = 721648 },
-    { url = "https://files.pythonhosted.org/packages/71/45/eaa3dba5253a7c6931230dc961641455710ab231f8a89cb3c4c2af70f8c8/safetensors-0.5.2-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:1506e4c2eda1431099cebe9abf6c76853e95d0b7a95addceaa74c6019c65d8cf", size = 659588 },
-    { url = "https://files.pythonhosted.org/packages/b0/71/2f9851164f821064d43b481ddbea0149c2d676c4f4e077b178e7eeaa6660/safetensors-0.5.2-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:5c5b5d9da594f638a259fca766046f44c97244cc7ab8bef161b3e80d04becc76", size = 632533 },
-    { url = "https://files.pythonhosted.org/packages/00/f1/5680e2ef61d9c61454fad82c344f0e40b8741a9dbd1e31484f0d31a9b1c3/safetensors-0.5.2-cp38-abi3-win32.whl", hash = "sha256:fe55c039d97090d1f85277d402954dd6ad27f63034fa81985a9cc59655ac3ee2", size = 291167 },
-    { url = "https://files.pythonhosted.org/packages/86/ca/aa489392ec6fb59223ffce825461e1f811a3affd417121a2088be7a5758b/safetensors-0.5.2-cp38-abi3-win_amd64.whl", hash = "sha256:78abdddd03a406646107f973c7843276e7b64e5e32623529dc17f3d94a20f589", size = 303756 },
+    { url = "https://files.pythonhosted.org/packages/4d/b1/3f5fd73c039fc87dba3ff8b5d528bfc5a32b597fea8e7a6a4800343a17c7/safetensors-0.6.2-cp38-abi3-macosx_10_12_x86_64.whl", hash = "sha256:9c85ede8ec58f120bad982ec47746981e210492a6db876882aa021446af8ffba", size = 454797, upload-time = "2025-08-08T13:13:52.066Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/c9/bb114c158540ee17907ec470d01980957fdaf87b4aa07914c24eba87b9c6/safetensors-0.6.2-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:d6675cf4b39c98dbd7d940598028f3742e0375a6b4d4277e76beb0c35f4b843b", size = 432206, upload-time = "2025-08-08T13:13:50.931Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/8e/f70c34e47df3110e8e0bb268d90db8d4be8958a54ab0336c9be4fe86dac8/safetensors-0.6.2-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1d2d2b3ce1e2509c68932ca03ab8f20570920cd9754b05063d4368ee52833ecd", size = 473261, upload-time = "2025-08-08T13:13:41.259Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/f5/be9c6a7c7ef773e1996dc214e73485286df1836dbd063e8085ee1976f9cb/safetensors-0.6.2-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:93de35a18f46b0f5a6a1f9e26d91b442094f2df02e9fd7acf224cfec4238821a", size = 485117, upload-time = "2025-08-08T13:13:43.506Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/55/23f2d0a2c96ed8665bf17a30ab4ce5270413f4d74b6d87dd663258b9af31/safetensors-0.6.2-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:89a89b505f335640f9120fac65ddeb83e40f1fd081cb8ed88b505bdccec8d0a1", size = 616154, upload-time = "2025-08-08T13:13:45.096Z" },
+    { url = "https://files.pythonhosted.org/packages/98/c6/affb0bd9ce02aa46e7acddbe087912a04d953d7a4d74b708c91b5806ef3f/safetensors-0.6.2-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fc4d0d0b937e04bdf2ae6f70cd3ad51328635fe0e6214aa1fc811f3b576b3bda", size = 520713, upload-time = "2025-08-08T13:13:46.25Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/5d/5a514d7b88e310c8b146e2404e0dc161282e78634d9358975fd56dfd14be/safetensors-0.6.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8045db2c872db8f4cbe3faa0495932d89c38c899c603f21e9b6486951a5ecb8f", size = 485835, upload-time = "2025-08-08T13:13:49.373Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/7b/4fc3b2ba62c352b2071bea9cfbad330fadda70579f617506ae1a2f129cab/safetensors-0.6.2-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:81e67e8bab9878bb568cffbc5f5e655adb38d2418351dc0859ccac158f753e19", size = 521503, upload-time = "2025-08-08T13:13:47.651Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/50/0057e11fe1f3cead9254315a6c106a16dd4b1a19cd247f7cc6414f6b7866/safetensors-0.6.2-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:b0e4d029ab0a0e0e4fdf142b194514695b1d7d3735503ba700cf36d0fc7136ce", size = 652256, upload-time = "2025-08-08T13:13:53.167Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/29/473f789e4ac242593ac1656fbece6e1ecd860bb289e635e963667807afe3/safetensors-0.6.2-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:fa48268185c52bfe8771e46325a1e21d317207bcabcb72e65c6e28e9ffeb29c7", size = 747281, upload-time = "2025-08-08T13:13:54.656Z" },
+    { url = "https://files.pythonhosted.org/packages/68/52/f7324aad7f2df99e05525c84d352dc217e0fa637a4f603e9f2eedfbe2c67/safetensors-0.6.2-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:d83c20c12c2d2f465997c51b7ecb00e407e5f94d7dec3ea0cc11d86f60d3fde5", size = 692286, upload-time = "2025-08-08T13:13:55.884Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/fe/cad1d9762868c7c5dc70c8620074df28ebb1a8e4c17d4c0cb031889c457e/safetensors-0.6.2-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:d944cea65fad0ead848b6ec2c37cc0b197194bec228f8020054742190e9312ac", size = 655957, upload-time = "2025-08-08T13:13:57.029Z" },
+    { url = "https://files.pythonhosted.org/packages/59/a7/e2158e17bbe57d104f0abbd95dff60dda916cf277c9f9663b4bf9bad8b6e/safetensors-0.6.2-cp38-abi3-win32.whl", hash = "sha256:cab75ca7c064d3911411461151cb69380c9225798a20e712b102edda2542ddb1", size = 308926, upload-time = "2025-08-08T13:14:01.095Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/c3/c0be1135726618dc1e28d181b8c442403d8dbb9e273fd791de2d4384bcdd/safetensors-0.6.2-cp38-abi3-win_amd64.whl", hash = "sha256:c7b214870df923cbc1593c3faee16bec59ea462758699bd3fee399d00aac072c", size = 320192, upload-time = "2025-08-08T13:13:59.467Z" },
 ]
 
 [[package]]
@@ -3006,365 +3790,618 @@ name = "scipy"
 version = "1.13.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.9.17' and python_full_version < '3.10'",
-    "python_full_version < '3.9.17'",
-]
-dependencies = [
-    { name = "numpy", marker = "python_full_version < '3.10'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/ae/00/48c2f661e2816ccf2ecd77982f6605b2950afe60f60a52b4cbbc2504aa8f/scipy-1.13.1.tar.gz", hash = "sha256:095a87a0312b08dfd6a6155cbbd310a8c51800fc931b8c0b84003014b874ed3c", size = 57210720 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/33/59/41b2529908c002ade869623b87eecff3e11e3ce62e996d0bdcb536984187/scipy-1.13.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:20335853b85e9a49ff7572ab453794298bcf0354d8068c5f6775a0eabf350aca", size = 39328076 },
-    { url = "https://files.pythonhosted.org/packages/d5/33/f1307601f492f764062ce7dd471a14750f3360e33cd0f8c614dae208492c/scipy-1.13.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:d605e9c23906d1994f55ace80e0125c587f96c020037ea6aa98d01b4bd2e222f", size = 30306232 },
-    { url = "https://files.pythonhosted.org/packages/c0/66/9cd4f501dd5ea03e4a4572ecd874936d0da296bd04d1c45ae1a4a75d9c3a/scipy-1.13.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cfa31f1def5c819b19ecc3a8b52d28ffdcc7ed52bb20c9a7589669dd3c250989", size = 33743202 },
-    { url = "https://files.pythonhosted.org/packages/a3/ba/7255e5dc82a65adbe83771c72f384d99c43063648456796436c9a5585ec3/scipy-1.13.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f26264b282b9da0952a024ae34710c2aff7d27480ee91a2e82b7b7073c24722f", size = 38577335 },
-    { url = "https://files.pythonhosted.org/packages/49/a5/bb9ded8326e9f0cdfdc412eeda1054b914dfea952bda2097d174f8832cc0/scipy-1.13.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:eccfa1906eacc02de42d70ef4aecea45415f5be17e72b61bafcfd329bdc52e94", size = 38820728 },
-    { url = "https://files.pythonhosted.org/packages/12/30/df7a8fcc08f9b4a83f5f27cfaaa7d43f9a2d2ad0b6562cced433e5b04e31/scipy-1.13.1-cp310-cp310-win_amd64.whl", hash = "sha256:2831f0dc9c5ea9edd6e51e6e769b655f08ec6db6e2e10f86ef39bd32eb11da54", size = 46210588 },
-    { url = "https://files.pythonhosted.org/packages/b4/15/4a4bb1b15bbd2cd2786c4f46e76b871b28799b67891f23f455323a0cdcfb/scipy-1.13.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:27e52b09c0d3a1d5b63e1105f24177e544a222b43611aaf5bc44d4a0979e32f9", size = 39333805 },
-    { url = "https://files.pythonhosted.org/packages/ba/92/42476de1af309c27710004f5cdebc27bec62c204db42e05b23a302cb0c9a/scipy-1.13.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:54f430b00f0133e2224c3ba42b805bfd0086fe488835effa33fa291561932326", size = 30317687 },
-    { url = "https://files.pythonhosted.org/packages/80/ba/8be64fe225360a4beb6840f3cbee494c107c0887f33350d0a47d55400b01/scipy-1.13.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e89369d27f9e7b0884ae559a3a956e77c02114cc60a6058b4e5011572eea9299", size = 33694638 },
-    { url = "https://files.pythonhosted.org/packages/36/07/035d22ff9795129c5a847c64cb43c1fa9188826b59344fee28a3ab02e283/scipy-1.13.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a78b4b3345f1b6f68a763c6e25c0c9a23a9fd0f39f5f3d200efe8feda560a5fa", size = 38569931 },
-    { url = "https://files.pythonhosted.org/packages/d9/10/f9b43de37e5ed91facc0cfff31d45ed0104f359e4f9a68416cbf4e790241/scipy-1.13.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:45484bee6d65633752c490404513b9ef02475b4284c4cfab0ef946def50b3f59", size = 38838145 },
-    { url = "https://files.pythonhosted.org/packages/4a/48/4513a1a5623a23e95f94abd675ed91cfb19989c58e9f6f7d03990f6caf3d/scipy-1.13.1-cp311-cp311-win_amd64.whl", hash = "sha256:5713f62f781eebd8d597eb3f88b8bf9274e79eeabf63afb4a737abc6c84ad37b", size = 46196227 },
-    { url = "https://files.pythonhosted.org/packages/f2/7b/fb6b46fbee30fc7051913068758414f2721003a89dd9a707ad49174e3843/scipy-1.13.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:5d72782f39716b2b3509cd7c33cdc08c96f2f4d2b06d51e52fb45a19ca0c86a1", size = 39357301 },
-    { url = "https://files.pythonhosted.org/packages/dc/5a/2043a3bde1443d94014aaa41e0b50c39d046dda8360abd3b2a1d3f79907d/scipy-1.13.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:017367484ce5498445aade74b1d5ab377acdc65e27095155e448c88497755a5d", size = 30363348 },
-    { url = "https://files.pythonhosted.org/packages/e7/cb/26e4a47364bbfdb3b7fb3363be6d8a1c543bcd70a7753ab397350f5f189a/scipy-1.13.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:949ae67db5fa78a86e8fa644b9a6b07252f449dcf74247108c50e1d20d2b4627", size = 33406062 },
-    { url = "https://files.pythonhosted.org/packages/88/ab/6ecdc526d509d33814835447bbbeedbebdec7cca46ef495a61b00a35b4bf/scipy-1.13.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de3ade0e53bc1f21358aa74ff4830235d716211d7d077e340c7349bc3542e884", size = 38218311 },
-    { url = "https://files.pythonhosted.org/packages/0b/00/9f54554f0f8318100a71515122d8f4f503b1a2c4b4cfab3b4b68c0eb08fa/scipy-1.13.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:2ac65fb503dad64218c228e2dc2d0a0193f7904747db43014645ae139c8fad16", size = 38442493 },
-    { url = "https://files.pythonhosted.org/packages/3e/df/963384e90733e08eac978cd103c34df181d1fec424de383cdc443f418dd4/scipy-1.13.1-cp312-cp312-win_amd64.whl", hash = "sha256:cdd7dacfb95fea358916410ec61bbc20440f7860333aee6d882bb8046264e949", size = 45910955 },
-    { url = "https://files.pythonhosted.org/packages/7f/29/c2ea58c9731b9ecb30b6738113a95d147e83922986b34c685b8f6eefde21/scipy-1.13.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:436bbb42a94a8aeef855d755ce5a465479c721e9d684de76bf61a62e7c2b81d5", size = 39352927 },
-    { url = "https://files.pythonhosted.org/packages/5c/c0/e71b94b20ccf9effb38d7147c0064c08c622309fd487b1b677771a97d18c/scipy-1.13.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:8335549ebbca860c52bf3d02f80784e91a004b71b059e3eea9678ba994796a24", size = 30324538 },
-    { url = "https://files.pythonhosted.org/packages/6d/0f/aaa55b06d474817cea311e7b10aab2ea1fd5d43bc6a2861ccc9caec9f418/scipy-1.13.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d533654b7d221a6a97304ab63c41c96473ff04459e404b83275b60aa8f4b7004", size = 33732190 },
-    { url = "https://files.pythonhosted.org/packages/35/f5/d0ad1a96f80962ba65e2ce1de6a1e59edecd1f0a7b55990ed208848012e0/scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:637e98dcf185ba7f8e663e122ebf908c4702420477ae52a04f9908707456ba4d", size = 38612244 },
-    { url = "https://files.pythonhosted.org/packages/8d/02/1165905f14962174e6569076bcc3315809ae1291ed14de6448cc151eedfd/scipy-1.13.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a014c2b3697bde71724244f63de2476925596c24285c7a637364761f8710891c", size = 38845637 },
-    { url = "https://files.pythonhosted.org/packages/3e/77/dab54fe647a08ee4253963bcd8f9cf17509c8ca64d6335141422fe2e2114/scipy-1.13.1-cp39-cp39-win_amd64.whl", hash = "sha256:392e4ec766654852c25ebad4f64e4e584cf19820b980bc04960bca0b0cd6eaa2", size = 46227440 },
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ae/00/48c2f661e2816ccf2ecd77982f6605b2950afe60f60a52b4cbbc2504aa8f/scipy-1.13.1.tar.gz", hash = "sha256:095a87a0312b08dfd6a6155cbbd310a8c51800fc931b8c0b84003014b874ed3c", size = 57210720, upload-time = "2024-05-23T03:29:26.079Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/33/59/41b2529908c002ade869623b87eecff3e11e3ce62e996d0bdcb536984187/scipy-1.13.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:20335853b85e9a49ff7572ab453794298bcf0354d8068c5f6775a0eabf350aca", size = 39328076, upload-time = "2024-05-23T03:19:01.687Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/33/f1307601f492f764062ce7dd471a14750f3360e33cd0f8c614dae208492c/scipy-1.13.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:d605e9c23906d1994f55ace80e0125c587f96c020037ea6aa98d01b4bd2e222f", size = 30306232, upload-time = "2024-05-23T03:19:09.089Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/66/9cd4f501dd5ea03e4a4572ecd874936d0da296bd04d1c45ae1a4a75d9c3a/scipy-1.13.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cfa31f1def5c819b19ecc3a8b52d28ffdcc7ed52bb20c9a7589669dd3c250989", size = 33743202, upload-time = "2024-05-23T03:19:15.138Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/ba/7255e5dc82a65adbe83771c72f384d99c43063648456796436c9a5585ec3/scipy-1.13.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f26264b282b9da0952a024ae34710c2aff7d27480ee91a2e82b7b7073c24722f", size = 38577335, upload-time = "2024-05-23T03:19:21.984Z" },
+    { url = "https://files.pythonhosted.org/packages/49/a5/bb9ded8326e9f0cdfdc412eeda1054b914dfea952bda2097d174f8832cc0/scipy-1.13.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:eccfa1906eacc02de42d70ef4aecea45415f5be17e72b61bafcfd329bdc52e94", size = 38820728, upload-time = "2024-05-23T03:19:28.225Z" },
+    { url = "https://files.pythonhosted.org/packages/12/30/df7a8fcc08f9b4a83f5f27cfaaa7d43f9a2d2ad0b6562cced433e5b04e31/scipy-1.13.1-cp310-cp310-win_amd64.whl", hash = "sha256:2831f0dc9c5ea9edd6e51e6e769b655f08ec6db6e2e10f86ef39bd32eb11da54", size = 46210588, upload-time = "2024-05-23T03:19:35.661Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/15/4a4bb1b15bbd2cd2786c4f46e76b871b28799b67891f23f455323a0cdcfb/scipy-1.13.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:27e52b09c0d3a1d5b63e1105f24177e544a222b43611aaf5bc44d4a0979e32f9", size = 39333805, upload-time = "2024-05-23T03:19:43.081Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/92/42476de1af309c27710004f5cdebc27bec62c204db42e05b23a302cb0c9a/scipy-1.13.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:54f430b00f0133e2224c3ba42b805bfd0086fe488835effa33fa291561932326", size = 30317687, upload-time = "2024-05-23T03:19:48.799Z" },
+    { url = "https://files.pythonhosted.org/packages/80/ba/8be64fe225360a4beb6840f3cbee494c107c0887f33350d0a47d55400b01/scipy-1.13.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e89369d27f9e7b0884ae559a3a956e77c02114cc60a6058b4e5011572eea9299", size = 33694638, upload-time = "2024-05-23T03:19:55.104Z" },
+    { url = "https://files.pythonhosted.org/packages/36/07/035d22ff9795129c5a847c64cb43c1fa9188826b59344fee28a3ab02e283/scipy-1.13.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a78b4b3345f1b6f68a763c6e25c0c9a23a9fd0f39f5f3d200efe8feda560a5fa", size = 38569931, upload-time = "2024-05-23T03:20:01.82Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/10/f9b43de37e5ed91facc0cfff31d45ed0104f359e4f9a68416cbf4e790241/scipy-1.13.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:45484bee6d65633752c490404513b9ef02475b4284c4cfab0ef946def50b3f59", size = 38838145, upload-time = "2024-05-23T03:20:09.173Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/48/4513a1a5623a23e95f94abd675ed91cfb19989c58e9f6f7d03990f6caf3d/scipy-1.13.1-cp311-cp311-win_amd64.whl", hash = "sha256:5713f62f781eebd8d597eb3f88b8bf9274e79eeabf63afb4a737abc6c84ad37b", size = 46196227, upload-time = "2024-05-23T03:20:16.433Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/7b/fb6b46fbee30fc7051913068758414f2721003a89dd9a707ad49174e3843/scipy-1.13.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:5d72782f39716b2b3509cd7c33cdc08c96f2f4d2b06d51e52fb45a19ca0c86a1", size = 39357301, upload-time = "2024-05-23T03:20:23.538Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/5a/2043a3bde1443d94014aaa41e0b50c39d046dda8360abd3b2a1d3f79907d/scipy-1.13.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:017367484ce5498445aade74b1d5ab377acdc65e27095155e448c88497755a5d", size = 30363348, upload-time = "2024-05-23T03:20:29.885Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/cb/26e4a47364bbfdb3b7fb3363be6d8a1c543bcd70a7753ab397350f5f189a/scipy-1.13.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:949ae67db5fa78a86e8fa644b9a6b07252f449dcf74247108c50e1d20d2b4627", size = 33406062, upload-time = "2024-05-23T03:20:36.012Z" },
+    { url = "https://files.pythonhosted.org/packages/88/ab/6ecdc526d509d33814835447bbbeedbebdec7cca46ef495a61b00a35b4bf/scipy-1.13.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:de3ade0e53bc1f21358aa74ff4830235d716211d7d077e340c7349bc3542e884", size = 38218311, upload-time = "2024-05-23T03:20:42.086Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/00/9f54554f0f8318100a71515122d8f4f503b1a2c4b4cfab3b4b68c0eb08fa/scipy-1.13.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:2ac65fb503dad64218c228e2dc2d0a0193f7904747db43014645ae139c8fad16", size = 38442493, upload-time = "2024-05-23T03:20:48.292Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/df/963384e90733e08eac978cd103c34df181d1fec424de383cdc443f418dd4/scipy-1.13.1-cp312-cp312-win_amd64.whl", hash = "sha256:cdd7dacfb95fea358916410ec61bbc20440f7860333aee6d882bb8046264e949", size = 45910955, upload-time = "2024-05-23T03:20:55.091Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/29/c2ea58c9731b9ecb30b6738113a95d147e83922986b34c685b8f6eefde21/scipy-1.13.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:436bbb42a94a8aeef855d755ce5a465479c721e9d684de76bf61a62e7c2b81d5", size = 39352927, upload-time = "2024-05-23T03:21:01.95Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/c0/e71b94b20ccf9effb38d7147c0064c08c622309fd487b1b677771a97d18c/scipy-1.13.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:8335549ebbca860c52bf3d02f80784e91a004b71b059e3eea9678ba994796a24", size = 30324538, upload-time = "2024-05-23T03:21:07.634Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/0f/aaa55b06d474817cea311e7b10aab2ea1fd5d43bc6a2861ccc9caec9f418/scipy-1.13.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d533654b7d221a6a97304ab63c41c96473ff04459e404b83275b60aa8f4b7004", size = 33732190, upload-time = "2024-05-23T03:21:14.41Z" },
+    { url = "https://files.pythonhosted.org/packages/35/f5/d0ad1a96f80962ba65e2ce1de6a1e59edecd1f0a7b55990ed208848012e0/scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:637e98dcf185ba7f8e663e122ebf908c4702420477ae52a04f9908707456ba4d", size = 38612244, upload-time = "2024-05-23T03:21:21.827Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/02/1165905f14962174e6569076bcc3315809ae1291ed14de6448cc151eedfd/scipy-1.13.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a014c2b3697bde71724244f63de2476925596c24285c7a637364761f8710891c", size = 38845637, upload-time = "2024-05-23T03:21:28.729Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/77/dab54fe647a08ee4253963bcd8f9cf17509c8ca64d6335141422fe2e2114/scipy-1.13.1-cp39-cp39-win_amd64.whl", hash = "sha256:392e4ec766654852c25ebad4f64e4e584cf19820b980bc04960bca0b0cd6eaa2", size = 46227440, upload-time = "2024-05-23T03:21:35.888Z" },
 ]
 
 [[package]]
 name = "scipy"
-version = "1.15.1"
+version = "1.15.3"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.13'",
-    "python_full_version == '3.12.*'",
-    "python_full_version == '3.11.*'",
     "python_full_version == '3.10.*'",
 ]
 dependencies = [
-    { name = "numpy", marker = "python_full_version >= '3.10'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/76/c6/8eb0654ba0c7d0bb1bf67bf8fbace101a8e4f250f7722371105e8b6f68fc/scipy-1.15.1.tar.gz", hash = "sha256:033a75ddad1463970c96a88063a1df87ccfddd526437136b6ee81ff0312ebdf6", size = 59407493 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/86/53/b204ce5a4433f1864001b9d16f103b9c25f5002a602ae83585d0ea5f9c4a/scipy-1.15.1-cp310-cp310-macosx_10_13_x86_64.whl", hash = "sha256:c64ded12dcab08afff9e805a67ff4480f5e69993310e093434b10e85dc9d43e1", size = 41414518 },
-    { url = "https://files.pythonhosted.org/packages/c7/fc/54ffa7a8847f7f303197a6ba65a66104724beba2e38f328135a78f0dc480/scipy-1.15.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:5b190b935e7db569960b48840e5bef71dc513314cc4e79a1b7d14664f57fd4ff", size = 32519265 },
-    { url = "https://files.pythonhosted.org/packages/f1/77/a98b8ba03d6f371dc31a38719affd53426d4665729dcffbed4afe296784a/scipy-1.15.1-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:4b17d4220df99bacb63065c76b0d1126d82bbf00167d1730019d2a30d6ae01ea", size = 24792859 },
-    { url = "https://files.pythonhosted.org/packages/a7/78/70bb9f0df7444b18b108580934bfef774822e28fd34a68e5c263c7d2828a/scipy-1.15.1-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:63b9b6cd0333d0eb1a49de6f834e8aeaefe438df8f6372352084535ad095219e", size = 27886506 },
-    { url = "https://files.pythonhosted.org/packages/14/a7/f40f6033e06de4176ddd6cc8c3ae9f10a226c3bca5d6b4ab883bc9914a14/scipy-1.15.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9f151e9fb60fbf8e52426132f473221a49362091ce7a5e72f8aa41f8e0da4f25", size = 38375041 },
-    { url = "https://files.pythonhosted.org/packages/17/03/390a1c5c61fd76b0fa4b3c5aa3bdd7e60f6c46f712924f1a9df5705ec046/scipy-1.15.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:21e10b1dd56ce92fba3e786007322542361984f8463c6d37f6f25935a5a6ef52", size = 40597556 },
-    { url = "https://files.pythonhosted.org/packages/4e/70/fa95b3ae026b97eeca58204a90868802e5155ac71b9d7bdee92b68115dd3/scipy-1.15.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5dff14e75cdbcf07cdaa1c7707db6017d130f0af9ac41f6ce443a93318d6c6e0", size = 42938505 },
-    { url = "https://files.pythonhosted.org/packages/d6/07/427859116bdd71847c898180f01802691f203c3e2455a1eb496130ff07c5/scipy-1.15.1-cp310-cp310-win_amd64.whl", hash = "sha256:f82fcf4e5b377f819542fbc8541f7b5fbcf1c0017d0df0bc22c781bf60abc4d8", size = 43909663 },
-    { url = "https://files.pythonhosted.org/packages/8e/2e/7b71312da9c2dabff53e7c9a9d08231bc34d9d8fdabe88a6f1155b44591c/scipy-1.15.1-cp311-cp311-macosx_10_13_x86_64.whl", hash = "sha256:5bd8d27d44e2c13d0c1124e6a556454f52cd3f704742985f6b09e75e163d20d2", size = 41424362 },
-    { url = "https://files.pythonhosted.org/packages/81/8c/ab85f1aa1cc200c796532a385b6ebf6a81089747adc1da7482a062acc46c/scipy-1.15.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:be3deeb32844c27599347faa077b359584ba96664c5c79d71a354b80a0ad0ce0", size = 32535910 },
-    { url = "https://files.pythonhosted.org/packages/3b/9c/6f4b787058daa8d8da21ddff881b4320e28de4704a65ec147adb50cb2230/scipy-1.15.1-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:5eb0ca35d4b08e95da99a9f9c400dc9f6c21c424298a0ba876fdc69c7afacedf", size = 24809398 },
-    { url = "https://files.pythonhosted.org/packages/16/2b/949460a796df75fc7a1ee1becea202cf072edbe325ebe29f6d2029947aa7/scipy-1.15.1-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:74bb864ff7640dea310a1377d8567dc2cb7599c26a79ca852fc184cc851954ac", size = 27918045 },
-    { url = "https://files.pythonhosted.org/packages/5f/36/67fe249dd7ccfcd2a38b25a640e3af7e59d9169c802478b6035ba91dfd6d/scipy-1.15.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:667f950bf8b7c3a23b4199db24cb9bf7512e27e86d0e3813f015b74ec2c6e3df", size = 38332074 },
-    { url = "https://files.pythonhosted.org/packages/fc/da/452e1119e6f720df3feb588cce3c42c5e3d628d4bfd4aec097bd30b7de0c/scipy-1.15.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:395be70220d1189756068b3173853029a013d8c8dd5fd3d1361d505b2aa58fa7", size = 40588469 },
-    { url = "https://files.pythonhosted.org/packages/7f/71/5f94aceeac99a4941478af94fe9f459c6752d497035b6b0761a700f5f9ff/scipy-1.15.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ce3a000cd28b4430426db2ca44d96636f701ed12e2b3ca1f2b1dd7abdd84b39a", size = 42965214 },
-    { url = "https://files.pythonhosted.org/packages/af/25/caa430865749d504271757cafd24066d596217e83326155993980bc22f97/scipy-1.15.1-cp311-cp311-win_amd64.whl", hash = "sha256:3fe1d95944f9cf6ba77aa28b82dd6bb2a5b52f2026beb39ecf05304b8392864b", size = 43896034 },
-    { url = "https://files.pythonhosted.org/packages/d8/6e/a9c42d0d39e09ed7fd203d0ac17adfea759cba61ab457671fe66e523dbec/scipy-1.15.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c09aa9d90f3500ea4c9b393ee96f96b0ccb27f2f350d09a47f533293c78ea776", size = 41478318 },
-    { url = "https://files.pythonhosted.org/packages/04/ee/e3e535c81828618878a7433992fecc92fa4df79393f31a8fea1d05615091/scipy-1.15.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:0ac102ce99934b162914b1e4a6b94ca7da0f4058b6d6fd65b0cef330c0f3346f", size = 32596696 },
-    { url = "https://files.pythonhosted.org/packages/c4/5e/b1b0124be8e76f87115f16b8915003eec4b7060298117715baf13f51942c/scipy-1.15.1-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:09c52320c42d7f5c7748b69e9f0389266fd4f82cf34c38485c14ee976cb8cb04", size = 24870366 },
-    { url = "https://files.pythonhosted.org/packages/14/36/c00cb73eefda85946172c27913ab995c6ad4eee00fa4f007572e8c50cd51/scipy-1.15.1-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:cdde8414154054763b42b74fe8ce89d7f3d17a7ac5dd77204f0e142cdc9239e9", size = 28007461 },
-    { url = "https://files.pythonhosted.org/packages/68/94/aff5c51b3799349a9d1e67a056772a0f8a47db371e83b498d43467806557/scipy-1.15.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4c9d8fc81d6a3b6844235e6fd175ee1d4c060163905a2becce8e74cb0d7554ce", size = 38068174 },
-    { url = "https://files.pythonhosted.org/packages/b0/3c/0de11ca154e24a57b579fb648151d901326d3102115bc4f9a7a86526ce54/scipy-1.15.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0fb57b30f0017d4afa5fe5f5b150b8f807618819287c21cbe51130de7ccdaed2", size = 40249869 },
-    { url = "https://files.pythonhosted.org/packages/15/09/472e8d0a6b33199d1bb95e49bedcabc0976c3724edd9b0ef7602ccacf41e/scipy-1.15.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:491d57fe89927fa1aafbe260f4cfa5ffa20ab9f1435025045a5315006a91b8f5", size = 42629068 },
-    { url = "https://files.pythonhosted.org/packages/ff/ba/31c7a8131152822b3a2cdeba76398ffb404d81d640de98287d236da90c49/scipy-1.15.1-cp312-cp312-win_amd64.whl", hash = "sha256:900f3fa3db87257510f011c292a5779eb627043dd89731b9c461cd16ef76ab3d", size = 43621992 },
-    { url = "https://files.pythonhosted.org/packages/2b/bf/dd68965a4c5138a630eeed0baec9ae96e5d598887835bdde96cdd2fe4780/scipy-1.15.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:100193bb72fbff37dbd0bf14322314fc7cbe08b7ff3137f11a34d06dc0ee6b85", size = 41441136 },
-    { url = "https://files.pythonhosted.org/packages/ef/5e/4928581312922d7e4d416d74c416a660addec4dd5ea185401df2269ba5a0/scipy-1.15.1-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:2114a08daec64980e4b4cbdf5bee90935af66d750146b1d2feb0d3ac30613692", size = 32533699 },
-    { url = "https://files.pythonhosted.org/packages/32/90/03f99c43041852837686898c66767787cd41c5843d7a1509c39ffef683e9/scipy-1.15.1-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:6b3e71893c6687fc5e29208d518900c24ea372a862854c9888368c0b267387ab", size = 24807289 },
-    { url = "https://files.pythonhosted.org/packages/9d/52/bfe82b42ae112eaba1af2f3e556275b8727d55ac6e4932e7aef337a9d9d4/scipy-1.15.1-cp313-cp313-macosx_14_0_x86_64.whl", hash = "sha256:837299eec3d19b7e042923448d17d95a86e43941104d33f00da7e31a0f715d3c", size = 27929844 },
-    { url = "https://files.pythonhosted.org/packages/f6/77/54ff610bad600462c313326acdb035783accc6a3d5f566d22757ad297564/scipy-1.15.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:82add84e8a9fb12af5c2c1a3a3f1cb51849d27a580cb9e6bd66226195142be6e", size = 38031272 },
-    { url = "https://files.pythonhosted.org/packages/f1/26/98585cbf04c7cf503d7eb0a1966df8a268154b5d923c5fe0c1ed13154c49/scipy-1.15.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:070d10654f0cb6abd295bc96c12656f948e623ec5f9a4eab0ddb1466c000716e", size = 40210217 },
-    { url = "https://files.pythonhosted.org/packages/fd/3f/3d2285eb6fece8bc5dbb2f9f94d61157d61d155e854fd5fea825b8218f12/scipy-1.15.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:55cc79ce4085c702ac31e49b1e69b27ef41111f22beafb9b49fea67142b696c4", size = 42587785 },
-    { url = "https://files.pythonhosted.org/packages/48/7d/5b5251984bf0160d6533695a74a5fddb1fa36edd6f26ffa8c871fbd4782a/scipy-1.15.1-cp313-cp313-win_amd64.whl", hash = "sha256:c352c1b6d7cac452534517e022f8f7b8d139cd9f27e6fbd9f3cbd0bfd39f5bef", size = 43640439 },
-    { url = "https://files.pythonhosted.org/packages/e7/b8/0e092f592d280496de52e152582030f8a270b194f87f890e1a97c5599b81/scipy-1.15.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:0458839c9f873062db69a03de9a9765ae2e694352c76a16be44f93ea45c28d2b", size = 41619862 },
-    { url = "https://files.pythonhosted.org/packages/f6/19/0b6e1173aba4db9e0b7aa27fe45019857fb90d6904038b83927cbe0a6c1d/scipy-1.15.1-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:af0b61c1de46d0565b4b39c6417373304c1d4f5220004058bdad3061c9fa8a95", size = 32610387 },
-    { url = "https://files.pythonhosted.org/packages/e7/02/754aae3bd1fa0f2479ade3cfdf1732ecd6b05853f63eee6066a32684563a/scipy-1.15.1-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:71ba9a76c2390eca6e359be81a3e879614af3a71dfdabb96d1d7ab33da6f2364", size = 24883814 },
-    { url = "https://files.pythonhosted.org/packages/1f/ac/d7906201604a2ea3b143bb0de51b3966f66441ba50b7dc182c4505b3edf9/scipy-1.15.1-cp313-cp313t-macosx_14_0_x86_64.whl", hash = "sha256:14eaa373c89eaf553be73c3affb11ec6c37493b7eaaf31cf9ac5dffae700c2e0", size = 27944865 },
-    { url = "https://files.pythonhosted.org/packages/84/9d/8f539002b5e203723af6a6f513a45e0a7671e9dabeedb08f417ac17e4edc/scipy-1.15.1-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f735bc41bd1c792c96bc426dece66c8723283695f02df61dcc4d0a707a42fc54", size = 39883261 },
-    { url = "https://files.pythonhosted.org/packages/97/c0/62fd3bab828bcccc9b864c5997645a3b86372a35941cdaf677565c25c98d/scipy-1.15.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:2722a021a7929d21168830790202a75dbb20b468a8133c74a2c0230c72626b6c", size = 42093299 },
-    { url = "https://files.pythonhosted.org/packages/e4/1f/5d46a8d94e9f6d2c913cbb109e57e7eed914de38ea99e2c4d69a9fc93140/scipy-1.15.1-cp313-cp313t-win_amd64.whl", hash = "sha256:bc7136626261ac1ed988dca56cfc4ab5180f75e0ee52e58f1e6aa74b5f3eacd5", size = 43181730 },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/0f/37/6964b830433e654ec7485e45a00fc9a27cf868d622838f6b6d9c5ec0d532/scipy-1.15.3.tar.gz", hash = "sha256:eae3cf522bc7df64b42cad3925c876e1b0b6c35c1337c93e12c0f366f55b0eaf", size = 59419214, upload-time = "2025-05-08T16:13:05.955Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/78/2f/4966032c5f8cc7e6a60f1b2e0ad686293b9474b65246b0c642e3ef3badd0/scipy-1.15.3-cp310-cp310-macosx_10_13_x86_64.whl", hash = "sha256:a345928c86d535060c9c2b25e71e87c39ab2f22fc96e9636bd74d1dbf9de448c", size = 38702770, upload-time = "2025-05-08T16:04:20.849Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/6e/0c3bf90fae0e910c274db43304ebe25a6b391327f3f10b5dcc638c090795/scipy-1.15.3-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:ad3432cb0f9ed87477a8d97f03b763fd1d57709f1bbde3c9369b1dff5503b253", size = 30094511, upload-time = "2025-05-08T16:04:27.103Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/b1/4deb37252311c1acff7f101f6453f0440794f51b6eacb1aad4459a134081/scipy-1.15.3-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:aef683a9ae6eb00728a542b796f52a5477b78252edede72b8327a886ab63293f", size = 22368151, upload-time = "2025-05-08T16:04:31.731Z" },
+    { url = "https://files.pythonhosted.org/packages/38/7d/f457626e3cd3c29b3a49ca115a304cebb8cc6f31b04678f03b216899d3c6/scipy-1.15.3-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:1c832e1bd78dea67d5c16f786681b28dd695a8cb1fb90af2e27580d3d0967e92", size = 25121732, upload-time = "2025-05-08T16:04:36.596Z" },
+    { url = "https://files.pythonhosted.org/packages/db/0a/92b1de4a7adc7a15dcf5bddc6e191f6f29ee663b30511ce20467ef9b82e4/scipy-1.15.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:263961f658ce2165bbd7b99fa5135195c3a12d9bef045345016b8b50c315cb82", size = 35547617, upload-time = "2025-05-08T16:04:43.546Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/6d/41991e503e51fc1134502694c5fa7a1671501a17ffa12716a4a9151af3df/scipy-1.15.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9e2abc762b0811e09a0d3258abee2d98e0c703eee49464ce0069590846f31d40", size = 37662964, upload-time = "2025-05-08T16:04:49.431Z" },
+    { url = "https://files.pythonhosted.org/packages/25/e1/3df8f83cb15f3500478c889be8fb18700813b95e9e087328230b98d547ff/scipy-1.15.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ed7284b21a7a0c8f1b6e5977ac05396c0d008b89e05498c8b7e8f4a1423bba0e", size = 37238749, upload-time = "2025-05-08T16:04:55.215Z" },
+    { url = "https://files.pythonhosted.org/packages/93/3e/b3257cf446f2a3533ed7809757039016b74cd6f38271de91682aa844cfc5/scipy-1.15.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5380741e53df2c566f4d234b100a484b420af85deb39ea35a1cc1be84ff53a5c", size = 40022383, upload-time = "2025-05-08T16:05:01.914Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/84/55bc4881973d3f79b479a5a2e2df61c8c9a04fcb986a213ac9c02cfb659b/scipy-1.15.3-cp310-cp310-win_amd64.whl", hash = "sha256:9d61e97b186a57350f6d6fd72640f9e99d5a4a2b8fbf4b9ee9a841eab327dc13", size = 41259201, upload-time = "2025-05-08T16:05:08.166Z" },
+    { url = "https://files.pythonhosted.org/packages/96/ab/5cc9f80f28f6a7dff646c5756e559823614a42b1939d86dd0ed550470210/scipy-1.15.3-cp311-cp311-macosx_10_13_x86_64.whl", hash = "sha256:993439ce220d25e3696d1b23b233dd010169b62f6456488567e830654ee37a6b", size = 38714255, upload-time = "2025-05-08T16:05:14.596Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/4a/66ba30abe5ad1a3ad15bfb0b59d22174012e8056ff448cb1644deccbfed2/scipy-1.15.3-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:34716e281f181a02341ddeaad584205bd2fd3c242063bd3423d61ac259ca7eba", size = 30111035, upload-time = "2025-05-08T16:05:20.152Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/fa/a7e5b95afd80d24313307f03624acc65801846fa75599034f8ceb9e2cbf6/scipy-1.15.3-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:3b0334816afb8b91dab859281b1b9786934392aa3d527cd847e41bb6f45bee65", size = 22384499, upload-time = "2025-05-08T16:05:24.494Z" },
+    { url = "https://files.pythonhosted.org/packages/17/99/f3aaddccf3588bb4aea70ba35328c204cadd89517a1612ecfda5b2dd9d7a/scipy-1.15.3-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:6db907c7368e3092e24919b5e31c76998b0ce1684d51a90943cb0ed1b4ffd6c1", size = 25152602, upload-time = "2025-05-08T16:05:29.313Z" },
+    { url = "https://files.pythonhosted.org/packages/56/c5/1032cdb565f146109212153339f9cb8b993701e9fe56b1c97699eee12586/scipy-1.15.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:721d6b4ef5dc82ca8968c25b111e307083d7ca9091bc38163fb89243e85e3889", size = 35503415, upload-time = "2025-05-08T16:05:34.699Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/37/89f19c8c05505d0601ed5650156e50eb881ae3918786c8fd7262b4ee66d3/scipy-1.15.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:39cb9c62e471b1bb3750066ecc3a3f3052b37751c7c3dfd0fd7e48900ed52982", size = 37652622, upload-time = "2025-05-08T16:05:40.762Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/31/be59513aa9695519b18e1851bb9e487de66f2d31f835201f1b42f5d4d475/scipy-1.15.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:795c46999bae845966368a3c013e0e00947932d68e235702b5c3f6ea799aa8c9", size = 37244796, upload-time = "2025-05-08T16:05:48.119Z" },
+    { url = "https://files.pythonhosted.org/packages/10/c0/4f5f3eeccc235632aab79b27a74a9130c6c35df358129f7ac8b29f562ac7/scipy-1.15.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:18aaacb735ab38b38db42cb01f6b92a2d0d4b6aabefeb07f02849e47f8fb3594", size = 40047684, upload-time = "2025-05-08T16:05:54.22Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/a7/0ddaf514ce8a8714f6ed243a2b391b41dbb65251affe21ee3077ec45ea9a/scipy-1.15.3-cp311-cp311-win_amd64.whl", hash = "sha256:ae48a786a28412d744c62fd7816a4118ef97e5be0bee968ce8f0a2fba7acf3bb", size = 41246504, upload-time = "2025-05-08T16:06:00.437Z" },
+    { url = "https://files.pythonhosted.org/packages/37/4b/683aa044c4162e10ed7a7ea30527f2cbd92e6999c10a8ed8edb253836e9c/scipy-1.15.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:6ac6310fdbfb7aa6612408bd2f07295bcbd3fda00d2d702178434751fe48e019", size = 38766735, upload-time = "2025-05-08T16:06:06.471Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/7e/f30be3d03de07f25dc0ec926d1681fed5c732d759ac8f51079708c79e680/scipy-1.15.3-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:185cd3d6d05ca4b44a8f1595af87f9c372bb6acf9c808e99aa3e9aa03bd98cf6", size = 30173284, upload-time = "2025-05-08T16:06:11.686Z" },
+    { url = "https://files.pythonhosted.org/packages/07/9c/0ddb0d0abdabe0d181c1793db51f02cd59e4901da6f9f7848e1f96759f0d/scipy-1.15.3-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:05dc6abcd105e1a29f95eada46d4a3f251743cfd7d3ae8ddb4088047f24ea477", size = 22446958, upload-time = "2025-05-08T16:06:15.97Z" },
+    { url = "https://files.pythonhosted.org/packages/af/43/0bce905a965f36c58ff80d8bea33f1f9351b05fad4beaad4eae34699b7a1/scipy-1.15.3-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:06efcba926324df1696931a57a176c80848ccd67ce6ad020c810736bfd58eb1c", size = 25242454, upload-time = "2025-05-08T16:06:20.394Z" },
+    { url = "https://files.pythonhosted.org/packages/56/30/a6f08f84ee5b7b28b4c597aca4cbe545535c39fe911845a96414700b64ba/scipy-1.15.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c05045d8b9bfd807ee1b9f38761993297b10b245f012b11b13b91ba8945f7e45", size = 35210199, upload-time = "2025-05-08T16:06:26.159Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/1f/03f52c282437a168ee2c7c14a1a0d0781a9a4a8962d84ac05c06b4c5b555/scipy-1.15.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:271e3713e645149ea5ea3e97b57fdab61ce61333f97cfae392c28ba786f9bb49", size = 37309455, upload-time = "2025-05-08T16:06:32.778Z" },
+    { url = "https://files.pythonhosted.org/packages/89/b1/fbb53137f42c4bf630b1ffdfc2151a62d1d1b903b249f030d2b1c0280af8/scipy-1.15.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:6cfd56fc1a8e53f6e89ba3a7a7251f7396412d655bca2aa5611c8ec9a6784a1e", size = 36885140, upload-time = "2025-05-08T16:06:39.249Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/2e/025e39e339f5090df1ff266d021892694dbb7e63568edcfe43f892fa381d/scipy-1.15.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:0ff17c0bb1cb32952c09217d8d1eed9b53d1463e5f1dd6052c7857f83127d539", size = 39710549, upload-time = "2025-05-08T16:06:45.729Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/eb/3bf6ea8ab7f1503dca3a10df2e4b9c3f6b3316df07f6c0ded94b281c7101/scipy-1.15.3-cp312-cp312-win_amd64.whl", hash = "sha256:52092bc0472cfd17df49ff17e70624345efece4e1a12b23783a1ac59a1b728ed", size = 40966184, upload-time = "2025-05-08T16:06:52.623Z" },
+]
+
+[[package]]
+name = "scipy"
+version = "1.16.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+]
+dependencies = [
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f5/4a/b927028464795439faec8eaf0b03b011005c487bb2d07409f28bf30879c4/scipy-1.16.1.tar.gz", hash = "sha256:44c76f9e8b6e8e488a586190ab38016e4ed2f8a038af7cd3defa903c0a2238b3", size = 30580861, upload-time = "2025-07-27T16:33:30.834Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/da/91/812adc6f74409b461e3a5fa97f4f74c769016919203138a3bf6fc24ba4c5/scipy-1.16.1-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:c033fa32bab91dc98ca59d0cf23bb876454e2bb02cbe592d5023138778f70030", size = 36552519, upload-time = "2025-07-27T16:26:29.658Z" },
+    { url = "https://files.pythonhosted.org/packages/47/18/8e355edcf3b71418d9e9f9acd2708cc3a6c27e8f98fde0ac34b8a0b45407/scipy-1.16.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:6e5c2f74e5df33479b5cd4e97a9104c511518fbd979aa9b8f6aec18b2e9ecae7", size = 28638010, upload-time = "2025-07-27T16:26:38.196Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/eb/e931853058607bdfbc11b86df19ae7a08686121c203483f62f1ecae5989c/scipy-1.16.1-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:0a55ffe0ba0f59666e90951971a884d1ff6f4ec3275a48f472cfb64175570f77", size = 20909790, upload-time = "2025-07-27T16:26:43.93Z" },
+    { url = "https://files.pythonhosted.org/packages/45/0c/be83a271d6e96750cd0be2e000f35ff18880a46f05ce8b5d3465dc0f7a2a/scipy-1.16.1-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:f8a5d6cd147acecc2603fbd382fed6c46f474cccfcf69ea32582e033fb54dcfe", size = 23513352, upload-time = "2025-07-27T16:26:50.017Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/bf/fe6eb47e74f762f933cca962db7f2c7183acfdc4483bd1c3813cfe83e538/scipy-1.16.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:cb18899127278058bcc09e7b9966d41a5a43740b5bb8dcba401bd983f82e885b", size = 33534643, upload-time = "2025-07-27T16:26:57.503Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/ba/63f402e74875486b87ec6506a4f93f6d8a0d94d10467280f3d9d7837ce3a/scipy-1.16.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:adccd93a2fa937a27aae826d33e3bfa5edf9aa672376a4852d23a7cd67a2e5b7", size = 35376776, upload-time = "2025-07-27T16:27:06.639Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/b4/04eb9d39ec26a1b939689102da23d505ea16cdae3dbb18ffc53d1f831044/scipy-1.16.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:18aca1646a29ee9a0625a1be5637fa798d4d81fdf426481f06d69af828f16958", size = 35698906, upload-time = "2025-07-27T16:27:14.943Z" },
+    { url = "https://files.pythonhosted.org/packages/04/d6/bb5468da53321baeb001f6e4e0d9049eadd175a4a497709939128556e3ec/scipy-1.16.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:d85495cef541729a70cdddbbf3e6b903421bc1af3e8e3a9a72a06751f33b7c39", size = 38129275, upload-time = "2025-07-27T16:27:23.873Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/94/994369978509f227cba7dfb9e623254d0d5559506fe994aef4bea3ed469c/scipy-1.16.1-cp311-cp311-win_amd64.whl", hash = "sha256:226652fca853008119c03a8ce71ffe1b3f6d2844cc1686e8f9806edafae68596", size = 38644572, upload-time = "2025-07-27T16:27:32.637Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/d9/ec4864f5896232133f51382b54a08de91a9d1af7a76dfa372894026dfee2/scipy-1.16.1-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:81b433bbeaf35728dad619afc002db9b189e45eebe2cd676effe1fb93fef2b9c", size = 36575194, upload-time = "2025-07-27T16:27:41.321Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/6d/40e81ecfb688e9d25d34a847dca361982a6addf8e31f0957b1a54fbfa994/scipy-1.16.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:886cc81fdb4c6903a3bb0464047c25a6d1016fef77bb97949817d0c0d79f9e04", size = 28594590, upload-time = "2025-07-27T16:27:49.204Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/37/9f65178edfcc629377ce9a64fc09baebea18c80a9e57ae09a52edf84880b/scipy-1.16.1-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:15240c3aac087a522b4eaedb09f0ad061753c5eebf1ea430859e5bf8640d5919", size = 20866458, upload-time = "2025-07-27T16:27:54.98Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/7b/749a66766871ea4cb1d1ea10f27004db63023074c22abed51f22f09770e0/scipy-1.16.1-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:65f81a25805f3659b48126b5053d9e823d3215e4a63730b5e1671852a1705921", size = 23539318, upload-time = "2025-07-27T16:28:01.604Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/db/8d4afec60eb833a666434d4541a3151eedbf2494ea6d4d468cbe877f00cd/scipy-1.16.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:6c62eea7f607f122069b9bad3f99489ddca1a5173bef8a0c75555d7488b6f725", size = 33292899, upload-time = "2025-07-27T16:28:09.147Z" },
+    { url = "https://files.pythonhosted.org/packages/51/1e/79023ca3bbb13a015d7d2757ecca3b81293c663694c35d6541b4dca53e98/scipy-1.16.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f965bbf3235b01c776115ab18f092a95aa74c271a52577bcb0563e85738fd618", size = 35162637, upload-time = "2025-07-27T16:28:17.535Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/49/0648665f9c29fdaca4c679182eb972935b3b4f5ace41d323c32352f29816/scipy-1.16.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:f006e323874ffd0b0b816d8c6a8e7f9a73d55ab3b8c3f72b752b226d0e3ac83d", size = 35490507, upload-time = "2025-07-27T16:28:25.705Z" },
+    { url = "https://files.pythonhosted.org/packages/62/8f/66cbb9d6bbb18d8c658f774904f42a92078707a7c71e5347e8bf2f52bb89/scipy-1.16.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e8fd15fc5085ab4cca74cb91fe0a4263b1f32e4420761ddae531ad60934c2119", size = 37923998, upload-time = "2025-07-27T16:28:34.339Z" },
+    { url = "https://files.pythonhosted.org/packages/14/c3/61f273ae550fbf1667675701112e380881905e28448c080b23b5a181df7c/scipy-1.16.1-cp312-cp312-win_amd64.whl", hash = "sha256:f7b8013c6c066609577d910d1a2a077021727af07b6fab0ee22c2f901f22352a", size = 38508060, upload-time = "2025-07-27T16:28:43.242Z" },
 ]
 
 [[package]]
 name = "sentencepiece"
-version = "0.2.0"
+version = "0.2.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/c9/d2/b9c7ca067c26d8ff085d252c89b5f69609ca93fb85a00ede95f4857865d4/sentencepiece-0.2.0.tar.gz", hash = "sha256:a52c19171daaf2e697dc6cbe67684e0fa341b1248966f6aebb541de654d15843", size = 2632106 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/f6/71/98648c3b64b23edb5403f74bcc906ad21766872a6e1ada26ea3f1eb941ab/sentencepiece-0.2.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:188779e1298a1c8b8253c7d3ad729cb0a9891e5cef5e5d07ce4592c54869e227", size = 2408979 },
-    { url = "https://files.pythonhosted.org/packages/77/9f/7efbaa6d4c0c718a9affbecc536b03ca62f99f421bdffb531c16030e2d2b/sentencepiece-0.2.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:bed9cf85b296fa2b76fc2547b9cbb691a523864cebaee86304c43a7b4cb1b452", size = 1238845 },
-    { url = "https://files.pythonhosted.org/packages/1c/e4/c2541027a43ec6962ba9b601805d17ba3f86b38bdeae0e8ac65a2981e248/sentencepiece-0.2.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d7b67e724bead13f18db6e1d10b6bbdc454af574d70efbb36f27d90387be1ca3", size = 1181472 },
-    { url = "https://files.pythonhosted.org/packages/fd/46/316c1ba6c52b97de76aff7b9da678f7afbb52136afb2987c474d95630e65/sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2fde4b08cfe237be4484c6c7c2e2c75fb862cfeab6bd5449ce4caeafd97b767a", size = 1259151 },
-    { url = "https://files.pythonhosted.org/packages/aa/5a/3c48738a0835d76dd06c62b6ac48d39c923cde78dd0f587353bdcbb99851/sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4c378492056202d1c48a4979650981635fd97875a00eabb1f00c6a236b013b5e", size = 1355931 },
-    { url = "https://files.pythonhosted.org/packages/a6/27/33019685023221ca8ed98e8ceb7ae5e166032686fa3662c68f1f1edf334e/sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1380ce6540a368de2ef6d7e6ba14ba8f3258df650d39ba7d833b79ee68a52040", size = 1301537 },
-    { url = "https://files.pythonhosted.org/packages/ca/e4/55f97cef14293171fef5f96e96999919ab5b4d1ce95b53547ad653d7e3bf/sentencepiece-0.2.0-cp310-cp310-win32.whl", hash = "sha256:a1151d6a6dd4b43e552394aed0edfe9292820272f0194bd56c7c1660a0c06c3d", size = 936747 },
-    { url = "https://files.pythonhosted.org/packages/85/f4/4ef1a6e0e9dbd8a60780a91df8b7452ada14cfaa0e17b3b8dfa42cecae18/sentencepiece-0.2.0-cp310-cp310-win_amd64.whl", hash = "sha256:d490142b0521ef22bc1085f061d922a2a6666175bb6b42e588ff95c0db6819b2", size = 991525 },
-    { url = "https://files.pythonhosted.org/packages/32/43/8f8885168a47a02eba1455bd3f4f169f50ad5b8cebd2402d0f5e20854d04/sentencepiece-0.2.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:17982700c4f6dbb55fa3594f3d7e5dd1c8659a274af3738e33c987d2a27c9d5c", size = 2409036 },
-    { url = "https://files.pythonhosted.org/packages/0f/35/e63ba28062af0a3d688a9f128e407a1a2608544b2f480cb49bf7f4b1cbb9/sentencepiece-0.2.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:7c867012c0e8bcd5bdad0f791609101cb5c66acb303ab3270218d6debc68a65e", size = 1238921 },
-    { url = "https://files.pythonhosted.org/packages/de/42/ae30952c4a0bd773e90c9bf2579f5533037c886dfc8ec68133d5694f4dd2/sentencepiece-0.2.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7fd6071249c74f779c5b27183295b9202f8dedb68034e716784364443879eaa6", size = 1181477 },
-    { url = "https://files.pythonhosted.org/packages/e3/ac/2f2ab1d60bb2d795d054eebe5e3f24b164bc21b5a9b75fba7968b3b91b5a/sentencepiece-0.2.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:27f90c55a65013cbb8f4d7aab0599bf925cde4adc67ae43a0d323677b5a1c6cb", size = 1259182 },
-    { url = "https://files.pythonhosted.org/packages/45/fb/14633c6ecf262c468759ffcdb55c3a7ee38fe4eda6a70d75ee7c7d63c58b/sentencepiece-0.2.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b293734059ef656dcd65be62ff771507bea8fed0a711b6733976e1ed3add4553", size = 1355537 },
-    { url = "https://files.pythonhosted.org/packages/fb/12/2f5c8d4764b00033cf1c935b702d3bb878d10be9f0b87f0253495832d85f/sentencepiece-0.2.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e58b47f933aca74c6a60a79dcb21d5b9e47416256c795c2d58d55cec27f9551d", size = 1301464 },
-    { url = "https://files.pythonhosted.org/packages/4e/b1/67afc0bde24f6dcb3acdea0dd8dcdf4b8b0db240f6bacd39378bd32d09f8/sentencepiece-0.2.0-cp311-cp311-win32.whl", hash = "sha256:c581258cf346b327c62c4f1cebd32691826306f6a41d8c4bec43b010dee08e75", size = 936749 },
-    { url = "https://files.pythonhosted.org/packages/a2/f6/587c62fd21fc988555b85351f50bbde43a51524caafd63bc69240ded14fd/sentencepiece-0.2.0-cp311-cp311-win_amd64.whl", hash = "sha256:0993dbc665f4113017892f1b87c3904a44d0640eda510abcacdfb07f74286d36", size = 991520 },
-    { url = "https://files.pythonhosted.org/packages/27/5a/141b227ed54293360a9ffbb7bf8252b4e5efc0400cdeac5809340e5d2b21/sentencepiece-0.2.0-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:ea5f536e32ea8ec96086ee00d7a4a131ce583a1b18d130711707c10e69601cb2", size = 2409370 },
-    { url = "https://files.pythonhosted.org/packages/2e/08/a4c135ad6fc2ce26798d14ab72790d66e813efc9589fd30a5316a88ca8d5/sentencepiece-0.2.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:d0cb51f53b6aae3c36bafe41e86167c71af8370a039f542c43b0cce5ef24a68c", size = 1239288 },
-    { url = "https://files.pythonhosted.org/packages/49/0a/2fe387f825ac5aad5a0bfe221904882106cac58e1b693ba7818785a882b6/sentencepiece-0.2.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3212121805afc58d8b00ab4e7dd1f8f76c203ddb9dc94aa4079618a31cf5da0f", size = 1181597 },
-    { url = "https://files.pythonhosted.org/packages/cc/38/e4698ee2293fe4835dc033c49796a39b3eebd8752098f6bd0aa53a14af1f/sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2a3149e3066c2a75e0d68a43eb632d7ae728c7925b517f4c05c40f6f7280ce08", size = 1259220 },
-    { url = "https://files.pythonhosted.org/packages/12/24/fd7ef967c9dad2f6e6e5386d0cadaf65cda8b7be6e3861a9ab3121035139/sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:632f3594d3e7ac8b367bca204cb3fd05a01d5b21455acd097ea4c0e30e2f63d7", size = 1355962 },
-    { url = "https://files.pythonhosted.org/packages/4f/d2/18246f43ca730bb81918f87b7e886531eda32d835811ad9f4657c54eee35/sentencepiece-0.2.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f295105c6bdbb05bd5e1b0cafbd78ff95036f5d3641e7949455a3f4e5e7c3109", size = 1301706 },
-    { url = "https://files.pythonhosted.org/packages/8a/47/ca237b562f420044ab56ddb4c278672f7e8c866e183730a20e413b38a989/sentencepiece-0.2.0-cp312-cp312-win32.whl", hash = "sha256:fb89f811e5efd18bab141afc3fea3de141c3f69f3fe9e898f710ae7fe3aab251", size = 936941 },
-    { url = "https://files.pythonhosted.org/packages/c6/97/d159c32642306ee2b70732077632895438867b3b6df282354bd550cf2a67/sentencepiece-0.2.0-cp312-cp312-win_amd64.whl", hash = "sha256:7a673a72aab81fef5ebe755c6e0cc60087d1f3a4700835d40537183c1703a45f", size = 991994 },
-    { url = "https://files.pythonhosted.org/packages/e9/18/eb620d94d63f62ca69cecccf4459529864ac3fbb35ec123190bd58dadb46/sentencepiece-0.2.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:1e0f9c4d0a6b0af59b613175f019916e28ade076e21242fd5be24340d8a2f64a", size = 2409003 },
-    { url = "https://files.pythonhosted.org/packages/6e/a6/df28bc0b6a2a86416232c0a5f0d69a9cb7244bb95cb5dcdfcbf01cced8a6/sentencepiece-0.2.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:298f21cc1366eb60311aedba3169d30f885c363ddbf44214b0a587d2908141ad", size = 1238898 },
-    { url = "https://files.pythonhosted.org/packages/79/91/b54a528e0789cd7986341ed3909bec56365c3b672daef8b10aa4098238f0/sentencepiece-0.2.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3f1ec95aa1e5dab11f37ac7eff190493fd87770f7a8b81ebc9dd768d1a3c8704", size = 1181534 },
-    { url = "https://files.pythonhosted.org/packages/a3/69/e96ef68261fa5b82379fdedb325ceaf1d353c6e839ec346d8244e0da5f2f/sentencepiece-0.2.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7b06b70af54daa4b4904cbb90b4eb6d35c9f3252fdc86c9c32d5afd4d30118d8", size = 1259161 },
-    { url = "https://files.pythonhosted.org/packages/45/de/461d15856c29ba1ce778cf76e0462572661f647abc8a5373690c52e98a00/sentencepiece-0.2.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:22e37bac44dd6603388cb598c64ff7a76e41ca774646f21c23aadfbf5a2228ab", size = 1355945 },
-    { url = "https://files.pythonhosted.org/packages/5f/01/c95e42eb86282b2c79305d3e0b0ca5a743f85a61262bb7130999c70b9374/sentencepiece-0.2.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0461324897735512a32d222e3d886e24ad6a499761952b6bda2a9ee6e4313ea5", size = 1301596 },
-    { url = "https://files.pythonhosted.org/packages/be/47/e16f368fe6327e873e8029aa539115025e9f61a4e8ca8f0f8eaf8e6a4c1c/sentencepiece-0.2.0-cp39-cp39-win32.whl", hash = "sha256:38aed822fb76435fa1f12185f10465a94ab9e51d5e8a9159e9a540ce926f0ffd", size = 936757 },
-    { url = "https://files.pythonhosted.org/packages/4b/36/497e6407700efd6b97f81bc160913a70d33b9b09227429f68fc86f387bbe/sentencepiece-0.2.0-cp39-cp39-win_amd64.whl", hash = "sha256:d8cf876516548b5a1d6ac4745d8b554f5c07891d55da557925e5c13ff0b4e6ad", size = 991541 },
+sdist = { url = "https://files.pythonhosted.org/packages/15/15/2e7a025fc62d764b151ae6d0f2a92f8081755ebe8d4a64099accc6f77ba6/sentencepiece-0.2.1.tar.gz", hash = "sha256:8138cec27c2f2282f4a34d9a016e3374cd40e5c6e9cb335063db66a0a3b71fad", size = 3228515, upload-time = "2025-08-12T07:00:51.718Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/af/31/5b7cccb307b485db1a2372d6d2980b0a65d067f8be5ca943a103b4acd5b3/sentencepiece-0.2.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e10fa50bdbaa5e2445dbd387979980d391760faf0ec99a09bd7780ff37eaec44", size = 1942557, upload-time = "2025-08-12T06:59:12.379Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/41/0ac923a8e685ad290c5afc8ae55c5844977b8d75076fcc04302b9a324274/sentencepiece-0.2.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2f27ae6deea72efdb6f361750c92f6c21fd0ad087445082770cc34015213c526", size = 1325384, upload-time = "2025-08-12T06:59:14.334Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/ef/3751555d67daf9003384978f169d31c775cb5c7baf28633caaf1eb2b2b4d/sentencepiece-0.2.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:60937c959e6f44159fdd9f56fbdd302501f96114a5ba436829496d5f32d8de3f", size = 1253317, upload-time = "2025-08-12T06:59:16.247Z" },
+    { url = "https://files.pythonhosted.org/packages/46/a5/742c69b7bd144eb32b6e5fd50dbd8abbbc7a95fce2fe16e50156fa400e3b/sentencepiece-0.2.1-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d8b1d91545578852f128650b8cce4ec20f93d39b378ff554ebe66290f2dabb92", size = 1316379, upload-time = "2025-08-12T06:59:17.825Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/89/8deeafbba2871e8fa10f20f17447786f4ac38085925335728d360eaf4cae/sentencepiece-0.2.1-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:27e38eee653abc3d387862e67bc5c8b6f428cd604e688b85d29170b7e725c26c", size = 1387926, upload-time = "2025-08-12T06:59:19.395Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/ca/67fe73005f0ab617c6a970b199754e28e524b6873aa7025224fad3cda252/sentencepiece-0.2.1-cp310-cp310-win32.whl", hash = "sha256:251874d720ac7f28024a168501f3c7bb15d1802245f6e66de565f18bbb9b5eaa", size = 999550, upload-time = "2025-08-12T06:59:20.844Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/33/dc5b54042050d2dda4229c3ce1f862541c99966390b6aa20f54d520d2dc2/sentencepiece-0.2.1-cp310-cp310-win_amd64.whl", hash = "sha256:e52144670738b4b477fade6c2a9b6af71a8d0094514c9853ac9f6fc1fcfabae7", size = 1054613, upload-time = "2025-08-12T06:59:22.255Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/19/1ea47f46ff97fe04422b78997da1a37cd632f414aae042d27a9009c5b733/sentencepiece-0.2.1-cp310-cp310-win_arm64.whl", hash = "sha256:9076430ac25dfa7147d9d05751dbc66a04bc1aaac371c07f84952979ea59f0d0", size = 1033884, upload-time = "2025-08-12T06:59:24.194Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/15/46afbab00733d81788b64be430ca1b93011bb9388527958e26cc31832de5/sentencepiece-0.2.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6356d0986b8b8dc351b943150fcd81a1c6e6e4d439772e8584c64230e58ca987", size = 1942560, upload-time = "2025-08-12T06:59:25.82Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/79/7c01b8ef98a0567e9d84a4e7a910f8e7074fcbf398a5cd76f93f4b9316f9/sentencepiece-0.2.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:8f8ba89a3acb3dc1ae90f65ec1894b0b9596fdb98ab003ff38e058f898b39bc7", size = 1325385, upload-time = "2025-08-12T06:59:27.722Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/88/2b41e07bd24f33dcf2f18ec3b74247aa4af3526bad8907b8727ea3caba03/sentencepiece-0.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:02593eca45440ef39247cee8c47322a34bdcc1d8ae83ad28ba5a899a2cf8d79a", size = 1253319, upload-time = "2025-08-12T06:59:29.306Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/54/38a1af0c6210a3c6f95aa46d23d6640636d020fba7135cd0d9a84ada05a7/sentencepiece-0.2.1-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a0d15781a171d188b661ae4bde1d998c303f6bd8621498c50c671bd45a4798e", size = 1316162, upload-time = "2025-08-12T06:59:30.914Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/66/fb191403ade791ad2c3c1e72fe8413e63781b08cfa3aa4c9dfc536d6e795/sentencepiece-0.2.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4f5a3e0d9f445ed9d66c0fec47d4b23d12cfc858b407a03c194c1b26c2ac2a63", size = 1387785, upload-time = "2025-08-12T06:59:32.491Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/2d/3bd9b08e70067b2124518b308db6a84a4f8901cc8a4317e2e4288cdd9b4d/sentencepiece-0.2.1-cp311-cp311-win32.whl", hash = "sha256:6d297a1748d429ba8534eebe5535448d78b8acc32d00a29b49acf28102eeb094", size = 999555, upload-time = "2025-08-12T06:59:34.475Z" },
+    { url = "https://files.pythonhosted.org/packages/32/b8/f709977f5fda195ae1ea24f24e7c581163b6f142b1005bc3d0bbfe4d7082/sentencepiece-0.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:82d9ead6591015f009cb1be1cb1c015d5e6f04046dbb8c9588b931e869a29728", size = 1054617, upload-time = "2025-08-12T06:59:36.461Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/40/a1fc23be23067da0f703709797b464e8a30a1c78cc8a687120cd58d4d509/sentencepiece-0.2.1-cp311-cp311-win_arm64.whl", hash = "sha256:39f8651bd10974eafb9834ce30d9bcf5b73e1fc798a7f7d2528f9820ca86e119", size = 1033877, upload-time = "2025-08-12T06:59:38.391Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/be/32ce495aa1d0e0c323dcb1ba87096037358edee539cac5baf8755a6bd396/sentencepiece-0.2.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:57cae326c8727de58c85977b175af132a7138d84c764635d7e71bbee7e774133", size = 1943152, upload-time = "2025-08-12T06:59:40.048Z" },
+    { url = "https://files.pythonhosted.org/packages/88/7e/ff23008899a58678e98c6ff592bf4d368eee5a71af96d0df6b38a039dd4f/sentencepiece-0.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:56dd39a3c4d6493db3cdca7e8cc68c6b633f0d4195495cbadfcf5af8a22d05a6", size = 1325651, upload-time = "2025-08-12T06:59:41.536Z" },
+    { url = "https://files.pythonhosted.org/packages/19/84/42eb3ce4796777a1b5d3699dfd4dca85113e68b637f194a6c8d786f16a04/sentencepiece-0.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:d9381351182ff9888cc80e41c632e7e274b106f450de33d67a9e8f6043da6f76", size = 1253645, upload-time = "2025-08-12T06:59:42.903Z" },
+    { url = "https://files.pythonhosted.org/packages/89/fa/d3d5ebcba3cb9e6d3775a096251860c41a6bc53a1b9461151df83fe93255/sentencepiece-0.2.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:99f955df238021bf11f0fc37cdb54fd5e5b5f7fd30ecc3d93fb48b6815437167", size = 1316273, upload-time = "2025-08-12T06:59:44.476Z" },
+    { url = "https://files.pythonhosted.org/packages/04/88/14f2f4a2b922d8b39be45bf63d79e6cd3a9b2f248b2fcb98a69b12af12f5/sentencepiece-0.2.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0cdfecef430d985f1c2bcbfff3defd1d95dae876fbd0173376012d2d7d24044b", size = 1387881, upload-time = "2025-08-12T06:59:46.09Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/b8/903e5ccb77b4ef140605d5d71b4f9e0ad95d456d6184688073ed11712809/sentencepiece-0.2.1-cp312-cp312-win32.whl", hash = "sha256:a483fd29a34c3e34c39ac5556b0a90942bec253d260235729e50976f5dba1068", size = 999540, upload-time = "2025-08-12T06:59:48.023Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/81/92df5673c067148c2545b1bfe49adfd775bcc3a169a047f5a0e6575ddaca/sentencepiece-0.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:4cdc7c36234fda305e85c32949c5211faaf8dd886096c7cea289ddc12a2d02de", size = 1054671, upload-time = "2025-08-12T06:59:49.895Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/02/c5e3bc518655d714622bec87d83db9cdba1cd0619a4a04e2109751c4f47f/sentencepiece-0.2.1-cp312-cp312-win_arm64.whl", hash = "sha256:daeb5e9e9fcad012324807856113708614d534f596d5008638eb9b40112cd9e4", size = 1033923, upload-time = "2025-08-12T06:59:51.952Z" },
+    { url = "https://files.pythonhosted.org/packages/98/df/76390cc0bb812687f5fa7574b555354197b8e8fdb9aaf5c9c28ae58e148b/sentencepiece-0.2.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:afefe50a0cdcb4f2fd9733cb52001a2c164181ee2d82c32d38f5b1b326a8528c", size = 1942582, upload-time = "2025-08-12T06:58:59.807Z" },
+    { url = "https://files.pythonhosted.org/packages/24/51/ff7ff849a7f139535d0309fe2a351379459bac5332127eb9a3a955ef2847/sentencepiece-0.2.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:891ade6503dd93d418c03993f7d6a8aa20260c422cefff5096b9068185e67642", size = 1325367, upload-time = "2025-08-12T06:59:01.677Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/29/bca460c34ebc79a1f3706d0eebceecd1187b2b9b93bc211f177e3b520eb0/sentencepiece-0.2.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:814978ac05130dd5812b4b03215c766bc6abaef13e7bd72bc534e4d1e12e9a4c", size = 1253366, upload-time = "2025-08-12T06:59:03.252Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/98/f279374c1e86d2a735c7d063d5b800608fc7b8370319121a9fc501cf9091/sentencepiece-0.2.1-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:017f97b274d4b0baa84b2dc743bf4517be81156f413bb24f12aacacde378e5ab", size = 1316437, upload-time = "2025-08-12T06:59:04.832Z" },
+    { url = "https://files.pythonhosted.org/packages/32/d7/efd7e172875f1329d8ae43e55f853776e332542ecb881a11a6b67d49c98c/sentencepiece-0.2.1-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:22c4ebcb3c6ab1496ab1c37c79ef7bb563b8726f29548c30773b7a4cb152df1a", size = 1387621, upload-time = "2025-08-12T06:59:06.445Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/a5/cc46d5308e1056d756b20211f039f609c86d3073aa66088cd5af57a652fa/sentencepiece-0.2.1-cp39-cp39-win32.whl", hash = "sha256:caa4e560c72c151da80036aecc2159e51a7fd8ae9efebefd96860460ce6bd025", size = 999563, upload-time = "2025-08-12T06:59:07.833Z" },
+    { url = "https://files.pythonhosted.org/packages/87/06/94bc50dc6b113bbf5f4ba7bfe85cfa7d62c679f9fb68f658d48f3ea556cf/sentencepiece-0.2.1-cp39-cp39-win_amd64.whl", hash = "sha256:2af5a1fb05013332ad94343b8b5f3973e006a2dde2dfba55a819549e054e2f0f", size = 1054663, upload-time = "2025-08-12T06:59:09.234Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/21/55267fcdc1a99612248c299c12d8f7c51d6e5ae992f32694d17adcf13fda/sentencepiece-0.2.1-cp39-cp39-win_arm64.whl", hash = "sha256:3d165fbb9bf8fba35f1946ba2617c3f9995679f07438325f07c026d53f33e746", size = 1033914, upload-time = "2025-08-12T06:59:10.705Z" },
+]
+
+[[package]]
+name = "setuptools"
+version = "80.9.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/18/5d/3bf57dcd21979b887f014ea83c24ae194cfcd12b9e0fda66b957c69d1fca/setuptools-80.9.0.tar.gz", hash = "sha256:f36b47402ecde768dbfafc46e8e4207b4360c654f1f3bb84475f0a28628fb19c", size = 1319958, upload-time = "2025-05-27T00:56:51.443Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a3/dc/17031897dae0efacfea57dfd3a82fdd2a2aeb58e0ff71b77b87e44edc772/setuptools-80.9.0-py3-none-any.whl", hash = "sha256:062d34222ad13e0cc312a4c02d73f059e86a4acbfbdea8f8f76b28c99f306922", size = 1201486, upload-time = "2025-05-27T00:56:49.664Z" },
+]
+
+[[package]]
+name = "shortuuid"
+version = "1.0.13"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8c/e2/bcf761f3bff95856203f9559baf3741c416071dd200c0fc19fad7f078f86/shortuuid-1.0.13.tar.gz", hash = "sha256:3bb9cf07f606260584b1df46399c0b87dd84773e7b25912b7e391e30797c5e72", size = 9662, upload-time = "2024-03-11T20:11:06.879Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c0/44/21d6bf170bf40b41396480d8d49ad640bca3f2b02139cd52aa1e272830a5/shortuuid-1.0.13-py3-none-any.whl", hash = "sha256:a482a497300b49b4953e15108a7913244e1bb0d41f9d332f5e9925dba33a3c5a", size = 10529, upload-time = "2024-03-11T20:11:04.807Z" },
+]
+
+[[package]]
+name = "simplejson"
+version = "3.20.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/af/92/51b417685abd96b31308b61b9acce7ec50d8e1de8fbc39a7fd4962c60689/simplejson-3.20.1.tar.gz", hash = "sha256:e64139b4ec4f1f24c142ff7dcafe55a22b811a74d86d66560c8815687143037d", size = 85591, upload-time = "2025-02-15T05:18:53.15Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/25/c4/627214fb418cd4a17fb0230ff0b6c3bb4a85cbb48dd69c85dcc3b85df828/simplejson-3.20.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e580aa65d5f6c3bf41b9b4afe74be5d5ddba9576701c107c772d936ea2b5043a", size = 93790, upload-time = "2025-02-15T05:15:32.954Z" },
+    { url = "https://files.pythonhosted.org/packages/15/ca/56a6a2a33cbcf330c4d71af3f827c47e4e0ba791e78f2642f3d1ab02ff31/simplejson-3.20.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:4a586ce4f78cec11f22fe55c5bee0f067e803aab9bad3441afe2181693b5ebb5", size = 75707, upload-time = "2025-02-15T05:15:34.954Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/c8/3d92b67e03a3b6207d97202669f9454ed700b35ade9bd4428265a078fb6c/simplejson-3.20.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:74a1608f9e6e8c27a4008d70a54270868306d80ed48c9df7872f9f4b8ac87808", size = 75700, upload-time = "2025-02-15T05:15:37.144Z" },
+    { url = "https://files.pythonhosted.org/packages/74/30/20001219d6fdca4aaa3974c96dfb6955a766b4e2cc950505a5b51fd050b0/simplejson-3.20.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:03db8cb64154189a92a7786209f24e391644f3a3fa335658be2df2af1960b8d8", size = 138672, upload-time = "2025-02-15T05:15:38.547Z" },
+    { url = "https://files.pythonhosted.org/packages/21/47/50157810876c2a7ebbd6e6346ec25eda841fe061fecaa02538a7742a3d2a/simplejson-3.20.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:eea7e2b7d858f6fdfbf0fe3cb846d6bd8a45446865bc09960e51f3d473c2271b", size = 146616, upload-time = "2025-02-15T05:15:39.871Z" },
+    { url = "https://files.pythonhosted.org/packages/95/60/8c97cdc93096437b0aca2745aca63c880fe2315fd7f6a6ce6edbb344a2ae/simplejson-3.20.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e66712b17d8425bb7ff8968d4c7c7fd5a2dd7bd63728b28356223c000dd2f91f", size = 134344, upload-time = "2025-02-15T05:15:42.091Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/9e/da184f0e9bb3a5d7ffcde713bd41b4fe46cca56b6f24d9bd155fac56805a/simplejson-3.20.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a2cc4f6486f9f515b62f5831ff1888886619b84fc837de68f26d919ba7bbdcbc", size = 138017, upload-time = "2025-02-15T05:15:43.542Z" },
+    { url = "https://files.pythonhosted.org/packages/31/db/00d1a8d9b036db98f678c8a3c69ed17d2894d1768d7a00576e787ad3e546/simplejson-3.20.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:a3c2df555ee4016148fa192e2b9cd9e60bc1d40769366134882685e90aee2a1e", size = 140118, upload-time = "2025-02-15T05:15:45.7Z" },
+    { url = "https://files.pythonhosted.org/packages/52/21/57fc47eab8c1c73390b933a5ba9271f08e3e1ec83162c580357f28f5b97c/simplejson-3.20.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:78520f04b7548a5e476b5396c0847e066f1e0a4c0c5e920da1ad65e95f410b11", size = 140314, upload-time = "2025-02-15T05:16:07.949Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/cc/7cfd78d1e0fa5e57350b98cfe77353b6dfa13dce21afa4060e1019223852/simplejson-3.20.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:f4bd49ecde87b0fe9f55cc971449a32832bca9910821f7072bbfae1155eaa007", size = 148544, upload-time = "2025-02-15T05:16:09.455Z" },
+    { url = "https://files.pythonhosted.org/packages/63/26/1c894a1c2bd95dc8be0cf5a2fa73b0d173105b6ca18c90cb981ff10443d0/simplejson-3.20.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:7eaae2b88eb5da53caaffdfa50e2e12022553949b88c0df4f9a9663609373f72", size = 141172, upload-time = "2025-02-15T05:16:10.966Z" },
+    { url = "https://files.pythonhosted.org/packages/93/27/0717dccc10cd9988dbf1314def52ab32678a95a95328bb37cafacf499400/simplejson-3.20.1-cp310-cp310-win32.whl", hash = "sha256:e836fb88902799eac8debc2b642300748f4860a197fa3d9ea502112b6bb8e142", size = 74181, upload-time = "2025-02-15T05:16:12.361Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/af/593f896573f306519332d4287b1ab8b7b888c239bbd5159f7054d7055c2d/simplejson-3.20.1-cp310-cp310-win_amd64.whl", hash = "sha256:b122a19b552b212fc3b5b96fc5ce92333d4a9ac0a800803e1f17ebb16dac4be5", size = 75738, upload-time = "2025-02-15T05:16:14.438Z" },
+    { url = "https://files.pythonhosted.org/packages/76/59/74bc90d1c051bc2432c96b34bd4e8036875ab58b4fcbe4d6a5a76985f853/simplejson-3.20.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:325b8c107253d3217e89d7b50c71015b5b31e2433e6c5bf38967b2f80630a8ca", size = 92132, upload-time = "2025-02-15T05:16:15.743Z" },
+    { url = "https://files.pythonhosted.org/packages/71/c7/1970916e0c51794fff89f76da2f632aaf0b259b87753c88a8c409623d3e1/simplejson-3.20.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:88a7baa8211089b9e58d78fbc1b0b322103f3f3d459ff16f03a36cece0d0fcf0", size = 74956, upload-time = "2025-02-15T05:16:17.062Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/0d/98cc5909180463f1d75fac7180de62d4cdb4e82c4fef276b9e591979372c/simplejson-3.20.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:299b1007b8101d50d95bc0db1bf5c38dc372e85b504cf77f596462083ee77e3f", size = 74772, upload-time = "2025-02-15T05:16:19.204Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/94/a30a5211a90d67725a3e8fcc1c788189f2ae2ed2b96b63ed15d0b7f5d6bb/simplejson-3.20.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:03ec618ed65caab48e81e3ed29586236a8e57daef792f1f3bb59504a7e98cd10", size = 143575, upload-time = "2025-02-15T05:16:21.337Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/08/cdb6821f1058eb5db46d252de69ff7e6c53f05f1bae6368fe20d5b51d37e/simplejson-3.20.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cd2cdead1d3197f0ff43373cf4730213420523ba48697743e135e26f3d179f38", size = 153241, upload-time = "2025-02-15T05:16:22.859Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/2d/ca3caeea0bdc5efc5503d5f57a2dfb56804898fb196dfada121323ee0ccb/simplejson-3.20.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3466d2839fdc83e1af42e07b90bc8ff361c4e8796cd66722a40ba14e458faddd", size = 141500, upload-time = "2025-02-15T05:16:25.068Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/33/d3e0779d5c58245e7370c98eb969275af6b7a4a5aec3b97cbf85f09ad328/simplejson-3.20.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d492ed8e92f3a9f9be829205f44b1d0a89af6582f0cf43e0d129fa477b93fe0c", size = 144757, upload-time = "2025-02-15T05:16:28.301Z" },
+    { url = "https://files.pythonhosted.org/packages/54/53/2d93128bb55861b2fa36c5944f38da51a0bc6d83e513afc6f7838440dd15/simplejson-3.20.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:f924b485537b640dc69434565463fd6fc0c68c65a8c6e01a823dd26c9983cf79", size = 144409, upload-time = "2025-02-15T05:16:29.687Z" },
+    { url = "https://files.pythonhosted.org/packages/99/4c/dac310a98f897ad3435b4bdc836d92e78f09e38c5dbf28211ed21dc59fa2/simplejson-3.20.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:9e8eacf6a3491bf76ea91a8d46726368a6be0eb94993f60b8583550baae9439e", size = 146082, upload-time = "2025-02-15T05:16:31.064Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/22/d7ba958cfed39827335b82656b1c46f89678faecda9a7677b47e87b48ee6/simplejson-3.20.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:d34d04bf90b4cea7c22d8b19091633908f14a096caa301b24c2f3d85b5068fb8", size = 154339, upload-time = "2025-02-15T05:16:32.719Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/c8/b072b741129406a7086a0799c6f5d13096231bf35fdd87a0cffa789687fc/simplejson-3.20.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:69dd28d4ce38390ea4aaf212902712c0fd1093dc4c1ff67e09687c3c3e15a749", size = 147915, upload-time = "2025-02-15T05:16:34.291Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/46/8347e61e9cf3db5342a42f7fd30a81b4f5cf85977f916852d7674a540907/simplejson-3.20.1-cp311-cp311-win32.whl", hash = "sha256:dfe7a9da5fd2a3499436cd350f31539e0a6ded5da6b5b3d422df016444d65e43", size = 73972, upload-time = "2025-02-15T05:16:35.712Z" },
+    { url = "https://files.pythonhosted.org/packages/01/85/b52f24859237b4e9d523d5655796d911ba3d46e242eb1959c45b6af5aedd/simplejson-3.20.1-cp311-cp311-win_amd64.whl", hash = "sha256:896a6c04d7861d507d800da7642479c3547060bf97419d9ef73d98ced8258766", size = 75595, upload-time = "2025-02-15T05:16:36.957Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/eb/34c16a1ac9ba265d024dc977ad84e1659d931c0a700967c3e59a98ed7514/simplejson-3.20.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:f31c4a3a7ab18467ee73a27f3e59158255d1520f3aad74315edde7a940f1be23", size = 93100, upload-time = "2025-02-15T05:16:38.801Z" },
+    { url = "https://files.pythonhosted.org/packages/41/fc/2c2c007d135894971e6814e7c0806936e5bade28f8db4dd7e2a58b50debd/simplejson-3.20.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:884e6183d16b725e113b83a6fc0230152ab6627d4d36cb05c89c2c5bccfa7bc6", size = 75464, upload-time = "2025-02-15T05:16:40.905Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/05/2b5ecb33b776c34bb5cace5de5d7669f9b60e3ca13c113037b2ca86edfbd/simplejson-3.20.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:03d7a426e416fe0d3337115f04164cd9427eb4256e843a6b8751cacf70abc832", size = 75112, upload-time = "2025-02-15T05:16:42.246Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/36/1f3609a2792f06cd4b71030485f78e91eb09cfd57bebf3116bf2980a8bac/simplejson-3.20.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:000602141d0bddfcff60ea6a6e97d5e10c9db6b17fd2d6c66199fa481b6214bb", size = 150182, upload-time = "2025-02-15T05:16:43.557Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/b0/053fbda38b8b602a77a4f7829def1b4f316cd8deb5440a6d3ee90790d2a4/simplejson-3.20.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:af8377a8af78226e82e3a4349efdde59ffa421ae88be67e18cef915e4023a595", size = 158363, upload-time = "2025-02-15T05:16:45.748Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/4b/2eb84ae867539a80822e92f9be4a7200dffba609275faf99b24141839110/simplejson-3.20.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:15c7de4c88ab2fbcb8781a3b982ef883696736134e20b1210bca43fb42ff1acf", size = 148415, upload-time = "2025-02-15T05:16:47.861Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/bd/400b0bd372a5666addf2540c7358bfc3841b9ce5cdbc5cc4ad2f61627ad8/simplejson-3.20.1-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:455a882ff3f97d810709f7b620007d4e0aca8da71d06fc5c18ba11daf1c4df49", size = 152213, upload-time = "2025-02-15T05:16:49.25Z" },
+    { url = "https://files.pythonhosted.org/packages/50/12/143f447bf6a827ee9472693768dc1a5eb96154f8feb140a88ce6973a3cfa/simplejson-3.20.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:fc0f523ce923e7f38eb67804bc80e0a028c76d7868500aa3f59225574b5d0453", size = 150048, upload-time = "2025-02-15T05:16:51.5Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/ea/dd9b3e8e8ed710a66f24a22c16a907c9b539b6f5f45fd8586bd5c231444e/simplejson-3.20.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:76461ec929282dde4a08061071a47281ad939d0202dc4e63cdd135844e162fbc", size = 151668, upload-time = "2025-02-15T05:16:53Z" },
+    { url = "https://files.pythonhosted.org/packages/99/af/ee52a8045426a0c5b89d755a5a70cc821815ef3c333b56fbcad33c4435c0/simplejson-3.20.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:ab19c2da8c043607bde4d4ef3a6b633e668a7d2e3d56f40a476a74c5ea71949f", size = 158840, upload-time = "2025-02-15T05:16:54.851Z" },
+    { url = "https://files.pythonhosted.org/packages/68/db/ab32869acea6b5de7d75fa0dac07a112ded795d41eaa7e66c7813b17be95/simplejson-3.20.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b2578bedaedf6294415197b267d4ef678fea336dd78ee2a6d2f4b028e9d07be3", size = 154212, upload-time = "2025-02-15T05:16:56.318Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/7a/e3132d454977d75a3bf9a6d541d730f76462ebf42a96fea2621498166f41/simplejson-3.20.1-cp312-cp312-win32.whl", hash = "sha256:339f407373325a36b7fd744b688ba5bae0666b5d340ec6d98aebc3014bf3d8ea", size = 74101, upload-time = "2025-02-15T05:16:57.746Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/5d/4e243e937fa3560107c69f6f7c2eed8589163f5ed14324e864871daa2dd9/simplejson-3.20.1-cp312-cp312-win_amd64.whl", hash = "sha256:627d4486a1ea7edf1f66bb044ace1ce6b4c1698acd1b05353c97ba4864ea2e17", size = 75736, upload-time = "2025-02-15T05:16:59.017Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/ba/d32fe890a5edaf4a8518adf043bccf7866b600123f512a6de0988cf36810/simplejson-3.20.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:a8011f1dd1d676befcd4d675ebdbfdbbefd3bf350052b956ba8c699fca7d8cef", size = 93773, upload-time = "2025-02-15T05:18:28.231Z" },
+    { url = "https://files.pythonhosted.org/packages/48/c7/361e7f6695b56001a04e0a5cc623cd6c82ea2f45e872e61213e405cc8a24/simplejson-3.20.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:e91703a4c5fec53e36875ae426ad785f4120bd1d93b65bed4752eeccd1789e0c", size = 75697, upload-time = "2025-02-15T05:18:30.006Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/2f/d0ff0b772d4ef092876eb85c99bc591c446b0502715551dad7dfc7f7c2c0/simplejson-3.20.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e39eaa57c7757daa25bcd21f976c46be443b73dd6c3da47fe5ce7b7048ccefe2", size = 75692, upload-time = "2025-02-15T05:18:31.424Z" },
+    { url = "https://files.pythonhosted.org/packages/26/94/cab4db9530b6ca9d62f16a260e8311b04130ccd670dab75e958fcb44590e/simplejson-3.20.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ceab2ce2acdc7fbaa433a93006758db6ba9a659e80c4faa13b80b9d2318e9b17", size = 138106, upload-time = "2025-02-15T05:18:32.907Z" },
+    { url = "https://files.pythonhosted.org/packages/40/22/11c0f746bdb44c297cea8a37d8f7ccb75ea6681132aadfb9f820d9a52647/simplejson-3.20.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6d4f320c33277a5b715db5bf5b10dae10c19076bd6d66c2843e04bd12d1f1ea5", size = 146242, upload-time = "2025-02-15T05:18:35.223Z" },
+    { url = "https://files.pythonhosted.org/packages/78/e9/b7c4c26f29b41cc41ba5f0224c47adbfa7f28427418edfd58ab122f3b584/simplejson-3.20.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2b6436c48e64378fa844d8c9e58a5ed0352bbcfd4028369a9b46679b7ab79d2d", size = 133866, upload-time = "2025-02-15T05:18:36.998Z" },
+    { url = "https://files.pythonhosted.org/packages/09/68/1e81ed83f38906c8859f2b973afb19302357d6003e724a6105cee0f61ec7/simplejson-3.20.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6e18345c8dda5d699be8166b61f9d80aaee4545b709f1363f60813dc032dac53", size = 137444, upload-time = "2025-02-15T05:18:38.763Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/6b/8d1e076c543277c1d603230eec24f4dd75ebce46d351c0679526d202981f/simplejson-3.20.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:90b573693d1526bed576f6817e2a492eaaef68f088b57d7a9e83d122bbb49e51", size = 139617, upload-time = "2025-02-15T05:18:40.36Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/46/7b74803de10d4157c5cd2e89028897fa733374667bc5520a44b23b6c887a/simplejson-3.20.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:272cc767826e924a6bd369ea3dbf18e166ded29059c7a4d64d21a9a22424b5b5", size = 139725, upload-time = "2025-02-15T05:18:42.012Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/8f/9991582665a7b6d95415e439bb4fbaa4faf0f77231666675a0fd1de54107/simplejson-3.20.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:51b41f284d603c4380732d7d619f8b34bd04bc4aa0ed0ed5f4ffd0539b14da44", size = 148010, upload-time = "2025-02-15T05:18:43.749Z" },
+    { url = "https://files.pythonhosted.org/packages/54/ee/3c6e91989cdf65ec75e75662d9f15cfe167a792b893806169ea5b1da6fd2/simplejson-3.20.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:6e6697a3067d281f01de0fe96fc7cba4ea870d96d7deb7bfcf85186d74456503", size = 140624, upload-time = "2025-02-15T05:18:45.498Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/bd/05e13ebb7ead81c8b555f4ccc741ea7dfa0ef5c2a0c183d6a7bc50a02bca/simplejson-3.20.1-cp39-cp39-win32.whl", hash = "sha256:6dd3a1d5aca87bf947f3339b0f8e8e329f1badf548bdbff37fac63c17936da8e", size = 74148, upload-time = "2025-02-15T05:18:47.27Z" },
+    { url = "https://files.pythonhosted.org/packages/88/c9/d8bf87aaebec5a4c3ccfd5228689578e2fe77027d6114a259255d54969bf/simplejson-3.20.1-cp39-cp39-win_amd64.whl", hash = "sha256:463f1fca8fbf23d088e5850fdd0dd4d5faea8900a9f9680270bd98fd649814ca", size = 75732, upload-time = "2025-02-15T05:18:49.598Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/30/00f02a0a921556dd5a6db1ef2926a1bc7a8bbbfb1c49cfed68a275b8ab2b/simplejson-3.20.1-py3-none-any.whl", hash = "sha256:8a6c1bbac39fa4a79f83cbf1df6ccd8ff7069582a9fd8db1e52cea073bc2c697", size = 57121, upload-time = "2025-02-15T05:18:51.243Z" },
 ]
 
 [[package]]
 name = "six"
 version = "1.17.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/94/e7/b2c673351809dca68a0e064b6af791aa332cf192da575fd474ed7d6f16a2/six-1.17.0.tar.gz", hash = "sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81", size = 34031 }
+sdist = { url = "https://files.pythonhosted.org/packages/94/e7/b2c673351809dca68a0e064b6af791aa332cf192da575fd474ed7d6f16a2/six-1.17.0.tar.gz", hash = "sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81", size = 34031, upload-time = "2024-12-04T17:35:28.174Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b7/ce/149a00dd41f10bc29e5921b496af8b574d8413afcd5e30dfa0ed46c2cc5e/six-1.17.0-py2.py3-none-any.whl", hash = "sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274", size = 11050 },
+    { url = "https://files.pythonhosted.org/packages/b7/ce/149a00dd41f10bc29e5921b496af8b574d8413afcd5e30dfa0ed46c2cc5e/six-1.17.0-py2.py3-none-any.whl", hash = "sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274", size = 11050, upload-time = "2024-12-04T17:35:26.475Z" },
 ]
 
 [[package]]
 name = "sniffio"
 version = "1.3.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/a2/87/a6771e1546d97e7e041b6ae58d80074f81b7d5121207425c964ddf5cfdbd/sniffio-1.3.1.tar.gz", hash = "sha256:f4324edc670a0f49750a81b895f35c3adb843cca46f0530f79fc1babb23789dc", size = 20372 }
+sdist = { url = "https://files.pythonhosted.org/packages/a2/87/a6771e1546d97e7e041b6ae58d80074f81b7d5121207425c964ddf5cfdbd/sniffio-1.3.1.tar.gz", hash = "sha256:f4324edc670a0f49750a81b895f35c3adb843cca46f0530f79fc1babb23789dc", size = 20372, upload-time = "2024-02-25T23:20:04.057Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2", size = 10235 },
+    { url = "https://files.pythonhosted.org/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2", size = 10235, upload-time = "2024-02-25T23:20:01.196Z" },
 ]
 
 [[package]]
 name = "starlette"
-version = "0.41.3"
+version = "0.47.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
-    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/1a/4c/9b5764bd22eec91c4039ef4c55334e9187085da2d8a2df7bd570869aae18/starlette-0.41.3.tar.gz", hash = "sha256:0e4ab3d16522a255be6b28260b938eae2482f98ce5cc934cb08dce8dc3ba5835", size = 2574159 }
+sdist = { url = "https://files.pythonhosted.org/packages/15/b9/cc3017f9a9c9b6e27c5106cc10cc7904653c3eec0729793aec10479dd669/starlette-0.47.3.tar.gz", hash = "sha256:6bc94f839cc176c4858894f1f8908f0ab79dfec1a6b8402f6da9be26ebea52e9", size = 2584144, upload-time = "2025-08-24T13:36:42.122Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ce/fd/901cfa59aaa5b30a99e16876f11abe38b59a1a2c51ffb3d7142bb6089069/starlette-0.47.3-py3-none-any.whl", hash = "sha256:89c0778ca62a76b826101e7c709e70680a1699ca7da6b44d38eb0a7e61fe4b51", size = 72991, upload-time = "2025-08-24T13:36:40.887Z" },
+]
+
+[[package]]
+name = "svgwrite"
+version = "1.4.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/16/c1/263d4e93b543390d86d8eb4fc23d9ce8a8d6efd146f9427364109004fa9b/svgwrite-1.4.3.zip", hash = "sha256:a8fbdfd4443302a6619a7f76bc937fc683daf2628d9b737c891ec08b8ce524c3", size = 189516, upload-time = "2022-07-14T14:05:26.107Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/96/00/2b325970b3060c7cecebab6d295afe763365822b1306a12eeab198f74323/starlette-0.41.3-py3-none-any.whl", hash = "sha256:44cedb2b7c77a9de33a8b74b2b90e9f50d11fcf25d8270ea525ad71a25374ff7", size = 73225 },
+    { url = "https://files.pythonhosted.org/packages/84/15/640e399579024a6875918839454025bb1d5f850bb70d96a11eabb644d11c/svgwrite-1.4.3-py3-none-any.whl", hash = "sha256:bb6b2b5450f1edbfa597d924f9ac2dd099e625562e492021d7dd614f65f8a22d", size = 67122, upload-time = "2022-07-14T14:05:24.459Z" },
 ]
 
 [[package]]
 name = "sympy"
-version = "1.13.3"
+version = "1.14.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "mpmath" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/11/8a/5a7fd6284fa8caac23a26c9ddf9c30485a48169344b4bd3b0f02fef1890f/sympy-1.13.3.tar.gz", hash = "sha256:b27fd2c6530e0ab39e275fc9b683895367e51d5da91baa8d3d64db2565fec4d9", size = 7533196 }
+sdist = { url = "https://files.pythonhosted.org/packages/83/d3/803453b36afefb7c2bb238361cd4ae6125a569b4db67cd9e79846ba2d68c/sympy-1.14.0.tar.gz", hash = "sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517", size = 7793921, upload-time = "2025-04-27T18:05:01.611Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a2/09/77d55d46fd61b4a135c444fc97158ef34a095e5681d0a6c10b75bf356191/sympy-1.14.0-py3-none-any.whl", hash = "sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5", size = 6299353, upload-time = "2025-04-27T18:04:59.103Z" },
+]
+
+[[package]]
+name = "tenacity"
+version = "9.1.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0a/d4/2b0cd0fe285e14b36db076e78c93766ff1d529d70408bd1d2a5a84f1d929/tenacity-9.1.2.tar.gz", hash = "sha256:1169d376c297e7de388d18b4481760d478b0e99a777cad3a9c86e556f4b697cb", size = 48036, upload-time = "2025-04-02T08:25:09.966Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e5/30/643397144bfbfec6f6ef821f36f33e57d35946c44a2352d3c9f0ae847619/tenacity-9.1.2-py3-none-any.whl", hash = "sha256:f77bf36710d8b73a50b2dd155c97b870017ad21afe6ab300326b0371b3b05138", size = 28248, upload-time = "2025-04-02T08:25:07.678Z" },
+]
+
+[[package]]
+name = "tensorboard"
+version = "2.19.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "absl-py" },
+    { name = "grpcio" },
+    { name = "markdown" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "packaging" },
+    { name = "protobuf" },
+    { name = "setuptools" },
+    { name = "six" },
+    { name = "tensorboard-data-server" },
+    { name = "werkzeug" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/5d/12/4f70e8e2ba0dbe72ea978429d8530b0333f0ed2140cc571a48802878ef99/tensorboard-2.19.0-py3-none-any.whl", hash = "sha256:5e71b98663a641a7ce8a6e70b0be8e1a4c0c45d48760b076383ac4755c35b9a0", size = 5503412, upload-time = "2025-02-12T08:17:27.21Z" },
+]
+
+[[package]]
+name = "tensorboard-data-server"
+version = "0.7.2"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7a/13/e503968fefabd4c6b2650af21e110aa8466fe21432cd7c43a84577a89438/tensorboard_data_server-0.7.2-py3-none-any.whl", hash = "sha256:7e0610d205889588983836ec05dc098e80f97b7e7bbff7e994ebb78f578d0ddb", size = 2356, upload-time = "2023-10-23T21:23:32.16Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/85/dabeaf902892922777492e1d253bb7e1264cadce3cea932f7ff599e53fea/tensorboard_data_server-0.7.2-py3-none-macosx_10_9_x86_64.whl", hash = "sha256:9fe5d24221b29625dbc7328b0436ca7fc1c23de4acf4d272f1180856e32f9f60", size = 4823598, upload-time = "2023-10-23T21:23:33.714Z" },
+    { url = "https://files.pythonhosted.org/packages/73/c6/825dab04195756cf8ff2e12698f22513b3db2f64925bdd41671bfb33aaa5/tensorboard_data_server-0.7.2-py3-none-manylinux_2_31_x86_64.whl", hash = "sha256:ef687163c24185ae9754ed5650eb5bc4d84ff257aabdc33f0cc6f74d8ba54530", size = 6590363, upload-time = "2023-10-23T21:23:35.583Z" },
+]
+
+[[package]]
+name = "tensorflow"
+version = "2.19.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "absl-py" },
+    { name = "astunparse" },
+    { name = "flatbuffers" },
+    { name = "gast" },
+    { name = "google-pasta" },
+    { name = "grpcio" },
+    { name = "h5py" },
+    { name = "keras", version = "3.10.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "keras", version = "3.11.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "libclang" },
+    { name = "ml-dtypes" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "opt-einsum" },
+    { name = "packaging" },
+    { name = "protobuf" },
+    { name = "requests" },
+    { name = "setuptools" },
+    { name = "six" },
+    { name = "tensorboard" },
+    { name = "tensorflow-io-gcs-filesystem", marker = "python_full_version < '3.12'" },
+    { name = "termcolor" },
+    { name = "typing-extensions" },
+    { name = "wrapt" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cb/6e/8613ef1cea8013b899adae32bb7cc5218434d84a3a2d6f41937e6d7251af/tensorflow-2.19.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:ed3148ee523cb46badd30ea04a6b7fb6056b9bdb7e082de294a86b2fdc01f9e4", size = 252466041, upload-time = "2025-08-13T16:27:10.511Z" },
+    { url = "https://files.pythonhosted.org/packages/65/4a/cdbcef3a959a2eddf5d3ee7a355454e8d18c89a2e197e96a1b4f55e58d86/tensorflow-2.19.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9077404d1928ee58bfb82ea1b0fce25353420338c259742227c27c4e00377f4b", size = 252498918, upload-time = "2025-08-13T16:27:19.812Z" },
+    { url = "https://files.pythonhosted.org/packages/77/60/51d921b17f7b1547db32018d6a933627d4e2a762d2fc2ca6c0032cc8b062/tensorflow-2.19.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:11377d4cecdc665515e370524b650a8ceb80b461f40f5e63993ed27d0ef931da", size = 644754761, upload-time = "2025-08-13T16:27:36.011Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/80/b74fef28ea980bf6a65dcd9cdc55e2c205b1fa7e8c253a6ddf9f260d990e/tensorflow-2.19.1-cp310-cp310-win_amd64.whl", hash = "sha256:a61c209ec594b2ed4ee2d55a905945bd748df6a1d73ab7ccecf7a9234f52045f", size = 375725444, upload-time = "2025-08-13T16:27:55.406Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/91/70ff01474ff0f2266f47ab292819b97088b512d7e9093e416d16c7aac5ea/tensorflow-2.19.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:8a274a1336051bef0e7e042fa5818a9d342e300d9c311c5d7933eb60e2269770", size = 252591021, upload-time = "2025-08-13T16:28:08.623Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/c6/fc6416ac841106f3b6f2b4d16f50793a71d675c3558c482fd4b3cb44f95e/tensorflow-2.19.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:22deedc251837f2382d2b287c705705cb089bfaf79b2274b78c08740e02bf188", size = 252660426, upload-time = "2025-08-13T16:28:17.267Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/58/34f49a802d6d1f1eb6fe89bc9ff52e2bee19d5c3461059a55702fb7e5fb3/tensorflow-2.19.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8e5524eb492160576e5eeea9bb472ddc653123503f2290f83123d5ac9545750e", size = 644896980, upload-time = "2025-08-13T16:28:32.949Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/f7/b0f6e55174217a59e0b850909dc2691cb8afd8f7b38a99151d0a0d81e3ef/tensorflow-2.19.1-cp311-cp311-win_amd64.whl", hash = "sha256:1729bc2ae90a100a87d5a462b5e9b85a088261a03f0a7179eebb8735938bed57", size = 375912087, upload-time = "2025-08-13T16:28:49.636Z" },
+    { url = "https://files.pythonhosted.org/packages/73/c9/7f6430b944b3683d448f2b3ddace47701021116f6891ffc4926ebb957a88/tensorflow-2.19.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:02c69369a82709827c6a5218c6697ee94dcd1397230319cc17c47a3342249354", size = 252656033, upload-time = "2025-08-13T16:29:05.097Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/3e/717bb5b6ec69de7c08917326470432156b5d0fe888e43775de95d8322b07/tensorflow-2.19.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4c0e78b6100545fd90934c557f1ce802e6a9a7fdb031025efc298efb4aa2e3bb", size = 252718124, upload-time = "2025-08-13T16:29:15.802Z" },
+    { url = "https://files.pythonhosted.org/packages/50/96/b33fdf567332f65e78f02df3a69fbea911eec328e81dc4941d890f0777f5/tensorflow-2.19.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:35a6950f3fee0ad2d1668241d31684676f2e6eceb613cbeef960ea22014ef8fb", size = 645010075, upload-time = "2025-08-13T16:29:37.407Z" },
+    { url = "https://files.pythonhosted.org/packages/27/be/e303c6b55d3bc1d392c555f3e00773a5b2a1c82c8f5af052148e3f0d60ee/tensorflow-2.19.1-cp312-cp312-win_amd64.whl", hash = "sha256:14ec6dee7ab3b241aae3537616e6f1ed1c7c67eff5d853e586eb66a2e9276ccc", size = 375963361, upload-time = "2025-08-13T16:29:54.216Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/a2/cb060b8944a03321efbf1a56cfe49144e5ae648fbe68f2eff275ca297feb/tensorflow-2.19.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:04f3b4dbb3ce4b507d853decfb2d116f8d0344d23413e2ad889e1e069fbeb54d", size = 252468489, upload-time = "2025-08-13T16:30:05.684Z" },
+    { url = "https://files.pythonhosted.org/packages/50/0f/e457ff412f20226516bebaeaf28f6a86531c15dde39fbf4f49b849e30cd7/tensorflow-2.19.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:938fb49b3299a77ccdaa26c0824dfcca3002b260c7bc994cb129d4ec12f31619", size = 252500289, upload-time = "2025-08-13T16:30:15.237Z" },
+    { url = "https://files.pythonhosted.org/packages/93/45/d9480e6140fdbdff8400c132c99a221787d66803739bc8664d3ec3f70633/tensorflow-2.19.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7a66b66fcc3eee0cc4d6c405cb5918972afeeb148684e174693d5d0449c3fdaf", size = 644755371, upload-time = "2025-08-13T16:30:31.327Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/8a/a854c04228d41861c77493a7da5510c9b0969fe44c7903745afb8437188a/tensorflow-2.19.1-cp39-cp39-win_amd64.whl", hash = "sha256:7e770ce558ceeed129537683d3bc7ec9214de8bb296354e1c4198169f4deda1f", size = 375723379, upload-time = "2025-08-13T16:30:51.58Z" },
+]
+
+[[package]]
+name = "tensorflow-io-gcs-filesystem"
+version = "0.37.1"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e9/a3/12d7e7326a707919b321e2d6e4c88eb61596457940fd2b8ff3e9b7fac8a7/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:249c12b830165841411ba71e08215d0e94277a49c551e6dd5d72aab54fe5491b", size = 2470224, upload-time = "2024-07-01T23:44:15.341Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/55/3849a188cc15e58fefde20e9524d124a629a67a06b4dc0f6c881cb3c6e39/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:257aab23470a0796978efc9c2bcf8b0bc80f22e6298612a4c0a50d3f4e88060c", size = 3479613, upload-time = "2024-07-01T23:44:17.445Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/19/9095c69e22c879cb3896321e676c69273a549a3148c4f62aa4bc5ebdb20f/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8febbfcc67c61e542a5ac1a98c7c20a91a5e1afc2e14b1ef0cb7c28bc3b6aa70", size = 4842078, upload-time = "2024-07-01T23:44:18.977Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/48/47b7d25572961a48b1de3729b7a11e835b888e41e0203cca82df95d23b91/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9679b36e3a80921876f31685ab6f7270f3411a4cc51bc2847e80d0e4b5291e27", size = 5085736, upload-time = "2024-07-01T23:44:21.034Z" },
+    { url = "https://files.pythonhosted.org/packages/40/9b/b2fb82d0da673b17a334f785fc19c23483165019ddc33b275ef25ca31173/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:32c50ab4e29a23c1f91cd0f9ab8c381a0ab10f45ef5c5252e94965916041737c", size = 2470224, upload-time = "2024-07-01T23:44:23.039Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/cc/16634e76f3647fbec18187258da3ba11184a6232dcf9073dc44579076d36/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:b02f9c5f94fd62773954a04f69b68c4d576d076fd0db4ca25d5479f0fbfcdbad", size = 3479613, upload-time = "2024-07-01T23:44:24.399Z" },
+    { url = "https://files.pythonhosted.org/packages/de/bf/ba597d3884c77d05a78050f3c178933d69e3f80200a261df6eaa920656cd/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6e1f2796b57e799a8ca1b75bf47c2aaa437c968408cc1a402a9862929e104cda", size = 4842079, upload-time = "2024-07-01T23:44:26.825Z" },
+    { url = "https://files.pythonhosted.org/packages/66/7f/e36ae148c2f03d61ca1bff24bc13a0fef6d6825c966abef73fc6f880a23b/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ee7c8ee5fe2fd8cb6392669ef16e71841133041fee8a330eff519ad9b36e4556", size = 5085736, upload-time = "2024-07-01T23:44:28.618Z" },
+    { url = "https://files.pythonhosted.org/packages/70/83/4422804257fe2942ae0af4ea5bcc9df59cb6cb1bd092202ef240751d16aa/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:ffebb6666a7bfc28005f4fbbb111a455b5e7d6cd3b12752b7050863ecb27d5cc", size = 2470224, upload-time = "2024-07-01T23:44:30.232Z" },
+    { url = "https://files.pythonhosted.org/packages/43/9b/be27588352d7bd971696874db92d370f578715c17c0ccb27e4b13e16751e/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:fe8dcc6d222258a080ac3dfcaaaa347325ce36a7a046277f6b3e19abc1efb3c5", size = 3479614, upload-time = "2024-07-01T23:44:32.316Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/46/962f47af08bd39fc9feb280d3192825431a91a078c856d17a78ae4884eb1/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fbb33f1745f218464a59cecd9a18e32ca927b0f4d77abd8f8671b645cc1a182f", size = 4842077, upload-time = "2024-07-01T23:44:33.86Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/9b/790d290c232bce9b691391cf16e95a96e469669c56abfb1d9d0f35fa437c/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:286389a203a5aee1a4fa2e53718c661091aa5fea797ff4fa6715ab8436b02e6c", size = 5085733, upload-time = "2024-07-01T23:44:36.663Z" },
+    { url = "https://files.pythonhosted.org/packages/12/4f/798df777498fab9dc683a658688e962f0af56454eb040c90f836fd9fa67c/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:ee5da49019670ed364f3e5fb86b46420841a6c3cb52a300553c63841671b3e6d", size = 2470221, upload-time = "2024-07-01T23:44:39.105Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/f9/ce6a0efde262a79361f0d67392fdf0d0406781a1ee4fc48d0d8b0553b311/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:8943036bbf84e7a2be3705cb56f9c9df7c48c9e614bb941f0936c58e3ca89d6f", size = 3479613, upload-time = "2024-07-01T23:44:40.583Z" },
+    { url = "https://files.pythonhosted.org/packages/66/5f/334a011caa1eb97689274d1141df8e6b7a25e389f0390bdcd90235de9783/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:426de1173cb81fbd62becec2012fc00322a295326d90eb6c737fab636f182aed", size = 4842075, upload-time = "2024-07-01T23:44:42.094Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/cb/7dcee55fc5a7d7d8a862e12519322851cd5fe5b086f946fd71e4ae1ef281/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0df00891669390078a003cedbdd3b8e645c718b111917535fa1d7725e95cdb95", size = 5087496, upload-time = "2024-07-01T23:44:43.797Z" },
+]
+
+[[package]]
+name = "tensorstore"
+version = "0.1.69"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "ml-dtypes", marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/12/0e/03b2797d25f70b30c807bfef1aa8ce09731e7a82a406ba84bb67ddfc34df/tensorstore-0.1.69.tar.gz", hash = "sha256:150cdc7e2044b7629ea466bfc8425ab50e52c8300950dbdd4a64445a2d4fbab1", size = 6585496, upload-time = "2024-11-25T23:09:43.073Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f6/34/1b0821a0389cd4a10fefde2ef29e067b619dcecf3f6df548ad59f74bb554/tensorstore-0.1.69-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:9b6bf6938a3a9ac1a415d06479496f0e3180d487be6e218d223e53203b12c208", size = 14836993, upload-time = "2024-11-25T23:08:42.902Z" },
+    { url = "https://files.pythonhosted.org/packages/47/d7/f13a89ccf22a025456b34f188ffbd1a8abf8c834b389c7b84f1a6c9aff1e/tensorstore-0.1.69-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0a030706644c501aba8def8c453a26148d1c76e11a1020a3aa6a2737d0d8f982", size = 12820268, upload-time = "2024-11-25T23:08:45.69Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/2a/699156262efcbcfc2091fd0f8485c97b090938ea91637f0fe6c7cc93dc88/tensorstore-0.1.69-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1b0b9dce8d35e74bfb7d1f96ef86dc6f7de6a8ce13dbb2166fbc30ec19fdbbe7", size = 16204089, upload-time = "2024-11-25T23:08:48.26Z" },
+    { url = "https://files.pythonhosted.org/packages/96/e0/bd20ddda24e36e3cb68b674a8d320b7208ab9c7b90a61b159c2471468e11/tensorstore-0.1.69-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b796eae8466fe8a4b9f5636e4ea68ead47dd50454d95f27365ffed63ab73e64", size = 17534514, upload-time = "2024-11-25T23:08:51.558Z" },
+    { url = "https://files.pythonhosted.org/packages/62/f9/af4523a5cc0aeb5aa4ffd4bb0c381ca004354022c8cb7994afd7f213f138/tensorstore-0.1.69-cp310-cp310-win_amd64.whl", hash = "sha256:26c8bbc03537d236d393e816872d2ee8166f1b9662e6dd6bc36b38588884f6b7", size = 11997600, upload-time = "2024-11-25T23:08:54.824Z" },
+    { url = "https://files.pythonhosted.org/packages/15/8b/e71ae481056d0429940ab23a7b2bb1faf2a2a63a4f2a3fd2c7738e43275b/tensorstore-0.1.69-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:760f9841e5694ceb2be2ffc93961646de8e60e147d150322f95a396c2b063785", size = 14839086, upload-time = "2024-11-25T23:08:58.083Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/8c/d5ba205c47c54c71c137dddc75d9f4c16020ab05700f9f9314c958d34f0d/tensorstore-0.1.69-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:819ccd6d68cb5c1a60ac2cd1fa6f89d5797ce90b795493acf4e45a4242ae1738", size = 12820531, upload-time = "2024-11-25T23:09:00.498Z" },
+    { url = "https://files.pythonhosted.org/packages/32/a7/76deb854937fcdb1f56c18ee5ed261f6520c2ca1b22d330a4e337533d2af/tensorstore-0.1.69-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e99ed4bf3c81435e5348c2f404756499dbb4d9e2d9b18490c71db7664363058a", size = 16205478, upload-time = "2024-11-25T23:09:03.757Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/3b/44b16c200637d4c55b5a26cb994af8b3af656d56dc277a761c1e1000ab97/tensorstore-0.1.69-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:172549769bf46ef454f9f9bca42eb1c9c75d8d822bbaf573b18397b53f7124b3", size = 17536793, upload-time = "2024-11-25T23:09:07.059Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/88/e828b97b9971cd1b71e99b2c3b154686b5ee893982f02e8886590b380ab5/tensorstore-0.1.69-cp311-cp311-win_amd64.whl", hash = "sha256:3d702d7a3fe4043a47d8cc5abc31a937a2073503dae97f92cf2bcbad6622b1ab", size = 11997580, upload-time = "2024-11-25T23:09:09.543Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/71/faf0e954dd6eb612f652c2f110cc292b1f3071b2a15179d98d9367221866/tensorstore-0.1.69-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:8e328b6c201d32339422838a97b67157470a34d2827b880c149d9102afb07e02", size = 14878020, upload-time = "2024-11-25T23:09:12.998Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/ea/67e55d3dce8ff473b42063a6fea639f21b8132f8e7072330f81df2af819c/tensorstore-0.1.69-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:f57f184e4fbf9175561914bd54045ca4fa75ea4be1206ebe4fe22b19b540ce5f", size = 12848724, upload-time = "2024-11-25T23:09:15.367Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/9e/48339e916035472a3307eb2746f731100834715aec1023f07d0d6a7e9fbc/tensorstore-0.1.69-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:368e84a9b67f153f18cefa8d6557098c52300b44ea633cbe00a1f23b6124373a", size = 16194513, upload-time = "2024-11-25T23:09:18.698Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/a9/f3dbd91a93a3f4d9c427676d5e11a9cf8b5485ff27eb13cbc57b0b937fa6/tensorstore-0.1.69-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3221a1954f1721c178b453a9bd9839c6d80a36a2f45afe43ec9d7f21de97ebd", size = 17529065, upload-time = "2024-11-25T23:09:21.2Z" },
+    { url = "https://files.pythonhosted.org/packages/00/ca/8080e4bd57028aface79b212de13d9fecdebe905e644823c5cc96e172484/tensorstore-0.1.69-cp312-cp312-win_amd64.whl", hash = "sha256:3d88bc1732cc2aa84f2ae98ed00586ce1ce9f149f885982f097248d77986d5ef", size = 11999274, upload-time = "2024-11-25T23:09:23.87Z" },
+    { url = "https://files.pythonhosted.org/packages/94/62/59cd878a73476a1e75ee96cc0b1f120bf23c590b1a4b6bdc5cb8c5418c52/tensorstore-0.1.69-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:9579ed3b3fcf13d64c5274e76b4cc987edda3134dca3532b8a0a36a9b66af486", size = 14837313, upload-time = "2024-11-25T23:09:26.378Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/58/859aaae8f7e94f94235c642face4f289162761b98b3770c7aece8950597d/tensorstore-0.1.69-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:25954820b75bed339bd3df87fd35596dc63773e192d7e63918092af9d3328146", size = 12820471, upload-time = "2024-11-25T23:09:30.36Z" },
+    { url = "https://files.pythonhosted.org/packages/64/65/98983608fff396ad46bfcce48e503a88aeb74cb47a9fe38ede9d2cbacd30/tensorstore-0.1.69-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:de18eefc23d75a0c438576985ca9a516ce5c1dc757a2052ce24e36c145239ff9", size = 16204805, upload-time = "2024-11-25T23:09:32.881Z" },
+    { url = "https://files.pythonhosted.org/packages/09/33/1b24d6eb58d9b50f1600eb500c020d4c87da43637d9ab1ee134e4d7d7dc0/tensorstore-0.1.69-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a645a770b6a5134e51869c215aa60bd29c565a07c0f12d1cd03352dcd9f1b6c0", size = 17537436, upload-time = "2024-11-25T23:09:36.542Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/25/36e7aad6c50572abc69668b8302252667e86f6d410e45cfa36ea1a35f79b/tensorstore-0.1.69-cp39-cp39-win_amd64.whl", hash = "sha256:041eaa3b875dd692882efe1ccf2e52eb67fcc212399f30d381f2e1c1a2932597", size = 11959410, upload-time = "2024-11-25T23:09:39.995Z" },
+]
+
+[[package]]
+name = "tensorstore"
+version = "0.1.76"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "ml-dtypes", marker = "python_full_version >= '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ff/ae/947a9f232de7319b664ed8d278e9e0363e9294da73fd422c687ac4eb070e/tensorstore-0.1.76.tar.gz", hash = "sha256:ed0d565e7a038a84b1b5b5d9f7397caec200b53941d8889f44b7f63dd6abffe7", size = 6869230, upload-time = "2025-07-02T21:34:03.773Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ca/66/6da79c349c3576836efec0195851bb71fed307dd9855de0692bf3e4df6cb/tensorstore-0.1.76-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:97e9eeab539369b99828c07e1d7da93062da91b65503dafe06d941208b0869d3", size = 15674263, upload-time = "2025-07-02T21:33:13.387Z" },
+    { url = "https://files.pythonhosted.org/packages/24/e7/014f188a5bd2087cdf0df154ee756ca95bae62ba76765f8f180e9dbc4e78/tensorstore-0.1.76-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a9c6f207a0b5f5138f7844307e5cb96587a73e9ba7b5edb6429524cca985e2c3", size = 13601077, upload-time = "2025-07-02T21:33:16.354Z" },
+    { url = "https://files.pythonhosted.org/packages/48/63/38410a20016c6c0b28d9970bcb63762f4df4b62a53eeddb27044deeb4792/tensorstore-0.1.76-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:05aae734cfaf5c1356b6f2c56226526607361c6e9aa6751df395dda53c3783b1", size = 17545411, upload-time = "2025-07-02T21:33:18.83Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/83/17fb39326adfee5d7048093ee248b3aa69c47031e62832825e29da81d0b7/tensorstore-0.1.76-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7574f837d8bfccf374795100be23f155eff3040f8b0f2010b1a0b2e1d4d6f791", size = 18923422, upload-time = "2025-07-02T21:33:21.38Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/a1/8932f2871e1f390e29f587c3295f3f00113d1a32a05b0723ef06f0134630/tensorstore-0.1.76-cp310-cp310-win_amd64.whl", hash = "sha256:9c964c7538805867894f6ba4a03dfa57e434b73d0c563d93fcbf6245eb67cc2c", size = 12609199, upload-time = "2025-07-02T21:33:24.337Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/9e/b3d691d14122064e16a3a47c14ce3b1178d749e59b3afec91a8656125c29/tensorstore-0.1.76-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:1c882dcf30049952cb6b183c70bd1922815cdbceca8d4115a7fbeb3b5513f9e4", size = 15675667, upload-time = "2025-07-02T21:33:27.409Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/bb/16d97d8b31912f27019115eb23b7feb0b83bf520858b97aec64064653329/tensorstore-0.1.76-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:262b856b21626688cefd11913737a94800b1ba3061d56d70babc439ef69587bc", size = 13602503, upload-time = "2025-07-02T21:33:29.872Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/5a/4b675941a73bc46959f24b3f5a68c246422278022a0e121f9c3f226a7a2b/tensorstore-0.1.76-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:daf1c7fb71cd07bfdf37ae4f297c0a6895b011562f4050f5c8f52f5753cc24cc", size = 17548043, upload-time = "2025-07-02T21:33:32.34Z" },
+    { url = "https://files.pythonhosted.org/packages/96/b9/8e306cbccb12ce4c241ac69fb98a4fc1bad5fc0311112f579bc24bee9c42/tensorstore-0.1.76-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b24f21bbd4830822422e984bc023a37ce7db08be02138c01c96870e62d041e7f", size = 18925296, upload-time = "2025-07-02T21:33:35.061Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/48/b542e9a4fa6f82b00e9a7c41c30003c195178fa78f835ea205b346a45baf/tensorstore-0.1.76-cp311-cp311-win_amd64.whl", hash = "sha256:36ba59d99d8279802793405fb8615803ea0e136ad439e6fe0ab3c3d7df22179d", size = 12609617, upload-time = "2025-07-02T21:33:37.153Z" },
+    { url = "https://files.pythonhosted.org/packages/09/37/f2254b4ae1dabd95e258fa3eb4783ac4db4261bb8c90ff9bfe15549d1238/tensorstore-0.1.76-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:b68450983ccad9e7774e81b2fa37daef1b72c774fd939d9eb4065d6aa70e666a", size = 15712650, upload-time = "2025-07-02T21:33:39.716Z" },
+    { url = "https://files.pythonhosted.org/packages/93/3c/1cae56cbbe9610ff48cb2d7c0921a4d4c333a0540918e3b2db08b521c5f6/tensorstore-0.1.76-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6b7a3856f884279e40f90bad87d0da70869879e124835e650c6b16c80f64fbc4", size = 13624138, upload-time = "2025-07-02T21:33:41.758Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/d2/b92d34a896f608a59dc76c290d4ec9f7d0264a02e4d74864987a6adbd3c9/tensorstore-0.1.76-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8709a98ae0b453eb23525c07372c2be1f6bbd978bba53319f26a1f2a83a77c2a", size = 17538270, upload-time = "2025-07-02T21:33:44.911Z" },
+    { url = "https://files.pythonhosted.org/packages/21/66/142b803541552b02a2fa033b1f48bcb50e1d2df6ac10131aab1857c5141d/tensorstore-0.1.76-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:267edea8f1596f2bd67017ff97b7b350bf3f95ff84947a8babadc5e17ca53663", size = 18910782, upload-time = "2025-07-02T21:33:47.401Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/3e/c264cf1435c04fb998a1f30dd1f066deb370b841412f89e1cb36d37ee4fc/tensorstore-0.1.76-cp312-cp312-win_amd64.whl", hash = "sha256:f66ac63d0c63c3336ac4dc61f1f97b6afe8b512e586ddfdbc91f19175787f321", size = 12611059, upload-time = "2025-07-02T21:33:49.596Z" },
+]
+
+[[package]]
+name = "termcolor"
+version = "3.1.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ca/6c/3d75c196ac07ac8749600b60b03f4f6094d54e132c4d94ebac6ee0e0add0/termcolor-3.1.0.tar.gz", hash = "sha256:6a6dd7fbee581909eeec6a756cff1d7f7c376063b14e4a298dc4980309e55970", size = 14324, upload-time = "2025-04-30T11:37:53.791Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/99/ff/c87e0622b1dadea79d2fb0b25ade9ed98954c9033722eb707053d310d4f3/sympy-1.13.3-py3-none-any.whl", hash = "sha256:54612cf55a62755ee71824ce692986f23c88ffa77207b30c1368eda4a7060f73", size = 6189483 },
+    { url = "https://files.pythonhosted.org/packages/4f/bd/de8d508070629b6d84a30d01d57e4a65c69aa7f5abe7560b8fad3b50ea59/termcolor-3.1.0-py3-none-any.whl", hash = "sha256:591dd26b5c2ce03b9e43f391264626557873ce1d379019786f99b0c2bee140aa", size = 7684, upload-time = "2025-04-30T11:37:52.382Z" },
 ]
 
 [[package]]
-name = "tbb"
-version = "2021.13.1"
+name = "tf-keras"
+version = "2.19.0"
 source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "tensorflow" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a9/71/a6e82b8e8ce04a8ab1ff0560791698b70d1b2dda1a0270f1d69a834e8880/tf_keras-2.19.0.tar.gz", hash = "sha256:b09a407d87a4571ce1e8ca985cfc68483e3d63b2518a5d79a97ad92cb64dbe9c", size = 1261172, upload-time = "2025-03-03T21:24:13.582Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/fb/8a/5062b00c378c051e26507e5eca8d3b5c91ed63f8a2139f6f0f422be84b02/tbb-2021.13.1-py3-none-win32.whl", hash = "sha256:00f5e5a70051650ddd0ab6247c0549521968339ec21002e475cd23b1cbf46d66", size = 248994 },
-    { url = "https://files.pythonhosted.org/packages/9b/24/84ce997e8ae6296168a74d0d9c4dde572d90fb23fd7c0b219c30ff71e00e/tbb-2021.13.1-py3-none-win_amd64.whl", hash = "sha256:cbf024b2463fdab3ebe3fa6ff453026358e6b903839c80d647e08ad6d0796ee9", size = 286908 },
+    { url = "https://files.pythonhosted.org/packages/45/6b/d245122d108a94df5969ee7408ad343af1627730e91478e01ef098976bfa/tf_keras-2.19.0-py3-none-any.whl", hash = "sha256:4f339e800987b39d1548a8c76a7b33b6801a97ec7fcd89c299ec29741f7890bd", size = 1726787, upload-time = "2025-03-03T21:24:11.826Z" },
 ]
 
 [[package]]
 name = "tiktoken"
-version = "0.8.0"
+version = "0.11.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "regex" },
     { name = "requests" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/37/02/576ff3a6639e755c4f70997b2d315f56d6d71e0d046f4fb64cb81a3fb099/tiktoken-0.8.0.tar.gz", hash = "sha256:9ccbb2740f24542534369c5635cfd9b2b3c2490754a78ac8831d99f89f94eeb2", size = 35107 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/c9/ba/a35fad753bbca8ba0cc1b0f3402a70256a110ced7ac332cf84ba89fc87ab/tiktoken-0.8.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b07e33283463089c81ef1467180e3e00ab00d46c2c4bbcef0acab5f771d6695e", size = 1039905 },
-    { url = "https://files.pythonhosted.org/packages/91/05/13dab8fd7460391c387b3e69e14bf1e51ff71fe0a202cd2933cc3ea93fb6/tiktoken-0.8.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9269348cb650726f44dd3bbb3f9110ac19a8dcc8f54949ad3ef652ca22a38e21", size = 982417 },
-    { url = "https://files.pythonhosted.org/packages/e9/98/18ec4a8351a6cf4537e40cd6e19a422c10cce1ef00a2fcb716e0a96af58b/tiktoken-0.8.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:25e13f37bc4ef2d012731e93e0fef21dc3b7aea5bb9009618de9a4026844e560", size = 1144915 },
-    { url = "https://files.pythonhosted.org/packages/2e/28/cf3633018cbcc6deb7805b700ccd6085c9a5a7f72b38974ee0bffd56d311/tiktoken-0.8.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f13d13c981511331eac0d01a59b5df7c0d4060a8be1e378672822213da51e0a2", size = 1177221 },
-    { url = "https://files.pythonhosted.org/packages/57/81/8a5be305cbd39d4e83a794f9e80c7f2c84b524587b7feb27c797b2046d51/tiktoken-0.8.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:6b2ddbc79a22621ce8b1166afa9f9a888a664a579350dc7c09346a3b5de837d9", size = 1237398 },
-    { url = "https://files.pythonhosted.org/packages/dc/da/8d1cc3089a83f5cf11c2e489332752981435280285231924557350523a59/tiktoken-0.8.0-cp310-cp310-win_amd64.whl", hash = "sha256:d8c2d0e5ba6453a290b86cd65fc51fedf247e1ba170191715b049dac1f628005", size = 884215 },
-    { url = "https://files.pythonhosted.org/packages/f6/1e/ca48e7bfeeccaf76f3a501bd84db1fa28b3c22c9d1a1f41af9fb7579c5f6/tiktoken-0.8.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:d622d8011e6d6f239297efa42a2657043aaed06c4f68833550cac9e9bc723ef1", size = 1039700 },
-    { url = "https://files.pythonhosted.org/packages/8c/f8/f0101d98d661b34534769c3818f5af631e59c36ac6d07268fbfc89e539ce/tiktoken-0.8.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2efaf6199717b4485031b4d6edb94075e4d79177a172f38dd934d911b588d54a", size = 982413 },
-    { url = "https://files.pythonhosted.org/packages/ac/3c/2b95391d9bd520a73830469f80a96e3790e6c0a5ac2444f80f20b4b31051/tiktoken-0.8.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5637e425ce1fc49cf716d88df3092048359a4b3bbb7da762840426e937ada06d", size = 1144242 },
-    { url = "https://files.pythonhosted.org/packages/01/c4/c4a4360de845217b6aa9709c15773484b50479f36bb50419c443204e5de9/tiktoken-0.8.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9fb0e352d1dbe15aba082883058b3cce9e48d33101bdaac1eccf66424feb5b47", size = 1176588 },
-    { url = "https://files.pythonhosted.org/packages/f8/a3/ef984e976822cd6c2227c854f74d2e60cf4cd6fbfca46251199914746f78/tiktoken-0.8.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:56edfefe896c8f10aba372ab5706b9e3558e78db39dd497c940b47bf228bc419", size = 1237261 },
-    { url = "https://files.pythonhosted.org/packages/1e/86/eea2309dc258fb86c7d9b10db536434fc16420feaa3b6113df18b23db7c2/tiktoken-0.8.0-cp311-cp311-win_amd64.whl", hash = "sha256:326624128590def898775b722ccc327e90b073714227175ea8febbc920ac0a99", size = 884537 },
-    { url = "https://files.pythonhosted.org/packages/c1/22/34b2e136a6f4af186b6640cbfd6f93400783c9ef6cd550d9eab80628d9de/tiktoken-0.8.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:881839cfeae051b3628d9823b2e56b5cc93a9e2efb435f4cf15f17dc45f21586", size = 1039357 },
-    { url = "https://files.pythonhosted.org/packages/04/d2/c793cf49c20f5855fd6ce05d080c0537d7418f22c58e71f392d5e8c8dbf7/tiktoken-0.8.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:fe9399bdc3f29d428f16a2f86c3c8ec20be3eac5f53693ce4980371c3245729b", size = 982616 },
-    { url = "https://files.pythonhosted.org/packages/b3/a1/79846e5ef911cd5d75c844de3fa496a10c91b4b5f550aad695c5df153d72/tiktoken-0.8.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9a58deb7075d5b69237a3ff4bb51a726670419db6ea62bdcd8bd80c78497d7ab", size = 1144011 },
-    { url = "https://files.pythonhosted.org/packages/26/32/e0e3a859136e95c85a572e4806dc58bf1ddf651108ae8b97d5f3ebe1a244/tiktoken-0.8.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d2908c0d043a7d03ebd80347266b0e58440bdef5564f84f4d29fb235b5df3b04", size = 1175432 },
-    { url = "https://files.pythonhosted.org/packages/c7/89/926b66e9025b97e9fbabeaa59048a736fe3c3e4530a204109571104f921c/tiktoken-0.8.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:294440d21a2a51e12d4238e68a5972095534fe9878be57d905c476017bff99fc", size = 1236576 },
-    { url = "https://files.pythonhosted.org/packages/45/e2/39d4aa02a52bba73b2cd21ba4533c84425ff8786cc63c511d68c8897376e/tiktoken-0.8.0-cp312-cp312-win_amd64.whl", hash = "sha256:d8f3192733ac4d77977432947d563d7e1b310b96497acd3c196c9bddb36ed9db", size = 883824 },
-    { url = "https://files.pythonhosted.org/packages/e3/38/802e79ba0ee5fcbf240cd624143f57744e5d411d2e9d9ad2db70d8395986/tiktoken-0.8.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:02be1666096aff7da6cbd7cdaa8e7917bfed3467cd64b38b1f112e96d3b06a24", size = 1039648 },
-    { url = "https://files.pythonhosted.org/packages/b1/da/24cdbfc302c98663fbea66f5866f7fa1048405c7564ab88483aea97c3b1a/tiktoken-0.8.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:c94ff53c5c74b535b2cbf431d907fc13c678bbd009ee633a2aca269a04389f9a", size = 982763 },
-    { url = "https://files.pythonhosted.org/packages/e4/f0/0ecf79a279dfa41fc97d00adccf976ecc2556d3c08ef3e25e45eb31f665b/tiktoken-0.8.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6b231f5e8982c245ee3065cd84a4712d64692348bc609d84467c57b4b72dcbc5", size = 1144417 },
-    { url = "https://files.pythonhosted.org/packages/ab/d3/155d2d4514f3471a25dc1d6d20549ef254e2aa9bb5b1060809b1d3b03d3a/tiktoken-0.8.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4177faa809bd55f699e88c96d9bb4635d22e3f59d635ba6fd9ffedf7150b9953", size = 1175108 },
-    { url = "https://files.pythonhosted.org/packages/19/eb/5989e16821ee8300ef8ee13c16effc20dfc26c777d05fbb6825e3c037b81/tiktoken-0.8.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5376b6f8dc4753cd81ead935c5f518fa0fbe7e133d9e25f648d8c4dabdd4bad7", size = 1236520 },
-    { url = "https://files.pythonhosted.org/packages/40/59/14b20465f1d1cb89cfbc96ec27e5617b2d41c79da12b5e04e96d689be2a7/tiktoken-0.8.0-cp313-cp313-win_amd64.whl", hash = "sha256:18228d624807d66c87acd8f25fc135665617cab220671eb65b50f5d70fa51f69", size = 883849 },
-    { url = "https://files.pythonhosted.org/packages/08/f3/8a8ba9329e6b426d822c974d58fc6477f3f7b3b8deef651813d275cbe75f/tiktoken-0.8.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7e17807445f0cf1f25771c9d86496bd8b5c376f7419912519699f3cc4dc5c12e", size = 1040915 },
-    { url = "https://files.pythonhosted.org/packages/42/7a/914bd98100449422778f9222d00b3a4ee654211c40784e57541fa46311ab/tiktoken-0.8.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:886f80bd339578bbdba6ed6d0567a0d5c6cfe198d9e587ba6c447654c65b8edc", size = 983753 },
-    { url = "https://files.pythonhosted.org/packages/f7/01/1483856d84827c5fe541cb160f07914c6b063b8d961146e9c3557c4730c0/tiktoken-0.8.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6adc8323016d7758d6de7313527f755b0fc6c72985b7d9291be5d96d73ecd1e1", size = 1145913 },
-    { url = "https://files.pythonhosted.org/packages/c2/e1/6c7a772e0200131e960e3381f1d7b26406bc5612c70677989c1498af2a60/tiktoken-0.8.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b591fb2b30d6a72121a80be24ec7a0e9eb51c5500ddc7e4c2496516dd5e3816b", size = 1178505 },
-    { url = "https://files.pythonhosted.org/packages/3e/6b/3ae00f0bff5d0b6925bf6370cf0ff606f56daed76210c2b0a156017b78dc/tiktoken-0.8.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:845287b9798e476b4d762c3ebda5102be87ca26e5d2c9854002825d60cdb815d", size = 1239111 },
-    { url = "https://files.pythonhosted.org/packages/d5/3b/7c8812952ca55e1bab08afc1dda3c5991804c71b550b9402e82a082ab795/tiktoken-0.8.0-cp39-cp39-win_amd64.whl", hash = "sha256:1473cfe584252dc3fa62adceb5b1c763c1874e04511b197da4e6de51d6ce5a02", size = 884803 },
+sdist = { url = "https://files.pythonhosted.org/packages/a7/86/ad0155a37c4f310935d5ac0b1ccf9bdb635dcb906e0a9a26b616dd55825a/tiktoken-0.11.0.tar.gz", hash = "sha256:3c518641aee1c52247c2b97e74d8d07d780092af79d5911a6ab5e79359d9b06a", size = 37648, upload-time = "2025-08-08T23:58:08.495Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/dd/f4/ec8d43338d28d53513004ebf4cd83732a135d11011433c58bf045890cc10/tiktoken-0.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:10331d08b5ecf7a780b4fe4d0281328b23ab22cdb4ff65e68d56caeda9940ecc", size = 1130076, upload-time = "2025-08-08T23:57:31.706Z" },
+    { url = "https://files.pythonhosted.org/packages/94/80/fb0ada0a882cb453caf519a4bf0d117c2a3ee2e852c88775abff5413c176/tiktoken-0.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b062c82300341dc87e0258c69f79bed725f87e753c21887aea90d272816be882", size = 1183942, upload-time = "2025-08-08T23:57:33.142Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/e9/6c104355b463601719582823f3ea658bc3aa7c73d1b3b7553ebdc48468ce/tiktoken-0.11.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:195d84bec46169af3b1349a1495c151d37a0ff4cba73fd08282736be7f92cc6c", size = 1244705, upload-time = "2025-08-08T23:57:34.594Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/3e/a05d1547cf7db9dc75d1461cfa7b556a3b48e0516ec29dfc81d984a145f6/tiktoken-0.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2130127471e293d385179c1f3f9cd445070c0772be73cdafb7cec9a3684c0458", size = 1129402, upload-time = "2025-08-08T23:57:39.627Z" },
+    { url = "https://files.pythonhosted.org/packages/34/9a/db7a86b829e05a01fd4daa492086f708e0a8b53952e1dbc9d380d2b03677/tiktoken-0.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:21e43022bf2c33f733ea9b54f6a3f6b4354b909f5a73388fb1b9347ca54a069c", size = 1184046, upload-time = "2025-08-08T23:57:40.689Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/bb/52edc8e078cf062ed749248f1454e9e5cfd09979baadb830b3940e522015/tiktoken-0.11.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:adb4e308eb64380dc70fa30493e21c93475eaa11669dea313b6bbf8210bfd013", size = 1244691, upload-time = "2025-08-08T23:57:42.251Z" },
+    { url = "https://files.pythonhosted.org/packages/65/8e/c769b45ef379bc360c9978c4f6914c79fd432400a6733a8afc7ed7b0726a/tiktoken-0.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6ef72aab3ea240646e642413cb363b73869fed4e604dcfd69eec63dc54d603e8", size = 1128867, upload-time = "2025-08-08T23:57:47.438Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/2d/4d77f6feb9292bfdd23d5813e442b3bba883f42d0ac78ef5fdc56873f756/tiktoken-0.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7f929255c705efec7a28bf515e29dc74220b2f07544a8c81b8d69e8efc4578bd", size = 1183308, upload-time = "2025-08-08T23:57:48.566Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/65/7ff0a65d3bb0fc5a1fb6cc71b03e0f6e71a68c5eea230d1ff1ba3fd6df49/tiktoken-0.11.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:61f1d15822e4404953d499fd1dcc62817a12ae9fb1e4898033ec8fe3915fdf8e", size = 1244301, upload-time = "2025-08-08T23:57:49.642Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/7f/5f950528b54cb3025af4bc3522c23dbfb691afe8ffb292aa1e8dc2e6bddf/tiktoken-0.11.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2302772f035dceb2bcf8e55a735e4604a0b51a6dd50f38218ff664d46ec43807", size = 1130777, upload-time = "2025-08-08T23:58:03.256Z" },
+    { url = "https://files.pythonhosted.org/packages/27/a4/e82ddf0773835ba24536ac8c0dce561e697698ec020a93212a1e041d39b4/tiktoken-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:20b977989afe44c94bcc50db1f76971bb26dca44218bd203ba95925ef56f8e7a", size = 1185692, upload-time = "2025-08-08T23:58:04.476Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/c2/06361e41d176e62797ae65fa678111cdd30553321cf4d83e7b84107ea95f/tiktoken-0.11.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:669a1aa1ad6ebf1b3c26b45deb346f345da7680f845b5ea700bba45c20dea24c", size = 1246518, upload-time = "2025-08-08T23:58:06.126Z" },
 ]
 
 [[package]]
 name = "tokenizers"
-version = "0.21.0"
+version = "0.21.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "huggingface-hub" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/20/41/c2be10975ca37f6ec40d7abd7e98a5213bb04f284b869c1a24e6504fd94d/tokenizers-0.21.0.tar.gz", hash = "sha256:ee0894bf311b75b0c03079f33859ae4b2334d675d4e93f5a4132e1eae2834fe4", size = 343021 }
+sdist = { url = "https://files.pythonhosted.org/packages/c2/2f/402986d0823f8d7ca139d969af2917fefaa9b947d1fb32f6168c509f2492/tokenizers-0.21.4.tar.gz", hash = "sha256:fa23f85fbc9a02ec5c6978da172cdcbac23498c3ca9f3645c5c68740ac007880", size = 351253, upload-time = "2025-07-28T15:48:54.325Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b0/5c/8b09607b37e996dc47e70d6a7b6f4bdd4e4d5ab22fe49d7374565c7fefaf/tokenizers-0.21.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:3c4c93eae637e7d2aaae3d376f06085164e1660f89304c0ab2b1d08a406636b2", size = 2647461 },
-    { url = "https://files.pythonhosted.org/packages/22/7a/88e58bb297c22633ed1c9d16029316e5b5ac5ee44012164c2edede599a5e/tokenizers-0.21.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:f53ea537c925422a2e0e92a24cce96f6bc5046bbef24a1652a5edc8ba975f62e", size = 2563639 },
-    { url = "https://files.pythonhosted.org/packages/f7/14/83429177c19364df27d22bc096d4c2e431e0ba43e56c525434f1f9b0fd00/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6b177fb54c4702ef611de0c069d9169f0004233890e0c4c5bd5508ae05abf193", size = 2903304 },
-    { url = "https://files.pythonhosted.org/packages/7e/db/3433eab42347e0dc5452d8fcc8da03f638c9accffefe5a7c78146666964a/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:6b43779a269f4629bebb114e19c3fca0223296ae9fea8bb9a7a6c6fb0657ff8e", size = 2804378 },
-    { url = "https://files.pythonhosted.org/packages/57/8b/7da5e6f89736c2ade02816b4733983fca1c226b0c42980b1ae9dc8fcf5cc/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9aeb255802be90acfd363626753fda0064a8df06031012fe7d52fd9a905eb00e", size = 3095488 },
-    { url = "https://files.pythonhosted.org/packages/4d/f6/5ed6711093dc2c04a4e03f6461798b12669bc5a17c8be7cce1240e0b5ce8/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d8b09dbeb7a8d73ee204a70f94fc06ea0f17dcf0844f16102b9f414f0b7463ba", size = 3121410 },
-    { url = "https://files.pythonhosted.org/packages/81/42/07600892d48950c5e80505b81411044a2d969368cdc0d929b1c847bf6697/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:400832c0904f77ce87c40f1a8a27493071282f785724ae62144324f171377273", size = 3388821 },
-    { url = "https://files.pythonhosted.org/packages/22/06/69d7ce374747edaf1695a4f61b83570d91cc8bbfc51ccfecf76f56ab4aac/tokenizers-0.21.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e84ca973b3a96894d1707e189c14a774b701596d579ffc7e69debfc036a61a04", size = 3008868 },
-    { url = "https://files.pythonhosted.org/packages/c8/69/54a0aee4d576045b49a0eb8bffdc495634309c823bf886042e6f46b80058/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:eb7202d231b273c34ec67767378cd04c767e967fda12d4a9e36208a34e2f137e", size = 8975831 },
-    { url = "https://files.pythonhosted.org/packages/f7/f3/b776061e4f3ebf2905ba1a25d90380aafd10c02d406437a8ba22d1724d76/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:089d56db6782a73a27fd8abf3ba21779f5b85d4a9f35e3b493c7bbcbbf0d539b", size = 8920746 },
-    { url = "https://files.pythonhosted.org/packages/d8/ee/ce83d5ec8b6844ad4c3ecfe3333d58ecc1adc61f0878b323a15355bcab24/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:c87ca3dc48b9b1222d984b6b7490355a6fdb411a2d810f6f05977258400ddb74", size = 9161814 },
-    { url = "https://files.pythonhosted.org/packages/18/07/3e88e65c0ed28fa93aa0c4d264988428eef3df2764c3126dc83e243cb36f/tokenizers-0.21.0-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:4145505a973116f91bc3ac45988a92e618a6f83eb458f49ea0790df94ee243ff", size = 9357138 },
-    { url = "https://files.pythonhosted.org/packages/15/b0/dc4572ca61555fc482ebc933f26cb407c6aceb3dc19c301c68184f8cad03/tokenizers-0.21.0-cp39-abi3-win32.whl", hash = "sha256:eb1702c2f27d25d9dd5b389cc1f2f51813e99f8ca30d9e25348db6585a97e24a", size = 2202266 },
-    { url = "https://files.pythonhosted.org/packages/44/69/d21eb253fa91622da25585d362a874fa4710be600f0ea9446d8d0217cec1/tokenizers-0.21.0-cp39-abi3-win_amd64.whl", hash = "sha256:87841da5a25a3a5f70c102de371db120f41873b854ba65e52bccd57df5a3780c", size = 2389192 },
+    { url = "https://files.pythonhosted.org/packages/98/c6/fdb6f72bf6454f52eb4a2510be7fb0f614e541a2554d6210e370d85efff4/tokenizers-0.21.4-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:2ccc10a7c3bcefe0f242867dc914fc1226ee44321eb618cfe3019b5df3400133", size = 2863987, upload-time = "2025-07-28T15:48:44.877Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/a6/28975479e35ddc751dc1ddc97b9b69bf7fcf074db31548aab37f8116674c/tokenizers-0.21.4-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:5e2f601a8e0cd5be5cc7506b20a79112370b9b3e9cb5f13f68ab11acd6ca7d60", size = 2732457, upload-time = "2025-07-28T15:48:43.265Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/8f/24f39d7b5c726b7b0be95dca04f344df278a3fe3a4deb15a975d194cbb32/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:39b376f5a1aee67b4d29032ee85511bbd1b99007ec735f7f35c8a2eb104eade5", size = 3012624, upload-time = "2025-07-28T13:22:43.895Z" },
+    { url = "https://files.pythonhosted.org/packages/58/47/26358925717687a58cb74d7a508de96649544fad5778f0cd9827398dc499/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:2107ad649e2cda4488d41dfd031469e9da3fcbfd6183e74e4958fa729ffbf9c6", size = 2939681, upload-time = "2025-07-28T13:22:47.499Z" },
+    { url = "https://files.pythonhosted.org/packages/99/6f/cc300fea5db2ab5ddc2c8aea5757a27b89c84469899710c3aeddc1d39801/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3c73012da95afafdf235ba80047699df4384fdc481527448a078ffd00e45a7d9", size = 3247445, upload-time = "2025-07-28T15:48:39.711Z" },
+    { url = "https://files.pythonhosted.org/packages/be/bf/98cb4b9c3c4afd8be89cfa6423704337dc20b73eb4180397a6e0d456c334/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f23186c40395fc390d27f519679a58023f368a0aad234af145e0f39ad1212732", size = 3428014, upload-time = "2025-07-28T13:22:49.569Z" },
+    { url = "https://files.pythonhosted.org/packages/75/c7/96c1cc780e6ca7f01a57c13235dd05b7bc1c0f3588512ebe9d1331b5f5ae/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cc88bb34e23a54cc42713d6d98af5f1bf79c07653d24fe984d2d695ba2c922a2", size = 3193197, upload-time = "2025-07-28T13:22:51.471Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/90/273b6c7ec78af547694eddeea9e05de771278bd20476525ab930cecaf7d8/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:51b7eabb104f46c1c50b486520555715457ae833d5aee9ff6ae853d1130506ff", size = 3115426, upload-time = "2025-07-28T15:48:41.439Z" },
+    { url = "https://files.pythonhosted.org/packages/91/43/c640d5a07e95f1cf9d2c92501f20a25f179ac53a4f71e1489a3dcfcc67ee/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:714b05b2e1af1288bd1bc56ce496c4cebb64a20d158ee802887757791191e6e2", size = 9089127, upload-time = "2025-07-28T15:48:46.472Z" },
+    { url = "https://files.pythonhosted.org/packages/44/a1/dd23edd6271d4dca788e5200a807b49ec3e6987815cd9d0a07ad9c96c7c2/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:1340ff877ceedfa937544b7d79f5b7becf33a4cfb58f89b3b49927004ef66f78", size = 9055243, upload-time = "2025-07-28T15:48:48.539Z" },
+    { url = "https://files.pythonhosted.org/packages/21/2b/b410d6e9021c4b7ddb57248304dc817c4d4970b73b6ee343674914701197/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:3c1f4317576e465ac9ef0d165b247825a2a4078bcd01cba6b54b867bdf9fdd8b", size = 9298237, upload-time = "2025-07-28T15:48:50.443Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/0a/42348c995c67e2e6e5c89ffb9cfd68507cbaeb84ff39c49ee6e0a6dd0fd2/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:c212aa4e45ec0bb5274b16b6f31dd3f1c41944025c2358faaa5782c754e84c24", size = 9461980, upload-time = "2025-07-28T15:48:52.325Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/d3/dacccd834404cd71b5c334882f3ba40331ad2120e69ded32cf5fda9a7436/tokenizers-0.21.4-cp39-abi3-win32.whl", hash = "sha256:6c42a930bc5f4c47f4ea775c91de47d27910881902b0f20e4990ebe045a415d0", size = 2329871, upload-time = "2025-07-28T15:48:56.841Z" },
+    { url = "https://files.pythonhosted.org/packages/41/f2/fd673d979185f5dcbac4be7d09461cbb99751554ffb6718d0013af8604cb/tokenizers-0.21.4-cp39-abi3-win_amd64.whl", hash = "sha256:475d807a5c3eb72c59ad9b5fcdb254f6e17f53dfcbb9903233b0dfa9c943b597", size = 2507568, upload-time = "2025-07-28T15:48:55.456Z" },
 ]
 
 [[package]]
 name = "tomli"
 version = "2.2.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/18/87/302344fed471e44a87289cf4967697d07e532f2421fdaf868a303cbae4ff/tomli-2.2.1.tar.gz", hash = "sha256:cd45e1dc79c835ce60f7404ec8119f2eb06d38b1deba146f07ced3bbc44505ff", size = 17175 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/43/ca/75707e6efa2b37c77dadb324ae7d9571cb424e61ea73fad7c56c2d14527f/tomli-2.2.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:678e4fa69e4575eb77d103de3df8a895e1591b48e740211bd1067378c69e8249", size = 131077 },
-    { url = "https://files.pythonhosted.org/packages/c7/16/51ae563a8615d472fdbffc43a3f3d46588c264ac4f024f63f01283becfbb/tomli-2.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:023aa114dd824ade0100497eb2318602af309e5a55595f76b626d6d9f3b7b0a6", size = 123429 },
-    { url = "https://files.pythonhosted.org/packages/f1/dd/4f6cd1e7b160041db83c694abc78e100473c15d54620083dbd5aae7b990e/tomli-2.2.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ece47d672db52ac607a3d9599a9d48dcb2f2f735c6c2d1f34130085bb12b112a", size = 226067 },
-    { url = "https://files.pythonhosted.org/packages/a9/6b/c54ede5dc70d648cc6361eaf429304b02f2871a345bbdd51e993d6cdf550/tomli-2.2.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6972ca9c9cc9f0acaa56a8ca1ff51e7af152a9f87fb64623e31d5c83700080ee", size = 236030 },
-    { url = "https://files.pythonhosted.org/packages/1f/47/999514fa49cfaf7a92c805a86c3c43f4215621855d151b61c602abb38091/tomli-2.2.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c954d2250168d28797dd4e3ac5cf812a406cd5a92674ee4c8f123c889786aa8e", size = 240898 },
-    { url = "https://files.pythonhosted.org/packages/73/41/0a01279a7ae09ee1573b423318e7934674ce06eb33f50936655071d81a24/tomli-2.2.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8dd28b3e155b80f4d54beb40a441d366adcfe740969820caf156c019fb5c7ec4", size = 229894 },
-    { url = "https://files.pythonhosted.org/packages/55/18/5d8bc5b0a0362311ce4d18830a5d28943667599a60d20118074ea1b01bb7/tomli-2.2.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:e59e304978767a54663af13c07b3d1af22ddee3bb2fb0618ca1593e4f593a106", size = 245319 },
-    { url = "https://files.pythonhosted.org/packages/92/a3/7ade0576d17f3cdf5ff44d61390d4b3febb8a9fc2b480c75c47ea048c646/tomli-2.2.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:33580bccab0338d00994d7f16f4c4ec25b776af3ffaac1ed74e0b3fc95e885a8", size = 238273 },
-    { url = "https://files.pythonhosted.org/packages/72/6f/fa64ef058ac1446a1e51110c375339b3ec6be245af9d14c87c4a6412dd32/tomli-2.2.1-cp311-cp311-win32.whl", hash = "sha256:465af0e0875402f1d226519c9904f37254b3045fc5084697cefb9bdde1ff99ff", size = 98310 },
-    { url = "https://files.pythonhosted.org/packages/6a/1c/4a2dcde4a51b81be3530565e92eda625d94dafb46dbeb15069df4caffc34/tomli-2.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:2d0f2fdd22b02c6d81637a3c95f8cd77f995846af7414c5c4b8d0545afa1bc4b", size = 108309 },
-    { url = "https://files.pythonhosted.org/packages/52/e1/f8af4c2fcde17500422858155aeb0d7e93477a0d59a98e56cbfe75070fd0/tomli-2.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:4a8f6e44de52d5e6c657c9fe83b562f5f4256d8ebbfe4ff922c495620a7f6cea", size = 132762 },
-    { url = "https://files.pythonhosted.org/packages/03/b8/152c68bb84fc00396b83e7bbddd5ec0bd3dd409db4195e2a9b3e398ad2e3/tomli-2.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8d57ca8095a641b8237d5b079147646153d22552f1c637fd3ba7f4b0b29167a8", size = 123453 },
-    { url = "https://files.pythonhosted.org/packages/c8/d6/fc9267af9166f79ac528ff7e8c55c8181ded34eb4b0e93daa767b8841573/tomli-2.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e340144ad7ae1533cb897d406382b4b6fede8890a03738ff1683af800d54192", size = 233486 },
-    { url = "https://files.pythonhosted.org/packages/5c/51/51c3f2884d7bab89af25f678447ea7d297b53b5a3b5730a7cb2ef6069f07/tomli-2.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:db2b95f9de79181805df90bedc5a5ab4c165e6ec3fe99f970d0e302f384ad222", size = 242349 },
-    { url = "https://files.pythonhosted.org/packages/ab/df/bfa89627d13a5cc22402e441e8a931ef2108403db390ff3345c05253935e/tomli-2.2.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:40741994320b232529c802f8bc86da4e1aa9f413db394617b9a256ae0f9a7f77", size = 252159 },
-    { url = "https://files.pythonhosted.org/packages/9e/6e/fa2b916dced65763a5168c6ccb91066f7639bdc88b48adda990db10c8c0b/tomli-2.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:400e720fe168c0f8521520190686ef8ef033fb19fc493da09779e592861b78c6", size = 237243 },
-    { url = "https://files.pythonhosted.org/packages/b4/04/885d3b1f650e1153cbb93a6a9782c58a972b94ea4483ae4ac5cedd5e4a09/tomli-2.2.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:02abe224de6ae62c19f090f68da4e27b10af2b93213d36cf44e6e1c5abd19fdd", size = 259645 },
-    { url = "https://files.pythonhosted.org/packages/9c/de/6b432d66e986e501586da298e28ebeefd3edc2c780f3ad73d22566034239/tomli-2.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b82ebccc8c8a36f2094e969560a1b836758481f3dc360ce9a3277c65f374285e", size = 244584 },
-    { url = "https://files.pythonhosted.org/packages/1c/9a/47c0449b98e6e7d1be6cbac02f93dd79003234ddc4aaab6ba07a9a7482e2/tomli-2.2.1-cp312-cp312-win32.whl", hash = "sha256:889f80ef92701b9dbb224e49ec87c645ce5df3fa2cc548664eb8a25e03127a98", size = 98875 },
-    { url = "https://files.pythonhosted.org/packages/ef/60/9b9638f081c6f1261e2688bd487625cd1e660d0a85bd469e91d8db969734/tomli-2.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:7fc04e92e1d624a4a63c76474610238576942d6b8950a2d7f908a340494e67e4", size = 109418 },
-    { url = "https://files.pythonhosted.org/packages/04/90/2ee5f2e0362cb8a0b6499dc44f4d7d48f8fff06d28ba46e6f1eaa61a1388/tomli-2.2.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:f4039b9cbc3048b2416cc57ab3bda989a6fcf9b36cf8937f01a6e731b64f80d7", size = 132708 },
-    { url = "https://files.pythonhosted.org/packages/c0/ec/46b4108816de6b385141f082ba99e315501ccd0a2ea23db4a100dd3990ea/tomli-2.2.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:286f0ca2ffeeb5b9bd4fcc8d6c330534323ec51b2f52da063b11c502da16f30c", size = 123582 },
-    { url = "https://files.pythonhosted.org/packages/a0/bd/b470466d0137b37b68d24556c38a0cc819e8febe392d5b199dcd7f578365/tomli-2.2.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a92ef1a44547e894e2a17d24e7557a5e85a9e1d0048b0b5e7541f76c5032cb13", size = 232543 },
-    { url = "https://files.pythonhosted.org/packages/d9/e5/82e80ff3b751373f7cead2815bcbe2d51c895b3c990686741a8e56ec42ab/tomli-2.2.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9316dc65bed1684c9a98ee68759ceaed29d229e985297003e494aa825ebb0281", size = 241691 },
-    { url = "https://files.pythonhosted.org/packages/05/7e/2a110bc2713557d6a1bfb06af23dd01e7dde52b6ee7dadc589868f9abfac/tomli-2.2.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e85e99945e688e32d5a35c1ff38ed0b3f41f43fad8df0bdf79f72b2ba7bc5272", size = 251170 },
-    { url = "https://files.pythonhosted.org/packages/64/7b/22d713946efe00e0adbcdfd6d1aa119ae03fd0b60ebed51ebb3fa9f5a2e5/tomli-2.2.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ac065718db92ca818f8d6141b5f66369833d4a80a9d74435a268c52bdfa73140", size = 236530 },
-    { url = "https://files.pythonhosted.org/packages/38/31/3a76f67da4b0cf37b742ca76beaf819dca0ebef26d78fc794a576e08accf/tomli-2.2.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:d920f33822747519673ee656a4b6ac33e382eca9d331c87770faa3eef562aeb2", size = 258666 },
-    { url = "https://files.pythonhosted.org/packages/07/10/5af1293da642aded87e8a988753945d0cf7e00a9452d3911dd3bb354c9e2/tomli-2.2.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a198f10c4d1b1375d7687bc25294306e551bf1abfa4eace6650070a5c1ae2744", size = 243954 },
-    { url = "https://files.pythonhosted.org/packages/5b/b9/1ed31d167be802da0fc95020d04cd27b7d7065cc6fbefdd2f9186f60d7bd/tomli-2.2.1-cp313-cp313-win32.whl", hash = "sha256:d3f5614314d758649ab2ab3a62d4f2004c825922f9e370b29416484086b264ec", size = 98724 },
-    { url = "https://files.pythonhosted.org/packages/c7/32/b0963458706accd9afcfeb867c0f9175a741bf7b19cd424230714d722198/tomli-2.2.1-cp313-cp313-win_amd64.whl", hash = "sha256:a38aa0308e754b0e3c67e344754dff64999ff9b513e691d0e786265c93583c69", size = 109383 },
-    { url = "https://files.pythonhosted.org/packages/6e/c2/61d3e0f47e2b74ef40a68b9e6ad5984f6241a942f7cd3bbfbdbd03861ea9/tomli-2.2.1-py3-none-any.whl", hash = "sha256:cb55c73c5f4408779d0cf3eef9f762b9c9f147a77de7b258bef0a5628adc85cc", size = 14257 },
+sdist = { url = "https://files.pythonhosted.org/packages/18/87/302344fed471e44a87289cf4967697d07e532f2421fdaf868a303cbae4ff/tomli-2.2.1.tar.gz", hash = "sha256:cd45e1dc79c835ce60f7404ec8119f2eb06d38b1deba146f07ced3bbc44505ff", size = 17175, upload-time = "2024-11-27T22:38:36.873Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/ca/75707e6efa2b37c77dadb324ae7d9571cb424e61ea73fad7c56c2d14527f/tomli-2.2.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:678e4fa69e4575eb77d103de3df8a895e1591b48e740211bd1067378c69e8249", size = 131077, upload-time = "2024-11-27T22:37:54.956Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/16/51ae563a8615d472fdbffc43a3f3d46588c264ac4f024f63f01283becfbb/tomli-2.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:023aa114dd824ade0100497eb2318602af309e5a55595f76b626d6d9f3b7b0a6", size = 123429, upload-time = "2024-11-27T22:37:56.698Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/dd/4f6cd1e7b160041db83c694abc78e100473c15d54620083dbd5aae7b990e/tomli-2.2.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ece47d672db52ac607a3d9599a9d48dcb2f2f735c6c2d1f34130085bb12b112a", size = 226067, upload-time = "2024-11-27T22:37:57.63Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/6b/c54ede5dc70d648cc6361eaf429304b02f2871a345bbdd51e993d6cdf550/tomli-2.2.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6972ca9c9cc9f0acaa56a8ca1ff51e7af152a9f87fb64623e31d5c83700080ee", size = 236030, upload-time = "2024-11-27T22:37:59.344Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/47/999514fa49cfaf7a92c805a86c3c43f4215621855d151b61c602abb38091/tomli-2.2.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c954d2250168d28797dd4e3ac5cf812a406cd5a92674ee4c8f123c889786aa8e", size = 240898, upload-time = "2024-11-27T22:38:00.429Z" },
+    { url = "https://files.pythonhosted.org/packages/73/41/0a01279a7ae09ee1573b423318e7934674ce06eb33f50936655071d81a24/tomli-2.2.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8dd28b3e155b80f4d54beb40a441d366adcfe740969820caf156c019fb5c7ec4", size = 229894, upload-time = "2024-11-27T22:38:02.094Z" },
+    { url = "https://files.pythonhosted.org/packages/55/18/5d8bc5b0a0362311ce4d18830a5d28943667599a60d20118074ea1b01bb7/tomli-2.2.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:e59e304978767a54663af13c07b3d1af22ddee3bb2fb0618ca1593e4f593a106", size = 245319, upload-time = "2024-11-27T22:38:03.206Z" },
+    { url = "https://files.pythonhosted.org/packages/92/a3/7ade0576d17f3cdf5ff44d61390d4b3febb8a9fc2b480c75c47ea048c646/tomli-2.2.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:33580bccab0338d00994d7f16f4c4ec25b776af3ffaac1ed74e0b3fc95e885a8", size = 238273, upload-time = "2024-11-27T22:38:04.217Z" },
+    { url = "https://files.pythonhosted.org/packages/72/6f/fa64ef058ac1446a1e51110c375339b3ec6be245af9d14c87c4a6412dd32/tomli-2.2.1-cp311-cp311-win32.whl", hash = "sha256:465af0e0875402f1d226519c9904f37254b3045fc5084697cefb9bdde1ff99ff", size = 98310, upload-time = "2024-11-27T22:38:05.908Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/1c/4a2dcde4a51b81be3530565e92eda625d94dafb46dbeb15069df4caffc34/tomli-2.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:2d0f2fdd22b02c6d81637a3c95f8cd77f995846af7414c5c4b8d0545afa1bc4b", size = 108309, upload-time = "2024-11-27T22:38:06.812Z" },
+    { url = "https://files.pythonhosted.org/packages/52/e1/f8af4c2fcde17500422858155aeb0d7e93477a0d59a98e56cbfe75070fd0/tomli-2.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:4a8f6e44de52d5e6c657c9fe83b562f5f4256d8ebbfe4ff922c495620a7f6cea", size = 132762, upload-time = "2024-11-27T22:38:07.731Z" },
+    { url = "https://files.pythonhosted.org/packages/03/b8/152c68bb84fc00396b83e7bbddd5ec0bd3dd409db4195e2a9b3e398ad2e3/tomli-2.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8d57ca8095a641b8237d5b079147646153d22552f1c637fd3ba7f4b0b29167a8", size = 123453, upload-time = "2024-11-27T22:38:09.384Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/d6/fc9267af9166f79ac528ff7e8c55c8181ded34eb4b0e93daa767b8841573/tomli-2.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e340144ad7ae1533cb897d406382b4b6fede8890a03738ff1683af800d54192", size = 233486, upload-time = "2024-11-27T22:38:10.329Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/51/51c3f2884d7bab89af25f678447ea7d297b53b5a3b5730a7cb2ef6069f07/tomli-2.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:db2b95f9de79181805df90bedc5a5ab4c165e6ec3fe99f970d0e302f384ad222", size = 242349, upload-time = "2024-11-27T22:38:11.443Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/df/bfa89627d13a5cc22402e441e8a931ef2108403db390ff3345c05253935e/tomli-2.2.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:40741994320b232529c802f8bc86da4e1aa9f413db394617b9a256ae0f9a7f77", size = 252159, upload-time = "2024-11-27T22:38:13.099Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/6e/fa2b916dced65763a5168c6ccb91066f7639bdc88b48adda990db10c8c0b/tomli-2.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:400e720fe168c0f8521520190686ef8ef033fb19fc493da09779e592861b78c6", size = 237243, upload-time = "2024-11-27T22:38:14.766Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/04/885d3b1f650e1153cbb93a6a9782c58a972b94ea4483ae4ac5cedd5e4a09/tomli-2.2.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:02abe224de6ae62c19f090f68da4e27b10af2b93213d36cf44e6e1c5abd19fdd", size = 259645, upload-time = "2024-11-27T22:38:15.843Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/de/6b432d66e986e501586da298e28ebeefd3edc2c780f3ad73d22566034239/tomli-2.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b82ebccc8c8a36f2094e969560a1b836758481f3dc360ce9a3277c65f374285e", size = 244584, upload-time = "2024-11-27T22:38:17.645Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/9a/47c0449b98e6e7d1be6cbac02f93dd79003234ddc4aaab6ba07a9a7482e2/tomli-2.2.1-cp312-cp312-win32.whl", hash = "sha256:889f80ef92701b9dbb224e49ec87c645ce5df3fa2cc548664eb8a25e03127a98", size = 98875, upload-time = "2024-11-27T22:38:19.159Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/60/9b9638f081c6f1261e2688bd487625cd1e660d0a85bd469e91d8db969734/tomli-2.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:7fc04e92e1d624a4a63c76474610238576942d6b8950a2d7f908a340494e67e4", size = 109418, upload-time = "2024-11-27T22:38:20.064Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/c2/61d3e0f47e2b74ef40a68b9e6ad5984f6241a942f7cd3bbfbdbd03861ea9/tomli-2.2.1-py3-none-any.whl", hash = "sha256:cb55c73c5f4408779d0cf3eef9f762b9c9f147a77de7b258bef0a5628adc85cc", size = 14257, upload-time = "2024-11-27T22:38:35.385Z" },
+]
+
+[[package]]
+name = "toolz"
+version = "1.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8a/0b/d80dfa675bf592f636d1ea0b835eab4ec8df6e9415d8cfd766df54456123/toolz-1.0.0.tar.gz", hash = "sha256:2c86e3d9a04798ac556793bced838816296a2f085017664e4995cb40a1047a02", size = 66790, upload-time = "2024-10-04T16:17:04.001Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/03/98/eb27cc78ad3af8e302c9d8ff4977f5026676e130d28dd7578132a457170c/toolz-1.0.0-py3-none-any.whl", hash = "sha256:292c8f1c4e7516bf9086f8850935c799a874039c8bcf959d47b600e4c44a6236", size = 56383, upload-time = "2024-10-04T16:17:01.533Z" },
 ]
 
 [[package]]
 name = "torch"
-version = "2.3.0"
+version = "2.8.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "filelock" },
     { name = "fsspec" },
     { name = "jinja2" },
-    { name = "mkl", marker = "sys_platform == 'win32'" },
     { name = "networkx", version = "3.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "networkx", version = "3.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "networkx", version = "3.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "networkx", version = "3.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "nvidia-cublas-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-cuda-cupti-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-cuda-nvrtc-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-cuda-runtime-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-cudnn-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-cufft-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufile-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-curand-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-cusolver-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-cusparse-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparselt-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-nccl-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvjitlink-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "nvidia-nvtx-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "setuptools", marker = "python_full_version >= '3.12'" },
     { name = "sympy" },
-    { name = "triton", marker = "python_full_version < '3.12' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "triton", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "typing-extensions" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/43/e5/2ddae60ae999b224aceb74490abeb885ee118227f866cb12046f0481d4c9/torch-2.3.0-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:d8ea5a465dbfd8501f33c937d1f693176c9aef9d1c1b0ca1d44ed7b0a18c52ac", size = 779141888 },
-    { url = "https://files.pythonhosted.org/packages/dc/82/7ab793eba3f4c3c84244e4c578ea838536d526db01c787448e8177404598/torch-2.3.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:09c81c5859a5b819956c6925a405ef1cdda393c9d8a01ce3851453f699d3358c", size = 88506981 },
-    { url = "https://files.pythonhosted.org/packages/2e/f7/503bab04f4e7a0a43f2ff05c3635cee9dfcf2e09656020d29502b87a94a3/torch-2.3.0-cp310-cp310-win_amd64.whl", hash = "sha256:1bf023aa20902586f614f7682fedfa463e773e26c58820b74158a72470259459", size = 159799740 },
-    { url = "https://files.pythonhosted.org/packages/01/c1/c6b42224122989ec95a820974aee92bdd4308380a7bb6ffa9a9d2429765d/torch-2.3.0-cp310-none-macosx_11_0_arm64.whl", hash = "sha256:758ef938de87a2653bba74b91f703458c15569f1562bf4b6c63c62d9c5a0c1f5", size = 60993271 },
-    { url = "https://files.pythonhosted.org/packages/35/3a/a39f354fa3119785be87e2f94ffa2620f8a270c8560f7356358ee62fb4c5/torch-2.3.0-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:493d54ee2f9df100b5ce1d18c96dbb8d14908721f76351e908c9d2622773a788", size = 779160265 },
-    { url = "https://files.pythonhosted.org/packages/91/3c/98a9b901b40278b40a9ff5b9c6be387b20997269f929f2182d8845c94085/torch-2.3.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:bce43af735c3da16cc14c7de2be7ad038e2fbf75654c2e274e575c6c05772ace", size = 88536251 },
-    { url = "https://files.pythonhosted.org/packages/2a/b7/a3cf5fd40334b9785cc83ee0c96b50603026eb3aa70210a33729018e7029/torch-2.3.0-cp311-cp311-win_amd64.whl", hash = "sha256:729804e97b7cf19ae9ab4181f91f5e612af07956f35c8b2c8e9d9f3596a8e877", size = 159803952 },
-    { url = "https://files.pythonhosted.org/packages/ad/08/c5e41eb22323db4a52260607598a207a2e1918916ae8201aa7a8ae005fcd/torch-2.3.0-cp311-none-macosx_11_0_arm64.whl", hash = "sha256:d24e328226d8e2af7cf80fcb1d2f1d108e0de32777fab4aaa2b37b9765d8be73", size = 60998957 },
-    { url = "https://files.pythonhosted.org/packages/51/03/1abad10990c76bee3703857b1617563b241f87d297ee466dbad922b0c308/torch-2.3.0-cp312-cp312-manylinux1_x86_64.whl", hash = "sha256:b0de2bdc0486ea7b14fc47ff805172df44e421a7318b7c4d92ef589a75d27410", size = 779062531 },
-    { url = "https://files.pythonhosted.org/packages/f1/9d/dfe273e19b7165148208bd182fac87488c5a0f7a3318d1646f5f37af872f/torch-2.3.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:a306c87a3eead1ed47457822c01dfbd459fe2920f2d38cbdf90de18f23f72542", size = 88437165 },
-    { url = "https://files.pythonhosted.org/packages/37/04/a5cd83baccbf2d4329990ec06b8abf3a644e1559a7b1f764f42d2cb77d51/torch-2.3.0-cp312-cp312-win_amd64.whl", hash = "sha256:f9b98bf1a3c8af2d4c41f0bf1433920900896c446d1ddc128290ff146d1eb4bd", size = 159749140 },
-    { url = "https://files.pythonhosted.org/packages/55/51/4bdee83e6fa9cca8e3a6cdf81a2695ede9d3fd7148e4fd4188dff142d7b0/torch-2.3.0-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:dca986214267b34065a79000cee54232e62b41dff1ec2cab9abc3fc8b3dee0ad", size = 60968873 },
-    { url = "https://files.pythonhosted.org/packages/da/af/abfe44a9ac7fd9705f145483e9e6c73a23d0533223082468eb2d978a494c/torch-2.3.0-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:cd0dc498b961ab19cb3f8dbf0c6c50e244f2f37dbfa05754ab44ea057c944ef9", size = 779140324 },
-    { url = "https://files.pythonhosted.org/packages/e6/b3/0ba7ab81889a964b027f9f46f7e1ddc49ac5532d5f7f09fc0730a9295bde/torch-2.3.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:e05f836559251e4096f3786ee99f4a8cbe67bc7fbedba8ad5e799681e47c5e80", size = 88508656 },
-    { url = "https://files.pythonhosted.org/packages/af/73/9fbe55b0db8e329b124df415bbdac760f4592e030d2cf20b05fd9aec68f5/torch-2.3.0-cp39-cp39-win_amd64.whl", hash = "sha256:4fb27b35dbb32303c2927da86e27b54a92209ddfb7234afb1949ea2b3effffea", size = 159734927 },
-    { url = "https://files.pythonhosted.org/packages/5b/e0/12a8ef2d7a00060ca5cbacef7b12f2b3e35fe3472973882e4a87165fea67/torch-2.3.0-cp39-none-macosx_11_0_arm64.whl", hash = "sha256:760f8bedff506ce9e6e103498f9b1e9e15809e008368594c3a66bf74a8a51380", size = 60993508 },
-]
-
-[[package]]
-name = "torchvision"
-version = "0.18.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "numpy" },
-    { name = "pillow" },
-    { name = "torch" },
-]
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/c5/e7/19e06d609f44f2c670db1056817deb928412ed329ff3f63cc2a27231b029/torchvision-0.18.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:dd61628a3d189c6852a12dc5ed4cd2eece66d2d67f35a866cb16f1dcb06c8c62", size = 1554228 },
-    { url = "https://files.pythonhosted.org/packages/d4/7e/d41b771dbffa927b9cc37372b1e18c881348cd18a0e4ad73f2c6bdf56c0e/torchvision-0.18.0-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:493c45f9937dad37aa1b64b14da17c7a589c72b91adc4837d431009cfe29bd53", size = 6954946 },
-    { url = "https://files.pythonhosted.org/packages/b2/dd/e5d39496413a5e5c2ca69d333bc241e7c8e8e412778c8309d54ce27cb9ec/torchvision-0.18.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:5337f6acfa1fe959d5cb340d01a00614d6b31ce7a4824ccb95435a85c5273b95", size = 13995854 },
-    { url = "https://files.pythonhosted.org/packages/f3/ef/f9559f3fc09362eafea5937521a2013a3ae67e38a7c6c3c9b51b3751c320/torchvision-0.18.0-cp310-cp310-win_amd64.whl", hash = "sha256:bd8e6f3b5beb49965f15c461302488edfa3d8c2d01d3bb79b150d6fb62711e3a", size = 1174147 },
-    { url = "https://files.pythonhosted.org/packages/b5/14/c05da13c98f528ba5fd99897320a7684df5dd136ec6faa6a5766f25e4a7e/torchvision-0.18.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6896a52168befe1105fb3c9335287390ed227e71d1e4ec4d68b62e8a3099fc09", size = 1554225 },
-    { url = "https://files.pythonhosted.org/packages/6e/7d/bc67ec2d1011226e75cdcc45a2c85d97b8eaac32a7d648b71c432d584367/torchvision-0.18.0-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:3d7955398d4ceaad77c487c2c44f6f7813112402c9bab8cd906d346005891048", size = 6955003 },
-    { url = "https://files.pythonhosted.org/packages/70/1d/107894816bf5ebecbc5a8556743fd89c0a1dd6da82b1c562d6becd5a108a/torchvision-0.18.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:e5a24d620cea14a4bb89f24aa2b506230c0a16a3ada57fc53ad80cfd256a2128", size = 13995855 },
-    { url = "https://files.pythonhosted.org/packages/12/c2/7c89c62f2b0a606070aa7cdb8af8af0c905562942778ebdd77600642c3b9/torchvision-0.18.0-cp311-cp311-win_amd64.whl", hash = "sha256:6ad70ddfa879bda5ed886b2518fe562640e0059787cbd65cb2bffa7674541410", size = 1174125 },
-    { url = "https://files.pythonhosted.org/packages/7c/12/49d4fd4ae7a48c6d33babb01a523594aca38365d378518320a10f9a5baa1/torchvision-0.18.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:eb9d83c0e1dbb54ecb0fb04c87f786333e3a6fb8b9c400aca7c31081f9aa5707", size = 1554232 },
-    { url = "https://files.pythonhosted.org/packages/6d/fe/729256fec03403b0bfdc229d4350936e29020d618f1740009c6a4f995b06/torchvision-0.18.0-cp312-cp312-manylinux1_x86_64.whl", hash = "sha256:b657d052d146f24cb3b2a78219bfc82ae70a9706671c50f632528907d10cccec", size = 6955047 },
-    { url = "https://files.pythonhosted.org/packages/51/2d/30883e9c6734546f9e7e0c429b76bd2b651aa25f6ced087c9c11cbf2ef41/torchvision-0.18.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:a964afbc7ddf50a46b941477f6c35729b416deedd139756befd488245e2e226d", size = 13996688 },
-    { url = "https://files.pythonhosted.org/packages/53/8a/864c3969af219a95213a5065d453313a96598e7c744b859e99b6ac134e16/torchvision-0.18.0-cp312-cp312-win_amd64.whl", hash = "sha256:7c770f0f748e0b17f57c0297508d7254f686cdf03fc2e2949f422b20574f4c0f", size = 1174123 },
-    { url = "https://files.pythonhosted.org/packages/d1/dc/bace7633ce37bc165fb6509e741d3cd73535f36e86b3ee5a95b2b647af92/torchvision-0.18.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:75e22ecf44a13b8f95b8ad421c0261282d859c61816badaca1959e073ccdd691", size = 1554270 },
-    { url = "https://files.pythonhosted.org/packages/54/20/309df7711dd17c399b7cd233e78bcee0f6fcdfbef5f2ec0930910c1884df/torchvision-0.18.0-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:4c334b3e719ba0a9ba6e15d4aff1178f5e6d029174f346163fed525f0ccfffd3", size = 6956173 },
-    { url = "https://files.pythonhosted.org/packages/72/73/b50f27719a9d0d9decae9f4264967a5c54a84046d1986772316d2471b6a2/torchvision-0.18.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:36efd87001c6bee2383e043e46a025affb03179747c8f4777b9918527ffce756", size = 1626203 },
-    { url = "https://files.pythonhosted.org/packages/db/4a/33a91dfcce5d62f826b15691a3bda2cdf62e4fbaf164f469af6a6756ebb4/torchvision-0.18.0-cp39-cp39-win_amd64.whl", hash = "sha256:ccc292e093771d5baacf5535ac4416306b6b5f15676341cd4d010d8542eace25", size = 1174149 },
+    { url = "https://files.pythonhosted.org/packages/63/28/110f7274254f1b8476c561dada127173f994afa2b1ffc044efb773c15650/torch-2.8.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:0be92c08b44009d4131d1ff7a8060d10bafdb7ddcb7359ef8d8c5169007ea905", size = 102052793, upload-time = "2025-08-06T14:53:15.852Z" },
+    { url = "https://files.pythonhosted.org/packages/70/1c/58da560016f81c339ae14ab16c98153d51c941544ae568da3cb5b1ceb572/torch-2.8.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:89aa9ee820bb39d4d72b794345cccef106b574508dd17dbec457949678c76011", size = 888025420, upload-time = "2025-08-06T14:54:18.014Z" },
+    { url = "https://files.pythonhosted.org/packages/70/87/f69752d0dd4ba8218c390f0438130c166fa264a33b7025adb5014b92192c/torch-2.8.0-cp310-cp310-win_amd64.whl", hash = "sha256:e8e5bf982e87e2b59d932769938b698858c64cc53753894be25629bdf5cf2f46", size = 241363614, upload-time = "2025-08-06T14:53:31.496Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/d6/e6d4c57e61c2b2175d3aafbfb779926a2cfd7c32eeda7c543925dceec923/torch-2.8.0-cp310-none-macosx_11_0_arm64.whl", hash = "sha256:a3f16a58a9a800f589b26d47ee15aca3acf065546137fc2af039876135f4c760", size = 73611154, upload-time = "2025-08-06T14:53:10.919Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/c4/3e7a3887eba14e815e614db70b3b529112d1513d9dae6f4d43e373360b7f/torch-2.8.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:220a06fd7af8b653c35d359dfe1aaf32f65aa85befa342629f716acb134b9710", size = 102073391, upload-time = "2025-08-06T14:53:20.937Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/63/4fdc45a0304536e75a5e1b1bbfb1b56dd0e2743c48ee83ca729f7ce44162/torch-2.8.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:c12fa219f51a933d5f80eeb3a7a5d0cbe9168c0a14bbb4055f1979431660879b", size = 888063640, upload-time = "2025-08-06T14:55:05.325Z" },
+    { url = "https://files.pythonhosted.org/packages/84/57/2f64161769610cf6b1c5ed782bd8a780e18a3c9d48931319f2887fa9d0b1/torch-2.8.0-cp311-cp311-win_amd64.whl", hash = "sha256:8c7ef765e27551b2fbfc0f41bcf270e1292d9bf79f8e0724848b1682be6e80aa", size = 241366752, upload-time = "2025-08-06T14:53:38.692Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/5e/05a5c46085d9b97e928f3f037081d3d2b87fb4b4195030fc099aaec5effc/torch-2.8.0-cp311-none-macosx_11_0_arm64.whl", hash = "sha256:5ae0524688fb6707c57a530c2325e13bb0090b745ba7b4a2cd6a3ce262572916", size = 73621174, upload-time = "2025-08-06T14:53:25.44Z" },
+    { url = "https://files.pythonhosted.org/packages/49/0c/2fd4df0d83a495bb5e54dca4474c4ec5f9c62db185421563deeb5dabf609/torch-2.8.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:e2fab4153768d433f8ed9279c8133a114a034a61e77a3a104dcdf54388838705", size = 101906089, upload-time = "2025-08-06T14:53:52.631Z" },
+    { url = "https://files.pythonhosted.org/packages/99/a8/6acf48d48838fb8fe480597d98a0668c2beb02ee4755cc136de92a0a956f/torch-2.8.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:b2aca0939fb7e4d842561febbd4ffda67a8e958ff725c1c27e244e85e982173c", size = 887913624, upload-time = "2025-08-06T14:56:44.33Z" },
+    { url = "https://files.pythonhosted.org/packages/af/8a/5c87f08e3abd825c7dfecef5a0f1d9aa5df5dd0e3fd1fa2f490a8e512402/torch-2.8.0-cp312-cp312-win_amd64.whl", hash = "sha256:2f4ac52f0130275d7517b03a33d2493bab3693c83dcfadf4f81688ea82147d2e", size = 241326087, upload-time = "2025-08-06T14:53:46.503Z" },
+    { url = "https://files.pythonhosted.org/packages/be/66/5c9a321b325aaecb92d4d1855421e3a055abd77903b7dab6575ca07796db/torch-2.8.0-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:619c2869db3ada2c0105487ba21b5008defcc472d23f8b80ed91ac4a380283b0", size = 73630478, upload-time = "2025-08-06T14:53:57.144Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/b0/a321f27270049baa12f5c3fb0d6ceea005634787e3af9a8d75dce8306b0a/torch-2.8.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:da6afa31c13b669d4ba49d8a2169f0db2c3ec6bec4af898aa714f401d4c38904", size = 102059214, upload-time = "2025-08-06T14:55:33.433Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/dd/1630cb51b10d3d2e97db95e5a84c32def81fc26b005bce6fc880b0e6db81/torch-2.8.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:06fcee8000e5c62a9f3e52a688b9c5abb7c6228d0e56e3452983416025c41381", size = 888024302, upload-time = "2025-08-06T14:57:28.23Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/dc/1f1f621afe15e3c496e1e8f94f8903f75f87e7d642d5a985e92210cc208d/torch-2.8.0-cp39-cp39-win_amd64.whl", hash = "sha256:5128fe752a355d9308e56af1ad28b15266fe2da5948660fad44de9e3a9e36e8c", size = 241249338, upload-time = "2025-08-06T14:57:05.669Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/95/ae26263aceb3d57b821179f827d0e321373ed49423e603dd5906ab14a730/torch-2.8.0-cp39-none-macosx_11_0_arm64.whl", hash = "sha256:e9f071f5b52a9f6970dc8a919694b27a91ae9dc08898b2b988abbef5eddfd1ae", size = 73610795, upload-time = "2025-08-06T14:57:11.513Z" },
 ]
 
 [[package]]
@@ -3374,19 +4411,20 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "colorama", marker = "sys_platform == 'win32'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a8/4b/29b4ef32e036bb34e4ab51796dd745cdba7ed47ad142a9f4a1eb8e0c744d/tqdm-4.67.1.tar.gz", hash = "sha256:f8aef9c52c08c13a65f30ea34f4e5aac3fd1a34959879d7e59e63027286627f2", size = 169737 }
+sdist = { url = "https://files.pythonhosted.org/packages/a8/4b/29b4ef32e036bb34e4ab51796dd745cdba7ed47ad142a9f4a1eb8e0c744d/tqdm-4.67.1.tar.gz", hash = "sha256:f8aef9c52c08c13a65f30ea34f4e5aac3fd1a34959879d7e59e63027286627f2", size = 169737, upload-time = "2024-11-24T20:12:22.481Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d0/30/dc54f88dd4a2b5dc8a0279bdd7270e735851848b762aeb1c1184ed1f6b14/tqdm-4.67.1-py3-none-any.whl", hash = "sha256:26445eca388f82e72884e0d580d5464cd801a3ea01e63e5601bdff9ba6a48de2", size = 78540 },
+    { url = "https://files.pythonhosted.org/packages/d0/30/dc54f88dd4a2b5dc8a0279bdd7270e735851848b762aeb1c1184ed1f6b14/tqdm-4.67.1-py3-none-any.whl", hash = "sha256:26445eca388f82e72884e0d580d5464cd801a3ea01e63e5601bdff9ba6a48de2", size = 78540, upload-time = "2024-11-24T20:12:19.698Z" },
 ]
 
 [[package]]
 name = "transformers"
-version = "4.48.0"
+version = "4.55.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "filelock" },
     { name = "huggingface-hub" },
-    { name = "numpy" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "packaging" },
     { name = "pyyaml" },
     { name = "regex" },
@@ -3395,557 +4433,535 @@ dependencies = [
     { name = "tokenizers" },
     { name = "tqdm" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ea/71/93a6331682d6f15adf7d646956db0c43e5f1759bbbd05f2ef53029bae107/transformers-4.48.0.tar.gz", hash = "sha256:03fdfcbfb8b0367fb6c9fbe9d1c9aa54dfd847618be9b52400b2811d22799cb1", size = 8372101 }
+sdist = { url = "https://files.pythonhosted.org/packages/2b/43/3cb831d5f28cc723516e5bb43a8c6042aca3038bb36b6bd6016b40dfd1e8/transformers-4.55.4.tar.gz", hash = "sha256:574a30559bc273c7a4585599ff28ab6b676e96dc56ffd2025ecfce2fd0ab915d", size = 9573015, upload-time = "2025-08-22T15:18:43.192Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/45/d6/a69764e89fc5c2c957aa473881527c8c35521108d553df703e9ba703daeb/transformers-4.48.0-py3-none-any.whl", hash = "sha256:6d3de6d71cb5f2a10f9775ccc17abce9620195caaf32ec96542bd2a6937f25b0", size = 9673380 },
+    { url = "https://files.pythonhosted.org/packages/fa/0a/8791a6ee0529c45f669566969e99b75e2ab20eb0bfee8794ce295c18bdad/transformers-4.55.4-py3-none-any.whl", hash = "sha256:df28f3849665faba4af5106f0db4510323277c4bb595055340544f7e59d06458", size = 11269659, upload-time = "2025-08-22T15:18:40.025Z" },
 ]
 
-[package.optional-dependencies]
-sentencepiece = [
-    { name = "protobuf" },
-    { name = "sentencepiece" },
+[[package]]
+name = "treescope"
+version = "0.1.10"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f0/2a/d13d3c38862632742d2fe2f7ae307c431db06538fd05ca03020d207b5dcc/treescope-0.1.10.tar.gz", hash = "sha256:20f74656f34ab2d8716715013e8163a0da79bdc2554c16d5023172c50d27ea95", size = 138870, upload-time = "2025-08-08T05:43:48.048Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/2b/36e984399089c026a6499ac8f7401d38487cf0183839a4aa78140d373771/treescope-0.1.10-py3-none-any.whl", hash = "sha256:dde52f5314f4c29d22157a6fe4d3bd103f9cae02791c9e672eefa32c9aa1da51", size = 182255, upload-time = "2025-08-08T05:43:46.673Z" },
 ]
 
 [[package]]
 name = "triton"
-version = "2.3.0"
+version = "3.4.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "filelock", marker = "python_full_version < '3.12'" },
+    { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
+    { name = "setuptools" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/db/ee/8d50d44ed5b63677bb387f4ee67a7dbaaded0189b320ffe82685a6827728/triton-2.3.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5ce4b8ff70c48e47274c66f269cce8861cf1dc347ceeb7a67414ca151b1822d8", size = 168055246 },
-    { url = "https://files.pythonhosted.org/packages/3c/00/84e0006f2025260fa111ddfc66194bd1af731b3ee18e2fd611a00f290b5e/triton-2.3.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3c3d9607f85103afdb279938fc1dd2a66e4f5999a58eb48a346bd42738f986dd", size = 168079300 },
-    { url = "https://files.pythonhosted.org/packages/90/2f/3e8f0ea4ef7ac6d8720a48b9b9700fd32c9909ee83b2eb1f25209ace0767/triton-2.3.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:218d742e67480d9581bafb73ed598416cc8a56f6316152e5562ee65e33de01c0", size = 168091361 },
-    { url = "https://files.pythonhosted.org/packages/8e/7b/8557e97f0f39992eaf390aacdf36c38a4de059caa9eda99321f7d3576d82/triton-2.3.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6d8f636e0341ac348899a47a057c3daea99ea7db31528a225a3ba4ded28ccc65", size = 168051036 },
+    { url = "https://files.pythonhosted.org/packages/62/ee/0ee5f64a87eeda19bbad9bc54ae5ca5b98186ed00055281fd40fb4beb10e/triton-3.4.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7ff2785de9bc02f500e085420273bb5cc9c9bb767584a4aa28d6e360cec70128", size = 155430069, upload-time = "2025-07-30T19:58:21.715Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/39/43325b3b651d50187e591eefa22e236b2981afcebaefd4f2fc0ea99df191/triton-3.4.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7b70f5e6a41e52e48cfc087436c8a28c17ff98db369447bcaff3b887a3ab4467", size = 155531138, upload-time = "2025-07-30T19:58:29.908Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/66/b1eb52839f563623d185f0927eb3530ee4d5ffe9d377cdaf5346b306689e/triton-3.4.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:31c1d84a5c0ec2c0f8e8a072d7fd150cab84a9c239eaddc6706c081bfae4eb04", size = 155560068, upload-time = "2025-07-30T19:58:37.081Z" },
+    { url = "https://files.pythonhosted.org/packages/12/34/1251beb5a3cb93f3950ebe68732752014646003ef6eb11eb5f1a37ca78cd/triton-3.4.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:98e5c1442eaeabae2e2452ae765801bd53cd4ce873cab0d1bdd59a32ab2d9397", size = 155430799, upload-time = "2025-07-30T19:58:57.664Z" },
 ]
 
 [[package]]
 name = "typing-extensions"
-version = "4.12.2"
+version = "4.15.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/72/94/1a15dd82efb362ac84269196e94cf00f187f7ed21c242792a923cdb1c61f/typing_extensions-4.15.0.tar.gz", hash = "sha256:0cea48d173cc12fa28ecabc3b837ea3cf6f38c6d1136f85cbaaf598984861466", size = 109391, upload-time = "2025-08-25T13:49:26.313Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/18/67/36e9267722cc04a6b9f15c7f3441c2363321a3ea07da7ae0c0707beb2a9c/typing_extensions-4.15.0-py3-none-any.whl", hash = "sha256:f0fa19c6845758ab08074a0cfa8b7aecb71c999ca73d62883bc25cc018c4e548", size = 44614, upload-time = "2025-08-25T13:49:24.86Z" },
+]
+
+[[package]]
+name = "typing-inspection"
+version = "0.4.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/df/db/f35a00659bc03fec321ba8bce9420de607a1d37f8342eee1863174c69557/typing_extensions-4.12.2.tar.gz", hash = "sha256:1a7ead55c7e559dd4dee8856e3a88b41225abfe1ce8df57b7c13915fe121ffb8", size = 85321 }
+dependencies = [
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f8/b1/0c11f5058406b3af7609f121aaa6b609744687f1d158b3c3a5bf4cc94238/typing_inspection-0.4.1.tar.gz", hash = "sha256:6ae134cc0203c33377d43188d4064e9b357dba58cff3185f22924610e70a9d28", size = 75726, upload-time = "2025-05-21T18:55:23.885Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/26/9f/ad63fc0248c5379346306f8668cda6e2e2e9c95e01216d2b8ffd9ff037d0/typing_extensions-4.12.2-py3-none-any.whl", hash = "sha256:04e5ca0351e0f3f85c6853954072df659d0d13fac324d0072316b67d7794700d", size = 37438 },
+    { url = "https://files.pythonhosted.org/packages/17/69/cd203477f944c353c31bade965f880aa1061fd6bf05ded0726ca845b6ff7/typing_inspection-0.4.1-py3-none-any.whl", hash = "sha256:389055682238f53b04f7badcb49b989835495a96700ced5dab2d8feae4b26f51", size = 14552, upload-time = "2025-05-21T18:55:22.152Z" },
 ]
 
 [[package]]
 name = "tzdata"
-version = "2024.2"
+version = "2025.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/e1/34/943888654477a574a86a98e9896bae89c7aa15078ec29f490fef2f1e5384/tzdata-2024.2.tar.gz", hash = "sha256:7d85cc416e9382e69095b7bdf4afd9e3880418a2413feec7069d533d6b4e31cc", size = 193282 }
+sdist = { url = "https://files.pythonhosted.org/packages/95/32/1a225d6164441be760d75c2c42e2780dc0873fe382da3e98a2e1e48361e5/tzdata-2025.2.tar.gz", hash = "sha256:b60a638fcc0daffadf82fe0f57e53d06bdec2f36c4df66280ae79bce6bd6f2b9", size = 196380, upload-time = "2025-03-23T13:54:43.652Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a6/ab/7e5f53c3b9d14972843a647d8d7a853969a58aecc7559cb3267302c94774/tzdata-2024.2-py2.py3-none-any.whl", hash = "sha256:a48093786cdcde33cad18c2555e8532f34422074448fbc874186f0abd79565cd", size = 346586 },
+    { url = "https://files.pythonhosted.org/packages/5c/23/c7abc0ca0a1526a0774eca151daeb8de62ec457e77262b66b359c3c7679e/tzdata-2025.2-py2.py3-none-any.whl", hash = "sha256:1a403fada01ff9221ca8044d701868fa132215d84beb92242d9acd2147f667a8", size = 347839, upload-time = "2025-03-23T13:54:41.845Z" },
 ]
 
 [[package]]
 name = "urllib3"
-version = "2.3.0"
+version = "2.5.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/aa/63/e53da845320b757bf29ef6a9062f5c669fe997973f966045cb019c3f4b66/urllib3-2.3.0.tar.gz", hash = "sha256:f8c5449b3cf0861679ce7e0503c7b44b5ec981bec0d1d3795a07f1ba96f0204d", size = 307268 }
+sdist = { url = "https://files.pythonhosted.org/packages/15/22/9ee70a2574a4f4599c47dd506532914ce044817c7752a79b6a51286319bc/urllib3-2.5.0.tar.gz", hash = "sha256:3fc47733c7e419d4bc3f6b3dc2b4f890bb743906a30d56ba4a5bfa4bbff92760", size = 393185, upload-time = "2025-06-18T14:07:41.644Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c8/19/4ec628951a74043532ca2cf5d97b7b14863931476d117c471e8e2b1eb39f/urllib3-2.3.0-py3-none-any.whl", hash = "sha256:1cee9ad369867bfdbbb48b7dd50374c0967a0bb7710050facf0dd6911440e3df", size = 128369 },
+    { url = "https://files.pythonhosted.org/packages/a7/c2/fe1e52489ae3122415c51f387e221dd0773709bad6c6cdaa599e8a2c5185/urllib3-2.5.0-py3-none-any.whl", hash = "sha256:e6b01673c0fa6a13e374b50871808eb3bf7046c4b125b216f6bf1cc604cff0dc", size = 129795, upload-time = "2025-06-18T14:07:40.39Z" },
 ]
 
 [[package]]
 name = "uvicorn"
-version = "0.34.0"
+version = "0.35.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "click" },
+    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "h11" },
     { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/4b/4d/938bd85e5bf2edeec766267a5015ad969730bb91e31b44021dfe8b22df6c/uvicorn-0.34.0.tar.gz", hash = "sha256:404051050cd7e905de2c9a7e61790943440b3416f49cb409f965d9dcd0fa73e9", size = 76568 }
+sdist = { url = "https://files.pythonhosted.org/packages/5e/42/e0e305207bb88c6b8d3061399c6a961ffe5fbb7e2aa63c9234df7259e9cd/uvicorn-0.35.0.tar.gz", hash = "sha256:bc662f087f7cf2ce11a1d7fd70b90c9f98ef2e2831556dd078d131b96cc94a01", size = 78473, upload-time = "2025-06-28T16:15:46.058Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/61/14/33a3a1352cfa71812a3a21e8c9bfb83f60b0011f5e36f2b1399d51928209/uvicorn-0.34.0-py3-none-any.whl", hash = "sha256:023dc038422502fa28a09c7a30bf2b6991512da7dcdb8fd35fe57cfc154126f4", size = 62315 },
-]
-
-[package.optional-dependencies]
-standard = [
-    { name = "colorama", marker = "sys_platform == 'win32'" },
-    { name = "httptools" },
-    { name = "python-dotenv" },
-    { name = "pyyaml" },
-    { name = "uvloop", marker = "platform_python_implementation != 'PyPy' and sys_platform != 'cygwin' and sys_platform != 'win32'" },
-    { name = "watchfiles" },
-    { name = "websockets" },
-]
-
-[[package]]
-name = "uvloop"
-version = "0.21.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/af/c0/854216d09d33c543f12a44b393c402e89a920b1a0a7dc634c42de91b9cf6/uvloop-0.21.0.tar.gz", hash = "sha256:3bf12b0fda68447806a7ad847bfa591613177275d35b6724b1ee573faa3704e3", size = 2492741 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/3d/76/44a55515e8c9505aa1420aebacf4dd82552e5e15691654894e90d0bd051a/uvloop-0.21.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:ec7e6b09a6fdded42403182ab6b832b71f4edaf7f37a9a0e371a01db5f0cb45f", size = 1442019 },
-    { url = "https://files.pythonhosted.org/packages/35/5a/62d5800358a78cc25c8a6c72ef8b10851bdb8cca22e14d9c74167b7f86da/uvloop-0.21.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:196274f2adb9689a289ad7d65700d37df0c0930fd8e4e743fa4834e850d7719d", size = 801898 },
-    { url = "https://files.pythonhosted.org/packages/f3/96/63695e0ebd7da6c741ccd4489b5947394435e198a1382349c17b1146bb97/uvloop-0.21.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f38b2e090258d051d68a5b14d1da7203a3c3677321cf32a95a6f4db4dd8b6f26", size = 3827735 },
-    { url = "https://files.pythonhosted.org/packages/61/e0/f0f8ec84979068ffae132c58c79af1de9cceeb664076beea86d941af1a30/uvloop-0.21.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:87c43e0f13022b998eb9b973b5e97200c8b90823454d4bc06ab33829e09fb9bb", size = 3825126 },
-    { url = "https://files.pythonhosted.org/packages/bf/fe/5e94a977d058a54a19df95f12f7161ab6e323ad49f4dabc28822eb2df7ea/uvloop-0.21.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:10d66943def5fcb6e7b37310eb6b5639fd2ccbc38df1177262b0640c3ca68c1f", size = 3705789 },
-    { url = "https://files.pythonhosted.org/packages/26/dd/c7179618e46092a77e036650c1f056041a028a35c4d76945089fcfc38af8/uvloop-0.21.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:67dd654b8ca23aed0a8e99010b4c34aca62f4b7fce88f39d452ed7622c94845c", size = 3800523 },
-    { url = "https://files.pythonhosted.org/packages/57/a7/4cf0334105c1160dd6819f3297f8700fda7fc30ab4f61fbf3e725acbc7cc/uvloop-0.21.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:c0f3fa6200b3108919f8bdabb9a7f87f20e7097ea3c543754cabc7d717d95cf8", size = 1447410 },
-    { url = "https://files.pythonhosted.org/packages/8c/7c/1517b0bbc2dbe784b563d6ab54f2ef88c890fdad77232c98ed490aa07132/uvloop-0.21.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:0878c2640cf341b269b7e128b1a5fed890adc4455513ca710d77d5e93aa6d6a0", size = 805476 },
-    { url = "https://files.pythonhosted.org/packages/ee/ea/0bfae1aceb82a503f358d8d2fa126ca9dbdb2ba9c7866974faec1cb5875c/uvloop-0.21.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b9fb766bb57b7388745d8bcc53a359b116b8a04c83a2288069809d2b3466c37e", size = 3960855 },
-    { url = "https://files.pythonhosted.org/packages/8a/ca/0864176a649838b838f36d44bf31c451597ab363b60dc9e09c9630619d41/uvloop-0.21.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a375441696e2eda1c43c44ccb66e04d61ceeffcd76e4929e527b7fa401b90fb", size = 3973185 },
-    { url = "https://files.pythonhosted.org/packages/30/bf/08ad29979a936d63787ba47a540de2132169f140d54aa25bc8c3df3e67f4/uvloop-0.21.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:baa0e6291d91649c6ba4ed4b2f982f9fa165b5bbd50a9e203c416a2797bab3c6", size = 3820256 },
-    { url = "https://files.pythonhosted.org/packages/da/e2/5cf6ef37e3daf2f06e651aae5ea108ad30df3cb269102678b61ebf1fdf42/uvloop-0.21.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4509360fcc4c3bd2c70d87573ad472de40c13387f5fda8cb58350a1d7475e58d", size = 3937323 },
-    { url = "https://files.pythonhosted.org/packages/8c/4c/03f93178830dc7ce8b4cdee1d36770d2f5ebb6f3d37d354e061eefc73545/uvloop-0.21.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:359ec2c888397b9e592a889c4d72ba3d6befba8b2bb01743f72fffbde663b59c", size = 1471284 },
-    { url = "https://files.pythonhosted.org/packages/43/3e/92c03f4d05e50f09251bd8b2b2b584a2a7f8fe600008bcc4523337abe676/uvloop-0.21.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f7089d2dc73179ce5ac255bdf37c236a9f914b264825fdaacaded6990a7fb4c2", size = 821349 },
-    { url = "https://files.pythonhosted.org/packages/a6/ef/a02ec5da49909dbbfb1fd205a9a1ac4e88ea92dcae885e7c961847cd51e2/uvloop-0.21.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:baa4dcdbd9ae0a372f2167a207cd98c9f9a1ea1188a8a526431eef2f8116cc8d", size = 4580089 },
-    { url = "https://files.pythonhosted.org/packages/06/a7/b4e6a19925c900be9f98bec0a75e6e8f79bb53bdeb891916609ab3958967/uvloop-0.21.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86975dca1c773a2c9864f4c52c5a55631038e387b47eaf56210f873887b6c8dc", size = 4693770 },
-    { url = "https://files.pythonhosted.org/packages/ce/0c/f07435a18a4b94ce6bd0677d8319cd3de61f3a9eeb1e5f8ab4e8b5edfcb3/uvloop-0.21.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:461d9ae6660fbbafedd07559c6a2e57cd553b34b0065b6550685f6653a98c1cb", size = 4451321 },
-    { url = "https://files.pythonhosted.org/packages/8f/eb/f7032be105877bcf924709c97b1bf3b90255b4ec251f9340cef912559f28/uvloop-0.21.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:183aef7c8730e54c9a3ee3227464daed66e37ba13040bb3f350bc2ddc040f22f", size = 4659022 },
-    { url = "https://files.pythonhosted.org/packages/3f/8d/2cbef610ca21539f0f36e2b34da49302029e7c9f09acef0b1c3b5839412b/uvloop-0.21.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:bfd55dfcc2a512316e65f16e503e9e450cab148ef11df4e4e679b5e8253a5281", size = 1468123 },
-    { url = "https://files.pythonhosted.org/packages/93/0d/b0038d5a469f94ed8f2b2fce2434a18396d8fbfb5da85a0a9781ebbdec14/uvloop-0.21.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:787ae31ad8a2856fc4e7c095341cccc7209bd657d0e71ad0dc2ea83c4a6fa8af", size = 819325 },
-    { url = "https://files.pythonhosted.org/packages/50/94/0a687f39e78c4c1e02e3272c6b2ccdb4e0085fda3b8352fecd0410ccf915/uvloop-0.21.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5ee4d4ef48036ff6e5cfffb09dd192c7a5027153948d85b8da7ff705065bacc6", size = 4582806 },
-    { url = "https://files.pythonhosted.org/packages/d2/19/f5b78616566ea68edd42aacaf645adbf71fbd83fc52281fba555dc27e3f1/uvloop-0.21.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f3df876acd7ec037a3d005b3ab85a7e4110422e4d9c1571d4fc89b0fc41b6816", size = 4701068 },
-    { url = "https://files.pythonhosted.org/packages/47/57/66f061ee118f413cd22a656de622925097170b9380b30091b78ea0c6ea75/uvloop-0.21.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:bd53ecc9a0f3d87ab847503c2e1552b690362e005ab54e8a48ba97da3924c0dc", size = 4454428 },
-    { url = "https://files.pythonhosted.org/packages/63/9a/0962b05b308494e3202d3f794a6e85abe471fe3cafdbcf95c2e8c713aabd/uvloop-0.21.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a5c39f217ab3c663dc699c04cbd50c13813e31d917642d459fdcec07555cc553", size = 4660018 },
-    { url = "https://files.pythonhosted.org/packages/3c/a4/646a9d0edff7cde25fc1734695d3dfcee0501140dd0e723e4df3f0a50acb/uvloop-0.21.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:c097078b8031190c934ed0ebfee8cc5f9ba9642e6eb88322b9958b649750f72b", size = 1439646 },
-    { url = "https://files.pythonhosted.org/packages/01/2e/e128c66106af9728f86ebfeeb52af27ecd3cb09336f3e2f3e06053707a15/uvloop-0.21.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:46923b0b5ee7fc0020bef24afe7836cb068f5050ca04caf6b487c513dc1a20b2", size = 800931 },
-    { url = "https://files.pythonhosted.org/packages/2d/1a/9fbc2b1543d0df11f7aed1632f64bdf5ecc4053cf98cdc9edb91a65494f9/uvloop-0.21.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:53e420a3afe22cdcf2a0f4846e377d16e718bc70103d7088a4f7623567ba5fb0", size = 3829660 },
-    { url = "https://files.pythonhosted.org/packages/b8/c0/392e235e4100ae3b95b5c6dac77f82b529d2760942b1e7e0981e5d8e895d/uvloop-0.21.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88cb67cdbc0e483da00af0b2c3cdad4b7c61ceb1ee0f33fe00e09c81e3a6cb75", size = 3827185 },
-    { url = "https://files.pythonhosted.org/packages/e1/24/a5da6aba58f99aed5255eca87d58d1760853e8302d390820cc29058408e3/uvloop-0.21.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:221f4f2a1f46032b403bf3be628011caf75428ee3cc204a22addf96f586b19fd", size = 3705833 },
-    { url = "https://files.pythonhosted.org/packages/1a/5c/6ba221bb60f1e6474474102e17e38612ec7a06dc320e22b687ab563d877f/uvloop-0.21.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:2d1f581393673ce119355d56da84fe1dd9d2bb8b3d13ce792524e1607139feff", size = 3804696 },
+    { url = "https://files.pythonhosted.org/packages/d2/e2/dc81b1bd1dcfe91735810265e9d26bc8ec5da45b4c0f6237e286819194c3/uvicorn-0.35.0-py3-none-any.whl", hash = "sha256:197535216b25ff9b785e29a0b79199f55222193d47f820816e7da751e9bc8d4a", size = 66406, upload-time = "2025-06-28T16:15:44.816Z" },
 ]
 
 [[package]]
 name = "virtualenv"
-version = "20.29.1"
+version = "20.34.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "distlib" },
     { name = "filelock" },
     { name = "platformdirs" },
+    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a7/ca/f23dcb02e161a9bba141b1c08aa50e8da6ea25e6d780528f1d385a3efe25/virtualenv-20.29.1.tar.gz", hash = "sha256:b8b8970138d32fb606192cb97f6cd4bb644fa486be9308fb9b63f81091b5dc35", size = 7658028 }
+sdist = { url = "https://files.pythonhosted.org/packages/1c/14/37fcdba2808a6c615681cd216fecae00413c9dab44fb2e57805ecf3eaee3/virtualenv-20.34.0.tar.gz", hash = "sha256:44815b2c9dee7ed86e387b842a84f20b93f7f417f95886ca1996a72a4138eb1a", size = 6003808, upload-time = "2025-08-13T14:24:07.464Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/89/9b/599bcfc7064fbe5740919e78c5df18e5dceb0887e676256a1061bb5ae232/virtualenv-20.29.1-py3-none-any.whl", hash = "sha256:4e4cb403c0b0da39e13b46b1b2476e505cb0046b25f242bee80f62bf990b2779", size = 4282379 },
+    { url = "https://files.pythonhosted.org/packages/76/06/04c8e804f813cf972e3262f3f8584c232de64f0cde9f703b46cf53a45090/virtualenv-20.34.0-py3-none-any.whl", hash = "sha256:341f5afa7eee943e4984a9207c025feedd768baff6753cd660c857ceb3e36026", size = 5983279, upload-time = "2025-08-13T14:24:05.111Z" },
 ]
 
 [[package]]
 name = "vllm"
-version = "0.5.1"
+version = "0.1.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "aiohttp" },
-    { name = "cmake" },
     { name = "fastapi" },
-    { name = "filelock" },
-    { name = "lm-format-enforcer" },
+    { name = "fschat" },
     { name = "ninja" },
-    { name = "numpy" },
-    { name = "nvidia-ml-py" },
-    { name = "openai" },
-    { name = "outlines" },
-    { name = "pillow" },
-    { name = "prometheus-client" },
-    { name = "prometheus-fastapi-instrumentator" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "psutil" },
-    { name = "py-cpuinfo" },
     { name = "pydantic" },
     { name = "ray" },
-    { name = "requests" },
     { name = "sentencepiece" },
-    { name = "tiktoken" },
-    { name = "tokenizers" },
     { name = "torch" },
-    { name = "torchvision" },
-    { name = "tqdm" },
     { name = "transformers" },
-    { name = "typing-extensions" },
-    { name = "uvicorn", extra = ["standard"] },
-    { name = "vllm-flash-attn" },
+    { name = "uvicorn" },
     { name = "xformers" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fc/d2/38993991ee61bacaccd49182ce597bf4496bc4ac96e92a3b011fcde90bb7/vllm-0.5.1.tar.gz", hash = "sha256:c7b6d01ec0644dd251e00a52bd08f9a172da2185dce87538deb56c577120e0eb", size = 790562 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/82/43/a0fbf45a2ccf038ea05774fba715633c70f7a26c7b3270c337038f9f77c0/vllm-0.5.1-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:562e3dd1d54ad0afb299e64412322b372976faaebb70a7db8e4a6c4ef5ed67c4", size = 146906281 },
-    { url = "https://files.pythonhosted.org/packages/13/79/e212b0eae62716be0d75994b71b7f97f4a45a7ffd8533408bcd3cde6a7e0/vllm-0.5.1-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:8b7483adcfc115dc7237663636a0976943fa65cde6f616a3839f0de69a419b5c", size = 146906182 },
-    { url = "https://files.pythonhosted.org/packages/72/89/671a176610d4ffe04216dccde6bffd07a8c67f4c0c45ce8be60266d7f574/vllm-0.5.1-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:6e6cad87335713cfc6710bc44c2688a6158abc3184d3530731e89b948617b2c7", size = 146906268 },
-]
+sdist = { url = "https://files.pythonhosted.org/packages/f9/ed/1e4625f290e24c1381aeeb66620b32755b17f2894a4227fb49fb20fa0ec9/vllm-0.1.2.tar.gz", hash = "sha256:6fff59a918ec822c3d7a9060c3af5604a1bbf26ec4db15656678768019aefad7", size = 93849, upload-time = "2023-07-05T04:53:21.847Z" }
 
 [[package]]
-name = "vllm-flash-attn"
-version = "2.5.9"
+name = "watchdog"
+version = "6.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/db/7d/7f3d619e951c88ed75c6037b246ddcf2d322812ee8ea189be89511721d54/watchdog-6.0.0.tar.gz", hash = "sha256:9ddf7c82fda3ae8e24decda1338ede66e1c99883db93711d8fb941eaa2d8c282", size = 131220, upload-time = "2024-11-01T14:07:13.037Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0c/56/90994d789c61df619bfc5ce2ecdabd5eeff564e1eb47512bd01b5e019569/watchdog-6.0.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d1cdb490583ebd691c012b3d6dae011000fe42edb7a82ece80965b42abd61f26", size = 96390, upload-time = "2024-11-01T14:06:24.793Z" },
+    { url = "https://files.pythonhosted.org/packages/55/46/9a67ee697342ddf3c6daa97e3a587a56d6c4052f881ed926a849fcf7371c/watchdog-6.0.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:bc64ab3bdb6a04d69d4023b29422170b74681784ffb9463ed4870cf2f3e66112", size = 88389, upload-time = "2024-11-01T14:06:27.112Z" },
+    { url = "https://files.pythonhosted.org/packages/44/65/91b0985747c52064d8701e1075eb96f8c40a79df889e59a399453adfb882/watchdog-6.0.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c897ac1b55c5a1461e16dae288d22bb2e412ba9807df8397a635d88f671d36c3", size = 89020, upload-time = "2024-11-01T14:06:29.876Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/24/d9be5cd6642a6aa68352ded4b4b10fb0d7889cb7f45814fb92cecd35f101/watchdog-6.0.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6eb11feb5a0d452ee41f824e271ca311a09e250441c262ca2fd7ebcf2461a06c", size = 96393, upload-time = "2024-11-01T14:06:31.756Z" },
+    { url = "https://files.pythonhosted.org/packages/63/7a/6013b0d8dbc56adca7fdd4f0beed381c59f6752341b12fa0886fa7afc78b/watchdog-6.0.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:ef810fbf7b781a5a593894e4f439773830bdecb885e6880d957d5b9382a960d2", size = 88392, upload-time = "2024-11-01T14:06:32.99Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/40/b75381494851556de56281e053700e46bff5b37bf4c7267e858640af5a7f/watchdog-6.0.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:afd0fe1b2270917c5e23c2a65ce50c2a4abb63daafb0d419fde368e272a76b7c", size = 89019, upload-time = "2024-11-01T14:06:34.963Z" },
+    { url = "https://files.pythonhosted.org/packages/39/ea/3930d07dafc9e286ed356a679aa02d777c06e9bfd1164fa7c19c288a5483/watchdog-6.0.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:bdd4e6f14b8b18c334febb9c4425a878a2ac20efd1e0b231978e7b150f92a948", size = 96471, upload-time = "2024-11-01T14:06:37.745Z" },
+    { url = "https://files.pythonhosted.org/packages/12/87/48361531f70b1f87928b045df868a9fd4e253d9ae087fa4cf3f7113be363/watchdog-6.0.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c7c15dda13c4eb00d6fb6fc508b3c0ed88b9d5d374056b239c4ad1611125c860", size = 88449, upload-time = "2024-11-01T14:06:39.748Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/7e/8f322f5e600812e6f9a31b75d242631068ca8f4ef0582dd3ae6e72daecc8/watchdog-6.0.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6f10cb2d5902447c7d0da897e2c6768bca89174d0c6e1e30abec5421af97a5b0", size = 89054, upload-time = "2024-11-01T14:06:41.009Z" },
+    { url = "https://files.pythonhosted.org/packages/05/52/7223011bb760fce8ddc53416beb65b83a3ea6d7d13738dde75eeb2c89679/watchdog-6.0.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:e6f0e77c9417e7cd62af82529b10563db3423625c5fce018430b249bf977f9e8", size = 96390, upload-time = "2024-11-01T14:06:49.325Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/62/d2b21bc4e706d3a9d467561f487c2938cbd881c69f3808c43ac1ec242391/watchdog-6.0.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:90c8e78f3b94014f7aaae121e6b909674df5b46ec24d6bebc45c44c56729af2a", size = 88386, upload-time = "2024-11-01T14:06:50.536Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/22/1c90b20eda9f4132e4603a26296108728a8bfe9584b006bd05dd94548853/watchdog-6.0.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e7631a77ffb1f7d2eefa4445ebbee491c720a5661ddf6df3498ebecae5ed375c", size = 89017, upload-time = "2024-11-01T14:06:51.717Z" },
+    { url = "https://files.pythonhosted.org/packages/30/ad/d17b5d42e28a8b91f8ed01cb949da092827afb9995d4559fd448d0472763/watchdog-6.0.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:c7ac31a19f4545dd92fc25d200694098f42c9a8e391bc00bdd362c5736dbf881", size = 87902, upload-time = "2024-11-01T14:06:53.119Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/ca/c3649991d140ff6ab67bfc85ab42b165ead119c9e12211e08089d763ece5/watchdog-6.0.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:9513f27a1a582d9808cf21a07dae516f0fab1cf2d7683a742c498b93eedabb11", size = 88380, upload-time = "2024-11-01T14:06:55.19Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/79/69f2b0e8d3f2afd462029031baafb1b75d11bb62703f0e1022b2e54d49ee/watchdog-6.0.0-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:7a0e56874cfbc4b9b05c60c8a1926fedf56324bb08cfbc188969777940aef3aa", size = 87903, upload-time = "2024-11-01T14:06:57.052Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/2b/dc048dd71c2e5f0f7ebc04dd7912981ec45793a03c0dc462438e0591ba5d/watchdog-6.0.0-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:e6439e374fc012255b4ec786ae3c4bc838cd7309a540e5fe0952d03687d8804e", size = 88381, upload-time = "2024-11-01T14:06:58.193Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/c7/ca4bf3e518cb57a686b2feb4f55a1892fd9a3dd13f470fca14e00f80ea36/watchdog-6.0.0-py3-none-manylinux2014_aarch64.whl", hash = "sha256:7607498efa04a3542ae3e05e64da8202e58159aa1fa4acddf7678d34a35d4f13", size = 79079, upload-time = "2024-11-01T14:06:59.472Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/51/d46dc9332f9a647593c947b4b88e2381c8dfc0942d15b8edc0310fa4abb1/watchdog-6.0.0-py3-none-manylinux2014_armv7l.whl", hash = "sha256:9041567ee8953024c83343288ccc458fd0a2d811d6a0fd68c4c22609e3490379", size = 79078, upload-time = "2024-11-01T14:07:01.431Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/57/04edbf5e169cd318d5f07b4766fee38e825d64b6913ca157ca32d1a42267/watchdog-6.0.0-py3-none-manylinux2014_i686.whl", hash = "sha256:82dc3e3143c7e38ec49d61af98d6558288c415eac98486a5c581726e0737c00e", size = 79076, upload-time = "2024-11-01T14:07:02.568Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/cc/da8422b300e13cb187d2203f20b9253e91058aaf7db65b74142013478e66/watchdog-6.0.0-py3-none-manylinux2014_ppc64.whl", hash = "sha256:212ac9b8bf1161dc91bd09c048048a95ca3a4c4f5e5d4a7d1b1a7d5752a7f96f", size = 79077, upload-time = "2024-11-01T14:07:03.893Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/3b/b8964e04ae1a025c44ba8e4291f86e97fac443bca31de8bd98d3263d2fcf/watchdog-6.0.0-py3-none-manylinux2014_ppc64le.whl", hash = "sha256:e3df4cbb9a450c6d49318f6d14f4bbc80d763fa587ba46ec86f99f9e6876bb26", size = 79078, upload-time = "2024-11-01T14:07:05.189Z" },
+    { url = "https://files.pythonhosted.org/packages/62/ae/a696eb424bedff7407801c257d4b1afda455fe40821a2be430e173660e81/watchdog-6.0.0-py3-none-manylinux2014_s390x.whl", hash = "sha256:2cce7cfc2008eb51feb6aab51251fd79b85d9894e98ba847408f662b3395ca3c", size = 79077, upload-time = "2024-11-01T14:07:06.376Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/e8/dbf020b4d98251a9860752a094d09a65e1b436ad181faf929983f697048f/watchdog-6.0.0-py3-none-manylinux2014_x86_64.whl", hash = "sha256:20ffe5b202af80ab4266dcd3e91aae72bf2da48c0d33bdb15c66658e685e94e2", size = 79078, upload-time = "2024-11-01T14:07:07.547Z" },
+    { url = "https://files.pythonhosted.org/packages/07/f6/d0e5b343768e8bcb4cda79f0f2f55051bf26177ecd5651f84c07567461cf/watchdog-6.0.0-py3-none-win32.whl", hash = "sha256:07df1fdd701c5d4c8e55ef6cf55b8f0120fe1aef7ef39a1c6fc6bc2e606d517a", size = 79065, upload-time = "2024-11-01T14:07:09.525Z" },
+    { url = "https://files.pythonhosted.org/packages/db/d9/c495884c6e548fce18a8f40568ff120bc3a4b7b99813081c8ac0c936fa64/watchdog-6.0.0-py3-none-win_amd64.whl", hash = "sha256:cbafb470cf848d93b5d013e2ecb245d4aa1c8fd0504e863ccefa32445359d680", size = 79070, upload-time = "2024-11-01T14:07:10.686Z" },
+    { url = "https://files.pythonhosted.org/packages/33/e8/e40370e6d74ddba47f002a32919d91310d6074130fe4e17dabcafc15cbf1/watchdog-6.0.0-py3-none-win_ia64.whl", hash = "sha256:a1914259fa9e1454315171103c6a30961236f508b9b623eae470268bbcc6a22f", size = 79067, upload-time = "2024-11-01T14:07:11.845Z" },
+]
+
+[[package]]
+name = "wavedrom"
+version = "2.0.3.post3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "torch" },
+    { name = "pyyaml" },
+    { name = "six" },
+    { name = "svgwrite" },
 ]
+sdist = { url = "https://files.pythonhosted.org/packages/be/71/6739e3abac630540aaeaaece4584c39f88b5f8658ce6ca517efec455e3de/wavedrom-2.0.3.post3.tar.gz", hash = "sha256:327b4d5dca593c81257c202fea516f7a908747fb11527c359f034f5b7af7f47b", size = 137737, upload-time = "2022-05-27T11:53:09.675Z" }
+
+[[package]]
+name = "wcwidth"
+version = "0.2.13"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/6c/63/53559446a878410fc5a5974feb13d31d78d752eb18aeba59c7fef1af7598/wcwidth-0.2.13.tar.gz", hash = "sha256:72ea0c06399eb286d978fdedb6923a9eb47e1c486ce63e9b4e64fc18303972b5", size = 101301, upload-time = "2024-01-06T02:10:57.829Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/2c/35/b00f367f3655bfed85a0df3c32fe6af34c0fd38b28ec7f0f9363991615f1/vllm_flash_attn-2.5.9-cp310-cp310-manylinux1_x86_64.whl", hash = "sha256:3e33af65880e99c47ff636f465d45c08e5e4c2eedbc06c9970599a6408768944", size = 37095920 },
-    { url = "https://files.pythonhosted.org/packages/36/42/307d9afa8c884efd369f7e402e69a588dea22b23fb23ce540b9ef6a284b0/vllm_flash_attn-2.5.9-cp311-cp311-manylinux1_x86_64.whl", hash = "sha256:88a63365db204859131fd8b3a0e85ea3b8b726f4be5ddbf2c6211a94a2cf4258", size = 37112678 },
-    { url = "https://files.pythonhosted.org/packages/f4/f2/155a33878f747b09250cf5a73e0449f52d23e723c770913952aa5340d66a/vllm_flash_attn-2.5.9-cp39-cp39-manylinux1_x86_64.whl", hash = "sha256:384845ae339946cb46e11cdfb5fadda1c0e519d2312b595e04da0df67a6d5383", size = 37092795 },
+    { url = "https://files.pythonhosted.org/packages/fd/84/fd2ba7aafacbad3c4201d395674fc6348826569da3c0937e75505ead3528/wcwidth-0.2.13-py2.py3-none-any.whl", hash = "sha256:3da69048e4540d84af32131829ff948f1e022c1c6bdb8d6102117aac784f6859", size = 34166, upload-time = "2024-01-06T02:10:55.763Z" },
 ]
 
 [[package]]
-name = "watchfiles"
-version = "1.0.4"
+name = "websockets"
+version = "15.0.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/21/e6/26d09fab466b7ca9c7737474c52be4f76a40301b08362eb2dbc19dcc16c1/websockets-15.0.1.tar.gz", hash = "sha256:82544de02076bafba038ce055ee6412d68da13ab47f0c60cab827346de828dee", size = 177016, upload-time = "2025-03-05T20:03:41.606Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1e/da/6462a9f510c0c49837bbc9345aca92d767a56c1fb2939e1579df1e1cdcf7/websockets-15.0.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d63efaa0cd96cf0c5fe4d581521d9fa87744540d4bc999ae6e08595a1014b45b", size = 175423, upload-time = "2025-03-05T20:01:35.363Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/9f/9d11c1a4eb046a9e106483b9ff69bce7ac880443f00e5ce64261b47b07e7/websockets-15.0.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:ac60e3b188ec7574cb761b08d50fcedf9d77f1530352db4eef1707fe9dee7205", size = 173080, upload-time = "2025-03-05T20:01:37.304Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/4f/b462242432d93ea45f297b6179c7333dd0402b855a912a04e7fc61c0d71f/websockets-15.0.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:5756779642579d902eed757b21b0164cd6fe338506a8083eb58af5c372e39d9a", size = 173329, upload-time = "2025-03-05T20:01:39.668Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/0c/6afa1f4644d7ed50284ac59cc70ef8abd44ccf7d45850d989ea7310538d0/websockets-15.0.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0fdfe3e2a29e4db3659dbd5bbf04560cea53dd9610273917799f1cde46aa725e", size = 182312, upload-time = "2025-03-05T20:01:41.815Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/d4/ffc8bd1350b229ca7a4db2a3e1c482cf87cea1baccd0ef3e72bc720caeec/websockets-15.0.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4c2529b320eb9e35af0fa3016c187dffb84a3ecc572bcee7c3ce302bfeba52bf", size = 181319, upload-time = "2025-03-05T20:01:43.967Z" },
+    { url = "https://files.pythonhosted.org/packages/97/3a/5323a6bb94917af13bbb34009fac01e55c51dfde354f63692bf2533ffbc2/websockets-15.0.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ac1e5c9054fe23226fb11e05a6e630837f074174c4c2f0fe442996112a6de4fb", size = 181631, upload-time = "2025-03-05T20:01:46.104Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/cc/1aeb0f7cee59ef065724041bb7ed667b6ab1eeffe5141696cccec2687b66/websockets-15.0.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:5df592cd503496351d6dc14f7cdad49f268d8e618f80dce0cd5a36b93c3fc08d", size = 182016, upload-time = "2025-03-05T20:01:47.603Z" },
+    { url = "https://files.pythonhosted.org/packages/79/f9/c86f8f7af208e4161a7f7e02774e9d0a81c632ae76db2ff22549e1718a51/websockets-15.0.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:0a34631031a8f05657e8e90903e656959234f3a04552259458aac0b0f9ae6fd9", size = 181426, upload-time = "2025-03-05T20:01:48.949Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/b9/828b0bc6753db905b91df6ae477c0b14a141090df64fb17f8a9d7e3516cf/websockets-15.0.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:3d00075aa65772e7ce9e990cab3ff1de702aa09be3940d1dc88d5abf1ab8a09c", size = 181360, upload-time = "2025-03-05T20:01:50.938Z" },
+    { url = "https://files.pythonhosted.org/packages/89/fb/250f5533ec468ba6327055b7d98b9df056fb1ce623b8b6aaafb30b55d02e/websockets-15.0.1-cp310-cp310-win32.whl", hash = "sha256:1234d4ef35db82f5446dca8e35a7da7964d02c127b095e172e54397fb6a6c256", size = 176388, upload-time = "2025-03-05T20:01:52.213Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/46/aca7082012768bb98e5608f01658ff3ac8437e563eca41cf068bd5849a5e/websockets-15.0.1-cp310-cp310-win_amd64.whl", hash = "sha256:39c1fec2c11dc8d89bba6b2bf1556af381611a173ac2b511cf7231622058af41", size = 176830, upload-time = "2025-03-05T20:01:53.922Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/32/18fcd5919c293a398db67443acd33fde142f283853076049824fc58e6f75/websockets-15.0.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:823c248b690b2fd9303ba00c4f66cd5e2d8c3ba4aa968b2779be9532a4dad431", size = 175423, upload-time = "2025-03-05T20:01:56.276Z" },
+    { url = "https://files.pythonhosted.org/packages/76/70/ba1ad96b07869275ef42e2ce21f07a5b0148936688c2baf7e4a1f60d5058/websockets-15.0.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:678999709e68425ae2593acf2e3ebcbcf2e69885a5ee78f9eb80e6e371f1bf57", size = 173082, upload-time = "2025-03-05T20:01:57.563Z" },
+    { url = "https://files.pythonhosted.org/packages/86/f2/10b55821dd40eb696ce4704a87d57774696f9451108cff0d2824c97e0f97/websockets-15.0.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:d50fd1ee42388dcfb2b3676132c78116490976f1300da28eb629272d5d93e905", size = 173330, upload-time = "2025-03-05T20:01:59.063Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/90/1c37ae8b8a113d3daf1065222b6af61cc44102da95388ac0018fcb7d93d9/websockets-15.0.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d99e5546bf73dbad5bf3547174cd6cb8ba7273062a23808ffea025ecb1cf8562", size = 182878, upload-time = "2025-03-05T20:02:00.305Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/8d/96e8e288b2a41dffafb78e8904ea7367ee4f891dafc2ab8d87e2124cb3d3/websockets-15.0.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:66dd88c918e3287efc22409d426c8f729688d89a0c587c88971a0faa2c2f3792", size = 181883, upload-time = "2025-03-05T20:02:03.148Z" },
+    { url = "https://files.pythonhosted.org/packages/93/1f/5d6dbf551766308f6f50f8baf8e9860be6182911e8106da7a7f73785f4c4/websockets-15.0.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8dd8327c795b3e3f219760fa603dcae1dcc148172290a8ab15158cf85a953413", size = 182252, upload-time = "2025-03-05T20:02:05.29Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/78/2d4fed9123e6620cbf1706c0de8a1632e1a28e7774d94346d7de1bba2ca3/websockets-15.0.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8fdc51055e6ff4adeb88d58a11042ec9a5eae317a0a53d12c062c8a8865909e8", size = 182521, upload-time = "2025-03-05T20:02:07.458Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/3b/66d4c1b444dd1a9823c4a81f50231b921bab54eee2f69e70319b4e21f1ca/websockets-15.0.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:693f0192126df6c2327cce3baa7c06f2a117575e32ab2308f7f8216c29d9e2e3", size = 181958, upload-time = "2025-03-05T20:02:09.842Z" },
+    { url = "https://files.pythonhosted.org/packages/08/ff/e9eed2ee5fed6f76fdd6032ca5cd38c57ca9661430bb3d5fb2872dc8703c/websockets-15.0.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:54479983bd5fb469c38f2f5c7e3a24f9a4e70594cd68cd1fa6b9340dadaff7cf", size = 181918, upload-time = "2025-03-05T20:02:11.968Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/75/994634a49b7e12532be6a42103597b71098fd25900f7437d6055ed39930a/websockets-15.0.1-cp311-cp311-win32.whl", hash = "sha256:16b6c1b3e57799b9d38427dda63edcbe4926352c47cf88588c0be4ace18dac85", size = 176388, upload-time = "2025-03-05T20:02:13.32Z" },
+    { url = "https://files.pythonhosted.org/packages/98/93/e36c73f78400a65f5e236cd376713c34182e6663f6889cd45a4a04d8f203/websockets-15.0.1-cp311-cp311-win_amd64.whl", hash = "sha256:27ccee0071a0e75d22cb35849b1db43f2ecd3e161041ac1ee9d2352ddf72f065", size = 176828, upload-time = "2025-03-05T20:02:14.585Z" },
+    { url = "https://files.pythonhosted.org/packages/51/6b/4545a0d843594f5d0771e86463606a3988b5a09ca5123136f8a76580dd63/websockets-15.0.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:3e90baa811a5d73f3ca0bcbf32064d663ed81318ab225ee4f427ad4e26e5aff3", size = 175437, upload-time = "2025-03-05T20:02:16.706Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/71/809a0f5f6a06522af902e0f2ea2757f71ead94610010cf570ab5c98e99ed/websockets-15.0.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:592f1a9fe869c778694f0aa806ba0374e97648ab57936f092fd9d87f8bc03665", size = 173096, upload-time = "2025-03-05T20:02:18.832Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/69/1a681dd6f02180916f116894181eab8b2e25b31e484c5d0eae637ec01f7c/websockets-15.0.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:0701bc3cfcb9164d04a14b149fd74be7347a530ad3bbf15ab2c678a2cd3dd9a2", size = 173332, upload-time = "2025-03-05T20:02:20.187Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/02/0073b3952f5bce97eafbb35757f8d0d54812b6174ed8dd952aa08429bcc3/websockets-15.0.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e8b56bdcdb4505c8078cb6c7157d9811a85790f2f2b3632c7d1462ab5783d215", size = 183152, upload-time = "2025-03-05T20:02:22.286Z" },
+    { url = "https://files.pythonhosted.org/packages/74/45/c205c8480eafd114b428284840da0b1be9ffd0e4f87338dc95dc6ff961a1/websockets-15.0.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0af68c55afbd5f07986df82831c7bff04846928ea8d1fd7f30052638788bc9b5", size = 182096, upload-time = "2025-03-05T20:02:24.368Z" },
+    { url = "https://files.pythonhosted.org/packages/14/8f/aa61f528fba38578ec553c145857a181384c72b98156f858ca5c8e82d9d3/websockets-15.0.1-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:64dee438fed052b52e4f98f76c5790513235efaa1ef7f3f2192c392cd7c91b65", size = 182523, upload-time = "2025-03-05T20:02:25.669Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/6d/0267396610add5bc0d0d3e77f546d4cd287200804fe02323797de77dbce9/websockets-15.0.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d5f6b181bb38171a8ad1d6aa58a67a6aa9d4b38d0f8c5f496b9e42561dfc62fe", size = 182790, upload-time = "2025-03-05T20:02:26.99Z" },
+    { url = "https://files.pythonhosted.org/packages/02/05/c68c5adbf679cf610ae2f74a9b871ae84564462955d991178f95a1ddb7dd/websockets-15.0.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:5d54b09eba2bada6011aea5375542a157637b91029687eb4fdb2dab11059c1b4", size = 182165, upload-time = "2025-03-05T20:02:30.291Z" },
+    { url = "https://files.pythonhosted.org/packages/29/93/bb672df7b2f5faac89761cb5fa34f5cec45a4026c383a4b5761c6cea5c16/websockets-15.0.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:3be571a8b5afed347da347bfcf27ba12b069d9d7f42cb8c7028b5e98bbb12597", size = 182160, upload-time = "2025-03-05T20:02:31.634Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/83/de1f7709376dc3ca9b7eeb4b9a07b4526b14876b6d372a4dc62312bebee0/websockets-15.0.1-cp312-cp312-win32.whl", hash = "sha256:c338ffa0520bdb12fbc527265235639fb76e7bc7faafbb93f6ba80d9c06578a9", size = 176395, upload-time = "2025-03-05T20:02:33.017Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/71/abf2ebc3bbfa40f391ce1428c7168fb20582d0ff57019b69ea20fa698043/websockets-15.0.1-cp312-cp312-win_amd64.whl", hash = "sha256:fcd5cf9e305d7b8338754470cf69cf81f420459dbae8a3b40cee57417f4614a7", size = 176841, upload-time = "2025-03-05T20:02:34.498Z" },
+    { url = "https://files.pythonhosted.org/packages/36/db/3fff0bcbe339a6fa6a3b9e3fbc2bfb321ec2f4cd233692272c5a8d6cf801/websockets-15.0.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:5f4c04ead5aed67c8a1a20491d54cdfba5884507a48dd798ecaf13c74c4489f5", size = 175424, upload-time = "2025-03-05T20:02:56.505Z" },
+    { url = "https://files.pythonhosted.org/packages/46/e6/519054c2f477def4165b0ec060ad664ed174e140b0d1cbb9fafa4a54f6db/websockets-15.0.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:abdc0c6c8c648b4805c5eacd131910d2a7f6455dfd3becab248ef108e89ab16a", size = 173077, upload-time = "2025-03-05T20:02:58.37Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/21/c0712e382df64c93a0d16449ecbf87b647163485ca1cc3f6cbadb36d2b03/websockets-15.0.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a625e06551975f4b7ea7102bc43895b90742746797e2e14b70ed61c43a90f09b", size = 173324, upload-time = "2025-03-05T20:02:59.773Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/cb/51ba82e59b3a664df54beed8ad95517c1b4dc1a913730e7a7db778f21291/websockets-15.0.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d591f8de75824cbb7acad4e05d2d710484f15f29d4a915092675ad3456f11770", size = 182094, upload-time = "2025-03-05T20:03:01.827Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/0f/bf3788c03fec679bcdaef787518dbe60d12fe5615a544a6d4cf82f045193/websockets-15.0.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:47819cea040f31d670cc8d324bb6435c6f133b8c7a19ec3d61634e62f8d8f9eb", size = 181094, upload-time = "2025-03-05T20:03:03.123Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/da/9fb8c21edbc719b66763a571afbaf206cb6d3736d28255a46fc2fe20f902/websockets-15.0.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ac017dd64572e5c3bd01939121e4d16cf30e5d7e110a119399cf3133b63ad054", size = 181397, upload-time = "2025-03-05T20:03:04.443Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/65/65f379525a2719e91d9d90c38fe8b8bc62bd3c702ac651b7278609b696c4/websockets-15.0.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:4a9fac8e469d04ce6c25bb2610dc535235bd4aa14996b4e6dbebf5e007eba5ee", size = 181794, upload-time = "2025-03-05T20:03:06.708Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/26/31ac2d08f8e9304d81a1a7ed2851c0300f636019a57cbaa91342015c72cc/websockets-15.0.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:363c6f671b761efcb30608d24925a382497c12c506b51661883c3e22337265ed", size = 181194, upload-time = "2025-03-05T20:03:08.844Z" },
+    { url = "https://files.pythonhosted.org/packages/98/72/1090de20d6c91994cd4b357c3f75a4f25ee231b63e03adea89671cc12a3f/websockets-15.0.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:2034693ad3097d5355bfdacfffcbd3ef5694f9718ab7f29c29689a9eae841880", size = 181164, upload-time = "2025-03-05T20:03:10.242Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/37/098f2e1c103ae8ed79b0e77f08d83b0ec0b241cf4b7f2f10edd0126472e1/websockets-15.0.1-cp39-cp39-win32.whl", hash = "sha256:3b1ac0d3e594bf121308112697cf4b32be538fb1444468fb0a6ae4feebc83411", size = 176381, upload-time = "2025-03-05T20:03:12.77Z" },
+    { url = "https://files.pythonhosted.org/packages/75/8b/a32978a3ab42cebb2ebdd5b05df0696a09f4d436ce69def11893afa301f0/websockets-15.0.1-cp39-cp39-win_amd64.whl", hash = "sha256:b7643a03db5c95c799b89b31c036d5f27eeb4d259c798e878d6937d71832b1e4", size = 176841, upload-time = "2025-03-05T20:03:14.367Z" },
+    { url = "https://files.pythonhosted.org/packages/02/9e/d40f779fa16f74d3468357197af8d6ad07e7c5a27ea1ca74ceb38986f77a/websockets-15.0.1-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:0c9e74d766f2818bb95f84c25be4dea09841ac0f734d1966f415e4edfc4ef1c3", size = 173109, upload-time = "2025-03-05T20:03:17.769Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/cd/5b887b8585a593073fd92f7c23ecd3985cd2c3175025a91b0d69b0551372/websockets-15.0.1-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:1009ee0c7739c08a0cd59de430d6de452a55e42d6b522de7aa15e6f67db0b8e1", size = 173343, upload-time = "2025-03-05T20:03:19.094Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/ae/d34f7556890341e900a95acf4886833646306269f899d58ad62f588bf410/websockets-15.0.1-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:76d1f20b1c7a2fa82367e04982e708723ba0e7b8d43aa643d3dcd404d74f1475", size = 174599, upload-time = "2025-03-05T20:03:21.1Z" },
+    { url = "https://files.pythonhosted.org/packages/71/e6/5fd43993a87db364ec60fc1d608273a1a465c0caba69176dd160e197ce42/websockets-15.0.1-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f29d80eb9a9263b8d109135351caf568cc3f80b9928bccde535c235de55c22d9", size = 174207, upload-time = "2025-03-05T20:03:23.221Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/fb/c492d6daa5ec067c2988ac80c61359ace5c4c674c532985ac5a123436cec/websockets-15.0.1-pp310-pypy310_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b359ed09954d7c18bbc1680f380c7301f92c60bf924171629c5db97febb12f04", size = 174155, upload-time = "2025-03-05T20:03:25.321Z" },
+    { url = "https://files.pythonhosted.org/packages/68/a1/dcb68430b1d00b698ae7a7e0194433bce4f07ded185f0ee5fb21e2a2e91e/websockets-15.0.1-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:cad21560da69f4ce7658ca2cb83138fb4cf695a2ba3e475e0559e05991aa8122", size = 176884, upload-time = "2025-03-05T20:03:27.934Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/48/4b67623bac4d79beb3a6bb27b803ba75c1bdedc06bd827e465803690a4b2/websockets-15.0.1-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:7f493881579c90fc262d9cdbaa05a6b54b3811c2f300766748db79f098db9940", size = 173106, upload-time = "2025-03-05T20:03:29.404Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/f0/adb07514a49fe5728192764e04295be78859e4a537ab8fcc518a3dbb3281/websockets-15.0.1-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:47b099e1f4fbc95b701b6e85768e1fcdaf1630f3cbe4765fa216596f12310e2e", size = 173339, upload-time = "2025-03-05T20:03:30.755Z" },
+    { url = "https://files.pythonhosted.org/packages/87/28/bd23c6344b18fb43df40d0700f6d3fffcd7cef14a6995b4f976978b52e62/websockets-15.0.1-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:67f2b6de947f8c757db2db9c71527933ad0019737ec374a8a6be9a956786aaf9", size = 174597, upload-time = "2025-03-05T20:03:32.247Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/79/ca288495863d0f23a60f546f0905ae8f3ed467ad87f8b6aceb65f4c013e4/websockets-15.0.1-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d08eb4c2b7d6c41da6ca0600c077e93f5adcfd979cd777d747e9ee624556da4b", size = 174205, upload-time = "2025-03-05T20:03:33.731Z" },
+    { url = "https://files.pythonhosted.org/packages/04/e4/120ff3180b0872b1fe6637f6f995bcb009fb5c87d597c1fc21456f50c848/websockets-15.0.1-pp39-pypy39_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b826973a4a2ae47ba357e4e82fa44a463b8f168e1ca775ac64521442b19e87f", size = 174150, upload-time = "2025-03-05T20:03:35.757Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/c3/30e2f9c539b8da8b1d76f64012f3b19253271a63413b2d3adb94b143407f/websockets-15.0.1-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:21c1fa28a6a7e3cbdc171c694398b6df4744613ce9b36b1a498e816787e28123", size = 176877, upload-time = "2025-03-05T20:03:37.199Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/a8/5b41e0da817d64113292ab1f8247140aac61cbf6cfd085d6a0fa77f4984f/websockets-15.0.1-py3-none-any.whl", hash = "sha256:f7a866fbc1e97b5c617ee4116daaa09b722101d4a3c170c787450ba409f9736f", size = 169743, upload-time = "2025-03-05T20:03:39.41Z" },
+]
+
+[[package]]
+name = "werkzeug"
+version = "3.1.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "anyio" },
+    { name = "markupsafe" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f5/26/c705fc77d0a9ecdb9b66f1e2976d95b81df3cae518967431e7dbf9b5e219/watchfiles-1.0.4.tar.gz", hash = "sha256:6ba473efd11062d73e4f00c2b730255f9c1bdd73cd5f9fe5b5da8dbd4a717205", size = 94625 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/14/02/22fcaed0396730b0d362bc8d1ffb3be2658fd473eecbb2ba84243e157f11/watchfiles-1.0.4-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:ba5bb3073d9db37c64520681dd2650f8bd40902d991e7b4cfaeece3e32561d08", size = 395212 },
-    { url = "https://files.pythonhosted.org/packages/e9/3d/ec5a2369a46edf3ebe092c39d9ae48e8cb6dacbde51c4b4f98936c524269/watchfiles-1.0.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9f25d0ba0fe2b6d2c921cf587b2bf4c451860086534f40c384329fb96e2044d1", size = 384815 },
-    { url = "https://files.pythonhosted.org/packages/df/b4/898991cececbe171e67142c31905510203649569d9817848f47c4177ee42/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:47eb32ef8c729dbc4f4273baece89398a4d4b5d21a1493efea77a17059f4df8a", size = 450680 },
-    { url = "https://files.pythonhosted.org/packages/58/f7/d4aa3000e812cfb5e5c2c6c0a3ec9d0a46a42489a8727edd160631c4e210/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:076f293100db3b0b634514aa0d294b941daa85fc777f9c698adb1009e5aca0b1", size = 455923 },
-    { url = "https://files.pythonhosted.org/packages/dd/95/7e2e4c6aba1b02fb5c76d2f6a450b85215921ec5f8f7ad5efd075369563f/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1eacd91daeb5158c598fe22d7ce66d60878b6294a86477a4715154990394c9b3", size = 482339 },
-    { url = "https://files.pythonhosted.org/packages/bb/67/4265b0fabcc2ef2c9e3e8802ba7908cf718a357ebfb49c72e53787156a48/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:13c2ce7b72026cfbca120d652f02c7750f33b4c9395d79c9790b27f014c8a5a2", size = 519908 },
-    { url = "https://files.pythonhosted.org/packages/0d/96/b57802d5f8164bdf070befb4fd3dec4edba5a364ec0670965a97eb8098ce/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:90192cdc15ab7254caa7765a98132a5a41471cf739513cc9bcf7d2ffcc0ec7b2", size = 501410 },
-    { url = "https://files.pythonhosted.org/packages/8b/18/6db0de4e8911ba14e31853201b40c0fa9fea5ecf3feb86b0ad58f006dfc3/watchfiles-1.0.4-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:278aaa395f405972e9f523bd786ed59dfb61e4b827856be46a42130605fd0899", size = 452876 },
-    { url = "https://files.pythonhosted.org/packages/df/df/092a961815edf723a38ba2638c49491365943919c3526cc9cf82c42786a6/watchfiles-1.0.4-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:a462490e75e466edbb9fc4cd679b62187153b3ba804868452ef0577ec958f5ff", size = 615353 },
-    { url = "https://files.pythonhosted.org/packages/f3/cf/b85fe645de4ff82f3f436c5e9032379fce37c303f6396a18f9726cc34519/watchfiles-1.0.4-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:8d0d0630930f5cd5af929040e0778cf676a46775753e442a3f60511f2409f48f", size = 613187 },
-    { url = "https://files.pythonhosted.org/packages/f6/d4/a9fea27aef4dd69689bc3556718c1157a7accb72aa035ece87c1fa8483b5/watchfiles-1.0.4-cp310-cp310-win32.whl", hash = "sha256:cc27a65069bcabac4552f34fd2dce923ce3fcde0721a16e4fb1b466d63ec831f", size = 270799 },
-    { url = "https://files.pythonhosted.org/packages/df/02/dbe9d4439f15dd4ad0720b6e039bde9d66d1f830331f34c18eb70fa6608e/watchfiles-1.0.4-cp310-cp310-win_amd64.whl", hash = "sha256:8b1f135238e75d075359cf506b27bf3f4ca12029c47d3e769d8593a2024ce161", size = 284145 },
-    { url = "https://files.pythonhosted.org/packages/0f/bb/8461adc4b1fed009546fb797fc0d5698dcfe5e289cb37e1b8f16a93cdc30/watchfiles-1.0.4-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:2a9f93f8439639dc244c4d2902abe35b0279102bca7bbcf119af964f51d53c19", size = 394869 },
-    { url = "https://files.pythonhosted.org/packages/55/88/9ebf36b3547176d1709c320de78c1fa3263a46be31b5b1267571d9102686/watchfiles-1.0.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:9eea33ad8c418847dd296e61eb683cae1c63329b6d854aefcd412e12d94ee235", size = 384905 },
-    { url = "https://files.pythonhosted.org/packages/03/8a/04335ce23ef78d8c69f0913e8b20cf7d9233e3986543aeef95ef2d6e43d2/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:31f1a379c9dcbb3f09cf6be1b7e83b67c0e9faabed0471556d9438a4a4e14202", size = 449944 },
-    { url = "https://files.pythonhosted.org/packages/17/4e/c8d5dcd14fe637f4633616dabea8a4af0a10142dccf3b43e0f081ba81ab4/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ab594e75644421ae0a2484554832ca5895f8cab5ab62de30a1a57db460ce06c6", size = 456020 },
-    { url = "https://files.pythonhosted.org/packages/5e/74/3e91e09e1861dd7fbb1190ce7bd786700dc0fbc2ccd33bb9fff5de039229/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fc2eb5d14a8e0d5df7b36288979176fbb39672d45184fc4b1c004d7c3ce29317", size = 482983 },
-    { url = "https://files.pythonhosted.org/packages/a1/3d/e64de2d1ce4eb6a574fd78ce3a28c279da263be9ef3cfcab6f708df192f2/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3f68d8e9d5a321163ddacebe97091000955a1b74cd43724e346056030b0bacee", size = 520320 },
-    { url = "https://files.pythonhosted.org/packages/2c/bd/52235f7063b57240c66a991696ed27e2a18bd6fcec8a1ea5a040b70d0611/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f9ce064e81fe79faa925ff03b9f4c1a98b0bbb4a1b8c1b015afa93030cb21a49", size = 500988 },
-    { url = "https://files.pythonhosted.org/packages/3a/b0/ff04194141a5fe650c150400dd9e42667916bc0f52426e2e174d779b8a74/watchfiles-1.0.4-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b77d5622ac5cc91d21ae9c2b284b5d5c51085a0bdb7b518dba263d0af006132c", size = 452573 },
-    { url = "https://files.pythonhosted.org/packages/3d/9d/966164332c5a178444ae6d165082d4f351bd56afd9c3ec828eecbf190e6a/watchfiles-1.0.4-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:1941b4e39de9b38b868a69b911df5e89dc43767feeda667b40ae032522b9b5f1", size = 615114 },
-    { url = "https://files.pythonhosted.org/packages/94/df/f569ae4c1877f96ad4086c153a8eee5a19a3b519487bf5c9454a3438c341/watchfiles-1.0.4-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:4f8c4998506241dedf59613082d1c18b836e26ef2a4caecad0ec41e2a15e4226", size = 613076 },
-    { url = "https://files.pythonhosted.org/packages/15/ae/8ce5f29e65d5fa5790e3c80c289819c55e12be2e1b9f5b6a0e55e169b97d/watchfiles-1.0.4-cp311-cp311-win32.whl", hash = "sha256:4ebbeca9360c830766b9f0df3640b791be569d988f4be6c06d6fae41f187f105", size = 271013 },
-    { url = "https://files.pythonhosted.org/packages/a4/c6/79dc4a7c598a978e5fafa135090aaf7bbb03b8dec7bada437dfbe578e7ed/watchfiles-1.0.4-cp311-cp311-win_amd64.whl", hash = "sha256:05d341c71f3d7098920f8551d4df47f7b57ac5b8dad56558064c3431bdfc0b74", size = 284229 },
-    { url = "https://files.pythonhosted.org/packages/37/3d/928633723211753f3500bfb138434f080363b87a1b08ca188b1ce54d1e05/watchfiles-1.0.4-cp311-cp311-win_arm64.whl", hash = "sha256:32b026a6ab64245b584acf4931fe21842374da82372d5c039cba6bf99ef722f3", size = 276824 },
-    { url = "https://files.pythonhosted.org/packages/5b/1a/8f4d9a1461709756ace48c98f07772bc6d4519b1e48b5fa24a4061216256/watchfiles-1.0.4-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:229e6ec880eca20e0ba2f7e2249c85bae1999d330161f45c78d160832e026ee2", size = 391345 },
-    { url = "https://files.pythonhosted.org/packages/bc/d2/6750b7b3527b1cdaa33731438432e7238a6c6c40a9924049e4cebfa40805/watchfiles-1.0.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:5717021b199e8353782dce03bd8a8f64438832b84e2885c4a645f9723bf656d9", size = 381515 },
-    { url = "https://files.pythonhosted.org/packages/4e/17/80500e42363deef1e4b4818729ed939aaddc56f82f4e72b2508729dd3c6b/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0799ae68dfa95136dde7c472525700bd48777875a4abb2ee454e3ab18e9fc712", size = 449767 },
-    { url = "https://files.pythonhosted.org/packages/10/37/1427fa4cfa09adbe04b1e97bced19a29a3462cc64c78630787b613a23f18/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:43b168bba889886b62edb0397cab5b6490ffb656ee2fcb22dec8bfeb371a9e12", size = 455677 },
-    { url = "https://files.pythonhosted.org/packages/c5/7a/39e9397f3a19cb549a7d380412fd9e507d4854eddc0700bfad10ef6d4dba/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fb2c46e275fbb9f0c92e7654b231543c7bbfa1df07cdc4b99fa73bedfde5c844", size = 482219 },
-    { url = "https://files.pythonhosted.org/packages/45/2d/7113931a77e2ea4436cad0c1690c09a40a7f31d366f79c6f0a5bc7a4f6d5/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:857f5fc3aa027ff5e57047da93f96e908a35fe602d24f5e5d8ce64bf1f2fc733", size = 518830 },
-    { url = "https://files.pythonhosted.org/packages/f9/1b/50733b1980fa81ef3c70388a546481ae5fa4c2080040100cd7bf3bf7b321/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:55ccfd27c497b228581e2838d4386301227fc0cb47f5a12923ec2fe4f97b95af", size = 497997 },
-    { url = "https://files.pythonhosted.org/packages/2b/b4/9396cc61b948ef18943e7c85ecfa64cf940c88977d882da57147f62b34b1/watchfiles-1.0.4-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5c11ea22304d17d4385067588123658e9f23159225a27b983f343fcffc3e796a", size = 452249 },
-    { url = "https://files.pythonhosted.org/packages/fb/69/0c65a5a29e057ad0dc691c2fa6c23b2983c7dabaa190ba553b29ac84c3cc/watchfiles-1.0.4-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:74cb3ca19a740be4caa18f238298b9d472c850f7b2ed89f396c00a4c97e2d9ff", size = 614412 },
-    { url = "https://files.pythonhosted.org/packages/7f/b9/319fcba6eba5fad34327d7ce16a6b163b39741016b1996f4a3c96b8dd0e1/watchfiles-1.0.4-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:c7cce76c138a91e720d1df54014a047e680b652336e1b73b8e3ff3158e05061e", size = 611982 },
-    { url = "https://files.pythonhosted.org/packages/f1/47/143c92418e30cb9348a4387bfa149c8e0e404a7c5b0585d46d2f7031b4b9/watchfiles-1.0.4-cp312-cp312-win32.whl", hash = "sha256:b045c800d55bc7e2cadd47f45a97c7b29f70f08a7c2fa13241905010a5493f94", size = 271822 },
-    { url = "https://files.pythonhosted.org/packages/ea/94/b0165481bff99a64b29e46e07ac2e0df9f7a957ef13bec4ceab8515f44e3/watchfiles-1.0.4-cp312-cp312-win_amd64.whl", hash = "sha256:c2acfa49dd0ad0bf2a9c0bb9a985af02e89345a7189be1efc6baa085e0f72d7c", size = 285441 },
-    { url = "https://files.pythonhosted.org/packages/11/de/09fe56317d582742d7ca8c2ca7b52a85927ebb50678d9b0fa8194658f536/watchfiles-1.0.4-cp312-cp312-win_arm64.whl", hash = "sha256:22bb55a7c9e564e763ea06c7acea24fc5d2ee5dfc5dafc5cfbedfe58505e9f90", size = 277141 },
-    { url = "https://files.pythonhosted.org/packages/08/98/f03efabec64b5b1fa58c0daab25c68ef815b0f320e54adcacd0d6847c339/watchfiles-1.0.4-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:8012bd820c380c3d3db8435e8cf7592260257b378b649154a7948a663b5f84e9", size = 390954 },
-    { url = "https://files.pythonhosted.org/packages/16/09/4dd49ba0a32a45813debe5fb3897955541351ee8142f586303b271a02b40/watchfiles-1.0.4-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:aa216f87594f951c17511efe5912808dfcc4befa464ab17c98d387830ce07b60", size = 381133 },
-    { url = "https://files.pythonhosted.org/packages/76/59/5aa6fc93553cd8d8ee75c6247763d77c02631aed21551a97d94998bf1dae/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:62c9953cf85529c05b24705639ffa390f78c26449e15ec34d5339e8108c7c407", size = 449516 },
-    { url = "https://files.pythonhosted.org/packages/4c/aa/df4b6fe14b6317290b91335b23c96b488d365d65549587434817e06895ea/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7cf684aa9bba4cd95ecb62c822a56de54e3ae0598c1a7f2065d51e24637a3c5d", size = 454820 },
-    { url = "https://files.pythonhosted.org/packages/5e/71/185f8672f1094ce48af33252c73e39b48be93b761273872d9312087245f6/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f44a39aee3cbb9b825285ff979ab887a25c5d336e5ec3574f1506a4671556a8d", size = 481550 },
-    { url = "https://files.pythonhosted.org/packages/85/d7/50ebba2c426ef1a5cb17f02158222911a2e005d401caf5d911bfca58f4c4/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a38320582736922be8c865d46520c043bff350956dfc9fbaee3b2df4e1740a4b", size = 518647 },
-    { url = "https://files.pythonhosted.org/packages/f0/7a/4c009342e393c545d68987e8010b937f72f47937731225b2b29b7231428f/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:39f4914548b818540ef21fd22447a63e7be6e24b43a70f7642d21f1e73371590", size = 497547 },
-    { url = "https://files.pythonhosted.org/packages/0f/7c/1cf50b35412d5c72d63b2bf9a4fffee2e1549a245924960dd087eb6a6de4/watchfiles-1.0.4-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f12969a3765909cf5dc1e50b2436eb2c0e676a3c75773ab8cc3aa6175c16e902", size = 452179 },
-    { url = "https://files.pythonhosted.org/packages/d6/a9/3db1410e1c1413735a9a472380e4f431ad9a9e81711cda2aaf02b7f62693/watchfiles-1.0.4-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:0986902677a1a5e6212d0c49b319aad9cc48da4bd967f86a11bde96ad9676ca1", size = 614125 },
-    { url = "https://files.pythonhosted.org/packages/f2/e1/0025d365cf6248c4d1ee4c3d2e3d373bdd3f6aff78ba4298f97b4fad2740/watchfiles-1.0.4-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:308ac265c56f936636e3b0e3f59e059a40003c655228c131e1ad439957592303", size = 611911 },
-    { url = "https://files.pythonhosted.org/packages/55/55/035838277d8c98fc8c917ac9beeb0cd6c59d675dc2421df5f9fcf44a0070/watchfiles-1.0.4-cp313-cp313-win32.whl", hash = "sha256:aee397456a29b492c20fda2d8961e1ffb266223625346ace14e4b6d861ba9c80", size = 271152 },
-    { url = "https://files.pythonhosted.org/packages/f0/e5/96b8e55271685ddbadc50ce8bc53aa2dff278fb7ac4c2e473df890def2dc/watchfiles-1.0.4-cp313-cp313-win_amd64.whl", hash = "sha256:d6097538b0ae5c1b88c3b55afa245a66793a8fec7ada6755322e465fb1a0e8cc", size = 285216 },
-    { url = "https://files.pythonhosted.org/packages/15/81/54484fc2fa715abe79694b975692af963f0878fb9d72b8251aa542bf3f10/watchfiles-1.0.4-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:d3452c1ec703aa1c61e15dfe9d482543e4145e7c45a6b8566978fbb044265a21", size = 394967 },
-    { url = "https://files.pythonhosted.org/packages/14/b3/557f0cd90add86586fe3deeebd11e8299db6bc3452b44a534f844c6ab831/watchfiles-1.0.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:7b75fee5a16826cf5c46fe1c63116e4a156924d668c38b013e6276f2582230f0", size = 384707 },
-    { url = "https://files.pythonhosted.org/packages/03/a3/34638e1bffcb85a405e7b005e30bb211fd9be2ab2cb1847f2ceb81bef27b/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e997802d78cdb02623b5941830ab06f8860038faf344f0d288d325cc9c5d2ff", size = 450442 },
-    { url = "https://files.pythonhosted.org/packages/8f/9f/6a97460dd11a606003d634c7158d9fea8517e98daffc6f56d0f5fde2e86a/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e0611d244ce94d83f5b9aff441ad196c6e21b55f77f3c47608dcf651efe54c4a", size = 455959 },
-    { url = "https://files.pythonhosted.org/packages/9d/bb/e0648c6364e4d37ec692bc3f0c77507d17d8bb8f75689148819142010bbf/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9745a4210b59e218ce64c91deb599ae8775c8a9da4e95fb2ee6fe745fc87d01a", size = 483187 },
-    { url = "https://files.pythonhosted.org/packages/dd/ad/d9290586a25288a81dfa8ad6329cf1de32aa1a9798ace45259eb95dcfb37/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4810ea2ae622add560f4aa50c92fef975e475f7ac4900ce5ff5547b2434642d8", size = 519733 },
-    { url = "https://files.pythonhosted.org/packages/4e/a9/150c1666825cc9637093f8cae7fc6f53b3296311ab8bd65f1389acb717cb/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:740d103cd01458f22462dedeb5a3382b7f2c57d07ff033fbc9465919e5e1d0f3", size = 502275 },
-    { url = "https://files.pythonhosted.org/packages/44/dc/5bfd21e20a330aca1706ac44713bc322838061938edf4b53130f97a7b211/watchfiles-1.0.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cdbd912a61543a36aef85e34f212e5d2486e7c53ebfdb70d1e0b060cc50dd0bf", size = 452907 },
-    { url = "https://files.pythonhosted.org/packages/50/fe/8f4fc488f1699f564687b697456eb5c0cb8e2b0b8538150511c234c62094/watchfiles-1.0.4-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:0bc80d91ddaf95f70258cf78c471246846c1986bcc5fd33ccc4a1a67fcb40f9a", size = 615927 },
-    { url = "https://files.pythonhosted.org/packages/ad/19/2e45f6f6eec89dd97a4d281635e3d73c17e5f692e7432063bdfdf9562c89/watchfiles-1.0.4-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:ab0311bb2ffcd9f74b6c9de2dda1612c13c84b996d032cd74799adb656af4e8b", size = 613435 },
-    { url = "https://files.pythonhosted.org/packages/91/17/dc5ac62ca377827c24321d68050efc2eaee2ebaf3f21d055bbce2206d309/watchfiles-1.0.4-cp39-cp39-win32.whl", hash = "sha256:02a526ee5b5a09e8168314c905fc545c9bc46509896ed282aeb5a8ba9bd6ca27", size = 270810 },
-    { url = "https://files.pythonhosted.org/packages/82/2b/dad851342492d538e7ffe72a8c756f747dd147988abb039ac9d6577d2235/watchfiles-1.0.4-cp39-cp39-win_amd64.whl", hash = "sha256:a5ae5706058b27c74bac987d615105da17724172d5aaacc6c362a40599b6de43", size = 284866 },
-    { url = "https://files.pythonhosted.org/packages/6f/06/175d5ac6b838fb319008c0cd981d7bf289317c510154d411d3584ca2b67b/watchfiles-1.0.4-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:cdcc92daeae268de1acf5b7befcd6cfffd9a047098199056c72e4623f531de18", size = 396269 },
-    { url = "https://files.pythonhosted.org/packages/86/ee/5db93b0b57dc0587abdbac4149296ee73275f615d790a82cb5598af0557f/watchfiles-1.0.4-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:d8d3d9203705b5797f0af7e7e5baa17c8588030aaadb7f6a86107b7247303817", size = 386010 },
-    { url = "https://files.pythonhosted.org/packages/75/61/fe0dc5fedf152bfc085a53711f740701f6bdb8ab6b5c950402b681d4858b/watchfiles-1.0.4-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bdef5a1be32d0b07dcea3318a0be95d42c98ece24177820226b56276e06b63b0", size = 450913 },
-    { url = "https://files.pythonhosted.org/packages/9f/dd/3c7731af3baf1a9957afc643d176f94480921a690ec3237c9f9d11301c08/watchfiles-1.0.4-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:342622287b5604ddf0ed2d085f3a589099c9ae8b7331df3ae9845571586c4f3d", size = 453474 },
-    { url = "https://files.pythonhosted.org/packages/6b/b4/c3998f54c91a35cee60ee6d3a855a069c5dff2bae6865147a46e9090dccd/watchfiles-1.0.4-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:9fe37a2de80aa785d340f2980276b17ef697ab8db6019b07ee4fd28a8359d2f3", size = 395565 },
-    { url = "https://files.pythonhosted.org/packages/3f/05/ac1a4d235beb9ddfb8ac26ce93a00ba6bd1b1b43051ef12d7da957b4a9d1/watchfiles-1.0.4-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:9d1ef56b56ed7e8f312c934436dea93bfa3e7368adfcf3df4c0da6d4de959a1e", size = 385406 },
-    { url = "https://files.pythonhosted.org/packages/4c/ea/36532e7d86525f4e52a10efed182abf33efb106a93d49f5fbc994b256bcd/watchfiles-1.0.4-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:95b42cac65beae3a362629950c444077d1b44f1790ea2772beaea95451c086bb", size = 450424 },
-    { url = "https://files.pythonhosted.org/packages/7a/e9/3cbcf4d70cd0b6d3f30631deae1bf37cc0be39887ca327a44462fe546bf5/watchfiles-1.0.4-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5e0227b8ed9074c6172cf55d85b5670199c99ab11fd27d2c473aa30aec67ee42", size = 452488 },
+sdist = { url = "https://files.pythonhosted.org/packages/9f/69/83029f1f6300c5fb2471d621ab06f6ec6b3324685a2ce0f9777fd4a8b71e/werkzeug-3.1.3.tar.gz", hash = "sha256:60723ce945c19328679790e3282cc758aa4a6040e4bb330f53d30fa546d44746", size = 806925, upload-time = "2024-11-08T15:52:18.093Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/52/24/ab44c871b0f07f491e5d2ad12c9bd7358e527510618cb1b803a88e986db1/werkzeug-3.1.3-py3-none-any.whl", hash = "sha256:54b78bf3716d19a65be4fceccc0d1d7b89e608834989dfae50ea87564639213e", size = 224498, upload-time = "2024-11-08T15:52:16.132Z" },
+]
+
+[[package]]
+name = "wheel"
+version = "0.45.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8a/98/2d9906746cdc6a6ef809ae6338005b3f21bb568bea3165cfc6a243fdc25c/wheel-0.45.1.tar.gz", hash = "sha256:661e1abd9198507b1409a20c02106d9670b2576e916d58f520316666abca6729", size = 107545, upload-time = "2024-11-23T00:18:23.513Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0b/2c/87f3254fd8ffd29e4c02732eee68a83a1d3c346ae39bc6822dcbcb697f2b/wheel-0.45.1-py3-none-any.whl", hash = "sha256:708e7481cc80179af0e556bbf0cc00b8444c7321e2700b8d8580231d13017248", size = 72494, upload-time = "2024-11-23T00:18:21.207Z" },
+]
+
+[[package]]
+name = "wrapt"
+version = "1.17.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/95/8f/aeb76c5b46e273670962298c23e7ddde79916cb74db802131d49a85e4b7d/wrapt-1.17.3.tar.gz", hash = "sha256:f66eb08feaa410fe4eebd17f2a2c8e2e46d3476e9f8c783daa8e09e0faa666d0", size = 55547, upload-time = "2025-08-12T05:53:21.714Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3f/23/bb82321b86411eb51e5a5db3fb8f8032fd30bd7c2d74bfe936136b2fa1d6/wrapt-1.17.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:88bbae4d40d5a46142e70d58bf664a89b6b4befaea7b2ecc14e03cedb8e06c04", size = 53482, upload-time = "2025-08-12T05:51:44.467Z" },
+    { url = "https://files.pythonhosted.org/packages/45/69/f3c47642b79485a30a59c63f6d739ed779fb4cc8323205d047d741d55220/wrapt-1.17.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:e6b13af258d6a9ad602d57d889f83b9d5543acd471eee12eb51f5b01f8eb1bc2", size = 38676, upload-time = "2025-08-12T05:51:32.636Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/71/e7e7f5670c1eafd9e990438e69d8fb46fa91a50785332e06b560c869454f/wrapt-1.17.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:fd341868a4b6714a5962c1af0bd44f7c404ef78720c7de4892901e540417111c", size = 38957, upload-time = "2025-08-12T05:51:54.655Z" },
+    { url = "https://files.pythonhosted.org/packages/de/17/9f8f86755c191d6779d7ddead1a53c7a8aa18bccb7cea8e7e72dfa6a8a09/wrapt-1.17.3-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f9b2601381be482f70e5d1051a5965c25fb3625455a2bf520b5a077b22afb775", size = 81975, upload-time = "2025-08-12T05:52:30.109Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/15/dd576273491f9f43dd09fce517f6c2ce6eb4fe21681726068db0d0467096/wrapt-1.17.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:343e44b2a8e60e06a7e0d29c1671a0d9951f59174f3709962b5143f60a2a98bd", size = 83149, upload-time = "2025-08-12T05:52:09.316Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/c4/5eb4ce0d4814521fee7aa806264bf7a114e748ad05110441cd5b8a5c744b/wrapt-1.17.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:33486899acd2d7d3066156b03465b949da3fd41a5da6e394ec49d271baefcf05", size = 82209, upload-time = "2025-08-12T05:52:10.331Z" },
+    { url = "https://files.pythonhosted.org/packages/31/4b/819e9e0eb5c8dc86f60dfc42aa4e2c0d6c3db8732bce93cc752e604bb5f5/wrapt-1.17.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:e6f40a8aa5a92f150bdb3e1c44b7e98fb7113955b2e5394122fa5532fec4b418", size = 81551, upload-time = "2025-08-12T05:52:31.137Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/83/ed6baf89ba3a56694700139698cf703aac9f0f9eb03dab92f57551bd5385/wrapt-1.17.3-cp310-cp310-win32.whl", hash = "sha256:a36692b8491d30a8c75f1dfee65bef119d6f39ea84ee04d9f9311f83c5ad9390", size = 36464, upload-time = "2025-08-12T05:53:01.204Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/90/ee61d36862340ad7e9d15a02529df6b948676b9a5829fd5e16640156627d/wrapt-1.17.3-cp310-cp310-win_amd64.whl", hash = "sha256:afd964fd43b10c12213574db492cb8f73b2f0826c8df07a68288f8f19af2ebe6", size = 38748, upload-time = "2025-08-12T05:53:00.209Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/c3/cefe0bd330d389c9983ced15d326f45373f4073c9f4a8c2f99b50bfea329/wrapt-1.17.3-cp310-cp310-win_arm64.whl", hash = "sha256:af338aa93554be859173c39c85243970dc6a289fa907402289eeae7543e1ae18", size = 36810, upload-time = "2025-08-12T05:52:51.906Z" },
+    { url = "https://files.pythonhosted.org/packages/52/db/00e2a219213856074a213503fdac0511203dceefff26e1daa15250cc01a0/wrapt-1.17.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:273a736c4645e63ac582c60a56b0acb529ef07f78e08dc6bfadf6a46b19c0da7", size = 53482, upload-time = "2025-08-12T05:51:45.79Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/30/ca3c4a5eba478408572096fe9ce36e6e915994dd26a4e9e98b4f729c06d9/wrapt-1.17.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:5531d911795e3f935a9c23eb1c8c03c211661a5060aab167065896bbf62a5f85", size = 38674, upload-time = "2025-08-12T05:51:34.629Z" },
+    { url = "https://files.pythonhosted.org/packages/31/25/3e8cc2c46b5329c5957cec959cb76a10718e1a513309c31399a4dad07eb3/wrapt-1.17.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:0610b46293c59a3adbae3dee552b648b984176f8562ee0dba099a56cfbe4df1f", size = 38959, upload-time = "2025-08-12T05:51:56.074Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/8f/a32a99fc03e4b37e31b57cb9cefc65050ea08147a8ce12f288616b05ef54/wrapt-1.17.3-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:b32888aad8b6e68f83a8fdccbf3165f5469702a7544472bdf41f582970ed3311", size = 82376, upload-time = "2025-08-12T05:52:32.134Z" },
+    { url = "https://files.pythonhosted.org/packages/31/57/4930cb8d9d70d59c27ee1332a318c20291749b4fba31f113c2f8ac49a72e/wrapt-1.17.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8cccf4f81371f257440c88faed6b74f1053eef90807b77e31ca057b2db74edb1", size = 83604, upload-time = "2025-08-12T05:52:11.663Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/f3/1afd48de81d63dd66e01b263a6fbb86e1b5053b419b9b33d13e1f6d0f7d0/wrapt-1.17.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d8a210b158a34164de8bb68b0e7780041a903d7b00c87e906fb69928bf7890d5", size = 82782, upload-time = "2025-08-12T05:52:12.626Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/d7/4ad5327612173b144998232f98a85bb24b60c352afb73bc48e3e0d2bdc4e/wrapt-1.17.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:79573c24a46ce11aab457b472efd8d125e5a51da2d1d24387666cd85f54c05b2", size = 82076, upload-time = "2025-08-12T05:52:33.168Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/59/e0adfc831674a65694f18ea6dc821f9fcb9ec82c2ce7e3d73a88ba2e8718/wrapt-1.17.3-cp311-cp311-win32.whl", hash = "sha256:c31eebe420a9a5d2887b13000b043ff6ca27c452a9a22fa71f35f118e8d4bf89", size = 36457, upload-time = "2025-08-12T05:53:03.936Z" },
+    { url = "https://files.pythonhosted.org/packages/83/88/16b7231ba49861b6f75fc309b11012ede4d6b0a9c90969d9e0db8d991aeb/wrapt-1.17.3-cp311-cp311-win_amd64.whl", hash = "sha256:0b1831115c97f0663cb77aa27d381237e73ad4f721391a9bfb2fe8bc25fa6e77", size = 38745, upload-time = "2025-08-12T05:53:02.885Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/1e/c4d4f3398ec073012c51d1c8d87f715f56765444e1a4b11e5180577b7e6e/wrapt-1.17.3-cp311-cp311-win_arm64.whl", hash = "sha256:5a7b3c1ee8265eb4c8f1b7d29943f195c00673f5ab60c192eba2d4a7eae5f46a", size = 36806, upload-time = "2025-08-12T05:52:53.368Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/41/cad1aba93e752f1f9268c77270da3c469883d56e2798e7df6240dcb2287b/wrapt-1.17.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:ab232e7fdb44cdfbf55fc3afa31bcdb0d8980b9b95c38b6405df2acb672af0e0", size = 53998, upload-time = "2025-08-12T05:51:47.138Z" },
+    { url = "https://files.pythonhosted.org/packages/60/f8/096a7cc13097a1869fe44efe68dace40d2a16ecb853141394047f0780b96/wrapt-1.17.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:9baa544e6acc91130e926e8c802a17f3b16fbea0fd441b5a60f5cf2cc5c3deba", size = 39020, upload-time = "2025-08-12T05:51:35.906Z" },
+    { url = "https://files.pythonhosted.org/packages/33/df/bdf864b8997aab4febb96a9ae5c124f700a5abd9b5e13d2a3214ec4be705/wrapt-1.17.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6b538e31eca1a7ea4605e44f81a48aa24c4632a277431a6ed3f328835901f4fd", size = 39098, upload-time = "2025-08-12T05:51:57.474Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/81/5d931d78d0eb732b95dc3ddaeeb71c8bb572fb01356e9133916cd729ecdd/wrapt-1.17.3-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:042ec3bb8f319c147b1301f2393bc19dba6e176b7da446853406d041c36c7828", size = 88036, upload-time = "2025-08-12T05:52:34.784Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/38/2e1785df03b3d72d34fc6252d91d9d12dc27a5c89caef3335a1bbb8908ca/wrapt-1.17.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3af60380ba0b7b5aeb329bc4e402acd25bd877e98b3727b0135cb5c2efdaefe9", size = 88156, upload-time = "2025-08-12T05:52:13.599Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/8b/48cdb60fe0603e34e05cffda0b2a4adab81fd43718e11111a4b0100fd7c1/wrapt-1.17.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:0b02e424deef65c9f7326d8c19220a2c9040c51dc165cddb732f16198c168396", size = 87102, upload-time = "2025-08-12T05:52:14.56Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/51/d81abca783b58f40a154f1b2c56db1d2d9e0d04fa2d4224e357529f57a57/wrapt-1.17.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:74afa28374a3c3a11b3b5e5fca0ae03bef8450d6aa3ab3a1e2c30e3a75d023dc", size = 87732, upload-time = "2025-08-12T05:52:36.165Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/b1/43b286ca1392a006d5336412d41663eeef1ad57485f3e52c767376ba7e5a/wrapt-1.17.3-cp312-cp312-win32.whl", hash = "sha256:4da9f45279fff3543c371d5ababc57a0384f70be244de7759c85a7f989cb4ebe", size = 36705, upload-time = "2025-08-12T05:53:07.123Z" },
+    { url = "https://files.pythonhosted.org/packages/28/de/49493f962bd3c586ab4b88066e967aa2e0703d6ef2c43aa28cb83bf7b507/wrapt-1.17.3-cp312-cp312-win_amd64.whl", hash = "sha256:e71d5c6ebac14875668a1e90baf2ea0ef5b7ac7918355850c0908ae82bcb297c", size = 38877, upload-time = "2025-08-12T05:53:05.436Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/48/0f7102fe9cb1e8a5a77f80d4f0956d62d97034bbe88d33e94699f99d181d/wrapt-1.17.3-cp312-cp312-win_arm64.whl", hash = "sha256:604d076c55e2fdd4c1c03d06dc1a31b95130010517b5019db15365ec4a405fc6", size = 36885, upload-time = "2025-08-12T05:52:54.367Z" },
+    { url = "https://files.pythonhosted.org/packages/41/be/be9b3b0a461ee3e30278706f3f3759b9b69afeedef7fe686036286c04ac6/wrapt-1.17.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:30ce38e66630599e1193798285706903110d4f057aab3168a34b7fdc85569afc", size = 53485, upload-time = "2025-08-12T05:51:53.11Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/a8/8f61d6b8f526efc8c10e12bf80b4206099fea78ade70427846a37bc9cbea/wrapt-1.17.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:65d1d00fbfb3ea5f20add88bbc0f815150dbbde3b026e6c24759466c8b5a9ef9", size = 38675, upload-time = "2025-08-12T05:51:42.885Z" },
+    { url = "https://files.pythonhosted.org/packages/48/f1/23950c29a25637b74b322f9e425a17cc01a478f6afb35138ecb697f9558d/wrapt-1.17.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a7c06742645f914f26c7f1fa47b8bc4c91d222f76ee20116c43d5ef0912bba2d", size = 38956, upload-time = "2025-08-12T05:52:03.149Z" },
+    { url = "https://files.pythonhosted.org/packages/43/46/dd0791943613885f62619f18ee6107e6133237a6b6ed8a9ecfac339d0b4f/wrapt-1.17.3-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7e18f01b0c3e4a07fe6dfdb00e29049ba17eadbc5e7609a2a3a4af83ab7d710a", size = 81745, upload-time = "2025-08-12T05:52:49.62Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/ec/bb2d19bd1a614cc4f438abac13ae26c57186197920432d2a915183b15a8b/wrapt-1.17.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0f5f51a6466667a5a356e6381d362d259125b57f059103dd9fdc8c0cf1d14139", size = 82833, upload-time = "2025-08-12T05:52:27.738Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/eb/66579aea6ad36f07617fedca8e282e49c7c9bab64c63b446cfe4f7f47a49/wrapt-1.17.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:59923aa12d0157f6b82d686c3fd8e1166fa8cdfb3e17b42ce3b6147ff81528df", size = 81889, upload-time = "2025-08-12T05:52:29.023Z" },
+    { url = "https://files.pythonhosted.org/packages/04/9c/a56b5ac0e2473bdc3fb11b22dd69ff423154d63861cf77911cdde5e38fd2/wrapt-1.17.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:46acc57b331e0b3bcb3e1ca3b421d65637915cfcd65eb783cb2f78a511193f9b", size = 81344, upload-time = "2025-08-12T05:52:50.869Z" },
+    { url = "https://files.pythonhosted.org/packages/93/4c/9bd735c42641d81cb58d7bfb142c58f95c833962d15113026705add41a07/wrapt-1.17.3-cp39-cp39-win32.whl", hash = "sha256:3e62d15d3cfa26e3d0788094de7b64efa75f3a53875cdbccdf78547aed547a81", size = 36462, upload-time = "2025-08-12T05:53:19.623Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/ea/0b72f29cb5ebc16eb55c57dc0c98e5de76fc97f435fd407f7d409459c0a6/wrapt-1.17.3-cp39-cp39-win_amd64.whl", hash = "sha256:1f23fa283f51c890eda8e34e4937079114c74b4c81d2b2f1f1d94948f5cc3d7f", size = 38740, upload-time = "2025-08-12T05:53:18.271Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/8b/9eae65fb92321e38dbfec7719b87d840a4b92fde83fd1bbf238c5488d055/wrapt-1.17.3-cp39-cp39-win_arm64.whl", hash = "sha256:24c2ed34dc222ed754247a2702b1e1e89fdbaa4016f324b4b8f1a802d4ffe87f", size = 36806, upload-time = "2025-08-12T05:52:58.765Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/f6/a933bd70f98e9cf3e08167fc5cd7aaaca49147e48411c0bd5ae701bb2194/wrapt-1.17.3-py3-none-any.whl", hash = "sha256:7171ae35d2c33d326ac19dd8facb1e82e5fd04ef8c6c0e394d7af55a55051c22", size = 23591, upload-time = "2025-08-12T05:53:20.674Z" },
 ]
 
 [[package]]
-name = "websockets"
-version = "14.2"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/94/54/8359678c726243d19fae38ca14a334e740782336c9f19700858c4eb64a1e/websockets-14.2.tar.gz", hash = "sha256:5059ed9c54945efb321f097084b4c7e52c246f2c869815876a69d1efc4ad6eb5", size = 164394 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/28/fa/76607eb7dcec27b2d18d63f60a32e60e2b8629780f343bb83a4dbb9f4350/websockets-14.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e8179f95323b9ab1c11723e5d91a89403903f7b001828161b480a7810b334885", size = 163089 },
-    { url = "https://files.pythonhosted.org/packages/9e/00/ad2246b5030575b79e7af0721810fdaecaf94c4b2625842ef7a756fa06dd/websockets-14.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0d8c3e2cdb38f31d8bd7d9d28908005f6fa9def3324edb9bf336d7e4266fd397", size = 160741 },
-    { url = "https://files.pythonhosted.org/packages/72/f7/60f10924d333a28a1ff3fcdec85acf226281331bdabe9ad74947e1b7fc0a/websockets-14.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:714a9b682deb4339d39ffa674f7b674230227d981a37d5d174a4a83e3978a610", size = 160996 },
-    { url = "https://files.pythonhosted.org/packages/63/7c/c655789cf78648c01ac6ecbe2d6c18f91b75bdc263ffee4d08ce628d12f0/websockets-14.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f2e53c72052f2596fb792a7acd9704cbc549bf70fcde8a99e899311455974ca3", size = 169974 },
-    { url = "https://files.pythonhosted.org/packages/fb/5b/013ed8b4611857ac92ac631079c08d9715b388bd1d88ec62e245f87a39df/websockets-14.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e3fbd68850c837e57373d95c8fe352203a512b6e49eaae4c2f4088ef8cf21980", size = 168985 },
-    { url = "https://files.pythonhosted.org/packages/cd/33/aa3e32fd0df213a5a442310754fe3f89dd87a0b8e5b4e11e0991dd3bcc50/websockets-14.2-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b27ece32f63150c268593d5fdb82819584831a83a3f5809b7521df0685cd5d8", size = 169297 },
-    { url = "https://files.pythonhosted.org/packages/93/17/dae0174883d6399f57853ac44abf5f228eaba86d98d160f390ffabc19b6e/websockets-14.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4daa0faea5424d8713142b33825fff03c736f781690d90652d2c8b053345b0e7", size = 169677 },
-    { url = "https://files.pythonhosted.org/packages/42/e2/0375af7ac00169b98647c804651c515054b34977b6c1354f1458e4116c1e/websockets-14.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:bc63cee8596a6ec84d9753fd0fcfa0452ee12f317afe4beae6b157f0070c6c7f", size = 169089 },
-    { url = "https://files.pythonhosted.org/packages/73/8d/80f71d2a351a44b602859af65261d3dde3a0ce4e76cf9383738a949e0cc3/websockets-14.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:7a570862c325af2111343cc9b0257b7119b904823c675b22d4ac547163088d0d", size = 169026 },
-    { url = "https://files.pythonhosted.org/packages/48/97/173b1fa6052223e52bb4054a141433ad74931d94c575e04b654200b98ca4/websockets-14.2-cp310-cp310-win32.whl", hash = "sha256:75862126b3d2d505e895893e3deac0a9339ce750bd27b4ba515f008b5acf832d", size = 163967 },
-    { url = "https://files.pythonhosted.org/packages/c0/5b/2fcf60f38252a4562b28b66077e0d2b48f91fef645d5f78874cd1dec807b/websockets-14.2-cp310-cp310-win_amd64.whl", hash = "sha256:cc45afb9c9b2dc0852d5c8b5321759cf825f82a31bfaf506b65bf4668c96f8b2", size = 164413 },
-    { url = "https://files.pythonhosted.org/packages/15/b6/504695fb9a33df0ca56d157f5985660b5fc5b4bf8c78f121578d2d653392/websockets-14.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:3bdc8c692c866ce5fefcaf07d2b55c91d6922ac397e031ef9b774e5b9ea42166", size = 163088 },
-    { url = "https://files.pythonhosted.org/packages/81/26/ebfb8f6abe963c795122439c6433c4ae1e061aaedfc7eff32d09394afbae/websockets-14.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c93215fac5dadc63e51bcc6dceca72e72267c11def401d6668622b47675b097f", size = 160745 },
-    { url = "https://files.pythonhosted.org/packages/a1/c6/1435ad6f6dcbff80bb95e8986704c3174da8866ddb751184046f5c139ef6/websockets-14.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1c9b6535c0e2cf8a6bf938064fb754aaceb1e6a4a51a80d884cd5db569886910", size = 160995 },
-    { url = "https://files.pythonhosted.org/packages/96/63/900c27cfe8be1a1f2433fc77cd46771cf26ba57e6bdc7cf9e63644a61863/websockets-14.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0a52a6d7cf6938e04e9dceb949d35fbdf58ac14deea26e685ab6368e73744e4c", size = 170543 },
-    { url = "https://files.pythonhosted.org/packages/00/8b/bec2bdba92af0762d42d4410593c1d7d28e9bfd952c97a3729df603dc6ea/websockets-14.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9f05702e93203a6ff5226e21d9b40c037761b2cfb637187c9802c10f58e40473", size = 169546 },
-    { url = "https://files.pythonhosted.org/packages/6b/a9/37531cb5b994f12a57dec3da2200ef7aadffef82d888a4c29a0d781568e4/websockets-14.2-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:22441c81a6748a53bfcb98951d58d1af0661ab47a536af08920d129b4d1c3473", size = 169911 },
-    { url = "https://files.pythonhosted.org/packages/60/d5/a6eadba2ed9f7e65d677fec539ab14a9b83de2b484ab5fe15d3d6d208c28/websockets-14.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:efd9b868d78b194790e6236d9cbc46d68aba4b75b22497eb4ab64fa640c3af56", size = 170183 },
-    { url = "https://files.pythonhosted.org/packages/76/57/a338ccb00d1df881c1d1ee1f2a20c9c1b5b29b51e9e0191ee515d254fea6/websockets-14.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:1a5a20d5843886d34ff8c57424cc65a1deda4375729cbca4cb6b3353f3ce4142", size = 169623 },
-    { url = "https://files.pythonhosted.org/packages/64/22/e5f7c33db0cb2c1d03b79fd60d189a1da044e2661f5fd01d629451e1db89/websockets-14.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:34277a29f5303d54ec6468fb525d99c99938607bc96b8d72d675dee2b9f5bf1d", size = 169583 },
-    { url = "https://files.pythonhosted.org/packages/aa/2e/2b4662237060063a22e5fc40d46300a07142afe30302b634b4eebd717c07/websockets-14.2-cp311-cp311-win32.whl", hash = "sha256:02687db35dbc7d25fd541a602b5f8e451a238ffa033030b172ff86a93cb5dc2a", size = 163969 },
-    { url = "https://files.pythonhosted.org/packages/94/a5/0cda64e1851e73fc1ecdae6f42487babb06e55cb2f0dc8904b81d8ef6857/websockets-14.2-cp311-cp311-win_amd64.whl", hash = "sha256:862e9967b46c07d4dcd2532e9e8e3c2825e004ffbf91a5ef9dde519ee2effb0b", size = 164408 },
-    { url = "https://files.pythonhosted.org/packages/c1/81/04f7a397653dc8bec94ddc071f34833e8b99b13ef1a3804c149d59f92c18/websockets-14.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:1f20522e624d7ffbdbe259c6b6a65d73c895045f76a93719aa10cd93b3de100c", size = 163096 },
-    { url = "https://files.pythonhosted.org/packages/ec/c5/de30e88557e4d70988ed4d2eabd73fd3e1e52456b9f3a4e9564d86353b6d/websockets-14.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:647b573f7d3ada919fd60e64d533409a79dcf1ea21daeb4542d1d996519ca967", size = 160758 },
-    { url = "https://files.pythonhosted.org/packages/e5/8c/d130d668781f2c77d106c007b6c6c1d9db68239107c41ba109f09e6c218a/websockets-14.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6af99a38e49f66be5a64b1e890208ad026cda49355661549c507152113049990", size = 160995 },
-    { url = "https://files.pythonhosted.org/packages/a6/bc/f6678a0ff17246df4f06765e22fc9d98d1b11a258cc50c5968b33d6742a1/websockets-14.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:091ab63dfc8cea748cc22c1db2814eadb77ccbf82829bac6b2fbe3401d548eda", size = 170815 },
-    { url = "https://files.pythonhosted.org/packages/d8/b2/8070cb970c2e4122a6ef38bc5b203415fd46460e025652e1ee3f2f43a9a3/websockets-14.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b374e8953ad477d17e4851cdc66d83fdc2db88d9e73abf755c94510ebddceb95", size = 169759 },
-    { url = "https://files.pythonhosted.org/packages/81/da/72f7caabd94652e6eb7e92ed2d3da818626e70b4f2b15a854ef60bf501ec/websockets-14.2-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a39d7eceeea35db85b85e1169011bb4321c32e673920ae9c1b6e0978590012a3", size = 170178 },
-    { url = "https://files.pythonhosted.org/packages/31/e0/812725b6deca8afd3a08a2e81b3c4c120c17f68c9b84522a520b816cda58/websockets-14.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:0a6f3efd47ffd0d12080594f434faf1cd2549b31e54870b8470b28cc1d3817d9", size = 170453 },
-    { url = "https://files.pythonhosted.org/packages/66/d3/8275dbc231e5ba9bb0c4f93144394b4194402a7a0c8ffaca5307a58ab5e3/websockets-14.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:065ce275e7c4ffb42cb738dd6b20726ac26ac9ad0a2a48e33ca632351a737267", size = 169830 },
-    { url = "https://files.pythonhosted.org/packages/a3/ae/e7d1a56755ae15ad5a94e80dd490ad09e345365199600b2629b18ee37bc7/websockets-14.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e9d0e53530ba7b8b5e389c02282f9d2aa47581514bd6049d3a7cffe1385cf5fe", size = 169824 },
-    { url = "https://files.pythonhosted.org/packages/b6/32/88ccdd63cb261e77b882e706108d072e4f1c839ed723bf91a3e1f216bf60/websockets-14.2-cp312-cp312-win32.whl", hash = "sha256:20e6dd0984d7ca3037afcb4494e48c74ffb51e8013cac71cf607fffe11df7205", size = 163981 },
-    { url = "https://files.pythonhosted.org/packages/b3/7d/32cdb77990b3bdc34a306e0a0f73a1275221e9a66d869f6ff833c95b56ef/websockets-14.2-cp312-cp312-win_amd64.whl", hash = "sha256:44bba1a956c2c9d268bdcdf234d5e5ff4c9b6dc3e300545cbe99af59dda9dcce", size = 164421 },
-    { url = "https://files.pythonhosted.org/packages/82/94/4f9b55099a4603ac53c2912e1f043d6c49d23e94dd82a9ce1eb554a90215/websockets-14.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:6f1372e511c7409a542291bce92d6c83320e02c9cf392223272287ce55bc224e", size = 163102 },
-    { url = "https://files.pythonhosted.org/packages/8e/b7/7484905215627909d9a79ae07070057afe477433fdacb59bf608ce86365a/websockets-14.2-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:4da98b72009836179bb596a92297b1a61bb5a830c0e483a7d0766d45070a08ad", size = 160766 },
-    { url = "https://files.pythonhosted.org/packages/a3/a4/edb62efc84adb61883c7d2c6ad65181cb087c64252138e12d655989eec05/websockets-14.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f8a86a269759026d2bde227652b87be79f8a734e582debf64c9d302faa1e9f03", size = 160998 },
-    { url = "https://files.pythonhosted.org/packages/f5/79/036d320dc894b96af14eac2529967a6fc8b74f03b83c487e7a0e9043d842/websockets-14.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:86cf1aaeca909bf6815ea714d5c5736c8d6dd3a13770e885aafe062ecbd04f1f", size = 170780 },
-    { url = "https://files.pythonhosted.org/packages/63/75/5737d21ee4dd7e4b9d487ee044af24a935e36a9ff1e1419d684feedcba71/websockets-14.2-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a9b0f6c3ba3b1240f602ebb3971d45b02cc12bd1845466dd783496b3b05783a5", size = 169717 },
-    { url = "https://files.pythonhosted.org/packages/2c/3c/bf9b2c396ed86a0b4a92ff4cdaee09753d3ee389be738e92b9bbd0330b64/websockets-14.2-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:669c3e101c246aa85bc8534e495952e2ca208bd87994650b90a23d745902db9a", size = 170155 },
-    { url = "https://files.pythonhosted.org/packages/75/2d/83a5aca7247a655b1da5eb0ee73413abd5c3a57fc8b92915805e6033359d/websockets-14.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:eabdb28b972f3729348e632ab08f2a7b616c7e53d5414c12108c29972e655b20", size = 170495 },
-    { url = "https://files.pythonhosted.org/packages/79/dd/699238a92761e2f943885e091486378813ac8f43e3c84990bc394c2be93e/websockets-14.2-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:2066dc4cbcc19f32c12a5a0e8cc1b7ac734e5b64ac0a325ff8353451c4b15ef2", size = 169880 },
-    { url = "https://files.pythonhosted.org/packages/c8/c9/67a8f08923cf55ce61aadda72089e3ed4353a95a3a4bc8bf42082810e580/websockets-14.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:ab95d357cd471df61873dadf66dd05dd4709cae001dd6342edafc8dc6382f307", size = 169856 },
-    { url = "https://files.pythonhosted.org/packages/17/b1/1ffdb2680c64e9c3921d99db460546194c40d4acbef999a18c37aa4d58a3/websockets-14.2-cp313-cp313-win32.whl", hash = "sha256:a9e72fb63e5f3feacdcf5b4ff53199ec8c18d66e325c34ee4c551ca748623bbc", size = 163974 },
-    { url = "https://files.pythonhosted.org/packages/14/13/8b7fc4cb551b9cfd9890f0fd66e53c18a06240319915533b033a56a3d520/websockets-14.2-cp313-cp313-win_amd64.whl", hash = "sha256:b439ea828c4ba99bb3176dc8d9b933392a2413c0f6b149fdcba48393f573377f", size = 164420 },
-    { url = "https://files.pythonhosted.org/packages/6f/eb/367e0ed7b8a960b4fc12c7c6bf3ebddf06875037de641637994849560d47/websockets-14.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:7cd5706caec1686c5d233bc76243ff64b1c0dc445339bd538f30547e787c11fe", size = 163087 },
-    { url = "https://files.pythonhosted.org/packages/96/f7/1f18d028ec4a2c14598dfec6a73381a915c27464b693873198c1de872095/websockets-14.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ec607328ce95a2f12b595f7ae4c5d71bf502212bddcea528290b35c286932b12", size = 160740 },
-    { url = "https://files.pythonhosted.org/packages/5c/db/b4b353fb9c3f0eaa8138ea4c76e6fa555b6d2821ed2d51d0ac3c320bc57e/websockets-14.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:da85651270c6bfb630136423037dd4975199e5d4114cae6d3066641adcc9d1c7", size = 160992 },
-    { url = "https://files.pythonhosted.org/packages/b9/b1/9149e420c61f375e432654d5c1545e563b90ac1f829ee1a8d1dccaf0869d/websockets-14.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c3ecadc7ce90accf39903815697917643f5b7cfb73c96702318a096c00aa71f5", size = 169757 },
-    { url = "https://files.pythonhosted.org/packages/2b/33/0bb58204191e113212360f1392b6b1e9f85f62c7ca5b3b15f52f2f835516/websockets-14.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1979bee04af6a78608024bad6dfcc0cc930ce819f9e10342a29a05b5320355d0", size = 168762 },
-    { url = "https://files.pythonhosted.org/packages/be/3d/c3c192f16210d7b7535fbf4ee9a299612f4dccff665587617b13fa0a6aa3/websockets-14.2-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2dddacad58e2614a24938a50b85969d56f88e620e3f897b7d80ac0d8a5800258", size = 169060 },
-    { url = "https://files.pythonhosted.org/packages/a6/73/75efa8d9e4b1b257818a7b7a0b9ac84a07c91120b52148941370ef2c8f16/websockets-14.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:89a71173caaf75fa71a09a5f614f450ba3ec84ad9fca47cb2422a860676716f0", size = 169457 },
-    { url = "https://files.pythonhosted.org/packages/a4/11/300cf36cfd6990ffb218394862f0513be8c21917c9ff5e362f94599caedd/websockets-14.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6af6a4b26eea4fc06c6818a6b962a952441e0e39548b44773502761ded8cc1d4", size = 168860 },
-    { url = "https://files.pythonhosted.org/packages/c0/3d/5fd82500714ab7c09f003bde671dad1a3a131ac77b6b11ada72e466de4f6/websockets-14.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:80c8efa38957f20bba0117b48737993643204645e9ec45512579132508477cfc", size = 168825 },
-    { url = "https://files.pythonhosted.org/packages/88/16/715580eb6caaacc232f303e9619103a42dcd354b0854baa5ed26aacaf828/websockets-14.2-cp39-cp39-win32.whl", hash = "sha256:2e20c5f517e2163d76e2729104abc42639c41cf91f7b1839295be43302713661", size = 163960 },
-    { url = "https://files.pythonhosted.org/packages/63/a7/a1035cb198eaa12eaa9621aaaa3ec021b0e3bac96e1df9ceb6bfe5e53e5f/websockets-14.2-cp39-cp39-win_amd64.whl", hash = "sha256:b4c8cef610e8d7c70dea92e62b6814a8cd24fbd01d7103cc89308d2bfe1659ef", size = 164424 },
-    { url = "https://files.pythonhosted.org/packages/10/3d/91d3d2bb1325cd83e8e2c02d0262c7d4426dc8fa0831ef1aa4d6bf2041af/websockets-14.2-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:d7d9cafbccba46e768be8a8ad4635fa3eae1ffac4c6e7cb4eb276ba41297ed29", size = 160773 },
-    { url = "https://files.pythonhosted.org/packages/33/7c/cdedadfef7381939577858b1b5718a4ab073adbb584e429dd9d9dc9bfe16/websockets-14.2-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:c76193c1c044bd1e9b3316dcc34b174bbf9664598791e6fb606d8d29000e070c", size = 161007 },
-    { url = "https://files.pythonhosted.org/packages/ca/35/7a20a3c450b27c04e50fbbfc3dfb161ed8e827b2a26ae31c4b59b018b8c6/websockets-14.2-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fd475a974d5352390baf865309fe37dec6831aafc3014ffac1eea99e84e83fc2", size = 162264 },
-    { url = "https://files.pythonhosted.org/packages/e8/9c/e3f9600564b0c813f2448375cf28b47dc42c514344faed3a05d71fb527f9/websockets-14.2-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2c6c0097a41968b2e2b54ed3424739aab0b762ca92af2379f152c1aef0187e1c", size = 161873 },
-    { url = "https://files.pythonhosted.org/packages/3f/37/260f189b16b2b8290d6ae80c9f96d8b34692cf1bb3475df54c38d3deb57d/websockets-14.2-pp310-pypy310_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6d7ff794c8b36bc402f2e07c0b2ceb4a2424147ed4785ff03e2a7af03711d60a", size = 161818 },
-    { url = "https://files.pythonhosted.org/packages/ff/1e/e47dedac8bf7140e59aa6a679e850c4df9610ae844d71b6015263ddea37b/websockets-14.2-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:dec254fcabc7bd488dab64846f588fc5b6fe0d78f641180030f8ea27b76d72c3", size = 164465 },
-    { url = "https://files.pythonhosted.org/packages/f7/c0/8e9325c4987dcf66d4a0d63ec380d4aefe8dcc1e521af71ad17adf2c1ae2/websockets-14.2-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:bbe03eb853e17fd5b15448328b4ec7fb2407d45fb0245036d06a3af251f8e48f", size = 160773 },
-    { url = "https://files.pythonhosted.org/packages/5a/6e/c9a7f2edd4afddc4f8cccfc4e12468b7f6ec40f28d1b1e966a8d0298b875/websockets-14.2-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:a3c4aa3428b904d5404a0ed85f3644d37e2cb25996b7f096d77caeb0e96a3b42", size = 161006 },
-    { url = "https://files.pythonhosted.org/packages/f3/10/b90ece894828c954e674a81cb0db250e6c324c54db30a8b19e96431f928f/websockets-14.2-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:577a4cebf1ceaf0b65ffc42c54856214165fb8ceeba3935852fc33f6b0c55e7f", size = 162260 },
-    { url = "https://files.pythonhosted.org/packages/52/93/1147b6b5464a5fb6e8987da3ec7991dcc44f9090f67d9c841d7382fed429/websockets-14.2-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ad1c1d02357b7665e700eca43a31d52814ad9ad9b89b58118bdabc365454b574", size = 161868 },
-    { url = "https://files.pythonhosted.org/packages/32/ab/f7d80b4049bff0aa617507330db3a27389d0e70df54e29f7a3d76bbd2086/websockets-14.2-pp39-pypy39_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f390024a47d904613577df83ba700bd189eedc09c57af0a904e5c39624621270", size = 161813 },
-    { url = "https://files.pythonhosted.org/packages/cd/cc/adc9fb85f031b8df8e9f3d96cc004df25d2643e503953af5223c5b6825b7/websockets-14.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:3c1426c021c38cf92b453cdf371228d3430acd775edee6bac5a4d577efc72365", size = 164457 },
-    { url = "https://files.pythonhosted.org/packages/7b/c8/d529f8a32ce40d98309f4470780631e971a5a842b60aec864833b3615786/websockets-14.2-py3-none-any.whl", hash = "sha256:7a6ceec4ea84469f15cf15807a747e9efe57e369c384fa86e022b3bea679b79b", size = 157416 },
+name = "xformers"
+version = "0.0.32.post2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "torch" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/47/99/68d2e7a17b65a08180d6b120d1d293d5855a0c5999cb30a79fb466b49642/xformers-0.0.32.post2.tar.gz", hash = "sha256:9538be803969c6e1ca16a3ece921e472c24f79970b10be1087a389dcb66e412a", size = 12105990, upload-time = "2025-08-15T11:48:56.584Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ce/16/4ed2fbf7c20ba20565d9a1fc544c2bef39d409c8eca1949654c6ff9f8452/xformers-0.0.32.post2-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:3d1fefe0006eb00a730b3b197ff41aab085f1209b50419baab6152445f09e508", size = 117196673, upload-time = "2025-08-15T11:48:46.726Z" },
 ]
 
 [[package]]
-name = "xformers"
-version = "0.0.26.post1"
+name = "xgrammar"
+version = "0.1.23"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "numpy" },
+    { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
+    { name = "ninja" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pydantic" },
     { name = "torch" },
+    { name = "transformers" },
+    { name = "triton", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fc/d3/5eab1cb3171fc9d4e5613025e965b72727f6d12ae820d2d7ab9c37320eb6/xformers-0.0.26.post1.tar.gz", hash = "sha256:1d14b5f999ede649198379b0470ebdd25007ba224ae336ef958124158a6de8b1", size = 4087142 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/32/e7/27003645ef99e7571fb6964cd2f39da3f1b3f3011aa00bb2d3ac9b790757/xformers-0.0.26.post1-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:1fe0f9a3dddb7f6175c2e34de2f318d6742eec28c068b8334b093016b2c9c2c8", size = 222745400 },
-    { url = "https://files.pythonhosted.org/packages/7a/53/f4a36d24f6c2d06fccc356986bb45c2cb2fd06609986d44a925d57f38070/xformers-0.0.26.post1-cp310-cp310-win_amd64.whl", hash = "sha256:e34b8dd6982077bee0c8eb2db8bc1513177201bfe0af890a4db42d8d31c966a5", size = 208730032 },
-    { url = "https://files.pythonhosted.org/packages/ce/72/58e13d633efada336f65e23290dfc0bc9497c60fd9b8fdaab59b555b5f28/xformers-0.0.26.post1-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:0d35615870b9237077aec51802a5a821f9e9d2708730c8914d5cff301fb29558", size = 222780107 },
-    { url = "https://files.pythonhosted.org/packages/bb/6b/ac99df9545d965c11172ec3613b4652d6f0f9098328551cfe0bbbfa634fe/xformers-0.0.26.post1-cp311-cp311-win_amd64.whl", hash = "sha256:d05c547b4ba603fc8e21fad03a342138eaaece35fe0a1692e6ee0d061ddd21ac", size = 208727774 },
-    { url = "https://files.pythonhosted.org/packages/1d/42/7728b1d16425d3f5505eeb70749e15a71257e126a90ab44184d231d00341/xformers-0.0.26.post1-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:16dcfa801450a03f148d6d5f7ba2b5540a27c52517946570fe0e1f75238d73a1", size = 222743742 },
-    { url = "https://files.pythonhosted.org/packages/28/f7/452f15a1a69ef553f05119e03f4c26fa683793491520c88d0055ca90c9ec/xformers-0.0.26.post1-cp39-cp39-win_amd64.whl", hash = "sha256:cf267bac8f4454db1056e4e6ff9e4df1e02b2b31d8116d50913af15fa6ae7132", size = 208757765 },
+sdist = { url = "https://files.pythonhosted.org/packages/e9/12/958457553a87c31bdb18c8395b88bb3255f7c7373ab3a0b046d3b7f37f86/xgrammar-0.1.23.tar.gz", hash = "sha256:5ef280455c1ac008f052d7ea92286f0ca3a3d7ab360224894ac69277c8827113", size = 2263693, upload-time = "2025-08-15T07:31:42.792Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/95/f0/bdd2c9b97ad8db283c702119a23f3e338a3b9a5a0f6a7469d6d65938fe5a/xgrammar-0.1.23-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:7a4f47812ea93f79634b2ab14580ccd20893e9c81ff43929e1ea1a6b32e56015", size = 620720, upload-time = "2025-08-15T07:31:06.428Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/ee/bdd8783903437807856cbacd23ba48b4eea1b8b4e54057103961031dbd7b/xgrammar-0.1.23-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a40d28e5011f4e2a9d874ad00f7c4b3c1e816690b3f2d25904c64fd0c32cbbf9", size = 564940, upload-time = "2025-08-15T07:31:08.271Z" },
+    { url = "https://files.pythonhosted.org/packages/43/c2/d18e5af82cd18c2ff908f2438c5329e27ff6af3c11656bfb2582e96d8700/xgrammar-0.1.23-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c5439d9a7819b98bb82a699e63e9b233569eacd1d8d0d5366f4cca7d61cc2742", size = 7680078, upload-time = "2025-08-15T07:31:09.922Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/83/9fd0d2d4e06ef27b13f9343269c0110574622c543c6527d7dcafc55a194e/xgrammar-0.1.23-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41d262c48354f2d845e8654c05be681c17fb9fc037cde6883ef434a07b1e9ace", size = 7854128, upload-time = "2025-08-15T07:31:11.458Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/b7/0a328254411983f223baa33f10b54793570512e0868e6ba7b47cf2c7caae/xgrammar-0.1.23-cp310-cp310-win_amd64.whl", hash = "sha256:b2a40e4dd242dedc83c52f9699569efe04b448bd1cd06627d12a7093b4d800ed", size = 646685, upload-time = "2025-08-15T07:31:13.218Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/d5/b5b2dde17b1fbda9240af6cbb19523704eccf876610331c3317c08654106/xgrammar-0.1.23-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:af31d12105e31afa257870e9016da3c631300d040e63684b5aabae074916f552", size = 620555, upload-time = "2025-08-15T07:31:15.217Z" },
+    { url = "https://files.pythonhosted.org/packages/04/55/b92e490ddb7f510fbd2d8e2b2fa5a8da440cfb80221cb16a9a3d5f33322c/xgrammar-0.1.23-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ba90f257f75f12014360d9512f95caf731b18efe0562f91206b44b28388533d6", size = 564886, upload-time = "2025-08-15T07:31:16.44Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/61/07dd19e914512fbb9cb08a3e0479f75fa827c325abc23cf9fef48dd42515/xgrammar-0.1.23-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8612ed64bcf86526837df9c4c1061a2b00a39a51cb1ab1fe61f881c1c2ab5f1e", size = 7681629, upload-time = "2025-08-15T07:31:17.795Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/d1/500caef952bea84f335453fee01836d1b8c8b700b173e07a37c6a4b62d67/xgrammar-0.1.23-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:280d6e114b6ea57ee9dc4b5e31b5f897132e74d279b8064e7679893804498652", size = 7855408, upload-time = "2025-08-15T07:31:19.647Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/55/084f9ec87174220914f7b96c59f20e64ae2bc7938765c8b1d0191ae45d90/xgrammar-0.1.23-cp311-cp311-win_amd64.whl", hash = "sha256:a21b9f85fa321a2731106ffc167fde15d90e5af1779b64bc75c36fa809ea7925", size = 646693, upload-time = "2025-08-15T07:31:21.072Z" },
+    { url = "https://files.pythonhosted.org/packages/37/40/4fd2f41c603bbe4e60d51c7956ace6f386dd69b20ad62c443a1dbcac279a/xgrammar-0.1.23-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:6605ae8f7d7748ac00b55f4377ddc895bc9b7f8252760b1265145be9a2bc87e4", size = 620111, upload-time = "2025-08-15T07:31:22.652Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/07/55b50afb50813dab89a7ef52afdf73d8c5e22a15636edc27ca3cd4114665/xgrammar-0.1.23-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:20692f98e79fe85e1f33482e2f5ae92ba204a7ac95c3322795d39b910292df46", size = 564421, upload-time = "2025-08-15T07:31:23.875Z" },
+    { url = "https://files.pythonhosted.org/packages/96/d8/be7caa3a56ce96defc69b3d9de8538924a0a7114da4e89fe61c5bd557961/xgrammar-0.1.23-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7bde8472e2a28e529bff26f3ca2db56679c8dccbae68813c0ae61db1edff95a5", size = 7683250, upload-time = "2025-08-15T07:31:25.276Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/13/53d950b93a361ef73e5930050916fa36c23fade80ee05cfb0339c044e951/xgrammar-0.1.23-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e0ff9c0a1d46c95d82345a5bf026956ef6d98f1aac7115b57ce88d1d93c4a374", size = 7858258, upload-time = "2025-08-15T07:31:26.782Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/11/2ac0d10c88196f6fc6efa085a175483a72532cc7c2027cbd0583c759a748/xgrammar-0.1.23-cp312-cp312-win_amd64.whl", hash = "sha256:b330a5d673a53a657beb641af8e90146bba4ec18203c440ee7b7766856f5991c", size = 645937, upload-time = "2025-08-15T07:31:28.677Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/d5/797da4ec365d7efdfae47fed89dbc6b408ed2ac397ed08549c3d8a45e72c/xgrammar-0.1.23-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:0583caae0d1478e9c8fc485322c2d5ac9f6766fd995bfe3b5460704b610a12fc", size = 621015, upload-time = "2025-08-15T07:31:35.452Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/46/493805568645b4ea5be1bdcfff1cf845c084fa1f2e04a5bfe4f295a07793/xgrammar-0.1.23-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3dcad40085f1e280ed8bbe5ea19f152e3704fb6456760085481c4e6a376c288c", size = 565126, upload-time = "2025-08-15T07:31:36.68Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/67/733c5425ca0902335593ca9d1756fc131c5934f5fbff1ba28fc37bfef24a/xgrammar-0.1.23-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e7fdb6cf4106ad1b2dc21e3d0dd11b76bbaf3071712b8845124eb2db3171e32", size = 7681247, upload-time = "2025-08-15T07:31:38.409Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/62/b62f1defb085eeea0bbbc1d4a22fb9ea2418175719c44d2d618508976dfa/xgrammar-0.1.23-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f7feb92eb95f093392f05d7b4dd76a49f43909f27df02710f04bc7b717e7e4da", size = 7855034, upload-time = "2025-08-15T07:31:40.03Z" },
+    { url = "https://files.pythonhosted.org/packages/95/5b/8d5fa56f276ccf4f52c5ee6d0a61abbe4defd19f5f8837ede18b16cd313c/xgrammar-0.1.23-cp39-cp39-win_amd64.whl", hash = "sha256:3fa7214b19b1258d980262ce6fce20de3dbc64a85c70d706097c74ed816191c7", size = 646935, upload-time = "2025-08-15T07:31:41.524Z" },
 ]
 
 [[package]]
 name = "xxhash"
 version = "3.5.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/00/5e/d6e5258d69df8b4ed8c83b6664f2b47d30d2dec551a29ad72a6c69eafd31/xxhash-3.5.0.tar.gz", hash = "sha256:84f2caddf951c9cbf8dc2e22a89d4ccf5d86391ac6418fe81e3c67d0cf60b45f", size = 84241 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/bb/8a/0e9feca390d512d293afd844d31670e25608c4a901e10202aa98785eab09/xxhash-3.5.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:ece616532c499ee9afbb83078b1b952beffef121d989841f7f4b3dc5ac0fd212", size = 31970 },
-    { url = "https://files.pythonhosted.org/packages/16/e6/be5aa49580cd064a18200ab78e29b88b1127e1a8c7955eb8ecf81f2626eb/xxhash-3.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:3171f693dbc2cef6477054a665dc255d996646b4023fe56cb4db80e26f4cc520", size = 30801 },
-    { url = "https://files.pythonhosted.org/packages/20/ee/b8a99ebbc6d1113b3a3f09e747fa318c3cde5b04bd9c197688fadf0eeae8/xxhash-3.5.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7c5d3e570ef46adaf93fc81b44aca6002b5a4d8ca11bd0580c07eac537f36680", size = 220927 },
-    { url = "https://files.pythonhosted.org/packages/58/62/15d10582ef159283a5c2b47f6d799fc3303fe3911d5bb0bcc820e1ef7ff4/xxhash-3.5.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7cb29a034301e2982df8b1fe6328a84f4b676106a13e9135a0d7e0c3e9f806da", size = 200360 },
-    { url = "https://files.pythonhosted.org/packages/23/41/61202663ea9b1bd8e53673b8ec9e2619989353dba8cfb68e59a9cbd9ffe3/xxhash-3.5.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5d0d307d27099bb0cbeea7260eb39ed4fdb99c5542e21e94bb6fd29e49c57a23", size = 428528 },
-    { url = "https://files.pythonhosted.org/packages/f2/07/d9a3059f702dec5b3b703737afb6dda32f304f6e9da181a229dafd052c29/xxhash-3.5.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0342aafd421795d740e514bc9858ebddfc705a75a8c5046ac56d85fe97bf196", size = 194149 },
-    { url = "https://files.pythonhosted.org/packages/eb/58/27caadf78226ecf1d62dbd0c01d152ed381c14c1ee4ad01f0d460fc40eac/xxhash-3.5.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3dbbd9892c5ebffeca1ed620cf0ade13eb55a0d8c84e0751a6653adc6ac40d0c", size = 207703 },
-    { url = "https://files.pythonhosted.org/packages/b1/08/32d558ce23e1e068453c39aed7b3c1cdc690c177873ec0ca3a90d5808765/xxhash-3.5.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4cc2d67fdb4d057730c75a64c5923abfa17775ae234a71b0200346bfb0a7f482", size = 216255 },
-    { url = "https://files.pythonhosted.org/packages/3f/d4/2b971e2d2b0a61045f842b622ef11e94096cf1f12cd448b6fd426e80e0e2/xxhash-3.5.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:ec28adb204b759306a3d64358a5e5c07d7b1dd0ccbce04aa76cb9377b7b70296", size = 202744 },
-    { url = "https://files.pythonhosted.org/packages/19/ae/6a6438864a8c4c39915d7b65effd85392ebe22710412902487e51769146d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:1328f6d8cca2b86acb14104e381225a3d7b42c92c4b86ceae814e5c400dbb415", size = 210115 },
-    { url = "https://files.pythonhosted.org/packages/48/7d/b3c27c27d1fc868094d02fe4498ccce8cec9fcc591825c01d6bcb0b4fc49/xxhash-3.5.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:8d47ebd9f5d9607fd039c1fbf4994e3b071ea23eff42f4ecef246ab2b7334198", size = 414247 },
-    { url = "https://files.pythonhosted.org/packages/a1/05/918f9e7d2fbbd334b829997045d341d6239b563c44e683b9a7ef8fe50f5d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b96d559e0fcddd3343c510a0fe2b127fbff16bf346dd76280b82292567523442", size = 191419 },
-    { url = "https://files.pythonhosted.org/packages/08/29/dfe393805b2f86bfc47c290b275f0b7c189dc2f4e136fd4754f32eb18a8d/xxhash-3.5.0-cp310-cp310-win32.whl", hash = "sha256:61c722ed8d49ac9bc26c7071eeaa1f6ff24053d553146d5df031802deffd03da", size = 30114 },
-    { url = "https://files.pythonhosted.org/packages/7b/d7/aa0b22c4ebb7c3ccb993d4c565132abc641cd11164f8952d89eb6a501909/xxhash-3.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:9bed5144c6923cc902cd14bb8963f2d5e034def4486ab0bbe1f58f03f042f9a9", size = 30003 },
-    { url = "https://files.pythonhosted.org/packages/69/12/f969b81541ee91b55f1ce469d7ab55079593c80d04fd01691b550e535000/xxhash-3.5.0-cp310-cp310-win_arm64.whl", hash = "sha256:893074d651cf25c1cc14e3bea4fceefd67f2921b1bb8e40fcfeba56820de80c6", size = 26773 },
-    { url = "https://files.pythonhosted.org/packages/b8/c7/afed0f131fbda960ff15eee7f304fa0eeb2d58770fade99897984852ef23/xxhash-3.5.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:02c2e816896dc6f85922ced60097bcf6f008dedfc5073dcba32f9c8dd786f3c1", size = 31969 },
-    { url = "https://files.pythonhosted.org/packages/8c/0c/7c3bc6d87e5235672fcc2fb42fd5ad79fe1033925f71bf549ee068c7d1ca/xxhash-3.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6027dcd885e21581e46d3c7f682cfb2b870942feeed58a21c29583512c3f09f8", size = 30800 },
-    { url = "https://files.pythonhosted.org/packages/04/9e/01067981d98069eec1c20201f8c145367698e9056f8bc295346e4ea32dd1/xxhash-3.5.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1308fa542bbdbf2fa85e9e66b1077eea3a88bef38ee8a06270b4298a7a62a166", size = 221566 },
-    { url = "https://files.pythonhosted.org/packages/d4/09/d4996de4059c3ce5342b6e1e6a77c9d6c91acce31f6ed979891872dd162b/xxhash-3.5.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c28b2fdcee797e1c1961cd3bcd3d545cab22ad202c846235197935e1df2f8ef7", size = 201214 },
-    { url = "https://files.pythonhosted.org/packages/62/f5/6d2dc9f8d55a7ce0f5e7bfef916e67536f01b85d32a9fbf137d4cadbee38/xxhash-3.5.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:924361811732ddad75ff23e90efd9ccfda4f664132feecb90895bade6a1b4623", size = 429433 },
-    { url = "https://files.pythonhosted.org/packages/d9/72/9256303f10e41ab004799a4aa74b80b3c5977d6383ae4550548b24bd1971/xxhash-3.5.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:89997aa1c4b6a5b1e5b588979d1da048a3c6f15e55c11d117a56b75c84531f5a", size = 194822 },
-    { url = "https://files.pythonhosted.org/packages/34/92/1a3a29acd08248a34b0e6a94f4e0ed9b8379a4ff471f1668e4dce7bdbaa8/xxhash-3.5.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:685c4f4e8c59837de103344eb1c8a3851f670309eb5c361f746805c5471b8c88", size = 208538 },
-    { url = "https://files.pythonhosted.org/packages/53/ad/7fa1a109663366de42f724a1cdb8e796a260dbac45047bce153bc1e18abf/xxhash-3.5.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dbd2ecfbfee70bc1a4acb7461fa6af7748ec2ab08ac0fa298f281c51518f982c", size = 216953 },
-    { url = "https://files.pythonhosted.org/packages/35/02/137300e24203bf2b2a49b48ce898ecce6fd01789c0fcd9c686c0a002d129/xxhash-3.5.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:25b5a51dc3dfb20a10833c8eee25903fd2e14059e9afcd329c9da20609a307b2", size = 203594 },
-    { url = "https://files.pythonhosted.org/packages/23/03/aeceb273933d7eee248c4322b98b8e971f06cc3880e5f7602c94e5578af5/xxhash-3.5.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a8fb786fb754ef6ff8c120cb96629fb518f8eb5a61a16aac3a979a9dbd40a084", size = 210971 },
-    { url = "https://files.pythonhosted.org/packages/e3/64/ed82ec09489474cbb35c716b189ddc1521d8b3de12b1b5ab41ce7f70253c/xxhash-3.5.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:a905ad00ad1e1c34fe4e9d7c1d949ab09c6fa90c919860c1534ff479f40fd12d", size = 415050 },
-    { url = "https://files.pythonhosted.org/packages/71/43/6db4c02dcb488ad4e03bc86d70506c3d40a384ee73c9b5c93338eb1f3c23/xxhash-3.5.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:963be41bcd49f53af6d795f65c0da9b4cc518c0dd9c47145c98f61cb464f4839", size = 192216 },
-    { url = "https://files.pythonhosted.org/packages/22/6d/db4abec29e7a567455344433d095fdb39c97db6955bb4a2c432e486b4d28/xxhash-3.5.0-cp311-cp311-win32.whl", hash = "sha256:109b436096d0a2dd039c355fa3414160ec4d843dfecc64a14077332a00aeb7da", size = 30120 },
-    { url = "https://files.pythonhosted.org/packages/52/1c/fa3b61c0cf03e1da4767213672efe186b1dfa4fc901a4a694fb184a513d1/xxhash-3.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:b702f806693201ad6c0a05ddbbe4c8f359626d0b3305f766077d51388a6bac58", size = 30003 },
-    { url = "https://files.pythonhosted.org/packages/6b/8e/9e6fc572acf6e1cc7ccb01973c213f895cb8668a9d4c2b58a99350da14b7/xxhash-3.5.0-cp311-cp311-win_arm64.whl", hash = "sha256:c4dcb4120d0cc3cc448624147dba64e9021b278c63e34a38789b688fd0da9bf3", size = 26777 },
-    { url = "https://files.pythonhosted.org/packages/07/0e/1bfce2502c57d7e2e787600b31c83535af83746885aa1a5f153d8c8059d6/xxhash-3.5.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:14470ace8bd3b5d51318782cd94e6f94431974f16cb3b8dc15d52f3b69df8e00", size = 31969 },
-    { url = "https://files.pythonhosted.org/packages/3f/d6/8ca450d6fe5b71ce521b4e5db69622383d039e2b253e9b2f24f93265b52c/xxhash-3.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:59aa1203de1cb96dbeab595ded0ad0c0056bb2245ae11fac11c0ceea861382b9", size = 30787 },
-    { url = "https://files.pythonhosted.org/packages/5b/84/de7c89bc6ef63d750159086a6ada6416cc4349eab23f76ab870407178b93/xxhash-3.5.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:08424f6648526076e28fae6ea2806c0a7d504b9ef05ae61d196d571e5c879c84", size = 220959 },
-    { url = "https://files.pythonhosted.org/packages/fe/86/51258d3e8a8545ff26468c977101964c14d56a8a37f5835bc0082426c672/xxhash-3.5.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:61a1ff00674879725b194695e17f23d3248998b843eb5e933007ca743310f793", size = 200006 },
-    { url = "https://files.pythonhosted.org/packages/02/0a/96973bd325412feccf23cf3680fd2246aebf4b789122f938d5557c54a6b2/xxhash-3.5.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f2f2c61bee5844d41c3eb015ac652a0229e901074951ae48581d58bfb2ba01be", size = 428326 },
-    { url = "https://files.pythonhosted.org/packages/11/a7/81dba5010f7e733de88af9555725146fc133be97ce36533867f4c7e75066/xxhash-3.5.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9d32a592cac88d18cc09a89172e1c32d7f2a6e516c3dfde1b9adb90ab5df54a6", size = 194380 },
-    { url = "https://files.pythonhosted.org/packages/fb/7d/f29006ab398a173f4501c0e4977ba288f1c621d878ec217b4ff516810c04/xxhash-3.5.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:70dabf941dede727cca579e8c205e61121afc9b28516752fd65724be1355cc90", size = 207934 },
-    { url = "https://files.pythonhosted.org/packages/8a/6e/6e88b8f24612510e73d4d70d9b0c7dff62a2e78451b9f0d042a5462c8d03/xxhash-3.5.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e5d0ddaca65ecca9c10dcf01730165fd858533d0be84c75c327487c37a906a27", size = 216301 },
-    { url = "https://files.pythonhosted.org/packages/af/51/7862f4fa4b75a25c3b4163c8a873f070532fe5f2d3f9b3fc869c8337a398/xxhash-3.5.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:3e5b5e16c5a480fe5f59f56c30abdeba09ffd75da8d13f6b9b6fd224d0b4d0a2", size = 203351 },
-    { url = "https://files.pythonhosted.org/packages/22/61/8d6a40f288f791cf79ed5bb113159abf0c81d6efb86e734334f698eb4c59/xxhash-3.5.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:149b7914451eb154b3dfaa721315117ea1dac2cc55a01bfbd4df7c68c5dd683d", size = 210294 },
-    { url = "https://files.pythonhosted.org/packages/17/02/215c4698955762d45a8158117190261b2dbefe9ae7e5b906768c09d8bc74/xxhash-3.5.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:eade977f5c96c677035ff39c56ac74d851b1cca7d607ab3d8f23c6b859379cab", size = 414674 },
-    { url = "https://files.pythonhosted.org/packages/31/5c/b7a8db8a3237cff3d535261325d95de509f6a8ae439a5a7a4ffcff478189/xxhash-3.5.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:fa9f547bd98f5553d03160967866a71056a60960be00356a15ecc44efb40ba8e", size = 192022 },
-    { url = "https://files.pythonhosted.org/packages/78/e3/dd76659b2811b3fd06892a8beb850e1996b63e9235af5a86ea348f053e9e/xxhash-3.5.0-cp312-cp312-win32.whl", hash = "sha256:f7b58d1fd3551b8c80a971199543379be1cee3d0d409e1f6d8b01c1a2eebf1f8", size = 30170 },
-    { url = "https://files.pythonhosted.org/packages/d9/6b/1c443fe6cfeb4ad1dcf231cdec96eb94fb43d6498b4469ed8b51f8b59a37/xxhash-3.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:fa0cafd3a2af231b4e113fba24a65d7922af91aeb23774a8b78228e6cd785e3e", size = 30040 },
-    { url = "https://files.pythonhosted.org/packages/0f/eb/04405305f290173acc0350eba6d2f1a794b57925df0398861a20fbafa415/xxhash-3.5.0-cp312-cp312-win_arm64.whl", hash = "sha256:586886c7e89cb9828bcd8a5686b12e161368e0064d040e225e72607b43858ba2", size = 26796 },
-    { url = "https://files.pythonhosted.org/packages/c9/b8/e4b3ad92d249be5c83fa72916c9091b0965cb0faeff05d9a0a3870ae6bff/xxhash-3.5.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:37889a0d13b0b7d739cfc128b1c902f04e32de17b33d74b637ad42f1c55101f6", size = 31795 },
-    { url = "https://files.pythonhosted.org/packages/fc/d8/b3627a0aebfbfa4c12a41e22af3742cf08c8ea84f5cc3367b5de2d039cce/xxhash-3.5.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:97a662338797c660178e682f3bc180277b9569a59abfb5925e8620fba00b9fc5", size = 30792 },
-    { url = "https://files.pythonhosted.org/packages/c3/cc/762312960691da989c7cd0545cb120ba2a4148741c6ba458aa723c00a3f8/xxhash-3.5.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7f85e0108d51092bdda90672476c7d909c04ada6923c14ff9d913c4f7dc8a3bc", size = 220950 },
-    { url = "https://files.pythonhosted.org/packages/fe/e9/cc266f1042c3c13750e86a535496b58beb12bf8c50a915c336136f6168dc/xxhash-3.5.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cd2fd827b0ba763ac919440042302315c564fdb797294d86e8cdd4578e3bc7f3", size = 199980 },
-    { url = "https://files.pythonhosted.org/packages/bf/85/a836cd0dc5cc20376de26b346858d0ac9656f8f730998ca4324921a010b9/xxhash-3.5.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:82085c2abec437abebf457c1d12fccb30cc8b3774a0814872511f0f0562c768c", size = 428324 },
-    { url = "https://files.pythonhosted.org/packages/b4/0e/15c243775342ce840b9ba34aceace06a1148fa1630cd8ca269e3223987f5/xxhash-3.5.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:07fda5de378626e502b42b311b049848c2ef38784d0d67b6f30bb5008642f8eb", size = 194370 },
-    { url = "https://files.pythonhosted.org/packages/87/a1/b028bb02636dfdc190da01951d0703b3d904301ed0ef6094d948983bef0e/xxhash-3.5.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c279f0d2b34ef15f922b77966640ade58b4ccdfef1c4d94b20f2a364617a493f", size = 207911 },
-    { url = "https://files.pythonhosted.org/packages/80/d5/73c73b03fc0ac73dacf069fdf6036c9abad82de0a47549e9912c955ab449/xxhash-3.5.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:89e66ceed67b213dec5a773e2f7a9e8c58f64daeb38c7859d8815d2c89f39ad7", size = 216352 },
-    { url = "https://files.pythonhosted.org/packages/b6/2a/5043dba5ddbe35b4fe6ea0a111280ad9c3d4ba477dd0f2d1fe1129bda9d0/xxhash-3.5.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:bcd51708a633410737111e998ceb3b45d3dbc98c0931f743d9bb0a209033a326", size = 203410 },
-    { url = "https://files.pythonhosted.org/packages/a2/b2/9a8ded888b7b190aed75b484eb5c853ddd48aa2896e7b59bbfbce442f0a1/xxhash-3.5.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:3ff2c0a34eae7df88c868be53a8dd56fbdf592109e21d4bfa092a27b0bf4a7bf", size = 210322 },
-    { url = "https://files.pythonhosted.org/packages/98/62/440083fafbc917bf3e4b67c2ade621920dd905517e85631c10aac955c1d2/xxhash-3.5.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:4e28503dccc7d32e0b9817aa0cbfc1f45f563b2c995b7a66c4c8a0d232e840c7", size = 414725 },
-    { url = "https://files.pythonhosted.org/packages/75/db/009206f7076ad60a517e016bb0058381d96a007ce3f79fa91d3010f49cc2/xxhash-3.5.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a6c50017518329ed65a9e4829154626f008916d36295b6a3ba336e2458824c8c", size = 192070 },
-    { url = "https://files.pythonhosted.org/packages/1f/6d/c61e0668943a034abc3a569cdc5aeae37d686d9da7e39cf2ed621d533e36/xxhash-3.5.0-cp313-cp313-win32.whl", hash = "sha256:53a068fe70301ec30d868ece566ac90d873e3bb059cf83c32e76012c889b8637", size = 30172 },
-    { url = "https://files.pythonhosted.org/packages/96/14/8416dce965f35e3d24722cdf79361ae154fa23e2ab730e5323aa98d7919e/xxhash-3.5.0-cp313-cp313-win_amd64.whl", hash = "sha256:80babcc30e7a1a484eab952d76a4f4673ff601f54d5142c26826502740e70b43", size = 30041 },
-    { url = "https://files.pythonhosted.org/packages/27/ee/518b72faa2073f5aa8e3262408d284892cb79cf2754ba0c3a5870645ef73/xxhash-3.5.0-cp313-cp313-win_arm64.whl", hash = "sha256:4811336f1ce11cac89dcbd18f3a25c527c16311709a89313c3acaf771def2d4b", size = 26801 },
-    { url = "https://files.pythonhosted.org/packages/d4/f6/531dd6858adf8877675270b9d6989b6dacfd1c2d7135b17584fc29866df3/xxhash-3.5.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bfc8cdd7f33d57f0468b0614ae634cc38ab9202c6957a60e31d285a71ebe0301", size = 31971 },
-    { url = "https://files.pythonhosted.org/packages/7c/a8/b2a42b6c9ae46e233f474f3d307c2e7bca8d9817650babeca048d2ad01d6/xxhash-3.5.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e0c48b6300cd0b0106bf49169c3e0536408dfbeb1ccb53180068a18b03c662ab", size = 30801 },
-    { url = "https://files.pythonhosted.org/packages/b4/92/9ac297e3487818f429bcf369c1c6a097edf5b56ed6fc1feff4c1882e87ef/xxhash-3.5.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fe1a92cfbaa0a1253e339ccec42dbe6db262615e52df591b68726ab10338003f", size = 220644 },
-    { url = "https://files.pythonhosted.org/packages/86/48/c1426dd3c86fc4a52f983301867463472f6a9013fb32d15991e60c9919b6/xxhash-3.5.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:33513d6cc3ed3b559134fb307aae9bdd94d7e7c02907b37896a6c45ff9ce51bd", size = 200021 },
-    { url = "https://files.pythonhosted.org/packages/f3/de/0ab8c79993765c94fc0d0c1a22b454483c58a0161e1b562f58b654f47660/xxhash-3.5.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eefc37f6138f522e771ac6db71a6d4838ec7933939676f3753eafd7d3f4c40bc", size = 428217 },
-    { url = "https://files.pythonhosted.org/packages/b4/b4/332647451ed7d2c021294b7c1e9c144dbb5586b1fb214ad4f5a404642835/xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a606c8070ada8aa2a88e181773fa1ef17ba65ce5dd168b9d08038e2a61b33754", size = 193868 },
-    { url = "https://files.pythonhosted.org/packages/f4/1c/a42c0a6cac752f84f7b44a90d1a9fa9047cf70bdba5198a304fde7cc471f/xxhash-3.5.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:42eca420c8fa072cc1dd62597635d140e78e384a79bb4944f825fbef8bfeeef6", size = 207403 },
-    { url = "https://files.pythonhosted.org/packages/c4/d7/04e1b0daae9dc9b02c73c1664cc8aa527498c3f66ccbc586eeb25bbe9f14/xxhash-3.5.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:604253b2143e13218ff1ef0b59ce67f18b8bd1c4205d2ffda22b09b426386898", size = 215978 },
-    { url = "https://files.pythonhosted.org/packages/c4/f4/05e15e67505228fc19ee98a79e427b3a0b9695f5567cd66ced5d66389883/xxhash-3.5.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6e93a5ad22f434d7876665444a97e713a8f60b5b1a3521e8df11b98309bff833", size = 202416 },
-    { url = "https://files.pythonhosted.org/packages/94/fb/e9028d3645bba5412a09de13ee36df276a567e60bdb31d499dafa46d76ae/xxhash-3.5.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:7a46e1d6d2817ba8024de44c4fd79913a90e5f7265434cef97026215b7d30df6", size = 209853 },
-    { url = "https://files.pythonhosted.org/packages/02/2c/18c6a622429368274739372d2f86c8125413ec169025c7d8ffb051784bba/xxhash-3.5.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:30eb2efe6503c379b7ab99c81ba4a779748e3830241f032ab46bd182bf5873af", size = 413926 },
-    { url = "https://files.pythonhosted.org/packages/72/bb/5b55c391084a0321c3809632a018b9b657e59d5966289664f85a645942ac/xxhash-3.5.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c8aa771ff2c13dd9cda8166d685d7333d389fae30a4d2bb39d63ab5775de8606", size = 191156 },
-    { url = "https://files.pythonhosted.org/packages/86/2b/915049db13401792fec159f57e4f4a5ca7a9768e83ef71d6645b9d0cd749/xxhash-3.5.0-cp39-cp39-win32.whl", hash = "sha256:5ed9ebc46f24cf91034544b26b131241b699edbfc99ec5e7f8f3d02d6eb7fba4", size = 30122 },
-    { url = "https://files.pythonhosted.org/packages/d5/87/382ef7b24917d7cf4c540ee30f29b283bc87ac5893d2f89b23ea3cdf7d77/xxhash-3.5.0-cp39-cp39-win_amd64.whl", hash = "sha256:220f3f896c6b8d0316f63f16c077d52c412619e475f9372333474ee15133a558", size = 30021 },
-    { url = "https://files.pythonhosted.org/packages/e2/47/d06b24e2d9c3dcabccfd734d11b5bbebfdf59ceac2c61509d8205dd20ac6/xxhash-3.5.0-cp39-cp39-win_arm64.whl", hash = "sha256:a7b1d8315d9b5e9f89eb2933b73afae6ec9597a258d52190944437158b49d38e", size = 26780 },
-    { url = "https://files.pythonhosted.org/packages/ab/9a/233606bada5bd6f50b2b72c45de3d9868ad551e83893d2ac86dc7bb8553a/xxhash-3.5.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:2014c5b3ff15e64feecb6b713af12093f75b7926049e26a580e94dcad3c73d8c", size = 29732 },
-    { url = "https://files.pythonhosted.org/packages/0c/67/f75276ca39e2c6604e3bee6c84e9db8a56a4973fde9bf35989787cf6e8aa/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fab81ef75003eda96239a23eda4e4543cedc22e34c373edcaf744e721a163986", size = 36214 },
-    { url = "https://files.pythonhosted.org/packages/0f/f8/f6c61fd794229cc3848d144f73754a0c107854372d7261419dcbbd286299/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4e2febf914ace002132aa09169cc572e0d8959d0f305f93d5828c4836f9bc5a6", size = 32020 },
-    { url = "https://files.pythonhosted.org/packages/79/d3/c029c99801526f859e6b38d34ab87c08993bf3dcea34b11275775001638a/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5d3a10609c51da2a1c0ea0293fc3968ca0a18bd73838455b5bca3069d7f8e32b", size = 40515 },
-    { url = "https://files.pythonhosted.org/packages/62/e3/bef7b82c1997579c94de9ac5ea7626d01ae5858aa22bf4fcb38bf220cb3e/xxhash-3.5.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:5a74f23335b9689b66eb6dbe2a931a88fcd7a4c2cc4b1cb0edba8ce381c7a1da", size = 30064 },
-    { url = "https://files.pythonhosted.org/packages/c2/56/30d3df421814947f9d782b20c9b7e5e957f3791cbd89874578011daafcbd/xxhash-3.5.0-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:531af8845aaadcadf951b7e0c1345c6b9c68a990eeb74ff9acd8501a0ad6a1c9", size = 29734 },
-    { url = "https://files.pythonhosted.org/packages/82/dd/3c42a1f022ad0d82c852d3cb65493ebac03dcfa8c994465a5fb052b00e3c/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7ce379bcaa9fcc00f19affa7773084dd09f5b59947b3fb47a1ceb0179f91aaa1", size = 36216 },
-    { url = "https://files.pythonhosted.org/packages/b2/40/8f902ab3bebda228a9b4de69eba988280285a7f7f167b942bc20bb562df9/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1b2281d01723f076df3c8188f43f2472248a6b63118b036e641243656b1b0f", size = 32042 },
-    { url = "https://files.pythonhosted.org/packages/db/87/bd06beb8ccaa0e9e577c9b909a49cfa5c5cd2ca46034342d72dd9ce5bc56/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9c770750cc80e8694492244bca7251385188bc5597b6a39d98a9f30e8da984e0", size = 40516 },
-    { url = "https://files.pythonhosted.org/packages/bb/f8/505385e2fbd753ddcaafd5550eabe86f6232cbebabad3b2508d411b19153/xxhash-3.5.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:b150b8467852e1bd844387459aa6fbe11d7f38b56e901f9f3b3e6aba0d660240", size = 30108 },
+sdist = { url = "https://files.pythonhosted.org/packages/00/5e/d6e5258d69df8b4ed8c83b6664f2b47d30d2dec551a29ad72a6c69eafd31/xxhash-3.5.0.tar.gz", hash = "sha256:84f2caddf951c9cbf8dc2e22a89d4ccf5d86391ac6418fe81e3c67d0cf60b45f", size = 84241, upload-time = "2024-08-17T09:20:38.972Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/bb/8a/0e9feca390d512d293afd844d31670e25608c4a901e10202aa98785eab09/xxhash-3.5.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:ece616532c499ee9afbb83078b1b952beffef121d989841f7f4b3dc5ac0fd212", size = 31970, upload-time = "2024-08-17T09:17:35.675Z" },
+    { url = "https://files.pythonhosted.org/packages/16/e6/be5aa49580cd064a18200ab78e29b88b1127e1a8c7955eb8ecf81f2626eb/xxhash-3.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:3171f693dbc2cef6477054a665dc255d996646b4023fe56cb4db80e26f4cc520", size = 30801, upload-time = "2024-08-17T09:17:37.353Z" },
+    { url = "https://files.pythonhosted.org/packages/20/ee/b8a99ebbc6d1113b3a3f09e747fa318c3cde5b04bd9c197688fadf0eeae8/xxhash-3.5.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7c5d3e570ef46adaf93fc81b44aca6002b5a4d8ca11bd0580c07eac537f36680", size = 220927, upload-time = "2024-08-17T09:17:38.835Z" },
+    { url = "https://files.pythonhosted.org/packages/58/62/15d10582ef159283a5c2b47f6d799fc3303fe3911d5bb0bcc820e1ef7ff4/xxhash-3.5.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7cb29a034301e2982df8b1fe6328a84f4b676106a13e9135a0d7e0c3e9f806da", size = 200360, upload-time = "2024-08-17T09:17:40.851Z" },
+    { url = "https://files.pythonhosted.org/packages/23/41/61202663ea9b1bd8e53673b8ec9e2619989353dba8cfb68e59a9cbd9ffe3/xxhash-3.5.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5d0d307d27099bb0cbeea7260eb39ed4fdb99c5542e21e94bb6fd29e49c57a23", size = 428528, upload-time = "2024-08-17T09:17:42.545Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/07/d9a3059f702dec5b3b703737afb6dda32f304f6e9da181a229dafd052c29/xxhash-3.5.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0342aafd421795d740e514bc9858ebddfc705a75a8c5046ac56d85fe97bf196", size = 194149, upload-time = "2024-08-17T09:17:44.361Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/58/27caadf78226ecf1d62dbd0c01d152ed381c14c1ee4ad01f0d460fc40eac/xxhash-3.5.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3dbbd9892c5ebffeca1ed620cf0ade13eb55a0d8c84e0751a6653adc6ac40d0c", size = 207703, upload-time = "2024-08-17T09:17:46.656Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/08/32d558ce23e1e068453c39aed7b3c1cdc690c177873ec0ca3a90d5808765/xxhash-3.5.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4cc2d67fdb4d057730c75a64c5923abfa17775ae234a71b0200346bfb0a7f482", size = 216255, upload-time = "2024-08-17T09:17:48.031Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/d4/2b971e2d2b0a61045f842b622ef11e94096cf1f12cd448b6fd426e80e0e2/xxhash-3.5.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:ec28adb204b759306a3d64358a5e5c07d7b1dd0ccbce04aa76cb9377b7b70296", size = 202744, upload-time = "2024-08-17T09:17:50.045Z" },
+    { url = "https://files.pythonhosted.org/packages/19/ae/6a6438864a8c4c39915d7b65effd85392ebe22710412902487e51769146d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:1328f6d8cca2b86acb14104e381225a3d7b42c92c4b86ceae814e5c400dbb415", size = 210115, upload-time = "2024-08-17T09:17:51.834Z" },
+    { url = "https://files.pythonhosted.org/packages/48/7d/b3c27c27d1fc868094d02fe4498ccce8cec9fcc591825c01d6bcb0b4fc49/xxhash-3.5.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:8d47ebd9f5d9607fd039c1fbf4994e3b071ea23eff42f4ecef246ab2b7334198", size = 414247, upload-time = "2024-08-17T09:17:53.094Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/05/918f9e7d2fbbd334b829997045d341d6239b563c44e683b9a7ef8fe50f5d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b96d559e0fcddd3343c510a0fe2b127fbff16bf346dd76280b82292567523442", size = 191419, upload-time = "2024-08-17T09:17:54.906Z" },
+    { url = "https://files.pythonhosted.org/packages/08/29/dfe393805b2f86bfc47c290b275f0b7c189dc2f4e136fd4754f32eb18a8d/xxhash-3.5.0-cp310-cp310-win32.whl", hash = "sha256:61c722ed8d49ac9bc26c7071eeaa1f6ff24053d553146d5df031802deffd03da", size = 30114, upload-time = "2024-08-17T09:17:56.566Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/d7/aa0b22c4ebb7c3ccb993d4c565132abc641cd11164f8952d89eb6a501909/xxhash-3.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:9bed5144c6923cc902cd14bb8963f2d5e034def4486ab0bbe1f58f03f042f9a9", size = 30003, upload-time = "2024-08-17T09:17:57.596Z" },
+    { url = "https://files.pythonhosted.org/packages/69/12/f969b81541ee91b55f1ce469d7ab55079593c80d04fd01691b550e535000/xxhash-3.5.0-cp310-cp310-win_arm64.whl", hash = "sha256:893074d651cf25c1cc14e3bea4fceefd67f2921b1bb8e40fcfeba56820de80c6", size = 26773, upload-time = "2024-08-17T09:17:59.169Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/c7/afed0f131fbda960ff15eee7f304fa0eeb2d58770fade99897984852ef23/xxhash-3.5.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:02c2e816896dc6f85922ced60097bcf6f008dedfc5073dcba32f9c8dd786f3c1", size = 31969, upload-time = "2024-08-17T09:18:00.852Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/0c/7c3bc6d87e5235672fcc2fb42fd5ad79fe1033925f71bf549ee068c7d1ca/xxhash-3.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6027dcd885e21581e46d3c7f682cfb2b870942feeed58a21c29583512c3f09f8", size = 30800, upload-time = "2024-08-17T09:18:01.863Z" },
+    { url = "https://files.pythonhosted.org/packages/04/9e/01067981d98069eec1c20201f8c145367698e9056f8bc295346e4ea32dd1/xxhash-3.5.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1308fa542bbdbf2fa85e9e66b1077eea3a88bef38ee8a06270b4298a7a62a166", size = 221566, upload-time = "2024-08-17T09:18:03.461Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/09/d4996de4059c3ce5342b6e1e6a77c9d6c91acce31f6ed979891872dd162b/xxhash-3.5.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c28b2fdcee797e1c1961cd3bcd3d545cab22ad202c846235197935e1df2f8ef7", size = 201214, upload-time = "2024-08-17T09:18:05.616Z" },
+    { url = "https://files.pythonhosted.org/packages/62/f5/6d2dc9f8d55a7ce0f5e7bfef916e67536f01b85d32a9fbf137d4cadbee38/xxhash-3.5.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:924361811732ddad75ff23e90efd9ccfda4f664132feecb90895bade6a1b4623", size = 429433, upload-time = "2024-08-17T09:18:06.957Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/72/9256303f10e41ab004799a4aa74b80b3c5977d6383ae4550548b24bd1971/xxhash-3.5.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:89997aa1c4b6a5b1e5b588979d1da048a3c6f15e55c11d117a56b75c84531f5a", size = 194822, upload-time = "2024-08-17T09:18:08.331Z" },
+    { url = "https://files.pythonhosted.org/packages/34/92/1a3a29acd08248a34b0e6a94f4e0ed9b8379a4ff471f1668e4dce7bdbaa8/xxhash-3.5.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:685c4f4e8c59837de103344eb1c8a3851f670309eb5c361f746805c5471b8c88", size = 208538, upload-time = "2024-08-17T09:18:10.332Z" },
+    { url = "https://files.pythonhosted.org/packages/53/ad/7fa1a109663366de42f724a1cdb8e796a260dbac45047bce153bc1e18abf/xxhash-3.5.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dbd2ecfbfee70bc1a4acb7461fa6af7748ec2ab08ac0fa298f281c51518f982c", size = 216953, upload-time = "2024-08-17T09:18:11.707Z" },
+    { url = "https://files.pythonhosted.org/packages/35/02/137300e24203bf2b2a49b48ce898ecce6fd01789c0fcd9c686c0a002d129/xxhash-3.5.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:25b5a51dc3dfb20a10833c8eee25903fd2e14059e9afcd329c9da20609a307b2", size = 203594, upload-time = "2024-08-17T09:18:13.799Z" },
+    { url = "https://files.pythonhosted.org/packages/23/03/aeceb273933d7eee248c4322b98b8e971f06cc3880e5f7602c94e5578af5/xxhash-3.5.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a8fb786fb754ef6ff8c120cb96629fb518f8eb5a61a16aac3a979a9dbd40a084", size = 210971, upload-time = "2024-08-17T09:18:15.824Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/64/ed82ec09489474cbb35c716b189ddc1521d8b3de12b1b5ab41ce7f70253c/xxhash-3.5.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:a905ad00ad1e1c34fe4e9d7c1d949ab09c6fa90c919860c1534ff479f40fd12d", size = 415050, upload-time = "2024-08-17T09:18:17.142Z" },
+    { url = "https://files.pythonhosted.org/packages/71/43/6db4c02dcb488ad4e03bc86d70506c3d40a384ee73c9b5c93338eb1f3c23/xxhash-3.5.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:963be41bcd49f53af6d795f65c0da9b4cc518c0dd9c47145c98f61cb464f4839", size = 192216, upload-time = "2024-08-17T09:18:18.779Z" },
+    { url = "https://files.pythonhosted.org/packages/22/6d/db4abec29e7a567455344433d095fdb39c97db6955bb4a2c432e486b4d28/xxhash-3.5.0-cp311-cp311-win32.whl", hash = "sha256:109b436096d0a2dd039c355fa3414160ec4d843dfecc64a14077332a00aeb7da", size = 30120, upload-time = "2024-08-17T09:18:20.009Z" },
+    { url = "https://files.pythonhosted.org/packages/52/1c/fa3b61c0cf03e1da4767213672efe186b1dfa4fc901a4a694fb184a513d1/xxhash-3.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:b702f806693201ad6c0a05ddbbe4c8f359626d0b3305f766077d51388a6bac58", size = 30003, upload-time = "2024-08-17T09:18:21.052Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/8e/9e6fc572acf6e1cc7ccb01973c213f895cb8668a9d4c2b58a99350da14b7/xxhash-3.5.0-cp311-cp311-win_arm64.whl", hash = "sha256:c4dcb4120d0cc3cc448624147dba64e9021b278c63e34a38789b688fd0da9bf3", size = 26777, upload-time = "2024-08-17T09:18:22.809Z" },
+    { url = "https://files.pythonhosted.org/packages/07/0e/1bfce2502c57d7e2e787600b31c83535af83746885aa1a5f153d8c8059d6/xxhash-3.5.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:14470ace8bd3b5d51318782cd94e6f94431974f16cb3b8dc15d52f3b69df8e00", size = 31969, upload-time = "2024-08-17T09:18:24.025Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/d6/8ca450d6fe5b71ce521b4e5db69622383d039e2b253e9b2f24f93265b52c/xxhash-3.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:59aa1203de1cb96dbeab595ded0ad0c0056bb2245ae11fac11c0ceea861382b9", size = 30787, upload-time = "2024-08-17T09:18:25.318Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/84/de7c89bc6ef63d750159086a6ada6416cc4349eab23f76ab870407178b93/xxhash-3.5.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:08424f6648526076e28fae6ea2806c0a7d504b9ef05ae61d196d571e5c879c84", size = 220959, upload-time = "2024-08-17T09:18:26.518Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/86/51258d3e8a8545ff26468c977101964c14d56a8a37f5835bc0082426c672/xxhash-3.5.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:61a1ff00674879725b194695e17f23d3248998b843eb5e933007ca743310f793", size = 200006, upload-time = "2024-08-17T09:18:27.905Z" },
+    { url = "https://files.pythonhosted.org/packages/02/0a/96973bd325412feccf23cf3680fd2246aebf4b789122f938d5557c54a6b2/xxhash-3.5.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f2f2c61bee5844d41c3eb015ac652a0229e901074951ae48581d58bfb2ba01be", size = 428326, upload-time = "2024-08-17T09:18:29.335Z" },
+    { url = "https://files.pythonhosted.org/packages/11/a7/81dba5010f7e733de88af9555725146fc133be97ce36533867f4c7e75066/xxhash-3.5.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9d32a592cac88d18cc09a89172e1c32d7f2a6e516c3dfde1b9adb90ab5df54a6", size = 194380, upload-time = "2024-08-17T09:18:30.706Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/7d/f29006ab398a173f4501c0e4977ba288f1c621d878ec217b4ff516810c04/xxhash-3.5.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:70dabf941dede727cca579e8c205e61121afc9b28516752fd65724be1355cc90", size = 207934, upload-time = "2024-08-17T09:18:32.133Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/6e/6e88b8f24612510e73d4d70d9b0c7dff62a2e78451b9f0d042a5462c8d03/xxhash-3.5.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e5d0ddaca65ecca9c10dcf01730165fd858533d0be84c75c327487c37a906a27", size = 216301, upload-time = "2024-08-17T09:18:33.474Z" },
+    { url = "https://files.pythonhosted.org/packages/af/51/7862f4fa4b75a25c3b4163c8a873f070532fe5f2d3f9b3fc869c8337a398/xxhash-3.5.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:3e5b5e16c5a480fe5f59f56c30abdeba09ffd75da8d13f6b9b6fd224d0b4d0a2", size = 203351, upload-time = "2024-08-17T09:18:34.889Z" },
+    { url = "https://files.pythonhosted.org/packages/22/61/8d6a40f288f791cf79ed5bb113159abf0c81d6efb86e734334f698eb4c59/xxhash-3.5.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:149b7914451eb154b3dfaa721315117ea1dac2cc55a01bfbd4df7c68c5dd683d", size = 210294, upload-time = "2024-08-17T09:18:36.355Z" },
+    { url = "https://files.pythonhosted.org/packages/17/02/215c4698955762d45a8158117190261b2dbefe9ae7e5b906768c09d8bc74/xxhash-3.5.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:eade977f5c96c677035ff39c56ac74d851b1cca7d607ab3d8f23c6b859379cab", size = 414674, upload-time = "2024-08-17T09:18:38.536Z" },
+    { url = "https://files.pythonhosted.org/packages/31/5c/b7a8db8a3237cff3d535261325d95de509f6a8ae439a5a7a4ffcff478189/xxhash-3.5.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:fa9f547bd98f5553d03160967866a71056a60960be00356a15ecc44efb40ba8e", size = 192022, upload-time = "2024-08-17T09:18:40.138Z" },
+    { url = "https://files.pythonhosted.org/packages/78/e3/dd76659b2811b3fd06892a8beb850e1996b63e9235af5a86ea348f053e9e/xxhash-3.5.0-cp312-cp312-win32.whl", hash = "sha256:f7b58d1fd3551b8c80a971199543379be1cee3d0d409e1f6d8b01c1a2eebf1f8", size = 30170, upload-time = "2024-08-17T09:18:42.163Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/6b/1c443fe6cfeb4ad1dcf231cdec96eb94fb43d6498b4469ed8b51f8b59a37/xxhash-3.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:fa0cafd3a2af231b4e113fba24a65d7922af91aeb23774a8b78228e6cd785e3e", size = 30040, upload-time = "2024-08-17T09:18:43.699Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/eb/04405305f290173acc0350eba6d2f1a794b57925df0398861a20fbafa415/xxhash-3.5.0-cp312-cp312-win_arm64.whl", hash = "sha256:586886c7e89cb9828bcd8a5686b12e161368e0064d040e225e72607b43858ba2", size = 26796, upload-time = "2024-08-17T09:18:45.29Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/f6/531dd6858adf8877675270b9d6989b6dacfd1c2d7135b17584fc29866df3/xxhash-3.5.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bfc8cdd7f33d57f0468b0614ae634cc38ab9202c6957a60e31d285a71ebe0301", size = 31971, upload-time = "2024-08-17T09:19:47.447Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/a8/b2a42b6c9ae46e233f474f3d307c2e7bca8d9817650babeca048d2ad01d6/xxhash-3.5.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e0c48b6300cd0b0106bf49169c3e0536408dfbeb1ccb53180068a18b03c662ab", size = 30801, upload-time = "2024-08-17T09:19:48.911Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/92/9ac297e3487818f429bcf369c1c6a097edf5b56ed6fc1feff4c1882e87ef/xxhash-3.5.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fe1a92cfbaa0a1253e339ccec42dbe6db262615e52df591b68726ab10338003f", size = 220644, upload-time = "2024-08-17T09:19:51.081Z" },
+    { url = "https://files.pythonhosted.org/packages/86/48/c1426dd3c86fc4a52f983301867463472f6a9013fb32d15991e60c9919b6/xxhash-3.5.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:33513d6cc3ed3b559134fb307aae9bdd94d7e7c02907b37896a6c45ff9ce51bd", size = 200021, upload-time = "2024-08-17T09:19:52.923Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/de/0ab8c79993765c94fc0d0c1a22b454483c58a0161e1b562f58b654f47660/xxhash-3.5.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eefc37f6138f522e771ac6db71a6d4838ec7933939676f3753eafd7d3f4c40bc", size = 428217, upload-time = "2024-08-17T09:19:54.349Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/b4/332647451ed7d2c021294b7c1e9c144dbb5586b1fb214ad4f5a404642835/xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a606c8070ada8aa2a88e181773fa1ef17ba65ce5dd168b9d08038e2a61b33754", size = 193868, upload-time = "2024-08-17T09:19:55.763Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/1c/a42c0a6cac752f84f7b44a90d1a9fa9047cf70bdba5198a304fde7cc471f/xxhash-3.5.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:42eca420c8fa072cc1dd62597635d140e78e384a79bb4944f825fbef8bfeeef6", size = 207403, upload-time = "2024-08-17T09:19:57.945Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/d7/04e1b0daae9dc9b02c73c1664cc8aa527498c3f66ccbc586eeb25bbe9f14/xxhash-3.5.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:604253b2143e13218ff1ef0b59ce67f18b8bd1c4205d2ffda22b09b426386898", size = 215978, upload-time = "2024-08-17T09:19:59.381Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/f4/05e15e67505228fc19ee98a79e427b3a0b9695f5567cd66ced5d66389883/xxhash-3.5.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6e93a5ad22f434d7876665444a97e713a8f60b5b1a3521e8df11b98309bff833", size = 202416, upload-time = "2024-08-17T09:20:01.534Z" },
+    { url = "https://files.pythonhosted.org/packages/94/fb/e9028d3645bba5412a09de13ee36df276a567e60bdb31d499dafa46d76ae/xxhash-3.5.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:7a46e1d6d2817ba8024de44c4fd79913a90e5f7265434cef97026215b7d30df6", size = 209853, upload-time = "2024-08-17T09:20:03.376Z" },
+    { url = "https://files.pythonhosted.org/packages/02/2c/18c6a622429368274739372d2f86c8125413ec169025c7d8ffb051784bba/xxhash-3.5.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:30eb2efe6503c379b7ab99c81ba4a779748e3830241f032ab46bd182bf5873af", size = 413926, upload-time = "2024-08-17T09:20:04.946Z" },
+    { url = "https://files.pythonhosted.org/packages/72/bb/5b55c391084a0321c3809632a018b9b657e59d5966289664f85a645942ac/xxhash-3.5.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c8aa771ff2c13dd9cda8166d685d7333d389fae30a4d2bb39d63ab5775de8606", size = 191156, upload-time = "2024-08-17T09:20:06.318Z" },
+    { url = "https://files.pythonhosted.org/packages/86/2b/915049db13401792fec159f57e4f4a5ca7a9768e83ef71d6645b9d0cd749/xxhash-3.5.0-cp39-cp39-win32.whl", hash = "sha256:5ed9ebc46f24cf91034544b26b131241b699edbfc99ec5e7f8f3d02d6eb7fba4", size = 30122, upload-time = "2024-08-17T09:20:07.691Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/87/382ef7b24917d7cf4c540ee30f29b283bc87ac5893d2f89b23ea3cdf7d77/xxhash-3.5.0-cp39-cp39-win_amd64.whl", hash = "sha256:220f3f896c6b8d0316f63f16c077d52c412619e475f9372333474ee15133a558", size = 30021, upload-time = "2024-08-17T09:20:08.832Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/47/d06b24e2d9c3dcabccfd734d11b5bbebfdf59ceac2c61509d8205dd20ac6/xxhash-3.5.0-cp39-cp39-win_arm64.whl", hash = "sha256:a7b1d8315d9b5e9f89eb2933b73afae6ec9597a258d52190944437158b49d38e", size = 26780, upload-time = "2024-08-17T09:20:09.989Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/9a/233606bada5bd6f50b2b72c45de3d9868ad551e83893d2ac86dc7bb8553a/xxhash-3.5.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:2014c5b3ff15e64feecb6b713af12093f75b7926049e26a580e94dcad3c73d8c", size = 29732, upload-time = "2024-08-17T09:20:11.175Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/67/f75276ca39e2c6604e3bee6c84e9db8a56a4973fde9bf35989787cf6e8aa/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fab81ef75003eda96239a23eda4e4543cedc22e34c373edcaf744e721a163986", size = 36214, upload-time = "2024-08-17T09:20:12.335Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/f8/f6c61fd794229cc3848d144f73754a0c107854372d7261419dcbbd286299/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4e2febf914ace002132aa09169cc572e0d8959d0f305f93d5828c4836f9bc5a6", size = 32020, upload-time = "2024-08-17T09:20:13.537Z" },
+    { url = "https://files.pythonhosted.org/packages/79/d3/c029c99801526f859e6b38d34ab87c08993bf3dcea34b11275775001638a/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5d3a10609c51da2a1c0ea0293fc3968ca0a18bd73838455b5bca3069d7f8e32b", size = 40515, upload-time = "2024-08-17T09:20:14.669Z" },
+    { url = "https://files.pythonhosted.org/packages/62/e3/bef7b82c1997579c94de9ac5ea7626d01ae5858aa22bf4fcb38bf220cb3e/xxhash-3.5.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:5a74f23335b9689b66eb6dbe2a931a88fcd7a4c2cc4b1cb0edba8ce381c7a1da", size = 30064, upload-time = "2024-08-17T09:20:15.925Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/56/30d3df421814947f9d782b20c9b7e5e957f3791cbd89874578011daafcbd/xxhash-3.5.0-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:531af8845aaadcadf951b7e0c1345c6b9c68a990eeb74ff9acd8501a0ad6a1c9", size = 29734, upload-time = "2024-08-17T09:20:30.457Z" },
+    { url = "https://files.pythonhosted.org/packages/82/dd/3c42a1f022ad0d82c852d3cb65493ebac03dcfa8c994465a5fb052b00e3c/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7ce379bcaa9fcc00f19affa7773084dd09f5b59947b3fb47a1ceb0179f91aaa1", size = 36216, upload-time = "2024-08-17T09:20:32.116Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/40/8f902ab3bebda228a9b4de69eba988280285a7f7f167b942bc20bb562df9/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1b2281d01723f076df3c8188f43f2472248a6b63118b036e641243656b1b0f", size = 32042, upload-time = "2024-08-17T09:20:33.562Z" },
+    { url = "https://files.pythonhosted.org/packages/db/87/bd06beb8ccaa0e9e577c9b909a49cfa5c5cd2ca46034342d72dd9ce5bc56/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9c770750cc80e8694492244bca7251385188bc5597b6a39d98a9f30e8da984e0", size = 40516, upload-time = "2024-08-17T09:20:36.004Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/f8/505385e2fbd753ddcaafd5550eabe86f6232cbebabad3b2508d411b19153/xxhash-3.5.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:b150b8467852e1bd844387459aa6fbe11d7f38b56e901f9f3b3e6aba0d660240", size = 30108, upload-time = "2024-08-17T09:20:37.214Z" },
 ]
 
 [[package]]
 name = "yarl"
-version = "1.18.3"
+version = "1.20.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "idna" },
     { name = "multidict" },
     { name = "propcache" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b7/9d/4b94a8e6d2b51b599516a5cb88e5bc99b4d8d4583e468057eaa29d5f0918/yarl-1.18.3.tar.gz", hash = "sha256:ac1801c45cbf77b6c99242eeff4fffb5e4e73a800b5c4ad4fc0be5def634d2e1", size = 181062 }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/d2/98/e005bc608765a8a5569f58e650961314873c8469c333616eb40bff19ae97/yarl-1.18.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7df647e8edd71f000a5208fe6ff8c382a1de8edfbccdbbfe649d263de07d8c34", size = 141458 },
-    { url = "https://files.pythonhosted.org/packages/df/5d/f8106b263b8ae8a866b46d9be869ac01f9b3fb7f2325f3ecb3df8003f796/yarl-1.18.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:c69697d3adff5aa4f874b19c0e4ed65180ceed6318ec856ebc423aa5850d84f7", size = 94365 },
-    { url = "https://files.pythonhosted.org/packages/56/3e/d8637ddb9ba69bf851f765a3ee288676f7cf64fb3be13760c18cbc9d10bd/yarl-1.18.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:602d98f2c2d929f8e697ed274fbadc09902c4025c5a9963bf4e9edfc3ab6f7ed", size = 92181 },
-    { url = "https://files.pythonhosted.org/packages/76/f9/d616a5c2daae281171de10fba41e1c0e2d8207166fc3547252f7d469b4e1/yarl-1.18.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c654d5207c78e0bd6d749f6dae1dcbbfde3403ad3a4b11f3c5544d9906969dde", size = 315349 },
-    { url = "https://files.pythonhosted.org/packages/bb/b4/3ea5e7b6f08f698b3769a06054783e434f6d59857181b5c4e145de83f59b/yarl-1.18.3-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5094d9206c64181d0f6e76ebd8fb2f8fe274950a63890ee9e0ebfd58bf9d787b", size = 330494 },
-    { url = "https://files.pythonhosted.org/packages/55/f1/e0fc810554877b1b67420568afff51b967baed5b53bcc983ab164eebf9c9/yarl-1.18.3-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:35098b24e0327fc4ebdc8ffe336cee0a87a700c24ffed13161af80124b7dc8e5", size = 326927 },
-    { url = "https://files.pythonhosted.org/packages/a9/42/b1753949b327b36f210899f2dd0a0947c0c74e42a32de3f8eb5c7d93edca/yarl-1.18.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3236da9272872443f81fedc389bace88408f64f89f75d1bdb2256069a8730ccc", size = 319703 },
-    { url = "https://files.pythonhosted.org/packages/f0/6d/e87c62dc9635daefb064b56f5c97df55a2e9cc947a2b3afd4fd2f3b841c7/yarl-1.18.3-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e2c08cc9b16f4f4bc522771d96734c7901e7ebef70c6c5c35dd0f10845270bcd", size = 310246 },
-    { url = "https://files.pythonhosted.org/packages/e3/ef/e2e8d1785cdcbd986f7622d7f0098205f3644546da7919c24b95790ec65a/yarl-1.18.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:80316a8bd5109320d38eef8833ccf5f89608c9107d02d2a7f985f98ed6876990", size = 319730 },
-    { url = "https://files.pythonhosted.org/packages/fc/15/8723e22345bc160dfde68c4b3ae8b236e868f9963c74015f1bc8a614101c/yarl-1.18.3-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:c1e1cc06da1491e6734f0ea1e6294ce00792193c463350626571c287c9a704db", size = 321681 },
-    { url = "https://files.pythonhosted.org/packages/86/09/bf764e974f1516efa0ae2801494a5951e959f1610dd41edbfc07e5e0f978/yarl-1.18.3-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fea09ca13323376a2fdfb353a5fa2e59f90cd18d7ca4eaa1fd31f0a8b4f91e62", size = 324812 },
-    { url = "https://files.pythonhosted.org/packages/f6/4c/20a0187e3b903c97d857cf0272d687c1b08b03438968ae8ffc50fe78b0d6/yarl-1.18.3-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:e3b9fd71836999aad54084906f8663dffcd2a7fb5cdafd6c37713b2e72be1760", size = 337011 },
-    { url = "https://files.pythonhosted.org/packages/c9/71/6244599a6e1cc4c9f73254a627234e0dad3883ece40cc33dce6265977461/yarl-1.18.3-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:757e81cae69244257d125ff31663249b3013b5dc0a8520d73694aed497fb195b", size = 338132 },
-    { url = "https://files.pythonhosted.org/packages/af/f5/e0c3efaf74566c4b4a41cb76d27097df424052a064216beccae8d303c90f/yarl-1.18.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b1771de9944d875f1b98a745bc547e684b863abf8f8287da8466cf470ef52690", size = 331849 },
-    { url = "https://files.pythonhosted.org/packages/8a/b8/3d16209c2014c2f98a8f658850a57b716efb97930aebf1ca0d9325933731/yarl-1.18.3-cp310-cp310-win32.whl", hash = "sha256:8874027a53e3aea659a6d62751800cf6e63314c160fd607489ba5c2edd753cf6", size = 84309 },
-    { url = "https://files.pythonhosted.org/packages/fd/b7/2e9a5b18eb0fe24c3a0e8bae994e812ed9852ab4fd067c0107fadde0d5f0/yarl-1.18.3-cp310-cp310-win_amd64.whl", hash = "sha256:93b2e109287f93db79210f86deb6b9bbb81ac32fc97236b16f7433db7fc437d8", size = 90484 },
-    { url = "https://files.pythonhosted.org/packages/40/93/282b5f4898d8e8efaf0790ba6d10e2245d2c9f30e199d1a85cae9356098c/yarl-1.18.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:8503ad47387b8ebd39cbbbdf0bf113e17330ffd339ba1144074da24c545f0069", size = 141555 },
-    { url = "https://files.pythonhosted.org/packages/6d/9c/0a49af78df099c283ca3444560f10718fadb8a18dc8b3edf8c7bd9fd7d89/yarl-1.18.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:02ddb6756f8f4517a2d5e99d8b2f272488e18dd0bfbc802f31c16c6c20f22193", size = 94351 },
-    { url = "https://files.pythonhosted.org/packages/5a/a1/205ab51e148fdcedad189ca8dd587794c6f119882437d04c33c01a75dece/yarl-1.18.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:67a283dd2882ac98cc6318384f565bffc751ab564605959df4752d42483ad889", size = 92286 },
-    { url = "https://files.pythonhosted.org/packages/ed/fe/88b690b30f3f59275fb674f5f93ddd4a3ae796c2b62e5bb9ece8a4914b83/yarl-1.18.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d980e0325b6eddc81331d3f4551e2a333999fb176fd153e075c6d1c2530aa8a8", size = 340649 },
-    { url = "https://files.pythonhosted.org/packages/07/eb/3b65499b568e01f36e847cebdc8d7ccb51fff716dbda1ae83c3cbb8ca1c9/yarl-1.18.3-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b643562c12680b01e17239be267bc306bbc6aac1f34f6444d1bded0c5ce438ca", size = 356623 },
-    { url = "https://files.pythonhosted.org/packages/33/46/f559dc184280b745fc76ec6b1954de2c55595f0ec0a7614238b9ebf69618/yarl-1.18.3-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c017a3b6df3a1bd45b9fa49a0f54005e53fbcad16633870104b66fa1a30a29d8", size = 354007 },
-    { url = "https://files.pythonhosted.org/packages/af/ba/1865d85212351ad160f19fb99808acf23aab9a0f8ff31c8c9f1b4d671fc9/yarl-1.18.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:75674776d96d7b851b6498f17824ba17849d790a44d282929c42dbb77d4f17ae", size = 344145 },
-    { url = "https://files.pythonhosted.org/packages/94/cb/5c3e975d77755d7b3d5193e92056b19d83752ea2da7ab394e22260a7b824/yarl-1.18.3-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ccaa3a4b521b780a7e771cc336a2dba389a0861592bbce09a476190bb0c8b4b3", size = 336133 },
-    { url = "https://files.pythonhosted.org/packages/19/89/b77d3fd249ab52a5c40859815765d35c91425b6bb82e7427ab2f78f5ff55/yarl-1.18.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:2d06d3005e668744e11ed80812e61efd77d70bb7f03e33c1598c301eea20efbb", size = 347967 },
-    { url = "https://files.pythonhosted.org/packages/35/bd/f6b7630ba2cc06c319c3235634c582a6ab014d52311e7d7c22f9518189b5/yarl-1.18.3-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:9d41beda9dc97ca9ab0b9888cb71f7539124bc05df02c0cff6e5acc5a19dcc6e", size = 346397 },
-    { url = "https://files.pythonhosted.org/packages/18/1a/0b4e367d5a72d1f095318344848e93ea70da728118221f84f1bf6c1e39e7/yarl-1.18.3-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:ba23302c0c61a9999784e73809427c9dbedd79f66a13d84ad1b1943802eaaf59", size = 350206 },
-    { url = "https://files.pythonhosted.org/packages/b5/cf/320fff4367341fb77809a2d8d7fe75b5d323a8e1b35710aafe41fdbf327b/yarl-1.18.3-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:6748dbf9bfa5ba1afcc7556b71cda0d7ce5f24768043a02a58846e4a443d808d", size = 362089 },
-    { url = "https://files.pythonhosted.org/packages/57/cf/aadba261d8b920253204085268bad5e8cdd86b50162fcb1b10c10834885a/yarl-1.18.3-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:0b0cad37311123211dc91eadcb322ef4d4a66008d3e1bdc404808992260e1a0e", size = 366267 },
-    { url = "https://files.pythonhosted.org/packages/54/58/fb4cadd81acdee6dafe14abeb258f876e4dd410518099ae9a35c88d8097c/yarl-1.18.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0fb2171a4486bb075316ee754c6d8382ea6eb8b399d4ec62fde2b591f879778a", size = 359141 },
-    { url = "https://files.pythonhosted.org/packages/9a/7a/4c571597589da4cd5c14ed2a0b17ac56ec9ee7ee615013f74653169e702d/yarl-1.18.3-cp311-cp311-win32.whl", hash = "sha256:61b1a825a13bef4a5f10b1885245377d3cd0bf87cba068e1d9a88c2ae36880e1", size = 84402 },
-    { url = "https://files.pythonhosted.org/packages/ae/7b/8600250b3d89b625f1121d897062f629883c2f45339623b69b1747ec65fa/yarl-1.18.3-cp311-cp311-win_amd64.whl", hash = "sha256:b9d60031cf568c627d028239693fd718025719c02c9f55df0a53e587aab951b5", size = 91030 },
-    { url = "https://files.pythonhosted.org/packages/33/85/bd2e2729752ff4c77338e0102914897512e92496375e079ce0150a6dc306/yarl-1.18.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:1dd4bdd05407ced96fed3d7f25dbbf88d2ffb045a0db60dbc247f5b3c5c25d50", size = 142644 },
-    { url = "https://files.pythonhosted.org/packages/ff/74/1178322cc0f10288d7eefa6e4a85d8d2e28187ccab13d5b844e8b5d7c88d/yarl-1.18.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:7c33dd1931a95e5d9a772d0ac5e44cac8957eaf58e3c8da8c1414de7dd27c576", size = 94962 },
-    { url = "https://files.pythonhosted.org/packages/be/75/79c6acc0261e2c2ae8a1c41cf12265e91628c8c58ae91f5ff59e29c0787f/yarl-1.18.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:25b411eddcfd56a2f0cd6a384e9f4f7aa3efee14b188de13048c25b5e91f1640", size = 92795 },
-    { url = "https://files.pythonhosted.org/packages/6b/32/927b2d67a412c31199e83fefdce6e645247b4fb164aa1ecb35a0f9eb2058/yarl-1.18.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:436c4fc0a4d66b2badc6c5fc5ef4e47bb10e4fd9bf0c79524ac719a01f3607c2", size = 332368 },
-    { url = "https://files.pythonhosted.org/packages/19/e5/859fca07169d6eceeaa4fde1997c91d8abde4e9a7c018e371640c2da2b71/yarl-1.18.3-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e35ef8683211db69ffe129a25d5634319a677570ab6b2eba4afa860f54eeaf75", size = 342314 },
-    { url = "https://files.pythonhosted.org/packages/08/75/76b63ccd91c9e03ab213ef27ae6add2e3400e77e5cdddf8ed2dbc36e3f21/yarl-1.18.3-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:84b2deecba4a3f1a398df819151eb72d29bfeb3b69abb145a00ddc8d30094512", size = 341987 },
-    { url = "https://files.pythonhosted.org/packages/1a/e1/a097d5755d3ea8479a42856f51d97eeff7a3a7160593332d98f2709b3580/yarl-1.18.3-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:00e5a1fea0fd4f5bfa7440a47eff01d9822a65b4488f7cff83155a0f31a2ecba", size = 336914 },
-    { url = "https://files.pythonhosted.org/packages/0b/42/e1b4d0e396b7987feceebe565286c27bc085bf07d61a59508cdaf2d45e63/yarl-1.18.3-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d0e883008013c0e4aef84dcfe2a0b172c4d23c2669412cf5b3371003941f72bb", size = 325765 },
-    { url = "https://files.pythonhosted.org/packages/7e/18/03a5834ccc9177f97ca1bbb245b93c13e58e8225276f01eedc4cc98ab820/yarl-1.18.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5a3f356548e34a70b0172d8890006c37be92995f62d95a07b4a42e90fba54272", size = 344444 },
-    { url = "https://files.pythonhosted.org/packages/c8/03/a713633bdde0640b0472aa197b5b86e90fbc4c5bc05b727b714cd8a40e6d/yarl-1.18.3-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:ccd17349166b1bee6e529b4add61727d3f55edb7babbe4069b5764c9587a8cc6", size = 340760 },
-    { url = "https://files.pythonhosted.org/packages/eb/99/f6567e3f3bbad8fd101886ea0276c68ecb86a2b58be0f64077396cd4b95e/yarl-1.18.3-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:b958ddd075ddba5b09bb0be8a6d9906d2ce933aee81100db289badbeb966f54e", size = 346484 },
-    { url = "https://files.pythonhosted.org/packages/8e/a9/84717c896b2fc6cb15bd4eecd64e34a2f0a9fd6669e69170c73a8b46795a/yarl-1.18.3-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:c7d79f7d9aabd6011004e33b22bc13056a3e3fb54794d138af57f5ee9d9032cb", size = 359864 },
-    { url = "https://files.pythonhosted.org/packages/1e/2e/d0f5f1bef7ee93ed17e739ec8dbcb47794af891f7d165fa6014517b48169/yarl-1.18.3-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:4891ed92157e5430874dad17b15eb1fda57627710756c27422200c52d8a4e393", size = 364537 },
-    { url = "https://files.pythonhosted.org/packages/97/8a/568d07c5d4964da5b02621a517532adb8ec5ba181ad1687191fffeda0ab6/yarl-1.18.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ce1af883b94304f493698b00d0f006d56aea98aeb49d75ec7d98cd4a777e9285", size = 357861 },
-    { url = "https://files.pythonhosted.org/packages/7d/e3/924c3f64b6b3077889df9a1ece1ed8947e7b61b0a933f2ec93041990a677/yarl-1.18.3-cp312-cp312-win32.whl", hash = "sha256:f91c4803173928a25e1a55b943c81f55b8872f0018be83e3ad4938adffb77dd2", size = 84097 },
-    { url = "https://files.pythonhosted.org/packages/34/45/0e055320daaabfc169b21ff6174567b2c910c45617b0d79c68d7ab349b02/yarl-1.18.3-cp312-cp312-win_amd64.whl", hash = "sha256:7e2ee16578af3b52ac2f334c3b1f92262f47e02cc6193c598502bd46f5cd1477", size = 90399 },
-    { url = "https://files.pythonhosted.org/packages/30/c7/c790513d5328a8390be8f47be5d52e141f78b66c6c48f48d241ca6bd5265/yarl-1.18.3-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:90adb47ad432332d4f0bc28f83a5963f426ce9a1a8809f5e584e704b82685dcb", size = 140789 },
-    { url = "https://files.pythonhosted.org/packages/30/aa/a2f84e93554a578463e2edaaf2300faa61c8701f0898725842c704ba5444/yarl-1.18.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:913829534200eb0f789d45349e55203a091f45c37a2674678744ae52fae23efa", size = 94144 },
-    { url = "https://files.pythonhosted.org/packages/c6/fc/d68d8f83714b221a85ce7866832cba36d7c04a68fa6a960b908c2c84f325/yarl-1.18.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:ef9f7768395923c3039055c14334ba4d926f3baf7b776c923c93d80195624782", size = 91974 },
-    { url = "https://files.pythonhosted.org/packages/56/4e/d2563d8323a7e9a414b5b25341b3942af5902a2263d36d20fb17c40411e2/yarl-1.18.3-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:88a19f62ff30117e706ebc9090b8ecc79aeb77d0b1f5ec10d2d27a12bc9f66d0", size = 333587 },
-    { url = "https://files.pythonhosted.org/packages/25/c9/cfec0bc0cac8d054be223e9f2c7909d3e8442a856af9dbce7e3442a8ec8d/yarl-1.18.3-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e17c9361d46a4d5addf777c6dd5eab0715a7684c2f11b88c67ac37edfba6c482", size = 344386 },
-    { url = "https://files.pythonhosted.org/packages/ab/5d/4c532190113b25f1364d25f4c319322e86232d69175b91f27e3ebc2caf9a/yarl-1.18.3-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1a74a13a4c857a84a845505fd2d68e54826a2cd01935a96efb1e9d86c728e186", size = 345421 },
-    { url = "https://files.pythonhosted.org/packages/23/d1/6cdd1632da013aa6ba18cee4d750d953104a5e7aac44e249d9410a972bf5/yarl-1.18.3-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41f7ce59d6ee7741af71d82020346af364949314ed3d87553763a2df1829cc58", size = 339384 },
-    { url = "https://files.pythonhosted.org/packages/9a/c4/6b3c39bec352e441bd30f432cda6ba51681ab19bb8abe023f0d19777aad1/yarl-1.18.3-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f52a265001d830bc425f82ca9eabda94a64a4d753b07d623a9f2863fde532b53", size = 326689 },
-    { url = "https://files.pythonhosted.org/packages/23/30/07fb088f2eefdc0aa4fc1af4e3ca4eb1a3aadd1ce7d866d74c0f124e6a85/yarl-1.18.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:82123d0c954dc58db301f5021a01854a85bf1f3bb7d12ae0c01afc414a882ca2", size = 345453 },
-    { url = "https://files.pythonhosted.org/packages/63/09/d54befb48f9cd8eec43797f624ec37783a0266855f4930a91e3d5c7717f8/yarl-1.18.3-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:2ec9bbba33b2d00999af4631a3397d1fd78290c48e2a3e52d8dd72db3a067ac8", size = 341872 },
-    { url = "https://files.pythonhosted.org/packages/91/26/fd0ef9bf29dd906a84b59f0cd1281e65b0c3e08c6aa94b57f7d11f593518/yarl-1.18.3-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:fbd6748e8ab9b41171bb95c6142faf068f5ef1511935a0aa07025438dd9a9bc1", size = 347497 },
-    { url = "https://files.pythonhosted.org/packages/d9/b5/14ac7a256d0511b2ac168d50d4b7d744aea1c1aa20c79f620d1059aab8b2/yarl-1.18.3-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:877d209b6aebeb5b16c42cbb377f5f94d9e556626b1bfff66d7b0d115be88d0a", size = 359981 },
-    { url = "https://files.pythonhosted.org/packages/ca/b3/d493221ad5cbd18bc07e642894030437e405e1413c4236dd5db6e46bcec9/yarl-1.18.3-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:b464c4ab4bfcb41e3bfd3f1c26600d038376c2de3297760dfe064d2cb7ea8e10", size = 366229 },
-    { url = "https://files.pythonhosted.org/packages/04/56/6a3e2a5d9152c56c346df9b8fb8edd2c8888b1e03f96324d457e5cf06d34/yarl-1.18.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:8d39d351e7faf01483cc7ff7c0213c412e38e5a340238826be7e0e4da450fdc8", size = 360383 },
-    { url = "https://files.pythonhosted.org/packages/fd/b7/4b3c7c7913a278d445cc6284e59b2e62fa25e72758f888b7a7a39eb8423f/yarl-1.18.3-cp313-cp313-win32.whl", hash = "sha256:61ee62ead9b68b9123ec24bc866cbef297dd266175d53296e2db5e7f797f902d", size = 310152 },
-    { url = "https://files.pythonhosted.org/packages/f5/d5/688db678e987c3e0fb17867970700b92603cadf36c56e5fb08f23e822a0c/yarl-1.18.3-cp313-cp313-win_amd64.whl", hash = "sha256:578e281c393af575879990861823ef19d66e2b1d0098414855dd367e234f5b3c", size = 315723 },
-    { url = "https://files.pythonhosted.org/packages/6a/3b/fec4b08f5e88f68e56ee698a59284a73704df2e0e0b5bdf6536c86e76c76/yarl-1.18.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:61e5e68cb65ac8f547f6b5ef933f510134a6bf31bb178be428994b0cb46c2a04", size = 142780 },
-    { url = "https://files.pythonhosted.org/packages/ed/85/796b0d6a22d536ec8e14bdbb86519250bad980cec450b6e299b1c2a9079e/yarl-1.18.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:fe57328fbc1bfd0bd0514470ac692630f3901c0ee39052ae47acd1d90a436719", size = 94981 },
-    { url = "https://files.pythonhosted.org/packages/ee/0e/a830fd2238f7a29050f6dd0de748b3d6f33a7dbb67dbbc081a970b2bbbeb/yarl-1.18.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a440a2a624683108a1b454705ecd7afc1c3438a08e890a1513d468671d90a04e", size = 92789 },
-    { url = "https://files.pythonhosted.org/packages/0f/4f/438c9fd668954779e48f08c0688ee25e0673380a21bb1e8ccc56de5b55d7/yarl-1.18.3-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:09c7907c8548bcd6ab860e5f513e727c53b4a714f459b084f6580b49fa1b9cee", size = 317327 },
-    { url = "https://files.pythonhosted.org/packages/bd/79/a78066f06179b4ed4581186c136c12fcfb928c475cbeb23743e71a991935/yarl-1.18.3-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b4f6450109834af88cb4cc5ecddfc5380ebb9c228695afc11915a0bf82116789", size = 336999 },
-    { url = "https://files.pythonhosted.org/packages/55/02/527963cf65f34a06aed1e766ff9a3b3e7d0eaa1c90736b2948a62e528e1d/yarl-1.18.3-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a9ca04806f3be0ac6d558fffc2fdf8fcef767e0489d2684a21912cc4ed0cd1b8", size = 331693 },
-    { url = "https://files.pythonhosted.org/packages/a2/2a/167447ae39252ba624b98b8c13c0ba35994d40d9110e8a724c83dbbb5822/yarl-1.18.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:77a6e85b90a7641d2e07184df5557132a337f136250caafc9ccaa4a2a998ca2c", size = 321473 },
-    { url = "https://files.pythonhosted.org/packages/55/03/07955fabb20082373be311c91fd78abe458bc7ff9069d34385e8bddad20e/yarl-1.18.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6333c5a377c8e2f5fae35e7b8f145c617b02c939d04110c76f29ee3676b5f9a5", size = 313571 },
-    { url = "https://files.pythonhosted.org/packages/95/e2/67c8d3ec58a8cd8ddb1d63bd06eb7e7b91c9f148707a3eeb5a7ed87df0ef/yarl-1.18.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:0b3c92fa08759dbf12b3a59579a4096ba9af8dd344d9a813fc7f5070d86bbab1", size = 325004 },
-    { url = "https://files.pythonhosted.org/packages/06/43/51ceb3e427368fe6ccd9eccd162be227fd082523e02bad1fd3063daf68da/yarl-1.18.3-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:4ac515b860c36becb81bb84b667466885096b5fc85596948548b667da3bf9f24", size = 322677 },
-    { url = "https://files.pythonhosted.org/packages/e4/0e/7ef286bfb23267739a703f7b967a858e2128c10bea898de8fa027e962521/yarl-1.18.3-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:045b8482ce9483ada4f3f23b3774f4e1bf4f23a2d5c912ed5170f68efb053318", size = 332806 },
-    { url = "https://files.pythonhosted.org/packages/c8/94/2d1f060f4bfa47c8bd0bcb652bfe71fba881564bcac06ebb6d8ced9ac3bc/yarl-1.18.3-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:a4bb030cf46a434ec0225bddbebd4b89e6471814ca851abb8696170adb163985", size = 339919 },
-    { url = "https://files.pythonhosted.org/packages/8e/8d/73b5f9a6ab69acddf1ca1d5e7bc92f50b69124512e6c26b36844531d7f23/yarl-1.18.3-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:54d6921f07555713b9300bee9c50fb46e57e2e639027089b1d795ecd9f7fa910", size = 340960 },
-    { url = "https://files.pythonhosted.org/packages/41/13/ce6bc32be4476b60f4f8694831f49590884b2c975afcffc8d533bf2be7ec/yarl-1.18.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1d407181cfa6e70077df3377938c08012d18893f9f20e92f7d2f314a437c30b1", size = 336592 },
-    { url = "https://files.pythonhosted.org/packages/81/d5/6e0460292d6299ac3919945f912b16b104f4e81ab20bf53e0872a1296daf/yarl-1.18.3-cp39-cp39-win32.whl", hash = "sha256:ac36703a585e0929b032fbaab0707b75dc12703766d0b53486eabd5139ebadd5", size = 84833 },
-    { url = "https://files.pythonhosted.org/packages/b2/fc/a8aef69156ad5508165d8ae956736d55c3a68890610834bd985540966008/yarl-1.18.3-cp39-cp39-win_amd64.whl", hash = "sha256:ba87babd629f8af77f557b61e49e7c7cac36f22f871156b91e10a6e9d4f829e9", size = 90968 },
-    { url = "https://files.pythonhosted.org/packages/f5/4b/a06e0ec3d155924f77835ed2d167ebd3b211a7b0853da1cf8d8414d784ef/yarl-1.18.3-py3-none-any.whl", hash = "sha256:b57f4f58099328dfb26c6a771d09fb20dbbae81d20cfb66141251ea063bd101b", size = 45109 },
+sdist = { url = "https://files.pythonhosted.org/packages/3c/fb/efaa23fa4e45537b827620f04cf8f3cd658b76642205162e072703a5b963/yarl-1.20.1.tar.gz", hash = "sha256:d017a4997ee50c91fd5466cef416231bb82177b93b029906cefc542ce14c35ac", size = 186428, upload-time = "2025-06-10T00:46:09.923Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cb/65/7fed0d774abf47487c64be14e9223749468922817b5e8792b8a64792a1bb/yarl-1.20.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:6032e6da6abd41e4acda34d75a816012717000fa6839f37124a47fcefc49bec4", size = 132910, upload-time = "2025-06-10T00:42:31.108Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/7b/988f55a52da99df9e56dc733b8e4e5a6ae2090081dc2754fc8fd34e60aa0/yarl-1.20.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2c7b34d804b8cf9b214f05015c4fee2ebe7ed05cf581e7192c06555c71f4446a", size = 90644, upload-time = "2025-06-10T00:42:33.851Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/de/30d98f03e95d30c7e3cc093759982d038c8833ec2451001d45ef4854edc1/yarl-1.20.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0c869f2651cc77465f6cd01d938d91a11d9ea5d798738c1dc077f3de0b5e5fed", size = 89322, upload-time = "2025-06-10T00:42:35.688Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/7a/f2f314f5ebfe9200724b0b748de2186b927acb334cf964fd312eb86fc286/yarl-1.20.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:62915e6688eb4d180d93840cda4110995ad50c459bf931b8b3775b37c264af1e", size = 323786, upload-time = "2025-06-10T00:42:37.817Z" },
+    { url = "https://files.pythonhosted.org/packages/15/3f/718d26f189db96d993d14b984ce91de52e76309d0fd1d4296f34039856aa/yarl-1.20.1-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:41ebd28167bc6af8abb97fec1a399f412eec5fd61a3ccbe2305a18b84fb4ca73", size = 319627, upload-time = "2025-06-10T00:42:39.937Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/76/8fcfbf5fa2369157b9898962a4a7d96764b287b085b5b3d9ffae69cdefd1/yarl-1.20.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:21242b4288a6d56f04ea193adde174b7e347ac46ce6bc84989ff7c1b1ecea84e", size = 339149, upload-time = "2025-06-10T00:42:42.627Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/95/d7fc301cc4661785967acc04f54a4a42d5124905e27db27bb578aac49b5c/yarl-1.20.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bea21cdae6c7eb02ba02a475f37463abfe0a01f5d7200121b03e605d6a0439f8", size = 333327, upload-time = "2025-06-10T00:42:44.842Z" },
+    { url = "https://files.pythonhosted.org/packages/65/94/e21269718349582eee81efc5c1c08ee71c816bfc1585b77d0ec3f58089eb/yarl-1.20.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1f8a891e4a22a89f5dde7862994485e19db246b70bb288d3ce73a34422e55b23", size = 326054, upload-time = "2025-06-10T00:42:47.149Z" },
+    { url = "https://files.pythonhosted.org/packages/32/ae/8616d1f07853704523519f6131d21f092e567c5af93de7e3e94b38d7f065/yarl-1.20.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:dd803820d44c8853a109a34e3660e5a61beae12970da479cf44aa2954019bf70", size = 315035, upload-time = "2025-06-10T00:42:48.852Z" },
+    { url = "https://files.pythonhosted.org/packages/48/aa/0ace06280861ef055855333707db5e49c6e3a08840a7ce62682259d0a6c0/yarl-1.20.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:b982fa7f74c80d5c0c7b5b38f908971e513380a10fecea528091405f519b9ebb", size = 338962, upload-time = "2025-06-10T00:42:51.024Z" },
+    { url = "https://files.pythonhosted.org/packages/20/52/1e9d0e6916f45a8fb50e6844f01cb34692455f1acd548606cbda8134cd1e/yarl-1.20.1-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:33f29ecfe0330c570d997bcf1afd304377f2e48f61447f37e846a6058a4d33b2", size = 335399, upload-time = "2025-06-10T00:42:53.007Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/65/60452df742952c630e82f394cd409de10610481d9043aa14c61bf846b7b1/yarl-1.20.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:835ab2cfc74d5eb4a6a528c57f05688099da41cf4957cf08cad38647e4a83b30", size = 338649, upload-time = "2025-06-10T00:42:54.964Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/f5/6cd4ff38dcde57a70f23719a838665ee17079640c77087404c3d34da6727/yarl-1.20.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:46b5e0ccf1943a9a6e766b2c2b8c732c55b34e28be57d8daa2b3c1d1d4009309", size = 358563, upload-time = "2025-06-10T00:42:57.28Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/90/c42eefd79d0d8222cb3227bdd51b640c0c1d0aa33fe4cc86c36eccba77d3/yarl-1.20.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:df47c55f7d74127d1b11251fe6397d84afdde0d53b90bedb46a23c0e534f9d24", size = 357609, upload-time = "2025-06-10T00:42:59.055Z" },
+    { url = "https://files.pythonhosted.org/packages/03/c8/cea6b232cb4617514232e0f8a718153a95b5d82b5290711b201545825532/yarl-1.20.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:76d12524d05841276b0e22573f28d5fbcb67589836772ae9244d90dd7d66aa13", size = 350224, upload-time = "2025-06-10T00:43:01.248Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/a3/eaa0ab9712f1f3d01faf43cf6f1f7210ce4ea4a7e9b28b489a2261ca8db9/yarl-1.20.1-cp310-cp310-win32.whl", hash = "sha256:6c4fbf6b02d70e512d7ade4b1f998f237137f1417ab07ec06358ea04f69134f8", size = 81753, upload-time = "2025-06-10T00:43:03.486Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/34/e4abde70a9256465fe31c88ed02c3f8502b7b5dead693a4f350a06413f28/yarl-1.20.1-cp310-cp310-win_amd64.whl", hash = "sha256:aef6c4d69554d44b7f9d923245f8ad9a707d971e6209d51279196d8e8fe1ae16", size = 86817, upload-time = "2025-06-10T00:43:05.231Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/18/893b50efc2350e47a874c5c2d67e55a0ea5df91186b2a6f5ac52eff887cd/yarl-1.20.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:47ee6188fea634bdfaeb2cc420f5b3b17332e6225ce88149a17c413c77ff269e", size = 133833, upload-time = "2025-06-10T00:43:07.393Z" },
+    { url = "https://files.pythonhosted.org/packages/89/ed/b8773448030e6fc47fa797f099ab9eab151a43a25717f9ac043844ad5ea3/yarl-1.20.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:d0f6500f69e8402d513e5eedb77a4e1818691e8f45e6b687147963514d84b44b", size = 91070, upload-time = "2025-06-10T00:43:09.538Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/e3/409bd17b1e42619bf69f60e4f031ce1ccb29bd7380117a55529e76933464/yarl-1.20.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7a8900a42fcdaad568de58887c7b2f602962356908eedb7628eaf6021a6e435b", size = 89818, upload-time = "2025-06-10T00:43:11.575Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/77/64d8431a4d77c856eb2d82aa3de2ad6741365245a29b3a9543cd598ed8c5/yarl-1.20.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bad6d131fda8ef508b36be3ece16d0902e80b88ea7200f030a0f6c11d9e508d4", size = 347003, upload-time = "2025-06-10T00:43:14.088Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/d2/0c7e4def093dcef0bd9fa22d4d24b023788b0a33b8d0088b51aa51e21e99/yarl-1.20.1-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:df018d92fe22aaebb679a7f89fe0c0f368ec497e3dda6cb81a567610f04501f1", size = 336537, upload-time = "2025-06-10T00:43:16.431Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/f3/fc514f4b2cf02cb59d10cbfe228691d25929ce8f72a38db07d3febc3f706/yarl-1.20.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8f969afbb0a9b63c18d0feecf0db09d164b7a44a053e78a7d05f5df163e43833", size = 362358, upload-time = "2025-06-10T00:43:18.704Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/6d/a313ac8d8391381ff9006ac05f1d4331cee3b1efaa833a53d12253733255/yarl-1.20.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:812303eb4aa98e302886ccda58d6b099e3576b1b9276161469c25803a8db277d", size = 357362, upload-time = "2025-06-10T00:43:20.888Z" },
+    { url = "https://files.pythonhosted.org/packages/00/70/8f78a95d6935a70263d46caa3dd18e1f223cf2f2ff2037baa01a22bc5b22/yarl-1.20.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98c4a7d166635147924aa0bf9bfe8d8abad6fffa6102de9c99ea04a1376f91e8", size = 348979, upload-time = "2025-06-10T00:43:23.169Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/05/42773027968968f4f15143553970ee36ead27038d627f457cc44bbbeecf3/yarl-1.20.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:12e768f966538e81e6e7550f9086a6236b16e26cd964cf4df35349970f3551cf", size = 337274, upload-time = "2025-06-10T00:43:27.111Z" },
+    { url = "https://files.pythonhosted.org/packages/05/be/665634aa196954156741ea591d2f946f1b78ceee8bb8f28488bf28c0dd62/yarl-1.20.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:fe41919b9d899661c5c28a8b4b0acf704510b88f27f0934ac7a7bebdd8938d5e", size = 363294, upload-time = "2025-06-10T00:43:28.96Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/90/73448401d36fa4e210ece5579895731f190d5119c4b66b43b52182e88cd5/yarl-1.20.1-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:8601bc010d1d7780592f3fc1bdc6c72e2b6466ea34569778422943e1a1f3c389", size = 358169, upload-time = "2025-06-10T00:43:30.701Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/b0/fce922d46dc1eb43c811f1889f7daa6001b27a4005587e94878570300881/yarl-1.20.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:daadbdc1f2a9033a2399c42646fbd46da7992e868a5fe9513860122d7fe7a73f", size = 362776, upload-time = "2025-06-10T00:43:32.51Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/0d/b172628fce039dae8977fd22caeff3eeebffd52e86060413f5673767c427/yarl-1.20.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:03aa1e041727cb438ca762628109ef1333498b122e4c76dd858d186a37cec845", size = 381341, upload-time = "2025-06-10T00:43:34.543Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/9b/5b886d7671f4580209e855974fe1cecec409aa4a89ea58b8f0560dc529b1/yarl-1.20.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:642980ef5e0fa1de5fa96d905c7e00cb2c47cb468bfcac5a18c58e27dbf8d8d1", size = 379988, upload-time = "2025-06-10T00:43:36.489Z" },
+    { url = "https://files.pythonhosted.org/packages/73/be/75ef5fd0fcd8f083a5d13f78fd3f009528132a1f2a1d7c925c39fa20aa79/yarl-1.20.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:86971e2795584fe8c002356d3b97ef6c61862720eeff03db2a7c86b678d85b3e", size = 371113, upload-time = "2025-06-10T00:43:38.592Z" },
+    { url = "https://files.pythonhosted.org/packages/50/4f/62faab3b479dfdcb741fe9e3f0323e2a7d5cd1ab2edc73221d57ad4834b2/yarl-1.20.1-cp311-cp311-win32.whl", hash = "sha256:597f40615b8d25812f14562699e287f0dcc035d25eb74da72cae043bb884d773", size = 81485, upload-time = "2025-06-10T00:43:41.038Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/09/d9c7942f8f05c32ec72cd5c8e041c8b29b5807328b68b4801ff2511d4d5e/yarl-1.20.1-cp311-cp311-win_amd64.whl", hash = "sha256:26ef53a9e726e61e9cd1cda6b478f17e350fb5800b4bd1cd9fe81c4d91cfeb2e", size = 86686, upload-time = "2025-06-10T00:43:42.692Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/9a/cb7fad7d73c69f296eda6815e4a2c7ed53fc70c2f136479a91c8e5fbdb6d/yarl-1.20.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:bdcc4cd244e58593a4379fe60fdee5ac0331f8eb70320a24d591a3be197b94a9", size = 133667, upload-time = "2025-06-10T00:43:44.369Z" },
+    { url = "https://files.pythonhosted.org/packages/67/38/688577a1cb1e656e3971fb66a3492501c5a5df56d99722e57c98249e5b8a/yarl-1.20.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:b29a2c385a5f5b9c7d9347e5812b6f7ab267193c62d282a540b4fc528c8a9d2a", size = 91025, upload-time = "2025-06-10T00:43:46.295Z" },
+    { url = "https://files.pythonhosted.org/packages/50/ec/72991ae51febeb11a42813fc259f0d4c8e0507f2b74b5514618d8b640365/yarl-1.20.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1112ae8154186dfe2de4732197f59c05a83dc814849a5ced892b708033f40dc2", size = 89709, upload-time = "2025-06-10T00:43:48.22Z" },
+    { url = "https://files.pythonhosted.org/packages/99/da/4d798025490e89426e9f976702e5f9482005c548c579bdae792a4c37769e/yarl-1.20.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:90bbd29c4fe234233f7fa2b9b121fb63c321830e5d05b45153a2ca68f7d310ee", size = 352287, upload-time = "2025-06-10T00:43:49.924Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/26/54a15c6a567aac1c61b18aa0f4b8aa2e285a52d547d1be8bf48abe2b3991/yarl-1.20.1-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:680e19c7ce3710ac4cd964e90dad99bf9b5029372ba0c7cbfcd55e54d90ea819", size = 345429, upload-time = "2025-06-10T00:43:51.7Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/95/9dcf2386cb875b234353b93ec43e40219e14900e046bf6ac118f94b1e353/yarl-1.20.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4a979218c1fdb4246a05efc2cc23859d47c89af463a90b99b7c56094daf25a16", size = 365429, upload-time = "2025-06-10T00:43:53.494Z" },
+    { url = "https://files.pythonhosted.org/packages/91/b2/33a8750f6a4bc224242a635f5f2cff6d6ad5ba651f6edcccf721992c21a0/yarl-1.20.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:255b468adf57b4a7b65d8aad5b5138dce6a0752c139965711bdcb81bc370e1b6", size = 363862, upload-time = "2025-06-10T00:43:55.766Z" },
+    { url = "https://files.pythonhosted.org/packages/98/28/3ab7acc5b51f4434b181b0cee8f1f4b77a65919700a355fb3617f9488874/yarl-1.20.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a97d67108e79cfe22e2b430d80d7571ae57d19f17cda8bb967057ca8a7bf5bfd", size = 355616, upload-time = "2025-06-10T00:43:58.056Z" },
+    { url = "https://files.pythonhosted.org/packages/36/a3/f666894aa947a371724ec7cd2e5daa78ee8a777b21509b4252dd7bd15e29/yarl-1.20.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:8570d998db4ddbfb9a590b185a0a33dbf8aafb831d07a5257b4ec9948df9cb0a", size = 339954, upload-time = "2025-06-10T00:43:59.773Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/81/5f466427e09773c04219d3450d7a1256138a010b6c9f0af2d48565e9ad13/yarl-1.20.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:97c75596019baae7c71ccf1d8cc4738bc08134060d0adfcbe5642f778d1dca38", size = 365575, upload-time = "2025-06-10T00:44:02.051Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/e3/e4b0ad8403e97e6c9972dd587388940a032f030ebec196ab81a3b8e94d31/yarl-1.20.1-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:1c48912653e63aef91ff988c5432832692ac5a1d8f0fb8a33091520b5bbe19ef", size = 365061, upload-time = "2025-06-10T00:44:04.196Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/99/b8a142e79eb86c926f9f06452eb13ecb1bb5713bd01dc0038faf5452e544/yarl-1.20.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:4c3ae28f3ae1563c50f3d37f064ddb1511ecc1d5584e88c6b7c63cf7702a6d5f", size = 364142, upload-time = "2025-06-10T00:44:06.527Z" },
+    { url = "https://files.pythonhosted.org/packages/34/f2/08ed34a4a506d82a1a3e5bab99ccd930a040f9b6449e9fd050320e45845c/yarl-1.20.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:c5e9642f27036283550f5f57dc6156c51084b458570b9d0d96100c8bebb186a8", size = 381894, upload-time = "2025-06-10T00:44:08.379Z" },
+    { url = "https://files.pythonhosted.org/packages/92/f8/9a3fbf0968eac704f681726eff595dce9b49c8a25cd92bf83df209668285/yarl-1.20.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:2c26b0c49220d5799f7b22c6838409ee9bc58ee5c95361a4d7831f03cc225b5a", size = 383378, upload-time = "2025-06-10T00:44:10.51Z" },
+    { url = "https://files.pythonhosted.org/packages/af/85/9363f77bdfa1e4d690957cd39d192c4cacd1c58965df0470a4905253b54f/yarl-1.20.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:564ab3d517e3d01c408c67f2e5247aad4019dcf1969982aba3974b4093279004", size = 374069, upload-time = "2025-06-10T00:44:12.834Z" },
+    { url = "https://files.pythonhosted.org/packages/35/99/9918c8739ba271dcd935400cff8b32e3cd319eaf02fcd023d5dcd487a7c8/yarl-1.20.1-cp312-cp312-win32.whl", hash = "sha256:daea0d313868da1cf2fac6b2d3a25c6e3a9e879483244be38c8e6a41f1d876a5", size = 81249, upload-time = "2025-06-10T00:44:14.731Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/83/5d9092950565481b413b31a23e75dd3418ff0a277d6e0abf3729d4d1ce25/yarl-1.20.1-cp312-cp312-win_amd64.whl", hash = "sha256:48ea7d7f9be0487339828a4de0360d7ce0efc06524a48e1810f945c45b813698", size = 86710, upload-time = "2025-06-10T00:44:16.716Z" },
+    { url = "https://files.pythonhosted.org/packages/01/75/0d37402d208d025afa6b5b8eb80e466d267d3fd1927db8e317d29a94a4cb/yarl-1.20.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:e42ba79e2efb6845ebab49c7bf20306c4edf74a0b20fc6b2ccdd1a219d12fad3", size = 134259, upload-time = "2025-06-10T00:45:29.882Z" },
+    { url = "https://files.pythonhosted.org/packages/73/84/1fb6c85ae0cf9901046f07d0ac9eb162f7ce6d95db541130aa542ed377e6/yarl-1.20.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:41493b9b7c312ac448b7f0a42a089dffe1d6e6e981a2d76205801a023ed26a2b", size = 91269, upload-time = "2025-06-10T00:45:32.917Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/9c/eae746b24c4ea29a5accba9a06c197a70fa38a49c7df244e0d3951108861/yarl-1.20.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:f5a5928ff5eb13408c62a968ac90d43f8322fd56d87008b8f9dabf3c0f6ee983", size = 89995, upload-time = "2025-06-10T00:45:35.066Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/30/693e71003ec4bc1daf2e4cf7c478c417d0985e0a8e8f00b2230d517876fc/yarl-1.20.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:30c41ad5d717b3961b2dd785593b67d386b73feca30522048d37298fee981805", size = 325253, upload-time = "2025-06-10T00:45:37.052Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/a2/5264dbebf90763139aeb0b0b3154763239398400f754ae19a0518b654117/yarl-1.20.1-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:59febc3969b0781682b469d4aca1a5cab7505a4f7b85acf6db01fa500fa3f6ba", size = 320897, upload-time = "2025-06-10T00:45:39.962Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/17/77c7a89b3c05856489777e922f41db79ab4faf58621886df40d812c7facd/yarl-1.20.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d2b6fb3622b7e5bf7a6e5b679a69326b4279e805ed1699d749739a61d242449e", size = 340696, upload-time = "2025-06-10T00:45:41.915Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/55/28409330b8ef5f2f681f5b478150496ec9cf3309b149dab7ec8ab5cfa3f0/yarl-1.20.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:749d73611db8d26a6281086f859ea7ec08f9c4c56cec864e52028c8b328db723", size = 335064, upload-time = "2025-06-10T00:45:43.893Z" },
+    { url = "https://files.pythonhosted.org/packages/85/58/cb0257cbd4002828ff735f44d3c5b6966c4fd1fc8cc1cd3cd8a143fbc513/yarl-1.20.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9427925776096e664c39e131447aa20ec738bdd77c049c48ea5200db2237e000", size = 327256, upload-time = "2025-06-10T00:45:46.393Z" },
+    { url = "https://files.pythonhosted.org/packages/53/f6/c77960370cfa46f6fb3d6a5a79a49d3abfdb9ef92556badc2dcd2748bc2a/yarl-1.20.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ff70f32aa316393eaf8222d518ce9118148eddb8a53073c2403863b41033eed5", size = 316389, upload-time = "2025-06-10T00:45:48.358Z" },
+    { url = "https://files.pythonhosted.org/packages/64/ab/be0b10b8e029553c10905b6b00c64ecad3ebc8ace44b02293a62579343f6/yarl-1.20.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:c7ddf7a09f38667aea38801da8b8d6bfe81df767d9dfc8c88eb45827b195cd1c", size = 340481, upload-time = "2025-06-10T00:45:50.663Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/c3/3f327bd3905a4916029bf5feb7f86dcf864c7704f099715f62155fb386b2/yarl-1.20.1-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:57edc88517d7fc62b174fcfb2e939fbc486a68315d648d7e74d07fac42cec240", size = 336941, upload-time = "2025-06-10T00:45:52.554Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/42/040bdd5d3b3bb02b4a6ace4ed4075e02f85df964d6e6cb321795d2a6496a/yarl-1.20.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:dab096ce479d5894d62c26ff4f699ec9072269d514b4edd630a393223f45a0ee", size = 339936, upload-time = "2025-06-10T00:45:54.919Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/1c/911867b8e8c7463b84dfdc275e0d99b04b66ad5132b503f184fe76be8ea4/yarl-1.20.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:14a85f3bd2d7bb255be7183e5d7d6e70add151a98edf56a770d6140f5d5f4010", size = 360163, upload-time = "2025-06-10T00:45:56.87Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/31/8c389f6c6ca0379b57b2da87f1f126c834777b4931c5ee8427dd65d0ff6b/yarl-1.20.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:2c89b5c792685dd9cd3fa9761c1b9f46fc240c2a3265483acc1565769996a3f8", size = 359108, upload-time = "2025-06-10T00:45:58.869Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/09/ae4a649fb3964324c70a3e2b61f45e566d9ffc0affd2b974cbf628957673/yarl-1.20.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:69e9b141de5511021942a6866990aea6d111c9042235de90e08f94cf972ca03d", size = 351875, upload-time = "2025-06-10T00:46:01.45Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/43/bbb4ed4c34d5bb62b48bf957f68cd43f736f79059d4f85225ab1ef80f4b9/yarl-1.20.1-cp39-cp39-win32.whl", hash = "sha256:b5f307337819cdfdbb40193cad84978a029f847b0a357fbe49f712063cfc4f06", size = 82293, upload-time = "2025-06-10T00:46:03.763Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/cd/ce185848a7dba68ea69e932674b5c1a42a1852123584bccc5443120f857c/yarl-1.20.1-cp39-cp39-win_amd64.whl", hash = "sha256:eae7bfe2069f9c1c5b05fc7fe5d612e5bbc089a39309904ee8b829e322dcad00", size = 87385, upload-time = "2025-06-10T00:46:05.655Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/2d/2345fce04cfd4bee161bf1e7d9cdc702e3e16109021035dbb24db654a622/yarl-1.20.1-py3-none-any.whl", hash = "sha256:83b8eb083fe4683c6115795d9fc1cfaf2cbbefb19b3a1cb68f6527460f483a77", size = 46542, upload-time = "2025-06-10T00:46:07.521Z" },
 ]
 
 [[package]]
 name = "zipp"
-version = "3.21.0"
+version = "3.23.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/3f/50/bad581df71744867e9468ebd0bcd6505de3b275e06f202c2cb016e3ff56f/zipp-3.21.0.tar.gz", hash = "sha256:2c9958f6430a2040341a52eb608ed6dd93ef4392e02ffe219417c1b28b5dd1f4", size = 24545 }
+sdist = { url = "https://files.pythonhosted.org/packages/e3/02/0f2892c661036d50ede074e376733dca2ae7c6eb617489437771209d4180/zipp-3.23.0.tar.gz", hash = "sha256:a07157588a12518c9d4034df3fbbee09c814741a33ff63c05fa29d26a2404166", size = 25547, upload-time = "2025-06-08T17:06:39.4Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b7/1a/7e4798e9339adc931158c9d69ecc34f5e6791489d469f5e50ec15e35f458/zipp-3.21.0-py3-none-any.whl", hash = "sha256:ac1bbe05fd2991f160ebce24ffbac5f6d11d83dc90891255885223d42b3cd931", size = 9630 },
+    { url = "https://files.pythonhosted.org/packages/2e/54/647ade08bf0db230bfea292f893923872fd20be6ac6f53b2b936ba839d75/zipp-3.23.0-py3-none-any.whl", hash = "sha256:071652d6115ed432f5ce1d34c336c0adfd6a884660d1e9712a256d3d3bd4b14e", size = 10276, upload-time = "2025-06-08T17:06:38.034Z" },
 ]

From 849a969d56775b998990784cedede1c4e1a00f27 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 1 Sep 2025 14:02:29 +0200
Subject: [PATCH 441/505] Remove tests for flax and tensorflow for the
 transformers model

---
 outlines/models/transformers.py   |  4 ++--
 tests/models/test_transformers.py | 24 ------------------------
 2 files changed, 2 insertions(+), 26 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index cf677cdde..964dbce60 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -242,7 +242,7 @@ def __init__(
         if (
             FlaxPreTrainedModel is not None
             and isinstance(model, FlaxPreTrainedModel)
-        ):
+        ):  # pragma: no cover
             self.tensor_library_name = "jax"
             warnings.warn("""
                 Support for `jax` has been deprecated and will be removed in
@@ -256,7 +256,7 @@ def __init__(
         elif (
             TFPreTrainedModel is not None
             and isinstance(model, TFPreTrainedModel)
-        ):
+        ):  # pragma: no cover
             self.tensor_library_name = "tensorflow"
             warnings.warn("""
                 Support for `tensorflow` has been deprecated and will be removed in
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index d98176c22..10f572794 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -39,30 +39,6 @@ def test_transformers_instantiate_simple():
     assert model.tensor_library_name == "torch"
 
 
-def test_transformers_instantiate_flax_model():
-    with pytest.warns(DeprecationWarning):
-        model = outlines.from_transformers(
-            transformers.FlaxAutoModelForCausalLM.from_pretrained(TEST_MODEL),
-            transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-        )
-        assert isinstance(model, Transformers)
-        assert isinstance(model.tokenizer, TransformerTokenizer)
-        assert isinstance(model.type_adapter, TransformersTypeAdapter)
-        assert model.tensor_library_name == "jax"
-
-
-def test_transformers_instantiate_tensorflow_model():
-    with pytest.warns(DeprecationWarning):
-        model = outlines.from_transformers(
-            transformers.TFAutoModelForCausalLM.from_pretrained(TEST_MODEL),
-            transformers.AutoTokenizer.from_pretrained(TEST_MODEL),
-        )
-        assert isinstance(model, Transformers)
-        assert isinstance(model.tokenizer, TransformerTokenizer)
-        assert isinstance(model.type_adapter, TransformersTypeAdapter)
-        assert model.tensor_library_name == "tensorflow"
-
-
 def test_transformers_instantiate_mamba():
     model = outlines.from_transformers(
         transformers.MambaForCausalLM.from_pretrained(TEST_MODEL_MAMBA),

From 6533d5d80df32b0d2606e144b6a31dec018eaf32 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Mon, 1 Sep 2025 14:45:06 +0200
Subject: [PATCH 442/505] Add support for the Chat input to the MLXLM model

---
 docs/features/models/llamacpp.md        |  2 +-
 docs/features/models/mlxlm.md           | 55 ++++++++++++----
 outlines/models/mlxlm.py                | 25 +++++++-
 tests/models/test_mlxlm_type_adapter.py | 84 +++++++++++++++++++++++++
 4 files changed, 150 insertions(+), 16 deletions(-)
 create mode 100644 tests/models/test_mlxlm_type_adapter.py

diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
index 6f3c83173..1dd2c31f2 100644
--- a/docs/features/models/llamacpp.md
+++ b/docs/features/models/llamacpp.md
@@ -81,7 +81,7 @@ prompt = Chat([
 
 # Call the model to generate a response
 response = model(prompt, max_tokens=50)
-print(response) # 'This is a picture of a black dog.'
+print(response) # 'Riga.'
 ```
 
 #### Streaming
diff --git a/docs/features/models/mlxlm.md b/docs/features/models/mlxlm.md
index d2cba95fc..6ccf7f38a 100644
--- a/docs/features/models/mlxlm.md
+++ b/docs/features/models/mlxlm.md
@@ -29,7 +29,7 @@ import mlx_lm
 
 # Create the model
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 ```
 
@@ -45,7 +45,7 @@ import mlx_lm
 
 # Load the model
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 
 # Call it to generate text
@@ -53,6 +53,35 @@ result = model("What's the capital of Latvia?", max_tokens=20)
 print(result) # 'Riga'
 ```
 
+#### Chat
+
+You can use chat inputs with the `MLXLM` model. To do so, call the model with a `Chat` instance.
+
+For instance:
+
+```python
+import outlines
+import mlx_lm
+from outlines.inputs import Chat
+
+# Load the model
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
+)
+
+# Create the prompt containing the text and the image
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "assistant", "content": "What's the capital of Latvia?"},
+])
+
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'Riga.'
+```
+
+#### Streaming
+
 The `MLXLM` model also supports streaming. For instance:
 
 ```python
@@ -61,7 +90,7 @@ import mlx_lm
 
 # Load the model
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 
 # Stream text
@@ -73,7 +102,7 @@ for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
 
 As a local model, `MLXLM` supports all forms of structured generation available in Outlines.
 
-### Basic Type
+#### Basic Type
 
 ```python
 import outlines
@@ -82,14 +111,14 @@ import mlx_lm
 output_type = int
 
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 
 result = model("How many countries are there in the world?", output_type)
 print(result) # '200'
 ```
 
-### JSON Schema
+#### JSON Schema
 
 ```python
 from pydantic import BaseModel
@@ -103,7 +132,7 @@ class Character(BaseModel):
     skills: List[str]
 
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 
 result = model("Create a character.", output_type=Character)
@@ -111,7 +140,7 @@ print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth",
 print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
 
-### Multiple Choice
+#### Multiple Choice
 
 ```python
 from typing import Literal
@@ -121,14 +150,14 @@ import mlx_lm
 output_type = Literal["Paris", "London", "Rome", "Berlin"]
 
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 
 result = model("What is the capital of France?", output_type)
 print(result) # 'Paris'
 ```
 
-### Regex
+#### Regex
 
 ```python
 from outlines.types import Regex
@@ -138,14 +167,14 @@ import mlx_lm
 output_type = Regex(r"\d{3}-\d{2}-\d{4}")
 
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 
 result = model("Generate a fake social security number.", output_type)
 print(result) # '782-32-3789'
 ```
 
-### Context-Free Grammar
+#### Context-Free Grammar
 
 ```python
 from outlines.types import CFG
@@ -175,7 +204,7 @@ arithmetic_grammar = """
 output_type = CFG(arithmetic_grammar)
 
 model = outlines.from_mlxlm(
-    *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
 )
 
 result = model("Write an addition.", output_type, max_tokens=20)
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 775a56222..498fec2dc 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -3,6 +3,7 @@
 from functools import singledispatchmethod
 from typing import TYPE_CHECKING, Iterator, List, Optional
 
+from outlines.inputs import Chat
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.models.transformers import TransformerTokenizer
 from outlines.processors import OutlinesLogitsProcessor
@@ -17,6 +18,9 @@
 class MLXLMTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `MLXLM` model."""
 
+    def __init__(self, **kwargs):
+        self.tokenizer = kwargs.get("tokenizer")
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
@@ -34,13 +38,30 @@ def format_input(self, model_input):
         """
         raise NotImplementedError(
             f"The input type {input} is not available with mlx-lm. "
-            "The only available type is `str`."
+            "The available types are `str` and `Chat`."
         )
 
     @format_input.register(str)
     def format_str_input(self, model_input: str):
         return model_input
 
+    @format_input.register(Chat)
+    def format_chat_input(self, model_input: Chat) -> str:
+        if not all(
+            isinstance(message["content"], str)
+            for message in model_input.messages
+        ):
+            raise ValueError(
+                "mlx-lm does not support multi-modal messages."
+                + "The content of each message must be a string."
+            )
+
+        return self.tokenizer.apply_chat_template(
+            model_input.messages,
+            tokenize=False,
+            add_generation_prompt=True,
+        )
+
     def format_output_type(
         self, output_type: Optional[OutlinesLogitsProcessor] = None,
     ) -> Optional[List[OutlinesLogitsProcessor]]:
@@ -92,7 +113,7 @@ def __init__(
         self.mlx_tokenizer = tokenizer
         # self.tokenizer is used by the logits processor
         self.tokenizer = TransformerTokenizer(tokenizer._tokenizer)
-        self.type_adapter = MLXLMTypeAdapter()
+        self.type_adapter = MLXLMTypeAdapter(tokenizer=tokenizer)
 
     def generate(
         self,
diff --git a/tests/models/test_mlxlm_type_adapter.py b/tests/models/test_mlxlm_type_adapter.py
new file mode 100644
index 000000000..dd2feb16d
--- /dev/null
+++ b/tests/models/test_mlxlm_type_adapter.py
@@ -0,0 +1,84 @@
+import pytest
+import io
+
+from outlines_core import Index, Vocabulary
+from PIL import Image as PILImage
+
+from outlines.backends.outlines_core import OutlinesCoreLogitsProcessor
+from outlines.inputs import Chat, Image
+from outlines.models.mlxlm import MLXLMTypeAdapter
+
+try:
+    import mlx_lm
+    import mlx.core as mx
+
+    HAS_MLX = mx.metal.is_available()
+except ImportError:
+    HAS_MLX = False
+
+
+MODEL_NAME = "mlx-community/SmolLM-135M-Instruct-4bit"
+
+
+@pytest.fixture
+def adapter():
+    _, tokenizer = mlx_lm.load(MODEL_NAME)
+    return MLXLMTypeAdapter(tokenizer=tokenizer)
+
+
+@pytest.fixture
+def logits_processor():
+    vocabulary = Vocabulary.from_pretrained(MODEL_NAME)
+    index = Index(r"[0-9]{3}", vocabulary)
+    return OutlinesCoreLogitsProcessor(index, "mlx")
+
+
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
+
+    return image
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_type_adapter_format_input(adapter, image):
+    # Anything else than a string/Chat (invalid)
+    with pytest.raises(NotImplementedError):
+        adapter.format_input(["Hello, world!"])
+
+    # String
+    assert adapter.format_input("Hello, world!") == "Hello, world!"
+
+    # Chat
+    messages = [
+        {"role": "user", "content": "Hello, world!"},
+        {"role": "assistant", "content": "Hello, world!"},
+    ]
+    expected = (
+        "<|im_start|>user\nHello, world!<|im_end|>\n<|im_start|>assistant\n"
+        + "Hello, world!<|im_end|>\n<|im_start|>assistant\n"
+    )
+    assert adapter.format_input(Chat(messages=messages)) == expected
+
+    # Multi-modal (invalid)
+    with pytest.raises(
+        ValueError,
+        match="mlx-lm does not support multi-modal messages."
+    ):
+        adapter.format_input(Chat(messages=[
+            {"role": "user", "content": ["prompt", Image(image)]},
+        ]))
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_type_adapter_format_output_type(adapter, logits_processor):
+    formatted = adapter.format_output_type(logits_processor)
+    assert isinstance(formatted, list)
+    assert len(formatted) == 1
+    assert isinstance(formatted[0], OutlinesCoreLogitsProcessor)

From 96ea0702f56cc38e6e882f0c17812de083702dcc Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 2 Sep 2025 11:35:12 +0200
Subject: [PATCH 443/505] Update uv.lock

---
 uv.lock | 361 ++++++++++++++++++++++++++++++--------------------------
 1 file changed, 192 insertions(+), 169 deletions(-)

diff --git a/uv.lock b/uv.lock
index cdd505bd0..7791d4dad 100644
--- a/uv.lock
+++ b/uv.lock
@@ -355,25 +355,47 @@ wheels = [
 name = "chex"
 version = "0.1.90"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version == '3.10.*'",
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "absl-py" },
+    { name = "absl-py", marker = "python_full_version < '3.11'" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jaxlib", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "setuptools", marker = "python_full_version >= '3.12'" },
-    { name = "toolz" },
-    { name = "typing-extensions" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "toolz", marker = "python_full_version < '3.11'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/77/70/53c7d404ce9e2a94009aea7f77ef6e392f6740e071c62683a506647c520f/chex-0.1.90.tar.gz", hash = "sha256:d3c375aeb6154b08f1cccd2bee4ed83659ee2198a6acf1160d2fe2e4a6c87b5c", size = 92363, upload-time = "2025-07-23T19:50:47.945Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/6f/3d/46bb04776c465cea2dd8aa2d4b61ab610b707f798f47838ef7e6105b025c/chex-0.1.90-py3-none-any.whl", hash = "sha256:fce3de82588f72d4796e545e574a433aa29229cbdcf792555e41bead24b704ae", size = 101047, upload-time = "2025-07-23T19:50:46.603Z" },
 ]
 
+[[package]]
+name = "chex"
+version = "0.1.91"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+]
+dependencies = [
+    { name = "absl-py", marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "toolz", marker = "python_full_version >= '3.11'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/5b/7d/812f01e7b2ddf28a0caa8dde56bd951a2c8f691c9bbfce38d469458d1502/chex-0.1.91.tar.gz", hash = "sha256:65367a521415ada905b8c0222b0a41a68337fcadf79a1fb6fc992dbd95dd9f76", size = 90302, upload-time = "2025-09-01T21:49:32.834Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/12/0c/96102c01dd02ae740d4afc3644d5c7d7fc51d3feefd67300a2aa1ddbf7cb/chex-0.1.91-py3-none-any.whl", hash = "sha256:6fc4cbfc22301c08d4a7ef706045668410100962eba8ba6af03fa07f4e5dcf9b", size = 100965, upload-time = "2025-09-01T21:49:31.141Z" },
+]
+
 [[package]]
 name = "click"
 version = "8.1.8"
@@ -426,53 +448,53 @@ wheels = [
 
 [[package]]
 name = "coverage"
-version = "7.10.5"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/61/83/153f54356c7c200013a752ce1ed5448573dca546ce125801afca9e1ac1a4/coverage-7.10.5.tar.gz", hash = "sha256:f2e57716a78bc3ae80b2207be0709a3b2b63b9f2dcf9740ee6ac03588a2015b6", size = 821662, upload-time = "2025-08-23T14:42:44.78Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/af/70/e77b0061a6c7157bfce645c6b9a715a08d4c86b3360a7b3252818080b817/coverage-7.10.5-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:c6a5c3414bfc7451b879141ce772c546985163cf553f08e0f135f0699a911801", size = 216774, upload-time = "2025-08-23T14:40:26.301Z" },
-    { url = "https://files.pythonhosted.org/packages/91/08/2a79de5ecf37ee40f2d898012306f11c161548753391cec763f92647837b/coverage-7.10.5-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:bc8e4d99ce82f1710cc3c125adc30fd1487d3cf6c2cd4994d78d68a47b16989a", size = 217175, upload-time = "2025-08-23T14:40:29.142Z" },
-    { url = "https://files.pythonhosted.org/packages/64/57/0171d69a699690149a6ba6a4eb702814448c8d617cf62dbafa7ce6bfdf63/coverage-7.10.5-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:02252dc1216e512a9311f596b3169fad54abcb13827a8d76d5630c798a50a754", size = 243931, upload-time = "2025-08-23T14:40:30.735Z" },
-    { url = "https://files.pythonhosted.org/packages/15/06/3a67662c55656702bd398a727a7f35df598eb11104fcb34f1ecbb070291a/coverage-7.10.5-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:73269df37883e02d460bee0cc16be90509faea1e3bd105d77360b512d5bb9c33", size = 245740, upload-time = "2025-08-23T14:40:32.302Z" },
-    { url = "https://files.pythonhosted.org/packages/00/f4/f8763aabf4dc30ef0d0012522d312f0b7f9fede6246a1f27dbcc4a1e523c/coverage-7.10.5-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1f8a81b0614642f91c9effd53eec284f965577591f51f547a1cbeb32035b4c2f", size = 247600, upload-time = "2025-08-23T14:40:33.66Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/31/6632219a9065e1b83f77eda116fed4c76fb64908a6a9feae41816dab8237/coverage-7.10.5-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:6a29f8e0adb7f8c2b95fa2d4566a1d6e6722e0a637634c6563cb1ab844427dd9", size = 245640, upload-time = "2025-08-23T14:40:35.248Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/e2/3dba9b86037b81649b11d192bb1df11dde9a81013e434af3520222707bc8/coverage-7.10.5-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:fcf6ab569436b4a647d4e91accba12509ad9f2554bc93d3aee23cc596e7f99c3", size = 243659, upload-time = "2025-08-23T14:40:36.815Z" },
-    { url = "https://files.pythonhosted.org/packages/02/b9/57170bd9f3e333837fc24ecc88bc70fbc2eb7ccfd0876854b0c0407078c3/coverage-7.10.5-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:90dc3d6fb222b194a5de60af8d190bedeeddcbc7add317e4a3cd333ee6b7c879", size = 244537, upload-time = "2025-08-23T14:40:38.737Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/1c/93ac36ef1e8b06b8d5777393a3a40cb356f9f3dab980be40a6941e443588/coverage-7.10.5-cp310-cp310-win32.whl", hash = "sha256:414a568cd545f9dc75f0686a0049393de8098414b58ea071e03395505b73d7a8", size = 219285, upload-time = "2025-08-23T14:40:40.342Z" },
-    { url = "https://files.pythonhosted.org/packages/30/95/23252277e6e5fe649d6cd3ed3f35d2307e5166de4e75e66aa7f432abc46d/coverage-7.10.5-cp310-cp310-win_amd64.whl", hash = "sha256:e551f9d03347196271935fd3c0c165f0e8c049220280c1120de0084d65e9c7ff", size = 220185, upload-time = "2025-08-23T14:40:42.026Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/f2/336d34d2fc1291ca7c18eeb46f64985e6cef5a1a7ef6d9c23720c6527289/coverage-7.10.5-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c177e6ffe2ebc7c410785307758ee21258aa8e8092b44d09a2da767834f075f2", size = 216890, upload-time = "2025-08-23T14:40:43.627Z" },
-    { url = "https://files.pythonhosted.org/packages/39/ea/92448b07cc1cf2b429d0ce635f59cf0c626a5d8de21358f11e92174ff2a6/coverage-7.10.5-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:14d6071c51ad0f703d6440827eaa46386169b5fdced42631d5a5ac419616046f", size = 217287, upload-time = "2025-08-23T14:40:45.214Z" },
-    { url = "https://files.pythonhosted.org/packages/96/ba/ad5b36537c5179c808d0ecdf6e4aa7630b311b3c12747ad624dcd43a9b6b/coverage-7.10.5-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:61f78c7c3bc272a410c5ae3fde7792b4ffb4acc03d35a7df73ca8978826bb7ab", size = 247683, upload-time = "2025-08-23T14:40:46.791Z" },
-    { url = "https://files.pythonhosted.org/packages/28/e5/fe3bbc8d097029d284b5fb305b38bb3404895da48495f05bff025df62770/coverage-7.10.5-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f39071caa126f69d63f99b324fb08c7b1da2ec28cbb1fe7b5b1799926492f65c", size = 249614, upload-time = "2025-08-23T14:40:48.082Z" },
-    { url = "https://files.pythonhosted.org/packages/69/9c/a1c89a8c8712799efccb32cd0a1ee88e452f0c13a006b65bb2271f1ac767/coverage-7.10.5-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:343a023193f04d46edc46b2616cdbee68c94dd10208ecd3adc56fcc54ef2baa1", size = 251719, upload-time = "2025-08-23T14:40:49.349Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/be/5576b5625865aa95b5633315f8f4142b003a70c3d96e76f04487c3b5cc95/coverage-7.10.5-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:585ffe93ae5894d1ebdee69fc0b0d4b7c75d8007983692fb300ac98eed146f78", size = 249411, upload-time = "2025-08-23T14:40:50.624Z" },
-    { url = "https://files.pythonhosted.org/packages/94/0a/e39a113d4209da0dbbc9385608cdb1b0726a4d25f78672dc51c97cfea80f/coverage-7.10.5-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:b0ef4e66f006ed181df29b59921bd8fc7ed7cd6a9289295cd8b2824b49b570df", size = 247466, upload-time = "2025-08-23T14:40:52.362Z" },
-    { url = "https://files.pythonhosted.org/packages/40/cb/aebb2d8c9e3533ee340bea19b71c5b76605a0268aa49808e26fe96ec0a07/coverage-7.10.5-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:eb7b0bbf7cc1d0453b843eca7b5fa017874735bef9bfdfa4121373d2cc885ed6", size = 248104, upload-time = "2025-08-23T14:40:54.064Z" },
-    { url = "https://files.pythonhosted.org/packages/08/e6/26570d6ccce8ff5de912cbfd268e7f475f00597cb58da9991fa919c5e539/coverage-7.10.5-cp311-cp311-win32.whl", hash = "sha256:1d043a8a06987cc0c98516e57c4d3fc2c1591364831e9deb59c9e1b4937e8caf", size = 219327, upload-time = "2025-08-23T14:40:55.424Z" },
-    { url = "https://files.pythonhosted.org/packages/79/79/5f48525e366e518b36e66167e3b6e5db6fd54f63982500c6a5abb9d3dfbd/coverage-7.10.5-cp311-cp311-win_amd64.whl", hash = "sha256:fefafcca09c3ac56372ef64a40f5fe17c5592fab906e0fdffd09543f3012ba50", size = 220213, upload-time = "2025-08-23T14:40:56.724Z" },
-    { url = "https://files.pythonhosted.org/packages/40/3c/9058128b7b0bf333130c320b1eb1ae485623014a21ee196d68f7737f8610/coverage-7.10.5-cp311-cp311-win_arm64.whl", hash = "sha256:7e78b767da8b5fc5b2faa69bb001edafcd6f3995b42a331c53ef9572c55ceb82", size = 218893, upload-time = "2025-08-23T14:40:58.011Z" },
-    { url = "https://files.pythonhosted.org/packages/27/8e/40d75c7128f871ea0fd829d3e7e4a14460cad7c3826e3b472e6471ad05bd/coverage-7.10.5-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c2d05c7e73c60a4cecc7d9b60dbfd603b4ebc0adafaef371445b47d0f805c8a9", size = 217077, upload-time = "2025-08-23T14:40:59.329Z" },
-    { url = "https://files.pythonhosted.org/packages/18/a8/f333f4cf3fb5477a7f727b4d603a2eb5c3c5611c7fe01329c2e13b23b678/coverage-7.10.5-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:32ddaa3b2c509778ed5373b177eb2bf5662405493baeff52278a0b4f9415188b", size = 217310, upload-time = "2025-08-23T14:41:00.628Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/2c/fbecd8381e0a07d1547922be819b4543a901402f63930313a519b937c668/coverage-7.10.5-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:dd382410039fe062097aa0292ab6335a3f1e7af7bba2ef8d27dcda484918f20c", size = 248802, upload-time = "2025-08-23T14:41:02.012Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/bc/1011da599b414fb6c9c0f34086736126f9ff71f841755786a6b87601b088/coverage-7.10.5-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7fa22800f3908df31cea6fb230f20ac49e343515d968cc3a42b30d5c3ebf9b5a", size = 251550, upload-time = "2025-08-23T14:41:03.438Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/6f/b5c03c0c721c067d21bc697accc3642f3cef9f087dac429c918c37a37437/coverage-7.10.5-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f366a57ac81f5e12797136552f5b7502fa053c861a009b91b80ed51f2ce651c6", size = 252684, upload-time = "2025-08-23T14:41:04.85Z" },
-    { url = "https://files.pythonhosted.org/packages/f9/50/d474bc300ebcb6a38a1047d5c465a227605d6473e49b4e0d793102312bc5/coverage-7.10.5-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5f1dc8f1980a272ad4a6c84cba7981792344dad33bf5869361576b7aef42733a", size = 250602, upload-time = "2025-08-23T14:41:06.719Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/2d/548c8e04249cbba3aba6bd799efdd11eee3941b70253733f5d355d689559/coverage-7.10.5-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:2285c04ee8676f7938b02b4936d9b9b672064daab3187c20f73a55f3d70e6b4a", size = 248724, upload-time = "2025-08-23T14:41:08.429Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/96/a7c3c0562266ac39dcad271d0eec8fc20ab576e3e2f64130a845ad2a557b/coverage-7.10.5-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:c2492e4dd9daab63f5f56286f8a04c51323d237631eb98505d87e4c4ff19ec34", size = 250158, upload-time = "2025-08-23T14:41:09.749Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/75/74d4be58c70c42ef0b352d597b022baf12dbe2b43e7cb1525f56a0fb1d4b/coverage-7.10.5-cp312-cp312-win32.whl", hash = "sha256:38a9109c4ee8135d5df5505384fc2f20287a47ccbe0b3f04c53c9a1989c2bbaf", size = 219493, upload-time = "2025-08-23T14:41:11.095Z" },
-    { url = "https://files.pythonhosted.org/packages/4f/08/364e6012d1d4d09d1e27437382967efed971d7613f94bca9add25f0c1f2b/coverage-7.10.5-cp312-cp312-win_amd64.whl", hash = "sha256:6b87f1ad60b30bc3c43c66afa7db6b22a3109902e28c5094957626a0143a001f", size = 220302, upload-time = "2025-08-23T14:41:12.449Z" },
-    { url = "https://files.pythonhosted.org/packages/db/d5/7c8a365e1f7355c58af4fe5faf3f90cc8e587590f5854808d17ccb4e7077/coverage-7.10.5-cp312-cp312-win_arm64.whl", hash = "sha256:672a6c1da5aea6c629819a0e1461e89d244f78d7b60c424ecf4f1f2556c041d8", size = 218936, upload-time = "2025-08-23T14:41:13.872Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/21/05248e8bc74683488cb7477e6b6b878decadd15af0ec96f56381d3d7ff2d/coverage-7.10.5-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:62835c1b00c4a4ace24c1a88561a5a59b612fbb83a525d1c70ff5720c97c0610", size = 216763, upload-time = "2025-08-23T14:42:26.75Z" },
-    { url = "https://files.pythonhosted.org/packages/a9/7f/161a0ad40cb1c7e19dc1aae106d3430cc88dac3d651796d6cf3f3730c800/coverage-7.10.5-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:5255b3bbcc1d32a4069d6403820ac8e6dbcc1d68cb28a60a1ebf17e47028e898", size = 217154, upload-time = "2025-08-23T14:42:28.238Z" },
-    { url = "https://files.pythonhosted.org/packages/de/31/41929ee53af829ea5a88e71d335ea09d0bb587a3da1c5e58e59b48473ed8/coverage-7.10.5-cp39-cp39-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:3876385722e335d6e991c430302c24251ef9c2a9701b2b390f5473199b1b8ebf", size = 243588, upload-time = "2025-08-23T14:42:29.798Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/4e/2649344e33eeb3567041e8255a1942173cae81817fe06b60f3fafaafe111/coverage-7.10.5-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:8048ce4b149c93447a55d279078c8ae98b08a6951a3c4d2d7e87f4efc7bfe100", size = 245412, upload-time = "2025-08-23T14:42:31.296Z" },
-    { url = "https://files.pythonhosted.org/packages/ac/b1/b21e1e69986ad89b051dd42c3ef06d9326e03ac3c0c844fc33385d1d9e35/coverage-7.10.5-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4028e7558e268dd8bcf4d9484aad393cafa654c24b4885f6f9474bf53183a82a", size = 247182, upload-time = "2025-08-23T14:42:33.155Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/b5/80837be411ae092e03fcc2a7877bd9a659c531eff50453e463057a9eee44/coverage-7.10.5-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:03f47dc870eec0367fcdd603ca6a01517d2504e83dc18dbfafae37faec66129a", size = 245066, upload-time = "2025-08-23T14:42:34.754Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/ed/fcb0838ddf149d68d09f89af57397b0dd9d26b100cc729daf1b0caf0b2d3/coverage-7.10.5-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:2d488d7d42b6ded7ea0704884f89dcabd2619505457de8fc9a6011c62106f6e5", size = 243138, upload-time = "2025-08-23T14:42:36.311Z" },
-    { url = "https://files.pythonhosted.org/packages/75/0f/505c6af24a9ae5d8919d209b9c31b7092815f468fa43bec3b1118232c62a/coverage-7.10.5-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:b3dcf2ead47fa8be14224ee817dfc1df98043af568fe120a22f81c0eb3c34ad2", size = 244095, upload-time = "2025-08-23T14:42:38.227Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/7e/c82a8bede46217c1d944bd19b65e7106633b998640f00ab49c5f747a5844/coverage-7.10.5-cp39-cp39-win32.whl", hash = "sha256:02650a11324b80057b8c9c29487020073d5e98a498f1857f37e3f9b6ea1b2426", size = 219289, upload-time = "2025-08-23T14:42:39.827Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/ac/46645ef6be543f2e7de08cc2601a0b67e130c816be3b749ab741be689fb9/coverage-7.10.5-cp39-cp39-win_amd64.whl", hash = "sha256:b45264dd450a10f9e03237b41a9a24e85cbb1e278e5a32adb1a303f58f0017f3", size = 220199, upload-time = "2025-08-23T14:42:41.363Z" },
-    { url = "https://files.pythonhosted.org/packages/08/b6/fff6609354deba9aeec466e4bcaeb9d1ed3e5d60b14b57df2a36fb2273f2/coverage-7.10.5-py3-none-any.whl", hash = "sha256:0be24d35e4db1d23d0db5c0f6a74a962e2ec83c426b5cac09f4234aadef38e4a", size = 208736, upload-time = "2025-08-23T14:42:43.145Z" },
+version = "7.10.6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/14/70/025b179c993f019105b79575ac6edb5e084fb0f0e63f15cdebef4e454fb5/coverage-7.10.6.tar.gz", hash = "sha256:f644a3ae5933a552a29dbb9aa2f90c677a875f80ebea028e5a52a4f429044b90", size = 823736, upload-time = "2025-08-29T15:35:16.668Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a8/1d/2e64b43d978b5bd184e0756a41415597dfef30fcbd90b747474bd749d45f/coverage-7.10.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:70e7bfbd57126b5554aa482691145f798d7df77489a177a6bef80de78860a356", size = 217025, upload-time = "2025-08-29T15:32:57.169Z" },
+    { url = "https://files.pythonhosted.org/packages/23/62/b1e0f513417c02cc10ef735c3ee5186df55f190f70498b3702d516aad06f/coverage-7.10.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e41be6f0f19da64af13403e52f2dec38bbc2937af54df8ecef10850ff8d35301", size = 217419, upload-time = "2025-08-29T15:32:59.908Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/16/b800640b7a43e7c538429e4d7223e0a94fd72453a1a048f70bf766f12e96/coverage-7.10.6-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:c61fc91ab80b23f5fddbee342d19662f3d3328173229caded831aa0bd7595460", size = 244180, upload-time = "2025-08-29T15:33:01.608Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/6f/5e03631c3305cad187eaf76af0b559fff88af9a0b0c180d006fb02413d7a/coverage-7.10.6-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:10356fdd33a7cc06e8051413140bbdc6f972137508a3572e3f59f805cd2832fd", size = 245992, upload-time = "2025-08-29T15:33:03.239Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/a1/f30ea0fb400b080730125b490771ec62b3375789f90af0bb68bfb8a921d7/coverage-7.10.6-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:80b1695cf7c5ebe7b44bf2521221b9bb8cdf69b1f24231149a7e3eb1ae5fa2fb", size = 247851, upload-time = "2025-08-29T15:33:04.603Z" },
+    { url = "https://files.pythonhosted.org/packages/02/8e/cfa8fee8e8ef9a6bb76c7bef039f3302f44e615d2194161a21d3d83ac2e9/coverage-7.10.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:2e4c33e6378b9d52d3454bd08847a8651f4ed23ddbb4a0520227bd346382bbc6", size = 245891, upload-time = "2025-08-29T15:33:06.176Z" },
+    { url = "https://files.pythonhosted.org/packages/93/a9/51be09b75c55c4f6c16d8d73a6a1d46ad764acca0eab48fa2ffaef5958fe/coverage-7.10.6-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:c8a3ec16e34ef980a46f60dc6ad86ec60f763c3f2fa0db6d261e6e754f72e945", size = 243909, upload-time = "2025-08-29T15:33:07.74Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/a6/ba188b376529ce36483b2d585ca7bdac64aacbe5aa10da5978029a9c94db/coverage-7.10.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:7d79dabc0a56f5af990cc6da9ad1e40766e82773c075f09cc571e2076fef882e", size = 244786, upload-time = "2025-08-29T15:33:08.965Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/4c/37ed872374a21813e0d3215256180c9a382c3f5ced6f2e5da0102fc2fd3e/coverage-7.10.6-cp310-cp310-win32.whl", hash = "sha256:86b9b59f2b16e981906e9d6383eb6446d5b46c278460ae2c36487667717eccf1", size = 219521, upload-time = "2025-08-29T15:33:10.599Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/36/9311352fdc551dec5b973b61f4e453227ce482985a9368305880af4f85dd/coverage-7.10.6-cp310-cp310-win_amd64.whl", hash = "sha256:e132b9152749bd33534e5bd8565c7576f135f157b4029b975e15ee184325f528", size = 220417, upload-time = "2025-08-29T15:33:11.907Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/16/2bea27e212c4980753d6d563a0803c150edeaaddb0771a50d2afc410a261/coverage-7.10.6-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c706db3cabb7ceef779de68270150665e710b46d56372455cd741184f3868d8f", size = 217129, upload-time = "2025-08-29T15:33:13.575Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/51/e7159e068831ab37e31aac0969d47b8c5ee25b7d307b51e310ec34869315/coverage-7.10.6-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:8e0c38dc289e0508ef68ec95834cb5d2e96fdbe792eaccaa1bccac3966bbadcc", size = 217532, upload-time = "2025-08-29T15:33:14.872Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/c0/246ccbea53d6099325d25cd208df94ea435cd55f0db38099dd721efc7a1f/coverage-7.10.6-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:752a3005a1ded28f2f3a6e8787e24f28d6abe176ca64677bcd8d53d6fe2ec08a", size = 247931, upload-time = "2025-08-29T15:33:16.142Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/fb/7435ef8ab9b2594a6e3f58505cc30e98ae8b33265d844007737946c59389/coverage-7.10.6-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:689920ecfd60f992cafca4f5477d55720466ad2c7fa29bb56ac8d44a1ac2b47a", size = 249864, upload-time = "2025-08-29T15:33:17.434Z" },
+    { url = "https://files.pythonhosted.org/packages/51/f8/d9d64e8da7bcddb094d511154824038833c81e3a039020a9d6539bf303e9/coverage-7.10.6-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ec98435796d2624d6905820a42f82149ee9fc4f2d45c2c5bc5a44481cc50db62", size = 251969, upload-time = "2025-08-29T15:33:18.822Z" },
+    { url = "https://files.pythonhosted.org/packages/43/28/c43ba0ef19f446d6463c751315140d8f2a521e04c3e79e5c5fe211bfa430/coverage-7.10.6-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b37201ce4a458c7a758ecc4efa92fa8ed783c66e0fa3c42ae19fc454a0792153", size = 249659, upload-time = "2025-08-29T15:33:20.407Z" },
+    { url = "https://files.pythonhosted.org/packages/79/3e/53635bd0b72beaacf265784508a0b386defc9ab7fad99ff95f79ce9db555/coverage-7.10.6-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:2904271c80898663c810a6b067920a61dd8d38341244a3605bd31ab55250dad5", size = 247714, upload-time = "2025-08-29T15:33:21.751Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/55/0964aa87126624e8c159e32b0bc4e84edef78c89a1a4b924d28dd8265625/coverage-7.10.6-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:5aea98383463d6e1fa4e95416d8de66f2d0cb588774ee20ae1b28df826bcb619", size = 248351, upload-time = "2025-08-29T15:33:23.105Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/ab/6cfa9dc518c6c8e14a691c54e53a9433ba67336c760607e299bfcf520cb1/coverage-7.10.6-cp311-cp311-win32.whl", hash = "sha256:e3fb1fa01d3598002777dd259c0c2e6d9d5e10e7222976fc8e03992f972a2cba", size = 219562, upload-time = "2025-08-29T15:33:24.717Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/18/99b25346690cbc55922e7cfef06d755d4abee803ef335baff0014268eff4/coverage-7.10.6-cp311-cp311-win_amd64.whl", hash = "sha256:f35ed9d945bece26553d5b4c8630453169672bea0050a564456eb88bdffd927e", size = 220453, upload-time = "2025-08-29T15:33:26.482Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/ed/81d86648a07ccb124a5cf1f1a7788712b8d7216b593562683cd5c9b0d2c1/coverage-7.10.6-cp311-cp311-win_arm64.whl", hash = "sha256:99e1a305c7765631d74b98bf7dbf54eeea931f975e80f115437d23848ee8c27c", size = 219127, upload-time = "2025-08-29T15:33:27.777Z" },
+    { url = "https://files.pythonhosted.org/packages/26/06/263f3305c97ad78aab066d116b52250dd316e74fcc20c197b61e07eb391a/coverage-7.10.6-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:5b2dd6059938063a2c9fee1af729d4f2af28fd1a545e9b7652861f0d752ebcea", size = 217324, upload-time = "2025-08-29T15:33:29.06Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/60/1e1ded9a4fe80d843d7d53b3e395c1db3ff32d6c301e501f393b2e6c1c1f/coverage-7.10.6-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:388d80e56191bf846c485c14ae2bc8898aa3124d9d35903fef7d907780477634", size = 217560, upload-time = "2025-08-29T15:33:30.748Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/25/52136173c14e26dfed8b106ed725811bb53c30b896d04d28d74cb64318b3/coverage-7.10.6-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:90cb5b1a4670662719591aa92d0095bb41714970c0b065b02a2610172dbf0af6", size = 249053, upload-time = "2025-08-29T15:33:32.041Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/1d/ae25a7dc58fcce8b172d42ffe5313fc267afe61c97fa872b80ee72d9515a/coverage-7.10.6-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:961834e2f2b863a0e14260a9a273aff07ff7818ab6e66d2addf5628590c628f9", size = 251802, upload-time = "2025-08-29T15:33:33.625Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/7a/1f561d47743710fe996957ed7c124b421320f150f1d38523d8d9102d3e2a/coverage-7.10.6-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bf9a19f5012dab774628491659646335b1928cfc931bf8d97b0d5918dd58033c", size = 252935, upload-time = "2025-08-29T15:33:34.909Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/ad/8b97cd5d28aecdfde792dcbf646bac141167a5cacae2cd775998b45fabb5/coverage-7.10.6-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:99c4283e2a0e147b9c9cc6bc9c96124de9419d6044837e9799763a0e29a7321a", size = 250855, upload-time = "2025-08-29T15:33:36.922Z" },
+    { url = "https://files.pythonhosted.org/packages/33/6a/95c32b558d9a61858ff9d79580d3877df3eb5bc9eed0941b1f187c89e143/coverage-7.10.6-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:282b1b20f45df57cc508c1e033403f02283adfb67d4c9c35a90281d81e5c52c5", size = 248974, upload-time = "2025-08-29T15:33:38.175Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/9c/8ce95dee640a38e760d5b747c10913e7a06554704d60b41e73fdea6a1ffd/coverage-7.10.6-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:8cdbe264f11afd69841bd8c0d83ca10b5b32853263ee62e6ac6a0ab63895f972", size = 250409, upload-time = "2025-08-29T15:33:39.447Z" },
+    { url = "https://files.pythonhosted.org/packages/04/12/7a55b0bdde78a98e2eb2356771fd2dcddb96579e8342bb52aa5bc52e96f0/coverage-7.10.6-cp312-cp312-win32.whl", hash = "sha256:a517feaf3a0a3eca1ee985d8373135cfdedfbba3882a5eab4362bda7c7cf518d", size = 219724, upload-time = "2025-08-29T15:33:41.172Z" },
+    { url = "https://files.pythonhosted.org/packages/36/4a/32b185b8b8e327802c9efce3d3108d2fe2d9d31f153a0f7ecfd59c773705/coverage-7.10.6-cp312-cp312-win_amd64.whl", hash = "sha256:856986eadf41f52b214176d894a7de05331117f6035a28ac0016c0f63d887629", size = 220536, upload-time = "2025-08-29T15:33:42.524Z" },
+    { url = "https://files.pythonhosted.org/packages/08/3a/d5d8dc703e4998038c3099eaf77adddb00536a3cec08c8dcd556a36a3eb4/coverage-7.10.6-cp312-cp312-win_arm64.whl", hash = "sha256:acf36b8268785aad739443fa2780c16260ee3fa09d12b3a70f772ef100939d80", size = 219171, upload-time = "2025-08-29T15:33:43.974Z" },
+    { url = "https://files.pythonhosted.org/packages/91/70/f73ad83b1d2fd2d5825ac58c8f551193433a7deaf9b0d00a8b69ef61cd9a/coverage-7.10.6-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:90558c35af64971d65fbd935c32010f9a2f52776103a259f1dee865fe8259352", size = 217009, upload-time = "2025-08-29T15:34:57.381Z" },
+    { url = "https://files.pythonhosted.org/packages/01/e8/099b55cd48922abbd4b01ddd9ffa352408614413ebfc965501e981aced6b/coverage-7.10.6-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:8953746d371e5695405806c46d705a3cd170b9cc2b9f93953ad838f6c1e58612", size = 217400, upload-time = "2025-08-29T15:34:58.985Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/d1/c6bac7c9e1003110a318636fef3b5c039df57ab44abcc41d43262a163c28/coverage-7.10.6-cp39-cp39-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:c83f6afb480eae0313114297d29d7c295670a41c11b274e6bca0c64540c1ce7b", size = 243835, upload-time = "2025-08-29T15:35:00.541Z" },
+    { url = "https://files.pythonhosted.org/packages/01/f9/82c6c061838afbd2172e773156c0aa84a901d59211b4975a4e93accf5c89/coverage-7.10.6-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7eb68d356ba0cc158ca535ce1381dbf2037fa8cb5b1ae5ddfc302e7317d04144", size = 245658, upload-time = "2025-08-29T15:35:02.135Z" },
+    { url = "https://files.pythonhosted.org/packages/81/6a/35674445b1d38161148558a3ff51b0aa7f0b54b1def3abe3fbd34efe05bc/coverage-7.10.6-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5b15a87265e96307482746d86995f4bff282f14b027db75469c446da6127433b", size = 247433, upload-time = "2025-08-29T15:35:03.777Z" },
+    { url = "https://files.pythonhosted.org/packages/18/27/98c99e7cafb288730a93535092eb433b5503d529869791681c4f2e2012a8/coverage-7.10.6-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:fc53ba868875bfbb66ee447d64d6413c2db91fddcfca57025a0e7ab5b07d5862", size = 245315, upload-time = "2025-08-29T15:35:05.629Z" },
+    { url = "https://files.pythonhosted.org/packages/09/05/123e0dba812408c719c319dea05782433246f7aa7b67e60402d90e847545/coverage-7.10.6-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:efeda443000aa23f276f4df973cb82beca682fd800bb119d19e80504ffe53ec2", size = 243385, upload-time = "2025-08-29T15:35:07.494Z" },
+    { url = "https://files.pythonhosted.org/packages/67/52/d57a42502aef05c6325f28e2e81216c2d9b489040132c18725b7a04d1448/coverage-7.10.6-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:9702b59d582ff1e184945d8b501ffdd08d2cee38d93a2206aa5f1365ce0b8d78", size = 244343, upload-time = "2025-08-29T15:35:09.55Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/22/7f6fad7dbb37cf99b542c5e157d463bd96b797078b1ec506691bc836f476/coverage-7.10.6-cp39-cp39-win32.whl", hash = "sha256:2195f8e16ba1a44651ca684db2ea2b2d4b5345da12f07d9c22a395202a05b23c", size = 219530, upload-time = "2025-08-29T15:35:11.167Z" },
+    { url = "https://files.pythonhosted.org/packages/62/30/e2fda29bfe335026027e11e6a5e57a764c9df13127b5cf42af4c3e99b937/coverage-7.10.6-cp39-cp39-win_amd64.whl", hash = "sha256:f32ff80e7ef6a5b5b606ea69a36e97b219cd9dc799bcf2963018a4d8f788cfbf", size = 220432, upload-time = "2025-08-29T15:35:12.902Z" },
+    { url = "https://files.pythonhosted.org/packages/44/0c/50db5379b615854b5cf89146f8f5bd1d5a9693d7f3a987e269693521c404/coverage-7.10.6-py3-none-any.whl", hash = "sha256:92c4ecf6bf11b2e85fd4d8204814dc26e6a19f0c9d938c207c5cb0eadfcabbe3", size = 208986, upload-time = "2025-08-29T15:35:14.506Z" },
 ]
 
 [package.optional-dependencies]
@@ -700,7 +722,7 @@ dependencies = [
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "msgpack", marker = "python_full_version == '3.10.*'" },
     { name = "optax", version = "0.2.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "orbax-checkpoint", version = "0.11.23", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "orbax-checkpoint", version = "0.11.24", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "pyyaml", marker = "python_full_version == '3.10.*'" },
     { name = "rich", marker = "python_full_version == '3.10.*'" },
     { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
@@ -725,7 +747,7 @@ dependencies = [
     { name = "msgpack", marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "optax", version = "0.2.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "orbax-checkpoint", version = "0.11.23", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "orbax-checkpoint", version = "0.11.24", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "pyyaml", marker = "python_full_version >= '3.11'" },
     { name = "rich", marker = "python_full_version >= '3.11'" },
     { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
@@ -1448,11 +1470,11 @@ wheels = [
 
 [[package]]
 name = "latex2mathml"
-version = "3.78.0"
+version = "3.78.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/69/33/ad2c3929494ad160f5130ea132ca298627a6c81c70be6bedd1bc806b5b01/latex2mathml-3.78.0.tar.gz", hash = "sha256:712193aa4c6ade1a8e0145dac7bc1f9aafbd54f93046a2356a7e1c05fa0f8b31", size = 73737, upload-time = "2025-05-03T16:51:53.563Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/1a/26/57b1034c08922d0aefea79430a5e0006ffaee4f0ec59d566613f667ab2f7/latex2mathml-3.78.1.tar.gz", hash = "sha256:f941db80bf41db33f31df87b304e8b588f8166b813b0257c11c98f7a9d0aac71", size = 74030, upload-time = "2025-08-29T23:34:23.178Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/1e/fd/aba08bb9e527168efad57985d7db9a853eb2384b1efa5ca5f3a3794c9cef/latex2mathml-3.78.0-py3-none-any.whl", hash = "sha256:1aeca3dc027b3006ad7b301b7f4a15ffbb4c1451e3dc8c3389e97b37b497e1d6", size = 73673, upload-time = "2025-05-03T16:51:51.991Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/76/d661ea2e529c3d464f9efd73f9ac31626b45279eb4306e684054ea20e3d4/latex2mathml-3.78.1-py3-none-any.whl", hash = "sha256:f089b6d75e85b937f99693c93e8c16c0804008672c3dd2a3d25affd36f238100", size = 73892, upload-time = "2025-08-29T23:34:21.98Z" },
 ]
 
 [[package]]
@@ -1770,33 +1792,33 @@ wheels = [
 
 [[package]]
 name = "mlx"
-version = "0.28.0"
+version = "0.29.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "mlx-metal", marker = "sys_platform == 'darwin'" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/1e/ac/3d78ad9a4625b2684ad4356958d5353c33e06eeb54cf6c625bdb167044b9/mlx-0.28.0-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:5167b4c8d4aae9361afb5d770c75b653d66aa0ebbb8b0099ddd6da5d55d8e6b6", size = 565787, upload-time = "2025-08-07T07:50:42.798Z" },
-    { url = "https://files.pythonhosted.org/packages/61/16/fd564926a3e2b96ee339933a46d3e47db9177c038a028916bb43cc0fefba/mlx-0.28.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:168e945bd0e6de19fd0d73ecd0123ba99108342347be2c5aacd387cf1c889310", size = 540079, upload-time = "2025-08-07T07:49:07.185Z" },
-    { url = "https://files.pythonhosted.org/packages/15/f6/9cfb7e021992f7c8ece0bff6684b585acae0513a11c70fb14b0271bec724/mlx-0.28.0-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:74e7069c12b3f6aef0f75dc8ec5c76248c4776b86b6ee64a0f28d699b2412a1a", size = 540078, upload-time = "2025-08-07T07:50:47.728Z" },
-    { url = "https://files.pythonhosted.org/packages/42/dd/3b7adf81affa0b3336a8b28088bb69d7f2459ba9f66d1db864a6d204f3e6/mlx-0.28.0-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:e1e4734ac951ddc64bdc635417ea5ab98feeac2a2791e88759a20e7bf107e30e", size = 636545, upload-time = "2025-08-07T07:52:30.117Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/e1/c4a654b58af8af5ed7605a47a513de21716210adafac956bfff84b9e4df1/mlx-0.28.0-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:f969f50da4e709f21240765347cf28b4fcb721d21eed5801d895c694eb315e9f", size = 565785, upload-time = "2025-08-07T07:50:30.893Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/6f/82c0a988deb13a303cc8c8b6c2c34c9ab21c634a4b02b405a3a6731ffe98/mlx-0.28.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:a19046afe989565ce383d82d9b87bb539f355be75765d00b51ff8a4c7c038104", size = 540283, upload-time = "2025-08-07T07:48:57.715Z" },
-    { url = "https://files.pythonhosted.org/packages/22/c6/bbc28fbd10ddad7727d239a29bfc0a0c9f9258ae6762cec8ee50b6d7e939/mlx-0.28.0-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:be3cbc4bf21debed9b6f7a3755b41b007333d289636a3fd4ea9040b0eb20e22c", size = 540282, upload-time = "2025-08-07T07:50:48.883Z" },
-    { url = "https://files.pythonhosted.org/packages/86/24/58b6e434dac2bc384d24a7e44bda86c8f0e1308421e137e4b0583ec40f28/mlx-0.28.0-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:bd37c1ee9fbf91879e02bb68c874061badc893509567c3392ee7be91414a6dba", size = 636553, upload-time = "2025-08-07T07:52:29.75Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/c9/d12ed6a8393450e28eb1f552b50200f83f138b1268b5f4e8074a76d745a2/mlx-0.28.0-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:97866d5e454e8f2d7bc42aadcbfd7565d40f4755564785e4fb964812fbad604b", size = 564160, upload-time = "2025-08-07T07:50:34.652Z" },
-    { url = "https://files.pythonhosted.org/packages/71/4f/3951766a5edb75c0d2d860381f592d271b4c3b7241e730e78dd63926f5b4/mlx-0.28.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:5204ebf399439e5da374295f6c1b6961355824604eed7026c18edfe4c83e9243", size = 540098, upload-time = "2025-08-07T07:50:52.67Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/52/cb8eb03544eace055a500bd4a3b776a3ce48198d7b7b398e21a5a3256e89/mlx-0.28.0-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:34776bd3fe97bca7c6c76d77f6104e0d6b05b3626bb3cf9ed48d3a9bbd46c180", size = 540100, upload-time = "2025-08-07T07:50:49.095Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/23/cb2703724f011d9aabc5a5a8ddf20481fdbdf251227468de89885b71f0ed/mlx-0.28.0-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:ead9a6c13b704239b5ca50d68f5effc505c8f15a6017f35d8b4d0e25832e29dd", size = 632943, upload-time = "2025-08-07T07:52:35.695Z" },
-    { url = "https://files.pythonhosted.org/packages/cc/e5/f1edc60f2176f8600e4a040c9c92ba564030298199830ae0982c95c1fb8b/mlx-0.28.0-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:9a954e5f8f0c3c92d76e679162c8eb8e4618f8d3662cbd41e9b5cbdc087a298b", size = 566000, upload-time = "2025-08-07T07:53:04.304Z" },
-    { url = "https://files.pythonhosted.org/packages/56/8d/35dee5df087f6e313754c9ef15eadf48de21bff6030139b1af662f5ae763/mlx-0.28.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:56eaa39961c3ef23d4915466380d80407221134d12c0eef8d2391e55bcfe8802", size = 540561, upload-time = "2025-08-07T07:52:58.193Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/dd/54d8e9e82656537938fc8bc903352d731f5f1d025e8601354e0aed1ba692/mlx-0.28.0-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:3bc6dbe0d6b42c5bfcd783632b2217483fccb9994f7ce3b864d2b9ad2dfea7aa", size = 540561, upload-time = "2025-08-07T07:53:02.354Z" },
-    { url = "https://files.pythonhosted.org/packages/77/b2/254057fb1dce8070e789f14a724a514679646b9045b18f6c83435bc33f61/mlx-0.28.0-cp39-cp39-manylinux_2_35_x86_64.whl", hash = "sha256:f7903324346ac69c60f7e316624779c2ddf1ad41dd0c9670f978126b88ac5c6a", size = 636768, upload-time = "2025-08-07T07:55:33.109Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/13/da859e935a0f4179c4e599afe9f4add8d594e0433ac131460746f6a7cbdf/mlx-0.29.0-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:430b85d5444247b80417fd3a2b966676af43549f0926095bac25866b2154c9f0", size = 545687, upload-time = "2025-08-29T17:13:54.532Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/77/3c6d307c62cec555c2c6e81a978b41c3db9d1244653578df6159c7872ef8/mlx-0.29.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:afe33a3feefc62622bc71dd10c4e7e75bc8278780351fd82f1c33145b072d88b", size = 545686, upload-time = "2025-08-29T17:14:20.195Z" },
+    { url = "https://files.pythonhosted.org/packages/11/42/90b5d0d52ed733b57de7fbde2d972ecd20b819553a36bd8a4b78750539d6/mlx-0.29.0-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:4e055878e54afb6ac258c145674d59682c8c565e3fd1dde5e28ca3ebc52d3434", size = 545689, upload-time = "2025-08-29T17:14:14.423Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/03/d5186edf170e7866c4bf37baa50c8e0fc26b6d8668f9fa79f34d23e66ed6/mlx-0.29.0-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:03a738104b50c38e939ba8037b6a04ee937ebd4e77a9792f1f922d021cad0010", size = 650149, upload-time = "2025-08-29T17:19:36.852Z" },
+    { url = "https://files.pythonhosted.org/packages/00/b3/490ae792034a95895a70d04feadb1158786d96605a45555c1cae0a1158d3/mlx-0.29.0-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:96d49e76a152a202c146ce4b71538b59c9eb6c98b40557ee26a3cf9753b5fb34", size = 545948, upload-time = "2025-08-29T17:14:52.863Z" },
+    { url = "https://files.pythonhosted.org/packages/57/00/277ba0f1f384275d122b0af4ebc5866a452689e3b0144d03b46a549a7142/mlx-0.29.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:05c390811f6514013535b5fea9fa33b63daa8cdc17881deadca7fbca38e42c93", size = 545950, upload-time = "2025-08-29T17:14:35.294Z" },
+    { url = "https://files.pythonhosted.org/packages/de/a3/337461b75168cf701a1190eabd331de4fb752dc9dc5759fbede76f744aab/mlx-0.29.0-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:7800f87fc5eb9861c6444b0bb38175c8f0b12d48746a980264fe744f6eb53a50", size = 545946, upload-time = "2025-08-29T17:13:45.738Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/57/d1d0cdd5d194ec8b7b55062407d8e04bcdf62de73c2ef0db075fba0ce7f3/mlx-0.29.0-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:6f2f969c6ad6259f4422c7cd7bfcb49705679585991109220de0b758feaedc25", size = 650033, upload-time = "2025-08-29T17:18:23.646Z" },
+    { url = "https://files.pythonhosted.org/packages/60/c0/a95f24f4b78fdf96f57aaa1a8919f5688603fd4fa0294c2f868f60bc98af/mlx-0.29.0-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:a9f9f760a179d96fa7fecc0b31a9885a9ab4ce9e2914f7f34c2980edd7f012fa", size = 546443, upload-time = "2025-08-29T17:14:17.997Z" },
+    { url = "https://files.pythonhosted.org/packages/61/c7/9f6a5cf1a0b0eb1ace3892c930f64fa84351cecc75221ce4ca5abd0ad64d/mlx-0.29.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:446ea784a4717c9b06eee9c91ee038245ebb231760a2cf61b677dc3d256cbd83", size = 546442, upload-time = "2025-08-29T17:14:19.035Z" },
+    { url = "https://files.pythonhosted.org/packages/85/40/32b1fe76d80f559ff3af18d104a22c4d07ecad62a88e3c313083722868e4/mlx-0.29.0-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:593a3d71c5859f83bc55a9bdc612fe5f8b495c14112e0e7a96c027c11ce5ab52", size = 546444, upload-time = "2025-08-29T17:14:58.946Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/dd/5e1c4672a551f475b5305e3e21da808aacb080a114af83a38e5b9551c212/mlx-0.29.0-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:a1174dd755da6a443bd34616b6ef6bf70718648420b9efafb37d1b20d70877d0", size = 645004, upload-time = "2025-08-29T17:18:10.974Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/78/1526222cb0fbb9a3ed53493416259c8816bbccb59e9c0ac72bbaa3b00731/mlx-0.29.0-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:e5b3aa72b8ab380e131189427275b8fa0a292bd96f7389d847915a621697824e", size = 546024, upload-time = "2025-08-29T17:16:37.472Z" },
+    { url = "https://files.pythonhosted.org/packages/78/23/5d2929d048eaa12caab2c38936e96205c174dd919bc6db835ab55b3526cb/mlx-0.29.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:b21447542d4082b9de5dfc14253a09574b9df6518133b38d111afbd68457b2e7", size = 546023, upload-time = "2025-08-29T17:16:34.408Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/16/53d7ad4e9b7d2c990440ea1574b2d1ddf8fae3af6c6c78596fd8ed2088d3/mlx-0.29.0-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:b883d3546bc1ef2484c9350ef04bbec685bda8e78f37a81b0520f7dbc5de369b", size = 546018, upload-time = "2025-08-29T17:15:27.79Z" },
+    { url = "https://files.pythonhosted.org/packages/37/1a/d10424c2ffdb7cdc6b1cccda7763d5b72b5ddb6483d8d32206fd76053e04/mlx-0.29.0-cp39-cp39-manylinux_2_35_x86_64.whl", hash = "sha256:ce2a1efc3556cfa2ad99ae17191c352a94648ab1f8ec22d370a94709f428de5b", size = 650315, upload-time = "2025-08-29T17:21:04.237Z" },
 ]
 
 [[package]]
 name = "mlx-lm"
-version = "0.26.4"
+version = "0.27.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jinja2" },
@@ -1807,19 +1829,19 @@ dependencies = [
     { name = "pyyaml" },
     { name = "transformers" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/88/20/f3af9d99a5ad6ac42419a3d381290a28bf6d9899ed517a7ccc9fea08546e/mlx_lm-0.26.4.tar.gz", hash = "sha256:1bf21ede1d2d7b660ae312868790df9d73a8553dc50655cf7ae867a36ebcc08c", size = 176384, upload-time = "2025-08-25T15:57:41.723Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/8c/4b/68a0b7cde71279ef47a3cbbbe663ea2007824d7ad93a19dc8a26ee68341a/mlx_lm-0.27.0.tar.gz", hash = "sha256:91da83708bc63e5985ff862bc1d66b1348e8af98baa8b7f873661661d8cabfd2", size = 178986, upload-time = "2025-08-29T18:00:07.404Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/de/6a/4d20d1b20cd690a3eeaf609c7cb9058f2d52c6d1081394f0d91bd12d08f7/mlx_lm-0.26.4-py3-none-any.whl", hash = "sha256:79bf3afb399ae3bb6073bf0fa6c04f33d70c831ccc6bbbc206c10567d4eef162", size = 242038, upload-time = "2025-08-25T15:57:40.181Z" },
+    { url = "https://files.pythonhosted.org/packages/82/9a/5d27af907b579495c5bbff3f42a4dbbe5834aaa62db1cd5f50465fa28a3d/mlx_lm-0.27.0-py3-none-any.whl", hash = "sha256:296e2fdd331db11fa14652161eca3622b2311439309b639d0832dec33fb47d55", size = 245622, upload-time = "2025-08-29T18:00:06.077Z" },
 ]
 
 [[package]]
 name = "mlx-metal"
-version = "0.28.0"
+version = "0.29.0"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f1/71/879284c71dfb12ded986a6532a4ab7df5c2794385ccf2766c1b40aee74cb/mlx_metal-0.28.0-py3-none-macosx_13_0_arm64.whl", hash = "sha256:ce08d40f1fad4f0b3bc87bfff5d603c7fe7dd141c082ba9ce9328b41e8f8d46b", size = 33840007, upload-time = "2025-08-07T07:53:07.437Z" },
-    { url = "https://files.pythonhosted.org/packages/06/90/44a261ccb9f6052c93c9da4faa4fc6d4f914938c51ecbbb68c546ab521b9/mlx_metal-0.28.0-py3-none-macosx_14_0_arm64.whl", hash = "sha256:424142ab843e2ac0b14edb58cf88d96723823c565291f46ddeeaa072abcc991e", size = 33196759, upload-time = "2025-08-07T07:52:59.436Z" },
-    { url = "https://files.pythonhosted.org/packages/72/59/8e4dee2893a56fc68a27eec7ec7ed9559c7ea01099313a9b8196373bf3cf/mlx_metal-0.28.0-py3-none-macosx_15_0_arm64.whl", hash = "sha256:214ece3781d44f57eb9686561594b28915ec5568df4a5a73da59c66880b204ed", size = 33167706, upload-time = "2025-08-07T07:53:03.852Z" },
+    { url = "https://files.pythonhosted.org/packages/08/28/3fb942b27d53061cacd12883629e5c48821f9e661f01210f55d89dabe3a6/mlx_metal-0.29.0-py3-none-macosx_13_0_arm64.whl", hash = "sha256:7e8279851504a8394122497d491276db37ef85d29f2f54d8a711ac90edfef7bf", size = 34964891, upload-time = "2025-08-29T17:16:35.72Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/82/582555232044bbdc0d74186077bcd79f1ce45f07a79e49b36afc25f9e313/mlx_metal-0.29.0-py3-none-macosx_14_0_arm64.whl", hash = "sha256:674fc7df4b9502dffe56c92cc95bae3a44950c70aa50f5b53291f228cfbbb631", size = 34694819, upload-time = "2025-08-29T17:16:32.636Z" },
+    { url = "https://files.pythonhosted.org/packages/30/d6/b3fc6d07b9bc814551078e53438ebfae8e44c47e4e5f762df244150e4e3a/mlx_metal-0.29.0-py3-none-macosx_15_0_arm64.whl", hash = "sha256:f69eb71128dcc55b5d9baa7606c31641bdd214b859d809deac51be4248d6fddb", size = 34685946, upload-time = "2025-08-29T17:15:25.028Z" },
 ]
 
 [[package]]
@@ -2437,7 +2459,7 @@ resolution-markers = [
 ]
 dependencies = [
     { name = "absl-py", marker = "python_full_version < '3.10'" },
-    { name = "chex", marker = "python_full_version < '3.10'" },
+    { name = "chex", version = "0.1.90", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "etils", version = "1.5.2", source = { registry = "https://pypi.org/simple" }, extra = ["epy"], marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
@@ -2459,7 +2481,8 @@ resolution-markers = [
 ]
 dependencies = [
     { name = "absl-py", marker = "python_full_version >= '3.10'" },
-    { name = "chex", marker = "python_full_version >= '3.10'" },
+    { name = "chex", version = "0.1.90", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "chex", version = "0.1.91", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "jaxlib", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
@@ -2558,7 +2581,7 @@ wheels = [
 
 [[package]]
 name = "orbax-checkpoint"
-version = "0.11.23"
+version = "0.11.24"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -2581,9 +2604,9 @@ dependencies = [
     { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/25/07/f533bb7a47de27c8666b00ebaeba597b1a725d6734c6f0e22d2e074a5d47/orbax_checkpoint-0.11.23.tar.gz", hash = "sha256:f59491f7c8c671cf255df7f467389e5317baba637ff9d57b2b4af83fe8e8526f", size = 366861, upload-time = "2025-08-18T20:13:19.555Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/52/24/44915f33cbea4cfd35f654a5ba01d248447bc007f1d049dd20bf58592820/orbax_checkpoint-0.11.24.tar.gz", hash = "sha256:4e7afe927d1ed6d8160bacf5ed4fef56c1370320e0ebdfda213c6351a2e3c0d0", size = 372123, upload-time = "2025-08-28T20:51:49.089Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/0e/fb/1f909c74aca659eb001f7416cc8a856b633d23ba84b2a88aa0890c14983b/orbax_checkpoint-0.11.23-py3-none-any.whl", hash = "sha256:db91516f574428cd6734042bbf380cf095fc297cefe00ec55601c68a3ecb60b5", size = 523142, upload-time = "2025-08-18T20:13:18.062Z" },
+    { url = "https://files.pythonhosted.org/packages/99/e8/acef62ec5c4b8658eacfce23d3f4e866d3cf5d07c509ca03eebb0c420a6c/orbax_checkpoint-0.11.24-py3-none-any.whl", hash = "sha256:a94178c9ba9fd3d6fd8fc511b6a0f34f7d89798bfdb79661e258cd32ada7650b", size = 529268, upload-time = "2025-08-28T20:51:47.917Z" },
 ]
 
 [[package]]
@@ -3540,68 +3563,68 @@ wheels = [
 
 [[package]]
 name = "regex"
-version = "2025.7.34"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/0b/de/e13fa6dc61d78b30ba47481f99933a3b49a57779d625c392d8036770a60d/regex-2025.7.34.tar.gz", hash = "sha256:9ead9765217afd04a86822dfcd4ed2747dfe426e887da413b15ff0ac2457e21a", size = 400714, upload-time = "2025-07-31T00:21:16.262Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/50/d2/0a44a9d92370e5e105f16669acf801b215107efea9dea4317fe96e9aad67/regex-2025.7.34-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:d856164d25e2b3b07b779bfed813eb4b6b6ce73c2fd818d46f47c1eb5cd79bd6", size = 484591, upload-time = "2025-07-31T00:18:46.675Z" },
-    { url = "https://files.pythonhosted.org/packages/2e/b1/00c4f83aa902f1048495de9f2f33638ce970ce1cf9447b477d272a0e22bb/regex-2025.7.34-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2d15a9da5fad793e35fb7be74eec450d968e05d2e294f3e0e77ab03fa7234a83", size = 289293, upload-time = "2025-07-31T00:18:53.069Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/b0/5bc5c8ddc418e8be5530b43ae1f7c9303f43aeff5f40185c4287cf6732f2/regex-2025.7.34-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:95b4639c77d414efa93c8de14ce3f7965a94d007e068a94f9d4997bb9bd9c81f", size = 285932, upload-time = "2025-07-31T00:18:54.673Z" },
-    { url = "https://files.pythonhosted.org/packages/46/c7/a1a28d050b23665a5e1eeb4d7f13b83ea86f0bc018da7b8f89f86ff7f094/regex-2025.7.34-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5d7de1ceed5a5f84f342ba4a9f4ae589524adf9744b2ee61b5da884b5b659834", size = 780361, upload-time = "2025-07-31T00:18:56.13Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/0d/82e7afe7b2c9fe3d488a6ab6145d1d97e55f822dfb9b4569aba2497e3d09/regex-2025.7.34-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:02e5860a250cd350c4933cf376c3bc9cb28948e2c96a8bc042aee7b985cfa26f", size = 849176, upload-time = "2025-07-31T00:18:57.483Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/16/3036e16903d8194f1490af457a7e33b06d9e9edd9576b1fe6c7ac660e9ed/regex-2025.7.34-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0a5966220b9a1a88691282b7e4350e9599cf65780ca60d914a798cb791aa1177", size = 897222, upload-time = "2025-07-31T00:18:58.721Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/c2/010e089ae00d31418e7d2c6601760eea1957cde12be719730c7133b8c165/regex-2025.7.34-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:48fb045bbd4aab2418dc1ba2088a5e32de4bfe64e1457b948bb328a8dc2f1c2e", size = 789831, upload-time = "2025-07-31T00:19:00.436Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/86/b312b7bf5c46d21dbd9a3fdc4a80fde56ea93c9c0b89cf401879635e094d/regex-2025.7.34-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:20ff8433fa45e131f7316594efe24d4679c5449c0ca69d91c2f9d21846fdf064", size = 780665, upload-time = "2025-07-31T00:19:01.828Z" },
-    { url = "https://files.pythonhosted.org/packages/40/e5/674b82bfff112c820b09e3c86a423d4a568143ede7f8440fdcbce259e895/regex-2025.7.34-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:c436fd1e95c04c19039668cfb548450a37c13f051e8659f40aed426e36b3765f", size = 773511, upload-time = "2025-07-31T00:19:03.654Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/18/39e7c578eb6cf1454db2b64e4733d7e4f179714867a75d84492ec44fa9b2/regex-2025.7.34-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:0b85241d3cfb9f8a13cefdfbd58a2843f208f2ed2c88181bf84e22e0c7fc066d", size = 843990, upload-time = "2025-07-31T00:19:05.61Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/d9/522a6715aefe2f463dc60c68924abeeb8ab6893f01adf5720359d94ede8c/regex-2025.7.34-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:075641c94126b064c65ab86e7e71fc3d63e7ff1bea1fb794f0773c97cdad3a03", size = 834676, upload-time = "2025-07-31T00:19:07.023Z" },
-    { url = "https://files.pythonhosted.org/packages/59/53/c4d5284cb40543566542e24f1badc9f72af68d01db21e89e36e02292eee0/regex-2025.7.34-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:70645cad3407d103d1dbcb4841839d2946f7d36cf38acbd40120fee1682151e5", size = 778420, upload-time = "2025-07-31T00:19:08.511Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/4a/b779a7707d4a44a7e6ee9d0d98e40b2a4de74d622966080e9c95e25e2d24/regex-2025.7.34-cp310-cp310-win32.whl", hash = "sha256:3b836eb4a95526b263c2a3359308600bd95ce7848ebd3c29af0c37c4f9627cd3", size = 263999, upload-time = "2025-07-31T00:19:10.072Z" },
-    { url = "https://files.pythonhosted.org/packages/ef/6e/33c7583f5427aa039c28bff7f4103c2de5b6aa5b9edc330c61ec576b1960/regex-2025.7.34-cp310-cp310-win_amd64.whl", hash = "sha256:cbfaa401d77334613cf434f723c7e8ba585df162be76474bccc53ae4e5520b3a", size = 276023, upload-time = "2025-07-31T00:19:11.34Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/fc/00b32e0ac14213d76d806d952826402b49fd06d42bfabacdf5d5d016bc47/regex-2025.7.34-cp310-cp310-win_arm64.whl", hash = "sha256:bca11d3c38a47c621769433c47f364b44e8043e0de8e482c5968b20ab90a3986", size = 268357, upload-time = "2025-07-31T00:19:12.729Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/85/f497b91577169472f7c1dc262a5ecc65e39e146fc3a52c571e5daaae4b7d/regex-2025.7.34-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:da304313761b8500b8e175eb2040c4394a875837d5635f6256d6fa0377ad32c8", size = 484594, upload-time = "2025-07-31T00:19:13.927Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/c5/ad2a5c11ce9e6257fcbfd6cd965d07502f6054aaa19d50a3d7fd991ec5d1/regex-2025.7.34-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:35e43ebf5b18cd751ea81455b19acfdec402e82fe0dc6143edfae4c5c4b3909a", size = 289294, upload-time = "2025-07-31T00:19:15.395Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/01/83ffd9641fcf5e018f9b51aa922c3e538ac9439424fda3df540b643ecf4f/regex-2025.7.34-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:96bbae4c616726f4661fe7bcad5952e10d25d3c51ddc388189d8864fbc1b3c68", size = 285933, upload-time = "2025-07-31T00:19:16.704Z" },
-    { url = "https://files.pythonhosted.org/packages/77/20/5edab2e5766f0259bc1da7381b07ce6eb4401b17b2254d02f492cd8a81a8/regex-2025.7.34-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9feab78a1ffa4f2b1e27b1bcdaad36f48c2fed4870264ce32f52a393db093c78", size = 792335, upload-time = "2025-07-31T00:19:18.561Z" },
-    { url = "https://files.pythonhosted.org/packages/30/bd/744d3ed8777dce8487b2606b94925e207e7c5931d5870f47f5b643a4580a/regex-2025.7.34-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f14b36e6d4d07f1a5060f28ef3b3561c5d95eb0651741474ce4c0a4c56ba8719", size = 858605, upload-time = "2025-07-31T00:19:20.204Z" },
-    { url = "https://files.pythonhosted.org/packages/99/3d/93754176289718d7578c31d151047e7b8acc7a8c20e7706716f23c49e45e/regex-2025.7.34-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:85c3a958ef8b3d5079c763477e1f09e89d13ad22198a37e9d7b26b4b17438b33", size = 905780, upload-time = "2025-07-31T00:19:21.876Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/2e/c689f274a92deffa03999a430505ff2aeace408fd681a90eafa92fdd6930/regex-2025.7.34-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:37555e4ae0b93358fa7c2d240a4291d4a4227cc7c607d8f85596cdb08ec0a083", size = 798868, upload-time = "2025-07-31T00:19:23.222Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/9e/39673688805d139b33b4a24851a71b9978d61915c4d72b5ffda324d0668a/regex-2025.7.34-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ee38926f31f1aa61b0232a3a11b83461f7807661c062df9eb88769d86e6195c3", size = 781784, upload-time = "2025-07-31T00:19:24.59Z" },
-    { url = "https://files.pythonhosted.org/packages/18/bd/4c1cab12cfabe14beaa076523056b8ab0c882a8feaf0a6f48b0a75dab9ed/regex-2025.7.34-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a664291c31cae9c4a30589bd8bc2ebb56ef880c9c6264cb7643633831e606a4d", size = 852837, upload-time = "2025-07-31T00:19:25.911Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/21/663d983cbb3bba537fc213a579abbd0f263fb28271c514123f3c547ab917/regex-2025.7.34-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:f3e5c1e0925e77ec46ddc736b756a6da50d4df4ee3f69536ffb2373460e2dafd", size = 844240, upload-time = "2025-07-31T00:19:27.688Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/2d/9beeeb913bc5d32faa913cf8c47e968da936af61ec20af5d269d0f84a100/regex-2025.7.34-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:d428fc7731dcbb4e2ffe43aeb8f90775ad155e7db4347a639768bc6cd2df881a", size = 787139, upload-time = "2025-07-31T00:19:29.475Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/f5/9b9384415fdc533551be2ba805dd8c4621873e5df69c958f403bfd3b2b6e/regex-2025.7.34-cp311-cp311-win32.whl", hash = "sha256:e154a7ee7fa18333ad90b20e16ef84daaeac61877c8ef942ec8dfa50dc38b7a1", size = 264019, upload-time = "2025-07-31T00:19:31.129Z" },
-    { url = "https://files.pythonhosted.org/packages/18/9d/e069ed94debcf4cc9626d652a48040b079ce34c7e4fb174f16874958d485/regex-2025.7.34-cp311-cp311-win_amd64.whl", hash = "sha256:24257953d5c1d6d3c129ab03414c07fc1a47833c9165d49b954190b2b7f21a1a", size = 276047, upload-time = "2025-07-31T00:19:32.497Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/cf/3bafbe9d1fd1db77355e7fbbbf0d0cfb34501a8b8e334deca14f94c7b315/regex-2025.7.34-cp311-cp311-win_arm64.whl", hash = "sha256:3157aa512b9e606586900888cd469a444f9b898ecb7f8931996cb715f77477f0", size = 268362, upload-time = "2025-07-31T00:19:34.094Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/f0/31d62596c75a33f979317658e8d261574785c6cd8672c06741ce2e2e2070/regex-2025.7.34-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:7f7211a746aced993bef487de69307a38c5ddd79257d7be83f7b202cb59ddb50", size = 485492, upload-time = "2025-07-31T00:19:35.57Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/16/b818d223f1c9758c3434be89aa1a01aae798e0e0df36c1f143d1963dd1ee/regex-2025.7.34-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:fb31080f2bd0681484b275461b202b5ad182f52c9ec606052020fe13eb13a72f", size = 290000, upload-time = "2025-07-31T00:19:37.175Z" },
-    { url = "https://files.pythonhosted.org/packages/cd/70/69506d53397b4bd6954061bae75677ad34deb7f6ca3ba199660d6f728ff5/regex-2025.7.34-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:0200a5150c4cf61e407038f4b4d5cdad13e86345dac29ff9dab3d75d905cf130", size = 286072, upload-time = "2025-07-31T00:19:38.612Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/73/536a216d5f66084fb577bb0543b5cb7de3272eb70a157f0c3a542f1c2551/regex-2025.7.34-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:739a74970e736df0773788377969c9fea3876c2fc13d0563f98e5503e5185f46", size = 797341, upload-time = "2025-07-31T00:19:40.119Z" },
-    { url = "https://files.pythonhosted.org/packages/26/af/733f8168449e56e8f404bb807ea7189f59507cbea1b67a7bbcd92f8bf844/regex-2025.7.34-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4fef81b2f7ea6a2029161ed6dea9ae13834c28eb5a95b8771828194a026621e4", size = 862556, upload-time = "2025-07-31T00:19:41.556Z" },
-    { url = "https://files.pythonhosted.org/packages/19/dd/59c464d58c06c4f7d87de4ab1f590e430821345a40c5d345d449a636d15f/regex-2025.7.34-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:ea74cf81fe61a7e9d77989050d0089a927ab758c29dac4e8e1b6c06fccf3ebf0", size = 910762, upload-time = "2025-07-31T00:19:43Z" },
-    { url = "https://files.pythonhosted.org/packages/37/a8/b05ccf33ceca0815a1e253693b2c86544932ebcc0049c16b0fbdf18b688b/regex-2025.7.34-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e4636a7f3b65a5f340ed9ddf53585c42e3ff37101d383ed321bfe5660481744b", size = 801892, upload-time = "2025-07-31T00:19:44.645Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/9a/b993cb2e634cc22810afd1652dba0cae156c40d4864285ff486c73cd1996/regex-2025.7.34-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:6cef962d7834437fe8d3da6f9bfc6f93f20f218266dcefec0560ed7765f5fe01", size = 786551, upload-time = "2025-07-31T00:19:46.127Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/79/7849d67910a0de4e26834b5bb816e028e35473f3d7ae563552ea04f58ca2/regex-2025.7.34-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:cbe1698e5b80298dbce8df4d8d1182279fbdaf1044e864cbc9d53c20e4a2be77", size = 856457, upload-time = "2025-07-31T00:19:47.562Z" },
-    { url = "https://files.pythonhosted.org/packages/91/c6/de516bc082524b27e45cb4f54e28bd800c01efb26d15646a65b87b13a91e/regex-2025.7.34-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:32b9f9bcf0f605eb094b08e8da72e44badabb63dde6b83bd530580b488d1c6da", size = 848902, upload-time = "2025-07-31T00:19:49.312Z" },
-    { url = "https://files.pythonhosted.org/packages/7d/22/519ff8ba15f732db099b126f039586bd372da6cd4efb810d5d66a5daeda1/regex-2025.7.34-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:524c868ba527eab4e8744a9287809579f54ae8c62fbf07d62aacd89f6026b282", size = 788038, upload-time = "2025-07-31T00:19:50.794Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/7d/aabb467d8f57d8149895d133c88eb809a1a6a0fe262c1d508eb9dfabb6f9/regex-2025.7.34-cp312-cp312-win32.whl", hash = "sha256:d600e58ee6d036081c89696d2bdd55d507498a7180df2e19945c6642fac59588", size = 264417, upload-time = "2025-07-31T00:19:52.292Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/39/bd922b55a4fc5ad5c13753274e5b536f5b06ec8eb9747675668491c7ab7a/regex-2025.7.34-cp312-cp312-win_amd64.whl", hash = "sha256:9a9ab52a466a9b4b91564437b36417b76033e8778e5af8f36be835d8cb370d62", size = 275387, upload-time = "2025-07-31T00:19:53.593Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/3c/c61d2fdcecb754a40475a3d1ef9a000911d3e3fc75c096acf44b0dfb786a/regex-2025.7.34-cp312-cp312-win_arm64.whl", hash = "sha256:c83aec91af9c6fbf7c743274fd952272403ad9a9db05fe9bfc9df8d12b45f176", size = 268482, upload-time = "2025-07-31T00:19:55.183Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/7f/8333b894499c1172c0378bb45a80146c420621e5c7b27a1d8fc5456f7038/regex-2025.7.34-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:fd5edc3f453de727af267c7909d083e19f6426fc9dd149e332b6034f2a5611e6", size = 484602, upload-time = "2025-07-31T00:20:48.184Z" },
-    { url = "https://files.pythonhosted.org/packages/14/47/58aac4758b659df3835e73bda070f78ec6620a028484a1fcb81daf7443ec/regex-2025.7.34-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:fa1cdfb8db96ef20137de5587954c812821966c3e8b48ffc871e22d7ec0a4938", size = 289289, upload-time = "2025-07-31T00:20:49.79Z" },
-    { url = "https://files.pythonhosted.org/packages/46/cc/5c9ebdc23b34458a41b559e0ae1b759196b2212920164b9d8aae4b25aa26/regex-2025.7.34-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:89c9504fc96268e8e74b0283e548f53a80c421182a2007e3365805b74ceef936", size = 285931, upload-time = "2025-07-31T00:20:51.362Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/da/467a851615b040d3be478ef60fd2d54e7e2f44eeda65dc02866ad4e404df/regex-2025.7.34-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:33be70d75fa05a904ee0dc43b650844e067d14c849df7e82ad673541cd465b5f", size = 779782, upload-time = "2025-07-31T00:20:52.997Z" },
-    { url = "https://files.pythonhosted.org/packages/a0/47/6eab7100b7ded84e94312c6791ab72581950b7adaa5ad48cdd3dfa329ab8/regex-2025.7.34-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:57d25b6732ea93eeb1d090e8399b6235ca84a651b52d52d272ed37d3d2efa0f1", size = 848838, upload-time = "2025-07-31T00:20:54.991Z" },
-    { url = "https://files.pythonhosted.org/packages/17/86/3b07305698e7ff21cc472efae816a56e77c5d45c6b7fe250a56dd67a114e/regex-2025.7.34-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:baf2fe122a3db1c0b9f161aa44463d8f7e33eeeda47bb0309923deb743a18276", size = 896648, upload-time = "2025-07-31T00:20:56.655Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/9a/c8f4f0535bf953e34e068c9a30c946e7affa06a48c48c1eda6d3a7562c49/regex-2025.7.34-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1a764a83128af9c1a54be81485b34dca488cbcacefe1e1d543ef11fbace191e1", size = 789367, upload-time = "2025-07-31T00:20:58.359Z" },
-    { url = "https://files.pythonhosted.org/packages/c1/4e/1892685a0e053d376fbcb8aa618e38afc5882bd69d94e9712171b9f2a412/regex-2025.7.34-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:c7f663ccc4093877f55b51477522abd7299a14c5bb7626c5238599db6a0cb95d", size = 780029, upload-time = "2025-07-31T00:21:00.383Z" },
-    { url = "https://files.pythonhosted.org/packages/98/12/af86906b9342d37b051b076a3ccc925c4f33ff2a96328b3009e7b93dfc53/regex-2025.7.34-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:4913f52fbc7a744aaebf53acd8d3dc1b519e46ba481d4d7596de3c862e011ada", size = 773039, upload-time = "2025-07-31T00:21:02.093Z" },
-    { url = "https://files.pythonhosted.org/packages/97/d1/03c21fb12daf73819f39927b533d09f162e8e452bd415993607242c1cd68/regex-2025.7.34-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:efac4db9e044d47fd3b6b0d40b6708f4dfa2d8131a5ac1d604064147c0f552fd", size = 843438, upload-time = "2025-07-31T00:21:04.248Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/7f/53569415d23dc47122c9f669db5d1e7aa2bd8954723e5c1050548cb7622e/regex-2025.7.34-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:7373afae7cfb716e3b8e15d0184510d518f9d21471f2d62918dbece85f2c588f", size = 834053, upload-time = "2025-07-31T00:21:06.298Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/7a/9b6b75778f7af6306ad9dcd9860be3f9c4123385cc856b6e9d099a6403b2/regex-2025.7.34-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:9960d162f3fecf6af252534a1ae337e9c2e20d74469fed782903b24e2cc9d3d7", size = 777909, upload-time = "2025-07-31T00:21:08.302Z" },
-    { url = "https://files.pythonhosted.org/packages/54/34/ebdf85bef946c63dc7995e95710364de0e3e2791bc28afc1a9642373d6c1/regex-2025.7.34-cp39-cp39-win32.whl", hash = "sha256:95d538b10eb4621350a54bf14600cc80b514211d91a019dc74b8e23d2159ace5", size = 264039, upload-time = "2025-07-31T00:21:10.346Z" },
-    { url = "https://files.pythonhosted.org/packages/82/0b/fba6f0dee661b838c09c85bf598a43a915d310648d62f704ece237aa3d73/regex-2025.7.34-cp39-cp39-win_amd64.whl", hash = "sha256:f7f3071b5faa605b0ea51ec4bb3ea7257277446b053f4fd3ad02b1dcb4e64353", size = 276120, upload-time = "2025-07-31T00:21:12.321Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/6d/183f0cf19bd8ac7628f4c3b2ca99033a5ad417ad010f86c61d11d27b4968/regex-2025.7.34-cp39-cp39-win_arm64.whl", hash = "sha256:716a47515ba1d03f8e8a61c5013041c8c90f2e21f055203498105d7571b44531", size = 268390, upload-time = "2025-07-31T00:21:14.293Z" },
+version = "2025.9.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/b2/5a/4c63457fbcaf19d138d72b2e9b39405954f98c0349b31c601bfcb151582c/regex-2025.9.1.tar.gz", hash = "sha256:88ac07b38d20b54d79e704e38aa3bd2c0f8027432164226bdee201a1c0c9c9ff", size = 400852, upload-time = "2025-09-01T22:10:10.479Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/46/c1/ed9ef923156105a78aa004f9390e5dd87eadc29f5ca8840f172cadb638de/regex-2025.9.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:c5aa2a6a73bf218515484b36a0d20c6ad9dc63f6339ff6224147b0e2c095ee55", size = 484813, upload-time = "2025-09-01T22:07:45.528Z" },
+    { url = "https://files.pythonhosted.org/packages/05/de/97957618a774c67f892609eee2fafe3e30703fbbba66de5e6b79d7196dbc/regex-2025.9.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:8c2ff5c01d5e47ad5fc9d31bcd61e78c2fa0068ed00cab86b7320214446da766", size = 288981, upload-time = "2025-09-01T22:07:48.464Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/b0/441afadd0a6ffccbd58a9663e5bdd182daa237893e5f8ceec6ff9df4418a/regex-2025.9.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d49dc84e796b666181de8a9973284cad6616335f01b52bf099643253094920fc", size = 286608, upload-time = "2025-09-01T22:07:50.484Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/cf/d89aecaf17e999ab11a3ef73fc9ab8b64f4e156f121250ef84340b35338d/regex-2025.9.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d9914fe1040874f83c15fcea86d94ea54091b0666eab330aaab69e30d106aabe", size = 780459, upload-time = "2025-09-01T22:07:52.34Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/05/05884594a9975a29597917bbdd6837f7b97e8ac23faf22d628aa781e58f7/regex-2025.9.1-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e71bceb3947362ec5eabd2ca0870bb78eae4edfc60c6c21495133c01b6cd2df4", size = 849276, upload-time = "2025-09-01T22:07:54.591Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/8d/2b3067506838d02096bf107beb129b2ce328cdf776d6474b7f542c0a7bfd/regex-2025.9.1-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:67a74456f410fe5e869239ee7a5423510fe5121549af133809d9591a8075893f", size = 897320, upload-time = "2025-09-01T22:07:56.129Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/b3/0f9f7766e980b900df0ba9901b52871a2e4203698fb35cdebd219240d5f7/regex-2025.9.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5c3b96ed0223b32dbdc53a83149b6de7ca3acd5acd9c8e64b42a166228abe29c", size = 789931, upload-time = "2025-09-01T22:07:57.834Z" },
+    { url = "https://files.pythonhosted.org/packages/47/9f/7b2f29c8f8b698eb44be5fc68e8b9c8d32e99635eac5defc98de114e9f35/regex-2025.9.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:113d5aa950f428faf46fd77d452df62ebb4cc6531cb619f6cc30a369d326bfbd", size = 780764, upload-time = "2025-09-01T22:07:59.413Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/ac/56176caa86155c14462531eb0a4ddc450d17ba8875001122b3b7c0cb01bf/regex-2025.9.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:fcdeb38de4f7f3d69d798f4f371189061446792a84e7c92b50054c87aae9c07c", size = 773610, upload-time = "2025-09-01T22:08:01.042Z" },
+    { url = "https://files.pythonhosted.org/packages/39/e8/9d6b9bd43998268a9de2f35602077519cacc9cb149f7381758cf8f502ba7/regex-2025.9.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:4bcdff370509164b67a6c8ec23c9fb40797b72a014766fdc159bb809bd74f7d8", size = 844090, upload-time = "2025-09-01T22:08:02.94Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/92/d89743b089005cae4cb81cc2fe177e180b7452e60f29de53af34349640f8/regex-2025.9.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:7383efdf6e8e8c61d85e00cfb2e2e18da1a621b8bfb4b0f1c2747db57b942b8f", size = 834775, upload-time = "2025-09-01T22:08:04.781Z" },
+    { url = "https://files.pythonhosted.org/packages/01/8f/86a3e0aaa89295d2a3445bb238e56369963ef6b02a5b4aa3362f4e687413/regex-2025.9.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:1ec2bd3bdf0f73f7e9f48dca550ba7d973692d5e5e9a90ac42cc5f16c4432d8b", size = 778521, upload-time = "2025-09-01T22:08:06.596Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/df/72072acb370ee8577c255717f8a58264f1d0de40aa3c9e6ebd5271cac633/regex-2025.9.1-cp310-cp310-win32.whl", hash = "sha256:9627e887116c4e9c0986d5c3b4f52bcfe3df09850b704f62ec3cbf177a0ae374", size = 264105, upload-time = "2025-09-01T22:08:08.708Z" },
+    { url = "https://files.pythonhosted.org/packages/97/73/fb82faaf0375aeaa1bb675008246c79b6779fa5688585a35327610ea0e2e/regex-2025.9.1-cp310-cp310-win_amd64.whl", hash = "sha256:94533e32dc0065eca43912ee6649c90ea0681d59f56d43c45b5bcda9a740b3dd", size = 276131, upload-time = "2025-09-01T22:08:10.156Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/3a/77d7718a2493e54725494f44da1a1e55704743dc4b8fabe5b0596f7b8014/regex-2025.9.1-cp310-cp310-win_arm64.whl", hash = "sha256:a874a61bb580d48642ffd338570ee24ab13fa023779190513fcacad104a6e251", size = 268462, upload-time = "2025-09-01T22:08:11.651Z" },
+    { url = "https://files.pythonhosted.org/packages/06/4d/f741543c0c59f96c6625bc6c11fea1da2e378b7d293ffff6f318edc0ce14/regex-2025.9.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:e5bcf112b09bfd3646e4db6bf2e598534a17d502b0c01ea6550ba4eca780c5e6", size = 484811, upload-time = "2025-09-01T22:08:12.834Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/bd/27e73e92635b6fbd51afc26a414a3133243c662949cd1cda677fe7bb09bd/regex-2025.9.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:67a0295a3c31d675a9ee0238d20238ff10a9a2fdb7a1323c798fc7029578b15c", size = 288977, upload-time = "2025-09-01T22:08:14.499Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/7d/7dc0c6efc8bc93cd6e9b947581f5fde8a5dbaa0af7c4ec818c5729fdc807/regex-2025.9.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ea8267fbadc7d4bd7c1301a50e85c2ff0de293ff9452a1a9f8d82c6cafe38179", size = 286606, upload-time = "2025-09-01T22:08:15.881Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/01/9b5c6dd394f97c8f2c12f6e8f96879c9ac27292a718903faf2e27a0c09f6/regex-2025.9.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6aeff21de7214d15e928fb5ce757f9495214367ba62875100d4c18d293750cc1", size = 792436, upload-time = "2025-09-01T22:08:17.38Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/24/b7430cfc6ee34bbb3db6ff933beb5e7692e5cc81e8f6f4da63d353566fb0/regex-2025.9.1-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:d89f1bbbbbc0885e1c230f7770d5e98f4f00b0ee85688c871d10df8b184a6323", size = 858705, upload-time = "2025-09-01T22:08:19.037Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/98/155f914b4ea6ae012663188545c4f5216c11926d09b817127639d618b003/regex-2025.9.1-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:ca3affe8ddea498ba9d294ab05f5f2d3b5ad5d515bc0d4a9016dd592a03afe52", size = 905881, upload-time = "2025-09-01T22:08:20.377Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/a7/a470e7bc8259c40429afb6d6a517b40c03f2f3e455c44a01abc483a1c512/regex-2025.9.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:91892a7a9f0a980e4c2c85dd19bc14de2b219a3a8867c4b5664b9f972dcc0c78", size = 798968, upload-time = "2025-09-01T22:08:22.081Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/fa/33f6fec4d41449fea5f62fdf5e46d668a1c046730a7f4ed9f478331a8e3a/regex-2025.9.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:e1cb40406f4ae862710615f9f636c1e030fd6e6abe0e0f65f6a695a2721440c6", size = 781884, upload-time = "2025-09-01T22:08:23.832Z" },
+    { url = "https://files.pythonhosted.org/packages/42/de/2b45f36ab20da14eedddf5009d370625bc5942d9953fa7e5037a32d66843/regex-2025.9.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:94f6cff6f7e2149c7e6499a6ecd4695379eeda8ccbccb9726e8149f2fe382e92", size = 852935, upload-time = "2025-09-01T22:08:25.536Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/f9/878f4fc92c87e125e27aed0f8ee0d1eced9b541f404b048f66f79914475a/regex-2025.9.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:6c0226fb322b82709e78c49cc33484206647f8a39954d7e9de1567f5399becd0", size = 844340, upload-time = "2025-09-01T22:08:27.141Z" },
+    { url = "https://files.pythonhosted.org/packages/90/c2/5b6f2bce6ece5f8427c718c085eca0de4bbb4db59f54db77aa6557aef3e9/regex-2025.9.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a12f59c7c380b4fcf7516e9cbb126f95b7a9518902bcf4a852423ff1dcd03e6a", size = 787238, upload-time = "2025-09-01T22:08:28.75Z" },
+    { url = "https://files.pythonhosted.org/packages/47/66/1ef1081c831c5b611f6f55f6302166cfa1bc9574017410ba5595353f846a/regex-2025.9.1-cp311-cp311-win32.whl", hash = "sha256:49865e78d147a7a4f143064488da5d549be6bfc3f2579e5044cac61f5c92edd4", size = 264118, upload-time = "2025-09-01T22:08:30.388Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/e0/8adc550d7169df1d6b9be8ff6019cda5291054a0107760c2f30788b6195f/regex-2025.9.1-cp311-cp311-win_amd64.whl", hash = "sha256:d34b901f6f2f02ef60f4ad3855d3a02378c65b094efc4b80388a3aeb700a5de7", size = 276151, upload-time = "2025-09-01T22:08:32.073Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/bd/46fef29341396d955066e55384fb93b0be7d64693842bf4a9a398db6e555/regex-2025.9.1-cp311-cp311-win_arm64.whl", hash = "sha256:47d7c2dab7e0b95b95fd580087b6ae196039d62306a592fa4e162e49004b6299", size = 268460, upload-time = "2025-09-01T22:08:33.281Z" },
+    { url = "https://files.pythonhosted.org/packages/39/ef/a0372febc5a1d44c1be75f35d7e5aff40c659ecde864d7fa10e138f75e74/regex-2025.9.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:84a25164bd8dcfa9f11c53f561ae9766e506e580b70279d05a7946510bdd6f6a", size = 486317, upload-time = "2025-09-01T22:08:34.529Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/25/d64543fb7eb41a1024786d518cc57faf1ce64aa6e9ddba097675a0c2f1d2/regex-2025.9.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:645e88a73861c64c1af558dd12294fb4e67b5c1eae0096a60d7d8a2143a611c7", size = 289698, upload-time = "2025-09-01T22:08:36.162Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/dc/fbf31fc60be317bd9f6f87daa40a8a9669b3b392aa8fe4313df0a39d0722/regex-2025.9.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:10a450cba5cd5409526ee1d4449f42aad38dd83ac6948cbd6d7f71ca7018f7db", size = 287242, upload-time = "2025-09-01T22:08:37.794Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/74/f933a607a538f785da5021acf5323961b4620972e2c2f1f39b6af4b71db7/regex-2025.9.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e9dc5991592933a4192c166eeb67b29d9234f9c86344481173d1bc52f73a7104", size = 797441, upload-time = "2025-09-01T22:08:39.108Z" },
+    { url = "https://files.pythonhosted.org/packages/89/d0/71fc49b4f20e31e97f199348b8c4d6e613e7b6a54a90eb1b090c2b8496d7/regex-2025.9.1-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:a32291add816961aab472f4fad344c92871a2ee33c6c219b6598e98c1f0108f2", size = 862654, upload-time = "2025-09-01T22:08:40.586Z" },
+    { url = "https://files.pythonhosted.org/packages/59/05/984edce1411a5685ba9abbe10d42cdd9450aab4a022271f9585539788150/regex-2025.9.1-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:588c161a68a383478e27442a678e3b197b13c5ba51dbba40c1ccb8c4c7bee9e9", size = 910862, upload-time = "2025-09-01T22:08:42.416Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/02/5c891bb5fe0691cc1bad336e3a94b9097fbcf9707ec8ddc1dce9f0397289/regex-2025.9.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:47829ffaf652f30d579534da9085fe30c171fa2a6744a93d52ef7195dc38218b", size = 801991, upload-time = "2025-09-01T22:08:44.072Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/ae/fd10d6ad179910f7a1b3e0a7fde1ef8bb65e738e8ac4fd6ecff3f52252e4/regex-2025.9.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:1e978e5a35b293ea43f140c92a3269b6ab13fe0a2bf8a881f7ac740f5a6ade85", size = 786651, upload-time = "2025-09-01T22:08:46.079Z" },
+    { url = "https://files.pythonhosted.org/packages/30/cf/9d686b07bbc5bf94c879cc168db92542d6bc9fb67088d03479fef09ba9d3/regex-2025.9.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:4cf09903e72411f4bf3ac1eddd624ecfd423f14b2e4bf1c8b547b72f248b7bf7", size = 856556, upload-time = "2025-09-01T22:08:48.376Z" },
+    { url = "https://files.pythonhosted.org/packages/91/9d/302f8a29bb8a49528abbab2d357a793e2a59b645c54deae0050f8474785b/regex-2025.9.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:d016b0f77be63e49613c9e26aaf4a242f196cd3d7a4f15898f5f0ab55c9b24d2", size = 849001, upload-time = "2025-09-01T22:08:50.067Z" },
+    { url = "https://files.pythonhosted.org/packages/93/fa/b4c6dbdedc85ef4caec54c817cd5f4418dbfa2453214119f2538082bf666/regex-2025.9.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:656563e620de6908cd1c9d4f7b9e0777e3341ca7db9d4383bcaa44709c90281e", size = 788138, upload-time = "2025-09-01T22:08:51.933Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/1b/91ee17a3cbf87f81e8c110399279d0e57f33405468f6e70809100f2ff7d8/regex-2025.9.1-cp312-cp312-win32.whl", hash = "sha256:df33f4ef07b68f7ab637b1dbd70accbf42ef0021c201660656601e8a9835de45", size = 264524, upload-time = "2025-09-01T22:08:53.75Z" },
+    { url = "https://files.pythonhosted.org/packages/92/28/6ba31cce05b0f1ec6b787921903f83bd0acf8efde55219435572af83c350/regex-2025.9.1-cp312-cp312-win_amd64.whl", hash = "sha256:5aba22dfbc60cda7c0853516104724dc904caa2db55f2c3e6e984eb858d3edf3", size = 275489, upload-time = "2025-09-01T22:08:55.037Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/ed/ea49f324db00196e9ef7fe00dd13c6164d5173dd0f1bbe495e61bb1fb09d/regex-2025.9.1-cp312-cp312-win_arm64.whl", hash = "sha256:ec1efb4c25e1849c2685fa95da44bfde1b28c62d356f9c8d861d4dad89ed56e9", size = 268589, upload-time = "2025-09-01T22:08:56.369Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/28/3f2b211686160f15ccf2ecd770e25c588fb776332956006e02ab93982a5f/regex-2025.9.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:a13d20007dce3c4b00af5d84f6c191ed1c0f70928c6d9b6cd7b8d2f125df7f46", size = 484825, upload-time = "2025-09-01T22:09:44.502Z" },
+    { url = "https://files.pythonhosted.org/packages/be/8d/88539f6ca5967022942ab64f2038a037b344db636c80852d2b4b147f6b1a/regex-2025.9.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:d6b046b0a01cb713fd53ef36cb59db4b0062b343db28e83b52ac6aa01ee5b368", size = 288972, upload-time = "2025-09-01T22:09:46.415Z" },
+    { url = "https://files.pythonhosted.org/packages/88/ad/0b9316cfe91d8a2917fbcd8cb7c990051745b5852f9f3088b93b4267fe68/regex-2025.9.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:0fa9a7477288717f42dbd02ff5d13057549e9a8cdb81f224c313154cc10bab52", size = 286624, upload-time = "2025-09-01T22:09:47.823Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/6e/199003d956dac19ddaebb2ef3c068379918539a497dd4d4eb5561a1c5e3f/regex-2025.9.1-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b2b3ad150c6bc01a8cd5030040675060e2adbe6cbc50aadc4da42c6d32ec266e", size = 779882, upload-time = "2025-09-01T22:09:49.601Z" },
+    { url = "https://files.pythonhosted.org/packages/69/0a/a4bed3424beefbf8322b8b581aa2e2a65cbfe358d968f58a2ab910d62927/regex-2025.9.1-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:aa88d5a82dfe80deaf04e8c39c8b0ad166d5d527097eb9431cb932c44bf88715", size = 848938, upload-time = "2025-09-01T22:09:51.155Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/01/c6c2bc65e97387523742c54daec1384a50d2575d1189e844e4e320e145e9/regex-2025.9.1-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:6f1dae2cf6c2dbc6fd2526653692c144721b3cf3f769d2a3c3aa44d0f38b9a58", size = 896747, upload-time = "2025-09-01T22:09:52.965Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/56/25480407a452198414a68d6d1f4f9920652810ba9f2c7f6279e5bd9b15ef/regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ff62a3022914fc19adaa76b65e03cf62bc67ea16326cbbeb170d280710a7d719", size = 789466, upload-time = "2025-09-01T22:09:54.564Z" },
+    { url = "https://files.pythonhosted.org/packages/82/5c/207422b3a2c3923eb09a1e807435effcbd8b756cd8ece24c153d525463d6/regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:a34ef82216189d823bc82f614d1031cb0b919abef27cecfd7b07d1e9a8bdeeb4", size = 780130, upload-time = "2025-09-01T22:09:56.413Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/e3/74a9da339f043006a1c0e0daae143c2d8b70e8987da3c967dbb8b6e8c9d2/regex-2025.9.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:6d40e6b49daae9ebbd7fa4e600697372cba85b826592408600068e83a3c47211", size = 773140, upload-time = "2025-09-01T22:09:57.99Z" },
+    { url = "https://files.pythonhosted.org/packages/45/11/d2ca4dccec797f9325cfae0cfff8ee2977e1d5db43296ed949f8c63b6db0/regex-2025.9.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:0aeb0fe80331059c152a002142699a89bf3e44352aee28261315df0c9874759b", size = 843538, upload-time = "2025-09-01T22:09:59.74Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/9a/652095a5df5a8cd6b550280dc4b885f06beb68e2dc8728b380639b40446d/regex-2025.9.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:a90014d29cb3098403d82a879105d1418edbbdf948540297435ea6e377023ea7", size = 834151, upload-time = "2025-09-01T22:10:01.658Z" },
+    { url = "https://files.pythonhosted.org/packages/17/27/f1ca20401df0544a627188fede42ef723b7d88fed8364b0d80fa9833bb84/regex-2025.9.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:6ff623271e0b0cc5a95b802666bbd70f17ddd641582d65b10fb260cc0c003529", size = 778009, upload-time = "2025-09-01T22:10:03.609Z" },
+    { url = "https://files.pythonhosted.org/packages/20/28/c6c94f642c115e72142e8404464835f03dee55266a158455aeb8216d6d4d/regex-2025.9.1-cp39-cp39-win32.whl", hash = "sha256:d161bfdeabe236290adfd8c7588da7f835d67e9e7bf2945f1e9e120622839ba6", size = 264141, upload-time = "2025-09-01T22:10:05.457Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/7e/4ef3cf970b4c7ef9abee4c5a42f3405a452e5abe21095ee36f9a3a83af9d/regex-2025.9.1-cp39-cp39-win_amd64.whl", hash = "sha256:43ebc77a7dfe36661192afd8d7df5e8be81ec32d2ad0c65b536f66ebfec3dece", size = 276219, upload-time = "2025-09-01T22:10:07.106Z" },
+    { url = "https://files.pythonhosted.org/packages/60/42/86a45b36b0d8f37388485a2c93d0f3c571732ac4e17b92d8ccd5da53b792/regex-2025.9.1-cp39-cp39-win_arm64.whl", hash = "sha256:5d74b557cf5554001a869cda60b9a619be307df4d10155894aeaad3ee67c9899", size = 268497, upload-time = "2025-09-01T22:10:08.595Z" },
 ]
 
 [[package]]
@@ -4294,27 +4317,27 @@ wheels = [
 
 [[package]]
 name = "tokenizers"
-version = "0.21.4"
+version = "0.22.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "huggingface-hub" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/c2/2f/402986d0823f8d7ca139d969af2917fefaa9b947d1fb32f6168c509f2492/tokenizers-0.21.4.tar.gz", hash = "sha256:fa23f85fbc9a02ec5c6978da172cdcbac23498c3ca9f3645c5c68740ac007880", size = 351253, upload-time = "2025-07-28T15:48:54.325Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/5e/b4/c1ce3699e81977da2ace8b16d2badfd42b060e7d33d75c4ccdbf9dc920fa/tokenizers-0.22.0.tar.gz", hash = "sha256:2e33b98525be8453f355927f3cab312c36cd3e44f4d7e9e97da2fa94d0a49dcb", size = 362771, upload-time = "2025-08-29T10:25:33.914Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/98/c6/fdb6f72bf6454f52eb4a2510be7fb0f614e541a2554d6210e370d85efff4/tokenizers-0.21.4-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:2ccc10a7c3bcefe0f242867dc914fc1226ee44321eb618cfe3019b5df3400133", size = 2863987, upload-time = "2025-07-28T15:48:44.877Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/a6/28975479e35ddc751dc1ddc97b9b69bf7fcf074db31548aab37f8116674c/tokenizers-0.21.4-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:5e2f601a8e0cd5be5cc7506b20a79112370b9b3e9cb5f13f68ab11acd6ca7d60", size = 2732457, upload-time = "2025-07-28T15:48:43.265Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/8f/24f39d7b5c726b7b0be95dca04f344df278a3fe3a4deb15a975d194cbb32/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:39b376f5a1aee67b4d29032ee85511bbd1b99007ec735f7f35c8a2eb104eade5", size = 3012624, upload-time = "2025-07-28T13:22:43.895Z" },
-    { url = "https://files.pythonhosted.org/packages/58/47/26358925717687a58cb74d7a508de96649544fad5778f0cd9827398dc499/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:2107ad649e2cda4488d41dfd031469e9da3fcbfd6183e74e4958fa729ffbf9c6", size = 2939681, upload-time = "2025-07-28T13:22:47.499Z" },
-    { url = "https://files.pythonhosted.org/packages/99/6f/cc300fea5db2ab5ddc2c8aea5757a27b89c84469899710c3aeddc1d39801/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3c73012da95afafdf235ba80047699df4384fdc481527448a078ffd00e45a7d9", size = 3247445, upload-time = "2025-07-28T15:48:39.711Z" },
-    { url = "https://files.pythonhosted.org/packages/be/bf/98cb4b9c3c4afd8be89cfa6423704337dc20b73eb4180397a6e0d456c334/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f23186c40395fc390d27f519679a58023f368a0aad234af145e0f39ad1212732", size = 3428014, upload-time = "2025-07-28T13:22:49.569Z" },
-    { url = "https://files.pythonhosted.org/packages/75/c7/96c1cc780e6ca7f01a57c13235dd05b7bc1c0f3588512ebe9d1331b5f5ae/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cc88bb34e23a54cc42713d6d98af5f1bf79c07653d24fe984d2d695ba2c922a2", size = 3193197, upload-time = "2025-07-28T13:22:51.471Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/90/273b6c7ec78af547694eddeea9e05de771278bd20476525ab930cecaf7d8/tokenizers-0.21.4-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:51b7eabb104f46c1c50b486520555715457ae833d5aee9ff6ae853d1130506ff", size = 3115426, upload-time = "2025-07-28T15:48:41.439Z" },
-    { url = "https://files.pythonhosted.org/packages/91/43/c640d5a07e95f1cf9d2c92501f20a25f179ac53a4f71e1489a3dcfcc67ee/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:714b05b2e1af1288bd1bc56ce496c4cebb64a20d158ee802887757791191e6e2", size = 9089127, upload-time = "2025-07-28T15:48:46.472Z" },
-    { url = "https://files.pythonhosted.org/packages/44/a1/dd23edd6271d4dca788e5200a807b49ec3e6987815cd9d0a07ad9c96c7c2/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:1340ff877ceedfa937544b7d79f5b7becf33a4cfb58f89b3b49927004ef66f78", size = 9055243, upload-time = "2025-07-28T15:48:48.539Z" },
-    { url = "https://files.pythonhosted.org/packages/21/2b/b410d6e9021c4b7ddb57248304dc817c4d4970b73b6ee343674914701197/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:3c1f4317576e465ac9ef0d165b247825a2a4078bcd01cba6b54b867bdf9fdd8b", size = 9298237, upload-time = "2025-07-28T15:48:50.443Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/0a/42348c995c67e2e6e5c89ffb9cfd68507cbaeb84ff39c49ee6e0a6dd0fd2/tokenizers-0.21.4-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:c212aa4e45ec0bb5274b16b6f31dd3f1c41944025c2358faaa5782c754e84c24", size = 9461980, upload-time = "2025-07-28T15:48:52.325Z" },
-    { url = "https://files.pythonhosted.org/packages/3d/d3/dacccd834404cd71b5c334882f3ba40331ad2120e69ded32cf5fda9a7436/tokenizers-0.21.4-cp39-abi3-win32.whl", hash = "sha256:6c42a930bc5f4c47f4ea775c91de47d27910881902b0f20e4990ebe045a415d0", size = 2329871, upload-time = "2025-07-28T15:48:56.841Z" },
-    { url = "https://files.pythonhosted.org/packages/41/f2/fd673d979185f5dcbac4be7d09461cbb99751554ffb6718d0013af8604cb/tokenizers-0.21.4-cp39-abi3-win_amd64.whl", hash = "sha256:475d807a5c3eb72c59ad9b5fcdb254f6e17f53dfcbb9903233b0dfa9c943b597", size = 2507568, upload-time = "2025-07-28T15:48:55.456Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/b1/18c13648edabbe66baa85fe266a478a7931ddc0cd1ba618802eb7b8d9865/tokenizers-0.22.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:eaa9620122a3fb99b943f864af95ed14c8dfc0f47afa3b404ac8c16b3f2bb484", size = 3081954, upload-time = "2025-08-29T10:25:24.993Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/02/c3c454b641bd7c4f79e4464accfae9e7dfc913a777d2e561e168ae060362/tokenizers-0.22.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:71784b9ab5bf0ff3075bceeb198149d2c5e068549c0d18fe32d06ba0deb63f79", size = 2945644, upload-time = "2025-08-29T10:25:23.405Z" },
+    { url = "https://files.pythonhosted.org/packages/55/02/d10185ba2fd8c2d111e124c9d92de398aee0264b35ce433f79fb8472f5d0/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ec5b71f668a8076802b0241a42387d48289f25435b86b769ae1837cad4172a17", size = 3254764, upload-time = "2025-08-29T10:25:12.445Z" },
+    { url = "https://files.pythonhosted.org/packages/13/89/17514bd7ef4bf5bfff58e2b131cec0f8d5cea2b1c8ffe1050a2c8de88dbb/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ea8562fa7498850d02a16178105b58803ea825b50dc9094d60549a7ed63654bb", size = 3161654, upload-time = "2025-08-29T10:25:15.493Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/d8/bac9f3a7ef6dcceec206e3857c3b61bb16c6b702ed7ae49585f5bd85c0ef/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4136e1558a9ef2e2f1de1555dcd573e1cbc4a320c1a06c4107a3d46dc8ac6e4b", size = 3511484, upload-time = "2025-08-29T10:25:20.477Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/27/9c9800eb6763683010a4851db4d1802d8cab9cec114c17056eccb4d4a6e0/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cdf5954de3962a5fd9781dc12048d24a1a6f1f5df038c6e95db328cd22964206", size = 3712829, upload-time = "2025-08-29T10:25:17.154Z" },
+    { url = "https://files.pythonhosted.org/packages/10/e3/b1726dbc1f03f757260fa21752e1921445b5bc350389a8314dd3338836db/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8337ca75d0731fc4860e6204cc24bb36a67d9736142aa06ed320943b50b1e7ed", size = 3408934, upload-time = "2025-08-29T10:25:18.76Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/61/aeab3402c26874b74bb67a7f2c4b569dde29b51032c5384db592e7b216f4/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a89264e26f63c449d8cded9061adea7b5de53ba2346fc7e87311f7e4117c1cc8", size = 3345585, upload-time = "2025-08-29T10:25:22.08Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/d3/498b4a8a8764cce0900af1add0f176ff24f475d4413d55b760b8cdf00893/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:790bad50a1b59d4c21592f9c3cf5e5cf9c3c7ce7e1a23a739f13e01fb1be377a", size = 9322986, upload-time = "2025-08-29T10:25:26.607Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/62/92378eb1c2c565837ca3cb5f9569860d132ab9d195d7950c1ea2681dffd0/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:76cf6757c73a10ef10bf06fa937c0ec7393d90432f543f49adc8cab3fb6f26cb", size = 9276630, upload-time = "2025-08-29T10:25:28.349Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/f0/342d80457aa1cda7654327460f69db0d69405af1e4c453f4dc6ca7c4a76e/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:1626cb186e143720c62c6c6b5371e62bbc10af60481388c0da89bc903f37ea0c", size = 9547175, upload-time = "2025-08-29T10:25:29.989Z" },
+    { url = "https://files.pythonhosted.org/packages/14/84/8aa9b4adfc4fbd09381e20a5bc6aa27040c9c09caa89988c01544e008d18/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:da589a61cbfea18ae267723d6b029b84598dc8ca78db9951d8f5beff72d8507c", size = 9692735, upload-time = "2025-08-29T10:25:32.089Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/24/83ee2b1dc76bfe05c3142e7d0ccdfe69f0ad2f1ebf6c726cea7f0874c0d0/tokenizers-0.22.0-cp39-abi3-win32.whl", hash = "sha256:dbf9d6851bddae3e046fedfb166f47743c1c7bd11c640f0691dd35ef0bcad3be", size = 2471915, upload-time = "2025-08-29T10:25:36.411Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/9b/0e0bf82214ee20231845b127aa4a8015936ad5a46779f30865d10e404167/tokenizers-0.22.0-cp39-abi3-win_amd64.whl", hash = "sha256:c78174859eeaee96021f248a56c801e36bfb6bd5b067f2e95aa82445ca324f00", size = 2680494, upload-time = "2025-08-29T10:25:35.14Z" },
 ]
 
 [[package]]
@@ -4418,7 +4441,7 @@ wheels = [
 
 [[package]]
 name = "transformers"
-version = "4.55.4"
+version = "4.56.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "filelock" },
@@ -4433,9 +4456,9 @@ dependencies = [
     { name = "tokenizers" },
     { name = "tqdm" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/2b/43/3cb831d5f28cc723516e5bb43a8c6042aca3038bb36b6bd6016b40dfd1e8/transformers-4.55.4.tar.gz", hash = "sha256:574a30559bc273c7a4585599ff28ab6b676e96dc56ffd2025ecfce2fd0ab915d", size = 9573015, upload-time = "2025-08-22T15:18:43.192Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/e9/40/2898c1a5f8b2fff2f94dffd856d6793320e590e1cfdc3a1453c31c364094/transformers-4.56.0.tar.gz", hash = "sha256:6ca9c3f38aa4da93ebf877db7156368c1c188c7465f09dbe70951e7622e987fa", size = 9840426, upload-time = "2025-08-29T18:23:46.247Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/fa/0a/8791a6ee0529c45f669566969e99b75e2ab20eb0bfee8794ce295c18bdad/transformers-4.55.4-py3-none-any.whl", hash = "sha256:df28f3849665faba4af5106f0db4510323277c4bb595055340544f7e59d06458", size = 11269659, upload-time = "2025-08-22T15:18:40.025Z" },
+    { url = "https://files.pythonhosted.org/packages/98/41/6ba7c2eafa839069f1f949c30cd22e791b14b90c7af8cb0c65cc47702dca/transformers-4.56.0-py3-none-any.whl", hash = "sha256:bacf539c38dd850690856881c4974321af93a22f2ee96bcc994741a2121d8e71", size = 11607938, upload-time = "2025-08-29T18:23:33.92Z" },
 ]
 
 [[package]]

From f9e0a15833a3d434c8ad774a81f9831217374d53 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Tue, 2 Sep 2025 13:22:23 +0200
Subject: [PATCH 444/505] Add setuptools to the list of transformers
 dependencies

---
 pyproject.toml | 2 +-
 uv.lock        | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 8fecdaa7c..c08b56f5d 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -47,7 +47,7 @@ ollama = ["ollama"]
 openai = ["openai"]
 sglang = ["openai"]
 tgi = ["huggingface_hub"]
-transformers = ["accelerate", "datasets", "transformers"]
+transformers = ["accelerate", "datasets", "transformers", "setuptools"]
 vllm = ["openai"]
 xgrammar = ["xgrammar"]
 llguidance = ["llguidance"]
diff --git a/uv.lock b/uv.lock
index 7791d4dad..5b8721b9a 100644
--- a/uv.lock
+++ b/uv.lock
@@ -2759,6 +2759,7 @@ tgi = [
 transformers = [
     { name = "accelerate" },
     { name = "datasets" },
+    { name = "setuptools" },
     { name = "transformers" },
 ]
 vllm = [
@@ -2830,6 +2831,7 @@ requires-dist = [
     { name = "requests", marker = "extra == 'test'" },
     { name = "responses", marker = "extra == 'test'" },
     { name = "sentencepiece", marker = "extra == 'test'" },
+    { name = "setuptools", marker = "extra == 'transformers'" },
     { name = "tensorflow", marker = "extra == 'test'" },
     { name = "tf-keras", marker = "extra == 'test'" },
     { name = "torch", marker = "extra == 'test'", specifier = ">2.3.0" },

From f12bb75881900de801deef36f711fc2b4cd33dd9 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 15 Sep 2025 20:21:02 +0200
Subject: [PATCH 445/505] Update the output_type formatting of the llamacpp
 model

---
 outlines/models/llamacpp.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index c6a3bf8b0..8b788a7f9 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -197,7 +197,7 @@ def format_chat_input(self, model_input: Chat) -> list:
 
     def format_output_type(
         self, output_type: Optional[OutlinesLogitsProcessor] = None,
-    ) -> "LogitsProcessorList":
+    ) -> Optional["LogitsProcessorList"]:
         """Generate the logits processor argument to pass to the model.
 
         Parameters
@@ -213,7 +213,9 @@ def format_output_type(
         """
         from llama_cpp import LogitsProcessorList
 
-        return LogitsProcessorList([output_type])
+        if output_type is not None:
+            return LogitsProcessorList([output_type])
+        return None
 
 
 class LlamaCpp(Model):

From 38263107019ba8e584d160fb2481382e1196443a Mon Sep 17 00:00:00 2001
From: Francesco Laiti <25352428+laitifranz@users.noreply.github.com>
Date: Thu, 14 Aug 2025 17:55:54 +0200
Subject: [PATCH 446/505] Fix correct handling of chat multimodal inputs in
 TransformersMultiModal model and update docs for the usage of Chat in
 TransformersMultiModal

---
 .../models/transformers_multimodal.md         | 52 +++++++++++++++++++
 outlines/models/transformers.py               | 26 ++++------
 2 files changed, 63 insertions(+), 15 deletions(-)

diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index 24d119de3..42bc75e0e 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -120,6 +120,58 @@ result = model.batch(
 print(result) # ['The image shows a cat', 'The image shows an astronaut']
 ```
 
+### Chat
+You can use chat inputs with the `TransformersMultiModal` model. To do so, call the model with a `Chat` instance. 
+
+For instance:
+
+```python
+import outlines
+from outlines.inputs import Chat, Image
+from transformers import AutoModelForImageTextToText, AutoProcessor
+from PIL import Image as PILImage
+from io import BytesIO
+from urllib.request import urlopen
+import torch
+
+model_kwargs = {
+        "torch_dtype": torch.bfloat16,
+        "attn_implementation": "flash_attention_2",
+        "device_map": "auto",
+    }
+
+def get_image_from_url(image_url):
+    img_byte_stream = BytesIO(urlopen(image_url).read())
+    image = PILImage.open(img_byte_stream).convert("RGB")
+    image.format = "PNG"
+    image.save("image.png")
+    return image
+    
+# Create the model
+model = outlines.from_transformers(
+    AutoModelForImageTextToText.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", **model_kwargs),
+    AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", **model_kwargs)
+)
+
+IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+
+# Create the chat mutimodal input
+prompt = Chat([
+    {
+        "role": "user",
+        "content": [
+            {"type": "image", "image": Image(get_image_from_url(IMAGE_URL))},
+            {"type": "text", "text": "Describe the image in few words."}
+        ],
+    }
+])
+
+# Call the model to generate a response
+response = model(prompt, max_new_tokens=50)
+print(response) # 'A Siamese cat with blue eyes is sitting on a cat tree, looking alert and curious.'
+```
+
+
 !!! Warning
 
     Make sure your prompt contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are image assets included in your model input.
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 964dbce60..3dfae601c 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -459,26 +459,22 @@ def format_dict_input(self, model_input: dict) -> dict:
 
     @format_input.register(Chat)
     def format_chat_input(self, model_input: Chat) -> dict:
-        # we need to separate the images from the messages
-        # to apply the chat template to the messages without images
+        # we need to separate the assets from the messages
+        # to apply the chat template to the messages without assets
         messages = model_input.messages
-        images = []
-        messages_without_images = []
+        assets = []
         for message in messages:
             if isinstance(message["content"], list):
-                images.extend(message["content"][1:])
-                messages_without_images.append({
-                    "role": message["role"],
-                    "content": message["content"][0],
-                })
-            else:
-                messages_without_images.append(message)
+                for item in message["content"]:
+                    if item["type"] != "text":
+                        assets.append(item[item["type"]])
         formatted_prompt = self.tokenizer.apply_chat_template(
-            messages_without_images,
-            tokenize=False
+            messages, # full message for applying chat template
+            tokenize=False,
+            add_generation_prompt=True
         )
-        # use the formatted prompt and the images to format the input
-        return self.format_list_input([formatted_prompt, *images])
+        # use the formatted prompt and the assets to format the input
+        return self.format_list_input([formatted_prompt, *assets])
 
     @format_input.register(list)
     def format_list_input(self, model_input: list) -> dict:

From 32ccbc1d73c581fc05c8043d7bae0c73d82dc03d Mon Sep 17 00:00:00 2001
From: Francesco Laiti <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 16 Aug 2025 10:49:04 +0200
Subject: [PATCH 447/505] Handle empty assets cases where onyl text is provided
 to MLLMs

---
 outlines/models/transformers.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 3dfae601c..4cfeba27e 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -481,6 +481,9 @@ def format_list_input(self, model_input: list) -> dict:
         prompt = model_input[0]
         assets = model_input[1:]
 
+        if not assets:  # handle empty assets case
+            return {"text": prompt}
+        
         asset_types = set(type(asset) for asset in assets)
         if len(asset_types) > 1:
             raise ValueError(

From 350b196c65099de3ab8c4960793fef3c090aae9d Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Fri, 22 Aug 2025 20:38:43 +0200
Subject: [PATCH 448/505] Remove deprecated function and update test
 accordingly since Outlines version is now > 1.2.0

---
 outlines/models/transformers.py               | 27 -------------------
 ...st_transformers_multimodal_type_adapter.py |  5 ++--
 2 files changed, 2 insertions(+), 30 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 4cfeba27e..3729e5074 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -430,33 +430,6 @@ def format_input(self, model_input):
             + "model or a `Chat` instance."
         )
 
-    @format_input.register(dict)
-    def format_dict_input(self, model_input: dict) -> dict:
-        warnings.warn("""
-            Providing the input as a dict is deprecated. Support for this will
-            be removed in the v1.2.0 release of Outlines. Use a list containing
-            a text prompt and assets (`Image`, `Audio` or `Video` instances)
-            instead.
-            For instance:
-            ```python
-            from outlines import Image
-            model = from_transformers(mymodel, myprocessor)
-            response = model([
-                "A beautiful image of a cat",
-                Image(my_image),
-            ])
-            ```
-            """,
-            DeprecationWarning,
-            stacklevel=2,
-        )
-        if "text" not in model_input:
-            raise ValueError(
-                "The input must contain the 'text' key along with the other "
-                + "keys required by your processor."
-            )
-        return model_input
-
     @format_input.register(Chat)
     def format_chat_input(self, model_input: Chat) -> dict:
         # we need to separate the assets from the messages
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index 10df1b0e9..d8dc49f03 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -43,9 +43,8 @@ def test_transformers_multimodal_type_adapter_format_input(adapter, image):
     with pytest.raises(TypeError):
         adapter.format_input("hello")
 
-    with pytest.raises(ValueError):
-        with pytest.deprecated_call():
-            adapter.format_input({"foo": "bar"})
+    with pytest.raises(TypeError):
+        adapter.format_input({"foo": "bar"})
 
     with pytest.raises(ValueError, match="All assets must be of the same type"):
         adapter.format_input(["foo", Image(image), Video("")])

From eafa8c427de51dac7200dfd149ad5ac1723b5197 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 14:30:03 +0200
Subject: [PATCH 449/505] =?UTF-8?q?Update=20docstring=20with=20OpenAI?=
 =?UTF-8?q?=E2=80=99s=20multi-modal=20chat=20format=20support,=20and=20upd?=
 =?UTF-8?q?ate=20tests=20to=20verify=20new=20functionality=20for=20multimo?=
 =?UTF-8?q?dal=20chat?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 outlines/inputs.py   |  9 ++++++---
 tests/test_inputs.py | 21 +++++++++++++++++++++
 2 files changed, 27 insertions(+), 3 deletions(-)

diff --git a/outlines/inputs.py b/outlines/inputs.py
index 50a6a6741..37c93e097 100644
--- a/outlines/inputs.py
+++ b/outlines/inputs.py
@@ -80,8 +80,11 @@ class Chat:
 
     Each message contained in the messages list must be a dict with 'role' and
     'content' keys. The role can be 'user', 'assistant', or 'system'. The content
-    can be a string or a list containing a str and assets (images, videos,
-    audios, etc.) in the case of multimodal models.
+    supports either:
+    - a text string, or
+    - a list containing text and assets (e.g., ["Describe...", Image(...)]), or
+    - a list of dict items with explicit types (e.g.,
+      [{"type": "text", "text": "Describe..."}, {"type": "image", "image": Image(...)}])
 
     Examples
     --------
@@ -95,7 +98,7 @@ class Chat:
     chat_prompt.add_user_message(["Describe the image below", Image(image)])
 
     # Add as an assistant message the response from the model.
-    chat_prompt.add_assistant_message("The is a black cat sitting on a couch.")
+    chat_prompt.add_assistant_message("There is a black cat sitting on a couch.")
     ```
 
     Parameters
diff --git a/tests/test_inputs.py b/tests/test_inputs.py
index d020a48c9..e3b93c91f 100644
--- a/tests/test_inputs.py
+++ b/tests/test_inputs.py
@@ -141,6 +141,13 @@ def test_chat_add_system_message(image_input):
     assert chat.messages[0]["role"] == "system"
     assert chat.messages[0]["content"] == ["prompt", image_input]
 
+    # Add a list of dict items with explicit types
+    chat = Chat(messages=[])
+    chat.add_system_message([{"type": "text", "text": "prompt"}, {"type": "image", "image": image_input}])
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "system"
+    assert chat.messages[0]["content"] == [{"type": "text", "text": "prompt"}, {"type": "image", "image": image_input}]
+
 
 def test_add_user_message_string(image_input):
     # Add a string
@@ -157,6 +164,13 @@ def test_add_user_message_string(image_input):
     assert chat.messages[0]["role"] == "user"
     assert chat.messages[0]["content"] == ["prompt", image_input]
 
+    # Add a list of dict items with explicit types
+    chat = Chat(messages=[])
+    chat.add_user_message([{"type": "text", "text": "prompt"}, {"type": "image", "image": image_input}])
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "user"
+    assert chat.messages[0]["content"] == [{"type": "text", "text": "prompt"}, {"type": "image", "image": image_input}]
+
 
 def test_add_assistant_message_string(image_input):
     # Add a string
@@ -172,3 +186,10 @@ def test_add_assistant_message_string(image_input):
     assert len(chat.messages) == 1
     assert chat.messages[0]["role"] == "assistant"
     assert chat.messages[0]["content"] == ["prompt", image_input]
+
+    # Add a list of dict items with explicit types
+    chat = Chat(messages=[])
+    chat.add_assistant_message([{"type": "text", "text": "prompt"}, {"type": "image", "image": image_input}])
+    assert len(chat.messages) == 1
+    assert chat.messages[0]["role"] == "assistant"
+    assert chat.messages[0]["content"] == [{"type": "text", "text": "prompt"}, {"type": "image", "image": image_input}]

From 95dd983bd2c60d0a90228f20d360ecf31940ea2d Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 17:11:05 +0200
Subject: [PATCH 450/505] Clarify docstring for Chat class

---
 outlines/inputs.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/inputs.py b/outlines/inputs.py
index 37c93e097..658242662 100644
--- a/outlines/inputs.py
+++ b/outlines/inputs.py
@@ -83,7 +83,7 @@ class Chat:
     supports either:
     - a text string, or
     - a list containing text and assets (e.g., ["Describe...", Image(...)]), or
-    - a list of dict items with explicit types (e.g.,
+    - only for HuggingFace transformers models, a list of dict items with explicit types (e.g.,
       [{"type": "text", "text": "Describe..."}, {"type": "image", "image": Image(...)}])
 
     Examples

From b1244756a2dcd8b2a702171bad9465bff02dd561 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 17:16:29 +0200
Subject: [PATCH 451/505] Enhance asset handling in format_chat_input method to
 validate asset types and raise errors for unsupported content types in
 TransformersMultiModalTypeAdapter

---
 outlines/models/transformers.py | 26 ++++++++++++++++++++++----
 1 file changed, 22 insertions(+), 4 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 3729e5074..b95090e9f 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -432,15 +432,33 @@ def format_input(self, model_input):
 
     @format_input.register(Chat)
     def format_chat_input(self, model_input: Chat) -> dict:
-        # we need to separate the assets from the messages
-        # to apply the chat template to the messages without assets
+        SUPPORTED_ASSETS = {
+            "type": ["image", "video", "audio"],
+            "class": (Image, Video, Audio)
+        }
         messages = model_input.messages
         assets = []
+
+        # we need to collect assets from model_input messages for HF processor
         for message in messages:
             if isinstance(message["content"], list):
                 for item in message["content"]:
-                    if item["type"] != "text":
-                        assets.append(item[item["type"]])
+                    if item["type"] == "text":
+                        continue
+                    elif item["type"] in SUPPORTED_ASSETS["type"]:
+                        asset_key = item["type"]
+                        if isinstance(item[asset_key], SUPPORTED_ASSETS["class"]):
+                            assets.append(item[asset_key])
+                        else:
+                            raise ValueError(
+                                f"Assets must be of type {SUPPORTED_ASSETS['class']}. "
+                                + f"Unsupported asset type: {asset_key}"
+                            )
+                    else:
+                        raise ValueError(
+                            f"Content must be 'text', {SUPPORTED_ASSETS["type"]}. "
+                            + f"Unsupported content type: {item['type']}")
+
         formatted_prompt = self.tokenizer.apply_chat_template(
             messages, # full message for applying chat template
             tokenize=False,

From 1bf51e84ef1652ce74135af23628f818a58f0bbc Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 17:17:26 +0200
Subject: [PATCH 452/505] Update tests for transformers MultiModal to the
 multimodal chat format

---
 tests/models/test_transformers_multimodal.py  | 27 +++++--------------
 ...st_transformers_multimodal_type_adapter.py |  2 +-
 2 files changed, 7 insertions(+), 22 deletions(-)

diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 41ff058e9..42a6ff44d 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -43,8 +43,6 @@ def model():
         LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
     )
-    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
-    model.type_adapter.tokenizer.chat_template = chat_template
 
     return model
 
@@ -100,8 +98,8 @@ def test_transformers_multimodal_chat(model, image):
             {
                 "role": "user",
                 "content": [
-                    "Describe this image in one sentence:<image>",
-                    Image(image),
+                    {"type": "text", "text": "What's on this image?"},
+                    {"type": "image", "image": Image(images[0])},
                 ],
             },
         ]),
@@ -221,8 +219,8 @@ def test_transformers_multimodal_batch(model, image):
                 {
                     "role": "user",
                     "content": [
-                        "Describe this image in one sentence:<image>",
-                        Image(image),
+                        {"type": "text", "text": "What's on this image?"},
+                        {"type": "image", "image": Image(images[0])},
                     ],
                 },
             ]),
@@ -231,8 +229,8 @@ def test_transformers_multimodal_batch(model, image):
                 {
                     "role": "user",
                     "content": [
-                        "Describe this image in one sentence:<image>",
-                        Image(image),
+                        {"type": "text", "text": "What's on this image?"},
+                        {"type": "image", "image": Image(images[1])},
                     ],
                 },
             ]),
@@ -241,16 +239,3 @@ def test_transformers_multimodal_batch(model, image):
     )
     assert isinstance(result, list)
     assert len(result) == 2
-
-
-def test_transformers_multimodal_deprecated_input_type(model, image):
-    with pytest.warns(DeprecationWarning):
-        result = model.generate(
-            {
-                "text": "<image>Describe this image in one sentence:",
-                "image": image,
-            },
-            None,
-            max_new_tokens=2,
-        )
-        assert isinstance(result, str)
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index d8dc49f03..b715eae38 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -63,7 +63,7 @@ class MockAsset:
 
     chat_prompt = Chat(messages=[
         {"role": "system", "content": "foo"},
-        {"role": "user", "content": ["bar", image_asset]},
+        {"role": "user", "content": [{"type": "text", "text": "bar"}, {"type": "image", "image": image_asset}]},
     ])
     result = adapter.format_input(chat_prompt)
     assert isinstance(result, dict)

From 64e0709e1dcbf2412ebe95e56c01242eabc7f374 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 18:07:41 +0200
Subject: [PATCH 453/505] Update transformers multimodal docs to reflect new
 changes with Chat and Batching examples

---
 .../models/transformers_multimodal.md         | 155 +++++++++++++-----
 1 file changed, 111 insertions(+), 44 deletions(-)

diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index 42bc75e0e..d8399d6b2 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -10,18 +10,18 @@ The Outlines `TransformersMultiModal` model inherits from `Transformers` and sha
 
 To load the model, you can use the `from_transformers` function. It takes 2 arguments:
 
-- `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
+- `model`: a `transformers` model (created with `AutoModelForImageTextToText` for instance)
 - `tokenizer_or_processor`: a `transformers` processor (created with `AutoProcessor` for instance, it must be an instance of `ProcessorMixin`)
 
 For instance:
 
 ```python
 import outlines
-from transformers import AutoModelForCausalLM, AutoProcessor
+from transformers import AutoModelForImageTextToText, AutoProcessor
 
 # Create the transformers model and processor
-hf_model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
-hf_processor = AutoProcessor.from_pretrained("microsoft/Phi-3-mini-4k-instruct")
+hf_model = AutoModelForImageTextToText.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct")
+hf_processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct")
 
 # Create the Outlines model
 model = outlines.from_transformers(hf_model, hf_processor)
@@ -76,54 +76,66 @@ result = model(
 print(result) # '{"specie": "cat", "color": "white", "weight": 4}'
 print(Animal.model_validate_json(result)) # specie=cat, color=white, weight=4
 ```
+!!! Warning
+
+    Make sure your prompt contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are image assets included in your model input. `Chat` method, instead, does not require this step.
+
 
-The `TransformersMultiModal` model supports batch generation. To use it, invoke the `batch` method with a list of lists. You will receive as a result a list of completions.
+### Chat
+The `Chat` interface offers a more convenient way to work with multimodal inputs. You don't need to manually add asset tags like `<image>`. The model's HF processor handles the chat templating and asset placement for you automatically. 
+To do so, call the model with a `Chat` instance using a multimodal chat format. Assets must be pre-processed as `outlines.inputs.{Image, Audio, Video}` format, and only `image`, `video`, and `audio` types are supported.
 
 For instance:
 
 ```python
+import outlines
+from outlines.inputs import Chat, Image
+from transformers import AutoModelForImageTextToText, AutoProcessor
+from PIL import Image as PILImage
 from io import BytesIO
 from urllib.request import urlopen
+import torch
 
-from PIL import Image as PILImage
-from transformers import (
-    LlavaForConditionalGeneration,
-    AutoProcessor,
-)
-
-import outlines
-from outlines.inputs import Image
-
-TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
-IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
-IMAGE_URL_2 ="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
+model_kwargs = {
+        "torch_dtype": torch.bfloat16,
+        "attn_implementation": "flash_attention_2",
+        "device_map": "auto",
+    }
 
 def get_image_from_url(image_url):
     img_byte_stream = BytesIO(urlopen(image_url).read())
     image = PILImage.open(img_byte_stream).convert("RGB")
     image.format = "PNG"
     return image
-
-# Create a model
+    
+# Create the model
 model = outlines.from_transformers(
-    LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
-    AutoProcessor.from_pretrained(TEST_MODEL),
+    AutoModelForImageTextToText.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", **model_kwargs),
+    AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", **model_kwargs)
 )
 
-# Call the batch method with a list of model input dicts
-result = model.batch(
-    [
-        ["<image>Describe the image.", Image(get_image_from_url(IMAGE_URL))],
-        ["<image>Describe the image.", Image(get_image_from_url(IMAGE_URL_2))],
-    ]
-)
-print(result) # ['The image shows a cat', 'The image shows an astronaut']
+IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+
+# Create the chat mutimodal input
+prompt = Chat([
+    {
+        "role": "user",
+        "content": [
+            {"type": "image", "image": Image(get_image_from_url(IMAGE_URL))},
+            {"type": "text", "text": "Describe the image in few words."}
+        ],
+    }
+])
+
+# Call the model to generate a response
+response = model(prompt, max_new_tokens=50)
+print(response) # 'A Siamese cat with blue eyes is sitting on a cat tree, looking alert and curious.'
 ```
 
-### Chat
-You can use chat inputs with the `TransformersMultiModal` model. To do so, call the model with a `Chat` instance. 
+### Batching
+The `TransformersMultiModal` model supports batching through the `batch` method. To use it, provide a list of prompts (using the formats described above) to the `batch` method. You will receive as a result a list of completions.
 
-For instance:
+An example using the Chat format:
 
 ```python
 import outlines
@@ -133,6 +145,7 @@ from PIL import Image as PILImage
 from io import BytesIO
 from urllib.request import urlopen
 import torch
+from pydantic import BaseModel
 
 model_kwargs = {
         "torch_dtype": torch.bfloat16,
@@ -140,11 +153,14 @@ model_kwargs = {
         "device_map": "auto",
     }
 
+class Animal(BaseModel):
+    animal: str
+    color: str
+
 def get_image_from_url(image_url):
     img_byte_stream = BytesIO(urlopen(image_url).read())
     image = PILImage.open(img_byte_stream).convert("RGB")
     image.format = "PNG"
-    image.save("image.png")
     return image
     
 # Create the model
@@ -153,25 +169,76 @@ model = outlines.from_transformers(
     AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", **model_kwargs)
 )
 
-IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+IMAGE_URL_1 = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+IMAGE_URL_2 = "https://upload.wikimedia.org/wikipedia/commons/a/af/Golden_retriever_eating_pigs_foot.jpg"
 
-# Create the chat mutimodal input
-prompt = Chat([
+# Create the chat mutimodal messages
+messages = [
     {
         "role": "user",
         "content": [
-            {"type": "image", "image": Image(get_image_from_url(IMAGE_URL))},
-            {"type": "text", "text": "Describe the image in few words."}
+            {"type": "text", "text": "Describe the image in few words."},
+            {"type": "image", "image": Image(get_image_from_url(IMAGE_URL_1))},
         ],
-    }
-])
+    },
+]
+
+messages_2 = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "text", "text": "Describe the image in few words."},
+            {"type": "image", "image": Image(get_image_from_url(IMAGE_URL_2))},
+        ],
+    },
+]
+
+prompts = [Chat(messages), Chat(messages_2)]
 
 # Call the model to generate a response
-response = model(prompt, max_new_tokens=50)
-print(response) # 'A Siamese cat with blue eyes is sitting on a cat tree, looking alert and curious.'
+responses = model.batch(prompts, output_type=Animal, max_new_tokens=100)
+print(responses) # ['{ "animal": "cat", "color": "white and gray" }', '{ "animal": "dog", "color": "white" }']
+print([Animal.model_validate_json(i) for i in responses]) # [Animal(animal='cat', color='white and gray'), Animal(animal='dog', color='white')]
 ```
 
 
-!!! Warning
+An example using a list of lists with tag assets:
 
-    Make sure your prompt contains the tags expected by your processor to correctly inject the assets in the prompt. For some vision multimodal models for instance, you need to add as many `<image>` tags in your prompt as there are image assets included in your model input.
+```python
+from io import BytesIO
+from urllib.request import urlopen
+
+from PIL import Image as PILImage
+from transformers import (
+    LlavaForConditionalGeneration,
+    AutoProcessor,
+)
+
+import outlines
+from outlines.inputs import Image
+
+TEST_MODEL = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
+IMAGE_URL = "https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"
+IMAGE_URL_2 ="https://upload.wikimedia.org/wikipedia/commons/9/98/Aldrin_Apollo_11_original.jpg"
+
+def get_image_from_url(image_url):
+    img_byte_stream = BytesIO(urlopen(image_url).read())
+    image = PILImage.open(img_byte_stream).convert("RGB")
+    image.format = "PNG"
+    return image
+
+# Create a model
+model = outlines.from_transformers(
+    LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
+    AutoProcessor.from_pretrained(TEST_MODEL),
+)
+
+# Call the batch method with a list of model input dicts
+result = model.batch(
+    [
+        ["<image>Describe the image.", Image(get_image_from_url(IMAGE_URL))],
+        ["<image>Describe the image.", Image(get_image_from_url(IMAGE_URL_2))],
+    ]
+)
+print(result) # ['The image shows a cat', 'The image shows an astronaut']
+```

From c860b8892e9597c20c0d90b75ad90d259018c274 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 20:55:56 +0200
Subject: [PATCH 454/505] Use consistent string quoting for SUPPORTED_ASSETS to
 avoid Python f-string errors

---
 outlines/models/transformers.py | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index b95090e9f..76da40fcd 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -445,9 +445,9 @@ def format_chat_input(self, model_input: Chat) -> dict:
                 for item in message["content"]:
                     if item["type"] == "text":
                         continue
-                    elif item["type"] in SUPPORTED_ASSETS["type"]:
+                    elif item["type"] in SUPPORTED_ASSETS['type']:
                         asset_key = item["type"]
-                        if isinstance(item[asset_key], SUPPORTED_ASSETS["class"]):
+                        if isinstance(item[asset_key], SUPPORTED_ASSETS['class']):
                             assets.append(item[asset_key])
                         else:
                             raise ValueError(
@@ -456,7 +456,7 @@ def format_chat_input(self, model_input: Chat) -> dict:
                             )
                     else:
                         raise ValueError(
-                            f"Content must be 'text', {SUPPORTED_ASSETS["type"]}. "
+                            f"Content must be 'text', {SUPPORTED_ASSETS['type']}. "
                             + f"Unsupported content type: {item['type']}")
 
         formatted_prompt = self.tokenizer.apply_chat_template(

From 1ef8f5c58106a70ec04c2627f8e9e1cc2dbdca30 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 21:43:31 +0200
Subject: [PATCH 455/505] Add tests for handling empty assets and invalid
 content types in transformers multimodal type adapter

---
 ...st_transformers_multimodal_type_adapter.py | 32 +++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index b715eae38..4202a82ea 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -73,6 +73,38 @@ class MockAsset:
     assert result["images"][0] == image_asset.image
 
 
+def test_transformers_multimodal_type_adapter_format_input_empty_assets(adapter):
+    result = adapter.format_input(["Just text prompt"])
+    assert result == {"text": "Just text prompt"}
+
+
+def test_transformers_multimodal_type_adapter_format_input_chat_invalid_asset_type(adapter, image):
+    class MockAsset:
+        pass
+    
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": [
+            {"type": "text", "text": "Hello"},
+            {"type": "image", "image": MockAsset()}  # Wrong type
+        ]}
+    ])
+    
+    with pytest.raises(ValueError, match="Assets must be of type"):
+        adapter.format_input(chat_prompt)
+
+
+def test_transformers_multimodal_type_adapter_format_input_chat_unsupported_content_type(adapter):
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": [
+            {"type": "text", "text": "Hello"},
+            {"type": "unsupported", "data": "some_data"}  # Unsupported type
+        ]}
+    ])
+    
+    with pytest.raises(ValueError, match="Content must be 'text'"):
+        adapter.format_input(chat_prompt)
+
+
 def test_transformers_multimodal_type_adapter_format_output_type(
     adapter, logits_processor
 ):

From a14151f9d838e97190e018736e93a90040c0c3c1 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 22:00:17 +0200
Subject: [PATCH 456/505] Removing SUPPORTED_ASSETS constant for clarity and to
 avoid pre-commit errors

---
 outlines/models/transformers.py | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 76da40fcd..b827fcdd2 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -432,10 +432,6 @@ def format_input(self, model_input):
 
     @format_input.register(Chat)
     def format_chat_input(self, model_input: Chat) -> dict:
-        SUPPORTED_ASSETS = {
-            "type": ["image", "video", "audio"],
-            "class": (Image, Video, Audio)
-        }
         messages = model_input.messages
         assets = []
 
@@ -445,18 +441,18 @@ def format_chat_input(self, model_input: Chat) -> dict:
                 for item in message["content"]:
                     if item["type"] == "text":
                         continue
-                    elif item["type"] in SUPPORTED_ASSETS['type']:
+                    elif item["type"] in ["image", "video", "audio"]:
                         asset_key = item["type"]
-                        if isinstance(item[asset_key], SUPPORTED_ASSETS['class']):
+                        if isinstance(item[asset_key], (Image, Video, Audio)):
                             assets.append(item[asset_key])
                         else:
                             raise ValueError(
-                                f"Assets must be of type {SUPPORTED_ASSETS['class']}. "
+                                "Assets must be of type `Image`, `Video` or `Audio`. "
                                 + f"Unsupported asset type: {asset_key}"
                             )
                     else:
                         raise ValueError(
-                            f"Content must be 'text', {SUPPORTED_ASSETS['type']}. "
+                            "Content must be 'text', 'image', 'video' or 'audio'. "
                             + f"Unsupported content type: {item['type']}")
 
         formatted_prompt = self.tokenizer.apply_chat_template(
@@ -474,7 +470,7 @@ def format_list_input(self, model_input: list) -> dict:
 
         if not assets:  # handle empty assets case
             return {"text": prompt}
-        
+
         asset_types = set(type(asset) for asset in assets)
         if len(asset_types) > 1:
             raise ValueError(

From 79e2332744c88abc2729ebe54a3e12198b356368 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Sat, 23 Aug 2025 22:09:12 +0200
Subject: [PATCH 457/505] Apply pre-commit suggestions: trim trailing
 whitespace

---
 docs/features/models/transformers_multimodal.md           | 6 +++---
 tests/models/test_transformers_multimodal_type_adapter.py | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index d8399d6b2..5375773fc 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -82,7 +82,7 @@ print(Animal.model_validate_json(result)) # specie=cat, color=white, weight=4
 
 
 ### Chat
-The `Chat` interface offers a more convenient way to work with multimodal inputs. You don't need to manually add asset tags like `<image>`. The model's HF processor handles the chat templating and asset placement for you automatically. 
+The `Chat` interface offers a more convenient way to work with multimodal inputs. You don't need to manually add asset tags like `<image>`. The model's HF processor handles the chat templating and asset placement for you automatically.
 To do so, call the model with a `Chat` instance using a multimodal chat format. Assets must be pre-processed as `outlines.inputs.{Image, Audio, Video}` format, and only `image`, `video`, and `audio` types are supported.
 
 For instance:
@@ -107,7 +107,7 @@ def get_image_from_url(image_url):
     image = PILImage.open(img_byte_stream).convert("RGB")
     image.format = "PNG"
     return image
-    
+
 # Create the model
 model = outlines.from_transformers(
     AutoModelForImageTextToText.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", **model_kwargs),
@@ -162,7 +162,7 @@ def get_image_from_url(image_url):
     image = PILImage.open(img_byte_stream).convert("RGB")
     image.format = "PNG"
     return image
-    
+
 # Create the model
 model = outlines.from_transformers(
     AutoModelForImageTextToText.from_pretrained("Qwen/Qwen2.5-VL-3B-Instruct", **model_kwargs),
diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index 4202a82ea..af4cbafe1 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -81,14 +81,14 @@ def test_transformers_multimodal_type_adapter_format_input_empty_assets(adapter)
 def test_transformers_multimodal_type_adapter_format_input_chat_invalid_asset_type(adapter, image):
     class MockAsset:
         pass
-    
+
     chat_prompt = Chat(messages=[
         {"role": "user", "content": [
             {"type": "text", "text": "Hello"},
             {"type": "image", "image": MockAsset()}  # Wrong type
         ]}
     ])
-    
+
     with pytest.raises(ValueError, match="Assets must be of type"):
         adapter.format_input(chat_prompt)
 
@@ -100,7 +100,7 @@ def test_transformers_multimodal_type_adapter_format_input_chat_unsupported_cont
             {"type": "unsupported", "data": "some_data"}  # Unsupported type
         ]}
     ])
-    
+
     with pytest.raises(ValueError, match="Content must be 'text'"):
         adapter.format_input(chat_prompt)
 

From dedd1b953cd457b61557081368a8147517e949f1 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Tue, 26 Aug 2025 19:27:10 +0200
Subject: [PATCH 458/505] Remove or from Chat docstring for better readability

---
 outlines/inputs.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/inputs.py b/outlines/inputs.py
index 658242662..54b17dd49 100644
--- a/outlines/inputs.py
+++ b/outlines/inputs.py
@@ -81,8 +81,8 @@ class Chat:
     Each message contained in the messages list must be a dict with 'role' and
     'content' keys. The role can be 'user', 'assistant', or 'system'. The content
     supports either:
-    - a text string, or
-    - a list containing text and assets (e.g., ["Describe...", Image(...)]), or
+    - a text string,
+    - a list containing text and assets (e.g., ["Describe...", Image(...)]),
     - only for HuggingFace transformers models, a list of dict items with explicit types (e.g.,
       [{"type": "text", "text": "Describe..."}, {"type": "image", "image": Image(...)}])
 

From edcff4f19f9d5e066800370916d54ef25b21d1d7 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 10:12:10 +0200
Subject: [PATCH 459/505] Update format_chat_input for Chat class to support
 list of string + assets and the mm chat template format

---
 outlines/models/transformers.py | 102 +++++++++++++++++++++++++-------
 1 file changed, 81 insertions(+), 21 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index b827fcdd2..e0a5c753f 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -432,37 +432,97 @@ def format_input(self, model_input):
 
     @format_input.register(Chat)
     def format_chat_input(self, model_input: Chat) -> dict:
-        messages = model_input.messages
+        conversation = []
         assets = []
 
-        # we need to collect assets from model_input messages for HF processor
-        for message in messages:
-            if isinstance(message["content"], list):
-                for item in message["content"]:
-                    if item["type"] == "text":
-                        continue
-                    elif item["type"] in ["image", "video", "audio"]:
-                        asset_key = item["type"]
-                        if isinstance(item[asset_key], (Image, Video, Audio)):
-                            assets.append(item[asset_key])
-                        else:
-                            raise ValueError(
-                                "Assets must be of type `Image`, `Video` or `Audio`. "
-                                + f"Unsupported asset type: {asset_key}"
-                            )
-                    else:
-                        raise ValueError(
-                            "Content must be 'text', 'image', 'video' or 'audio'. "
-                            + f"Unsupported content type: {item['type']}")
+        # process each message, convert if needed to standardized multimodal chat template format
+        # and collect assets for HF processor
+        for message in model_input.messages:
+            processed_message, message_assets = self._prepare_message(
+                message["role"], message["content"]
+            )
+            conversation.append(processed_message)
+            assets.extend(message_assets)
 
         formatted_prompt = self.tokenizer.apply_chat_template(
-            messages, # full message for applying chat template
+            conversation,
             tokenize=False,
             add_generation_prompt=True
         )
         # use the formatted prompt and the assets to format the input
         return self.format_list_input([formatted_prompt, *assets])
 
+    def _prepare_message(self, role: str, content: str | list) -> tuple[dict, list]:
+        """Create a message."""
+        if isinstance(content, str):
+            return {"role": role, "content": content}, []
+
+        elif isinstance(content, list):
+            if all(isinstance(item, dict) for item in content): # HF multimodal chat template
+                return {"role": role, "content": content}, self._extract_assets_from_content(content)
+            else: # list of string + assets
+                prompt = content[0]
+                assets = content[1:]
+                assets_dict = [self._format_asset_for_template(asset) for asset in assets]
+
+                return {"role": role, "content": [
+                    {"type": "text", "text": prompt},
+                    *assets_dict
+                ]}, assets
+        else:
+            raise ValueError(
+                f"Invalid content type: {type(content)}. "
+                "The content must be a string or a list containing text and assets "
+                "or a list of dict items with explicit types."
+            )
+
+    def _extract_assets_from_content(self, content: list) -> list:
+        """Process a list of dict items."""
+        assets = []
+        for item in content:
+            if "type" not in item:
+                raise ValueError(
+                    "Each item in the content list must be a dictionary with a 'type' key. "
+                    "Valid types are 'text', 'image', 'video', or 'audio'. "
+                    f"For instance {{'type': 'text', 'text': 'your message'}}. "
+                    f"Found item without 'type' key: {item}"
+                )
+            if item["type"] == "text":
+                continue
+            elif item["type"] in ["image", "video", "audio"]:
+                asset_key = item["type"]
+                if asset_key not in item:
+                    raise ValueError(
+                        f"Item with type '{asset_key}' must contain a '{asset_key}' key. "
+                        f"Found item: {item}"
+                    )
+                if isinstance(item[asset_key], (Image, Video, Audio)):
+                    assets.append(item[asset_key])
+                else:
+                    raise ValueError(
+                        "Assets must be of type `Image`, `Video` or `Audio`. "
+                        + f"Unsupported asset type: {type(item[asset_key])}"
+                    )
+            else:
+                raise ValueError(
+                    "Content must be 'text', 'image', 'video' or 'audio'. "
+                    + f"Unsupported content type: {item['type']}")
+        return assets
+
+    def _format_asset_for_template(self, asset: Image | Video | Audio) -> dict:
+        """Process an asset."""
+        if isinstance(asset, Image):
+            return {"type": "image", "image": asset}
+        elif isinstance(asset, Video):
+            return {"type": "video", "video": asset}
+        elif isinstance(asset, Audio):
+            return {"type": "audio", "audio": asset}
+        else:
+            raise ValueError(
+                "Assets must be of type `Image`, `Video` or `Audio`. "
+                + f"Unsupported asset type: {type(asset)}"
+            )
+
     @format_input.register(list)
     def format_list_input(self, model_input: list) -> dict:
         prompt = model_input[0]

From 68d17fd257c0061e3813d76511181ca74ec2aaa9 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 10:13:40 +0200
Subject: [PATCH 460/505] Reintroduce test for list of string+assets

---
 tests/models/test_transformers_multimodal.py | 49 ++++++++++++++++++--
 1 file changed, 46 insertions(+), 3 deletions(-)

diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index 42a6ff44d..f17c0473a 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -92,6 +92,21 @@ def test_transformers_multimodal_wrong_input_type(model):
 
 
 def test_transformers_multimodal_chat(model, image):
+    result = model(
+        Chat(messages=[
+            {"role": "system", "content": "You are a helpful assistant."},
+            {
+                "role": "user",
+                "content": [
+                    "What's on this image?",
+                    Image(image),
+                ],
+            },
+        ]),
+        max_new_tokens=2,
+    )
+    assert isinstance(result, str)
+
     result = model(
         Chat(messages=[
             {"role": "system", "content": "You are a helpful assistant."},
@@ -99,7 +114,7 @@ def test_transformers_multimodal_chat(model, image):
                 "role": "user",
                 "content": [
                     {"type": "text", "text": "What's on this image?"},
-                    {"type": "image", "image": Image(images[0])},
+                    {"type": "image", "image": Image(image)},
                 ],
             },
         ]),
@@ -212,6 +227,34 @@ def test_transformers_multimodal_batch(model, image):
         assert isinstance(item, list)
         assert len(item) == 2
 
+    result = model.batch(
+        [
+            Chat(messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {
+                    "role": "user",
+                    "content": [
+                        "What's on this image?",
+                        Image(image),
+                    ],
+                },
+            ]),
+            Chat(messages=[
+                {"role": "system", "content": "You are a helpful assistant."},
+                {
+                    "role": "user",
+                    "content": [
+                        "What's on this image?",
+                        Image(image),
+                    ],
+                },
+            ]),
+        ],
+        max_new_tokens=2,
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+
     result = model.batch(
         [
             Chat(messages=[
@@ -220,7 +263,7 @@ def test_transformers_multimodal_batch(model, image):
                     "role": "user",
                     "content": [
                         {"type": "text", "text": "What's on this image?"},
-                        {"type": "image", "image": Image(images[0])},
+                        {"type": "image", "image": Image(image)},
                     ],
                 },
             ]),
@@ -230,7 +273,7 @@ def test_transformers_multimodal_batch(model, image):
                     "role": "user",
                     "content": [
                         {"type": "text", "text": "What's on this image?"},
-                        {"type": "image", "image": Image(images[1])},
+                        {"type": "image", "image": Image(image)},
                     ],
                 },
             ]),

From 5673580e93360cf40d292ced502e0b70f3bc6f4d Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 10:15:00 +0200
Subject: [PATCH 461/505] Reintroduce test for list of string+assets and add
 new tests for missing asset and type keys

---
 ...st_transformers_multimodal_type_adapter.py | 54 +++++++++++++++++++
 1 file changed, 54 insertions(+)

diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index af4cbafe1..0e479c2aa 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -61,6 +61,17 @@ class MockAsset:
         "images": [image_asset.image],
     }
 
+    chat_prompt = Chat(messages=[
+        {"role": "system", "content": "foo"},
+        {"role": "user", "content": ["bar", image_asset]},
+    ])
+    result = adapter.format_input(chat_prompt)
+    assert isinstance(result, dict)
+    assert isinstance(result["text"], str)
+    assert isinstance(result["images"], list)
+    assert len(result["images"]) == 1
+    assert result["images"][0] == image_asset.image
+
     chat_prompt = Chat(messages=[
         {"role": "system", "content": "foo"},
         {"role": "user", "content": [{"type": "text", "text": "bar"}, {"type": "image", "image": image_asset}]},
@@ -73,6 +84,8 @@ class MockAsset:
     assert result["images"][0] == image_asset.image
 
 
+
+
 def test_transformers_multimodal_type_adapter_format_input_empty_assets(adapter):
     result = adapter.format_input(["Just text prompt"])
     assert result == {"text": "Just text prompt"}
@@ -116,3 +129,44 @@ def test_transformers_multimodal_type_adapter_format_output_type(
 
     formatted = adapter.format_output_type(None)
     assert formatted is None
+
+
+def test_transformers_multimodal_type_adapter_format_input_chat_missing_asset_key(adapter, image):
+    image_asset = Image(image)
+
+    # Test missing 'image' key when type is 'image'
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": [
+            {"type": "text", "text": "What's in this image?"},
+            {"type": "image", "txt": image_asset}  # Wrong key: 'txt' instead of 'image'
+        ]}
+    ])
+
+    with pytest.raises(ValueError, match="Item with type 'image' must contain a 'image' key"):
+        adapter.format_input(chat_prompt)
+
+    # Test missing 'video' key when type is 'video'
+    video_asset = Video("dummy_video")
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": [
+            {"type": "text", "text": "What's in this video?"},
+            {"type": "video", "vid": video_asset}  # Wrong key: 'vid' instead of 'video'
+        ]}
+    ])
+
+    with pytest.raises(ValueError, match="Item with type 'video' must contain a 'video' key"):
+        adapter.format_input(chat_prompt)
+
+
+def test_transformers_multimodal_type_adapter_format_input_chat_missing_type_key(adapter, image):
+    image_asset = Image(image)
+
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": [
+            {"text": "What's in this image?"},  # Missing 'type' key
+            {"type": "image", "image": image_asset}
+        ]}
+    ])
+
+    with pytest.raises(ValueError, match="Each item in the content list must be a dictionary with a 'type' key"):
+        adapter.format_input(chat_prompt)

From 3180cbad4a8d9cb8039defbd51f1ed3edaf354e8 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 10:15:34 +0200
Subject: [PATCH 462/505] Add example for using list of string + assets with
 Chat class

---
 .../models/transformers_multimodal.md         | 52 +++++++++++++++++++
 1 file changed, 52 insertions(+)

diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index 5375773fc..3d34a0f78 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -132,6 +132,58 @@ response = model(prompt, max_new_tokens=50)
 print(response) # 'A Siamese cat with blue eyes is sitting on a cat tree, looking alert and curious.'
 ```
 
+Or using a list containing text and assets:
+
+```python
+import outlines
+from outlines.inputs import Chat, Image
+from transformers import AutoModelForImageTextToText, AutoProcessor
+from PIL import Image as PILImage
+from io import BytesIO
+import requests
+import torch
+
+
+TEST_MODEL = "Qwen/Qwen2.5-VL-7B-Instruct"
+
+# Function to get an image
+def get_image(url):
+    headers = {
+        'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
+    }
+    r = requests.get(url, headers=headers)
+    image = PILImage.open(BytesIO(r.content)).convert("RGB")
+    image.format = "PNG"
+    return image
+
+model_kwargs = {
+        "torch_dtype": torch.bfloat16,
+        # "attn_implementation": "flash_attention_2",
+        "device_map": "auto",
+    }
+
+# Create a model
+model = outlines.from_transformers(
+    AutoModelForImageTextToText.from_pretrained(TEST_MODEL, **model_kwargs),
+    AutoProcessor.from_pretrained(TEST_MODEL, **model_kwargs),
+)
+
+# Create the chat input
+prompt = Chat([
+    {"role": "user", "content": "You are a helpful assistant that helps me described pictures."},
+    {"role": "assistant", "content": "I'd be happy to help you describe pictures! Please go ahead and share an image"},
+    {
+        "role": "user",
+        "content": ["Describe briefly the image", Image(get_image("https://upload.wikimedia.org/wikipedia/commons/2/25/Siam_lilacpoint.jpg"))]
+    },
+])
+
+# Call the model to generate a response
+response = model(prompt, max_new_tokens=50)
+print(response) # 'The image shows a light-colored cat with a white chest...'
+```
+
+
 ### Batching
 The `TransformersMultiModal` model supports batching through the `batch` method. To use it, provide a list of prompts (using the formats described above) to the `batch` method. You will receive as a result a list of completions.
 

From 6d87de7c674415f13169a06e9796716d4fbc5dab Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 10:30:45 +0200
Subject: [PATCH 463/505] Add tests for invalid content types and for
 unsupported asset types

---
 ...st_transformers_multimodal_type_adapter.py | 29 +++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index 0e479c2aa..ecc15fac7 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -170,3 +170,32 @@ def test_transformers_multimodal_type_adapter_format_input_chat_missing_type_key
 
     with pytest.raises(ValueError, match="Each item in the content list must be a dictionary with a 'type' key"):
         adapter.format_input(chat_prompt)
+
+
+def test_transformers_multimodal_type_adapter_format_input_invalid_content_type(adapter):
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": 42}  # Invalid content type (integer)
+    ])
+
+    with pytest.raises(ValueError, match="Invalid content type"):
+        adapter.format_input(chat_prompt)
+
+    # Test with another invalid type
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": {"invalid": "dict"}}  # Invalid content type (dict not in list)
+    ])
+
+    with pytest.raises(ValueError, match="Invalid content type"):
+        adapter.format_input(chat_prompt)
+
+
+def test_transformers_multimodal_type_adapter_format_asset_for_template_invalid_type(adapter):
+    class MockUnsupportedAsset:
+        pass
+
+    # This test requires accessing the private method directly since the error
+    # would normally be caught earlier in the validation chain
+    unsupported_asset = MockUnsupportedAsset()
+
+    with pytest.raises(ValueError, match="Assets must be of type `Image`, `Video` or `Audio`"):
+        adapter._format_asset_for_template(unsupported_asset)

From 91ae7fa29a270948c715b92561842fb3906a19ac Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 11:04:14 +0200
Subject: [PATCH 464/505] Migrate multimodal type adapter tests to use
 AutoProcessor for a MM scenario, and added Audio and Video tests

---
 ...st_transformers_multimodal_type_adapter.py | 43 ++++++++++++++++---
 1 file changed, 37 insertions(+), 6 deletions(-)

diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index ecc15fac7..b2344cb99 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -2,22 +2,24 @@
 
 from PIL import Image as PILImage
 from outlines_core import Index, Vocabulary
-from transformers import AutoTokenizer, LogitsProcessorList
+from transformers import (
+    AutoProcessor,
+    LogitsProcessorList,
+)
 
-from outlines.inputs import Chat, Image, Video
+from outlines.inputs import Audio, Chat, Image, Video
 from outlines.models.transformers import TransformersMultiModalTypeAdapter
 from outlines.backends.outlines_core import OutlinesCoreLogitsProcessor
 
 
-MODEL_NAME = "erwanf/gpt2-mini"
+MODEL_NAME = "trl-internal-testing/tiny-LlavaForConditionalGeneration"
 
 
 @pytest.fixture
 def adapter():
-    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+    processor = AutoProcessor.from_pretrained(MODEL_NAME)
+    tokenizer = processor.tokenizer
     type_adapter = TransformersMultiModalTypeAdapter(tokenizer=tokenizer)
-    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
-    type_adapter.tokenizer.chat_template = chat_template
 
     return type_adapter
 
@@ -39,6 +41,18 @@ def image():
     return image
 
 
+@pytest.fixture
+def video():
+    # Simple mock video data
+    return "mock_video_data"
+
+
+@pytest.fixture
+def audio():
+    # Simple mock audio data
+    return "mock_audio_data"
+
+
 def test_transformers_multimodal_type_adapter_format_input(adapter, image):
     with pytest.raises(TypeError):
         adapter.format_input("hello")
@@ -189,6 +203,23 @@ def test_transformers_multimodal_type_adapter_format_input_invalid_content_type(
         adapter.format_input(chat_prompt)
 
 
+def test_transformers_multimodal_type_adapter_format_asset_for_template(adapter, image, video, audio):
+    # Test Image asset
+    image_asset = Image(image)
+    formatted_image = adapter._format_asset_for_template(image_asset)
+    assert formatted_image == {"type": "image", "image": image_asset}
+
+    # Test Video asset
+    video_asset = Video(video)
+    formatted_video = adapter._format_asset_for_template(video_asset)
+    assert formatted_video == {"type": "video", "video": video_asset}
+
+    # Test Audio asset
+    audio_asset = Audio(audio)
+    formatted_audio = adapter._format_asset_for_template(audio_asset)
+    assert formatted_audio == {"type": "audio", "audio": audio_asset}
+
+
 def test_transformers_multimodal_type_adapter_format_asset_for_template_invalid_type(adapter):
     class MockUnsupportedAsset:
         pass

From 6bec76f81d571a2763e60617cb3793afae6bdeb3 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 12:25:20 +0200
Subject: [PATCH 465/505] Added validation for items with multiple keys

---
 outlines/models/transformers.py | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index e0a5c753f..4d6387a40 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -472,20 +472,29 @@ def _prepare_message(self, role: str, content: str | list) -> tuple[dict, list]:
         else:
             raise ValueError(
                 f"Invalid content type: {type(content)}. "
-                "The content must be a string or a list containing text and assets "
-                "or a list of dict items with explicit types."
+                + "The content must be a string or a list containing text and assets "
+                + "or a list of dict items with explicit types."
             )
 
     def _extract_assets_from_content(self, content: list) -> list:
         """Process a list of dict items."""
         assets = []
+
         for item in content:
+            if len(item) > 2:
+                raise ValueError(
+                    f"Found item with multiple keys: {item}. "
+                    + "Each item in the content list must be a dictionary with a 'type' key and a single asset key. "
+                    + "To include multiple assets, use separate dictionary items. "
+                    + "For example: [{{'type': 'image', 'image': image1}}, {{'type': 'image', 'image': image2}}]. "
+                )
+
             if "type" not in item:
                 raise ValueError(
                     "Each item in the content list must be a dictionary with a 'type' key. "
-                    "Valid types are 'text', 'image', 'video', or 'audio'. "
-                    f"For instance {{'type': 'text', 'text': 'your message'}}. "
-                    f"Found item without 'type' key: {item}"
+                    + "Valid types are 'text', 'image', 'video', or 'audio'. "
+                    + "For instance {{'type': 'text', 'text': 'your message'}}. "
+                    + f"Found item without 'type' key: {item}"
                 )
             if item["type"] == "text":
                 continue
@@ -494,7 +503,7 @@ def _extract_assets_from_content(self, content: list) -> list:
                 if asset_key not in item:
                     raise ValueError(
                         f"Item with type '{asset_key}' must contain a '{asset_key}' key. "
-                        f"Found item: {item}"
+                        + f"Found item: {item}"
                     )
                 if isinstance(item[asset_key], (Image, Video, Audio)):
                     assets.append(item[asset_key])

From 99093e05cd5494c5b94a8c45afc73261d1f3170e Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Wed, 27 Aug 2025 12:26:40 +0200
Subject: [PATCH 466/505] Add tests for mutiple assets usage in multimodal chat
 template format

---
 ...st_transformers_multimodal_type_adapter.py | 36 +++++++++++++++++++
 1 file changed, 36 insertions(+)

diff --git a/tests/models/test_transformers_multimodal_type_adapter.py b/tests/models/test_transformers_multimodal_type_adapter.py
index b2344cb99..5518ef553 100644
--- a/tests/models/test_transformers_multimodal_type_adapter.py
+++ b/tests/models/test_transformers_multimodal_type_adapter.py
@@ -230,3 +230,39 @@ class MockUnsupportedAsset:
 
     with pytest.raises(ValueError, match="Assets must be of type `Image`, `Video` or `Audio`"):
         adapter._format_asset_for_template(unsupported_asset)
+
+
+def test_transformers_multimodal_type_adapter_multiple_assets_in_single_item(adapter, image):
+    image_asset = Image(image)
+    video_asset = Video("dummy_video")
+
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": [
+            {"type": "text", "text": "What's in this?"},
+            {"type": "image", "image": image_asset, "video": video_asset}  # Multiple asset types
+        ]}
+    ])
+
+    with pytest.raises(ValueError, match="Found item with multiple keys:"):
+        adapter.format_input(chat_prompt)
+
+
+
+def test_transformers_multimodal_type_adapter_correct_multiple_assets_usage(adapter, image):
+    image_asset1 = Image(image)
+    image_asset2 = Image(image)
+
+    # Correct way: separate dictionary items for each asset
+    chat_prompt = Chat(messages=[
+        {"role": "user", "content": [
+            {"type": "text", "text": "What's in these images?"},
+            {"type": "image", "image": image_asset1},
+            {"type": "image", "image": image_asset2}
+        ]}
+    ])
+
+    result = adapter.format_input(chat_prompt)
+    assert isinstance(result, dict)
+    assert "text" in result
+    assert "images" in result
+    assert len(result["images"]) == 2

From d975d1340a9435ea9ea58c994ceeaf1b0a749cd9 Mon Sep 17 00:00:00 2001
From: laitifranz <25352428+laitifranz@users.noreply.github.com>
Date: Fri, 12 Sep 2025 18:37:29 +0200
Subject: [PATCH 467/505] Update text input prompt to align with author's
 preference

---
 tests/models/test_transformers_multimodal.py | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index f17c0473a..ce968e1f6 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -98,7 +98,7 @@ def test_transformers_multimodal_chat(model, image):
             {
                 "role": "user",
                 "content": [
-                    "What's on this image?",
+                    "Describe this image in one sentence:",
                     Image(image),
                 ],
             },
@@ -113,7 +113,7 @@ def test_transformers_multimodal_chat(model, image):
             {
                 "role": "user",
                 "content": [
-                    {"type": "text", "text": "What's on this image?"},
+                    {"type": "text", "text": "Describe this image in one sentence:"},
                     {"type": "image", "image": Image(image)},
                 ],
             },
@@ -234,7 +234,7 @@ def test_transformers_multimodal_batch(model, image):
                 {
                     "role": "user",
                     "content": [
-                        "What's on this image?",
+                        "Describe this image in one sentence:",
                         Image(image),
                     ],
                 },
@@ -244,7 +244,7 @@ def test_transformers_multimodal_batch(model, image):
                 {
                     "role": "user",
                     "content": [
-                        "What's on this image?",
+                        "Describe this image in one sentence:",
                         Image(image),
                     ],
                 },
@@ -262,7 +262,7 @@ def test_transformers_multimodal_batch(model, image):
                 {
                     "role": "user",
                     "content": [
-                        {"type": "text", "text": "What's on this image?"},
+                        {"type": "text", "text": "Describe this image in one sentence:"},
                         {"type": "image", "image": Image(image)},
                     ],
                 },
@@ -272,7 +272,7 @@ def test_transformers_multimodal_batch(model, image):
                 {
                     "role": "user",
                     "content": [
-                        {"type": "text", "text": "What's on this image?"},
+                        {"type": "text", "text": "Describe this image in one sentence:"},
                         {"type": "image", "image": Image(image)},
                     ],
                 },

From 2a19bdd0ebd04e767a00615653e6caf23bc57840 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Tue, 23 Sep 2025 14:48:23 +0200
Subject: [PATCH 468/505] Add batch generation to the MLXLM model

---
 docs/features/models/index.md |  2 +-
 docs/features/models/mlxlm.md | 18 +++++++++++++++++
 outlines/models/mlxlm.py      | 38 ++++++++++++++++++++++++++++-------
 tests/models/test_mlxlm.py    | 11 ++++++----
 4 files changed, 57 insertions(+), 12 deletions(-)

diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index 8049107e7..f7969eec8 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -96,7 +96,7 @@ In alphabetical order:
 | Async | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
 | Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
-| Batching | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ✅ |
+| Batching | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ✅ |
 
 ## Model Types
 
diff --git a/docs/features/models/mlxlm.md b/docs/features/models/mlxlm.md
index 6ccf7f38a..dc34257ac 100644
--- a/docs/features/models/mlxlm.md
+++ b/docs/features/models/mlxlm.md
@@ -98,6 +98,24 @@ for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
     print(chunk) # 'In...'
 ```
 
+#### Batch Generation
+
+The `MLXLM` model supports generating text in batches. To do so, use the `batch` method and provide a list of strings as a model input. For instance:
+
+```python
+import outlines
+import mlx_lm
+
+# Load the model
+model = outlines.from_mlxlm(
+    *mlx_lm.load("mlx-community/TinyLlama-1.1B-Chat-v1.0-4bit")
+)
+
+# Generate text in batches
+result = model.batch(["What's the capital of Lithuania?", "What's the capital of Latvia?"], max_tokens=20)
+print(result) # ['Vilnius', 'Riga']
+```
+
 ## Structured Generation
 
 As a local model, `MLXLM` supports all forms of structured generation available in Outlines.
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 498fec2dc..0a91e4bd1 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -37,8 +37,8 @@ def format_input(self, model_input):
 
         """
         raise NotImplementedError(
-            f"The input type {input} is not available with mlx-lm. "
-            "The available types are `str` and `Chat`."
+            f"The input type {type(model_input)} is not available with "
+            "mlx-lm. The available types are `str` and `Chat`."
         )
 
     @format_input.register(str)
@@ -151,12 +151,36 @@ def generate(
 
     def generate_batch(
         self,
-        model_input,
-        output_type = None,
+        model_input: list[str],
+        output_type: Optional[OutlinesLogitsProcessor] = None,
         **kwargs,
-    ):
-        raise NotImplementedError(
-            "The `mlx_lm` library does not support batch inference."
+    ) -> list[str]:
+        """Generate a batch of text using `mlx-lm`.
+
+        Parameters
+        ----------
+        model_input
+            The list of prompts based on which the model will generate a response.
+        output_type
+            The logits processor the model will use to constrain the format of
+            the generated text.
+        kwargs
+            Additional keyword arguments to pass to the `mlx-lm` library.
+
+        Returns
+        -------
+        list[str]
+            The list of text generated by the model.
+
+        """
+        from mlx_lm import generate_batch
+
+        return generate_batch(
+            self.model,
+            self.mlx_tokenizer,
+            [self.type_adapter.format_input(item) for item in model_input],
+            logits_processors=self.type_adapter.format_output_type(output_type),
+            **kwargs,
         )
 
     def generate_stream(
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index fc2ec6f39..924a54551 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -121,7 +121,10 @@ def test_mlxlm_stream_text_stop(model):
 
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
 def test_mlxlm_batch(model):
-    with pytest.raises(NotImplementedError, match="does not support"):
-        model.batch(
-            ["Respond with one word.", "Respond with one word."],
-        )
+    result = model.batch(
+        ["Respond with one word.", "Respond with one word."],
+    )
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)

From 87ee5c9db1698275a91c87b7c655819f4b00a7b6 Mon Sep 17 00:00:00 2001
From: "Steven E. Elliott" <seeyallc6c@gmail.com>
Date: Fri, 26 Sep 2025 21:46:10 -0400
Subject: [PATCH 469/505] Add Mistral AI integration (models, tests, docs)

---
 docs/features/models/mistral.md           | 515 +++++++++++++++
 outlines/models/__init__.py               |   5 +
 outlines/models/mistral.py                | 631 +++++++++++++++++++
 pyproject.toml                            |   4 +
 tests/models/test_mistral.py              | 723 ++++++++++++++++++++++
 tests/models/test_mistral_type_adapter.py | 553 +++++++++++++++++
 6 files changed, 2431 insertions(+)
 create mode 100644 docs/features/models/mistral.md
 create mode 100644 outlines/models/mistral.py
 create mode 100644 tests/models/test_mistral.py
 create mode 100644 tests/models/test_mistral_type_adapter.py

diff --git a/docs/features/models/mistral.md b/docs/features/models/mistral.md
new file mode 100644
index 000000000..481e58576
--- /dev/null
+++ b/docs/features/models/mistral.md
@@ -0,0 +1,515 @@
+// # Mistral AI Models
+
+Outlines now supports models from [Mistral AI](https://mistral.ai) via the official `mistralai` package.
+This implementation provides a thin wrapper around the `mistralai.Mistral` client, handling prompt formatting, structured outputs, and streaming.
+This document describes how to use the **Mistral SDK** with Outlines, both in
+synchronous and asynchronous contexts.
+
+---
+
+## 0. Installation
+
+```bash
+pip install mistralai
+export MISTRAL_API_KEY=YOUR_API_KEY_GOES_HERE # Or in code if done securely
+```
+
+
+---
+
+## 1. Open Models
+
+Below is a list of the open-weight models available via Apache2 license released by Mistral. These can be
+used directly via the API or downloaded where applicable (as of 2025-09-19).
+
+| API Endpoint             | Version | Model                | Max Tokens | Description                                                                | Outlines Support   | Multi-Modal        |
+|--------------------------|---------|----------------------|------------|----------------------------------------------------------------------------|--------------------|--------------------|
+| `magistral-small-2509`   | 25.09   | Magistral Small 1.2  | 128k       | Small reasoning model (Sept 2025) with vision support.                     | ✔️                 | ✔️                 |
+| `magistral-small-2507`   | 25.07   | Magistral Small 1.1  | 40k        | Small reasoning model (July 2025).                                         | ✔️                 | ❌                 |
+| `voxtral-mini-2507`      | 25.07   | Voxtral Mini         | 32k        | Mini audio input model.                                                    | ❓                 | ❓                 |
+| `mistral-small-2506`     | 25.06   | Mistral Small 3.2    | 128k       | Updated small model (June 2025).                                           | ✔️                 | ✔️                 |
+| `magistral-small-2506`   | 25.06   | Magistral Small 1    | 40k        | First small reasoning model (June 2025).                                   | ✔️                 | ✔️                 |
+| `devstral-small-2507`    | 25.07   | Devstral Small 1.1   | 128k       | Tool-use model for code exploration and editing (July 2025).               | ❓                 | ❓                 |
+| `mistral-small-2503`     | 25.03   | Mistral Small 4    | 128k       | Small model with image understanding (March 2025).                         | ✔️                 | ✔️                 |
+| `mistral-small-2501`     | 25.01   | Mistral Small 3      | 32k        | Small model (Jan 2025).                                                    | ✔️                 | ❌                 |
+| `devstral-small-2505`    | 25.05   | Devstral Small 1     | 128k       | Open source 24B text model for tool-use and software agents (June 2025).   | ❓                 | ❓                 |
+| `pixtral-12b-2409`       | 24.09   | Pixtral 12B          | 128k       | 12B multimodal model with image understanding (Sept 2024).                 | ✔️                 | ✔️                 |
+| `open-mistral-nemo`      | 24.07   | Mistral Nemo 12B     | 128k       | Multilingual open source model (July 2024).                                | ✔️                 | ❓                 |
+
+
+**Mistral SDK Limitations:**
+- Regex and CFG-based structured outputs are not supported by the Mistral API.
+- Structured outputs (including Pydantic/dataclass-based ones) are unavailable for codestral-mamba models, because they lack JSON Schema support.
+
+**Outlines Expansion:**
+- Outlines provides a drop-in for Regex and CFG-based structured outputs, reducing the SDK limitations.
+
+---
+## 2. Listing Available Models via SDK
+
+The mistralai python SDK provides methods to query the API for a list of [all available models](https://docs.mistral.ai/getting-started/models/models_overview/#api-versioning),
+including **paid endpoints** for [premium models](https://docs.mistral.ai/getting-started/models/models_overview/) in addition to open weights.
+
+##### Sync Example
+```python
+#!/usr/bin/env python
+import os
+from mistralai import Mistral
+
+def main():
+    api_key = os.environ["MISTRAL_API_KEY"]
+    client = Mistral(api_key=api_key)
+    list_models_response = client.models.list()
+    print(list_models_response)
+
+if __name__ == "__main__":
+    main()
+```
+
+##### Async Example
+```python
+#!/usr/bin/env python
+import asyncio
+import os
+from mistralai import Mistral
+
+async def main():
+    api_key = os.environ["MISTRAL_API_KEY"]
+    client = Mistral(api_key=api_key)
+    list_models_response = await client.models.list_async()
+    print(list_models_response)
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+---
+## 3. Basic Usage
+
+
+### 3.1 mistralai SDK wrapper
+
+Generate outputs for the given model name :
+
+#### Sync
+```python
+from mistralai import Mistral
+from outlines.models.mistral import from_mistral
+
+def main():
+    client = Mistral(api_key="YOUR_API_KEY")
+    model = from_mistral(client, "mistral-small-2506")
+    result = model.generate("What is 2+2? Answer in one sentence.")
+    print(result)
+
+if __name__ == "__main__":
+    main()
+```
+
+#### Async
+```python
+import asyncio
+from mistralai import Mistral
+from outlines.models.mistral import from_mistral
+
+async def main():
+    client = Mistral(api_key="YOUR_API_KEY")
+    model = from_mistral(client, "mistral-small-2506", async_client=True)
+    result = await model.generate("What is 2+2? Answer in one sentence.")
+    print(result)
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+
+### 3.2 Structured Generation
+
+Outlines automatically converts model outputs into your desired format.
+
+#### Sync
+```python
+from typing import Literal
+from pydantic import BaseModel
+
+# Multiple choice (via enum in JSON schema)
+sentiment = model.generate(
+    "The movie was absolutely fantastic!",
+    output_type=Literal["positive", "negative", "neutral"]
+)
+
+# Structured object (Pydantic model)
+class Person(BaseModel):
+    name: str
+    age: int
+    occupation: str
+
+person = model.generate(
+    "Generate information about a software engineer",
+    output_type=Person
+)
+```
+
+#### Async
+```python
+import asyncio
+from typing import Literal
+from pydantic import BaseModel
+
+async def main():
+    # Multiple choice (via enum in JSON schema)
+    sentiment = await model.generate(
+        "The movie was absolutely fantastic!",
+        output_type=Literal["positive", "negative", "neutral"]
+    )
+
+    # Structured object (Pydantic model)
+    class Person(BaseModel):
+        name: str
+        age: int
+        occupation: str
+
+    person = await model.generate(
+        "Generate information about a software engineer",
+        output_type=Person
+    )
+```
+
+- Structured outputs may be **unavailable** for `codestral-mamba` models (they lack JSON schema support).
+- For best results with complex/nested schemas, use `mistral-large-latest` or higher; simpler models may require retries.
+- Check model compatibility:
+```python
+if model.supports_structured_output():
+    # Safe to use Pydantic/JSON schemas
+    pass
+else:
+    print("Use text-only generation")
+```
+
+---
+
+### 3.3 Streaming
+
+Streaming methods (`generate_stream` / async `generate_stream`) allow you to iterate over tokens as they are generated.
+
+#### Sync
+```python
+for chunk in model.generate_stream("Write a short story about AI"):
+    print(chunk, end="", flush=True)
+```
+
+#### Async
+```python
+async for chunk in model.generate_stream("Write a short story about AI"):
+    print(chunk, end="", flush=True)
+```
+
+### 3.4 Structured Output with Streaming
+
+Structured JSON is streamed as text chunks that can be assembled and parsed:
+
+#### Sync Structured Streaming
+```python
+import json
+from pydantic import BaseModel
+
+class Person(BaseModel):
+    name: str
+    age: int
+
+model = from_mistral(mistral_client, "mistral-large-latest")
+prompt = """Generate a JSON object representing a person with name and age.
+Return only the JSON, no other text."""
+
+chunks = []
+for chunk in model.generate_stream(prompt, output_type=Person):
+    chunks.append(chunk)
+
+# Assemble and parse the complete JSON
+result = "".join(chunks)
+parsed = json.loads(result)
+print(f"Name: {parsed['name']}, Age: {parsed['age']}")
+```
+
+#### Async Structured Streaming
+```python
+import asyncio
+import json
+from pydantic import BaseModel
+
+class Person(BaseModel):
+    name: str
+    age: int
+
+async def main():
+    model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
+    prompt = """Generate a JSON object representing a person with name and age.
+    Return only the JSON, no other text."""
+
+    chunks = []
+    async for chunk in model.generate_stream(prompt, output_type=Person):
+        chunks.append(chunk)
+
+    # Assemble and parse the complete JSON
+    result = "".join(chunks)
+    parsed = json.loads(result)
+    print(f"Name: {parsed['name']}, Age: {parsed['age']}")
+
+if __name__ == "__main__":
+    asyncio.run(main())
+```
+
+**Note:** With structured outputs, individual chunks may not be valid JSON until the complete response is assembled.
+
+
+### 3.5 Batch Processing
+
+Batch inference is currently **not supported** by the `mistralai` library:
+
+```python
+# This will raise NotImplementedError
+# model.generate_batch(prompts)
+
+# Use sequential processing instead
+results = [model.generate(prompt) for prompt in prompts]
+```
+
+### 3.6 Error Handling
+
+Usage of this component can raise specific exceptions for common issues:
+
+```python
+try:
+    result = model.generate("Hello", output_type=Person)
+except RuntimeError as e:
+    print(f"API Error: {e}")  # e.g., rate limits, network issues
+except TypeError as e:
+    print(f"Schema/Output Error: {e}")  # e.g., unsupported schema, model incompatibility
+except ValueError as e:
+    print(f"Input Error: {e}")  # e.g., invalid message roles or content, empty lists
+```
+
+---
+
+## 4. Advanced Usage
+
+### 4.0 Custom Inference Parameters
+
+Pass inference parameters directly to `generate`:
+
+```python
+# Sync
+result = model.generate(
+    "Write a poem",
+    temperature=0.9,  # More creative
+    max_tokens=200,
+    top_p=0.95
+)
+
+# Async
+result = await model.generate(
+    "Write a poem",
+    temperature=0.9,
+    max_tokens=200,
+    top_p=0.95
+)
+```
+
+### 4.2 Multiple Samples
+
+Generate multiple responses using the `n` parameter:
+
+```python
+# Sync
+results = model.generate("Respond with one word.", n=2)
+assert isinstance(results, list)
+assert len(results) == 2
+
+# Async
+results = await model.generate("Respond with one word.", n=2)
+assert isinstance(results, list)
+assert len(results) == 2
+```
+
+### 4.3 Multimodal Inputs (Images)
+
+Mistral supports image inputs via `content` lists (e.g., for `pixtral-*` model). Outlines handles this in `Chat` or list inputs:
+
+#### Sync Image Processing
+```python
+from outlines.inputs import Image
+from PIL import Image as PILImage
+
+# Load image
+pil_image = PILImage.open("path/to/image.jpg")
+image = Image(pil_image)
+
+# Text + image prompt
+result = model.generate(
+    ["Describe this image", image],
+    model_name="pixtral-12b-2409"
+)
+
+# Or via Chat
+chat = Chat([
+    {"role": "user", "content": ["What do you see?", image]}
+])
+result = model.generate(chat)
+```
+
+#### Async Image Processing
+```python
+import asyncio
+from outlines.inputs import Image
+from PIL import Image as PILImage
+
+async def main():
+    # Load image
+    pil_image = PILImage.open("path/to/image.jpg")
+    image = Image(pil_image)
+
+    # Text + image prompt
+    result = await model.generate(
+        ["Describe this image", image],
+        model_name="pixtral-12b-2409"
+    )
+
+    # Or via Chat
+    chat = Chat([
+        {"role": "user", "content": ["What do you see?", image]}
+    ])
+    result = await model.generate(chat)
+```
+
+**Note:** The first item must be a string, followed by optional `Image` objects.
+Text-only models will raise errors.
+
+### 4.4 Chat Messages
+
+Use `Chat` objects for multi-turn conversations:
+
+#### Sync Chat
+```python
+from outlines.inputs import Chat
+
+chat = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {"role": "user", "content": "What is Python?"},
+    {"role": "assistant", "content": "Python is a programming language..."},
+    {"role": "user", "content": "Tell me more about its uses."}
+])
+
+response = model.generate(chat)
+```
+
+#### Async Chat
+```python
+import asyncio
+from outlines.inputs import Chat
+
+async def main():
+    chat = Chat([
+        {"role": "system", "content": "You are a helpful assistant."},
+        {"role": "user", "content": "What is Python?"},
+        {"role": "assistant", "content": "Python is a programming language..."},
+        {"role": "user", "content": "Tell me more about its uses."}
+    ])
+
+    response = await model.generate(chat)
+```
+
+**Supported roles:** `system`, `user`, `assistant`. Other roles raise `ValueError`.
+
+### 4.5 Complex Nested Structures
+
+Based on integration tests, Outlines handles complex nested Pydantic models:
+
+#### Sync Complex Structure
+```python
+import json
+from pydantic import BaseModel
+from typing import List
+from outlines.inputs import Chat
+
+class Address(BaseModel):
+    street: str
+    city: str
+    zip_code: str
+
+class Employee(BaseModel):
+    name: str
+    role: str
+    department: str
+    address: Address
+
+class Company(BaseModel):
+    company_name: str
+    employees: List[Employee]
+
+model = from_mistral(mistral_client, "mistral-large-latest")
+prompt = Chat([
+    {"role": "system", "content": "You are a business data generator. All employees must work in the IT department."},
+    {"role": "user", "content": "Generate a company with at least two employees. Return only JSON."}
+])
+
+result = model.generate(prompt, output_type=Company)
+parsed = json.loads(result)
+print(f"Company: {parsed['company_name']}")
+for emp in parsed['employees']:
+    print(f"Employee: {emp['name']} - {emp['role']}")
+```
+
+#### Async Complex Structure
+```python
+import asyncio
+import json
+
+async def main():
+    # Same model definitions as above
+    model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
+    prompt = Chat([
+        {"role": "system", "content": "You are a business data generator. All employees must work in the IT department."},
+        {"role": "user", "content": "Generate a company with at least two employees. Return only JSON."}
+    ])
+
+    result = await model.generate(prompt, output_type=Company)
+    parsed = json.loads(result)
+    print(f"Company: {parsed['company_name']}")
+    for emp in parsed['employees']:
+        print(f"Employee: {emp['name']} - {emp['role']}")
+```
+
+---
+
+## 5. Troubleshooting
+
+- **ImportError: No module named 'mistralai'**
+  → Run `pip install mistralai`.
+
+- **Authentication Error**
+  → Verify `MISTRAL_API_KEY` is set and valid. Test with the [Mistral Playground](https://chat.mistral.ai).
+
+- **Schema Error (e.g., "Mistral does not support your schema")**
+  → Avoid regex/CFG; ensure no `pattern` fields in Pydantic (Outlines sets `additionalProperties: false`); try a simpler schema or different model.
+
+- **Model Not Found Error**
+  → Confirm the model name (e.g., `"mistral-small-latest"`) and your subscription tier. Check [docs](https://docs.mistral.ai/getting-started/models/).
+
+- **Rate Limits or Quotas**
+  → Monitor usage in the Mistral console; upgrade your plan for higher limits.
+
+- **Input Validation Errors**
+  → Ensure Chat messages use valid roles (`system`, `user`, `assistant`); list inputs start with strings.
+
+---
+
+## 6. Notes
+
+- Outlines automatically sets `additionalProperties: false` in JSON schemas for stricter parsing.
+- For multimodal inputs, use compatible models
+- `generate_batch` raises `NotImplementedError` due to API limitations.
+- For more examples (e.g., vision classification), see the integration tests at tests/models/test_mistral.py
+- **Author:** Steven E. Elliott aka see (seeyallc6c@gmail.com)
+
+---
+
+*Last updated: September 19, 2025*
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index d4b9cd842..c11525b82 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -13,6 +13,7 @@
 from .dottxt import Dottxt, from_dottxt
 from .gemini import Gemini, from_gemini
 from .llamacpp import LlamaCpp, from_llamacpp
+from .mistral import Mistral, from_mistral
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import AsyncOllama, Ollama, from_ollama
 from .openai import AsyncOpenAI, OpenAI, from_openai
@@ -34,6 +35,7 @@
     Gemini,
     Ollama,
     OpenAI,
+    Mistral,
     SGLang,
     TGI,
     VLLM,
@@ -48,6 +50,7 @@
 ]
 
 __all__ = [
+
     "Anthropic",
     "from_anthropic",
     "Model",
@@ -58,6 +61,8 @@
     "from_gemini",
     "LlamaCpp",
     "from_llamacpp",
+    "Mistral",
+    "from_mistral",
     "MLXLM",
     "from_mlxlm",
     "AsyncOllama",
diff --git a/outlines/models/mistral.py b/outlines/models/mistral.py
new file mode 100644
index 000000000..806aca800
--- /dev/null
+++ b/outlines/models/mistral.py
@@ -0,0 +1,631 @@
+"""
+outlines/models/mistral.py
+Integration with Mistral AI API.
+"""
+
+import json
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    AsyncGenerator,
+    AsyncIterator,
+    Coroutine,
+    Iterator,
+    List,
+    Dict,
+    Optional,
+    Union,
+)
+from functools import singledispatchmethod
+
+from pydantic import BaseModel, TypeAdapter
+
+from outlines.inputs import Chat, Image
+from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
+from outlines.models.utils import set_additional_properties_false_json_schema
+from outlines.types import JsonSchema, Regex, CFG
+from outlines.types.utils import (
+    is_dataclass,
+    is_typed_dict,
+    is_pydantic_model,
+    is_genson_schema_builder,
+    is_native_dict,
+    is_literal,
+)
+
+if TYPE_CHECKING:
+    from mistralai import Mistral as MistralClient
+
+__all__ = ["Mistral", "AsyncMistral", "from_mistral"]
+
+
+class MistralTypeAdapter(ModelTypeAdapter):
+    """Type adapter for the `Mistral` model.
+
+    Prepares arguments for Mistral's client `chat.complete`, `chat.complete_async`, or `chat.stream` methods.
+    Handles input (prompt or chat messages) and output type (JSON schema, Pydantic model, or enum-based choices).
+    """
+
+    @singledispatchmethod
+    def format_input(self, model_input):
+        """Generate the `messages` argument to pass to the client.
+
+        Parameters
+        ----------
+        model_input
+            The input provided by the user.
+
+        Returns
+        -------
+        list
+            The `messages` argument to pass to the client.
+        """
+        raise TypeError(
+            f"The input type {type(model_input)} is not available with "
+            "Mistral. The only available types are `str`, `list` and `Chat`."
+        )
+
+    @format_input.register(str)
+    def format_str_model_input(self, model_input: str) -> list:
+        """Format a string input into a list of messages.
+
+        Parameters
+        ----------
+        model_input : str
+            The input string prompt.
+
+        Returns
+        -------
+        list
+            A list of Mistral message objects.
+        """
+        from mistralai import UserMessage
+        return [UserMessage(content=model_input)]
+
+    @format_input.register(list)
+    def format_list_model_input(self, model_input: list) -> list:
+        """Format a list input into a list of messages.
+
+        Parameters
+        ----------
+        model_input : list
+            The input list, containing a string prompt and optionally Image objects.
+
+        Returns
+        -------
+        list
+            A list of Mistral message objects.
+        """
+        from mistralai import UserMessage
+        return [UserMessage(content=self._create_message_content(model_input))]
+
+    @format_input.register(Chat)
+    def format_chat_model_input(self, model_input: Chat) -> list:
+        """Format a Chat input into a list of messages.
+
+        Parameters
+        ----------
+        model_input : Chat
+            The Chat object containing a list of message dictionaries.
+
+        Returns
+        -------
+        list
+            A list of Mistral message objects.
+        """
+        from mistralai import UserMessage, AssistantMessage, SystemMessage
+        messages = []
+        for message in model_input.messages:
+            role = message["role"]
+            content = message["content"]
+            if role == "user":
+                messages.append(UserMessage(content=self._create_message_content(content)))
+            elif role == "assistant":
+                messages.append(AssistantMessage(content=content))
+            elif role == "system":
+                messages.append(SystemMessage(content=content))
+            else:
+                raise ValueError(f"Unsupported role: {role}")
+        return messages
+
+    def _create_message_content(self, content: Union[str, list]) -> Union[str, List[Dict[str, Union[str, Dict[str, str]]]]]:
+        """Create message content from an input.
+
+        Parameters
+        ----------
+        content : Union[str, list]
+            The content to format, either a string or a list containing a string and optionally Image objects.
+
+        Returns
+        -------
+        Union[str, List[Dict[str, Union[str, Dict[str, str]]]]]
+            The formatted content, either a string or a list of content parts (text and image URLs).
+
+        Raises
+        ------
+        ValueError
+            If the content type is invalid, the list is empty, or contains unsupported types.
+        """
+        if isinstance(content, str):
+            return content
+        elif isinstance(content, list):
+            if not content:
+                raise ValueError("Content list cannot be empty.")
+            if not isinstance(content[0], str):
+                raise ValueError("The first item in the list should be a string.")
+            if len(content) == 1:
+                return content[0]
+            content_parts: List[Dict[str, Union[str, Dict[str, str]]]] = [{"type": "text", "text": content[0]}]
+            for item in content[1:]:
+                if isinstance(item, Image):
+                    data_url = f"data:{item.image_format};base64,{item.image_str}"
+                    content_parts.append({
+                        "type": "image_url",
+                        "image_url": {"url": data_url}
+                    })
+                else:
+                    raise ValueError(
+                        f"Invalid item type in content list: {type(item)}. "
+                        "Expected Image objects after the first string."
+                    )
+            return content_parts
+        else:
+            raise ValueError(
+                f"Invalid content type: {type(content)}. "
+                "Content must be a string or a list starting with a string followed by optional Image objects."
+            )
+
+    def format_output_type(self, output_type: Optional[Any] = None) -> dict:
+        """Generate the `response_format` argument to pass to the client.
+
+        Parameters
+        ----------
+        output_type : Optional[Any]
+            The desired output type provided by the user.
+
+        Returns
+        -------
+        dict
+            The `response_format` dict to pass to the client.
+        """
+        if isinstance(output_type, Regex):
+            raise TypeError(
+                "Neither regex-based structured outputs nor the pattern keyword "
+                "in Json Schema are available with Mistral. Use an open source "
+                "model or dottxt instead."
+            )
+        elif isinstance(output_type, CFG):
+            raise TypeError(
+                "CFG-based structured outputs are not available with Mistral. "
+                "Use an open source model or dottxt instead."
+            )
+        elif is_literal(output_type):
+            raise TypeError(
+                "Literal types are not supported with Mistral. "
+                "Use a JSON schema with enum properties instead."
+            )
+        elif is_pydantic_model(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+        elif output_type is None:
+            return {}
+        elif is_native_dict(output_type):
+            return {"type": "json_object"}
+        elif is_dataclass(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+        elif is_typed_dict(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+        elif is_genson_schema_builder(output_type):
+            schema = json.loads(output_type.to_json())
+            return self.format_json_schema_type(schema, "Schema")
+        elif isinstance(output_type, JsonSchema):
+            schema = json.loads(output_type.schema)
+            return self.format_json_schema_type(schema, "Schema")
+        else:
+            type_name = getattr(output_type, "__name__", None) or str(output_type)
+            raise TypeError(
+                f"The type {type_name} is not available with Mistral. "
+                "Use an open source model or dottxt instead."
+            )
+
+    def format_json_schema_type(self, schema: dict, schema_name: str = "Schema") -> dict:
+        """Format a JSON schema for the client.
+
+        Parameters
+        ----------
+        schema : dict
+            The JSON schema to format.
+        schema_name : str
+            The name of the schema (default: "Schema").
+
+        Returns
+        -------
+        dict
+            The formatted JSON schema configuration.
+        """
+        schema = set_additional_properties_false_json_schema(schema)
+        return {
+            "type": "json_schema",
+            "json_schema": {
+                "schema": schema,
+                "name": schema_name.lower(),
+                "strict": True
+            }
+        }
+
+    def format_json_mode_type(self) -> dict:
+        """Generate configuration for JSON mode output.
+
+        Returns
+        -------
+        dict
+            The JSON mode configuration.
+        """
+        return {"type": "json_object"}
+
+
+class Mistral(Model):
+    """Thin wrapper around the `mistralai.Mistral` client.
+
+    Converts input and output types to arguments for the `mistralai.Mistral` client's `chat.complete` or `chat.stream` methods.
+    """
+
+    def __init__(self, client: "MistralClient", model_name: Optional[str] = None):
+        """
+        Parameters
+        ----------
+        client : MistralClient
+            A mistralai.Mistral client instance.
+        model_name : Optional[str]
+            The name of the model to use.
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = MistralTypeAdapter()
+
+    def generate(
+        self,
+        model_input: Union[Chat, list, str],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate a response from the model.
+
+        Parameters
+        ----------
+        model_input : Union[Chat, list, str]
+            The prompt or chat messages to generate a response from.
+        output_type : Optional[Any]
+            The desired format of the response (e.g., JSON schema).
+        **inference_kwargs : Any
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The response generated by the model as text.
+        """
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        try:
+            result = self.client.chat.complete(
+                messages=messages,
+                response_format=response_format,
+                **inference_kwargs,
+            )
+        except Exception as e:
+            if "schema" in str(e).lower() or "json_schema" in str(e).lower():
+                raise TypeError(
+                    f"Mistral does not support your schema: {e}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise RuntimeError(f"Error calling Mistral API: {e}") from e
+
+        messages = [choice.message for choice in result.choices]
+        return messages[0].content if len(messages) == 1 else [m.content for m in messages]
+
+    def generate_batch(
+        self,
+        model_input,
+        output_type=None,
+        **inference_kwargs,
+    ):
+        """Generate responses for a batch of inputs.
+
+        Parameters
+        ----------
+        model_input
+            The batch of inputs to generate responses from.
+        output_type
+            The desired format of the responses.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Raises
+        ------
+        NotImplementedError
+            Batch inference is not supported by the Mistral API.
+        """
+        raise NotImplementedError(
+            "The mistralai library does not support batch inference."
+        )
+
+    def generate_stream(
+        self,
+        model_input: Union[Chat, list, str],
+        output_type: Optional[Any] = None,
+        **inference_kwargs,
+    ) -> Iterator[str]:
+        """Generate a stream of responses from the model.
+
+        Parameters
+        ----------
+        model_input : Union[Chat, list, str]
+            The prompt or chat messages to generate a response from.
+        output_type : Optional[Any]
+            The desired format of the response (e.g., JSON schema).
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Iterator[str]
+            An iterator that yields the text chunks generated by the model.
+        """
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        try:
+            stream = self.client.chat.stream(
+                messages=messages,
+                response_format=response_format,
+                **inference_kwargs
+            )
+        except Exception as e:
+            if "schema" in str(e).lower() or "json_schema" in str(e).lower():
+                raise TypeError(
+                    f"Mistral does not support your schema: {e}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise RuntimeError(f"Error calling Mistral API: {e}") from e
+
+        for chunk in stream:
+            if (
+                hasattr(chunk, "data")
+                and chunk.data.choices
+                and chunk.data.choices[0].delta.content is not None
+            ):
+                yield chunk.data.choices[0].delta.content
+
+    def supports_structured_output(self, model_name: Optional[str] = None) -> bool:
+        """Check if the model supports structured output.
+
+        Parameters
+        ----------
+        model_name : Optional[str]
+            The name of the model to check (defaults to instance's model_name).
+
+        Returns
+        -------
+        bool
+            True if the model supports structured output, False otherwise.
+        """
+        model = model_name or self.model_name
+        if model is None:
+            return False
+
+        # Codestral Mamba does NOT support structured output
+        if "codestral-mamba" in model.lower():
+            return False
+
+        # All other models support structured output
+        return True
+
+
+class AsyncMistral(AsyncModel):
+    """Async thin wrapper around the `mistralai.Mistral` client.
+
+    Converts input and output types to arguments for the `mistralai.Mistral` client's async methods (`chat.complete_async` or `chat.stream`).
+    """
+
+    def __init__(self, client: "MistralClient", model_name: Optional[str] = None):
+        """
+        Parameters
+        ----------
+        client : MistralClient
+            A mistralai.Mistral client instance.
+        model_name : Optional[str]
+            The name of the model to use.
+        """
+        self.client = client
+        self.model_name = model_name
+        self.type_adapter = MistralTypeAdapter()
+
+    async def generate(
+        self,
+        model_input: Union[Chat, list, str],
+        output_type: Optional[Any] = None,
+        **inference_kwargs: Any,
+    ) -> Union[str, list[str]]:
+        """Generate a response from the model asynchronously.
+
+        Parameters
+        ----------
+        model_input : Union[Chat, list, str]
+            The prompt or chat messages to generate a response from.
+        output_type : Optional[Any]
+            The desired format of the response (e.g., JSON schema).
+        **inference_kwargs : Any
+            Additional keyword arguments to pass to the client.
+
+        Returns
+        -------
+        Union[str, list[str]]
+            The response generated by the model as text.
+        """
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        try:
+            result = await self.client.chat.complete_async(
+                messages=messages,
+                response_format=response_format,
+                stream=False,
+                **inference_kwargs,
+            )
+        except Exception as e:
+            if "schema" in str(e).lower() or "json_schema" in str(e).lower():
+                raise TypeError(
+                    f"Mistral does not support your schema: {e}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise RuntimeError(f"Mistral API error: {e}") from e
+
+        messages = [choice.message for choice in result.choices]
+        return messages[0].content if len(messages) == 1 else [m.content for m in messages]
+
+    async def generate_batch(
+        self,
+        model_input,
+        output_type=None,
+        **inference_kwargs,
+    ):
+        """Generate responses for a batch of inputs asynchronously.
+
+        Parameters
+        ----------
+        model_input
+            The batch of inputs to generate responses from.
+        output_type
+            The desired format of the responses.
+        **inference_kwargs
+            Additional keyword arguments to pass to the client.
+
+        Raises
+        ------
+        NotImplementedError
+            Batch inference is not supported by the Mistral API.
+        """
+        raise NotImplementedError(
+            "The mistralai library does not support batch inference."
+        )
+
+    async def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+        """
+        Generate text from the model as an async stream of chunks.
+
+        Uses `stream_async` if streaming, otherwise `complete_async`.
+
+        Args:
+            model_input: str, list, or chat input to generate from.
+            output_type: Optional type for structured output.
+            **inference_kwargs: Extra kwargs like "model" name.
+
+        Yields:
+            str: chunks of text as they are streamed.
+        """
+        messages = self.type_adapter.format_input(model_input)
+        response_format = self.type_adapter.format_output_type(output_type)
+
+        model_name = inference_kwargs.get("model", self.model_name)
+        stream = inference_kwargs.get("stream", True)
+
+        try:
+            if stream:
+                stream_kwargs = {
+                    "model": model_name,
+                    "messages": messages,
+                    **inference_kwargs
+                }
+
+                if response_format:
+                    stream_kwargs["response_format"] = response_format
+
+                response = await self.client.chat.stream_async(**stream_kwargs)
+
+                async for chunk in response:
+                    if (
+                        hasattr(chunk, "data")
+                        and chunk.data.choices
+                        and len(chunk.data.choices) > 0
+                        and hasattr(chunk.data.choices[0], "delta")
+                        and chunk.data.choices[0].delta.content is not None
+                    ):
+                        yield chunk.data.choices[0].delta.content
+            else:
+                complete_kwargs = {
+                    "model": model_name,
+                    "messages": messages,
+                    **inference_kwargs
+                }
+
+                if response_format:
+                    complete_kwargs["response_format"] = response_format
+
+                res = await self.client.chat.complete_async(**complete_kwargs)
+                content = res.choices[0].message.content
+                yield content
+
+        except Exception as e:
+            raise RuntimeError(f"Mistral API error: {e}") from e
+
+    def supports_structured_output(self, model_name: Optional[str] = None) -> bool:
+        """Check if the model supports structured output.
+
+        Parameters
+        ----------
+        model_name : Optional[str]
+            The name of the model to check (defaults to instance's model_name).
+
+        Returns
+        -------
+        bool
+            True if the model supports structured output, False otherwise.
+        """
+        model = model_name or self.model_name
+        if model is None:
+            return True
+        return "codestral-mamba" not in model.lower()
+
+
+def from_mistral(
+    client: "MistralClient",
+    model_name: Optional[str] = None,
+    async_client: bool = False,
+) -> Union[Mistral, AsyncMistral]:
+    """Create an Outlines Mistral model instance from a mistralai.Mistral client.
+
+    Parameters
+    ----------
+    client : MistralClient
+        A mistralai.Mistral client instance.
+    model_name : Optional[str]
+        The name of the model to use.
+    async_client : bool
+        If True, return an AsyncMistral instance; otherwise, return a Mistral instance.
+
+    Returns
+    -------
+    Union[Mistral, AsyncMistral]
+        An Outlines Mistral or AsyncMistral model instance.
+    """
+    from mistralai import Mistral as MistralClient
+
+    if not isinstance(client, MistralClient):
+        raise ValueError(
+            "Invalid client type. The client must be an instance of mistralai.Mistral."
+        )
+
+    return AsyncMistral(client, model_name) if async_client else Mistral(client, model_name)
diff --git a/pyproject.toml b/pyproject.toml
index c08b56f5d..027d0a074 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -45,6 +45,7 @@ llamacpp = ["huggingface-hub", "llama-cpp-python", "numba"]
 mlxlm = ["datasets", "mlx", "mlx-lm"]
 ollama = ["ollama"]
 openai = ["openai"]
+mistral = ["mistralai"]
 sglang = ["openai"]
 tgi = ["huggingface_hub"]
 transformers = ["accelerate", "datasets", "transformers", "setuptools"]
@@ -72,6 +73,7 @@ test = [
     "datasets",
     "anthropic",
     "google-genai",
+    "mistralai",
     "transformers",
     "pillow",
     "jax",
@@ -137,6 +139,7 @@ module = [
     "jsonschema.*",
     "anthropic.*",
     "google.*",
+    "mistralai.*",
     "mamba_ssm.*",
     "mlx_lm.*",
     "mlx.*",
@@ -183,6 +186,7 @@ omit = [
     "outlines/models/gemini.py",
     "outlines/models/mlxlm.py",
     "outlines/models/openai.py",
+    "outlines/models/mistral.py",
     "outlines/models/vllm_offline.py",
     "outlines/processors/tensor_adapters/mlx.py",
     "tests/*",
diff --git a/tests/models/test_mistral.py b/tests/models/test_mistral.py
new file mode 100644
index 000000000..8653811f9
--- /dev/null
+++ b/tests/models/test_mistral.py
@@ -0,0 +1,723 @@
+"""
+outlines/tests/models/test_mistral.py
+Integration tests for outlines mistral api
+"""
+
+
+import json
+import base64
+import os
+from typing import Dict, List, Literal, Iterator, AsyncIterator
+import requests
+import io
+from PIL import Image as PILImage
+import pytest
+from pydantic import BaseModel, Field
+
+from outlines.inputs import Chat, Image
+from outlines.models.mistral import from_mistral
+from outlines.types import JsonSchema
+
+# List of models to try in sequence
+MODEL_LIST = [
+    "mistral-small-2506",
+    "magistral-small-2506",
+    "mistral-large-latest",
+    "magistral-small-2509",
+    "magistral-small-2507",
+    "voxtral-small-2507",
+    "voxtral-mini-2507",
+    "devstral-small-2507",
+    "mistral-small-2503",
+    "mistral-small-2501",
+    "devstral-small-2505",
+    "pixtral-12b-2409",
+    "open-mistral-nemo",
+]
+
+# Vision-capable models
+VISION_MODELS = [
+    "magistral-small-2509",
+    "pixtral-12b-2409",
+    "mistral-small-2503",
+]
+
+class AnimalClassification(BaseModel):
+    image_subject: Literal["dog", "not an animal", "cat", "other"]
+    specific_kind: str
+
+def try_models(mistral_client, model_list, async_client=False, **kwargs):
+    """Helper function to try generating with each model until one succeeds."""
+    last_exception = None
+    for model_name in model_list:
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=async_client)
+            generate_func = kwargs.pop("generate_func")
+            # Extract model_input (prompt or prompts) from kwargs
+            model_input = kwargs.pop("prompt", kwargs.pop("prompts", None))
+            if model_input is None:
+                raise ValueError("No 'prompt' or 'prompts' provided in kwargs")
+            # Call the generate function with model_input as positional arg
+            return model, generate_func(model, model_input=model_input, **kwargs)
+        except Exception as e:
+            if "429" in str(e):  # Handle rate limit errors
+                last_exception = e
+                continue
+            raise  # Re-raise non-429 errors
+    raise last_exception or Exception("No models succeeded")
+
+async def try_models_async(mistral_client, model_list, **kwargs):
+    """Helper function for async tests to try generating with each model."""
+    last_exception = None
+    for model_name in model_list:
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
+            generate_func = kwargs.pop("generate_func")
+            # Extract model_input (prompt or prompts) from kwargs
+            model_input = kwargs.pop("prompt", kwargs.pop("prompts", None))
+            if model_input is None:
+                raise ValueError("No 'prompt' or 'prompts' provided in kwargs")
+            # Call the async generate function with model_input as positional arg
+            return model, await generate_func(model, model_input=model_input, **kwargs)
+        except Exception as e:
+            if "429" in str(e):  # Handle rate limit errors
+                last_exception = e
+                continue
+            raise
+    raise last_exception or Exception("No models succeeded")
+
+@pytest.mark.integration
+class TestMistralIntegration:
+    @pytest.fixture
+    def api_key(self):
+        api_key = os.getenv("MISTRAL_API_KEY")
+        if not api_key:
+            pytest.skip("MISTRAL_API_KEY environment variable not set")
+        return api_key
+
+    @pytest.fixture
+    def mistral_client(self, api_key):
+        try:
+            from mistralai import Mistral as MistralClient
+            return MistralClient(api_key=api_key)
+        except ImportError:
+            pytest.skip("mistralai package not installed")
+
+    def test_simple_generation(self, mistral_client):
+        def generate_func(model, model_input, **kwargs):
+            return model.generate(model_input, **kwargs)
+
+        model, result = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="What is 2+2? Answer in one sentence."
+        )
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert "4" in result
+
+    def test_structured_generation_json(self, mistral_client):
+        class Person(BaseModel):
+            name: str
+            age: int
+
+        def generate_func(model, model_input, **kwargs):
+            return model.generate(model_input, **kwargs)
+
+        model, result = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="""Generate a JSON object representing a person with name and age.
+            Return only the JSON, no other text.""",
+            output_type=Person
+        )
+        assert isinstance(result, str)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)
+
+    def test_streaming_generation(self, mistral_client):
+        def generate_func(model, model_input, **kwargs):
+            return list(model.generate_stream(model_input, **kwargs))
+
+        model, chunks = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="Write a short story about a robot."
+        )
+        assert len(chunks) > 0
+        assert all(isinstance(chunk, str) for chunk in chunks)
+        full_text = "".join(chunks)
+        assert len(full_text) > 0
+
+    def test_mistral_with_outlines(self, mistral_client):
+        class Person(BaseModel):
+            name: str
+            age: int
+
+        def generate_func(model, model_input, **kwargs):
+            return model.generate(model_input, **kwargs)
+
+        model, result = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="""Generate a JSON object representing a person with name and age.
+            Return only the JSON, no other text.""",
+            output_type=Person
+        )
+        assert isinstance(result, str)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert len(parsed["name"]) > 0
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)
+
+    def test_mistral_with_outlines_real(self, mistral_client):
+        class Address(BaseModel):
+            street: str
+            city: str
+            zip_code: str
+
+        class Employee(BaseModel):
+            name: str
+            role: str
+            department: str
+            address: Address
+
+        class Company(BaseModel):
+            company_name: str
+            employees: list[Employee]
+
+        def generate_func(model, model_input, **kwargs):
+            return model.generate(model_input, **kwargs)
+
+        model, result = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt=Chat([
+                {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
+                {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
+            ]),
+            output_type=Company
+        )
+        assert isinstance(result, str)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "company_name" in parsed
+        assert "employees" in parsed
+        assert isinstance(parsed["employees"], list)
+        assert len(parsed["employees"]) >= 2
+        for employee in parsed["employees"]:
+            assert employee["department"] == "IT"
+            assert isinstance(employee["name"], str)
+            assert len(employee["name"]) > 0
+            assert isinstance(employee["role"], str)
+            assert len(employee["role"]) > 0
+            assert isinstance(employee["address"], dict)
+            assert isinstance(employee["address"]["city"], str)
+            assert len(employee["address"]["city"]) > 0
+            assert isinstance(employee["address"]["street"], str)
+            assert len(employee["address"]["street"]) > 0
+            assert isinstance(employee["address"]["zip_code"], str)
+            assert len(employee["address"]["zip_code"]) > 0
+
+    def test_real_image_classification(self, mistral_client):
+        image_url = "https://upload.wikimedia.org/wikipedia/commons/7/71/Charmonty_Norfolkterrier.jpg"
+        headers = {
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
+                          'AppleWebKit/537.36 (KHTML, like Gecko) '
+                          'Chrome/91.0.4472.124 Safari/537.36'
+        }
+        response = requests.get(image_url, headers=headers, timeout=30)
+        response.raise_for_status()
+        content_type = response.headers.get('content-type', '').lower()
+        if not any(img_type in content_type for img_type in ['image/', 'jpeg', 'jpg', 'png']):
+            raise ValueError(f"Expected image content, got: {content_type}")
+        pil_image = PILImage.open(io.BytesIO(response.content))
+        image = Image(pil_image)
+
+        last_exception = None
+        for model_name in VISION_MODELS:
+            try:
+                model = from_mistral(mistral_client, model_name)
+                prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
+                result = model.generate(prompt, output_type=AnimalClassification)
+                parsed = json.loads(result)
+                assert parsed["image_subject"] == "dog"
+                assert isinstance(parsed["specific_kind"], str)
+                assert len(parsed["specific_kind"]) > 0
+                return  # Success, exit the function
+            except Exception as e:
+                last_exception = e
+                continue
+
+        # If all models failed, raise the last exception
+        raise last_exception or Exception("No vision models succeeded")
+
+    def test_image_classification_simple(self, mistral_client):
+        pil_image = PILImage.new('RGB', (100, 100), color='red')
+        buffer = io.BytesIO()
+        pil_image.save(buffer, format="PNG")
+        buffer.seek(0)
+        pil_image = PILImage.open(buffer)
+        image = Image(pil_image)
+
+        last_exception = None
+        for model_name in VISION_MODELS:
+            try:
+                model = from_mistral(mistral_client, model_name)
+                prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
+                result = model.generate(prompt, output_type=AnimalClassification)
+                parsed = json.loads(result)
+                assert parsed["image_subject"] == "not an animal"
+                assert parsed["specific_kind"] == "hotdog"
+                return  # Success, exit the function
+            except Exception as e:
+                last_exception = e
+                continue
+
+        # If all models failed, raise the last exception
+        raise last_exception or Exception("No vision models succeeded")
+
+
+    def test_sync_streaming_structured_generation(self, mistral_client):
+        class Person(BaseModel):
+            name: str
+            age: int
+
+        def generate_func(model, model_input, **kwargs):
+            return list(model.generate_stream(model_input, **kwargs))
+
+        model, chunks = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="""Generate a JSON object representing a person with name and age.
+            Return only the JSON, no other text.""",
+            output_type=Person,
+            async_client=False
+        )
+        result = "".join(chunks)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)
+
+@pytest.mark.integration
+class TestAsyncMistralIntegration:
+    @pytest.fixture
+    def api_key(self):
+        api_key = os.getenv("MISTRAL_API_KEY")
+        if not api_key:
+            pytest.skip("MISTRAL_API_KEY environment variable not set")
+        return api_key
+
+    @pytest.fixture
+    def mistral_client(self, api_key):
+        try:
+            from mistralai import Mistral as MistralClient
+            return MistralClient(api_key=api_key)
+        except ImportError:
+            pytest.skip("mistralai package not installed")
+
+    @pytest.mark.asyncio
+    async def test_async_simple_generation(self, mistral_client):
+        async def generate_func(model, model_input, **kwargs):
+            return await model.generate(model_input, **kwargs)
+
+        model, result = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="What is 2+2? Answer in one sentence."
+        )
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert "4" in result
+
+    @pytest.mark.asyncio
+    async def test_async_structured_generation(self, mistral_client):
+        class Person(BaseModel):
+            name: str
+            age: int
+
+        async def generate_func(model, model_input, **kwargs):
+            return await model.generate(model_input, **kwargs)
+
+        model, result = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="""Generate a JSON object representing a person with name and age.
+            Return only the JSON, no other text.""",
+            output_type=Person
+        )
+        assert isinstance(result, str)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)
+
+    @pytest.mark.asyncio
+    async def test_async_streaming_generation(self, mistral_client):
+        async def generate_func(model, model_input, **kwargs):
+            chunks = []
+            async for chunk in model.generate_stream(model_input, **kwargs):
+                chunks.append(chunk)
+            return chunks
+
+        model, chunks = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="Write a short story about a robot."
+        )
+        assert len(chunks) > 0
+        assert all(isinstance(chunk, str) for chunk in chunks)
+
+    @pytest.mark.asyncio
+    async def test_async_vision_classification(self, mistral_client):
+        image_url = "https://upload.wikimedia.org/wikipedia/commons/7/71/Charmonty_Norfolkterrier.jpg"
+        headers = {
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
+                          'AppleWebKit/537.36 (KHTML, like Gecko) '
+                          'Chrome/91.0.4472.124 Safari/537.36'
+        }
+        response = requests.get(image_url, headers=headers, timeout=30)
+        response.raise_for_status()
+        content_type = response.headers.get('content-type', '').lower()
+        if not any(img_type in content_type for img_type in ['image/', 'jpeg', 'jpg', 'png']):
+            raise ValueError(f"Expected image content, got: {content_type}")
+        pil_image = PILImage.open(io.BytesIO(response.content))
+        image = Image(pil_image)
+
+        last_exception = None
+        for model_name in VISION_MODELS:
+            try:
+                model = from_mistral(mistral_client, model_name, async_client=True)
+                prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
+                result = await model.generate(prompt, output_type=AnimalClassification)
+                parsed = json.loads(result)
+                assert parsed["image_subject"] == "dog"
+                assert isinstance(parsed["specific_kind"], str)
+                assert len(parsed["specific_kind"]) > 0
+                return  # Success, exit the function
+            except Exception as e:
+                last_exception = e
+                continue
+
+        # If all models failed, raise the last exception
+        raise last_exception or Exception("No vision models succeeded")
+
+
+    @pytest.mark.asyncio
+    async def test_async_structured_schema_error(self, mistral_client):
+        class InvalidModel(BaseModel):
+            value: str = Field(pattern=r"^\d+$")
+
+        # Just use the first model; we expect the error for all models
+        model = from_mistral(mistral_client, MODEL_LIST[0], async_client=True)
+        with pytest.raises(TypeError, match="Mistral does not support your schema"):
+            await model.generate("Generate a number string", output_type=InvalidModel)
+
+
+    @pytest.mark.asyncio
+    async def test_async_multiple_samples(self, mistral_client):
+        async def generate_func(model, model_input, **kwargs):
+            return await model.generate(model_input, **kwargs)
+
+        model, result = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="Respond with one word.",
+            n=2
+        )
+        assert isinstance(result, list)
+        assert len(result) == 2
+        assert all(isinstance(r, str) for r in result)
+
+    @pytest.mark.asyncio
+    async def test_async_nested_structured(self, mistral_client):
+        class Address(BaseModel):
+            street: str
+            city: str
+            zip_code: str
+
+        class Employee(BaseModel):
+            name: str
+            role: str
+            department: str
+            address: Address
+
+        class Company(BaseModel):
+            company_name: str
+            employees: List[Employee]
+
+        async def generate_func(model, model_input, **kwargs):
+            return await model.generate(model_input, **kwargs)
+
+        model, result = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt=Chat([
+                {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
+                {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
+            ]),
+            output_type=Company
+        )
+        assert isinstance(result, str)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "company_name" in parsed
+        assert "employees" in parsed
+        assert isinstance(parsed["employees"], list)
+        assert len(parsed["employees"]) >= 2
+        for employee in parsed["employees"]:
+            assert employee["department"] == "IT"
+            assert isinstance(employee["name"], str)
+            assert len(employee["name"]) > 0
+            assert isinstance(employee["role"], str)
+            assert len(employee["role"]) > 0
+            assert isinstance(employee["address"], dict)
+            assert isinstance(employee["address"]["city"], str)
+            assert len(employee["address"]["city"]) > 0
+            assert isinstance(employee["address"]["street"], str)
+            assert len(employee["address"]["street"]) > 0
+            assert isinstance(employee["address"]["zip_code"], str)
+            assert len(employee["address"]["zip_code"]) > 0
+
+    @pytest.mark.asyncio
+    async def test_async_streaming_structured_generation(self, mistral_client):
+        class Person(BaseModel):
+            name: str
+            age: int
+
+        async def generate_func(model, model_input, **kwargs):
+            chunks = []
+            async for chunk in model.generate_stream(model_input, **kwargs):
+                chunks.append(chunk)
+            return chunks
+
+        model, chunks = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="""Generate a JSON object representing a person with name and age.
+            Return only the JSON, no other text.""",
+            output_type=Person
+        )
+        result = "".join(chunks)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)
+
+@pytest.mark.integration
+class TestMistralAdditionalFeatures:
+    @pytest.fixture
+    def api_key(self):
+        api_key = os.getenv("MISTRAL_API_KEY")
+        if not api_key:
+            pytest.skip("MISTRAL_API_KEY environment variable not set")
+        return api_key
+
+    @pytest.fixture
+    def mistral_client(self, api_key):
+        try:
+            from mistralai import Mistral as MistralClient
+            return MistralClient(api_key=api_key)
+        except ImportError:
+            pytest.skip("mistralai package not installed")
+
+    def test_custom_inference_parameters(self, mistral_client):
+        """Test custom inference parameters like temperature, max_tokens, top_p"""
+        def generate_func(model, model_input, **kwargs):
+            return model.generate(model_input, **kwargs)
+
+        model, result = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="Write a very short creative story about a robot.",
+            temperature=0.9,
+            max_tokens=50,
+            top_p=0.95
+        )
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert len(result.split()) <= 100
+
+    def test_text_only_streaming(self, mistral_client):
+        """Test streaming without structured output"""
+        def generate_func(model, model_input, **kwargs):
+            return list(model.generate_stream(model_input, **kwargs))
+
+        model, chunks = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="Count from 1 to 5 with spaces: "
+        )
+        assert len(chunks) > 0
+        assert all(isinstance(chunk, str) for chunk in chunks)
+        full_text = "".join(chunks)
+        assert len(full_text) > 0
+        assert any(str(i) in full_text for i in range(1, 6))
+
+    def test_multi_role_chat(self, mistral_client):
+        """Test Chat with system, user, and assistant roles"""
+        from outlines.inputs import Chat
+
+        def generate_func(model, model_input, **kwargs):
+            return model.generate(model_input, **kwargs)
+
+        model, result = try_models(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt=Chat([
+                {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
+                {"role": "user", "content": "What is 5 + 3?"},
+                {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
+                {"role": "user", "content": "What about 10 - 4?"}
+            ])
+        )
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert "6" in result
+        assert "Happy learning!" in result
+
+    def test_model_compatibility_check(self, mistral_client):
+        """Test supports_structured_output method"""
+        # Just use the first model; no prompt needed
+        model = from_mistral(mistral_client, MODEL_LIST[0])
+        supports_structured = model.supports_structured_output()
+        assert isinstance(supports_structured, bool)
+
+    @pytest.mark.asyncio
+    async def test_async_custom_inference_parameters(self, mistral_client):
+        """Test async custom inference parameters"""
+        async def generate_func(model, model_input, **kwargs):
+            return await model.generate(model_input, **kwargs)
+
+        model, result = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="Write a very short creative story about a robot.",
+            temperature=0.9,
+            max_tokens=50,
+            top_p=0.95
+        )
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert len(result.split()) <= 100
+
+    @pytest.mark.asyncio
+    async def test_async_text_only_streaming(self, mistral_client):
+        """Test async streaming without structured output"""
+        async def generate_func(model, model_input, **kwargs):
+            chunks = []
+            async for chunk in model.generate_stream(model_input, **kwargs):
+                chunks.append(chunk)
+            return chunks
+
+        model, chunks = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt="Count from 1 to 5 with spaces: "
+        )
+        assert len(chunks) > 0
+        assert all(isinstance(chunk, str) for chunk in chunks)
+        full_text = "".join(chunks)
+        assert len(full_text) > 0
+        assert any(str(i) in full_text for i in range(1, 6))
+
+    @pytest.mark.asyncio
+    async def test_async_multi_role_chat(self, mistral_client):
+        """Test async Chat with multiple roles"""
+        from outlines.inputs import Chat
+
+        async def generate_func(model, model_input, **kwargs):
+            return await model.generate(model_input, **kwargs)
+
+        model, result = await try_models_async(
+            mistral_client,
+            MODEL_LIST,
+            generate_func=generate_func,
+            prompt=Chat([
+                {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
+                {"role": "user", "content": "What is 5 + 3?"},
+                {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
+                {"role": "user", "content": "What about 10 - 4?"}
+            ])
+        )
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert "6" in result
+        assert "Happy learning!" in result
+
+    def test_empty_input_error(self, mistral_client):
+        """Test behavior with empty inputs"""
+        def generate_func(model, model_input, **kwargs):
+            return model.generate(model_input, **kwargs)
+
+        model, _ = try_models(mistral_client, MODEL_LIST, generate_func=generate_func, prompt="")
+        try:
+            result = model.generate("")
+            print(f"Empty string result: '{result}'")
+            print(f"Result type: {type(result)}")
+            print(f"Result length: {len(result)}")
+            assert isinstance(result, str)
+        except Exception as e:
+            print(f"Empty string exception raised: {type(e).__name__}: {e}")
+            assert False, f"Empty string unexpectedly failed with result: '{e}'"
+
+        with pytest.raises(ValueError):
+            model.generate([])
+
+    def test_mixed_content_list_validation(self, mistral_client):
+        from outlines.inputs import Image
+        from PIL import Image as PILImage
+        import io
+
+        pil_image = PILImage.new('RGB', (100, 100), color='blue')
+        buffer = io.BytesIO()
+        pil_image.save(buffer, format="PNG")
+        buffer.seek(0)
+        pil_image = PILImage.open(buffer)
+        image = Image(pil_image)
+
+        last_exception = None
+        for model_name in VISION_MODELS:
+            try:
+                model = from_mistral(mistral_client, model_name)
+                result = model.generate(["Describe this color", image])
+                assert isinstance(result, str)
+                assert len(result) > 0
+                return  # Success, exit the function
+            except Exception as e:
+                last_exception = e
+                continue
+
+        # If all models failed, raise the last exception
+        raise last_exception or Exception("No vision models succeeded")
diff --git a/tests/models/test_mistral_type_adapter.py b/tests/models/test_mistral_type_adapter.py
new file mode 100644
index 000000000..5bbc76285
--- /dev/null
+++ b/tests/models/test_mistral_type_adapter.py
@@ -0,0 +1,553 @@
+
+"""
+outlines/tests/models/test_mistral_type_adapter.py
+Tests for MistralTypeAdapter class.
+"""
+
+import os
+import json
+import pytest
+import sys
+from dataclasses import dataclass
+from typing import Literal
+from unittest.mock import patch
+
+from genson import SchemaBuilder
+from pydantic import BaseModel, Field
+import io
+from PIL import Image as PILImage
+
+from outlines import cfg, json_schema, regex
+from outlines.inputs import Chat, Image
+from outlines.models.mistral import from_mistral
+from outlines.models.mistral import MistralTypeAdapter
+from outlines.types import JsonSchema
+
+MODEL_NAME = "mistral-small-2506"
+
+
+if sys.version_info >= (3, 12):
+    from typing import TypedDict
+else:
+    from typing_extensions import TypedDict
+
+
+@pytest.fixture
+def schema():
+    return {
+        "properties": {
+            "user_id": {"title": "User Id", "type": "integer"},
+            "name": {"title": "Name", "type": "string"},
+        },
+        "required": ["user_id", "name"],
+        "title": "User",
+        "type": "object",
+        "additionalProperties": False,
+    }
+
+
+@pytest.fixture
+def adapter():
+    return MistralTypeAdapter()
+
+
+def test_mistral_type_adapter_init(adapter):
+    """Test MistralTypeAdapter initialization."""
+    assert isinstance(adapter, MistralTypeAdapter)
+
+
+@patch('mistralai.UserMessage')
+def test_mistral_type_adapter_input_text(mock_user_msg, adapter):
+    """Test formatting string input."""
+    message = "Hello world"
+    result = adapter.format_input(message)
+    assert result is not None
+    mock_user_msg.assert_called_once_with(content=message)
+
+
+@patch('mistralai.UserMessage')
+def test_mistral_type_adapter_input_list(mock_user_msg, adapter):
+    """Test formatting list input."""
+    message_list = ["Hello world"]
+    result = adapter.format_input(message_list)
+    assert result is not None
+    mock_user_msg.assert_called_once_with(content="Hello world")
+
+
+@patch('mistralai.UserMessage')
+@patch('mistralai.AssistantMessage')
+@patch('mistralai.SystemMessage')
+def test_mistral_type_adapter_input_chat(mock_system_msg, mock_assistant_msg, mock_user_msg, adapter):
+    """Test formatting Chat input with system message."""
+    chat = Chat([
+        {"role": "system", "content": "You are helpful"},
+        {"role": "user", "content": "Hello"},
+        {"role": "assistant", "content": "Hi there"},
+        {"role": "user", "content": "How are you?"}
+    ])
+    result = adapter.format_input(chat)
+    assert result is not None
+    mock_system_msg.assert_called_once_with(content="You are helpful")
+    assert mock_user_msg.call_count == 2
+    mock_assistant_msg.assert_called_once_with(content="Hi there")
+
+
+def test_mistral_type_adapter_input_invalid(adapter):
+    """Test formatting invalid input types."""
+    @dataclass
+    class Audio:
+        file: str
+
+    with pytest.raises(TypeError, match="The input type .* is not available"):
+        adapter.format_input(Audio("file"))
+
+    with pytest.raises(TypeError, match="The input type .* is not available"):
+        adapter.format_input(123)
+
+
+def test_mistral_type_adapter_input_list_invalid_content(adapter):
+    """Test formatting list input with invalid content."""
+    with pytest.raises(ValueError, match="The first item in the list should be a string"):
+        adapter.format_input([123])
+
+    with pytest.raises(ValueError, match="Content list cannot be empty."):
+        adapter.format_input([])
+
+
+def test_mistral_type_adapter_input_chat_invalid_role(adapter):
+    """Test formatting Chat input with invalid role."""
+    chat = Chat([{"role": "invalid", "content": "Hello"}])
+    with pytest.raises(ValueError, match="Unsupported role: invalid"):
+        adapter.format_input(chat)
+
+
+def test_mistral_type_adapter_output_none(adapter):
+    """Test formatting None output type."""
+    result = adapter.format_output_type(None)
+    assert result == {}
+
+
+def test_mistral_type_adapter_output_dict(adapter):
+    """Test formatting dict output type."""
+    result = adapter.format_output_type(dict)
+    assert result == {"type": "json_object"}
+
+
+def test_mistral_type_adapter_output_pydantic(adapter):
+    """Test formatting Pydantic model output type."""
+    class TestModel(BaseModel):
+        name: str
+        age: int
+
+    result = adapter.format_output_type(TestModel)
+    assert isinstance(result, dict)
+    assert result["type"] == "json_schema"
+    assert result["json_schema"]["name"] == "testmodel"
+    assert result["json_schema"]["strict"] is True
+    assert "properties" in result["json_schema"]["schema"]
+    assert "name" in result["json_schema"]["schema"]["properties"]
+    assert "age" in result["json_schema"]["schema"]["properties"]
+
+
+def test_mistral_type_adapter_output_dataclass(adapter, schema):
+    """Test formatting dataclass output type."""
+    @dataclass
+    class User:
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["type"] == "json_schema"
+    assert result["json_schema"]["strict"] is True
+    assert result["json_schema"]["name"] == "user"
+    assert result["json_schema"]["schema"] == schema
+
+
+def test_mistral_type_adapter_output_typed_dict(adapter, schema):
+    """Test formatting TypedDict output type."""
+    class User(TypedDict):
+        user_id: int
+        name: str
+
+    result = adapter.format_output_type(User)
+    assert isinstance(result, dict)
+    assert result["type"] == "json_schema"
+    assert result["json_schema"]["strict"] is True
+    assert result["json_schema"]["name"] == "user"
+    assert result["json_schema"]["schema"] == schema
+
+
+def test_mistral_type_adapter_output_genson_schema_builder(adapter):
+    """Test formatting Genson SchemaBuilder output type."""
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    assert isinstance(result, dict)
+    assert result["type"] == "json_schema"
+    assert result["json_schema"]["strict"] is True
+    assert result["json_schema"]["name"] == "schema"
+    expected_schema = {
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"],
+        "additionalProperties": False
+    }
+    assert result["json_schema"]["schema"] == expected_schema
+
+
+def test_mistral_type_adapter_output_json_schema_str(adapter, schema):
+    """Test formatting JsonSchema output type from string."""
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(JsonSchema(schema_str))
+    assert isinstance(result, dict)
+    assert result["type"] == "json_schema"
+    assert result["json_schema"]["strict"] is True
+    assert result["json_schema"]["name"] == "schema"
+    assert result["json_schema"]["schema"] == schema
+
+
+def test_mistral_type_adapter_output_json_schema_dict(adapter, schema):
+    """Test formatting JsonSchema output type from dict."""
+    result = adapter.format_output_type(JsonSchema(json.dumps(schema)))
+    assert isinstance(result, dict)
+    assert result["type"] == "json_schema"
+    assert result["json_schema"]["strict"] is True
+    assert result["json_schema"]["name"] == "schema"
+    assert result["json_schema"]["schema"] == schema
+
+
+def test_mistral_type_adapter_output_literal_error(adapter):
+    """Test that Literal output type raises TypeError."""
+    with pytest.raises(TypeError, match="Literal types are not supported with Mistral"):
+        adapter.format_output_type(Literal["Yes", "Maybe", "No"])
+
+
+def test_mistral_type_adapter_output_unsupported_regex(adapter):
+    """Test formatting unsupported regex output type."""
+    with pytest.raises(TypeError, match="Neither regex-based structured outputs.*dottxt instead"):
+        adapter.format_output_type(regex(r"\d+"))
+
+
+def test_mistral_type_adapter_output_unsupported_cfg(adapter):
+    """Test formatting unsupported CFG output type."""
+    with pytest.raises(TypeError, match="CFG-based structured outputs.*not available"):
+        adapter.format_output_type(cfg("grammar"))
+
+
+def test_mistral_type_adapter_output_unsupported_type(adapter):
+    """Test formatting unsupported output types."""
+    with pytest.raises(TypeError, match="The type str is not available"):
+        adapter.format_output_type(str)
+
+    with pytest.raises(TypeError, match="The type int is not available"):
+        adapter.format_output_type(int)
+
+
+def test_mistral_type_adapter_format_json_schema_type(adapter):
+    """Test formatting JSON schema type."""
+    schema_dict = {
+        "type": "object",
+        "properties": {"name": {"type": "string"}},
+        "required": ["name"]
+    }
+    result = adapter.format_json_schema_type(schema_dict, "TestSchema")
+    expected = {
+        "type": "json_schema",
+        "json_schema": {
+            "schema": {
+                "type": "object",
+                "properties": {"name": {"type": "string"}},
+                "required": ["name"],
+                "additionalProperties": False
+            },
+            "name": "testschema",
+            "strict": True
+        }
+    }
+    assert result == expected
+
+
+def test_mistral_type_adapter_format_json_mode_type(adapter):
+    """Test formatting JSON mode output type."""
+    result = adapter.format_json_mode_type()
+    assert result == {"type": "json_object"}
+
+
+def test_mistral_type_adapter_create_message_content_string(adapter):
+    """Test creating message content from string."""
+    result = adapter._create_message_content("Hello")
+    assert result == "Hello"
+
+
+def test_mistral_type_adapter_create_message_content_list_valid(adapter):
+    """Test creating message content from valid list."""
+    # Test with single string (should return the string)
+    result = adapter._create_message_content(["Hello"])
+    assert result == "Hello"
+
+
+def test_mistral_type_adapter_create_message_content_list_invalid(adapter):
+    """Test creating message content from invalid list."""
+    with pytest.raises(ValueError, match="The first item in the list should be a string"):
+        adapter._create_message_content([123, "world"])
+
+    with pytest.raises(ValueError, match="Content list cannot be empty."):
+        adapter._create_message_content([])
+
+    # Test that additional non-Image items are rejected
+    with pytest.raises(ValueError, match="Invalid item type in content list"):
+        adapter._create_message_content(["Hello", "world"])
+
+
+def test_mistral_type_adapter_create_message_content_invalid_type(adapter):
+    """Test creating message content from invalid type."""
+    with pytest.raises(ValueError, match="Invalid content type"):
+        adapter._create_message_content(123)
+
+    def test_batch_generation_not_supported(self, mistral_client):
+        """Test batch generation raises NotImplementedError like dottxt."""
+        model = from_mistral(mistral_client, "mistral-large-latest")
+        with pytest.raises(NotImplementedError, match="does not support batch inference"):
+            model.generate_batch(["What is Python?", "What is JavaScript?"])
+
+    def test_streaming_generation(self, mistral_client):
+        """Test basic streaming functionality."""
+        model = from_mistral(mistral_client, "mistral-large-latest")
+        chunks = []
+        for chunk in model.generate_stream("Count from 1 to 5: "):
+            chunks.append(chunk)
+
+        assert len(chunks) > 0
+        assert all(isinstance(chunk, str) for chunk in chunks)
+        full_text = "".join(chunks)
+        assert len(full_text) > 0
+
+    def test_nested_structured_generation(self, mistral_client):
+        """Test nested Pydantic model generation."""
+        class Address(BaseModel):
+            street: str
+            city: str
+            zip_code: str
+
+        class Employee(BaseModel):
+            name: str
+            role: str
+            department: str
+            address: Address
+
+        class Company(BaseModel):
+            company_name: str
+            employees: list[Employee]
+
+        model = from_mistral(mistral_client, model_name="mistral-large-latest")
+        prompt = Chat([
+            {"role": "system", "content": "You are a business data generator. All employees must work in the IT department."},
+            {"role": "user", "content": "Generate a JSON object representing a company with at least two employees. Return only the JSON, no other text."}
+        ])
+        result = model.generate(prompt, output_type=Company)
+        assert isinstance(result, str)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "company_name" in parsed
+        assert "employees" in parsed
+        assert isinstance(parsed["employees"], list)
+        assert len(parsed["employees"]) >= 2
+
+    def test_simple_vision_generation(self, mistral_client):
+        """Test basic vision functionality with simple image."""
+        pil_image = PILImage.new('RGB', (100, 100), color='red')
+        buffer = io.BytesIO()
+        pil_image.save(buffer, format="PNG")
+        buffer.seek(0)
+        pil_image = PILImage.open(buffer)
+
+        image = Image(pil_image)
+        model = from_mistral(mistral_client, "mistral-small-2503")
+        result = model.generate(["Describe this color", image])
+        assert isinstance(result, str)
+        assert len(result) > 0
+
+    def test_chat_with_multiple_roles(self, mistral_client):
+        """Test Chat with system, user, and assistant roles."""
+        model = from_mistral(mistral_client, "mistral-large-latest")
+
+        chat = Chat([
+            {"role": "system", "content": "You are a helpful math tutor."},
+            {"role": "user", "content": "What is 5 + 3?"},
+            {"role": "assistant", "content": "5 + 3 = 8."},
+            {"role": "user", "content": "What about 10 - 4?"}
+        ])
+
+        result = model.generate(chat)
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert "6" in result
+
+    def test_invalid_chat_role_error(self, mistral_client):
+        """Test that invalid chat roles raise ValueError."""
+        model = from_mistral(mistral_client, "mistral-large-latest")
+
+        invalid_chat = Chat([
+            {"role": "invalid_role", "content": "This should fail"}
+        ])
+
+        with pytest.raises(ValueError, match="role"):
+            model.generate(invalid_chat)
+
+    def test_structured_output_compatibility_check(self, mistral_client):
+        """Test supports_structured_output method."""
+        model = from_mistral(mistral_client, "mistral-large-latest")
+        supports_structured = model.supports_structured_output()
+        assert isinstance(supports_structured, bool)
+        assert supports_structured is True
+
+    def test_streaming_with_structured_output(self, mistral_client):
+        """Test streaming with JSON schema output."""
+        class Person(BaseModel):
+            name: str
+            age: int
+
+        model = from_mistral(mistral_client, "mistral-large-latest")
+        prompt = """Generate a JSON object representing a person with name and age.
+        Return only the JSON, no other text."""
+
+        chunks = []
+        for chunk in model.generate_stream(prompt, output_type=Person):
+            chunks.append(chunk)
+
+        result = "".join(chunks)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)
+
+
+@pytest.mark.integration
+class TestAsyncMistralIntegration:
+    @pytest.fixture
+    def api_key(self):
+        api_key = os.getenv("MISTRAL_API_KEY")
+        if not api_key:
+            pytest.skip("MISTRAL_API_KEY environment variable not set")
+        return api_key
+
+    @pytest.fixture
+    def mistral_client(self, api_key):
+        try:
+            from mistralai import Mistral as MistralClient
+            return MistralClient(api_key=api_key)
+        except ImportError:
+            pytest.skip("mistralai package not installed")
+
+    @pytest.mark.asyncio
+    async def test_async_init_from_client(self, api_key):
+        """Test async client initialization."""
+        from mistralai import Mistral as MistralClient
+
+        client = MistralClient(api_key=api_key)
+        model = from_mistral(client, MODEL_NAME, async_client=True)
+        assert model.client == client
+        assert model.model_name == MODEL_NAME
+
+    @pytest.mark.asyncio
+    async def test_async_wrong_input_type(self, mistral_client):
+        """Test async wrong input type error."""
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
+
+        with pytest.raises(TypeError, match="is not available"):
+            await model.generate(123)
+
+    @pytest.mark.asyncio
+    async def test_async_simple_generation(self, mistral_client):
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
+        result = await model.generate("What is 2+2? Answer in one sentence.")
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert "4" in result
+
+    @pytest.mark.asyncio
+    async def test_async_structured_generation(self, mistral_client):
+        class Person(BaseModel):
+            name: str
+            age: int
+        model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
+        prompt = """Generate a JSON object representing a person with name and age.
+        Return only the JSON, no other text."""
+        result = await model.generate(prompt, output_type=Person)
+        assert isinstance(result, str)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)
+
+    @pytest.mark.asyncio
+    async def test_async_streaming_generation(self, mistral_client):
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
+        chunks = []
+        async for chunk in model.generate_stream("Write a short story about a robot."):
+            chunks.append(chunk)
+        assert len(chunks) > 0
+        assert all(isinstance(chunk, str) for chunk in chunks)
+
+    @pytest.mark.asyncio
+    async def test_async_vision_generation(self, mistral_client):
+        """Test async vision generation."""
+        pil_image = PILImage.new('RGB', (100, 100), color='blue')
+        buffer = io.BytesIO()
+        pil_image.save(buffer, format="PNG")
+        buffer.seek(0)
+        pil_image = PILImage.open(buffer)
+
+        image = Image(pil_image)
+        model = from_mistral(mistral_client, "mistral-small-2503", async_client=True)
+        result = await model.generate(["Describe this color", image])
+        assert isinstance(result, str)
+        assert len(result) > 0
+
+    @pytest.mark.asyncio
+    async def test_async_structured_schema_error(self, mistral_client):
+        """Test schema error handling in async mode."""
+        class InvalidModel(BaseModel):
+            value: str = Field(pattern=r"^\d+$")
+        model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
+        with pytest.raises(TypeError, match="Mistral does not support your schema"):
+            await model.generate("Generate a number string", output_type=InvalidModel)
+
+    @pytest.mark.asyncio
+    async def test_async_batch_not_supported(self, mistral_client):
+        """Test async batch generation raises NotImplementedError."""
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
+        with pytest.raises(NotImplementedError, match="does not support batch inference"):
+            await model.generate_batch(["Hello", "World"])
+
+    @pytest.mark.asyncio
+    async def test_async_streaming_structured_generation(self, mistral_client):
+        """Test async streaming with structured output."""
+        class Person(BaseModel):
+            name: str
+            age: int
+
+        model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
+        prompt = """Generate a JSON object representing a person with name and age.
+        Return only the JSON, no other text."""
+
+        chunks = []
+        async for chunk in model.generate_stream(prompt, output_type=Person):
+            chunks.append(chunk)
+
+        result = "".join(chunks)
+        parsed = json.loads(result)
+        assert isinstance(parsed, dict)
+        assert "name" in parsed
+        assert isinstance(parsed["name"], str)
+        assert "age" in parsed
+        assert isinstance(parsed["age"], int)

From 9dd1412ec799dd13aa130592d4140dcb69e1d66c Mon Sep 17 00:00:00 2001
From: "Steven E. Elliott" <seeyallc6c@gmail.com>
Date: Mon, 29 Sep 2025 20:53:15 -0400
Subject: [PATCH 470/505] Another pass to address the issues rasied in PR 1755

---
 outlines/models/mistral.py                | 214 +++--
 tests/models/test_mistral.py              | 956 +++++++++++-----------
 tests/models/test_mistral_type_adapter.py | 419 +---------
 3 files changed, 583 insertions(+), 1006 deletions(-)

diff --git a/outlines/models/mistral.py b/outlines/models/mistral.py
index 806aca800..301b11dd0 100644
--- a/outlines/models/mistral.py
+++ b/outlines/models/mistral.py
@@ -16,6 +16,7 @@
     Optional,
     Union,
 )
+
 from functools import singledispatchmethod
 
 from pydantic import BaseModel, TypeAdapter
@@ -29,7 +30,6 @@
     is_typed_dict,
     is_pydantic_model,
     is_genson_schema_builder,
-    is_native_dict,
     is_literal,
 )
 
@@ -80,8 +80,10 @@ def format_str_model_input(self, model_input: str) -> list:
             A list of Mistral message objects.
         """
         from mistralai import UserMessage
+
         return [UserMessage(content=model_input)]
 
+
     @format_input.register(list)
     def format_list_model_input(self, model_input: list) -> list:
         """Format a list input into a list of messages.
@@ -89,16 +91,22 @@ def format_list_model_input(self, model_input: list) -> list:
         Parameters
         ----------
         model_input : list
-            The input list, containing a string prompt and optionally Image objects.
+            The input list, containing a string prompt and optionally Image objects (vision models only).
 
         Returns
         -------
         list
             A list of Mistral message objects.
+
+        Raises
+        ------
+        ValueError
+            If multimodal on non-vision model.
         """
         from mistralai import UserMessage
         return [UserMessage(content=self._create_message_content(model_input))]
 
+
     @format_input.register(Chat)
     def format_chat_model_input(self, model_input: Chat) -> list:
         """Format a Chat input into a list of messages.
@@ -114,6 +122,7 @@ def format_chat_model_input(self, model_input: Chat) -> list:
             A list of Mistral message objects.
         """
         from mistralai import UserMessage, AssistantMessage, SystemMessage
+
         messages = []
         for message in model_input.messages:
             role = message["role"]
@@ -128,6 +137,7 @@ def format_chat_model_input(self, model_input: Chat) -> list:
                 raise ValueError(f"Unsupported role: {role}")
         return messages
 
+
     def _create_message_content(self, content: Union[str, list]) -> Union[str, List[Dict[str, Union[str, Dict[str, str]]]]]:
         """Create message content from an input.
 
@@ -175,6 +185,7 @@ def _create_message_content(self, content: Union[str, list]) -> Union[str, List[
                 "Content must be a string or a list starting with a string followed by optional Image objects."
             )
 
+
     def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the `response_format` argument to pass to the client.
 
@@ -188,7 +199,35 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         dict
             The `response_format` dict to pass to the client.
         """
-        if isinstance(output_type, Regex):
+        # Handle None first
+        if output_type is None:
+            return {}
+
+        # Group JSON schema equivalent formats thematically
+        elif is_pydantic_model(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+        elif is_dataclass(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+        elif is_typed_dict(output_type):
+            schema = TypeAdapter(output_type).json_schema()
+            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+        elif is_genson_schema_builder(output_type):
+            schema = json.loads(output_type.to_json())
+            return self.format_json_schema_type(schema, "Schema")
+        elif isinstance(output_type, JsonSchema):
+            # CHANGE: Handle dict or str schema (common use)
+            if isinstance(output_type.schema, str):
+                schema = json.loads(output_type.schema)
+            else:
+                schema = output_type.schema
+            return self.format_json_schema_type(schema, "Schema")
+        elif output_type is dict or isinstance(output_type, dict):  # Covers type and instances
+            return {"type": "json_object"}
+
+        # Unsupported Outlines-defined types
+        elif isinstance(output_type, Regex):
             raise TypeError(
                 "Neither regex-based structured outputs nor the pattern keyword "
                 "in Json Schema are available with Mistral. Use an open source "
@@ -204,25 +243,6 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
                 "Literal types are not supported with Mistral. "
                 "Use a JSON schema with enum properties instead."
             )
-        elif is_pydantic_model(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
-        elif output_type is None:
-            return {}
-        elif is_native_dict(output_type):
-            return {"type": "json_object"}
-        elif is_dataclass(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
-        elif is_typed_dict(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
-        elif is_genson_schema_builder(output_type):
-            schema = json.loads(output_type.to_json())
-            return self.format_json_schema_type(schema, "Schema")
-        elif isinstance(output_type, JsonSchema):
-            schema = json.loads(output_type.schema)
-            return self.format_json_schema_type(schema, "Schema")
         else:
             type_name = getattr(output_type, "__name__", None) or str(output_type)
             raise TypeError(
@@ -230,6 +250,7 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
                 "Use an open source model or dottxt instead."
             )
 
+
     def format_json_schema_type(self, schema: dict, schema_name: str = "Schema") -> dict:
         """Format a JSON schema for the client.
 
@@ -255,15 +276,7 @@ def format_json_schema_type(self, schema: dict, schema_name: str = "Schema") ->
             }
         }
 
-    def format_json_mode_type(self) -> dict:
-        """Generate configuration for JSON mode output.
-
-        Returns
-        -------
-        dict
-            The JSON mode configuration.
-        """
-        return {"type": "json_object"}
+    # Removed: format_json_mode_type (unused, adds no value)
 
 
 class Mistral(Model):
@@ -312,6 +325,10 @@ def generate(
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
+        # CHANGE: Vision check (addresses 85 without breaking tests)
+        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
+            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
+
         try:
             result = self.client.chat.complete(
                 messages=messages,
@@ -325,7 +342,7 @@ def generate(
                     "Try a local model or dottxt instead."
                 )
             else:
-                raise RuntimeError(f"Error calling Mistral API: {e}") from e
+                raise RuntimeError(f"Mistral API error: {e}") from e
 
         messages = [choice.message for choice in result.choices]
         return messages[0].content if len(messages) == 1 else [m.content for m in messages]
@@ -383,6 +400,10 @@ def generate_stream(
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
+        # CHANGE: Vision check (addresses 85 without breaking tests)
+        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
+            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
+
         try:
             stream = self.client.chat.stream(
                 messages=messages,
@@ -396,7 +417,7 @@ def generate_stream(
                     "Try a local model or dottxt instead."
                 )
             else:
-                raise RuntimeError(f"Error calling Mistral API: {e}") from e
+                raise RuntimeError(f"Mistral API error: {e}") from e
 
         for chunk in stream:
             if (
@@ -406,30 +427,6 @@ def generate_stream(
             ):
                 yield chunk.data.choices[0].delta.content
 
-    def supports_structured_output(self, model_name: Optional[str] = None) -> bool:
-        """Check if the model supports structured output.
-
-        Parameters
-        ----------
-        model_name : Optional[str]
-            The name of the model to check (defaults to instance's model_name).
-
-        Returns
-        -------
-        bool
-            True if the model supports structured output, False otherwise.
-        """
-        model = model_name or self.model_name
-        if model is None:
-            return False
-
-        # Codestral Mamba does NOT support structured output
-        if "codestral-mamba" in model.lower():
-            return False
-
-        # All other models support structured output
-        return True
-
 
 class AsyncMistral(AsyncModel):
     """Async thin wrapper around the `mistralai.Mistral` client.
@@ -477,6 +474,10 @@ async def generate(
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
+        # CHANGE: Vision check (async)
+        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
+            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
+
         try:
             result = await self.client.chat.complete_async(
                 messages=messages,
@@ -523,82 +524,57 @@ async def generate_batch(
         )
 
     async def generate_stream(self, model_input, output_type=None, **inference_kwargs):
-        """
-        Generate text from the model as an async stream of chunks.
+        """Generate text from the model as an async stream of chunks.
 
-        Uses `stream_async` if streaming, otherwise `complete_async`.
-
-        Args:
-            model_input: str, list, or chat input to generate from.
-            output_type: Optional type for structured output.
-            **inference_kwargs: Extra kwargs like "model" name.
+        Parameters
+        ----------
+        model_input
+            str, list, or chat input to generate from.
+        output_type
+            Optional type for structured output.
+        **inference_kwargs
+            Extra kwargs like "model" name.
 
-        Yields:
-            str: chunks of text as they are streamed.
+        Yields
+        ------
+        str
+            Chunks of text as they are streamed.
         """
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
 
-        model_name = inference_kwargs.get("model", self.model_name)
-        stream = inference_kwargs.get("stream", True)
+        if "model" not in inference_kwargs and self.model_name is not None:
+            inference_kwargs["model"] = self.model_name
+
+        # CHANGE: Vision check (async stream)
+        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
+            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
 
         try:
-            if stream:
-                stream_kwargs = {
-                    "model": model_name,
-                    "messages": messages,
-                    **inference_kwargs
-                }
-
-                if response_format:
-                    stream_kwargs["response_format"] = response_format
-
-                response = await self.client.chat.stream_async(**stream_kwargs)
-
-                async for chunk in response:
-                    if (
-                        hasattr(chunk, "data")
-                        and chunk.data.choices
-                        and len(chunk.data.choices) > 0
-                        and hasattr(chunk.data.choices[0], "delta")
-                        and chunk.data.choices[0].delta.content is not None
-                    ):
-                        yield chunk.data.choices[0].delta.content
-            else:
-                complete_kwargs = {
-                    "model": model_name,
-                    "messages": messages,
-                    **inference_kwargs
-                }
+            # Build kwargs for stream_async
+            stream_kwargs = {
+                "messages": messages,
+                **inference_kwargs
+            }
+
+            if response_format:
+                stream_kwargs["response_format"] = response_format
 
-                if response_format:
-                    complete_kwargs["response_format"] = response_format
+            response = await self.client.chat.stream_async(**stream_kwargs)
 
-                res = await self.client.chat.complete_async(**complete_kwargs)
-                content = res.choices[0].message.content
-                yield content
+            async for chunk in response:
+                if (
+                    hasattr(chunk, "data")
+                    and chunk.data.choices
+                    and len(chunk.data.choices) > 0
+                    and hasattr(chunk.data.choices[0], "delta")
+                    and chunk.data.choices[0].delta.content is not None
+                ):
+                    yield chunk.data.choices[0].delta.content
 
         except Exception as e:
             raise RuntimeError(f"Mistral API error: {e}") from e
 
-    def supports_structured_output(self, model_name: Optional[str] = None) -> bool:
-        """Check if the model supports structured output.
-
-        Parameters
-        ----------
-        model_name : Optional[str]
-            The name of the model to check (defaults to instance's model_name).
-
-        Returns
-        -------
-        bool
-            True if the model supports structured output, False otherwise.
-        """
-        model = model_name or self.model_name
-        if model is None:
-            return True
-        return "codestral-mamba" not in model.lower()
-
 
 def from_mistral(
     client: "MistralClient",
diff --git a/tests/models/test_mistral.py b/tests/models/test_mistral.py
index 8653811f9..e4e2c291a 100644
--- a/tests/models/test_mistral.py
+++ b/tests/models/test_mistral.py
@@ -3,7 +3,6 @@
 Integration tests for outlines mistral api
 """
 
-
 import json
 import base64
 import os
@@ -18,76 +17,28 @@
 from outlines.models.mistral import from_mistral
 from outlines.types import JsonSchema
 
-# List of models to try in sequence
-MODEL_LIST = [
-    "mistral-small-2506",
-    "magistral-small-2506",
-    "mistral-large-latest",
-    "magistral-small-2509",
-    "magistral-small-2507",
-    "voxtral-small-2507",
-    "voxtral-mini-2507",
-    "devstral-small-2507",
-    "mistral-small-2503",
-    "mistral-small-2501",
-    "devstral-small-2505",
-    "pixtral-12b-2409",
-    "open-mistral-nemo",
-]
-
-# Vision-capable models
-VISION_MODELS = [
-    "magistral-small-2509",
-    "pixtral-12b-2409",
-    "mistral-small-2503",
-]
+MODEL_NAME = "mistral-large-latest"
+VISION_MODEL = "pixtral-large-latest"
 
 class AnimalClassification(BaseModel):
     image_subject: Literal["dog", "not an animal", "cat", "other"]
     specific_kind: str
 
-def try_models(mistral_client, model_list, async_client=False, **kwargs):
-    """Helper function to try generating with each model until one succeeds."""
-    last_exception = None
-    for model_name in model_list:
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=async_client)
-            generate_func = kwargs.pop("generate_func")
-            # Extract model_input (prompt or prompts) from kwargs
-            model_input = kwargs.pop("prompt", kwargs.pop("prompts", None))
-            if model_input is None:
-                raise ValueError("No 'prompt' or 'prompts' provided in kwargs")
-            # Call the generate function with model_input as positional arg
-            return model, generate_func(model, model_input=model_input, **kwargs)
-        except Exception as e:
-            if "429" in str(e):  # Handle rate limit errors
-                last_exception = e
-                continue
-            raise  # Re-raise non-429 errors
-    raise last_exception or Exception("No models succeeded")
-
-async def try_models_async(mistral_client, model_list, **kwargs):
-    """Helper function for async tests to try generating with each model."""
-    last_exception = None
-    for model_name in model_list:
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            generate_func = kwargs.pop("generate_func")
-            # Extract model_input (prompt or prompts) from kwargs
-            model_input = kwargs.pop("prompt", kwargs.pop("prompts", None))
-            if model_input is None:
-                raise ValueError("No 'prompt' or 'prompts' provided in kwargs")
-            # Call the async generate function with model_input as positional arg
-            return model, await generate_func(model, model_input=model_input, **kwargs)
-        except Exception as e:
-            if "429" in str(e):  # Handle rate limit errors
-                last_exception = e
-                continue
-            raise
-    raise last_exception or Exception("No models succeeded")
+def get_fallback_model(mistral_client):
+    """Dynamic fallback: Pick first stable chat model if primary rate-limits."""
+    try:
+        available_models = mistral_client.models.list()
+        # Prefer large/stable; avoid codestral (non-chat)
+        fallback = next((m.id for m in available_models if "large" in m.id.lower() and "codestral" not in m.id.lower()),
+                        next((m.id for m in available_models if "codestral" not in m.id.lower()), None))
+        return fallback
+    except Exception:
+        return None
 
 @pytest.mark.integration
+@pytest.mark.api_call
 class TestMistralIntegration:
+    @pytest.mark.api_call
     @pytest.fixture
     def api_key(self):
         api_key = os.getenv("MISTRAL_API_KEY")
@@ -95,6 +46,7 @@ def api_key(self):
             pytest.skip("MISTRAL_API_KEY environment variable not set")
         return api_key
 
+    @pytest.mark.api_call
     @pytest.fixture
     def mistral_client(self, api_key):
         try:
@@ -103,36 +55,54 @@ def mistral_client(self, api_key):
         except ImportError:
             pytest.skip("mistralai package not installed")
 
+    @pytest.mark.api_call
     def test_simple_generation(self, mistral_client):
-        def generate_func(model, model_input, **kwargs):
-            return model.generate(model_input, **kwargs)
-
-        model, result = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="What is 2+2? Answer in one sentence."
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            result = model.generate("What is 2+2? Answer in one sentence.")
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    result = model.generate("What is 2+2? Answer in one sentence.")
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         assert len(result) > 0
         assert "4" in result
 
+    @pytest.mark.api_call
     def test_structured_generation_json(self, mistral_client):
         class Person(BaseModel):
             name: str
             age: int
 
-        def generate_func(model, model_input, **kwargs):
-            return model.generate(model_input, **kwargs)
-
-        model, result = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="""Generate a JSON object representing a person with name and age.
-            Return only the JSON, no other text.""",
-            output_type=Person
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            result = model.generate(
+                """Generate a JSON object representing a person with name and age.
+                Return only the JSON, no other text.""",
+                output_type=Person
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    result = model.generate(
+                        """Generate a JSON object representing a person with name and age.
+                        Return only the JSON, no other text.""",
+                        output_type=Person
+                    )
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         parsed = json.loads(result)
         assert isinstance(parsed, dict)
@@ -141,37 +111,55 @@ def generate_func(model, model_input, **kwargs):
         assert "age" in parsed
         assert isinstance(parsed["age"], int)
 
+    @pytest.mark.api_call
     def test_streaming_generation(self, mistral_client):
-        def generate_func(model, model_input, **kwargs):
-            return list(model.generate_stream(model_input, **kwargs))
-
-        model, chunks = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="Write a short story about a robot."
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            chunks = list(model.generate_stream("Write a short story about a robot."))
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    chunks = list(model.generate_stream("Write a short story about a robot."))
+                else:
+                    raise
+            else:
+                raise
         assert len(chunks) > 0
         assert all(isinstance(chunk, str) for chunk in chunks)
         full_text = "".join(chunks)
         assert len(full_text) > 0
 
+    @pytest.mark.api_call
     def test_mistral_with_outlines(self, mistral_client):
         class Person(BaseModel):
             name: str
             age: int
 
-        def generate_func(model, model_input, **kwargs):
-            return model.generate(model_input, **kwargs)
-
-        model, result = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="""Generate a JSON object representing a person with name and age.
-            Return only the JSON, no other text.""",
-            output_type=Person
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            result = model.generate(
+                """Generate a JSON object representing a person with name and age.
+                Return only the JSON, no other text.""",
+                output_type=Person
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    result = model.generate(
+                        """Generate a JSON object representing a person with name and age.
+                        Return only the JSON, no other text.""",
+                        output_type=Person
+                    )
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         parsed = json.loads(result)
         assert isinstance(parsed, dict)
@@ -181,6 +169,7 @@ def generate_func(model, model_input, **kwargs):
         assert "age" in parsed
         assert isinstance(parsed["age"], int)
 
+    @pytest.mark.api_call
     def test_mistral_with_outlines_real(self, mistral_client):
         class Address(BaseModel):
             street: str
@@ -197,19 +186,32 @@ class Company(BaseModel):
             company_name: str
             employees: list[Employee]
 
-        def generate_func(model, model_input, **kwargs):
-            return model.generate(model_input, **kwargs)
-
-        model, result = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt=Chat([
-                {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
-                {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
-            ]),
-            output_type=Company
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            result = model.generate(
+                Chat([
+                    {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
+                    {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
+                ]),
+                output_type=Company
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    result = model.generate(
+                        Chat([
+                            {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
+                            {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
+                        ]),
+                        output_type=Company
+                    )
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         parsed = json.loads(result)
         assert isinstance(parsed, dict)
@@ -231,7 +233,112 @@ def generate_func(model, model_input, **kwargs):
             assert isinstance(employee["address"]["zip_code"], str)
             assert len(employee["address"]["zip_code"]) > 0
 
+    @pytest.mark.api_call
+    def test_custom_inference_parameters(self, mistral_client):
+        """Test custom inference parameters like temperature, max_tokens, top_p"""
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            result = model.generate(
+                "Write a very short creative story about a robot.",
+                temperature=0.9,
+                max_tokens=50,
+                top_p=0.95
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    result = model.generate(
+                        "Write a very short creative story about a robot.",
+                        temperature=0.9,
+                        max_tokens=50,
+                        top_p=0.95
+                    )
+                else:
+                    raise
+            else:
+                raise
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert len(result.split()) <= 100
+
+    @pytest.mark.api_call
+    def test_text_only_streaming(self, mistral_client):
+        """Test streaming without structured output"""
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            chunks = list(model.generate_stream("Count from 1 to 5 with spaces: "))
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    chunks = list(model.generate_stream("Count from 1 to 5 with spaces: "))
+                else:
+                    raise
+            else:
+                raise
+        assert len(chunks) > 0
+        assert all(isinstance(chunk, str) for chunk in chunks)
+        full_text = "".join(chunks)
+        assert len(full_text) > 0
+        assert any(str(i) in full_text for i in range(1, 6))
+
+    @pytest.mark.api_call
+    def test_multi_role_chat(self, mistral_client):
+        """Test Chat with system, user, and assistant roles"""
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name)
+            result = model.generate(
+                Chat([
+                    {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
+                    {"role": "user", "content": "What is 5 + 3?"},
+                    {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
+                    {"role": "user", "content": "What about 10 - 4?"}
+                ])
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback)
+                    result = model.generate(
+                        Chat([
+                            {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
+                            {"role": "user", "content": "What is 5 + 3?"},
+                            {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
+                            {"role": "user", "content": "What about 10 - 4?"}
+                        ])
+                    )
+                else:
+                    raise
+            else:
+                raise
+        assert isinstance(result, str)
+        assert len(result) > 0
+        assert "6" in result
+        assert "Happy learning!" in result
+
+    @pytest.mark.api_call
+    def test_empty_input_error(self, mistral_client):
+        """Test behavior with empty inputs"""
+        model = from_mistral(mistral_client, MODEL_NAME)
+        try:
+            result = model.generate("")
+            assert isinstance(result, str)
+        except Exception as e:
+            assert False, f"Empty string unexpectedly failed: {e}"
+
+        with pytest.raises(ValueError):
+            model.generate([])
+
+    @pytest.mark.api_call
     def test_real_image_classification(self, mistral_client):
+        # CHANGE: Incorporated Claude's real image test for better coverage
         image_url = "https://upload.wikimedia.org/wikipedia/commons/7/71/Charmonty_Norfolkterrier.jpg"
         headers = {
             'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
@@ -246,25 +353,34 @@ def test_real_image_classification(self, mistral_client):
         pil_image = PILImage.open(io.BytesIO(response.content))
         image = Image(pil_image)
 
-        last_exception = None
-        for model_name in VISION_MODELS:
-            try:
-                model = from_mistral(mistral_client, model_name)
-                prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
-                result = model.generate(prompt, output_type=AnimalClassification)
-                parsed = json.loads(result)
-                assert parsed["image_subject"] == "dog"
-                assert isinstance(parsed["specific_kind"], str)
-                assert len(parsed["specific_kind"]) > 0
-                return  # Success, exit the function
-            except Exception as e:
-                last_exception = e
-                continue
-
-        # If all models failed, raise the last exception
-        raise last_exception or Exception("No vision models succeeded")
-
+        model_name = VISION_MODEL
+        try:
+            model = from_mistral(mistral_client, model_name)
+            prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
+            result = model.generate(prompt, output_type=AnimalClassification)
+            parsed = json.loads(result)
+            assert parsed["image_subject"] == "dog"
+            assert isinstance(parsed["specific_kind"], str)
+            assert len(parsed["specific_kind"]) > 0
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback and "pixtral" in fallback.lower():
+                    model = from_mistral(mistral_client, fallback)
+                    prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
+                    result = model.generate(prompt, output_type=AnimalClassification)
+                    parsed = json.loads(result)
+                    assert parsed["image_subject"] == "dog"
+                    assert isinstance(parsed["specific_kind"], str)
+                    assert len(parsed["specific_kind"]) > 0
+                else:
+                    raise
+            else:
+                raise
+
+    @pytest.mark.api_call
     def test_image_classification_simple(self, mistral_client):
+        # CHANGE: Incorporated Claude's simple image test
         pil_image = PILImage.new('RGB', (100, 100), color='red')
         buffer = io.BytesIO()
         pil_image.save(buffer, format="PNG")
@@ -272,51 +388,34 @@ def test_image_classification_simple(self, mistral_client):
         pil_image = PILImage.open(buffer)
         image = Image(pil_image)
 
-        last_exception = None
-        for model_name in VISION_MODELS:
-            try:
-                model = from_mistral(mistral_client, model_name)
-                prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
-                result = model.generate(prompt, output_type=AnimalClassification)
-                parsed = json.loads(result)
-                assert parsed["image_subject"] == "not an animal"
-                assert parsed["specific_kind"] == "hotdog"
-                return  # Success, exit the function
-            except Exception as e:
-                last_exception = e
-                continue
-
-        # If all models failed, raise the last exception
-        raise last_exception or Exception("No vision models succeeded")
-
-
-    def test_sync_streaming_structured_generation(self, mistral_client):
-        class Person(BaseModel):
-            name: str
-            age: int
+        model_name = VISION_MODEL
+        try:
+            model = from_mistral(mistral_client, model_name)
+            prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
+            result = model.generate(prompt, output_type=AnimalClassification)
+            parsed = json.loads(result)
+            assert parsed["image_subject"] == "not an animal"
+            assert parsed["specific_kind"] == "hotdog"
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback and "pixtral" in fallback.lower():
+                    model = from_mistral(mistral_client, fallback)
+                    prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
+                    result = model.generate(prompt, output_type=AnimalClassification)
+                    parsed = json.loads(result)
+                    assert parsed["image_subject"] == "not an animal"
+                    assert parsed["specific_kind"] == "hotdog"
+                else:
+                    raise
+            else:
+                raise
 
-        def generate_func(model, model_input, **kwargs):
-            return list(model.generate_stream(model_input, **kwargs))
-
-        model, chunks = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="""Generate a JSON object representing a person with name and age.
-            Return only the JSON, no other text.""",
-            output_type=Person,
-            async_client=False
-        )
-        result = "".join(chunks)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
 
 @pytest.mark.integration
+@pytest.mark.api_call
 class TestAsyncMistralIntegration:
+    @pytest.mark.api_call
     @pytest.fixture
     def api_key(self):
         api_key = os.getenv("MISTRAL_API_KEY")
@@ -324,6 +423,7 @@ def api_key(self):
             pytest.skip("MISTRAL_API_KEY environment variable not set")
         return api_key
 
+    @pytest.mark.api_call
     @pytest.fixture
     def mistral_client(self, api_key):
         try:
@@ -332,38 +432,75 @@ def mistral_client(self, api_key):
         except ImportError:
             pytest.skip("mistralai package not installed")
 
+    @pytest.mark.api_call
+    @pytest.mark.asyncio
+    async def test_async_init_from_client(self, api_key):
+        """Test async client initialization."""
+        from mistralai import Mistral as MistralClient
+
+        client = MistralClient(api_key=api_key)
+        model = from_mistral(client, MODEL_NAME, async_client=True)
+        assert model.client == client
+        assert model.model_name == MODEL_NAME
+
+    @pytest.mark.api_call
+    @pytest.mark.asyncio
+    async def test_async_wrong_input_type(self, mistral_client):
+        """Test async wrong input type error."""
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
+
+        with pytest.raises(TypeError, match="is not available"):
+            await model.generate(123)
+
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_simple_generation(self, mistral_client):
-        async def generate_func(model, model_input, **kwargs):
-            return await model.generate(model_input, **kwargs)
-
-        model, result = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="What is 2+2? Answer in one sentence."
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
+            result = await model.generate("What is 2+2? Answer in one sentence.")
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback, async_client=True)
+                    result = await model.generate("What is 2+2? Answer in one sentence.")
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         assert len(result) > 0
         assert "4" in result
 
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_structured_generation(self, mistral_client):
         class Person(BaseModel):
             name: str
             age: int
-
-        async def generate_func(model, model_input, **kwargs):
-            return await model.generate(model_input, **kwargs)
-
-        model, result = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="""Generate a JSON object representing a person with name and age.
-            Return only the JSON, no other text.""",
-            output_type=Person
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
+            result = await model.generate(
+                """Generate a JSON object representing a person with name and age.
+                Return only the JSON, no other text.""",
+                output_type=Person
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback, async_client=True)
+                    result = await model.generate(
+                        """Generate a JSON object representing a person with name and age.
+                        Return only the JSON, no other text.""",
+                        output_type=Person
+                    )
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         parsed = json.loads(result)
         assert isinstance(parsed, dict)
@@ -372,156 +509,94 @@ async def generate_func(model, model_input, **kwargs):
         assert "age" in parsed
         assert isinstance(parsed["age"], int)
 
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_streaming_generation(self, mistral_client):
-        async def generate_func(model, model_input, **kwargs):
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
             chunks = []
-            async for chunk in model.generate_stream(model_input, **kwargs):
+            async for chunk in model.generate_stream("Write a short story about a robot."):
                 chunks.append(chunk)
-            return chunks
-
-        model, chunks = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="Write a short story about a robot."
-        )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback, async_client=True)
+                    chunks = []
+                    async for chunk in model.generate_stream("Write a short story about a robot."):
+                        chunks.append(chunk)
+                else:
+                    raise
+            else:
+                raise
         assert len(chunks) > 0
         assert all(isinstance(chunk, str) for chunk in chunks)
 
+    @pytest.mark.api_call
     @pytest.mark.asyncio
-    async def test_async_vision_classification(self, mistral_client):
-        image_url = "https://upload.wikimedia.org/wikipedia/commons/7/71/Charmonty_Norfolkterrier.jpg"
-        headers = {
-            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
-                          'AppleWebKit/537.36 (KHTML, like Gecko) '
-                          'Chrome/91.0.4472.124 Safari/537.36'
-        }
-        response = requests.get(image_url, headers=headers, timeout=30)
-        response.raise_for_status()
-        content_type = response.headers.get('content-type', '').lower()
-        if not any(img_type in content_type for img_type in ['image/', 'jpeg', 'jpg', 'png']):
-            raise ValueError(f"Expected image content, got: {content_type}")
-        pil_image = PILImage.open(io.BytesIO(response.content))
-        image = Image(pil_image)
-
-        last_exception = None
-        for model_name in VISION_MODELS:
-            try:
-                model = from_mistral(mistral_client, model_name, async_client=True)
-                prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
-                result = await model.generate(prompt, output_type=AnimalClassification)
-                parsed = json.loads(result)
-                assert parsed["image_subject"] == "dog"
-                assert isinstance(parsed["specific_kind"], str)
-                assert len(parsed["specific_kind"]) > 0
-                return  # Success, exit the function
-            except Exception as e:
-                last_exception = e
-                continue
-
-        # If all models failed, raise the last exception
-        raise last_exception or Exception("No vision models succeeded")
-
+    async def test_async_vision_generation(self, mistral_client):
+        """Test async vision generation."""
+        pil_image = PILImage.new('RGB', (100, 100), color='blue')
+        buffer = io.BytesIO()
+        pil_image.save(buffer, format="PNG")
+        buffer.seek(0)
+        pil_image = PILImage.open(buffer)
 
+        image = Image(pil_image)
+        model_name = VISION_MODEL
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
+            result = await model.generate(["Describe this color", image])
+            assert isinstance(result, str)
+            assert len(result) > 0
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback and "pixtral" in fallback.lower():
+                    model = from_mistral(mistral_client, fallback, async_client=True)
+                    result = await model.generate(["Describe this color", image])
+                    assert isinstance(result, str)
+                    assert len(result) > 0
+                else:
+                    raise
+            else:
+                raise
+
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_structured_schema_error(self, mistral_client):
+        """Test schema error handling in async mode."""
         class InvalidModel(BaseModel):
             value: str = Field(pattern=r"^\d+$")
-
-        # Just use the first model; we expect the error for all models
-        model = from_mistral(mistral_client, MODEL_LIST[0], async_client=True)
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
         with pytest.raises(TypeError, match="Mistral does not support your schema"):
             await model.generate("Generate a number string", output_type=InvalidModel)
 
-
-    @pytest.mark.asyncio
-    async def test_async_multiple_samples(self, mistral_client):
-        async def generate_func(model, model_input, **kwargs):
-            return await model.generate(model_input, **kwargs)
-
-        model, result = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="Respond with one word.",
-            n=2
-        )
-        assert isinstance(result, list)
-        assert len(result) == 2
-        assert all(isinstance(r, str) for r in result)
-
+    @pytest.mark.api_call
     @pytest.mark.asyncio
-    async def test_async_nested_structured(self, mistral_client):
-        class Address(BaseModel):
-            street: str
-            city: str
-            zip_code: str
-
-        class Employee(BaseModel):
-            name: str
-            role: str
-            department: str
-            address: Address
-
-        class Company(BaseModel):
-            company_name: str
-            employees: List[Employee]
-
-        async def generate_func(model, model_input, **kwargs):
-            return await model.generate(model_input, **kwargs)
-
-        model, result = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt=Chat([
-                {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
-                {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
-            ]),
-            output_type=Company
-        )
-        assert isinstance(result, str)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "company_name" in parsed
-        assert "employees" in parsed
-        assert isinstance(parsed["employees"], list)
-        assert len(parsed["employees"]) >= 2
-        for employee in parsed["employees"]:
-            assert employee["department"] == "IT"
-            assert isinstance(employee["name"], str)
-            assert len(employee["name"]) > 0
-            assert isinstance(employee["role"], str)
-            assert len(employee["role"]) > 0
-            assert isinstance(employee["address"], dict)
-            assert isinstance(employee["address"]["city"], str)
-            assert len(employee["address"]["city"]) > 0
-            assert isinstance(employee["address"]["street"], str)
-            assert len(employee["address"]["street"]) > 0
-            assert isinstance(employee["address"]["zip_code"], str)
-            assert len(employee["address"]["zip_code"]) > 0
+    async def test_async_batch_not_supported(self, mistral_client):
+        """Test async batch generation raises NotImplementedError."""
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
+        with pytest.raises(NotImplementedError, match="does not support batch inference"):
+            await model.generate_batch(["Hello", "World"])
 
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_streaming_structured_generation(self, mistral_client):
+        """Test async streaming with structured output."""
         class Person(BaseModel):
             name: str
             age: int
 
-        async def generate_func(model, model_input, **kwargs):
-            chunks = []
-            async for chunk in model.generate_stream(model_input, **kwargs):
-                chunks.append(chunk)
-            return chunks
-
-        model, chunks = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="""Generate a JSON object representing a person with name and age.
-            Return only the JSON, no other text.""",
-            output_type=Person
-        )
+        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
+        prompt = """Generate a JSON object representing a person with name and age.
+        Return only the JSON, no other text."""
+
+        chunks = []
+        async for chunk in model.generate_stream(prompt, output_type=Person):
+            chunks.append(chunk)
+
         result = "".join(chunks)
         parsed = json.loads(result)
         assert isinstance(parsed, dict)
@@ -530,194 +605,99 @@ async def generate_func(model, model_input, **kwargs):
         assert "age" in parsed
         assert isinstance(parsed["age"], int)
 
-@pytest.mark.integration
-class TestMistralAdditionalFeatures:
-    @pytest.fixture
-    def api_key(self):
-        api_key = os.getenv("MISTRAL_API_KEY")
-        if not api_key:
-            pytest.skip("MISTRAL_API_KEY environment variable not set")
-        return api_key
-
-    @pytest.fixture
-    def mistral_client(self, api_key):
-        try:
-            from mistralai import Mistral as MistralClient
-            return MistralClient(api_key=api_key)
-        except ImportError:
-            pytest.skip("mistralai package not installed")
-
-    def test_custom_inference_parameters(self, mistral_client):
-        """Test custom inference parameters like temperature, max_tokens, top_p"""
-        def generate_func(model, model_input, **kwargs):
-            return model.generate(model_input, **kwargs)
-
-        model, result = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="Write a very short creative story about a robot.",
-            temperature=0.9,
-            max_tokens=50,
-            top_p=0.95
-        )
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert len(result.split()) <= 100
-
-    def test_text_only_streaming(self, mistral_client):
-        """Test streaming without structured output"""
-        def generate_func(model, model_input, **kwargs):
-            return list(model.generate_stream(model_input, **kwargs))
-
-        model, chunks = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="Count from 1 to 5 with spaces: "
-        )
-        assert len(chunks) > 0
-        assert all(isinstance(chunk, str) for chunk in chunks)
-        full_text = "".join(chunks)
-        assert len(full_text) > 0
-        assert any(str(i) in full_text for i in range(1, 6))
-
-    def test_multi_role_chat(self, mistral_client):
-        """Test Chat with system, user, and assistant roles"""
-        from outlines.inputs import Chat
-
-        def generate_func(model, model_input, **kwargs):
-            return model.generate(model_input, **kwargs)
-
-        model, result = try_models(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt=Chat([
-                {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
-                {"role": "user", "content": "What is 5 + 3?"},
-                {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
-                {"role": "user", "content": "What about 10 - 4?"}
-            ])
-        )
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert "6" in result
-        assert "Happy learning!" in result
-
-    def test_model_compatibility_check(self, mistral_client):
-        """Test supports_structured_output method"""
-        # Just use the first model; no prompt needed
-        model = from_mistral(mistral_client, MODEL_LIST[0])
-        supports_structured = model.supports_structured_output()
-        assert isinstance(supports_structured, bool)
-
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_custom_inference_parameters(self, mistral_client):
         """Test async custom inference parameters"""
-        async def generate_func(model, model_input, **kwargs):
-            return await model.generate(model_input, **kwargs)
-
-        model, result = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="Write a very short creative story about a robot.",
-            temperature=0.9,
-            max_tokens=50,
-            top_p=0.95
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
+            result = await model.generate(
+                "Write a very short creative story about a robot.",
+                temperature=0.9,
+                max_tokens=50,
+                top_p=0.95
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback, async_client=True)
+                    result = await model.generate(
+                        "Write a very short creative story about a robot.",
+                        temperature=0.9,
+                        max_tokens=50,
+                        top_p=0.95
+                    )
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         assert len(result) > 0
         assert len(result.split()) <= 100
 
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_text_only_streaming(self, mistral_client):
         """Test async streaming without structured output"""
-        async def generate_func(model, model_input, **kwargs):
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
             chunks = []
-            async for chunk in model.generate_stream(model_input, **kwargs):
+            async for chunk in model.generate_stream("Count from 1 to 5 with spaces: "):
                 chunks.append(chunk)
-            return chunks
-
-        model, chunks = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt="Count from 1 to 5 with spaces: "
-        )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback, async_client=True)
+                    chunks = []
+                    async for chunk in model.generate_stream("Count from 1 to 5 with spaces: "):
+                        chunks.append(chunk)
+                else:
+                    raise
+            else:
+                raise
         assert len(chunks) > 0
         assert all(isinstance(chunk, str) for chunk in chunks)
         full_text = "".join(chunks)
         assert len(full_text) > 0
         assert any(str(i) in full_text for i in range(1, 6))
 
+    @pytest.mark.api_call
     @pytest.mark.asyncio
     async def test_async_multi_role_chat(self, mistral_client):
         """Test async Chat with multiple roles"""
-        from outlines.inputs import Chat
-
-        async def generate_func(model, model_input, **kwargs):
-            return await model.generate(model_input, **kwargs)
-
-        model, result = await try_models_async(
-            mistral_client,
-            MODEL_LIST,
-            generate_func=generate_func,
-            prompt=Chat([
-                {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
-                {"role": "user", "content": "What is 5 + 3?"},
-                {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
-                {"role": "user", "content": "What about 10 - 4?"}
-            ])
-        )
+        model_name = MODEL_NAME
+        try:
+            model = from_mistral(mistral_client, model_name, async_client=True)
+            result = await model.generate(
+                Chat([
+                    {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
+                    {"role": "user", "content": "What is 5 + 3?"},
+                    {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
+                    {"role": "user", "content": "What about 10 - 4?"}
+                ])
+            )
+        except Exception as e:
+            if "429" in str(e):
+                fallback = get_fallback_model(mistral_client)
+                if fallback:
+                    model = from_mistral(mistral_client, fallback, async_client=True)
+                    result = await model.generate(
+                        Chat([
+                            {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
+                            {"role": "user", "content": "What is 5 + 3?"},
+                            {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
+                            {"role": "user", "content": "What about 10 - 4?"}
+                        ])
+                    )
+                else:
+                    raise
+            else:
+                raise
         assert isinstance(result, str)
         assert len(result) > 0
         assert "6" in result
         assert "Happy learning!" in result
-
-    def test_empty_input_error(self, mistral_client):
-        """Test behavior with empty inputs"""
-        def generate_func(model, model_input, **kwargs):
-            return model.generate(model_input, **kwargs)
-
-        model, _ = try_models(mistral_client, MODEL_LIST, generate_func=generate_func, prompt="")
-        try:
-            result = model.generate("")
-            print(f"Empty string result: '{result}'")
-            print(f"Result type: {type(result)}")
-            print(f"Result length: {len(result)}")
-            assert isinstance(result, str)
-        except Exception as e:
-            print(f"Empty string exception raised: {type(e).__name__}: {e}")
-            assert False, f"Empty string unexpectedly failed with result: '{e}'"
-
-        with pytest.raises(ValueError):
-            model.generate([])
-
-    def test_mixed_content_list_validation(self, mistral_client):
-        from outlines.inputs import Image
-        from PIL import Image as PILImage
-        import io
-
-        pil_image = PILImage.new('RGB', (100, 100), color='blue')
-        buffer = io.BytesIO()
-        pil_image.save(buffer, format="PNG")
-        buffer.seek(0)
-        pil_image = PILImage.open(buffer)
-        image = Image(pil_image)
-
-        last_exception = None
-        for model_name in VISION_MODELS:
-            try:
-                model = from_mistral(mistral_client, model_name)
-                result = model.generate(["Describe this color", image])
-                assert isinstance(result, str)
-                assert len(result) > 0
-                return  # Success, exit the function
-            except Exception as e:
-                last_exception = e
-                continue
-
-        # If all models failed, raise the last exception
-        raise last_exception or Exception("No vision models succeeded")
diff --git a/tests/models/test_mistral_type_adapter.py b/tests/models/test_mistral_type_adapter.py
index 5bbc76285..0e6819983 100644
--- a/tests/models/test_mistral_type_adapter.py
+++ b/tests/models/test_mistral_type_adapter.py
@@ -1,16 +1,14 @@
-
 """
 outlines/tests/models/test_mistral_type_adapter.py
 Tests for MistralTypeAdapter class.
 """
 
-import os
 import json
 import pytest
 import sys
 from dataclasses import dataclass
-from typing import Literal
-from unittest.mock import patch
+from typing import Literal, get_origin  # CHANGE: Added get_origin for dict type check
+from unittest.mock import patch  # CHANGE: Kept for minimal use if needed, but not in input tests
 
 from genson import SchemaBuilder
 from pydantic import BaseModel, Field
@@ -19,12 +17,9 @@
 
 from outlines import cfg, json_schema, regex
 from outlines.inputs import Chat, Image
-from outlines.models.mistral import from_mistral
 from outlines.models.mistral import MistralTypeAdapter
 from outlines.types import JsonSchema
 
-MODEL_NAME = "mistral-small-2506"
-
 
 if sys.version_info >= (3, 12):
     from typing import TypedDict
@@ -56,29 +51,29 @@ def test_mistral_type_adapter_init(adapter):
     assert isinstance(adapter, MistralTypeAdapter)
 
 
-@patch('mistralai.UserMessage')
-def test_mistral_type_adapter_input_text(mock_user_msg, adapter):
+def test_mistral_type_adapter_input_text(adapter):
     """Test formatting string input."""
+    from mistralai import UserMessage  # Minimal import for type check
     message = "Hello world"
     result = adapter.format_input(message)
-    assert result is not None
-    mock_user_msg.assert_called_once_with(content=message)
+    assert len(result) == 1
+    assert isinstance(result[0], UserMessage)
+    assert result[0].content == message
 
 
-@patch('mistralai.UserMessage')
-def test_mistral_type_adapter_input_list(mock_user_msg, adapter):
+def test_mistral_type_adapter_input_list(adapter):
     """Test formatting list input."""
+    from mistralai import UserMessage  # Minimal import for type check
     message_list = ["Hello world"]
     result = adapter.format_input(message_list)
-    assert result is not None
-    mock_user_msg.assert_called_once_with(content="Hello world")
+    assert len(result) == 1
+    assert isinstance(result[0], UserMessage)
+    assert result[0].content == "Hello world"
 
 
-@patch('mistralai.UserMessage')
-@patch('mistralai.AssistantMessage')
-@patch('mistralai.SystemMessage')
-def test_mistral_type_adapter_input_chat(mock_system_msg, mock_assistant_msg, mock_user_msg, adapter):
+def test_mistral_type_adapter_input_chat(adapter):
     """Test formatting Chat input with system message."""
+    from mistralai import UserMessage, AssistantMessage, SystemMessage  # Minimal imports for type checks
     chat = Chat([
         {"role": "system", "content": "You are helpful"},
         {"role": "user", "content": "Hello"},
@@ -86,10 +81,11 @@ def test_mistral_type_adapter_input_chat(mock_system_msg, mock_assistant_msg, mo
         {"role": "user", "content": "How are you?"}
     ])
     result = adapter.format_input(chat)
-    assert result is not None
-    mock_system_msg.assert_called_once_with(content="You are helpful")
-    assert mock_user_msg.call_count == 2
-    mock_assistant_msg.assert_called_once_with(content="Hi there")
+    assert len(result) == 4
+    assert isinstance(result[0], SystemMessage) and result[0].content == "You are helpful"
+    assert isinstance(result[1], UserMessage) and result[1].content == "Hello"
+    assert isinstance(result[2], AssistantMessage) and result[2].content == "Hi there"
+    assert isinstance(result[3], UserMessage) and result[3].content == "How are you?"
 
 
 def test_mistral_type_adapter_input_invalid(adapter):
@@ -175,379 +171,4 @@ class User(TypedDict):
     assert result["type"] == "json_schema"
     assert result["json_schema"]["strict"] is True
     assert result["json_schema"]["name"] == "user"
-    assert result["json_schema"]["schema"] == schema
-
-
-def test_mistral_type_adapter_output_genson_schema_builder(adapter):
-    """Test formatting Genson SchemaBuilder output type."""
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    builder.add_object({"hi": "there"})
-    builder.add_object({"hi": 5})
-
-    result = adapter.format_output_type(builder)
-    assert isinstance(result, dict)
-    assert result["type"] == "json_schema"
-    assert result["json_schema"]["strict"] is True
-    assert result["json_schema"]["name"] == "schema"
-    expected_schema = {
-        "$schema": "http://json-schema.org/schema#",
-        "type": "object",
-        "properties": {"hi": {"type": ["integer", "string"]}},
-        "required": ["hi"],
-        "additionalProperties": False
-    }
-    assert result["json_schema"]["schema"] == expected_schema
-
-
-def test_mistral_type_adapter_output_json_schema_str(adapter, schema):
-    """Test formatting JsonSchema output type from string."""
-    schema_str = json.dumps(schema)
-    result = adapter.format_output_type(JsonSchema(schema_str))
-    assert isinstance(result, dict)
-    assert result["type"] == "json_schema"
-    assert result["json_schema"]["strict"] is True
-    assert result["json_schema"]["name"] == "schema"
-    assert result["json_schema"]["schema"] == schema
-
-
-def test_mistral_type_adapter_output_json_schema_dict(adapter, schema):
-    """Test formatting JsonSchema output type from dict."""
-    result = adapter.format_output_type(JsonSchema(json.dumps(schema)))
-    assert isinstance(result, dict)
-    assert result["type"] == "json_schema"
-    assert result["json_schema"]["strict"] is True
-    assert result["json_schema"]["name"] == "schema"
-    assert result["json_schema"]["schema"] == schema
-
-
-def test_mistral_type_adapter_output_literal_error(adapter):
-    """Test that Literal output type raises TypeError."""
-    with pytest.raises(TypeError, match="Literal types are not supported with Mistral"):
-        adapter.format_output_type(Literal["Yes", "Maybe", "No"])
-
-
-def test_mistral_type_adapter_output_unsupported_regex(adapter):
-    """Test formatting unsupported regex output type."""
-    with pytest.raises(TypeError, match="Neither regex-based structured outputs.*dottxt instead"):
-        adapter.format_output_type(regex(r"\d+"))
-
-
-def test_mistral_type_adapter_output_unsupported_cfg(adapter):
-    """Test formatting unsupported CFG output type."""
-    with pytest.raises(TypeError, match="CFG-based structured outputs.*not available"):
-        adapter.format_output_type(cfg("grammar"))
-
-
-def test_mistral_type_adapter_output_unsupported_type(adapter):
-    """Test formatting unsupported output types."""
-    with pytest.raises(TypeError, match="The type str is not available"):
-        adapter.format_output_type(str)
-
-    with pytest.raises(TypeError, match="The type int is not available"):
-        adapter.format_output_type(int)
-
-
-def test_mistral_type_adapter_format_json_schema_type(adapter):
-    """Test formatting JSON schema type."""
-    schema_dict = {
-        "type": "object",
-        "properties": {"name": {"type": "string"}},
-        "required": ["name"]
-    }
-    result = adapter.format_json_schema_type(schema_dict, "TestSchema")
-    expected = {
-        "type": "json_schema",
-        "json_schema": {
-            "schema": {
-                "type": "object",
-                "properties": {"name": {"type": "string"}},
-                "required": ["name"],
-                "additionalProperties": False
-            },
-            "name": "testschema",
-            "strict": True
-        }
-    }
-    assert result == expected
-
-
-def test_mistral_type_adapter_format_json_mode_type(adapter):
-    """Test formatting JSON mode output type."""
-    result = adapter.format_json_mode_type()
-    assert result == {"type": "json_object"}
-
-
-def test_mistral_type_adapter_create_message_content_string(adapter):
-    """Test creating message content from string."""
-    result = adapter._create_message_content("Hello")
-    assert result == "Hello"
-
-
-def test_mistral_type_adapter_create_message_content_list_valid(adapter):
-    """Test creating message content from valid list."""
-    # Test with single string (should return the string)
-    result = adapter._create_message_content(["Hello"])
-    assert result == "Hello"
-
-
-def test_mistral_type_adapter_create_message_content_list_invalid(adapter):
-    """Test creating message content from invalid list."""
-    with pytest.raises(ValueError, match="The first item in the list should be a string"):
-        adapter._create_message_content([123, "world"])
-
-    with pytest.raises(ValueError, match="Content list cannot be empty."):
-        adapter._create_message_content([])
-
-    # Test that additional non-Image items are rejected
-    with pytest.raises(ValueError, match="Invalid item type in content list"):
-        adapter._create_message_content(["Hello", "world"])
-
-
-def test_mistral_type_adapter_create_message_content_invalid_type(adapter):
-    """Test creating message content from invalid type."""
-    with pytest.raises(ValueError, match="Invalid content type"):
-        adapter._create_message_content(123)
-
-    def test_batch_generation_not_supported(self, mistral_client):
-        """Test batch generation raises NotImplementedError like dottxt."""
-        model = from_mistral(mistral_client, "mistral-large-latest")
-        with pytest.raises(NotImplementedError, match="does not support batch inference"):
-            model.generate_batch(["What is Python?", "What is JavaScript?"])
-
-    def test_streaming_generation(self, mistral_client):
-        """Test basic streaming functionality."""
-        model = from_mistral(mistral_client, "mistral-large-latest")
-        chunks = []
-        for chunk in model.generate_stream("Count from 1 to 5: "):
-            chunks.append(chunk)
-
-        assert len(chunks) > 0
-        assert all(isinstance(chunk, str) for chunk in chunks)
-        full_text = "".join(chunks)
-        assert len(full_text) > 0
-
-    def test_nested_structured_generation(self, mistral_client):
-        """Test nested Pydantic model generation."""
-        class Address(BaseModel):
-            street: str
-            city: str
-            zip_code: str
-
-        class Employee(BaseModel):
-            name: str
-            role: str
-            department: str
-            address: Address
-
-        class Company(BaseModel):
-            company_name: str
-            employees: list[Employee]
-
-        model = from_mistral(mistral_client, model_name="mistral-large-latest")
-        prompt = Chat([
-            {"role": "system", "content": "You are a business data generator. All employees must work in the IT department."},
-            {"role": "user", "content": "Generate a JSON object representing a company with at least two employees. Return only the JSON, no other text."}
-        ])
-        result = model.generate(prompt, output_type=Company)
-        assert isinstance(result, str)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "company_name" in parsed
-        assert "employees" in parsed
-        assert isinstance(parsed["employees"], list)
-        assert len(parsed["employees"]) >= 2
-
-    def test_simple_vision_generation(self, mistral_client):
-        """Test basic vision functionality with simple image."""
-        pil_image = PILImage.new('RGB', (100, 100), color='red')
-        buffer = io.BytesIO()
-        pil_image.save(buffer, format="PNG")
-        buffer.seek(0)
-        pil_image = PILImage.open(buffer)
-
-        image = Image(pil_image)
-        model = from_mistral(mistral_client, "mistral-small-2503")
-        result = model.generate(["Describe this color", image])
-        assert isinstance(result, str)
-        assert len(result) > 0
-
-    def test_chat_with_multiple_roles(self, mistral_client):
-        """Test Chat with system, user, and assistant roles."""
-        model = from_mistral(mistral_client, "mistral-large-latest")
-
-        chat = Chat([
-            {"role": "system", "content": "You are a helpful math tutor."},
-            {"role": "user", "content": "What is 5 + 3?"},
-            {"role": "assistant", "content": "5 + 3 = 8."},
-            {"role": "user", "content": "What about 10 - 4?"}
-        ])
-
-        result = model.generate(chat)
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert "6" in result
-
-    def test_invalid_chat_role_error(self, mistral_client):
-        """Test that invalid chat roles raise ValueError."""
-        model = from_mistral(mistral_client, "mistral-large-latest")
-
-        invalid_chat = Chat([
-            {"role": "invalid_role", "content": "This should fail"}
-        ])
-
-        with pytest.raises(ValueError, match="role"):
-            model.generate(invalid_chat)
-
-    def test_structured_output_compatibility_check(self, mistral_client):
-        """Test supports_structured_output method."""
-        model = from_mistral(mistral_client, "mistral-large-latest")
-        supports_structured = model.supports_structured_output()
-        assert isinstance(supports_structured, bool)
-        assert supports_structured is True
-
-    def test_streaming_with_structured_output(self, mistral_client):
-        """Test streaming with JSON schema output."""
-        class Person(BaseModel):
-            name: str
-            age: int
-
-        model = from_mistral(mistral_client, "mistral-large-latest")
-        prompt = """Generate a JSON object representing a person with name and age.
-        Return only the JSON, no other text."""
-
-        chunks = []
-        for chunk in model.generate_stream(prompt, output_type=Person):
-            chunks.append(chunk)
-
-        result = "".join(chunks)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
-
-
-@pytest.mark.integration
-class TestAsyncMistralIntegration:
-    @pytest.fixture
-    def api_key(self):
-        api_key = os.getenv("MISTRAL_API_KEY")
-        if not api_key:
-            pytest.skip("MISTRAL_API_KEY environment variable not set")
-        return api_key
-
-    @pytest.fixture
-    def mistral_client(self, api_key):
-        try:
-            from mistralai import Mistral as MistralClient
-            return MistralClient(api_key=api_key)
-        except ImportError:
-            pytest.skip("mistralai package not installed")
-
-    @pytest.mark.asyncio
-    async def test_async_init_from_client(self, api_key):
-        """Test async client initialization."""
-        from mistralai import Mistral as MistralClient
-
-        client = MistralClient(api_key=api_key)
-        model = from_mistral(client, MODEL_NAME, async_client=True)
-        assert model.client == client
-        assert model.model_name == MODEL_NAME
-
-    @pytest.mark.asyncio
-    async def test_async_wrong_input_type(self, mistral_client):
-        """Test async wrong input type error."""
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-
-        with pytest.raises(TypeError, match="is not available"):
-            await model.generate(123)
-
-    @pytest.mark.asyncio
-    async def test_async_simple_generation(self, mistral_client):
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-        result = await model.generate("What is 2+2? Answer in one sentence.")
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert "4" in result
-
-    @pytest.mark.asyncio
-    async def test_async_structured_generation(self, mistral_client):
-        class Person(BaseModel):
-            name: str
-            age: int
-        model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
-        prompt = """Generate a JSON object representing a person with name and age.
-        Return only the JSON, no other text."""
-        result = await model.generate(prompt, output_type=Person)
-        assert isinstance(result, str)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
-
-    @pytest.mark.asyncio
-    async def test_async_streaming_generation(self, mistral_client):
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-        chunks = []
-        async for chunk in model.generate_stream("Write a short story about a robot."):
-            chunks.append(chunk)
-        assert len(chunks) > 0
-        assert all(isinstance(chunk, str) for chunk in chunks)
-
-    @pytest.mark.asyncio
-    async def test_async_vision_generation(self, mistral_client):
-        """Test async vision generation."""
-        pil_image = PILImage.new('RGB', (100, 100), color='blue')
-        buffer = io.BytesIO()
-        pil_image.save(buffer, format="PNG")
-        buffer.seek(0)
-        pil_image = PILImage.open(buffer)
-
-        image = Image(pil_image)
-        model = from_mistral(mistral_client, "mistral-small-2503", async_client=True)
-        result = await model.generate(["Describe this color", image])
-        assert isinstance(result, str)
-        assert len(result) > 0
-
-    @pytest.mark.asyncio
-    async def test_async_structured_schema_error(self, mistral_client):
-        """Test schema error handling in async mode."""
-        class InvalidModel(BaseModel):
-            value: str = Field(pattern=r"^\d+$")
-        model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
-        with pytest.raises(TypeError, match="Mistral does not support your schema"):
-            await model.generate("Generate a number string", output_type=InvalidModel)
-
-    @pytest.mark.asyncio
-    async def test_async_batch_not_supported(self, mistral_client):
-        """Test async batch generation raises NotImplementedError."""
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-        with pytest.raises(NotImplementedError, match="does not support batch inference"):
-            await model.generate_batch(["Hello", "World"])
-
-    @pytest.mark.asyncio
-    async def test_async_streaming_structured_generation(self, mistral_client):
-        """Test async streaming with structured output."""
-        class Person(BaseModel):
-            name: str
-            age: int
-
-        model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
-        prompt = """Generate a JSON object representing a person with name and age.
-        Return only the JSON, no other text."""
-
-        chunks = []
-        async for chunk in model.generate_stream(prompt, output_type=Person):
-            chunks.append(chunk)
-
-        result = "".join(chunks)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
+    assert result

From 9367ccddc296845df41c54bdd5643defd38f80a7 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Thu, 2 Oct 2025 12:13:06 +0200
Subject: [PATCH 471/505] Make cosmethic changes to the Mistral model

---
 docs/features/models/mistral.md           |  586 ++++--------
 outlines/models/__init__.py               |    4 +-
 outlines/models/mistral.py                |  292 +++---
 tests/models/test_mistral.py              | 1035 +++++++--------------
 tests/models/test_mistral_type_adapter.py |  202 ++--
 5 files changed, 781 insertions(+), 1338 deletions(-)

diff --git a/docs/features/models/mistral.md b/docs/features/models/mistral.md
index 481e58576..8aec916f4 100644
--- a/docs/features/models/mistral.md
+++ b/docs/features/models/mistral.md
@@ -1,486 +1,265 @@
-// # Mistral AI Models
+# Mistral
 
-Outlines now supports models from [Mistral AI](https://mistral.ai) via the official `mistralai` package.
-This implementation provides a thin wrapper around the `mistralai.Mistral` client, handling prompt formatting, structured outputs, and streaming.
-This document describes how to use the **Mistral SDK** with Outlines, both in
-synchronous and asynchronous contexts.
+!!! Installation
 
----
+    You need to install the `mistralai` library to be able to use the Mistral API in Outlines. Install all optional dependencies of the `Mistral` model with: `pip install outlines[mistral]`.
 
-## 0. Installation
+    You also need to have an Mistral API key. This API key must either be set as an environment variable called `MISTRAL_API_KEY` or be provided to the `mistralai.Mistral` class when instantiating it.
 
-```bash
-pip install mistralai
-export MISTRAL_API_KEY=YOUR_API_KEY_GOES_HERE # Or in code if done securely
-```
-
-
----
-
-## 1. Open Models
+## Model Initialization
 
-Below is a list of the open-weight models available via Apache2 license released by Mistral. These can be
-used directly via the API or downloaded where applicable (as of 2025-09-19).
+To create an `Mistral` or `AsyncMistral` model instance, you can use the `from_mistral` function. It takes 3 arguments:
 
-| API Endpoint             | Version | Model                | Max Tokens | Description                                                                | Outlines Support   | Multi-Modal        |
-|--------------------------|---------|----------------------|------------|----------------------------------------------------------------------------|--------------------|--------------------|
-| `magistral-small-2509`   | 25.09   | Magistral Small 1.2  | 128k       | Small reasoning model (Sept 2025) with vision support.                     | ✔️                 | ✔️                 |
-| `magistral-small-2507`   | 25.07   | Magistral Small 1.1  | 40k        | Small reasoning model (July 2025).                                         | ✔️                 | ❌                 |
-| `voxtral-mini-2507`      | 25.07   | Voxtral Mini         | 32k        | Mini audio input model.                                                    | ❓                 | ❓                 |
-| `mistral-small-2506`     | 25.06   | Mistral Small 3.2    | 128k       | Updated small model (June 2025).                                           | ✔️                 | ✔️                 |
-| `magistral-small-2506`   | 25.06   | Magistral Small 1    | 40k        | First small reasoning model (June 2025).                                   | ✔️                 | ✔️                 |
-| `devstral-small-2507`    | 25.07   | Devstral Small 1.1   | 128k       | Tool-use model for code exploration and editing (July 2025).               | ❓                 | ❓                 |
-| `mistral-small-2503`     | 25.03   | Mistral Small 4    | 128k       | Small model with image understanding (March 2025).                         | ✔️                 | ✔️                 |
-| `mistral-small-2501`     | 25.01   | Mistral Small 3      | 32k        | Small model (Jan 2025).                                                    | ✔️                 | ❌                 |
-| `devstral-small-2505`    | 25.05   | Devstral Small 1     | 128k       | Open source 24B text model for tool-use and software agents (June 2025).   | ❓                 | ❓                 |
-| `pixtral-12b-2409`       | 24.09   | Pixtral 12B          | 128k       | 12B multimodal model with image understanding (Sept 2024).                 | ✔️                 | ✔️                 |
-| `open-mistral-nemo`      | 24.07   | Mistral Nemo 12B     | 128k       | Multilingual open source model (July 2024).                                | ✔️                 | ❓                 |
+- `client`: a `mistralai.Mistral` instance
+- `model_name` (optional): the name of the model you want to use
+- `async_client` (optional): whether it should create a sync or an async model
 
+As the `mistralai` library uses a single class to handle both sync and async requests, you must set the `async_client` argument to True to get an `AsyncMistral` model.
 
-**Mistral SDK Limitations:**
-- Regex and CFG-based structured outputs are not supported by the Mistral API.
-- Structured outputs (including Pydantic/dataclass-based ones) are unavailable for codestral-mamba models, because they lack JSON Schema support.
-
-**Outlines Expansion:**
-- Outlines provides a drop-in for Regex and CFG-based structured outputs, reducing the SDK limitations.
-
----
-## 2. Listing Available Models via SDK
-
-The mistralai python SDK provides methods to query the API for a list of [all available models](https://docs.mistral.ai/getting-started/models/models_overview/#api-versioning),
-including **paid endpoints** for [premium models](https://docs.mistral.ai/getting-started/models/models_overview/) in addition to open weights.
+For instance:
 
-##### Sync Example
 ```python
-#!/usr/bin/env python
-import os
-from mistralai import Mistral
-
-def main():
-    api_key = os.environ["MISTRAL_API_KEY"]
-    client = Mistral(api_key=api_key)
-    list_models_response = client.models.list()
-    print(list_models_response)
-
-if __name__ == "__main__":
-    main()
-```
+import mistralai
+import outlines
 
-##### Async Example
-```python
-#!/usr/bin/env python
-import asyncio
-import os
-from mistralai import Mistral
+# Create the Mistral client
+client = mistral.Mistral()
 
-async def main():
-    api_key = os.environ["MISTRAL_API_KEY"]
-    client = Mistral(api_key=api_key)
-    list_models_response = await client.models.list_async()
-    print(list_models_response)
+# Create a sync model
+model = outlines.from_mistral(
+    client,
+    "mistral-large-latest"
+)
 
-if __name__ == "__main__":
-    asyncio.run(main())
+# Create aa async model
+model = outlines.from_mistral(
+    client,
+    "mistral-large-latest",
+    True
+)
 ```
----
-## 3. Basic Usage
-
-
-### 3.1 mistralai SDK wrapper
 
-Generate outputs for the given model name :
+The mistralai python SDK provides methods to query the API for a list of [all available models](https://docs.mistral.ai/getting-started/models/models_overview/#api-versioning),
+including paid endpoints for [premium models](https://docs.mistral.ai/getting-started/models/models_overview/) in addition to open weights.
 
-#### Sync
-```python
-from mistralai import Mistral
-from outlines.models.mistral import from_mistral
+## Text Generation
 
-def main():
-    client = Mistral(api_key="YOUR_API_KEY")
-    model = from_mistral(client, "mistral-small-2506")
-    result = model.generate("What is 2+2? Answer in one sentence.")
-    print(result)
+Once you've created your Outlines `Mistral` model instance, you're all set to generate text with this provider. You can simply call the model with a prompt.
 
-if __name__ == "__main__":
-    main()
-```
+For instance:
 
-#### Async
 ```python
-import asyncio
-from mistralai import Mistral
-from outlines.models.mistral import from_mistral
+import mistralai
+import outlines
 
-async def main():
-    client = Mistral(api_key="YOUR_API_KEY")
-    model = from_mistral(client, "mistral-small-2506", async_client=True)
-    result = await model.generate("What is 2+2? Answer in one sentence.")
-    print(result)
+# Create the model
+model = outlines.from_mistral(
+    mistralai.Mistral(),
+    "mistral-large-latest"
+)
 
-if __name__ == "__main__":
-    asyncio.run(main())
+# Call it to generate text
+response = model("What's the capital of Latvia?", max_tokens=20)
+print(response) # 'Riga'
 ```
 
+#### Vision
 
-### 3.2 Structured Generation
+Some Mistral models support vision input. To use this feature, provide a list containing a text prompt and `Image` instances.
 
-Outlines automatically converts model outputs into your desired format.
+For instance:
 
-#### Sync
 ```python
-from typing import Literal
-from pydantic import BaseModel
+import io
+import requests
+import PIL
+import outlines
+import mistralai
+from outlines.inputs import Image
 
-# Multiple choice (via enum in JSON schema)
-sentiment = model.generate(
-    "The movie was absolutely fantastic!",
-    output_type=Literal["positive", "negative", "neutral"]
+# Create the model
+model = outlines.from_mistral(
+    mistralai.Mistral(),
+    "mistral-large-latest"
 )
 
-# Structured object (Pydantic model)
-class Person(BaseModel):
-    name: str
-    age: int
-    occupation: str
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
 
-person = model.generate(
-    "Generate information about a software engineer",
-    output_type=Person
-)
-```
+# Create the prompt containing the text and the image
+prompt = [
+    "Describe the image",
+    Image(get_image("https://picsum.photos/id/237/400/300"))
+]
 
-#### Async
-```python
-import asyncio
-from typing import Literal
-from pydantic import BaseModel
-
-async def main():
-    # Multiple choice (via enum in JSON schema)
-    sentiment = await model.generate(
-        "The movie was absolutely fantastic!",
-        output_type=Literal["positive", "negative", "neutral"]
-    )
-
-    # Structured object (Pydantic model)
-    class Person(BaseModel):
-        name: str
-        age: int
-        occupation: str
-
-    person = await model.generate(
-        "Generate information about a software engineer",
-        output_type=Person
-    )
-```
-
-- Structured outputs may be **unavailable** for `codestral-mamba` models (they lack JSON schema support).
-- For best results with complex/nested schemas, use `mistral-large-latest` or higher; simpler models may require retries.
-- Check model compatibility:
-```python
-if model.supports_structured_output():
-    # Safe to use Pydantic/JSON schemas
-    pass
-else:
-    print("Use text-only generation")
-```
-
----
-
-### 3.3 Streaming
-
-Streaming methods (`generate_stream` / async `generate_stream`) allow you to iterate over tokens as they are generated.
-
-#### Sync
-```python
-for chunk in model.generate_stream("Write a short story about AI"):
-    print(chunk, end="", flush=True)
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
 ```
 
-#### Async
-```python
-async for chunk in model.generate_stream("Write a short story about AI"):
-    print(chunk, end="", flush=True)
-```
-
-### 3.4 Structured Output with Streaming
-
-Structured JSON is streamed as text chunks that can be assembled and parsed:
+#### Chat
 
-#### Sync Structured Streaming
-```python
-import json
-from pydantic import BaseModel
-
-class Person(BaseModel):
-    name: str
-    age: int
-
-model = from_mistral(mistral_client, "mistral-large-latest")
-prompt = """Generate a JSON object representing a person with name and age.
-Return only the JSON, no other text."""
-
-chunks = []
-for chunk in model.generate_stream(prompt, output_type=Person):
-    chunks.append(chunk)
+You can also use chat inputs with the `Mistral` model. To do so, call the model with a `Chat` instance. The content of messsage within the chat can be vision inputs as described above.
 
-# Assemble and parse the complete JSON
-result = "".join(chunks)
-parsed = json.loads(result)
-print(f"Name: {parsed['name']}, Age: {parsed['age']}")
-```
+For instance:
 
-#### Async Structured Streaming
 ```python
-import asyncio
-import json
-from pydantic import BaseModel
-
-class Person(BaseModel):
-    name: str
-    age: int
-
-async def main():
-    model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
-    prompt = """Generate a JSON object representing a person with name and age.
-    Return only the JSON, no other text."""
+import io
+import requests
+import PIL
+import mistralai
+import outlines
+from outlines.inputs import Chat, Image
+
+# Create the model
+model = outlines.from_mistral(
+    mistralai.Mistral(),
+    "mistral-large-latest"
+)
 
-    chunks = []
-    async for chunk in model.generate_stream(prompt, output_type=Person):
-        chunks.append(chunk)
+# Function to get an image
+def get_image(url):
+    r = requests.get(url)
+    return PIL.Image.open(io.BytesIO(r.content))
 
-    # Assemble and parse the complete JSON
-    result = "".join(chunks)
-    parsed = json.loads(result)
-    print(f"Name: {parsed['name']}, Age: {parsed['age']}")
+# Create the chat input
+prompt = Chat([
+    {"role": "system", "content": "You are a helpful assistant."},
+    {
+        "role": "user",
+        "content": ["Describe the image", Image(get_image("https://picsum.photos/id/237/400/300"))]
+    },
+])
 
-if __name__ == "__main__":
-    asyncio.run(main())
+# Call the model to generate a response
+response = model(prompt, max_tokens=50)
+print(response) # 'This is a picture of a black dog.'
 ```
 
-**Note:** With structured outputs, individual chunks may not be valid JSON until the complete response is assembled.
+#### Streaming
 
+Finally, the `Mistral` model supports streaming through the `stream` method.
 
-### 3.5 Batch Processing
-
-Batch inference is currently **not supported** by the `mistralai` library:
+For instance:
 
 ```python
-# This will raise NotImplementedError
-# model.generate_batch(prompts)
-
-# Use sequential processing instead
-results = [model.generate(prompt) for prompt in prompts]
-```
-
-### 3.6 Error Handling
+import mistralai
+import outlines
 
-Usage of this component can raise specific exceptions for common issues:
+# Create the model
+model = outlines.from_mistral(
+    mistralai.Mistral(),
+    "mistral-large-latest"
+)
 
-```python
-try:
-    result = model.generate("Hello", output_type=Person)
-except RuntimeError as e:
-    print(f"API Error: {e}")  # e.g., rate limits, network issues
-except TypeError as e:
-    print(f"Schema/Output Error: {e}")  # e.g., unsupported schema, model incompatibility
-except ValueError as e:
-    print(f"Input Error: {e}")  # e.g., invalid message roles or content, empty lists
+# Stream the response
+for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
+    print(chunk) # 'Once...'
 ```
 
----
+## Structured Generation
 
-## 4. Advanced Usage
+Mistral provides supports for some forms of structured output: JSON schemas and JSON syntax. To use it, call the model with an `output_type` on top of your prompt.
 
-### 4.0 Custom Inference Parameters
-
-Pass inference parameters directly to `generate`:
+#### JSON Schema
 
 ```python
-# Sync
-result = model.generate(
-    "Write a poem",
-    temperature=0.9,  # More creative
-    max_tokens=200,
-    top_p=0.95
-)
-
-# Async
-result = await model.generate(
-    "Write a poem",
-    temperature=0.9,
-    max_tokens=200,
-    top_p=0.95
-)
-```
+from typing import List
+from pydantic import BaseModel
+import mistralai
+import outlines
 
-### 4.2 Multiple Samples
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
 
-Generate multiple responses using the `n` parameter:
+# Create the model
+model = outlines.from_mistral(
+    mistralai.Mistral(),
+    "mistral-large-latest"
+)
 
-```python
-# Sync
-results = model.generate("Respond with one word.", n=2)
-assert isinstance(results, list)
-assert len(results) == 2
-
-# Async
-results = await model.generate("Respond with one word.", n=2)
-assert isinstance(results, list)
-assert len(results) == 2
+# Call it with the output type to generate structured text
+result = model("Create a character, use the json format.", Character, top_p=0.1)
+print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 ```
 
-### 4.3 Multimodal Inputs (Images)
+#### JSON Syntax
 
-Mistral supports image inputs via `content` lists (e.g., for `pixtral-*` model). Outlines handles this in `Chat` or list inputs:
+What we mean by JSON syntax is what is sometimes called JSON mode, meaning that the model will return a valid JSON, but you do not get to specify its structure. To use this JSON mode, provide the `dict` type as an output type.
 
-#### Sync Image Processing
 ```python
-from outlines.inputs import Image
-from PIL import Image as PILImage
+import mistralai
+import outlines
 
-# Load image
-pil_image = PILImage.open("path/to/image.jpg")
-image = Image(pil_image)
-
-# Text + image prompt
-result = model.generate(
-    ["Describe this image", image],
-    model_name="pixtral-12b-2409"
+## Create the model
+model = outlines.from_mistral(
+    mistralai.Mistral(),
+    "mistral-large-latest"
 )
 
-# Or via Chat
-chat = Chat([
-    {"role": "user", "content": ["What do you see?", image]}
-])
-result = model.generate(chat)
-```
 
-#### Async Image Processing
-```python
-import asyncio
-from outlines.inputs import Image
-from PIL import Image as PILImage
-
-async def main():
-    # Load image
-    pil_image = PILImage.open("path/to/image.jpg")
-    image = Image(pil_image)
-
-    # Text + image prompt
-    result = await model.generate(
-        ["Describe this image", image],
-        model_name="pixtral-12b-2409"
-    )
-
-    # Or via Chat
-    chat = Chat([
-        {"role": "user", "content": ["What do you see?", image]}
-    ])
-    result = await model.generate(chat)
+# Call it with the output type to generate structured text
+result = model("Create a character, use the json format.", dict, temperature=0.5)
+print(result) # '{"first_name": "Henri", "last_name": "Smith", "height": "170"}'
 ```
 
-**Note:** The first item must be a string, followed by optional `Image` objects.
-Text-only models will raise errors.
+## Asynchronous Calls
 
-### 4.4 Chat Messages
+All features presented above for the sync model are also available for the async model.
 
-Use `Chat` objects for multi-turn conversations:
+For instance:
 
-#### Sync Chat
-```python
-from outlines.inputs import Chat
-
-chat = Chat([
-    {"role": "system", "content": "You are a helpful assistant."},
-    {"role": "user", "content": "What is Python?"},
-    {"role": "assistant", "content": "Python is a programming language..."},
-    {"role": "user", "content": "Tell me more about its uses."}
-])
-
-response = model.generate(chat)
-```
-
-#### Async Chat
 ```python
 import asyncio
-from outlines.inputs import Chat
-
-async def main():
-    chat = Chat([
-        {"role": "system", "content": "You are a helpful assistant."},
-        {"role": "user", "content": "What is Python?"},
-        {"role": "assistant", "content": "Python is a programming language..."},
-        {"role": "user", "content": "Tell me more about its uses."}
-    ])
-
-    response = await model.generate(chat)
-```
+import mistralai
+import outlines
+from pydantic import BaseModel
+from typing import List
 
-**Supported roles:** `system`, `user`, `assistant`. Other roles raise `ValueError`.
+class Character(BaseModel):
+    name: str
+    age: int
+    skills: List[str]
 
-### 4.5 Complex Nested Structures
+# Create the model
+model = outlines.from_mistral(
+    mistralai.Mistral(),
+    "mistral-large-latest",
+    True
+)
 
-Based on integration tests, Outlines handles complex nested Pydantic models:
+async def text_generation():
+    # Regular generation
+    response = await model("What's the capital of Latvia?", max_tokens=20)
+    print(response) # 'Riga'
 
-#### Sync Complex Structure
-```python
-import json
-from pydantic import BaseModel
-from typing import List
-from outlines.inputs import Chat
+    # Streaming
+    async for chunk in  model.stream("Tell me a short story about a cat.", max_tokens=50):
+        print(chunk, end="") # 'Once...'
 
-class Address(BaseModel):
-    street: str
-    city: str
-    zip_code: str
+    # Structured generation
+    result = await model("Create a character, use the json format.", Character, top_p=0.1)
+    print(result) # '{"name": "Evelyn", "age": 34, "skills": ["archery", "stealth", "alchemy"]}'
+    print(Character.model_validate_json(result)) # name=Evelyn, age=34, skills=['archery', 'stealth', 'alchemy']
 
-class Employee(BaseModel):
-    name: str
-    role: str
-    department: str
-    address: Address
+asyncio.run(text_generation())
+```
 
-class Company(BaseModel):
-    company_name: str
-    employees: List[Employee]
+## Inference arguments
 
-model = from_mistral(mistral_client, "mistral-large-latest")
-prompt = Chat([
-    {"role": "system", "content": "You are a business data generator. All employees must work in the IT department."},
-    {"role": "user", "content": "Generate a company with at least two employees. Return only JSON."}
-])
+When calling the model, you can provide keyword arguments that will be passed down to the `chat.complete` method of the Mistral client and its async and streaming equivalents. Some of the most common arguments include `max_tokens`, `temperature`, `stop` and `top_p`.
 
-result = model.generate(prompt, output_type=Company)
-parsed = json.loads(result)
-print(f"Company: {parsed['company_name']}")
-for emp in parsed['employees']:
-    print(f"Employee: {emp['name']} - {emp['role']}")
-```
+Another keyword argument of interest is `n`. If set with an integer value superior to 1, Mistral will generate several sample responses and you will receive a list of strings as a response to your model call.
 
-#### Async Complex Structure
-```python
-import asyncio
-import json
-
-async def main():
-    # Same model definitions as above
-    model = from_mistral(mistral_client, "mistral-large-latest", async_client=True)
-    prompt = Chat([
-        {"role": "system", "content": "You are a business data generator. All employees must work in the IT department."},
-        {"role": "user", "content": "Generate a company with at least two employees. Return only JSON."}
-    ])
-
-    result = await model.generate(prompt, output_type=Company)
-    parsed = json.loads(result)
-    print(f"Company: {parsed['company_name']}")
-    for emp in parsed['employees']:
-        print(f"Employee: {emp['name']} - {emp['role']}")
-```
+See the [Mistral API documentation](https://docs.mistral.ai/api/#tag/chat) for the full list of available arguments.
 
----
 
-## 5. Troubleshooting
+## Troubleshooting
 
 - **ImportError: No module named 'mistralai'**
   → Run `pip install mistralai`.
@@ -489,7 +268,7 @@ async def main():
   → Verify `MISTRAL_API_KEY` is set and valid. Test with the [Mistral Playground](https://chat.mistral.ai).
 
 - **Schema Error (e.g., "Mistral does not support your schema")**
-  → Avoid regex/CFG; ensure no `pattern` fields in Pydantic (Outlines sets `additionalProperties: false`); try a simpler schema or different model.
+  → Ensure no `pattern` fields in Pydantic (Outlines sets `additionalProperties: false`); try a simpler schema or a different Outlines model (local models in particular).
 
 - **Model Not Found Error**
   → Confirm the model name (e.g., `"mistral-small-latest"`) and your subscription tier. Check [docs](https://docs.mistral.ai/getting-started/models/).
@@ -500,16 +279,5 @@ async def main():
 - **Input Validation Errors**
   → Ensure Chat messages use valid roles (`system`, `user`, `assistant`); list inputs start with strings.
 
----
-
-## 6. Notes
-
-- Outlines automatically sets `additionalProperties: false` in JSON schemas for stricter parsing.
-- For multimodal inputs, use compatible models
-- `generate_batch` raises `NotImplementedError` due to API limitations.
-- For more examples (e.g., vision classification), see the integration tests at tests/models/test_mistral.py
-- **Author:** Steven E. Elliott aka see (seeyallc6c@gmail.com)
-
----
 
-*Last updated: September 19, 2025*
+*Last updated: October 2, 2025*
diff --git a/outlines/models/__init__.py b/outlines/models/__init__.py
index c11525b82..c707d7406 100644
--- a/outlines/models/__init__.py
+++ b/outlines/models/__init__.py
@@ -13,7 +13,7 @@
 from .dottxt import Dottxt, from_dottxt
 from .gemini import Gemini, from_gemini
 from .llamacpp import LlamaCpp, from_llamacpp
-from .mistral import Mistral, from_mistral
+from .mistral import AsyncMistral, Mistral, from_mistral
 from .mlxlm import MLXLM, from_mlxlm
 from .ollama import AsyncOllama, Ollama, from_ollama
 from .openai import AsyncOpenAI, OpenAI, from_openai
@@ -42,6 +42,7 @@
     VLLMOffline,
 ]
 AsyncBlackBoxModel = Union[
+    AsyncMistral,
     AsyncOllama,
     AsyncOpenAI,
     AsyncTGI,
@@ -61,6 +62,7 @@
     "from_gemini",
     "LlamaCpp",
     "from_llamacpp",
+    "AsyncMistral",
     "Mistral",
     "from_mistral",
     "MLXLM",
diff --git a/outlines/models/mistral.py b/outlines/models/mistral.py
index 301b11dd0..ea832c951 100644
--- a/outlines/models/mistral.py
+++ b/outlines/models/mistral.py
@@ -1,15 +1,10 @@
-"""
-outlines/models/mistral.py
-Integration with Mistral AI API.
-"""
+"""Integration with Mistral AI API."""
 
 import json
+from functools import singledispatchmethod
 from typing import (
     TYPE_CHECKING,
     Any,
-    AsyncGenerator,
-    AsyncIterator,
-    Coroutine,
     Iterator,
     List,
     Dict,
@@ -17,9 +12,7 @@
     Union,
 )
 
-from functools import singledispatchmethod
-
-from pydantic import BaseModel, TypeAdapter
+from pydantic import TypeAdapter
 
 from outlines.inputs import Chat, Image
 from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
@@ -27,23 +20,24 @@
 from outlines.types import JsonSchema, Regex, CFG
 from outlines.types.utils import (
     is_dataclass,
-    is_typed_dict,
-    is_pydantic_model,
     is_genson_schema_builder,
-    is_literal,
+    is_native_dict,
+    is_pydantic_model,
+    is_typed_dict,
 )
 
 if TYPE_CHECKING:
     from mistralai import Mistral as MistralClient
 
-__all__ = ["Mistral", "AsyncMistral", "from_mistral"]
+__all__ = ["AsyncMistral", "Mistral", "from_mistral"]
 
 
 class MistralTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `Mistral` model.
 
-    Prepares arguments for Mistral's client `chat.complete`, `chat.complete_async`, or `chat.stream` methods.
-    Handles input (prompt or chat messages) and output type (JSON schema, Pydantic model, or enum-based choices).
+    Prepares arguments for Mistral's client `chat.complete`,
+    `chat.complete_async`, or `chat.stream` methods. Handles input (prompt or
+    chat messages) and output type (JSON schema types).
     """
 
     @singledispatchmethod
@@ -59,6 +53,7 @@ def format_input(self, model_input):
         -------
         list
             The `messages` argument to pass to the client.
+
         """
         raise TypeError(
             f"The input type {type(model_input)} is not available with "
@@ -78,12 +73,12 @@ def format_str_model_input(self, model_input: str) -> list:
         -------
         list
             A list of Mistral message objects.
+
         """
         from mistralai import UserMessage
 
         return [UserMessage(content=model_input)]
 
-
     @format_input.register(list)
     def format_list_model_input(self, model_input: list) -> list:
         """Format a list input into a list of messages.
@@ -91,21 +86,18 @@ def format_list_model_input(self, model_input: list) -> list:
         Parameters
         ----------
         model_input : list
-            The input list, containing a string prompt and optionally Image objects (vision models only).
+            The input list, containing a string prompt and optionally Image
+            objects (vision models only).
 
         Returns
         -------
         list
             A list of Mistral message objects.
 
-        Raises
-        ------
-        ValueError
-            If multimodal on non-vision model.
         """
         from mistralai import UserMessage
-        return [UserMessage(content=self._create_message_content(model_input))]
 
+        return [UserMessage(content=self._create_message_content(model_input))]
 
     @format_input.register(Chat)
     def format_chat_model_input(self, model_input: Chat) -> list:
@@ -120,41 +112,45 @@ def format_chat_model_input(self, model_input: Chat) -> list:
         -------
         list
             A list of Mistral message objects.
+
         """
         from mistralai import UserMessage, AssistantMessage, SystemMessage
 
         messages = []
+
         for message in model_input.messages:
             role = message["role"]
             content = message["content"]
             if role == "user":
-                messages.append(UserMessage(content=self._create_message_content(content)))
+                messages.append(
+                    UserMessage(content=self._create_message_content(content))
+                )
             elif role == "assistant":
                 messages.append(AssistantMessage(content=content))
             elif role == "system":
                 messages.append(SystemMessage(content=content))
             else:
                 raise ValueError(f"Unsupported role: {role}")
-        return messages
 
+        return messages
 
-    def _create_message_content(self, content: Union[str, list]) -> Union[str, List[Dict[str, Union[str, Dict[str, str]]]]]:
+    def _create_message_content(
+        self, content: Union[str, list]
+    ) -> Union[str, List[Dict[str, Union[str, Dict[str, str]]]]]:
         """Create message content from an input.
 
         Parameters
         ----------
         content : Union[str, list]
-            The content to format, either a string or a list containing a string and optionally Image objects.
+            The content to format, either a string or a list containing a
+            string and optionally Image objects.
 
         Returns
         -------
         Union[str, List[Dict[str, Union[str, Dict[str, str]]]]]
-            The formatted content, either a string or a list of content parts (text and image URLs).
+            The formatted content, either a string or a list of content parts
+            (text and image URLs).
 
-        Raises
-        ------
-        ValueError
-            If the content type is invalid, the list is empty, or contains unsupported types.
         """
         if isinstance(content, str):
             return content
@@ -162,10 +158,14 @@ def _create_message_content(self, content: Union[str, list]) -> Union[str, List[
             if not content:
                 raise ValueError("Content list cannot be empty.")
             if not isinstance(content[0], str):
-                raise ValueError("The first item in the list should be a string.")
+                raise ValueError(
+                    "The first item in the list should be a string."
+                )
             if len(content) == 1:
                 return content[0]
-            content_parts: List[Dict[str, Union[str, Dict[str, str]]]] = [{"type": "text", "text": content[0]}]
+            content_parts: List[Dict[str, Union[str, Dict[str, str]]]] = [
+                {"type": "text", "text": content[0]}
+            ]
             for item in content[1:]:
                 if isinstance(item, Image):
                     data_url = f"data:{item.image_format};base64,{item.image_str}"
@@ -176,16 +176,16 @@ def _create_message_content(self, content: Union[str, list]) -> Union[str, List[
                 else:
                     raise ValueError(
                         f"Invalid item type in content list: {type(item)}. "
-                        "Expected Image objects after the first string."
+                        + "Expected Image objects after the first string."
                     )
             return content_parts
         else:
-            raise ValueError(
+            raise TypeError(
                 f"Invalid content type: {type(content)}. "
-                "Content must be a string or a list starting with a string followed by optional Image objects."
+                + "Content must be a string or a list starting with a string "
+                + "followed by optional Image objects."
             )
 
-
     def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the `response_format` argument to pass to the client.
 
@@ -198,75 +198,68 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         -------
         dict
             The `response_format` dict to pass to the client.
+
         """
-        # Handle None first
         if output_type is None:
             return {}
 
-        # Group JSON schema equivalent formats thematically
+        # JSON schema types
         elif is_pydantic_model(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+            schema = output_type.model_json_schema()
+            return self.format_json_schema_type(schema, output_type.__name__)
         elif is_dataclass(output_type):
             schema = TypeAdapter(output_type).json_schema()
-            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+            return self.format_json_schema_type(schema, output_type.__name__)
         elif is_typed_dict(output_type):
             schema = TypeAdapter(output_type).json_schema()
-            return self.format_json_schema_type(schema, getattr(output_type, "__name__", "Schema"))
+            return self.format_json_schema_type(schema, output_type.__name__)
         elif is_genson_schema_builder(output_type):
             schema = json.loads(output_type.to_json())
-            return self.format_json_schema_type(schema, "Schema")
+            return self.format_json_schema_type(schema)
         elif isinstance(output_type, JsonSchema):
-            # CHANGE: Handle dict or str schema (common use)
-            if isinstance(output_type.schema, str):
-                schema = json.loads(output_type.schema)
-            else:
-                schema = output_type.schema
-            return self.format_json_schema_type(schema, "Schema")
-        elif output_type is dict or isinstance(output_type, dict):  # Covers type and instances
+            return self.format_json_schema_type(json.loads(output_type.schema))
+
+        # Json mode
+        elif is_native_dict(output_type):
             return {"type": "json_object"}
 
-        # Unsupported Outlines-defined types
+        # Unsupported types
         elif isinstance(output_type, Regex):
             raise TypeError(
-                "Neither regex-based structured outputs nor the pattern keyword "
-                "in Json Schema are available with Mistral. Use an open source "
-                "model or dottxt instead."
+                "Regex-based structured outputs are not available with "
+                "Mistral."
             )
         elif isinstance(output_type, CFG):
             raise TypeError(
-                "CFG-based structured outputs are not available with Mistral. "
-                "Use an open source model or dottxt instead."
-            )
-        elif is_literal(output_type):
-            raise TypeError(
-                "Literal types are not supported with Mistral. "
-                "Use a JSON schema with enum properties instead."
+                "CFG-based structured outputs are not available with Mistral."
             )
         else:
-            type_name = getattr(output_type, "__name__", None) or str(output_type)
+            type_name = getattr(output_type, "__name__", str(output_type))
             raise TypeError(
-                f"The type {type_name} is not available with Mistral. "
-                "Use an open source model or dottxt instead."
+                f"The type {type_name} is not available with Mistral."
             )
 
-
-    def format_json_schema_type(self, schema: dict, schema_name: str = "Schema") -> dict:
-        """Format a JSON schema for the client.
+    def format_json_schema_type(
+        self, schema: dict, schema_name: str = "default"
+    ) -> dict:
+        """Create the `response_format` argument to pass to the client from a
+        JSON schema dictionary.
 
         Parameters
         ----------
         schema : dict
             The JSON schema to format.
         schema_name : str
-            The name of the schema (default: "Schema").
+            The name of the schema.
 
         Returns
         -------
         dict
-            The formatted JSON schema configuration.
+            The value of the `response_format` argument to pass to the client.
+
         """
         schema = set_additional_properties_false_json_schema(schema)
+
         return {
             "type": "json_schema",
             "json_schema": {
@@ -276,16 +269,18 @@ def format_json_schema_type(self, schema: dict, schema_name: str = "Schema") ->
             }
         }
 
-    # Removed: format_json_mode_type (unused, adds no value)
-
 
 class Mistral(Model):
     """Thin wrapper around the `mistralai.Mistral` client.
 
-    Converts input and output types to arguments for the `mistralai.Mistral` client's `chat.complete` or `chat.stream` methods.
+    Converts input and output types to arguments for the `mistralai.Mistral`
+    client's `chat.complete` or `chat.stream` methods.
+
     """
 
-    def __init__(self, client: "MistralClient", model_name: Optional[str] = None):
+    def __init__(
+        self, client: "MistralClient", model_name: Optional[str] = None
+    ):
         """
         Parameters
         ----------
@@ -293,6 +288,7 @@ def __init__(self, client: "MistralClient", model_name: Optional[str] = None):
             A mistralai.Mistral client instance.
         model_name : Optional[str]
             The name of the model to use.
+
         """
         self.client = client
         self.model_name = model_name
@@ -319,16 +315,14 @@ def generate(
         -------
         Union[str, list[str]]
             The response generated by the model as text.
+
         """
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
+
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
-        # CHANGE: Vision check (addresses 85 without breaking tests)
-        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
-            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
-
         try:
             result = self.client.chat.complete(
                 messages=messages,
@@ -344,8 +338,12 @@ def generate(
             else:
                 raise RuntimeError(f"Mistral API error: {e}") from e
 
-        messages = [choice.message for choice in result.choices]
-        return messages[0].content if len(messages) == 1 else [m.content for m in messages]
+        outputs = [choice.message for choice in result.choices]
+
+        if len(outputs) == 1:
+            return outputs[0].content
+        else:
+            return [m.content for m in outputs]
 
     def generate_batch(
         self,
@@ -353,24 +351,8 @@ def generate_batch(
         output_type=None,
         **inference_kwargs,
     ):
-        """Generate responses for a batch of inputs.
-
-        Parameters
-        ----------
-        model_input
-            The batch of inputs to generate responses from.
-        output_type
-            The desired format of the responses.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Raises
-        ------
-        NotImplementedError
-            Batch inference is not supported by the Mistral API.
-        """
         raise NotImplementedError(
-            "The mistralai library does not support batch inference."
+            "The `mistralai` library does not support batch inference."
         )
 
     def generate_stream(
@@ -394,16 +376,14 @@ def generate_stream(
         -------
         Iterator[str]
             An iterator that yields the text chunks generated by the model.
+
         """
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
+
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
-        # CHANGE: Vision check (addresses 85 without breaking tests)
-        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
-            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
-
         try:
             stream = self.client.chat.stream(
                 messages=messages,
@@ -431,10 +411,14 @@ def generate_stream(
 class AsyncMistral(AsyncModel):
     """Async thin wrapper around the `mistralai.Mistral` client.
 
-    Converts input and output types to arguments for the `mistralai.Mistral` client's async methods (`chat.complete_async` or `chat.stream`).
+    Converts input and output types to arguments for the `mistralai.Mistral`
+    client's async methods (`chat.complete_async` or `chat.stream_async`).
+
     """
 
-    def __init__(self, client: "MistralClient", model_name: Optional[str] = None):
+    def __init__(
+        self, client: "MistralClient", model_name: Optional[str] = None
+    ):
         """
         Parameters
         ----------
@@ -442,6 +426,7 @@ def __init__(self, client: "MistralClient", model_name: Optional[str] = None):
             A mistralai.Mistral client instance.
         model_name : Optional[str]
             The name of the model to use.
+
         """
         self.client = client
         self.model_name = model_name
@@ -468,16 +453,14 @@ async def generate(
         -------
         Union[str, list[str]]
             The response generated by the model as text.
+
         """
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
+
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
-        # CHANGE: Vision check (async)
-        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
-            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
-
         try:
             result = await self.client.chat.complete_async(
                 messages=messages,
@@ -494,8 +477,12 @@ async def generate(
             else:
                 raise RuntimeError(f"Mistral API error: {e}") from e
 
-        messages = [choice.message for choice in result.choices]
-        return messages[0].content if len(messages) == 1 else [m.content for m in messages]
+        outputs = [choice.message for choice in result.choices]
+
+        if len(outputs) == 1:
+            return outputs[0].content
+        else:
+            return [m.content for m in outputs]
 
     async def generate_batch(
         self,
@@ -503,27 +490,16 @@ async def generate_batch(
         output_type=None,
         **inference_kwargs,
     ):
-        """Generate responses for a batch of inputs asynchronously.
-
-        Parameters
-        ----------
-        model_input
-            The batch of inputs to generate responses from.
-        output_type
-            The desired format of the responses.
-        **inference_kwargs
-            Additional keyword arguments to pass to the client.
-
-        Raises
-        ------
-        NotImplementedError
-            Batch inference is not supported by the Mistral API.
-        """
         raise NotImplementedError(
             "The mistralai library does not support batch inference."
         )
 
-    async def generate_stream(self, model_input, output_type=None, **inference_kwargs):
+    async def generate_stream(
+        self,
+        model_input,
+        output_type=None,
+        **inference_kwargs,
+    ):
         """Generate text from the model as an async stream of chunks.
 
         Parameters
@@ -539,6 +515,7 @@ async def generate_stream(self, model_input, output_type=None, **inference_kwarg
         ------
         str
             Chunks of text as they are streamed.
+
         """
         messages = self.type_adapter.format_input(model_input)
         response_format = self.type_adapter.format_output_type(output_type)
@@ -546,34 +523,30 @@ async def generate_stream(self, model_input, output_type=None, **inference_kwarg
         if "model" not in inference_kwargs and self.model_name is not None:
             inference_kwargs["model"] = self.model_name
 
-        # CHANGE: Vision check (async stream)
-        if isinstance(model_input, list) and len(model_input) > 1 and self.model_name and "pixtral" not in self.model_name.lower():
-            raise ValueError(f"Model {self.model_name} does not support multimodal inputs. Use a vision model like pixtral-large-2411.")
-
         try:
-            # Build kwargs for stream_async
-            stream_kwargs = {
-                "messages": messages,
+            response = await self.client.chat.stream_async(
+                messages=messages,
+                response_format=response_format,
                 **inference_kwargs
-            }
-
-            if response_format:
-                stream_kwargs["response_format"] = response_format
-
-            response = await self.client.chat.stream_async(**stream_kwargs)
-
-            async for chunk in response:
-                if (
-                    hasattr(chunk, "data")
-                    and chunk.data.choices
-                    and len(chunk.data.choices) > 0
-                    and hasattr(chunk.data.choices[0], "delta")
-                    and chunk.data.choices[0].delta.content is not None
-                ):
-                    yield chunk.data.choices[0].delta.content
-
+            )
         except Exception as e:
-            raise RuntimeError(f"Mistral API error: {e}") from e
+            if "schema" in str(e).lower() or "json_schema" in str(e).lower():
+                raise TypeError(
+                    f"Mistral does not support your schema: {e}. "
+                    "Try a local model or dottxt instead."
+                )
+            else:
+                raise RuntimeError(f"Mistral API error: {e}") from e
+
+        async for chunk in response:
+            if (
+                hasattr(chunk, "data")
+                and chunk.data.choices
+                and len(chunk.data.choices) > 0
+                and hasattr(chunk.data.choices[0], "delta")
+                and chunk.data.choices[0].delta.content is not None
+            ):
+                yield chunk.data.choices[0].delta.content
 
 
 def from_mistral(
@@ -581,7 +554,8 @@ def from_mistral(
     model_name: Optional[str] = None,
     async_client: bool = False,
 ) -> Union[Mistral, AsyncMistral]:
-    """Create an Outlines Mistral model instance from a mistralai.Mistral client.
+    """Create an Outlines Mistral model instance from a mistralai.Mistral
+    client.
 
     Parameters
     ----------
@@ -590,18 +564,24 @@ def from_mistral(
     model_name : Optional[str]
         The name of the model to use.
     async_client : bool
-        If True, return an AsyncMistral instance; otherwise, return a Mistral instance.
+        If True, return an AsyncMistral instance;
+        otherwise, return a Mistral instance.
 
     Returns
     -------
     Union[Mistral, AsyncMistral]
         An Outlines Mistral or AsyncMistral model instance.
+
     """
     from mistralai import Mistral as MistralClient
 
     if not isinstance(client, MistralClient):
         raise ValueError(
-            "Invalid client type. The client must be an instance of mistralai.Mistral."
+            "Invalid client type. The client must be an instance of "
+            "`mistralai.Mistral`."
         )
 
-    return AsyncMistral(client, model_name) if async_client else Mistral(client, model_name)
+    if async_client:
+        return AsyncMistral(client, model_name)
+    else:
+        return Mistral(client, model_name)
diff --git a/tests/models/test_mistral.py b/tests/models/test_mistral.py
index e4e2c291a..ec8f7b0c2 100644
--- a/tests/models/test_mistral.py
+++ b/tests/models/test_mistral.py
@@ -1,703 +1,354 @@
-"""
-outlines/tests/models/test_mistral.py
-Integration tests for outlines mistral api
-"""
-
+import io
 import json
-import base64
 import os
-from typing import Dict, List, Literal, Iterator, AsyncIterator
-import requests
-import io
-from PIL import Image as PILImage
+from typing import Annotated, Generator, AsyncGenerator
+
 import pytest
+from PIL import Image as PILImage
+from mistralai import Mistral as MistralClient
 from pydantic import BaseModel, Field
 
-from outlines.inputs import Chat, Image
-from outlines.models.mistral import from_mistral
-from outlines.types import JsonSchema
+import outlines
+from outlines.inputs import Chat, Image, Video
+from outlines.models.mistral import AsyncMistral, Mistral
+from outlines.types import JsonSchema, Regex
+
 
 MODEL_NAME = "mistral-large-latest"
 VISION_MODEL = "pixtral-large-latest"
 
-class AnimalClassification(BaseModel):
-    image_subject: Literal["dog", "not an animal", "cat", "other"]
-    specific_kind: str
-
-def get_fallback_model(mistral_client):
-    """Dynamic fallback: Pick first stable chat model if primary rate-limits."""
-    try:
-        available_models = mistral_client.models.list()
-        # Prefer large/stable; avoid codestral (non-chat)
-        fallback = next((m.id for m in available_models if "large" in m.id.lower() and "codestral" not in m.id.lower()),
-                        next((m.id for m in available_models if "codestral" not in m.id.lower()), None))
-        return fallback
-    except Exception:
-        return None
-
-@pytest.mark.integration
+
+@pytest.fixture(scope="session")
+def api_key():
+    """Get the Mistral API key from the environment, providing a default value if not found.
+
+    This fixture should be used for tests that do not make actual api calls,
+    but still require to initialize the Mistral client.
+
+    """
+    api_key = os.getenv("MISTRAL_API_KEY")
+    if not api_key:
+        return "MOCK_VALUE"
+    return api_key
+
+
+@pytest.fixture(scope="session")
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
+
+    return image
+
+
+@pytest.fixture(scope="session")
+def model(api_key):
+    return Mistral(MistralClient(api_key=api_key), MODEL_NAME)
+
+
+@pytest.fixture(scope="session")
+def vision_model(api_key):
+    return Mistral(MistralClient(api_key=api_key), VISION_MODEL)
+
+
+@pytest.fixture(scope="session")
+def async_model(api_key):
+    return AsyncMistral(MistralClient(api_key=api_key), MODEL_NAME)
+
+
+@pytest.fixture(scope="session")
+def async_vision_model(api_key):
+    return AsyncMistral(MistralClient(api_key=api_key), VISION_MODEL)
+
+
+@pytest.fixture(scope="session")
+def model_no_model_name(api_key):
+    return Mistral(MistralClient(api_key=api_key))
+
+
+@pytest.fixture(scope="session")
+def async_model_no_model_name(api_key):
+    return AsyncMistral(MistralClient(api_key=api_key))
+
+
+def test_mistral_init_from_client(api_key):
+    client = MistralClient(api_key=api_key)
+
+    # With model name
+    model = outlines.from_mistral(client, MODEL_NAME)
+    assert isinstance(model, Mistral)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+    # Without model name
+    model = outlines.from_mistral(client)
+    assert isinstance(model, Mistral)
+    assert model.client == client
+    assert model.model_name is None
+
+
+def test_mistral_wrong_inference_parameters(model):
+    with pytest.raises(RuntimeError, match="got an unexpected"):
+        model("prompt", foo=10)
+
+
+def test_mistral_wrong_input_type(model):
+    with pytest.raises(TypeError, match="is not available"):
+        model(123)
+
+
+def test_mistral_wrong_output_type(model):
+    with pytest.raises(
+        TypeError,
+        match="Regex-based structured outputs are not available with Mistral.",
+    ):
+        model("prompt", Regex("^.*$"))
+
+
+@pytest.mark.api_call
+def test_mistral_call(model):
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_mistral_call_model_name(model_no_model_name):
+    result = model_no_model_name(
+        "Respond with one word. Not more.",
+        model=MODEL_NAME
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_mistral_multiple_samples(model):
+    result = model("Respond with one word. Not more.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+@pytest.mark.api_call
+def test_mistral_vision(image, vision_model):
+    result = vision_model(["What does this logo represent?", Image(image)])
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_mistral_chat(image, vision_model):
+    result = vision_model(Chat(messages=[
+        {"role": "system", "content": "You are a helpful assistant."},
+        {
+            "role": "user",
+            "content": ["What does this logo represent?", Image(image)]
+        },
+    ]), max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.api_call
+def test_mistral_pydantic(model):
+    class Foo(BaseModel):
+        bar: int
+
+    result = model("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
 @pytest.mark.api_call
-class TestMistralIntegration:
-    @pytest.mark.api_call
-    @pytest.fixture
-    def api_key(self):
-        api_key = os.getenv("MISTRAL_API_KEY")
-        if not api_key:
-            pytest.skip("MISTRAL_API_KEY environment variable not set")
-        return api_key
-
-    @pytest.mark.api_call
-    @pytest.fixture
-    def mistral_client(self, api_key):
-        try:
-            from mistralai import Mistral as MistralClient
-            return MistralClient(api_key=api_key)
-        except ImportError:
-            pytest.skip("mistralai package not installed")
-
-    @pytest.mark.api_call
-    def test_simple_generation(self, mistral_client):
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            result = model.generate("What is 2+2? Answer in one sentence.")
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    result = model.generate("What is 2+2? Answer in one sentence.")
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert "4" in result
-
-    @pytest.mark.api_call
-    def test_structured_generation_json(self, mistral_client):
-        class Person(BaseModel):
-            name: str
-            age: int
-
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            result = model.generate(
-                """Generate a JSON object representing a person with name and age.
-                Return only the JSON, no other text.""",
-                output_type=Person
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    result = model.generate(
-                        """Generate a JSON object representing a person with name and age.
-                        Return only the JSON, no other text.""",
-                        output_type=Person
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
-
-    @pytest.mark.api_call
-    def test_streaming_generation(self, mistral_client):
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            chunks = list(model.generate_stream("Write a short story about a robot."))
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    chunks = list(model.generate_stream("Write a short story about a robot."))
-                else:
-                    raise
-            else:
-                raise
-        assert len(chunks) > 0
-        assert all(isinstance(chunk, str) for chunk in chunks)
-        full_text = "".join(chunks)
-        assert len(full_text) > 0
-
-    @pytest.mark.api_call
-    def test_mistral_with_outlines(self, mistral_client):
-        class Person(BaseModel):
-            name: str
-            age: int
-
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            result = model.generate(
-                """Generate a JSON object representing a person with name and age.
-                Return only the JSON, no other text.""",
-                output_type=Person
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    result = model.generate(
-                        """Generate a JSON object representing a person with name and age.
-                        Return only the JSON, no other text.""",
-                        output_type=Person
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert len(parsed["name"]) > 0
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
-
-    @pytest.mark.api_call
-    def test_mistral_with_outlines_real(self, mistral_client):
-        class Address(BaseModel):
-            street: str
-            city: str
-            zip_code: str
-
-        class Employee(BaseModel):
-            name: str
-            role: str
-            department: str
-            address: Address
-
-        class Company(BaseModel):
-            company_name: str
-            employees: list[Employee]
-
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            result = model.generate(
-                Chat([
-                    {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
-                    {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
-                ]),
-                output_type=Company
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    result = model.generate(
-                        Chat([
-                            {"role": "system", "content": "You are a business data generator. All employees must work in the IT department, and addresses must include a valid US city."},
-                            {"role": "user", "content": "Generate a JSON object representing a company with at least two employees, including their names, roles, IT department, and addresses with valid US cities. Return only the JSON, no other text."}
-                        ]),
-                        output_type=Company
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "company_name" in parsed
-        assert "employees" in parsed
-        assert isinstance(parsed["employees"], list)
-        assert len(parsed["employees"]) >= 2
-        for employee in parsed["employees"]:
-            assert employee["department"] == "IT"
-            assert isinstance(employee["name"], str)
-            assert len(employee["name"]) > 0
-            assert isinstance(employee["role"], str)
-            assert len(employee["role"]) > 0
-            assert isinstance(employee["address"], dict)
-            assert isinstance(employee["address"]["city"], str)
-            assert len(employee["address"]["city"]) > 0
-            assert isinstance(employee["address"]["street"], str)
-            assert len(employee["address"]["street"]) > 0
-            assert isinstance(employee["address"]["zip_code"], str)
-            assert len(employee["address"]["zip_code"]) > 0
-
-    @pytest.mark.api_call
-    def test_custom_inference_parameters(self, mistral_client):
-        """Test custom inference parameters like temperature, max_tokens, top_p"""
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            result = model.generate(
-                "Write a very short creative story about a robot.",
-                temperature=0.9,
-                max_tokens=50,
-                top_p=0.95
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    result = model.generate(
-                        "Write a very short creative story about a robot.",
-                        temperature=0.9,
-                        max_tokens=50,
-                        top_p=0.95
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert len(result.split()) <= 100
-
-    @pytest.mark.api_call
-    def test_text_only_streaming(self, mistral_client):
-        """Test streaming without structured output"""
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            chunks = list(model.generate_stream("Count from 1 to 5 with spaces: "))
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    chunks = list(model.generate_stream("Count from 1 to 5 with spaces: "))
-                else:
-                    raise
-            else:
-                raise
-        assert len(chunks) > 0
-        assert all(isinstance(chunk, str) for chunk in chunks)
-        full_text = "".join(chunks)
-        assert len(full_text) > 0
-        assert any(str(i) in full_text for i in range(1, 6))
-
-    @pytest.mark.api_call
-    def test_multi_role_chat(self, mistral_client):
-        """Test Chat with system, user, and assistant roles"""
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name)
-            result = model.generate(
-                Chat([
-                    {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
-                    {"role": "user", "content": "What is 5 + 3?"},
-                    {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
-                    {"role": "user", "content": "What about 10 - 4?"}
-                ])
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback)
-                    result = model.generate(
-                        Chat([
-                            {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
-                            {"role": "user", "content": "What is 5 + 3?"},
-                            {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
-                            {"role": "user", "content": "What about 10 - 4?"}
-                        ])
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert "6" in result
-        assert "Happy learning!" in result
-
-    @pytest.mark.api_call
-    def test_empty_input_error(self, mistral_client):
-        """Test behavior with empty inputs"""
-        model = from_mistral(mistral_client, MODEL_NAME)
-        try:
-            result = model.generate("")
-            assert isinstance(result, str)
-        except Exception as e:
-            assert False, f"Empty string unexpectedly failed: {e}"
-
-        with pytest.raises(ValueError):
-            model.generate([])
-
-    @pytest.mark.api_call
-    def test_real_image_classification(self, mistral_client):
-        # CHANGE: Incorporated Claude's real image test for better coverage
-        image_url = "https://upload.wikimedia.org/wikipedia/commons/7/71/Charmonty_Norfolkterrier.jpg"
-        headers = {
-            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) '
-                          'AppleWebKit/537.36 (KHTML, like Gecko) '
-                          'Chrome/91.0.4472.124 Safari/537.36'
-        }
-        response = requests.get(image_url, headers=headers, timeout=30)
-        response.raise_for_status()
-        content_type = response.headers.get('content-type', '').lower()
-        if not any(img_type in content_type for img_type in ['image/', 'jpeg', 'jpg', 'png']):
-            raise ValueError(f"Expected image content, got: {content_type}")
-        pil_image = PILImage.open(io.BytesIO(response.content))
-        image = Image(pil_image)
-
-        model_name = VISION_MODEL
-        try:
-            model = from_mistral(mistral_client, model_name)
-            prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
-            result = model.generate(prompt, output_type=AnimalClassification)
-            parsed = json.loads(result)
-            assert parsed["image_subject"] == "dog"
-            assert isinstance(parsed["specific_kind"], str)
-            assert len(parsed["specific_kind"]) > 0
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback and "pixtral" in fallback.lower():
-                    model = from_mistral(mistral_client, fallback)
-                    prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
-                    result = model.generate(prompt, output_type=AnimalClassification)
-                    parsed = json.loads(result)
-                    assert parsed["image_subject"] == "dog"
-                    assert isinstance(parsed["specific_kind"], str)
-                    assert len(parsed["specific_kind"]) > 0
-                else:
-                    raise
-            else:
-                raise
-
-    @pytest.mark.api_call
-    def test_image_classification_simple(self, mistral_client):
-        # CHANGE: Incorporated Claude's simple image test
-        pil_image = PILImage.new('RGB', (100, 100), color='red')
-        buffer = io.BytesIO()
-        pil_image.save(buffer, format="PNG")
-        buffer.seek(0)
-        pil_image = PILImage.open(buffer)
-        image = Image(pil_image)
-
-        model_name = VISION_MODEL
-        try:
-            model = from_mistral(mistral_client, model_name)
-            prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
-            result = model.generate(prompt, output_type=AnimalClassification)
-            parsed = json.loads(result)
-            assert parsed["image_subject"] == "not an animal"
-            assert parsed["specific_kind"] == "hotdog"
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback and "pixtral" in fallback.lower():
-                    model = from_mistral(mistral_client, fallback)
-                    prompt = ["Classify this image as a cat, dog, other, or not an animal. For a cat and dog, classify the kind of species; otherwise the kind is hotdog", image]
-                    result = model.generate(prompt, output_type=AnimalClassification)
-                    parsed = json.loads(result)
-                    assert parsed["image_subject"] == "not an animal"
-                    assert parsed["specific_kind"] == "hotdog"
-                else:
-                    raise
-            else:
-                raise
-
-
-@pytest.mark.integration
+def test_mistral_pydantic_refusal(model):
+    class Foo(BaseModel):
+        bar: Annotated[str, Field(int, pattern=r"^\d+$")]
+
+    with pytest.raises(TypeError, match="Mistral does not support your schema"):
+        _ = model("foo?", Foo)
+
+
+@pytest.mark.api_call
+def test_mistral_vision_pydantic(vision_model, image):
+    class Logo(BaseModel):
+        name: int
+
+    result = vision_model(["What does this logo represent?", Image(image)], Logo)
+    assert isinstance(result, str)
+    assert "name" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_mistral_json_schema(model):
+    class Foo(BaseModel):
+        bar: int
+
+    schema = json.dumps(Foo.model_json_schema())
+
+    result = model("foo?", JsonSchema(schema))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.api_call
+def test_mistral_streaming(model):
+    result = model.stream("Respond with one word. Not more.")
+    assert isinstance(result, Generator)
+    assert isinstance(next(result), str)
+
+
+def test_mistral_batch(model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
+
+
+def test_mistral_async_init_from_client(api_key):
+    client = MistralClient(api_key=api_key)
+
+    # Async with model name
+    model = outlines.from_mistral(client, MODEL_NAME, async_client=True)
+    assert isinstance(model, AsyncMistral)
+    assert model.client == client
+    assert model.model_name == MODEL_NAME
+
+    # Async without model name
+    model = outlines.from_mistral(client, async_client=True)
+    assert isinstance(model, AsyncMistral)
+    assert model.client == client
+    assert model.model_name is None
+
+
+@pytest.mark.asyncio
+async def test_mistral_async_wrong_inference_parameters(async_model):
+    with pytest.raises(RuntimeError, match="got an unexpected"):
+        await async_model("prompt", foo=10)
+
+
+@pytest.mark.asyncio
+async def test_mistral_async_wrong_input_type(async_model):
+    with pytest.raises(TypeError, match="is not available"):
+        await async_model(123)
+
+
+@pytest.mark.asyncio
+async def test_mistral_async_wrong_output_type(async_model):
+    with pytest.raises(
+        TypeError,
+        match="Regex-based structured outputs are not available with Mistral.",
+    ):
+        await async_model("prompt", Regex("^.*$"))
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_call(async_model):
+    result = await async_model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_call_model_name(async_model_no_model_name):
+    result = await async_model_no_model_name(
+        "Respond with one word. Not more.",
+        model=MODEL_NAME,
+    )
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_multiple_samples(async_model):
+    result = await async_model("Respond with one word. Not more.", n=2)
+    assert isinstance(result, list)
+    assert len(result) == 2
+    assert isinstance(result[0], str)
+    assert isinstance(result[1], str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_vision(async_vision_model, image):
+    result = await async_vision_model(["What does this logo represent?", Image(image)])
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_chat(async_vision_model, image):
+    result = await async_vision_model(Chat(messages=[
+        {"role": "system", "content": "You are a helpful assistant."},
+        {
+            "role": "user",
+            "content": ["What does this logo represent?", Image(image)]
+        },
+    ]), max_tokens=10)
+    assert isinstance(result, str)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_pydantic(async_model):
+    class Foo(BaseModel):
+        bar: int
+
+    result = await async_model("foo?", Foo)
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_pydantic_refusal(async_model):
+    class Foo(BaseModel):
+        bar: Annotated[str, Field(int, pattern=r"^\d+$")]
+
+    with pytest.raises(TypeError, match="Mistral does not support your schema"):
+        _ = await async_model("foo?", Foo)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_vision_pydantic(async_vision_model, image):
+    class Logo(BaseModel):
+        name: int
+
+    result = await async_vision_model(["What does this logo represent?", Image(image)], Logo)
+    assert isinstance(result, str)
+    assert "name" in json.loads(result)
+
+
+@pytest.mark.asyncio
+@pytest.mark.api_call
+async def test_mistral_async_json_schema(async_model):
+    class Foo(BaseModel):
+        bar: int
+
+    schema = json.dumps(Foo.model_json_schema())
+
+    result = await async_model("foo?", JsonSchema(schema))
+    assert isinstance(result, str)
+    assert "bar" in json.loads(result)
+
+
+@pytest.mark.asyncio
 @pytest.mark.api_call
-class TestAsyncMistralIntegration:
-    @pytest.mark.api_call
-    @pytest.fixture
-    def api_key(self):
-        api_key = os.getenv("MISTRAL_API_KEY")
-        if not api_key:
-            pytest.skip("MISTRAL_API_KEY environment variable not set")
-        return api_key
-
-    @pytest.mark.api_call
-    @pytest.fixture
-    def mistral_client(self, api_key):
-        try:
-            from mistralai import Mistral as MistralClient
-            return MistralClient(api_key=api_key)
-        except ImportError:
-            pytest.skip("mistralai package not installed")
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_init_from_client(self, api_key):
-        """Test async client initialization."""
-        from mistralai import Mistral as MistralClient
-
-        client = MistralClient(api_key=api_key)
-        model = from_mistral(client, MODEL_NAME, async_client=True)
-        assert model.client == client
-        assert model.model_name == MODEL_NAME
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_wrong_input_type(self, mistral_client):
-        """Test async wrong input type error."""
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-
-        with pytest.raises(TypeError, match="is not available"):
-            await model.generate(123)
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_simple_generation(self, mistral_client):
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            result = await model.generate("What is 2+2? Answer in one sentence.")
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback, async_client=True)
-                    result = await model.generate("What is 2+2? Answer in one sentence.")
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert "4" in result
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_structured_generation(self, mistral_client):
-        class Person(BaseModel):
-            name: str
-            age: int
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            result = await model.generate(
-                """Generate a JSON object representing a person with name and age.
-                Return only the JSON, no other text.""",
-                output_type=Person
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback, async_client=True)
-                    result = await model.generate(
-                        """Generate a JSON object representing a person with name and age.
-                        Return only the JSON, no other text.""",
-                        output_type=Person
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_streaming_generation(self, mistral_client):
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            chunks = []
-            async for chunk in model.generate_stream("Write a short story about a robot."):
-                chunks.append(chunk)
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback, async_client=True)
-                    chunks = []
-                    async for chunk in model.generate_stream("Write a short story about a robot."):
-                        chunks.append(chunk)
-                else:
-                    raise
-            else:
-                raise
-        assert len(chunks) > 0
-        assert all(isinstance(chunk, str) for chunk in chunks)
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_vision_generation(self, mistral_client):
-        """Test async vision generation."""
-        pil_image = PILImage.new('RGB', (100, 100), color='blue')
-        buffer = io.BytesIO()
-        pil_image.save(buffer, format="PNG")
-        buffer.seek(0)
-        pil_image = PILImage.open(buffer)
-
-        image = Image(pil_image)
-        model_name = VISION_MODEL
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            result = await model.generate(["Describe this color", image])
-            assert isinstance(result, str)
-            assert len(result) > 0
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback and "pixtral" in fallback.lower():
-                    model = from_mistral(mistral_client, fallback, async_client=True)
-                    result = await model.generate(["Describe this color", image])
-                    assert isinstance(result, str)
-                    assert len(result) > 0
-                else:
-                    raise
-            else:
-                raise
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_structured_schema_error(self, mistral_client):
-        """Test schema error handling in async mode."""
-        class InvalidModel(BaseModel):
-            value: str = Field(pattern=r"^\d+$")
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-        with pytest.raises(TypeError, match="Mistral does not support your schema"):
-            await model.generate("Generate a number string", output_type=InvalidModel)
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_batch_not_supported(self, mistral_client):
-        """Test async batch generation raises NotImplementedError."""
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-        with pytest.raises(NotImplementedError, match="does not support batch inference"):
-            await model.generate_batch(["Hello", "World"])
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_streaming_structured_generation(self, mistral_client):
-        """Test async streaming with structured output."""
-        class Person(BaseModel):
-            name: str
-            age: int
-
-        model = from_mistral(mistral_client, MODEL_NAME, async_client=True)
-        prompt = """Generate a JSON object representing a person with name and age.
-        Return only the JSON, no other text."""
-
-        chunks = []
-        async for chunk in model.generate_stream(prompt, output_type=Person):
-            chunks.append(chunk)
-
-        result = "".join(chunks)
-        parsed = json.loads(result)
-        assert isinstance(parsed, dict)
-        assert "name" in parsed
-        assert isinstance(parsed["name"], str)
-        assert "age" in parsed
-        assert isinstance(parsed["age"], int)
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_custom_inference_parameters(self, mistral_client):
-        """Test async custom inference parameters"""
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            result = await model.generate(
-                "Write a very short creative story about a robot.",
-                temperature=0.9,
-                max_tokens=50,
-                top_p=0.95
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback, async_client=True)
-                    result = await model.generate(
-                        "Write a very short creative story about a robot.",
-                        temperature=0.9,
-                        max_tokens=50,
-                        top_p=0.95
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert len(result.split()) <= 100
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_text_only_streaming(self, mistral_client):
-        """Test async streaming without structured output"""
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            chunks = []
-            async for chunk in model.generate_stream("Count from 1 to 5 with spaces: "):
-                chunks.append(chunk)
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback, async_client=True)
-                    chunks = []
-                    async for chunk in model.generate_stream("Count from 1 to 5 with spaces: "):
-                        chunks.append(chunk)
-                else:
-                    raise
-            else:
-                raise
-        assert len(chunks) > 0
-        assert all(isinstance(chunk, str) for chunk in chunks)
-        full_text = "".join(chunks)
-        assert len(full_text) > 0
-        assert any(str(i) in full_text for i in range(1, 6))
-
-    @pytest.mark.api_call
-    @pytest.mark.asyncio
-    async def test_async_multi_role_chat(self, mistral_client):
-        """Test async Chat with multiple roles"""
-        model_name = MODEL_NAME
-        try:
-            model = from_mistral(mistral_client, model_name, async_client=True)
-            result = await model.generate(
-                Chat([
-                    {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
-                    {"role": "user", "content": "What is 5 + 3?"},
-                    {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
-                    {"role": "user", "content": "What about 10 - 4?"}
-                ])
-            )
-        except Exception as e:
-            if "429" in str(e):
-                fallback = get_fallback_model(mistral_client)
-                if fallback:
-                    model = from_mistral(mistral_client, fallback, async_client=True)
-                    result = await model.generate(
-                        Chat([
-                            {"role": "system", "content": "You are a helpful math tutor. Always end responses with 'Happy learning!'"},
-                            {"role": "user", "content": "What is 5 + 3?"},
-                            {"role": "assistant", "content": "5 + 3 = 8. Happy learning!"},
-                            {"role": "user", "content": "What about 10 - 4?"}
-                        ])
-                    )
-                else:
-                    raise
-            else:
-                raise
-        assert isinstance(result, str)
-        assert len(result) > 0
-        assert "6" in result
-        assert "Happy learning!" in result
+async def test_mistral_async_streaming(async_model):
+    result = async_model.stream("Respond with one word. Not more.")
+    assert isinstance(result, AsyncGenerator)
+    async for chunk in result:
+        assert isinstance(chunk, str)
+        break  # Just check the first chunk
+
+
+@pytest.mark.asyncio
+async def test_mistral_async_batch(async_model):
+    with pytest.raises(NotImplementedError, match="does not support"):
+        _ = await async_model.batch(
+            ["Respond with one word.", "Respond with one word."],
+        )
diff --git a/tests/models/test_mistral_type_adapter.py b/tests/models/test_mistral_type_adapter.py
index 0e6819983..33bdba058 100644
--- a/tests/models/test_mistral_type_adapter.py
+++ b/tests/models/test_mistral_type_adapter.py
@@ -1,25 +1,22 @@
-"""
-outlines/tests/models/test_mistral_type_adapter.py
-Tests for MistralTypeAdapter class.
-"""
-
+import io
 import json
-import pytest
 import sys
 from dataclasses import dataclass
-from typing import Literal, get_origin  # CHANGE: Added get_origin for dict type check
-from unittest.mock import patch  # CHANGE: Kept for minimal use if needed, but not in input tests
+from typing import Literal
 
-from genson import SchemaBuilder
-from pydantic import BaseModel, Field
-import io
+import pytest
 from PIL import Image as PILImage
+from genson import SchemaBuilder
+from mistralai import (
+    AssistantMessage,
+    SystemMessage,
+    UserMessage,
+)
+from pydantic import BaseModel
 
-from outlines import cfg, json_schema, regex
 from outlines.inputs import Chat, Image
 from outlines.models.mistral import MistralTypeAdapter
-from outlines.types import JsonSchema
-
+from outlines.types import CFG, JsonSchema, Regex
 
 if sys.version_info >= (3, 12):
     from typing import TypedDict
@@ -40,20 +37,26 @@ def schema():
         "additionalProperties": False,
     }
 
+@pytest.fixture
+def image():
+    width, height = 1, 1
+    white_background = (255, 255, 255)
+    image = PILImage.new("RGB", (width, height), white_background)
+
+    # Save to an in-memory bytes buffer and read as png
+    buffer = io.BytesIO()
+    image.save(buffer, format="PNG")
+    buffer.seek(0)
+    image = PILImage.open(buffer)
+
+    return image
 
 @pytest.fixture
 def adapter():
     return MistralTypeAdapter()
 
 
-def test_mistral_type_adapter_init(adapter):
-    """Test MistralTypeAdapter initialization."""
-    assert isinstance(adapter, MistralTypeAdapter)
-
-
 def test_mistral_type_adapter_input_text(adapter):
-    """Test formatting string input."""
-    from mistralai import UserMessage  # Minimal import for type check
     message = "Hello world"
     result = adapter.format_input(message)
     assert len(result) == 1
@@ -61,114 +64,153 @@ def test_mistral_type_adapter_input_text(adapter):
     assert result[0].content == message
 
 
-def test_mistral_type_adapter_input_list(adapter):
-    """Test formatting list input."""
-    from mistralai import UserMessage  # Minimal import for type check
-    message_list = ["Hello world"]
+def test_mistral_type_adapter_input_list(adapter, image):
+    image_input = Image(image)
+    message_list = ["Hello world", image_input]
     result = adapter.format_input(message_list)
     assert len(result) == 1
     assert isinstance(result[0], UserMessage)
-    assert result[0].content == "Hello world"
+    message_content = result[0].content
+    assert dict(message_content[0]) == {"type": "text", "text": "Hello world"}
+    assert message_content[1].type == "image_url"
+    assert hasattr(message_content[1], "image_url")
 
 
-def test_mistral_type_adapter_input_chat(adapter):
-    """Test formatting Chat input with system message."""
-    from mistralai import UserMessage, AssistantMessage, SystemMessage  # Minimal imports for type checks
+def test_mistral_type_adapter_input_chat(adapter, image):
+    image_input = Image(image)
     chat = Chat([
         {"role": "system", "content": "You are helpful"},
-        {"role": "user", "content": "Hello"},
+        {"role": "user", "content": ["Hello world", image_input]},
         {"role": "assistant", "content": "Hi there"},
-        {"role": "user", "content": "How are you?"}
     ])
     result = adapter.format_input(chat)
-    assert len(result) == 4
-    assert isinstance(result[0], SystemMessage) and result[0].content == "You are helpful"
-    assert isinstance(result[1], UserMessage) and result[1].content == "Hello"
-    assert isinstance(result[2], AssistantMessage) and result[2].content == "Hi there"
-    assert isinstance(result[3], UserMessage) and result[3].content == "How are you?"
-
-
-def test_mistral_type_adapter_input_invalid(adapter):
-    """Test formatting invalid input types."""
+    assert len(result) == 3
+    assert isinstance(result[0], SystemMessage)
+    assert result[0].content == "You are helpful"
+    assert isinstance(result[1], UserMessage)
+    assert dict(result[1].content[0]) == {"type": "text", "text": "Hello world"}
+    assert result[1].content[1].type == "image_url"
+    assert hasattr(result[1].content[1], "image_url")
+    assert isinstance(result[2], AssistantMessage)
+    assert result[2].content == "Hi there"
+
+
+def test_mistral_type_adapter_input_invalid(adapter, image):
     @dataclass
     class Audio:
         file: str
 
-    with pytest.raises(TypeError, match="The input type .* is not available"):
-        adapter.format_input(Audio("file"))
-
-    with pytest.raises(TypeError, match="The input type .* is not available"):
+    with pytest.raises(TypeError, match="is not available"):
         adapter.format_input(123)
 
-
-def test_mistral_type_adapter_input_list_invalid_content(adapter):
-    """Test formatting list input with invalid content."""
-    with pytest.raises(ValueError, match="The first item in the list should be a string"):
-        adapter.format_input([123])
-
     with pytest.raises(ValueError, match="Content list cannot be empty."):
         adapter.format_input([])
 
+    with pytest.raises(
+        ValueError,
+        match="The first item in the list should be a string.",
+    ):
+        adapter.format_input([Image(image)])
 
-def test_mistral_type_adapter_input_chat_invalid_role(adapter):
-    """Test formatting Chat input with invalid role."""
-    chat = Chat([{"role": "invalid", "content": "Hello"}])
-    with pytest.raises(ValueError, match="Unsupported role: invalid"):
-        adapter.format_input(chat)
+    with pytest.raises(
+        ValueError,
+        match="Expected Image objects after the first string"
+    ):
+        adapter.format_input(["hello", Audio("file")])
+
+    with pytest.raises(
+        TypeError,
+        match="Invalid content type",
+    ):
+        adapter.format_input(Chat([{"role": "user", "content": {}}]))
+
+    with pytest.raises(ValueError, match="Unsupported role"):
+        adapter.format_input(Chat([{"role": "invalid", "content": "Hello"}]))
 
 
 def test_mistral_type_adapter_output_none(adapter):
-    """Test formatting None output type."""
     result = adapter.format_output_type(None)
     assert result == {}
 
 
-def test_mistral_type_adapter_output_dict(adapter):
-    """Test formatting dict output type."""
+def test_mistral_type_adapter_output_json_mode(adapter):
     result = adapter.format_output_type(dict)
     assert result == {"type": "json_object"}
 
 
-def test_mistral_type_adapter_output_pydantic(adapter):
-    """Test formatting Pydantic model output type."""
-    class TestModel(BaseModel):
+def test_mistral_type_adapter_dataclass(adapter, schema):
+    @dataclass
+    class User:
+        user_id: int
         name: str
-        age: int
 
-    result = adapter.format_output_type(TestModel)
+    result = adapter.format_output_type(User)
     assert isinstance(result, dict)
-    assert result["type"] == "json_schema"
-    assert result["json_schema"]["name"] == "testmodel"
     assert result["json_schema"]["strict"] is True
-    assert "properties" in result["json_schema"]["schema"]
-    assert "name" in result["json_schema"]["schema"]["properties"]
-    assert "age" in result["json_schema"]["schema"]["properties"]
+    assert result["json_schema"]["schema"] == schema
 
 
-def test_mistral_type_adapter_output_dataclass(adapter, schema):
-    """Test formatting dataclass output type."""
-    @dataclass
-    class User:
+def test_mistral_type_adapter_typed_dict(adapter, schema):
+    class User(TypedDict):
         user_id: int
         name: str
 
     result = adapter.format_output_type(User)
     assert isinstance(result, dict)
-    assert result["type"] == "json_schema"
     assert result["json_schema"]["strict"] is True
-    assert result["json_schema"]["name"] == "user"
     assert result["json_schema"]["schema"] == schema
 
 
-def test_mistral_type_adapter_output_typed_dict(adapter, schema):
-    """Test formatting TypedDict output type."""
-    class User(TypedDict):
+def test_mistral_type_adapter_pydantic(adapter, schema):
+    class User(BaseModel):
         user_id: int
         name: str
 
     result = adapter.format_output_type(User)
     assert isinstance(result, dict)
-    assert result["type"] == "json_schema"
     assert result["json_schema"]["strict"] is True
-    assert result["json_schema"]["name"] == "user"
-    assert result
+    assert result["json_schema"]["schema"] == schema
+
+
+def test_mistral_type_adapter_genson_schema_builder(adapter, schema):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {}})
+    builder.add_object({"hi": "there"})
+    builder.add_object({"hi": 5})
+
+    result = adapter.format_output_type(builder)
+    assert isinstance(result, dict)
+    assert result["json_schema"]["strict"] is True
+    expected_schema = {
+        "$schema": "http://json-schema.org/schema#",
+        "type": "object",
+        "properties": {"hi": {"type": ["integer", "string"]}},
+        "required": ["hi"],
+        "additionalProperties": False
+    }
+    assert result["json_schema"]["schema"] == expected_schema
+
+
+def test_mistral_type_adapter_json_schema_str(adapter, schema):
+    schema_str = json.dumps(schema)
+    result = adapter.format_output_type(JsonSchema(schema_str))
+    assert isinstance(result, dict)
+    assert result["json_schema"]["strict"] is True
+    assert result["json_schema"]["schema"] == schema
+
+
+def test_mistral_type_adapter_output_unsupported(adapter):
+    with pytest.raises(
+        TypeError,
+        match="Regex-based structured outputs are not available with Mistral.",
+    ):
+        adapter.format_output_type(Regex("[0-9]"))
+
+    with pytest.raises(
+        TypeError,
+        match="CFG-based structured outputs are not available with Mistral.",
+    ):
+        adapter.format_output_type(CFG(""))
+
+    with pytest.raises(TypeError, match="is not available with Mistral."):
+        adapter.format_output_type(Literal["foo", "bar"])

From 62be0f21b7c48c539f9e5c446812dd658a918155 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@gmail.com>
Date: Mon, 6 Oct 2025 11:39:37 +0200
Subject: [PATCH 472/505] Add design partner CTA

---
 README.md | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index bcbb97ee2..f21a48681 100644
--- a/README.md
+++ b/README.md
@@ -21,7 +21,13 @@ Made with ❤👷️ by the team at [.txt](https://dottxt.co)
 
 </div>
 
-Need a high-performance commercial solution for structured outputs? Email us at [contact@dottxt.co](mailto:contact@dottxt.co), or [schedule a call](https://cal.com/team/dottxt/sales).
+## 🚀 Building the future of structured generation
+
+We're working with select partners to develop new interfaces to structured generation.
+
+Need XML, FHIR, custom schemas or grammars? Let's talk.
+
+[Become a design partner](mailto:contact@dottxt.ai)
 
 ## Table of Contents
 

From 605fe14237ee357e022c7705214177f88fddc133 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 24 Sep 2025 10:01:47 +0200
Subject: [PATCH 473/505] Remove json schema union types workaround

---
 outlines/types/dsl.py                       |   4 +-
 outlines/types/json_schema_utils.py         | 139 -----------------
 scripts/reproduce_1383.py                   |  83 ----------
 tests/types/test_json_schema_union_types.py | 158 --------------------
 4 files changed, 1 insertion(+), 383 deletions(-)
 delete mode 100644 scripts/reproduce_1383.py
 delete mode 100644 tests/types/test_json_schema_union_types.py

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index 5a12a9b68..ce992f9e2 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -41,7 +41,6 @@
 
 import outlines.types as types
 from outlines import grammars
-from outlines.types.json_schema_utils import preprocess_schema_for_union_types
 from outlines.types.utils import (
     get_schema_from_signature,
     is_int,
@@ -324,8 +323,7 @@ def __init__(
                 + "specification"
             )
 
-        # Preprocess the schema to handle union types
-        self.schema = preprocess_schema_for_union_types(schema_str, ensure_ascii=ensure_ascii)
+        self.schema = schema_str
         self.whitespace_pattern = whitespace_pattern
 
     def __post_init__(self):
diff --git a/outlines/types/json_schema_utils.py b/outlines/types/json_schema_utils.py
index 4c99e14a5..6f0a29967 100644
--- a/outlines/types/json_schema_utils.py
+++ b/outlines/types/json_schema_utils.py
@@ -1,140 +1 @@
 """Utilities for handling JSON schema compatibility."""
-
-import json
-from typing import Any, Union
-
-
-def preprocess_schema_for_union_types(
-    schema: Union[str, dict], ensure_ascii: bool = True
-) -> str:
-    """
-    Preprocess a JSON schema to handle union types (array type specifications).
-
-    This is a temporary workaround for the limitation in outlines-core 0.1.26
-    which doesn't support JSON schema type arrays like ["string", "null"].
-    This function converts such arrays into the equivalent anyOf format.
-
-    Parameters
-    ----------
-    schema
-        The JSON schema as a string or dictionary
-    ensure_ascii
-        Whether to ensure the output JSON is ASCII-only
-
-    Returns
-    -------
-    str
-        The preprocessed JSON schema string
-
-    Examples
-    --------
-    >>> schema = {"type": ["string", "null"]}
-    >>> preprocess_schema_for_union_types(schema)
-    '{"anyOf":[{"type":"string"},{"type":"null"}]}'
-    """
-    # Convert to dict if string
-    if isinstance(schema, str):
-        original_str = schema
-        try:
-            schema_dict = json.loads(schema)
-        except (json.JSONDecodeError, ValueError):
-            # If JSON parsing fails, return original string unchanged
-            # This preserves original error handling behavior
-            return original_str
-    else:
-        original_str = None
-        schema_dict = schema
-
-    # Process the schema
-    preprocessed = _convert_type_arrays_to_anyof(schema_dict)
-
-    # If no changes were made, return the original string (if it was a string)
-    if preprocessed == schema_dict and original_str is not None: # pragma: no cover
-        return original_str
-
-    # Return as JSON string with proper formatting
-    return json.dumps(preprocessed, ensure_ascii=ensure_ascii, separators=(",", ":"))
-
-
-def _convert_type_arrays_to_anyof(obj: Any) -> Any:
-    """
-    Recursively convert type arrays to anyOf format.
-
-    Parameters
-    ----------
-    obj
-        The JSON schema object or sub-object to process
-
-    Returns
-    -------
-    Any
-        The processed object with type arrays converted to anyOf
-    """
-    if isinstance(obj, dict):
-        # Check if this object has a type array that needs conversion
-        if "type" in obj and isinstance(obj["type"], list):
-            # Convert type array to anyOf
-            types = obj["type"]
-            new_obj = {}
-
-            # Copy over non-type-specific properties
-            type_specific_keys = {
-                "properties", "required", "additionalProperties",  # object
-                "items", "minItems", "maxItems", "uniqueItems",  # array
-                "minLength", "maxLength", "pattern", "format",  # string
-                "minimum", "maximum", "exclusiveMinimum", "exclusiveMaximum", "multipleOf"  # number/integer
-            }
-
-            for k, v in obj.items():
-                if k != "type" and k not in type_specific_keys:
-                    new_obj[k] = v
-
-            # Create anyOf array
-            any_of = []
-            for t in types:
-                if t == "null":
-                    # null is a special case - just type
-                    any_of.append({"type": "null"})
-                else:
-                    # For other types, preserve any type-specific constraints
-                    type_schema = {"type": t}
-
-                    # Copy over type-specific constraints
-                    if t == "string":
-                        for k in ["minLength", "maxLength", "pattern", "format"]:
-                            if k in obj:
-                                type_schema[k] = obj[k]
-                    elif t == "number" or t == "integer":
-                        for k in ["minimum", "maximum", "exclusiveMinimum", "exclusiveMaximum", "multipleOf"]:
-                            if k in obj:
-                                type_schema[k] = obj[k]
-                    elif t == "array":
-                        for k in ["items", "minItems", "maxItems", "uniqueItems"]:
-                            if k in obj:
-                                # Recursively process items if present
-                                if k == "items":
-                                    type_schema[k] = _convert_type_arrays_to_anyof(obj[k])
-                                else:
-                                    type_schema[k] = obj[k]
-                    elif t == "object": # pragma: no cover
-                        for k in ["properties", "required", "additionalProperties"]:
-                            if k in obj:
-                                # Recursively process properties if present
-                                if k == "properties":
-                                    type_schema[k] = _convert_type_arrays_to_anyof(obj[k])
-                                else:
-                                    type_schema[k] = obj[k]
-
-                    any_of.append(type_schema)
-
-            new_obj["anyOf"] = any_of
-            return new_obj
-        else:
-            # Recursively process all values
-            return {k: _convert_type_arrays_to_anyof(v) for k, v in obj.items()}
-    elif isinstance(obj, list):
-        # Process each item in the list
-        return [_convert_type_arrays_to_anyof(item) for item in obj]
-    else:
-        # Return primitive values as-is
-        return obj
diff --git a/scripts/reproduce_1383.py b/scripts/reproduce_1383.py
deleted file mode 100644
index af4e36dcf..000000000
--- a/scripts/reproduce_1383.py
+++ /dev/null
@@ -1,83 +0,0 @@
-#!/usr/bin/env python3
-"""
-Reproduction script for issue #1383: Dynamic schema creation fails with nested optionals
-
-This script demonstrates the issue where JSON schemas with union types specified as arrays
-(e.g., {"type": ["string", "null"]}) fail with outlines-core 0.1.26, and shows how the
-Python-side preprocessing resolves the issue.
-"""
-
-import json
-from typing import Optional
-from pydantic import BaseModel
-# TODO: change this once the import issue is fixed in outlines_core
-from outlines_core import outlines_core
-from outlines.types import JsonSchema
-
-# Test case 1: Simple optional field (this should work)
-simple_schema = {
-    "type": "object",
-    "properties": {"name": {"type": "string"}, "age": {"type": ["integer", "null"]}},
-    "required": ["name"],
-}
-
-# Test case 2: Nested optional field (this should fail)
-nested_optional_schema = {
-    "type": "object",
-    "properties": {
-        "person": {
-            "type": ["object", "null"],
-            "properties": {
-                "name": {"type": "string"},
-                "age": {"type": ["integer", "null"]},
-            },
-        }
-    },
-}
-
-# Test case 3: Array with optional items
-array_optional_schema = {
-    "type": "object",
-    "properties": {"items": {"type": "array", "items": {"type": ["string", "null"]}}},
-}
-
-
-def test_schema(schema_dict, name):
-    print(f"\n=== Testing {name} ===")
-    print(f"Schema: {json.dumps(schema_dict, indent=2)}")
-
-    print("\n--- Testing direct build_regex_from_schema ---")
-    try:
-        # Try to create regex from schema
-        schema_str = json.dumps(schema_dict)
-        regex = outlines_core.json_schema.build_regex_from_schema(schema_str)
-        print(f"✓ Success! Generated regex: {regex[:100]}...")
-    except Exception as e:
-        print(f"✗ Error: {type(e).__name__}: {e}")
-
-    print("\n--- Testing JsonSchema class (with workaround) ---")
-    try:
-        # Try to create JsonSchema object
-        json_schema = JsonSchema(schema_dict)
-        print("✓ JsonSchema object created successfully")
-        print(f"  Preprocessed schema: {json_schema.schema[:200]}...")
-
-        # Try to convert to regex
-        from outlines.types.dsl import to_regex
-
-        regex = to_regex(json_schema)
-        print(f"✓ Regex generated successfully: {regex[:100]}...")
-
-    except Exception as e:
-        print(f"✗ Error: {type(e).__name__}: {e}")
-        import traceback
-
-        traceback.print_exc()
-
-
-if __name__ == "__main__":
-    print("Testing issue #1383: Dynamic schema creation fails with nested optionals")
-
-    test_schema(simple_schema, "Simple optional field")
-    test_schema(nested_optional_schema, "Nested optional field")
-    test_schema(array_optional_schema, "Array with optional items")
diff --git a/tests/types/test_json_schema_union_types.py b/tests/types/test_json_schema_union_types.py
deleted file mode 100644
index 681b781f3..000000000
--- a/tests/types/test_json_schema_union_types.py
+++ /dev/null
@@ -1,158 +0,0 @@
-"""Test cases for JSON schemas with union types."""
-
-import json
-
-from outlines.types.json_schema_utils import preprocess_schema_for_union_types
-
-
-class TestJsonSchemaUnionTypes:
-    """Test cases for JSON schemas with union types."""
-
-    def test_invalid_schema(self):
-        """Test invalid schema."""
-        schema = 'foo'
-        result = preprocess_schema_for_union_types(schema)
-        assert result == schema
-
-    def test_simple_union_type(self):
-        """Test simple union type conversion."""
-        schema = {"type": ["string", "null"]}
-        result = json.loads(preprocess_schema_for_union_types(schema))
-
-        assert "anyOf" in result
-        assert len(result["anyOf"]) == 2
-        assert {"type": "string"} in result["anyOf"]
-        assert {"type": "null"} in result["anyOf"]
-
-    def test_union_with_constraints(self):
-        """Test union type with type-specific constraints."""
-        schema = {
-            "type": ["string", "null"],
-            "minLength": 5,
-            "maxLength": 10
-        }
-        result = json.loads(preprocess_schema_for_union_types(schema))
-
-        assert "anyOf" in result
-        assert len(result["anyOf"]) == 2
-        assert {"type": "string", "minLength": 5, "maxLength": 10} in result["anyOf"]
-        assert {"type": "null"} in result["anyOf"]
-
-    def test_nested_union_types(self):
-        """Test nested objects with union types."""
-        schema = {
-            "type": "object",
-            "properties": {
-                "name": {"type": ["string", "null"]},
-                "age": {"type": ["integer", "null"]}
-            }
-        }
-        result = json.loads(preprocess_schema_for_union_types(schema))
-
-        assert result["properties"]["name"]["anyOf"][0] == {"type": "string"}
-        assert result["properties"]["name"]["anyOf"][1] == {"type": "null"}
-        assert result["properties"]["age"]["anyOf"][0] == {"type": "integer"}
-        assert result["properties"]["age"]["anyOf"][1] == {"type": "null"}
-
-    def test_array_with_union_items(self):
-        """Test array with union type items."""
-        schema = {
-            "type": "array",
-            "items": {"type": ["string", "number", "null"]}
-        }
-        result = json.loads(preprocess_schema_for_union_types(schema))
-
-        assert "anyOf" in result["items"]
-        assert len(result["items"]["anyOf"]) == 3
-        assert {"type": "string"} in result["items"]["anyOf"]
-        assert {"type": "number"} in result["items"]["anyOf"]
-        assert {"type": "null"} in result["items"]["anyOf"]
-
-    def test_no_union_types(self):
-        """Test schema without union types remains unchanged."""
-        schema = {
-            "type": "object",
-            "properties": {
-                "name": {"type": "string"},
-                "age": {"type": "integer"}
-            }
-        }
-        result = json.loads(preprocess_schema_for_union_types(schema))
-        assert result == schema
-
-    def test_string_input(self):
-        """Test that string input is properly handled."""
-        schema_str = '{"type": ["string", "null"]}'
-        result = json.loads(preprocess_schema_for_union_types(schema_str))
-
-        assert "anyOf" in result
-        assert len(result["anyOf"]) == 2
-
-    def test_ensure_ascii_parameter(self):
-        """Test that ensure_ascii parameter is respected."""
-        schema = {"type": "string", "pattern": "café"}
-
-        # With ensure_ascii=True (default)
-        result_ascii = preprocess_schema_for_union_types(schema, ensure_ascii=True)
-        assert "caf\\u00e9" in result_ascii
-
-        # With ensure_ascii=False
-        result_unicode = preprocess_schema_for_union_types(schema, ensure_ascii=False)
-        assert "café" in result_unicode
-
-    def test_union_array_type(self):
-        """Test union array type."""
-        schema = {
-            "type": ["array", "null"],
-            "items": {"type": "string"},
-            "minItems": 1,
-        }
-        result = json.loads(preprocess_schema_for_union_types(schema))
-
-        assert "anyOf" in result
-        assert len(result["anyOf"]) == 2
-        assert {"type": "array", "items": {"type": "string"}, "minItems": 1} in result["anyOf"]
-        assert {"type": "null"} in result["anyOf"]
-
-    def test_complex_nested_structure(self):
-        """Test a more complex nested structure with multiple union types."""
-        schema = {
-            "type": "object",
-            "properties": {
-                "id": {"type": "integer"},
-                "data": {
-                    "type": "object",
-                    "properties": {
-                        "value": {"type": ["string", "number", "null"], "minimum": 0},
-                        "metadata": {
-                            "type": ["object", "null"],
-                            "description": "Some metadata",
-                            "additionalProperties": False,
-                            "properties": {
-                                "created": {"type": "string"},
-                                "tags": {
-                                    "type": "array",
-                                    "items": {"type": ["string", "null"]}
-                                }
-                            }
-                        }
-                    }
-                }
-            }
-        }
-        result = json.loads(preprocess_schema_for_union_types(schema))
-
-        # Check value was converted
-        assert "anyOf" in result["properties"]["data"]["properties"]["value"]
-        assert len(result["properties"]["data"]["properties"]["value"]["anyOf"]) == 3
-
-        # Check metadata was converted
-        assert "anyOf" in result["properties"]["data"]["properties"]["metadata"]
-        assert len(result["properties"]["data"]["properties"]["metadata"]["anyOf"]) == 2
-
-        # Check tags items were converted
-        metadata_object = next(
-            item for item in result["properties"]["data"]["properties"]["metadata"]["anyOf"]
-            if item["type"] == "object"
-        )
-        assert "anyOf" in metadata_object["properties"]["tags"]["items"]

From 9c5ef8f56b741cd9818991d961b03fe8e742f975 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 26 Sep 2025 10:55:21 +0200
Subject: [PATCH 474/505] Create the json schema conversion methods

---
 outlines/types/dsl.py                 | 112 +++++++++
 outlines/types/json_schema_utils.py   | 176 +++++++++++++-
 tests/types/test_dsl.py               |  72 ++++--
 tests/types/test_json_schema_utils.py | 320 ++++++++++++++++++++++++++
 4 files changed, 660 insertions(+), 20 deletions(-)
 create mode 100644 tests/types/test_json_schema_utils.py

diff --git a/outlines/types/dsl.py b/outlines/types/dsl.py
index ce992f9e2..5081a6cba 100644
--- a/outlines/types/dsl.py
+++ b/outlines/types/dsl.py
@@ -22,6 +22,7 @@
 from typing import (
     Any,
     List,
+    Literal,
     Optional as OptionalType,
     Union,
     get_args,
@@ -41,6 +42,11 @@
 
 import outlines.types as types
 from outlines import grammars
+from outlines.types.json_schema_utils import (
+    json_schema_dict_to_pydantic,
+    json_schema_dict_to_typeddict,
+    json_schema_dict_to_dataclass,
+)
 from outlines.types.utils import (
     get_schema_from_signature,
     is_int,
@@ -282,6 +288,9 @@ class JsonSchema(Term):
     genSON schema builder.
 
     """
+    schema: str
+    whitespace_pattern: OptionalType[str]
+
     def __init__(
         self,
         schema: Union[
@@ -329,6 +338,109 @@ def __init__(
     def __post_init__(self):
         jsonschema.Draft7Validator.check_schema(json.loads(self.schema))
 
+    @classmethod
+    def is_json_schema(cls, obj: Any) -> bool:
+        """Check if the object provided is a JSON schema type.
+
+        Parameters
+        ----------
+        obj: Any
+            The object to check
+
+        Returns
+        -------
+        bool
+            True if the object is a JSON schema type, False otherwise
+
+        """
+        return (
+            isinstance(obj, cls)
+            or is_pydantic_model(obj)
+            or is_typed_dict(obj)
+            or is_dataclass(obj)
+            or is_genson_schema_builder(obj)
+        )
+
+    @classmethod
+    def convert_to(
+        cls,
+        schema: Union[
+            "JsonSchema",
+            type[BaseModel],
+            _TypedDictMeta,
+            type,
+            SchemaBuilder,
+        ],
+        target_types: List[Literal[
+            "str",
+            "dict",
+            "pydantic",
+            "typeddict",
+            "dataclass",
+            "genson",
+        ]],
+    ) -> Union[str, dict, type[BaseModel], _TypedDictMeta, type, SchemaBuilder]:
+        """Convert a JSON schema type to a different JSON schema type.
+
+        If the schema provided is already of a type in the target_types, return
+        it unchanged.
+
+        Parameters
+        ----------
+        schema: Union[JsonSchema, type[BaseModel], _TypedDictMeta, type, SchemaBuilder]
+            The schema to convert
+        target_types: List[Literal["str", "dict", "pydantic", "typeddict", "dataclass", "genson"]]
+            The target types to convert to
+
+        """
+        # If the schema provided is already of a type in the target_types,
+        # just return it
+        if isinstance(schema, cls):
+            if "str" in target_types:
+                return schema.schema
+            elif "dict" in target_types:
+                return json.loads(schema.schema)
+        elif is_pydantic_model(schema) and "pydantic" in target_types:
+            return schema
+        elif is_typed_dict(schema) and "typeddict" in target_types:
+            return schema
+        elif is_dataclass(schema) and "dataclass" in target_types:
+            return schema
+        elif is_genson_schema_builder(schema) and "genson" in target_types:
+            return schema
+
+        # Convert the schema to a JSON schema string/dict
+        if isinstance(schema, cls):
+            schema_str = schema.schema
+        else:
+            schema_str = cls(schema).schema
+        schema_dict = json.loads(schema_str)
+
+        for target_type in target_types:
+            try:
+                # Convert the JSON schema string to the target type
+                if target_type == "str":
+                    return schema_str
+                elif target_type == "dict":
+                    return schema_dict
+                elif target_type == "pydantic":
+                    return json_schema_dict_to_pydantic(schema_dict)
+                elif target_type == "typeddict":
+                    return json_schema_dict_to_typeddict(schema_dict)
+                elif target_type == "dataclass":
+                    return json_schema_dict_to_dataclass(schema_dict)
+                # No conversion available for genson
+            except Exception as e:  # pragma: no cover
+                warnings.warn(
+                    f"Cannot convert schema type {type(schema)} to {target_type}: {e}"
+                )
+                continue
+
+        raise ValueError(
+            f"Cannot convert schema type {type(schema)} to any of the target "
+            f"types {target_types}"
+        )
+
     def _display_node(self) -> str:
         return f"JsonSchema('{self.schema}')"
 
diff --git a/outlines/types/json_schema_utils.py b/outlines/types/json_schema_utils.py
index 6f0a29967..4e898d727 100644
--- a/outlines/types/json_schema_utils.py
+++ b/outlines/types/json_schema_utils.py
@@ -1 +1,175 @@
-"""Utilities for handling JSON schema compatibility."""
+"""Convert JSON Schema dicts to Python types."""
+
+import sys
+from dataclasses import dataclass, field
+from typing import Any, Dict, List, Literal, Optional
+
+from pydantic import BaseModel, create_model
+
+if sys.version_info >= (3, 12):  # pragma: no cover
+    from typing import _TypedDictMeta, TypedDict  # type: ignore
+else:  # pragma: no cover
+    from typing_extensions import _TypedDictMeta, TypedDict  # type: ignore
+
+
+def schema_type_to_python(
+    schema: dict,
+    caller_target_type: Literal["pydantic", "typeddict", "dataclass"]
+) -> Any:
+    """Get a Python type from a JSON Schema dict.
+
+    Parameters
+    ----------
+    schema: dict
+        The JSON Schema dict to convert to a Python type
+    caller_target_type: Literal["pydantic", "typeddict", "dataclass"]
+        The type of the caller
+
+    Returns
+    -------
+    Any
+        The Python type
+
+    """
+    if "enum" in schema:
+        values = schema["enum"]
+        return Literal[tuple(values)]
+
+    t = schema.get("type")
+
+    if t == "string":
+        return str
+    elif t == "integer":
+        return int
+    elif t == "number":
+        return float
+    elif t == "boolean":
+        return bool
+    elif t == "array":
+        items = schema.get("items", {})
+        if items:
+            item_type = schema_type_to_python(items, caller_target_type)
+        else:
+            item_type = Any
+        return List[item_type]  # type: ignore
+    elif t == "object":
+        name = schema.get("title")
+        if caller_target_type == "pydantic":
+            return json_schema_dict_to_pydantic(schema, name)
+        elif caller_target_type == "typeddict":
+            return json_schema_dict_to_typeddict(schema, name)
+        elif caller_target_type == "dataclass":
+            return json_schema_dict_to_dataclass(schema, name)
+
+    return Any
+
+
+def json_schema_dict_to_typeddict(
+    schema: dict,
+    name: Optional[str] = None
+) -> _TypedDictMeta:
+    """Convert a JSON Schema dict into a TypedDict class.
+
+    Parameters
+    ----------
+    schema: dict
+        The JSON Schema dict to convert to a TypedDict
+    name: Optional[str]
+        The name of the TypedDict
+
+    Returns
+    -------
+    _TypedDictMeta
+        The TypedDict class
+
+    """
+    required = set(schema.get("required", []))
+    properties = schema.get("properties", {})
+
+    annotations: Dict[str, Any] = {}
+
+    for property, details in properties.items():
+        typ = schema_type_to_python(details, "typeddict")
+        if property not in required:
+            typ = Optional[typ]
+        annotations[property] = typ
+
+    return TypedDict(name or "AnonymousTypedDict", annotations)  # type: ignore
+
+
+def json_schema_dict_to_pydantic(
+    schema: dict,
+    name: Optional[str] = None
+) -> type[BaseModel]:
+    """Convert a JSON Schema dict into a Pydantic BaseModel class.
+
+    Parameters
+    ----------
+    schema: dict
+        The JSON Schema dict to convert to a Pydantic BaseModel
+    name: Optional[str]
+        The name of the Pydantic BaseModel
+
+    Returns
+    -------
+    type[BaseModel]
+        The Pydantic BaseModel class
+
+    """
+    required = set(schema.get("required", []))
+    properties = schema.get("properties", {})
+
+    field_definitions: Dict[str, Any] = {}
+
+    for property, details in properties.items():
+        typ = schema_type_to_python(details, "pydantic")
+        if property not in required:
+            field_definitions[property] = (Optional[typ], None)
+        else:
+            field_definitions[property] = (typ, ...)
+
+    return create_model(name or "AnonymousPydanticModel", **field_definitions)
+
+
+def json_schema_dict_to_dataclass(
+    schema: dict,
+    name: Optional[str] = None
+) -> type:
+    """Convert a JSON Schema dict into a dataclass.
+
+    Parameters
+    ----------
+    schema: dict
+        The JSON Schema dict to convert to a dataclass
+    name: Optional[str]
+        The name of the dataclass
+
+    Returns
+    -------
+    type
+        The dataclass
+
+    """
+    required = set(schema.get("required", []))
+    properties = schema.get("properties", {})
+
+    annotations: Dict[str, Any] = {}
+    defaults: Dict[str, Any] = {}
+
+    for property, details in properties.items():
+        typ = schema_type_to_python(details, "dataclass")
+        annotations[property] = typ
+
+        if property not in required:
+            defaults[property] = None
+
+    class_dict = {
+        '__annotations__': annotations,
+        '__module__': __name__,
+    }
+
+    for property, default_val in defaults.items():
+        class_dict[property] = field(default=default_val)
+
+    cls = type(name or "AnonymousDataclass", (), class_dict)
+    return dataclass(cls)
diff --git a/tests/types/test_dsl.py b/tests/types/test_dsl.py
index e4175e221..a4c1593ea 100644
--- a/tests/types/test_dsl.py
+++ b/tests/types/test_dsl.py
@@ -5,8 +5,6 @@
 from dataclasses import dataclass
 from enum import Enum
 from typing import (
-    Dict,
-    List,
     Literal,
     Tuple,
     Union,
@@ -53,6 +51,11 @@
     python_types_to_terms,
     to_regex,
 )
+from outlines.types.utils import (
+    is_pydantic_model,
+    is_typed_dict,
+    is_dataclass,
+)
 
 if sys.version_info >= (3, 12):
     from typing import TypedDict
@@ -401,6 +404,21 @@ def test_cfg():
 
 
 def test_json_schema():
+    # variables to be used in the tests
+    json_schema = types.json_schema('{"type": "object", "properties": {"foo": {"type": "string"}, "bar": {"type": "integer"}}, "required": ["foo"]}')
+    schema_builder_instance = SchemaBuilder()
+    schema_builder_instance.add_schema({"type": "object", "properties": {"foo": {"type": "string"}, "bar": {"type": "integer"}}, "required": ["foo"]})
+    class MyPydanticModel(BaseModel):
+        foo: str
+        bar: PyOptional[int] = None
+    class MyTypedDict(TypedDict):
+        foo: str
+        bar: int
+    @dataclass
+    class MyDataClass:
+        foo: str
+        bar: PyOptional[int] = None
+
     # init dict
     schema = types.json_schema({"type": "string"})
     assert schema.schema == '{"type": "string"}'
@@ -410,34 +428,50 @@ def test_json_schema():
     assert schema.schema == '{"type": "string"}'
 
     # init Pydantic model
-    class PydanticModel(BaseModel):
-        foo: str
-    schema = types.json_schema(PydanticModel)
-    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "PydanticModel", "type": "object"}'
+    schema = types.json_schema(MyPydanticModel)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}, "bar": {"anyOf": [{"type": "integer"}, {"type": "null"}], "default": null, "title": "Bar"}}, "required": ["foo"], "title": "MyPydanticModel", "type": "object"}'
 
     # init TypedDict
-    class SomeTypedDict(TypedDict):
-        foo: str
-    schema = types.json_schema(SomeTypedDict)
-    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "SomeTypedDict", "type": "object"}'
+    schema = types.json_schema(MyTypedDict)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}, "bar": {"title": "Bar", "type": "integer"}}, "required": ["foo", "bar"], "title": "MyTypedDict", "type": "object"}'
 
     # init dataclass
-    @dataclass
-    class DataClass:
-        foo: str
-    schema = types.json_schema(DataClass)
-    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}}, "required": ["foo"], "title": "DataClass", "type": "object"}'
+    schema = types.json_schema(MyDataClass)
+    assert schema.schema == '{"properties": {"foo": {"title": "Foo", "type": "string"}, "bar": {"anyOf": [{"type": "integer"}, {"type": "null"}], "default": null, "title": "Bar"}}, "required": ["foo"], "title": "MyDataClass", "type": "object"}'
 
     # init SchemaBuilder
-    builder = SchemaBuilder()
-    builder.add_schema({"type": "object", "properties": {}})
-    schema = types.json_schema(builder)
-    assert schema.schema == '{"$schema": "http://json-schema.org/schema#", "type": "object"}'
+    schema = types.json_schema(schema_builder_instance)
+    assert schema.schema == '{"$schema": "http://json-schema.org/schema#", "type": "object", "properties": {"foo": {"type": "string"}, "bar": {"type": "integer"}}, "required": ["foo"]}'
 
     # init unsupported type
     with pytest.raises(ValueError, match="Cannot parse schema"):
         types.json_schema(1)
 
+    # is_json_schema
+    assert not JsonSchema.is_json_schema(None)
+    assert not JsonSchema.is_json_schema('{"type": "string"}')
+    assert not JsonSchema.is_json_schema({"type": "string"})
+    assert JsonSchema.is_json_schema(json_schema)
+    assert JsonSchema.is_json_schema(schema_builder_instance)
+    assert JsonSchema.is_json_schema(MyPydanticModel)
+    assert JsonSchema.is_json_schema(MyTypedDict)
+    assert JsonSchema.is_json_schema(MyDataClass)
+
+    # convert_to
+    assert JsonSchema.convert_to(json_schema, ["str"]) == json_schema.schema
+    assert JsonSchema.convert_to(json_schema, ["dict"]) == json.loads(json_schema.schema)
+    assert JsonSchema.convert_to(MyPydanticModel, ["pydantic"]) == MyPydanticModel
+    assert JsonSchema.convert_to(MyTypedDict, ["typeddict"]) == MyTypedDict
+    assert JsonSchema.convert_to(MyDataClass, ["dataclass"]) == MyDataClass
+    assert JsonSchema.convert_to(schema_builder_instance, ["genson"]) == schema_builder_instance
+    assert JsonSchema.convert_to(MyPydanticModel, ["str"]) == JsonSchema(MyPydanticModel).schema
+    assert JsonSchema.convert_to(MyPydanticModel, ["dict"]) == json.loads(JsonSchema(MyPydanticModel).schema)
+    assert is_pydantic_model(JsonSchema.convert_to(json_schema, ["pydantic"]))
+    assert is_typed_dict(JsonSchema.convert_to(json_schema, ["typeddict"]))
+    assert is_dataclass(JsonSchema.convert_to(json_schema, ["dataclass"]))
+    with pytest.raises(ValueError, match="Cannot convert schema type"):
+        JsonSchema.convert_to(json_schema, ["genson"])
+
     # other methods
     schema = types.json_schema('{"type": "string"}')
     schema.__post_init__()
diff --git a/tests/types/test_json_schema_utils.py b/tests/types/test_json_schema_utils.py
new file mode 100644
index 000000000..2985b5bc4
--- /dev/null
+++ b/tests/types/test_json_schema_utils.py
@@ -0,0 +1,320 @@
+import sys
+from dataclasses import is_dataclass
+from typing import Any, List, Literal, Optional
+
+from pydantic import BaseModel, TypeAdapter
+from pydantic_core import PydanticUndefined
+
+from outlines.types.json_schema_utils import (
+    schema_type_to_python,
+    json_schema_dict_to_typeddict,
+    json_schema_dict_to_pydantic,
+    json_schema_dict_to_dataclass
+)
+
+if sys.version_info >= (3, 12):
+    from typing import _TypedDictMeta  # type: ignore
+else:
+    from typing_extensions import _TypedDictMeta  # type: ignore
+
+
+def test_schema_type_to_python_simple_types():
+    assert schema_type_to_python({"type": "string"}, "pydantic") is str
+    assert schema_type_to_python({"type": "integer"}, "pydantic") is int
+    assert schema_type_to_python({"type": "number"}, "pydantic") is float
+    assert schema_type_to_python({"type": "boolean"}, "pydantic") is bool
+    assert schema_type_to_python({"type": "object"}, "foo") is Any
+    assert schema_type_to_python({}, "pydantic") is Any
+
+
+def test_schema_type_to_python_enum():
+    schema = {"enum": ["red", "green", "blue"]}
+    result = schema_type_to_python(schema, "pydantic")
+    assert result == Literal[("red", "green", "blue")]
+
+
+def test_schema_type_to_python_array():
+    # String items
+    schema = {"type": "array", "items": {"type": "string"}}
+    result = schema_type_to_python(schema, "pydantic")
+    assert result == List[str]
+
+    # Integer items
+    schema = {"type": "array", "items": {"type": "integer"}}
+    result = schema_type_to_python(schema, "pydantic")
+    assert result == List[int]
+
+    # Without items specification
+    schema = {"type": "array"}
+    result = schema_type_to_python(schema, "pydantic")
+    assert result == List[Any]
+
+
+def test_schema_type_to_python_object():
+    schema = {
+        "type": "object",
+        "title": "TestObject",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"}
+        },
+        "required": ["name"]
+    }
+
+    # Pydantic caller
+    pydantic_result = schema_type_to_python(schema, "pydantic")
+    assert issubclass(pydantic_result, BaseModel)
+    assert pydantic_result.__name__ == "TestObject"
+    assert pydantic_result.model_fields["name"].annotation is str
+    assert pydantic_result.model_fields["age"].annotation == Optional[int]
+
+    # Typeddict caller
+    typeddict_result = schema_type_to_python(schema, "typeddict")
+    assert isinstance(typeddict_result, _TypedDictMeta)
+    assert typeddict_result.__name__ == "TestObject"
+    assert typeddict_result.__annotations__["name"] is str
+    assert typeddict_result.__annotations__["age"] == Optional[int]
+
+    # Dataclass caller
+    dataclass_result = schema_type_to_python(schema, "dataclass")
+    print(TypeAdapter(dataclass_result).json_schema())
+    assert hasattr(dataclass_result, "__dataclass_fields__")
+    assert dataclass_result.__annotations__["name"] is str
+    assert not hasattr(dataclass_result, "name")
+    assert dataclass_result.__annotations__["age"] is int
+    assert dataclass_result.age is None
+
+
+def test_schema_type_to_python_unknown_type():
+    # Unknown type
+    schema = {"type": "unknown"}
+    result = schema_type_to_python(schema, "pydantic")
+    assert result == Any
+
+    # Schema without type
+    schema = {}
+    result = schema_type_to_python(schema, "pydantic")
+    assert result == Any
+
+
+def test_json_schema_dict_to_typeddict_basic():
+    schema = {
+        "type": "object",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"}
+        },
+        "required": ["name"]
+    }
+
+    result = json_schema_dict_to_typeddict(schema, "Person")
+    assert isinstance(result, _TypedDictMeta)
+    assert result.__name__ == "Person"
+
+    annotations = result.__annotations__
+    assert annotations["name"] is str
+    assert annotations["age"] == Optional[int]
+
+
+def test_json_schema_dict_to_typeddict_array_enum():
+    schema = {
+        "type": "object",
+        "properties": {
+            "tags": {
+                "type": "array",
+                "items": {"type": "string"}
+            },
+            "preferences": {
+                "enum": ["light", "dark"]
+            }
+        },
+        "required": ["tags"]
+    }
+
+    result = json_schema_dict_to_typeddict(schema)
+    assert isinstance(result, _TypedDictMeta)
+    assert result.__name__ == "AnonymousTypedDict"
+
+    annotations = result.__annotations__
+    assert annotations["tags"] == List[str]
+    assert annotations["preferences"] == Optional[Literal[("light", "dark")]]
+
+
+def test_json_schema_dict_to_typeddict_nested_object():
+    schema = {
+        "type": "object",
+        "properties": {
+            "field": {
+                "type": "object",
+                "properties": {
+                    "name": {"type": "string"},
+                    "age": {"type": "integer"}
+                },
+                "required": ["name"]
+            }
+        },
+        "required": ["field"]
+    }
+
+    result = json_schema_dict_to_typeddict(schema)
+    assert isinstance(result, _TypedDictMeta)
+    assert result.__name__ == "AnonymousTypedDict"
+
+    annotations = result.__annotations__
+    assert isinstance(annotations["field"], _TypedDictMeta)
+    assert annotations["field"].__name__ == "AnonymousTypedDict"
+    assert annotations["field"].__annotations__["name"] is str
+    assert annotations["field"].__annotations__["age"] == Optional[int]
+
+
+def test_json_schema_dict_to_pydantic_basic():
+    schema = {
+        "type": "object",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"}
+        },
+        "required": ["name"]
+    }
+
+    result = json_schema_dict_to_pydantic(schema, "Person")
+    assert issubclass(result, BaseModel)
+    assert result.__name__ == "Person"
+
+    assert result.model_fields["name"].annotation is str
+    assert result.model_fields["age"].annotation == Optional[int]
+    assert result.model_fields["name"].default == PydanticUndefined
+    result.model_fields["age"].default is None
+
+
+def test_json_schema_dict_to_pydantic_array_enum():
+    schema = {
+        "type": "object",
+        "properties": {
+            "tags": {
+                "type": "array",
+                "items": {"type": "string"}
+            },
+            "status": {
+                "enum": ["active", "inactive", "pending"]
+            },
+        },
+        "required": ["status"]
+    }
+
+    result = json_schema_dict_to_pydantic(schema)
+    assert issubclass(result, BaseModel)
+    assert result.__name__ == "AnonymousPydanticModel"
+
+    assert result.model_fields["tags"].annotation == Optional[List[str]]
+    assert result.model_fields["status"].annotation == Literal[("active", "inactive", "pending")]
+    assert result.model_fields["tags"].default is None
+    assert result.model_fields["status"].default == PydanticUndefined
+
+
+def test_json_schema_dict_to_pydantic_nested_object():
+    schema = {
+        "type": "object",
+        "properties": {
+            "field": {
+                "type": "object",
+                "properties": {
+                    "name": {"type": "string"},
+                    "age": {"type": "integer"}
+                },
+                "required": ["name"]
+            }
+        },
+        "required": ["field"]
+    }
+
+    result = json_schema_dict_to_pydantic(schema)
+    assert issubclass(result, BaseModel)
+    assert result.__name__ == "AnonymousPydanticModel"
+
+    assert issubclass(result.model_fields["field"].annotation, BaseModel)
+    assert result.model_fields["field"].annotation.__name__ == "AnonymousPydanticModel"
+
+    field = result.model_fields["field"].annotation
+    assert field.model_fields["name"].annotation is str
+    assert field.model_fields["age"].annotation == Optional[int]
+    assert field.model_fields["name"].default == PydanticUndefined
+    assert field.model_fields["age"].default is None
+
+
+def test_json_schema_dict_to_dataclass_basic():
+    schema = {
+        "type": "object",
+        "properties": {
+            "name": {"type": "string"},
+            "age": {"type": "integer"}
+        },
+        "required": ["name"]
+    }
+
+    result = json_schema_dict_to_dataclass(schema, "Person")
+    assert is_dataclass(result)
+    assert result.__name__ == "Person"
+
+    annotations = result.__annotations__
+    assert annotations["name"] is str
+    assert annotations["age"] is int
+    assert not hasattr(result, "name")
+    assert result.age is None
+
+
+def test_json_schema_dict_to_dataclass_array_enum():
+    schema = {
+        "type": "object",
+        "properties": {
+            "status": {
+                "enum": ["active", "inactive", "pending"]
+            },
+            "tags": {
+                "type": "array",
+                "items": {"type": "string"}
+            },
+        },
+        "required": ["status"]
+    }
+
+    result = json_schema_dict_to_dataclass(schema)
+    assert is_dataclass(result)
+    assert result.__name__ == "AnonymousDataclass"
+
+    annotations = result.__annotations__
+    assert annotations["tags"] == List[str]
+    assert annotations["status"] == Literal[("active", "inactive", "pending")]
+    assert not hasattr(result, "status")
+    assert result.tags is None
+
+
+def test_json_schema_dict_to_dataclass_nested_object():
+    schema = {
+        "type": "object",
+        "properties": {
+            "field": {
+                "type": "object",
+                "properties": {
+                    "name": {"type": "string"},
+                    "age": {"type": "integer"}
+                },
+                "required": ["name"]
+            }
+        },
+        "required": ["field"]
+    }
+
+    result = json_schema_dict_to_dataclass(schema)
+    assert is_dataclass(result)
+    assert result.__name__ == "AnonymousDataclass"
+
+    annotations = result.__annotations__
+    assert is_dataclass(annotations["field"])
+    assert annotations["field"].__name__ == "AnonymousDataclass"
+
+    field = annotations["field"]
+    assert field.__annotations__["name"] is str
+    assert field.__annotations__["age"] is int
+    assert not hasattr(field, "name")
+    assert field.age is None

From 2c8c9d7924f5cb2e664515dde2d4c85d0b85a56d Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 26 Sep 2025 10:55:29 +0200
Subject: [PATCH 475/505] Use the JsonSchema conversion methods in models

---
 outlines/models/dottxt.py                | 27 ++---------
 outlines/models/gemini.py                | 48 +++++++-------------
 outlines/models/ollama.py                | 43 ++++++------------
 outlines/models/openai.py                | 30 +++---------
 tests/models/test_gemini_type_adapter.py | 58 ++++++++++++++++++------
 tests/models/test_ollama_type_adapter.py |  6 +--
 6 files changed, 89 insertions(+), 123 deletions(-)

diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index f88d3d5d6..f0eef2458 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -1,18 +1,9 @@
 """Integration with Dottxt's API."""
 
-import json
-from typing import TYPE_CHECKING, Any, Optional
-
-from pydantic import TypeAdapter
+from typing import TYPE_CHECKING, Any, Optional, cast
 
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.types import CFG, JsonSchema, Regex
-from outlines.types.utils import (
-    is_dataclass,
-    is_genson_schema_builder,
-    is_pydantic_model,
-    is_typed_dict,
-)
 
 if TYPE_CHECKING:
     from dottxt import Dottxt as DottxtClient
@@ -77,20 +68,8 @@ def format_output_type(self, output_type: Optional[Any] = None) -> str:
                 "CFG-based structured outputs will soon be available with "
                 "Dottxt. Use an open source model in the meantime."
             )
-
-        elif isinstance(output_type, JsonSchema):
-            return output_type.schema
-        elif is_dataclass(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return json.dumps(schema)
-        elif is_typed_dict(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return json.dumps(schema)
-        elif is_pydantic_model(output_type):
-            schema = output_type.model_json_schema()
-            return json.dumps(schema)
-        elif is_genson_schema_builder(output_type):
-            return output_type.to_json()
+        elif JsonSchema.is_json_schema(output_type):
+            return cast(str, JsonSchema.convert_to(output_type, ["str"]))
         else:
             type_name = getattr(output_type, "__name__", output_type)
             raise TypeError(
diff --git a/outlines/models/gemini.py b/outlines/models/gemini.py
index 742dfa7a1..d8ccf272f 100644
--- a/outlines/models/gemini.py
+++ b/outlines/models/gemini.py
@@ -14,14 +14,11 @@
 from outlines.models.base import Model, ModelTypeAdapter
 from outlines.types import CFG, Choice, JsonSchema, Regex
 from outlines.types.utils import (
-    is_dataclass,
     is_enum,
     get_enum_from_choice,
     get_enum_from_literal,
     is_genson_schema_builder,
     is_literal,
-    is_pydantic_model,
-    is_typed_dict,
     is_typing_list,
 )
 
@@ -171,28 +168,18 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
                 "CFG-based structured outputs are not available with Gemini. "
                 "Use an open source model or dottxt instead."
             )
-        elif is_genson_schema_builder(output_type):
-            raise TypeError(
-                "The Gemini SDK does not accept Genson schema builders as an "
-                "input. Pass a Pydantic model, typed dict or dataclass "
-                "instead."
-            )
-        elif isinstance(output_type, JsonSchema):
-            raise TypeError(
-                "The Gemini SDK does not accept Json Schemas as an input. "
-                "Pass a Pydantic model, typed dict or dataclass instead."
-            )
 
         if output_type is None:
             return {}
 
-        # Structured types
-        elif is_dataclass(output_type):
-            return self.format_json_output_type(output_type)
-        elif is_typed_dict(output_type):
-            return self.format_json_output_type(output_type)
-        elif is_pydantic_model(output_type):
-            return self.format_json_output_type(output_type)
+        # JSON schema types
+        elif JsonSchema.is_json_schema(output_type):
+            return self.format_json_output_type(
+                JsonSchema.convert_to(
+                    output_type,
+                    ["dataclass", "typeddict", "pydantic"]
+                )
+            )
 
         # List of structured types
         elif is_typing_list(output_type):
@@ -233,21 +220,20 @@ def format_list_output_type(self, output_type: Optional[Any]) -> dict:
         if len(args) == 1:
             item_type = args[0]
 
-            # Check if list item type is supported
-            if (
-                is_pydantic_model(item_type)
-                or is_typed_dict(item_type)
-                or is_dataclass(item_type)
-            ):
+            if JsonSchema.is_json_schema(item_type):
                 return {
                     "response_mime_type": "application/json",
-                    "response_schema": output_type,
+                    "response_schema": list[  # type: ignore
+                        JsonSchema.convert_to(
+                            item_type,
+                            ["dataclass", "typeddict", "pydantic"]
+                        )
+                    ],
                 }
-
             else:
                 raise TypeError(
-                    "The only supported types for list items are Pydantic "
-                    + "models, typed dicts and dataclasses."
+                    "The list items output type must contain a JSON schema "
+                    "type."
                 )
 
         raise TypeError(
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 24ea77eac..cc693854d 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -1,20 +1,19 @@
 """Integration with the `ollama` library."""
 
-import json
 from functools import singledispatchmethod
-from typing import TYPE_CHECKING, Any, AsyncIterator, Iterator, Optional, Union
-
-from pydantic import TypeAdapter
+from typing import (
+    TYPE_CHECKING,
+    Any,
+    AsyncIterator,
+    Iterator,
+    Optional,
+    Union,
+    cast,
+)
 
 from outlines.inputs import Chat, Image
 from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.types import CFG, JsonSchema, Regex
-from outlines.types.utils import (
-    is_dataclass,
-    is_genson_schema_builder,
-    is_pydantic_model,
-    is_typed_dict,
-)
 
 if TYPE_CHECKING:
     from ollama import Client
@@ -109,7 +108,7 @@ def _create_message(self, role: str, content: str | list) -> dict:
 
     def format_output_type(
         self, output_type: Optional[Any] = None
-    ) -> Optional[str]:
+    ) -> Optional[dict]:
         """Format the output type to pass to the client.
 
         TODO: `int`, `float` and other Python types could be supported via
@@ -126,7 +125,9 @@ def format_output_type(
             The formatted output type to be passed to the model.
 
         """
-        if isinstance(output_type, Regex):
+        if output_type is None:
+            return None
+        elif isinstance(output_type, Regex):
             raise TypeError(
                 "Regex-based structured outputs are not supported by Ollama. "
                 "Use an open source model in the meantime."
@@ -136,22 +137,8 @@ def format_output_type(
                 "CFG-based structured outputs are not supported by Ollama. "
                 "Use an open source model in the meantime."
             )
-
-        if output_type is None:
-            return None
-        elif isinstance(output_type, JsonSchema):
-            return json.loads(output_type.schema)
-        elif is_dataclass(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return schema
-        elif is_typed_dict(output_type):
-            schema = TypeAdapter(output_type).json_schema()
-            return schema
-        elif is_pydantic_model(output_type):
-            schema = output_type.model_json_schema()
-            return schema
-        elif is_genson_schema_builder(output_type):
-            return output_type.to_json()
+        elif JsonSchema.is_json_schema(output_type):
+            return cast(dict, JsonSchema.convert_to(output_type, ["dict"]))
         else:
             type_name = getattr(output_type, "__name__", output_type)
             raise TypeError(
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index dad987b93..4626bffa8 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -1,6 +1,5 @@
 """Integration with OpenAI's API."""
 
-import json
 from typing import (
     TYPE_CHECKING,
     Any,
@@ -8,22 +7,17 @@
     Iterator,
     Optional,
     Union,
+    cast,
 )
 from functools import singledispatchmethod
 
-from pydantic import BaseModel, TypeAdapter
+from pydantic import BaseModel
 
 from outlines.inputs import Chat, Image
 from outlines.models.base import AsyncModel, Model, ModelTypeAdapter
 from outlines.models.utils import set_additional_properties_false_json_schema
 from outlines.types import JsonSchema, Regex, CFG
-from outlines.types.utils import (
-    is_dataclass,
-    is_typed_dict,
-    is_pydantic_model,
-    is_genson_schema_builder,
-    is_native_dict
-)
+from outlines.types.utils import is_native_dict
 
 if TYPE_CHECKING:
     from openai import (
@@ -176,20 +170,10 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
             return {}
         elif is_native_dict(output_type):
             return self.format_json_mode_type()
-        elif is_dataclass(output_type):
-            output_type = TypeAdapter(output_type).json_schema()
-            return self.format_json_output_type(output_type)
-        elif is_typed_dict(output_type):
-            output_type = TypeAdapter(output_type).json_schema()
-            return self.format_json_output_type(output_type)
-        elif is_pydantic_model(output_type):
-            output_type = output_type.model_json_schema()
-            return self.format_json_output_type(output_type)
-        elif is_genson_schema_builder(output_type):
-            schema = json.loads(output_type.to_json())
-            return self.format_json_output_type(schema)
-        elif isinstance(output_type, JsonSchema):
-            return self.format_json_output_type(json.loads(output_type.schema))
+        elif JsonSchema.is_json_schema(output_type):
+            return self.format_json_output_type(
+                cast(dict, JsonSchema.convert_to(output_type, ["dict"]))
+            )
         else:
             type_name = getattr(output_type, "__name__", output_type)
             raise TypeError(
diff --git a/tests/models/test_gemini_type_adapter.py b/tests/models/test_gemini_type_adapter.py
index 0f1985991..897cfd726 100644
--- a/tests/models/test_gemini_type_adapter.py
+++ b/tests/models/test_gemini_type_adapter.py
@@ -3,16 +3,16 @@
 import sys
 from dataclasses import dataclass
 from enum import Enum, EnumMeta
-from typing import Literal
+from typing import Literal, get_args
 
 from PIL import Image as PILImage
 from genson import SchemaBuilder
-from google.genai import types
 from pydantic import BaseModel
 
 from outlines import cfg, json_schema, regex
 from outlines.inputs import Chat, Image
 from outlines.models.gemini import GeminiTypeAdapter
+from outlines.types.utils import is_dataclass
 
 if sys.version_info >= (3, 12):
     from typing import TypedDict
@@ -135,19 +135,29 @@ def test_gemini_type_adapter_output_invalid(adapter):
     with pytest.raises(TypeError, match="CFG-based structured outputs"):
         adapter.format_output_type(cfg(""))
 
-    with pytest.raises(TypeError, match="The Gemini SDK does not accept"):
-        adapter.format_output_type(SchemaBuilder())
-
-    with pytest.raises(TypeError, match="The Gemini SDK does not"):
-        adapter.format_output_type(json_schema(""))
-
 
 def test_gemini_type_adapter_output_none(adapter):
     result = adapter.format_output_type(None)
     assert result == {}
 
 
-def test_gemini_type_adapter_output_dataclass(adapter, schema):
+def test_gemini_type_adapter_output_json_schema(adapter, schema):
+    result = adapter.format_output_type(json_schema(schema))
+    assert isinstance(result, dict)
+    assert result["response_mime_type"] == "application/json"
+    assert is_dataclass(result["response_schema"])
+
+
+def test_gemini_type_adapter_output_list_json_schema(adapter, schema):
+    result = adapter.format_output_type(list[json_schema(schema)])
+    assert isinstance(result, dict)
+    assert result["response_mime_type"] == "application/json"
+    args = get_args(result["response_schema"])
+    assert len(args) == 1
+    assert is_dataclass(args[0])
+
+
+def test_gemini_type_adapter_output_dataclass(adapter):
     @dataclass
     class User:
         user_id: int
@@ -160,7 +170,7 @@ class User:
     }
 
 
-def test_gemini_type_adapter_output_list_dataclass(adapter, schema):
+def test_gemini_type_adapter_output_list_dataclass(adapter):
     class User(BaseModel):
         user_id: int
         name: str
@@ -172,7 +182,7 @@ class User(BaseModel):
     }
 
 
-def test_gemini_type_adapter_output_typed_dict(adapter, schema):
+def test_gemini_type_adapter_output_typed_dict(adapter):
     class User(TypedDict):
         user_id: int
         name: str
@@ -184,7 +194,7 @@ class User(TypedDict):
     }
 
 
-def test_gemini_type_adapter_output_list_typed_dict(adapter, schema):
+def test_gemini_type_adapter_output_list_typed_dict(adapter):
     class User(BaseModel):
         user_id: int
         name: str
@@ -196,7 +206,7 @@ class User(BaseModel):
     }
 
 
-def test_gemini_type_adapter_output_pydantic(adapter, schema):
+def test_gemini_type_adapter_output_pydantic(adapter):
     class User(BaseModel):
         user_id: int
         name: str
@@ -208,7 +218,7 @@ class User(BaseModel):
     }
 
 
-def test_gemini_type_adapter_output_list_pydantic(adapter, schema):
+def test_gemini_type_adapter_output_list_pydantic(adapter):
     class User(BaseModel):
         user_id: int
         name: str
@@ -220,6 +230,26 @@ class User(BaseModel):
     }
 
 
+def test_gemini_type_adapter_output_genson_schema_builder(adapter):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {"foo": {"type": "string"}, "bar": {"type": "integer"}}, "required": ["foo"]})
+    result = adapter.format_output_type(builder)
+    assert isinstance(result, dict)
+    assert result["response_mime_type"] == "application/json"
+    assert is_dataclass(result["response_schema"])
+
+
+def test_gemini_type_adapter_output_list_genson_schema_builder(adapter):
+    builder = SchemaBuilder()
+    builder.add_schema({"type": "object", "properties": {"foo": {"type": "string"}, "bar": {"type": "integer"}}, "required": ["foo"]})
+    result = adapter.format_output_type(list[builder])
+    assert isinstance(result, dict)
+    assert result["response_mime_type"] == "application/json"
+    args = get_args(result["response_schema"])
+    assert len(args) == 1
+    assert is_dataclass(args[0])
+
+
 def test_gemini_type_adapter_output_enum(adapter):
     class Foo(Enum):
         Bar = "bar"
diff --git a/tests/models/test_ollama_type_adapter.py b/tests/models/test_ollama_type_adapter.py
index 2061831a6..c949e3a14 100644
--- a/tests/models/test_ollama_type_adapter.py
+++ b/tests/models/test_ollama_type_adapter.py
@@ -144,19 +144,19 @@ class User(BaseModel):
     assert result == schema
 
 
-def test_ollama_type_adapter_output_genson_schema_builder(adapter, schema):
+def test_ollama_type_adapter_output_genson_schema_builder(adapter):
     builder = SchemaBuilder()
     builder.add_schema({"type": "object", "properties": {}})
     builder.add_object({"hi": "there"})
     builder.add_object({"hi": 5})
 
     result = adapter.format_output_type(builder)
-    assert result == json.dumps({
+    assert result == {
         "$schema": "http://json-schema.org/schema#",
         "type": "object",
         "properties": {"hi": {"type": ["integer", "string"]}},
         "required": ["hi"]
-    })
+    }
 
 
 def test_ollama_type_adapter_json_schema_str(adapter, schema):

From e72ec50c481cb35e9ccb794075cf43961c7621ad Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 3 Oct 2025 10:16:51 +0200
Subject: [PATCH 476/505] Fix dependencies installation instructions

---
 .github/workflows/tests.yml               | 2 +-
 .github/workflows/tests_api_models.yml    | 2 +-
 docs/features/models/anthropic.md         | 2 +-
 docs/features/models/dottxt.md            | 2 +-
 docs/features/models/gemini.md            | 2 +-
 docs/features/models/llamacpp.md          | 2 +-
 docs/features/models/mistral.md           | 2 +-
 docs/features/models/mlxlm.md             | 2 +-
 docs/features/models/ollama.md            | 2 +-
 docs/features/models/openai.md            | 2 +-
 docs/features/models/openai_compatible.md | 2 +-
 docs/features/models/openrouter.md        | 2 +-
 docs/features/models/sglang.md            | 2 +-
 docs/features/models/tgi.md               | 2 +-
 docs/features/models/transformers.md      | 2 +-
 15 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 309a754d8..7a8c03778 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -39,7 +39,7 @@ jobs:
         python -m pip install --upgrade pip
         pip install uv
         uv venv
-        uv pip install -e .[test]
+        uv pip install -e ".[test]"
     - name: Create matrix id
       id: matrix-id
       env:
diff --git a/.github/workflows/tests_api_models.yml b/.github/workflows/tests_api_models.yml
index 4d8633ac9..255a1d01c 100644
--- a/.github/workflows/tests_api_models.yml
+++ b/.github/workflows/tests_api_models.yml
@@ -38,7 +38,7 @@ jobs:
         python -m pip install --upgrade pip
         pip install uv
         uv venv
-        uv pip install -e .[test]
+        uv pip install -e ".[test]"
 
     - name: Create matrix id
       id: matrix-id
diff --git a/docs/features/models/anthropic.md b/docs/features/models/anthropic.md
index 403040ca4..6d49e2e92 100644
--- a/docs/features/models/anthropic.md
+++ b/docs/features/models/anthropic.md
@@ -6,7 +6,7 @@ title: Anthropic
 
 !!! Installation
 
-    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines. Install all optional dependencies of the `Anthropic` model with: `pip install outlines[anthropic]`.
+    You need to install the `anthropic` library to be able to use the Anthropic API in Outlines. Install all optional dependencies of the `Anthropic` model with: `pip install "outlines[anthropic]"`.
 
     You also need to have an Anthropic API key. This API key must either be set as an environment variable called `ANTHROPIC_API_KEY` or be provided to the `anthropic.Anthropic` class when instantiating it.
 
diff --git a/docs/features/models/dottxt.md b/docs/features/models/dottxt.md
index 306ef2c62..31dce1cec 100644
--- a/docs/features/models/dottxt.md
+++ b/docs/features/models/dottxt.md
@@ -6,7 +6,7 @@ title: Dottxt
 
 !!! Installation
 
-    You need to install the `dottxt` python sdk to be able to use the Dottxt API in Outlines. Install all optional dependencies of the `Dottxt` model with: `pip install outlines[dottxt]`.
+    You need to install the `dottxt` python sdk to be able to use the Dottxt API in Outlines. Install all optional dependencies of the `Dottxt` model with: `pip install "outlines[dottxt]"`.
 
     You also need to have a Dottxt API key. This API key must either be set as an environment variable called `DOTTXT_API_KEY` or be provided to the `dottxt.client.Dottxt` class when instantiating it.
 
diff --git a/docs/features/models/gemini.md b/docs/features/models/gemini.md
index ee48041aa..df125b67c 100644
--- a/docs/features/models/gemini.md
+++ b/docs/features/models/gemini.md
@@ -2,7 +2,7 @@
 
 !!! Installation
 
-    You need to install the `google.genai` libray to be able to use the Gemini API in Outlines. Install all optional dependencies of the `Gemini` model with: `pip install outlines[gemini]`.
+    You need to install the `google.genai` libray to be able to use the Gemini API in Outlines. Install all optional dependencies of the `Gemini` model with: `pip install "outlines[gemini]"`.
 
     You also need to have a Gemini API key. This API key must either be set as an environment variable called `GEMINI_API_KEY` or be provided to the `google.genai.Client` class when instantiating it.
 
diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
index 1dd2c31f2..0f1717e24 100644
--- a/docs/features/models/llamacpp.md
+++ b/docs/features/models/llamacpp.md
@@ -8,7 +8,7 @@ Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/l
 
 !!! Installation
 
-    You need to install the `llama-cpp-python` library to use the llama.cpp integration. Install all optional dependencies of the `LlamaCpp` model with: `pip install outlines[llamacpp]`.
+    You need to install the `llama-cpp-python` library to use the llama.cpp integration. Install all optional dependencies of the `LlamaCpp` model with: `"pip install "outlines[llamacpp]"`.
 
     See the [llama-cpp-python Github page](https://github.com/abetlen/llama-cpp-python?tab=readme-ov-file#supported-backends) for instructions on installing with CUDA, Metal, ROCm and other backends.
 
diff --git a/docs/features/models/mistral.md b/docs/features/models/mistral.md
index 8aec916f4..70be9eb8e 100644
--- a/docs/features/models/mistral.md
+++ b/docs/features/models/mistral.md
@@ -2,7 +2,7 @@
 
 !!! Installation
 
-    You need to install the `mistralai` library to be able to use the Mistral API in Outlines. Install all optional dependencies of the `Mistral` model with: `pip install outlines[mistral]`.
+    You need to install the `mistralai` library to be able to use the Mistral API in Outlines. Install all optional dependencies of the `Mistral` model with: `pip install "outlines[mistral]"`.
 
     You also need to have an Mistral API key. This API key must either be set as an environment variable called `MISTRAL_API_KEY` or be provided to the `mistralai.Mistral` class when instantiating it.
 
diff --git a/docs/features/models/mlxlm.md b/docs/features/models/mlxlm.md
index dc34257ac..546bb6a37 100644
--- a/docs/features/models/mlxlm.md
+++ b/docs/features/models/mlxlm.md
@@ -10,7 +10,7 @@ Outlines provides an integration with [mlx-lm](https://github.com/ml-explore/mlx
 
     You need a device that [supports Metal](https://support.apple.com/en-us/102894) to use the mlx-lm integration.
 
-    You need to install the `mlx` and `mlx-lm` libraries to be able to use mlx in Outlines. Install all optional dependencies of the `MLXLM` model with: `pip install outlines[mlxlm]`.
+    You need to install the `mlx` and `mlx-lm` libraries to be able to use mlx in Outlines. Install all optional dependencies of the `MLXLM` model with: `pip install "outlines[mlxlm]"`.
 
 ## Model Initialization
 
diff --git a/docs/features/models/ollama.md b/docs/features/models/ollama.md
index 202927901..7e8cef5f6 100644
--- a/docs/features/models/ollama.md
+++ b/docs/features/models/ollama.md
@@ -9,7 +9,7 @@ title: Ollama
     To be able to use Ollama in Outlines, you must install both Ollama and the optional dependency libraries of the model.
 
     - To download Ollama: https://ollama.com/download
-    - To install the ollama python sdk: `pip install outlines[ollama]`
+    - To install the ollama python sdk: `pip install "outlines[ollama]"`
 
     Consult the [`ollama` documentation](https://github.com/ollama/ollama-python) for detailed information on installation and client initialization.
 
diff --git a/docs/features/models/openai.md b/docs/features/models/openai.md
index 07a7fa08f..93b6d8091 100644
--- a/docs/features/models/openai.md
+++ b/docs/features/models/openai.md
@@ -2,7 +2,7 @@
 
 !!! Installation
 
-    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
+    You need to install the `openai` library to be able to use the OpenAI API in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install "outlines[openai]"`.
 
     You also need to have an OpenAI API key. This API key must either be set as an environment variable called `OPENAI_API_KEY` or be provided to the `openai.OpenAI` class when instantiating it.
 
diff --git a/docs/features/models/openai_compatible.md b/docs/features/models/openai_compatible.md
index 4925d53d4..7ab85d224 100644
--- a/docs/features/models/openai_compatible.md
+++ b/docs/features/models/openai_compatible.md
@@ -8,7 +8,7 @@ OpenAI-compatible APIs implement the same REST endpoints and request/response fo
 
 !!! Installation
 
-    You need to install the `openai` library to be able to use the OpenAI-compatible APIs in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
+    You need to install the `openai` library to be able to use the OpenAI-compatible APIs in Outlines. Install all optional dependencies of the `OpenAI` model with: `pip install "outlines[openai]"`.
 
 ## General Usage Pattern
 
diff --git a/docs/features/models/openrouter.md b/docs/features/models/openrouter.md
index 948cf9b8a..29354d58c 100644
--- a/docs/features/models/openrouter.md
+++ b/docs/features/models/openrouter.md
@@ -2,7 +2,7 @@
 
 !!! Installation
 
-    [OpenRouter](https://openrouter.ai/docs/api-reference/overview) uses the same API as OpenAI, so both services are [interoperable](./openai_compatible.md) using the `openai` library. Install all optional dependencies of the `OpenAI` model with: `pip install outlines[openai]`.
+    [OpenRouter](https://openrouter.ai/docs/api-reference/overview) uses the same API as OpenAI, so both services are [interoperable](./openai_compatible.md) using the `openai` library. Install all optional dependencies of the `OpenAI` model with: `pip install "outlines[openai]"`.
 
     You also need to have an Openrouter API key. This API key must either be set as an environment variable called `OPENAI_API_KEY` or be provided to the `openai.OpenAI` class when instantiating it.
 
diff --git a/docs/features/models/sglang.md b/docs/features/models/sglang.md
index 428b6f69c..14a6b37d9 100644
--- a/docs/features/models/sglang.md
+++ b/docs/features/models/sglang.md
@@ -19,7 +19,7 @@ python -m sglang.launch_server \
 
 Follow the [Installation instructions](https://docs.sglang.ai/start/install.html) for more information on how to set up a SGLang server for your particular setup.
 
-As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies of the `SGLang` model with: `pip install outlines[sglang]`.
+As the SGLang client relies on the `openai` python sdk, you need to have the `openai` package installed. Install all optional dependencies of the `SGLang` model with: `pip install "outlines[sglang]"`.
 
 When launching your SGLang server, you can specify the backend engine to use for structured generation through the `grammar-backend` cli argument. Add `--grammar-backend outlines` to your command to use Outlines instead of the default engine.
 
diff --git a/docs/features/models/tgi.md b/docs/features/models/tgi.md
index f71e8dad6..7523f5129 100644
--- a/docs/features/models/tgi.md
+++ b/docs/features/models/tgi.md
@@ -18,7 +18,7 @@ docker run \
 ```
 
 Please consult the [installation guide](https://huggingface.co/docs/text-generation-inference/en/quicktour) for more information about how to run TGI with your particular setup.
-As the TGI client relies on the `huggingface_hub` python package, you need to have it installed. Install all optional dependencoes of the `TGI` model with: `pip install outlines[tgi]`
+As the TGI client relies on the `huggingface_hub` python package, you need to have it installed. Install all optional dependencoes of the `TGI` model with: `pip install "outlines[tgi]"`
 
 ## Model Initialization
 
diff --git a/docs/features/models/transformers.md b/docs/features/models/transformers.md
index 77c5d4241..25b5a851e 100644
--- a/docs/features/models/transformers.md
+++ b/docs/features/models/transformers.md
@@ -6,7 +6,7 @@ title: Transformers
 
 !!! Installation
 
-    You need to install the `transformers` library to be able to use the Transformers in Outlines. Install all optional dependencies of the `Transformers` model with: `pip install outlines[transformers]`.
+    You need to install the `transformers` library to be able to use the Transformers in Outlines. Install all optional dependencies of the `Transformers` model with: `pip install "outlines[transformers]"`.
 
     See the [HuggingFace documentation](https://huggingface.co/docs/transformers/en/installation) for more information on installing `transformers` with CPU, GPU...
 

From 1a8d2c0c0255ae250d7e6fe731dbf1dba0bcf25f Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 3 Oct 2025 10:34:35 +0200
Subject: [PATCH 477/505] Add Mistral to the list of available models in the
 doc

---
 docs/features/models/index.md | 28 +++++++++++++++-------------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/docs/features/models/index.md b/docs/features/models/index.md
index f7969eec8..2aa2fea7a 100644
--- a/docs/features/models/index.md
+++ b/docs/features/models/index.md
@@ -84,19 +84,19 @@ print(response) # ['Riga', 'Tallinn']
 
 In alphabetical order:
 
-| | [Anthropic](../../models/anthropic) | [Dottxt](../../models/dottxt) | [Gemini](../../models/gemini) | [LlamaCpp](../../models/llamacpp) | [MLXLM](../../models/mlxlm) | [Ollama](../../models/ollama) | [OpenAI](../../models/openai) | [SGLang](../../models/sglang) | [TGI](../../models/tgi) | [Transformers](../../models/transformers) | [Transformers MultiModal](../../models/transformers_multimodal) | [VLLM](../../models/vllm) | [VLLMOffline](../../models/vllm_offline) |
-|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
-| **Output Types** | | | | | | | | | | | | | |
-| Simple Types | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| JSON Schema | ❌ | ✅ | 🟠 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| Multiple Choice | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| Regex | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
-| Grammar | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | 🟠 | ❌ | ✅ | ✅ | ✅ | ✅ |
-| **Generation Features** | | | | | | | | | | | | | |
-| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
-| Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
-| Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
-| Batching | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ✅ |
+| | [Anthropic](../../models/anthropic) | [Dottxt](../../models/dottxt) | [Gemini](../../models/gemini) | [LlamaCpp](../../models/llamacpp) | [MLXLM](../../models/mlxlm) | [Mistral](../../models/mistral) | [Ollama](../../models/ollama) | [OpenAI](../../models/openai) | [SGLang](../../models/sglang) | [TGI](../../models/tgi) | [Transformers](../../models/transformers) | [Transformers MultiModal](../../models/transformers_multimodal) | [VLLM](../../models/vllm) | [VLLMOffline](../../models/vllm_offline) |
+|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
+| **Output Types** | | | | | | | | | | | | | | |
+| Simple Types | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| JSON Schema | ❌ | ✅ | 🟠 | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Multiple Choice | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Regex | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ |
+| Grammar | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | 🟠 | ❌ | ✅ | ✅ | ✅ | ✅ |
+| **Generation Features** | | | | | | | | | | | | | | |
+| Async | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Streaming | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ✅ | ❌ |
+| Vision | ✅ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ |
+| Batching | ❌ | ❌ | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ | ❌ | ✅ |
 
 ## Model Types
 
@@ -119,6 +119,7 @@ The server-based models available are the following:
 - Anthropic
 - Dottxt
 - Gemini
+- Mistral
 - Ollama
 - OpenAI
 - SgLang
@@ -141,6 +142,7 @@ print(type(model)) # outlines.models.tgi.AsyncTGI
 
 The models that have an async version are the following:
 
+- Mistral
 - Ollama
 - OpenAI
 - SgLang

From b393c8874514f6f9a767646f1ad84bd1486ae36e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9mi=20Louf?= <remilouf@protonmail.com>
Date: Mon, 6 Oct 2025 11:42:23 +0200
Subject: [PATCH 478/505] Add design partner CTA in docs

---
 docs/index.md | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/docs/index.md b/docs/index.md
index 8e90f2daf..e9609c326 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -11,6 +11,7 @@ hide:
 ![](assets/images/logo-dark-mode.svg#only-dark){ width="500" }
 </figure>
 
+
 LLMs are powerful but their outputs are unpredictable. Most solutions attempt to fix bad outputs after generation using parsing, regex, or fragile code that breaks easily.
 
 Outlines guarantees structured outputs during generation — directly from any LLM.
@@ -28,6 +29,14 @@ Outlines guarantees structured outputs during generation — directly from any L
 [GitHub](https://github.com/dottxt-ai/outlines){ .md-button }
 </figure>
 
+## 🚀 Building the future of structured generation
+
+We're working with select partners to develop new interfaces to structured generation.
+
+Need XML, FHIR, custom schemas or grammars? Let's talk.
+
+[Become a design partner](mailto:contact@dottxt.ai)
+
 ## See it in action
 
 ```python

From c81e06ee2229e5a523678c95f1e30b9f75410a5b Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Fri, 26 Sep 2025 16:10:34 +0200
Subject: [PATCH 479/505] Add the device_dtype init parameter to the
 transformers model

---
 docs/features/models/transformers.md          |  3 +-
 .../models/transformers_multimodal.md         |  1 +
 outlines/models/transformers.py               | 40 ++++++++++++++++---
 tests/models/test_transformers.py             |  9 ++++-
 tests/models/test_transformers_multimodal.py  | 12 +++++-
 5 files changed, 56 insertions(+), 9 deletions(-)

diff --git a/docs/features/models/transformers.md b/docs/features/models/transformers.md
index 25b5a851e..d001d0ab5 100644
--- a/docs/features/models/transformers.md
+++ b/docs/features/models/transformers.md
@@ -12,10 +12,11 @@ title: Transformers
 
 ## Model Initialization
 
-To load the model, you can use the `from_transformers` function. It takes 2 arguments:
+To load the model, you can use the `from_transformers` function. It takes 3 arguments:
 
 - `model`: a `transformers` model (created with `AutoModelForCausalLM` for instance)
 - `tokenizer_or_processor`: a `transformers` tokenizer (created with `AutoTokenizer` for instance, it must be an instance of either `PreTrainedTokenizer` or `PreTrainedTokenizerFast`)
+- `device_dtype` (optional): the tensor dtype to use for inference. If not provided, the model will use the default dtype.
 
 For instance:
 
diff --git a/docs/features/models/transformers_multimodal.md b/docs/features/models/transformers_multimodal.md
index 3d34a0f78..78e6c51a1 100644
--- a/docs/features/models/transformers_multimodal.md
+++ b/docs/features/models/transformers_multimodal.md
@@ -12,6 +12,7 @@ To load the model, you can use the `from_transformers` function. It takes 2 argu
 
 - `model`: a `transformers` model (created with `AutoModelForImageTextToText` for instance)
 - `tokenizer_or_processor`: a `transformers` processor (created with `AutoProcessor` for instance, it must be an instance of `ProcessorMixin`)
+- `device_dtype` (optional): the tensor dtype to use for inference. If not provided, the model will use the default dtype.
 
 For instance:
 
diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 4d6387a40..42f05553c 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -209,6 +209,8 @@ def __init__(
         self,
         model: "PreTrainedModel",
         tokenizer: "PreTrainedTokenizer",
+        *,
+        device_dtype: Optional["torch.dtype"] = None,
     ):
         """
         Parameters:
@@ -219,6 +221,9 @@ def __init__(
         tokenizer
             A `PreTrainedTokenizer`, or any tokenizer that is compatible with
             the `transformers` API for tokenizers.
+        device_dtype
+            The dtype to use for the model. If not provided, the model will use
+            the default dtype.
 
         """
         # We need to handle the cases in which jax/flax or tensorflow
@@ -237,6 +242,7 @@ def __init__(
         self.model = model
         self.hf_tokenizer = tokenizer
         self.tokenizer = TransformerTokenizer(tokenizer)
+        self.device_dtype = device_dtype
         self.type_adapter = TransformersTypeAdapter(tokenizer=tokenizer)
 
         if (
@@ -287,7 +293,11 @@ def _prepare_model_inputs(
         input_ids, attention_mask = self.tokenizer.encode(prompts)
         inputs = {
             "input_ids": input_ids.to(self.model.device),
-            "attention_mask": attention_mask.to(self.model.device),
+            "attention_mask": (
+                attention_mask.to(self.model.device, dtype=self.device_dtype)
+                if self.device_dtype is not None
+                else attention_mask.to(self.model.device)
+            ),
         }
 
         return prompts, inputs
@@ -600,7 +610,13 @@ class TransformersMultiModal(Transformers):
 
     """
 
-    def __init__(self, model: "PreTrainedModel", processor):
+    def __init__(
+        self,
+        model: "PreTrainedModel",
+        processor,
+        *,
+        device_dtype: Optional["torch.dtype"] = None,
+    ):
         """Create a TransformersMultiModal model instance
 
         We rely on the `__init__` method of the `Transformers` class to handle
@@ -614,6 +630,9 @@ def __init__(self, model: "PreTrainedModel", processor):
             `transformers` API for models.
         processor
             A `ProcessorMixin` instance.
+        device_dtype
+            The dtype to use for the model. If not provided, the model will use
+            the default dtype.
 
         """
         self.processor = processor
@@ -622,7 +641,7 @@ def __init__(self, model: "PreTrainedModel", processor):
 
         tokenizer: "PreTrainedTokenizer" = self.processor.tokenizer
 
-        super().__init__(model, tokenizer)
+        super().__init__(model, tokenizer, device_dtype=device_dtype)
 
         self.type_adapter = TransformersMultiModalTypeAdapter(
             tokenizer=tokenizer
@@ -655,7 +674,11 @@ def _prepare_model_inputs(
 
         inputs = self.processor(
             **merged_prompts, padding=True, return_tensors="pt"
-        ).to(self.model.device)
+        )
+        if self.device_dtype is not None:
+            inputs = inputs.to(self.model.device, dtype=self.device_dtype)
+        else:
+            inputs = inputs.to(self.model.device)
 
         return merged_prompts["text"], inputs
 
@@ -663,6 +686,8 @@ def _prepare_model_inputs(
 def from_transformers(
     model: "PreTrainedModel",
     tokenizer_or_processor: Union["PreTrainedTokenizer", "ProcessorMixin"],
+    *,
+    device_dtype: Optional["torch.dtype"] = None,
 ) -> Union[Transformers, TransformersMultiModal]:
     """Create an Outlines `Transformers` or `TransformersMultiModal` model
     instance from a `PreTrainedModel` instance and a `PreTrainedTokenizer` or
@@ -679,6 +704,9 @@ def from_transformers(
     tokenizer_or_processor
         A `transformers.PreTrainedTokenizer` or
         `transformers.ProcessorMixin` instance.
+    device_dtype
+        The dtype to use for the model. If not provided, the model will use
+        the default dtype.
 
     Returns
     -------
@@ -693,10 +721,10 @@ def from_transformers(
         tokenizer_or_processor, (PreTrainedTokenizer, PreTrainedTokenizerFast)
     ):
         tokenizer = tokenizer_or_processor
-        return Transformers(model, tokenizer)
+        return Transformers(model, tokenizer, device_dtype=device_dtype)
     elif isinstance(tokenizer_or_processor, ProcessorMixin):
         processor = tokenizer_or_processor
-        return TransformersMultiModal(model, processor)
+        return TransformersMultiModal(model, processor, device_dtype=device_dtype)
     else:
         raise ValueError(
             "We could determine whether the model passed to `from_transformers`"
diff --git a/tests/models/test_transformers.py b/tests/models/test_transformers.py
index 10f572794..91b7ff7ab 100644
--- a/tests/models/test_transformers.py
+++ b/tests/models/test_transformers.py
@@ -3,6 +3,7 @@
 
 from pydantic import BaseModel
 import pytest
+import torch
 import transformers
 
 import outlines
@@ -47,15 +48,17 @@ def test_transformers_instantiate_mamba():
     assert isinstance(model, Transformers)
 
 
-def test_transformers_instantiate_tokenizer_kwargs():
+def test_transformers_instantiate_tokenizer_kwargs_dtype():
     model = outlines.from_transformers(
         transformers.AutoModelForCausalLM.from_pretrained(TEST_MODEL),
         transformers.AutoTokenizer.from_pretrained(
             TEST_MODEL, additional_special_tokens=["<t1>", "<t2>"]
         ),
+        device_dtype=torch.bfloat16,
     )
     assert "<t1>" in model.tokenizer.special_tokens
     assert "<t2>" in model.tokenizer.special_tokens
+    assert model.device_dtype == torch.bfloat16
 
 
 @pytest.fixture
@@ -88,6 +91,10 @@ def test_transformers_call(model, model_bart):
     result = model("Respond with one word. Not more.")
     assert isinstance(result, str)
 
+    model.device_dtype = torch.bfloat16
+    result = model("Respond with one word. Not more.")
+    assert isinstance(result, str)
+
     result = model_bart("Respond with one word. Not more.")
     assert isinstance(result, str)
 
diff --git a/tests/models/test_transformers_multimodal.py b/tests/models/test_transformers_multimodal.py
index ce968e1f6..4bdc7282e 100644
--- a/tests/models/test_transformers_multimodal.py
+++ b/tests/models/test_transformers_multimodal.py
@@ -2,6 +2,7 @@
 
 import io
 import re
+import torch
 from enum import Enum
 
 import pytest
@@ -47,15 +48,17 @@ def model():
     return model
 
 
-def test_transformers_multimodal_instantiate_simple():
+def test_transformers_multimodal_instantiate():
     model = outlines.from_transformers(
         LlavaForConditionalGeneration.from_pretrained(TEST_MODEL),
         AutoProcessor.from_pretrained(TEST_MODEL),
+        device_dtype=torch.bfloat16,
     )
     assert isinstance(model, TransformersMultiModal)
     assert isinstance(model.tokenizer, TransformerTokenizer)
     assert isinstance(model.type_adapter, TransformersMultiModalTypeAdapter)
     assert model.tensor_library_name == "torch"
+    assert model.device_dtype == torch.bfloat16
 
 
 def test_transformers_multimodal_simple(model, image):
@@ -74,6 +77,13 @@ def test_transformers_multimodal_call(model, image):
     )
     assert isinstance(result, str)
 
+    model.device_dtype = torch.bfloat16
+    result = model(
+        ["<image>Describe this image in one sentence:", Image(image)],
+        max_new_tokens=2,
+    )
+    assert isinstance(result, str)
+
 
 def test_transformers_multimodal_wrong_number_image(model, image):
     with pytest.raises(ValueError):

From ffe92a731b02690845e185173046fb1ab9467e28 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 8 Oct 2025 09:56:40 +0200
Subject: [PATCH 480/505] Update the test workflows in the CI to use uv instead
 of pip

---
 .github/workflows/tests.yml            | 3 +--
 .github/workflows/tests_api_models.yml | 3 +--
 pyproject.toml                         | 4 +++-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 7a8c03778..a8e578e0a 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -38,8 +38,7 @@ jobs:
       run: |
         python -m pip install --upgrade pip
         pip install uv
-        uv venv
-        uv pip install -e ".[test]"
+        uv sync --no-group test-gpu --extra test
     - name: Create matrix id
       id: matrix-id
       env:
diff --git a/.github/workflows/tests_api_models.yml b/.github/workflows/tests_api_models.yml
index 255a1d01c..3d0610135 100644
--- a/.github/workflows/tests_api_models.yml
+++ b/.github/workflows/tests_api_models.yml
@@ -37,8 +37,7 @@ jobs:
       run: |
         python -m pip install --upgrade pip
         pip install uv
-        uv venv
-        uv pip install -e ".[test]"
+        uv sync --no-group test-gpu --extra test
 
     - name: Create matrix id
       id: matrix-id
diff --git a/pyproject.toml b/pyproject.toml
index 027d0a074..46281cfbf 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -93,7 +93,9 @@ test = [
     "iso3166",
     "requests",
 ]
-test-gpu=["outlines[test]", "vllm; sys_platform == 'linux'"]
+
+[dependency-groups]
+test-gpu = ["outlines[test]", "vllm; sys_platform == 'linux'"]
 
 [project.urls]
 homepage = "https://github.com/dottxt-ai/outlines"

From ca0cf180a6c5c8478ab4348e944b54f5343dface Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Wed, 8 Oct 2025 09:56:51 +0200
Subject: [PATCH 481/505] Update the uv lock

---
 uv.lock | 3564 +++++++++++++++++++++++++++++++++++--------------------
 1 file changed, 2253 insertions(+), 1311 deletions(-)

diff --git a/uv.lock b/uv.lock
index 5b8721b9a..b69731165 100644
--- a/uv.lock
+++ b/uv.lock
@@ -56,7 +56,7 @@ wheels = [
 
 [[package]]
 name = "aiohttp"
-version = "3.12.15"
+version = "3.13.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "aiohappyeyeballs" },
@@ -68,76 +68,76 @@ dependencies = [
     { name = "propcache" },
     { name = "yarl" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/9b/e7/d92a237d8802ca88483906c388f7c201bbe96cd80a165ffd0ac2f6a8d59f/aiohttp-3.12.15.tar.gz", hash = "sha256:4fc61385e9c98d72fcdf47e6dd81833f47b2f77c114c29cd64a361be57a763a2", size = 7823716, upload-time = "2025-07-29T05:52:32.215Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/47/dc/ef9394bde9080128ad401ac7ede185267ed637df03b51f05d14d1c99ad67/aiohttp-3.12.15-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:b6fc902bff74d9b1879ad55f5404153e2b33a82e72a95c89cec5eb6cc9e92fbc", size = 703921, upload-time = "2025-07-29T05:49:43.584Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/42/63fccfc3a7ed97eb6e1a71722396f409c46b60a0552d8a56d7aad74e0df5/aiohttp-3.12.15-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:098e92835b8119b54c693f2f88a1dec690e20798ca5f5fe5f0520245253ee0af", size = 480288, upload-time = "2025-07-29T05:49:47.851Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/a2/7b8a020549f66ea2a68129db6960a762d2393248f1994499f8ba9728bbed/aiohttp-3.12.15-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:40b3fee496a47c3b4a39a731954c06f0bd9bd3e8258c059a4beb76ac23f8e421", size = 468063, upload-time = "2025-07-29T05:49:49.789Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/f5/d11e088da9176e2ad8220338ae0000ed5429a15f3c9dfd983f39105399cd/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2ce13fcfb0bb2f259fb42106cdc63fa5515fb85b7e87177267d89a771a660b79", size = 1650122, upload-time = "2025-07-29T05:49:51.874Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/6b/b60ce2757e2faed3d70ed45dafee48cee7bfb878785a9423f7e883f0639c/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3beb14f053222b391bf9cf92ae82e0171067cc9c8f52453a0f1ec7c37df12a77", size = 1624176, upload-time = "2025-07-29T05:49:53.805Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/de/8c9fde2072a1b72c4fadecf4f7d4be7a85b1d9a4ab333d8245694057b4c6/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4c39e87afe48aa3e814cac5f535bc6199180a53e38d3f51c5e2530f5aa4ec58c", size = 1696583, upload-time = "2025-07-29T05:49:55.338Z" },
-    { url = "https://files.pythonhosted.org/packages/0c/ad/07f863ca3d895a1ad958a54006c6dafb4f9310f8c2fdb5f961b8529029d3/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:d5f1b4ce5bc528a6ee38dbf5f39bbf11dd127048726323b72b8e85769319ffc4", size = 1738896, upload-time = "2025-07-29T05:49:57.045Z" },
-    { url = "https://files.pythonhosted.org/packages/20/43/2bd482ebe2b126533e8755a49b128ec4e58f1a3af56879a3abdb7b42c54f/aiohttp-3.12.15-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1004e67962efabbaf3f03b11b4c43b834081c9e3f9b32b16a7d97d4708a9abe6", size = 1643561, upload-time = "2025-07-29T05:49:58.762Z" },
-    { url = "https://files.pythonhosted.org/packages/23/40/2fa9f514c4cf4cbae8d7911927f81a1901838baf5e09a8b2c299de1acfe5/aiohttp-3.12.15-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:8faa08fcc2e411f7ab91d1541d9d597d3a90e9004180edb2072238c085eac8c2", size = 1583685, upload-time = "2025-07-29T05:50:00.375Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/c3/94dc7357bc421f4fb978ca72a201a6c604ee90148f1181790c129396ceeb/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:fe086edf38b2222328cdf89af0dde2439ee173b8ad7cb659b4e4c6f385b2be3d", size = 1627533, upload-time = "2025-07-29T05:50:02.306Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/3f/1f8911fe1844a07001e26593b5c255a685318943864b27b4e0267e840f95/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:79b26fe467219add81d5e47b4a4ba0f2394e8b7c7c3198ed36609f9ba161aecb", size = 1638319, upload-time = "2025-07-29T05:50:04.282Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/46/27bf57a99168c4e145ffee6b63d0458b9c66e58bb70687c23ad3d2f0bd17/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:b761bac1192ef24e16706d761aefcb581438b34b13a2f069a6d343ec8fb693a5", size = 1613776, upload-time = "2025-07-29T05:50:05.863Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/7e/1d2d9061a574584bb4ad3dbdba0da90a27fdc795bc227def3a46186a8bc1/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:e153e8adacfe2af562861b72f8bc47f8a5c08e010ac94eebbe33dc21d677cd5b", size = 1693359, upload-time = "2025-07-29T05:50:07.563Z" },
-    { url = "https://files.pythonhosted.org/packages/08/98/bee429b52233c4a391980a5b3b196b060872a13eadd41c3a34be9b1469ed/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:fc49c4de44977aa8601a00edbf157e9a421f227aa7eb477d9e3df48343311065", size = 1716598, upload-time = "2025-07-29T05:50:09.33Z" },
-    { url = "https://files.pythonhosted.org/packages/57/39/b0314c1ea774df3392751b686104a3938c63ece2b7ce0ba1ed7c0b4a934f/aiohttp-3.12.15-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:2776c7ec89c54a47029940177e75c8c07c29c66f73464784971d6a81904ce9d1", size = 1644940, upload-time = "2025-07-29T05:50:11.334Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/83/3dacb8d3f8f512c8ca43e3fa8a68b20583bd25636ffa4e56ee841ffd79ae/aiohttp-3.12.15-cp310-cp310-win32.whl", hash = "sha256:2c7d81a277fa78b2203ab626ced1487420e8c11a8e373707ab72d189fcdad20a", size = 429239, upload-time = "2025-07-29T05:50:12.803Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/f9/470b5daba04d558c9673ca2034f28d067f3202a40e17804425f0c331c89f/aiohttp-3.12.15-cp310-cp310-win_amd64.whl", hash = "sha256:83603f881e11f0f710f8e2327817c82e79431ec976448839f3cd05d7afe8f830", size = 452297, upload-time = "2025-07-29T05:50:14.266Z" },
-    { url = "https://files.pythonhosted.org/packages/20/19/9e86722ec8e835959bd97ce8c1efa78cf361fa4531fca372551abcc9cdd6/aiohttp-3.12.15-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:d3ce17ce0220383a0f9ea07175eeaa6aa13ae5a41f30bc61d84df17f0e9b1117", size = 711246, upload-time = "2025-07-29T05:50:15.937Z" },
-    { url = "https://files.pythonhosted.org/packages/71/f9/0a31fcb1a7d4629ac9d8f01f1cb9242e2f9943f47f5d03215af91c3c1a26/aiohttp-3.12.15-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:010cc9bbd06db80fe234d9003f67e97a10fe003bfbedb40da7d71c1008eda0fe", size = 483515, upload-time = "2025-07-29T05:50:17.442Z" },
-    { url = "https://files.pythonhosted.org/packages/62/6c/94846f576f1d11df0c2e41d3001000527c0fdf63fce7e69b3927a731325d/aiohttp-3.12.15-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3f9d7c55b41ed687b9d7165b17672340187f87a773c98236c987f08c858145a9", size = 471776, upload-time = "2025-07-29T05:50:19.568Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/6c/f766d0aaafcee0447fad0328da780d344489c042e25cd58fde566bf40aed/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bc4fbc61bb3548d3b482f9ac7ddd0f18c67e4225aaa4e8552b9f1ac7e6bda9e5", size = 1741977, upload-time = "2025-07-29T05:50:21.665Z" },
-    { url = "https://files.pythonhosted.org/packages/17/e5/fb779a05ba6ff44d7bc1e9d24c644e876bfff5abe5454f7b854cace1b9cc/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:7fbc8a7c410bb3ad5d595bb7118147dfbb6449d862cc1125cf8867cb337e8728", size = 1690645, upload-time = "2025-07-29T05:50:23.333Z" },
-    { url = "https://files.pythonhosted.org/packages/37/4e/a22e799c2035f5d6a4ad2cf8e7c1d1bd0923192871dd6e367dafb158b14c/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:74dad41b3458dbb0511e760fb355bb0b6689e0630de8a22b1b62a98777136e16", size = 1789437, upload-time = "2025-07-29T05:50:25.007Z" },
-    { url = "https://files.pythonhosted.org/packages/28/e5/55a33b991f6433569babb56018b2fb8fb9146424f8b3a0c8ecca80556762/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3b6f0af863cf17e6222b1735a756d664159e58855da99cfe965134a3ff63b0b0", size = 1828482, upload-time = "2025-07-29T05:50:26.693Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/82/1ddf0ea4f2f3afe79dffed5e8a246737cff6cbe781887a6a170299e33204/aiohttp-3.12.15-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b5b7fe4972d48a4da367043b8e023fb70a04d1490aa7d68800e465d1b97e493b", size = 1730944, upload-time = "2025-07-29T05:50:28.382Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/96/784c785674117b4cb3877522a177ba1b5e4db9ce0fd519430b5de76eec90/aiohttp-3.12.15-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6443cca89553b7a5485331bc9bedb2342b08d073fa10b8c7d1c60579c4a7b9bd", size = 1668020, upload-time = "2025-07-29T05:50:30.032Z" },
-    { url = "https://files.pythonhosted.org/packages/12/8a/8b75f203ea7e5c21c0920d84dd24a5c0e971fe1e9b9ebbf29ae7e8e39790/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:6c5f40ec615e5264f44b4282ee27628cea221fcad52f27405b80abb346d9f3f8", size = 1716292, upload-time = "2025-07-29T05:50:31.983Z" },
-    { url = "https://files.pythonhosted.org/packages/47/0b/a1451543475bb6b86a5cfc27861e52b14085ae232896a2654ff1231c0992/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:2abbb216a1d3a2fe86dbd2edce20cdc5e9ad0be6378455b05ec7f77361b3ab50", size = 1711451, upload-time = "2025-07-29T05:50:33.989Z" },
-    { url = "https://files.pythonhosted.org/packages/55/fd/793a23a197cc2f0d29188805cfc93aa613407f07e5f9da5cd1366afd9d7c/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:db71ce547012a5420a39c1b744d485cfb823564d01d5d20805977f5ea1345676", size = 1691634, upload-time = "2025-07-29T05:50:35.846Z" },
-    { url = "https://files.pythonhosted.org/packages/ca/bf/23a335a6670b5f5dfc6d268328e55a22651b440fca341a64fccf1eada0c6/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:ced339d7c9b5030abad5854aa5413a77565e5b6e6248ff927d3e174baf3badf7", size = 1785238, upload-time = "2025-07-29T05:50:37.597Z" },
-    { url = "https://files.pythonhosted.org/packages/57/4f/ed60a591839a9d85d40694aba5cef86dde9ee51ce6cca0bb30d6eb1581e7/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:7c7dd29c7b5bda137464dc9bfc738d7ceea46ff70309859ffde8c022e9b08ba7", size = 1805701, upload-time = "2025-07-29T05:50:39.591Z" },
-    { url = "https://files.pythonhosted.org/packages/85/e0/444747a9455c5de188c0f4a0173ee701e2e325d4b2550e9af84abb20cdba/aiohttp-3.12.15-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:421da6fd326460517873274875c6c5a18ff225b40da2616083c5a34a7570b685", size = 1718758, upload-time = "2025-07-29T05:50:41.292Z" },
-    { url = "https://files.pythonhosted.org/packages/36/ab/1006278d1ffd13a698e5dd4bfa01e5878f6bddefc296c8b62649753ff249/aiohttp-3.12.15-cp311-cp311-win32.whl", hash = "sha256:4420cf9d179ec8dfe4be10e7d0fe47d6d606485512ea2265b0d8c5113372771b", size = 428868, upload-time = "2025-07-29T05:50:43.063Z" },
-    { url = "https://files.pythonhosted.org/packages/10/97/ad2b18700708452400278039272032170246a1bf8ec5d832772372c71f1a/aiohttp-3.12.15-cp311-cp311-win_amd64.whl", hash = "sha256:edd533a07da85baa4b423ee8839e3e91681c7bfa19b04260a469ee94b778bf6d", size = 453273, upload-time = "2025-07-29T05:50:44.613Z" },
-    { url = "https://files.pythonhosted.org/packages/63/97/77cb2450d9b35f517d6cf506256bf4f5bda3f93a66b4ad64ba7fc917899c/aiohttp-3.12.15-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:802d3868f5776e28f7bf69d349c26fc0efadb81676d0afa88ed00d98a26340b7", size = 702333, upload-time = "2025-07-29T05:50:46.507Z" },
-    { url = "https://files.pythonhosted.org/packages/83/6d/0544e6b08b748682c30b9f65640d006e51f90763b41d7c546693bc22900d/aiohttp-3.12.15-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f2800614cd560287be05e33a679638e586a2d7401f4ddf99e304d98878c29444", size = 476948, upload-time = "2025-07-29T05:50:48.067Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/1d/c8c40e611e5094330284b1aea8a4b02ca0858f8458614fa35754cab42b9c/aiohttp-3.12.15-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8466151554b593909d30a0a125d638b4e5f3836e5aecde85b66b80ded1cb5b0d", size = 469787, upload-time = "2025-07-29T05:50:49.669Z" },
-    { url = "https://files.pythonhosted.org/packages/38/7d/b76438e70319796bfff717f325d97ce2e9310f752a267bfdf5192ac6082b/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2e5a495cb1be69dae4b08f35a6c4579c539e9b5706f606632102c0f855bcba7c", size = 1716590, upload-time = "2025-07-29T05:50:51.368Z" },
-    { url = "https://files.pythonhosted.org/packages/79/b1/60370d70cdf8b269ee1444b390cbd72ce514f0d1cd1a715821c784d272c9/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:6404dfc8cdde35c69aaa489bb3542fb86ef215fc70277c892be8af540e5e21c0", size = 1699241, upload-time = "2025-07-29T05:50:53.628Z" },
-    { url = "https://files.pythonhosted.org/packages/a3/2b/4968a7b8792437ebc12186db31523f541943e99bda8f30335c482bea6879/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3ead1c00f8521a5c9070fcb88f02967b1d8a0544e6d85c253f6968b785e1a2ab", size = 1754335, upload-time = "2025-07-29T05:50:55.394Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/c1/49524ed553f9a0bec1a11fac09e790f49ff669bcd14164f9fab608831c4d/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6990ef617f14450bc6b34941dba4f12d5613cbf4e33805932f853fbd1cf18bfb", size = 1800491, upload-time = "2025-07-29T05:50:57.202Z" },
-    { url = "https://files.pythonhosted.org/packages/de/5e/3bf5acea47a96a28c121b167f5ef659cf71208b19e52a88cdfa5c37f1fcc/aiohttp-3.12.15-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd736ed420f4db2b8148b52b46b88ed038d0354255f9a73196b7bbce3ea97545", size = 1719929, upload-time = "2025-07-29T05:50:59.192Z" },
-    { url = "https://files.pythonhosted.org/packages/39/94/8ae30b806835bcd1cba799ba35347dee6961a11bd507db634516210e91d8/aiohttp-3.12.15-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3c5092ce14361a73086b90c6efb3948ffa5be2f5b6fbcf52e8d8c8b8848bb97c", size = 1635733, upload-time = "2025-07-29T05:51:01.394Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/46/06cdef71dd03acd9da7f51ab3a9107318aee12ad38d273f654e4f981583a/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:aaa2234bb60c4dbf82893e934d8ee8dea30446f0647e024074237a56a08c01bd", size = 1696790, upload-time = "2025-07-29T05:51:03.657Z" },
-    { url = "https://files.pythonhosted.org/packages/02/90/6b4cfaaf92ed98d0ec4d173e78b99b4b1a7551250be8937d9d67ecb356b4/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:6d86a2fbdd14192e2f234a92d3b494dd4457e683ba07e5905a0b3ee25389ac9f", size = 1718245, upload-time = "2025-07-29T05:51:05.911Z" },
-    { url = "https://files.pythonhosted.org/packages/2e/e6/2593751670fa06f080a846f37f112cbe6f873ba510d070136a6ed46117c6/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:a041e7e2612041a6ddf1c6a33b883be6a421247c7afd47e885969ee4cc58bd8d", size = 1658899, upload-time = "2025-07-29T05:51:07.753Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/28/c15bacbdb8b8eb5bf39b10680d129ea7410b859e379b03190f02fa104ffd/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:5015082477abeafad7203757ae44299a610e89ee82a1503e3d4184e6bafdd519", size = 1738459, upload-time = "2025-07-29T05:51:09.56Z" },
-    { url = "https://files.pythonhosted.org/packages/00/de/c269cbc4faa01fb10f143b1670633a8ddd5b2e1ffd0548f7aa49cb5c70e2/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:56822ff5ddfd1b745534e658faba944012346184fbfe732e0d6134b744516eea", size = 1766434, upload-time = "2025-07-29T05:51:11.423Z" },
-    { url = "https://files.pythonhosted.org/packages/52/b0/4ff3abd81aa7d929b27d2e1403722a65fc87b763e3a97b3a2a494bfc63bc/aiohttp-3.12.15-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b2acbbfff69019d9014508c4ba0401822e8bae5a5fdc3b6814285b71231b60f3", size = 1726045, upload-time = "2025-07-29T05:51:13.689Z" },
-    { url = "https://files.pythonhosted.org/packages/71/16/949225a6a2dd6efcbd855fbd90cf476052e648fb011aa538e3b15b89a57a/aiohttp-3.12.15-cp312-cp312-win32.whl", hash = "sha256:d849b0901b50f2185874b9a232f38e26b9b3d4810095a7572eacea939132d4e1", size = 423591, upload-time = "2025-07-29T05:51:15.452Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/d8/fa65d2a349fe938b76d309db1a56a75c4fb8cc7b17a398b698488a939903/aiohttp-3.12.15-cp312-cp312-win_amd64.whl", hash = "sha256:b390ef5f62bb508a9d67cb3bba9b8356e23b3996da7062f1a57ce1a79d2b3d34", size = 450266, upload-time = "2025-07-29T05:51:17.239Z" },
-    { url = "https://files.pythonhosted.org/packages/18/8d/da08099af8db234d1cd43163e6ffc8e9313d0e988cee1901610f2fa5c764/aiohttp-3.12.15-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:691d203c2bdf4f4637792efbbcdcd157ae11e55eaeb5e9c360c1206fb03d4d98", size = 706829, upload-time = "2025-07-29T05:51:54.434Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/94/8eed385cfb60cf4fdb5b8a165f6148f3bebeb365f08663d83c35a5f273ef/aiohttp-3.12.15-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:8e995e1abc4ed2a454c731385bf4082be06f875822adc4c6d9eaadf96e20d406", size = 481806, upload-time = "2025-07-29T05:51:56.355Z" },
-    { url = "https://files.pythonhosted.org/packages/38/68/b13e1a34584fbf263151b3a72a084e89f2102afe38df1dce5a05a15b83e9/aiohttp-3.12.15-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:bd44d5936ab3193c617bfd6c9a7d8d1085a8dc8c3f44d5f1dcf554d17d04cf7d", size = 469205, upload-time = "2025-07-29T05:51:58.277Z" },
-    { url = "https://files.pythonhosted.org/packages/38/14/3d7348bf53aa4af54416bc64cbef3a2ac5e8b9bfa97cc45f1cf9a94d9c8d/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46749be6e89cd78d6068cdf7da51dbcfa4321147ab8e4116ee6678d9a056a0cf", size = 1644174, upload-time = "2025-07-29T05:52:00.23Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/ed/fd9b5b22b0f6ca1a85c33bb4868cbcc6ae5eae070a0f4c9c5cad003c89d7/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:0c643f4d75adea39e92c0f01b3fb83d57abdec8c9279b3078b68a3a52b3933b6", size = 1618672, upload-time = "2025-07-29T05:52:02.272Z" },
-    { url = "https://files.pythonhosted.org/packages/39/f7/f6530ab5f8c8c409e44a63fcad35e839c87aabecdfe5b8e96d671ed12f64/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0a23918fedc05806966a2438489dcffccbdf83e921a1170773b6178d04ade142", size = 1692295, upload-time = "2025-07-29T05:52:04.546Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/dc/3cf483bb0106566dc97ebaa2bb097f5e44d4bc4ab650a6f107151cd7b193/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:74bdd8c864b36c3673741023343565d95bfbd778ffe1eb4d412c135a28a8dc89", size = 1731609, upload-time = "2025-07-29T05:52:06.552Z" },
-    { url = "https://files.pythonhosted.org/packages/de/a4/fd04bf807851197077d9cac9381d58f86d91c95c06cbaf9d3a776ac4467a/aiohttp-3.12.15-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0a146708808c9b7a988a4af3821379e379e0f0e5e466ca31a73dbdd0325b0263", size = 1637852, upload-time = "2025-07-29T05:52:08.975Z" },
-    { url = "https://files.pythonhosted.org/packages/98/03/29d626ca3bcdcafbd74b45d77ca42645a5c94d396f2ee3446880ad2405fb/aiohttp-3.12.15-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b7011a70b56facde58d6d26da4fec3280cc8e2a78c714c96b7a01a87930a9530", size = 1572852, upload-time = "2025-07-29T05:52:11.508Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/cd/b4777a9e204f4e01091091027e5d1e2fa86decd0fee5067bc168e4fa1e76/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:3bdd6e17e16e1dbd3db74d7f989e8af29c4d2e025f9828e6ef45fbdee158ec75", size = 1620813, upload-time = "2025-07-29T05:52:13.891Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/26/1a44a6e8417e84057beaf8c462529b9e05d4b53b8605784f1eb571f0ff68/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:57d16590a351dfc914670bd72530fd78344b885a00b250e992faea565b7fdc05", size = 1630951, upload-time = "2025-07-29T05:52:15.955Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/7f/10c605dbd01c40e2b27df7ef9004bec75d156f0705141e11047ecdfe264d/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:bc9a0f6569ff990e0bbd75506c8d8fe7214c8f6579cca32f0546e54372a3bb54", size = 1607595, upload-time = "2025-07-29T05:52:18.089Z" },
-    { url = "https://files.pythonhosted.org/packages/66/f6/2560dcb01731c1d7df1d34b64de95bc4b3ed02bb78830fd82299c1eb314e/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:536ad7234747a37e50e7b6794ea868833d5220b49c92806ae2d7e8a9d6b5de02", size = 1695194, upload-time = "2025-07-29T05:52:20.255Z" },
-    { url = "https://files.pythonhosted.org/packages/e7/02/ee105ae82dc2b981039fd25b0cf6eaa52b493731960f9bc861375a72b463/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:f0adb4177fa748072546fb650d9bd7398caaf0e15b370ed3317280b13f4083b0", size = 1710872, upload-time = "2025-07-29T05:52:22.769Z" },
-    { url = "https://files.pythonhosted.org/packages/88/16/70c4e42ed6a04f78fb58d1a46500a6ce560741d13afde2a5f33840746a5f/aiohttp-3.12.15-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:14954a2988feae3987f1eb49c706bff39947605f4b6fa4027c1d75743723eb09", size = 1640539, upload-time = "2025-07-29T05:52:25.733Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/1d/a7eb5fa8a6967117c5c0ad5ab4b1dec0d21e178c89aa08bc442a0b836392/aiohttp-3.12.15-cp39-cp39-win32.whl", hash = "sha256:b784d6ed757f27574dca1c336f968f4e81130b27595e458e69457e6878251f5d", size = 430164, upload-time = "2025-07-29T05:52:27.905Z" },
-    { url = "https://files.pythonhosted.org/packages/14/25/e0cf8793aedc41c6d7f2aad646a27e27bdacafe3b402bb373d7651c94d73/aiohttp-3.12.15-cp39-cp39-win_amd64.whl", hash = "sha256:86ceded4e78a992f835209e236617bffae649371c4a50d5e5a3987f237db84b8", size = 453370, upload-time = "2025-07-29T05:52:29.936Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/62/f1/8515650ac3121a9e55c7b217c60e7fae3e0134b5acfe65691781b5356929/aiohttp-3.13.0.tar.gz", hash = "sha256:378dbc57dd8cf341ce243f13fa1fa5394d68e2e02c15cd5f28eae35a70ec7f67", size = 7832348, upload-time = "2025-10-06T19:58:48.089Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/25/18/a3a9c9b7c8d400f71d1ff93c3e1520a5d53dba170f829ca9c6b2b070677b/aiohttp-3.13.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:ca69ec38adf5cadcc21d0b25e2144f6a25b7db7bea7e730bac25075bc305eff0", size = 734428, upload-time = "2025-10-06T19:54:40.285Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/02/f1eac06d78997e015030130ccf1c7cf864a919f97d77ff27e89c82fc3186/aiohttp-3.13.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:240f99f88a9a6beb53ebadac79a2e3417247aa756202ed234b1dbae13d248092", size = 491939, upload-time = "2025-10-06T19:54:42.113Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/db/5d65af7cbe5f302e23b1ea5cfc156cd0c7738a0d2db531a3837d2754de94/aiohttp-3.13.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a4676b978a9711531e7cea499d4cdc0794c617a1c0579310ab46c9fdf5877702", size = 487229, upload-time = "2025-10-06T19:54:43.978Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/d5/56c622ad3bd57ff4adc2b701f298dcc0408735a8af998cec1c66a9ce224e/aiohttp-3.13.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:48fcdd5bc771cbbab8ccc9588b8b6447f6a30f9fe00898b1a5107098e00d6793", size = 1666118, upload-time = "2025-10-06T19:54:46.569Z" },
+    { url = "https://files.pythonhosted.org/packages/44/16/db236671ec3758e3a6be6977009e74016470368012a58fea4b3799546549/aiohttp-3.13.0-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:eeea0cdd2f687e210c8f605f322d7b0300ba55145014a5dbe98bd4be6fff1f6c", size = 1633983, upload-time = "2025-10-06T19:54:48.244Z" },
+    { url = "https://files.pythonhosted.org/packages/19/ad/d96d7d7023e7f5215b8737cad21a7637f6d9d10fbfbfef0435d0277f71a2/aiohttp-3.13.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:10b3f01d5aeb632adaaf39c5e93f040a550464a768d54c514050c635adcbb9d0", size = 1725922, upload-time = "2025-10-06T19:54:49.885Z" },
+    { url = "https://files.pythonhosted.org/packages/88/d7/e8a5ba2bbd929ed587b2a8ea9390765daede2d8cd28dfae3a0773c6d3fbc/aiohttp-3.13.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a4dc0b83e25267f42ef065ea57653de4365b56d7bc4e4cfc94fabe56998f8ee6", size = 1813770, upload-time = "2025-10-06T19:54:51.648Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/ca/135c21e85ffeff66b80ecd8a647ca104f2e5a91c37dc86649244ddbf87ab/aiohttp-3.13.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:72714919ed9b90f030f761c20670e529c4af96c31bd000917dd0c9afd1afb731", size = 1667322, upload-time = "2025-10-06T19:54:53.668Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/38/348c4343052a400968dbf2051ee3dc222bdefd95af5874cf0f04cc7a8c92/aiohttp-3.13.0-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:564be41e85318403fdb176e9e5b3e852d528392f42f2c1d1efcbeeed481126d7", size = 1553270, upload-time = "2025-10-06T19:54:56.054Z" },
+    { url = "https://files.pythonhosted.org/packages/47/89/71cbda30f0900ab16084769960c467a355d6b1db51668fbb821c4a4ad5ed/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:84912962071087286333f70569362e10793f73f45c48854e6859df11001eb2d3", size = 1637087, upload-time = "2025-10-06T19:54:58.548Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/b1/5ff5fcaecccdcd5be7ff717cbde6e630760a8130e89167c3aa05b6b57707/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:90b570f1a146181c3d6ae8f755de66227ded49d30d050479b5ae07710f7894c5", size = 1643443, upload-time = "2025-10-06T19:55:00.856Z" },
+    { url = "https://files.pythonhosted.org/packages/87/e2/1d1f202f43c8be1956f05196159064cc05dc6842a33c1397cbb1b99610af/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:2d71ca30257ce756e37a6078b1dff2d9475fee13609ad831eac9a6531bea903b", size = 1695571, upload-time = "2025-10-06T19:55:03.006Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/b9/53c1df2991686f947a9651265757ea12c4afc29b351a249b73a0fc81dd3c/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:cd45eb70eca63f41bb156b7dffbe1a7760153b69892d923bdb79a74099e2ed90", size = 1539975, upload-time = "2025-10-06T19:55:04.839Z" },
+    { url = "https://files.pythonhosted.org/packages/93/24/345166f9c4cd2f5cc1d2173131998ee4adab0db8729126db32a7f91ed400/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:5ae3a19949a27982c7425a7a5a963c1268fdbabf0be15ab59448cbcf0f992519", size = 1712866, upload-time = "2025-10-06T19:55:06.905Z" },
+    { url = "https://files.pythonhosted.org/packages/09/f1/e8f70462848b74d49b3115050623ecbd697889713c2c93c96616da56b2de/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:ea6df292013c9f050cbf3f93eee9953d6e5acd9e64a0bf4ca16404bfd7aa9bcc", size = 1654058, upload-time = "2025-10-06T19:55:08.51Z" },
+    { url = "https://files.pythonhosted.org/packages/23/ba/47fd065510a8bfab5d5f6e1d97c0de672447c0a941c5021298bd7210afc3/aiohttp-3.13.0-cp310-cp310-win32.whl", hash = "sha256:3b64f22fbb6dcd5663de5ef2d847a5638646ef99112503e6f7704bdecb0d1c4d", size = 430230, upload-time = "2025-10-06T19:55:10.178Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/38/f5385cb79afa1f31bcaa3625a9e8d849b782edaeac09f894f46439e006a1/aiohttp-3.13.0-cp310-cp310-win_amd64.whl", hash = "sha256:f8d877aa60d80715b2afc565f0f1aea66565824c229a2d065b31670e09fed6d7", size = 453013, upload-time = "2025-10-06T19:55:11.623Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/db/df80cacac46cd548a736c5535b13cc18925cf6f9f83cd128cf3839842219/aiohttp-3.13.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:99eb94e97a42367fef5fc11e28cb2362809d3e70837f6e60557816c7106e2e20", size = 741374, upload-time = "2025-10-06T19:55:13.095Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/f9/2d6d93fd57ab4726e18a7cdab083772eda8302d682620fbf2aef48322351/aiohttp-3.13.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4696665b2713021c6eba3e2b882a86013763b442577fe5d2056a42111e732eca", size = 494956, upload-time = "2025-10-06T19:55:14.687Z" },
+    { url = "https://files.pythonhosted.org/packages/89/a6/e1c061b079fed04ffd6777950c82f2e8246fd08b7b3c4f56fdd47f697e5a/aiohttp-3.13.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3e6a38366f7f0d0f6ed7a1198055150c52fda552b107dad4785c0852ad7685d1", size = 491154, upload-time = "2025-10-06T19:55:16.661Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/4d/ee8913c0d2c7da37fdc98673a342b51611eaa0871682b37b8430084e35b5/aiohttp-3.13.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:aab715b1a0c37f7f11f9f1f579c6fbaa51ef569e47e3c0a4644fba46077a9409", size = 1745707, upload-time = "2025-10-06T19:55:18.376Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/70/26b2c97e8fa68644aec43d788940984c5f3b53a8d1468d5baaa328f809c9/aiohttp-3.13.0-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:7972c82bed87d7bd8e374b60a6b6e816d75ba4f7c2627c2d14eed216e62738e1", size = 1702404, upload-time = "2025-10-06T19:55:20.098Z" },
+    { url = "https://files.pythonhosted.org/packages/65/1e/c8aa3c293a0e8b18968b1b88e9bd8fb269eb67eb7449f504a4c3e175b159/aiohttp-3.13.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ca8313cb852af788c78d5afdea24c40172cbfff8b35e58b407467732fde20390", size = 1805519, upload-time = "2025-10-06T19:55:21.811Z" },
+    { url = "https://files.pythonhosted.org/packages/51/b6/a3753fe86249eb441768658cfc00f8c4e0913b255c13be00ddb8192775e1/aiohttp-3.13.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:6c333a2385d2a6298265f4b3e960590f787311b87f6b5e6e21bb8375914ef504", size = 1893904, upload-time = "2025-10-06T19:55:23.462Z" },
+    { url = "https://files.pythonhosted.org/packages/51/6d/7b1e020fe1d2a2be7cf0ce5e35922f345e3507cf337faa1a6563c42065c1/aiohttp-3.13.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:cc6d5fc5edbfb8041d9607f6a417997fa4d02de78284d386bea7ab767b5ea4f3", size = 1745043, upload-time = "2025-10-06T19:55:25.208Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/df/aad5dce268f9d4f29759c3eeb5fb5995c569d76abb267468dc1075218d5b/aiohttp-3.13.0-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:7ddedba3d0043349edc79df3dc2da49c72b06d59a45a42c1c8d987e6b8d175b8", size = 1604765, upload-time = "2025-10-06T19:55:27.157Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/19/a84a0e97b2da2224c8b85e1aef5cac834d07b2903c17bff1a6bdbc7041d2/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:23ca762140159417a6bbc959ca1927f6949711851e56f2181ddfe8d63512b5ad", size = 1721737, upload-time = "2025-10-06T19:55:28.854Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/61/ca6ad390128d964a08554fd63d6df5810fb5fbc7e599cb9e617f1729ae19/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:bfe824d6707a5dc3c5676685f624bc0c63c40d79dc0239a7fd6c034b98c25ebe", size = 1716052, upload-time = "2025-10-06T19:55:30.563Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/71/769e249e6625372c7d14be79b8b8c3b0592963a09793fb3d36758e60952c/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:3c11fa5dd2ef773a8a5a6daa40243d83b450915992eab021789498dc87acc114", size = 1783532, upload-time = "2025-10-06T19:55:32.798Z" },
+    { url = "https://files.pythonhosted.org/packages/66/64/b9cd03cdbb629bc492e4a744fbe96550a8340b0cd7a0cc4a9c90cfecd8d3/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:00fdfe370cffede3163ba9d3f190b32c0cfc8c774f6f67395683d7b0e48cdb8a", size = 1593072, upload-time = "2025-10-06T19:55:34.686Z" },
+    { url = "https://files.pythonhosted.org/packages/24/0e/87922c8cfdbd09f5e2197e9d87714a98c99c423560d44739e3af55400fe3/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:6475e42ef92717a678bfbf50885a682bb360a6f9c8819fb1a388d98198fdcb80", size = 1798613, upload-time = "2025-10-06T19:55:36.393Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/bb/a3adfe2af76e1ee9e3b5464522004b148b266bc99d7ec424ca7843d64a3c/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:77da5305a410910218b99f2a963092f4277d8a9c1f429c1ff1b026d1826bd0b6", size = 1737480, upload-time = "2025-10-06T19:55:38.043Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/53/e124dcbd64e6365602f3493fe37a11ca5b7ac0a40822a6e2bc8260cd08e0/aiohttp-3.13.0-cp311-cp311-win32.whl", hash = "sha256:2f9d9ea547618d907f2ee6670c9a951f059c5994e4b6de8dcf7d9747b420c820", size = 429824, upload-time = "2025-10-06T19:55:39.595Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/bd/485d98b372a2cd6998484a93ddd401ec6b6031657661c36846a10e2a1f6e/aiohttp-3.13.0-cp311-cp311-win_amd64.whl", hash = "sha256:0f19f7798996d4458c669bd770504f710014926e9970f4729cf55853ae200469", size = 454137, upload-time = "2025-10-06T19:55:41.617Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/95/7e8bdfa6e79099a086d59d42589492f1fe9d29aae3cefb58b676015ce278/aiohttp-3.13.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:1c272a9a18a5ecc48a7101882230046b83023bb2a662050ecb9bfcb28d9ab53a", size = 735585, upload-time = "2025-10-06T19:55:43.401Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/20/2f1d3ee06ee94eafe516810705219bff234d09f135d6951661661d5595ae/aiohttp-3.13.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:97891a23d7fd4e1afe9c2f4473e04595e4acb18e4733b910b6577b74e7e21985", size = 490613, upload-time = "2025-10-06T19:55:45.237Z" },
+    { url = "https://files.pythonhosted.org/packages/74/15/ab8600ef6dc1dcd599009a81acfed2ea407037e654d32e47e344e0b08c34/aiohttp-3.13.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:475bd56492ce5f4cffe32b5533c6533ee0c406d1d0e6924879f83adcf51da0ae", size = 489750, upload-time = "2025-10-06T19:55:46.937Z" },
+    { url = "https://files.pythonhosted.org/packages/33/59/752640c2b86ca987fe5703a01733b00d375e6cd2392bc7574489934e64e5/aiohttp-3.13.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c32ada0abb4bc94c30be2b681c42f058ab104d048da6f0148280a51ce98add8c", size = 1736812, upload-time = "2025-10-06T19:55:48.917Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/c6/dd6b86ddb852a7fdbcdc7a45b6bdc80178aef713c08279afcaee7a5a9f07/aiohttp-3.13.0-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:4af1f8877ca46ecdd0bc0d4a6b66d4b2bddc84a79e2e8366bc0d5308e76bceb8", size = 1698535, upload-time = "2025-10-06T19:55:50.75Z" },
+    { url = "https://files.pythonhosted.org/packages/33/e2/27c92d205b9e8cee7661670e8e9f187931b71e26d42796b153d2a0ba6949/aiohttp-3.13.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e04ab827ec4f775817736b20cdc8350f40327f9b598dec4e18c9ffdcbea88a93", size = 1766573, upload-time = "2025-10-06T19:55:53.106Z" },
+    { url = "https://files.pythonhosted.org/packages/df/6a/1fc1ad71d130a30f7a207d8d958a41224c29b834463b5185efb2dbff6ad4/aiohttp-3.13.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a6d9487b9471ec36b0faedf52228cd732e89be0a2bbd649af890b5e2ce422353", size = 1865229, upload-time = "2025-10-06T19:55:55.01Z" },
+    { url = "https://files.pythonhosted.org/packages/14/51/d0c1701a79fcb0109cff5304da16226581569b89a282d8e7f1549a7e3ec0/aiohttp-3.13.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2e66c57416352f36bf98f6641ddadd47c93740a22af7150d3e9a1ef6e983f9a8", size = 1750379, upload-time = "2025-10-06T19:55:57.219Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/3d/2ec4b934f85856de1c0c18e90adc8902adadbfac2b3c0b831bfeb7214fc8/aiohttp-3.13.0-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:469167d5372f5bb3aedff4fc53035d593884fff2617a75317740e885acd48b04", size = 1560798, upload-time = "2025-10-06T19:55:58.888Z" },
+    { url = "https://files.pythonhosted.org/packages/38/56/e23d9c3e13006e599fdce3851517c70279e177871e3e567d22cf3baf5d6c/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a9f3546b503975a69b547c9fd1582cad10ede1ce6f3e313a2f547c73a3d7814f", size = 1697552, upload-time = "2025-10-06T19:56:01.172Z" },
+    { url = "https://files.pythonhosted.org/packages/56/cb/caa32c2ccaeca0a3dc39129079fd2ad02f9406c3a5f7924340435b87d4cd/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:6b4174fcec98601f0cfdf308ee29a6ae53c55f14359e848dab4e94009112ee7d", size = 1718609, upload-time = "2025-10-06T19:56:03.102Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/c0/5911856fef9e40fd1ccbb8c54a90116875d5753a92c1cac66ce2059b390d/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:a533873a7a4ec2270fb362ee5a0d3b98752e4e1dc9042b257cd54545a96bd8ed", size = 1735887, upload-time = "2025-10-06T19:56:04.841Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/48/8d6f4757a24c02f0a454c043556593a00645d10583859f7156db44d8b7d3/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:ce887c5e54411d607ee0959cac15bb31d506d86a9bcaddf0b7e9d63325a7a802", size = 1553079, upload-time = "2025-10-06T19:56:07.197Z" },
+    { url = "https://files.pythonhosted.org/packages/39/fa/e82c9445e40b50e46770702b5b6ca2f767966d53e1a5eef03583ceac6df6/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:d871f6a30d43e32fc9252dc7b9febe1a042b3ff3908aa83868d7cf7c9579a59b", size = 1762750, upload-time = "2025-10-06T19:56:09.376Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/e6/9d30554e7f1e700bfeae4ab6b153d5dc7441606a9ec5e929288fa93a1477/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:222c828243b4789d79a706a876910f656fad4381661691220ba57b2ab4547865", size = 1717461, upload-time = "2025-10-06T19:56:11.551Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/e5/29cca547990a59ea54f0674fc01de98519fc628cfceeab6175711750eca7/aiohttp-3.13.0-cp312-cp312-win32.whl", hash = "sha256:682d2e434ff2f1108314ff7f056ce44e457f12dbed0249b24e106e385cf154b9", size = 424633, upload-time = "2025-10-06T19:56:13.316Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/68/46dd042d7bc62eab30bafdb8569f55ef125c3a88bb174270324224f8df56/aiohttp-3.13.0-cp312-cp312-win_amd64.whl", hash = "sha256:0a2be20eb23888df130214b91c262a90e2de1553d6fb7de9e9010cec994c0ff2", size = 451401, upload-time = "2025-10-06T19:56:15.188Z" },
+    { url = "https://files.pythonhosted.org/packages/40/26/6625e909ea770910d9901cb04336d0280a5a578daa352434eba5e482a1eb/aiohttp-3.13.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:fcc425fb6fd2a00c6d91c85d084c6b75a61bc8bc12159d08e17c5711df6c5ba4", size = 736970, upload-time = "2025-10-06T19:58:07.563Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/bc/ce07cc194f416f2d2539fec1237507b97496752f350738a051258be3dc08/aiohttp-3.13.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7c2c4c9ce834801651f81d6760d0a51035b8b239f58f298de25162fcf6f8bb64", size = 493252, upload-time = "2025-10-06T19:58:10.146Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/70/f51f30694aa4f9de3fad5b254609c263bf873a36dfa36f0370b37acb6b18/aiohttp-3.13.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:f91e8f9053a07177868e813656ec57599cd2a63238844393cd01bd69c2e40147", size = 488585, upload-time = "2025-10-06T19:58:12.438Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/89/4dd10ab758ba0dee025fc53345d24f5943f4d8741c1064ed3e92ca3274a8/aiohttp-3.13.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:df46d9a3d78ec19b495b1107bf26e4fcf97c900279901f4f4819ac5bb2a02a4c", size = 1660915, upload-time = "2025-10-06T19:58:14.359Z" },
+    { url = "https://files.pythonhosted.org/packages/49/72/f3a894182b625c7f242ab90ff95d6fbf4319d39dc854568a33c8768220ec/aiohttp-3.13.0-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3b1eb9871cbe43b6ca6fac3544682971539d8a1d229e6babe43446279679609d", size = 1625065, upload-time = "2025-10-06T19:58:16.754Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/3d/31f310c67b33e30d1f2fe11bf5aeb82d178189e8abfcf696bed69a9713cb/aiohttp-3.13.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:62a3cddf8d9a2eae1f79585fa81d32e13d0c509bb9e7ad47d33c83b45a944df7", size = 1721983, upload-time = "2025-10-06T19:58:19.275Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/6a/64f23ffd040409ac148a389be5c377f8edc4ab2aebdd601dbe786a30e05d/aiohttp-3.13.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0f735e680c323ee7e9ef8e2ea26425c7dbc2ede0086fa83ce9d7ccab8a089f26", size = 1810847, upload-time = "2025-10-06T19:58:21.419Z" },
+    { url = "https://files.pythonhosted.org/packages/51/64/9951b85fdf53aea1e9f41c715e9f4fc3e60f430fb58e0a46bc7837688e39/aiohttp-3.13.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6a51839f778b0e283b43cd82bb17f1835ee2cc1bf1101765e90ae886e53e751c", size = 1658909, upload-time = "2025-10-06T19:58:23.406Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/66/4d2c26a8a86b81a4493beb222457c6cba9f3853b6efa476a97edb4d52bb9/aiohttp-3.13.0-cp39-cp39-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:ac90cfab65bc281d6752f22db5fa90419e33220af4b4fa53b51f5948f414c0e7", size = 1551803, upload-time = "2025-10-06T19:58:25.568Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/9a/c6b4eb325f473f3a117fab90008348fc7094ccbaeeedf811d89738454996/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:62fd54f3e6f17976962ba67f911d62723c760a69d54f5d7b74c3ceb1a4e9ef8d", size = 1632205, upload-time = "2025-10-06T19:58:27.976Z" },
+    { url = "https://files.pythonhosted.org/packages/55/27/6fa2a583d45f711cf2f719f0d69274739a94f3b6d3deddfd19106d2ef3b9/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:cf2b60b65df05b6b2fa0d887f2189991a0dbf44a0dd18359001dc8fcdb7f1163", size = 1637447, upload-time = "2025-10-06T19:58:30.143Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/2a/ca6f2ac71b6c7e6f93ad42616e0a75660d8b9400977053126b7eaeaa5132/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:1ccedfe280e804d9a9d7fe8b8c4309d28e364b77f40309c86596baa754af50b1", size = 1692038, upload-time = "2025-10-06T19:58:32.227Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/fa/221c09563a11ce0d22408bf41e4bd8385f2093a22ceb60eeddd1f5019bab/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:ea01ffbe23df53ece0c8732d1585b3d6079bb8c9ee14f3745daf000051415a31", size = 1538478, upload-time = "2025-10-06T19:58:34.776Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/09/06b8b45874ba26d7b012b9b21b6593d9a7a0a4e3375da83c74eb044caafa/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:19ba8625fa69523627b67f7e9901b587a4952470f68814d79cdc5bc460e9b885", size = 1708317, upload-time = "2025-10-06T19:58:38.746Z" },
+    { url = "https://files.pythonhosted.org/packages/63/b2/ee8c8bae8d2f17a146089bba8130cf18eb0e929d4c5262beae944da82b3a/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:4b14bfae90598d331b5061fd15a7c290ea0c15b34aeb1cf620464bb5ec02a602", size = 1647241, upload-time = "2025-10-06T19:58:41.233Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/e4/455240c0433ebc7c9400c6014d5e4d09509d0aeddb8c9d22aff1f3d40500/aiohttp-3.13.0-cp39-cp39-win32.whl", hash = "sha256:cf7a4b976da219e726d0043fc94ae8169c0dba1d3a059b3c1e2c964bafc5a77d", size = 430922, upload-time = "2025-10-06T19:58:43.699Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/e6/f49ed1869b961f41aa0421d35401dd4ee4c8ad74b3da618f7bde1a8ddc29/aiohttp-3.13.0-cp39-cp39-win_amd64.whl", hash = "sha256:6b9697d15231aeaed4786f090c9c8bc3ab5f0e0a6da1e76c135a310def271020", size = 453958, upload-time = "2025-10-06T19:58:45.868Z" },
 ]
 
 [[package]]
@@ -155,11 +155,11 @@ wheels = [
 
 [[package]]
 name = "airportsdata"
-version = "20250811"
+version = "20250909"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/6b/61/3751adeb92d3197268ad5db00d6fb7ba9c2db72757f0d47462338c25dd0a/airportsdata-20250811.tar.gz", hash = "sha256:844b663c33e85c8f3b932c14c798ef314bec53bda3b14f5694eddc24c7ab4b25", size = 905387, upload-time = "2025-08-11T18:38:24.684Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/6d/c6/17ae8a65f7fa5bbbeee166f8070063eb8b70c89501a65c2e6885db61fc08/airportsdata-20250909.tar.gz", hash = "sha256:f39974fe1101817ced4ccf7c6ed336408469e5e778395d0a3e7a5112ec298f90", size = 907204, upload-time = "2025-09-09T01:07:31.256Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/15/52/f229acde71b1b9c05f06e504d3dcfcc91bc0614946a35d68cbc36b746dfa/airportsdata-20250811-py3-none-any.whl", hash = "sha256:96f37fcf3f7732350b6dd606d2a8faac7e3b610d0c0b2c1637a3d2fab6b5bf39", size = 912696, upload-time = "2025-08-11T18:38:22.426Z" },
+    { url = "https://files.pythonhosted.org/packages/16/87/59b570b9c4b014532777dc3baffc9bea10cf0cc8b232cf3c17e4bd0754a6/airportsdata-20250909-py3-none-any.whl", hash = "sha256:ce7dc6e1485afe3915e708212c7024ad158470c1c934e6a6cb217cf28b798ac7", size = 914391, upload-time = "2025-09-09T01:07:29.364Z" },
 ]
 
 [[package]]
@@ -173,25 +173,26 @@ wheels = [
 
 [[package]]
 name = "anthropic"
-version = "0.64.0"
+version = "0.69.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
     { name = "distro" },
+    { name = "docstring-parser" },
     { name = "httpx" },
     { name = "jiter" },
     { name = "pydantic" },
     { name = "sniffio" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/d8/4f/f2b880cba1a76f3acc7d5eb2ae217632eac1b8cef5ed3027493545c59eba/anthropic-0.64.0.tar.gz", hash = "sha256:3d496c91a63dff64f451b3e8e4b238a9640bf87b0c11d0b74ddc372ba5a3fe58", size = 427893, upload-time = "2025-08-13T17:09:49.915Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/c8/9d/9ad1778b95f15c5b04e7d328c1b5f558f1e893857b7c33cd288c19c0057a/anthropic-0.69.0.tar.gz", hash = "sha256:c604d287f4d73640f40bd2c0f3265a2eb6ce034217ead0608f6b07a8bc5ae5f2", size = 480622, upload-time = "2025-09-29T16:53:45.282Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a9/b2/2d268bcd5d6441df9dc0ebebc67107657edb8b0150d3fda1a5b81d1bec45/anthropic-0.64.0-py3-none-any.whl", hash = "sha256:6f5f7d913a6a95eb7f8e1bda4e75f76670e8acd8d4cd965e02e2a256b0429dd1", size = 297244, upload-time = "2025-08-13T17:09:47.908Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/38/75129688de5637eb5b383e5f2b1570a5cc3aecafa4de422da8eea4b90a6c/anthropic-0.69.0-py3-none-any.whl", hash = "sha256:1f73193040f33f11e27c2cd6ec25f24fe7c3f193dc1c5cde6b7a08b18a16bcc5", size = 337265, upload-time = "2025-09-29T16:53:43.686Z" },
 ]
 
 [[package]]
 name = "anyio"
-version = "4.10.0"
+version = "4.11.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "exceptiongroup", marker = "python_full_version < '3.11'" },
@@ -199,9 +200,18 @@ dependencies = [
     { name = "sniffio" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f1/b4/636b3b65173d3ce9a38ef5f0522789614e590dab6a8d505340a4efe4c567/anyio-4.10.0.tar.gz", hash = "sha256:3f3fae35c96039744587aa5b8371e7e8e603c0702999535961dd336026973ba6", size = 213252, upload-time = "2025-08-04T08:54:26.451Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/c6/78/7d432127c41b50bccba979505f272c16cbcadcc33645d5fa3a738110ae75/anyio-4.11.0.tar.gz", hash = "sha256:82a8d0b81e318cc5ce71a5f1f8b5c4e63619620b63141ef8c995fa0db95a57c4", size = 219094, upload-time = "2025-09-23T09:19:12.58Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6f/12/e5e0282d673bb9746bacfb6e2dba8719989d3660cdb2ea79aee9a9651afb/anyio-4.10.0-py3-none-any.whl", hash = "sha256:60e474ac86736bbfd6f210f7a61218939c318f43f9972497381f1c5e930ed3d1", size = 107213, upload-time = "2025-08-04T08:54:24.882Z" },
+    { url = "https://files.pythonhosted.org/packages/15/b3/9b1a8074496371342ec1e796a96f99c82c945a339cd81a8e73de28b4cf9e/anyio-4.11.0-py3-none-any.whl", hash = "sha256:0287e96f4d26d4149305414d4e3bc32f0dcd0862365a4bddea19d7a1ec38c4fc", size = 109097, upload-time = "2025-09-23T09:19:10.601Z" },
+]
+
+[[package]]
+name = "astor"
+version = "0.8.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/5a/21/75b771132fee241dfe601d39ade629548a9626d1d39f333fde31bc46febe/astor-0.8.1.tar.gz", hash = "sha256:6a6effda93f4e1ce9f618779b2dd1d9d84f1e32812c23a29b3fff6fd7f63fa5e", size = 35090, upload-time = "2019-12-10T01:50:35.51Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c3/88/97eef84f48fa04fbd6750e62dcceafba6c63c81b7ac1420856c8dcc0a3f9/astor-0.8.1-py2.py3-none-any.whl", hash = "sha256:070a54e890cefb5b3739d19f30f5a5ec840ffc9c50ffa7d23cc9fc1a38ebbfc5", size = 27488, upload-time = "2019-12-10T01:50:33.628Z" },
 ]
 
 [[package]]
@@ -228,11 +238,11 @@ wheels = [
 
 [[package]]
 name = "attrs"
-version = "25.3.0"
+version = "25.4.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/5a/b0/1367933a8532ee6ff8d63537de4f1177af4bff9f3e829baf7331f595bb24/attrs-25.3.0.tar.gz", hash = "sha256:75d7cefc7fb576747b2c81b4442d4d4a1ce0900973527c011d1030fd3bf4af1b", size = 812032, upload-time = "2025-03-13T11:10:22.779Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/6b/5c/685e6633917e101e5dcb62b9dd76946cbb57c26e133bae9e0cd36033c0a9/attrs-25.4.0.tar.gz", hash = "sha256:16d5969b87f0859ef33a48b35d55ac1be6e42ae49d5e853b597db70c35c57e11", size = 934251, upload-time = "2025-10-06T13:54:44.725Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/77/06/bb80f5f86020c4551da315d78b3ab75e8228f89f0162f2c3a819e407941a/attrs-25.3.0-py3-none-any.whl", hash = "sha256:427318ce031701fea540783410126f03899a97ffc6f61596ad581ac2e40e3bc3", size = 63815, upload-time = "2025-03-13T11:10:21.14Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/2a/7cc015f5b9f5db42b7d48157e23356022889fc354a2813c15934b7cb5c0e/attrs-25.4.0-py3-none-any.whl", hash = "sha256:adcf7e2a1fb3b36ac48d97835bb6d8ade15b8dcce26aba8bf1d14847b57a3373", size = 67615, upload-time = "2025-10-06T13:54:43.17Z" },
 ]
 
 [[package]]
@@ -262,22 +272,132 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/6f/e7/9220de010e015fdfb7ce22e7d9a846bfcd6e5d686c4fc555fa76a22846ba/beartype-0.15.0-py3-none-any.whl", hash = "sha256:52cd2edea72fdd84e4e7f8011a9e3007bf0125c3d6d7219e937b9d8868169177", size = 777565, upload-time = "2023-07-22T05:25:41.659Z" },
 ]
 
+[[package]]
+name = "blake3"
+version = "1.0.7"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions", marker = "python_full_version < '3.12'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/14/aa/e767c874392e95c42231a71072e6d47b9411b0c5fc5b221ed1304d63e4c1/blake3-1.0.7.tar.gz", hash = "sha256:228c617d4a92b45ef06c63236348001976245eb8efd561dbe181a0def524173e", size = 114008, upload-time = "2025-09-29T16:41:23.083Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ce/45/68acac32a1d8e41312c48ba0103515baf09dd79201deb6dd0262faa47eea/blake3-1.0.7-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:924f410e2d5f07a383473645b9c5da0b08220e1cbb54cfcf272ceaf53004e24b", size = 370989, upload-time = "2025-09-29T16:39:28.385Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/48/0e12f0f7603e69564247f4e1d5648e658fe1d3a6afd23a658985edbec672/blake3-1.0.7-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:10d2aa630238a6fa25a812d7d178f29b353c41eb34b2336dd11a7d2dce804bdf", size = 375310, upload-time = "2025-09-29T16:39:29.884Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/51/7d43e4e90aed90bccf72d3502bbea7a17d7eaabd8ce7f9d7f17de3bdddc6/blake3-1.0.7-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:385de4f30d90242b7b4bd2ce597d2cfdcd4d10292b410fc683c32e6bbf3d1acf", size = 448625, upload-time = "2025-09-29T16:39:31.448Z" },
+    { url = "https://files.pythonhosted.org/packages/47/ef/7ae2f8a1bac8e0201aa4e1416a4357b6f22092a51a36b0d8045e9e644f60/blake3-1.0.7-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bbe44233e96d39082eadb41cb3cd7f6f0ebca3948a5bb0529166afd965b091e2", size = 506388, upload-time = "2025-09-29T16:39:33.009Z" },
+    { url = "https://files.pythonhosted.org/packages/da/98/90bc7a158d9dfa734644f8474bd1a34ab1b7c2670ed3dbf3c5acab8f2da3/blake3-1.0.7-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:dc4000c194b3537b6ce97107f56ff881f9d8c7f91203698ce63978f597dfced5", size = 393443, upload-time = "2025-09-29T16:39:34.242Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/ee/642b3d7abbd09541717d4b96ebfbb2c79eacceefcffa99e44105be5ef249/blake3-1.0.7-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4147dbe2a6c76351cd2c122e55c06d8ae19cc94ef519b6a5be7d27d516c9e897", size = 387774, upload-time = "2025-09-29T16:39:35.455Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/3f/d73b00827135182bdbd8f071ef7f7bf0819fdad1837f91cc9baa75358f01/blake3-1.0.7-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:21fd54ff74959dedf148afee42a84b50e91a813ed4957bf0cceacfc1ebcf28fa", size = 549894, upload-time = "2025-09-29T16:39:37.039Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/5d/7c2b12832094b782ddb7f03936c289482b8a2180c76d97300550f8e78b54/blake3-1.0.7-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:f4f93d6d4e3e8b6366e706199bc71043d60cff235489c6a0fd9fd5aa1b9442d7", size = 554126, upload-time = "2025-09-29T16:39:38.35Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/0e/9e085fc2a19a4f81ad772c08193f4f95d3dea6ad25e713dafaec81e4f6ad/blake3-1.0.7-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e88d048e41035b0630bd5f2998224ebd50130d707496f19c7ec0b78946de1fc1", size = 371023, upload-time = "2025-09-29T16:39:44.627Z" },
+    { url = "https://files.pythonhosted.org/packages/18/88/63c5e8a3b4d3b1c54a6250c9d2674e2ed7bc6daf5d51d5633ef7d757ba52/blake3-1.0.7-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:532591a10c36e8de04d2caee128f4491136119f1224a11435b07c87f5da841b8", size = 375277, upload-time = "2025-09-29T16:39:45.829Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/15/cd5925106cc0f8d28dd141ecc131495717bb7b1a288197a92cade39c1a18/blake3-1.0.7-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a36ba39662bd504a3363d70e0654726e295c080e267f758a4801303991f30d00", size = 448409, upload-time = "2025-09-29T16:39:47.324Z" },
+    { url = "https://files.pythonhosted.org/packages/27/81/f06e2078cb1202c0aa86f3d6257a069a80a9acc370b1656b89073f5c2384/blake3-1.0.7-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:47d82d24acb22c7c68eeaac5bdf7dedb0cce2831cf00fcee92d459bf07ac51b1", size = 506520, upload-time = "2025-09-29T16:39:48.933Z" },
+    { url = "https://files.pythonhosted.org/packages/11/70/566513a733dfc1691df76d437c1e6b389261cd44c57bd361d9daddd68575/blake3-1.0.7-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c87822a2f7ea23aa5536256414d6c552db7d027941e3e25ab57807714877b40b", size = 393138, upload-time = "2025-09-29T16:39:50.508Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/58/16b74904befa9404bf3df2d6608b72b64d9d91209b7af491309f5644eb26/blake3-1.0.7-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8807d115255f235aa536269691581e59e9463bf1ece973e0e1e89d250a3f01c4", size = 387867, upload-time = "2025-09-29T16:39:51.694Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/2c/4930b54db10506ea65f0237578e4c69c3a98fd9bf2df01c8d1e193686308/blake3-1.0.7-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:dc97d1ed59edc26fb6e274cd00b4d3a3dc3d9ac7dbbcc0515dd2907c7a96cc6d", size = 550087, upload-time = "2025-09-29T16:39:53.316Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/f2/f94e27ab505ed5f25312f6d0b49cd8f5fd1258f162c27f19664b42eb04f4/blake3-1.0.7-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:3c515ab26dcc5d433bb5bf047b4e3d392ded742e7bc3255f1fa26014e20c93b2", size = 553940, upload-time = "2025-09-29T16:39:54.852Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/35/1fc595de42168569a83c49aff0aab94feef3c10c4152ad68f8046ad9839a/blake3-1.0.7-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5a7ef1c2c0da34c336e6ee4fa02b22cf142dc1a171e84d1cc46e9948c4e2c76d", size = 370790, upload-time = "2025-09-29T16:40:01.476Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/cb/58c5e33dde29b86e8a196d3acb2c1e25436ed58f48cd8ab5d4668cc085ec/blake3-1.0.7-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:46bfeeac655420f7421897a6d4754aafa9cd02bbcda869fc0bc358b19d90d864", size = 375207, upload-time = "2025-09-29T16:40:02.673Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/a9/4e59c99d8d4d137424e2175dbd67be48862a6543880dd9d78bc0f42be4ed/blake3-1.0.7-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:55095edd17183028c5be4fab204ee48e635ce78bd9f499ca1f95c0ed011fbf12", size = 448151, upload-time = "2025-09-29T16:40:03.888Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/55/d2e7d53a654193aad9ac27f3748f903791a9ebb88ba2fbfc1fd3e2629b59/blake3-1.0.7-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:05f8430d31aa11aa9b488b4a6ca07e35cc402b1c104d39ccbc99decdeb895130", size = 504463, upload-time = "2025-09-29T16:40:05.71Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/d0/3a2e712202ab85e972a0419caa3465153234adeb8429ba7a0bd4c23b936e/blake3-1.0.7-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a5d19e9a88274c85eb5c9985e58dd3f2cf56b1de0f71a77a4ba7fac2a2c70bf8", size = 393701, upload-time = "2025-09-29T16:40:06.928Z" },
+    { url = "https://files.pythonhosted.org/packages/db/30/addf808816833de5e497a1bf9d10221b1fc8048703b3de2c21ac6450326f/blake3-1.0.7-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4ef46ca98068d0153a8fd12215df8a24844800762fd5a3145658d89a3335990c", size = 387936, upload-time = "2025-09-29T16:40:08.214Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/70/48588e41c62d4ae1164acab51ba17bb15eca7499e18731645831a2e3459f/blake3-1.0.7-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:f4c327bdb172c08b935a98d1f527c07437d825ab317469f35ea43243b58898bc", size = 550142, upload-time = "2025-09-29T16:40:09.898Z" },
+    { url = "https://files.pythonhosted.org/packages/73/48/1aa83fd4671886eee2ab1497dcfa7089797975b927c79aefc418a2b928f3/blake3-1.0.7-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:bbfad4c447e8f910b4f2442e79768e5a3f6d9b1a31453fc780bab25e3e03d9de", size = 553949, upload-time = "2025-09-29T16:40:11.183Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/a6/ef3c2d14bf723173b72faeb3d79d487da1ed5160878d2f2e80911e821f8f/blake3-1.0.7-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:26bf2b2376cab4160f91759324dc5ebe3e4c61fd316781690bbba04c6f6d1574", size = 372798, upload-time = "2025-09-29T16:41:08.594Z" },
+    { url = "https://files.pythonhosted.org/packages/85/1e/2d347a865d890d139f6c4f99af1f012d1d3c0540c677a11eca6c04944fab/blake3-1.0.7-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:4959f47978cbad9a82774bc4efb02e2d4243e99db6f358ad5aa3df7704474bc0", size = 376973, upload-time = "2025-09-29T16:41:10.038Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/d2/6920923b207e62a644ff76ea23c6a077b1e6ef1e0b559350364448c72de6/blake3-1.0.7-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:abbc90d68f6918361821f107aec2de585d141e35deb362bc93854d2bfb105338", size = 450561, upload-time = "2025-09-29T16:41:11.381Z" },
+    { url = "https://files.pythonhosted.org/packages/df/d8/d6f4c495e1d5cf415d4115c6530cdfe0198031f1428ec27f0465917933d1/blake3-1.0.7-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d0a4fed89b9674f9651f1a9d88f8ae9b247e80d36e9b60222ea278cd2cf71b7d", size = 506581, upload-time = "2025-09-29T16:41:12.987Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/59/4ad0640d3c8b2b52f211902a5d814e5e9d9025ce6f6e2c6178c0d3238236/blake3-1.0.7-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8bdc842630cc6b27c8d038fce15f2b91684a95b65afb24f2075b241332dc0962", size = 394619, upload-time = "2025-09-29T16:41:14.349Z" },
+    { url = "https://files.pythonhosted.org/packages/64/34/21f8207496e8f26c9158438b7eae3cb62d270997d4c04e0363d58fcba408/blake3-1.0.7-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f6322f73e2255897ef31a3a10901d056f7215354dee68a3b8f8e18e4da2e694b", size = 389447, upload-time = "2025-09-29T16:41:15.687Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/26/f75a2dc988fbe2d1e632580d7b85275bd8dd711ce037fa1d44f866b874fa/blake3-1.0.7-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:e0d6f20b0e24c8438a565c0fcc81e376d5f1c0eaf28932cb229166736bb6e6ed", size = 551815, upload-time = "2025-09-29T16:41:17.287Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/66/388874447798e68e0b486e5f9e9cf069d27bff78f0f2132d6dd51c16fba7/blake3-1.0.7-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:f30a9c8c8f4c9423cb76ccce17f33e7a3f6ff5cde2d204591ff9a13b826aa129", size = 555876, upload-time = "2025-09-29T16:41:18.906Z" },
+]
+
 [[package]]
 name = "cachetools"
-version = "5.5.2"
+version = "6.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/9d/61/e4fad8155db4a04bfb4734c7c8ff0882f078f24294d42798b3568eb63bff/cachetools-6.2.0.tar.gz", hash = "sha256:38b328c0889450f05f5e120f56ab68c8abaf424e1275522b138ffc93253f7e32", size = 30988, upload-time = "2025-08-25T18:57:30.924Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6c/56/3124f61d37a7a4e7cc96afc5492c78ba0cb551151e530b54669ddd1436ef/cachetools-6.2.0-py3-none-any.whl", hash = "sha256:1c76a8960c0041fcc21097e357f882197c79da0dbff766e7317890a65d7d8ba6", size = 11276, upload-time = "2025-08-25T18:57:29.684Z" },
+]
+
+[[package]]
+name = "cbor2"
+version = "5.7.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/6c/81/3747dad6b14fa2cf53fcf10548cf5aea6913e96fab41a3c198676f8948a5/cachetools-5.5.2.tar.gz", hash = "sha256:1a661caa9175d26759571b2e19580f9d6393969e5dfca11fdb1f947a23e640d4", size = 28380, upload-time = "2025-02-20T21:01:19.524Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/3a/89/01df16cdc9c60c07956756c90fe92c684021003079e358a78e213bce45a2/cbor2-5.7.0.tar.gz", hash = "sha256:3f6d843f4db4d0ec501c46453c22a4fbebb1abfb5b740e1bcab34c615cd7406b", size = 102374, upload-time = "2025-08-14T08:59:47.294Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/72/76/20fa66124dbe6be5cafeb312ece67de6b61dd91a0247d1ea13db4ebb33c2/cachetools-5.5.2-py3-none-any.whl", hash = "sha256:d26a22bcc62eb95c3beabd9f1ee5e820d3d2704fe2967cbe350e20c8ffcd3f0a", size = 10080, upload-time = "2025-02-20T21:01:16.647Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/b7/17fe6e502b9884f3a7bdf6e9331688e290a546110dd14575aa09527b238f/cbor2-5.7.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:36b6e229b36147051ea1063ba0cd6225bfe6b5398ca0ac7b33fa91407ca75081", size = 254198, upload-time = "2025-08-14T08:58:40.293Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/fb/6cc1dd51beb35f899f9f6621fd0df9ec733e1f4c27f334212107f7a05822/cbor2-5.7.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:06f8e9d8fa6d0ed7b8a81f4613a60e740be9bd5087de620e3b64007832dfd815", size = 247202, upload-time = "2025-08-14T08:58:42.101Z" },
+    { url = "https://files.pythonhosted.org/packages/34/2d/e51cd823cb918eed3180bbe0169955ccefa03817c0e4788024e9e56d3657/cbor2-5.7.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:90c1236e5bfad19493183f9a9ecd3705e3ad3eb02fce6d6381c12ece86147b15", size = 250001, upload-time = "2025-08-14T08:58:43.82Z" },
+    { url = "https://files.pythonhosted.org/packages/de/45/79a0320a4562537df3e589526d636963a06a95a66d7b6c366bdc8d9b3755/cbor2-5.7.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:afc6c6611b7529136ea4a04cf8711786640b687859c9688ae172726c305a83a1", size = 243743, upload-time = "2025-08-14T08:58:45.618Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/52/6092e45a6b164d22bc2828aa14f422fbcf5af4bd9201a3e27b62b512f828/cbor2-5.7.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5fc1005b412ace94bbf905a8c4214e639557568551d9b5474645789e976e91e4", size = 261081, upload-time = "2025-08-14T08:58:53.747Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/12/9eef73bcfcdd110df7e585f8034399d1cf2006dc7576487e0509ba50e4f8/cbor2-5.7.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d981d691dd721dd712fc824d04a01347955a206107fbee1d75803fa58de172c7", size = 254132, upload-time = "2025-08-14T08:58:55.465Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/e1/00bc92353ecacef0b8eaba285168ca517c18539c12ddebe06a9bbab03e47/cbor2-5.7.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b3b9730928163d02e7f1267e48a39ed75560ff3e56cdc6768d670d3e96028623", size = 256825, upload-time = "2025-08-14T08:58:57.239Z" },
+    { url = "https://files.pythonhosted.org/packages/25/01/454c93f360c404a3e6531ce0e2e4cc5818c5746a10b1cb295b6abd3057cd/cbor2-5.7.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:01cc4d840943b4c2e468b8560830235d044a8cb45e5d66ba3ae933c3e21b8d49", size = 250639, upload-time = "2025-08-14T08:58:58.594Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/6a/9aed5b716407c1d48425ba55c6022a01a9abdbf58a691f50416461fa371d/cbor2-5.7.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:16bea83598a1eeedbd50c2e9fdf3685bae78ca9d9ec8cd8010777db14a315578", size = 285685, upload-time = "2025-08-14T08:59:05.165Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/6e/3499eaa2b858c7695a447b6311303f06ffc90fc2c45851337121661f1f5c/cbor2-5.7.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8e837825a16e60ace6e284095aa9fbe504bf87a8f4494bf7d95931e37fb01a70", size = 284948, upload-time = "2025-08-14T08:59:06.64Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/3e/ae67866ef65717665e0acf2873d466c5d4a1d965b0d0348f2269b73f28fb/cbor2-5.7.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:27396c5e275ff7c7cd87fe8aaadf781e6194903921f250934af7c86d5efec82e", size = 276375, upload-time = "2025-08-14T08:59:08.845Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/3d/2f8e9671111661dd571de206344ecd7706f6d458aab191e06834c89aa58e/cbor2-5.7.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:c84bfef78c4e9c81eb0a10cec340222ba4e39498a63fc2e3d5f982a3f4efa4a7", size = 277680, upload-time = "2025-08-14T08:59:10.292Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/40/dd8123fef6b18220f78e64a4bb4f8e4255c516db390ae8e8db0084474503/cbor2-5.7.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f94dbfd77ae591010429d645e4e7fd4f387183193f14c4f645dd7ba03668f9a9", size = 252958, upload-time = "2025-08-14T08:59:38.279Z" },
+    { url = "https://files.pythonhosted.org/packages/df/1a/644bf42929915875212744100131200bc17cf526235c28747c78b30588db/cbor2-5.7.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:cd36c935adf56ccf83c6b85949cb2ac8b991672d241ec1f645b557ae4730221b", size = 246023, upload-time = "2025-08-14T08:59:39.645Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/c1/32ad36a8e036b10a49412c0e55a3f96ecf1bbfc7b640ee80cca011731eb2/cbor2-5.7.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1198cbfa5194410b35b7fb5c0e58a0ee22c20b44197605e26f14e320e1750c5e", size = 248762, upload-time = "2025-08-14T08:59:41.048Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/db/847c604f876bb5990707b42e52c00960bdb81df800e4dfa81050d5f4ef5c/cbor2-5.7.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:78dca93abdbea701caa068e412e54923f3b4d117f5b46b39f44456e01b047534", size = 242520, upload-time = "2025-08-14T08:59:42.406Z" },
+    { url = "https://files.pythonhosted.org/packages/41/cc/0ce73676d2a0c9e5a9330c301940c50eb325dacf5f6d9690fd43a8817fe9/cbor2-5.7.0-py3-none-any.whl", hash = "sha256:a871e7a6f7cba1ddb02503ea974f15f6524c95078fbfe0b860fd4193d7c8f27a", size = 23828, upload-time = "2025-08-14T08:59:46.129Z" },
 ]
 
 [[package]]
 name = "certifi"
-version = "2025.8.3"
+version = "2025.10.5"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/dc/67/960ebe6bf230a96cda2e0abcf73af550ec4f090005363542f0765df162e0/certifi-2025.8.3.tar.gz", hash = "sha256:e564105f78ded564e3ae7c923924435e1daa7463faeab5bb932bc53ffae63407", size = 162386, upload-time = "2025-08-03T03:07:47.08Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/4c/5b/b6ce21586237c77ce67d01dc5507039d444b630dd76611bbca2d8e5dcd91/certifi-2025.10.5.tar.gz", hash = "sha256:47c09d31ccf2acf0be3f701ea53595ee7e0b8fa08801c6624be771df09ae7b43", size = 164519, upload-time = "2025-10-05T04:12:15.808Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e5/48/1549795ba7742c948d2ad169c1c8cdbae65bc450d6cd753d124b17c8cd32/certifi-2025.8.3-py3-none-any.whl", hash = "sha256:f6c12493cfb1b06ba2ff328595af9350c65d6644968e5d3a2ffd78699af217a5", size = 161216, upload-time = "2025-08-03T03:07:45.777Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/37/af0d2ef3967ac0d6113837b44a4f0bfe1328c2b9763bd5b1744520e5cfed/certifi-2025.10.5-py3-none-any.whl", hash = "sha256:0f212c2744a9bb6de0c56639a6f68afe01ecd92d91f14ae897c4fe7bbeeef0de", size = 163286, upload-time = "2025-10-05T04:12:14.03Z" },
+]
+
+[[package]]
+name = "cffi"
+version = "2.0.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pycparser", marker = "implementation_name != 'PyPy'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/eb/56/b1ba7935a17738ae8453301356628e8147c79dbb825bcbc73dc7401f9846/cffi-2.0.0.tar.gz", hash = "sha256:44d1b5909021139fe36001ae048dbdde8214afa20200eda0f64c068cac5d5529", size = 523588, upload-time = "2025-09-08T23:24:04.541Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/50/bd/b1a6362b80628111e6653c961f987faa55262b4002fcec42308cad1db680/cffi-2.0.0-cp310-cp310-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:53f77cbe57044e88bbd5ed26ac1d0514d2acf0591dd6bb02a3ae37f76811b80c", size = 208811, upload-time = "2025-09-08T23:22:12.267Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/27/6933a8b2562d7bd1fb595074cf99cc81fc3789f6a6c05cdabb46284a3188/cffi-2.0.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:3e837e369566884707ddaf85fc1744b47575005c0a229de3327f8f9a20f4efeb", size = 216402, upload-time = "2025-09-08T23:22:13.455Z" },
+    { url = "https://files.pythonhosted.org/packages/05/eb/b86f2a2645b62adcfff53b0dd97e8dfafb5c8aa864bd0d9a2c2049a0d551/cffi-2.0.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:5eda85d6d1879e692d546a078b44251cdd08dd1cfb98dfb77b670c97cee49ea0", size = 203217, upload-time = "2025-09-08T23:22:14.596Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/e0/6cbe77a53acf5acc7c08cc186c9928864bd7c005f9efd0d126884858a5fe/cffi-2.0.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:9332088d75dc3241c702d852d4671613136d90fa6881da7d770a483fd05248b4", size = 203079, upload-time = "2025-09-08T23:22:15.769Z" },
+    { url = "https://files.pythonhosted.org/packages/98/29/9b366e70e243eb3d14a5cb488dfd3a0b6b2f1fb001a203f653b93ccfac88/cffi-2.0.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fc7de24befaeae77ba923797c7c87834c73648a05a4bde34b3b7e5588973a453", size = 216475, upload-time = "2025-09-08T23:22:17.427Z" },
+    { url = "https://files.pythonhosted.org/packages/21/7a/13b24e70d2f90a322f2900c5d8e1f14fa7e2a6b3332b7309ba7b2ba51a5a/cffi-2.0.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:cf364028c016c03078a23b503f02058f1814320a56ad535686f90565636a9495", size = 218829, upload-time = "2025-09-08T23:22:19.069Z" },
+    { url = "https://files.pythonhosted.org/packages/60/99/c9dc110974c59cc981b1f5b66e1d8af8af764e00f0293266824d9c4254bc/cffi-2.0.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e11e82b744887154b182fd3e7e8512418446501191994dbf9c9fc1f32cc8efd5", size = 211211, upload-time = "2025-09-08T23:22:20.588Z" },
+    { url = "https://files.pythonhosted.org/packages/49/72/ff2d12dbf21aca1b32a40ed792ee6b40f6dc3a9cf1644bd7ef6e95e0ac5e/cffi-2.0.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:8ea985900c5c95ce9db1745f7933eeef5d314f0565b27625d9a10ec9881e1bfb", size = 218036, upload-time = "2025-09-08T23:22:22.143Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/b7/1200d354378ef52ec227395d95c2576330fd22a869f7a70e88e1447eb234/cffi-2.0.0-cp311-cp311-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:baf5215e0ab74c16e2dd324e8ec067ef59e41125d3eade2b863d294fd5035c92", size = 209613, upload-time = "2025-09-08T23:22:29.475Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/56/6033f5e86e8cc9bb629f0077ba71679508bdf54a9a5e112a3c0b91870332/cffi-2.0.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:730cacb21e1bdff3ce90babf007d0a0917cc3e6492f336c2f0134101e0944f93", size = 216476, upload-time = "2025-09-08T23:22:31.063Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/7f/55fecd70f7ece178db2f26128ec41430d8720f2d12ca97bf8f0a628207d5/cffi-2.0.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:6824f87845e3396029f3820c206e459ccc91760e8fa24422f8b0c3d1731cbec5", size = 203374, upload-time = "2025-09-08T23:22:32.507Z" },
+    { url = "https://files.pythonhosted.org/packages/84/ef/a7b77c8bdc0f77adc3b46888f1ad54be8f3b7821697a7b89126e829e676a/cffi-2.0.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:9de40a7b0323d889cf8d23d1ef214f565ab154443c42737dfe52ff82cf857664", size = 202597, upload-time = "2025-09-08T23:22:34.132Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/91/500d892b2bf36529a75b77958edfcd5ad8e2ce4064ce2ecfeab2125d72d1/cffi-2.0.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:8941aaadaf67246224cee8c3803777eed332a19d909b47e29c9842ef1e79ac26", size = 215574, upload-time = "2025-09-08T23:22:35.443Z" },
+    { url = "https://files.pythonhosted.org/packages/44/64/58f6255b62b101093d5df22dcb752596066c7e89dd725e0afaed242a61be/cffi-2.0.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:a05d0c237b3349096d3981b727493e22147f934b20f6f125a3eba8f994bec4a9", size = 218971, upload-time = "2025-09-08T23:22:36.805Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/49/fa72cebe2fd8a55fbe14956f9970fe8eb1ac59e5df042f603ef7c8ba0adc/cffi-2.0.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:94698a9c5f91f9d138526b48fe26a199609544591f859c870d477351dc7b2414", size = 211972, upload-time = "2025-09-08T23:22:38.436Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/28/dd0967a76aab36731b6ebfe64dec4e981aff7e0608f60c2d46b46982607d/cffi-2.0.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:5fed36fccc0612a53f1d4d9a816b50a36702c28a2aa880cb8a122b3466638743", size = 217078, upload-time = "2025-09-08T23:22:39.776Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/df/a4f0fbd47331ceeba3d37c2e51e9dfc9722498becbeec2bd8bc856c9538a/cffi-2.0.0-cp312-cp312-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:21d1152871b019407d8ac3985f6775c079416c282e431a4da6afe7aefd2bccbe", size = 212529, upload-time = "2025-09-08T23:22:47.349Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/72/12b5f8d3865bf0f87cf1404d8c374e7487dcf097a1c91c436e72e6badd83/cffi-2.0.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:b21e08af67b8a103c71a250401c78d5e0893beff75e28c53c98f4de42f774062", size = 220097, upload-time = "2025-09-08T23:22:48.677Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/95/7a135d52a50dfa7c882ab0ac17e8dc11cec9d55d2c18dda414c051c5e69e/cffi-2.0.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:1e3a615586f05fc4065a8b22b8152f0c1b00cdbc60596d187c2a74f9e3036e4e", size = 207983, upload-time = "2025-09-08T23:22:50.06Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/c8/15cb9ada8895957ea171c62dc78ff3e99159ee7adb13c0123c001a2546c1/cffi-2.0.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:81afed14892743bbe14dacb9e36d9e0e504cd204e0b165062c488942b9718037", size = 206519, upload-time = "2025-09-08T23:22:51.364Z" },
+    { url = "https://files.pythonhosted.org/packages/78/2d/7fa73dfa841b5ac06c7b8855cfc18622132e365f5b81d02230333ff26e9e/cffi-2.0.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3e17ed538242334bf70832644a32a7aae3d83b57567f9fd60a26257e992b79ba", size = 219572, upload-time = "2025-09-08T23:22:52.902Z" },
+    { url = "https://files.pythonhosted.org/packages/07/e0/267e57e387b4ca276b90f0434ff88b2c2241ad72b16d31836adddfd6031b/cffi-2.0.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3925dd22fa2b7699ed2617149842d2e6adde22b262fcbfada50e3d195e4b3a94", size = 222963, upload-time = "2025-09-08T23:22:54.518Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/75/1f2747525e06f53efbd878f4d03bac5b859cbc11c633d0fb81432d98a795/cffi-2.0.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:2c8f814d84194c9ea681642fd164267891702542f028a15fc97d4674b6206187", size = 221361, upload-time = "2025-09-08T23:22:55.867Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/13/c92e36358fbcc39cf0962e83223c9522154ee8630e1df7c0b3a39a8124e2/cffi-2.0.0-cp39-cp39-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:4647afc2f90d1ddd33441e5b0e85b16b12ddec4fca55f0d9671fef036ecca27c", size = 208813, upload-time = "2025-09-08T23:23:51.263Z" },
+    { url = "https://files.pythonhosted.org/packages/15/12/a7a79bd0df4c3bff744b2d7e52cc1b68d5e7e427b384252c42366dc1ecbc/cffi-2.0.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:3f4d46d8b35698056ec29bca21546e1551a205058ae1a181d871e278b0b28165", size = 216498, upload-time = "2025-09-08T23:23:52.494Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/ad/5c51c1c7600bdd7ed9a24a203ec255dccdd0ebf4527f7b922a0bde2fb6ed/cffi-2.0.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:e6e73b9e02893c764e7e8d5bb5ce277f1a009cd5243f8228f75f842bf937c534", size = 203243, upload-time = "2025-09-08T23:23:53.836Z" },
+    { url = "https://files.pythonhosted.org/packages/32/f2/81b63e288295928739d715d00952c8c6034cb6c6a516b17d37e0c8be5600/cffi-2.0.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:cb527a79772e5ef98fb1d700678fe031e353e765d1ca2d409c92263c6d43e09f", size = 203158, upload-time = "2025-09-08T23:23:55.169Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/74/cc4096ce66f5939042ae094e2e96f53426a979864aa1f96a621ad128be27/cffi-2.0.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:61d028e90346df14fedc3d1e5441df818d095f3b87d286825dfcbd6459b7ef63", size = 216548, upload-time = "2025-09-08T23:23:56.506Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/be/f6424d1dc46b1091ffcc8964fa7c0ab0cd36839dd2761b49c90481a6ba1b/cffi-2.0.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:0f6084a0ea23d05d20c3edcda20c3d006f9b6f3fefeac38f59262e10cef47ee2", size = 218897, upload-time = "2025-09-08T23:23:57.825Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/e0/dda537c2309817edf60109e39265f24f24aa7f050767e22c98c53fe7f48b/cffi-2.0.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:1cd13c99ce269b3ed80b417dcd591415d3372bcac067009b6e0f59c7d4015e65", size = 211249, upload-time = "2025-09-08T23:23:59.139Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/e7/7c769804eb75e4c4b35e658dba01de1640a351a9653c3d49ca89d16ccc91/cffi-2.0.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:89472c9762729b5ae1ad974b777416bfda4ac5642423fa93bd57a09204712322", size = 218041, upload-time = "2025-09-08T23:24:00.496Z" },
 ]
 
 [[package]]
@@ -385,8 +505,8 @@ resolution-markers = [
 ]
 dependencies = [
     { name = "absl-py", marker = "python_full_version >= '3.11'" },
-    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "toolz", marker = "python_full_version >= '3.11'" },
     { name = "typing-extensions", marker = "python_full_version >= '3.11'" },
@@ -413,7 +533,7 @@ wheels = [
 
 [[package]]
 name = "click"
-version = "8.2.1"
+version = "8.3.0"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -423,9 +543,9 @@ resolution-markers = [
 dependencies = [
     { name = "colorama", marker = "python_full_version >= '3.10' and sys_platform == 'win32'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/60/6c/8ca2efa64cf75a977a0d7fac081354553ebe483345c734fb6b6515d96bbc/click-8.2.1.tar.gz", hash = "sha256:27c491cc05d968d271d5a1db13e3b5a184636d9d930f148c50b038f0d0646202", size = 286342, upload-time = "2025-05-20T23:19:49.832Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/46/61/de6cd827efad202d7057d93e0fed9294b96952e188f7384832791c7b2254/click-8.3.0.tar.gz", hash = "sha256:e7b8232224eba16f4ebe410c25ced9f7875cb5f3263ffc93cc3e8da705e229c4", size = 276943, upload-time = "2025-09-18T17:32:23.696Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/85/32/10bb5764d90a8eee674e9dc6f4db6a0ab47c8c4d0d83c27f7c39ac415a4d/click-8.2.1-py3-none-any.whl", hash = "sha256:61a3265b914e850b85317d0b3109c7f8cd35a670f963866005d6ef1d5175a12b", size = 102215, upload-time = "2025-05-20T23:19:47.796Z" },
+    { url = "https://files.pythonhosted.org/packages/db/d3/9dcc0f5797f070ec8edf30fbadfb200e71d9db6b84d211e3b2085a7589a0/click-8.3.0-py3-none-any.whl", hash = "sha256:9b9f285302c6e3064f4330c05f05b81945b2a39544279343e6e7c5f27a9baddc", size = 107295, upload-time = "2025-09-18T17:32:22.42Z" },
 ]
 
 [[package]]
@@ -446,55 +566,78 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6", size = 25335, upload-time = "2022-10-25T02:36:20.889Z" },
 ]
 
+[[package]]
+name = "compressed-tensors"
+version = "0.11.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "frozendict" },
+    { name = "pydantic" },
+    { name = "torch" },
+    { name = "transformers" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b8/99/3fdabfc95609d6efdf02fa7f1ed0245524cb1209d3d4a17109d3205d2eed/compressed_tensors-0.11.0.tar.gz", hash = "sha256:95ddf19699f775df6494dd864e5f52e8a24f8015496520190c1a22c6cfc44b1f", size = 187566, upload-time = "2025-08-19T18:59:31.854Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d2/81/e3073017a8f5c75169e79108eda209e6089e3f96c9f197d307cbda7df71c/compressed_tensors-0.11.0-py3-none-any.whl", hash = "sha256:e1cbc46e1ae032b7ceea915fe18c8d2de5a54d3a50a607969b6bdfe703b6cb83", size = 179951, upload-time = "2025-08-19T18:59:29.308Z" },
+]
+
 [[package]]
 name = "coverage"
-version = "7.10.6"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/14/70/025b179c993f019105b79575ac6edb5e084fb0f0e63f15cdebef4e454fb5/coverage-7.10.6.tar.gz", hash = "sha256:f644a3ae5933a552a29dbb9aa2f90c677a875f80ebea028e5a52a4f429044b90", size = 823736, upload-time = "2025-08-29T15:35:16.668Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/a8/1d/2e64b43d978b5bd184e0756a41415597dfef30fcbd90b747474bd749d45f/coverage-7.10.6-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:70e7bfbd57126b5554aa482691145f798d7df77489a177a6bef80de78860a356", size = 217025, upload-time = "2025-08-29T15:32:57.169Z" },
-    { url = "https://files.pythonhosted.org/packages/23/62/b1e0f513417c02cc10ef735c3ee5186df55f190f70498b3702d516aad06f/coverage-7.10.6-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e41be6f0f19da64af13403e52f2dec38bbc2937af54df8ecef10850ff8d35301", size = 217419, upload-time = "2025-08-29T15:32:59.908Z" },
-    { url = "https://files.pythonhosted.org/packages/e7/16/b800640b7a43e7c538429e4d7223e0a94fd72453a1a048f70bf766f12e96/coverage-7.10.6-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:c61fc91ab80b23f5fddbee342d19662f3d3328173229caded831aa0bd7595460", size = 244180, upload-time = "2025-08-29T15:33:01.608Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/6f/5e03631c3305cad187eaf76af0b559fff88af9a0b0c180d006fb02413d7a/coverage-7.10.6-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:10356fdd33a7cc06e8051413140bbdc6f972137508a3572e3f59f805cd2832fd", size = 245992, upload-time = "2025-08-29T15:33:03.239Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/a1/f30ea0fb400b080730125b490771ec62b3375789f90af0bb68bfb8a921d7/coverage-7.10.6-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:80b1695cf7c5ebe7b44bf2521221b9bb8cdf69b1f24231149a7e3eb1ae5fa2fb", size = 247851, upload-time = "2025-08-29T15:33:04.603Z" },
-    { url = "https://files.pythonhosted.org/packages/02/8e/cfa8fee8e8ef9a6bb76c7bef039f3302f44e615d2194161a21d3d83ac2e9/coverage-7.10.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:2e4c33e6378b9d52d3454bd08847a8651f4ed23ddbb4a0520227bd346382bbc6", size = 245891, upload-time = "2025-08-29T15:33:06.176Z" },
-    { url = "https://files.pythonhosted.org/packages/93/a9/51be09b75c55c4f6c16d8d73a6a1d46ad764acca0eab48fa2ffaef5958fe/coverage-7.10.6-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:c8a3ec16e34ef980a46f60dc6ad86ec60f763c3f2fa0db6d261e6e754f72e945", size = 243909, upload-time = "2025-08-29T15:33:07.74Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/a6/ba188b376529ce36483b2d585ca7bdac64aacbe5aa10da5978029a9c94db/coverage-7.10.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:7d79dabc0a56f5af990cc6da9ad1e40766e82773c075f09cc571e2076fef882e", size = 244786, upload-time = "2025-08-29T15:33:08.965Z" },
-    { url = "https://files.pythonhosted.org/packages/d0/4c/37ed872374a21813e0d3215256180c9a382c3f5ced6f2e5da0102fc2fd3e/coverage-7.10.6-cp310-cp310-win32.whl", hash = "sha256:86b9b59f2b16e981906e9d6383eb6446d5b46c278460ae2c36487667717eccf1", size = 219521, upload-time = "2025-08-29T15:33:10.599Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/36/9311352fdc551dec5b973b61f4e453227ce482985a9368305880af4f85dd/coverage-7.10.6-cp310-cp310-win_amd64.whl", hash = "sha256:e132b9152749bd33534e5bd8565c7576f135f157b4029b975e15ee184325f528", size = 220417, upload-time = "2025-08-29T15:33:11.907Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/16/2bea27e212c4980753d6d563a0803c150edeaaddb0771a50d2afc410a261/coverage-7.10.6-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c706db3cabb7ceef779de68270150665e710b46d56372455cd741184f3868d8f", size = 217129, upload-time = "2025-08-29T15:33:13.575Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/51/e7159e068831ab37e31aac0969d47b8c5ee25b7d307b51e310ec34869315/coverage-7.10.6-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:8e0c38dc289e0508ef68ec95834cb5d2e96fdbe792eaccaa1bccac3966bbadcc", size = 217532, upload-time = "2025-08-29T15:33:14.872Z" },
-    { url = "https://files.pythonhosted.org/packages/e7/c0/246ccbea53d6099325d25cd208df94ea435cd55f0db38099dd721efc7a1f/coverage-7.10.6-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:752a3005a1ded28f2f3a6e8787e24f28d6abe176ca64677bcd8d53d6fe2ec08a", size = 247931, upload-time = "2025-08-29T15:33:16.142Z" },
-    { url = "https://files.pythonhosted.org/packages/7d/fb/7435ef8ab9b2594a6e3f58505cc30e98ae8b33265d844007737946c59389/coverage-7.10.6-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:689920ecfd60f992cafca4f5477d55720466ad2c7fa29bb56ac8d44a1ac2b47a", size = 249864, upload-time = "2025-08-29T15:33:17.434Z" },
-    { url = "https://files.pythonhosted.org/packages/51/f8/d9d64e8da7bcddb094d511154824038833c81e3a039020a9d6539bf303e9/coverage-7.10.6-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ec98435796d2624d6905820a42f82149ee9fc4f2d45c2c5bc5a44481cc50db62", size = 251969, upload-time = "2025-08-29T15:33:18.822Z" },
-    { url = "https://files.pythonhosted.org/packages/43/28/c43ba0ef19f446d6463c751315140d8f2a521e04c3e79e5c5fe211bfa430/coverage-7.10.6-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b37201ce4a458c7a758ecc4efa92fa8ed783c66e0fa3c42ae19fc454a0792153", size = 249659, upload-time = "2025-08-29T15:33:20.407Z" },
-    { url = "https://files.pythonhosted.org/packages/79/3e/53635bd0b72beaacf265784508a0b386defc9ab7fad99ff95f79ce9db555/coverage-7.10.6-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:2904271c80898663c810a6b067920a61dd8d38341244a3605bd31ab55250dad5", size = 247714, upload-time = "2025-08-29T15:33:21.751Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/55/0964aa87126624e8c159e32b0bc4e84edef78c89a1a4b924d28dd8265625/coverage-7.10.6-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:5aea98383463d6e1fa4e95416d8de66f2d0cb588774ee20ae1b28df826bcb619", size = 248351, upload-time = "2025-08-29T15:33:23.105Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/ab/6cfa9dc518c6c8e14a691c54e53a9433ba67336c760607e299bfcf520cb1/coverage-7.10.6-cp311-cp311-win32.whl", hash = "sha256:e3fb1fa01d3598002777dd259c0c2e6d9d5e10e7222976fc8e03992f972a2cba", size = 219562, upload-time = "2025-08-29T15:33:24.717Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/18/99b25346690cbc55922e7cfef06d755d4abee803ef335baff0014268eff4/coverage-7.10.6-cp311-cp311-win_amd64.whl", hash = "sha256:f35ed9d945bece26553d5b4c8630453169672bea0050a564456eb88bdffd927e", size = 220453, upload-time = "2025-08-29T15:33:26.482Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/ed/81d86648a07ccb124a5cf1f1a7788712b8d7216b593562683cd5c9b0d2c1/coverage-7.10.6-cp311-cp311-win_arm64.whl", hash = "sha256:99e1a305c7765631d74b98bf7dbf54eeea931f975e80f115437d23848ee8c27c", size = 219127, upload-time = "2025-08-29T15:33:27.777Z" },
-    { url = "https://files.pythonhosted.org/packages/26/06/263f3305c97ad78aab066d116b52250dd316e74fcc20c197b61e07eb391a/coverage-7.10.6-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:5b2dd6059938063a2c9fee1af729d4f2af28fd1a545e9b7652861f0d752ebcea", size = 217324, upload-time = "2025-08-29T15:33:29.06Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/60/1e1ded9a4fe80d843d7d53b3e395c1db3ff32d6c301e501f393b2e6c1c1f/coverage-7.10.6-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:388d80e56191bf846c485c14ae2bc8898aa3124d9d35903fef7d907780477634", size = 217560, upload-time = "2025-08-29T15:33:30.748Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/25/52136173c14e26dfed8b106ed725811bb53c30b896d04d28d74cb64318b3/coverage-7.10.6-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:90cb5b1a4670662719591aa92d0095bb41714970c0b065b02a2610172dbf0af6", size = 249053, upload-time = "2025-08-29T15:33:32.041Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/1d/ae25a7dc58fcce8b172d42ffe5313fc267afe61c97fa872b80ee72d9515a/coverage-7.10.6-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:961834e2f2b863a0e14260a9a273aff07ff7818ab6e66d2addf5628590c628f9", size = 251802, upload-time = "2025-08-29T15:33:33.625Z" },
-    { url = "https://files.pythonhosted.org/packages/f5/7a/1f561d47743710fe996957ed7c124b421320f150f1d38523d8d9102d3e2a/coverage-7.10.6-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bf9a19f5012dab774628491659646335b1928cfc931bf8d97b0d5918dd58033c", size = 252935, upload-time = "2025-08-29T15:33:34.909Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/ad/8b97cd5d28aecdfde792dcbf646bac141167a5cacae2cd775998b45fabb5/coverage-7.10.6-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:99c4283e2a0e147b9c9cc6bc9c96124de9419d6044837e9799763a0e29a7321a", size = 250855, upload-time = "2025-08-29T15:33:36.922Z" },
-    { url = "https://files.pythonhosted.org/packages/33/6a/95c32b558d9a61858ff9d79580d3877df3eb5bc9eed0941b1f187c89e143/coverage-7.10.6-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:282b1b20f45df57cc508c1e033403f02283adfb67d4c9c35a90281d81e5c52c5", size = 248974, upload-time = "2025-08-29T15:33:38.175Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/9c/8ce95dee640a38e760d5b747c10913e7a06554704d60b41e73fdea6a1ffd/coverage-7.10.6-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:8cdbe264f11afd69841bd8c0d83ca10b5b32853263ee62e6ac6a0ab63895f972", size = 250409, upload-time = "2025-08-29T15:33:39.447Z" },
-    { url = "https://files.pythonhosted.org/packages/04/12/7a55b0bdde78a98e2eb2356771fd2dcddb96579e8342bb52aa5bc52e96f0/coverage-7.10.6-cp312-cp312-win32.whl", hash = "sha256:a517feaf3a0a3eca1ee985d8373135cfdedfbba3882a5eab4362bda7c7cf518d", size = 219724, upload-time = "2025-08-29T15:33:41.172Z" },
-    { url = "https://files.pythonhosted.org/packages/36/4a/32b185b8b8e327802c9efce3d3108d2fe2d9d31f153a0f7ecfd59c773705/coverage-7.10.6-cp312-cp312-win_amd64.whl", hash = "sha256:856986eadf41f52b214176d894a7de05331117f6035a28ac0016c0f63d887629", size = 220536, upload-time = "2025-08-29T15:33:42.524Z" },
-    { url = "https://files.pythonhosted.org/packages/08/3a/d5d8dc703e4998038c3099eaf77adddb00536a3cec08c8dcd556a36a3eb4/coverage-7.10.6-cp312-cp312-win_arm64.whl", hash = "sha256:acf36b8268785aad739443fa2780c16260ee3fa09d12b3a70f772ef100939d80", size = 219171, upload-time = "2025-08-29T15:33:43.974Z" },
-    { url = "https://files.pythonhosted.org/packages/91/70/f73ad83b1d2fd2d5825ac58c8f551193433a7deaf9b0d00a8b69ef61cd9a/coverage-7.10.6-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:90558c35af64971d65fbd935c32010f9a2f52776103a259f1dee865fe8259352", size = 217009, upload-time = "2025-08-29T15:34:57.381Z" },
-    { url = "https://files.pythonhosted.org/packages/01/e8/099b55cd48922abbd4b01ddd9ffa352408614413ebfc965501e981aced6b/coverage-7.10.6-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:8953746d371e5695405806c46d705a3cd170b9cc2b9f93953ad838f6c1e58612", size = 217400, upload-time = "2025-08-29T15:34:58.985Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/d1/c6bac7c9e1003110a318636fef3b5c039df57ab44abcc41d43262a163c28/coverage-7.10.6-cp39-cp39-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:c83f6afb480eae0313114297d29d7c295670a41c11b274e6bca0c64540c1ce7b", size = 243835, upload-time = "2025-08-29T15:35:00.541Z" },
-    { url = "https://files.pythonhosted.org/packages/01/f9/82c6c061838afbd2172e773156c0aa84a901d59211b4975a4e93accf5c89/coverage-7.10.6-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7eb68d356ba0cc158ca535ce1381dbf2037fa8cb5b1ae5ddfc302e7317d04144", size = 245658, upload-time = "2025-08-29T15:35:02.135Z" },
-    { url = "https://files.pythonhosted.org/packages/81/6a/35674445b1d38161148558a3ff51b0aa7f0b54b1def3abe3fbd34efe05bc/coverage-7.10.6-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5b15a87265e96307482746d86995f4bff282f14b027db75469c446da6127433b", size = 247433, upload-time = "2025-08-29T15:35:03.777Z" },
-    { url = "https://files.pythonhosted.org/packages/18/27/98c99e7cafb288730a93535092eb433b5503d529869791681c4f2e2012a8/coverage-7.10.6-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:fc53ba868875bfbb66ee447d64d6413c2db91fddcfca57025a0e7ab5b07d5862", size = 245315, upload-time = "2025-08-29T15:35:05.629Z" },
-    { url = "https://files.pythonhosted.org/packages/09/05/123e0dba812408c719c319dea05782433246f7aa7b67e60402d90e847545/coverage-7.10.6-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:efeda443000aa23f276f4df973cb82beca682fd800bb119d19e80504ffe53ec2", size = 243385, upload-time = "2025-08-29T15:35:07.494Z" },
-    { url = "https://files.pythonhosted.org/packages/67/52/d57a42502aef05c6325f28e2e81216c2d9b489040132c18725b7a04d1448/coverage-7.10.6-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:9702b59d582ff1e184945d8b501ffdd08d2cee38d93a2206aa5f1365ce0b8d78", size = 244343, upload-time = "2025-08-29T15:35:09.55Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/22/7f6fad7dbb37cf99b542c5e157d463bd96b797078b1ec506691bc836f476/coverage-7.10.6-cp39-cp39-win32.whl", hash = "sha256:2195f8e16ba1a44651ca684db2ea2b2d4b5345da12f07d9c22a395202a05b23c", size = 219530, upload-time = "2025-08-29T15:35:11.167Z" },
-    { url = "https://files.pythonhosted.org/packages/62/30/e2fda29bfe335026027e11e6a5e57a764c9df13127b5cf42af4c3e99b937/coverage-7.10.6-cp39-cp39-win_amd64.whl", hash = "sha256:f32ff80e7ef6a5b5b606ea69a36e97b219cd9dc799bcf2963018a4d8f788cfbf", size = 220432, upload-time = "2025-08-29T15:35:12.902Z" },
-    { url = "https://files.pythonhosted.org/packages/44/0c/50db5379b615854b5cf89146f8f5bd1d5a9693d7f3a987e269693521c404/coverage-7.10.6-py3-none-any.whl", hash = "sha256:92c4ecf6bf11b2e85fd4d8204814dc26e6a19f0c9d938c207c5cb0eadfcabbe3", size = 208986, upload-time = "2025-08-29T15:35:14.506Z" },
+version = "7.10.7"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/51/26/d22c300112504f5f9a9fd2297ce33c35f3d353e4aeb987c8419453b2a7c2/coverage-7.10.7.tar.gz", hash = "sha256:f4ab143ab113be368a3e9b795f9cd7906c5ef407d6173fe9675a902e1fffc239", size = 827704, upload-time = "2025-09-21T20:03:56.815Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e5/6c/3a3f7a46888e69d18abe3ccc6fe4cb16cccb1e6a2f99698931dafca489e6/coverage-7.10.7-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:fc04cc7a3db33664e0c2d10eb8990ff6b3536f6842c9590ae8da4c614b9ed05a", size = 217987, upload-time = "2025-09-21T20:00:57.218Z" },
+    { url = "https://files.pythonhosted.org/packages/03/94/952d30f180b1a916c11a56f5c22d3535e943aa22430e9e3322447e520e1c/coverage-7.10.7-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e201e015644e207139f7e2351980feb7040e6f4b2c2978892f3e3789d1c125e5", size = 218388, upload-time = "2025-09-21T20:01:00.081Z" },
+    { url = "https://files.pythonhosted.org/packages/50/2b/9e0cf8ded1e114bcd8b2fd42792b57f1c4e9e4ea1824cde2af93a67305be/coverage-7.10.7-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:240af60539987ced2c399809bd34f7c78e8abe0736af91c3d7d0e795df633d17", size = 245148, upload-time = "2025-09-21T20:01:01.768Z" },
+    { url = "https://files.pythonhosted.org/packages/19/20/d0384ac06a6f908783d9b6aa6135e41b093971499ec488e47279f5b846e6/coverage-7.10.7-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:8421e088bc051361b01c4b3a50fd39a4b9133079a2229978d9d30511fd05231b", size = 246958, upload-time = "2025-09-21T20:01:03.355Z" },
+    { url = "https://files.pythonhosted.org/packages/60/83/5c283cff3d41285f8eab897651585db908a909c572bdc014bcfaf8a8b6ae/coverage-7.10.7-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6be8ed3039ae7f7ac5ce058c308484787c86e8437e72b30bf5e88b8ea10f3c87", size = 248819, upload-time = "2025-09-21T20:01:04.968Z" },
+    { url = "https://files.pythonhosted.org/packages/60/22/02eb98fdc5ff79f423e990d877693e5310ae1eab6cb20ae0b0b9ac45b23b/coverage-7.10.7-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:e28299d9f2e889e6d51b1f043f58d5f997c373cc12e6403b90df95b8b047c13e", size = 245754, upload-time = "2025-09-21T20:01:06.321Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/bc/25c83bcf3ad141b32cd7dc45485ef3c01a776ca3aa8ef0a93e77e8b5bc43/coverage-7.10.7-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:c4e16bd7761c5e454f4efd36f345286d6f7c5fa111623c355691e2755cae3b9e", size = 246860, upload-time = "2025-09-21T20:01:07.605Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/b7/95574702888b58c0928a6e982038c596f9c34d52c5e5107f1eef729399b5/coverage-7.10.7-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:b1c81d0e5e160651879755c9c675b974276f135558cf4ba79fee7b8413a515df", size = 244877, upload-time = "2025-09-21T20:01:08.829Z" },
+    { url = "https://files.pythonhosted.org/packages/47/b6/40095c185f235e085df0e0b158f6bd68cc6e1d80ba6c7721dc81d97ec318/coverage-7.10.7-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:606cc265adc9aaedcc84f1f064f0e8736bc45814f15a357e30fca7ecc01504e0", size = 245108, upload-time = "2025-09-21T20:01:10.527Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/50/4aea0556da7a4b93ec9168420d170b55e2eb50ae21b25062513d020c6861/coverage-7.10.7-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:10b24412692df990dbc34f8fb1b6b13d236ace9dfdd68df5b28c2e39cafbba13", size = 245752, upload-time = "2025-09-21T20:01:11.857Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/28/ea1a84a60828177ae3b100cb6723838523369a44ec5742313ed7db3da160/coverage-7.10.7-cp310-cp310-win32.whl", hash = "sha256:b51dcd060f18c19290d9b8a9dd1e0181538df2ce0717f562fff6cf74d9fc0b5b", size = 220497, upload-time = "2025-09-21T20:01:13.459Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/1a/a81d46bbeb3c3fd97b9602ebaa411e076219a150489bcc2c025f151bd52d/coverage-7.10.7-cp310-cp310-win_amd64.whl", hash = "sha256:3a622ac801b17198020f09af3eaf45666b344a0d69fc2a6ffe2ea83aeef1d807", size = 221392, upload-time = "2025-09-21T20:01:14.722Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/5d/c1a17867b0456f2e9ce2d8d4708a4c3a089947d0bec9c66cdf60c9e7739f/coverage-7.10.7-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:a609f9c93113be646f44c2a0256d6ea375ad047005d7f57a5c15f614dc1b2f59", size = 218102, upload-time = "2025-09-21T20:01:16.089Z" },
+    { url = "https://files.pythonhosted.org/packages/54/f0/514dcf4b4e3698b9a9077f084429681bf3aad2b4a72578f89d7f643eb506/coverage-7.10.7-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:65646bb0359386e07639c367a22cf9b5bf6304e8630b565d0626e2bdf329227a", size = 218505, upload-time = "2025-09-21T20:01:17.788Z" },
+    { url = "https://files.pythonhosted.org/packages/20/f6/9626b81d17e2a4b25c63ac1b425ff307ecdeef03d67c9a147673ae40dc36/coverage-7.10.7-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:5f33166f0dfcce728191f520bd2692914ec70fac2713f6bf3ce59c3deacb4699", size = 248898, upload-time = "2025-09-21T20:01:19.488Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/ef/bd8e719c2f7417ba03239052e099b76ea1130ac0cbb183ee1fcaa58aaff3/coverage-7.10.7-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:35f5e3f9e455bb17831876048355dca0f758b6df22f49258cb5a91da23ef437d", size = 250831, upload-time = "2025-09-21T20:01:20.817Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/b6/bf054de41ec948b151ae2b79a55c107f5760979538f5fb80c195f2517718/coverage-7.10.7-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4da86b6d62a496e908ac2898243920c7992499c1712ff7c2b6d837cc69d9467e", size = 252937, upload-time = "2025-09-21T20:01:22.171Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/e5/3860756aa6f9318227443c6ce4ed7bf9e70bb7f1447a0353f45ac5c7974b/coverage-7.10.7-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:6b8b09c1fad947c84bbbc95eca841350fad9cbfa5a2d7ca88ac9f8d836c92e23", size = 249021, upload-time = "2025-09-21T20:01:23.907Z" },
+    { url = "https://files.pythonhosted.org/packages/26/0f/bd08bd042854f7fd07b45808927ebcce99a7ed0f2f412d11629883517ac2/coverage-7.10.7-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:4376538f36b533b46f8971d3a3e63464f2c7905c9800db97361c43a2b14792ab", size = 250626, upload-time = "2025-09-21T20:01:25.721Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/a7/4777b14de4abcc2e80c6b1d430f5d51eb18ed1d75fca56cbce5f2db9b36e/coverage-7.10.7-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:121da30abb574f6ce6ae09840dae322bef734480ceafe410117627aa54f76d82", size = 248682, upload-time = "2025-09-21T20:01:27.105Z" },
+    { url = "https://files.pythonhosted.org/packages/34/72/17d082b00b53cd45679bad682fac058b87f011fd8b9fe31d77f5f8d3a4e4/coverage-7.10.7-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:88127d40df529336a9836870436fc2751c339fbaed3a836d42c93f3e4bd1d0a2", size = 248402, upload-time = "2025-09-21T20:01:28.629Z" },
+    { url = "https://files.pythonhosted.org/packages/81/7a/92367572eb5bdd6a84bfa278cc7e97db192f9f45b28c94a9ca1a921c3577/coverage-7.10.7-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ba58bbcd1b72f136080c0bccc2400d66cc6115f3f906c499013d065ac33a4b61", size = 249320, upload-time = "2025-09-21T20:01:30.004Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/88/a23cc185f6a805dfc4fdf14a94016835eeb85e22ac3a0e66d5e89acd6462/coverage-7.10.7-cp311-cp311-win32.whl", hash = "sha256:972b9e3a4094b053a4e46832b4bc829fc8a8d347160eb39d03f1690316a99c14", size = 220536, upload-time = "2025-09-21T20:01:32.184Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/ef/0b510a399dfca17cec7bc2f05ad8bd78cf55f15c8bc9a73ab20c5c913c2e/coverage-7.10.7-cp311-cp311-win_amd64.whl", hash = "sha256:a7b55a944a7f43892e28ad4bc0561dfd5f0d73e605d1aa5c3c976b52aea121d2", size = 221425, upload-time = "2025-09-21T20:01:33.557Z" },
+    { url = "https://files.pythonhosted.org/packages/51/7f/023657f301a276e4ba1850f82749bc136f5a7e8768060c2e5d9744a22951/coverage-7.10.7-cp311-cp311-win_arm64.whl", hash = "sha256:736f227fb490f03c6488f9b6d45855f8e0fd749c007f9303ad30efab0e73c05a", size = 220103, upload-time = "2025-09-21T20:01:34.929Z" },
+    { url = "https://files.pythonhosted.org/packages/13/e4/eb12450f71b542a53972d19117ea5a5cea1cab3ac9e31b0b5d498df1bd5a/coverage-7.10.7-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:7bb3b9ddb87ef7725056572368040c32775036472d5a033679d1fa6c8dc08417", size = 218290, upload-time = "2025-09-21T20:01:36.455Z" },
+    { url = "https://files.pythonhosted.org/packages/37/66/593f9be12fc19fb36711f19a5371af79a718537204d16ea1d36f16bd78d2/coverage-7.10.7-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:18afb24843cbc175687225cab1138c95d262337f5473512010e46831aa0c2973", size = 218515, upload-time = "2025-09-21T20:01:37.982Z" },
+    { url = "https://files.pythonhosted.org/packages/66/80/4c49f7ae09cafdacc73fbc30949ffe77359635c168f4e9ff33c9ebb07838/coverage-7.10.7-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:399a0b6347bcd3822be369392932884b8216d0944049ae22925631a9b3d4ba4c", size = 250020, upload-time = "2025-09-21T20:01:39.617Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/90/a64aaacab3b37a17aaedd83e8000142561a29eb262cede42d94a67f7556b/coverage-7.10.7-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:314f2c326ded3f4b09be11bc282eb2fc861184bc95748ae67b360ac962770be7", size = 252769, upload-time = "2025-09-21T20:01:41.341Z" },
+    { url = "https://files.pythonhosted.org/packages/98/2e/2dda59afd6103b342e096f246ebc5f87a3363b5412609946c120f4e7750d/coverage-7.10.7-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c41e71c9cfb854789dee6fc51e46743a6d138b1803fab6cb860af43265b42ea6", size = 253901, upload-time = "2025-09-21T20:01:43.042Z" },
+    { url = "https://files.pythonhosted.org/packages/53/dc/8d8119c9051d50f3119bb4a75f29f1e4a6ab9415cd1fa8bf22fcc3fb3b5f/coverage-7.10.7-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:bc01f57ca26269c2c706e838f6422e2a8788e41b3e3c65e2f41148212e57cd59", size = 250413, upload-time = "2025-09-21T20:01:44.469Z" },
+    { url = "https://files.pythonhosted.org/packages/98/b3/edaff9c5d79ee4d4b6d3fe046f2b1d799850425695b789d491a64225d493/coverage-7.10.7-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a6442c59a8ac8b85812ce33bc4d05bde3fb22321fa8294e2a5b487c3505f611b", size = 251820, upload-time = "2025-09-21T20:01:45.915Z" },
+    { url = "https://files.pythonhosted.org/packages/11/25/9a0728564bb05863f7e513e5a594fe5ffef091b325437f5430e8cfb0d530/coverage-7.10.7-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:78a384e49f46b80fb4c901d52d92abe098e78768ed829c673fbb53c498bef73a", size = 249941, upload-time = "2025-09-21T20:01:47.296Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/fd/ca2650443bfbef5b0e74373aac4df67b08180d2f184b482c41499668e258/coverage-7.10.7-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:5e1e9802121405ede4b0133aa4340ad8186a1d2526de5b7c3eca519db7bb89fb", size = 249519, upload-time = "2025-09-21T20:01:48.73Z" },
+    { url = "https://files.pythonhosted.org/packages/24/79/f692f125fb4299b6f963b0745124998ebb8e73ecdfce4ceceb06a8c6bec5/coverage-7.10.7-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:d41213ea25a86f69efd1575073d34ea11aabe075604ddf3d148ecfec9e1e96a1", size = 251375, upload-time = "2025-09-21T20:01:50.529Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/75/61b9bbd6c7d24d896bfeec57acba78e0f8deac68e6baf2d4804f7aae1f88/coverage-7.10.7-cp312-cp312-win32.whl", hash = "sha256:77eb4c747061a6af8d0f7bdb31f1e108d172762ef579166ec84542f711d90256", size = 220699, upload-time = "2025-09-21T20:01:51.941Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/f3/3bf7905288b45b075918d372498f1cf845b5b579b723c8fd17168018d5f5/coverage-7.10.7-cp312-cp312-win_amd64.whl", hash = "sha256:f51328ffe987aecf6d09f3cd9d979face89a617eacdaea43e7b3080777f647ba", size = 221512, upload-time = "2025-09-21T20:01:53.481Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/44/3e32dbe933979d05cf2dac5e697c8599cfe038aaf51223ab901e208d5a62/coverage-7.10.7-cp312-cp312-win_arm64.whl", hash = "sha256:bda5e34f8a75721c96085903c6f2197dc398c20ffd98df33f866a9c8fd95f4bf", size = 220147, upload-time = "2025-09-21T20:01:55.2Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/ad/d1c25053764b4c42eb294aae92ab617d2e4f803397f9c7c8295caa77a260/coverage-7.10.7-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:fff7b9c3f19957020cac546c70025331113d2e61537f6e2441bc7657913de7d3", size = 217978, upload-time = "2025-09-21T20:03:30.362Z" },
+    { url = "https://files.pythonhosted.org/packages/52/2f/b9f9daa39b80ece0b9548bbb723381e29bc664822d9a12c2135f8922c22b/coverage-7.10.7-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:bc91b314cef27742da486d6839b677b3f2793dfe52b51bbbb7cf736d5c29281c", size = 218370, upload-time = "2025-09-21T20:03:32.147Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/6e/30d006c3b469e58449650642383dddf1c8fb63d44fdf92994bfd46570695/coverage-7.10.7-cp39-cp39-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:567f5c155eda8df1d3d439d40a45a6a5f029b429b06648235f1e7e51b522b396", size = 244802, upload-time = "2025-09-21T20:03:33.919Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/49/8a070782ce7e6b94ff6a0b6d7c65ba6bc3091d92a92cef4cd4eb0767965c/coverage-7.10.7-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:2af88deffcc8a4d5974cf2d502251bc3b2db8461f0b66d80a449c33757aa9f40", size = 246625, upload-time = "2025-09-21T20:03:36.09Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/92/1c1c5a9e8677ce56d42b97bdaca337b2d4d9ebe703d8c174ede52dbabd5f/coverage-7.10.7-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c7315339eae3b24c2d2fa1ed7d7a38654cba34a13ef19fbcb9425da46d3dc594", size = 248399, upload-time = "2025-09-21T20:03:38.342Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/54/b140edee7257e815de7426d5d9846b58505dffc29795fff2dfb7f8a1c5a0/coverage-7.10.7-cp39-cp39-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:912e6ebc7a6e4adfdbb1aec371ad04c68854cd3bf3608b3514e7ff9062931d8a", size = 245142, upload-time = "2025-09-21T20:03:40.591Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/9e/6d6b8295940b118e8b7083b29226c71f6154f7ff41e9ca431f03de2eac0d/coverage-7.10.7-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:f49a05acd3dfe1ce9715b657e28d138578bc40126760efb962322c56e9ca344b", size = 246284, upload-time = "2025-09-21T20:03:42.355Z" },
+    { url = "https://files.pythonhosted.org/packages/db/e5/5e957ca747d43dbe4d9714358375c7546cb3cb533007b6813fc20fce37ad/coverage-7.10.7-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:cce2109b6219f22ece99db7644b9622f54a4e915dad65660ec435e89a3ea7cc3", size = 244353, upload-time = "2025-09-21T20:03:44.218Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/45/540fc5cc92536a1b783b7ef99450bd55a4b3af234aae35a18a339973ce30/coverage-7.10.7-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:f3c887f96407cea3916294046fc7dab611c2552beadbed4ea901cbc6a40cc7a0", size = 244430, upload-time = "2025-09-21T20:03:46.065Z" },
+    { url = "https://files.pythonhosted.org/packages/75/0b/8287b2e5b38c8fe15d7e3398849bb58d382aedc0864ea0fa1820e8630491/coverage-7.10.7-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:635adb9a4507c9fd2ed65f39693fa31c9a3ee3a8e6dc64df033e8fdf52a7003f", size = 245311, upload-time = "2025-09-21T20:03:48.19Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/1d/29724999984740f0c86d03e6420b942439bf5bd7f54d4382cae386a9d1e9/coverage-7.10.7-cp39-cp39-win32.whl", hash = "sha256:5a02d5a850e2979b0a014c412573953995174743a3f7fa4ea5a6e9a3c5617431", size = 220500, upload-time = "2025-09-21T20:03:50.024Z" },
+    { url = "https://files.pythonhosted.org/packages/43/11/4b1e6b129943f905ca54c339f343877b55b365ae2558806c1be4f7476ed5/coverage-7.10.7-cp39-cp39-win_amd64.whl", hash = "sha256:c134869d5ffe34547d14e174c866fd8fe2254918cc0a95e99052903bc1543e07", size = 221408, upload-time = "2025-09-21T20:03:51.803Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/16/114df1c291c22cac3b0c127a73e0af5c12ed7bbb6558d310429a0ae24023/coverage-7.10.7-py3-none-any.whl", hash = "sha256:f7941f6f2fe6dd6807a1208737b8a0cbcf1cc6d7b07d24998ad2d63590868260", size = 209952, upload-time = "2025-09-21T20:03:53.918Z" },
 ]
 
 [package.optional-dependencies]
@@ -502,9 +645,29 @@ toml = [
     { name = "tomli", marker = "python_full_version <= '3.11'" },
 ]
 
+[[package]]
+name = "cupy-cuda12x"
+version = "13.6.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "fastrlock" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f7/2e/db22c5148884e4e384f6ebbc7971fa3710f3ba67ca492798890a0fdebc45/cupy_cuda12x-13.6.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:9e37f60f27ff9625dfdccc4688a09852707ec613e32ea9404f425dd22a386d14", size = 126341714, upload-time = "2025-08-18T08:24:08.335Z" },
+    { url = "https://files.pythonhosted.org/packages/53/2b/8064d94a6ab6b5c4e643d8535ab6af6cabe5455765540931f0ef60a0bc3b/cupy_cuda12x-13.6.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:e78409ea72f5ac7d6b6f3d33d99426a94005254fa57e10617f430f9fd7c3a0a1", size = 112238589, upload-time = "2025-08-18T08:24:15.541Z" },
+    { url = "https://files.pythonhosted.org/packages/54/64/71c6e08f76c06639e5112f69ee3bc1129be00054ad5f906d7fd3138af579/cupy_cuda12x-13.6.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:c790d012fd4d86872b9c89af9f5f15d91c30b8e3a4aa4dd04c2610f45f06ac44", size = 128016458, upload-time = "2025-08-18T08:24:26.394Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/d9/5c5077243cd92368c3eccecdbf91d76db15db338169042ffd1647533c6b1/cupy_cuda12x-13.6.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:77ba6745a130d880c962e687e4e146ebbb9014f290b0a80dbc4e4634eb5c3b48", size = 113039337, upload-time = "2025-08-18T08:24:31.814Z" },
+    { url = "https://files.pythonhosted.org/packages/12/c5/7e7fc4816d0de0154e5d9053242c3a08a0ca8b43ee656a6f7b3b95055a7b/cupy_cuda12x-13.6.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:a6970ceefe40f9acbede41d7fe17416bd277b1bd2093adcde457b23b578c5a59", size = 127334633, upload-time = "2025-08-18T08:24:43.065Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/95/d7e1295141e7d530674a3cc567e13ed0eb6b81524cb122d797ed996b5bea/cupy_cuda12x-13.6.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:79b0cacb5e8b190ef409f9e03f06ac8de1b021b0c0dda47674d446f5557e0eb1", size = 112886268, upload-time = "2025-08-18T08:24:49.294Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/23/aec6590ce32f0b0d284081192e9d25a07a410e57aaa28e2c3f2911881d2d/cupy_cuda12x-13.6.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:6ccd2fc75b0e0e24493531b8f8d8f978efecddb45f8479a48890c40d3805eb87", size = 126427104, upload-time = "2025-08-18T08:25:16.124Z" },
+    { url = "https://files.pythonhosted.org/packages/17/56/683967510b4392b518ba32a8560ef587d841f91b20afea76c6c642eb8b68/cupy_cuda12x-13.6.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:771f3135861b68199c18b49345210180d4fcdce4681b51c28224db389c4aac5d", size = 112305805, upload-time = "2025-08-18T08:25:24.097Z" },
+]
+
 [[package]]
 name = "datasets"
-version = "4.0.0"
+version = "4.1.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "dill" },
@@ -522,14 +685,27 @@ dependencies = [
     { name = "tqdm" },
     { name = "xxhash" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/e3/9d/348ed92110ba5f9b70b51ca1078d4809767a835aa2b7ce7e74ad2b98323d/datasets-4.0.0.tar.gz", hash = "sha256:9657e7140a9050db13443ba21cb5de185af8af944479b00e7ff1e00a61c8dbf1", size = 569566, upload-time = "2025-07-09T14:35:52.431Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/91/a4/73f8e6ef52c535e1d20d5b2ca83bfe6de399d8b8b8a61ccc8d63d60735aa/datasets-4.1.1.tar.gz", hash = "sha256:7d8d5ba8b12861d2c44bfff9c83484ebfafff1ff553371e5901a8d3aab5450e2", size = 579324, upload-time = "2025-09-18T13:14:27.108Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/eb/62/eb8157afb21bd229c864521c1ab4fa8e9b4f1b06bafdd8c4668a7a31b5dd/datasets-4.0.0-py3-none-any.whl", hash = "sha256:7ef95e62025fd122882dbce6cb904c8cd3fbc829de6669a5eb939c77d50e203d", size = 494825, upload-time = "2025-07-09T14:35:50.658Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/c8/09012ac195a0aab58755800d2efdc0e7d5905053509f12cb5d136c911cda/datasets-4.1.1-py3-none-any.whl", hash = "sha256:62e4f6899a36be9ec74a7e759a6951253cc85b3fcfa0a759b0efa8353b149dac", size = 503623, upload-time = "2025-09-18T13:14:25.111Z" },
+]
+
+[[package]]
+name = "depyf"
+version = "0.19.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "astor" },
+    { name = "dill" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/19/38/69157d711be575f1b9cf3177b64ef4ade44373fc02839f183fdd98ec2dd6/depyf-0.19.0.tar.gz", hash = "sha256:afed0916b32d141cc90fa6220df01885eda442ca43b297d5050eeb90b4a5cb44", size = 6171405, upload-time = "2025-04-20T08:07:41.224Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/28/4d/1192acbcdc5e843f5e5d51f6e8788f2b60a9fe0b578ac385ded67a0b0b26/depyf-0.19.0-py3-none-any.whl", hash = "sha256:040b35fc0997d49df024b7d094f2a7836f91e9ed02f49982dd37e70aa3285ad5", size = 39034, upload-time = "2025-04-20T08:07:37.036Z" },
 ]
 
 [[package]]
 name = "diff-cover"
-version = "9.6.0"
+version = "9.7.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "chardet" },
@@ -537,18 +713,18 @@ dependencies = [
     { name = "pluggy" },
     { name = "pygments" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/4c/81/0496938bb72606231b2518b6fd404a5adfde3c1b60a97d0dc39a92b25570/diff_cover-9.6.0.tar.gz", hash = "sha256:75e5bc056dcaa68c6c87c9fb4e07c9e60daef15b6e8d034d56d2da9e2c84a872", size = 99756, upload-time = "2025-07-17T02:18:44.286Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/eb/b7/3124f2038eb9d2842665c109d4516110e3e5570615243ee1065fdc65ac7d/diff_cover-9.7.1.tar.gz", hash = "sha256:a03b78bc64b7dc27dda7f11c46bc6d44f4e121c8fb9a5068d41f20a05d463548", size = 100366, upload-time = "2025-09-26T01:02:43.81Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/30/c5/5f3b69c74328c6a3a3a52a084afbd0c2d15c4b95e7784b752072c133f01d/diff_cover-9.6.0-py3-none-any.whl", hash = "sha256:29fbeb52d77a0b8c811e5580d5dbf41801a838da2ed54319a599da8f7233c547", size = 55671, upload-time = "2025-07-17T02:18:42.572Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/c2/36d747a14dd3834af7e2f4f86daf8d6e21762ffbdb2b177a5bb3be8f0e8f/diff_cover-9.7.1-py3-none-any.whl", hash = "sha256:a662f3cacf57ba494dcb0235349c4912ae7f44345b162894ce5949745501c9c7", size = 56008, upload-time = "2025-09-26T01:02:42.399Z" },
 ]
 
 [[package]]
 name = "dill"
-version = "0.3.8"
+version = "0.4.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/17/4d/ac7ffa80c69ea1df30a8aa11b3578692a5118e7cd1aa157e3ef73b092d15/dill-0.3.8.tar.gz", hash = "sha256:3ebe3c479ad625c4553aca177444d89b486b1d84982eeacded644afc0cf797ca", size = 184847, upload-time = "2024-01-27T23:42:16.145Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/12/80/630b4b88364e9a8c8c5797f4602d0f76ef820909ee32f0bacb9f90654042/dill-0.4.0.tar.gz", hash = "sha256:0633f1d2df477324f53a895b02c901fb961bdbf65a17122586ea7019292cbcf0", size = 186976, upload-time = "2025-04-16T00:41:48.867Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c9/7a/cef76fd8438a42f96db64ddaa85280485a9c395e7df3db8158cfec1eee34/dill-0.3.8-py3-none-any.whl", hash = "sha256:c36ca9ffb54365bdd2f8eb3eff7d2a21237f8452b57ace88b1ac615b7e815bd7", size = 116252, upload-time = "2024-01-27T23:42:14.239Z" },
+    { url = "https://files.pythonhosted.org/packages/50/3d/9373ad9c56321fdab5b41197068e1d8c25883b3fea29dd361f9b55116869/dill-0.4.0-py3-none-any.whl", hash = "sha256:44f54bf6412c2c8464c14e8243eb163690a9800dbe2c367330883b19c7561049", size = 119668, upload-time = "2025-04-16T00:41:47.671Z" },
 ]
 
 [[package]]
@@ -578,6 +754,41 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/12/b3/231ffd4ab1fc9d679809f356cebee130ac7daa00d6d6f3206dd4fd137e9e/distro-1.9.0-py3-none-any.whl", hash = "sha256:7bffd925d65168f85027d8da9af6bddab658135b840670a223589bc0c8ef02b2", size = 20277, upload-time = "2023-12-24T09:54:30.421Z" },
 ]
 
+[[package]]
+name = "dnspython"
+version = "2.7.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b5/4a/263763cb2ba3816dd94b08ad3a33d5fdae34ecb856678773cc40a3605829/dnspython-2.7.0.tar.gz", hash = "sha256:ce9c432eda0dc91cf618a5cedf1a4e142651196bbcd2c80e89ed5a907e5cfaf1", size = 345197, upload-time = "2024-10-05T20:14:59.362Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/68/1b/e0a87d256e40e8c888847551b20a017a6b98139178505dc7ffb96f04e954/dnspython-2.7.0-py3-none-any.whl", hash = "sha256:b4c34b7d10b51bcc3a5071e7b8dee77939f1e878477eeecc965e9835f63c6c86", size = 313632, upload-time = "2024-10-05T20:14:57.687Z" },
+]
+
+[[package]]
+name = "dnspython"
+version = "2.8.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/8c/8b/57666417c0f90f08bcafa776861060426765fdb422eb10212086fb811d26/dnspython-2.8.0.tar.gz", hash = "sha256:181d3c6996452cb1189c4046c61599b84a5a86e099562ffde77d26984ff26d0f", size = 368251, upload-time = "2025-09-07T18:58:00.022Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ba/5a/18ad964b0086c6e62e2e7500f7edc89e3faa45033c71c1893d34eed2b2de/dnspython-2.8.0-py3-none-any.whl", hash = "sha256:01d9bbc4a2d76bf0db7c1f729812ded6d912bd318d3b1cf81d30c0f845dbf3af", size = 331094, upload-time = "2025-09-07T18:57:58.071Z" },
+]
+
+[[package]]
+name = "docstring-parser"
+version = "0.17.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/b2/9d/c3b43da9515bd270df0f80548d9944e389870713cc1fe2b8fb35fe2bcefd/docstring_parser-0.17.0.tar.gz", hash = "sha256:583de4a309722b3315439bb31d64ba3eebada841f2e2cee23b99df001434c912", size = 27442, upload-time = "2025-07-21T07:35:01.868Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/55/e2/2537ebcff11c1ee1ff17d8d0b6f4db75873e3b0fb32c2d4a2ee31ecb310a/docstring_parser-0.17.0-py3-none-any.whl", hash = "sha256:cf2569abd23dce8099b300f9b4fa8191e9582dda731fd533daf54c4551658708", size = 36896, upload-time = "2025-07-21T07:35:00.684Z" },
+]
+
 [[package]]
 name = "dottxt"
 version = "0.1.5"
@@ -585,7 +796,7 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "backoff" },
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pydantic" },
     { name = "pydantic-settings" },
     { name = "python-dateutil" },
@@ -596,6 +807,29 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/46/0c/c0507e1641185890ef173bce883d9a02d60b446441fd674db352d9cf52e5/dottxt-0.1.5-py3-none-any.whl", hash = "sha256:645af5f9c43721893b7be91c3a18d1a45bee0f7544caf370f54587df0e127dd2", size = 65293, upload-time = "2025-02-14T17:10:12.353Z" },
 ]
 
+[[package]]
+name = "einops"
+version = "0.8.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/e5/81/df4fbe24dff8ba3934af99044188e20a98ed441ad17a274539b74e82e126/einops-0.8.1.tar.gz", hash = "sha256:de5d960a7a761225532e0f1959e5315ebeafc0cd43394732f103ca44b9837e84", size = 54805, upload-time = "2025-02-09T03:17:00.434Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/87/62/9773de14fe6c45c23649e98b83231fffd7b9892b6cf863251dc2afa73643/einops-0.8.1-py3-none-any.whl", hash = "sha256:919387eb55330f5757c6bea9165c5ff5cfe63a642682ea788a6d472576d81737", size = 64359, upload-time = "2025-02-09T03:17:01.998Z" },
+]
+
+[[package]]
+name = "email-validator"
+version = "2.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "dnspython", version = "2.7.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "dnspython", version = "2.8.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "idna" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f5/22/900cb125c76b7aaa450ce02fd727f452243f2e91a61af068b40adba60ea9/email_validator-2.3.0.tar.gz", hash = "sha256:9fc05c37f2f6cf439ff414f8fc46d917929974a82244c20eb10231ba60c54426", size = 51238, upload-time = "2025-08-26T13:09:06.831Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/de/15/545e2b6cf2e3be84bc1ed85613edd75b8aea69807a71c26f4ca6a9258e82/email_validator-2.3.0-py3-none-any.whl", hash = "sha256:80f13f623413e6b197ae73bb10bf4eb0908faf509ad8362c5edeb0be7fd450b4", size = 35604, upload-time = "2025-08-26T13:09:05.858Z" },
+]
+
 [[package]]
 name = "etils"
 version = "1.5.2"
@@ -644,6 +878,15 @@ epy = [
     { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
 
+[[package]]
+name = "eval-type-backport"
+version = "0.2.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/30/ea/8b0ac4469d4c347c6a385ff09dc3c048c2d021696664e26c7ee6791631b5/eval_type_backport-0.2.2.tar.gz", hash = "sha256:f0576b4cf01ebb5bd358d02314d31846af5e07678387486e2c798af0e7d849c1", size = 9079, upload-time = "2024-12-21T20:09:46.005Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ce/31/55cd413eaccd39125368be33c46de24a1f639f2e12349b0361b4678f3915/eval_type_backport-0.2.2-py3-none-any.whl", hash = "sha256:cb6ad7c393517f476f96d456d0412ea80f0a8cf96f6892834cd9340149111b0a", size = 5830, upload-time = "2024-12-21T20:09:44.175Z" },
+]
+
 [[package]]
 name = "exceptiongroup"
 version = "1.3.0"
@@ -658,16 +901,96 @@ wheels = [
 
 [[package]]
 name = "fastapi"
-version = "0.116.1"
+version = "0.118.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "pydantic" },
     { name = "starlette" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/78/d7/6c8b3bfe33eeffa208183ec037fee0cce9f7f024089ab1c5d12ef04bd27c/fastapi-0.116.1.tar.gz", hash = "sha256:ed52cbf946abfd70c5a0dccb24673f0670deeb517a88b3544d03c2a6bf283143", size = 296485, upload-time = "2025-07-11T16:22:32.057Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/28/3c/2b9345a6504e4055eaa490e0b41c10e338ad61d9aeaae41d97807873cdf2/fastapi-0.118.0.tar.gz", hash = "sha256:5e81654d98c4d2f53790a7d32d25a7353b30c81441be7d0958a26b5d761fa1c8", size = 310536, upload-time = "2025-09-29T03:37:23.126Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e5/47/d63c60f59a59467fda0f93f46335c9d18526d7071f025cb5b89d5353ea42/fastapi-0.116.1-py3-none-any.whl", hash = "sha256:c46ac7c312df840f0c9e220f7964bada936781bc4e2e6eb71f1c4d7553786565", size = 95631, upload-time = "2025-07-11T16:22:30.485Z" },
+    { url = "https://files.pythonhosted.org/packages/54/20/54e2bdaad22ca91a59455251998d43094d5c3d3567c52c7c04774b3f43f2/fastapi-0.118.0-py3-none-any.whl", hash = "sha256:705137a61e2ef71019d2445b123aa8845bd97273c395b744d5a7dfe559056855", size = 97694, upload-time = "2025-09-29T03:37:21.338Z" },
+]
+
+[package.optional-dependencies]
+standard = [
+    { name = "email-validator" },
+    { name = "fastapi-cli", extra = ["standard"] },
+    { name = "httpx" },
+    { name = "jinja2" },
+    { name = "python-multipart" },
+    { name = "uvicorn", extra = ["standard"] },
+]
+
+[[package]]
+name = "fastapi-cli"
+version = "0.0.13"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "rich-toolkit" },
+    { name = "typer" },
+    { name = "uvicorn", extra = ["standard"] },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/32/4e/3f61850012473b097fc5297d681bd85788e186fadb8555b67baf4c7707f4/fastapi_cli-0.0.13.tar.gz", hash = "sha256:312addf3f57ba7139457cf0d345c03e2170cc5a034057488259c33cd7e494529", size = 17780, upload-time = "2025-09-20T16:37:31.089Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/08/36/7432750f3638324b055496d2c952000bea824259fca70df5577a6a3c172f/fastapi_cli-0.0.13-py3-none-any.whl", hash = "sha256:219b73ccfde7622559cef1d43197da928516acb4f21f2ec69128c4b90057baba", size = 11142, upload-time = "2025-09-20T16:37:29.695Z" },
+]
+
+[package.optional-dependencies]
+standard = [
+    { name = "fastapi-cloud-cli" },
+    { name = "uvicorn", extra = ["standard"] },
+]
+
+[[package]]
+name = "fastapi-cloud-cli"
+version = "0.3.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "httpx" },
+    { name = "pydantic", extra = ["email"] },
+    { name = "rich-toolkit" },
+    { name = "rignore" },
+    { name = "sentry-sdk" },
+    { name = "typer" },
+    { name = "uvicorn", extra = ["standard"] },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a6/5f/17b403148a23dd708e3166f534136f4d3918942e168aca66659311eb0678/fastapi_cloud_cli-0.3.0.tar.gz", hash = "sha256:17c7f8baa16b2f907696bf77d49df4a04e8715bbf5233024f273870f3ff1ca4d", size = 24388, upload-time = "2025-10-02T13:25:52.361Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/58/59/7d12c5173fe2eed21e99bb1a6eb7e4f301951db870a4d915d126e0b6062d/fastapi_cloud_cli-0.3.0-py3-none-any.whl", hash = "sha256:572677dbe38b6d4712d30097a8807b383d648ca09eb58e4a07cef4a517020832", size = 19921, upload-time = "2025-10-02T13:25:51.164Z" },
+]
+
+[[package]]
+name = "fastrlock"
+version = "0.8.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/73/b1/1c3d635d955f2b4bf34d45abf8f35492e04dbd7804e94ce65d9f928ef3ec/fastrlock-0.8.3.tar.gz", hash = "sha256:4af6734d92eaa3ab4373e6c9a1dd0d5ad1304e172b1521733c6c3b3d73c8fa5d", size = 79327, upload-time = "2024-12-17T11:03:39.638Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/be/b4/aae7ed94b8122c325d89eb91336084596cebc505dc629b795fcc9629606d/fastrlock-0.8.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:7a77ebb0a24535ef4f167da2c5ee35d9be1e96ae192137e9dc3ff75b8dfc08a5", size = 48220, upload-time = "2024-12-17T11:01:51.071Z" },
+    { url = "https://files.pythonhosted.org/packages/96/87/9807af47617fdd65c68b0fcd1e714542c1d4d3a1f1381f591f1aa7383a53/fastrlock-0.8.3-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_24_i686.whl", hash = "sha256:d51f7fb0db8dab341b7f03a39a3031678cf4a98b18533b176c533c122bfce47d", size = 49551, upload-time = "2024-12-17T11:01:52.316Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/12/e201634810ac9aee59f93e3953cb39f98157d17c3fc9d44900f1209054e9/fastrlock-0.8.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:767ec79b7f6ed9b9a00eb9ff62f2a51f56fdb221c5092ab2dadec34a9ccbfc6e", size = 49398, upload-time = "2024-12-17T11:01:53.514Z" },
+    { url = "https://files.pythonhosted.org/packages/15/a1/439962ed439ff6f00b7dce14927e7830e02618f26f4653424220a646cd1c/fastrlock-0.8.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0d6a77b3f396f7d41094ef09606f65ae57feeb713f4285e8e417f4021617ca62", size = 53334, upload-time = "2024-12-17T11:01:55.518Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/9e/1ae90829dd40559ab104e97ebe74217d9da794c4bb43016da8367ca7a596/fastrlock-0.8.3-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:92577ff82ef4a94c5667d6d2841f017820932bc59f31ffd83e4a2c56c1738f90", size = 52495, upload-time = "2024-12-17T11:01:57.76Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/8c/5e746ee6f3d7afbfbb0d794c16c71bfd5259a4e3fb1dda48baf31e46956c/fastrlock-0.8.3-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:3df8514086e16bb7c66169156a8066dc152f3be892c7817e85bf09a27fa2ada2", size = 51972, upload-time = "2024-12-17T11:02:01.384Z" },
+    { url = "https://files.pythonhosted.org/packages/be/91/5f3afba7d14b8b7d60ac651375f50fff9220d6ccc3bef233d2bd74b73ec7/fastrlock-0.8.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:55d42f6286b9d867370af4c27bc70d04ce2d342fe450c4a4fcce14440514e695", size = 48911, upload-time = "2024-12-17T11:02:06.173Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/7a/e37bd72d7d70a8a551b3b4610d028bd73ff5d6253201d5d3cf6296468bee/fastrlock-0.8.3-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_24_i686.whl", hash = "sha256:bbc3bf96dcbd68392366c477f78c9d5c47e5d9290cb115feea19f20a43ef6d05", size = 50357, upload-time = "2024-12-17T11:02:07.418Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/ef/a13b8bab8266840bf38831d7bf5970518c02603d00a548a678763322d5bf/fastrlock-0.8.3-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:77ab8a98417a1f467dafcd2226718f7ca0cf18d4b64732f838b8c2b3e4b55cb5", size = 50222, upload-time = "2024-12-17T11:02:08.745Z" },
+    { url = "https://files.pythonhosted.org/packages/01/e2/5e5515562b2e9a56d84659377176aef7345da2c3c22909a1897fe27e14dd/fastrlock-0.8.3-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:04bb5eef8f460d13b8c0084ea5a9d3aab2c0573991c880c0a34a56bb14951d30", size = 54553, upload-time = "2024-12-17T11:02:10.925Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/8f/65907405a8cdb2fc8beaf7d09a9a07bb58deff478ff391ca95be4f130b70/fastrlock-0.8.3-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:8c9d459ce344c21ff03268212a1845aa37feab634d242131bc16c2a2355d5f65", size = 53362, upload-time = "2024-12-17T11:02:12.476Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/b9/ae6511e52738ba4e3a6adb7c6a20158573fbc98aab448992ece25abb0b07/fastrlock-0.8.3-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:33e6fa4af4f3af3e9c747ec72d1eadc0b7ba2035456c2afb51c24d9e8a56f8fd", size = 52836, upload-time = "2024-12-17T11:02:13.74Z" },
+    { url = "https://files.pythonhosted.org/packages/57/21/ea1511b0ef0d5457efca3bf1823effb9c5cad4fc9dca86ce08e4d65330ce/fastrlock-0.8.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:85a49a1f1e020097d087e1963e42cea6f307897d5ebe2cb6daf4af47ffdd3eed", size = 52201, upload-time = "2024-12-17T11:02:19.512Z" },
+    { url = "https://files.pythonhosted.org/packages/80/07/cdecb7aa976f34328372f1c4efd6c9dc1b039b3cc8d3f38787d640009a25/fastrlock-0.8.3-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5f13ec08f1adb1aa916c384b05ecb7dbebb8df9ea81abd045f60941c6283a670", size = 53924, upload-time = "2024-12-17T11:02:20.85Z" },
+    { url = "https://files.pythonhosted.org/packages/88/6d/59c497f8db9a125066dd3a7442fab6aecbe90d6fec344c54645eaf311666/fastrlock-0.8.3-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:0ea4e53a04980d646def0f5e4b5e8bd8c7884288464acab0b37ca0c65c482bfe", size = 52140, upload-time = "2024-12-17T11:02:22.263Z" },
+    { url = "https://files.pythonhosted.org/packages/62/04/9138943c2ee803d62a48a3c17b69de2f6fa27677a6896c300369e839a550/fastrlock-0.8.3-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:38340f6635bd4ee2a4fb02a3a725759fe921f2ca846cb9ca44531ba739cc17b4", size = 53261, upload-time = "2024-12-17T11:02:24.418Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/ad/c8fb45d5efcdf791f0dba5c09896b39eabbdc108f5b518941a2caae52f23/fastrlock-0.8.3-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:40b328369005a0b32de14b699192aed32f549c2d2b27a5e1f614fb7ac4cec4e9", size = 49804, upload-time = "2024-12-17T11:03:25.14Z" },
+    { url = "https://files.pythonhosted.org/packages/47/15/365918306c30132bd63ae27b154e2aadb4e71c178297fc635e613aa4e767/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_24_i686.whl", hash = "sha256:6cbfb6f7731b5a280851c93883624424068fa5b22c2f546d8ae6f1fd9311e36d", size = 51763, upload-time = "2024-12-17T11:03:27.546Z" },
+    { url = "https://files.pythonhosted.org/packages/84/39/74fda02c3edeb6cc69cf5a4616e394f5636a227262788f4d33fee8401941/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1fced4cb0b3f1616be68092b70a56e9173713a4a943d02e90eb9c7897a7b5e07", size = 38277, upload-time = "2024-12-17T11:03:28.879Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/8f/86cf1dfd1d0d027110d0177946ddb34a28a6d0040331899df6dabcf9f332/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:387b2ac642938a20170a50f528817026c561882ea33306c5cbe750ae10d0a7c2", size = 51561, upload-time = "2024-12-17T11:03:30.184Z" },
+    { url = "https://files.pythonhosted.org/packages/09/5a/eabdde19fee480da1e0b3af4aef7f285d544c1ea733dc0f3df22a620df23/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5a0d31840a28d66573047d2df410eb971135a2461fb952894bf51c9533cbfea5", size = 55191, upload-time = "2024-12-17T11:03:31.604Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/e3/bdbe97b6d0d25b44bb2141c8e6be5f5bf573cf6413c9e23a7029af2d8922/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:0a9dc6fa73174f974dfb22778d05a44445b611a41d5d3776b0d5daa9e50225c6", size = 39536, upload-time = "2024-12-17T11:03:32.958Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/d0/aa12b01ea28606398bcd781b01c07dad388616029a14e065b1f0ae64d8ca/fastrlock-0.8.3-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:9842b7722e4923fe76b08d8c58a9415a9a50d4c29b80673cffeae4874ea6626a", size = 54474, upload-time = "2024-12-17T11:03:34.199Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/fb/e82f40aa6a4844107f6ace90f70b72c0cd26838a5d1984e44ec4a5d72f30/fastrlock-0.8.3-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:05029d7080c0c61a81d5fee78e842c9a1bf22552cd56129451a252655290dcef", size = 54142, upload-time = "2024-12-17T11:03:35.587Z" },
 ]
 
 [[package]]
@@ -681,11 +1004,11 @@ wheels = [
 
 [[package]]
 name = "flatbuffers"
-version = "25.2.10"
+version = "25.9.23"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/e4/30/eb5dce7994fc71a2f685d98ec33cc660c0a5887db5610137e60d8cbc4489/flatbuffers-25.2.10.tar.gz", hash = "sha256:97e451377a41262f8d9bd4295cc836133415cc03d8cb966410a4af92eb00d26e", size = 22170, upload-time = "2025-02-11T04:26:46.257Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/9d/1f/3ee70b0a55137442038f2a33469cc5fddd7e0ad2abf83d7497c18a2b6923/flatbuffers-25.9.23.tar.gz", hash = "sha256:676f9fa62750bb50cf531b42a0a2a118ad8f7f797a511eda12881c016f093b12", size = 22067, upload-time = "2025-09-24T05:25:30.106Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b8/25/155f9f080d5e4bc0082edfda032ea2bc2b8fab3f4d25d46c1e9dd22a1a89/flatbuffers-25.2.10-py2.py3-none-any.whl", hash = "sha256:ebba5f4d5ea615af3f7fd70fc310636fbb2bbd1f566ac0a23d98dd412de50051", size = 30953, upload-time = "2025-02-11T04:26:44.484Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/1b/00a78aa2e8fbd63f9af08c9c19e6deb3d5d66b4dda677a0f61654680ee89/flatbuffers-25.9.23-py2.py3-none-any.whl", hash = "sha256:255538574d6cb6d0a79a17ec8bc0d30985913b87513a01cce8bcdb6b4c44d0e2", size = 30869, upload-time = "2025-09-24T05:25:28.912Z" },
 ]
 
 [[package]]
@@ -721,11 +1044,11 @@ resolution-markers = [
 dependencies = [
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "msgpack", marker = "python_full_version == '3.10.*'" },
-    { name = "optax", version = "0.2.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "orbax-checkpoint", version = "0.11.24", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "optax", version = "0.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "orbax-checkpoint", version = "0.11.25", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "pyyaml", marker = "python_full_version == '3.10.*'" },
     { name = "rich", marker = "python_full_version == '3.10.*'" },
-    { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "tensorstore", version = "0.1.78", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "treescope", marker = "python_full_version == '3.10.*'" },
     { name = "typing-extensions", marker = "python_full_version == '3.10.*'" },
 ]
@@ -736,138 +1059,127 @@ wheels = [
 
 [[package]]
 name = "flax"
-version = "0.11.2"
+version = "0.12.0"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "msgpack", marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "optax", version = "0.2.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "orbax-checkpoint", version = "0.11.24", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "optax", version = "0.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "orbax-checkpoint", version = "0.11.25", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "pyyaml", marker = "python_full_version >= '3.11'" },
     { name = "rich", marker = "python_full_version >= '3.11'" },
-    { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "tensorstore", version = "0.1.78", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "treescope", marker = "python_full_version >= '3.11'" },
     { name = "typing-extensions", marker = "python_full_version >= '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ac/b9/3d2a5f94eb4aa9ff2e5981ac73ea91cfedef5f9de35d903e56cdcf539f31/flax-0.11.2.tar.gz", hash = "sha256:55452529b70c704128075a17f7a54d94f1f90b0f2d9498bdc9578cae3646e460", size = 5201170, upload-time = "2025-08-28T17:56:34.658Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/8b/02/2d4fbc31dcf5fb02e2b8ce733f8eecaffc7f24e905df2da39539809b195b/flax-0.12.0.tar.gz", hash = "sha256:cb3d4a2028666640c1a2e5b267dadfbd42ef14c6db41896fb4c765a7f05ebe74", size = 5038989, upload-time = "2025-09-25T23:59:00.919Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/bb/b2/5f520253823c84ca87ca0fe5a7b9d96417b1e549b98a38f5069555590d07/flax-0.11.2-py3-none-any.whl", hash = "sha256:61a5f60c27ab4e6420fda7f593f93b5ec1adba629fd33b2f0e52740ab7c04da9", size = 458093, upload-time = "2025-08-28T17:56:32.588Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/cf/d118be069d0fbbfbb6c5591069233678b573e49d81e2a387afb0297528a0/flax-0.12.0-py3-none-any.whl", hash = "sha256:13dee5f2658e8b51e22cef54b788ddbad1b6e14a9d7c5b6d1ae9c3a0110d645e", size = 466282, upload-time = "2025-09-25T23:58:58.631Z" },
 ]
 
 [[package]]
-name = "frozenlist"
-version = "1.7.0"
+name = "frozendict"
+version = "2.4.6"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/79/b1/b64018016eeb087db503b038296fd782586432b9c077fc5c7839e9cb6ef6/frozenlist-1.7.0.tar.gz", hash = "sha256:2e310d81923c2437ea8670467121cc3e9b0f76d3043cc1d2331d56c7fb7a3a8f", size = 45078, upload-time = "2025-06-09T23:02:35.538Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/af/36/0da0a49409f6b47cc2d060dc8c9040b897b5902a8a4e37d9bc1deb11f680/frozenlist-1.7.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:cc4df77d638aa2ed703b878dd093725b72a824c3c546c076e8fdf276f78ee84a", size = 81304, upload-time = "2025-06-09T22:59:46.226Z" },
-    { url = "https://files.pythonhosted.org/packages/77/f0/77c11d13d39513b298e267b22eb6cb559c103d56f155aa9a49097221f0b6/frozenlist-1.7.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:716a9973a2cc963160394f701964fe25012600f3d311f60c790400b00e568b61", size = 47735, upload-time = "2025-06-09T22:59:48.133Z" },
-    { url = "https://files.pythonhosted.org/packages/37/12/9d07fa18971a44150593de56b2f2947c46604819976784bcf6ea0d5db43b/frozenlist-1.7.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a0fd1bad056a3600047fb9462cff4c5322cebc59ebf5d0a3725e0ee78955001d", size = 46775, upload-time = "2025-06-09T22:59:49.564Z" },
-    { url = "https://files.pythonhosted.org/packages/70/34/f73539227e06288fcd1f8a76853e755b2b48bca6747e99e283111c18bcd4/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3789ebc19cb811163e70fe2bd354cea097254ce6e707ae42e56f45e31e96cb8e", size = 224644, upload-time = "2025-06-09T22:59:51.35Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/68/c1d9c2f4a6e438e14613bad0f2973567586610cc22dcb1e1241da71de9d3/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:af369aa35ee34f132fcfad5be45fbfcde0e3a5f6a1ec0712857f286b7d20cca9", size = 222125, upload-time = "2025-06-09T22:59:52.884Z" },
-    { url = "https://files.pythonhosted.org/packages/b9/d0/98e8f9a515228d708344d7c6986752be3e3192d1795f748c24bcf154ad99/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ac64b6478722eeb7a3313d494f8342ef3478dff539d17002f849101b212ef97c", size = 233455, upload-time = "2025-06-09T22:59:54.74Z" },
-    { url = "https://files.pythonhosted.org/packages/79/df/8a11bcec5600557f40338407d3e5bea80376ed1c01a6c0910fcfdc4b8993/frozenlist-1.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f89f65d85774f1797239693cef07ad4c97fdd0639544bad9ac4b869782eb1981", size = 227339, upload-time = "2025-06-09T22:59:56.187Z" },
-    { url = "https://files.pythonhosted.org/packages/50/82/41cb97d9c9a5ff94438c63cc343eb7980dac4187eb625a51bdfdb7707314/frozenlist-1.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1073557c941395fdfcfac13eb2456cb8aad89f9de27bae29fabca8e563b12615", size = 212969, upload-time = "2025-06-09T22:59:57.604Z" },
-    { url = "https://files.pythonhosted.org/packages/13/47/f9179ee5ee4f55629e4f28c660b3fdf2775c8bfde8f9c53f2de2d93f52a9/frozenlist-1.7.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1ed8d2fa095aae4bdc7fdd80351009a48d286635edffee66bf865e37a9125c50", size = 222862, upload-time = "2025-06-09T22:59:59.498Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/52/df81e41ec6b953902c8b7e3a83bee48b195cb0e5ec2eabae5d8330c78038/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:24c34bea555fe42d9f928ba0a740c553088500377448febecaa82cc3e88aa1fa", size = 222492, upload-time = "2025-06-09T23:00:01.026Z" },
-    { url = "https://files.pythonhosted.org/packages/84/17/30d6ea87fa95a9408245a948604b82c1a4b8b3e153cea596421a2aef2754/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:69cac419ac6a6baad202c85aaf467b65ac860ac2e7f2ac1686dc40dbb52f6577", size = 238250, upload-time = "2025-06-09T23:00:03.401Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/00/ecbeb51669e3c3df76cf2ddd66ae3e48345ec213a55e3887d216eb4fbab3/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:960d67d0611f4c87da7e2ae2eacf7ea81a5be967861e0c63cf205215afbfac59", size = 218720, upload-time = "2025-06-09T23:00:05.282Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/c0/c224ce0e0eb31cc57f67742071bb470ba8246623c1823a7530be0e76164c/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:41be2964bd4b15bf575e5daee5a5ce7ed3115320fb3c2b71fca05582ffa4dc9e", size = 232585, upload-time = "2025-06-09T23:00:07.962Z" },
-    { url = "https://files.pythonhosted.org/packages/55/3c/34cb694abf532f31f365106deebdeac9e45c19304d83cf7d51ebbb4ca4d1/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:46d84d49e00c9429238a7ce02dc0be8f6d7cd0cd405abd1bebdc991bf27c15bd", size = 234248, upload-time = "2025-06-09T23:00:09.428Z" },
-    { url = "https://files.pythonhosted.org/packages/98/c0/2052d8b6cecda2e70bd81299e3512fa332abb6dcd2969b9c80dfcdddbf75/frozenlist-1.7.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:15900082e886edb37480335d9d518cec978afc69ccbc30bd18610b7c1b22a718", size = 221621, upload-time = "2025-06-09T23:00:11.32Z" },
-    { url = "https://files.pythonhosted.org/packages/c5/bf/7dcebae315436903b1d98ffb791a09d674c88480c158aa171958a3ac07f0/frozenlist-1.7.0-cp310-cp310-win32.whl", hash = "sha256:400ddd24ab4e55014bba442d917203c73b2846391dd42ca5e38ff52bb18c3c5e", size = 39578, upload-time = "2025-06-09T23:00:13.526Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/5f/f69818f017fa9a3d24d1ae39763e29b7f60a59e46d5f91b9c6b21622f4cd/frozenlist-1.7.0-cp310-cp310-win_amd64.whl", hash = "sha256:6eb93efb8101ef39d32d50bce242c84bcbddb4f7e9febfa7b524532a239b4464", size = 43830, upload-time = "2025-06-09T23:00:14.98Z" },
-    { url = "https://files.pythonhosted.org/packages/34/7e/803dde33760128acd393a27eb002f2020ddb8d99d30a44bfbaab31c5f08a/frozenlist-1.7.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:aa51e147a66b2d74de1e6e2cf5921890de6b0f4820b257465101d7f37b49fb5a", size = 82251, upload-time = "2025-06-09T23:00:16.279Z" },
-    { url = "https://files.pythonhosted.org/packages/75/a9/9c2c5760b6ba45eae11334db454c189d43d34a4c0b489feb2175e5e64277/frozenlist-1.7.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:9b35db7ce1cd71d36ba24f80f0c9e7cff73a28d7a74e91fe83e23d27c7828750", size = 48183, upload-time = "2025-06-09T23:00:17.698Z" },
-    { url = "https://files.pythonhosted.org/packages/47/be/4038e2d869f8a2da165f35a6befb9158c259819be22eeaf9c9a8f6a87771/frozenlist-1.7.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:34a69a85e34ff37791e94542065c8416c1afbf820b68f720452f636d5fb990cd", size = 47107, upload-time = "2025-06-09T23:00:18.952Z" },
-    { url = "https://files.pythonhosted.org/packages/79/26/85314b8a83187c76a37183ceed886381a5f992975786f883472fcb6dc5f2/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4a646531fa8d82c87fe4bb2e596f23173caec9185bfbca5d583b4ccfb95183e2", size = 237333, upload-time = "2025-06-09T23:00:20.275Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/fd/e5b64f7d2c92a41639ffb2ad44a6a82f347787abc0c7df5f49057cf11770/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:79b2ffbba483f4ed36a0f236ccb85fbb16e670c9238313709638167670ba235f", size = 231724, upload-time = "2025-06-09T23:00:21.705Z" },
-    { url = "https://files.pythonhosted.org/packages/20/fb/03395c0a43a5976af4bf7534759d214405fbbb4c114683f434dfdd3128ef/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a26f205c9ca5829cbf82bb2a84b5c36f7184c4316617d7ef1b271a56720d6b30", size = 245842, upload-time = "2025-06-09T23:00:23.148Z" },
-    { url = "https://files.pythonhosted.org/packages/d0/15/c01c8e1dffdac5d9803507d824f27aed2ba76b6ed0026fab4d9866e82f1f/frozenlist-1.7.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bcacfad3185a623fa11ea0e0634aac7b691aa925d50a440f39b458e41c561d98", size = 239767, upload-time = "2025-06-09T23:00:25.103Z" },
-    { url = "https://files.pythonhosted.org/packages/14/99/3f4c6fe882c1f5514b6848aa0a69b20cb5e5d8e8f51a339d48c0e9305ed0/frozenlist-1.7.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:72c1b0fe8fe451b34f12dce46445ddf14bd2a5bcad7e324987194dc8e3a74c86", size = 224130, upload-time = "2025-06-09T23:00:27.061Z" },
-    { url = "https://files.pythonhosted.org/packages/4d/83/220a374bd7b2aeba9d0725130665afe11de347d95c3620b9b82cc2fcab97/frozenlist-1.7.0-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:61d1a5baeaac6c0798ff6edfaeaa00e0e412d49946c53fae8d4b8e8b3566c4ae", size = 235301, upload-time = "2025-06-09T23:00:29.02Z" },
-    { url = "https://files.pythonhosted.org/packages/03/3c/3e3390d75334a063181625343e8daab61b77e1b8214802cc4e8a1bb678fc/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:7edf5c043c062462f09b6820de9854bf28cc6cc5b6714b383149745e287181a8", size = 234606, upload-time = "2025-06-09T23:00:30.514Z" },
-    { url = "https://files.pythonhosted.org/packages/23/1e/58232c19608b7a549d72d9903005e2d82488f12554a32de2d5fb59b9b1ba/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:d50ac7627b3a1bd2dcef6f9da89a772694ec04d9a61b66cf87f7d9446b4a0c31", size = 248372, upload-time = "2025-06-09T23:00:31.966Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/a4/e4a567e01702a88a74ce8a324691e62a629bf47d4f8607f24bf1c7216e7f/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:ce48b2fece5aeb45265bb7a58259f45027db0abff478e3077e12b05b17fb9da7", size = 229860, upload-time = "2025-06-09T23:00:33.375Z" },
-    { url = "https://files.pythonhosted.org/packages/73/a6/63b3374f7d22268b41a9db73d68a8233afa30ed164c46107b33c4d18ecdd/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:fe2365ae915a1fafd982c146754e1de6ab3478def8a59c86e1f7242d794f97d5", size = 245893, upload-time = "2025-06-09T23:00:35.002Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/eb/d18b3f6e64799a79673c4ba0b45e4cfbe49c240edfd03a68be20002eaeaa/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:45a6f2fdbd10e074e8814eb98b05292f27bad7d1883afbe009d96abdcf3bc898", size = 246323, upload-time = "2025-06-09T23:00:36.468Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/f5/720f3812e3d06cd89a1d5db9ff6450088b8f5c449dae8ffb2971a44da506/frozenlist-1.7.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:21884e23cffabb157a9dd7e353779077bf5b8f9a58e9b262c6caad2ef5f80a56", size = 233149, upload-time = "2025-06-09T23:00:37.963Z" },
-    { url = "https://files.pythonhosted.org/packages/69/68/03efbf545e217d5db8446acfd4c447c15b7c8cf4dbd4a58403111df9322d/frozenlist-1.7.0-cp311-cp311-win32.whl", hash = "sha256:284d233a8953d7b24f9159b8a3496fc1ddc00f4db99c324bd5fb5f22d8698ea7", size = 39565, upload-time = "2025-06-09T23:00:39.753Z" },
-    { url = "https://files.pythonhosted.org/packages/58/17/fe61124c5c333ae87f09bb67186d65038834a47d974fc10a5fadb4cc5ae1/frozenlist-1.7.0-cp311-cp311-win_amd64.whl", hash = "sha256:387cbfdcde2f2353f19c2f66bbb52406d06ed77519ac7ee21be0232147c2592d", size = 44019, upload-time = "2025-06-09T23:00:40.988Z" },
-    { url = "https://files.pythonhosted.org/packages/ef/a2/c8131383f1e66adad5f6ecfcce383d584ca94055a34d683bbb24ac5f2f1c/frozenlist-1.7.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:3dbf9952c4bb0e90e98aec1bd992b3318685005702656bc6f67c1a32b76787f2", size = 81424, upload-time = "2025-06-09T23:00:42.24Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/9d/02754159955088cb52567337d1113f945b9e444c4960771ea90eb73de8db/frozenlist-1.7.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:1f5906d3359300b8a9bb194239491122e6cf1444c2efb88865426f170c262cdb", size = 47952, upload-time = "2025-06-09T23:00:43.481Z" },
-    { url = "https://files.pythonhosted.org/packages/01/7a/0046ef1bd6699b40acd2067ed6d6670b4db2f425c56980fa21c982c2a9db/frozenlist-1.7.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3dabd5a8f84573c8d10d8859a50ea2dec01eea372031929871368c09fa103478", size = 46688, upload-time = "2025-06-09T23:00:44.793Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/a2/a910bafe29c86997363fb4c02069df4ff0b5bc39d33c5198b4e9dd42d8f8/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:aa57daa5917f1738064f302bf2626281a1cb01920c32f711fbc7bc36111058a8", size = 243084, upload-time = "2025-06-09T23:00:46.125Z" },
-    { url = "https://files.pythonhosted.org/packages/64/3e/5036af9d5031374c64c387469bfcc3af537fc0f5b1187d83a1cf6fab1639/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:c193dda2b6d49f4c4398962810fa7d7c78f032bf45572b3e04dd5249dff27e08", size = 233524, upload-time = "2025-06-09T23:00:47.73Z" },
-    { url = "https://files.pythonhosted.org/packages/06/39/6a17b7c107a2887e781a48ecf20ad20f1c39d94b2a548c83615b5b879f28/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bfe2b675cf0aaa6d61bf8fbffd3c274b3c9b7b1623beb3809df8a81399a4a9c4", size = 248493, upload-time = "2025-06-09T23:00:49.742Z" },
-    { url = "https://files.pythonhosted.org/packages/be/00/711d1337c7327d88c44d91dd0f556a1c47fb99afc060ae0ef66b4d24793d/frozenlist-1.7.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8fc5d5cda37f62b262405cf9652cf0856839c4be8ee41be0afe8858f17f4c94b", size = 244116, upload-time = "2025-06-09T23:00:51.352Z" },
-    { url = "https://files.pythonhosted.org/packages/24/fe/74e6ec0639c115df13d5850e75722750adabdc7de24e37e05a40527ca539/frozenlist-1.7.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b0d5ce521d1dd7d620198829b87ea002956e4319002ef0bc8d3e6d045cb4646e", size = 224557, upload-time = "2025-06-09T23:00:52.855Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/db/48421f62a6f77c553575201e89048e97198046b793f4a089c79a6e3268bd/frozenlist-1.7.0-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:488d0a7d6a0008ca0db273c542098a0fa9e7dfaa7e57f70acef43f32b3f69dca", size = 241820, upload-time = "2025-06-09T23:00:54.43Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/fa/cb4a76bea23047c8462976ea7b7a2bf53997a0ca171302deae9d6dd12096/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:15a7eaba63983d22c54d255b854e8108e7e5f3e89f647fc854bd77a237e767df", size = 236542, upload-time = "2025-06-09T23:00:56.409Z" },
-    { url = "https://files.pythonhosted.org/packages/5d/32/476a4b5cfaa0ec94d3f808f193301debff2ea42288a099afe60757ef6282/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:1eaa7e9c6d15df825bf255649e05bd8a74b04a4d2baa1ae46d9c2d00b2ca2cb5", size = 249350, upload-time = "2025-06-09T23:00:58.468Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/ba/9a28042f84a6bf8ea5dbc81cfff8eaef18d78b2a1ad9d51c7bc5b029ad16/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:e4389e06714cfa9d47ab87f784a7c5be91d3934cd6e9a7b85beef808297cc025", size = 225093, upload-time = "2025-06-09T23:01:00.015Z" },
-    { url = "https://files.pythonhosted.org/packages/bc/29/3a32959e68f9cf000b04e79ba574527c17e8842e38c91d68214a37455786/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:73bd45e1488c40b63fe5a7df892baf9e2a4d4bb6409a2b3b78ac1c6236178e01", size = 245482, upload-time = "2025-06-09T23:01:01.474Z" },
-    { url = "https://files.pythonhosted.org/packages/80/e8/edf2f9e00da553f07f5fa165325cfc302dead715cab6ac8336a5f3d0adc2/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:99886d98e1643269760e5fe0df31e5ae7050788dd288947f7f007209b8c33f08", size = 249590, upload-time = "2025-06-09T23:01:02.961Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/80/9a0eb48b944050f94cc51ee1c413eb14a39543cc4f760ed12657a5a3c45a/frozenlist-1.7.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:290a172aae5a4c278c6da8a96222e6337744cd9c77313efe33d5670b9f65fc43", size = 237785, upload-time = "2025-06-09T23:01:05.095Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/74/87601e0fb0369b7a2baf404ea921769c53b7ae00dee7dcfe5162c8c6dbf0/frozenlist-1.7.0-cp312-cp312-win32.whl", hash = "sha256:426c7bc70e07cfebc178bc4c2bf2d861d720c4fff172181eeb4a4c41d4ca2ad3", size = 39487, upload-time = "2025-06-09T23:01:06.54Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/15/c026e9a9fc17585a9d461f65d8593d281fedf55fbf7eb53f16c6df2392f9/frozenlist-1.7.0-cp312-cp312-win_amd64.whl", hash = "sha256:563b72efe5da92e02eb68c59cb37205457c977aa7a449ed1b37e6939e5c47c6a", size = 43874, upload-time = "2025-06-09T23:01:07.752Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/b1/ee59496f51cd244039330015d60f13ce5a54a0f2bd8d79e4a4a375ab7469/frozenlist-1.7.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:cea3dbd15aea1341ea2de490574a4a37ca080b2ae24e4b4f4b51b9057b4c3630", size = 82434, upload-time = "2025-06-09T23:02:05.195Z" },
-    { url = "https://files.pythonhosted.org/packages/75/e1/d518391ce36a6279b3fa5bc14327dde80bcb646bb50d059c6ca0756b8d05/frozenlist-1.7.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7d536ee086b23fecc36c2073c371572374ff50ef4db515e4e503925361c24f71", size = 48232, upload-time = "2025-06-09T23:02:07.728Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/8d/a0d04f28b6e821a9685c22e67b5fb798a5a7b68752f104bfbc2dccf080c4/frozenlist-1.7.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:dfcebf56f703cb2e346315431699f00db126d158455e513bd14089d992101e44", size = 47186, upload-time = "2025-06-09T23:02:09.243Z" },
-    { url = "https://files.pythonhosted.org/packages/93/3a/a5334c0535c8b7c78eeabda1579179e44fe3d644e07118e59a2276dedaf1/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:974c5336e61d6e7eb1ea5b929cb645e882aadab0095c5a6974a111e6479f8878", size = 226617, upload-time = "2025-06-09T23:02:10.949Z" },
-    { url = "https://files.pythonhosted.org/packages/0a/67/8258d971f519dc3f278c55069a775096cda6610a267b53f6248152b72b2f/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:c70db4a0ab5ab20878432c40563573229a7ed9241506181bba12f6b7d0dc41cb", size = 224179, upload-time = "2025-06-09T23:02:12.603Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/89/8225905bf889b97c6d935dd3aeb45668461e59d415cb019619383a8a7c3b/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1137b78384eebaf70560a36b7b229f752fb64d463d38d1304939984d5cb887b6", size = 235783, upload-time = "2025-06-09T23:02:14.678Z" },
-    { url = "https://files.pythonhosted.org/packages/54/6e/ef52375aa93d4bc510d061df06205fa6dcfd94cd631dd22956b09128f0d4/frozenlist-1.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e793a9f01b3e8b5c0bc646fb59140ce0efcc580d22a3468d70766091beb81b35", size = 229210, upload-time = "2025-06-09T23:02:16.313Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/55/62c87d1a6547bfbcd645df10432c129100c5bd0fd92a384de6e3378b07c1/frozenlist-1.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:74739ba8e4e38221d2c5c03d90a7e542cb8ad681915f4ca8f68d04f810ee0a87", size = 215994, upload-time = "2025-06-09T23:02:17.9Z" },
-    { url = "https://files.pythonhosted.org/packages/45/d2/263fea1f658b8ad648c7d94d18a87bca7e8c67bd6a1bbf5445b1bd5b158c/frozenlist-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1e63344c4e929b1a01e29bc184bbb5fd82954869033765bfe8d65d09e336a677", size = 225122, upload-time = "2025-06-09T23:02:19.479Z" },
-    { url = "https://files.pythonhosted.org/packages/7b/22/7145e35d12fb368d92124f679bea87309495e2e9ddf14c6533990cb69218/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:2ea2a7369eb76de2217a842f22087913cdf75f63cf1307b9024ab82dfb525938", size = 224019, upload-time = "2025-06-09T23:02:20.969Z" },
-    { url = "https://files.pythonhosted.org/packages/44/1e/7dae8c54301beb87bcafc6144b9a103bfd2c8f38078c7902984c9a0c4e5b/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:836b42f472a0e006e02499cef9352ce8097f33df43baaba3e0a28a964c26c7d2", size = 239925, upload-time = "2025-06-09T23:02:22.466Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/1e/99c93e54aa382e949a98976a73b9b20c3aae6d9d893f31bbe4991f64e3a8/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:e22b9a99741294b2571667c07d9f8cceec07cb92aae5ccda39ea1b6052ed4319", size = 220881, upload-time = "2025-06-09T23:02:24.521Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/9c/ca5105fa7fb5abdfa8837581be790447ae051da75d32f25c8f81082ffc45/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:9a19e85cc503d958abe5218953df722748d87172f71b73cf3c9257a91b999890", size = 234046, upload-time = "2025-06-09T23:02:26.206Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/4d/e99014756093b4ddbb67fb8f0df11fe7a415760d69ace98e2ac6d5d43402/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:f22dac33bb3ee8fe3e013aa7b91dc12f60d61d05b7fe32191ffa84c3aafe77bd", size = 235756, upload-time = "2025-06-09T23:02:27.79Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/72/a19a40bcdaa28a51add2aaa3a1a294ec357f36f27bd836a012e070c5e8a5/frozenlist-1.7.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:9ccec739a99e4ccf664ea0775149f2749b8a6418eb5b8384b4dc0a7d15d304cb", size = 222894, upload-time = "2025-06-09T23:02:29.848Z" },
-    { url = "https://files.pythonhosted.org/packages/08/49/0042469993e023a758af81db68c76907cd29e847d772334d4d201cbe9a42/frozenlist-1.7.0-cp39-cp39-win32.whl", hash = "sha256:b3950f11058310008a87757f3eee16a8e1ca97979833239439586857bc25482e", size = 39848, upload-time = "2025-06-09T23:02:31.413Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/45/827d86ee475c877f5f766fbc23fb6acb6fada9e52f1c9720e2ba3eae32da/frozenlist-1.7.0-cp39-cp39-win_amd64.whl", hash = "sha256:43a82fce6769c70f2f5a06248b614a7d268080a9d20f7457ef10ecee5af82b63", size = 44102, upload-time = "2025-06-09T23:02:32.808Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/45/b82e3c16be2182bff01179db177fe144d58b5dc787a7d4492c6ed8b9317f/frozenlist-1.7.0-py3-none-any.whl", hash = "sha256:9a5af342e34f7e97caf8c995864c7a396418ae2859cc6fdf1b1073020d516a7e", size = 13106, upload-time = "2025-06-09T23:02:34.204Z" },
-]
-
-[[package]]
-name = "fschat"
-version = "0.2.36"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "aiohttp" },
-    { name = "fastapi" },
-    { name = "httpx" },
-    { name = "markdown2", extra = ["all"] },
-    { name = "nh3" },
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "prompt-toolkit" },
-    { name = "pydantic" },
-    { name = "requests" },
-    { name = "rich" },
-    { name = "shortuuid" },
-    { name = "tiktoken" },
-    { name = "uvicorn" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/22/ad/8336e2dc6323944d59e888d41ff8833993ebc9c79bdd8d07997c54d92d55/fschat-0.2.36.tar.gz", hash = "sha256:ac724458bb105cb5b3b9ec852150345dfa7f61fd7c275d6635d66fa1ed904d90", size = 195280, upload-time = "2024-02-11T15:42:27.02Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/bb/59/19eb300ba28e7547538bdf603f1c6c34793240a90e1a7b61b65d8517e35e/frozendict-2.4.6.tar.gz", hash = "sha256:df7cd16470fbd26fc4969a208efadc46319334eb97def1ddf48919b351192b8e", size = 316416, upload-time = "2024-10-13T12:15:32.449Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/43/cd/3651634ab25f74e64eaf6a6e08e359320d6b26e637afe622762fe99d127b/fschat-0.2.36-py3-none-any.whl", hash = "sha256:a66ecf1e90f6db1d015096f6aa8a305d20d0a13e067951285aecdb56a33e9cb1", size = 256902, upload-time = "2024-02-11T15:42:24.819Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/f1/a10be024a9d53441c997b3661ea80ecba6e3130adc53812a4b95b607cdd1/frozendict-2.4.6-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f4c789fd70879ccb6289a603cdebdc4953e7e5dea047d30c1b180529b28257b5", size = 117656, upload-time = "2024-10-13T12:14:22.038Z" },
+    { url = "https://files.pythonhosted.org/packages/46/a6/34c760975e6f1cb4db59a990d58dcf22287e10241c851804670c74c6a27a/frozendict-2.4.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:da6a10164c8a50b34b9ab508a9420df38f4edf286b9ca7b7df8a91767baecb34", size = 117444, upload-time = "2024-10-13T12:14:24.251Z" },
+    { url = "https://files.pythonhosted.org/packages/62/dd/64bddd1ffa9617f50e7e63656b2a7ad7f0a46c86b5f4a3d2c714d0006277/frozendict-2.4.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:9a8a43036754a941601635ea9c788ebd7a7efbed2becba01b54a887b41b175b9", size = 116801, upload-time = "2024-10-13T12:14:26.518Z" },
+    { url = "https://files.pythonhosted.org/packages/45/ae/af06a8bde1947277aad895c2f26c3b8b8b6ee9c0c2ad988fb58a9d1dde3f/frozendict-2.4.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:c9905dcf7aa659e6a11b8051114c9fa76dfde3a6e50e6dc129d5aece75b449a2", size = 117329, upload-time = "2024-10-13T12:14:28.485Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/a2/5a178339345edff643240e48dd276581df64b1dd93eaa7d26556396a145b/frozendict-2.4.6-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:669237c571856be575eca28a69e92a3d18f8490511eff184937283dc6093bd67", size = 117093, upload-time = "2024-10-13T12:15:15.621Z" },
+    { url = "https://files.pythonhosted.org/packages/41/df/09a752239eb0661eeda0f34f14577c10edc6f3e4deb7652b3a3efff22ad4/frozendict-2.4.6-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0aaa11e7c472150efe65adbcd6c17ac0f586896096ab3963775e1c5c58ac0098", size = 116883, upload-time = "2024-10-13T12:15:17.521Z" },
+    { url = "https://files.pythonhosted.org/packages/22/d4/619d1cfbc74be5641d839a5a2e292f9eac494aa557bfe7c266542c4014a2/frozendict-2.4.6-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:b8f2829048f29fe115da4a60409be2130e69402e29029339663fac39c90e6e2b", size = 116314, upload-time = "2024-10-13T12:15:19.689Z" },
+    { url = "https://files.pythonhosted.org/packages/41/b9/40042606a4ac458046ebeecc34cec2971e78e029ea3b6ad4e35833c7f8e6/frozendict-2.4.6-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:94321e646cc39bebc66954a31edd1847d3a2a3483cf52ff051cd0996e7db07db", size = 117017, upload-time = "2024-10-13T12:15:21.718Z" },
+    { url = "https://files.pythonhosted.org/packages/04/13/d9839089b900fa7b479cce495d62110cddc4bd5630a04d8469916c0e79c5/frozendict-2.4.6-py311-none-any.whl", hash = "sha256:d065db6a44db2e2375c23eac816f1a022feb2fa98cbb50df44a9e83700accbea", size = 16148, upload-time = "2024-10-13T12:15:26.839Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/d0/d482c39cee2ab2978a892558cf130681d4574ea208e162da8958b31e9250/frozendict-2.4.6-py312-none-any.whl", hash = "sha256:49344abe90fb75f0f9fdefe6d4ef6d4894e640fadab71f11009d52ad97f370b9", size = 16146, upload-time = "2024-10-13T12:15:28.16Z" },
+]
+
+[[package]]
+name = "frozenlist"
+version = "1.8.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/2d/f5/c831fac6cc817d26fd54c7eaccd04ef7e0288806943f7cc5bbf69f3ac1f0/frozenlist-1.8.0.tar.gz", hash = "sha256:3ede829ed8d842f6cd48fc7081d7a41001a56f1f38603f9d49bf3020d59a31ad", size = 45875, upload-time = "2025-10-06T05:38:17.865Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/83/4a/557715d5047da48d54e659203b9335be7bfaafda2c3f627b7c47e0b3aaf3/frozenlist-1.8.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:b37f6d31b3dcea7deb5e9696e529a6aa4a898adc33db82da12e4c60a7c4d2011", size = 86230, upload-time = "2025-10-06T05:35:23.699Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/fb/c85f9fed3ea8fe8740e5b46a59cc141c23b842eca617da8876cfce5f760e/frozenlist-1.8.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:ef2b7b394f208233e471abc541cc6991f907ffd47dc72584acee3147899d6565", size = 49621, upload-time = "2025-10-06T05:35:25.341Z" },
+    { url = "https://files.pythonhosted.org/packages/63/70/26ca3f06aace16f2352796b08704338d74b6d1a24ca38f2771afbb7ed915/frozenlist-1.8.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a88f062f072d1589b7b46e951698950e7da00442fc1cacbe17e19e025dc327ad", size = 49889, upload-time = "2025-10-06T05:35:26.797Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/ed/c7895fd2fde7f3ee70d248175f9b6cdf792fb741ab92dc59cd9ef3bd241b/frozenlist-1.8.0-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f57fb59d9f385710aa7060e89410aeb5058b99e62f4d16b08b91986b9a2140c2", size = 219464, upload-time = "2025-10-06T05:35:28.254Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/83/4d587dccbfca74cb8b810472392ad62bfa100bf8108c7223eb4c4fa2f7b3/frozenlist-1.8.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:799345ab092bee59f01a915620b5d014698547afd011e691a208637312db9186", size = 221649, upload-time = "2025-10-06T05:35:29.454Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/c6/fd3b9cd046ec5fff9dab66831083bc2077006a874a2d3d9247dea93ddf7e/frozenlist-1.8.0-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:c23c3ff005322a6e16f71bf8692fcf4d5a304aaafe1e262c98c6d4adc7be863e", size = 219188, upload-time = "2025-10-06T05:35:30.951Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/80/6693f55eb2e085fc8afb28cf611448fb5b90e98e068fa1d1b8d8e66e5c7d/frozenlist-1.8.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8a76ea0f0b9dfa06f254ee06053d93a600865b3274358ca48a352ce4f0798450", size = 231748, upload-time = "2025-10-06T05:35:32.101Z" },
+    { url = "https://files.pythonhosted.org/packages/97/d6/e9459f7c5183854abd989ba384fe0cc1a0fb795a83c033f0571ec5933ca4/frozenlist-1.8.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:c7366fe1418a6133d5aa824ee53d406550110984de7637d65a178010f759c6ef", size = 236351, upload-time = "2025-10-06T05:35:33.834Z" },
+    { url = "https://files.pythonhosted.org/packages/97/92/24e97474b65c0262e9ecd076e826bfd1d3074adcc165a256e42e7b8a7249/frozenlist-1.8.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:13d23a45c4cebade99340c4165bd90eeb4a56c6d8a9d8aa49568cac19a6d0dc4", size = 218767, upload-time = "2025-10-06T05:35:35.205Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/bf/dc394a097508f15abff383c5108cb8ad880d1f64a725ed3b90d5c2fbf0bb/frozenlist-1.8.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:e4a3408834f65da56c83528fb52ce7911484f0d1eaf7b761fc66001db1646eff", size = 235887, upload-time = "2025-10-06T05:35:36.354Z" },
+    { url = "https://files.pythonhosted.org/packages/40/90/25b201b9c015dbc999a5baf475a257010471a1fa8c200c843fd4abbee725/frozenlist-1.8.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:42145cd2748ca39f32801dad54aeea10039da6f86e303659db90db1c4b614c8c", size = 228785, upload-time = "2025-10-06T05:35:37.949Z" },
+    { url = "https://files.pythonhosted.org/packages/84/f4/b5bc148df03082f05d2dd30c089e269acdbe251ac9a9cf4e727b2dbb8a3d/frozenlist-1.8.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:e2de870d16a7a53901e41b64ffdf26f2fbb8917b3e6ebf398098d72c5b20bd7f", size = 230312, upload-time = "2025-10-06T05:35:39.178Z" },
+    { url = "https://files.pythonhosted.org/packages/db/4b/87e95b5d15097c302430e647136b7d7ab2398a702390cf4c8601975709e7/frozenlist-1.8.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:20e63c9493d33ee48536600d1a5c95eefc870cd71e7ab037763d1fbb89cc51e7", size = 217650, upload-time = "2025-10-06T05:35:40.377Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/70/78a0315d1fea97120591a83e0acd644da638c872f142fd72a6cebee825f3/frozenlist-1.8.0-cp310-cp310-win32.whl", hash = "sha256:adbeebaebae3526afc3c96fad434367cafbfd1b25d72369a9e5858453b1bb71a", size = 39659, upload-time = "2025-10-06T05:35:41.863Z" },
+    { url = "https://files.pythonhosted.org/packages/66/aa/3f04523fb189a00e147e60c5b2205126118f216b0aa908035c45336e27e4/frozenlist-1.8.0-cp310-cp310-win_amd64.whl", hash = "sha256:667c3777ca571e5dbeb76f331562ff98b957431df140b54c85fd4d52eea8d8f6", size = 43837, upload-time = "2025-10-06T05:35:43.205Z" },
+    { url = "https://files.pythonhosted.org/packages/39/75/1135feecdd7c336938bd55b4dc3b0dfc46d85b9be12ef2628574b28de776/frozenlist-1.8.0-cp310-cp310-win_arm64.whl", hash = "sha256:80f85f0a7cc86e7a54c46d99c9e1318ff01f4687c172ede30fd52d19d1da1c8e", size = 39989, upload-time = "2025-10-06T05:35:44.596Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/03/077f869d540370db12165c0aa51640a873fb661d8b315d1d4d67b284d7ac/frozenlist-1.8.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:09474e9831bc2b2199fad6da3c14c7b0fbdd377cce9d3d77131be28906cb7d84", size = 86912, upload-time = "2025-10-06T05:35:45.98Z" },
+    { url = "https://files.pythonhosted.org/packages/df/b5/7610b6bd13e4ae77b96ba85abea1c8cb249683217ef09ac9e0ae93f25a91/frozenlist-1.8.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:17c883ab0ab67200b5f964d2b9ed6b00971917d5d8a92df149dc2c9779208ee9", size = 50046, upload-time = "2025-10-06T05:35:47.009Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/ef/0e8f1fe32f8a53dd26bdd1f9347efe0778b0fddf62789ea683f4cc7d787d/frozenlist-1.8.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:fa47e444b8ba08fffd1c18e8cdb9a75db1b6a27f17507522834ad13ed5922b93", size = 50119, upload-time = "2025-10-06T05:35:48.38Z" },
+    { url = "https://files.pythonhosted.org/packages/11/b1/71a477adc7c36e5fb628245dfbdea2166feae310757dea848d02bd0689fd/frozenlist-1.8.0-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:2552f44204b744fba866e573be4c1f9048d6a324dfe14475103fd51613eb1d1f", size = 231067, upload-time = "2025-10-06T05:35:49.97Z" },
+    { url = "https://files.pythonhosted.org/packages/45/7e/afe40eca3a2dc19b9904c0f5d7edfe82b5304cb831391edec0ac04af94c2/frozenlist-1.8.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:957e7c38f250991e48a9a73e6423db1bb9dd14e722a10f6b8bb8e16a0f55f695", size = 233160, upload-time = "2025-10-06T05:35:51.729Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/aa/7416eac95603ce428679d273255ffc7c998d4132cfae200103f164b108aa/frozenlist-1.8.0-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:8585e3bb2cdea02fc88ffa245069c36555557ad3609e83be0ec71f54fd4abb52", size = 228544, upload-time = "2025-10-06T05:35:53.246Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/3d/2a2d1f683d55ac7e3875e4263d28410063e738384d3adc294f5ff3d7105e/frozenlist-1.8.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:edee74874ce20a373d62dc28b0b18b93f645633c2943fd90ee9d898550770581", size = 243797, upload-time = "2025-10-06T05:35:54.497Z" },
+    { url = "https://files.pythonhosted.org/packages/78/1e/2d5565b589e580c296d3bb54da08d206e797d941a83a6fdea42af23be79c/frozenlist-1.8.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:c9a63152fe95756b85f31186bddf42e4c02c6321207fd6601a1c89ebac4fe567", size = 247923, upload-time = "2025-10-06T05:35:55.861Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/c3/65872fcf1d326a7f101ad4d86285c403c87be7d832b7470b77f6d2ed5ddc/frozenlist-1.8.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b6db2185db9be0a04fecf2f241c70b63b1a242e2805be291855078f2b404dd6b", size = 230886, upload-time = "2025-10-06T05:35:57.399Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/76/ac9ced601d62f6956f03cc794f9e04c81719509f85255abf96e2510f4265/frozenlist-1.8.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:f4be2e3d8bc8aabd566f8d5b8ba7ecc09249d74ba3c9ed52e54dc23a293f0b92", size = 245731, upload-time = "2025-10-06T05:35:58.563Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/49/ecccb5f2598daf0b4a1415497eba4c33c1e8ce07495eb07d2860c731b8d5/frozenlist-1.8.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:c8d1634419f39ea6f5c427ea2f90ca85126b54b50837f31497f3bf38266e853d", size = 241544, upload-time = "2025-10-06T05:35:59.719Z" },
+    { url = "https://files.pythonhosted.org/packages/53/4b/ddf24113323c0bbcc54cb38c8b8916f1da7165e07b8e24a717b4a12cbf10/frozenlist-1.8.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:1a7fa382a4a223773ed64242dbe1c9c326ec09457e6b8428efb4118c685c3dfd", size = 241806, upload-time = "2025-10-06T05:36:00.959Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/fb/9b9a084d73c67175484ba2789a59f8eebebd0827d186a8102005ce41e1ba/frozenlist-1.8.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:11847b53d722050808926e785df837353bd4d75f1d494377e59b23594d834967", size = 229382, upload-time = "2025-10-06T05:36:02.22Z" },
+    { url = "https://files.pythonhosted.org/packages/95/a3/c8fb25aac55bf5e12dae5c5aa6a98f85d436c1dc658f21c3ac73f9fa95e5/frozenlist-1.8.0-cp311-cp311-win32.whl", hash = "sha256:27c6e8077956cf73eadd514be8fb04d77fc946a7fe9f7fe167648b0b9085cc25", size = 39647, upload-time = "2025-10-06T05:36:03.409Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/f5/603d0d6a02cfd4c8f2a095a54672b3cf967ad688a60fb9faf04fc4887f65/frozenlist-1.8.0-cp311-cp311-win_amd64.whl", hash = "sha256:ac913f8403b36a2c8610bbfd25b8013488533e71e62b4b4adce9c86c8cea905b", size = 44064, upload-time = "2025-10-06T05:36:04.368Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/16/c2c9ab44e181f043a86f9a8f84d5124b62dbcb3a02c0977ec72b9ac1d3e0/frozenlist-1.8.0-cp311-cp311-win_arm64.whl", hash = "sha256:d4d3214a0f8394edfa3e303136d0575eece0745ff2b47bd2cb2e66dd92d4351a", size = 39937, upload-time = "2025-10-06T05:36:05.669Z" },
+    { url = "https://files.pythonhosted.org/packages/69/29/948b9aa87e75820a38650af445d2ef2b6b8a6fab1a23b6bb9e4ef0be2d59/frozenlist-1.8.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:78f7b9e5d6f2fdb88cdde9440dc147259b62b9d3b019924def9f6478be254ac1", size = 87782, upload-time = "2025-10-06T05:36:06.649Z" },
+    { url = "https://files.pythonhosted.org/packages/64/80/4f6e318ee2a7c0750ed724fa33a4bdf1eacdc5a39a7a24e818a773cd91af/frozenlist-1.8.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:229bf37d2e4acdaf808fd3f06e854a4a7a3661e871b10dc1f8f1896a3b05f18b", size = 50594, upload-time = "2025-10-06T05:36:07.69Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/94/5c8a2b50a496b11dd519f4a24cb5496cf125681dd99e94c604ccdea9419a/frozenlist-1.8.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:f833670942247a14eafbb675458b4e61c82e002a148f49e68257b79296e865c4", size = 50448, upload-time = "2025-10-06T05:36:08.78Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/bd/d91c5e39f490a49df14320f4e8c80161cfcce09f1e2cde1edd16a551abb3/frozenlist-1.8.0-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:494a5952b1c597ba44e0e78113a7266e656b9794eec897b19ead706bd7074383", size = 242411, upload-time = "2025-10-06T05:36:09.801Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/83/f61505a05109ef3293dfb1ff594d13d64a2324ac3482be2cedc2be818256/frozenlist-1.8.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:96f423a119f4777a4a056b66ce11527366a8bb92f54e541ade21f2374433f6d4", size = 243014, upload-time = "2025-10-06T05:36:11.394Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/cb/cb6c7b0f7d4023ddda30cf56b8b17494eb3a79e3fda666bf735f63118b35/frozenlist-1.8.0-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3462dd9475af2025c31cc61be6652dfa25cbfb56cbbf52f4ccfe029f38decaf8", size = 234909, upload-time = "2025-10-06T05:36:12.598Z" },
+    { url = "https://files.pythonhosted.org/packages/31/c5/cd7a1f3b8b34af009fb17d4123c5a778b44ae2804e3ad6b86204255f9ec5/frozenlist-1.8.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c4c800524c9cd9bac5166cd6f55285957fcfc907db323e193f2afcd4d9abd69b", size = 250049, upload-time = "2025-10-06T05:36:14.065Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/01/2f95d3b416c584a1e7f0e1d6d31998c4a795f7544069ee2e0962a4b60740/frozenlist-1.8.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:d6a5df73acd3399d893dafc71663ad22534b5aa4f94e8a2fabfe856c3c1b6a52", size = 256485, upload-time = "2025-10-06T05:36:15.39Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/03/024bf7720b3abaebcff6d0793d73c154237b85bdf67b7ed55e5e9596dc9a/frozenlist-1.8.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:405e8fe955c2280ce66428b3ca55e12b3c4e9c336fb2103a4937e891c69a4a29", size = 237619, upload-time = "2025-10-06T05:36:16.558Z" },
+    { url = "https://files.pythonhosted.org/packages/69/fa/f8abdfe7d76b731f5d8bd217827cf6764d4f1d9763407e42717b4bed50a0/frozenlist-1.8.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:908bd3f6439f2fef9e85031b59fd4f1297af54415fb60e4254a95f75b3cab3f3", size = 250320, upload-time = "2025-10-06T05:36:17.821Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/3c/b051329f718b463b22613e269ad72138cc256c540f78a6de89452803a47d/frozenlist-1.8.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:294e487f9ec720bd8ffcebc99d575f7eff3568a08a253d1ee1a0378754b74143", size = 246820, upload-time = "2025-10-06T05:36:19.046Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/ae/58282e8f98e444b3f4dd42448ff36fa38bef29e40d40f330b22e7108f565/frozenlist-1.8.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:74c51543498289c0c43656701be6b077f4b265868fa7f8a8859c197006efb608", size = 250518, upload-time = "2025-10-06T05:36:20.763Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/96/007e5944694d66123183845a106547a15944fbbb7154788cbf7272789536/frozenlist-1.8.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:776f352e8329135506a1d6bf16ac3f87bc25b28e765949282dcc627af36123aa", size = 239096, upload-time = "2025-10-06T05:36:22.129Z" },
+    { url = "https://files.pythonhosted.org/packages/66/bb/852b9d6db2fa40be96f29c0d1205c306288f0684df8fd26ca1951d461a56/frozenlist-1.8.0-cp312-cp312-win32.whl", hash = "sha256:433403ae80709741ce34038da08511d4a77062aa924baf411ef73d1146e74faf", size = 39985, upload-time = "2025-10-06T05:36:23.661Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/af/38e51a553dd66eb064cdf193841f16f077585d4d28394c2fa6235cb41765/frozenlist-1.8.0-cp312-cp312-win_amd64.whl", hash = "sha256:34187385b08f866104f0c0617404c8eb08165ab1272e884abc89c112e9c00746", size = 44591, upload-time = "2025-10-06T05:36:24.958Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/06/1dc65480ab147339fecc70797e9c2f69d9cea9cf38934ce08df070fdb9cb/frozenlist-1.8.0-cp312-cp312-win_arm64.whl", hash = "sha256:fe3c58d2f5db5fbd18c2987cba06d51b0529f52bc3a6cdc33d3f4eab725104bd", size = 40102, upload-time = "2025-10-06T05:36:26.333Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/59/ae5cdac87a00962122ea37bb346d41b66aec05f9ce328fa2b9e216f8967b/frozenlist-1.8.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:d8b7138e5cd0647e4523d6685b0eac5d4be9a184ae9634492f25c6eb38c12a47", size = 86967, upload-time = "2025-10-06T05:37:55.607Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/10/17059b2db5a032fd9323c41c39e9d1f5f9d0c8f04d1e4e3e788573086e61/frozenlist-1.8.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:a6483e309ca809f1efd154b4d37dc6d9f61037d6c6a81c2dc7a15cb22c8c5dca", size = 49984, upload-time = "2025-10-06T05:37:57.049Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/de/ad9d82ca8e5fa8f0c636e64606553c79e2b859ad253030b62a21fe9986f5/frozenlist-1.8.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:1b9290cf81e95e93fdf90548ce9d3c1211cf574b8e3f4b3b7cb0537cf2227068", size = 50240, upload-time = "2025-10-06T05:37:58.145Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/45/3dfb7767c2a67d123650122b62ce13c731b6c745bc14424eea67678b508c/frozenlist-1.8.0-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:59a6a5876ca59d1b63af8cd5e7ffffb024c3dc1e9cf9301b21a2e76286505c95", size = 219472, upload-time = "2025-10-06T05:37:59.239Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/bf/5bf23d913a741b960d5c1dac7c1985d8a2a1d015772b2d18ea168b08e7ff/frozenlist-1.8.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6dc4126390929823e2d2d9dc79ab4046ed74680360fc5f38b585c12c66cdf459", size = 221531, upload-time = "2025-10-06T05:38:00.521Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/03/27ec393f3b55860859f4b74cdc8c2a4af3dbf3533305e8eacf48a4fd9a54/frozenlist-1.8.0-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:332db6b2563333c5671fecacd085141b5800cb866be16d5e3eb15a2086476675", size = 219211, upload-time = "2025-10-06T05:38:01.842Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/ad/0fd00c404fa73fe9b169429e9a972d5ed807973c40ab6b3cf9365a33d360/frozenlist-1.8.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:9ff15928d62a0b80bb875655c39bf517938c7d589554cbd2669be42d97c2cb61", size = 231775, upload-time = "2025-10-06T05:38:03.384Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/c3/86962566154cb4d2995358bc8331bfc4ea19d07db1a96f64935a1607f2b6/frozenlist-1.8.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:7bf6cdf8e07c8151fba6fe85735441240ec7f619f935a5205953d58009aef8c6", size = 236631, upload-time = "2025-10-06T05:38:04.609Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/9e/6ffad161dbd83782d2c66dc4d378a9103b31770cb1e67febf43aea42d202/frozenlist-1.8.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:48e6d3f4ec5c7273dfe83ff27c91083c6c9065af655dc2684d2c200c94308bb5", size = 218632, upload-time = "2025-10-06T05:38:05.917Z" },
+    { url = "https://files.pythonhosted.org/packages/58/b2/4677eee46e0a97f9b30735e6ad0bf6aba3e497986066eb68807ac85cf60f/frozenlist-1.8.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:1a7607e17ad33361677adcd1443edf6f5da0ce5e5377b798fba20fae194825f3", size = 235967, upload-time = "2025-10-06T05:38:07.614Z" },
+    { url = "https://files.pythonhosted.org/packages/05/f3/86e75f8639c5a93745ca7addbbc9de6af56aebb930d233512b17e46f6493/frozenlist-1.8.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:5a3a935c3a4e89c733303a2d5a7c257ea44af3a56c8202df486b7f5de40f37e1", size = 228799, upload-time = "2025-10-06T05:38:08.845Z" },
+    { url = "https://files.pythonhosted.org/packages/30/00/39aad3a7f0d98f5eb1d99a3c311215674ed87061aecee7851974b335c050/frozenlist-1.8.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:940d4a017dbfed9daf46a3b086e1d2167e7012ee297fef9e1c545c4d022f5178", size = 230566, upload-time = "2025-10-06T05:38:10.52Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/4d/aa144cac44568d137846ddc4d5210fb5d9719eb1d7ec6fa2728a54b5b94a/frozenlist-1.8.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:b9be22a69a014bc47e78072d0ecae716f5eb56c15238acca0f43d6eb8e4a5bda", size = 217715, upload-time = "2025-10-06T05:38:11.832Z" },
+    { url = "https://files.pythonhosted.org/packages/64/4c/8f665921667509d25a0dd72540513bc86b356c95541686f6442a3283019f/frozenlist-1.8.0-cp39-cp39-win32.whl", hash = "sha256:1aa77cb5697069af47472e39612976ed05343ff2e84a3dcf15437b232cbfd087", size = 39933, upload-time = "2025-10-06T05:38:13.061Z" },
+    { url = "https://files.pythonhosted.org/packages/79/bd/bcc926f87027fad5e59926ff12d136e1082a115025d33c032d1cd69ab377/frozenlist-1.8.0-cp39-cp39-win_amd64.whl", hash = "sha256:7398c222d1d405e796970320036b1b563892b65809d9e5261487bb2c7f7b5c6a", size = 44121, upload-time = "2025-10-06T05:38:14.572Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/07/9c2e4eb7584af4b705237b971b89a4155a8e57599c4483a131a39256a9a0/frozenlist-1.8.0-cp39-cp39-win_arm64.whl", hash = "sha256:b4f3b365f31c6cd4af24545ca0a244a53688cad8834e32f56831c4923b50a103", size = 40312, upload-time = "2025-10-06T05:38:15.699Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/9a/e35b4a917281c0b8419d4207f4334c8e8c5dbf4f3f5f9ada73958d937dcc/frozenlist-1.8.0-py3-none-any.whl", hash = "sha256:0c18a16eab41e82c295618a77502e17b195883241c563b00f0aa5106fc4eaa0d", size = 13409, upload-time = "2025-10-06T05:38:16.721Z" },
 ]
 
 [[package]]
 name = "fsspec"
-version = "2025.3.0"
+version = "2025.9.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/34/f4/5721faf47b8c499e776bc34c6a8fc17efdf7fdef0b00f398128bc5dcb4ac/fsspec-2025.3.0.tar.gz", hash = "sha256:a935fd1ea872591f2b5148907d103488fc523295e6c64b835cfad8c3eca44972", size = 298491, upload-time = "2025-03-07T21:47:56.461Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/de/e0/bab50af11c2d75c9c4a2a26a5254573c0bd97cea152254401510950486fa/fsspec-2025.9.0.tar.gz", hash = "sha256:19fd429483d25d28b65ec68f9f4adc16c17ea2c7c7bf54ec61360d478fb19c19", size = 304847, upload-time = "2025-09-02T19:10:49.215Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/56/53/eb690efa8513166adef3e0669afd31e95ffde69fb3c52ec2ac7223ed6018/fsspec-2025.3.0-py3-none-any.whl", hash = "sha256:efb87af3efa9103f94ca91a7f8cb7a4df91af9f74fc106c9c7ea0efd7277c1b3", size = 193615, upload-time = "2025-03-07T21:47:54.809Z" },
+    { url = "https://files.pythonhosted.org/packages/47/71/70db47e4f6ce3e5c37a607355f80da8860a33226be640226ac52cb05ef2e/fsspec-2025.9.0-py3-none-any.whl", hash = "sha256:530dc2a2af60a414a832059574df4a6e10cce927f6f4a78209390fe38955cfb7", size = 199289, upload-time = "2025-09-02T19:10:47.708Z" },
 ]
 
 [package.optional-dependencies]
@@ -893,6 +1205,21 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f8/5c/e226de133afd8bb267ec27eead9ae3d784b95b39a287ed404caab39a5f50/genson-1.3.0-py3-none-any.whl", hash = "sha256:468feccd00274cc7e4c09e84b08704270ba8d95232aa280f65b986139cec67f7", size = 21470, upload-time = "2024-05-15T22:08:47.056Z" },
 ]
 
+[[package]]
+name = "gguf"
+version = "0.17.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pyyaml" },
+    { name = "tqdm" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/08/08/7de1ca4b71e7bf33b547f82bb22505e221b5fa42f67d635e200e0ad22ad6/gguf-0.17.1.tar.gz", hash = "sha256:36ad71aad900a3e75fc94ebe96ea6029f03a4e44be7627ef7ad3d03e8c7bcb53", size = 89338, upload-time = "2025-06-19T14:00:33.705Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fc/31/6a93a887617ee7deeaa602ca3d02d1c12a6cb8a742a695de5d128f5fa46a/gguf-0.17.1-py3-none-any.whl", hash = "sha256:7bc5aa7eeb1931f7d39b48fdc5b38fda6b294b9dca75cf607ac69557840a3943", size = 96224, upload-time = "2025-06-19T14:00:32.88Z" },
+]
+
 [[package]]
 name = "ghp-import"
 version = "2.1.0"
@@ -907,21 +1234,21 @@ wheels = [
 
 [[package]]
 name = "google-auth"
-version = "2.40.3"
+version = "2.41.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "cachetools" },
     { name = "pyasn1-modules" },
     { name = "rsa" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/9e/9b/e92ef23b84fa10a64ce4831390b7a4c2e53c0132568d99d4ae61d04c8855/google_auth-2.40.3.tar.gz", hash = "sha256:500c3a29adedeb36ea9cf24b8d10858e152f2412e3ca37829b3fa18e33d63b77", size = 281029, upload-time = "2025-06-04T18:04:57.577Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/a8/af/5129ce5b2f9688d2fa49b463e544972a7c82b0fdb50980dafee92e121d9f/google_auth-2.41.1.tar.gz", hash = "sha256:b76b7b1f9e61f0cb7e88870d14f6a94aeef248959ef6992670efee37709cbfd2", size = 292284, upload-time = "2025-09-30T22:51:26.363Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/17/63/b19553b658a1692443c62bd07e5868adaa0ad746a0751ba62c59568cd45b/google_auth-2.40.3-py2.py3-none-any.whl", hash = "sha256:1370d4593e86213563547f97a92752fc658456fe4514c809544f330fed45a7ca", size = 216137, upload-time = "2025-06-04T18:04:55.573Z" },
+    { url = "https://files.pythonhosted.org/packages/be/a4/7319a2a8add4cc352be9e3efeff5e2aacee917c85ca2fa1647e29089983c/google_auth-2.41.1-py2.py3-none-any.whl", hash = "sha256:754843be95575b9a19c604a848a41be03f7f2afd8c019f716dc1f51ee41c639d", size = 221302, upload-time = "2025-09-30T22:51:24.212Z" },
 ]
 
 [[package]]
 name = "google-genai"
-version = "1.32.0"
+version = "1.41.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
@@ -933,9 +1260,9 @@ dependencies = [
     { name = "typing-extensions" },
     { name = "websockets" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/03/ab/e6cdd8fa957c647ef00c4da7c59d0e734354bd49ed8d98c860732d8e1944/google_genai-1.32.0.tar.gz", hash = "sha256:349da3f5ff0e981066bd508585fcdd308d28fc4646f318c8f6d1aa6041f4c7e3", size = 240802, upload-time = "2025-08-27T22:16:32.781Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/72/8b/ee20bcf707769b3b0e1106c3b5c811507736af7e8a60f29a70af1750ba19/google_genai-1.41.0.tar.gz", hash = "sha256:134f861bb0ace4e34af0501ecb75ceee15f7662fd8120698cd185e8cb39f2800", size = 245812, upload-time = "2025-10-02T22:30:29.699Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/59/55/be09472f7a656af1208196d2ef9a3d2710f3cbcf695f51acbcbe28b9472b/google_genai-1.32.0-py3-none-any.whl", hash = "sha256:c0c4b1d45adf3aa99501050dd73da2f0dea09374002231052d81a6765d15e7f6", size = 241680, upload-time = "2025-08-27T22:16:31.409Z" },
+    { url = "https://files.pythonhosted.org/packages/15/14/e5e8fbca8863fee718208566c4e927b8e9f45fd46ec5cf89e24759da545b/google_genai-1.41.0-py3-none-any.whl", hash = "sha256:111a3ee64c1a0927d3879faddb368234594432479a40c311e5fe4db338ca8778", size = 245931, upload-time = "2025-10-02T22:30:27.885Z" },
 ]
 
 [[package]]
@@ -952,50 +1279,53 @@ wheels = [
 
 [[package]]
 name = "grpcio"
-version = "1.74.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/38/b4/35feb8f7cab7239c5b94bd2db71abb3d6adb5f335ad8f131abb6060840b6/grpcio-1.74.0.tar.gz", hash = "sha256:80d1f4fbb35b0742d3e3d3bb654b7381cd5f015f8497279a1e9c21ba623e01b1", size = 12756048, upload-time = "2025-07-24T18:54:23.039Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/66/54/68e51a90797ad7afc5b0a7881426c337f6a9168ebab73c3210b76aa7c90d/grpcio-1.74.0-cp310-cp310-linux_armv7l.whl", hash = "sha256:85bd5cdf4ed7b2d6438871adf6afff9af7096486fcf51818a81b77ef4dd30907", size = 5481935, upload-time = "2025-07-24T18:52:43.756Z" },
-    { url = "https://files.pythonhosted.org/packages/32/2a/af817c7e9843929e93e54d09c9aee2555c2e8d81b93102a9426b36e91833/grpcio-1.74.0-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:68c8ebcca945efff9d86d8d6d7bfb0841cf0071024417e2d7f45c5e46b5b08eb", size = 10986796, upload-time = "2025-07-24T18:52:47.219Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/94/d67756638d7bb07750b07d0826c68e414124574b53840ba1ff777abcd388/grpcio-1.74.0-cp310-cp310-manylinux_2_17_aarch64.whl", hash = "sha256:e154d230dc1bbbd78ad2fdc3039fa50ad7ffcf438e4eb2fa30bce223a70c7486", size = 5983663, upload-time = "2025-07-24T18:52:49.463Z" },
-    { url = "https://files.pythonhosted.org/packages/35/f5/c5e4853bf42148fea8532d49e919426585b73eafcf379a712934652a8de9/grpcio-1.74.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e8978003816c7b9eabe217f88c78bc26adc8f9304bf6a594b02e5a49b2ef9c11", size = 6653765, upload-time = "2025-07-24T18:52:51.094Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/75/a1991dd64b331d199935e096cc9daa3415ee5ccbe9f909aa48eded7bba34/grpcio-1.74.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c3d7bd6e3929fd2ea7fbc3f562e4987229ead70c9ae5f01501a46701e08f1ad9", size = 6215172, upload-time = "2025-07-24T18:52:53.282Z" },
-    { url = "https://files.pythonhosted.org/packages/01/a4/7cef3dbb3b073d0ce34fd507efc44ac4c9442a0ef9fba4fb3f5c551efef5/grpcio-1.74.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:136b53c91ac1d02c8c24201bfdeb56f8b3ac3278668cbb8e0ba49c88069e1bdc", size = 6329142, upload-time = "2025-07-24T18:52:54.927Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/d3/587920f882b46e835ad96014087054655312400e2f1f1446419e5179a383/grpcio-1.74.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:fe0f540750a13fd8e5da4b3eaba91a785eea8dca5ccd2bc2ffe978caa403090e", size = 7018632, upload-time = "2025-07-24T18:52:56.523Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/95/c70a3b15a0bc83334b507e3d2ae20ee8fa38d419b8758a4d838f5c2a7d32/grpcio-1.74.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:4e4181bfc24413d1e3a37a0b7889bea68d973d4b45dd2bc68bb766c140718f82", size = 6509641, upload-time = "2025-07-24T18:52:58.495Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/06/2e7042d06247d668ae69ea6998eca33f475fd4e2855f94dcb2aa5daef334/grpcio-1.74.0-cp310-cp310-win32.whl", hash = "sha256:1733969040989f7acc3d94c22f55b4a9501a30f6aaacdbccfaba0a3ffb255ab7", size = 3817478, upload-time = "2025-07-24T18:53:00.128Z" },
-    { url = "https://files.pythonhosted.org/packages/93/20/e02b9dcca3ee91124060b65bbf5b8e1af80b3b76a30f694b44b964ab4d71/grpcio-1.74.0-cp310-cp310-win_amd64.whl", hash = "sha256:9e912d3c993a29df6c627459af58975b2e5c897d93287939b9d5065f000249b5", size = 4493971, upload-time = "2025-07-24T18:53:02.068Z" },
-    { url = "https://files.pythonhosted.org/packages/e7/77/b2f06db9f240a5abeddd23a0e49eae2b6ac54d85f0e5267784ce02269c3b/grpcio-1.74.0-cp311-cp311-linux_armv7l.whl", hash = "sha256:69e1a8180868a2576f02356565f16635b99088da7df3d45aaa7e24e73a054e31", size = 5487368, upload-time = "2025-07-24T18:53:03.548Z" },
-    { url = "https://files.pythonhosted.org/packages/48/99/0ac8678a819c28d9a370a663007581744a9f2a844e32f0fa95e1ddda5b9e/grpcio-1.74.0-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:8efe72fde5500f47aca1ef59495cb59c885afe04ac89dd11d810f2de87d935d4", size = 10999804, upload-time = "2025-07-24T18:53:05.095Z" },
-    { url = "https://files.pythonhosted.org/packages/45/c6/a2d586300d9e14ad72e8dc211c7aecb45fe9846a51e558c5bca0c9102c7f/grpcio-1.74.0-cp311-cp311-manylinux_2_17_aarch64.whl", hash = "sha256:a8f0302f9ac4e9923f98d8e243939a6fb627cd048f5cd38595c97e38020dffce", size = 5987667, upload-time = "2025-07-24T18:53:07.157Z" },
-    { url = "https://files.pythonhosted.org/packages/c9/57/5f338bf56a7f22584e68d669632e521f0de460bb3749d54533fc3d0fca4f/grpcio-1.74.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2f609a39f62a6f6f05c7512746798282546358a37ea93c1fcbadf8b2fed162e3", size = 6655612, upload-time = "2025-07-24T18:53:09.244Z" },
-    { url = "https://files.pythonhosted.org/packages/82/ea/a4820c4c44c8b35b1903a6c72a5bdccec92d0840cf5c858c498c66786ba5/grpcio-1.74.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c98e0b7434a7fa4e3e63f250456eaef52499fba5ae661c58cc5b5477d11e7182", size = 6219544, upload-time = "2025-07-24T18:53:11.221Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/17/0537630a921365928f5abb6d14c79ba4dcb3e662e0dbeede8af4138d9dcf/grpcio-1.74.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:662456c4513e298db6d7bd9c3b8df6f75f8752f0ba01fb653e252ed4a59b5a5d", size = 6334863, upload-time = "2025-07-24T18:53:12.925Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/a6/85ca6cb9af3f13e1320d0a806658dca432ff88149d5972df1f7b51e87127/grpcio-1.74.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:3d14e3c4d65e19d8430a4e28ceb71ace4728776fd6c3ce34016947474479683f", size = 7019320, upload-time = "2025-07-24T18:53:15.002Z" },
-    { url = "https://files.pythonhosted.org/packages/4f/a7/fe2beab970a1e25d2eff108b3cf4f7d9a53c185106377a3d1989216eba45/grpcio-1.74.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:1bf949792cee20d2078323a9b02bacbbae002b9e3b9e2433f2741c15bdeba1c4", size = 6514228, upload-time = "2025-07-24T18:53:16.999Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/c2/2f9c945c8a248cebc3ccda1b7a1bf1775b9d7d59e444dbb18c0014e23da6/grpcio-1.74.0-cp311-cp311-win32.whl", hash = "sha256:55b453812fa7c7ce2f5c88be3018fb4a490519b6ce80788d5913f3f9d7da8c7b", size = 3817216, upload-time = "2025-07-24T18:53:20.564Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/d1/a9cf9c94b55becda2199299a12b9feef0c79946b0d9d34c989de6d12d05d/grpcio-1.74.0-cp311-cp311-win_amd64.whl", hash = "sha256:86ad489db097141a907c559988c29718719aa3e13370d40e20506f11b4de0d11", size = 4495380, upload-time = "2025-07-24T18:53:22.058Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/5d/e504d5d5c4469823504f65687d6c8fb97b7f7bf0b34873b7598f1df24630/grpcio-1.74.0-cp312-cp312-linux_armv7l.whl", hash = "sha256:8533e6e9c5bd630ca98062e3a1326249e6ada07d05acf191a77bc33f8948f3d8", size = 5445551, upload-time = "2025-07-24T18:53:23.641Z" },
-    { url = "https://files.pythonhosted.org/packages/43/01/730e37056f96f2f6ce9f17999af1556df62ee8dab7fa48bceeaab5fd3008/grpcio-1.74.0-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:2918948864fec2a11721d91568effffbe0a02b23ecd57f281391d986847982f6", size = 10979810, upload-time = "2025-07-24T18:53:25.349Z" },
-    { url = "https://files.pythonhosted.org/packages/79/3d/09fd100473ea5c47083889ca47ffd356576173ec134312f6aa0e13111dee/grpcio-1.74.0-cp312-cp312-manylinux_2_17_aarch64.whl", hash = "sha256:60d2d48b0580e70d2e1954d0d19fa3c2e60dd7cbed826aca104fff518310d1c5", size = 5941946, upload-time = "2025-07-24T18:53:27.387Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/99/12d2cca0a63c874c6d3d195629dcd85cdf5d6f98a30d8db44271f8a97b93/grpcio-1.74.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3601274bc0523f6dc07666c0e01682c94472402ac2fd1226fd96e079863bfa49", size = 6621763, upload-time = "2025-07-24T18:53:29.193Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/2c/930b0e7a2f1029bbc193443c7bc4dc2a46fedb0203c8793dcd97081f1520/grpcio-1.74.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:176d60a5168d7948539def20b2a3adcce67d72454d9ae05969a2e73f3a0feee7", size = 6180664, upload-time = "2025-07-24T18:53:30.823Z" },
-    { url = "https://files.pythonhosted.org/packages/db/d5/ff8a2442180ad0867717e670f5ec42bfd8d38b92158ad6bcd864e6d4b1ed/grpcio-1.74.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:e759f9e8bc908aaae0412642afe5416c9f983a80499448fcc7fab8692ae044c3", size = 6301083, upload-time = "2025-07-24T18:53:32.454Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/ba/b361d390451a37ca118e4ec7dccec690422e05bc85fba2ec72b06cefec9f/grpcio-1.74.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:9e7c4389771855a92934b2846bd807fc25a3dfa820fd912fe6bd8136026b2707", size = 6994132, upload-time = "2025-07-24T18:53:34.506Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/0c/3a5fa47d2437a44ced74141795ac0251bbddeae74bf81df3447edd767d27/grpcio-1.74.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:cce634b10aeab37010449124814b05a62fb5f18928ca878f1bf4750d1f0c815b", size = 6489616, upload-time = "2025-07-24T18:53:36.217Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/95/ab64703b436d99dc5217228babc76047d60e9ad14df129e307b5fec81fd0/grpcio-1.74.0-cp312-cp312-win32.whl", hash = "sha256:885912559974df35d92219e2dc98f51a16a48395f37b92865ad45186f294096c", size = 3807083, upload-time = "2025-07-24T18:53:37.911Z" },
-    { url = "https://files.pythonhosted.org/packages/84/59/900aa2445891fc47a33f7d2f76e00ca5d6ae6584b20d19af9c06fa09bf9a/grpcio-1.74.0-cp312-cp312-win_amd64.whl", hash = "sha256:42f8fee287427b94be63d916c90399ed310ed10aadbf9e2e5538b3e497d269bc", size = 4490123, upload-time = "2025-07-24T18:53:39.528Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/de/dd7db504703c3b669f1b83265e2fbb5d79c8d3da86ea52cbd9202b9a8b05/grpcio-1.74.0-cp39-cp39-linux_armv7l.whl", hash = "sha256:4bc5fca10aaf74779081e16c2bcc3d5ec643ffd528d9e7b1c9039000ead73bae", size = 5480998, upload-time = "2025-07-24T18:54:01.868Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/57/6537ace3af4c97f2b013ceff1f2e789c52b8448334ca3a0c36e7421cf6ed/grpcio-1.74.0-cp39-cp39-macosx_11_0_universal2.whl", hash = "sha256:6bab67d15ad617aff094c382c882e0177637da73cbc5532d52c07b4ee887a87b", size = 10990945, upload-time = "2025-07-24T18:54:03.854Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/f2/579410017de16cd27f35326df6fecde81bff6e9b43c871d28263fa8a77a4/grpcio-1.74.0-cp39-cp39-manylinux_2_17_aarch64.whl", hash = "sha256:655726919b75ab3c34cdad39da5c530ac6fa32696fb23119e36b64adcfca174a", size = 5983968, upload-time = "2025-07-24T18:54:06.434Z" },
-    { url = "https://files.pythonhosted.org/packages/a0/6a/0f3571003663d991f4ea953b82dc518fed094c182decc48c9b0242bec7e3/grpcio-1.74.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1a2b06afe2e50ebfd46247ac3ba60cac523f54ec7792ae9ba6073c12daf26f0a", size = 6654768, upload-time = "2025-07-24T18:54:08.632Z" },
-    { url = "https://files.pythonhosted.org/packages/24/e3/1d42cb00e0390bacab3c9ee79e37416140d907c8c7c7a92654c535805963/grpcio-1.74.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5f251c355167b2360537cf17bea2cf0197995e551ab9da6a0a59b3da5e8704f9", size = 6215627, upload-time = "2025-07-24T18:54:10.584Z" },
-    { url = "https://files.pythonhosted.org/packages/77/84/4f8312bc4430eda1cdbc4e8689f54daa807b5d304d4ea53e9d27c448889b/grpcio-1.74.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:8f7b5882fb50632ab1e48cb3122d6df55b9afabc265582808036b6e51b9fd6b7", size = 6330938, upload-time = "2025-07-24T18:54:12.557Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/c0/422d2b40110716a4775212256a56ac71586be2403a7b7055818bfd0fc203/grpcio-1.74.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:834988b6c34515545b3edd13e902c1acdd9f2465d386ea5143fb558f153a7176", size = 7019216, upload-time = "2025-07-24T18:54:14.475Z" },
-    { url = "https://files.pythonhosted.org/packages/6f/84/668ab6df27fb35886dfa1242f2d302d0cd319c72e3dd3845a322ecabf61b/grpcio-1.74.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:22b834cef33429ca6cc28303c9c327ba9a3fafecbf62fae17e9a7b7163cc43ac", size = 6510719, upload-time = "2025-07-24T18:54:16.775Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/6a/981150f20dd435b9c46cd504038b4fbae2171b43fe70019914d80159e156/grpcio-1.74.0-cp39-cp39-win32.whl", hash = "sha256:7d95d71ff35291bab3f1c52f52f474c632db26ea12700c2ff0ea0532cb0b5854", size = 3819185, upload-time = "2025-07-24T18:54:18.673Z" },
-    { url = "https://files.pythonhosted.org/packages/75/5f/d64b9745bb9def186e1be11b42d4d310570799d6170ac75829ef1c67c176/grpcio-1.74.0-cp39-cp39-win_amd64.whl", hash = "sha256:ecde9ab49f58433abe02f9ed076c7b5be839cf0153883a6d23995937a82392fa", size = 4495789, upload-time = "2025-07-24T18:54:20.582Z" },
+version = "1.75.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9d/f7/8963848164c7604efb3a3e6ee457fdb3a469653e19002bd24742473254f8/grpcio-1.75.1.tar.gz", hash = "sha256:3e81d89ece99b9ace23a6916880baca613c03a799925afb2857887efa8b1b3d2", size = 12731327, upload-time = "2025-09-26T09:03:36.887Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/51/57/89fd829fb00a6d0bee3fbcb2c8a7aa0252d908949b6ab58bfae99d39d77e/grpcio-1.75.1-cp310-cp310-linux_armv7l.whl", hash = "sha256:1712b5890b22547dd29f3215c5788d8fc759ce6dd0b85a6ba6e2731f2d04c088", size = 5705534, upload-time = "2025-09-26T09:00:52.225Z" },
+    { url = "https://files.pythonhosted.org/packages/76/dd/2f8536e092551cf804e96bcda79ecfbc51560b214a0f5b7ebc253f0d4664/grpcio-1.75.1-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:8d04e101bba4b55cea9954e4aa71c24153ba6182481b487ff376da28d4ba46cf", size = 11484103, upload-time = "2025-09-26T09:00:59.457Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/3d/affe2fb897804c98d56361138e73786af8f4dd876b9d9851cfe6342b53c8/grpcio-1.75.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:683cfc70be0c1383449097cba637317e4737a357cfc185d887fd984206380403", size = 6289953, upload-time = "2025-09-26T09:01:03.699Z" },
+    { url = "https://files.pythonhosted.org/packages/87/aa/0f40b7f47a0ff10d7e482bc3af22dac767c7ff27205915f08962d5ca87a2/grpcio-1.75.1-cp310-cp310-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:491444c081a54dcd5e6ada57314321ae526377f498d4aa09d975c3241c5b9e1c", size = 6949785, upload-time = "2025-09-26T09:01:07.504Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/45/b04407e44050781821c84f26df71b3f7bc469923f92f9f8bc27f1406dbcc/grpcio-1.75.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ce08d4e112d0d38487c2b631ec8723deac9bc404e9c7b1011426af50a79999e4", size = 6465708, upload-time = "2025-09-26T09:01:11.028Z" },
+    { url = "https://files.pythonhosted.org/packages/09/3e/4ae3ec0a4d20dcaafbb6e597defcde06399ccdc5b342f607323f3b47f0a3/grpcio-1.75.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:5a2acda37fc926ccc4547977ac3e56b1df48fe200de968e8c8421f6e3093df6c", size = 7100912, upload-time = "2025-09-26T09:01:14.393Z" },
+    { url = "https://files.pythonhosted.org/packages/34/3f/a9085dab5c313bb0cb853f222d095e2477b9b8490a03634cdd8d19daa5c3/grpcio-1.75.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:745c5fe6bf05df6a04bf2d11552c7d867a2690759e7ab6b05c318a772739bd75", size = 8042497, upload-time = "2025-09-26T09:01:17.759Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/87/ea54eba931ab9ed3f999ba95f5d8d01a20221b664725bab2fe93e3dee848/grpcio-1.75.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:259526a7159d39e2db40d566fe3e8f8e034d0fb2db5bf9c00e09aace655a4c2b", size = 7493284, upload-time = "2025-09-26T09:01:20.896Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/5e/287f1bf1a998f4ac46ef45d518de3b5da08b4e86c7cb5e1108cee30b0282/grpcio-1.75.1-cp310-cp310-win32.whl", hash = "sha256:f4b29b9aabe33fed5df0a85e5f13b09ff25e2c05bd5946d25270a8bd5682dac9", size = 3950809, upload-time = "2025-09-26T09:01:23.695Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/a2/3cbfc06a4ec160dc77403b29ecb5cf76ae329eb63204fea6a7c715f1dfdb/grpcio-1.75.1-cp310-cp310-win_amd64.whl", hash = "sha256:cf2e760978dcce7ff7d465cbc7e276c3157eedc4c27aa6de7b594c7a295d3d61", size = 4644704, upload-time = "2025-09-26T09:01:25.763Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/3c/35ca9747473a306bfad0cee04504953f7098527cd112a4ab55c55af9e7bd/grpcio-1.75.1-cp311-cp311-linux_armv7l.whl", hash = "sha256:573855ca2e58e35032aff30bfbd1ee103fbcf4472e4b28d4010757700918e326", size = 5709761, upload-time = "2025-09-26T09:01:28.528Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/2c/ecbcb4241e4edbe85ac2663f885726fea0e947767401288b50d8fdcb9200/grpcio-1.75.1-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:6a4996a2c8accc37976dc142d5991adf60733e223e5c9a2219e157dc6a8fd3a2", size = 11496691, upload-time = "2025-09-26T09:01:31.214Z" },
+    { url = "https://files.pythonhosted.org/packages/81/40/bc07aee2911f0d426fa53fe636216100c31a8ea65a400894f280274cb023/grpcio-1.75.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:b1ea1bbe77ecbc1be00af2769f4ae4a88ce93be57a4f3eebd91087898ed749f9", size = 6296084, upload-time = "2025-09-26T09:01:34.596Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/d1/10c067f6c67396cbf46448b80f27583b5e8c4b46cdfbe18a2a02c2c2f290/grpcio-1.75.1-cp311-cp311-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:e5b425aee54cc5e3e3c58f00731e8a33f5567965d478d516d35ef99fd648ab68", size = 6950403, upload-time = "2025-09-26T09:01:36.736Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/42/5f628abe360b84dfe8dd8f32be6b0606dc31dc04d3358eef27db791ea4d5/grpcio-1.75.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:0049a7bf547dafaeeb1db17079ce79596c298bfe308fc084d023c8907a845b9a", size = 6470166, upload-time = "2025-09-26T09:01:39.474Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/93/a24035080251324019882ee2265cfde642d6476c0cf8eb207fc693fcebdc/grpcio-1.75.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:5b8ea230c7f77c0a1a3208a04a1eda164633fb0767b4cefd65a01079b65e5b1f", size = 7107828, upload-time = "2025-09-26T09:01:41.782Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/f8/d18b984c1c9ba0318e3628dbbeb6af77a5007f02abc378c845070f2d3edd/grpcio-1.75.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:36990d629c3c9fb41e546414e5af52d0a7af37ce7113d9682c46d7e2919e4cca", size = 8045421, upload-time = "2025-09-26T09:01:45.835Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/b6/4bf9aacff45deca5eac5562547ed212556b831064da77971a4e632917da3/grpcio-1.75.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:b10ad908118d38c2453ade7ff790e5bce36580c3742919007a2a78e3a1e521ca", size = 7503290, upload-time = "2025-09-26T09:01:49.28Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/15/d8d69d10223cb54c887a2180bd29fe5fa2aec1d4995c8821f7aa6eaf72e4/grpcio-1.75.1-cp311-cp311-win32.whl", hash = "sha256:d6be2b5ee7bea656c954dcf6aa8093c6f0e6a3ef9945c99d99fcbfc88c5c0bfe", size = 3950631, upload-time = "2025-09-26T09:01:51.23Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/40/7b8642d45fff6f83300c24eaac0380a840e5e7fe0e8d80afd31b99d7134e/grpcio-1.75.1-cp311-cp311-win_amd64.whl", hash = "sha256:61c692fb05956b17dd6d1ab480f7f10ad0536dba3bc8fd4e3c7263dc244ed772", size = 4646131, upload-time = "2025-09-26T09:01:53.266Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/81/42be79e73a50aaa20af66731c2defeb0e8c9008d9935a64dd8ea8e8c44eb/grpcio-1.75.1-cp312-cp312-linux_armv7l.whl", hash = "sha256:7b888b33cd14085d86176b1628ad2fcbff94cfbbe7809465097aa0132e58b018", size = 5668314, upload-time = "2025-09-26T09:01:55.424Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/a7/3686ed15822fedc58c22f82b3a7403d9faf38d7c33de46d4de6f06e49426/grpcio-1.75.1-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:8775036efe4ad2085975531d221535329f5dac99b6c2a854a995456098f99546", size = 11476125, upload-time = "2025-09-26T09:01:57.927Z" },
+    { url = "https://files.pythonhosted.org/packages/14/85/21c71d674f03345ab183c634ecd889d3330177e27baea8d5d247a89b6442/grpcio-1.75.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:bb658f703468d7fbb5dcc4037c65391b7dc34f808ac46ed9136c24fc5eeb041d", size = 6246335, upload-time = "2025-09-26T09:02:00.76Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/db/3beb661bc56a385ae4fa6b0e70f6b91ac99d47afb726fe76aaff87ebb116/grpcio-1.75.1-cp312-cp312-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:4b7177a1cdb3c51b02b0c0a256b0a72fdab719600a693e0e9037949efffb200b", size = 6916309, upload-time = "2025-09-26T09:02:02.894Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/9c/eda9fe57f2b84343d44c1b66cf3831c973ba29b078b16a27d4587a1fdd47/grpcio-1.75.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:7d4fa6ccc3ec2e68a04f7b883d354d7fea22a34c44ce535a2f0c0049cf626ddf", size = 6435419, upload-time = "2025-09-26T09:02:05.055Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/b8/090c98983e0a9d602e3f919a6e2d4e470a8b489452905f9a0fa472cac059/grpcio-1.75.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3d86880ecaeb5b2f0a8afa63824de93adb8ebe4e49d0e51442532f4e08add7d6", size = 7064893, upload-time = "2025-09-26T09:02:07.275Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/c0/6d53d4dbbd00f8bd81571f5478d8a95528b716e0eddb4217cc7cb45aae5f/grpcio-1.75.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:a8041d2f9e8a742aeae96f4b047ee44e73619f4f9d24565e84d5446c623673b6", size = 8011922, upload-time = "2025-09-26T09:02:09.527Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/7c/48455b2d0c5949678d6982c3e31ea4d89df4e16131b03f7d5c590811cbe9/grpcio-1.75.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:3652516048bf4c314ce12be37423c79829f46efffb390ad64149a10c6071e8de", size = 7466181, upload-time = "2025-09-26T09:02:12.279Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/12/04a0e79081e3170b6124f8cba9b6275871276be06c156ef981033f691880/grpcio-1.75.1-cp312-cp312-win32.whl", hash = "sha256:44b62345d8403975513af88da2f3d5cc76f73ca538ba46596f92a127c2aea945", size = 3938543, upload-time = "2025-09-26T09:02:14.77Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/d7/11350d9d7fb5adc73d2b0ebf6ac1cc70135577701e607407fe6739a90021/grpcio-1.75.1-cp312-cp312-win_amd64.whl", hash = "sha256:b1e191c5c465fa777d4cafbaacf0c01e0d5278022082c0abbd2ee1d6454ed94d", size = 4641938, upload-time = "2025-09-26T09:02:16.927Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/e2/33efd823a879dc7b60c10192df1900ee5c200f8e782663a41a3b2aecd143/grpcio-1.75.1-cp39-cp39-linux_armv7l.whl", hash = "sha256:c09fba33327c3ac11b5c33dbdd8218eef8990d78f83b1656d628831812a8c0fb", size = 5706679, upload-time = "2025-09-26T09:03:10.218Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/13/17e39ee4897f1cd12dd463e863b830e64643b13e9a4af5062b4a6f0790be/grpcio-1.75.1-cp39-cp39-macosx_11_0_universal2.whl", hash = "sha256:7e21400b037be29545704889e72e586c238e346dcb2d08d8a7288d16c883a9ec", size = 11490271, upload-time = "2025-09-26T09:03:12.778Z" },
+    { url = "https://files.pythonhosted.org/packages/77/90/b80e75f8cce758425b2772742eed4e9db765a965d902ba4b7f239b2513de/grpcio-1.75.1-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:c12121e509b9f8b0914d10054d24120237d19e870b1cd82acbb8a9b9ddd198a3", size = 6291926, upload-time = "2025-09-26T09:03:16.282Z" },
+    { url = "https://files.pythonhosted.org/packages/40/5f/e6033d8f99063350e20873a46225468b73045b9ef2c8cba73d66a87c3fd5/grpcio-1.75.1-cp39-cp39-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:73577a93e692b3474b1bfe84285d098de36705dbd838bb4d6a056d326e4dc880", size = 6950040, upload-time = "2025-09-26T09:03:18.874Z" },
+    { url = "https://files.pythonhosted.org/packages/01/12/34076c079b45af5aed40f037fffe388d7fbe90dd539ed01e4744c926d227/grpcio-1.75.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e19e7dfa0d7ca7dea22be464339e18ac608fd75d88c56770c646cdabe54bc724", size = 6465780, upload-time = "2025-09-26T09:03:21.219Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/c5/ee6fd69a9f6e7288d04da010ad7480a0566d2aac81097ff4dafbc5ffa9b6/grpcio-1.75.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:4e1c28f51c1cf67eccdfc1065e8e866c9ed622f09773ca60947089c117f848a1", size = 7098308, upload-time = "2025-09-26T09:03:23.875Z" },
+    { url = "https://files.pythonhosted.org/packages/78/32/f2be13f13035361768923159fe20470a7d22db2c7c692b952e21284f56e5/grpcio-1.75.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:030a6164bc2ca726052778c0cf8e3249617a34e368354f9e6107c27ad4af8c28", size = 8042268, upload-time = "2025-09-26T09:03:26.268Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/2d/1bb0572f0a2eaab100b4635c6c2cd0d37e3cda5554037e3f90b1bc428d56/grpcio-1.75.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:67697efef5a98d46d5db7b1720fa4043536f8b8e5072a5d61cfca762f287e939", size = 7491470, upload-time = "2025-09-26T09:03:28.906Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/e0/1e962dcb64019bbd87eedcfacdedb83af0f66da01f2f6e03d69b0aa1b7f0/grpcio-1.75.1-cp39-cp39-win32.whl", hash = "sha256:52015cf73eb5d76f6404e0ce0505a69b51fd1f35810b3a01233b34b10baafb41", size = 3951697, upload-time = "2025-09-26T09:03:31.535Z" },
+    { url = "https://files.pythonhosted.org/packages/87/bc/47fb3aaa77e7d657999937ec1026beba9e37f3199599fe510f762d31da97/grpcio-1.75.1-cp39-cp39-win_amd64.whl", hash = "sha256:9fe51e4a1f896ea84ac750900eae34d9e9b896b5b1e4a30b02dc31ad29f36383", size = 4645764, upload-time = "2025-09-26T09:03:34.071Z" },
 ]
 
 [[package]]
@@ -1041,17 +1371,17 @@ wheels = [
 
 [[package]]
 name = "hf-xet"
-version = "1.1.9"
+version = "1.1.10"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/23/0f/5b60fc28ee7f8cc17a5114a584fd6b86e11c3e0a6e142a7f97a161e9640a/hf_xet-1.1.9.tar.gz", hash = "sha256:c99073ce404462e909f1d5839b2d14a3827b8fe75ed8aed551ba6609c026c803", size = 484242, upload-time = "2025-08-27T23:05:19.441Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/74/31/feeddfce1748c4a233ec1aa5b7396161c07ae1aa9b7bdbc9a72c3c7dd768/hf_xet-1.1.10.tar.gz", hash = "sha256:408aef343800a2102374a883f283ff29068055c111f003ff840733d3b715bb97", size = 487910, upload-time = "2025-09-12T20:10:27.12Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/de/12/56e1abb9a44cdef59a411fe8a8673313195711b5ecce27880eb9c8fa90bd/hf_xet-1.1.9-cp37-abi3-macosx_10_12_x86_64.whl", hash = "sha256:a3b6215f88638dd7a6ff82cb4e738dcbf3d863bf667997c093a3c990337d1160", size = 2762553, upload-time = "2025-08-27T23:05:15.153Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/e6/2d0d16890c5f21b862f5df3146519c182e7f0ae49b4b4bf2bd8a40d0b05e/hf_xet-1.1.9-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:9b486de7a64a66f9a172f4b3e0dfe79c9f0a93257c501296a2521a13495a698a", size = 2623216, upload-time = "2025-08-27T23:05:13.778Z" },
-    { url = "https://files.pythonhosted.org/packages/81/42/7e6955cf0621e87491a1fb8cad755d5c2517803cea174229b0ec00ff0166/hf_xet-1.1.9-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a4c5a840c2c4e6ec875ed13703a60e3523bc7f48031dfd750923b2a4d1a5fc3c", size = 3186789, upload-time = "2025-08-27T23:05:12.368Z" },
-    { url = "https://files.pythonhosted.org/packages/df/8b/759233bce05457f5f7ec062d63bbfd2d0c740b816279eaaa54be92aa452a/hf_xet-1.1.9-cp37-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:96a6139c9e44dad1c52c52520db0fffe948f6bce487cfb9d69c125f254bb3790", size = 3088747, upload-time = "2025-08-27T23:05:10.439Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/3c/28cc4db153a7601a996985bcb564f7b8f5b9e1a706c7537aad4b4809f358/hf_xet-1.1.9-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:ad1022e9a998e784c97b2173965d07fe33ee26e4594770b7785a8cc8f922cd95", size = 3251429, upload-time = "2025-08-27T23:05:16.471Z" },
-    { url = "https://files.pythonhosted.org/packages/84/17/7caf27a1d101bfcb05be85850d4aa0a265b2e1acc2d4d52a48026ef1d299/hf_xet-1.1.9-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:86754c2d6d5afb11b0a435e6e18911a4199262fe77553f8c50d75e21242193ea", size = 3354643, upload-time = "2025-08-27T23:05:17.828Z" },
-    { url = "https://files.pythonhosted.org/packages/cd/50/0c39c9eed3411deadcc98749a6699d871b822473f55fe472fad7c01ec588/hf_xet-1.1.9-cp37-abi3-win_amd64.whl", hash = "sha256:5aad3933de6b725d61d51034e04174ed1dce7a57c63d530df0014dea15a40127", size = 2804797, upload-time = "2025-08-27T23:05:20.77Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/a2/343e6d05de96908366bdc0081f2d8607d61200be2ac802769c4284cc65bd/hf_xet-1.1.10-cp37-abi3-macosx_10_12_x86_64.whl", hash = "sha256:686083aca1a6669bc85c21c0563551cbcdaa5cf7876a91f3d074a030b577231d", size = 2761466, upload-time = "2025-09-12T20:10:22.836Z" },
+    { url = "https://files.pythonhosted.org/packages/31/f9/6215f948ac8f17566ee27af6430ea72045e0418ce757260248b483f4183b/hf_xet-1.1.10-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:71081925383b66b24eedff3013f8e6bbd41215c3338be4b94ba75fd75b21513b", size = 2623807, upload-time = "2025-09-12T20:10:21.118Z" },
+    { url = "https://files.pythonhosted.org/packages/15/07/86397573efefff941e100367bbda0b21496ffcdb34db7ab51912994c32a2/hf_xet-1.1.10-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b6bceb6361c80c1cc42b5a7b4e3efd90e64630bcf11224dcac50ef30a47e435", size = 3186960, upload-time = "2025-09-12T20:10:19.336Z" },
+    { url = "https://files.pythonhosted.org/packages/01/a7/0b2e242b918cc30e1f91980f3c4b026ff2eedaf1e2ad96933bca164b2869/hf_xet-1.1.10-cp37-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:eae7c1fc8a664e54753ffc235e11427ca61f4b0477d757cc4eb9ae374b69f09c", size = 3087167, upload-time = "2025-09-12T20:10:17.255Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/25/3e32ab61cc7145b11eee9d745988e2f0f4fafda81b25980eebf97d8cff15/hf_xet-1.1.10-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:0a0005fd08f002180f7a12d4e13b22be277725bc23ed0529f8add5c7a6309c06", size = 3248612, upload-time = "2025-09-12T20:10:24.093Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/3d/ab7109e607ed321afaa690f557a9ada6d6d164ec852fd6bf9979665dc3d6/hf_xet-1.1.10-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:f900481cf6e362a6c549c61ff77468bd59d6dd082f3170a36acfef2eb6a6793f", size = 3353360, upload-time = "2025-09-12T20:10:25.563Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/0e/471f0a21db36e71a2f1752767ad77e92d8cde24e974e03d662931b1305ec/hf_xet-1.1.10-cp37-abi3-win_amd64.whl", hash = "sha256:5f54b19cc347c13235ae7ee98b330c26dd65ef1df47e5316ffb1e87713ca7045", size = 2804691, upload-time = "2025-09-12T20:10:28.433Z" },
 ]
 
 [[package]]
@@ -1067,6 +1397,30 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/7e/f5/f66802a942d491edb555dd61e3a9961140fd64c90bce1eafd741609d334d/httpcore-1.0.9-py3-none-any.whl", hash = "sha256:2d400746a40668fc9dec9810239072b40b4484b640a8c38fd654a024c7a1bf55", size = 78784, upload-time = "2025-04-24T22:06:20.566Z" },
 ]
 
+[[package]]
+name = "httptools"
+version = "0.6.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/a7/9a/ce5e1f7e131522e6d3426e8e7a490b3a01f39a6696602e1c4f33f9e94277/httptools-0.6.4.tar.gz", hash = "sha256:4e93eee4add6493b59a5c514da98c939b244fce4a0d8879cd3f466562f4b7d5c", size = 240639, upload-time = "2024-10-16T19:45:08.902Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/92/5e/b4a826fe91971a0b68e8c2bd4e7db3e7519882f5a8ccdb1194be2b3ab98f/httptools-0.6.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:deee0e3343f98ee8047e9f4c5bc7cedbf69f5734454a94c38ee829fb2d5fa3c1", size = 443130, upload-time = "2024-10-16T19:44:09.45Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/51/ce61e531e40289a681a463e1258fa1e05e0be54540e40d91d065a264cd8f/httptools-0.6.4-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ca80b7485c76f768a3bc83ea58373f8db7b015551117375e4918e2aa77ea9b50", size = 442148, upload-time = "2024-10-16T19:44:11.539Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/9e/270b7d767849b0c96f275c695d27ca76c30671f8eb8cc1bab6ced5c5e1d0/httptools-0.6.4-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:90d96a385fa941283ebd231464045187a31ad932ebfa541be8edf5b3c2328959", size = 415949, upload-time = "2024-10-16T19:44:13.388Z" },
+    { url = "https://files.pythonhosted.org/packages/81/86/ced96e3179c48c6f656354e106934e65c8963d48b69be78f355797f0e1b3/httptools-0.6.4-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:59e724f8b332319e2875efd360e61ac07f33b492889284a3e05e6d13746876f4", size = 417591, upload-time = "2024-10-16T19:44:15.258Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/24/0fe235d7b69c42423c7698d086d4db96475f9b50b6ad26a718ef27a0bce6/httptools-0.6.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f8787367fbdfccae38e35abf7641dafc5310310a5987b689f4c32cc8cc3ee975", size = 462891, upload-time = "2024-10-16T19:44:21.067Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/2f/205d1f2a190b72da6ffb5f41a3736c26d6fa7871101212b15e9b5cd8f61d/httptools-0.6.4-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:40b0f7fe4fd38e6a507bdb751db0379df1e99120c65fbdc8ee6c1d044897a636", size = 459788, upload-time = "2024-10-16T19:44:22.958Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/4c/d09ce0eff09057a206a74575ae8f1e1e2f0364d20e2442224f9e6612c8b9/httptools-0.6.4-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:40a5ec98d3f49904b9fe36827dcf1aadfef3b89e2bd05b0e35e94f97c2b14721", size = 433214, upload-time = "2024-10-16T19:44:24.513Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/d2/84c9e23edbccc4a4c6f96a1b8d99dfd2350289e94f00e9ccc7aadde26fb5/httptools-0.6.4-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:dacdd3d10ea1b4ca9df97a0a303cbacafc04b5cd375fa98732678151643d4988", size = 434120, upload-time = "2024-10-16T19:44:26.295Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/01/6fb20be3196ffdc8eeec4e653bc2a275eca7f36634c86302242c4fbb2760/httptools-0.6.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:69422b7f458c5af875922cdb5bd586cc1f1033295aa9ff63ee196a87519ac8e1", size = 508796, upload-time = "2024-10-16T19:44:32.825Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/d8/b644c44acc1368938317d76ac991c9bba1166311880bcc0ac297cb9d6bd7/httptools-0.6.4-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:16e603a3bff50db08cd578d54f07032ca1631450ceb972c2f834c2b860c28ea2", size = 510837, upload-time = "2024-10-16T19:44:33.974Z" },
+    { url = "https://files.pythonhosted.org/packages/52/d8/254d16a31d543073a0e57f1c329ca7378d8924e7e292eda72d0064987486/httptools-0.6.4-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:ec4f178901fa1834d4a060320d2f3abc5c9e39766953d038f1458cb885f47e81", size = 485289, upload-time = "2024-10-16T19:44:35.111Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/3c/4aee161b4b7a971660b8be71a92c24d6c64372c1ab3ae7f366b3680df20f/httptools-0.6.4-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:f9eb89ecf8b290f2e293325c646a211ff1c2493222798bb80a530c5e7502494f", size = 489779, upload-time = "2024-10-16T19:44:36.253Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/24/c70c34119d209bf08199d938dc9c69164f585ed3029237b4bdb90f673cb9/httptools-0.6.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d1ffd262a73d7c28424252381a5b854c19d9de5f56f075445d33919a637e3547", size = 449615, upload-time = "2024-10-16T19:45:01.351Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/62/e7f317fed3703bd81053840cacba4e40bcf424b870e4197f94bd1cf9fe7a/httptools-0.6.4-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:703c346571fa50d2e9856a37d7cd9435a25e7fd15e236c397bf224afaa355fe9", size = 448819, upload-time = "2024-10-16T19:45:02.652Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/13/68337d3be6b023260139434c49d7aa466aaa98f9aee7ed29270ac7dde6a2/httptools-0.6.4-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:aafe0f1918ed07b67c1e838f950b1c1fabc683030477e60b335649b8020e1076", size = 422093, upload-time = "2024-10-16T19:45:03.765Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/b3/3a1bc45be03dda7a60c7858e55b6cd0489a81613c1908fb81cf21d34ae50/httptools-0.6.4-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:0e563e54979e97b6d13f1bbc05a96109923e76b901f786a5eae36e99c01237bd", size = 423898, upload-time = "2024-10-16T19:45:05.683Z" },
+]
+
 [[package]]
 name = "httpx"
 version = "0.28.1"
@@ -1084,7 +1438,7 @@ wheels = [
 
 [[package]]
 name = "huggingface-hub"
-version = "0.34.4"
+version = "0.35.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "filelock" },
@@ -1096,9 +1450,9 @@ dependencies = [
     { name = "tqdm" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/45/c9/bdbe19339f76d12985bc03572f330a01a93c04dffecaaea3061bdd7fb892/huggingface_hub-0.34.4.tar.gz", hash = "sha256:a4228daa6fb001be3f4f4bdaf9a0db00e1739235702848df00885c9b5742c85c", size = 459768, upload-time = "2025-08-08T09:14:52.365Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/10/7e/a0a97de7c73671863ca6b3f61fa12518caf35db37825e43d63a70956738c/huggingface_hub-0.35.3.tar.gz", hash = "sha256:350932eaa5cc6a4747efae85126ee220e4ef1b54e29d31c3b45c5612ddf0b32a", size = 461798, upload-time = "2025-09-29T14:29:58.625Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/39/7b/bb06b061991107cd8783f300adff3e7b7f284e330fd82f507f2a1417b11d/huggingface_hub-0.34.4-py3-none-any.whl", hash = "sha256:9b365d781739c93ff90c359844221beef048403f1bc1f1c123c191257c3c890a", size = 561452, upload-time = "2025-08-08T09:14:50.159Z" },
+    { url = "https://files.pythonhosted.org/packages/31/a0/651f93d154cb72323358bf2bbae3e642bdb5d2f1bfc874d096f7cb159fa0/huggingface_hub-0.35.3-py3-none-any.whl", hash = "sha256:0e3a01829c19d86d03793e4577816fe3bdfc1602ac62c7fb220d593d351224ba", size = 564262, upload-time = "2025-09-29T14:29:55.813Z" },
 ]
 
 [[package]]
@@ -1112,11 +1466,11 @@ wheels = [
 
 [[package]]
 name = "identify"
-version = "2.6.13"
+version = "2.6.15"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/82/ca/ffbabe3635bb839aa36b3a893c91a9b0d368cb4d8073e03a12896970af82/identify-2.6.13.tar.gz", hash = "sha256:da8d6c828e773620e13bfa86ea601c5a5310ba4bcd65edf378198b56a1f9fb32", size = 99243, upload-time = "2025-08-09T19:35:00.6Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/ff/e7/685de97986c916a6d93b3876139e00eef26ad5bbbd61925d670ae8013449/identify-2.6.15.tar.gz", hash = "sha256:e4f4864b96c6557ef2a1e1c951771838f4edc9df3a72ec7118b338801b11c7bf", size = 99311, upload-time = "2025-10-02T17:43:40.631Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e7/ce/461b60a3ee109518c055953729bf9ed089a04db895d47e95444071dcdef2/identify-2.6.13-py2.py3-none-any.whl", hash = "sha256:60381139b3ae39447482ecc406944190f690d4a2997f2584062089848361b33b", size = 99153, upload-time = "2025-08-09T19:34:59.1Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/1c/e5fd8f973d4f375adb21565739498e2e9a1e54c858a97b9a8ccfdc81da9b/identify-2.6.15-py2.py3-none-any.whl", hash = "sha256:1181ef7608e00704db228516541eb83a88a9f94433a8c80bb9b5bd54b1d81757", size = 99183, upload-time = "2025-10-02T17:43:39.137Z" },
 ]
 
 [[package]]
@@ -1161,6 +1515,24 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/2c/e1/e6716421ea10d38022b952c159d5161ca1193197fb744506875fbb87ea7b/iniconfig-2.1.0-py3-none-any.whl", hash = "sha256:9deba5723312380e77435581c6bf4935c94cbfab9b1ed33ef8d238ea168eb760", size = 6050, upload-time = "2025-03-19T20:10:01.071Z" },
 ]
 
+[[package]]
+name = "interegular"
+version = "0.3.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/dc/9d/8b6dde58a028a3962ce17e84d5fe73758df61378e00ef8ac3d85da34b0ff/interegular-0.3.3.tar.gz", hash = "sha256:d9b697b21b34884711399ba0f0376914b81899ce670032486d0d048344a76600", size = 24705, upload-time = "2024-01-06T23:01:22.372Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c4/01/72d6472f80651673716d1deda2a5bbb633e563ecf94f4479da5519d69d25/interegular-0.3.3-py37-none-any.whl", hash = "sha256:b0c07007d48c89d6d19f7204972d369b2a77222722e126b6aa63aa721dc3b19c", size = 23635, upload-time = "2024-01-06T23:01:20.829Z" },
+]
+
+[[package]]
+name = "invoke"
+version = "2.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f9/42/127e6d792884ab860defc3f4d80a8f9812e48ace584ffc5a346de58cdc6c/invoke-2.2.0.tar.gz", hash = "sha256:ee6cbb101af1a859c7fe84f2a264c059020b0cb7fe3535f9424300ab568f6bd5", size = 299835, upload-time = "2023-07-12T18:05:17.998Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0a/66/7f8c48009c72d73bc6bbe6eb87ac838d6a526146f7dab14af671121eb379/invoke-2.2.0-py3-none-any.whl", hash = "sha256:6ea924cc53d4f78e3d98bc436b08069a03077e6f85ad1ddaa8a116d7dad15820", size = 160274, upload-time = "2023-07-12T18:05:16.294Z" },
+]
+
 [[package]]
 name = "iso3166"
 version = "2.1.1"
@@ -1211,22 +1583,22 @@ wheels = [
 
 [[package]]
 name = "jax"
-version = "0.7.1"
+version = "0.7.2"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "ml-dtypes", marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "opt-einsum", marker = "python_full_version >= '3.11'" },
-    { name = "scipy", version = "1.16.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "scipy", version = "1.16.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/bc/e8/b393ee314d3b042bd66b986d38e52f4e6046590399d916381265c20467d3/jax-0.7.1.tar.gz", hash = "sha256:118f56338c503361d2791f069d24339d8d44a8db442ed851d2e591222fb7a56d", size = 2428411, upload-time = "2025-08-20T15:55:46.098Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/8f/e7/1e8e8af59b7659c83dc07dfa1dc23bc13551e5ef89bdef19ced044a497fc/jax-0.7.2.tar.gz", hash = "sha256:71a42b964bc6d52e819311429e6c0f5742e2a4650226dab1a1dd26fd986ca70d", size = 2434085, upload-time = "2025-09-16T16:48:53.108Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/83/81/793d78c91b0546b3b1f08e55fdd97437174171cd7d70e46098f1a4d94b7b/jax-0.7.1-py3-none-any.whl", hash = "sha256:056e576e0e58465506125699f48111ac8891cce4c9ebf034704c42b219dfd4a6", size = 2827341, upload-time = "2025-08-20T15:55:44.576Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/e6/5fd0f6fff79eb47469ff9c4fa27125b661517a2fbf8884689b02e9fdfaa8/jax-0.7.2-py3-none-any.whl", hash = "sha256:e7e32f9be51ae5cc6854225958c57de8cca2187d279844338465b15e8a1fe7f2", size = 2835570, upload-time = "2025-09-16T16:48:51.33Z" },
 ]
 
 [[package]]
@@ -1293,7 +1665,7 @@ wheels = [
 
 [[package]]
 name = "jaxlib"
-version = "0.7.1"
+version = "0.7.2"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -1302,17 +1674,17 @@ resolution-markers = [
 dependencies = [
     { name = "ml-dtypes", marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "scipy", version = "1.16.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "scipy", version = "1.16.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/8b/af/5058d545e95f99a54289648f5430cc3c23263dd70a1391e7491f24ed328d/jaxlib-0.7.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3f32c3e4c167b7327c342e82d3df84079714ea0b43718be871d039999670b3c9", size = 57686934, upload-time = "2025-08-20T15:55:58.989Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/77/ef7f6cd03e699da7d9755f88741c29b3015654473fc9d5f906da19edcb47/jaxlib-0.7.1-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:9fb189c3b39470c4394ffcb18b71e47cffc5bf85e8fcb1e33692686b0c3e04dd", size = 85134885, upload-time = "2025-08-20T15:56:03.484Z" },
-    { url = "https://files.pythonhosted.org/packages/4d/72/304018d46703f337787f010735f70d17212f86778fcba8bb5cf678f8e460/jaxlib-0.7.1-cp311-cp311-manylinux_2_27_x86_64.whl", hash = "sha256:eaf5f68f53bf4dcb93b6512538547667625588e4f3ccaeef048788fd18d8c0d5", size = 81147868, upload-time = "2025-08-20T15:56:07.214Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/b7/0f0df407518691099d659ba6e19db01320dfb58e49d80594eaddd57d77c1/jaxlib-0.7.1-cp311-cp311-win_amd64.whl", hash = "sha256:ab4510fbaeafac6c794ab335f23e71200d824c48f6a0ab20553db8deab8805c5", size = 61185342, upload-time = "2025-08-20T15:56:10.452Z" },
-    { url = "https://files.pythonhosted.org/packages/ef/1f/10543d7a3f7e76dd4bbdc77134890ac2f41bc8570c565961464f6320009b/jaxlib-0.7.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:127c07c727703e5d59f84f655169bec849f4422e52f8546349cecc30a8a13e1d", size = 57682851, upload-time = "2025-08-20T15:56:13.395Z" },
-    { url = "https://files.pythonhosted.org/packages/de/4d/76ee71959311fe3da9951aa6f55af8f98eb3572bb322f5a7c89faf7ab933/jaxlib-0.7.1-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:f0f1f52956b8c2518ab000a4d3d8c21be777e1d47f926ba03640e391061a41ee", size = 85133707, upload-time = "2025-08-20T15:56:16.908Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/50/e37d02e250f5feb755112ec95b1c012a36d48a99209277267037d100f630/jaxlib-0.7.1-cp312-cp312-manylinux_2_27_x86_64.whl", hash = "sha256:74abd3135797f82440dd3711a35cba16c430d1bba65474b85bb70e41733a52e9", size = 81156916, upload-time = "2025-08-20T15:56:20.41Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/97/c6c28dfe57cccffd85512615416024b52dd327d78270204caba9311e71f1/jaxlib-0.7.1-cp312-cp312-win_amd64.whl", hash = "sha256:c4023863b14f280516f24ecb7539b4300a3236ea81ed69ad82595beceed1ba1f", size = 61212445, upload-time = "2025-08-20T15:56:23.929Z" },
+    { url = "https://files.pythonhosted.org/packages/55/fe/41e1df8a02acdc12472b2922e41495cf0a9b805f5fd1f163c9aaf81b30df/jaxlib-0.7.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:9fbf90afce30e34eba2ea929a506f5907bdd4062358122de499ce9e671a0ba1f", size = 53550443, upload-time = "2025-09-16T16:49:04.061Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/9b/c86d5c3cf354326b6d0c89f7a1562299e0cdcf3f507c82a6aca8ce77dcb6/jaxlib-0.7.2-cp311-cp311-manylinux_2_27_aarch64.whl", hash = "sha256:8ca7003351fbe8ccfa2fa5a493ec2dfbf2df92441306cf5c3b970508eedb92ab", size = 71551113, upload-time = "2025-09-16T16:49:07.95Z" },
+    { url = "https://files.pythonhosted.org/packages/80/dd/4a0565f636c0e702777bdb0afeeb262a019039869268194cbd0440ad47da/jaxlib-0.7.2-cp311-cp311-manylinux_2_27_x86_64.whl", hash = "sha256:4382006235cced59d2f795acc983c1bedcfbca4fea8f9461311d61c6a793ae66", size = 78159922, upload-time = "2025-09-16T16:49:11.656Z" },
+    { url = "https://files.pythonhosted.org/packages/28/03/199d9ce30981c6c8d2ae3a442d2fc3eeb1f31fa6aab9620ff1d3d3fdfd7f/jaxlib-0.7.2-cp311-cp311-win_amd64.whl", hash = "sha256:3b37eb937e0b8ed4f9b265fdf46a3cf64e1decd4f41c0053e96540d39bd7050c", size = 58013064, upload-time = "2025-09-16T16:49:15.063Z" },
+    { url = "https://files.pythonhosted.org/packages/df/82/324ffb86a6de8c149689b499b6dc6e4f83684fe43ae9e81c8e5eb91bc50f/jaxlib-0.7.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:bd6d1c53bd475e0e768a54af98b1642fb49d7304cf055ceebb1d01e89d38a1cb", size = 53563644, upload-time = "2025-09-16T16:49:18.04Z" },
+    { url = "https://files.pythonhosted.org/packages/89/6a/689b177de0672a7b4fe5489eec7cbb11305959c3af8da7d646e7f3aeb754/jaxlib-0.7.2-cp312-cp312-manylinux_2_27_aarch64.whl", hash = "sha256:9b5a0d357497611a113d207fb2c1997f01ab7a175870700812220f0bcaa31822", size = 71557871, upload-time = "2025-09-16T16:49:20.964Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/73/b44fbe943c9e02e25c99eb64e6b86e2dde8d918d064326813b5bbe620951/jaxlib-0.7.2-cp312-cp312-manylinux_2_27_x86_64.whl", hash = "sha256:11f32319e662ccff66859eb393757050d8971bd880bc4dd70dec6434d890fb59", size = 78167979, upload-time = "2025-09-16T16:49:24.607Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/0e/13d5264bbe08f03ae9e7abe7aa3041b5c29f9daf5493b8156ae7baa6cbf1/jaxlib-0.7.2-cp312-cp312-win_amd64.whl", hash = "sha256:b53cf3a7ed342ca1a9f419cc7a3c387598fc743043ba4a7c5895ebc4d62fa05a", size = 58040487, upload-time = "2025-09-16T16:49:28.101Z" },
 ]
 
 [[package]]
@@ -1329,58 +1701,59 @@ wheels = [
 
 [[package]]
 name = "jiter"
-version = "0.10.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/ee/9d/ae7ddb4b8ab3fb1b51faf4deb36cb48a4fbbd7cb36bad6a5fca4741306f7/jiter-0.10.0.tar.gz", hash = "sha256:07a7142c38aacc85194391108dc91b5b57093c978a9932bd86a36862759d9500", size = 162759, upload-time = "2025-05-18T19:04:59.73Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/be/7e/4011b5c77bec97cb2b572f566220364e3e21b51c48c5bd9c4a9c26b41b67/jiter-0.10.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:cd2fb72b02478f06a900a5782de2ef47e0396b3e1f7d5aba30daeb1fce66f303", size = 317215, upload-time = "2025-05-18T19:03:04.303Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/4f/144c1b57c39692efc7ea7d8e247acf28e47d0912800b34d0ad815f6b2824/jiter-0.10.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:32bb468e3af278f095d3fa5b90314728a6916d89ba3d0ffb726dd9bf7367285e", size = 322814, upload-time = "2025-05-18T19:03:06.433Z" },
-    { url = "https://files.pythonhosted.org/packages/63/1f/db977336d332a9406c0b1f0b82be6f71f72526a806cbb2281baf201d38e3/jiter-0.10.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:aa8b3e0068c26ddedc7abc6fac37da2d0af16b921e288a5a613f4b86f050354f", size = 345237, upload-time = "2025-05-18T19:03:07.833Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/1c/aa30a4a775e8a672ad7f21532bdbfb269f0706b39c6ff14e1f86bdd9e5ff/jiter-0.10.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:286299b74cc49e25cd42eea19b72aa82c515d2f2ee12d11392c56d8701f52224", size = 370999, upload-time = "2025-05-18T19:03:09.338Z" },
-    { url = "https://files.pythonhosted.org/packages/35/df/f8257abc4207830cb18880781b5f5b716bad5b2a22fb4330cfd357407c5b/jiter-0.10.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6ed5649ceeaeffc28d87fb012d25a4cd356dcd53eff5acff1f0466b831dda2a7", size = 491109, upload-time = "2025-05-18T19:03:11.13Z" },
-    { url = "https://files.pythonhosted.org/packages/06/76/9e1516fd7b4278aa13a2cc7f159e56befbea9aa65c71586305e7afa8b0b3/jiter-0.10.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b2ab0051160cb758a70716448908ef14ad476c3774bd03ddce075f3c1f90a3d6", size = 388608, upload-time = "2025-05-18T19:03:12.911Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/64/67750672b4354ca20ca18d3d1ccf2c62a072e8a2d452ac3cf8ced73571ef/jiter-0.10.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:03997d2f37f6b67d2f5c475da4412be584e1cec273c1cfc03d642c46db43f8cf", size = 352454, upload-time = "2025-05-18T19:03:14.741Z" },
-    { url = "https://files.pythonhosted.org/packages/96/4d/5c4e36d48f169a54b53a305114be3efa2bbffd33b648cd1478a688f639c1/jiter-0.10.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c404a99352d839fed80d6afd6c1d66071f3bacaaa5c4268983fc10f769112e90", size = 391833, upload-time = "2025-05-18T19:03:16.426Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/de/ce4a6166a78810bd83763d2fa13f85f73cbd3743a325469a4a9289af6dae/jiter-0.10.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:66e989410b6666d3ddb27a74c7e50d0829704ede652fd4c858e91f8d64b403d0", size = 523646, upload-time = "2025-05-18T19:03:17.704Z" },
-    { url = "https://files.pythonhosted.org/packages/a2/a6/3bc9acce53466972964cf4ad85efecb94f9244539ab6da1107f7aed82934/jiter-0.10.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:b532d3af9ef4f6374609a3bcb5e05a1951d3bf6190dc6b176fdb277c9bbf15ee", size = 514735, upload-time = "2025-05-18T19:03:19.44Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/d8/243c2ab8426a2a4dea85ba2a2ba43df379ccece2145320dfd4799b9633c5/jiter-0.10.0-cp310-cp310-win32.whl", hash = "sha256:da9be20b333970e28b72edc4dff63d4fec3398e05770fb3205f7fb460eb48dd4", size = 210747, upload-time = "2025-05-18T19:03:21.184Z" },
-    { url = "https://files.pythonhosted.org/packages/37/7a/8021bd615ef7788b98fc76ff533eaac846322c170e93cbffa01979197a45/jiter-0.10.0-cp310-cp310-win_amd64.whl", hash = "sha256:f59e533afed0c5b0ac3eba20d2548c4a550336d8282ee69eb07b37ea526ee4e5", size = 207484, upload-time = "2025-05-18T19:03:23.046Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/dd/6cefc6bd68b1c3c979cecfa7029ab582b57690a31cd2f346c4d0ce7951b6/jiter-0.10.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:3bebe0c558e19902c96e99217e0b8e8b17d570906e72ed8a87170bc290b1e978", size = 317473, upload-time = "2025-05-18T19:03:25.942Z" },
-    { url = "https://files.pythonhosted.org/packages/be/cf/fc33f5159ce132be1d8dd57251a1ec7a631c7df4bd11e1cd198308c6ae32/jiter-0.10.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:558cc7e44fd8e507a236bee6a02fa17199ba752874400a0ca6cd6e2196cdb7dc", size = 321971, upload-time = "2025-05-18T19:03:27.255Z" },
-    { url = "https://files.pythonhosted.org/packages/68/a4/da3f150cf1d51f6c472616fb7650429c7ce053e0c962b41b68557fdf6379/jiter-0.10.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4d613e4b379a07d7c8453c5712ce7014e86c6ac93d990a0b8e7377e18505e98d", size = 345574, upload-time = "2025-05-18T19:03:28.63Z" },
-    { url = "https://files.pythonhosted.org/packages/84/34/6e8d412e60ff06b186040e77da5f83bc158e9735759fcae65b37d681f28b/jiter-0.10.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:f62cf8ba0618eda841b9bf61797f21c5ebd15a7a1e19daab76e4e4b498d515b2", size = 371028, upload-time = "2025-05-18T19:03:30.292Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/d9/9ee86173aae4576c35a2f50ae930d2ccb4c4c236f6cb9353267aa1d626b7/jiter-0.10.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:919d139cdfa8ae8945112398511cb7fca58a77382617d279556b344867a37e61", size = 491083, upload-time = "2025-05-18T19:03:31.654Z" },
-    { url = "https://files.pythonhosted.org/packages/d9/2c/f955de55e74771493ac9e188b0f731524c6a995dffdcb8c255b89c6fb74b/jiter-0.10.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:13ddbc6ae311175a3b03bd8994881bc4635c923754932918e18da841632349db", size = 388821, upload-time = "2025-05-18T19:03:33.184Z" },
-    { url = "https://files.pythonhosted.org/packages/81/5a/0e73541b6edd3f4aada586c24e50626c7815c561a7ba337d6a7eb0a915b4/jiter-0.10.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4c440ea003ad10927a30521a9062ce10b5479592e8a70da27f21eeb457b4a9c5", size = 352174, upload-time = "2025-05-18T19:03:34.965Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/c0/61eeec33b8c75b31cae42be14d44f9e6fe3ac15a4e58010256ac3abf3638/jiter-0.10.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:dc347c87944983481e138dea467c0551080c86b9d21de6ea9306efb12ca8f606", size = 391869, upload-time = "2025-05-18T19:03:36.436Z" },
-    { url = "https://files.pythonhosted.org/packages/41/22/5beb5ee4ad4ef7d86f5ea5b4509f680a20706c4a7659e74344777efb7739/jiter-0.10.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:13252b58c1f4d8c5b63ab103c03d909e8e1e7842d302473f482915d95fefd605", size = 523741, upload-time = "2025-05-18T19:03:38.168Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/10/768e8818538e5817c637b0df52e54366ec4cebc3346108a4457ea7a98f32/jiter-0.10.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:7d1bbf3c465de4a24ab12fb7766a0003f6f9bce48b8b6a886158c4d569452dc5", size = 514527, upload-time = "2025-05-18T19:03:39.577Z" },
-    { url = "https://files.pythonhosted.org/packages/73/6d/29b7c2dc76ce93cbedabfd842fc9096d01a0550c52692dfc33d3cc889815/jiter-0.10.0-cp311-cp311-win32.whl", hash = "sha256:db16e4848b7e826edca4ccdd5b145939758dadf0dc06e7007ad0e9cfb5928ae7", size = 210765, upload-time = "2025-05-18T19:03:41.271Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/c9/d394706deb4c660137caf13e33d05a031d734eb99c051142e039d8ceb794/jiter-0.10.0-cp311-cp311-win_amd64.whl", hash = "sha256:9c9c1d5f10e18909e993f9641f12fe1c77b3e9b533ee94ffa970acc14ded3812", size = 209234, upload-time = "2025-05-18T19:03:42.918Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/b5/348b3313c58f5fbfb2194eb4d07e46a35748ba6e5b3b3046143f3040bafa/jiter-0.10.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:1e274728e4a5345a6dde2d343c8da018b9d4bd4350f5a472fa91f66fda44911b", size = 312262, upload-time = "2025-05-18T19:03:44.637Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/4a/6a2397096162b21645162825f058d1709a02965606e537e3304b02742e9b/jiter-0.10.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:7202ae396446c988cb2a5feb33a543ab2165b786ac97f53b59aafb803fef0744", size = 320124, upload-time = "2025-05-18T19:03:46.341Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/85/1ce02cade7516b726dd88f59a4ee46914bf79d1676d1228ef2002ed2f1c9/jiter-0.10.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:23ba7722d6748b6920ed02a8f1726fb4b33e0fd2f3f621816a8b486c66410ab2", size = 345330, upload-time = "2025-05-18T19:03:47.596Z" },
-    { url = "https://files.pythonhosted.org/packages/75/d0/bb6b4f209a77190ce10ea8d7e50bf3725fc16d3372d0a9f11985a2b23eff/jiter-0.10.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:371eab43c0a288537d30e1f0b193bc4eca90439fc08a022dd83e5e07500ed026", size = 369670, upload-time = "2025-05-18T19:03:49.334Z" },
-    { url = "https://files.pythonhosted.org/packages/a0/f5/a61787da9b8847a601e6827fbc42ecb12be2c925ced3252c8ffcb56afcaf/jiter-0.10.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6c675736059020365cebc845a820214765162728b51ab1e03a1b7b3abb70f74c", size = 489057, upload-time = "2025-05-18T19:03:50.66Z" },
-    { url = "https://files.pythonhosted.org/packages/12/e4/6f906272810a7b21406c760a53aadbe52e99ee070fc5c0cb191e316de30b/jiter-0.10.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0c5867d40ab716e4684858e4887489685968a47e3ba222e44cde6e4a2154f959", size = 389372, upload-time = "2025-05-18T19:03:51.98Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/ba/77013b0b8ba904bf3762f11e0129b8928bff7f978a81838dfcc958ad5728/jiter-0.10.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:395bb9a26111b60141757d874d27fdea01b17e8fac958b91c20128ba8f4acc8a", size = 352038, upload-time = "2025-05-18T19:03:53.703Z" },
-    { url = "https://files.pythonhosted.org/packages/67/27/c62568e3ccb03368dbcc44a1ef3a423cb86778a4389e995125d3d1aaa0a4/jiter-0.10.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:6842184aed5cdb07e0c7e20e5bdcfafe33515ee1741a6835353bb45fe5d1bd95", size = 391538, upload-time = "2025-05-18T19:03:55.046Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/72/0d6b7e31fc17a8fdce76164884edef0698ba556b8eb0af9546ae1a06b91d/jiter-0.10.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:62755d1bcea9876770d4df713d82606c8c1a3dca88ff39046b85a048566d56ea", size = 523557, upload-time = "2025-05-18T19:03:56.386Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/09/bc1661fbbcbeb6244bd2904ff3a06f340aa77a2b94e5a7373fd165960ea3/jiter-0.10.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:533efbce2cacec78d5ba73a41756beff8431dfa1694b6346ce7af3a12c42202b", size = 514202, upload-time = "2025-05-18T19:03:57.675Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/84/5a5d5400e9d4d54b8004c9673bbe4403928a00d28529ff35b19e9d176b19/jiter-0.10.0-cp312-cp312-win32.whl", hash = "sha256:8be921f0cadd245e981b964dfbcd6fd4bc4e254cdc069490416dd7a2632ecc01", size = 211781, upload-time = "2025-05-18T19:03:59.025Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/52/7ec47455e26f2d6e5f2ea4951a0652c06e5b995c291f723973ae9e724a65/jiter-0.10.0-cp312-cp312-win_amd64.whl", hash = "sha256:a7c7d785ae9dda68c2678532a5a1581347e9c15362ae9f6e68f3fdbfb64f2e49", size = 206176, upload-time = "2025-05-18T19:04:00.305Z" },
-    { url = "https://files.pythonhosted.org/packages/98/fd/aced428e2bd3c6c1132f67c5a708f9e7fd161d0ca8f8c5862b17b93cdf0a/jiter-0.10.0-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:bd6292a43c0fc09ce7c154ec0fa646a536b877d1e8f2f96c19707f65355b5a4d", size = 317665, upload-time = "2025-05-18T19:04:43.417Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/2e/47d42f15d53ed382aef8212a737101ae2720e3697a954f9b95af06d34e89/jiter-0.10.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:39de429dcaeb6808d75ffe9effefe96a4903c6a4b376b2f6d08d77c1aaee2f18", size = 312152, upload-time = "2025-05-18T19:04:44.797Z" },
-    { url = "https://files.pythonhosted.org/packages/7b/02/aae834228ef4834fc18718724017995ace8da5f70aa1ec225b9bc2b2d7aa/jiter-0.10.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:52ce124f13a7a616fad3bb723f2bfb537d78239d1f7f219566dc52b6f2a9e48d", size = 346708, upload-time = "2025-05-18T19:04:46.127Z" },
-    { url = "https://files.pythonhosted.org/packages/35/d4/6ff39dee2d0a9abd69d8a3832ce48a3aa644eed75e8515b5ff86c526ca9a/jiter-0.10.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:166f3606f11920f9a1746b2eea84fa2c0a5d50fd313c38bdea4edc072000b0af", size = 371360, upload-time = "2025-05-18T19:04:47.448Z" },
-    { url = "https://files.pythonhosted.org/packages/a9/67/c749d962b4eb62445867ae4e64a543cbb5d63cc7d78ada274ac515500a7f/jiter-0.10.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:28dcecbb4ba402916034fc14eba7709f250c4d24b0c43fc94d187ee0580af181", size = 492105, upload-time = "2025-05-18T19:04:48.792Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/d3/8fe1b1bae5161f27b1891c256668f598fa4c30c0a7dacd668046a6215fca/jiter-0.10.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:86c5aa6910f9bebcc7bc4f8bc461aff68504388b43bfe5e5c0bd21efa33b52f4", size = 389577, upload-time = "2025-05-18T19:04:50.13Z" },
-    { url = "https://files.pythonhosted.org/packages/ef/28/ecb19d789b4777898a4252bfaac35e3f8caf16c93becd58dcbaac0dc24ad/jiter-0.10.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ceeb52d242b315d7f1f74b441b6a167f78cea801ad7c11c36da77ff2d42e8a28", size = 353849, upload-time = "2025-05-18T19:04:51.443Z" },
-    { url = "https://files.pythonhosted.org/packages/77/69/261f798f84790da6482ebd8c87ec976192b8c846e79444d0a2e0d33ebed8/jiter-0.10.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:ff76d8887c8c8ee1e772274fcf8cc1071c2c58590d13e33bd12d02dc9a560397", size = 392029, upload-time = "2025-05-18T19:04:52.792Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/08/b8d15140d4d91f16faa2f5d416c1a71ab1bbe2b66c57197b692d04c0335f/jiter-0.10.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:a9be4d0fa2b79f7222a88aa488bd89e2ae0a0a5b189462a12def6ece2faa45f1", size = 524386, upload-time = "2025-05-18T19:04:54.203Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/1d/23c41765cc95c0e23ac492a88450d34bf0fd87a37218d1b97000bffe0f53/jiter-0.10.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:9ab7fd8738094139b6c1ab1822d6f2000ebe41515c537235fd45dabe13ec9324", size = 515234, upload-time = "2025-05-18T19:04:55.838Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/14/381d8b151132e79790579819c3775be32820569f23806769658535fe467f/jiter-0.10.0-cp39-cp39-win32.whl", hash = "sha256:5f51e048540dd27f204ff4a87f5d79294ea0aa3aa552aca34934588cf27023cf", size = 211436, upload-time = "2025-05-18T19:04:57.183Z" },
-    { url = "https://files.pythonhosted.org/packages/59/66/f23ae51dea8ee8ce429027b60008ca895d0fa0704f0c7fe5f09014a6cffb/jiter-0.10.0-cp39-cp39-win_amd64.whl", hash = "sha256:1b28302349dc65703a9e4ead16f163b1c339efffbe1049c30a44b001a2a4fff9", size = 208777, upload-time = "2025-05-18T19:04:58.454Z" },
+version = "0.11.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/9d/c0/a3bb4cc13aced219dd18191ea66e874266bd8aa7b96744e495e1c733aa2d/jiter-0.11.0.tar.gz", hash = "sha256:1d9637eaf8c1d6a63d6562f2a6e5ab3af946c66037eb1b894e8fad75422266e4", size = 167094, upload-time = "2025-09-15T09:20:38.212Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/25/21/7dd1235a19e26979be6098e87e4cced2e061752f3a40a17bbce6dea7fae1/jiter-0.11.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:3893ce831e1c0094a83eeaf56c635a167d6fa8cc14393cc14298fd6fdc2a2449", size = 309875, upload-time = "2025-09-15T09:18:48.41Z" },
+    { url = "https://files.pythonhosted.org/packages/71/f9/462b54708aa85b135733ccba70529dd68a18511bf367a87c5fd28676c841/jiter-0.11.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:25c625b9b61b5a8725267fdf867ef2e51b429687f6a4eef211f4612e95607179", size = 316505, upload-time = "2025-09-15T09:18:51.057Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/40/14e2eeaac6a47bff27d213834795472355fd39769272eb53cb7aa83d5aa8/jiter-0.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dd4ca85fb6a62cf72e1c7f5e34ddef1b660ce4ed0886ec94a1ef9777d35eaa1f", size = 337613, upload-time = "2025-09-15T09:18:52.358Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/ed/a5f1f8419c92b150a7c7fb5ccba1fb1e192887ad713d780e70874f0ce996/jiter-0.11.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:572208127034725e79c28437b82414028c3562335f2b4f451d98136d0fc5f9cd", size = 361438, upload-time = "2025-09-15T09:18:54.637Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/f5/70682c023dfcdd463a53faf5d30205a7d99c51d70d3e303c932d0936e5a2/jiter-0.11.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:494ba627c7f550ad3dabb21862864b8f2216098dc18ff62f37b37796f2f7c325", size = 486180, upload-time = "2025-09-15T09:18:56.158Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/39/020d08cbab4eab48142ad88b837c41eb08a15c0767fdb7c0d3265128a44b/jiter-0.11.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b8da18a99f58bca3ecc2d2bba99cac000a924e115b6c4f0a2b98f752b6fbf39a", size = 376681, upload-time = "2025-09-15T09:18:57.553Z" },
+    { url = "https://files.pythonhosted.org/packages/52/10/b86733f6e594cf51dd142f37c602d8df87c554c5844958deaab0de30eb5d/jiter-0.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e4ffd3b0fff3fabbb02cc09910c08144db6bb5697a98d227a074401e01ee63dd", size = 348685, upload-time = "2025-09-15T09:18:59.208Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/ee/8861665e83a9e703aa5f65fddddb6225428e163e6b0baa95a7f9a8fb9aae/jiter-0.11.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:8fe6530aa738a4f7d4e4702aa8f9581425d04036a5f9e25af65ebe1f708f23be", size = 385573, upload-time = "2025-09-15T09:19:00.593Z" },
+    { url = "https://files.pythonhosted.org/packages/25/74/05afec03600951f128293813b5a208c9ba1bf587c57a344c05a42a69e1b1/jiter-0.11.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:e35d66681c133a03d7e974e7eedae89720fe8ca3bd09f01a4909b86a8adf31f5", size = 516669, upload-time = "2025-09-15T09:19:02.369Z" },
+    { url = "https://files.pythonhosted.org/packages/93/d1/2e5bfe147cfbc2a5eef7f73eb75dc5c6669da4fa10fc7937181d93af9495/jiter-0.11.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:c59459beca2fbc9718b6f1acb7bfb59ebc3eb4294fa4d40e9cb679dafdcc6c60", size = 508767, upload-time = "2025-09-15T09:19:04.011Z" },
+    { url = "https://files.pythonhosted.org/packages/87/50/597f71307e10426b5c082fd05d38c615ddbdd08c3348d8502963307f0652/jiter-0.11.0-cp310-cp310-win32.whl", hash = "sha256:b7b0178417b0dcfc5f259edbc6db2b1f5896093ed9035ee7bab0f2be8854726d", size = 205476, upload-time = "2025-09-15T09:19:05.594Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/86/1e5214b3272e311754da26e63edec93a183811d4fc2e0118addec365df8b/jiter-0.11.0-cp310-cp310-win_amd64.whl", hash = "sha256:11df2bf99fb4754abddd7f5d940a48e51f9d11624d6313ca4314145fcad347f0", size = 204708, upload-time = "2025-09-15T09:19:06.955Z" },
+    { url = "https://files.pythonhosted.org/packages/38/55/a69fefeef09c2eaabae44b935a1aa81517e49639c0a0c25d861cb18cd7ac/jiter-0.11.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:cb5d9db02979c3f49071fce51a48f4b4e4cf574175fb2b11c7a535fa4867b222", size = 309503, upload-time = "2025-09-15T09:19:08.191Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/d5/a6aba9e6551f32f9c127184f398208e4eddb96c59ac065c8a92056089d28/jiter-0.11.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1dc6a123f3471c4730db7ca8ba75f1bb3dcb6faeb8d46dd781083e7dee88b32d", size = 317688, upload-time = "2025-09-15T09:19:09.918Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/f3/5e86f57c1883971cdc8535d0429c2787bf734840a231da30a3be12850562/jiter-0.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:09858f8d230f031c7b8e557429102bf050eea29c77ad9c34c8fe253c5329acb7", size = 337418, upload-time = "2025-09-15T09:19:11.078Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/4f/a71d8a24c2a70664970574a8e0b766663f5ef788f7fe1cc20ee0c016d488/jiter-0.11.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:dbe2196c4a0ce760925a74ab4456bf644748ab0979762139626ad138f6dac72d", size = 361423, upload-time = "2025-09-15T09:19:13.286Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/e5/b09076f4e7fd9471b91e16f9f3dc7330b161b738f3b39b2c37054a36e26a/jiter-0.11.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5beb56d22b63647bafd0b74979216fdee80c580c0c63410be8c11053860ffd09", size = 486367, upload-time = "2025-09-15T09:19:14.546Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/f1/98cb3a36f5e62f80cd860f0179f948d9eab5a316d55d3e1bab98d9767af5/jiter-0.11.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:97025d09ef549795d8dc720a824312cee3253c890ac73c621721ddfc75066789", size = 376335, upload-time = "2025-09-15T09:19:15.939Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/d8/ec74886497ea393c29dbd7651ddecc1899e86404a6b1f84a3ddab0ab59fd/jiter-0.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d50880a6da65d8c23a2cf53c412847d9757e74cc9a3b95c5704a1d1a24667347", size = 348981, upload-time = "2025-09-15T09:19:17.568Z" },
+    { url = "https://files.pythonhosted.org/packages/24/93/d22ad7fa3b86ade66c86153ceea73094fc2af8b20c59cb7fceab9fea4704/jiter-0.11.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:452d80a1c86c095a242007bd9fc5d21b8a8442307193378f891cb8727e469648", size = 385797, upload-time = "2025-09-15T09:19:19.121Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/bd/e25ff4a4df226e9b885f7cb01ee4b9dc74e3000e612d6f723860d71a1f34/jiter-0.11.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:e84e58198d4894668eec2da660ffff60e0f3e60afa790ecc50cb12b0e02ca1d4", size = 516597, upload-time = "2025-09-15T09:19:20.301Z" },
+    { url = "https://files.pythonhosted.org/packages/be/fb/beda613db7d93ffa2fdd2683f90f2f5dce8daf4bc2d0d2829e7de35308c6/jiter-0.11.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:df64edcfc5dd5279a791eea52aa113d432c933119a025b0b5739f90d2e4e75f1", size = 508853, upload-time = "2025-09-15T09:19:22.075Z" },
+    { url = "https://files.pythonhosted.org/packages/20/64/c5b0d93490634e41e38e2a15de5d54fdbd2c9f64a19abb0f95305b63373c/jiter-0.11.0-cp311-cp311-win32.whl", hash = "sha256:144fc21337d21b1d048f7f44bf70881e1586401d405ed3a98c95a114a9994982", size = 205140, upload-time = "2025-09-15T09:19:23.351Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/e6/c347c0e6f5796e97d4356b7e5ff0ce336498b7f4ef848fae621a56f1ccf3/jiter-0.11.0-cp311-cp311-win_amd64.whl", hash = "sha256:b0f32e644d241293b892b1a6dd8f0b9cc029bfd94c97376b2681c36548aabab7", size = 204311, upload-time = "2025-09-15T09:19:24.591Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/b5/3009b112b8f673e568ef79af9863d8309a15f0a8cdcc06ed6092051f377e/jiter-0.11.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:2fb7b377688cc3850bbe5c192a6bd493562a0bc50cbc8b047316428fbae00ada", size = 305510, upload-time = "2025-09-15T09:19:25.893Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/82/15514244e03b9e71e086bbe2a6de3e4616b48f07d5f834200c873956fb8c/jiter-0.11.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a1b7cbe3f25bd0d8abb468ba4302a5d45617ee61b2a7a638f63fee1dc086be99", size = 316521, upload-time = "2025-09-15T09:19:27.525Z" },
+    { url = "https://files.pythonhosted.org/packages/92/94/7a2e905f40ad2d6d660e00b68d818f9e29fb87ffe82774f06191e93cbe4a/jiter-0.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c0a7f0ec81d5b7588c5cade1eb1925b91436ae6726dc2df2348524aeabad5de6", size = 338214, upload-time = "2025-09-15T09:19:28.727Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/9c/5791ed5bdc76f12110158d3316a7a3ec0b1413d018b41c5ed399549d3ad5/jiter-0.11.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:07630bb46ea2a6b9c6ed986c6e17e35b26148cce2c535454b26ee3f0e8dcaba1", size = 361280, upload-time = "2025-09-15T09:19:30.013Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/7f/b7d82d77ff0d2cb06424141000176b53a9e6b16a1125525bb51ea4990c2e/jiter-0.11.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7764f27d28cd4a9cbc61704dfcd80c903ce3aad106a37902d3270cd6673d17f4", size = 487895, upload-time = "2025-09-15T09:19:31.424Z" },
+    { url = "https://files.pythonhosted.org/packages/42/44/10a1475d46f1fc1fd5cc2e82c58e7bca0ce5852208e0fa5df2f949353321/jiter-0.11.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1d4a6c4a737d486f77f842aeb22807edecb4a9417e6700c7b981e16d34ba7c72", size = 378421, upload-time = "2025-09-15T09:19:32.746Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/5f/0dc34563d8164d31d07bc09d141d3da08157a68dcd1f9b886fa4e917805b/jiter-0.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cf408d2a0abd919b60de8c2e7bc5eeab72d4dafd18784152acc7c9adc3291591", size = 347932, upload-time = "2025-09-15T09:19:34.612Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/de/b68f32a4fcb7b4a682b37c73a0e5dae32180140cd1caf11aef6ad40ddbf2/jiter-0.11.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:cdef53eda7d18e799625023e1e250dbc18fbc275153039b873ec74d7e8883e09", size = 386959, upload-time = "2025-09-15T09:19:35.994Z" },
+    { url = "https://files.pythonhosted.org/packages/76/0a/c08c92e713b6e28972a846a81ce374883dac2f78ec6f39a0dad9f2339c3a/jiter-0.11.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:53933a38ef7b551dd9c7f1064f9d7bb235bb3168d0fa5f14f0798d1b7ea0d9c5", size = 517187, upload-time = "2025-09-15T09:19:37.426Z" },
+    { url = "https://files.pythonhosted.org/packages/89/b5/4a283bec43b15aad54fcae18d951f06a2ec3f78db5708d3b59a48e9c3fbd/jiter-0.11.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:11840d2324c9ab5162fc1abba23bc922124fedcff0d7b7f85fffa291e2f69206", size = 509461, upload-time = "2025-09-15T09:19:38.761Z" },
+    { url = "https://files.pythonhosted.org/packages/34/a5/f8bad793010534ea73c985caaeef8cc22dfb1fedb15220ecdf15c623c07a/jiter-0.11.0-cp312-cp312-win32.whl", hash = "sha256:4f01a744d24a5f2bb4a11657a1b27b61dc038ae2e674621a74020406e08f749b", size = 206664, upload-time = "2025-09-15T09:19:40.096Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/42/5823ec2b1469395a160b4bf5f14326b4a098f3b6898fbd327366789fa5d3/jiter-0.11.0-cp312-cp312-win_amd64.whl", hash = "sha256:29fff31190ab3a26de026da2f187814f4b9c6695361e20a9ac2123e4d4378a4c", size = 203520, upload-time = "2025-09-15T09:19:41.798Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/d9/51cf35d92bea21f2051da8ca2328831589e67e2bf971e85b1a6e6c0d2030/jiter-0.11.0-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:719891c2fb7628a41adff4f2f54c19380a27e6fdfdb743c24680ef1a54c67bd0", size = 312764, upload-time = "2025-09-15T09:20:21.378Z" },
+    { url = "https://files.pythonhosted.org/packages/da/48/eae309ce5c180faa1bb45e378a503717da22ceb2b0488f78e548f97c2b6b/jiter-0.11.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:df7f1927cbdf34cb91262a5418ca06920fd42f1cf733936d863aeb29b45a14ef", size = 305861, upload-time = "2025-09-15T09:20:22.603Z" },
+    { url = "https://files.pythonhosted.org/packages/83/4f/13b80e18b0331f0fecc09cb2f09f722530b9a395006941b01491fe58baea/jiter-0.11.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e71ae6d969d0c9bab336c5e9e2fabad31e74d823f19e3604eaf96d9a97f463df", size = 339507, upload-time = "2025-09-15T09:20:23.857Z" },
+    { url = "https://files.pythonhosted.org/packages/97/6d/c2fd1512873d3f23d24537e97765e7090a00de466516aa442b994b064918/jiter-0.11.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5661469a7b2be25ade3a4bb6c21ffd1e142e13351a0759f264dfdd3ad99af1ab", size = 363751, upload-time = "2025-09-15T09:20:25.12Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/99/48d156c742e75d33b9c8be44b1142d233823be491acdb1009629e4109e6a/jiter-0.11.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:76c15ef0d3d02f8b389066fa4c410a0b89e9cc6468a1f0674c5925d2f3c3e890", size = 488591, upload-time = "2025-09-15T09:20:26.397Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/fd/214452149f63847b791b1f6e9558f59e94674c47418c03e9787236ac8656/jiter-0.11.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:63782a1350917a27817030716566ed3d5b3c731500fd42d483cbd7094e2c5b25", size = 378906, upload-time = "2025-09-15T09:20:27.637Z" },
+    { url = "https://files.pythonhosted.org/packages/de/91/25e38fbbfc17111d7b70b24290a41d611cc2a27fa6cd0ed84ddae38ec3e6/jiter-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5a7092b699646a1ddc03a7b112622d9c066172627c7382659befb0d2996f1659", size = 350288, upload-time = "2025-09-15T09:20:28.951Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/d8/d6d2eefa9f0ff6ac6b725f5164a94f15bb4dee68584b5054043d735803da/jiter-0.11.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f637b8e818f6d75540f350a6011ce21252573c0998ea1b4365ee54b7672c23c5", size = 388442, upload-time = "2025-09-15T09:20:30.223Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/e4/cd7e27852de498d441a575a147ac7a15cf66768ae2cde8c43ea5b464c827/jiter-0.11.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:a624d87719e1b5d09c15286eaee7e1532a40c692a096ea7ca791121365f548c1", size = 518273, upload-time = "2025-09-15T09:20:31.569Z" },
+    { url = "https://files.pythonhosted.org/packages/77/a2/6681a9a503141752b33c92c58512ed8da13849ed3dbf88a3f8aba9bfb255/jiter-0.11.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a9d0146d8d9b3995821bb586fc8256636258947c2f39da5bab709f3a28fb1a0b", size = 510254, upload-time = "2025-09-15T09:20:32.889Z" },
+    { url = "https://files.pythonhosted.org/packages/38/32/df1a06f397074da35cf8fe79ec07da203358a2912b2a6349a0d4a88a1e0a/jiter-0.11.0-cp39-cp39-win32.whl", hash = "sha256:d067655a7cf0831eb8ec3e39cbd752995e9b69a2206df3535b3a067fac23b032", size = 207076, upload-time = "2025-09-15T09:20:34.196Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/91/11bc61fa76fd6197f5baa8576614852ee8586a16c2f25085edc6b47a369d/jiter-0.11.0-cp39-cp39-win_amd64.whl", hash = "sha256:f05d03775a11aaf132c447436983169958439f1219069abf24662a672851f94e", size = 206077, upload-time = "2025-09-15T09:20:35.598Z" },
+    { url = "https://files.pythonhosted.org/packages/70/f3/ce100253c80063a7b8b406e1d1562657fd4b9b4e1b562db40e68645342fb/jiter-0.11.0-graalpy311-graalpy242_311_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:902b43386c04739229076bd1c4c69de5d115553d982ab442a8ae82947c72ede7", size = 336380, upload-time = "2025-09-15T09:20:36.867Z" },
 ]
 
 [[package]]
@@ -1412,14 +1785,14 @@ wheels = [
 
 [[package]]
 name = "jsonschema-specifications"
-version = "2025.4.1"
+version = "2025.9.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "referencing" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/bf/ce/46fbd9c8119cfc3581ee5643ea49464d168028cfb5caff5fc0596d0cf914/jsonschema_specifications-2025.4.1.tar.gz", hash = "sha256:630159c9f4dbea161a6a2205c3011cc4f18ff381b189fff48bb39b9bf26ae608", size = 15513, upload-time = "2025-04-23T12:34:07.418Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/19/74/a633ee74eb36c44aa6d1095e7cc5569bebf04342ee146178e2d36600708b/jsonschema_specifications-2025.9.1.tar.gz", hash = "sha256:b540987f239e745613c7a9176f3edb72b832a4ac465cf02712288397832b5e8d", size = 32855, upload-time = "2025-09-08T01:34:59.186Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/01/0e/b27cdbaccf30b890c40ed1da9fd4a3593a5cf94dae54fb34f8a4b74fcd3f/jsonschema_specifications-2025.4.1-py3-none-any.whl", hash = "sha256:4653bffbd6584f7de83a67e0d620ef16900b390ddc7939d56684d6c81e33f1af", size = 18437, upload-time = "2025-04-23T12:34:05.422Z" },
+    { url = "https://files.pythonhosted.org/packages/41/45/1a4ed80516f02155c51f51e8cedb3c1902296743db0bbc66608a0db2814f/jsonschema_specifications-2025.9.1-py3-none-any.whl", hash = "sha256:98802fee3a11ee76ecaca44429fda8a41bff98b00a0f2838151b113f210cc6fe", size = 18437, upload-time = "2025-09-08T01:34:57.871Z" },
 ]
 
 [[package]]
@@ -1469,12 +1842,12 @@ wheels = [
 ]
 
 [[package]]
-name = "latex2mathml"
-version = "3.78.1"
+name = "lark"
+version = "1.2.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/1a/26/57b1034c08922d0aefea79430a5e0006ffaee4f0ec59d566613f667ab2f7/latex2mathml-3.78.1.tar.gz", hash = "sha256:f941db80bf41db33f31df87b304e8b588f8166b813b0257c11c98f7a9d0aac71", size = 74030, upload-time = "2025-08-29T23:34:23.178Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/af/60/bc7622aefb2aee1c0b4ba23c1446d3e30225c8770b38d7aedbfb65ca9d5a/lark-1.2.2.tar.gz", hash = "sha256:ca807d0162cd16cef15a8feecb862d7319e7a09bdb13aef927968e45040fed80", size = 252132, upload-time = "2024-08-13T19:49:00.652Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/3e/76/d661ea2e529c3d464f9efd73f9ac31626b45279eb4306e684054ea20e3d4/latex2mathml-3.78.1-py3-none-any.whl", hash = "sha256:f089b6d75e85b937f99693c93e8c16c0804008672c3dd2a3d25affd36f238100", size = 73892, upload-time = "2025-08-29T23:34:21.98Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/00/d90b10b962b4277f5e64a78b6609968859ff86889f5b898c1a778c06ec00/lark-1.2.2-py3-none-any.whl", hash = "sha256:c2276486b02f0f1b90be155f2c8ba4a8e194d42775786db622faccd652d8e80c", size = 111036, upload-time = "2024-08-13T19:48:58.603Z" },
 ]
 
 [[package]]
@@ -1509,17 +1882,17 @@ sdist = { url = "https://files.pythonhosted.org/packages/e4/b4/c8cd17629ced0b964
 
 [[package]]
 name = "llguidance"
-version = "1.2.0"
+version = "0.7.30"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/14/d2/e7f7176a04a4fe17277fc0a467477b6f41117f9217c83d8bc84065d452b5/llguidance-1.2.0.tar.gz", hash = "sha256:124aaf546381e76189deb6d0b324cf55163983a84dfa016032ba5b3d5b243290", size = 1069002, upload-time = "2025-08-12T20:07:15.939Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/bf/38/d1ef3ae08d8d857e5e0690c5b1e07bf7eb4a1cae5881d87215826dc6cadb/llguidance-0.7.30.tar.gz", hash = "sha256:e93bf75f2b6e48afb86a5cee23038746975e1654672bf5ba0ae75f7d4d4a2248", size = 1055528, upload-time = "2025-06-23T00:23:49.247Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6c/cb/1f9672f9ed0474781c52334c756137799a219144dd2517956904de51cd4e/llguidance-1.2.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:f6831a232afe79303bf41e4e73f9e715e8fe201bed00c896ddd8e20b3ff245e3", size = 3227696, upload-time = "2025-08-12T20:07:14.424Z" },
-    { url = "https://files.pythonhosted.org/packages/30/e4/f0b3c16d936448efe4334fe0f35c35d914b6a1f25dd1c1e1e1f7a053ead6/llguidance-1.2.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:475b376b52681e96432ba026aa94b736efdd4428618248a041a910c23037edf7", size = 3095669, upload-time = "2025-08-12T20:07:12.553Z" },
-    { url = "https://files.pythonhosted.org/packages/93/a4/af9e3f09b754f0e0ed87746e8409d32f08ffc2aa6a70b7fa04f9393098be/llguidance-1.2.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:52102a6e65d7fcd08214d6392653b813f8c4e15736da46738c304dc3a5ea5ab4", size = 2815374, upload-time = "2025-08-12T20:07:06.984Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/04/34c2766bcda7df428b63e9936bf7b685a99d022884807f9f339204ffc682/llguidance-1.2.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a06356cf3315562fc395c4b15d41ebae0ef6833703db3a4b3c7125447160d05b", size = 3299527, upload-time = "2025-08-12T20:07:09.091Z" },
-    { url = "https://files.pythonhosted.org/packages/00/06/e42e64e0fc1f33877e41a55556322fee17e73efb7b1b58f60d1f40d6b31e/llguidance-1.2.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c144da83cc1927b5742ac9301b6e39c5fa9e04aefbd543fa4b9fc88c28bbb78c", size = 3044367, upload-time = "2025-08-12T20:07:10.891Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/3b/fa95e6ba555b5c652f19e5bf632a90fcbd43865d8722277642873fe63bae/llguidance-1.2.0-cp39-abi3-win32.whl", hash = "sha256:d5da82f54d98e155f8cb24425cc29bba7ee4e0750daacb84f5ea845390c0d5d9", size = 2607705, upload-time = "2025-08-12T20:07:19.676Z" },
-    { url = "https://files.pythonhosted.org/packages/08/69/d41416280d457e787aa77b446269539f11ff902298375fbfc95751c0ee9d/llguidance-1.2.0-cp39-abi3-win_amd64.whl", hash = "sha256:0a985c600f7f0c6b483fdbcf6586bfd45d8412a0115755d37d3089dc4726abf5", size = 2806088, upload-time = "2025-08-12T20:07:18.011Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/e1/694c89986fcae7777184fc8b22baa0976eba15a6847221763f6ad211fc1f/llguidance-0.7.30-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:c80af02c118d2b0526bcecaab389af2ed094537a069b0fc724cd2a2f2ba3990f", size = 3327974, upload-time = "2025-06-23T00:23:47.556Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/77/ab7a548ae189dc23900fdd37803c115c2339b1223af9e8eb1f4329b5935a/llguidance-0.7.30-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:00a256d532911d2cf5ba4ef63e182944e767dd2402f38d63002016bc37755958", size = 3210709, upload-time = "2025-06-23T00:23:45.872Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/5b/6a166564b14f9f805f0ea01ec233a84f55789cb7eeffe1d6224ccd0e6cdd/llguidance-0.7.30-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:af8741c867e4bc7e42f7cdc68350c076b4edd0ca10ecefbde75f15a9f6bc25d0", size = 14867038, upload-time = "2025-06-23T00:23:39.571Z" },
+    { url = "https://files.pythonhosted.org/packages/17/ec/69507bdb36767f9b6ff2e290660a9b5afdda0fb8a7903faa37f37c6c2a72/llguidance-0.7.30-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f4a327a30dd37d86dd6347861ac8de3521fc1dbef9475296c06744e5b40ffc54", size = 15142936, upload-time = "2025-06-23T00:23:41.944Z" },
+    { url = "https://files.pythonhosted.org/packages/af/80/5a40b9689f17612434b820854cba9b8cabd5142072c491b5280fe5f7a35e/llguidance-0.7.30-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9edc409b9decd6cffba5f5bf3b4fbd7541f95daa8cbc9510cbf96c6ab1ffc153", size = 15004926, upload-time = "2025-06-23T00:23:43.965Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/bc/2d2f9b446bb3e51e4dd4db290590afee03ae29163f417168569f0361204c/llguidance-0.7.30-cp39-abi3-win32.whl", hash = "sha256:a0d52b8d1b2d3b0e661e3f953ecccfa16644f302026b3067a4815c1baa2ae643", size = 2585627, upload-time = "2025-06-23T00:23:52.39Z" },
+    { url = "https://files.pythonhosted.org/packages/99/47/58e49a118b514855b245f8a962c6aaf9a5cc95a0f61eac7e230e691c7b7e/llguidance-0.7.30-cp39-abi3-win_amd64.whl", hash = "sha256:05234ecceea7c9c6ff13b9739112043173a3bcb88cae860249b20335a07b3075", size = 2796878, upload-time = "2025-06-23T00:23:51Z" },
 ]
 
 [[package]]
@@ -1581,16 +1954,31 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/e2/3b/a9a17366af80127bd09decbe2a54d8974b6d8b274b39bf47fbaedeec6307/llvmlite-0.44.0-cp312-cp312-win_amd64.whl", hash = "sha256:eae7e2d4ca8f88f89d315b48c6b741dcb925d6a1042da694aa16ab3dd4cbd3a1", size = 30332380, upload-time = "2025-01-20T11:14:02.442Z" },
 ]
 
+[[package]]
+name = "lm-format-enforcer"
+version = "0.11.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "interegular" },
+    { name = "packaging" },
+    { name = "pydantic" },
+    { name = "pyyaml" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/84/d5/41cd417ba7dfdbbcfe46cebf81fb3dfd7c591b89897560ad05bb410a465d/lm_format_enforcer-0.11.3.tar.gz", hash = "sha256:e68081c108719cce284a9bcc889709b26ffb085a1945b5eba3a12cfa96d528da", size = 40258, upload-time = "2025-08-24T19:37:47.527Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a0/ef/11292bb0b85cf4c93447cab5a29f64576ed14d3ab4280e35ddd23486594a/lm_format_enforcer-0.11.3-py3-none-any.whl", hash = "sha256:cf586350875def1ae7a8fba84fcbbfc8371424b6c9d05c1fcba70aa233fbf06f", size = 45418, upload-time = "2025-08-24T19:37:46.325Z" },
+]
+
 [[package]]
 name = "markdown"
-version = "3.8.2"
+version = "3.9"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/d7/c2/4ab49206c17f75cb08d6311171f2d65798988db4360c4d1485bd0eedd67c/markdown-3.8.2.tar.gz", hash = "sha256:247b9a70dd12e27f67431ce62523e675b866d254f900c4fe75ce3dda62237c45", size = 362071, upload-time = "2025-06-19T17:12:44.483Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/8d/37/02347f6d6d8279247a5837082ebc26fc0d5aaeaf75aa013fcbb433c777ab/markdown-3.9.tar.gz", hash = "sha256:d2900fe1782bd33bdbbd56859defef70c2e78fc46668f8eb9df3128138f2cb6a", size = 364585, upload-time = "2025-09-04T20:25:22.885Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/96/2b/34cc11786bc00d0f04d0f5fdc3a2b1ae0b6239eef72d3d345805f9ad92a1/markdown-3.8.2-py3-none-any.whl", hash = "sha256:5c83764dbd4e00bdd94d85a19b8d55ccca20fe35b2e678a1422b380324dd5f24", size = 106827, upload-time = "2025-06-19T17:12:42.994Z" },
+    { url = "https://files.pythonhosted.org/packages/70/ae/44c4a6a4cbb496d93c6257954260fe3a6e91b7bed2240e5dad2a717f5111/markdown-3.9-py3-none-any.whl", hash = "sha256:9f4d91ed810864ea88a6f32c07ba8bee1346c0cc1f6b1f9f6c822f2a9667d280", size = 107441, upload-time = "2025-09-04T20:25:21.784Z" },
 ]
 
 [[package]]
@@ -1625,68 +2013,56 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/94/54/e7d793b573f298e1c9013b8c4dade17d481164aa517d1d7148619c2cedbf/markdown_it_py-4.0.0-py3-none-any.whl", hash = "sha256:87327c59b172c5011896038353a81343b6754500a08cd7a4973bb48c6d578147", size = 87321, upload-time = "2025-08-11T12:57:51.923Z" },
 ]
 
-[[package]]
-name = "markdown2"
-version = "2.5.4"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/42/f8/b2ae8bf5f28f9b510ae097415e6e4cb63226bb28d7ee01aec03a755ba03b/markdown2-2.5.4.tar.gz", hash = "sha256:a09873f0b3c23dbfae589b0080587df52ad75bb09a5fa6559147554736676889", size = 145652, upload-time = "2025-07-27T16:16:24.307Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/b8/06/2697b5043c3ecb720ce0d243fc7cf5024c0b5b1e450506e9b21939019963/markdown2-2.5.4-py3-none-any.whl", hash = "sha256:3c4b2934e677be7fec0e6f2de4410e116681f4ad50ec8e5ba7557be506d3f439", size = 49954, upload-time = "2025-07-27T16:16:23.026Z" },
-]
-
-[package.optional-dependencies]
-all = [
-    { name = "latex2mathml" },
-    { name = "pygments" },
-    { name = "wavedrom" },
-]
-
 [[package]]
 name = "markupsafe"
-version = "3.0.2"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/b2/97/5d42485e71dfc078108a86d6de8fa46db44a1a9295e89c5d6d4a06e23a62/markupsafe-3.0.2.tar.gz", hash = "sha256:ee55d3edf80167e48ea11a923c7386f4669df67d7994554387f84e7d8b0a2bf0", size = 20537, upload-time = "2024-10-18T15:21:54.129Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/04/90/d08277ce111dd22f77149fd1a5d4653eeb3b3eaacbdfcbae5afb2600eebd/MarkupSafe-3.0.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7e94c425039cde14257288fd61dcfb01963e658efbc0ff54f5306b06054700f8", size = 14357, upload-time = "2024-10-18T15:20:51.44Z" },
-    { url = "https://files.pythonhosted.org/packages/04/e1/6e2194baeae0bca1fae6629dc0cbbb968d4d941469cbab11a3872edff374/MarkupSafe-3.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:9e2d922824181480953426608b81967de705c3cef4d1af983af849d7bd619158", size = 12393, upload-time = "2024-10-18T15:20:52.426Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/69/35fa85a8ece0a437493dc61ce0bb6d459dcba482c34197e3efc829aa357f/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:38a9ef736c01fccdd6600705b09dc574584b89bea478200c5fbf112a6b0d5579", size = 21732, upload-time = "2024-10-18T15:20:53.578Z" },
-    { url = "https://files.pythonhosted.org/packages/22/35/137da042dfb4720b638d2937c38a9c2df83fe32d20e8c8f3185dbfef05f7/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bbcb445fa71794da8f178f0f6d66789a28d7319071af7a496d4d507ed566270d", size = 20866, upload-time = "2024-10-18T15:20:55.06Z" },
-    { url = "https://files.pythonhosted.org/packages/29/28/6d029a903727a1b62edb51863232152fd335d602def598dade38996887f0/MarkupSafe-3.0.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:57cb5a3cf367aeb1d316576250f65edec5bb3be939e9247ae594b4bcbc317dfb", size = 20964, upload-time = "2024-10-18T15:20:55.906Z" },
-    { url = "https://files.pythonhosted.org/packages/cc/cd/07438f95f83e8bc028279909d9c9bd39e24149b0d60053a97b2bc4f8aa51/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:3809ede931876f5b2ec92eef964286840ed3540dadf803dd570c3b7e13141a3b", size = 21977, upload-time = "2024-10-18T15:20:57.189Z" },
-    { url = "https://files.pythonhosted.org/packages/29/01/84b57395b4cc062f9c4c55ce0df7d3108ca32397299d9df00fedd9117d3d/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e07c3764494e3776c602c1e78e298937c3315ccc9043ead7e685b7f2b8d47b3c", size = 21366, upload-time = "2024-10-18T15:20:58.235Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/6e/61ebf08d8940553afff20d1fb1ba7294b6f8d279df9fd0c0db911b4bbcfd/MarkupSafe-3.0.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b424c77b206d63d500bcb69fa55ed8d0e6a3774056bdc4839fc9298a7edca171", size = 21091, upload-time = "2024-10-18T15:20:59.235Z" },
-    { url = "https://files.pythonhosted.org/packages/11/23/ffbf53694e8c94ebd1e7e491de185124277964344733c45481f32ede2499/MarkupSafe-3.0.2-cp310-cp310-win32.whl", hash = "sha256:fcabf5ff6eea076f859677f5f0b6b5c1a51e70a376b0579e0eadef8db48c6b50", size = 15065, upload-time = "2024-10-18T15:21:00.307Z" },
-    { url = "https://files.pythonhosted.org/packages/44/06/e7175d06dd6e9172d4a69a72592cb3f7a996a9c396eee29082826449bbc3/MarkupSafe-3.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:6af100e168aa82a50e186c82875a5893c5597a0c1ccdb0d8b40240b1f28b969a", size = 15514, upload-time = "2024-10-18T15:21:01.122Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/28/bbf83e3f76936960b850435576dd5e67034e200469571be53f69174a2dfd/MarkupSafe-3.0.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:9025b4018f3a1314059769c7bf15441064b2207cb3f065e6ea1e7359cb46db9d", size = 14353, upload-time = "2024-10-18T15:21:02.187Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/30/316d194b093cde57d448a4c3209f22e3046c5bb2fb0820b118292b334be7/MarkupSafe-3.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:93335ca3812df2f366e80509ae119189886b0f3c2b81325d39efdb84a1e2ae93", size = 12392, upload-time = "2024-10-18T15:21:02.941Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/96/9cdafba8445d3a53cae530aaf83c38ec64c4d5427d975c974084af5bc5d2/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2cb8438c3cbb25e220c2ab33bb226559e7afb3baec11c4f218ffa7308603c832", size = 23984, upload-time = "2024-10-18T15:21:03.953Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/a4/aefb044a2cd8d7334c8a47d3fb2c9f328ac48cb349468cc31c20b539305f/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a123e330ef0853c6e822384873bef7507557d8e4a082961e1defa947aa59ba84", size = 23120, upload-time = "2024-10-18T15:21:06.495Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/21/5e4851379f88f3fad1de30361db501300d4f07bcad047d3cb0449fc51f8c/MarkupSafe-3.0.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1e084f686b92e5b83186b07e8a17fc09e38fff551f3602b249881fec658d3eca", size = 23032, upload-time = "2024-10-18T15:21:07.295Z" },
-    { url = "https://files.pythonhosted.org/packages/00/7b/e92c64e079b2d0d7ddf69899c98842f3f9a60a1ae72657c89ce2655c999d/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d8213e09c917a951de9d09ecee036d5c7d36cb6cb7dbaece4c71a60d79fb9798", size = 24057, upload-time = "2024-10-18T15:21:08.073Z" },
-    { url = "https://files.pythonhosted.org/packages/f9/ac/46f960ca323037caa0a10662ef97d0a4728e890334fc156b9f9e52bcc4ca/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:5b02fb34468b6aaa40dfc198d813a641e3a63b98c2b05a16b9f80b7ec314185e", size = 23359, upload-time = "2024-10-18T15:21:09.318Z" },
-    { url = "https://files.pythonhosted.org/packages/69/84/83439e16197337b8b14b6a5b9c2105fff81d42c2a7c5b58ac7b62ee2c3b1/MarkupSafe-3.0.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0bff5e0ae4ef2e1ae4fdf2dfd5b76c75e5c2fa4132d05fc1b0dabcd20c7e28c4", size = 23306, upload-time = "2024-10-18T15:21:10.185Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/34/a15aa69f01e2181ed8d2b685c0d2f6655d5cca2c4db0ddea775e631918cd/MarkupSafe-3.0.2-cp311-cp311-win32.whl", hash = "sha256:6c89876f41da747c8d3677a2b540fb32ef5715f97b66eeb0c6b66f5e3ef6f59d", size = 15094, upload-time = "2024-10-18T15:21:11.005Z" },
-    { url = "https://files.pythonhosted.org/packages/da/b8/3a3bd761922d416f3dc5d00bfbed11f66b1ab89a0c2b6e887240a30b0f6b/MarkupSafe-3.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:70a87b411535ccad5ef2f1df5136506a10775d267e197e4cf531ced10537bd6b", size = 15521, upload-time = "2024-10-18T15:21:12.911Z" },
-    { url = "https://files.pythonhosted.org/packages/22/09/d1f21434c97fc42f09d290cbb6350d44eb12f09cc62c9476effdb33a18aa/MarkupSafe-3.0.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:9778bd8ab0a994ebf6f84c2b949e65736d5575320a17ae8984a77fab08db94cf", size = 14274, upload-time = "2024-10-18T15:21:13.777Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/b0/18f76bba336fa5aecf79d45dcd6c806c280ec44538b3c13671d49099fdd0/MarkupSafe-3.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:846ade7b71e3536c4e56b386c2a47adf5741d2d8b94ec9dc3e92e5e1ee1e2225", size = 12348, upload-time = "2024-10-18T15:21:14.822Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/25/dd5c0f6ac1311e9b40f4af06c78efde0f3b5cbf02502f8ef9501294c425b/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1c99d261bd2d5f6b59325c92c73df481e05e57f19837bdca8413b9eac4bd8028", size = 24149, upload-time = "2024-10-18T15:21:15.642Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/f0/89e7aadfb3749d0f52234a0c8c7867877876e0a20b60e2188e9850794c17/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e17c96c14e19278594aa4841ec148115f9c7615a47382ecb6b82bd8fea3ab0c8", size = 23118, upload-time = "2024-10-18T15:21:17.133Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/da/f2eeb64c723f5e3777bc081da884b414671982008c47dcc1873d81f625b6/MarkupSafe-3.0.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:88416bd1e65dcea10bc7569faacb2c20ce071dd1f87539ca2ab364bf6231393c", size = 22993, upload-time = "2024-10-18T15:21:18.064Z" },
-    { url = "https://files.pythonhosted.org/packages/da/0e/1f32af846df486dce7c227fe0f2398dc7e2e51d4a370508281f3c1c5cddc/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2181e67807fc2fa785d0592dc2d6206c019b9502410671cc905d132a92866557", size = 24178, upload-time = "2024-10-18T15:21:18.859Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/f6/bb3ca0532de8086cbff5f06d137064c8410d10779c4c127e0e47d17c0b71/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:52305740fe773d09cffb16f8ed0427942901f00adedac82ec8b67752f58a1b22", size = 23319, upload-time = "2024-10-18T15:21:19.671Z" },
-    { url = "https://files.pythonhosted.org/packages/a2/82/8be4c96ffee03c5b4a034e60a31294daf481e12c7c43ab8e34a1453ee48b/MarkupSafe-3.0.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ad10d3ded218f1039f11a75f8091880239651b52e9bb592ca27de44eed242a48", size = 23352, upload-time = "2024-10-18T15:21:20.971Z" },
-    { url = "https://files.pythonhosted.org/packages/51/ae/97827349d3fcffee7e184bdf7f41cd6b88d9919c80f0263ba7acd1bbcb18/MarkupSafe-3.0.2-cp312-cp312-win32.whl", hash = "sha256:0f4ca02bea9a23221c0182836703cbf8930c5e9454bacce27e767509fa286a30", size = 15097, upload-time = "2024-10-18T15:21:22.646Z" },
-    { url = "https://files.pythonhosted.org/packages/c1/80/a61f99dc3a936413c3ee4e1eecac96c0da5ed07ad56fd975f1a9da5bc630/MarkupSafe-3.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:8e06879fc22a25ca47312fbe7c8264eb0b662f6db27cb2d3bbbc74b1df4b9b87", size = 15601, upload-time = "2024-10-18T15:21:23.499Z" },
-    { url = "https://files.pythonhosted.org/packages/a7/ea/9b1530c3fdeeca613faeb0fb5cbcf2389d816072fab72a71b45749ef6062/MarkupSafe-3.0.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:eaa0a10b7f72326f1372a713e73c3f739b524b3af41feb43e4921cb529f5929a", size = 14344, upload-time = "2024-10-18T15:21:43.721Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/c2/fbdbfe48848e7112ab05e627e718e854d20192b674952d9042ebd8c9e5de/MarkupSafe-3.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:48032821bbdf20f5799ff537c7ac3d1fba0ba032cfc06194faffa8cda8b560ff", size = 12389, upload-time = "2024-10-18T15:21:44.666Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/25/7a7c6e4dbd4f867d95d94ca15449e91e52856f6ed1905d58ef1de5e211d0/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1a9d3f5f0901fdec14d8d2f66ef7d035f2157240a433441719ac9a3fba440b13", size = 21607, upload-time = "2024-10-18T15:21:45.452Z" },
-    { url = "https://files.pythonhosted.org/packages/53/8f/f339c98a178f3c1e545622206b40986a4c3307fe39f70ccd3d9df9a9e425/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88b49a3b9ff31e19998750c38e030fc7bb937398b1f78cfa599aaef92d693144", size = 20728, upload-time = "2024-10-18T15:21:46.295Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/03/8496a1a78308456dbd50b23a385c69b41f2e9661c67ea1329849a598a8f9/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:cfad01eed2c2e0c01fd0ecd2ef42c492f7f93902e39a42fc9ee1692961443a29", size = 20826, upload-time = "2024-10-18T15:21:47.134Z" },
-    { url = "https://files.pythonhosted.org/packages/e6/cf/0a490a4bd363048c3022f2f475c8c05582179bb179defcee4766fb3dcc18/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1225beacc926f536dc82e45f8a4d68502949dc67eea90eab715dea3a21c1b5f0", size = 21843, upload-time = "2024-10-18T15:21:48.334Z" },
-    { url = "https://files.pythonhosted.org/packages/19/a3/34187a78613920dfd3cdf68ef6ce5e99c4f3417f035694074beb8848cd77/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:3169b1eefae027567d1ce6ee7cae382c57fe26e82775f460f0b2778beaad66c0", size = 21219, upload-time = "2024-10-18T15:21:49.587Z" },
-    { url = "https://files.pythonhosted.org/packages/17/d8/5811082f85bb88410ad7e452263af048d685669bbbfb7b595e8689152498/MarkupSafe-3.0.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:eb7972a85c54febfb25b5c4b4f3af4dcc731994c7da0d8a0b4a6eb0640e1d178", size = 20946, upload-time = "2024-10-18T15:21:50.441Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/31/bd635fb5989440d9365c5e3c47556cfea121c7803f5034ac843e8f37c2f2/MarkupSafe-3.0.2-cp39-cp39-win32.whl", hash = "sha256:8c4e8c3ce11e1f92f6536ff07154f9d49677ebaaafc32db9db4620bc11ed480f", size = 15063, upload-time = "2024-10-18T15:21:51.385Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/73/085399401383ce949f727afec55ec3abd76648d04b9f22e1c0e99cb4bec3/MarkupSafe-3.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:6e296a513ca3d94054c2c881cc913116e90fd030ad1c656b3869762b754f5f8a", size = 15506, upload-time = "2024-10-18T15:21:52.974Z" },
+version = "3.0.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/7e/99/7690b6d4034fffd95959cbe0c02de8deb3098cc577c67bb6a24fe5d7caa7/markupsafe-3.0.3.tar.gz", hash = "sha256:722695808f4b6457b320fdc131280796bdceb04ab50fe1795cd540799ebe1698", size = 80313, upload-time = "2025-09-27T18:37:40.426Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e8/4b/3541d44f3937ba468b75da9eebcae497dcf67adb65caa16760b0a6807ebb/markupsafe-3.0.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2f981d352f04553a7171b8e44369f2af4055f888dfb147d55e42d29e29e74559", size = 11631, upload-time = "2025-09-27T18:36:05.558Z" },
+    { url = "https://files.pythonhosted.org/packages/98/1b/fbd8eed11021cabd9226c37342fa6ca4e8a98d8188a8d9b66740494960e4/markupsafe-3.0.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e1c1493fb6e50ab01d20a22826e57520f1284df32f2d8601fdd90b6304601419", size = 12057, upload-time = "2025-09-27T18:36:07.165Z" },
+    { url = "https://files.pythonhosted.org/packages/40/01/e560d658dc0bb8ab762670ece35281dec7b6c1b33f5fbc09ebb57a185519/markupsafe-3.0.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1ba88449deb3de88bd40044603fafffb7bc2b055d626a330323a9ed736661695", size = 22050, upload-time = "2025-09-27T18:36:08.005Z" },
+    { url = "https://files.pythonhosted.org/packages/af/cd/ce6e848bbf2c32314c9b237839119c5a564a59725b53157c856e90937b7a/markupsafe-3.0.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f42d0984e947b8adf7dd6dde396e720934d12c506ce84eea8476409563607591", size = 20681, upload-time = "2025-09-27T18:36:08.881Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/2a/b5c12c809f1c3045c4d580b035a743d12fcde53cf685dbc44660826308da/markupsafe-3.0.3-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:c0c0b3ade1c0b13b936d7970b1d37a57acde9199dc2aecc4c336773e1d86049c", size = 20705, upload-time = "2025-09-27T18:36:10.131Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/e3/9427a68c82728d0a88c50f890d0fc072a1484de2f3ac1ad0bfc1a7214fd5/markupsafe-3.0.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:0303439a41979d9e74d18ff5e2dd8c43ed6c6001fd40e5bf2e43f7bd9bbc523f", size = 21524, upload-time = "2025-09-27T18:36:11.324Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/36/23578f29e9e582a4d0278e009b38081dbe363c5e7165113fad546918a232/markupsafe-3.0.3-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:d2ee202e79d8ed691ceebae8e0486bd9a2cd4794cec4824e1c99b6f5009502f6", size = 20282, upload-time = "2025-09-27T18:36:12.573Z" },
+    { url = "https://files.pythonhosted.org/packages/56/21/dca11354e756ebd03e036bd8ad58d6d7168c80ce1fe5e75218e4945cbab7/markupsafe-3.0.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:177b5253b2834fe3678cb4a5f0059808258584c559193998be2601324fdeafb1", size = 20745, upload-time = "2025-09-27T18:36:13.504Z" },
+    { url = "https://files.pythonhosted.org/packages/87/99/faba9369a7ad6e4d10b6a5fbf71fa2a188fe4a593b15f0963b73859a1bbd/markupsafe-3.0.3-cp310-cp310-win32.whl", hash = "sha256:2a15a08b17dd94c53a1da0438822d70ebcd13f8c3a95abe3a9ef9f11a94830aa", size = 14571, upload-time = "2025-09-27T18:36:14.779Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/25/55dc3ab959917602c96985cb1253efaa4ff42f71194bddeb61eb7278b8be/markupsafe-3.0.3-cp310-cp310-win_amd64.whl", hash = "sha256:c4ffb7ebf07cfe8931028e3e4c85f0357459a3f9f9490886198848f4fa002ec8", size = 15056, upload-time = "2025-09-27T18:36:16.125Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/9e/0a02226640c255d1da0b8d12e24ac2aa6734da68bff14c05dd53b94a0fc3/markupsafe-3.0.3-cp310-cp310-win_arm64.whl", hash = "sha256:e2103a929dfa2fcaf9bb4e7c091983a49c9ac3b19c9061b6d5427dd7d14d81a1", size = 13932, upload-time = "2025-09-27T18:36:17.311Z" },
+    { url = "https://files.pythonhosted.org/packages/08/db/fefacb2136439fc8dd20e797950e749aa1f4997ed584c62cfb8ef7c2be0e/markupsafe-3.0.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:1cc7ea17a6824959616c525620e387f6dd30fec8cb44f649e31712db02123dad", size = 11631, upload-time = "2025-09-27T18:36:18.185Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/2e/5898933336b61975ce9dc04decbc0a7f2fee78c30353c5efba7f2d6ff27a/markupsafe-3.0.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4bd4cd07944443f5a265608cc6aab442e4f74dff8088b0dfc8238647b8f6ae9a", size = 12058, upload-time = "2025-09-27T18:36:19.444Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/09/adf2df3699d87d1d8184038df46a9c80d78c0148492323f4693df54e17bb/markupsafe-3.0.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6b5420a1d9450023228968e7e6a9ce57f65d148ab56d2313fcd589eee96a7a50", size = 24287, upload-time = "2025-09-27T18:36:20.768Z" },
+    { url = "https://files.pythonhosted.org/packages/30/ac/0273f6fcb5f42e314c6d8cd99effae6a5354604d461b8d392b5ec9530a54/markupsafe-3.0.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0bf2a864d67e76e5c9a34dc26ec616a66b9888e25e7b9460e1c76d3293bd9dbf", size = 22940, upload-time = "2025-09-27T18:36:22.249Z" },
+    { url = "https://files.pythonhosted.org/packages/19/ae/31c1be199ef767124c042c6c3e904da327a2f7f0cd63a0337e1eca2967a8/markupsafe-3.0.3-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:bc51efed119bc9cfdf792cdeaa4d67e8f6fcccab66ed4bfdd6bde3e59bfcbb2f", size = 21887, upload-time = "2025-09-27T18:36:23.535Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/76/7edcab99d5349a4532a459e1fe64f0b0467a3365056ae550d3bcf3f79e1e/markupsafe-3.0.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:068f375c472b3e7acbe2d5318dea141359e6900156b5b2ba06a30b169086b91a", size = 23692, upload-time = "2025-09-27T18:36:24.823Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/28/6e74cdd26d7514849143d69f0bf2399f929c37dc2b31e6829fd2045b2765/markupsafe-3.0.3-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:7be7b61bb172e1ed687f1754f8e7484f1c8019780f6f6b0786e76bb01c2ae115", size = 21471, upload-time = "2025-09-27T18:36:25.95Z" },
+    { url = "https://files.pythonhosted.org/packages/62/7e/a145f36a5c2945673e590850a6f8014318d5577ed7e5920a4b3448e0865d/markupsafe-3.0.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:f9e130248f4462aaa8e2552d547f36ddadbeaa573879158d721bbd33dfe4743a", size = 22923, upload-time = "2025-09-27T18:36:27.109Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/62/d9c46a7f5c9adbeeeda52f5b8d802e1094e9717705a645efc71b0913a0a8/markupsafe-3.0.3-cp311-cp311-win32.whl", hash = "sha256:0db14f5dafddbb6d9208827849fad01f1a2609380add406671a26386cdf15a19", size = 14572, upload-time = "2025-09-27T18:36:28.045Z" },
+    { url = "https://files.pythonhosted.org/packages/83/8a/4414c03d3f891739326e1783338e48fb49781cc915b2e0ee052aa490d586/markupsafe-3.0.3-cp311-cp311-win_amd64.whl", hash = "sha256:de8a88e63464af587c950061a5e6a67d3632e36df62b986892331d4620a35c01", size = 15077, upload-time = "2025-09-27T18:36:29.025Z" },
+    { url = "https://files.pythonhosted.org/packages/35/73/893072b42e6862f319b5207adc9ae06070f095b358655f077f69a35601f0/markupsafe-3.0.3-cp311-cp311-win_arm64.whl", hash = "sha256:3b562dd9e9ea93f13d53989d23a7e775fdfd1066c33494ff43f5418bc8c58a5c", size = 13876, upload-time = "2025-09-27T18:36:29.954Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/72/147da192e38635ada20e0a2e1a51cf8823d2119ce8883f7053879c2199b5/markupsafe-3.0.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:d53197da72cc091b024dd97249dfc7794d6a56530370992a5e1a08983ad9230e", size = 11615, upload-time = "2025-09-27T18:36:30.854Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/81/7e4e08678a1f98521201c3079f77db69fb552acd56067661f8c2f534a718/markupsafe-3.0.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1872df69a4de6aead3491198eaf13810b565bdbeec3ae2dc8780f14458ec73ce", size = 12020, upload-time = "2025-09-27T18:36:31.971Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/2c/799f4742efc39633a1b54a92eec4082e4f815314869865d876824c257c1e/markupsafe-3.0.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3a7e8ae81ae39e62a41ec302f972ba6ae23a5c5396c8e60113e9066ef893da0d", size = 24332, upload-time = "2025-09-27T18:36:32.813Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/2e/8d0c2ab90a8c1d9a24f0399058ab8519a3279d1bd4289511d74e909f060e/markupsafe-3.0.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d6dd0be5b5b189d31db7cda48b91d7e0a9795f31430b7f271219ab30f1d3ac9d", size = 22947, upload-time = "2025-09-27T18:36:33.86Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/54/887f3092a85238093a0b2154bd629c89444f395618842e8b0c41783898ea/markupsafe-3.0.3-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:94c6f0bb423f739146aec64595853541634bde58b2135f27f61c1ffd1cd4d16a", size = 21962, upload-time = "2025-09-27T18:36:35.099Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/2f/336b8c7b6f4a4d95e91119dc8521402461b74a485558d8f238a68312f11c/markupsafe-3.0.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:be8813b57049a7dc738189df53d69395eba14fb99345e0a5994914a3864c8a4b", size = 23760, upload-time = "2025-09-27T18:36:36.001Z" },
+    { url = "https://files.pythonhosted.org/packages/32/43/67935f2b7e4982ffb50a4d169b724d74b62a3964bc1a9a527f5ac4f1ee2b/markupsafe-3.0.3-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:83891d0e9fb81a825d9a6d61e3f07550ca70a076484292a70fde82c4b807286f", size = 21529, upload-time = "2025-09-27T18:36:36.906Z" },
+    { url = "https://files.pythonhosted.org/packages/89/e0/4486f11e51bbba8b0c041098859e869e304d1c261e59244baa3d295d47b7/markupsafe-3.0.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:77f0643abe7495da77fb436f50f8dab76dbc6e5fd25d39589a0f1fe6548bfa2b", size = 23015, upload-time = "2025-09-27T18:36:37.868Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/e1/78ee7a023dac597a5825441ebd17170785a9dab23de95d2c7508ade94e0e/markupsafe-3.0.3-cp312-cp312-win32.whl", hash = "sha256:d88b440e37a16e651bda4c7c2b930eb586fd15ca7406cb39e211fcff3bf3017d", size = 14540, upload-time = "2025-09-27T18:36:38.761Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/5b/bec5aa9bbbb2c946ca2733ef9c4ca91c91b6a24580193e891b5f7dbe8e1e/markupsafe-3.0.3-cp312-cp312-win_amd64.whl", hash = "sha256:26a5784ded40c9e318cfc2bdb30fe164bdb8665ded9cd64d500a34fb42067b1c", size = 15105, upload-time = "2025-09-27T18:36:39.701Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/f1/216fc1bbfd74011693a4fd837e7026152e89c4bcf3e77b6692fba9923123/markupsafe-3.0.3-cp312-cp312-win_arm64.whl", hash = "sha256:35add3b638a5d900e807944a078b51922212fb3dedb01633a8defc4b01a3c85f", size = 13906, upload-time = "2025-09-27T18:36:40.689Z" },
+    { url = "https://files.pythonhosted.org/packages/56/23/0d8c13a44bde9154821586520840643467aee574d8ce79a17da539ee7fed/markupsafe-3.0.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:15d939a21d546304880945ca1ecb8a039db6b4dc49b2c5a400387cdae6a62e26", size = 11623, upload-time = "2025-09-27T18:37:29.296Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/23/07a2cb9a8045d5f3f0890a8c3bc0859d7a47bfd9a560b563899bec7b72ed/markupsafe-3.0.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:f71a396b3bf33ecaa1626c255855702aca4d3d9fea5e051b41ac59a9c1c41edc", size = 12049, upload-time = "2025-09-27T18:37:30.234Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/e4/6be85eb81503f8e11b61c0b6369b6e077dcf0a74adbd9ebf6b349937b4e9/markupsafe-3.0.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0f4b68347f8c5eab4a13419215bdfd7f8c9b19f2b25520968adfad23eb0ce60c", size = 21923, upload-time = "2025-09-27T18:37:31.177Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/bc/4dc914ead3fe6ddaef035341fee0fc956949bbd27335b611829292b89ee2/markupsafe-3.0.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e8fc20152abba6b83724d7ff268c249fa196d8259ff481f3b1476383f8f24e42", size = 20543, upload-time = "2025-09-27T18:37:32.168Z" },
+    { url = "https://files.pythonhosted.org/packages/89/6e/5fe81fbcfba4aef4093d5f856e5c774ec2057946052d18d168219b7bd9f9/markupsafe-3.0.3-cp39-cp39-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:949b8d66bc381ee8b007cd945914c721d9aba8e27f71959d750a46f7c282b20b", size = 20585, upload-time = "2025-09-27T18:37:33.166Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/f6/e0e5a3d3ae9c4020f696cd055f940ef86b64fe88de26f3a0308b9d3d048c/markupsafe-3.0.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:3537e01efc9d4dccdf77221fb1cb3b8e1a38d5428920e0657ce299b20324d758", size = 21387, upload-time = "2025-09-27T18:37:34.185Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/25/651753ef4dea08ea790f4fbb65146a9a44a014986996ca40102e237aa49a/markupsafe-3.0.3-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:591ae9f2a647529ca990bc681daebdd52c8791ff06c2bfa05b65163e28102ef2", size = 20133, upload-time = "2025-09-27T18:37:35.138Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/0a/c3cf2b4fef5f0426e8a6d7fce3cb966a17817c568ce59d76b92a233fdbec/markupsafe-3.0.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:a320721ab5a1aba0a233739394eb907f8c8da5c98c9181d1161e77a0c8e36f2d", size = 20588, upload-time = "2025-09-27T18:37:36.096Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/1b/a7782984844bd519ad4ffdbebbba2671ec5d0ebbeac34736c15fb86399e8/markupsafe-3.0.3-cp39-cp39-win32.whl", hash = "sha256:df2449253ef108a379b8b5d6b43f4b1a8e81a061d6537becd5582fba5f9196d7", size = 14566, upload-time = "2025-09-27T18:37:37.09Z" },
+    { url = "https://files.pythonhosted.org/packages/18/1f/8d9c20e1c9440e215a44be5ab64359e207fcb4f675543f1cf9a2a7f648d0/markupsafe-3.0.3-cp39-cp39-win_amd64.whl", hash = "sha256:7c3fb7d25180895632e5d3148dbdc29ea38ccb7fd210aa27acbd1201a1902c6e", size = 15053, upload-time = "2025-09-27T18:37:38.054Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/d3/fe08482b5cd995033556d45041a4f4e76e7f0521112a9c9991d40d39825f/markupsafe-3.0.3-cp39-cp39-win_arm64.whl", hash = "sha256:38664109c14ffc9e7437e86b4dceb442b0096dfe3541d7864d9cbe1da4cf36c8", size = 13928, upload-time = "2025-09-27T18:37:39.037Z" },
 ]
 
 [[package]]
@@ -1707,13 +2083,60 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/2c/19/04f9b178c2d8a15b076c8b5140708fa6ffc5601fb6f1e975537072df5b2a/mergedeep-1.3.4-py3-none-any.whl", hash = "sha256:70775750742b25c0d8f36c55aed03d24c3384d17c951b3175d898bd778ef0307", size = 6354, upload-time = "2021-02-05T18:55:29.583Z" },
 ]
 
+[[package]]
+name = "mistral-common"
+version = "1.8.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "jsonschema" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pillow" },
+    { name = "pydantic" },
+    { name = "pydantic-extra-types", extra = ["pycountry"] },
+    { name = "requests" },
+    { name = "tiktoken" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ce/ff/1992a00ccc936f2c6e69ecb1f2cac678e0fd46c53c71bdab99eda4f89dfd/mistral_common-1.8.5.tar.gz", hash = "sha256:9f6204ede9c807f09040a208a9381ae78ef93e2e5a9cd5202dc12e712a025de8", size = 6331923, upload-time = "2025-09-12T06:43:01.937Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/91/4a/54e19c5e75939fd9418c7b806c21d12cf252ea2ba38f122b597272b459dd/mistral_common-1.8.5-py3-none-any.whl", hash = "sha256:f3cf87b61958a00485e603f3fe0530eb509d7e9b2f7178329dcd260e307eced1", size = 6515140, upload-time = "2025-09-12T06:42:59.622Z" },
+]
+
+[package.optional-dependencies]
+audio = [
+    { name = "soundfile" },
+    { name = "soxr" },
+]
+image = [
+    { name = "opencv-python-headless" },
+]
+
+[[package]]
+name = "mistralai"
+version = "1.9.11"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "eval-type-backport" },
+    { name = "httpx" },
+    { name = "invoke" },
+    { name = "pydantic" },
+    { name = "python-dateutil" },
+    { name = "pyyaml" },
+    { name = "typing-inspection" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/5a/8d/d8b7af67a966b6f227024e1cb7287fc19901a434f87a5a391dcfe635d338/mistralai-1.9.11.tar.gz", hash = "sha256:3df9e403c31a756ec79e78df25ee73cea3eb15f86693773e16b16adaf59c9b8a", size = 208051, upload-time = "2025-10-02T15:53:40.473Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fe/76/4ce12563aea5a76016f8643eff30ab731e6656c845e9e4d090ef10c7b925/mistralai-1.9.11-py3-none-any.whl", hash = "sha256:7a3dc2b8ef3fceaa3582220234261b5c4e3e03a972563b07afa150e44a25a6d3", size = 442796, upload-time = "2025-10-02T15:53:39.134Z" },
+]
+
 [[package]]
 name = "mkdocs"
 version = "1.6.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "colorama", marker = "sys_platform == 'win32'" },
     { name = "ghp-import" },
     { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
@@ -1792,33 +2215,33 @@ wheels = [
 
 [[package]]
 name = "mlx"
-version = "0.29.0"
+version = "0.29.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "mlx-metal", marker = "sys_platform == 'darwin'" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/cb/13/da859e935a0f4179c4e599afe9f4add8d594e0433ac131460746f6a7cbdf/mlx-0.29.0-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:430b85d5444247b80417fd3a2b966676af43549f0926095bac25866b2154c9f0", size = 545687, upload-time = "2025-08-29T17:13:54.532Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/77/3c6d307c62cec555c2c6e81a978b41c3db9d1244653578df6159c7872ef8/mlx-0.29.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:afe33a3feefc62622bc71dd10c4e7e75bc8278780351fd82f1c33145b072d88b", size = 545686, upload-time = "2025-08-29T17:14:20.195Z" },
-    { url = "https://files.pythonhosted.org/packages/11/42/90b5d0d52ed733b57de7fbde2d972ecd20b819553a36bd8a4b78750539d6/mlx-0.29.0-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:4e055878e54afb6ac258c145674d59682c8c565e3fd1dde5e28ca3ebc52d3434", size = 545689, upload-time = "2025-08-29T17:14:14.423Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/03/d5186edf170e7866c4bf37baa50c8e0fc26b6d8668f9fa79f34d23e66ed6/mlx-0.29.0-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:03a738104b50c38e939ba8037b6a04ee937ebd4e77a9792f1f922d021cad0010", size = 650149, upload-time = "2025-08-29T17:19:36.852Z" },
-    { url = "https://files.pythonhosted.org/packages/00/b3/490ae792034a95895a70d04feadb1158786d96605a45555c1cae0a1158d3/mlx-0.29.0-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:96d49e76a152a202c146ce4b71538b59c9eb6c98b40557ee26a3cf9753b5fb34", size = 545948, upload-time = "2025-08-29T17:14:52.863Z" },
-    { url = "https://files.pythonhosted.org/packages/57/00/277ba0f1f384275d122b0af4ebc5866a452689e3b0144d03b46a549a7142/mlx-0.29.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:05c390811f6514013535b5fea9fa33b63daa8cdc17881deadca7fbca38e42c93", size = 545950, upload-time = "2025-08-29T17:14:35.294Z" },
-    { url = "https://files.pythonhosted.org/packages/de/a3/337461b75168cf701a1190eabd331de4fb752dc9dc5759fbede76f744aab/mlx-0.29.0-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:7800f87fc5eb9861c6444b0bb38175c8f0b12d48746a980264fe744f6eb53a50", size = 545946, upload-time = "2025-08-29T17:13:45.738Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/57/d1d0cdd5d194ec8b7b55062407d8e04bcdf62de73c2ef0db075fba0ce7f3/mlx-0.29.0-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:6f2f969c6ad6259f4422c7cd7bfcb49705679585991109220de0b758feaedc25", size = 650033, upload-time = "2025-08-29T17:18:23.646Z" },
-    { url = "https://files.pythonhosted.org/packages/60/c0/a95f24f4b78fdf96f57aaa1a8919f5688603fd4fa0294c2f868f60bc98af/mlx-0.29.0-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:a9f9f760a179d96fa7fecc0b31a9885a9ab4ce9e2914f7f34c2980edd7f012fa", size = 546443, upload-time = "2025-08-29T17:14:17.997Z" },
-    { url = "https://files.pythonhosted.org/packages/61/c7/9f6a5cf1a0b0eb1ace3892c930f64fa84351cecc75221ce4ca5abd0ad64d/mlx-0.29.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:446ea784a4717c9b06eee9c91ee038245ebb231760a2cf61b677dc3d256cbd83", size = 546442, upload-time = "2025-08-29T17:14:19.035Z" },
-    { url = "https://files.pythonhosted.org/packages/85/40/32b1fe76d80f559ff3af18d104a22c4d07ecad62a88e3c313083722868e4/mlx-0.29.0-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:593a3d71c5859f83bc55a9bdc612fe5f8b495c14112e0e7a96c027c11ce5ab52", size = 546444, upload-time = "2025-08-29T17:14:58.946Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/dd/5e1c4672a551f475b5305e3e21da808aacb080a114af83a38e5b9551c212/mlx-0.29.0-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:a1174dd755da6a443bd34616b6ef6bf70718648420b9efafb37d1b20d70877d0", size = 645004, upload-time = "2025-08-29T17:18:10.974Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/78/1526222cb0fbb9a3ed53493416259c8816bbccb59e9c0ac72bbaa3b00731/mlx-0.29.0-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:e5b3aa72b8ab380e131189427275b8fa0a292bd96f7389d847915a621697824e", size = 546024, upload-time = "2025-08-29T17:16:37.472Z" },
-    { url = "https://files.pythonhosted.org/packages/78/23/5d2929d048eaa12caab2c38936e96205c174dd919bc6db835ab55b3526cb/mlx-0.29.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:b21447542d4082b9de5dfc14253a09574b9df6518133b38d111afbd68457b2e7", size = 546023, upload-time = "2025-08-29T17:16:34.408Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/16/53d7ad4e9b7d2c990440ea1574b2d1ddf8fae3af6c6c78596fd8ed2088d3/mlx-0.29.0-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:b883d3546bc1ef2484c9350ef04bbec685bda8e78f37a81b0520f7dbc5de369b", size = 546018, upload-time = "2025-08-29T17:15:27.79Z" },
-    { url = "https://files.pythonhosted.org/packages/37/1a/d10424c2ffdb7cdc6b1cccda7763d5b72b5ddb6483d8d32206fd76053e04/mlx-0.29.0-cp39-cp39-manylinux_2_35_x86_64.whl", hash = "sha256:ce2a1efc3556cfa2ad99ae17191c352a94648ab1f8ec22d370a94709f428de5b", size = 650315, upload-time = "2025-08-29T17:21:04.237Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/f0/2c2f99a91ed9dfcc78d31d9e5d3bb2f5305a8d65953cbc41f34f8056c49a/mlx-0.29.2-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:b46c1a24b9b8f7145e4d84410552ddfa03f40f9afdbe8f819f6b4b52b4db5d30", size = 547369, upload-time = "2025-09-26T22:21:33.668Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/06/0edddf0a5facb58c17616cd33da6de2722636da8d8d3927272a5a88658e4/mlx-0.29.2-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:18c5b63c6e4b35f75f477f74d3942870e0bc9c9f1c6a071d8a058bbd46681bf4", size = 547367, upload-time = "2025-09-26T22:21:36.63Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/cd/8c089cf1678a752ed4175a7ad5f08823ceef75d797217e12a44a71d6c062/mlx-0.29.2-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:ed53b8383ad4ee4311400558e0a5ec61105fc4553950b5732c66e7081cd1a9e8", size = 547365, upload-time = "2025-09-26T22:21:32.585Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/cc/9a14ac57a251f0ee5047f42214fa50d1d5223d805f0b8b6627639c3692eb/mlx-0.29.2-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:fff27af8dca74546422e8400d32ab848baf42405123cbcfdf62674a9c0560ebe", size = 652302, upload-time = "2025-09-26T22:26:24.575Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/f0/f57349f37cf5dd53f95127e141fc59fc435e4b6bfabba5a84c65de4d3597/mlx-0.29.2-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:e74965369227230374b3e8e8c8d46e209e5221a9b76bbb0fa788617e2c68f73c", size = 547581, upload-time = "2025-09-26T22:21:39.24Z" },
+    { url = "https://files.pythonhosted.org/packages/66/04/e016ca28dc9e0738a2541581420125cfe6bba24466a64420600bdd6fd52c/mlx-0.29.2-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:0f79194eeac78e85b96439d3bbc17aae5aba045a2af083c000b4fbbc501f253e", size = 547581, upload-time = "2025-09-26T22:21:47.706Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/b3/e2595e70ef8d4438dff694857745b0e108911e5b5fb83259dde6e5dc5bd1/mlx-0.29.2-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:33bbbb0fd24895d5ff080bb4d10e3e77017bba675d9a12466c8866eaf9b47854", size = 547578, upload-time = "2025-09-26T22:21:22.041Z" },
+    { url = "https://files.pythonhosted.org/packages/11/8c/5d51543ab128c2dff5e4b44ca799db8db5aa4f4ffc34af6531fd73627b54/mlx-0.29.2-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:32e159f2772be893bec580d2d50c0e6b32ad71a19ded7307bf6c871c8aaa9cf2", size = 651900, upload-time = "2025-09-26T22:26:11.133Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/84/7250237039e91d8e44ca0cf3522f189164844c196f262509afd29ef54710/mlx-0.29.2-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:eec950bf7118ad0865d0fc4686bd85d99bf8463fc717d836a5132e1a08b4f129", size = 548336, upload-time = "2025-09-26T22:21:44.914Z" },
+    { url = "https://files.pythonhosted.org/packages/13/47/428ac8d9b0cb5c136e5ce6c726cfdd55caa5b9497dafb6221acfee18f145/mlx-0.29.2-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:bef7333268d6d02e50a9ac6b10f661b711cd02da4a5e2d7619cf198a7e530308", size = 548334, upload-time = "2025-09-26T22:21:21.41Z" },
+    { url = "https://files.pythonhosted.org/packages/14/f0/7d5d3527ca3fdc664c900b4b822028691739e58c8e8f7975b33df4d3536e/mlx-0.29.2-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:f622fc6a84542a08ad2136e9251822d2c08106e5a1a0bd5d249a2d72bccd6577", size = 548330, upload-time = "2025-09-26T22:21:41.182Z" },
+    { url = "https://files.pythonhosted.org/packages/09/18/e202e0f6232822f6768995cdbf50eda202137bb6547368f6e3993dbee00b/mlx-0.29.2-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:a1aa1aee8e1b6bd1e51361e6b692c70d281b8187b2e859e70ecc11daab306dac", size = 648728, upload-time = "2025-09-26T22:25:49.159Z" },
+    { url = "https://files.pythonhosted.org/packages/da/cb/6dc30495c985a69b3ca51b267a822168b678dd37a580da011a38e1c4ea86/mlx-0.29.2-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:b9b1ece61e25c3e728ce760da1bbb11aa377ea0df84b92fd69cf60af02286969", size = 547851, upload-time = "2025-09-26T22:23:49.215Z" },
+    { url = "https://files.pythonhosted.org/packages/50/61/e7aa84522bce713f0bfc5a85e44f16fbdec00bbc5f96d06a82264859dba2/mlx-0.29.2-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:00cbf0dacf3d7a345581b49ad350b50f419788983dfbf6ee03624e485237c476", size = 547849, upload-time = "2025-09-26T22:23:33.569Z" },
+    { url = "https://files.pythonhosted.org/packages/df/c2/214211fb2e5b0813594287b304852d7e038c571c06a4cbd492068d8b49a8/mlx-0.29.2-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:bd8c031889fdd6e40bfb831490ab6d4776491f7e91909851941489b5329e80d4", size = 547849, upload-time = "2025-09-26T22:23:31.468Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/a0/3fe20a277b8f96131223ea53f739cd8845b6dc27caa6bf6576d4feb880f9/mlx-0.29.2-cp39-cp39-manylinux_2_35_x86_64.whl", hash = "sha256:eb319abca8cfc5a8a9d9c3b020abcc942d12683186d71ec6b6b795eb75206096", size = 652549, upload-time = "2025-09-26T22:28:39.486Z" },
 ]
 
 [[package]]
 name = "mlx-lm"
-version = "0.27.0"
+version = "0.28.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jinja2" },
@@ -1829,19 +2252,19 @@ dependencies = [
     { name = "pyyaml" },
     { name = "transformers" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/8c/4b/68a0b7cde71279ef47a3cbbbe663ea2007824d7ad93a19dc8a26ee68341a/mlx_lm-0.27.0.tar.gz", hash = "sha256:91da83708bc63e5985ff862bc1d66b1348e8af98baa8b7f873661661d8cabfd2", size = 178986, upload-time = "2025-08-29T18:00:07.404Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/1c/d7/fdde445c7bd443a2ed23badda6064f1477c4051543922106f365e94082cd/mlx_lm-0.28.2.tar.gz", hash = "sha256:d28752635ed5c89ff2b41361916c928e6b16f765c07b2908044e1dcaf921ed9b", size = 209374, upload-time = "2025-10-02T14:23:57.497Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/82/9a/5d27af907b579495c5bbff3f42a4dbbe5834aaa62db1cd5f50465fa28a3d/mlx_lm-0.27.0-py3-none-any.whl", hash = "sha256:296e2fdd331db11fa14652161eca3622b2311439309b639d0832dec33fb47d55", size = 245622, upload-time = "2025-08-29T18:00:06.077Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/1c/89e0f60d45e364de8507065f73aeb8d2fd810d6cb95a9a512880b09399d5/mlx_lm-0.28.2-py3-none-any.whl", hash = "sha256:1501529e625d0d648216f7bb543b8b449d5fd17bd598f635536dbc1fbde6d1d6", size = 284600, upload-time = "2025-10-02T14:23:56.395Z" },
 ]
 
 [[package]]
 name = "mlx-metal"
-version = "0.29.0"
+version = "0.29.2"
 source = { registry = "https://pypi.org/simple" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/08/28/3fb942b27d53061cacd12883629e5c48821f9e661f01210f55d89dabe3a6/mlx_metal-0.29.0-py3-none-macosx_13_0_arm64.whl", hash = "sha256:7e8279851504a8394122497d491276db37ef85d29f2f54d8a711ac90edfef7bf", size = 34964891, upload-time = "2025-08-29T17:16:35.72Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/82/582555232044bbdc0d74186077bcd79f1ce45f07a79e49b36afc25f9e313/mlx_metal-0.29.0-py3-none-macosx_14_0_arm64.whl", hash = "sha256:674fc7df4b9502dffe56c92cc95bae3a44950c70aa50f5b53291f228cfbbb631", size = 34694819, upload-time = "2025-08-29T17:16:32.636Z" },
-    { url = "https://files.pythonhosted.org/packages/30/d6/b3fc6d07b9bc814551078e53438ebfae8e44c47e4e5f762df244150e4e3a/mlx_metal-0.29.0-py3-none-macosx_15_0_arm64.whl", hash = "sha256:f69eb71128dcc55b5d9baa7606c31641bdd214b859d809deac51be4248d6fddb", size = 34685946, upload-time = "2025-08-29T17:15:25.028Z" },
+    { url = "https://files.pythonhosted.org/packages/31/a5/a045006546fed791f6e9a74ed4451dac871d3c35f9e54a3a25d820668a85/mlx_metal-0.29.2-py3-none-macosx_13_0_arm64.whl", hash = "sha256:cf8f83a521e620357185c57945142718d526b9312ee112e5a89eb5600480f4d6", size = 35056194, upload-time = "2025-09-26T22:23:47.201Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/8c/4bdd3a7d04ed477b32aec30d30236dfca9f9ac27706cb309511278ddd281/mlx_metal-0.29.2-py3-none-macosx_14_0_arm64.whl", hash = "sha256:fa944001970813b296e8aff5616f2fa9daeda6bc1d190c17fbe8a7ca838ecef0", size = 34791708, upload-time = "2025-09-26T22:23:30.599Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/11/12e158848fe4d3316c999ffb6c2d88f554bde98d69022b3385e25ece997e/mlx_metal-0.29.2-py3-none-macosx_15_0_arm64.whl", hash = "sha256:08d8b7fe305425a14b74ebf36cee176575bfd4cd8d34a2aaae8f05b9983d2d71", size = 34784506, upload-time = "2025-09-26T22:23:29.207Z" },
 ]
 
 [[package]]
@@ -1901,88 +2324,112 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/20/d6/cd62cded572e5e25892747a5d27850170bcd03c855e9c69c538e024de6f9/msgpack-1.1.1-cp39-cp39-win_amd64.whl", hash = "sha256:40eae974c873b2992fd36424a5d9407f93e97656d999f43fca9d29f820899084", size = 71700, upload-time = "2025-06-13T06:52:50.244Z" },
 ]
 
+[[package]]
+name = "msgspec"
+version = "0.19.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/cf/9b/95d8ce458462b8b71b8a70fa94563b2498b89933689f3a7b8911edfae3d7/msgspec-0.19.0.tar.gz", hash = "sha256:604037e7cd475345848116e89c553aa9a233259733ab51986ac924ab1b976f8e", size = 216934, upload-time = "2024-12-27T17:40:28.597Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/e5/27/322badde18eb234e36d4a14122b89edd4e2973cdbc3da61ca7edf40a1ccd/msgspec-0.19.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fe2c4bf29bf4e89790b3117470dea2c20b59932772483082c468b990d45fb947", size = 209334, upload-time = "2024-12-27T17:39:19.065Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/65/080509c5774a1592b2779d902a70b5fe008532759927e011f068145a16cb/msgspec-0.19.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:00e87ecfa9795ee5214861eab8326b0e75475c2e68a384002aa135ea2a27d909", size = 211551, upload-time = "2024-12-27T17:39:21.767Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/2e/1c23c6b4ca6f4285c30a39def1054e2bee281389e4b681b5e3711bd5a8c9/msgspec-0.19.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:3c4ec642689da44618f68c90855a10edbc6ac3ff7c1d94395446c65a776e712a", size = 215099, upload-time = "2024-12-27T17:39:24.71Z" },
+    { url = "https://files.pythonhosted.org/packages/83/fe/95f9654518879f3359d1e76bc41189113aa9102452170ab7c9a9a4ee52f6/msgspec-0.19.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:2719647625320b60e2d8af06b35f5b12d4f4d281db30a15a1df22adb2295f633", size = 218211, upload-time = "2024-12-27T17:39:27.396Z" },
+    { url = "https://files.pythonhosted.org/packages/81/25/3a4b24d468203d8af90d1d351b77ea3cffb96b29492855cf83078f16bfe4/msgspec-0.19.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e78f46ff39a427e10b4a61614a2777ad69559cc8d603a7c05681f5a595ea98f7", size = 209029, upload-time = "2024-12-27T17:39:35.023Z" },
+    { url = "https://files.pythonhosted.org/packages/85/2e/db7e189b57901955239f7689b5dcd6ae9458637a9c66747326726c650523/msgspec-0.19.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6c7adf191e4bd3be0e9231c3b6dc20cf1199ada2af523885efc2ed218eafd011", size = 210682, upload-time = "2024-12-27T17:39:36.384Z" },
+    { url = "https://files.pythonhosted.org/packages/03/97/7c8895c9074a97052d7e4a1cc1230b7b6e2ca2486714eb12c3f08bb9d284/msgspec-0.19.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:f04cad4385e20be7c7176bb8ae3dca54a08e9756cfc97bcdb4f18560c3042063", size = 214003, upload-time = "2024-12-27T17:39:39.097Z" },
+    { url = "https://files.pythonhosted.org/packages/61/61/e892997bcaa289559b4d5869f066a8021b79f4bf8e955f831b095f47a4cd/msgspec-0.19.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:45c8fb410670b3b7eb884d44a75589377c341ec1392b778311acdbfa55187716", size = 216833, upload-time = "2024-12-27T17:39:41.203Z" },
+    { url = "https://files.pythonhosted.org/packages/87/81/0c8c93f0b92c97e326b279795f9c5b956c5a97af28ca0fbb9fd86c83737a/msgspec-0.19.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4cfc033c02c3e0aec52b71710d7f84cb3ca5eb407ab2ad23d75631153fdb1f12", size = 210633, upload-time = "2024-12-27T17:39:49.099Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/ef/c5422ce8af73928d194a6606f8ae36e93a52fd5e8df5abd366903a5ca8da/msgspec-0.19.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d911c442571605e17658ca2b416fd8579c5050ac9adc5e00c2cb3126c97f73bc", size = 213594, upload-time = "2024-12-27T17:39:51.204Z" },
+    { url = "https://files.pythonhosted.org/packages/19/2b/4137bc2ed45660444842d042be2cf5b18aa06efd2cda107cff18253b9653/msgspec-0.19.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:757b501fa57e24896cf40a831442b19a864f56d253679f34f260dcb002524a6c", size = 214053, upload-time = "2024-12-27T17:39:52.866Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/e6/8ad51bdc806aac1dc501e8fe43f759f9ed7284043d722b53323ea421c360/msgspec-0.19.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:5f0f65f29b45e2816d8bded36e6b837a4bf5fb60ec4bc3c625fa2c6da4124537", size = 219081, upload-time = "2024-12-27T17:39:55.142Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/a2/48f2c15c7644668e51f4dce99d5f709bd55314e47acb02e90682f5880f35/msgspec-0.19.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0f5c043ace7962ef188746e83b99faaa9e3e699ab857ca3f367b309c8e2c6b12", size = 209272, upload-time = "2024-12-27T17:40:21.534Z" },
+    { url = "https://files.pythonhosted.org/packages/25/3c/aa339cf08b990c3f07e67b229a3a8aa31bf129ed974b35e5daa0df7d9d56/msgspec-0.19.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ca06aa08e39bf57e39a258e1996474f84d0dd8130d486c00bec26d797b8c5446", size = 211396, upload-time = "2024-12-27T17:40:22.897Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/00/c7fb9d524327c558b2803973cc3f988c5100a1708879970a9e377bdf6f4f/msgspec-0.19.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:e695dad6897896e9384cf5e2687d9ae9feaef50e802f93602d35458e20d1fb19", size = 215002, upload-time = "2024-12-27T17:40:24.341Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/bf/d9f9fff026c1248cde84a5ce62b3742e8a63a3c4e811f99f00c8babf7615/msgspec-0.19.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:3be5c02e1fee57b54130316a08fe40cca53af92999a302a6054cd451700ea7db", size = 218132, upload-time = "2024-12-27T17:40:25.744Z" },
+]
+
 [[package]]
 name = "multidict"
-version = "6.6.4"
+version = "6.7.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/69/7f/0652e6ed47ab288e3756ea9c0df8b14950781184d4bd7883f4d87dd41245/multidict-6.6.4.tar.gz", hash = "sha256:d2d4e4787672911b48350df02ed3fa3fffdc2f2e8ca06dd6afdf34189b76a9dd", size = 101843, upload-time = "2025-08-11T12:08:48.217Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/eb/6b/86f353088c1358e76fd30b0146947fddecee812703b604ee901e85cd2a80/multidict-6.6.4-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:b8aa6f0bd8125ddd04a6593437bad6a7e70f300ff4180a531654aa2ab3f6d58f", size = 77054, upload-time = "2025-08-11T12:06:02.99Z" },
-    { url = "https://files.pythonhosted.org/packages/19/5d/c01dc3d3788bb877bd7f5753ea6eb23c1beeca8044902a8f5bfb54430f63/multidict-6.6.4-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:b9e5853bbd7264baca42ffc53391b490d65fe62849bf2c690fa3f6273dbcd0cb", size = 44914, upload-time = "2025-08-11T12:06:05.264Z" },
-    { url = "https://files.pythonhosted.org/packages/46/44/964dae19ea42f7d3e166474d8205f14bb811020e28bc423d46123ddda763/multidict-6.6.4-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0af5f9dee472371e36d6ae38bde009bd8ce65ac7335f55dcc240379d7bed1495", size = 44601, upload-time = "2025-08-11T12:06:06.627Z" },
-    { url = "https://files.pythonhosted.org/packages/31/20/0616348a1dfb36cb2ab33fc9521de1f27235a397bf3f59338e583afadd17/multidict-6.6.4-cp310-cp310-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:d24f351e4d759f5054b641c81e8291e5d122af0fca5c72454ff77f7cbe492de8", size = 224821, upload-time = "2025-08-11T12:06:08.06Z" },
-    { url = "https://files.pythonhosted.org/packages/14/26/5d8923c69c110ff51861af05bd27ca6783011b96725d59ccae6d9daeb627/multidict-6.6.4-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:db6a3810eec08280a172a6cd541ff4a5f6a97b161d93ec94e6c4018917deb6b7", size = 242608, upload-time = "2025-08-11T12:06:09.697Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/cc/e2ad3ba9459aa34fa65cf1f82a5c4a820a2ce615aacfb5143b8817f76504/multidict-6.6.4-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:a1b20a9d56b2d81e2ff52ecc0670d583eaabaa55f402e8d16dd062373dbbe796", size = 222324, upload-time = "2025-08-11T12:06:10.905Z" },
-    { url = "https://files.pythonhosted.org/packages/19/db/4ed0f65701afbc2cb0c140d2d02928bb0fe38dd044af76e58ad7c54fd21f/multidict-6.6.4-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8c9854df0eaa610a23494c32a6f44a3a550fb398b6b51a56e8c6b9b3689578db", size = 253234, upload-time = "2025-08-11T12:06:12.658Z" },
-    { url = "https://files.pythonhosted.org/packages/94/c1/5160c9813269e39ae14b73debb907bfaaa1beee1762da8c4fb95df4764ed/multidict-6.6.4-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4bb7627fd7a968f41905a4d6343b0d63244a0623f006e9ed989fa2b78f4438a0", size = 251613, upload-time = "2025-08-11T12:06:13.97Z" },
-    { url = "https://files.pythonhosted.org/packages/05/a9/48d1bd111fc2f8fb98b2ed7f9a115c55a9355358432a19f53c0b74d8425d/multidict-6.6.4-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:caebafea30ed049c57c673d0b36238b1748683be2593965614d7b0e99125c877", size = 241649, upload-time = "2025-08-11T12:06:15.204Z" },
-    { url = "https://files.pythonhosted.org/packages/85/2a/f7d743df0019408768af8a70d2037546a2be7b81fbb65f040d76caafd4c5/multidict-6.6.4-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ad887a8250eb47d3ab083d2f98db7f48098d13d42eb7a3b67d8a5c795f224ace", size = 239238, upload-time = "2025-08-11T12:06:16.467Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/b8/4f4bb13323c2d647323f7919201493cf48ebe7ded971717bfb0f1a79b6bf/multidict-6.6.4-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:ed8358ae7d94ffb7c397cecb62cbac9578a83ecefc1eba27b9090ee910e2efb6", size = 233517, upload-time = "2025-08-11T12:06:18.107Z" },
-    { url = "https://files.pythonhosted.org/packages/33/29/4293c26029ebfbba4f574febd2ed01b6f619cfa0d2e344217d53eef34192/multidict-6.6.4-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:ecab51ad2462197a4c000b6d5701fc8585b80eecb90583635d7e327b7b6923eb", size = 243122, upload-time = "2025-08-11T12:06:19.361Z" },
-    { url = "https://files.pythonhosted.org/packages/20/60/a1c53628168aa22447bfde3a8730096ac28086704a0d8c590f3b63388d0c/multidict-6.6.4-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:c5c97aa666cf70e667dfa5af945424ba1329af5dd988a437efeb3a09430389fb", size = 248992, upload-time = "2025-08-11T12:06:20.661Z" },
-    { url = "https://files.pythonhosted.org/packages/a3/3b/55443a0c372f33cae5d9ec37a6a973802884fa0ab3586659b197cf8cc5e9/multidict-6.6.4-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:9a950b7cf54099c1209f455ac5970b1ea81410f2af60ed9eb3c3f14f0bfcf987", size = 243708, upload-time = "2025-08-11T12:06:21.891Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/60/a18c6900086769312560b2626b18e8cca22d9e85b1186ba77f4755b11266/multidict-6.6.4-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:163c7ea522ea9365a8a57832dea7618e6cbdc3cd75f8c627663587459a4e328f", size = 237498, upload-time = "2025-08-11T12:06:23.206Z" },
-    { url = "https://files.pythonhosted.org/packages/11/3d/8bdd8bcaff2951ce2affccca107a404925a2beafedd5aef0b5e4a71120a6/multidict-6.6.4-cp310-cp310-win32.whl", hash = "sha256:17d2cbbfa6ff20821396b25890f155f40c986f9cfbce5667759696d83504954f", size = 41415, upload-time = "2025-08-11T12:06:24.77Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/53/cab1ad80356a4cd1b685a254b680167059b433b573e53872fab245e9fc95/multidict-6.6.4-cp310-cp310-win_amd64.whl", hash = "sha256:ce9a40fbe52e57e7edf20113a4eaddfacac0561a0879734e636aa6d4bb5e3fb0", size = 46046, upload-time = "2025-08-11T12:06:25.893Z" },
-    { url = "https://files.pythonhosted.org/packages/cf/9a/874212b6f5c1c2d870d0a7adc5bb4cfe9b0624fa15cdf5cf757c0f5087ae/multidict-6.6.4-cp310-cp310-win_arm64.whl", hash = "sha256:01d0959807a451fe9fdd4da3e139cb5b77f7328baf2140feeaf233e1d777b729", size = 43147, upload-time = "2025-08-11T12:06:27.534Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/7f/90a7f01e2d005d6653c689039977f6856718c75c5579445effb7e60923d1/multidict-6.6.4-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:c7a0e9b561e6460484318a7612e725df1145d46b0ef57c6b9866441bf6e27e0c", size = 76472, upload-time = "2025-08-11T12:06:29.006Z" },
-    { url = "https://files.pythonhosted.org/packages/54/a3/bed07bc9e2bb302ce752f1dabc69e884cd6a676da44fb0e501b246031fdd/multidict-6.6.4-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:6bf2f10f70acc7a2446965ffbc726e5fc0b272c97a90b485857e5c70022213eb", size = 44634, upload-time = "2025-08-11T12:06:30.374Z" },
-    { url = "https://files.pythonhosted.org/packages/a7/4b/ceeb4f8f33cf81277da464307afeaf164fb0297947642585884f5cad4f28/multidict-6.6.4-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:66247d72ed62d5dd29752ffc1d3b88f135c6a8de8b5f63b7c14e973ef5bda19e", size = 44282, upload-time = "2025-08-11T12:06:31.958Z" },
-    { url = "https://files.pythonhosted.org/packages/03/35/436a5da8702b06866189b69f655ffdb8f70796252a8772a77815f1812679/multidict-6.6.4-cp311-cp311-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:105245cc6b76f51e408451a844a54e6823bbd5a490ebfe5bdfc79798511ceded", size = 229696, upload-time = "2025-08-11T12:06:33.087Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/0e/915160be8fecf1fca35f790c08fb74ca684d752fcba62c11daaf3d92c216/multidict-6.6.4-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cbbc54e58b34c3bae389ef00046be0961f30fef7cb0dd9c7756aee376a4f7683", size = 246665, upload-time = "2025-08-11T12:06:34.448Z" },
-    { url = "https://files.pythonhosted.org/packages/08/ee/2f464330acd83f77dcc346f0b1a0eaae10230291450887f96b204b8ac4d3/multidict-6.6.4-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:56c6b3652f945c9bc3ac6c8178cd93132b8d82dd581fcbc3a00676c51302bc1a", size = 225485, upload-time = "2025-08-11T12:06:35.672Z" },
-    { url = "https://files.pythonhosted.org/packages/71/cc/9a117f828b4d7fbaec6adeed2204f211e9caf0a012692a1ee32169f846ae/multidict-6.6.4-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b95494daf857602eccf4c18ca33337dd2be705bccdb6dddbfc9d513e6addb9d9", size = 257318, upload-time = "2025-08-11T12:06:36.98Z" },
-    { url = "https://files.pythonhosted.org/packages/25/77/62752d3dbd70e27fdd68e86626c1ae6bccfebe2bb1f84ae226363e112f5a/multidict-6.6.4-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:e5b1413361cef15340ab9dc61523e653d25723e82d488ef7d60a12878227ed50", size = 254689, upload-time = "2025-08-11T12:06:38.233Z" },
-    { url = "https://files.pythonhosted.org/packages/00/6e/fac58b1072a6fc59af5e7acb245e8754d3e1f97f4f808a6559951f72a0d4/multidict-6.6.4-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e167bf899c3d724f9662ef00b4f7fef87a19c22b2fead198a6f68b263618df52", size = 246709, upload-time = "2025-08-11T12:06:39.517Z" },
-    { url = "https://files.pythonhosted.org/packages/01/ef/4698d6842ef5e797c6db7744b0081e36fb5de3d00002cc4c58071097fac3/multidict-6.6.4-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:aaea28ba20a9026dfa77f4b80369e51cb767c61e33a2d4043399c67bd95fb7c6", size = 243185, upload-time = "2025-08-11T12:06:40.796Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/c9/d82e95ae1d6e4ef396934e9b0e942dfc428775f9554acf04393cce66b157/multidict-6.6.4-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:8c91cdb30809a96d9ecf442ec9bc45e8cfaa0f7f8bdf534e082c2443a196727e", size = 237838, upload-time = "2025-08-11T12:06:42.595Z" },
-    { url = "https://files.pythonhosted.org/packages/57/cf/f94af5c36baaa75d44fab9f02e2a6bcfa0cd90acb44d4976a80960759dbc/multidict-6.6.4-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:1a0ccbfe93ca114c5d65a2471d52d8829e56d467c97b0e341cf5ee45410033b3", size = 246368, upload-time = "2025-08-11T12:06:44.304Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/fe/29f23460c3d995f6a4b678cb2e9730e7277231b981f0b234702f0177818a/multidict-6.6.4-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:55624b3f321d84c403cb7d8e6e982f41ae233d85f85db54ba6286f7295dc8a9c", size = 253339, upload-time = "2025-08-11T12:06:45.597Z" },
-    { url = "https://files.pythonhosted.org/packages/29/b6/fd59449204426187b82bf8a75f629310f68c6adc9559dc922d5abe34797b/multidict-6.6.4-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:4a1fb393a2c9d202cb766c76208bd7945bc194eba8ac920ce98c6e458f0b524b", size = 246933, upload-time = "2025-08-11T12:06:46.841Z" },
-    { url = "https://files.pythonhosted.org/packages/19/52/d5d6b344f176a5ac3606f7a61fb44dc746e04550e1a13834dff722b8d7d6/multidict-6.6.4-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:43868297a5759a845fa3a483fb4392973a95fb1de891605a3728130c52b8f40f", size = 242225, upload-time = "2025-08-11T12:06:48.588Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/d3/5b2281ed89ff4d5318d82478a2a2450fcdfc3300da48ff15c1778280ad26/multidict-6.6.4-cp311-cp311-win32.whl", hash = "sha256:ed3b94c5e362a8a84d69642dbeac615452e8af9b8eb825b7bc9f31a53a1051e2", size = 41306, upload-time = "2025-08-11T12:06:49.95Z" },
-    { url = "https://files.pythonhosted.org/packages/74/7d/36b045c23a1ab98507aefd44fd8b264ee1dd5e5010543c6fccf82141ccef/multidict-6.6.4-cp311-cp311-win_amd64.whl", hash = "sha256:d8c112f7a90d8ca5d20213aa41eac690bb50a76da153e3afb3886418e61cb22e", size = 46029, upload-time = "2025-08-11T12:06:51.082Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/5e/553d67d24432c5cd52b49047f2d248821843743ee6d29a704594f656d182/multidict-6.6.4-cp311-cp311-win_arm64.whl", hash = "sha256:3bb0eae408fa1996d87247ca0d6a57b7fc1dcf83e8a5c47ab82c558c250d4adf", size = 43017, upload-time = "2025-08-11T12:06:52.243Z" },
-    { url = "https://files.pythonhosted.org/packages/05/f6/512ffd8fd8b37fb2680e5ac35d788f1d71bbaf37789d21a820bdc441e565/multidict-6.6.4-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:0ffb87be160942d56d7b87b0fdf098e81ed565add09eaa1294268c7f3caac4c8", size = 76516, upload-time = "2025-08-11T12:06:53.393Z" },
-    { url = "https://files.pythonhosted.org/packages/99/58/45c3e75deb8855c36bd66cc1658007589662ba584dbf423d01df478dd1c5/multidict-6.6.4-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:d191de6cbab2aff5de6c5723101705fd044b3e4c7cfd587a1929b5028b9714b3", size = 45394, upload-time = "2025-08-11T12:06:54.555Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/ca/e8c4472a93a26e4507c0b8e1f0762c0d8a32de1328ef72fd704ef9cc5447/multidict-6.6.4-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:38a0956dd92d918ad5feff3db8fcb4a5eb7dba114da917e1a88475619781b57b", size = 43591, upload-time = "2025-08-11T12:06:55.672Z" },
-    { url = "https://files.pythonhosted.org/packages/05/51/edf414f4df058574a7265034d04c935aa84a89e79ce90fcf4df211f47b16/multidict-6.6.4-cp312-cp312-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:6865f6d3b7900ae020b495d599fcf3765653bc927951c1abb959017f81ae8287", size = 237215, upload-time = "2025-08-11T12:06:57.213Z" },
-    { url = "https://files.pythonhosted.org/packages/c8/45/8b3d6dbad8cf3252553cc41abea09ad527b33ce47a5e199072620b296902/multidict-6.6.4-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a2088c126b6f72db6c9212ad827d0ba088c01d951cee25e758c450da732c138", size = 258299, upload-time = "2025-08-11T12:06:58.946Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/e8/8ca2e9a9f5a435fc6db40438a55730a4bf4956b554e487fa1b9ae920f825/multidict-6.6.4-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:0f37bed7319b848097085d7d48116f545985db988e2256b2e6f00563a3416ee6", size = 242357, upload-time = "2025-08-11T12:07:00.301Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/84/80c77c99df05a75c28490b2af8f7cba2a12621186e0a8b0865d8e745c104/multidict-6.6.4-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:01368e3c94032ba6ca0b78e7ccb099643466cf24f8dc8eefcfdc0571d56e58f9", size = 268369, upload-time = "2025-08-11T12:07:01.638Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/e9/920bfa46c27b05fb3e1ad85121fd49f441492dca2449c5bcfe42e4565d8a/multidict-6.6.4-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:8fe323540c255db0bffee79ad7f048c909f2ab0edb87a597e1c17da6a54e493c", size = 269341, upload-time = "2025-08-11T12:07:02.943Z" },
-    { url = "https://files.pythonhosted.org/packages/af/65/753a2d8b05daf496f4a9c367fe844e90a1b2cac78e2be2c844200d10cc4c/multidict-6.6.4-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b8eb3025f17b0a4c3cd08cda49acf312a19ad6e8a4edd9dbd591e6506d999402", size = 256100, upload-time = "2025-08-11T12:07:04.564Z" },
-    { url = "https://files.pythonhosted.org/packages/09/54/655be13ae324212bf0bc15d665a4e34844f34c206f78801be42f7a0a8aaa/multidict-6.6.4-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:bbc14f0365534d35a06970d6a83478b249752e922d662dc24d489af1aa0d1be7", size = 253584, upload-time = "2025-08-11T12:07:05.914Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/74/ab2039ecc05264b5cec73eb018ce417af3ebb384ae9c0e9ed42cb33f8151/multidict-6.6.4-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:75aa52fba2d96bf972e85451b99d8e19cc37ce26fd016f6d4aa60da9ab2b005f", size = 251018, upload-time = "2025-08-11T12:07:08.301Z" },
-    { url = "https://files.pythonhosted.org/packages/af/0a/ccbb244ac848e56c6427f2392741c06302bbfba49c0042f1eb3c5b606497/multidict-6.6.4-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:4fefd4a815e362d4f011919d97d7b4a1e566f1dde83dc4ad8cfb5b41de1df68d", size = 251477, upload-time = "2025-08-11T12:07:10.248Z" },
-    { url = "https://files.pythonhosted.org/packages/0e/b0/0ed49bba775b135937f52fe13922bc64a7eaf0a3ead84a36e8e4e446e096/multidict-6.6.4-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:db9801fe021f59a5b375ab778973127ca0ac52429a26e2fd86aa9508f4d26eb7", size = 263575, upload-time = "2025-08-11T12:07:11.928Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/d9/7fb85a85e14de2e44dfb6a24f03c41e2af8697a6df83daddb0e9b7569f73/multidict-6.6.4-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:a650629970fa21ac1fb06ba25dabfc5b8a2054fcbf6ae97c758aa956b8dba802", size = 259649, upload-time = "2025-08-11T12:07:13.244Z" },
-    { url = "https://files.pythonhosted.org/packages/03/9e/b3a459bcf9b6e74fa461a5222a10ff9b544cb1cd52fd482fb1b75ecda2a2/multidict-6.6.4-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:452ff5da78d4720d7516a3a2abd804957532dd69296cb77319c193e3ffb87e24", size = 251505, upload-time = "2025-08-11T12:07:14.57Z" },
-    { url = "https://files.pythonhosted.org/packages/86/a2/8022f78f041dfe6d71e364001a5cf987c30edfc83c8a5fb7a3f0974cff39/multidict-6.6.4-cp312-cp312-win32.whl", hash = "sha256:8c2fcb12136530ed19572bbba61b407f655e3953ba669b96a35036a11a485793", size = 41888, upload-time = "2025-08-11T12:07:15.904Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/eb/d88b1780d43a56db2cba24289fa744a9d216c1a8546a0dc3956563fd53ea/multidict-6.6.4-cp312-cp312-win_amd64.whl", hash = "sha256:047d9425860a8c9544fed1b9584f0c8bcd31bcde9568b047c5e567a1025ecd6e", size = 46072, upload-time = "2025-08-11T12:07:17.045Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/16/b929320bf5750e2d9d4931835a4c638a19d2494a5b519caaaa7492ebe105/multidict-6.6.4-cp312-cp312-win_arm64.whl", hash = "sha256:14754eb72feaa1e8ae528468f24250dd997b8e2188c3d2f593f9eba259e4b364", size = 43222, upload-time = "2025-08-11T12:07:18.328Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/d3/f04c5db316caee9b5b2cbba66270b358c922a959855995bedde87134287c/multidict-6.6.4-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:af7618b591bae552b40dbb6f93f5518328a949dac626ee75927bba1ecdeea9f4", size = 76977, upload-time = "2025-08-11T12:08:16.667Z" },
-    { url = "https://files.pythonhosted.org/packages/70/39/a6200417d883e510728ab3caec02d3b66ff09e1c85e0aab2ba311abfdf06/multidict-6.6.4-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:b6819f83aef06f560cb15482d619d0e623ce9bf155115150a85ab11b8342a665", size = 44878, upload-time = "2025-08-11T12:08:18.157Z" },
-    { url = "https://files.pythonhosted.org/packages/6f/7e/815be31ed35571b137d65232816f61513fcd97b2717d6a9d7800b5a0c6e0/multidict-6.6.4-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:4d09384e75788861e046330308e7af54dd306aaf20eb760eb1d0de26b2bea2cb", size = 44546, upload-time = "2025-08-11T12:08:19.694Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/f1/21b5bff6a8c3e2aff56956c241941ace6b8820e1abe6b12d3c52868a773d/multidict-6.6.4-cp39-cp39-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:a59c63061f1a07b861c004e53869eb1211ffd1a4acbca330e3322efa6dd02978", size = 223020, upload-time = "2025-08-11T12:08:21.554Z" },
-    { url = "https://files.pythonhosted.org/packages/15/59/37083f1dd3439979a0ffeb1906818d978d88b4cc7f4600a9f89b1cb6713c/multidict-6.6.4-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:350f6b0fe1ced61e778037fdc7613f4051c8baf64b1ee19371b42a3acdb016a0", size = 240528, upload-time = "2025-08-11T12:08:23.45Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/f0/f054d123c87784307a27324c829eb55bcfd2e261eb785fcabbd832c8dc4a/multidict-6.6.4-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:0c5cbac6b55ad69cb6aa17ee9343dfbba903118fd530348c330211dc7aa756d1", size = 219540, upload-time = "2025-08-11T12:08:24.965Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/26/8f78ce17b7118149c17f238f28fba2a850b660b860f9b024a34d0191030f/multidict-6.6.4-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:630f70c32b8066ddfd920350bc236225814ad94dfa493fe1910ee17fe4365cbb", size = 251182, upload-time = "2025-08-11T12:08:26.511Z" },
-    { url = "https://files.pythonhosted.org/packages/00/c3/a21466322d69f6594fe22d9379200f99194d21c12a5bbf8c2a39a46b83b6/multidict-6.6.4-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f8d4916a81697faec6cb724a273bd5457e4c6c43d82b29f9dc02c5542fd21fc9", size = 249371, upload-time = "2025-08-11T12:08:28.075Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/8e/2e673124eb05cf8dc82e9265eccde01a36bcbd3193e27799b8377123c976/multidict-6.6.4-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8e42332cf8276bb7645d310cdecca93a16920256a5b01bebf747365f86a1675b", size = 239235, upload-time = "2025-08-11T12:08:29.937Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/2d/bdd9f05e7c89e30a4b0e4faf0681a30748f8d1310f68cfdc0e3571e75bd5/multidict-6.6.4-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:f3be27440f7644ab9a13a6fc86f09cdd90b347c3c5e30c6d6d860de822d7cb53", size = 237410, upload-time = "2025-08-11T12:08:31.872Z" },
-    { url = "https://files.pythonhosted.org/packages/46/4c/3237b83f8ca9a2673bb08fc340c15da005a80f5cc49748b587c8ae83823b/multidict-6.6.4-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:21f216669109e02ef3e2415ede07f4f8987f00de8cdfa0cc0b3440d42534f9f0", size = 232979, upload-time = "2025-08-11T12:08:33.399Z" },
-    { url = "https://files.pythonhosted.org/packages/55/a6/a765decff625ae9bc581aed303cd1837955177dafc558859a69f56f56ba8/multidict-6.6.4-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:d9890d68c45d1aeac5178ded1d1cccf3bc8d7accf1f976f79bf63099fb16e4bd", size = 240979, upload-time = "2025-08-11T12:08:35.02Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/2d/9c75975cb0c66ea33cae1443bb265b2b3cd689bffcbc68872565f401da23/multidict-6.6.4-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:edfdcae97cdc5d1a89477c436b61f472c4d40971774ac4729c613b4b133163cb", size = 246849, upload-time = "2025-08-11T12:08:37.038Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/71/d21ac0843c1d8751fb5dcf8a1f436625d39d4577bc27829799d09b419af7/multidict-6.6.4-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:0b2e886624be5773e69cf32bcb8534aecdeb38943520b240fed3d5596a430f2f", size = 241798, upload-time = "2025-08-11T12:08:38.669Z" },
-    { url = "https://files.pythonhosted.org/packages/94/3d/1d8911e53092837bd11b1c99d71de3e2a9a26f8911f864554677663242aa/multidict-6.6.4-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:be5bf4b3224948032a845d12ab0f69f208293742df96dc14c4ff9b09e508fc17", size = 235315, upload-time = "2025-08-11T12:08:40.266Z" },
-    { url = "https://files.pythonhosted.org/packages/86/c5/4b758df96376f73e936b1942c6c2dfc17e37ed9d5ff3b01a811496966ca0/multidict-6.6.4-cp39-cp39-win32.whl", hash = "sha256:10a68a9191f284fe9d501fef4efe93226e74df92ce7a24e301371293bd4918ae", size = 41434, upload-time = "2025-08-11T12:08:41.965Z" },
-    { url = "https://files.pythonhosted.org/packages/58/16/f1dfa2a0f25f2717a5e9e5fe8fd30613f7fe95e3530cec8d11f5de0b709c/multidict-6.6.4-cp39-cp39-win_amd64.whl", hash = "sha256:ee25f82f53262f9ac93bd7e58e47ea1bdcc3393cef815847e397cba17e284210", size = 46186, upload-time = "2025-08-11T12:08:43.367Z" },
-    { url = "https://files.pythonhosted.org/packages/88/7d/a0568bac65438c494cb6950b29f394d875a796a237536ac724879cf710c9/multidict-6.6.4-cp39-cp39-win_arm64.whl", hash = "sha256:f9867e55590e0855bcec60d4f9a092b69476db64573c9fe17e92b0c50614c16a", size = 43115, upload-time = "2025-08-11T12:08:45.126Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/69/b547032297c7e63ba2af494edba695d781af8a0c6e89e4d06cf848b21d80/multidict-6.6.4-py3-none-any.whl", hash = "sha256:27d8f8e125c07cb954e54d75d04905a9bba8a439c1d84aca94949d4d03d8601c", size = 12313, upload-time = "2025-08-11T12:08:46.891Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/80/1e/5492c365f222f907de1039b91f922b93fa4f764c713ee858d235495d8f50/multidict-6.7.0.tar.gz", hash = "sha256:c6e99d9a65ca282e578dfea819cfa9c0a62b2499d8677392e09feaf305e9e6f5", size = 101834, upload-time = "2025-10-06T14:52:30.657Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a9/63/7bdd4adc330abcca54c85728db2327130e49e52e8c3ce685cec44e0f2e9f/multidict-6.7.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:9f474ad5acda359c8758c8accc22032c6abe6dc87a8be2440d097785e27a9349", size = 77153, upload-time = "2025-10-06T14:48:26.409Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/bb/b6c35ff175ed1a3142222b78455ee31be71a8396ed3ab5280fbe3ebe4e85/multidict-6.7.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:4b7a9db5a870f780220e931d0002bbfd88fb53aceb6293251e2c839415c1b20e", size = 44993, upload-time = "2025-10-06T14:48:28.4Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/1f/064c77877c5fa6df6d346e68075c0f6998547afe952d6471b4c5f6a7345d/multidict-6.7.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:03ca744319864e92721195fa28c7a3b2bc7b686246b35e4078c1e4d0eb5466d3", size = 44607, upload-time = "2025-10-06T14:48:29.581Z" },
+    { url = "https://files.pythonhosted.org/packages/04/7a/bf6aa92065dd47f287690000b3d7d332edfccb2277634cadf6a810463c6a/multidict-6.7.0-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:f0e77e3c0008bc9316e662624535b88d360c3a5d3f81e15cf12c139a75250046", size = 241847, upload-time = "2025-10-06T14:48:32.107Z" },
+    { url = "https://files.pythonhosted.org/packages/94/39/297a8de920f76eda343e4ce05f3b489f0ab3f9504f2576dfb37b7c08ca08/multidict-6.7.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:08325c9e5367aa379a3496aa9a022fe8837ff22e00b94db256d3a1378c76ab32", size = 242616, upload-time = "2025-10-06T14:48:34.054Z" },
+    { url = "https://files.pythonhosted.org/packages/39/3a/d0eee2898cfd9d654aea6cb8c4addc2f9756e9a7e09391cfe55541f917f7/multidict-6.7.0-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:e2862408c99f84aa571ab462d25236ef9cb12a602ea959ba9c9009a54902fc73", size = 222333, upload-time = "2025-10-06T14:48:35.9Z" },
+    { url = "https://files.pythonhosted.org/packages/05/48/3b328851193c7a4240815b71eea165b49248867bbb6153a0aee227a0bb47/multidict-6.7.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4d72a9a2d885f5c208b0cb91ff2ed43636bb7e345ec839ff64708e04f69a13cc", size = 253239, upload-time = "2025-10-06T14:48:37.302Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/ca/0706a98c8d126a89245413225ca4a3fefc8435014de309cf8b30acb68841/multidict-6.7.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:478cc36476687bac1514d651cbbaa94b86b0732fb6855c60c673794c7dd2da62", size = 251618, upload-time = "2025-10-06T14:48:38.963Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/4f/9c7992f245554d8b173f6f0a048ad24b3e645d883f096857ec2c0822b8bd/multidict-6.7.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6843b28b0364dc605f21481c90fadb5f60d9123b442eb8a726bb74feef588a84", size = 241655, upload-time = "2025-10-06T14:48:40.312Z" },
+    { url = "https://files.pythonhosted.org/packages/31/79/26a85991ae67efd1c0b1fc2e0c275b8a6aceeb155a68861f63f87a798f16/multidict-6.7.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:23bfeee5316266e5ee2d625df2d2c602b829435fc3a235c2ba2131495706e4a0", size = 239245, upload-time = "2025-10-06T14:48:41.848Z" },
+    { url = "https://files.pythonhosted.org/packages/14/1e/75fa96394478930b79d0302eaf9a6c69f34005a1a5251ac8b9c336486ec9/multidict-6.7.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:680878b9f3d45c31e1f730eef731f9b0bc1da456155688c6745ee84eb818e90e", size = 233523, upload-time = "2025-10-06T14:48:43.749Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/5e/085544cb9f9c4ad2b5d97467c15f856df8d9bac410cffd5c43991a5d878b/multidict-6.7.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:eb866162ef2f45063acc7a53a88ef6fe8bf121d45c30ea3c9cd87ce7e191a8d4", size = 243129, upload-time = "2025-10-06T14:48:45.225Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/c3/e9d9e2f20c9474e7a8fcef28f863c5cbd29bb5adce6b70cebe8bdad0039d/multidict-6.7.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:df0e3bf7993bdbeca5ac25aa859cf40d39019e015c9c91809ba7093967f7a648", size = 248999, upload-time = "2025-10-06T14:48:46.703Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/3f/df171b6efa3239ae33b97b887e42671cd1d94d460614bfb2c30ffdab3b95/multidict-6.7.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:661709cdcd919a2ece2234f9bae7174e5220c80b034585d7d8a755632d3e2111", size = 243711, upload-time = "2025-10-06T14:48:48.146Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/2f/9b5564888c4e14b9af64c54acf149263721a283aaf4aa0ae89b091d5d8c1/multidict-6.7.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:096f52730c3fb8ed419db2d44391932b63891b2c5ed14850a7e215c0ba9ade36", size = 237504, upload-time = "2025-10-06T14:48:49.447Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/3a/0bd6ca0f7d96d790542d591c8c3354c1e1b6bfd2024d4d92dc3d87485ec7/multidict-6.7.0-cp310-cp310-win32.whl", hash = "sha256:afa8a2978ec65d2336305550535c9c4ff50ee527914328c8677b3973ade52b85", size = 41422, upload-time = "2025-10-06T14:48:50.789Z" },
+    { url = "https://files.pythonhosted.org/packages/00/35/f6a637ea2c75f0d3b7c7d41b1189189acff0d9deeb8b8f35536bb30f5e33/multidict-6.7.0-cp310-cp310-win_amd64.whl", hash = "sha256:b15b3afff74f707b9275d5ba6a91ae8f6429c3ffb29bbfd216b0b375a56f13d7", size = 46050, upload-time = "2025-10-06T14:48:51.938Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/b8/f7bf8329b39893d02d9d95cf610c75885d12fc0f402b1c894e1c8e01c916/multidict-6.7.0-cp310-cp310-win_arm64.whl", hash = "sha256:4b73189894398d59131a66ff157837b1fafea9974be486d036bb3d32331fdbf0", size = 43153, upload-time = "2025-10-06T14:48:53.146Z" },
+    { url = "https://files.pythonhosted.org/packages/34/9e/5c727587644d67b2ed479041e4b1c58e30afc011e3d45d25bbe35781217c/multidict-6.7.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:4d409aa42a94c0b3fa617708ef5276dfe81012ba6753a0370fcc9d0195d0a1fc", size = 76604, upload-time = "2025-10-06T14:48:54.277Z" },
+    { url = "https://files.pythonhosted.org/packages/17/e4/67b5c27bd17c085a5ea8f1ec05b8a3e5cba0ca734bfcad5560fb129e70ca/multidict-6.7.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:14c9e076eede3b54c636f8ce1c9c252b5f057c62131211f0ceeec273810c9721", size = 44715, upload-time = "2025-10-06T14:48:55.445Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/e1/866a5d77be6ea435711bef2a4291eed11032679b6b28b56b4776ab06ba3e/multidict-6.7.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:4c09703000a9d0fa3c3404b27041e574cc7f4df4c6563873246d0e11812a94b6", size = 44332, upload-time = "2025-10-06T14:48:56.706Z" },
+    { url = "https://files.pythonhosted.org/packages/31/61/0c2d50241ada71ff61a79518db85ada85fdabfcf395d5968dae1cbda04e5/multidict-6.7.0-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:a265acbb7bb33a3a2d626afbe756371dce0279e7b17f4f4eda406459c2b5ff1c", size = 245212, upload-time = "2025-10-06T14:48:58.042Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/e0/919666a4e4b57fff1b57f279be1c9316e6cdc5de8a8b525d76f6598fefc7/multidict-6.7.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:51cb455de290ae462593e5b1cb1118c5c22ea7f0d3620d9940bf695cea5a4bd7", size = 246671, upload-time = "2025-10-06T14:49:00.004Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/cc/d027d9c5a520f3321b65adea289b965e7bcbd2c34402663f482648c716ce/multidict-6.7.0-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:db99677b4457c7a5c5a949353e125ba72d62b35f74e26da141530fbb012218a7", size = 225491, upload-time = "2025-10-06T14:49:01.393Z" },
+    { url = "https://files.pythonhosted.org/packages/75/c4/bbd633980ce6155a28ff04e6a6492dd3335858394d7bb752d8b108708558/multidict-6.7.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f470f68adc395e0183b92a2f4689264d1ea4b40504a24d9882c27375e6662bb9", size = 257322, upload-time = "2025-10-06T14:49:02.745Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/6d/d622322d344f1f053eae47e033b0b3f965af01212de21b10bcf91be991fb/multidict-6.7.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0db4956f82723cc1c270de9c6e799b4c341d327762ec78ef82bb962f79cc07d8", size = 254694, upload-time = "2025-10-06T14:49:04.15Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/9f/78f8761c2705d4c6d7516faed63c0ebdac569f6db1bef95e0d5218fdc146/multidict-6.7.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3e56d780c238f9e1ae66a22d2adf8d16f485381878250db8d496623cd38b22bd", size = 246715, upload-time = "2025-10-06T14:49:05.967Z" },
+    { url = "https://files.pythonhosted.org/packages/78/59/950818e04f91b9c2b95aab3d923d9eabd01689d0dcd889563988e9ea0fd8/multidict-6.7.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:9d14baca2ee12c1a64740d4531356ba50b82543017f3ad6de0deb943c5979abb", size = 243189, upload-time = "2025-10-06T14:49:07.37Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/3d/77c79e1934cad2ee74991840f8a0110966d9599b3af95964c0cd79bb905b/multidict-6.7.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:295a92a76188917c7f99cda95858c822f9e4aae5824246bba9b6b44004ddd0a6", size = 237845, upload-time = "2025-10-06T14:49:08.759Z" },
+    { url = "https://files.pythonhosted.org/packages/63/1b/834ce32a0a97a3b70f86437f685f880136677ac00d8bce0027e9fd9c2db7/multidict-6.7.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:39f1719f57adbb767ef592a50ae5ebb794220d1188f9ca93de471336401c34d2", size = 246374, upload-time = "2025-10-06T14:49:10.574Z" },
+    { url = "https://files.pythonhosted.org/packages/23/ef/43d1c3ba205b5dec93dc97f3fba179dfa47910fc73aaaea4f7ceb41cec2a/multidict-6.7.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:0a13fb8e748dfc94749f622de065dd5c1def7e0d2216dba72b1d8069a389c6ff", size = 253345, upload-time = "2025-10-06T14:49:12.331Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/03/eaf95bcc2d19ead522001f6a650ef32811aa9e3624ff0ad37c445c7a588c/multidict-6.7.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:e3aa16de190d29a0ea1b48253c57d99a68492c8dd8948638073ab9e74dc9410b", size = 246940, upload-time = "2025-10-06T14:49:13.821Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/df/ec8a5fd66ea6cd6f525b1fcbb23511b033c3e9bc42b81384834ffa484a62/multidict-6.7.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a048ce45dcdaaf1defb76b2e684f997fb5abf74437b6cb7b22ddad934a964e34", size = 242229, upload-time = "2025-10-06T14:49:15.603Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/a2/59b405d59fd39ec86d1142630e9049243015a5f5291ba49cadf3c090c541/multidict-6.7.0-cp311-cp311-win32.whl", hash = "sha256:a90af66facec4cebe4181b9e62a68be65e45ac9b52b67de9eec118701856e7ff", size = 41308, upload-time = "2025-10-06T14:49:16.871Z" },
+    { url = "https://files.pythonhosted.org/packages/32/0f/13228f26f8b882c34da36efa776c3b7348455ec383bab4a66390e42963ae/multidict-6.7.0-cp311-cp311-win_amd64.whl", hash = "sha256:95b5ffa4349df2887518bb839409bcf22caa72d82beec453216802f475b23c81", size = 46037, upload-time = "2025-10-06T14:49:18.457Z" },
+    { url = "https://files.pythonhosted.org/packages/84/1f/68588e31b000535a3207fd3c909ebeec4fb36b52c442107499c18a896a2a/multidict-6.7.0-cp311-cp311-win_arm64.whl", hash = "sha256:329aa225b085b6f004a4955271a7ba9f1087e39dcb7e65f6284a988264a63912", size = 43023, upload-time = "2025-10-06T14:49:19.648Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/9e/9f61ac18d9c8b475889f32ccfa91c9f59363480613fc807b6e3023d6f60b/multidict-6.7.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:8a3862568a36d26e650a19bb5cbbba14b71789032aebc0423f8cc5f150730184", size = 76877, upload-time = "2025-10-06T14:49:20.884Z" },
+    { url = "https://files.pythonhosted.org/packages/38/6f/614f09a04e6184f8824268fce4bc925e9849edfa654ddd59f0b64508c595/multidict-6.7.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:960c60b5849b9b4f9dcc9bea6e3626143c252c74113df2c1540aebce70209b45", size = 45467, upload-time = "2025-10-06T14:49:22.054Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/93/c4f67a436dd026f2e780c433277fff72be79152894d9fc36f44569cab1a6/multidict-6.7.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:2049be98fb57a31b4ccf870bf377af2504d4ae35646a19037ec271e4c07998aa", size = 43834, upload-time = "2025-10-06T14:49:23.566Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/f5/013798161ca665e4a422afbc5e2d9e4070142a9ff8905e482139cd09e4d0/multidict-6.7.0-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:0934f3843a1860dd465d38895c17fce1f1cb37295149ab05cd1b9a03afacb2a7", size = 250545, upload-time = "2025-10-06T14:49:24.882Z" },
+    { url = "https://files.pythonhosted.org/packages/71/2f/91dbac13e0ba94669ea5119ba267c9a832f0cb65419aca75549fcf09a3dc/multidict-6.7.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b3e34f3a1b8131ba06f1a73adab24f30934d148afcd5f5de9a73565a4404384e", size = 258305, upload-time = "2025-10-06T14:49:26.778Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/b0/754038b26f6e04488b48ac621f779c341338d78503fb45403755af2df477/multidict-6.7.0-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:efbb54e98446892590dc2458c19c10344ee9a883a79b5cec4bc34d6656e8d546", size = 242363, upload-time = "2025-10-06T14:49:28.562Z" },
+    { url = "https://files.pythonhosted.org/packages/87/15/9da40b9336a7c9fa606c4cf2ed80a649dffeb42b905d4f63a1d7eb17d746/multidict-6.7.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:a35c5fc61d4f51eb045061e7967cfe3123d622cd500e8868e7c0c592a09fedc4", size = 268375, upload-time = "2025-10-06T14:49:29.96Z" },
+    { url = "https://files.pythonhosted.org/packages/82/72/c53fcade0cc94dfaad583105fd92b3a783af2091eddcb41a6d5a52474000/multidict-6.7.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:29fe6740ebccba4175af1b9b87bf553e9c15cd5868ee967e010efcf94e4fd0f1", size = 269346, upload-time = "2025-10-06T14:49:31.404Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/e2/9baffdae21a76f77ef8447f1a05a96ec4bc0a24dae08767abc0a2fe680b8/multidict-6.7.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:123e2a72e20537add2f33a79e605f6191fba2afda4cbb876e35c1a7074298a7d", size = 256107, upload-time = "2025-10-06T14:49:32.974Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/06/3f06f611087dc60d65ef775f1fb5aca7c6d61c6db4990e7cda0cef9b1651/multidict-6.7.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:b284e319754366c1aee2267a2036248b24eeb17ecd5dc16022095e747f2f4304", size = 253592, upload-time = "2025-10-06T14:49:34.52Z" },
+    { url = "https://files.pythonhosted.org/packages/20/24/54e804ec7945b6023b340c412ce9c3f81e91b3bf5fa5ce65558740141bee/multidict-6.7.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:803d685de7be4303b5a657b76e2f6d1240e7e0a8aa2968ad5811fa2285553a12", size = 251024, upload-time = "2025-10-06T14:49:35.956Z" },
+    { url = "https://files.pythonhosted.org/packages/14/48/011cba467ea0b17ceb938315d219391d3e421dfd35928e5dbdc3f4ae76ef/multidict-6.7.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:c04a328260dfd5db8c39538f999f02779012268f54614902d0afc775d44e0a62", size = 251484, upload-time = "2025-10-06T14:49:37.631Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/2f/919258b43bb35b99fa127435cfb2d91798eb3a943396631ef43e3720dcf4/multidict-6.7.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:8a19cdb57cd3df4cd865849d93ee14920fb97224300c88501f16ecfa2604b4e0", size = 263579, upload-time = "2025-10-06T14:49:39.502Z" },
+    { url = "https://files.pythonhosted.org/packages/31/22/a0e884d86b5242b5a74cf08e876bdf299e413016b66e55511f7a804a366e/multidict-6.7.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:9b2fd74c52accced7e75de26023b7dccee62511a600e62311b918ec5c168fc2a", size = 259654, upload-time = "2025-10-06T14:49:41.32Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/e5/17e10e1b5c5f5a40f2fcbb45953c9b215f8a4098003915e46a93f5fcaa8f/multidict-6.7.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:3e8bfdd0e487acf992407a140d2589fe598238eaeffa3da8448d63a63cd363f8", size = 251511, upload-time = "2025-10-06T14:49:46.021Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/9a/201bb1e17e7af53139597069c375e7b0dcbd47594604f65c2d5359508566/multidict-6.7.0-cp312-cp312-win32.whl", hash = "sha256:dd32a49400a2c3d52088e120ee00c1e3576cbff7e10b98467962c74fdb762ed4", size = 41895, upload-time = "2025-10-06T14:49:48.718Z" },
+    { url = "https://files.pythonhosted.org/packages/46/e2/348cd32faad84eaf1d20cce80e2bb0ef8d312c55bca1f7fa9865e7770aaf/multidict-6.7.0-cp312-cp312-win_amd64.whl", hash = "sha256:92abb658ef2d7ef22ac9f8bb88e8b6c3e571671534e029359b6d9e845923eb1b", size = 46073, upload-time = "2025-10-06T14:49:50.28Z" },
+    { url = "https://files.pythonhosted.org/packages/25/ec/aad2613c1910dce907480e0c3aa306905830f25df2e54ccc9dea450cb5aa/multidict-6.7.0-cp312-cp312-win_arm64.whl", hash = "sha256:490dab541a6a642ce1a9d61a4781656b346a55c13038f0b1244653828e3a83ec", size = 43226, upload-time = "2025-10-06T14:49:52.304Z" },
+    { url = "https://files.pythonhosted.org/packages/90/d7/4cf84257902265c4250769ac49f4eaab81c182ee9aff8bf59d2714dbb174/multidict-6.7.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:363eb68a0a59bd2303216d2346e6c441ba10d36d1f9969fcb6f1ba700de7bb5c", size = 77073, upload-time = "2025-10-06T14:51:57.386Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/51/194e999630a656e76c2965a1590d12faa5cd528170f2abaa04423e09fe8d/multidict-6.7.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:d874eb056410ca05fed180b6642e680373688efafc7f077b2a2f61811e873a40", size = 44928, upload-time = "2025-10-06T14:51:58.791Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/6b/2a195373c33068c9158e0941d0b46cfcc9c1d894ca2eb137d1128081dff0/multidict-6.7.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:8b55d5497b51afdfde55925e04a022f1de14d4f4f25cdfd4f5d9b0aa96166851", size = 44581, upload-time = "2025-10-06T14:52:00.174Z" },
+    { url = "https://files.pythonhosted.org/packages/69/7b/7f4f2e644b6978bf011a5fd9a5ebb7c21de3f38523b1f7897d36a1ac1311/multidict-6.7.0-cp39-cp39-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:f8e5c0031b90ca9ce555e2e8fd5c3b02a25f14989cbc310701823832c99eb687", size = 239901, upload-time = "2025-10-06T14:52:02.416Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/b5/952c72786710a031aa204a9adf7db66d7f97a2c6573889d58b9e60fe6702/multidict-6.7.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9cf41880c991716f3c7cec48e2f19ae4045fc9db5fc9cff27347ada24d710bb5", size = 240534, upload-time = "2025-10-06T14:52:04.105Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/ef/109fe1f2471e4c458c74242c7e4a833f2d9fc8a6813cd7ee345b0bad18f9/multidict-6.7.0-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:8cfc12a8630a29d601f48d47787bd7eb730e475e83edb5d6c5084317463373eb", size = 219545, upload-time = "2025-10-06T14:52:06.208Z" },
+    { url = "https://files.pythonhosted.org/packages/42/bd/327d91288114967f9fe90dc53de70aa3fec1b9073e46aa32c4828f771a87/multidict-6.7.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3996b50c3237c4aec17459217c1e7bbdead9a22a0fcd3c365564fbd16439dde6", size = 251187, upload-time = "2025-10-06T14:52:08.049Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/13/a8b078ebbaceb7819fd28cd004413c33b98f1b70d542a62e6a00b74fb09f/multidict-6.7.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:7f5170993a0dd3ab871c74f45c0a21a4e2c37a2f2b01b5f722a2ad9c6650469e", size = 249379, upload-time = "2025-10-06T14:52:09.831Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/6d/ab12e1246be4d65d1f55de1e6f6aaa9b8120eddcfdd1d290439c7833d5ce/multidict-6.7.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ec81878ddf0e98817def1e77d4f50dae5ef5b0e4fe796fae3bd674304172416e", size = 239241, upload-time = "2025-10-06T14:52:11.561Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/d7/079a93625208c173b8fa756396814397c0fd9fee61ef87b75a748820b86e/multidict-6.7.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:9281bf5b34f59afbc6b1e477a372e9526b66ca446f4bf62592839c195a718b32", size = 237418, upload-time = "2025-10-06T14:52:13.671Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/29/03777c2212274aa9440918d604dc9d6af0e6b4558c611c32c3dcf1a13870/multidict-6.7.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:68af405971779d8b37198726f2b6fe3955db846fee42db7a4286fc542203934c", size = 232987, upload-time = "2025-10-06T14:52:15.708Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/00/11188b68d85a84e8050ee34724d6ded19ad03975caebe0c8dcb2829b37bf/multidict-6.7.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:3ba3ef510467abb0667421a286dc906e30eb08569365f5cdb131d7aff7c2dd84", size = 240985, upload-time = "2025-10-06T14:52:17.317Z" },
+    { url = "https://files.pythonhosted.org/packages/df/0c/12eef6aeda21859c6cdf7d75bd5516d83be3efe3d8cc45fd1a3037f5b9dc/multidict-6.7.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:b61189b29081a20c7e4e0b49b44d5d44bb0dc92be3c6d06a11cc043f81bf9329", size = 246855, upload-time = "2025-10-06T14:52:19.096Z" },
+    { url = "https://files.pythonhosted.org/packages/69/f6/076120fd8bb3975f09228e288e08bff6b9f1bfd5166397c7ba284f622ab2/multidict-6.7.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:fb287618b9c7aa3bf8d825f02d9201b2f13078a5ed3b293c8f4d953917d84d5e", size = 241804, upload-time = "2025-10-06T14:52:21.166Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/51/41bb950c81437b88a93e6ddfca1d8763569ae861e638442838c4375f7497/multidict-6.7.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:521f33e377ff64b96c4c556b81c55d0cfffb96a11c194fd0c3f1e56f3d8dd5a4", size = 235321, upload-time = "2025-10-06T14:52:23.208Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/cf/5bbd31f055199d56c1f6b04bbadad3ccb24e6d5d4db75db774fc6d6674b8/multidict-6.7.0-cp39-cp39-win32.whl", hash = "sha256:ce8fdc2dca699f8dbf055a61d73eaa10482569ad20ee3c36ef9641f69afa8c91", size = 41435, upload-time = "2025-10-06T14:52:24.735Z" },
+    { url = "https://files.pythonhosted.org/packages/af/01/547ffe9c2faec91c26965c152f3fea6cff068b6037401f61d310cc861ff4/multidict-6.7.0-cp39-cp39-win_amd64.whl", hash = "sha256:7e73299c99939f089dd9b2120a04a516b95cdf8c1cd2b18c53ebf0de80b1f18f", size = 46193, upload-time = "2025-10-06T14:52:26.101Z" },
+    { url = "https://files.pythonhosted.org/packages/27/77/cfa5461d1d2651d6fc24216c92b4a21d4e385a41c46e0d9f3b070675167b/multidict-6.7.0-cp39-cp39-win_arm64.whl", hash = "sha256:6bdce131e14b04fd34a809b6380dbfd826065c3e2fe8a50dbae659fa0c390546", size = 43118, upload-time = "2025-10-06T14:52:27.876Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/da/7d22601b625e241d4f23ef1ebff8acfc60da633c9e7e7922e24d10f592b3/multidict-6.7.0-py3-none-any.whl", hash = "sha256:394fc5c42a333c9ffc3e421a4c85e08580d990e08b99f6bf35b4132114c5dcb3", size = 12317, upload-time = "2025-10-06T14:52:29.272Z" },
 ]
 
 [[package]]
@@ -2060,25 +2507,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/eb/8d/776adee7bbf76365fdd7f2552710282c79a4ead5d2a46408c9043a2b70ba/networkx-3.5-py3-none-any.whl", hash = "sha256:0030d386a9a06dee3565298b4a734b68589749a544acbb6c412dc9e2489ec6ec", size = 2034406, upload-time = "2025-05-29T11:35:04.961Z" },
 ]
 
-[[package]]
-name = "nh3"
-version = "0.3.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/c3/a4/96cff0977357f60f06ec4368c4c7a7a26cccfe7c9fcd54f5378bf0428fd3/nh3-0.3.0.tar.gz", hash = "sha256:d8ba24cb31525492ea71b6aac11a4adac91d828aadeff7c4586541bf5dc34d2f", size = 19655, upload-time = "2025-07-17T14:43:37.05Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/5c/86/a96b1453c107b815f9ab8fac5412407c33cc5c7580a4daf57aabeb41b774/nh3-0.3.0-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ce5e7185599f89b0e391e2f29cc12dc2e206167380cea49b33beda4891be2fe1", size = 823791, upload-time = "2025-07-17T14:43:19.721Z" },
-    { url = "https://files.pythonhosted.org/packages/97/33/11e7273b663839626f714cb68f6eb49899da5a0d9b6bc47b41fe870259c2/nh3-0.3.0-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:389d93d59b8214d51c400fb5b07866c2a4f79e4e14b071ad66c92184fec3a392", size = 811143, upload-time = "2025-07-17T14:43:20.779Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/1b/b15bd1ce201a1a610aeb44afd478d55ac018b4475920a3118ffd806e2483/nh3-0.3.0-cp38-abi3-manylinux_2_17_ppc64.manylinux2014_ppc64.whl", hash = "sha256:e9e6a7e4d38f7e8dda9edd1433af5170c597336c1a74b4693c5cb75ab2b30f2a", size = 1064661, upload-time = "2025-07-17T14:43:21.839Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/14/079670fb2e848c4ba2476c5a7a2d1319826053f4f0368f61fca9bb4227ae/nh3-0.3.0-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7852f038a054e0096dac12b8141191e02e93e0b4608c4b993ec7d4ffafea4e49", size = 997061, upload-time = "2025-07-17T14:43:23.179Z" },
-    { url = "https://files.pythonhosted.org/packages/a3/e5/ac7fc565f5d8bce7f979d1afd68e8cb415020d62fa6507133281c7d49f91/nh3-0.3.0-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:af5aa8127f62bbf03d68f67a956627b1bd0469703a35b3dad28d0c1195e6c7fb", size = 924761, upload-time = "2025-07-17T14:43:24.23Z" },
-    { url = "https://files.pythonhosted.org/packages/39/2c/6394301428b2017a9d5644af25f487fa557d06bc8a491769accec7524d9a/nh3-0.3.0-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f416c35efee3e6a6c9ab7716d9e57aa0a49981be915963a82697952cba1353e1", size = 803959, upload-time = "2025-07-17T14:43:26.377Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/9a/344b9f9c4bd1c2413a397f38ee6a3d5db30f1a507d4976e046226f12b297/nh3-0.3.0-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:37d3003d98dedca6cd762bf88f2e70b67f05100f6b949ffe540e189cc06887f9", size = 844073, upload-time = "2025-07-17T14:43:27.375Z" },
-    { url = "https://files.pythonhosted.org/packages/66/3f/cd37f76c8ca277b02a84aa20d7bd60fbac85b4e2cbdae77cb759b22de58b/nh3-0.3.0-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:634e34e6162e0408e14fb61d5e69dbaea32f59e847cfcfa41b66100a6b796f62", size = 1000680, upload-time = "2025-07-17T14:43:28.452Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/db/7aa11b44bae4e7474feb1201d8dee04fabe5651c7cb51409ebda94a4ed67/nh3-0.3.0-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:b0612ccf5de8a480cf08f047b08f9d3fecc12e63d2ee91769cb19d7290614c23", size = 1076613, upload-time = "2025-07-17T14:43:30.031Z" },
-    { url = "https://files.pythonhosted.org/packages/97/03/03f79f7e5178eb1ad5083af84faff471e866801beb980cc72943a4397368/nh3-0.3.0-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:c7a32a7f0d89f7d30cb8f4a84bdbd56d1eb88b78a2434534f62c71dac538c450", size = 1001418, upload-time = "2025-07-17T14:43:31.429Z" },
-    { url = "https://files.pythonhosted.org/packages/ce/55/1974bcc16884a397ee699cebd3914e1f59be64ab305533347ca2d983756f/nh3-0.3.0-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:3f1b4f8a264a0c86ea01da0d0c390fe295ea0bcacc52c2103aca286f6884f518", size = 986499, upload-time = "2025-07-17T14:43:32.459Z" },
-]
-
 [[package]]
 name = "ninja"
 version = "1.13.0"
@@ -2411,20 +2839,20 @@ wheels = [
 
 [[package]]
 name = "ollama"
-version = "0.5.3"
+version = "0.6.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "httpx" },
     { name = "pydantic" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/91/6d/ae96027416dcc2e98c944c050c492789502d7d7c0b95a740f0bb39268632/ollama-0.5.3.tar.gz", hash = "sha256:40b6dff729df3b24e56d4042fd9d37e231cee8e528677e0d085413a1d6692394", size = 43331, upload-time = "2025-08-07T21:44:10.422Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/d6/47/f9ee32467fe92744474a8c72e138113f3b529fc266eea76abfdec9a33f3b/ollama-0.6.0.tar.gz", hash = "sha256:da2b2d846b5944cfbcee1ca1e6ee0585f6c9d45a2fe9467cbcd096a37383da2f", size = 50811, upload-time = "2025-09-24T22:46:02.417Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/be/f6/2091e50b8b6c3e6901f6eab283d5efd66fb71c86ddb1b4d68766c3eeba0f/ollama-0.5.3-py3-none-any.whl", hash = "sha256:a8303b413d99a9043dbf77ebf11ced672396b59bec27e6d5db67c88f01b279d2", size = 13490, upload-time = "2025-08-07T21:44:09.353Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/c1/edc9f41b425ca40b26b7c104c5f6841a4537bb2552bfa6ca66e81405bb95/ollama-0.6.0-py3-none-any.whl", hash = "sha256:534511b3ccea2dff419ae06c3b58d7f217c55be7897c8ce5868dfb6b219cf7a0", size = 14130, upload-time = "2025-09-24T22:46:01.19Z" },
 ]
 
 [[package]]
 name = "openai"
-version = "1.102.0"
+version = "2.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
@@ -2436,9 +2864,43 @@ dependencies = [
     { name = "tqdm" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/07/55/da5598ed5c6bdd9939633854049cddc5cbac0da938dfcfcb3c6b119c16c0/openai-1.102.0.tar.gz", hash = "sha256:2e0153bcd64a6523071e90211cbfca1f2bbc5ceedd0993ba932a5869f93b7fc9", size = 519027, upload-time = "2025-08-26T20:50:29.397Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/b8/b1/8201e321a7d64a25c6f5a560320272d8be70547add40311fceb916518632/openai-2.2.0.tar.gz", hash = "sha256:bc49d077a8bf0e370eec4d038bc05e232c20855a19df0b58e5b3e5a8da7d33e0", size = 588512, upload-time = "2025-10-06T18:08:13.665Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/bd/0d/c9e7016d82c53c5b5e23e2bad36daebb8921ed44f69c0a985c6529a35106/openai-1.102.0-py3-none-any.whl", hash = "sha256:d751a7e95e222b5325306362ad02a7aa96e1fab3ed05b5888ce1c7ca63451345", size = 812015, upload-time = "2025-08-26T20:50:27.219Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/92/6aeef1836e66dfec7f7f160a4f06d7041be7f6ccfc47a2f0f5738b332245/openai-2.2.0-py3-none-any.whl", hash = "sha256:d222e63436e33f3134a3d7ce490dc2d2f146fa98036eb65cc225df3ce163916f", size = 998972, upload-time = "2025-10-06T18:08:11.775Z" },
+]
+
+[[package]]
+name = "openai-harmony"
+version = "0.0.4"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/92/94/01509d510bebf6606614e51113e5a415ced15b8f34aa98a8bf2539314650/openai_harmony-0.0.4.tar.gz", hash = "sha256:5c67ac6df349236fb7b64f57c3dbb0273efcdca24314daa108f2a482c427106c", size = 279848, upload-time = "2025-08-09T01:43:24.974Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/05/88/ade63bd8f36603610040e7cc086bc134d57a99a742e05f7fcddfdf822ee1/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3cf2344366f10981bbc0f6d9949a0b2bb87151d209ed295943ed6ad8eda37932", size = 2963206, upload-time = "2025-08-09T01:43:02.433Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/ef/a65a0ff177fdf67bc0afd18bb9e7ad690d1b553a8eb5ebf27f601b22dbd0/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:2d8d16d84702059833fb03b841b28c25600c54e83cadccef79af44e1c81166b1", size = 2724854, upload-time = "2025-08-09T01:43:04.606Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/a1/ebaf0f55601a98609641283884d52dbfe9a1cf34b04f1cf80acb1560ab74/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:97f1fe3909733212cc6b36f0f199b1421a9c57b79ec665f0322bd604cec47340", size = 2984312, upload-time = "2025-08-09T01:43:08.908Z" },
+    { url = "https://files.pythonhosted.org/packages/45/24/246f6f470bfbc89a117714b68f27cdaee12b31166237a227cc657780cc1d/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:567cc568b6bf7b4d041b0c9aa7d6b2c9394f8af6065bc87fa6d23f207b5af9a7", size = 3447870, upload-time = "2025-08-09T01:43:06.734Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/ec/dcdcace0ffcf3a532cca910e0c351b62d3a7decf0b091ea8cf856d2a67a6/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:31e9bcac0902a309e2fc688e52f247eec7fffcd00d17e958b9a83a8fea6519c2", size = 3049306, upload-time = "2025-08-09T01:43:11.019Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/39/172f1048d935db1523a82b45fee5231ad6c622645e566706e6bcf3731da8/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:96a63199c0d81095b5d5d1ae8ca82b64c1c13d18d4e30323ae9e8ab31bc80a3d", size = 3121347, upload-time = "2025-08-09T01:43:16.705Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/36/8ee4ca5d0b25587121fd3621e6a6106fba80218cb6d159e1670aeb2b22ef/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:d38f2639f6bf7c3c34a5dfd79e29075811ae2fa9b895a63e76767f74a47a971e", size = 2952326, upload-time = "2025-08-09T01:43:18.841Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/a0/ec8906393968679e269e23e957e11ff419978d1d077fb9af9561b161c988/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:038f1d6772d1be5213b36ae76e5d042022395ec35c428a73ccb8b839b2cecf6a", size = 3015832, upload-time = "2025-08-09T01:43:21.076Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/bd/aa9e6e5cf140716dbcae17402fac2a81a9ebb3f934059ac0eec61cb447fc/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:15e6d53a66502491a3675a536df30e271f976e6c5efe68250a65191efcb85c4f", size = 3221129, upload-time = "2025-08-09T01:43:23.146Z" },
+]
+
+[[package]]
+name = "opencv-python-headless"
+version = "4.12.0.88"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a4/63/6861102ec149c3cd298f4d1ea7ce9d6adbc7529221606ff1dab991a19adb/opencv-python-headless-4.12.0.88.tar.gz", hash = "sha256:cfdc017ddf2e59b6c2f53bc12d74b6b0be7ded4ec59083ea70763921af2b6c09", size = 95379675, upload-time = "2025-07-07T09:21:06.815Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/69/4e/116720df7f1f7f3b59abc608ca30fbec9d2b3ae810afe4e4d26483d9dfa0/opencv_python_headless-4.12.0.88-cp37-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:aeb4b13ecb8b4a0beb2668ea07928160ea7c2cd2d9b5ef571bbee6bafe9cc8d0", size = 33145800, upload-time = "2025-07-07T09:15:00.367Z" },
+    { url = "https://files.pythonhosted.org/packages/89/53/e19c21e0c4eb1275c3e2c97b081103b6dfb3938172264d283a519bf728b9/opencv_python_headless-4.12.0.88-cp37-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:236c8df54a90f4d02076e6f9c1cc763d794542e886c576a6fee46ec8ff75a7a9", size = 54023419, upload-time = "2025-07-07T09:15:10.164Z" },
 ]
 
 [[package]]
@@ -2472,7 +2934,7 @@ wheels = [
 
 [[package]]
 name = "optax"
-version = "0.2.5"
+version = "0.2.6"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -2484,14 +2946,14 @@ dependencies = [
     { name = "chex", version = "0.1.90", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "chex", version = "0.1.91", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "jaxlib", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jaxlib", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/c0/75/1e011953c48be502d4d84fa8458e91be7c6f983002511669bddd7b1a065f/optax-0.2.5.tar.gz", hash = "sha256:b2e38c7aea376186deae758ba7a258e6ef760c6f6131e9e11bc561c65386d594", size = 258548, upload-time = "2025-06-10T17:00:47.544Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/6d/3b/90c11f740a3538200b61cd2b7d9346959cb9e31e0bdea3d2f886b7262203/optax-0.2.6.tar.gz", hash = "sha256:ba8d1e12678eba2657484d6feeca4fb281b8066bdfd5efbfc0f41b87663109c0", size = 269660, upload-time = "2025-09-15T22:41:24.76Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b9/33/f86091c706db1a5459f501830241afff2ecab3532725c188ea57be6e54de/optax-0.2.5-py3-none-any.whl", hash = "sha256:966deae936207f268ac8f564d8ed228d645ac1aaddefbbf194096d2299b24ba8", size = 354324, upload-time = "2025-06-10T17:00:46.062Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/ec/19c6cc6064c7fc8f0cd6d5b37c4747849e66040c6ca98f86565efc2c227c/optax-0.2.6-py3-none-any.whl", hash = "sha256:f875251a5ab20f179d4be57478354e8e21963373b10f9c3b762b94dcb8c36d91", size = 367782, upload-time = "2025-09-15T22:41:22.825Z" },
 ]
 
 [[package]]
@@ -2581,7 +3043,7 @@ wheels = [
 
 [[package]]
 name = "orbax-checkpoint"
-version = "0.11.24"
+version = "0.11.25"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -2594,19 +3056,19 @@ dependencies = [
     { name = "etils", version = "1.13.0", source = { registry = "https://pypi.org/simple" }, extra = ["epath", "epy"], marker = "python_full_version >= '3.10'" },
     { name = "humanize", marker = "python_full_version >= '3.10'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "msgpack", marker = "python_full_version >= '3.10'" },
     { name = "nest-asyncio", marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "protobuf", marker = "python_full_version >= '3.10'" },
     { name = "pyyaml", marker = "python_full_version >= '3.10'" },
     { name = "simplejson", marker = "python_full_version >= '3.10'" },
-    { name = "tensorstore", version = "0.1.76", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "tensorstore", version = "0.1.78", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/52/24/44915f33cbea4cfd35f654a5ba01d248447bc007f1d049dd20bf58592820/orbax_checkpoint-0.11.24.tar.gz", hash = "sha256:4e7afe927d1ed6d8160bacf5ed4fef56c1370320e0ebdfda213c6351a2e3c0d0", size = 372123, upload-time = "2025-08-28T20:51:49.089Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/00/44/63d4a064fddc9f75cf622427f01333ed109096b6af8db039c00c12b2a85b/orbax_checkpoint-0.11.25.tar.gz", hash = "sha256:15d35e4ee165aad8edccc25bc462a71c921bdaecc79ce0a2a5a0f784fce091be", size = 389724, upload-time = "2025-09-11T17:40:50.771Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/99/e8/acef62ec5c4b8658eacfce23d3f4e866d3cf5d07c509ca03eebb0c420a6c/orbax_checkpoint-0.11.24-py3-none-any.whl", hash = "sha256:a94178c9ba9fd3d6fd8fc511b6a0f34f7d89798bfdb79661e258cd32ada7650b", size = 529268, upload-time = "2025-08-28T20:51:47.917Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/96/510dff16d2b75d9debb77fcfc4e6a4749059024a290953032f3ac793f74a/orbax_checkpoint-0.11.25-py3-none-any.whl", hash = "sha256:af8f5d16f1185c54ead7859309c57de28bf3cc9b4928ebd5a526e3ba549256c4", size = 563103, upload-time = "2025-09-11T17:40:49.368Z" },
 ]
 
 [[package]]
@@ -2650,6 +3112,9 @@ llamacpp = [
 llguidance = [
     { name = "llguidance" },
 ]
+mistral = [
+    { name = "mistralai" },
+]
 mlxlm = [
     { name = "datasets" },
     { name = "mlx" },
@@ -2675,59 +3140,16 @@ test = [
     { name = "dottxt" },
     { name = "flax", version = "0.8.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "flax", version = "0.10.7", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "flax", version = "0.11.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "google-genai" },
-    { name = "huggingface-hub" },
-    { name = "iso3166" },
-    { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "llama-cpp-python" },
-    { name = "llguidance" },
-    { name = "mkdocs-gen-files" },
-    { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
-    { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "ollama" },
-    { name = "openai" },
-    { name = "pillow" },
-    { name = "pre-commit" },
-    { name = "pytest" },
-    { name = "pytest-asyncio" },
-    { name = "pytest-benchmark" },
-    { name = "pytest-cov" },
-    { name = "pytest-mock" },
-    { name = "requests" },
-    { name = "responses" },
-    { name = "sentencepiece" },
-    { name = "tensorflow" },
-    { name = "tf-keras" },
-    { name = "torch" },
-    { name = "transformers" },
-    { name = "xgrammar" },
-]
-test-gpu = [
-    { name = "accelerate" },
-    { name = "airportsdata" },
-    { name = "anthropic" },
-    { name = "beartype" },
-    { name = "coverage", extra = ["toml"] },
-    { name = "datasets" },
-    { name = "diff-cover" },
-    { name = "dottxt" },
-    { name = "flax", version = "0.8.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "flax", version = "0.10.7", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "flax", version = "0.11.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "flax", version = "0.12.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "google-genai" },
     { name = "huggingface-hub" },
     { name = "iso3166" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "llama-cpp-python" },
     { name = "llguidance" },
+    { name = "mistralai" },
     { name = "mkdocs-gen-files" },
     { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
     { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
@@ -2750,7 +3172,6 @@ test-gpu = [
     { name = "tf-keras" },
     { name = "torch" },
     { name = "transformers" },
-    { name = "vllm", marker = "sys_platform == 'linux'" },
     { name = "xgrammar" },
 ]
 tgi = [
@@ -2769,6 +3190,12 @@ xgrammar = [
     { name = "xgrammar" },
 ]
 
+[package.dev-dependencies]
+test-gpu = [
+    { name = "outlines", extra = ["test"] },
+    { name = "vllm", marker = "sys_platform == 'linux'" },
+]
+
 [package.metadata]
 requires-dist = [
     { name = "accelerate", marker = "extra == 'test'" },
@@ -2804,6 +3231,8 @@ requires-dist = [
     { name = "llama-cpp-python", marker = "extra == 'test'" },
     { name = "llguidance", marker = "extra == 'llguidance'" },
     { name = "llguidance", marker = "extra == 'test'" },
+    { name = "mistralai", marker = "extra == 'mistral'" },
+    { name = "mistralai", marker = "extra == 'test'" },
     { name = "mkdocs-gen-files", marker = "extra == 'test'" },
     { name = "mlx", marker = "extra == 'mlxlm'" },
     { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin' and extra == 'test'", specifier = ">=0.19.2" },
@@ -2817,7 +3246,6 @@ requires-dist = [
     { name = "openai", marker = "extra == 'sglang'" },
     { name = "openai", marker = "extra == 'test'", specifier = ">=1.0.0" },
     { name = "openai", marker = "extra == 'vllm'" },
-    { name = "outlines", extras = ["test"], marker = "extra == 'test-gpu'" },
     { name = "outlines-core", specifier = "==0.2.11" },
     { name = "pillow" },
     { name = "pillow", marker = "extra == 'test'" },
@@ -2838,11 +3266,16 @@ requires-dist = [
     { name = "transformers", marker = "extra == 'test'" },
     { name = "transformers", marker = "extra == 'transformers'" },
     { name = "typing-extensions" },
-    { name = "vllm", marker = "sys_platform == 'linux' and extra == 'test-gpu'" },
     { name = "xgrammar", marker = "extra == 'test'" },
     { name = "xgrammar", marker = "extra == 'xgrammar'" },
 ]
-provides-extras = ["anthropic", "dottxt", "gemini", "llamacpp", "mlxlm", "ollama", "openai", "sglang", "tgi", "transformers", "vllm", "xgrammar", "llguidance", "airports", "countries", "test", "test-gpu"]
+provides-extras = ["anthropic", "dottxt", "gemini", "llamacpp", "mlxlm", "ollama", "openai", "mistral", "sglang", "tgi", "transformers", "vllm", "xgrammar", "llguidance", "airports", "countries", "test"]
+
+[package.metadata.requires-dev]
+test-gpu = [
+    { name = "outlines", extras = ["test"] },
+    { name = "vllm", marker = "sys_platform == 'linux'" },
+]
 
 [[package]]
 name = "outlines-core"
@@ -2895,7 +3328,7 @@ wheels = [
 
 [[package]]
 name = "pandas"
-version = "2.3.2"
+version = "2.3.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
@@ -2904,36 +3337,45 @@ dependencies = [
     { name = "pytz" },
     { name = "tzdata" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/79/8e/0e90233ac205ad182bd6b422532695d2b9414944a280488105d598c70023/pandas-2.3.2.tar.gz", hash = "sha256:ab7b58f8f82706890924ccdfb5f48002b83d2b5a3845976a9fb705d36c34dcdb", size = 4488684, upload-time = "2025-08-21T10:28:29.257Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/2e/16/a8eeb70aad84ccbf14076793f90e0031eded63c1899aeae9fdfbf37881f4/pandas-2.3.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:52bc29a946304c360561974c6542d1dd628ddafa69134a7131fdfd6a5d7a1a35", size = 11539648, upload-time = "2025-08-21T10:26:36.236Z" },
-    { url = "https://files.pythonhosted.org/packages/47/f1/c5bdaea13bf3708554d93e948b7ea74121ce6e0d59537ca4c4f77731072b/pandas-2.3.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:220cc5c35ffaa764dd5bb17cf42df283b5cb7fdf49e10a7b053a06c9cb48ee2b", size = 10786923, upload-time = "2025-08-21T10:26:40.518Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/10/811fa01476d29ffed692e735825516ad0e56d925961819e6126b4ba32147/pandas-2.3.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:42c05e15111221384019897df20c6fe893b2f697d03c811ee67ec9e0bb5a3424", size = 11726241, upload-time = "2025-08-21T10:26:43.175Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/6a/40b043b06e08df1ea1b6d20f0e0c2f2c4ec8c4f07d1c92948273d943a50b/pandas-2.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cc03acc273c5515ab69f898df99d9d4f12c4d70dbfc24c3acc6203751d0804cf", size = 12349533, upload-time = "2025-08-21T10:26:46.611Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/ea/2e081a2302e41a9bca7056659fdd2b85ef94923723e41665b42d65afd347/pandas-2.3.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d25c20a03e8870f6339bcf67281b946bd20b86f1a544ebbebb87e66a8d642cba", size = 13202407, upload-time = "2025-08-21T10:26:49.068Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/12/7ff9f6a79e2ee8869dcf70741ef998b97ea20050fe25f83dc759764c1e32/pandas-2.3.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:21bb612d148bb5860b7eb2c10faacf1a810799245afd342cf297d7551513fbb6", size = 13837212, upload-time = "2025-08-21T10:26:51.832Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/df/5ab92fcd76455a632b3db34a746e1074d432c0cdbbd28d7cd1daba46a75d/pandas-2.3.2-cp310-cp310-win_amd64.whl", hash = "sha256:b62d586eb25cb8cb70a5746a378fc3194cb7f11ea77170d59f889f5dfe3cec7a", size = 11338099, upload-time = "2025-08-21T10:26:54.382Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/59/f3e010879f118c2d400902d2d871c2226cef29b08c09fb8dc41111730400/pandas-2.3.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:1333e9c299adcbb68ee89a9bb568fc3f20f9cbb419f1dd5225071e6cddb2a743", size = 11563308, upload-time = "2025-08-21T10:26:56.656Z" },
-    { url = "https://files.pythonhosted.org/packages/38/18/48f10f1cc5c397af59571d638d211f494dba481f449c19adbd282aa8f4ca/pandas-2.3.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:76972bcbd7de8e91ad5f0ca884a9f2c477a2125354af624e022c49e5bd0dfff4", size = 10820319, upload-time = "2025-08-21T10:26:59.162Z" },
-    { url = "https://files.pythonhosted.org/packages/95/3b/1e9b69632898b048e223834cd9702052bcf06b15e1ae716eda3196fb972e/pandas-2.3.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b98bdd7c456a05eef7cd21fd6b29e3ca243591fe531c62be94a2cc987efb5ac2", size = 11790097, upload-time = "2025-08-21T10:27:02.204Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/ef/0e2ffb30b1f7fbc9a588bd01e3c14a0d96854d09a887e15e30cc19961227/pandas-2.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1d81573b3f7db40d020983f78721e9bfc425f411e616ef019a10ebf597aedb2e", size = 12397958, upload-time = "2025-08-21T10:27:05.409Z" },
-    { url = "https://files.pythonhosted.org/packages/23/82/e6b85f0d92e9afb0e7f705a51d1399b79c7380c19687bfbf3d2837743249/pandas-2.3.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:e190b738675a73b581736cc8ec71ae113d6c3768d0bd18bffa5b9a0927b0b6ea", size = 13225600, upload-time = "2025-08-21T10:27:07.791Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/f1/f682015893d9ed51611948bd83683670842286a8edd4f68c2c1c3b231eef/pandas-2.3.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:c253828cb08f47488d60f43c5fc95114c771bbfff085da54bfc79cb4f9e3a372", size = 13879433, upload-time = "2025-08-21T10:27:10.347Z" },
-    { url = "https://files.pythonhosted.org/packages/a7/e7/ae86261695b6c8a36d6a4c8d5f9b9ede8248510d689a2f379a18354b37d7/pandas-2.3.2-cp311-cp311-win_amd64.whl", hash = "sha256:9467697b8083f9667b212633ad6aa4ab32436dcbaf4cd57325debb0ddef2012f", size = 11336557, upload-time = "2025-08-21T10:27:12.983Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/db/614c20fb7a85a14828edd23f1c02db58a30abf3ce76f38806155d160313c/pandas-2.3.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:3fbb977f802156e7a3f829e9d1d5398f6192375a3e2d1a9ee0803e35fe70a2b9", size = 11587652, upload-time = "2025-08-21T10:27:15.888Z" },
-    { url = "https://files.pythonhosted.org/packages/99/b0/756e52f6582cade5e746f19bad0517ff27ba9c73404607c0306585c201b3/pandas-2.3.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1b9b52693123dd234b7c985c68b709b0b009f4521000d0525f2b95c22f15944b", size = 10717686, upload-time = "2025-08-21T10:27:18.486Z" },
-    { url = "https://files.pythonhosted.org/packages/37/4c/dd5ccc1e357abfeee8353123282de17997f90ff67855f86154e5a13b81e5/pandas-2.3.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0bd281310d4f412733f319a5bc552f86d62cddc5f51d2e392c8787335c994175", size = 11278722, upload-time = "2025-08-21T10:27:21.149Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/a4/f7edcfa47e0a88cda0be8b068a5bae710bf264f867edfdf7b71584ace362/pandas-2.3.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:96d31a6b4354e3b9b8a2c848af75d31da390657e3ac6f30c05c82068b9ed79b9", size = 11987803, upload-time = "2025-08-21T10:27:23.767Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/61/1bce4129f93ab66f1c68b7ed1c12bac6a70b1b56c5dab359c6bbcd480b52/pandas-2.3.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:df4df0b9d02bb873a106971bb85d448378ef14b86ba96f035f50bbd3688456b4", size = 12766345, upload-time = "2025-08-21T10:27:26.6Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/46/80d53de70fee835531da3a1dae827a1e76e77a43ad22a8cd0f8142b61587/pandas-2.3.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:213a5adf93d020b74327cb2c1b842884dbdd37f895f42dcc2f09d451d949f811", size = 13439314, upload-time = "2025-08-21T10:27:29.213Z" },
-    { url = "https://files.pythonhosted.org/packages/28/30/8114832daff7489f179971dbc1d854109b7f4365a546e3ea75b6516cea95/pandas-2.3.2-cp312-cp312-win_amd64.whl", hash = "sha256:8c13b81a9347eb8c7548f53fd9a4f08d4dfe996836543f805c987bafa03317ae", size = 10983326, upload-time = "2025-08-21T10:27:31.901Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/c3/b37e090d0aceda9b4dd85c8dbd1bea65b1de9e7a4f690d6bd3a40bd16390/pandas-2.3.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:88080a0ff8a55eac9c84e3ff3c7665b3b5476c6fbc484775ca1910ce1c3e0b87", size = 11551511, upload-time = "2025-08-21T10:28:11.111Z" },
-    { url = "https://files.pythonhosted.org/packages/b9/47/381fb1e7adcfcf4230fa6dc3a741acbac6c6fe072f19f4e7a46bddf3e5f6/pandas-2.3.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:d4a558c7620340a0931828d8065688b3cc5b4c8eb674bcaf33d18ff4a6870b4a", size = 10797930, upload-time = "2025-08-21T10:28:13.436Z" },
-    { url = "https://files.pythonhosted.org/packages/36/ca/d42467829080b92fc46d451288af8068f129fbcfb6578d573f45120de5cf/pandas-2.3.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:45178cf09d1858a1509dc73ec261bf5b25a625a389b65be2e47b559905f0ab6a", size = 11738470, upload-time = "2025-08-21T10:28:16.065Z" },
-    { url = "https://files.pythonhosted.org/packages/60/76/7d0f0a0deed7867c51163982d7b79c0a089096cd7ad50e1b87c2c82220e9/pandas-2.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:77cefe00e1b210f9c76c697fedd8fdb8d3dd86563e9c8adc9fa72b90f5e9e4c2", size = 12366640, upload-time = "2025-08-21T10:28:18.557Z" },
-    { url = "https://files.pythonhosted.org/packages/21/31/56784743e421cf51e34358fe7e5954345e5942168897bf8eb5707b71eedb/pandas-2.3.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:13bd629c653856f00c53dc495191baa59bcafbbf54860a46ecc50d3a88421a96", size = 13211567, upload-time = "2025-08-21T10:28:20.998Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/4e/50a399dc7d9dd4aa09a03b163751d428026cf0f16c419b4010f6aca26ebd/pandas-2.3.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:36d627906fd44b5fd63c943264e11e96e923f8de77d6016dc2f667b9ad193438", size = 13854073, upload-time = "2025-08-21T10:28:24.056Z" },
-    { url = "https://files.pythonhosted.org/packages/29/72/8978a84861a5124e56ce1048376569545412501fcb9a83f035393d6d85bc/pandas-2.3.2-cp39-cp39-win_amd64.whl", hash = "sha256:a9d7ec92d71a420185dec44909c32e9a362248c4ae2238234b76d5be37f208cc", size = 11346452, upload-time = "2025-08-21T10:28:26.691Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/33/01/d40b85317f86cf08d853a4f495195c73815fdf205eef3993821720274518/pandas-2.3.3.tar.gz", hash = "sha256:e05e1af93b977f7eafa636d043f9f94c7ee3ac81af99c13508215942e64c993b", size = 4495223, upload-time = "2025-09-29T23:34:51.853Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3d/f7/f425a00df4fcc22b292c6895c6831c0c8ae1d9fac1e024d16f98a9ce8749/pandas-2.3.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:376c6446ae31770764215a6c937f72d917f214b43560603cd60da6408f183b6c", size = 11555763, upload-time = "2025-09-29T23:16:53.287Z" },
+    { url = "https://files.pythonhosted.org/packages/13/4f/66d99628ff8ce7857aca52fed8f0066ce209f96be2fede6cef9f84e8d04f/pandas-2.3.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e19d192383eab2f4ceb30b412b22ea30690c9e618f78870357ae1d682912015a", size = 10801217, upload-time = "2025-09-29T23:17:04.522Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/03/3fc4a529a7710f890a239cc496fc6d50ad4a0995657dccc1d64695adb9f4/pandas-2.3.3-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5caf26f64126b6c7aec964f74266f435afef1c1b13da3b0636c7518a1fa3e2b1", size = 12148791, upload-time = "2025-09-29T23:17:18.444Z" },
+    { url = "https://files.pythonhosted.org/packages/40/a8/4dac1f8f8235e5d25b9955d02ff6f29396191d4e665d71122c3722ca83c5/pandas-2.3.3-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:dd7478f1463441ae4ca7308a70e90b33470fa593429f9d4c578dd00d1fa78838", size = 12769373, upload-time = "2025-09-29T23:17:35.846Z" },
+    { url = "https://files.pythonhosted.org/packages/df/91/82cc5169b6b25440a7fc0ef3a694582418d875c8e3ebf796a6d6470aa578/pandas-2.3.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4793891684806ae50d1288c9bae9330293ab4e083ccd1c5e383c34549c6e4250", size = 13200444, upload-time = "2025-09-29T23:17:49.341Z" },
+    { url = "https://files.pythonhosted.org/packages/10/ae/89b3283800ab58f7af2952704078555fa60c807fff764395bb57ea0b0dbd/pandas-2.3.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:28083c648d9a99a5dd035ec125d42439c6c1c525098c58af0fc38dd1a7a1b3d4", size = 13858459, upload-time = "2025-09-29T23:18:03.722Z" },
+    { url = "https://files.pythonhosted.org/packages/85/72/530900610650f54a35a19476eca5104f38555afccda1aa11a92ee14cb21d/pandas-2.3.3-cp310-cp310-win_amd64.whl", hash = "sha256:503cf027cf9940d2ceaa1a93cfb5f8c8c7e6e90720a2850378f0b3f3b1e06826", size = 11346086, upload-time = "2025-09-29T23:18:18.505Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/fa/7ac648108144a095b4fb6aa3de1954689f7af60a14cf25583f4960ecb878/pandas-2.3.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:602b8615ebcc4a0c1751e71840428ddebeb142ec02c786e8ad6b1ce3c8dec523", size = 11578790, upload-time = "2025-09-29T23:18:30.065Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/35/74442388c6cf008882d4d4bdfc4109be87e9b8b7ccd097ad1e7f006e2e95/pandas-2.3.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:8fe25fc7b623b0ef6b5009149627e34d2a4657e880948ec3c840e9402e5c1b45", size = 10833831, upload-time = "2025-09-29T23:38:56.071Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/e4/de154cbfeee13383ad58d23017da99390b91d73f8c11856f2095e813201b/pandas-2.3.3-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b468d3dad6ff947df92dcb32ede5b7bd41a9b3cceef0a30ed925f6d01fb8fa66", size = 12199267, upload-time = "2025-09-29T23:18:41.627Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/c9/63f8d545568d9ab91476b1818b4741f521646cbdd151c6efebf40d6de6f7/pandas-2.3.3-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b98560e98cb334799c0b07ca7967ac361a47326e9b4e5a7dfb5ab2b1c9d35a1b", size = 12789281, upload-time = "2025-09-29T23:18:56.834Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/00/a5ac8c7a0e67fd1a6059e40aa08fa1c52cc00709077d2300e210c3ce0322/pandas-2.3.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:1d37b5848ba49824e5c30bedb9c830ab9b7751fd049bc7914533e01c65f79791", size = 13240453, upload-time = "2025-09-29T23:19:09.247Z" },
+    { url = "https://files.pythonhosted.org/packages/27/4d/5c23a5bc7bd209231618dd9e606ce076272c9bc4f12023a70e03a86b4067/pandas-2.3.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:db4301b2d1f926ae677a751eb2bd0e8c5f5319c9cb3f88b0becbbb0b07b34151", size = 13890361, upload-time = "2025-09-29T23:19:25.342Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/59/712db1d7040520de7a4965df15b774348980e6df45c129b8c64d0dbe74ef/pandas-2.3.3-cp311-cp311-win_amd64.whl", hash = "sha256:f086f6fe114e19d92014a1966f43a3e62285109afe874f067f5abbdcbb10e59c", size = 11348702, upload-time = "2025-09-29T23:19:38.296Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/fb/231d89e8637c808b997d172b18e9d4a4bc7bf31296196c260526055d1ea0/pandas-2.3.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:6d21f6d74eb1725c2efaa71a2bfc661a0689579b58e9c0ca58a739ff0b002b53", size = 11597846, upload-time = "2025-09-29T23:19:48.856Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/bd/bf8064d9cfa214294356c2d6702b716d3cf3bb24be59287a6a21e24cae6b/pandas-2.3.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3fd2f887589c7aa868e02632612ba39acb0b8948faf5cc58f0850e165bd46f35", size = 10729618, upload-time = "2025-09-29T23:39:08.659Z" },
+    { url = "https://files.pythonhosted.org/packages/57/56/cf2dbe1a3f5271370669475ead12ce77c61726ffd19a35546e31aa8edf4e/pandas-2.3.3-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ecaf1e12bdc03c86ad4a7ea848d66c685cb6851d807a26aa245ca3d2017a1908", size = 11737212, upload-time = "2025-09-29T23:19:59.765Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/63/cd7d615331b328e287d8233ba9fdf191a9c2d11b6af0c7a59cfcec23de68/pandas-2.3.3-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b3d11d2fda7eb164ef27ffc14b4fcab16a80e1ce67e9f57e19ec0afaf715ba89", size = 12362693, upload-time = "2025-09-29T23:20:14.098Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/de/8b1895b107277d52f2b42d3a6806e69cfef0d5cf1d0ba343470b9d8e0a04/pandas-2.3.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a68e15f780eddf2b07d242e17a04aa187a7ee12b40b930bfdd78070556550e98", size = 12771002, upload-time = "2025-09-29T23:20:26.76Z" },
+    { url = "https://files.pythonhosted.org/packages/87/21/84072af3187a677c5893b170ba2c8fbe450a6ff911234916da889b698220/pandas-2.3.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:371a4ab48e950033bcf52b6527eccb564f52dc826c02afd9a1bc0ab731bba084", size = 13450971, upload-time = "2025-09-29T23:20:41.344Z" },
+    { url = "https://files.pythonhosted.org/packages/86/41/585a168330ff063014880a80d744219dbf1dd7a1c706e75ab3425a987384/pandas-2.3.3-cp312-cp312-win_amd64.whl", hash = "sha256:a16dcec078a01eeef8ee61bf64074b4e524a2a3f4b3be9326420cabe59c4778b", size = 10992722, upload-time = "2025-09-29T23:20:54.139Z" },
+    { url = "https://files.pythonhosted.org/packages/56/b4/52eeb530a99e2a4c55ffcd352772b599ed4473a0f892d127f4147cf0f88e/pandas-2.3.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:c503ba5216814e295f40711470446bc3fd00f0faea8a086cbc688808e26f92a2", size = 11567720, upload-time = "2025-09-29T23:33:06.209Z" },
+    { url = "https://files.pythonhosted.org/packages/48/4a/2d8b67632a021bced649ba940455ed441ca854e57d6e7658a6024587b083/pandas-2.3.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a637c5cdfa04b6d6e2ecedcb81fc52ffb0fd78ce2ebccc9ea964df9f658de8c8", size = 10810302, upload-time = "2025-09-29T23:33:35.846Z" },
+    { url = "https://files.pythonhosted.org/packages/13/e6/d2465010ee0569a245c975dc6967b801887068bc893e908239b1f4b6c1ac/pandas-2.3.3-cp39-cp39-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:854d00d556406bffe66a4c0802f334c9ad5a96b4f1f868adf036a21b11ef13ff", size = 12154874, upload-time = "2025-09-29T23:33:49.939Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/18/aae8c0aa69a386a3255940e9317f793808ea79d0a525a97a903366bb2569/pandas-2.3.3-cp39-cp39-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:bf1f8a81d04ca90e32a0aceb819d34dbd378a98bf923b6398b9a3ec0bf44de29", size = 12790141, upload-time = "2025-09-29T23:34:05.655Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/26/617f98de789de00c2a444fbe6301bb19e66556ac78cff933d2c98f62f2b4/pandas-2.3.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:23ebd657a4d38268c7dfbdf089fbc31ea709d82e4923c5ffd4fbd5747133ce73", size = 13208697, upload-time = "2025-09-29T23:34:21.835Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/fb/25709afa4552042bd0e15717c75e9b4a2294c3dc4f7e6ea50f03c5136600/pandas-2.3.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:5554c929ccc317d41a5e3d1234f3be588248e61f08a74dd17c9eabb535777dc9", size = 13879233, upload-time = "2025-09-29T23:34:35.079Z" },
+    { url = "https://files.pythonhosted.org/packages/98/af/7be05277859a7bc399da8ba68b88c96b27b48740b6cf49688899c6eb4176/pandas-2.3.3-cp39-cp39-win_amd64.whl", hash = "sha256:d3e28b3e83862ccf4d85ff19cf8c20b2ae7e503881711ff2d534dc8f761131aa", size = 11359119, upload-time = "2025-09-29T23:34:46.339Z" },
+]
+
+[[package]]
+name = "partial-json-parser"
+version = "0.2.1.1.post6"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/86/13/459e86c9c67a006651803a3df3d0b08f7708bc5483fdc482582d75562949/partial_json_parser-0.2.1.1.post6.tar.gz", hash = "sha256:43896b68929678224cbbe4884a6a5fe9251ded4b30b8b7d7eb569e5feea93afc", size = 10299, upload-time = "2025-06-23T17:51:45.372Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/cb/40/1f922794af3dc7503f19319a8804b398a161a2cd54183cff8b12225b8d85/partial_json_parser-0.2.1.1.post6-py3-none-any.whl", hash = "sha256:abc332f09b13ef5233384dbfe7128a0e9ea3fa4b8f8be9b37ac1b433c810e99e", size = 10876, upload-time = "2025-06-23T17:51:44.332Z" },
 ]
 
 [[package]]
@@ -3055,119 +3497,126 @@ wheels = [
 ]
 
 [[package]]
-name = "prompt-toolkit"
-version = "3.0.52"
+name = "prometheus-client"
+version = "0.23.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/23/53/3edb5d68ecf6b38fcbcc1ad28391117d2a322d9a1a3eff04bfdb184d8c3b/prometheus_client-0.23.1.tar.gz", hash = "sha256:6ae8f9081eaaaf153a2e959d2e6c4f4fb57b12ef76c8c7980202f1e57b48b2ce", size = 80481, upload-time = "2025-09-18T20:47:25.043Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b8/db/14bafcb4af2139e046d03fd00dea7873e48eafe18b7d2797e73d6681f210/prometheus_client-0.23.1-py3-none-any.whl", hash = "sha256:dd1913e6e76b59cfe44e7a4b83e01afc9873c1bdfd2ed8739f1e76aeca115f99", size = 61145, upload-time = "2025-09-18T20:47:23.875Z" },
+]
+
+[[package]]
+name = "prometheus-fastapi-instrumentator"
+version = "7.1.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "wcwidth" },
+    { name = "prometheus-client" },
+    { name = "starlette" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a1/96/06e01a7b38dce6fe1db213e061a4602dd6032a8a97ef6c1a862537732421/prompt_toolkit-3.0.52.tar.gz", hash = "sha256:28cde192929c8e7321de85de1ddbe736f1375148b02f2e17edd840042b1be855", size = 434198, upload-time = "2025-08-27T15:24:02.057Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/69/6d/24d53033cf93826aa7857699a4450c1c67e5b9c710e925b1ed2b320c04df/prometheus_fastapi_instrumentator-7.1.0.tar.gz", hash = "sha256:be7cd61eeea4e5912aeccb4261c6631b3f227d8924542d79eaf5af3f439cbe5e", size = 20220, upload-time = "2025-03-19T19:35:05.351Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/84/03/0d3ce49e2505ae70cf43bc5bb3033955d2fc9f932163e84dc0779cc47f48/prompt_toolkit-3.0.52-py3-none-any.whl", hash = "sha256:9aac639a3bbd33284347de5ad8d68ecc044b91a762dc39b7c21095fcd6a19955", size = 391431, upload-time = "2025-08-27T15:23:59.498Z" },
+    { url = "https://files.pythonhosted.org/packages/27/72/0824c18f3bc75810f55dacc2dd933f6ec829771180245ae3cc976195dec0/prometheus_fastapi_instrumentator-7.1.0-py3-none-any.whl", hash = "sha256:978130f3c0bb7b8ebcc90d35516a6fe13e02d2eb358c8f83887cdef7020c31e9", size = 19296, upload-time = "2025-03-19T19:35:04.323Z" },
 ]
 
 [[package]]
 name = "propcache"
-version = "0.3.2"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/a6/16/43264e4a779dd8588c21a70f0709665ee8f611211bdd2c87d952cfa7c776/propcache-0.3.2.tar.gz", hash = "sha256:20d7d62e4e7ef05f221e0db2856b979540686342e7dd9973b815599c7057e168", size = 44139, upload-time = "2025-06-09T22:56:06.081Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/ab/14/510deed325e262afeb8b360043c5d7c960da7d3ecd6d6f9496c9c56dc7f4/propcache-0.3.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:22d9962a358aedbb7a2e36187ff273adeaab9743373a272976d2e348d08c7770", size = 73178, upload-time = "2025-06-09T22:53:40.126Z" },
-    { url = "https://files.pythonhosted.org/packages/cd/4e/ad52a7925ff01c1325653a730c7ec3175a23f948f08626a534133427dcff/propcache-0.3.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0d0fda578d1dc3f77b6b5a5dce3b9ad69a8250a891760a548df850a5e8da87f3", size = 43133, upload-time = "2025-06-09T22:53:41.965Z" },
-    { url = "https://files.pythonhosted.org/packages/63/7c/e9399ba5da7780871db4eac178e9c2e204c23dd3e7d32df202092a1ed400/propcache-0.3.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:3def3da3ac3ce41562d85db655d18ebac740cb3fa4367f11a52b3da9d03a5cc3", size = 43039, upload-time = "2025-06-09T22:53:43.268Z" },
-    { url = "https://files.pythonhosted.org/packages/22/e1/58da211eb8fdc6fc854002387d38f415a6ca5f5c67c1315b204a5d3e9d7a/propcache-0.3.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9bec58347a5a6cebf239daba9bda37dffec5b8d2ce004d9fe4edef3d2815137e", size = 201903, upload-time = "2025-06-09T22:53:44.872Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/0a/550ea0f52aac455cb90111c8bab995208443e46d925e51e2f6ebdf869525/propcache-0.3.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:55ffda449a507e9fbd4aca1a7d9aa6753b07d6166140e5a18d2ac9bc49eac220", size = 213362, upload-time = "2025-06-09T22:53:46.707Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/af/9893b7d878deda9bb69fcf54600b247fba7317761b7db11fede6e0f28bd0/propcache-0.3.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:64a67fb39229a8a8491dd42f864e5e263155e729c2e7ff723d6e25f596b1e8cb", size = 210525, upload-time = "2025-06-09T22:53:48.547Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/bb/38fd08b278ca85cde36d848091ad2b45954bc5f15cce494bb300b9285831/propcache-0.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9da1cf97b92b51253d5b68cf5a2b9e0dafca095e36b7f2da335e27dc6172a614", size = 198283, upload-time = "2025-06-09T22:53:50.067Z" },
-    { url = "https://files.pythonhosted.org/packages/78/8c/9fe55bd01d362bafb413dfe508c48753111a1e269737fa143ba85693592c/propcache-0.3.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5f559e127134b07425134b4065be45b166183fdcb433cb6c24c8e4149056ad50", size = 191872, upload-time = "2025-06-09T22:53:51.438Z" },
-    { url = "https://files.pythonhosted.org/packages/54/14/4701c33852937a22584e08abb531d654c8bcf7948a8f87ad0a4822394147/propcache-0.3.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:aff2e4e06435d61f11a428360a932138d0ec288b0a31dd9bd78d200bd4a2b339", size = 199452, upload-time = "2025-06-09T22:53:53.229Z" },
-    { url = "https://files.pythonhosted.org/packages/16/44/447f2253d859602095356007657ee535e0093215ea0b3d1d6a41d16e5201/propcache-0.3.2-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:4927842833830942a5d0a56e6f4839bc484785b8e1ce8d287359794818633ba0", size = 191567, upload-time = "2025-06-09T22:53:54.541Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/b3/e4756258749bb2d3b46defcff606a2f47410bab82be5824a67e84015b267/propcache-0.3.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:6107ddd08b02654a30fb8ad7a132021759d750a82578b94cd55ee2772b6ebea2", size = 193015, upload-time = "2025-06-09T22:53:56.44Z" },
-    { url = "https://files.pythonhosted.org/packages/1e/df/e6d3c7574233164b6330b9fd697beeac402afd367280e6dc377bb99b43d9/propcache-0.3.2-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:70bd8b9cd6b519e12859c99f3fc9a93f375ebd22a50296c3a295028bea73b9e7", size = 204660, upload-time = "2025-06-09T22:53:57.839Z" },
-    { url = "https://files.pythonhosted.org/packages/b2/53/e4d31dd5170b4a0e2e6b730f2385a96410633b4833dc25fe5dffd1f73294/propcache-0.3.2-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:2183111651d710d3097338dd1893fcf09c9f54e27ff1a8795495a16a469cc90b", size = 206105, upload-time = "2025-06-09T22:53:59.638Z" },
-    { url = "https://files.pythonhosted.org/packages/7f/fe/74d54cf9fbe2a20ff786e5f7afcfde446588f0cf15fb2daacfbc267b866c/propcache-0.3.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:fb075ad271405dcad8e2a7ffc9a750a3bf70e533bd86e89f0603e607b93aa64c", size = 196980, upload-time = "2025-06-09T22:54:01.071Z" },
-    { url = "https://files.pythonhosted.org/packages/22/ec/c469c9d59dada8a7679625e0440b544fe72e99311a4679c279562051f6fc/propcache-0.3.2-cp310-cp310-win32.whl", hash = "sha256:404d70768080d3d3bdb41d0771037da19d8340d50b08e104ca0e7f9ce55fce70", size = 37679, upload-time = "2025-06-09T22:54:03.003Z" },
-    { url = "https://files.pythonhosted.org/packages/38/35/07a471371ac89d418f8d0b699c75ea6dca2041fbda360823de21f6a9ce0a/propcache-0.3.2-cp310-cp310-win_amd64.whl", hash = "sha256:7435d766f978b4ede777002e6b3b6641dd229cd1da8d3d3106a45770365f9ad9", size = 41459, upload-time = "2025-06-09T22:54:04.134Z" },
-    { url = "https://files.pythonhosted.org/packages/80/8d/e8b436717ab9c2cfc23b116d2c297305aa4cd8339172a456d61ebf5669b8/propcache-0.3.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:0b8d2f607bd8f80ddc04088bc2a037fdd17884a6fcadc47a96e334d72f3717be", size = 74207, upload-time = "2025-06-09T22:54:05.399Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/29/1e34000e9766d112171764b9fa3226fa0153ab565d0c242c70e9945318a7/propcache-0.3.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:06766d8f34733416e2e34f46fea488ad5d60726bb9481d3cddf89a6fa2d9603f", size = 43648, upload-time = "2025-06-09T22:54:08.023Z" },
-    { url = "https://files.pythonhosted.org/packages/46/92/1ad5af0df781e76988897da39b5f086c2bf0f028b7f9bd1f409bb05b6874/propcache-0.3.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a2dc1f4a1df4fecf4e6f68013575ff4af84ef6f478fe5344317a65d38a8e6dc9", size = 43496, upload-time = "2025-06-09T22:54:09.228Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/ce/e96392460f9fb68461fabab3e095cb00c8ddf901205be4eae5ce246e5b7e/propcache-0.3.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:be29c4f4810c5789cf10ddf6af80b041c724e629fa51e308a7a0fb19ed1ef7bf", size = 217288, upload-time = "2025-06-09T22:54:10.466Z" },
-    { url = "https://files.pythonhosted.org/packages/c5/2a/866726ea345299f7ceefc861a5e782b045545ae6940851930a6adaf1fca6/propcache-0.3.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:59d61f6970ecbd8ff2e9360304d5c8876a6abd4530cb752c06586849ac8a9dc9", size = 227456, upload-time = "2025-06-09T22:54:11.828Z" },
-    { url = "https://files.pythonhosted.org/packages/de/03/07d992ccb6d930398689187e1b3c718339a1c06b8b145a8d9650e4726166/propcache-0.3.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:62180e0b8dbb6b004baec00a7983e4cc52f5ada9cd11f48c3528d8cfa7b96a66", size = 225429, upload-time = "2025-06-09T22:54:13.823Z" },
-    { url = "https://files.pythonhosted.org/packages/5d/e6/116ba39448753b1330f48ab8ba927dcd6cf0baea8a0ccbc512dfb49ba670/propcache-0.3.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c144ca294a204c470f18cf4c9d78887810d04a3e2fbb30eea903575a779159df", size = 213472, upload-time = "2025-06-09T22:54:15.232Z" },
-    { url = "https://files.pythonhosted.org/packages/a6/85/f01f5d97e54e428885a5497ccf7f54404cbb4f906688a1690cd51bf597dc/propcache-0.3.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c5c2a784234c28854878d68978265617aa6dc0780e53d44b4d67f3651a17a9a2", size = 204480, upload-time = "2025-06-09T22:54:17.104Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/79/7bf5ab9033b8b8194cc3f7cf1aaa0e9c3256320726f64a3e1f113a812dce/propcache-0.3.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:5745bc7acdafa978ca1642891b82c19238eadc78ba2aaa293c6863b304e552d7", size = 214530, upload-time = "2025-06-09T22:54:18.512Z" },
-    { url = "https://files.pythonhosted.org/packages/31/0b/bd3e0c00509b609317df4a18e6b05a450ef2d9a963e1d8bc9c9415d86f30/propcache-0.3.2-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:c0075bf773d66fa8c9d41f66cc132ecc75e5bb9dd7cce3cfd14adc5ca184cb95", size = 205230, upload-time = "2025-06-09T22:54:19.947Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/23/fae0ff9b54b0de4e819bbe559508da132d5683c32d84d0dc2ccce3563ed4/propcache-0.3.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:5f57aa0847730daceff0497f417c9de353c575d8da3579162cc74ac294c5369e", size = 206754, upload-time = "2025-06-09T22:54:21.716Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/7f/ad6a3c22630aaa5f618b4dc3c3598974a72abb4c18e45a50b3cdd091eb2f/propcache-0.3.2-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:eef914c014bf72d18efb55619447e0aecd5fb7c2e3fa7441e2e5d6099bddff7e", size = 218430, upload-time = "2025-06-09T22:54:23.17Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/2c/ba4f1c0e8a4b4c75910742f0d333759d441f65a1c7f34683b4a74c0ee015/propcache-0.3.2-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:2a4092e8549031e82facf3decdbc0883755d5bbcc62d3aea9d9e185549936dcf", size = 223884, upload-time = "2025-06-09T22:54:25.539Z" },
-    { url = "https://files.pythonhosted.org/packages/88/e4/ebe30fc399e98572019eee82ad0caf512401661985cbd3da5e3140ffa1b0/propcache-0.3.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:85871b050f174bc0bfb437efbdb68aaf860611953ed12418e4361bc9c392749e", size = 211480, upload-time = "2025-06-09T22:54:26.892Z" },
-    { url = "https://files.pythonhosted.org/packages/96/0a/7d5260b914e01d1d0906f7f38af101f8d8ed0dc47426219eeaf05e8ea7c2/propcache-0.3.2-cp311-cp311-win32.whl", hash = "sha256:36c8d9b673ec57900c3554264e630d45980fd302458e4ac801802a7fd2ef7897", size = 37757, upload-time = "2025-06-09T22:54:28.241Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/2d/89fe4489a884bc0da0c3278c552bd4ffe06a1ace559db5ef02ef24ab446b/propcache-0.3.2-cp311-cp311-win_amd64.whl", hash = "sha256:e53af8cb6a781b02d2ea079b5b853ba9430fcbe18a8e3ce647d5982a3ff69f39", size = 41500, upload-time = "2025-06-09T22:54:29.4Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/42/9ca01b0a6f48e81615dca4765a8f1dd2c057e0540f6116a27dc5ee01dfb6/propcache-0.3.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:8de106b6c84506b31c27168582cd3cb3000a6412c16df14a8628e5871ff83c10", size = 73674, upload-time = "2025-06-09T22:54:30.551Z" },
-    { url = "https://files.pythonhosted.org/packages/af/6e/21293133beb550f9c901bbece755d582bfaf2176bee4774000bd4dd41884/propcache-0.3.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:28710b0d3975117239c76600ea351934ac7b5ff56e60953474342608dbbb6154", size = 43570, upload-time = "2025-06-09T22:54:32.296Z" },
-    { url = "https://files.pythonhosted.org/packages/0c/c8/0393a0a3a2b8760eb3bde3c147f62b20044f0ddac81e9d6ed7318ec0d852/propcache-0.3.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce26862344bdf836650ed2487c3d724b00fbfec4233a1013f597b78c1cb73615", size = 43094, upload-time = "2025-06-09T22:54:33.929Z" },
-    { url = "https://files.pythonhosted.org/packages/37/2c/489afe311a690399d04a3e03b069225670c1d489eb7b044a566511c1c498/propcache-0.3.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bca54bd347a253af2cf4544bbec232ab982f4868de0dd684246b67a51bc6b1db", size = 226958, upload-time = "2025-06-09T22:54:35.186Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/ca/63b520d2f3d418c968bf596839ae26cf7f87bead026b6192d4da6a08c467/propcache-0.3.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:55780d5e9a2ddc59711d727226bb1ba83a22dd32f64ee15594b9392b1f544eb1", size = 234894, upload-time = "2025-06-09T22:54:36.708Z" },
-    { url = "https://files.pythonhosted.org/packages/11/60/1d0ed6fff455a028d678df30cc28dcee7af77fa2b0e6962ce1df95c9a2a9/propcache-0.3.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:035e631be25d6975ed87ab23153db6a73426a48db688070d925aa27e996fe93c", size = 233672, upload-time = "2025-06-09T22:54:38.062Z" },
-    { url = "https://files.pythonhosted.org/packages/37/7c/54fd5301ef38505ab235d98827207176a5c9b2aa61939b10a460ca53e123/propcache-0.3.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ee6f22b6eaa39297c751d0e80c0d3a454f112f5c6481214fcf4c092074cecd67", size = 224395, upload-time = "2025-06-09T22:54:39.634Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/1a/89a40e0846f5de05fdc6779883bf46ba980e6df4d2ff8fb02643de126592/propcache-0.3.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:7ca3aee1aa955438c4dba34fc20a9f390e4c79967257d830f137bd5a8a32ed3b", size = 212510, upload-time = "2025-06-09T22:54:41.565Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/33/ca98368586c9566a6b8d5ef66e30484f8da84c0aac3f2d9aec6d31a11bd5/propcache-0.3.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:7a4f30862869fa2b68380d677cc1c5fcf1e0f2b9ea0cf665812895c75d0ca3b8", size = 222949, upload-time = "2025-06-09T22:54:43.038Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/11/ace870d0aafe443b33b2f0b7efdb872b7c3abd505bfb4890716ad7865e9d/propcache-0.3.2-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:b77ec3c257d7816d9f3700013639db7491a434644c906a2578a11daf13176251", size = 217258, upload-time = "2025-06-09T22:54:44.376Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/d2/86fd6f7adffcfc74b42c10a6b7db721d1d9ca1055c45d39a1a8f2a740a21/propcache-0.3.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:cab90ac9d3f14b2d5050928483d3d3b8fb6b4018893fc75710e6aa361ecb2474", size = 213036, upload-time = "2025-06-09T22:54:46.243Z" },
-    { url = "https://files.pythonhosted.org/packages/07/94/2d7d1e328f45ff34a0a284cf5a2847013701e24c2a53117e7c280a4316b3/propcache-0.3.2-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:0b504d29f3c47cf6b9e936c1852246c83d450e8e063d50562115a6be6d3a2535", size = 227684, upload-time = "2025-06-09T22:54:47.63Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/05/37ae63a0087677e90b1d14710e532ff104d44bc1efa3b3970fff99b891dc/propcache-0.3.2-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:ce2ac2675a6aa41ddb2a0c9cbff53780a617ac3d43e620f8fd77ba1c84dcfc06", size = 234562, upload-time = "2025-06-09T22:54:48.982Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/7c/3f539fcae630408d0bd8bf3208b9a647ccad10976eda62402a80adf8fc34/propcache-0.3.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:62b4239611205294cc433845b914131b2a1f03500ff3c1ed093ed216b82621e1", size = 222142, upload-time = "2025-06-09T22:54:50.424Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/d2/34b9eac8c35f79f8a962546b3e97e9d4b990c420ee66ac8255d5d9611648/propcache-0.3.2-cp312-cp312-win32.whl", hash = "sha256:df4a81b9b53449ebc90cc4deefb052c1dd934ba85012aa912c7ea7b7e38b60c1", size = 37711, upload-time = "2025-06-09T22:54:52.072Z" },
-    { url = "https://files.pythonhosted.org/packages/19/61/d582be5d226cf79071681d1b46b848d6cb03d7b70af7063e33a2787eaa03/propcache-0.3.2-cp312-cp312-win_amd64.whl", hash = "sha256:7046e79b989d7fe457bb755844019e10f693752d169076138abf17f31380800c", size = 41479, upload-time = "2025-06-09T22:54:53.234Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/39/8ea9bcfaaff16fd0b0fc901ee522e24c9ec44b4ca0229cfffb8066a06959/propcache-0.3.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:a7fad897f14d92086d6b03fdd2eb844777b0c4d7ec5e3bac0fbae2ab0602bbe5", size = 74678, upload-time = "2025-06-09T22:55:41.227Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/85/cab84c86966e1d354cf90cdc4ba52f32f99a5bca92a1529d666d957d7686/propcache-0.3.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1f43837d4ca000243fd7fd6301947d7cb93360d03cd08369969450cc6b2ce3b4", size = 43829, upload-time = "2025-06-09T22:55:42.417Z" },
-    { url = "https://files.pythonhosted.org/packages/23/f7/9cb719749152d8b26d63801b3220ce2d3931312b2744d2b3a088b0ee9947/propcache-0.3.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:261df2e9474a5949c46e962065d88eb9b96ce0f2bd30e9d3136bcde84befd8f2", size = 43729, upload-time = "2025-06-09T22:55:43.651Z" },
-    { url = "https://files.pythonhosted.org/packages/a2/a2/0b2b5a210ff311260002a315f6f9531b65a36064dfb804655432b2f7d3e3/propcache-0.3.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e514326b79e51f0a177daab1052bc164d9d9e54133797a3a58d24c9c87a3fe6d", size = 204483, upload-time = "2025-06-09T22:55:45.327Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/e0/7aff5de0c535f783b0c8be5bdb750c305c1961d69fbb136939926e155d98/propcache-0.3.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d4a996adb6904f85894570301939afeee65f072b4fd265ed7e569e8d9058e4ec", size = 217425, upload-time = "2025-06-09T22:55:46.729Z" },
-    { url = "https://files.pythonhosted.org/packages/92/1d/65fa889eb3b2a7d6e4ed3c2b568a9cb8817547a1450b572de7bf24872800/propcache-0.3.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:76cace5d6b2a54e55b137669b30f31aa15977eeed390c7cbfb1dafa8dfe9a701", size = 214723, upload-time = "2025-06-09T22:55:48.342Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/e2/eecf6989870988dfd731de408a6fa366e853d361a06c2133b5878ce821ad/propcache-0.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:31248e44b81d59d6addbb182c4720f90b44e1efdc19f58112a3c3a1615fb47ef", size = 200166, upload-time = "2025-06-09T22:55:49.775Z" },
-    { url = "https://files.pythonhosted.org/packages/12/06/c32be4950967f18f77489268488c7cdc78cbfc65a8ba8101b15e526b83dc/propcache-0.3.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:abb7fa19dbf88d3857363e0493b999b8011eea856b846305d8c0512dfdf8fbb1", size = 194004, upload-time = "2025-06-09T22:55:51.335Z" },
-    { url = "https://files.pythonhosted.org/packages/46/6c/17b521a6b3b7cbe277a4064ff0aa9129dd8c89f425a5a9b6b4dd51cc3ff4/propcache-0.3.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:d81ac3ae39d38588ad0549e321e6f773a4e7cc68e7751524a22885d5bbadf886", size = 203075, upload-time = "2025-06-09T22:55:52.681Z" },
-    { url = "https://files.pythonhosted.org/packages/62/cb/3bdba2b736b3e45bc0e40f4370f745b3e711d439ffbffe3ae416393eece9/propcache-0.3.2-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:cc2782eb0f7a16462285b6f8394bbbd0e1ee5f928034e941ffc444012224171b", size = 195407, upload-time = "2025-06-09T22:55:54.048Z" },
-    { url = "https://files.pythonhosted.org/packages/29/bd/760c5c6a60a4a2c55a421bc34a25ba3919d49dee411ddb9d1493bb51d46e/propcache-0.3.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:db429c19a6c7e8a1c320e6a13c99799450f411b02251fb1b75e6217cf4a14fcb", size = 196045, upload-time = "2025-06-09T22:55:55.485Z" },
-    { url = "https://files.pythonhosted.org/packages/76/58/ced2757a46f55b8c84358d6ab8de4faf57cba831c51e823654da7144b13a/propcache-0.3.2-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:21d8759141a9e00a681d35a1f160892a36fb6caa715ba0b832f7747da48fb6ea", size = 208432, upload-time = "2025-06-09T22:55:56.884Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/ec/d98ea8d5a4d8fe0e372033f5254eddf3254344c0c5dc6c49ab84349e4733/propcache-0.3.2-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:2ca6d378f09adb13837614ad2754fa8afaee330254f404299611bce41a8438cb", size = 210100, upload-time = "2025-06-09T22:55:58.498Z" },
-    { url = "https://files.pythonhosted.org/packages/56/84/b6d8a7ecf3f62d7dd09d9d10bbf89fad6837970ef868b35b5ffa0d24d9de/propcache-0.3.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:34a624af06c048946709f4278b4176470073deda88d91342665d95f7c6270fbe", size = 200712, upload-time = "2025-06-09T22:55:59.906Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/32/889f4903ddfe4a9dc61da71ee58b763758cf2d608fe1decede06e6467f8d/propcache-0.3.2-cp39-cp39-win32.whl", hash = "sha256:4ba3fef1c30f306b1c274ce0b8baaa2c3cdd91f645c48f06394068f37d3837a1", size = 38187, upload-time = "2025-06-09T22:56:01.212Z" },
-    { url = "https://files.pythonhosted.org/packages/67/74/d666795fb9ba1dc139d30de64f3b6fd1ff9c9d3d96ccfdb992cd715ce5d2/propcache-0.3.2-cp39-cp39-win_amd64.whl", hash = "sha256:7a2368eed65fc69a7a7a40b27f22e85e7627b74216f0846b04ba5c116e191ec9", size = 42025, upload-time = "2025-06-09T22:56:02.875Z" },
-    { url = "https://files.pythonhosted.org/packages/cc/35/cc0aaecf278bb4575b8555f2b137de5ab821595ddae9da9d3cd1da4072c7/propcache-0.3.2-py3-none-any.whl", hash = "sha256:98f1ec44fb675f5052cccc8e609c46ed23a35a1cfd18545ad4e29002d858a43f", size = 12663, upload-time = "2025-06-09T22:56:04.484Z" },
+version = "0.4.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ea/c8/d70cd26d845c6d85479d8f5a11a0fd7151e9bc4794cc5e6eb5a790f12df8/propcache-0.4.0.tar.gz", hash = "sha256:c1ad731253eb738f9cadd9fa1844e019576c70bca6a534252e97cf33a57da529", size = 45187, upload-time = "2025-10-04T21:57:39.546Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d1/7b/4bd85fea3dc58b6f246abf0e6c9e44adca26f6817e6c136780315d723b82/propcache-0.4.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:779aaae64089e2f4992e993faea801925395d26bb5de4a47df7ef7f942c14f80", size = 79437, upload-time = "2025-10-04T21:54:49.766Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/91/379ecc1ab37fe33648c7cb2d2252f58969adac1edcd6ec74682d7fb2d920/propcache-0.4.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:566552ed9b003030745e5bc7b402b83cf3cecae1bade95262d78543741786db5", size = 45369, upload-time = "2025-10-04T21:54:51.688Z" },
+    { url = "https://files.pythonhosted.org/packages/de/8e/2e002e59e359bbc6ababbb7da168226f93e0533429ea1e93989a7eedcb2a/propcache-0.4.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:944de70384c62d16d4a00c686b422aa75efbc67c4addaebefbb56475d1c16034", size = 47191, upload-time = "2025-10-04T21:54:52.915Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/9a/f56eef9932dc3cbc63df4716f09fbaefec7a475608b643842784a01351b6/propcache-0.4.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e878553543ece1f8006d0ba4d096b40290580db173bfb18e16158045b9371335", size = 201000, upload-time = "2025-10-04T21:54:54.556Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/84/e7ad1e09c13f0574dbad261441f6a7f1fb8cc1e2fcb23ec4d4b3e4c7dc67/propcache-0.4.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8659f995b19185179474b18de8755689e1f71e1334d05c14e1895caa4e409cf7", size = 209175, upload-time = "2025-10-04T21:54:55.996Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/74/0b785ac0fbb44a5a7c267efc409b7a62d7a03b17c6442ecb52fd29152314/propcache-0.4.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:7aa8cc5c94e682dce91cb4d12d7b81c01641f4ef5b3b3dc53325d43f0e3b9f2e", size = 214874, upload-time = "2025-10-04T21:54:57.831Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/fd/e8d795def2b1d8dc1dc4731d36da1f6111d7c73212909e79462172d0434c/propcache-0.4.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:da584d917a1a17f690fc726617fd2c3f3006ea959dae5bb07a5630f7b16f9f5f", size = 196686, upload-time = "2025-10-04T21:54:59.218Z" },
+    { url = "https://files.pythonhosted.org/packages/79/c2/dc992c712c3a1bfaa11d13ff177dbdf9b8b272e7bd443601e37f35728338/propcache-0.4.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:892a072e5b19c3f324a4f8543c9f7e8fc2b0aa08579e46f69bdf0cfc1b440454", size = 192000, upload-time = "2025-10-04T21:55:00.645Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/de/bb108dbdfae594148b033ff283d9fa6e4b0906a99f2c03b98b526883149d/propcache-0.4.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:c20d796210720455086ef3f85adc413d1e41d374742f9b439354f122bbc3b528", size = 190310, upload-time = "2025-10-04T21:55:02.107Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/7b/1bdb5d44ba4c87d270bcf11354950f8f7fbc9ace1fbe7745e683fcb57b5a/propcache-0.4.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:df7107a91126a495880576610ae989f19106e1900dd5218d08498391fa43b31d", size = 199646, upload-time = "2025-10-04T21:55:03.55Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/04/44beda877f779f49f5b8c0ff4817a62b5f90a2dfac1ec5311df15a9dfceb/propcache-0.4.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:0b04ac2120c161416c866d0b6a4259e47e92231ff166b518cc0efb95777367c3", size = 200507, upload-time = "2025-10-04T21:55:04.914Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/62/a13ad0a63e06f3695fcaeaeeeb62e2cc685181a1248b23a2bc877c8b7111/propcache-0.4.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:1e7fa29c71ffa8d6a37324258737d09475f84715a6e8c350f67f0bc8e5e44993", size = 192787, upload-time = "2025-10-04T21:55:06.385Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/07/386246b3b4a6b11208bcbf57580210fb8c923ab26759389fe594e5615cd7/propcache-0.4.0-cp310-cp310-win32.whl", hash = "sha256:01c0ebc172ca28e9d62876832befbf7f36080eee6ed9c9e00243de2a8089ad57", size = 38004, upload-time = "2025-10-04T21:55:07.692Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/f2/e1fcb9694f590bc443ae5044f982546bb01cbaa3cdf05286e9473a9874bf/propcache-0.4.0-cp310-cp310-win_amd64.whl", hash = "sha256:84f847e64f4d1a232e50460eebc1196642ee9b4c983612f41cd2d44fd2fe7c71", size = 41516, upload-time = "2025-10-04T21:55:08.854Z" },
+    { url = "https://files.pythonhosted.org/packages/15/f4/d211744d41d72fbb89d3ee53963c1dc26892c49f53ae3c49fbc15cfb2548/propcache-0.4.0-cp310-cp310-win_arm64.whl", hash = "sha256:2166466a666a5bebc332cd209cad77d996fad925ca7e8a2a6310ba9e851ae641", size = 38122, upload-time = "2025-10-04T21:55:10.044Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/c4/72b8d41bdbae8aea9c25b869d7cdc3ab5f281f979d8aea30f4646ad12743/propcache-0.4.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6a6a36b94c09711d6397d79006ca47901539fbc602c853d794c39abd6a326549", size = 80035, upload-time = "2025-10-04T21:55:11.266Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/f8/f87115733e221408a363f3a9753419cf2d4be7a8a7ec9dc0788325cd23f1/propcache-0.4.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:da47070e1340a1639aca6b1c18fe1f1f3d8d64d3a1f9ddc67b94475f44cd40f3", size = 45622, upload-time = "2025-10-04T21:55:12.41Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/cc/391f883248faa2efdf6886bdb12ac8edf20eac0863770d8d925450d8cc76/propcache-0.4.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:de536cf796abc5b58d11c0ad56580215d231d9554ea4bb6b8b1b3bed80aa3234", size = 47517, upload-time = "2025-10-04T21:55:13.819Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/d2/5593b59999f42d1044c5ab5f238be1f9d537ab91b0c910727986d520a6e9/propcache-0.4.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f5c82af8e329c3cdc3e717dd3c7b2ff1a218b6de611f6ce76ee34967570a9de9", size = 214540, upload-time = "2025-10-04T21:55:15.206Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/5d/028cdc0eaa1a66ee2ec339a08b5e6ec15e7e71dac86103bebe53ba10dc0f/propcache-0.4.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:abe04e7aa5ab2e4056fcf3255ebee2071e4a427681f76d4729519e292c46ecc1", size = 221603, upload-time = "2025-10-04T21:55:16.704Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/f8/e30aee5f59ea21647faef9c82bd67fa510295c34908a7a38571def555881/propcache-0.4.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:075ca32384294434344760fdcb95f7833e1d7cf7c4e55f0e726358140179da35", size = 227749, upload-time = "2025-10-04T21:55:18.082Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/85/0757dfc73931bea63b18d26b2c5e7bf13113ca60fe0e5f19905f104bcf6a/propcache-0.4.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:626ec13592928b677f48ff5861040b604b635e93d8e2162fb638397ea83d07e8", size = 209792, upload-time = "2025-10-04T21:55:19.475Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/45/35a6a6241f46948c0ac2418d5bf50cfbcd9735739f42028a1c11e9066a72/propcache-0.4.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:02e071548b6a376e173b0102c3f55dc16e7d055b5307d487e844c320e38cacf2", size = 207979, upload-time = "2025-10-04T21:55:21.164Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/d1/5930396e75c9ed477958eac1496e6fb08794d823e9b14a459f1c0e20f338/propcache-0.4.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:2af6de831a26f42a3f94592964becd8d7f238551786d7525807f02e53defbd13", size = 201923, upload-time = "2025-10-04T21:55:22.5Z" },
+    { url = "https://files.pythonhosted.org/packages/98/72/675455f22bcefeda16907461f9a9a4a93709ff2095e8cf799bdb6c78e030/propcache-0.4.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:bd6c6dba1a3b8949e08c4280071c86e38cb602f02e0ed6659234108c7a7cd710", size = 212117, upload-time = "2025-10-04T21:55:23.858Z" },
+    { url = "https://files.pythonhosted.org/packages/13/27/c533302ff80a49a848c3dbd01bb18f87b06826602b3b37043ff00d6b5005/propcache-0.4.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:783e91595cf9b66c2deda17f2e8748ae8591aa9f7c65dcab038872bfe83c5bb1", size = 216594, upload-time = "2025-10-04T21:55:25.169Z" },
+    { url = "https://files.pythonhosted.org/packages/63/91/8250fbb601fd16c427e5f469132f27e175c6692dbfa784ef1266dc652e55/propcache-0.4.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:c3f4b125285d354a627eb37f3ea7c13b8842c7c0d47783581d0df0e272dbf5f0", size = 204863, upload-time = "2025-10-04T21:55:26.511Z" },
+    { url = "https://files.pythonhosted.org/packages/34/c4/fd945a9a25845aafb6094b9fa6a88286e4e1c55686e60172c60fe669e0d1/propcache-0.4.0-cp311-cp311-win32.whl", hash = "sha256:71c45f02ffbb8a21040ae816ceff7f6cd749ffac29fc0f9daa42dc1a9652d577", size = 37948, upload-time = "2025-10-04T21:55:27.719Z" },
+    { url = "https://files.pythonhosted.org/packages/42/02/f30e7304661ffe8d51ff4050e06765ac2df6d95cf23c999dfe5a0cd0eb4c/propcache-0.4.0-cp311-cp311-win_amd64.whl", hash = "sha256:7d51f70f77950f8efafed4383865d3533eeee52d8a0dd1c35b65f24de41de4e0", size = 41511, upload-time = "2025-10-04T21:55:29.15Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/f2/edd329d86085438a1ba32cf4cf45fc982d18343bed1f16b218b516c3340d/propcache-0.4.0-cp311-cp311-win_arm64.whl", hash = "sha256:858eaabd2191dd0da5272993ad08a748b5d3ae1aefabea8aee619b45c2af4a64", size = 37957, upload-time = "2025-10-04T21:55:30.31Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/cf/3f88344261d69f8021256f20e82e820c5df3aba96e5ba9b5fdd3685d3a9f/propcache-0.4.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:381c84a445efb8c9168f1393a5a7c566de22edc42bfe207a142fff919b37f5d9", size = 79846, upload-time = "2025-10-04T21:55:31.447Z" },
+    { url = "https://files.pythonhosted.org/packages/be/fa/0286fc92764eead9dcfee639b67828daa32e61dd0f1618831547141eb28b/propcache-0.4.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:5a531d29d7b873b12730972237c48b1a4e5980b98cf21b3f09fa4710abd3a8c3", size = 45850, upload-time = "2025-10-04T21:55:32.637Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/83/57840656f972f8a67992eee40781e4066657776dcb889f49df0e8eecb112/propcache-0.4.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:cd6e22255ed73efeaaeb1765505a66a48a9ec9ebc919fce5ad490fe5e33b1555", size = 47171, upload-time = "2025-10-04T21:55:33.819Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/8e/e0a0bd376c3440476b924eca517589ee535bb4520420d178268bf88558ba/propcache-0.4.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d9a8d277dc218ddf04ec243a53ac309b1afcebe297c0526a8f82320139b56289", size = 225306, upload-time = "2025-10-04T21:55:35.312Z" },
+    { url = "https://files.pythonhosted.org/packages/84/fe/76884442da1bab6d4353ba1c43fdc4a770c3b3973f3ac7620a7205402fdd/propcache-0.4.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:399c73201d88c856a994916200d7cba41d7687096f8eb5139eb68f02785dc3f7", size = 230013, upload-time = "2025-10-04T21:55:37.005Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/b7/322af273bd1136bb7e13628821fb855c9f61d64651c73fea71dded68dda5/propcache-0.4.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a1d5e474d43c238035b74ecf997f655afa67f979bae591ac838bb3fbe3076392", size = 238331, upload-time = "2025-10-04T21:55:38.713Z" },
+    { url = "https://files.pythonhosted.org/packages/84/5e/036d2b105927ae7f179346c9911d16c345f4dba5a19a063f23a8d28acfbd/propcache-0.4.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:22f589652ee38de96aa58dd219335604e09666092bc250c1d9c26a55bcef9932", size = 221461, upload-time = "2025-10-04T21:55:40.034Z" },
+    { url = "https://files.pythonhosted.org/packages/63/0d/babd038efb12a87a46ab070438c52daeac6bed0a930693a418feef8cb8a6/propcache-0.4.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e5227da556b2939da6125cda1d5eecf9e412e58bc97b41e2f192605c3ccbb7c2", size = 216707, upload-time = "2025-10-04T21:55:41.455Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/68/dd075a037381581f16e7e504a6da9c1d7e415e945dd8ed67905d608f0687/propcache-0.4.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:92bc43a1ab852310721ce856f40a3a352254aa6f5e26f0fad870b31be45bba2e", size = 212591, upload-time = "2025-10-04T21:55:42.938Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/43/22698f28fc8e04c32b109cb9cb81305a4873b77c907b17484566b6133aef/propcache-0.4.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:83ae2f5343f6f06f4c91ae530d95f56b415f768f9c401a5ee2a10459cf74370b", size = 220188, upload-time = "2025-10-04T21:55:44.53Z" },
+    { url = "https://files.pythonhosted.org/packages/96/7a/27886e4a4c69598a38fbeeed64f9b8ddfa6f08fe3452035845a1fe90336f/propcache-0.4.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:077a32977399dc05299b16e793210341a0b511eb0a86d1796873e83ce47334cc", size = 226736, upload-time = "2025-10-04T21:55:46.348Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/c7/313c632b5888db3c9f4cb262420dcd5e57cf858d939d6ad9c3b1b90c12af/propcache-0.4.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:94a278c45e6463031b5a8278e40a07edf2bcc3b5379510e22b6c1a6e6498c194", size = 216363, upload-time = "2025-10-04T21:55:47.768Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/5d/5aaf82bd1542aedb47d10483b84f49ee8f00d970a58e27534cd241e9c5ac/propcache-0.4.0-cp312-cp312-win32.whl", hash = "sha256:4c491462e1dc80f9deb93f428aad8d83bb286de212837f58eb48e75606e7726c", size = 37945, upload-time = "2025-10-04T21:55:49.104Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/67/47ffff6eb176f383f56319f31c0e1bcf7500cb94ffb7582efc600c6b3c73/propcache-0.4.0-cp312-cp312-win_amd64.whl", hash = "sha256:cdb0cecafb528ab15ed89cdfed183074d15912d046d3e304955513b50a34b907", size = 41530, upload-time = "2025-10-04T21:55:50.261Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/7e/61b70306b9d7527286ce887a8ff28c304ab2514e5893eea36b5bdf7a21af/propcache-0.4.0-cp312-cp312-win_arm64.whl", hash = "sha256:b2f29697d1110e8cdf7a39cc630498df0082d7898b79b731c1c863f77c6e8cfc", size = 37662, upload-time = "2025-10-04T21:55:51.35Z" },
+    { url = "https://files.pythonhosted.org/packages/84/24/31f6efbca9211550377731642131ad274b01b2dc4788c0a7d5c47fcd9e1c/propcache-0.4.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:0cd30341142c68377cf3c4e2d9f0581e6e528694b2d57c62c786be441053d2fc", size = 80114, upload-time = "2025-10-04T21:57:16.407Z" },
+    { url = "https://files.pythonhosted.org/packages/21/96/295cb8cef28d50cfccf445689a1dc8de0b2e3fa923921404560ae7a423cc/propcache-0.4.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:2c46d37955820dd883cf9156ceb7825b8903e910bdd869902e20a5ac4ecd2c8b", size = 45697, upload-time = "2025-10-04T21:57:17.6Z" },
+    { url = "https://files.pythonhosted.org/packages/97/c0/3468b29275579cf533b25de083072b5fe11a7666106827a5ea704eb5bfc3/propcache-0.4.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:0b12df77eb19266efd146627a65b8ad414f9d15672d253699a50c8205661a820", size = 47510, upload-time = "2025-10-04T21:57:18.807Z" },
+    { url = "https://files.pythonhosted.org/packages/29/4e/fd5f46077a435e0866eb27a1a9771067ab41bfd35e14b1aa51c097bcd035/propcache-0.4.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1cdabd60e109506462e6a7b37008e57979e737dc6e7dfbe1437adcfe354d1a0a", size = 201095, upload-time = "2025-10-04T21:57:20.104Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/d0/cdbc660a9946e3d389db2541145046073f9ce767e0dbea13bced2657fcf5/propcache-0.4.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:65ff56a31f25925ef030b494fe63289bf07ef0febe6da181b8219146c590e185", size = 209505, upload-time = "2025-10-04T21:57:21.509Z" },
+    { url = "https://files.pythonhosted.org/packages/42/77/9ee6d0266bc33b5ee7c3eabddb48be30b141f6acddf964d0844ff521be6a/propcache-0.4.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:96153e037ae065bb71cae889f23c933190d81ae183f3696a030b47352fd8655d", size = 215226, upload-time = "2025-10-04T21:57:22.88Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/c6/72c4ed7751a82f60d14fb713734efcfcf190942288dd2056898bac4824d2/propcache-0.4.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4bf95be277fbb51513895c2cecc81ab12a421cdbd8837f159828a919a0167f96", size = 196960, upload-time = "2025-10-04T21:57:24.562Z" },
+    { url = "https://files.pythonhosted.org/packages/82/7e/e01179669192b94a640d942ad42535c38ba6cad70518db7354585e051d1c/propcache-0.4.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:8d18d796ffecdc8253742fd53a94ceee2e77ad149eb9ed5960c2856b5f692f71", size = 192189, upload-time = "2025-10-04T21:57:25.973Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/ea/0c2be8a7fdf012598583df72efd38220ef0db671fe9a9d8ff82f790bf313/propcache-0.4.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:4a52c25a51d5894ba60c567b0dbcf73de2f3cd642cf5343679e07ca3a768b085", size = 190520, upload-time = "2025-10-04T21:57:27.394Z" },
+    { url = "https://files.pythonhosted.org/packages/17/68/288e1e53c4f677906eac137d047720165aa1e10cd4d8b28f18d6b4c5c29a/propcache-0.4.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:e0ce7f3d1faf7ad58652ed758cc9753049af5308b38f89948aa71793282419c5", size = 199773, upload-time = "2025-10-04T21:57:28.733Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/52/eff7750f9d7a20000143fb857b6215fc6ac1e897ca62eeb20ef6b62af4ea/propcache-0.4.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:545987971b2aded25ba4698135ea0ae128836e7deb6e18c29a581076aaef44aa", size = 200567, upload-time = "2025-10-04T21:57:30.241Z" },
+    { url = "https://files.pythonhosted.org/packages/27/a5/1ecedd68f72554c7c9e5508158b7634eb84a24cb8880183d1cd74a433d6e/propcache-0.4.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:7da5c4c72ae40fd3ce87213ab057db66df53e55600d0b9e72e2b7f5a470a2cc4", size = 192963, upload-time = "2025-10-04T21:57:32.382Z" },
+    { url = "https://files.pythonhosted.org/packages/67/0a/aeaf727f5a0dcea135e2f90469d2410767d84ef386005219a50825381021/propcache-0.4.0-cp39-cp39-win32.whl", hash = "sha256:2015218812ee8f13bbaebc9f52b1e424cc130b68d4857bef018e65e3834e1c4d", size = 38223, upload-time = "2025-10-04T21:57:33.889Z" },
+    { url = "https://files.pythonhosted.org/packages/45/0b/604ba6b7ad807312a2a0125a64c46cd83a12399fc29c1646cc42cf814f9e/propcache-0.4.0-cp39-cp39-win_amd64.whl", hash = "sha256:39f0f6a3b56e82dc91d84c763b783c5c33720a33c70ee48a1c13ba800ac1fa69", size = 41782, upload-time = "2025-10-04T21:57:35.522Z" },
+    { url = "https://files.pythonhosted.org/packages/77/30/2107b450136c26fc5b3c3a811e80898b6343a4c514bedee4c6a3d098723e/propcache-0.4.0-cp39-cp39-win_arm64.whl", hash = "sha256:236c8da353ea7c22a8e963ab78cddb1126f700ae9538e2c4c6ef471e5545494b", size = 38329, upload-time = "2025-10-04T21:57:36.959Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/16/794c114f6041bbe2de23eb418ef58a0f45de27224d5540f5dbb266a73d72/propcache-0.4.0-py3-none-any.whl", hash = "sha256:015b2ca2f98ea9e08ac06eecc409d5d988f78c5fd5821b2ad42bc9afcd6b1557", size = 13183, upload-time = "2025-10-04T21:57:38.054Z" },
 ]
 
 [[package]]
 name = "protobuf"
-version = "5.29.5"
+version = "6.32.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/43/29/d09e70352e4e88c9c7a198d5645d7277811448d76c23b00345670f7c8a38/protobuf-5.29.5.tar.gz", hash = "sha256:bc1463bafd4b0929216c35f437a8e28731a2b7fe3d98bb77a600efced5a15c84", size = 425226, upload-time = "2025-05-28T23:51:59.82Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/fa/a4/cc17347aa2897568beece2e674674359f911d6fe21b0b8d6268cd42727ac/protobuf-6.32.1.tar.gz", hash = "sha256:ee2469e4a021474ab9baafea6cd070e5bf27c7d29433504ddea1a4ee5850f68d", size = 440635, upload-time = "2025-09-11T21:38:42.935Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/5f/11/6e40e9fc5bba02988a214c07cf324595789ca7820160bfd1f8be96e48539/protobuf-5.29.5-cp310-abi3-win32.whl", hash = "sha256:3f1c6468a2cfd102ff4703976138844f78ebd1fb45f49011afc5139e9e283079", size = 422963, upload-time = "2025-05-28T23:51:41.204Z" },
-    { url = "https://files.pythonhosted.org/packages/81/7f/73cefb093e1a2a7c3ffd839e6f9fcafb7a427d300c7f8aef9c64405d8ac6/protobuf-5.29.5-cp310-abi3-win_amd64.whl", hash = "sha256:3f76e3a3675b4a4d867b52e4a5f5b78a2ef9565549d4037e06cf7b0942b1d3fc", size = 434818, upload-time = "2025-05-28T23:51:44.297Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/73/10e1661c21f139f2c6ad9b23040ff36fee624310dc28fba20d33fdae124c/protobuf-5.29.5-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:e38c5add5a311f2a6eb0340716ef9b039c1dfa428b28f25a7838ac329204a671", size = 418091, upload-time = "2025-05-28T23:51:45.907Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/04/98f6f8cf5b07ab1294c13f34b4e69b3722bb609c5b701d6c169828f9f8aa/protobuf-5.29.5-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:fa18533a299d7ab6c55a238bf8629311439995f2e7eca5caaff08663606e9015", size = 319824, upload-time = "2025-05-28T23:51:47.545Z" },
-    { url = "https://files.pythonhosted.org/packages/85/e4/07c80521879c2d15f321465ac24c70efe2381378c00bf5e56a0f4fbac8cd/protobuf-5.29.5-cp38-abi3-manylinux2014_x86_64.whl", hash = "sha256:63848923da3325e1bf7e9003d680ce6e14b07e55d0473253a690c3a8b8fd6e61", size = 319942, upload-time = "2025-05-28T23:51:49.11Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/59/ca89678bb0352f094fc92f2b358daa40e3acc91a93aa8f922b24762bf841/protobuf-5.29.5-cp39-cp39-win32.whl", hash = "sha256:6f642dc9a61782fa72b90878af134c5afe1917c89a568cd3476d758d3c3a0736", size = 423025, upload-time = "2025-05-28T23:51:54.003Z" },
-    { url = "https://files.pythonhosted.org/packages/96/8b/2c62731fe3e92ddbbeca0174f78f0f8739197cdeb7c75ceb5aad3706963b/protobuf-5.29.5-cp39-cp39-win_amd64.whl", hash = "sha256:470f3af547ef17847a28e1f47200a1cbf0ba3ff57b7de50d22776607cd2ea353", size = 434906, upload-time = "2025-05-28T23:51:55.782Z" },
-    { url = "https://files.pythonhosted.org/packages/7e/cc/7e77861000a0691aeea8f4566e5d3aa716f2b1dece4a24439437e41d3d25/protobuf-5.29.5-py3-none-any.whl", hash = "sha256:6cf42630262c59b2d8de33954443d94b746c952b01434fc58a417fdbd2e84bd5", size = 172823, upload-time = "2025-05-28T23:51:58.157Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/98/645183ea03ab3995d29086b8bf4f7562ebd3d10c9a4b14ee3f20d47cfe50/protobuf-6.32.1-cp310-abi3-win32.whl", hash = "sha256:a8a32a84bc9f2aad712041b8b366190f71dde248926da517bde9e832e4412085", size = 424411, upload-time = "2025-09-11T21:38:27.427Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/f3/6f58f841f6ebafe076cebeae33fc336e900619d34b1c93e4b5c97a81fdfa/protobuf-6.32.1-cp310-abi3-win_amd64.whl", hash = "sha256:b00a7d8c25fa471f16bc8153d0e53d6c9e827f0953f3c09aaa4331c718cae5e1", size = 435738, upload-time = "2025-09-11T21:38:30.959Z" },
+    { url = "https://files.pythonhosted.org/packages/10/56/a8a3f4e7190837139e68c7002ec749190a163af3e330f65d90309145a210/protobuf-6.32.1-cp39-abi3-macosx_10_9_universal2.whl", hash = "sha256:d8c7e6eb619ffdf105ee4ab76af5a68b60a9d0f66da3ea12d1640e6d8dab7281", size = 426454, upload-time = "2025-09-11T21:38:34.076Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/be/8dd0a927c559b37d7a6c8ab79034fd167dcc1f851595f2e641ad62be8643/protobuf-6.32.1-cp39-abi3-manylinux2014_aarch64.whl", hash = "sha256:2f5b80a49e1eb7b86d85fcd23fe92df154b9730a725c3b38c4e43b9d77018bf4", size = 322874, upload-time = "2025-09-11T21:38:35.509Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/f6/88d77011b605ef979aace37b7703e4eefad066f7e84d935e5a696515c2dd/protobuf-6.32.1-cp39-abi3-manylinux2014_x86_64.whl", hash = "sha256:b1864818300c297265c83a4982fd3169f97122c299f56a56e2445c3698d34710", size = 322013, upload-time = "2025-09-11T21:38:37.017Z" },
+    { url = "https://files.pythonhosted.org/packages/05/9d/d6f1a8b6657296920c58f6b85f7bca55fa27e3ca7fc5914604d89cd0250b/protobuf-6.32.1-cp39-cp39-win32.whl", hash = "sha256:68ff170bac18c8178f130d1ccb94700cf72852298e016a2443bdb9502279e5f1", size = 424505, upload-time = "2025-09-11T21:38:38.415Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/cd/891bd2d23558f52392a5687b2406a741e2e28d629524c88aade457029acd/protobuf-6.32.1-cp39-cp39-win_amd64.whl", hash = "sha256:d0975d0b2f3e6957111aa3935d08a0eb7e006b1505d825f862a1fffc8348e122", size = 435825, upload-time = "2025-09-11T21:38:39.773Z" },
+    { url = "https://files.pythonhosted.org/packages/97/b7/15cc7d93443d6c6a84626ae3258a91f4c6ac8c0edd5df35ea7658f71b79c/protobuf-6.32.1-py3-none-any.whl", hash = "sha256:2601b779fc7d32a866c6b4404f9d42a3f67c5b9f3f15b4db3cccabe06b95c346", size = 169289, upload-time = "2025-09-11T21:38:41.234Z" },
 ]
 
 [[package]]
 name = "psutil"
-version = "7.0.0"
+version = "7.1.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/2a/80/336820c1ad9286a4ded7e845b2eccfcb27851ab8ac6abece774a6ff4d3de/psutil-7.0.0.tar.gz", hash = "sha256:7be9c3eba38beccb6495ea33afd982a44074b78f28c434a1f51cc07fd315c456", size = 497003, upload-time = "2025-02-13T21:54:07.946Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/b3/31/4723d756b59344b643542936e37a31d1d3204bcdc42a7daa8ee9eb06fb50/psutil-7.1.0.tar.gz", hash = "sha256:655708b3c069387c8b77b072fc429a57d0e214221d01c0a772df7dfedcb3bcd2", size = 497660, upload-time = "2025-09-17T20:14:52.902Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ed/e6/2d26234410f8b8abdbf891c9da62bee396583f713fb9f3325a4760875d22/psutil-7.0.0-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:101d71dc322e3cffd7cea0650b09b3d08b8e7c4109dd6809fe452dfd00e58b25", size = 238051, upload-time = "2025-02-13T21:54:12.36Z" },
-    { url = "https://files.pythonhosted.org/packages/04/8b/30f930733afe425e3cbfc0e1468a30a18942350c1a8816acfade80c005c4/psutil-7.0.0-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:39db632f6bb862eeccf56660871433e111b6ea58f2caea825571951d4b6aa3da", size = 239535, upload-time = "2025-02-13T21:54:16.07Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/ed/d362e84620dd22876b55389248e522338ed1bf134a5edd3b8231d7207f6d/psutil-7.0.0-cp36-abi3-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1fcee592b4c6f146991ca55919ea3d1f8926497a713ed7faaf8225e174581e91", size = 275004, upload-time = "2025-02-13T21:54:18.662Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/b9/b0eb3f3cbcb734d930fdf839431606844a825b23eaf9a6ab371edac8162c/psutil-7.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4b1388a4f6875d7e2aff5c4ca1cc16c545ed41dd8bb596cefea80111db353a34", size = 277986, upload-time = "2025-02-13T21:54:21.811Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/a2/709e0fe2f093556c17fbafda93ac032257242cabcc7ff3369e2cb76a97aa/psutil-7.0.0-cp36-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a5f098451abc2828f7dc6b58d44b532b22f2088f4999a937557b603ce72b1993", size = 279544, upload-time = "2025-02-13T21:54:24.68Z" },
-    { url = "https://files.pythonhosted.org/packages/50/e6/eecf58810b9d12e6427369784efe814a1eec0f492084ce8eb8f4d89d6d61/psutil-7.0.0-cp37-abi3-win32.whl", hash = "sha256:ba3fcef7523064a6c9da440fc4d6bd07da93ac726b5733c29027d7dc95b39d99", size = 241053, upload-time = "2025-02-13T21:54:34.31Z" },
-    { url = "https://files.pythonhosted.org/packages/50/1b/6921afe68c74868b4c9fa424dad3be35b095e16687989ebbb50ce4fceb7c/psutil-7.0.0-cp37-abi3-win_amd64.whl", hash = "sha256:4cf3d4eb1aa9b348dec30105c55cd9b7d4629285735a102beb4441e38db90553", size = 244885, upload-time = "2025-02-13T21:54:37.486Z" },
+    { url = "https://files.pythonhosted.org/packages/46/62/ce4051019ee20ce0ed74432dd73a5bb087a6704284a470bb8adff69a0932/psutil-7.1.0-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:76168cef4397494250e9f4e73eb3752b146de1dd950040b29186d0cce1d5ca13", size = 245242, upload-time = "2025-09-17T20:14:56.126Z" },
+    { url = "https://files.pythonhosted.org/packages/38/61/f76959fba841bf5b61123fbf4b650886dc4094c6858008b5bf73d9057216/psutil-7.1.0-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:5d007560c8c372efdff9e4579c2846d71de737e4605f611437255e81efcca2c5", size = 246682, upload-time = "2025-09-17T20:14:58.25Z" },
+    { url = "https://files.pythonhosted.org/packages/88/7a/37c99d2e77ec30d63398ffa6a660450b8a62517cabe44b3e9bae97696e8d/psutil-7.1.0-cp36-abi3-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:22e4454970b32472ce7deaa45d045b34d3648ce478e26a04c7e858a0a6e75ff3", size = 287994, upload-time = "2025-09-17T20:14:59.901Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/de/04c8c61232f7244aa0a4b9a9fbd63a89d5aeaf94b2fc9d1d16e2faa5cbb0/psutil-7.1.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8c70e113920d51e89f212dd7be06219a9b88014e63a4cec69b684c327bc474e3", size = 291163, upload-time = "2025-09-17T20:15:01.481Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/58/c4f976234bf6d4737bc8c02a81192f045c307b72cf39c9e5c5a2d78927f6/psutil-7.1.0-cp36-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7d4a113425c037300de3ac8b331637293da9be9713855c4fc9d2d97436d7259d", size = 293625, upload-time = "2025-09-17T20:15:04.492Z" },
+    { url = "https://files.pythonhosted.org/packages/79/87/157c8e7959ec39ced1b11cc93c730c4fb7f9d408569a6c59dbd92ceb35db/psutil-7.1.0-cp37-abi3-win32.whl", hash = "sha256:09ad740870c8d219ed8daae0ad3b726d3bf9a028a198e7f3080f6a1888b99bca", size = 244812, upload-time = "2025-09-17T20:15:07.462Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/e9/b44c4f697276a7a95b8e94d0e320a7bf7f3318521b23de69035540b39838/psutil-7.1.0-cp37-abi3-win_amd64.whl", hash = "sha256:57f5e987c36d3146c0dd2528cd42151cf96cd359b9d67cfff836995cc5df9a3d", size = 247965, upload-time = "2025-09-17T20:15:09.673Z" },
+    { url = "https://files.pythonhosted.org/packages/26/65/1070a6e3c036f39142c2820c4b52e9243246fcfc3f96239ac84472ba361e/psutil-7.1.0-cp37-abi3-win_arm64.whl", hash = "sha256:6937cb68133e7c97b6cc9649a570c9a18ba0efebed46d8c5dae4c07fa1b67a07", size = 244971, upload-time = "2025-09-17T20:15:12.262Z" },
 ]
 
 [[package]]
@@ -3236,9 +3685,102 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/47/8d/d529b5d697919ba8c11ad626e835d4039be708a35b0d22de83a269a6682c/pyasn1_modules-0.4.2-py3-none-any.whl", hash = "sha256:29253a9207ce32b64c3ac6600edc75368f98473906e8fd1043bd6b5b1de2c14a", size = 181259, upload-time = "2025-03-28T02:41:19.028Z" },
 ]
 
+[[package]]
+name = "pybase64"
+version = "1.4.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/04/14/43297a7b7f0c1bf0c00b596f754ee3ac946128c64d21047ccf9c9bbc5165/pybase64-1.4.2.tar.gz", hash = "sha256:46cdefd283ed9643315d952fe44de80dc9b9a811ce6e3ec97fd1827af97692d0", size = 137246, upload-time = "2025-07-27T13:08:57.808Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1d/d8/30ea35dc2c8c568be93e1379efcaa35092e37efa2ce7f1985ccc63babee7/pybase64-1.4.2-cp310-cp310-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:1eef93c29cc5567480d168f9cc1ebd3fc3107c65787aed2019a8ea68575a33e0", size = 65963, upload-time = "2025-07-27T13:02:22.376Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/da/1c22f2a21d6bb9ec2a214d15ae02d5b20a95335de218a0ecbf769c535a5c/pybase64-1.4.2-cp310-cp310-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:17b871a34aaeb0644145cb6bf28feb163f593abea11aec3dbcc34a006edfc828", size = 68887, upload-time = "2025-07-27T13:02:23.606Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/8d/e04d489ba99b444ce94b4d5b232365d00b0f0e8564275d7ba7434dcabe72/pybase64-1.4.2-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:1f734e16293637a35d282ce594eb05a7a90ea3ae2bc84a3496a5df9e6b890725", size = 57503, upload-time = "2025-07-27T13:02:24.83Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/b8/5ec9c334f30cf898709a084d596bf4b47aec2e07870f07bac5cf39754eca/pybase64-1.4.2-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:22bd38db2d990d5545dde83511edeec366630d00679dbd945472315c09041dc6", size = 54517, upload-time = "2025-07-27T13:02:26.006Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/5a/6e4424ecca041e53aa7c14525f99edd43d0117c23c5d9cb14e931458a536/pybase64-1.4.2-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:dc65cee686dda72007b7541b2014f33ee282459c781b9b61305bd8b9cfadc8e1", size = 57167, upload-time = "2025-07-27T13:02:27.47Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/d0/13f1a9467cf565eecc21dce89fb0723458d8c563d2ccfb99b96e8318dfd5/pybase64-1.4.2-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:1e79641c420a22e49c67c046895efad05bf5f8b1dbe0dd78b4af3ab3f2923fe2", size = 57718, upload-time = "2025-07-27T13:02:28.631Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/34/d80335c36ad9400b18b4f92e9f680cf7646102fe4919f7bce5786a2ccb7b/pybase64-1.4.2-cp310-cp310-manylinux_2_31_riscv64.whl", hash = "sha256:12f5e7db522ef780a8b333dab5f7d750d270b23a1684bc2235ba50756c7ba428", size = 53021, upload-time = "2025-07-27T13:02:29.823Z" },
+    { url = "https://files.pythonhosted.org/packages/68/57/504ff75f7c78df28be126fe6634083d28d7f84c17e04a74a7dcb50ab2377/pybase64-1.4.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:a618b1e1a63e75dd40c2a397d875935ed0835464dc55cb1b91e8f880113d0444", size = 56306, upload-time = "2025-07-27T13:02:31.314Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/bc/2d21cda8b73c8c9f5cd3d7e6e26dd6dfc96491052112f282332a3d5bf1d9/pybase64-1.4.2-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:89b0a51702c7746fa914e75e680ad697b979cdead6b418603f56a6fc9de2f50f", size = 50101, upload-time = "2025-07-27T13:02:32.662Z" },
+    { url = "https://files.pythonhosted.org/packages/88/6d/51942e7737bb0711ca3e55db53924fd7f07166d79da5508ab8f5fd5972a8/pybase64-1.4.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:c5161b8b82f8ba5dbbc3f76e0270622a2c2fdb9ffaf092d8f774ad7ec468c027", size = 66555, upload-time = "2025-07-27T13:02:34.122Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/c8/c46024d196402e7be4d3fad85336863a34816c3436c51fcf9c7c0781bf11/pybase64-1.4.2-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:2168de920c9b1e57850e9ff681852923a953601f73cc96a0742a42236695c316", size = 55684, upload-time = "2025-07-27T13:02:35.427Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/c5/953782c9d599ff5217ee87f19e317c494cd4840afcab4c48f99cb78ca201/pybase64-1.4.2-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:7a1e3dc977562abe40ab43483223013be71b215a5d5f3c78a666e70a5076eeec", size = 52475, upload-time = "2025-07-27T13:02:36.634Z" },
+    { url = "https://files.pythonhosted.org/packages/05/fb/57d36173631aab67ca4558cdbde1047fc67a09b77f9c53addd57c7e9fdd4/pybase64-1.4.2-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:4cf1e8a57449e48137ef4de00a005e24c3f1cffc0aafc488e36ceb5bb2cbb1da", size = 53943, upload-time = "2025-07-27T13:02:37.777Z" },
+    { url = "https://files.pythonhosted.org/packages/75/73/23e5bb0bffac0cabe2d11d1c618f6ef73da9f430da03c5249931e3c49b63/pybase64-1.4.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:d8e1a381ba124f26a93d5925efbf6e6c36287fc2c93d74958e8b677c30a53fc0", size = 68411, upload-time = "2025-07-27T13:02:39.302Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/8f/f9b49926a60848ba98350dd648227ec524fb78340b47a450c4dbaf24b1bb/pybase64-1.4.2-cp311-cp311-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:37f133e8c96427995480bb6d396d9d49e949a3e829591845bb6a5a7f215ca177", size = 68318, upload-time = "2025-07-27T13:02:46.644Z" },
+    { url = "https://files.pythonhosted.org/packages/29/9b/6ed2dd2bc8007f33b8316d6366b0901acbdd5665b419c2893b3dd48708de/pybase64-1.4.2-cp311-cp311-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:a6ee3874b0abbdd4c903d3989682a3f016fd84188622879f6f95a5dc5718d7e5", size = 71357, upload-time = "2025-07-27T13:02:47.937Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/69/be9ac8127da8d8339db7129683bd2975cecb0bf40a82731e1a492577a177/pybase64-1.4.2-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5c69f177b1e404b22b05802127d6979acf4cb57f953c7de9472410f9c3fdece7", size = 59817, upload-time = "2025-07-27T13:02:49.163Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/a2/e3e09e000b509609276ee28b71beb0b61462d4a43b3e0db0a44c8652880c/pybase64-1.4.2-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:80c817e88ef2ca3cc9a285fde267690a1cb821ce0da4848c921c16f0fec56fda", size = 56639, upload-time = "2025-07-27T13:02:50.384Z" },
+    { url = "https://files.pythonhosted.org/packages/01/70/ad7eff88aa4f1be06db705812e1f01749606933bf8fe9df553bb04b703e6/pybase64-1.4.2-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:7a4bb6e7e45bfdaea0f2aaf022fc9a013abe6e46ccea31914a77e10f44098688", size = 59368, upload-time = "2025-07-27T13:02:51.883Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/82/0cd1b4bcd2a4da7805cfa04587be783bf9583b34ac16cadc29cf119a4fa2/pybase64-1.4.2-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:2710a80d41a2b41293cb0e5b84b5464f54aa3f28f7c43de88784d2d9702b8a1c", size = 59981, upload-time = "2025-07-27T13:02:53.16Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/4c/8029a03468307dfaf0f9694d31830487ee43af5f8a73407004907724e8ac/pybase64-1.4.2-cp311-cp311-manylinux_2_31_riscv64.whl", hash = "sha256:aa6122c8a81f6597e1c1116511f03ed42cf377c2100fe7debaae7ca62521095a", size = 54908, upload-time = "2025-07-27T13:02:54.363Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/8b/70bd0fe659e242efd0f60895a8ce1fe88e3a4084fd1be368974c561138c9/pybase64-1.4.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b7e22b02505d64db308e9feeb6cb52f1d554ede5983de0befa59ac2d2ffb6a5f", size = 58650, upload-time = "2025-07-27T13:02:55.905Z" },
+    { url = "https://files.pythonhosted.org/packages/64/ca/9c1d23cbc4b9beac43386a32ad53903c816063cef3f14c10d7c3d6d49a23/pybase64-1.4.2-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:edfe4a3c8c4007f09591f49b46a89d287ef5e8cd6630339536fe98ff077263c2", size = 52323, upload-time = "2025-07-27T13:02:57.192Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/29/a6292e9047248c8616dc53131a49da6c97a61616f80e1e36c73d7ef895fe/pybase64-1.4.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:b79b4a53dd117ffbd03e96953f2e6bd2827bfe11afeb717ea16d9b0893603077", size = 68979, upload-time = "2025-07-27T13:02:58.594Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/e0/cfec7b948e170395d8e88066e01f50e71195db9837151db10c14965d6222/pybase64-1.4.2-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:fd9afa7a61d89d170607faf22287290045757e782089f0357b8f801d228d52c3", size = 58037, upload-time = "2025-07-27T13:02:59.753Z" },
+    { url = "https://files.pythonhosted.org/packages/74/7e/0ac1850198c9c35ef631174009cee576f4d8afff3bf493ce310582976ab4/pybase64-1.4.2-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:5c17b092e4da677a595178d2db17a5d2fafe5c8e418d46c0c4e4cde5adb8cff3", size = 54416, upload-time = "2025-07-27T13:03:00.978Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/45/b0b037f27e86c50e62d927f0bc1bde8b798dd55ab39197b116702e508d05/pybase64-1.4.2-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:120799274cf55f3f5bb8489eaa85142f26170564baafa7cf3e85541c46b6ab13", size = 56257, upload-time = "2025-07-27T13:03:02.201Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/0d/5034598aac56336d88fd5aaf6f34630330643b51d399336b8c788d798fc5/pybase64-1.4.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:522e4e712686acec2d25de9759dda0b0618cb9f6588523528bc74715c0245c7b", size = 70889, upload-time = "2025-07-27T13:03:03.437Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/2a/a24c810e7a61d2cc6f73fe9ee4872a03030887fa8654150901b15f376f65/pybase64-1.4.2-cp312-cp312-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:f48c32ac6a16cbf57a5a96a073fef6ff7e3526f623cd49faa112b7f9980bafba", size = 68192, upload-time = "2025-07-27T13:03:11.467Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/87/d9baf98cbfc37b8657290ad4421f3a3c36aa0eafe4872c5859cfb52f3448/pybase64-1.4.2-cp312-cp312-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:ace8b23093a6bb862477080d9059b784096ab2f97541e8bfc40d42f062875149", size = 71587, upload-time = "2025-07-27T13:03:12.719Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/89/3df043cc56ef3b91b7aa0c26ae822a2d7ec8da0b0fd7c309c879b0eb5988/pybase64-1.4.2-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:1772c7532a7fb6301baea3dd3e010148dbf70cd1136a83c2f5f91bdc94822145", size = 59910, upload-time = "2025-07-27T13:03:14.266Z" },
+    { url = "https://files.pythonhosted.org/packages/75/4f/6641e9edf37aeb4d4524dc7ba2168eff8d96c90e77f6283c2be3400ab380/pybase64-1.4.2-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:f86f7faddcba5cbfea475f8ab96567834c28bf09ca6c7c3d66ee445adac80d8f", size = 56701, upload-time = "2025-07-27T13:03:15.6Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/7f/20d8ac1046f12420a0954a45a13033e75f98aade36eecd00c64e3549b071/pybase64-1.4.2-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:0b8c8e275b5294089f314814b4a50174ab90af79d6a4850f6ae11261ff6a7372", size = 59288, upload-time = "2025-07-27T13:03:16.823Z" },
+    { url = "https://files.pythonhosted.org/packages/17/ea/9c0ca570e3e50b3c6c3442e280c83b321a0464c86a9db1f982a4ff531550/pybase64-1.4.2-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:864d85a0470c615807ae8b97d724d068b940a2d10ac13a5f1b9e75a3ce441758", size = 60267, upload-time = "2025-07-27T13:03:18.132Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/ac/46894929d71ccedebbfb0284173b0fea96bc029cd262654ba8451a7035d6/pybase64-1.4.2-cp312-cp312-manylinux_2_31_riscv64.whl", hash = "sha256:47254d97ed2d8351e30ecfdb9e2414547f66ba73f8a09f932c9378ff75cd10c5", size = 54801, upload-time = "2025-07-27T13:03:19.669Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/1e/02c95218ea964f0b2469717c2c69b48e63f4ca9f18af01a5b2a29e4c1216/pybase64-1.4.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:264b65ecc4f0ee73f3298ab83bbd8008f7f9578361b8df5b448f985d8c63e02a", size = 58599, upload-time = "2025-07-27T13:03:20.951Z" },
+    { url = "https://files.pythonhosted.org/packages/15/45/ccc21004930789b8fb439d43e3212a6c260ccddb2bf450c39a20db093f33/pybase64-1.4.2-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:fbcc2b30cd740c16c9699f596f22c7a9e643591311ae72b1e776f2d539e9dd9d", size = 52388, upload-time = "2025-07-27T13:03:23.064Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/45/22e46e549710c4c237d77785b6fb1bc4c44c288a5c44237ba9daf5c34b82/pybase64-1.4.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:cda9f79c22d51ee4508f5a43b673565f1d26af4330c99f114e37e3186fdd3607", size = 68802, upload-time = "2025-07-27T13:03:24.673Z" },
+    { url = "https://files.pythonhosted.org/packages/55/0c/232c6261b81296e5593549b36e6e7884a5da008776d12665923446322c36/pybase64-1.4.2-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:0c91c6d2a7232e2a1cd10b3b75a8bb657defacd4295a1e5e80455df2dfc84d4f", size = 57841, upload-time = "2025-07-27T13:03:25.948Z" },
+    { url = "https://files.pythonhosted.org/packages/20/8a/b35a615ae6f04550d696bb179c414538b3b477999435fdd4ad75b76139e4/pybase64-1.4.2-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:a370dea7b1cee2a36a4d5445d4e09cc243816c5bc8def61f602db5a6f5438e52", size = 54320, upload-time = "2025-07-27T13:03:27.495Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/a9/8bd4f9bcc53689f1b457ecefed1eaa080e4949d65a62c31a38b7253d5226/pybase64-1.4.2-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:9aa4de83f02e462a6f4e066811c71d6af31b52d7484de635582d0e3ec3d6cc3e", size = 56482, upload-time = "2025-07-27T13:03:28.942Z" },
+    { url = "https://files.pythonhosted.org/packages/75/e5/4a7735b54a1191f61c3f5c2952212c85c2d6b06eb5fb3671c7603395f70c/pybase64-1.4.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:83a1c2f9ed00fee8f064d548c8654a480741131f280e5750bb32475b7ec8ee38", size = 70959, upload-time = "2025-07-27T13:03:30.171Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/93/fd6f0cb1356ef849cef347ae3192a91c5beb944d485b2b97fd760265b7ad/pybase64-1.4.2-cp39-cp39-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:2089a72b04a62f63e0eac202ecff4440fb52fd05cd5f4ab9fe7e07839fedb9e9", size = 65747, upload-time = "2025-07-27T13:06:39.216Z" },
+    { url = "https://files.pythonhosted.org/packages/57/fc/75f8567866efcbc1c984e1bfb966ce92790fc38d5197f2d01a6131cb77f8/pybase64-1.4.2-cp39-cp39-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:bad101c24dcd23ed6fd6ea24c4a1b36ac7abc5eb07447dd7fa98b33859aed871", size = 68664, upload-time = "2025-07-27T13:06:41.908Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/19/8dbd3b29d54cce1ee82a0da1ff7473e2958057932af5788aa55f6f4d4ab5/pybase64-1.4.2-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:28592c88a9cf6fd27c9f191fb41688c1c27f57493d874cbc50e72e1cc2a3b854", size = 57325, upload-time = "2025-07-27T13:06:44.271Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/f7/e20978705feb84242c402ce7e626d9e8f7de638d2a6abf60e1384769d7f1/pybase64-1.4.2-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:0b5639fa2ceb3095393bd56dca8c16079717c361dd3a75439c9a8b8d679f4cf0", size = 54286, upload-time = "2025-07-27T13:06:46.974Z" },
+    { url = "https://files.pythonhosted.org/packages/39/f5/815813fa85168b3afcc56eb085ef7acb8f03f36e1b470837b98864dc847b/pybase64-1.4.2-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:49630338d4c321336d0dfc4c2c23162a87d9ebc8bb8879348ae019ac8a4366de", size = 56951, upload-time = "2025-07-27T13:06:49.582Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/eb/dc61ec538c6fa2cf86318e7e393fa4772cc363f5bc13a760bc241ef3f942/pybase64-1.4.2-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:c3d9f9881d7315e1d04d72aa7b3f40e2059bdbfdcec51939016409417725c952", size = 57442, upload-time = "2025-07-27T13:06:52.347Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/80/9d7286b25ff9952dcedd84456966916429d2bdb89cc826f7a11e6c9e8507/pybase64-1.4.2-cp39-cp39-manylinux_2_31_riscv64.whl", hash = "sha256:8e1226939eac9ce1f080d1b0a8afafee3140e277a4c40ccb306d82de396a41a8", size = 52774, upload-time = "2025-07-27T13:06:54.863Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/5e/7bd80d6c8ebb88918749a0a177e0328a0954b8d8f35dbd55c3bc395ed050/pybase64-1.4.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:69f424a227ec503742bac69b89e232c474dc199cd98c3e58e91020c1c4bad0ad", size = 56061, upload-time = "2025-07-27T13:06:57.547Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/29/a2e7b8b90acbeb3d8fc58589212ee6124fb102d6db6e52c37d76f5f679d9/pybase64-1.4.2-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:171ae85837de14d3691d5c4f29f5bb551209930c063a2cab6f5feb270aec66db", size = 49865, upload-time = "2025-07-27T13:07:00.076Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/ed/2d02e78c4f77ae70210a16399c734f8adfbfb4247c37a6924ae0285b23b3/pybase64-1.4.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:a55a13493fd165c3a619080149eda6f31c05c04c0577da9c9ef63d23f3abf374", size = 66341, upload-time = "2025-07-27T13:07:02.9Z" },
+    { url = "https://files.pythonhosted.org/packages/64/44/8ba4a9ec8a4bbb7b36068c6b4f85f6443960456f60ddd694d92de39d5b55/pybase64-1.4.2-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:06801fdc7fa83eac5cb7d1c7051bb623a25af8cb40e088671fa51a393d1053ad", size = 55470, upload-time = "2025-07-27T13:07:05.82Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/2f/d2b233ce4cefe80c5dbaedd8f2d11860ae2ef0b00233f70d2c7a46e59974/pybase64-1.4.2-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:7f2fbd6870228e9c8c3e2e2622ed7615a8d0159125b85e9d6c2d8e9ead74cdf0", size = 52242, upload-time = "2025-07-27T13:07:08.923Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/40/bcabd4398565a15329d93588aa0f15af5d0ae02a05f3388937bebb56846e/pybase64-1.4.2-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:1726017f04da880d10a57f078d117fe62532b5ed7bd58bd3318f3364b9767d91", size = 53739, upload-time = "2025-07-27T13:07:11.825Z" },
+    { url = "https://files.pythonhosted.org/packages/59/47/7f947e3efe660a8dc14c8c956e8165e8631533cf2dc1bbd3920f8028a4b0/pybase64-1.4.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1264f7fa417de7183732761f37c8ceb4652662a84f04538a28dadd5d84bf9a4a", size = 68149, upload-time = "2025-07-27T13:07:14.372Z" },
+    { url = "https://files.pythonhosted.org/packages/51/34/f40d3262c3953814b9bcdcf858436bd5bc1133a698be4bcc7ed2a8c0730d/pybase64-1.4.2-graalpy311-graalpy242_311_native-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:eef9255d926c64e2fca021d3aee98023bacb98e1518e5986d6aab04102411b04", size = 43212, upload-time = "2025-07-27T13:07:31.327Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/2a/5e05d25718cb8ffd68bd46553ddfd2b660893d937feda1716b8a3b21fb38/pybase64-1.4.2-graalpy311-graalpy242_311_native-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:89614ea2d2329b6708746c540e0f14d692125df99fb1203ff0de948d9e68dfc9", size = 35789, upload-time = "2025-07-27T13:07:34.026Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/da/2b7e69abfc62abe4d54b10d1e09ec78021a6b9b2d7e6e7b632243a19433e/pybase64-1.4.2-pp310-pypy310_pp73-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:4f98c5c6152d3c01d933fcde04322cd9ddcf65b5346034aac69a04c1a7cbb012", size = 40667, upload-time = "2025-07-27T13:07:46.715Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/11/ba738655fb3ba85c7a0605eddd2709fef606e654840c72ee5c5ff7ab29bf/pybase64-1.4.2-pp310-pypy310_pp73-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:9096a4977b7aff7ef250f759fb6a4b6b7b6199d99c84070c7fc862dd3b208b34", size = 41290, upload-time = "2025-07-27T13:07:49.534Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/38/2d5502fcaf712297b95c1b6ca924656dd7d17501fd7f9c9e0b3bbf8892ef/pybase64-1.4.2-pp310-pypy310_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:49d8597e2872966399410502310b1e2a5b7e8d8ba96766ee1fe242e00bd80775", size = 35438, upload-time = "2025-07-27T13:07:52.327Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/dd/c819003b59b2832256b72ad23cbeadbd95d083ef0318d07149a58b7a88af/pybase64-1.4.2-pp311-pypy311_pp73-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:1159e70cba8e76c3d8f334bd1f8fd52a1bb7384f4c3533831b23ab2df84a6ef3", size = 40668, upload-time = "2025-07-27T13:08:04.176Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/c5/38c6aba28678c4a4db49312a6b8171b93a0ffe9f21362cf4c0f325caa850/pybase64-1.4.2-pp311-pypy311_pp73-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7d943bc5dad8388971494554b97f22ae06a46cc7779ad0de3d4bfdf7d0bbea30", size = 41281, upload-time = "2025-07-27T13:08:07.395Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/23/5927bd9e59714e4e8cefd1d21ccd7216048bb1c6c3e7104b1b200afdc63d/pybase64-1.4.2-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:10b99182c561d86422c5de4265fd1f8f172fb38efaed9d72c71fb31e279a7f94", size = 35433, upload-time = "2025-07-27T13:08:10.551Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/0a/7d19d99844f6cd4d67119a7d781e7997a5254dab86c47736c4b138f4e148/pybase64-1.4.2-pp39-pypy39_pp73-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:45f078139d76194024e59b4bcfa64d42e5a5f8a5a4ea55ca4d27df46989c5e32", size = 40661, upload-time = "2025-07-27T13:08:43.922Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/78/20e0271084d3687e33c3858520c8808ad47128d3153e0f760226f36abf88/pybase64-1.4.2-pp39-pypy39_pp73-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:06305e602f128b289b98490a2d07d9d78e7e781e32e7b0252c2e71084fd19edf", size = 41286, upload-time = "2025-07-27T13:08:47.264Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/b1/b8120705861192e9ffb2c73d77b581607dbfe0d41d81e086744b4f134d6e/pybase64-1.4.2-pp39-pypy39_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d58eb4cb50b6466cef2e25761a5c915a8d57feda53165cced537a7ce0421b928", size = 35432, upload-time = "2025-07-27T13:08:50.798Z" },
+]
+
+[[package]]
+name = "pycountry"
+version = "24.6.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/76/57/c389fa68c50590881a75b7883eeb3dc15e9e73a0fdc001cdd45c13290c92/pycountry-24.6.1.tar.gz", hash = "sha256:b61b3faccea67f87d10c1f2b0fc0be714409e8fcdcc1315613174f6466c10221", size = 6043910, upload-time = "2024-06-01T04:12:15.05Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b1/ec/1fb891d8a2660716aadb2143235481d15ed1cbfe3ad669194690b0604492/pycountry-24.6.1-py3-none-any.whl", hash = "sha256:f1a4fb391cd7214f8eefd39556d740adcc233c778a27f8942c8dca351d6ce06f", size = 6335189, upload-time = "2024-06-01T04:11:49.711Z" },
+]
+
+[[package]]
+name = "pycparser"
+version = "2.23"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/fe/cf/d2d3b9f5699fb1e4615c8e32ff220203e43b248e1dfcc6736ad9057731ca/pycparser-2.23.tar.gz", hash = "sha256:78816d4f24add8f10a06d6f05b4d424ad9e96cfebf68a4ddc99c65c0720d00c2", size = 173734, upload-time = "2025-09-09T13:23:47.91Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a0/e3/59cd50310fc9b59512193629e1984c1f95e5c8ae6e5d8c69532ccc65a7fe/pycparser-2.23-py3-none-any.whl", hash = "sha256:e5c6e8d3fbad53479cab09ac03729e0a9faf2bee3db8208a550daf5af81a5934", size = 118140, upload-time = "2025-09-09T13:23:46.651Z" },
+]
+
 [[package]]
 name = "pydantic"
-version = "2.11.7"
+version = "2.11.10"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "annotated-types" },
@@ -3246,9 +3788,14 @@ dependencies = [
     { name = "typing-extensions" },
     { name = "typing-inspection" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/00/dd/4325abf92c39ba8623b5af936ddb36ffcfe0beae70405d456ab1fb2f5b8c/pydantic-2.11.7.tar.gz", hash = "sha256:d989c3c6cb79469287b1569f7447a17848c998458d49ebe294e975b9baf0f0db", size = 788350, upload-time = "2025-06-14T08:33:17.137Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/ae/54/ecab642b3bed45f7d5f59b38443dcb36ef50f85af192e6ece103dbfe9587/pydantic-2.11.10.tar.gz", hash = "sha256:dc280f0982fbda6c38fada4e476dc0a4f3aeaf9c6ad4c28df68a666ec3c61423", size = 788494, upload-time = "2025-10-04T10:40:41.338Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6a/c0/ec2b1c8712ca690e5d61979dee872603e92b8a32f94cc1b72d53beab008a/pydantic-2.11.7-py3-none-any.whl", hash = "sha256:dde5df002701f6de26248661f6835bbe296a47bf73990135c7d07ce741b9623b", size = 444782, upload-time = "2025-06-14T08:33:14.905Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/1f/73c53fcbfb0b5a78f91176df41945ca466e71e9d9d836e5c522abda39ee7/pydantic-2.11.10-py3-none-any.whl", hash = "sha256:802a655709d49bd004c31e865ef37da30b540786a46bfce02333e0e24b5fe29a", size = 444823, upload-time = "2025-10-04T10:40:39.055Z" },
+]
+
+[package.optional-dependencies]
+email = [
+    { name = "email-validator" },
 ]
 
 [[package]]
@@ -3343,18 +3890,36 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/d4/29/3cade8a924a61f60ccfa10842f75eb12787e1440e2b8660ceffeb26685e7/pydantic_core-2.33.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:2807668ba86cb38c6817ad9bc66215ab8584d1d304030ce4f0887336f28a5e27", size = 2066661, upload-time = "2025-04-23T18:33:49.995Z" },
 ]
 
+[[package]]
+name = "pydantic-extra-types"
+version = "2.10.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "pydantic" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/7e/ba/4178111ec4116c54e1dc7ecd2a1ff8f54256cdbd250e576882911e8f710a/pydantic_extra_types-2.10.5.tar.gz", hash = "sha256:1dcfa2c0cf741a422f088e0dbb4690e7bfadaaf050da3d6f80d6c3cf58a2bad8", size = 138429, upload-time = "2025-06-02T09:31:52.713Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/70/1a/5f4fd9e7285f10c44095a4f9fe17d0f358d1702a7c74a9278c794e8a7537/pydantic_extra_types-2.10.5-py3-none-any.whl", hash = "sha256:b60c4e23d573a69a4f1a16dd92888ecc0ef34fb0e655b4f305530377fa70e7a8", size = 38315, upload-time = "2025-06-02T09:31:51.229Z" },
+]
+
+[package.optional-dependencies]
+pycountry = [
+    { name = "pycountry" },
+]
+
 [[package]]
 name = "pydantic-settings"
-version = "2.10.1"
+version = "2.11.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "pydantic" },
     { name = "python-dotenv" },
     { name = "typing-inspection" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/68/85/1ea668bbab3c50071ca613c6ab30047fb36ab0da1b92fa8f17bbc38fd36c/pydantic_settings-2.10.1.tar.gz", hash = "sha256:06f0062169818d0f5524420a360d632d5857b83cffd4d42fe29597807a1614ee", size = 172583, upload-time = "2025-06-24T13:26:46.841Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/20/c5/dbbc27b814c71676593d1c3f718e6cd7d4f00652cefa24b75f7aa3efb25e/pydantic_settings-2.11.0.tar.gz", hash = "sha256:d0e87a1c7d33593beb7194adb8470fc426e95ba02af83a0f23474a04c9a08180", size = 188394, upload-time = "2025-09-24T14:19:11.764Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/58/f0/427018098906416f580e3cf1366d3b1abfb408a0652e9f31600c24a1903c/pydantic_settings-2.10.1-py3-none-any.whl", hash = "sha256:a60952460b99cf661dc25c29c0ef171721f98bfcb52ef8d9ea4c943d7c8cc796", size = 45235, upload-time = "2025-06-24T13:26:45.485Z" },
+    { url = "https://files.pythonhosted.org/packages/83/d6/887a1ff844e64aa823fb4905978d882a633cfe295c32eacad582b78a7d8b/pydantic_settings-2.11.0-py3-none-any.whl", hash = "sha256:fe2cea3413b9530d10f3a5875adffb17ada5c1e1bab0b2885546d7310415207c", size = 48608, upload-time = "2025-09-24T14:19:10.015Z" },
 ]
 
 [[package]]
@@ -3368,7 +3933,7 @@ wheels = [
 
 [[package]]
 name = "pytest"
-version = "8.4.1"
+version = "8.4.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "colorama", marker = "sys_platform == 'win32'" },
@@ -3379,23 +3944,23 @@ dependencies = [
     { name = "pygments" },
     { name = "tomli", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/08/ba/45911d754e8eba3d5a841a5ce61a65a685ff1798421ac054f85aa8747dfb/pytest-8.4.1.tar.gz", hash = "sha256:7c67fd69174877359ed9371ec3af8a3d2b04741818c51e5e99cc1742251fa93c", size = 1517714, upload-time = "2025-06-18T05:48:06.109Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/a3/5c/00a0e072241553e1a7496d638deababa67c5058571567b92a7eaa258397c/pytest-8.4.2.tar.gz", hash = "sha256:86c0d0b93306b961d58d62a4db4879f27fe25513d4b969df351abdddb3c30e01", size = 1519618, upload-time = "2025-09-04T14:34:22.711Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/29/16/c8a903f4c4dffe7a12843191437d7cd8e32751d5de349d45d3fe69544e87/pytest-8.4.1-py3-none-any.whl", hash = "sha256:539c70ba6fcead8e78eebbf1115e8b589e7565830d7d006a8723f19ac8a0afb7", size = 365474, upload-time = "2025-06-18T05:48:03.955Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/a4/20da314d277121d6534b3a980b29035dcd51e6744bd79075a6ce8fa4eb8d/pytest-8.4.2-py3-none-any.whl", hash = "sha256:872f880de3fc3a5bdc88a11b39c9710c3497a547cfa9320bc3c5e62fbf272e79", size = 365750, upload-time = "2025-09-04T14:34:20.226Z" },
 ]
 
 [[package]]
 name = "pytest-asyncio"
-version = "1.1.0"
+version = "1.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "backports-asyncio-runner", marker = "python_full_version < '3.11'" },
     { name = "pytest" },
-    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/4e/51/f8794af39eeb870e87a8c8068642fc07bce0c854d6865d7dd0f2a9d338c2/pytest_asyncio-1.1.0.tar.gz", hash = "sha256:796aa822981e01b68c12e4827b8697108f7205020f24b5793b3c41555dab68ea", size = 46652, upload-time = "2025-07-16T04:29:26.393Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/42/86/9e3c5f48f7b7b638b216e4b9e645f54d199d7abbbab7a64a13b4e12ba10f/pytest_asyncio-1.2.0.tar.gz", hash = "sha256:c609a64a2a8768462d0c99811ddb8bd2583c33fd33cf7f21af1c142e824ffb57", size = 50119, upload-time = "2025-09-12T07:33:53.816Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c7/9d/bf86eddabf8c6c9cb1ea9a869d6873b46f105a5d292d3a6f7071f5b07935/pytest_asyncio-1.1.0-py3-none-any.whl", hash = "sha256:5fe2d69607b0bd75c656d1211f969cadba035030156745ee09e7d71740e58ecf", size = 15157, upload-time = "2025-07-16T04:29:24.929Z" },
+    { url = "https://files.pythonhosted.org/packages/04/93/2fa34714b7a4ae72f2f8dad66ba17dd9a2c793220719e736dda28b7aec27/pytest_asyncio-1.2.0-py3-none-any.whl", hash = "sha256:8e17ae5e46d8e7efe51ab6494dd2010f4ca8dae51652aa3c8d55acf50bfb2e99", size = 15095, upload-time = "2025-09-12T07:33:52.639Z" },
 ]
 
 [[package]]
@@ -3413,28 +3978,28 @@ wheels = [
 
 [[package]]
 name = "pytest-cov"
-version = "6.2.1"
+version = "7.0.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "coverage", extra = ["toml"] },
     { name = "pluggy" },
     { name = "pytest" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/18/99/668cade231f434aaa59bbfbf49469068d2ddd945000621d3d165d2e7dd7b/pytest_cov-6.2.1.tar.gz", hash = "sha256:25cc6cc0a5358204b8108ecedc51a9b57b34cc6b8c967cc2c01a4e00d8a67da2", size = 69432, upload-time = "2025-06-12T10:47:47.684Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/5e/f7/c933acc76f5208b3b00089573cf6a2bc26dc80a8aece8f52bb7d6b1855ca/pytest_cov-7.0.0.tar.gz", hash = "sha256:33c97eda2e049a0c5298e91f519302a1334c26ac65c1a483d6206fd458361af1", size = 54328, upload-time = "2025-09-09T10:57:02.113Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/bc/16/4ea354101abb1287856baa4af2732be351c7bee728065aed451b678153fd/pytest_cov-6.2.1-py3-none-any.whl", hash = "sha256:f5bc4c23f42f1cdd23c70b1dab1bbaef4fc505ba950d53e0081d0730dd7e86d5", size = 24644, upload-time = "2025-06-12T10:47:45.932Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/49/1377b49de7d0c1ce41292161ea0f721913fa8722c19fb9c1e3aa0367eecb/pytest_cov-7.0.0-py3-none-any.whl", hash = "sha256:3b8e9558b16cc1479da72058bdecf8073661c7f57f7d3c5f22a1c23507f2d861", size = 22424, upload-time = "2025-09-09T10:57:00.695Z" },
 ]
 
 [[package]]
 name = "pytest-mock"
-version = "3.14.1"
+version = "3.15.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "pytest" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/71/28/67172c96ba684058a4d24ffe144d64783d2a270d0af0d9e792737bddc75c/pytest_mock-3.14.1.tar.gz", hash = "sha256:159e9edac4c451ce77a5cdb9fc5d1100708d2dd4ba3c3df572f14097351af80e", size = 33241, upload-time = "2025-05-26T13:58:45.167Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/68/14/eb014d26be205d38ad5ad20d9a80f7d201472e08167f0bb4361e251084a9/pytest_mock-3.15.1.tar.gz", hash = "sha256:1849a238f6f396da19762269de72cb1814ab44416fa73a8686deac10b0d87a0f", size = 34036, upload-time = "2025-09-16T16:37:27.081Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b2/05/77b60e520511c53d1c1ca75f1930c7dd8e971d0c4379b7f4b3f9644685ba/pytest_mock-3.14.1-py3-none-any.whl", hash = "sha256:178aefcd11307d874b4cd3100344e7e2d888d9791a6a1d9bfe90fbc1b74fd1d0", size = 9923, upload-time = "2025-05-26T13:58:43.487Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/cc/06253936f4a7fa2e0f48dfe6d851d9c56df896a9ab09ac019d70b760619c/pytest_mock-3.15.1-py3-none-any.whl", hash = "sha256:0a25e2eb88fe5168d535041d09a4529a188176ae608a6d249ee65abc0949630d", size = 10095, upload-time = "2025-09-16T16:37:25.734Z" },
 ]
 
 [[package]]
@@ -3458,6 +4023,27 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/5f/ed/539768cf28c661b5b068d66d96a2f155c4971a5d55684a514c1a0e0dec2f/python_dotenv-1.1.1-py3-none-any.whl", hash = "sha256:31f23644fe2602f88ff55e1f5c79ba497e01224ee7737937930c448e4d0e24dc", size = 20556, upload-time = "2025-06-24T04:21:06.073Z" },
 ]
 
+[[package]]
+name = "python-json-logger"
+version = "4.0.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/29/bf/eca6a3d43db1dae7070f70e160ab20b807627ba953663ba07928cdd3dc58/python_json_logger-4.0.0.tar.gz", hash = "sha256:f58e68eb46e1faed27e0f574a55a0455eecd7b8a5b88b85a784519ba3cff047f", size = 17683, upload-time = "2025-10-06T04:15:18.984Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/51/e5/fecf13f06e5e5f67e8837d777d1bc43fac0ed2b77a676804df5c34744727/python_json_logger-4.0.0-py3-none-any.whl", hash = "sha256:af09c9daf6a813aa4cc7180395f50f2a9e5fa056034c9953aec92e381c5ba1e2", size = 15548, upload-time = "2025-10-06T04:15:17.553Z" },
+]
+
+[[package]]
+name = "python-multipart"
+version = "0.0.20"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/87/f44d7c9f274c7ee665a29b885ec97089ec5dc034c7f3fafa03da9e39a09e/python_multipart-0.0.20.tar.gz", hash = "sha256:8dd0cab45b8e23064ae09147625994d090fa46f5b0d1e13af944c331a7fa9d13", size = 37158, upload-time = "2024-12-16T19:45:46.972Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/45/58/38b5afbc1a800eeea951b9285d3912613f2603bdf897a4ab0f4bd7f405fc/python_multipart-0.0.20-py3-none-any.whl", hash = "sha256:8a62d3a8335e06589fe01f2a3e178cdcc632f3fbe0d492ad9ee0ec35aab1f104", size = 24546, upload-time = "2024-12-16T19:45:44.423Z" },
+]
+
 [[package]]
 name = "pytz"
 version = "2025.2"
@@ -3469,46 +4055,47 @@ wheels = [
 
 [[package]]
 name = "pyyaml"
-version = "6.0.2"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/54/ed/79a089b6be93607fa5cdaedf301d7dfb23af5f25c398d5ead2525b063e17/pyyaml-6.0.2.tar.gz", hash = "sha256:d584d9ec91ad65861cc08d42e834324ef890a082e591037abe114850ff7bbc3e", size = 130631, upload-time = "2024-08-06T20:33:50.674Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/9b/95/a3fac87cb7158e231b5a6012e438c647e1a87f09f8e0d123acec8ab8bf71/PyYAML-6.0.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0a9a2848a5b7feac301353437eb7d5957887edbf81d56e903999a75a3d743086", size = 184199, upload-time = "2024-08-06T20:31:40.178Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/7a/68bd47624dab8fd4afbfd3c48e3b79efe09098ae941de5b58abcbadff5cb/PyYAML-6.0.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:29717114e51c84ddfba879543fb232a6ed60086602313ca38cce623c1d62cfbf", size = 171758, upload-time = "2024-08-06T20:31:42.173Z" },
-    { url = "https://files.pythonhosted.org/packages/49/ee/14c54df452143b9ee9f0f29074d7ca5516a36edb0b4cc40c3f280131656f/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8824b5a04a04a047e72eea5cec3bc266db09e35de6bdfe34c9436ac5ee27d237", size = 718463, upload-time = "2024-08-06T20:31:44.263Z" },
-    { url = "https://files.pythonhosted.org/packages/4d/61/de363a97476e766574650d742205be468921a7b532aa2499fcd886b62530/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7c36280e6fb8385e520936c3cb3b8042851904eba0e58d277dca80a5cfed590b", size = 719280, upload-time = "2024-08-06T20:31:50.199Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/4e/1523cb902fd98355e2e9ea5e5eb237cbc5f3ad5f3075fa65087aa0ecb669/PyYAML-6.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ec031d5d2feb36d1d1a24380e4db6d43695f3748343d99434e6f5f9156aaa2ed", size = 751239, upload-time = "2024-08-06T20:31:52.292Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/33/5504b3a9a4464893c32f118a9cc045190a91637b119a9c881da1cf6b7a72/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:936d68689298c36b53b29f23c6dbb74de12b4ac12ca6cfe0e047bedceea56180", size = 695802, upload-time = "2024-08-06T20:31:53.836Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/20/8347dcabd41ef3a3cdc4f7b7a2aff3d06598c8779faa189cdbf878b626a4/PyYAML-6.0.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:23502f431948090f597378482b4812b0caae32c22213aecf3b55325e049a6c68", size = 720527, upload-time = "2024-08-06T20:31:55.565Z" },
-    { url = "https://files.pythonhosted.org/packages/be/aa/5afe99233fb360d0ff37377145a949ae258aaab831bde4792b32650a4378/PyYAML-6.0.2-cp310-cp310-win32.whl", hash = "sha256:2e99c6826ffa974fe6e27cdb5ed0021786b03fc98e5ee3c5bfe1fd5015f42b99", size = 144052, upload-time = "2024-08-06T20:31:56.914Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/84/0fa4b06f6d6c958d207620fc60005e241ecedceee58931bb20138e1e5776/PyYAML-6.0.2-cp310-cp310-win_amd64.whl", hash = "sha256:a4d3091415f010369ae4ed1fc6b79def9416358877534caf6a0fdd2146c87a3e", size = 161774, upload-time = "2024-08-06T20:31:58.304Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/aa/7af4e81f7acba21a4c6be026da38fd2b872ca46226673c89a758ebdc4fd2/PyYAML-6.0.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:cc1c1159b3d456576af7a3e4d1ba7e6924cb39de8f67111c735f6fc832082774", size = 184612, upload-time = "2024-08-06T20:32:03.408Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/62/b9faa998fd185f65c1371643678e4d58254add437edb764a08c5a98fb986/PyYAML-6.0.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1e2120ef853f59c7419231f3bf4e7021f1b936f6ebd222406c3b60212205d2ee", size = 172040, upload-time = "2024-08-06T20:32:04.926Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/0c/c804f5f922a9a6563bab712d8dcc70251e8af811fce4524d57c2c0fd49a4/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5d225db5a45f21e78dd9358e58a98702a0302f2659a3c6cd320564b75b86f47c", size = 736829, upload-time = "2024-08-06T20:32:06.459Z" },
-    { url = "https://files.pythonhosted.org/packages/51/16/6af8d6a6b210c8e54f1406a6b9481febf9c64a3109c541567e35a49aa2e7/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5ac9328ec4831237bec75defaf839f7d4564be1e6b25ac710bd1a96321cc8317", size = 764167, upload-time = "2024-08-06T20:32:08.338Z" },
-    { url = "https://files.pythonhosted.org/packages/75/e4/2c27590dfc9992f73aabbeb9241ae20220bd9452df27483b6e56d3975cc5/PyYAML-6.0.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3ad2a3decf9aaba3d29c8f537ac4b243e36bef957511b4766cb0057d32b0be85", size = 762952, upload-time = "2024-08-06T20:32:14.124Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/97/ecc1abf4a823f5ac61941a9c00fe501b02ac3ab0e373c3857f7d4b83e2b6/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:ff3824dc5261f50c9b0dfb3be22b4567a6f938ccce4587b38952d85fd9e9afe4", size = 735301, upload-time = "2024-08-06T20:32:16.17Z" },
-    { url = "https://files.pythonhosted.org/packages/45/73/0f49dacd6e82c9430e46f4a027baa4ca205e8b0a9dce1397f44edc23559d/PyYAML-6.0.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:797b4f722ffa07cc8d62053e4cff1486fa6dc094105d13fea7b1de7d8bf71c9e", size = 756638, upload-time = "2024-08-06T20:32:18.555Z" },
-    { url = "https://files.pythonhosted.org/packages/22/5f/956f0f9fc65223a58fbc14459bf34b4cc48dec52e00535c79b8db361aabd/PyYAML-6.0.2-cp311-cp311-win32.whl", hash = "sha256:11d8f3dd2b9c1207dcaf2ee0bbbfd5991f571186ec9cc78427ba5bd32afae4b5", size = 143850, upload-time = "2024-08-06T20:32:19.889Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/23/8da0bbe2ab9dcdd11f4f4557ccaf95c10b9811b13ecced089d43ce59c3c8/PyYAML-6.0.2-cp311-cp311-win_amd64.whl", hash = "sha256:e10ce637b18caea04431ce14fabcf5c64a1c61ec9c56b071a4b7ca131ca52d44", size = 161980, upload-time = "2024-08-06T20:32:21.273Z" },
-    { url = "https://files.pythonhosted.org/packages/86/0c/c581167fc46d6d6d7ddcfb8c843a4de25bdd27e4466938109ca68492292c/PyYAML-6.0.2-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:c70c95198c015b85feafc136515252a261a84561b7b1d51e3384e0655ddf25ab", size = 183873, upload-time = "2024-08-06T20:32:25.131Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/0c/38374f5bb272c051e2a69281d71cba6fdb983413e6758b84482905e29a5d/PyYAML-6.0.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:ce826d6ef20b1bc864f0a68340c8b3287705cae2f8b4b1d932177dcc76721725", size = 173302, upload-time = "2024-08-06T20:32:26.511Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/93/9916574aa8c00aa06bbac729972eb1071d002b8e158bd0e83a3b9a20a1f7/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1f71ea527786de97d1a0cc0eacd1defc0985dcf6b3f17bb77dcfc8c34bec4dc5", size = 739154, upload-time = "2024-08-06T20:32:28.363Z" },
-    { url = "https://files.pythonhosted.org/packages/95/0f/b8938f1cbd09739c6da569d172531567dbcc9789e0029aa070856f123984/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9b22676e8097e9e22e36d6b7bda33190d0d400f345f23d4065d48f4ca7ae0425", size = 766223, upload-time = "2024-08-06T20:32:30.058Z" },
-    { url = "https://files.pythonhosted.org/packages/b9/2b/614b4752f2e127db5cc206abc23a8c19678e92b23c3db30fc86ab731d3bd/PyYAML-6.0.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:80bab7bfc629882493af4aa31a4cfa43a4c57c83813253626916b8c7ada83476", size = 767542, upload-time = "2024-08-06T20:32:31.881Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/00/dd137d5bcc7efea1836d6264f049359861cf548469d18da90cd8216cf05f/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:0833f8694549e586547b576dcfaba4a6b55b9e96098b36cdc7ebefe667dfed48", size = 731164, upload-time = "2024-08-06T20:32:37.083Z" },
-    { url = "https://files.pythonhosted.org/packages/c9/1f/4f998c900485e5c0ef43838363ba4a9723ac0ad73a9dc42068b12aaba4e4/PyYAML-6.0.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:8b9c7197f7cb2738065c481a0461e50ad02f18c78cd75775628afb4d7137fb3b", size = 756611, upload-time = "2024-08-06T20:32:38.898Z" },
-    { url = "https://files.pythonhosted.org/packages/df/d1/f5a275fdb252768b7a11ec63585bc38d0e87c9e05668a139fea92b80634c/PyYAML-6.0.2-cp312-cp312-win32.whl", hash = "sha256:ef6107725bd54b262d6dedcc2af448a266975032bc85ef0172c5f059da6325b4", size = 140591, upload-time = "2024-08-06T20:32:40.241Z" },
-    { url = "https://files.pythonhosted.org/packages/0c/e8/4f648c598b17c3d06e8753d7d13d57542b30d56e6c2dedf9c331ae56312e/PyYAML-6.0.2-cp312-cp312-win_amd64.whl", hash = "sha256:7e7401d0de89a9a855c839bc697c079a4af81cf878373abd7dc625847d25cbd8", size = 156338, upload-time = "2024-08-06T20:32:41.93Z" },
-    { url = "https://files.pythonhosted.org/packages/65/d8/b7a1db13636d7fb7d4ff431593c510c8b8fca920ade06ca8ef20015493c5/PyYAML-6.0.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:688ba32a1cffef67fd2e9398a2efebaea461578b0923624778664cc1c914db5d", size = 184777, upload-time = "2024-08-06T20:33:25.896Z" },
-    { url = "https://files.pythonhosted.org/packages/0a/02/6ec546cd45143fdf9840b2c6be8d875116a64076218b61d68e12548e5839/PyYAML-6.0.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a8786accb172bd8afb8be14490a16625cbc387036876ab6ba70912730faf8e1f", size = 172318, upload-time = "2024-08-06T20:33:27.212Z" },
-    { url = "https://files.pythonhosted.org/packages/0e/9a/8cc68be846c972bda34f6c2a93abb644fb2476f4dcc924d52175786932c9/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d8e03406cac8513435335dbab54c0d385e4a49e4945d2909a581c83647ca0290", size = 720891, upload-time = "2024-08-06T20:33:28.974Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/6c/6e1b7f40181bc4805e2e07f4abc10a88ce4648e7e95ff1abe4ae4014a9b2/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f753120cb8181e736c57ef7636e83f31b9c0d1722c516f7e86cf15b7aa57ff12", size = 722614, upload-time = "2024-08-06T20:33:34.157Z" },
-    { url = "https://files.pythonhosted.org/packages/3d/32/e7bd8535d22ea2874cef6a81021ba019474ace0d13a4819c2a4bce79bd6a/PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3b1fdb9dc17f5a7677423d508ab4f243a726dea51fa5e70992e59a7411c89d19", size = 737360, upload-time = "2024-08-06T20:33:35.84Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/12/7322c1e30b9be969670b672573d45479edef72c9a0deac3bb2868f5d7469/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:0b69e4ce7a131fe56b7e4d770c67429700908fc0752af059838b1cfb41960e4e", size = 699006, upload-time = "2024-08-06T20:33:37.501Z" },
-    { url = "https://files.pythonhosted.org/packages/82/72/04fcad41ca56491995076630c3ec1e834be241664c0c09a64c9a2589b507/PyYAML-6.0.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a9f8c2e67970f13b16084e04f134610fd1d374bf477b17ec1599185cf611d725", size = 723577, upload-time = "2024-08-06T20:33:39.389Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/5e/46168b1f2757f1fcd442bc3029cd8767d88a98c9c05770d8b420948743bb/PyYAML-6.0.2-cp39-cp39-win32.whl", hash = "sha256:6395c297d42274772abc367baaa79683958044e5d3835486c16da75d2a694631", size = 144593, upload-time = "2024-08-06T20:33:46.63Z" },
-    { url = "https://files.pythonhosted.org/packages/19/87/5124b1c1f2412bb95c59ec481eaf936cd32f0fe2a7b16b97b81c4c017a6a/PyYAML-6.0.2-cp39-cp39-win_amd64.whl", hash = "sha256:39693e1f8320ae4f43943590b49779ffb98acb81f788220ea932a6b6c51004d8", size = 162312, upload-time = "2024-08-06T20:33:49.073Z" },
+version = "6.0.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/05/8e/961c0007c59b8dd7729d542c61a4d537767a59645b82a0b521206e1e25c2/pyyaml-6.0.3.tar.gz", hash = "sha256:d76623373421df22fb4cf8817020cbb7ef15c725b9d5e45f17e189bfc384190f", size = 130960, upload-time = "2025-09-25T21:33:16.546Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f4/a0/39350dd17dd6d6c6507025c0e53aef67a9293a6d37d3511f23ea510d5800/pyyaml-6.0.3-cp310-cp310-macosx_10_13_x86_64.whl", hash = "sha256:214ed4befebe12df36bcc8bc2b64b396ca31be9304b8f59e25c11cf94a4c033b", size = 184227, upload-time = "2025-09-25T21:31:46.04Z" },
+    { url = "https://files.pythonhosted.org/packages/05/14/52d505b5c59ce73244f59c7a50ecf47093ce4765f116cdb98286a71eeca2/pyyaml-6.0.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:02ea2dfa234451bbb8772601d7b8e426c2bfa197136796224e50e35a78777956", size = 174019, upload-time = "2025-09-25T21:31:47.706Z" },
+    { url = "https://files.pythonhosted.org/packages/43/f7/0e6a5ae5599c838c696adb4e6330a59f463265bfa1e116cfd1fbb0abaaae/pyyaml-6.0.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b30236e45cf30d2b8e7b3e85881719e98507abed1011bf463a8fa23e9c3e98a8", size = 740646, upload-time = "2025-09-25T21:31:49.21Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/3a/61b9db1d28f00f8fd0ae760459a5c4bf1b941baf714e207b6eb0657d2578/pyyaml-6.0.3-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:66291b10affd76d76f54fad28e22e51719ef9ba22b29e1d7d03d6777a9174198", size = 840793, upload-time = "2025-09-25T21:31:50.735Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/1e/7acc4f0e74c4b3d9531e24739e0ab832a5edf40e64fbae1a9c01941cabd7/pyyaml-6.0.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9c7708761fccb9397fe64bbc0395abcae8c4bf7b0eac081e12b809bf47700d0b", size = 770293, upload-time = "2025-09-25T21:31:51.828Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/ef/abd085f06853af0cd59fa5f913d61a8eab65d7639ff2a658d18a25d6a89d/pyyaml-6.0.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:418cf3f2111bc80e0933b2cd8cd04f286338bb88bdc7bc8e6dd775ebde60b5e0", size = 732872, upload-time = "2025-09-25T21:31:53.282Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/15/2bc9c8faf6450a8b3c9fc5448ed869c599c0a74ba2669772b1f3a0040180/pyyaml-6.0.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:5e0b74767e5f8c593e8c9b5912019159ed0533c70051e9cce3e8b6aa699fcd69", size = 758828, upload-time = "2025-09-25T21:31:54.807Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/00/531e92e88c00f4333ce359e50c19b8d1de9fe8d581b1534e35ccfbc5f393/pyyaml-6.0.3-cp310-cp310-win32.whl", hash = "sha256:28c8d926f98f432f88adc23edf2e6d4921ac26fb084b028c733d01868d19007e", size = 142415, upload-time = "2025-09-25T21:31:55.885Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/fa/926c003379b19fca39dd4634818b00dec6c62d87faf628d1394e137354d4/pyyaml-6.0.3-cp310-cp310-win_amd64.whl", hash = "sha256:bdb2c67c6c1390b63c6ff89f210c8fd09d9a1217a465701eac7316313c915e4c", size = 158561, upload-time = "2025-09-25T21:31:57.406Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/16/a95b6757765b7b031c9374925bb718d55e0a9ba8a1b6a12d25962ea44347/pyyaml-6.0.3-cp311-cp311-macosx_10_13_x86_64.whl", hash = "sha256:44edc647873928551a01e7a563d7452ccdebee747728c1080d881d68af7b997e", size = 185826, upload-time = "2025-09-25T21:31:58.655Z" },
+    { url = "https://files.pythonhosted.org/packages/16/19/13de8e4377ed53079ee996e1ab0a9c33ec2faf808a4647b7b4c0d46dd239/pyyaml-6.0.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:652cb6edd41e718550aad172851962662ff2681490a8a711af6a4d288dd96824", size = 175577, upload-time = "2025-09-25T21:32:00.088Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/62/d2eb46264d4b157dae1275b573017abec435397aa59cbcdab6fc978a8af4/pyyaml-6.0.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:10892704fc220243f5305762e276552a0395f7beb4dbf9b14ec8fd43b57f126c", size = 775556, upload-time = "2025-09-25T21:32:01.31Z" },
+    { url = "https://files.pythonhosted.org/packages/10/cb/16c3f2cf3266edd25aaa00d6c4350381c8b012ed6f5276675b9eba8d9ff4/pyyaml-6.0.3-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:850774a7879607d3a6f50d36d04f00ee69e7fc816450e5f7e58d7f17f1ae5c00", size = 882114, upload-time = "2025-09-25T21:32:03.376Z" },
+    { url = "https://files.pythonhosted.org/packages/71/60/917329f640924b18ff085ab889a11c763e0b573da888e8404ff486657602/pyyaml-6.0.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b8bb0864c5a28024fac8a632c443c87c5aa6f215c0b126c449ae1a150412f31d", size = 806638, upload-time = "2025-09-25T21:32:04.553Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/6f/529b0f316a9fd167281a6c3826b5583e6192dba792dd55e3203d3f8e655a/pyyaml-6.0.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:1d37d57ad971609cf3c53ba6a7e365e40660e3be0e5175fa9f2365a379d6095a", size = 767463, upload-time = "2025-09-25T21:32:06.152Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/6a/b627b4e0c1dd03718543519ffb2f1deea4a1e6d42fbab8021936a4d22589/pyyaml-6.0.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:37503bfbfc9d2c40b344d06b2199cf0e96e97957ab1c1b546fd4f87e53e5d3e4", size = 794986, upload-time = "2025-09-25T21:32:07.367Z" },
+    { url = "https://files.pythonhosted.org/packages/45/91/47a6e1c42d9ee337c4839208f30d9f09caa9f720ec7582917b264defc875/pyyaml-6.0.3-cp311-cp311-win32.whl", hash = "sha256:8098f252adfa6c80ab48096053f512f2321f0b998f98150cea9bd23d83e1467b", size = 142543, upload-time = "2025-09-25T21:32:08.95Z" },
+    { url = "https://files.pythonhosted.org/packages/da/e3/ea007450a105ae919a72393cb06f122f288ef60bba2dc64b26e2646fa315/pyyaml-6.0.3-cp311-cp311-win_amd64.whl", hash = "sha256:9f3bfb4965eb874431221a3ff3fdcddc7e74e3b07799e0e84ca4a0f867d449bf", size = 158763, upload-time = "2025-09-25T21:32:09.96Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/33/422b98d2195232ca1826284a76852ad5a86fe23e31b009c9886b2d0fb8b2/pyyaml-6.0.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:7f047e29dcae44602496db43be01ad42fc6f1cc0d8cd6c83d342306c32270196", size = 182063, upload-time = "2025-09-25T21:32:11.445Z" },
+    { url = "https://files.pythonhosted.org/packages/89/a0/6cf41a19a1f2f3feab0e9c0b74134aa2ce6849093d5517a0c550fe37a648/pyyaml-6.0.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:fc09d0aa354569bc501d4e787133afc08552722d3ab34836a80547331bb5d4a0", size = 173973, upload-time = "2025-09-25T21:32:12.492Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/23/7a778b6bd0b9a8039df8b1b1d80e2e2ad78aa04171592c8a5c43a56a6af4/pyyaml-6.0.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9149cad251584d5fb4981be1ecde53a1ca46c891a79788c0df828d2f166bda28", size = 775116, upload-time = "2025-09-25T21:32:13.652Z" },
+    { url = "https://files.pythonhosted.org/packages/65/30/d7353c338e12baef4ecc1b09e877c1970bd3382789c159b4f89d6a70dc09/pyyaml-6.0.3-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:5fdec68f91a0c6739b380c83b951e2c72ac0197ace422360e6d5a959d8d97b2c", size = 844011, upload-time = "2025-09-25T21:32:15.21Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/9d/b3589d3877982d4f2329302ef98a8026e7f4443c765c46cfecc8858c6b4b/pyyaml-6.0.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ba1cc08a7ccde2d2ec775841541641e4548226580ab850948cbfda66a1befcdc", size = 807870, upload-time = "2025-09-25T21:32:16.431Z" },
+    { url = "https://files.pythonhosted.org/packages/05/c0/b3be26a015601b822b97d9149ff8cb5ead58c66f981e04fedf4e762f4bd4/pyyaml-6.0.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:8dc52c23056b9ddd46818a57b78404882310fb473d63f17b07d5c40421e47f8e", size = 761089, upload-time = "2025-09-25T21:32:17.56Z" },
+    { url = "https://files.pythonhosted.org/packages/be/8e/98435a21d1d4b46590d5459a22d88128103f8da4c2d4cb8f14f2a96504e1/pyyaml-6.0.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:41715c910c881bc081f1e8872880d3c650acf13dfa8214bad49ed4cede7c34ea", size = 790181, upload-time = "2025-09-25T21:32:18.834Z" },
+    { url = "https://files.pythonhosted.org/packages/74/93/7baea19427dcfbe1e5a372d81473250b379f04b1bd3c4c5ff825e2327202/pyyaml-6.0.3-cp312-cp312-win32.whl", hash = "sha256:96b533f0e99f6579b3d4d4995707cf36df9100d67e0c8303a0c55b27b5f99bc5", size = 137658, upload-time = "2025-09-25T21:32:20.209Z" },
+    { url = "https://files.pythonhosted.org/packages/86/bf/899e81e4cce32febab4fb42bb97dcdf66bc135272882d1987881a4b519e9/pyyaml-6.0.3-cp312-cp312-win_amd64.whl", hash = "sha256:5fcd34e47f6e0b794d17de1b4ff496c00986e1c83f7ab2fb8fcfe9616ff7477b", size = 154003, upload-time = "2025-09-25T21:32:21.167Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/08/67bd04656199bbb51dbed1439b7f27601dfb576fb864099c7ef0c3e55531/pyyaml-6.0.3-cp312-cp312-win_arm64.whl", hash = "sha256:64386e5e707d03a7e172c0701abfb7e10f0fb753ee1d773128192742712a98fd", size = 140344, upload-time = "2025-09-25T21:32:22.617Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/62/67fc8e68a75f738c9200422bf65693fb79a4cd0dc5b23310e5202e978090/pyyaml-6.0.3-cp39-cp39-macosx_10_13_x86_64.whl", hash = "sha256:b865addae83924361678b652338317d1bd7e79b1f4596f96b96c77a5a34b34da", size = 184450, upload-time = "2025-09-25T21:33:00.618Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/92/861f152ce87c452b11b9d0977952259aa7df792d71c1053365cc7b09cc08/pyyaml-6.0.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:c3355370a2c156cffb25e876646f149d5d68f5e0a3ce86a5084dd0b64a994917", size = 174319, upload-time = "2025-09-25T21:33:02.086Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/cd/f0cfc8c74f8a030017a2b9c771b7f47e5dd702c3e28e5b2071374bda2948/pyyaml-6.0.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3c5677e12444c15717b902a5798264fa7909e41153cdf9ef7ad571b704a63dd9", size = 737631, upload-time = "2025-09-25T21:33:03.25Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/b2/18f2bd28cd2055a79a46c9b0895c0b3d987ce40ee471cecf58a1a0199805/pyyaml-6.0.3-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:5ed875a24292240029e4483f9d4a4b8a1ae08843b9c54f43fcc11e404532a8a5", size = 836795, upload-time = "2025-09-25T21:33:05.014Z" },
+    { url = "https://files.pythonhosted.org/packages/73/b9/793686b2d54b531203c160ef12bec60228a0109c79bae6c1277961026770/pyyaml-6.0.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0150219816b6a1fa26fb4699fb7daa9caf09eb1999f3b70fb6e786805e80375a", size = 750767, upload-time = "2025-09-25T21:33:06.398Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/86/a137b39a611def2ed78b0e66ce2fe13ee701a07c07aebe55c340ed2a050e/pyyaml-6.0.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:fa160448684b4e94d80416c0fa4aac48967a969efe22931448d853ada8baf926", size = 727982, upload-time = "2025-09-25T21:33:08.708Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/62/71c27c94f457cf4418ef8ccc71735324c549f7e3ea9d34aba50874563561/pyyaml-6.0.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:27c0abcb4a5dac13684a37f76e701e054692a9b2d3064b70f5e4eb54810553d7", size = 755677, upload-time = "2025-09-25T21:33:09.876Z" },
+    { url = "https://files.pythonhosted.org/packages/29/3d/6f5e0d58bd924fb0d06c3a6bad00effbdae2de5adb5cda5648006ffbd8d3/pyyaml-6.0.3-cp39-cp39-win32.whl", hash = "sha256:1ebe39cb5fc479422b83de611d14e2c0d3bb2a18bbcb01f229ab3cfbd8fee7a0", size = 142592, upload-time = "2025-09-25T21:33:10.983Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/0c/25113e0b5e103d7f1490c0e947e303fe4a696c10b501dea7a9f49d4e876c/pyyaml-6.0.3-cp39-cp39-win_amd64.whl", hash = "sha256:2e71d11abed7344e42a8849600193d15b6def118602c4c176f748e4583246007", size = 158777, upload-time = "2025-09-25T21:33:15.55Z" },
 ]
 
 [[package]]
@@ -3523,13 +4110,57 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/04/11/432f32f8097b03e3cd5fe57e88efb685d964e2e5178a48ed61e841f7fdce/pyyaml_env_tag-1.1-py3-none-any.whl", hash = "sha256:17109e1a528561e32f026364712fee1264bc2ea6715120891174ed1b980d2e04", size = 4722, upload-time = "2025-05-13T15:23:59.629Z" },
 ]
 
+[[package]]
+name = "pyzmq"
+version = "27.1.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "cffi", marker = "implementation_name == 'pypy'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/04/0b/3c9baedbdf613ecaa7aa07027780b8867f57b6293b6ee50de316c9f3222b/pyzmq-27.1.0.tar.gz", hash = "sha256:ac0765e3d44455adb6ddbf4417dcce460fc40a05978c08efdf2948072f6db540", size = 281750, upload-time = "2025-09-08T23:10:18.157Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/99/64/5653e7b7425b169f994835a2b2abf9486264401fdef18df91ddae47ce2cc/pyzmq-27.1.0-cp310-cp310-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:507b6f430bdcf0ee48c0d30e734ea89ce5567fd7b8a0f0044a369c176aa44556", size = 906380, upload-time = "2025-09-08T23:07:29.78Z" },
+    { url = "https://files.pythonhosted.org/packages/73/78/7d713284dbe022f6440e391bd1f3c48d9185673878034cfb3939cdf333b2/pyzmq-27.1.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bf7b38f9fd7b81cb6d9391b2946382c8237fd814075c6aa9c3b746d53076023b", size = 666421, upload-time = "2025-09-08T23:07:31.263Z" },
+    { url = "https://files.pythonhosted.org/packages/30/76/8f099f9d6482450428b17c4d6b241281af7ce6a9de8149ca8c1c649f6792/pyzmq-27.1.0-cp310-cp310-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:03ff0b279b40d687691a6217c12242ee71f0fba28bf8626ff50e3ef0f4410e1e", size = 854149, upload-time = "2025-09-08T23:07:33.17Z" },
+    { url = "https://files.pythonhosted.org/packages/59/f0/37fbfff06c68016019043897e4c969ceab18bde46cd2aca89821fcf4fb2e/pyzmq-27.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:677e744fee605753eac48198b15a2124016c009a11056f93807000ab11ce6526", size = 1655070, upload-time = "2025-09-08T23:07:35.205Z" },
+    { url = "https://files.pythonhosted.org/packages/47/14/7254be73f7a8edc3587609554fcaa7bfd30649bf89cd260e4487ca70fdaa/pyzmq-27.1.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:dd2fec2b13137416a1c5648b7009499bcc8fea78154cd888855fa32514f3dad1", size = 2033441, upload-time = "2025-09-08T23:07:37.432Z" },
+    { url = "https://files.pythonhosted.org/packages/22/dc/49f2be26c6f86f347e796a4d99b19167fc94503f0af3fd010ad262158822/pyzmq-27.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:08e90bb4b57603b84eab1d0ca05b3bbb10f60c1839dc471fc1c9e1507bef3386", size = 1891529, upload-time = "2025-09-08T23:07:39.047Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/a0/fc7e78a23748ad5443ac3275943457e8452da67fda347e05260261108cbc/pyzmq-27.1.0-cp311-cp311-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:0790a0161c281ca9723f804871b4027f2e8b5a528d357c8952d08cd1a9c15581", size = 908803, upload-time = "2025-09-08T23:07:47.551Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/22/37d15eb05f3bdfa4abea6f6d96eb3bb58585fbd3e4e0ded4e743bc650c97/pyzmq-27.1.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c895a6f35476b0c3a54e3eb6ccf41bf3018de937016e6e18748317f25d4e925f", size = 668836, upload-time = "2025-09-08T23:07:49.436Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/c4/2a6fe5111a01005fc7af3878259ce17684fabb8852815eda6225620f3c59/pyzmq-27.1.0-cp311-cp311-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5bbf8d3630bf96550b3be8e1fc0fea5cbdc8d5466c1192887bd94869da17a63e", size = 857038, upload-time = "2025-09-08T23:07:51.234Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/eb/bfdcb41d0db9cd233d6fb22dc131583774135505ada800ebf14dfb0a7c40/pyzmq-27.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:15c8bd0fe0dabf808e2d7a681398c4e5ded70a551ab47482067a572c054c8e2e", size = 1657531, upload-time = "2025-09-08T23:07:52.795Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/21/e3180ca269ed4a0de5c34417dfe71a8ae80421198be83ee619a8a485b0c7/pyzmq-27.1.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:bafcb3dd171b4ae9f19ee6380dfc71ce0390fefaf26b504c0e5f628d7c8c54f2", size = 2034786, upload-time = "2025-09-08T23:07:55.047Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/b1/5e21d0b517434b7f33588ff76c177c5a167858cc38ef740608898cd329f2/pyzmq-27.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:e829529fcaa09937189178115c49c504e69289abd39967cd8a4c215761373394", size = 1894220, upload-time = "2025-09-08T23:07:57.172Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/5e/c3c49fdd0f535ef45eefcc16934648e9e59dace4a37ee88fc53f6cd8e641/pyzmq-27.1.0-cp312-abi3-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:1c179799b118e554b66da67d88ed66cd37a169f1f23b5d9f0a231b4e8d44a113", size = 895645, upload-time = "2025-09-08T23:08:05.301Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/e5/b0b2504cb4e903a74dcf1ebae157f9e20ebb6ea76095f6cfffea28c42ecd/pyzmq-27.1.0-cp312-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3837439b7f99e60312f0c926a6ad437b067356dc2bc2ec96eb395fd0fe804233", size = 652574, upload-time = "2025-09-08T23:08:06.828Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/9b/c108cdb55560eaf253f0cbdb61b29971e9fb34d9c3499b0e96e4e60ed8a5/pyzmq-27.1.0-cp312-abi3-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:43ad9a73e3da1fab5b0e7e13402f0b2fb934ae1c876c51d0afff0e7c052eca31", size = 840995, upload-time = "2025-09-08T23:08:08.396Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/bb/b79798ca177b9eb0825b4c9998c6af8cd2a7f15a6a1a4272c1d1a21d382f/pyzmq-27.1.0-cp312-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:0de3028d69d4cdc475bfe47a6128eb38d8bc0e8f4d69646adfbcd840facbac28", size = 1642070, upload-time = "2025-09-08T23:08:09.989Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/80/2df2e7977c4ede24c79ae39dcef3899bfc5f34d1ca7a5b24f182c9b7a9ca/pyzmq-27.1.0-cp312-abi3-musllinux_1_2_i686.whl", hash = "sha256:cf44a7763aea9298c0aa7dbf859f87ed7012de8bda0f3977b6fb1d96745df856", size = 2021121, upload-time = "2025-09-08T23:08:11.907Z" },
+    { url = "https://files.pythonhosted.org/packages/46/bd/2d45ad24f5f5ae7e8d01525eb76786fa7557136555cac7d929880519e33a/pyzmq-27.1.0-cp312-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:f30f395a9e6fbca195400ce833c731e7b64c3919aa481af4d88c3759e0cb7496", size = 1878550, upload-time = "2025-09-08T23:08:13.513Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/ca/2b8693d06b1db4e0c084871e4c9d7842b561d0a6ff9d780640f5e3e9eb55/pyzmq-27.1.0-cp39-cp39-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:49d3980544447f6bd2968b6ac913ab963a49dcaa2d4a2990041f16057b04c429", size = 906559, upload-time = "2025-09-08T23:09:22.983Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/b3/b99b39e2cfdcebd512959780e4d299447fd7f46010b1d88d63324e2481ec/pyzmq-27.1.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:849ca054d81aa1c175c49484afaaa5db0622092b5eccb2055f9f3bb8f703782d", size = 863816, upload-time = "2025-09-08T23:09:24.556Z" },
+    { url = "https://files.pythonhosted.org/packages/61/b2/018fa8e8eefb34a625b1a45e2effcbc9885645b22cdd0a68283f758351e7/pyzmq-27.1.0-cp39-cp39-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3970778e74cb7f85934d2b926b9900e92bfe597e62267d7499acc39c9c28e345", size = 666735, upload-time = "2025-09-08T23:09:26.297Z" },
+    { url = "https://files.pythonhosted.org/packages/01/05/8ae778f7cd7c94030731ae2305e6a38f3a333b6825f56c0c03f2134ccf1b/pyzmq-27.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:da96ecdcf7d3919c3be2de91a8c513c186f6762aa6cf7c01087ed74fad7f0968", size = 1655425, upload-time = "2025-09-08T23:09:28.172Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/ad/d69478a97a3f3142f9dbbbd9daa4fcf42541913a85567c36d4cfc19b2218/pyzmq-27.1.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:9541c444cfe1b1c0156c5c86ece2bb926c7079a18e7b47b0b1b3b1b875e5d098", size = 2033729, upload-time = "2025-09-08T23:09:30.097Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/6d/e3c6ad05bc1cddd25094e66cc15ae8924e15c67e231e93ed2955c401007e/pyzmq-27.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:e30a74a39b93e2e1591b58eb1acef4902be27c957a8720b0e368f579b82dc22f", size = 1891803, upload-time = "2025-09-08T23:09:31.875Z" },
+    { url = "https://files.pythonhosted.org/packages/58/ed/0202ca350f4f2b69faa95c6d931e3c05c3a397c184cacb84cb4f8f42f287/pyzmq-27.1.0-pp310-pypy310_pp73-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:f328d01128373cb6763823b2b4e7f73bdf767834268c565151eacb3b7a392f90", size = 800206, upload-time = "2025-09-08T23:09:41.902Z" },
+    { url = "https://files.pythonhosted.org/packages/47/42/1ff831fa87fe8f0a840ddb399054ca0009605d820e2b44ea43114f5459f4/pyzmq-27.1.0-pp310-pypy310_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9c1790386614232e1b3a40a958454bdd42c6d1811837b15ddbb052a032a43f62", size = 567747, upload-time = "2025-09-08T23:09:43.741Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/db/5c4d6807434751e3f21231bee98109aa57b9b9b55e058e450d0aef59b70f/pyzmq-27.1.0-pp310-pypy310_pp73-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:448f9cb54eb0cee4732b46584f2710c8bc178b0e5371d9e4fc8125201e413a74", size = 747371, upload-time = "2025-09-08T23:09:45.575Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/79/f38c92eeaeb03a2ccc2ba9866f0439593bb08c5e3b714ac1d553e5c96e25/pyzmq-27.1.0-pp311-pypy311_pp73-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:ac25465d42f92e990f8d8b0546b01c391ad431c3bf447683fdc40565941d0604", size = 800208, upload-time = "2025-09-08T23:09:51.073Z" },
+    { url = "https://files.pythonhosted.org/packages/49/0e/3f0d0d335c6b3abb9b7b723776d0b21fa7f3a6c819a0db6097059aada160/pyzmq-27.1.0-pp311-pypy311_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:53b40f8ae006f2734ee7608d59ed661419f087521edbfc2149c3932e9c14808c", size = 567747, upload-time = "2025-09-08T23:09:52.698Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/cf/f2b3784d536250ffd4be70e049f3b60981235d70c6e8ce7e3ef21e1adb25/pyzmq-27.1.0-pp311-pypy311_pp73-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f605d884e7c8be8fe1aa94e0a783bf3f591b84c24e4bc4f3e7564c82ac25e271", size = 747371, upload-time = "2025-09-08T23:09:54.563Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/5f/4e10c7f57a4c92ab0fbb2396297aa8d618e6f5b9b8f8e9756d56f3e6fc52/pyzmq-27.1.0-pp39-pypy39_pp73-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:01f9437501886d3a1dd4b02ef59fb8cc384fa718ce066d52f175ee49dd5b7ed8", size = 800203, upload-time = "2025-09-08T23:10:09.436Z" },
+    { url = "https://files.pythonhosted.org/packages/19/72/a74a007cd636f903448c6ab66628104b1fc5f2ba018733d5eabb94a0a6fb/pyzmq-27.1.0-pp39-pypy39_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4a19387a3dddcc762bfd2f570d14e2395b2c9701329b266f83dd87a2b3cbd381", size = 758756, upload-time = "2025-09-08T23:10:11.733Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/d4/30c25b91f2b4786026372f5ef454134d7f576fcf4ac58539ad7dd5de4762/pyzmq-27.1.0-pp39-pypy39_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4c618fbcd069e3a29dcd221739cacde52edcc681f041907867e0f5cc7e85f172", size = 567742, upload-time = "2025-09-08T23:10:14.732Z" },
+]
+
 [[package]]
 name = "ray"
-version = "2.49.0"
+version = "2.49.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "filelock" },
     { name = "jsonschema" },
     { name = "msgpack" },
@@ -3539,14 +4170,19 @@ dependencies = [
     { name = "requests" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ca/2d/67ac2d5700dbe51c5a3162ec8937299ba100ea016b540997d52776de08e6/ray-2.49.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:8c55a7a462a0cdeb1ed58fa3610f52e3a284ce4e0c55566359bd1158a9683f44", size = 69124127, upload-time = "2025-08-26T19:07:26.774Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/56/b954e0a81a94fad760e57185bc18f0955a3574ad4bf63ea2756e8d588462/ray-2.49.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:6c540a6dfa8e582e0725057c137245ccd3fa0d697b47648533176f5a8ab30a30", size = 69933848, upload-time = "2025-08-26T19:07:32.234Z" },
-    { url = "https://files.pythonhosted.org/packages/0e/60/36b5773a14dda16c2ff997cbc6b3ee2700dca42550c026c7108c7bc7a2b4/ray-2.49.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:2a75d2fd8bcab1a85091ea30e5932c5e2897c7122ae23f921f4fafeec22c3828", size = 69264900, upload-time = "2025-08-26T19:07:52.476Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/47/5bc38d9d86b6fcea65dda30ea6d128c7e18bd45687fca9c5038ac8276d1a/ray-2.49.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:25ae4c208dc78b4a653c311cb26b93c74401e6762a193ced5fabaa910a817d81", size = 70068495, upload-time = "2025-08-26T19:07:57.491Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/81/d14e0d58358ca145f173b7b0604434f368cd90243172955d6cbf4104ce59/ray-2.49.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:363567f56112f3a1518ecd8d0e3a0ce4f053705cc8f4d43e933caf69fb35971c", size = 69285849, upload-time = "2025-08-26T19:08:23.849Z" },
-    { url = "https://files.pythonhosted.org/packages/de/8b/1de4d89347857d78f3ffd10f1b1eef767bfc6ec22bf7721ff61719a6b56e/ray-2.49.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:95285d23a303b174d070a6b3ea8723e9dabf07a4147a01cb9331d5eadf8ca0b2", size = 70113049, upload-time = "2025-08-26T19:08:29.453Z" },
-    { url = "https://files.pythonhosted.org/packages/07/89/857a879351dcc8b299084eba05e5d2d0e47c973b399ef769c80f45c62503/ray-2.49.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:d3a3f04d8975db0c1e9e59ff4169cae4e6f1026ad27dafb9fb9baf0121c32427", size = 69124048, upload-time = "2025-08-26T19:09:19.197Z" },
-    { url = "https://files.pythonhosted.org/packages/de/cc/9ace3df475404f45199c7bf6587d4760c0add58c7d32002f23d6f81da988/ray-2.49.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:0ff4a0c4fb5f11fe9dcb8266bb98e2d43b436c1e654fdfabd8a22892b4ae48e1", size = 69931493, upload-time = "2025-08-26T19:09:24.642Z" },
+    { url = "https://files.pythonhosted.org/packages/63/0e/830df5a0f7e2b582422ee8ad0cdf2a2a9563aa63bb8e60be9ceec494981c/ray-2.49.2-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:eae07b3fed45f5b041a8bf9795cd26fad2464be5126efd447e4484905a29b677", size = 69125462, upload-time = "2025-09-19T19:14:51.029Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/85/a340eba596db3f66d3a338aff43942d8bac32732fb4cf4a20ed4bbbd07eb/ray-2.49.2-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:74566876af7bf4e48ea4b9b3b75b34db053d1064cc4d4b1670dc4ce78f6894af", size = 69935752, upload-time = "2025-09-19T19:14:56.191Z" },
+    { url = "https://files.pythonhosted.org/packages/38/67/93f0d6d558874a730581059eb6dfa8860991a5410502ea0685dba5e788e4/ray-2.49.2-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:eada9dd89ccda643a3c6c2cba7016b59898432d126e10b38fed52d74165364f4", size = 69266231, upload-time = "2025-09-19T19:15:16.92Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/2b/f2efd0e7bcef06d51422db1af48cc5695a3f9b40a444f9d270a2d4663252/ray-2.49.2-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:54077dde338c5ffba349a4ab61b72352a3c3be69ea5b4f1b436d98d40b312763", size = 70070382, upload-time = "2025-09-19T19:15:22.048Z" },
+    { url = "https://files.pythonhosted.org/packages/da/99/23b732c0b7b2ee2ffd28bf632257fb98924a03251d251810cb637512fcab/ray-2.49.2-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:dd0d8d8641d142fafe6d83e87d3c19bd5637d21e34608d3ff69ad71ea3e2f462", size = 69287193, upload-time = "2025-09-19T19:15:42.093Z" },
+    { url = "https://files.pythonhosted.org/packages/69/ca/94791be5c3b68ed0df85589a8ca558334818a47bf2978000f85533245aed/ray-2.49.2-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:2ecaaa51f588ccdda2b61563a8be3843bf65dfaaa83a240588a307f4ebb82471", size = 70114942, upload-time = "2025-09-19T19:15:47.536Z" },
+    { url = "https://files.pythonhosted.org/packages/25/ca/04ae817999fc805fc68c0819c0fd8fe23c0b830520c74cf84e9768d31b5c/ray-2.49.2-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:df99b7aafbfbe122befebfdca78facb297273260a1026752ff43af613b237939", size = 69125379, upload-time = "2025-09-19T19:16:29.125Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/fd/ac5da30b701cc6647b0c15da5107543932509c327fb541aef6c4b38c11bf/ray-2.49.2-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:057e2e1408790d11d8931c46ed0ca6a0c125bc129811ea1c8a833ee945c36e13", size = 69933379, upload-time = "2025-09-19T19:16:34.774Z" },
+]
+
+[package.optional-dependencies]
+cgraph = [
+    { name = "cupy-cuda12x", marker = "sys_platform != 'darwin'" },
 ]
 
 [[package]]
@@ -3565,68 +4201,68 @@ wheels = [
 
 [[package]]
 name = "regex"
-version = "2025.9.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/b2/5a/4c63457fbcaf19d138d72b2e9b39405954f98c0349b31c601bfcb151582c/regex-2025.9.1.tar.gz", hash = "sha256:88ac07b38d20b54d79e704e38aa3bd2c0f8027432164226bdee201a1c0c9c9ff", size = 400852, upload-time = "2025-09-01T22:10:10.479Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/46/c1/ed9ef923156105a78aa004f9390e5dd87eadc29f5ca8840f172cadb638de/regex-2025.9.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:c5aa2a6a73bf218515484b36a0d20c6ad9dc63f6339ff6224147b0e2c095ee55", size = 484813, upload-time = "2025-09-01T22:07:45.528Z" },
-    { url = "https://files.pythonhosted.org/packages/05/de/97957618a774c67f892609eee2fafe3e30703fbbba66de5e6b79d7196dbc/regex-2025.9.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:8c2ff5c01d5e47ad5fc9d31bcd61e78c2fa0068ed00cab86b7320214446da766", size = 288981, upload-time = "2025-09-01T22:07:48.464Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/b0/441afadd0a6ffccbd58a9663e5bdd182daa237893e5f8ceec6ff9df4418a/regex-2025.9.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d49dc84e796b666181de8a9973284cad6616335f01b52bf099643253094920fc", size = 286608, upload-time = "2025-09-01T22:07:50.484Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/cf/d89aecaf17e999ab11a3ef73fc9ab8b64f4e156f121250ef84340b35338d/regex-2025.9.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d9914fe1040874f83c15fcea86d94ea54091b0666eab330aaab69e30d106aabe", size = 780459, upload-time = "2025-09-01T22:07:52.34Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/05/05884594a9975a29597917bbdd6837f7b97e8ac23faf22d628aa781e58f7/regex-2025.9.1-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e71bceb3947362ec5eabd2ca0870bb78eae4edfc60c6c21495133c01b6cd2df4", size = 849276, upload-time = "2025-09-01T22:07:54.591Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/8d/2b3067506838d02096bf107beb129b2ce328cdf776d6474b7f542c0a7bfd/regex-2025.9.1-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:67a74456f410fe5e869239ee7a5423510fe5121549af133809d9591a8075893f", size = 897320, upload-time = "2025-09-01T22:07:56.129Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/b3/0f9f7766e980b900df0ba9901b52871a2e4203698fb35cdebd219240d5f7/regex-2025.9.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5c3b96ed0223b32dbdc53a83149b6de7ca3acd5acd9c8e64b42a166228abe29c", size = 789931, upload-time = "2025-09-01T22:07:57.834Z" },
-    { url = "https://files.pythonhosted.org/packages/47/9f/7b2f29c8f8b698eb44be5fc68e8b9c8d32e99635eac5defc98de114e9f35/regex-2025.9.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:113d5aa950f428faf46fd77d452df62ebb4cc6531cb619f6cc30a369d326bfbd", size = 780764, upload-time = "2025-09-01T22:07:59.413Z" },
-    { url = "https://files.pythonhosted.org/packages/ac/ac/56176caa86155c14462531eb0a4ddc450d17ba8875001122b3b7c0cb01bf/regex-2025.9.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:fcdeb38de4f7f3d69d798f4f371189061446792a84e7c92b50054c87aae9c07c", size = 773610, upload-time = "2025-09-01T22:08:01.042Z" },
-    { url = "https://files.pythonhosted.org/packages/39/e8/9d6b9bd43998268a9de2f35602077519cacc9cb149f7381758cf8f502ba7/regex-2025.9.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:4bcdff370509164b67a6c8ec23c9fb40797b72a014766fdc159bb809bd74f7d8", size = 844090, upload-time = "2025-09-01T22:08:02.94Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/92/d89743b089005cae4cb81cc2fe177e180b7452e60f29de53af34349640f8/regex-2025.9.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:7383efdf6e8e8c61d85e00cfb2e2e18da1a621b8bfb4b0f1c2747db57b942b8f", size = 834775, upload-time = "2025-09-01T22:08:04.781Z" },
-    { url = "https://files.pythonhosted.org/packages/01/8f/86a3e0aaa89295d2a3445bb238e56369963ef6b02a5b4aa3362f4e687413/regex-2025.9.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:1ec2bd3bdf0f73f7e9f48dca550ba7d973692d5e5e9a90ac42cc5f16c4432d8b", size = 778521, upload-time = "2025-09-01T22:08:06.596Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/df/72072acb370ee8577c255717f8a58264f1d0de40aa3c9e6ebd5271cac633/regex-2025.9.1-cp310-cp310-win32.whl", hash = "sha256:9627e887116c4e9c0986d5c3b4f52bcfe3df09850b704f62ec3cbf177a0ae374", size = 264105, upload-time = "2025-09-01T22:08:08.708Z" },
-    { url = "https://files.pythonhosted.org/packages/97/73/fb82faaf0375aeaa1bb675008246c79b6779fa5688585a35327610ea0e2e/regex-2025.9.1-cp310-cp310-win_amd64.whl", hash = "sha256:94533e32dc0065eca43912ee6649c90ea0681d59f56d43c45b5bcda9a740b3dd", size = 276131, upload-time = "2025-09-01T22:08:10.156Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/3a/77d7718a2493e54725494f44da1a1e55704743dc4b8fabe5b0596f7b8014/regex-2025.9.1-cp310-cp310-win_arm64.whl", hash = "sha256:a874a61bb580d48642ffd338570ee24ab13fa023779190513fcacad104a6e251", size = 268462, upload-time = "2025-09-01T22:08:11.651Z" },
-    { url = "https://files.pythonhosted.org/packages/06/4d/f741543c0c59f96c6625bc6c11fea1da2e378b7d293ffff6f318edc0ce14/regex-2025.9.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:e5bcf112b09bfd3646e4db6bf2e598534a17d502b0c01ea6550ba4eca780c5e6", size = 484811, upload-time = "2025-09-01T22:08:12.834Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/bd/27e73e92635b6fbd51afc26a414a3133243c662949cd1cda677fe7bb09bd/regex-2025.9.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:67a0295a3c31d675a9ee0238d20238ff10a9a2fdb7a1323c798fc7029578b15c", size = 288977, upload-time = "2025-09-01T22:08:14.499Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/7d/7dc0c6efc8bc93cd6e9b947581f5fde8a5dbaa0af7c4ec818c5729fdc807/regex-2025.9.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ea8267fbadc7d4bd7c1301a50e85c2ff0de293ff9452a1a9f8d82c6cafe38179", size = 286606, upload-time = "2025-09-01T22:08:15.881Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/01/9b5c6dd394f97c8f2c12f6e8f96879c9ac27292a718903faf2e27a0c09f6/regex-2025.9.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6aeff21de7214d15e928fb5ce757f9495214367ba62875100d4c18d293750cc1", size = 792436, upload-time = "2025-09-01T22:08:17.38Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/24/b7430cfc6ee34bbb3db6ff933beb5e7692e5cc81e8f6f4da63d353566fb0/regex-2025.9.1-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:d89f1bbbbbc0885e1c230f7770d5e98f4f00b0ee85688c871d10df8b184a6323", size = 858705, upload-time = "2025-09-01T22:08:19.037Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/98/155f914b4ea6ae012663188545c4f5216c11926d09b817127639d618b003/regex-2025.9.1-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:ca3affe8ddea498ba9d294ab05f5f2d3b5ad5d515bc0d4a9016dd592a03afe52", size = 905881, upload-time = "2025-09-01T22:08:20.377Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/a7/a470e7bc8259c40429afb6d6a517b40c03f2f3e455c44a01abc483a1c512/regex-2025.9.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:91892a7a9f0a980e4c2c85dd19bc14de2b219a3a8867c4b5664b9f972dcc0c78", size = 798968, upload-time = "2025-09-01T22:08:22.081Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/fa/33f6fec4d41449fea5f62fdf5e46d668a1c046730a7f4ed9f478331a8e3a/regex-2025.9.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:e1cb40406f4ae862710615f9f636c1e030fd6e6abe0e0f65f6a695a2721440c6", size = 781884, upload-time = "2025-09-01T22:08:23.832Z" },
-    { url = "https://files.pythonhosted.org/packages/42/de/2b45f36ab20da14eedddf5009d370625bc5942d9953fa7e5037a32d66843/regex-2025.9.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:94f6cff6f7e2149c7e6499a6ecd4695379eeda8ccbccb9726e8149f2fe382e92", size = 852935, upload-time = "2025-09-01T22:08:25.536Z" },
-    { url = "https://files.pythonhosted.org/packages/1e/f9/878f4fc92c87e125e27aed0f8ee0d1eced9b541f404b048f66f79914475a/regex-2025.9.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:6c0226fb322b82709e78c49cc33484206647f8a39954d7e9de1567f5399becd0", size = 844340, upload-time = "2025-09-01T22:08:27.141Z" },
-    { url = "https://files.pythonhosted.org/packages/90/c2/5b6f2bce6ece5f8427c718c085eca0de4bbb4db59f54db77aa6557aef3e9/regex-2025.9.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a12f59c7c380b4fcf7516e9cbb126f95b7a9518902bcf4a852423ff1dcd03e6a", size = 787238, upload-time = "2025-09-01T22:08:28.75Z" },
-    { url = "https://files.pythonhosted.org/packages/47/66/1ef1081c831c5b611f6f55f6302166cfa1bc9574017410ba5595353f846a/regex-2025.9.1-cp311-cp311-win32.whl", hash = "sha256:49865e78d147a7a4f143064488da5d549be6bfc3f2579e5044cac61f5c92edd4", size = 264118, upload-time = "2025-09-01T22:08:30.388Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/e0/8adc550d7169df1d6b9be8ff6019cda5291054a0107760c2f30788b6195f/regex-2025.9.1-cp311-cp311-win_amd64.whl", hash = "sha256:d34b901f6f2f02ef60f4ad3855d3a02378c65b094efc4b80388a3aeb700a5de7", size = 276151, upload-time = "2025-09-01T22:08:32.073Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/bd/46fef29341396d955066e55384fb93b0be7d64693842bf4a9a398db6e555/regex-2025.9.1-cp311-cp311-win_arm64.whl", hash = "sha256:47d7c2dab7e0b95b95fd580087b6ae196039d62306a592fa4e162e49004b6299", size = 268460, upload-time = "2025-09-01T22:08:33.281Z" },
-    { url = "https://files.pythonhosted.org/packages/39/ef/a0372febc5a1d44c1be75f35d7e5aff40c659ecde864d7fa10e138f75e74/regex-2025.9.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:84a25164bd8dcfa9f11c53f561ae9766e506e580b70279d05a7946510bdd6f6a", size = 486317, upload-time = "2025-09-01T22:08:34.529Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/25/d64543fb7eb41a1024786d518cc57faf1ce64aa6e9ddba097675a0c2f1d2/regex-2025.9.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:645e88a73861c64c1af558dd12294fb4e67b5c1eae0096a60d7d8a2143a611c7", size = 289698, upload-time = "2025-09-01T22:08:36.162Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/dc/fbf31fc60be317bd9f6f87daa40a8a9669b3b392aa8fe4313df0a39d0722/regex-2025.9.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:10a450cba5cd5409526ee1d4449f42aad38dd83ac6948cbd6d7f71ca7018f7db", size = 287242, upload-time = "2025-09-01T22:08:37.794Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/74/f933a607a538f785da5021acf5323961b4620972e2c2f1f39b6af4b71db7/regex-2025.9.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e9dc5991592933a4192c166eeb67b29d9234f9c86344481173d1bc52f73a7104", size = 797441, upload-time = "2025-09-01T22:08:39.108Z" },
-    { url = "https://files.pythonhosted.org/packages/89/d0/71fc49b4f20e31e97f199348b8c4d6e613e7b6a54a90eb1b090c2b8496d7/regex-2025.9.1-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:a32291add816961aab472f4fad344c92871a2ee33c6c219b6598e98c1f0108f2", size = 862654, upload-time = "2025-09-01T22:08:40.586Z" },
-    { url = "https://files.pythonhosted.org/packages/59/05/984edce1411a5685ba9abbe10d42cdd9450aab4a022271f9585539788150/regex-2025.9.1-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:588c161a68a383478e27442a678e3b197b13c5ba51dbba40c1ccb8c4c7bee9e9", size = 910862, upload-time = "2025-09-01T22:08:42.416Z" },
-    { url = "https://files.pythonhosted.org/packages/b2/02/5c891bb5fe0691cc1bad336e3a94b9097fbcf9707ec8ddc1dce9f0397289/regex-2025.9.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:47829ffaf652f30d579534da9085fe30c171fa2a6744a93d52ef7195dc38218b", size = 801991, upload-time = "2025-09-01T22:08:44.072Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/ae/fd10d6ad179910f7a1b3e0a7fde1ef8bb65e738e8ac4fd6ecff3f52252e4/regex-2025.9.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:1e978e5a35b293ea43f140c92a3269b6ab13fe0a2bf8a881f7ac740f5a6ade85", size = 786651, upload-time = "2025-09-01T22:08:46.079Z" },
-    { url = "https://files.pythonhosted.org/packages/30/cf/9d686b07bbc5bf94c879cc168db92542d6bc9fb67088d03479fef09ba9d3/regex-2025.9.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:4cf09903e72411f4bf3ac1eddd624ecfd423f14b2e4bf1c8b547b72f248b7bf7", size = 856556, upload-time = "2025-09-01T22:08:48.376Z" },
-    { url = "https://files.pythonhosted.org/packages/91/9d/302f8a29bb8a49528abbab2d357a793e2a59b645c54deae0050f8474785b/regex-2025.9.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:d016b0f77be63e49613c9e26aaf4a242f196cd3d7a4f15898f5f0ab55c9b24d2", size = 849001, upload-time = "2025-09-01T22:08:50.067Z" },
-    { url = "https://files.pythonhosted.org/packages/93/fa/b4c6dbdedc85ef4caec54c817cd5f4418dbfa2453214119f2538082bf666/regex-2025.9.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:656563e620de6908cd1c9d4f7b9e0777e3341ca7db9d4383bcaa44709c90281e", size = 788138, upload-time = "2025-09-01T22:08:51.933Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/1b/91ee17a3cbf87f81e8c110399279d0e57f33405468f6e70809100f2ff7d8/regex-2025.9.1-cp312-cp312-win32.whl", hash = "sha256:df33f4ef07b68f7ab637b1dbd70accbf42ef0021c201660656601e8a9835de45", size = 264524, upload-time = "2025-09-01T22:08:53.75Z" },
-    { url = "https://files.pythonhosted.org/packages/92/28/6ba31cce05b0f1ec6b787921903f83bd0acf8efde55219435572af83c350/regex-2025.9.1-cp312-cp312-win_amd64.whl", hash = "sha256:5aba22dfbc60cda7c0853516104724dc904caa2db55f2c3e6e984eb858d3edf3", size = 275489, upload-time = "2025-09-01T22:08:55.037Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/ed/ea49f324db00196e9ef7fe00dd13c6164d5173dd0f1bbe495e61bb1fb09d/regex-2025.9.1-cp312-cp312-win_arm64.whl", hash = "sha256:ec1efb4c25e1849c2685fa95da44bfde1b28c62d356f9c8d861d4dad89ed56e9", size = 268589, upload-time = "2025-09-01T22:08:56.369Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/28/3f2b211686160f15ccf2ecd770e25c588fb776332956006e02ab93982a5f/regex-2025.9.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:a13d20007dce3c4b00af5d84f6c191ed1c0f70928c6d9b6cd7b8d2f125df7f46", size = 484825, upload-time = "2025-09-01T22:09:44.502Z" },
-    { url = "https://files.pythonhosted.org/packages/be/8d/88539f6ca5967022942ab64f2038a037b344db636c80852d2b4b147f6b1a/regex-2025.9.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:d6b046b0a01cb713fd53ef36cb59db4b0062b343db28e83b52ac6aa01ee5b368", size = 288972, upload-time = "2025-09-01T22:09:46.415Z" },
-    { url = "https://files.pythonhosted.org/packages/88/ad/0b9316cfe91d8a2917fbcd8cb7c990051745b5852f9f3088b93b4267fe68/regex-2025.9.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:0fa9a7477288717f42dbd02ff5d13057549e9a8cdb81f224c313154cc10bab52", size = 286624, upload-time = "2025-09-01T22:09:47.823Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/6e/199003d956dac19ddaebb2ef3c068379918539a497dd4d4eb5561a1c5e3f/regex-2025.9.1-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b2b3ad150c6bc01a8cd5030040675060e2adbe6cbc50aadc4da42c6d32ec266e", size = 779882, upload-time = "2025-09-01T22:09:49.601Z" },
-    { url = "https://files.pythonhosted.org/packages/69/0a/a4bed3424beefbf8322b8b581aa2e2a65cbfe358d968f58a2ab910d62927/regex-2025.9.1-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:aa88d5a82dfe80deaf04e8c39c8b0ad166d5d527097eb9431cb932c44bf88715", size = 848938, upload-time = "2025-09-01T22:09:51.155Z" },
-    { url = "https://files.pythonhosted.org/packages/d2/01/c6c2bc65e97387523742c54daec1384a50d2575d1189e844e4e320e145e9/regex-2025.9.1-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:6f1dae2cf6c2dbc6fd2526653692c144721b3cf3f769d2a3c3aa44d0f38b9a58", size = 896747, upload-time = "2025-09-01T22:09:52.965Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/56/25480407a452198414a68d6d1f4f9920652810ba9f2c7f6279e5bd9b15ef/regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ff62a3022914fc19adaa76b65e03cf62bc67ea16326cbbeb170d280710a7d719", size = 789466, upload-time = "2025-09-01T22:09:54.564Z" },
-    { url = "https://files.pythonhosted.org/packages/82/5c/207422b3a2c3923eb09a1e807435effcbd8b756cd8ece24c153d525463d6/regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:a34ef82216189d823bc82f614d1031cb0b919abef27cecfd7b07d1e9a8bdeeb4", size = 780130, upload-time = "2025-09-01T22:09:56.413Z" },
-    { url = "https://files.pythonhosted.org/packages/2c/e3/74a9da339f043006a1c0e0daae143c2d8b70e8987da3c967dbb8b6e8c9d2/regex-2025.9.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:6d40e6b49daae9ebbd7fa4e600697372cba85b826592408600068e83a3c47211", size = 773140, upload-time = "2025-09-01T22:09:57.99Z" },
-    { url = "https://files.pythonhosted.org/packages/45/11/d2ca4dccec797f9325cfae0cfff8ee2977e1d5db43296ed949f8c63b6db0/regex-2025.9.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:0aeb0fe80331059c152a002142699a89bf3e44352aee28261315df0c9874759b", size = 843538, upload-time = "2025-09-01T22:09:59.74Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/9a/652095a5df5a8cd6b550280dc4b885f06beb68e2dc8728b380639b40446d/regex-2025.9.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:a90014d29cb3098403d82a879105d1418edbbdf948540297435ea6e377023ea7", size = 834151, upload-time = "2025-09-01T22:10:01.658Z" },
-    { url = "https://files.pythonhosted.org/packages/17/27/f1ca20401df0544a627188fede42ef723b7d88fed8364b0d80fa9833bb84/regex-2025.9.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:6ff623271e0b0cc5a95b802666bbd70f17ddd641582d65b10fb260cc0c003529", size = 778009, upload-time = "2025-09-01T22:10:03.609Z" },
-    { url = "https://files.pythonhosted.org/packages/20/28/c6c94f642c115e72142e8404464835f03dee55266a158455aeb8216d6d4d/regex-2025.9.1-cp39-cp39-win32.whl", hash = "sha256:d161bfdeabe236290adfd8c7588da7f835d67e9e7bf2945f1e9e120622839ba6", size = 264141, upload-time = "2025-09-01T22:10:05.457Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/7e/4ef3cf970b4c7ef9abee4c5a42f3405a452e5abe21095ee36f9a3a83af9d/regex-2025.9.1-cp39-cp39-win_amd64.whl", hash = "sha256:43ebc77a7dfe36661192afd8d7df5e8be81ec32d2ad0c65b536f66ebfec3dece", size = 276219, upload-time = "2025-09-01T22:10:07.106Z" },
-    { url = "https://files.pythonhosted.org/packages/60/42/86a45b36b0d8f37388485a2c93d0f3c571732ac4e17b92d8ccd5da53b792/regex-2025.9.1-cp39-cp39-win_arm64.whl", hash = "sha256:5d74b557cf5554001a869cda60b9a619be307df4d10155894aeaad3ee67c9899", size = 268497, upload-time = "2025-09-01T22:10:08.595Z" },
+version = "2025.9.18"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/49/d3/eaa0d28aba6ad1827ad1e716d9a93e1ba963ada61887498297d3da715133/regex-2025.9.18.tar.gz", hash = "sha256:c5ba23274c61c6fef447ba6a39333297d0c247f53059dba0bca415cac511edc4", size = 400917, upload-time = "2025-09-19T00:38:35.79Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7e/d8/7e06171db8e55f917c5b8e89319cea2d86982e3fc46b677f40358223dece/regex-2025.9.18-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:12296202480c201c98a84aecc4d210592b2f55e200a1d193235c4db92b9f6788", size = 484829, upload-time = "2025-09-19T00:35:05.215Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/70/bf91bb39e5bedf75ce730ffbaa82ca585584d13335306d637458946b8b9f/regex-2025.9.18-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:220381f1464a581f2ea988f2220cf2a67927adcef107d47d6897ba5a2f6d51a4", size = 288993, upload-time = "2025-09-19T00:35:08.154Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/89/69f79b28365eda2c46e64c39d617d5f65a2aa451a4c94de7d9b34c2dc80f/regex-2025.9.18-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:87f681bfca84ebd265278b5daa1dcb57f4db315da3b5d044add7c30c10442e61", size = 286624, upload-time = "2025-09-19T00:35:09.717Z" },
+    { url = "https://files.pythonhosted.org/packages/44/31/81e62955726c3a14fcc1049a80bc716765af6c055706869de5e880ddc783/regex-2025.9.18-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:34d674cbba70c9398074c8a1fcc1a79739d65d1105de2a3c695e2b05ea728251", size = 780473, upload-time = "2025-09-19T00:35:11.013Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/23/07072b7e191fbb6e213dc03b2f5b96f06d3c12d7deaded84679482926fc7/regex-2025.9.18-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:385c9b769655cb65ea40b6eea6ff763cbb6d69b3ffef0b0db8208e1833d4e746", size = 849290, upload-time = "2025-09-19T00:35:12.348Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/f0/aec7f6a01f2a112210424d77c6401b9015675fb887ced7e18926df4ae51e/regex-2025.9.18-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:8900b3208e022570ae34328712bef6696de0804c122933414014bae791437ab2", size = 897335, upload-time = "2025-09-19T00:35:14.058Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/90/2e5f9da89d260de7d0417ead91a1bc897f19f0af05f4f9323313b76c47f2/regex-2025.9.18-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c204e93bf32cd7a77151d44b05eb36f469d0898e3fba141c026a26b79d9914a0", size = 789946, upload-time = "2025-09-19T00:35:15.403Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/d5/1c712c7362f2563d389be66bae131c8bab121a3fabfa04b0b5bfc9e73c51/regex-2025.9.18-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3acc471d1dd7e5ff82e6cacb3b286750decd949ecd4ae258696d04f019817ef8", size = 780787, upload-time = "2025-09-19T00:35:17.061Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/92/c54cdb4aa41009632e69817a5aa452673507f07e341076735a2f6c46a37c/regex-2025.9.18-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:6479d5555122433728760e5f29edb4c2b79655a8deb681a141beb5c8a025baea", size = 773632, upload-time = "2025-09-19T00:35:18.57Z" },
+    { url = "https://files.pythonhosted.org/packages/db/99/75c996dc6a2231a8652d7ad0bfbeaf8a8c77612d335580f520f3ec40e30b/regex-2025.9.18-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:431bd2a8726b000eb6f12429c9b438a24062a535d06783a93d2bcbad3698f8a8", size = 844104, upload-time = "2025-09-19T00:35:20.259Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/f7/25aba34cc130cb6844047dbfe9716c9b8f9629fee8b8bec331aa9241b97b/regex-2025.9.18-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:0cc3521060162d02bd36927e20690129200e5ac9d2c6d32b70368870b122db25", size = 834794, upload-time = "2025-09-19T00:35:22.002Z" },
+    { url = "https://files.pythonhosted.org/packages/51/eb/64e671beafa0ae29712268421597596d781704973551312b2425831d4037/regex-2025.9.18-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:a021217b01be2d51632ce056d7a837d3fa37c543ede36e39d14063176a26ae29", size = 778535, upload-time = "2025-09-19T00:35:23.298Z" },
+    { url = "https://files.pythonhosted.org/packages/26/33/c0ebc0b07bd0bf88f716cca240546b26235a07710ea58e271cfe390ae273/regex-2025.9.18-cp310-cp310-win32.whl", hash = "sha256:4a12a06c268a629cb67cc1d009b7bb0be43e289d00d5111f86a2efd3b1949444", size = 264115, upload-time = "2025-09-19T00:35:25.206Z" },
+    { url = "https://files.pythonhosted.org/packages/59/39/aeb11a4ae68faaec2498512cadae09f2d8a91f1f65730fe62b9bffeea150/regex-2025.9.18-cp310-cp310-win_amd64.whl", hash = "sha256:47acd811589301298c49db2c56bde4f9308d6396da92daf99cba781fa74aa450", size = 276143, upload-time = "2025-09-19T00:35:26.785Z" },
+    { url = "https://files.pythonhosted.org/packages/29/04/37f2d3fc334a1031fc2767c9d89cec13c2e72207c7e7f6feae8a47f4e149/regex-2025.9.18-cp310-cp310-win_arm64.whl", hash = "sha256:16bd2944e77522275e5ee36f867e19995bcaa533dcb516753a26726ac7285442", size = 268473, upload-time = "2025-09-19T00:35:28.39Z" },
+    { url = "https://files.pythonhosted.org/packages/58/61/80eda662fc4eb32bfedc331f42390974c9e89c7eac1b79cd9eea4d7c458c/regex-2025.9.18-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:51076980cd08cd13c88eb7365427ae27f0d94e7cebe9ceb2bb9ffdae8fc4d82a", size = 484832, upload-time = "2025-09-19T00:35:30.011Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/d9/33833d9abddf3f07ad48504ddb53fe3b22f353214bbb878a72eee1e3ddbf/regex-2025.9.18-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:828446870bd7dee4e0cbeed767f07961aa07f0ea3129f38b3ccecebc9742e0b8", size = 288994, upload-time = "2025-09-19T00:35:31.733Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/b3/526ee96b0d70ea81980cbc20c3496fa582f775a52e001e2743cc33b2fa75/regex-2025.9.18-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:c28821d5637866479ec4cc23b8c990f5bc6dd24e5e4384ba4a11d38a526e1414", size = 286619, upload-time = "2025-09-19T00:35:33.221Z" },
+    { url = "https://files.pythonhosted.org/packages/65/4f/c2c096b02a351b33442aed5895cdd8bf87d372498d2100927c5a053d7ba3/regex-2025.9.18-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:726177ade8e481db669e76bf99de0b278783be8acd11cef71165327abd1f170a", size = 792454, upload-time = "2025-09-19T00:35:35.361Z" },
+    { url = "https://files.pythonhosted.org/packages/24/15/b562c9d6e47c403c4b5deb744f8b4bf6e40684cf866c7b077960a925bdff/regex-2025.9.18-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f5cca697da89b9f8ea44115ce3130f6c54c22f541943ac8e9900461edc2b8bd4", size = 858723, upload-time = "2025-09-19T00:35:36.949Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/01/dba305409849e85b8a1a681eac4c03ed327d8de37895ddf9dc137f59c140/regex-2025.9.18-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:dfbde38f38004703c35666a1e1c088b778e35d55348da2b7b278914491698d6a", size = 905899, upload-time = "2025-09-19T00:35:38.723Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/d0/c51d1e6a80eab11ef96a4cbad17fc0310cf68994fb01a7283276b7e5bbd6/regex-2025.9.18-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f2f422214a03fab16bfa495cfec72bee4aaa5731843b771860a471282f1bf74f", size = 798981, upload-time = "2025-09-19T00:35:40.416Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/5e/72db90970887bbe02296612bd61b0fa31e6d88aa24f6a4853db3e96c575e/regex-2025.9.18-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:a295916890f4df0902e4286bc7223ee7f9e925daa6dcdec4192364255b70561a", size = 781900, upload-time = "2025-09-19T00:35:42.077Z" },
+    { url = "https://files.pythonhosted.org/packages/50/ff/596be45eea8e9bc31677fde243fa2904d00aad1b32c31bce26c3dbba0b9e/regex-2025.9.18-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:5db95ff632dbabc8c38c4e82bf545ab78d902e81160e6e455598014f0abe66b9", size = 852952, upload-time = "2025-09-19T00:35:43.751Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/1b/2dfa348fa551e900ed3f5f63f74185b6a08e8a76bc62bc9c106f4f92668b/regex-2025.9.18-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:fb967eb441b0f15ae610b7069bdb760b929f267efbf522e814bbbfffdf125ce2", size = 844355, upload-time = "2025-09-19T00:35:45.309Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/bf/aefb1def27fe33b8cbbb19c75c13aefccfbef1c6686f8e7f7095705969c7/regex-2025.9.18-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:f04d2f20da4053d96c08f7fde6e1419b7ec9dbcee89c96e3d731fca77f411b95", size = 787254, upload-time = "2025-09-19T00:35:46.904Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/4e/8ef042e7cf0dbbb401e784e896acfc1b367b95dfbfc9ada94c2ed55a081f/regex-2025.9.18-cp311-cp311-win32.whl", hash = "sha256:895197241fccf18c0cea7550c80e75f185b8bd55b6924fcae269a1a92c614a07", size = 264129, upload-time = "2025-09-19T00:35:48.597Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/7d/c4fcabf80dcdd6821c0578ad9b451f8640b9110fb3dcb74793dd077069ff/regex-2025.9.18-cp311-cp311-win_amd64.whl", hash = "sha256:7e2b414deae99166e22c005e154a5513ac31493db178d8aec92b3269c9cce8c9", size = 276160, upload-time = "2025-09-19T00:36:00.45Z" },
+    { url = "https://files.pythonhosted.org/packages/64/f8/0e13c8ae4d6df9d128afaba138342d532283d53a4c1e7a8c93d6756c8f4a/regex-2025.9.18-cp311-cp311-win_arm64.whl", hash = "sha256:fb137ec7c5c54f34a25ff9b31f6b7b0c2757be80176435bf367111e3f71d72df", size = 268471, upload-time = "2025-09-19T00:36:02.149Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/99/05859d87a66ae7098222d65748f11ef7f2dff51bfd7482a4e2256c90d72b/regex-2025.9.18-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:436e1b31d7efd4dcd52091d076482031c611dde58bf9c46ca6d0a26e33053a7e", size = 486335, upload-time = "2025-09-19T00:36:03.661Z" },
+    { url = "https://files.pythonhosted.org/packages/97/7e/d43d4e8b978890932cf7b0957fce58c5b08c66f32698f695b0c2c24a48bf/regex-2025.9.18-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c190af81e5576b9c5fdc708f781a52ff20f8b96386c6e2e0557a78402b029f4a", size = 289720, upload-time = "2025-09-19T00:36:05.471Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/3b/ff80886089eb5dcf7e0d2040d9aaed539e25a94300403814bb24cc775058/regex-2025.9.18-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:e4121f1ce2b2b5eec4b397cc1b277686e577e658d8f5870b7eb2d726bd2300ab", size = 287257, upload-time = "2025-09-19T00:36:07.072Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/66/243edf49dd8720cba8d5245dd4d6adcb03a1defab7238598c0c97cf549b8/regex-2025.9.18-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:300e25dbbf8299d87205e821a201057f2ef9aa3deb29caa01cd2cac669e508d5", size = 797463, upload-time = "2025-09-19T00:36:08.399Z" },
+    { url = "https://files.pythonhosted.org/packages/df/71/c9d25a1142c70432e68bb03211d4a82299cd1c1fbc41db9409a394374ef5/regex-2025.9.18-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7b47fcf9f5316c0bdaf449e879407e1b9937a23c3b369135ca94ebc8d74b1742", size = 862670, upload-time = "2025-09-19T00:36:10.101Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/8f/329b1efc3a64375a294e3a92d43372bf1a351aa418e83c21f2f01cf6ec41/regex-2025.9.18-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:57a161bd3acaa4b513220b49949b07e252165e6b6dc910ee7617a37ff4f5b425", size = 910881, upload-time = "2025-09-19T00:36:12.223Z" },
+    { url = "https://files.pythonhosted.org/packages/35/9e/a91b50332a9750519320ed30ec378b74c996f6befe282cfa6bb6cea7e9fd/regex-2025.9.18-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4f130c3a7845ba42de42f380fff3c8aebe89a810747d91bcf56d40a069f15352", size = 802011, upload-time = "2025-09-19T00:36:13.901Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/1d/6be3b8d7856b6e0d7ee7f942f437d0a76e0d5622983abbb6d21e21ab9a17/regex-2025.9.18-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5f96fa342b6f54dcba928dd452e8d8cb9f0d63e711d1721cd765bb9f73bb048d", size = 786668, upload-time = "2025-09-19T00:36:15.391Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/ce/4a60e53df58bd157c5156a1736d3636f9910bdcc271d067b32b7fcd0c3a8/regex-2025.9.18-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:0f0d676522d68c207828dcd01fb6f214f63f238c283d9f01d85fc664c7c85b56", size = 856578, upload-time = "2025-09-19T00:36:16.845Z" },
+    { url = "https://files.pythonhosted.org/packages/86/e8/162c91bfe7217253afccde112868afb239f94703de6580fb235058d506a6/regex-2025.9.18-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:40532bff8a1a0621e7903ae57fce88feb2e8a9a9116d341701302c9302aef06e", size = 849017, upload-time = "2025-09-19T00:36:18.597Z" },
+    { url = "https://files.pythonhosted.org/packages/35/34/42b165bc45289646ea0959a1bc7531733e90b47c56a72067adfe6b3251f6/regex-2025.9.18-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:039f11b618ce8d71a1c364fdee37da1012f5a3e79b1b2819a9f389cd82fd6282", size = 788150, upload-time = "2025-09-19T00:36:20.464Z" },
+    { url = "https://files.pythonhosted.org/packages/79/5d/cdd13b1f3c53afa7191593a7ad2ee24092a5a46417725ffff7f64be8342d/regex-2025.9.18-cp312-cp312-win32.whl", hash = "sha256:e1dd06f981eb226edf87c55d523131ade7285137fbde837c34dc9d1bf309f459", size = 264536, upload-time = "2025-09-19T00:36:21.922Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/f5/4a7770c9a522e7d2dc1fa3ffc83ab2ab33b0b22b447e62cffef186805302/regex-2025.9.18-cp312-cp312-win_amd64.whl", hash = "sha256:3d86b5247bf25fa3715e385aa9ff272c307e0636ce0c9595f64568b41f0a9c77", size = 275501, upload-time = "2025-09-19T00:36:23.4Z" },
+    { url = "https://files.pythonhosted.org/packages/df/05/9ce3e110e70d225ecbed455b966003a3afda5e58e8aec2964042363a18f4/regex-2025.9.18-cp312-cp312-win_arm64.whl", hash = "sha256:032720248cbeeae6444c269b78cb15664458b7bb9ed02401d3da59fe4d68c3a5", size = 268601, upload-time = "2025-09-19T00:36:25.092Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/d2/5b0ded10467d6e96f78de5e6f195b7f9b57251f411b1090004597cffe5d9/regex-2025.9.18-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:3dbcfcaa18e9480669030d07371713c10b4f1a41f791ffa5cb1a99f24e777f40", size = 484847, upload-time = "2025-09-19T00:38:07.367Z" },
+    { url = "https://files.pythonhosted.org/packages/55/35/051da2c0ae6124e3f1aa1442ecc2bb4e2de930e95433bce1301a2e7ae255/regex-2025.9.18-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1e85f73ef7095f0380208269055ae20524bfde3f27c5384126ddccf20382a638", size = 288995, upload-time = "2025-09-19T00:38:09.253Z" },
+    { url = "https://files.pythonhosted.org/packages/22/4b/4bfc51cad95263d25b6ed8c5253831b2536e8e279e6736d0a08c9f7ffe98/regex-2025.9.18-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:9098e29b3ea4ffffeade423f6779665e2a4f8db64e699c0ed737ef0db6ba7b12", size = 286642, upload-time = "2025-09-19T00:38:11.012Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/67/d2f3e2483e09d1e9f7d93b4fe106b04933fba5e619bc901530d1c90d62da/regex-2025.9.18-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:90b6b7a2d0f45b7ecaaee1aec6b362184d6596ba2092dd583ffba1b78dd0231c", size = 779896, upload-time = "2025-09-19T00:38:12.732Z" },
+    { url = "https://files.pythonhosted.org/packages/14/5e/49a4f07ce6f5563de02b0e321220b9534f3fd3bae275311b785dd618aea5/regex-2025.9.18-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c81b892af4a38286101502eae7aec69f7cd749a893d9987a92776954f3943408", size = 848954, upload-time = "2025-09-19T00:38:14.716Z" },
+    { url = "https://files.pythonhosted.org/packages/00/8d/f5995ae51225c77ca9215d78ceb1dc30c52fa2b22c41dac977214e8b4bbd/regex-2025.9.18-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:3b524d010973f2e1929aeb635418d468d869a5f77b52084d9f74c272189c251d", size = 896770, upload-time = "2025-09-19T00:38:16.381Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/15/2a3a744d73a557337c7561db2114bab10b4e9941c626c03169ea62f42c8f/regex-2025.9.18-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6b498437c026a3d5d0be0020023ff76d70ae4d77118e92f6f26c9d0423452446", size = 789484, upload-time = "2025-09-19T00:38:18.183Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/27/e425f3d17d32062a657b836d0c8a68f5e71a9e6295fa637159f265eaa609/regex-2025.9.18-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:0716e4d6e58853d83f6563f3cf25c281ff46cf7107e5f11879e32cb0b59797d9", size = 780150, upload-time = "2025-09-19T00:38:19.879Z" },
+    { url = "https://files.pythonhosted.org/packages/62/28/79dfae89b6fd7901b82611ac1a96ec25deceb7e918e9c5eb3f96cf5ad654/regex-2025.9.18-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:065b6956749379d41db2625f880b637d4acc14c0a4de0d25d609a62850e96d36", size = 773160, upload-time = "2025-09-19T00:38:21.641Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/67/df83d6ae608f487448e9be7ac26211af2afa2b6e34465fde3e07d1f11290/regex-2025.9.18-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:d4a691494439287c08ddb9b5793da605ee80299dd31e95fa3f323fac3c33d9d4", size = 843555, upload-time = "2025-09-19T00:38:23.696Z" },
+    { url = "https://files.pythonhosted.org/packages/32/67/c65f56f3edd3f213d3aa41e9b9b07cc2247721a23d34bcfb2947dc0f4685/regex-2025.9.18-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:ef8d10cc0989565bcbe45fb4439f044594d5c2b8919d3d229ea2c4238f1d55b0", size = 834169, upload-time = "2025-09-19T00:38:25.997Z" },
+    { url = "https://files.pythonhosted.org/packages/95/90/7fca37435e3aa1a032c38fa1e171fdaf809c8dbf2717508e3f6a92c75446/regex-2025.9.18-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:4baeb1b16735ac969a7eeecc216f1f8b7caf60431f38a2671ae601f716a32d25", size = 778024, upload-time = "2025-09-19T00:38:28.043Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/05/c2ee512cdf34d6be5ac5cf938a58c1b79a9d96cbad404bc4d70404212edb/regex-2025.9.18-cp39-cp39-win32.whl", hash = "sha256:8e5f41ad24a1e0b5dfcf4c4e5d9f5bd54c895feb5708dd0c1d0d35693b24d478", size = 264151, upload-time = "2025-09-19T00:38:30.23Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/2f/8414fb46181b6108484f04d670ece196db6734cc4c683f41125043fd3280/regex-2025.9.18-cp39-cp39-win_amd64.whl", hash = "sha256:50e8290707f2fb8e314ab3831e594da71e062f1d623b05266f8cfe4db4949afd", size = 276232, upload-time = "2025-09-19T00:38:31.981Z" },
+    { url = "https://files.pythonhosted.org/packages/61/63/f40931d477e1ed4b53105d506758a58cfec1b052c12972054930ec743ee5/regex-2025.9.18-cp39-cp39-win_arm64.whl", hash = "sha256:039a9d7195fd88c943d7c777d4941e8ef736731947becce773c31a1009cb3c35", size = 268505, upload-time = "2025-09-19T00:38:34.015Z" },
 ]
 
 [[package]]
@@ -3672,6 +4308,95 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/e3/30/3c4d035596d3cf444529e0b2953ad0466f6049528a879d27534700580395/rich-14.1.0-py3-none-any.whl", hash = "sha256:536f5f1785986d6dbdea3c75205c473f970777b4a0d6c6dd1b696aa05a3fa04f", size = 243368, upload-time = "2025-07-25T07:32:56.73Z" },
 ]
 
+[[package]]
+name = "rich-toolkit"
+version = "0.15.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "rich" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/67/33/1a18839aaa8feef7983590c05c22c9c09d245ada6017d118325bbfcc7651/rich_toolkit-0.15.1.tar.gz", hash = "sha256:6f9630eb29f3843d19d48c3bd5706a086d36d62016687f9d0efa027ddc2dd08a", size = 115322, upload-time = "2025-09-04T09:28:11.789Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c8/49/42821d55ead7b5a87c8d121edf323cb393d8579f63e933002ade900b784f/rich_toolkit-0.15.1-py3-none-any.whl", hash = "sha256:36a0b1d9a135d26776e4b78f1d5c2655da6e0ef432380b5c6b523c8d8ab97478", size = 29412, upload-time = "2025-09-04T09:28:10.587Z" },
+]
+
+[[package]]
+name = "rignore"
+version = "0.7.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ae/46/e5ef3423a3746f91d3a3d9a68c499fde983be7dbab7d874efa8d3bb139ba/rignore-0.7.0.tar.gz", hash = "sha256:cfe6a2cbec855b440d7550d53e670246fce43ca5847e46557b6d4577c9cdb540", size = 12796, upload-time = "2025-10-02T13:26:22.194Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b3/19/04831e4d3db0d828f9cf497b53c944b9c56c26fba764c98747013aae0585/rignore-0.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7e011b6412690e34113d5cb133844bfe087fe9a57b37c63cb68671dfbf6080ed", size = 890938, upload-time = "2025-10-02T13:23:16.526Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/3a/da748c8ec25fa15a855fdb6f66c86fc1b1756f5cbe354389d4311d84022e/rignore-0.7.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:c177b8267aa361bf04f9e28fa948881ff01e98e2556bf9d39b088e42de23b190", size = 865825, upload-time = "2025-10-02T13:23:35.145Z" },
+    { url = "https://files.pythonhosted.org/packages/58/8a/8cd9415da272e94a5b306e37f4cc3c0631f08b884836d5c92cf90cecc7a8/rignore-0.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6cfe5873ac415f62d221d8bd04d88f9a70e73fe7aea0c094b9974e530628d8ea", size = 1168074, upload-time = "2025-10-02T13:23:52.542Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/98/008476632a518463875d44dba429a03d59333194ed3a0d08b29c0348f7c3/rignore-0.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:20d62988d4f565ab101ec33c501527fde52693eced4ea34d5da61b88db602616", size = 936248, upload-time = "2025-10-02T13:24:08.962Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/ed/6d5c345ba0de67ff4096f0ebcfd2bfdad72335ab9dabe1c665a9579ba687/rignore-0.7.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:427ba9cfc424aedb0b569d659d2f2ea88ed308d8eb245446db10733a73db0fb1", size = 951260, upload-time = "2025-10-02T13:24:38.712Z" },
+    { url = "https://files.pythonhosted.org/packages/44/ca/c56e097b091313b723416de9e28826ac92d731af5c4b51c4892e04286efc/rignore-0.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:86280aae7d0980debe5ed6785e3fc9a68ca627ac84e7c84048d7d3fe6d80ef7a", size = 975261, upload-time = "2025-10-02T13:24:25.553Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/dc/2e6987f8f6c8c96c29074901d7d5624de9d1741b4cab6c15975ad159f959/rignore-0.7.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:389c5fe844ad1fd5fba46c0cba0d9e68bfdcaae12e943b135395730efe45bcfe", size = 1071689, upload-time = "2025-10-02T13:25:15.11Z" },
+    { url = "https://files.pythonhosted.org/packages/61/83/fda08b5e11e98f9e96e8c94b7cd5c21ec50193e2861784eb6e21a61f6ccf/rignore-0.7.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:34a34d5e86fda5355b55d927f43ff515c7371e8880b8f16cb92c3278c21327ee", size = 1129324, upload-time = "2025-10-02T13:25:31.669Z" },
+    { url = "https://files.pythonhosted.org/packages/57/74/e5ee481d4f7ccc373a1ebeec2d03415d1cf2b45522ac7424fe773b454f17/rignore-0.7.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:f7aacce87c710a3f254eb8b28a0cec1801362e7cf4f8258cceb8f36fc9cc695a", size = 1108242, upload-time = "2025-10-02T13:25:48.119Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/89/195c5a909303c841ad5e1de300f1d3dd2177768cd3369318654f92b95d8a/rignore-0.7.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b3b88dfeb0d08902fe28eb5d75cbfac1d130ccdaeaca742eab1628bab7960294", size = 1116370, upload-time = "2025-10-02T13:26:06.084Z" },
+    { url = "https://files.pythonhosted.org/packages/31/08/d64298cec32d5df121968b3ab75d17d2a30ff02f080a3457893e57689809/rignore-0.7.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:686c162f945ede315b7b63958d83531b18226cad4fae9170a5787dd8b8b4be89", size = 891607, upload-time = "2025-10-02T13:23:18.739Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/b3/602bb25ba0c862dd3f7f52af0f5e3fce4321207a1b76c0b3b7f17aed0146/rignore-0.7.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:a3c8b62a00c1b6e0ed73412ba8d37d05e214e6a8757f2779d313078d2bdec209", size = 865644, upload-time = "2025-10-02T13:23:36.604Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/fc/18f5ac22714bdd0437aaa59ff2ded2ba3caff2745c89e671bc9c91c52947/rignore-0.7.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f115666738614cdb0ef122c2b48806043b9b6c603dc03a4708b2eb1df5a44514", size = 1167949, upload-time = "2025-10-02T13:23:54.257Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/1b/6409b434420995b8897c3d6b5a2701343857d2d36d159bd9305287c33634/rignore-0.7.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9ffaf2047304b97bc648592f82c0aeba3468f43546a918994411b8f1d79d42d6", size = 935950, upload-time = "2025-10-02T13:24:10.463Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/56/c0a03cb643ca41091f0377ffea3a35ae3f3cff39b075ca94eec35fae6ed0/rignore-0.7.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:04678c2f1787eb07378754d6aa50e66ce712e0b75e8b843fd9e5e4da35130617", size = 951418, upload-time = "2025-10-02T13:24:40.222Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/3b/33783bc1681662789f71614dee496fb0dd96de4887eb8d5d2cb9f365d1ff/rignore-0.7.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:53a4c4a43558f34b32732efcee9c79c7948ff26673bb764aa0e9bbe951e435fa", size = 975421, upload-time = "2025-10-02T13:24:27.049Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/e2/af19c05288c2afb5b79f73c68e88a34b88245b66e5cf358417461a72c8c5/rignore-0.7.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:794f72ce7195cad1fb41c03b3e3484396c404498b73855004ebea965a697edd9", size = 1071989, upload-time = "2025-10-02T13:25:17.248Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/ea/6ab6d1afafcd3f6e5ba898646bcfe3a6f69eb8f4ac264dd82848ab7f2c5b/rignore-0.7.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:989f35a152bc508c52d63d7d4527215c5dabe7981e5744bcf35f96c99f3758f7", size = 1129150, upload-time = "2025-10-02T13:25:33.352Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/49/a327d54cbd5f9f34ed383057ee1c9a044571878045cbd37a129f27f13ab0/rignore-0.7.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:b945b29a995fdcf669dc098ec40237131742de2cf49484011ba3f81d0fff23a3", size = 1107917, upload-time = "2025-10-02T13:25:49.702Z" },
+    { url = "https://files.pythonhosted.org/packages/86/f8/89a1269911e7895e3c4a5c1fb1abb3b9b255362035fa54c593287cf38b15/rignore-0.7.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:e4deda4c3e5cec1ebfb714094cd9af79e8840680187537d13a216377d6aa2ed6", size = 1116013, upload-time = "2025-10-02T13:26:07.597Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/e6/fd2cbc71f725ea10892c85ea56bd8f54426557cf5ac2924f9c27b771ee45/rignore-0.7.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0525ccf3e8b9ccd6f1dfc87ecc78218a83605070b247633636d144acdf6b73be", size = 892031, upload-time = "2025-10-02T13:23:20.558Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/c8/0dfd755f57515d34ca26de011e016f62db86f7bef0586f2ab0d9f6e18136/rignore-0.7.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:570bcf51fd9f78ec79ec33f2f852e6665027fae80cc3e5e2523c97d3f4220369", size = 865496, upload-time = "2025-10-02T13:23:37.965Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/b9/f73af8509842d74788fc26feca25db1eade9291fae79540872c130407340/rignore-0.7.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:32f5d3d90a520d61e43c2a23724852c689c3ed36b38264c77b613f967e2d1f68", size = 1165555, upload-time = "2025-10-02T13:23:56.009Z" },
+    { url = "https://files.pythonhosted.org/packages/44/22/67d2fb589cedd7bf3a01e16617f2da10f172165b3ecdaa8fa0707043e9ed/rignore-0.7.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a7d189cfb9059dfa497e5480c411bd2aba838124b50b93abf7e92556221b7956", size = 936631, upload-time = "2025-10-02T13:24:11.97Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/6b/e0f969a1cb3ff2caa0dd342e512d7a0a6f1b737b6f5373c04606aa946e80/rignore-0.7.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6c871a31596476ac4343f6b803ee8ddca068425e1837cf6849ebe46c498c73c5", size = 951058, upload-time = "2025-10-02T13:24:41.742Z" },
+    { url = "https://files.pythonhosted.org/packages/45/cf/ccf053fb87601332e8b2e2da707f2801bee66ee5fe843687183f45c2e768/rignore-0.7.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:5b7d8ce1efbd8fa865712d34753ce4eb8e0732874df95351244e14308fb87d0a", size = 974638, upload-time = "2025-10-02T13:24:29Z" },
+    { url = "https://files.pythonhosted.org/packages/de/ae/a00181c0d2dc437a3729dbebcfffd67bb849d1c53e45850c7b4428f5fba4/rignore-0.7.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d261aea1a51ef93c262b52ad195a1092a8bae17577e8192473d1b5fd30379346", size = 1072970, upload-time = "2025-10-02T13:25:18.888Z" },
+    { url = "https://files.pythonhosted.org/packages/81/30/3011207fc9f26f9eb21d2282dfedd8f2d66cf7a9a3053370c9b4b87601e1/rignore-0.7.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:034bef935e3734b4ad2dada59c96717f3e3d0b48551a0c79379c4d3280b4a397", size = 1128833, upload-time = "2025-10-02T13:25:34.987Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/be/4c6a860f851db6cb0b96a3ec62dd4fe95290ee36e67b845ffab58908c6cc/rignore-0.7.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:5f816b65c9bf97093d792c9b50369d5a81a5f95b4ed5f003d4091bd1db3b70d8", size = 1106909, upload-time = "2025-10-02T13:25:51.266Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/8a/691d79e72f000968e1e3457ff53634760dac24fa6c6b5663d994362b8a99/rignore-0.7.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b88479f0a89828781d25a9acd485be88abf4f1f1c14e455b6530da265adb593c", size = 1115733, upload-time = "2025-10-02T13:26:09.256Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/9d/288216c28dd35e80a56cd6f68952a1bb0915d05b429ad8003b735a3af31a/rignore-0.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e39eccbb92f1b562b72aefeec93bfae38c8f302a58aaf9b2d77362b57e1cf27", size = 892385, upload-time = "2025-10-02T13:23:27.369Z" },
+    { url = "https://files.pythonhosted.org/packages/61/19/655cd781fe7ede2c61280905f9694a4094db42b106d2f92a07197b7a3716/rignore-0.7.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:bc34644c1d42ea0ec7b7b14a56203dbf19df6c258d75937191f871da3cd1c1ca", size = 866836, upload-time = "2025-10-02T13:23:45.714Z" },
+    { url = "https://files.pythonhosted.org/packages/72/4c/b20543b6ca5fb192a04504995b140fa0b7a3f143e9a5960710d093cc3bae/rignore-0.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7cd512a48d9b0900b297fc0c2227e81096eed21e42dae5a9e82e332500fb8730", size = 1169748, upload-time = "2025-10-02T13:24:02.446Z" },
+    { url = "https://files.pythonhosted.org/packages/97/60/ef01964969ec70995c0b30cb2e90dd16dfa9792259cf5060a5cdba437f25/rignore-0.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:baee13481bf5d335aa48cb3607637f842edaf418f5dde252fe9d99f360533549", size = 937629, upload-time = "2025-10-02T13:24:18.393Z" },
+    { url = "https://files.pythonhosted.org/packages/16/cb/73fc5d9f99cf76120ec9b97b61bfc49967f37695c0ed3f61fca16a4eb1c9/rignore-0.7.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f7ae8d7ec27ce07599aef2e83e7cff831888e381db0f0ee066d9fb4873ba14ab", size = 952303, upload-time = "2025-10-02T13:24:48.683Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/8e/c7c48fbad072a9d7ca8c2af5058540db85c57d82257323df6a088c0f3205/rignore-0.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:7a68d1a180d5af3e1ca2e3e1f2b318e6925677d06f5f4d6b3cbd2f63eaad9255", size = 977032, upload-time = "2025-10-02T13:24:35.361Z" },
+    { url = "https://files.pythonhosted.org/packages/29/ca/42293e524c772ac1ab929496f395a5105e6c54687656320ea478d7c04909/rignore-0.7.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:0714ee0759fe57068ea8dc234672f939b5d07dbc89d4e0b59468fdb88dbca830", size = 1072977, upload-time = "2025-10-02T13:25:25.24Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/c4/94ec3ee3e86051c5c09a196298f6d2fad9208bcd7bb0f228ef35118a92aa/rignore-0.7.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:44cda8590511c9a9aa32e253267f1742b41260f3b71d9bcabdb07911cca41a9a", size = 1130563, upload-time = "2025-10-02T13:25:41.694Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/a7/f975a509194b1daf02326d30a417b13939ef1faebbad260d5858de5aed2d/rignore-0.7.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:2f680188a2cfae1e64a863223e6a702c212d28b77fa3b9cc0547a4957b5845b0", size = 1109816, upload-time = "2025-10-02T13:25:58.105Z" },
+    { url = "https://files.pythonhosted.org/packages/59/db/99b695509ad1f9d067b0382b8f595ae547d163b2aea90b8948fa2ee45347/rignore-0.7.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:236003f4d348e4e2992aa7936d38aeffbaf17e43e3b068623096ee4b8f7efa2b", size = 1117778, upload-time = "2025-10-02T13:26:15.608Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/28/59aa850097283f3ae651e13ced4a8beadab8bab2f193a6e6d32d4235fc79/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:11831ac0c3bc656667848abd0cb869d288b13ad69a976cac307b447a4f79c9d3", size = 893706, upload-time = "2025-10-02T13:23:29.65Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/5e/0bf7c2101cd557374805372ae8a230ba83b4aa460c2f2f327580f79774f9/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:1f18bdc1bfd73da6a43bcf2c08f94e1ecddabf234e47e0f95daf6107cf937fb3", size = 867651, upload-time = "2025-10-02T13:23:47.193Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/ac/33080dba026b863a43e43a4c861278e51eb1b03c90f1a108f35a74b85d2d/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e6268a12ecb6d25caf0adb166732940bd9113e7dddd46018e457f1fb9408a707", size = 1168395, upload-time = "2025-10-02T13:24:03.882Z" },
+    { url = "https://files.pythonhosted.org/packages/83/b1/0c62eb8df324c00ef65c02c24dee30cc5a491ba224728916703761ee7c80/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7c1402445b24c8904b6cef124f2798d55a2ba84b41397d7fdab6fe316b1f20e6", size = 938744, upload-time = "2025-10-02T13:24:20.188Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/c3/965ab1d3674e790429a63d8486e2432fe120b29d4f4682a4171b3b3efac2/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:8aa1c3215cc9587e55b3326357637e17a2ed713ddb2c59339f908aae98ae01d6", size = 1074476, upload-time = "2025-10-02T13:25:26.864Z" },
+    { url = "https://files.pythonhosted.org/packages/36/2d/2673af40f46c2182c34d06e0662c035130008b47f74e4b5c72cddbbb50b6/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_armv7l.whl", hash = "sha256:78c8f56aaae18406699026e26f9c3b4721adc95f08ac4e76972aed0efb5eb91d", size = 1131270, upload-time = "2025-10-02T13:25:43.319Z" },
+    { url = "https://files.pythonhosted.org/packages/98/6c/7fcd680db36c2e34c0d5cceefd7a423772a9fbf25bb468b5748fedacda94/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_i686.whl", hash = "sha256:89ab6d73ffd48be27032def1decef83faebee891519e7f2006df5657b8ba2f4a", size = 1110257, upload-time = "2025-10-02T13:26:00.717Z" },
+    { url = "https://files.pythonhosted.org/packages/22/95/8ce27268d27fd12bc1b80d3e1840402f3ef3c205e788975d61c7a4077ef8/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:cf4eebeebeabd27467b51d5dbc92adc7177fcfd73a29c86f649853ba476d98fa", size = 1118831, upload-time = "2025-10-02T13:26:17.187Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/60/b02edbf5059f7947e375dc46583283aad579505e9e07775277e7fd6e04db/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f40142a34a7f08cd90fb4e74e43deffe3381fa3b164fb59857fa4e3996d4716d", size = 892600, upload-time = "2025-10-02T13:23:31.158Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/c5/3caa7732a91623110bc80c30f592efc6571a1c610b94f36083601ebf2392/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ccbc0b6285bb981316e5646ac96be7bca9665ee2444427d8d170fda5eda6f022", size = 866500, upload-time = "2025-10-02T13:23:49.099Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/66/943300886972b2dded2e0e851c1da1ad36565d40b5e55833b049cbf9285b/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:77cdf15a8b0ab80cd1d05a754b3237330e60e8731c255b7eb2a5d240a68df9f8", size = 1167255, upload-time = "2025-10-02T13:24:05.583Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/26/2f8cb5a546ce7056fe0fb8afbfc887431f9ba986cd7b4c65821dac13afa8/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:14e7e5ac99d60dd1993032205de7e79c36687825c45a7caa704620a0e9fde03f", size = 937991, upload-time = "2025-10-02T13:24:21.694Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/29/f97d581fc4d1013a42fe51154f820a7ccb97c679a2c2ea0c73072aa8935e/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98fae67456f053942ccda2cb2677a55fd34397e6674eaa403ab7c1c4930dcb12", size = 951972, upload-time = "2025-10-02T13:24:50.199Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/06/18da8ea8fc217fce872f81de23217c7ae011dd6e396dff026a262b499a4b/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2b55d2dcee6808f677ef25219ec0bb4852fbf2edb0b5010a5f18fe5feee276d6", size = 976002, upload-time = "2025-10-02T13:24:36.851Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/11/2f998fccb85a31f8dbd94b31123b48645067d4ca55b49c033987286475e7/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:7ff87634a648f17a9992ac4ce2fb48397696e3ab4a80154a895b9d1f6fc606cf", size = 1073180, upload-time = "2025-10-02T13:25:28.424Z" },
+    { url = "https://files.pythonhosted.org/packages/01/bf/ee6927f8dd8644f4c9c44d364380ab49629d259cc9611224512b161d7bef/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_armv7l.whl", hash = "sha256:c5721daa569fae74f5bf060165f96c6fec0a963ed008213e778259945406ec53", size = 1130056, upload-time = "2025-10-02T13:25:45.019Z" },
+    { url = "https://files.pythonhosted.org/packages/33/89/b231f432caced14303055c8611b34c5e2910c48b882de1c79eff4ce177d0/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_i686.whl", hash = "sha256:5770e783e08403b02c052b8b74a3e9431142aca93c78ccd1cc389b4dc60c2846", size = 1108603, upload-time = "2025-10-02T13:26:02.539Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/33/d331a0aea9e4a00ff530ad18421c46e213da1a608ad05463a2e5ae6cc572/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:504f66805fcc2a684cd1cda460d9f15b8b08997f06d9281efa221007072c53f5", size = 1117330, upload-time = "2025-10-02T13:26:18.741Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/cd/c5b39846a51d9dbc90ec6a0ac3502c9c827b4fcf2ce70b80ac9ca346cbb1/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:280fa78100c94640f36fe40304b5fdda1f14487f306f20b884f4dc230ad86e7b", size = 893459, upload-time = "2025-10-02T13:23:33.318Z" },
+    { url = "https://files.pythonhosted.org/packages/65/00/226cd6516bbf1b8a74fec65d2c1f19bc38866e41099d970e35f13984812f/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:3f9ba817b3ef8560c34d18764f32b7165027a8b82503a1f57ccfdc821e689d30", size = 867819, upload-time = "2025-10-02T13:23:50.609Z" },
+    { url = "https://files.pythonhosted.org/packages/28/18/07e6ff331a227e67bd2ccb9bfd823cb40434db26e96c406935a31af6f484/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c2150dbfb0227762a0af2decd7cc67a1c8b5cb4d35f8b83c674143fefc9b76f3", size = 1170040, upload-time = "2025-10-02T13:24:07.079Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/83/bcf95d4fcb8653c8556b4d50873830253ac46c34f005b135819979a41884/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:4e9af74b3b90266d7700b44a08217af636421b6e823f9280f9521a06e4fc18b9", size = 938346, upload-time = "2025-10-02T13:24:24.098Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/4b/0ba560f01a4cc22ecf5058311132b91e76dfef91f95a99072aae7948abf0/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:165890c27b1fd8ec71d47a05e6a11c6cf42693f3b479f3fc8d0047b5c2a46191", size = 1074376, upload-time = "2025-10-02T13:25:30.023Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/6f/3003ef871fac5f6ee7599695078c9d9165fee8171f9bf794cd5cddd4377f/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_armv7l.whl", hash = "sha256:3830a252b7e82ba61595489a819af721f645dd42548d5801000583044745a41c", size = 1131436, upload-time = "2025-10-02T13:25:46.557Z" },
+    { url = "https://files.pythonhosted.org/packages/02/56/a63bf6c3499a08713d491dba0543b47176f389c873e03d70b72bf0252216/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_i686.whl", hash = "sha256:d5a50ab60948bd7339ab50329991acfe7afb5fec4a444f0e19939ee6e18d364a", size = 1110413, upload-time = "2025-10-02T13:26:04.518Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/a9/678ac02cb63517335ce442507438c06d2e21940ae8baca3e4457de6426d1/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:648c052ba530ff43ecb516dd33831563ecac9878cd54897171237198870c48d4", size = 1118680, upload-time = "2025-10-02T13:26:20.907Z" },
+]
+
 [[package]]
 name = "rpds-py"
 version = "0.27.1"
@@ -3891,7 +4616,7 @@ wheels = [
 
 [[package]]
 name = "scipy"
-version = "1.16.1"
+version = "1.16.2"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -3900,26 +4625,28 @@ resolution-markers = [
 dependencies = [
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f5/4a/b927028464795439faec8eaf0b03b011005c487bb2d07409f28bf30879c4/scipy-1.16.1.tar.gz", hash = "sha256:44c76f9e8b6e8e488a586190ab38016e4ed2f8a038af7cd3defa903c0a2238b3", size = 30580861, upload-time = "2025-07-27T16:33:30.834Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/da/91/812adc6f74409b461e3a5fa97f4f74c769016919203138a3bf6fc24ba4c5/scipy-1.16.1-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:c033fa32bab91dc98ca59d0cf23bb876454e2bb02cbe592d5023138778f70030", size = 36552519, upload-time = "2025-07-27T16:26:29.658Z" },
-    { url = "https://files.pythonhosted.org/packages/47/18/8e355edcf3b71418d9e9f9acd2708cc3a6c27e8f98fde0ac34b8a0b45407/scipy-1.16.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:6e5c2f74e5df33479b5cd4e97a9104c511518fbd979aa9b8f6aec18b2e9ecae7", size = 28638010, upload-time = "2025-07-27T16:26:38.196Z" },
-    { url = "https://files.pythonhosted.org/packages/d9/eb/e931853058607bdfbc11b86df19ae7a08686121c203483f62f1ecae5989c/scipy-1.16.1-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:0a55ffe0ba0f59666e90951971a884d1ff6f4ec3275a48f472cfb64175570f77", size = 20909790, upload-time = "2025-07-27T16:26:43.93Z" },
-    { url = "https://files.pythonhosted.org/packages/45/0c/be83a271d6e96750cd0be2e000f35ff18880a46f05ce8b5d3465dc0f7a2a/scipy-1.16.1-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:f8a5d6cd147acecc2603fbd382fed6c46f474cccfcf69ea32582e033fb54dcfe", size = 23513352, upload-time = "2025-07-27T16:26:50.017Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/bf/fe6eb47e74f762f933cca962db7f2c7183acfdc4483bd1c3813cfe83e538/scipy-1.16.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:cb18899127278058bcc09e7b9966d41a5a43740b5bb8dcba401bd983f82e885b", size = 33534643, upload-time = "2025-07-27T16:26:57.503Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/ba/63f402e74875486b87ec6506a4f93f6d8a0d94d10467280f3d9d7837ce3a/scipy-1.16.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:adccd93a2fa937a27aae826d33e3bfa5edf9aa672376a4852d23a7cd67a2e5b7", size = 35376776, upload-time = "2025-07-27T16:27:06.639Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/b4/04eb9d39ec26a1b939689102da23d505ea16cdae3dbb18ffc53d1f831044/scipy-1.16.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:18aca1646a29ee9a0625a1be5637fa798d4d81fdf426481f06d69af828f16958", size = 35698906, upload-time = "2025-07-27T16:27:14.943Z" },
-    { url = "https://files.pythonhosted.org/packages/04/d6/bb5468da53321baeb001f6e4e0d9049eadd175a4a497709939128556e3ec/scipy-1.16.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:d85495cef541729a70cdddbbf3e6b903421bc1af3e8e3a9a72a06751f33b7c39", size = 38129275, upload-time = "2025-07-27T16:27:23.873Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/94/994369978509f227cba7dfb9e623254d0d5559506fe994aef4bea3ed469c/scipy-1.16.1-cp311-cp311-win_amd64.whl", hash = "sha256:226652fca853008119c03a8ce71ffe1b3f6d2844cc1686e8f9806edafae68596", size = 38644572, upload-time = "2025-07-27T16:27:32.637Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/d9/ec4864f5896232133f51382b54a08de91a9d1af7a76dfa372894026dfee2/scipy-1.16.1-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:81b433bbeaf35728dad619afc002db9b189e45eebe2cd676effe1fb93fef2b9c", size = 36575194, upload-time = "2025-07-27T16:27:41.321Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/6d/40e81ecfb688e9d25d34a847dca361982a6addf8e31f0957b1a54fbfa994/scipy-1.16.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:886cc81fdb4c6903a3bb0464047c25a6d1016fef77bb97949817d0c0d79f9e04", size = 28594590, upload-time = "2025-07-27T16:27:49.204Z" },
-    { url = "https://files.pythonhosted.org/packages/0e/37/9f65178edfcc629377ce9a64fc09baebea18c80a9e57ae09a52edf84880b/scipy-1.16.1-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:15240c3aac087a522b4eaedb09f0ad061753c5eebf1ea430859e5bf8640d5919", size = 20866458, upload-time = "2025-07-27T16:27:54.98Z" },
-    { url = "https://files.pythonhosted.org/packages/2c/7b/749a66766871ea4cb1d1ea10f27004db63023074c22abed51f22f09770e0/scipy-1.16.1-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:65f81a25805f3659b48126b5053d9e823d3215e4a63730b5e1671852a1705921", size = 23539318, upload-time = "2025-07-27T16:28:01.604Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/db/8d4afec60eb833a666434d4541a3151eedbf2494ea6d4d468cbe877f00cd/scipy-1.16.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:6c62eea7f607f122069b9bad3f99489ddca1a5173bef8a0c75555d7488b6f725", size = 33292899, upload-time = "2025-07-27T16:28:09.147Z" },
-    { url = "https://files.pythonhosted.org/packages/51/1e/79023ca3bbb13a015d7d2757ecca3b81293c663694c35d6541b4dca53e98/scipy-1.16.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f965bbf3235b01c776115ab18f092a95aa74c271a52577bcb0563e85738fd618", size = 35162637, upload-time = "2025-07-27T16:28:17.535Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/49/0648665f9c29fdaca4c679182eb972935b3b4f5ace41d323c32352f29816/scipy-1.16.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:f006e323874ffd0b0b816d8c6a8e7f9a73d55ab3b8c3f72b752b226d0e3ac83d", size = 35490507, upload-time = "2025-07-27T16:28:25.705Z" },
-    { url = "https://files.pythonhosted.org/packages/62/8f/66cbb9d6bbb18d8c658f774904f42a92078707a7c71e5347e8bf2f52bb89/scipy-1.16.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:e8fd15fc5085ab4cca74cb91fe0a4263b1f32e4420761ddae531ad60934c2119", size = 37923998, upload-time = "2025-07-27T16:28:34.339Z" },
-    { url = "https://files.pythonhosted.org/packages/14/c3/61f273ae550fbf1667675701112e380881905e28448c080b23b5a181df7c/scipy-1.16.1-cp312-cp312-win_amd64.whl", hash = "sha256:f7b8013c6c066609577d910d1a2a077021727af07b6fab0ee22c2f901f22352a", size = 38508060, upload-time = "2025-07-27T16:28:43.242Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/4c/3b/546a6f0bfe791bbb7f8d591613454d15097e53f906308ec6f7c1ce588e8e/scipy-1.16.2.tar.gz", hash = "sha256:af029b153d243a80afb6eabe40b0a07f8e35c9adc269c019f364ad747f826a6b", size = 30580599, upload-time = "2025-09-11T17:48:08.271Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0b/ef/37ed4b213d64b48422df92560af7300e10fe30b5d665dd79932baebee0c6/scipy-1.16.2-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:6ab88ea43a57da1af33292ebd04b417e8e2eaf9d5aa05700be8d6e1b6501cd92", size = 36619956, upload-time = "2025-09-11T17:39:20.5Z" },
+    { url = "https://files.pythonhosted.org/packages/85/ab/5c2eba89b9416961a982346a4d6a647d78c91ec96ab94ed522b3b6baf444/scipy-1.16.2-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:c95e96c7305c96ede73a7389f46ccd6c659c4da5ef1b2789466baeaed3622b6e", size = 28931117, upload-time = "2025-09-11T17:39:29.06Z" },
+    { url = "https://files.pythonhosted.org/packages/80/d1/eed51ab64d227fe60229a2d57fb60ca5898cfa50ba27d4f573e9e5f0b430/scipy-1.16.2-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:87eb178db04ece7c698220d523c170125dbffebb7af0345e66c3554f6f60c173", size = 20921997, upload-time = "2025-09-11T17:39:34.892Z" },
+    { url = "https://files.pythonhosted.org/packages/be/7c/33ea3e23bbadde96726edba6bf9111fb1969d14d9d477ffa202c67bec9da/scipy-1.16.2-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:4e409eac067dcee96a57fbcf424c13f428037827ec7ee3cb671ff525ca4fc34d", size = 23523374, upload-time = "2025-09-11T17:39:40.846Z" },
+    { url = "https://files.pythonhosted.org/packages/96/0b/7399dc96e1e3f9a05e258c98d716196a34f528eef2ec55aad651ed136d03/scipy-1.16.2-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:e574be127bb760f0dad24ff6e217c80213d153058372362ccb9555a10fc5e8d2", size = 33583702, upload-time = "2025-09-11T17:39:49.011Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/bc/a5c75095089b96ea72c1bd37a4497c24b581ec73db4ef58ebee142ad2d14/scipy-1.16.2-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f5db5ba6188d698ba7abab982ad6973265b74bb40a1efe1821b58c87f73892b9", size = 35883427, upload-time = "2025-09-11T17:39:57.406Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/66/e25705ca3d2b87b97fe0a278a24b7f477b4023a926847935a1a71488a6a6/scipy-1.16.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ec6e74c4e884104ae006d34110677bfe0098203a3fec2f3faf349f4cb05165e3", size = 36212940, upload-time = "2025-09-11T17:40:06.013Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/fd/0bb911585e12f3abdd603d721d83fc1c7492835e1401a0e6d498d7822b4b/scipy-1.16.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:912f46667d2d3834bc3d57361f854226475f695eb08c08a904aadb1c936b6a88", size = 38865092, upload-time = "2025-09-11T17:40:15.143Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/73/c449a7d56ba6e6f874183759f8483cde21f900a8be117d67ffbb670c2958/scipy-1.16.2-cp311-cp311-win_amd64.whl", hash = "sha256:91e9e8a37befa5a69e9cacbe0bcb79ae5afb4a0b130fd6db6ee6cc0d491695fa", size = 38687626, upload-time = "2025-09-11T17:40:24.041Z" },
+    { url = "https://files.pythonhosted.org/packages/68/72/02f37316adf95307f5d9e579023c6899f89ff3a051fa079dbd6faafc48e5/scipy-1.16.2-cp311-cp311-win_arm64.whl", hash = "sha256:f3bf75a6dcecab62afde4d1f973f1692be013110cad5338007927db8da73249c", size = 25503506, upload-time = "2025-09-11T17:40:30.703Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/8d/6396e00db1282279a4ddd507c5f5e11f606812b608ee58517ce8abbf883f/scipy-1.16.2-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:89d6c100fa5c48472047632e06f0876b3c4931aac1f4291afc81a3644316bb0d", size = 36646259, upload-time = "2025-09-11T17:40:39.329Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/93/ea9edd7e193fceb8eef149804491890bde73fb169c896b61aa3e2d1e4e77/scipy-1.16.2-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:ca748936cd579d3f01928b30a17dc474550b01272d8046e3e1ee593f23620371", size = 28888976, upload-time = "2025-09-11T17:40:46.82Z" },
+    { url = "https://files.pythonhosted.org/packages/91/4d/281fddc3d80fd738ba86fd3aed9202331180b01e2c78eaae0642f22f7e83/scipy-1.16.2-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:fac4f8ce2ddb40e2e3d0f7ec36d2a1e7f92559a2471e59aec37bd8d9de01fec0", size = 20879905, upload-time = "2025-09-11T17:40:52.545Z" },
+    { url = "https://files.pythonhosted.org/packages/69/40/b33b74c84606fd301b2915f0062e45733c6ff5708d121dd0deaa8871e2d0/scipy-1.16.2-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:033570f1dcefd79547a88e18bccacff025c8c647a330381064f561d43b821232", size = 23553066, upload-time = "2025-09-11T17:40:59.014Z" },
+    { url = "https://files.pythonhosted.org/packages/55/a7/22c739e2f21a42cc8f16bc76b47cff4ed54fbe0962832c589591c2abec34/scipy-1.16.2-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:ea3421209bf00c8a5ef2227de496601087d8f638a2363ee09af059bd70976dc1", size = 33336407, upload-time = "2025-09-11T17:41:06.796Z" },
+    { url = "https://files.pythonhosted.org/packages/53/11/a0160990b82999b45874dc60c0c183d3a3a969a563fffc476d5a9995c407/scipy-1.16.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f66bd07ba6f84cd4a380b41d1bf3c59ea488b590a2ff96744845163309ee8e2f", size = 35673281, upload-time = "2025-09-11T17:41:15.055Z" },
+    { url = "https://files.pythonhosted.org/packages/96/53/7ef48a4cfcf243c3d0f1643f5887c81f29fdf76911c4e49331828e19fc0a/scipy-1.16.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5e9feab931bd2aea4a23388c962df6468af3d808ddf2d40f94a81c5dc38f32ef", size = 36004222, upload-time = "2025-09-11T17:41:23.868Z" },
+    { url = "https://files.pythonhosted.org/packages/49/7f/71a69e0afd460049d41c65c630c919c537815277dfea214031005f474d78/scipy-1.16.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:03dfc75e52f72cf23ec2ced468645321407faad8f0fe7b1f5b49264adbc29cb1", size = 38664586, upload-time = "2025-09-11T17:41:31.021Z" },
+    { url = "https://files.pythonhosted.org/packages/34/95/20e02ca66fb495a95fba0642fd48e0c390d0ece9b9b14c6e931a60a12dea/scipy-1.16.2-cp312-cp312-win_amd64.whl", hash = "sha256:0ce54e07bbb394b417457409a64fd015be623f36e330ac49306433ffe04bc97e", size = 38550641, upload-time = "2025-09-11T17:41:36.61Z" },
+    { url = "https://files.pythonhosted.org/packages/92/ad/13646b9beb0a95528ca46d52b7babafbe115017814a611f2065ee4e61d20/scipy-1.16.2-cp312-cp312-win_arm64.whl", hash = "sha256:2a8ffaa4ac0df81a0b94577b18ee079f13fecdb924df3328fc44a7dc5ac46851", size = 25456070, upload-time = "2025-09-11T17:41:41.3Z" },
 ]
 
 [[package]]
@@ -3962,83 +4689,130 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/8d/21/55267fcdc1a99612248c299c12d8f7c51d6e5ae992f32694d17adcf13fda/sentencepiece-0.2.1-cp39-cp39-win_arm64.whl", hash = "sha256:3d165fbb9bf8fba35f1946ba2617c3f9995679f07438325f07c026d53f33e746", size = 1033914, upload-time = "2025-08-12T06:59:10.705Z" },
 ]
 
+[[package]]
+name = "sentry-sdk"
+version = "2.40.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "certifi" },
+    { name = "urllib3" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4f/b5/ce879ce3292e5ca41fa3ebf68f60645032eca813c9ed8f92dcf09804c0e3/sentry_sdk-2.40.0.tar.gz", hash = "sha256:b9c4672fb2cafabcc28586ab8fd0ceeff9b2352afcf2b936e13d5ba06d141b9f", size = 351703, upload-time = "2025-10-06T12:27:29.207Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a4/d1/a54bd3622c6e742e6a01bc3bac45966b7ba886e29827da6b8ca7ae234e21/sentry_sdk-2.40.0-py2.py3-none-any.whl", hash = "sha256:d5f6ae0f27ea73e7b09c70ad7d42242326eb44765e87a15d8c5aab96b80013e6", size = 374747, upload-time = "2025-10-06T12:27:27.051Z" },
+]
+
+[[package]]
+name = "setproctitle"
+version = "1.3.7"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/8d/48/49393a96a2eef1ab418b17475fb92b8fcfad83d099e678751b05472e69de/setproctitle-1.3.7.tar.gz", hash = "sha256:bc2bc917691c1537d5b9bca1468437176809c7e11e5694ca79a9ca12345dcb9e", size = 27002, upload-time = "2025-09-05T12:51:25.278Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0e/f8/17bda581c517678260e6541b600eeb67745f53596dc077174141ba2f6702/setproctitle-1.3.7-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:00afa6fc507967d8c9d592a887cdc6c1f5742ceac6a4354d111ca0214847732c", size = 31793, upload-time = "2025-09-05T12:49:10.297Z" },
+    { url = "https://files.pythonhosted.org/packages/27/d1/76a33ae80d4e788ecab9eb9b53db03e81cfc95367ec7e3fbf4989962fedd/setproctitle-1.3.7-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9e02667f6b9fc1238ba753c0f4b0a37ae184ce8f3bbbc38e115d99646b3f4cd3", size = 32779, upload-time = "2025-09-05T12:49:12.157Z" },
+    { url = "https://files.pythonhosted.org/packages/59/27/1a07c38121967061564f5e0884414a5ab11a783260450172d4fc68c15621/setproctitle-1.3.7-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:83fcd271567d133eb9532d3b067c8a75be175b2b3b271e2812921a05303a693f", size = 34578, upload-time = "2025-09-05T12:49:13.393Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/d4/725e6353935962d8bb12cbf7e7abba1d0d738c7f6935f90239d8e1ccf913/setproctitle-1.3.7-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:13fe37951dda1a45c35d77d06e3da5d90e4f875c4918a7312b3b4556cfa7ff64", size = 32030, upload-time = "2025-09-05T12:49:15.362Z" },
+    { url = "https://files.pythonhosted.org/packages/67/24/e4677ae8e1cb0d549ab558b12db10c175a889be0974c589c428fece5433e/setproctitle-1.3.7-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:a05509cfb2059e5d2ddff701d38e474169e9ce2a298cf1b6fd5f3a213a553fe5", size = 33363, upload-time = "2025-09-05T12:49:16.829Z" },
+    { url = "https://files.pythonhosted.org/packages/55/d4/69ce66e4373a48fdbb37489f3ded476bb393e27f514968c3a69a67343ae0/setproctitle-1.3.7-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:6da835e76ae18574859224a75db6e15c4c2aaa66d300a57efeaa4c97ca4c7381", size = 31508, upload-time = "2025-09-05T12:49:18.032Z" },
+    { url = "https://files.pythonhosted.org/packages/18/2e/bd03ff02432a181c1787f6fc2a678f53b7dacdd5ded69c318fe1619556e8/setproctitle-1.3.7-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:1607b963e7b53e24ec8a2cb4e0ab3ae591d7c6bf0a160feef0551da63452b37f", size = 32191, upload-time = "2025-09-05T12:49:24.567Z" },
+    { url = "https://files.pythonhosted.org/packages/28/78/1e62fc0937a8549f2220445ed2175daacee9b6764c7963b16148119b016d/setproctitle-1.3.7-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a20fb1a3974e2dab857870cf874b325b8705605cb7e7e8bcbb915bca896f52a9", size = 33203, upload-time = "2025-09-05T12:49:25.871Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/3c/65edc65db3fa3df400cf13b05e9d41a3c77517b4839ce873aa6b4043184f/setproctitle-1.3.7-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f8d961bba676e07d77665204f36cffaa260f526e7b32d07ab3df6a2c1dfb44ba", size = 34963, upload-time = "2025-09-05T12:49:27.044Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/32/89157e3de997973e306e44152522385f428e16f92f3cf113461489e1e2ee/setproctitle-1.3.7-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:db0fd964fbd3a9f8999b502f65bd2e20883fdb5b1fae3a424e66db9a793ed307", size = 32398, upload-time = "2025-09-05T12:49:28.909Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/18/77a765a339ddf046844cb4513353d8e9dcd8183da9cdba6e078713e6b0b2/setproctitle-1.3.7-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:db116850fcf7cca19492030f8d3b4b6e231278e8fe097a043957d22ce1bdf3ee", size = 33657, upload-time = "2025-09-05T12:49:30.323Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/63/f0b6205c64d74d2a24a58644a38ec77bdbaa6afc13747e75973bf8904932/setproctitle-1.3.7-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:316664d8b24a5c91ee244460bdaf7a74a707adaa9e14fbe0dc0a53168bb9aba1", size = 31836, upload-time = "2025-09-05T12:49:32.309Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/99/71630546b9395b095f4082be41165d1078204d1696c2d9baade3de3202d0/setproctitle-1.3.7-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:2906b6c7959cdb75f46159bf0acd8cc9906cf1361c9e1ded0d065fe8f9039629", size = 32932, upload-time = "2025-09-05T12:49:39.271Z" },
+    { url = "https://files.pythonhosted.org/packages/50/22/cee06af4ffcfb0e8aba047bd44f5262e644199ae7527ae2c1f672b86495c/setproctitle-1.3.7-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6915964a6dda07920a1159321dcd6d94fc7fc526f815ca08a8063aeca3c204f1", size = 33736, upload-time = "2025-09-05T12:49:40.565Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/00/a5949a8bb06ef5e7df214fc393bb2fb6aedf0479b17214e57750dfdd0f24/setproctitle-1.3.7-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:cff72899861c765bd4021d1ff1c68d60edc129711a2fdba77f9cb69ef726a8b6", size = 35605, upload-time = "2025-09-05T12:49:42.362Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/3a/50caca532a9343828e3bf5778c7a84d6c737a249b1796d50dd680290594d/setproctitle-1.3.7-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:b7cb05bd446687ff816a3aaaf831047fc4c364feff7ada94a66024f1367b448c", size = 33143, upload-time = "2025-09-05T12:49:43.515Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/14/b843a251296ce55e2e17c017d6b9f11ce0d3d070e9265de4ecad948b913d/setproctitle-1.3.7-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:3a57b9a00de8cae7e2a1f7b9f0c2ac7b69372159e16a7708aa2f38f9e5cc987a", size = 34434, upload-time = "2025-09-05T12:49:45.31Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/b7/06145c238c0a6d2c4bc881f8be230bb9f36d2bf51aff7bddcb796d5eed67/setproctitle-1.3.7-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:d8828b356114f6b308b04afe398ed93803d7fca4a955dd3abe84430e28d33739", size = 32795, upload-time = "2025-09-05T12:49:46.419Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/b7/1abfcfe0de1b50c96002e7ecfd6a2c109798bc4284f0dc02635c7da22e9e/setproctitle-1.3.7-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:134e7f66703a1d92c0a9a0a417c580f2cc04b93d31d3fc0dd43c3aa194b706e1", size = 31551, upload-time = "2025-09-05T12:51:05.088Z" },
+    { url = "https://files.pythonhosted.org/packages/67/fa/2b25384c154bfb36112d7ec377281776e337e5124590bd96c4f3f076285a/setproctitle-1.3.7-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9796732a040f617fc933f9531c9a84bb73c5c27b8074abbe52907076e804b2b7", size = 32489, upload-time = "2025-09-05T12:51:06.207Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/24/e4f2a3467cd6322462134f1918a565b99e8bd43c1d47040082e874a3f512/setproctitle-1.3.7-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ff3c1c32382fb71a200db8bab3df22f32e6ac7ec3170e92fa5b542cf42eed9a2", size = 34328, upload-time = "2025-09-05T12:51:07.504Z" },
+    { url = "https://files.pythonhosted.org/packages/33/75/7ca8f54e35c62f99562fcf1e052d4f162f310e2e14ab3e34d5965b5a7f71/setproctitle-1.3.7-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:01f27b5b72505b304152cb0bd7ff410cc4f2d69ac70c21a7fdfa64400a68642d", size = 31744, upload-time = "2025-09-05T12:51:08.698Z" },
+    { url = "https://files.pythonhosted.org/packages/19/74/080caa13f1dad1ffab1ef7b7dbef97f278c23d4aa45bdbd209aac54e9b21/setproctitle-1.3.7-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:80b6a562cbc92b289c28f34ce709a16b26b1696e9b9a0542a675ce3a788bdf3f", size = 33100, upload-time = "2025-09-05T12:51:10.53Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/bc/478b45ccbcf4be609d4ae5b46b4601960d7a7194cca2075cfae57b0cbb76/setproctitle-1.3.7-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c4fb90174d176473122e7eef7c6492d53761826f34ff61c81a1c1d66905025d3", size = 31233, upload-time = "2025-09-05T12:51:12.03Z" },
+    { url = "https://files.pythonhosted.org/packages/41/89/5b6f2faedd6ced3d3c085a5efbd91380fb1f61f4c12bc42acad37932f4e9/setproctitle-1.3.7-pp310-pypy310_pp73-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:502b902a0e4c69031b87870ff4986c290ebbb12d6038a70639f09c331b18efb2", size = 14284, upload-time = "2025-09-05T12:51:18.393Z" },
+    { url = "https://files.pythonhosted.org/packages/73/02/b9eadc226195dcfa90eed37afe56b5dd6fa2f0e5220ab8b7867b8862b926/setproctitle-1.3.7-pp311-pypy311_pp73-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f1704c9e041f2b1dc38f5be4552e141e1432fba3dd52c72eeffd5bc2db04dc65", size = 14286, upload-time = "2025-09-05T12:51:22.61Z" },
+]
+
 [[package]]
 name = "setuptools"
-version = "80.9.0"
+version = "79.0.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/18/5d/3bf57dcd21979b887f014ea83c24ae194cfcd12b9e0fda66b957c69d1fca/setuptools-80.9.0.tar.gz", hash = "sha256:f36b47402ecde768dbfafc46e8e4207b4360c654f1f3bb84475f0a28628fb19c", size = 1319958, upload-time = "2025-05-27T00:56:51.443Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/bb/71/b6365e6325b3290e14957b2c3a804a529968c77a049b2ed40c095f749707/setuptools-79.0.1.tar.gz", hash = "sha256:128ce7b8f33c3079fd1b067ecbb4051a66e8526e7b65f6cec075dfc650ddfa88", size = 1367909, upload-time = "2025-04-23T22:20:59.241Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a3/dc/17031897dae0efacfea57dfd3a82fdd2a2aeb58e0ff71b77b87e44edc772/setuptools-80.9.0-py3-none-any.whl", hash = "sha256:062d34222ad13e0cc312a4c02d73f059e86a4acbfbdea8f8f76b28c99f306922", size = 1201486, upload-time = "2025-05-27T00:56:49.664Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/6d/b4752b044bf94cb802d88a888dc7d288baaf77d7910b7dedda74b5ceea0c/setuptools-79.0.1-py3-none-any.whl", hash = "sha256:e147c0549f27767ba362f9da434eab9c5dc0045d5304feb602a0af001089fc51", size = 1256281, upload-time = "2025-04-23T22:20:56.768Z" },
 ]
 
 [[package]]
-name = "shortuuid"
-version = "1.0.13"
+name = "shellingham"
+version = "1.5.4"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/8c/e2/bcf761f3bff95856203f9559baf3741c416071dd200c0fc19fad7f078f86/shortuuid-1.0.13.tar.gz", hash = "sha256:3bb9cf07f606260584b1df46399c0b87dd84773e7b25912b7e391e30797c5e72", size = 9662, upload-time = "2024-03-11T20:11:06.879Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/58/15/8b3609fd3830ef7b27b655beb4b4e9c62313a4e8da8c676e142cc210d58e/shellingham-1.5.4.tar.gz", hash = "sha256:8dbca0739d487e5bd35ab3ca4b36e11c4078f3a234bfce294b0a0291363404de", size = 10310, upload-time = "2023-10-24T04:13:40.426Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c0/44/21d6bf170bf40b41396480d8d49ad640bca3f2b02139cd52aa1e272830a5/shortuuid-1.0.13-py3-none-any.whl", hash = "sha256:a482a497300b49b4953e15108a7913244e1bb0d41f9d332f5e9925dba33a3c5a", size = 10529, upload-time = "2024-03-11T20:11:04.807Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/f9/0595336914c5619e5f28a1fb793285925a8cd4b432c9da0a987836c7f822/shellingham-1.5.4-py2.py3-none-any.whl", hash = "sha256:7ecfff8f2fd72616f7481040475a65b2bf8af90a56c89140852d1120324e8686", size = 9755, upload-time = "2023-10-24T04:13:38.866Z" },
 ]
 
 [[package]]
 name = "simplejson"
-version = "3.20.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/af/92/51b417685abd96b31308b61b9acce7ec50d8e1de8fbc39a7fd4962c60689/simplejson-3.20.1.tar.gz", hash = "sha256:e64139b4ec4f1f24c142ff7dcafe55a22b811a74d86d66560c8815687143037d", size = 85591, upload-time = "2025-02-15T05:18:53.15Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/25/c4/627214fb418cd4a17fb0230ff0b6c3bb4a85cbb48dd69c85dcc3b85df828/simplejson-3.20.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e580aa65d5f6c3bf41b9b4afe74be5d5ddba9576701c107c772d936ea2b5043a", size = 93790, upload-time = "2025-02-15T05:15:32.954Z" },
-    { url = "https://files.pythonhosted.org/packages/15/ca/56a6a2a33cbcf330c4d71af3f827c47e4e0ba791e78f2642f3d1ab02ff31/simplejson-3.20.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:4a586ce4f78cec11f22fe55c5bee0f067e803aab9bad3441afe2181693b5ebb5", size = 75707, upload-time = "2025-02-15T05:15:34.954Z" },
-    { url = "https://files.pythonhosted.org/packages/a9/c8/3d92b67e03a3b6207d97202669f9454ed700b35ade9bd4428265a078fb6c/simplejson-3.20.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:74a1608f9e6e8c27a4008d70a54270868306d80ed48c9df7872f9f4b8ac87808", size = 75700, upload-time = "2025-02-15T05:15:37.144Z" },
-    { url = "https://files.pythonhosted.org/packages/74/30/20001219d6fdca4aaa3974c96dfb6955a766b4e2cc950505a5b51fd050b0/simplejson-3.20.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:03db8cb64154189a92a7786209f24e391644f3a3fa335658be2df2af1960b8d8", size = 138672, upload-time = "2025-02-15T05:15:38.547Z" },
-    { url = "https://files.pythonhosted.org/packages/21/47/50157810876c2a7ebbd6e6346ec25eda841fe061fecaa02538a7742a3d2a/simplejson-3.20.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:eea7e2b7d858f6fdfbf0fe3cb846d6bd8a45446865bc09960e51f3d473c2271b", size = 146616, upload-time = "2025-02-15T05:15:39.871Z" },
-    { url = "https://files.pythonhosted.org/packages/95/60/8c97cdc93096437b0aca2745aca63c880fe2315fd7f6a6ce6edbb344a2ae/simplejson-3.20.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e66712b17d8425bb7ff8968d4c7c7fd5a2dd7bd63728b28356223c000dd2f91f", size = 134344, upload-time = "2025-02-15T05:15:42.091Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/9e/da184f0e9bb3a5d7ffcde713bd41b4fe46cca56b6f24d9bd155fac56805a/simplejson-3.20.1-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a2cc4f6486f9f515b62f5831ff1888886619b84fc837de68f26d919ba7bbdcbc", size = 138017, upload-time = "2025-02-15T05:15:43.542Z" },
-    { url = "https://files.pythonhosted.org/packages/31/db/00d1a8d9b036db98f678c8a3c69ed17d2894d1768d7a00576e787ad3e546/simplejson-3.20.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:a3c2df555ee4016148fa192e2b9cd9e60bc1d40769366134882685e90aee2a1e", size = 140118, upload-time = "2025-02-15T05:15:45.7Z" },
-    { url = "https://files.pythonhosted.org/packages/52/21/57fc47eab8c1c73390b933a5ba9271f08e3e1ec83162c580357f28f5b97c/simplejson-3.20.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:78520f04b7548a5e476b5396c0847e066f1e0a4c0c5e920da1ad65e95f410b11", size = 140314, upload-time = "2025-02-15T05:16:07.949Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/cc/7cfd78d1e0fa5e57350b98cfe77353b6dfa13dce21afa4060e1019223852/simplejson-3.20.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:f4bd49ecde87b0fe9f55cc971449a32832bca9910821f7072bbfae1155eaa007", size = 148544, upload-time = "2025-02-15T05:16:09.455Z" },
-    { url = "https://files.pythonhosted.org/packages/63/26/1c894a1c2bd95dc8be0cf5a2fa73b0d173105b6ca18c90cb981ff10443d0/simplejson-3.20.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:7eaae2b88eb5da53caaffdfa50e2e12022553949b88c0df4f9a9663609373f72", size = 141172, upload-time = "2025-02-15T05:16:10.966Z" },
-    { url = "https://files.pythonhosted.org/packages/93/27/0717dccc10cd9988dbf1314def52ab32678a95a95328bb37cafacf499400/simplejson-3.20.1-cp310-cp310-win32.whl", hash = "sha256:e836fb88902799eac8debc2b642300748f4860a197fa3d9ea502112b6bb8e142", size = 74181, upload-time = "2025-02-15T05:16:12.361Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/af/593f896573f306519332d4287b1ab8b7b888c239bbd5159f7054d7055c2d/simplejson-3.20.1-cp310-cp310-win_amd64.whl", hash = "sha256:b122a19b552b212fc3b5b96fc5ce92333d4a9ac0a800803e1f17ebb16dac4be5", size = 75738, upload-time = "2025-02-15T05:16:14.438Z" },
-    { url = "https://files.pythonhosted.org/packages/76/59/74bc90d1c051bc2432c96b34bd4e8036875ab58b4fcbe4d6a5a76985f853/simplejson-3.20.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:325b8c107253d3217e89d7b50c71015b5b31e2433e6c5bf38967b2f80630a8ca", size = 92132, upload-time = "2025-02-15T05:16:15.743Z" },
-    { url = "https://files.pythonhosted.org/packages/71/c7/1970916e0c51794fff89f76da2f632aaf0b259b87753c88a8c409623d3e1/simplejson-3.20.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:88a7baa8211089b9e58d78fbc1b0b322103f3f3d459ff16f03a36cece0d0fcf0", size = 74956, upload-time = "2025-02-15T05:16:17.062Z" },
-    { url = "https://files.pythonhosted.org/packages/c8/0d/98cc5909180463f1d75fac7180de62d4cdb4e82c4fef276b9e591979372c/simplejson-3.20.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:299b1007b8101d50d95bc0db1bf5c38dc372e85b504cf77f596462083ee77e3f", size = 74772, upload-time = "2025-02-15T05:16:19.204Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/94/a30a5211a90d67725a3e8fcc1c788189f2ae2ed2b96b63ed15d0b7f5d6bb/simplejson-3.20.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:03ec618ed65caab48e81e3ed29586236a8e57daef792f1f3bb59504a7e98cd10", size = 143575, upload-time = "2025-02-15T05:16:21.337Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/08/cdb6821f1058eb5db46d252de69ff7e6c53f05f1bae6368fe20d5b51d37e/simplejson-3.20.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cd2cdead1d3197f0ff43373cf4730213420523ba48697743e135e26f3d179f38", size = 153241, upload-time = "2025-02-15T05:16:22.859Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/2d/ca3caeea0bdc5efc5503d5f57a2dfb56804898fb196dfada121323ee0ccb/simplejson-3.20.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3466d2839fdc83e1af42e07b90bc8ff361c4e8796cd66722a40ba14e458faddd", size = 141500, upload-time = "2025-02-15T05:16:25.068Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/33/d3e0779d5c58245e7370c98eb969275af6b7a4a5aec3b97cbf85f09ad328/simplejson-3.20.1-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d492ed8e92f3a9f9be829205f44b1d0a89af6582f0cf43e0d129fa477b93fe0c", size = 144757, upload-time = "2025-02-15T05:16:28.301Z" },
-    { url = "https://files.pythonhosted.org/packages/54/53/2d93128bb55861b2fa36c5944f38da51a0bc6d83e513afc6f7838440dd15/simplejson-3.20.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:f924b485537b640dc69434565463fd6fc0c68c65a8c6e01a823dd26c9983cf79", size = 144409, upload-time = "2025-02-15T05:16:29.687Z" },
-    { url = "https://files.pythonhosted.org/packages/99/4c/dac310a98f897ad3435b4bdc836d92e78f09e38c5dbf28211ed21dc59fa2/simplejson-3.20.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:9e8eacf6a3491bf76ea91a8d46726368a6be0eb94993f60b8583550baae9439e", size = 146082, upload-time = "2025-02-15T05:16:31.064Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/22/d7ba958cfed39827335b82656b1c46f89678faecda9a7677b47e87b48ee6/simplejson-3.20.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:d34d04bf90b4cea7c22d8b19091633908f14a096caa301b24c2f3d85b5068fb8", size = 154339, upload-time = "2025-02-15T05:16:32.719Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/c8/b072b741129406a7086a0799c6f5d13096231bf35fdd87a0cffa789687fc/simplejson-3.20.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:69dd28d4ce38390ea4aaf212902712c0fd1093dc4c1ff67e09687c3c3e15a749", size = 147915, upload-time = "2025-02-15T05:16:34.291Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/46/8347e61e9cf3db5342a42f7fd30a81b4f5cf85977f916852d7674a540907/simplejson-3.20.1-cp311-cp311-win32.whl", hash = "sha256:dfe7a9da5fd2a3499436cd350f31539e0a6ded5da6b5b3d422df016444d65e43", size = 73972, upload-time = "2025-02-15T05:16:35.712Z" },
-    { url = "https://files.pythonhosted.org/packages/01/85/b52f24859237b4e9d523d5655796d911ba3d46e242eb1959c45b6af5aedd/simplejson-3.20.1-cp311-cp311-win_amd64.whl", hash = "sha256:896a6c04d7861d507d800da7642479c3547060bf97419d9ef73d98ced8258766", size = 75595, upload-time = "2025-02-15T05:16:36.957Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/eb/34c16a1ac9ba265d024dc977ad84e1659d931c0a700967c3e59a98ed7514/simplejson-3.20.1-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:f31c4a3a7ab18467ee73a27f3e59158255d1520f3aad74315edde7a940f1be23", size = 93100, upload-time = "2025-02-15T05:16:38.801Z" },
-    { url = "https://files.pythonhosted.org/packages/41/fc/2c2c007d135894971e6814e7c0806936e5bade28f8db4dd7e2a58b50debd/simplejson-3.20.1-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:884e6183d16b725e113b83a6fc0230152ab6627d4d36cb05c89c2c5bccfa7bc6", size = 75464, upload-time = "2025-02-15T05:16:40.905Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/05/2b5ecb33b776c34bb5cace5de5d7669f9b60e3ca13c113037b2ca86edfbd/simplejson-3.20.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:03d7a426e416fe0d3337115f04164cd9427eb4256e843a6b8751cacf70abc832", size = 75112, upload-time = "2025-02-15T05:16:42.246Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/36/1f3609a2792f06cd4b71030485f78e91eb09cfd57bebf3116bf2980a8bac/simplejson-3.20.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:000602141d0bddfcff60ea6a6e97d5e10c9db6b17fd2d6c66199fa481b6214bb", size = 150182, upload-time = "2025-02-15T05:16:43.557Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/b0/053fbda38b8b602a77a4f7829def1b4f316cd8deb5440a6d3ee90790d2a4/simplejson-3.20.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:af8377a8af78226e82e3a4349efdde59ffa421ae88be67e18cef915e4023a595", size = 158363, upload-time = "2025-02-15T05:16:45.748Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/4b/2eb84ae867539a80822e92f9be4a7200dffba609275faf99b24141839110/simplejson-3.20.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:15c7de4c88ab2fbcb8781a3b982ef883696736134e20b1210bca43fb42ff1acf", size = 148415, upload-time = "2025-02-15T05:16:47.861Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/bd/400b0bd372a5666addf2540c7358bfc3841b9ce5cdbc5cc4ad2f61627ad8/simplejson-3.20.1-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:455a882ff3f97d810709f7b620007d4e0aca8da71d06fc5c18ba11daf1c4df49", size = 152213, upload-time = "2025-02-15T05:16:49.25Z" },
-    { url = "https://files.pythonhosted.org/packages/50/12/143f447bf6a827ee9472693768dc1a5eb96154f8feb140a88ce6973a3cfa/simplejson-3.20.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:fc0f523ce923e7f38eb67804bc80e0a028c76d7868500aa3f59225574b5d0453", size = 150048, upload-time = "2025-02-15T05:16:51.5Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/ea/dd9b3e8e8ed710a66f24a22c16a907c9b539b6f5f45fd8586bd5c231444e/simplejson-3.20.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:76461ec929282dde4a08061071a47281ad939d0202dc4e63cdd135844e162fbc", size = 151668, upload-time = "2025-02-15T05:16:53Z" },
-    { url = "https://files.pythonhosted.org/packages/99/af/ee52a8045426a0c5b89d755a5a70cc821815ef3c333b56fbcad33c4435c0/simplejson-3.20.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:ab19c2da8c043607bde4d4ef3a6b633e668a7d2e3d56f40a476a74c5ea71949f", size = 158840, upload-time = "2025-02-15T05:16:54.851Z" },
-    { url = "https://files.pythonhosted.org/packages/68/db/ab32869acea6b5de7d75fa0dac07a112ded795d41eaa7e66c7813b17be95/simplejson-3.20.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b2578bedaedf6294415197b267d4ef678fea336dd78ee2a6d2f4b028e9d07be3", size = 154212, upload-time = "2025-02-15T05:16:56.318Z" },
-    { url = "https://files.pythonhosted.org/packages/fa/7a/e3132d454977d75a3bf9a6d541d730f76462ebf42a96fea2621498166f41/simplejson-3.20.1-cp312-cp312-win32.whl", hash = "sha256:339f407373325a36b7fd744b688ba5bae0666b5d340ec6d98aebc3014bf3d8ea", size = 74101, upload-time = "2025-02-15T05:16:57.746Z" },
-    { url = "https://files.pythonhosted.org/packages/bc/5d/4e243e937fa3560107c69f6f7c2eed8589163f5ed14324e864871daa2dd9/simplejson-3.20.1-cp312-cp312-win_amd64.whl", hash = "sha256:627d4486a1ea7edf1f66bb044ace1ce6b4c1698acd1b05353c97ba4864ea2e17", size = 75736, upload-time = "2025-02-15T05:16:59.017Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/ba/d32fe890a5edaf4a8518adf043bccf7866b600123f512a6de0988cf36810/simplejson-3.20.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:a8011f1dd1d676befcd4d675ebdbfdbbefd3bf350052b956ba8c699fca7d8cef", size = 93773, upload-time = "2025-02-15T05:18:28.231Z" },
-    { url = "https://files.pythonhosted.org/packages/48/c7/361e7f6695b56001a04e0a5cc623cd6c82ea2f45e872e61213e405cc8a24/simplejson-3.20.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:e91703a4c5fec53e36875ae426ad785f4120bd1d93b65bed4752eeccd1789e0c", size = 75697, upload-time = "2025-02-15T05:18:30.006Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/2f/d0ff0b772d4ef092876eb85c99bc591c446b0502715551dad7dfc7f7c2c0/simplejson-3.20.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e39eaa57c7757daa25bcd21f976c46be443b73dd6c3da47fe5ce7b7048ccefe2", size = 75692, upload-time = "2025-02-15T05:18:31.424Z" },
-    { url = "https://files.pythonhosted.org/packages/26/94/cab4db9530b6ca9d62f16a260e8311b04130ccd670dab75e958fcb44590e/simplejson-3.20.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ceab2ce2acdc7fbaa433a93006758db6ba9a659e80c4faa13b80b9d2318e9b17", size = 138106, upload-time = "2025-02-15T05:18:32.907Z" },
-    { url = "https://files.pythonhosted.org/packages/40/22/11c0f746bdb44c297cea8a37d8f7ccb75ea6681132aadfb9f820d9a52647/simplejson-3.20.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6d4f320c33277a5b715db5bf5b10dae10c19076bd6d66c2843e04bd12d1f1ea5", size = 146242, upload-time = "2025-02-15T05:18:35.223Z" },
-    { url = "https://files.pythonhosted.org/packages/78/e9/b7c4c26f29b41cc41ba5f0224c47adbfa7f28427418edfd58ab122f3b584/simplejson-3.20.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2b6436c48e64378fa844d8c9e58a5ed0352bbcfd4028369a9b46679b7ab79d2d", size = 133866, upload-time = "2025-02-15T05:18:36.998Z" },
-    { url = "https://files.pythonhosted.org/packages/09/68/1e81ed83f38906c8859f2b973afb19302357d6003e724a6105cee0f61ec7/simplejson-3.20.1-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6e18345c8dda5d699be8166b61f9d80aaee4545b709f1363f60813dc032dac53", size = 137444, upload-time = "2025-02-15T05:18:38.763Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/6b/8d1e076c543277c1d603230eec24f4dd75ebce46d351c0679526d202981f/simplejson-3.20.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:90b573693d1526bed576f6817e2a492eaaef68f088b57d7a9e83d122bbb49e51", size = 139617, upload-time = "2025-02-15T05:18:40.36Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/46/7b74803de10d4157c5cd2e89028897fa733374667bc5520a44b23b6c887a/simplejson-3.20.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:272cc767826e924a6bd369ea3dbf18e166ded29059c7a4d64d21a9a22424b5b5", size = 139725, upload-time = "2025-02-15T05:18:42.012Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/8f/9991582665a7b6d95415e439bb4fbaa4faf0f77231666675a0fd1de54107/simplejson-3.20.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:51b41f284d603c4380732d7d619f8b34bd04bc4aa0ed0ed5f4ffd0539b14da44", size = 148010, upload-time = "2025-02-15T05:18:43.749Z" },
-    { url = "https://files.pythonhosted.org/packages/54/ee/3c6e91989cdf65ec75e75662d9f15cfe167a792b893806169ea5b1da6fd2/simplejson-3.20.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:6e6697a3067d281f01de0fe96fc7cba4ea870d96d7deb7bfcf85186d74456503", size = 140624, upload-time = "2025-02-15T05:18:45.498Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/bd/05e13ebb7ead81c8b555f4ccc741ea7dfa0ef5c2a0c183d6a7bc50a02bca/simplejson-3.20.1-cp39-cp39-win32.whl", hash = "sha256:6dd3a1d5aca87bf947f3339b0f8e8e329f1badf548bdbff37fac63c17936da8e", size = 74148, upload-time = "2025-02-15T05:18:47.27Z" },
-    { url = "https://files.pythonhosted.org/packages/88/c9/d8bf87aaebec5a4c3ccfd5228689578e2fe77027d6114a259255d54969bf/simplejson-3.20.1-cp39-cp39-win_amd64.whl", hash = "sha256:463f1fca8fbf23d088e5850fdd0dd4d5faea8900a9f9680270bd98fd649814ca", size = 75732, upload-time = "2025-02-15T05:18:49.598Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/30/00f02a0a921556dd5a6db1ef2926a1bc7a8bbbfb1c49cfed68a275b8ab2b/simplejson-3.20.1-py3-none-any.whl", hash = "sha256:8a6c1bbac39fa4a79f83cbf1df6ccd8ff7069582a9fd8db1e52cea073bc2c697", size = 57121, upload-time = "2025-02-15T05:18:51.243Z" },
+version = "3.20.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/41/f4/a1ac5ed32f7ed9a088d62a59d410d4c204b3b3815722e2ccfb491fa8251b/simplejson-3.20.2.tar.gz", hash = "sha256:5fe7a6ce14d1c300d80d08695b7f7e633de6cd72c80644021874d985b3393649", size = 85784, upload-time = "2025-09-26T16:29:36.64Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/78/09/2bf3761de89ea2d91bdce6cf107dcd858892d0adc22c995684878826cc6b/simplejson-3.20.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:6d7286dc11af60a2f76eafb0c2acde2d997e87890e37e24590bb513bec9f1bc5", size = 94039, upload-time = "2025-09-26T16:27:29.283Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/33/c3277db8931f0ae9e54b9292668863365672d90fb0f632f4cf9829cb7d68/simplejson-3.20.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:c01379b4861c3b0aa40cba8d44f2b448f5743999aa68aaa5d3ef7049d4a28a2d", size = 75894, upload-time = "2025-09-26T16:27:30.378Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/ea/ae47b04d03c7c8a7b7b1a8b39a6e27c3bd424e52f4988d70aca6293ff5e5/simplejson-3.20.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a16b029ca25645b3bc44e84a4f941efa51bf93c180b31bd704ce6349d1fc77c1", size = 76116, upload-time = "2025-09-26T16:27:31.42Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/42/6c9af551e5a8d0f171d6dce3d9d1260068927f7b80f1f09834e07887c8c4/simplejson-3.20.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3e22a5fb7b1437ffb057e02e1936a3bfb19084ae9d221ec5e9f4cf85f69946b6", size = 138827, upload-time = "2025-09-26T16:27:32.486Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/22/5e268bbcbe9f75577491e406ec0a5536f5b2fa91a3b52031fea51cd83e1d/simplejson-3.20.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d8b6ff02fc7b8555c906c24735908854819b0d0dc85883d453e23ca4c0445d01", size = 146772, upload-time = "2025-09-26T16:27:34.036Z" },
+    { url = "https://files.pythonhosted.org/packages/71/b4/800f14728e2ad666f420dfdb57697ca128aeae7f991b35759c09356b829a/simplejson-3.20.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2bfc1c396ad972ba4431130b42307b2321dba14d988580c1ac421ec6a6b7cee3", size = 134497, upload-time = "2025-09-26T16:27:35.211Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/b9/c54eef4226c6ac8e9a389bbe5b21fef116768f97a2dc1a683c716ffe66ef/simplejson-3.20.2-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3a97249ee1aee005d891b5a211faf58092a309f3d9d440bc269043b08f662eda", size = 138172, upload-time = "2025-09-26T16:27:36.44Z" },
+    { url = "https://files.pythonhosted.org/packages/09/36/4e282f5211b34620f1b2e4b51d9ddaab5af82219b9b7b78360a33f7e5387/simplejson-3.20.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:f1036be00b5edaddbddbb89c0f80ed229714a941cfd21e51386dc69c237201c2", size = 140272, upload-time = "2025-09-26T16:27:37.605Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/b0/94ad2cf32f477c449e1f63c863d8a513e2408d651c4e58fe4b6a7434e168/simplejson-3.20.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:5d6f5bacb8cdee64946b45f2680afa3f54cd38e62471ceda89f777693aeca4e4", size = 140468, upload-time = "2025-09-26T16:27:39.015Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/46/827731e4163be3f987cb8ee90f5d444161db8f540b5e735355faa098d9bc/simplejson-3.20.2-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:8db6841fb796ec5af632f677abf21c6425a1ebea0d9ac3ef1a340b8dc69f52b8", size = 148700, upload-time = "2025-09-26T16:27:40.171Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/28/c32121064b1ec2fb7b5d872d9a1abda62df064d35e0160eddfa907118343/simplejson-3.20.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:c0a341f7cc2aae82ee2b31f8a827fd2e51d09626f8b3accc441a6907c88aedb7", size = 141323, upload-time = "2025-09-26T16:27:41.324Z" },
+    { url = "https://files.pythonhosted.org/packages/46/b6/c897c54326fe86dd12d101981171a49361949f4728294f418c3b86a1af77/simplejson-3.20.2-cp310-cp310-win32.whl", hash = "sha256:27f9c01a6bc581d32ab026f515226864576da05ef322d7fc141cd8a15a95ce53", size = 74377, upload-time = "2025-09-26T16:27:42.533Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/87/a6e03d4d80cca99c1fee4e960f3440e2f21be9470e537970f960ca5547f1/simplejson-3.20.2-cp310-cp310-win_amd64.whl", hash = "sha256:c0a63ec98a4547ff366871bf832a7367ee43d047bcec0b07b66c794e2137b476", size = 76081, upload-time = "2025-09-26T16:27:43.945Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/3e/96898c6c66d9dca3f9bd14d7487bf783b4acc77471b42f979babbb68d4ca/simplejson-3.20.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:06190b33cd7849efc413a5738d3da00b90e4a5382fd3d584c841ac20fb828c6f", size = 92633, upload-time = "2025-09-26T16:27:45.028Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/a2/cd2e10b880368305d89dd540685b8bdcc136df2b3c76b5ddd72596254539/simplejson-3.20.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4ad4eac7d858947a30d2c404e61f16b84d16be79eb6fb316341885bdde864fa8", size = 75309, upload-time = "2025-09-26T16:27:46.142Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/02/290f7282eaa6ebe945d35c47e6534348af97472446951dce0d144e013f4c/simplejson-3.20.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:b392e11c6165d4a0fde41754a0e13e1d88a5ad782b245a973dd4b2bdb4e5076a", size = 75308, upload-time = "2025-09-26T16:27:47.542Z" },
+    { url = "https://files.pythonhosted.org/packages/43/91/43695f17b69e70c4b0b03247aa47fb3989d338a70c4b726bbdc2da184160/simplejson-3.20.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:51eccc4e353eed3c50e0ea2326173acdc05e58f0c110405920b989d481287e51", size = 143733, upload-time = "2025-09-26T16:27:48.673Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/4b/fdcaf444ac1c3cbf1c52bf00320c499e1cf05d373a58a3731ae627ba5e2d/simplejson-3.20.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:306e83d7c331ad833d2d43c76a67f476c4b80c4a13334f6e34bb110e6105b3bd", size = 153397, upload-time = "2025-09-26T16:27:49.89Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/83/21550f81a50cd03599f048a2d588ffb7f4c4d8064ae091511e8e5848eeaa/simplejson-3.20.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f820a6ac2ef0bc338ae4963f4f82ccebdb0824fe9caf6d660670c578abe01013", size = 141654, upload-time = "2025-09-26T16:27:51.168Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/54/d76c0e72ad02450a3e723b65b04f49001d0e73218ef6a220b158a64639cb/simplejson-3.20.2-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:21e7a066528a5451433eb3418184f05682ea0493d14e9aae690499b7e1eb6b81", size = 144913, upload-time = "2025-09-26T16:27:52.331Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/49/976f59b42a6956d4aeb075ada16ad64448a985704bc69cd427a2245ce835/simplejson-3.20.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:438680ddde57ea87161a4824e8de04387b328ad51cfdf1eaf723623a3014b7aa", size = 144568, upload-time = "2025-09-26T16:27:53.41Z" },
+    { url = "https://files.pythonhosted.org/packages/60/c7/30bae30424ace8cd791ca660fed454ed9479233810fe25c3f3eab3d9dc7b/simplejson-3.20.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:cac78470ae68b8d8c41b6fca97f5bf8e024ca80d5878c7724e024540f5cdaadb", size = 146239, upload-time = "2025-09-26T16:27:54.502Z" },
+    { url = "https://files.pythonhosted.org/packages/79/3e/7f3b7b97351c53746e7b996fcd106986cda1954ab556fd665314756618d2/simplejson-3.20.2-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:7524e19c2da5ef281860a3d74668050c6986be15c9dd99966034ba47c68828c2", size = 154497, upload-time = "2025-09-26T16:27:55.885Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/48/7241daa91d0bf19126589f6a8dcbe8287f4ed3d734e76fd4a092708947be/simplejson-3.20.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:0e9b6d845a603b2eef3394eb5e21edb8626cd9ae9a8361d14e267eb969dbe413", size = 148069, upload-time = "2025-09-26T16:27:57.039Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/f4/ef18d2962fe53e7be5123d3784e623859eec7ed97060c9c8536c69d34836/simplejson-3.20.2-cp311-cp311-win32.whl", hash = "sha256:47d8927e5ac927fdd34c99cc617938abb3624b06ff86e8e219740a86507eb961", size = 74158, upload-time = "2025-09-26T16:27:58.265Z" },
+    { url = "https://files.pythonhosted.org/packages/35/fd/3d1158ecdc573fdad81bf3cc78df04522bf3959758bba6597ba4c956c74d/simplejson-3.20.2-cp311-cp311-win_amd64.whl", hash = "sha256:ba4edf3be8e97e4713d06c3d302cba1ff5c49d16e9d24c209884ac1b8455520c", size = 75911, upload-time = "2025-09-26T16:27:59.292Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/9e/1a91e7614db0416885eab4136d49b7303de20528860ffdd798ce04d054db/simplejson-3.20.2-cp312-cp312-macosx_10_9_universal2.whl", hash = "sha256:4376d5acae0d1e91e78baeba4ee3cf22fbf6509d81539d01b94e0951d28ec2b6", size = 93523, upload-time = "2025-09-26T16:28:00.356Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/2b/d2413f5218fc25608739e3d63fe321dfa85c5f097aa6648dbe72513a5f12/simplejson-3.20.2-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:f8fe6de652fcddae6dec8f281cc1e77e4e8f3575249e1800090aab48f73b4259", size = 75844, upload-time = "2025-09-26T16:28:01.756Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/f1/efd09efcc1e26629e120fef59be059ce7841cc6e1f949a4db94f1ae8a918/simplejson-3.20.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:25ca2663d99328d51e5a138f22018e54c9162438d831e26cfc3458688616eca8", size = 75655, upload-time = "2025-09-26T16:28:03.037Z" },
+    { url = "https://files.pythonhosted.org/packages/97/ec/5c6db08e42f380f005d03944be1af1a6bd501cc641175429a1cbe7fb23b9/simplejson-3.20.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:12a6b2816b6cab6c3fd273d43b1948bc9acf708272074c8858f579c394f4cbc9", size = 150335, upload-time = "2025-09-26T16:28:05.027Z" },
+    { url = "https://files.pythonhosted.org/packages/81/f5/808a907485876a9242ec67054da7cbebefe0ee1522ef1c0be3bfc90f96f6/simplejson-3.20.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:ac20dc3fcdfc7b8415bfc3d7d51beccd8695c3f4acb7f74e3a3b538e76672868", size = 158519, upload-time = "2025-09-26T16:28:06.5Z" },
+    { url = "https://files.pythonhosted.org/packages/66/af/b8a158246834645ea890c36136584b0cc1c0e4b83a73b11ebd9c2a12877c/simplejson-3.20.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:db0804d04564e70862ef807f3e1ace2cc212ef0e22deb1b3d6f80c45e5882c6b", size = 148571, upload-time = "2025-09-26T16:28:07.715Z" },
+    { url = "https://files.pythonhosted.org/packages/20/05/ed9b2571bbf38f1a2425391f18e3ac11cb1e91482c22d644a1640dea9da7/simplejson-3.20.2-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:979ce23ea663895ae39106946ef3d78527822d918a136dbc77b9e2b7f006237e", size = 152367, upload-time = "2025-09-26T16:28:08.921Z" },
+    { url = "https://files.pythonhosted.org/packages/81/2c/bad68b05dd43e93f77994b920505634d31ed239418eb6a88997d06599983/simplejson-3.20.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a2ba921b047bb029805726800819675249ef25d2f65fd0edb90639c5b1c3033c", size = 150205, upload-time = "2025-09-26T16:28:10.086Z" },
+    { url = "https://files.pythonhosted.org/packages/69/46/90c7fc878061adafcf298ce60cecdee17a027486e9dce507e87396d68255/simplejson-3.20.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:12d3d4dc33770069b780cc8f5abef909fe4a3f071f18f55f6d896a370fd0f970", size = 151823, upload-time = "2025-09-26T16:28:11.329Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/27/b85b03349f825ae0f5d4f780cdde0bbccd4f06c3d8433f6a3882df887481/simplejson-3.20.2-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:aff032a59a201b3683a34be1169e71ddda683d9c3b43b261599c12055349251e", size = 158997, upload-time = "2025-09-26T16:28:12.917Z" },
+    { url = "https://files.pythonhosted.org/packages/71/ad/d7f3c331fb930638420ac6d236db68e9f4c28dab9c03164c3cd0e7967e15/simplejson-3.20.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:30e590e133b06773f0dc9c3f82e567463df40598b660b5adf53eb1c488202544", size = 154367, upload-time = "2025-09-26T16:28:14.393Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/46/5c67324addd40fa2966f6e886cacbbe0407c03a500db94fb8bb40333fcdf/simplejson-3.20.2-cp312-cp312-win32.whl", hash = "sha256:8d7be7c99939cc58e7c5bcf6bb52a842a58e6c65e1e9cdd2a94b697b24cddb54", size = 74285, upload-time = "2025-09-26T16:28:15.931Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/c9/5cc2189f4acd3a6e30ffa9775bf09b354302dbebab713ca914d7134d0f29/simplejson-3.20.2-cp312-cp312-win_amd64.whl", hash = "sha256:2c0b4a67e75b945489052af6590e7dca0ed473ead5d0f3aad61fa584afe814ab", size = 75969, upload-time = "2025-09-26T16:28:17.017Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/2d/7c4968c60ddc8b504b77301cc80d6e75cd0269b81a779b01d66d8f36dcb8/simplejson-3.20.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:b3bf76512ccb07d47944ebdca44c65b781612d38b9098566b4bb40f713fc4047", size = 94039, upload-time = "2025-09-26T16:29:17.406Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/e4/d96b56fb87f245240b514c1fe552e76c17e09f0faa1f61137b2296f81529/simplejson-3.20.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:214e26acf2dfb9ff3314e65c4e168a6b125bced0e2d99a65ea7b0f169db1e562", size = 75893, upload-time = "2025-09-26T16:29:18.534Z" },
+    { url = "https://files.pythonhosted.org/packages/09/4f/be411eeb52ab21d6d4c00722b632dd2bd430c01a47dfed3c15ef5ad7ee6e/simplejson-3.20.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2fb1259ca9c385b0395bad59cdbf79535a5a84fb1988f339a49bfbc57455a35a", size = 76104, upload-time = "2025-09-26T16:29:19.66Z" },
+    { url = "https://files.pythonhosted.org/packages/66/6f/3bd0007b64881a90a058c59a4869b1b4f130ddb86a726f884fafc67e5ef7/simplejson-3.20.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c34e028a2ba8553a208ded1da5fa8501833875078c4c00a50dffc33622057881", size = 138261, upload-time = "2025-09-26T16:29:20.822Z" },
+    { url = "https://files.pythonhosted.org/packages/15/5d/b6d0b71508e503c759a0a7563cb2c28716ec8af9828ca9f5b59023011406/simplejson-3.20.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b538f9d9e503b0dd43af60496780cb50755e4d8e5b34e5647b887675c1ae9fee", size = 146397, upload-time = "2025-09-26T16:29:22.363Z" },
+    { url = "https://files.pythonhosted.org/packages/19/24/40b3e5a3ca5e6f80cc1c639fcd5565ae087e72e8656dea780f02302ddc97/simplejson-3.20.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ab998e416ded6c58f549a22b6a8847e75a9e1ef98eb9fbb2863e1f9e61a4105b", size = 134020, upload-time = "2025-09-26T16:29:23.615Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/8c/8fc2c2734ac9e514124635b25ca8f7e347db1ded4a30417ee41e78e6d61c/simplejson-3.20.2-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6a8f1c307edf5fbf0c6db3396c5d3471409c4a40c7a2a466fbc762f20d46601a", size = 137598, upload-time = "2025-09-26T16:29:24.835Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/d9/15036d7f43c6208fb0fbc827f9f897c1f577fba02aeb7a8a223581da4925/simplejson-3.20.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:5a7bbac80bdb82a44303f5630baee140aee208e5a4618e8b9fde3fc400a42671", size = 139770, upload-time = "2025-09-26T16:29:26.244Z" },
+    { url = "https://files.pythonhosted.org/packages/73/cc/18374fb9dfcb4827b692ca5a33bdb607384ca06cdb645e0b863022dae8a3/simplejson-3.20.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:5ef70ec8fe1569872e5a3e4720c1e1dcb823879a3c78bc02589eb88fab920b1f", size = 139884, upload-time = "2025-09-26T16:29:28.51Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/a2/1526d4152806670124dd499ff831726a92bd7e029e8349c4affa78ea8845/simplejson-3.20.2-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:cb11c09c99253a74c36925d461c86ea25f0140f3b98ff678322734ddc0f038d7", size = 148166, upload-time = "2025-09-26T16:29:29.789Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/77/fc16d41b5f67a2591c9b6ff7b0f6aed2b2aed1b6912bb346b61279697638/simplejson-3.20.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:66f7c78c6ef776f8bd9afaad455e88b8197a51e95617bcc44b50dd974a7825ba", size = 140778, upload-time = "2025-09-26T16:29:31.025Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/97/a26ef6b7387349623c042f329df70a4f3baf3a365fe6d1154d73da1dcf5a/simplejson-3.20.2-cp39-cp39-win32.whl", hash = "sha256:619ada86bfe3a5aa02b8222ca6bfc5aa3e1075c1fb5b3263d24ba579382df472", size = 74339, upload-time = "2025-09-26T16:29:32.648Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/b7/94c6049a99e3c04eed2064e91295370b7429e2361188e35a78df562312e0/simplejson-3.20.2-cp39-cp39-win_amd64.whl", hash = "sha256:44a6235e09ca5cc41aa5870a952489c06aa4aee3361ae46daa947d8398e57502", size = 76067, upload-time = "2025-09-26T16:29:34.184Z" },
+    { url = "https://files.pythonhosted.org/packages/05/5b/83e1ff87eb60ca706972f7e02e15c0b33396e7bdbd080069a5d1b53cf0d8/simplejson-3.20.2-py3-none-any.whl", hash = "sha256:3b6bb7fb96efd673eac2e4235200bfffdc2353ad12c54117e1e4e2fc485ac017", size = 57309, upload-time = "2025-09-26T16:29:35.312Z" },
 ]
 
 [[package]]
@@ -4060,25 +4834,52 @@ wheels = [
 ]
 
 [[package]]
-name = "starlette"
-version = "0.47.3"
+name = "soundfile"
+version = "0.13.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "anyio" },
-    { name = "typing-extensions" },
+    { name = "cffi" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/e1/41/9b873a8c055582859b239be17902a85339bec6a30ad162f98c9b0288a2cc/soundfile-0.13.1.tar.gz", hash = "sha256:b2c68dab1e30297317080a5b43df57e302584c49e2942defdde0acccc53f0e5b", size = 46156, upload-time = "2025-01-25T09:17:04.831Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/64/28/e2a36573ccbcf3d57c00626a21fe51989380636e821b341d36ccca0c1c3a/soundfile-0.13.1-py2.py3-none-any.whl", hash = "sha256:a23c717560da2cf4c7b5ae1142514e0fd82d6bbd9dfc93a50423447142f2c445", size = 25751, upload-time = "2025-01-25T09:16:44.235Z" },
+    { url = "https://files.pythonhosted.org/packages/58/ae/c0e4a53d77cf6e9a04179535766b3321b0b9ced5f70522e4caf9329f0046/soundfile-0.13.1-py2.py3-none-manylinux_2_28_aarch64.whl", hash = "sha256:9c9e855f5a4d06ce4213f31918653ab7de0c5a8d8107cd2427e44b42df547deb", size = 1235729, upload-time = "2025-01-25T09:16:53.018Z" },
+    { url = "https://files.pythonhosted.org/packages/57/5e/70bdd9579b35003a489fc850b5047beeda26328053ebadc1fb60f320f7db/soundfile-0.13.1-py2.py3-none-manylinux_2_28_x86_64.whl", hash = "sha256:03267c4e493315294834a0870f31dbb3b28a95561b80b134f0bd3cf2d5f0e618", size = 1313646, upload-time = "2025-01-25T09:16:54.872Z" },
+]
+
+[[package]]
+name = "soxr"
+version = "1.0.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/15/b9/cc3017f9a9c9b6e27c5106cc10cc7904653c3eec0729793aec10479dd669/starlette-0.47.3.tar.gz", hash = "sha256:6bc94f839cc176c4858894f1f8908f0ab79dfec1a6b8402f6da9be26ebea52e9", size = 2584144, upload-time = "2025-08-24T13:36:42.122Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/42/7e/f4b461944662ad75036df65277d6130f9411002bfb79e9df7dff40a31db9/soxr-1.0.0.tar.gz", hash = "sha256:e07ee6c1d659bc6957034f4800c60cb8b98de798823e34d2a2bba1caa85a4509", size = 171415, upload-time = "2025-09-07T13:22:21.317Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ce/fd/901cfa59aaa5b30a99e16876f11abe38b59a1a2c51ffb3d7142bb6089069/starlette-0.47.3-py3-none-any.whl", hash = "sha256:89c0778ca62a76b826101e7c709e70680a1699ca7da6b44d38eb0a7e61fe4b51", size = 72991, upload-time = "2025-08-24T13:36:40.887Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/91/00384166f110a3888ea8efd44523ba7168dd2dc39e3e43c931cc2d069fa9/soxr-1.0.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b89685faedebc45af71f08f9957b61cc6143bc94ba43fe38e97067f81e272969", size = 208586, upload-time = "2025-09-07T13:21:50.341Z" },
+    { url = "https://files.pythonhosted.org/packages/75/34/e18f1003e242aabed44ed8902534814d3e64209e4d1d874f5b9b67d73cde/soxr-1.0.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d255741b2f0084fd02d4a2ddd77cd495be9e7e7b6f9dba1c9494f86afefac65b", size = 242310, upload-time = "2025-09-07T13:21:51.56Z" },
+    { url = "https://files.pythonhosted.org/packages/75/12/4b49611c9ba5e9fe6f807d0a83352516808e8e573f8b4e712fc0c17f3363/soxr-1.0.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3f15450e6f65f22f02fcd4c5a9219c873b1e583a73e232805ff160c759a6b586", size = 208867, upload-time = "2025-09-07T13:21:57.076Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/70/92146ab970a3ef8c43ac160035b1e52fde5417f89adb10572f7e788d9596/soxr-1.0.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1f73f57452f9df37b4de7a4052789fcbd474a5b28f38bba43278ae4b489d4384", size = 242633, upload-time = "2025-09-07T13:21:58.621Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/80/10640970998a1d2199bef6c4d92205f36968cddaf3e4d0e9fe35ddd405bd/soxr-1.0.0-cp312-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e8ce273cca101aff3d8c387db5a5a41001ba76ef1837883438d3c652507a9ccc", size = 204707, upload-time = "2025-09-07T13:22:05.125Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/87/2726603c13c2126cb8ded9e57381b7377f4f0df6ba4408e1af5ddbfdc3dd/soxr-1.0.0-cp312-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e8f2a69686f2856d37823bbb7b78c3d44904f311fe70ba49b893af11d6b6047b", size = 238032, upload-time = "2025-09-07T13:22:06.428Z" },
+    { url = "https://files.pythonhosted.org/packages/64/e4/449c59781d9d942bfa88f691099935b41b207f39704515cc7d368c017f9f/soxr-1.0.0-cp39-cp39-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bb86c342862697dbd4a44043f275e5196f2d2c49dca374c78f19b7893988675d", size = 208806, upload-time = "2025-09-07T13:22:17.487Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/89/9af231937ba3bc0c6d42955431290eeab0efcf37987bf0d69f5734bda78a/soxr-1.0.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3d2a4fadd88207c2991fb08c29fc189e7b2e298b598a94ea1747e42c8acb7a01", size = 242642, upload-time = "2025-09-07T13:22:18.727Z" },
 ]
 
 [[package]]
-name = "svgwrite"
-version = "1.4.3"
+name = "starlette"
+version = "0.48.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/16/c1/263d4e93b543390d86d8eb4fc23d9ce8a8d6efd146f9427364109004fa9b/svgwrite-1.4.3.zip", hash = "sha256:a8fbdfd4443302a6619a7f76bc937fc683daf2628d9b737c891ec08b8ce524c3", size = 189516, upload-time = "2022-07-14T14:05:26.107Z" }
+dependencies = [
+    { name = "anyio" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a7/a5/d6f429d43394057b67a6b5bbe6eae2f77a6bf7459d961fdb224bf206eee6/starlette-0.48.0.tar.gz", hash = "sha256:7e8cee469a8ab2352911528110ce9088fdc6a37d9876926e73da7ce4aa4c7a46", size = 2652949, upload-time = "2025-09-13T08:41:05.699Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/84/15/640e399579024a6875918839454025bb1d5f850bb70d96a11eabb644d11c/svgwrite-1.4.3-py3-none-any.whl", hash = "sha256:bb6b2b5450f1edbfa597d924f9ac2dd099e625562e492021d7dd614f65f8a22d", size = 67122, upload-time = "2022-07-14T14:05:24.459Z" },
+    { url = "https://files.pythonhosted.org/packages/be/72/2db2f49247d0a18b4f1bb9a5a39a0162869acf235f3a96418363947b3d46/starlette-0.48.0-py3-none-any.whl", hash = "sha256:0764ca97b097582558ecb498132ed0c7d942f233f365b86ba37770e026510659", size = 73736, upload-time = "2025-09-13T08:41:03.869Z" },
 ]
 
 [[package]]
@@ -4104,7 +4905,7 @@ wheels = [
 
 [[package]]
 name = "tensorboard"
-version = "2.19.0"
+version = "2.20.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "absl-py" },
@@ -4113,14 +4914,14 @@ dependencies = [
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "packaging" },
+    { name = "pillow" },
     { name = "protobuf" },
     { name = "setuptools" },
-    { name = "six" },
     { name = "tensorboard-data-server" },
     { name = "werkzeug" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/5d/12/4f70e8e2ba0dbe72ea978429d8530b0333f0ed2140cc571a48802878ef99/tensorboard-2.19.0-py3-none-any.whl", hash = "sha256:5e71b98663a641a7ce8a6e70b0be8e1a4c0c45d48760b076383ac4755c35b9a0", size = 5503412, upload-time = "2025-02-12T08:17:27.21Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/d9/a5db55f88f258ac669a92858b70a714bbbd5acd993820b41ec4a96a4d77f/tensorboard-2.20.0-py3-none-any.whl", hash = "sha256:9dc9f978cb84c0723acf9a345d96c184f0293d18f166bb8d59ee098e6cfaaba6", size = 5525680, upload-time = "2025-07-17T19:20:49.638Z" },
 ]
 
 [[package]]
@@ -4135,7 +4936,7 @@ wheels = [
 
 [[package]]
 name = "tensorflow"
-version = "2.19.1"
+version = "2.20.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "absl-py" },
@@ -4158,51 +4959,27 @@ dependencies = [
     { name = "setuptools" },
     { name = "six" },
     { name = "tensorboard" },
-    { name = "tensorflow-io-gcs-filesystem", marker = "python_full_version < '3.12'" },
     { name = "termcolor" },
     { name = "typing-extensions" },
     { name = "wrapt" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/cb/6e/8613ef1cea8013b899adae32bb7cc5218434d84a3a2d6f41937e6d7251af/tensorflow-2.19.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:ed3148ee523cb46badd30ea04a6b7fb6056b9bdb7e082de294a86b2fdc01f9e4", size = 252466041, upload-time = "2025-08-13T16:27:10.511Z" },
-    { url = "https://files.pythonhosted.org/packages/65/4a/cdbcef3a959a2eddf5d3ee7a355454e8d18c89a2e197e96a1b4f55e58d86/tensorflow-2.19.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9077404d1928ee58bfb82ea1b0fce25353420338c259742227c27c4e00377f4b", size = 252498918, upload-time = "2025-08-13T16:27:19.812Z" },
-    { url = "https://files.pythonhosted.org/packages/77/60/51d921b17f7b1547db32018d6a933627d4e2a762d2fc2ca6c0032cc8b062/tensorflow-2.19.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:11377d4cecdc665515e370524b650a8ceb80b461f40f5e63993ed27d0ef931da", size = 644754761, upload-time = "2025-08-13T16:27:36.011Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/80/b74fef28ea980bf6a65dcd9cdc55e2c205b1fa7e8c253a6ddf9f260d990e/tensorflow-2.19.1-cp310-cp310-win_amd64.whl", hash = "sha256:a61c209ec594b2ed4ee2d55a905945bd748df6a1d73ab7ccecf7a9234f52045f", size = 375725444, upload-time = "2025-08-13T16:27:55.406Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/91/70ff01474ff0f2266f47ab292819b97088b512d7e9093e416d16c7aac5ea/tensorflow-2.19.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:8a274a1336051bef0e7e042fa5818a9d342e300d9c311c5d7933eb60e2269770", size = 252591021, upload-time = "2025-08-13T16:28:08.623Z" },
-    { url = "https://files.pythonhosted.org/packages/7e/c6/fc6416ac841106f3b6f2b4d16f50793a71d675c3558c482fd4b3cb44f95e/tensorflow-2.19.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:22deedc251837f2382d2b287c705705cb089bfaf79b2274b78c08740e02bf188", size = 252660426, upload-time = "2025-08-13T16:28:17.267Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/58/34f49a802d6d1f1eb6fe89bc9ff52e2bee19d5c3461059a55702fb7e5fb3/tensorflow-2.19.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8e5524eb492160576e5eeea9bb472ddc653123503f2290f83123d5ac9545750e", size = 644896980, upload-time = "2025-08-13T16:28:32.949Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/f7/b0f6e55174217a59e0b850909dc2691cb8afd8f7b38a99151d0a0d81e3ef/tensorflow-2.19.1-cp311-cp311-win_amd64.whl", hash = "sha256:1729bc2ae90a100a87d5a462b5e9b85a088261a03f0a7179eebb8735938bed57", size = 375912087, upload-time = "2025-08-13T16:28:49.636Z" },
-    { url = "https://files.pythonhosted.org/packages/73/c9/7f6430b944b3683d448f2b3ddace47701021116f6891ffc4926ebb957a88/tensorflow-2.19.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:02c69369a82709827c6a5218c6697ee94dcd1397230319cc17c47a3342249354", size = 252656033, upload-time = "2025-08-13T16:29:05.097Z" },
-    { url = "https://files.pythonhosted.org/packages/6f/3e/717bb5b6ec69de7c08917326470432156b5d0fe888e43775de95d8322b07/tensorflow-2.19.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4c0e78b6100545fd90934c557f1ce802e6a9a7fdb031025efc298efb4aa2e3bb", size = 252718124, upload-time = "2025-08-13T16:29:15.802Z" },
-    { url = "https://files.pythonhosted.org/packages/50/96/b33fdf567332f65e78f02df3a69fbea911eec328e81dc4941d890f0777f5/tensorflow-2.19.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:35a6950f3fee0ad2d1668241d31684676f2e6eceb613cbeef960ea22014ef8fb", size = 645010075, upload-time = "2025-08-13T16:29:37.407Z" },
-    { url = "https://files.pythonhosted.org/packages/27/be/e303c6b55d3bc1d392c555f3e00773a5b2a1c82c8f5af052148e3f0d60ee/tensorflow-2.19.1-cp312-cp312-win_amd64.whl", hash = "sha256:14ec6dee7ab3b241aae3537616e6f1ed1c7c67eff5d853e586eb66a2e9276ccc", size = 375963361, upload-time = "2025-08-13T16:29:54.216Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/a2/cb060b8944a03321efbf1a56cfe49144e5ae648fbe68f2eff275ca297feb/tensorflow-2.19.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:04f3b4dbb3ce4b507d853decfb2d116f8d0344d23413e2ad889e1e069fbeb54d", size = 252468489, upload-time = "2025-08-13T16:30:05.684Z" },
-    { url = "https://files.pythonhosted.org/packages/50/0f/e457ff412f20226516bebaeaf28f6a86531c15dde39fbf4f49b849e30cd7/tensorflow-2.19.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:938fb49b3299a77ccdaa26c0824dfcca3002b260c7bc994cb129d4ec12f31619", size = 252500289, upload-time = "2025-08-13T16:30:15.237Z" },
-    { url = "https://files.pythonhosted.org/packages/93/45/d9480e6140fdbdff8400c132c99a221787d66803739bc8664d3ec3f70633/tensorflow-2.19.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7a66b66fcc3eee0cc4d6c405cb5918972afeeb148684e174693d5d0449c3fdaf", size = 644755371, upload-time = "2025-08-13T16:30:31.327Z" },
-    { url = "https://files.pythonhosted.org/packages/a2/8a/a854c04228d41861c77493a7da5510c9b0969fe44c7903745afb8437188a/tensorflow-2.19.1-cp39-cp39-win_amd64.whl", hash = "sha256:7e770ce558ceeed129537683d3bc7ec9214de8bb296354e1c4198169f4deda1f", size = 375723379, upload-time = "2025-08-13T16:30:51.58Z" },
-]
-
-[[package]]
-name = "tensorflow-io-gcs-filesystem"
-version = "0.37.1"
-source = { registry = "https://pypi.org/simple" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/e9/a3/12d7e7326a707919b321e2d6e4c88eb61596457940fd2b8ff3e9b7fac8a7/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:249c12b830165841411ba71e08215d0e94277a49c551e6dd5d72aab54fe5491b", size = 2470224, upload-time = "2024-07-01T23:44:15.341Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/55/3849a188cc15e58fefde20e9524d124a629a67a06b4dc0f6c881cb3c6e39/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:257aab23470a0796978efc9c2bcf8b0bc80f22e6298612a4c0a50d3f4e88060c", size = 3479613, upload-time = "2024-07-01T23:44:17.445Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/19/9095c69e22c879cb3896321e676c69273a549a3148c4f62aa4bc5ebdb20f/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8febbfcc67c61e542a5ac1a98c7c20a91a5e1afc2e14b1ef0cb7c28bc3b6aa70", size = 4842078, upload-time = "2024-07-01T23:44:18.977Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/48/47b7d25572961a48b1de3729b7a11e835b888e41e0203cca82df95d23b91/tensorflow_io_gcs_filesystem-0.37.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9679b36e3a80921876f31685ab6f7270f3411a4cc51bc2847e80d0e4b5291e27", size = 5085736, upload-time = "2024-07-01T23:44:21.034Z" },
-    { url = "https://files.pythonhosted.org/packages/40/9b/b2fb82d0da673b17a334f785fc19c23483165019ddc33b275ef25ca31173/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:32c50ab4e29a23c1f91cd0f9ab8c381a0ab10f45ef5c5252e94965916041737c", size = 2470224, upload-time = "2024-07-01T23:44:23.039Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/cc/16634e76f3647fbec18187258da3ba11184a6232dcf9073dc44579076d36/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:b02f9c5f94fd62773954a04f69b68c4d576d076fd0db4ca25d5479f0fbfcdbad", size = 3479613, upload-time = "2024-07-01T23:44:24.399Z" },
-    { url = "https://files.pythonhosted.org/packages/de/bf/ba597d3884c77d05a78050f3c178933d69e3f80200a261df6eaa920656cd/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6e1f2796b57e799a8ca1b75bf47c2aaa437c968408cc1a402a9862929e104cda", size = 4842079, upload-time = "2024-07-01T23:44:26.825Z" },
-    { url = "https://files.pythonhosted.org/packages/66/7f/e36ae148c2f03d61ca1bff24bc13a0fef6d6825c966abef73fc6f880a23b/tensorflow_io_gcs_filesystem-0.37.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ee7c8ee5fe2fd8cb6392669ef16e71841133041fee8a330eff519ad9b36e4556", size = 5085736, upload-time = "2024-07-01T23:44:28.618Z" },
-    { url = "https://files.pythonhosted.org/packages/70/83/4422804257fe2942ae0af4ea5bcc9df59cb6cb1bd092202ef240751d16aa/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:ffebb6666a7bfc28005f4fbbb111a455b5e7d6cd3b12752b7050863ecb27d5cc", size = 2470224, upload-time = "2024-07-01T23:44:30.232Z" },
-    { url = "https://files.pythonhosted.org/packages/43/9b/be27588352d7bd971696874db92d370f578715c17c0ccb27e4b13e16751e/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:fe8dcc6d222258a080ac3dfcaaaa347325ce36a7a046277f6b3e19abc1efb3c5", size = 3479614, upload-time = "2024-07-01T23:44:32.316Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/46/962f47af08bd39fc9feb280d3192825431a91a078c856d17a78ae4884eb1/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fbb33f1745f218464a59cecd9a18e32ca927b0f4d77abd8f8671b645cc1a182f", size = 4842077, upload-time = "2024-07-01T23:44:33.86Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/9b/790d290c232bce9b691391cf16e95a96e469669c56abfb1d9d0f35fa437c/tensorflow_io_gcs_filesystem-0.37.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:286389a203a5aee1a4fa2e53718c661091aa5fea797ff4fa6715ab8436b02e6c", size = 5085733, upload-time = "2024-07-01T23:44:36.663Z" },
-    { url = "https://files.pythonhosted.org/packages/12/4f/798df777498fab9dc683a658688e962f0af56454eb040c90f836fd9fa67c/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:ee5da49019670ed364f3e5fb86b46420841a6c3cb52a300553c63841671b3e6d", size = 2470221, upload-time = "2024-07-01T23:44:39.105Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/f9/ce6a0efde262a79361f0d67392fdf0d0406781a1ee4fc48d0d8b0553b311/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:8943036bbf84e7a2be3705cb56f9c9df7c48c9e614bb941f0936c58e3ca89d6f", size = 3479613, upload-time = "2024-07-01T23:44:40.583Z" },
-    { url = "https://files.pythonhosted.org/packages/66/5f/334a011caa1eb97689274d1141df8e6b7a25e389f0390bdcd90235de9783/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:426de1173cb81fbd62becec2012fc00322a295326d90eb6c737fab636f182aed", size = 4842075, upload-time = "2024-07-01T23:44:42.094Z" },
-    { url = "https://files.pythonhosted.org/packages/3d/cb/7dcee55fc5a7d7d8a862e12519322851cd5fe5b086f946fd71e4ae1ef281/tensorflow_io_gcs_filesystem-0.37.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0df00891669390078a003cedbdd3b8e645c718b111917535fa1d7725e95cdb95", size = 5087496, upload-time = "2024-07-01T23:44:43.797Z" },
+    { url = "https://files.pythonhosted.org/packages/16/0e/9408083cb80d85024829eb78aa0aa799ca9f030a348acac35631b5191d4b/tensorflow-2.20.0-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:e5f169f8f5130ab255bbe854c5f0ae152e93d3d1ac44f42cb1866003b81a5357", size = 200387116, upload-time = "2025-08-13T16:50:38.945Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/07/ea91ac67a9fd36d3372099f5a3e69860ded544f877f5f2117802388f4212/tensorflow-2.20.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:02a0293d94f5c8b7125b66abf622cc4854a33ae9d618a0d41309f95e091bbaea", size = 259307122, upload-time = "2025-08-13T16:50:47.909Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/9e/0d57922cf46b9e91de636cd5b5e0d7a424ebe98f3245380a713f1f6c2a0b/tensorflow-2.20.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7abd7f3a010e0d354dc804182372779a722d474c4d8a3db8f4a3f5baef2a591e", size = 620425510, upload-time = "2025-08-13T16:51:02.608Z" },
+    { url = "https://files.pythonhosted.org/packages/74/b5/d40e1e389e07de9d113cf8e5d294c04d06124441d57606febfd0fb2cf5a6/tensorflow-2.20.0-cp310-cp310-win_amd64.whl", hash = "sha256:4a69ac2c2ce20720abf3abf917b4e86376326c0976fcec3df330e184b81e4088", size = 331664937, upload-time = "2025-08-13T16:51:17.719Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/69/de33bd90dbddc8eede8f99ddeccfb374f7e18f84beb404bfe2cbbdf8df90/tensorflow-2.20.0-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:5f964016c5035d09b85a246a6b739be89282a7839743f3ea63640224f0c63aee", size = 200507363, upload-time = "2025-08-13T16:51:28.27Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/b7/a3d455db88ab5b35ce53ab885ec0dd9f28d905a86a2250423048bc8cafa0/tensorflow-2.20.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3e9568c8efcb05c0266be223e3269c62ebf7ad3498f156438311735f6fa5ced5", size = 259465882, upload-time = "2025-08-13T16:51:39.546Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/0c/7df285ee8a88139fab0b237003634d90690759fae9c18f55ddb7c04656ec/tensorflow-2.20.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:481499fd0f824583de8945be61d5e827898cdaa4f5ea1bc2cc28ca2ccff8229e", size = 620570129, upload-time = "2025-08-13T16:51:55.104Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/f8/9246d3c7e185a29d7359d8b12b3d70bf2c3150ecf1427ec1382290e71a56/tensorflow-2.20.0-cp311-cp311-win_amd64.whl", hash = "sha256:7551558a48c2e2f6c32a1537f06c654a9df1408a1c18e7b99c3caafbd03edfe3", size = 331845735, upload-time = "2025-08-13T16:52:12.863Z" },
+    { url = "https://files.pythonhosted.org/packages/35/31/47712f425c09cc8b8dba39c6c45aee939c4636a6feb8c81376a4eae653e0/tensorflow-2.20.0-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:52b122f0232fd7ab10f28d537ce08470d0b6dcac7fff9685432daac7f8a06c8f", size = 200540302, upload-time = "2025-08-13T16:52:22.146Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/b4/f028a5de27d0fda10ba6145bc76e40c37ff6d2d1e95b601adb5ae17d635e/tensorflow-2.20.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2bfbfb3dd0e22bffc45fe1e922390d27753e99261fab8a882e802cf98a0e078f", size = 259533109, upload-time = "2025-08-13T16:52:31.513Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/d1/6aa15085d672056d5f08b5f28b1c7ce01c4e12149a23b0c98e3c79d04441/tensorflow-2.20.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:25265b0bc527e0d54b1e9cc60c44a24f44a809fe27666b905f0466471f9c52ec", size = 620682547, upload-time = "2025-08-13T16:52:46.396Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/37/b97abb360b551fbf5870a0ee07e39ff9c655e6e3e2f839bc88be81361842/tensorflow-2.20.0-cp312-cp312-win_amd64.whl", hash = "sha256:1590cbf87b6bcbd34d8e9ad70d0c696135e0aa71be31803b27358cf7ed63f8fc", size = 331887041, upload-time = "2025-08-13T16:53:05.532Z" },
+    { url = "https://files.pythonhosted.org/packages/68/ca/9709f20f9b3c93536d549ef890d1b0dcae6cbfd1e863195f261ec59fc4c6/tensorflow-2.20.0-cp39-cp39-macosx_12_0_arm64.whl", hash = "sha256:a66cbd1b19209d3fbc45cbea80de92514ba455434013937251d65d444779783c", size = 200387750, upload-time = "2025-08-13T16:54:08.084Z" },
+    { url = "https://files.pythonhosted.org/packages/83/ff/a26d49895586207b2704403366ef976dcaa6ed07514699dae9a4fc3fa1a9/tensorflow-2.20.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:28bc33759249c98eabcee9debd24e74506bbe29ac139e050cf0c74aa9888ebdf", size = 259307564, upload-time = "2025-08-13T16:54:17.691Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/fe/f3d738dc7c93ed5f67f9ace8dd3ed66971dab7c5a47f2d1c504ef0d0cf1d/tensorflow-2.20.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0deb5c583dfc53b54fd158a194ce0087b406bb6518af400ca3809735e4548ec3", size = 620427169, upload-time = "2025-08-13T16:54:33.431Z" },
+    { url = "https://files.pythonhosted.org/packages/16/25/b2d7ef3b6570d2827d06066cdfdbc719367c5fe4bce7910901951e5652eb/tensorflow-2.20.0-cp39-cp39-win_amd64.whl", hash = "sha256:dd71a7e7c3270239f4185915e8f2c5d39608c5e18973d6e1d101b153993841eb", size = 331661805, upload-time = "2025-08-13T16:54:50.512Z" },
 ]
 
 [[package]]
@@ -4242,7 +5019,7 @@ wheels = [
 
 [[package]]
 name = "tensorstore"
-version = "0.1.76"
+version = "0.1.78"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -4253,23 +5030,23 @@ dependencies = [
     { name = "ml-dtypes", marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ff/ae/947a9f232de7319b664ed8d278e9e0363e9294da73fd422c687ac4eb070e/tensorstore-0.1.76.tar.gz", hash = "sha256:ed0d565e7a038a84b1b5b5d9f7397caec200b53941d8889f44b7f63dd6abffe7", size = 6869230, upload-time = "2025-07-02T21:34:03.773Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/9f/ee/05eb424437f4db63331c90e4605025eedc0f71da3faff97161d5d7b405af/tensorstore-0.1.78.tar.gz", hash = "sha256:e26074ffe462394cf54197eb76d6569b500f347573cd74da3f4dd5f510a4ad7c", size = 6913502, upload-time = "2025-10-06T17:44:29.649Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ca/66/6da79c349c3576836efec0195851bb71fed307dd9855de0692bf3e4df6cb/tensorstore-0.1.76-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:97e9eeab539369b99828c07e1d7da93062da91b65503dafe06d941208b0869d3", size = 15674263, upload-time = "2025-07-02T21:33:13.387Z" },
-    { url = "https://files.pythonhosted.org/packages/24/e7/014f188a5bd2087cdf0df154ee756ca95bae62ba76765f8f180e9dbc4e78/tensorstore-0.1.76-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a9c6f207a0b5f5138f7844307e5cb96587a73e9ba7b5edb6429524cca985e2c3", size = 13601077, upload-time = "2025-07-02T21:33:16.354Z" },
-    { url = "https://files.pythonhosted.org/packages/48/63/38410a20016c6c0b28d9970bcb63762f4df4b62a53eeddb27044deeb4792/tensorstore-0.1.76-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:05aae734cfaf5c1356b6f2c56226526607361c6e9aa6751df395dda53c3783b1", size = 17545411, upload-time = "2025-07-02T21:33:18.83Z" },
-    { url = "https://files.pythonhosted.org/packages/ca/83/17fb39326adfee5d7048093ee248b3aa69c47031e62832825e29da81d0b7/tensorstore-0.1.76-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7574f837d8bfccf374795100be23f155eff3040f8b0f2010b1a0b2e1d4d6f791", size = 18923422, upload-time = "2025-07-02T21:33:21.38Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/a1/8932f2871e1f390e29f587c3295f3f00113d1a32a05b0723ef06f0134630/tensorstore-0.1.76-cp310-cp310-win_amd64.whl", hash = "sha256:9c964c7538805867894f6ba4a03dfa57e434b73d0c563d93fcbf6245eb67cc2c", size = 12609199, upload-time = "2025-07-02T21:33:24.337Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/9e/b3d691d14122064e16a3a47c14ce3b1178d749e59b3afec91a8656125c29/tensorstore-0.1.76-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:1c882dcf30049952cb6b183c70bd1922815cdbceca8d4115a7fbeb3b5513f9e4", size = 15675667, upload-time = "2025-07-02T21:33:27.409Z" },
-    { url = "https://files.pythonhosted.org/packages/0c/bb/16d97d8b31912f27019115eb23b7feb0b83bf520858b97aec64064653329/tensorstore-0.1.76-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:262b856b21626688cefd11913737a94800b1ba3061d56d70babc439ef69587bc", size = 13602503, upload-time = "2025-07-02T21:33:29.872Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/5a/4b675941a73bc46959f24b3f5a68c246422278022a0e121f9c3f226a7a2b/tensorstore-0.1.76-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:daf1c7fb71cd07bfdf37ae4f297c0a6895b011562f4050f5c8f52f5753cc24cc", size = 17548043, upload-time = "2025-07-02T21:33:32.34Z" },
-    { url = "https://files.pythonhosted.org/packages/96/b9/8e306cbccb12ce4c241ac69fb98a4fc1bad5fc0311112f579bc24bee9c42/tensorstore-0.1.76-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b24f21bbd4830822422e984bc023a37ce7db08be02138c01c96870e62d041e7f", size = 18925296, upload-time = "2025-07-02T21:33:35.061Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/48/b542e9a4fa6f82b00e9a7c41c30003c195178fa78f835ea205b346a45baf/tensorstore-0.1.76-cp311-cp311-win_amd64.whl", hash = "sha256:36ba59d99d8279802793405fb8615803ea0e136ad439e6fe0ab3c3d7df22179d", size = 12609617, upload-time = "2025-07-02T21:33:37.153Z" },
-    { url = "https://files.pythonhosted.org/packages/09/37/f2254b4ae1dabd95e258fa3eb4783ac4db4261bb8c90ff9bfe15549d1238/tensorstore-0.1.76-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:b68450983ccad9e7774e81b2fa37daef1b72c774fd939d9eb4065d6aa70e666a", size = 15712650, upload-time = "2025-07-02T21:33:39.716Z" },
-    { url = "https://files.pythonhosted.org/packages/93/3c/1cae56cbbe9610ff48cb2d7c0921a4d4c333a0540918e3b2db08b521c5f6/tensorstore-0.1.76-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6b7a3856f884279e40f90bad87d0da70869879e124835e650c6b16c80f64fbc4", size = 13624138, upload-time = "2025-07-02T21:33:41.758Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/d2/b92d34a896f608a59dc76c290d4ec9f7d0264a02e4d74864987a6adbd3c9/tensorstore-0.1.76-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8709a98ae0b453eb23525c07372c2be1f6bbd978bba53319f26a1f2a83a77c2a", size = 17538270, upload-time = "2025-07-02T21:33:44.911Z" },
-    { url = "https://files.pythonhosted.org/packages/21/66/142b803541552b02a2fa033b1f48bcb50e1d2df6ac10131aab1857c5141d/tensorstore-0.1.76-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:267edea8f1596f2bd67017ff97b7b350bf3f95ff84947a8babadc5e17ca53663", size = 18910782, upload-time = "2025-07-02T21:33:47.401Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/3e/c264cf1435c04fb998a1f30dd1f066deb370b841412f89e1cb36d37ee4fc/tensorstore-0.1.76-cp312-cp312-win_amd64.whl", hash = "sha256:f66ac63d0c63c3336ac4dc61f1f97b6afe8b512e586ddfdbc91f19175787f321", size = 12611059, upload-time = "2025-07-02T21:33:49.596Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/1e/77eff7bb320f72a9cb6e9a19eee4d78bee4a6ac1c28ceef60df28b4ab670/tensorstore-0.1.78-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:f1bc58164ad964d9cc298d20b62ca704ab6241639a21015e47ce6ea5b5cae27f", size = 15710776, upload-time = "2025-10-06T17:43:47.469Z" },
+    { url = "https://files.pythonhosted.org/packages/55/df/f74f8004b246006ae03c90c28e32d71eb8a86a5b325d2d84dda327babdcc/tensorstore-0.1.78-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:1910101ea85b6507958da28628ef53712c5311df19a795f449604f82bae6a24b", size = 13771121, upload-time = "2025-10-06T17:43:49.88Z" },
+    { url = "https://files.pythonhosted.org/packages/be/b8/ab0d0b2afc53f47fbfd95c10d9ae21d393019aca45c8513657b8d7002f1f/tensorstore-0.1.78-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1e92195db0c8c3ca749f24b1e930ab93382ac27430ac4ad2e3f53fc8f739323f", size = 18154513, upload-time = "2025-10-06T17:43:51.694Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/ea/c1b4cc6a089a39f63e8d189a55c715e393995628b12b4c8560b3ae4874ba/tensorstore-0.1.78-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:90570b867f9100f7405e4116c73910d0bd283a101500ea5680c5a8a881ea05c6", size = 20048971, upload-time = "2025-10-06T17:43:54.358Z" },
+    { url = "https://files.pythonhosted.org/packages/58/2a/7167087885b12473f20ae4fddb9a8feeed6bd44ea8d42c73ae29ad3d1591/tensorstore-0.1.78-cp310-cp310-win_amd64.whl", hash = "sha256:4de9d4ee93d712cb665890af0738f4d74cac3b9b9a0492d477a3ee63fbbf445b", size = 12707793, upload-time = "2025-10-06T17:43:56.405Z" },
+    { url = "https://files.pythonhosted.org/packages/33/b1/45070c393586306cef44c7bfc47ed2eddfb8930e648aaa847f615e3ae797/tensorstore-0.1.78-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:1c91e7ff93561612bd9868f3ee56702b0e4fecb45079a4c152dff9a6aa751913", size = 15712387, upload-time = "2025-10-06T17:43:58.458Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/d8/c045da71460301f37704e1ab1eec9e7e480dc711dbd281d86dc3d792c50e/tensorstore-0.1.78-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:781e123d392b2d9115e94b01849797a4540f54cd6d34c6ee32b9491f2f2a399c", size = 13773158, upload-time = "2025-10-06T17:44:00.285Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/e8/2b0d48100816649ec516fca31d02ad8028c090324e77b1c309c09a172350/tensorstore-0.1.78-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e650d363ad43754626a828a242785e6359a59fedb171276e9a0c66c0bd963cd4", size = 18154388, upload-time = "2025-10-06T17:44:02.428Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/a1/d9be82de18afe764c0fc7fb21b3d3bb0ad12845d202861fff7189afdb99d/tensorstore-0.1.78-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:33fed0ffa7a42ad24ce203486cf039f81b211723b45bd54859ba237a9d3aedb9", size = 20050304, upload-time = "2025-10-06T17:44:04.673Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/fc/b980958f91a9780e4dbc1038da723d2ad91307dbe30563359606f78926e5/tensorstore-0.1.78-cp311-cp311-win_amd64.whl", hash = "sha256:c02df3d8de4703d9ee42c8f620b2288f41c19a0fd5ffa907b72a736678e22188", size = 12708115, upload-time = "2025-10-06T17:44:06.574Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/5f/5853c04bebaed2d3c0ada9245328ffe3fff8b0f0f1c64f4776f67b42033f/tensorstore-0.1.78-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:ce375a8f6621cdb94638b9cdc5266519db16a58353d4c6920e8b9d6bdd419e21", size = 15727539, upload-time = "2025-10-06T17:44:08.631Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/e2/f67fcca8f90258c1cf1326aa366fe10f559f4c60102f53fdcc6614159c45/tensorstore-0.1.78-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:82f68fa5a3b4c84365a667ea0a7465a53d5d969c4d3909ac990f314d1569ffc3", size = 13780753, upload-time = "2025-10-06T17:44:10.488Z" },
+    { url = "https://files.pythonhosted.org/packages/57/de/95013db6ef3b6a14b4237b95184c21becdf56d16605bf42903bb141f729e/tensorstore-0.1.78-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5dc0bd6361d73e3f67d70980f96f4e8bcbd8e810b5475a01333ca9c37f0785a5", size = 18157446, upload-time = "2025-10-06T17:44:12.831Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/75/6e7cef68cab3a672c6668cc80c399ae6626a498a3ef04b35b3704b41e9cc/tensorstore-0.1.78-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:75a17cef99f05fad9cc6fda37f1a1868d5f1502fd577af13174382931481c948", size = 20060211, upload-time = "2025-10-06T17:44:15.189Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/46/4ff3e395c44348c7442523c8ddd8ccc72d9ac81838e7a8f6afdd92131c3e/tensorstore-0.1.78-cp312-cp312-win_amd64.whl", hash = "sha256:56271d4652a7cb445879089f620af47801c091765d35a005505d6bfb8d00c535", size = 12711274, upload-time = "2025-10-06T17:44:17.586Z" },
 ]
 
 [[package]]
@@ -4283,63 +5060,67 @@ wheels = [
 
 [[package]]
 name = "tf-keras"
-version = "2.19.0"
+version = "2.20.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "tensorflow" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a9/71/a6e82b8e8ce04a8ab1ff0560791698b70d1b2dda1a0270f1d69a834e8880/tf_keras-2.19.0.tar.gz", hash = "sha256:b09a407d87a4571ce1e8ca985cfc68483e3d63b2518a5d79a97ad92cb64dbe9c", size = 1261172, upload-time = "2025-03-03T21:24:13.582Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/42/38/6060f6c7472439bb3890b9094d69d31d9f8d5da123b16c738773e70fff91/tf_keras-2.20.1.tar.gz", hash = "sha256:884be5938fb0b2b53b1583c1ae2b660ef87215377c29b5b6a77fd221b472aeaf", size = 1254487, upload-time = "2025-09-04T21:23:41.81Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/45/6b/d245122d108a94df5969ee7408ad343af1627730e91478e01ef098976bfa/tf_keras-2.19.0-py3-none-any.whl", hash = "sha256:4f339e800987b39d1548a8c76a7b33b6801a97ec7fcd89c299ec29741f7890bd", size = 1726787, upload-time = "2025-03-03T21:24:11.826Z" },
+    { url = "https://files.pythonhosted.org/packages/85/6b/d9a8202bfe5c9e3b078cf550bafab962aa9d6b1a1f1180f0065399d4c9b2/tf_keras-2.20.1-py3-none-any.whl", hash = "sha256:3f0e0a34d9a4c8758f24fdc1053e6e335f16ab5534c7d34f1899b8924779760c", size = 1694335, upload-time = "2025-09-04T21:23:40.153Z" },
 ]
 
 [[package]]
 name = "tiktoken"
-version = "0.11.0"
+version = "0.12.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "regex" },
     { name = "requests" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a7/86/ad0155a37c4f310935d5ac0b1ccf9bdb635dcb906e0a9a26b616dd55825a/tiktoken-0.11.0.tar.gz", hash = "sha256:3c518641aee1c52247c2b97e74d8d07d780092af79d5911a6ab5e79359d9b06a", size = 37648, upload-time = "2025-08-08T23:58:08.495Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/7d/ab/4d017d0f76ec3171d469d80fc03dfbb4e48a4bcaddaa831b31d526f05edc/tiktoken-0.12.0.tar.gz", hash = "sha256:b18ba7ee2b093863978fcb14f74b3707cdc8d4d4d3836853ce7ec60772139931", size = 37806, upload-time = "2025-10-06T20:22:45.419Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/dd/f4/ec8d43338d28d53513004ebf4cd83732a135d11011433c58bf045890cc10/tiktoken-0.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:10331d08b5ecf7a780b4fe4d0281328b23ab22cdb4ff65e68d56caeda9940ecc", size = 1130076, upload-time = "2025-08-08T23:57:31.706Z" },
-    { url = "https://files.pythonhosted.org/packages/94/80/fb0ada0a882cb453caf519a4bf0d117c2a3ee2e852c88775abff5413c176/tiktoken-0.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b062c82300341dc87e0258c69f79bed725f87e753c21887aea90d272816be882", size = 1183942, upload-time = "2025-08-08T23:57:33.142Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/e9/6c104355b463601719582823f3ea658bc3aa7c73d1b3b7553ebdc48468ce/tiktoken-0.11.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:195d84bec46169af3b1349a1495c151d37a0ff4cba73fd08282736be7f92cc6c", size = 1244705, upload-time = "2025-08-08T23:57:34.594Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/3e/a05d1547cf7db9dc75d1461cfa7b556a3b48e0516ec29dfc81d984a145f6/tiktoken-0.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2130127471e293d385179c1f3f9cd445070c0772be73cdafb7cec9a3684c0458", size = 1129402, upload-time = "2025-08-08T23:57:39.627Z" },
-    { url = "https://files.pythonhosted.org/packages/34/9a/db7a86b829e05a01fd4daa492086f708e0a8b53952e1dbc9d380d2b03677/tiktoken-0.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:21e43022bf2c33f733ea9b54f6a3f6b4354b909f5a73388fb1b9347ca54a069c", size = 1184046, upload-time = "2025-08-08T23:57:40.689Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/bb/52edc8e078cf062ed749248f1454e9e5cfd09979baadb830b3940e522015/tiktoken-0.11.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:adb4e308eb64380dc70fa30493e21c93475eaa11669dea313b6bbf8210bfd013", size = 1244691, upload-time = "2025-08-08T23:57:42.251Z" },
-    { url = "https://files.pythonhosted.org/packages/65/8e/c769b45ef379bc360c9978c4f6914c79fd432400a6733a8afc7ed7b0726a/tiktoken-0.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6ef72aab3ea240646e642413cb363b73869fed4e604dcfd69eec63dc54d603e8", size = 1128867, upload-time = "2025-08-08T23:57:47.438Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/2d/4d77f6feb9292bfdd23d5813e442b3bba883f42d0ac78ef5fdc56873f756/tiktoken-0.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7f929255c705efec7a28bf515e29dc74220b2f07544a8c81b8d69e8efc4578bd", size = 1183308, upload-time = "2025-08-08T23:57:48.566Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/65/7ff0a65d3bb0fc5a1fb6cc71b03e0f6e71a68c5eea230d1ff1ba3fd6df49/tiktoken-0.11.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:61f1d15822e4404953d499fd1dcc62817a12ae9fb1e4898033ec8fe3915fdf8e", size = 1244301, upload-time = "2025-08-08T23:57:49.642Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/7f/5f950528b54cb3025af4bc3522c23dbfb691afe8ffb292aa1e8dc2e6bddf/tiktoken-0.11.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2302772f035dceb2bcf8e55a735e4604a0b51a6dd50f38218ff664d46ec43807", size = 1130777, upload-time = "2025-08-08T23:58:03.256Z" },
-    { url = "https://files.pythonhosted.org/packages/27/a4/e82ddf0773835ba24536ac8c0dce561e697698ec020a93212a1e041d39b4/tiktoken-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:20b977989afe44c94bcc50db1f76971bb26dca44218bd203ba95925ef56f8e7a", size = 1185692, upload-time = "2025-08-08T23:58:04.476Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/c2/06361e41d176e62797ae65fa678111cdd30553321cf4d83e7b84107ea95f/tiktoken-0.11.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:669a1aa1ad6ebf1b3c26b45deb346f345da7680f845b5ea700bba45c20dea24c", size = 1246518, upload-time = "2025-08-08T23:58:06.126Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/5d/26a691f28ab220d5edc09b9b787399b130f24327ef824de15e5d85ef21aa/tiktoken-0.12.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:cde24cdb1b8a08368f709124f15b36ab5524aac5fa830cc3fdce9c03d4fb8030", size = 1129865, upload-time = "2025-10-06T20:21:36.675Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/94/443fab3d4e5ebecac895712abd3849b8da93b7b7dec61c7db5c9c7ebe40c/tiktoken-0.12.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:6de0da39f605992649b9cfa6f84071e3f9ef2cec458d08c5feb1b6f0ff62e134", size = 1152856, upload-time = "2025-10-06T20:21:37.873Z" },
+    { url = "https://files.pythonhosted.org/packages/54/35/388f941251b2521c70dd4c5958e598ea6d2c88e28445d2fb8189eecc1dfc/tiktoken-0.12.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:6faa0534e0eefbcafaccb75927a4a380463a2eaa7e26000f0173b920e98b720a", size = 1195308, upload-time = "2025-10-06T20:21:39.577Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/00/c6681c7f833dd410576183715a530437a9873fa910265817081f65f9105f/tiktoken-0.12.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:82991e04fc860afb933efb63957affc7ad54f83e2216fe7d319007dab1ba5892", size = 1255697, upload-time = "2025-10-06T20:21:41.154Z" },
+    { url = "https://files.pythonhosted.org/packages/01/84/961106c37b8e49b9fdcf33fe007bb3a8fdcc380c528b20cc7fbba80578b8/tiktoken-0.12.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:f18f249b041851954217e9fd8e5c00b024ab2315ffda5ed77665a05fa91f42dc", size = 1129201, upload-time = "2025-10-06T20:21:47.074Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/d0/3d9275198e067f8b65076a68894bb52fd253875f3644f0a321a720277b8a/tiktoken-0.12.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:47a5bc270b8c3db00bb46ece01ef34ad050e364b51d406b6f9730b64ac28eded", size = 1152444, upload-time = "2025-10-06T20:21:48.139Z" },
+    { url = "https://files.pythonhosted.org/packages/78/db/a58e09687c1698a7c592e1038e01c206569b86a0377828d51635561f8ebf/tiktoken-0.12.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:508fa71810c0efdcd1b898fda574889ee62852989f7c1667414736bcb2b9a4bd", size = 1195080, upload-time = "2025-10-06T20:21:49.246Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/1b/a9e4d2bf91d515c0f74afc526fd773a812232dd6cda33ebea7f531202325/tiktoken-0.12.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a1af81a6c44f008cba48494089dd98cccb8b313f55e961a52f5b222d1e507967", size = 1255240, upload-time = "2025-10-06T20:21:50.274Z" },
+    { url = "https://files.pythonhosted.org/packages/66/c5/ed88504d2f4a5fd6856990b230b56d85a777feab84e6129af0822f5d0f70/tiktoken-0.12.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:65b26c7a780e2139e73acc193e5c63ac754021f160df919add909c1492c0fb37", size = 1129008, upload-time = "2025-10-06T20:21:54.832Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/90/3dae6cc5436137ebd38944d396b5849e167896fc2073da643a49f372dc4f/tiktoken-0.12.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:edde1ec917dfd21c1f2f8046b86348b0f54a2c0547f68149d8600859598769ad", size = 1152665, upload-time = "2025-10-06T20:21:56.129Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/fe/26df24ce53ffde419a42f5f53d755b995c9318908288c17ec3f3448313a3/tiktoken-0.12.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:35a2f8ddd3824608b3d650a000c1ef71f730d0c56486845705a8248da00f9fe5", size = 1194230, upload-time = "2025-10-06T20:21:57.546Z" },
+    { url = "https://files.pythonhosted.org/packages/20/cc/b064cae1a0e9fac84b0d2c46b89f4e57051a5f41324e385d10225a984c24/tiktoken-0.12.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:83d16643edb7fa2c99eff2ab7733508aae1eebb03d5dfc46f5565862810f24e3", size = 1254688, upload-time = "2025-10-06T20:21:58.619Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/1a/62d1d36b167eccd441aff2f0091551ca834295541b949d161021aa658167/tiktoken-0.12.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:df37684ace87d10895acb44b7f447d4700349b12197a526da0d4a4149fde074c", size = 1131575, upload-time = "2025-10-06T20:22:39.023Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/16/544207d63c8c50edd2321228f21d236e4e49d235128bb7e3e0f69eed0807/tiktoken-0.12.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:4c9614597ac94bb294544345ad8cf30dac2129c05e2db8dc53e082f355857af7", size = 1154920, upload-time = "2025-10-06T20:22:40.175Z" },
+    { url = "https://files.pythonhosted.org/packages/99/4c/0a3504157c81364fc0c64cada54efef0567961357e786706ea63bc8946e1/tiktoken-0.12.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:20cf97135c9a50de0b157879c3c4accbb29116bcf001283d26e073ff3b345946", size = 1196766, upload-time = "2025-10-06T20:22:41.365Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/46/8e6a258ae65447c75770fe5ea8968acab369e8c9f537f727c91f83772325/tiktoken-0.12.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:15d875454bbaa3728be39880ddd11a5a2a9e548c29418b41e8fd8a767172b5ec", size = 1258278, upload-time = "2025-10-06T20:22:42.846Z" },
 ]
 
 [[package]]
 name = "tokenizers"
-version = "0.22.0"
+version = "0.22.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "huggingface-hub" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/5e/b4/c1ce3699e81977da2ace8b16d2badfd42b060e7d33d75c4ccdbf9dc920fa/tokenizers-0.22.0.tar.gz", hash = "sha256:2e33b98525be8453f355927f3cab312c36cd3e44f4d7e9e97da2fa94d0a49dcb", size = 362771, upload-time = "2025-08-29T10:25:33.914Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/1c/46/fb6854cec3278fbfa4a75b50232c77622bc517ac886156e6afbfa4d8fc6e/tokenizers-0.22.1.tar.gz", hash = "sha256:61de6522785310a309b3407bac22d99c4db5dba349935e99e4d15ea2226af2d9", size = 363123, upload-time = "2025-09-19T09:49:23.424Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6d/b1/18c13648edabbe66baa85fe266a478a7931ddc0cd1ba618802eb7b8d9865/tokenizers-0.22.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:eaa9620122a3fb99b943f864af95ed14c8dfc0f47afa3b404ac8c16b3f2bb484", size = 3081954, upload-time = "2025-08-29T10:25:24.993Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/02/c3c454b641bd7c4f79e4464accfae9e7dfc913a777d2e561e168ae060362/tokenizers-0.22.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:71784b9ab5bf0ff3075bceeb198149d2c5e068549c0d18fe32d06ba0deb63f79", size = 2945644, upload-time = "2025-08-29T10:25:23.405Z" },
-    { url = "https://files.pythonhosted.org/packages/55/02/d10185ba2fd8c2d111e124c9d92de398aee0264b35ce433f79fb8472f5d0/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ec5b71f668a8076802b0241a42387d48289f25435b86b769ae1837cad4172a17", size = 3254764, upload-time = "2025-08-29T10:25:12.445Z" },
-    { url = "https://files.pythonhosted.org/packages/13/89/17514bd7ef4bf5bfff58e2b131cec0f8d5cea2b1c8ffe1050a2c8de88dbb/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ea8562fa7498850d02a16178105b58803ea825b50dc9094d60549a7ed63654bb", size = 3161654, upload-time = "2025-08-29T10:25:15.493Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/d8/bac9f3a7ef6dcceec206e3857c3b61bb16c6b702ed7ae49585f5bd85c0ef/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:4136e1558a9ef2e2f1de1555dcd573e1cbc4a320c1a06c4107a3d46dc8ac6e4b", size = 3511484, upload-time = "2025-08-29T10:25:20.477Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/27/9c9800eb6763683010a4851db4d1802d8cab9cec114c17056eccb4d4a6e0/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cdf5954de3962a5fd9781dc12048d24a1a6f1f5df038c6e95db328cd22964206", size = 3712829, upload-time = "2025-08-29T10:25:17.154Z" },
-    { url = "https://files.pythonhosted.org/packages/10/e3/b1726dbc1f03f757260fa21752e1921445b5bc350389a8314dd3338836db/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8337ca75d0731fc4860e6204cc24bb36a67d9736142aa06ed320943b50b1e7ed", size = 3408934, upload-time = "2025-08-29T10:25:18.76Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/61/aeab3402c26874b74bb67a7f2c4b569dde29b51032c5384db592e7b216f4/tokenizers-0.22.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a89264e26f63c449d8cded9061adea7b5de53ba2346fc7e87311f7e4117c1cc8", size = 3345585, upload-time = "2025-08-29T10:25:22.08Z" },
-    { url = "https://files.pythonhosted.org/packages/bc/d3/498b4a8a8764cce0900af1add0f176ff24f475d4413d55b760b8cdf00893/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:790bad50a1b59d4c21592f9c3cf5e5cf9c3c7ce7e1a23a739f13e01fb1be377a", size = 9322986, upload-time = "2025-08-29T10:25:26.607Z" },
-    { url = "https://files.pythonhosted.org/packages/a2/62/92378eb1c2c565837ca3cb5f9569860d132ab9d195d7950c1ea2681dffd0/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:76cf6757c73a10ef10bf06fa937c0ec7393d90432f543f49adc8cab3fb6f26cb", size = 9276630, upload-time = "2025-08-29T10:25:28.349Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/f0/342d80457aa1cda7654327460f69db0d69405af1e4c453f4dc6ca7c4a76e/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:1626cb186e143720c62c6c6b5371e62bbc10af60481388c0da89bc903f37ea0c", size = 9547175, upload-time = "2025-08-29T10:25:29.989Z" },
-    { url = "https://files.pythonhosted.org/packages/14/84/8aa9b4adfc4fbd09381e20a5bc6aa27040c9c09caa89988c01544e008d18/tokenizers-0.22.0-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:da589a61cbfea18ae267723d6b029b84598dc8ca78db9951d8f5beff72d8507c", size = 9692735, upload-time = "2025-08-29T10:25:32.089Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/24/83ee2b1dc76bfe05c3142e7d0ccdfe69f0ad2f1ebf6c726cea7f0874c0d0/tokenizers-0.22.0-cp39-abi3-win32.whl", hash = "sha256:dbf9d6851bddae3e046fedfb166f47743c1c7bd11c640f0691dd35ef0bcad3be", size = 2471915, upload-time = "2025-08-29T10:25:36.411Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/9b/0e0bf82214ee20231845b127aa4a8015936ad5a46779f30865d10e404167/tokenizers-0.22.0-cp39-abi3-win_amd64.whl", hash = "sha256:c78174859eeaee96021f248a56c801e36bfb6bd5b067f2e95aa82445ca324f00", size = 2680494, upload-time = "2025-08-29T10:25:35.14Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/33/f4b2d94ada7ab297328fc671fed209368ddb82f965ec2224eb1892674c3a/tokenizers-0.22.1-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:59fdb013df17455e5f950b4b834a7b3ee2e0271e6378ccb33aa74d178b513c73", size = 3069318, upload-time = "2025-09-19T09:49:11.848Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/58/2aa8c874d02b974990e89ff95826a4852a8b2a273c7d1b4411cdd45a4565/tokenizers-0.22.1-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:8d4e484f7b0827021ac5f9f71d4794aaef62b979ab7608593da22b1d2e3c4edc", size = 2926478, upload-time = "2025-09-19T09:49:09.759Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/3b/55e64befa1e7bfea963cf4b787b2cea1011362c4193f5477047532ce127e/tokenizers-0.22.1-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:19d2962dd28bc67c1f205ab180578a78eef89ac60ca7ef7cbe9635a46a56422a", size = 3256994, upload-time = "2025-09-19T09:48:56.701Z" },
+    { url = "https://files.pythonhosted.org/packages/71/0b/fbfecf42f67d9b7b80fde4aabb2b3110a97fac6585c9470b5bff103a80cb/tokenizers-0.22.1-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:38201f15cdb1f8a6843e6563e6e79f4abd053394992b9bbdf5213ea3469b4ae7", size = 3153141, upload-time = "2025-09-19T09:48:59.749Z" },
+    { url = "https://files.pythonhosted.org/packages/17/a9/b38f4e74e0817af8f8ef925507c63c6ae8171e3c4cb2d5d4624bf58fca69/tokenizers-0.22.1-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d1cbe5454c9a15df1b3443c726063d930c16f047a3cc724b9e6e1a91140e5a21", size = 3508049, upload-time = "2025-09-19T09:49:05.868Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/48/dd2b3dac46bb9134a88e35d72e1aa4869579eacc1a27238f1577270773ff/tokenizers-0.22.1-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e7d094ae6312d69cc2a872b54b91b309f4f6fbce871ef28eb27b52a98e4d0214", size = 3710730, upload-time = "2025-09-19T09:49:01.832Z" },
+    { url = "https://files.pythonhosted.org/packages/93/0e/ccabc8d16ae4ba84a55d41345207c1e2ea88784651a5a487547d80851398/tokenizers-0.22.1-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:afd7594a56656ace95cdd6df4cca2e4059d294c5cfb1679c57824b605556cb2f", size = 3412560, upload-time = "2025-09-19T09:49:03.867Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/c6/dc3a0db5a6766416c32c034286d7c2d406da1f498e4de04ab1b8959edd00/tokenizers-0.22.1-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e2ef6063d7a84994129732b47e7915e8710f27f99f3a3260b8a38fc7ccd083f4", size = 3250221, upload-time = "2025-09-19T09:49:07.664Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/a6/2c8486eef79671601ff57b093889a345dd3d576713ef047776015dc66de7/tokenizers-0.22.1-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:ba0a64f450b9ef412c98f6bcd2a50c6df6e2443b560024a09fa6a03189726879", size = 9345569, upload-time = "2025-09-19T09:49:14.214Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/16/32ce667f14c35537f5f605fe9bea3e415ea1b0a646389d2295ec348d5657/tokenizers-0.22.1-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:331d6d149fa9c7d632cde4490fb8bbb12337fa3a0232e77892be656464f4b446", size = 9271599, upload-time = "2025-09-19T09:49:16.639Z" },
+    { url = "https://files.pythonhosted.org/packages/51/7c/a5f7898a3f6baa3fc2685c705e04c98c1094c523051c805cdd9306b8f87e/tokenizers-0.22.1-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:607989f2ea68a46cb1dfbaf3e3aabdf3f21d8748312dbeb6263d1b3b66c5010a", size = 9533862, upload-time = "2025-09-19T09:49:19.146Z" },
+    { url = "https://files.pythonhosted.org/packages/36/65/7e75caea90bc73c1dd8d40438adf1a7bc26af3b8d0a6705ea190462506e1/tokenizers-0.22.1-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:a0f307d490295717726598ef6fa4f24af9d484809223bbc253b201c740a06390", size = 9681250, upload-time = "2025-09-19T09:49:21.501Z" },
+    { url = "https://files.pythonhosted.org/packages/30/2c/959dddef581b46e6209da82df3b78471e96260e2bc463f89d23b1bf0e52a/tokenizers-0.22.1-cp39-abi3-win32.whl", hash = "sha256:b5120eed1442765cd90b903bb6cfef781fd8fe64e34ccaecbae4c619b7b12a82", size = 2472003, upload-time = "2025-09-19T09:49:27.089Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/46/e33a8c93907b631a99377ef4c5f817ab453d0b34f93529421f42ff559671/tokenizers-0.22.1-cp39-abi3-win_amd64.whl", hash = "sha256:65fd6e3fb11ca1e78a6a93602490f134d1fdeb13bcef99389d5102ea318ed138", size = 2674684, upload-time = "2025-09-19T09:49:24.953Z" },
 ]
 
 [[package]]
@@ -4429,6 +5210,45 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/ae/95/ae26263aceb3d57b821179f827d0e321373ed49423e603dd5906ab14a730/torch-2.8.0-cp39-none-macosx_11_0_arm64.whl", hash = "sha256:e9f071f5b52a9f6970dc8a919694b27a91ae9dc08898b2b988abbef5eddfd1ae", size = 73610795, upload-time = "2025-08-06T14:57:11.513Z" },
 ]
 
+[[package]]
+name = "torchaudio"
+version = "2.8.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "torch" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/95/c8/e46c22a3c059844bb0f1b670317c9e538b51728558326dcd9e5fffbf2ec2/torchaudio-2.8.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:d3c1b85b26a09832d139f6d6da6b66caeb51d2e16e08f8587665c44a9e1aa8f9", size = 1685620, upload-time = "2025-08-06T14:58:34.045Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/f5/69db76b564263f22c1788cc298ab1c4e2391a79fa8ba7b4a3e76d945292a/torchaudio-2.8.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:58f912bf2d289c709b42a55475b2b483becec79d9affb7684b606bb1f896b434", size = 4001714, upload-time = "2025-08-06T14:58:38.951Z" },
+    { url = "https://files.pythonhosted.org/packages/75/ca/da5d0a3bb7d114a8b590ecce14859ea0a05102bb4de68cdd1ed7a90634d6/torchaudio-2.8.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:4573c6042950c20278e3608a9a38050ba0bc72e0049e1bbfd249caf859a8029b", size = 1692033, upload-time = "2025-08-06T14:58:37.393Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/ef/62ac736d8f906cc414181050e08a495a637dab985186c34bd76ea37efbc0/torchaudio-2.8.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:776c0b4ba84b9e3ddf6304b9c47cd63549d7896a6f3d5184ece074cc3d76ed6b", size = 4011716, upload-time = "2025-08-06T14:58:40.138Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/0d/24dad878784f1edd62862f27173781669f0c71eb46368636787d1e364188/torchaudio-2.8.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:862e2e40bf09d865e5df080a84c1a39bbcef40e43140f4b1737eb3a389d3b38f", size = 1692930, upload-time = "2025-08-06T14:58:41.312Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/a6/84d80f34472503e9eb82245d7df501c59602d75d7360e717fb9b84f91c5e/torchaudio-2.8.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:93a8583f280fe83ba021aa713319381ea71362cc87b67ee38e97a43cb2254aee", size = 4014607, upload-time = "2025-08-06T14:58:47.234Z" },
+    { url = "https://files.pythonhosted.org/packages/76/34/89203e1c7c553e463b8f0fe6746842182652bff2208607853d59ba836001/torchaudio-2.8.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:ef7f7ffa828b8d8ba5d3a569b825fc0469688e1e8962bf6577d538bd8af1387d", size = 1685905, upload-time = "2025-08-06T14:58:48.547Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/a4/fab5dbbb2720fc775656b6a81f4daace069ff28a008a73d1b99e0fa995a2/torchaudio-2.8.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:a1f4bde9ce9316b6b24304c73ea9f82a2aabc4dbf8c6c2598117ea7c6ecc4db2", size = 4002156, upload-time = "2025-08-06T14:58:53.289Z" },
+]
+
+[[package]]
+name = "torchvision"
+version = "0.23.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pillow" },
+    { name = "torch" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/25/44/ddd56d1637bac42a8c5da2c8c440d8a28c431f996dd9790f32dd9a96ca6e/torchvision-0.23.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:31c583ba27426a3a04eca8c05450524105c1564db41be6632f7536ef405a6de2", size = 2394251, upload-time = "2025-08-06T14:58:01.725Z" },
+    { url = "https://files.pythonhosted.org/packages/93/f3/3cdf55bbf0f737304d997561c34ab0176222e0496b6743b0feab5995182c/torchvision-0.23.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:3932bf67256f2d095ce90a9f826f6033694c818856f4bb26794cf2ce64253e53", size = 8627497, upload-time = "2025-08-06T14:58:09.317Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/14/7b44fe766b7d11e064c539d92a172fa9689a53b69029e24f2f1f51e7dc56/torchvision-0.23.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:01dc33ee24c79148aee7cdbcf34ae8a3c9da1674a591e781577b716d233b1fa6", size = 2395543, upload-time = "2025-08-06T14:58:04.373Z" },
+    { url = "https://files.pythonhosted.org/packages/79/9c/fcb09aff941c8147d9e6aa6c8f67412a05622b0c750bcf796be4c85a58d4/torchvision-0.23.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:35c27941831b653f5101edfe62c03d196c13f32139310519e8228f35eae0e96a", size = 8628388, upload-time = "2025-08-06T14:58:07.802Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/00/2f6454decc0cd67158c7890364e446aad4b91797087a57a78e72e1a8f8bc/torchvision-0.23.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:6dd7c4d329a0e03157803031bc856220c6155ef08c26d4f5bbac938acecf0948", size = 2396614, upload-time = "2025-08-06T14:58:03.116Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/b5/3e580dcbc16f39a324f3dd71b90edbf02a42548ad44d2b4893cc92b1194b/torchvision-0.23.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:4e7d31c43bc7cbecbb1a5652ac0106b436aa66e26437585fc2c4b2cf04d6014c", size = 8627108, upload-time = "2025-08-06T14:58:12.956Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/e2/aafc6af854e792d212ff58e459f8d5d807568dc3f2b49ec41b677275e5a9/torchvision-0.23.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:6c74cbc1cbee26dd4f35f989cd80dccc40411f258dee476b29871dee4b483af0", size = 2392870, upload-time = "2025-08-06T14:58:21.303Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/06/09b6a917b3759ef000428af0aa2597f983e20d9fbbcfeb826750f778fe6d/torchvision-0.23.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:a9e9d7552d34547b80843eaf64ab0737b19b2e8bec2514286b8cfd30861ca8b5", size = 8630400, upload-time = "2025-08-06T14:58:24.139Z" },
+]
+
 [[package]]
 name = "tqdm"
 version = "4.67.1"
@@ -4443,7 +5263,7 @@ wheels = [
 
 [[package]]
 name = "transformers"
-version = "4.56.0"
+version = "4.57.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "filelock" },
@@ -4458,9 +5278,9 @@ dependencies = [
     { name = "tokenizers" },
     { name = "tqdm" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/e9/40/2898c1a5f8b2fff2f94dffd856d6793320e590e1cfdc3a1453c31c364094/transformers-4.56.0.tar.gz", hash = "sha256:6ca9c3f38aa4da93ebf877db7156368c1c188c7465f09dbe70951e7622e987fa", size = 9840426, upload-time = "2025-08-29T18:23:46.247Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/5c/a22c39dac2687f3fe2a6b97e2c1ae516e91cd4d3976a7a2b7c24ff2fae48/transformers-4.57.0.tar.gz", hash = "sha256:d045753f3d93f9216e693cdb168698dfd2e9d3aad1bb72579a5d60ebf1545a8b", size = 10142956, upload-time = "2025-10-03T17:03:47.177Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/98/41/6ba7c2eafa839069f1f949c30cd22e791b14b90c7af8cb0c65cc47702dca/transformers-4.56.0-py3-none-any.whl", hash = "sha256:bacf539c38dd850690856881c4974321af93a22f2ee96bcc994741a2121d8e71", size = 11607938, upload-time = "2025-08-29T18:23:33.92Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/2b/4d2708ac1ff5cd708b6548f4c5812d0ae40d1c28591c4c1c762b6dbdef2d/transformers-4.57.0-py3-none-any.whl", hash = "sha256:9d7c6d098c026e40d897e017ed1f481ab803cbac041021dbc6ae6100e4949b55", size = 11990588, upload-time = "2025-10-03T17:03:43.629Z" },
 ]
 
 [[package]]
@@ -4490,6 +5310,22 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/12/34/1251beb5a3cb93f3950ebe68732752014646003ef6eb11eb5f1a37ca78cd/triton-3.4.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:98e5c1442eaeabae2e2452ae765801bd53cd4ce873cab0d1bdd59a32ab2d9397", size = 155430799, upload-time = "2025-07-30T19:58:57.664Z" },
 ]
 
+[[package]]
+name = "typer"
+version = "0.19.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "rich" },
+    { name = "shellingham" },
+    { name = "typing-extensions" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/21/ca/950278884e2ca20547ff3eb109478c6baf6b8cf219318e6bc4f666fad8e8/typer-0.19.2.tar.gz", hash = "sha256:9ad824308ded0ad06cc716434705f691d4ee0bfd0fb081839d2e426860e7fdca", size = 104755, upload-time = "2025-09-23T09:47:48.256Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/00/22/35617eee79080a5d071d0f14ad698d325ee6b3bf824fc0467c03b30e7fa8/typer-0.19.2-py3-none-any.whl", hash = "sha256:755e7e19670ffad8283db353267cb81ef252f595aa6834a0d1ca9312d9326cb9", size = 46748, upload-time = "2025-09-23T09:47:46.777Z" },
+]
+
 [[package]]
 name = "typing-extensions"
 version = "4.15.0"
@@ -4501,14 +5337,14 @@ wheels = [
 
 [[package]]
 name = "typing-inspection"
-version = "0.4.1"
+version = "0.4.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f8/b1/0c11f5058406b3af7609f121aaa6b609744687f1d158b3c3a5bf4cc94238/typing_inspection-0.4.1.tar.gz", hash = "sha256:6ae134cc0203c33377d43188d4064e9b357dba58cff3185f22924610e70a9d28", size = 75726, upload-time = "2025-05-21T18:55:23.885Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/55/e3/70399cb7dd41c10ac53367ae42139cf4b1ca5f36bb3dc6c9d33acdb43655/typing_inspection-0.4.2.tar.gz", hash = "sha256:ba561c48a67c5958007083d386c3295464928b01faa735ab8547c5692e87f464", size = 75949, upload-time = "2025-10-01T02:14:41.687Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/17/69/cd203477f944c353c31bade965f880aa1061fd6bf05ded0726ca845b6ff7/typing_inspection-0.4.1-py3-none-any.whl", hash = "sha256:389055682238f53b04f7badcb49b989835495a96700ced5dab2d8feae4b26f51", size = 14552, upload-time = "2025-05-21T18:55:22.152Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/9b/47798a6c91d8bdb567fe2698fe81e0c6b7cb7ef4d13da4114b41d239f65d/typing_inspection-0.4.2-py3-none-any.whl", hash = "sha256:4ed1cacbdc298c220f1bd249ed5287caa16f34d44ef4e9c3d0cbad5b521545e7", size = 14611, upload-time = "2025-10-01T02:14:40.154Z" },
 ]
 
 [[package]]
@@ -4531,17 +5367,52 @@ wheels = [
 
 [[package]]
 name = "uvicorn"
-version = "0.35.0"
+version = "0.37.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "h11" },
     { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/5e/42/e0e305207bb88c6b8d3061399c6a961ffe5fbb7e2aa63c9234df7259e9cd/uvicorn-0.35.0.tar.gz", hash = "sha256:bc662f087f7cf2ce11a1d7fd70b90c9f98ef2e2831556dd078d131b96cc94a01", size = 78473, upload-time = "2025-06-28T16:15:46.058Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/71/57/1616c8274c3442d802621abf5deb230771c7a0fec9414cb6763900eb3868/uvicorn-0.37.0.tar.gz", hash = "sha256:4115c8add6d3fd536c8ee77f0e14a7fd2ebba939fed9b02583a97f80648f9e13", size = 80367, upload-time = "2025-09-23T13:33:47.486Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d2/e2/dc81b1bd1dcfe91735810265e9d26bc8ec5da45b4c0f6237e286819194c3/uvicorn-0.35.0-py3-none-any.whl", hash = "sha256:197535216b25ff9b785e29a0b79199f55222193d47f820816e7da751e9bc8d4a", size = 66406, upload-time = "2025-06-28T16:15:44.816Z" },
+    { url = "https://files.pythonhosted.org/packages/85/cd/584a2ceb5532af99dd09e50919e3615ba99aa127e9850eafe5f31ddfdb9a/uvicorn-0.37.0-py3-none-any.whl", hash = "sha256:913b2b88672343739927ce381ff9e2ad62541f9f8289664fa1d1d3803fa2ce6c", size = 67976, upload-time = "2025-09-23T13:33:45.842Z" },
+]
+
+[package.optional-dependencies]
+standard = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "httptools" },
+    { name = "python-dotenv" },
+    { name = "pyyaml" },
+    { name = "uvloop", marker = "platform_python_implementation != 'PyPy' and sys_platform != 'cygwin' and sys_platform != 'win32'" },
+    { name = "watchfiles" },
+    { name = "websockets" },
+]
+
+[[package]]
+name = "uvloop"
+version = "0.21.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/af/c0/854216d09d33c543f12a44b393c402e89a920b1a0a7dc634c42de91b9cf6/uvloop-0.21.0.tar.gz", hash = "sha256:3bf12b0fda68447806a7ad847bfa591613177275d35b6724b1ee573faa3704e3", size = 2492741, upload-time = "2024-10-14T23:38:35.489Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f3/96/63695e0ebd7da6c741ccd4489b5947394435e198a1382349c17b1146bb97/uvloop-0.21.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f38b2e090258d051d68a5b14d1da7203a3c3677321cf32a95a6f4db4dd8b6f26", size = 3827735, upload-time = "2024-10-14T23:37:25.129Z" },
+    { url = "https://files.pythonhosted.org/packages/61/e0/f0f8ec84979068ffae132c58c79af1de9cceeb664076beea86d941af1a30/uvloop-0.21.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:87c43e0f13022b998eb9b973b5e97200c8b90823454d4bc06ab33829e09fb9bb", size = 3825126, upload-time = "2024-10-14T23:37:27.59Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/fe/5e94a977d058a54a19df95f12f7161ab6e323ad49f4dabc28822eb2df7ea/uvloop-0.21.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:10d66943def5fcb6e7b37310eb6b5639fd2ccbc38df1177262b0640c3ca68c1f", size = 3705789, upload-time = "2024-10-14T23:37:29.385Z" },
+    { url = "https://files.pythonhosted.org/packages/26/dd/c7179618e46092a77e036650c1f056041a028a35c4d76945089fcfc38af8/uvloop-0.21.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:67dd654b8ca23aed0a8e99010b4c34aca62f4b7fce88f39d452ed7622c94845c", size = 3800523, upload-time = "2024-10-14T23:37:32.048Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/ea/0bfae1aceb82a503f358d8d2fa126ca9dbdb2ba9c7866974faec1cb5875c/uvloop-0.21.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b9fb766bb57b7388745d8bcc53a359b116b8a04c83a2288069809d2b3466c37e", size = 3960855, upload-time = "2024-10-14T23:37:37.683Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/ca/0864176a649838b838f36d44bf31c451597ab363b60dc9e09c9630619d41/uvloop-0.21.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a375441696e2eda1c43c44ccb66e04d61ceeffcd76e4929e527b7fa401b90fb", size = 3973185, upload-time = "2024-10-14T23:37:40.226Z" },
+    { url = "https://files.pythonhosted.org/packages/30/bf/08ad29979a936d63787ba47a540de2132169f140d54aa25bc8c3df3e67f4/uvloop-0.21.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:baa0e6291d91649c6ba4ed4b2f982f9fa165b5bbd50a9e203c416a2797bab3c6", size = 3820256, upload-time = "2024-10-14T23:37:42.839Z" },
+    { url = "https://files.pythonhosted.org/packages/da/e2/5cf6ef37e3daf2f06e651aae5ea108ad30df3cb269102678b61ebf1fdf42/uvloop-0.21.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4509360fcc4c3bd2c70d87573ad472de40c13387f5fda8cb58350a1d7475e58d", size = 3937323, upload-time = "2024-10-14T23:37:45.337Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/ef/a02ec5da49909dbbfb1fd205a9a1ac4e88ea92dcae885e7c961847cd51e2/uvloop-0.21.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:baa4dcdbd9ae0a372f2167a207cd98c9f9a1ea1188a8a526431eef2f8116cc8d", size = 4580089, upload-time = "2024-10-14T23:37:51.703Z" },
+    { url = "https://files.pythonhosted.org/packages/06/a7/b4e6a19925c900be9f98bec0a75e6e8f79bb53bdeb891916609ab3958967/uvloop-0.21.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86975dca1c773a2c9864f4c52c5a55631038e387b47eaf56210f873887b6c8dc", size = 4693770, upload-time = "2024-10-14T23:37:54.122Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/0c/f07435a18a4b94ce6bd0677d8319cd3de61f3a9eeb1e5f8ab4e8b5edfcb3/uvloop-0.21.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:461d9ae6660fbbafedd07559c6a2e57cd553b34b0065b6550685f6653a98c1cb", size = 4451321, upload-time = "2024-10-14T23:37:55.766Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/eb/f7032be105877bcf924709c97b1bf3b90255b4ec251f9340cef912559f28/uvloop-0.21.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:183aef7c8730e54c9a3ee3227464daed66e37ba13040bb3f350bc2ddc040f22f", size = 4659022, upload-time = "2024-10-14T23:37:58.195Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/1a/9fbc2b1543d0df11f7aed1632f64bdf5ecc4053cf98cdc9edb91a65494f9/uvloop-0.21.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:53e420a3afe22cdcf2a0f4846e377d16e718bc70103d7088a4f7623567ba5fb0", size = 3829660, upload-time = "2024-10-14T23:38:27.905Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/c0/392e235e4100ae3b95b5c6dac77f82b529d2760942b1e7e0981e5d8e895d/uvloop-0.21.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88cb67cdbc0e483da00af0b2c3cdad4b7c61ceb1ee0f33fe00e09c81e3a6cb75", size = 3827185, upload-time = "2024-10-14T23:38:29.458Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/24/a5da6aba58f99aed5255eca87d58d1760853e8302d390820cc29058408e3/uvloop-0.21.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:221f4f2a1f46032b403bf3be628011caf75428ee3cc204a22addf96f586b19fd", size = 3705833, upload-time = "2024-10-14T23:38:31.155Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/5c/6ba221bb60f1e6474474102e17e38612ec7a06dc320e22b687ab563d877f/uvloop-0.21.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:2d1f581393673ce119355d56da84fe1dd9d2bb8b3d13ce792524e1607139feff", size = 3804696, upload-time = "2024-10-14T23:38:33.633Z" },
 ]
 
 [[package]]
@@ -4561,24 +5432,75 @@ wheels = [
 
 [[package]]
 name = "vllm"
-version = "0.1.2"
+version = "0.11.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "fastapi" },
-    { name = "fschat" },
+    { name = "aiohttp" },
+    { name = "blake3" },
+    { name = "cachetools" },
+    { name = "cbor2" },
+    { name = "cloudpickle" },
+    { name = "compressed-tensors" },
+    { name = "depyf" },
+    { name = "diskcache" },
+    { name = "einops" },
+    { name = "fastapi", extra = ["standard"] },
+    { name = "filelock" },
+    { name = "gguf" },
+    { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
+    { name = "lark" },
+    { name = "llguidance", marker = "platform_machine == 'aarch64' or platform_machine == 'arm64' or platform_machine == 'x86_64'" },
+    { name = "lm-format-enforcer" },
+    { name = "mistral-common", extra = ["audio", "image"] },
+    { name = "msgspec" },
     { name = "ninja" },
+    { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "openai" },
+    { name = "openai-harmony" },
+    { name = "opencv-python-headless" },
+    { name = "outlines-core" },
+    { name = "partial-json-parser" },
+    { name = "pillow" },
+    { name = "prometheus-client" },
+    { name = "prometheus-fastapi-instrumentator" },
+    { name = "protobuf" },
     { name = "psutil" },
+    { name = "py-cpuinfo" },
+    { name = "pybase64" },
     { name = "pydantic" },
-    { name = "ray" },
+    { name = "python-json-logger" },
+    { name = "pyyaml" },
+    { name = "pyzmq" },
+    { name = "ray", extra = ["cgraph"] },
+    { name = "regex" },
+    { name = "requests" },
+    { name = "scipy", version = "1.13.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "scipy", version = "1.15.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "scipy", version = "1.16.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "sentencepiece" },
+    { name = "setproctitle" },
+    { name = "setuptools", marker = "python_full_version >= '3.12'" },
+    { name = "six", marker = "python_full_version >= '3.12'" },
+    { name = "tiktoken" },
+    { name = "tokenizers" },
     { name = "torch" },
+    { name = "torchaudio" },
+    { name = "torchvision" },
+    { name = "tqdm" },
     { name = "transformers" },
-    { name = "uvicorn" },
-    { name = "xformers" },
+    { name = "typing-extensions" },
+    { name = "watchfiles" },
+    { name = "xformers", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "xgrammar", marker = "platform_machine == 'aarch64' or platform_machine == 'arm64' or platform_machine == 'x86_64'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/82/5a/36d2351206f4d8d871b10780f874d03957985e08298d430cc837723e07af/vllm-0.11.0.tar.gz", hash = "sha256:f435a64c24e9c4178d657a76f8edd8548ddc444012f7d06a9f79ac3a6392bfae", size = 10822208, upload-time = "2025-10-04T01:39:57.798Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/47/33/d19e0763c34392ec956534536fa837c060495bfff31ed83452135ea7608d/vllm-0.11.0-cp38-abi3-manylinux1_x86_64.whl", hash = "sha256:3861c75ff2b12e24f6d179ff5c084d791b42ded8675d76c8706697c79f68cd62", size = 438217982, upload-time = "2025-10-04T01:39:32.382Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/bf/973444bb959fc7acbbeb3d226bd4d135dcd49b6af174b29aab1b50e2d710/vllm-0.11.0-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:52369c9ee949944354bdc7afc88ded2d1ed02b098bf90db06cf80098a19787b7", size = 401003969, upload-time = "2025-10-04T01:39:50.251Z" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f9/ed/1e4625f290e24c1381aeeb66620b32755b17f2894a4227fb49fb20fa0ec9/vllm-0.1.2.tar.gz", hash = "sha256:6fff59a918ec822c3d7a9060c3af5604a1bbf26ec4db15656678768019aefad7", size = 93849, upload-time = "2023-07-05T04:53:21.847Z" }
 
 [[package]]
 name = "watchdog"
@@ -4615,23 +5537,52 @@ wheels = [
 ]
 
 [[package]]
-name = "wavedrom"
-version = "2.0.3.post3"
+name = "watchfiles"
+version = "1.1.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "pyyaml" },
-    { name = "six" },
-    { name = "svgwrite" },
+    { name = "anyio" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/be/71/6739e3abac630540aaeaaece4584c39f88b5f8658ce6ca517efec455e3de/wavedrom-2.0.3.post3.tar.gz", hash = "sha256:327b4d5dca593c81257c202fea516f7a908747fb11527c359f034f5b7af7f47b", size = 137737, upload-time = "2022-05-27T11:53:09.675Z" }
-
-[[package]]
-name = "wcwidth"
-version = "0.2.13"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/6c/63/53559446a878410fc5a5974feb13d31d78d752eb18aeba59c7fef1af7598/wcwidth-0.2.13.tar.gz", hash = "sha256:72ea0c06399eb286d978fdedb6923a9eb47e1c486ce63e9b4e64fc18303972b5", size = 101301, upload-time = "2024-01-06T02:10:57.829Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/fd/84/fd2ba7aafacbad3c4201d395674fc6348826569da3c0937e75505ead3528/wcwidth-0.2.13-py2.py3-none-any.whl", hash = "sha256:3da69048e4540d84af32131829ff948f1e022c1c6bdb8d6102117aac784f6859", size = 34166, upload-time = "2024-01-06T02:10:55.763Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/2a/9a/d451fcc97d029f5812e898fd30a53fd8c15c7bbd058fd75cfc6beb9bd761/watchfiles-1.1.0.tar.gz", hash = "sha256:693ed7ec72cbfcee399e92c895362b6e66d63dac6b91e2c11ae03d10d503e575", size = 94406, upload-time = "2025-06-15T19:06:59.42Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/06/be/503514656d0555ec2195f60d810eca29b938772e9bfb112d5cd5ad6f6a9e/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8412eacef34cae2836d891836a7fff7b754d6bcac61f6c12ba5ca9bc7e427b68", size = 450739, upload-time = "2025-06-15T19:04:54.203Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/0d/a05dd9e5f136cdc29751816d0890d084ab99f8c17b86f25697288ca09bc7/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:df670918eb7dd719642e05979fc84704af913d563fd17ed636f7c4783003fdcc", size = 458106, upload-time = "2025-06-15T19:04:55.607Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/fa/9cd16e4dfdb831072b7ac39e7bea986e52128526251038eb481effe9f48e/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d7642b9bc4827b5518ebdb3b82698ada8c14c7661ddec5fe719f3e56ccd13c97", size = 484264, upload-time = "2025-06-15T19:04:57.009Z" },
+    { url = "https://files.pythonhosted.org/packages/32/04/1da8a637c7e2b70e750a0308e9c8e662ada0cca46211fa9ef24a23937e0b/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:199207b2d3eeaeb80ef4411875a6243d9ad8bc35b07fc42daa6b801cc39cc41c", size = 597612, upload-time = "2025-06-15T19:04:58.409Z" },
+    { url = "https://files.pythonhosted.org/packages/30/01/109f2762e968d3e58c95731a206e5d7d2a7abaed4299dd8a94597250153c/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a479466da6db5c1e8754caee6c262cd373e6e6c363172d74394f4bff3d84d7b5", size = 477242, upload-time = "2025-06-15T19:04:59.786Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/b8/46f58cf4969d3b7bc3ca35a98e739fa4085b0657a1540ccc29a1a0bc016f/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:935f9edd022ec13e447e5723a7d14456c8af254544cefbc533f6dd276c9aa0d9", size = 453148, upload-time = "2025-06-15T19:05:01.103Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/cd/8267594263b1770f1eb76914940d7b2d03ee55eca212302329608208e061/watchfiles-1.1.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:8076a5769d6bdf5f673a19d51da05fc79e2bbf25e9fe755c47595785c06a8c72", size = 626574, upload-time = "2025-06-15T19:05:02.582Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/2f/7f2722e85899bed337cba715723e19185e288ef361360718973f891805be/watchfiles-1.1.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:86b1e28d4c37e89220e924305cd9f82866bb0ace666943a6e4196c5df4d58dcc", size = 624378, upload-time = "2025-06-15T19:05:03.719Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/a2/8afa359ff52e99af1632f90cbf359da46184207e893a5f179301b0c8d6df/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cb45350fd1dc75cd68d3d72c47f5b513cb0578da716df5fba02fff31c69d5f2d", size = 450792, upload-time = "2025-06-15T19:05:09.869Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/bf/7446b401667f5c64972a57a0233be1104157fc3abf72c4ef2666c1bd09b2/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:11ee4444250fcbeb47459a877e5e80ed994ce8e8d20283857fc128be1715dac7", size = 458196, upload-time = "2025-06-15T19:05:11.91Z" },
+    { url = "https://files.pythonhosted.org/packages/58/2f/501ddbdfa3fa874ea5597c77eeea3d413579c29af26c1091b08d0c792280/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bda8136e6a80bdea23e5e74e09df0362744d24ffb8cd59c4a95a6ce3d142f79c", size = 484788, upload-time = "2025-06-15T19:05:13.373Z" },
+    { url = "https://files.pythonhosted.org/packages/61/1e/9c18eb2eb5c953c96bc0e5f626f0e53cfef4bd19bd50d71d1a049c63a575/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b915daeb2d8c1f5cee4b970f2e2c988ce6514aace3c9296e58dd64dc9aa5d575", size = 597879, upload-time = "2025-06-15T19:05:14.725Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/6c/1467402e5185d89388b4486745af1e0325007af0017c3384cc786fff0542/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ed8fc66786de8d0376f9f913c09e963c66e90ced9aa11997f93bdb30f7c872a8", size = 477447, upload-time = "2025-06-15T19:05:15.775Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/a1/ec0a606bde4853d6c4a578f9391eeb3684a9aea736a8eb217e3e00aa89a1/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fe4371595edf78c41ef8ac8df20df3943e13defd0efcb732b2e393b5a8a7a71f", size = 453145, upload-time = "2025-06-15T19:05:17.17Z" },
+    { url = "https://files.pythonhosted.org/packages/90/b9/ef6f0c247a6a35d689fc970dc7f6734f9257451aefb30def5d100d6246a5/watchfiles-1.1.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:b7c5f6fe273291f4d414d55b2c80d33c457b8a42677ad14b4b47ff025d0893e4", size = 626539, upload-time = "2025-06-15T19:05:18.557Z" },
+    { url = "https://files.pythonhosted.org/packages/34/44/6ffda5537085106ff5aaa762b0d130ac6c75a08015dd1621376f708c94de/watchfiles-1.1.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:7738027989881e70e3723c75921f1efa45225084228788fc59ea8c6d732eb30d", size = 624472, upload-time = "2025-06-15T19:05:19.588Z" },
+    { url = "https://files.pythonhosted.org/packages/86/50/dee79968566c03190677c26f7f47960aff738d32087087bdf63a5473e7df/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:130fc497b8ee68dce163e4254d9b0356411d1490e868bd8790028bc46c5cc297", size = 450939, upload-time = "2025-06-15T19:05:26.494Z" },
+    { url = "https://files.pythonhosted.org/packages/40/45/a7b56fb129700f3cfe2594a01aa38d033b92a33dddce86c8dfdfc1247b72/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:50a51a90610d0845a5931a780d8e51d7bd7f309ebc25132ba975aca016b576a0", size = 457270, upload-time = "2025-06-15T19:05:27.466Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/c8/fa5ef9476b1d02dc6b5e258f515fcaaecf559037edf8b6feffcbc097c4b8/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:dc44678a72ac0910bac46fa6a0de6af9ba1355669b3dfaf1ce5f05ca7a74364e", size = 483370, upload-time = "2025-06-15T19:05:28.548Z" },
+    { url = "https://files.pythonhosted.org/packages/98/68/42cfcdd6533ec94f0a7aab83f759ec11280f70b11bfba0b0f885e298f9bd/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a543492513a93b001975ae283a51f4b67973662a375a403ae82f420d2c7205ee", size = 598654, upload-time = "2025-06-15T19:05:29.997Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/74/b2a1544224118cc28df7e59008a929e711f9c68ce7d554e171b2dc531352/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8ac164e20d17cc285f2b94dc31c384bc3aa3dd5e7490473b3db043dd70fbccfd", size = 478667, upload-time = "2025-06-15T19:05:31.172Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/77/e3362fe308358dc9f8588102481e599c83e1b91c2ae843780a7ded939a35/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f7590d5a455321e53857892ab8879dce62d1f4b04748769f5adf2e707afb9d4f", size = 452213, upload-time = "2025-06-15T19:05:32.299Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/17/c8f1a36540c9a1558d4faf08e909399e8133599fa359bf52ec8fcee5be6f/watchfiles-1.1.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:37d3d3f7defb13f62ece99e9be912afe9dd8a0077b7c45ee5a57c74811d581a4", size = 626718, upload-time = "2025-06-15T19:05:33.415Z" },
+    { url = "https://files.pythonhosted.org/packages/26/45/fb599be38b4bd38032643783d7496a26a6f9ae05dea1a42e58229a20ac13/watchfiles-1.1.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:7080c4bb3efd70a07b1cc2df99a7aa51d98685be56be6038c3169199d0a1c69f", size = 623098, upload-time = "2025-06-15T19:05:34.534Z" },
+    { url = "https://files.pythonhosted.org/packages/79/e6/e10d5675af653b1b07d4156906858041149ca222edaf8995877f2605ba9e/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:aa0cc8365ab29487eb4f9979fd41b22549853389e22d5de3f134a6796e1b05a4", size = 451196, upload-time = "2025-06-15T19:06:32.435Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/8a/facd6988100cd0f39e89f6c550af80edb28e3a529e1ee662e750663e6b36/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:90ebb429e933645f3da534c89b29b665e285048973b4d2b6946526888c3eb2c7", size = 458218, upload-time = "2025-06-15T19:06:33.503Z" },
+    { url = "https://files.pythonhosted.org/packages/90/26/34cbcbc4d0f2f8f9cc243007e65d741ae039f7a11ef8ec6e9cd25bee08d1/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c588c45da9b08ab3da81d08d7987dae6d2a3badd63acdb3e206a42dbfa7cb76f", size = 484851, upload-time = "2025-06-15T19:06:34.541Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/1f/f59faa9fc4b0e36dbcdd28a18c430416443b309d295d8b82e18192d120ad/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7c55b0f9f68590115c25272b06e63f0824f03d4fc7d6deed43d8ad5660cabdbf", size = 599520, upload-time = "2025-06-15T19:06:35.785Z" },
+    { url = "https://files.pythonhosted.org/packages/83/72/3637abecb3bf590529f5154ca000924003e5f4bbb9619744feeaf6f0b70b/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cd17a1e489f02ce9117b0de3c0b1fab1c3e2eedc82311b299ee6b6faf6c23a29", size = 477956, upload-time = "2025-06-15T19:06:36.965Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/f3/d14ffd9acc0c1bd4790378995e320981423263a5d70bd3929e2e0dc87fff/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:da71945c9ace018d8634822f16cbc2a78323ef6c876b1d34bbf5d5222fd6a72e", size = 453196, upload-time = "2025-06-15T19:06:38.024Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/38/78ad77bd99e20c0fdc82262be571ef114fc0beef9b43db52adb939768c38/watchfiles-1.1.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:51556d5004887045dba3acdd1fdf61dddea2be0a7e18048b5e853dcd37149b86", size = 627479, upload-time = "2025-06-15T19:06:39.442Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/cf/549d50a22fcc83f1017c6427b1c76c053233f91b526f4ad7a45971e70c0b/watchfiles-1.1.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:04e4ed5d1cd3eae68c89bcc1a485a109f39f2fd8de05f705e98af6b5f1861f1f", size = 624414, upload-time = "2025-06-15T19:06:40.859Z" },
+    { url = "https://files.pythonhosted.org/packages/70/9c/9a6a42e97f92eeed77c3485a43ea96723900aefa3ac739a8c73f4bff2cd7/watchfiles-1.1.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a2726d7bfd9f76158c84c10a409b77a320426540df8c35be172444394b17f7ea", size = 451528, upload-time = "2025-06-15T19:06:46.791Z" },
+    { url = "https://files.pythonhosted.org/packages/51/7b/98c7f4f7ce7ff03023cf971cd84a3ee3b790021ae7584ffffa0eb2554b96/watchfiles-1.1.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:df32d59cb9780f66d165a9a7a26f19df2c7d24e3bd58713108b41d0ff4f929c6", size = 454095, upload-time = "2025-06-15T19:06:48.211Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/fa/12269467b2fc006f8fce4cd6c3acfa77491dd0777d2a747415f28ccc8c60/watchfiles-1.1.0-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f2bcdc54ea267fe72bfc7d83c041e4eb58d7d8dc6f578dfddb52f037ce62f432", size = 451584, upload-time = "2025-06-15T19:06:51.834Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/d3/254cea30f918f489db09d6a8435a7de7047f8cb68584477a515f160541d6/watchfiles-1.1.0-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:923fec6e5461c42bd7e3fd5ec37492c6f3468be0499bc0707b4bbbc16ac21792", size = 454009, upload-time = "2025-06-15T19:06:52.896Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/dc/6f324a6f32c5ab73b54311b5f393a79df34c1584b8d2404cf7e6d780aa5d/watchfiles-1.1.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:54062ef956807ba806559b3c3d52105ae1827a0d4ab47b621b31132b6b7e2866", size = 451787, upload-time = "2025-06-15T19:06:56.998Z" },
+    { url = "https://files.pythonhosted.org/packages/45/5d/1d02ef4caa4ec02389e72d5594cdf9c67f1800a7c380baa55063c30c6598/watchfiles-1.1.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7a7bd57a1bb02f9d5c398c0c1675384e7ab1dd39da0ca50b7f09af45fa435277", size = 454272, upload-time = "2025-06-15T19:06:58.055Z" },
 ]
 
 [[package]]
@@ -4771,21 +5722,21 @@ wheels = [
 
 [[package]]
 name = "xformers"
-version = "0.0.32.post2"
+version = "0.0.32.post1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "torch" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/47/99/68d2e7a17b65a08180d6b120d1d293d5855a0c5999cb30a79fb466b49642/xformers-0.0.32.post2.tar.gz", hash = "sha256:9538be803969c6e1ca16a3ece921e472c24f79970b10be1087a389dcb66e412a", size = 12105990, upload-time = "2025-08-15T11:48:56.584Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/6f/33/3b9c4d3d5b2da453d27de891df4ad653ac5795324961aa3a5c15b0353fe6/xformers-0.0.32.post1.tar.gz", hash = "sha256:1de84a45c497c8d92326986508d81f4b0a8c6be4d3d62a29b8ad6048a6ab51e1", size = 12106196, upload-time = "2025-08-14T18:07:45.486Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ce/16/4ed2fbf7c20ba20565d9a1fc544c2bef39d409c8eca1949654c6ff9f8452/xformers-0.0.32.post2-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:3d1fefe0006eb00a730b3b197ff41aab085f1209b50419baab6152445f09e508", size = 117196673, upload-time = "2025-08-15T11:48:46.726Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/df/6817346f1a77278315d5fe1fc9f239ba3282ba36e8ab3256babd448dde62/xformers-0.0.32.post1-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:5f245b5555188da112070d8fefb6b7ae1ae47422856521d66c837e9d2352fbe4", size = 117199943, upload-time = "2025-08-14T18:07:34.78Z" },
 ]
 
 [[package]]
 name = "xgrammar"
-version = "0.1.23"
+version = "0.1.25"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
@@ -4798,188 +5749,179 @@ dependencies = [
     { name = "triton", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/e9/12/958457553a87c31bdb18c8395b88bb3255f7c7373ab3a0b046d3b7f37f86/xgrammar-0.1.23.tar.gz", hash = "sha256:5ef280455c1ac008f052d7ea92286f0ca3a3d7ab360224894ac69277c8827113", size = 2263693, upload-time = "2025-08-15T07:31:42.792Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/95/f0/bdd2c9b97ad8db283c702119a23f3e338a3b9a5a0f6a7469d6d65938fe5a/xgrammar-0.1.23-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:7a4f47812ea93f79634b2ab14580ccd20893e9c81ff43929e1ea1a6b32e56015", size = 620720, upload-time = "2025-08-15T07:31:06.428Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/ee/bdd8783903437807856cbacd23ba48b4eea1b8b4e54057103961031dbd7b/xgrammar-0.1.23-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a40d28e5011f4e2a9d874ad00f7c4b3c1e816690b3f2d25904c64fd0c32cbbf9", size = 564940, upload-time = "2025-08-15T07:31:08.271Z" },
-    { url = "https://files.pythonhosted.org/packages/43/c2/d18e5af82cd18c2ff908f2438c5329e27ff6af3c11656bfb2582e96d8700/xgrammar-0.1.23-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c5439d9a7819b98bb82a699e63e9b233569eacd1d8d0d5366f4cca7d61cc2742", size = 7680078, upload-time = "2025-08-15T07:31:09.922Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/83/9fd0d2d4e06ef27b13f9343269c0110574622c543c6527d7dcafc55a194e/xgrammar-0.1.23-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:41d262c48354f2d845e8654c05be681c17fb9fc037cde6883ef434a07b1e9ace", size = 7854128, upload-time = "2025-08-15T07:31:11.458Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/b7/0a328254411983f223baa33f10b54793570512e0868e6ba7b47cf2c7caae/xgrammar-0.1.23-cp310-cp310-win_amd64.whl", hash = "sha256:b2a40e4dd242dedc83c52f9699569efe04b448bd1cd06627d12a7093b4d800ed", size = 646685, upload-time = "2025-08-15T07:31:13.218Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/d5/b5b2dde17b1fbda9240af6cbb19523704eccf876610331c3317c08654106/xgrammar-0.1.23-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:af31d12105e31afa257870e9016da3c631300d040e63684b5aabae074916f552", size = 620555, upload-time = "2025-08-15T07:31:15.217Z" },
-    { url = "https://files.pythonhosted.org/packages/04/55/b92e490ddb7f510fbd2d8e2b2fa5a8da440cfb80221cb16a9a3d5f33322c/xgrammar-0.1.23-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ba90f257f75f12014360d9512f95caf731b18efe0562f91206b44b28388533d6", size = 564886, upload-time = "2025-08-15T07:31:16.44Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/61/07dd19e914512fbb9cb08a3e0479f75fa827c325abc23cf9fef48dd42515/xgrammar-0.1.23-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8612ed64bcf86526837df9c4c1061a2b00a39a51cb1ab1fe61f881c1c2ab5f1e", size = 7681629, upload-time = "2025-08-15T07:31:17.795Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/d1/500caef952bea84f335453fee01836d1b8c8b700b173e07a37c6a4b62d67/xgrammar-0.1.23-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:280d6e114b6ea57ee9dc4b5e31b5f897132e74d279b8064e7679893804498652", size = 7855408, upload-time = "2025-08-15T07:31:19.647Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/55/084f9ec87174220914f7b96c59f20e64ae2bc7938765c8b1d0191ae45d90/xgrammar-0.1.23-cp311-cp311-win_amd64.whl", hash = "sha256:a21b9f85fa321a2731106ffc167fde15d90e5af1779b64bc75c36fa809ea7925", size = 646693, upload-time = "2025-08-15T07:31:21.072Z" },
-    { url = "https://files.pythonhosted.org/packages/37/40/4fd2f41c603bbe4e60d51c7956ace6f386dd69b20ad62c443a1dbcac279a/xgrammar-0.1.23-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:6605ae8f7d7748ac00b55f4377ddc895bc9b7f8252760b1265145be9a2bc87e4", size = 620111, upload-time = "2025-08-15T07:31:22.652Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/07/55b50afb50813dab89a7ef52afdf73d8c5e22a15636edc27ca3cd4114665/xgrammar-0.1.23-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:20692f98e79fe85e1f33482e2f5ae92ba204a7ac95c3322795d39b910292df46", size = 564421, upload-time = "2025-08-15T07:31:23.875Z" },
-    { url = "https://files.pythonhosted.org/packages/96/d8/be7caa3a56ce96defc69b3d9de8538924a0a7114da4e89fe61c5bd557961/xgrammar-0.1.23-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7bde8472e2a28e529bff26f3ca2db56679c8dccbae68813c0ae61db1edff95a5", size = 7683250, upload-time = "2025-08-15T07:31:25.276Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/13/53d950b93a361ef73e5930050916fa36c23fade80ee05cfb0339c044e951/xgrammar-0.1.23-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e0ff9c0a1d46c95d82345a5bf026956ef6d98f1aac7115b57ce88d1d93c4a374", size = 7858258, upload-time = "2025-08-15T07:31:26.782Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/11/2ac0d10c88196f6fc6efa085a175483a72532cc7c2027cbd0583c759a748/xgrammar-0.1.23-cp312-cp312-win_amd64.whl", hash = "sha256:b330a5d673a53a657beb641af8e90146bba4ec18203c440ee7b7766856f5991c", size = 645937, upload-time = "2025-08-15T07:31:28.677Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/d5/797da4ec365d7efdfae47fed89dbc6b408ed2ac397ed08549c3d8a45e72c/xgrammar-0.1.23-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:0583caae0d1478e9c8fc485322c2d5ac9f6766fd995bfe3b5460704b610a12fc", size = 621015, upload-time = "2025-08-15T07:31:35.452Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/46/493805568645b4ea5be1bdcfff1cf845c084fa1f2e04a5bfe4f295a07793/xgrammar-0.1.23-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3dcad40085f1e280ed8bbe5ea19f152e3704fb6456760085481c4e6a376c288c", size = 565126, upload-time = "2025-08-15T07:31:36.68Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/67/733c5425ca0902335593ca9d1756fc131c5934f5fbff1ba28fc37bfef24a/xgrammar-0.1.23-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e7fdb6cf4106ad1b2dc21e3d0dd11b76bbaf3071712b8845124eb2db3171e32", size = 7681247, upload-time = "2025-08-15T07:31:38.409Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/62/b62f1defb085eeea0bbbc1d4a22fb9ea2418175719c44d2d618508976dfa/xgrammar-0.1.23-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f7feb92eb95f093392f05d7b4dd76a49f43909f27df02710f04bc7b717e7e4da", size = 7855034, upload-time = "2025-08-15T07:31:40.03Z" },
-    { url = "https://files.pythonhosted.org/packages/95/5b/8d5fa56f276ccf4f52c5ee6d0a61abbe4defd19f5f8837ede18b16cd313c/xgrammar-0.1.23-cp39-cp39-win_amd64.whl", hash = "sha256:3fa7214b19b1258d980262ce6fce20de3dbc64a85c70d706097c74ed816191c7", size = 646935, upload-time = "2025-08-15T07:31:41.524Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/f2/a9/dc3c63cf7f082d183711e46ef34d10d8a135c2319dc581905d79449f52ea/xgrammar-0.1.25.tar.gz", hash = "sha256:70ce16b27e8082f20808ed759b0733304316facc421656f0f30cfce514b5b77a", size = 2297187, upload-time = "2025-09-21T05:58:58.942Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c0/b4/8f78b56ebf64f161258f339cc5898bf761b4fb6c6805d0bca1bcaaaef4a1/xgrammar-0.1.25-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:d12d1078ee2b5c1531610489b433b77694a7786210ceb2c0c1c1eb058e9053c7", size = 679074, upload-time = "2025-09-21T05:58:20.344Z" },
+    { url = "https://files.pythonhosted.org/packages/52/38/b57120b73adcd342ef974bff14b2b584e7c47edf28d91419cb9325fd5ef2/xgrammar-0.1.25-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c2e940541b7cddf3ef55a70f20d4c872af7f0d900bc0ed36f434bf7212e2e729", size = 622668, upload-time = "2025-09-21T05:58:22.269Z" },
+    { url = "https://files.pythonhosted.org/packages/19/8d/64430d01c21ca2b1d8c5a1ed47c90f8ac43717beafc9440d01d81acd5cfc/xgrammar-0.1.25-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2063e1c72f0c00f47ac8ce7ce0fcbff6fa77f79012e063369683844e2570c266", size = 8517569, upload-time = "2025-09-21T05:58:23.77Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/c4/137d0e9cd038ff4141752c509dbeea0ec5093eb80815620c01b1f1c26d0a/xgrammar-0.1.25-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9785eafa251c996ebaa441f3b8a6c037538930104e265a64a013da0e6fd2ad86", size = 8709188, upload-time = "2025-09-21T05:58:26.246Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/3d/c228c470d50865c9db3fb1e75a95449d0183a8248519b89e86dc481d6078/xgrammar-0.1.25-cp310-cp310-win_amd64.whl", hash = "sha256:42ecefd020038b3919a473fe5b9bb9d8d809717b8689a736b81617dec4acc59b", size = 698919, upload-time = "2025-09-21T05:58:28.368Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/b7/ca0ff7c91f24b2302e94b0e6c2a234cc5752b10da51eb937e7f2aa257fde/xgrammar-0.1.25-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:27d7ac4be05cf9aa258c109a8647092ae47cb1e28df7d27caced6ab44b72b799", size = 678801, upload-time = "2025-09-21T05:58:29.936Z" },
+    { url = "https://files.pythonhosted.org/packages/43/cd/fdf4fb1b5f9c301d381656a600ad95255a76fa68132978af6f06e50a46e1/xgrammar-0.1.25-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:151c1636188bc8c5cdf318cefc5ba23221c9c8cc07cb392317fb3f7635428150", size = 622565, upload-time = "2025-09-21T05:58:31.185Z" },
+    { url = "https://files.pythonhosted.org/packages/55/04/55a87e814bcab771d3e4159281fa382b3d5f14a36114f2f9e572728da831/xgrammar-0.1.25-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:35fc135650aa204bf84db7fe9c0c0f480b6b11419fe47d89f4bd21602ac33be9", size = 8517238, upload-time = "2025-09-21T05:58:32.835Z" },
+    { url = "https://files.pythonhosted.org/packages/31/f6/3c5210bc41b61fb32b66bf5c9fd8ec5edacfeddf9860e95baa9caa9a2c82/xgrammar-0.1.25-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fc19d6d7e8e51b6c9a266e949ac7fb3d2992447efeec7df32cca109149afac18", size = 8709514, upload-time = "2025-09-21T05:58:34.727Z" },
+    { url = "https://files.pythonhosted.org/packages/21/de/85714f307536b328cc16cc6755151865e8875378c8557c15447ca07dff98/xgrammar-0.1.25-cp311-cp311-win_amd64.whl", hash = "sha256:8fcb24f5a7acd5876165c50bd51ce4bf8e6ff897344a5086be92d1fe6695f7fe", size = 698722, upload-time = "2025-09-21T05:58:36.411Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/d7/a7bdb158afa88af7e6e0d312e9677ba5fb5e423932008c9aa2c45af75d5d/xgrammar-0.1.25-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:96500d7578c46e8551253b9211b02e02f54e147bc290479a64717d80dcf4f7e3", size = 678250, upload-time = "2025-09-21T05:58:37.936Z" },
+    { url = "https://files.pythonhosted.org/packages/10/9d/b20588a3209d544a3432ebfcf2e3b1a455833ee658149b08c18eef0c6f59/xgrammar-0.1.25-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:73ba9031e359447af53ce89dfb0775e7b9f4b358d513bcc28a6b4deace661dd5", size = 621550, upload-time = "2025-09-21T05:58:39.464Z" },
+    { url = "https://files.pythonhosted.org/packages/99/9c/39bb38680be3b6d6aa11b8a46a69fb43e2537d6728710b299fa9fc231ff0/xgrammar-0.1.25-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c519518ebc65f75053123baaf23776a21bda58f64101a64c2fc4aa467c9cd480", size = 8519097, upload-time = "2025-09-21T05:58:40.831Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/c2/695797afa9922c30c45aa94e087ad33a9d87843f269461b622a65a39022a/xgrammar-0.1.25-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:47fdbfc6007df47de2142613220292023e88e4a570546b39591f053e4d9ec33f", size = 8712184, upload-time = "2025-09-21T05:58:43.142Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/7f/aa80d1d4c4632cd3d8d083f1de8b470fcb3df23d9165992a3ced019f1b93/xgrammar-0.1.25-cp312-cp312-win_amd64.whl", hash = "sha256:c9b3defb6b45272e896da401f43b513f5ac12104ec3101bbe4d3a7d02bcf4a27", size = 698264, upload-time = "2025-09-21T05:58:44.787Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/85/83401a12a029bb1ea8cff08b19c47aa7077b359ad04251cc1725c0dce926/xgrammar-0.1.25-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:a62dea5d73147a254e71e07ceae4a48c0f5a294cce2fa3e028159f48da19a39d", size = 679391, upload-time = "2025-09-21T05:58:51.019Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/68/9854e947333698b282bd0e7c074e4433d45750cb37a0ae9d5753e66f5695/xgrammar-0.1.25-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2b309807ad837c1cbe2f833830b665a15309e11692b53795363c59041c65074f", size = 622817, upload-time = "2025-09-21T05:58:52.242Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/c3/1e85087460cb752ab5581fbc7d8e604b9de27d2b0312cfbc4a90126cea02/xgrammar-0.1.25-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:073829d8a53ef482e6b51469316f6e505962460bb576ae4d4a606769c4c37678", size = 8517691, upload-time = "2025-09-21T05:58:53.66Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/b9/531d823576559c4528ca4b3907dd6baed11f9f5487ad10997b2200e1f361/xgrammar-0.1.25-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:241529d6104d97738b3e29c573bffa6d0fa89a8d0299b2c685358cc13858205c", size = 8707414, upload-time = "2025-09-21T05:58:55.457Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/49/abdc00753a3eb30c0c85f3eb670420c6c294e2ac569e0f6210fbaca8e41e/xgrammar-0.1.25-cp39-cp39-win_amd64.whl", hash = "sha256:f5d46e1749d9324684d2462e428bc63652096addc1e2c21db2ae66ca88e76a1c", size = 699327, upload-time = "2025-09-21T05:58:57.541Z" },
 ]
 
 [[package]]
 name = "xxhash"
-version = "3.5.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/00/5e/d6e5258d69df8b4ed8c83b6664f2b47d30d2dec551a29ad72a6c69eafd31/xxhash-3.5.0.tar.gz", hash = "sha256:84f2caddf951c9cbf8dc2e22a89d4ccf5d86391ac6418fe81e3c67d0cf60b45f", size = 84241, upload-time = "2024-08-17T09:20:38.972Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/bb/8a/0e9feca390d512d293afd844d31670e25608c4a901e10202aa98785eab09/xxhash-3.5.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:ece616532c499ee9afbb83078b1b952beffef121d989841f7f4b3dc5ac0fd212", size = 31970, upload-time = "2024-08-17T09:17:35.675Z" },
-    { url = "https://files.pythonhosted.org/packages/16/e6/be5aa49580cd064a18200ab78e29b88b1127e1a8c7955eb8ecf81f2626eb/xxhash-3.5.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:3171f693dbc2cef6477054a665dc255d996646b4023fe56cb4db80e26f4cc520", size = 30801, upload-time = "2024-08-17T09:17:37.353Z" },
-    { url = "https://files.pythonhosted.org/packages/20/ee/b8a99ebbc6d1113b3a3f09e747fa318c3cde5b04bd9c197688fadf0eeae8/xxhash-3.5.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7c5d3e570ef46adaf93fc81b44aca6002b5a4d8ca11bd0580c07eac537f36680", size = 220927, upload-time = "2024-08-17T09:17:38.835Z" },
-    { url = "https://files.pythonhosted.org/packages/58/62/15d10582ef159283a5c2b47f6d799fc3303fe3911d5bb0bcc820e1ef7ff4/xxhash-3.5.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7cb29a034301e2982df8b1fe6328a84f4b676106a13e9135a0d7e0c3e9f806da", size = 200360, upload-time = "2024-08-17T09:17:40.851Z" },
-    { url = "https://files.pythonhosted.org/packages/23/41/61202663ea9b1bd8e53673b8ec9e2619989353dba8cfb68e59a9cbd9ffe3/xxhash-3.5.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:5d0d307d27099bb0cbeea7260eb39ed4fdb99c5542e21e94bb6fd29e49c57a23", size = 428528, upload-time = "2024-08-17T09:17:42.545Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/07/d9a3059f702dec5b3b703737afb6dda32f304f6e9da181a229dafd052c29/xxhash-3.5.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0342aafd421795d740e514bc9858ebddfc705a75a8c5046ac56d85fe97bf196", size = 194149, upload-time = "2024-08-17T09:17:44.361Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/58/27caadf78226ecf1d62dbd0c01d152ed381c14c1ee4ad01f0d460fc40eac/xxhash-3.5.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3dbbd9892c5ebffeca1ed620cf0ade13eb55a0d8c84e0751a6653adc6ac40d0c", size = 207703, upload-time = "2024-08-17T09:17:46.656Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/08/32d558ce23e1e068453c39aed7b3c1cdc690c177873ec0ca3a90d5808765/xxhash-3.5.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4cc2d67fdb4d057730c75a64c5923abfa17775ae234a71b0200346bfb0a7f482", size = 216255, upload-time = "2024-08-17T09:17:48.031Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/d4/2b971e2d2b0a61045f842b622ef11e94096cf1f12cd448b6fd426e80e0e2/xxhash-3.5.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:ec28adb204b759306a3d64358a5e5c07d7b1dd0ccbce04aa76cb9377b7b70296", size = 202744, upload-time = "2024-08-17T09:17:50.045Z" },
-    { url = "https://files.pythonhosted.org/packages/19/ae/6a6438864a8c4c39915d7b65effd85392ebe22710412902487e51769146d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:1328f6d8cca2b86acb14104e381225a3d7b42c92c4b86ceae814e5c400dbb415", size = 210115, upload-time = "2024-08-17T09:17:51.834Z" },
-    { url = "https://files.pythonhosted.org/packages/48/7d/b3c27c27d1fc868094d02fe4498ccce8cec9fcc591825c01d6bcb0b4fc49/xxhash-3.5.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:8d47ebd9f5d9607fd039c1fbf4994e3b071ea23eff42f4ecef246ab2b7334198", size = 414247, upload-time = "2024-08-17T09:17:53.094Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/05/918f9e7d2fbbd334b829997045d341d6239b563c44e683b9a7ef8fe50f5d/xxhash-3.5.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b96d559e0fcddd3343c510a0fe2b127fbff16bf346dd76280b82292567523442", size = 191419, upload-time = "2024-08-17T09:17:54.906Z" },
-    { url = "https://files.pythonhosted.org/packages/08/29/dfe393805b2f86bfc47c290b275f0b7c189dc2f4e136fd4754f32eb18a8d/xxhash-3.5.0-cp310-cp310-win32.whl", hash = "sha256:61c722ed8d49ac9bc26c7071eeaa1f6ff24053d553146d5df031802deffd03da", size = 30114, upload-time = "2024-08-17T09:17:56.566Z" },
-    { url = "https://files.pythonhosted.org/packages/7b/d7/aa0b22c4ebb7c3ccb993d4c565132abc641cd11164f8952d89eb6a501909/xxhash-3.5.0-cp310-cp310-win_amd64.whl", hash = "sha256:9bed5144c6923cc902cd14bb8963f2d5e034def4486ab0bbe1f58f03f042f9a9", size = 30003, upload-time = "2024-08-17T09:17:57.596Z" },
-    { url = "https://files.pythonhosted.org/packages/69/12/f969b81541ee91b55f1ce469d7ab55079593c80d04fd01691b550e535000/xxhash-3.5.0-cp310-cp310-win_arm64.whl", hash = "sha256:893074d651cf25c1cc14e3bea4fceefd67f2921b1bb8e40fcfeba56820de80c6", size = 26773, upload-time = "2024-08-17T09:17:59.169Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/c7/afed0f131fbda960ff15eee7f304fa0eeb2d58770fade99897984852ef23/xxhash-3.5.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:02c2e816896dc6f85922ced60097bcf6f008dedfc5073dcba32f9c8dd786f3c1", size = 31969, upload-time = "2024-08-17T09:18:00.852Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/0c/7c3bc6d87e5235672fcc2fb42fd5ad79fe1033925f71bf549ee068c7d1ca/xxhash-3.5.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6027dcd885e21581e46d3c7f682cfb2b870942feeed58a21c29583512c3f09f8", size = 30800, upload-time = "2024-08-17T09:18:01.863Z" },
-    { url = "https://files.pythonhosted.org/packages/04/9e/01067981d98069eec1c20201f8c145367698e9056f8bc295346e4ea32dd1/xxhash-3.5.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1308fa542bbdbf2fa85e9e66b1077eea3a88bef38ee8a06270b4298a7a62a166", size = 221566, upload-time = "2024-08-17T09:18:03.461Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/09/d4996de4059c3ce5342b6e1e6a77c9d6c91acce31f6ed979891872dd162b/xxhash-3.5.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c28b2fdcee797e1c1961cd3bcd3d545cab22ad202c846235197935e1df2f8ef7", size = 201214, upload-time = "2024-08-17T09:18:05.616Z" },
-    { url = "https://files.pythonhosted.org/packages/62/f5/6d2dc9f8d55a7ce0f5e7bfef916e67536f01b85d32a9fbf137d4cadbee38/xxhash-3.5.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:924361811732ddad75ff23e90efd9ccfda4f664132feecb90895bade6a1b4623", size = 429433, upload-time = "2024-08-17T09:18:06.957Z" },
-    { url = "https://files.pythonhosted.org/packages/d9/72/9256303f10e41ab004799a4aa74b80b3c5977d6383ae4550548b24bd1971/xxhash-3.5.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:89997aa1c4b6a5b1e5b588979d1da048a3c6f15e55c11d117a56b75c84531f5a", size = 194822, upload-time = "2024-08-17T09:18:08.331Z" },
-    { url = "https://files.pythonhosted.org/packages/34/92/1a3a29acd08248a34b0e6a94f4e0ed9b8379a4ff471f1668e4dce7bdbaa8/xxhash-3.5.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:685c4f4e8c59837de103344eb1c8a3851f670309eb5c361f746805c5471b8c88", size = 208538, upload-time = "2024-08-17T09:18:10.332Z" },
-    { url = "https://files.pythonhosted.org/packages/53/ad/7fa1a109663366de42f724a1cdb8e796a260dbac45047bce153bc1e18abf/xxhash-3.5.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dbd2ecfbfee70bc1a4acb7461fa6af7748ec2ab08ac0fa298f281c51518f982c", size = 216953, upload-time = "2024-08-17T09:18:11.707Z" },
-    { url = "https://files.pythonhosted.org/packages/35/02/137300e24203bf2b2a49b48ce898ecce6fd01789c0fcd9c686c0a002d129/xxhash-3.5.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:25b5a51dc3dfb20a10833c8eee25903fd2e14059e9afcd329c9da20609a307b2", size = 203594, upload-time = "2024-08-17T09:18:13.799Z" },
-    { url = "https://files.pythonhosted.org/packages/23/03/aeceb273933d7eee248c4322b98b8e971f06cc3880e5f7602c94e5578af5/xxhash-3.5.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:a8fb786fb754ef6ff8c120cb96629fb518f8eb5a61a16aac3a979a9dbd40a084", size = 210971, upload-time = "2024-08-17T09:18:15.824Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/64/ed82ec09489474cbb35c716b189ddc1521d8b3de12b1b5ab41ce7f70253c/xxhash-3.5.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:a905ad00ad1e1c34fe4e9d7c1d949ab09c6fa90c919860c1534ff479f40fd12d", size = 415050, upload-time = "2024-08-17T09:18:17.142Z" },
-    { url = "https://files.pythonhosted.org/packages/71/43/6db4c02dcb488ad4e03bc86d70506c3d40a384ee73c9b5c93338eb1f3c23/xxhash-3.5.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:963be41bcd49f53af6d795f65c0da9b4cc518c0dd9c47145c98f61cb464f4839", size = 192216, upload-time = "2024-08-17T09:18:18.779Z" },
-    { url = "https://files.pythonhosted.org/packages/22/6d/db4abec29e7a567455344433d095fdb39c97db6955bb4a2c432e486b4d28/xxhash-3.5.0-cp311-cp311-win32.whl", hash = "sha256:109b436096d0a2dd039c355fa3414160ec4d843dfecc64a14077332a00aeb7da", size = 30120, upload-time = "2024-08-17T09:18:20.009Z" },
-    { url = "https://files.pythonhosted.org/packages/52/1c/fa3b61c0cf03e1da4767213672efe186b1dfa4fc901a4a694fb184a513d1/xxhash-3.5.0-cp311-cp311-win_amd64.whl", hash = "sha256:b702f806693201ad6c0a05ddbbe4c8f359626d0b3305f766077d51388a6bac58", size = 30003, upload-time = "2024-08-17T09:18:21.052Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/8e/9e6fc572acf6e1cc7ccb01973c213f895cb8668a9d4c2b58a99350da14b7/xxhash-3.5.0-cp311-cp311-win_arm64.whl", hash = "sha256:c4dcb4120d0cc3cc448624147dba64e9021b278c63e34a38789b688fd0da9bf3", size = 26777, upload-time = "2024-08-17T09:18:22.809Z" },
-    { url = "https://files.pythonhosted.org/packages/07/0e/1bfce2502c57d7e2e787600b31c83535af83746885aa1a5f153d8c8059d6/xxhash-3.5.0-cp312-cp312-macosx_10_9_x86_64.whl", hash = "sha256:14470ace8bd3b5d51318782cd94e6f94431974f16cb3b8dc15d52f3b69df8e00", size = 31969, upload-time = "2024-08-17T09:18:24.025Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/d6/8ca450d6fe5b71ce521b4e5db69622383d039e2b253e9b2f24f93265b52c/xxhash-3.5.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:59aa1203de1cb96dbeab595ded0ad0c0056bb2245ae11fac11c0ceea861382b9", size = 30787, upload-time = "2024-08-17T09:18:25.318Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/84/de7c89bc6ef63d750159086a6ada6416cc4349eab23f76ab870407178b93/xxhash-3.5.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:08424f6648526076e28fae6ea2806c0a7d504b9ef05ae61d196d571e5c879c84", size = 220959, upload-time = "2024-08-17T09:18:26.518Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/86/51258d3e8a8545ff26468c977101964c14d56a8a37f5835bc0082426c672/xxhash-3.5.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:61a1ff00674879725b194695e17f23d3248998b843eb5e933007ca743310f793", size = 200006, upload-time = "2024-08-17T09:18:27.905Z" },
-    { url = "https://files.pythonhosted.org/packages/02/0a/96973bd325412feccf23cf3680fd2246aebf4b789122f938d5557c54a6b2/xxhash-3.5.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f2f2c61bee5844d41c3eb015ac652a0229e901074951ae48581d58bfb2ba01be", size = 428326, upload-time = "2024-08-17T09:18:29.335Z" },
-    { url = "https://files.pythonhosted.org/packages/11/a7/81dba5010f7e733de88af9555725146fc133be97ce36533867f4c7e75066/xxhash-3.5.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9d32a592cac88d18cc09a89172e1c32d7f2a6e516c3dfde1b9adb90ab5df54a6", size = 194380, upload-time = "2024-08-17T09:18:30.706Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/7d/f29006ab398a173f4501c0e4977ba288f1c621d878ec217b4ff516810c04/xxhash-3.5.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:70dabf941dede727cca579e8c205e61121afc9b28516752fd65724be1355cc90", size = 207934, upload-time = "2024-08-17T09:18:32.133Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/6e/6e88b8f24612510e73d4d70d9b0c7dff62a2e78451b9f0d042a5462c8d03/xxhash-3.5.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e5d0ddaca65ecca9c10dcf01730165fd858533d0be84c75c327487c37a906a27", size = 216301, upload-time = "2024-08-17T09:18:33.474Z" },
-    { url = "https://files.pythonhosted.org/packages/af/51/7862f4fa4b75a25c3b4163c8a873f070532fe5f2d3f9b3fc869c8337a398/xxhash-3.5.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:3e5b5e16c5a480fe5f59f56c30abdeba09ffd75da8d13f6b9b6fd224d0b4d0a2", size = 203351, upload-time = "2024-08-17T09:18:34.889Z" },
-    { url = "https://files.pythonhosted.org/packages/22/61/8d6a40f288f791cf79ed5bb113159abf0c81d6efb86e734334f698eb4c59/xxhash-3.5.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:149b7914451eb154b3dfaa721315117ea1dac2cc55a01bfbd4df7c68c5dd683d", size = 210294, upload-time = "2024-08-17T09:18:36.355Z" },
-    { url = "https://files.pythonhosted.org/packages/17/02/215c4698955762d45a8158117190261b2dbefe9ae7e5b906768c09d8bc74/xxhash-3.5.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:eade977f5c96c677035ff39c56ac74d851b1cca7d607ab3d8f23c6b859379cab", size = 414674, upload-time = "2024-08-17T09:18:38.536Z" },
-    { url = "https://files.pythonhosted.org/packages/31/5c/b7a8db8a3237cff3d535261325d95de509f6a8ae439a5a7a4ffcff478189/xxhash-3.5.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:fa9f547bd98f5553d03160967866a71056a60960be00356a15ecc44efb40ba8e", size = 192022, upload-time = "2024-08-17T09:18:40.138Z" },
-    { url = "https://files.pythonhosted.org/packages/78/e3/dd76659b2811b3fd06892a8beb850e1996b63e9235af5a86ea348f053e9e/xxhash-3.5.0-cp312-cp312-win32.whl", hash = "sha256:f7b58d1fd3551b8c80a971199543379be1cee3d0d409e1f6d8b01c1a2eebf1f8", size = 30170, upload-time = "2024-08-17T09:18:42.163Z" },
-    { url = "https://files.pythonhosted.org/packages/d9/6b/1c443fe6cfeb4ad1dcf231cdec96eb94fb43d6498b4469ed8b51f8b59a37/xxhash-3.5.0-cp312-cp312-win_amd64.whl", hash = "sha256:fa0cafd3a2af231b4e113fba24a65d7922af91aeb23774a8b78228e6cd785e3e", size = 30040, upload-time = "2024-08-17T09:18:43.699Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/eb/04405305f290173acc0350eba6d2f1a794b57925df0398861a20fbafa415/xxhash-3.5.0-cp312-cp312-win_arm64.whl", hash = "sha256:586886c7e89cb9828bcd8a5686b12e161368e0064d040e225e72607b43858ba2", size = 26796, upload-time = "2024-08-17T09:18:45.29Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/f6/531dd6858adf8877675270b9d6989b6dacfd1c2d7135b17584fc29866df3/xxhash-3.5.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:bfc8cdd7f33d57f0468b0614ae634cc38ab9202c6957a60e31d285a71ebe0301", size = 31971, upload-time = "2024-08-17T09:19:47.447Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/a8/b2a42b6c9ae46e233f474f3d307c2e7bca8d9817650babeca048d2ad01d6/xxhash-3.5.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:e0c48b6300cd0b0106bf49169c3e0536408dfbeb1ccb53180068a18b03c662ab", size = 30801, upload-time = "2024-08-17T09:19:48.911Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/92/9ac297e3487818f429bcf369c1c6a097edf5b56ed6fc1feff4c1882e87ef/xxhash-3.5.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fe1a92cfbaa0a1253e339ccec42dbe6db262615e52df591b68726ab10338003f", size = 220644, upload-time = "2024-08-17T09:19:51.081Z" },
-    { url = "https://files.pythonhosted.org/packages/86/48/c1426dd3c86fc4a52f983301867463472f6a9013fb32d15991e60c9919b6/xxhash-3.5.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:33513d6cc3ed3b559134fb307aae9bdd94d7e7c02907b37896a6c45ff9ce51bd", size = 200021, upload-time = "2024-08-17T09:19:52.923Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/de/0ab8c79993765c94fc0d0c1a22b454483c58a0161e1b562f58b654f47660/xxhash-3.5.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:eefc37f6138f522e771ac6db71a6d4838ec7933939676f3753eafd7d3f4c40bc", size = 428217, upload-time = "2024-08-17T09:19:54.349Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/b4/332647451ed7d2c021294b7c1e9c144dbb5586b1fb214ad4f5a404642835/xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a606c8070ada8aa2a88e181773fa1ef17ba65ce5dd168b9d08038e2a61b33754", size = 193868, upload-time = "2024-08-17T09:19:55.763Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/1c/a42c0a6cac752f84f7b44a90d1a9fa9047cf70bdba5198a304fde7cc471f/xxhash-3.5.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:42eca420c8fa072cc1dd62597635d140e78e384a79bb4944f825fbef8bfeeef6", size = 207403, upload-time = "2024-08-17T09:19:57.945Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/d7/04e1b0daae9dc9b02c73c1664cc8aa527498c3f66ccbc586eeb25bbe9f14/xxhash-3.5.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:604253b2143e13218ff1ef0b59ce67f18b8bd1c4205d2ffda22b09b426386898", size = 215978, upload-time = "2024-08-17T09:19:59.381Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/f4/05e15e67505228fc19ee98a79e427b3a0b9695f5567cd66ced5d66389883/xxhash-3.5.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:6e93a5ad22f434d7876665444a97e713a8f60b5b1a3521e8df11b98309bff833", size = 202416, upload-time = "2024-08-17T09:20:01.534Z" },
-    { url = "https://files.pythonhosted.org/packages/94/fb/e9028d3645bba5412a09de13ee36df276a567e60bdb31d499dafa46d76ae/xxhash-3.5.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:7a46e1d6d2817ba8024de44c4fd79913a90e5f7265434cef97026215b7d30df6", size = 209853, upload-time = "2024-08-17T09:20:03.376Z" },
-    { url = "https://files.pythonhosted.org/packages/02/2c/18c6a622429368274739372d2f86c8125413ec169025c7d8ffb051784bba/xxhash-3.5.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:30eb2efe6503c379b7ab99c81ba4a779748e3830241f032ab46bd182bf5873af", size = 413926, upload-time = "2024-08-17T09:20:04.946Z" },
-    { url = "https://files.pythonhosted.org/packages/72/bb/5b55c391084a0321c3809632a018b9b657e59d5966289664f85a645942ac/xxhash-3.5.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c8aa771ff2c13dd9cda8166d685d7333d389fae30a4d2bb39d63ab5775de8606", size = 191156, upload-time = "2024-08-17T09:20:06.318Z" },
-    { url = "https://files.pythonhosted.org/packages/86/2b/915049db13401792fec159f57e4f4a5ca7a9768e83ef71d6645b9d0cd749/xxhash-3.5.0-cp39-cp39-win32.whl", hash = "sha256:5ed9ebc46f24cf91034544b26b131241b699edbfc99ec5e7f8f3d02d6eb7fba4", size = 30122, upload-time = "2024-08-17T09:20:07.691Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/87/382ef7b24917d7cf4c540ee30f29b283bc87ac5893d2f89b23ea3cdf7d77/xxhash-3.5.0-cp39-cp39-win_amd64.whl", hash = "sha256:220f3f896c6b8d0316f63f16c077d52c412619e475f9372333474ee15133a558", size = 30021, upload-time = "2024-08-17T09:20:08.832Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/47/d06b24e2d9c3dcabccfd734d11b5bbebfdf59ceac2c61509d8205dd20ac6/xxhash-3.5.0-cp39-cp39-win_arm64.whl", hash = "sha256:a7b1d8315d9b5e9f89eb2933b73afae6ec9597a258d52190944437158b49d38e", size = 26780, upload-time = "2024-08-17T09:20:09.989Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/9a/233606bada5bd6f50b2b72c45de3d9868ad551e83893d2ac86dc7bb8553a/xxhash-3.5.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:2014c5b3ff15e64feecb6b713af12093f75b7926049e26a580e94dcad3c73d8c", size = 29732, upload-time = "2024-08-17T09:20:11.175Z" },
-    { url = "https://files.pythonhosted.org/packages/0c/67/f75276ca39e2c6604e3bee6c84e9db8a56a4973fde9bf35989787cf6e8aa/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fab81ef75003eda96239a23eda4e4543cedc22e34c373edcaf744e721a163986", size = 36214, upload-time = "2024-08-17T09:20:12.335Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/f8/f6c61fd794229cc3848d144f73754a0c107854372d7261419dcbbd286299/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4e2febf914ace002132aa09169cc572e0d8959d0f305f93d5828c4836f9bc5a6", size = 32020, upload-time = "2024-08-17T09:20:13.537Z" },
-    { url = "https://files.pythonhosted.org/packages/79/d3/c029c99801526f859e6b38d34ab87c08993bf3dcea34b11275775001638a/xxhash-3.5.0-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5d3a10609c51da2a1c0ea0293fc3968ca0a18bd73838455b5bca3069d7f8e32b", size = 40515, upload-time = "2024-08-17T09:20:14.669Z" },
-    { url = "https://files.pythonhosted.org/packages/62/e3/bef7b82c1997579c94de9ac5ea7626d01ae5858aa22bf4fcb38bf220cb3e/xxhash-3.5.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:5a74f23335b9689b66eb6dbe2a931a88fcd7a4c2cc4b1cb0edba8ce381c7a1da", size = 30064, upload-time = "2024-08-17T09:20:15.925Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/56/30d3df421814947f9d782b20c9b7e5e957f3791cbd89874578011daafcbd/xxhash-3.5.0-pp39-pypy39_pp73-macosx_10_15_x86_64.whl", hash = "sha256:531af8845aaadcadf951b7e0c1345c6b9c68a990eeb74ff9acd8501a0ad6a1c9", size = 29734, upload-time = "2024-08-17T09:20:30.457Z" },
-    { url = "https://files.pythonhosted.org/packages/82/dd/3c42a1f022ad0d82c852d3cb65493ebac03dcfa8c994465a5fb052b00e3c/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7ce379bcaa9fcc00f19affa7773084dd09f5b59947b3fb47a1ceb0179f91aaa1", size = 36216, upload-time = "2024-08-17T09:20:32.116Z" },
-    { url = "https://files.pythonhosted.org/packages/b2/40/8f902ab3bebda228a9b4de69eba988280285a7f7f167b942bc20bb562df9/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fd1b2281d01723f076df3c8188f43f2472248a6b63118b036e641243656b1b0f", size = 32042, upload-time = "2024-08-17T09:20:33.562Z" },
-    { url = "https://files.pythonhosted.org/packages/db/87/bd06beb8ccaa0e9e577c9b909a49cfa5c5cd2ca46034342d72dd9ce5bc56/xxhash-3.5.0-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9c770750cc80e8694492244bca7251385188bc5597b6a39d98a9f30e8da984e0", size = 40516, upload-time = "2024-08-17T09:20:36.004Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/f8/505385e2fbd753ddcaafd5550eabe86f6232cbebabad3b2508d411b19153/xxhash-3.5.0-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:b150b8467852e1bd844387459aa6fbe11d7f38b56e901f9f3b3e6aba0d660240", size = 30108, upload-time = "2024-08-17T09:20:37.214Z" },
+version = "3.6.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/02/84/30869e01909fb37a6cc7e18688ee8bf1e42d57e7e0777636bd47524c43c7/xxhash-3.6.0.tar.gz", hash = "sha256:f0162a78b13a0d7617b2845b90c763339d1f1d82bb04a4b07f4ab535cc5e05d6", size = 85160, upload-time = "2025-10-02T14:37:08.097Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/34/ee/f9f1d656ad168681bb0f6b092372c1e533c4416b8069b1896a175c46e484/xxhash-3.6.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:87ff03d7e35c61435976554477a7f4cd1704c3596a89a8300d5ce7fc83874a71", size = 32845, upload-time = "2025-10-02T14:33:51.573Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/b1/93508d9460b292c74a09b83d16750c52a0ead89c51eea9951cb97a60d959/xxhash-3.6.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:f572dfd3d0e2eb1a57511831cf6341242f5a9f8298a45862d085f5b93394a27d", size = 30807, upload-time = "2025-10-02T14:33:52.964Z" },
+    { url = "https://files.pythonhosted.org/packages/07/55/28c93a3662f2d200c70704efe74aab9640e824f8ce330d8d3943bf7c9b3c/xxhash-3.6.0-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:89952ea539566b9fed2bbd94e589672794b4286f342254fad28b149f9615fef8", size = 193786, upload-time = "2025-10-02T14:33:54.272Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/96/fec0be9bb4b8f5d9c57d76380a366f31a1781fb802f76fc7cda6c84893c7/xxhash-3.6.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:48e6f2ffb07a50b52465a1032c3cf1f4a5683f944acaca8a134a2f23674c2058", size = 212830, upload-time = "2025-10-02T14:33:55.706Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/a0/c706845ba77b9611f81fd2e93fad9859346b026e8445e76f8c6fd057cc6d/xxhash-3.6.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b5b848ad6c16d308c3ac7ad4ba6bede80ed5df2ba8ed382f8932df63158dd4b2", size = 211606, upload-time = "2025-10-02T14:33:57.133Z" },
+    { url = "https://files.pythonhosted.org/packages/67/1e/164126a2999e5045f04a69257eea946c0dc3e86541b400d4385d646b53d7/xxhash-3.6.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a034590a727b44dd8ac5914236a7b8504144447a9682586c3327e935f33ec8cc", size = 444872, upload-time = "2025-10-02T14:33:58.446Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/4b/55ab404c56cd70a2cf5ecfe484838865d0fea5627365c6c8ca156bd09c8f/xxhash-3.6.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8a8f1972e75ebdd161d7896743122834fe87378160c20e97f8b09166213bf8cc", size = 193217, upload-time = "2025-10-02T14:33:59.724Z" },
+    { url = "https://files.pythonhosted.org/packages/45/e6/52abf06bac316db33aa269091ae7311bd53cfc6f4b120ae77bac1b348091/xxhash-3.6.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:ee34327b187f002a596d7b167ebc59a1b729e963ce645964bbc050d2f1b73d07", size = 210139, upload-time = "2025-10-02T14:34:02.041Z" },
+    { url = "https://files.pythonhosted.org/packages/34/37/db94d490b8691236d356bc249c08819cbcef9273a1a30acf1254ff9ce157/xxhash-3.6.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:339f518c3c7a850dd033ab416ea25a692759dc7478a71131fe8869010d2b75e4", size = 197669, upload-time = "2025-10-02T14:34:03.664Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/36/c4f219ef4a17a4f7a64ed3569bc2b5a9c8311abdb22249ac96093625b1a4/xxhash-3.6.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:bf48889c9630542d4709192578aebbd836177c9f7a4a2778a7d6340107c65f06", size = 210018, upload-time = "2025-10-02T14:34:05.325Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/06/bfac889a374fc2fc439a69223d1750eed2e18a7db8514737ab630534fa08/xxhash-3.6.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:5576b002a56207f640636056b4160a378fe36a58db73ae5c27a7ec8db35f71d4", size = 413058, upload-time = "2025-10-02T14:34:06.925Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/d1/555d8447e0dd32ad0930a249a522bb2e289f0d08b6b16204cfa42c1f5a0c/xxhash-3.6.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:af1f3278bd02814d6dedc5dec397993b549d6f16c19379721e5a1d31e132c49b", size = 190628, upload-time = "2025-10-02T14:34:08.669Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/15/8751330b5186cedc4ed4b597989882ea05e0408b53fa47bcb46a6125bfc6/xxhash-3.6.0-cp310-cp310-win32.whl", hash = "sha256:aed058764db109dc9052720da65fafe84873b05eb8b07e5e653597951af57c3b", size = 30577, upload-time = "2025-10-02T14:34:10.234Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/cc/53f87e8b5871a6eb2ff7e89c48c66093bda2be52315a8161ddc54ea550c4/xxhash-3.6.0-cp310-cp310-win_amd64.whl", hash = "sha256:e82da5670f2d0d98950317f82a0e4a0197150ff19a6df2ba40399c2a3b9ae5fb", size = 31487, upload-time = "2025-10-02T14:34:11.618Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/00/60f9ea3bb697667a14314d7269956f58bf56bb73864f8f8d52a3c2535e9a/xxhash-3.6.0-cp310-cp310-win_arm64.whl", hash = "sha256:4a082ffff8c6ac07707fb6b671caf7c6e020c75226c561830b73d862060f281d", size = 27863, upload-time = "2025-10-02T14:34:12.619Z" },
+    { url = "https://files.pythonhosted.org/packages/17/d4/cc2f0400e9154df4b9964249da78ebd72f318e35ccc425e9f403c392f22a/xxhash-3.6.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:b47bbd8cf2d72797f3c2772eaaac0ded3d3af26481a26d7d7d41dc2d3c46b04a", size = 32844, upload-time = "2025-10-02T14:34:14.037Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/ec/1cc11cd13e26ea8bc3cb4af4eaadd8d46d5014aebb67be3f71fb0b68802a/xxhash-3.6.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2b6821e94346f96db75abaa6e255706fb06ebd530899ed76d32cd99f20dc52fa", size = 30809, upload-time = "2025-10-02T14:34:15.484Z" },
+    { url = "https://files.pythonhosted.org/packages/04/5f/19fe357ea348d98ca22f456f75a30ac0916b51c753e1f8b2e0e6fb884cce/xxhash-3.6.0-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d0a9751f71a1a65ce3584e9cae4467651c7e70c9d31017fa57574583a4540248", size = 194665, upload-time = "2025-10-02T14:34:16.541Z" },
+    { url = "https://files.pythonhosted.org/packages/90/3b/d1f1a8f5442a5fd8beedae110c5af7604dc37349a8e16519c13c19a9a2de/xxhash-3.6.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8b29ee68625ab37b04c0b40c3fafdf24d2f75ccd778333cfb698f65f6c463f62", size = 213550, upload-time = "2025-10-02T14:34:17.878Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/ef/3a9b05eb527457d5db13a135a2ae1a26c80fecd624d20f3e8dcc4cb170f3/xxhash-3.6.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:6812c25fe0d6c36a46ccb002f40f27ac903bf18af9f6dd8f9669cb4d176ab18f", size = 212384, upload-time = "2025-10-02T14:34:19.182Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/18/ccc194ee698c6c623acbf0f8c2969811a8a4b6185af5e824cd27b9e4fd3e/xxhash-3.6.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4ccbff013972390b51a18ef1255ef5ac125c92dc9143b2d1909f59abc765540e", size = 445749, upload-time = "2025-10-02T14:34:20.659Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/86/cf2c0321dc3940a7aa73076f4fd677a0fb3e405cb297ead7d864fd90847e/xxhash-3.6.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:297b7fbf86c82c550e12e8fb71968b3f033d27b874276ba3624ea868c11165a8", size = 193880, upload-time = "2025-10-02T14:34:22.431Z" },
+    { url = "https://files.pythonhosted.org/packages/82/fb/96213c8560e6f948a1ecc9a7613f8032b19ee45f747f4fca4eb31bb6d6ed/xxhash-3.6.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:dea26ae1eb293db089798d3973a5fc928a18fdd97cc8801226fae705b02b14b0", size = 210912, upload-time = "2025-10-02T14:34:23.937Z" },
+    { url = "https://files.pythonhosted.org/packages/40/aa/4395e669b0606a096d6788f40dbdf2b819d6773aa290c19e6e83cbfc312f/xxhash-3.6.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:7a0b169aafb98f4284f73635a8e93f0735f9cbde17bd5ec332480484241aaa77", size = 198654, upload-time = "2025-10-02T14:34:25.644Z" },
+    { url = "https://files.pythonhosted.org/packages/67/74/b044fcd6b3d89e9b1b665924d85d3f400636c23590226feb1eb09e1176ce/xxhash-3.6.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:08d45aef063a4531b785cd72de4887766d01dc8f362a515693df349fdb825e0c", size = 210867, upload-time = "2025-10-02T14:34:27.203Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/fd/3ce73bf753b08cb19daee1eb14aa0d7fe331f8da9c02dd95316ddfe5275e/xxhash-3.6.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:929142361a48ee07f09121fe9e96a84950e8d4df3bb298ca5d88061969f34d7b", size = 414012, upload-time = "2025-10-02T14:34:28.409Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/b3/5a4241309217c5c876f156b10778f3ab3af7ba7e3259e6d5f5c7d0129eb2/xxhash-3.6.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:51312c768403d8540487dbbfb557454cfc55589bbde6424456951f7fcd4facb3", size = 191409, upload-time = "2025-10-02T14:34:29.696Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/01/99bfbc15fb9abb9a72b088c1d95219fc4782b7d01fc835bd5744d66dd0b8/xxhash-3.6.0-cp311-cp311-win32.whl", hash = "sha256:d1927a69feddc24c987b337ce81ac15c4720955b667fe9b588e02254b80446fd", size = 30574, upload-time = "2025-10-02T14:34:31.028Z" },
+    { url = "https://files.pythonhosted.org/packages/65/79/9d24d7f53819fe301b231044ea362ce64e86c74f6e8c8e51320de248b3e5/xxhash-3.6.0-cp311-cp311-win_amd64.whl", hash = "sha256:26734cdc2d4ffe449b41d186bbeac416f704a482ed835d375a5c0cb02bc63fef", size = 31481, upload-time = "2025-10-02T14:34:32.062Z" },
+    { url = "https://files.pythonhosted.org/packages/30/4e/15cd0e3e8772071344eab2961ce83f6e485111fed8beb491a3f1ce100270/xxhash-3.6.0-cp311-cp311-win_arm64.whl", hash = "sha256:d72f67ef8bf36e05f5b6c65e8524f265bd61071471cd4cf1d36743ebeeeb06b7", size = 27861, upload-time = "2025-10-02T14:34:33.555Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/07/d9412f3d7d462347e4511181dea65e47e0d0e16e26fbee2ea86a2aefb657/xxhash-3.6.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:01362c4331775398e7bb34e3ab403bc9ee9f7c497bc7dee6272114055277dd3c", size = 32744, upload-time = "2025-10-02T14:34:34.622Z" },
+    { url = "https://files.pythonhosted.org/packages/79/35/0429ee11d035fc33abe32dca1b2b69e8c18d236547b9a9b72c1929189b9a/xxhash-3.6.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b7b2df81a23f8cb99656378e72501b2cb41b1827c0f5a86f87d6b06b69f9f204", size = 30816, upload-time = "2025-10-02T14:34:36.043Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/f2/57eb99aa0f7d98624c0932c5b9a170e1806406cdbcdb510546634a1359e0/xxhash-3.6.0-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:dc94790144e66b14f67b10ac8ed75b39ca47536bf8800eb7c24b50271ea0c490", size = 194035, upload-time = "2025-10-02T14:34:37.354Z" },
+    { url = "https://files.pythonhosted.org/packages/4c/ed/6224ba353690d73af7a3f1c7cdb1fc1b002e38f783cb991ae338e1eb3d79/xxhash-3.6.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:93f107c673bccf0d592cdba077dedaf52fe7f42dcd7676eba1f6d6f0c3efffd2", size = 212914, upload-time = "2025-10-02T14:34:38.6Z" },
+    { url = "https://files.pythonhosted.org/packages/38/86/fb6b6130d8dd6b8942cc17ab4d90e223653a89aa32ad2776f8af7064ed13/xxhash-3.6.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:2aa5ee3444c25b69813663c9f8067dcfaa2e126dc55e8dddf40f4d1c25d7effa", size = 212163, upload-time = "2025-10-02T14:34:39.872Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/dc/e84875682b0593e884ad73b2d40767b5790d417bde603cceb6878901d647/xxhash-3.6.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f7f99123f0e1194fa59cc69ad46dbae2e07becec5df50a0509a808f90a0f03f0", size = 445411, upload-time = "2025-10-02T14:34:41.569Z" },
+    { url = "https://files.pythonhosted.org/packages/11/4f/426f91b96701ec2f37bb2b8cec664eff4f658a11f3fa9d94f0a887ea6d2b/xxhash-3.6.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:49e03e6fe2cac4a1bc64952dd250cf0dbc5ef4ebb7b8d96bce82e2de163c82a2", size = 193883, upload-time = "2025-10-02T14:34:43.249Z" },
+    { url = "https://files.pythonhosted.org/packages/53/5a/ddbb83eee8e28b778eacfc5a85c969673e4023cdeedcfcef61f36731610b/xxhash-3.6.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:bd17fede52a17a4f9a7bc4472a5867cb0b160deeb431795c0e4abe158bc784e9", size = 210392, upload-time = "2025-10-02T14:34:45.042Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/c2/ff69efd07c8c074ccdf0a4f36fcdd3d27363665bcdf4ba399abebe643465/xxhash-3.6.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:6fb5f5476bef678f69db04f2bd1efbed3030d2aba305b0fc1773645f187d6a4e", size = 197898, upload-time = "2025-10-02T14:34:46.302Z" },
+    { url = "https://files.pythonhosted.org/packages/58/ca/faa05ac19b3b622c7c9317ac3e23954187516298a091eb02c976d0d3dd45/xxhash-3.6.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:843b52f6d88071f87eba1631b684fcb4b2068cd2180a0224122fe4ef011a9374", size = 210655, upload-time = "2025-10-02T14:34:47.571Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/7a/06aa7482345480cc0cb597f5c875b11a82c3953f534394f620b0be2f700c/xxhash-3.6.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:7d14a6cfaf03b1b6f5f9790f76880601ccc7896aff7ab9cd8978a939c1eb7e0d", size = 414001, upload-time = "2025-10-02T14:34:49.273Z" },
+    { url = "https://files.pythonhosted.org/packages/23/07/63ffb386cd47029aa2916b3d2f454e6cc5b9f5c5ada3790377d5430084e7/xxhash-3.6.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:418daf3db71e1413cfe211c2f9a528456936645c17f46b5204705581a45390ae", size = 191431, upload-time = "2025-10-02T14:34:50.798Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/93/14fde614cadb4ddf5e7cebf8918b7e8fac5ae7861c1875964f17e678205c/xxhash-3.6.0-cp312-cp312-win32.whl", hash = "sha256:50fc255f39428a27299c20e280d6193d8b63b8ef8028995323bf834a026b4fbb", size = 30617, upload-time = "2025-10-02T14:34:51.954Z" },
+    { url = "https://files.pythonhosted.org/packages/13/5d/0d125536cbe7565a83d06e43783389ecae0c0f2ed037b48ede185de477c0/xxhash-3.6.0-cp312-cp312-win_amd64.whl", hash = "sha256:c0f2ab8c715630565ab8991b536ecded9416d615538be8ecddce43ccf26cbc7c", size = 31534, upload-time = "2025-10-02T14:34:53.276Z" },
+    { url = "https://files.pythonhosted.org/packages/54/85/6ec269b0952ec7e36ba019125982cf11d91256a778c7c3f98a4c5043d283/xxhash-3.6.0-cp312-cp312-win_arm64.whl", hash = "sha256:eae5c13f3bc455a3bbb68bdc513912dc7356de7e2280363ea235f71f54064829", size = 27876, upload-time = "2025-10-02T14:34:54.371Z" },
+    { url = "https://files.pythonhosted.org/packages/03/ff/1b4bb3f397552116c1df6266c1b83a21aeeb26061ab1f462984b499a3870/xxhash-3.6.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:cc604dc06027dbeb8281aeac5899c35fcfe7c77b25212833709f0bff4ce74d2a", size = 32844, upload-time = "2025-10-02T14:36:39.157Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/db/27146d0bee4346a9a31f7b498a81fc02747f6f1e6c52a2e7989504278051/xxhash-3.6.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:277175a73900ad43a8caeb8b99b9604f21fe8d7c842f2f9061a364a7e220ddb7", size = 30806, upload-time = "2025-10-02T14:36:40.621Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/2b/4896188df564908817a75de19bf7f2384b99a75af2d528f9c49326f76458/xxhash-3.6.0-cp39-cp39-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:cfbc5b91397c8c2972fdac13fb3e4ed2f7f8ccac85cd2c644887557780a9b6e2", size = 193448, upload-time = "2025-10-02T14:36:41.797Z" },
+    { url = "https://files.pythonhosted.org/packages/51/c5/be8953f62e772340319a826ce1e07489935600089756cf83b628cd36ebe3/xxhash-3.6.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:2762bfff264c4e73c0e507274b40634ff465e025f0eaf050897e88ec8367575d", size = 212547, upload-time = "2025-10-02T14:36:43.581Z" },
+    { url = "https://files.pythonhosted.org/packages/51/1a/1e9f0b911d1cf00dd537c074ae3fae15b535a7f0d9e7edd42a9d2c4f78ce/xxhash-3.6.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:2f171a900d59d51511209f7476933c34a0c2c711078d3c80e74e0fe4f38680ec", size = 211309, upload-time = "2025-10-02T14:36:45.307Z" },
+    { url = "https://files.pythonhosted.org/packages/63/88/b284c6a128d88dc47f201957f926e707db79fb7415a87072e15c0e490de0/xxhash-3.6.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:780b90c313348f030b811efc37b0fa1431163cb8db8064cf88a7936b6ce5f222", size = 444480, upload-time = "2025-10-02T14:36:47.226Z" },
+    { url = "https://files.pythonhosted.org/packages/87/e4/798293a2bf9e4fac5f6d53ce59cba4739930778dfc6c7c73f40044ab0e6e/xxhash-3.6.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:18b242455eccdfcd1fa4134c431a30737d2b4f045770f8fe84356b3469d4b919", size = 192957, upload-time = "2025-10-02T14:36:48.968Z" },
+    { url = "https://files.pythonhosted.org/packages/78/55/bfd0d7db447a927897469048b953caececa3532e743b940dd1f5c1032d24/xxhash-3.6.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:a75ffc1bd5def584129774c158e108e5d768e10b75813f2b32650bb041066ed6", size = 209850, upload-time = "2025-10-02T14:36:50.258Z" },
+    { url = "https://files.pythonhosted.org/packages/31/06/d08ef9a792bfebfd2fb2bcbf04a541ad283bef74749ead6f089a0809d288/xxhash-3.6.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:1fc1ed882d1e8df932a66e2999429ba6cc4d5172914c904ab193381fba825360", size = 197342, upload-time = "2025-10-02T14:36:51.651Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/1a/aebf90797c94e9ca407c28e23f54d71f7149d91a93406a08a09e44d06994/xxhash-3.6.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:44e342e8cc11b4e79dae5c57f2fb6360c3c20cc57d32049af8f567f5b4bcb5f4", size = 209757, upload-time = "2025-10-02T14:36:53.009Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/80/799eec3d0a144dc3edf8c19b4f139c27fb923c50b34352796089ca206429/xxhash-3.6.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:c2f9ccd5c4be370939a2e17602fbc49995299203da72a3429db013d44d590e86", size = 412773, upload-time = "2025-10-02T14:36:54.691Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/f9/09df7545699de09219a205123b8463ce9ea83f48acc7aeeba0269507f9d3/xxhash-3.6.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:02ea4cb627c76f48cd9fb37cf7ab22bd51e57e1b519807234b473faebe526796", size = 190357, upload-time = "2025-10-02T14:36:56.363Z" },
+    { url = "https://files.pythonhosted.org/packages/07/40/2f8327f94e64a3f34d6ce3347c55207c322abbc80ae486ea45df4c62e7b3/xxhash-3.6.0-cp39-cp39-win32.whl", hash = "sha256:6551880383f0e6971dc23e512c9ccc986147ce7bfa1cd2e4b520b876c53e9f3d", size = 30585, upload-time = "2025-10-02T14:36:57.664Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/c8/2ecbc6799be9c02e8bf7b5a66cd94832b6ac13d59808746f0d402481c6ad/xxhash-3.6.0-cp39-cp39-win_amd64.whl", hash = "sha256:7c35c4cdc65f2a29f34425c446f2f5cdcd0e3c34158931e1cc927ece925ab802", size = 31512, upload-time = "2025-10-02T14:36:58.837Z" },
+    { url = "https://files.pythonhosted.org/packages/19/94/1d5459a9c587c94d7b8bcc710bd08bbfa145cbd814ebde41b48494362a21/xxhash-3.6.0-cp39-cp39-win_arm64.whl", hash = "sha256:ffc578717a347baf25be8397cb10d2528802d24f94cfc005c0e44fef44b5cdd6", size = 27878, upload-time = "2025-10-02T14:37:00.201Z" },
+    { url = "https://files.pythonhosted.org/packages/93/1e/8aec23647a34a249f62e2398c42955acd9b4c6ed5cf08cbea94dc46f78d2/xxhash-3.6.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:0f7b7e2ec26c1666ad5fc9dbfa426a6a3367ceaf79db5dd76264659d509d73b0", size = 30662, upload-time = "2025-10-02T14:37:01.743Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/0b/b14510b38ba91caf43006209db846a696ceea6a847a0c9ba0a5b1adc53d6/xxhash-3.6.0-pp311-pypy311_pp73-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:5dc1e14d14fa0f5789ec29a7062004b5933964bb9b02aae6622b8f530dc40296", size = 41056, upload-time = "2025-10-02T14:37:02.879Z" },
+    { url = "https://files.pythonhosted.org/packages/50/55/15a7b8a56590e66ccd374bbfa3f9ffc45b810886c8c3b614e3f90bd2367c/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:881b47fc47e051b37d94d13e7455131054b56749b91b508b0907eb07900d1c13", size = 36251, upload-time = "2025-10-02T14:37:04.44Z" },
+    { url = "https://files.pythonhosted.org/packages/62/b2/5ac99a041a29e58e95f907876b04f7067a0242cb85b5f39e726153981503/xxhash-3.6.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c6dc31591899f5e5666f04cc2e529e69b4072827085c1ef15294d91a004bc1bd", size = 32481, upload-time = "2025-10-02T14:37:05.869Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/d9/8d95e906764a386a3d3b596f3c68bb63687dfca806373509f51ce8eea81f/xxhash-3.6.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:15e0dac10eb9309508bfc41f7f9deaa7755c69e35af835db9cb10751adebc35d", size = 31565, upload-time = "2025-10-02T14:37:06.966Z" },
 ]
 
 [[package]]
 name = "yarl"
-version = "1.20.1"
+version = "1.22.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "idna" },
     { name = "multidict" },
     { name = "propcache" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/3c/fb/efaa23fa4e45537b827620f04cf8f3cd658b76642205162e072703a5b963/yarl-1.20.1.tar.gz", hash = "sha256:d017a4997ee50c91fd5466cef416231bb82177b93b029906cefc542ce14c35ac", size = 186428, upload-time = "2025-06-10T00:46:09.923Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/cb/65/7fed0d774abf47487c64be14e9223749468922817b5e8792b8a64792a1bb/yarl-1.20.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:6032e6da6abd41e4acda34d75a816012717000fa6839f37124a47fcefc49bec4", size = 132910, upload-time = "2025-06-10T00:42:31.108Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/7b/988f55a52da99df9e56dc733b8e4e5a6ae2090081dc2754fc8fd34e60aa0/yarl-1.20.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2c7b34d804b8cf9b214f05015c4fee2ebe7ed05cf581e7192c06555c71f4446a", size = 90644, upload-time = "2025-06-10T00:42:33.851Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/de/30d98f03e95d30c7e3cc093759982d038c8833ec2451001d45ef4854edc1/yarl-1.20.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0c869f2651cc77465f6cd01d938d91a11d9ea5d798738c1dc077f3de0b5e5fed", size = 89322, upload-time = "2025-06-10T00:42:35.688Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/7a/f2f314f5ebfe9200724b0b748de2186b927acb334cf964fd312eb86fc286/yarl-1.20.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:62915e6688eb4d180d93840cda4110995ad50c459bf931b8b3775b37c264af1e", size = 323786, upload-time = "2025-06-10T00:42:37.817Z" },
-    { url = "https://files.pythonhosted.org/packages/15/3f/718d26f189db96d993d14b984ce91de52e76309d0fd1d4296f34039856aa/yarl-1.20.1-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:41ebd28167bc6af8abb97fec1a399f412eec5fd61a3ccbe2305a18b84fb4ca73", size = 319627, upload-time = "2025-06-10T00:42:39.937Z" },
-    { url = "https://files.pythonhosted.org/packages/a5/76/8fcfbf5fa2369157b9898962a4a7d96764b287b085b5b3d9ffae69cdefd1/yarl-1.20.1-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:21242b4288a6d56f04ea193adde174b7e347ac46ce6bc84989ff7c1b1ecea84e", size = 339149, upload-time = "2025-06-10T00:42:42.627Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/95/d7fc301cc4661785967acc04f54a4a42d5124905e27db27bb578aac49b5c/yarl-1.20.1-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bea21cdae6c7eb02ba02a475f37463abfe0a01f5d7200121b03e605d6a0439f8", size = 333327, upload-time = "2025-06-10T00:42:44.842Z" },
-    { url = "https://files.pythonhosted.org/packages/65/94/e21269718349582eee81efc5c1c08ee71c816bfc1585b77d0ec3f58089eb/yarl-1.20.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1f8a891e4a22a89f5dde7862994485e19db246b70bb288d3ce73a34422e55b23", size = 326054, upload-time = "2025-06-10T00:42:47.149Z" },
-    { url = "https://files.pythonhosted.org/packages/32/ae/8616d1f07853704523519f6131d21f092e567c5af93de7e3e94b38d7f065/yarl-1.20.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:dd803820d44c8853a109a34e3660e5a61beae12970da479cf44aa2954019bf70", size = 315035, upload-time = "2025-06-10T00:42:48.852Z" },
-    { url = "https://files.pythonhosted.org/packages/48/aa/0ace06280861ef055855333707db5e49c6e3a08840a7ce62682259d0a6c0/yarl-1.20.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:b982fa7f74c80d5c0c7b5b38f908971e513380a10fecea528091405f519b9ebb", size = 338962, upload-time = "2025-06-10T00:42:51.024Z" },
-    { url = "https://files.pythonhosted.org/packages/20/52/1e9d0e6916f45a8fb50e6844f01cb34692455f1acd548606cbda8134cd1e/yarl-1.20.1-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:33f29ecfe0330c570d997bcf1afd304377f2e48f61447f37e846a6058a4d33b2", size = 335399, upload-time = "2025-06-10T00:42:53.007Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/65/60452df742952c630e82f394cd409de10610481d9043aa14c61bf846b7b1/yarl-1.20.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:835ab2cfc74d5eb4a6a528c57f05688099da41cf4957cf08cad38647e4a83b30", size = 338649, upload-time = "2025-06-10T00:42:54.964Z" },
-    { url = "https://files.pythonhosted.org/packages/7b/f5/6cd4ff38dcde57a70f23719a838665ee17079640c77087404c3d34da6727/yarl-1.20.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:46b5e0ccf1943a9a6e766b2c2b8c732c55b34e28be57d8daa2b3c1d1d4009309", size = 358563, upload-time = "2025-06-10T00:42:57.28Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/90/c42eefd79d0d8222cb3227bdd51b640c0c1d0aa33fe4cc86c36eccba77d3/yarl-1.20.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:df47c55f7d74127d1b11251fe6397d84afdde0d53b90bedb46a23c0e534f9d24", size = 357609, upload-time = "2025-06-10T00:42:59.055Z" },
-    { url = "https://files.pythonhosted.org/packages/03/c8/cea6b232cb4617514232e0f8a718153a95b5d82b5290711b201545825532/yarl-1.20.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:76d12524d05841276b0e22573f28d5fbcb67589836772ae9244d90dd7d66aa13", size = 350224, upload-time = "2025-06-10T00:43:01.248Z" },
-    { url = "https://files.pythonhosted.org/packages/ce/a3/eaa0ab9712f1f3d01faf43cf6f1f7210ce4ea4a7e9b28b489a2261ca8db9/yarl-1.20.1-cp310-cp310-win32.whl", hash = "sha256:6c4fbf6b02d70e512d7ade4b1f998f237137f1417ab07ec06358ea04f69134f8", size = 81753, upload-time = "2025-06-10T00:43:03.486Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/34/e4abde70a9256465fe31c88ed02c3f8502b7b5dead693a4f350a06413f28/yarl-1.20.1-cp310-cp310-win_amd64.whl", hash = "sha256:aef6c4d69554d44b7f9d923245f8ad9a707d971e6209d51279196d8e8fe1ae16", size = 86817, upload-time = "2025-06-10T00:43:05.231Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/18/893b50efc2350e47a874c5c2d67e55a0ea5df91186b2a6f5ac52eff887cd/yarl-1.20.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:47ee6188fea634bdfaeb2cc420f5b3b17332e6225ce88149a17c413c77ff269e", size = 133833, upload-time = "2025-06-10T00:43:07.393Z" },
-    { url = "https://files.pythonhosted.org/packages/89/ed/b8773448030e6fc47fa797f099ab9eab151a43a25717f9ac043844ad5ea3/yarl-1.20.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:d0f6500f69e8402d513e5eedb77a4e1818691e8f45e6b687147963514d84b44b", size = 91070, upload-time = "2025-06-10T00:43:09.538Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/e3/409bd17b1e42619bf69f60e4f031ce1ccb29bd7380117a55529e76933464/yarl-1.20.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7a8900a42fcdaad568de58887c7b2f602962356908eedb7628eaf6021a6e435b", size = 89818, upload-time = "2025-06-10T00:43:11.575Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/77/64d8431a4d77c856eb2d82aa3de2ad6741365245a29b3a9543cd598ed8c5/yarl-1.20.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bad6d131fda8ef508b36be3ece16d0902e80b88ea7200f030a0f6c11d9e508d4", size = 347003, upload-time = "2025-06-10T00:43:14.088Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/d2/0c7e4def093dcef0bd9fa22d4d24b023788b0a33b8d0088b51aa51e21e99/yarl-1.20.1-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:df018d92fe22aaebb679a7f89fe0c0f368ec497e3dda6cb81a567610f04501f1", size = 336537, upload-time = "2025-06-10T00:43:16.431Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/f3/fc514f4b2cf02cb59d10cbfe228691d25929ce8f72a38db07d3febc3f706/yarl-1.20.1-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:8f969afbb0a9b63c18d0feecf0db09d164b7a44a053e78a7d05f5df163e43833", size = 362358, upload-time = "2025-06-10T00:43:18.704Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/6d/a313ac8d8391381ff9006ac05f1d4331cee3b1efaa833a53d12253733255/yarl-1.20.1-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:812303eb4aa98e302886ccda58d6b099e3576b1b9276161469c25803a8db277d", size = 357362, upload-time = "2025-06-10T00:43:20.888Z" },
-    { url = "https://files.pythonhosted.org/packages/00/70/8f78a95d6935a70263d46caa3dd18e1f223cf2f2ff2037baa01a22bc5b22/yarl-1.20.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98c4a7d166635147924aa0bf9bfe8d8abad6fffa6102de9c99ea04a1376f91e8", size = 348979, upload-time = "2025-06-10T00:43:23.169Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/05/42773027968968f4f15143553970ee36ead27038d627f457cc44bbbeecf3/yarl-1.20.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:12e768f966538e81e6e7550f9086a6236b16e26cd964cf4df35349970f3551cf", size = 337274, upload-time = "2025-06-10T00:43:27.111Z" },
-    { url = "https://files.pythonhosted.org/packages/05/be/665634aa196954156741ea591d2f946f1b78ceee8bb8f28488bf28c0dd62/yarl-1.20.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:fe41919b9d899661c5c28a8b4b0acf704510b88f27f0934ac7a7bebdd8938d5e", size = 363294, upload-time = "2025-06-10T00:43:28.96Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/90/73448401d36fa4e210ece5579895731f190d5119c4b66b43b52182e88cd5/yarl-1.20.1-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:8601bc010d1d7780592f3fc1bdc6c72e2b6466ea34569778422943e1a1f3c389", size = 358169, upload-time = "2025-06-10T00:43:30.701Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/b0/fce922d46dc1eb43c811f1889f7daa6001b27a4005587e94878570300881/yarl-1.20.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:daadbdc1f2a9033a2399c42646fbd46da7992e868a5fe9513860122d7fe7a73f", size = 362776, upload-time = "2025-06-10T00:43:32.51Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/0d/b172628fce039dae8977fd22caeff3eeebffd52e86060413f5673767c427/yarl-1.20.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:03aa1e041727cb438ca762628109ef1333498b122e4c76dd858d186a37cec845", size = 381341, upload-time = "2025-06-10T00:43:34.543Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/9b/5b886d7671f4580209e855974fe1cecec409aa4a89ea58b8f0560dc529b1/yarl-1.20.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:642980ef5e0fa1de5fa96d905c7e00cb2c47cb468bfcac5a18c58e27dbf8d8d1", size = 379988, upload-time = "2025-06-10T00:43:36.489Z" },
-    { url = "https://files.pythonhosted.org/packages/73/be/75ef5fd0fcd8f083a5d13f78fd3f009528132a1f2a1d7c925c39fa20aa79/yarl-1.20.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:86971e2795584fe8c002356d3b97ef6c61862720eeff03db2a7c86b678d85b3e", size = 371113, upload-time = "2025-06-10T00:43:38.592Z" },
-    { url = "https://files.pythonhosted.org/packages/50/4f/62faab3b479dfdcb741fe9e3f0323e2a7d5cd1ab2edc73221d57ad4834b2/yarl-1.20.1-cp311-cp311-win32.whl", hash = "sha256:597f40615b8d25812f14562699e287f0dcc035d25eb74da72cae043bb884d773", size = 81485, upload-time = "2025-06-10T00:43:41.038Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/09/d9c7942f8f05c32ec72cd5c8e041c8b29b5807328b68b4801ff2511d4d5e/yarl-1.20.1-cp311-cp311-win_amd64.whl", hash = "sha256:26ef53a9e726e61e9cd1cda6b478f17e350fb5800b4bd1cd9fe81c4d91cfeb2e", size = 86686, upload-time = "2025-06-10T00:43:42.692Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/9a/cb7fad7d73c69f296eda6815e4a2c7ed53fc70c2f136479a91c8e5fbdb6d/yarl-1.20.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:bdcc4cd244e58593a4379fe60fdee5ac0331f8eb70320a24d591a3be197b94a9", size = 133667, upload-time = "2025-06-10T00:43:44.369Z" },
-    { url = "https://files.pythonhosted.org/packages/67/38/688577a1cb1e656e3971fb66a3492501c5a5df56d99722e57c98249e5b8a/yarl-1.20.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:b29a2c385a5f5b9c7d9347e5812b6f7ab267193c62d282a540b4fc528c8a9d2a", size = 91025, upload-time = "2025-06-10T00:43:46.295Z" },
-    { url = "https://files.pythonhosted.org/packages/50/ec/72991ae51febeb11a42813fc259f0d4c8e0507f2b74b5514618d8b640365/yarl-1.20.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1112ae8154186dfe2de4732197f59c05a83dc814849a5ced892b708033f40dc2", size = 89709, upload-time = "2025-06-10T00:43:48.22Z" },
-    { url = "https://files.pythonhosted.org/packages/99/da/4d798025490e89426e9f976702e5f9482005c548c579bdae792a4c37769e/yarl-1.20.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:90bbd29c4fe234233f7fa2b9b121fb63c321830e5d05b45153a2ca68f7d310ee", size = 352287, upload-time = "2025-06-10T00:43:49.924Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/26/54a15c6a567aac1c61b18aa0f4b8aa2e285a52d547d1be8bf48abe2b3991/yarl-1.20.1-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:680e19c7ce3710ac4cd964e90dad99bf9b5029372ba0c7cbfcd55e54d90ea819", size = 345429, upload-time = "2025-06-10T00:43:51.7Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/95/9dcf2386cb875b234353b93ec43e40219e14900e046bf6ac118f94b1e353/yarl-1.20.1-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:4a979218c1fdb4246a05efc2cc23859d47c89af463a90b99b7c56094daf25a16", size = 365429, upload-time = "2025-06-10T00:43:53.494Z" },
-    { url = "https://files.pythonhosted.org/packages/91/b2/33a8750f6a4bc224242a635f5f2cff6d6ad5ba651f6edcccf721992c21a0/yarl-1.20.1-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:255b468adf57b4a7b65d8aad5b5138dce6a0752c139965711bdcb81bc370e1b6", size = 363862, upload-time = "2025-06-10T00:43:55.766Z" },
-    { url = "https://files.pythonhosted.org/packages/98/28/3ab7acc5b51f4434b181b0cee8f1f4b77a65919700a355fb3617f9488874/yarl-1.20.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a97d67108e79cfe22e2b430d80d7571ae57d19f17cda8bb967057ca8a7bf5bfd", size = 355616, upload-time = "2025-06-10T00:43:58.056Z" },
-    { url = "https://files.pythonhosted.org/packages/36/a3/f666894aa947a371724ec7cd2e5daa78ee8a777b21509b4252dd7bd15e29/yarl-1.20.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:8570d998db4ddbfb9a590b185a0a33dbf8aafb831d07a5257b4ec9948df9cb0a", size = 339954, upload-time = "2025-06-10T00:43:59.773Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/81/5f466427e09773c04219d3450d7a1256138a010b6c9f0af2d48565e9ad13/yarl-1.20.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:97c75596019baae7c71ccf1d8cc4738bc08134060d0adfcbe5642f778d1dca38", size = 365575, upload-time = "2025-06-10T00:44:02.051Z" },
-    { url = "https://files.pythonhosted.org/packages/2e/e3/e4b0ad8403e97e6c9972dd587388940a032f030ebec196ab81a3b8e94d31/yarl-1.20.1-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:1c48912653e63aef91ff988c5432832692ac5a1d8f0fb8a33091520b5bbe19ef", size = 365061, upload-time = "2025-06-10T00:44:04.196Z" },
-    { url = "https://files.pythonhosted.org/packages/ac/99/b8a142e79eb86c926f9f06452eb13ecb1bb5713bd01dc0038faf5452e544/yarl-1.20.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:4c3ae28f3ae1563c50f3d37f064ddb1511ecc1d5584e88c6b7c63cf7702a6d5f", size = 364142, upload-time = "2025-06-10T00:44:06.527Z" },
-    { url = "https://files.pythonhosted.org/packages/34/f2/08ed34a4a506d82a1a3e5bab99ccd930a040f9b6449e9fd050320e45845c/yarl-1.20.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:c5e9642f27036283550f5f57dc6156c51084b458570b9d0d96100c8bebb186a8", size = 381894, upload-time = "2025-06-10T00:44:08.379Z" },
-    { url = "https://files.pythonhosted.org/packages/92/f8/9a3fbf0968eac704f681726eff595dce9b49c8a25cd92bf83df209668285/yarl-1.20.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:2c26b0c49220d5799f7b22c6838409ee9bc58ee5c95361a4d7831f03cc225b5a", size = 383378, upload-time = "2025-06-10T00:44:10.51Z" },
-    { url = "https://files.pythonhosted.org/packages/af/85/9363f77bdfa1e4d690957cd39d192c4cacd1c58965df0470a4905253b54f/yarl-1.20.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:564ab3d517e3d01c408c67f2e5247aad4019dcf1969982aba3974b4093279004", size = 374069, upload-time = "2025-06-10T00:44:12.834Z" },
-    { url = "https://files.pythonhosted.org/packages/35/99/9918c8739ba271dcd935400cff8b32e3cd319eaf02fcd023d5dcd487a7c8/yarl-1.20.1-cp312-cp312-win32.whl", hash = "sha256:daea0d313868da1cf2fac6b2d3a25c6e3a9e879483244be38c8e6a41f1d876a5", size = 81249, upload-time = "2025-06-10T00:44:14.731Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/83/5d9092950565481b413b31a23e75dd3418ff0a277d6e0abf3729d4d1ce25/yarl-1.20.1-cp312-cp312-win_amd64.whl", hash = "sha256:48ea7d7f9be0487339828a4de0360d7ce0efc06524a48e1810f945c45b813698", size = 86710, upload-time = "2025-06-10T00:44:16.716Z" },
-    { url = "https://files.pythonhosted.org/packages/01/75/0d37402d208d025afa6b5b8eb80e466d267d3fd1927db8e317d29a94a4cb/yarl-1.20.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:e42ba79e2efb6845ebab49c7bf20306c4edf74a0b20fc6b2ccdd1a219d12fad3", size = 134259, upload-time = "2025-06-10T00:45:29.882Z" },
-    { url = "https://files.pythonhosted.org/packages/73/84/1fb6c85ae0cf9901046f07d0ac9eb162f7ce6d95db541130aa542ed377e6/yarl-1.20.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:41493b9b7c312ac448b7f0a42a089dffe1d6e6e981a2d76205801a023ed26a2b", size = 91269, upload-time = "2025-06-10T00:45:32.917Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/9c/eae746b24c4ea29a5accba9a06c197a70fa38a49c7df244e0d3951108861/yarl-1.20.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:f5a5928ff5eb13408c62a968ac90d43f8322fd56d87008b8f9dabf3c0f6ee983", size = 89995, upload-time = "2025-06-10T00:45:35.066Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/30/693e71003ec4bc1daf2e4cf7c478c417d0985e0a8e8f00b2230d517876fc/yarl-1.20.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:30c41ad5d717b3961b2dd785593b67d386b73feca30522048d37298fee981805", size = 325253, upload-time = "2025-06-10T00:45:37.052Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/a2/5264dbebf90763139aeb0b0b3154763239398400f754ae19a0518b654117/yarl-1.20.1-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:59febc3969b0781682b469d4aca1a5cab7505a4f7b85acf6db01fa500fa3f6ba", size = 320897, upload-time = "2025-06-10T00:45:39.962Z" },
-    { url = "https://files.pythonhosted.org/packages/e7/17/77c7a89b3c05856489777e922f41db79ab4faf58621886df40d812c7facd/yarl-1.20.1-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d2b6fb3622b7e5bf7a6e5b679a69326b4279e805ed1699d749739a61d242449e", size = 340696, upload-time = "2025-06-10T00:45:41.915Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/55/28409330b8ef5f2f681f5b478150496ec9cf3309b149dab7ec8ab5cfa3f0/yarl-1.20.1-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:749d73611db8d26a6281086f859ea7ec08f9c4c56cec864e52028c8b328db723", size = 335064, upload-time = "2025-06-10T00:45:43.893Z" },
-    { url = "https://files.pythonhosted.org/packages/85/58/cb0257cbd4002828ff735f44d3c5b6966c4fd1fc8cc1cd3cd8a143fbc513/yarl-1.20.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9427925776096e664c39e131447aa20ec738bdd77c049c48ea5200db2237e000", size = 327256, upload-time = "2025-06-10T00:45:46.393Z" },
-    { url = "https://files.pythonhosted.org/packages/53/f6/c77960370cfa46f6fb3d6a5a79a49d3abfdb9ef92556badc2dcd2748bc2a/yarl-1.20.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ff70f32aa316393eaf8222d518ce9118148eddb8a53073c2403863b41033eed5", size = 316389, upload-time = "2025-06-10T00:45:48.358Z" },
-    { url = "https://files.pythonhosted.org/packages/64/ab/be0b10b8e029553c10905b6b00c64ecad3ebc8ace44b02293a62579343f6/yarl-1.20.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:c7ddf7a09f38667aea38801da8b8d6bfe81df767d9dfc8c88eb45827b195cd1c", size = 340481, upload-time = "2025-06-10T00:45:50.663Z" },
-    { url = "https://files.pythonhosted.org/packages/c5/c3/3f327bd3905a4916029bf5feb7f86dcf864c7704f099715f62155fb386b2/yarl-1.20.1-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:57edc88517d7fc62b174fcfb2e939fbc486a68315d648d7e74d07fac42cec240", size = 336941, upload-time = "2025-06-10T00:45:52.554Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/42/040bdd5d3b3bb02b4a6ace4ed4075e02f85df964d6e6cb321795d2a6496a/yarl-1.20.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:dab096ce479d5894d62c26ff4f699ec9072269d514b4edd630a393223f45a0ee", size = 339936, upload-time = "2025-06-10T00:45:54.919Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/1c/911867b8e8c7463b84dfdc275e0d99b04b66ad5132b503f184fe76be8ea4/yarl-1.20.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:14a85f3bd2d7bb255be7183e5d7d6e70add151a98edf56a770d6140f5d5f4010", size = 360163, upload-time = "2025-06-10T00:45:56.87Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/31/8c389f6c6ca0379b57b2da87f1f126c834777b4931c5ee8427dd65d0ff6b/yarl-1.20.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:2c89b5c792685dd9cd3fa9761c1b9f46fc240c2a3265483acc1565769996a3f8", size = 359108, upload-time = "2025-06-10T00:45:58.869Z" },
-    { url = "https://files.pythonhosted.org/packages/7f/09/ae4a649fb3964324c70a3e2b61f45e566d9ffc0affd2b974cbf628957673/yarl-1.20.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:69e9b141de5511021942a6866990aea6d111c9042235de90e08f94cf972ca03d", size = 351875, upload-time = "2025-06-10T00:46:01.45Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/43/bbb4ed4c34d5bb62b48bf957f68cd43f736f79059d4f85225ab1ef80f4b9/yarl-1.20.1-cp39-cp39-win32.whl", hash = "sha256:b5f307337819cdfdbb40193cad84978a029f847b0a357fbe49f712063cfc4f06", size = 82293, upload-time = "2025-06-10T00:46:03.763Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/cd/ce185848a7dba68ea69e932674b5c1a42a1852123584bccc5443120f857c/yarl-1.20.1-cp39-cp39-win_amd64.whl", hash = "sha256:eae7bfe2069f9c1c5b05fc7fe5d612e5bbc089a39309904ee8b829e322dcad00", size = 87385, upload-time = "2025-06-10T00:46:05.655Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/2d/2345fce04cfd4bee161bf1e7d9cdc702e3e16109021035dbb24db654a622/yarl-1.20.1-py3-none-any.whl", hash = "sha256:83b8eb083fe4683c6115795d9fc1cfaf2cbbefb19b3a1cb68f6527460f483a77", size = 46542, upload-time = "2025-06-10T00:46:07.521Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/57/63/0c6ebca57330cd313f6102b16dd57ffaf3ec4c83403dcb45dbd15c6f3ea1/yarl-1.22.0.tar.gz", hash = "sha256:bebf8557577d4401ba8bd9ff33906f1376c877aa78d1fe216ad01b4d6745af71", size = 187169, upload-time = "2025-10-06T14:12:55.963Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d1/43/a2204825342f37c337f5edb6637040fa14e365b2fcc2346960201d457579/yarl-1.22.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:c7bd6683587567e5a49ee6e336e0612bec8329be1b7d4c8af5687dcdeb67ee1e", size = 140517, upload-time = "2025-10-06T14:08:42.494Z" },
+    { url = "https://files.pythonhosted.org/packages/44/6f/674f3e6f02266428c56f704cd2501c22f78e8b2eeb23f153117cc86fb28a/yarl-1.22.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:5cdac20da754f3a723cceea5b3448e1a2074866406adeb4ef35b469d089adb8f", size = 93495, upload-time = "2025-10-06T14:08:46.2Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/12/5b274d8a0f30c07b91b2f02cba69152600b47830fcfb465c108880fcee9c/yarl-1.22.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:07a524d84df0c10f41e3ee918846e1974aba4ec017f990dc735aad487a0bdfdf", size = 94400, upload-time = "2025-10-06T14:08:47.855Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/7f/df1b6949b1fa1aa9ff6de6e2631876ad4b73c4437822026e85d8acb56bb1/yarl-1.22.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e1b329cb8146d7b736677a2440e422eadd775d1806a81db2d4cded80a48efc1a", size = 347545, upload-time = "2025-10-06T14:08:49.683Z" },
+    { url = "https://files.pythonhosted.org/packages/84/09/f92ed93bd6cd77872ab6c3462df45ca45cd058d8f1d0c9b4f54c1704429f/yarl-1.22.0-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:75976c6945d85dbb9ee6308cd7ff7b1fb9409380c82d6119bd778d8fcfe2931c", size = 319598, upload-time = "2025-10-06T14:08:51.215Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/97/ac3f3feae7d522cf7ccec3d340bb0b2b61c56cb9767923df62a135092c6b/yarl-1.22.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:80ddf7a5f8c86cb3eb4bc9028b07bbbf1f08a96c5c0bc1244be5e8fefcb94147", size = 363893, upload-time = "2025-10-06T14:08:53.144Z" },
+    { url = "https://files.pythonhosted.org/packages/06/49/f3219097403b9c84a4d079b1d7bda62dd9b86d0d6e4428c02d46ab2c77fc/yarl-1.22.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:d332fc2e3c94dad927f2112395772a4e4fedbcf8f80efc21ed7cdfae4d574fdb", size = 371240, upload-time = "2025-10-06T14:08:55.036Z" },
+    { url = "https://files.pythonhosted.org/packages/35/9f/06b765d45c0e44e8ecf0fe15c9eacbbde342bb5b7561c46944f107bfb6c3/yarl-1.22.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0cf71bf877efeac18b38d3930594c0948c82b64547c1cf420ba48722fe5509f6", size = 346965, upload-time = "2025-10-06T14:08:56.722Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/69/599e7cea8d0fcb1694323b0db0dda317fa3162f7b90166faddecf532166f/yarl-1.22.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:663e1cadaddae26be034a6ab6072449a8426ddb03d500f43daf952b74553bba0", size = 342026, upload-time = "2025-10-06T14:08:58.563Z" },
+    { url = "https://files.pythonhosted.org/packages/95/6f/9dfd12c8bc90fea9eab39832ee32ea48f8e53d1256252a77b710c065c89f/yarl-1.22.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:6dcbb0829c671f305be48a7227918cfcd11276c2d637a8033a99a02b67bf9eda", size = 335637, upload-time = "2025-10-06T14:09:00.506Z" },
+    { url = "https://files.pythonhosted.org/packages/57/2e/34c5b4eb9b07e16e873db5b182c71e5f06f9b5af388cdaa97736d79dd9a6/yarl-1.22.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:f0d97c18dfd9a9af4490631905a3f131a8e4c9e80a39353919e2cfed8f00aedc", size = 359082, upload-time = "2025-10-06T14:09:01.936Z" },
+    { url = "https://files.pythonhosted.org/packages/31/71/fa7e10fb772d273aa1f096ecb8ab8594117822f683bab7d2c5a89914c92a/yarl-1.22.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:437840083abe022c978470b942ff832c3940b2ad3734d424b7eaffcd07f76737", size = 357811, upload-time = "2025-10-06T14:09:03.445Z" },
+    { url = "https://files.pythonhosted.org/packages/26/da/11374c04e8e1184a6a03cf9c8f5688d3e5cec83ed6f31ad3481b3207f709/yarl-1.22.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:a899cbd98dce6f5d8de1aad31cb712ec0a530abc0a86bd6edaa47c1090138467", size = 351223, upload-time = "2025-10-06T14:09:05.401Z" },
+    { url = "https://files.pythonhosted.org/packages/82/8f/e2d01f161b0c034a30410e375e191a5d27608c1f8693bab1a08b089ca096/yarl-1.22.0-cp310-cp310-win32.whl", hash = "sha256:595697f68bd1f0c1c159fcb97b661fc9c3f5db46498043555d04805430e79bea", size = 82118, upload-time = "2025-10-06T14:09:11.148Z" },
+    { url = "https://files.pythonhosted.org/packages/62/46/94c76196642dbeae634c7a61ba3da88cd77bed875bf6e4a8bed037505aa6/yarl-1.22.0-cp310-cp310-win_amd64.whl", hash = "sha256:cb95a9b1adaa48e41815a55ae740cfda005758104049a640a398120bf02515ca", size = 86852, upload-time = "2025-10-06T14:09:12.958Z" },
+    { url = "https://files.pythonhosted.org/packages/af/af/7df4f179d3b1a6dcb9a4bd2ffbc67642746fcafdb62580e66876ce83fff4/yarl-1.22.0-cp310-cp310-win_arm64.whl", hash = "sha256:b85b982afde6df99ecc996990d4ad7ccbdbb70e2a4ba4de0aecde5922ba98a0b", size = 82012, upload-time = "2025-10-06T14:09:14.664Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/27/5ab13fc84c76a0250afd3d26d5936349a35be56ce5785447d6c423b26d92/yarl-1.22.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:1ab72135b1f2db3fed3997d7e7dc1b80573c67138023852b6efb336a5eae6511", size = 141607, upload-time = "2025-10-06T14:09:16.298Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/a1/d065d51d02dc02ce81501d476b9ed2229d9a990818332242a882d5d60340/yarl-1.22.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:669930400e375570189492dc8d8341301578e8493aec04aebc20d4717f899dd6", size = 94027, upload-time = "2025-10-06T14:09:17.786Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/da/8da9f6a53f67b5106ffe902c6fa0164e10398d4e150d85838b82f424072a/yarl-1.22.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:792a2af6d58177ef7c19cbf0097aba92ca1b9cb3ffdd9c7470e156c8f9b5e028", size = 94963, upload-time = "2025-10-06T14:09:19.662Z" },
+    { url = "https://files.pythonhosted.org/packages/68/fe/2c1f674960c376e29cb0bec1249b117d11738db92a6ccc4a530b972648db/yarl-1.22.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3ea66b1c11c9150f1372f69afb6b8116f2dd7286f38e14ea71a44eee9ec51b9d", size = 368406, upload-time = "2025-10-06T14:09:21.402Z" },
+    { url = "https://files.pythonhosted.org/packages/95/26/812a540e1c3c6418fec60e9bbd38e871eaba9545e94fa5eff8f4a8e28e1e/yarl-1.22.0-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3e2daa88dc91870215961e96a039ec73e4937da13cf77ce17f9cad0c18df3503", size = 336581, upload-time = "2025-10-06T14:09:22.98Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/f5/5777b19e26fdf98563985e481f8be3d8a39f8734147a6ebf459d0dab5a6b/yarl-1.22.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ba440ae430c00eee41509353628600212112cd5018d5def7e9b05ea7ac34eb65", size = 388924, upload-time = "2025-10-06T14:09:24.655Z" },
+    { url = "https://files.pythonhosted.org/packages/86/08/24bd2477bd59c0bbd994fe1d93b126e0472e4e3df5a96a277b0a55309e89/yarl-1.22.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:e6438cc8f23a9c1478633d216b16104a586b9761db62bfacb6425bac0a36679e", size = 392890, upload-time = "2025-10-06T14:09:26.617Z" },
+    { url = "https://files.pythonhosted.org/packages/46/00/71b90ed48e895667ecfb1eaab27c1523ee2fa217433ed77a73b13205ca4b/yarl-1.22.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4c52a6e78aef5cf47a98ef8e934755abf53953379b7d53e68b15ff4420e6683d", size = 365819, upload-time = "2025-10-06T14:09:28.544Z" },
+    { url = "https://files.pythonhosted.org/packages/30/2d/f715501cae832651d3282387c6a9236cd26bd00d0ff1e404b3dc52447884/yarl-1.22.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:3b06bcadaac49c70f4c88af4ffcfbe3dc155aab3163e75777818092478bcbbe7", size = 363601, upload-time = "2025-10-06T14:09:30.568Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/f9/a678c992d78e394e7126ee0b0e4e71bd2775e4334d00a9278c06a6cce96a/yarl-1.22.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:6944b2dc72c4d7f7052683487e3677456050ff77fcf5e6204e98caf785ad1967", size = 358072, upload-time = "2025-10-06T14:09:32.528Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/d1/b49454411a60edb6fefdcad4f8e6dbba7d8019e3a508a1c5836cba6d0781/yarl-1.22.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:d5372ca1df0f91a86b047d1277c2aaf1edb32d78bbcefffc81b40ffd18f027ed", size = 385311, upload-time = "2025-10-06T14:09:34.634Z" },
+    { url = "https://files.pythonhosted.org/packages/87/e5/40d7a94debb8448c7771a916d1861d6609dddf7958dc381117e7ba36d9e8/yarl-1.22.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:51af598701f5299012b8416486b40fceef8c26fc87dc6d7d1f6fc30609ea0aa6", size = 381094, upload-time = "2025-10-06T14:09:36.268Z" },
+    { url = "https://files.pythonhosted.org/packages/35/d8/611cc282502381ad855448643e1ad0538957fc82ae83dfe7762c14069e14/yarl-1.22.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:b266bd01fedeffeeac01a79ae181719ff848a5a13ce10075adbefc8f1daee70e", size = 370944, upload-time = "2025-10-06T14:09:37.872Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/df/fadd00fb1c90e1a5a8bd731fa3d3de2e165e5a3666a095b04e31b04d9cb6/yarl-1.22.0-cp311-cp311-win32.whl", hash = "sha256:a9b1ba5610a4e20f655258d5a1fdc7ebe3d837bb0e45b581398b99eb98b1f5ca", size = 81804, upload-time = "2025-10-06T14:09:39.359Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/f7/149bb6f45f267cb5c074ac40c01c6b3ea6d8a620d34b337f6321928a1b4d/yarl-1.22.0-cp311-cp311-win_amd64.whl", hash = "sha256:078278b9b0b11568937d9509b589ee83ef98ed6d561dfe2020e24a9fd08eaa2b", size = 86858, upload-time = "2025-10-06T14:09:41.068Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/13/88b78b93ad3f2f0b78e13bfaaa24d11cbc746e93fe76d8c06bf139615646/yarl-1.22.0-cp311-cp311-win_arm64.whl", hash = "sha256:b6a6f620cfe13ccec221fa312139135166e47ae169f8253f72a0abc0dae94376", size = 81637, upload-time = "2025-10-06T14:09:42.712Z" },
+    { url = "https://files.pythonhosted.org/packages/75/ff/46736024fee3429b80a165a732e38e5d5a238721e634ab41b040d49f8738/yarl-1.22.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:e340382d1afa5d32b892b3ff062436d592ec3d692aeea3bef3a5cfe11bbf8c6f", size = 142000, upload-time = "2025-10-06T14:09:44.631Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/9a/b312ed670df903145598914770eb12de1bac44599549b3360acc96878df8/yarl-1.22.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f1e09112a2c31ffe8d80be1b0988fa6a18c5d5cad92a9ffbb1c04c91bfe52ad2", size = 94338, upload-time = "2025-10-06T14:09:46.372Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/f5/0601483296f09c3c65e303d60c070a5c19fcdbc72daa061e96170785bc7d/yarl-1.22.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:939fe60db294c786f6b7c2d2e121576628468f65453d86b0fe36cb52f987bd74", size = 94909, upload-time = "2025-10-06T14:09:48.648Z" },
+    { url = "https://files.pythonhosted.org/packages/60/41/9a1fe0b73dbcefce72e46cf149b0e0a67612d60bfc90fb59c2b2efdfbd86/yarl-1.22.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e1651bf8e0398574646744c1885a41198eba53dc8a9312b954073f845c90a8df", size = 372940, upload-time = "2025-10-06T14:09:50.089Z" },
+    { url = "https://files.pythonhosted.org/packages/17/7a/795cb6dfee561961c30b800f0ed616b923a2ec6258b5def2a00bf8231334/yarl-1.22.0-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:b8a0588521a26bf92a57a1705b77b8b59044cdceccac7151bd8d229e66b8dedb", size = 345825, upload-time = "2025-10-06T14:09:52.142Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/93/a58f4d596d2be2ae7bab1a5846c4d270b894958845753b2c606d666744d3/yarl-1.22.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:42188e6a615c1a75bcaa6e150c3fe8f3e8680471a6b10150c5f7e83f47cc34d2", size = 386705, upload-time = "2025-10-06T14:09:54.128Z" },
+    { url = "https://files.pythonhosted.org/packages/61/92/682279d0e099d0e14d7fd2e176bd04f48de1484f56546a3e1313cd6c8e7c/yarl-1.22.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f6d2cb59377d99718913ad9a151030d6f83ef420a2b8f521d94609ecc106ee82", size = 396518, upload-time = "2025-10-06T14:09:55.762Z" },
+    { url = "https://files.pythonhosted.org/packages/db/0f/0d52c98b8a885aeda831224b78f3be7ec2e1aa4a62091f9f9188c3c65b56/yarl-1.22.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:50678a3b71c751d58d7908edc96d332af328839eea883bb554a43f539101277a", size = 377267, upload-time = "2025-10-06T14:09:57.958Z" },
+    { url = "https://files.pythonhosted.org/packages/22/42/d2685e35908cbeaa6532c1fc73e89e7f2efb5d8a7df3959ea8e37177c5a3/yarl-1.22.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:1e8fbaa7cec507aa24ea27a01456e8dd4b6fab829059b69844bd348f2d467124", size = 365797, upload-time = "2025-10-06T14:09:59.527Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/83/cf8c7bcc6355631762f7d8bdab920ad09b82efa6b722999dfb05afa6cfac/yarl-1.22.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:433885ab5431bc3d3d4f2f9bd15bfa1614c522b0f1405d62c4f926ccd69d04fa", size = 365535, upload-time = "2025-10-06T14:10:01.139Z" },
+    { url = "https://files.pythonhosted.org/packages/25/e1/5302ff9b28f0c59cac913b91fe3f16c59a033887e57ce9ca5d41a3a94737/yarl-1.22.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:b790b39c7e9a4192dc2e201a282109ed2985a1ddbd5ac08dc56d0e121400a8f7", size = 382324, upload-time = "2025-10-06T14:10:02.756Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/cd/4617eb60f032f19ae3a688dc990d8f0d89ee0ea378b61cac81ede3e52fae/yarl-1.22.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:31f0b53913220599446872d757257be5898019c85e7971599065bc55065dc99d", size = 383803, upload-time = "2025-10-06T14:10:04.552Z" },
+    { url = "https://files.pythonhosted.org/packages/59/65/afc6e62bb506a319ea67b694551dab4a7e6fb7bf604e9bd9f3e11d575fec/yarl-1.22.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:a49370e8f711daec68d09b821a34e1167792ee2d24d405cbc2387be4f158b520", size = 374220, upload-time = "2025-10-06T14:10:06.489Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/3d/68bf18d50dc674b942daec86a9ba922d3113d8399b0e52b9897530442da2/yarl-1.22.0-cp312-cp312-win32.whl", hash = "sha256:70dfd4f241c04bd9239d53b17f11e6ab672b9f1420364af63e8531198e3f5fe8", size = 81589, upload-time = "2025-10-06T14:10:09.254Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/9a/6ad1a9b37c2f72874f93e691b2e7ecb6137fb2b899983125db4204e47575/yarl-1.22.0-cp312-cp312-win_amd64.whl", hash = "sha256:8884d8b332a5e9b88e23f60bb166890009429391864c685e17bd73a9eda9105c", size = 87213, upload-time = "2025-10-06T14:10:11.369Z" },
+    { url = "https://files.pythonhosted.org/packages/44/c5/c21b562d1680a77634d748e30c653c3ca918beb35555cff24986fff54598/yarl-1.22.0-cp312-cp312-win_arm64.whl", hash = "sha256:ea70f61a47f3cc93bdf8b2f368ed359ef02a01ca6393916bc8ff877427181e74", size = 81330, upload-time = "2025-10-06T14:10:13.112Z" },
+    { url = "https://files.pythonhosted.org/packages/94/fd/6480106702a79bcceda5fd9c63cb19a04a6506bd5ce7fd8d9b63742f0021/yarl-1.22.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:3aa27acb6de7a23785d81557577491f6c38a5209a254d1191519d07d8fe51748", size = 141301, upload-time = "2025-10-06T14:12:19.01Z" },
+    { url = "https://files.pythonhosted.org/packages/42/e1/6d95d21b17a93e793e4ec420a925fe1f6a9342338ca7a563ed21129c0990/yarl-1.22.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:af74f05666a5e531289cb1cc9c883d1de2088b8e5b4de48004e5ca8a830ac859", size = 93864, upload-time = "2025-10-06T14:12:21.05Z" },
+    { url = "https://files.pythonhosted.org/packages/32/58/b8055273c203968e89808413ea4c984988b6649baabf10f4522e67c22d2f/yarl-1.22.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:62441e55958977b8167b2709c164c91a6363e25da322d87ae6dd9c6019ceecf9", size = 94706, upload-time = "2025-10-06T14:12:23.287Z" },
+    { url = "https://files.pythonhosted.org/packages/18/91/d7bfbc28a88c2895ecd0da6a874def0c147de78afc52c773c28e1aa233a3/yarl-1.22.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b580e71cac3f8113d3135888770903eaf2f507e9421e5697d6ee6d8cd1c7f054", size = 347100, upload-time = "2025-10-06T14:12:28.527Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/e8/37a1e7b99721c0564b1fc7b0a4d1f595ef6fb8060d82ca61775b644185f7/yarl-1.22.0-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:e81fda2fb4a07eda1a2252b216aa0df23ebcd4d584894e9612e80999a78fd95b", size = 318902, upload-time = "2025-10-06T14:12:30.528Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/ef/34724449d7ef2db4f22df644f2dac0b8a275d20f585e526937b3ae47b02d/yarl-1.22.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:99b6fc1d55782461b78221e95fc357b47ad98b041e8e20f47c1411d0aacddc60", size = 363302, upload-time = "2025-10-06T14:12:32.295Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/04/88a39a5dad39889f192cce8d66cc4c58dbeca983e83f9b6bf23822a7ed91/yarl-1.22.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:088e4e08f033db4be2ccd1f34cf29fe994772fb54cfe004bbf54db320af56890", size = 370816, upload-time = "2025-10-06T14:12:34.01Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/1f/5e895e547129413f56c76be2c3ce4b96c797d2d0ff3e16a817d9269b12e6/yarl-1.22.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2e4e1f6f0b4da23e61188676e3ed027ef0baa833a2e633c29ff8530800edccba", size = 346465, upload-time = "2025-10-06T14:12:35.977Z" },
+    { url = "https://files.pythonhosted.org/packages/11/13/a750e9fd6f9cc9ed3a52a70fe58ffe505322f0efe0d48e1fd9ffe53281f5/yarl-1.22.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:84fc3ec96fce86ce5aa305eb4aa9358279d1aa644b71fab7b8ed33fe3ba1a7ca", size = 341506, upload-time = "2025-10-06T14:12:37.788Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/67/bb6024de76e7186611ebe626aec5b71a2d2ecf9453e795f2dbd80614784c/yarl-1.22.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:5dbeefd6ca588b33576a01b0ad58aa934bc1b41ef89dee505bf2932b22ddffba", size = 335030, upload-time = "2025-10-06T14:12:39.775Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/be/50b38447fd94a7992996a62b8b463d0579323fcfc08c61bdba949eef8a5d/yarl-1.22.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:14291620375b1060613f4aab9ebf21850058b6b1b438f386cc814813d901c60b", size = 358560, upload-time = "2025-10-06T14:12:41.547Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/89/c020b6f547578c4e3dbb6335bf918f26e2f34ad0d1e515d72fd33ac0c635/yarl-1.22.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:a4fcfc8eb2c34148c118dfa02e6427ca278bfd0f3df7c5f99e33d2c0e81eae3e", size = 357290, upload-time = "2025-10-06T14:12:43.861Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/52/c49a619ee35a402fa3a7019a4fa8d26878fec0d1243f6968bbf516789578/yarl-1.22.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:029866bde8d7b0878b9c160e72305bbf0a7342bcd20b9999381704ae03308dc8", size = 350700, upload-time = "2025-10-06T14:12:46.868Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/c9/f5042d87777bf6968435f04a2bbb15466b2f142e6e47fa4f34d1a3f32f0c/yarl-1.22.0-cp39-cp39-win32.whl", hash = "sha256:4dcc74149ccc8bba31ce1944acee24813e93cfdee2acda3c172df844948ddf7b", size = 82323, upload-time = "2025-10-06T14:12:48.633Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/58/d00f7cad9eba20c4eefac2682f34661d1d1b3a942fc0092eb60e78cfb733/yarl-1.22.0-cp39-cp39-win_amd64.whl", hash = "sha256:10619d9fdee46d20edc49d3479e2f8269d0779f1b031e6f7c2aa1c76be04b7ed", size = 87145, upload-time = "2025-10-06T14:12:50.241Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/a3/70904f365080780d38b919edd42d224b8c4ce224a86950d2eaa2a24366ad/yarl-1.22.0-cp39-cp39-win_arm64.whl", hash = "sha256:dd7afd3f8b0bfb4e0d9fc3c31bfe8a4ec7debe124cfd90619305def3c8ca8cd2", size = 82173, upload-time = "2025-10-06T14:12:51.869Z" },
+    { url = "https://files.pythonhosted.org/packages/73/ae/b48f95715333080afb75a4504487cbe142cae1268afc482d06692d605ae6/yarl-1.22.0-py3-none-any.whl", hash = "sha256:1380560bdba02b6b6c90de54133c81c9f2a453dee9912fe58c1dcced1edb7cff", size = 46814, upload-time = "2025-10-06T14:12:53.872Z" },
 ]
 
 [[package]]

From 87234d202924acce84ead694f8d06748608fd5f9 Mon Sep 17 00:00:00 2001
From: Robin Picard <robin.picard@sciencespo.fr>
Date: Sat, 25 Oct 2025 08:50:56 +0800
Subject: [PATCH 482/505] Fix error in the batch method of the MLXLM model

---
 docs/features/models/mlxlm.md |  2 +-
 outlines/models/mlxlm.py      | 33 +++++++++++++++++++++++++++++----
 tests/models/test_mlxlm.py    | 12 ++++++++++++
 3 files changed, 42 insertions(+), 5 deletions(-)

diff --git a/docs/features/models/mlxlm.md b/docs/features/models/mlxlm.md
index 546bb6a37..90b5951b0 100644
--- a/docs/features/models/mlxlm.md
+++ b/docs/features/models/mlxlm.md
@@ -100,7 +100,7 @@ for chunk in model.stream("Write a short story about a cat.", max_tokens=100):
 
 #### Batch Generation
 
-The `MLXLM` model supports generating text in batches. To do so, use the `batch` method and provide a list of strings as a model input. For instance:
+The `MLXLM` model supports generating text in batches. To do so, use the `batch` method and provide a list of strings as a model input. However, constrained generation is not supported with batching, so you cannot provide an `output_type`. For instance:
 
 ```python
 import outlines
diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 0a91e4bd1..5111eccaa 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -173,16 +173,41 @@ def generate_batch(
             The list of text generated by the model.
 
         """
-        from mlx_lm import generate_batch
+        from mlx_lm import batch_generate
 
-        return generate_batch(
+        if output_type:
+            raise NotImplementedError(
+                "mlx-lm does not support constrained generation with batching."
+                + "You cannot provide an `output_type` with this method."
+            )
+
+        model_input = [self.type_adapter.format_input(item) for item in model_input]
+
+        # Contrarily to the other generate methods, batch_generate requires
+        # tokenized prompts
+        add_special_tokens = [
+            (
+                self.mlx_tokenizer.bos_token is None
+                or not prompt.startswith(self.mlx_tokenizer.bos_token)
+            )
+            for prompt in model_input
+        ]
+        tokenized_model_input = [
+            self.mlx_tokenizer.encode(
+                model_input[i], add_special_tokens=add_special_tokens[i]
+            )
+            for i in range(len(model_input))
+        ]
+
+        response = batch_generate(
             self.model,
             self.mlx_tokenizer,
-            [self.type_adapter.format_input(item) for item in model_input],
-            logits_processors=self.type_adapter.format_output_type(output_type),
+            tokenized_model_input,
             **kwargs,
         )
 
+        return response.texts
+
     def generate_stream(
         self,
         model_input: str,
diff --git a/tests/models/test_mlxlm.py b/tests/models/test_mlxlm.py
index 924a54551..b1b0217f4 100644
--- a/tests/models/test_mlxlm.py
+++ b/tests/models/test_mlxlm.py
@@ -128,3 +128,15 @@ def test_mlxlm_batch(model):
     assert len(result) == 2
     assert isinstance(result[0], str)
     assert isinstance(result[1], str)
+
+
+@pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
+def test_mlxlm_batch_output_type(model):
+    with pytest.raises(
+        NotImplementedError,
+        match="mlx-lm does not support constrained generation with batching."
+    ):
+        model.batch(
+            ["Respond with one word.", "Respond with one word."],
+            Regex(r"[0-9]")
+        )

From 0f9f05f17eb33aedf8d73fd0bc494494ea756bb4 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 20 Nov 2025 17:07:37 +0800
Subject: [PATCH 483/505] fix: Refactor sampling parameters in VLLMOffline to
 use StructuredOutputsParams

---
 outlines/models/vllm_offline.py | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index 5e38e72f4..9d1ea08f3 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -118,7 +118,7 @@ def _build_generation_args(
     ) -> "SamplingParams":
         """Create the `SamplingParams` object to pass to the `generate` method
         of the `vllm.LLM` model."""
-        from vllm.sampling_params import GuidedDecodingParams, SamplingParams
+        from vllm.sampling_params import StructuredOutputsParams, SamplingParams
 
         sampling_params = inference_kwargs.pop("sampling_params", None)
 
@@ -127,7 +127,9 @@ def _build_generation_args(
 
         output_type_args = self.type_adapter.format_output_type(output_type)
         if output_type_args:
-            sampling_params.guided_decoding = GuidedDecodingParams(**output_type_args)
+            original_sampling_params_dict = {f: getattr(sampling_params, f) for f in sampling_params.__struct_fields__}
+            sampling_params_dict = {**original_sampling_params_dict, "structured_outputs": StructuredOutputsParams(**output_type_args)}
+            sampling_params = SamplingParams(**sampling_params_dict)
 
         return sampling_params
 

From 9f909bdec831dfd31b429f6db812d0cda32f0366 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 20 Nov 2025 10:31:11 +0800
Subject: [PATCH 484/505] Add SamplingParams to model response example

Updated example to include SamplingParams in response call.
---
 docs/features/models/vllm_offline.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
index 91b454d68..693305fcb 100644
--- a/docs/features/models/vllm_offline.md
+++ b/docs/features/models/vllm_offline.md
@@ -44,7 +44,7 @@ For instance:
 
 ```python
 import outlines
-from vllm import LLM
+from vllm import LLM, SamplingParams
 
 # Create the model
 model = outlines.from_vllm_offline(
@@ -52,7 +52,7 @@ model = outlines.from_vllm_offline(
 )
 
 # Call it to generate text
-response = model("What's the capital of Latvia?", max_tokens=20)
+response = model("What's the capital of Latvia?", sampling_params=SamplingParams(max_tokens=20))
 print(response) # 'Riga'
 ```
 

From e12c99518335725f656eeace7400930cbb4a30be Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 20 Nov 2025 11:02:47 +0800
Subject: [PATCH 485/505] fix: Add SamplingParams to model response examples

Updated usage examples to include SamplingParams for response generation and streaming.
---
 docs/features/models/vllm_offline.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/features/models/vllm_offline.md b/docs/features/models/vllm_offline.md
index 693305fcb..5a2a8d205 100644
--- a/docs/features/models/vllm_offline.md
+++ b/docs/features/models/vllm_offline.md
@@ -64,7 +64,7 @@ For instance:
 
 ```python
 import outlines
-from vllm import LLM
+from vllm import LLM, SamplingParams
 from outlines.inputs import Chat
 
 # Create the model
@@ -79,7 +79,7 @@ prompt = Chat([
 ])
 
 # Call the model to generate a response
-response = model(prompt, max_tokens=50)
+response = model(prompt, sampling_params=SamplingParams(max_tokens=50))
 print(response) # 'Riga'
 ```
 
@@ -91,7 +91,7 @@ For instance:
 
 ```python
 import outlines
-from vllm import LLM
+from vllm import LLM, SamplingParams
 
 # Create the model
 model = outlines.from_vllm_offline(
@@ -99,7 +99,7 @@ model = outlines.from_vllm_offline(
 )
 
 # Stream the response
-for chunk in model.stream("Tell me a short story about a cat.", max_tokens=50):
+for chunk in model.stream("Tell me a short story about a cat.", sampling_params=SamplingParams(max_tokens=50)):
     print(chunk) # 'Once...'
 ```
 

From c9dd2963ab1bba6cb4cc2fc0a3cfa60d7e4cde71 Mon Sep 17 00:00:00 2001
From: RobinPicard <robin.picard@sciencespo.fr>
Date: Mon, 24 Nov 2025 12:03:33 +0100
Subject: [PATCH 486/505] Draft

---
 uv.lock | 4022 +++++++++++++++++++++++++++----------------------------
 1 file changed, 1974 insertions(+), 2048 deletions(-)

diff --git a/uv.lock b/uv.lock
index b69731165..5bb950a75 100644
--- a/uv.lock
+++ b/uv.lock
@@ -1,5 +1,5 @@
 version = 1
-revision = 2
+revision = 3
 requires-python = ">=3.9, <3.13"
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -21,28 +21,53 @@ wheels = [
 name = "accelerate"
 version = "1.10.1"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "huggingface-hub" },
+    { name = "huggingface-hub", marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "packaging" },
-    { name = "psutil" },
-    { name = "pyyaml" },
-    { name = "safetensors" },
-    { name = "torch" },
+    { name = "packaging", marker = "python_full_version < '3.10'" },
+    { name = "psutil", marker = "python_full_version < '3.10'" },
+    { name = "pyyaml", marker = "python_full_version < '3.10'" },
+    { name = "safetensors", marker = "python_full_version < '3.10'" },
+    { name = "torch", version = "2.8.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/b1/72/ff3961c19ee395c3d30ac630ee77bfb0e1b46b87edc504d4f83bb4a89705/accelerate-1.10.1.tar.gz", hash = "sha256:3dea89e433420e4bfac0369cae7e36dcd6a56adfcfd38cdda145c6225eab5df8", size = 392446, upload-time = "2025-08-25T13:57:06.21Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/5f/a0/d9ef19f780f319c21ee90ecfef4431cbeeca95bec7f14071785c17b6029b/accelerate-1.10.1-py3-none-any.whl", hash = "sha256:3621cff60b9a27ce798857ece05e2b9f56fcc71631cfb31ccf71f0359c311f11", size = 374909, upload-time = "2025-08-25T13:57:04.55Z" },
 ]
 
+[[package]]
+name = "accelerate"
+version = "1.12.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "huggingface-hub", marker = "python_full_version >= '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "packaging", marker = "python_full_version >= '3.10'" },
+    { name = "psutil", marker = "python_full_version >= '3.10'" },
+    { name = "pyyaml", marker = "python_full_version >= '3.10'" },
+    { name = "safetensors", marker = "python_full_version >= '3.10'" },
+    { name = "torch", version = "2.9.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4a/8e/ac2a9566747a93f8be36ee08532eb0160558b07630a081a6056a9f89bf1d/accelerate-1.12.0.tar.gz", hash = "sha256:70988c352feb481887077d2ab845125024b2a137a5090d6d7a32b57d03a45df6", size = 398399, upload-time = "2025-11-21T11:27:46.973Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9f/d2/c581486aa6c4fbd7394c23c47b83fa1a919d34194e16944241daf9e762dd/accelerate-1.12.0-py3-none-any.whl", hash = "sha256:3e2091cd341423207e2f084a6654b1efcd250dc326f2a37d6dde446e07cabb11", size = 380935, upload-time = "2025-11-21T11:27:44.522Z" },
+]
+
 [[package]]
 name = "aiofiles"
-version = "24.1.0"
+version = "25.1.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/0b/03/a88171e277e8caa88a4c77808c20ebb04ba74cc4681bf1e9416c862de237/aiofiles-24.1.0.tar.gz", hash = "sha256:22a075c9e5a3810f0c2e48f3008c94d68c65d763b9b03857924c99e57355166c", size = 30247, upload-time = "2024-06-24T11:02:03.584Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/41/c3/534eac40372d8ee36ef40df62ec129bee4fdb5ad9706e58a29be53b2c970/aiofiles-25.1.0.tar.gz", hash = "sha256:a8d728f0a29de45dc521f18f07297428d56992a742f0cd2701ba86e44d23d5b2", size = 46354, upload-time = "2025-10-09T20:51:04.358Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a5/45/30bb92d442636f570cb5651bc661f52b610e2eec3f891a5dc3a4c3667db0/aiofiles-24.1.0-py3-none-any.whl", hash = "sha256:b4ec55f4195e3eb5d7abd1bf7e061763e864dd4954231fb8539a0ef8bb8260e5", size = 15896, upload-time = "2024-06-24T11:02:01.529Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/8a/340a1555ae33d7354dbca4faa54948d76d89a27ceef032c8c3bc661d003e/aiofiles-25.1.0-py3-none-any.whl", hash = "sha256:abe311e527c862958650f9438e859c1fa7568a141b22abcd015e120e86a85695", size = 14668, upload-time = "2025-10-09T20:51:03.174Z" },
 ]
 
 [[package]]
@@ -56,7 +81,7 @@ wheels = [
 
 [[package]]
 name = "aiohttp"
-version = "3.13.0"
+version = "3.13.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "aiohappyeyeballs" },
@@ -68,76 +93,76 @@ dependencies = [
     { name = "propcache" },
     { name = "yarl" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/62/f1/8515650ac3121a9e55c7b217c60e7fae3e0134b5acfe65691781b5356929/aiohttp-3.13.0.tar.gz", hash = "sha256:378dbc57dd8cf341ce243f13fa1fa5394d68e2e02c15cd5f28eae35a70ec7f67", size = 7832348, upload-time = "2025-10-06T19:58:48.089Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/25/18/a3a9c9b7c8d400f71d1ff93c3e1520a5d53dba170f829ca9c6b2b070677b/aiohttp-3.13.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:ca69ec38adf5cadcc21d0b25e2144f6a25b7db7bea7e730bac25075bc305eff0", size = 734428, upload-time = "2025-10-06T19:54:40.285Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/02/f1eac06d78997e015030130ccf1c7cf864a919f97d77ff27e89c82fc3186/aiohttp-3.13.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:240f99f88a9a6beb53ebadac79a2e3417247aa756202ed234b1dbae13d248092", size = 491939, upload-time = "2025-10-06T19:54:42.113Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/db/5d65af7cbe5f302e23b1ea5cfc156cd0c7738a0d2db531a3837d2754de94/aiohttp-3.13.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a4676b978a9711531e7cea499d4cdc0794c617a1c0579310ab46c9fdf5877702", size = 487229, upload-time = "2025-10-06T19:54:43.978Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/d5/56c622ad3bd57ff4adc2b701f298dcc0408735a8af998cec1c66a9ce224e/aiohttp-3.13.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:48fcdd5bc771cbbab8ccc9588b8b6447f6a30f9fe00898b1a5107098e00d6793", size = 1666118, upload-time = "2025-10-06T19:54:46.569Z" },
-    { url = "https://files.pythonhosted.org/packages/44/16/db236671ec3758e3a6be6977009e74016470368012a58fea4b3799546549/aiohttp-3.13.0-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:eeea0cdd2f687e210c8f605f322d7b0300ba55145014a5dbe98bd4be6fff1f6c", size = 1633983, upload-time = "2025-10-06T19:54:48.244Z" },
-    { url = "https://files.pythonhosted.org/packages/19/ad/d96d7d7023e7f5215b8737cad21a7637f6d9d10fbfbfef0435d0277f71a2/aiohttp-3.13.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:10b3f01d5aeb632adaaf39c5e93f040a550464a768d54c514050c635adcbb9d0", size = 1725922, upload-time = "2025-10-06T19:54:49.885Z" },
-    { url = "https://files.pythonhosted.org/packages/88/d7/e8a5ba2bbd929ed587b2a8ea9390765daede2d8cd28dfae3a0773c6d3fbc/aiohttp-3.13.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a4dc0b83e25267f42ef065ea57653de4365b56d7bc4e4cfc94fabe56998f8ee6", size = 1813770, upload-time = "2025-10-06T19:54:51.648Z" },
-    { url = "https://files.pythonhosted.org/packages/f9/ca/135c21e85ffeff66b80ecd8a647ca104f2e5a91c37dc86649244ddbf87ab/aiohttp-3.13.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:72714919ed9b90f030f761c20670e529c4af96c31bd000917dd0c9afd1afb731", size = 1667322, upload-time = "2025-10-06T19:54:53.668Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/38/348c4343052a400968dbf2051ee3dc222bdefd95af5874cf0f04cc7a8c92/aiohttp-3.13.0-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:564be41e85318403fdb176e9e5b3e852d528392f42f2c1d1efcbeeed481126d7", size = 1553270, upload-time = "2025-10-06T19:54:56.054Z" },
-    { url = "https://files.pythonhosted.org/packages/47/89/71cbda30f0900ab16084769960c467a355d6b1db51668fbb821c4a4ad5ed/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:84912962071087286333f70569362e10793f73f45c48854e6859df11001eb2d3", size = 1637087, upload-time = "2025-10-06T19:54:58.548Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/b1/5ff5fcaecccdcd5be7ff717cbde6e630760a8130e89167c3aa05b6b57707/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:90b570f1a146181c3d6ae8f755de66227ded49d30d050479b5ae07710f7894c5", size = 1643443, upload-time = "2025-10-06T19:55:00.856Z" },
-    { url = "https://files.pythonhosted.org/packages/87/e2/1d1f202f43c8be1956f05196159064cc05dc6842a33c1397cbb1b99610af/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:2d71ca30257ce756e37a6078b1dff2d9475fee13609ad831eac9a6531bea903b", size = 1695571, upload-time = "2025-10-06T19:55:03.006Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/b9/53c1df2991686f947a9651265757ea12c4afc29b351a249b73a0fc81dd3c/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:cd45eb70eca63f41bb156b7dffbe1a7760153b69892d923bdb79a74099e2ed90", size = 1539975, upload-time = "2025-10-06T19:55:04.839Z" },
-    { url = "https://files.pythonhosted.org/packages/93/24/345166f9c4cd2f5cc1d2173131998ee4adab0db8729126db32a7f91ed400/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:5ae3a19949a27982c7425a7a5a963c1268fdbabf0be15ab59448cbcf0f992519", size = 1712866, upload-time = "2025-10-06T19:55:06.905Z" },
-    { url = "https://files.pythonhosted.org/packages/09/f1/e8f70462848b74d49b3115050623ecbd697889713c2c93c96616da56b2de/aiohttp-3.13.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:ea6df292013c9f050cbf3f93eee9953d6e5acd9e64a0bf4ca16404bfd7aa9bcc", size = 1654058, upload-time = "2025-10-06T19:55:08.51Z" },
-    { url = "https://files.pythonhosted.org/packages/23/ba/47fd065510a8bfab5d5f6e1d97c0de672447c0a941c5021298bd7210afc3/aiohttp-3.13.0-cp310-cp310-win32.whl", hash = "sha256:3b64f22fbb6dcd5663de5ef2d847a5638646ef99112503e6f7704bdecb0d1c4d", size = 430230, upload-time = "2025-10-06T19:55:10.178Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/38/f5385cb79afa1f31bcaa3625a9e8d849b782edaeac09f894f46439e006a1/aiohttp-3.13.0-cp310-cp310-win_amd64.whl", hash = "sha256:f8d877aa60d80715b2afc565f0f1aea66565824c229a2d065b31670e09fed6d7", size = 453013, upload-time = "2025-10-06T19:55:11.623Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/db/df80cacac46cd548a736c5535b13cc18925cf6f9f83cd128cf3839842219/aiohttp-3.13.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:99eb94e97a42367fef5fc11e28cb2362809d3e70837f6e60557816c7106e2e20", size = 741374, upload-time = "2025-10-06T19:55:13.095Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/f9/2d6d93fd57ab4726e18a7cdab083772eda8302d682620fbf2aef48322351/aiohttp-3.13.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:4696665b2713021c6eba3e2b882a86013763b442577fe5d2056a42111e732eca", size = 494956, upload-time = "2025-10-06T19:55:14.687Z" },
-    { url = "https://files.pythonhosted.org/packages/89/a6/e1c061b079fed04ffd6777950c82f2e8246fd08b7b3c4f56fdd47f697e5a/aiohttp-3.13.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3e6a38366f7f0d0f6ed7a1198055150c52fda552b107dad4785c0852ad7685d1", size = 491154, upload-time = "2025-10-06T19:55:16.661Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/4d/ee8913c0d2c7da37fdc98673a342b51611eaa0871682b37b8430084e35b5/aiohttp-3.13.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:aab715b1a0c37f7f11f9f1f579c6fbaa51ef569e47e3c0a4644fba46077a9409", size = 1745707, upload-time = "2025-10-06T19:55:18.376Z" },
-    { url = "https://files.pythonhosted.org/packages/f9/70/26b2c97e8fa68644aec43d788940984c5f3b53a8d1468d5baaa328f809c9/aiohttp-3.13.0-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:7972c82bed87d7bd8e374b60a6b6e816d75ba4f7c2627c2d14eed216e62738e1", size = 1702404, upload-time = "2025-10-06T19:55:20.098Z" },
-    { url = "https://files.pythonhosted.org/packages/65/1e/c8aa3c293a0e8b18968b1b88e9bd8fb269eb67eb7449f504a4c3e175b159/aiohttp-3.13.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ca8313cb852af788c78d5afdea24c40172cbfff8b35e58b407467732fde20390", size = 1805519, upload-time = "2025-10-06T19:55:21.811Z" },
-    { url = "https://files.pythonhosted.org/packages/51/b6/a3753fe86249eb441768658cfc00f8c4e0913b255c13be00ddb8192775e1/aiohttp-3.13.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:6c333a2385d2a6298265f4b3e960590f787311b87f6b5e6e21bb8375914ef504", size = 1893904, upload-time = "2025-10-06T19:55:23.462Z" },
-    { url = "https://files.pythonhosted.org/packages/51/6d/7b1e020fe1d2a2be7cf0ce5e35922f345e3507cf337faa1a6563c42065c1/aiohttp-3.13.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:cc6d5fc5edbfb8041d9607f6a417997fa4d02de78284d386bea7ab767b5ea4f3", size = 1745043, upload-time = "2025-10-06T19:55:25.208Z" },
-    { url = "https://files.pythonhosted.org/packages/e6/df/aad5dce268f9d4f29759c3eeb5fb5995c569d76abb267468dc1075218d5b/aiohttp-3.13.0-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:7ddedba3d0043349edc79df3dc2da49c72b06d59a45a42c1c8d987e6b8d175b8", size = 1604765, upload-time = "2025-10-06T19:55:27.157Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/19/a84a0e97b2da2224c8b85e1aef5cac834d07b2903c17bff1a6bdbc7041d2/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:23ca762140159417a6bbc959ca1927f6949711851e56f2181ddfe8d63512b5ad", size = 1721737, upload-time = "2025-10-06T19:55:28.854Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/61/ca6ad390128d964a08554fd63d6df5810fb5fbc7e599cb9e617f1729ae19/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:bfe824d6707a5dc3c5676685f624bc0c63c40d79dc0239a7fd6c034b98c25ebe", size = 1716052, upload-time = "2025-10-06T19:55:30.563Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/71/769e249e6625372c7d14be79b8b8c3b0592963a09793fb3d36758e60952c/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:3c11fa5dd2ef773a8a5a6daa40243d83b450915992eab021789498dc87acc114", size = 1783532, upload-time = "2025-10-06T19:55:32.798Z" },
-    { url = "https://files.pythonhosted.org/packages/66/64/b9cd03cdbb629bc492e4a744fbe96550a8340b0cd7a0cc4a9c90cfecd8d3/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:00fdfe370cffede3163ba9d3f190b32c0cfc8c774f6f67395683d7b0e48cdb8a", size = 1593072, upload-time = "2025-10-06T19:55:34.686Z" },
-    { url = "https://files.pythonhosted.org/packages/24/0e/87922c8cfdbd09f5e2197e9d87714a98c99c423560d44739e3af55400fe3/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:6475e42ef92717a678bfbf50885a682bb360a6f9c8819fb1a388d98198fdcb80", size = 1798613, upload-time = "2025-10-06T19:55:36.393Z" },
-    { url = "https://files.pythonhosted.org/packages/c5/bb/a3adfe2af76e1ee9e3b5464522004b148b266bc99d7ec424ca7843d64a3c/aiohttp-3.13.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:77da5305a410910218b99f2a963092f4277d8a9c1f429c1ff1b026d1826bd0b6", size = 1737480, upload-time = "2025-10-06T19:55:38.043Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/53/e124dcbd64e6365602f3493fe37a11ca5b7ac0a40822a6e2bc8260cd08e0/aiohttp-3.13.0-cp311-cp311-win32.whl", hash = "sha256:2f9d9ea547618d907f2ee6670c9a951f059c5994e4b6de8dcf7d9747b420c820", size = 429824, upload-time = "2025-10-06T19:55:39.595Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/bd/485d98b372a2cd6998484a93ddd401ec6b6031657661c36846a10e2a1f6e/aiohttp-3.13.0-cp311-cp311-win_amd64.whl", hash = "sha256:0f19f7798996d4458c669bd770504f710014926e9970f4729cf55853ae200469", size = 454137, upload-time = "2025-10-06T19:55:41.617Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/95/7e8bdfa6e79099a086d59d42589492f1fe9d29aae3cefb58b676015ce278/aiohttp-3.13.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:1c272a9a18a5ecc48a7101882230046b83023bb2a662050ecb9bfcb28d9ab53a", size = 735585, upload-time = "2025-10-06T19:55:43.401Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/20/2f1d3ee06ee94eafe516810705219bff234d09f135d6951661661d5595ae/aiohttp-3.13.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:97891a23d7fd4e1afe9c2f4473e04595e4acb18e4733b910b6577b74e7e21985", size = 490613, upload-time = "2025-10-06T19:55:45.237Z" },
-    { url = "https://files.pythonhosted.org/packages/74/15/ab8600ef6dc1dcd599009a81acfed2ea407037e654d32e47e344e0b08c34/aiohttp-3.13.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:475bd56492ce5f4cffe32b5533c6533ee0c406d1d0e6924879f83adcf51da0ae", size = 489750, upload-time = "2025-10-06T19:55:46.937Z" },
-    { url = "https://files.pythonhosted.org/packages/33/59/752640c2b86ca987fe5703a01733b00d375e6cd2392bc7574489934e64e5/aiohttp-3.13.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c32ada0abb4bc94c30be2b681c42f058ab104d048da6f0148280a51ce98add8c", size = 1736812, upload-time = "2025-10-06T19:55:48.917Z" },
-    { url = "https://files.pythonhosted.org/packages/3d/c6/dd6b86ddb852a7fdbcdc7a45b6bdc80178aef713c08279afcaee7a5a9f07/aiohttp-3.13.0-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:4af1f8877ca46ecdd0bc0d4a6b66d4b2bddc84a79e2e8366bc0d5308e76bceb8", size = 1698535, upload-time = "2025-10-06T19:55:50.75Z" },
-    { url = "https://files.pythonhosted.org/packages/33/e2/27c92d205b9e8cee7661670e8e9f187931b71e26d42796b153d2a0ba6949/aiohttp-3.13.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e04ab827ec4f775817736b20cdc8350f40327f9b598dec4e18c9ffdcbea88a93", size = 1766573, upload-time = "2025-10-06T19:55:53.106Z" },
-    { url = "https://files.pythonhosted.org/packages/df/6a/1fc1ad71d130a30f7a207d8d958a41224c29b834463b5185efb2dbff6ad4/aiohttp-3.13.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a6d9487b9471ec36b0faedf52228cd732e89be0a2bbd649af890b5e2ce422353", size = 1865229, upload-time = "2025-10-06T19:55:55.01Z" },
-    { url = "https://files.pythonhosted.org/packages/14/51/d0c1701a79fcb0109cff5304da16226581569b89a282d8e7f1549a7e3ec0/aiohttp-3.13.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2e66c57416352f36bf98f6641ddadd47c93740a22af7150d3e9a1ef6e983f9a8", size = 1750379, upload-time = "2025-10-06T19:55:57.219Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/3d/2ec4b934f85856de1c0c18e90adc8902adadbfac2b3c0b831bfeb7214fc8/aiohttp-3.13.0-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:469167d5372f5bb3aedff4fc53035d593884fff2617a75317740e885acd48b04", size = 1560798, upload-time = "2025-10-06T19:55:58.888Z" },
-    { url = "https://files.pythonhosted.org/packages/38/56/e23d9c3e13006e599fdce3851517c70279e177871e3e567d22cf3baf5d6c/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a9f3546b503975a69b547c9fd1582cad10ede1ce6f3e313a2f547c73a3d7814f", size = 1697552, upload-time = "2025-10-06T19:56:01.172Z" },
-    { url = "https://files.pythonhosted.org/packages/56/cb/caa32c2ccaeca0a3dc39129079fd2ad02f9406c3a5f7924340435b87d4cd/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:6b4174fcec98601f0cfdf308ee29a6ae53c55f14359e848dab4e94009112ee7d", size = 1718609, upload-time = "2025-10-06T19:56:03.102Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/c0/5911856fef9e40fd1ccbb8c54a90116875d5753a92c1cac66ce2059b390d/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:a533873a7a4ec2270fb362ee5a0d3b98752e4e1dc9042b257cd54545a96bd8ed", size = 1735887, upload-time = "2025-10-06T19:56:04.841Z" },
-    { url = "https://files.pythonhosted.org/packages/0e/48/8d6f4757a24c02f0a454c043556593a00645d10583859f7156db44d8b7d3/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:ce887c5e54411d607ee0959cac15bb31d506d86a9bcaddf0b7e9d63325a7a802", size = 1553079, upload-time = "2025-10-06T19:56:07.197Z" },
-    { url = "https://files.pythonhosted.org/packages/39/fa/e82c9445e40b50e46770702b5b6ca2f767966d53e1a5eef03583ceac6df6/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:d871f6a30d43e32fc9252dc7b9febe1a042b3ff3908aa83868d7cf7c9579a59b", size = 1762750, upload-time = "2025-10-06T19:56:09.376Z" },
-    { url = "https://files.pythonhosted.org/packages/3d/e6/9d30554e7f1e700bfeae4ab6b153d5dc7441606a9ec5e929288fa93a1477/aiohttp-3.13.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:222c828243b4789d79a706a876910f656fad4381661691220ba57b2ab4547865", size = 1717461, upload-time = "2025-10-06T19:56:11.551Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/e5/29cca547990a59ea54f0674fc01de98519fc628cfceeab6175711750eca7/aiohttp-3.13.0-cp312-cp312-win32.whl", hash = "sha256:682d2e434ff2f1108314ff7f056ce44e457f12dbed0249b24e106e385cf154b9", size = 424633, upload-time = "2025-10-06T19:56:13.316Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/68/46dd042d7bc62eab30bafdb8569f55ef125c3a88bb174270324224f8df56/aiohttp-3.13.0-cp312-cp312-win_amd64.whl", hash = "sha256:0a2be20eb23888df130214b91c262a90e2de1553d6fb7de9e9010cec994c0ff2", size = 451401, upload-time = "2025-10-06T19:56:15.188Z" },
-    { url = "https://files.pythonhosted.org/packages/40/26/6625e909ea770910d9901cb04336d0280a5a578daa352434eba5e482a1eb/aiohttp-3.13.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:fcc425fb6fd2a00c6d91c85d084c6b75a61bc8bc12159d08e17c5711df6c5ba4", size = 736970, upload-time = "2025-10-06T19:58:07.563Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/bc/ce07cc194f416f2d2539fec1237507b97496752f350738a051258be3dc08/aiohttp-3.13.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7c2c4c9ce834801651f81d6760d0a51035b8b239f58f298de25162fcf6f8bb64", size = 493252, upload-time = "2025-10-06T19:58:10.146Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/70/f51f30694aa4f9de3fad5b254609c263bf873a36dfa36f0370b37acb6b18/aiohttp-3.13.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:f91e8f9053a07177868e813656ec57599cd2a63238844393cd01bd69c2e40147", size = 488585, upload-time = "2025-10-06T19:58:12.438Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/89/4dd10ab758ba0dee025fc53345d24f5943f4d8741c1064ed3e92ca3274a8/aiohttp-3.13.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:df46d9a3d78ec19b495b1107bf26e4fcf97c900279901f4f4819ac5bb2a02a4c", size = 1660915, upload-time = "2025-10-06T19:58:14.359Z" },
-    { url = "https://files.pythonhosted.org/packages/49/72/f3a894182b625c7f242ab90ff95d6fbf4319d39dc854568a33c8768220ec/aiohttp-3.13.0-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3b1eb9871cbe43b6ca6fac3544682971539d8a1d229e6babe43446279679609d", size = 1625065, upload-time = "2025-10-06T19:58:16.754Z" },
-    { url = "https://files.pythonhosted.org/packages/a5/3d/31f310c67b33e30d1f2fe11bf5aeb82d178189e8abfcf696bed69a9713cb/aiohttp-3.13.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:62a3cddf8d9a2eae1f79585fa81d32e13d0c509bb9e7ad47d33c83b45a944df7", size = 1721983, upload-time = "2025-10-06T19:58:19.275Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/6a/64f23ffd040409ac148a389be5c377f8edc4ab2aebdd601dbe786a30e05d/aiohttp-3.13.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0f735e680c323ee7e9ef8e2ea26425c7dbc2ede0086fa83ce9d7ccab8a089f26", size = 1810847, upload-time = "2025-10-06T19:58:21.419Z" },
-    { url = "https://files.pythonhosted.org/packages/51/64/9951b85fdf53aea1e9f41c715e9f4fc3e60f430fb58e0a46bc7837688e39/aiohttp-3.13.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6a51839f778b0e283b43cd82bb17f1835ee2cc1bf1101765e90ae886e53e751c", size = 1658909, upload-time = "2025-10-06T19:58:23.406Z" },
-    { url = "https://files.pythonhosted.org/packages/ac/66/4d2c26a8a86b81a4493beb222457c6cba9f3853b6efa476a97edb4d52bb9/aiohttp-3.13.0-cp39-cp39-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:ac90cfab65bc281d6752f22db5fa90419e33220af4b4fa53b51f5948f414c0e7", size = 1551803, upload-time = "2025-10-06T19:58:25.568Z" },
-    { url = "https://files.pythonhosted.org/packages/a6/9a/c6b4eb325f473f3a117fab90008348fc7094ccbaeeedf811d89738454996/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:62fd54f3e6f17976962ba67f911d62723c760a69d54f5d7b74c3ceb1a4e9ef8d", size = 1632205, upload-time = "2025-10-06T19:58:27.976Z" },
-    { url = "https://files.pythonhosted.org/packages/55/27/6fa2a583d45f711cf2f719f0d69274739a94f3b6d3deddfd19106d2ef3b9/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:cf2b60b65df05b6b2fa0d887f2189991a0dbf44a0dd18359001dc8fcdb7f1163", size = 1637447, upload-time = "2025-10-06T19:58:30.143Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/2a/ca6f2ac71b6c7e6f93ad42616e0a75660d8b9400977053126b7eaeaa5132/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:1ccedfe280e804d9a9d7fe8b8c4309d28e364b77f40309c86596baa754af50b1", size = 1692038, upload-time = "2025-10-06T19:58:32.227Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/fa/221c09563a11ce0d22408bf41e4bd8385f2093a22ceb60eeddd1f5019bab/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:ea01ffbe23df53ece0c8732d1585b3d6079bb8c9ee14f3745daf000051415a31", size = 1538478, upload-time = "2025-10-06T19:58:34.776Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/09/06b8b45874ba26d7b012b9b21b6593d9a7a0a4e3375da83c74eb044caafa/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:19ba8625fa69523627b67f7e9901b587a4952470f68814d79cdc5bc460e9b885", size = 1708317, upload-time = "2025-10-06T19:58:38.746Z" },
-    { url = "https://files.pythonhosted.org/packages/63/b2/ee8c8bae8d2f17a146089bba8130cf18eb0e929d4c5262beae944da82b3a/aiohttp-3.13.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:4b14bfae90598d331b5061fd15a7c290ea0c15b34aeb1cf620464bb5ec02a602", size = 1647241, upload-time = "2025-10-06T19:58:41.233Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/e4/455240c0433ebc7c9400c6014d5e4d09509d0aeddb8c9d22aff1f3d40500/aiohttp-3.13.0-cp39-cp39-win32.whl", hash = "sha256:cf7a4b976da219e726d0043fc94ae8169c0dba1d3a059b3c1e2c964bafc5a77d", size = 430922, upload-time = "2025-10-06T19:58:43.699Z" },
-    { url = "https://files.pythonhosted.org/packages/e6/e6/f49ed1869b961f41aa0421d35401dd4ee4c8ad74b3da618f7bde1a8ddc29/aiohttp-3.13.0-cp39-cp39-win_amd64.whl", hash = "sha256:6b9697d15231aeaed4786f090c9c8bc3ab5f0e0a6da1e76c135a310def271020", size = 453958, upload-time = "2025-10-06T19:58:45.868Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/1c/ce/3b83ebba6b3207a7135e5fcaba49706f8a4b6008153b4e30540c982fae26/aiohttp-3.13.2.tar.gz", hash = "sha256:40176a52c186aefef6eb3cad2cdd30cd06e3afbe88fe8ab2af9c0b90f228daca", size = 7837994, upload-time = "2025-10-28T20:59:39.937Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6d/34/939730e66b716b76046dedfe0842995842fa906ccc4964bba414ff69e429/aiohttp-3.13.2-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:2372b15a5f62ed37789a6b383ff7344fc5b9f243999b0cd9b629d8bc5f5b4155", size = 736471, upload-time = "2025-10-28T20:55:27.924Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/cf/dcbdf2df7f6ca72b0bb4c0b4509701f2d8942cf54e29ca197389c214c07f/aiohttp-3.13.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:e7f8659a48995edee7229522984bd1009c1213929c769c2daa80b40fe49a180c", size = 493985, upload-time = "2025-10-28T20:55:29.456Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/87/71c8867e0a1d0882dcbc94af767784c3cb381c1c4db0943ab4aae4fed65e/aiohttp-3.13.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:939ced4a7add92296b0ad38892ce62b98c619288a081170695c6babe4f50e636", size = 489274, upload-time = "2025-10-28T20:55:31.134Z" },
+    { url = "https://files.pythonhosted.org/packages/38/0f/46c24e8dae237295eaadd113edd56dee96ef6462adf19b88592d44891dc5/aiohttp-3.13.2-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6315fb6977f1d0dd41a107c527fee2ed5ab0550b7d885bc15fee20ccb17891da", size = 1668171, upload-time = "2025-10-28T20:55:36.065Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/c6/4cdfb4440d0e28483681a48f69841fa5e39366347d66ef808cbdadddb20e/aiohttp-3.13.2-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:6e7352512f763f760baaed2637055c49134fd1d35b37c2dedfac35bfe5cf8725", size = 1636036, upload-time = "2025-10-28T20:55:37.576Z" },
+    { url = "https://files.pythonhosted.org/packages/84/37/8708cf678628216fb678ab327a4e1711c576d6673998f4f43e86e9ae90dd/aiohttp-3.13.2-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e09a0a06348a2dd73e7213353c90d709502d9786219f69b731f6caa0efeb46f5", size = 1727975, upload-time = "2025-10-28T20:55:39.457Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/2e/3ebfe12fdcb9b5f66e8a0a42dffcd7636844c8a018f261efb2419f68220b/aiohttp-3.13.2-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a09a6d073fb5789456545bdee2474d14395792faa0527887f2f4ec1a486a59d3", size = 1815823, upload-time = "2025-10-28T20:55:40.958Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/4f/ca2ef819488cbb41844c6cf92ca6dd15b9441e6207c58e5ae0e0fc8d70ad/aiohttp-3.13.2-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b59d13c443f8e049d9e94099c7e412e34610f1f49be0f230ec656a10692a5802", size = 1669374, upload-time = "2025-10-28T20:55:42.745Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/fe/1fe2e1179a0d91ce09c99069684aab619bf2ccde9b20bd6ca44f8837203e/aiohttp-3.13.2-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:20db2d67985d71ca033443a1ba2001c4b5693fe09b0e29f6d9358a99d4d62a8a", size = 1555315, upload-time = "2025-10-28T20:55:44.264Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/2b/f3781899b81c45d7cbc7140cddb8a3481c195e7cbff8e36374759d2ab5a5/aiohttp-3.13.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:960c2fc686ba27b535f9fd2b52d87ecd7e4fd1cf877f6a5cba8afb5b4a8bd204", size = 1639140, upload-time = "2025-10-28T20:55:46.626Z" },
+    { url = "https://files.pythonhosted.org/packages/72/27/c37e85cd3ece6f6c772e549bd5a253d0c122557b25855fb274224811e4f2/aiohttp-3.13.2-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:6c00dbcf5f0d88796151e264a8eab23de2997c9303dd7c0bf622e23b24d3ce22", size = 1645496, upload-time = "2025-10-28T20:55:48.933Z" },
+    { url = "https://files.pythonhosted.org/packages/66/20/3af1ab663151bd3780b123e907761cdb86ec2c4e44b2d9b195ebc91fbe37/aiohttp-3.13.2-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:fed38a5edb7945f4d1bcabe2fcd05db4f6ec7e0e82560088b754f7e08d93772d", size = 1697625, upload-time = "2025-10-28T20:55:50.377Z" },
+    { url = "https://files.pythonhosted.org/packages/95/eb/ae5cab15efa365e13d56b31b0d085a62600298bf398a7986f8388f73b598/aiohttp-3.13.2-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:b395bbca716c38bef3c764f187860e88c724b342c26275bc03e906142fc5964f", size = 1542025, upload-time = "2025-10-28T20:55:51.861Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/2d/1683e8d67ec72d911397fe4e575688d2a9b8f6a6e03c8fdc9f3fd3d4c03f/aiohttp-3.13.2-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:204ffff2426c25dfda401ba08da85f9c59525cdc42bda26660463dd1cbcfec6f", size = 1714918, upload-time = "2025-10-28T20:55:53.515Z" },
+    { url = "https://files.pythonhosted.org/packages/99/a2/ffe8e0e1c57c5e542d47ffa1fcf95ef2b3ea573bf7c4d2ee877252431efc/aiohttp-3.13.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:05c4dd3c48fb5f15db31f57eb35374cb0c09afdde532e7fb70a75aede0ed30f6", size = 1656113, upload-time = "2025-10-28T20:55:55.438Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/42/d511aff5c3a2b06c09d7d214f508a4ad8ac7799817f7c3d23e7336b5e896/aiohttp-3.13.2-cp310-cp310-win32.whl", hash = "sha256:e574a7d61cf10351d734bcddabbe15ede0eaa8a02070d85446875dc11189a251", size = 432290, upload-time = "2025-10-28T20:55:56.96Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/ea/1c2eb7098b5bad4532994f2b7a8228d27674035c9b3234fe02c37469ef14/aiohttp-3.13.2-cp310-cp310-win_amd64.whl", hash = "sha256:364f55663085d658b8462a1c3f17b2b84a5c2e1ba858e1b79bff7b2e24ad1514", size = 455075, upload-time = "2025-10-28T20:55:58.373Z" },
+    { url = "https://files.pythonhosted.org/packages/35/74/b321e7d7ca762638cdf8cdeceb39755d9c745aff7a64c8789be96ddf6e96/aiohttp-3.13.2-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:4647d02df098f6434bafd7f32ad14942f05a9caa06c7016fdcc816f343997dd0", size = 743409, upload-time = "2025-10-28T20:56:00.354Z" },
+    { url = "https://files.pythonhosted.org/packages/99/3d/91524b905ec473beaf35158d17f82ef5a38033e5809fe8742e3657cdbb97/aiohttp-3.13.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:e3403f24bcb9c3b29113611c3c16a2a447c3953ecf86b79775e7be06f7ae7ccb", size = 497006, upload-time = "2025-10-28T20:56:01.85Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/d3/7f68bc02a67716fe80f063e19adbd80a642e30682ce74071269e17d2dba1/aiohttp-3.13.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:43dff14e35aba17e3d6d5ba628858fb8cb51e30f44724a2d2f0c75be492c55e9", size = 493195, upload-time = "2025-10-28T20:56:03.314Z" },
+    { url = "https://files.pythonhosted.org/packages/98/31/913f774a4708775433b7375c4f867d58ba58ead833af96c8af3621a0d243/aiohttp-3.13.2-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e2a9ea08e8c58bb17655630198833109227dea914cd20be660f52215f6de5613", size = 1747759, upload-time = "2025-10-28T20:56:04.904Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/63/04efe156f4326f31c7c4a97144f82132c3bb21859b7bb84748d452ccc17c/aiohttp-3.13.2-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:53b07472f235eb80e826ad038c9d106c2f653584753f3ddab907c83f49eedead", size = 1704456, upload-time = "2025-10-28T20:56:06.986Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/02/4e16154d8e0a9cf4ae76f692941fd52543bbb148f02f098ca73cab9b1c1b/aiohttp-3.13.2-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e736c93e9c274fce6419af4aac199984d866e55f8a4cec9114671d0ea9688780", size = 1807572, upload-time = "2025-10-28T20:56:08.558Z" },
+    { url = "https://files.pythonhosted.org/packages/34/58/b0583defb38689e7f06798f0285b1ffb3a6fb371f38363ce5fd772112724/aiohttp-3.13.2-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:ff5e771f5dcbc81c64898c597a434f7682f2259e0cd666932a913d53d1341d1a", size = 1895954, upload-time = "2025-10-28T20:56:10.545Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/f3/083907ee3437425b4e376aa58b2c915eb1a33703ec0dc30040f7ae3368c6/aiohttp-3.13.2-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a3b6fb0c207cc661fa0bf8c66d8d9b657331ccc814f4719468af61034b478592", size = 1747092, upload-time = "2025-10-28T20:56:12.118Z" },
+    { url = "https://files.pythonhosted.org/packages/ac/61/98a47319b4e425cc134e05e5f3fc512bf9a04bf65aafd9fdcda5d57ec693/aiohttp-3.13.2-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:97a0895a8e840ab3520e2288db7cace3a1981300d48babeb50e7425609e2e0ab", size = 1606815, upload-time = "2025-10-28T20:56:14.191Z" },
+    { url = "https://files.pythonhosted.org/packages/97/4b/e78b854d82f66bb974189135d31fce265dee0f5344f64dd0d345158a5973/aiohttp-3.13.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:9e8f8afb552297aca127c90cb840e9a1d4bfd6a10d7d8f2d9176e1acc69bad30", size = 1723789, upload-time = "2025-10-28T20:56:16.101Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/fc/9d2ccc794fc9b9acd1379d625c3a8c64a45508b5091c546dea273a41929e/aiohttp-3.13.2-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:ed2f9c7216e53c3df02264f25d824b079cc5914f9e2deba94155190ef648ee40", size = 1718104, upload-time = "2025-10-28T20:56:17.655Z" },
+    { url = "https://files.pythonhosted.org/packages/66/65/34564b8765ea5c7d79d23c9113135d1dd3609173da13084830f1507d56cf/aiohttp-3.13.2-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:99c5280a329d5fa18ef30fd10c793a190d996567667908bef8a7f81f8202b948", size = 1785584, upload-time = "2025-10-28T20:56:19.238Z" },
+    { url = "https://files.pythonhosted.org/packages/30/be/f6a7a426e02fc82781afd62016417b3948e2207426d90a0e478790d1c8a4/aiohttp-3.13.2-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:2ca6ffef405fc9c09a746cb5d019c1672cd7f402542e379afc66b370833170cf", size = 1595126, upload-time = "2025-10-28T20:56:20.836Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/c7/8e22d5d28f94f67d2af496f14a83b3c155d915d1fe53d94b66d425ec5b42/aiohttp-3.13.2-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:47f438b1a28e926c37632bff3c44df7d27c9b57aaf4e34b1def3c07111fdb782", size = 1800665, upload-time = "2025-10-28T20:56:22.922Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/11/91133c8b68b1da9fc16555706aa7276fdf781ae2bb0876c838dd86b8116e/aiohttp-3.13.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:9acda8604a57bb60544e4646a4615c1866ee6c04a8edef9b8ee6fd1d8fa2ddc8", size = 1739532, upload-time = "2025-10-28T20:56:25.924Z" },
+    { url = "https://files.pythonhosted.org/packages/17/6b/3747644d26a998774b21a616016620293ddefa4d63af6286f389aedac844/aiohttp-3.13.2-cp311-cp311-win32.whl", hash = "sha256:868e195e39b24aaa930b063c08bb0c17924899c16c672a28a65afded9c46c6ec", size = 431876, upload-time = "2025-10-28T20:56:27.524Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/63/688462108c1a00eb9f05765331c107f95ae86f6b197b865d29e930b7e462/aiohttp-3.13.2-cp311-cp311-win_amd64.whl", hash = "sha256:7fd19df530c292542636c2a9a85854fab93474396a52f1695e799186bbd7f24c", size = 456205, upload-time = "2025-10-28T20:56:29.062Z" },
+    { url = "https://files.pythonhosted.org/packages/29/9b/01f00e9856d0a73260e86dd8ed0c2234a466c5c1712ce1c281548df39777/aiohttp-3.13.2-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:b1e56bab2e12b2b9ed300218c351ee2a3d8c8fdab5b1ec6193e11a817767e47b", size = 737623, upload-time = "2025-10-28T20:56:30.797Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/1b/4be39c445e2b2bd0aab4ba736deb649fabf14f6757f405f0c9685019b9e9/aiohttp-3.13.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:364e25edaabd3d37b1db1f0cbcee8c73c9a3727bfa262b83e5e4cf3489a2a9dc", size = 492664, upload-time = "2025-10-28T20:56:32.708Z" },
+    { url = "https://files.pythonhosted.org/packages/28/66/d35dcfea8050e131cdd731dff36434390479b4045a8d0b9d7111b0a968f1/aiohttp-3.13.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:c5c94825f744694c4b8db20b71dba9a257cd2ba8e010a803042123f3a25d50d7", size = 491808, upload-time = "2025-10-28T20:56:34.57Z" },
+    { url = "https://files.pythonhosted.org/packages/00/29/8e4609b93e10a853b65f8291e64985de66d4f5848c5637cddc70e98f01f8/aiohttp-3.13.2-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ba2715d842ffa787be87cbfce150d5e88c87a98e0b62e0f5aa489169a393dbbb", size = 1738863, upload-time = "2025-10-28T20:56:36.377Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/fa/4ebdf4adcc0def75ced1a0d2d227577cd7b1b85beb7edad85fcc87693c75/aiohttp-3.13.2-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:585542825c4bc662221fb257889e011a5aa00f1ae4d75d1d246a5225289183e3", size = 1700586, upload-time = "2025-10-28T20:56:38.034Z" },
+    { url = "https://files.pythonhosted.org/packages/da/04/73f5f02ff348a3558763ff6abe99c223381b0bace05cd4530a0258e52597/aiohttp-3.13.2-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:39d02cb6025fe1aabca329c5632f48c9532a3dabccd859e7e2f110668972331f", size = 1768625, upload-time = "2025-10-28T20:56:39.75Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/49/a825b79ffec124317265ca7d2344a86bcffeb960743487cb11988ffb3494/aiohttp-3.13.2-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:e67446b19e014d37342f7195f592a2a948141d15a312fe0e700c2fd2f03124f6", size = 1867281, upload-time = "2025-10-28T20:56:41.471Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/48/adf56e05f81eac31edcfae45c90928f4ad50ef2e3ea72cb8376162a368f8/aiohttp-3.13.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4356474ad6333e41ccefd39eae869ba15a6c5299c9c01dfdcfdd5c107be4363e", size = 1752431, upload-time = "2025-10-28T20:56:43.162Z" },
+    { url = "https://files.pythonhosted.org/packages/30/ab/593855356eead019a74e862f21523db09c27f12fd24af72dbc3555b9bfd9/aiohttp-3.13.2-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:eeacf451c99b4525f700f078becff32c32ec327b10dcf31306a8a52d78166de7", size = 1562846, upload-time = "2025-10-28T20:56:44.85Z" },
+    { url = "https://files.pythonhosted.org/packages/39/0f/9f3d32271aa8dc35036e9668e31870a9d3b9542dd6b3e2c8a30931cb27ae/aiohttp-3.13.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d8a9b889aeabd7a4e9af0b7f4ab5ad94d42e7ff679aaec6d0db21e3b639ad58d", size = 1699606, upload-time = "2025-10-28T20:56:46.519Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/3c/52d2658c5699b6ef7692a3f7128b2d2d4d9775f2a68093f74bca06cf01e1/aiohttp-3.13.2-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:fa89cb11bc71a63b69568d5b8a25c3ca25b6d54c15f907ca1c130d72f320b76b", size = 1720663, upload-time = "2025-10-28T20:56:48.528Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/d4/8f8f3ff1fb7fb9e3f04fcad4e89d8a1cd8fc7d05de67e3de5b15b33008ff/aiohttp-3.13.2-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:8aa7c807df234f693fed0ecd507192fc97692e61fee5702cdc11155d2e5cadc8", size = 1737939, upload-time = "2025-10-28T20:56:50.77Z" },
+    { url = "https://files.pythonhosted.org/packages/03/d3/ddd348f8a27a634daae39a1b8e291ff19c77867af438af844bf8b7e3231b/aiohttp-3.13.2-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:9eb3e33fdbe43f88c3c75fa608c25e7c47bbd80f48d012763cb67c47f39a7e16", size = 1555132, upload-time = "2025-10-28T20:56:52.568Z" },
+    { url = "https://files.pythonhosted.org/packages/39/b8/46790692dc46218406f94374903ba47552f2f9f90dad554eed61bfb7b64c/aiohttp-3.13.2-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:9434bc0d80076138ea986833156c5a48c9c7a8abb0c96039ddbb4afc93184169", size = 1764802, upload-time = "2025-10-28T20:56:54.292Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/e4/19ce547b58ab2a385e5f0b8aa3db38674785085abcf79b6e0edd1632b12f/aiohttp-3.13.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ff15c147b2ad66da1f2cbb0622313f2242d8e6e8f9b79b5206c84523a4473248", size = 1719512, upload-time = "2025-10-28T20:56:56.428Z" },
+    { url = "https://files.pythonhosted.org/packages/70/30/6355a737fed29dcb6dfdd48682d5790cb5eab050f7b4e01f49b121d3acad/aiohttp-3.13.2-cp312-cp312-win32.whl", hash = "sha256:27e569eb9d9e95dbd55c0fc3ec3a9335defbf1d8bc1d20171a49f3c4c607b93e", size = 426690, upload-time = "2025-10-28T20:56:58.736Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/0d/b10ac09069973d112de6ef980c1f6bb31cb7dcd0bc363acbdad58f927873/aiohttp-3.13.2-cp312-cp312-win_amd64.whl", hash = "sha256:8709a0f05d59a71f33fd05c17fc11fcb8c30140506e13c2f5e8ee1b8964e1b45", size = 453465, upload-time = "2025-10-28T20:57:00.795Z" },
+    { url = "https://files.pythonhosted.org/packages/04/4a/3da532fdf51b5e58fffa1a86d6569184cb1bf4bf81cd4434b6541a8d14fd/aiohttp-3.13.2-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:7fbdf5ad6084f1940ce88933de34b62358d0f4a0b6ec097362dcd3e5a65a4989", size = 739009, upload-time = "2025-10-28T20:58:55.682Z" },
+    { url = "https://files.pythonhosted.org/packages/89/74/fefa6f7939cdc1d77e5cad712004e675a8847dccc589dcc3abca7feaed73/aiohttp-3.13.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:7c3a50345635a02db61792c85bb86daffac05330f6473d524f1a4e3ef9d0046d", size = 495308, upload-time = "2025-10-28T20:58:58.408Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/b4/a0638ae1f12d09a0dc558870968a2f19a1eba1b10ad0a85ef142ddb40b50/aiohttp-3.13.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:0e87dff73f46e969af38ab3f7cb75316a7c944e2e574ff7c933bc01b10def7f5", size = 490624, upload-time = "2025-10-28T20:59:00.479Z" },
+    { url = "https://files.pythonhosted.org/packages/02/73/361cd4cac9d98a5a4183d1f26faf7b777330f8dba838c5aae2412862bdd0/aiohttp-3.13.2-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:2adebd4577724dcae085665f294cc57c8701ddd4d26140504db622b8d566d7aa", size = 1662968, upload-time = "2025-10-28T20:59:03.105Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/93/ce2ca7584555a6c7dd78f2e6b539a96c5172d88815e13a05a576e14a5a22/aiohttp-3.13.2-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:e036a3a645fe92309ec34b918394bb377950cbb43039a97edae6c08db64b23e2", size = 1627117, upload-time = "2025-10-28T20:59:05.274Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/42/7ee0e699111f5fc20a69b3203e8f5d5da0b681f270b90bc088d15e339980/aiohttp-3.13.2-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:23ad365e30108c422d0b4428cf271156dd56790f6dd50d770b8e360e6c5ab2e6", size = 1724037, upload-time = "2025-10-28T20:59:07.522Z" },
+    { url = "https://files.pythonhosted.org/packages/66/88/67ad5ff11dd61dd1d7882cda39f085d5fca31cf7e2143f5173429d8a591e/aiohttp-3.13.2-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:1f9b2c2d4b9d958b1f9ae0c984ec1dd6b6689e15c75045be8ccb4011426268ca", size = 1812899, upload-time = "2025-10-28T20:59:11.698Z" },
+    { url = "https://files.pythonhosted.org/packages/60/1b/a46f6e1c2a347b9c7a789292279c159b327fadecbf8340f3b05fffff1151/aiohttp-3.13.2-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3a92cf4b9bea33e15ecbaa5c59921be0f23222608143d025c989924f7e3e0c07", size = 1660961, upload-time = "2025-10-28T20:59:14.425Z" },
+    { url = "https://files.pythonhosted.org/packages/44/cc/1af9e466eafd9b5d8922238c69aaf95b656137add4c5db65f63ee129bf3c/aiohttp-3.13.2-cp39-cp39-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:070599407f4954021509193404c4ac53153525a19531051661440644728ba9a7", size = 1553851, upload-time = "2025-10-28T20:59:17.044Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/d1/9e5f4f40f9d0ee5668e9b5e7ebfb0eaf371cc09da03785decdc5da56f4b3/aiohttp-3.13.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:29562998ec66f988d49fb83c9b01694fa927186b781463f376c5845c121e4e0b", size = 1634260, upload-time = "2025-10-28T20:59:19.378Z" },
+    { url = "https://files.pythonhosted.org/packages/83/2e/5d065091c4ae8b55a153f458f19308191bad3b62a89496aa081385486338/aiohttp-3.13.2-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:4dd3db9d0f4ebca1d887d76f7cdbcd1116ac0d05a9221b9dad82c64a62578c4d", size = 1639499, upload-time = "2025-10-28T20:59:22.013Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/de/58ae6dc73691a51ff16f69a94d13657bf417456fa0fdfed2b59dd6b4c293/aiohttp-3.13.2-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:d7bc4b7f9c4921eba72677cd9fedd2308f4a4ca3e12fab58935295ad9ea98700", size = 1694087, upload-time = "2025-10-28T20:59:24.773Z" },
+    { url = "https://files.pythonhosted.org/packages/45/fe/4d9df516268867d83041b6c073ee15cd532dbea58b82d675a7e1cf2ec24c/aiohttp-3.13.2-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:dacd50501cd017f8cccb328da0c90823511d70d24a323196826d923aad865901", size = 1540532, upload-time = "2025-10-28T20:59:27.982Z" },
+    { url = "https://files.pythonhosted.org/packages/24/e7/a802619308232499482bf30b3530efb5d141481cfd61850368350fb1acb5/aiohttp-3.13.2-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:8b2f1414f6a1e0683f212ec80e813f4abef94c739fd090b66c9adf9d2a05feac", size = 1710369, upload-time = "2025-10-28T20:59:30.363Z" },
+    { url = "https://files.pythonhosted.org/packages/62/08/e8593f39f025efe96ef59550d17cf097222d84f6f84798bedac5bf037fce/aiohttp-3.13.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:04c3971421576ed24c191f610052bcb2f059e395bc2489dd99e397f9bc466329", size = 1649296, upload-time = "2025-10-28T20:59:33.285Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/fd/ffbc1b6aa46fc6c284af4a438b2c7eab79af1c8ac4b6d2ced185c17f403e/aiohttp-3.13.2-cp39-cp39-win32.whl", hash = "sha256:9f377d0a924e5cc94dc620bc6366fc3e889586a7f18b748901cf016c916e2084", size = 432980, upload-time = "2025-10-28T20:59:35.515Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/a9/d47e7873175a4d8aed425f2cdea2df700b2dd44fac024ffbd83455a69a50/aiohttp-3.13.2-cp39-cp39-win_amd64.whl", hash = "sha256:9c705601e16c03466cb72011bd1af55d68fa65b045356d8f96c216e5f6db0fa5", size = 456021, upload-time = "2025-10-28T20:59:37.659Z" },
 ]
 
 [[package]]
@@ -162,6 +187,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/16/87/59b570b9c4b014532777dc3baffc9bea10cf0cc8b232cf3c17e4bd0754a6/airportsdata-20250909-py3-none-any.whl", hash = "sha256:ce7dc6e1485afe3915e708212c7024ad158470c1c934e6a6cb217cf28b798ac7", size = 914391, upload-time = "2025-09-09T01:07:29.364Z" },
 ]
 
+[[package]]
+name = "annotated-doc"
+version = "0.0.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/57/ba/046ceea27344560984e26a590f90bc7f4a75b06701f653222458922b558c/annotated_doc-0.0.4.tar.gz", hash = "sha256:fbcda96e87e9c92ad167c2e53839e57503ecfda18804ea28102353485033faa4", size = 7288, upload-time = "2025-11-10T22:07:42.062Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1e/d3/26bf1008eb3d2daa8ef4cacc7f3bfdc11818d111f7e2d0201bc6e3b49d45/annotated_doc-0.0.4-py3-none-any.whl", hash = "sha256:571ac1dc6991c450b25a9c2d84a3705e2ae7a53467b5d111c24fa8baabbed320", size = 5303, upload-time = "2025-11-10T22:07:40.673Z" },
+]
+
 [[package]]
 name = "annotated-types"
 version = "0.7.0"
@@ -173,7 +207,7 @@ wheels = [
 
 [[package]]
 name = "anthropic"
-version = "0.69.0"
+version = "0.74.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
@@ -185,9 +219,9 @@ dependencies = [
     { name = "sniffio" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/c8/9d/9ad1778b95f15c5b04e7d328c1b5f558f1e893857b7c33cd288c19c0057a/anthropic-0.69.0.tar.gz", hash = "sha256:c604d287f4d73640f40bd2c0f3265a2eb6ce034217ead0608f6b07a8bc5ae5f2", size = 480622, upload-time = "2025-09-29T16:53:45.282Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/d7/7b/609eea5c54ae69b1a4a94169d4b0c86dc5c41b43509989913f6cdc61b81d/anthropic-0.74.1.tar.gz", hash = "sha256:04c087b2751385c524f6d332d066a913870e4de8b3e335fb0a0c595f1f88dc6e", size = 428981, upload-time = "2025-11-19T22:17:31.533Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/9b/38/75129688de5637eb5b383e5f2b1570a5cc3aecafa4de422da8eea4b90a6c/anthropic-0.69.0-py3-none-any.whl", hash = "sha256:1f73193040f33f11e27c2cd6ec25f24fe7c3f193dc1c5cde6b7a08b18a16bcc5", size = 337265, upload-time = "2025-09-29T16:53:43.686Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/45/6b18d0692302b8cbc01a10c35b43953d3c4172fbd4f83337b8ed21a8eaa4/anthropic-0.74.1-py3-none-any.whl", hash = "sha256:b07b998d1cee7f41d9f02530597d7411672b362cc2417760a40c0167b81c6e65", size = 371473, upload-time = "2025-11-19T22:17:29.998Z" },
 ]
 
 [[package]]
@@ -205,15 +239,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/15/b3/9b1a8074496371342ec1e796a96f99c82c945a339cd81a8e73de28b4cf9e/anyio-4.11.0-py3-none-any.whl", hash = "sha256:0287e96f4d26d4149305414d4e3bc32f0dcd0862365a4bddea19d7a1ec38c4fc", size = 109097, upload-time = "2025-09-23T09:19:10.601Z" },
 ]
 
-[[package]]
-name = "astor"
-version = "0.8.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/5a/21/75b771132fee241dfe601d39ade629548a9626d1d39f333fde31bc46febe/astor-0.8.1.tar.gz", hash = "sha256:6a6effda93f4e1ce9f618779b2dd1d9d84f1e32812c23a29b3fff6fd7f63fa5e", size = 35090, upload-time = "2019-12-10T01:50:35.51Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/c3/88/97eef84f48fa04fbd6750e62dcceafba6c63c81b7ac1420856c8dcc0a3f9/astor-0.8.1-py2.py3-none-any.whl", hash = "sha256:070a54e890cefb5b3739d19f30f5a5ec840ffc9c50ffa7d23cc9fc1a38ebbfc5", size = 27488, upload-time = "2019-12-10T01:50:33.628Z" },
-]
-
 [[package]]
 name = "astunparse"
 version = "1.6.3"
@@ -272,141 +297,48 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/6f/e7/9220de010e015fdfb7ce22e7d9a846bfcd6e5d686c4fc555fa76a22846ba/beartype-0.15.0-py3-none-any.whl", hash = "sha256:52cd2edea72fdd84e4e7f8011a9e3007bf0125c3d6d7219e937b9d8868169177", size = 777565, upload-time = "2023-07-22T05:25:41.659Z" },
 ]
 
-[[package]]
-name = "blake3"
-version = "1.0.7"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "typing-extensions", marker = "python_full_version < '3.12'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/14/aa/e767c874392e95c42231a71072e6d47b9411b0c5fc5b221ed1304d63e4c1/blake3-1.0.7.tar.gz", hash = "sha256:228c617d4a92b45ef06c63236348001976245eb8efd561dbe181a0def524173e", size = 114008, upload-time = "2025-09-29T16:41:23.083Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/ce/45/68acac32a1d8e41312c48ba0103515baf09dd79201deb6dd0262faa47eea/blake3-1.0.7-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:924f410e2d5f07a383473645b9c5da0b08220e1cbb54cfcf272ceaf53004e24b", size = 370989, upload-time = "2025-09-29T16:39:28.385Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/48/0e12f0f7603e69564247f4e1d5648e658fe1d3a6afd23a658985edbec672/blake3-1.0.7-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:10d2aa630238a6fa25a812d7d178f29b353c41eb34b2336dd11a7d2dce804bdf", size = 375310, upload-time = "2025-09-29T16:39:29.884Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/51/7d43e4e90aed90bccf72d3502bbea7a17d7eaabd8ce7f9d7f17de3bdddc6/blake3-1.0.7-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:385de4f30d90242b7b4bd2ce597d2cfdcd4d10292b410fc683c32e6bbf3d1acf", size = 448625, upload-time = "2025-09-29T16:39:31.448Z" },
-    { url = "https://files.pythonhosted.org/packages/47/ef/7ae2f8a1bac8e0201aa4e1416a4357b6f22092a51a36b0d8045e9e644f60/blake3-1.0.7-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:bbe44233e96d39082eadb41cb3cd7f6f0ebca3948a5bb0529166afd965b091e2", size = 506388, upload-time = "2025-09-29T16:39:33.009Z" },
-    { url = "https://files.pythonhosted.org/packages/da/98/90bc7a158d9dfa734644f8474bd1a34ab1b7c2670ed3dbf3c5acab8f2da3/blake3-1.0.7-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:dc4000c194b3537b6ce97107f56ff881f9d8c7f91203698ce63978f597dfced5", size = 393443, upload-time = "2025-09-29T16:39:34.242Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/ee/642b3d7abbd09541717d4b96ebfbb2c79eacceefcffa99e44105be5ef249/blake3-1.0.7-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4147dbe2a6c76351cd2c122e55c06d8ae19cc94ef519b6a5be7d27d516c9e897", size = 387774, upload-time = "2025-09-29T16:39:35.455Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/3f/d73b00827135182bdbd8f071ef7f7bf0819fdad1837f91cc9baa75358f01/blake3-1.0.7-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:21fd54ff74959dedf148afee42a84b50e91a813ed4957bf0cceacfc1ebcf28fa", size = 549894, upload-time = "2025-09-29T16:39:37.039Z" },
-    { url = "https://files.pythonhosted.org/packages/dc/5d/7c2b12832094b782ddb7f03936c289482b8a2180c76d97300550f8e78b54/blake3-1.0.7-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:f4f93d6d4e3e8b6366e706199bc71043d60cff235489c6a0fd9fd5aa1b9442d7", size = 554126, upload-time = "2025-09-29T16:39:38.35Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/0e/9e085fc2a19a4f81ad772c08193f4f95d3dea6ad25e713dafaec81e4f6ad/blake3-1.0.7-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e88d048e41035b0630bd5f2998224ebd50130d707496f19c7ec0b78946de1fc1", size = 371023, upload-time = "2025-09-29T16:39:44.627Z" },
-    { url = "https://files.pythonhosted.org/packages/18/88/63c5e8a3b4d3b1c54a6250c9d2674e2ed7bc6daf5d51d5633ef7d757ba52/blake3-1.0.7-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:532591a10c36e8de04d2caee128f4491136119f1224a11435b07c87f5da841b8", size = 375277, upload-time = "2025-09-29T16:39:45.829Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/15/cd5925106cc0f8d28dd141ecc131495717bb7b1a288197a92cade39c1a18/blake3-1.0.7-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:a36ba39662bd504a3363d70e0654726e295c080e267f758a4801303991f30d00", size = 448409, upload-time = "2025-09-29T16:39:47.324Z" },
-    { url = "https://files.pythonhosted.org/packages/27/81/f06e2078cb1202c0aa86f3d6257a069a80a9acc370b1656b89073f5c2384/blake3-1.0.7-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:47d82d24acb22c7c68eeaac5bdf7dedb0cce2831cf00fcee92d459bf07ac51b1", size = 506520, upload-time = "2025-09-29T16:39:48.933Z" },
-    { url = "https://files.pythonhosted.org/packages/11/70/566513a733dfc1691df76d437c1e6b389261cd44c57bd361d9daddd68575/blake3-1.0.7-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c87822a2f7ea23aa5536256414d6c552db7d027941e3e25ab57807714877b40b", size = 393138, upload-time = "2025-09-29T16:39:50.508Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/58/16b74904befa9404bf3df2d6608b72b64d9d91209b7af491309f5644eb26/blake3-1.0.7-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8807d115255f235aa536269691581e59e9463bf1ece973e0e1e89d250a3f01c4", size = 387867, upload-time = "2025-09-29T16:39:51.694Z" },
-    { url = "https://files.pythonhosted.org/packages/ac/2c/4930b54db10506ea65f0237578e4c69c3a98fd9bf2df01c8d1e193686308/blake3-1.0.7-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:dc97d1ed59edc26fb6e274cd00b4d3a3dc3d9ac7dbbcc0515dd2907c7a96cc6d", size = 550087, upload-time = "2025-09-29T16:39:53.316Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/f2/f94e27ab505ed5f25312f6d0b49cd8f5fd1258f162c27f19664b42eb04f4/blake3-1.0.7-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:3c515ab26dcc5d433bb5bf047b4e3d392ded742e7bc3255f1fa26014e20c93b2", size = 553940, upload-time = "2025-09-29T16:39:54.852Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/35/1fc595de42168569a83c49aff0aab94feef3c10c4152ad68f8046ad9839a/blake3-1.0.7-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5a7ef1c2c0da34c336e6ee4fa02b22cf142dc1a171e84d1cc46e9948c4e2c76d", size = 370790, upload-time = "2025-09-29T16:40:01.476Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/cb/58c5e33dde29b86e8a196d3acb2c1e25436ed58f48cd8ab5d4668cc085ec/blake3-1.0.7-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:46bfeeac655420f7421897a6d4754aafa9cd02bbcda869fc0bc358b19d90d864", size = 375207, upload-time = "2025-09-29T16:40:02.673Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/a9/4e59c99d8d4d137424e2175dbd67be48862a6543880dd9d78bc0f42be4ed/blake3-1.0.7-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:55095edd17183028c5be4fab204ee48e635ce78bd9f499ca1f95c0ed011fbf12", size = 448151, upload-time = "2025-09-29T16:40:03.888Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/55/d2e7d53a654193aad9ac27f3748f903791a9ebb88ba2fbfc1fd3e2629b59/blake3-1.0.7-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:05f8430d31aa11aa9b488b4a6ca07e35cc402b1c104d39ccbc99decdeb895130", size = 504463, upload-time = "2025-09-29T16:40:05.71Z" },
-    { url = "https://files.pythonhosted.org/packages/a5/d0/3a2e712202ab85e972a0419caa3465153234adeb8429ba7a0bd4c23b936e/blake3-1.0.7-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a5d19e9a88274c85eb5c9985e58dd3f2cf56b1de0f71a77a4ba7fac2a2c70bf8", size = 393701, upload-time = "2025-09-29T16:40:06.928Z" },
-    { url = "https://files.pythonhosted.org/packages/db/30/addf808816833de5e497a1bf9d10221b1fc8048703b3de2c21ac6450326f/blake3-1.0.7-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4ef46ca98068d0153a8fd12215df8a24844800762fd5a3145658d89a3335990c", size = 387936, upload-time = "2025-09-29T16:40:08.214Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/70/48588e41c62d4ae1164acab51ba17bb15eca7499e18731645831a2e3459f/blake3-1.0.7-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:f4c327bdb172c08b935a98d1f527c07437d825ab317469f35ea43243b58898bc", size = 550142, upload-time = "2025-09-29T16:40:09.898Z" },
-    { url = "https://files.pythonhosted.org/packages/73/48/1aa83fd4671886eee2ab1497dcfa7089797975b927c79aefc418a2b928f3/blake3-1.0.7-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:bbfad4c447e8f910b4f2442e79768e5a3f6d9b1a31453fc780bab25e3e03d9de", size = 553949, upload-time = "2025-09-29T16:40:11.183Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/a6/ef3c2d14bf723173b72faeb3d79d487da1ed5160878d2f2e80911e821f8f/blake3-1.0.7-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:26bf2b2376cab4160f91759324dc5ebe3e4c61fd316781690bbba04c6f6d1574", size = 372798, upload-time = "2025-09-29T16:41:08.594Z" },
-    { url = "https://files.pythonhosted.org/packages/85/1e/2d347a865d890d139f6c4f99af1f012d1d3c0540c677a11eca6c04944fab/blake3-1.0.7-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:4959f47978cbad9a82774bc4efb02e2d4243e99db6f358ad5aa3df7704474bc0", size = 376973, upload-time = "2025-09-29T16:41:10.038Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/d2/6920923b207e62a644ff76ea23c6a077b1e6ef1e0b559350364448c72de6/blake3-1.0.7-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:abbc90d68f6918361821f107aec2de585d141e35deb362bc93854d2bfb105338", size = 450561, upload-time = "2025-09-29T16:41:11.381Z" },
-    { url = "https://files.pythonhosted.org/packages/df/d8/d6f4c495e1d5cf415d4115c6530cdfe0198031f1428ec27f0465917933d1/blake3-1.0.7-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d0a4fed89b9674f9651f1a9d88f8ae9b247e80d36e9b60222ea278cd2cf71b7d", size = 506581, upload-time = "2025-09-29T16:41:12.987Z" },
-    { url = "https://files.pythonhosted.org/packages/a0/59/4ad0640d3c8b2b52f211902a5d814e5e9d9025ce6f6e2c6178c0d3238236/blake3-1.0.7-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8bdc842630cc6b27c8d038fce15f2b91684a95b65afb24f2075b241332dc0962", size = 394619, upload-time = "2025-09-29T16:41:14.349Z" },
-    { url = "https://files.pythonhosted.org/packages/64/34/21f8207496e8f26c9158438b7eae3cb62d270997d4c04e0363d58fcba408/blake3-1.0.7-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f6322f73e2255897ef31a3a10901d056f7215354dee68a3b8f8e18e4da2e694b", size = 389447, upload-time = "2025-09-29T16:41:15.687Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/26/f75a2dc988fbe2d1e632580d7b85275bd8dd711ce037fa1d44f866b874fa/blake3-1.0.7-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:e0d6f20b0e24c8438a565c0fcc81e376d5f1c0eaf28932cb229166736bb6e6ed", size = 551815, upload-time = "2025-09-29T16:41:17.287Z" },
-    { url = "https://files.pythonhosted.org/packages/4f/66/388874447798e68e0b486e5f9e9cf069d27bff78f0f2132d6dd51c16fba7/blake3-1.0.7-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:f30a9c8c8f4c9423cb76ccce17f33e7a3f6ff5cde2d204591ff9a13b826aa129", size = 555876, upload-time = "2025-09-29T16:41:18.906Z" },
-]
-
 [[package]]
 name = "cachetools"
-version = "6.2.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/9d/61/e4fad8155db4a04bfb4734c7c8ff0882f078f24294d42798b3568eb63bff/cachetools-6.2.0.tar.gz", hash = "sha256:38b328c0889450f05f5e120f56ab68c8abaf424e1275522b138ffc93253f7e32", size = 30988, upload-time = "2025-08-25T18:57:30.924Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/6c/56/3124f61d37a7a4e7cc96afc5492c78ba0cb551151e530b54669ddd1436ef/cachetools-6.2.0-py3-none-any.whl", hash = "sha256:1c76a8960c0041fcc21097e357f882197c79da0dbff766e7317890a65d7d8ba6", size = 11276, upload-time = "2025-08-25T18:57:29.684Z" },
-]
-
-[[package]]
-name = "cbor2"
-version = "5.7.0"
+version = "6.2.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/3a/89/01df16cdc9c60c07956756c90fe92c684021003079e358a78e213bce45a2/cbor2-5.7.0.tar.gz", hash = "sha256:3f6d843f4db4d0ec501c46453c22a4fbebb1abfb5b740e1bcab34c615cd7406b", size = 102374, upload-time = "2025-08-14T08:59:47.294Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/fb/44/ca1675be2a83aeee1886ab745b28cda92093066590233cc501890eb8417a/cachetools-6.2.2.tar.gz", hash = "sha256:8e6d266b25e539df852251cfd6f990b4bc3a141db73b939058d809ebd2590fc6", size = 31571, upload-time = "2025-11-13T17:42:51.465Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/7f/b7/17fe6e502b9884f3a7bdf6e9331688e290a546110dd14575aa09527b238f/cbor2-5.7.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:36b6e229b36147051ea1063ba0cd6225bfe6b5398ca0ac7b33fa91407ca75081", size = 254198, upload-time = "2025-08-14T08:58:40.293Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/fb/6cc1dd51beb35f899f9f6621fd0df9ec733e1f4c27f334212107f7a05822/cbor2-5.7.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:06f8e9d8fa6d0ed7b8a81f4613a60e740be9bd5087de620e3b64007832dfd815", size = 247202, upload-time = "2025-08-14T08:58:42.101Z" },
-    { url = "https://files.pythonhosted.org/packages/34/2d/e51cd823cb918eed3180bbe0169955ccefa03817c0e4788024e9e56d3657/cbor2-5.7.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:90c1236e5bfad19493183f9a9ecd3705e3ad3eb02fce6d6381c12ece86147b15", size = 250001, upload-time = "2025-08-14T08:58:43.82Z" },
-    { url = "https://files.pythonhosted.org/packages/de/45/79a0320a4562537df3e589526d636963a06a95a66d7b6c366bdc8d9b3755/cbor2-5.7.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:afc6c6611b7529136ea4a04cf8711786640b687859c9688ae172726c305a83a1", size = 243743, upload-time = "2025-08-14T08:58:45.618Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/52/6092e45a6b164d22bc2828aa14f422fbcf5af4bd9201a3e27b62b512f828/cbor2-5.7.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5fc1005b412ace94bbf905a8c4214e639557568551d9b5474645789e976e91e4", size = 261081, upload-time = "2025-08-14T08:58:53.747Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/12/9eef73bcfcdd110df7e585f8034399d1cf2006dc7576487e0509ba50e4f8/cbor2-5.7.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d981d691dd721dd712fc824d04a01347955a206107fbee1d75803fa58de172c7", size = 254132, upload-time = "2025-08-14T08:58:55.465Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/e1/00bc92353ecacef0b8eaba285168ca517c18539c12ddebe06a9bbab03e47/cbor2-5.7.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b3b9730928163d02e7f1267e48a39ed75560ff3e56cdc6768d670d3e96028623", size = 256825, upload-time = "2025-08-14T08:58:57.239Z" },
-    { url = "https://files.pythonhosted.org/packages/25/01/454c93f360c404a3e6531ce0e2e4cc5818c5746a10b1cb295b6abd3057cd/cbor2-5.7.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:01cc4d840943b4c2e468b8560830235d044a8cb45e5d66ba3ae933c3e21b8d49", size = 250639, upload-time = "2025-08-14T08:58:58.594Z" },
-    { url = "https://files.pythonhosted.org/packages/1e/6a/9aed5b716407c1d48425ba55c6022a01a9abdbf58a691f50416461fa371d/cbor2-5.7.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:16bea83598a1eeedbd50c2e9fdf3685bae78ca9d9ec8cd8010777db14a315578", size = 285685, upload-time = "2025-08-14T08:59:05.165Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/6e/3499eaa2b858c7695a447b6311303f06ffc90fc2c45851337121661f1f5c/cbor2-5.7.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8e837825a16e60ace6e284095aa9fbe504bf87a8f4494bf7d95931e37fb01a70", size = 284948, upload-time = "2025-08-14T08:59:06.64Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/3e/ae67866ef65717665e0acf2873d466c5d4a1d965b0d0348f2269b73f28fb/cbor2-5.7.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:27396c5e275ff7c7cd87fe8aaadf781e6194903921f250934af7c86d5efec82e", size = 276375, upload-time = "2025-08-14T08:59:08.845Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/3d/2f8e9671111661dd571de206344ecd7706f6d458aab191e06834c89aa58e/cbor2-5.7.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:c84bfef78c4e9c81eb0a10cec340222ba4e39498a63fc2e3d5f982a3f4efa4a7", size = 277680, upload-time = "2025-08-14T08:59:10.292Z" },
-    { url = "https://files.pythonhosted.org/packages/cd/40/dd8123fef6b18220f78e64a4bb4f8e4255c516db390ae8e8db0084474503/cbor2-5.7.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f94dbfd77ae591010429d645e4e7fd4f387183193f14c4f645dd7ba03668f9a9", size = 252958, upload-time = "2025-08-14T08:59:38.279Z" },
-    { url = "https://files.pythonhosted.org/packages/df/1a/644bf42929915875212744100131200bc17cf526235c28747c78b30588db/cbor2-5.7.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:cd36c935adf56ccf83c6b85949cb2ac8b991672d241ec1f645b557ae4730221b", size = 246023, upload-time = "2025-08-14T08:59:39.645Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/c1/32ad36a8e036b10a49412c0e55a3f96ecf1bbfc7b640ee80cca011731eb2/cbor2-5.7.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1198cbfa5194410b35b7fb5c0e58a0ee22c20b44197605e26f14e320e1750c5e", size = 248762, upload-time = "2025-08-14T08:59:41.048Z" },
-    { url = "https://files.pythonhosted.org/packages/ef/db/847c604f876bb5990707b42e52c00960bdb81df800e4dfa81050d5f4ef5c/cbor2-5.7.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:78dca93abdbea701caa068e412e54923f3b4d117f5b46b39f44456e01b047534", size = 242520, upload-time = "2025-08-14T08:59:42.406Z" },
-    { url = "https://files.pythonhosted.org/packages/41/cc/0ce73676d2a0c9e5a9330c301940c50eb325dacf5f6d9690fd43a8817fe9/cbor2-5.7.0-py3-none-any.whl", hash = "sha256:a871e7a6f7cba1ddb02503ea974f15f6524c95078fbfe0b860fd4193d7c8f27a", size = 23828, upload-time = "2025-08-14T08:59:46.129Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/46/eb6eca305c77a4489affe1c5d8f4cae82f285d9addd8de4ec084a7184221/cachetools-6.2.2-py3-none-any.whl", hash = "sha256:6c09c98183bf58560c97b2abfcedcbaf6a896a490f534b031b661d3723b45ace", size = 11503, upload-time = "2025-11-13T17:42:50.232Z" },
 ]
 
 [[package]]
 name = "certifi"
-version = "2025.10.5"
+version = "2025.11.12"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/4c/5b/b6ce21586237c77ce67d01dc5507039d444b630dd76611bbca2d8e5dcd91/certifi-2025.10.5.tar.gz", hash = "sha256:47c09d31ccf2acf0be3f701ea53595ee7e0b8fa08801c6624be771df09ae7b43", size = 164519, upload-time = "2025-10-05T04:12:15.808Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/a2/8c/58f469717fa48465e4a50c014a0400602d3c437d7c0c468e17ada824da3a/certifi-2025.11.12.tar.gz", hash = "sha256:d8ab5478f2ecd78af242878415affce761ca6bc54a22a27e026d7c25357c3316", size = 160538, upload-time = "2025-11-12T02:54:51.517Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e4/37/af0d2ef3967ac0d6113837b44a4f0bfe1328c2b9763bd5b1744520e5cfed/certifi-2025.10.5-py3-none-any.whl", hash = "sha256:0f212c2744a9bb6de0c56639a6f68afe01ecd92d91f14ae897c4fe7bbeeef0de", size = 163286, upload-time = "2025-10-05T04:12:14.03Z" },
+    { url = "https://files.pythonhosted.org/packages/70/7d/9bc192684cea499815ff478dfcdc13835ddf401365057044fb721ec6bddb/certifi-2025.11.12-py3-none-any.whl", hash = "sha256:97de8790030bbd5c2d96b7ec782fc2f7820ef8dba6db909ccf95449f2d062d4b", size = 159438, upload-time = "2025-11-12T02:54:49.735Z" },
 ]
 
 [[package]]
-name = "cffi"
-version = "2.0.0"
+name = "cfgv"
+version = "3.4.0"
 source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "pycparser", marker = "implementation_name != 'PyPy'" },
+resolution-markers = [
+    "python_full_version < '3.10'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/eb/56/b1ba7935a17738ae8453301356628e8147c79dbb825bcbc73dc7401f9846/cffi-2.0.0.tar.gz", hash = "sha256:44d1b5909021139fe36001ae048dbdde8214afa20200eda0f64c068cac5d5529", size = 523588, upload-time = "2025-09-08T23:24:04.541Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/11/74/539e56497d9bd1d484fd863dd69cbbfa653cd2aa27abfe35653494d85e94/cfgv-3.4.0.tar.gz", hash = "sha256:e52591d4c5f5dead8e0f673fb16db7949d2cfb3f7da4582893288f0ded8fe560", size = 7114, upload-time = "2023-08-12T20:38:17.776Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/50/bd/b1a6362b80628111e6653c961f987faa55262b4002fcec42308cad1db680/cffi-2.0.0-cp310-cp310-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:53f77cbe57044e88bbd5ed26ac1d0514d2acf0591dd6bb02a3ae37f76811b80c", size = 208811, upload-time = "2025-09-08T23:22:12.267Z" },
-    { url = "https://files.pythonhosted.org/packages/4f/27/6933a8b2562d7bd1fb595074cf99cc81fc3789f6a6c05cdabb46284a3188/cffi-2.0.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:3e837e369566884707ddaf85fc1744b47575005c0a229de3327f8f9a20f4efeb", size = 216402, upload-time = "2025-09-08T23:22:13.455Z" },
-    { url = "https://files.pythonhosted.org/packages/05/eb/b86f2a2645b62adcfff53b0dd97e8dfafb5c8aa864bd0d9a2c2049a0d551/cffi-2.0.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:5eda85d6d1879e692d546a078b44251cdd08dd1cfb98dfb77b670c97cee49ea0", size = 203217, upload-time = "2025-09-08T23:22:14.596Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/e0/6cbe77a53acf5acc7c08cc186c9928864bd7c005f9efd0d126884858a5fe/cffi-2.0.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:9332088d75dc3241c702d852d4671613136d90fa6881da7d770a483fd05248b4", size = 203079, upload-time = "2025-09-08T23:22:15.769Z" },
-    { url = "https://files.pythonhosted.org/packages/98/29/9b366e70e243eb3d14a5cb488dfd3a0b6b2f1fb001a203f653b93ccfac88/cffi-2.0.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:fc7de24befaeae77ba923797c7c87834c73648a05a4bde34b3b7e5588973a453", size = 216475, upload-time = "2025-09-08T23:22:17.427Z" },
-    { url = "https://files.pythonhosted.org/packages/21/7a/13b24e70d2f90a322f2900c5d8e1f14fa7e2a6b3332b7309ba7b2ba51a5a/cffi-2.0.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:cf364028c016c03078a23b503f02058f1814320a56ad535686f90565636a9495", size = 218829, upload-time = "2025-09-08T23:22:19.069Z" },
-    { url = "https://files.pythonhosted.org/packages/60/99/c9dc110974c59cc981b1f5b66e1d8af8af764e00f0293266824d9c4254bc/cffi-2.0.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:e11e82b744887154b182fd3e7e8512418446501191994dbf9c9fc1f32cc8efd5", size = 211211, upload-time = "2025-09-08T23:22:20.588Z" },
-    { url = "https://files.pythonhosted.org/packages/49/72/ff2d12dbf21aca1b32a40ed792ee6b40f6dc3a9cf1644bd7ef6e95e0ac5e/cffi-2.0.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:8ea985900c5c95ce9db1745f7933eeef5d314f0565b27625d9a10ec9881e1bfb", size = 218036, upload-time = "2025-09-08T23:22:22.143Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/b7/1200d354378ef52ec227395d95c2576330fd22a869f7a70e88e1447eb234/cffi-2.0.0-cp311-cp311-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:baf5215e0ab74c16e2dd324e8ec067ef59e41125d3eade2b863d294fd5035c92", size = 209613, upload-time = "2025-09-08T23:22:29.475Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/56/6033f5e86e8cc9bb629f0077ba71679508bdf54a9a5e112a3c0b91870332/cffi-2.0.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:730cacb21e1bdff3ce90babf007d0a0917cc3e6492f336c2f0134101e0944f93", size = 216476, upload-time = "2025-09-08T23:22:31.063Z" },
-    { url = "https://files.pythonhosted.org/packages/dc/7f/55fecd70f7ece178db2f26128ec41430d8720f2d12ca97bf8f0a628207d5/cffi-2.0.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:6824f87845e3396029f3820c206e459ccc91760e8fa24422f8b0c3d1731cbec5", size = 203374, upload-time = "2025-09-08T23:22:32.507Z" },
-    { url = "https://files.pythonhosted.org/packages/84/ef/a7b77c8bdc0f77adc3b46888f1ad54be8f3b7821697a7b89126e829e676a/cffi-2.0.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:9de40a7b0323d889cf8d23d1ef214f565ab154443c42737dfe52ff82cf857664", size = 202597, upload-time = "2025-09-08T23:22:34.132Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/91/500d892b2bf36529a75b77958edfcd5ad8e2ce4064ce2ecfeab2125d72d1/cffi-2.0.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:8941aaadaf67246224cee8c3803777eed332a19d909b47e29c9842ef1e79ac26", size = 215574, upload-time = "2025-09-08T23:22:35.443Z" },
-    { url = "https://files.pythonhosted.org/packages/44/64/58f6255b62b101093d5df22dcb752596066c7e89dd725e0afaed242a61be/cffi-2.0.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:a05d0c237b3349096d3981b727493e22147f934b20f6f125a3eba8f994bec4a9", size = 218971, upload-time = "2025-09-08T23:22:36.805Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/49/fa72cebe2fd8a55fbe14956f9970fe8eb1ac59e5df042f603ef7c8ba0adc/cffi-2.0.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:94698a9c5f91f9d138526b48fe26a199609544591f859c870d477351dc7b2414", size = 211972, upload-time = "2025-09-08T23:22:38.436Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/28/dd0967a76aab36731b6ebfe64dec4e981aff7e0608f60c2d46b46982607d/cffi-2.0.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:5fed36fccc0612a53f1d4d9a816b50a36702c28a2aa880cb8a122b3466638743", size = 217078, upload-time = "2025-09-08T23:22:39.776Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/df/a4f0fbd47331ceeba3d37c2e51e9dfc9722498becbeec2bd8bc856c9538a/cffi-2.0.0-cp312-cp312-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:21d1152871b019407d8ac3985f6775c079416c282e431a4da6afe7aefd2bccbe", size = 212529, upload-time = "2025-09-08T23:22:47.349Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/72/12b5f8d3865bf0f87cf1404d8c374e7487dcf097a1c91c436e72e6badd83/cffi-2.0.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:b21e08af67b8a103c71a250401c78d5e0893beff75e28c53c98f4de42f774062", size = 220097, upload-time = "2025-09-08T23:22:48.677Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/95/7a135d52a50dfa7c882ab0ac17e8dc11cec9d55d2c18dda414c051c5e69e/cffi-2.0.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:1e3a615586f05fc4065a8b22b8152f0c1b00cdbc60596d187c2a74f9e3036e4e", size = 207983, upload-time = "2025-09-08T23:22:50.06Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/c8/15cb9ada8895957ea171c62dc78ff3e99159ee7adb13c0123c001a2546c1/cffi-2.0.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:81afed14892743bbe14dacb9e36d9e0e504cd204e0b165062c488942b9718037", size = 206519, upload-time = "2025-09-08T23:22:51.364Z" },
-    { url = "https://files.pythonhosted.org/packages/78/2d/7fa73dfa841b5ac06c7b8855cfc18622132e365f5b81d02230333ff26e9e/cffi-2.0.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3e17ed538242334bf70832644a32a7aae3d83b57567f9fd60a26257e992b79ba", size = 219572, upload-time = "2025-09-08T23:22:52.902Z" },
-    { url = "https://files.pythonhosted.org/packages/07/e0/267e57e387b4ca276b90f0434ff88b2c2241ad72b16d31836adddfd6031b/cffi-2.0.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3925dd22fa2b7699ed2617149842d2e6adde22b262fcbfada50e3d195e4b3a94", size = 222963, upload-time = "2025-09-08T23:22:54.518Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/75/1f2747525e06f53efbd878f4d03bac5b859cbc11c633d0fb81432d98a795/cffi-2.0.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:2c8f814d84194c9ea681642fd164267891702542f028a15fc97d4674b6206187", size = 221361, upload-time = "2025-09-08T23:22:55.867Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/13/c92e36358fbcc39cf0962e83223c9522154ee8630e1df7c0b3a39a8124e2/cffi-2.0.0-cp39-cp39-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:4647afc2f90d1ddd33441e5b0e85b16b12ddec4fca55f0d9671fef036ecca27c", size = 208813, upload-time = "2025-09-08T23:23:51.263Z" },
-    { url = "https://files.pythonhosted.org/packages/15/12/a7a79bd0df4c3bff744b2d7e52cc1b68d5e7e427b384252c42366dc1ecbc/cffi-2.0.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:3f4d46d8b35698056ec29bca21546e1551a205058ae1a181d871e278b0b28165", size = 216498, upload-time = "2025-09-08T23:23:52.494Z" },
-    { url = "https://files.pythonhosted.org/packages/a3/ad/5c51c1c7600bdd7ed9a24a203ec255dccdd0ebf4527f7b922a0bde2fb6ed/cffi-2.0.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:e6e73b9e02893c764e7e8d5bb5ce277f1a009cd5243f8228f75f842bf937c534", size = 203243, upload-time = "2025-09-08T23:23:53.836Z" },
-    { url = "https://files.pythonhosted.org/packages/32/f2/81b63e288295928739d715d00952c8c6034cb6c6a516b17d37e0c8be5600/cffi-2.0.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:cb527a79772e5ef98fb1d700678fe031e353e765d1ca2d409c92263c6d43e09f", size = 203158, upload-time = "2025-09-08T23:23:55.169Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/74/cc4096ce66f5939042ae094e2e96f53426a979864aa1f96a621ad128be27/cffi-2.0.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:61d028e90346df14fedc3d1e5441df818d095f3b87d286825dfcbd6459b7ef63", size = 216548, upload-time = "2025-09-08T23:23:56.506Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/be/f6424d1dc46b1091ffcc8964fa7c0ab0cd36839dd2761b49c90481a6ba1b/cffi-2.0.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:0f6084a0ea23d05d20c3edcda20c3d006f9b6f3fefeac38f59262e10cef47ee2", size = 218897, upload-time = "2025-09-08T23:23:57.825Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/e0/dda537c2309817edf60109e39265f24f24aa7f050767e22c98c53fe7f48b/cffi-2.0.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:1cd13c99ce269b3ed80b417dcd591415d3372bcac067009b6e0f59c7d4015e65", size = 211249, upload-time = "2025-09-08T23:23:59.139Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/e7/7c769804eb75e4c4b35e658dba01de1640a351a9653c3d49ca89d16ccc91/cffi-2.0.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:89472c9762729b5ae1ad974b777416bfda4ac5642423fa93bd57a09204712322", size = 218041, upload-time = "2025-09-08T23:24:00.496Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/55/51844dd50c4fc7a33b653bfaba4c2456f06955289ca770a5dbd5fd267374/cfgv-3.4.0-py2.py3-none-any.whl", hash = "sha256:b7265b1f29fd3316bfcd2b330d63d024f2bfd8bcb8b0272f8e19a504856c48f9", size = 7249, upload-time = "2023-08-12T20:38:16.269Z" },
 ]
 
 [[package]]
 name = "cfgv"
-version = "3.4.0"
+version = "3.5.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/11/74/539e56497d9bd1d484fd863dd69cbbfa653cd2aa27abfe35653494d85e94/cfgv-3.4.0.tar.gz", hash = "sha256:e52591d4c5f5dead8e0f673fb16db7949d2cfb3f7da4582893288f0ded8fe560", size = 7114, upload-time = "2023-08-12T20:38:17.776Z" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4e/b5/721b8799b04bf9afe054a3899c6cf4e880fcf8563cc71c15610242490a0c/cfgv-3.5.0.tar.gz", hash = "sha256:d5b1034354820651caa73ede66a6294d6e95c1b00acc5e9b098e917404669132", size = 7334, upload-time = "2025-11-19T20:55:51.612Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c5/55/51844dd50c4fc7a33b653bfaba4c2456f06955289ca770a5dbd5fd267374/cfgv-3.4.0-py2.py3-none-any.whl", hash = "sha256:b7265b1f29fd3316bfcd2b330d63d024f2bfd8bcb8b0272f8e19a504856c48f9", size = 7249, upload-time = "2023-08-12T20:38:16.269Z" },
+    { url = "https://files.pythonhosted.org/packages/db/3c/33bac158f8ab7f89b2e59426d5fe2e4f63f7ed25df84c036890172b412b5/cfgv-3.5.0-py2.py3-none-any.whl", hash = "sha256:a8dc6b26ad22ff227d2634a65cb388215ce6cc96bbcc5cfde7641ae87e8dacc0", size = 7445, upload-time = "2025-11-19T20:55:50.744Z" },
 ]
 
 [[package]]
@@ -420,55 +352,75 @@ wheels = [
 
 [[package]]
 name = "charset-normalizer"
-version = "3.4.3"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/83/2d/5fd176ceb9b2fc619e63405525573493ca23441330fcdaee6bef9460e924/charset_normalizer-3.4.3.tar.gz", hash = "sha256:6fce4b8500244f6fcb71465d4a4930d132ba9ab8e71a7859e6a5d59851068d14", size = 122371, upload-time = "2025-08-09T07:57:28.46Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/d6/98/f3b8013223728a99b908c9344da3aa04ee6e3fa235f19409033eda92fb78/charset_normalizer-3.4.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:fb7f67a1bfa6e40b438170ebdc8158b78dc465a5a67b6dde178a46987b244a72", size = 207695, upload-time = "2025-08-09T07:55:36.452Z" },
-    { url = "https://files.pythonhosted.org/packages/21/40/5188be1e3118c82dcb7c2a5ba101b783822cfb413a0268ed3be0468532de/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:cc9370a2da1ac13f0153780040f465839e6cccb4a1e44810124b4e22483c93fe", size = 147153, upload-time = "2025-08-09T07:55:38.467Z" },
-    { url = "https://files.pythonhosted.org/packages/37/60/5d0d74bc1e1380f0b72c327948d9c2aca14b46a9efd87604e724260f384c/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:07a0eae9e2787b586e129fdcbe1af6997f8d0e5abaa0bc98c0e20e124d67e601", size = 160428, upload-time = "2025-08-09T07:55:40.072Z" },
-    { url = "https://files.pythonhosted.org/packages/85/9a/d891f63722d9158688de58d050c59dc3da560ea7f04f4c53e769de5140f5/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:74d77e25adda8581ffc1c720f1c81ca082921329452eba58b16233ab1842141c", size = 157627, upload-time = "2025-08-09T07:55:41.706Z" },
-    { url = "https://files.pythonhosted.org/packages/65/1a/7425c952944a6521a9cfa7e675343f83fd82085b8af2b1373a2409c683dc/charset_normalizer-3.4.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d0e909868420b7049dafd3a31d45125b31143eec59235311fc4c57ea26a4acd2", size = 152388, upload-time = "2025-08-09T07:55:43.262Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/c9/a2c9c2a355a8594ce2446085e2ec97fd44d323c684ff32042e2a6b718e1d/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:c6f162aabe9a91a309510d74eeb6507fab5fff92337a15acbe77753d88d9dcf0", size = 150077, upload-time = "2025-08-09T07:55:44.903Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/38/20a1f44e4851aa1c9105d6e7110c9d020e093dfa5836d712a5f074a12bf7/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:4ca4c094de7771a98d7fbd67d9e5dbf1eb73efa4f744a730437d8a3a5cf994f0", size = 161631, upload-time = "2025-08-09T07:55:46.346Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/fa/384d2c0f57edad03d7bec3ebefb462090d8905b4ff5a2d2525f3bb711fac/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:02425242e96bcf29a49711b0ca9f37e451da7c70562bc10e8ed992a5a7a25cc0", size = 159210, upload-time = "2025-08-09T07:55:47.539Z" },
-    { url = "https://files.pythonhosted.org/packages/33/9e/eca49d35867ca2db336b6ca27617deed4653b97ebf45dfc21311ce473c37/charset_normalizer-3.4.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:78deba4d8f9590fe4dae384aeff04082510a709957e968753ff3c48399f6f92a", size = 153739, upload-time = "2025-08-09T07:55:48.744Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/91/26c3036e62dfe8de8061182d33be5025e2424002125c9500faff74a6735e/charset_normalizer-3.4.3-cp310-cp310-win32.whl", hash = "sha256:d79c198e27580c8e958906f803e63cddb77653731be08851c7df0b1a14a8fc0f", size = 99825, upload-time = "2025-08-09T07:55:50.305Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/c6/f05db471f81af1fa01839d44ae2a8bfeec8d2a8b4590f16c4e7393afd323/charset_normalizer-3.4.3-cp310-cp310-win_amd64.whl", hash = "sha256:c6e490913a46fa054e03699c70019ab869e990270597018cef1d8562132c2669", size = 107452, upload-time = "2025-08-09T07:55:51.461Z" },
-    { url = "https://files.pythonhosted.org/packages/7f/b5/991245018615474a60965a7c9cd2b4efbaabd16d582a5547c47ee1c7730b/charset_normalizer-3.4.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:b256ee2e749283ef3ddcff51a675ff43798d92d746d1a6e4631bf8c707d22d0b", size = 204483, upload-time = "2025-08-09T07:55:53.12Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/2a/ae245c41c06299ec18262825c1569c5d3298fc920e4ddf56ab011b417efd/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:13faeacfe61784e2559e690fc53fa4c5ae97c6fcedb8eb6fb8d0a15b475d2c64", size = 145520, upload-time = "2025-08-09T07:55:54.712Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/a4/b3b6c76e7a635748c4421d2b92c7b8f90a432f98bda5082049af37ffc8e3/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:00237675befef519d9af72169d8604a067d92755e84fe76492fef5441db05b91", size = 158876, upload-time = "2025-08-09T07:55:56.024Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/e6/63bb0e10f90a8243c5def74b5b105b3bbbfb3e7bb753915fe333fb0c11ea/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:585f3b2a80fbd26b048a0be90c5aae8f06605d3c92615911c3a2b03a8a3b796f", size = 156083, upload-time = "2025-08-09T07:55:57.582Z" },
-    { url = "https://files.pythonhosted.org/packages/87/df/b7737ff046c974b183ea9aa111b74185ac8c3a326c6262d413bd5a1b8c69/charset_normalizer-3.4.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0e78314bdc32fa80696f72fa16dc61168fda4d6a0c014e0380f9d02f0e5d8a07", size = 150295, upload-time = "2025-08-09T07:55:59.147Z" },
-    { url = "https://files.pythonhosted.org/packages/61/f1/190d9977e0084d3f1dc169acd060d479bbbc71b90bf3e7bf7b9927dec3eb/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:96b2b3d1a83ad55310de8c7b4a2d04d9277d5591f40761274856635acc5fcb30", size = 148379, upload-time = "2025-08-09T07:56:00.364Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/92/27dbe365d34c68cfe0ca76f1edd70e8705d82b378cb54ebbaeabc2e3029d/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:939578d9d8fd4299220161fdd76e86c6a251987476f5243e8864a7844476ba14", size = 160018, upload-time = "2025-08-09T07:56:01.678Z" },
-    { url = "https://files.pythonhosted.org/packages/99/04/baae2a1ea1893a01635d475b9261c889a18fd48393634b6270827869fa34/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:fd10de089bcdcd1be95a2f73dbe6254798ec1bda9f450d5828c96f93e2536b9c", size = 157430, upload-time = "2025-08-09T07:56:02.87Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/36/77da9c6a328c54d17b960c89eccacfab8271fdaaa228305330915b88afa9/charset_normalizer-3.4.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:1e8ac75d72fa3775e0b7cb7e4629cec13b7514d928d15ef8ea06bca03ef01cae", size = 151600, upload-time = "2025-08-09T07:56:04.089Z" },
-    { url = "https://files.pythonhosted.org/packages/64/d4/9eb4ff2c167edbbf08cdd28e19078bf195762e9bd63371689cab5ecd3d0d/charset_normalizer-3.4.3-cp311-cp311-win32.whl", hash = "sha256:6cf8fd4c04756b6b60146d98cd8a77d0cdae0e1ca20329da2ac85eed779b6849", size = 99616, upload-time = "2025-08-09T07:56:05.658Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/9c/996a4a028222e7761a96634d1820de8a744ff4327a00ada9c8942033089b/charset_normalizer-3.4.3-cp311-cp311-win_amd64.whl", hash = "sha256:31a9a6f775f9bcd865d88ee350f0ffb0e25936a7f930ca98995c05abf1faf21c", size = 107108, upload-time = "2025-08-09T07:56:07.176Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/5e/14c94999e418d9b87682734589404a25854d5f5d0408df68bc15b6ff54bb/charset_normalizer-3.4.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:e28e334d3ff134e88989d90ba04b47d84382a828c061d0d1027b1b12a62b39b1", size = 205655, upload-time = "2025-08-09T07:56:08.475Z" },
-    { url = "https://files.pythonhosted.org/packages/7d/a8/c6ec5d389672521f644505a257f50544c074cf5fc292d5390331cd6fc9c3/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0cacf8f7297b0c4fcb74227692ca46b4a5852f8f4f24b3c766dd94a1075c4884", size = 146223, upload-time = "2025-08-09T07:56:09.708Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/eb/a2ffb08547f4e1e5415fb69eb7db25932c52a52bed371429648db4d84fb1/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c6fd51128a41297f5409deab284fecbe5305ebd7e5a1f959bee1c054622b7018", size = 159366, upload-time = "2025-08-09T07:56:11.326Z" },
-    { url = "https://files.pythonhosted.org/packages/82/10/0fd19f20c624b278dddaf83b8464dcddc2456cb4b02bb902a6da126b87a1/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:3cfb2aad70f2c6debfbcb717f23b7eb55febc0bb23dcffc0f076009da10c6392", size = 157104, upload-time = "2025-08-09T07:56:13.014Z" },
-    { url = "https://files.pythonhosted.org/packages/16/ab/0233c3231af734f5dfcf0844aa9582d5a1466c985bbed6cedab85af9bfe3/charset_normalizer-3.4.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1606f4a55c0fd363d754049cdf400175ee96c992b1f8018b993941f221221c5f", size = 151830, upload-time = "2025-08-09T07:56:14.428Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/02/e29e22b4e02839a0e4a06557b1999d0a47db3567e82989b5bb21f3fbbd9f/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:027b776c26d38b7f15b26a5da1044f376455fb3766df8fc38563b4efbc515154", size = 148854, upload-time = "2025-08-09T07:56:16.051Z" },
-    { url = "https://files.pythonhosted.org/packages/05/6b/e2539a0a4be302b481e8cafb5af8792da8093b486885a1ae4d15d452bcec/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:42e5088973e56e31e4fa58eb6bd709e42fc03799c11c42929592889a2e54c491", size = 160670, upload-time = "2025-08-09T07:56:17.314Z" },
-    { url = "https://files.pythonhosted.org/packages/31/e7/883ee5676a2ef217a40ce0bffcc3d0dfbf9e64cbcfbdf822c52981c3304b/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:cc34f233c9e71701040d772aa7490318673aa7164a0efe3172b2981218c26d93", size = 158501, upload-time = "2025-08-09T07:56:18.641Z" },
-    { url = "https://files.pythonhosted.org/packages/c1/35/6525b21aa0db614cf8b5792d232021dca3df7f90a1944db934efa5d20bb1/charset_normalizer-3.4.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:320e8e66157cc4e247d9ddca8e21f427efc7a04bbd0ac8a9faf56583fa543f9f", size = 153173, upload-time = "2025-08-09T07:56:20.289Z" },
-    { url = "https://files.pythonhosted.org/packages/50/ee/f4704bad8201de513fdc8aac1cabc87e38c5818c93857140e06e772b5892/charset_normalizer-3.4.3-cp312-cp312-win32.whl", hash = "sha256:fb6fecfd65564f208cbf0fba07f107fb661bcd1a7c389edbced3f7a493f70e37", size = 99822, upload-time = "2025-08-09T07:56:21.551Z" },
-    { url = "https://files.pythonhosted.org/packages/39/f5/3b3836ca6064d0992c58c7561c6b6eee1b3892e9665d650c803bd5614522/charset_normalizer-3.4.3-cp312-cp312-win_amd64.whl", hash = "sha256:86df271bf921c2ee3818f0522e9a5b8092ca2ad8b065ece5d7d9d0e9f4849bcc", size = 107543, upload-time = "2025-08-09T07:56:23.115Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/ca/9a0983dd5c8e9733565cf3db4df2b0a2e9a82659fd8aa2a868ac6e4a991f/charset_normalizer-3.4.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:70bfc5f2c318afece2f5838ea5e4c3febada0be750fcf4775641052bbba14d05", size = 207520, upload-time = "2025-08-09T07:57:11.026Z" },
-    { url = "https://files.pythonhosted.org/packages/39/c6/99271dc37243a4f925b09090493fb96c9333d7992c6187f5cfe5312008d2/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:23b6b24d74478dc833444cbd927c338349d6ae852ba53a0d02a2de1fce45b96e", size = 147307, upload-time = "2025-08-09T07:57:12.4Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/69/132eab043356bba06eb333cc2cc60c6340857d0a2e4ca6dc2b51312886b3/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:34a7f768e3f985abdb42841e20e17b330ad3aaf4bb7e7aeeb73db2e70f077b99", size = 160448, upload-time = "2025-08-09T07:57:13.712Z" },
-    { url = "https://files.pythonhosted.org/packages/04/9a/914d294daa4809c57667b77470533e65def9c0be1ef8b4c1183a99170e9d/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:fb731e5deb0c7ef82d698b0f4c5bb724633ee2a489401594c5c88b02e6cb15f7", size = 157758, upload-time = "2025-08-09T07:57:14.979Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/a8/6f5bcf1bcf63cb45625f7c5cadca026121ff8a6c8a3256d8d8cd59302663/charset_normalizer-3.4.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:257f26fed7d7ff59921b78244f3cd93ed2af1800ff048c33f624c87475819dd7", size = 152487, upload-time = "2025-08-09T07:57:16.332Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/72/d3d0e9592f4e504f9dea08b8db270821c909558c353dc3b457ed2509f2fb/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:1ef99f0456d3d46a50945c98de1774da86f8e992ab5c77865ea8b8195341fc19", size = 150054, upload-time = "2025-08-09T07:57:17.576Z" },
-    { url = "https://files.pythonhosted.org/packages/20/30/5f64fe3981677fe63fa987b80e6c01042eb5ff653ff7cec1b7bd9268e54e/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:2c322db9c8c89009a990ef07c3bcc9f011a3269bc06782f916cd3d9eed7c9312", size = 161703, upload-time = "2025-08-09T07:57:20.012Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/ef/dd08b2cac9284fd59e70f7d97382c33a3d0a926e45b15fc21b3308324ffd/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:511729f456829ef86ac41ca78c63a5cb55240ed23b4b737faca0eb1abb1c41bc", size = 159096, upload-time = "2025-08-09T07:57:21.329Z" },
-    { url = "https://files.pythonhosted.org/packages/45/8c/dcef87cfc2b3f002a6478f38906f9040302c68aebe21468090e39cde1445/charset_normalizer-3.4.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:88ab34806dea0671532d3f82d82b85e8fc23d7b2dd12fa837978dad9bb392a34", size = 153852, upload-time = "2025-08-09T07:57:22.608Z" },
-    { url = "https://files.pythonhosted.org/packages/63/86/9cbd533bd37883d467fcd1bd491b3547a3532d0fbb46de2b99feeebf185e/charset_normalizer-3.4.3-cp39-cp39-win32.whl", hash = "sha256:16a8770207946ac75703458e2c743631c79c59c5890c80011d536248f8eaa432", size = 99840, upload-time = "2025-08-09T07:57:23.883Z" },
-    { url = "https://files.pythonhosted.org/packages/ce/d6/7e805c8e5c46ff9729c49950acc4ee0aeb55efb8b3a56687658ad10c3216/charset_normalizer-3.4.3-cp39-cp39-win_amd64.whl", hash = "sha256:d22dbedd33326a4a5190dd4fe9e9e693ef12160c77382d9e87919bce54f3d4ca", size = 107438, upload-time = "2025-08-09T07:57:25.287Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/1f/f041989e93b001bc4e44bb1669ccdcf54d3f00e628229a85b08d330615c5/charset_normalizer-3.4.3-py3-none-any.whl", hash = "sha256:ce571ab16d890d23b5c278547ba694193a45011ff86a9162a71307ed9f86759a", size = 53175, upload-time = "2025-08-09T07:57:26.864Z" },
+version = "3.4.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/13/69/33ddede1939fdd074bce5434295f38fae7136463422fe4fd3e0e89b98062/charset_normalizer-3.4.4.tar.gz", hash = "sha256:94537985111c35f28720e43603b8e7b43a6ecfb2ce1d3058bbe955b73404e21a", size = 129418, upload-time = "2025-10-14T04:42:32.879Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/1f/b8/6d51fc1d52cbd52cd4ccedd5b5b2f0f6a11bbf6765c782298b0f3e808541/charset_normalizer-3.4.4-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:e824f1492727fa856dd6eda4f7cee25f8518a12f3c4a56a74e8095695089cf6d", size = 209709, upload-time = "2025-10-14T04:40:11.385Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/af/1f9d7f7faafe2ddfb6f72a2e07a548a629c61ad510fe60f9630309908fef/charset_normalizer-3.4.4-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4bd5d4137d500351a30687c2d3971758aac9a19208fc110ccb9d7188fbe709e8", size = 148814, upload-time = "2025-10-14T04:40:13.135Z" },
+    { url = "https://files.pythonhosted.org/packages/79/3d/f2e3ac2bbc056ca0c204298ea4e3d9db9b4afe437812638759db2c976b5f/charset_normalizer-3.4.4-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:027f6de494925c0ab2a55eab46ae5129951638a49a34d87f4c3eda90f696b4ad", size = 144467, upload-time = "2025-10-14T04:40:14.728Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/85/1bf997003815e60d57de7bd972c57dc6950446a3e4ccac43bc3070721856/charset_normalizer-3.4.4-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f820802628d2694cb7e56db99213f930856014862f3fd943d290ea8438d07ca8", size = 162280, upload-time = "2025-10-14T04:40:16.14Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/8e/6aa1952f56b192f54921c436b87f2aaf7c7a7c3d0d1a765547d64fd83c13/charset_normalizer-3.4.4-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:798d75d81754988d2565bff1b97ba5a44411867c0cf32b77a7e8f8d84796b10d", size = 159454, upload-time = "2025-10-14T04:40:17.567Z" },
+    { url = "https://files.pythonhosted.org/packages/36/3b/60cbd1f8e93aa25d1c669c649b7a655b0b5fb4c571858910ea9332678558/charset_normalizer-3.4.4-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9d1bb833febdff5c8927f922386db610b49db6e0d4f4ee29601d71e7c2694313", size = 153609, upload-time = "2025-10-14T04:40:19.08Z" },
+    { url = "https://files.pythonhosted.org/packages/64/91/6a13396948b8fd3c4b4fd5bc74d045f5637d78c9675585e8e9fbe5636554/charset_normalizer-3.4.4-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:9cd98cdc06614a2f768d2b7286d66805f94c48cde050acdbbb7db2600ab3197e", size = 151849, upload-time = "2025-10-14T04:40:20.607Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/7a/59482e28b9981d105691e968c544cc0df3b7d6133152fb3dcdc8f135da7a/charset_normalizer-3.4.4-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:077fbb858e903c73f6c9db43374fd213b0b6a778106bc7032446a8e8b5b38b93", size = 151586, upload-time = "2025-10-14T04:40:21.719Z" },
+    { url = "https://files.pythonhosted.org/packages/92/59/f64ef6a1c4bdd2baf892b04cd78792ed8684fbc48d4c2afe467d96b4df57/charset_normalizer-3.4.4-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:244bfb999c71b35de57821b8ea746b24e863398194a4014e4c76adc2bbdfeff0", size = 145290, upload-time = "2025-10-14T04:40:23.069Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/63/3bf9f279ddfa641ffa1962b0db6a57a9c294361cc2f5fcac997049a00e9c/charset_normalizer-3.4.4-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:64b55f9dce520635f018f907ff1b0df1fdc31f2795a922fb49dd14fbcdf48c84", size = 163663, upload-time = "2025-10-14T04:40:24.17Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/09/c9e38fc8fa9e0849b172b581fd9803bdf6e694041127933934184e19f8c3/charset_normalizer-3.4.4-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:faa3a41b2b66b6e50f84ae4a68c64fcd0c44355741c6374813a800cd6695db9e", size = 151964, upload-time = "2025-10-14T04:40:25.368Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/d1/d28b747e512d0da79d8b6a1ac18b7ab2ecfd81b2944c4c710e166d8dd09c/charset_normalizer-3.4.4-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:6515f3182dbe4ea06ced2d9e8666d97b46ef4c75e326b79bb624110f122551db", size = 161064, upload-time = "2025-10-14T04:40:26.806Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/9a/31d62b611d901c3b9e5500c36aab0ff5eb442043fb3a1c254200d3d397d9/charset_normalizer-3.4.4-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:cc00f04ed596e9dc0da42ed17ac5e596c6ccba999ba6bd92b0e0aef2f170f2d6", size = 155015, upload-time = "2025-10-14T04:40:28.284Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/f3/107e008fa2bff0c8b9319584174418e5e5285fef32f79d8ee6a430d0039c/charset_normalizer-3.4.4-cp310-cp310-win32.whl", hash = "sha256:f34be2938726fc13801220747472850852fe6b1ea75869a048d6f896838c896f", size = 99792, upload-time = "2025-10-14T04:40:29.613Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/66/e396e8a408843337d7315bab30dbf106c38966f1819f123257f5520f8a96/charset_normalizer-3.4.4-cp310-cp310-win_amd64.whl", hash = "sha256:a61900df84c667873b292c3de315a786dd8dac506704dea57bc957bd31e22c7d", size = 107198, upload-time = "2025-10-14T04:40:30.644Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/58/01b4f815bf0312704c267f2ccb6e5d42bcc7752340cd487bc9f8c3710597/charset_normalizer-3.4.4-cp310-cp310-win_arm64.whl", hash = "sha256:cead0978fc57397645f12578bfd2d5ea9138ea0fac82b2f63f7f7c6877986a69", size = 100262, upload-time = "2025-10-14T04:40:32.108Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/27/c6491ff4954e58a10f69ad90aca8a1b6fe9c5d3c6f380907af3c37435b59/charset_normalizer-3.4.4-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6e1fcf0720908f200cd21aa4e6750a48ff6ce4afe7ff5a79a90d5ed8a08296f8", size = 206988, upload-time = "2025-10-14T04:40:33.79Z" },
+    { url = "https://files.pythonhosted.org/packages/94/59/2e87300fe67ab820b5428580a53cad894272dbb97f38a7a814a2a1ac1011/charset_normalizer-3.4.4-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5f819d5fe9234f9f82d75bdfa9aef3a3d72c4d24a6e57aeaebba32a704553aa0", size = 147324, upload-time = "2025-10-14T04:40:34.961Z" },
+    { url = "https://files.pythonhosted.org/packages/07/fb/0cf61dc84b2b088391830f6274cb57c82e4da8bbc2efeac8c025edb88772/charset_normalizer-3.4.4-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:a59cb51917aa591b1c4e6a43c132f0cdc3c76dbad6155df4e28ee626cc77a0a3", size = 142742, upload-time = "2025-10-14T04:40:36.105Z" },
+    { url = "https://files.pythonhosted.org/packages/62/8b/171935adf2312cd745d290ed93cf16cf0dfe320863ab7cbeeae1dcd6535f/charset_normalizer-3.4.4-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8ef3c867360f88ac904fd3f5e1f902f13307af9052646963ee08ff4f131adafc", size = 160863, upload-time = "2025-10-14T04:40:37.188Z" },
+    { url = "https://files.pythonhosted.org/packages/09/73/ad875b192bda14f2173bfc1bc9a55e009808484a4b256748d931b6948442/charset_normalizer-3.4.4-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:d9e45d7faa48ee908174d8fe84854479ef838fc6a705c9315372eacbc2f02897", size = 157837, upload-time = "2025-10-14T04:40:38.435Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/fc/de9cce525b2c5b94b47c70a4b4fb19f871b24995c728e957ee68ab1671ea/charset_normalizer-3.4.4-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:840c25fb618a231545cbab0564a799f101b63b9901f2569faecd6b222ac72381", size = 151550, upload-time = "2025-10-14T04:40:40.053Z" },
+    { url = "https://files.pythonhosted.org/packages/55/c2/43edd615fdfba8c6f2dfbd459b25a6b3b551f24ea21981e23fb768503ce1/charset_normalizer-3.4.4-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:ca5862d5b3928c4940729dacc329aa9102900382fea192fc5e52eb69d6093815", size = 149162, upload-time = "2025-10-14T04:40:41.163Z" },
+    { url = "https://files.pythonhosted.org/packages/03/86/bde4ad8b4d0e9429a4e82c1e8f5c659993a9a863ad62c7df05cf7b678d75/charset_normalizer-3.4.4-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d9c7f57c3d666a53421049053eaacdd14bbd0a528e2186fcb2e672effd053bb0", size = 150019, upload-time = "2025-10-14T04:40:42.276Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/86/a151eb2af293a7e7bac3a739b81072585ce36ccfb4493039f49f1d3cae8c/charset_normalizer-3.4.4-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:277e970e750505ed74c832b4bf75dac7476262ee2a013f5574dd49075879e161", size = 143310, upload-time = "2025-10-14T04:40:43.439Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/fe/43dae6144a7e07b87478fdfc4dbe9efd5defb0e7ec29f5f58a55aeef7bf7/charset_normalizer-3.4.4-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:31fd66405eaf47bb62e8cd575dc621c56c668f27d46a61d975a249930dd5e2a4", size = 162022, upload-time = "2025-10-14T04:40:44.547Z" },
+    { url = "https://files.pythonhosted.org/packages/80/e6/7aab83774f5d2bca81f42ac58d04caf44f0cc2b65fc6db2b3b2e8a05f3b3/charset_normalizer-3.4.4-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:0d3d8f15c07f86e9ff82319b3d9ef6f4bf907608f53fe9d92b28ea9ae3d1fd89", size = 149383, upload-time = "2025-10-14T04:40:46.018Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/e8/b289173b4edae05c0dde07f69f8db476a0b511eac556dfe0d6bda3c43384/charset_normalizer-3.4.4-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:9f7fcd74d410a36883701fafa2482a6af2ff5ba96b9a620e9e0721e28ead5569", size = 159098, upload-time = "2025-10-14T04:40:47.081Z" },
+    { url = "https://files.pythonhosted.org/packages/d8/df/fe699727754cae3f8478493c7f45f777b17c3ef0600e28abfec8619eb49c/charset_normalizer-3.4.4-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ebf3e58c7ec8a8bed6d66a75d7fb37b55e5015b03ceae72a8e7c74495551e224", size = 152991, upload-time = "2025-10-14T04:40:48.246Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/86/584869fe4ddb6ffa3bd9f491b87a01568797fb9bd8933f557dba9771beaf/charset_normalizer-3.4.4-cp311-cp311-win32.whl", hash = "sha256:eecbc200c7fd5ddb9a7f16c7decb07b566c29fa2161a16cf67b8d068bd21690a", size = 99456, upload-time = "2025-10-14T04:40:49.376Z" },
+    { url = "https://files.pythonhosted.org/packages/65/f6/62fdd5feb60530f50f7e38b4f6a1d5203f4d16ff4f9f0952962c044e919a/charset_normalizer-3.4.4-cp311-cp311-win_amd64.whl", hash = "sha256:5ae497466c7901d54b639cf42d5b8c1b6a4fead55215500d2f486d34db48d016", size = 106978, upload-time = "2025-10-14T04:40:50.844Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/9d/0710916e6c82948b3be62d9d398cb4fcf4e97b56d6a6aeccd66c4b2f2bd5/charset_normalizer-3.4.4-cp311-cp311-win_arm64.whl", hash = "sha256:65e2befcd84bc6f37095f5961e68a6f077bf44946771354a28ad434c2cce0ae1", size = 99969, upload-time = "2025-10-14T04:40:52.272Z" },
+    { url = "https://files.pythonhosted.org/packages/f3/85/1637cd4af66fa687396e757dec650f28025f2a2f5a5531a3208dc0ec43f2/charset_normalizer-3.4.4-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:0a98e6759f854bd25a58a73fa88833fba3b7c491169f86ce1180c948ab3fd394", size = 208425, upload-time = "2025-10-14T04:40:53.353Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/6a/04130023fef2a0d9c62d0bae2649b69f7b7d8d24ea5536feef50551029df/charset_normalizer-3.4.4-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b5b290ccc2a263e8d185130284f8501e3e36c5e02750fc6b6bdeb2e9e96f1e25", size = 148162, upload-time = "2025-10-14T04:40:54.558Z" },
+    { url = "https://files.pythonhosted.org/packages/78/29/62328d79aa60da22c9e0b9a66539feae06ca0f5a4171ac4f7dc285b83688/charset_normalizer-3.4.4-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:74bb723680f9f7a6234dcf67aea57e708ec1fbdf5699fb91dfd6f511b0a320ef", size = 144558, upload-time = "2025-10-14T04:40:55.677Z" },
+    { url = "https://files.pythonhosted.org/packages/86/bb/b32194a4bf15b88403537c2e120b817c61cd4ecffa9b6876e941c3ee38fe/charset_normalizer-3.4.4-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f1e34719c6ed0b92f418c7c780480b26b5d9c50349e9a9af7d76bf757530350d", size = 161497, upload-time = "2025-10-14T04:40:57.217Z" },
+    { url = "https://files.pythonhosted.org/packages/19/89/a54c82b253d5b9b111dc74aca196ba5ccfcca8242d0fb64146d4d3183ff1/charset_normalizer-3.4.4-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:2437418e20515acec67d86e12bf70056a33abdacb5cb1655042f6538d6b085a8", size = 159240, upload-time = "2025-10-14T04:40:58.358Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/10/d20b513afe03acc89ec33948320a5544d31f21b05368436d580dec4e234d/charset_normalizer-3.4.4-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:11d694519d7f29d6cd09f6ac70028dba10f92f6cdd059096db198c283794ac86", size = 153471, upload-time = "2025-10-14T04:40:59.468Z" },
+    { url = "https://files.pythonhosted.org/packages/61/fa/fbf177b55bdd727010f9c0a3c49eefa1d10f960e5f09d1d887bf93c2e698/charset_normalizer-3.4.4-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:ac1c4a689edcc530fc9d9aa11f5774b9e2f33f9a0c6a57864e90908f5208d30a", size = 150864, upload-time = "2025-10-14T04:41:00.623Z" },
+    { url = "https://files.pythonhosted.org/packages/05/12/9fbc6a4d39c0198adeebbde20b619790e9236557ca59fc40e0e3cebe6f40/charset_normalizer-3.4.4-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:21d142cc6c0ec30d2efee5068ca36c128a30b0f2c53c1c07bd78cb6bc1d3be5f", size = 150647, upload-time = "2025-10-14T04:41:01.754Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/1f/6a9a593d52e3e8c5d2b167daf8c6b968808efb57ef4c210acb907c365bc4/charset_normalizer-3.4.4-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:5dbe56a36425d26d6cfb40ce79c314a2e4dd6211d51d6d2191c00bed34f354cc", size = 145110, upload-time = "2025-10-14T04:41:03.231Z" },
+    { url = "https://files.pythonhosted.org/packages/30/42/9a52c609e72471b0fc54386dc63c3781a387bb4fe61c20231a4ebcd58bdd/charset_normalizer-3.4.4-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:5bfbb1b9acf3334612667b61bd3002196fe2a1eb4dd74d247e0f2a4d50ec9bbf", size = 162839, upload-time = "2025-10-14T04:41:04.715Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/5b/c0682bbf9f11597073052628ddd38344a3d673fda35a36773f7d19344b23/charset_normalizer-3.4.4-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:d055ec1e26e441f6187acf818b73564e6e6282709e9bcb5b63f5b23068356a15", size = 150667, upload-time = "2025-10-14T04:41:05.827Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/24/a41afeab6f990cf2daf6cb8c67419b63b48cf518e4f56022230840c9bfb2/charset_normalizer-3.4.4-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:af2d8c67d8e573d6de5bc30cdb27e9b95e49115cd9baad5ddbd1a6207aaa82a9", size = 160535, upload-time = "2025-10-14T04:41:06.938Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/e5/6a4ce77ed243c4a50a1fecca6aaaab419628c818a49434be428fe24c9957/charset_normalizer-3.4.4-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:780236ac706e66881f3b7f2f32dfe90507a09e67d1d454c762cf642e6e1586e0", size = 154816, upload-time = "2025-10-14T04:41:08.101Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/ef/89297262b8092b312d29cdb2517cb1237e51db8ecef2e9af5edbe7b683b1/charset_normalizer-3.4.4-cp312-cp312-win32.whl", hash = "sha256:5833d2c39d8896e4e19b689ffc198f08ea58116bee26dea51e362ecc7cd3ed26", size = 99694, upload-time = "2025-10-14T04:41:09.23Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/2d/1e5ed9dd3b3803994c155cd9aacb60c82c331bad84daf75bcb9c91b3295e/charset_normalizer-3.4.4-cp312-cp312-win_amd64.whl", hash = "sha256:a79cfe37875f822425b89a82333404539ae63dbdddf97f84dcbc3d339aae9525", size = 107131, upload-time = "2025-10-14T04:41:10.467Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/d9/0ed4c7098a861482a7b6a95603edce4c0d9db2311af23da1fb2b75ec26fc/charset_normalizer-3.4.4-cp312-cp312-win_arm64.whl", hash = "sha256:376bec83a63b8021bb5c8ea75e21c4ccb86e7e45ca4eb81146091b56599b80c3", size = 100390, upload-time = "2025-10-14T04:41:11.915Z" },
+    { url = "https://files.pythonhosted.org/packages/46/7c/0c4760bccf082737ca7ab84a4c2034fcc06b1f21cf3032ea98bd6feb1725/charset_normalizer-3.4.4-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:a9768c477b9d7bd54bc0c86dbaebdec6f03306675526c9927c0e8a04e8f94af9", size = 209609, upload-time = "2025-10-14T04:42:10.922Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/a4/69719daef2f3d7f1819de60c9a6be981b8eeead7542d5ec4440f3c80e111/charset_normalizer-3.4.4-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1bee1e43c28aa63cb16e5c14e582580546b08e535299b8b6158a7c9c768a1f3d", size = 149029, upload-time = "2025-10-14T04:42:12.38Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/21/8d4e1d6c1e6070d3672908b8e4533a71b5b53e71d16828cc24d0efec564c/charset_normalizer-3.4.4-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:fd44c878ea55ba351104cb93cc85e74916eb8fa440ca7903e57575e97394f608", size = 144580, upload-time = "2025-10-14T04:42:13.549Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/0a/a616d001b3f25647a9068e0b9199f697ce507ec898cacb06a0d5a1617c99/charset_normalizer-3.4.4-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:0f04b14ffe5fdc8c4933862d8306109a2c51e0704acfa35d51598eb45a1e89fc", size = 162340, upload-time = "2025-10-14T04:42:14.892Z" },
+    { url = "https://files.pythonhosted.org/packages/85/93/060b52deb249a5450460e0585c88a904a83aec474ab8e7aba787f45e79f2/charset_normalizer-3.4.4-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:cd09d08005f958f370f539f186d10aec3377d55b9eeb0d796025d4886119d76e", size = 159619, upload-time = "2025-10-14T04:42:16.676Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/21/0274deb1cc0632cd587a9a0ec6b4674d9108e461cb4cd40d457adaeb0564/charset_normalizer-3.4.4-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4fe7859a4e3e8457458e2ff592f15ccb02f3da787fcd31e0183879c3ad4692a1", size = 153980, upload-time = "2025-10-14T04:42:17.917Z" },
+    { url = "https://files.pythonhosted.org/packages/28/2b/e3d7d982858dccc11b31906976323d790dded2017a0572f093ff982d692f/charset_normalizer-3.4.4-cp39-cp39-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:fa09f53c465e532f4d3db095e0c55b615f010ad81803d383195b6b5ca6cbf5f3", size = 152174, upload-time = "2025-10-14T04:42:19.018Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/ff/4a269f8e35f1e58b2df52c131a1fa019acb7ef3f8697b7d464b07e9b492d/charset_normalizer-3.4.4-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:7fa17817dc5625de8a027cb8b26d9fefa3ea28c8253929b8d6649e705d2835b6", size = 151666, upload-time = "2025-10-14T04:42:20.171Z" },
+    { url = "https://files.pythonhosted.org/packages/da/c9/ec39870f0b330d58486001dd8e532c6b9a905f5765f58a6f8204926b4a93/charset_normalizer-3.4.4-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:5947809c8a2417be3267efc979c47d76a079758166f7d43ef5ae8e9f92751f88", size = 145550, upload-time = "2025-10-14T04:42:21.324Z" },
+    { url = "https://files.pythonhosted.org/packages/75/8f/d186ab99e40e0ed9f82f033d6e49001701c81244d01905dd4a6924191a30/charset_normalizer-3.4.4-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:4902828217069c3c5c71094537a8e623f5d097858ac6ca8252f7b4d10b7560f1", size = 163721, upload-time = "2025-10-14T04:42:22.46Z" },
+    { url = "https://files.pythonhosted.org/packages/96/b1/6047663b9744df26a7e479ac1e77af7134b1fcf9026243bb48ee2d18810f/charset_normalizer-3.4.4-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:7c308f7e26e4363d79df40ca5b2be1c6ba9f02bdbccfed5abddb7859a6ce72cf", size = 152127, upload-time = "2025-10-14T04:42:23.712Z" },
+    { url = "https://files.pythonhosted.org/packages/59/78/e5a6eac9179f24f704d1be67d08704c3c6ab9f00963963524be27c18ed87/charset_normalizer-3.4.4-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:2c9d3c380143a1fedbff95a312aa798578371eb29da42106a29019368a475318", size = 161175, upload-time = "2025-10-14T04:42:24.87Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/43/0e626e42d54dd2f8dd6fc5e1c5ff00f05fbca17cb699bedead2cae69c62f/charset_normalizer-3.4.4-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:cb01158d8b88ee68f15949894ccc6712278243d95f344770fa7593fa2d94410c", size = 155375, upload-time = "2025-10-14T04:42:27.246Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/91/d9615bf2e06f35e4997616ff31248c3657ed649c5ab9d35ea12fce54e380/charset_normalizer-3.4.4-cp39-cp39-win32.whl", hash = "sha256:2677acec1a2f8ef614c6888b5b4ae4060cc184174a938ed4e8ef690e15d3e505", size = 99692, upload-time = "2025-10-14T04:42:28.425Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/a9/6c040053909d9d1ef4fcab45fddec083aedc9052c10078339b47c8573ea8/charset_normalizer-3.4.4-cp39-cp39-win_amd64.whl", hash = "sha256:f8e160feb2aed042cd657a72acc0b481212ed28b1b9a95c0cee1621b524e1966", size = 107192, upload-time = "2025-10-14T04:42:29.482Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/c6/4fa536b2c0cd3edfb7ccf8469fa0f363ea67b7213a842b90909ca33dd851/charset_normalizer-3.4.4-cp39-cp39-win_arm64.whl", hash = "sha256:b5d84d37db046c5ca74ee7bb47dd6cbc13f80665fdde3e8040bdd3fb015ecb50", size = 100220, upload-time = "2025-10-14T04:42:30.632Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/4c/925909008ed5a988ccbb72dcc897407e5d6d3bd72410d69e051fc0c14647/charset_normalizer-3.4.4-py3-none-any.whl", hash = "sha256:7a32c560861a02ff789ad905a2fe94e3f840803362c84fecf1851cb4cf3dc37f", size = 53402, upload-time = "2025-10-14T04:42:31.76Z" },
 ]
 
 [[package]]
@@ -505,8 +457,8 @@ resolution-markers = [
 ]
 dependencies = [
     { name = "absl-py", marker = "python_full_version >= '3.11'" },
-    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "jaxlib", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "toolz", marker = "python_full_version >= '3.11'" },
     { name = "typing-extensions", marker = "python_full_version >= '3.11'" },
@@ -533,7 +485,7 @@ wheels = [
 
 [[package]]
 name = "click"
-version = "8.3.0"
+version = "8.2.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -543,18 +495,18 @@ resolution-markers = [
 dependencies = [
     { name = "colorama", marker = "python_full_version >= '3.10' and sys_platform == 'win32'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/46/61/de6cd827efad202d7057d93e0fed9294b96952e188f7384832791c7b2254/click-8.3.0.tar.gz", hash = "sha256:e7b8232224eba16f4ebe410c25ced9f7875cb5f3263ffc93cc3e8da705e229c4", size = 276943, upload-time = "2025-09-18T17:32:23.696Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/60/6c/8ca2efa64cf75a977a0d7fac081354553ebe483345c734fb6b6515d96bbc/click-8.2.1.tar.gz", hash = "sha256:27c491cc05d968d271d5a1db13e3b5a184636d9d930f148c50b038f0d0646202", size = 286342, upload-time = "2025-05-20T23:19:49.832Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/db/d3/9dcc0f5797f070ec8edf30fbadfb200e71d9db6b84d211e3b2085a7589a0/click-8.3.0-py3-none-any.whl", hash = "sha256:9b9f285302c6e3064f4330c05f05b81945b2a39544279343e6e7c5f27a9baddc", size = 107295, upload-time = "2025-09-18T17:32:22.42Z" },
+    { url = "https://files.pythonhosted.org/packages/85/32/10bb5764d90a8eee674e9dc6f4db6a0ab47c8c4d0d83c27f7c39ac415a4d/click-8.2.1-py3-none-any.whl", hash = "sha256:61a3265b914e850b85317d0b3109c7f8cd35a670f963866005d6ef1d5175a12b", size = 102215, upload-time = "2025-05-20T23:19:47.796Z" },
 ]
 
 [[package]]
 name = "cloudpickle"
-version = "3.1.1"
+version = "3.1.2"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/52/39/069100b84d7418bc358d81669d5748efb14b9cceacd2f9c75f550424132f/cloudpickle-3.1.1.tar.gz", hash = "sha256:b216fa8ae4019d5482a8ac3c95d8f6346115d8835911fd4aefd1a445e4242c64", size = 22113, upload-time = "2025-01-14T17:02:05.085Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/27/fb/576f067976d320f5f0114a8d9fa1215425441bb35627b1993e5afd8111e5/cloudpickle-3.1.2.tar.gz", hash = "sha256:7fda9eb655c9c230dab534f1983763de5835249750e85fbcef43aaa30a9a2414", size = 22330, upload-time = "2025-11-03T09:25:26.604Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/7e/e8/64c37fadfc2816a7701fa8a6ed8d87327c7d54eacfbfb6edab14a2f2be75/cloudpickle-3.1.1-py3-none-any.whl", hash = "sha256:c8c5a44295039331ee9dad40ba100a9c7297b6f988e50e87ccdf3765a668350e", size = 20992, upload-time = "2025-01-14T17:02:02.417Z" },
+    { url = "https://files.pythonhosted.org/packages/88/39/799be3f2f0f38cc727ee3b4f1445fe6d5e4133064ec2e4115069418a5bb6/cloudpickle-3.1.2-py3-none-any.whl", hash = "sha256:9acb47f6afd73f60dc1df93bb801b472f05ff42fa6c84167d25cb206be1fbf4a", size = 22228, upload-time = "2025-11-03T09:25:25.534Z" },
 ]
 
 [[package]]
@@ -566,25 +518,13 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6", size = 25335, upload-time = "2022-10-25T02:36:20.889Z" },
 ]
 
-[[package]]
-name = "compressed-tensors"
-version = "0.11.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "frozendict" },
-    { name = "pydantic" },
-    { name = "torch" },
-    { name = "transformers" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/b8/99/3fdabfc95609d6efdf02fa7f1ed0245524cb1209d3d4a17109d3205d2eed/compressed_tensors-0.11.0.tar.gz", hash = "sha256:95ddf19699f775df6494dd864e5f52e8a24f8015496520190c1a22c6cfc44b1f", size = 187566, upload-time = "2025-08-19T18:59:31.854Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/d2/81/e3073017a8f5c75169e79108eda209e6089e3f96c9f197d307cbda7df71c/compressed_tensors-0.11.0-py3-none-any.whl", hash = "sha256:e1cbc46e1ae032b7ceea915fe18c8d2de5a54d3a50a607969b6bdfe703b6cb83", size = 179951, upload-time = "2025-08-19T18:59:29.308Z" },
-]
-
 [[package]]
 name = "coverage"
 version = "7.10.7"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/51/26/d22c300112504f5f9a9fd2297ce33c35f3d353e4aeb987c8419453b2a7c2/coverage-7.10.7.tar.gz", hash = "sha256:f4ab143ab113be368a3e9b795f9cd7906c5ef407d6173fe9675a902e1fffc239", size = 827704, upload-time = "2025-09-21T20:03:56.815Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/e5/6c/3a3f7a46888e69d18abe3ccc6fe4cb16cccb1e6a2f99698931dafca489e6/coverage-7.10.7-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:fc04cc7a3db33664e0c2d10eb8990ff6b3536f6842c9590ae8da4c614b9ed05a", size = 217987, upload-time = "2025-09-21T20:00:57.218Z" },
@@ -642,70 +582,97 @@ wheels = [
 
 [package.optional-dependencies]
 toml = [
-    { name = "tomli", marker = "python_full_version <= '3.11'" },
+    { name = "tomli", marker = "python_full_version < '3.10'" },
 ]
 
 [[package]]
-name = "cupy-cuda12x"
-version = "13.6.0"
+name = "coverage"
+version = "7.12.0"
 source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "fastrlock" },
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
 ]
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/f7/2e/db22c5148884e4e384f6ebbc7971fa3710f3ba67ca492798890a0fdebc45/cupy_cuda12x-13.6.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:9e37f60f27ff9625dfdccc4688a09852707ec613e32ea9404f425dd22a386d14", size = 126341714, upload-time = "2025-08-18T08:24:08.335Z" },
-    { url = "https://files.pythonhosted.org/packages/53/2b/8064d94a6ab6b5c4e643d8535ab6af6cabe5455765540931f0ef60a0bc3b/cupy_cuda12x-13.6.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:e78409ea72f5ac7d6b6f3d33d99426a94005254fa57e10617f430f9fd7c3a0a1", size = 112238589, upload-time = "2025-08-18T08:24:15.541Z" },
-    { url = "https://files.pythonhosted.org/packages/54/64/71c6e08f76c06639e5112f69ee3bc1129be00054ad5f906d7fd3138af579/cupy_cuda12x-13.6.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:c790d012fd4d86872b9c89af9f5f15d91c30b8e3a4aa4dd04c2610f45f06ac44", size = 128016458, upload-time = "2025-08-18T08:24:26.394Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/d9/5c5077243cd92368c3eccecdbf91d76db15db338169042ffd1647533c6b1/cupy_cuda12x-13.6.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:77ba6745a130d880c962e687e4e146ebbb9014f290b0a80dbc4e4634eb5c3b48", size = 113039337, upload-time = "2025-08-18T08:24:31.814Z" },
-    { url = "https://files.pythonhosted.org/packages/12/c5/7e7fc4816d0de0154e5d9053242c3a08a0ca8b43ee656a6f7b3b95055a7b/cupy_cuda12x-13.6.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:a6970ceefe40f9acbede41d7fe17416bd277b1bd2093adcde457b23b578c5a59", size = 127334633, upload-time = "2025-08-18T08:24:43.065Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/95/d7e1295141e7d530674a3cc567e13ed0eb6b81524cb122d797ed996b5bea/cupy_cuda12x-13.6.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:79b0cacb5e8b190ef409f9e03f06ac8de1b021b0c0dda47674d446f5557e0eb1", size = 112886268, upload-time = "2025-08-18T08:24:49.294Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/23/aec6590ce32f0b0d284081192e9d25a07a410e57aaa28e2c3f2911881d2d/cupy_cuda12x-13.6.0-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:6ccd2fc75b0e0e24493531b8f8d8f978efecddb45f8479a48890c40d3805eb87", size = 126427104, upload-time = "2025-08-18T08:25:16.124Z" },
-    { url = "https://files.pythonhosted.org/packages/17/56/683967510b4392b518ba32a8560ef587d841f91b20afea76c6c642eb8b68/cupy_cuda12x-13.6.0-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:771f3135861b68199c18b49345210180d4fcdce4681b51c28224db389c4aac5d", size = 112305805, upload-time = "2025-08-18T08:25:24.097Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/89/26/4a96807b193b011588099c3b5c89fbb05294e5b90e71018e065465f34eb6/coverage-7.12.0.tar.gz", hash = "sha256:fc11e0a4e372cb5f282f16ef90d4a585034050ccda536451901abfb19a57f40c", size = 819341, upload-time = "2025-11-18T13:34:20.766Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/26/4a/0dc3de1c172d35abe512332cfdcc43211b6ebce629e4cc42e6cd25ed8f4d/coverage-7.12.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:32b75c2ba3f324ee37af3ccee5b30458038c50b349ad9b88cee85096132a575b", size = 217409, upload-time = "2025-11-18T13:31:53.122Z" },
+    { url = "https://files.pythonhosted.org/packages/01/c3/086198b98db0109ad4f84241e8e9ea7e5fb2db8c8ffb787162d40c26cc76/coverage-7.12.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:cb2a1b6ab9fe833714a483a915de350abc624a37149649297624c8d57add089c", size = 217927, upload-time = "2025-11-18T13:31:54.458Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/5f/34614dbf5ce0420828fc6c6f915126a0fcb01e25d16cf141bf5361e6aea6/coverage-7.12.0-cp310-cp310-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:5734b5d913c3755e72f70bf6cc37a0518d4f4745cde760c5d8e12005e62f9832", size = 244678, upload-time = "2025-11-18T13:31:55.805Z" },
+    { url = "https://files.pythonhosted.org/packages/55/7b/6b26fb32e8e4a6989ac1d40c4e132b14556131493b1d06bc0f2be169c357/coverage-7.12.0-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:b527a08cdf15753279b7afb2339a12073620b761d79b81cbe2cdebdb43d90daa", size = 246507, upload-time = "2025-11-18T13:31:57.05Z" },
+    { url = "https://files.pythonhosted.org/packages/06/42/7d70e6603d3260199b90fb48b537ca29ac183d524a65cc31366b2e905fad/coverage-7.12.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9bb44c889fb68004e94cab71f6a021ec83eac9aeabdbb5a5a88821ec46e1da73", size = 248366, upload-time = "2025-11-18T13:31:58.362Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/4a/d86b837923878424c72458c5b25e899a3c5ca73e663082a915f5b3c4d749/coverage-7.12.0-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:4b59b501455535e2e5dde5881739897967b272ba25988c89145c12d772810ccb", size = 245366, upload-time = "2025-11-18T13:31:59.572Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/c2/2adec557e0aa9721875f06ced19730fdb7fc58e31b02b5aa56f2ebe4944d/coverage-7.12.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:d8842f17095b9868a05837b7b1b73495293091bed870e099521ada176aa3e00e", size = 246408, upload-time = "2025-11-18T13:32:00.784Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/4b/8bd1f1148260df11c618e535fdccd1e5aaf646e55b50759006a4f41d8a26/coverage-7.12.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:c5a6f20bf48b8866095c6820641e7ffbe23f2ac84a2efc218d91235e404c7777", size = 244416, upload-time = "2025-11-18T13:32:01.963Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/13/3a248dd6a83df90414c54a4e121fd081fb20602ca43955fbe1d60e2312a9/coverage-7.12.0-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:5f3738279524e988d9da2893f307c2093815c623f8d05a8f79e3eff3a7a9e553", size = 244681, upload-time = "2025-11-18T13:32:03.408Z" },
+    { url = "https://files.pythonhosted.org/packages/76/30/aa833827465a5e8c938935f5d91ba055f70516941078a703740aaf1aa41f/coverage-7.12.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:e0d68c1f7eabbc8abe582d11fa393ea483caf4f44b0af86881174769f185c94d", size = 245300, upload-time = "2025-11-18T13:32:04.686Z" },
+    { url = "https://files.pythonhosted.org/packages/38/24/f85b3843af1370fb3739fa7571819b71243daa311289b31214fe3e8c9d68/coverage-7.12.0-cp310-cp310-win32.whl", hash = "sha256:7670d860e18b1e3ee5930b17a7d55ae6287ec6e55d9799982aa103a2cc1fa2ef", size = 220008, upload-time = "2025-11-18T13:32:05.806Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/a2/c7da5b9566f7164db9eefa133d17761ecb2c2fde9385d754e5b5c80f710d/coverage-7.12.0-cp310-cp310-win_amd64.whl", hash = "sha256:f999813dddeb2a56aab5841e687b68169da0d3f6fc78ccf50952fa2463746022", size = 220943, upload-time = "2025-11-18T13:32:07.166Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/0c/0dfe7f0487477d96432e4815537263363fb6dd7289743a796e8e51eabdf2/coverage-7.12.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:aa124a3683d2af98bd9d9c2bfa7a5076ca7e5ab09fdb96b81fa7d89376ae928f", size = 217535, upload-time = "2025-11-18T13:32:08.812Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/f5/f9a4a053a5bbff023d3bec259faac8f11a1e5a6479c2ccf586f910d8dac7/coverage-7.12.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:d93fbf446c31c0140208dcd07c5d882029832e8ed7891a39d6d44bd65f2316c3", size = 218044, upload-time = "2025-11-18T13:32:10.329Z" },
+    { url = "https://files.pythonhosted.org/packages/95/c5/84fc3697c1fa10cd8571919bf9693f693b7373278daaf3b73e328d502bc8/coverage-7.12.0-cp311-cp311-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:52ca620260bd8cd6027317bdd8b8ba929be1d741764ee765b42c4d79a408601e", size = 248440, upload-time = "2025-11-18T13:32:12.536Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/36/2d93fbf6a04670f3874aed397d5a5371948a076e3249244a9e84fb0e02d6/coverage-7.12.0-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f3433ffd541380f3a0e423cff0f4926d55b0cc8c1d160fdc3be24a4c03aa65f7", size = 250361, upload-time = "2025-11-18T13:32:13.852Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/49/66dc65cc456a6bfc41ea3d0758c4afeaa4068a2b2931bf83be6894cf1058/coverage-7.12.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f7bbb321d4adc9f65e402c677cd1c8e4c2d0105d3ce285b51b4d87f1d5db5245", size = 252472, upload-time = "2025-11-18T13:32:15.068Z" },
+    { url = "https://files.pythonhosted.org/packages/35/1f/ebb8a18dffd406db9fcd4b3ae42254aedcaf612470e8712f12041325930f/coverage-7.12.0-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:22a7aade354a72dff3b59c577bfd18d6945c61f97393bc5fb7bd293a4237024b", size = 248592, upload-time = "2025-11-18T13:32:16.328Z" },
+    { url = "https://files.pythonhosted.org/packages/da/a8/67f213c06e5ea3b3d4980df7dc344d7fea88240b5fe878a5dcbdfe0e2315/coverage-7.12.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:3ff651dcd36d2fea66877cd4a82de478004c59b849945446acb5baf9379a1b64", size = 250167, upload-time = "2025-11-18T13:32:17.687Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/00/e52aef68154164ea40cc8389c120c314c747fe63a04b013a5782e989b77f/coverage-7.12.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:31b8b2e38391a56e3cea39d22a23faaa7c3fc911751756ef6d2621d2a9daf742", size = 248238, upload-time = "2025-11-18T13:32:19.2Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/a4/4d88750bcf9d6d66f77865e5a05a20e14db44074c25fd22519777cb69025/coverage-7.12.0-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:297bc2da28440f5ae51c845a47c8175a4db0553a53827886e4fb25c66633000c", size = 247964, upload-time = "2025-11-18T13:32:21.027Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/6b/b74693158899d5b47b0bf6238d2c6722e20ba749f86b74454fac0696bb00/coverage-7.12.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:6ff7651cc01a246908eac162a6a86fc0dbab6de1ad165dfb9a1e2ec660b44984", size = 248862, upload-time = "2025-11-18T13:32:22.304Z" },
+    { url = "https://files.pythonhosted.org/packages/18/de/6af6730227ce0e8ade307b1cc4a08e7f51b419a78d02083a86c04ccceb29/coverage-7.12.0-cp311-cp311-win32.whl", hash = "sha256:313672140638b6ddb2c6455ddeda41c6a0b208298034544cfca138978c6baed6", size = 220033, upload-time = "2025-11-18T13:32:23.714Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/a1/e7f63021a7c4fe20994359fcdeae43cbef4a4d0ca36a5a1639feeea5d9e1/coverage-7.12.0-cp311-cp311-win_amd64.whl", hash = "sha256:a1783ed5bd0d5938d4435014626568dc7f93e3cb99bc59188cc18857c47aa3c4", size = 220966, upload-time = "2025-11-18T13:32:25.599Z" },
+    { url = "https://files.pythonhosted.org/packages/77/e8/deae26453f37c20c3aa0c4433a1e32cdc169bf415cce223a693117aa3ddd/coverage-7.12.0-cp311-cp311-win_arm64.whl", hash = "sha256:4648158fd8dd9381b5847622df1c90ff314efbfc1df4550092ab6013c238a5fc", size = 219637, upload-time = "2025-11-18T13:32:27.265Z" },
+    { url = "https://files.pythonhosted.org/packages/02/bf/638c0427c0f0d47638242e2438127f3c8ee3cfc06c7fdeb16778ed47f836/coverage-7.12.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:29644c928772c78512b48e14156b81255000dcfd4817574ff69def189bcb3647", size = 217704, upload-time = "2025-11-18T13:32:28.906Z" },
+    { url = "https://files.pythonhosted.org/packages/08/e1/706fae6692a66c2d6b871a608bbde0da6281903fa0e9f53a39ed441da36a/coverage-7.12.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8638cbb002eaa5d7c8d04da667813ce1067080b9a91099801a0053086e52b736", size = 218064, upload-time = "2025-11-18T13:32:30.161Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/8b/eb0231d0540f8af3ffda39720ff43cb91926489d01524e68f60e961366e4/coverage-7.12.0-cp312-cp312-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:083631eeff5eb9992c923e14b810a179798bb598e6a0dd60586819fc23be6e60", size = 249560, upload-time = "2025-11-18T13:32:31.835Z" },
+    { url = "https://files.pythonhosted.org/packages/e9/a1/67fb52af642e974d159b5b379e4d4c59d0ebe1288677fbd04bbffe665a82/coverage-7.12.0-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:99d5415c73ca12d558e07776bd957c4222c687b9f1d26fa0e1b57e3598bdcde8", size = 252318, upload-time = "2025-11-18T13:32:33.178Z" },
+    { url = "https://files.pythonhosted.org/packages/41/e5/38228f31b2c7665ebf9bdfdddd7a184d56450755c7e43ac721c11a4b8dab/coverage-7.12.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e949ebf60c717c3df63adb4a1a366c096c8d7fd8472608cd09359e1bd48ef59f", size = 253403, upload-time = "2025-11-18T13:32:34.45Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/4b/df78e4c8188f9960684267c5a4897836f3f0f20a20c51606ee778a1d9749/coverage-7.12.0-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:6d907ddccbca819afa2cd014bc69983b146cca2735a0b1e6259b2a6c10be1e70", size = 249984, upload-time = "2025-11-18T13:32:35.747Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/51/bb163933d195a345c6f63eab9e55743413d064c291b6220df754075c2769/coverage-7.12.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:b1518ecbad4e6173f4c6e6c4a46e49555ea5679bf3feda5edb1b935c7c44e8a0", size = 251339, upload-time = "2025-11-18T13:32:37.352Z" },
+    { url = "https://files.pythonhosted.org/packages/15/40/c9b29cdb8412c837cdcbc2cfa054547dd83affe6cbbd4ce4fdb92b6ba7d1/coverage-7.12.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:51777647a749abdf6f6fd8c7cffab12de68ab93aab15efc72fbbb83036c2a068", size = 249489, upload-time = "2025-11-18T13:32:39.212Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/da/b3131e20ba07a0de4437a50ef3b47840dfabf9293675b0cd5c2c7f66dd61/coverage-7.12.0-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:42435d46d6461a3b305cdfcad7cdd3248787771f53fe18305548cba474e6523b", size = 249070, upload-time = "2025-11-18T13:32:40.598Z" },
+    { url = "https://files.pythonhosted.org/packages/70/81/b653329b5f6302c08d683ceff6785bc60a34be9ae92a5c7b63ee7ee7acec/coverage-7.12.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:5bcead88c8423e1855e64b8057d0544e33e4080b95b240c2a355334bb7ced937", size = 250929, upload-time = "2025-11-18T13:32:42.915Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/00/250ac3bca9f252a5fb1338b5ad01331ebb7b40223f72bef5b1b2cb03aa64/coverage-7.12.0-cp312-cp312-win32.whl", hash = "sha256:dcbb630ab034e86d2a0f79aefd2be07e583202f41e037602d438c80044957baa", size = 220241, upload-time = "2025-11-18T13:32:44.665Z" },
+    { url = "https://files.pythonhosted.org/packages/64/1c/77e79e76d37ce83302f6c21980b45e09f8aa4551965213a10e62d71ce0ab/coverage-7.12.0-cp312-cp312-win_amd64.whl", hash = "sha256:2fd8354ed5d69775ac42986a691fbf68b4084278710cee9d7c3eaa0c28fa982a", size = 221051, upload-time = "2025-11-18T13:32:46.008Z" },
+    { url = "https://files.pythonhosted.org/packages/31/f5/641b8a25baae564f9e52cac0e2667b123de961985709a004e287ee7663cc/coverage-7.12.0-cp312-cp312-win_arm64.whl", hash = "sha256:737c3814903be30695b2de20d22bcc5428fdae305c61ba44cdc8b3252984c49c", size = 219692, upload-time = "2025-11-18T13:32:47.372Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/a3/43b749004e3c09452e39bb56347a008f0a0668aad37324a99b5c8ca91d9e/coverage-7.12.0-py3-none-any.whl", hash = "sha256:159d50c0b12e060b15ed3d39f87ed43d4f7f7ad40b8a534f4dd331adbb51104a", size = 209503, upload-time = "2025-11-18T13:34:18.892Z" },
+]
+
+[package.optional-dependencies]
+toml = [
+    { name = "tomli", marker = "python_full_version >= '3.10' and python_full_version <= '3.11'" },
 ]
 
 [[package]]
 name = "datasets"
-version = "4.1.1"
+version = "4.4.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "dill" },
-    { name = "filelock" },
+    { name = "filelock", version = "3.19.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "filelock", version = "3.20.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "fsspec", extra = ["http"] },
+    { name = "httpx" },
     { name = "huggingface-hub" },
     { name = "multiprocess" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "packaging" },
     { name = "pandas" },
-    { name = "pyarrow" },
+    { name = "pyarrow", version = "21.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pyarrow", version = "22.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pyyaml" },
     { name = "requests" },
     { name = "tqdm" },
     { name = "xxhash" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/91/a4/73f8e6ef52c535e1d20d5b2ca83bfe6de399d8b8b8a61ccc8d63d60735aa/datasets-4.1.1.tar.gz", hash = "sha256:7d8d5ba8b12861d2c44bfff9c83484ebfafff1ff553371e5901a8d3aab5450e2", size = 579324, upload-time = "2025-09-18T13:14:27.108Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/f4/c8/09012ac195a0aab58755800d2efdc0e7d5905053509f12cb5d136c911cda/datasets-4.1.1-py3-none-any.whl", hash = "sha256:62e4f6899a36be9ec74a7e759a6951253cc85b3fcfa0a759b0efa8353b149dac", size = 503623, upload-time = "2025-09-18T13:14:25.111Z" },
-]
-
-[[package]]
-name = "depyf"
-version = "0.19.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "astor" },
-    { name = "dill" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/19/38/69157d711be575f1b9cf3177b64ef4ade44373fc02839f183fdd98ec2dd6/depyf-0.19.0.tar.gz", hash = "sha256:afed0916b32d141cc90fa6220df01885eda442ca43b297d5050eeb90b4a5cb44", size = 6171405, upload-time = "2025-04-20T08:07:41.224Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/93/bf/0dae295d6d1ba0b1a200a9dd216838464b5bbd05da01407cb1330b377445/datasets-4.4.1.tar.gz", hash = "sha256:80322699aa8c0bbbdb7caa87906da689c3c2e29523cff698775c67f28fdab1fc", size = 585341, upload-time = "2025-11-05T16:00:38.162Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/28/4d/1192acbcdc5e843f5e5d51f6e8788f2b60a9fe0b578ac385ded67a0b0b26/depyf-0.19.0-py3-none-any.whl", hash = "sha256:040b35fc0997d49df024b7d094f2a7836f91e9ed02f49982dd37e70aa3285ad5", size = 39034, upload-time = "2025-04-20T08:07:37.036Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/5e/6f8d874366788ad5d549e9ba258037d974dda6e004843be1bda794571701/datasets-4.4.1-py3-none-any.whl", hash = "sha256:c1163de5211e42546079ab355cc0250c7e6db16eb209ac5ac6252f801f596c44", size = 511591, upload-time = "2025-11-05T16:00:36.365Z" },
 ]
 
 [[package]]
 name = "diff-cover"
-version = "9.7.1"
+version = "9.7.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "chardet" },
@@ -713,9 +680,9 @@ dependencies = [
     { name = "pluggy" },
     { name = "pygments" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/eb/b7/3124f2038eb9d2842665c109d4516110e3e5570615243ee1065fdc65ac7d/diff_cover-9.7.1.tar.gz", hash = "sha256:a03b78bc64b7dc27dda7f11c46bc6d44f4e121c8fb9a5068d41f20a05d463548", size = 100366, upload-time = "2025-09-26T01:02:43.81Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/e1/66/c200febf00c8b0115825a99c3fdb2e77798419b8a632b1f8e4c81a75e734/diff_cover-9.7.2.tar.gz", hash = "sha256:872c820d2ecbf79c61d52c7dc70419015e0ab9289589566c791dd270fc0c6e3b", size = 100374, upload-time = "2025-11-11T02:49:34.883Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/1d/c2/36d747a14dd3834af7e2f4f86daf8d6e21762ffbdb2b177a5bb3be8f0e8f/diff_cover-9.7.1-py3-none-any.whl", hash = "sha256:a662f3cacf57ba494dcb0235349c4912ae7f44345b162894ce5949745501c9c7", size = 56008, upload-time = "2025-09-26T01:02:42.399Z" },
+    { url = "https://files.pythonhosted.org/packages/21/dd/6db45462ffbfe3791c6de81be5db1c9e883e6b664ae39fc8adf947c738a6/diff_cover-9.7.2-py3-none-any.whl", hash = "sha256:cd6498620c747c2493a6c83c14362c32868bfd91cd8d0dd093f136070ec4ffc5", size = 56015, upload-time = "2025-11-11T02:49:33.294Z" },
 ]
 
 [[package]]
@@ -754,32 +721,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/12/b3/231ffd4ab1fc9d679809f356cebee130ac7daa00d6d6f3206dd4fd137e9e/distro-1.9.0-py3-none-any.whl", hash = "sha256:7bffd925d65168f85027d8da9af6bddab658135b840670a223589bc0c8ef02b2", size = 20277, upload-time = "2023-12-24T09:54:30.421Z" },
 ]
 
-[[package]]
-name = "dnspython"
-version = "2.7.0"
-source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "python_full_version < '3.10'",
-]
-sdist = { url = "https://files.pythonhosted.org/packages/b5/4a/263763cb2ba3816dd94b08ad3a33d5fdae34ecb856678773cc40a3605829/dnspython-2.7.0.tar.gz", hash = "sha256:ce9c432eda0dc91cf618a5cedf1a4e142651196bbcd2c80e89ed5a907e5cfaf1", size = 345197, upload-time = "2024-10-05T20:14:59.362Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/68/1b/e0a87d256e40e8c888847551b20a017a6b98139178505dc7ffb96f04e954/dnspython-2.7.0-py3-none-any.whl", hash = "sha256:b4c34b7d10b51bcc3a5071e7b8dee77939f1e878477eeecc965e9835f63c6c86", size = 313632, upload-time = "2024-10-05T20:14:57.687Z" },
-]
-
-[[package]]
-name = "dnspython"
-version = "2.8.0"
-source = { registry = "https://pypi.org/simple" }
-resolution-markers = [
-    "python_full_version >= '3.12'",
-    "python_full_version == '3.11.*'",
-    "python_full_version == '3.10.*'",
-]
-sdist = { url = "https://files.pythonhosted.org/packages/8c/8b/57666417c0f90f08bcafa776861060426765fdb422eb10212086fb811d26/dnspython-2.8.0.tar.gz", hash = "sha256:181d3c6996452cb1189c4046c61599b84a5a86e099562ffde77d26984ff26d0f", size = 368251, upload-time = "2025-09-07T18:58:00.022Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/ba/5a/18ad964b0086c6e62e2e7500f7edc89e3faa45033c71c1893d34eed2b2de/dnspython-2.8.0-py3-none-any.whl", hash = "sha256:01d9bbc4a2d76bf0db7c1f729812ded6d912bd318d3b1cf81d30c0f845dbf3af", size = 331094, upload-time = "2025-09-07T18:57:58.071Z" },
-]
-
 [[package]]
 name = "docstring-parser"
 version = "0.17.0"
@@ -796,9 +737,10 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "backoff" },
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pydantic" },
-    { name = "pydantic-settings" },
+    { name = "pydantic-settings", version = "2.11.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pydantic-settings", version = "2.12.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "python-dateutil" },
     { name = "typing-extensions" },
     { name = "urllib3" },
@@ -807,29 +749,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/46/0c/c0507e1641185890ef173bce883d9a02d60b446441fd674db352d9cf52e5/dottxt-0.1.5-py3-none-any.whl", hash = "sha256:645af5f9c43721893b7be91c3a18d1a45bee0f7544caf370f54587df0e127dd2", size = 65293, upload-time = "2025-02-14T17:10:12.353Z" },
 ]
 
-[[package]]
-name = "einops"
-version = "0.8.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/e5/81/df4fbe24dff8ba3934af99044188e20a98ed441ad17a274539b74e82e126/einops-0.8.1.tar.gz", hash = "sha256:de5d960a7a761225532e0f1959e5315ebeafc0cd43394732f103ca44b9837e84", size = 54805, upload-time = "2025-02-09T03:17:00.434Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/87/62/9773de14fe6c45c23649e98b83231fffd7b9892b6cf863251dc2afa73643/einops-0.8.1-py3-none-any.whl", hash = "sha256:919387eb55330f5757c6bea9165c5ff5cfe63a642682ea788a6d472576d81737", size = 64359, upload-time = "2025-02-09T03:17:01.998Z" },
-]
-
-[[package]]
-name = "email-validator"
-version = "2.3.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "dnspython", version = "2.7.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "dnspython", version = "2.8.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "idna" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/f5/22/900cb125c76b7aaa450ce02fd727f452243f2e91a61af068b40adba60ea9/email_validator-2.3.0.tar.gz", hash = "sha256:9fc05c37f2f6cf439ff414f8fc46d917929974a82244c20eb10231ba60c54426", size = 51238, upload-time = "2025-08-26T13:09:06.831Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/de/15/545e2b6cf2e3be84bc1ed85613edd75b8aea69807a71c26f4ca6a9258e82/email_validator-2.3.0-py3-none-any.whl", hash = "sha256:80f13f623413e6b197ae73bb10bf4eb0908faf509ad8362c5edeb0be7fd450b4", size = 35604, upload-time = "2025-08-26T13:09:05.858Z" },
-]
-
 [[package]]
 name = "etils"
 version = "1.5.2"
@@ -880,126 +799,65 @@ epy = [
 
 [[package]]
 name = "eval-type-backport"
-version = "0.2.2"
+version = "0.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/30/ea/8b0ac4469d4c347c6a385ff09dc3c048c2d021696664e26c7ee6791631b5/eval_type_backport-0.2.2.tar.gz", hash = "sha256:f0576b4cf01ebb5bd358d02314d31846af5e07678387486e2c798af0e7d849c1", size = 9079, upload-time = "2024-12-21T20:09:46.005Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/51/23/079e39571d6dd8d90d7a369ecb55ad766efb6bae4e77389629e14458c280/eval_type_backport-0.3.0.tar.gz", hash = "sha256:1638210401e184ff17f877e9a2fa076b60b5838790f4532a21761cc2be67aea1", size = 9272, upload-time = "2025-11-13T20:56:50.845Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ce/31/55cd413eaccd39125368be33c46de24a1f639f2e12349b0361b4678f3915/eval_type_backport-0.2.2-py3-none-any.whl", hash = "sha256:cb6ad7c393517f476f96d456d0412ea80f0a8cf96f6892834cd9340149111b0a", size = 5830, upload-time = "2024-12-21T20:09:44.175Z" },
+    { url = "https://files.pythonhosted.org/packages/19/d8/2a1c638d9e0aa7e269269a1a1bf423ddd94267f1a01bbe3ad03432b67dd4/eval_type_backport-0.3.0-py3-none-any.whl", hash = "sha256:975a10a0fe333c8b6260d7fdb637698c9a16c3a9e3b6eb943fee6a6f67a37fe8", size = 6061, upload-time = "2025-11-13T20:56:49.499Z" },
 ]
 
 [[package]]
 name = "exceptiongroup"
-version = "1.3.0"
+version = "1.3.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/0b/9f/a65090624ecf468cdca03533906e7c69ed7588582240cfe7cc9e770b50eb/exceptiongroup-1.3.0.tar.gz", hash = "sha256:b241f5885f560bc56a59ee63ca4c6a8bfa46ae4ad651af316d4e81817bb9fd88", size = 29749, upload-time = "2025-05-10T17:42:51.123Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/50/79/66800aadf48771f6b62f7eb014e352e5d06856655206165d775e675a02c9/exceptiongroup-1.3.1.tar.gz", hash = "sha256:8b412432c6055b0b7d14c310000ae93352ed6754f70fa8f7c34141f91c4e3219", size = 30371, upload-time = "2025-11-21T23:01:54.787Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/36/f4/c6e662dade71f56cd2f3735141b265c3c79293c109549c1e6933b0651ffc/exceptiongroup-1.3.0-py3-none-any.whl", hash = "sha256:4d111e6e0c13d0644cad6ddaa7ed0261a0b36971f6d23e7ec9b4b9097da78a10", size = 16674, upload-time = "2025-05-10T17:42:49.33Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/0e/97c33bf5009bdbac74fd2beace167cab3f978feb69cc36f1ef79360d6c4e/exceptiongroup-1.3.1-py3-none-any.whl", hash = "sha256:a7a39a3bd276781e98394987d3a5701d0c4edffb633bb7a5144577f82c773598", size = 16740, upload-time = "2025-11-21T23:01:53.443Z" },
 ]
 
 [[package]]
 name = "fastapi"
-version = "0.118.0"
+version = "0.121.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
+    { name = "annotated-doc" },
     { name = "pydantic" },
-    { name = "starlette" },
+    { name = "starlette", version = "0.49.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "starlette", version = "0.50.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/28/3c/2b9345a6504e4055eaa490e0b41c10e338ad61d9aeaae41d97807873cdf2/fastapi-0.118.0.tar.gz", hash = "sha256:5e81654d98c4d2f53790a7d32d25a7353b30c81441be7d0958a26b5d761fa1c8", size = 310536, upload-time = "2025-09-29T03:37:23.126Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/80/f0/086c442c6516195786131b8ca70488c6ef11d2f2e33c9a893576b2b0d3f7/fastapi-0.121.3.tar.gz", hash = "sha256:0055bc24fe53e56a40e9e0ad1ae2baa81622c406e548e501e717634e2dfbc40b", size = 344501, upload-time = "2025-11-19T16:53:39.243Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/54/20/54e2bdaad22ca91a59455251998d43094d5c3d3567c52c7c04774b3f43f2/fastapi-0.118.0-py3-none-any.whl", hash = "sha256:705137a61e2ef71019d2445b123aa8845bd97273c395b744d5a7dfe559056855", size = 97694, upload-time = "2025-09-29T03:37:21.338Z" },
-]
-
-[package.optional-dependencies]
-standard = [
-    { name = "email-validator" },
-    { name = "fastapi-cli", extra = ["standard"] },
-    { name = "httpx" },
-    { name = "jinja2" },
-    { name = "python-multipart" },
-    { name = "uvicorn", extra = ["standard"] },
+    { url = "https://files.pythonhosted.org/packages/98/b6/4f620d7720fc0a754c8c1b7501d73777f6ba43b57c8ab99671f4d7441eb8/fastapi-0.121.3-py3-none-any.whl", hash = "sha256:0c78fc87587fcd910ca1bbf5bc8ba37b80e119b388a7206b39f0ecc95ebf53e9", size = 109801, upload-time = "2025-11-19T16:53:37.918Z" },
 ]
 
 [[package]]
-name = "fastapi-cli"
-version = "0.0.13"
+name = "filelock"
+version = "3.19.1"
 source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "rich-toolkit" },
-    { name = "typer" },
-    { name = "uvicorn", extra = ["standard"] },
+resolution-markers = [
+    "python_full_version < '3.10'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/32/4e/3f61850012473b097fc5297d681bd85788e186fadb8555b67baf4c7707f4/fastapi_cli-0.0.13.tar.gz", hash = "sha256:312addf3f57ba7139457cf0d345c03e2170cc5a034057488259c33cd7e494529", size = 17780, upload-time = "2025-09-20T16:37:31.089Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/40/bb/0ab3e58d22305b6f5440629d20683af28959bf793d98d11950e305c1c326/filelock-3.19.1.tar.gz", hash = "sha256:66eda1888b0171c998b35be2bcc0f6d75c388a7ce20c3f3f37aa8e96c2dddf58", size = 17687, upload-time = "2025-08-14T16:56:03.016Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/08/36/7432750f3638324b055496d2c952000bea824259fca70df5577a6a3c172f/fastapi_cli-0.0.13-py3-none-any.whl", hash = "sha256:219b73ccfde7622559cef1d43197da928516acb4f21f2ec69128c4b90057baba", size = 11142, upload-time = "2025-09-20T16:37:29.695Z" },
-]
-
-[package.optional-dependencies]
-standard = [
-    { name = "fastapi-cloud-cli" },
-    { name = "uvicorn", extra = ["standard"] },
-]
-
-[[package]]
-name = "fastapi-cloud-cli"
-version = "0.3.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "httpx" },
-    { name = "pydantic", extra = ["email"] },
-    { name = "rich-toolkit" },
-    { name = "rignore" },
-    { name = "sentry-sdk" },
-    { name = "typer" },
-    { name = "uvicorn", extra = ["standard"] },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/a6/5f/17b403148a23dd708e3166f534136f4d3918942e168aca66659311eb0678/fastapi_cloud_cli-0.3.0.tar.gz", hash = "sha256:17c7f8baa16b2f907696bf77d49df4a04e8715bbf5233024f273870f3ff1ca4d", size = 24388, upload-time = "2025-10-02T13:25:52.361Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/58/59/7d12c5173fe2eed21e99bb1a6eb7e4f301951db870a4d915d126e0b6062d/fastapi_cloud_cli-0.3.0-py3-none-any.whl", hash = "sha256:572677dbe38b6d4712d30097a8807b383d648ca09eb58e4a07cef4a517020832", size = 19921, upload-time = "2025-10-02T13:25:51.164Z" },
-]
-
-[[package]]
-name = "fastrlock"
-version = "0.8.3"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/73/b1/1c3d635d955f2b4bf34d45abf8f35492e04dbd7804e94ce65d9f928ef3ec/fastrlock-0.8.3.tar.gz", hash = "sha256:4af6734d92eaa3ab4373e6c9a1dd0d5ad1304e172b1521733c6c3b3d73c8fa5d", size = 79327, upload-time = "2024-12-17T11:03:39.638Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/be/b4/aae7ed94b8122c325d89eb91336084596cebc505dc629b795fcc9629606d/fastrlock-0.8.3-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:7a77ebb0a24535ef4f167da2c5ee35d9be1e96ae192137e9dc3ff75b8dfc08a5", size = 48220, upload-time = "2024-12-17T11:01:51.071Z" },
-    { url = "https://files.pythonhosted.org/packages/96/87/9807af47617fdd65c68b0fcd1e714542c1d4d3a1f1381f591f1aa7383a53/fastrlock-0.8.3-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_24_i686.whl", hash = "sha256:d51f7fb0db8dab341b7f03a39a3031678cf4a98b18533b176c533c122bfce47d", size = 49551, upload-time = "2024-12-17T11:01:52.316Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/12/e201634810ac9aee59f93e3953cb39f98157d17c3fc9d44900f1209054e9/fastrlock-0.8.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:767ec79b7f6ed9b9a00eb9ff62f2a51f56fdb221c5092ab2dadec34a9ccbfc6e", size = 49398, upload-time = "2024-12-17T11:01:53.514Z" },
-    { url = "https://files.pythonhosted.org/packages/15/a1/439962ed439ff6f00b7dce14927e7830e02618f26f4653424220a646cd1c/fastrlock-0.8.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0d6a77b3f396f7d41094ef09606f65ae57feeb713f4285e8e417f4021617ca62", size = 53334, upload-time = "2024-12-17T11:01:55.518Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/9e/1ae90829dd40559ab104e97ebe74217d9da794c4bb43016da8367ca7a596/fastrlock-0.8.3-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:92577ff82ef4a94c5667d6d2841f017820932bc59f31ffd83e4a2c56c1738f90", size = 52495, upload-time = "2024-12-17T11:01:57.76Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/8c/5e746ee6f3d7afbfbb0d794c16c71bfd5259a4e3fb1dda48baf31e46956c/fastrlock-0.8.3-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:3df8514086e16bb7c66169156a8066dc152f3be892c7817e85bf09a27fa2ada2", size = 51972, upload-time = "2024-12-17T11:02:01.384Z" },
-    { url = "https://files.pythonhosted.org/packages/be/91/5f3afba7d14b8b7d60ac651375f50fff9220d6ccc3bef233d2bd74b73ec7/fastrlock-0.8.3-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:55d42f6286b9d867370af4c27bc70d04ce2d342fe450c4a4fcce14440514e695", size = 48911, upload-time = "2024-12-17T11:02:06.173Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/7a/e37bd72d7d70a8a551b3b4610d028bd73ff5d6253201d5d3cf6296468bee/fastrlock-0.8.3-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_24_i686.whl", hash = "sha256:bbc3bf96dcbd68392366c477f78c9d5c47e5d9290cb115feea19f20a43ef6d05", size = 50357, upload-time = "2024-12-17T11:02:07.418Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/ef/a13b8bab8266840bf38831d7bf5970518c02603d00a548a678763322d5bf/fastrlock-0.8.3-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:77ab8a98417a1f467dafcd2226718f7ca0cf18d4b64732f838b8c2b3e4b55cb5", size = 50222, upload-time = "2024-12-17T11:02:08.745Z" },
-    { url = "https://files.pythonhosted.org/packages/01/e2/5e5515562b2e9a56d84659377176aef7345da2c3c22909a1897fe27e14dd/fastrlock-0.8.3-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:04bb5eef8f460d13b8c0084ea5a9d3aab2c0573991c880c0a34a56bb14951d30", size = 54553, upload-time = "2024-12-17T11:02:10.925Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/8f/65907405a8cdb2fc8beaf7d09a9a07bb58deff478ff391ca95be4f130b70/fastrlock-0.8.3-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:8c9d459ce344c21ff03268212a1845aa37feab634d242131bc16c2a2355d5f65", size = 53362, upload-time = "2024-12-17T11:02:12.476Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/b9/ae6511e52738ba4e3a6adb7c6a20158573fbc98aab448992ece25abb0b07/fastrlock-0.8.3-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:33e6fa4af4f3af3e9c747ec72d1eadc0b7ba2035456c2afb51c24d9e8a56f8fd", size = 52836, upload-time = "2024-12-17T11:02:13.74Z" },
-    { url = "https://files.pythonhosted.org/packages/57/21/ea1511b0ef0d5457efca3bf1823effb9c5cad4fc9dca86ce08e4d65330ce/fastrlock-0.8.3-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:85a49a1f1e020097d087e1963e42cea6f307897d5ebe2cb6daf4af47ffdd3eed", size = 52201, upload-time = "2024-12-17T11:02:19.512Z" },
-    { url = "https://files.pythonhosted.org/packages/80/07/cdecb7aa976f34328372f1c4efd6c9dc1b039b3cc8d3f38787d640009a25/fastrlock-0.8.3-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5f13ec08f1adb1aa916c384b05ecb7dbebb8df9ea81abd045f60941c6283a670", size = 53924, upload-time = "2024-12-17T11:02:20.85Z" },
-    { url = "https://files.pythonhosted.org/packages/88/6d/59c497f8db9a125066dd3a7442fab6aecbe90d6fec344c54645eaf311666/fastrlock-0.8.3-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:0ea4e53a04980d646def0f5e4b5e8bd8c7884288464acab0b37ca0c65c482bfe", size = 52140, upload-time = "2024-12-17T11:02:22.263Z" },
-    { url = "https://files.pythonhosted.org/packages/62/04/9138943c2ee803d62a48a3c17b69de2f6fa27677a6896c300369e839a550/fastrlock-0.8.3-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:38340f6635bd4ee2a4fb02a3a725759fe921f2ca846cb9ca44531ba739cc17b4", size = 53261, upload-time = "2024-12-17T11:02:24.418Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/ad/c8fb45d5efcdf791f0dba5c09896b39eabbdc108f5b518941a2caae52f23/fastrlock-0.8.3-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.manylinux_2_24_aarch64.whl", hash = "sha256:40b328369005a0b32de14b699192aed32f549c2d2b27a5e1f614fb7ac4cec4e9", size = 49804, upload-time = "2024-12-17T11:03:25.14Z" },
-    { url = "https://files.pythonhosted.org/packages/47/15/365918306c30132bd63ae27b154e2aadb4e71c178297fc635e613aa4e767/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_24_i686.whl", hash = "sha256:6cbfb6f7731b5a280851c93883624424068fa5b22c2f546d8ae6f1fd9311e36d", size = 51763, upload-time = "2024-12-17T11:03:27.546Z" },
-    { url = "https://files.pythonhosted.org/packages/84/39/74fda02c3edeb6cc69cf5a4616e394f5636a227262788f4d33fee8401941/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1fced4cb0b3f1616be68092b70a56e9173713a4a943d02e90eb9c7897a7b5e07", size = 38277, upload-time = "2024-12-17T11:03:28.879Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/8f/86cf1dfd1d0d027110d0177946ddb34a28a6d0040331899df6dabcf9f332/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_24_x86_64.whl", hash = "sha256:387b2ac642938a20170a50f528817026c561882ea33306c5cbe750ae10d0a7c2", size = 51561, upload-time = "2024-12-17T11:03:30.184Z" },
-    { url = "https://files.pythonhosted.org/packages/09/5a/eabdde19fee480da1e0b3af4aef7f285d544c1ea733dc0f3df22a620df23/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5a0d31840a28d66573047d2df410eb971135a2461fb952894bf51c9533cbfea5", size = 55191, upload-time = "2024-12-17T11:03:31.604Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/e3/bdbe97b6d0d25b44bb2141c8e6be5f5bf573cf6413c9e23a7029af2d8922/fastrlock-0.8.3-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.whl", hash = "sha256:0a9dc6fa73174f974dfb22778d05a44445b611a41d5d3776b0d5daa9e50225c6", size = 39536, upload-time = "2024-12-17T11:03:32.958Z" },
-    { url = "https://files.pythonhosted.org/packages/0a/d0/aa12b01ea28606398bcd781b01c07dad388616029a14e065b1f0ae64d8ca/fastrlock-0.8.3-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:9842b7722e4923fe76b08d8c58a9415a9a50d4c29b80673cffeae4874ea6626a", size = 54474, upload-time = "2024-12-17T11:03:34.199Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/fb/e82f40aa6a4844107f6ace90f70b72c0cd26838a5d1984e44ec4a5d72f30/fastrlock-0.8.3-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:05029d7080c0c61a81d5fee78e842c9a1bf22552cd56129451a252655290dcef", size = 54142, upload-time = "2024-12-17T11:03:35.587Z" },
+    { url = "https://files.pythonhosted.org/packages/42/14/42b2651a2f46b022ccd948bca9f2d5af0fd8929c4eec235b8d6d844fbe67/filelock-3.19.1-py3-none-any.whl", hash = "sha256:d38e30481def20772f5baf097c122c3babc4fcdb7e14e57049eb9d88c6dc017d", size = 15988, upload-time = "2025-08-14T16:56:01.633Z" },
 ]
 
 [[package]]
 name = "filelock"
-version = "3.19.1"
+version = "3.20.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/40/bb/0ab3e58d22305b6f5440629d20683af28959bf793d98d11950e305c1c326/filelock-3.19.1.tar.gz", hash = "sha256:66eda1888b0171c998b35be2bcc0f6d75c388a7ce20c3f3f37aa8e96c2dddf58", size = 17687, upload-time = "2025-08-14T16:56:03.016Z" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/58/46/0028a82567109b5ef6e4d2a1f04a583fb513e6cf9527fcdd09afd817deeb/filelock-3.20.0.tar.gz", hash = "sha256:711e943b4ec6be42e1d4e6690b48dc175c822967466bb31c0c293f34334c13f4", size = 18922, upload-time = "2025-10-08T18:03:50.056Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/42/14/42b2651a2f46b022ccd948bca9f2d5af0fd8929c4eec235b8d6d844fbe67/filelock-3.19.1-py3-none-any.whl", hash = "sha256:d38e30481def20772f5baf097c122c3babc4fcdb7e14e57049eb9d88c6dc017d", size = 15988, upload-time = "2025-08-14T16:56:01.633Z" },
+    { url = "https://files.pythonhosted.org/packages/76/91/7216b27286936c16f5b4d0c530087e4a54eead683e6b0b73dd0c64844af6/filelock-3.20.0-py3-none-any.whl", hash = "sha256:339b4732ffda5cd79b13f4e2711a31b0365ce445d95d243bb996273d072546a2", size = 16054, upload-time = "2025-10-08T18:03:48.35Z" },
 ]
 
 [[package]]
@@ -1045,7 +903,7 @@ dependencies = [
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "msgpack", marker = "python_full_version == '3.10.*'" },
     { name = "optax", version = "0.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "orbax-checkpoint", version = "0.11.25", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "orbax-checkpoint", version = "0.11.28", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "pyyaml", marker = "python_full_version == '3.10.*'" },
     { name = "rich", marker = "python_full_version == '3.10.*'" },
     { name = "tensorstore", version = "0.1.78", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
@@ -1059,45 +917,27 @@ wheels = [
 
 [[package]]
 name = "flax"
-version = "0.12.0"
+version = "0.12.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "msgpack", marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "optax", version = "0.2.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "orbax-checkpoint", version = "0.11.25", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "orbax-checkpoint", version = "0.11.28", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "pyyaml", marker = "python_full_version >= '3.11'" },
     { name = "rich", marker = "python_full_version >= '3.11'" },
-    { name = "tensorstore", version = "0.1.78", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "tensorstore", version = "0.1.79", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "treescope", marker = "python_full_version >= '3.11'" },
     { name = "typing-extensions", marker = "python_full_version >= '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/8b/02/2d4fbc31dcf5fb02e2b8ce733f8eecaffc7f24e905df2da39539809b195b/flax-0.12.0.tar.gz", hash = "sha256:cb3d4a2028666640c1a2e5b267dadfbd42ef14c6db41896fb4c765a7f05ebe74", size = 5038989, upload-time = "2025-09-25T23:59:00.919Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/e6/cf/d118be069d0fbbfbb6c5591069233678b573e49d81e2a387afb0297528a0/flax-0.12.0-py3-none-any.whl", hash = "sha256:13dee5f2658e8b51e22cef54b788ddbad1b6e14a9d7c5b6d1ae9c3a0110d645e", size = 466282, upload-time = "2025-09-25T23:58:58.631Z" },
-]
-
-[[package]]
-name = "frozendict"
-version = "2.4.6"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/bb/59/19eb300ba28e7547538bdf603f1c6c34793240a90e1a7b61b65d8517e35e/frozendict-2.4.6.tar.gz", hash = "sha256:df7cd16470fbd26fc4969a208efadc46319334eb97def1ddf48919b351192b8e", size = 316416, upload-time = "2024-10-13T12:15:32.449Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/fb/9c/4f8860d0baf9014c66d68e8f7ca491126aa5f7977fa21c269dc99c693cb0/flax-0.12.1.tar.gz", hash = "sha256:ce2229a2ecb689678ba619e822346447c173c2ac1635e674e8249b97ed816cab", size = 5021765, upload-time = "2025-11-19T21:56:34.542Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ac/f1/a10be024a9d53441c997b3661ea80ecba6e3130adc53812a4b95b607cdd1/frozendict-2.4.6-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f4c789fd70879ccb6289a603cdebdc4953e7e5dea047d30c1b180529b28257b5", size = 117656, upload-time = "2024-10-13T12:14:22.038Z" },
-    { url = "https://files.pythonhosted.org/packages/46/a6/34c760975e6f1cb4db59a990d58dcf22287e10241c851804670c74c6a27a/frozendict-2.4.6-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:da6a10164c8a50b34b9ab508a9420df38f4edf286b9ca7b7df8a91767baecb34", size = 117444, upload-time = "2024-10-13T12:14:24.251Z" },
-    { url = "https://files.pythonhosted.org/packages/62/dd/64bddd1ffa9617f50e7e63656b2a7ad7f0a46c86b5f4a3d2c714d0006277/frozendict-2.4.6-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:9a8a43036754a941601635ea9c788ebd7a7efbed2becba01b54a887b41b175b9", size = 116801, upload-time = "2024-10-13T12:14:26.518Z" },
-    { url = "https://files.pythonhosted.org/packages/45/ae/af06a8bde1947277aad895c2f26c3b8b8b6ee9c0c2ad988fb58a9d1dde3f/frozendict-2.4.6-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:c9905dcf7aa659e6a11b8051114c9fa76dfde3a6e50e6dc129d5aece75b449a2", size = 117329, upload-time = "2024-10-13T12:14:28.485Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/a2/5a178339345edff643240e48dd276581df64b1dd93eaa7d26556396a145b/frozendict-2.4.6-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:669237c571856be575eca28a69e92a3d18f8490511eff184937283dc6093bd67", size = 117093, upload-time = "2024-10-13T12:15:15.621Z" },
-    { url = "https://files.pythonhosted.org/packages/41/df/09a752239eb0661eeda0f34f14577c10edc6f3e4deb7652b3a3efff22ad4/frozendict-2.4.6-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0aaa11e7c472150efe65adbcd6c17ac0f586896096ab3963775e1c5c58ac0098", size = 116883, upload-time = "2024-10-13T12:15:17.521Z" },
-    { url = "https://files.pythonhosted.org/packages/22/d4/619d1cfbc74be5641d839a5a2e292f9eac494aa557bfe7c266542c4014a2/frozendict-2.4.6-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:b8f2829048f29fe115da4a60409be2130e69402e29029339663fac39c90e6e2b", size = 116314, upload-time = "2024-10-13T12:15:19.689Z" },
-    { url = "https://files.pythonhosted.org/packages/41/b9/40042606a4ac458046ebeecc34cec2971e78e029ea3b6ad4e35833c7f8e6/frozendict-2.4.6-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:94321e646cc39bebc66954a31edd1847d3a2a3483cf52ff051cd0996e7db07db", size = 117017, upload-time = "2024-10-13T12:15:21.718Z" },
-    { url = "https://files.pythonhosted.org/packages/04/13/d9839089b900fa7b479cce495d62110cddc4bd5630a04d8469916c0e79c5/frozendict-2.4.6-py311-none-any.whl", hash = "sha256:d065db6a44db2e2375c23eac816f1a022feb2fa98cbb50df44a9e83700accbea", size = 16148, upload-time = "2024-10-13T12:15:26.839Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/d0/d482c39cee2ab2978a892558cf130681d4574ea208e162da8958b31e9250/frozendict-2.4.6-py312-none-any.whl", hash = "sha256:49344abe90fb75f0f9fdefe6d4ef6d4894e640fadab71f11009d52ad97f370b9", size = 16146, upload-time = "2024-10-13T12:15:28.16Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/c2/5157167a7ba03d4ca65bb1e7d830b31f7d45ee26dcc786786694ce47582e/flax-0.12.1-py3-none-any.whl", hash = "sha256:5e4d00d77e8a0d8a52eaab9d1837b2206dcc2f068feedd2d0b0187dabaad5d0a", size = 488158, upload-time = "2025-11-19T21:56:33.009Z" },
 ]
 
 [[package]]
@@ -1173,13 +1013,38 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/9a/9a/e35b4a917281c0b8419d4207f4334c8e8c5dbf4f3f5f9ada73958d937dcc/frozenlist-1.8.0-py3-none-any.whl", hash = "sha256:0c18a16eab41e82c295618a77502e17b195883241c563b00f0aa5106fc4eaa0d", size = 13409, upload-time = "2025-10-06T05:38:16.721Z" },
 ]
 
+[[package]]
+name = "fschat"
+version = "0.2.36"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "aiohttp" },
+    { name = "fastapi" },
+    { name = "httpx" },
+    { name = "markdown2", extra = ["all"] },
+    { name = "nh3" },
+    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "prompt-toolkit" },
+    { name = "pydantic" },
+    { name = "requests" },
+    { name = "rich" },
+    { name = "shortuuid" },
+    { name = "tiktoken" },
+    { name = "uvicorn" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/22/ad/8336e2dc6323944d59e888d41ff8833993ebc9c79bdd8d07997c54d92d55/fschat-0.2.36.tar.gz", hash = "sha256:ac724458bb105cb5b3b9ec852150345dfa7f61fd7c275d6635d66fa1ed904d90", size = 195280, upload-time = "2024-02-11T15:42:27.02Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/43/cd/3651634ab25f74e64eaf6a6e08e359320d6b26e637afe622762fe99d127b/fschat-0.2.36-py3-none-any.whl", hash = "sha256:a66ecf1e90f6db1d015096f6aa8a305d20d0a13e067951285aecdb56a33e9cb1", size = 256902, upload-time = "2024-02-11T15:42:24.819Z" },
+]
+
 [[package]]
 name = "fsspec"
-version = "2025.9.0"
+version = "2025.10.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/de/e0/bab50af11c2d75c9c4a2a26a5254573c0bd97cea152254401510950486fa/fsspec-2025.9.0.tar.gz", hash = "sha256:19fd429483d25d28b65ec68f9f4adc16c17ea2c7c7bf54ec61360d478fb19c19", size = 304847, upload-time = "2025-09-02T19:10:49.215Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/24/7f/2747c0d332b9acfa75dc84447a066fdf812b5a6b8d30472b74d309bfe8cb/fsspec-2025.10.0.tar.gz", hash = "sha256:b6789427626f068f9a83ca4e8a3cc050850b6c0f71f99ddb4f542b8266a26a59", size = 309285, upload-time = "2025-10-30T14:58:44.036Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/47/71/70db47e4f6ce3e5c37a607355f80da8860a33226be640226ac52cb05ef2e/fsspec-2025.9.0-py3-none-any.whl", hash = "sha256:530dc2a2af60a414a832059574df4a6e10cce927f6f4a78209390fe38955cfb7", size = 199289, upload-time = "2025-09-02T19:10:47.708Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/02/a6b21098b1d5d6249b7c5ab69dde30108a71e4e819d4a9778f1de1d5b70d/fsspec-2025.10.0-py3-none-any.whl", hash = "sha256:7c7712353ae7d875407f97715f0e1ffcc21e33d5b24556cb1e090ae9409ec61d", size = 200966, upload-time = "2025-10-30T14:58:42.53Z" },
 ]
 
 [package.optional-dependencies]
@@ -1205,21 +1070,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f8/5c/e226de133afd8bb267ec27eead9ae3d784b95b39a287ed404caab39a5f50/genson-1.3.0-py3-none-any.whl", hash = "sha256:468feccd00274cc7e4c09e84b08704270ba8d95232aa280f65b986139cec67f7", size = 21470, upload-time = "2024-05-15T22:08:47.056Z" },
 ]
 
-[[package]]
-name = "gguf"
-version = "0.17.1"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "pyyaml" },
-    { name = "tqdm" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/08/08/7de1ca4b71e7bf33b547f82bb22505e221b5fa42f67d635e200e0ad22ad6/gguf-0.17.1.tar.gz", hash = "sha256:36ad71aad900a3e75fc94ebe96ea6029f03a4e44be7627ef7ad3d03e8c7bcb53", size = 89338, upload-time = "2025-06-19T14:00:33.705Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/fc/31/6a93a887617ee7deeaa602ca3d02d1c12a6cb8a742a695de5d128f5fa46a/gguf-0.17.1-py3-none-any.whl", hash = "sha256:7bc5aa7eeb1931f7d39b48fdc5b38fda6b294b9dca75cf607ac69557840a3943", size = 96224, upload-time = "2025-06-19T14:00:32.88Z" },
-]
-
 [[package]]
 name = "ghp-import"
 version = "2.1.0"
@@ -1234,35 +1084,62 @@ wheels = [
 
 [[package]]
 name = "google-auth"
-version = "2.41.1"
+version = "2.43.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "cachetools" },
     { name = "pyasn1-modules" },
     { name = "rsa" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a8/af/5129ce5b2f9688d2fa49b463e544972a7c82b0fdb50980dafee92e121d9f/google_auth-2.41.1.tar.gz", hash = "sha256:b76b7b1f9e61f0cb7e88870d14f6a94aeef248959ef6992670efee37709cbfd2", size = 292284, upload-time = "2025-09-30T22:51:26.363Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/ff/ef/66d14cf0e01b08d2d51ffc3c20410c4e134a1548fc246a6081eae585a4fe/google_auth-2.43.0.tar.gz", hash = "sha256:88228eee5fc21b62a1b5fe773ca15e67778cb07dc8363adcb4a8827b52d81483", size = 296359, upload-time = "2025-11-06T00:13:36.587Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/be/a4/7319a2a8add4cc352be9e3efeff5e2aacee917c85ca2fa1647e29089983c/google_auth-2.41.1-py2.py3-none-any.whl", hash = "sha256:754843be95575b9a19c604a848a41be03f7f2afd8c019f716dc1f51ee41c639d", size = 221302, upload-time = "2025-09-30T22:51:24.212Z" },
+    { url = "https://files.pythonhosted.org/packages/6f/d1/385110a9ae86d91cc14c5282c61fe9f4dc41c0b9f7d423c6ad77038c4448/google_auth-2.43.0-py2.py3-none-any.whl", hash = "sha256:af628ba6fa493f75c7e9dbe9373d148ca9f4399b5ea29976519e0a3848eddd16", size = 223114, upload-time = "2025-11-06T00:13:35.209Z" },
 ]
 
 [[package]]
 name = "google-genai"
-version = "1.41.0"
+version = "1.47.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "anyio" },
-    { name = "google-auth" },
-    { name = "httpx" },
-    { name = "pydantic" },
-    { name = "requests" },
-    { name = "tenacity" },
-    { name = "typing-extensions" },
-    { name = "websockets" },
+    { name = "anyio", marker = "python_full_version < '3.10'" },
+    { name = "google-auth", marker = "python_full_version < '3.10'" },
+    { name = "httpx", marker = "python_full_version < '3.10'" },
+    { name = "pydantic", marker = "python_full_version < '3.10'" },
+    { name = "requests", marker = "python_full_version < '3.10'" },
+    { name = "tenacity", marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+    { name = "websockets", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/9f/97/784fba9bc6c41263ff90cb9063eadfdd755dde79cfa5a8d0e397b067dcf9/google_genai-1.47.0.tar.gz", hash = "sha256:ecece00d0a04e6739ea76cc8dad82ec9593d9380aaabef078990e60574e5bf59", size = 241471, upload-time = "2025-10-29T22:01:02.88Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/89/ef/e080e8d67c270ea320956bb911a9359664fc46d3b87d1f029decd33e5c4c/google_genai-1.47.0-py3-none-any.whl", hash = "sha256:e3851237556cbdec96007d8028b4b1f2425cdc5c099a8dc36b72a57e42821b60", size = 241506, upload-time = "2025-10-29T22:01:00.982Z" },
+]
+
+[[package]]
+name = "google-genai"
+version = "1.52.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/72/8b/ee20bcf707769b3b0e1106c3b5c811507736af7e8a60f29a70af1750ba19/google_genai-1.41.0.tar.gz", hash = "sha256:134f861bb0ace4e34af0501ecb75ceee15f7662fd8120698cd185e8cb39f2800", size = 245812, upload-time = "2025-10-02T22:30:29.699Z" }
+dependencies = [
+    { name = "anyio", marker = "python_full_version >= '3.10'" },
+    { name = "google-auth", marker = "python_full_version >= '3.10'" },
+    { name = "httpx", marker = "python_full_version >= '3.10'" },
+    { name = "pydantic", marker = "python_full_version >= '3.10'" },
+    { name = "requests", marker = "python_full_version >= '3.10'" },
+    { name = "tenacity", marker = "python_full_version >= '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
+    { name = "websockets", marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/09/4e/0ad8585d05312074bb69711b2d81cfed69ce0ae441913d57bf169bed20a7/google_genai-1.52.0.tar.gz", hash = "sha256:a74e8a4b3025f23aa98d6a0f84783119012ca6c336fd68f73c5d2b11465d7fc5", size = 258743, upload-time = "2025-11-21T02:18:55.742Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/15/14/e5e8fbca8863fee718208566c4e927b8e9f45fd46ec5cf89e24759da545b/google_genai-1.41.0-py3-none-any.whl", hash = "sha256:111a3ee64c1a0927d3879faddb368234594432479a40c311e5fe4db338ca8778", size = 245931, upload-time = "2025-10-02T22:30:27.885Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/66/03f663e7bca7abe9ccfebe6cb3fe7da9a118fd723a5abb278d6117e7990e/google_genai-1.52.0-py3-none-any.whl", hash = "sha256:c8352b9f065ae14b9322b949c7debab8562982f03bf71d44130cd2b798c20743", size = 261219, upload-time = "2025-11-21T02:18:54.515Z" },
 ]
 
 [[package]]
@@ -1279,53 +1156,53 @@ wheels = [
 
 [[package]]
 name = "grpcio"
-version = "1.75.1"
+version = "1.76.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/9d/f7/8963848164c7604efb3a3e6ee457fdb3a469653e19002bd24742473254f8/grpcio-1.75.1.tar.gz", hash = "sha256:3e81d89ece99b9ace23a6916880baca613c03a799925afb2857887efa8b1b3d2", size = 12731327, upload-time = "2025-09-26T09:03:36.887Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/51/57/89fd829fb00a6d0bee3fbcb2c8a7aa0252d908949b6ab58bfae99d39d77e/grpcio-1.75.1-cp310-cp310-linux_armv7l.whl", hash = "sha256:1712b5890b22547dd29f3215c5788d8fc759ce6dd0b85a6ba6e2731f2d04c088", size = 5705534, upload-time = "2025-09-26T09:00:52.225Z" },
-    { url = "https://files.pythonhosted.org/packages/76/dd/2f8536e092551cf804e96bcda79ecfbc51560b214a0f5b7ebc253f0d4664/grpcio-1.75.1-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:8d04e101bba4b55cea9954e4aa71c24153ba6182481b487ff376da28d4ba46cf", size = 11484103, upload-time = "2025-09-26T09:00:59.457Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/3d/affe2fb897804c98d56361138e73786af8f4dd876b9d9851cfe6342b53c8/grpcio-1.75.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:683cfc70be0c1383449097cba637317e4737a357cfc185d887fd984206380403", size = 6289953, upload-time = "2025-09-26T09:01:03.699Z" },
-    { url = "https://files.pythonhosted.org/packages/87/aa/0f40b7f47a0ff10d7e482bc3af22dac767c7ff27205915f08962d5ca87a2/grpcio-1.75.1-cp310-cp310-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:491444c081a54dcd5e6ada57314321ae526377f498d4aa09d975c3241c5b9e1c", size = 6949785, upload-time = "2025-09-26T09:01:07.504Z" },
-    { url = "https://files.pythonhosted.org/packages/a5/45/b04407e44050781821c84f26df71b3f7bc469923f92f9f8bc27f1406dbcc/grpcio-1.75.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ce08d4e112d0d38487c2b631ec8723deac9bc404e9c7b1011426af50a79999e4", size = 6465708, upload-time = "2025-09-26T09:01:11.028Z" },
-    { url = "https://files.pythonhosted.org/packages/09/3e/4ae3ec0a4d20dcaafbb6e597defcde06399ccdc5b342f607323f3b47f0a3/grpcio-1.75.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:5a2acda37fc926ccc4547977ac3e56b1df48fe200de968e8c8421f6e3093df6c", size = 7100912, upload-time = "2025-09-26T09:01:14.393Z" },
-    { url = "https://files.pythonhosted.org/packages/34/3f/a9085dab5c313bb0cb853f222d095e2477b9b8490a03634cdd8d19daa5c3/grpcio-1.75.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:745c5fe6bf05df6a04bf2d11552c7d867a2690759e7ab6b05c318a772739bd75", size = 8042497, upload-time = "2025-09-26T09:01:17.759Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/87/ea54eba931ab9ed3f999ba95f5d8d01a20221b664725bab2fe93e3dee848/grpcio-1.75.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:259526a7159d39e2db40d566fe3e8f8e034d0fb2db5bf9c00e09aace655a4c2b", size = 7493284, upload-time = "2025-09-26T09:01:20.896Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/5e/287f1bf1a998f4ac46ef45d518de3b5da08b4e86c7cb5e1108cee30b0282/grpcio-1.75.1-cp310-cp310-win32.whl", hash = "sha256:f4b29b9aabe33fed5df0a85e5f13b09ff25e2c05bd5946d25270a8bd5682dac9", size = 3950809, upload-time = "2025-09-26T09:01:23.695Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/a2/3cbfc06a4ec160dc77403b29ecb5cf76ae329eb63204fea6a7c715f1dfdb/grpcio-1.75.1-cp310-cp310-win_amd64.whl", hash = "sha256:cf2e760978dcce7ff7d465cbc7e276c3157eedc4c27aa6de7b594c7a295d3d61", size = 4644704, upload-time = "2025-09-26T09:01:25.763Z" },
-    { url = "https://files.pythonhosted.org/packages/0c/3c/35ca9747473a306bfad0cee04504953f7098527cd112a4ab55c55af9e7bd/grpcio-1.75.1-cp311-cp311-linux_armv7l.whl", hash = "sha256:573855ca2e58e35032aff30bfbd1ee103fbcf4472e4b28d4010757700918e326", size = 5709761, upload-time = "2025-09-26T09:01:28.528Z" },
-    { url = "https://files.pythonhosted.org/packages/c9/2c/ecbcb4241e4edbe85ac2663f885726fea0e947767401288b50d8fdcb9200/grpcio-1.75.1-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:6a4996a2c8accc37976dc142d5991adf60733e223e5c9a2219e157dc6a8fd3a2", size = 11496691, upload-time = "2025-09-26T09:01:31.214Z" },
-    { url = "https://files.pythonhosted.org/packages/81/40/bc07aee2911f0d426fa53fe636216100c31a8ea65a400894f280274cb023/grpcio-1.75.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:b1ea1bbe77ecbc1be00af2769f4ae4a88ce93be57a4f3eebd91087898ed749f9", size = 6296084, upload-time = "2025-09-26T09:01:34.596Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/d1/10c067f6c67396cbf46448b80f27583b5e8c4b46cdfbe18a2a02c2c2f290/grpcio-1.75.1-cp311-cp311-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:e5b425aee54cc5e3e3c58f00731e8a33f5567965d478d516d35ef99fd648ab68", size = 6950403, upload-time = "2025-09-26T09:01:36.736Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/42/5f628abe360b84dfe8dd8f32be6b0606dc31dc04d3358eef27db791ea4d5/grpcio-1.75.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:0049a7bf547dafaeeb1db17079ce79596c298bfe308fc084d023c8907a845b9a", size = 6470166, upload-time = "2025-09-26T09:01:39.474Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/93/a24035080251324019882ee2265cfde642d6476c0cf8eb207fc693fcebdc/grpcio-1.75.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:5b8ea230c7f77c0a1a3208a04a1eda164633fb0767b4cefd65a01079b65e5b1f", size = 7107828, upload-time = "2025-09-26T09:01:41.782Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/f8/d18b984c1c9ba0318e3628dbbeb6af77a5007f02abc378c845070f2d3edd/grpcio-1.75.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:36990d629c3c9fb41e546414e5af52d0a7af37ce7113d9682c46d7e2919e4cca", size = 8045421, upload-time = "2025-09-26T09:01:45.835Z" },
-    { url = "https://files.pythonhosted.org/packages/7e/b6/4bf9aacff45deca5eac5562547ed212556b831064da77971a4e632917da3/grpcio-1.75.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:b10ad908118d38c2453ade7ff790e5bce36580c3742919007a2a78e3a1e521ca", size = 7503290, upload-time = "2025-09-26T09:01:49.28Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/15/d8d69d10223cb54c887a2180bd29fe5fa2aec1d4995c8821f7aa6eaf72e4/grpcio-1.75.1-cp311-cp311-win32.whl", hash = "sha256:d6be2b5ee7bea656c954dcf6aa8093c6f0e6a3ef9945c99d99fcbfc88c5c0bfe", size = 3950631, upload-time = "2025-09-26T09:01:51.23Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/40/7b8642d45fff6f83300c24eaac0380a840e5e7fe0e8d80afd31b99d7134e/grpcio-1.75.1-cp311-cp311-win_amd64.whl", hash = "sha256:61c692fb05956b17dd6d1ab480f7f10ad0536dba3bc8fd4e3c7263dc244ed772", size = 4646131, upload-time = "2025-09-26T09:01:53.266Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/81/42be79e73a50aaa20af66731c2defeb0e8c9008d9935a64dd8ea8e8c44eb/grpcio-1.75.1-cp312-cp312-linux_armv7l.whl", hash = "sha256:7b888b33cd14085d86176b1628ad2fcbff94cfbbe7809465097aa0132e58b018", size = 5668314, upload-time = "2025-09-26T09:01:55.424Z" },
-    { url = "https://files.pythonhosted.org/packages/c5/a7/3686ed15822fedc58c22f82b3a7403d9faf38d7c33de46d4de6f06e49426/grpcio-1.75.1-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:8775036efe4ad2085975531d221535329f5dac99b6c2a854a995456098f99546", size = 11476125, upload-time = "2025-09-26T09:01:57.927Z" },
-    { url = "https://files.pythonhosted.org/packages/14/85/21c71d674f03345ab183c634ecd889d3330177e27baea8d5d247a89b6442/grpcio-1.75.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:bb658f703468d7fbb5dcc4037c65391b7dc34f808ac46ed9136c24fc5eeb041d", size = 6246335, upload-time = "2025-09-26T09:02:00.76Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/db/3beb661bc56a385ae4fa6b0e70f6b91ac99d47afb726fe76aaff87ebb116/grpcio-1.75.1-cp312-cp312-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:4b7177a1cdb3c51b02b0c0a256b0a72fdab719600a693e0e9037949efffb200b", size = 6916309, upload-time = "2025-09-26T09:02:02.894Z" },
-    { url = "https://files.pythonhosted.org/packages/1e/9c/eda9fe57f2b84343d44c1b66cf3831c973ba29b078b16a27d4587a1fdd47/grpcio-1.75.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:7d4fa6ccc3ec2e68a04f7b883d354d7fea22a34c44ce535a2f0c0049cf626ddf", size = 6435419, upload-time = "2025-09-26T09:02:05.055Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/b8/090c98983e0a9d602e3f919a6e2d4e470a8b489452905f9a0fa472cac059/grpcio-1.75.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3d86880ecaeb5b2f0a8afa63824de93adb8ebe4e49d0e51442532f4e08add7d6", size = 7064893, upload-time = "2025-09-26T09:02:07.275Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/c0/6d53d4dbbd00f8bd81571f5478d8a95528b716e0eddb4217cc7cb45aae5f/grpcio-1.75.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:a8041d2f9e8a742aeae96f4b047ee44e73619f4f9d24565e84d5446c623673b6", size = 8011922, upload-time = "2025-09-26T09:02:09.527Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/7c/48455b2d0c5949678d6982c3e31ea4d89df4e16131b03f7d5c590811cbe9/grpcio-1.75.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:3652516048bf4c314ce12be37423c79829f46efffb390ad64149a10c6071e8de", size = 7466181, upload-time = "2025-09-26T09:02:12.279Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/12/04a0e79081e3170b6124f8cba9b6275871276be06c156ef981033f691880/grpcio-1.75.1-cp312-cp312-win32.whl", hash = "sha256:44b62345d8403975513af88da2f3d5cc76f73ca538ba46596f92a127c2aea945", size = 3938543, upload-time = "2025-09-26T09:02:14.77Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/d7/11350d9d7fb5adc73d2b0ebf6ac1cc70135577701e607407fe6739a90021/grpcio-1.75.1-cp312-cp312-win_amd64.whl", hash = "sha256:b1e191c5c465fa777d4cafbaacf0c01e0d5278022082c0abbd2ee1d6454ed94d", size = 4641938, upload-time = "2025-09-26T09:02:16.927Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/e2/33efd823a879dc7b60c10192df1900ee5c200f8e782663a41a3b2aecd143/grpcio-1.75.1-cp39-cp39-linux_armv7l.whl", hash = "sha256:c09fba33327c3ac11b5c33dbdd8218eef8990d78f83b1656d628831812a8c0fb", size = 5706679, upload-time = "2025-09-26T09:03:10.218Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/13/17e39ee4897f1cd12dd463e863b830e64643b13e9a4af5062b4a6f0790be/grpcio-1.75.1-cp39-cp39-macosx_11_0_universal2.whl", hash = "sha256:7e21400b037be29545704889e72e586c238e346dcb2d08d8a7288d16c883a9ec", size = 11490271, upload-time = "2025-09-26T09:03:12.778Z" },
-    { url = "https://files.pythonhosted.org/packages/77/90/b80e75f8cce758425b2772742eed4e9db765a965d902ba4b7f239b2513de/grpcio-1.75.1-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:c12121e509b9f8b0914d10054d24120237d19e870b1cd82acbb8a9b9ddd198a3", size = 6291926, upload-time = "2025-09-26T09:03:16.282Z" },
-    { url = "https://files.pythonhosted.org/packages/40/5f/e6033d8f99063350e20873a46225468b73045b9ef2c8cba73d66a87c3fd5/grpcio-1.75.1-cp39-cp39-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:73577a93e692b3474b1bfe84285d098de36705dbd838bb4d6a056d326e4dc880", size = 6950040, upload-time = "2025-09-26T09:03:18.874Z" },
-    { url = "https://files.pythonhosted.org/packages/01/12/34076c079b45af5aed40f037fffe388d7fbe90dd539ed01e4744c926d227/grpcio-1.75.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e19e7dfa0d7ca7dea22be464339e18ac608fd75d88c56770c646cdabe54bc724", size = 6465780, upload-time = "2025-09-26T09:03:21.219Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/c5/ee6fd69a9f6e7288d04da010ad7480a0566d2aac81097ff4dafbc5ffa9b6/grpcio-1.75.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:4e1c28f51c1cf67eccdfc1065e8e866c9ed622f09773ca60947089c117f848a1", size = 7098308, upload-time = "2025-09-26T09:03:23.875Z" },
-    { url = "https://files.pythonhosted.org/packages/78/32/f2be13f13035361768923159fe20470a7d22db2c7c692b952e21284f56e5/grpcio-1.75.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:030a6164bc2ca726052778c0cf8e3249617a34e368354f9e6107c27ad4af8c28", size = 8042268, upload-time = "2025-09-26T09:03:26.268Z" },
-    { url = "https://files.pythonhosted.org/packages/e7/2d/1bb0572f0a2eaab100b4635c6c2cd0d37e3cda5554037e3f90b1bc428d56/grpcio-1.75.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:67697efef5a98d46d5db7b1720fa4043536f8b8e5072a5d61cfca762f287e939", size = 7491470, upload-time = "2025-09-26T09:03:28.906Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/e0/1e962dcb64019bbd87eedcfacdedb83af0f66da01f2f6e03d69b0aa1b7f0/grpcio-1.75.1-cp39-cp39-win32.whl", hash = "sha256:52015cf73eb5d76f6404e0ce0505a69b51fd1f35810b3a01233b34b10baafb41", size = 3951697, upload-time = "2025-09-26T09:03:31.535Z" },
-    { url = "https://files.pythonhosted.org/packages/87/bc/47fb3aaa77e7d657999937ec1026beba9e37f3199599fe510f762d31da97/grpcio-1.75.1-cp39-cp39-win_amd64.whl", hash = "sha256:9fe51e4a1f896ea84ac750900eae34d9e9b896b5b1e4a30b02dc31ad29f36383", size = 4645764, upload-time = "2025-09-26T09:03:34.071Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/b6/e0/318c1ce3ae5a17894d5791e87aea147587c9e702f24122cc7a5c8bbaeeb1/grpcio-1.76.0.tar.gz", hash = "sha256:7be78388d6da1a25c0d5ec506523db58b18be22d9c37d8d3a32c08be4987bd73", size = 12785182, upload-time = "2025-10-21T16:23:12.106Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/88/17/ff4795dc9a34b6aee6ec379f1b66438a3789cd1315aac0cbab60d92f74b3/grpcio-1.76.0-cp310-cp310-linux_armv7l.whl", hash = "sha256:65a20de41e85648e00305c1bb09a3598f840422e522277641145a32d42dcefcc", size = 5840037, upload-time = "2025-10-21T16:20:25.069Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/ff/35f9b96e3fa2f12e1dcd58a4513a2e2294a001d64dec81677361b7040c9a/grpcio-1.76.0-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:40ad3afe81676fd9ec6d9d406eda00933f218038433980aa19d401490e46ecde", size = 11836482, upload-time = "2025-10-21T16:20:30.113Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/1c/8374990f9545e99462caacea5413ed783014b3b66ace49e35c533f07507b/grpcio-1.76.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:035d90bc79eaa4bed83f524331d55e35820725c9fbb00ffa1904d5550ed7ede3", size = 6407178, upload-time = "2025-10-21T16:20:32.733Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/77/36fd7d7c75a6c12542c90a6d647a27935a1ecaad03e0ffdb7c42db6b04d2/grpcio-1.76.0-cp310-cp310-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:4215d3a102bd95e2e11b5395c78562967959824156af11fa93d18fdd18050990", size = 7075684, upload-time = "2025-10-21T16:20:35.435Z" },
+    { url = "https://files.pythonhosted.org/packages/38/f7/e3cdb252492278e004722306c5a8935eae91e64ea11f0af3437a7de2e2b7/grpcio-1.76.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:49ce47231818806067aea3324d4bf13825b658ad662d3b25fada0bdad9b8a6af", size = 6611133, upload-time = "2025-10-21T16:20:37.541Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/20/340db7af162ccd20a0893b5f3c4a5d676af7b71105517e62279b5b61d95a/grpcio-1.76.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:8cc3309d8e08fd79089e13ed4819d0af72aa935dd8f435a195fd152796752ff2", size = 7195507, upload-time = "2025-10-21T16:20:39.643Z" },
+    { url = "https://files.pythonhosted.org/packages/10/f0/b2160addc1487bd8fa4810857a27132fb4ce35c1b330c2f3ac45d697b106/grpcio-1.76.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:971fd5a1d6e62e00d945423a567e42eb1fa678ba89072832185ca836a94daaa6", size = 8160651, upload-time = "2025-10-21T16:20:42.492Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/2c/ac6f98aa113c6ef111b3f347854e99ebb7fb9d8f7bb3af1491d438f62af4/grpcio-1.76.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:9d9adda641db7207e800a7f089068f6f645959f2df27e870ee81d44701dd9db3", size = 7620568, upload-time = "2025-10-21T16:20:45.995Z" },
+    { url = "https://files.pythonhosted.org/packages/90/84/7852f7e087285e3ac17a2703bc4129fafee52d77c6c82af97d905566857e/grpcio-1.76.0-cp310-cp310-win32.whl", hash = "sha256:063065249d9e7e0782d03d2bca50787f53bd0fb89a67de9a7b521c4a01f1989b", size = 3998879, upload-time = "2025-10-21T16:20:48.592Z" },
+    { url = "https://files.pythonhosted.org/packages/10/30/d3d2adcbb6dd3ff59d6ac3df6ef830e02b437fb5c90990429fd180e52f30/grpcio-1.76.0-cp310-cp310-win_amd64.whl", hash = "sha256:a6ae758eb08088d36812dd5d9af7a9859c05b1e0f714470ea243694b49278e7b", size = 4706892, upload-time = "2025-10-21T16:20:50.697Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/00/8163a1beeb6971f66b4bbe6ac9457b97948beba8dd2fc8e1281dce7f79ec/grpcio-1.76.0-cp311-cp311-linux_armv7l.whl", hash = "sha256:2e1743fbd7f5fa713a1b0a8ac8ebabf0ec980b5d8809ec358d488e273b9cf02a", size = 5843567, upload-time = "2025-10-21T16:20:52.829Z" },
+    { url = "https://files.pythonhosted.org/packages/10/c1/934202f5cf335e6d852530ce14ddb0fef21be612ba9ecbbcbd4d748ca32d/grpcio-1.76.0-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:a8c2cf1209497cf659a667d7dea88985e834c24b7c3b605e6254cbb5076d985c", size = 11848017, upload-time = "2025-10-21T16:20:56.705Z" },
+    { url = "https://files.pythonhosted.org/packages/11/0b/8dec16b1863d74af6eb3543928600ec2195af49ca58b16334972f6775663/grpcio-1.76.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:08caea849a9d3c71a542827d6df9d5a69067b0a1efbea8a855633ff5d9571465", size = 6412027, upload-time = "2025-10-21T16:20:59.3Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/64/7b9e6e7ab910bea9d46f2c090380bab274a0b91fb0a2fe9b0cd399fffa12/grpcio-1.76.0-cp311-cp311-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:f0e34c2079d47ae9f6188211db9e777c619a21d4faba6977774e8fa43b085e48", size = 7075913, upload-time = "2025-10-21T16:21:01.645Z" },
+    { url = "https://files.pythonhosted.org/packages/68/86/093c46e9546073cefa789bd76d44c5cb2abc824ca62af0c18be590ff13ba/grpcio-1.76.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:8843114c0cfce61b40ad48df65abcfc00d4dba82eae8718fab5352390848c5da", size = 6615417, upload-time = "2025-10-21T16:21:03.844Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/b6/5709a3a68500a9c03da6fb71740dcdd5ef245e39266461a03f31a57036d8/grpcio-1.76.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8eddfb4d203a237da6f3cc8a540dad0517d274b5a1e9e636fd8d2c79b5c1d397", size = 7199683, upload-time = "2025-10-21T16:21:06.195Z" },
+    { url = "https://files.pythonhosted.org/packages/91/d3/4b1f2bf16ed52ce0b508161df3a2d186e4935379a159a834cb4a7d687429/grpcio-1.76.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:32483fe2aab2c3794101c2a159070584e5db11d0aa091b2c0ea9c4fc43d0d749", size = 8163109, upload-time = "2025-10-21T16:21:08.498Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/61/d9043f95f5f4cf085ac5dd6137b469d41befb04bd80280952ffa2a4c3f12/grpcio-1.76.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:dcfe41187da8992c5f40aa8c5ec086fa3672834d2be57a32384c08d5a05b4c00", size = 7626676, upload-time = "2025-10-21T16:21:10.693Z" },
+    { url = "https://files.pythonhosted.org/packages/36/95/fd9a5152ca02d8881e4dd419cdd790e11805979f499a2e5b96488b85cf27/grpcio-1.76.0-cp311-cp311-win32.whl", hash = "sha256:2107b0c024d1b35f4083f11245c0e23846ae64d02f40b2b226684840260ed054", size = 3997688, upload-time = "2025-10-21T16:21:12.746Z" },
+    { url = "https://files.pythonhosted.org/packages/60/9c/5c359c8d4c9176cfa3c61ecd4efe5affe1f38d9bae81e81ac7186b4c9cc8/grpcio-1.76.0-cp311-cp311-win_amd64.whl", hash = "sha256:522175aba7af9113c48ec10cc471b9b9bd4f6ceb36aeb4544a8e2c80ed9d252d", size = 4709315, upload-time = "2025-10-21T16:21:15.26Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/05/8e29121994b8d959ffa0afd28996d452f291b48cfc0875619de0bde2c50c/grpcio-1.76.0-cp312-cp312-linux_armv7l.whl", hash = "sha256:81fd9652b37b36f16138611c7e884eb82e0cec137c40d3ef7c3f9b3ed00f6ed8", size = 5799718, upload-time = "2025-10-21T16:21:17.939Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/75/11d0e66b3cdf998c996489581bdad8900db79ebd83513e45c19548f1cba4/grpcio-1.76.0-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:04bbe1bfe3a68bbfd4e52402ab7d4eb59d72d02647ae2042204326cf4bbad280", size = 11825627, upload-time = "2025-10-21T16:21:20.466Z" },
+    { url = "https://files.pythonhosted.org/packages/28/50/2f0aa0498bc188048f5d9504dcc5c2c24f2eb1a9337cd0fa09a61a2e75f0/grpcio-1.76.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d388087771c837cdb6515539f43b9d4bf0b0f23593a24054ac16f7a960be16f4", size = 6359167, upload-time = "2025-10-21T16:21:23.122Z" },
+    { url = "https://files.pythonhosted.org/packages/66/e5/bbf0bb97d29ede1d59d6588af40018cfc345b17ce979b7b45424628dc8bb/grpcio-1.76.0-cp312-cp312-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:9f8f757bebaaea112c00dba718fc0d3260052ce714e25804a03f93f5d1c6cc11", size = 7044267, upload-time = "2025-10-21T16:21:25.995Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/86/f6ec2164f743d9609691115ae8ece098c76b894ebe4f7c94a655c6b03e98/grpcio-1.76.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:980a846182ce88c4f2f7e2c22c56aefd515daeb36149d1c897f83cf57999e0b6", size = 6573963, upload-time = "2025-10-21T16:21:28.631Z" },
+    { url = "https://files.pythonhosted.org/packages/60/bc/8d9d0d8505feccfdf38a766d262c71e73639c165b311c9457208b56d92ae/grpcio-1.76.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:f92f88e6c033db65a5ae3d97905c8fea9c725b63e28d5a75cb73b49bda5024d8", size = 7164484, upload-time = "2025-10-21T16:21:30.837Z" },
+    { url = "https://files.pythonhosted.org/packages/67/e6/5d6c2fc10b95edf6df9b8f19cf10a34263b7fd48493936fffd5085521292/grpcio-1.76.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:4baf3cbe2f0be3289eb68ac8ae771156971848bb8aaff60bad42005539431980", size = 8127777, upload-time = "2025-10-21T16:21:33.577Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/c8/dce8ff21c86abe025efe304d9e31fdb0deaaa3b502b6a78141080f206da0/grpcio-1.76.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:615ba64c208aaceb5ec83bfdce7728b80bfeb8be97562944836a7a0a9647d882", size = 7594014, upload-time = "2025-10-21T16:21:41.882Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/42/ad28191ebf983a5d0ecef90bab66baa5a6b18f2bfdef9d0a63b1973d9f75/grpcio-1.76.0-cp312-cp312-win32.whl", hash = "sha256:45d59a649a82df5718fd9527ce775fd66d1af35e6d31abdcdc906a49c6822958", size = 3984750, upload-time = "2025-10-21T16:21:44.006Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/00/7bd478cbb851c04a48baccaa49b75abaa8e4122f7d86da797500cccdd771/grpcio-1.76.0-cp312-cp312-win_amd64.whl", hash = "sha256:c088e7a90b6017307f423efbb9d1ba97a22aa2170876223f9709e9d1de0b5347", size = 4704003, upload-time = "2025-10-21T16:21:46.244Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/d5/301e71c7d22a5c7aabf1953dd1106987bd47f883377d528355f898a850f2/grpcio-1.76.0-cp39-cp39-linux_armv7l.whl", hash = "sha256:8ebe63ee5f8fa4296b1b8cfc743f870d10e902ca18afc65c68cf46fd39bb0783", size = 5840371, upload-time = "2025-10-21T16:22:42.468Z" },
+    { url = "https://files.pythonhosted.org/packages/00/55/e3181adccff8808301dd9214b5e03c6db5a404b5ae8a6ec5768a5a65ed63/grpcio-1.76.0-cp39-cp39-macosx_11_0_universal2.whl", hash = "sha256:3bf0f392c0b806905ed174dcd8bdd5e418a40d5567a05615a030a5aeddea692d", size = 11840384, upload-time = "2025-10-21T16:22:45.508Z" },
+    { url = "https://files.pythonhosted.org/packages/65/36/db1dfe943bce7180f5b6d9be564366ca1024a005e914a1f10212c24a840b/grpcio-1.76.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:0b7604868b38c1bfd5cf72d768aedd7db41d78cb6a4a18585e33fb0f9f2363fd", size = 6408765, upload-time = "2025-10-21T16:22:48.761Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/79/a8452764aa4b5ca30a970e514ec2fc5cf75451571793f6b276b6807f67dc/grpcio-1.76.0-cp39-cp39-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:e6d1db20594d9daba22f90da738b1a0441a7427552cc6e2e3d1297aeddc00378", size = 7076220, upload-time = "2025-10-21T16:22:51.546Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/61/4cca38c4e7bb3ac5a1e0be6cf700a4dd85c61cbd8a9c5e076c224967084e/grpcio-1.76.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d099566accf23d21037f18a2a63d323075bebace807742e4b0ac210971d4dd70", size = 6610195, upload-time = "2025-10-21T16:22:54.688Z" },
+    { url = "https://files.pythonhosted.org/packages/54/3d/3f8bfae264c22c95fa702c35aa2a8105b754b4ace049c66a8b2230c97671/grpcio-1.76.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:ebea5cc3aa8ea72e04df9913492f9a96d9348db876f9dda3ad729cfedf7ac416", size = 7193343, upload-time = "2025-10-21T16:22:57.434Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/cd/89f9254782b6cd94aa7c93fde370862877113b7189fb49900eaf9a706c82/grpcio-1.76.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:0c37db8606c258e2ee0c56b78c62fc9dee0e901b5dbdcf816c2dd4ad652b8b0c", size = 8161922, upload-time = "2025-10-21T16:23:00.135Z" },
+    { url = "https://files.pythonhosted.org/packages/af/e0/99eb899d7cb9c676afea70ab6d02a72a9e6ce24d0300f625773fafe6d547/grpcio-1.76.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:ebebf83299b0cb1721a8859ea98f3a77811e35dce7609c5c963b9ad90728f886", size = 7617951, upload-time = "2025-10-21T16:23:03.68Z" },
+    { url = "https://files.pythonhosted.org/packages/79/26/dca1b2bfaa9981cc28fa995730c80eedb0b86c912c30d1b676f08232e6ab/grpcio-1.76.0-cp39-cp39-win32.whl", hash = "sha256:0aaa82d0813fd4c8e589fac9b65d7dd88702555f702fb10417f96e2a2a6d4c0f", size = 3999306, upload-time = "2025-10-21T16:23:06.187Z" },
+    { url = "https://files.pythonhosted.org/packages/de/d1/fb90564a981eedd3cd87dc6bfd7c249e8a515cfad1ed8e9af73be223cd3b/grpcio-1.76.0-cp39-cp39-win_amd64.whl", hash = "sha256:acab0277c40eff7143c2323190ea57b9ee5fd353d8190ee9652369fae735668a", size = 4708771, upload-time = "2025-10-21T16:23:08.902Z" },
 ]
 
 [[package]]
@@ -1341,9 +1218,11 @@ wheels = [
 name = "h5py"
 version = "3.14.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/5d/57/dfb3c5c3f1bf5f5ef2e59a22dec4ff1f3d7408b55bfcefcfb0ea69ef21c6/h5py-3.14.0.tar.gz", hash = "sha256:2372116b2e0d5d3e5e705b7f663f7c8d96fa79a4052d250484ef91d24d6a08f4", size = 424323, upload-time = "2025-06-06T14:06:15.01Z" }
 wheels = [
@@ -1369,19 +1248,58 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/b4/e4/fb8032d0e5480b1db9b419b5b50737b61bb3c7187c49d809975d62129fb0/h5py-3.14.0-cp39-cp39-win_amd64.whl", hash = "sha256:4f025cf30ae738c4c4e38c7439a761a71ccfcce04c2b87b2a2ac64e8c5171d43", size = 2877166, upload-time = "2025-06-06T14:06:13.05Z" },
 ]
 
+[[package]]
+name = "h5py"
+version = "3.15.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/4d/6a/0d79de0b025aa85dc8864de8e97659c94cf3d23148394a954dc5ca52f8c8/h5py-3.15.1.tar.gz", hash = "sha256:c86e3ed45c4473564de55aa83b6fc9e5ead86578773dfbd93047380042e26b69", size = 426236, upload-time = "2025-10-16T10:35:27.404Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/86/30/8fa61698b438dd751fa46a359792e801191dadab560d0a5f1c709443ef8e/h5py-3.15.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:67e59f6c2f19a32973a40f43d9a088ae324fe228c8366e25ebc57ceebf093a6b", size = 3414477, upload-time = "2025-10-16T10:33:24.201Z" },
+    { url = "https://files.pythonhosted.org/packages/16/16/db2f63302937337c4e9e51d97a5984b769bdb7488e3d37632a6ac297f8ef/h5py-3.15.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0e2f471688402c3404fa4e13466e373e622fd4b74b47b56cfdff7cc688209422", size = 2850298, upload-time = "2025-10-16T10:33:27.747Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/2e/f1bb7de9b05112bfd14d5206090f0f92f1e75bbb412fbec5d4653c3d44dd/h5py-3.15.1-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4c45802bcb711e128a6839cb6c01e9ac648dc55df045c9542a675c771f15c8d5", size = 4523605, upload-time = "2025-10-16T10:33:31.168Z" },
+    { url = "https://files.pythonhosted.org/packages/05/8a/63f4b08f3628171ce8da1a04681a65ee7ac338fde3cb3e9e3c9f7818e4da/h5py-3.15.1-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:64ce3f6470adb87c06e3a8dd1b90e973699f1759ad79bfa70c230939bff356c9", size = 4735346, upload-time = "2025-10-16T10:33:34.759Z" },
+    { url = "https://files.pythonhosted.org/packages/74/48/f16d12d9de22277605bcc11c0dcab5e35f06a54be4798faa2636b5d44b3c/h5py-3.15.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:4411c1867b9899a25e983fff56d820a66f52ac326bbe10c7cdf7d832c9dcd883", size = 4175305, upload-time = "2025-10-16T10:33:38.83Z" },
+    { url = "https://files.pythonhosted.org/packages/d6/2f/47cdbff65b2ce53c27458c6df63a232d7bb1644b97df37b2342442342c84/h5py-3.15.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:2cbc4104d3d4aca9d6db8c0c694555e255805bfeacf9eb1349bda871e26cacbe", size = 4653602, upload-time = "2025-10-16T10:33:42.188Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/28/dc08de359c2f43a67baa529cb70d7f9599848750031975eed92d6ae78e1d/h5py-3.15.1-cp310-cp310-win_amd64.whl", hash = "sha256:01f55111ca516f5568ae7a7fc8247dfce607de331b4467ee8a9a6ed14e5422c7", size = 2873601, upload-time = "2025-10-16T10:33:45.323Z" },
+    { url = "https://files.pythonhosted.org/packages/41/fd/8349b48b15b47768042cff06ad6e1c229f0a4bd89225bf6b6894fea27e6d/h5py-3.15.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:5aaa330bcbf2830150c50897ea5dcbed30b5b6d56897289846ac5b9e529ec243", size = 3434135, upload-time = "2025-10-16T10:33:47.954Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/b0/1c628e26a0b95858f54aba17e1599e7f6cd241727596cc2580b72cb0a9bf/h5py-3.15.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:c970fb80001fffabb0109eaf95116c8e7c0d3ca2de854e0901e8a04c1f098509", size = 2870958, upload-time = "2025-10-16T10:33:50.907Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/e3/c255cafc9b85e6ea04e2ad1bba1416baa1d7f57fc98a214be1144087690c/h5py-3.15.1-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:80e5bb5b9508d5d9da09f81fd00abbb3f85da8143e56b1585d59bc8ceb1dba8b", size = 4504770, upload-time = "2025-10-16T10:33:54.357Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/23/4ab1108e87851ccc69694b03b817d92e142966a6c4abd99e17db77f2c066/h5py-3.15.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5b849ba619a066196169763c33f9f0f02e381156d61c03e000bb0100f9950faf", size = 4700329, upload-time = "2025-10-16T10:33:57.616Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/e4/932a3a8516e4e475b90969bf250b1924dbe3612a02b897e426613aed68f4/h5py-3.15.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:e7f6c841efd4e6e5b7e82222eaf90819927b6d256ab0f3aca29675601f654f3c", size = 4152456, upload-time = "2025-10-16T10:34:00.843Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/0a/f74d589883b13737021b2049ac796328f188dbb60c2ed35b101f5b95a3fc/h5py-3.15.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ca8a3a22458956ee7b40d8e39c9a9dc01f82933e4c030c964f8b875592f4d831", size = 4617295, upload-time = "2025-10-16T10:34:04.154Z" },
+    { url = "https://files.pythonhosted.org/packages/23/95/499b4e56452ef8b6c95a271af0dde08dac4ddb70515a75f346d4f400579b/h5py-3.15.1-cp311-cp311-win_amd64.whl", hash = "sha256:550e51131376889656feec4aff2170efc054a7fe79eb1da3bb92e1625d1ac878", size = 2882129, upload-time = "2025-10-16T10:34:06.886Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/bb/cfcc70b8a42222ba3ad4478bcef1791181ea908e2adbd7d53c66395edad5/h5py-3.15.1-cp311-cp311-win_arm64.whl", hash = "sha256:b39239947cb36a819147fc19e86b618dcb0953d1cd969f5ed71fc0de60392427", size = 2477121, upload-time = "2025-10-16T10:34:09.579Z" },
+    { url = "https://files.pythonhosted.org/packages/62/b8/c0d9aa013ecfa8b7057946c080c0c07f6fa41e231d2e9bd306a2f8110bdc/h5py-3.15.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:316dd0f119734f324ca7ed10b5627a2de4ea42cc4dfbcedbee026aaa361c238c", size = 3399089, upload-time = "2025-10-16T10:34:12.135Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/5e/3c6f6e0430813c7aefe784d00c6711166f46225f5d229546eb53032c3707/h5py-3.15.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:b51469890e58e85d5242e43aab29f5e9c7e526b951caab354f3ded4ac88e7b76", size = 2847803, upload-time = "2025-10-16T10:34:14.564Z" },
+    { url = "https://files.pythonhosted.org/packages/00/69/ba36273b888a4a48d78f9268d2aee05787e4438557450a8442946ab8f3ec/h5py-3.15.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8a33bfd5dfcea037196f7778534b1ff7e36a7f40a89e648c8f2967292eb6898e", size = 4914884, upload-time = "2025-10-16T10:34:18.452Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/30/d1c94066343a98bb2cea40120873193a4fed68c4ad7f8935c11caf74c681/h5py-3.15.1-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:25c8843fec43b2cc368aa15afa1cdf83fc5e17b1c4e10cd3771ef6c39b72e5ce", size = 5109965, upload-time = "2025-10-16T10:34:21.853Z" },
+    { url = "https://files.pythonhosted.org/packages/81/3d/d28172116eafc3bc9f5991b3cb3fd2c8a95f5984f50880adfdf991de9087/h5py-3.15.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:a308fd8681a864c04423c0324527237a0484e2611e3441f8089fd00ed56a8171", size = 4561870, upload-time = "2025-10-16T10:34:26.69Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/83/393a7226024238b0f51965a7156004eaae1fcf84aa4bfecf7e582676271b/h5py-3.15.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:f4a016df3f4a8a14d573b496e4d1964deb380e26031fc85fb40e417e9131888a", size = 5037161, upload-time = "2025-10-16T10:34:30.383Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/51/329e7436bf87ca6b0fe06dd0a3795c34bebe4ed8d6c44450a20565d57832/h5py-3.15.1-cp312-cp312-win_amd64.whl", hash = "sha256:59b25cf02411bf12e14f803fef0b80886444c7fe21a5ad17c6a28d3f08098a1e", size = 2874165, upload-time = "2025-10-16T10:34:33.461Z" },
+    { url = "https://files.pythonhosted.org/packages/09/a8/2d02b10a66747c54446e932171dd89b8b4126c0111b440e6bc05a7c852ec/h5py-3.15.1-cp312-cp312-win_arm64.whl", hash = "sha256:61d5a58a9851e01ee61c932bbbb1c98fe20aba0a5674776600fb9a361c0aa652", size = 2458214, upload-time = "2025-10-16T10:34:35.733Z" },
+]
+
 [[package]]
 name = "hf-xet"
-version = "1.1.10"
+version = "1.2.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/74/31/feeddfce1748c4a233ec1aa5b7396161c07ae1aa9b7bdbc9a72c3c7dd768/hf_xet-1.1.10.tar.gz", hash = "sha256:408aef343800a2102374a883f283ff29068055c111f003ff840733d3b715bb97", size = 487910, upload-time = "2025-09-12T20:10:27.12Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/5e/6e/0f11bacf08a67f7fb5ee09740f2ca54163863b07b70d579356e9222ce5d8/hf_xet-1.2.0.tar.gz", hash = "sha256:a8c27070ca547293b6890c4bf389f713f80e8c478631432962bb7f4bc0bd7d7f", size = 506020, upload-time = "2025-10-24T19:04:32.129Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f7/a2/343e6d05de96908366bdc0081f2d8607d61200be2ac802769c4284cc65bd/hf_xet-1.1.10-cp37-abi3-macosx_10_12_x86_64.whl", hash = "sha256:686083aca1a6669bc85c21c0563551cbcdaa5cf7876a91f3d074a030b577231d", size = 2761466, upload-time = "2025-09-12T20:10:22.836Z" },
-    { url = "https://files.pythonhosted.org/packages/31/f9/6215f948ac8f17566ee27af6430ea72045e0418ce757260248b483f4183b/hf_xet-1.1.10-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:71081925383b66b24eedff3013f8e6bbd41215c3338be4b94ba75fd75b21513b", size = 2623807, upload-time = "2025-09-12T20:10:21.118Z" },
-    { url = "https://files.pythonhosted.org/packages/15/07/86397573efefff941e100367bbda0b21496ffcdb34db7ab51912994c32a2/hf_xet-1.1.10-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6b6bceb6361c80c1cc42b5a7b4e3efd90e64630bcf11224dcac50ef30a47e435", size = 3186960, upload-time = "2025-09-12T20:10:19.336Z" },
-    { url = "https://files.pythonhosted.org/packages/01/a7/0b2e242b918cc30e1f91980f3c4b026ff2eedaf1e2ad96933bca164b2869/hf_xet-1.1.10-cp37-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:eae7c1fc8a664e54753ffc235e11427ca61f4b0477d757cc4eb9ae374b69f09c", size = 3087167, upload-time = "2025-09-12T20:10:17.255Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/25/3e32ab61cc7145b11eee9d745988e2f0f4fafda81b25980eebf97d8cff15/hf_xet-1.1.10-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:0a0005fd08f002180f7a12d4e13b22be277725bc23ed0529f8add5c7a6309c06", size = 3248612, upload-time = "2025-09-12T20:10:24.093Z" },
-    { url = "https://files.pythonhosted.org/packages/2c/3d/ab7109e607ed321afaa690f557a9ada6d6d164ec852fd6bf9979665dc3d6/hf_xet-1.1.10-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:f900481cf6e362a6c549c61ff77468bd59d6dd082f3170a36acfef2eb6a6793f", size = 3353360, upload-time = "2025-09-12T20:10:25.563Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/0e/471f0a21db36e71a2f1752767ad77e92d8cde24e974e03d662931b1305ec/hf_xet-1.1.10-cp37-abi3-win_amd64.whl", hash = "sha256:5f54b19cc347c13235ae7ee98b330c26dd65ef1df47e5316ffb1e87713ca7045", size = 2804691, upload-time = "2025-09-12T20:10:28.433Z" },
+    { url = "https://files.pythonhosted.org/packages/96/2d/22338486473df5923a9ab7107d375dbef9173c338ebef5098ef593d2b560/hf_xet-1.2.0-cp37-abi3-macosx_10_12_x86_64.whl", hash = "sha256:46740d4ac024a7ca9b22bebf77460ff43332868b661186a8e46c227fdae01848", size = 2866099, upload-time = "2025-10-24T19:04:15.366Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/8c/c5becfa53234299bc2210ba314eaaae36c2875e0045809b82e40a9544f0c/hf_xet-1.2.0-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:27df617a076420d8845bea087f59303da8be17ed7ec0cd7ee3b9b9f579dff0e4", size = 2722178, upload-time = "2025-10-24T19:04:13.695Z" },
+    { url = "https://files.pythonhosted.org/packages/9a/92/cf3ab0b652b082e66876d08da57fcc6fa2f0e6c70dfbbafbd470bb73eb47/hf_xet-1.2.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3651fd5bfe0281951b988c0facbe726aa5e347b103a675f49a3fa8144c7968fd", size = 3320214, upload-time = "2025-10-24T19:04:03.596Z" },
+    { url = "https://files.pythonhosted.org/packages/46/92/3f7ec4a1b6a65bf45b059b6d4a5d38988f63e193056de2f420137e3c3244/hf_xet-1.2.0-cp37-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:d06fa97c8562fb3ee7a378dd9b51e343bc5bc8190254202c9771029152f5e08c", size = 3229054, upload-time = "2025-10-24T19:04:01.949Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/dd/7ac658d54b9fb7999a0ccb07ad863b413cbaf5cf172f48ebcd9497ec7263/hf_xet-1.2.0-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:4c1428c9ae73ec0939410ec73023c4f842927f39db09b063b9482dac5a3bb737", size = 3413812, upload-time = "2025-10-24T19:04:24.585Z" },
+    { url = "https://files.pythonhosted.org/packages/92/68/89ac4e5b12a9ff6286a12174c8538a5930e2ed662091dd2572bbe0a18c8a/hf_xet-1.2.0-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:a55558084c16b09b5ed32ab9ed38421e2d87cf3f1f89815764d1177081b99865", size = 3508920, upload-time = "2025-10-24T19:04:26.927Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/44/870d44b30e1dcfb6a65932e3e1506c103a8a5aea9103c337e7a53180322c/hf_xet-1.2.0-cp37-abi3-win_amd64.whl", hash = "sha256:e6584a52253f72c9f52f9e549d5895ca7a471608495c4ecaa6cc73dba2b24d69", size = 2905735, upload-time = "2025-10-24T19:04:35.928Z" },
 ]
 
 [[package]]
@@ -1397,30 +1315,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/7e/f5/f66802a942d491edb555dd61e3a9961140fd64c90bce1eafd741609d334d/httpcore-1.0.9-py3-none-any.whl", hash = "sha256:2d400746a40668fc9dec9810239072b40b4484b640a8c38fd654a024c7a1bf55", size = 78784, upload-time = "2025-04-24T22:06:20.566Z" },
 ]
 
-[[package]]
-name = "httptools"
-version = "0.6.4"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/a7/9a/ce5e1f7e131522e6d3426e8e7a490b3a01f39a6696602e1c4f33f9e94277/httptools-0.6.4.tar.gz", hash = "sha256:4e93eee4add6493b59a5c514da98c939b244fce4a0d8879cd3f466562f4b7d5c", size = 240639, upload-time = "2024-10-16T19:45:08.902Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/92/5e/b4a826fe91971a0b68e8c2bd4e7db3e7519882f5a8ccdb1194be2b3ab98f/httptools-0.6.4-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:deee0e3343f98ee8047e9f4c5bc7cedbf69f5734454a94c38ee829fb2d5fa3c1", size = 443130, upload-time = "2024-10-16T19:44:09.45Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/51/ce61e531e40289a681a463e1258fa1e05e0be54540e40d91d065a264cd8f/httptools-0.6.4-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ca80b7485c76f768a3bc83ea58373f8db7b015551117375e4918e2aa77ea9b50", size = 442148, upload-time = "2024-10-16T19:44:11.539Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/9e/270b7d767849b0c96f275c695d27ca76c30671f8eb8cc1bab6ced5c5e1d0/httptools-0.6.4-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:90d96a385fa941283ebd231464045187a31ad932ebfa541be8edf5b3c2328959", size = 415949, upload-time = "2024-10-16T19:44:13.388Z" },
-    { url = "https://files.pythonhosted.org/packages/81/86/ced96e3179c48c6f656354e106934e65c8963d48b69be78f355797f0e1b3/httptools-0.6.4-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:59e724f8b332319e2875efd360e61ac07f33b492889284a3e05e6d13746876f4", size = 417591, upload-time = "2024-10-16T19:44:15.258Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/24/0fe235d7b69c42423c7698d086d4db96475f9b50b6ad26a718ef27a0bce6/httptools-0.6.4-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f8787367fbdfccae38e35abf7641dafc5310310a5987b689f4c32cc8cc3ee975", size = 462891, upload-time = "2024-10-16T19:44:21.067Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/2f/205d1f2a190b72da6ffb5f41a3736c26d6fa7871101212b15e9b5cd8f61d/httptools-0.6.4-cp311-cp311-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:40b0f7fe4fd38e6a507bdb751db0379df1e99120c65fbdc8ee6c1d044897a636", size = 459788, upload-time = "2024-10-16T19:44:22.958Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/4c/d09ce0eff09057a206a74575ae8f1e1e2f0364d20e2442224f9e6612c8b9/httptools-0.6.4-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:40a5ec98d3f49904b9fe36827dcf1aadfef3b89e2bd05b0e35e94f97c2b14721", size = 433214, upload-time = "2024-10-16T19:44:24.513Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/d2/84c9e23edbccc4a4c6f96a1b8d99dfd2350289e94f00e9ccc7aadde26fb5/httptools-0.6.4-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:dacdd3d10ea1b4ca9df97a0a303cbacafc04b5cd375fa98732678151643d4988", size = 434120, upload-time = "2024-10-16T19:44:26.295Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/01/6fb20be3196ffdc8eeec4e653bc2a275eca7f36634c86302242c4fbb2760/httptools-0.6.4-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:69422b7f458c5af875922cdb5bd586cc1f1033295aa9ff63ee196a87519ac8e1", size = 508796, upload-time = "2024-10-16T19:44:32.825Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/d8/b644c44acc1368938317d76ac991c9bba1166311880bcc0ac297cb9d6bd7/httptools-0.6.4-cp312-cp312-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:16e603a3bff50db08cd578d54f07032ca1631450ceb972c2f834c2b860c28ea2", size = 510837, upload-time = "2024-10-16T19:44:33.974Z" },
-    { url = "https://files.pythonhosted.org/packages/52/d8/254d16a31d543073a0e57f1c329ca7378d8924e7e292eda72d0064987486/httptools-0.6.4-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:ec4f178901fa1834d4a060320d2f3abc5c9e39766953d038f1458cb885f47e81", size = 485289, upload-time = "2024-10-16T19:44:35.111Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/3c/4aee161b4b7a971660b8be71a92c24d6c64372c1ab3ae7f366b3680df20f/httptools-0.6.4-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:f9eb89ecf8b290f2e293325c646a211ff1c2493222798bb80a530c5e7502494f", size = 489779, upload-time = "2024-10-16T19:44:36.253Z" },
-    { url = "https://files.pythonhosted.org/packages/dc/24/c70c34119d209bf08199d938dc9c69164f585ed3029237b4bdb90f673cb9/httptools-0.6.4-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d1ffd262a73d7c28424252381a5b854c19d9de5f56f075445d33919a637e3547", size = 449615, upload-time = "2024-10-16T19:45:01.351Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/62/e7f317fed3703bd81053840cacba4e40bcf424b870e4197f94bd1cf9fe7a/httptools-0.6.4-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:703c346571fa50d2e9856a37d7cd9435a25e7fd15e236c397bf224afaa355fe9", size = 448819, upload-time = "2024-10-16T19:45:02.652Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/13/68337d3be6b023260139434c49d7aa466aaa98f9aee7ed29270ac7dde6a2/httptools-0.6.4-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:aafe0f1918ed07b67c1e838f950b1c1fabc683030477e60b335649b8020e1076", size = 422093, upload-time = "2024-10-16T19:45:03.765Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/b3/3a1bc45be03dda7a60c7858e55b6cd0489a81613c1908fb81cf21d34ae50/httptools-0.6.4-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:0e563e54979e97b6d13f1bbc05a96109923e76b901f786a5eae36e99c01237bd", size = 423898, upload-time = "2024-10-16T19:45:05.683Z" },
-]
-
 [[package]]
 name = "httpx"
 version = "0.28.1"
@@ -1438,10 +1332,11 @@ wheels = [
 
 [[package]]
 name = "huggingface-hub"
-version = "0.35.3"
+version = "0.36.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "filelock" },
+    { name = "filelock", version = "3.19.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "filelock", version = "3.20.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "fsspec" },
     { name = "hf-xet", marker = "platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'arm64' or platform_machine == 'x86_64'" },
     { name = "packaging" },
@@ -1450,20 +1345,37 @@ dependencies = [
     { name = "tqdm" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/10/7e/a0a97de7c73671863ca6b3f61fa12518caf35db37825e43d63a70956738c/huggingface_hub-0.35.3.tar.gz", hash = "sha256:350932eaa5cc6a4747efae85126ee220e4ef1b54e29d31c3b45c5612ddf0b32a", size = 461798, upload-time = "2025-09-29T14:29:58.625Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/98/63/4910c5fa9128fdadf6a9c5ac138e8b1b6cee4ca44bf7915bbfbce4e355ee/huggingface_hub-0.36.0.tar.gz", hash = "sha256:47b3f0e2539c39bf5cde015d63b72ec49baff67b6931c3d97f3f84532e2b8d25", size = 463358, upload-time = "2025-10-23T12:12:01.413Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/31/a0/651f93d154cb72323358bf2bbae3e642bdb5d2f1bfc874d096f7cb159fa0/huggingface_hub-0.35.3-py3-none-any.whl", hash = "sha256:0e3a01829c19d86d03793e4577816fe3bdfc1602ac62c7fb220d593d351224ba", size = 564262, upload-time = "2025-09-29T14:29:55.813Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/bd/1a875e0d592d447cbc02805fd3fe0f497714d6a2583f59d14fa9ebad96eb/huggingface_hub-0.36.0-py3-none-any.whl", hash = "sha256:7bcc9ad17d5b3f07b57c78e79d527102d08313caa278a641993acddcb894548d", size = 566094, upload-time = "2025-10-23T12:11:59.557Z" },
 ]
 
 [[package]]
 name = "humanize"
 version = "4.13.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/98/1d/3062fcc89ee05a715c0b9bfe6490c00c576314f27ffee3a704122c6fd259/humanize-4.13.0.tar.gz", hash = "sha256:78f79e68f76f0b04d711c4e55d32bebef5be387148862cb1ef83d2b58e7935a0", size = 81884, upload-time = "2025-08-25T09:39:20.04Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/1e/c7/316e7ca04d26695ef0635dc81683d628350810eb8e9b2299fc08ba49f366/humanize-4.13.0-py3-none-any.whl", hash = "sha256:b810820b31891813b1673e8fec7f1ed3312061eab2f26e3fa192c393d11ed25f", size = 128869, upload-time = "2025-08-25T09:39:18.54Z" },
 ]
 
+[[package]]
+name = "humanize"
+version = "4.14.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/b6/43/50033d25ad96a7f3845f40999b4778f753c3901a11808a584fed7c00d9f5/humanize-4.14.0.tar.gz", hash = "sha256:2fa092705ea640d605c435b1ca82b2866a1b601cdf96f076d70b79a855eba90d", size = 82939, upload-time = "2025-10-15T13:04:51.214Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c3/5b/9512c5fb6c8218332b530f13500c6ff5f3ce3342f35e0dd7be9ac3856fd3/humanize-4.14.0-py3-none-any.whl", hash = "sha256:d57701248d040ad456092820e6fde56c930f17749956ac47f4f655c0c547bfff", size = 132092, upload-time = "2025-10-15T13:04:49.404Z" },
+]
+
 [[package]]
 name = "identify"
 version = "2.6.15"
@@ -1475,11 +1387,11 @@ wheels = [
 
 [[package]]
 name = "idna"
-version = "3.10"
+version = "3.11"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f1/70/7703c29685631f5a7590aa73f1f1d3fa9a380e654b86af429e0934a32f7d/idna-3.10.tar.gz", hash = "sha256:12f65c9b470abda6dc35cf8e63cc574b1c52b11df2c86030af0ac09b01b13ea9", size = 190490, upload-time = "2024-09-15T18:07:39.745Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/6f/6d/0703ccc57f3a7233505399edb88de3cbd678da106337b9fcde432b65ed60/idna-3.11.tar.gz", hash = "sha256:795dafcc9c04ed0c1fb032c2aa73654d8e8c5023a7df64a53f39190ada629902", size = 194582, upload-time = "2025-10-12T14:55:20.501Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/76/c6/c88e154df9c4e1a2a66ccf0005a88dfb2650c1dffb6f5ce603dfbd452ce3/idna-3.10-py3-none-any.whl", hash = "sha256:946d195a0d259cbba61165e88e65941f16e9b36ea6ddb97f00452bae8b1287d3", size = 70442, upload-time = "2024-09-15T18:07:37.964Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/61/66938bbb5fc52dbdf84594873d5b51fb1f7c7794e9c0f5bd885f30bc507b/idna-3.11-py3-none-any.whl", hash = "sha256:771a87f49d9defaf64091e6e6fe9c18d4833f140bd19464795bc32d966ca37ea", size = 71008, upload-time = "2025-10-12T14:55:18.883Z" },
 ]
 
 [[package]]
@@ -1510,27 +1422,35 @@ wheels = [
 name = "iniconfig"
 version = "2.1.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/f2/97/ebf4da567aa6827c909642694d71c9fcf53e5b504f2d96afea02718862f3/iniconfig-2.1.0.tar.gz", hash = "sha256:3abbd2e30b36733fee78f9c7f7308f2d0050e88f0087fd25c2645f63c773e1c7", size = 4793, upload-time = "2025-03-19T20:09:59.721Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/2c/e1/e6716421ea10d38022b952c159d5161ca1193197fb744506875fbb87ea7b/iniconfig-2.1.0-py3-none-any.whl", hash = "sha256:9deba5723312380e77435581c6bf4935c94cbfab9b1ed33ef8d238ea168eb760", size = 6050, upload-time = "2025-03-19T20:10:01.071Z" },
 ]
 
 [[package]]
-name = "interegular"
-version = "0.3.3"
+name = "iniconfig"
+version = "2.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/dc/9d/8b6dde58a028a3962ce17e84d5fe73758df61378e00ef8ac3d85da34b0ff/interegular-0.3.3.tar.gz", hash = "sha256:d9b697b21b34884711399ba0f0376914b81899ce670032486d0d048344a76600", size = 24705, upload-time = "2024-01-06T23:01:22.372Z" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/72/34/14ca021ce8e5dfedc35312d08ba8bf51fdd999c576889fc2c24cb97f4f10/iniconfig-2.3.0.tar.gz", hash = "sha256:c76315c77db068650d49c5b56314774a7804df16fee4402c1f19d6d15d8c4730", size = 20503, upload-time = "2025-10-18T21:55:43.219Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c4/01/72d6472f80651673716d1deda2a5bbb633e563ecf94f4479da5519d69d25/interegular-0.3.3-py37-none-any.whl", hash = "sha256:b0c07007d48c89d6d19f7204972d369b2a77222722e126b6aa63aa721dc3b19c", size = 23635, upload-time = "2024-01-06T23:01:20.829Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/b1/3846dd7f199d53cb17f49cba7e651e9ce294d8497c8c150530ed11865bb8/iniconfig-2.3.0-py3-none-any.whl", hash = "sha256:f631c04d2c48c52b84d0d0549c99ff3859c98df65b3101406327ecc7d53fbf12", size = 7484, upload-time = "2025-10-18T21:55:41.639Z" },
 ]
 
 [[package]]
 name = "invoke"
-version = "2.2.0"
+version = "2.2.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f9/42/127e6d792884ab860defc3f4d80a8f9812e48ace584ffc5a346de58cdc6c/invoke-2.2.0.tar.gz", hash = "sha256:ee6cbb101af1a859c7fe84f2a264c059020b0cb7fe3535f9424300ab568f6bd5", size = 299835, upload-time = "2023-07-12T18:05:17.998Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/de/bd/b461d3424a24c80490313fd77feeb666ca4f6a28c7e72713e3d9095719b4/invoke-2.2.1.tar.gz", hash = "sha256:515bf49b4a48932b79b024590348da22f39c4942dff991ad1fb8b8baea1be707", size = 304762, upload-time = "2025-10-11T00:36:35.172Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/0a/66/7f8c48009c72d73bc6bbe6eb87ac838d6a526146f7dab14af671121eb379/invoke-2.2.0-py3-none-any.whl", hash = "sha256:6ea924cc53d4f78e3d98bc436b08069a03077e6f85ad1ddaa8a116d7dad15820", size = 160274, upload-time = "2023-07-12T18:05:16.294Z" },
+    { url = "https://files.pythonhosted.org/packages/32/4b/b99e37f88336009971405cbb7630610322ed6fbfa31e1d7ab3fbf3049a2d/invoke-2.2.1-py3-none-any.whl", hash = "sha256:2413bc441b376e5cd3f55bb5d364f973ad8bdd7bf87e53c79de3c11bf3feecc8", size = 160287, upload-time = "2025-10-11T00:36:33.703Z" },
 ]
 
 [[package]]
@@ -1583,22 +1503,22 @@ wheels = [
 
 [[package]]
 name = "jax"
-version = "0.7.2"
+version = "0.8.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
     "python_full_version == '3.11.*'",
 ]
 dependencies = [
-    { name = "jaxlib", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "ml-dtypes", marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "opt-einsum", marker = "python_full_version >= '3.11'" },
-    { name = "scipy", version = "1.16.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "scipy", version = "1.16.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/8f/e7/1e8e8af59b7659c83dc07dfa1dc23bc13551e5ef89bdef19ced044a497fc/jax-0.7.2.tar.gz", hash = "sha256:71a42b964bc6d52e819311429e6c0f5742e2a4650226dab1a1dd26fd986ca70d", size = 2434085, upload-time = "2025-09-16T16:48:53.108Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/32/82/84fd2c662e4d410a34b0402de9b56bb69d7f72d1b875c3ae0edc07df18cc/jax-0.8.1.tar.gz", hash = "sha256:e53f67b15315f5e154851a7fd77a192b59c6c75b3f7ac56e214296765391cca7", size = 2509320, upload-time = "2025-11-18T19:50:02.609Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/d5/e6/5fd0f6fff79eb47469ff9c4fa27125b661517a2fbf8884689b02e9fdfaa8/jax-0.7.2-py3-none-any.whl", hash = "sha256:e7e32f9be51ae5cc6854225958c57de8cca2187d279844338465b15e8a1fe7f2", size = 2835570, upload-time = "2025-09-16T16:48:51.33Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/e7/19b8cfc8963b2e10a01a4db7bb27ec5fa39ecd024bc62f8e2d1de5625a9d/jax-0.8.1-py3-none-any.whl", hash = "sha256:4cbdc5548f3095cdd69d38e4337950b2fc1f250a740a0234d190e4a319077564", size = 2922137, upload-time = "2025-11-18T19:47:43.693Z" },
 ]
 
 [[package]]
@@ -1665,7 +1585,7 @@ wheels = [
 
 [[package]]
 name = "jaxlib"
-version = "0.7.2"
+version = "0.8.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -1674,17 +1594,17 @@ resolution-markers = [
 dependencies = [
     { name = "ml-dtypes", marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "scipy", version = "1.16.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "scipy", version = "1.16.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/55/fe/41e1df8a02acdc12472b2922e41495cf0a9b805f5fd1f163c9aaf81b30df/jaxlib-0.7.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:9fbf90afce30e34eba2ea929a506f5907bdd4062358122de499ce9e671a0ba1f", size = 53550443, upload-time = "2025-09-16T16:49:04.061Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/9b/c86d5c3cf354326b6d0c89f7a1562299e0cdcf3f507c82a6aca8ce77dcb6/jaxlib-0.7.2-cp311-cp311-manylinux_2_27_aarch64.whl", hash = "sha256:8ca7003351fbe8ccfa2fa5a493ec2dfbf2df92441306cf5c3b970508eedb92ab", size = 71551113, upload-time = "2025-09-16T16:49:07.95Z" },
-    { url = "https://files.pythonhosted.org/packages/80/dd/4a0565f636c0e702777bdb0afeeb262a019039869268194cbd0440ad47da/jaxlib-0.7.2-cp311-cp311-manylinux_2_27_x86_64.whl", hash = "sha256:4382006235cced59d2f795acc983c1bedcfbca4fea8f9461311d61c6a793ae66", size = 78159922, upload-time = "2025-09-16T16:49:11.656Z" },
-    { url = "https://files.pythonhosted.org/packages/28/03/199d9ce30981c6c8d2ae3a442d2fc3eeb1f31fa6aab9620ff1d3d3fdfd7f/jaxlib-0.7.2-cp311-cp311-win_amd64.whl", hash = "sha256:3b37eb937e0b8ed4f9b265fdf46a3cf64e1decd4f41c0053e96540d39bd7050c", size = 58013064, upload-time = "2025-09-16T16:49:15.063Z" },
-    { url = "https://files.pythonhosted.org/packages/df/82/324ffb86a6de8c149689b499b6dc6e4f83684fe43ae9e81c8e5eb91bc50f/jaxlib-0.7.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:bd6d1c53bd475e0e768a54af98b1642fb49d7304cf055ceebb1d01e89d38a1cb", size = 53563644, upload-time = "2025-09-16T16:49:18.04Z" },
-    { url = "https://files.pythonhosted.org/packages/89/6a/689b177de0672a7b4fe5489eec7cbb11305959c3af8da7d646e7f3aeb754/jaxlib-0.7.2-cp312-cp312-manylinux_2_27_aarch64.whl", hash = "sha256:9b5a0d357497611a113d207fb2c1997f01ab7a175870700812220f0bcaa31822", size = 71557871, upload-time = "2025-09-16T16:49:20.964Z" },
-    { url = "https://files.pythonhosted.org/packages/7b/73/b44fbe943c9e02e25c99eb64e6b86e2dde8d918d064326813b5bbe620951/jaxlib-0.7.2-cp312-cp312-manylinux_2_27_x86_64.whl", hash = "sha256:11f32319e662ccff66859eb393757050d8971bd880bc4dd70dec6434d890fb59", size = 78167979, upload-time = "2025-09-16T16:49:24.607Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/0e/13d5264bbe08f03ae9e7abe7aa3041b5c29f9daf5493b8156ae7baa6cbf1/jaxlib-0.7.2-cp312-cp312-win_amd64.whl", hash = "sha256:b53cf3a7ed342ca1a9f419cc7a3c387598fc743043ba4a7c5895ebc4d62fa05a", size = 58040487, upload-time = "2025-09-16T16:49:28.101Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/8b/9babcf487c6f1b533bca9611124c4d9593367c058a96d326c7e70db7d334/jaxlib-0.8.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:865add56139883405f3f15c9b0de6a64ab8f4aa549dff196b72dbc86be6ccc1f", size = 55719927, upload-time = "2025-11-18T19:48:42.679Z" },
+    { url = "https://files.pythonhosted.org/packages/df/0c/b8c67272647ea151b0ac651e43faa846b4987d971058683dcce8abf68bca/jaxlib-0.8.1-cp311-cp311-manylinux_2_27_aarch64.whl", hash = "sha256:ff32b6320d729131efaf22939825b52d75957c84c32af2b0b1bdb33cf27ba75f", size = 74208199, upload-time = "2025-11-18T19:48:45.848Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/d0/5b83d614eddb58a2cc97fb948bfeb84509b90da04e808273bf9ae89ad6c1/jaxlib-0.8.1-cp311-cp311-manylinux_2_27_x86_64.whl", hash = "sha256:22f489fb5c8be0da7be5e4957a10936b3760a169668f8b25c5d09c51c3ef47f6", size = 80247963, upload-time = "2025-11-18T19:48:49.443Z" },
+    { url = "https://files.pythonhosted.org/packages/52/b2/4b8507d23cfd9e895dce0204f684dc9db4f746da5a46c2e586a50fb3f1ee/jaxlib-0.8.1-cp311-cp311-win_amd64.whl", hash = "sha256:c14c8c19a7eb694aa14092b6d2fffb9d2bdd8a603b63d6f26fbeaf129c204f9f", size = 59760834, upload-time = "2025-11-18T19:48:52.886Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/9d/59b36e2f348e599d5812743f263ca54aa03be1a4c9dfc11504d19864b72d/jaxlib-0.8.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:88bde0f535eeea6689e0cd57d40b7660d5206ac95c7d42e09562a109b963a49f", size = 55728156, upload-time = "2025-11-18T19:48:56.254Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/73/2aa891de9f5f4c60ba3c63bda97ec4ace50ffb900ff3bf750ce42c514a3b/jaxlib-0.8.1-cp312-cp312-manylinux_2_27_aarch64.whl", hash = "sha256:bed1e94ae8c7c16bca4476d8d7f582f0d1a102a4e69c3a9bd2069a0dc42274a9", size = 74209108, upload-time = "2025-11-18T19:48:59.572Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/4b/3c7e373d81219ee7493c1581c85a926c413ddeb3794cff87a37023a337e4/jaxlib-0.8.1-cp312-cp312-manylinux_2_27_x86_64.whl", hash = "sha256:af4924189fc53b69237715b56ebcbfc71bb91ca16184143dcef0d430c8173de6", size = 80256943, upload-time = "2025-11-18T19:49:02.92Z" },
+    { url = "https://files.pythonhosted.org/packages/07/6c/a6f449a7d1c7f91d73c3b8e00ceba92dff9dfd642508bbe1ddba9cb9ea57/jaxlib-0.8.1-cp312-cp312-win_amd64.whl", hash = "sha256:24ec3f3a9c45d6de060020dc94c444d69e18099fab927ea3979ff8cedf0ed2c9", size = 59787068, upload-time = "2025-11-18T19:49:06.275Z" },
 ]
 
 [[package]]
@@ -1701,59 +1621,68 @@ wheels = [
 
 [[package]]
 name = "jiter"
-version = "0.11.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/9d/c0/a3bb4cc13aced219dd18191ea66e874266bd8aa7b96744e495e1c733aa2d/jiter-0.11.0.tar.gz", hash = "sha256:1d9637eaf8c1d6a63d6562f2a6e5ab3af946c66037eb1b894e8fad75422266e4", size = 167094, upload-time = "2025-09-15T09:20:38.212Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/25/21/7dd1235a19e26979be6098e87e4cced2e061752f3a40a17bbce6dea7fae1/jiter-0.11.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:3893ce831e1c0094a83eeaf56c635a167d6fa8cc14393cc14298fd6fdc2a2449", size = 309875, upload-time = "2025-09-15T09:18:48.41Z" },
-    { url = "https://files.pythonhosted.org/packages/71/f9/462b54708aa85b135733ccba70529dd68a18511bf367a87c5fd28676c841/jiter-0.11.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:25c625b9b61b5a8725267fdf867ef2e51b429687f6a4eef211f4612e95607179", size = 316505, upload-time = "2025-09-15T09:18:51.057Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/40/14e2eeaac6a47bff27d213834795472355fd39769272eb53cb7aa83d5aa8/jiter-0.11.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dd4ca85fb6a62cf72e1c7f5e34ddef1b660ce4ed0886ec94a1ef9777d35eaa1f", size = 337613, upload-time = "2025-09-15T09:18:52.358Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/ed/a5f1f8419c92b150a7c7fb5ccba1fb1e192887ad713d780e70874f0ce996/jiter-0.11.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:572208127034725e79c28437b82414028c3562335f2b4f451d98136d0fc5f9cd", size = 361438, upload-time = "2025-09-15T09:18:54.637Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/f5/70682c023dfcdd463a53faf5d30205a7d99c51d70d3e303c932d0936e5a2/jiter-0.11.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:494ba627c7f550ad3dabb21862864b8f2216098dc18ff62f37b37796f2f7c325", size = 486180, upload-time = "2025-09-15T09:18:56.158Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/39/020d08cbab4eab48142ad88b837c41eb08a15c0767fdb7c0d3265128a44b/jiter-0.11.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:b8da18a99f58bca3ecc2d2bba99cac000a924e115b6c4f0a2b98f752b6fbf39a", size = 376681, upload-time = "2025-09-15T09:18:57.553Z" },
-    { url = "https://files.pythonhosted.org/packages/52/10/b86733f6e594cf51dd142f37c602d8df87c554c5844958deaab0de30eb5d/jiter-0.11.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e4ffd3b0fff3fabbb02cc09910c08144db6bb5697a98d227a074401e01ee63dd", size = 348685, upload-time = "2025-09-15T09:18:59.208Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/ee/8861665e83a9e703aa5f65fddddb6225428e163e6b0baa95a7f9a8fb9aae/jiter-0.11.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:8fe6530aa738a4f7d4e4702aa8f9581425d04036a5f9e25af65ebe1f708f23be", size = 385573, upload-time = "2025-09-15T09:19:00.593Z" },
-    { url = "https://files.pythonhosted.org/packages/25/74/05afec03600951f128293813b5a208c9ba1bf587c57a344c05a42a69e1b1/jiter-0.11.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:e35d66681c133a03d7e974e7eedae89720fe8ca3bd09f01a4909b86a8adf31f5", size = 516669, upload-time = "2025-09-15T09:19:02.369Z" },
-    { url = "https://files.pythonhosted.org/packages/93/d1/2e5bfe147cfbc2a5eef7f73eb75dc5c6669da4fa10fc7937181d93af9495/jiter-0.11.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:c59459beca2fbc9718b6f1acb7bfb59ebc3eb4294fa4d40e9cb679dafdcc6c60", size = 508767, upload-time = "2025-09-15T09:19:04.011Z" },
-    { url = "https://files.pythonhosted.org/packages/87/50/597f71307e10426b5c082fd05d38c615ddbdd08c3348d8502963307f0652/jiter-0.11.0-cp310-cp310-win32.whl", hash = "sha256:b7b0178417b0dcfc5f259edbc6db2b1f5896093ed9035ee7bab0f2be8854726d", size = 205476, upload-time = "2025-09-15T09:19:05.594Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/86/1e5214b3272e311754da26e63edec93a183811d4fc2e0118addec365df8b/jiter-0.11.0-cp310-cp310-win_amd64.whl", hash = "sha256:11df2bf99fb4754abddd7f5d940a48e51f9d11624d6313ca4314145fcad347f0", size = 204708, upload-time = "2025-09-15T09:19:06.955Z" },
-    { url = "https://files.pythonhosted.org/packages/38/55/a69fefeef09c2eaabae44b935a1aa81517e49639c0a0c25d861cb18cd7ac/jiter-0.11.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:cb5d9db02979c3f49071fce51a48f4b4e4cf574175fb2b11c7a535fa4867b222", size = 309503, upload-time = "2025-09-15T09:19:08.191Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/d5/a6aba9e6551f32f9c127184f398208e4eddb96c59ac065c8a92056089d28/jiter-0.11.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:1dc6a123f3471c4730db7ca8ba75f1bb3dcb6faeb8d46dd781083e7dee88b32d", size = 317688, upload-time = "2025-09-15T09:19:09.918Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/f3/5e86f57c1883971cdc8535d0429c2787bf734840a231da30a3be12850562/jiter-0.11.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:09858f8d230f031c7b8e557429102bf050eea29c77ad9c34c8fe253c5329acb7", size = 337418, upload-time = "2025-09-15T09:19:11.078Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/4f/a71d8a24c2a70664970574a8e0b766663f5ef788f7fe1cc20ee0c016d488/jiter-0.11.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:dbe2196c4a0ce760925a74ab4456bf644748ab0979762139626ad138f6dac72d", size = 361423, upload-time = "2025-09-15T09:19:13.286Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/e5/b09076f4e7fd9471b91e16f9f3dc7330b161b738f3b39b2c37054a36e26a/jiter-0.11.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5beb56d22b63647bafd0b74979216fdee80c580c0c63410be8c11053860ffd09", size = 486367, upload-time = "2025-09-15T09:19:14.546Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/f1/98cb3a36f5e62f80cd860f0179f948d9eab5a316d55d3e1bab98d9767af5/jiter-0.11.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:97025d09ef549795d8dc720a824312cee3253c890ac73c621721ddfc75066789", size = 376335, upload-time = "2025-09-15T09:19:15.939Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/d8/ec74886497ea393c29dbd7651ddecc1899e86404a6b1f84a3ddab0ab59fd/jiter-0.11.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d50880a6da65d8c23a2cf53c412847d9757e74cc9a3b95c5704a1d1a24667347", size = 348981, upload-time = "2025-09-15T09:19:17.568Z" },
-    { url = "https://files.pythonhosted.org/packages/24/93/d22ad7fa3b86ade66c86153ceea73094fc2af8b20c59cb7fceab9fea4704/jiter-0.11.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:452d80a1c86c095a242007bd9fc5d21b8a8442307193378f891cb8727e469648", size = 385797, upload-time = "2025-09-15T09:19:19.121Z" },
-    { url = "https://files.pythonhosted.org/packages/c8/bd/e25ff4a4df226e9b885f7cb01ee4b9dc74e3000e612d6f723860d71a1f34/jiter-0.11.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:e84e58198d4894668eec2da660ffff60e0f3e60afa790ecc50cb12b0e02ca1d4", size = 516597, upload-time = "2025-09-15T09:19:20.301Z" },
-    { url = "https://files.pythonhosted.org/packages/be/fb/beda613db7d93ffa2fdd2683f90f2f5dce8daf4bc2d0d2829e7de35308c6/jiter-0.11.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:df64edcfc5dd5279a791eea52aa113d432c933119a025b0b5739f90d2e4e75f1", size = 508853, upload-time = "2025-09-15T09:19:22.075Z" },
-    { url = "https://files.pythonhosted.org/packages/20/64/c5b0d93490634e41e38e2a15de5d54fdbd2c9f64a19abb0f95305b63373c/jiter-0.11.0-cp311-cp311-win32.whl", hash = "sha256:144fc21337d21b1d048f7f44bf70881e1586401d405ed3a98c95a114a9994982", size = 205140, upload-time = "2025-09-15T09:19:23.351Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/e6/c347c0e6f5796e97d4356b7e5ff0ce336498b7f4ef848fae621a56f1ccf3/jiter-0.11.0-cp311-cp311-win_amd64.whl", hash = "sha256:b0f32e644d241293b892b1a6dd8f0b9cc029bfd94c97376b2681c36548aabab7", size = 204311, upload-time = "2025-09-15T09:19:24.591Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/b5/3009b112b8f673e568ef79af9863d8309a15f0a8cdcc06ed6092051f377e/jiter-0.11.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:2fb7b377688cc3850bbe5c192a6bd493562a0bc50cbc8b047316428fbae00ada", size = 305510, upload-time = "2025-09-15T09:19:25.893Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/82/15514244e03b9e71e086bbe2a6de3e4616b48f07d5f834200c873956fb8c/jiter-0.11.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:a1b7cbe3f25bd0d8abb468ba4302a5d45617ee61b2a7a638f63fee1dc086be99", size = 316521, upload-time = "2025-09-15T09:19:27.525Z" },
-    { url = "https://files.pythonhosted.org/packages/92/94/7a2e905f40ad2d6d660e00b68d818f9e29fb87ffe82774f06191e93cbe4a/jiter-0.11.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c0a7f0ec81d5b7588c5cade1eb1925b91436ae6726dc2df2348524aeabad5de6", size = 338214, upload-time = "2025-09-15T09:19:28.727Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/9c/5791ed5bdc76f12110158d3316a7a3ec0b1413d018b41c5ed399549d3ad5/jiter-0.11.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:07630bb46ea2a6b9c6ed986c6e17e35b26148cce2c535454b26ee3f0e8dcaba1", size = 361280, upload-time = "2025-09-15T09:19:30.013Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/7f/b7d82d77ff0d2cb06424141000176b53a9e6b16a1125525bb51ea4990c2e/jiter-0.11.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7764f27d28cd4a9cbc61704dfcd80c903ce3aad106a37902d3270cd6673d17f4", size = 487895, upload-time = "2025-09-15T09:19:31.424Z" },
-    { url = "https://files.pythonhosted.org/packages/42/44/10a1475d46f1fc1fd5cc2e82c58e7bca0ce5852208e0fa5df2f949353321/jiter-0.11.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:1d4a6c4a737d486f77f842aeb22807edecb4a9417e6700c7b981e16d34ba7c72", size = 378421, upload-time = "2025-09-15T09:19:32.746Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/5f/0dc34563d8164d31d07bc09d141d3da08157a68dcd1f9b886fa4e917805b/jiter-0.11.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cf408d2a0abd919b60de8c2e7bc5eeab72d4dafd18784152acc7c9adc3291591", size = 347932, upload-time = "2025-09-15T09:19:34.612Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/de/b68f32a4fcb7b4a682b37c73a0e5dae32180140cd1caf11aef6ad40ddbf2/jiter-0.11.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:cdef53eda7d18e799625023e1e250dbc18fbc275153039b873ec74d7e8883e09", size = 386959, upload-time = "2025-09-15T09:19:35.994Z" },
-    { url = "https://files.pythonhosted.org/packages/76/0a/c08c92e713b6e28972a846a81ce374883dac2f78ec6f39a0dad9f2339c3a/jiter-0.11.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:53933a38ef7b551dd9c7f1064f9d7bb235bb3168d0fa5f14f0798d1b7ea0d9c5", size = 517187, upload-time = "2025-09-15T09:19:37.426Z" },
-    { url = "https://files.pythonhosted.org/packages/89/b5/4a283bec43b15aad54fcae18d951f06a2ec3f78db5708d3b59a48e9c3fbd/jiter-0.11.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:11840d2324c9ab5162fc1abba23bc922124fedcff0d7b7f85fffa291e2f69206", size = 509461, upload-time = "2025-09-15T09:19:38.761Z" },
-    { url = "https://files.pythonhosted.org/packages/34/a5/f8bad793010534ea73c985caaeef8cc22dfb1fedb15220ecdf15c623c07a/jiter-0.11.0-cp312-cp312-win32.whl", hash = "sha256:4f01a744d24a5f2bb4a11657a1b27b61dc038ae2e674621a74020406e08f749b", size = 206664, upload-time = "2025-09-15T09:19:40.096Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/42/5823ec2b1469395a160b4bf5f14326b4a098f3b6898fbd327366789fa5d3/jiter-0.11.0-cp312-cp312-win_amd64.whl", hash = "sha256:29fff31190ab3a26de026da2f187814f4b9c6695361e20a9ac2123e4d4378a4c", size = 203520, upload-time = "2025-09-15T09:19:41.798Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/d9/51cf35d92bea21f2051da8ca2328831589e67e2bf971e85b1a6e6c0d2030/jiter-0.11.0-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:719891c2fb7628a41adff4f2f54c19380a27e6fdfdb743c24680ef1a54c67bd0", size = 312764, upload-time = "2025-09-15T09:20:21.378Z" },
-    { url = "https://files.pythonhosted.org/packages/da/48/eae309ce5c180faa1bb45e378a503717da22ceb2b0488f78e548f97c2b6b/jiter-0.11.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:df7f1927cbdf34cb91262a5418ca06920fd42f1cf733936d863aeb29b45a14ef", size = 305861, upload-time = "2025-09-15T09:20:22.603Z" },
-    { url = "https://files.pythonhosted.org/packages/83/4f/13b80e18b0331f0fecc09cb2f09f722530b9a395006941b01491fe58baea/jiter-0.11.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e71ae6d969d0c9bab336c5e9e2fabad31e74d823f19e3604eaf96d9a97f463df", size = 339507, upload-time = "2025-09-15T09:20:23.857Z" },
-    { url = "https://files.pythonhosted.org/packages/97/6d/c2fd1512873d3f23d24537e97765e7090a00de466516aa442b994b064918/jiter-0.11.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5661469a7b2be25ade3a4bb6c21ffd1e142e13351a0759f264dfdd3ad99af1ab", size = 363751, upload-time = "2025-09-15T09:20:25.12Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/99/48d156c742e75d33b9c8be44b1142d233823be491acdb1009629e4109e6a/jiter-0.11.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:76c15ef0d3d02f8b389066fa4c410a0b89e9cc6468a1f0674c5925d2f3c3e890", size = 488591, upload-time = "2025-09-15T09:20:26.397Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/fd/214452149f63847b791b1f6e9558f59e94674c47418c03e9787236ac8656/jiter-0.11.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:63782a1350917a27817030716566ed3d5b3c731500fd42d483cbd7094e2c5b25", size = 378906, upload-time = "2025-09-15T09:20:27.637Z" },
-    { url = "https://files.pythonhosted.org/packages/de/91/25e38fbbfc17111d7b70b24290a41d611cc2a27fa6cd0ed84ddae38ec3e6/jiter-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5a7092b699646a1ddc03a7b112622d9c066172627c7382659befb0d2996f1659", size = 350288, upload-time = "2025-09-15T09:20:28.951Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/d8/d6d2eefa9f0ff6ac6b725f5164a94f15bb4dee68584b5054043d735803da/jiter-0.11.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f637b8e818f6d75540f350a6011ce21252573c0998ea1b4365ee54b7672c23c5", size = 388442, upload-time = "2025-09-15T09:20:30.223Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/e4/cd7e27852de498d441a575a147ac7a15cf66768ae2cde8c43ea5b464c827/jiter-0.11.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:a624d87719e1b5d09c15286eaee7e1532a40c692a096ea7ca791121365f548c1", size = 518273, upload-time = "2025-09-15T09:20:31.569Z" },
-    { url = "https://files.pythonhosted.org/packages/77/a2/6681a9a503141752b33c92c58512ed8da13849ed3dbf88a3f8aba9bfb255/jiter-0.11.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:a9d0146d8d9b3995821bb586fc8256636258947c2f39da5bab709f3a28fb1a0b", size = 510254, upload-time = "2025-09-15T09:20:32.889Z" },
-    { url = "https://files.pythonhosted.org/packages/38/32/df1a06f397074da35cf8fe79ec07da203358a2912b2a6349a0d4a88a1e0a/jiter-0.11.0-cp39-cp39-win32.whl", hash = "sha256:d067655a7cf0831eb8ec3e39cbd752995e9b69a2206df3535b3a067fac23b032", size = 207076, upload-time = "2025-09-15T09:20:34.196Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/91/11bc61fa76fd6197f5baa8576614852ee8586a16c2f25085edc6b47a369d/jiter-0.11.0-cp39-cp39-win_amd64.whl", hash = "sha256:f05d03775a11aaf132c447436983169958439f1219069abf24662a672851f94e", size = 206077, upload-time = "2025-09-15T09:20:35.598Z" },
-    { url = "https://files.pythonhosted.org/packages/70/f3/ce100253c80063a7b8b406e1d1562657fd4b9b4e1b562db40e68645342fb/jiter-0.11.0-graalpy311-graalpy242_311_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:902b43386c04739229076bd1c4c69de5d115553d982ab442a8ae82947c72ede7", size = 336380, upload-time = "2025-09-15T09:20:36.867Z" },
+version = "0.12.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/45/9d/e0660989c1370e25848bb4c52d061c71837239738ad937e83edca174c273/jiter-0.12.0.tar.gz", hash = "sha256:64dfcd7d5c168b38d3f9f8bba7fc639edb3418abcc74f22fdbe6b8938293f30b", size = 168294, upload-time = "2025-11-09T20:49:23.302Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3b/91/13cb9505f7be74a933f37da3af22e029f6ba64f5669416cb8b2774bc9682/jiter-0.12.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:e7acbaba9703d5de82a2c98ae6a0f59ab9770ab5af5fa35e43a303aee962cf65", size = 316652, upload-time = "2025-11-09T20:46:41.021Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/76/4e9185e5d9bb4e482cf6dec6410d5f78dfeb374cfcecbbe9888d07c52daa/jiter-0.12.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:364f1a7294c91281260364222f535bc427f56d4de1d8ffd718162d21fbbd602e", size = 319829, upload-time = "2025-11-09T20:46:43.281Z" },
+    { url = "https://files.pythonhosted.org/packages/86/af/727de50995d3a153138139f259baae2379d8cb0522c0c00419957bc478a6/jiter-0.12.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:85ee4d25805d4fb23f0a5167a962ef8e002dbfb29c0989378488e32cf2744b62", size = 350568, upload-time = "2025-11-09T20:46:45.075Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/c1/d6e9f4b7a3d5ac63bcbdfddeb50b2dcfbdc512c86cffc008584fdc350233/jiter-0.12.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:796f466b7942107eb889c08433b6e31b9a7ed31daceaecf8af1be26fb26c0ca8", size = 369052, upload-time = "2025-11-09T20:46:46.818Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/be/00824cd530f30ed73fa8a4f9f3890a705519e31ccb9e929f1e22062e7c76/jiter-0.12.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:35506cb71f47dba416694e67af996bbdefb8e3608f1f78799c2e1f9058b01ceb", size = 481585, upload-time = "2025-11-09T20:46:48.319Z" },
+    { url = "https://files.pythonhosted.org/packages/74/b6/2ad7990dff9504d4b5052eef64aa9574bd03d722dc7edced97aad0d47be7/jiter-0.12.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:726c764a90c9218ec9e4f99a33d6bf5ec169163f2ca0fc21b654e88c2abc0abc", size = 380541, upload-time = "2025-11-09T20:46:49.643Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/c7/f3c26ecbc1adbf1db0d6bba99192143d8fe8504729d9594542ecc4445784/jiter-0.12.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:baa47810c5565274810b726b0dc86d18dce5fd17b190ebdc3890851d7b2a0e74", size = 364423, upload-time = "2025-11-09T20:46:51.731Z" },
+    { url = "https://files.pythonhosted.org/packages/18/51/eac547bf3a2d7f7e556927278e14c56a0604b8cddae75815d5739f65f81d/jiter-0.12.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f8ec0259d3f26c62aed4d73b198c53e316ae11f0f69c8fbe6682c6dcfa0fcce2", size = 389958, upload-time = "2025-11-09T20:46:53.432Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/1f/9ca592e67175f2db156cff035e0d817d6004e293ee0c1d73692d38fcb596/jiter-0.12.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:79307d74ea83465b0152fa23e5e297149506435535282f979f18b9033c0bb025", size = 522084, upload-time = "2025-11-09T20:46:54.848Z" },
+    { url = "https://files.pythonhosted.org/packages/83/ff/597d9cdc3028f28224f53e1a9d063628e28b7a5601433e3196edda578cdd/jiter-0.12.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:cf6e6dd18927121fec86739f1a8906944703941d000f0639f3eb6281cc601dca", size = 513054, upload-time = "2025-11-09T20:46:56.487Z" },
+    { url = "https://files.pythonhosted.org/packages/24/6d/1970bce1351bd02e3afcc5f49e4f7ef3dabd7fb688f42be7e8091a5b809a/jiter-0.12.0-cp310-cp310-win32.whl", hash = "sha256:b6ae2aec8217327d872cbfb2c1694489057b9433afce447955763e6ab015b4c4", size = 206368, upload-time = "2025-11-09T20:46:58.638Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/6b/eb1eb505b2d86709b59ec06681a2b14a94d0941db091f044b9f0e16badc0/jiter-0.12.0-cp310-cp310-win_amd64.whl", hash = "sha256:c7f49ce90a71e44f7e1aa9e7ec415b9686bbc6a5961e57eab511015e6759bc11", size = 204847, upload-time = "2025-11-09T20:47:00.295Z" },
+    { url = "https://files.pythonhosted.org/packages/32/f9/eaca4633486b527ebe7e681c431f529b63fe2709e7c5242fc0f43f77ce63/jiter-0.12.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:d8f8a7e317190b2c2d60eb2e8aa835270b008139562d70fe732e1c0020ec53c9", size = 316435, upload-time = "2025-11-09T20:47:02.087Z" },
+    { url = "https://files.pythonhosted.org/packages/10/c1/40c9f7c22f5e6ff715f28113ebaba27ab85f9af2660ad6e1dd6425d14c19/jiter-0.12.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:2218228a077e784c6c8f1a8e5d6b8cb1dea62ce25811c356364848554b2056cd", size = 320548, upload-time = "2025-11-09T20:47:03.409Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/1b/efbb68fe87e7711b00d2cfd1f26bb4bfc25a10539aefeaa7727329ffb9cb/jiter-0.12.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9354ccaa2982bf2188fd5f57f79f800ef622ec67beb8329903abf6b10da7d423", size = 351915, upload-time = "2025-11-09T20:47:05.171Z" },
+    { url = "https://files.pythonhosted.org/packages/15/2d/c06e659888c128ad1e838123d0638f0efad90cc30860cb5f74dd3f2fc0b3/jiter-0.12.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:8f2607185ea89b4af9a604d4c7ec40e45d3ad03ee66998b031134bc510232bb7", size = 368966, upload-time = "2025-11-09T20:47:06.508Z" },
+    { url = "https://files.pythonhosted.org/packages/6b/20/058db4ae5fb07cf6a4ab2e9b9294416f606d8e467fb74c2184b2a1eeacba/jiter-0.12.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:3a585a5e42d25f2e71db5f10b171f5e5ea641d3aa44f7df745aa965606111cc2", size = 482047, upload-time = "2025-11-09T20:47:08.382Z" },
+    { url = "https://files.pythonhosted.org/packages/49/bb/dc2b1c122275e1de2eb12905015d61e8316b2f888bdaac34221c301495d6/jiter-0.12.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bd9e21d34edff5a663c631f850edcb786719c960ce887a5661e9c828a53a95d9", size = 380835, upload-time = "2025-11-09T20:47:09.81Z" },
+    { url = "https://files.pythonhosted.org/packages/23/7d/38f9cd337575349de16da575ee57ddb2d5a64d425c9367f5ef9e4612e32e/jiter-0.12.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4a612534770470686cd5431478dc5a1b660eceb410abade6b1b74e320ca98de6", size = 364587, upload-time = "2025-11-09T20:47:11.529Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/a3/b13e8e61e70f0bb06085099c4e2462647f53cc2ca97614f7fedcaa2bb9f3/jiter-0.12.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:3985aea37d40a908f887b34d05111e0aae822943796ebf8338877fee2ab67725", size = 390492, upload-time = "2025-11-09T20:47:12.993Z" },
+    { url = "https://files.pythonhosted.org/packages/07/71/e0d11422ed027e21422f7bc1883c61deba2d9752b720538430c1deadfbca/jiter-0.12.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:b1207af186495f48f72529f8d86671903c8c10127cac6381b11dddc4aaa52df6", size = 522046, upload-time = "2025-11-09T20:47:14.6Z" },
+    { url = "https://files.pythonhosted.org/packages/9f/59/b968a9aa7102a8375dbbdfbd2aeebe563c7e5dddf0f47c9ef1588a97e224/jiter-0.12.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:ef2fb241de583934c9915a33120ecc06d94aa3381a134570f59eed784e87001e", size = 513392, upload-time = "2025-11-09T20:47:16.011Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/e4/7df62002499080dbd61b505c5cb351aa09e9959d176cac2aa8da6f93b13b/jiter-0.12.0-cp311-cp311-win32.whl", hash = "sha256:453b6035672fecce8007465896a25b28a6b59cfe8fbc974b2563a92f5a92a67c", size = 206096, upload-time = "2025-11-09T20:47:17.344Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/60/1032b30ae0572196b0de0e87dce3b6c26a1eff71aad5fe43dee3082d32e0/jiter-0.12.0-cp311-cp311-win_amd64.whl", hash = "sha256:ca264b9603973c2ad9435c71a8ec8b49f8f715ab5ba421c85a51cde9887e421f", size = 204899, upload-time = "2025-11-09T20:47:19.365Z" },
+    { url = "https://files.pythonhosted.org/packages/49/d5/c145e526fccdb834063fb45c071df78b0cc426bbaf6de38b0781f45d956f/jiter-0.12.0-cp311-cp311-win_arm64.whl", hash = "sha256:cb00ef392e7d684f2754598c02c409f376ddcef857aae796d559e6cacc2d78a5", size = 188070, upload-time = "2025-11-09T20:47:20.75Z" },
+    { url = "https://files.pythonhosted.org/packages/92/c9/5b9f7b4983f1b542c64e84165075335e8a236fa9e2ea03a0c79780062be8/jiter-0.12.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:305e061fa82f4680607a775b2e8e0bcb071cd2205ac38e6ef48c8dd5ebe1cf37", size = 314449, upload-time = "2025-11-09T20:47:22.999Z" },
+    { url = "https://files.pythonhosted.org/packages/98/6e/e8efa0e78de00db0aee82c0cf9e8b3f2027efd7f8a71f859d8f4be8e98ef/jiter-0.12.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:5c1860627048e302a528333c9307c818c547f214d8659b0705d2195e1a94b274", size = 319855, upload-time = "2025-11-09T20:47:24.779Z" },
+    { url = "https://files.pythonhosted.org/packages/20/26/894cd88e60b5d58af53bec5c6759d1292bd0b37a8b5f60f07abf7a63ae5f/jiter-0.12.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:df37577a4f8408f7e0ec3205d2a8f87672af8f17008358063a4d6425b6081ce3", size = 350171, upload-time = "2025-11-09T20:47:26.469Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/27/a7b818b9979ac31b3763d25f3653ec3a954044d5e9f5d87f2f247d679fd1/jiter-0.12.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:75fdd787356c1c13a4f40b43c2156276ef7a71eb487d98472476476d803fb2cf", size = 365590, upload-time = "2025-11-09T20:47:27.918Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/7e/e46195801a97673a83746170b17984aa8ac4a455746354516d02ca5541b4/jiter-0.12.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1eb5db8d9c65b112aacf14fcd0faae9913d07a8afea5ed06ccdd12b724e966a1", size = 479462, upload-time = "2025-11-09T20:47:29.654Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/75/f833bfb009ab4bd11b1c9406d333e3b4357709ed0570bb48c7c06d78c7dd/jiter-0.12.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:73c568cc27c473f82480abc15d1301adf333a7ea4f2e813d6a2c7d8b6ba8d0df", size = 378983, upload-time = "2025-11-09T20:47:31.026Z" },
+    { url = "https://files.pythonhosted.org/packages/71/b3/7a69d77943cc837d30165643db753471aff5df39692d598da880a6e51c24/jiter-0.12.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4321e8a3d868919bcb1abb1db550d41f2b5b326f72df29e53b2df8b006eb9403", size = 361328, upload-time = "2025-11-09T20:47:33.286Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/ac/a78f90caf48d65ba70d8c6efc6f23150bc39dc3389d65bbec2a95c7bc628/jiter-0.12.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:0a51bad79f8cc9cac2b4b705039f814049142e0050f30d91695a2d9a6611f126", size = 386740, upload-time = "2025-11-09T20:47:34.703Z" },
+    { url = "https://files.pythonhosted.org/packages/39/b6/5d31c2cc8e1b6a6bcf3c5721e4ca0a3633d1ab4754b09bc7084f6c4f5327/jiter-0.12.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:2a67b678f6a5f1dd6c36d642d7db83e456bc8b104788262aaefc11a22339f5a9", size = 520875, upload-time = "2025-11-09T20:47:36.058Z" },
+    { url = "https://files.pythonhosted.org/packages/30/b5/4df540fae4e9f68c54b8dab004bd8c943a752f0b00efd6e7d64aa3850339/jiter-0.12.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:efe1a211fe1fd14762adea941e3cfd6c611a136e28da6c39272dbb7a1bbe6a86", size = 511457, upload-time = "2025-11-09T20:47:37.932Z" },
+    { url = "https://files.pythonhosted.org/packages/07/65/86b74010e450a1a77b2c1aabb91d4a91dd3cd5afce99f34d75fd1ac64b19/jiter-0.12.0-cp312-cp312-win32.whl", hash = "sha256:d779d97c834b4278276ec703dc3fc1735fca50af63eb7262f05bdb4e62203d44", size = 204546, upload-time = "2025-11-09T20:47:40.47Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/c7/6659f537f9562d963488e3e55573498a442503ced01f7e169e96a6110383/jiter-0.12.0-cp312-cp312-win_amd64.whl", hash = "sha256:e8269062060212b373316fe69236096aaf4c49022d267c6736eebd66bbbc60bb", size = 205196, upload-time = "2025-11-09T20:47:41.794Z" },
+    { url = "https://files.pythonhosted.org/packages/21/f4/935304f5169edadfec7f9c01eacbce4c90bb9a82035ac1de1f3bd2d40be6/jiter-0.12.0-cp312-cp312-win_arm64.whl", hash = "sha256:06cb970936c65de926d648af0ed3d21857f026b1cf5525cb2947aa5e01e05789", size = 186100, upload-time = "2025-11-09T20:47:43.007Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/da/3e1fbd1f03f89ff0b4469d481be0b5cf2880c8e7b56fd80303b3ab5ae52d/jiter-0.12.0-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:c9d28b218d5f9e5f69a0787a196322a5056540cb378cac8ff542b4fa7219966c", size = 319378, upload-time = "2025-11-09T20:48:51.761Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/4e/e07d69285e9e19a153050a6d281d2f0968600753a8fed8a3a141d6ffc140/jiter-0.12.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:d0ee12028daf8cfcf880dd492349a122a64f42c059b6c62a2b0c96a83a8da820", size = 312195, upload-time = "2025-11-09T20:48:53.547Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/82/1f1cb5231b36af9f3d6d5b6030e70110faf14fd143419fc5fe7d852e691a/jiter-0.12.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1b135ebe757a82d67ed2821526e72d0acf87dd61f6013e20d3c45b8048af927b", size = 352777, upload-time = "2025-11-09T20:48:55.058Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/5e/728393bbbc99b31e8f7a4fdd8fa55e455a0a9648f79097d9088baf1f676f/jiter-0.12.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:15d7fafb81af8a9e3039fc305529a61cd933eecee33b4251878a1c89859552a3", size = 370738, upload-time = "2025-11-09T20:48:56.632Z" },
+    { url = "https://files.pythonhosted.org/packages/30/08/ac92f0df7b14ac82f2fe0a382a8000e600ab90af95798d4a7db0c1bd0736/jiter-0.12.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:92d1f41211d8a8fe412faad962d424d334764c01dac6691c44691c2e4d3eedaf", size = 483744, upload-time = "2025-11-09T20:48:58.006Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/f4/dbfa4e759a2b82e969a14c3d0a91b176f1ed94717183a2f495cf94a651b9/jiter-0.12.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3a64a48d7c917b8f32f25c176df8749ecf08cec17c466114727efe7441e17f6d", size = 382888, upload-time = "2025-11-09T20:48:59.471Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/d9/b86fff7f748b0bb54222a8f132ffaf4d1be56b4591fa76d3cfdd701a33e5/jiter-0.12.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:122046f3b3710b85de99d9aa2f3f0492a8233a2f54a64902b096efc27ea747b5", size = 366465, upload-time = "2025-11-09T20:49:01.408Z" },
+    { url = "https://files.pythonhosted.org/packages/93/3c/1152d8b433317a568927e13c1b125c680e6c058ff5d304833be8469bd4f2/jiter-0.12.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:27ec39225e03c32c6b863ba879deb427882f243ae46f0d82d68b695fa5b48b40", size = 392603, upload-time = "2025-11-09T20:49:02.784Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/92/ff19d8fb87f3f9438eb7464862c8d0126455bc046b345d59b21443640c62/jiter-0.12.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:26b9e155ddc132225a39b1995b3b9f0fe0f79a6d5cbbeacf103271e7d309b404", size = 523780, upload-time = "2025-11-09T20:49:04.42Z" },
+    { url = "https://files.pythonhosted.org/packages/87/3a/4260e2d84e4a293c36d2a8e8b8dcd69609c671f3bd310e4625359217c517/jiter-0.12.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:9ab05b7c58e29bb9e60b70c2e0094c98df79a1e42e397b9bb6eaa989b7a66dd0", size = 514874, upload-time = "2025-11-09T20:49:05.844Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/f7/574d2cb79e86feb035ade18c2254da71d04417555907c9df51dd6b183426/jiter-0.12.0-cp39-cp39-win32.whl", hash = "sha256:59f9f9df87ed499136db1c2b6c9efb902f964bed42a582ab7af413b6a293e7b0", size = 208329, upload-time = "2025-11-09T20:49:07.444Z" },
+    { url = "https://files.pythonhosted.org/packages/05/ce/50725ec39782d8c973f19ae2d7dd3d192d01332c7cbde48c75e16a3e85a9/jiter-0.12.0-cp39-cp39-win_amd64.whl", hash = "sha256:d3719596a1ebe7a48a498e8d5d0c4bf7553321d4c3eee1d620628d51351a3928", size = 206557, upload-time = "2025-11-09T20:49:08.888Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/54/5339ef1ecaa881c6948669956567a64d2670941925f245c434f494ffb0e5/jiter-0.12.0-graalpy311-graalpy242_311_native-macosx_10_12_x86_64.whl", hash = "sha256:4739a4657179ebf08f85914ce50332495811004cc1747852e8b2041ed2aab9b8", size = 311144, upload-time = "2025-11-09T20:49:10.503Z" },
+    { url = "https://files.pythonhosted.org/packages/27/74/3446c652bffbd5e81ab354e388b1b5fc1d20daac34ee0ed11ff096b1b01a/jiter-0.12.0-graalpy311-graalpy242_311_native-macosx_11_0_arm64.whl", hash = "sha256:41da8def934bf7bec16cb24bd33c0ca62126d2d45d81d17b864bd5ad721393c3", size = 305877, upload-time = "2025-11-09T20:49:12.269Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/f4/ed76ef9043450f57aac2d4fbeb27175aa0eb9c38f833be6ef6379b3b9a86/jiter-0.12.0-graalpy311-graalpy242_311_native-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9c44ee814f499c082e69872d426b624987dbc5943ab06e9bbaa4f81989fdb79e", size = 340419, upload-time = "2025-11-09T20:49:13.803Z" },
+    { url = "https://files.pythonhosted.org/packages/21/01/857d4608f5edb0664aa791a3d45702e1a5bcfff9934da74035e7b9803846/jiter-0.12.0-graalpy311-graalpy242_311_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cd2097de91cf03eaa27b3cbdb969addf83f0179c6afc41bbc4513705e013c65d", size = 347212, upload-time = "2025-11-09T20:49:15.643Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/f5/12efb8ada5f5c9edc1d4555fe383c1fb2eac05ac5859258a72d61981d999/jiter-0.12.0-graalpy312-graalpy250_312_native-macosx_10_12_x86_64.whl", hash = "sha256:e8547883d7b96ef2e5fe22b88f8a4c8725a56e7f4abafff20fd5272d634c7ecb", size = 309974, upload-time = "2025-11-09T20:49:17.187Z" },
+    { url = "https://files.pythonhosted.org/packages/85/15/d6eb3b770f6a0d332675141ab3962fd4a7c270ede3515d9f3583e1d28276/jiter-0.12.0-graalpy312-graalpy250_312_native-macosx_11_0_arm64.whl", hash = "sha256:89163163c0934854a668ed783a2546a0617f71706a2551a4a0666d91ab365d6b", size = 304233, upload-time = "2025-11-09T20:49:18.734Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/3e/e7e06743294eea2cf02ced6aa0ff2ad237367394e37a0e2b4a1108c67a36/jiter-0.12.0-graalpy312-graalpy250_312_native-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d96b264ab7d34bbb2312dedc47ce07cd53f06835eacbc16dde3761f47c3a9e7f", size = 338537, upload-time = "2025-11-09T20:49:20.317Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/9c/6753e6522b8d0ef07d3a3d239426669e984fb0eba15a315cdbc1253904e4/jiter-0.12.0-graalpy312-graalpy250_312_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c24e864cb30ab82311c6425655b0cdab0a98c5d973b065c66a3f020740c2324c", size = 346110, upload-time = "2025-11-09T20:49:21.817Z" },
 ]
 
 [[package]]
@@ -1775,8 +1704,10 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "attrs" },
     { name = "jsonschema-specifications" },
-    { name = "referencing" },
-    { name = "rpds-py" },
+    { name = "referencing", version = "0.36.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "referencing", version = "0.37.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "rpds-py", version = "0.27.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "rpds-py", version = "0.29.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/74/69/f7185de793a29082a9f3c7728268ffb31cb5095131a9c139a74078e27336/jsonschema-4.25.1.tar.gz", hash = "sha256:e4a9655ce0da0c0b67a085847e00a3a51449e1157f4f75e9fb5aa545e122eb85", size = 357342, upload-time = "2025-08-18T17:03:50.038Z" }
 wheels = [
@@ -1788,7 +1719,8 @@ name = "jsonschema-specifications"
 version = "2025.9.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "referencing" },
+    { name = "referencing", version = "0.36.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "referencing", version = "0.37.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/19/74/a633ee74eb36c44aa6d1095e7cc5569bebf04342ee146178e2d36600708b/jsonschema_specifications-2025.9.1.tar.gz", hash = "sha256:b540987f239e745613c7a9176f3edb72b832a4ac465cf02712288397832b5e8d", size = 32855, upload-time = "2025-09-08T01:34:59.186Z" }
 wheels = [
@@ -1804,7 +1736,7 @@ resolution-markers = [
 ]
 dependencies = [
     { name = "absl-py", marker = "python_full_version < '3.10'" },
-    { name = "h5py", marker = "python_full_version < '3.10'" },
+    { name = "h5py", version = "3.14.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "ml-dtypes", marker = "python_full_version < '3.10'" },
     { name = "namex", marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
@@ -1819,7 +1751,7 @@ wheels = [
 
 [[package]]
 name = "keras"
-version = "3.11.3"
+version = "3.12.0"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -1828,7 +1760,7 @@ resolution-markers = [
 ]
 dependencies = [
     { name = "absl-py", marker = "python_full_version >= '3.10'" },
-    { name = "h5py", marker = "python_full_version >= '3.10'" },
+    { name = "h5py", version = "3.15.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "ml-dtypes", marker = "python_full_version >= '3.10'" },
     { name = "namex", marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
@@ -1836,18 +1768,18 @@ dependencies = [
     { name = "packaging", marker = "python_full_version >= '3.10'" },
     { name = "rich", marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/6a/89/646425fe9a46f9053430e1271f817c36041c6f33469950a3caafc3d2591e/keras-3.11.3.tar.gz", hash = "sha256:efda616835c31b7d916d72303ef9adec1257320bc9fd4b2b0138840fc65fb5b7", size = 1065906, upload-time = "2025-08-21T22:08:57.643Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/9b/b8/8df141314a64a31d3a21762658826f716cdf4c261c7fcdb3f729958def55/keras-3.12.0.tar.gz", hash = "sha256:536e3f8385a05ae04e82e08715a1a59988578087e187b04cb0a6fad11743f07f", size = 1129187, upload-time = "2025-10-27T20:23:11.574Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/94/5b/4c778cc921ce4b864b238f63f8e3ff6e954ab19b80c9fa680593ad8093d4/keras-3.11.3-py3-none-any.whl", hash = "sha256:f484f050e05ee400455b05ec8c36ed35edc34de94256b6073f56cfe68f65491f", size = 1408438, upload-time = "2025-08-21T22:08:55.858Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/61/cc8be27bd65082440754be443b17b6f7c185dec5e00dfdaeab4f8662e4a8/keras-3.12.0-py3-none-any.whl", hash = "sha256:02b69e007d5df8042286c3bcc2a888539e3e487590ffb08f6be1b4354df50aa8", size = 1474424, upload-time = "2025-10-27T20:23:09.571Z" },
 ]
 
 [[package]]
-name = "lark"
-version = "1.2.2"
+name = "latex2mathml"
+version = "3.78.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/af/60/bc7622aefb2aee1c0b4ba23c1446d3e30225c8770b38d7aedbfb65ca9d5a/lark-1.2.2.tar.gz", hash = "sha256:ca807d0162cd16cef15a8feecb862d7319e7a09bdb13aef927968e45040fed80", size = 252132, upload-time = "2024-08-13T19:49:00.652Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/1a/26/57b1034c08922d0aefea79430a5e0006ffaee4f0ec59d566613f667ab2f7/latex2mathml-3.78.1.tar.gz", hash = "sha256:f941db80bf41db33f31df87b304e8b588f8166b813b0257c11c98f7a9d0aac71", size = 74030, upload-time = "2025-08-29T23:34:23.178Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/2d/00/d90b10b962b4277f5e64a78b6609968859ff86889f5b898c1a778c06ec00/lark-1.2.2-py3-none-any.whl", hash = "sha256:c2276486b02f0f1b90be155f2c8ba4a8e194d42775786db622faccd652d8e80c", size = 111036, upload-time = "2024-08-13T19:48:58.603Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/76/d661ea2e529c3d464f9efd73f9ac31626b45279eb4306e684054ea20e3d4/latex2mathml-3.78.1-py3-none-any.whl", hash = "sha256:f089b6d75e85b937f99693c93e8c16c0804008672c3dd2a3d25affd36f238100", size = 73892, upload-time = "2025-08-29T23:34:21.98Z" },
 ]
 
 [[package]]
@@ -1882,17 +1814,40 @@ sdist = { url = "https://files.pythonhosted.org/packages/e4/b4/c8cd17629ced0b964
 
 [[package]]
 name = "llguidance"
-version = "0.7.30"
+version = "1.3.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/95/48/3f7a9d3ff1b36bba92b5107a3a21286821227afe9ea464736133994d61fb/llguidance-1.3.0.tar.gz", hash = "sha256:861249afd51dc325646834462ea827e57a5c2b2042e108e6aae7059fdad9104d", size = 1070460, upload-time = "2025-10-20T19:58:44.164Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3b/33/be5acb85cd8cdc4afde33d9c234eece9f318e087920255af3c05864cd3e7/llguidance-1.3.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:f7685222660a762e481ac633d49cc559c64980fe2ee59c8f932a5bb5cbc0c2c2", size = 3220647, upload-time = "2025-10-20T19:58:42.542Z" },
+    { url = "https://files.pythonhosted.org/packages/82/e6/b48bda5b15efeaeb62bd0dba8fc6a01d4ae5457a85dbb5d18632385fe15c/llguidance-1.3.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:098030ff0687261a3f1bd54cf21fe951fc861d56d37a0671250dd36677eaf224", size = 3099830, upload-time = "2025-10-20T19:58:40.826Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/11/44389d3d1526d7a5c38ffd587a5ebc61d7bee443ac1dea95f2089ad58f5f/llguidance-1.3.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6f6caca5d78db7f76e1fbb0fff8607b861c32d47fa3d5dee2fc49de27ee269df", size = 2835242, upload-time = "2025-10-20T19:58:34.518Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/ca/53ea256396405e4dee70d5a4a35e18543408e18bb16b251d6ca6b5d80310/llguidance-1.3.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:0612bb3f034d2487b6e8f9561f02a94a6039d88273bf0c5c539a3bd3895e47d2", size = 3297480, upload-time = "2025-10-20T19:58:37.033Z" },
+    { url = "https://files.pythonhosted.org/packages/83/a8/1ff2bedb8f9acb46a2d2d603415d272bb622c142ea86f5b95445cc6e366c/llguidance-1.3.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc17e9dd602c3879bf91664a64bf72f54c74dbfbeb24ccfab6a5fe435b12f7aa", size = 3033133, upload-time = "2025-10-20T19:58:38.721Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/a7/9b8086c0cfdddf3f6d47b173a404fa7ac46272f7affbee082c36740f4f1c/llguidance-1.3.0-cp39-abi3-win32.whl", hash = "sha256:2f6f558485a43e273fc5c6c974a9a3ace5d5e170076db9b40e0560e41c3ff18f", size = 2598109, upload-time = "2025-10-20T19:58:47.656Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/7e/809349638231f469b9056c0e1bfd924d5ef5558b3b3ec72d093b6fad33b1/llguidance-1.3.0-cp39-abi3-win_amd64.whl", hash = "sha256:1d1cd1c8618d1a13605d3e057c978651e551c8c469b481ee4041f1d6c436002d", size = 2789946, upload-time = "2025-10-20T19:58:45.958Z" },
+]
+
+[[package]]
+name = "llguidance"
+version = "1.4.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/bf/38/d1ef3ae08d8d857e5e0690c5b1e07bf7eb4a1cae5881d87215826dc6cadb/llguidance-0.7.30.tar.gz", hash = "sha256:e93bf75f2b6e48afb86a5cee23038746975e1654672bf5ba0ae75f7d4d4a2248", size = 1055528, upload-time = "2025-06-23T00:23:49.247Z" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/a7/dd/9edb6b87b0e52702b0f98362f08806b151a0c73216bec203dfc470ac16a2/llguidance-1.4.0.tar.gz", hash = "sha256:f1371639ff1e96b1cb4e60ad64b8e514a533b12025e83954025c79c5c9cad8f7", size = 1075906, upload-time = "2025-11-18T18:58:56.702Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b3/e1/694c89986fcae7777184fc8b22baa0976eba15a6847221763f6ad211fc1f/llguidance-0.7.30-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:c80af02c118d2b0526bcecaab389af2ed094537a069b0fc724cd2a2f2ba3990f", size = 3327974, upload-time = "2025-06-23T00:23:47.556Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/77/ab7a548ae189dc23900fdd37803c115c2339b1223af9e8eb1f4329b5935a/llguidance-0.7.30-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:00a256d532911d2cf5ba4ef63e182944e767dd2402f38d63002016bc37755958", size = 3210709, upload-time = "2025-06-23T00:23:45.872Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/5b/6a166564b14f9f805f0ea01ec233a84f55789cb7eeffe1d6224ccd0e6cdd/llguidance-0.7.30-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:af8741c867e4bc7e42f7cdc68350c076b4edd0ca10ecefbde75f15a9f6bc25d0", size = 14867038, upload-time = "2025-06-23T00:23:39.571Z" },
-    { url = "https://files.pythonhosted.org/packages/17/ec/69507bdb36767f9b6ff2e290660a9b5afdda0fb8a7903faa37f37c6c2a72/llguidance-0.7.30-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f4a327a30dd37d86dd6347861ac8de3521fc1dbef9475296c06744e5b40ffc54", size = 15142936, upload-time = "2025-06-23T00:23:41.944Z" },
-    { url = "https://files.pythonhosted.org/packages/af/80/5a40b9689f17612434b820854cba9b8cabd5142072c491b5280fe5f7a35e/llguidance-0.7.30-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9edc409b9decd6cffba5f5bf3b4fbd7541f95daa8cbc9510cbf96c6ab1ffc153", size = 15004926, upload-time = "2025-06-23T00:23:43.965Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/bc/2d2f9b446bb3e51e4dd4db290590afee03ae29163f417168569f0361204c/llguidance-0.7.30-cp39-abi3-win32.whl", hash = "sha256:a0d52b8d1b2d3b0e661e3f953ecccfa16644f302026b3067a4815c1baa2ae643", size = 2585627, upload-time = "2025-06-23T00:23:52.39Z" },
-    { url = "https://files.pythonhosted.org/packages/99/47/58e49a118b514855b245f8a962c6aaf9a5cc95a0f61eac7e230e691c7b7e/llguidance-0.7.30-cp39-abi3-win_amd64.whl", hash = "sha256:05234ecceea7c9c6ff13b9739112043173a3bcb88cae860249b20335a07b3075", size = 2796878, upload-time = "2025-06-23T00:23:51Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/a7/c5e5e31b2ea2d606638653df5a1f4033d1f060c57665a1b00eb53501cdfb/llguidance-1.4.0-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:117859752f35f61d42d8d7c244fe5f93f84f336d6e7f69b1039e716ec6c269f4", size = 3280406, upload-time = "2025-11-18T18:58:55.021Z" },
+    { url = "https://files.pythonhosted.org/packages/19/e9/3c58aa751bef282f107c10a5c5993196f289bf14f6373122cd90063ebac1/llguidance-1.4.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:1dff9e138aa0ae09e8249e45cf4ab666d5d4d32a3b9afff03be22f0a561e6440", size = 3179692, upload-time = "2025-11-18T18:58:53.245Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/a5/f20f061c957f4978c396f0feea4b3d938674b4c8a8f9915b684712beb66e/llguidance-1.4.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6a88ea5e856a1f9cbeee0995446b409b031b58e6a08440e79d89ee6352d0a475", size = 2930081, upload-time = "2025-11-18T18:58:47.604Z" },
+    { url = "https://files.pythonhosted.org/packages/14/85/521fd49e11de7e6ecd11160fc424ebfe48b07c641d3e9bd8ff821acbaae9/llguidance-1.4.0-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:544210256ce5b5e3163af6e1ce6e8f13c38b49ac875cd936ac37837bea132059", size = 3392301, upload-time = "2025-11-18T18:58:49.607Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/d9/6754c45e37e533007abe327f3b168f1b70fb6af2838420c8d09398eeb322/llguidance-1.4.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1f96770da4b919f1f3bda2395a6e131134c911496d62e7328f853604d077c1c7", size = 3120112, upload-time = "2025-11-18T18:58:51.269Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/cf/52132eed1096393b432e2ce18d9363796a7003e8a843922c7f7a4630fb61/llguidance-1.4.0-cp39-abi3-win32.whl", hash = "sha256:1aa61e5990f3ee389e98efe0e46c46d7fe938158dfaaaf196c8ff4d9bfe2d030", size = 2667519, upload-time = "2025-11-18T18:59:01.521Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/62/773e42e4d57679cd2fd35c5dfb6bb028302653670ca189f8bfe3a7a0071b/llguidance-1.4.0-cp39-abi3-win_amd64.whl", hash = "sha256:f94b3ca562900947bac233a045d17ffa5f49f9499b2797f798c1def42cd3503a", size = 2909873, upload-time = "2025-11-18T18:58:58.554Z" },
 ]
 
 [[package]]
@@ -1928,51 +1883,39 @@ wheels = [
 
 [[package]]
 name = "llvmlite"
-version = "0.44.0"
+version = "0.45.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
-    "python_full_version == '3.11.*'",
-    "python_full_version == '3.10.*'",
-]
-sdist = { url = "https://files.pythonhosted.org/packages/89/6a/95a3d3610d5c75293d5dbbb2a76480d5d4eeba641557b69fe90af6c5b84e/llvmlite-0.44.0.tar.gz", hash = "sha256:07667d66a5d150abed9157ab6c0b9393c9356f229784a4385c02f99e94fc94d4", size = 171880, upload-time = "2025-01-20T11:14:41.342Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/41/75/d4863ddfd8ab5f6e70f4504cf8cc37f4e986ec6910f4ef8502bb7d3c1c71/llvmlite-0.44.0-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:9fbadbfba8422123bab5535b293da1cf72f9f478a65645ecd73e781f962ca614", size = 28132306, upload-time = "2025-01-20T11:12:18.634Z" },
-    { url = "https://files.pythonhosted.org/packages/37/d9/6e8943e1515d2f1003e8278819ec03e4e653e2eeb71e4d00de6cfe59424e/llvmlite-0.44.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:cccf8eb28f24840f2689fb1a45f9c0f7e582dd24e088dcf96e424834af11f791", size = 26201096, upload-time = "2025-01-20T11:12:24.544Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/46/8ffbc114def88cc698906bf5acab54ca9fdf9214fe04aed0e71731fb3688/llvmlite-0.44.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7202b678cdf904823c764ee0fe2dfe38a76981f4c1e51715b4cb5abb6cf1d9e8", size = 42361859, upload-time = "2025-01-20T11:12:31.839Z" },
-    { url = "https://files.pythonhosted.org/packages/30/1c/9366b29ab050a726af13ebaae8d0dff00c3c58562261c79c635ad4f5eb71/llvmlite-0.44.0-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:40526fb5e313d7b96bda4cbb2c85cd5374e04d80732dd36a282d72a560bb6408", size = 41184199, upload-time = "2025-01-20T11:12:40.049Z" },
-    { url = "https://files.pythonhosted.org/packages/69/07/35e7c594b021ecb1938540f5bce543ddd8713cff97f71d81f021221edc1b/llvmlite-0.44.0-cp310-cp310-win_amd64.whl", hash = "sha256:41e3839150db4330e1b2716c0be3b5c4672525b4c9005e17c7597f835f351ce2", size = 30332381, upload-time = "2025-01-20T11:12:47.054Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/e2/86b245397052386595ad726f9742e5223d7aea999b18c518a50e96c3aca4/llvmlite-0.44.0-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:eed7d5f29136bda63b6d7804c279e2b72e08c952b7c5df61f45db408e0ee52f3", size = 28132305, upload-time = "2025-01-20T11:12:53.936Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/ec/506902dc6870249fbe2466d9cf66d531265d0f3a1157213c8f986250c033/llvmlite-0.44.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:ace564d9fa44bb91eb6e6d8e7754977783c68e90a471ea7ce913bff30bd62427", size = 26201090, upload-time = "2025-01-20T11:12:59.847Z" },
-    { url = "https://files.pythonhosted.org/packages/99/fe/d030f1849ebb1f394bb3f7adad5e729b634fb100515594aca25c354ffc62/llvmlite-0.44.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c5d22c3bfc842668168a786af4205ec8e3ad29fb1bc03fd11fd48460d0df64c1", size = 42361858, upload-time = "2025-01-20T11:13:07.623Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/7a/ce6174664b9077fc673d172e4c888cb0b128e707e306bc33fff8c2035f0d/llvmlite-0.44.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f01a394e9c9b7b1d4e63c327b096d10f6f0ed149ef53d38a09b3749dcf8c9610", size = 41184200, upload-time = "2025-01-20T11:13:20.058Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/c6/258801143975a6d09a373f2641237992496e15567b907a4d401839d671b8/llvmlite-0.44.0-cp311-cp311-win_amd64.whl", hash = "sha256:d8489634d43c20cd0ad71330dde1d5bc7b9966937a263ff1ec1cebb90dc50955", size = 30331193, upload-time = "2025-01-20T11:13:26.976Z" },
-    { url = "https://files.pythonhosted.org/packages/15/86/e3c3195b92e6e492458f16d233e58a1a812aa2bfbef9bdd0fbafcec85c60/llvmlite-0.44.0-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:1d671a56acf725bf1b531d5ef76b86660a5ab8ef19bb6a46064a705c6ca80aad", size = 28132297, upload-time = "2025-01-20T11:13:32.57Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/53/373b6b8be67b9221d12b24125fd0ec56b1078b660eeae266ec388a6ac9a0/llvmlite-0.44.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:5f79a728e0435493611c9f405168682bb75ffd1fbe6fc360733b850c80a026db", size = 26201105, upload-time = "2025-01-20T11:13:38.744Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/da/8341fd3056419441286c8e26bf436923021005ece0bff5f41906476ae514/llvmlite-0.44.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c0143a5ef336da14deaa8ec26c5449ad5b6a2b564df82fcef4be040b9cacfea9", size = 42361901, upload-time = "2025-01-20T11:13:46.711Z" },
-    { url = "https://files.pythonhosted.org/packages/53/ad/d79349dc07b8a395a99153d7ce8b01d6fcdc9f8231355a5df55ded649b61/llvmlite-0.44.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d752f89e31b66db6f8da06df8b39f9b91e78c5feea1bf9e8c1fba1d1c24c065d", size = 41184247, upload-time = "2025-01-20T11:13:56.159Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/3b/a9a17366af80127bd09decbe2a54d8974b6d8b274b39bf47fbaedeec6307/llvmlite-0.44.0-cp312-cp312-win_amd64.whl", hash = "sha256:eae7e2d4ca8f88f89d315b48c6b741dcb925d6a1042da694aa16ab3dd4cbd3a1", size = 30332380, upload-time = "2025-01-20T11:14:02.442Z" },
-]
-
-[[package]]
-name = "lm-format-enforcer"
-version = "0.11.3"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "interegular" },
-    { name = "packaging" },
-    { name = "pydantic" },
-    { name = "pyyaml" },
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/84/d5/41cd417ba7dfdbbcfe46cebf81fb3dfd7c591b89897560ad05bb410a465d/lm_format_enforcer-0.11.3.tar.gz", hash = "sha256:e68081c108719cce284a9bcc889709b26ffb085a1945b5eba3a12cfa96d528da", size = 40258, upload-time = "2025-08-24T19:37:47.527Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/99/8d/5baf1cef7f9c084fb35a8afbde88074f0d6a727bc63ef764fe0e7543ba40/llvmlite-0.45.1.tar.gz", hash = "sha256:09430bb9d0bb58fc45a45a57c7eae912850bedc095cd0810a57de109c69e1c32", size = 185600, upload-time = "2025-10-01T17:59:52.046Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/a0/ef/11292bb0b85cf4c93447cab5a29f64576ed14d3ab4280e35ddd23486594a/lm_format_enforcer-0.11.3-py3-none-any.whl", hash = "sha256:cf586350875def1ae7a8fba84fcbbfc8371424b6c9d05c1fcba70aa233fbf06f", size = 45418, upload-time = "2025-08-24T19:37:46.325Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/6d/585c84ddd9d2a539a3c3487792b3cf3f988e28ec4fa281bf8b0e055e1166/llvmlite-0.45.1-cp310-cp310-macosx_10_15_x86_64.whl", hash = "sha256:1b1af0c910af0978aa55fa4f60bbb3e9f39b41e97c2a6d94d199897be62ba07a", size = 43043523, upload-time = "2025-10-01T18:02:58.621Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/34/992bd12d3ff245e0801bcf6013961daa8c19c9b9c2e61cb4b8bce94566f9/llvmlite-0.45.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:02a164db2d79088bbd6e0d9633b4fe4021d6379d7e4ac7cc85ed5f44b06a30c5", size = 37253122, upload-time = "2025-10-01T18:03:55.159Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/7b/6d7585998a5991fa74dc925aae57913ba8c7c2efff909de9d34cc1cd3c27/llvmlite-0.45.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f2d47f34e4029e6df3395de34cc1c66440a8d72712993a6e6168db228686711b", size = 56288210, upload-time = "2025-10-01T18:00:41.978Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/e2/a4abea058633bfc82eb08fd69ce242c118fdb9b0abad1fdcbe0bc6aedab5/llvmlite-0.45.1-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f7319e5f9f90720578a7f56fbc805bdfb4bc071b507c7611f170d631c3c0f1e0", size = 55140958, upload-time = "2025-10-01T18:01:55.694Z" },
+    { url = "https://files.pythonhosted.org/packages/74/c0/233468e96ed287b953239c3b24b1d69df47c6ba9262bfdca98eda7e83a04/llvmlite-0.45.1-cp310-cp310-win_amd64.whl", hash = "sha256:4edb62e685867799e336723cb9787ec6598d51d0b1ed9af0f38e692aa757e898", size = 38132232, upload-time = "2025-10-01T18:04:41.538Z" },
+    { url = "https://files.pythonhosted.org/packages/04/ad/9bdc87b2eb34642c1cfe6bcb4f5db64c21f91f26b010f263e7467e7536a3/llvmlite-0.45.1-cp311-cp311-macosx_10_15_x86_64.whl", hash = "sha256:60f92868d5d3af30b4239b50e1717cb4e4e54f6ac1c361a27903b318d0f07f42", size = 43043526, upload-time = "2025-10-01T18:03:15.051Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/ea/c25c6382f452a943b4082da5e8c1665ce29a62884e2ec80608533e8e82d5/llvmlite-0.45.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:98baab513e19beb210f1ef39066288784839a44cd504e24fff5d17f1b3cf0860", size = 37253118, upload-time = "2025-10-01T18:04:06.783Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/af/85fc237de98b181dbbe8647324331238d6c52a3554327ccdc83ced28efba/llvmlite-0.45.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3adc2355694d6a6fbcc024d59bb756677e7de506037c878022d7b877e7613a36", size = 56288209, upload-time = "2025-10-01T18:01:00.168Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/df/3daf95302ff49beff4230065e3178cd40e71294968e8d55baf4a9e560814/llvmlite-0.45.1-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:2f3377a6db40f563058c9515dedcc8a3e562d8693a106a28f2ddccf2c8fcf6ca", size = 55140958, upload-time = "2025-10-01T18:02:11.199Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/56/4c0d503fe03bac820ecdeb14590cf9a248e120f483bcd5c009f2534f23f0/llvmlite-0.45.1-cp311-cp311-win_amd64.whl", hash = "sha256:f9c272682d91e0d57f2a76c6d9ebdfccc603a01828cdbe3d15273bdca0c3363a", size = 38132232, upload-time = "2025-10-01T18:04:52.181Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/7c/82cbd5c656e8991bcc110c69d05913be2229302a92acb96109e166ae31fb/llvmlite-0.45.1-cp312-cp312-macosx_10_15_x86_64.whl", hash = "sha256:28e763aba92fe9c72296911e040231d486447c01d4f90027c8e893d89d49b20e", size = 43043524, upload-time = "2025-10-01T18:03:30.666Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/bc/5314005bb2c7ee9f33102c6456c18cc81745d7055155d1218f1624463774/llvmlite-0.45.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:1a53f4b74ee9fd30cb3d27d904dadece67a7575198bd80e687ee76474620735f", size = 37253123, upload-time = "2025-10-01T18:04:18.177Z" },
+    { url = "https://files.pythonhosted.org/packages/96/76/0f7154952f037cb320b83e1c952ec4a19d5d689cf7d27cb8a26887d7bbc1/llvmlite-0.45.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5b3796b1b1e1c14dcae34285d2f4ea488402fbd2c400ccf7137603ca3800864f", size = 56288211, upload-time = "2025-10-01T18:01:24.079Z" },
+    { url = "https://files.pythonhosted.org/packages/00/b1/0b581942be2683ceb6862d558979e87387e14ad65a1e4db0e7dd671fa315/llvmlite-0.45.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:779e2f2ceefef0f4368548685f0b4adde34e5f4b457e90391f570a10b348d433", size = 55140958, upload-time = "2025-10-01T18:02:30.482Z" },
+    { url = "https://files.pythonhosted.org/packages/33/94/9ba4ebcf4d541a325fd8098ddc073b663af75cc8b065b6059848f7d4dce7/llvmlite-0.45.1-cp312-cp312-win_amd64.whl", hash = "sha256:9e6c9949baf25d9aa9cd7cf0f6d011b9ca660dd17f5ba2b23bdbdb77cc86b116", size = 38132231, upload-time = "2025-10-01T18:05:03.664Z" },
 ]
 
 [[package]]
 name = "markdown"
 version = "3.9"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
     { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
 ]
@@ -1981,6 +1924,20 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/70/ae/44c4a6a4cbb496d93c6257954260fe3a6e91b7bed2240e5dad2a717f5111/markdown-3.9-py3-none-any.whl", hash = "sha256:9f4d91ed810864ea88a6f32c07ba8bee1346c0cc1f6b1f9f6c822f2a9667d280", size = 107441, upload-time = "2025-09-04T20:25:21.784Z" },
 ]
 
+[[package]]
+name = "markdown"
+version = "3.10"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/7d/ab/7dd27d9d863b3376fcf23a5a13cb5d024aed1db46f963f1b5735ae43b3be/markdown-3.10.tar.gz", hash = "sha256:37062d4f2aa4b2b6b32aefb80faa300f82cc790cb949a35b8caede34f2b68c0e", size = 364931, upload-time = "2025-11-03T19:51:15.007Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/70/81/54e3ce63502cd085a0c556652a4e1b919c45a446bd1e5300e10c44c8c521/markdown-3.10-py3-none-any.whl", hash = "sha256:b5b99d6951e2e4948d939255596523444c0e677c669700b1d17aa4a8a464cb7c", size = 107678, upload-time = "2025-11-03T19:51:13.887Z" },
+]
+
 [[package]]
 name = "markdown-it-py"
 version = "3.0.0"
@@ -2013,6 +1970,22 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/94/54/e7d793b573f298e1c9013b8c4dade17d481164aa517d1d7148619c2cedbf/markdown_it_py-4.0.0-py3-none-any.whl", hash = "sha256:87327c59b172c5011896038353a81343b6754500a08cd7a4973bb48c6d578147", size = 87321, upload-time = "2025-08-11T12:57:51.923Z" },
 ]
 
+[[package]]
+name = "markdown2"
+version = "2.5.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/42/f8/b2ae8bf5f28f9b510ae097415e6e4cb63226bb28d7ee01aec03a755ba03b/markdown2-2.5.4.tar.gz", hash = "sha256:a09873f0b3c23dbfae589b0080587df52ad75bb09a5fa6559147554736676889", size = 145652, upload-time = "2025-07-27T16:16:24.307Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b8/06/2697b5043c3ecb720ce0d243fc7cf5024c0b5b1e450506e9b21939019963/markdown2-2.5.4-py3-none-any.whl", hash = "sha256:3c4b2934e677be7fec0e6f2de4410e116681f4ad50ec8e5ba7557be506d3f439", size = 49954, upload-time = "2025-07-27T16:16:23.026Z" },
+]
+
+[package.optional-dependencies]
+all = [
+    { name = "latex2mathml" },
+    { name = "pygments" },
+    { name = "wavedrom" },
+]
+
 [[package]]
 name = "markupsafe"
 version = "3.0.3"
@@ -2083,35 +2056,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/2c/19/04f9b178c2d8a15b076c8b5140708fa6ffc5601fb6f1e975537072df5b2a/mergedeep-1.3.4-py3-none-any.whl", hash = "sha256:70775750742b25c0d8f36c55aed03d24c3384d17c951b3175d898bd778ef0307", size = 6354, upload-time = "2021-02-05T18:55:29.583Z" },
 ]
 
-[[package]]
-name = "mistral-common"
-version = "1.8.5"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "jsonschema" },
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "pillow" },
-    { name = "pydantic" },
-    { name = "pydantic-extra-types", extra = ["pycountry"] },
-    { name = "requests" },
-    { name = "tiktoken" },
-    { name = "typing-extensions" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/ce/ff/1992a00ccc936f2c6e69ecb1f2cac678e0fd46c53c71bdab99eda4f89dfd/mistral_common-1.8.5.tar.gz", hash = "sha256:9f6204ede9c807f09040a208a9381ae78ef93e2e5a9cd5202dc12e712a025de8", size = 6331923, upload-time = "2025-09-12T06:43:01.937Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/91/4a/54e19c5e75939fd9418c7b806c21d12cf252ea2ba38f122b597272b459dd/mistral_common-1.8.5-py3-none-any.whl", hash = "sha256:f3cf87b61958a00485e603f3fe0530eb509d7e9b2f7178329dcd260e307eced1", size = 6515140, upload-time = "2025-09-12T06:42:59.622Z" },
-]
-
-[package.optional-dependencies]
-audio = [
-    { name = "soundfile" },
-    { name = "soxr" },
-]
-image = [
-    { name = "opencv-python-headless" },
-]
-
 [[package]]
 name = "mistralai"
 version = "1.9.11"
@@ -2136,12 +2080,13 @@ version = "1.6.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "colorama", marker = "sys_platform == 'win32'" },
     { name = "ghp-import" },
     { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
     { name = "jinja2" },
-    { name = "markdown" },
+    { name = "markdown", version = "3.9", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "markdown", version = "3.10", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "markupsafe" },
     { name = "mergedeep" },
     { name = "mkdocs-get-deps" },
@@ -2158,14 +2103,14 @@ wheels = [
 
 [[package]]
 name = "mkdocs-gen-files"
-version = "0.5.0"
+version = "0.6.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "mkdocs" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/48/85/2d634462fd59136197d3126ca431ffb666f412e3db38fd5ce3a60566303e/mkdocs_gen_files-0.5.0.tar.gz", hash = "sha256:4c7cf256b5d67062a788f6b1d035e157fc1a9498c2399be9af5257d4ff4d19bc", size = 7539, upload-time = "2023-04-27T19:48:04.894Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/61/35/f26349f7fa18414eb2e25d75a6fa9c7e3186c36e1d227c0b2d785a7bd5c4/mkdocs_gen_files-0.6.0.tar.gz", hash = "sha256:52022dc14dcc0451e05e54a8f5d5e7760351b6701eff816d1e9739577ec5635e", size = 8642, upload-time = "2025-11-23T12:13:22.124Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e7/0f/1e55b3fd490ad2cecb6e7b31892d27cb9fc4218ec1dab780440ba8579e74/mkdocs_gen_files-0.5.0-py3-none-any.whl", hash = "sha256:7ac060096f3f40bd19039e7277dd3050be9a453c8ac578645844d4d91d7978ea", size = 8380, upload-time = "2023-04-27T19:48:07.059Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/ec/72417415563c60ae01b36f0d497f1f4c803972f447ef4fb7f7746d6e07db/mkdocs_gen_files-0.6.0-py3-none-any.whl", hash = "sha256:815af15f3e2dbfda379629c1b95c02c8e6f232edf2a901186ea3b204ab1135b2", size = 8182, upload-time = "2025-11-23T12:13:20.756Z" },
 ]
 
 [[package]]
@@ -2175,7 +2120,8 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
     { name = "mergedeep" },
-    { name = "platformdirs" },
+    { name = "platformdirs", version = "4.4.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "platformdirs", version = "4.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pyyaml" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/98/f5/ed29cd50067784976f25ed0ed6fcd3c2ce9eb90650aa3b2796ddf7b6870b/mkdocs_get_deps-0.2.0.tar.gz", hash = "sha256:162b3d129c7fad9b19abfdcb9c1458a651628e4b1dea628ac68790fb3061c60c", size = 10239, upload-time = "2023-11-20T17:51:09.981Z" }
@@ -2185,86 +2131,135 @@ wheels = [
 
 [[package]]
 name = "ml-dtypes"
-version = "0.5.3"
+version = "0.5.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/78/a7/aad060393123cfb383956dca68402aff3db1e1caffd5764887ed5153f41b/ml_dtypes-0.5.3.tar.gz", hash = "sha256:95ce33057ba4d05df50b1f3cfefab22e351868a843b3b15a46c65836283670c9", size = 692316, upload-time = "2025-07-29T18:39:19.454Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/ac/bb/1f32124ab6d3a279ea39202fe098aea95b2d81ef0ce1d48612b6bf715e82/ml_dtypes-0.5.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:0a1d68a7cb53e3f640b2b6a34d12c0542da3dd935e560fdf463c0c77f339fc20", size = 667409, upload-time = "2025-07-29T18:38:17.321Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/ac/e002d12ae19136e25bb41c7d14d7e1a1b08f3c0e99a44455ff6339796507/ml_dtypes-0.5.3-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0cd5a6c711b5350f3cbc2ac28def81cd1c580075ccb7955e61e9d8f4bfd40d24", size = 4960702, upload-time = "2025-07-29T18:38:19.616Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/12/79e9954e6b3255a4b1becb191a922d6e2e94d03d16a06341ae9261963ae8/ml_dtypes-0.5.3-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:bdcf26c2dbc926b8a35ec8cbfad7eff1a8bd8239e12478caca83a1fc2c400dc2", size = 4933471, upload-time = "2025-07-29T18:38:21.809Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/aa/d1eff619e83cd1ddf6b561d8240063d978e5d887d1861ba09ef01778ec3a/ml_dtypes-0.5.3-cp310-cp310-win_amd64.whl", hash = "sha256:aecbd7c5272c82e54d5b99d8435fd10915d1bc704b7df15e4d9ca8dc3902be61", size = 206330, upload-time = "2025-07-29T18:38:23.663Z" },
-    { url = "https://files.pythonhosted.org/packages/af/f1/720cb1409b5d0c05cff9040c0e9fba73fa4c67897d33babf905d5d46a070/ml_dtypes-0.5.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:4a177b882667c69422402df6ed5c3428ce07ac2c1f844d8a1314944651439458", size = 667412, upload-time = "2025-07-29T18:38:25.275Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/d5/05861ede5d299f6599f86e6bc1291714e2116d96df003cfe23cc54bcc568/ml_dtypes-0.5.3-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9849ce7267444c0a717c80c6900997de4f36e2815ce34ac560a3edb2d9a64cd2", size = 4964606, upload-time = "2025-07-29T18:38:27.045Z" },
-    { url = "https://files.pythonhosted.org/packages/db/dc/72992b68de367741bfab8df3b3fe7c29f982b7279d341aa5bf3e7ef737ea/ml_dtypes-0.5.3-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c3f5ae0309d9f888fd825c2e9d0241102fadaca81d888f26f845bc8c13c1e4ee", size = 4938435, upload-time = "2025-07-29T18:38:29.193Z" },
-    { url = "https://files.pythonhosted.org/packages/81/1c/d27a930bca31fb07d975a2d7eaf3404f9388114463b9f15032813c98f893/ml_dtypes-0.5.3-cp311-cp311-win_amd64.whl", hash = "sha256:58e39349d820b5702bb6f94ea0cb2dc8ec62ee81c0267d9622067d8333596a46", size = 206334, upload-time = "2025-07-29T18:38:30.687Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/d8/6922499effa616012cb8dc445280f66d100a7ff39b35c864cfca019b3f89/ml_dtypes-0.5.3-cp311-cp311-win_arm64.whl", hash = "sha256:66c2756ae6cfd7f5224e355c893cfd617fa2f747b8bbd8996152cbdebad9a184", size = 157584, upload-time = "2025-07-29T18:38:32.187Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/eb/bc07c88a6ab002b4635e44585d80fa0b350603f11a2097c9d1bfacc03357/ml_dtypes-0.5.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:156418abeeda48ea4797db6776db3c5bdab9ac7be197c1233771e0880c304057", size = 663864, upload-time = "2025-07-29T18:38:33.777Z" },
-    { url = "https://files.pythonhosted.org/packages/cf/89/11af9b0f21b99e6386b6581ab40fb38d03225f9de5f55cf52097047e2826/ml_dtypes-0.5.3-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1db60c154989af253f6c4a34e8a540c2c9dce4d770784d426945e09908fbb177", size = 4951313, upload-time = "2025-07-29T18:38:36.45Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/a9/b98b86426c24900b0c754aad006dce2863df7ce0bb2bcc2c02f9cc7e8489/ml_dtypes-0.5.3-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1b255acada256d1fa8c35ed07b5f6d18bc21d1556f842fbc2d5718aea2cd9e55", size = 4928805, upload-time = "2025-07-29T18:38:38.29Z" },
-    { url = "https://files.pythonhosted.org/packages/50/c1/85e6be4fc09c6175f36fb05a45917837f30af9a5146a5151cb3a3f0f9e09/ml_dtypes-0.5.3-cp312-cp312-win_amd64.whl", hash = "sha256:da65e5fd3eea434ccb8984c3624bc234ddcc0d9f4c81864af611aaebcc08a50e", size = 208182, upload-time = "2025-07-29T18:38:39.72Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/17/cf5326d6867be057f232d0610de1458f70a8ce7b6290e4b4a277ea62b4cd/ml_dtypes-0.5.3-cp312-cp312-win_arm64.whl", hash = "sha256:8bb9cd1ce63096567f5f42851f5843b5a0ea11511e50039a7649619abfb4ba6d", size = 161560, upload-time = "2025-07-29T18:38:41.072Z" },
-    { url = "https://files.pythonhosted.org/packages/19/2d/c61af51173083bbf2a3b0f1a1a01d50ef1830436880027433d1b75271083/ml_dtypes-0.5.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:5ee72568d46b9533ad54f78b1e1f3067c0534c5065120ea8ecc6f210d22748b3", size = 663552, upload-time = "2025-07-29T18:39:13.102Z" },
-    { url = "https://files.pythonhosted.org/packages/61/0e/a628f2aefd719745e8a13492375a55cedea77c0cfc917b1ce11bde435c68/ml_dtypes-0.5.3-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:01de48de4537dc3c46e684b969a40ec36594e7eeb7c69e9a093e7239f030a28a", size = 4952704, upload-time = "2025-07-29T18:39:14.829Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/2e/5ba92f1f99d1f5f62bffec614a5b8161e55c3961257c902fa26dbe909baa/ml_dtypes-0.5.3-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8b1a6e231b0770f2894910f1dce6d2f31d65884dbf7668f9b08d73623cdca909", size = 4923538, upload-time = "2025-07-29T18:39:16.581Z" },
-    { url = "https://files.pythonhosted.org/packages/70/3b/f801c69027866ea6e387224551185fedef62ad8e2e71181ec0d9dda905f7/ml_dtypes-0.5.3-cp39-cp39-win_amd64.whl", hash = "sha256:a4f39b9bf6555fab9bfb536cf5fdd1c1c727e8d22312078702e9ff005354b37f", size = 206567, upload-time = "2025-07-29T18:39:18.047Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/0e/4a/c27b42ed9b1c7d13d9ba8b6905dece787d6259152f2309338aed29b2447b/ml_dtypes-0.5.4.tar.gz", hash = "sha256:8ab06a50fb9bf9666dd0fe5dfb4676fa2b0ac0f31ecff72a6c3af8e22c063453", size = 692314, upload-time = "2025-11-17T22:32:31.031Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fe/3a/c5b855752a70267ff729c349e650263adb3c206c29d28cc8ea7ace30a1d5/ml_dtypes-0.5.4-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:b95e97e470fe60ed493fd9ae3911d8da4ebac16bd21f87ffa2b7c588bf22ea2c", size = 679735, upload-time = "2025-11-17T22:31:31.367Z" },
+    { url = "https://files.pythonhosted.org/packages/41/79/7433f30ee04bd4faa303844048f55e1eb939131c8e5195a00a96a0939b64/ml_dtypes-0.5.4-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b4b801ebe0b477be666696bda493a9be8356f1f0057a57f1e35cd26928823e5a", size = 5051883, upload-time = "2025-11-17T22:31:33.658Z" },
+    { url = "https://files.pythonhosted.org/packages/10/b1/8938e8830b0ee2e167fc75a094dea766a1152bde46752cd9bfc57ee78a82/ml_dtypes-0.5.4-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:388d399a2152dd79a3f0456a952284a99ee5c93d3e2f8dfe25977511e0515270", size = 5030369, upload-time = "2025-11-17T22:31:35.595Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/a3/51886727bd16e2f47587997b802dd56398692ce8c6c03c2e5bb32ecafe26/ml_dtypes-0.5.4-cp310-cp310-win_amd64.whl", hash = "sha256:4ff7f3e7ca2972e7de850e7b8fcbb355304271e2933dd90814c1cb847414d6e2", size = 210738, upload-time = "2025-11-17T22:31:37.43Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/5e/712092cfe7e5eb667b8ad9ca7c54442f21ed7ca8979745f1000e24cf8737/ml_dtypes-0.5.4-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6c7ecb74c4bd71db68a6bea1edf8da8c34f3d9fe218f038814fd1d310ac76c90", size = 679734, upload-time = "2025-11-17T22:31:39.223Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/cf/912146dfd4b5c0eea956836c01dcd2fce6c9c844b2691f5152aca196ce4f/ml_dtypes-0.5.4-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bc11d7e8c44a65115d05e2ab9989d1e045125d7be8e05a071a48bc76eb6d6040", size = 5056165, upload-time = "2025-11-17T22:31:41.071Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/80/19189ea605017473660e43762dc853d2797984b3c7bf30ce656099add30c/ml_dtypes-0.5.4-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:19b9a53598f21e453ea2fbda8aa783c20faff8e1eeb0d7ab899309a0053f1483", size = 5034975, upload-time = "2025-11-17T22:31:42.758Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/24/70bd59276883fdd91600ca20040b41efd4902a923283c4d6edcb1de128d2/ml_dtypes-0.5.4-cp311-cp311-win_amd64.whl", hash = "sha256:7c23c54a00ae43edf48d44066a7ec31e05fdc2eee0be2b8b50dd1903a1db94bb", size = 210742, upload-time = "2025-11-17T22:31:44.068Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/c9/64230ef14e40aa3f1cb254ef623bf812735e6bec7772848d19131111ac0d/ml_dtypes-0.5.4-cp311-cp311-win_arm64.whl", hash = "sha256:557a31a390b7e9439056644cb80ed0735a6e3e3bb09d67fd5687e4b04238d1de", size = 160709, upload-time = "2025-11-17T22:31:46.557Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/b8/3c70881695e056f8a32f8b941126cf78775d9a4d7feba8abcb52cb7b04f2/ml_dtypes-0.5.4-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:a174837a64f5b16cab6f368171a1a03a27936b31699d167684073ff1c4237dac", size = 676927, upload-time = "2025-11-17T22:31:48.182Z" },
+    { url = "https://files.pythonhosted.org/packages/54/0f/428ef6881782e5ebb7eca459689448c0394fa0a80bea3aa9262cba5445ea/ml_dtypes-0.5.4-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a7f7c643e8b1320fd958bf098aa7ecf70623a42ec5154e3be3be673f4c34d900", size = 5028464, upload-time = "2025-11-17T22:31:50.135Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/cb/28ce52eb94390dda42599c98ea0204d74799e4d8047a0eb559b6fd648056/ml_dtypes-0.5.4-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9ad459e99793fa6e13bd5b7e6792c8f9190b4e5a1b45c63aba14a4d0a7f1d5ff", size = 5009002, upload-time = "2025-11-17T22:31:52.001Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/f0/0cfadd537c5470378b1b32bd859cf2824972174b51b873c9d95cfd7475a5/ml_dtypes-0.5.4-cp312-cp312-win_amd64.whl", hash = "sha256:c1a953995cccb9e25a4ae19e34316671e4e2edaebe4cf538229b1fc7109087b7", size = 212222, upload-time = "2025-11-17T22:31:53.742Z" },
+    { url = "https://files.pythonhosted.org/packages/16/2e/9acc86985bfad8f2c2d30291b27cd2bb4c74cea08695bd540906ed744249/ml_dtypes-0.5.4-cp312-cp312-win_arm64.whl", hash = "sha256:9bad06436568442575beb2d03389aa7456c690a5b05892c471215bfd8cf39460", size = 160793, upload-time = "2025-11-17T22:31:55.358Z" },
+    { url = "https://files.pythonhosted.org/packages/af/a1/4f20f56ba9c21c7ee78505dc9f782017ffc9ae9ff261179e28da710e3900/ml_dtypes-0.5.4-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:d81fdb088defa30eb37bf390bb7dde35d3a83ec112ac8e33d75ab28cc29dd8b0", size = 676875, upload-time = "2025-11-17T22:32:24.954Z" },
+    { url = "https://files.pythonhosted.org/packages/71/85/846992d38a1f3ca561ac5d05f7bd8654695f2a3c202fcdc4f9e53951f211/ml_dtypes-0.5.4-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:88c982aac7cb1cbe8cbb4e7f253072b1df872701fcaf48d84ffbb433b6568f24", size = 5046025, upload-time = "2025-11-17T22:32:26.767Z" },
+    { url = "https://files.pythonhosted.org/packages/22/08/f9aaafa02f46b1d81bf3b7a158b1b9df24df6e4b8ec0082a26eaf16ce229/ml_dtypes-0.5.4-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a9b61c19040397970d18d7737375cffd83b1f36a11dd4ad19f83a016f736c3ef", size = 5018614, upload-time = "2025-11-17T22:32:28.231Z" },
+    { url = "https://files.pythonhosted.org/packages/63/8a/bc7f9c8c358214dba25f70077dbc85aac85f92d255a6f20dd3ae64026a43/ml_dtypes-0.5.4-cp39-cp39-win_amd64.whl", hash = "sha256:3d277bf3637f2a62176f4575512e9ff9ef51d00e39626d9fe4a161992f355af2", size = 210704, upload-time = "2025-11-17T22:32:29.696Z" },
+]
+
+[[package]]
+name = "mlx"
+version = "0.29.3"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "mlx-metal", version = "0.29.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10' and sys_platform == 'darwin'" },
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a4/8a/743ff24a07f8cfd6fb14b3fe05f122f1d8e04e8a912b2f6d0e14369c8caf/mlx-0.29.3-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:340d46443fe0b1e5d84c1e36aa633310de70365ce79aefcaa6f618e62bd4b045", size = 548930, upload-time = "2025-10-17T19:16:49.872Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/2a/af1b8391b6f543e59ca595f63aaddc33e320d3cc57a4c86ded6932d9dc3c/mlx-0.29.3-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:8449c0e7c221e38368a734a471e0c4b1a7fea072947c75e893a1ee214f208d34", size = 548928, upload-time = "2025-10-17T19:16:58.275Z" },
+    { url = "https://files.pythonhosted.org/packages/49/85/0c58bdc5733ba92f78f067fc25e131e34db46562719d7909cebfad9313c5/mlx-0.29.3-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:1bba5203ed3f785167f5b8891c2e91ede23401586b0a723bfaf815a3ed450e3d", size = 548931, upload-time = "2025-10-17T19:16:52.198Z" },
+    { url = "https://files.pythonhosted.org/packages/1b/53/a9648dff9544a201e9ed483f1e0b18cdd92d9be453d58f1fedfd99cde6e1/mlx-0.29.3-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:ffb3a167da52baeb05756895298a2d582b5e9b9b9364b4a454b78b824a9fa482", size = 652400, upload-time = "2025-10-17T19:20:48.146Z" },
+    { url = "https://files.pythonhosted.org/packages/94/13/3e91a37fa55dc0e9114620729ab61b27f45ed59053fc77846cad2df54f21/mlx-0.29.3-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:0ffdf1f171c903adeaa688210ba39063059b102f3dcc52a64c2200d95d237f15", size = 549089, upload-time = "2025-10-17T19:17:00.446Z" },
+    { url = "https://files.pythonhosted.org/packages/13/01/ce008d14fbd2e22b694f568ab4014e14c979a2262c5f8c10e06d4806709f/mlx-0.29.3-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:e7d1d815be0d4a41e598bdb2992822dafd9ab0d59d4b88af760ee0b6584506b7", size = 549091, upload-time = "2025-10-17T19:16:54.428Z" },
+    { url = "https://files.pythonhosted.org/packages/72/1c/45642746d36e91e26f3401e9b7931f92d8cc1eb6015cc40218628f320747/mlx-0.29.3-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:d33bff69887fadfd85ce67b8e11318c2319984f3ad4157f871aa9d3beb9de972", size = 549092, upload-time = "2025-10-17T19:17:10.963Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/17/9c85fc6ebe6b8ad30c3e75c0cb869939df82146aa8728de1261adacc731d/mlx-0.29.3-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:8ead74126ffcc1ae49f3b1e0e988620ffbb059c38184f4e9390e294808e2c614", size = 652523, upload-time = "2025-10-17T19:24:01.159Z" },
+    { url = "https://files.pythonhosted.org/packages/07/f5/14e12e219a2715296150d35f930dc3a6ff319cd60126408e563f03100113/mlx-0.29.3-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:86c62791ce930028d75c41b88b4e3ceb58f5f2e263ff9bfacda998b0c03d9544", size = 549516, upload-time = "2025-10-17T19:18:13.831Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/e2/5177c80e8c33a8be89fa45fa0a839d5b6a5578687d0ec973bf03638a4e73/mlx-0.29.3-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:cddf6bcdc561094af6b3f0706f8768ecc5216a97eb6973e838c3ac2e2fca2cc8", size = 549509, upload-time = "2025-10-17T19:17:21.517Z" },
+    { url = "https://files.pythonhosted.org/packages/11/89/aa424217a7a0291b84f8969d504ac63f5af0ef60f248fe5562c3d6e44048/mlx-0.29.3-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:b2e1a249437d017a7425358420d28e641b7bc9c2650f3e013c1b1f4f239d8533", size = 549511, upload-time = "2025-10-17T19:16:54.227Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/c7/af484ab5a4864384dc8a5f6f8ade9a29bd6e7a652e535f2ca39cf473ce26/mlx-0.29.3-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:f09f71ee958f04824b7c7b275a1c1deb052740f5e69eccbff6672e43d9d7f890", size = 649834, upload-time = "2025-10-17T19:23:59.474Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/eb/3ad14e1acb45bcccb3031247cb9fb3ad18d73fa647fed59fe238de99a68e/mlx-0.29.3-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:c6778d023cc540a83a08caa750c9752523c24ddebd4201b2f34c9e0a2e109394", size = 549302, upload-time = "2025-10-17T19:19:28.002Z" },
+    { url = "https://files.pythonhosted.org/packages/59/30/93cab1b5e59b879a5a4346684d7564a7e5d74e6f9448163c6b118f167cad/mlx-0.29.3-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:30a1da8b2aeb4f4b405759b108abbbe3b4f8c5924ac3d550006509bc7bcbfc49", size = 549300, upload-time = "2025-10-17T19:18:44.123Z" },
+    { url = "https://files.pythonhosted.org/packages/79/95/af0ebe08bc917b3ef2bd91e7c149c31000ce72deffbf55cd8d96e43ea208/mlx-0.29.3-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:ac2409a1ba79149143232a02ae5b1a201864413979a76d8f9d63107022e0dce5", size = 549301, upload-time = "2025-10-17T19:18:39.994Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/3b/050b8cbe6e326a8ad982ef7690913cf8da6303e439029ce45a6de0476743/mlx-0.29.3-cp39-cp39-manylinux_2_35_x86_64.whl", hash = "sha256:efefa29663387f9340602d4b5906881398eda12f48882a60994fd2b379848d96", size = 653164, upload-time = "2025-10-17T19:24:01.274Z" },
 ]
 
 [[package]]
 name = "mlx"
-version = "0.29.2"
+version = "0.30.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
 dependencies = [
-    { name = "mlx-metal", marker = "sys_platform == 'darwin'" },
+    { name = "mlx-metal", version = "0.30.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10' and sys_platform == 'darwin'" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/3f/f0/2c2f99a91ed9dfcc78d31d9e5d3bb2f5305a8d65953cbc41f34f8056c49a/mlx-0.29.2-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:b46c1a24b9b8f7145e4d84410552ddfa03f40f9afdbe8f819f6b4b52b4db5d30", size = 547369, upload-time = "2025-09-26T22:21:33.668Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/06/0edddf0a5facb58c17616cd33da6de2722636da8d8d3927272a5a88658e4/mlx-0.29.2-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:18c5b63c6e4b35f75f477f74d3942870e0bc9c9f1c6a071d8a058bbd46681bf4", size = 547367, upload-time = "2025-09-26T22:21:36.63Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/cd/8c089cf1678a752ed4175a7ad5f08823ceef75d797217e12a44a71d6c062/mlx-0.29.2-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:ed53b8383ad4ee4311400558e0a5ec61105fc4553950b5732c66e7081cd1a9e8", size = 547365, upload-time = "2025-09-26T22:21:32.585Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/cc/9a14ac57a251f0ee5047f42214fa50d1d5223d805f0b8b6627639c3692eb/mlx-0.29.2-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:fff27af8dca74546422e8400d32ab848baf42405123cbcfdf62674a9c0560ebe", size = 652302, upload-time = "2025-09-26T22:26:24.575Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/f0/f57349f37cf5dd53f95127e141fc59fc435e4b6bfabba5a84c65de4d3597/mlx-0.29.2-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:e74965369227230374b3e8e8c8d46e209e5221a9b76bbb0fa788617e2c68f73c", size = 547581, upload-time = "2025-09-26T22:21:39.24Z" },
-    { url = "https://files.pythonhosted.org/packages/66/04/e016ca28dc9e0738a2541581420125cfe6bba24466a64420600bdd6fd52c/mlx-0.29.2-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:0f79194eeac78e85b96439d3bbc17aae5aba045a2af083c000b4fbbc501f253e", size = 547581, upload-time = "2025-09-26T22:21:47.706Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/b3/e2595e70ef8d4438dff694857745b0e108911e5b5fb83259dde6e5dc5bd1/mlx-0.29.2-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:33bbbb0fd24895d5ff080bb4d10e3e77017bba675d9a12466c8866eaf9b47854", size = 547578, upload-time = "2025-09-26T22:21:22.041Z" },
-    { url = "https://files.pythonhosted.org/packages/11/8c/5d51543ab128c2dff5e4b44ca799db8db5aa4f4ffc34af6531fd73627b54/mlx-0.29.2-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:32e159f2772be893bec580d2d50c0e6b32ad71a19ded7307bf6c871c8aaa9cf2", size = 651900, upload-time = "2025-09-26T22:26:11.133Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/84/7250237039e91d8e44ca0cf3522f189164844c196f262509afd29ef54710/mlx-0.29.2-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:eec950bf7118ad0865d0fc4686bd85d99bf8463fc717d836a5132e1a08b4f129", size = 548336, upload-time = "2025-09-26T22:21:44.914Z" },
-    { url = "https://files.pythonhosted.org/packages/13/47/428ac8d9b0cb5c136e5ce6c726cfdd55caa5b9497dafb6221acfee18f145/mlx-0.29.2-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:bef7333268d6d02e50a9ac6b10f661b711cd02da4a5e2d7619cf198a7e530308", size = 548334, upload-time = "2025-09-26T22:21:21.41Z" },
-    { url = "https://files.pythonhosted.org/packages/14/f0/7d5d3527ca3fdc664c900b4b822028691739e58c8e8f7975b33df4d3536e/mlx-0.29.2-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:f622fc6a84542a08ad2136e9251822d2c08106e5a1a0bd5d249a2d72bccd6577", size = 548330, upload-time = "2025-09-26T22:21:41.182Z" },
-    { url = "https://files.pythonhosted.org/packages/09/18/e202e0f6232822f6768995cdbf50eda202137bb6547368f6e3993dbee00b/mlx-0.29.2-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:a1aa1aee8e1b6bd1e51361e6b692c70d281b8187b2e859e70ecc11daab306dac", size = 648728, upload-time = "2025-09-26T22:25:49.159Z" },
-    { url = "https://files.pythonhosted.org/packages/da/cb/6dc30495c985a69b3ca51b267a822168b678dd37a580da011a38e1c4ea86/mlx-0.29.2-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:b9b1ece61e25c3e728ce760da1bbb11aa377ea0df84b92fd69cf60af02286969", size = 547851, upload-time = "2025-09-26T22:23:49.215Z" },
-    { url = "https://files.pythonhosted.org/packages/50/61/e7aa84522bce713f0bfc5a85e44f16fbdec00bbc5f96d06a82264859dba2/mlx-0.29.2-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:00cbf0dacf3d7a345581b49ad350b50f419788983dfbf6ee03624e485237c476", size = 547849, upload-time = "2025-09-26T22:23:33.569Z" },
-    { url = "https://files.pythonhosted.org/packages/df/c2/214211fb2e5b0813594287b304852d7e038c571c06a4cbd492068d8b49a8/mlx-0.29.2-cp39-cp39-macosx_15_0_arm64.whl", hash = "sha256:bd8c031889fdd6e40bfb831490ab6d4776491f7e91909851941489b5329e80d4", size = 547849, upload-time = "2025-09-26T22:23:31.468Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/a0/3fe20a277b8f96131223ea53f739cd8845b6dc27caa6bf6576d4feb880f9/mlx-0.29.2-cp39-cp39-manylinux_2_35_x86_64.whl", hash = "sha256:eb319abca8cfc5a8a9d9c3b020abcc942d12683186d71ec6b6b795eb75206096", size = 652549, upload-time = "2025-09-26T22:28:39.486Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/21/2f95936bdffeed3a5769b1677be0d17fa4c6131d76cad40b4d60ab5c2f22/mlx-0.30.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:f861392df1ee0ee216188060a8ff2cd2ed4d2b34d4366102863d01d02acf2a7e", size = 554578, upload-time = "2025-11-20T06:42:23.059Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/11/85f679159f58770a02e4e61bd418c4a441b2a5df7efee1003e6301be5211/mlx-0.30.0-cp310-cp310-macosx_15_0_arm64.whl", hash = "sha256:7f4a0c37cc3eee051ee5efade3200cb49bc7e9973b38e53db2dddacdb3896a71", size = 554576, upload-time = "2025-11-20T01:16:34.564Z" },
+    { url = "https://files.pythonhosted.org/packages/49/18/e11f5b6f617b69b5b4ad8b83489e920adbabde90a8453f561edd774416af/mlx-0.30.0-cp310-cp310-macosx_26_0_arm64.whl", hash = "sha256:df9a48d181ba52da86b524b5e21f8c4eebd73393649127cde19ace75b26daa2c", size = 554517, upload-time = "2025-11-20T05:33:22.645Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/a7/e18ebe7c9572106e1e1fe7bf9cd91d446c2a9b828c4a062ccfbaa4e5f3dd/mlx-0.30.0-cp310-cp310-manylinux_2_35_x86_64.whl", hash = "sha256:3e2488ec30a1b17138050f0a33d9202c3cbdf67307b411d4ba3e116bd14484d0", size = 659670, upload-time = "2025-11-20T01:16:36.291Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/e8/69ebac29536c026489ded1ad58a6f5163b8fc10ab5eac21228f57ea9e83f/mlx-0.30.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:d5871a5f1ce5cba2f690d2c630a6672cc65a62326bcaa6db258185957a1f073f", size = 554825, upload-time = "2025-11-20T16:45:11.135Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/c7/db80b1e9f613baf99745e9920a3a7fe7b6c61398420ed308f24f60877a15/mlx-0.30.0-cp311-cp311-macosx_15_0_arm64.whl", hash = "sha256:302c3d52b4f68b80a8ee259489369524a3381fcc9aed12b17c1c537870c81fce", size = 554821, upload-time = "2025-11-20T01:16:38.53Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/68/55cd5cab5a9b8b958ea995295aa3428ad65b3e893f04c27e2cdfcdc176af/mlx-0.30.0-cp311-cp311-macosx_26_0_arm64.whl", hash = "sha256:d72443a461f2e92329edf344ded8460df53fd7ec3c04ddd2353ee4ceb34c0ff2", size = 554791, upload-time = "2025-11-20T05:33:23.851Z" },
+    { url = "https://files.pythonhosted.org/packages/80/17/16868ac1ea36ea3baf61f84721b4dfd98dd247a230b951ddd715981db79b/mlx-0.30.0-cp311-cp311-manylinux_2_35_x86_64.whl", hash = "sha256:0beb015a6afef2f86dfbefe65c9022dce91dbe1651e1b4330ef32434602323e0", size = 659735, upload-time = "2025-11-20T01:16:40.594Z" },
+    { url = "https://files.pythonhosted.org/packages/94/a3/32c4c05d8967591e2a1a1e7e3fc9cece8821f5aea8ac8f3bcfdb203f4722/mlx-0.30.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:11fbae58b1e992afdec6709d5e281932871a0138582a794cdcc82ff895a28670", size = 554567, upload-time = "2025-11-20T16:45:12.73Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/b3/b6143f1c078fbc873e40e624dc428a3ada240721001414955f584afa866d/mlx-0.30.0-cp312-cp312-macosx_15_0_arm64.whl", hash = "sha256:fc1994a4c9e60ccfe2ab9211c846be2ef7abc6fc3e53e1addb2cdf7468f61e7b", size = 554566, upload-time = "2025-11-20T01:16:41.877Z" },
+    { url = "https://files.pythonhosted.org/packages/96/cb/aec36297ef76b4b190cc6d4cea1ed995b458bbc21cb91c58b0862f8cae7d/mlx-0.30.0-cp312-cp312-macosx_26_0_arm64.whl", hash = "sha256:a9f30c5c94b30b65f026162e91256473187e1de57c13dcc2aedb2fc33c07f2c1", size = 554593, upload-time = "2025-11-20T05:33:25.312Z" },
+    { url = "https://files.pythonhosted.org/packages/23/37/b5dd68da0e79e258f5d6a0e9f5fd4f9e5452e92c20b13b64948a965ea429/mlx-0.30.0-cp312-cp312-manylinux_2_35_x86_64.whl", hash = "sha256:1c4a2f5285bdd585aa6485a4fb5759ebc2721ba9381404ff867c136e84764e9b", size = 653994, upload-time = "2025-11-20T01:16:43.492Z" },
 ]
 
 [[package]]
 name = "mlx-lm"
-version = "0.28.2"
+version = "0.28.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jinja2" },
-    { name = "mlx" },
+    { name = "mlx", version = "0.29.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10' and sys_platform == 'darwin'" },
+    { name = "mlx", version = "0.30.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10' and sys_platform == 'darwin'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "protobuf" },
     { name = "pyyaml" },
     { name = "transformers" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/1c/d7/fdde445c7bd443a2ed23badda6064f1477c4051543922106f365e94082cd/mlx_lm-0.28.2.tar.gz", hash = "sha256:d28752635ed5c89ff2b41361916c928e6b16f765c07b2908044e1dcaf921ed9b", size = 209374, upload-time = "2025-10-02T14:23:57.497Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/51/f6/15e002d52c28d8c544ec3aaf9053677468333e6ef0e76ea68579fd77b76d/mlx_lm-0.28.3.tar.gz", hash = "sha256:75df2b925d343ebaf50b63008dede4fe98cd3b02b1b24b7da71ebeb198d674f0", size = 214455, upload-time = "2025-10-17T21:44:33.921Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c2/a6/db3b44a5ac1a1174605628b0a477fbe4632d4fad1f94cf08647e27cc79ad/mlx_lm-0.28.3-py3-none-any.whl", hash = "sha256:ec103e2c9a06bd2cbafd41aafc975e40262176f7360d4f53ec342cebb9e0e6ea", size = 294506, upload-time = "2025-10-17T21:44:32.447Z" },
+]
+
+[[package]]
+name = "mlx-metal"
+version = "0.29.3"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f2/1c/89e0f60d45e364de8507065f73aeb8d2fd810d6cb95a9a512880b09399d5/mlx_lm-0.28.2-py3-none-any.whl", hash = "sha256:1501529e625d0d648216f7bb543b8b449d5fd17bd598f635536dbc1fbde6d1d6", size = 284600, upload-time = "2025-10-02T14:23:56.395Z" },
+    { url = "https://files.pythonhosted.org/packages/41/95/a00054a006df82bb1b5b8f666ae44a676b259146fadbff90fe654309fefc/mlx_metal-0.29.3-py3-none-macosx_13_0_arm64.whl", hash = "sha256:27b5a4d905202a71e84d9fd559ea0236813f6f960ef494e5cafe9c45df4c9d7c", size = 36817352, upload-time = "2025-10-17T19:19:25.801Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/d8/5ee91eac16dfcf0334103120b47d4abd8c890ccc0d73d3eee4770ce8810f/mlx_metal-0.29.3-py3-none-macosx_14_0_arm64.whl", hash = "sha256:f426d4b67f96b4d6f0ed50d5992933595aadb370dc3e9ed2410bafbc16229882", size = 36555573, upload-time = "2025-10-17T19:18:42.098Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/9a/39b7ecdf21cf2a39ced8d7933eed65c6cb38295cadfd0907dd1abd4d1ded/mlx_metal-0.29.3-py3-none-macosx_15_0_arm64.whl", hash = "sha256:106616f7f825851043c53d3dc186965c003985da9cbb6e5c034f35108fc1fc27", size = 36549163, upload-time = "2025-10-17T19:18:37.701Z" },
 ]
 
 [[package]]
 name = "mlx-metal"
-version = "0.29.2"
+version = "0.30.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/31/a5/a045006546fed791f6e9a74ed4451dac871d3c35f9e54a3a25d820668a85/mlx_metal-0.29.2-py3-none-macosx_13_0_arm64.whl", hash = "sha256:cf8f83a521e620357185c57945142718d526b9312ee112e5a89eb5600480f4d6", size = 35056194, upload-time = "2025-09-26T22:23:47.201Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/8c/4bdd3a7d04ed477b32aec30d30236dfca9f9ac27706cb309511278ddd281/mlx_metal-0.29.2-py3-none-macosx_14_0_arm64.whl", hash = "sha256:fa944001970813b296e8aff5616f2fa9daeda6bc1d190c17fbe8a7ca838ecef0", size = 34791708, upload-time = "2025-09-26T22:23:30.599Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/11/12e158848fe4d3316c999ffb6c2d88f554bde98d69022b3385e25ece997e/mlx_metal-0.29.2-py3-none-macosx_15_0_arm64.whl", hash = "sha256:08d8b7fe305425a14b74ebf36cee176575bfd4cd8d34a2aaae8f05b9983d2d71", size = 34784506, upload-time = "2025-09-26T22:23:29.207Z" },
+    { url = "https://files.pythonhosted.org/packages/64/9f/47ebb6e9b2c33371c6ca3733e70324ed064f49e790ee4e194b713d6d7d84/mlx_metal-0.30.0-py3-none-macosx_14_0_arm64.whl", hash = "sha256:f48543b10d13bf0591b3f99c6eb585dd2c2e5db379edae5df0f19a728cb41742", size = 36819863, upload-time = "2025-11-20T06:42:23.927Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/91/c04e420326390c37d4c13f58960956d698cd34c7432ae3860bc5c6be71a0/mlx_metal-0.30.0-py3-none-macosx_15_0_arm64.whl", hash = "sha256:74bb5d10e0f24e21973d39430557bbd5d733c2a6599c3f1b87f9a0ff73fed2c8", size = 36817633, upload-time = "2025-11-20T01:16:27.77Z" },
+    { url = "https://files.pythonhosted.org/packages/86/18/7af11eac0f488b68c436d249ffcf4f76003326e691994db3f720d21f21bb/mlx_metal-0.30.0-py3-none-macosx_26_0_arm64.whl", hash = "sha256:310fda8b2f7345865f3dd75a9b478e974f28d22e8ebf05f26af5adc0e8979cee", size = 44878409, upload-time = "2025-11-20T05:33:20.221Z" },
 ]
 
 [[package]]
@@ -2278,74 +2273,44 @@ wheels = [
 
 [[package]]
 name = "msgpack"
-version = "1.1.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/45/b1/ea4f68038a18c77c9467400d166d74c4ffa536f34761f7983a104357e614/msgpack-1.1.1.tar.gz", hash = "sha256:77b79ce34a2bdab2594f490c8e80dd62a02d650b91a75159a63ec413b8d104cd", size = 173555, upload-time = "2025-06-13T06:52:51.324Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/33/52/f30da112c1dc92cf64f57d08a273ac771e7b29dea10b4b30369b2d7e8546/msgpack-1.1.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:353b6fc0c36fde68b661a12949d7d49f8f51ff5fa019c1e47c87c4ff34b080ed", size = 81799, upload-time = "2025-06-13T06:51:37.228Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/35/7bfc0def2f04ab4145f7f108e3563f9b4abae4ab0ed78a61f350518cc4d2/msgpack-1.1.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:79c408fcf76a958491b4e3b103d1c417044544b68e96d06432a189b43d1215c8", size = 78278, upload-time = "2025-06-13T06:51:38.534Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/c5/df5d6c1c39856bc55f800bf82778fd4c11370667f9b9e9d51b2f5da88f20/msgpack-1.1.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:78426096939c2c7482bf31ef15ca219a9e24460289c00dd0b94411040bb73ad2", size = 402805, upload-time = "2025-06-13T06:51:39.538Z" },
-    { url = "https://files.pythonhosted.org/packages/20/8e/0bb8c977efecfe6ea7116e2ed73a78a8d32a947f94d272586cf02a9757db/msgpack-1.1.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8b17ba27727a36cb73aabacaa44b13090feb88a01d012c0f4be70c00f75048b4", size = 408642, upload-time = "2025-06-13T06:51:41.092Z" },
-    { url = "https://files.pythonhosted.org/packages/59/a1/731d52c1aeec52006be6d1f8027c49fdc2cfc3ab7cbe7c28335b2910d7b6/msgpack-1.1.1-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:7a17ac1ea6ec3c7687d70201cfda3b1e8061466f28f686c24f627cae4ea8efd0", size = 395143, upload-time = "2025-06-13T06:51:42.575Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/92/b42911c52cda2ba67a6418ffa7d08969edf2e760b09015593c8a8a27a97d/msgpack-1.1.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:88d1e966c9235c1d4e2afac21ca83933ba59537e2e2727a999bf3f515ca2af26", size = 395986, upload-time = "2025-06-13T06:51:43.807Z" },
-    { url = "https://files.pythonhosted.org/packages/61/dc/8ae165337e70118d4dab651b8b562dd5066dd1e6dd57b038f32ebc3e2f07/msgpack-1.1.1-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:f6d58656842e1b2ddbe07f43f56b10a60f2ba5826164910968f5933e5178af75", size = 402682, upload-time = "2025-06-13T06:51:45.534Z" },
-    { url = "https://files.pythonhosted.org/packages/58/27/555851cb98dcbd6ce041df1eacb25ac30646575e9cd125681aa2f4b1b6f1/msgpack-1.1.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:96decdfc4adcbc087f5ea7ebdcfd3dee9a13358cae6e81d54be962efc38f6338", size = 406368, upload-time = "2025-06-13T06:51:46.97Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/64/39a26add4ce16f24e99eabb9005e44c663db00e3fce17d4ae1ae9d61df99/msgpack-1.1.1-cp310-cp310-win32.whl", hash = "sha256:6640fd979ca9a212e4bcdf6eb74051ade2c690b862b679bfcb60ae46e6dc4bfd", size = 65004, upload-time = "2025-06-13T06:51:48.582Z" },
-    { url = "https://files.pythonhosted.org/packages/7d/18/73dfa3e9d5d7450d39debde5b0d848139f7de23bd637a4506e36c9800fd6/msgpack-1.1.1-cp310-cp310-win_amd64.whl", hash = "sha256:8b65b53204fe1bd037c40c4148d00ef918eb2108d24c9aaa20bc31f9810ce0a8", size = 71548, upload-time = "2025-06-13T06:51:49.558Z" },
-    { url = "https://files.pythonhosted.org/packages/7f/83/97f24bf9848af23fe2ba04380388216defc49a8af6da0c28cc636d722502/msgpack-1.1.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:71ef05c1726884e44f8b1d1773604ab5d4d17729d8491403a705e649116c9558", size = 82728, upload-time = "2025-06-13T06:51:50.68Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/7f/2eaa388267a78401f6e182662b08a588ef4f3de6f0eab1ec09736a7aaa2b/msgpack-1.1.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:36043272c6aede309d29d56851f8841ba907a1a3d04435e43e8a19928e243c1d", size = 79279, upload-time = "2025-06-13T06:51:51.72Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/46/31eb60f4452c96161e4dfd26dbca562b4ec68c72e4ad07d9566d7ea35e8a/msgpack-1.1.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a32747b1b39c3ac27d0670122b57e6e57f28eefb725e0b625618d1b59bf9d1e0", size = 423859, upload-time = "2025-06-13T06:51:52.749Z" },
-    { url = "https://files.pythonhosted.org/packages/45/16/a20fa8c32825cc7ae8457fab45670c7a8996d7746ce80ce41cc51e3b2bd7/msgpack-1.1.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a8b10fdb84a43e50d38057b06901ec9da52baac6983d3f709d8507f3889d43f", size = 429975, upload-time = "2025-06-13T06:51:53.97Z" },
-    { url = "https://files.pythonhosted.org/packages/86/ea/6c958e07692367feeb1a1594d35e22b62f7f476f3c568b002a5ea09d443d/msgpack-1.1.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:ba0c325c3f485dc54ec298d8b024e134acf07c10d494ffa24373bea729acf704", size = 413528, upload-time = "2025-06-13T06:51:55.507Z" },
-    { url = "https://files.pythonhosted.org/packages/75/05/ac84063c5dae79722bda9f68b878dc31fc3059adb8633c79f1e82c2cd946/msgpack-1.1.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:88daaf7d146e48ec71212ce21109b66e06a98e5e44dca47d853cbfe171d6c8d2", size = 413338, upload-time = "2025-06-13T06:51:57.023Z" },
-    { url = "https://files.pythonhosted.org/packages/69/e8/fe86b082c781d3e1c09ca0f4dacd457ede60a13119b6ce939efe2ea77b76/msgpack-1.1.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:d8b55ea20dc59b181d3f47103f113e6f28a5e1c89fd5b67b9140edb442ab67f2", size = 422658, upload-time = "2025-06-13T06:51:58.419Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/2b/bafc9924df52d8f3bb7c00d24e57be477f4d0f967c0a31ef5e2225e035c7/msgpack-1.1.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4a28e8072ae9779f20427af07f53bbb8b4aa81151054e882aee333b158da8752", size = 427124, upload-time = "2025-06-13T06:51:59.969Z" },
-    { url = "https://files.pythonhosted.org/packages/a2/3b/1f717e17e53e0ed0b68fa59e9188f3f610c79d7151f0e52ff3cd8eb6b2dc/msgpack-1.1.1-cp311-cp311-win32.whl", hash = "sha256:7da8831f9a0fdb526621ba09a281fadc58ea12701bc709e7b8cbc362feabc295", size = 65016, upload-time = "2025-06-13T06:52:01.294Z" },
-    { url = "https://files.pythonhosted.org/packages/48/45/9d1780768d3b249accecc5a38c725eb1e203d44a191f7b7ff1941f7df60c/msgpack-1.1.1-cp311-cp311-win_amd64.whl", hash = "sha256:5fd1b58e1431008a57247d6e7cc4faa41c3607e8e7d4aaf81f7c29ea013cb458", size = 72267, upload-time = "2025-06-13T06:52:02.568Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/26/389b9c593eda2b8551b2e7126ad3a06af6f9b44274eb3a4f054d48ff7e47/msgpack-1.1.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:ae497b11f4c21558d95de9f64fff7053544f4d1a17731c866143ed6bb4591238", size = 82359, upload-time = "2025-06-13T06:52:03.909Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/65/7d1de38c8a22cf8b1551469159d4b6cf49be2126adc2482de50976084d78/msgpack-1.1.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:33be9ab121df9b6b461ff91baac6f2731f83d9b27ed948c5b9d1978ae28bf157", size = 79172, upload-time = "2025-06-13T06:52:05.246Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/bd/cacf208b64d9577a62c74b677e1ada005caa9b69a05a599889d6fc2ab20a/msgpack-1.1.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6f64ae8fe7ffba251fecb8408540c34ee9df1c26674c50c4544d72dbf792e5ce", size = 425013, upload-time = "2025-06-13T06:52:06.341Z" },
-    { url = "https://files.pythonhosted.org/packages/4d/ec/fd869e2567cc9c01278a736cfd1697941ba0d4b81a43e0aa2e8d71dab208/msgpack-1.1.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a494554874691720ba5891c9b0b39474ba43ffb1aaf32a5dac874effb1619e1a", size = 426905, upload-time = "2025-06-13T06:52:07.501Z" },
-    { url = "https://files.pythonhosted.org/packages/55/2a/35860f33229075bce803a5593d046d8b489d7ba2fc85701e714fc1aaf898/msgpack-1.1.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:cb643284ab0ed26f6957d969fe0dd8bb17beb567beb8998140b5e38a90974f6c", size = 407336, upload-time = "2025-06-13T06:52:09.047Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/16/69ed8f3ada150bf92745fb4921bd621fd2cdf5a42e25eb50bcc57a5328f0/msgpack-1.1.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d275a9e3c81b1093c060c3837e580c37f47c51eca031f7b5fb76f7b8470f5f9b", size = 409485, upload-time = "2025-06-13T06:52:10.382Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/b6/0c398039e4c6d0b2e37c61d7e0e9d13439f91f780686deb8ee64ecf1ae71/msgpack-1.1.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:4fd6b577e4541676e0cc9ddc1709d25014d3ad9a66caa19962c4f5de30fc09ef", size = 412182, upload-time = "2025-06-13T06:52:11.644Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/d0/0cf4a6ecb9bc960d624c93effaeaae75cbf00b3bc4a54f35c8507273cda1/msgpack-1.1.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:bb29aaa613c0a1c40d1af111abf025f1732cab333f96f285d6a93b934738a68a", size = 419883, upload-time = "2025-06-13T06:52:12.806Z" },
-    { url = "https://files.pythonhosted.org/packages/62/83/9697c211720fa71a2dfb632cad6196a8af3abea56eece220fde4674dc44b/msgpack-1.1.1-cp312-cp312-win32.whl", hash = "sha256:870b9a626280c86cff9c576ec0d9cbcc54a1e5ebda9cd26dab12baf41fee218c", size = 65406, upload-time = "2025-06-13T06:52:14.271Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/23/0abb886e80eab08f5e8c485d6f13924028602829f63b8f5fa25a06636628/msgpack-1.1.1-cp312-cp312-win_amd64.whl", hash = "sha256:5692095123007180dca3e788bb4c399cc26626da51629a31d40207cb262e67f4", size = 72558, upload-time = "2025-06-13T06:52:15.252Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/bd/0792be119d7fe7dc2148689ef65c90507d82d20a204aab3b98c74a1f8684/msgpack-1.1.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f5be6b6bc52fad84d010cb45433720327ce886009d862f46b26d4d154001994b", size = 81882, upload-time = "2025-06-13T06:52:39.316Z" },
-    { url = "https://files.pythonhosted.org/packages/75/77/ce06c8e26a816ae8730a8e030d263c5289adcaff9f0476f9b270bdd7c5c2/msgpack-1.1.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:3a89cd8c087ea67e64844287ea52888239cbd2940884eafd2dcd25754fb72232", size = 78414, upload-time = "2025-06-13T06:52:40.341Z" },
-    { url = "https://files.pythonhosted.org/packages/73/27/190576c497677fb4a0d05d896b24aea6cdccd910f206aaa7b511901befed/msgpack-1.1.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1d75f3807a9900a7d575d8d6674a3a47e9f227e8716256f35bc6f03fc597ffbf", size = 400927, upload-time = "2025-06-13T06:52:41.399Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/af/6a0aa5a06762e70726ec3c10fb966600d84a7220b52635cb0ab2dc64d32f/msgpack-1.1.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d182dac0221eb8faef2e6f44701812b467c02674a322c739355c39e94730cdbf", size = 405903, upload-time = "2025-06-13T06:52:42.699Z" },
-    { url = "https://files.pythonhosted.org/packages/1e/80/3f3da358cecbbe8eb12360814bd1277d59d2608485934742a074d99894a9/msgpack-1.1.1-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1b13fe0fb4aac1aa5320cd693b297fe6fdef0e7bea5518cbc2dd5299f873ae90", size = 393192, upload-time = "2025-06-13T06:52:43.986Z" },
-    { url = "https://files.pythonhosted.org/packages/98/c6/3a0ec7fdebbb4f3f8f254696cd91d491c29c501dbebd86286c17e8f68cd7/msgpack-1.1.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:435807eeb1bc791ceb3247d13c79868deb22184e1fc4224808750f0d7d1affc1", size = 393851, upload-time = "2025-06-13T06:52:45.177Z" },
-    { url = "https://files.pythonhosted.org/packages/39/37/df50d5f8e68514b60fbe70f6e8337ea2b32ae2be030871bcd9d1cf7d4b62/msgpack-1.1.1-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:4835d17af722609a45e16037bb1d4d78b7bdf19d6c0128116d178956618c4e88", size = 400292, upload-time = "2025-06-13T06:52:46.381Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/ec/1e067292e02d2ceb4c8cb5ba222c4f7bb28730eef5676740609dc2627e0f/msgpack-1.1.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:a8ef6e342c137888ebbfb233e02b8fbd689bb5b5fcc59b34711ac47ebd504478", size = 401873, upload-time = "2025-06-13T06:52:47.957Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/31/e8c9c6b5b58d64c9efa99c8d181fcc25f38ead357b0360379fbc8a4234ad/msgpack-1.1.1-cp39-cp39-win32.whl", hash = "sha256:61abccf9de335d9efd149e2fff97ed5974f2481b3353772e8e2dd3402ba2bd57", size = 65028, upload-time = "2025-06-13T06:52:49.166Z" },
-    { url = "https://files.pythonhosted.org/packages/20/d6/cd62cded572e5e25892747a5d27850170bcd03c855e9c69c538e024de6f9/msgpack-1.1.1-cp39-cp39-win_amd64.whl", hash = "sha256:40eae974c873b2992fd36424a5d9407f93e97656d999f43fca9d29f820899084", size = 71700, upload-time = "2025-06-13T06:52:50.244Z" },
-]
-
-[[package]]
-name = "msgspec"
-version = "0.19.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/cf/9b/95d8ce458462b8b71b8a70fa94563b2498b89933689f3a7b8911edfae3d7/msgspec-0.19.0.tar.gz", hash = "sha256:604037e7cd475345848116e89c553aa9a233259733ab51986ac924ab1b976f8e", size = 216934, upload-time = "2024-12-27T17:40:28.597Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/e5/27/322badde18eb234e36d4a14122b89edd4e2973cdbc3da61ca7edf40a1ccd/msgspec-0.19.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fe2c4bf29bf4e89790b3117470dea2c20b59932772483082c468b990d45fb947", size = 209334, upload-time = "2024-12-27T17:39:19.065Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/65/080509c5774a1592b2779d902a70b5fe008532759927e011f068145a16cb/msgspec-0.19.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:00e87ecfa9795ee5214861eab8326b0e75475c2e68a384002aa135ea2a27d909", size = 211551, upload-time = "2024-12-27T17:39:21.767Z" },
-    { url = "https://files.pythonhosted.org/packages/6f/2e/1c23c6b4ca6f4285c30a39def1054e2bee281389e4b681b5e3711bd5a8c9/msgspec-0.19.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:3c4ec642689da44618f68c90855a10edbc6ac3ff7c1d94395446c65a776e712a", size = 215099, upload-time = "2024-12-27T17:39:24.71Z" },
-    { url = "https://files.pythonhosted.org/packages/83/fe/95f9654518879f3359d1e76bc41189113aa9102452170ab7c9a9a4ee52f6/msgspec-0.19.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:2719647625320b60e2d8af06b35f5b12d4f4d281db30a15a1df22adb2295f633", size = 218211, upload-time = "2024-12-27T17:39:27.396Z" },
-    { url = "https://files.pythonhosted.org/packages/81/25/3a4b24d468203d8af90d1d351b77ea3cffb96b29492855cf83078f16bfe4/msgspec-0.19.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e78f46ff39a427e10b4a61614a2777ad69559cc8d603a7c05681f5a595ea98f7", size = 209029, upload-time = "2024-12-27T17:39:35.023Z" },
-    { url = "https://files.pythonhosted.org/packages/85/2e/db7e189b57901955239f7689b5dcd6ae9458637a9c66747326726c650523/msgspec-0.19.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6c7adf191e4bd3be0e9231c3b6dc20cf1199ada2af523885efc2ed218eafd011", size = 210682, upload-time = "2024-12-27T17:39:36.384Z" },
-    { url = "https://files.pythonhosted.org/packages/03/97/7c8895c9074a97052d7e4a1cc1230b7b6e2ca2486714eb12c3f08bb9d284/msgspec-0.19.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:f04cad4385e20be7c7176bb8ae3dca54a08e9756cfc97bcdb4f18560c3042063", size = 214003, upload-time = "2024-12-27T17:39:39.097Z" },
-    { url = "https://files.pythonhosted.org/packages/61/61/e892997bcaa289559b4d5869f066a8021b79f4bf8e955f831b095f47a4cd/msgspec-0.19.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:45c8fb410670b3b7eb884d44a75589377c341ec1392b778311acdbfa55187716", size = 216833, upload-time = "2024-12-27T17:39:41.203Z" },
-    { url = "https://files.pythonhosted.org/packages/87/81/0c8c93f0b92c97e326b279795f9c5b956c5a97af28ca0fbb9fd86c83737a/msgspec-0.19.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4cfc033c02c3e0aec52b71710d7f84cb3ca5eb407ab2ad23d75631153fdb1f12", size = 210633, upload-time = "2024-12-27T17:39:49.099Z" },
-    { url = "https://files.pythonhosted.org/packages/d0/ef/c5422ce8af73928d194a6606f8ae36e93a52fd5e8df5abd366903a5ca8da/msgspec-0.19.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d911c442571605e17658ca2b416fd8579c5050ac9adc5e00c2cb3126c97f73bc", size = 213594, upload-time = "2024-12-27T17:39:51.204Z" },
-    { url = "https://files.pythonhosted.org/packages/19/2b/4137bc2ed45660444842d042be2cf5b18aa06efd2cda107cff18253b9653/msgspec-0.19.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:757b501fa57e24896cf40a831442b19a864f56d253679f34f260dcb002524a6c", size = 214053, upload-time = "2024-12-27T17:39:52.866Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/e6/8ad51bdc806aac1dc501e8fe43f759f9ed7284043d722b53323ea421c360/msgspec-0.19.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:5f0f65f29b45e2816d8bded36e6b837a4bf5fb60ec4bc3c625fa2c6da4124537", size = 219081, upload-time = "2024-12-27T17:39:55.142Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/a2/48f2c15c7644668e51f4dce99d5f709bd55314e47acb02e90682f5880f35/msgspec-0.19.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0f5c043ace7962ef188746e83b99faaa9e3e699ab857ca3f367b309c8e2c6b12", size = 209272, upload-time = "2024-12-27T17:40:21.534Z" },
-    { url = "https://files.pythonhosted.org/packages/25/3c/aa339cf08b990c3f07e67b229a3a8aa31bf129ed974b35e5daa0df7d9d56/msgspec-0.19.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ca06aa08e39bf57e39a258e1996474f84d0dd8130d486c00bec26d797b8c5446", size = 211396, upload-time = "2024-12-27T17:40:22.897Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/00/c7fb9d524327c558b2803973cc3f988c5100a1708879970a9e377bdf6f4f/msgspec-0.19.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:e695dad6897896e9384cf5e2687d9ae9feaef50e802f93602d35458e20d1fb19", size = 215002, upload-time = "2024-12-27T17:40:24.341Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/bf/d9f9fff026c1248cde84a5ce62b3742e8a63a3c4e811f99f00c8babf7615/msgspec-0.19.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:3be5c02e1fee57b54130316a08fe40cca53af92999a302a6054cd451700ea7db", size = 218132, upload-time = "2024-12-27T17:40:25.744Z" },
+version = "1.1.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/4d/f2/bfb55a6236ed8725a96b0aa3acbd0ec17588e6a2c3b62a93eb513ed8783f/msgpack-1.1.2.tar.gz", hash = "sha256:3b60763c1373dd60f398488069bcdc703cd08a711477b5d480eecc9f9626f47e", size = 173581, upload-time = "2025-10-08T09:15:56.596Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f5/a2/3b68a9e769db68668b25c6108444a35f9bd163bb848c0650d516761a59c0/msgpack-1.1.2-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:0051fffef5a37ca2cd16978ae4f0aef92f164df86823871b5162812bebecd8e2", size = 81318, upload-time = "2025-10-08T09:14:38.722Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/e1/2b720cc341325c00be44e1ed59e7cfeae2678329fbf5aa68f5bda57fe728/msgpack-1.1.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:a605409040f2da88676e9c9e5853b3449ba8011973616189ea5ee55ddbc5bc87", size = 83786, upload-time = "2025-10-08T09:14:40.082Z" },
+    { url = "https://files.pythonhosted.org/packages/71/e5/c2241de64bfceac456b140737812a2ab310b10538a7b34a1d393b748e095/msgpack-1.1.2-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8b696e83c9f1532b4af884045ba7f3aa741a63b2bc22617293a2c6a7c645f251", size = 398240, upload-time = "2025-10-08T09:14:41.151Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/09/2a06956383c0fdebaef5aa9246e2356776f12ea6f2a44bd1368abf0e46c4/msgpack-1.1.2-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:365c0bbe981a27d8932da71af63ef86acc59ed5c01ad929e09a0b88c6294e28a", size = 406070, upload-time = "2025-10-08T09:14:42.821Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/74/2957703f0e1ef20637d6aead4fbb314330c26f39aa046b348c7edcf6ca6b/msgpack-1.1.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:41d1a5d875680166d3ac5c38573896453bbbea7092936d2e107214daf43b1d4f", size = 393403, upload-time = "2025-10-08T09:14:44.38Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/09/3bfc12aa90f77b37322fc33e7a8a7c29ba7c8edeadfa27664451801b9860/msgpack-1.1.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:354e81bcdebaab427c3df4281187edc765d5d76bfb3a7c125af9da7a27e8458f", size = 398947, upload-time = "2025-10-08T09:14:45.56Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/4f/05fcebd3b4977cb3d840f7ef6b77c51f8582086de5e642f3fefee35c86fc/msgpack-1.1.2-cp310-cp310-win32.whl", hash = "sha256:e64c8d2f5e5d5fda7b842f55dec6133260ea8f53c4257d64494c534f306bf7a9", size = 64769, upload-time = "2025-10-08T09:14:47.334Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/3e/b4547e3a34210956382eed1c85935fff7e0f9b98be3106b3745d7dec9c5e/msgpack-1.1.2-cp310-cp310-win_amd64.whl", hash = "sha256:db6192777d943bdaaafb6ba66d44bf65aa0e9c5616fa1d2da9bb08828c6b39aa", size = 71293, upload-time = "2025-10-08T09:14:48.665Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/97/560d11202bcd537abca693fd85d81cebe2107ba17301de42b01ac1677b69/msgpack-1.1.2-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:2e86a607e558d22985d856948c12a3fa7b42efad264dca8a3ebbcfa2735d786c", size = 82271, upload-time = "2025-10-08T09:14:49.967Z" },
+    { url = "https://files.pythonhosted.org/packages/83/04/28a41024ccbd67467380b6fb440ae916c1e4f25e2cd4c63abe6835ac566e/msgpack-1.1.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:283ae72fc89da59aa004ba147e8fc2f766647b1251500182fac0350d8af299c0", size = 84914, upload-time = "2025-10-08T09:14:50.958Z" },
+    { url = "https://files.pythonhosted.org/packages/71/46/b817349db6886d79e57a966346cf0902a426375aadc1e8e7a86a75e22f19/msgpack-1.1.2-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:61c8aa3bd513d87c72ed0b37b53dd5c5a0f58f2ff9f26e1555d3bd7948fb7296", size = 416962, upload-time = "2025-10-08T09:14:51.997Z" },
+    { url = "https://files.pythonhosted.org/packages/da/e0/6cc2e852837cd6086fe7d8406af4294e66827a60a4cf60b86575a4a65ca8/msgpack-1.1.2-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:454e29e186285d2ebe65be34629fa0e8605202c60fbc7c4c650ccd41870896ef", size = 426183, upload-time = "2025-10-08T09:14:53.477Z" },
+    { url = "https://files.pythonhosted.org/packages/25/98/6a19f030b3d2ea906696cedd1eb251708e50a5891d0978b012cb6107234c/msgpack-1.1.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:7bc8813f88417599564fafa59fd6f95be417179f76b40325b500b3c98409757c", size = 411454, upload-time = "2025-10-08T09:14:54.648Z" },
+    { url = "https://files.pythonhosted.org/packages/b7/cd/9098fcb6adb32187a70b7ecaabf6339da50553351558f37600e53a4a2a23/msgpack-1.1.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:bafca952dc13907bdfdedfc6a5f579bf4f292bdd506fadb38389afa3ac5b208e", size = 422341, upload-time = "2025-10-08T09:14:56.328Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/ae/270cecbcf36c1dc85ec086b33a51a4d7d08fc4f404bdbc15b582255d05ff/msgpack-1.1.2-cp311-cp311-win32.whl", hash = "sha256:602b6740e95ffc55bfb078172d279de3773d7b7db1f703b2f1323566b878b90e", size = 64747, upload-time = "2025-10-08T09:14:57.882Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/79/309d0e637f6f37e83c711f547308b91af02b72d2326ddd860b966080ef29/msgpack-1.1.2-cp311-cp311-win_amd64.whl", hash = "sha256:d198d275222dc54244bf3327eb8cbe00307d220241d9cec4d306d49a44e85f68", size = 71633, upload-time = "2025-10-08T09:14:59.177Z" },
+    { url = "https://files.pythonhosted.org/packages/73/4d/7c4e2b3d9b1106cd0aa6cb56cc57c6267f59fa8bfab7d91df5adc802c847/msgpack-1.1.2-cp311-cp311-win_arm64.whl", hash = "sha256:86f8136dfa5c116365a8a651a7d7484b65b13339731dd6faebb9a0242151c406", size = 64755, upload-time = "2025-10-08T09:15:00.48Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/bd/8b0d01c756203fbab65d265859749860682ccd2a59594609aeec3a144efa/msgpack-1.1.2-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:70a0dff9d1f8da25179ffcf880e10cf1aad55fdb63cd59c9a49a1b82290062aa", size = 81939, upload-time = "2025-10-08T09:15:01.472Z" },
+    { url = "https://files.pythonhosted.org/packages/34/68/ba4f155f793a74c1483d4bdef136e1023f7bcba557f0db4ef3db3c665cf1/msgpack-1.1.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:446abdd8b94b55c800ac34b102dffd2f6aa0ce643c55dfc017ad89347db3dbdb", size = 85064, upload-time = "2025-10-08T09:15:03.764Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/60/a064b0345fc36c4c3d2c743c82d9100c40388d77f0b48b2f04d6041dbec1/msgpack-1.1.2-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c63eea553c69ab05b6747901b97d620bb2a690633c77f23feb0c6a947a8a7b8f", size = 417131, upload-time = "2025-10-08T09:15:05.136Z" },
+    { url = "https://files.pythonhosted.org/packages/65/92/a5100f7185a800a5d29f8d14041f61475b9de465ffcc0f3b9fba606e4505/msgpack-1.1.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:372839311ccf6bdaf39b00b61288e0557916c3729529b301c52c2d88842add42", size = 427556, upload-time = "2025-10-08T09:15:06.837Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/87/ffe21d1bf7d9991354ad93949286f643b2bb6ddbeab66373922b44c3b8cc/msgpack-1.1.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2929af52106ca73fcb28576218476ffbb531a036c2adbcf54a3664de124303e9", size = 404920, upload-time = "2025-10-08T09:15:08.179Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/41/8543ed2b8604f7c0d89ce066f42007faac1eaa7d79a81555f206a5cdb889/msgpack-1.1.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:be52a8fc79e45b0364210eef5234a7cf8d330836d0a64dfbb878efa903d84620", size = 415013, upload-time = "2025-10-08T09:15:09.83Z" },
+    { url = "https://files.pythonhosted.org/packages/41/0d/2ddfaa8b7e1cee6c490d46cb0a39742b19e2481600a7a0e96537e9c22f43/msgpack-1.1.2-cp312-cp312-win32.whl", hash = "sha256:1fff3d825d7859ac888b0fbda39a42d59193543920eda9d9bea44d958a878029", size = 65096, upload-time = "2025-10-08T09:15:11.11Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/ec/d431eb7941fb55a31dd6ca3404d41fbb52d99172df2e7707754488390910/msgpack-1.1.2-cp312-cp312-win_amd64.whl", hash = "sha256:1de460f0403172cff81169a30b9a92b260cb809c4cb7e2fc79ae8d0510c78b6b", size = 72708, upload-time = "2025-10-08T09:15:12.554Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/31/5b1a1f70eb0e87d1678e9624908f86317787b536060641d6798e3cf70ace/msgpack-1.1.2-cp312-cp312-win_arm64.whl", hash = "sha256:be5980f3ee0e6bd44f3a9e9dea01054f175b50c3e6cdb692bc9424c0bbb8bf69", size = 64119, upload-time = "2025-10-08T09:15:13.589Z" },
+    { url = "https://files.pythonhosted.org/packages/46/73/85469b4aa71d25e5949fee50d3c2cf46f69cea619fe97cfe309058080f75/msgpack-1.1.2-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:ea5405c46e690122a76531ab97a079e184c0daf491e588592d6a23d3e32af99e", size = 81529, upload-time = "2025-10-08T09:15:46.069Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/3a/7d4077e8ae720b29d2b299a9591969f0d105146960681ea6f4121e6d0f8d/msgpack-1.1.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:9fba231af7a933400238cb357ecccf8ab5d51535ea95d94fc35b7806218ff844", size = 84106, upload-time = "2025-10-08T09:15:47.064Z" },
+    { url = "https://files.pythonhosted.org/packages/df/c0/da451c74746ed9388dca1b4ec647c82945f4e2f8ce242c25fb7c0e12181f/msgpack-1.1.2-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a8f6e7d30253714751aa0b0c84ae28948e852ee7fb0524082e6716769124bc23", size = 396656, upload-time = "2025-10-08T09:15:48.118Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/a1/20486c29a31ec9f0f88377fdf7eb7a67f30bcb5e0f89b7550f6f16d9373b/msgpack-1.1.2-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:94fd7dc7d8cb0a54432f296f2246bc39474e017204ca6f4ff345941d4ed285a7", size = 404722, upload-time = "2025-10-08T09:15:49.328Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/ae/e613b0a526d54ce85447d9665c2ff8c3210a784378d50573321d43d324b8/msgpack-1.1.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:350ad5353a467d9e3b126d8d1b90fe05ad081e2e1cef5753f8c345217c37e7b8", size = 391838, upload-time = "2025-10-08T09:15:50.517Z" },
+    { url = "https://files.pythonhosted.org/packages/49/6a/07f3e10ed4503045b882ef7bf8512d01d8a9e25056950a977bd5f50df1c2/msgpack-1.1.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:6bde749afe671dc44893f8d08e83bf475a1a14570d67c4bb5cec5573463c8833", size = 397516, upload-time = "2025-10-08T09:15:51.646Z" },
+    { url = "https://files.pythonhosted.org/packages/76/9b/a86828e75986c12a3809c1e5062f5eba8e0cae3dfa2bf724ed2b1bb72b4c/msgpack-1.1.2-cp39-cp39-win32.whl", hash = "sha256:ad09b984828d6b7bb52d1d1d0c9be68ad781fa004ca39216c8a1e63c0f34ba3c", size = 64863, upload-time = "2025-10-08T09:15:53.118Z" },
+    { url = "https://files.pythonhosted.org/packages/14/a7/b1992b4fb3da3b413f5fb78a63bad42f256c3be2352eb69273c3789c2c96/msgpack-1.1.2-cp39-cp39-win_amd64.whl", hash = "sha256:67016ae8c8965124fdede9d3769528ad8284f14d635337ffa6a713a580f6c030", size = 71540, upload-time = "2025-10-08T09:15:55.573Z" },
 ]
 
 [[package]]
@@ -2434,22 +2399,27 @@ wheels = [
 
 [[package]]
 name = "multiprocess"
-version = "0.70.16"
+version = "0.70.18"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "dill" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b5/ae/04f39c5d0d0def03247c2893d6f2b83c136bf3320a2154d7b8858f2ba72d/multiprocess-0.70.16.tar.gz", hash = "sha256:161af703d4652a0e1410be6abccecde4a7ddffd19341be0a7011b94aeb171ac1", size = 1772603, upload-time = "2024-01-28T18:52:34.85Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/72/fd/2ae3826f5be24c6ed87266bc4e59c46ea5b059a103f3d7e7eb76a52aeecb/multiprocess-0.70.18.tar.gz", hash = "sha256:f9597128e6b3e67b23956da07cf3d2e5cba79e2f4e0fba8d7903636663ec6d0d", size = 1798503, upload-time = "2025-04-17T03:11:27.742Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/ef/76/6e712a2623d146d314f17598df5de7224c85c0060ef63fd95cc15a25b3fa/multiprocess-0.70.16-pp310-pypy310_pp73-macosx_10_13_x86_64.whl", hash = "sha256:476887be10e2f59ff183c006af746cb6f1fd0eadcfd4ef49e605cbe2659920ee", size = 134980, upload-time = "2024-01-28T18:52:15.731Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/ab/1e6e8009e380e22254ff539ebe117861e5bdb3bff1fc977920972237c6c7/multiprocess-0.70.16-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:d951bed82c8f73929ac82c61f01a7b5ce8f3e5ef40f5b52553b4f547ce2b08ec", size = 134982, upload-time = "2024-01-28T18:52:17.783Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/94/8638a89f93c80df329116e6781a060506c7e91e1f4370dc831e9d17a041d/multiprocess-0.70.16-pp39-pypy39_pp73-macosx_10_13_x86_64.whl", hash = "sha256:0dfd078c306e08d46d7a8d06fb120313d87aa43af60d66da43ffff40b44d2f41", size = 133497, upload-time = "2024-01-28T18:52:22.644Z" },
-    { url = "https://files.pythonhosted.org/packages/89/21/222066f6bb8d8af287923ae3bd26cf4699a9ce020228ac273caca1de8250/multiprocess-0.70.16-pp39-pypy39_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:e7b9d0f307cd9bd50851afaac0dba2cb6c44449efff697df7c7645f7d3f2be3a", size = 133498, upload-time = "2024-01-28T18:52:24.576Z" },
-    { url = "https://files.pythonhosted.org/packages/bc/f7/7ec7fddc92e50714ea3745631f79bd9c96424cb2702632521028e57d3a36/multiprocess-0.70.16-py310-none-any.whl", hash = "sha256:c4a9944c67bd49f823687463660a2d6daae94c289adff97e0f9d696ba6371d02", size = 134824, upload-time = "2024-01-28T18:52:26.062Z" },
-    { url = "https://files.pythonhosted.org/packages/50/15/b56e50e8debaf439f44befec5b2af11db85f6e0f344c3113ae0be0593a91/multiprocess-0.70.16-py311-none-any.whl", hash = "sha256:af4cabb0dac72abfb1e794fa7855c325fd2b55a10a44628a3c1ad3311c04127a", size = 143519, upload-time = "2024-01-28T18:52:28.115Z" },
-    { url = "https://files.pythonhosted.org/packages/0a/7d/a988f258104dcd2ccf1ed40fdc97e26c4ac351eeaf81d76e266c52d84e2f/multiprocess-0.70.16-py312-none-any.whl", hash = "sha256:fc0544c531920dde3b00c29863377f87e1632601092ea2daca74e4beb40faa2e", size = 146741, upload-time = "2024-01-28T18:52:29.395Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/89/38df130f2c799090c978b366cfdf5b96d08de5b29a4a293df7f7429fa50b/multiprocess-0.70.16-py38-none-any.whl", hash = "sha256:a71d82033454891091a226dfc319d0cfa8019a4e888ef9ca910372a446de4435", size = 132628, upload-time = "2024-01-28T18:52:30.853Z" },
-    { url = "https://files.pythonhosted.org/packages/da/d9/f7f9379981e39b8c2511c9e0326d212accacb82f12fbfdc1aa2ce2a7b2b6/multiprocess-0.70.16-py39-none-any.whl", hash = "sha256:a0bafd3ae1b732eac64be2e72038231c1ba97724b60b09400d68f229fcc2fbf3", size = 133351, upload-time = "2024-01-28T18:52:31.981Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/f8/7f9a8f08bf98cea1dfaa181e05cc8bbcb59cecf044b5a9ac3cce39f9c449/multiprocess-0.70.18-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:25d4012dcaaf66b9e8e955f58482b42910c2ee526d532844d8bcf661bbc604df", size = 135083, upload-time = "2025-04-17T03:11:04.223Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/03/b7b10dbfc17b2b3ce07d4d30b3ba8367d0ed32d6d46cd166e298f161dd46/multiprocess-0.70.18-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:06b19433de0d02afe5869aec8931dd5c01d99074664f806c73896b0d9e527213", size = 135128, upload-time = "2025-04-17T03:11:06.045Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/a3/5f8d3b9690ea5580bee5868ab7d7e2cfca74b7e826b28192b40aa3881cdc/multiprocess-0.70.18-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:6fa1366f994373aaf2d4738b0f56e707caeaa05486e97a7f71ee0853823180c2", size = 135132, upload-time = "2025-04-17T03:11:07.533Z" },
+    { url = "https://files.pythonhosted.org/packages/55/4d/9af0d1279c84618bcd35bf5fd7e371657358c7b0a523e54a9cffb87461f8/multiprocess-0.70.18-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:8b8940ae30139e04b076da6c5b83e9398585ebdf0f2ad3250673fef5b2ff06d6", size = 144695, upload-time = "2025-04-17T03:11:09.161Z" },
+    { url = "https://files.pythonhosted.org/packages/17/bf/87323e79dd0562474fad3373c21c66bc6c3c9963b68eb2a209deb4c8575e/multiprocess-0.70.18-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:0929ba95831adb938edbd5fb801ac45e705ecad9d100b3e653946b7716cb6bd3", size = 144742, upload-time = "2025-04-17T03:11:10.072Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/74/cb8c831e58dc6d5cf450b17c7db87f14294a1df52eb391da948b5e0a0b94/multiprocess-0.70.18-pp311-pypy311_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:4d77f8e4bfe6c6e2e661925bbf9aed4d5ade9a1c6502d5dfc10129b9d1141797", size = 144745, upload-time = "2025-04-17T03:11:11.453Z" },
+    { url = "https://files.pythonhosted.org/packages/12/89/733ebfc487a4381590d863222f3f0c8bac105d032102d7a96d6566823d67/multiprocess-0.70.18-pp39-pypy39_pp73-macosx_10_13_arm64.whl", hash = "sha256:9fd8d662f7524a95a1be7cbea271f0b33089fe792baabec17d93103d368907da", size = 133534, upload-time = "2025-04-17T03:11:17.045Z" },
+    { url = "https://files.pythonhosted.org/packages/41/65/d74a7955593c2c56c79d01ef07be5aa103342441673657dc7387855703de/multiprocess-0.70.18-pp39-pypy39_pp73-macosx_10_13_x86_64.whl", hash = "sha256:3fbba48bfcd932747c33f0b152b26207c4e0840c35cab359afaff7a8672b1031", size = 133533, upload-time = "2025-04-17T03:11:18.378Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/5a/6f3ebcbc1508aa651cbe8deeca612b7915b97303410c93e9a4d83ba07e03/multiprocess-0.70.18-pp39-pypy39_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:5f9be0342e597dde86152c10442c5fb6c07994b1c29de441b7a3a08b0e6be2a0", size = 133537, upload-time = "2025-04-17T03:11:19.29Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/d8/0cba6cf51a1a31f20471fbc823a716170c73012ddc4fb85d706630ed6e8f/multiprocess-0.70.18-py310-none-any.whl", hash = "sha256:60c194974c31784019c1f459d984e8f33ee48f10fcf42c309ba97b30d9bd53ea", size = 134948, upload-time = "2025-04-17T03:11:20.223Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/88/9039f2fed1012ef584751d4ceff9ab4a51e5ae264898f0b7cbf44340a859/multiprocess-0.70.18-py311-none-any.whl", hash = "sha256:5aa6eef98e691281b3ad923be2832bf1c55dd2c859acd73e5ec53a66aae06a1d", size = 144462, upload-time = "2025-04-17T03:11:21.657Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/b6/5f922792be93b82ec6b5f270bbb1ef031fd0622847070bbcf9da816502cc/multiprocess-0.70.18-py312-none-any.whl", hash = "sha256:9b78f8e5024b573730bfb654783a13800c2c0f2dfc0c25e70b40d184d64adaa2", size = 150287, upload-time = "2025-04-17T03:11:22.69Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/c3/ca84c19bd14cdfc21c388fdcebf08b86a7a470ebc9f5c3c084fc2dbc50f7/multiprocess-0.70.18-py38-none-any.whl", hash = "sha256:dbf705e52a154fe5e90fb17b38f02556169557c2dd8bb084f2e06c2784d8279b", size = 132636, upload-time = "2025-04-17T03:11:24.936Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/28/dd72947e59a6a8c856448a5e74da6201cb5502ddff644fbc790e4bd40b9a/multiprocess-0.70.18-py39-none-any.whl", hash = "sha256:e78ca805a72b1b810c690b6b4cc32579eba34f403094bbbae962b7b5bf9dfcb8", size = 133478, upload-time = "2025-04-17T03:11:26.253Z" },
 ]
 
 [[package]]
@@ -2496,15 +2466,34 @@ wheels = [
 
 [[package]]
 name = "networkx"
-version = "3.5"
+version = "3.6"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
     "python_full_version == '3.11.*'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/6c/4f/ccdb8ad3a38e583f214547fd2f7ff1fc160c43a75af88e6aec213404b96a/networkx-3.5.tar.gz", hash = "sha256:d4c6f9cf81f52d69230866796b82afbccdec3db7ae4fbd1b65ea750feed50037", size = 2471065, upload-time = "2025-05-29T11:35:07.804Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/e8/fc/7b6fd4d22c8c4dc5704430140d8b3f520531d4fe7328b8f8d03f5a7950e8/networkx-3.6.tar.gz", hash = "sha256:285276002ad1f7f7da0f7b42f004bcba70d381e936559166363707fdad3d72ad", size = 2511464, upload-time = "2025-11-24T03:03:47.158Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/07/c7/d64168da60332c17d24c0d2f08bdf3987e8d1ae9d84b5bbd0eec2eb26a55/networkx-3.6-py3-none-any.whl", hash = "sha256:cdb395b105806062473d3be36458d8f1459a4e4b98e236a66c3a48996e07684f", size = 2063713, upload-time = "2025-11-24T03:03:45.21Z" },
+]
+
+[[package]]
+name = "nh3"
+version = "0.3.2"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/ca/a5/34c26015d3a434409f4d2a1cd8821a06c05238703f49283ffeb937bef093/nh3-0.3.2.tar.gz", hash = "sha256:f394759a06df8b685a4ebfb1874fb67a9cbfd58c64fc5ed587a663c0e63ec376", size = 19288, upload-time = "2025-10-30T11:17:45.948Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/eb/8d/776adee7bbf76365fdd7f2552710282c79a4ead5d2a46408c9043a2b70ba/networkx-3.5-py3-none-any.whl", hash = "sha256:0030d386a9a06dee3565298b4a734b68589749a544acbb6c412dc9e2489ec6ec", size = 2034406, upload-time = "2025-05-29T11:35:04.961Z" },
+    { url = "https://files.pythonhosted.org/packages/7f/f7/529a99324d7ef055de88b690858f4189379708abae92ace799365a797b7f/nh3-0.3.2-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c8745454cdd28bbbc90861b80a0111a195b0e3961b9fa2e672be89eb199fa5d8", size = 820805, upload-time = "2025-10-30T11:17:26.98Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/62/19b7c50ccd1fa7d0764822d2cea8f2a320f2fd77474c7a1805cb22cf69b0/nh3-0.3.2-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:72d67c25a84579f4a432c065e8b4274e53b7cf1df8f792cf846abfe2c3090866", size = 803527, upload-time = "2025-10-30T11:17:28.284Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/ca/f022273bab5440abff6302731a49410c5ef66b1a9502ba3fbb2df998d9ff/nh3-0.3.2-cp38-abi3-manylinux_2_17_ppc64.manylinux2014_ppc64.whl", hash = "sha256:13398e676a14d6233f372c75f52d5ae74f98210172991f7a3142a736bd92b131", size = 1051674, upload-time = "2025-10-30T11:17:29.909Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/f7/5728e3b32a11daf5bd21cf71d91c463f74305938bc3eb9e0ac1ce141646e/nh3-0.3.2-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:03d617e5c8aa7331bd2659c654e021caf9bba704b109e7b2b28b039a00949fe5", size = 1004737, upload-time = "2025-10-30T11:17:31.205Z" },
+    { url = "https://files.pythonhosted.org/packages/53/7f/f17e0dba0a99cee29e6cee6d4d52340ef9cb1f8a06946d3a01eb7ec2fb01/nh3-0.3.2-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f2f55c4d2d5a207e74eefe4d828067bbb01300e06e2a7436142f915c5928de07", size = 911745, upload-time = "2025-10-30T11:17:32.945Z" },
+    { url = "https://files.pythonhosted.org/packages/42/0f/c76bf3dba22c73c38e9b1113b017cf163f7696f50e003404ec5ecdb1e8a6/nh3-0.3.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7bb18403f02b655a1bbe4e3a4696c2ae1d6ae8f5991f7cacb684b1ae27e6c9f7", size = 797184, upload-time = "2025-10-30T11:17:34.226Z" },
+    { url = "https://files.pythonhosted.org/packages/08/a1/73d8250f888fb0ddf1b119b139c382f8903d8bb0c5bd1f64afc7e38dad1d/nh3-0.3.2-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:6d66f41672eb4060cf87c037f760bdbc6847852ca9ef8e9c5a5da18f090abf87", size = 838556, upload-time = "2025-10-30T11:17:35.875Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/09/deb57f1fb656a7a5192497f4a287b0ade5a2ff6b5d5de4736d13ef6d2c1f/nh3-0.3.2-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:f97f8b25cb2681d25e2338148159447e4d689aafdccfcf19e61ff7db3905768a", size = 1006695, upload-time = "2025-10-30T11:17:37.071Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/61/8f4d41c4ccdac30e4b1a4fa7be4b0f9914d8314a5058472f84c8e101a418/nh3-0.3.2-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:2ab70e8c6c7d2ce953d2a58102eefa90c2d0a5ed7aa40c7e29a487bc5e613131", size = 1075471, upload-time = "2025-10-30T11:17:38.225Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/c6/966aec0cb4705e69f6c3580422c239205d5d4d0e50fac380b21e87b6cf1b/nh3-0.3.2-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:1710f3901cd6440ca92494ba2eb6dc260f829fa8d9196b659fa10de825610ce0", size = 1002439, upload-time = "2025-10-30T11:17:39.553Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/c8/97a2d5f7a314cce2c5c49f30c6f161b7f3617960ade4bfc2fd1ee092cb20/nh3-0.3.2-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:91e9b001101fb4500a2aafe3e7c92928d85242d38bf5ac0aba0b7480da0a4cd6", size = 987439, upload-time = "2025-10-30T11:17:40.81Z" },
 ]
 
 [[package]]
@@ -2579,7 +2568,7 @@ wheels = [
 
 [[package]]
 name = "numba"
-version = "0.61.2"
+version = "0.62.1"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -2587,26 +2576,26 @@ resolution-markers = [
     "python_full_version == '3.10.*'",
 ]
 dependencies = [
-    { name = "llvmlite", version = "0.44.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "llvmlite", version = "0.45.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/1c/a0/e21f57604304aa03ebb8e098429222722ad99176a4f979d34af1d1ee80da/numba-0.61.2.tar.gz", hash = "sha256:8750ee147940a6637b80ecf7f95062185ad8726c8c28a2295b8ec1160a196f7d", size = 2820615, upload-time = "2025-04-09T02:58:07.659Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/a3/20/33dbdbfe60e5fd8e3dbfde299d106279a33d9f8308346022316781368591/numba-0.62.1.tar.gz", hash = "sha256:7b774242aa890e34c21200a1fc62e5b5757d5286267e71103257f4e2af0d5161", size = 2749817, upload-time = "2025-09-29T10:46:31.551Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/eb/ca/f470be59552ccbf9531d2d383b67ae0b9b524d435fb4a0d229fef135116e/numba-0.61.2-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:cf9f9fc00d6eca0c23fc840817ce9f439b9f03c8f03d6246c0e7f0cb15b7162a", size = 2775663, upload-time = "2025-04-09T02:57:34.143Z" },
-    { url = "https://files.pythonhosted.org/packages/f5/13/3bdf52609c80d460a3b4acfb9fdb3817e392875c0d6270cf3fd9546f138b/numba-0.61.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:ea0247617edcb5dd61f6106a56255baab031acc4257bddaeddb3a1003b4ca3fd", size = 2778344, upload-time = "2025-04-09T02:57:36.609Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/7d/bfb2805bcfbd479f04f835241ecf28519f6e3609912e3a985aed45e21370/numba-0.61.2-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ae8c7a522c26215d5f62ebec436e3d341f7f590079245a2f1008dfd498cc1642", size = 3824054, upload-time = "2025-04-09T02:57:38.162Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/27/797b2004745c92955470c73c82f0e300cf033c791f45bdecb4b33b12bdea/numba-0.61.2-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:bd1e74609855aa43661edffca37346e4e8462f6903889917e9f41db40907daa2", size = 3518531, upload-time = "2025-04-09T02:57:39.709Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/c6/c2fb11e50482cb310afae87a997707f6c7d8a48967b9696271347441f650/numba-0.61.2-cp310-cp310-win_amd64.whl", hash = "sha256:ae45830b129c6137294093b269ef0a22998ccc27bf7cf096ab8dcf7bca8946f9", size = 2831612, upload-time = "2025-04-09T02:57:41.559Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/97/c99d1056aed767503c228f7099dc11c402906b42a4757fec2819329abb98/numba-0.61.2-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:efd3db391df53aaa5cfbee189b6c910a5b471488749fd6606c3f33fc984c2ae2", size = 2775825, upload-time = "2025-04-09T02:57:43.442Z" },
-    { url = "https://files.pythonhosted.org/packages/95/9e/63c549f37136e892f006260c3e2613d09d5120672378191f2dc387ba65a2/numba-0.61.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:49c980e4171948ffebf6b9a2520ea81feed113c1f4890747ba7f59e74be84b1b", size = 2778695, upload-time = "2025-04-09T02:57:44.968Z" },
-    { url = "https://files.pythonhosted.org/packages/97/c8/8740616c8436c86c1b9a62e72cb891177d2c34c2d24ddcde4c390371bf4c/numba-0.61.2-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3945615cd73c2c7eba2a85ccc9c1730c21cd3958bfcf5a44302abae0fb07bb60", size = 3829227, upload-time = "2025-04-09T02:57:46.63Z" },
-    { url = "https://files.pythonhosted.org/packages/fc/06/66e99ae06507c31d15ff3ecd1f108f2f59e18b6e08662cd5f8a5853fbd18/numba-0.61.2-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:bbfdf4eca202cebade0b7d43896978e146f39398909a42941c9303f82f403a18", size = 3523422, upload-time = "2025-04-09T02:57:48.222Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/a4/2b309a6a9f6d4d8cfba583401c7c2f9ff887adb5d54d8e2e130274c0973f/numba-0.61.2-cp311-cp311-win_amd64.whl", hash = "sha256:76bcec9f46259cedf888041b9886e257ae101c6268261b19fda8cfbc52bec9d1", size = 2831505, upload-time = "2025-04-09T02:57:50.108Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/a0/c6b7b9c615cfa3b98c4c63f4316e3f6b3bbe2387740277006551784218cd/numba-0.61.2-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:34fba9406078bac7ab052efbf0d13939426c753ad72946baaa5bf9ae0ebb8dd2", size = 2776626, upload-time = "2025-04-09T02:57:51.857Z" },
-    { url = "https://files.pythonhosted.org/packages/92/4a/fe4e3c2ecad72d88f5f8cd04e7f7cff49e718398a2fac02d2947480a00ca/numba-0.61.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:4ddce10009bc097b080fc96876d14c051cc0c7679e99de3e0af59014dab7dfe8", size = 2779287, upload-time = "2025-04-09T02:57:53.658Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/2d/e518df036feab381c23a624dac47f8445ac55686ec7f11083655eb707da3/numba-0.61.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5b1bb509d01f23d70325d3a5a0e237cbc9544dd50e50588bc581ba860c213546", size = 3885928, upload-time = "2025-04-09T02:57:55.206Z" },
-    { url = "https://files.pythonhosted.org/packages/10/0f/23cced68ead67b75d77cfcca3df4991d1855c897ee0ff3fe25a56ed82108/numba-0.61.2-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:48a53a3de8f8793526cbe330f2a39fe9a6638efcbf11bd63f3d2f9757ae345cd", size = 3577115, upload-time = "2025-04-09T02:57:56.818Z" },
-    { url = "https://files.pythonhosted.org/packages/68/1d/ddb3e704c5a8fb90142bf9dc195c27db02a08a99f037395503bfbc1d14b3/numba-0.61.2-cp312-cp312-win_amd64.whl", hash = "sha256:97cf4f12c728cf77c9c1d7c23707e4d8fb4632b46275f8f3397de33e5877af18", size = 2831929, upload-time = "2025-04-09T02:57:58.45Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/27/a5a9a58f267ec3b72f609789b2a8eefd6156bd7117e41cc9b7cf5de30490/numba-0.62.1-cp310-cp310-macosx_10_15_x86_64.whl", hash = "sha256:a323df9d36a0da1ca9c592a6baaddd0176d9f417ef49a65bb81951dce69d941a", size = 2684281, upload-time = "2025-09-29T10:43:31.863Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/9d/ffc091c0bfd7b80f66df3887a7061b6af80c8c2649902444026ee1454391/numba-0.62.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:e1e1f4781d3f9f7c23f16eb04e76ca10b5a3516e959634bd226fc48d5d8e7a0a", size = 2687311, upload-time = "2025-09-29T10:43:54.441Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/13/9a27bcd0baeea236116070c7df458414336f25e9dd5a872b066cf36b74bf/numba-0.62.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:14432af305ea68627a084cd702124fd5d0c1f5b8a413b05f4e14757202d1cf6c", size = 3734548, upload-time = "2025-09-29T10:42:38.232Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/00/17a1ac4a60253c784ce59549375e047da98330b82de7df6ac7f4ecc90902/numba-0.62.1-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f180922adf159ae36c2fe79fb94ffaa74cf5cb3688cb72dba0a904b91e978507", size = 3441277, upload-time = "2025-09-29T10:43:06.124Z" },
+    { url = "https://files.pythonhosted.org/packages/86/94/20ae0ff78612c4697eaf942a639db01dd4e2d90f634ac41fa3e015c961fc/numba-0.62.1-cp310-cp310-win_amd64.whl", hash = "sha256:f41834909d411b4b8d1c68f745144136f21416547009c1e860cc2098754b4ca7", size = 2745647, upload-time = "2025-09-29T10:44:15.282Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/5f/8b3491dd849474f55e33c16ef55678ace1455c490555337899c35826836c/numba-0.62.1-cp311-cp311-macosx_10_15_x86_64.whl", hash = "sha256:f43e24b057714e480fe44bc6031de499e7cf8150c63eb461192caa6cc8530bc8", size = 2684279, upload-time = "2025-09-29T10:43:37.213Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/18/71969149bfeb65a629e652b752b80167fe8a6a6f6e084f1f2060801f7f31/numba-0.62.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:57cbddc53b9ee02830b828a8428757f5c218831ccc96490a314ef569d8342b7b", size = 2687330, upload-time = "2025-09-29T10:43:59.601Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/7d/403be3fecae33088027bc8a95dc80a2fda1e3beff3e0e5fc4374ada3afbe/numba-0.62.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:604059730c637c7885386521bb1b0ddcbc91fd56131a6dcc54163d6f1804c872", size = 3739727, upload-time = "2025-09-29T10:42:45.922Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/c3/3d910d08b659a6d4c62ab3cd8cd93c4d8b7709f55afa0d79a87413027ff6/numba-0.62.1-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d6c540880170bee817011757dc9049dba5a29db0c09b4d2349295991fe3ee55f", size = 3445490, upload-time = "2025-09-29T10:43:12.692Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/82/9d425c2f20d9f0a37f7cb955945a553a00fa06a2b025856c3550227c5543/numba-0.62.1-cp311-cp311-win_amd64.whl", hash = "sha256:03de6d691d6b6e2b76660ba0f38f37b81ece8b2cc524a62f2a0cfae2bfb6f9da", size = 2745550, upload-time = "2025-09-29T10:44:20.571Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/fa/30fa6873e9f821c0ae755915a3ca444e6ff8d6a7b6860b669a3d33377ac7/numba-0.62.1-cp312-cp312-macosx_10_15_x86_64.whl", hash = "sha256:1b743b32f8fa5fff22e19c2e906db2f0a340782caf024477b97801b918cf0494", size = 2685346, upload-time = "2025-09-29T10:43:43.677Z" },
+    { url = "https://files.pythonhosted.org/packages/a9/d5/504ce8dc46e0dba2790c77e6b878ee65b60fe3e7d6d0006483ef6fde5a97/numba-0.62.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:90fa21b0142bcf08ad8e32a97d25d0b84b1e921bc9423f8dda07d3652860eef6", size = 2688139, upload-time = "2025-09-29T10:44:04.894Z" },
+    { url = "https://files.pythonhosted.org/packages/50/5f/6a802741176c93f2ebe97ad90751894c7b0c922b52ba99a4395e79492205/numba-0.62.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:6ef84d0ac19f1bf80431347b6f4ce3c39b7ec13f48f233a48c01e2ec06ecbc59", size = 3796453, upload-time = "2025-09-29T10:42:52.771Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/df/efd21527d25150c4544eccc9d0b7260a5dec4b7e98b5a581990e05a133c0/numba-0.62.1-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9315cc5e441300e0ca07c828a627d92a6802bcbf27c5487f31ae73783c58da53", size = 3496451, upload-time = "2025-09-29T10:43:19.279Z" },
+    { url = "https://files.pythonhosted.org/packages/80/44/79bfdab12a02796bf4f1841630355c82b5a69933b1d50eb15c7fa37dabe8/numba-0.62.1-cp312-cp312-win_amd64.whl", hash = "sha256:44e3aa6228039992f058f5ebfcfd372c83798e9464297bdad8cc79febcf7891e", size = 2745552, upload-time = "2025-09-29T10:44:26.399Z" },
 ]
 
 [[package]]
@@ -2817,10 +2806,26 @@ wheels = [
 name = "nvidia-nccl-cu12"
 version = "2.27.3"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/5c/5b/4e4fff7bad39adf89f735f2bc87248c81db71205b62bcc0d5ca5b606b3c3/nvidia_nccl_cu12-2.27.3-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:adf27ccf4238253e0b826bce3ff5fa532d65fc42322c8bfdfaf28024c0fbe039", size = 322364134, upload-time = "2025-06-03T21:58:04.013Z" },
 ]
 
+[[package]]
+name = "nvidia-nccl-cu12"
+version = "2.27.5"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/6e/89/f7a07dc961b60645dbbf42e80f2bc85ade7feb9a491b11a1e973aa00071f/nvidia_nccl_cu12-2.27.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ad730cf15cb5d25fe849c6e6ca9eb5b76db16a80f13f425ac68d8e2e55624457", size = 322348229, upload-time = "2025-06-26T04:11:28.385Z" },
+]
+
 [[package]]
 name = "nvidia-nvjitlink-cu12"
 version = "12.8.93"
@@ -2829,6 +2834,14 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f6/74/86a07f1d0f42998ca31312f998bd3b9a7eff7f52378f4f270c8679c77fb9/nvidia_nvjitlink_cu12-12.8.93-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl", hash = "sha256:81ff63371a7ebd6e6451970684f916be2eab07321b73c9d244dc2b4da7f73b88", size = 39254836, upload-time = "2025-03-07T01:49:55.661Z" },
 ]
 
+[[package]]
+name = "nvidia-nvshmem-cu12"
+version = "3.3.20"
+source = { registry = "https://pypi.org/simple" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3b/6c/99acb2f9eb85c29fc6f3a7ac4dccfd992e22666dd08a642b303311326a97/nvidia_nvshmem_cu12-3.3.20-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d00f26d3f9b2e3c3065be895e3059d6479ea5c638a3f38c9fec49b1b9dd7c1e5", size = 124657145, upload-time = "2025-08-04T20:25:19.995Z" },
+]
+
 [[package]]
 name = "nvidia-nvtx-cu12"
 version = "12.8.90"
@@ -2839,20 +2852,20 @@ wheels = [
 
 [[package]]
 name = "ollama"
-version = "0.6.0"
+version = "0.6.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "httpx" },
     { name = "pydantic" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/d6/47/f9ee32467fe92744474a8c72e138113f3b529fc266eea76abfdec9a33f3b/ollama-0.6.0.tar.gz", hash = "sha256:da2b2d846b5944cfbcee1ca1e6ee0585f6c9d45a2fe9467cbcd096a37383da2f", size = 50811, upload-time = "2025-09-24T22:46:02.417Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/9d/5a/652dac4b7affc2b37b95386f8ae78f22808af09d720689e3d7a86b6ed98e/ollama-0.6.1.tar.gz", hash = "sha256:478c67546836430034b415ed64fa890fd3d1ff91781a9d548b3325274e69d7c6", size = 51620, upload-time = "2025-11-13T23:02:17.416Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b5/c1/edc9f41b425ca40b26b7c104c5f6841a4537bb2552bfa6ca66e81405bb95/ollama-0.6.0-py3-none-any.whl", hash = "sha256:534511b3ccea2dff419ae06c3b58d7f217c55be7897c8ce5868dfb6b219cf7a0", size = 14130, upload-time = "2025-09-24T22:46:01.19Z" },
+    { url = "https://files.pythonhosted.org/packages/47/4f/4a617ee93d8208d2bcf26b2d8b9402ceaed03e3853c754940e2290fed063/ollama-0.6.1-py3-none-any.whl", hash = "sha256:fc4c984b345735c5486faeee67d8a265214a31cbb828167782dc642ce0a2bf8c", size = 14354, upload-time = "2025-11-13T23:02:16.292Z" },
 ]
 
 [[package]]
 name = "openai"
-version = "2.2.0"
+version = "2.8.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
@@ -2864,43 +2877,9 @@ dependencies = [
     { name = "tqdm" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/b8/b1/8201e321a7d64a25c6f5a560320272d8be70547add40311fceb916518632/openai-2.2.0.tar.gz", hash = "sha256:bc49d077a8bf0e370eec4d038bc05e232c20855a19df0b58e5b3e5a8da7d33e0", size = 588512, upload-time = "2025-10-06T18:08:13.665Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/cb/92/6aeef1836e66dfec7f7f160a4f06d7041be7f6ccfc47a2f0f5738b332245/openai-2.2.0-py3-none-any.whl", hash = "sha256:d222e63436e33f3134a3d7ce490dc2d2f146fa98036eb65cc225df3ce163916f", size = 998972, upload-time = "2025-10-06T18:08:11.775Z" },
-]
-
-[[package]]
-name = "openai-harmony"
-version = "0.0.4"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "pydantic" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/92/94/01509d510bebf6606614e51113e5a415ced15b8f34aa98a8bf2539314650/openai_harmony-0.0.4.tar.gz", hash = "sha256:5c67ac6df349236fb7b64f57c3dbb0273efcdca24314daa108f2a482c427106c", size = 279848, upload-time = "2025-08-09T01:43:24.974Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/05/88/ade63bd8f36603610040e7cc086bc134d57a99a742e05f7fcddfdf822ee1/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:3cf2344366f10981bbc0f6d9949a0b2bb87151d209ed295943ed6ad8eda37932", size = 2963206, upload-time = "2025-08-09T01:43:02.433Z" },
-    { url = "https://files.pythonhosted.org/packages/8e/ef/a65a0ff177fdf67bc0afd18bb9e7ad690d1b553a8eb5ebf27f601b22dbd0/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:2d8d16d84702059833fb03b841b28c25600c54e83cadccef79af44e1c81166b1", size = 2724854, upload-time = "2025-08-09T01:43:04.606Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/a1/ebaf0f55601a98609641283884d52dbfe9a1cf34b04f1cf80acb1560ab74/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:97f1fe3909733212cc6b36f0f199b1421a9c57b79ec665f0322bd604cec47340", size = 2984312, upload-time = "2025-08-09T01:43:08.908Z" },
-    { url = "https://files.pythonhosted.org/packages/45/24/246f6f470bfbc89a117714b68f27cdaee12b31166237a227cc657780cc1d/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:567cc568b6bf7b4d041b0c9aa7d6b2c9394f8af6065bc87fa6d23f207b5af9a7", size = 3447870, upload-time = "2025-08-09T01:43:06.734Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/ec/dcdcace0ffcf3a532cca910e0c351b62d3a7decf0b091ea8cf856d2a67a6/openai_harmony-0.0.4-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:31e9bcac0902a309e2fc688e52f247eec7fffcd00d17e958b9a83a8fea6519c2", size = 3049306, upload-time = "2025-08-09T01:43:11.019Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/39/172f1048d935db1523a82b45fee5231ad6c622645e566706e6bcf3731da8/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:96a63199c0d81095b5d5d1ae8ca82b64c1c13d18d4e30323ae9e8ab31bc80a3d", size = 3121347, upload-time = "2025-08-09T01:43:16.705Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/36/8ee4ca5d0b25587121fd3621e6a6106fba80218cb6d159e1670aeb2b22ef/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:d38f2639f6bf7c3c34a5dfd79e29075811ae2fa9b895a63e76767f74a47a971e", size = 2952326, upload-time = "2025-08-09T01:43:18.841Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/a0/ec8906393968679e269e23e957e11ff419978d1d077fb9af9561b161c988/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:038f1d6772d1be5213b36ae76e5d042022395ec35c428a73ccb8b839b2cecf6a", size = 3015832, upload-time = "2025-08-09T01:43:21.076Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/bd/aa9e6e5cf140716dbcae17402fac2a81a9ebb3f934059ac0eec61cb447fc/openai_harmony-0.0.4-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:15e6d53a66502491a3675a536df30e271f976e6c5efe68250a65191efcb85c4f", size = 3221129, upload-time = "2025-08-09T01:43:23.146Z" },
-]
-
-[[package]]
-name = "opencv-python-headless"
-version = "4.12.0.88"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/a4/63/6861102ec149c3cd298f4d1ea7ce9d6adbc7529221606ff1dab991a19adb/opencv-python-headless-4.12.0.88.tar.gz", hash = "sha256:cfdc017ddf2e59b6c2f53bc12d74b6b0be7ded4ec59083ea70763921af2b6c09", size = 95379675, upload-time = "2025-07-07T09:21:06.815Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/d5/e4/42591e356f1d53c568418dc7e30dcda7be31dd5a4d570bca22acb0525862/openai-2.8.1.tar.gz", hash = "sha256:cb1b79eef6e809f6da326a7ef6038719e35aa944c42d081807bfa1be8060f15f", size = 602490, upload-time = "2025-11-17T22:39:59.549Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/69/4e/116720df7f1f7f3b59abc608ca30fbec9d2b3ae810afe4e4d26483d9dfa0/opencv_python_headless-4.12.0.88-cp37-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:aeb4b13ecb8b4a0beb2668ea07928160ea7c2cd2d9b5ef571bbee6bafe9cc8d0", size = 33145800, upload-time = "2025-07-07T09:15:00.367Z" },
-    { url = "https://files.pythonhosted.org/packages/89/53/e19c21e0c4eb1275c3e2c97b081103b6dfb3938172264d283a519bf728b9/opencv_python_headless-4.12.0.88-cp37-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:236c8df54a90f4d02076e6f9c1cc763d794542e886c576a6fee46ec8ff75a7a9", size = 54023419, upload-time = "2025-07-07T09:15:10.164Z" },
+    { url = "https://files.pythonhosted.org/packages/55/4f/dbc0c124c40cb390508a82770fb9f6e3ed162560181a85089191a851c59a/openai-2.8.1-py3-none-any.whl", hash = "sha256:c6c3b5a04994734386e8dad3c00a393f56d3b68a27cd2e8acae91a59e4122463", size = 1022688, upload-time = "2025-11-17T22:39:57.675Z" },
 ]
 
 [[package]]
@@ -2946,9 +2925,9 @@ dependencies = [
     { name = "chex", version = "0.1.90", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
     { name = "chex", version = "0.1.91", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "jaxlib", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jaxlib", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jaxlib", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/6d/3b/90c11f740a3538200b61cd2b7d9346959cb9e31e0bdea3d2f886b7262203/optax-0.2.6.tar.gz", hash = "sha256:ba8d1e12678eba2657484d6feeca4fb281b8066bdfd5efbfc0f41b87663109c0", size = 269660, upload-time = "2025-09-15T22:41:24.76Z" }
@@ -2958,61 +2937,60 @@ wheels = [
 
 [[package]]
 name = "optree"
-version = "0.17.0"
+version = "0.18.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/56/c7/0853e0c59b135dff770615d2713b547b6b3b5cde7c10995b4a5825244612/optree-0.17.0.tar.gz", hash = "sha256:5335a5ec44479920620d72324c66563bd705ab2a698605dd4b6ee67dbcad7ecd", size = 163111, upload-time = "2025-07-25T11:26:11.586Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/52/a0/d5795ac13390b04822f1c61699f684cde682b57bf0a2d6b406019e1762ae/optree-0.17.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:85ec183b8eec6efc9a5572c2a84c62214c949555efbc69ca2381aca6048d08df", size = 622371, upload-time = "2025-07-25T11:24:23.345Z" },
-    { url = "https://files.pythonhosted.org/packages/53/8b/ae8ddb511e680eb9d61edd2f5245be88ce050456658fb165550144f9a509/optree-0.17.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:6e77b6e0b7bb3ecfeb9a92ba605ef21b39bff38829b745af993e2e2b474322e2", size = 337260, upload-time = "2025-07-25T11:24:25.291Z" },
-    { url = "https://files.pythonhosted.org/packages/91/f9/6ca076fd4c6f16be031afdc711a2676c1ff15bd1717ee2e699179b1a29bc/optree-0.17.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:98990201f352dba253af1a995c1453818db5f08de4cae7355d85aa6023676a52", size = 350398, upload-time = "2025-07-25T11:24:26.672Z" },
-    { url = "https://files.pythonhosted.org/packages/95/4c/81344cbdcf8ea8525a21c9d65892d7529010ee2146c53423b2e9a84441ba/optree-0.17.0-cp310-cp310-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:e1a40adf6bb78a6a4b4f480879de2cb6b57d46d680a4d9834aa824f41e69c0d9", size = 404834, upload-time = "2025-07-25T11:24:28.988Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/c4/ac1880372a89f5c21514a7965dfa23b1afb2ad683fb9804d366727de9ecf/optree-0.17.0-cp310-cp310-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:78a113436a0a440f900b2799584f3cc2b2eea1b245d81c3583af42ac003e333c", size = 402116, upload-time = "2025-07-25T11:24:30.396Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/72/ad6be4d6a03805cf3921b492494cb3371ca28060d5ad19d5a36e10c4d67d/optree-0.17.0-cp310-cp310-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0e45c16018f4283f028cf839b707b7ac734e8056a31b7198a1577161fcbe146d", size = 398491, upload-time = "2025-07-25T11:24:31.725Z" },
-    { url = "https://files.pythonhosted.org/packages/d9/c1/6827fb504351f9a3935699b0eb31c8a6af59d775ee78289a25e0ba54f732/optree-0.17.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b698613d821d80cc216a2444ebc3145c8bf671b55a2223058a6574c1483a65f6", size = 387957, upload-time = "2025-07-25T11:24:32.759Z" },
-    { url = "https://files.pythonhosted.org/packages/21/3d/44b3cbe4c9245a13b2677e30db2aafadf00bda976a551d64a31dc92f4977/optree-0.17.0-cp310-cp310-win32.whl", hash = "sha256:d07bfd8ce803dbc005502a89fda5f5e078e237342eaa36fb0c46cfbdf750bc76", size = 280064, upload-time = "2025-07-25T11:24:33.875Z" },
-    { url = "https://files.pythonhosted.org/packages/74/fa/83d4cd387043483ee23617b048829a1289bf54afe2f6cb98ec7b27133369/optree-0.17.0-cp310-cp310-win_amd64.whl", hash = "sha256:d009d368ef06b8757891b772cad24d4f84122bd1877f7674fb8227d6e15340b4", size = 304398, upload-time = "2025-07-25T11:24:34.844Z" },
-    { url = "https://files.pythonhosted.org/packages/21/4f/752522f318683efa7bba1895667c9841165d0284f6dfadf601769f6398ce/optree-0.17.0-cp310-cp310-win_arm64.whl", hash = "sha256:3571085ed9a5f39ff78ef57def0e9607c6b3f0099b6910524a0b42f5d58e481e", size = 308260, upload-time = "2025-07-25T11:24:36.144Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/eb/389a7dae8b113064f53909707aea9d72372fdc2eb918c48783c443cb3438/optree-0.17.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:09fbc0e5e42b20cab11851dffb7abe2fdf289c45d29e5be2b50b4ea93d069a9f", size = 640773, upload-time = "2025-07-25T11:24:37.25Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/bb/2d78b524989cabb5720e85ea366addc8589b4bbd0ce3f5ea58e370e5636a/optree-0.17.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:90a5864689268eda75d90abded5d474ae0a7ae2608d510626724fb78a1955948", size = 346402, upload-time = "2025-07-25T11:24:38.25Z" },
-    { url = "https://files.pythonhosted.org/packages/73/5c/13a2a864b0c0b39c3c193be534a195a3ab2463c7d0443d4a76e749e3ff83/optree-0.17.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3080c564c9760711aa72d1b4d700ce1417f99ad087136f415c4eb8221169e2a3", size = 362797, upload-time = "2025-07-25T11:24:39.509Z" },
-    { url = "https://files.pythonhosted.org/packages/da/f5/ff7dcb5a0108ee89c2be09aed2ebd26a7e1333d8122031aa9d9322b24ee6/optree-0.17.0-cp311-cp311-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:834a8fb358b608240b3a38706a09b43974675624485fad64c8ee641dae2eb57d", size = 419450, upload-time = "2025-07-25T11:24:40.555Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/e6/48a97aefd18770b55e5ed456d8183891f325cdb6d90592e5f072ed6951f8/optree-0.17.0-cp311-cp311-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:1a2bd263e6b5621d000d0f94de1f245414fd5dbce365a24b7b89b1ed0ef56cf9", size = 417557, upload-time = "2025-07-25T11:24:42.396Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/b1/4e280edab8a86be47ec1f9bd9ed4b685d2e15f0950ae62b613b26d12a1da/optree-0.17.0-cp311-cp311-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:9b37daca4ad89339b1f5320cc61ac600dcf976adbb060769d36d5542d6ebfedf", size = 414174, upload-time = "2025-07-25T11:24:43.51Z" },
-    { url = "https://files.pythonhosted.org/packages/db/3b/49a9a1986215dd342525974deeb17c260a83fee8fad147276fd710ac8718/optree-0.17.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a146a6917f3e28cfdc268ff1770aa696c346482dd3da681c3ff92153d94450ea", size = 402000, upload-time = "2025-07-25T11:24:44.819Z" },
-    { url = "https://files.pythonhosted.org/packages/00/8d/13b79d3394b83f4b1c93daac336f0eca5cb1cd5f58e10618f2c2db779cb7/optree-0.17.0-cp311-cp311-win32.whl", hash = "sha256:6b0446803d08f6aaae84f82f03c51527f36dfa15850873fc0183792247bc0071", size = 285777, upload-time = "2025-07-25T11:24:45.976Z" },
-    { url = "https://files.pythonhosted.org/packages/90/32/da5191a347e33a78c2804a0cbfaed8eecb758818efda4b4d70bfd9b9b38d/optree-0.17.0-cp311-cp311-win_amd64.whl", hash = "sha256:e39f4f00b2967116badd9617ad6aa9845d8327fe13b6dbf5bc36d8c7b4a5ea03", size = 313761, upload-time = "2025-07-25T11:24:47.047Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/ea/7cae17a37a8ef67a33c354fce6f136d5f253d5afa40f68701252b1b2c2a0/optree-0.17.0-cp311-cp311-win_arm64.whl", hash = "sha256:50d4dbcbca3e379cc6b374f9b5a5626ff7ea41df8373e26c3af41d89d8a4b3d5", size = 318242, upload-time = "2025-07-25T11:24:48.708Z" },
-    { url = "https://files.pythonhosted.org/packages/79/ce/471ff57336630f2434238a8cb8401e0d714ee7d54a6117823fd85de5f656/optree-0.17.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:09156e2ea62cde66dcbd9a450a5517ad6bad07d4ffc98fab0982c1e4f538341a", size = 654627, upload-time = "2025-07-25T11:24:49.754Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/ef/3143b7840dd2daedf1257643119c0f3addd23cf90cc9d2efc88f8166931e/optree-0.17.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:750f24304d1d437c8b235d4bc9e4afda17d85950706c34a875c16049f707eeb4", size = 351124, upload-time = "2025-07-25T11:24:50.813Z" },
-    { url = "https://files.pythonhosted.org/packages/41/90/e12dea2cb5d8a5e17bbe3011ed4e972b89c027272a816db4897589751cad/optree-0.17.0-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e13ae51a63d69db445f269a3a4fd1d6edb064a705188d007ea47c9f034788fc5", size = 365869, upload-time = "2025-07-25T11:24:51.807Z" },
-    { url = "https://files.pythonhosted.org/packages/76/ee/21af214663960a479863cd6c03d7a0abc8123ea22a6ea34689c2eed88ccd/optree-0.17.0-cp312-cp312-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:5958f58423cc7870cb011c8c8f92687397380886e8c9d33adac752147e7bbc3f", size = 424465, upload-time = "2025-07-25T11:24:53.124Z" },
-    { url = "https://files.pythonhosted.org/packages/54/a3/64b184a79373753f4f46a5cd301ea581f71d6dc1a5c103bd2394f0925d40/optree-0.17.0-cp312-cp312-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:970ae4e47727b4c5526fc583b87d29190e576f6a2b6c19e8671589b73d256250", size = 420686, upload-time = "2025-07-25T11:24:54.212Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/6d/b6051b0b1ef9a49df96a66e9e62fc02620d2115d1ba659888c94e67fcfc9/optree-0.17.0-cp312-cp312-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:54177fd3e6e05c08b66329e26d7d44b85f24125f25c6b74c921499a1b31b8f70", size = 421225, upload-time = "2025-07-25T11:24:55.213Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/f1/940bc959aaef9eede8bb1b1127833b0929c6ffa9268ec0f6cb19877e2027/optree-0.17.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e1959cfbc38c228c8195354967cda64887b96219924b7b3759e5ee355582c1ec", size = 408819, upload-time = "2025-07-25T11:24:56.315Z" },
-    { url = "https://files.pythonhosted.org/packages/56/52/ce527556e27dbf77266c1b1bb313ca446c94bc6edd6d7a882dbded028197/optree-0.17.0-cp312-cp312-win32.whl", hash = "sha256:039ea98c0cd94a64040d6f6d21dbe5cd9731bb380d7893f78d6898672080a232", size = 289107, upload-time = "2025-07-25T11:24:57.357Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/f1/aecb0199d269ad8ea41a86182474f98378a72681facbd6a06e94c23a2d02/optree-0.17.0-cp312-cp312-win_amd64.whl", hash = "sha256:c3a21109f635ce353d116ed1d77a7dfd77b898bcdaccef3bf74881ce7d6d54d8", size = 314074, upload-time = "2025-07-25T11:24:58.499Z" },
-    { url = "https://files.pythonhosted.org/packages/3a/20/615ad64d24318709a236163dd8620fa7879a7720bfd0c755604d3dceeb76/optree-0.17.0-cp312-cp312-win_arm64.whl", hash = "sha256:1a39f957299426d2d4aa36cbc1acd71edb198ff0f28ddb43029bf58efe34a9a1", size = 316409, upload-time = "2025-07-25T11:24:59.855Z" },
-    { url = "https://files.pythonhosted.org/packages/74/b7/6efb4955f7f08425d67b38ffd001a28896f670109bdbf1b1745f632a2306/optree-0.17.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:1535fb8725178715315af0f2862668fb49030a5737d9f6c68bcb4747b029b20b", size = 622899, upload-time = "2025-07-25T11:25:49.771Z" },
-    { url = "https://files.pythonhosted.org/packages/df/41/0d6e4d066aa25c59170055a4236d4b52abc6ac09bd0821bb85a58db7f0e0/optree-0.17.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:4aec2d138baed1357ca1ded81e40140bafbfdfd09b73d3d9d96c6c3cc527bcd9", size = 337376, upload-time = "2025-07-25T11:25:50.864Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/29/3bb53de2de3b36a51e46b6d9ada7ee1a3a312ac461cd54292a023adc807c/optree-0.17.0-cp39-cp39-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:537498cf7bf7a4fe71f7ffd815e72b8672aea0fac82e1513f6b6e35e8569f5aa", size = 350302, upload-time = "2025-07-25T11:25:52.016Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/3b/d17a31447ed7ef6f10bd0caf40742b016fcdeaa3abb7568307b04a0f50cf/optree-0.17.0-cp39-cp39-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:3b3bb2326b550ddb048e3454fad40183b7fed74dda4351b016d20362809180af", size = 405358, upload-time = "2025-07-25T11:25:53.085Z" },
-    { url = "https://files.pythonhosted.org/packages/db/f3/b9f0a8c98fd0c7f53fa9d9a46d75bb1182aeecd7ecde6f353d3e69ec9618/optree-0.17.0-cp39-cp39-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c0d3d702044e5acbec2cf8349789f6b096057bd00dc8e1e1c97b990347279fda", size = 402694, upload-time = "2025-07-25T11:25:54.537Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/dd/0d9d7426fd6b5d90ad40e4d93717a955d4257d06574dfe7a1da0d24cb06c/optree-0.17.0-cp39-cp39-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a9155e82717be1dda1f3c1244e9cb5b3733d5dd3ba47702730c7816be083a5cb", size = 398857, upload-time = "2025-07-25T11:25:55.921Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/57/dacec3f8c70f4685bb07fce19cf3361037fde2b596f6f7228e1a4b39677b/optree-0.17.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a8e825501f55360e8381718623b094579dedc485e57010e01593d72a43b43e68", size = 387849, upload-time = "2025-07-25T11:25:57.046Z" },
-    { url = "https://files.pythonhosted.org/packages/71/66/659501896bf169ad5044bec00ac9dd4fe7144bd5448c5dae06977b596166/optree-0.17.0-cp39-cp39-win32.whl", hash = "sha256:80865cf4287ed86e65af9bacd98d5395f424ffc08dc0d784590763fc1a1576b9", size = 280368, upload-time = "2025-07-25T11:25:58.209Z" },
-    { url = "https://files.pythonhosted.org/packages/56/d4/0ff0b8f762fd79249df3ed48f82eb3e6144a219c02278e39e135c582224b/optree-0.17.0-cp39-cp39-win_amd64.whl", hash = "sha256:c361ee45a97d69a427d949db5f0d6a8d9ad5f703ac7cef57a206f7f3df13d6f9", size = 310964, upload-time = "2025-07-25T11:25:59.597Z" },
-    { url = "https://files.pythonhosted.org/packages/16/62/3720d2ac320c327f304c4ed50cf6aa6059fc4f85172c69616b6ddc3b7e7d/optree-0.17.0-cp39-cp39-win_arm64.whl", hash = "sha256:4ad585248f82896ac85681b9f36b33a791d4ebf8588f3126b4dbbe5c31edbefa", size = 306151, upload-time = "2025-07-25T11:26:00.632Z" },
-    { url = "https://files.pythonhosted.org/packages/ca/52/350c58dce327257afd77b92258e43d0bfe00416fc167b0c256ec86dcf9e7/optree-0.17.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:f365328450c1072e7a707dce67eaa6db3f63671907c866e3751e317b27ea187e", size = 342845, upload-time = "2025-07-25T11:26:01.651Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/d7/3036d15c028c447b1bd65dcf8f66cfd775bfa4e52daa74b82fb1d3c88faf/optree-0.17.0-pp310-pypy310_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:adde1427e0982cfc5f56939c26b4ebbd833091a176734c79fb95c78bdf833dff", size = 350952, upload-time = "2025-07-25T11:26:02.692Z" },
-    { url = "https://files.pythonhosted.org/packages/71/45/e710024ef77324e745de48efd64f6270d8c209f14107a48ffef4049ac57a/optree-0.17.0-pp310-pypy310_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a80b7e5de5dd09b9c8b62d501e29a3850b047565c336c9d004b07ee1c01f4ae1", size = 389568, upload-time = "2025-07-25T11:26:04.094Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/63/b5cd1309f76f53e8a3cfbc88642647e58b1d3dd39f7cb0daf60ec516a252/optree-0.17.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:3c2c79652c45d82f23cbe08349456b1067ea513234a086b9a6bf1bcf128962a9", size = 306686, upload-time = "2025-07-25T11:26:05.511Z" },
-    { url = "https://files.pythonhosted.org/packages/ca/40/afec131d9dd7a18d129190d407d97c95994f42b70c3d8ab897092d4de1d9/optree-0.17.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:bd92011cd0f2de40d28a95842819e778c476ab25c12731bfef1d1a0225554f83", size = 353955, upload-time = "2025-07-25T11:26:06.75Z" },
-    { url = "https://files.pythonhosted.org/packages/69/c4/94a187ed3ca71194b9da6a276790e1703c7544c8f695ac915214ae8ce934/optree-0.17.0-pp311-pypy311_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f87f6f39015fc82d7adeee19900d246b89911319726e93cb2dbd4d1a809899bd", size = 363728, upload-time = "2025-07-25T11:26:07.959Z" },
-    { url = "https://files.pythonhosted.org/packages/cd/99/23b7a484da8dfb814107b20ef2c93ef27c04f36aeb83bd976964a5b69e06/optree-0.17.0-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:58b0a83a967d2ef0f343db7182f0ad074eb1166bcaea909ae33909462013f151", size = 404649, upload-time = "2025-07-25T11:26:09.463Z" },
-    { url = "https://files.pythonhosted.org/packages/bc/1f/7eca6da47eadb9ff2183bc9169eadde3dda0518e9a0187b99d5926fb2994/optree-0.17.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:e1ae8cbbcfaa45c57f5e51c544afa554cefbbb9fe9586c108aaf2aebfadf5899", size = 316368, upload-time = "2025-07-25T11:26:10.572Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/83/8e/09d899ad531d50b79aa24e7558f604980fe4048350172e643bb1b9983aec/optree-0.18.0.tar.gz", hash = "sha256:3804fb6ddc923855db2dc4805b4524c66e00f1ef30b166be4aadd52822b13e06", size = 165178, upload-time = "2025-11-14T08:58:31.234Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b9/45/22dfc0a3a8ca568c2702d725cdce85e2e1767d4241f2ac7d28c518ce7019/optree-0.18.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:1f19867b02a547fc9f11d27c0413e7483cef89699e16f3b9e8af73a9b25e6061", size = 353272, upload-time = "2025-11-14T08:56:31.887Z" },
+    { url = "https://files.pythonhosted.org/packages/73/ba/f89cd7fe3fca9f434b6cd6ba80716f1ad59f40453482c04d6887e5c516de/optree-0.18.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:ad428ccdb2a40804919880dfe8d2a3021fd4418be15ea7ecb8434ab249badf9f", size = 330497, upload-time = "2025-11-14T08:56:33.447Z" },
+    { url = "https://files.pythonhosted.org/packages/73/93/463a531b863bffae92d6d1c7857f655234f12ad46fe088bf5bd5cd37cd67/optree-0.18.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e058cc51d9d57b45801060af9f74765b95bedfc59fd6df1c7489ae0825126be5", size = 349724, upload-time = "2025-11-14T08:56:34.475Z" },
+    { url = "https://files.pythonhosted.org/packages/b9/4f/7d54f0eeea24f5893422d65fce835d80e644ebac8a4570d762f994cfe97c/optree-0.18.0-cp310-cp310-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:daab231cf768937ce4675376ea3e214d399116d9867a6737372c31c58630bdfc", size = 404190, upload-time = "2025-11-14T08:56:35.622Z" },
+    { url = "https://files.pythonhosted.org/packages/66/af/044080368dc8ab809c5b089ad3132e7775043c5a8a165465df1f0c99dd62/optree-0.18.0-cp310-cp310-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ea357657143f364a764b63b2b1ce12d77156d48a1f32def990b696d755acb629", size = 401798, upload-time = "2025-11-14T08:56:37.001Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/97/c449712ccb50af3cb2608718e503e31b259aa55de40ee83bebb159b07593/optree-0.18.0-cp310-cp310-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f81f5340c8df50662abaf753ab07095901e40b934efb27da50032a4ae71c5a97", size = 397477, upload-time = "2025-11-14T08:56:38.298Z" },
+    { url = "https://files.pythonhosted.org/packages/66/8e/bdda357f31e2b7c80b61b9785e6dea25e94c01e7237c7a8b1af38f369ee4/optree-0.18.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:66f142c743732cd4e630ea84415f654a00c792793c7f80d4511167f0f89796a6", size = 386631, upload-time = "2025-11-14T08:56:39.707Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/2f/32575d5ac11993c19a6b60f6e15cafcc807adabf92caf06824bc3b5d50f9/optree-0.18.0-cp310-cp310-manylinux_2_39_riscv64.whl", hash = "sha256:55a2ccd121fccc9df961e982db2f4e8f2b4f7015e814ef70b1140514cdffe214", size = 347972, upload-time = "2025-11-14T08:56:40.656Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/7d/72bc9147ed59a6a1cab4e0c43a27d6542029a66c46c5301d05f990dd1991/optree-0.18.0-cp310-cp310-win32.whl", hash = "sha256:090a3f0ccafa0fe99d71e7d974ae52ff966ac26c409ec41f96556b96646054ef", size = 277698, upload-time = "2025-11-14T08:56:41.801Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/53/cf4e66ac4f134c2211a6f3f04e9d3a98effb6229316bcdadc3d3a9247362/optree-0.18.0-cp310-cp310-win_amd64.whl", hash = "sha256:0e0dbe995241efe70cfb522e89c1a7c968216926725a0e5e20cc72bd5d0311b1", size = 302673, upload-time = "2025-11-14T08:56:43.132Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/b2/66c12a8707722bad5f25cd917d528796544fc9045c5933532f5db071cb02/optree-0.18.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:80f28e4666aad66e5e20bdc2c47b5bf320250bb5407b3a39dfb1772787a7068f", size = 363110, upload-time = "2025-11-14T08:56:44.782Z" },
+    { url = "https://files.pythonhosted.org/packages/d4/f6/5377f265a8dcd61edabf8b87b657d78fca9051eeaf311ed77f73b43526a9/optree-0.18.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:72fa79be4d6515682417f103ae759a22345439eb1319886be936029215ee00dc", size = 337284, upload-time = "2025-11-14T08:56:46.039Z" },
+    { url = "https://files.pythonhosted.org/packages/5e/d7/3045564c1183c7b7cfb32d11a6250fbe9f904f723c02c80a91f71c150b78/optree-0.18.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4cc92339899acb685ee718fd22b25069dfa7be038c63274c54481d54ccc2f9e2", size = 361562, upload-time = "2025-11-14T08:56:47.162Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/32/5a7b5a441d82b44c6d2b31ad14f7b1a4d3be7afcc38437d07762212bc9c6/optree-0.18.0-cp311-cp311-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:1545c68299c0ce600e4ea1bc9112765dc4afe9a0b8ab43f955df6566bf78db42", size = 420660, upload-time = "2025-11-14T08:56:48.478Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/06/fc7aea4d6c72c4e0f42c157183b3e91c615fdc15da5a5e4e5f8b596a24f3/optree-0.18.0-cp311-cp311-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8a901666afc2d7a8d0c20decc8079763e3313457ee67210382162d90163c0007", size = 417586, upload-time = "2025-11-14T08:56:49.51Z" },
+    { url = "https://files.pythonhosted.org/packages/95/cc/bb0607eb8d20cf80ea6b122c059954fb525bbbb7150d650fd87696e4d141/optree-0.18.0-cp311-cp311-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:efd162e3bfc7812d75ebf2d0fb2783daee2407a92155af8a90650a6b0fa9342e", size = 413013, upload-time = "2025-11-14T08:56:50.842Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/6f/7f2238ec5e9d33e56252c30880bb8f44aec1415474b62b9e33b38594953d/optree-0.18.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b0986ff1267a3b44d3ed76c3efb8b7239371444143f6e0d79f9dd23dbe02c7f9", size = 400775, upload-time = "2025-11-14T08:56:52.249Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/42/f17e2977ecacdbca50c888731c4f0488f4d499fca1c48c3063bff0d5303b/optree-0.18.0-cp311-cp311-manylinux_2_39_riscv64.whl", hash = "sha256:382e5ca02cbd5b20d713d4da189a8613f828832e2af57ccbe04a9c6b0bd9497e", size = 359998, upload-time = "2025-11-14T08:56:53.321Z" },
+    { url = "https://files.pythonhosted.org/packages/73/f2/f5c5758e5db9d1b52e7b7809dcf876f071f64ec03bbd6007eee633bf0027/optree-0.18.0-cp311-cp311-win32.whl", hash = "sha256:056894ce6242cd1c7fed71325a7d9f633b2d3b4420c52af48f6a0c4560d74ca1", size = 283573, upload-time = "2025-11-14T08:56:54.426Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/a3/91942b7e6e365f4e05d196dbbb52909aae11f1e2f4b4c8aee5b506f93877/optree-0.18.0-cp311-cp311-win_amd64.whl", hash = "sha256:86f5bf05ad236f666e5395e989d6ac2cbfd02556526703e6c6f0a594c7fa081f", size = 312223, upload-time = "2025-11-14T08:56:55.812Z" },
+    { url = "https://files.pythonhosted.org/packages/64/af/a98364b1b27516942db07d9841d8ac2d7ba96039bcd8ed496f7b3f297dc4/optree-0.18.0-cp311-cp311-win_arm64.whl", hash = "sha256:9b1e7e8f9ddc85f05d542b74157bdb73ed0e49aded67d1775f721fcd6eb9be94", size = 317247, upload-time = "2025-11-14T08:56:57.795Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/9d/e32dba0876d6514b40cd30e57938e3605b7e07d827ac617f072ff64d9cff/optree-0.18.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:f02faeda66d531dc5f5356589afcf2a6bc41c8d00bc903efab60f9a2182b140d", size = 369756, upload-time = "2025-11-14T08:56:58.793Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/60/e643c3ab2cea904d24e56ab0060c905443ceac716e542392ad8db87f09ef/optree-0.18.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:e28024e6e343353285cf99ae9c74210f0e89e47b2f0f3af7c72c4a9e89dc3ebc", size = 342794, upload-time = "2025-11-14T08:56:59.819Z" },
+    { url = "https://files.pythonhosted.org/packages/79/af/c002cc41f6eba05fddcd6435e7c4ccc037a39a345de824d1a515386b52ee/optree-0.18.0-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:057b983a9526645133553184bed7090bb07855df986abd9e99c456922045c6bc", size = 364436, upload-time = "2025-11-14T08:57:01.17Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/32/2049f9597ae75a6a1b9c872ffcc31d96ebebf62ff1b20160eb972aea5456/optree-0.18.0-cp312-cp312-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:8a2003fab79694e04b5f260628511e441c248b46a9fc46138e2424038ac04ada", size = 425359, upload-time = "2025-11-14T08:57:02.509Z" },
+    { url = "https://files.pythonhosted.org/packages/d0/14/523cfb3a139df0d2a064b38497c7ee6e49edac6e07c777d2069cffd8545c/optree-0.18.0-cp312-cp312-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:02d9999840fabef85a6b22e757f336d5591f712f99c710d8b232d52e53115314", size = 421373, upload-time = "2025-11-14T08:57:03.953Z" },
+    { url = "https://files.pythonhosted.org/packages/65/48/febb5976fa4f37a805809757b050d95effb25a17d3fec439349d18f5d451/optree-0.18.0-cp312-cp312-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:79bbe14d6cad81f5840958589daa1b836864ada40031712a446dce8129917efd", size = 420798, upload-time = "2025-11-14T08:57:05.02Z" },
+    { url = "https://files.pythonhosted.org/packages/02/9c/c38b861e504829e0dcbed2e2b5bc865a98fb7c655e2ddd2bf8e1fd6ae712/optree-0.18.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a74c45f04def041504bd21682eaf7f359f1a50dc7cf42b548b6f19aab50596bd", size = 408195, upload-time = "2025-11-14T08:57:05.983Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/71/ff4b053ad1242f3d0b8792caa786c1c0138c1fb3d0c0a3720ccc21725739/optree-0.18.0-cp312-cp312-manylinux_2_39_riscv64.whl", hash = "sha256:d2844478690b5892159df0b2500e9d146dc8d3aa5b44e4564d05787b7330eca3", size = 364997, upload-time = "2025-11-14T08:57:07.306Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/09/a71dbc096e8195d0e0fe990b394884d061938e0b918b61d16d48d817c0d5/optree-0.18.0-cp312-cp312-win32.whl", hash = "sha256:cfa2e16993ba47e671a4e7ee1ad805f67b8d6744eb30a9d27ea0b07b3b7a22ed", size = 286802, upload-time = "2025-11-14T08:57:08.674Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/95/a487d5c93dcb50d768c4cd7d17476b4ce4360c5943bb7251d1e26d38e5cf/optree-0.18.0-cp312-cp312-win_amd64.whl", hash = "sha256:89e81afb11792d13d3777b503c6f21ec17b1a3b7de69cde1ae2c5471bcdcd4a0", size = 312381, upload-time = "2025-11-14T08:57:09.98Z" },
+    { url = "https://files.pythonhosted.org/packages/91/ba/0559af098b33103bc65ba9d0c38454e21c142279b913c0b87a89c59c520d/optree-0.18.0-cp312-cp312-win_arm64.whl", hash = "sha256:4eb146711d4cd0876bf93e0118d3e74050b6f633d756c269ce7cda907281b499", size = 315747, upload-time = "2025-11-14T08:57:11.033Z" },
+    { url = "https://files.pythonhosted.org/packages/83/63/58efd0916d8ccddfb5f786aa67c68937512e11d066eb424a5a9451e72fb3/optree-0.18.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:f04286908654ffb05455254ebf72fe69473fc4560fc7ea49410df94dea6783a2", size = 353280, upload-time = "2025-11-14T08:58:14.205Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/53/84bcac1e05ea49a5bfa3d52712fb788b507c2248a9d4266939d4c1d0188a/optree-0.18.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:421b839c7ff30df5791e66c89b2e9c2f68191dd6a5d6927c32bcc6b887090df8", size = 330590, upload-time = "2025-11-14T08:58:15.283Z" },
+    { url = "https://files.pythonhosted.org/packages/79/1c/8782c71b73db1d217b18f62cda888f6f976662ed630caa61fe9dd6fa2e8a/optree-0.18.0-cp39-cp39-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d20765efa494a80a8fd91c4de8890f34de8e9f234da5516e8f34f55703cfb93d", size = 349539, upload-time = "2025-11-14T08:58:16.359Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/26/1c3b2a77a0dbd58d9488ddabd66bcf9745e9eb64922433ed42f0224752e8/optree-0.18.0-cp39-cp39-manylinux_2_26_i686.manylinux_2_28_i686.whl", hash = "sha256:30f95279188f6b9300e17c1557989baa991c2d6f519013bd8fea13462a0e6a45", size = 404850, upload-time = "2025-11-14T08:58:17.409Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/4e/2b05367bc3a86179772bd1545647521254d00a10b2b546c9abea54ad646e/optree-0.18.0-cp39-cp39-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:bda4572392ac1dff3fc67b6d9a4b1084e1637972e8135ad3788b4ce7cf0a90f5", size = 402379, upload-time = "2025-11-14T08:58:18.508Z" },
+    { url = "https://files.pythonhosted.org/packages/da/d4/04c5c3a978db8e8ab2f75c27cfdce69dd8098a4f19daf0de2288ba11745f/optree-0.18.0-cp39-cp39-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:b45d7172c67fc8d2b69f77b384998b39793ee91f8b3b46c609297b781fb7eea5", size = 397642, upload-time = "2025-11-14T08:58:19.755Z" },
+    { url = "https://files.pythonhosted.org/packages/25/f0/3d322678cf6e691e8c044153506b699c6886df4f13adb4d885cc8922cdf4/optree-0.18.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:56bb19ff827c9a443202b52bf103705ce96ef14d045e0a30d0d7ee7dbcef6a0d", size = 386870, upload-time = "2025-11-14T08:58:20.79Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/94/da38e68e20fb34737d83808b2ad05422496f05ab98c550af6cebabdaadc4/optree-0.18.0-cp39-cp39-manylinux_2_39_riscv64.whl", hash = "sha256:b7aa0de08bbbfcef6e49c107f9f397f5d4742548500f16e3e6c5e0b9e4ff0faa", size = 347826, upload-time = "2025-11-14T08:58:22.008Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/fa/07a4b30a5e8b268c2f0e083812c568e3db9abf4392ea2d1611716737474a/optree-0.18.0-cp39-cp39-win32.whl", hash = "sha256:31539dec60af84e16e99574634811d38e34e1fb381f40d6f489a2e582bf41f03", size = 277949, upload-time = "2025-11-14T08:58:23.162Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/81/d419aa8ad240878e934e88546b51e0764faf0d68caf055fb3c53f90c2e16/optree-0.18.0-cp39-cp39-win_amd64.whl", hash = "sha256:7699957183f8d45402edd6266e175510317f5fcd7f0e623510f2eb7e1ebfc667", size = 309380, upload-time = "2025-11-14T08:58:24.229Z" },
+    { url = "https://files.pythonhosted.org/packages/79/5d/3c654144031d4dc28442bca8343bc02d177ab5da4c38ed521cad59e5d4b0/optree-0.18.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:30fefc84975ac41d9075993196c64ce0c240510f0539cff121d63b709e03846f", size = 371229, upload-time = "2025-11-14T08:58:25.702Z" },
+    { url = "https://files.pythonhosted.org/packages/d2/da/4ddcf4fbb75c5779f58b1f85657209f5e38eb34d41c038502e1d0bdae68d/optree-0.18.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:ba23caafd0e0c911bb7eab54e5cf69644af864d153e4b2abdab83ff0ef357ba1", size = 346077, upload-time = "2025-11-14T08:58:26.762Z" },
+    { url = "https://files.pythonhosted.org/packages/50/56/55e565d593d91a47ee70b91b693a763362aafac3bde06288036309c87949/optree-0.18.0-pp311-pypy311_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:10f29662d637b80363dc620da46ddc58def7acf7935e20595b23e216ea912367", size = 363316, upload-time = "2025-11-14T08:58:27.888Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/c9/19d5a8d99934f80bfa02d503587a5697a374324f95ccc4c1656a8a74fc3b/optree-0.18.0-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ff7326f36ed70d84c3fd62fb39bc6858f699640b8ab238c3cb8dafe1e200af59", size = 403560, upload-time = "2025-11-14T08:58:28.994Z" },
+    { url = "https://files.pythonhosted.org/packages/13/61/016ff1dcf63b97bfd182af8705f156e128a1e3adfcd94a2283fe04cf95d7/optree-0.18.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:01b79aaee544adf5bfa573db32b943030dfeb9fd1c6e7a97aa417db56a8127e7", size = 314844, upload-time = "2025-11-14T08:58:30.146Z" },
 ]
 
 [[package]]
@@ -3025,7 +3003,7 @@ resolution-markers = [
 dependencies = [
     { name = "absl-py", marker = "python_full_version < '3.10'" },
     { name = "etils", version = "1.5.2", source = { registry = "https://pypi.org/simple" }, extra = ["epath", "epy"], marker = "python_full_version < '3.10'" },
-    { name = "humanize", marker = "python_full_version < '3.10'" },
+    { name = "humanize", version = "4.13.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jaxlib", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "msgpack", marker = "python_full_version < '3.10'" },
@@ -3043,7 +3021,7 @@ wheels = [
 
 [[package]]
 name = "orbax-checkpoint"
-version = "0.11.25"
+version = "0.11.28"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -3054,21 +3032,23 @@ dependencies = [
     { name = "absl-py", marker = "python_full_version >= '3.10'" },
     { name = "aiofiles", marker = "python_full_version >= '3.10'" },
     { name = "etils", version = "1.13.0", source = { registry = "https://pypi.org/simple" }, extra = ["epath", "epy"], marker = "python_full_version >= '3.10'" },
-    { name = "humanize", marker = "python_full_version >= '3.10'" },
+    { name = "humanize", version = "4.14.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "msgpack", marker = "python_full_version >= '3.10'" },
     { name = "nest-asyncio", marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "protobuf", marker = "python_full_version >= '3.10'" },
+    { name = "psutil", marker = "python_full_version >= '3.10'" },
     { name = "pyyaml", marker = "python_full_version >= '3.10'" },
     { name = "simplejson", marker = "python_full_version >= '3.10'" },
-    { name = "tensorstore", version = "0.1.78", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "tensorstore", version = "0.1.78", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "tensorstore", version = "0.1.79", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/00/44/63d4a064fddc9f75cf622427f01333ed109096b6af8db039c00c12b2a85b/orbax_checkpoint-0.11.25.tar.gz", hash = "sha256:15d35e4ee165aad8edccc25bc462a71c921bdaecc79ce0a2a5a0f784fce091be", size = 389724, upload-time = "2025-09-11T17:40:50.771Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/c1/b7/cc0826b55de34f664486b6520bb23e473927676fc55cc53195b5230ec65d/orbax_checkpoint-0.11.28.tar.gz", hash = "sha256:a72075411c8aa4b48ddd9f2b247348bdfcf7b72f633b4d9c99541648089e90e9", size = 393676, upload-time = "2025-11-06T20:43:10.046Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/1b/96/510dff16d2b75d9debb77fcfc4e6a4749059024a290953032f3ac793f74a/orbax_checkpoint-0.11.25-py3-none-any.whl", hash = "sha256:af8f5d16f1185c54ead7859309c57de28bf3cc9b4928ebd5a526e3ba549256c4", size = 563103, upload-time = "2025-09-11T17:40:49.368Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/f0/23168b72535c846268761b498fd61a5166b4c730a7820c7dcb13efdae05d/orbax_checkpoint-0.11.28-py3-none-any.whl", hash = "sha256:094273495fca331da88ec845053a96a7f0152b0223c6b51b82a6c1b32f373c3b", size = 586314, upload-time = "2025-11-06T20:43:08.302Z" },
 ]
 
 [[package]]
@@ -3082,7 +3062,8 @@ dependencies = [
     { name = "jsonpath-ng" },
     { name = "jsonschema" },
     { name = "outlines-core" },
-    { name = "pillow" },
+    { name = "pillow", version = "11.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pillow", version = "12.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pydantic" },
     { name = "typing-extensions" },
 ]
@@ -3101,23 +3082,26 @@ dottxt = [
     { name = "dottxt" },
 ]
 gemini = [
-    { name = "google-genai" },
+    { name = "google-genai", version = "1.47.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "google-genai", version = "1.52.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 llamacpp = [
     { name = "huggingface-hub" },
     { name = "llama-cpp-python" },
     { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "numba", version = "0.62.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 llguidance = [
-    { name = "llguidance" },
+    { name = "llguidance", version = "1.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "llguidance", version = "1.4.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 mistral = [
     { name = "mistralai" },
 ]
 mlxlm = [
     { name = "datasets" },
-    { name = "mlx" },
+    { name = "mlx", version = "0.29.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "mlx", version = "0.30.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "mlx-lm" },
 ]
 ollama = [
@@ -3130,38 +3114,46 @@ sglang = [
     { name = "openai" },
 ]
 test = [
-    { name = "accelerate" },
+    { name = "accelerate", version = "1.10.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "accelerate", version = "1.12.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "airportsdata" },
     { name = "anthropic" },
     { name = "beartype" },
-    { name = "coverage", extra = ["toml"] },
+    { name = "coverage", version = "7.10.7", source = { registry = "https://pypi.org/simple" }, extra = ["toml"], marker = "python_full_version < '3.10'" },
+    { name = "coverage", version = "7.12.0", source = { registry = "https://pypi.org/simple" }, extra = ["toml"], marker = "python_full_version >= '3.10'" },
     { name = "datasets" },
     { name = "diff-cover" },
     { name = "dottxt" },
     { name = "flax", version = "0.8.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "flax", version = "0.10.7", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "flax", version = "0.12.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "google-genai" },
+    { name = "flax", version = "0.12.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "google-genai", version = "1.47.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "google-genai", version = "1.52.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "huggingface-hub" },
     { name = "iso3166" },
     { name = "jax", version = "0.4.30", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "jax", version = "0.6.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "jax", version = "0.7.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "jax", version = "0.8.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
     { name = "llama-cpp-python" },
-    { name = "llguidance" },
+    { name = "llguidance", version = "1.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "llguidance", version = "1.4.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "mistralai" },
     { name = "mkdocs-gen-files" },
     { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
     { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "numba", version = "0.62.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "ollama" },
     { name = "openai" },
-    { name = "pillow" },
-    { name = "pre-commit" },
-    { name = "pytest" },
-    { name = "pytest-asyncio" },
+    { name = "pillow", version = "11.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pillow", version = "12.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pre-commit", version = "4.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pre-commit", version = "4.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pytest", version = "8.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pytest", version = "9.0.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "pytest-asyncio", version = "1.2.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pytest-asyncio", version = "1.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pytest-benchmark" },
     { name = "pytest-cov" },
     { name = "pytest-mock" },
@@ -3170,7 +3162,8 @@ test = [
     { name = "sentencepiece" },
     { name = "tensorflow" },
     { name = "tf-keras" },
-    { name = "torch" },
+    { name = "torch", version = "2.8.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "torch", version = "2.9.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "transformers" },
     { name = "xgrammar" },
 ]
@@ -3178,7 +3171,8 @@ tgi = [
     { name = "huggingface-hub" },
 ]
 transformers = [
-    { name = "accelerate" },
+    { name = "accelerate", version = "1.10.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "accelerate", version = "1.12.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "datasets" },
     { name = "setuptools" },
     { name = "transformers" },
@@ -3369,15 +3363,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/98/af/7be05277859a7bc399da8ba68b88c96b27b48740b6cf49688899c6eb4176/pandas-2.3.3-cp39-cp39-win_amd64.whl", hash = "sha256:d3e28b3e83862ccf4d85ff19cf8c20b2ae7e503881711ff2d534dc8f761131aa", size = 11359119, upload-time = "2025-09-29T23:34:46.339Z" },
 ]
 
-[[package]]
-name = "partial-json-parser"
-version = "0.2.1.1.post6"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/86/13/459e86c9c67a006651803a3df3d0b08f7708bc5483fdc482582d75562949/partial_json_parser-0.2.1.1.post6.tar.gz", hash = "sha256:43896b68929678224cbbe4884a6a5fe9251ded4b30b8b7d7eb569e5feea93afc", size = 10299, upload-time = "2025-06-23T17:51:45.372Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/cb/40/1f922794af3dc7503f19319a8804b398a161a2cd54183cff8b12225b8d85/partial_json_parser-0.2.1.1.post6-py3-none-any.whl", hash = "sha256:abc332f09b13ef5233384dbfe7128a0e9ea3fa4b8f8be9b37ac1b433c810e99e", size = 10876, upload-time = "2025-06-23T17:51:44.332Z" },
-]
-
 [[package]]
 name = "pathspec"
 version = "0.12.1"
@@ -3391,6 +3376,9 @@ wheels = [
 name = "pillow"
 version = "11.3.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/f3/0d/d0d6dea55cd152ce3d6767bb38a8fc10e33796ba4ba210cbab9354b6d238/pillow-11.3.0.tar.gz", hash = "sha256:3828ee7586cd0b2091b6209e5ad53e20d0649bbe87164a459d0676e035e8f523", size = 47113069, upload-time = "2025-07-01T09:16:30.666Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/4c/5d/45a3553a253ac8763f3561371432a90bdbe6000fbdcf1397ffe502aa206c/pillow-11.3.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:1b9c17fd4ace828b3003dfd1e30bff24863e0eb59b535e8f80194d9cc7ecf860", size = 5316554, upload-time = "2025-07-01T09:13:39.342Z" },
@@ -3453,15 +3441,85 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/34/e7/ae39f538fd6844e982063c3a5e4598b8ced43b9633baa3a85ef33af8c05c/pillow-11.3.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:c84d689db21a1c397d001aa08241044aa2069e7587b398c8cc63020390b1c1b8", size = 6984598, upload-time = "2025-07-01T09:16:27.732Z" },
 ]
 
+[[package]]
+name = "pillow"
+version = "12.0.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/5a/b0/cace85a1b0c9775a9f8f5d5423c8261c858760e2466c79b2dd184638b056/pillow-12.0.0.tar.gz", hash = "sha256:87d4f8125c9988bfbed67af47dd7a953e2fc7b0cc1e7800ec6d2080d490bb353", size = 47008828, upload-time = "2025-10-15T18:24:14.008Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/5d/08/26e68b6b5da219c2a2cb7b563af008b53bb8e6b6fcb3fa40715fcdb2523a/pillow-12.0.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:3adfb466bbc544b926d50fe8f4a4e6abd8c6bffd28a26177594e6e9b2b76572b", size = 5289809, upload-time = "2025-10-15T18:21:27.791Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/e9/4e58fb097fb74c7b4758a680aacd558810a417d1edaa7000142976ef9d2f/pillow-12.0.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:1ac11e8ea4f611c3c0147424eae514028b5e9077dd99ab91e1bd7bc33ff145e1", size = 4650606, upload-time = "2025-10-15T18:21:29.823Z" },
+    { url = "https://files.pythonhosted.org/packages/4b/e0/1fa492aa9f77b3bc6d471c468e62bfea1823056bf7e5e4f1914d7ab2565e/pillow-12.0.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d49e2314c373f4c2b39446fb1a45ed333c850e09d0c59ac79b72eb3b95397363", size = 6221023, upload-time = "2025-10-15T18:21:31.415Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/09/4de7cd03e33734ccd0c876f0251401f1314e819cbfd89a0fcb6e77927cc6/pillow-12.0.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:c7b2a63fd6d5246349f3d3f37b14430d73ee7e8173154461785e43036ffa96ca", size = 8024937, upload-time = "2025-10-15T18:21:33.453Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/69/0688e7c1390666592876d9d474f5e135abb4acb39dcb583c4dc5490f1aff/pillow-12.0.0-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d64317d2587c70324b79861babb9c09f71fbb780bad212018874b2c013d8600e", size = 6334139, upload-time = "2025-10-15T18:21:35.395Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/1c/880921e98f525b9b44ce747ad1ea8f73fd7e992bafe3ca5e5644bf433dea/pillow-12.0.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d77153e14b709fd8b8af6f66a3afbb9ed6e9fc5ccf0b6b7e1ced7b036a228782", size = 7026074, upload-time = "2025-10-15T18:21:37.219Z" },
+    { url = "https://files.pythonhosted.org/packages/28/03/96f718331b19b355610ef4ebdbbde3557c726513030665071fd025745671/pillow-12.0.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:32ed80ea8a90ee3e6fa08c21e2e091bba6eda8eccc83dbc34c95169507a91f10", size = 6448852, upload-time = "2025-10-15T18:21:39.168Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/a0/6a193b3f0cc9437b122978d2c5cbce59510ccf9a5b48825096ed7472da2f/pillow-12.0.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:c828a1ae702fc712978bda0320ba1b9893d99be0badf2647f693cc01cf0f04fa", size = 7117058, upload-time = "2025-10-15T18:21:40.997Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/c4/043192375eaa4463254e8e61f0e2ec9a846b983929a8d0a7122e0a6d6fff/pillow-12.0.0-cp310-cp310-win32.whl", hash = "sha256:bd87e140e45399c818fac4247880b9ce719e4783d767e030a883a970be632275", size = 6295431, upload-time = "2025-10-15T18:21:42.518Z" },
+    { url = "https://files.pythonhosted.org/packages/92/c6/c2f2fc7e56301c21827e689bb8b0b465f1b52878b57471a070678c0c33cd/pillow-12.0.0-cp310-cp310-win_amd64.whl", hash = "sha256:455247ac8a4cfb7b9bc45b7e432d10421aea9fc2e74d285ba4072688a74c2e9d", size = 7000412, upload-time = "2025-10-15T18:21:44.404Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/d2/5f675067ba82da7a1c238a73b32e3fd78d67f9d9f80fbadd33a40b9c0481/pillow-12.0.0-cp310-cp310-win_arm64.whl", hash = "sha256:6ace95230bfb7cd79ef66caa064bbe2f2a1e63d93471c3a2e1f1348d9f22d6b7", size = 2435903, upload-time = "2025-10-15T18:21:46.29Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/5a/a2f6773b64edb921a756eb0729068acad9fc5208a53f4a349396e9436721/pillow-12.0.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:0fd00cac9c03256c8b2ff58f162ebcd2587ad3e1f2e397eab718c47e24d231cc", size = 5289798, upload-time = "2025-10-15T18:21:47.763Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/05/069b1f8a2e4b5a37493da6c5868531c3f77b85e716ad7a590ef87d58730d/pillow-12.0.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:a3475b96f5908b3b16c47533daaa87380c491357d197564e0ba34ae75c0f3257", size = 4650589, upload-time = "2025-10-15T18:21:49.515Z" },
+    { url = "https://files.pythonhosted.org/packages/61/e3/2c820d6e9a36432503ead175ae294f96861b07600a7156154a086ba7111a/pillow-12.0.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:110486b79f2d112cf6add83b28b627e369219388f64ef2f960fef9ebaf54c642", size = 6230472, upload-time = "2025-10-15T18:21:51.052Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/89/63427f51c64209c5e23d4d52071c8d0f21024d3a8a487737caaf614a5795/pillow-12.0.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5269cc1caeedb67e6f7269a42014f381f45e2e7cd42d834ede3c703a1d915fe3", size = 8033887, upload-time = "2025-10-15T18:21:52.604Z" },
+    { url = "https://files.pythonhosted.org/packages/f6/1b/c9711318d4901093c15840f268ad649459cd81984c9ec9887756cca049a5/pillow-12.0.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:aa5129de4e174daccbc59d0a3b6d20eaf24417d59851c07ebb37aeb02947987c", size = 6343964, upload-time = "2025-10-15T18:21:54.619Z" },
+    { url = "https://files.pythonhosted.org/packages/41/1e/db9470f2d030b4995083044cd8738cdd1bf773106819f6d8ba12597d5352/pillow-12.0.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:bee2a6db3a7242ea309aa7ee8e2780726fed67ff4e5b40169f2c940e7eb09227", size = 7034756, upload-time = "2025-10-15T18:21:56.151Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/b0/6177a8bdd5ee4ed87cba2de5a3cc1db55ffbbec6176784ce5bb75aa96798/pillow-12.0.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:90387104ee8400a7b4598253b4c406f8958f59fcf983a6cea2b50d59f7d63d0b", size = 6458075, upload-time = "2025-10-15T18:21:57.759Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/5e/61537aa6fa977922c6a03253a0e727e6e4a72381a80d63ad8eec350684f2/pillow-12.0.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:bc91a56697869546d1b8f0a3ff35224557ae7f881050e99f615e0119bf934b4e", size = 7125955, upload-time = "2025-10-15T18:21:59.372Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/3d/d5033539344ee3cbd9a4d69e12e63ca3a44a739eb2d4c8da350a3d38edd7/pillow-12.0.0-cp311-cp311-win32.whl", hash = "sha256:27f95b12453d165099c84f8a8bfdfd46b9e4bda9e0e4b65f0635430027f55739", size = 6298440, upload-time = "2025-10-15T18:22:00.982Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/42/aaca386de5cc8bd8a0254516957c1f265e3521c91515b16e286c662854c4/pillow-12.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:b583dc9070312190192631373c6c8ed277254aa6e6084b74bdd0a6d3b221608e", size = 6999256, upload-time = "2025-10-15T18:22:02.617Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/f1/9197c9c2d5708b785f631a6dfbfa8eb3fb9672837cb92ae9af812c13b4ed/pillow-12.0.0-cp311-cp311-win_arm64.whl", hash = "sha256:759de84a33be3b178a64c8ba28ad5c135900359e85fb662bc6e403ad4407791d", size = 2436025, upload-time = "2025-10-15T18:22:04.598Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/90/4fcce2c22caf044e660a198d740e7fbc14395619e3cb1abad12192c0826c/pillow-12.0.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:53561a4ddc36facb432fae7a9d8afbfaf94795414f5cdc5fc52f28c1dca90371", size = 5249377, upload-time = "2025-10-15T18:22:05.993Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/e0/ed960067543d080691d47d6938ebccbf3976a931c9567ab2fbfab983a5dd/pillow-12.0.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:71db6b4c1653045dacc1585c1b0d184004f0d7e694c7b34ac165ca70c0838082", size = 4650343, upload-time = "2025-10-15T18:22:07.718Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/a1/f81fdeddcb99c044bf7d6faa47e12850f13cee0849537a7d27eeab5534d4/pillow-12.0.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:2fa5f0b6716fc88f11380b88b31fe591a06c6315e955c096c35715788b339e3f", size = 6232981, upload-time = "2025-10-15T18:22:09.287Z" },
+    { url = "https://files.pythonhosted.org/packages/88/e1/9098d3ce341a8750b55b0e00c03f1630d6178f38ac191c81c97a3b047b44/pillow-12.0.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:82240051c6ca513c616f7f9da06e871f61bfd7805f566275841af15015b8f98d", size = 8041399, upload-time = "2025-10-15T18:22:10.872Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/62/a22e8d3b602ae8cc01446d0c57a54e982737f44b6f2e1e019a925143771d/pillow-12.0.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:55f818bd74fe2f11d4d7cbc65880a843c4075e0ac7226bc1a23261dbea531953", size = 6347740, upload-time = "2025-10-15T18:22:12.769Z" },
+    { url = "https://files.pythonhosted.org/packages/4f/87/424511bdcd02c8d7acf9f65caa09f291a519b16bd83c3fb3374b3d4ae951/pillow-12.0.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b87843e225e74576437fd5b6a4c2205d422754f84a06942cfaf1dc32243e45a8", size = 7040201, upload-time = "2025-10-15T18:22:14.813Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/4d/435c8ac688c54d11755aedfdd9f29c9eeddf68d150fe42d1d3dbd2365149/pillow-12.0.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:c607c90ba67533e1b2355b821fef6764d1dd2cbe26b8c1005ae84f7aea25ff79", size = 6462334, upload-time = "2025-10-15T18:22:16.375Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/f2/ad34167a8059a59b8ad10bc5c72d4d9b35acc6b7c0877af8ac885b5f2044/pillow-12.0.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:21f241bdd5080a15bc86d3466a9f6074a9c2c2b314100dd896ac81ee6db2f1ba", size = 7134162, upload-time = "2025-10-15T18:22:17.996Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/b1/a7391df6adacf0a5c2cf6ac1cf1fcc1369e7d439d28f637a847f8803beb3/pillow-12.0.0-cp312-cp312-win32.whl", hash = "sha256:dd333073e0cacdc3089525c7df7d39b211bcdf31fc2824e49d01c6b6187b07d0", size = 6298769, upload-time = "2025-10-15T18:22:19.923Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/0b/d87733741526541c909bbf159e338dcace4f982daac6e5a8d6be225ca32d/pillow-12.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:9fe611163f6303d1619bbcb653540a4d60f9e55e622d60a3108be0d5b441017a", size = 7001107, upload-time = "2025-10-15T18:22:21.644Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/96/aaa61ce33cc98421fb6088af2a03be4157b1e7e0e87087c888e2370a7f45/pillow-12.0.0-cp312-cp312-win_arm64.whl", hash = "sha256:7dfb439562f234f7d57b1ac6bc8fe7f838a4bd49c79230e0f6a1da93e82f1fad", size = 2436012, upload-time = "2025-10-15T18:22:23.621Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/b3/582327e6c9f86d037b63beebe981425d6811104cb443e8193824ef1a2f27/pillow-12.0.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:b22bd8c974942477156be55a768f7aa37c46904c175be4e158b6a86e3a6b7ca8", size = 5215068, upload-time = "2025-10-15T18:23:59.594Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/d6/67748211d119f3b6540baf90f92fae73ae51d5217b171b0e8b5f7e5d558f/pillow-12.0.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:805ebf596939e48dbb2e4922a1d3852cfc25c38160751ce02da93058b48d252a", size = 4614994, upload-time = "2025-10-15T18:24:01.669Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/e1/f8281e5d844c41872b273b9f2c34a4bf64ca08905668c8ae730eedc7c9fa/pillow-12.0.0-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:cae81479f77420d217def5f54b5b9d279804d17e982e0f2fa19b1d1e14ab5197", size = 5246639, upload-time = "2025-10-15T18:24:03.403Z" },
+    { url = "https://files.pythonhosted.org/packages/94/5a/0d8ab8ffe8a102ff5df60d0de5af309015163bf710c7bb3e8311dd3b3ad0/pillow-12.0.0-pp311-pypy311_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:aeaefa96c768fc66818730b952a862235d68825c178f1b3ffd4efd7ad2edcb7c", size = 6986839, upload-time = "2025-10-15T18:24:05.344Z" },
+    { url = "https://files.pythonhosted.org/packages/20/2e/3434380e8110b76cd9eb00a363c484b050f949b4bbe84ba770bb8508a02c/pillow-12.0.0-pp311-pypy311_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:09f2d0abef9e4e2f349305a4f8cc784a8a6c2f58a8c4892eea13b10a943bd26e", size = 5313505, upload-time = "2025-10-15T18:24:07.137Z" },
+    { url = "https://files.pythonhosted.org/packages/57/ca/5a9d38900d9d74785141d6580950fe705de68af735ff6e727cb911b64740/pillow-12.0.0-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:bdee52571a343d721fb2eb3b090a82d959ff37fc631e3f70422e0c2e029f3e76", size = 5963654, upload-time = "2025-10-15T18:24:09.579Z" },
+    { url = "https://files.pythonhosted.org/packages/95/7e/f896623c3c635a90537ac093c6a618ebe1a90d87206e42309cb5d98a1b9e/pillow-12.0.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:b290fd8aa38422444d4b50d579de197557f182ef1068b75f5aa8558638b8d0a5", size = 6997850, upload-time = "2025-10-15T18:24:11.495Z" },
+]
+
 [[package]]
 name = "platformdirs"
 version = "4.4.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/23/e8/21db9c9987b0e728855bd57bff6984f67952bea55d6f75e055c46b5383e8/platformdirs-4.4.0.tar.gz", hash = "sha256:ca753cf4d81dc309bc67b0ea38fd15dc97bc30ce419a7f58d13eb3bf14c4febf", size = 21634, upload-time = "2025-08-26T14:32:04.268Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/40/4b/2028861e724d3bd36227adfa20d3fd24c3fc6d52032f4a93c133be5d17ce/platformdirs-4.4.0-py3-none-any.whl", hash = "sha256:abd01743f24e5287cd7a5db3752faf1a2d65353f38ec26d98e25a6db65958c85", size = 18654, upload-time = "2025-08-26T14:32:02.735Z" },
 ]
 
+[[package]]
+name = "platformdirs"
+version = "4.5.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/61/33/9611380c2bdb1225fdef633e2a9610622310fed35ab11dac9620972ee088/platformdirs-4.5.0.tar.gz", hash = "sha256:70ddccdd7c99fc5942e9fc25636a8b34d04c24b335100223152c2803e4063312", size = 21632, upload-time = "2025-10-08T17:44:48.791Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/73/cb/ac7874b3e5d58441674fb70742e6c374b28b0c7cb988d37d991cde47166c/platformdirs-4.5.0-py3-none-any.whl", hash = "sha256:e578a81bb873cbb89a41fcc904c7ef523cc18284b7e3b3ccf06aca1403b7ebd3", size = 18651, upload-time = "2025-10-08T17:44:47.223Z" },
+]
+
 [[package]]
 name = "pluggy"
 version = "1.6.0"
@@ -3484,12 +3542,15 @@ wheels = [
 name = "pre-commit"
 version = "4.3.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "cfgv" },
-    { name = "identify" },
-    { name = "nodeenv" },
-    { name = "pyyaml" },
-    { name = "virtualenv" },
+    { name = "cfgv", version = "3.4.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "identify", marker = "python_full_version < '3.10'" },
+    { name = "nodeenv", marker = "python_full_version < '3.10'" },
+    { name = "pyyaml", marker = "python_full_version < '3.10'" },
+    { name = "virtualenv", marker = "python_full_version < '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/ff/29/7cf5bbc236333876e4b41f56e06857a87937ce4bf91e117a6991a2dbb02a/pre_commit-4.3.0.tar.gz", hash = "sha256:499fe450cc9d42e9d58e606262795ecb64dd05438943c62b66f6a8673da30b16", size = 193792, upload-time = "2025-08-09T18:56:14.651Z" }
 wheels = [
@@ -3497,126 +3558,136 @@ wheels = [
 ]
 
 [[package]]
-name = "prometheus-client"
-version = "0.23.1"
+name = "pre-commit"
+version = "4.5.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/23/53/3edb5d68ecf6b38fcbcc1ad28391117d2a322d9a1a3eff04bfdb184d8c3b/prometheus_client-0.23.1.tar.gz", hash = "sha256:6ae8f9081eaaaf153a2e959d2e6c4f4fb57b12ef76c8c7980202f1e57b48b2ce", size = 80481, upload-time = "2025-09-18T20:47:25.043Z" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "cfgv", version = "3.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "identify", marker = "python_full_version >= '3.10'" },
+    { name = "nodeenv", marker = "python_full_version >= '3.10'" },
+    { name = "pyyaml", marker = "python_full_version >= '3.10'" },
+    { name = "virtualenv", marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f4/9b/6a4ffb4ed980519da959e1cf3122fc6cb41211daa58dbae1c73c0e519a37/pre_commit-4.5.0.tar.gz", hash = "sha256:dc5a065e932b19fc1d4c653c6939068fe54325af8e741e74e88db4d28a4dd66b", size = 198428, upload-time = "2025-11-22T21:02:42.304Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/b8/db/14bafcb4af2139e046d03fd00dea7873e48eafe18b7d2797e73d6681f210/prometheus_client-0.23.1-py3-none-any.whl", hash = "sha256:dd1913e6e76b59cfe44e7a4b83e01afc9873c1bdfd2ed8739f1e76aeca115f99", size = 61145, upload-time = "2025-09-18T20:47:23.875Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/c4/b2d28e9d2edf4f1713eb3c29307f1a63f3d67cf09bdda29715a36a68921a/pre_commit-4.5.0-py2.py3-none-any.whl", hash = "sha256:25e2ce09595174d9c97860a95609f9f852c0614ba602de3561e267547f2335e1", size = 226429, upload-time = "2025-11-22T21:02:40.836Z" },
 ]
 
 [[package]]
-name = "prometheus-fastapi-instrumentator"
-version = "7.1.0"
+name = "prompt-toolkit"
+version = "3.0.52"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "prometheus-client" },
-    { name = "starlette" },
+    { name = "wcwidth" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/69/6d/24d53033cf93826aa7857699a4450c1c67e5b9c710e925b1ed2b320c04df/prometheus_fastapi_instrumentator-7.1.0.tar.gz", hash = "sha256:be7cd61eeea4e5912aeccb4261c6631b3f227d8924542d79eaf5af3f439cbe5e", size = 20220, upload-time = "2025-03-19T19:35:05.351Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/a1/96/06e01a7b38dce6fe1db213e061a4602dd6032a8a97ef6c1a862537732421/prompt_toolkit-3.0.52.tar.gz", hash = "sha256:28cde192929c8e7321de85de1ddbe736f1375148b02f2e17edd840042b1be855", size = 434198, upload-time = "2025-08-27T15:24:02.057Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/27/72/0824c18f3bc75810f55dacc2dd933f6ec829771180245ae3cc976195dec0/prometheus_fastapi_instrumentator-7.1.0-py3-none-any.whl", hash = "sha256:978130f3c0bb7b8ebcc90d35516a6fe13e02d2eb358c8f83887cdef7020c31e9", size = 19296, upload-time = "2025-03-19T19:35:04.323Z" },
+    { url = "https://files.pythonhosted.org/packages/84/03/0d3ce49e2505ae70cf43bc5bb3033955d2fc9f932163e84dc0779cc47f48/prompt_toolkit-3.0.52-py3-none-any.whl", hash = "sha256:9aac639a3bbd33284347de5ad8d68ecc044b91a762dc39b7c21095fcd6a19955", size = 391431, upload-time = "2025-08-27T15:23:59.498Z" },
 ]
 
 [[package]]
 name = "propcache"
-version = "0.4.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/ea/c8/d70cd26d845c6d85479d8f5a11a0fd7151e9bc4794cc5e6eb5a790f12df8/propcache-0.4.0.tar.gz", hash = "sha256:c1ad731253eb738f9cadd9fa1844e019576c70bca6a534252e97cf33a57da529", size = 45187, upload-time = "2025-10-04T21:57:39.546Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/d1/7b/4bd85fea3dc58b6f246abf0e6c9e44adca26f6817e6c136780315d723b82/propcache-0.4.0-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:779aaae64089e2f4992e993faea801925395d26bb5de4a47df7ef7f942c14f80", size = 79437, upload-time = "2025-10-04T21:54:49.766Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/91/379ecc1ab37fe33648c7cb2d2252f58969adac1edcd6ec74682d7fb2d920/propcache-0.4.0-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:566552ed9b003030745e5bc7b402b83cf3cecae1bade95262d78543741786db5", size = 45369, upload-time = "2025-10-04T21:54:51.688Z" },
-    { url = "https://files.pythonhosted.org/packages/de/8e/2e002e59e359bbc6ababbb7da168226f93e0533429ea1e93989a7eedcb2a/propcache-0.4.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:944de70384c62d16d4a00c686b422aa75efbc67c4addaebefbb56475d1c16034", size = 47191, upload-time = "2025-10-04T21:54:52.915Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/9a/f56eef9932dc3cbc63df4716f09fbaefec7a475608b643842784a01351b6/propcache-0.4.0-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e878553543ece1f8006d0ba4d096b40290580db173bfb18e16158045b9371335", size = 201000, upload-time = "2025-10-04T21:54:54.556Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/84/e7ad1e09c13f0574dbad261441f6a7f1fb8cc1e2fcb23ec4d4b3e4c7dc67/propcache-0.4.0-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:8659f995b19185179474b18de8755689e1f71e1334d05c14e1895caa4e409cf7", size = 209175, upload-time = "2025-10-04T21:54:55.996Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/74/0b785ac0fbb44a5a7c267efc409b7a62d7a03b17c6442ecb52fd29152314/propcache-0.4.0-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:7aa8cc5c94e682dce91cb4d12d7b81c01641f4ef5b3b3dc53325d43f0e3b9f2e", size = 214874, upload-time = "2025-10-04T21:54:57.831Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/fd/e8d795def2b1d8dc1dc4731d36da1f6111d7c73212909e79462172d0434c/propcache-0.4.0-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:da584d917a1a17f690fc726617fd2c3f3006ea959dae5bb07a5630f7b16f9f5f", size = 196686, upload-time = "2025-10-04T21:54:59.218Z" },
-    { url = "https://files.pythonhosted.org/packages/79/c2/dc992c712c3a1bfaa11d13ff177dbdf9b8b272e7bd443601e37f35728338/propcache-0.4.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:892a072e5b19c3f324a4f8543c9f7e8fc2b0aa08579e46f69bdf0cfc1b440454", size = 192000, upload-time = "2025-10-04T21:55:00.645Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/de/bb108dbdfae594148b033ff283d9fa6e4b0906a99f2c03b98b526883149d/propcache-0.4.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:c20d796210720455086ef3f85adc413d1e41d374742f9b439354f122bbc3b528", size = 190310, upload-time = "2025-10-04T21:55:02.107Z" },
-    { url = "https://files.pythonhosted.org/packages/ba/7b/1bdb5d44ba4c87d270bcf11354950f8f7fbc9ace1fbe7745e683fcb57b5a/propcache-0.4.0-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:df7107a91126a495880576610ae989f19106e1900dd5218d08498391fa43b31d", size = 199646, upload-time = "2025-10-04T21:55:03.55Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/04/44beda877f779f49f5b8c0ff4817a62b5f90a2dfac1ec5311df15a9dfceb/propcache-0.4.0-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:0b04ac2120c161416c866d0b6a4259e47e92231ff166b518cc0efb95777367c3", size = 200507, upload-time = "2025-10-04T21:55:04.914Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/62/a13ad0a63e06f3695fcaeaeeeb62e2cc685181a1248b23a2bc877c8b7111/propcache-0.4.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:1e7fa29c71ffa8d6a37324258737d09475f84715a6e8c350f67f0bc8e5e44993", size = 192787, upload-time = "2025-10-04T21:55:06.385Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/07/386246b3b4a6b11208bcbf57580210fb8c923ab26759389fe594e5615cd7/propcache-0.4.0-cp310-cp310-win32.whl", hash = "sha256:01c0ebc172ca28e9d62876832befbf7f36080eee6ed9c9e00243de2a8089ad57", size = 38004, upload-time = "2025-10-04T21:55:07.692Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/f2/e1fcb9694f590bc443ae5044f982546bb01cbaa3cdf05286e9473a9874bf/propcache-0.4.0-cp310-cp310-win_amd64.whl", hash = "sha256:84f847e64f4d1a232e50460eebc1196642ee9b4c983612f41cd2d44fd2fe7c71", size = 41516, upload-time = "2025-10-04T21:55:08.854Z" },
-    { url = "https://files.pythonhosted.org/packages/15/f4/d211744d41d72fbb89d3ee53963c1dc26892c49f53ae3c49fbc15cfb2548/propcache-0.4.0-cp310-cp310-win_arm64.whl", hash = "sha256:2166466a666a5bebc332cd209cad77d996fad925ca7e8a2a6310ba9e851ae641", size = 38122, upload-time = "2025-10-04T21:55:10.044Z" },
-    { url = "https://files.pythonhosted.org/packages/f9/c4/72b8d41bdbae8aea9c25b869d7cdc3ab5f281f979d8aea30f4646ad12743/propcache-0.4.0-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:6a6a36b94c09711d6397d79006ca47901539fbc602c853d794c39abd6a326549", size = 80035, upload-time = "2025-10-04T21:55:11.266Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/f8/f87115733e221408a363f3a9753419cf2d4be7a8a7ec9dc0788325cd23f1/propcache-0.4.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:da47070e1340a1639aca6b1c18fe1f1f3d8d64d3a1f9ddc67b94475f44cd40f3", size = 45622, upload-time = "2025-10-04T21:55:12.41Z" },
-    { url = "https://files.pythonhosted.org/packages/5d/cc/391f883248faa2efdf6886bdb12ac8edf20eac0863770d8d925450d8cc76/propcache-0.4.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:de536cf796abc5b58d11c0ad56580215d231d9554ea4bb6b8b1b3bed80aa3234", size = 47517, upload-time = "2025-10-04T21:55:13.819Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/d2/5593b59999f42d1044c5ab5f238be1f9d537ab91b0c910727986d520a6e9/propcache-0.4.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f5c82af8e329c3cdc3e717dd3c7b2ff1a218b6de611f6ce76ee34967570a9de9", size = 214540, upload-time = "2025-10-04T21:55:15.206Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/5d/028cdc0eaa1a66ee2ec339a08b5e6ec15e7e71dac86103bebe53ba10dc0f/propcache-0.4.0-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:abe04e7aa5ab2e4056fcf3255ebee2071e4a427681f76d4729519e292c46ecc1", size = 221603, upload-time = "2025-10-04T21:55:16.704Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/f8/e30aee5f59ea21647faef9c82bd67fa510295c34908a7a38571def555881/propcache-0.4.0-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:075ca32384294434344760fdcb95f7833e1d7cf7c4e55f0e726358140179da35", size = 227749, upload-time = "2025-10-04T21:55:18.082Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/85/0757dfc73931bea63b18d26b2c5e7bf13113ca60fe0e5f19905f104bcf6a/propcache-0.4.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:626ec13592928b677f48ff5861040b604b635e93d8e2162fb638397ea83d07e8", size = 209792, upload-time = "2025-10-04T21:55:19.475Z" },
-    { url = "https://files.pythonhosted.org/packages/d2/45/35a6a6241f46948c0ac2418d5bf50cfbcd9735739f42028a1c11e9066a72/propcache-0.4.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:02e071548b6a376e173b0102c3f55dc16e7d055b5307d487e844c320e38cacf2", size = 207979, upload-time = "2025-10-04T21:55:21.164Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/d1/5930396e75c9ed477958eac1496e6fb08794d823e9b14a459f1c0e20f338/propcache-0.4.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:2af6de831a26f42a3f94592964becd8d7f238551786d7525807f02e53defbd13", size = 201923, upload-time = "2025-10-04T21:55:22.5Z" },
-    { url = "https://files.pythonhosted.org/packages/98/72/675455f22bcefeda16907461f9a9a4a93709ff2095e8cf799bdb6c78e030/propcache-0.4.0-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:bd6c6dba1a3b8949e08c4280071c86e38cb602f02e0ed6659234108c7a7cd710", size = 212117, upload-time = "2025-10-04T21:55:23.858Z" },
-    { url = "https://files.pythonhosted.org/packages/13/27/c533302ff80a49a848c3dbd01bb18f87b06826602b3b37043ff00d6b5005/propcache-0.4.0-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:783e91595cf9b66c2deda17f2e8748ae8591aa9f7c65dcab038872bfe83c5bb1", size = 216594, upload-time = "2025-10-04T21:55:25.169Z" },
-    { url = "https://files.pythonhosted.org/packages/63/91/8250fbb601fd16c427e5f469132f27e175c6692dbfa784ef1266dc652e55/propcache-0.4.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:c3f4b125285d354a627eb37f3ea7c13b8842c7c0d47783581d0df0e272dbf5f0", size = 204863, upload-time = "2025-10-04T21:55:26.511Z" },
-    { url = "https://files.pythonhosted.org/packages/34/c4/fd945a9a25845aafb6094b9fa6a88286e4e1c55686e60172c60fe669e0d1/propcache-0.4.0-cp311-cp311-win32.whl", hash = "sha256:71c45f02ffbb8a21040ae816ceff7f6cd749ffac29fc0f9daa42dc1a9652d577", size = 37948, upload-time = "2025-10-04T21:55:27.719Z" },
-    { url = "https://files.pythonhosted.org/packages/42/02/f30e7304661ffe8d51ff4050e06765ac2df6d95cf23c999dfe5a0cd0eb4c/propcache-0.4.0-cp311-cp311-win_amd64.whl", hash = "sha256:7d51f70f77950f8efafed4383865d3533eeee52d8a0dd1c35b65f24de41de4e0", size = 41511, upload-time = "2025-10-04T21:55:29.15Z" },
-    { url = "https://files.pythonhosted.org/packages/a5/f2/edd329d86085438a1ba32cf4cf45fc982d18343bed1f16b218b516c3340d/propcache-0.4.0-cp311-cp311-win_arm64.whl", hash = "sha256:858eaabd2191dd0da5272993ad08a748b5d3ae1aefabea8aee619b45c2af4a64", size = 37957, upload-time = "2025-10-04T21:55:30.31Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/cf/3f88344261d69f8021256f20e82e820c5df3aba96e5ba9b5fdd3685d3a9f/propcache-0.4.0-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:381c84a445efb8c9168f1393a5a7c566de22edc42bfe207a142fff919b37f5d9", size = 79846, upload-time = "2025-10-04T21:55:31.447Z" },
-    { url = "https://files.pythonhosted.org/packages/be/fa/0286fc92764eead9dcfee639b67828daa32e61dd0f1618831547141eb28b/propcache-0.4.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:5a531d29d7b873b12730972237c48b1a4e5980b98cf21b3f09fa4710abd3a8c3", size = 45850, upload-time = "2025-10-04T21:55:32.637Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/83/57840656f972f8a67992eee40781e4066657776dcb889f49df0e8eecb112/propcache-0.4.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:cd6e22255ed73efeaaeb1765505a66a48a9ec9ebc919fce5ad490fe5e33b1555", size = 47171, upload-time = "2025-10-04T21:55:33.819Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/8e/e0a0bd376c3440476b924eca517589ee535bb4520420d178268bf88558ba/propcache-0.4.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d9a8d277dc218ddf04ec243a53ac309b1afcebe297c0526a8f82320139b56289", size = 225306, upload-time = "2025-10-04T21:55:35.312Z" },
-    { url = "https://files.pythonhosted.org/packages/84/fe/76884442da1bab6d4353ba1c43fdc4a770c3b3973f3ac7620a7205402fdd/propcache-0.4.0-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:399c73201d88c856a994916200d7cba41d7687096f8eb5139eb68f02785dc3f7", size = 230013, upload-time = "2025-10-04T21:55:37.005Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/b7/322af273bd1136bb7e13628821fb855c9f61d64651c73fea71dded68dda5/propcache-0.4.0-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a1d5e474d43c238035b74ecf997f655afa67f979bae591ac838bb3fbe3076392", size = 238331, upload-time = "2025-10-04T21:55:38.713Z" },
-    { url = "https://files.pythonhosted.org/packages/84/5e/036d2b105927ae7f179346c9911d16c345f4dba5a19a063f23a8d28acfbd/propcache-0.4.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:22f589652ee38de96aa58dd219335604e09666092bc250c1d9c26a55bcef9932", size = 221461, upload-time = "2025-10-04T21:55:40.034Z" },
-    { url = "https://files.pythonhosted.org/packages/63/0d/babd038efb12a87a46ab070438c52daeac6bed0a930693a418feef8cb8a6/propcache-0.4.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e5227da556b2939da6125cda1d5eecf9e412e58bc97b41e2f192605c3ccbb7c2", size = 216707, upload-time = "2025-10-04T21:55:41.455Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/68/dd075a037381581f16e7e504a6da9c1d7e415e945dd8ed67905d608f0687/propcache-0.4.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:92bc43a1ab852310721ce856f40a3a352254aa6f5e26f0fad870b31be45bba2e", size = 212591, upload-time = "2025-10-04T21:55:42.938Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/43/22698f28fc8e04c32b109cb9cb81305a4873b77c907b17484566b6133aef/propcache-0.4.0-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:83ae2f5343f6f06f4c91ae530d95f56b415f768f9c401a5ee2a10459cf74370b", size = 220188, upload-time = "2025-10-04T21:55:44.53Z" },
-    { url = "https://files.pythonhosted.org/packages/96/7a/27886e4a4c69598a38fbeeed64f9b8ddfa6f08fe3452035845a1fe90336f/propcache-0.4.0-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:077a32977399dc05299b16e793210341a0b511eb0a86d1796873e83ce47334cc", size = 226736, upload-time = "2025-10-04T21:55:46.348Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/c7/313c632b5888db3c9f4cb262420dcd5e57cf858d939d6ad9c3b1b90c12af/propcache-0.4.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:94a278c45e6463031b5a8278e40a07edf2bcc3b5379510e22b6c1a6e6498c194", size = 216363, upload-time = "2025-10-04T21:55:47.768Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/5d/5aaf82bd1542aedb47d10483b84f49ee8f00d970a58e27534cd241e9c5ac/propcache-0.4.0-cp312-cp312-win32.whl", hash = "sha256:4c491462e1dc80f9deb93f428aad8d83bb286de212837f58eb48e75606e7726c", size = 37945, upload-time = "2025-10-04T21:55:49.104Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/67/47ffff6eb176f383f56319f31c0e1bcf7500cb94ffb7582efc600c6b3c73/propcache-0.4.0-cp312-cp312-win_amd64.whl", hash = "sha256:cdb0cecafb528ab15ed89cdfed183074d15912d046d3e304955513b50a34b907", size = 41530, upload-time = "2025-10-04T21:55:50.261Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/7e/61b70306b9d7527286ce887a8ff28c304ab2514e5893eea36b5bdf7a21af/propcache-0.4.0-cp312-cp312-win_arm64.whl", hash = "sha256:b2f29697d1110e8cdf7a39cc630498df0082d7898b79b731c1c863f77c6e8cfc", size = 37662, upload-time = "2025-10-04T21:55:51.35Z" },
-    { url = "https://files.pythonhosted.org/packages/84/24/31f6efbca9211550377731642131ad274b01b2dc4788c0a7d5c47fcd9e1c/propcache-0.4.0-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:0cd30341142c68377cf3c4e2d9f0581e6e528694b2d57c62c786be441053d2fc", size = 80114, upload-time = "2025-10-04T21:57:16.407Z" },
-    { url = "https://files.pythonhosted.org/packages/21/96/295cb8cef28d50cfccf445689a1dc8de0b2e3fa923921404560ae7a423cc/propcache-0.4.0-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:2c46d37955820dd883cf9156ceb7825b8903e910bdd869902e20a5ac4ecd2c8b", size = 45697, upload-time = "2025-10-04T21:57:17.6Z" },
-    { url = "https://files.pythonhosted.org/packages/97/c0/3468b29275579cf533b25de083072b5fe11a7666106827a5ea704eb5bfc3/propcache-0.4.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:0b12df77eb19266efd146627a65b8ad414f9d15672d253699a50c8205661a820", size = 47510, upload-time = "2025-10-04T21:57:18.807Z" },
-    { url = "https://files.pythonhosted.org/packages/29/4e/fd5f46077a435e0866eb27a1a9771067ab41bfd35e14b1aa51c097bcd035/propcache-0.4.0-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1cdabd60e109506462e6a7b37008e57979e737dc6e7dfbe1437adcfe354d1a0a", size = 201095, upload-time = "2025-10-04T21:57:20.104Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/d0/cdbc660a9946e3d389db2541145046073f9ce767e0dbea13bced2657fcf5/propcache-0.4.0-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:65ff56a31f25925ef030b494fe63289bf07ef0febe6da181b8219146c590e185", size = 209505, upload-time = "2025-10-04T21:57:21.509Z" },
-    { url = "https://files.pythonhosted.org/packages/42/77/9ee6d0266bc33b5ee7c3eabddb48be30b141f6acddf964d0844ff521be6a/propcache-0.4.0-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:96153e037ae065bb71cae889f23c933190d81ae183f3696a030b47352fd8655d", size = 215226, upload-time = "2025-10-04T21:57:22.88Z" },
-    { url = "https://files.pythonhosted.org/packages/7d/c6/72c4ed7751a82f60d14fb713734efcfcf190942288dd2056898bac4824d2/propcache-0.4.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4bf95be277fbb51513895c2cecc81ab12a421cdbd8837f159828a919a0167f96", size = 196960, upload-time = "2025-10-04T21:57:24.562Z" },
-    { url = "https://files.pythonhosted.org/packages/82/7e/e01179669192b94a640d942ad42535c38ba6cad70518db7354585e051d1c/propcache-0.4.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:8d18d796ffecdc8253742fd53a94ceee2e77ad149eb9ed5960c2856b5f692f71", size = 192189, upload-time = "2025-10-04T21:57:25.973Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/ea/0c2be8a7fdf012598583df72efd38220ef0db671fe9a9d8ff82f790bf313/propcache-0.4.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:4a52c25a51d5894ba60c567b0dbcf73de2f3cd642cf5343679e07ca3a768b085", size = 190520, upload-time = "2025-10-04T21:57:27.394Z" },
-    { url = "https://files.pythonhosted.org/packages/17/68/288e1e53c4f677906eac137d047720165aa1e10cd4d8b28f18d6b4c5c29a/propcache-0.4.0-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:e0ce7f3d1faf7ad58652ed758cc9753049af5308b38f89948aa71793282419c5", size = 199773, upload-time = "2025-10-04T21:57:28.733Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/52/eff7750f9d7a20000143fb857b6215fc6ac1e897ca62eeb20ef6b62af4ea/propcache-0.4.0-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:545987971b2aded25ba4698135ea0ae128836e7deb6e18c29a581076aaef44aa", size = 200567, upload-time = "2025-10-04T21:57:30.241Z" },
-    { url = "https://files.pythonhosted.org/packages/27/a5/1ecedd68f72554c7c9e5508158b7634eb84a24cb8880183d1cd74a433d6e/propcache-0.4.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:7da5c4c72ae40fd3ce87213ab057db66df53e55600d0b9e72e2b7f5a470a2cc4", size = 192963, upload-time = "2025-10-04T21:57:32.382Z" },
-    { url = "https://files.pythonhosted.org/packages/67/0a/aeaf727f5a0dcea135e2f90469d2410767d84ef386005219a50825381021/propcache-0.4.0-cp39-cp39-win32.whl", hash = "sha256:2015218812ee8f13bbaebc9f52b1e424cc130b68d4857bef018e65e3834e1c4d", size = 38223, upload-time = "2025-10-04T21:57:33.889Z" },
-    { url = "https://files.pythonhosted.org/packages/45/0b/604ba6b7ad807312a2a0125a64c46cd83a12399fc29c1646cc42cf814f9e/propcache-0.4.0-cp39-cp39-win_amd64.whl", hash = "sha256:39f0f6a3b56e82dc91d84c763b783c5c33720a33c70ee48a1c13ba800ac1fa69", size = 41782, upload-time = "2025-10-04T21:57:35.522Z" },
-    { url = "https://files.pythonhosted.org/packages/77/30/2107b450136c26fc5b3c3a811e80898b6343a4c514bedee4c6a3d098723e/propcache-0.4.0-cp39-cp39-win_arm64.whl", hash = "sha256:236c8da353ea7c22a8e963ab78cddb1126f700ae9538e2c4c6ef471e5545494b", size = 38329, upload-time = "2025-10-04T21:57:36.959Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/16/794c114f6041bbe2de23eb418ef58a0f45de27224d5540f5dbb266a73d72/propcache-0.4.0-py3-none-any.whl", hash = "sha256:015b2ca2f98ea9e08ac06eecc409d5d988f78c5fd5821b2ad42bc9afcd6b1557", size = 13183, upload-time = "2025-10-04T21:57:38.054Z" },
+version = "0.4.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/9e/da/e9fc233cf63743258bff22b3dfa7ea5baef7b5bc324af47a0ad89b8ffc6f/propcache-0.4.1.tar.gz", hash = "sha256:f48107a8c637e80362555f37ecf49abe20370e557cc4ab374f04ec4423c97c3d", size = 46442, upload-time = "2025-10-08T19:49:02.291Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3c/0e/934b541323035566a9af292dba85a195f7b78179114f2c6ebb24551118a9/propcache-0.4.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:7c2d1fa3201efaf55d730400d945b5b3ab6e672e100ba0f9a409d950ab25d7db", size = 79534, upload-time = "2025-10-08T19:46:02.083Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/6b/db0d03d96726d995dc7171286c6ba9d8d14251f37433890f88368951a44e/propcache-0.4.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:1eb2994229cc8ce7fe9b3db88f5465f5fd8651672840b2e426b88cdb1a30aac8", size = 45526, upload-time = "2025-10-08T19:46:03.884Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/c3/82728404aea669e1600f304f2609cde9e665c18df5a11cdd57ed73c1dceb/propcache-0.4.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:66c1f011f45a3b33d7bcb22daed4b29c0c9e2224758b6be00686731e1b46f925", size = 47263, upload-time = "2025-10-08T19:46:05.405Z" },
+    { url = "https://files.pythonhosted.org/packages/df/1b/39313ddad2bf9187a1432654c38249bab4562ef535ef07f5eb6eb04d0b1b/propcache-0.4.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9a52009f2adffe195d0b605c25ec929d26b36ef986ba85244891dee3b294df21", size = 201012, upload-time = "2025-10-08T19:46:07.165Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/01/f1d0b57d136f294a142acf97f4ed58c8e5b974c21e543000968357115011/propcache-0.4.1-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:5d4e2366a9c7b837555cf02fb9be2e3167d333aff716332ef1b7c3a142ec40c5", size = 209491, upload-time = "2025-10-08T19:46:08.909Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/c8/038d909c61c5bb039070b3fb02ad5cccdb1dde0d714792e251cdb17c9c05/propcache-0.4.1-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:9d2b6caef873b4f09e26ea7e33d65f42b944837563a47a94719cc3544319a0db", size = 215319, upload-time = "2025-10-08T19:46:10.7Z" },
+    { url = "https://files.pythonhosted.org/packages/08/57/8c87e93142b2c1fa2408e45695205a7ba05fb5db458c0bf5c06ba0e09ea6/propcache-0.4.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:2b16ec437a8c8a965ecf95739448dd938b5c7f56e67ea009f4300d8df05f32b7", size = 196856, upload-time = "2025-10-08T19:46:12.003Z" },
+    { url = "https://files.pythonhosted.org/packages/42/df/5615fec76aa561987a534759b3686008a288e73107faa49a8ae5795a9f7a/propcache-0.4.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:296f4c8ed03ca7476813fe666c9ea97869a8d7aec972618671b33a38a5182ef4", size = 193241, upload-time = "2025-10-08T19:46:13.495Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/21/62949eb3a7a54afe8327011c90aca7e03547787a88fb8bd9726806482fea/propcache-0.4.1-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:1f0978529a418ebd1f49dad413a2b68af33f85d5c5ca5c6ca2a3bed375a7ac60", size = 190552, upload-time = "2025-10-08T19:46:14.938Z" },
+    { url = "https://files.pythonhosted.org/packages/30/ee/ab4d727dd70806e5b4de96a798ae7ac6e4d42516f030ee60522474b6b332/propcache-0.4.1-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:fd138803047fb4c062b1c1dd95462f5209456bfab55c734458f15d11da288f8f", size = 200113, upload-time = "2025-10-08T19:46:16.695Z" },
+    { url = "https://files.pythonhosted.org/packages/8a/0b/38b46208e6711b016aa8966a3ac793eee0d05c7159d8342aa27fc0bc365e/propcache-0.4.1-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:8c9b3cbe4584636d72ff556d9036e0c9317fa27b3ac1f0f558e7e84d1c9c5900", size = 200778, upload-time = "2025-10-08T19:46:18.023Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/81/5abec54355ed344476bee711e9f04815d4b00a311ab0535599204eecc257/propcache-0.4.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:f93243fdc5657247533273ac4f86ae106cc6445a0efacb9a1bfe982fcfefd90c", size = 193047, upload-time = "2025-10-08T19:46:19.449Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/b6/1f237c04e32063cb034acd5f6ef34ef3a394f75502e72703545631ab1ef6/propcache-0.4.1-cp310-cp310-win32.whl", hash = "sha256:a0ee98db9c5f80785b266eb805016e36058ac72c51a064040f2bc43b61101cdb", size = 38093, upload-time = "2025-10-08T19:46:20.643Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/67/354aac4e0603a15f76439caf0427781bcd6797f370377f75a642133bc954/propcache-0.4.1-cp310-cp310-win_amd64.whl", hash = "sha256:1cdb7988c4e5ac7f6d175a28a9aa0c94cb6f2ebe52756a3c0cda98d2809a9e37", size = 41638, upload-time = "2025-10-08T19:46:21.935Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/e1/74e55b9fd1a4c209ff1a9a824bf6c8b3d1fc5a1ac3eabe23462637466785/propcache-0.4.1-cp310-cp310-win_arm64.whl", hash = "sha256:d82ad62b19645419fe79dd63b3f9253e15b30e955c0170e5cebc350c1844e581", size = 38229, upload-time = "2025-10-08T19:46:23.368Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/d4/4e2c9aaf7ac2242b9358f98dccd8f90f2605402f5afeff6c578682c2c491/propcache-0.4.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:60a8fda9644b7dfd5dece8c61d8a85e271cb958075bfc4e01083c148b61a7caf", size = 80208, upload-time = "2025-10-08T19:46:24.597Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/21/d7b68e911f9c8e18e4ae43bdbc1e1e9bbd971f8866eb81608947b6f585ff/propcache-0.4.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:c30b53e7e6bda1d547cabb47c825f3843a0a1a42b0496087bb58d8fedf9f41b5", size = 45777, upload-time = "2025-10-08T19:46:25.733Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/1d/11605e99ac8ea9435651ee71ab4cb4bf03f0949586246476a25aadfec54a/propcache-0.4.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:6918ecbd897443087a3b7cd978d56546a812517dcaaca51b49526720571fa93e", size = 47647, upload-time = "2025-10-08T19:46:27.304Z" },
+    { url = "https://files.pythonhosted.org/packages/58/1a/3c62c127a8466c9c843bccb503d40a273e5cc69838805f322e2826509e0d/propcache-0.4.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3d902a36df4e5989763425a8ab9e98cd8ad5c52c823b34ee7ef307fd50582566", size = 214929, upload-time = "2025-10-08T19:46:28.62Z" },
+    { url = "https://files.pythonhosted.org/packages/56/b9/8fa98f850960b367c4b8fe0592e7fc341daa7a9462e925228f10a60cf74f/propcache-0.4.1-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:a9695397f85973bb40427dedddf70d8dc4a44b22f1650dd4af9eedf443d45165", size = 221778, upload-time = "2025-10-08T19:46:30.358Z" },
+    { url = "https://files.pythonhosted.org/packages/46/a6/0ab4f660eb59649d14b3d3d65c439421cf2f87fe5dd68591cbe3c1e78a89/propcache-0.4.1-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:2bb07ffd7eaad486576430c89f9b215f9e4be68c4866a96e97db9e97fead85dc", size = 228144, upload-time = "2025-10-08T19:46:32.607Z" },
+    { url = "https://files.pythonhosted.org/packages/52/6a/57f43e054fb3d3a56ac9fc532bc684fc6169a26c75c353e65425b3e56eef/propcache-0.4.1-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fd6f30fdcf9ae2a70abd34da54f18da086160e4d7d9251f81f3da0ff84fc5a48", size = 210030, upload-time = "2025-10-08T19:46:33.969Z" },
+    { url = "https://files.pythonhosted.org/packages/40/e2/27e6feebb5f6b8408fa29f5efbb765cd54c153ac77314d27e457a3e993b7/propcache-0.4.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:fc38cba02d1acba4e2869eef1a57a43dfbd3d49a59bf90dda7444ec2be6a5570", size = 208252, upload-time = "2025-10-08T19:46:35.309Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/f8/91c27b22ccda1dbc7967f921c42825564fa5336a01ecd72eb78a9f4f53c2/propcache-0.4.1-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:67fad6162281e80e882fb3ec355398cf72864a54069d060321f6cd0ade95fe85", size = 202064, upload-time = "2025-10-08T19:46:36.993Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/26/7f00bd6bd1adba5aafe5f4a66390f243acab58eab24ff1a08bebb2ef9d40/propcache-0.4.1-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:f10207adf04d08bec185bae14d9606a1444715bc99180f9331c9c02093e1959e", size = 212429, upload-time = "2025-10-08T19:46:38.398Z" },
+    { url = "https://files.pythonhosted.org/packages/84/89/fd108ba7815c1117ddca79c228f3f8a15fc82a73bca8b142eb5de13b2785/propcache-0.4.1-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:e9b0d8d0845bbc4cfcdcbcdbf5086886bc8157aa963c31c777ceff7846c77757", size = 216727, upload-time = "2025-10-08T19:46:39.732Z" },
+    { url = "https://files.pythonhosted.org/packages/79/37/3ec3f7e3173e73f1d600495d8b545b53802cbf35506e5732dd8578db3724/propcache-0.4.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:981333cb2f4c1896a12f4ab92a9cc8f09ea664e9b7dbdc4eff74627af3a11c0f", size = 205097, upload-time = "2025-10-08T19:46:41.025Z" },
+    { url = "https://files.pythonhosted.org/packages/61/b0/b2631c19793f869d35f47d5a3a56fb19e9160d3c119f15ac7344fc3ccae7/propcache-0.4.1-cp311-cp311-win32.whl", hash = "sha256:f1d2f90aeec838a52f1c1a32fe9a619fefd5e411721a9117fbf82aea638fe8a1", size = 38084, upload-time = "2025-10-08T19:46:42.693Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/78/6cce448e2098e9f3bfc91bb877f06aa24b6ccace872e39c53b2f707c4648/propcache-0.4.1-cp311-cp311-win_amd64.whl", hash = "sha256:364426a62660f3f699949ac8c621aad6977be7126c5807ce48c0aeb8e7333ea6", size = 41637, upload-time = "2025-10-08T19:46:43.778Z" },
+    { url = "https://files.pythonhosted.org/packages/9c/e9/754f180cccd7f51a39913782c74717c581b9cc8177ad0e949f4d51812383/propcache-0.4.1-cp311-cp311-win_arm64.whl", hash = "sha256:e53f3a38d3510c11953f3e6a33f205c6d1b001129f972805ca9b42fc308bc239", size = 38064, upload-time = "2025-10-08T19:46:44.872Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/0f/f17b1b2b221d5ca28b4b876e8bb046ac40466513960646bda8e1853cdfa2/propcache-0.4.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:e153e9cd40cc8945138822807139367f256f89c6810c2634a4f6902b52d3b4e2", size = 80061, upload-time = "2025-10-08T19:46:46.075Z" },
+    { url = "https://files.pythonhosted.org/packages/76/47/8ccf75935f51448ba9a16a71b783eb7ef6b9ee60f5d14c7f8a8a79fbeed7/propcache-0.4.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:cd547953428f7abb73c5ad82cbb32109566204260d98e41e5dfdc682eb7f8403", size = 46037, upload-time = "2025-10-08T19:46:47.23Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/b6/5c9a0e42df4d00bfb4a3cbbe5cf9f54260300c88a0e9af1f47ca5ce17ac0/propcache-0.4.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:f048da1b4f243fc44f205dfd320933a951b8d89e0afd4c7cacc762a8b9165207", size = 47324, upload-time = "2025-10-08T19:46:48.384Z" },
+    { url = "https://files.pythonhosted.org/packages/9e/d3/6c7ee328b39a81ee877c962469f1e795f9db87f925251efeb0545e0020d0/propcache-0.4.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ec17c65562a827bba85e3872ead335f95405ea1674860d96483a02f5c698fa72", size = 225505, upload-time = "2025-10-08T19:46:50.055Z" },
+    { url = "https://files.pythonhosted.org/packages/01/5d/1c53f4563490b1d06a684742cc6076ef944bc6457df6051b7d1a877c057b/propcache-0.4.1-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:405aac25c6394ef275dee4c709be43745d36674b223ba4eb7144bf4d691b7367", size = 230242, upload-time = "2025-10-08T19:46:51.815Z" },
+    { url = "https://files.pythonhosted.org/packages/20/e1/ce4620633b0e2422207c3cb774a0ee61cac13abc6217763a7b9e2e3f4a12/propcache-0.4.1-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0013cb6f8dde4b2a2f66903b8ba740bdfe378c943c4377a200551ceb27f379e4", size = 238474, upload-time = "2025-10-08T19:46:53.208Z" },
+    { url = "https://files.pythonhosted.org/packages/46/4b/3aae6835b8e5f44ea6a68348ad90f78134047b503765087be2f9912140ea/propcache-0.4.1-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:15932ab57837c3368b024473a525e25d316d8353016e7cc0e5ba9eb343fbb1cf", size = 221575, upload-time = "2025-10-08T19:46:54.511Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/a5/8a5e8678bcc9d3a1a15b9a29165640d64762d424a16af543f00629c87338/propcache-0.4.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:031dce78b9dc099f4c29785d9cf5577a3faf9ebf74ecbd3c856a7b92768c3df3", size = 216736, upload-time = "2025-10-08T19:46:56.212Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/63/b7b215eddeac83ca1c6b934f89d09a625aa9ee4ba158338854c87210cc36/propcache-0.4.1-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:ab08df6c9a035bee56e31af99be621526bd237bea9f32def431c656b29e41778", size = 213019, upload-time = "2025-10-08T19:46:57.595Z" },
+    { url = "https://files.pythonhosted.org/packages/57/74/f580099a58c8af587cac7ba19ee7cb418506342fbbe2d4a4401661cca886/propcache-0.4.1-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:4d7af63f9f93fe593afbf104c21b3b15868efb2c21d07d8732c0c4287e66b6a6", size = 220376, upload-time = "2025-10-08T19:46:59.067Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/ee/542f1313aff7eaf19c2bb758c5d0560d2683dac001a1c96d0774af799843/propcache-0.4.1-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:cfc27c945f422e8b5071b6e93169679e4eb5bf73bbcbf1ba3ae3a83d2f78ebd9", size = 226988, upload-time = "2025-10-08T19:47:00.544Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/18/9c6b015dd9c6930f6ce2229e1f02fb35298b847f2087ea2b436a5bfa7287/propcache-0.4.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:35c3277624a080cc6ec6f847cbbbb5b49affa3598c4535a0a4682a697aaa5c75", size = 215615, upload-time = "2025-10-08T19:47:01.968Z" },
+    { url = "https://files.pythonhosted.org/packages/80/9e/e7b85720b98c45a45e1fca6a177024934dc9bc5f4d5dd04207f216fc33ed/propcache-0.4.1-cp312-cp312-win32.whl", hash = "sha256:671538c2262dadb5ba6395e26c1731e1d52534bfe9ae56d0b5573ce539266aa8", size = 38066, upload-time = "2025-10-08T19:47:03.503Z" },
+    { url = "https://files.pythonhosted.org/packages/54/09/d19cff2a5aaac632ec8fc03737b223597b1e347416934c1b3a7df079784c/propcache-0.4.1-cp312-cp312-win_amd64.whl", hash = "sha256:cb2d222e72399fcf5890d1d5cc1060857b9b236adff2792ff48ca2dfd46c81db", size = 41655, upload-time = "2025-10-08T19:47:04.973Z" },
+    { url = "https://files.pythonhosted.org/packages/68/ab/6b5c191bb5de08036a8c697b265d4ca76148efb10fa162f14af14fb5f076/propcache-0.4.1-cp312-cp312-win_arm64.whl", hash = "sha256:204483131fb222bdaaeeea9f9e6c6ed0cac32731f75dfc1d4a567fc1926477c1", size = 37789, upload-time = "2025-10-08T19:47:06.077Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/01/0ebaec9003f5d619a7475165961f8e3083cf8644d704b60395df3601632d/propcache-0.4.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:3d233076ccf9e450c8b3bc6720af226b898ef5d051a2d145f7d765e6e9f9bcff", size = 80277, upload-time = "2025-10-08T19:48:36.647Z" },
+    { url = "https://files.pythonhosted.org/packages/34/58/04af97ac586b4ef6b9026c3fd36ee7798b737a832f5d3440a4280dcebd3a/propcache-0.4.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:357f5bb5c377a82e105e44bd3d52ba22b616f7b9773714bff93573988ef0a5fb", size = 45865, upload-time = "2025-10-08T19:48:37.859Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/19/b65d98ae21384518b291d9939e24a8aeac4fdb5101b732576f8f7540e834/propcache-0.4.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:cbc3b6dfc728105b2a57c06791eb07a94229202ea75c59db644d7d496b698cac", size = 47636, upload-time = "2025-10-08T19:48:39.038Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/0f/317048c6d91c356c7154dca5af019e6effeb7ee15fa6a6db327cc19e12b4/propcache-0.4.1-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:182b51b421f0501952d938dc0b0eb45246a5b5153c50d42b495ad5fb7517c888", size = 201126, upload-time = "2025-10-08T19:48:40.774Z" },
+    { url = "https://files.pythonhosted.org/packages/71/69/0b2a7a5a6ee83292b4b997dbd80549d8ce7d40b6397c1646c0d9495f5a85/propcache-0.4.1-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4b536b39c5199b96fc6245eb5fb796c497381d3942f169e44e8e392b29c9ebcc", size = 209837, upload-time = "2025-10-08T19:48:42.167Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/92/c699ac495a6698df6e497fc2de27af4b6ace10d8e76528357ce153722e45/propcache-0.4.1-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:db65d2af507bbfbdcedb254a11149f894169d90488dd3e7190f7cdcb2d6cd57a", size = 215578, upload-time = "2025-10-08T19:48:43.56Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/ee/14de81c5eb02c0ee4f500b4e39c4e1bd0677c06e72379e6ab18923c773fc/propcache-0.4.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fd2dbc472da1f772a4dae4fa24be938a6c544671a912e30529984dd80400cd88", size = 197187, upload-time = "2025-10-08T19:48:45.309Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/94/48dce9aaa6d8dd5a0859bad75158ec522546d4ac23f8e2f05fac469477dd/propcache-0.4.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:daede9cd44e0f8bdd9e6cc9a607fc81feb80fae7a5fc6cecaff0e0bb32e42d00", size = 193478, upload-time = "2025-10-08T19:48:47.743Z" },
+    { url = "https://files.pythonhosted.org/packages/60/b5/0516b563e801e1ace212afde869a0596a0d7115eec0b12d296d75633fb29/propcache-0.4.1-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:71b749281b816793678ae7f3d0d84bd36e694953822eaad408d682efc5ca18e0", size = 190650, upload-time = "2025-10-08T19:48:49.373Z" },
+    { url = "https://files.pythonhosted.org/packages/24/89/e0f7d4a5978cd56f8cd67735f74052f257dc471ec901694e430f0d1572fe/propcache-0.4.1-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:0002004213ee1f36cfb3f9a42b5066100c44276b9b72b4e1504cddd3d692e86e", size = 200251, upload-time = "2025-10-08T19:48:51.4Z" },
+    { url = "https://files.pythonhosted.org/packages/06/7d/a1fac863d473876ed4406c914f2e14aa82d2f10dd207c9e16fc383cc5a24/propcache-0.4.1-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:fe49d0a85038f36ba9e3ffafa1103e61170b28e95b16622e11be0a0ea07c6781", size = 200919, upload-time = "2025-10-08T19:48:53.227Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/4e/f86a256ff24944cf5743e4e6c6994e3526f6acfcfb55e21694c2424f758c/propcache-0.4.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:99d43339c83aaf4d32bda60928231848eee470c6bda8d02599cc4cebe872d183", size = 193211, upload-time = "2025-10-08T19:48:55.027Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/3f/3fbad5f4356b068f1b047d300a6ff2c66614d7030f078cd50be3fec04228/propcache-0.4.1-cp39-cp39-win32.whl", hash = "sha256:a129e76735bc792794d5177069691c3217898b9f5cee2b2661471e52ffe13f19", size = 38314, upload-time = "2025-10-08T19:48:56.792Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/45/d78d136c3a3d215677abb886785aae744da2c3005bcb99e58640c56529b1/propcache-0.4.1-cp39-cp39-win_amd64.whl", hash = "sha256:948dab269721ae9a87fd16c514a0a2c2a1bdb23a9a61b969b0f9d9ee2968546f", size = 41912, upload-time = "2025-10-08T19:48:57.995Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/2a/b0632941f25139f4e58450b307242951f7c2717a5704977c6d5323a800af/propcache-0.4.1-cp39-cp39-win_arm64.whl", hash = "sha256:5fd37c406dd6dc85aa743e214cef35dc54bbdd1419baac4f6ae5e5b1a2976938", size = 38450, upload-time = "2025-10-08T19:48:59.349Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/5a/bc7b4a4ef808fa59a816c17b20c4bef6884daebbdf627ff2a161da67da19/propcache-0.4.1-py3-none-any.whl", hash = "sha256:af2a6052aeb6cf17d3e46ee169099044fd8224cbaf75c76a2ef596e8163e2237", size = 13305, upload-time = "2025-10-08T19:49:00.792Z" },
 ]
 
 [[package]]
 name = "protobuf"
-version = "6.32.1"
+version = "6.33.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/fa/a4/cc17347aa2897568beece2e674674359f911d6fe21b0b8d6268cd42727ac/protobuf-6.32.1.tar.gz", hash = "sha256:ee2469e4a021474ab9baafea6cd070e5bf27c7d29433504ddea1a4ee5850f68d", size = 440635, upload-time = "2025-09-11T21:38:42.935Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/0a/03/a1440979a3f74f16cab3b75b0da1a1a7f922d56a8ddea96092391998edc0/protobuf-6.33.1.tar.gz", hash = "sha256:97f65757e8d09870de6fd973aeddb92f85435607235d20b2dfed93405d00c85b", size = 443432, upload-time = "2025-11-13T16:44:18.895Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c0/98/645183ea03ab3995d29086b8bf4f7562ebd3d10c9a4b14ee3f20d47cfe50/protobuf-6.32.1-cp310-abi3-win32.whl", hash = "sha256:a8a32a84bc9f2aad712041b8b366190f71dde248926da517bde9e832e4412085", size = 424411, upload-time = "2025-09-11T21:38:27.427Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/f3/6f58f841f6ebafe076cebeae33fc336e900619d34b1c93e4b5c97a81fdfa/protobuf-6.32.1-cp310-abi3-win_amd64.whl", hash = "sha256:b00a7d8c25fa471f16bc8153d0e53d6c9e827f0953f3c09aaa4331c718cae5e1", size = 435738, upload-time = "2025-09-11T21:38:30.959Z" },
-    { url = "https://files.pythonhosted.org/packages/10/56/a8a3f4e7190837139e68c7002ec749190a163af3e330f65d90309145a210/protobuf-6.32.1-cp39-abi3-macosx_10_9_universal2.whl", hash = "sha256:d8c7e6eb619ffdf105ee4ab76af5a68b60a9d0f66da3ea12d1640e6d8dab7281", size = 426454, upload-time = "2025-09-11T21:38:34.076Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/be/8dd0a927c559b37d7a6c8ab79034fd167dcc1f851595f2e641ad62be8643/protobuf-6.32.1-cp39-abi3-manylinux2014_aarch64.whl", hash = "sha256:2f5b80a49e1eb7b86d85fcd23fe92df154b9730a725c3b38c4e43b9d77018bf4", size = 322874, upload-time = "2025-09-11T21:38:35.509Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/f6/88d77011b605ef979aace37b7703e4eefad066f7e84d935e5a696515c2dd/protobuf-6.32.1-cp39-abi3-manylinux2014_x86_64.whl", hash = "sha256:b1864818300c297265c83a4982fd3169f97122c299f56a56e2445c3698d34710", size = 322013, upload-time = "2025-09-11T21:38:37.017Z" },
-    { url = "https://files.pythonhosted.org/packages/05/9d/d6f1a8b6657296920c58f6b85f7bca55fa27e3ca7fc5914604d89cd0250b/protobuf-6.32.1-cp39-cp39-win32.whl", hash = "sha256:68ff170bac18c8178f130d1ccb94700cf72852298e016a2443bdb9502279e5f1", size = 424505, upload-time = "2025-09-11T21:38:38.415Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/cd/891bd2d23558f52392a5687b2406a741e2e28d629524c88aade457029acd/protobuf-6.32.1-cp39-cp39-win_amd64.whl", hash = "sha256:d0975d0b2f3e6957111aa3935d08a0eb7e006b1505d825f862a1fffc8348e122", size = 435825, upload-time = "2025-09-11T21:38:39.773Z" },
-    { url = "https://files.pythonhosted.org/packages/97/b7/15cc7d93443d6c6a84626ae3258a91f4c6ac8c0edd5df35ea7658f71b79c/protobuf-6.32.1-py3-none-any.whl", hash = "sha256:2601b779fc7d32a866c6b4404f9d42a3f67c5b9f3f15b4db3cccabe06b95c346", size = 169289, upload-time = "2025-09-11T21:38:41.234Z" },
+    { url = "https://files.pythonhosted.org/packages/06/f1/446a9bbd2c60772ca36556bac8bfde40eceb28d9cc7838755bc41e001d8f/protobuf-6.33.1-cp310-abi3-win32.whl", hash = "sha256:f8d3fdbc966aaab1d05046d0240dd94d40f2a8c62856d41eaa141ff64a79de6b", size = 425593, upload-time = "2025-11-13T16:44:06.275Z" },
+    { url = "https://files.pythonhosted.org/packages/a6/79/8780a378c650e3df849b73de8b13cf5412f521ca2ff9b78a45c247029440/protobuf-6.33.1-cp310-abi3-win_amd64.whl", hash = "sha256:923aa6d27a92bf44394f6abf7ea0500f38769d4b07f4be41cb52bd8b1123b9ed", size = 436883, upload-time = "2025-11-13T16:44:09.222Z" },
+    { url = "https://files.pythonhosted.org/packages/cd/93/26213ff72b103ae55bb0d73e7fb91ea570ef407c3ab4fd2f1f27cac16044/protobuf-6.33.1-cp39-abi3-macosx_10_9_universal2.whl", hash = "sha256:fe34575f2bdde76ac429ec7b570235bf0c788883e70aee90068e9981806f2490", size = 427522, upload-time = "2025-11-13T16:44:10.475Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/32/df4a35247923393aa6b887c3b3244a8c941c32a25681775f96e2b418f90e/protobuf-6.33.1-cp39-abi3-manylinux2014_aarch64.whl", hash = "sha256:f8adba2e44cde2d7618996b3fc02341f03f5bc3f2748be72dc7b063319276178", size = 324445, upload-time = "2025-11-13T16:44:11.869Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/d0/d796e419e2ec93d2f3fa44888861c3f88f722cde02b7c3488fcc6a166820/protobuf-6.33.1-cp39-abi3-manylinux2014_s390x.whl", hash = "sha256:0f4cf01222c0d959c2b399142deb526de420be8236f22c71356e2a544e153c53", size = 339161, upload-time = "2025-11-13T16:44:12.778Z" },
+    { url = "https://files.pythonhosted.org/packages/1d/2a/3c5f05a4af06649547027d288747f68525755de692a26a7720dced3652c0/protobuf-6.33.1-cp39-abi3-manylinux2014_x86_64.whl", hash = "sha256:8fd7d5e0eb08cd5b87fd3df49bc193f5cfd778701f47e11d127d0afc6c39f1d1", size = 323171, upload-time = "2025-11-13T16:44:14.035Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/86/a801cbb316860004bd865b1ded691c53e41d4a8224e3e421f8394174aba7/protobuf-6.33.1-cp39-cp39-win32.whl", hash = "sha256:023af8449482fa884d88b4563d85e83accab54138ae098924a985bcbb734a213", size = 425689, upload-time = "2025-11-13T16:44:15.389Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/0f/77b5a12825d59af2596634f062eb1a472f44494965a05dcd97cb5daf3ae5/protobuf-6.33.1-cp39-cp39-win_amd64.whl", hash = "sha256:df051de4fd7e5e4371334e234c62ba43763f15ab605579e04c7008c05735cd82", size = 436877, upload-time = "2025-11-13T16:44:16.71Z" },
+    { url = "https://files.pythonhosted.org/packages/08/b4/46310463b4f6ceef310f8348786f3cff181cea671578e3d9743ba61a459e/protobuf-6.33.1-py3-none-any.whl", hash = "sha256:d595a9fd694fdeb061a62fbe10eb039cc1e444df81ec9bb70c7fc59ebcb1eafa", size = 170477, upload-time = "2025-11-13T16:44:17.633Z" },
 ]
 
 [[package]]
 name = "psutil"
-version = "7.1.0"
+version = "7.1.3"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/b3/31/4723d756b59344b643542936e37a31d1d3204bcdc42a7daa8ee9eb06fb50/psutil-7.1.0.tar.gz", hash = "sha256:655708b3c069387c8b77b072fc429a57d0e214221d01c0a772df7dfedcb3bcd2", size = 497660, upload-time = "2025-09-17T20:14:52.902Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/e1/88/bdd0a41e5857d5d703287598cbf08dad90aed56774ea52ae071bae9071b6/psutil-7.1.3.tar.gz", hash = "sha256:6c86281738d77335af7aec228328e944b30930899ea760ecf33a4dba66be5e74", size = 489059, upload-time = "2025-11-02T12:25:54.619Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/46/62/ce4051019ee20ce0ed74432dd73a5bb087a6704284a470bb8adff69a0932/psutil-7.1.0-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:76168cef4397494250e9f4e73eb3752b146de1dd950040b29186d0cce1d5ca13", size = 245242, upload-time = "2025-09-17T20:14:56.126Z" },
-    { url = "https://files.pythonhosted.org/packages/38/61/f76959fba841bf5b61123fbf4b650886dc4094c6858008b5bf73d9057216/psutil-7.1.0-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:5d007560c8c372efdff9e4579c2846d71de737e4605f611437255e81efcca2c5", size = 246682, upload-time = "2025-09-17T20:14:58.25Z" },
-    { url = "https://files.pythonhosted.org/packages/88/7a/37c99d2e77ec30d63398ffa6a660450b8a62517cabe44b3e9bae97696e8d/psutil-7.1.0-cp36-abi3-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:22e4454970b32472ce7deaa45d045b34d3648ce478e26a04c7e858a0a6e75ff3", size = 287994, upload-time = "2025-09-17T20:14:59.901Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/de/04c8c61232f7244aa0a4b9a9fbd63a89d5aeaf94b2fc9d1d16e2faa5cbb0/psutil-7.1.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8c70e113920d51e89f212dd7be06219a9b88014e63a4cec69b684c327bc474e3", size = 291163, upload-time = "2025-09-17T20:15:01.481Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/58/c4f976234bf6d4737bc8c02a81192f045c307b72cf39c9e5c5a2d78927f6/psutil-7.1.0-cp36-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7d4a113425c037300de3ac8b331637293da9be9713855c4fc9d2d97436d7259d", size = 293625, upload-time = "2025-09-17T20:15:04.492Z" },
-    { url = "https://files.pythonhosted.org/packages/79/87/157c8e7959ec39ced1b11cc93c730c4fb7f9d408569a6c59dbd92ceb35db/psutil-7.1.0-cp37-abi3-win32.whl", hash = "sha256:09ad740870c8d219ed8daae0ad3b726d3bf9a028a198e7f3080f6a1888b99bca", size = 244812, upload-time = "2025-09-17T20:15:07.462Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/e9/b44c4f697276a7a95b8e94d0e320a7bf7f3318521b23de69035540b39838/psutil-7.1.0-cp37-abi3-win_amd64.whl", hash = "sha256:57f5e987c36d3146c0dd2528cd42151cf96cd359b9d67cfff836995cc5df9a3d", size = 247965, upload-time = "2025-09-17T20:15:09.673Z" },
-    { url = "https://files.pythonhosted.org/packages/26/65/1070a6e3c036f39142c2820c4b52e9243246fcfc3f96239ac84472ba361e/psutil-7.1.0-cp37-abi3-win_arm64.whl", hash = "sha256:6937cb68133e7c97b6cc9649a570c9a18ba0efebed46d8c5dae4c07fa1b67a07", size = 244971, upload-time = "2025-09-17T20:15:12.262Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/94/46b9154a800253e7ecff5aaacdf8ebf43db99de4a2dfa18575b02548654e/psutil-7.1.3-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:2bdbcd0e58ca14996a42adf3621a6244f1bb2e2e528886959c72cf1e326677ab", size = 238359, upload-time = "2025-11-02T12:26:25.284Z" },
+    { url = "https://files.pythonhosted.org/packages/68/3a/9f93cff5c025029a36d9a92fef47220ab4692ee7f2be0fba9f92813d0cb8/psutil-7.1.3-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:bc31fa00f1fbc3c3802141eede66f3a2d51d89716a194bf2cd6fc68310a19880", size = 239171, upload-time = "2025-11-02T12:26:27.23Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/b1/5f49af514f76431ba4eea935b8ad3725cdeb397e9245ab919dbc1d1dc20f/psutil-7.1.3-cp36-abi3-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3bb428f9f05c1225a558f53e30ccbad9930b11c3fc206836242de1091d3e7dd3", size = 263261, upload-time = "2025-11-02T12:26:29.48Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/95/992c8816a74016eb095e73585d747e0a8ea21a061ed3689474fabb29a395/psutil-7.1.3-cp36-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:56d974e02ca2c8eb4812c3f76c30e28836fffc311d55d979f1465c1feeb2b68b", size = 264635, upload-time = "2025-11-02T12:26:31.74Z" },
+    { url = "https://files.pythonhosted.org/packages/55/4c/c3ed1a622b6ae2fd3c945a366e64eb35247a31e4db16cf5095e269e8eb3c/psutil-7.1.3-cp37-abi3-win_amd64.whl", hash = "sha256:f39c2c19fe824b47484b96f9692932248a54c43799a84282cfe58d05a6449efd", size = 247633, upload-time = "2025-11-02T12:26:33.887Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/ad/33b2ccec09bf96c2b2ef3f9a6f66baac8253d7565d8839e024a6b905d45d/psutil-7.1.3-cp37-abi3-win_arm64.whl", hash = "sha256:bd0d69cee829226a761e92f28140bec9a5ee9d5b4fb4b0cc589068dbfff559b1", size = 244608, upload-time = "2025-11-02T12:26:36.136Z" },
 ]
 
 [[package]]
@@ -3632,6 +3703,9 @@ wheels = [
 name = "pyarrow"
 version = "21.0.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/ef/c2/ea068b8f00905c06329a3dfcd40d0fcc2b7d0f2e355bdb25b65e0a0e4cd4/pyarrow-21.0.0.tar.gz", hash = "sha256:5051f2dccf0e283ff56335760cbc8622cf52264d67e359d5569541ac11b6d5bc", size = 1133487, upload-time = "2025-07-18T00:57:31.761Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/17/d9/110de31880016e2afc52d8580b397dbe47615defbf09ca8cf55f56c62165/pyarrow-21.0.0-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:e563271e2c5ff4d4a4cbeb2c83d5cf0d4938b891518e676025f7268c6fe5fe26", size = 31196837, upload-time = "2025-07-18T00:54:34.755Z" },
@@ -3664,6 +3738,40 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/10/15/6b30e77872012bbfe8265d42a01d5b3c17ef0ac0f2fae531ad91b6a6c02e/pyarrow-21.0.0-cp39-cp39-win_amd64.whl", hash = "sha256:cdc4c17afda4dab2a9c0b79148a43a7f4e1094916b3e18d8975bfd6d6d52241f", size = 26227521, upload-time = "2025-07-18T00:57:29.119Z" },
 ]
 
+[[package]]
+name = "pyarrow"
+version = "22.0.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/30/53/04a7fdc63e6056116c9ddc8b43bc28c12cdd181b85cbeadb79278475f3ae/pyarrow-22.0.0.tar.gz", hash = "sha256:3d600dc583260d845c7d8a6db540339dd883081925da2bd1c5cb808f720b3cd9", size = 1151151, upload-time = "2025-10-24T12:30:00.762Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d9/9b/cb3f7e0a345353def531ca879053e9ef6b9f38ed91aebcf68b09ba54dec0/pyarrow-22.0.0-cp310-cp310-macosx_12_0_arm64.whl", hash = "sha256:77718810bd3066158db1e95a63c160ad7ce08c6b0710bc656055033e39cdad88", size = 34223968, upload-time = "2025-10-24T10:03:31.21Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/41/3184b8192a120306270c5307f105b70320fdaa592c99843c5ef78aaefdcf/pyarrow-22.0.0-cp310-cp310-macosx_12_0_x86_64.whl", hash = "sha256:44d2d26cda26d18f7af7db71453b7b783788322d756e81730acb98f24eb90ace", size = 35942085, upload-time = "2025-10-24T10:03:38.146Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/3d/a1eab2f6f08001f9fb714b8ed5cfb045e2fe3e3e3c0c221f2c9ed1e6d67d/pyarrow-22.0.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:b9d71701ce97c95480fecb0039ec5bb889e75f110da72005743451339262f4ce", size = 44964613, upload-time = "2025-10-24T10:03:46.516Z" },
+    { url = "https://files.pythonhosted.org/packages/46/46/a1d9c24baf21cfd9ce994ac820a24608decf2710521b29223d4334985127/pyarrow-22.0.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:710624ab925dc2b05a6229d47f6f0dac1c1155e6ed559be7109f684eba048a48", size = 47627059, upload-time = "2025-10-24T10:03:55.353Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/4c/f711acb13075c1391fd54bc17e078587672c575f8de2a6e62509af026dcf/pyarrow-22.0.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:f963ba8c3b0199f9d6b794c90ec77545e05eadc83973897a4523c9e8d84e9340", size = 47947043, upload-time = "2025-10-24T10:04:05.408Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/70/1f3180dd7c2eab35c2aca2b29ace6c519f827dcd4cfeb8e0dca41612cf7a/pyarrow-22.0.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:bd0d42297ace400d8febe55f13fdf46e86754842b860c978dfec16f081e5c653", size = 50206505, upload-time = "2025-10-24T10:04:15.786Z" },
+    { url = "https://files.pythonhosted.org/packages/80/07/fea6578112c8c60ffde55883a571e4c4c6bc7049f119d6b09333b5cc6f73/pyarrow-22.0.0-cp310-cp310-win_amd64.whl", hash = "sha256:00626d9dc0f5ef3a75fe63fd68b9c7c8302d2b5bbc7f74ecaedba83447a24f84", size = 28101641, upload-time = "2025-10-24T10:04:22.57Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/b7/18f611a8cdc43417f9394a3ccd3eace2f32183c08b9eddc3d17681819f37/pyarrow-22.0.0-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:3e294c5eadfb93d78b0763e859a0c16d4051fc1c5231ae8956d61cb0b5666f5a", size = 34272022, upload-time = "2025-10-24T10:04:28.973Z" },
+    { url = "https://files.pythonhosted.org/packages/26/5c/f259e2526c67eb4b9e511741b19870a02363a47a35edbebc55c3178db22d/pyarrow-22.0.0-cp311-cp311-macosx_12_0_x86_64.whl", hash = "sha256:69763ab2445f632d90b504a815a2a033f74332997052b721002298ed6de40f2e", size = 35995834, upload-time = "2025-10-24T10:04:35.467Z" },
+    { url = "https://files.pythonhosted.org/packages/50/8d/281f0f9b9376d4b7f146913b26fac0aa2829cd1ee7e997f53a27411bbb92/pyarrow-22.0.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:b41f37cabfe2463232684de44bad753d6be08a7a072f6a83447eeaf0e4d2a215", size = 45030348, upload-time = "2025-10-24T10:04:43.366Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/e5/53c0a1c428f0976bf22f513d79c73000926cb00b9c138d8e02daf2102e18/pyarrow-22.0.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:35ad0f0378c9359b3f297299c3309778bb03b8612f987399a0333a560b43862d", size = 47699480, upload-time = "2025-10-24T10:04:51.486Z" },
+    { url = "https://files.pythonhosted.org/packages/95/e1/9dbe4c465c3365959d183e6345d0a8d1dc5b02ca3f8db4760b3bc834cf25/pyarrow-22.0.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8382ad21458075c2e66a82a29d650f963ce51c7708c7c0ff313a8c206c4fd5e8", size = 48011148, upload-time = "2025-10-24T10:04:59.585Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/b4/7caf5d21930061444c3cf4fa7535c82faf5263e22ce43af7c2759ceb5b8b/pyarrow-22.0.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:1a812a5b727bc09c3d7ea072c4eebf657c2f7066155506ba31ebf4792f88f016", size = 50276964, upload-time = "2025-10-24T10:05:08.175Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/f3/cec89bd99fa3abf826f14d4e53d3d11340ce6f6af4d14bdcd54cd83b6576/pyarrow-22.0.0-cp311-cp311-win_amd64.whl", hash = "sha256:ec5d40dd494882704fb876c16fa7261a69791e784ae34e6b5992e977bd2e238c", size = 28106517, upload-time = "2025-10-24T10:05:14.314Z" },
+    { url = "https://files.pythonhosted.org/packages/af/63/ba23862d69652f85b615ca14ad14f3bcfc5bf1b99ef3f0cd04ff93fdad5a/pyarrow-22.0.0-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:bea79263d55c24a32b0d79c00a1c58bb2ee5f0757ed95656b01c0fb310c5af3d", size = 34211578, upload-time = "2025-10-24T10:05:21.583Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/d0/f9ad86fe809efd2bcc8be32032fa72e8b0d112b01ae56a053006376c5930/pyarrow-22.0.0-cp312-cp312-macosx_12_0_x86_64.whl", hash = "sha256:12fe549c9b10ac98c91cf791d2945e878875d95508e1a5d14091a7aaa66d9cf8", size = 35989906, upload-time = "2025-10-24T10:05:29.485Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/a8/f910afcb14630e64d673f15904ec27dd31f1e009b77033c365c84e8c1e1d/pyarrow-22.0.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:334f900ff08ce0423407af97e6c26ad5d4e3b0763645559ece6fbf3747d6a8f5", size = 45021677, upload-time = "2025-10-24T10:05:38.274Z" },
+    { url = "https://files.pythonhosted.org/packages/13/95/aec81f781c75cd10554dc17a25849c720d54feafb6f7847690478dcf5ef8/pyarrow-22.0.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:c6c791b09c57ed76a18b03f2631753a4960eefbbca80f846da8baefc6491fcfe", size = 47726315, upload-time = "2025-10-24T10:05:47.314Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/d4/74ac9f7a54cfde12ee42734ea25d5a3c9a45db78f9def949307a92720d37/pyarrow-22.0.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:c3200cb41cdbc65156e5f8c908d739b0dfed57e890329413da2748d1a2cd1a4e", size = 47990906, upload-time = "2025-10-24T10:05:58.254Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/71/fedf2499bf7a95062eafc989ace56572f3343432570e1c54e6599d5b88da/pyarrow-22.0.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:ac93252226cf288753d8b46280f4edf3433bf9508b6977f8dd8526b521a1bbb9", size = 50306783, upload-time = "2025-10-24T10:06:08.08Z" },
+    { url = "https://files.pythonhosted.org/packages/68/ed/b202abd5a5b78f519722f3d29063dda03c114711093c1995a33b8e2e0f4b/pyarrow-22.0.0-cp312-cp312-win_amd64.whl", hash = "sha256:44729980b6c50a5f2bfcc2668d36c569ce17f8b17bccaf470c4313dcbbf13c9d", size = 27972883, upload-time = "2025-10-24T10:06:14.204Z" },
+]
+
 [[package]]
 name = "pyasn1"
 version = "0.6.1"
@@ -3685,102 +3793,9 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/47/8d/d529b5d697919ba8c11ad626e835d4039be708a35b0d22de83a269a6682c/pyasn1_modules-0.4.2-py3-none-any.whl", hash = "sha256:29253a9207ce32b64c3ac6600edc75368f98473906e8fd1043bd6b5b1de2c14a", size = 181259, upload-time = "2025-03-28T02:41:19.028Z" },
 ]
 
-[[package]]
-name = "pybase64"
-version = "1.4.2"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/04/14/43297a7b7f0c1bf0c00b596f754ee3ac946128c64d21047ccf9c9bbc5165/pybase64-1.4.2.tar.gz", hash = "sha256:46cdefd283ed9643315d952fe44de80dc9b9a811ce6e3ec97fd1827af97692d0", size = 137246, upload-time = "2025-07-27T13:08:57.808Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/1d/d8/30ea35dc2c8c568be93e1379efcaa35092e37efa2ce7f1985ccc63babee7/pybase64-1.4.2-cp310-cp310-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:1eef93c29cc5567480d168f9cc1ebd3fc3107c65787aed2019a8ea68575a33e0", size = 65963, upload-time = "2025-07-27T13:02:22.376Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/da/1c22f2a21d6bb9ec2a214d15ae02d5b20a95335de218a0ecbf769c535a5c/pybase64-1.4.2-cp310-cp310-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:17b871a34aaeb0644145cb6bf28feb163f593abea11aec3dbcc34a006edfc828", size = 68887, upload-time = "2025-07-27T13:02:23.606Z" },
-    { url = "https://files.pythonhosted.org/packages/ac/8d/e04d489ba99b444ce94b4d5b232365d00b0f0e8564275d7ba7434dcabe72/pybase64-1.4.2-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:1f734e16293637a35d282ce594eb05a7a90ea3ae2bc84a3496a5df9e6b890725", size = 57503, upload-time = "2025-07-27T13:02:24.83Z" },
-    { url = "https://files.pythonhosted.org/packages/7e/b8/5ec9c334f30cf898709a084d596bf4b47aec2e07870f07bac5cf39754eca/pybase64-1.4.2-cp310-cp310-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:22bd38db2d990d5545dde83511edeec366630d00679dbd945472315c09041dc6", size = 54517, upload-time = "2025-07-27T13:02:26.006Z" },
-    { url = "https://files.pythonhosted.org/packages/b9/5a/6e4424ecca041e53aa7c14525f99edd43d0117c23c5d9cb14e931458a536/pybase64-1.4.2-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:dc65cee686dda72007b7541b2014f33ee282459c781b9b61305bd8b9cfadc8e1", size = 57167, upload-time = "2025-07-27T13:02:27.47Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/d0/13f1a9467cf565eecc21dce89fb0723458d8c563d2ccfb99b96e8318dfd5/pybase64-1.4.2-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:1e79641c420a22e49c67c046895efad05bf5f8b1dbe0dd78b4af3ab3f2923fe2", size = 57718, upload-time = "2025-07-27T13:02:28.631Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/34/d80335c36ad9400b18b4f92e9f680cf7646102fe4919f7bce5786a2ccb7b/pybase64-1.4.2-cp310-cp310-manylinux_2_31_riscv64.whl", hash = "sha256:12f5e7db522ef780a8b333dab5f7d750d270b23a1684bc2235ba50756c7ba428", size = 53021, upload-time = "2025-07-27T13:02:29.823Z" },
-    { url = "https://files.pythonhosted.org/packages/68/57/504ff75f7c78df28be126fe6634083d28d7f84c17e04a74a7dcb50ab2377/pybase64-1.4.2-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:a618b1e1a63e75dd40c2a397d875935ed0835464dc55cb1b91e8f880113d0444", size = 56306, upload-time = "2025-07-27T13:02:31.314Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/bc/2d21cda8b73c8c9f5cd3d7e6e26dd6dfc96491052112f282332a3d5bf1d9/pybase64-1.4.2-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:89b0a51702c7746fa914e75e680ad697b979cdead6b418603f56a6fc9de2f50f", size = 50101, upload-time = "2025-07-27T13:02:32.662Z" },
-    { url = "https://files.pythonhosted.org/packages/88/6d/51942e7737bb0711ca3e55db53924fd7f07166d79da5508ab8f5fd5972a8/pybase64-1.4.2-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:c5161b8b82f8ba5dbbc3f76e0270622a2c2fdb9ffaf092d8f774ad7ec468c027", size = 66555, upload-time = "2025-07-27T13:02:34.122Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/c8/c46024d196402e7be4d3fad85336863a34816c3436c51fcf9c7c0781bf11/pybase64-1.4.2-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:2168de920c9b1e57850e9ff681852923a953601f73cc96a0742a42236695c316", size = 55684, upload-time = "2025-07-27T13:02:35.427Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/c5/953782c9d599ff5217ee87f19e317c494cd4840afcab4c48f99cb78ca201/pybase64-1.4.2-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:7a1e3dc977562abe40ab43483223013be71b215a5d5f3c78a666e70a5076eeec", size = 52475, upload-time = "2025-07-27T13:02:36.634Z" },
-    { url = "https://files.pythonhosted.org/packages/05/fb/57d36173631aab67ca4558cdbde1047fc67a09b77f9c53addd57c7e9fdd4/pybase64-1.4.2-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:4cf1e8a57449e48137ef4de00a005e24c3f1cffc0aafc488e36ceb5bb2cbb1da", size = 53943, upload-time = "2025-07-27T13:02:37.777Z" },
-    { url = "https://files.pythonhosted.org/packages/75/73/23e5bb0bffac0cabe2d11d1c618f6ef73da9f430da03c5249931e3c49b63/pybase64-1.4.2-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:d8e1a381ba124f26a93d5925efbf6e6c36287fc2c93d74958e8b677c30a53fc0", size = 68411, upload-time = "2025-07-27T13:02:39.302Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/8f/f9b49926a60848ba98350dd648227ec524fb78340b47a450c4dbaf24b1bb/pybase64-1.4.2-cp311-cp311-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:37f133e8c96427995480bb6d396d9d49e949a3e829591845bb6a5a7f215ca177", size = 68318, upload-time = "2025-07-27T13:02:46.644Z" },
-    { url = "https://files.pythonhosted.org/packages/29/9b/6ed2dd2bc8007f33b8316d6366b0901acbdd5665b419c2893b3dd48708de/pybase64-1.4.2-cp311-cp311-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:a6ee3874b0abbdd4c903d3989682a3f016fd84188622879f6f95a5dc5718d7e5", size = 71357, upload-time = "2025-07-27T13:02:47.937Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/69/be9ac8127da8d8339db7129683bd2975cecb0bf40a82731e1a492577a177/pybase64-1.4.2-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:5c69f177b1e404b22b05802127d6979acf4cb57f953c7de9472410f9c3fdece7", size = 59817, upload-time = "2025-07-27T13:02:49.163Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/a2/e3e09e000b509609276ee28b71beb0b61462d4a43b3e0db0a44c8652880c/pybase64-1.4.2-cp311-cp311-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:80c817e88ef2ca3cc9a285fde267690a1cb821ce0da4848c921c16f0fec56fda", size = 56639, upload-time = "2025-07-27T13:02:50.384Z" },
-    { url = "https://files.pythonhosted.org/packages/01/70/ad7eff88aa4f1be06db705812e1f01749606933bf8fe9df553bb04b703e6/pybase64-1.4.2-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:7a4bb6e7e45bfdaea0f2aaf022fc9a013abe6e46ccea31914a77e10f44098688", size = 59368, upload-time = "2025-07-27T13:02:51.883Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/82/0cd1b4bcd2a4da7805cfa04587be783bf9583b34ac16cadc29cf119a4fa2/pybase64-1.4.2-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:2710a80d41a2b41293cb0e5b84b5464f54aa3f28f7c43de88784d2d9702b8a1c", size = 59981, upload-time = "2025-07-27T13:02:53.16Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/4c/8029a03468307dfaf0f9694d31830487ee43af5f8a73407004907724e8ac/pybase64-1.4.2-cp311-cp311-manylinux_2_31_riscv64.whl", hash = "sha256:aa6122c8a81f6597e1c1116511f03ed42cf377c2100fe7debaae7ca62521095a", size = 54908, upload-time = "2025-07-27T13:02:54.363Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/8b/70bd0fe659e242efd0f60895a8ce1fe88e3a4084fd1be368974c561138c9/pybase64-1.4.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b7e22b02505d64db308e9feeb6cb52f1d554ede5983de0befa59ac2d2ffb6a5f", size = 58650, upload-time = "2025-07-27T13:02:55.905Z" },
-    { url = "https://files.pythonhosted.org/packages/64/ca/9c1d23cbc4b9beac43386a32ad53903c816063cef3f14c10d7c3d6d49a23/pybase64-1.4.2-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:edfe4a3c8c4007f09591f49b46a89d287ef5e8cd6630339536fe98ff077263c2", size = 52323, upload-time = "2025-07-27T13:02:57.192Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/29/a6292e9047248c8616dc53131a49da6c97a61616f80e1e36c73d7ef895fe/pybase64-1.4.2-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:b79b4a53dd117ffbd03e96953f2e6bd2827bfe11afeb717ea16d9b0893603077", size = 68979, upload-time = "2025-07-27T13:02:58.594Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/e0/cfec7b948e170395d8e88066e01f50e71195db9837151db10c14965d6222/pybase64-1.4.2-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:fd9afa7a61d89d170607faf22287290045757e782089f0357b8f801d228d52c3", size = 58037, upload-time = "2025-07-27T13:02:59.753Z" },
-    { url = "https://files.pythonhosted.org/packages/74/7e/0ac1850198c9c35ef631174009cee576f4d8afff3bf493ce310582976ab4/pybase64-1.4.2-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:5c17b092e4da677a595178d2db17a5d2fafe5c8e418d46c0c4e4cde5adb8cff3", size = 54416, upload-time = "2025-07-27T13:03:00.978Z" },
-    { url = "https://files.pythonhosted.org/packages/1b/45/b0b037f27e86c50e62d927f0bc1bde8b798dd55ab39197b116702e508d05/pybase64-1.4.2-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:120799274cf55f3f5bb8489eaa85142f26170564baafa7cf3e85541c46b6ab13", size = 56257, upload-time = "2025-07-27T13:03:02.201Z" },
-    { url = "https://files.pythonhosted.org/packages/d2/0d/5034598aac56336d88fd5aaf6f34630330643b51d399336b8c788d798fc5/pybase64-1.4.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:522e4e712686acec2d25de9759dda0b0618cb9f6588523528bc74715c0245c7b", size = 70889, upload-time = "2025-07-27T13:03:03.437Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/2a/a24c810e7a61d2cc6f73fe9ee4872a03030887fa8654150901b15f376f65/pybase64-1.4.2-cp312-cp312-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:f48c32ac6a16cbf57a5a96a073fef6ff7e3526f623cd49faa112b7f9980bafba", size = 68192, upload-time = "2025-07-27T13:03:11.467Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/87/d9baf98cbfc37b8657290ad4421f3a3c36aa0eafe4872c5859cfb52f3448/pybase64-1.4.2-cp312-cp312-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:ace8b23093a6bb862477080d9059b784096ab2f97541e8bfc40d42f062875149", size = 71587, upload-time = "2025-07-27T13:03:12.719Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/89/3df043cc56ef3b91b7aa0c26ae822a2d7ec8da0b0fd7c309c879b0eb5988/pybase64-1.4.2-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:1772c7532a7fb6301baea3dd3e010148dbf70cd1136a83c2f5f91bdc94822145", size = 59910, upload-time = "2025-07-27T13:03:14.266Z" },
-    { url = "https://files.pythonhosted.org/packages/75/4f/6641e9edf37aeb4d4524dc7ba2168eff8d96c90e77f6283c2be3400ab380/pybase64-1.4.2-cp312-cp312-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:f86f7faddcba5cbfea475f8ab96567834c28bf09ca6c7c3d66ee445adac80d8f", size = 56701, upload-time = "2025-07-27T13:03:15.6Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/7f/20d8ac1046f12420a0954a45a13033e75f98aade36eecd00c64e3549b071/pybase64-1.4.2-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:0b8c8e275b5294089f314814b4a50174ab90af79d6a4850f6ae11261ff6a7372", size = 59288, upload-time = "2025-07-27T13:03:16.823Z" },
-    { url = "https://files.pythonhosted.org/packages/17/ea/9c0ca570e3e50b3c6c3442e280c83b321a0464c86a9db1f982a4ff531550/pybase64-1.4.2-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:864d85a0470c615807ae8b97d724d068b940a2d10ac13a5f1b9e75a3ce441758", size = 60267, upload-time = "2025-07-27T13:03:18.132Z" },
-    { url = "https://files.pythonhosted.org/packages/f9/ac/46894929d71ccedebbfb0284173b0fea96bc029cd262654ba8451a7035d6/pybase64-1.4.2-cp312-cp312-manylinux_2_31_riscv64.whl", hash = "sha256:47254d97ed2d8351e30ecfdb9e2414547f66ba73f8a09f932c9378ff75cd10c5", size = 54801, upload-time = "2025-07-27T13:03:19.669Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/1e/02c95218ea964f0b2469717c2c69b48e63f4ca9f18af01a5b2a29e4c1216/pybase64-1.4.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:264b65ecc4f0ee73f3298ab83bbd8008f7f9578361b8df5b448f985d8c63e02a", size = 58599, upload-time = "2025-07-27T13:03:20.951Z" },
-    { url = "https://files.pythonhosted.org/packages/15/45/ccc21004930789b8fb439d43e3212a6c260ccddb2bf450c39a20db093f33/pybase64-1.4.2-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:fbcc2b30cd740c16c9699f596f22c7a9e643591311ae72b1e776f2d539e9dd9d", size = 52388, upload-time = "2025-07-27T13:03:23.064Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/45/22e46e549710c4c237d77785b6fb1bc4c44c288a5c44237ba9daf5c34b82/pybase64-1.4.2-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:cda9f79c22d51ee4508f5a43b673565f1d26af4330c99f114e37e3186fdd3607", size = 68802, upload-time = "2025-07-27T13:03:24.673Z" },
-    { url = "https://files.pythonhosted.org/packages/55/0c/232c6261b81296e5593549b36e6e7884a5da008776d12665923446322c36/pybase64-1.4.2-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:0c91c6d2a7232e2a1cd10b3b75a8bb657defacd4295a1e5e80455df2dfc84d4f", size = 57841, upload-time = "2025-07-27T13:03:25.948Z" },
-    { url = "https://files.pythonhosted.org/packages/20/8a/b35a615ae6f04550d696bb179c414538b3b477999435fdd4ad75b76139e4/pybase64-1.4.2-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:a370dea7b1cee2a36a4d5445d4e09cc243816c5bc8def61f602db5a6f5438e52", size = 54320, upload-time = "2025-07-27T13:03:27.495Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/a9/8bd4f9bcc53689f1b457ecefed1eaa080e4949d65a62c31a38b7253d5226/pybase64-1.4.2-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:9aa4de83f02e462a6f4e066811c71d6af31b52d7484de635582d0e3ec3d6cc3e", size = 56482, upload-time = "2025-07-27T13:03:28.942Z" },
-    { url = "https://files.pythonhosted.org/packages/75/e5/4a7735b54a1191f61c3f5c2952212c85c2d6b06eb5fb3671c7603395f70c/pybase64-1.4.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:83a1c2f9ed00fee8f064d548c8654a480741131f280e5750bb32475b7ec8ee38", size = 70959, upload-time = "2025-07-27T13:03:30.171Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/93/fd6f0cb1356ef849cef347ae3192a91c5beb944d485b2b97fd760265b7ad/pybase64-1.4.2-cp39-cp39-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:2089a72b04a62f63e0eac202ecff4440fb52fd05cd5f4ab9fe7e07839fedb9e9", size = 65747, upload-time = "2025-07-27T13:06:39.216Z" },
-    { url = "https://files.pythonhosted.org/packages/57/fc/75f8567866efcbc1c984e1bfb966ce92790fc38d5197f2d01a6131cb77f8/pybase64-1.4.2-cp39-cp39-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:bad101c24dcd23ed6fd6ea24c4a1b36ac7abc5eb07447dd7fa98b33859aed871", size = 68664, upload-time = "2025-07-27T13:06:41.908Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/19/8dbd3b29d54cce1ee82a0da1ff7473e2958057932af5788aa55f6f4d4ab5/pybase64-1.4.2-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:28592c88a9cf6fd27c9f191fb41688c1c27f57493d874cbc50e72e1cc2a3b854", size = 57325, upload-time = "2025-07-27T13:06:44.271Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/f7/e20978705feb84242c402ce7e626d9e8f7de638d2a6abf60e1384769d7f1/pybase64-1.4.2-cp39-cp39-manylinux2014_armv7l.manylinux_2_17_armv7l.whl", hash = "sha256:0b5639fa2ceb3095393bd56dca8c16079717c361dd3a75439c9a8b8d679f4cf0", size = 54286, upload-time = "2025-07-27T13:06:46.974Z" },
-    { url = "https://files.pythonhosted.org/packages/39/f5/815813fa85168b3afcc56eb085ef7acb8f03f36e1b470837b98864dc847b/pybase64-1.4.2-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:49630338d4c321336d0dfc4c2c23162a87d9ebc8bb8879348ae019ac8a4366de", size = 56951, upload-time = "2025-07-27T13:06:49.582Z" },
-    { url = "https://files.pythonhosted.org/packages/4f/eb/dc61ec538c6fa2cf86318e7e393fa4772cc363f5bc13a760bc241ef3f942/pybase64-1.4.2-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:c3d9f9881d7315e1d04d72aa7b3f40e2059bdbfdcec51939016409417725c952", size = 57442, upload-time = "2025-07-27T13:06:52.347Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/80/9d7286b25ff9952dcedd84456966916429d2bdb89cc826f7a11e6c9e8507/pybase64-1.4.2-cp39-cp39-manylinux_2_31_riscv64.whl", hash = "sha256:8e1226939eac9ce1f080d1b0a8afafee3140e277a4c40ccb306d82de396a41a8", size = 52774, upload-time = "2025-07-27T13:06:54.863Z" },
-    { url = "https://files.pythonhosted.org/packages/c9/5e/7bd80d6c8ebb88918749a0a177e0328a0954b8d8f35dbd55c3bc395ed050/pybase64-1.4.2-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:69f424a227ec503742bac69b89e232c474dc199cd98c3e58e91020c1c4bad0ad", size = 56061, upload-time = "2025-07-27T13:06:57.547Z" },
-    { url = "https://files.pythonhosted.org/packages/ae/29/a2e7b8b90acbeb3d8fc58589212ee6124fb102d6db6e52c37d76f5f679d9/pybase64-1.4.2-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:171ae85837de14d3691d5c4f29f5bb551209930c063a2cab6f5feb270aec66db", size = 49865, upload-time = "2025-07-27T13:07:00.076Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/ed/2d02e78c4f77ae70210a16399c734f8adfbfb4247c37a6924ae0285b23b3/pybase64-1.4.2-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:a55a13493fd165c3a619080149eda6f31c05c04c0577da9c9ef63d23f3abf374", size = 66341, upload-time = "2025-07-27T13:07:02.9Z" },
-    { url = "https://files.pythonhosted.org/packages/64/44/8ba4a9ec8a4bbb7b36068c6b4f85f6443960456f60ddd694d92de39d5b55/pybase64-1.4.2-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:06801fdc7fa83eac5cb7d1c7051bb623a25af8cb40e088671fa51a393d1053ad", size = 55470, upload-time = "2025-07-27T13:07:05.82Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/2f/d2b233ce4cefe80c5dbaedd8f2d11860ae2ef0b00233f70d2c7a46e59974/pybase64-1.4.2-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:7f2fbd6870228e9c8c3e2e2622ed7615a8d0159125b85e9d6c2d8e9ead74cdf0", size = 52242, upload-time = "2025-07-27T13:07:08.923Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/40/bcabd4398565a15329d93588aa0f15af5d0ae02a05f3388937bebb56846e/pybase64-1.4.2-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:1726017f04da880d10a57f078d117fe62532b5ed7bd58bd3318f3364b9767d91", size = 53739, upload-time = "2025-07-27T13:07:11.825Z" },
-    { url = "https://files.pythonhosted.org/packages/59/47/7f947e3efe660a8dc14c8c956e8165e8631533cf2dc1bbd3920f8028a4b0/pybase64-1.4.2-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1264f7fa417de7183732761f37c8ceb4652662a84f04538a28dadd5d84bf9a4a", size = 68149, upload-time = "2025-07-27T13:07:14.372Z" },
-    { url = "https://files.pythonhosted.org/packages/51/34/f40d3262c3953814b9bcdcf858436bd5bc1133a698be4bcc7ed2a8c0730d/pybase64-1.4.2-graalpy311-graalpy242_311_native-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:eef9255d926c64e2fca021d3aee98023bacb98e1518e5986d6aab04102411b04", size = 43212, upload-time = "2025-07-27T13:07:31.327Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/2a/5e05d25718cb8ffd68bd46553ddfd2b660893d937feda1716b8a3b21fb38/pybase64-1.4.2-graalpy311-graalpy242_311_native-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:89614ea2d2329b6708746c540e0f14d692125df99fb1203ff0de948d9e68dfc9", size = 35789, upload-time = "2025-07-27T13:07:34.026Z" },
-    { url = "https://files.pythonhosted.org/packages/cf/da/2b7e69abfc62abe4d54b10d1e09ec78021a6b9b2d7e6e7b632243a19433e/pybase64-1.4.2-pp310-pypy310_pp73-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:4f98c5c6152d3c01d933fcde04322cd9ddcf65b5346034aac69a04c1a7cbb012", size = 40667, upload-time = "2025-07-27T13:07:46.715Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/11/ba738655fb3ba85c7a0605eddd2709fef606e654840c72ee5c5ff7ab29bf/pybase64-1.4.2-pp310-pypy310_pp73-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:9096a4977b7aff7ef250f759fb6a4b6b7b6199d99c84070c7fc862dd3b208b34", size = 41290, upload-time = "2025-07-27T13:07:49.534Z" },
-    { url = "https://files.pythonhosted.org/packages/5d/38/2d5502fcaf712297b95c1b6ca924656dd7d17501fd7f9c9e0b3bbf8892ef/pybase64-1.4.2-pp310-pypy310_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:49d8597e2872966399410502310b1e2a5b7e8d8ba96766ee1fe242e00bd80775", size = 35438, upload-time = "2025-07-27T13:07:52.327Z" },
-    { url = "https://files.pythonhosted.org/packages/aa/dd/c819003b59b2832256b72ad23cbeadbd95d083ef0318d07149a58b7a88af/pybase64-1.4.2-pp311-pypy311_pp73-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:1159e70cba8e76c3d8f334bd1f8fd52a1bb7384f4c3533831b23ab2df84a6ef3", size = 40668, upload-time = "2025-07-27T13:08:04.176Z" },
-    { url = "https://files.pythonhosted.org/packages/0e/c5/38c6aba28678c4a4db49312a6b8171b93a0ffe9f21362cf4c0f325caa850/pybase64-1.4.2-pp311-pypy311_pp73-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7d943bc5dad8388971494554b97f22ae06a46cc7779ad0de3d4bfdf7d0bbea30", size = 41281, upload-time = "2025-07-27T13:08:07.395Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/23/5927bd9e59714e4e8cefd1d21ccd7216048bb1c6c3e7104b1b200afdc63d/pybase64-1.4.2-pp311-pypy311_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:10b99182c561d86422c5de4265fd1f8f172fb38efaed9d72c71fb31e279a7f94", size = 35433, upload-time = "2025-07-27T13:08:10.551Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/0a/7d19d99844f6cd4d67119a7d781e7997a5254dab86c47736c4b138f4e148/pybase64-1.4.2-pp39-pypy39_pp73-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:45f078139d76194024e59b4bcfa64d42e5a5f8a5a4ea55ca4d27df46989c5e32", size = 40661, upload-time = "2025-07-27T13:08:43.922Z" },
-    { url = "https://files.pythonhosted.org/packages/2e/78/20e0271084d3687e33c3858520c8808ad47128d3153e0f760226f36abf88/pybase64-1.4.2-pp39-pypy39_pp73-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:06305e602f128b289b98490a2d07d9d78e7e781e32e7b0252c2e71084fd19edf", size = 41286, upload-time = "2025-07-27T13:08:47.264Z" },
-    { url = "https://files.pythonhosted.org/packages/7c/b1/b8120705861192e9ffb2c73d77b581607dbfe0d41d81e086744b4f134d6e/pybase64-1.4.2-pp39-pypy39_pp73-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d58eb4cb50b6466cef2e25761a5c915a8d57feda53165cced537a7ce0421b928", size = 35432, upload-time = "2025-07-27T13:08:50.798Z" },
-]
-
-[[package]]
-name = "pycountry"
-version = "24.6.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/76/57/c389fa68c50590881a75b7883eeb3dc15e9e73a0fdc001cdd45c13290c92/pycountry-24.6.1.tar.gz", hash = "sha256:b61b3faccea67f87d10c1f2b0fc0be714409e8fcdcc1315613174f6466c10221", size = 6043910, upload-time = "2024-06-01T04:12:15.05Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/b1/ec/1fb891d8a2660716aadb2143235481d15ed1cbfe3ad669194690b0604492/pycountry-24.6.1-py3-none-any.whl", hash = "sha256:f1a4fb391cd7214f8eefd39556d740adcc233c778a27f8942c8dca351d6ce06f", size = 6335189, upload-time = "2024-06-01T04:11:49.711Z" },
-]
-
-[[package]]
-name = "pycparser"
-version = "2.23"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/fe/cf/d2d3b9f5699fb1e4615c8e32ff220203e43b248e1dfcc6736ad9057731ca/pycparser-2.23.tar.gz", hash = "sha256:78816d4f24add8f10a06d6f05b4d424ad9e96cfebf68a4ddc99c65c0720d00c2", size = 173734, upload-time = "2025-09-09T13:23:47.91Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/a0/e3/59cd50310fc9b59512193629e1984c1f95e5c8ae6e5d8c69532ccc65a7fe/pycparser-2.23-py3-none-any.whl", hash = "sha256:e5c6e8d3fbad53479cab09ac03729e0a9faf2bee3db8208a550daf5af81a5934", size = 118140, upload-time = "2025-09-09T13:23:46.651Z" },
-]
-
 [[package]]
 name = "pydantic"
-version = "2.11.10"
+version = "2.12.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "annotated-types" },
@@ -3788,138 +3803,134 @@ dependencies = [
     { name = "typing-extensions" },
     { name = "typing-inspection" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ae/54/ecab642b3bed45f7d5f59b38443dcb36ef50f85af192e6ece103dbfe9587/pydantic-2.11.10.tar.gz", hash = "sha256:dc280f0982fbda6c38fada4e476dc0a4f3aeaf9c6ad4c28df68a666ec3c61423", size = 788494, upload-time = "2025-10-04T10:40:41.338Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/96/ad/a17bc283d7d81837c061c49e3eaa27a45991759a1b7eae1031921c6bd924/pydantic-2.12.4.tar.gz", hash = "sha256:0f8cb9555000a4b5b617f66bfd2566264c4984b27589d3b845685983e8ea85ac", size = 821038, upload-time = "2025-11-05T10:50:08.59Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/bd/1f/73c53fcbfb0b5a78f91176df41945ca466e71e9d9d836e5c522abda39ee7/pydantic-2.11.10-py3-none-any.whl", hash = "sha256:802a655709d49bd004c31e865ef37da30b540786a46bfce02333e0e24b5fe29a", size = 444823, upload-time = "2025-10-04T10:40:39.055Z" },
-]
-
-[package.optional-dependencies]
-email = [
-    { name = "email-validator" },
+    { url = "https://files.pythonhosted.org/packages/82/2f/e68750da9b04856e2a7ec56fc6f034a5a79775e9b9a81882252789873798/pydantic-2.12.4-py3-none-any.whl", hash = "sha256:92d3d202a745d46f9be6df459ac5a064fdaa3c1c4cd8adcfa332ccf3c05f871e", size = 463400, upload-time = "2025-11-05T10:50:06.732Z" },
 ]
 
 [[package]]
 name = "pydantic-core"
-version = "2.33.2"
+version = "2.41.5"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/ad/88/5f2260bdfae97aabf98f1778d43f69574390ad787afb646292a638c923d4/pydantic_core-2.33.2.tar.gz", hash = "sha256:7cb8bc3605c29176e1b105350d2e6474142d7c1bd1d9327c4a9bdb46bf827acc", size = 435195, upload-time = "2025-04-23T18:33:52.104Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/e5/92/b31726561b5dae176c2d2c2dc43a9c5bfba5d32f96f8b4c0a600dd492447/pydantic_core-2.33.2-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:2b3d326aaef0c0399d9afffeb6367d5e26ddc24d351dbc9c636840ac355dc5d8", size = 2028817, upload-time = "2025-04-23T18:30:43.919Z" },
-    { url = "https://files.pythonhosted.org/packages/a3/44/3f0b95fafdaca04a483c4e685fe437c6891001bf3ce8b2fded82b9ea3aa1/pydantic_core-2.33.2-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:0e5b2671f05ba48b94cb90ce55d8bdcaaedb8ba00cc5359f6810fc918713983d", size = 1861357, upload-time = "2025-04-23T18:30:46.372Z" },
-    { url = "https://files.pythonhosted.org/packages/30/97/e8f13b55766234caae05372826e8e4b3b96e7b248be3157f53237682e43c/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0069c9acc3f3981b9ff4cdfaf088e98d83440a4c7ea1bc07460af3d4dc22e72d", size = 1898011, upload-time = "2025-04-23T18:30:47.591Z" },
-    { url = "https://files.pythonhosted.org/packages/9b/a3/99c48cf7bafc991cc3ee66fd544c0aae8dc907b752f1dad2d79b1b5a471f/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:d53b22f2032c42eaaf025f7c40c2e3b94568ae077a606f006d206a463bc69572", size = 1982730, upload-time = "2025-04-23T18:30:49.328Z" },
-    { url = "https://files.pythonhosted.org/packages/de/8e/a5b882ec4307010a840fb8b58bd9bf65d1840c92eae7534c7441709bf54b/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0405262705a123b7ce9f0b92f123334d67b70fd1f20a9372b907ce1080c7ba02", size = 2136178, upload-time = "2025-04-23T18:30:50.907Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/bb/71e35fc3ed05af6834e890edb75968e2802fe98778971ab5cba20a162315/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:4b25d91e288e2c4e0662b8038a28c6a07eaac3e196cfc4ff69de4ea3db992a1b", size = 2736462, upload-time = "2025-04-23T18:30:52.083Z" },
-    { url = "https://files.pythonhosted.org/packages/31/0d/c8f7593e6bc7066289bbc366f2235701dcbebcd1ff0ef8e64f6f239fb47d/pydantic_core-2.33.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6bdfe4b3789761f3bcb4b1ddf33355a71079858958e3a552f16d5af19768fef2", size = 2005652, upload-time = "2025-04-23T18:30:53.389Z" },
-    { url = "https://files.pythonhosted.org/packages/d2/7a/996d8bd75f3eda405e3dd219ff5ff0a283cd8e34add39d8ef9157e722867/pydantic_core-2.33.2-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:efec8db3266b76ef9607c2c4c419bdb06bf335ae433b80816089ea7585816f6a", size = 2113306, upload-time = "2025-04-23T18:30:54.661Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/84/daf2a6fb2db40ffda6578a7e8c5a6e9c8affb251a05c233ae37098118788/pydantic_core-2.33.2-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:031c57d67ca86902726e0fae2214ce6770bbe2f710dc33063187a68744a5ecac", size = 2073720, upload-time = "2025-04-23T18:30:56.11Z" },
-    { url = "https://files.pythonhosted.org/packages/77/fb/2258da019f4825128445ae79456a5499c032b55849dbd5bed78c95ccf163/pydantic_core-2.33.2-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:f8de619080e944347f5f20de29a975c2d815d9ddd8be9b9b7268e2e3ef68605a", size = 2244915, upload-time = "2025-04-23T18:30:57.501Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/7a/925ff73756031289468326e355b6fa8316960d0d65f8b5d6b3a3e7866de7/pydantic_core-2.33.2-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:73662edf539e72a9440129f231ed3757faab89630d291b784ca99237fb94db2b", size = 2241884, upload-time = "2025-04-23T18:30:58.867Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/b0/249ee6d2646f1cdadcb813805fe76265745c4010cf20a8eba7b0e639d9b2/pydantic_core-2.33.2-cp310-cp310-win32.whl", hash = "sha256:0a39979dcbb70998b0e505fb1556a1d550a0781463ce84ebf915ba293ccb7e22", size = 1910496, upload-time = "2025-04-23T18:31:00.078Z" },
-    { url = "https://files.pythonhosted.org/packages/66/ff/172ba8f12a42d4b552917aa65d1f2328990d3ccfc01d5b7c943ec084299f/pydantic_core-2.33.2-cp310-cp310-win_amd64.whl", hash = "sha256:b0379a2b24882fef529ec3b4987cb5d003b9cda32256024e6fe1586ac45fc640", size = 1955019, upload-time = "2025-04-23T18:31:01.335Z" },
-    { url = "https://files.pythonhosted.org/packages/3f/8d/71db63483d518cbbf290261a1fc2839d17ff89fce7089e08cad07ccfce67/pydantic_core-2.33.2-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:4c5b0a576fb381edd6d27f0a85915c6daf2f8138dc5c267a57c08a62900758c7", size = 2028584, upload-time = "2025-04-23T18:31:03.106Z" },
-    { url = "https://files.pythonhosted.org/packages/24/2f/3cfa7244ae292dd850989f328722d2aef313f74ffc471184dc509e1e4e5a/pydantic_core-2.33.2-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:e799c050df38a639db758c617ec771fd8fb7a5f8eaaa4b27b101f266b216a246", size = 1855071, upload-time = "2025-04-23T18:31:04.621Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/d3/4ae42d33f5e3f50dd467761304be2fa0a9417fbf09735bc2cce003480f2a/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dc46a01bf8d62f227d5ecee74178ffc448ff4e5197c756331f71efcc66dc980f", size = 1897823, upload-time = "2025-04-23T18:31:06.377Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/f3/aa5976e8352b7695ff808599794b1fba2a9ae2ee954a3426855935799488/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:a144d4f717285c6d9234a66778059f33a89096dfb9b39117663fd8413d582dcc", size = 1983792, upload-time = "2025-04-23T18:31:07.93Z" },
-    { url = "https://files.pythonhosted.org/packages/d5/7a/cda9b5a23c552037717f2b2a5257e9b2bfe45e687386df9591eff7b46d28/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:73cf6373c21bc80b2e0dc88444f41ae60b2f070ed02095754eb5a01df12256de", size = 2136338, upload-time = "2025-04-23T18:31:09.283Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/9f/b8f9ec8dd1417eb9da784e91e1667d58a2a4a7b7b34cf4af765ef663a7e5/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3dc625f4aa79713512d1976fe9f0bc99f706a9dee21dfd1810b4bbbf228d0e8a", size = 2730998, upload-time = "2025-04-23T18:31:11.7Z" },
-    { url = "https://files.pythonhosted.org/packages/47/bc/cd720e078576bdb8255d5032c5d63ee5c0bf4b7173dd955185a1d658c456/pydantic_core-2.33.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:881b21b5549499972441da4758d662aeea93f1923f953e9cbaff14b8b9565aef", size = 2003200, upload-time = "2025-04-23T18:31:13.536Z" },
-    { url = "https://files.pythonhosted.org/packages/ca/22/3602b895ee2cd29d11a2b349372446ae9727c32e78a94b3d588a40fdf187/pydantic_core-2.33.2-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:bdc25f3681f7b78572699569514036afe3c243bc3059d3942624e936ec93450e", size = 2113890, upload-time = "2025-04-23T18:31:15.011Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/e6/e3c5908c03cf00d629eb38393a98fccc38ee0ce8ecce32f69fc7d7b558a7/pydantic_core-2.33.2-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:fe5b32187cbc0c862ee201ad66c30cf218e5ed468ec8dc1cf49dec66e160cc4d", size = 2073359, upload-time = "2025-04-23T18:31:16.393Z" },
-    { url = "https://files.pythonhosted.org/packages/12/e7/6a36a07c59ebefc8777d1ffdaf5ae71b06b21952582e4b07eba88a421c79/pydantic_core-2.33.2-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:bc7aee6f634a6f4a95676fcb5d6559a2c2a390330098dba5e5a5f28a2e4ada30", size = 2245883, upload-time = "2025-04-23T18:31:17.892Z" },
-    { url = "https://files.pythonhosted.org/packages/16/3f/59b3187aaa6cc0c1e6616e8045b284de2b6a87b027cce2ffcea073adf1d2/pydantic_core-2.33.2-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:235f45e5dbcccf6bd99f9f472858849f73d11120d76ea8707115415f8e5ebebf", size = 2241074, upload-time = "2025-04-23T18:31:19.205Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/ed/55532bb88f674d5d8f67ab121a2a13c385df382de2a1677f30ad385f7438/pydantic_core-2.33.2-cp311-cp311-win32.whl", hash = "sha256:6368900c2d3ef09b69cb0b913f9f8263b03786e5b2a387706c5afb66800efd51", size = 1910538, upload-time = "2025-04-23T18:31:20.541Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/1b/25b7cccd4519c0b23c2dd636ad39d381abf113085ce4f7bec2b0dc755eb1/pydantic_core-2.33.2-cp311-cp311-win_amd64.whl", hash = "sha256:1e063337ef9e9820c77acc768546325ebe04ee38b08703244c1309cccc4f1bab", size = 1952909, upload-time = "2025-04-23T18:31:22.371Z" },
-    { url = "https://files.pythonhosted.org/packages/49/a9/d809358e49126438055884c4366a1f6227f0f84f635a9014e2deb9b9de54/pydantic_core-2.33.2-cp311-cp311-win_arm64.whl", hash = "sha256:6b99022f1d19bc32a4c2a0d544fc9a76e3be90f0b3f4af413f87d38749300e65", size = 1897786, upload-time = "2025-04-23T18:31:24.161Z" },
-    { url = "https://files.pythonhosted.org/packages/18/8a/2b41c97f554ec8c71f2a8a5f85cb56a8b0956addfe8b0efb5b3d77e8bdc3/pydantic_core-2.33.2-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:a7ec89dc587667f22b6a0b6579c249fca9026ce7c333fc142ba42411fa243cdc", size = 2009000, upload-time = "2025-04-23T18:31:25.863Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/02/6224312aacb3c8ecbaa959897af57181fb6cf3a3d7917fd44d0f2917e6f2/pydantic_core-2.33.2-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3c6db6e52c6d70aa0d00d45cdb9b40f0433b96380071ea80b09277dba021ddf7", size = 1847996, upload-time = "2025-04-23T18:31:27.341Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/46/6dcdf084a523dbe0a0be59d054734b86a981726f221f4562aed313dbcb49/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e61206137cbc65e6d5256e1166f88331d3b6238e082d9f74613b9b765fb9025", size = 1880957, upload-time = "2025-04-23T18:31:28.956Z" },
-    { url = "https://files.pythonhosted.org/packages/ec/6b/1ec2c03837ac00886ba8160ce041ce4e325b41d06a034adbef11339ae422/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:eb8c529b2819c37140eb51b914153063d27ed88e3bdc31b71198a198e921e011", size = 1964199, upload-time = "2025-04-23T18:31:31.025Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/1d/6bf34d6adb9debd9136bd197ca72642203ce9aaaa85cfcbfcf20f9696e83/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c52b02ad8b4e2cf14ca7b3d918f3eb0ee91e63b3167c32591e57c4317e134f8f", size = 2120296, upload-time = "2025-04-23T18:31:32.514Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/94/2bd0aaf5a591e974b32a9f7123f16637776c304471a0ab33cf263cf5591a/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:96081f1605125ba0855dfda83f6f3df5ec90c61195421ba72223de35ccfb2f88", size = 2676109, upload-time = "2025-04-23T18:31:33.958Z" },
-    { url = "https://files.pythonhosted.org/packages/f9/41/4b043778cf9c4285d59742281a769eac371b9e47e35f98ad321349cc5d61/pydantic_core-2.33.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8f57a69461af2a5fa6e6bbd7a5f60d3b7e6cebb687f55106933188e79ad155c1", size = 2002028, upload-time = "2025-04-23T18:31:39.095Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/d5/7bb781bf2748ce3d03af04d5c969fa1308880e1dca35a9bd94e1a96a922e/pydantic_core-2.33.2-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:572c7e6c8bb4774d2ac88929e3d1f12bc45714ae5ee6d9a788a9fb35e60bb04b", size = 2100044, upload-time = "2025-04-23T18:31:41.034Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/36/def5e53e1eb0ad896785702a5bbfd25eed546cdcf4087ad285021a90ed53/pydantic_core-2.33.2-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:db4b41f9bd95fbe5acd76d89920336ba96f03e149097365afe1cb092fceb89a1", size = 2058881, upload-time = "2025-04-23T18:31:42.757Z" },
-    { url = "https://files.pythonhosted.org/packages/01/6c/57f8d70b2ee57fc3dc8b9610315949837fa8c11d86927b9bb044f8705419/pydantic_core-2.33.2-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:fa854f5cf7e33842a892e5c73f45327760bc7bc516339fda888c75ae60edaeb6", size = 2227034, upload-time = "2025-04-23T18:31:44.304Z" },
-    { url = "https://files.pythonhosted.org/packages/27/b9/9c17f0396a82b3d5cbea4c24d742083422639e7bb1d5bf600e12cb176a13/pydantic_core-2.33.2-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:5f483cfb75ff703095c59e365360cb73e00185e01aaea067cd19acffd2ab20ea", size = 2234187, upload-time = "2025-04-23T18:31:45.891Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/6a/adf5734ffd52bf86d865093ad70b2ce543415e0e356f6cacabbc0d9ad910/pydantic_core-2.33.2-cp312-cp312-win32.whl", hash = "sha256:9cb1da0f5a471435a7bc7e439b8a728e8b61e59784b2af70d7c169f8dd8ae290", size = 1892628, upload-time = "2025-04-23T18:31:47.819Z" },
-    { url = "https://files.pythonhosted.org/packages/43/e4/5479fecb3606c1368d496a825d8411e126133c41224c1e7238be58b87d7e/pydantic_core-2.33.2-cp312-cp312-win_amd64.whl", hash = "sha256:f941635f2a3d96b2973e867144fde513665c87f13fe0e193c158ac51bfaaa7b2", size = 1955866, upload-time = "2025-04-23T18:31:49.635Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/24/8b11e8b3e2be9dd82df4b11408a67c61bb4dc4f8e11b5b0fc888b38118b5/pydantic_core-2.33.2-cp312-cp312-win_arm64.whl", hash = "sha256:cca3868ddfaccfbc4bfb1d608e2ccaaebe0ae628e1416aeb9c4d88c001bb45ab", size = 1888894, upload-time = "2025-04-23T18:31:51.609Z" },
-    { url = "https://files.pythonhosted.org/packages/53/ea/bbe9095cdd771987d13c82d104a9c8559ae9aec1e29f139e286fd2e9256e/pydantic_core-2.33.2-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:a2b911a5b90e0374d03813674bf0a5fbbb7741570dcd4b4e85a2e48d17def29d", size = 2028677, upload-time = "2025-04-23T18:32:27.227Z" },
-    { url = "https://files.pythonhosted.org/packages/49/1d/4ac5ed228078737d457a609013e8f7edc64adc37b91d619ea965758369e5/pydantic_core-2.33.2-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:6fa6dfc3e4d1f734a34710f391ae822e0a8eb8559a85c6979e14e65ee6ba2954", size = 1864735, upload-time = "2025-04-23T18:32:29.019Z" },
-    { url = "https://files.pythonhosted.org/packages/23/9a/2e70d6388d7cda488ae38f57bc2f7b03ee442fbcf0d75d848304ac7e405b/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c54c939ee22dc8e2d545da79fc5381f1c020d6d3141d3bd747eab59164dc89fb", size = 1898467, upload-time = "2025-04-23T18:32:31.119Z" },
-    { url = "https://files.pythonhosted.org/packages/ff/2e/1568934feb43370c1ffb78a77f0baaa5a8b6897513e7a91051af707ffdc4/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:53a57d2ed685940a504248187d5685e49eb5eef0f696853647bf37c418c538f7", size = 1983041, upload-time = "2025-04-23T18:32:33.655Z" },
-    { url = "https://files.pythonhosted.org/packages/01/1a/1a1118f38ab64eac2f6269eb8c120ab915be30e387bb561e3af904b12499/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:09fb9dd6571aacd023fe6aaca316bd01cf60ab27240d7eb39ebd66a3a15293b4", size = 2136503, upload-time = "2025-04-23T18:32:35.519Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/da/44754d1d7ae0f22d6d3ce6c6b1486fc07ac2c524ed8f6eca636e2e1ee49b/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0e6116757f7959a712db11f3e9c0a99ade00a5bbedae83cb801985aa154f071b", size = 2736079, upload-time = "2025-04-23T18:32:37.659Z" },
-    { url = "https://files.pythonhosted.org/packages/4d/98/f43cd89172220ec5aa86654967b22d862146bc4d736b1350b4c41e7c9c03/pydantic_core-2.33.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8d55ab81c57b8ff8548c3e4947f119551253f4e3787a7bbc0b6b3ca47498a9d3", size = 2006508, upload-time = "2025-04-23T18:32:39.637Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/cc/f77e8e242171d2158309f830f7d5d07e0531b756106f36bc18712dc439df/pydantic_core-2.33.2-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c20c462aa4434b33a2661701b861604913f912254e441ab8d78d30485736115a", size = 2113693, upload-time = "2025-04-23T18:32:41.818Z" },
-    { url = "https://files.pythonhosted.org/packages/54/7a/7be6a7bd43e0a47c147ba7fbf124fe8aaf1200bc587da925509641113b2d/pydantic_core-2.33.2-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:44857c3227d3fb5e753d5fe4a3420d6376fa594b07b621e220cd93703fe21782", size = 2074224, upload-time = "2025-04-23T18:32:44.033Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/07/31cf8fadffbb03be1cb520850e00a8490c0927ec456e8293cafda0726184/pydantic_core-2.33.2-cp39-cp39-musllinux_1_1_armv7l.whl", hash = "sha256:eb9b459ca4df0e5c87deb59d37377461a538852765293f9e6ee834f0435a93b9", size = 2245403, upload-time = "2025-04-23T18:32:45.836Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/8d/bbaf4c6721b668d44f01861f297eb01c9b35f612f6b8e14173cb204e6240/pydantic_core-2.33.2-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:9fcd347d2cc5c23b06de6d3b7b8275be558a0c90549495c699e379a80bf8379e", size = 2242331, upload-time = "2025-04-23T18:32:47.618Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/93/3cc157026bca8f5006250e74515119fcaa6d6858aceee8f67ab6dc548c16/pydantic_core-2.33.2-cp39-cp39-win32.whl", hash = "sha256:83aa99b1285bc8f038941ddf598501a86f1536789740991d7d8756e34f1e74d9", size = 1910571, upload-time = "2025-04-23T18:32:49.401Z" },
-    { url = "https://files.pythonhosted.org/packages/5b/90/7edc3b2a0d9f0dda8806c04e511a67b0b7a41d2187e2003673a996fb4310/pydantic_core-2.33.2-cp39-cp39-win_amd64.whl", hash = "sha256:f481959862f57f29601ccced557cc2e817bce7533ab8e01a797a48b49c9692b3", size = 1956504, upload-time = "2025-04-23T18:32:51.287Z" },
-    { url = "https://files.pythonhosted.org/packages/30/68/373d55e58b7e83ce371691f6eaa7175e3a24b956c44628eb25d7da007917/pydantic_core-2.33.2-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:5c4aa4e82353f65e548c476b37e64189783aa5384903bfea4f41580f255fddfa", size = 2023982, upload-time = "2025-04-23T18:32:53.14Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/16/145f54ac08c96a63d8ed6442f9dec17b2773d19920b627b18d4f10a061ea/pydantic_core-2.33.2-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:d946c8bf0d5c24bf4fe333af284c59a19358aa3ec18cb3dc4370080da1e8ad29", size = 1858412, upload-time = "2025-04-23T18:32:55.52Z" },
-    { url = "https://files.pythonhosted.org/packages/41/b1/c6dc6c3e2de4516c0bb2c46f6a373b91b5660312342a0cf5826e38ad82fa/pydantic_core-2.33.2-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:87b31b6846e361ef83fedb187bb5b4372d0da3f7e28d85415efa92d6125d6e6d", size = 1892749, upload-time = "2025-04-23T18:32:57.546Z" },
-    { url = "https://files.pythonhosted.org/packages/12/73/8cd57e20afba760b21b742106f9dbdfa6697f1570b189c7457a1af4cd8a0/pydantic_core-2.33.2-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:aa9d91b338f2df0508606f7009fde642391425189bba6d8c653afd80fd6bb64e", size = 2067527, upload-time = "2025-04-23T18:32:59.771Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/d5/0bb5d988cc019b3cba4a78f2d4b3854427fc47ee8ec8e9eaabf787da239c/pydantic_core-2.33.2-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2058a32994f1fde4ca0480ab9d1e75a0e8c87c22b53a3ae66554f9af78f2fe8c", size = 2108225, upload-time = "2025-04-23T18:33:04.51Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/c5/00c02d1571913d496aabf146106ad8239dc132485ee22efe08085084ff7c/pydantic_core-2.33.2-pp310-pypy310_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:0e03262ab796d986f978f79c943fc5f620381be7287148b8010b4097f79a39ec", size = 2069490, upload-time = "2025-04-23T18:33:06.391Z" },
-    { url = "https://files.pythonhosted.org/packages/22/a8/dccc38768274d3ed3a59b5d06f59ccb845778687652daa71df0cab4040d7/pydantic_core-2.33.2-pp310-pypy310_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:1a8695a8d00c73e50bff9dfda4d540b7dee29ff9b8053e38380426a85ef10052", size = 2237525, upload-time = "2025-04-23T18:33:08.44Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/e7/4f98c0b125dda7cf7ccd14ba936218397b44f50a56dd8c16a3091df116c3/pydantic_core-2.33.2-pp310-pypy310_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:fa754d1850735a0b0e03bcffd9d4b4343eb417e47196e4485d9cca326073a42c", size = 2238446, upload-time = "2025-04-23T18:33:10.313Z" },
-    { url = "https://files.pythonhosted.org/packages/ce/91/2ec36480fdb0b783cd9ef6795753c1dea13882f2e68e73bce76ae8c21e6a/pydantic_core-2.33.2-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:a11c8d26a50bfab49002947d3d237abe4d9e4b5bdc8846a63537b6488e197808", size = 2066678, upload-time = "2025-04-23T18:33:12.224Z" },
-    { url = "https://files.pythonhosted.org/packages/7b/27/d4ae6487d73948d6f20dddcd94be4ea43e74349b56eba82e9bdee2d7494c/pydantic_core-2.33.2-pp311-pypy311_pp73-macosx_10_12_x86_64.whl", hash = "sha256:dd14041875d09cc0f9308e37a6f8b65f5585cf2598a53aa0123df8b129d481f8", size = 2025200, upload-time = "2025-04-23T18:33:14.199Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/b8/b3cb95375f05d33801024079b9392a5ab45267a63400bf1866e7ce0f0de4/pydantic_core-2.33.2-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:d87c561733f66531dced0da6e864f44ebf89a8fba55f31407b00c2f7f9449593", size = 1859123, upload-time = "2025-04-23T18:33:16.555Z" },
-    { url = "https://files.pythonhosted.org/packages/05/bc/0d0b5adeda59a261cd30a1235a445bf55c7e46ae44aea28f7bd6ed46e091/pydantic_core-2.33.2-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2f82865531efd18d6e07a04a17331af02cb7a651583c418df8266f17a63c6612", size = 1892852, upload-time = "2025-04-23T18:33:18.513Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/11/d37bdebbda2e449cb3f519f6ce950927b56d62f0b84fd9cb9e372a26a3d5/pydantic_core-2.33.2-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2bfb5112df54209d820d7bf9317c7a6c9025ea52e49f46b6a2060104bba37de7", size = 2067484, upload-time = "2025-04-23T18:33:20.475Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/55/1f95f0a05ce72ecb02a8a8a1c3be0579bbc29b1d5ab68f1378b7bebc5057/pydantic_core-2.33.2-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:64632ff9d614e5eecfb495796ad51b0ed98c453e447a76bcbeeb69615079fc7e", size = 2108896, upload-time = "2025-04-23T18:33:22.501Z" },
-    { url = "https://files.pythonhosted.org/packages/53/89/2b2de6c81fa131f423246a9109d7b2a375e83968ad0800d6e57d0574629b/pydantic_core-2.33.2-pp311-pypy311_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:f889f7a40498cc077332c7ab6b4608d296d852182211787d4f3ee377aaae66e8", size = 2069475, upload-time = "2025-04-23T18:33:24.528Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/e9/1f7efbe20d0b2b10f6718944b5d8ece9152390904f29a78e68d4e7961159/pydantic_core-2.33.2-pp311-pypy311_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:de4b83bb311557e439b9e186f733f6c645b9417c84e2eb8203f3f820a4b988bf", size = 2239013, upload-time = "2025-04-23T18:33:26.621Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/b2/5309c905a93811524a49b4e031e9851a6b00ff0fb668794472ea7746b448/pydantic_core-2.33.2-pp311-pypy311_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:82f68293f055f51b51ea42fafc74b6aad03e70e191799430b90c13d643059ebb", size = 2238715, upload-time = "2025-04-23T18:33:28.656Z" },
-    { url = "https://files.pythonhosted.org/packages/32/56/8a7ca5d2cd2cda1d245d34b1c9a942920a718082ae8e54e5f3e5a58b7add/pydantic_core-2.33.2-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:329467cecfb529c925cf2bbd4d60d2c509bc2fb52a20c1045bf09bb70971a9c1", size = 2066757, upload-time = "2025-04-23T18:33:30.645Z" },
-    { url = "https://files.pythonhosted.org/packages/08/98/dbf3fdfabaf81cda5622154fda78ea9965ac467e3239078e0dcd6df159e7/pydantic_core-2.33.2-pp39-pypy39_pp73-macosx_10_12_x86_64.whl", hash = "sha256:87acbfcf8e90ca885206e98359d7dca4bcbb35abdc0ff66672a293e1d7a19101", size = 2024034, upload-time = "2025-04-23T18:33:32.843Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/99/7810aa9256e7f2ccd492590f86b79d370df1e9292f1f80b000b6a75bd2fb/pydantic_core-2.33.2-pp39-pypy39_pp73-macosx_11_0_arm64.whl", hash = "sha256:7f92c15cd1e97d4b12acd1cc9004fa092578acfa57b67ad5e43a197175d01a64", size = 1858578, upload-time = "2025-04-23T18:33:34.912Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/60/bc06fa9027c7006cc6dd21e48dbf39076dc39d9abbaf718a1604973a9670/pydantic_core-2.33.2-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d3f26877a748dc4251cfcfda9dfb5f13fcb034f5308388066bcfe9031b63ae7d", size = 1892858, upload-time = "2025-04-23T18:33:36.933Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/40/9d03997d9518816c68b4dfccb88969756b9146031b61cd37f781c74c9b6a/pydantic_core-2.33.2-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dac89aea9af8cd672fa7b510e7b8c33b0bba9a43186680550ccf23020f32d535", size = 2068498, upload-time = "2025-04-23T18:33:38.997Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/62/d490198d05d2d86672dc269f52579cad7261ced64c2df213d5c16e0aecb1/pydantic_core-2.33.2-pp39-pypy39_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:970919794d126ba8645f3837ab6046fb4e72bbc057b3709144066204c19a455d", size = 2108428, upload-time = "2025-04-23T18:33:41.18Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/ec/4cd215534fd10b8549015f12ea650a1a973da20ce46430b68fc3185573e8/pydantic_core-2.33.2-pp39-pypy39_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:3eb3fe62804e8f859c49ed20a8451342de53ed764150cb14ca71357c765dc2a6", size = 2069854, upload-time = "2025-04-23T18:33:43.446Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/1a/abbd63d47e1d9b0d632fee6bb15785d0889c8a6e0a6c3b5a8e28ac1ec5d2/pydantic_core-2.33.2-pp39-pypy39_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:3abcd9392a36025e3bd55f9bd38d908bd17962cc49bc6da8e7e96285336e2bca", size = 2237859, upload-time = "2025-04-23T18:33:45.56Z" },
-    { url = "https://files.pythonhosted.org/packages/80/1c/fa883643429908b1c90598fd2642af8839efd1d835b65af1f75fba4d94fe/pydantic_core-2.33.2-pp39-pypy39_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:3a1c81334778f9e3af2f8aeb7a960736e5cab1dfebfb26aabca09afd2906c039", size = 2239059, upload-time = "2025-04-23T18:33:47.735Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/29/3cade8a924a61f60ccfa10842f75eb12787e1440e2b8660ceffeb26685e7/pydantic_core-2.33.2-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:2807668ba86cb38c6817ad9bc66215ab8584d1d304030ce4f0887336f28a5e27", size = 2066661, upload-time = "2025-04-23T18:33:49.995Z" },
-]
-
-[[package]]
-name = "pydantic-extra-types"
-version = "2.10.5"
+sdist = { url = "https://files.pythonhosted.org/packages/71/70/23b021c950c2addd24ec408e9ab05d59b035b39d97cdc1130e1bce647bb6/pydantic_core-2.41.5.tar.gz", hash = "sha256:08daa51ea16ad373ffd5e7606252cc32f07bc72b28284b6bc9c6df804816476e", size = 460952, upload-time = "2025-11-04T13:43:49.098Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/c6/90/32c9941e728d564b411d574d8ee0cf09b12ec978cb22b294995bae5549a5/pydantic_core-2.41.5-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:77b63866ca88d804225eaa4af3e664c5faf3568cea95360d21f4725ab6e07146", size = 2107298, upload-time = "2025-11-04T13:39:04.116Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/a8/61c96a77fe28993d9a6fb0f4127e05430a267b235a124545d79fea46dd65/pydantic_core-2.41.5-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:dfa8a0c812ac681395907e71e1274819dec685fec28273a28905df579ef137e2", size = 1901475, upload-time = "2025-11-04T13:39:06.055Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/b6/338abf60225acc18cdc08b4faef592d0310923d19a87fba1faf05af5346e/pydantic_core-2.41.5-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5921a4d3ca3aee735d9fd163808f5e8dd6c6972101e4adbda9a4667908849b97", size = 1918815, upload-time = "2025-11-04T13:39:10.41Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/1c/2ed0433e682983d8e8cba9c8d8ef274d4791ec6a6f24c58935b90e780e0a/pydantic_core-2.41.5-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e25c479382d26a2a41b7ebea1043564a937db462816ea07afa8a44c0866d52f9", size = 2065567, upload-time = "2025-11-04T13:39:12.244Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/24/cf84974ee7d6eae06b9e63289b7b8f6549d416b5c199ca2d7ce13bbcf619/pydantic_core-2.41.5-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f547144f2966e1e16ae626d8ce72b4cfa0caedc7fa28052001c94fb2fcaa1c52", size = 2230442, upload-time = "2025-11-04T13:39:13.962Z" },
+    { url = "https://files.pythonhosted.org/packages/fd/21/4e287865504b3edc0136c89c9c09431be326168b1eb7841911cbc877a995/pydantic_core-2.41.5-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6f52298fbd394f9ed112d56f3d11aabd0d5bd27beb3084cc3d8ad069483b8941", size = 2350956, upload-time = "2025-11-04T13:39:15.889Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/76/7727ef2ffa4b62fcab916686a68a0426b9b790139720e1934e8ba797e238/pydantic_core-2.41.5-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:100baa204bb412b74fe285fb0f3a385256dad1d1879f0a5cb1499ed2e83d132a", size = 2068253, upload-time = "2025-11-04T13:39:17.403Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/8c/a4abfc79604bcb4c748e18975c44f94f756f08fb04218d5cb87eb0d3a63e/pydantic_core-2.41.5-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:05a2c8852530ad2812cb7914dc61a1125dc4e06252ee98e5638a12da6cc6fb6c", size = 2177050, upload-time = "2025-11-04T13:39:19.351Z" },
+    { url = "https://files.pythonhosted.org/packages/67/b1/de2e9a9a79b480f9cb0b6e8b6ba4c50b18d4e89852426364c66aa82bb7b3/pydantic_core-2.41.5-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:29452c56df2ed968d18d7e21f4ab0ac55e71dc59524872f6fc57dcf4a3249ed2", size = 2147178, upload-time = "2025-11-04T13:39:21Z" },
+    { url = "https://files.pythonhosted.org/packages/16/c1/dfb33f837a47b20417500efaa0378adc6635b3c79e8369ff7a03c494b4ac/pydantic_core-2.41.5-cp310-cp310-musllinux_1_1_armv7l.whl", hash = "sha256:d5160812ea7a8a2ffbe233d8da666880cad0cbaf5d4de74ae15c313213d62556", size = 2341833, upload-time = "2025-11-04T13:39:22.606Z" },
+    { url = "https://files.pythonhosted.org/packages/47/36/00f398642a0f4b815a9a558c4f1dca1b4020a7d49562807d7bc9ff279a6c/pydantic_core-2.41.5-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:df3959765b553b9440adfd3c795617c352154e497a4eaf3752555cfb5da8fc49", size = 2321156, upload-time = "2025-11-04T13:39:25.843Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/70/cad3acd89fde2010807354d978725ae111ddf6d0ea46d1ea1775b5c1bd0c/pydantic_core-2.41.5-cp310-cp310-win32.whl", hash = "sha256:1f8d33a7f4d5a7889e60dc39856d76d09333d8a6ed0f5f1190635cbec70ec4ba", size = 1989378, upload-time = "2025-11-04T13:39:27.92Z" },
+    { url = "https://files.pythonhosted.org/packages/76/92/d338652464c6c367e5608e4488201702cd1cbb0f33f7b6a85a60fe5f3720/pydantic_core-2.41.5-cp310-cp310-win_amd64.whl", hash = "sha256:62de39db01b8d593e45871af2af9e497295db8d73b085f6bfd0b18c83c70a8f9", size = 2013622, upload-time = "2025-11-04T13:39:29.848Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/72/74a989dd9f2084b3d9530b0915fdda64ac48831c30dbf7c72a41a5232db8/pydantic_core-2.41.5-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:a3a52f6156e73e7ccb0f8cced536adccb7042be67cb45f9562e12b319c119da6", size = 2105873, upload-time = "2025-11-04T13:39:31.373Z" },
+    { url = "https://files.pythonhosted.org/packages/12/44/37e403fd9455708b3b942949e1d7febc02167662bf1a7da5b78ee1ea2842/pydantic_core-2.41.5-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:7f3bf998340c6d4b0c9a2f02d6a400e51f123b59565d74dc60d252ce888c260b", size = 1899826, upload-time = "2025-11-04T13:39:32.897Z" },
+    { url = "https://files.pythonhosted.org/packages/33/7f/1d5cab3ccf44c1935a359d51a8a2a9e1a654b744b5e7f80d41b88d501eec/pydantic_core-2.41.5-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:378bec5c66998815d224c9ca994f1e14c0c21cb95d2f52b6021cc0b2a58f2a5a", size = 1917869, upload-time = "2025-11-04T13:39:34.469Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/6a/30d94a9674a7fe4f4744052ed6c5e083424510be1e93da5bc47569d11810/pydantic_core-2.41.5-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e7b576130c69225432866fe2f4a469a85a54ade141d96fd396dffcf607b558f8", size = 2063890, upload-time = "2025-11-04T13:39:36.053Z" },
+    { url = "https://files.pythonhosted.org/packages/50/be/76e5d46203fcb2750e542f32e6c371ffa9b8ad17364cf94bb0818dbfb50c/pydantic_core-2.41.5-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6cb58b9c66f7e4179a2d5e0f849c48eff5c1fca560994d6eb6543abf955a149e", size = 2229740, upload-time = "2025-11-04T13:39:37.753Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/ee/fed784df0144793489f87db310a6bbf8118d7b630ed07aa180d6067e653a/pydantic_core-2.41.5-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:88942d3a3dff3afc8288c21e565e476fc278902ae4d6d134f1eeda118cc830b1", size = 2350021, upload-time = "2025-11-04T13:39:40.94Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/be/8fed28dd0a180dca19e72c233cbf58efa36df055e5b9d90d64fd1740b828/pydantic_core-2.41.5-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f31d95a179f8d64d90f6831d71fa93290893a33148d890ba15de25642c5d075b", size = 2066378, upload-time = "2025-11-04T13:39:42.523Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/3b/698cf8ae1d536a010e05121b4958b1257f0b5522085e335360e53a6b1c8b/pydantic_core-2.41.5-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c1df3d34aced70add6f867a8cf413e299177e0c22660cc767218373d0779487b", size = 2175761, upload-time = "2025-11-04T13:39:44.553Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/ba/15d537423939553116dea94ce02f9c31be0fa9d0b806d427e0308ec17145/pydantic_core-2.41.5-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:4009935984bd36bd2c774e13f9a09563ce8de4abaa7226f5108262fa3e637284", size = 2146303, upload-time = "2025-11-04T13:39:46.238Z" },
+    { url = "https://files.pythonhosted.org/packages/58/7f/0de669bf37d206723795f9c90c82966726a2ab06c336deba4735b55af431/pydantic_core-2.41.5-cp311-cp311-musllinux_1_1_armv7l.whl", hash = "sha256:34a64bc3441dc1213096a20fe27e8e128bd3ff89921706e83c0b1ac971276594", size = 2340355, upload-time = "2025-11-04T13:39:48.002Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/de/e7482c435b83d7e3c3ee5ee4451f6e8973cff0eb6007d2872ce6383f6398/pydantic_core-2.41.5-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:c9e19dd6e28fdcaa5a1de679aec4141f691023916427ef9bae8584f9c2fb3b0e", size = 2319875, upload-time = "2025-11-04T13:39:49.705Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/e6/8c9e81bb6dd7560e33b9053351c29f30c8194b72f2d6932888581f503482/pydantic_core-2.41.5-cp311-cp311-win32.whl", hash = "sha256:2c010c6ded393148374c0f6f0bf89d206bf3217f201faa0635dcd56bd1520f6b", size = 1987549, upload-time = "2025-11-04T13:39:51.842Z" },
+    { url = "https://files.pythonhosted.org/packages/11/66/f14d1d978ea94d1bc21fc98fcf570f9542fe55bfcc40269d4e1a21c19bf7/pydantic_core-2.41.5-cp311-cp311-win_amd64.whl", hash = "sha256:76ee27c6e9c7f16f47db7a94157112a2f3a00e958bc626e2f4ee8bec5c328fbe", size = 2011305, upload-time = "2025-11-04T13:39:53.485Z" },
+    { url = "https://files.pythonhosted.org/packages/56/d8/0e271434e8efd03186c5386671328154ee349ff0354d83c74f5caaf096ed/pydantic_core-2.41.5-cp311-cp311-win_arm64.whl", hash = "sha256:4bc36bbc0b7584de96561184ad7f012478987882ebf9f9c389b23f432ea3d90f", size = 1972902, upload-time = "2025-11-04T13:39:56.488Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/5d/5f6c63eebb5afee93bcaae4ce9a898f3373ca23df3ccaef086d0233a35a7/pydantic_core-2.41.5-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:f41a7489d32336dbf2199c8c0a215390a751c5b014c2c1c5366e817202e9cdf7", size = 2110990, upload-time = "2025-11-04T13:39:58.079Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/32/9c2e8ccb57c01111e0fd091f236c7b371c1bccea0fa85247ac55b1e2b6b6/pydantic_core-2.41.5-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:070259a8818988b9a84a449a2a7337c7f430a22acc0859c6b110aa7212a6d9c0", size = 1896003, upload-time = "2025-11-04T13:39:59.956Z" },
+    { url = "https://files.pythonhosted.org/packages/68/b8/a01b53cb0e59139fbc9e4fda3e9724ede8de279097179be4ff31f1abb65a/pydantic_core-2.41.5-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e96cea19e34778f8d59fe40775a7a574d95816eb150850a85a7a4c8f4b94ac69", size = 1919200, upload-time = "2025-11-04T13:40:02.241Z" },
+    { url = "https://files.pythonhosted.org/packages/38/de/8c36b5198a29bdaade07b5985e80a233a5ac27137846f3bc2d3b40a47360/pydantic_core-2.41.5-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ed2e99c456e3fadd05c991f8f437ef902e00eedf34320ba2b0842bd1c3ca3a75", size = 2052578, upload-time = "2025-11-04T13:40:04.401Z" },
+    { url = "https://files.pythonhosted.org/packages/00/b5/0e8e4b5b081eac6cb3dbb7e60a65907549a1ce035a724368c330112adfdd/pydantic_core-2.41.5-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:65840751b72fbfd82c3c640cff9284545342a4f1eb1586ad0636955b261b0b05", size = 2208504, upload-time = "2025-11-04T13:40:06.072Z" },
+    { url = "https://files.pythonhosted.org/packages/77/56/87a61aad59c7c5b9dc8caad5a41a5545cba3810c3e828708b3d7404f6cef/pydantic_core-2.41.5-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e536c98a7626a98feb2d3eaf75944ef6f3dbee447e1f841eae16f2f0a72d8ddc", size = 2335816, upload-time = "2025-11-04T13:40:07.835Z" },
+    { url = "https://files.pythonhosted.org/packages/0d/76/941cc9f73529988688a665a5c0ecff1112b3d95ab48f81db5f7606f522d3/pydantic_core-2.41.5-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:eceb81a8d74f9267ef4081e246ffd6d129da5d87e37a77c9bde550cb04870c1c", size = 2075366, upload-time = "2025-11-04T13:40:09.804Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/43/ebef01f69baa07a482844faaa0a591bad1ef129253ffd0cdaa9d8a7f72d3/pydantic_core-2.41.5-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:d38548150c39b74aeeb0ce8ee1d8e82696f4a4e16ddc6de7b1d8823f7de4b9b5", size = 2171698, upload-time = "2025-11-04T13:40:12.004Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/87/41f3202e4193e3bacfc2c065fab7706ebe81af46a83d3e27605029c1f5a6/pydantic_core-2.41.5-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:c23e27686783f60290e36827f9c626e63154b82b116d7fe9adba1fda36da706c", size = 2132603, upload-time = "2025-11-04T13:40:13.868Z" },
+    { url = "https://files.pythonhosted.org/packages/49/7d/4c00df99cb12070b6bccdef4a195255e6020a550d572768d92cc54dba91a/pydantic_core-2.41.5-cp312-cp312-musllinux_1_1_armv7l.whl", hash = "sha256:482c982f814460eabe1d3bb0adfdc583387bd4691ef00b90575ca0d2b6fe2294", size = 2329591, upload-time = "2025-11-04T13:40:15.672Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/6a/ebf4b1d65d458f3cda6a7335d141305dfa19bdc61140a884d165a8a1bbc7/pydantic_core-2.41.5-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:bfea2a5f0b4d8d43adf9d7b8bf019fb46fdd10a2e5cde477fbcb9d1fa08c68e1", size = 2319068, upload-time = "2025-11-04T13:40:17.532Z" },
+    { url = "https://files.pythonhosted.org/packages/49/3b/774f2b5cd4192d5ab75870ce4381fd89cf218af999515baf07e7206753f0/pydantic_core-2.41.5-cp312-cp312-win32.whl", hash = "sha256:b74557b16e390ec12dca509bce9264c3bbd128f8a2c376eaa68003d7f327276d", size = 1985908, upload-time = "2025-11-04T13:40:19.309Z" },
+    { url = "https://files.pythonhosted.org/packages/86/45/00173a033c801cacf67c190fef088789394feaf88a98a7035b0e40d53dc9/pydantic_core-2.41.5-cp312-cp312-win_amd64.whl", hash = "sha256:1962293292865bca8e54702b08a4f26da73adc83dd1fcf26fbc875b35d81c815", size = 2020145, upload-time = "2025-11-04T13:40:21.548Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/22/91fbc821fa6d261b376a3f73809f907cec5ca6025642c463d3488aad22fb/pydantic_core-2.41.5-cp312-cp312-win_arm64.whl", hash = "sha256:1746d4a3d9a794cacae06a5eaaccb4b8643a131d45fbc9af23e353dc0a5ba5c3", size = 1976179, upload-time = "2025-11-04T13:40:23.393Z" },
+    { url = "https://files.pythonhosted.org/packages/54/db/160dffb57ed9a3705c4cbcbff0ac03bdae45f1ca7d58ab74645550df3fbd/pydantic_core-2.41.5-cp39-cp39-macosx_10_12_x86_64.whl", hash = "sha256:8bfeaf8735be79f225f3fefab7f941c712aaca36f1128c9d7e2352ee1aa87bdf", size = 2107999, upload-time = "2025-11-04T13:42:03.885Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/7d/88e7de946f60d9263cc84819f32513520b85c0f8322f9b8f6e4afc938383/pydantic_core-2.41.5-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:346285d28e4c8017da95144c7f3acd42740d637ff41946af5ce6e5e420502dd5", size = 1929745, upload-time = "2025-11-04T13:42:06.075Z" },
+    { url = "https://files.pythonhosted.org/packages/d5/c2/aef51e5b283780e85e99ff19db0f05842d2d4a8a8cd15e63b0280029b08f/pydantic_core-2.41.5-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a75dafbf87d6276ddc5b2bf6fae5254e3d0876b626eb24969a574fff9149ee5d", size = 1920220, upload-time = "2025-11-04T13:42:08.457Z" },
+    { url = "https://files.pythonhosted.org/packages/c7/97/492ab10f9ac8695cd76b2fdb24e9e61f394051df71594e9bcc891c9f586e/pydantic_core-2.41.5-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7b93a4d08587e2b7e7882de461e82b6ed76d9026ce91ca7915e740ecc7855f60", size = 2067296, upload-time = "2025-11-04T13:42:10.817Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/23/984149650e5269c59a2a4c41d234a9570adc68ab29981825cfaf4cfad8f4/pydantic_core-2.41.5-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e8465ab91a4bd96d36dde3263f06caa6a8a6019e4113f24dc753d79a8b3a3f82", size = 2231548, upload-time = "2025-11-04T13:42:13.843Z" },
+    { url = "https://files.pythonhosted.org/packages/71/0c/85bcbb885b9732c28bec67a222dbed5ed2d77baee1f8bba2002e8cd00c5c/pydantic_core-2.41.5-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:299e0a22e7ae2b85c1a57f104538b2656e8ab1873511fd718a1c1c6f149b77b5", size = 2362571, upload-time = "2025-11-04T13:42:16.208Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/4a/412d2048be12c334003e9b823a3fa3d038e46cc2d64dd8aab50b31b65499/pydantic_core-2.41.5-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:707625ef0983fcfb461acfaf14de2067c5942c6bb0f3b4c99158bed6fedd3cf3", size = 2068175, upload-time = "2025-11-04T13:42:18.911Z" },
+    { url = "https://files.pythonhosted.org/packages/73/f4/c58b6a776b502d0a5540ad02e232514285513572060f0d78f7832ca3c98b/pydantic_core-2.41.5-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:f41eb9797986d6ebac5e8edff36d5cef9de40def462311b3eb3eeded1431e425", size = 2177203, upload-time = "2025-11-04T13:42:22.578Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/ae/f06ea4c7e7a9eead3d165e7623cd2ea0cb788e277e4f935af63fc98fa4e6/pydantic_core-2.41.5-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:0384e2e1021894b1ff5a786dbf94771e2986ebe2869533874d7e43bc79c6f504", size = 2148191, upload-time = "2025-11-04T13:42:24.89Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/57/25a11dcdc656bf5f8b05902c3c2934ac3ea296257cc4a3f79a6319e61856/pydantic_core-2.41.5-cp39-cp39-musllinux_1_1_armv7l.whl", hash = "sha256:f0cd744688278965817fd0839c4a4116add48d23890d468bc436f78beb28abf5", size = 2343907, upload-time = "2025-11-04T13:42:27.683Z" },
+    { url = "https://files.pythonhosted.org/packages/96/82/e33d5f4933d7a03327c0c43c65d575e5919d4974ffc026bc917a5f7b9f61/pydantic_core-2.41.5-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:753e230374206729bf0a807954bcc6c150d3743928a73faffee51ac6557a03c3", size = 2322174, upload-time = "2025-11-04T13:42:30.776Z" },
+    { url = "https://files.pythonhosted.org/packages/81/45/4091be67ce9f469e81656f880f3506f6a5624121ec5eb3eab37d7581897d/pydantic_core-2.41.5-cp39-cp39-win32.whl", hash = "sha256:873e0d5b4fb9b89ef7c2d2a963ea7d02879d9da0da8d9d4933dee8ee86a8b460", size = 1990353, upload-time = "2025-11-04T13:42:33.111Z" },
+    { url = "https://files.pythonhosted.org/packages/44/8a/a98aede18db6e9cd5d66bcacd8a409fcf8134204cdede2e7de35c5a2c5ef/pydantic_core-2.41.5-cp39-cp39-win_amd64.whl", hash = "sha256:e4f4a984405e91527a0d62649ee21138f8e3d0ef103be488c1dc11a80d7f184b", size = 2015698, upload-time = "2025-11-04T13:42:35.484Z" },
+    { url = "https://files.pythonhosted.org/packages/11/72/90fda5ee3b97e51c494938a4a44c3a35a9c96c19bba12372fb9c634d6f57/pydantic_core-2.41.5-graalpy311-graalpy242_311_native-macosx_10_12_x86_64.whl", hash = "sha256:b96d5f26b05d03cc60f11a7761a5ded1741da411e7fe0909e27a5e6a0cb7b034", size = 2115441, upload-time = "2025-11-04T13:42:39.557Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/53/8942f884fa33f50794f119012dc6a1a02ac43a56407adaac20463df8e98f/pydantic_core-2.41.5-graalpy311-graalpy242_311_native-macosx_11_0_arm64.whl", hash = "sha256:634e8609e89ceecea15e2d61bc9ac3718caaaa71963717bf3c8f38bfde64242c", size = 1930291, upload-time = "2025-11-04T13:42:42.169Z" },
+    { url = "https://files.pythonhosted.org/packages/79/c8/ecb9ed9cd942bce09fc888ee960b52654fbdbede4ba6c2d6e0d3b1d8b49c/pydantic_core-2.41.5-graalpy311-graalpy242_311_native-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:93e8740d7503eb008aa2df04d3b9735f845d43ae845e6dcd2be0b55a2da43cd2", size = 1948632, upload-time = "2025-11-04T13:42:44.564Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/1b/687711069de7efa6af934e74f601e2a4307365e8fdc404703afc453eab26/pydantic_core-2.41.5-graalpy311-graalpy242_311_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f15489ba13d61f670dcc96772e733aad1a6f9c429cc27574c6cdaed82d0146ad", size = 2138905, upload-time = "2025-11-04T13:42:47.156Z" },
+    { url = "https://files.pythonhosted.org/packages/09/32/59b0c7e63e277fa7911c2fc70ccfb45ce4b98991e7ef37110663437005af/pydantic_core-2.41.5-graalpy312-graalpy250_312_native-macosx_10_12_x86_64.whl", hash = "sha256:7da7087d756b19037bc2c06edc6c170eeef3c3bafcb8f532ff17d64dc427adfd", size = 2110495, upload-time = "2025-11-04T13:42:49.689Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/81/05e400037eaf55ad400bcd318c05bb345b57e708887f07ddb2d20e3f0e98/pydantic_core-2.41.5-graalpy312-graalpy250_312_native-macosx_11_0_arm64.whl", hash = "sha256:aabf5777b5c8ca26f7824cb4a120a740c9588ed58df9b2d196ce92fba42ff8dc", size = 1915388, upload-time = "2025-11-04T13:42:52.215Z" },
+    { url = "https://files.pythonhosted.org/packages/6e/0d/e3549b2399f71d56476b77dbf3cf8937cec5cd70536bdc0e374a421d0599/pydantic_core-2.41.5-graalpy312-graalpy250_312_native-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c007fe8a43d43b3969e8469004e9845944f1a80e6acd47c150856bb87f230c56", size = 1942879, upload-time = "2025-11-04T13:42:56.483Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/07/34573da085946b6a313d7c42f82f16e8920bfd730665de2d11c0c37a74b5/pydantic_core-2.41.5-graalpy312-graalpy250_312_native-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:76d0819de158cd855d1cbb8fcafdf6f5cf1eb8e470abe056d5d161106e38062b", size = 2139017, upload-time = "2025-11-04T13:42:59.471Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/b0/1a2aa41e3b5a4ba11420aba2d091b2d17959c8d1519ece3627c371951e73/pydantic_core-2.41.5-pp310-pypy310_pp73-macosx_10_12_x86_64.whl", hash = "sha256:b5819cd790dbf0c5eb9f82c73c16b39a65dd6dd4d1439dcdea7816ec9adddab8", size = 2103351, upload-time = "2025-11-04T13:43:02.058Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/ee/31b1f0020baaf6d091c87900ae05c6aeae101fa4e188e1613c80e4f1ea31/pydantic_core-2.41.5-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:5a4e67afbc95fa5c34cf27d9089bca7fcab4e51e57278d710320a70b956d1b9a", size = 1925363, upload-time = "2025-11-04T13:43:05.159Z" },
+    { url = "https://files.pythonhosted.org/packages/e1/89/ab8e86208467e467a80deaca4e434adac37b10a9d134cd2f99b28a01e483/pydantic_core-2.41.5-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ece5c59f0ce7d001e017643d8d24da587ea1f74f6993467d85ae8a5ef9d4f42b", size = 2135615, upload-time = "2025-11-04T13:43:08.116Z" },
+    { url = "https://files.pythonhosted.org/packages/99/0a/99a53d06dd0348b2008f2f30884b34719c323f16c3be4e6cc1203b74a91d/pydantic_core-2.41.5-pp310-pypy310_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:16f80f7abe3351f8ea6858914ddc8c77e02578544a0ebc15b4c2e1a0e813b0b2", size = 2175369, upload-time = "2025-11-04T13:43:12.49Z" },
+    { url = "https://files.pythonhosted.org/packages/6d/94/30ca3b73c6d485b9bb0bc66e611cff4a7138ff9736b7e66bcf0852151636/pydantic_core-2.41.5-pp310-pypy310_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:33cb885e759a705b426baada1fe68cbb0a2e68e34c5d0d0289a364cf01709093", size = 2144218, upload-time = "2025-11-04T13:43:15.431Z" },
+    { url = "https://files.pythonhosted.org/packages/87/57/31b4f8e12680b739a91f472b5671294236b82586889ef764b5fbc6669238/pydantic_core-2.41.5-pp310-pypy310_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:c8d8b4eb992936023be7dee581270af5c6e0697a8559895f527f5b7105ecd36a", size = 2329951, upload-time = "2025-11-04T13:43:18.062Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/73/3c2c8edef77b8f7310e6fb012dbc4b8551386ed575b9eb6fb2506e28a7eb/pydantic_core-2.41.5-pp310-pypy310_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:242a206cd0318f95cd21bdacff3fcc3aab23e79bba5cac3db5a841c9ef9c6963", size = 2318428, upload-time = "2025-11-04T13:43:20.679Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/02/8559b1f26ee0d502c74f9cca5c0d2fd97e967e083e006bbbb4e97f3a043a/pydantic_core-2.41.5-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:d3a978c4f57a597908b7e697229d996d77a6d3c94901e9edee593adada95ce1a", size = 2147009, upload-time = "2025-11-04T13:43:23.286Z" },
+    { url = "https://files.pythonhosted.org/packages/5f/9b/1b3f0e9f9305839d7e84912f9e8bfbd191ed1b1ef48083609f0dabde978c/pydantic_core-2.41.5-pp311-pypy311_pp73-macosx_10_12_x86_64.whl", hash = "sha256:b2379fa7ed44ddecb5bfe4e48577d752db9fc10be00a6b7446e9663ba143de26", size = 2101980, upload-time = "2025-11-04T13:43:25.97Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/ed/d71fefcb4263df0da6a85b5d8a7508360f2f2e9b3bf5814be9c8bccdccc1/pydantic_core-2.41.5-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:266fb4cbf5e3cbd0b53669a6d1b039c45e3ce651fd5442eff4d07c2cc8d66808", size = 1923865, upload-time = "2025-11-04T13:43:28.763Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/3a/626b38db460d675f873e4444b4bb030453bbe7b4ba55df821d026a0493c4/pydantic_core-2.41.5-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:58133647260ea01e4d0500089a8c4f07bd7aa6ce109682b1426394988d8aaacc", size = 2134256, upload-time = "2025-11-04T13:43:31.71Z" },
+    { url = "https://files.pythonhosted.org/packages/83/d9/8412d7f06f616bbc053d30cb4e5f76786af3221462ad5eee1f202021eb4e/pydantic_core-2.41.5-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:287dad91cfb551c363dc62899a80e9e14da1f0e2b6ebde82c806612ca2a13ef1", size = 2174762, upload-time = "2025-11-04T13:43:34.744Z" },
+    { url = "https://files.pythonhosted.org/packages/55/4c/162d906b8e3ba3a99354e20faa1b49a85206c47de97a639510a0e673f5da/pydantic_core-2.41.5-pp311-pypy311_pp73-musllinux_1_1_aarch64.whl", hash = "sha256:03b77d184b9eb40240ae9fd676ca364ce1085f203e1b1256f8ab9984dca80a84", size = 2143141, upload-time = "2025-11-04T13:43:37.701Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/f2/f11dd73284122713f5f89fc940f370d035fa8e1e078d446b3313955157fe/pydantic_core-2.41.5-pp311-pypy311_pp73-musllinux_1_1_armv7l.whl", hash = "sha256:a668ce24de96165bb239160b3d854943128f4334822900534f2fe947930e5770", size = 2330317, upload-time = "2025-11-04T13:43:40.406Z" },
+    { url = "https://files.pythonhosted.org/packages/88/9d/b06ca6acfe4abb296110fb1273a4d848a0bfb2ff65f3ee92127b3244e16b/pydantic_core-2.41.5-pp311-pypy311_pp73-musllinux_1_1_x86_64.whl", hash = "sha256:f14f8f046c14563f8eb3f45f499cc658ab8d10072961e07225e507adb700e93f", size = 2316992, upload-time = "2025-11-04T13:43:43.602Z" },
+    { url = "https://files.pythonhosted.org/packages/36/c7/cfc8e811f061c841d7990b0201912c3556bfeb99cdcb7ed24adc8d6f8704/pydantic_core-2.41.5-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:56121965f7a4dc965bff783d70b907ddf3d57f6eba29b6d2e5dabfaf07799c51", size = 2145302, upload-time = "2025-11-04T13:43:46.64Z" },
+]
+
+[[package]]
+name = "pydantic-settings"
+version = "2.11.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "pydantic" },
-    { name = "typing-extensions" },
+    { name = "pydantic", marker = "python_full_version < '3.10'" },
+    { name = "python-dotenv", marker = "python_full_version < '3.10'" },
+    { name = "typing-inspection", marker = "python_full_version < '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/7e/ba/4178111ec4116c54e1dc7ecd2a1ff8f54256cdbd250e576882911e8f710a/pydantic_extra_types-2.10.5.tar.gz", hash = "sha256:1dcfa2c0cf741a422f088e0dbb4690e7bfadaaf050da3d6f80d6c3cf58a2bad8", size = 138429, upload-time = "2025-06-02T09:31:52.713Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/20/c5/dbbc27b814c71676593d1c3f718e6cd7d4f00652cefa24b75f7aa3efb25e/pydantic_settings-2.11.0.tar.gz", hash = "sha256:d0e87a1c7d33593beb7194adb8470fc426e95ba02af83a0f23474a04c9a08180", size = 188394, upload-time = "2025-09-24T14:19:11.764Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/70/1a/5f4fd9e7285f10c44095a4f9fe17d0f358d1702a7c74a9278c794e8a7537/pydantic_extra_types-2.10.5-py3-none-any.whl", hash = "sha256:b60c4e23d573a69a4f1a16dd92888ecc0ef34fb0e655b4f305530377fa70e7a8", size = 38315, upload-time = "2025-06-02T09:31:51.229Z" },
-]
-
-[package.optional-dependencies]
-pycountry = [
-    { name = "pycountry" },
+    { url = "https://files.pythonhosted.org/packages/83/d6/887a1ff844e64aa823fb4905978d882a633cfe295c32eacad582b78a7d8b/pydantic_settings-2.11.0-py3-none-any.whl", hash = "sha256:fe2cea3413b9530d10f3a5875adffb17ada5c1e1bab0b2885546d7310415207c", size = 48608, upload-time = "2025-09-24T14:19:10.015Z" },
 ]
 
 [[package]]
 name = "pydantic-settings"
-version = "2.11.0"
+version = "2.12.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
 dependencies = [
-    { name = "pydantic" },
-    { name = "python-dotenv" },
-    { name = "typing-inspection" },
+    { name = "pydantic", marker = "python_full_version >= '3.10'" },
+    { name = "python-dotenv", marker = "python_full_version >= '3.10'" },
+    { name = "typing-inspection", marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/20/c5/dbbc27b814c71676593d1c3f718e6cd7d4f00652cefa24b75f7aa3efb25e/pydantic_settings-2.11.0.tar.gz", hash = "sha256:d0e87a1c7d33593beb7194adb8470fc426e95ba02af83a0f23474a04c9a08180", size = 188394, upload-time = "2025-09-24T14:19:11.764Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/43/4b/ac7e0aae12027748076d72a8764ff1c9d82ca75a7a52622e67ed3f765c54/pydantic_settings-2.12.0.tar.gz", hash = "sha256:005538ef951e3c2a68e1c08b292b5f2e71490def8589d4221b95dab00dafcfd0", size = 194184, upload-time = "2025-11-10T14:25:47.013Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/83/d6/887a1ff844e64aa823fb4905978d882a633cfe295c32eacad582b78a7d8b/pydantic_settings-2.11.0-py3-none-any.whl", hash = "sha256:fe2cea3413b9530d10f3a5875adffb17ada5c1e1bab0b2885546d7310415207c", size = 48608, upload-time = "2025-09-24T14:19:10.015Z" },
+    { url = "https://files.pythonhosted.org/packages/c1/60/5d4751ba3f4a40a6891f24eec885f51afd78d208498268c734e256fb13c4/pydantic_settings-2.12.0-py3-none-any.whl", hash = "sha256:fddb9fd99a5b18da837b29710391e945b1e30c135477f484084ee513adb93809", size = 51880, upload-time = "2025-11-10T14:25:45.546Z" },
 ]
 
 [[package]]
@@ -3935,45 +3946,94 @@ wheels = [
 name = "pytest"
 version = "8.4.2"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "colorama", marker = "sys_platform == 'win32'" },
-    { name = "exceptiongroup", marker = "python_full_version < '3.11'" },
-    { name = "iniconfig" },
-    { name = "packaging" },
-    { name = "pluggy" },
-    { name = "pygments" },
-    { name = "tomli", marker = "python_full_version < '3.11'" },
+    { name = "colorama", marker = "python_full_version < '3.10' and sys_platform == 'win32'" },
+    { name = "exceptiongroup", marker = "python_full_version < '3.10'" },
+    { name = "iniconfig", version = "2.1.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "packaging", marker = "python_full_version < '3.10'" },
+    { name = "pluggy", marker = "python_full_version < '3.10'" },
+    { name = "pygments", marker = "python_full_version < '3.10'" },
+    { name = "tomli", marker = "python_full_version < '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/a3/5c/00a0e072241553e1a7496d638deababa67c5058571567b92a7eaa258397c/pytest-8.4.2.tar.gz", hash = "sha256:86c0d0b93306b961d58d62a4db4879f27fe25513d4b969df351abdddb3c30e01", size = 1519618, upload-time = "2025-09-04T14:34:22.711Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/a8/a4/20da314d277121d6534b3a980b29035dcd51e6744bd79075a6ce8fa4eb8d/pytest-8.4.2-py3-none-any.whl", hash = "sha256:872f880de3fc3a5bdc88a11b39c9710c3497a547cfa9320bc3c5e62fbf272e79", size = 365750, upload-time = "2025-09-04T14:34:20.226Z" },
 ]
 
+[[package]]
+name = "pytest"
+version = "9.0.1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "colorama", marker = "python_full_version >= '3.10' and sys_platform == 'win32'" },
+    { name = "exceptiongroup", marker = "python_full_version == '3.10.*'" },
+    { name = "iniconfig", version = "2.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "packaging", marker = "python_full_version >= '3.10'" },
+    { name = "pluggy", marker = "python_full_version >= '3.10'" },
+    { name = "pygments", marker = "python_full_version >= '3.10'" },
+    { name = "tomli", marker = "python_full_version == '3.10.*'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/07/56/f013048ac4bc4c1d9be45afd4ab209ea62822fb1598f40687e6bf45dcea4/pytest-9.0.1.tar.gz", hash = "sha256:3e9c069ea73583e255c3b21cf46b8d3c56f6e3a1a8f6da94ccb0fcf57b9d73c8", size = 1564125, upload-time = "2025-11-12T13:05:09.333Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/0b/8b/6300fb80f858cda1c51ffa17075df5d846757081d11ab4aa35cef9e6258b/pytest-9.0.1-py3-none-any.whl", hash = "sha256:67be0030d194df2dfa7b556f2e56fb3c3315bd5c8822c6951162b92b32ce7dad", size = 373668, upload-time = "2025-11-12T13:05:07.379Z" },
+]
+
 [[package]]
 name = "pytest-asyncio"
 version = "1.2.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
+dependencies = [
+    { name = "backports-asyncio-runner", marker = "python_full_version < '3.10'" },
+    { name = "pytest", version = "8.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/42/86/9e3c5f48f7b7b638b216e4b9e645f54d199d7abbbab7a64a13b4e12ba10f/pytest_asyncio-1.2.0.tar.gz", hash = "sha256:c609a64a2a8768462d0c99811ddb8bd2583c33fd33cf7f21af1c142e824ffb57", size = 50119, upload-time = "2025-09-12T07:33:53.816Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/04/93/2fa34714b7a4ae72f2f8dad66ba17dd9a2c793220719e736dda28b7aec27/pytest_asyncio-1.2.0-py3-none-any.whl", hash = "sha256:8e17ae5e46d8e7efe51ab6494dd2010f4ca8dae51652aa3c8d55acf50bfb2e99", size = 15095, upload-time = "2025-09-12T07:33:52.639Z" },
+]
+
+[[package]]
+name = "pytest-asyncio"
+version = "1.3.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
 dependencies = [
-    { name = "backports-asyncio-runner", marker = "python_full_version < '3.11'" },
-    { name = "pytest" },
-    { name = "typing-extensions" },
+    { name = "backports-asyncio-runner", marker = "python_full_version == '3.10.*'" },
+    { name = "pytest", version = "9.0.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/42/86/9e3c5f48f7b7b638b216e4b9e645f54d199d7abbbab7a64a13b4e12ba10f/pytest_asyncio-1.2.0.tar.gz", hash = "sha256:c609a64a2a8768462d0c99811ddb8bd2583c33fd33cf7f21af1c142e824ffb57", size = 50119, upload-time = "2025-09-12T07:33:53.816Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/90/2c/8af215c0f776415f3590cac4f9086ccefd6fd463befeae41cd4d3f193e5a/pytest_asyncio-1.3.0.tar.gz", hash = "sha256:d7f52f36d231b80ee124cd216ffb19369aa168fc10095013c6b014a34d3ee9e5", size = 50087, upload-time = "2025-11-10T16:07:47.256Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/04/93/2fa34714b7a4ae72f2f8dad66ba17dd9a2c793220719e736dda28b7aec27/pytest_asyncio-1.2.0-py3-none-any.whl", hash = "sha256:8e17ae5e46d8e7efe51ab6494dd2010f4ca8dae51652aa3c8d55acf50bfb2e99", size = 15095, upload-time = "2025-09-12T07:33:52.639Z" },
+    { url = "https://files.pythonhosted.org/packages/e5/35/f8b19922b6a25bc0880171a2f1a003eaeb93657475193ab516fd87cac9da/pytest_asyncio-1.3.0-py3-none-any.whl", hash = "sha256:611e26147c7f77640e6d0a92a38ed17c3e9848063698d5c93d5aa7aa11cebff5", size = 15075, upload-time = "2025-11-10T16:07:45.537Z" },
 ]
 
 [[package]]
 name = "pytest-benchmark"
-version = "5.1.0"
+version = "5.2.3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "py-cpuinfo" },
-    { name = "pytest" },
+    { name = "pytest", version = "8.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pytest", version = "9.0.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/39/d0/a8bd08d641b393db3be3819b03e2d9bb8760ca8479080a26a5f6e540e99c/pytest-benchmark-5.1.0.tar.gz", hash = "sha256:9ea661cdc292e8231f7cd4c10b0319e56a2118e2c09d9f50e1b3d150d2aca105", size = 337810, upload-time = "2024-10-30T11:51:48.521Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/24/34/9f732b76456d64faffbef6232f1f9dbec7a7c4999ff46282fa418bd1af66/pytest_benchmark-5.2.3.tar.gz", hash = "sha256:deb7317998a23c650fd4ff76e1230066a76cb45dcece0aca5607143c619e7779", size = 341340, upload-time = "2025-11-09T18:48:43.215Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/9e/d6/b41653199ea09d5969d4e385df9bbfd9a100f28ca7e824ce7c0a016e3053/pytest_benchmark-5.1.0-py3-none-any.whl", hash = "sha256:922de2dfa3033c227c96da942d1878191afa135a29485fb942e85dff1c592c89", size = 44259, upload-time = "2024-10-30T11:51:45.94Z" },
+    { url = "https://files.pythonhosted.org/packages/33/29/e756e715a48959f1c0045342088d7ca9762a2f509b945f362a316e9412b7/pytest_benchmark-5.2.3-py3-none-any.whl", hash = "sha256:bc839726ad20e99aaa0d11a127445457b4219bdb9e80a1afc4b51da7f96b0803", size = 45255, upload-time = "2025-11-09T18:48:39.765Z" },
 ]
 
 [[package]]
@@ -3981,9 +4041,11 @@ name = "pytest-cov"
 version = "7.0.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "coverage", extra = ["toml"] },
+    { name = "coverage", version = "7.10.7", source = { registry = "https://pypi.org/simple" }, extra = ["toml"], marker = "python_full_version < '3.10'" },
+    { name = "coverage", version = "7.12.0", source = { registry = "https://pypi.org/simple" }, extra = ["toml"], marker = "python_full_version >= '3.10'" },
     { name = "pluggy" },
-    { name = "pytest" },
+    { name = "pytest", version = "8.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pytest", version = "9.0.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/5e/f7/c933acc76f5208b3b00089573cf6a2bc26dc80a8aece8f52bb7d6b1855ca/pytest_cov-7.0.0.tar.gz", hash = "sha256:33c97eda2e049a0c5298e91f519302a1334c26ac65c1a483d6206fd458361af1", size = 54328, upload-time = "2025-09-09T10:57:02.113Z" }
 wheels = [
@@ -3995,7 +4057,8 @@ name = "pytest-mock"
 version = "3.15.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "pytest" },
+    { name = "pytest", version = "8.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pytest", version = "9.0.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/68/14/eb014d26be205d38ad5ad20d9a80f7d201472e08167f0bb4361e251084a9/pytest_mock-3.15.1.tar.gz", hash = "sha256:1849a238f6f396da19762269de72cb1814ab44416fa73a8686deac10b0d87a0f", size = 34036, upload-time = "2025-09-16T16:37:27.081Z" }
 wheels = [
@@ -4016,32 +4079,11 @@ wheels = [
 
 [[package]]
 name = "python-dotenv"
-version = "1.1.1"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f6/b0/4bc07ccd3572a2f9df7e6782f52b0c6c90dcbb803ac4a167702d7d0dfe1e/python_dotenv-1.1.1.tar.gz", hash = "sha256:a8a6399716257f45be6a007360200409fce5cda2661e3dec71d23dc15f6189ab", size = 41978, upload-time = "2025-06-24T04:21:07.341Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/5f/ed/539768cf28c661b5b068d66d96a2f155c4971a5d55684a514c1a0e0dec2f/python_dotenv-1.1.1-py3-none-any.whl", hash = "sha256:31f23644fe2602f88ff55e1f5c79ba497e01224ee7737937930c448e4d0e24dc", size = 20556, upload-time = "2025-06-24T04:21:06.073Z" },
-]
-
-[[package]]
-name = "python-json-logger"
-version = "4.0.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/29/bf/eca6a3d43db1dae7070f70e160ab20b807627ba953663ba07928cdd3dc58/python_json_logger-4.0.0.tar.gz", hash = "sha256:f58e68eb46e1faed27e0f574a55a0455eecd7b8a5b88b85a784519ba3cff047f", size = 17683, upload-time = "2025-10-06T04:15:18.984Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/51/e5/fecf13f06e5e5f67e8837d777d1bc43fac0ed2b77a676804df5c34744727/python_json_logger-4.0.0-py3-none-any.whl", hash = "sha256:af09c9daf6a813aa4cc7180395f50f2a9e5fa056034c9953aec92e381c5ba1e2", size = 15548, upload-time = "2025-10-06T04:15:17.553Z" },
-]
-
-[[package]]
-name = "python-multipart"
-version = "0.0.20"
+version = "1.2.1"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/f3/87/f44d7c9f274c7ee665a29b885ec97089ec5dc034c7f3fafa03da9e39a09e/python_multipart-0.0.20.tar.gz", hash = "sha256:8dd0cab45b8e23064ae09147625994d090fa46f5b0d1e13af944c331a7fa9d13", size = 37158, upload-time = "2024-12-16T19:45:46.972Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/f0/26/19cadc79a718c5edbec86fd4919a6b6d3f681039a2f6d66d14be94e75fb9/python_dotenv-1.2.1.tar.gz", hash = "sha256:42667e897e16ab0d66954af0e60a9caa94f0fd4ecf3aaf6d2d260eec1aa36ad6", size = 44221, upload-time = "2025-10-26T15:12:10.434Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/45/58/38b5afbc1a800eeea951b9285d3912613f2603bdf897a4ab0f4bd7f405fc/python_multipart-0.0.20-py3-none-any.whl", hash = "sha256:8a62d3a8335e06589fe01f2a3e178cdcc632f3fbe0d492ad9ee0ec35aab1f104", size = 24546, upload-time = "2024-12-16T19:45:44.423Z" },
+    { url = "https://files.pythonhosted.org/packages/14/1b/a298b06749107c305e1fe0f814c6c74aea7b2f1e10989cb30f544a1b3253/python_dotenv-1.2.1-py3-none-any.whl", hash = "sha256:b81ee9561e9ca4004139c6cbba3a238c32b03e4894671e181b671e8cb8425d61", size = 21230, upload-time = "2025-10-26T15:12:09.109Z" },
 ]
 
 [[package]]
@@ -4110,58 +4152,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/04/11/432f32f8097b03e3cd5fe57e88efb685d964e2e5178a48ed61e841f7fdce/pyyaml_env_tag-1.1-py3-none-any.whl", hash = "sha256:17109e1a528561e32f026364712fee1264bc2ea6715120891174ed1b980d2e04", size = 4722, upload-time = "2025-05-13T15:23:59.629Z" },
 ]
 
-[[package]]
-name = "pyzmq"
-version = "27.1.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "cffi", marker = "implementation_name == 'pypy'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/04/0b/3c9baedbdf613ecaa7aa07027780b8867f57b6293b6ee50de316c9f3222b/pyzmq-27.1.0.tar.gz", hash = "sha256:ac0765e3d44455adb6ddbf4417dcce460fc40a05978c08efdf2948072f6db540", size = 281750, upload-time = "2025-09-08T23:10:18.157Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/99/64/5653e7b7425b169f994835a2b2abf9486264401fdef18df91ddae47ce2cc/pyzmq-27.1.0-cp310-cp310-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:507b6f430bdcf0ee48c0d30e734ea89ce5567fd7b8a0f0044a369c176aa44556", size = 906380, upload-time = "2025-09-08T23:07:29.78Z" },
-    { url = "https://files.pythonhosted.org/packages/73/78/7d713284dbe022f6440e391bd1f3c48d9185673878034cfb3939cdf333b2/pyzmq-27.1.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bf7b38f9fd7b81cb6d9391b2946382c8237fd814075c6aa9c3b746d53076023b", size = 666421, upload-time = "2025-09-08T23:07:31.263Z" },
-    { url = "https://files.pythonhosted.org/packages/30/76/8f099f9d6482450428b17c4d6b241281af7ce6a9de8149ca8c1c649f6792/pyzmq-27.1.0-cp310-cp310-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:03ff0b279b40d687691a6217c12242ee71f0fba28bf8626ff50e3ef0f4410e1e", size = 854149, upload-time = "2025-09-08T23:07:33.17Z" },
-    { url = "https://files.pythonhosted.org/packages/59/f0/37fbfff06c68016019043897e4c969ceab18bde46cd2aca89821fcf4fb2e/pyzmq-27.1.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:677e744fee605753eac48198b15a2124016c009a11056f93807000ab11ce6526", size = 1655070, upload-time = "2025-09-08T23:07:35.205Z" },
-    { url = "https://files.pythonhosted.org/packages/47/14/7254be73f7a8edc3587609554fcaa7bfd30649bf89cd260e4487ca70fdaa/pyzmq-27.1.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:dd2fec2b13137416a1c5648b7009499bcc8fea78154cd888855fa32514f3dad1", size = 2033441, upload-time = "2025-09-08T23:07:37.432Z" },
-    { url = "https://files.pythonhosted.org/packages/22/dc/49f2be26c6f86f347e796a4d99b19167fc94503f0af3fd010ad262158822/pyzmq-27.1.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:08e90bb4b57603b84eab1d0ca05b3bbb10f60c1839dc471fc1c9e1507bef3386", size = 1891529, upload-time = "2025-09-08T23:07:39.047Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/a0/fc7e78a23748ad5443ac3275943457e8452da67fda347e05260261108cbc/pyzmq-27.1.0-cp311-cp311-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:0790a0161c281ca9723f804871b4027f2e8b5a528d357c8952d08cd1a9c15581", size = 908803, upload-time = "2025-09-08T23:07:47.551Z" },
-    { url = "https://files.pythonhosted.org/packages/7e/22/37d15eb05f3bdfa4abea6f6d96eb3bb58585fbd3e4e0ded4e743bc650c97/pyzmq-27.1.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c895a6f35476b0c3a54e3eb6ccf41bf3018de937016e6e18748317f25d4e925f", size = 668836, upload-time = "2025-09-08T23:07:49.436Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/c4/2a6fe5111a01005fc7af3878259ce17684fabb8852815eda6225620f3c59/pyzmq-27.1.0-cp311-cp311-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5bbf8d3630bf96550b3be8e1fc0fea5cbdc8d5466c1192887bd94869da17a63e", size = 857038, upload-time = "2025-09-08T23:07:51.234Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/eb/bfdcb41d0db9cd233d6fb22dc131583774135505ada800ebf14dfb0a7c40/pyzmq-27.1.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:15c8bd0fe0dabf808e2d7a681398c4e5ded70a551ab47482067a572c054c8e2e", size = 1657531, upload-time = "2025-09-08T23:07:52.795Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/21/e3180ca269ed4a0de5c34417dfe71a8ae80421198be83ee619a8a485b0c7/pyzmq-27.1.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:bafcb3dd171b4ae9f19ee6380dfc71ce0390fefaf26b504c0e5f628d7c8c54f2", size = 2034786, upload-time = "2025-09-08T23:07:55.047Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/b1/5e21d0b517434b7f33588ff76c177c5a167858cc38ef740608898cd329f2/pyzmq-27.1.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:e829529fcaa09937189178115c49c504e69289abd39967cd8a4c215761373394", size = 1894220, upload-time = "2025-09-08T23:07:57.172Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/5e/c3c49fdd0f535ef45eefcc16934648e9e59dace4a37ee88fc53f6cd8e641/pyzmq-27.1.0-cp312-abi3-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:1c179799b118e554b66da67d88ed66cd37a169f1f23b5d9f0a231b4e8d44a113", size = 895645, upload-time = "2025-09-08T23:08:05.301Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/e5/b0b2504cb4e903a74dcf1ebae157f9e20ebb6ea76095f6cfffea28c42ecd/pyzmq-27.1.0-cp312-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3837439b7f99e60312f0c926a6ad437b067356dc2bc2ec96eb395fd0fe804233", size = 652574, upload-time = "2025-09-08T23:08:06.828Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/9b/c108cdb55560eaf253f0cbdb61b29971e9fb34d9c3499b0e96e4e60ed8a5/pyzmq-27.1.0-cp312-abi3-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:43ad9a73e3da1fab5b0e7e13402f0b2fb934ae1c876c51d0afff0e7c052eca31", size = 840995, upload-time = "2025-09-08T23:08:08.396Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/bb/b79798ca177b9eb0825b4c9998c6af8cd2a7f15a6a1a4272c1d1a21d382f/pyzmq-27.1.0-cp312-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:0de3028d69d4cdc475bfe47a6128eb38d8bc0e8f4d69646adfbcd840facbac28", size = 1642070, upload-time = "2025-09-08T23:08:09.989Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/80/2df2e7977c4ede24c79ae39dcef3899bfc5f34d1ca7a5b24f182c9b7a9ca/pyzmq-27.1.0-cp312-abi3-musllinux_1_2_i686.whl", hash = "sha256:cf44a7763aea9298c0aa7dbf859f87ed7012de8bda0f3977b6fb1d96745df856", size = 2021121, upload-time = "2025-09-08T23:08:11.907Z" },
-    { url = "https://files.pythonhosted.org/packages/46/bd/2d45ad24f5f5ae7e8d01525eb76786fa7557136555cac7d929880519e33a/pyzmq-27.1.0-cp312-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:f30f395a9e6fbca195400ce833c731e7b64c3919aa481af4d88c3759e0cb7496", size = 1878550, upload-time = "2025-09-08T23:08:13.513Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/ca/2b8693d06b1db4e0c084871e4c9d7842b561d0a6ff9d780640f5e3e9eb55/pyzmq-27.1.0-cp39-cp39-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:49d3980544447f6bd2968b6ac913ab963a49dcaa2d4a2990041f16057b04c429", size = 906559, upload-time = "2025-09-08T23:09:22.983Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/b3/b99b39e2cfdcebd512959780e4d299447fd7f46010b1d88d63324e2481ec/pyzmq-27.1.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:849ca054d81aa1c175c49484afaaa5db0622092b5eccb2055f9f3bb8f703782d", size = 863816, upload-time = "2025-09-08T23:09:24.556Z" },
-    { url = "https://files.pythonhosted.org/packages/61/b2/018fa8e8eefb34a625b1a45e2effcbc9885645b22cdd0a68283f758351e7/pyzmq-27.1.0-cp39-cp39-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3970778e74cb7f85934d2b926b9900e92bfe597e62267d7499acc39c9c28e345", size = 666735, upload-time = "2025-09-08T23:09:26.297Z" },
-    { url = "https://files.pythonhosted.org/packages/01/05/8ae778f7cd7c94030731ae2305e6a38f3a333b6825f56c0c03f2134ccf1b/pyzmq-27.1.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:da96ecdcf7d3919c3be2de91a8c513c186f6762aa6cf7c01087ed74fad7f0968", size = 1655425, upload-time = "2025-09-08T23:09:28.172Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/ad/d69478a97a3f3142f9dbbbd9daa4fcf42541913a85567c36d4cfc19b2218/pyzmq-27.1.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:9541c444cfe1b1c0156c5c86ece2bb926c7079a18e7b47b0b1b3b1b875e5d098", size = 2033729, upload-time = "2025-09-08T23:09:30.097Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/6d/e3c6ad05bc1cddd25094e66cc15ae8924e15c67e231e93ed2955c401007e/pyzmq-27.1.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:e30a74a39b93e2e1591b58eb1acef4902be27c957a8720b0e368f579b82dc22f", size = 1891803, upload-time = "2025-09-08T23:09:31.875Z" },
-    { url = "https://files.pythonhosted.org/packages/58/ed/0202ca350f4f2b69faa95c6d931e3c05c3a397c184cacb84cb4f8f42f287/pyzmq-27.1.0-pp310-pypy310_pp73-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:f328d01128373cb6763823b2b4e7f73bdf767834268c565151eacb3b7a392f90", size = 800206, upload-time = "2025-09-08T23:09:41.902Z" },
-    { url = "https://files.pythonhosted.org/packages/47/42/1ff831fa87fe8f0a840ddb399054ca0009605d820e2b44ea43114f5459f4/pyzmq-27.1.0-pp310-pypy310_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9c1790386614232e1b3a40a958454bdd42c6d1811837b15ddbb052a032a43f62", size = 567747, upload-time = "2025-09-08T23:09:43.741Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/db/5c4d6807434751e3f21231bee98109aa57b9b9b55e058e450d0aef59b70f/pyzmq-27.1.0-pp310-pypy310_pp73-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:448f9cb54eb0cee4732b46584f2710c8bc178b0e5371d9e4fc8125201e413a74", size = 747371, upload-time = "2025-09-08T23:09:45.575Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/79/f38c92eeaeb03a2ccc2ba9866f0439593bb08c5e3b714ac1d553e5c96e25/pyzmq-27.1.0-pp311-pypy311_pp73-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:ac25465d42f92e990f8d8b0546b01c391ad431c3bf447683fdc40565941d0604", size = 800208, upload-time = "2025-09-08T23:09:51.073Z" },
-    { url = "https://files.pythonhosted.org/packages/49/0e/3f0d0d335c6b3abb9b7b723776d0b21fa7f3a6c819a0db6097059aada160/pyzmq-27.1.0-pp311-pypy311_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:53b40f8ae006f2734ee7608d59ed661419f087521edbfc2149c3932e9c14808c", size = 567747, upload-time = "2025-09-08T23:09:52.698Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/cf/f2b3784d536250ffd4be70e049f3b60981235d70c6e8ce7e3ef21e1adb25/pyzmq-27.1.0-pp311-pypy311_pp73-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f605d884e7c8be8fe1aa94e0a783bf3f591b84c24e4bc4f3e7564c82ac25e271", size = 747371, upload-time = "2025-09-08T23:09:54.563Z" },
-    { url = "https://files.pythonhosted.org/packages/5f/5f/4e10c7f57a4c92ab0fbb2396297aa8d618e6f5b9b8f8e9756d56f3e6fc52/pyzmq-27.1.0-pp39-pypy39_pp73-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:01f9437501886d3a1dd4b02ef59fb8cc384fa718ce066d52f175ee49dd5b7ed8", size = 800203, upload-time = "2025-09-08T23:10:09.436Z" },
-    { url = "https://files.pythonhosted.org/packages/19/72/a74a007cd636f903448c6ab66628104b1fc5f2ba018733d5eabb94a0a6fb/pyzmq-27.1.0-pp39-pypy39_pp73-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:4a19387a3dddcc762bfd2f570d14e2395b2c9701329b266f83dd87a2b3cbd381", size = 758756, upload-time = "2025-09-08T23:10:11.733Z" },
-    { url = "https://files.pythonhosted.org/packages/a9/d4/30c25b91f2b4786026372f5ef454134d7f576fcf4ac58539ad7dd5de4762/pyzmq-27.1.0-pp39-pypy39_pp73-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4c618fbcd069e3a29dcd221739cacde52edcc681f041907867e0f5cc7e85f172", size = 567742, upload-time = "2025-09-08T23:10:14.732Z" },
-]
-
 [[package]]
 name = "ray"
-version = "2.49.2"
+version = "2.52.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "filelock" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "filelock", version = "3.19.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "filelock", version = "3.20.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "jsonschema" },
     { name = "msgpack" },
     { name = "packaging" },
@@ -4170,99 +4169,114 @@ dependencies = [
     { name = "requests" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/63/0e/830df5a0f7e2b582422ee8ad0cdf2a2a9563aa63bb8e60be9ceec494981c/ray-2.49.2-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:eae07b3fed45f5b041a8bf9795cd26fad2464be5126efd447e4484905a29b677", size = 69125462, upload-time = "2025-09-19T19:14:51.029Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/85/a340eba596db3f66d3a338aff43942d8bac32732fb4cf4a20ed4bbbd07eb/ray-2.49.2-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:74566876af7bf4e48ea4b9b3b75b34db053d1064cc4d4b1670dc4ce78f6894af", size = 69935752, upload-time = "2025-09-19T19:14:56.191Z" },
-    { url = "https://files.pythonhosted.org/packages/38/67/93f0d6d558874a730581059eb6dfa8860991a5410502ea0685dba5e788e4/ray-2.49.2-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:eada9dd89ccda643a3c6c2cba7016b59898432d126e10b38fed52d74165364f4", size = 69266231, upload-time = "2025-09-19T19:15:16.92Z" },
-    { url = "https://files.pythonhosted.org/packages/c1/2b/f2efd0e7bcef06d51422db1af48cc5695a3f9b40a444f9d270a2d4663252/ray-2.49.2-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:54077dde338c5ffba349a4ab61b72352a3c3be69ea5b4f1b436d98d40b312763", size = 70070382, upload-time = "2025-09-19T19:15:22.048Z" },
-    { url = "https://files.pythonhosted.org/packages/da/99/23b732c0b7b2ee2ffd28bf632257fb98924a03251d251810cb637512fcab/ray-2.49.2-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:dd0d8d8641d142fafe6d83e87d3c19bd5637d21e34608d3ff69ad71ea3e2f462", size = 69287193, upload-time = "2025-09-19T19:15:42.093Z" },
-    { url = "https://files.pythonhosted.org/packages/69/ca/94791be5c3b68ed0df85589a8ca558334818a47bf2978000f85533245aed/ray-2.49.2-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:2ecaaa51f588ccdda2b61563a8be3843bf65dfaaa83a240588a307f4ebb82471", size = 70114942, upload-time = "2025-09-19T19:15:47.536Z" },
-    { url = "https://files.pythonhosted.org/packages/25/ca/04ae817999fc805fc68c0819c0fd8fe23c0b830520c74cf84e9768d31b5c/ray-2.49.2-cp39-cp39-manylinux2014_aarch64.whl", hash = "sha256:df99b7aafbfbe122befebfdca78facb297273260a1026752ff43af613b237939", size = 69125379, upload-time = "2025-09-19T19:16:29.125Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/fd/ac5da30b701cc6647b0c15da5107543932509c327fb541aef6c4b38c11bf/ray-2.49.2-cp39-cp39-manylinux2014_x86_64.whl", hash = "sha256:057e2e1408790d11d8931c46ed0ca6a0c125bc129811ea1c8a833ee945c36e13", size = 69933379, upload-time = "2025-09-19T19:16:34.774Z" },
-]
-
-[package.optional-dependencies]
-cgraph = [
-    { name = "cupy-cuda12x", marker = "sys_platform != 'darwin'" },
+    { url = "https://files.pythonhosted.org/packages/9d/7a/5fd23a04404b4ff38470d180e7a3d0db5f6f684c0507dd5befc4ec2f3e9a/ray-2.52.0-cp310-cp310-manylinux2014_aarch64.whl", hash = "sha256:0f55e217ee795cacb3884ba2eaf95e467fa29532626aa67f73c1a64b7f413d44", size = 71253101, upload-time = "2025-11-21T18:25:14.231Z" },
+    { url = "https://files.pythonhosted.org/packages/9d/2a/89b68cfa37e386093e91367ccfa67d899a837cb457a54d5a01070056491e/ray-2.52.0-cp310-cp310-manylinux2014_x86_64.whl", hash = "sha256:88f7ac912b78cb436b4a8e5b370ad116318f926d54f8731020032cfb09cfc7dc", size = 72084396, upload-time = "2025-11-21T18:25:19.587Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/58/397f375df6ac954d5cb4b335b38d4084608ba1d598fef3ccd22e1ff8f2f9/ray-2.52.0-cp311-cp311-manylinux2014_aarch64.whl", hash = "sha256:a990a3ae6cd13dba21edc3b564e2573af558bb82ab63d2d92532a436a878bf7c", size = 71373068, upload-time = "2025-11-21T18:25:34.838Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/b5/a60fb08b55a762458e6f999225908ba57eb8f7790cf7f888617cd01ebe6c/ray-2.52.0-cp311-cp311-manylinux2014_x86_64.whl", hash = "sha256:59a22d8ea21bad4e703ac5a3a1947c2fc09a43b4e9043666d9f8d396ceebb974", size = 72202426, upload-time = "2025-11-21T18:25:40.626Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/e2/77595b3e5d7879fc254cf29ba5b12662523220c20eedfc856d92b7d2fb88/ray-2.52.0-cp312-cp312-manylinux2014_aarch64.whl", hash = "sha256:f6edecbbce3c4eae335e97134d3ffac953f4633fa7c6f12e3dea69634918a8a2", size = 71411747, upload-time = "2025-11-21T18:25:55.741Z" },
+    { url = "https://files.pythonhosted.org/packages/db/ed/7539443ab9a33a2e276274e91460d3d588b00595771d33cc3fe970e759de/ray-2.52.0-cp312-cp312-manylinux2014_x86_64.whl", hash = "sha256:540c72d190ec827e3122488485a929a1a908460aa51fd2247696a85343b7ab16", size = 72267942, upload-time = "2025-11-21T18:26:01.339Z" },
 ]
 
 [[package]]
 name = "referencing"
 version = "0.36.2"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "attrs" },
-    { name = "rpds-py" },
-    { name = "typing-extensions" },
+    { name = "attrs", marker = "python_full_version < '3.10'" },
+    { name = "rpds-py", version = "0.27.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/2f/db/98b5c277be99dd18bfd91dd04e1b759cad18d1a338188c936e92f921c7e2/referencing-0.36.2.tar.gz", hash = "sha256:df2e89862cd09deabbdba16944cc3f10feb6b3e6f18e902f7cc25609a34775aa", size = 74744, upload-time = "2025-01-25T08:48:16.138Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/c1/b1/3baf80dc6d2b7bc27a95a67752d0208e410351e3feb4eb78de5f77454d8d/referencing-0.36.2-py3-none-any.whl", hash = "sha256:e8699adbbf8b5c7de96d8ffa0eb5c158b3beafce084968e2ea8bb08c6794dcd0", size = 26775, upload-time = "2025-01-25T08:48:14.241Z" },
 ]
 
+[[package]]
+name = "referencing"
+version = "0.37.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
+    { name = "attrs", marker = "python_full_version >= '3.10'" },
+    { name = "rpds-py", version = "0.29.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/22/f5/df4e9027acead3ecc63e50fe1e36aca1523e1719559c499951bb4b53188f/referencing-0.37.0.tar.gz", hash = "sha256:44aefc3142c5b842538163acb373e24cce6632bd54bdb01b21ad5863489f50d8", size = 78036, upload-time = "2025-10-13T15:30:48.871Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2c/58/ca301544e1fa93ed4f80d724bf5b194f6e4b945841c5bfd555878eea9fcb/referencing-0.37.0-py3-none-any.whl", hash = "sha256:381329a9f99628c9069361716891d34ad94af76e461dcb0335825aecc7692231", size = 26766, upload-time = "2025-10-13T15:30:47.625Z" },
+]
+
 [[package]]
 name = "regex"
-version = "2025.9.18"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/49/d3/eaa0d28aba6ad1827ad1e716d9a93e1ba963ada61887498297d3da715133/regex-2025.9.18.tar.gz", hash = "sha256:c5ba23274c61c6fef447ba6a39333297d0c247f53059dba0bca415cac511edc4", size = 400917, upload-time = "2025-09-19T00:38:35.79Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/7e/d8/7e06171db8e55f917c5b8e89319cea2d86982e3fc46b677f40358223dece/regex-2025.9.18-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:12296202480c201c98a84aecc4d210592b2f55e200a1d193235c4db92b9f6788", size = 484829, upload-time = "2025-09-19T00:35:05.215Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/70/bf91bb39e5bedf75ce730ffbaa82ca585584d13335306d637458946b8b9f/regex-2025.9.18-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:220381f1464a581f2ea988f2220cf2a67927adcef107d47d6897ba5a2f6d51a4", size = 288993, upload-time = "2025-09-19T00:35:08.154Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/89/69f79b28365eda2c46e64c39d617d5f65a2aa451a4c94de7d9b34c2dc80f/regex-2025.9.18-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:87f681bfca84ebd265278b5daa1dcb57f4db315da3b5d044add7c30c10442e61", size = 286624, upload-time = "2025-09-19T00:35:09.717Z" },
-    { url = "https://files.pythonhosted.org/packages/44/31/81e62955726c3a14fcc1049a80bc716765af6c055706869de5e880ddc783/regex-2025.9.18-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:34d674cbba70c9398074c8a1fcc1a79739d65d1105de2a3c695e2b05ea728251", size = 780473, upload-time = "2025-09-19T00:35:11.013Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/23/07072b7e191fbb6e213dc03b2f5b96f06d3c12d7deaded84679482926fc7/regex-2025.9.18-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:385c9b769655cb65ea40b6eea6ff763cbb6d69b3ffef0b0db8208e1833d4e746", size = 849290, upload-time = "2025-09-19T00:35:12.348Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/f0/aec7f6a01f2a112210424d77c6401b9015675fb887ced7e18926df4ae51e/regex-2025.9.18-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:8900b3208e022570ae34328712bef6696de0804c122933414014bae791437ab2", size = 897335, upload-time = "2025-09-19T00:35:14.058Z" },
-    { url = "https://files.pythonhosted.org/packages/cc/90/2e5f9da89d260de7d0417ead91a1bc897f19f0af05f4f9323313b76c47f2/regex-2025.9.18-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c204e93bf32cd7a77151d44b05eb36f469d0898e3fba141c026a26b79d9914a0", size = 789946, upload-time = "2025-09-19T00:35:15.403Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/d5/1c712c7362f2563d389be66bae131c8bab121a3fabfa04b0b5bfc9e73c51/regex-2025.9.18-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:3acc471d1dd7e5ff82e6cacb3b286750decd949ecd4ae258696d04f019817ef8", size = 780787, upload-time = "2025-09-19T00:35:17.061Z" },
-    { url = "https://files.pythonhosted.org/packages/4f/92/c54cdb4aa41009632e69817a5aa452673507f07e341076735a2f6c46a37c/regex-2025.9.18-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:6479d5555122433728760e5f29edb4c2b79655a8deb681a141beb5c8a025baea", size = 773632, upload-time = "2025-09-19T00:35:18.57Z" },
-    { url = "https://files.pythonhosted.org/packages/db/99/75c996dc6a2231a8652d7ad0bfbeaf8a8c77612d335580f520f3ec40e30b/regex-2025.9.18-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:431bd2a8726b000eb6f12429c9b438a24062a535d06783a93d2bcbad3698f8a8", size = 844104, upload-time = "2025-09-19T00:35:20.259Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/f7/25aba34cc130cb6844047dbfe9716c9b8f9629fee8b8bec331aa9241b97b/regex-2025.9.18-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:0cc3521060162d02bd36927e20690129200e5ac9d2c6d32b70368870b122db25", size = 834794, upload-time = "2025-09-19T00:35:22.002Z" },
-    { url = "https://files.pythonhosted.org/packages/51/eb/64e671beafa0ae29712268421597596d781704973551312b2425831d4037/regex-2025.9.18-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:a021217b01be2d51632ce056d7a837d3fa37c543ede36e39d14063176a26ae29", size = 778535, upload-time = "2025-09-19T00:35:23.298Z" },
-    { url = "https://files.pythonhosted.org/packages/26/33/c0ebc0b07bd0bf88f716cca240546b26235a07710ea58e271cfe390ae273/regex-2025.9.18-cp310-cp310-win32.whl", hash = "sha256:4a12a06c268a629cb67cc1d009b7bb0be43e289d00d5111f86a2efd3b1949444", size = 264115, upload-time = "2025-09-19T00:35:25.206Z" },
-    { url = "https://files.pythonhosted.org/packages/59/39/aeb11a4ae68faaec2498512cadae09f2d8a91f1f65730fe62b9bffeea150/regex-2025.9.18-cp310-cp310-win_amd64.whl", hash = "sha256:47acd811589301298c49db2c56bde4f9308d6396da92daf99cba781fa74aa450", size = 276143, upload-time = "2025-09-19T00:35:26.785Z" },
-    { url = "https://files.pythonhosted.org/packages/29/04/37f2d3fc334a1031fc2767c9d89cec13c2e72207c7e7f6feae8a47f4e149/regex-2025.9.18-cp310-cp310-win_arm64.whl", hash = "sha256:16bd2944e77522275e5ee36f867e19995bcaa533dcb516753a26726ac7285442", size = 268473, upload-time = "2025-09-19T00:35:28.39Z" },
-    { url = "https://files.pythonhosted.org/packages/58/61/80eda662fc4eb32bfedc331f42390974c9e89c7eac1b79cd9eea4d7c458c/regex-2025.9.18-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:51076980cd08cd13c88eb7365427ae27f0d94e7cebe9ceb2bb9ffdae8fc4d82a", size = 484832, upload-time = "2025-09-19T00:35:30.011Z" },
-    { url = "https://files.pythonhosted.org/packages/a6/d9/33833d9abddf3f07ad48504ddb53fe3b22f353214bbb878a72eee1e3ddbf/regex-2025.9.18-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:828446870bd7dee4e0cbeed767f07961aa07f0ea3129f38b3ccecebc9742e0b8", size = 288994, upload-time = "2025-09-19T00:35:31.733Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/b3/526ee96b0d70ea81980cbc20c3496fa582f775a52e001e2743cc33b2fa75/regex-2025.9.18-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:c28821d5637866479ec4cc23b8c990f5bc6dd24e5e4384ba4a11d38a526e1414", size = 286619, upload-time = "2025-09-19T00:35:33.221Z" },
-    { url = "https://files.pythonhosted.org/packages/65/4f/c2c096b02a351b33442aed5895cdd8bf87d372498d2100927c5a053d7ba3/regex-2025.9.18-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:726177ade8e481db669e76bf99de0b278783be8acd11cef71165327abd1f170a", size = 792454, upload-time = "2025-09-19T00:35:35.361Z" },
-    { url = "https://files.pythonhosted.org/packages/24/15/b562c9d6e47c403c4b5deb744f8b4bf6e40684cf866c7b077960a925bdff/regex-2025.9.18-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f5cca697da89b9f8ea44115ce3130f6c54c22f541943ac8e9900461edc2b8bd4", size = 858723, upload-time = "2025-09-19T00:35:36.949Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/01/dba305409849e85b8a1a681eac4c03ed327d8de37895ddf9dc137f59c140/regex-2025.9.18-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:dfbde38f38004703c35666a1e1c088b778e35d55348da2b7b278914491698d6a", size = 905899, upload-time = "2025-09-19T00:35:38.723Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/d0/c51d1e6a80eab11ef96a4cbad17fc0310cf68994fb01a7283276b7e5bbd6/regex-2025.9.18-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f2f422214a03fab16bfa495cfec72bee4aaa5731843b771860a471282f1bf74f", size = 798981, upload-time = "2025-09-19T00:35:40.416Z" },
-    { url = "https://files.pythonhosted.org/packages/c4/5e/72db90970887bbe02296612bd61b0fa31e6d88aa24f6a4853db3e96c575e/regex-2025.9.18-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:a295916890f4df0902e4286bc7223ee7f9e925daa6dcdec4192364255b70561a", size = 781900, upload-time = "2025-09-19T00:35:42.077Z" },
-    { url = "https://files.pythonhosted.org/packages/50/ff/596be45eea8e9bc31677fde243fa2904d00aad1b32c31bce26c3dbba0b9e/regex-2025.9.18-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:5db95ff632dbabc8c38c4e82bf545ab78d902e81160e6e455598014f0abe66b9", size = 852952, upload-time = "2025-09-19T00:35:43.751Z" },
-    { url = "https://files.pythonhosted.org/packages/e5/1b/2dfa348fa551e900ed3f5f63f74185b6a08e8a76bc62bc9c106f4f92668b/regex-2025.9.18-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:fb967eb441b0f15ae610b7069bdb760b929f267efbf522e814bbbfffdf125ce2", size = 844355, upload-time = "2025-09-19T00:35:45.309Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/bf/aefb1def27fe33b8cbbb19c75c13aefccfbef1c6686f8e7f7095705969c7/regex-2025.9.18-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:f04d2f20da4053d96c08f7fde6e1419b7ec9dbcee89c96e3d731fca77f411b95", size = 787254, upload-time = "2025-09-19T00:35:46.904Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/4e/8ef042e7cf0dbbb401e784e896acfc1b367b95dfbfc9ada94c2ed55a081f/regex-2025.9.18-cp311-cp311-win32.whl", hash = "sha256:895197241fccf18c0cea7550c80e75f185b8bd55b6924fcae269a1a92c614a07", size = 264129, upload-time = "2025-09-19T00:35:48.597Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/7d/c4fcabf80dcdd6821c0578ad9b451f8640b9110fb3dcb74793dd077069ff/regex-2025.9.18-cp311-cp311-win_amd64.whl", hash = "sha256:7e2b414deae99166e22c005e154a5513ac31493db178d8aec92b3269c9cce8c9", size = 276160, upload-time = "2025-09-19T00:36:00.45Z" },
-    { url = "https://files.pythonhosted.org/packages/64/f8/0e13c8ae4d6df9d128afaba138342d532283d53a4c1e7a8c93d6756c8f4a/regex-2025.9.18-cp311-cp311-win_arm64.whl", hash = "sha256:fb137ec7c5c54f34a25ff9b31f6b7b0c2757be80176435bf367111e3f71d72df", size = 268471, upload-time = "2025-09-19T00:36:02.149Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/99/05859d87a66ae7098222d65748f11ef7f2dff51bfd7482a4e2256c90d72b/regex-2025.9.18-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:436e1b31d7efd4dcd52091d076482031c611dde58bf9c46ca6d0a26e33053a7e", size = 486335, upload-time = "2025-09-19T00:36:03.661Z" },
-    { url = "https://files.pythonhosted.org/packages/97/7e/d43d4e8b978890932cf7b0957fce58c5b08c66f32698f695b0c2c24a48bf/regex-2025.9.18-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:c190af81e5576b9c5fdc708f781a52ff20f8b96386c6e2e0557a78402b029f4a", size = 289720, upload-time = "2025-09-19T00:36:05.471Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/3b/ff80886089eb5dcf7e0d2040d9aaed539e25a94300403814bb24cc775058/regex-2025.9.18-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:e4121f1ce2b2b5eec4b397cc1b277686e577e658d8f5870b7eb2d726bd2300ab", size = 287257, upload-time = "2025-09-19T00:36:07.072Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/66/243edf49dd8720cba8d5245dd4d6adcb03a1defab7238598c0c97cf549b8/regex-2025.9.18-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:300e25dbbf8299d87205e821a201057f2ef9aa3deb29caa01cd2cac669e508d5", size = 797463, upload-time = "2025-09-19T00:36:08.399Z" },
-    { url = "https://files.pythonhosted.org/packages/df/71/c9d25a1142c70432e68bb03211d4a82299cd1c1fbc41db9409a394374ef5/regex-2025.9.18-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7b47fcf9f5316c0bdaf449e879407e1b9937a23c3b369135ca94ebc8d74b1742", size = 862670, upload-time = "2025-09-19T00:36:10.101Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/8f/329b1efc3a64375a294e3a92d43372bf1a351aa418e83c21f2f01cf6ec41/regex-2025.9.18-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:57a161bd3acaa4b513220b49949b07e252165e6b6dc910ee7617a37ff4f5b425", size = 910881, upload-time = "2025-09-19T00:36:12.223Z" },
-    { url = "https://files.pythonhosted.org/packages/35/9e/a91b50332a9750519320ed30ec378b74c996f6befe282cfa6bb6cea7e9fd/regex-2025.9.18-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4f130c3a7845ba42de42f380fff3c8aebe89a810747d91bcf56d40a069f15352", size = 802011, upload-time = "2025-09-19T00:36:13.901Z" },
-    { url = "https://files.pythonhosted.org/packages/a4/1d/6be3b8d7856b6e0d7ee7f942f437d0a76e0d5622983abbb6d21e21ab9a17/regex-2025.9.18-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5f96fa342b6f54dcba928dd452e8d8cb9f0d63e711d1721cd765bb9f73bb048d", size = 786668, upload-time = "2025-09-19T00:36:15.391Z" },
-    { url = "https://files.pythonhosted.org/packages/cb/ce/4a60e53df58bd157c5156a1736d3636f9910bdcc271d067b32b7fcd0c3a8/regex-2025.9.18-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:0f0d676522d68c207828dcd01fb6f214f63f238c283d9f01d85fc664c7c85b56", size = 856578, upload-time = "2025-09-19T00:36:16.845Z" },
-    { url = "https://files.pythonhosted.org/packages/86/e8/162c91bfe7217253afccde112868afb239f94703de6580fb235058d506a6/regex-2025.9.18-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:40532bff8a1a0621e7903ae57fce88feb2e8a9a9116d341701302c9302aef06e", size = 849017, upload-time = "2025-09-19T00:36:18.597Z" },
-    { url = "https://files.pythonhosted.org/packages/35/34/42b165bc45289646ea0959a1bc7531733e90b47c56a72067adfe6b3251f6/regex-2025.9.18-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:039f11b618ce8d71a1c364fdee37da1012f5a3e79b1b2819a9f389cd82fd6282", size = 788150, upload-time = "2025-09-19T00:36:20.464Z" },
-    { url = "https://files.pythonhosted.org/packages/79/5d/cdd13b1f3c53afa7191593a7ad2ee24092a5a46417725ffff7f64be8342d/regex-2025.9.18-cp312-cp312-win32.whl", hash = "sha256:e1dd06f981eb226edf87c55d523131ade7285137fbde837c34dc9d1bf309f459", size = 264536, upload-time = "2025-09-19T00:36:21.922Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/f5/4a7770c9a522e7d2dc1fa3ffc83ab2ab33b0b22b447e62cffef186805302/regex-2025.9.18-cp312-cp312-win_amd64.whl", hash = "sha256:3d86b5247bf25fa3715e385aa9ff272c307e0636ce0c9595f64568b41f0a9c77", size = 275501, upload-time = "2025-09-19T00:36:23.4Z" },
-    { url = "https://files.pythonhosted.org/packages/df/05/9ce3e110e70d225ecbed455b966003a3afda5e58e8aec2964042363a18f4/regex-2025.9.18-cp312-cp312-win_arm64.whl", hash = "sha256:032720248cbeeae6444c269b78cb15664458b7bb9ed02401d3da59fe4d68c3a5", size = 268601, upload-time = "2025-09-19T00:36:25.092Z" },
-    { url = "https://files.pythonhosted.org/packages/ed/d2/5b0ded10467d6e96f78de5e6f195b7f9b57251f411b1090004597cffe5d9/regex-2025.9.18-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:3dbcfcaa18e9480669030d07371713c10b4f1a41f791ffa5cb1a99f24e777f40", size = 484847, upload-time = "2025-09-19T00:38:07.367Z" },
-    { url = "https://files.pythonhosted.org/packages/55/35/051da2c0ae6124e3f1aa1442ecc2bb4e2de930e95433bce1301a2e7ae255/regex-2025.9.18-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:1e85f73ef7095f0380208269055ae20524bfde3f27c5384126ddccf20382a638", size = 288995, upload-time = "2025-09-19T00:38:09.253Z" },
-    { url = "https://files.pythonhosted.org/packages/22/4b/4bfc51cad95263d25b6ed8c5253831b2536e8e279e6736d0a08c9f7ffe98/regex-2025.9.18-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:9098e29b3ea4ffffeade423f6779665e2a4f8db64e699c0ed737ef0db6ba7b12", size = 286642, upload-time = "2025-09-19T00:38:11.012Z" },
-    { url = "https://files.pythonhosted.org/packages/0e/67/d2f3e2483e09d1e9f7d93b4fe106b04933fba5e619bc901530d1c90d62da/regex-2025.9.18-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:90b6b7a2d0f45b7ecaaee1aec6b362184d6596ba2092dd583ffba1b78dd0231c", size = 779896, upload-time = "2025-09-19T00:38:12.732Z" },
-    { url = "https://files.pythonhosted.org/packages/14/5e/49a4f07ce6f5563de02b0e321220b9534f3fd3bae275311b785dd618aea5/regex-2025.9.18-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c81b892af4a38286101502eae7aec69f7cd749a893d9987a92776954f3943408", size = 848954, upload-time = "2025-09-19T00:38:14.716Z" },
-    { url = "https://files.pythonhosted.org/packages/00/8d/f5995ae51225c77ca9215d78ceb1dc30c52fa2b22c41dac977214e8b4bbd/regex-2025.9.18-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:3b524d010973f2e1929aeb635418d468d869a5f77b52084d9f74c272189c251d", size = 896770, upload-time = "2025-09-19T00:38:16.381Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/15/2a3a744d73a557337c7561db2114bab10b4e9941c626c03169ea62f42c8f/regex-2025.9.18-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6b498437c026a3d5d0be0020023ff76d70ae4d77118e92f6f26c9d0423452446", size = 789484, upload-time = "2025-09-19T00:38:18.183Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/27/e425f3d17d32062a657b836d0c8a68f5e71a9e6295fa637159f265eaa609/regex-2025.9.18-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:0716e4d6e58853d83f6563f3cf25c281ff46cf7107e5f11879e32cb0b59797d9", size = 780150, upload-time = "2025-09-19T00:38:19.879Z" },
-    { url = "https://files.pythonhosted.org/packages/62/28/79dfae89b6fd7901b82611ac1a96ec25deceb7e918e9c5eb3f96cf5ad654/regex-2025.9.18-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:065b6956749379d41db2625f880b637d4acc14c0a4de0d25d609a62850e96d36", size = 773160, upload-time = "2025-09-19T00:38:21.641Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/67/df83d6ae608f487448e9be7ac26211af2afa2b6e34465fde3e07d1f11290/regex-2025.9.18-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:d4a691494439287c08ddb9b5793da605ee80299dd31e95fa3f323fac3c33d9d4", size = 843555, upload-time = "2025-09-19T00:38:23.696Z" },
-    { url = "https://files.pythonhosted.org/packages/32/67/c65f56f3edd3f213d3aa41e9b9b07cc2247721a23d34bcfb2947dc0f4685/regex-2025.9.18-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:ef8d10cc0989565bcbe45fb4439f044594d5c2b8919d3d229ea2c4238f1d55b0", size = 834169, upload-time = "2025-09-19T00:38:25.997Z" },
-    { url = "https://files.pythonhosted.org/packages/95/90/7fca37435e3aa1a032c38fa1e171fdaf809c8dbf2717508e3f6a92c75446/regex-2025.9.18-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:4baeb1b16735ac969a7eeecc216f1f8b7caf60431f38a2671ae601f716a32d25", size = 778024, upload-time = "2025-09-19T00:38:28.043Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/05/c2ee512cdf34d6be5ac5cf938a58c1b79a9d96cbad404bc4d70404212edb/regex-2025.9.18-cp39-cp39-win32.whl", hash = "sha256:8e5f41ad24a1e0b5dfcf4c4e5d9f5bd54c895feb5708dd0c1d0d35693b24d478", size = 264151, upload-time = "2025-09-19T00:38:30.23Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/2f/8414fb46181b6108484f04d670ece196db6734cc4c683f41125043fd3280/regex-2025.9.18-cp39-cp39-win_amd64.whl", hash = "sha256:50e8290707f2fb8e314ab3831e594da71e062f1d623b05266f8cfe4db4949afd", size = 276232, upload-time = "2025-09-19T00:38:31.981Z" },
-    { url = "https://files.pythonhosted.org/packages/61/63/f40931d477e1ed4b53105d506758a58cfec1b052c12972054930ec743ee5/regex-2025.9.18-cp39-cp39-win_arm64.whl", hash = "sha256:039a9d7195fd88c943d7c777d4941e8ef736731947becce773c31a1009cb3c35", size = 268505, upload-time = "2025-09-19T00:38:34.015Z" },
+version = "2025.11.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/cc/a9/546676f25e573a4cf00fe8e119b78a37b6a8fe2dc95cda877b30889c9c45/regex-2025.11.3.tar.gz", hash = "sha256:1fedc720f9bb2494ce31a58a1631f9c82df6a09b49c19517ea5cc280b4541e01", size = 414669, upload-time = "2025-11-03T21:34:22.089Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8a/d6/d788d52da01280a30a3f6268aef2aa71043bff359c618fea4c5b536654d5/regex-2025.11.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:2b441a4ae2c8049106e8b39973bfbddfb25a179dda2bdb99b0eeb60c40a6a3af", size = 488087, upload-time = "2025-11-03T21:30:47.317Z" },
+    { url = "https://files.pythonhosted.org/packages/69/39/abec3bd688ec9bbea3562de0fd764ff802976185f5ff22807bf0a2697992/regex-2025.11.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:2fa2eed3f76677777345d2f81ee89f5de2f5745910e805f7af7386a920fa7313", size = 290544, upload-time = "2025-11-03T21:30:49.912Z" },
+    { url = "https://files.pythonhosted.org/packages/39/b3/9a231475d5653e60002508f41205c61684bb2ffbf2401351ae2186897fc4/regex-2025.11.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:d8b4a27eebd684319bdf473d39f1d79eed36bf2cd34bd4465cdb4618d82b3d56", size = 288408, upload-time = "2025-11-03T21:30:51.344Z" },
+    { url = "https://files.pythonhosted.org/packages/c3/c5/1929a0491bd5ac2d1539a866768b88965fa8c405f3e16a8cef84313098d6/regex-2025.11.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5cf77eac15bd264986c4a2c63353212c095b40f3affb2bc6b4ef80c4776c1a28", size = 781584, upload-time = "2025-11-03T21:30:52.596Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/fd/16aa16cf5d497ef727ec966f74164fbe75d6516d3d58ac9aa989bc9cdaad/regex-2025.11.3-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:b7f9ee819f94c6abfa56ec7b1dbab586f41ebbdc0a57e6524bd5e7f487a878c7", size = 850733, upload-time = "2025-11-03T21:30:53.825Z" },
+    { url = "https://files.pythonhosted.org/packages/e6/49/3294b988855a221cb6565189edf5dc43239957427df2d81d4a6b15244f64/regex-2025.11.3-cp310-cp310-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:838441333bc90b829406d4a03cb4b8bf7656231b84358628b0406d803931ef32", size = 898691, upload-time = "2025-11-03T21:30:55.575Z" },
+    { url = "https://files.pythonhosted.org/packages/14/62/b56d29e70b03666193369bdbdedfdc23946dbe9f81dd78ce262c74d988ab/regex-2025.11.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:cfe6d3f0c9e3b7e8c0c694b24d25e677776f5ca26dce46fd6b0489f9c8339391", size = 791662, upload-time = "2025-11-03T21:30:57.262Z" },
+    { url = "https://files.pythonhosted.org/packages/15/fc/e4c31d061eced63fbf1ce9d853975f912c61a7d406ea14eda2dd355f48e7/regex-2025.11.3-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:2ab815eb8a96379a27c3b6157fcb127c8f59c36f043c1678110cea492868f1d5", size = 782587, upload-time = "2025-11-03T21:30:58.788Z" },
+    { url = "https://files.pythonhosted.org/packages/b2/bb/5e30c7394bcf63f0537121c23e796be67b55a8847c3956ae6068f4c70702/regex-2025.11.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:728a9d2d173a65b62bdc380b7932dd8e74ed4295279a8fe1021204ce210803e7", size = 774709, upload-time = "2025-11-03T21:31:00.081Z" },
+    { url = "https://files.pythonhosted.org/packages/c5/c4/fce773710af81b0cb37cb4ff0947e75d5d17dee304b93d940b87a67fc2f4/regex-2025.11.3-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:509dc827f89c15c66a0c216331260d777dd6c81e9a4e4f830e662b0bb296c313", size = 845773, upload-time = "2025-11-03T21:31:01.583Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/5e/9466a7ec4b8ec282077095c6eb50a12a389d2e036581134d4919e8ca518c/regex-2025.11.3-cp310-cp310-musllinux_1_2_s390x.whl", hash = "sha256:849202cd789e5f3cf5dcc7822c34b502181b4824a65ff20ce82da5524e45e8e9", size = 836164, upload-time = "2025-11-03T21:31:03.244Z" },
+    { url = "https://files.pythonhosted.org/packages/95/18/82980a60e8ed1594eb3c89eb814fb276ef51b9af7caeab1340bfd8564af6/regex-2025.11.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b6f78f98741dcc89607c16b1e9426ee46ce4bf31ac5e6b0d40e81c89f3481ea5", size = 779832, upload-time = "2025-11-03T21:31:04.876Z" },
+    { url = "https://files.pythonhosted.org/packages/03/cc/90ab0fdbe6dce064a42015433f9152710139fb04a8b81b4fb57a1cb63ffa/regex-2025.11.3-cp310-cp310-win32.whl", hash = "sha256:149eb0bba95231fb4f6d37c8f760ec9fa6fabf65bab555e128dde5f2475193ec", size = 265802, upload-time = "2025-11-03T21:31:06.581Z" },
+    { url = "https://files.pythonhosted.org/packages/34/9d/e9e8493a85f3b1ddc4a5014465f5c2b78c3ea1cbf238dcfde78956378041/regex-2025.11.3-cp310-cp310-win_amd64.whl", hash = "sha256:ee3a83ce492074c35a74cc76cf8235d49e77b757193a5365ff86e3f2f93db9fd", size = 277722, upload-time = "2025-11-03T21:31:08.144Z" },
+    { url = "https://files.pythonhosted.org/packages/15/c4/b54b24f553966564506dbf873a3e080aef47b356a3b39b5d5aba992b50db/regex-2025.11.3-cp310-cp310-win_arm64.whl", hash = "sha256:38af559ad934a7b35147716655d4a2f79fcef2d695ddfe06a06ba40ae631fa7e", size = 270289, upload-time = "2025-11-03T21:31:10.267Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/90/4fb5056e5f03a7048abd2b11f598d464f0c167de4f2a51aa868c376b8c70/regex-2025.11.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:eadade04221641516fa25139273505a1c19f9bf97589a05bc4cfcd8b4a618031", size = 488081, upload-time = "2025-11-03T21:31:11.946Z" },
+    { url = "https://files.pythonhosted.org/packages/85/23/63e481293fac8b069d84fba0299b6666df720d875110efd0338406b5d360/regex-2025.11.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:feff9e54ec0dd3833d659257f5c3f5322a12eee58ffa360984b716f8b92983f4", size = 290554, upload-time = "2025-11-03T21:31:13.387Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/9d/b101d0262ea293a0066b4522dfb722eb6a8785a8c3e084396a5f2c431a46/regex-2025.11.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3b30bc921d50365775c09a7ed446359e5c0179e9e2512beec4a60cbcef6ddd50", size = 288407, upload-time = "2025-11-03T21:31:14.809Z" },
+    { url = "https://files.pythonhosted.org/packages/0c/64/79241c8209d5b7e00577ec9dca35cd493cc6be35b7d147eda367d6179f6d/regex-2025.11.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f99be08cfead2020c7ca6e396c13543baea32343b7a9a5780c462e323bd8872f", size = 793418, upload-time = "2025-11-03T21:31:16.556Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/e2/23cd5d3573901ce8f9757c92ca4db4d09600b865919b6d3e7f69f03b1afd/regex-2025.11.3-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:6dd329a1b61c0ee95ba95385fb0c07ea0d3fe1a21e1349fa2bec272636217118", size = 860448, upload-time = "2025-11-03T21:31:18.12Z" },
+    { url = "https://files.pythonhosted.org/packages/2a/4c/aecf31beeaa416d0ae4ecb852148d38db35391aac19c687b5d56aedf3a8b/regex-2025.11.3-cp311-cp311-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4c5238d32f3c5269d9e87be0cf096437b7622b6920f5eac4fd202468aaeb34d2", size = 907139, upload-time = "2025-11-03T21:31:20.753Z" },
+    { url = "https://files.pythonhosted.org/packages/61/22/b8cb00df7d2b5e0875f60628594d44dba283e951b1ae17c12f99e332cc0a/regex-2025.11.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:10483eefbfb0adb18ee9474498c9a32fcf4e594fbca0543bb94c48bac6183e2e", size = 800439, upload-time = "2025-11-03T21:31:22.069Z" },
+    { url = "https://files.pythonhosted.org/packages/02/a8/c4b20330a5cdc7a8eb265f9ce593f389a6a88a0c5f280cf4d978f33966bc/regex-2025.11.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:78c2d02bb6e1da0720eedc0bad578049cad3f71050ef8cd065ecc87691bed2b0", size = 782965, upload-time = "2025-11-03T21:31:23.598Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/4c/ae3e52988ae74af4b04d2af32fee4e8077f26e51b62ec2d12d246876bea2/regex-2025.11.3-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:e6b49cd2aad93a1790ce9cffb18964f6d3a4b0b3dbdbd5de094b65296fce6e58", size = 854398, upload-time = "2025-11-03T21:31:25.008Z" },
+    { url = "https://files.pythonhosted.org/packages/06/d1/a8b9cf45874eda14b2e275157ce3b304c87e10fb38d9fc26a6e14eb18227/regex-2025.11.3-cp311-cp311-musllinux_1_2_s390x.whl", hash = "sha256:885b26aa3ee56433b630502dc3d36ba78d186a00cc535d3806e6bfd9ed3c70ab", size = 845897, upload-time = "2025-11-03T21:31:26.427Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/fe/1830eb0236be93d9b145e0bd8ab499f31602fe0999b1f19e99955aa8fe20/regex-2025.11.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:ddd76a9f58e6a00f8772e72cff8ebcff78e022be95edf018766707c730593e1e", size = 788906, upload-time = "2025-11-03T21:31:28.078Z" },
+    { url = "https://files.pythonhosted.org/packages/66/47/dc2577c1f95f188c1e13e2e69d8825a5ac582ac709942f8a03af42ed6e93/regex-2025.11.3-cp311-cp311-win32.whl", hash = "sha256:3e816cc9aac1cd3cc9a4ec4d860f06d40f994b5c7b4d03b93345f44e08cc68bf", size = 265812, upload-time = "2025-11-03T21:31:29.72Z" },
+    { url = "https://files.pythonhosted.org/packages/50/1e/15f08b2f82a9bbb510621ec9042547b54d11e83cb620643ebb54e4eb7d71/regex-2025.11.3-cp311-cp311-win_amd64.whl", hash = "sha256:087511f5c8b7dfbe3a03f5d5ad0c2a33861b1fc387f21f6f60825a44865a385a", size = 277737, upload-time = "2025-11-03T21:31:31.422Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/fc/6500eb39f5f76c5e47a398df82e6b535a5e345f839581012a418b16f9cc3/regex-2025.11.3-cp311-cp311-win_arm64.whl", hash = "sha256:1ff0d190c7f68ae7769cd0313fe45820ba07ffebfddfaa89cc1eb70827ba0ddc", size = 270290, upload-time = "2025-11-03T21:31:33.041Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/74/18f04cb53e58e3fb107439699bd8375cf5a835eec81084e0bddbd122e4c2/regex-2025.11.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:bc8ab71e2e31b16e40868a40a69007bc305e1109bd4658eb6cad007e0bf67c41", size = 489312, upload-time = "2025-11-03T21:31:34.343Z" },
+    { url = "https://files.pythonhosted.org/packages/78/3f/37fcdd0d2b1e78909108a876580485ea37c91e1acf66d3bb8e736348f441/regex-2025.11.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:22b29dda7e1f7062a52359fca6e58e548e28c6686f205e780b02ad8ef710de36", size = 291256, upload-time = "2025-11-03T21:31:35.675Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/26/0a575f58eb23b7ebd67a45fccbc02ac030b737b896b7e7a909ffe43ffd6a/regex-2025.11.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3a91e4a29938bc1a082cc28fdea44be420bf2bebe2665343029723892eb073e1", size = 288921, upload-time = "2025-11-03T21:31:37.07Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/98/6a8dff667d1af907150432cf5abc05a17ccd32c72a3615410d5365ac167a/regex-2025.11.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:08b884f4226602ad40c5d55f52bf91a9df30f513864e0054bad40c0e9cf1afb7", size = 798568, upload-time = "2025-11-03T21:31:38.784Z" },
+    { url = "https://files.pythonhosted.org/packages/64/15/92c1db4fa4e12733dd5a526c2dd2b6edcbfe13257e135fc0f6c57f34c173/regex-2025.11.3-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3e0b11b2b2433d1c39c7c7a30e3f3d0aeeea44c2a8d0bae28f6b95f639927a69", size = 864165, upload-time = "2025-11-03T21:31:40.559Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/e7/3ad7da8cdee1ce66c7cd37ab5ab05c463a86ffeb52b1a25fe7bd9293b36c/regex-2025.11.3-cp312-cp312-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:87eb52a81ef58c7ba4d45c3ca74e12aa4b4e77816f72ca25258a85b3ea96cb48", size = 912182, upload-time = "2025-11-03T21:31:42.002Z" },
+    { url = "https://files.pythonhosted.org/packages/84/bd/9ce9f629fcb714ffc2c3faf62b6766ecb7a585e1e885eb699bcf130a5209/regex-2025.11.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a12ab1f5c29b4e93db518f5e3872116b7e9b1646c9f9f426f777b50d44a09e8c", size = 803501, upload-time = "2025-11-03T21:31:43.815Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/0f/8dc2e4349d8e877283e6edd6c12bdcebc20f03744e86f197ab6e4492bf08/regex-2025.11.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:7521684c8c7c4f6e88e35ec89680ee1aa8358d3f09d27dfbdf62c446f5d4c695", size = 787842, upload-time = "2025-11-03T21:31:45.353Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/73/cff02702960bc185164d5619c0c62a2f598a6abff6695d391b096237d4ab/regex-2025.11.3-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:7fe6e5440584e94cc4b3f5f4d98a25e29ca12dccf8873679a635638349831b98", size = 858519, upload-time = "2025-11-03T21:31:46.814Z" },
+    { url = "https://files.pythonhosted.org/packages/61/83/0e8d1ae71e15bc1dc36231c90b46ee35f9d52fab2e226b0e039e7ea9c10a/regex-2025.11.3-cp312-cp312-musllinux_1_2_s390x.whl", hash = "sha256:8e026094aa12b43f4fd74576714e987803a315c76edb6b098b9809db5de58f74", size = 850611, upload-time = "2025-11-03T21:31:48.289Z" },
+    { url = "https://files.pythonhosted.org/packages/c8/f5/70a5cdd781dcfaa12556f2955bf170cd603cb1c96a1827479f8faea2df97/regex-2025.11.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:435bbad13e57eb5606a68443af62bed3556de2f46deb9f7d4237bc2f1c9fb3a0", size = 789759, upload-time = "2025-11-03T21:31:49.759Z" },
+    { url = "https://files.pythonhosted.org/packages/59/9b/7c29be7903c318488983e7d97abcf8ebd3830e4c956c4c540005fcfb0462/regex-2025.11.3-cp312-cp312-win32.whl", hash = "sha256:3839967cf4dc4b985e1570fd8d91078f0c519f30491c60f9ac42a8db039be204", size = 266194, upload-time = "2025-11-03T21:31:51.53Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/67/3b92df89f179d7c367be654ab5626ae311cb28f7d5c237b6bb976cd5fbbb/regex-2025.11.3-cp312-cp312-win_amd64.whl", hash = "sha256:e721d1b46e25c481dc5ded6f4b3f66c897c58d2e8cfdf77bbced84339108b0b9", size = 277069, upload-time = "2025-11-03T21:31:53.151Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/55/85ba4c066fe5094d35b249c3ce8df0ba623cfd35afb22d6764f23a52a1c5/regex-2025.11.3-cp312-cp312-win_arm64.whl", hash = "sha256:64350685ff08b1d3a6fff33f45a9ca183dc1d58bbfe4981604e70ec9801bbc26", size = 270330, upload-time = "2025-11-03T21:31:54.514Z" },
+    { url = "https://files.pythonhosted.org/packages/13/a2/71d9e3de7a8ba4dd301ce6327d3feacc911413dc0660a4e2256a8b43401b/regex-2025.11.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:81519e25707fc076978c6143b81ea3dc853f176895af05bf7ec51effe818aeec", size = 488097, upload-time = "2025-11-03T21:33:47.455Z" },
+    { url = "https://files.pythonhosted.org/packages/68/c2/c74c874de5c3e73302ea1f47968f2a990c2517c6fb827c95d2b857e01098/regex-2025.11.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:3bf28b1873a8af8bbb58c26cc56ea6e534d80053b41fb511a35795b6de507e6a", size = 290540, upload-time = "2025-11-03T21:33:49.296Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/4d/97cc016a7f705a496328d4d569db2023a200c2c391b0a0ddf36e66dd80f5/regex-2025.11.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:856a25c73b697f2ce2a24e7968285579e62577a048526161a2c0f53090bea9f9", size = 288419, upload-time = "2025-11-03T21:33:51.309Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/e6/725cd713a926a13f55bb9c4585d4cca51ab7d9ef77cb96eba8398e207418/regex-2025.11.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8a3d571bd95fade53c86c0517f859477ff3a93c3fde10c9e669086f038e0f207", size = 781001, upload-time = "2025-11-03T21:33:53.375Z" },
+    { url = "https://files.pythonhosted.org/packages/15/15/ff1e7428b64a66c83dd53a88294c3e1c3e6a08653d9670efd24cbb068fa5/regex-2025.11.3-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:732aea6de26051af97b94bc98ed86448821f839d058e5d259c72bf6d73ad0fc0", size = 850430, upload-time = "2025-11-03T21:33:55.362Z" },
+    { url = "https://files.pythonhosted.org/packages/89/1d/d4045ff7c563eb73be4419784a06efc7de02e15959b917fee335de8c10de/regex-2025.11.3-cp39-cp39-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:51c1c1847128238f54930edb8805b660305dca164645a9fd29243f5610beea34", size = 898038, upload-time = "2025-11-03T21:33:57.73Z" },
+    { url = "https://files.pythonhosted.org/packages/60/ee/e9c71bdf334edc14ff769463bd6173966b0445e442a28b18f790b84032f5/regex-2025.11.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:22dd622a402aad4558277305350699b2be14bc59f64d64ae1d928ce7d072dced", size = 791201, upload-time = "2025-11-03T21:33:59.848Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/21/cc19ad5a091f921fea0caae5e4fb926d5b5350c376d09e2ada9f5d58b54c/regex-2025.11.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f3b5a391c7597ffa96b41bd5cbd2ed0305f515fcbb367dfa72735679d5502364", size = 781906, upload-time = "2025-11-03T21:34:02.28Z" },
+    { url = "https://files.pythonhosted.org/packages/98/cc/22c52f2aff7f7e2591fe18db4b6143236c18800752e8e4c2cf938b64fe06/regex-2025.11.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:cc4076a5b4f36d849fd709284b4a3b112326652f3b0466f04002a6c15a0c96c1", size = 774267, upload-time = "2025-11-03T21:34:05.232Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/db/424728b5d831ebd15e86ca55f13990e2bc3c5b417dddaaaa3bd712f12e32/regex-2025.11.3-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:a295ca2bba5c1c885826ce3125fa0b9f702a1be547d821c01d65f199e10c01e2", size = 845211, upload-time = "2025-11-03T21:34:07.57Z" },
+    { url = "https://files.pythonhosted.org/packages/91/fc/48d64dcef7a8f3ecdb9fe60bb3a6c97b317dea089ac72594eccf6cad1938/regex-2025.11.3-cp39-cp39-musllinux_1_2_s390x.whl", hash = "sha256:b4774ff32f18e0504bfc4e59a3e71e18d83bc1e171a3c8ed75013958a03b2f14", size = 835575, upload-time = "2025-11-03T21:34:10.781Z" },
+    { url = "https://files.pythonhosted.org/packages/48/0b/c0cbc34e933ed528bb225d298df572a2e75fbe1db1b36fe719867a5e582d/regex-2025.11.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:22e7d1cdfa88ef33a2ae6aa0d707f9255eb286ffbd90045f1088246833223aee", size = 779426, upload-time = "2025-11-03T21:34:13.498Z" },
+    { url = "https://files.pythonhosted.org/packages/90/54/4987210bc0d139d90b579fbbdaafca2d4acee9c413843d6c1c8ac0dd56bf/regex-2025.11.3-cp39-cp39-win32.whl", hash = "sha256:74d04244852ff73b32eeede4f76f51c5bcf44bc3c207bc3e6cf1c5c45b890708", size = 265836, upload-time = "2025-11-03T21:34:15.556Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/68/0a3a1bf596db98c369d0f10242b006f4a928633e001906889f80c9b4b639/regex-2025.11.3-cp39-cp39-win_amd64.whl", hash = "sha256:7a50cd39f73faa34ec18d6720ee25ef10c4c1839514186fcda658a06c06057a2", size = 277807, upload-time = "2025-11-03T21:34:17.502Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/98/4ed5b608130249810977b030b86abfc7fcd19b31aa75060d8ad11003826d/regex-2025.11.3-cp39-cp39-win_arm64.whl", hash = "sha256:43b4fb020e779ca81c1b5255015fe2b82816c76ec982354534ad9ec09ad7c9e3", size = 270328, upload-time = "2025-11-03T21:34:19.783Z" },
 ]
 
 [[package]]
@@ -4296,111 +4310,25 @@ wheels = [
 
 [[package]]
 name = "rich"
-version = "14.1.0"
+version = "14.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "markdown-it-py", version = "3.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "markdown-it-py", version = "4.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pygments" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/fe/75/af448d8e52bf1d8fa6a9d089ca6c07ff4453d86c65c145d0a300bb073b9b/rich-14.1.0.tar.gz", hash = "sha256:e497a48b844b0320d45007cdebfeaeed8db2a4f4bcf49f15e455cfc4af11eaa8", size = 224441, upload-time = "2025-07-25T07:32:58.125Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/e3/30/3c4d035596d3cf444529e0b2953ad0466f6049528a879d27534700580395/rich-14.1.0-py3-none-any.whl", hash = "sha256:536f5f1785986d6dbdea3c75205c473f970777b4a0d6c6dd1b696aa05a3fa04f", size = 243368, upload-time = "2025-07-25T07:32:56.73Z" },
-]
-
-[[package]]
-name = "rich-toolkit"
-version = "0.15.1"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "rich" },
-    { name = "typing-extensions" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/67/33/1a18839aaa8feef7983590c05c22c9c09d245ada6017d118325bbfcc7651/rich_toolkit-0.15.1.tar.gz", hash = "sha256:6f9630eb29f3843d19d48c3bd5706a086d36d62016687f9d0efa027ddc2dd08a", size = 115322, upload-time = "2025-09-04T09:28:11.789Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/fb/d2/8920e102050a0de7bfabeb4c4614a49248cf8d5d7a8d01885fbb24dc767a/rich-14.2.0.tar.gz", hash = "sha256:73ff50c7c0c1c77c8243079283f4edb376f0f6442433aecb8ce7e6d0b92d1fe4", size = 219990, upload-time = "2025-10-09T14:16:53.064Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/c8/49/42821d55ead7b5a87c8d121edf323cb393d8579f63e933002ade900b784f/rich_toolkit-0.15.1-py3-none-any.whl", hash = "sha256:36a0b1d9a135d26776e4b78f1d5c2655da6e0ef432380b5c6b523c8d8ab97478", size = 29412, upload-time = "2025-09-04T09:28:10.587Z" },
-]
-
-[[package]]
-name = "rignore"
-version = "0.7.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/ae/46/e5ef3423a3746f91d3a3d9a68c499fde983be7dbab7d874efa8d3bb139ba/rignore-0.7.0.tar.gz", hash = "sha256:cfe6a2cbec855b440d7550d53e670246fce43ca5847e46557b6d4577c9cdb540", size = 12796, upload-time = "2025-10-02T13:26:22.194Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/b3/19/04831e4d3db0d828f9cf497b53c944b9c56c26fba764c98747013aae0585/rignore-0.7.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7e011b6412690e34113d5cb133844bfe087fe9a57b37c63cb68671dfbf6080ed", size = 890938, upload-time = "2025-10-02T13:23:16.526Z" },
-    { url = "https://files.pythonhosted.org/packages/0f/3a/da748c8ec25fa15a855fdb6f66c86fc1b1756f5cbe354389d4311d84022e/rignore-0.7.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:c177b8267aa361bf04f9e28fa948881ff01e98e2556bf9d39b088e42de23b190", size = 865825, upload-time = "2025-10-02T13:23:35.145Z" },
-    { url = "https://files.pythonhosted.org/packages/58/8a/8cd9415da272e94a5b306e37f4cc3c0631f08b884836d5c92cf90cecc7a8/rignore-0.7.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:6cfe5873ac415f62d221d8bd04d88f9a70e73fe7aea0c094b9974e530628d8ea", size = 1168074, upload-time = "2025-10-02T13:23:52.542Z" },
-    { url = "https://files.pythonhosted.org/packages/f5/98/008476632a518463875d44dba429a03d59333194ed3a0d08b29c0348f7c3/rignore-0.7.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:20d62988d4f565ab101ec33c501527fde52693eced4ea34d5da61b88db602616", size = 936248, upload-time = "2025-10-02T13:24:08.962Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/ed/6d5c345ba0de67ff4096f0ebcfd2bfdad72335ab9dabe1c665a9579ba687/rignore-0.7.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:427ba9cfc424aedb0b569d659d2f2ea88ed308d8eb245446db10733a73db0fb1", size = 951260, upload-time = "2025-10-02T13:24:38.712Z" },
-    { url = "https://files.pythonhosted.org/packages/44/ca/c56e097b091313b723416de9e28826ac92d731af5c4b51c4892e04286efc/rignore-0.7.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:86280aae7d0980debe5ed6785e3fc9a68ca627ac84e7c84048d7d3fe6d80ef7a", size = 975261, upload-time = "2025-10-02T13:24:25.553Z" },
-    { url = "https://files.pythonhosted.org/packages/e0/dc/2e6987f8f6c8c96c29074901d7d5624de9d1741b4cab6c15975ad159f959/rignore-0.7.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:389c5fe844ad1fd5fba46c0cba0d9e68bfdcaae12e943b135395730efe45bcfe", size = 1071689, upload-time = "2025-10-02T13:25:15.11Z" },
-    { url = "https://files.pythonhosted.org/packages/61/83/fda08b5e11e98f9e96e8c94b7cd5c21ec50193e2861784eb6e21a61f6ccf/rignore-0.7.0-cp310-cp310-musllinux_1_2_armv7l.whl", hash = "sha256:34a34d5e86fda5355b55d927f43ff515c7371e8880b8f16cb92c3278c21327ee", size = 1129324, upload-time = "2025-10-02T13:25:31.669Z" },
-    { url = "https://files.pythonhosted.org/packages/57/74/e5ee481d4f7ccc373a1ebeec2d03415d1cf2b45522ac7424fe773b454f17/rignore-0.7.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:f7aacce87c710a3f254eb8b28a0cec1801362e7cf4f8258cceb8f36fc9cc695a", size = 1108242, upload-time = "2025-10-02T13:25:48.119Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/89/195c5a909303c841ad5e1de300f1d3dd2177768cd3369318654f92b95d8a/rignore-0.7.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b3b88dfeb0d08902fe28eb5d75cbfac1d130ccdaeaca742eab1628bab7960294", size = 1116370, upload-time = "2025-10-02T13:26:06.084Z" },
-    { url = "https://files.pythonhosted.org/packages/31/08/d64298cec32d5df121968b3ab75d17d2a30ff02f080a3457893e57689809/rignore-0.7.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:686c162f945ede315b7b63958d83531b18226cad4fae9170a5787dd8b8b4be89", size = 891607, upload-time = "2025-10-02T13:23:18.739Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/b3/602bb25ba0c862dd3f7f52af0f5e3fce4321207a1b76c0b3b7f17aed0146/rignore-0.7.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:a3c8b62a00c1b6e0ed73412ba8d37d05e214e6a8757f2779d313078d2bdec209", size = 865644, upload-time = "2025-10-02T13:23:36.604Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/fc/18f5ac22714bdd0437aaa59ff2ded2ba3caff2745c89e671bc9c91c52947/rignore-0.7.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:f115666738614cdb0ef122c2b48806043b9b6c603dc03a4708b2eb1df5a44514", size = 1167949, upload-time = "2025-10-02T13:23:54.257Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/1b/6409b434420995b8897c3d6b5a2701343857d2d36d159bd9305287c33634/rignore-0.7.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9ffaf2047304b97bc648592f82c0aeba3468f43546a918994411b8f1d79d42d6", size = 935950, upload-time = "2025-10-02T13:24:10.463Z" },
-    { url = "https://files.pythonhosted.org/packages/b9/56/c0a03cb643ca41091f0377ffea3a35ae3f3cff39b075ca94eec35fae6ed0/rignore-0.7.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:04678c2f1787eb07378754d6aa50e66ce712e0b75e8b843fd9e5e4da35130617", size = 951418, upload-time = "2025-10-02T13:24:40.222Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/3b/33783bc1681662789f71614dee496fb0dd96de4887eb8d5d2cb9f365d1ff/rignore-0.7.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:53a4c4a43558f34b32732efcee9c79c7948ff26673bb764aa0e9bbe951e435fa", size = 975421, upload-time = "2025-10-02T13:24:27.049Z" },
-    { url = "https://files.pythonhosted.org/packages/f3/e2/af19c05288c2afb5b79f73c68e88a34b88245b66e5cf358417461a72c8c5/rignore-0.7.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:794f72ce7195cad1fb41c03b3e3484396c404498b73855004ebea965a697edd9", size = 1071989, upload-time = "2025-10-02T13:25:17.248Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/ea/6ab6d1afafcd3f6e5ba898646bcfe3a6f69eb8f4ac264dd82848ab7f2c5b/rignore-0.7.0-cp311-cp311-musllinux_1_2_armv7l.whl", hash = "sha256:989f35a152bc508c52d63d7d4527215c5dabe7981e5744bcf35f96c99f3758f7", size = 1129150, upload-time = "2025-10-02T13:25:33.352Z" },
-    { url = "https://files.pythonhosted.org/packages/0d/49/a327d54cbd5f9f34ed383057ee1c9a044571878045cbd37a129f27f13ab0/rignore-0.7.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:b945b29a995fdcf669dc098ec40237131742de2cf49484011ba3f81d0fff23a3", size = 1107917, upload-time = "2025-10-02T13:25:49.702Z" },
-    { url = "https://files.pythonhosted.org/packages/86/f8/89a1269911e7895e3c4a5c1fb1abb3b9b255362035fa54c593287cf38b15/rignore-0.7.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:e4deda4c3e5cec1ebfb714094cd9af79e8840680187537d13a216377d6aa2ed6", size = 1116013, upload-time = "2025-10-02T13:26:07.597Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/e6/fd2cbc71f725ea10892c85ea56bd8f54426557cf5ac2924f9c27b771ee45/rignore-0.7.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0525ccf3e8b9ccd6f1dfc87ecc78218a83605070b247633636d144acdf6b73be", size = 892031, upload-time = "2025-10-02T13:23:20.558Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/c8/0dfd755f57515d34ca26de011e016f62db86f7bef0586f2ab0d9f6e18136/rignore-0.7.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:570bcf51fd9f78ec79ec33f2f852e6665027fae80cc3e5e2523c97d3f4220369", size = 865496, upload-time = "2025-10-02T13:23:37.965Z" },
-    { url = "https://files.pythonhosted.org/packages/a6/b9/f73af8509842d74788fc26feca25db1eade9291fae79540872c130407340/rignore-0.7.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:32f5d3d90a520d61e43c2a23724852c689c3ed36b38264c77b613f967e2d1f68", size = 1165555, upload-time = "2025-10-02T13:23:56.009Z" },
-    { url = "https://files.pythonhosted.org/packages/44/22/67d2fb589cedd7bf3a01e16617f2da10f172165b3ecdaa8fa0707043e9ed/rignore-0.7.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a7d189cfb9059dfa497e5480c411bd2aba838124b50b93abf7e92556221b7956", size = 936631, upload-time = "2025-10-02T13:24:11.97Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/6b/e0f969a1cb3ff2caa0dd342e512d7a0a6f1b737b6f5373c04606aa946e80/rignore-0.7.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6c871a31596476ac4343f6b803ee8ddca068425e1837cf6849ebe46c498c73c5", size = 951058, upload-time = "2025-10-02T13:24:41.742Z" },
-    { url = "https://files.pythonhosted.org/packages/45/cf/ccf053fb87601332e8b2e2da707f2801bee66ee5fe843687183f45c2e768/rignore-0.7.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:5b7d8ce1efbd8fa865712d34753ce4eb8e0732874df95351244e14308fb87d0a", size = 974638, upload-time = "2025-10-02T13:24:29Z" },
-    { url = "https://files.pythonhosted.org/packages/de/ae/a00181c0d2dc437a3729dbebcfffd67bb849d1c53e45850c7b4428f5fba4/rignore-0.7.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:d261aea1a51ef93c262b52ad195a1092a8bae17577e8192473d1b5fd30379346", size = 1072970, upload-time = "2025-10-02T13:25:18.888Z" },
-    { url = "https://files.pythonhosted.org/packages/81/30/3011207fc9f26f9eb21d2282dfedd8f2d66cf7a9a3053370c9b4b87601e1/rignore-0.7.0-cp312-cp312-musllinux_1_2_armv7l.whl", hash = "sha256:034bef935e3734b4ad2dada59c96717f3e3d0b48551a0c79379c4d3280b4a397", size = 1128833, upload-time = "2025-10-02T13:25:34.987Z" },
-    { url = "https://files.pythonhosted.org/packages/4b/be/4c6a860f851db6cb0b96a3ec62dd4fe95290ee36e67b845ffab58908c6cc/rignore-0.7.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:5f816b65c9bf97093d792c9b50369d5a81a5f95b4ed5f003d4091bd1db3b70d8", size = 1106909, upload-time = "2025-10-02T13:25:51.266Z" },
-    { url = "https://files.pythonhosted.org/packages/9d/8a/691d79e72f000968e1e3457ff53634760dac24fa6c6b5663d994362b8a99/rignore-0.7.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b88479f0a89828781d25a9acd485be88abf4f1f1c14e455b6530da265adb593c", size = 1115733, upload-time = "2025-10-02T13:26:09.256Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/9d/288216c28dd35e80a56cd6f68952a1bb0915d05b429ad8003b735a3af31a/rignore-0.7.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8e39eccbb92f1b562b72aefeec93bfae38c8f302a58aaf9b2d77362b57e1cf27", size = 892385, upload-time = "2025-10-02T13:23:27.369Z" },
-    { url = "https://files.pythonhosted.org/packages/61/19/655cd781fe7ede2c61280905f9694a4094db42b106d2f92a07197b7a3716/rignore-0.7.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:bc34644c1d42ea0ec7b7b14a56203dbf19df6c258d75937191f871da3cd1c1ca", size = 866836, upload-time = "2025-10-02T13:23:45.714Z" },
-    { url = "https://files.pythonhosted.org/packages/72/4c/b20543b6ca5fb192a04504995b140fa0b7a3f143e9a5960710d093cc3bae/rignore-0.7.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7cd512a48d9b0900b297fc0c2227e81096eed21e42dae5a9e82e332500fb8730", size = 1169748, upload-time = "2025-10-02T13:24:02.446Z" },
-    { url = "https://files.pythonhosted.org/packages/97/60/ef01964969ec70995c0b30cb2e90dd16dfa9792259cf5060a5cdba437f25/rignore-0.7.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:baee13481bf5d335aa48cb3607637f842edaf418f5dde252fe9d99f360533549", size = 937629, upload-time = "2025-10-02T13:24:18.393Z" },
-    { url = "https://files.pythonhosted.org/packages/16/cb/73fc5d9f99cf76120ec9b97b61bfc49967f37695c0ed3f61fca16a4eb1c9/rignore-0.7.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f7ae8d7ec27ce07599aef2e83e7cff831888e381db0f0ee066d9fb4873ba14ab", size = 952303, upload-time = "2025-10-02T13:24:48.683Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/8e/c7c48fbad072a9d7ca8c2af5058540db85c57d82257323df6a088c0f3205/rignore-0.7.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:7a68d1a180d5af3e1ca2e3e1f2b318e6925677d06f5f4d6b3cbd2f63eaad9255", size = 977032, upload-time = "2025-10-02T13:24:35.361Z" },
-    { url = "https://files.pythonhosted.org/packages/29/ca/42293e524c772ac1ab929496f395a5105e6c54687656320ea478d7c04909/rignore-0.7.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:0714ee0759fe57068ea8dc234672f939b5d07dbc89d4e0b59468fdb88dbca830", size = 1072977, upload-time = "2025-10-02T13:25:25.24Z" },
-    { url = "https://files.pythonhosted.org/packages/0a/c4/94ec3ee3e86051c5c09a196298f6d2fad9208bcd7bb0f228ef35118a92aa/rignore-0.7.0-cp39-cp39-musllinux_1_2_armv7l.whl", hash = "sha256:44cda8590511c9a9aa32e253267f1742b41260f3b71d9bcabdb07911cca41a9a", size = 1130563, upload-time = "2025-10-02T13:25:41.694Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/a7/f975a509194b1daf02326d30a417b13939ef1faebbad260d5858de5aed2d/rignore-0.7.0-cp39-cp39-musllinux_1_2_i686.whl", hash = "sha256:2f680188a2cfae1e64a863223e6a702c212d28b77fa3b9cc0547a4957b5845b0", size = 1109816, upload-time = "2025-10-02T13:25:58.105Z" },
-    { url = "https://files.pythonhosted.org/packages/59/db/99b695509ad1f9d067b0382b8f595ae547d163b2aea90b8948fa2ee45347/rignore-0.7.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:236003f4d348e4e2992aa7936d38aeffbaf17e43e3b068623096ee4b8f7efa2b", size = 1117778, upload-time = "2025-10-02T13:26:15.608Z" },
-    { url = "https://files.pythonhosted.org/packages/eb/28/59aa850097283f3ae651e13ced4a8beadab8bab2f193a6e6d32d4235fc79/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:11831ac0c3bc656667848abd0cb869d288b13ad69a976cac307b447a4f79c9d3", size = 893706, upload-time = "2025-10-02T13:23:29.65Z" },
-    { url = "https://files.pythonhosted.org/packages/0b/5e/0bf7c2101cd557374805372ae8a230ba83b4aa460c2f2f327580f79774f9/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:1f18bdc1bfd73da6a43bcf2c08f94e1ecddabf234e47e0f95daf6107cf937fb3", size = 867651, upload-time = "2025-10-02T13:23:47.193Z" },
-    { url = "https://files.pythonhosted.org/packages/c0/ac/33080dba026b863a43e43a4c861278e51eb1b03c90f1a108f35a74b85d2d/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e6268a12ecb6d25caf0adb166732940bd9113e7dddd46018e457f1fb9408a707", size = 1168395, upload-time = "2025-10-02T13:24:03.882Z" },
-    { url = "https://files.pythonhosted.org/packages/83/b1/0c62eb8df324c00ef65c02c24dee30cc5a491ba224728916703761ee7c80/rignore-0.7.0-pp310-pypy310_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:7c1402445b24c8904b6cef124f2798d55a2ba84b41397d7fdab6fe316b1f20e6", size = 938744, upload-time = "2025-10-02T13:24:20.188Z" },
-    { url = "https://files.pythonhosted.org/packages/5d/c3/965ab1d3674e790429a63d8486e2432fe120b29d4f4682a4171b3b3efac2/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:8aa1c3215cc9587e55b3326357637e17a2ed713ddb2c59339f908aae98ae01d6", size = 1074476, upload-time = "2025-10-02T13:25:26.864Z" },
-    { url = "https://files.pythonhosted.org/packages/36/2d/2673af40f46c2182c34d06e0662c035130008b47f74e4b5c72cddbbb50b6/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_armv7l.whl", hash = "sha256:78c8f56aaae18406699026e26f9c3b4721adc95f08ac4e76972aed0efb5eb91d", size = 1131270, upload-time = "2025-10-02T13:25:43.319Z" },
-    { url = "https://files.pythonhosted.org/packages/98/6c/7fcd680db36c2e34c0d5cceefd7a423772a9fbf25bb468b5748fedacda94/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_i686.whl", hash = "sha256:89ab6d73ffd48be27032def1decef83faebee891519e7f2006df5657b8ba2f4a", size = 1110257, upload-time = "2025-10-02T13:26:00.717Z" },
-    { url = "https://files.pythonhosted.org/packages/22/95/8ce27268d27fd12bc1b80d3e1840402f3ef3c205e788975d61c7a4077ef8/rignore-0.7.0-pp310-pypy310_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:cf4eebeebeabd27467b51d5dbc92adc7177fcfd73a29c86f649853ba476d98fa", size = 1118831, upload-time = "2025-10-02T13:26:17.187Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/60/b02edbf5059f7947e375dc46583283aad579505e9e07775277e7fd6e04db/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f40142a34a7f08cd90fb4e74e43deffe3381fa3b164fb59857fa4e3996d4716d", size = 892600, upload-time = "2025-10-02T13:23:31.158Z" },
-    { url = "https://files.pythonhosted.org/packages/cf/c5/3caa7732a91623110bc80c30f592efc6571a1c610b94f36083601ebf2392/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ccbc0b6285bb981316e5646ac96be7bca9665ee2444427d8d170fda5eda6f022", size = 866500, upload-time = "2025-10-02T13:23:49.099Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/66/943300886972b2dded2e0e851c1da1ad36565d40b5e55833b049cbf9285b/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:77cdf15a8b0ab80cd1d05a754b3237330e60e8731c255b7eb2a5d240a68df9f8", size = 1167255, upload-time = "2025-10-02T13:24:05.583Z" },
-    { url = "https://files.pythonhosted.org/packages/1e/26/2f8cb5a546ce7056fe0fb8afbfc887431f9ba986cd7b4c65821dac13afa8/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:14e7e5ac99d60dd1993032205de7e79c36687825c45a7caa704620a0e9fde03f", size = 937991, upload-time = "2025-10-02T13:24:21.694Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/29/f97d581fc4d1013a42fe51154f820a7ccb97c679a2c2ea0c73072aa8935e/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:98fae67456f053942ccda2cb2677a55fd34397e6674eaa403ab7c1c4930dcb12", size = 951972, upload-time = "2025-10-02T13:24:50.199Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/06/18da8ea8fc217fce872f81de23217c7ae011dd6e396dff026a262b499a4b/rignore-0.7.0-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:2b55d2dcee6808f677ef25219ec0bb4852fbf2edb0b5010a5f18fe5feee276d6", size = 976002, upload-time = "2025-10-02T13:24:36.851Z" },
-    { url = "https://files.pythonhosted.org/packages/ea/11/2f998fccb85a31f8dbd94b31123b48645067d4ca55b49c033987286475e7/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:7ff87634a648f17a9992ac4ce2fb48397696e3ab4a80154a895b9d1f6fc606cf", size = 1073180, upload-time = "2025-10-02T13:25:28.424Z" },
-    { url = "https://files.pythonhosted.org/packages/01/bf/ee6927f8dd8644f4c9c44d364380ab49629d259cc9611224512b161d7bef/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_armv7l.whl", hash = "sha256:c5721daa569fae74f5bf060165f96c6fec0a963ed008213e778259945406ec53", size = 1130056, upload-time = "2025-10-02T13:25:45.019Z" },
-    { url = "https://files.pythonhosted.org/packages/33/89/b231f432caced14303055c8611b34c5e2910c48b882de1c79eff4ce177d0/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_i686.whl", hash = "sha256:5770e783e08403b02c052b8b74a3e9431142aca93c78ccd1cc389b4dc60c2846", size = 1108603, upload-time = "2025-10-02T13:26:02.539Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/33/d331a0aea9e4a00ff530ad18421c46e213da1a608ad05463a2e5ae6cc572/rignore-0.7.0-pp311-pypy311_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:504f66805fcc2a684cd1cda460d9f15b8b08997f06d9281efa221007072c53f5", size = 1117330, upload-time = "2025-10-02T13:26:18.741Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/cd/c5b39846a51d9dbc90ec6a0ac3502c9c827b4fcf2ce70b80ac9ca346cbb1/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:280fa78100c94640f36fe40304b5fdda1f14487f306f20b884f4dc230ad86e7b", size = 893459, upload-time = "2025-10-02T13:23:33.318Z" },
-    { url = "https://files.pythonhosted.org/packages/65/00/226cd6516bbf1b8a74fec65d2c1f19bc38866e41099d970e35f13984812f/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:3f9ba817b3ef8560c34d18764f32b7165027a8b82503a1f57ccfdc821e689d30", size = 867819, upload-time = "2025-10-02T13:23:50.609Z" },
-    { url = "https://files.pythonhosted.org/packages/28/18/07e6ff331a227e67bd2ccb9bfd823cb40434db26e96c406935a31af6f484/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:c2150dbfb0227762a0af2decd7cc67a1c8b5cb4d35f8b83c674143fefc9b76f3", size = 1170040, upload-time = "2025-10-02T13:24:07.079Z" },
-    { url = "https://files.pythonhosted.org/packages/fb/83/bcf95d4fcb8653c8556b4d50873830253ac46c34f005b135819979a41884/rignore-0.7.0-pp39-pypy39_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:4e9af74b3b90266d7700b44a08217af636421b6e823f9280f9521a06e4fc18b9", size = 938346, upload-time = "2025-10-02T13:24:24.098Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/4b/0ba560f01a4cc22ecf5058311132b91e76dfef91f95a99072aae7948abf0/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:165890c27b1fd8ec71d47a05e6a11c6cf42693f3b479f3fc8d0047b5c2a46191", size = 1074376, upload-time = "2025-10-02T13:25:30.023Z" },
-    { url = "https://files.pythonhosted.org/packages/a6/6f/3003ef871fac5f6ee7599695078c9d9165fee8171f9bf794cd5cddd4377f/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_armv7l.whl", hash = "sha256:3830a252b7e82ba61595489a819af721f645dd42548d5801000583044745a41c", size = 1131436, upload-time = "2025-10-02T13:25:46.557Z" },
-    { url = "https://files.pythonhosted.org/packages/02/56/a63bf6c3499a08713d491dba0543b47176f389c873e03d70b72bf0252216/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_i686.whl", hash = "sha256:d5a50ab60948bd7339ab50329991acfe7afb5fec4a444f0e19939ee6e18d364a", size = 1110413, upload-time = "2025-10-02T13:26:04.518Z" },
-    { url = "https://files.pythonhosted.org/packages/c9/a9/678ac02cb63517335ce442507438c06d2e21940ae8baca3e4457de6426d1/rignore-0.7.0-pp39-pypy39_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:648c052ba530ff43ecb516dd33831563ecac9878cd54897171237198870c48d4", size = 1118680, upload-time = "2025-10-02T13:26:20.907Z" },
+    { url = "https://files.pythonhosted.org/packages/25/7a/b0178788f8dc6cafce37a212c99565fa1fe7872c70c6c9c1e1a372d9d88f/rich-14.2.0-py3-none-any.whl", hash = "sha256:76bc51fe2e57d2b1be1f96c524b890b816e334ab4c1e45888799bfaab0021edd", size = 243393, upload-time = "2025-10-09T14:16:51.245Z" },
 ]
 
 [[package]]
 name = "rpds-py"
 version = "0.27.1"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/e9/dd/2c0cbe774744272b0ae725f44032c77bdcab6e8bcf544bffa3b6e70c8dba/rpds_py-0.27.1.tar.gz", hash = "sha256:26a1c73171d10b7acccbded82bf6a586ab8203601e565badc74bbbf8bc5a10f8", size = 27479, upload-time = "2025-08-27T12:16:36.024Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/a5/ed/3aef893e2dd30e77e35d20d4ddb45ca459db59cead748cad9796ad479411/rpds_py-0.27.1-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:68afeec26d42ab3b47e541b272166a0b4400313946871cba3ed3a4fc0cab1cef", size = 371606, upload-time = "2025-08-27T12:12:25.189Z" },
@@ -4501,6 +4429,75 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/60/94/384cf54c430b9dac742bbd2ec26c23feb78ded0d43d6d78563a281aec017/rpds_py-0.27.1-pp39-pypy39_pp73-win_amd64.whl", hash = "sha256:4fc9b7fe29478824361ead6e14e4f5aed570d477e06088826537e202d25fe859", size = 228784, upload-time = "2025-08-27T12:16:34.428Z" },
 ]
 
+[[package]]
+name = "rpds-py"
+version = "0.29.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/98/33/23b3b3419b6a3e0f559c7c0d2ca8fc1b9448382b25245033788785921332/rpds_py-0.29.0.tar.gz", hash = "sha256:fe55fe686908f50154d1dc599232016e50c243b438c3b7432f24e2895b0e5359", size = 69359, upload-time = "2025-11-16T14:50:39.532Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9e/7a/c5b2ff381b74bc742768e8d870f26babac4ef256ba160bdbf8d57af56461/rpds_py-0.29.0-cp310-cp310-macosx_10_12_x86_64.whl", hash = "sha256:4ae4b88c6617e1b9e5038ab3fccd7bac0842fdda2b703117b2aa99bc85379113", size = 372385, upload-time = "2025-11-16T14:47:36.287Z" },
+    { url = "https://files.pythonhosted.org/packages/28/36/531f1eb4d5bed4a9c150f363a7ec4a98d2dc746151bba5473bc38ee85dec/rpds_py-0.29.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:7d9128ec9d8cecda6f044001fde4fb71ea7c24325336612ef8179091eb9596b9", size = 362869, upload-time = "2025-11-16T14:47:38.196Z" },
+    { url = "https://files.pythonhosted.org/packages/54/df/7e9c0493a2015d9c82807a2d5f023ea9774e27a4c15b33ef1cdb7456138d/rpds_py-0.29.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d37812c3da8e06f2bb35b3cf10e4a7b68e776a706c13058997238762b4e07f4f", size = 391582, upload-time = "2025-11-16T14:47:39.746Z" },
+    { url = "https://files.pythonhosted.org/packages/15/38/42a981c3592ef46fbd7e17adbf8730cc5ec87e6aa1770c658c44bbb52960/rpds_py-0.29.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:66786c3fb1d8de416a7fa8e1cb1ec6ba0a745b2b0eee42f9b7daa26f1a495545", size = 405685, upload-time = "2025-11-16T14:47:41.472Z" },
+    { url = "https://files.pythonhosted.org/packages/12/45/628b8c15856c3849c3f52ec6dac93c046ed5faeed4a435af03b70525fd29/rpds_py-0.29.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b58f5c77f1af888b5fd1876c9a0d9858f6f88a39c9dd7c073a88e57e577da66d", size = 527067, upload-time = "2025-11-16T14:47:43.036Z" },
+    { url = "https://files.pythonhosted.org/packages/dc/ba/6b56d09badeabd95098016d72a437d4a0fd82d4672ce92a7607df5d70a42/rpds_py-0.29.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:799156ef1f3529ed82c36eb012b5d7a4cf4b6ef556dd7cc192148991d07206ae", size = 412532, upload-time = "2025-11-16T14:47:44.484Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/39/2f1f3db92888314b50b8f9641f679188bd24b3665a8cb9923b7201ae8011/rpds_py-0.29.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:453783477aa4f2d9104c4b59b08c871431647cb7af51b549bbf2d9eb9c827756", size = 392736, upload-time = "2025-11-16T14:47:46.053Z" },
+    { url = "https://files.pythonhosted.org/packages/60/43/3c3b1dcd827e50f2ae28786d846b8a351080d8a69a3b49bc10ae44cc39b1/rpds_py-0.29.0-cp310-cp310-manylinux_2_31_riscv64.whl", hash = "sha256:24a7231493e3c4a4b30138b50cca089a598e52c34cf60b2f35cebf62f274fdea", size = 406300, upload-time = "2025-11-16T14:47:47.268Z" },
+    { url = "https://files.pythonhosted.org/packages/da/02/bc96021b67f8525e6bcdd68935c4543ada61e1f3dcb067ed037d68b8c6d2/rpds_py-0.29.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:7033c1010b1f57bb44d8067e8c25aa6fa2e944dbf46ccc8c92b25043839c3fd2", size = 423641, upload-time = "2025-11-16T14:47:48.878Z" },
+    { url = "https://files.pythonhosted.org/packages/38/e9/c435ddb602ced19a80b8277a41371734f33ad3f91cc4ceb4d82596800a3c/rpds_py-0.29.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:0248b19405422573621172ab8e3a1f29141362d13d9f72bafa2e28ea0cdca5a2", size = 574153, upload-time = "2025-11-16T14:47:50.435Z" },
+    { url = "https://files.pythonhosted.org/packages/84/82/dc3c32e1f89ecba8a59600d4cd65fe0ad81b6c636ccdbf6cd177fd6a7bac/rpds_py-0.29.0-cp310-cp310-musllinux_1_2_i686.whl", hash = "sha256:f9f436aee28d13b9ad2c764fc273e0457e37c2e61529a07b928346b219fcde3b", size = 600304, upload-time = "2025-11-16T14:47:51.599Z" },
+    { url = "https://files.pythonhosted.org/packages/35/98/785290e0b7142470735dc1b1f68fb33aae29e5296f062c88396eedf796c8/rpds_py-0.29.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:24a16cb7163933906c62c272de20ea3c228e4542c8c45c1d7dc2b9913e17369a", size = 562211, upload-time = "2025-11-16T14:47:53.094Z" },
+    { url = "https://files.pythonhosted.org/packages/30/58/4eeddcb0737c6875f3e30c65dc9d7e7a10dfd5779646a990fa602c6d56c5/rpds_py-0.29.0-cp310-cp310-win32.whl", hash = "sha256:1a409b0310a566bfd1be82119891fefbdce615ccc8aa558aff7835c27988cbef", size = 221803, upload-time = "2025-11-16T14:47:54.404Z" },
+    { url = "https://files.pythonhosted.org/packages/54/77/b35a8dbdcbeb32505500547cdafaa9f8863e85f8faac50ef34464ec5a256/rpds_py-0.29.0-cp310-cp310-win_amd64.whl", hash = "sha256:c5523b0009e7c3c1263471b69d8da1c7d41b3ecb4cb62ef72be206b92040a950", size = 235530, upload-time = "2025-11-16T14:47:56.061Z" },
+    { url = "https://files.pythonhosted.org/packages/36/ab/7fb95163a53ab122c74a7c42d2d2f012819af2cf3deb43fb0d5acf45cc1a/rpds_py-0.29.0-cp311-cp311-macosx_10_12_x86_64.whl", hash = "sha256:9b9c764a11fd637e0322a488560533112837f5334ffeb48b1be20f6d98a7b437", size = 372344, upload-time = "2025-11-16T14:47:57.279Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/45/f3c30084c03b0d0f918cb4c5ae2c20b0a148b51ba2b3f6456765b629bedd/rpds_py-0.29.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:3fd2164d73812026ce970d44c3ebd51e019d2a26a4425a5dcbdfa93a34abc383", size = 363041, upload-time = "2025-11-16T14:47:58.908Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/e9/4d044a1662608c47a87cbb37b999d4d5af54c6d6ebdda93a4d8bbf8b2a10/rpds_py-0.29.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4a097b7f7f7274164566ae90a221fd725363c0e9d243e2e9ed43d195ccc5495c", size = 391775, upload-time = "2025-11-16T14:48:00.197Z" },
+    { url = "https://files.pythonhosted.org/packages/50/c9/7616d3ace4e6731aeb6e3cd85123e03aec58e439044e214b9c5c60fd8eb1/rpds_py-0.29.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7cdc0490374e31cedefefaa1520d5fe38e82fde8748cbc926e7284574c714d6b", size = 405624, upload-time = "2025-11-16T14:48:01.496Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/e2/6d7d6941ca0843609fd2d72c966a438d6f22617baf22d46c3d2156c31350/rpds_py-0.29.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:89ca2e673ddd5bde9b386da9a0aac0cab0e76f40c8f0aaf0d6311b6bbf2aa311", size = 527894, upload-time = "2025-11-16T14:48:03.167Z" },
+    { url = "https://files.pythonhosted.org/packages/8d/f7/aee14dc2db61bb2ae1e3068f134ca9da5f28c586120889a70ff504bb026f/rpds_py-0.29.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a5d9da3ff5af1ca1249b1adb8ef0573b94c76e6ae880ba1852f033bf429d4588", size = 412720, upload-time = "2025-11-16T14:48:04.413Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/e2/2293f236e887c0360c2723d90c00d48dee296406994d6271faf1712e94ec/rpds_py-0.29.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8238d1d310283e87376c12f658b61e1ee23a14c0e54c7c0ce953efdbdc72deed", size = 392945, upload-time = "2025-11-16T14:48:06.252Z" },
+    { url = "https://files.pythonhosted.org/packages/14/cd/ceea6147acd3bd1fd028d1975228f08ff19d62098078d5ec3eed49703797/rpds_py-0.29.0-cp311-cp311-manylinux_2_31_riscv64.whl", hash = "sha256:2d6fb2ad1c36f91c4646989811e84b1ea5e0c3cf9690b826b6e32b7965853a63", size = 406385, upload-time = "2025-11-16T14:48:07.575Z" },
+    { url = "https://files.pythonhosted.org/packages/52/36/fe4dead19e45eb77a0524acfdbf51e6cda597b26fc5b6dddbff55fbbb1a5/rpds_py-0.29.0-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:534dc9df211387547267ccdb42253aa30527482acb38dd9b21c5c115d66a96d2", size = 423943, upload-time = "2025-11-16T14:48:10.175Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/7b/4551510803b582fa4abbc8645441a2d15aa0c962c3b21ebb380b7e74f6a1/rpds_py-0.29.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d456e64724a075441e4ed648d7f154dc62e9aabff29bcdf723d0c00e9e1d352f", size = 574204, upload-time = "2025-11-16T14:48:11.499Z" },
+    { url = "https://files.pythonhosted.org/packages/64/ba/071ccdd7b171e727a6ae079f02c26f75790b41555f12ca8f1151336d2124/rpds_py-0.29.0-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:a738f2da2f565989401bd6fd0b15990a4d1523c6d7fe83f300b7e7d17212feca", size = 600587, upload-time = "2025-11-16T14:48:12.822Z" },
+    { url = "https://files.pythonhosted.org/packages/03/09/96983d48c8cf5a1e03c7d9cc1f4b48266adfb858ae48c7c2ce978dbba349/rpds_py-0.29.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:a110e14508fd26fd2e472bb541f37c209409876ba601cf57e739e87d8a53cf95", size = 562287, upload-time = "2025-11-16T14:48:14.108Z" },
+    { url = "https://files.pythonhosted.org/packages/40/f0/8c01aaedc0fa92156f0391f39ea93b5952bc0ec56b897763858f95da8168/rpds_py-0.29.0-cp311-cp311-win32.whl", hash = "sha256:923248a56dd8d158389a28934f6f69ebf89f218ef96a6b216a9be6861804d3f4", size = 221394, upload-time = "2025-11-16T14:48:15.374Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/a5/a8b21c54c7d234efdc83dc034a4d7cd9668e3613b6316876a29b49dece71/rpds_py-0.29.0-cp311-cp311-win_amd64.whl", hash = "sha256:539eb77eb043afcc45314d1be09ea6d6cafb3addc73e0547c171c6d636957f60", size = 235713, upload-time = "2025-11-16T14:48:16.636Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/1f/df3c56219523947b1be402fa12e6323fe6d61d883cf35d6cb5d5bb6db9d9/rpds_py-0.29.0-cp311-cp311-win_arm64.whl", hash = "sha256:bdb67151ea81fcf02d8f494703fb728d4d34d24556cbff5f417d74f6f5792e7c", size = 229157, upload-time = "2025-11-16T14:48:17.891Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/50/bc0e6e736d94e420df79be4deb5c9476b63165c87bb8f19ef75d100d21b3/rpds_py-0.29.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:a0891cfd8db43e085c0ab93ab7e9b0c8fee84780d436d3b266b113e51e79f954", size = 376000, upload-time = "2025-11-16T14:48:19.141Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/3a/46676277160f014ae95f24de53bed0e3b7ea66c235e7de0b9df7bd5d68ba/rpds_py-0.29.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:3897924d3f9a0361472d884051f9a2460358f9a45b1d85a39a158d2f8f1ad71c", size = 360575, upload-time = "2025-11-16T14:48:20.443Z" },
+    { url = "https://files.pythonhosted.org/packages/75/ba/411d414ed99ea1afdd185bbabeeaac00624bd1e4b22840b5e9967ade6337/rpds_py-0.29.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2a21deb8e0d1571508c6491ce5ea5e25669b1dd4adf1c9d64b6314842f708b5d", size = 392159, upload-time = "2025-11-16T14:48:22.12Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/b1/e18aa3a331f705467a48d0296778dc1fea9d7f6cf675bd261f9a846c7e90/rpds_py-0.29.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:9efe71687d6427737a0a2de9ca1c0a216510e6cd08925c44162be23ed7bed2d5", size = 410602, upload-time = "2025-11-16T14:48:23.563Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/6c/04f27f0c9f2299274c76612ac9d2c36c5048bb2c6c2e52c38c60bf3868d9/rpds_py-0.29.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:40f65470919dc189c833e86b2c4bd21bd355f98436a2cef9e0a9a92aebc8e57e", size = 515808, upload-time = "2025-11-16T14:48:24.949Z" },
+    { url = "https://files.pythonhosted.org/packages/83/56/a8412aa464fb151f8bc0d91fb0bb888adc9039bd41c1c6ba8d94990d8cf8/rpds_py-0.29.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:def48ff59f181130f1a2cb7c517d16328efac3ec03951cca40c1dc2049747e83", size = 416015, upload-time = "2025-11-16T14:48:26.782Z" },
+    { url = "https://files.pythonhosted.org/packages/04/4c/f9b8a05faca3d9e0a6397c90d13acb9307c9792b2bff621430c58b1d6e76/rpds_py-0.29.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:ad7bd570be92695d89285a4b373006930715b78d96449f686af422debb4d3949", size = 395325, upload-time = "2025-11-16T14:48:28.055Z" },
+    { url = "https://files.pythonhosted.org/packages/34/60/869f3bfbf8ed7b54f1ad9a5543e0fdffdd40b5a8f587fe300ee7b4f19340/rpds_py-0.29.0-cp312-cp312-manylinux_2_31_riscv64.whl", hash = "sha256:5a572911cd053137bbff8e3a52d31c5d2dba51d3a67ad902629c70185f3f2181", size = 410160, upload-time = "2025-11-16T14:48:29.338Z" },
+    { url = "https://files.pythonhosted.org/packages/91/aa/e5b496334e3aba4fe4c8a80187b89f3c1294c5c36f2a926da74338fa5a73/rpds_py-0.29.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:d583d4403bcbf10cffc3ab5cee23d7643fcc960dff85973fd3c2d6c86e8dbb0c", size = 425309, upload-time = "2025-11-16T14:48:30.691Z" },
+    { url = "https://files.pythonhosted.org/packages/85/68/4e24a34189751ceb6d66b28f18159922828dd84155876551f7ca5b25f14f/rpds_py-0.29.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:070befbb868f257d24c3bb350dbd6e2f645e83731f31264b19d7231dd5c396c7", size = 574644, upload-time = "2025-11-16T14:48:31.964Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/cf/474a005ea4ea9c3b4f17b6108b6b13cebfc98ebaff11d6e1b193204b3a93/rpds_py-0.29.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:fc935f6b20b0c9f919a8ff024739174522abd331978f750a74bb68abd117bd19", size = 601605, upload-time = "2025-11-16T14:48:33.252Z" },
+    { url = "https://files.pythonhosted.org/packages/f4/b1/c56f6a9ab8c5f6bb5c65c4b5f8229167a3a525245b0773f2c0896686b64e/rpds_py-0.29.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:8c5a8ecaa44ce2d8d9d20a68a2483a74c07f05d72e94a4dff88906c8807e77b0", size = 564593, upload-time = "2025-11-16T14:48:34.643Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/13/0494cecce4848f68501e0a229432620b4b57022388b071eeff95f3e1e75b/rpds_py-0.29.0-cp312-cp312-win32.whl", hash = "sha256:ba5e1aeaf8dd6d8f6caba1f5539cddda87d511331714b7b5fc908b6cfc3636b7", size = 223853, upload-time = "2025-11-16T14:48:36.419Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/6a/51e9aeb444a00cdc520b032a28b07e5f8dc7bc328b57760c53e7f96997b4/rpds_py-0.29.0-cp312-cp312-win_amd64.whl", hash = "sha256:b5f6134faf54b3cb83375db0f113506f8b7770785be1f95a631e7e2892101977", size = 239895, upload-time = "2025-11-16T14:48:37.956Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/d4/8bce56cdad1ab873e3f27cb31c6a51d8f384d66b022b820525b879f8bed1/rpds_py-0.29.0-cp312-cp312-win_arm64.whl", hash = "sha256:b016eddf00dca7944721bf0cd85b6af7f6c4efaf83ee0b37c4133bd39757a8c7", size = 230321, upload-time = "2025-11-16T14:48:39.71Z" },
+    { url = "https://files.pythonhosted.org/packages/f2/ac/b97e80bf107159e5b9ba9c91df1ab95f69e5e41b435f27bdd737f0d583ac/rpds_py-0.29.0-pp311-pypy311_pp73-macosx_10_12_x86_64.whl", hash = "sha256:acd82a9e39082dc5f4492d15a6b6c8599aa21db5c35aaf7d6889aea16502c07d", size = 373963, upload-time = "2025-11-16T14:50:16.205Z" },
+    { url = "https://files.pythonhosted.org/packages/40/5a/55e72962d5d29bd912f40c594e68880d3c7a52774b0f75542775f9250712/rpds_py-0.29.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:715b67eac317bf1c7657508170a3e011a1ea6ccb1c9d5f296e20ba14196be6b3", size = 364644, upload-time = "2025-11-16T14:50:18.22Z" },
+    { url = "https://files.pythonhosted.org/packages/99/2a/6b6524d0191b7fc1351c3c0840baac42250515afb48ae40c7ed15499a6a2/rpds_py-0.29.0-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f3b1b87a237cb2dba4db18bcfaaa44ba4cd5936b91121b62292ff21df577fc43", size = 393847, upload-time = "2025-11-16T14:50:20.012Z" },
+    { url = "https://files.pythonhosted.org/packages/1c/b8/c5692a7df577b3c0c7faed7ac01ee3c608b81750fc5d89f84529229b6873/rpds_py-0.29.0-pp311-pypy311_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:1c3c3e8101bb06e337c88eb0c0ede3187131f19d97d43ea0e1c5407ea74c0cbf", size = 407281, upload-time = "2025-11-16T14:50:21.64Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/57/0546c6f84031b7ea08b76646a8e33e45607cc6bd879ff1917dc077bb881e/rpds_py-0.29.0-pp311-pypy311_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:2b8e54d6e61f3ecd3abe032065ce83ea63417a24f437e4a3d73d2f85ce7b7cfe", size = 529213, upload-time = "2025-11-16T14:50:23.219Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/c1/01dd5f444233605555bc11fe5fed6a5c18f379f02013870c176c8e630a23/rpds_py-0.29.0-pp311-pypy311_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:3fbd4e9aebf110473a420dea85a238b254cf8a15acb04b22a5a6b5ce8925b760", size = 413808, upload-time = "2025-11-16T14:50:25.262Z" },
+    { url = "https://files.pythonhosted.org/packages/aa/0a/60f98b06156ea2a7af849fb148e00fbcfdb540909a5174a5ed10c93745c7/rpds_py-0.29.0-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:80fdf53d36e6c72819993e35d1ebeeb8e8fc688d0c6c2b391b55e335b3afba5a", size = 394600, upload-time = "2025-11-16T14:50:26.956Z" },
+    { url = "https://files.pythonhosted.org/packages/37/f1/dc9312fc9bec040ece08396429f2bd9e0977924ba7a11c5ad7056428465e/rpds_py-0.29.0-pp311-pypy311_pp73-manylinux_2_31_riscv64.whl", hash = "sha256:ea7173df5d86f625f8dde6d5929629ad811ed8decda3b60ae603903839ac9ac0", size = 408634, upload-time = "2025-11-16T14:50:28.989Z" },
+    { url = "https://files.pythonhosted.org/packages/ed/41/65024c9fd40c89bb7d604cf73beda4cbdbcebe92d8765345dd65855b6449/rpds_py-0.29.0-pp311-pypy311_pp73-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:76054d540061eda273274f3d13a21a4abdde90e13eaefdc205db37c05230efce", size = 426064, upload-time = "2025-11-16T14:50:30.674Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/e0/cf95478881fc88ca2fdbf56381d7df36567cccc39a05394beac72182cd62/rpds_py-0.29.0-pp311-pypy311_pp73-musllinux_1_2_aarch64.whl", hash = "sha256:9f84c549746a5be3bc7415830747a3a0312573afc9f95785eb35228bb17742ec", size = 575871, upload-time = "2025-11-16T14:50:33.428Z" },
+    { url = "https://files.pythonhosted.org/packages/ea/c0/df88097e64339a0218b57bd5f9ca49898e4c394db756c67fccc64add850a/rpds_py-0.29.0-pp311-pypy311_pp73-musllinux_1_2_i686.whl", hash = "sha256:0ea962671af5cb9a260489e311fa22b2e97103e3f9f0caaea6f81390af96a9ed", size = 601702, upload-time = "2025-11-16T14:50:36.051Z" },
+    { url = "https://files.pythonhosted.org/packages/87/f4/09ffb3ebd0cbb9e2c7c9b84d252557ecf434cd71584ee1e32f66013824df/rpds_py-0.29.0-pp311-pypy311_pp73-musllinux_1_2_x86_64.whl", hash = "sha256:f7728653900035fb7b8d06e1e5900545d8088efc9d5d4545782da7df03ec803f", size = 564054, upload-time = "2025-11-16T14:50:37.733Z" },
+]
+
 [[package]]
 name = "rsa"
 version = "4.9.1"
@@ -4515,24 +4512,32 @@ wheels = [
 
 [[package]]
 name = "safetensors"
-version = "0.6.2"
+version = "0.7.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/ac/cc/738f3011628920e027a11754d9cae9abec1aed00f7ae860abbf843755233/safetensors-0.6.2.tar.gz", hash = "sha256:43ff2aa0e6fa2dc3ea5524ac7ad93a9839256b8703761e76e2d0b2a3fa4f15d9", size = 197968, upload-time = "2025-08-08T13:13:58.654Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/4d/b1/3f5fd73c039fc87dba3ff8b5d528bfc5a32b597fea8e7a6a4800343a17c7/safetensors-0.6.2-cp38-abi3-macosx_10_12_x86_64.whl", hash = "sha256:9c85ede8ec58f120bad982ec47746981e210492a6db876882aa021446af8ffba", size = 454797, upload-time = "2025-08-08T13:13:52.066Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/c9/bb114c158540ee17907ec470d01980957fdaf87b4aa07914c24eba87b9c6/safetensors-0.6.2-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:d6675cf4b39c98dbd7d940598028f3742e0375a6b4d4277e76beb0c35f4b843b", size = 432206, upload-time = "2025-08-08T13:13:50.931Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/8e/f70c34e47df3110e8e0bb268d90db8d4be8958a54ab0336c9be4fe86dac8/safetensors-0.6.2-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1d2d2b3ce1e2509c68932ca03ab8f20570920cd9754b05063d4368ee52833ecd", size = 473261, upload-time = "2025-08-08T13:13:41.259Z" },
-    { url = "https://files.pythonhosted.org/packages/2a/f5/be9c6a7c7ef773e1996dc214e73485286df1836dbd063e8085ee1976f9cb/safetensors-0.6.2-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:93de35a18f46b0f5a6a1f9e26d91b442094f2df02e9fd7acf224cfec4238821a", size = 485117, upload-time = "2025-08-08T13:13:43.506Z" },
-    { url = "https://files.pythonhosted.org/packages/c9/55/23f2d0a2c96ed8665bf17a30ab4ce5270413f4d74b6d87dd663258b9af31/safetensors-0.6.2-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:89a89b505f335640f9120fac65ddeb83e40f1fd081cb8ed88b505bdccec8d0a1", size = 616154, upload-time = "2025-08-08T13:13:45.096Z" },
-    { url = "https://files.pythonhosted.org/packages/98/c6/affb0bd9ce02aa46e7acddbe087912a04d953d7a4d74b708c91b5806ef3f/safetensors-0.6.2-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:fc4d0d0b937e04bdf2ae6f70cd3ad51328635fe0e6214aa1fc811f3b576b3bda", size = 520713, upload-time = "2025-08-08T13:13:46.25Z" },
-    { url = "https://files.pythonhosted.org/packages/fe/5d/5a514d7b88e310c8b146e2404e0dc161282e78634d9358975fd56dfd14be/safetensors-0.6.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8045db2c872db8f4cbe3faa0495932d89c38c899c603f21e9b6486951a5ecb8f", size = 485835, upload-time = "2025-08-08T13:13:49.373Z" },
-    { url = "https://files.pythonhosted.org/packages/7a/7b/4fc3b2ba62c352b2071bea9cfbad330fadda70579f617506ae1a2f129cab/safetensors-0.6.2-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:81e67e8bab9878bb568cffbc5f5e655adb38d2418351dc0859ccac158f753e19", size = 521503, upload-time = "2025-08-08T13:13:47.651Z" },
-    { url = "https://files.pythonhosted.org/packages/5a/50/0057e11fe1f3cead9254315a6c106a16dd4b1a19cd247f7cc6414f6b7866/safetensors-0.6.2-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:b0e4d029ab0a0e0e4fdf142b194514695b1d7d3735503ba700cf36d0fc7136ce", size = 652256, upload-time = "2025-08-08T13:13:53.167Z" },
-    { url = "https://files.pythonhosted.org/packages/e9/29/473f789e4ac242593ac1656fbece6e1ecd860bb289e635e963667807afe3/safetensors-0.6.2-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:fa48268185c52bfe8771e46325a1e21d317207bcabcb72e65c6e28e9ffeb29c7", size = 747281, upload-time = "2025-08-08T13:13:54.656Z" },
-    { url = "https://files.pythonhosted.org/packages/68/52/f7324aad7f2df99e05525c84d352dc217e0fa637a4f603e9f2eedfbe2c67/safetensors-0.6.2-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:d83c20c12c2d2f465997c51b7ecb00e407e5f94d7dec3ea0cc11d86f60d3fde5", size = 692286, upload-time = "2025-08-08T13:13:55.884Z" },
-    { url = "https://files.pythonhosted.org/packages/ad/fe/cad1d9762868c7c5dc70c8620074df28ebb1a8e4c17d4c0cb031889c457e/safetensors-0.6.2-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:d944cea65fad0ead848b6ec2c37cc0b197194bec228f8020054742190e9312ac", size = 655957, upload-time = "2025-08-08T13:13:57.029Z" },
-    { url = "https://files.pythonhosted.org/packages/59/a7/e2158e17bbe57d104f0abbd95dff60dda916cf277c9f9663b4bf9bad8b6e/safetensors-0.6.2-cp38-abi3-win32.whl", hash = "sha256:cab75ca7c064d3911411461151cb69380c9225798a20e712b102edda2542ddb1", size = 308926, upload-time = "2025-08-08T13:14:01.095Z" },
-    { url = "https://files.pythonhosted.org/packages/2c/c3/c0be1135726618dc1e28d181b8c442403d8dbb9e273fd791de2d4384bcdd/safetensors-0.6.2-cp38-abi3-win_amd64.whl", hash = "sha256:c7b214870df923cbc1593c3faee16bec59ea462758699bd3fee399d00aac072c", size = 320192, upload-time = "2025-08-08T13:13:59.467Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/29/9c/6e74567782559a63bd040a236edca26fd71bc7ba88de2ef35d75df3bca5e/safetensors-0.7.0.tar.gz", hash = "sha256:07663963b67e8bd9f0b8ad15bb9163606cd27cc5a1b96235a50d8369803b96b0", size = 200878, upload-time = "2025-11-19T15:18:43.199Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/fa/47/aef6c06649039accf914afef490268e1067ed82be62bcfa5b7e886ad15e8/safetensors-0.7.0-cp38-abi3-macosx_10_12_x86_64.whl", hash = "sha256:c82f4d474cf725255d9e6acf17252991c3c8aac038d6ef363a4bf8be2f6db517", size = 467781, upload-time = "2025-11-19T15:18:35.84Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/00/374c0c068e30cd31f1e1b46b4b5738168ec79e7689ca82ee93ddfea05109/safetensors-0.7.0-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:94fd4858284736bb67a897a41608b5b0c2496c9bdb3bf2af1fa3409127f20d57", size = 447058, upload-time = "2025-11-19T15:18:34.416Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/06/578ffed52c2296f93d7fd2d844cabfa92be51a587c38c8afbb8ae449ca89/safetensors-0.7.0-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e07d91d0c92a31200f25351f4acb2bc6aff7f48094e13ebb1d0fb995b54b6542", size = 491748, upload-time = "2025-11-19T15:18:09.79Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/33/1debbbb70e4791dde185edb9413d1fe01619255abb64b300157d7f15dddd/safetensors-0.7.0-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:8469155f4cb518bafb4acf4865e8bb9d6804110d2d9bdcaa78564b9fd841e104", size = 503881, upload-time = "2025-11-19T15:18:16.145Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/1c/40c2ca924d60792c3be509833df711b553c60effbd91da6f5284a83f7122/safetensors-0.7.0-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:54bef08bf00a2bff599982f6b08e8770e09cc012d7bba00783fc7ea38f1fb37d", size = 623463, upload-time = "2025-11-19T15:18:21.11Z" },
+    { url = "https://files.pythonhosted.org/packages/9b/3a/13784a9364bd43b0d61eef4bea2845039bc2030458b16594a1bd787ae26e/safetensors-0.7.0-cp38-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:42cb091236206bb2016d245c377ed383aa7f78691748f3bb6ee1bfa51ae2ce6a", size = 532855, upload-time = "2025-11-19T15:18:25.719Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/60/429e9b1cb3fc651937727befe258ea24122d9663e4d5709a48c9cbfceecb/safetensors-0.7.0-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dac7252938f0696ddea46f5e855dd3138444e82236e3be475f54929f0c510d48", size = 507152, upload-time = "2025-11-19T15:18:33.023Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/a8/4b45e4e059270d17af60359713ffd83f97900d45a6afa73aaa0d737d48b6/safetensors-0.7.0-cp38-abi3-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1d060c70284127fa805085d8f10fbd0962792aed71879d00864acda69dbab981", size = 541856, upload-time = "2025-11-19T15:18:31.075Z" },
+    { url = "https://files.pythonhosted.org/packages/06/87/d26d8407c44175d8ae164a95b5a62707fcc445f3c0c56108e37d98070a3d/safetensors-0.7.0-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:cdab83a366799fa730f90a4ebb563e494f28e9e92c4819e556152ad55e43591b", size = 674060, upload-time = "2025-11-19T15:18:37.211Z" },
+    { url = "https://files.pythonhosted.org/packages/11/f5/57644a2ff08dc6325816ba7217e5095f17269dada2554b658442c66aed51/safetensors-0.7.0-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:672132907fcad9f2aedcb705b2d7b3b93354a2aec1b2f706c4db852abe338f85", size = 771715, upload-time = "2025-11-19T15:18:38.689Z" },
+    { url = "https://files.pythonhosted.org/packages/86/31/17883e13a814bd278ae6e266b13282a01049b0c81341da7fd0e3e71a80a3/safetensors-0.7.0-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:5d72abdb8a4d56d4020713724ba81dac065fedb7f3667151c4a637f1d3fb26c0", size = 714377, upload-time = "2025-11-19T15:18:40.162Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/d8/0c8a7dc9b41dcac53c4cbf9df2b9c83e0e0097203de8b37a712b345c0be5/safetensors-0.7.0-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:b0f6d66c1c538d5a94a73aa9ddca8ccc4227e6c9ff555322ea40bdd142391dd4", size = 677368, upload-time = "2025-11-19T15:18:41.627Z" },
+    { url = "https://files.pythonhosted.org/packages/05/e5/cb4b713c8a93469e3c5be7c3f8d77d307e65fe89673e731f5c2bfd0a9237/safetensors-0.7.0-cp38-abi3-win32.whl", hash = "sha256:c74af94bf3ac15ac4d0f2a7c7b4663a15f8c2ab15ed0fc7531ca61d0835eccba", size = 326423, upload-time = "2025-11-19T15:18:45.74Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/e6/ec8471c8072382cb91233ba7267fd931219753bb43814cbc71757bfd4dab/safetensors-0.7.0-cp38-abi3-win_amd64.whl", hash = "sha256:d1239932053f56f3456f32eb9625590cc7582e905021f94636202a864d470755", size = 341380, upload-time = "2025-11-19T15:18:44.427Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/6a/4d08d89a6fcbe905c5ae68b8b34f0791850882fc19782d0d02c65abbdf3b/safetensors-0.7.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f4729811a6640d019a4b7ba8638ee2fd21fa5ca8c7e7bdf0fed62068fcaac737", size = 492430, upload-time = "2025-11-19T15:18:11.884Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/29/59ed8152b30f72c42d00d241e58eaca558ae9dbfa5695206e2e0f54c7063/safetensors-0.7.0-pp310-pypy310_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:12f49080303fa6bb424b362149a12949dfbbf1e06811a88f2307276b0c131afd", size = 503977, upload-time = "2025-11-19T15:18:17.523Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/0b/4811bfec67fa260e791369b16dab105e4bae82686120554cc484064e22b4/safetensors-0.7.0-pp310-pypy310_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0071bffba4150c2f46cae1432d31995d77acfd9f8db598b5d1a2ce67e8440ad2", size = 623890, upload-time = "2025-11-19T15:18:22.666Z" },
+    { url = "https://files.pythonhosted.org/packages/58/5b/632a58724221ef03d78ab65062e82a1010e1bef8e8e0b9d7c6d7b8044841/safetensors-0.7.0-pp310-pypy310_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:473b32699f4200e69801bf5abf93f1a4ecd432a70984df164fc22ccf39c4a6f3", size = 531885, upload-time = "2025-11-19T15:18:27.146Z" },
+    { url = "https://files.pythonhosted.org/packages/94/60/13ccb63ea85bfe2e4fe6af602cf1272155f048906556d5ec8509da9dba42/safetensors-0.7.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7b95a3fa7b3abb9b5b0e07668e808364d0d40f6bbbf9ae0faa8b5b210c97b140", size = 492627, upload-time = "2025-11-19T15:18:14.661Z" },
+    { url = "https://files.pythonhosted.org/packages/2e/2b/e2fde0d6334439908b0b0c4cba18b8ad76ea6a03b569d4a3388f423b4046/safetensors-0.7.0-pp39-pypy39_pp73-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:cfdead2f57330d76aa7234051dadfa7d4eedc0e5a27fd08e6f96714a92b00f09", size = 503861, upload-time = "2025-11-19T15:18:19.418Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/71/566e3dd559a9cef1b4775c239daae09e6b6a32ca8b45eb1db9a4dfa1ba81/safetensors-0.7.0-pp39-pypy39_pp73-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:dc92bc2db7b45bda4510e4f51c59b00fe80b2d6be88928346e4294ce1c2abe7c", size = 623577, upload-time = "2025-11-19T15:18:24.275Z" },
+    { url = "https://files.pythonhosted.org/packages/82/fc/3035c5c30c8a5a82c31c6b2ad6f8bcd45ea2ddd9a8088840406bcf997413/safetensors-0.7.0-pp39-pypy39_pp73-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:6999421eb8ba9df4450a16d9184fcb7bef26240b9f98e95401f17af6c2210b71", size = 532524, upload-time = "2025-11-19T15:18:29.334Z" },
 ]
 
 [[package]]
@@ -4616,7 +4621,7 @@ wheels = [
 
 [[package]]
 name = "scipy"
-version = "1.16.2"
+version = "1.16.3"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
     "python_full_version >= '3.12'",
@@ -4625,28 +4630,28 @@ resolution-markers = [
 dependencies = [
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/4c/3b/546a6f0bfe791bbb7f8d591613454d15097e53f906308ec6f7c1ce588e8e/scipy-1.16.2.tar.gz", hash = "sha256:af029b153d243a80afb6eabe40b0a07f8e35c9adc269c019f364ad747f826a6b", size = 30580599, upload-time = "2025-09-11T17:48:08.271Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/0b/ef/37ed4b213d64b48422df92560af7300e10fe30b5d665dd79932baebee0c6/scipy-1.16.2-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:6ab88ea43a57da1af33292ebd04b417e8e2eaf9d5aa05700be8d6e1b6501cd92", size = 36619956, upload-time = "2025-09-11T17:39:20.5Z" },
-    { url = "https://files.pythonhosted.org/packages/85/ab/5c2eba89b9416961a982346a4d6a647d78c91ec96ab94ed522b3b6baf444/scipy-1.16.2-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:c95e96c7305c96ede73a7389f46ccd6c659c4da5ef1b2789466baeaed3622b6e", size = 28931117, upload-time = "2025-09-11T17:39:29.06Z" },
-    { url = "https://files.pythonhosted.org/packages/80/d1/eed51ab64d227fe60229a2d57fb60ca5898cfa50ba27d4f573e9e5f0b430/scipy-1.16.2-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:87eb178db04ece7c698220d523c170125dbffebb7af0345e66c3554f6f60c173", size = 20921997, upload-time = "2025-09-11T17:39:34.892Z" },
-    { url = "https://files.pythonhosted.org/packages/be/7c/33ea3e23bbadde96726edba6bf9111fb1969d14d9d477ffa202c67bec9da/scipy-1.16.2-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:4e409eac067dcee96a57fbcf424c13f428037827ec7ee3cb671ff525ca4fc34d", size = 23523374, upload-time = "2025-09-11T17:39:40.846Z" },
-    { url = "https://files.pythonhosted.org/packages/96/0b/7399dc96e1e3f9a05e258c98d716196a34f528eef2ec55aad651ed136d03/scipy-1.16.2-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:e574be127bb760f0dad24ff6e217c80213d153058372362ccb9555a10fc5e8d2", size = 33583702, upload-time = "2025-09-11T17:39:49.011Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/bc/a5c75095089b96ea72c1bd37a4497c24b581ec73db4ef58ebee142ad2d14/scipy-1.16.2-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f5db5ba6188d698ba7abab982ad6973265b74bb40a1efe1821b58c87f73892b9", size = 35883427, upload-time = "2025-09-11T17:39:57.406Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/66/e25705ca3d2b87b97fe0a278a24b7f477b4023a926847935a1a71488a6a6/scipy-1.16.2-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:ec6e74c4e884104ae006d34110677bfe0098203a3fec2f3faf349f4cb05165e3", size = 36212940, upload-time = "2025-09-11T17:40:06.013Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/fd/0bb911585e12f3abdd603d721d83fc1c7492835e1401a0e6d498d7822b4b/scipy-1.16.2-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:912f46667d2d3834bc3d57361f854226475f695eb08c08a904aadb1c936b6a88", size = 38865092, upload-time = "2025-09-11T17:40:15.143Z" },
-    { url = "https://files.pythonhosted.org/packages/d6/73/c449a7d56ba6e6f874183759f8483cde21f900a8be117d67ffbb670c2958/scipy-1.16.2-cp311-cp311-win_amd64.whl", hash = "sha256:91e9e8a37befa5a69e9cacbe0bcb79ae5afb4a0b130fd6db6ee6cc0d491695fa", size = 38687626, upload-time = "2025-09-11T17:40:24.041Z" },
-    { url = "https://files.pythonhosted.org/packages/68/72/02f37316adf95307f5d9e579023c6899f89ff3a051fa079dbd6faafc48e5/scipy-1.16.2-cp311-cp311-win_arm64.whl", hash = "sha256:f3bf75a6dcecab62afde4d1f973f1692be013110cad5338007927db8da73249c", size = 25503506, upload-time = "2025-09-11T17:40:30.703Z" },
-    { url = "https://files.pythonhosted.org/packages/b7/8d/6396e00db1282279a4ddd507c5f5e11f606812b608ee58517ce8abbf883f/scipy-1.16.2-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:89d6c100fa5c48472047632e06f0876b3c4931aac1f4291afc81a3644316bb0d", size = 36646259, upload-time = "2025-09-11T17:40:39.329Z" },
-    { url = "https://files.pythonhosted.org/packages/3b/93/ea9edd7e193fceb8eef149804491890bde73fb169c896b61aa3e2d1e4e77/scipy-1.16.2-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:ca748936cd579d3f01928b30a17dc474550b01272d8046e3e1ee593f23620371", size = 28888976, upload-time = "2025-09-11T17:40:46.82Z" },
-    { url = "https://files.pythonhosted.org/packages/91/4d/281fddc3d80fd738ba86fd3aed9202331180b01e2c78eaae0642f22f7e83/scipy-1.16.2-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:fac4f8ce2ddb40e2e3d0f7ec36d2a1e7f92559a2471e59aec37bd8d9de01fec0", size = 20879905, upload-time = "2025-09-11T17:40:52.545Z" },
-    { url = "https://files.pythonhosted.org/packages/69/40/b33b74c84606fd301b2915f0062e45733c6ff5708d121dd0deaa8871e2d0/scipy-1.16.2-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:033570f1dcefd79547a88e18bccacff025c8c647a330381064f561d43b821232", size = 23553066, upload-time = "2025-09-11T17:40:59.014Z" },
-    { url = "https://files.pythonhosted.org/packages/55/a7/22c739e2f21a42cc8f16bc76b47cff4ed54fbe0962832c589591c2abec34/scipy-1.16.2-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:ea3421209bf00c8a5ef2227de496601087d8f638a2363ee09af059bd70976dc1", size = 33336407, upload-time = "2025-09-11T17:41:06.796Z" },
-    { url = "https://files.pythonhosted.org/packages/53/11/a0160990b82999b45874dc60c0c183d3a3a969a563fffc476d5a9995c407/scipy-1.16.2-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:f66bd07ba6f84cd4a380b41d1bf3c59ea488b590a2ff96744845163309ee8e2f", size = 35673281, upload-time = "2025-09-11T17:41:15.055Z" },
-    { url = "https://files.pythonhosted.org/packages/96/53/7ef48a4cfcf243c3d0f1643f5887c81f29fdf76911c4e49331828e19fc0a/scipy-1.16.2-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:5e9feab931bd2aea4a23388c962df6468af3d808ddf2d40f94a81c5dc38f32ef", size = 36004222, upload-time = "2025-09-11T17:41:23.868Z" },
-    { url = "https://files.pythonhosted.org/packages/49/7f/71a69e0afd460049d41c65c630c919c537815277dfea214031005f474d78/scipy-1.16.2-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:03dfc75e52f72cf23ec2ced468645321407faad8f0fe7b1f5b49264adbc29cb1", size = 38664586, upload-time = "2025-09-11T17:41:31.021Z" },
-    { url = "https://files.pythonhosted.org/packages/34/95/20e02ca66fb495a95fba0642fd48e0c390d0ece9b9b14c6e931a60a12dea/scipy-1.16.2-cp312-cp312-win_amd64.whl", hash = "sha256:0ce54e07bbb394b417457409a64fd015be623f36e330ac49306433ffe04bc97e", size = 38550641, upload-time = "2025-09-11T17:41:36.61Z" },
-    { url = "https://files.pythonhosted.org/packages/92/ad/13646b9beb0a95528ca46d52b7babafbe115017814a611f2065ee4e61d20/scipy-1.16.2-cp312-cp312-win_arm64.whl", hash = "sha256:2a8ffaa4ac0df81a0b94577b18ee079f13fecdb924df3328fc44a7dc5ac46851", size = 25456070, upload-time = "2025-09-11T17:41:41.3Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/0a/ca/d8ace4f98322d01abcd52d381134344bf7b431eba7ed8b42bdea5a3c2ac9/scipy-1.16.3.tar.gz", hash = "sha256:01e87659402762f43bd2fee13370553a17ada367d42e7487800bf2916535aecb", size = 30597883, upload-time = "2025-10-28T17:38:54.068Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9b/5f/6f37d7439de1455ce9c5a556b8d1db0979f03a796c030bafdf08d35b7bf9/scipy-1.16.3-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:40be6cf99e68b6c4321e9f8782e7d5ff8265af28ef2cd56e9c9b2638fa08ad97", size = 36630881, upload-time = "2025-10-28T17:31:47.104Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/89/d70e9f628749b7e4db2aa4cd89735502ff3f08f7b9b27d2e799485987cd9/scipy-1.16.3-cp311-cp311-macosx_12_0_arm64.whl", hash = "sha256:8be1ca9170fcb6223cc7c27f4305d680ded114a1567c0bd2bfcbf947d1b17511", size = 28941012, upload-time = "2025-10-28T17:31:53.411Z" },
+    { url = "https://files.pythonhosted.org/packages/a8/a8/0e7a9a6872a923505dbdf6bb93451edcac120363131c19013044a1e7cb0c/scipy-1.16.3-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:bea0a62734d20d67608660f69dcda23e7f90fb4ca20974ab80b6ed40df87a005", size = 20931935, upload-time = "2025-10-28T17:31:57.361Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/c7/020fb72bd79ad798e4dbe53938543ecb96b3a9ac3fe274b7189e23e27353/scipy-1.16.3-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:2a207a6ce9c24f1951241f4693ede2d393f59c07abc159b2cb2be980820e01fb", size = 23534466, upload-time = "2025-10-28T17:32:01.875Z" },
+    { url = "https://files.pythonhosted.org/packages/be/a0/668c4609ce6dbf2f948e167836ccaf897f95fb63fa231c87da7558a374cd/scipy-1.16.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:532fb5ad6a87e9e9cd9c959b106b73145a03f04c7d57ea3e6f6bb60b86ab0876", size = 33593618, upload-time = "2025-10-28T17:32:06.902Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/6e/8942461cf2636cdae083e3eb72622a7fbbfa5cf559c7d13ab250a5dbdc01/scipy-1.16.3-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:0151a0749efeaaab78711c78422d413c583b8cdd2011a3c1d6c794938ee9fdb2", size = 35899798, upload-time = "2025-10-28T17:32:12.665Z" },
+    { url = "https://files.pythonhosted.org/packages/79/e8/d0f33590364cdbd67f28ce79368b373889faa4ee959588beddf6daef9abe/scipy-1.16.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:b7180967113560cca57418a7bc719e30366b47959dd845a93206fbed693c867e", size = 36226154, upload-time = "2025-10-28T17:32:17.961Z" },
+    { url = "https://files.pythonhosted.org/packages/39/c1/1903de608c0c924a1749c590064e65810f8046e437aba6be365abc4f7557/scipy-1.16.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:deb3841c925eeddb6afc1e4e4a45e418d19ec7b87c5df177695224078e8ec733", size = 38878540, upload-time = "2025-10-28T17:32:23.907Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/d0/22ec7036ba0b0a35bccb7f25ab407382ed34af0b111475eb301c16f8a2e5/scipy-1.16.3-cp311-cp311-win_amd64.whl", hash = "sha256:53c3844d527213631e886621df5695d35e4f6a75f620dca412bcd292f6b87d78", size = 38722107, upload-time = "2025-10-28T17:32:29.921Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/60/8a00e5a524bb3bf8898db1650d350f50e6cffb9d7a491c561dc9826c7515/scipy-1.16.3-cp311-cp311-win_arm64.whl", hash = "sha256:9452781bd879b14b6f055b26643703551320aa8d79ae064a71df55c00286a184", size = 25506272, upload-time = "2025-10-28T17:32:34.577Z" },
+    { url = "https://files.pythonhosted.org/packages/40/41/5bf55c3f386b1643812f3a5674edf74b26184378ef0f3e7c7a09a7e2ca7f/scipy-1.16.3-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:81fc5827606858cf71446a5e98715ba0e11f0dbc83d71c7409d05486592a45d6", size = 36659043, upload-time = "2025-10-28T17:32:40.285Z" },
+    { url = "https://files.pythonhosted.org/packages/1e/0f/65582071948cfc45d43e9870bf7ca5f0e0684e165d7c9ef4e50d783073eb/scipy-1.16.3-cp312-cp312-macosx_12_0_arm64.whl", hash = "sha256:c97176013d404c7346bf57874eaac5187d969293bf40497140b0a2b2b7482e07", size = 28898986, upload-time = "2025-10-28T17:32:45.325Z" },
+    { url = "https://files.pythonhosted.org/packages/96/5e/36bf3f0ac298187d1ceadde9051177d6a4fe4d507e8f59067dc9dd39e650/scipy-1.16.3-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:2b71d93c8a9936046866acebc915e2af2e292b883ed6e2cbe5c34beb094b82d9", size = 20889814, upload-time = "2025-10-28T17:32:49.277Z" },
+    { url = "https://files.pythonhosted.org/packages/80/35/178d9d0c35394d5d5211bbff7ac4f2986c5488b59506fef9e1de13ea28d3/scipy-1.16.3-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:3d4a07a8e785d80289dfe66b7c27d8634a773020742ec7187b85ccc4b0e7b686", size = 23565795, upload-time = "2025-10-28T17:32:53.337Z" },
+    { url = "https://files.pythonhosted.org/packages/fa/46/d1146ff536d034d02f83c8afc3c4bab2eddb634624d6529a8512f3afc9da/scipy-1.16.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:0553371015692a898e1aa858fed67a3576c34edefa6b7ebdb4e9dde49ce5c203", size = 33349476, upload-time = "2025-10-28T17:32:58.353Z" },
+    { url = "https://files.pythonhosted.org/packages/79/2e/415119c9ab3e62249e18c2b082c07aff907a273741b3f8160414b0e9193c/scipy-1.16.3-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:72d1717fd3b5e6ec747327ce9bda32d5463f472c9dce9f54499e81fbd50245a1", size = 35676692, upload-time = "2025-10-28T17:33:03.88Z" },
+    { url = "https://files.pythonhosted.org/packages/27/82/df26e44da78bf8d2aeaf7566082260cfa15955a5a6e96e6a29935b64132f/scipy-1.16.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:1fb2472e72e24d1530debe6ae078db70fb1605350c88a3d14bc401d6306dbffe", size = 36019345, upload-time = "2025-10-28T17:33:09.773Z" },
+    { url = "https://files.pythonhosted.org/packages/82/31/006cbb4b648ba379a95c87262c2855cd0d09453e500937f78b30f02fa1cd/scipy-1.16.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:c5192722cffe15f9329a3948c4b1db789fbb1f05c97899187dcf009b283aea70", size = 38678975, upload-time = "2025-10-28T17:33:15.809Z" },
+    { url = "https://files.pythonhosted.org/packages/c2/7f/acbd28c97e990b421af7d6d6cd416358c9c293fc958b8529e0bd5d2a2a19/scipy-1.16.3-cp312-cp312-win_amd64.whl", hash = "sha256:56edc65510d1331dae01ef9b658d428e33ed48b4f77b1d51caf479a0253f96dc", size = 38555926, upload-time = "2025-10-28T17:33:21.388Z" },
+    { url = "https://files.pythonhosted.org/packages/ce/69/c5c7807fd007dad4f48e0a5f2153038dc96e8725d3345b9ee31b2b7bed46/scipy-1.16.3-cp312-cp312-win_arm64.whl", hash = "sha256:a8a26c78ef223d3e30920ef759e25625a0ecdd0d60e5a8818b7513c3e5384cf2", size = 25463014, upload-time = "2025-10-28T17:33:25.975Z" },
 ]
 
 [[package]]
@@ -4689,69 +4694,22 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/8d/21/55267fcdc1a99612248c299c12d8f7c51d6e5ae992f32694d17adcf13fda/sentencepiece-0.2.1-cp39-cp39-win_arm64.whl", hash = "sha256:3d165fbb9bf8fba35f1946ba2617c3f9995679f07438325f07c026d53f33e746", size = 1033914, upload-time = "2025-08-12T06:59:10.705Z" },
 ]
 
-[[package]]
-name = "sentry-sdk"
-version = "2.40.0"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "certifi" },
-    { name = "urllib3" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/4f/b5/ce879ce3292e5ca41fa3ebf68f60645032eca813c9ed8f92dcf09804c0e3/sentry_sdk-2.40.0.tar.gz", hash = "sha256:b9c4672fb2cafabcc28586ab8fd0ceeff9b2352afcf2b936e13d5ba06d141b9f", size = 351703, upload-time = "2025-10-06T12:27:29.207Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/a4/d1/a54bd3622c6e742e6a01bc3bac45966b7ba886e29827da6b8ca7ae234e21/sentry_sdk-2.40.0-py2.py3-none-any.whl", hash = "sha256:d5f6ae0f27ea73e7b09c70ad7d42242326eb44765e87a15d8c5aab96b80013e6", size = 374747, upload-time = "2025-10-06T12:27:27.051Z" },
-]
-
-[[package]]
-name = "setproctitle"
-version = "1.3.7"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/8d/48/49393a96a2eef1ab418b17475fb92b8fcfad83d099e678751b05472e69de/setproctitle-1.3.7.tar.gz", hash = "sha256:bc2bc917691c1537d5b9bca1468437176809c7e11e5694ca79a9ca12345dcb9e", size = 27002, upload-time = "2025-09-05T12:51:25.278Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/0e/f8/17bda581c517678260e6541b600eeb67745f53596dc077174141ba2f6702/setproctitle-1.3.7-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:00afa6fc507967d8c9d592a887cdc6c1f5742ceac6a4354d111ca0214847732c", size = 31793, upload-time = "2025-09-05T12:49:10.297Z" },
-    { url = "https://files.pythonhosted.org/packages/27/d1/76a33ae80d4e788ecab9eb9b53db03e81cfc95367ec7e3fbf4989962fedd/setproctitle-1.3.7-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9e02667f6b9fc1238ba753c0f4b0a37ae184ce8f3bbbc38e115d99646b3f4cd3", size = 32779, upload-time = "2025-09-05T12:49:12.157Z" },
-    { url = "https://files.pythonhosted.org/packages/59/27/1a07c38121967061564f5e0884414a5ab11a783260450172d4fc68c15621/setproctitle-1.3.7-cp310-cp310-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:83fcd271567d133eb9532d3b067c8a75be175b2b3b271e2812921a05303a693f", size = 34578, upload-time = "2025-09-05T12:49:13.393Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/d4/725e6353935962d8bb12cbf7e7abba1d0d738c7f6935f90239d8e1ccf913/setproctitle-1.3.7-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:13fe37951dda1a45c35d77d06e3da5d90e4f875c4918a7312b3b4556cfa7ff64", size = 32030, upload-time = "2025-09-05T12:49:15.362Z" },
-    { url = "https://files.pythonhosted.org/packages/67/24/e4677ae8e1cb0d549ab558b12db10c175a889be0974c589c428fece5433e/setproctitle-1.3.7-cp310-cp310-musllinux_1_2_ppc64le.whl", hash = "sha256:a05509cfb2059e5d2ddff701d38e474169e9ce2a298cf1b6fd5f3a213a553fe5", size = 33363, upload-time = "2025-09-05T12:49:16.829Z" },
-    { url = "https://files.pythonhosted.org/packages/55/d4/69ce66e4373a48fdbb37489f3ded476bb393e27f514968c3a69a67343ae0/setproctitle-1.3.7-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:6da835e76ae18574859224a75db6e15c4c2aaa66d300a57efeaa4c97ca4c7381", size = 31508, upload-time = "2025-09-05T12:49:18.032Z" },
-    { url = "https://files.pythonhosted.org/packages/18/2e/bd03ff02432a181c1787f6fc2a678f53b7dacdd5ded69c318fe1619556e8/setproctitle-1.3.7-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:1607b963e7b53e24ec8a2cb4e0ab3ae591d7c6bf0a160feef0551da63452b37f", size = 32191, upload-time = "2025-09-05T12:49:24.567Z" },
-    { url = "https://files.pythonhosted.org/packages/28/78/1e62fc0937a8549f2220445ed2175daacee9b6764c7963b16148119b016d/setproctitle-1.3.7-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a20fb1a3974e2dab857870cf874b325b8705605cb7e7e8bcbb915bca896f52a9", size = 33203, upload-time = "2025-09-05T12:49:25.871Z" },
-    { url = "https://files.pythonhosted.org/packages/a0/3c/65edc65db3fa3df400cf13b05e9d41a3c77517b4839ce873aa6b4043184f/setproctitle-1.3.7-cp311-cp311-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f8d961bba676e07d77665204f36cffaa260f526e7b32d07ab3df6a2c1dfb44ba", size = 34963, upload-time = "2025-09-05T12:49:27.044Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/32/89157e3de997973e306e44152522385f428e16f92f3cf113461489e1e2ee/setproctitle-1.3.7-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:db0fd964fbd3a9f8999b502f65bd2e20883fdb5b1fae3a424e66db9a793ed307", size = 32398, upload-time = "2025-09-05T12:49:28.909Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/18/77a765a339ddf046844cb4513353d8e9dcd8183da9cdba6e078713e6b0b2/setproctitle-1.3.7-cp311-cp311-musllinux_1_2_ppc64le.whl", hash = "sha256:db116850fcf7cca19492030f8d3b4b6e231278e8fe097a043957d22ce1bdf3ee", size = 33657, upload-time = "2025-09-05T12:49:30.323Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/63/f0b6205c64d74d2a24a58644a38ec77bdbaa6afc13747e75973bf8904932/setproctitle-1.3.7-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:316664d8b24a5c91ee244460bdaf7a74a707adaa9e14fbe0dc0a53168bb9aba1", size = 31836, upload-time = "2025-09-05T12:49:32.309Z" },
-    { url = "https://files.pythonhosted.org/packages/d0/99/71630546b9395b095f4082be41165d1078204d1696c2d9baade3de3202d0/setproctitle-1.3.7-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:2906b6c7959cdb75f46159bf0acd8cc9906cf1361c9e1ded0d065fe8f9039629", size = 32932, upload-time = "2025-09-05T12:49:39.271Z" },
-    { url = "https://files.pythonhosted.org/packages/50/22/cee06af4ffcfb0e8aba047bd44f5262e644199ae7527ae2c1f672b86495c/setproctitle-1.3.7-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6915964a6dda07920a1159321dcd6d94fc7fc526f815ca08a8063aeca3c204f1", size = 33736, upload-time = "2025-09-05T12:49:40.565Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/00/a5949a8bb06ef5e7df214fc393bb2fb6aedf0479b17214e57750dfdd0f24/setproctitle-1.3.7-cp312-cp312-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:cff72899861c765bd4021d1ff1c68d60edc129711a2fdba77f9cb69ef726a8b6", size = 35605, upload-time = "2025-09-05T12:49:42.362Z" },
-    { url = "https://files.pythonhosted.org/packages/b0/3a/50caca532a9343828e3bf5778c7a84d6c737a249b1796d50dd680290594d/setproctitle-1.3.7-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:b7cb05bd446687ff816a3aaaf831047fc4c364feff7ada94a66024f1367b448c", size = 33143, upload-time = "2025-09-05T12:49:43.515Z" },
-    { url = "https://files.pythonhosted.org/packages/ca/14/b843a251296ce55e2e17c017d6b9f11ce0d3d070e9265de4ecad948b913d/setproctitle-1.3.7-cp312-cp312-musllinux_1_2_ppc64le.whl", hash = "sha256:3a57b9a00de8cae7e2a1f7b9f0c2ac7b69372159e16a7708aa2f38f9e5cc987a", size = 34434, upload-time = "2025-09-05T12:49:45.31Z" },
-    { url = "https://files.pythonhosted.org/packages/c8/b7/06145c238c0a6d2c4bc881f8be230bb9f36d2bf51aff7bddcb796d5eed67/setproctitle-1.3.7-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:d8828b356114f6b308b04afe398ed93803d7fca4a955dd3abe84430e28d33739", size = 32795, upload-time = "2025-09-05T12:49:46.419Z" },
-    { url = "https://files.pythonhosted.org/packages/7b/b7/1abfcfe0de1b50c96002e7ecfd6a2c109798bc4284f0dc02635c7da22e9e/setproctitle-1.3.7-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:134e7f66703a1d92c0a9a0a417c580f2cc04b93d31d3fc0dd43c3aa194b706e1", size = 31551, upload-time = "2025-09-05T12:51:05.088Z" },
-    { url = "https://files.pythonhosted.org/packages/67/fa/2b25384c154bfb36112d7ec377281776e337e5124590bd96c4f3f076285a/setproctitle-1.3.7-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9796732a040f617fc933f9531c9a84bb73c5c27b8074abbe52907076e804b2b7", size = 32489, upload-time = "2025-09-05T12:51:06.207Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/24/e4f2a3467cd6322462134f1918a565b99e8bd43c1d47040082e874a3f512/setproctitle-1.3.7-cp39-cp39-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ff3c1c32382fb71a200db8bab3df22f32e6ac7ec3170e92fa5b542cf42eed9a2", size = 34328, upload-time = "2025-09-05T12:51:07.504Z" },
-    { url = "https://files.pythonhosted.org/packages/33/75/7ca8f54e35c62f99562fcf1e052d4f162f310e2e14ab3e34d5965b5a7f71/setproctitle-1.3.7-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:01f27b5b72505b304152cb0bd7ff410cc4f2d69ac70c21a7fdfa64400a68642d", size = 31744, upload-time = "2025-09-05T12:51:08.698Z" },
-    { url = "https://files.pythonhosted.org/packages/19/74/080caa13f1dad1ffab1ef7b7dbef97f278c23d4aa45bdbd209aac54e9b21/setproctitle-1.3.7-cp39-cp39-musllinux_1_2_ppc64le.whl", hash = "sha256:80b6a562cbc92b289c28f34ce709a16b26b1696e9b9a0542a675ce3a788bdf3f", size = 33100, upload-time = "2025-09-05T12:51:10.53Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/bc/478b45ccbcf4be609d4ae5b46b4601960d7a7194cca2075cfae57b0cbb76/setproctitle-1.3.7-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:c4fb90174d176473122e7eef7c6492d53761826f34ff61c81a1c1d66905025d3", size = 31233, upload-time = "2025-09-05T12:51:12.03Z" },
-    { url = "https://files.pythonhosted.org/packages/41/89/5b6f2faedd6ced3d3c085a5efbd91380fb1f61f4c12bc42acad37932f4e9/setproctitle-1.3.7-pp310-pypy310_pp73-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:502b902a0e4c69031b87870ff4986c290ebbb12d6038a70639f09c331b18efb2", size = 14284, upload-time = "2025-09-05T12:51:18.393Z" },
-    { url = "https://files.pythonhosted.org/packages/73/02/b9eadc226195dcfa90eed37afe56b5dd6fa2f0e5220ab8b7867b8862b926/setproctitle-1.3.7-pp311-pypy311_pp73-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f1704c9e041f2b1dc38f5be4552e141e1432fba3dd52c72eeffd5bc2db04dc65", size = 14286, upload-time = "2025-09-05T12:51:22.61Z" },
-]
-
 [[package]]
 name = "setuptools"
-version = "79.0.1"
+version = "80.9.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/bb/71/b6365e6325b3290e14957b2c3a804a529968c77a049b2ed40c095f749707/setuptools-79.0.1.tar.gz", hash = "sha256:128ce7b8f33c3079fd1b067ecbb4051a66e8526e7b65f6cec075dfc650ddfa88", size = 1367909, upload-time = "2025-04-23T22:20:59.241Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/18/5d/3bf57dcd21979b887f014ea83c24ae194cfcd12b9e0fda66b957c69d1fca/setuptools-80.9.0.tar.gz", hash = "sha256:f36b47402ecde768dbfafc46e8e4207b4360c654f1f3bb84475f0a28628fb19c", size = 1319958, upload-time = "2025-05-27T00:56:51.443Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/0d/6d/b4752b044bf94cb802d88a888dc7d288baaf77d7910b7dedda74b5ceea0c/setuptools-79.0.1-py3-none-any.whl", hash = "sha256:e147c0549f27767ba362f9da434eab9c5dc0045d5304feb602a0af001089fc51", size = 1256281, upload-time = "2025-04-23T22:20:56.768Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/dc/17031897dae0efacfea57dfd3a82fdd2a2aeb58e0ff71b77b87e44edc772/setuptools-80.9.0-py3-none-any.whl", hash = "sha256:062d34222ad13e0cc312a4c02d73f059e86a4acbfbdea8f8f76b28c99f306922", size = 1201486, upload-time = "2025-05-27T00:56:49.664Z" },
 ]
 
 [[package]]
-name = "shellingham"
-version = "1.5.4"
+name = "shortuuid"
+version = "1.0.13"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/58/15/8b3609fd3830ef7b27b655beb4b4e9c62313a4e8da8c676e142cc210d58e/shellingham-1.5.4.tar.gz", hash = "sha256:8dbca0739d487e5bd35ab3ca4b36e11c4078f3a234bfce294b0a0291363404de", size = 10310, upload-time = "2023-10-24T04:13:40.426Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/8c/e2/bcf761f3bff95856203f9559baf3741c416071dd200c0fc19fad7f078f86/shortuuid-1.0.13.tar.gz", hash = "sha256:3bb9cf07f606260584b1df46399c0b87dd84773e7b25912b7e391e30797c5e72", size = 9662, upload-time = "2024-03-11T20:11:06.879Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e0/f9/0595336914c5619e5f28a1fb793285925a8cd4b432c9da0a987836c7f822/shellingham-1.5.4-py2.py3-none-any.whl", hash = "sha256:7ecfff8f2fd72616f7481040475a65b2bf8af90a56c89140852d1120324e8686", size = 9755, upload-time = "2023-10-24T04:13:38.866Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/44/21d6bf170bf40b41396480d8d49ad640bca3f2b02139cd52aa1e272830a5/shortuuid-1.0.13-py3-none-any.whl", hash = "sha256:a482a497300b49b4953e15108a7913244e1bb0d41f9d332f5e9925dba33a3c5a", size = 10529, upload-time = "2024-03-11T20:11:04.807Z" },
 ]
 
 [[package]]
@@ -4834,52 +4792,46 @@ wheels = [
 ]
 
 [[package]]
-name = "soundfile"
-version = "0.13.1"
+name = "starlette"
+version = "0.49.3"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "cffi" },
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "anyio", marker = "python_full_version < '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/e1/41/9b873a8c055582859b239be17902a85339bec6a30ad162f98c9b0288a2cc/soundfile-0.13.1.tar.gz", hash = "sha256:b2c68dab1e30297317080a5b43df57e302584c49e2942defdde0acccc53f0e5b", size = 46156, upload-time = "2025-01-25T09:17:04.831Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/de/1a/608df0b10b53b0beb96a37854ee05864d182ddd4b1156a22f1ad3860425a/starlette-0.49.3.tar.gz", hash = "sha256:1c14546f299b5901a1ea0e34410575bc33bbd741377a10484a54445588d00284", size = 2655031, upload-time = "2025-11-01T15:12:26.13Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/64/28/e2a36573ccbcf3d57c00626a21fe51989380636e821b341d36ccca0c1c3a/soundfile-0.13.1-py2.py3-none-any.whl", hash = "sha256:a23c717560da2cf4c7b5ae1142514e0fd82d6bbd9dfc93a50423447142f2c445", size = 25751, upload-time = "2025-01-25T09:16:44.235Z" },
-    { url = "https://files.pythonhosted.org/packages/58/ae/c0e4a53d77cf6e9a04179535766b3321b0b9ced5f70522e4caf9329f0046/soundfile-0.13.1-py2.py3-none-manylinux_2_28_aarch64.whl", hash = "sha256:9c9e855f5a4d06ce4213f31918653ab7de0c5a8d8107cd2427e44b42df547deb", size = 1235729, upload-time = "2025-01-25T09:16:53.018Z" },
-    { url = "https://files.pythonhosted.org/packages/57/5e/70bdd9579b35003a489fc850b5047beeda26328053ebadc1fb60f320f7db/soundfile-0.13.1-py2.py3-none-manylinux_2_28_x86_64.whl", hash = "sha256:03267c4e493315294834a0870f31dbb3b28a95561b80b134f0bd3cf2d5f0e618", size = 1313646, upload-time = "2025-01-25T09:16:54.872Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/e0/021c772d6a662f43b63044ab481dc6ac7592447605b5b35a957785363122/starlette-0.49.3-py3-none-any.whl", hash = "sha256:b579b99715fdc2980cf88c8ec96d3bf1ce16f5a8051a7c2b84ef9b1cdecaea2f", size = 74340, upload-time = "2025-11-01T15:12:24.387Z" },
 ]
 
 [[package]]
-name = "soxr"
-version = "1.0.0"
+name = "starlette"
+version = "0.50.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
 dependencies = [
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "anyio", marker = "python_full_version >= '3.10'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/42/7e/f4b461944662ad75036df65277d6130f9411002bfb79e9df7dff40a31db9/soxr-1.0.0.tar.gz", hash = "sha256:e07ee6c1d659bc6957034f4800c60cb8b98de798823e34d2a2bba1caa85a4509", size = 171415, upload-time = "2025-09-07T13:22:21.317Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/ba/b8/73a0e6a6e079a9d9cfa64113d771e421640b6f679a52eeb9b32f72d871a1/starlette-0.50.0.tar.gz", hash = "sha256:a2a17b22203254bcbc2e1f926d2d55f3f9497f769416b3190768befe598fa3ca", size = 2646985, upload-time = "2025-11-01T15:25:27.516Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/fc/91/00384166f110a3888ea8efd44523ba7168dd2dc39e3e43c931cc2d069fa9/soxr-1.0.0-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b89685faedebc45af71f08f9957b61cc6143bc94ba43fe38e97067f81e272969", size = 208586, upload-time = "2025-09-07T13:21:50.341Z" },
-    { url = "https://files.pythonhosted.org/packages/75/34/e18f1003e242aabed44ed8902534814d3e64209e4d1d874f5b9b67d73cde/soxr-1.0.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d255741b2f0084fd02d4a2ddd77cd495be9e7e7b6f9dba1c9494f86afefac65b", size = 242310, upload-time = "2025-09-07T13:21:51.56Z" },
-    { url = "https://files.pythonhosted.org/packages/75/12/4b49611c9ba5e9fe6f807d0a83352516808e8e573f8b4e712fc0c17f3363/soxr-1.0.0-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3f15450e6f65f22f02fcd4c5a9219c873b1e583a73e232805ff160c759a6b586", size = 208867, upload-time = "2025-09-07T13:21:57.076Z" },
-    { url = "https://files.pythonhosted.org/packages/cc/70/92146ab970a3ef8c43ac160035b1e52fde5417f89adb10572f7e788d9596/soxr-1.0.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1f73f57452f9df37b4de7a4052789fcbd474a5b28f38bba43278ae4b489d4384", size = 242633, upload-time = "2025-09-07T13:21:58.621Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/80/10640970998a1d2199bef6c4d92205f36968cddaf3e4d0e9fe35ddd405bd/soxr-1.0.0-cp312-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e8ce273cca101aff3d8c387db5a5a41001ba76ef1837883438d3c652507a9ccc", size = 204707, upload-time = "2025-09-07T13:22:05.125Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/87/2726603c13c2126cb8ded9e57381b7377f4f0df6ba4408e1af5ddbfdc3dd/soxr-1.0.0-cp312-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e8f2a69686f2856d37823bbb7b78c3d44904f311fe70ba49b893af11d6b6047b", size = 238032, upload-time = "2025-09-07T13:22:06.428Z" },
-    { url = "https://files.pythonhosted.org/packages/64/e4/449c59781d9d942bfa88f691099935b41b207f39704515cc7d368c017f9f/soxr-1.0.0-cp39-cp39-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bb86c342862697dbd4a44043f275e5196f2d2c49dca374c78f19b7893988675d", size = 208806, upload-time = "2025-09-07T13:22:17.487Z" },
-    { url = "https://files.pythonhosted.org/packages/e8/89/9af231937ba3bc0c6d42955431290eeab0efcf37987bf0d69f5734bda78a/soxr-1.0.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3d2a4fadd88207c2991fb08c29fc189e7b2e298b598a94ea1747e42c8acb7a01", size = 242642, upload-time = "2025-09-07T13:22:18.727Z" },
+    { url = "https://files.pythonhosted.org/packages/d9/52/1064f510b141bd54025f9b55105e26d1fa970b9be67ad766380a3c9b74b0/starlette-0.50.0-py3-none-any.whl", hash = "sha256:9e5391843ec9b6e472eed1365a78c8098cfceb7a74bfd4d6b1c0c0095efb3bca", size = 74033, upload-time = "2025-11-01T15:25:25.461Z" },
 ]
 
 [[package]]
-name = "starlette"
-version = "0.48.0"
+name = "svgwrite"
+version = "1.4.3"
 source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "anyio" },
-    { name = "typing-extensions" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/a7/a5/d6f429d43394057b67a6b5bbe6eae2f77a6bf7459d961fdb224bf206eee6/starlette-0.48.0.tar.gz", hash = "sha256:7e8cee469a8ab2352911528110ce9088fdc6a37d9876926e73da7ce4aa4c7a46", size = 2652949, upload-time = "2025-09-13T08:41:05.699Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/16/c1/263d4e93b543390d86d8eb4fc23d9ce8a8d6efd146f9427364109004fa9b/svgwrite-1.4.3.zip", hash = "sha256:a8fbdfd4443302a6619a7f76bc937fc683daf2628d9b737c891ec08b8ce524c3", size = 189516, upload-time = "2022-07-14T14:05:26.107Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/be/72/2db2f49247d0a18b4f1bb9a5a39a0162869acf235f3a96418363947b3d46/starlette-0.48.0-py3-none-any.whl", hash = "sha256:0764ca97b097582558ecb498132ed0c7d942f233f365b86ba37770e026510659", size = 73736, upload-time = "2025-09-13T08:41:03.869Z" },
+    { url = "https://files.pythonhosted.org/packages/84/15/640e399579024a6875918839454025bb1d5f850bb70d96a11eabb644d11c/svgwrite-1.4.3-py3-none-any.whl", hash = "sha256:bb6b2b5450f1edbfa597d924f9ac2dd099e625562e492021d7dd614f65f8a22d", size = 67122, upload-time = "2022-07-14T14:05:24.459Z" },
 ]
 
 [[package]]
@@ -4910,11 +4862,13 @@ source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "absl-py" },
     { name = "grpcio" },
-    { name = "markdown" },
+    { name = "markdown", version = "3.9", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "markdown", version = "3.10", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "packaging" },
-    { name = "pillow" },
+    { name = "pillow", version = "11.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "pillow", version = "12.0.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "protobuf" },
     { name = "setuptools" },
     { name = "tensorboard-data-server" },
@@ -4945,9 +4899,10 @@ dependencies = [
     { name = "gast" },
     { name = "google-pasta" },
     { name = "grpcio" },
-    { name = "h5py" },
+    { name = "h5py", version = "3.14.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "h5py", version = "3.15.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "keras", version = "3.10.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "keras", version = "3.11.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "keras", version = "3.12.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "libclang" },
     { name = "ml-dtypes" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
@@ -4959,7 +4914,8 @@ dependencies = [
     { name = "setuptools" },
     { name = "six" },
     { name = "tensorboard" },
-    { name = "termcolor" },
+    { name = "termcolor", version = "3.1.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "termcolor", version = "3.2.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "typing-extensions" },
     { name = "wrapt" },
 ]
@@ -5022,13 +4978,11 @@ name = "tensorstore"
 version = "0.1.78"
 source = { registry = "https://pypi.org/simple" }
 resolution-markers = [
-    "python_full_version >= '3.12'",
-    "python_full_version == '3.11.*'",
     "python_full_version == '3.10.*'",
 ]
 dependencies = [
-    { name = "ml-dtypes", marker = "python_full_version >= '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "ml-dtypes", marker = "python_full_version == '3.10.*'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/9f/ee/05eb424437f4db63331c90e4605025eedc0f71da3faff97161d5d7b405af/tensorstore-0.1.78.tar.gz", hash = "sha256:e26074ffe462394cf54197eb76d6569b500f347573cd74da3f4dd5f510a4ad7c", size = 6913502, upload-time = "2025-10-06T17:44:29.649Z" }
 wheels = [
@@ -5049,15 +5003,58 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/1e/46/4ff3e395c44348c7442523c8ddd8ccc72d9ac81838e7a8f6afdd92131c3e/tensorstore-0.1.78-cp312-cp312-win_amd64.whl", hash = "sha256:56271d4652a7cb445879089f620af47801c091765d35a005505d6bfb8d00c535", size = 12711274, upload-time = "2025-10-06T17:44:17.586Z" },
 ]
 
+[[package]]
+name = "tensorstore"
+version = "0.1.79"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+]
+dependencies = [
+    { name = "ml-dtypes", marker = "python_full_version >= '3.11'" },
+    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/26/2c/50ab489a0862ca88d2d766130a6fec45ccd5174f0e04081d8b7b07a8aedd/tensorstore-0.1.79.tar.gz", hash = "sha256:8dad44a8a7f2952a5d0030a8bd868b3cfdff048bd40ab53e7226f3d8b0881c5e", size = 7075782, upload-time = "2025-11-11T22:05:23.824Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/68/a9/1695d7ea197c4568c2f02f34b203eef702ec8080422331f00a65c6fb2a37/tensorstore-0.1.79-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:11a2c62694ea9c21770bc5a09938d3d15c4b9662b738ae6e1e513c26ed96251a", size = 16466511, upload-time = "2025-11-11T22:04:18.614Z" },
+    { url = "https://files.pythonhosted.org/packages/db/0e/5ce8a615c7f9ad7cf8ed4ac6e182fe0ef46fd06fef89757e49ba84a6ba9e/tensorstore-0.1.79-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:5e152d334bf34fbabdfe8e5bc35b87d1f9947065924ff83c29e659308b36e948", size = 14499810, upload-time = "2025-11-11T22:04:21.725Z" },
+    { url = "https://files.pythonhosted.org/packages/c0/29/2cb9552138fe84ab29421489121350e4af0502eafff31ccd9017490be0d8/tensorstore-0.1.79-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c4230b8fd29795e88e441f749d881973eca8dadf33c5262b367839fb8891f79b", size = 18937510, upload-time = "2025-11-11T22:04:24.221Z" },
+    { url = "https://files.pythonhosted.org/packages/42/70/d2a672a93faebdd176cd8541405cd5614b14d3d8dc812fbeaf2cf46d390a/tensorstore-0.1.79-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:83072ee0e551d6dca582e154b64c8b8066d276ec0759784e3149c28212a61f18", size = 20910324, upload-time = "2025-11-11T22:04:26.769Z" },
+    { url = "https://files.pythonhosted.org/packages/91/d5/7958cbfb614c4ffa5070ae9575874d46937067c0d81a7739e67fb1d62de5/tensorstore-0.1.79-cp311-cp311-win_amd64.whl", hash = "sha256:6c98c6b74c00e00eba7969292144e471d5c45d67088f0dc08e3a4c60a15ee191", size = 13206191, upload-time = "2025-11-11T22:04:29.254Z" },
+    { url = "https://files.pythonhosted.org/packages/f1/a2/a77be16b4a882ace36da0748305795f35306bdad568472f208bd89b96b9d/tensorstore-0.1.79-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:71aa9b45436d888c37b965f7b71195916d15438119b7dccb66a3b0776bfba367", size = 16485740, upload-time = "2025-11-11T22:04:33.478Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/e4/7fe268ec41aa70b71a1c56b1ec83346fbcbf12f4bfbefc79d14fb9c03408/tensorstore-0.1.79-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:108c0e867aa2c87d4982cc6325a2de0c4f5bd63c2bea18adb193a370c40594ce", size = 14508736, upload-time = "2025-11-11T22:04:38.613Z" },
+    { url = "https://files.pythonhosted.org/packages/5a/f1/b1248dae02598ce534834413e841f915a32ab185c36ecd05e4c67bdc8d19/tensorstore-0.1.79-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:debd435042c00be68ba1fb3cf59325a7babb3f4a3cf4744c87dde346802cbbb4", size = 18947817, upload-time = "2025-11-11T22:04:40.768Z" },
+    { url = "https://files.pythonhosted.org/packages/87/4a/60e234147570e21bbab4ac70ab79dd794a5ef9a4945d36c34c1914a73205/tensorstore-0.1.79-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:608f7178ec6e4e4a3c26545b0a44f44bf83438d04bf2d960cd0e7699eaa99ef6", size = 20929832, upload-time = "2025-11-11T22:04:43.613Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/48/0531868bce12a2f520002e810d4200ec6f01ba33a2f27b6bd7289fbc197b/tensorstore-0.1.79-cp312-cp312-win_amd64.whl", hash = "sha256:a071c6c255b7e412957a6aa563bc4250242c7894edad06ae6358e3d30b7d88ce", size = 13211970, upload-time = "2025-11-11T22:04:46.179Z" },
+]
+
 [[package]]
 name = "termcolor"
 version = "3.1.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 sdist = { url = "https://files.pythonhosted.org/packages/ca/6c/3d75c196ac07ac8749600b60b03f4f6094d54e132c4d94ebac6ee0e0add0/termcolor-3.1.0.tar.gz", hash = "sha256:6a6dd7fbee581909eeec6a756cff1d7f7c376063b14e4a298dc4980309e55970", size = 14324, upload-time = "2025-04-30T11:37:53.791Z" }
 wheels = [
     { url = "https://files.pythonhosted.org/packages/4f/bd/de8d508070629b6d84a30d01d57e4a65c69aa7f5abe7560b8fad3b50ea59/termcolor-3.1.0-py3-none-any.whl", hash = "sha256:591dd26b5c2ce03b9e43f391264626557873ce1d379019786f99b0c2bee140aa", size = 7684, upload-time = "2025-04-30T11:37:52.382Z" },
 ]
 
+[[package]]
+name = "termcolor"
+version = "3.2.0"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+sdist = { url = "https://files.pythonhosted.org/packages/87/56/ab275c2b56a5e2342568838f0d5e3e66a32354adcc159b495e374cda43f5/termcolor-3.2.0.tar.gz", hash = "sha256:610e6456feec42c4bcd28934a8c87a06c3fa28b01561d46aa09a9881b8622c58", size = 14423, upload-time = "2025-10-25T19:11:42.586Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/f9/d5/141f53d7c1eb2a80e6d3e9a390228c3222c27705cbe7f048d3623053f3ca/termcolor-3.2.0-py3-none-any.whl", hash = "sha256:a10343879eba4da819353c55cb8049b0933890c2ebf9ad5d3ecd2bb32ea96ea6", size = 7698, upload-time = "2025-10-25T19:11:41.536Z" },
+]
+
 [[package]]
 name = "tf-keras"
 version = "2.20.1"
@@ -5125,71 +5122,67 @@ wheels = [
 
 [[package]]
 name = "tomli"
-version = "2.2.1"
+version = "2.3.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/18/87/302344fed471e44a87289cf4967697d07e532f2421fdaf868a303cbae4ff/tomli-2.2.1.tar.gz", hash = "sha256:cd45e1dc79c835ce60f7404ec8119f2eb06d38b1deba146f07ced3bbc44505ff", size = 17175, upload-time = "2024-11-27T22:38:36.873Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/43/ca/75707e6efa2b37c77dadb324ae7d9571cb424e61ea73fad7c56c2d14527f/tomli-2.2.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:678e4fa69e4575eb77d103de3df8a895e1591b48e740211bd1067378c69e8249", size = 131077, upload-time = "2024-11-27T22:37:54.956Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/16/51ae563a8615d472fdbffc43a3f3d46588c264ac4f024f63f01283becfbb/tomli-2.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:023aa114dd824ade0100497eb2318602af309e5a55595f76b626d6d9f3b7b0a6", size = 123429, upload-time = "2024-11-27T22:37:56.698Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/dd/4f6cd1e7b160041db83c694abc78e100473c15d54620083dbd5aae7b990e/tomli-2.2.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ece47d672db52ac607a3d9599a9d48dcb2f2f735c6c2d1f34130085bb12b112a", size = 226067, upload-time = "2024-11-27T22:37:57.63Z" },
-    { url = "https://files.pythonhosted.org/packages/a9/6b/c54ede5dc70d648cc6361eaf429304b02f2871a345bbdd51e993d6cdf550/tomli-2.2.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6972ca9c9cc9f0acaa56a8ca1ff51e7af152a9f87fb64623e31d5c83700080ee", size = 236030, upload-time = "2024-11-27T22:37:59.344Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/47/999514fa49cfaf7a92c805a86c3c43f4215621855d151b61c602abb38091/tomli-2.2.1-cp311-cp311-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c954d2250168d28797dd4e3ac5cf812a406cd5a92674ee4c8f123c889786aa8e", size = 240898, upload-time = "2024-11-27T22:38:00.429Z" },
-    { url = "https://files.pythonhosted.org/packages/73/41/0a01279a7ae09ee1573b423318e7934674ce06eb33f50936655071d81a24/tomli-2.2.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:8dd28b3e155b80f4d54beb40a441d366adcfe740969820caf156c019fb5c7ec4", size = 229894, upload-time = "2024-11-27T22:38:02.094Z" },
-    { url = "https://files.pythonhosted.org/packages/55/18/5d8bc5b0a0362311ce4d18830a5d28943667599a60d20118074ea1b01bb7/tomli-2.2.1-cp311-cp311-musllinux_1_2_i686.whl", hash = "sha256:e59e304978767a54663af13c07b3d1af22ddee3bb2fb0618ca1593e4f593a106", size = 245319, upload-time = "2024-11-27T22:38:03.206Z" },
-    { url = "https://files.pythonhosted.org/packages/92/a3/7ade0576d17f3cdf5ff44d61390d4b3febb8a9fc2b480c75c47ea048c646/tomli-2.2.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:33580bccab0338d00994d7f16f4c4ec25b776af3ffaac1ed74e0b3fc95e885a8", size = 238273, upload-time = "2024-11-27T22:38:04.217Z" },
-    { url = "https://files.pythonhosted.org/packages/72/6f/fa64ef058ac1446a1e51110c375339b3ec6be245af9d14c87c4a6412dd32/tomli-2.2.1-cp311-cp311-win32.whl", hash = "sha256:465af0e0875402f1d226519c9904f37254b3045fc5084697cefb9bdde1ff99ff", size = 98310, upload-time = "2024-11-27T22:38:05.908Z" },
-    { url = "https://files.pythonhosted.org/packages/6a/1c/4a2dcde4a51b81be3530565e92eda625d94dafb46dbeb15069df4caffc34/tomli-2.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:2d0f2fdd22b02c6d81637a3c95f8cd77f995846af7414c5c4b8d0545afa1bc4b", size = 108309, upload-time = "2024-11-27T22:38:06.812Z" },
-    { url = "https://files.pythonhosted.org/packages/52/e1/f8af4c2fcde17500422858155aeb0d7e93477a0d59a98e56cbfe75070fd0/tomli-2.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:4a8f6e44de52d5e6c657c9fe83b562f5f4256d8ebbfe4ff922c495620a7f6cea", size = 132762, upload-time = "2024-11-27T22:38:07.731Z" },
-    { url = "https://files.pythonhosted.org/packages/03/b8/152c68bb84fc00396b83e7bbddd5ec0bd3dd409db4195e2a9b3e398ad2e3/tomli-2.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:8d57ca8095a641b8237d5b079147646153d22552f1c637fd3ba7f4b0b29167a8", size = 123453, upload-time = "2024-11-27T22:38:09.384Z" },
-    { url = "https://files.pythonhosted.org/packages/c8/d6/fc9267af9166f79ac528ff7e8c55c8181ded34eb4b0e93daa767b8841573/tomli-2.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e340144ad7ae1533cb897d406382b4b6fede8890a03738ff1683af800d54192", size = 233486, upload-time = "2024-11-27T22:38:10.329Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/51/51c3f2884d7bab89af25f678447ea7d297b53b5a3b5730a7cb2ef6069f07/tomli-2.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:db2b95f9de79181805df90bedc5a5ab4c165e6ec3fe99f970d0e302f384ad222", size = 242349, upload-time = "2024-11-27T22:38:11.443Z" },
-    { url = "https://files.pythonhosted.org/packages/ab/df/bfa89627d13a5cc22402e441e8a931ef2108403db390ff3345c05253935e/tomli-2.2.1-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:40741994320b232529c802f8bc86da4e1aa9f413db394617b9a256ae0f9a7f77", size = 252159, upload-time = "2024-11-27T22:38:13.099Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/6e/fa2b916dced65763a5168c6ccb91066f7639bdc88b48adda990db10c8c0b/tomli-2.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:400e720fe168c0f8521520190686ef8ef033fb19fc493da09779e592861b78c6", size = 237243, upload-time = "2024-11-27T22:38:14.766Z" },
-    { url = "https://files.pythonhosted.org/packages/b4/04/885d3b1f650e1153cbb93a6a9782c58a972b94ea4483ae4ac5cedd5e4a09/tomli-2.2.1-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:02abe224de6ae62c19f090f68da4e27b10af2b93213d36cf44e6e1c5abd19fdd", size = 259645, upload-time = "2024-11-27T22:38:15.843Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/de/6b432d66e986e501586da298e28ebeefd3edc2c780f3ad73d22566034239/tomli-2.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:b82ebccc8c8a36f2094e969560a1b836758481f3dc360ce9a3277c65f374285e", size = 244584, upload-time = "2024-11-27T22:38:17.645Z" },
-    { url = "https://files.pythonhosted.org/packages/1c/9a/47c0449b98e6e7d1be6cbac02f93dd79003234ddc4aaab6ba07a9a7482e2/tomli-2.2.1-cp312-cp312-win32.whl", hash = "sha256:889f80ef92701b9dbb224e49ec87c645ce5df3fa2cc548664eb8a25e03127a98", size = 98875, upload-time = "2024-11-27T22:38:19.159Z" },
-    { url = "https://files.pythonhosted.org/packages/ef/60/9b9638f081c6f1261e2688bd487625cd1e660d0a85bd469e91d8db969734/tomli-2.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:7fc04e92e1d624a4a63c76474610238576942d6b8950a2d7f908a340494e67e4", size = 109418, upload-time = "2024-11-27T22:38:20.064Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/c2/61d3e0f47e2b74ef40a68b9e6ad5984f6241a942f7cd3bbfbdbd03861ea9/tomli-2.2.1-py3-none-any.whl", hash = "sha256:cb55c73c5f4408779d0cf3eef9f762b9c9f147a77de7b258bef0a5628adc85cc", size = 14257, upload-time = "2024-11-27T22:38:35.385Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/52/ed/3f73f72945444548f33eba9a87fc7a6e969915e7b1acc8260b30e1f76a2f/tomli-2.3.0.tar.gz", hash = "sha256:64be704a875d2a59753d80ee8a533c3fe183e3f06807ff7dc2232938ccb01549", size = 17392, upload-time = "2025-10-08T22:01:47.119Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b3/2e/299f62b401438d5fe1624119c723f5d877acc86a4c2492da405626665f12/tomli-2.3.0-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:88bd15eb972f3664f5ed4b57c1634a97153b4bac4479dcb6a495f41921eb7f45", size = 153236, upload-time = "2025-10-08T22:01:00.137Z" },
+    { url = "https://files.pythonhosted.org/packages/86/7f/d8fffe6a7aefdb61bced88fcb5e280cfd71e08939da5894161bd71bea022/tomli-2.3.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:883b1c0d6398a6a9d29b508c331fa56adbcdff647f6ace4dfca0f50e90dfd0ba", size = 148084, upload-time = "2025-10-08T22:01:01.63Z" },
+    { url = "https://files.pythonhosted.org/packages/47/5c/24935fb6a2ee63e86d80e4d3b58b222dafaf438c416752c8b58537c8b89a/tomli-2.3.0-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d1381caf13ab9f300e30dd8feadb3de072aeb86f1d34a8569453ff32a7dea4bf", size = 234832, upload-time = "2025-10-08T22:01:02.543Z" },
+    { url = "https://files.pythonhosted.org/packages/89/da/75dfd804fc11e6612846758a23f13271b76d577e299592b4371a4ca4cd09/tomli-2.3.0-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a0e285d2649b78c0d9027570d4da3425bdb49830a6156121360b3f8511ea3441", size = 242052, upload-time = "2025-10-08T22:01:03.836Z" },
+    { url = "https://files.pythonhosted.org/packages/70/8c/f48ac899f7b3ca7eb13af73bacbc93aec37f9c954df3c08ad96991c8c373/tomli-2.3.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:0a154a9ae14bfcf5d8917a59b51ffd5a3ac1fd149b71b47a3a104ca4edcfa845", size = 239555, upload-time = "2025-10-08T22:01:04.834Z" },
+    { url = "https://files.pythonhosted.org/packages/ba/28/72f8afd73f1d0e7829bfc093f4cb98ce0a40ffc0cc997009ee1ed94ba705/tomli-2.3.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:74bf8464ff93e413514fefd2be591c3b0b23231a77f901db1eb30d6f712fc42c", size = 245128, upload-time = "2025-10-08T22:01:05.84Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/eb/a7679c8ac85208706d27436e8d421dfa39d4c914dcf5fa8083a9305f58d9/tomli-2.3.0-cp311-cp311-win32.whl", hash = "sha256:00b5f5d95bbfc7d12f91ad8c593a1659b6387b43f054104cda404be6bda62456", size = 96445, upload-time = "2025-10-08T22:01:06.896Z" },
+    { url = "https://files.pythonhosted.org/packages/0a/fe/3d3420c4cb1ad9cb462fb52967080575f15898da97e21cb6f1361d505383/tomli-2.3.0-cp311-cp311-win_amd64.whl", hash = "sha256:4dc4ce8483a5d429ab602f111a93a6ab1ed425eae3122032db7e9acf449451be", size = 107165, upload-time = "2025-10-08T22:01:08.107Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/b7/40f36368fcabc518bb11c8f06379a0fd631985046c038aca08c6d6a43c6e/tomli-2.3.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:d7d86942e56ded512a594786a5ba0a5e521d02529b3826e7761a05138341a2ac", size = 154891, upload-time = "2025-10-08T22:01:09.082Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/3f/d9dd692199e3b3aab2e4e4dd948abd0f790d9ded8cd10cbaae276a898434/tomli-2.3.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:73ee0b47d4dad1c5e996e3cd33b8a76a50167ae5f96a2607cbe8cc773506ab22", size = 148796, upload-time = "2025-10-08T22:01:10.266Z" },
+    { url = "https://files.pythonhosted.org/packages/60/83/59bff4996c2cf9f9387a0f5a3394629c7efa5ef16142076a23a90f1955fa/tomli-2.3.0-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:792262b94d5d0a466afb5bc63c7daa9d75520110971ee269152083270998316f", size = 242121, upload-time = "2025-10-08T22:01:11.332Z" },
+    { url = "https://files.pythonhosted.org/packages/45/e5/7c5119ff39de8693d6baab6c0b6dcb556d192c165596e9fc231ea1052041/tomli-2.3.0-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4f195fe57ecceac95a66a75ac24d9d5fbc98ef0962e09b2eddec5d39375aae52", size = 250070, upload-time = "2025-10-08T22:01:12.498Z" },
+    { url = "https://files.pythonhosted.org/packages/45/12/ad5126d3a278f27e6701abde51d342aa78d06e27ce2bb596a01f7709a5a2/tomli-2.3.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:e31d432427dcbf4d86958c184b9bfd1e96b5b71f8eb17e6d02531f434fd335b8", size = 245859, upload-time = "2025-10-08T22:01:13.551Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/a1/4d6865da6a71c603cfe6ad0e6556c73c76548557a8d658f9e3b142df245f/tomli-2.3.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:7b0882799624980785240ab732537fcfc372601015c00f7fc367c55308c186f6", size = 250296, upload-time = "2025-10-08T22:01:14.614Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/b7/a7a7042715d55c9ba6e8b196d65d2cb662578b4d8cd17d882d45322b0d78/tomli-2.3.0-cp312-cp312-win32.whl", hash = "sha256:ff72b71b5d10d22ecb084d345fc26f42b5143c5533db5e2eaba7d2d335358876", size = 97124, upload-time = "2025-10-08T22:01:15.629Z" },
+    { url = "https://files.pythonhosted.org/packages/06/1e/f22f100db15a68b520664eb3328fb0ae4e90530887928558112c8d1f4515/tomli-2.3.0-cp312-cp312-win_amd64.whl", hash = "sha256:1cb4ed918939151a03f33d4242ccd0aa5f11b3547d0cf30f7c74a408a5b99878", size = 107698, upload-time = "2025-10-08T22:01:16.51Z" },
+    { url = "https://files.pythonhosted.org/packages/77/b8/0135fadc89e73be292b473cb820b4f5a08197779206b33191e801feeae40/tomli-2.3.0-py3-none-any.whl", hash = "sha256:e95b1af3c5b07d9e643909b5abbec77cd9f1217e6d0bca72b0234736b9fb1f1b", size = 14408, upload-time = "2025-10-08T22:01:46.04Z" },
 ]
 
 [[package]]
 name = "toolz"
-version = "1.0.0"
+version = "1.1.0"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/8a/0b/d80dfa675bf592f636d1ea0b835eab4ec8df6e9415d8cfd766df54456123/toolz-1.0.0.tar.gz", hash = "sha256:2c86e3d9a04798ac556793bced838816296a2f085017664e4995cb40a1047a02", size = 66790, upload-time = "2024-10-04T16:17:04.001Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/11/d6/114b492226588d6ff54579d95847662fc69196bdeec318eb45393b24c192/toolz-1.1.0.tar.gz", hash = "sha256:27a5c770d068c110d9ed9323f24f1543e83b2f300a687b7891c1a6d56b697b5b", size = 52613, upload-time = "2025-10-17T04:03:21.661Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/03/98/eb27cc78ad3af8e302c9d8ff4977f5026676e130d28dd7578132a457170c/toolz-1.0.0-py3-none-any.whl", hash = "sha256:292c8f1c4e7516bf9086f8850935c799a874039c8bcf959d47b600e4c44a6236", size = 56383, upload-time = "2024-10-04T16:17:01.533Z" },
+    { url = "https://files.pythonhosted.org/packages/fb/12/5911ae3eeec47800503a238d971e51722ccea5feb8569b735184d5fcdbc0/toolz-1.1.0-py3-none-any.whl", hash = "sha256:15ccc861ac51c53696de0a5d6d4607f99c210739caf987b5d2054f3efed429d8", size = 58093, upload-time = "2025-10-17T04:03:20.435Z" },
 ]
 
 [[package]]
 name = "torch"
 version = "2.8.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
-    { name = "filelock" },
-    { name = "fsspec" },
-    { name = "jinja2" },
+    { name = "filelock", version = "3.19.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "fsspec", marker = "python_full_version < '3.10'" },
+    { name = "jinja2", marker = "python_full_version < '3.10'" },
     { name = "networkx", version = "3.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "networkx", version = "3.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "networkx", version = "3.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
-    { name = "nvidia-cublas-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cuda-cupti-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cuda-nvrtc-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cuda-runtime-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cudnn-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cufft-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cufile-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-curand-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cusolver-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cusparse-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-cusparselt-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-nccl-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-nvjitlink-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "nvidia-nvtx-cu12", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "setuptools", marker = "python_full_version >= '3.12'" },
-    { name = "sympy" },
-    { name = "triton", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "typing-extensions" },
+    { name = "nvidia-cublas-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-cupti-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-nvrtc-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-runtime-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cudnn-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufft-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufile-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-curand-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusolver-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparse-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparselt-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nccl-cu12", version = "2.27.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvjitlink-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvtx-cu12", marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "sympy", marker = "python_full_version < '3.10'" },
+    { name = "triton", version = "3.4.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "typing-extensions", marker = "python_full_version < '3.10'" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/63/28/110f7274254f1b8476c561dada127173f994afa2b1ffc044efb773c15650/torch-2.8.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:0be92c08b44009d4131d1ff7a8060d10bafdb7ddcb7359ef8d8c5169007ea905", size = 102052793, upload-time = "2025-08-06T14:53:15.852Z" },
@@ -5211,42 +5204,53 @@ wheels = [
 ]
 
 [[package]]
-name = "torchaudio"
-version = "2.8.0"
+name = "torch"
+version = "2.9.0"
 source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "torch" },
-]
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/95/c8/e46c22a3c059844bb0f1b670317c9e538b51728558326dcd9e5fffbf2ec2/torchaudio-2.8.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:d3c1b85b26a09832d139f6d6da6b66caeb51d2e16e08f8587665c44a9e1aa8f9", size = 1685620, upload-time = "2025-08-06T14:58:34.045Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/f5/69db76b564263f22c1788cc298ab1c4e2391a79fa8ba7b4a3e76d945292a/torchaudio-2.8.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:58f912bf2d289c709b42a55475b2b483becec79d9affb7684b606bb1f896b434", size = 4001714, upload-time = "2025-08-06T14:58:38.951Z" },
-    { url = "https://files.pythonhosted.org/packages/75/ca/da5d0a3bb7d114a8b590ecce14859ea0a05102bb4de68cdd1ed7a90634d6/torchaudio-2.8.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:4573c6042950c20278e3608a9a38050ba0bc72e0049e1bbfd249caf859a8029b", size = 1692033, upload-time = "2025-08-06T14:58:37.393Z" },
-    { url = "https://files.pythonhosted.org/packages/b6/ef/62ac736d8f906cc414181050e08a495a637dab985186c34bd76ea37efbc0/torchaudio-2.8.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:776c0b4ba84b9e3ddf6304b9c47cd63549d7896a6f3d5184ece074cc3d76ed6b", size = 4011716, upload-time = "2025-08-06T14:58:40.138Z" },
-    { url = "https://files.pythonhosted.org/packages/c7/0d/24dad878784f1edd62862f27173781669f0c71eb46368636787d1e364188/torchaudio-2.8.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:862e2e40bf09d865e5df080a84c1a39bbcef40e43140f4b1737eb3a389d3b38f", size = 1692930, upload-time = "2025-08-06T14:58:41.312Z" },
-    { url = "https://files.pythonhosted.org/packages/c2/a6/84d80f34472503e9eb82245d7df501c59602d75d7360e717fb9b84f91c5e/torchaudio-2.8.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:93a8583f280fe83ba021aa713319381ea71362cc87b67ee38e97a43cb2254aee", size = 4014607, upload-time = "2025-08-06T14:58:47.234Z" },
-    { url = "https://files.pythonhosted.org/packages/76/34/89203e1c7c553e463b8f0fe6746842182652bff2208607853d59ba836001/torchaudio-2.8.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:ef7f7ffa828b8d8ba5d3a569b825fc0469688e1e8962bf6577d538bd8af1387d", size = 1685905, upload-time = "2025-08-06T14:58:48.547Z" },
-    { url = "https://files.pythonhosted.org/packages/6b/a4/fab5dbbb2720fc775656b6a81f4daace069ff28a008a73d1b99e0fa995a2/torchaudio-2.8.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:a1f4bde9ce9316b6b24304c73ea9f82a2aabc4dbf8c6c2598117ea7c6ecc4db2", size = 4002156, upload-time = "2025-08-06T14:58:53.289Z" },
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
 ]
-
-[[package]]
-name = "torchvision"
-version = "0.23.0"
-source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "pillow" },
-    { name = "torch" },
+    { name = "filelock", version = "3.20.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "fsspec", marker = "python_full_version >= '3.10'" },
+    { name = "jinja2", marker = "python_full_version >= '3.10'" },
+    { name = "networkx", version = "3.4.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
+    { name = "networkx", version = "3.6", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "nvidia-cublas-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-cupti-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-nvrtc-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cuda-runtime-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cudnn-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufft-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cufile-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-curand-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusolver-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparse-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-cusparselt-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nccl-cu12", version = "2.27.5", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvjitlink-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvshmem-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "nvidia-nvtx-cu12", marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "setuptools", marker = "python_full_version >= '3.12'" },
+    { name = "sympy", marker = "python_full_version >= '3.10'" },
+    { name = "triton", version = "3.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "typing-extensions", marker = "python_full_version >= '3.10'" },
 ]
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/25/44/ddd56d1637bac42a8c5da2c8c440d8a28c431f996dd9790f32dd9a96ca6e/torchvision-0.23.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:31c583ba27426a3a04eca8c05450524105c1564db41be6632f7536ef405a6de2", size = 2394251, upload-time = "2025-08-06T14:58:01.725Z" },
-    { url = "https://files.pythonhosted.org/packages/93/f3/3cdf55bbf0f737304d997561c34ab0176222e0496b6743b0feab5995182c/torchvision-0.23.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:3932bf67256f2d095ce90a9f826f6033694c818856f4bb26794cf2ce64253e53", size = 8627497, upload-time = "2025-08-06T14:58:09.317Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/14/7b44fe766b7d11e064c539d92a172fa9689a53b69029e24f2f1f51e7dc56/torchvision-0.23.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:01dc33ee24c79148aee7cdbcf34ae8a3c9da1674a591e781577b716d233b1fa6", size = 2395543, upload-time = "2025-08-06T14:58:04.373Z" },
-    { url = "https://files.pythonhosted.org/packages/79/9c/fcb09aff941c8147d9e6aa6c8f67412a05622b0c750bcf796be4c85a58d4/torchvision-0.23.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:35c27941831b653f5101edfe62c03d196c13f32139310519e8228f35eae0e96a", size = 8628388, upload-time = "2025-08-06T14:58:07.802Z" },
-    { url = "https://files.pythonhosted.org/packages/e2/00/2f6454decc0cd67158c7890364e446aad4b91797087a57a78e72e1a8f8bc/torchvision-0.23.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:6dd7c4d329a0e03157803031bc856220c6155ef08c26d4f5bbac938acecf0948", size = 2396614, upload-time = "2025-08-06T14:58:03.116Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/b5/3e580dcbc16f39a324f3dd71b90edbf02a42548ad44d2b4893cc92b1194b/torchvision-0.23.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:4e7d31c43bc7cbecbb1a5652ac0106b436aa66e26437585fc2c4b2cf04d6014c", size = 8627108, upload-time = "2025-08-06T14:58:12.956Z" },
-    { url = "https://files.pythonhosted.org/packages/f4/e2/aafc6af854e792d212ff58e459f8d5d807568dc3f2b49ec41b677275e5a9/torchvision-0.23.0-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:6c74cbc1cbee26dd4f35f989cd80dccc40411f258dee476b29871dee4b483af0", size = 2392870, upload-time = "2025-08-06T14:58:21.303Z" },
-    { url = "https://files.pythonhosted.org/packages/5d/06/09b6a917b3759ef000428af0aa2597f983e20d9fbbcfeb826750f778fe6d/torchvision-0.23.0-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:a9e9d7552d34547b80843eaf64ab0737b19b2e8bec2514286b8cfd30861ca8b5", size = 8630400, upload-time = "2025-08-06T14:58:24.139Z" },
+    { url = "https://files.pythonhosted.org/packages/bb/86/245c240d2138c17ed572c943c289056c2721abab70810d772c6bf5495b28/torch-2.9.0-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:030bbfe367379ae6a4ae4042b6c44da25383343b8b3c68abaa9c7231efbaf2dd", size = 104213554, upload-time = "2025-10-15T15:45:59.798Z" },
+    { url = "https://files.pythonhosted.org/packages/58/1d/fd1e88ae0948825efcab7dd66d12bec23f05d4d38ed81573c8d453c14c06/torch-2.9.0-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:51cb63902182a78e90886e8068befd8ea102af4b00e420263591a3d70c7d3c6c", size = 899795167, upload-time = "2025-10-15T15:47:12.695Z" },
+    { url = "https://files.pythonhosted.org/packages/63/5a/496197b45c14982bef4e079b24c61dc108e3ab0d0cc9718dba9f54f45a46/torch-2.9.0-cp310-cp310-win_amd64.whl", hash = "sha256:3f6aad4d2f0ee2248bac25339d74858ff846c3969b27d14ac235821f055af83d", size = 109310314, upload-time = "2025-10-15T15:46:16.633Z" },
+    { url = "https://files.pythonhosted.org/packages/58/b0/2b4e647b0fc706e88eb6c253d05511865578f5f67b55fad639bf3272a4a1/torch-2.9.0-cp310-none-macosx_11_0_arm64.whl", hash = "sha256:413e1654c9203733138858780e184d9fc59442f0b3b209e16f39354eb893db9b", size = 74452019, upload-time = "2025-10-15T15:46:04.296Z" },
+    { url = "https://files.pythonhosted.org/packages/58/fe/334225e6330e672b36aef23d77451fa906ea12881570c08638a91331a212/torch-2.9.0-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:c596708b5105d0b199215acf0c9be7c1db5f1680d88eddadf4b75a299259a677", size = 104230578, upload-time = "2025-10-15T15:46:08.182Z" },
+    { url = "https://files.pythonhosted.org/packages/05/cc/49566caaa218872ec9a2912456f470ff92649894a4bc2e5274aa9ef87c4a/torch-2.9.0-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:51de31219c97c51cf4bf2be94d622e3deb5dcc526c6dc00e97c17eaec0fc1d67", size = 899815990, upload-time = "2025-10-15T15:48:03.336Z" },
+    { url = "https://files.pythonhosted.org/packages/74/25/e9ab21d5925b642d008f139d4a3c9664fc9ee1faafca22913c080cc4c0a5/torch-2.9.0-cp311-cp311-win_amd64.whl", hash = "sha256:dd515c70059afd95f48b8192733764c08ca37a1d19803af6401b5ecad7c8676e", size = 109313698, upload-time = "2025-10-15T15:46:12.425Z" },
+    { url = "https://files.pythonhosted.org/packages/b3/b7/205ef3e94de636feffd64b28bb59a0dfac0771221201b9871acf9236f5ca/torch-2.9.0-cp311-none-macosx_11_0_arm64.whl", hash = "sha256:614a185e4986326d526a91210c8fc1397e76e8cfafa78baf6296a790e53a9eec", size = 74463678, upload-time = "2025-10-15T15:46:29.779Z" },
+    { url = "https://files.pythonhosted.org/packages/d1/d3/3985739f3b8e88675127bf70f82b3a48ae083e39cda56305dbd90398fec0/torch-2.9.0-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:e5f7af1dc4c0a7c4a260c2534f41ddaf209714f7c89145e644c44712fbd6b642", size = 104107898, upload-time = "2025-10-15T15:46:20.883Z" },
+    { url = "https://files.pythonhosted.org/packages/a5/4b/f4bb2e6c25d0272f798cd6d7a04ed315da76cec68c602d87040c7847287f/torch-2.9.0-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:01cff95ecd9a212ea2f141db28acccdceb6a4c54f64e6c51091146f5e2a772c6", size = 899738273, upload-time = "2025-10-15T15:50:04.188Z" },
+    { url = "https://files.pythonhosted.org/packages/66/11/c1c5ba6691cda6279087c35bd626536e4fd29521fe740abf5008377a9a02/torch-2.9.0-cp312-cp312-win_amd64.whl", hash = "sha256:4582b162f541651f0cb184d3e291c05c2f556c7117c64a9873e2ee158d40062b", size = 109280887, upload-time = "2025-10-15T15:46:26.228Z" },
+    { url = "https://files.pythonhosted.org/packages/dd/5f/b85bd8c05312d71de9402bf5868d217c38827cfd09d8f8514e5be128a52b/torch-2.9.0-cp312-none-macosx_11_0_arm64.whl", hash = "sha256:33f58e9a102a91259af289d50525c30323b5c9ae1d31322b6447c0814da68695", size = 74478983, upload-time = "2025-10-15T15:46:39.406Z" },
 ]
 
 [[package]]
@@ -5263,10 +5267,11 @@ wheels = [
 
 [[package]]
 name = "transformers"
-version = "4.57.0"
+version = "4.57.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "filelock" },
+    { name = "filelock", version = "3.19.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "filelock", version = "3.20.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "huggingface-hub" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
@@ -5278,9 +5283,9 @@ dependencies = [
     { name = "tokenizers" },
     { name = "tqdm" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f3/5c/a22c39dac2687f3fe2a6b97e2c1ae516e91cd4d3976a7a2b7c24ff2fae48/transformers-4.57.0.tar.gz", hash = "sha256:d045753f3d93f9216e693cdb168698dfd2e9d3aad1bb72579a5d60ebf1545a8b", size = 10142956, upload-time = "2025-10-03T17:03:47.177Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/d6/68/a39307bcc4116a30b2106f2e689130a48de8bd8a1e635b5e1030e46fcd9e/transformers-4.57.1.tar.gz", hash = "sha256:f06c837959196c75039809636cd964b959f6604b75b8eeec6fdfc0440b89cc55", size = 10142511, upload-time = "2025-10-14T15:39:26.18Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e5/2b/4d2708ac1ff5cd708b6548f4c5812d0ae40d1c28591c4c1c762b6dbdef2d/transformers-4.57.0-py3-none-any.whl", hash = "sha256:9d7c6d098c026e40d897e017ed1f481ab803cbac041021dbc6ae6100e4949b55", size = 11990588, upload-time = "2025-10-03T17:03:43.629Z" },
+    { url = "https://files.pythonhosted.org/packages/71/d3/c16c3b3cf7655a67db1144da94b021c200ac1303f82428f2beef6c2e72bb/transformers-4.57.1-py3-none-any.whl", hash = "sha256:b10d05da8fa67dc41644dbbf9bc45a44cb86ae33da6f9295f5fbf5b7890bd267", size = 11990925, upload-time = "2025-10-14T15:39:23.085Z" },
 ]
 
 [[package]]
@@ -5299,9 +5304,12 @@ wheels = [
 name = "triton"
 version = "3.4.0"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
     { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
-    { name = "setuptools" },
+    { name = "setuptools", marker = "python_full_version < '3.10'" },
 ]
 wheels = [
     { url = "https://files.pythonhosted.org/packages/62/ee/0ee5f64a87eeda19bbad9bc54ae5ca5b98186ed00055281fd40fb4beb10e/triton-3.4.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7ff2785de9bc02f500e085420273bb5cc9c9bb767584a4aa28d6e360cec70128", size = 155430069, upload-time = "2025-07-30T19:58:21.715Z" },
@@ -5311,19 +5319,18 @@ wheels = [
 ]
 
 [[package]]
-name = "typer"
-version = "0.19.2"
+name = "triton"
+version = "3.5.0"
 source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "rich" },
-    { name = "shellingham" },
-    { name = "typing-extensions" },
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/21/ca/950278884e2ca20547ff3eb109478c6baf6b8cf219318e6bc4f666fad8e8/typer-0.19.2.tar.gz", hash = "sha256:9ad824308ded0ad06cc716434705f691d4ee0bfd0fb081839d2e426860e7fdca", size = 104755, upload-time = "2025-09-23T09:47:48.256Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/00/22/35617eee79080a5d071d0f14ad698d325ee6b3bf824fc0467c03b30e7fa8/typer-0.19.2-py3-none-any.whl", hash = "sha256:755e7e19670ffad8283db353267cb81ef252f595aa6834a0d1ca9312d9326cb9", size = 46748, upload-time = "2025-09-23T09:47:46.777Z" },
+    { url = "https://files.pythonhosted.org/packages/0b/eb/09e31d107a5d00eb281aa7e6635ca463e9bca86515944e399480eadb71f8/triton-3.5.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d5d3b3d480debf24eaa739623c9a42446b0b77f95593d30eb1f64cd2278cc1f0", size = 170333110, upload-time = "2025-10-13T16:37:49.588Z" },
+    { url = "https://files.pythonhosted.org/packages/3d/78/949a04391c21956c816523678f0e5fa308eb5b1e7622d88c4e4ef5fceca0/triton-3.5.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:f34bfa21c5b3a203c0f0eab28dcc1e49bd1f67d22724e77fb6665a659200a4ec", size = 170433488, upload-time = "2025-10-13T16:37:57.132Z" },
+    { url = "https://files.pythonhosted.org/packages/f5/3a/e991574f3102147b642e49637e0281e9bb7c4ba254edb2bab78247c85e01/triton-3.5.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:c9e71db82261c4ffa3921cd050cd5faa18322d2d405c30eb56084afaff3b0833", size = 170476535, upload-time = "2025-10-13T16:38:05.18Z" },
 ]
 
 [[package]]
@@ -5367,140 +5374,58 @@ wheels = [
 
 [[package]]
 name = "uvicorn"
-version = "0.37.0"
+version = "0.38.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "click", version = "8.3.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "click", version = "8.2.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "h11" },
     { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/71/57/1616c8274c3442d802621abf5deb230771c7a0fec9414cb6763900eb3868/uvicorn-0.37.0.tar.gz", hash = "sha256:4115c8add6d3fd536c8ee77f0e14a7fd2ebba939fed9b02583a97f80648f9e13", size = 80367, upload-time = "2025-09-23T13:33:47.486Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/85/cd/584a2ceb5532af99dd09e50919e3615ba99aa127e9850eafe5f31ddfdb9a/uvicorn-0.37.0-py3-none-any.whl", hash = "sha256:913b2b88672343739927ce381ff9e2ad62541f9f8289664fa1d1d3803fa2ce6c", size = 67976, upload-time = "2025-09-23T13:33:45.842Z" },
-]
-
-[package.optional-dependencies]
-standard = [
-    { name = "colorama", marker = "sys_platform == 'win32'" },
-    { name = "httptools" },
-    { name = "python-dotenv" },
-    { name = "pyyaml" },
-    { name = "uvloop", marker = "platform_python_implementation != 'PyPy' and sys_platform != 'cygwin' and sys_platform != 'win32'" },
-    { name = "watchfiles" },
-    { name = "websockets" },
-]
-
-[[package]]
-name = "uvloop"
-version = "0.21.0"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/af/c0/854216d09d33c543f12a44b393c402e89a920b1a0a7dc634c42de91b9cf6/uvloop-0.21.0.tar.gz", hash = "sha256:3bf12b0fda68447806a7ad847bfa591613177275d35b6724b1ee573faa3704e3", size = 2492741, upload-time = "2024-10-14T23:38:35.489Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/cb/ce/f06b84e2697fef4688ca63bdb2fdf113ca0a3be33f94488f2cadb690b0cf/uvicorn-0.38.0.tar.gz", hash = "sha256:fd97093bdd120a2609fc0d3afe931d4d4ad688b6e75f0f929fde1bc36fe0e91d", size = 80605, upload-time = "2025-10-18T13:46:44.63Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/f3/96/63695e0ebd7da6c741ccd4489b5947394435e198a1382349c17b1146bb97/uvloop-0.21.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f38b2e090258d051d68a5b14d1da7203a3c3677321cf32a95a6f4db4dd8b6f26", size = 3827735, upload-time = "2024-10-14T23:37:25.129Z" },
-    { url = "https://files.pythonhosted.org/packages/61/e0/f0f8ec84979068ffae132c58c79af1de9cceeb664076beea86d941af1a30/uvloop-0.21.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:87c43e0f13022b998eb9b973b5e97200c8b90823454d4bc06ab33829e09fb9bb", size = 3825126, upload-time = "2024-10-14T23:37:27.59Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/fe/5e94a977d058a54a19df95f12f7161ab6e323ad49f4dabc28822eb2df7ea/uvloop-0.21.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:10d66943def5fcb6e7b37310eb6b5639fd2ccbc38df1177262b0640c3ca68c1f", size = 3705789, upload-time = "2024-10-14T23:37:29.385Z" },
-    { url = "https://files.pythonhosted.org/packages/26/dd/c7179618e46092a77e036650c1f056041a028a35c4d76945089fcfc38af8/uvloop-0.21.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:67dd654b8ca23aed0a8e99010b4c34aca62f4b7fce88f39d452ed7622c94845c", size = 3800523, upload-time = "2024-10-14T23:37:32.048Z" },
-    { url = "https://files.pythonhosted.org/packages/ee/ea/0bfae1aceb82a503f358d8d2fa126ca9dbdb2ba9c7866974faec1cb5875c/uvloop-0.21.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b9fb766bb57b7388745d8bcc53a359b116b8a04c83a2288069809d2b3466c37e", size = 3960855, upload-time = "2024-10-14T23:37:37.683Z" },
-    { url = "https://files.pythonhosted.org/packages/8a/ca/0864176a649838b838f36d44bf31c451597ab363b60dc9e09c9630619d41/uvloop-0.21.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8a375441696e2eda1c43c44ccb66e04d61ceeffcd76e4929e527b7fa401b90fb", size = 3973185, upload-time = "2024-10-14T23:37:40.226Z" },
-    { url = "https://files.pythonhosted.org/packages/30/bf/08ad29979a936d63787ba47a540de2132169f140d54aa25bc8c3df3e67f4/uvloop-0.21.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:baa0e6291d91649c6ba4ed4b2f982f9fa165b5bbd50a9e203c416a2797bab3c6", size = 3820256, upload-time = "2024-10-14T23:37:42.839Z" },
-    { url = "https://files.pythonhosted.org/packages/da/e2/5cf6ef37e3daf2f06e651aae5ea108ad30df3cb269102678b61ebf1fdf42/uvloop-0.21.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4509360fcc4c3bd2c70d87573ad472de40c13387f5fda8cb58350a1d7475e58d", size = 3937323, upload-time = "2024-10-14T23:37:45.337Z" },
-    { url = "https://files.pythonhosted.org/packages/a6/ef/a02ec5da49909dbbfb1fd205a9a1ac4e88ea92dcae885e7c961847cd51e2/uvloop-0.21.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:baa4dcdbd9ae0a372f2167a207cd98c9f9a1ea1188a8a526431eef2f8116cc8d", size = 4580089, upload-time = "2024-10-14T23:37:51.703Z" },
-    { url = "https://files.pythonhosted.org/packages/06/a7/b4e6a19925c900be9f98bec0a75e6e8f79bb53bdeb891916609ab3958967/uvloop-0.21.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86975dca1c773a2c9864f4c52c5a55631038e387b47eaf56210f873887b6c8dc", size = 4693770, upload-time = "2024-10-14T23:37:54.122Z" },
-    { url = "https://files.pythonhosted.org/packages/ce/0c/f07435a18a4b94ce6bd0677d8319cd3de61f3a9eeb1e5f8ab4e8b5edfcb3/uvloop-0.21.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:461d9ae6660fbbafedd07559c6a2e57cd553b34b0065b6550685f6653a98c1cb", size = 4451321, upload-time = "2024-10-14T23:37:55.766Z" },
-    { url = "https://files.pythonhosted.org/packages/8f/eb/f7032be105877bcf924709c97b1bf3b90255b4ec251f9340cef912559f28/uvloop-0.21.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:183aef7c8730e54c9a3ee3227464daed66e37ba13040bb3f350bc2ddc040f22f", size = 4659022, upload-time = "2024-10-14T23:37:58.195Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/1a/9fbc2b1543d0df11f7aed1632f64bdf5ecc4053cf98cdc9edb91a65494f9/uvloop-0.21.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:53e420a3afe22cdcf2a0f4846e377d16e718bc70103d7088a4f7623567ba5fb0", size = 3829660, upload-time = "2024-10-14T23:38:27.905Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/c0/392e235e4100ae3b95b5c6dac77f82b529d2760942b1e7e0981e5d8e895d/uvloop-0.21.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:88cb67cdbc0e483da00af0b2c3cdad4b7c61ceb1ee0f33fe00e09c81e3a6cb75", size = 3827185, upload-time = "2024-10-14T23:38:29.458Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/24/a5da6aba58f99aed5255eca87d58d1760853e8302d390820cc29058408e3/uvloop-0.21.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:221f4f2a1f46032b403bf3be628011caf75428ee3cc204a22addf96f586b19fd", size = 3705833, upload-time = "2024-10-14T23:38:31.155Z" },
-    { url = "https://files.pythonhosted.org/packages/1a/5c/6ba221bb60f1e6474474102e17e38612ec7a06dc320e22b687ab563d877f/uvloop-0.21.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:2d1f581393673ce119355d56da84fe1dd9d2bb8b3d13ce792524e1607139feff", size = 3804696, upload-time = "2024-10-14T23:38:33.633Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/d9/d88e73ca598f4f6ff671fb5fde8a32925c2e08a637303a1d12883c7305fa/uvicorn-0.38.0-py3-none-any.whl", hash = "sha256:48c0afd214ceb59340075b4a052ea1ee91c16fbc2a9b1469cca0e54566977b02", size = 68109, upload-time = "2025-10-18T13:46:42.958Z" },
 ]
 
 [[package]]
 name = "virtualenv"
-version = "20.34.0"
+version = "20.35.4"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "distlib" },
-    { name = "filelock" },
-    { name = "platformdirs" },
+    { name = "filelock", version = "3.19.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "filelock", version = "3.20.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
+    { name = "platformdirs", version = "4.4.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "platformdirs", version = "4.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "typing-extensions", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/1c/14/37fcdba2808a6c615681cd216fecae00413c9dab44fb2e57805ecf3eaee3/virtualenv-20.34.0.tar.gz", hash = "sha256:44815b2c9dee7ed86e387b842a84f20b93f7f417f95886ca1996a72a4138eb1a", size = 6003808, upload-time = "2025-08-13T14:24:07.464Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/20/28/e6f1a6f655d620846bd9df527390ecc26b3805a0c5989048c210e22c5ca9/virtualenv-20.35.4.tar.gz", hash = "sha256:643d3914d73d3eeb0c552cbb12d7e82adf0e504dbf86a3182f8771a153a1971c", size = 6028799, upload-time = "2025-10-29T06:57:40.511Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/76/06/04c8e804f813cf972e3262f3f8584c232de64f0cde9f703b46cf53a45090/virtualenv-20.34.0-py3-none-any.whl", hash = "sha256:341f5afa7eee943e4984a9207c025feedd768baff6753cd660c857ceb3e36026", size = 5983279, upload-time = "2025-08-13T14:24:05.111Z" },
+    { url = "https://files.pythonhosted.org/packages/79/0c/c05523fa3181fdf0c9c52a6ba91a23fbf3246cc095f26f6516f9c60e6771/virtualenv-20.35.4-py3-none-any.whl", hash = "sha256:c21c9cede36c9753eeade68ba7d523529f228a403463376cf821eaae2b650f1b", size = 6005095, upload-time = "2025-10-29T06:57:37.598Z" },
 ]
 
 [[package]]
 name = "vllm"
-version = "0.11.0"
+version = "0.1.2"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "aiohttp" },
-    { name = "blake3" },
-    { name = "cachetools" },
-    { name = "cbor2" },
-    { name = "cloudpickle" },
-    { name = "compressed-tensors" },
-    { name = "depyf" },
-    { name = "diskcache" },
-    { name = "einops" },
-    { name = "fastapi", extra = ["standard"] },
-    { name = "filelock" },
-    { name = "gguf" },
-    { name = "importlib-metadata", marker = "python_full_version < '3.10'" },
-    { name = "lark" },
-    { name = "llguidance", marker = "platform_machine == 'aarch64' or platform_machine == 'arm64' or platform_machine == 'x86_64'" },
-    { name = "lm-format-enforcer" },
-    { name = "mistral-common", extra = ["audio", "image"] },
-    { name = "msgspec" },
+    { name = "fastapi" },
+    { name = "fschat" },
     { name = "ninja" },
-    { name = "numba", version = "0.60.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "numba", version = "0.61.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "openai" },
-    { name = "openai-harmony" },
-    { name = "opencv-python-headless" },
-    { name = "outlines-core" },
-    { name = "partial-json-parser" },
-    { name = "pillow" },
-    { name = "prometheus-client" },
-    { name = "prometheus-fastapi-instrumentator" },
-    { name = "protobuf" },
     { name = "psutil" },
-    { name = "py-cpuinfo" },
-    { name = "pybase64" },
     { name = "pydantic" },
-    { name = "python-json-logger" },
-    { name = "pyyaml" },
-    { name = "pyzmq" },
-    { name = "ray", extra = ["cgraph"] },
-    { name = "regex" },
-    { name = "requests" },
-    { name = "scipy", version = "1.13.1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
-    { name = "scipy", version = "1.15.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version == '3.10.*'" },
-    { name = "scipy", version = "1.16.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.11'" },
+    { name = "ray" },
     { name = "sentencepiece" },
-    { name = "setproctitle" },
-    { name = "setuptools", marker = "python_full_version >= '3.12'" },
-    { name = "six", marker = "python_full_version >= '3.12'" },
-    { name = "tiktoken" },
-    { name = "tokenizers" },
-    { name = "torch" },
-    { name = "torchaudio" },
-    { name = "torchvision" },
-    { name = "tqdm" },
+    { name = "torch", version = "2.8.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "torch", version = "2.9.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "transformers" },
-    { name = "typing-extensions" },
-    { name = "watchfiles" },
-    { name = "xformers", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
-    { name = "xgrammar", marker = "platform_machine == 'aarch64' or platform_machine == 'arm64' or platform_machine == 'x86_64'" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/82/5a/36d2351206f4d8d871b10780f874d03957985e08298d430cc837723e07af/vllm-0.11.0.tar.gz", hash = "sha256:f435a64c24e9c4178d657a76f8edd8548ddc444012f7d06a9f79ac3a6392bfae", size = 10822208, upload-time = "2025-10-04T01:39:57.798Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/47/33/d19e0763c34392ec956534536fa837c060495bfff31ed83452135ea7608d/vllm-0.11.0-cp38-abi3-manylinux1_x86_64.whl", hash = "sha256:3861c75ff2b12e24f6d179ff5c084d791b42ded8675d76c8706697c79f68cd62", size = 438217982, upload-time = "2025-10-04T01:39:32.382Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/bf/973444bb959fc7acbbeb3d226bd4d135dcd49b6af174b29aab1b50e2d710/vllm-0.11.0-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:52369c9ee949944354bdc7afc88ded2d1ed02b098bf90db06cf80098a19787b7", size = 401003969, upload-time = "2025-10-04T01:39:50.251Z" },
+    { name = "uvicorn" },
+    { name = "xformers", version = "0.0.32.post2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "xformers", version = "0.0.33.post1", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
+sdist = { url = "https://files.pythonhosted.org/packages/f9/ed/1e4625f290e24c1381aeeb66620b32755b17f2894a4227fb49fb20fa0ec9/vllm-0.1.2.tar.gz", hash = "sha256:6fff59a918ec822c3d7a9060c3af5604a1bbf26ec4db15656678768019aefad7", size = 93849, upload-time = "2023-07-05T04:53:21.847Z" }
 
 [[package]]
 name = "watchdog"
@@ -5537,52 +5462,23 @@ wheels = [
 ]
 
 [[package]]
-name = "watchfiles"
-version = "1.1.0"
+name = "wavedrom"
+version = "2.0.3.post3"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "anyio" },
+    { name = "pyyaml" },
+    { name = "six" },
+    { name = "svgwrite" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/2a/9a/d451fcc97d029f5812e898fd30a53fd8c15c7bbd058fd75cfc6beb9bd761/watchfiles-1.1.0.tar.gz", hash = "sha256:693ed7ec72cbfcee399e92c895362b6e66d63dac6b91e2c11ae03d10d503e575", size = 94406, upload-time = "2025-06-15T19:06:59.42Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/06/be/503514656d0555ec2195f60d810eca29b938772e9bfb112d5cd5ad6f6a9e/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8412eacef34cae2836d891836a7fff7b754d6bcac61f6c12ba5ca9bc7e427b68", size = 450739, upload-time = "2025-06-15T19:04:54.203Z" },
-    { url = "https://files.pythonhosted.org/packages/4e/0d/a05dd9e5f136cdc29751816d0890d084ab99f8c17b86f25697288ca09bc7/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:df670918eb7dd719642e05979fc84704af913d563fd17ed636f7c4783003fdcc", size = 458106, upload-time = "2025-06-15T19:04:55.607Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/fa/9cd16e4dfdb831072b7ac39e7bea986e52128526251038eb481effe9f48e/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d7642b9bc4827b5518ebdb3b82698ada8c14c7661ddec5fe719f3e56ccd13c97", size = 484264, upload-time = "2025-06-15T19:04:57.009Z" },
-    { url = "https://files.pythonhosted.org/packages/32/04/1da8a637c7e2b70e750a0308e9c8e662ada0cca46211fa9ef24a23937e0b/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:199207b2d3eeaeb80ef4411875a6243d9ad8bc35b07fc42daa6b801cc39cc41c", size = 597612, upload-time = "2025-06-15T19:04:58.409Z" },
-    { url = "https://files.pythonhosted.org/packages/30/01/109f2762e968d3e58c95731a206e5d7d2a7abaed4299dd8a94597250153c/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a479466da6db5c1e8754caee6c262cd373e6e6c363172d74394f4bff3d84d7b5", size = 477242, upload-time = "2025-06-15T19:04:59.786Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/b8/46f58cf4969d3b7bc3ca35a98e739fa4085b0657a1540ccc29a1a0bc016f/watchfiles-1.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:935f9edd022ec13e447e5723a7d14456c8af254544cefbc533f6dd276c9aa0d9", size = 453148, upload-time = "2025-06-15T19:05:01.103Z" },
-    { url = "https://files.pythonhosted.org/packages/a5/cd/8267594263b1770f1eb76914940d7b2d03ee55eca212302329608208e061/watchfiles-1.1.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:8076a5769d6bdf5f673a19d51da05fc79e2bbf25e9fe755c47595785c06a8c72", size = 626574, upload-time = "2025-06-15T19:05:02.582Z" },
-    { url = "https://files.pythonhosted.org/packages/a1/2f/7f2722e85899bed337cba715723e19185e288ef361360718973f891805be/watchfiles-1.1.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:86b1e28d4c37e89220e924305cd9f82866bb0ace666943a6e4196c5df4d58dcc", size = 624378, upload-time = "2025-06-15T19:05:03.719Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/a2/8afa359ff52e99af1632f90cbf359da46184207e893a5f179301b0c8d6df/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cb45350fd1dc75cd68d3d72c47f5b513cb0578da716df5fba02fff31c69d5f2d", size = 450792, upload-time = "2025-06-15T19:05:09.869Z" },
-    { url = "https://files.pythonhosted.org/packages/1d/bf/7446b401667f5c64972a57a0233be1104157fc3abf72c4ef2666c1bd09b2/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:11ee4444250fcbeb47459a877e5e80ed994ce8e8d20283857fc128be1715dac7", size = 458196, upload-time = "2025-06-15T19:05:11.91Z" },
-    { url = "https://files.pythonhosted.org/packages/58/2f/501ddbdfa3fa874ea5597c77eeea3d413579c29af26c1091b08d0c792280/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bda8136e6a80bdea23e5e74e09df0362744d24ffb8cd59c4a95a6ce3d142f79c", size = 484788, upload-time = "2025-06-15T19:05:13.373Z" },
-    { url = "https://files.pythonhosted.org/packages/61/1e/9c18eb2eb5c953c96bc0e5f626f0e53cfef4bd19bd50d71d1a049c63a575/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:b915daeb2d8c1f5cee4b970f2e2c988ce6514aace3c9296e58dd64dc9aa5d575", size = 597879, upload-time = "2025-06-15T19:05:14.725Z" },
-    { url = "https://files.pythonhosted.org/packages/8b/6c/1467402e5185d89388b4486745af1e0325007af0017c3384cc786fff0542/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:ed8fc66786de8d0376f9f913c09e963c66e90ced9aa11997f93bdb30f7c872a8", size = 477447, upload-time = "2025-06-15T19:05:15.775Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/a1/ec0a606bde4853d6c4a578f9391eeb3684a9aea736a8eb217e3e00aa89a1/watchfiles-1.1.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fe4371595edf78c41ef8ac8df20df3943e13defd0efcb732b2e393b5a8a7a71f", size = 453145, upload-time = "2025-06-15T19:05:17.17Z" },
-    { url = "https://files.pythonhosted.org/packages/90/b9/ef6f0c247a6a35d689fc970dc7f6734f9257451aefb30def5d100d6246a5/watchfiles-1.1.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:b7c5f6fe273291f4d414d55b2c80d33c457b8a42677ad14b4b47ff025d0893e4", size = 626539, upload-time = "2025-06-15T19:05:18.557Z" },
-    { url = "https://files.pythonhosted.org/packages/34/44/6ffda5537085106ff5aaa762b0d130ac6c75a08015dd1621376f708c94de/watchfiles-1.1.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:7738027989881e70e3723c75921f1efa45225084228788fc59ea8c6d732eb30d", size = 624472, upload-time = "2025-06-15T19:05:19.588Z" },
-    { url = "https://files.pythonhosted.org/packages/86/50/dee79968566c03190677c26f7f47960aff738d32087087bdf63a5473e7df/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:130fc497b8ee68dce163e4254d9b0356411d1490e868bd8790028bc46c5cc297", size = 450939, upload-time = "2025-06-15T19:05:26.494Z" },
-    { url = "https://files.pythonhosted.org/packages/40/45/a7b56fb129700f3cfe2594a01aa38d033b92a33dddce86c8dfdfc1247b72/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:50a51a90610d0845a5931a780d8e51d7bd7f309ebc25132ba975aca016b576a0", size = 457270, upload-time = "2025-06-15T19:05:27.466Z" },
-    { url = "https://files.pythonhosted.org/packages/b5/c8/fa5ef9476b1d02dc6b5e258f515fcaaecf559037edf8b6feffcbc097c4b8/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:dc44678a72ac0910bac46fa6a0de6af9ba1355669b3dfaf1ce5f05ca7a74364e", size = 483370, upload-time = "2025-06-15T19:05:28.548Z" },
-    { url = "https://files.pythonhosted.org/packages/98/68/42cfcdd6533ec94f0a7aab83f759ec11280f70b11bfba0b0f885e298f9bd/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a543492513a93b001975ae283a51f4b67973662a375a403ae82f420d2c7205ee", size = 598654, upload-time = "2025-06-15T19:05:29.997Z" },
-    { url = "https://files.pythonhosted.org/packages/d3/74/b2a1544224118cc28df7e59008a929e711f9c68ce7d554e171b2dc531352/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:8ac164e20d17cc285f2b94dc31c384bc3aa3dd5e7490473b3db043dd70fbccfd", size = 478667, upload-time = "2025-06-15T19:05:31.172Z" },
-    { url = "https://files.pythonhosted.org/packages/8c/77/e3362fe308358dc9f8588102481e599c83e1b91c2ae843780a7ded939a35/watchfiles-1.1.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:f7590d5a455321e53857892ab8879dce62d1f4b04748769f5adf2e707afb9d4f", size = 452213, upload-time = "2025-06-15T19:05:32.299Z" },
-    { url = "https://files.pythonhosted.org/packages/6e/17/c8f1a36540c9a1558d4faf08e909399e8133599fa359bf52ec8fcee5be6f/watchfiles-1.1.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:37d3d3f7defb13f62ece99e9be912afe9dd8a0077b7c45ee5a57c74811d581a4", size = 626718, upload-time = "2025-06-15T19:05:33.415Z" },
-    { url = "https://files.pythonhosted.org/packages/26/45/fb599be38b4bd38032643783d7496a26a6f9ae05dea1a42e58229a20ac13/watchfiles-1.1.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:7080c4bb3efd70a07b1cc2df99a7aa51d98685be56be6038c3169199d0a1c69f", size = 623098, upload-time = "2025-06-15T19:05:34.534Z" },
-    { url = "https://files.pythonhosted.org/packages/79/e6/e10d5675af653b1b07d4156906858041149ca222edaf8995877f2605ba9e/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:aa0cc8365ab29487eb4f9979fd41b22549853389e22d5de3f134a6796e1b05a4", size = 451196, upload-time = "2025-06-15T19:06:32.435Z" },
-    { url = "https://files.pythonhosted.org/packages/f6/8a/facd6988100cd0f39e89f6c550af80edb28e3a529e1ee662e750663e6b36/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:90ebb429e933645f3da534c89b29b665e285048973b4d2b6946526888c3eb2c7", size = 458218, upload-time = "2025-06-15T19:06:33.503Z" },
-    { url = "https://files.pythonhosted.org/packages/90/26/34cbcbc4d0f2f8f9cc243007e65d741ae039f7a11ef8ec6e9cd25bee08d1/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c588c45da9b08ab3da81d08d7987dae6d2a3badd63acdb3e206a42dbfa7cb76f", size = 484851, upload-time = "2025-06-15T19:06:34.541Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/1f/f59faa9fc4b0e36dbcdd28a18c430416443b309d295d8b82e18192d120ad/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:7c55b0f9f68590115c25272b06e63f0824f03d4fc7d6deed43d8ad5660cabdbf", size = 599520, upload-time = "2025-06-15T19:06:35.785Z" },
-    { url = "https://files.pythonhosted.org/packages/83/72/3637abecb3bf590529f5154ca000924003e5f4bbb9619744feeaf6f0b70b/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cd17a1e489f02ce9117b0de3c0b1fab1c3e2eedc82311b299ee6b6faf6c23a29", size = 477956, upload-time = "2025-06-15T19:06:36.965Z" },
-    { url = "https://files.pythonhosted.org/packages/f7/f3/d14ffd9acc0c1bd4790378995e320981423263a5d70bd3929e2e0dc87fff/watchfiles-1.1.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:da71945c9ace018d8634822f16cbc2a78323ef6c876b1d34bbf5d5222fd6a72e", size = 453196, upload-time = "2025-06-15T19:06:38.024Z" },
-    { url = "https://files.pythonhosted.org/packages/7f/38/78ad77bd99e20c0fdc82262be571ef114fc0beef9b43db52adb939768c38/watchfiles-1.1.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:51556d5004887045dba3acdd1fdf61dddea2be0a7e18048b5e853dcd37149b86", size = 627479, upload-time = "2025-06-15T19:06:39.442Z" },
-    { url = "https://files.pythonhosted.org/packages/e6/cf/549d50a22fcc83f1017c6427b1c76c053233f91b526f4ad7a45971e70c0b/watchfiles-1.1.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:04e4ed5d1cd3eae68c89bcc1a485a109f39f2fd8de05f705e98af6b5f1861f1f", size = 624414, upload-time = "2025-06-15T19:06:40.859Z" },
-    { url = "https://files.pythonhosted.org/packages/70/9c/9a6a42e97f92eeed77c3485a43ea96723900aefa3ac739a8c73f4bff2cd7/watchfiles-1.1.0-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a2726d7bfd9f76158c84c10a409b77a320426540df8c35be172444394b17f7ea", size = 451528, upload-time = "2025-06-15T19:06:46.791Z" },
-    { url = "https://files.pythonhosted.org/packages/51/7b/98c7f4f7ce7ff03023cf971cd84a3ee3b790021ae7584ffffa0eb2554b96/watchfiles-1.1.0-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:df32d59cb9780f66d165a9a7a26f19df2c7d24e3bd58713108b41d0ff4f929c6", size = 454095, upload-time = "2025-06-15T19:06:48.211Z" },
-    { url = "https://files.pythonhosted.org/packages/b8/fa/12269467b2fc006f8fce4cd6c3acfa77491dd0777d2a747415f28ccc8c60/watchfiles-1.1.0-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f2bcdc54ea267fe72bfc7d83c041e4eb58d7d8dc6f578dfddb52f037ce62f432", size = 451584, upload-time = "2025-06-15T19:06:51.834Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/d3/254cea30f918f489db09d6a8435a7de7047f8cb68584477a515f160541d6/watchfiles-1.1.0-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:923fec6e5461c42bd7e3fd5ec37492c6f3468be0499bc0707b4bbbc16ac21792", size = 454009, upload-time = "2025-06-15T19:06:52.896Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/dc/6f324a6f32c5ab73b54311b5f393a79df34c1584b8d2404cf7e6d780aa5d/watchfiles-1.1.0-pp39-pypy39_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:54062ef956807ba806559b3c3d52105ae1827a0d4ab47b621b31132b6b7e2866", size = 451787, upload-time = "2025-06-15T19:06:56.998Z" },
-    { url = "https://files.pythonhosted.org/packages/45/5d/1d02ef4caa4ec02389e72d5594cdf9c67f1800a7c380baa55063c30c6598/watchfiles-1.1.0-pp39-pypy39_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7a7bd57a1bb02f9d5c398c0c1675384e7ab1dd39da0ca50b7f09af45fa435277", size = 454272, upload-time = "2025-06-15T19:06:58.055Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/be/71/6739e3abac630540aaeaaece4584c39f88b5f8658ce6ca517efec455e3de/wavedrom-2.0.3.post3.tar.gz", hash = "sha256:327b4d5dca593c81257c202fea516f7a908747fb11527c359f034f5b7af7f47b", size = 137737, upload-time = "2022-05-27T11:53:09.675Z" }
+
+[[package]]
+name = "wcwidth"
+version = "0.2.14"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/24/30/6b0809f4510673dc723187aeaf24c7f5459922d01e2f794277a3dfb90345/wcwidth-0.2.14.tar.gz", hash = "sha256:4d478375d31bc5395a3c55c40ccdf3354688364cd61c4f6adacaa9215d0b3605", size = 102293, upload-time = "2025-09-22T16:29:53.023Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/af/b5/123f13c975e9f27ab9c0770f514345bd406d0e8d3b7a0723af9d43f710af/wcwidth-0.2.14-py2.py3-none-any.whl", hash = "sha256:a7bb560c8aee30f9957e5f9895805edd20602f2d7f720186dfd906e82b4982e1", size = 37286, upload-time = "2025-09-22T16:29:51.641Z" },
 ]
 
 [[package]]
@@ -5673,70 +5569,98 @@ wheels = [
 
 [[package]]
 name = "wrapt"
-version = "1.17.3"
-source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/95/8f/aeb76c5b46e273670962298c23e7ddde79916cb74db802131d49a85e4b7d/wrapt-1.17.3.tar.gz", hash = "sha256:f66eb08feaa410fe4eebd17f2a2c8e2e46d3476e9f8c783daa8e09e0faa666d0", size = 55547, upload-time = "2025-08-12T05:53:21.714Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/3f/23/bb82321b86411eb51e5a5db3fb8f8032fd30bd7c2d74bfe936136b2fa1d6/wrapt-1.17.3-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:88bbae4d40d5a46142e70d58bf664a89b6b4befaea7b2ecc14e03cedb8e06c04", size = 53482, upload-time = "2025-08-12T05:51:44.467Z" },
-    { url = "https://files.pythonhosted.org/packages/45/69/f3c47642b79485a30a59c63f6d739ed779fb4cc8323205d047d741d55220/wrapt-1.17.3-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:e6b13af258d6a9ad602d57d889f83b9d5543acd471eee12eb51f5b01f8eb1bc2", size = 38676, upload-time = "2025-08-12T05:51:32.636Z" },
-    { url = "https://files.pythonhosted.org/packages/d1/71/e7e7f5670c1eafd9e990438e69d8fb46fa91a50785332e06b560c869454f/wrapt-1.17.3-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:fd341868a4b6714a5962c1af0bd44f7c404ef78720c7de4892901e540417111c", size = 38957, upload-time = "2025-08-12T05:51:54.655Z" },
-    { url = "https://files.pythonhosted.org/packages/de/17/9f8f86755c191d6779d7ddead1a53c7a8aa18bccb7cea8e7e72dfa6a8a09/wrapt-1.17.3-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:f9b2601381be482f70e5d1051a5965c25fb3625455a2bf520b5a077b22afb775", size = 81975, upload-time = "2025-08-12T05:52:30.109Z" },
-    { url = "https://files.pythonhosted.org/packages/f2/15/dd576273491f9f43dd09fce517f6c2ce6eb4fe21681726068db0d0467096/wrapt-1.17.3-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:343e44b2a8e60e06a7e0d29c1671a0d9951f59174f3709962b5143f60a2a98bd", size = 83149, upload-time = "2025-08-12T05:52:09.316Z" },
-    { url = "https://files.pythonhosted.org/packages/0c/c4/5eb4ce0d4814521fee7aa806264bf7a114e748ad05110441cd5b8a5c744b/wrapt-1.17.3-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:33486899acd2d7d3066156b03465b949da3fd41a5da6e394ec49d271baefcf05", size = 82209, upload-time = "2025-08-12T05:52:10.331Z" },
-    { url = "https://files.pythonhosted.org/packages/31/4b/819e9e0eb5c8dc86f60dfc42aa4e2c0d6c3db8732bce93cc752e604bb5f5/wrapt-1.17.3-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:e6f40a8aa5a92f150bdb3e1c44b7e98fb7113955b2e5394122fa5532fec4b418", size = 81551, upload-time = "2025-08-12T05:52:31.137Z" },
-    { url = "https://files.pythonhosted.org/packages/f8/83/ed6baf89ba3a56694700139698cf703aac9f0f9eb03dab92f57551bd5385/wrapt-1.17.3-cp310-cp310-win32.whl", hash = "sha256:a36692b8491d30a8c75f1dfee65bef119d6f39ea84ee04d9f9311f83c5ad9390", size = 36464, upload-time = "2025-08-12T05:53:01.204Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/90/ee61d36862340ad7e9d15a02529df6b948676b9a5829fd5e16640156627d/wrapt-1.17.3-cp310-cp310-win_amd64.whl", hash = "sha256:afd964fd43b10c12213574db492cb8f73b2f0826c8df07a68288f8f19af2ebe6", size = 38748, upload-time = "2025-08-12T05:53:00.209Z" },
-    { url = "https://files.pythonhosted.org/packages/bd/c3/cefe0bd330d389c9983ced15d326f45373f4073c9f4a8c2f99b50bfea329/wrapt-1.17.3-cp310-cp310-win_arm64.whl", hash = "sha256:af338aa93554be859173c39c85243970dc6a289fa907402289eeae7543e1ae18", size = 36810, upload-time = "2025-08-12T05:52:51.906Z" },
-    { url = "https://files.pythonhosted.org/packages/52/db/00e2a219213856074a213503fdac0511203dceefff26e1daa15250cc01a0/wrapt-1.17.3-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:273a736c4645e63ac582c60a56b0acb529ef07f78e08dc6bfadf6a46b19c0da7", size = 53482, upload-time = "2025-08-12T05:51:45.79Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/30/ca3c4a5eba478408572096fe9ce36e6e915994dd26a4e9e98b4f729c06d9/wrapt-1.17.3-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:5531d911795e3f935a9c23eb1c8c03c211661a5060aab167065896bbf62a5f85", size = 38674, upload-time = "2025-08-12T05:51:34.629Z" },
-    { url = "https://files.pythonhosted.org/packages/31/25/3e8cc2c46b5329c5957cec959cb76a10718e1a513309c31399a4dad07eb3/wrapt-1.17.3-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:0610b46293c59a3adbae3dee552b648b984176f8562ee0dba099a56cfbe4df1f", size = 38959, upload-time = "2025-08-12T05:51:56.074Z" },
-    { url = "https://files.pythonhosted.org/packages/5d/8f/a32a99fc03e4b37e31b57cb9cefc65050ea08147a8ce12f288616b05ef54/wrapt-1.17.3-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:b32888aad8b6e68f83a8fdccbf3165f5469702a7544472bdf41f582970ed3311", size = 82376, upload-time = "2025-08-12T05:52:32.134Z" },
-    { url = "https://files.pythonhosted.org/packages/31/57/4930cb8d9d70d59c27ee1332a318c20291749b4fba31f113c2f8ac49a72e/wrapt-1.17.3-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:8cccf4f81371f257440c88faed6b74f1053eef90807b77e31ca057b2db74edb1", size = 83604, upload-time = "2025-08-12T05:52:11.663Z" },
-    { url = "https://files.pythonhosted.org/packages/a8/f3/1afd48de81d63dd66e01b263a6fbb86e1b5053b419b9b33d13e1f6d0f7d0/wrapt-1.17.3-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:d8a210b158a34164de8bb68b0e7780041a903d7b00c87e906fb69928bf7890d5", size = 82782, upload-time = "2025-08-12T05:52:12.626Z" },
-    { url = "https://files.pythonhosted.org/packages/1e/d7/4ad5327612173b144998232f98a85bb24b60c352afb73bc48e3e0d2bdc4e/wrapt-1.17.3-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:79573c24a46ce11aab457b472efd8d125e5a51da2d1d24387666cd85f54c05b2", size = 82076, upload-time = "2025-08-12T05:52:33.168Z" },
-    { url = "https://files.pythonhosted.org/packages/bb/59/e0adfc831674a65694f18ea6dc821f9fcb9ec82c2ce7e3d73a88ba2e8718/wrapt-1.17.3-cp311-cp311-win32.whl", hash = "sha256:c31eebe420a9a5d2887b13000b043ff6ca27c452a9a22fa71f35f118e8d4bf89", size = 36457, upload-time = "2025-08-12T05:53:03.936Z" },
-    { url = "https://files.pythonhosted.org/packages/83/88/16b7231ba49861b6f75fc309b11012ede4d6b0a9c90969d9e0db8d991aeb/wrapt-1.17.3-cp311-cp311-win_amd64.whl", hash = "sha256:0b1831115c97f0663cb77aa27d381237e73ad4f721391a9bfb2fe8bc25fa6e77", size = 38745, upload-time = "2025-08-12T05:53:02.885Z" },
-    { url = "https://files.pythonhosted.org/packages/9a/1e/c4d4f3398ec073012c51d1c8d87f715f56765444e1a4b11e5180577b7e6e/wrapt-1.17.3-cp311-cp311-win_arm64.whl", hash = "sha256:5a7b3c1ee8265eb4c8f1b7d29943f195c00673f5ab60c192eba2d4a7eae5f46a", size = 36806, upload-time = "2025-08-12T05:52:53.368Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/41/cad1aba93e752f1f9268c77270da3c469883d56e2798e7df6240dcb2287b/wrapt-1.17.3-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:ab232e7fdb44cdfbf55fc3afa31bcdb0d8980b9b95c38b6405df2acb672af0e0", size = 53998, upload-time = "2025-08-12T05:51:47.138Z" },
-    { url = "https://files.pythonhosted.org/packages/60/f8/096a7cc13097a1869fe44efe68dace40d2a16ecb853141394047f0780b96/wrapt-1.17.3-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:9baa544e6acc91130e926e8c802a17f3b16fbea0fd441b5a60f5cf2cc5c3deba", size = 39020, upload-time = "2025-08-12T05:51:35.906Z" },
-    { url = "https://files.pythonhosted.org/packages/33/df/bdf864b8997aab4febb96a9ae5c124f700a5abd9b5e13d2a3214ec4be705/wrapt-1.17.3-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:6b538e31eca1a7ea4605e44f81a48aa24c4632a277431a6ed3f328835901f4fd", size = 39098, upload-time = "2025-08-12T05:51:57.474Z" },
-    { url = "https://files.pythonhosted.org/packages/9f/81/5d931d78d0eb732b95dc3ddaeeb71c8bb572fb01356e9133916cd729ecdd/wrapt-1.17.3-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:042ec3bb8f319c147b1301f2393bc19dba6e176b7da446853406d041c36c7828", size = 88036, upload-time = "2025-08-12T05:52:34.784Z" },
-    { url = "https://files.pythonhosted.org/packages/ca/38/2e1785df03b3d72d34fc6252d91d9d12dc27a5c89caef3335a1bbb8908ca/wrapt-1.17.3-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3af60380ba0b7b5aeb329bc4e402acd25bd877e98b3727b0135cb5c2efdaefe9", size = 88156, upload-time = "2025-08-12T05:52:13.599Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/8b/48cdb60fe0603e34e05cffda0b2a4adab81fd43718e11111a4b0100fd7c1/wrapt-1.17.3-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:0b02e424deef65c9f7326d8c19220a2c9040c51dc165cddb732f16198c168396", size = 87102, upload-time = "2025-08-12T05:52:14.56Z" },
-    { url = "https://files.pythonhosted.org/packages/3c/51/d81abca783b58f40a154f1b2c56db1d2d9e0d04fa2d4224e357529f57a57/wrapt-1.17.3-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:74afa28374a3c3a11b3b5e5fca0ae03bef8450d6aa3ab3a1e2c30e3a75d023dc", size = 87732, upload-time = "2025-08-12T05:52:36.165Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/b1/43b286ca1392a006d5336412d41663eeef1ad57485f3e52c767376ba7e5a/wrapt-1.17.3-cp312-cp312-win32.whl", hash = "sha256:4da9f45279fff3543c371d5ababc57a0384f70be244de7759c85a7f989cb4ebe", size = 36705, upload-time = "2025-08-12T05:53:07.123Z" },
-    { url = "https://files.pythonhosted.org/packages/28/de/49493f962bd3c586ab4b88066e967aa2e0703d6ef2c43aa28cb83bf7b507/wrapt-1.17.3-cp312-cp312-win_amd64.whl", hash = "sha256:e71d5c6ebac14875668a1e90baf2ea0ef5b7ac7918355850c0908ae82bcb297c", size = 38877, upload-time = "2025-08-12T05:53:05.436Z" },
-    { url = "https://files.pythonhosted.org/packages/f1/48/0f7102fe9cb1e8a5a77f80d4f0956d62d97034bbe88d33e94699f99d181d/wrapt-1.17.3-cp312-cp312-win_arm64.whl", hash = "sha256:604d076c55e2fdd4c1c03d06dc1a31b95130010517b5019db15365ec4a405fc6", size = 36885, upload-time = "2025-08-12T05:52:54.367Z" },
-    { url = "https://files.pythonhosted.org/packages/41/be/be9b3b0a461ee3e30278706f3f3759b9b69afeedef7fe686036286c04ac6/wrapt-1.17.3-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:30ce38e66630599e1193798285706903110d4f057aab3168a34b7fdc85569afc", size = 53485, upload-time = "2025-08-12T05:51:53.11Z" },
-    { url = "https://files.pythonhosted.org/packages/b3/a8/8f61d6b8f526efc8c10e12bf80b4206099fea78ade70427846a37bc9cbea/wrapt-1.17.3-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:65d1d00fbfb3ea5f20add88bbc0f815150dbbde3b026e6c24759466c8b5a9ef9", size = 38675, upload-time = "2025-08-12T05:51:42.885Z" },
-    { url = "https://files.pythonhosted.org/packages/48/f1/23950c29a25637b74b322f9e425a17cc01a478f6afb35138ecb697f9558d/wrapt-1.17.3-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:a7c06742645f914f26c7f1fa47b8bc4c91d222f76ee20116c43d5ef0912bba2d", size = 38956, upload-time = "2025-08-12T05:52:03.149Z" },
-    { url = "https://files.pythonhosted.org/packages/43/46/dd0791943613885f62619f18ee6107e6133237a6b6ed8a9ecfac339d0b4f/wrapt-1.17.3-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7e18f01b0c3e4a07fe6dfdb00e29049ba17eadbc5e7609a2a3a4af83ab7d710a", size = 81745, upload-time = "2025-08-12T05:52:49.62Z" },
-    { url = "https://files.pythonhosted.org/packages/dd/ec/bb2d19bd1a614cc4f438abac13ae26c57186197920432d2a915183b15a8b/wrapt-1.17.3-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0f5f51a6466667a5a356e6381d362d259125b57f059103dd9fdc8c0cf1d14139", size = 82833, upload-time = "2025-08-12T05:52:27.738Z" },
-    { url = "https://files.pythonhosted.org/packages/8d/eb/66579aea6ad36f07617fedca8e282e49c7c9bab64c63b446cfe4f7f47a49/wrapt-1.17.3-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:59923aa12d0157f6b82d686c3fd8e1166fa8cdfb3e17b42ce3b6147ff81528df", size = 81889, upload-time = "2025-08-12T05:52:29.023Z" },
-    { url = "https://files.pythonhosted.org/packages/04/9c/a56b5ac0e2473bdc3fb11b22dd69ff423154d63861cf77911cdde5e38fd2/wrapt-1.17.3-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:46acc57b331e0b3bcb3e1ca3b421d65637915cfcd65eb783cb2f78a511193f9b", size = 81344, upload-time = "2025-08-12T05:52:50.869Z" },
-    { url = "https://files.pythonhosted.org/packages/93/4c/9bd735c42641d81cb58d7bfb142c58f95c833962d15113026705add41a07/wrapt-1.17.3-cp39-cp39-win32.whl", hash = "sha256:3e62d15d3cfa26e3d0788094de7b64efa75f3a53875cdbccdf78547aed547a81", size = 36462, upload-time = "2025-08-12T05:53:19.623Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/ea/0b72f29cb5ebc16eb55c57dc0c98e5de76fc97f435fd407f7d409459c0a6/wrapt-1.17.3-cp39-cp39-win_amd64.whl", hash = "sha256:1f23fa283f51c890eda8e34e4937079114c74b4c81d2b2f1f1d94948f5cc3d7f", size = 38740, upload-time = "2025-08-12T05:53:18.271Z" },
-    { url = "https://files.pythonhosted.org/packages/c3/8b/9eae65fb92321e38dbfec7719b87d840a4b92fde83fd1bbf238c5488d055/wrapt-1.17.3-cp39-cp39-win_arm64.whl", hash = "sha256:24c2ed34dc222ed754247a2702b1e1e89fdbaa4016f324b4b8f1a802d4ffe87f", size = 36806, upload-time = "2025-08-12T05:52:58.765Z" },
-    { url = "https://files.pythonhosted.org/packages/1f/f6/a933bd70f98e9cf3e08167fc5cd7aaaca49147e48411c0bd5ae701bb2194/wrapt-1.17.3-py3-none-any.whl", hash = "sha256:7171ae35d2c33d326ac19dd8facb1e82e5fd04ef8c6c0e394d7af55a55051c22", size = 23591, upload-time = "2025-08-12T05:53:20.674Z" },
+version = "2.0.1"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/49/2a/6de8a50cb435b7f42c46126cf1a54b2aab81784e74c8595c8e025e8f36d3/wrapt-2.0.1.tar.gz", hash = "sha256:9c9c635e78497cacb81e84f8b11b23e0aacac7a136e73b8e5b2109a1d9fc468f", size = 82040, upload-time = "2025-11-07T00:45:33.312Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/61/0d/12d8c803ed2ce4e5e7d5b9f5f602721f9dfef82c95959f3ce97fa584bb5c/wrapt-2.0.1-cp310-cp310-macosx_10_9_universal2.whl", hash = "sha256:64b103acdaa53b7caf409e8d45d39a8442fe6dcfec6ba3f3d141e0cc2b5b4dbd", size = 77481, upload-time = "2025-11-07T00:43:11.103Z" },
+    { url = "https://files.pythonhosted.org/packages/05/3e/4364ebe221ebf2a44d9fc8695a19324692f7dd2795e64bd59090856ebf12/wrapt-2.0.1-cp310-cp310-macosx_10_9_x86_64.whl", hash = "sha256:91bcc576260a274b169c3098e9a3519fb01f2989f6d3d386ef9cbf8653de1374", size = 60692, upload-time = "2025-11-07T00:43:13.697Z" },
+    { url = "https://files.pythonhosted.org/packages/1f/ff/ae2a210022b521f86a8ddcdd6058d137c051003812b0388a5e9a03d3fe10/wrapt-2.0.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:ab594f346517010050126fcd822697b25a7031d815bb4fbc238ccbe568216489", size = 61574, upload-time = "2025-11-07T00:43:14.967Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/93/5cf92edd99617095592af919cb81d4bff61c5dbbb70d3c92099425a8ec34/wrapt-2.0.1-cp310-cp310-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:36982b26f190f4d737f04a492a68accbfc6fa042c3f42326fdfbb6c5b7a20a31", size = 113688, upload-time = "2025-11-07T00:43:18.275Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/0a/e38fc0cee1f146c9fb266d8ef96ca39fb14a9eef165383004019aa53f88a/wrapt-2.0.1-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:23097ed8bc4c93b7bf36fa2113c6c733c976316ce0ee2c816f64ca06102034ef", size = 115698, upload-time = "2025-11-07T00:43:19.407Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/85/bef44ea018b3925fb0bcbe9112715f665e4d5309bd945191da814c314fd1/wrapt-2.0.1-cp310-cp310-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:8bacfe6e001749a3b64db47bcf0341da757c95959f592823a93931a422395013", size = 112096, upload-time = "2025-11-07T00:43:16.5Z" },
+    { url = "https://files.pythonhosted.org/packages/7c/0b/733a2376e413117e497aa1a5b1b78e8f3a28c0e9537d26569f67d724c7c5/wrapt-2.0.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:8ec3303e8a81932171f455f792f8df500fc1a09f20069e5c16bd7049ab4e8e38", size = 114878, upload-time = "2025-11-07T00:43:20.81Z" },
+    { url = "https://files.pythonhosted.org/packages/da/03/d81dcb21bbf678fcda656495792b059f9d56677d119ca022169a12542bd0/wrapt-2.0.1-cp310-cp310-musllinux_1_2_riscv64.whl", hash = "sha256:3f373a4ab5dbc528a94334f9fe444395b23c2f5332adab9ff4ea82f5a9e33bc1", size = 111298, upload-time = "2025-11-07T00:43:22.229Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/d5/5e623040e8056e1108b787020d56b9be93dbbf083bf2324d42cde80f3a19/wrapt-2.0.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:f49027b0b9503bf6c8cdc297ca55006b80c2f5dd36cecc72c6835ab6e10e8a25", size = 113361, upload-time = "2025-11-07T00:43:24.301Z" },
+    { url = "https://files.pythonhosted.org/packages/a1/f3/de535ccecede6960e28c7b722e5744846258111d6c9f071aa7578ea37ad3/wrapt-2.0.1-cp310-cp310-win32.whl", hash = "sha256:8330b42d769965e96e01fa14034b28a2a7600fbf7e8f0cc90ebb36d492c993e4", size = 58035, upload-time = "2025-11-07T00:43:28.96Z" },
+    { url = "https://files.pythonhosted.org/packages/21/15/39d3ca5428a70032c2ec8b1f1c9d24c32e497e7ed81aed887a4998905fcc/wrapt-2.0.1-cp310-cp310-win_amd64.whl", hash = "sha256:1218573502a8235bb8a7ecaed12736213b22dcde9feab115fa2989d42b5ded45", size = 60383, upload-time = "2025-11-07T00:43:25.804Z" },
+    { url = "https://files.pythonhosted.org/packages/43/c2/dfd23754b7f7a4dce07e08f4309c4e10a40046a83e9ae1800f2e6b18d7c1/wrapt-2.0.1-cp310-cp310-win_arm64.whl", hash = "sha256:eda8e4ecd662d48c28bb86be9e837c13e45c58b8300e43ba3c9b4fa9900302f7", size = 58894, upload-time = "2025-11-07T00:43:27.074Z" },
+    { url = "https://files.pythonhosted.org/packages/98/60/553997acf3939079dab022e37b67b1904b5b0cc235503226898ba573b10c/wrapt-2.0.1-cp311-cp311-macosx_10_9_universal2.whl", hash = "sha256:0e17283f533a0d24d6e5429a7d11f250a58d28b4ae5186f8f47853e3e70d2590", size = 77480, upload-time = "2025-11-07T00:43:30.573Z" },
+    { url = "https://files.pythonhosted.org/packages/2d/50/e5b3d30895d77c52105c6d5cbf94d5b38e2a3dd4a53d22d246670da98f7c/wrapt-2.0.1-cp311-cp311-macosx_10_9_x86_64.whl", hash = "sha256:85df8d92158cb8f3965aecc27cf821461bb5f40b450b03facc5d9f0d4d6ddec6", size = 60690, upload-time = "2025-11-07T00:43:31.594Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/40/660b2898703e5cbbb43db10cdefcc294274458c3ca4c68637c2b99371507/wrapt-2.0.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:c1be685ac7700c966b8610ccc63c3187a72e33cab53526a27b2a285a662cd4f7", size = 61578, upload-time = "2025-11-07T00:43:32.918Z" },
+    { url = "https://files.pythonhosted.org/packages/5b/36/825b44c8a10556957bc0c1d84c7b29a40e05fcf1873b6c40aa9dbe0bd972/wrapt-2.0.1-cp311-cp311-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:df0b6d3b95932809c5b3fecc18fda0f1e07452d05e2662a0b35548985f256e28", size = 114115, upload-time = "2025-11-07T00:43:35.605Z" },
+    { url = "https://files.pythonhosted.org/packages/83/73/0a5d14bb1599677304d3c613a55457d34c344e9b60eda8a737c2ead7619e/wrapt-2.0.1-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4da7384b0e5d4cae05c97cd6f94faaf78cc8b0f791fc63af43436d98c4ab37bb", size = 116157, upload-time = "2025-11-07T00:43:37.058Z" },
+    { url = "https://files.pythonhosted.org/packages/01/22/1c158fe763dbf0a119f985d945711d288994fe5514c0646ebe0eb18b016d/wrapt-2.0.1-cp311-cp311-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:ec65a78fbd9d6f083a15d7613b2800d5663dbb6bb96003899c834beaa68b242c", size = 112535, upload-time = "2025-11-07T00:43:34.138Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/28/4f16861af67d6de4eae9927799b559c20ebdd4fe432e89ea7fe6fcd9d709/wrapt-2.0.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:7de3cc939be0e1174969f943f3b44e0d79b6f9a82198133a5b7fc6cc92882f16", size = 115404, upload-time = "2025-11-07T00:43:39.214Z" },
+    { url = "https://files.pythonhosted.org/packages/a0/8b/7960122e625fad908f189b59c4aae2d50916eb4098b0fb2819c5a177414f/wrapt-2.0.1-cp311-cp311-musllinux_1_2_riscv64.whl", hash = "sha256:fb1a5b72cbd751813adc02ef01ada0b0d05d3dcbc32976ce189a1279d80ad4a2", size = 111802, upload-time = "2025-11-07T00:43:40.476Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/73/7881eee5ac31132a713ab19a22c9e5f1f7365c8b1df50abba5d45b781312/wrapt-2.0.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:3fa272ca34332581e00bf7773e993d4f632594eb2d1b0b162a9038df0fd971dd", size = 113837, upload-time = "2025-11-07T00:43:42.921Z" },
+    { url = "https://files.pythonhosted.org/packages/45/00/9499a3d14e636d1f7089339f96c4409bbc7544d0889f12264efa25502ae8/wrapt-2.0.1-cp311-cp311-win32.whl", hash = "sha256:fc007fdf480c77301ab1afdbb6ab22a5deee8885f3b1ed7afcb7e5e84a0e27be", size = 58028, upload-time = "2025-11-07T00:43:47.369Z" },
+    { url = "https://files.pythonhosted.org/packages/70/5d/8f3d7eea52f22638748f74b102e38fdf88cb57d08ddeb7827c476a20b01b/wrapt-2.0.1-cp311-cp311-win_amd64.whl", hash = "sha256:47434236c396d04875180171ee1f3815ca1eada05e24a1ee99546320d54d1d1b", size = 60385, upload-time = "2025-11-07T00:43:44.34Z" },
+    { url = "https://files.pythonhosted.org/packages/14/e2/32195e57a8209003587bbbad44d5922f13e0ced2a493bb46ca882c5b123d/wrapt-2.0.1-cp311-cp311-win_arm64.whl", hash = "sha256:837e31620e06b16030b1d126ed78e9383815cbac914693f54926d816d35d8edf", size = 58893, upload-time = "2025-11-07T00:43:46.161Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/73/8cb252858dc8254baa0ce58ce382858e3a1cf616acebc497cb13374c95c6/wrapt-2.0.1-cp312-cp312-macosx_10_13_universal2.whl", hash = "sha256:1fdbb34da15450f2b1d735a0e969c24bdb8d8924892380126e2a293d9902078c", size = 78129, upload-time = "2025-11-07T00:43:48.852Z" },
+    { url = "https://files.pythonhosted.org/packages/19/42/44a0db2108526ee6e17a5ab72478061158f34b08b793df251d9fbb9a7eb4/wrapt-2.0.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:3d32794fe940b7000f0519904e247f902f0149edbe6316c710a8562fb6738841", size = 61205, upload-time = "2025-11-07T00:43:50.402Z" },
+    { url = "https://files.pythonhosted.org/packages/4d/8a/5b4b1e44b791c22046e90d9b175f9a7581a8cc7a0debbb930f81e6ae8e25/wrapt-2.0.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:386fb54d9cd903ee0012c09291336469eb7b244f7183d40dc3e86a16a4bace62", size = 61692, upload-time = "2025-11-07T00:43:51.678Z" },
+    { url = "https://files.pythonhosted.org/packages/11/53/3e794346c39f462bcf1f58ac0487ff9bdad02f9b6d5ee2dc84c72e0243b2/wrapt-2.0.1-cp312-cp312-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:7b219cb2182f230676308cdcacd428fa837987b89e4b7c5c9025088b8a6c9faf", size = 121492, upload-time = "2025-11-07T00:43:55.017Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/7e/10b7b0e8841e684c8ca76b462a9091c45d62e8f2de9c4b1390b690eadf16/wrapt-2.0.1-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:641e94e789b5f6b4822bb8d8ebbdfc10f4e4eae7756d648b717d980f657a9eb9", size = 123064, upload-time = "2025-11-07T00:43:56.323Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/d1/3c1e4321fc2f5ee7fd866b2d822aa89b84495f28676fd976c47327c5b6aa/wrapt-2.0.1-cp312-cp312-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:fe21b118b9f58859b5ebaa4b130dee18669df4bd111daad082b7beb8799ad16b", size = 117403, upload-time = "2025-11-07T00:43:53.258Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/b0/d2f0a413cf201c8c2466de08414a15420a25aa83f53e647b7255cc2fab5d/wrapt-2.0.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:17fb85fa4abc26a5184d93b3efd2dcc14deb4b09edcdb3535a536ad34f0b4dba", size = 121500, upload-time = "2025-11-07T00:43:57.468Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/45/bddb11d28ca39970a41ed48a26d210505120f925918592283369219f83cc/wrapt-2.0.1-cp312-cp312-musllinux_1_2_riscv64.whl", hash = "sha256:b89ef9223d665ab255ae42cc282d27d69704d94be0deffc8b9d919179a609684", size = 116299, upload-time = "2025-11-07T00:43:58.877Z" },
+    { url = "https://files.pythonhosted.org/packages/81/af/34ba6dd570ef7a534e7eec0c25e2615c355602c52aba59413411c025a0cb/wrapt-2.0.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:a453257f19c31b31ba593c30d997d6e5be39e3b5ad9148c2af5a7314061c63eb", size = 120622, upload-time = "2025-11-07T00:43:59.962Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/3e/693a13b4146646fb03254636f8bafd20c621955d27d65b15de07ab886187/wrapt-2.0.1-cp312-cp312-win32.whl", hash = "sha256:3e271346f01e9c8b1130a6a3b0e11908049fe5be2d365a5f402778049147e7e9", size = 58246, upload-time = "2025-11-07T00:44:03.169Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/36/715ec5076f925a6be95f37917b66ebbeaa1372d1862c2ccd7a751574b068/wrapt-2.0.1-cp312-cp312-win_amd64.whl", hash = "sha256:2da620b31a90cdefa9cd0c2b661882329e2e19d1d7b9b920189956b76c564d75", size = 60492, upload-time = "2025-11-07T00:44:01.027Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/3e/62451cd7d80f65cc125f2b426b25fbb6c514bf6f7011a0c3904fc8c8df90/wrapt-2.0.1-cp312-cp312-win_arm64.whl", hash = "sha256:aea9c7224c302bc8bfc892b908537f56c430802560e827b75ecbde81b604598b", size = 58987, upload-time = "2025-11-07T00:44:02.095Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/1f/5af0ae22368ec69067a577f9e07a0dd2619a1f63aabc2851263679942667/wrapt-2.0.1-cp39-cp39-macosx_10_9_universal2.whl", hash = "sha256:68424221a2dc00d634b54f92441914929c5ffb1c30b3b837343978343a3512a3", size = 77478, upload-time = "2025-11-07T00:45:16.65Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/b7/fd6b563aada859baabc55db6aa71b8afb4a3ceb8bc33d1053e4c7b5e0109/wrapt-2.0.1-cp39-cp39-macosx_10_9_x86_64.whl", hash = "sha256:6bd1a18f5a797fe740cb3d7a0e853a8ce6461cc62023b630caec80171a6b8097", size = 60687, upload-time = "2025-11-07T00:45:17.896Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/8c/9ededfff478af396bcd081076986904bdca336d9664d247094150c877dcb/wrapt-2.0.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:fb3a86e703868561c5cad155a15c36c716e1ab513b7065bd2ac8ed353c503333", size = 61563, upload-time = "2025-11-07T00:45:19.109Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/a7/d795a1aa2b6ab20ca21157fe03cbfc6aa7e870a88ac3b4ea189e2f6c79f0/wrapt-2.0.1-cp39-cp39-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:5dc1b852337c6792aa111ca8becff5bacf576bf4a0255b0f05eb749da6a1643e", size = 113395, upload-time = "2025-11-07T00:45:21.551Z" },
+    { url = "https://files.pythonhosted.org/packages/61/32/56cde2bbf95f2d5698a1850a765520aa86bc7ae0f95b8ec80b6f2e2049bb/wrapt-2.0.1-cp39-cp39-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c046781d422f0830de6329fa4b16796096f28a92c8aef3850674442cdcb87b7f", size = 115362, upload-time = "2025-11-07T00:45:22.809Z" },
+    { url = "https://files.pythonhosted.org/packages/cf/53/8d3cc433847c219212c133a3e8305bd087b386ef44442ff39189e8fa62ac/wrapt-2.0.1-cp39-cp39-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:f73f9f7a0ebd0db139253d27e5fc8d2866ceaeef19c30ab5d69dcbe35e1a6981", size = 111766, upload-time = "2025-11-07T00:45:20.294Z" },
+    { url = "https://files.pythonhosted.org/packages/b8/d3/14b50c2d0463c0dcef8f388cb1527ed7bbdf0972b9fd9976905f36c77ebf/wrapt-2.0.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:b667189cf8efe008f55bbda321890bef628a67ab4147ebf90d182f2dadc78790", size = 114560, upload-time = "2025-11-07T00:45:24.054Z" },
+    { url = "https://files.pythonhosted.org/packages/3a/b8/4f731ff178f77ae55385586de9ff4b4261e872cf2ced4875e6c976fbcb8b/wrapt-2.0.1-cp39-cp39-musllinux_1_2_riscv64.whl", hash = "sha256:a9a83618c4f0757557c077ef71d708ddd9847ed66b7cc63416632af70d3e2308", size = 110999, upload-time = "2025-11-07T00:45:25.596Z" },
+    { url = "https://files.pythonhosted.org/packages/fe/bb/5f1bb0f9ae9d12e19f1d71993d052082062603e83fe3e978377f918f054d/wrapt-2.0.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:1e9b121e9aeb15df416c2c960b8255a49d44b4038016ee17af03975992d03931", size = 113164, upload-time = "2025-11-07T00:45:26.8Z" },
+    { url = "https://files.pythonhosted.org/packages/ad/f6/f3a3c623d3065c7bf292ee0b73566236b562d5ed894891bd8e435762b618/wrapt-2.0.1-cp39-cp39-win32.whl", hash = "sha256:1f186e26ea0a55f809f232e92cc8556a0977e00183c3ebda039a807a42be1494", size = 58028, upload-time = "2025-11-07T00:45:30.943Z" },
+    { url = "https://files.pythonhosted.org/packages/24/78/647c609dfa18063a7fcd5c23f762dd006be401cc9206314d29c9b0b12078/wrapt-2.0.1-cp39-cp39-win_amd64.whl", hash = "sha256:bf4cb76f36be5de950ce13e22e7fdf462b35b04665a12b64f3ac5c1bbbcf3728", size = 60380, upload-time = "2025-11-07T00:45:28.341Z" },
+    { url = "https://files.pythonhosted.org/packages/07/90/0c14b241d18d80ddf4c847a5f52071e126e8a6a9e5a8a7952add8ef0d766/wrapt-2.0.1-cp39-cp39-win_arm64.whl", hash = "sha256:d6cc985b9c8b235bd933990cdbf0f891f8e010b65a3911f7a55179cd7b0fc57b", size = 58895, upload-time = "2025-11-07T00:45:29.527Z" },
+    { url = "https://files.pythonhosted.org/packages/15/d1/b51471c11592ff9c012bd3e2f7334a6ff2f42a7aed2caffcf0bdddc9cb89/wrapt-2.0.1-py3-none-any.whl", hash = "sha256:4d2ce1bf1a48c5277d7969259232b57645aae5686dba1eaeade39442277afbca", size = 44046, upload-time = "2025-11-07T00:45:32.116Z" },
 ]
 
 [[package]]
 name = "xformers"
-version = "0.0.32.post1"
+version = "0.0.32.post2"
 source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version < '3.10'",
+]
 dependencies = [
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "torch", version = "2.8.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/47/99/68d2e7a17b65a08180d6b120d1d293d5855a0c5999cb30a79fb466b49642/xformers-0.0.32.post2.tar.gz", hash = "sha256:9538be803969c6e1ca16a3ece921e472c24f79970b10be1087a389dcb66e412a", size = 12105990, upload-time = "2025-08-15T11:48:56.584Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ce/16/4ed2fbf7c20ba20565d9a1fc544c2bef39d409c8eca1949654c6ff9f8452/xformers-0.0.32.post2-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:3d1fefe0006eb00a730b3b197ff41aab085f1209b50419baab6152445f09e508", size = 117196673, upload-time = "2025-08-15T11:48:46.726Z" },
+]
+
+[[package]]
+name = "xformers"
+version = "0.0.33.post1"
+source = { registry = "https://pypi.org/simple" }
+resolution-markers = [
+    "python_full_version >= '3.12'",
+    "python_full_version == '3.11.*'",
+    "python_full_version == '3.10.*'",
+]
+dependencies = [
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
-    { name = "torch" },
+    { name = "torch", version = "2.9.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/6f/33/3b9c4d3d5b2da453d27de891df4ad653ac5795324961aa3a5c15b0353fe6/xformers-0.0.32.post1.tar.gz", hash = "sha256:1de84a45c497c8d92326986508d81f4b0a8c6be4d3d62a29b8ad6048a6ab51e1", size = 12106196, upload-time = "2025-08-14T18:07:45.486Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/6f/c1/cd0d6b89da38d8aa174e8eabf29530f8871daf53b886ec6b680ef9d3e71f/xformers-0.0.33.post1.tar.gz", hash = "sha256:e555258249b514ba117b3403523fe0bd7d3e92e930575f0e0dbf5f7db5b42677", size = 14784437, upload-time = "2025-11-13T20:16:14.793Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/6b/df/6817346f1a77278315d5fe1fc9f239ba3282ba36e8ab3256babd448dde62/xformers-0.0.32.post1-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:5f245b5555188da112070d8fefb6b7ae1ae47422856521d66c837e9d2352fbe4", size = 117199943, upload-time = "2025-08-14T18:07:34.78Z" },
+    { url = "https://files.pythonhosted.org/packages/39/94/3ad80d1070ddfb280c20a67dfbc094a93579a02910ef41f20631a9b566fe/xformers-0.0.33.post1-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:a8d72c6272453450eede2ed9aaa14448e6525569e14217573057ded146090db3", size = 122884756, upload-time = "2025-11-13T20:16:04.002Z" },
 ]
 
 [[package]]
 name = "xgrammar"
-version = "0.1.25"
+version = "0.1.27"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "mlx-lm", marker = "platform_machine == 'arm64' and sys_platform == 'darwin'" },
@@ -5744,33 +5668,35 @@ dependencies = [
     { name = "numpy", version = "2.0.2", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
     { name = "numpy", version = "2.1.3", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "pydantic" },
-    { name = "torch" },
+    { name = "torch", version = "2.8.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },
+    { name = "torch", version = "2.9.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10'" },
     { name = "transformers" },
-    { name = "triton", marker = "platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "triton", version = "3.4.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
+    { name = "triton", version = "3.5.0", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version >= '3.10' and platform_machine == 'x86_64' and sys_platform == 'linux'" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/f2/a9/dc3c63cf7f082d183711e46ef34d10d8a135c2319dc581905d79449f52ea/xgrammar-0.1.25.tar.gz", hash = "sha256:70ce16b27e8082f20808ed759b0733304316facc421656f0f30cfce514b5b77a", size = 2297187, upload-time = "2025-09-21T05:58:58.942Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/c0/b4/8f78b56ebf64f161258f339cc5898bf761b4fb6c6805d0bca1bcaaaef4a1/xgrammar-0.1.25-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:d12d1078ee2b5c1531610489b433b77694a7786210ceb2c0c1c1eb058e9053c7", size = 679074, upload-time = "2025-09-21T05:58:20.344Z" },
-    { url = "https://files.pythonhosted.org/packages/52/38/b57120b73adcd342ef974bff14b2b584e7c47edf28d91419cb9325fd5ef2/xgrammar-0.1.25-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:c2e940541b7cddf3ef55a70f20d4c872af7f0d900bc0ed36f434bf7212e2e729", size = 622668, upload-time = "2025-09-21T05:58:22.269Z" },
-    { url = "https://files.pythonhosted.org/packages/19/8d/64430d01c21ca2b1d8c5a1ed47c90f8ac43717beafc9440d01d81acd5cfc/xgrammar-0.1.25-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2063e1c72f0c00f47ac8ce7ce0fcbff6fa77f79012e063369683844e2570c266", size = 8517569, upload-time = "2025-09-21T05:58:23.77Z" },
-    { url = "https://files.pythonhosted.org/packages/b1/c4/137d0e9cd038ff4141752c509dbeea0ec5093eb80815620c01b1f1c26d0a/xgrammar-0.1.25-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9785eafa251c996ebaa441f3b8a6c037538930104e265a64a013da0e6fd2ad86", size = 8709188, upload-time = "2025-09-21T05:58:26.246Z" },
-    { url = "https://files.pythonhosted.org/packages/6c/3d/c228c470d50865c9db3fb1e75a95449d0183a8248519b89e86dc481d6078/xgrammar-0.1.25-cp310-cp310-win_amd64.whl", hash = "sha256:42ecefd020038b3919a473fe5b9bb9d8d809717b8689a736b81617dec4acc59b", size = 698919, upload-time = "2025-09-21T05:58:28.368Z" },
-    { url = "https://files.pythonhosted.org/packages/9e/b7/ca0ff7c91f24b2302e94b0e6c2a234cc5752b10da51eb937e7f2aa257fde/xgrammar-0.1.25-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:27d7ac4be05cf9aa258c109a8647092ae47cb1e28df7d27caced6ab44b72b799", size = 678801, upload-time = "2025-09-21T05:58:29.936Z" },
-    { url = "https://files.pythonhosted.org/packages/43/cd/fdf4fb1b5f9c301d381656a600ad95255a76fa68132978af6f06e50a46e1/xgrammar-0.1.25-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:151c1636188bc8c5cdf318cefc5ba23221c9c8cc07cb392317fb3f7635428150", size = 622565, upload-time = "2025-09-21T05:58:31.185Z" },
-    { url = "https://files.pythonhosted.org/packages/55/04/55a87e814bcab771d3e4159281fa382b3d5f14a36114f2f9e572728da831/xgrammar-0.1.25-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:35fc135650aa204bf84db7fe9c0c0f480b6b11419fe47d89f4bd21602ac33be9", size = 8517238, upload-time = "2025-09-21T05:58:32.835Z" },
-    { url = "https://files.pythonhosted.org/packages/31/f6/3c5210bc41b61fb32b66bf5c9fd8ec5edacfeddf9860e95baa9caa9a2c82/xgrammar-0.1.25-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:fc19d6d7e8e51b6c9a266e949ac7fb3d2992447efeec7df32cca109149afac18", size = 8709514, upload-time = "2025-09-21T05:58:34.727Z" },
-    { url = "https://files.pythonhosted.org/packages/21/de/85714f307536b328cc16cc6755151865e8875378c8557c15447ca07dff98/xgrammar-0.1.25-cp311-cp311-win_amd64.whl", hash = "sha256:8fcb24f5a7acd5876165c50bd51ce4bf8e6ff897344a5086be92d1fe6695f7fe", size = 698722, upload-time = "2025-09-21T05:58:36.411Z" },
-    { url = "https://files.pythonhosted.org/packages/bf/d7/a7bdb158afa88af7e6e0d312e9677ba5fb5e423932008c9aa2c45af75d5d/xgrammar-0.1.25-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:96500d7578c46e8551253b9211b02e02f54e147bc290479a64717d80dcf4f7e3", size = 678250, upload-time = "2025-09-21T05:58:37.936Z" },
-    { url = "https://files.pythonhosted.org/packages/10/9d/b20588a3209d544a3432ebfcf2e3b1a455833ee658149b08c18eef0c6f59/xgrammar-0.1.25-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:73ba9031e359447af53ce89dfb0775e7b9f4b358d513bcc28a6b4deace661dd5", size = 621550, upload-time = "2025-09-21T05:58:39.464Z" },
-    { url = "https://files.pythonhosted.org/packages/99/9c/39bb38680be3b6d6aa11b8a46a69fb43e2537d6728710b299fa9fc231ff0/xgrammar-0.1.25-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c519518ebc65f75053123baaf23776a21bda58f64101a64c2fc4aa467c9cd480", size = 8519097, upload-time = "2025-09-21T05:58:40.831Z" },
-    { url = "https://files.pythonhosted.org/packages/c6/c2/695797afa9922c30c45aa94e087ad33a9d87843f269461b622a65a39022a/xgrammar-0.1.25-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:47fdbfc6007df47de2142613220292023e88e4a570546b39591f053e4d9ec33f", size = 8712184, upload-time = "2025-09-21T05:58:43.142Z" },
-    { url = "https://files.pythonhosted.org/packages/e4/7f/aa80d1d4c4632cd3d8d083f1de8b470fcb3df23d9165992a3ced019f1b93/xgrammar-0.1.25-cp312-cp312-win_amd64.whl", hash = "sha256:c9b3defb6b45272e896da401f43b513f5ac12104ec3101bbe4d3a7d02bcf4a27", size = 698264, upload-time = "2025-09-21T05:58:44.787Z" },
-    { url = "https://files.pythonhosted.org/packages/d8/85/83401a12a029bb1ea8cff08b19c47aa7077b359ad04251cc1725c0dce926/xgrammar-0.1.25-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:a62dea5d73147a254e71e07ceae4a48c0f5a294cce2fa3e028159f48da19a39d", size = 679391, upload-time = "2025-09-21T05:58:51.019Z" },
-    { url = "https://files.pythonhosted.org/packages/e1/68/9854e947333698b282bd0e7c074e4433d45750cb37a0ae9d5753e66f5695/xgrammar-0.1.25-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:2b309807ad837c1cbe2f833830b665a15309e11692b53795363c59041c65074f", size = 622817, upload-time = "2025-09-21T05:58:52.242Z" },
-    { url = "https://files.pythonhosted.org/packages/4c/c3/1e85087460cb752ab5581fbc7d8e604b9de27d2b0312cfbc4a90126cea02/xgrammar-0.1.25-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:073829d8a53ef482e6b51469316f6e505962460bb576ae4d4a606769c4c37678", size = 8517691, upload-time = "2025-09-21T05:58:53.66Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/b9/531d823576559c4528ca4b3907dd6baed11f9f5487ad10997b2200e1f361/xgrammar-0.1.25-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:241529d6104d97738b3e29c573bffa6d0fa89a8d0299b2c685358cc13858205c", size = 8707414, upload-time = "2025-09-21T05:58:55.457Z" },
-    { url = "https://files.pythonhosted.org/packages/2e/49/abdc00753a3eb30c0c85f3eb670420c6c294e2ac569e0f6210fbaca8e41e/xgrammar-0.1.25-cp39-cp39-win_amd64.whl", hash = "sha256:f5d46e1749d9324684d2462e428bc63652096addc1e2c21db2ae66ca88e76a1c", size = 699327, upload-time = "2025-09-21T05:58:57.541Z" },
+sdist = { url = "https://files.pythonhosted.org/packages/62/e1/b522b1e50fddd773d368c2945ef5ed628aa90c0c972027f9aa5a51d6d4f9/xgrammar-0.1.27.tar.gz", hash = "sha256:40af7bb2891f1633ec7f660723c74a92a963307d283aca9e3b4e53a0feaf1d46", size = 2303435, upload-time = "2025-11-04T03:11:53.512Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b0/55/03a548a116fa5cc716ce70e15240ca61ddaae046ed34c711e63d3d91d047/xgrammar-0.1.27-cp310-cp310-macosx_10_14_x86_64.whl", hash = "sha256:6e3ea7cd74a7d4188744f90878507637ce9ac5f671cb9d1bda9d53305a46889e", size = 664256, upload-time = "2025-11-04T03:11:08.471Z" },
+    { url = "https://files.pythonhosted.org/packages/84/a5/45a430a7fb44f70303742c59e7d792ca6d4b7960e9252ec5238f1112bbcd/xgrammar-0.1.27-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:73ca9ec86e81a7f936c5668b7e6dda6929c078d1748b7615c8da504584b6c24a", size = 637358, upload-time = "2025-11-04T03:11:10.975Z" },
+    { url = "https://files.pythonhosted.org/packages/cb/2b/3867379f76b97fb7cb03bc0fa1d0f81f19d13df3c313bd22878b636b0f50/xgrammar-0.1.27-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:fa8b7cc167737a9b4c3e1012faa7b488cc5b451ea8403c4d77ec1d53b58e9266", size = 8687578, upload-time = "2025-11-04T03:11:13.304Z" },
+    { url = "https://files.pythonhosted.org/packages/ae/35/fe718ec90c210ab892a845af6d4e6e876a3d3c7dcc1bacaa98abfec42c0f/xgrammar-0.1.27-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:6e0c5899b59c8e45ba3a6f3b9e7fb2ef23243f09b164f724d59c7734173bb3db", size = 8869161, upload-time = "2025-11-04T03:11:15.572Z" },
+    { url = "https://files.pythonhosted.org/packages/e8/44/674724714407e0265d088ad40a17d367c00d72d206f2b15d559a644a36dc/xgrammar-0.1.27-cp310-cp310-win_amd64.whl", hash = "sha256:1ce2558992b0ffda65f46772bae94b051d139f0036968853078904bc167d4a8d", size = 709212, upload-time = "2025-11-04T03:11:17.572Z" },
+    { url = "https://files.pythonhosted.org/packages/93/bb/e6d30457c99a0ce11247154ecb1f3f9fab5960192a0564c2862ba9b98897/xgrammar-0.1.27-cp311-cp311-macosx_10_14_x86_64.whl", hash = "sha256:c995c71ea94b153eac0e08c36eb82a898d7d71e4b77ce93f3b9fe648bd2d3a04", size = 664112, upload-time = "2025-11-04T03:11:18.932Z" },
+    { url = "https://files.pythonhosted.org/packages/7e/81/caab5c46d314c1b005e36c9ec8aef124f7c52619d980f2bbd2d4cf4cd491/xgrammar-0.1.27-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:456f2f74135a414f44413599d90a382f5b22e6b515e4ae7e8938a28f7efacbaa", size = 637181, upload-time = "2025-11-04T03:11:20.29Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/29/7f78ed69b5f221206af0b68b0517335f9c09459def5d63065827a79fec74/xgrammar-0.1.27-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ed23e6960218e791ecaccbbbb66d7caa5c0ed8636aca85807d81b89ba87a7f33", size = 8674617, upload-time = "2025-11-04T03:11:22.255Z" },
+    { url = "https://files.pythonhosted.org/packages/cc/a2/afcce6a59b83644ffe19ffebe8107355febb15d8084ce5316eccd457e3c8/xgrammar-0.1.27-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:02fe3b137d041649b8f7a180a0aa7f3466d47579ce4e9fbdb77208b59621b2ab", size = 8869958, upload-time = "2025-11-04T03:11:24.751Z" },
+    { url = "https://files.pythonhosted.org/packages/76/fb/a4a3254041174013ff09e99c298f2bc6c03f34891df458839de7cbb53e4b/xgrammar-0.1.27-cp311-cp311-win_amd64.whl", hash = "sha256:db0c74f7cc4fb2b5d566eee873e4d18920ed5ee0fe500178b412408d0dad3686", size = 709137, upload-time = "2025-11-04T03:11:26.672Z" },
+    { url = "https://files.pythonhosted.org/packages/39/b6/09b43e2adff45d30ebcf9110d0ff753f4c96b368adaa2d166df3dee88d5f/xgrammar-0.1.27-cp312-cp312-macosx_10_14_x86_64.whl", hash = "sha256:6404a7714440eb86ab0379d749f33591274eeef04787dc00d61f22069f3ed51d", size = 663319, upload-time = "2025-11-04T03:11:28.682Z" },
+    { url = "https://files.pythonhosted.org/packages/88/8b/53eb5c6d0df8df9f6350f182516a5b8c7b8b11d62650300d2c04af2bc4ea/xgrammar-0.1.27-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:d01fa9894bc44a7f6a70b0301b59f3e310c0e0e7b7ea4cf5ce190b12d8220dd8", size = 636168, upload-time = "2025-11-04T03:11:30.373Z" },
+    { url = "https://files.pythonhosted.org/packages/08/1b/53d30395bb973f13255d3e3a72961f95fdfb4083877c3f93bb626e3d1522/xgrammar-0.1.27-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:906c0601bac9170e1bab77ca985259035ff9c386c347efcb191555eab86e984e", size = 8676340, upload-time = "2025-11-04T03:11:32.203Z" },
+    { url = "https://files.pythonhosted.org/packages/48/74/70cfac0171d9f309cfe18c5384330e3edc9466c436b258495fd30ecf29a3/xgrammar-0.1.27-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:eb68988a122f544301c496f2cac8ee82960ca7f5b3a42a952b2a00c0a55e6ca5", size = 8870650, upload-time = "2025-11-04T03:11:34.322Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/a1/0392aa9c7669c56f7f88e4423b246476a74a72c3bb9db944e1bfc029985e/xgrammar-0.1.27-cp312-cp312-win_amd64.whl", hash = "sha256:3aac335ea052afc8f8dc34b9f2afcb9462a68189423aed9f60b0941db6cfc310", size = 708811, upload-time = "2025-11-04T03:11:36.214Z" },
+    { url = "https://files.pythonhosted.org/packages/3b/8c/c9e5892ce94f3c79e77951e89a40369f253199f2a163dc6dd0c9db2d9bff/xgrammar-0.1.27-cp39-cp39-macosx_10_14_x86_64.whl", hash = "sha256:ee2f3fdbcfeb5520c96004243413171c44cbcf9af5d424cbd533e4943ef643bf", size = 664435, upload-time = "2025-11-04T03:11:43.833Z" },
+    { url = "https://files.pythonhosted.org/packages/c6/ae/c227370576f1fe006cb6a36602cedc196c0441a3e18d382867432b86185e/xgrammar-0.1.27-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:bac3882f313369991d0450b61791443fe34b1c61d6c0de734c4533eeb51ecf9d", size = 637537, upload-time = "2025-11-04T03:11:45.271Z" },
+    { url = "https://files.pythonhosted.org/packages/06/27/694b541e54399237d3d19d2aa62c10be1b0eed46a81be96f0d83168a7207/xgrammar-0.1.27-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c0fbe08b57f28c631abb759195af38daa82b7e8c4d8fbebadaa5f62aeb7e7589", size = 8673501, upload-time = "2025-11-04T03:11:47.499Z" },
+    { url = "https://files.pythonhosted.org/packages/80/d8/c48454f5bd7e607b2c062f5aacbcb19bf2ae852d0de3faed1a79b5c61cfe/xgrammar-0.1.27-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:11ea8668409b7db230f958365346c2a983ca1af60206e09bcc4e71c9e19ba8a2", size = 8866885, upload-time = "2025-11-04T03:11:49.613Z" },
+    { url = "https://files.pythonhosted.org/packages/67/50/0864675aa98f337dcd7c4a80ff5d64a3671e252aa97ff10229092d0a55be/xgrammar-0.1.27-cp39-cp39-win_amd64.whl", hash = "sha256:4cedbe5c7e18bba6894518aee97562f727590a5b8200df39502b81445286d96e", size = 709442, upload-time = "2025-11-04T03:11:51.992Z" },
 ]
 
 [[package]]

From 1e1c9176985a0b1edfc99cbd13ed3e40913601dc Mon Sep 17 00:00:00 2001
From: Anri Lombard <anri.m.lombard@gmail.com>
Date: Fri, 28 Nov 2025 17:41:15 +0200
Subject: [PATCH 487/505] Use positional unpacking for mlx_lm.load() tuple

mlx_lm.load() returns a tuple (model, tokenizer), not a dict.
The example incorrectly used ** (keyword unpacking) instead of * (positional unpacking).
---
 docs/guide/getting_started.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/docs/guide/getting_started.md b/docs/guide/getting_started.md
index 617e27e5c..34331c322 100644
--- a/docs/guide/getting_started.md
+++ b/docs/guide/getting_started.md
@@ -125,7 +125,7 @@ For a quick start, you can find below an example of how to initialize all suppor
     # Create an MLXLM model with the output of mlx_lm.load
     # The model will be downloaded from the HuggingFace hub
     model = outlines.from_mlxlm(
-        **mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
+        *mlx_lm.load("mlx-community/SmolLM-135M-Instruct-4bit")
     )
     ```
 

From be1c25f55323f5a0144e8af98c74e200941cf5ca Mon Sep 17 00:00:00 2001
From: Anri Lombard <anri.m.lombard@gmail.com>
Date: Mon, 1 Dec 2025 20:36:35 +0200
Subject: [PATCH 488/505] Remove todos

---
 outlines/models/dottxt.py | 3 ---
 outlines/models/ollama.py | 3 ---
 outlines/models/openai.py | 3 ---
 3 files changed, 9 deletions(-)

diff --git a/outlines/models/dottxt.py b/outlines/models/dottxt.py
index f0eef2458..7aab4c3c0 100644
--- a/outlines/models/dottxt.py
+++ b/outlines/models/dottxt.py
@@ -38,9 +38,6 @@ def format_input(self, model_input: str) -> str:
     def format_output_type(self, output_type: Optional[Any] = None) -> str:
         """Format the output type to pass to the client.
 
-        TODO: `int`, `float` and other Python types could be supported via
-        JSON Schema.
-
         Parameters
         ----------
         output_type
diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index cc693854d..8965e34fa 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -111,9 +111,6 @@ def format_output_type(
     ) -> Optional[dict]:
         """Format the output type to pass to the client.
 
-        TODO: `int`, `float` and other Python types could be supported via
-        JSON Schema.
-
         Parameters
         ----------
         output_type
diff --git a/outlines/models/openai.py b/outlines/models/openai.py
index 4626bffa8..a1831b530 100644
--- a/outlines/models/openai.py
+++ b/outlines/models/openai.py
@@ -139,9 +139,6 @@ def format_output_type(self, output_type: Optional[Any] = None) -> dict:
         """Generate the `response_format` argument to the client based on the
         output type specified by the user.
 
-        TODO: `int`, `float` and other Python types could be supported via
-        JSON Schema.
-
         Parameters
         ----------
         output_type

From 2fa37774752fc6c7249d5650a52be07127ecbf39 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Wed, 3 Dec 2025 15:45:42 +0800
Subject: [PATCH 489/505] fix: correct chat format usage

---
 docs/guide/migration.md                 | 2 ++
 tests/backends/test_llguidance.py       | 1 +
 tests/backends/test_outlines_core.py    | 1 +
 tests/backends/test_xgrammar.py         | 1 +
 tests/models/test_llamacpp.py           | 2 ++
 tests/models/test_llamacpp_tokenizer.py | 2 ++
 6 files changed, 9 insertions(+)

diff --git a/docs/guide/migration.md b/docs/guide/migration.md
index 729cceac1..0d705ba14 100644
--- a/docs/guide/migration.md
+++ b/docs/guide/migration.md
@@ -62,6 +62,7 @@ from outlines import models
 model = models.llamacpp(
     repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
     filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    chat_format="qwen",
 )
 ```
 
@@ -74,6 +75,7 @@ from outlines import from_llamacpp
 llamacpp_model = Llama.from_pretrained(
     repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
     filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+    chat_format="qwen",
 )
 model = from_llamacpp(llamacpp_model)
 ```
diff --git a/tests/backends/test_llguidance.py b/tests/backends/test_llguidance.py
index fb5faea91..465f54871 100644
--- a/tests/backends/test_llguidance.py
+++ b/tests/backends/test_llguidance.py
@@ -31,6 +31,7 @@ def model_llamacpp():
         llama_cpp.Llama.from_pretrained(
             repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
             filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            chat_format="qwen",
         )
     )
 
diff --git a/tests/backends/test_outlines_core.py b/tests/backends/test_outlines_core.py
index ce3242457..9dc3db04a 100644
--- a/tests/backends/test_outlines_core.py
+++ b/tests/backends/test_outlines_core.py
@@ -30,6 +30,7 @@ def model_llamacpp():
         llama_cpp.Llama.from_pretrained(
             repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
             filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            chat_format="qwen",
         )
     )
 
diff --git a/tests/backends/test_xgrammar.py b/tests/backends/test_xgrammar.py
index e25f66508..855213990 100644
--- a/tests/backends/test_xgrammar.py
+++ b/tests/backends/test_xgrammar.py
@@ -27,6 +27,7 @@ def model_llamacpp():
         llama_cpp.Llama.from_pretrained(
             repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
             filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            chat_format="qwen",
         )
     )
 
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 9b881b7de..776da800b 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -20,6 +20,7 @@ def test_load_model():
         Llama.from_pretrained(
             repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
             filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            chat_format="qwen"
         )
     )
 
@@ -36,6 +37,7 @@ def model(tmp_path_factory):
         Llama.from_pretrained(
             repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
             filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+            chat_format="qwen",
         )
     )
 
diff --git a/tests/models/test_llamacpp_tokenizer.py b/tests/models/test_llamacpp_tokenizer.py
index dbf2507f3..80b063464 100644
--- a/tests/models/test_llamacpp_tokenizer.py
+++ b/tests/models/test_llamacpp_tokenizer.py
@@ -13,6 +13,7 @@ def model():
     model = llama_cpp.Llama.from_pretrained(
         repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
         filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        chat_format="qwen",
     )
     setattr(
         model.tokenizer_,
@@ -27,6 +28,7 @@ def model_no_hf_tokenizer():
     model = llama_cpp.Llama.from_pretrained(
         repo_id="M4-ai/TinyMistral-248M-v2-Instruct-GGUF",
         filename="TinyMistral-248M-v2-Instruct.Q4_K_M.gguf",
+        chat_format="qwen",
     )
     del model.tokenizer_
     return model

From f9c72dd505f8190cc4cfd484db8b319ba117f34a Mon Sep 17 00:00:00 2001
From: Anri Lombard <anri.m.lombard@gmail.com>
Date: Fri, 5 Dec 2025 07:05:10 +0200
Subject: [PATCH 490/505] Add AsyncOllama to __all__ exports

The AsyncOllama class was missing from the module's __all__ list,
making it inaccessible via 'from outlines.models.ollama import *'.
All other async model classes (AsyncVLLM, AsyncOpenAI, AsyncTGI,
AsyncMistral, AsyncSGLang) are properly exported.
---
 outlines/models/ollama.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/outlines/models/ollama.py b/outlines/models/ollama.py
index 8965e34fa..b1b7330f3 100644
--- a/outlines/models/ollama.py
+++ b/outlines/models/ollama.py
@@ -19,7 +19,7 @@
     from ollama import Client
     from ollama import AsyncClient
 
-__all__ = ["Ollama", "from_ollama"]
+__all__ = ["AsyncOllama", "Ollama", "from_ollama"]
 
 
 class OllamaTypeAdapter(ModelTypeAdapter):

From 96c3c405ceb20ab7251874ccdb2330fed79cec47 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Tue, 2 Dec 2025 16:56:29 +0800
Subject: [PATCH 491/505] feat: add chat template checking func

---
 outlines/models/tokenizer.py | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/outlines/models/tokenizer.py b/outlines/models/tokenizer.py
index 5463c5eaa..37a918b93 100644
--- a/outlines/models/tokenizer.py
+++ b/outlines/models/tokenizer.py
@@ -4,6 +4,7 @@
 if TYPE_CHECKING:
     import numpy as np
     from numpy.typing import NDArray
+    from transformers import PreTrainedTokenizer, PreTrainedTokenizerFast
 
 
 class Tokenizer(Hashable, Protocol):
@@ -31,3 +32,12 @@ def convert_token_to_string(self, token: str) -> str:
         token that includes `Ġ` with a string.
         """
         ...
+
+
+def _check_hf_chat_template(tokenizer: "PreTrainedTokenizer | PreTrainedTokenizerFast") -> bool:
+    """Check if the HuggingFace tokenizer has a chat template."""
+    try:
+        tokenizer.get_chat_template()
+        return True
+    except ValueError:
+        return False

From baba491b21abb1e59e522129c03a3c7bc8168e71 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Tue, 2 Dec 2025 16:58:32 +0800
Subject: [PATCH 492/505] fix: Best-effort use of chat completion for
 VLLMOffline

---
 outlines/models/vllm_offline.py | 63 +++++++++++++++++++++++++--------
 1 file changed, 49 insertions(+), 14 deletions(-)

diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index 9d1ea08f3..c74a78570 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -12,6 +12,12 @@
 if TYPE_CHECKING:
     from vllm import LLM
     from vllm.sampling_params import SamplingParams
+    from vllm.transformers_utils.tokenizer import (
+        AnyTokenizer,
+        PreTrainedTokenizer,
+        PreTrainedTokenizerFast,
+        TokenizerBase
+    )
 
 __all__ = ["VLLMOffline", "from_vllm_offline"]
 
@@ -19,6 +25,9 @@
 class VLLMOfflineTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `VLLMOffline` model."""
 
+    def __init__(self, has_chat_template: bool = False):
+        self.has_chat_template = has_chat_template
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
@@ -36,10 +45,12 @@ def format_input(self, model_input):
         )
 
     @format_input.register(str)
-    def format_input_str(self, model_input: str) -> str:
+    def format_input_str(self, model_input: str) -> str | list:
         """Format a `str` input.
 
         """
+        if self.has_chat_template:
+            return self.format_input_chat(Chat([{"role": "user", "content": model_input}]))
         return model_input
 
     @format_input.register(Chat)
@@ -109,7 +120,8 @@ def __init__(self, model: "LLM"):
 
         """
         self.model = model
-        self.type_adapter = VLLMOfflineTypeAdapter()
+        self.tokenizer = self.model.get_tokenizer()
+        self.type_adapter = VLLMOfflineTypeAdapter(has_chat_template=self._check_chat_template())
 
     def _build_generation_args(
         self,
@@ -163,15 +175,17 @@ def generate(
             output_type,
         )
 
-        if isinstance(model_input, Chat):
+        model_input = self.type_adapter.format_input(model_input)
+
+        if isinstance(model_input, list):
             results = self.model.chat(
-                messages=self.type_adapter.format_input(model_input),
+                messages=model_input,
                 sampling_params=sampling_params,
                 **inference_kwargs,
             )
         else:
             results = self.model.generate(
-                prompts=self.type_adapter.format_input(model_input),
+                prompts=model_input,
                 sampling_params=sampling_params,
                 **inference_kwargs,
             )
@@ -213,16 +227,20 @@ def generate_batch(
             output_type,
         )
 
-        if any(isinstance(item, Chat) for item in model_input):
-            raise TypeError(
-                "Batch generation is not available for the `Chat` input type."
-            )
+        model_inputs = [self.type_adapter.format_input(item) for item in model_input]
 
-        results = self.model.generate(
-            prompts=[self.type_adapter.format_input(item) for item in model_input],
-            sampling_params=sampling_params,
-            **inference_kwargs,
-        )
+        if model_inputs and isinstance(model_inputs[0], list):
+            results = self.model.chat(
+                messages=model_inputs,
+                sampling_params=sampling_params,
+                **inference_kwargs,
+            )
+        else:
+            results = self.model.generate(
+                prompts=model_inputs,
+                sampling_params=sampling_params,
+                **inference_kwargs,
+            )
         return [[sample.text for sample in batch.outputs] for batch in results]
 
     def generate_stream(self, model_input, output_type, **inference_kwargs):
@@ -235,6 +253,23 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
             "Streaming is not available for the vLLM offline integration."
         )
 
+    def _check_chat_template(self) -> bool:
+        """Check if the tokenizer has a chat template."""
+        from outlines.models.tokenizer import _check_hf_chat_template
+
+        if isinstance(self.tokenizer, (PreTrainedTokenizer, PreTrainedTokenizerFast)):
+            return _check_hf_chat_template(self.tokenizer)
+        elif isinstance(self.tokenizer, TokenizerBase):
+            # vLLM defines its own TokenizerBase class, and only provides
+            # limited compatibility with HuggingFace tokenizers. So we
+            # need to check for chat template support differently.
+            try:
+                self.tokenizer.apply_chat_template([{"role": "user", "content": "test"}])
+                return True
+            except Exception:
+                return False
+        else:  # Never reached  # pragma: no cover
+            return False
 
 def from_vllm_offline(model: "LLM") -> VLLMOffline:
     """Create an Outlines `VLLMOffline` model instance from a `vllm.LLM`

From 9e762ca67145f64f919c338aff15b51232124e19 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Tue, 2 Dec 2025 16:59:08 +0800
Subject: [PATCH 493/505] fix: Best-effort use of chat completion for
 Transformers

---
 outlines/models/transformers.py | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/outlines/models/transformers.py b/outlines/models/transformers.py
index 42f05553c..1dc911f41 100644
--- a/outlines/models/transformers.py
+++ b/outlines/models/transformers.py
@@ -8,7 +8,7 @@
 
 from outlines.inputs import Audio, Chat, Image, Video
 from outlines.models.base import Model, ModelTypeAdapter
-from outlines.models.tokenizer import Tokenizer
+from outlines.models.tokenizer import Tokenizer, _check_hf_chat_template
 from outlines.processors import OutlinesLogitsProcessor
 
 if TYPE_CHECKING:
@@ -136,8 +136,9 @@ def __setstate__(self, state):
 class TransformersTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `Transformers` model."""
 
-    def __init__(self, **kwargs):
-        self.tokenizer = kwargs.get("tokenizer")
+    def __init__(self, tokenizer: "PreTrainedTokenizer", has_chat_template: bool = False):
+        self.tokenizer = tokenizer
+        self.has_chat_template = has_chat_template
 
     @singledispatchmethod
     def format_input(self, model_input):
@@ -161,6 +162,8 @@ def format_input(self, model_input):
 
     @format_input.register(str)
     def format_str_input(self, model_input: str) -> str:
+        if self.has_chat_template:
+            return self.format_chat_input(Chat([{"role": "user", "content": model_input}]))
         return model_input
 
     @format_input.register(Chat)
@@ -243,7 +246,10 @@ def __init__(
         self.hf_tokenizer = tokenizer
         self.tokenizer = TransformerTokenizer(tokenizer)
         self.device_dtype = device_dtype
-        self.type_adapter = TransformersTypeAdapter(tokenizer=tokenizer)
+        self.type_adapter = TransformersTypeAdapter(
+            tokenizer=tokenizer,
+            has_chat_template=_check_hf_chat_template(tokenizer)
+        )
 
         if (
             FlaxPreTrainedModel is not None

From 16e3b014d88d78ee1b1ddaaa980d2cf28a5609c9 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Tue, 2 Dec 2025 16:59:39 +0800
Subject: [PATCH 494/505] fix: Best-effort use of chat completion for mlxlm

---
 outlines/models/mlxlm.py | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/outlines/models/mlxlm.py b/outlines/models/mlxlm.py
index 5111eccaa..5f8bed685 100644
--- a/outlines/models/mlxlm.py
+++ b/outlines/models/mlxlm.py
@@ -5,6 +5,7 @@
 
 from outlines.inputs import Chat
 from outlines.models.base import Model, ModelTypeAdapter
+from outlines.models.tokenizer import _check_hf_chat_template
 from outlines.models.transformers import TransformerTokenizer
 from outlines.processors import OutlinesLogitsProcessor
 
@@ -18,8 +19,9 @@
 class MLXLMTypeAdapter(ModelTypeAdapter):
     """Type adapter for the `MLXLM` model."""
 
-    def __init__(self, **kwargs):
-        self.tokenizer = kwargs.get("tokenizer")
+    def __init__(self, tokenizer: "PreTrainedTokenizer", has_chat_template: bool = False):
+        self.tokenizer = tokenizer
+        self.has_chat_template = has_chat_template
 
     @singledispatchmethod
     def format_input(self, model_input):
@@ -42,7 +44,9 @@ def format_input(self, model_input):
         )
 
     @format_input.register(str)
-    def format_str_input(self, model_input: str):
+    def format_str_input(self, model_input: str) -> str:
+        if self.has_chat_template:
+            return self.format_chat_input(Chat([{"role": "user", "content": model_input}]))
         return model_input
 
     @format_input.register(Chat)
@@ -113,7 +117,10 @@ def __init__(
         self.mlx_tokenizer = tokenizer
         # self.tokenizer is used by the logits processor
         self.tokenizer = TransformerTokenizer(tokenizer._tokenizer)
-        self.type_adapter = MLXLMTypeAdapter(tokenizer=tokenizer)
+        self.type_adapter = MLXLMTypeAdapter(
+            tokenizer=tokenizer,
+            has_chat_template=_check_hf_chat_template(tokenizer)
+        )
 
     def generate(
         self,

From 6892803eae202fce48060305356b5a79be81ef6a Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Tue, 2 Dec 2025 17:00:24 +0800
Subject: [PATCH 495/505] fix: Best-effort use of chat completion for LlamaCpp

---
 outlines/models/llamacpp.py | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 8b788a7f9..a560bbb19 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -152,6 +152,15 @@ class LlamaCppTypeAdapter(ModelTypeAdapter):
 
     """
 
+    def __init__(self, has_chat_template: bool = False):
+        """
+        Parameters
+        ----------
+        has_chat_template
+            Whether the model has a chat template defined.
+        """
+        self.has_chat_template = has_chat_template
+
     @singledispatchmethod
     def format_input(self, model_input):
         """Generate the prompt argument to pass to the model.
@@ -173,7 +182,9 @@ def format_input(self, model_input):
         )
 
     @format_input.register(str)
-    def format_str_input(self, model_input: str) -> str:
+    def format_str_input(self, model_input: str) -> str | list:
+        if self.has_chat_template:
+            return [{"role": "user", "content": model_input}]
         return model_input
 
     @format_input.register(Chat)
@@ -237,7 +248,7 @@ def __init__(self, model: "Llama"):
         """
         self.model = model
         self.tokenizer = LlamaCppTokenizer(self.model)
-        self.type_adapter = LlamaCppTypeAdapter()
+        self.type_adapter = LlamaCppTypeAdapter(has_chat_template=self._check_chat_template())
 
     def generate(
         self,
@@ -340,6 +351,16 @@ def generate_stream(
             for chunk in generator:
                 yield chunk["choices"][0]["delta"].get("content", "")
 
+    def _check_chat_template(self) -> bool:
+        """Check if the model has a chat template defined."""
+        return (
+            self.model.chat_format is not None
+            or self.model.chat_handler is not None
+            or (
+                hasattr(self.model, "metadata")
+                and self.model.metadata.get("tokenizer.chat_template") is not None
+            )
+        )
 
 def from_llamacpp(model: "Llama"):
     """Create an Outlines `LlamaCpp` model instance from a

From afabd981d56f0dfa276837d0e8a3073105515850 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Tue, 2 Dec 2025 17:00:53 +0800
Subject: [PATCH 496/505] test: Add some initial tests

---
 tests/models/test_llamacpp_type_adapter.py    | 33 +++++++++++++++++++
 tests/models/test_mlxlm_type_adapter.py       | 29 ++++++++++++++++
 .../models/test_transformers_type_adapter.py  | 32 ++++++++++++++----
 .../models/test_vllm_offline_type_adapter.py  | 17 ++++++++++
 4 files changed, 104 insertions(+), 7 deletions(-)

diff --git a/tests/models/test_llamacpp_type_adapter.py b/tests/models/test_llamacpp_type_adapter.py
index 403b3589c..167d50f4e 100644
--- a/tests/models/test_llamacpp_type_adapter.py
+++ b/tests/models/test_llamacpp_type_adapter.py
@@ -1,5 +1,6 @@
 import pytest
 import io
+from unittest.mock import MagicMock
 
 from llama_cpp import LogitsProcessorList
 from PIL import Image as PILImage
@@ -62,6 +63,38 @@ def test_llamacpp_type_adapter_format_input(adapter, image):
         ]))
 
 
+def test_llamacpp_type_adapter_format_input_with_chat_format():
+    adapter = LlamaCppTypeAdapter(has_chat_template=True)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == [{"role": "user", "content": "prompt"}]
+
+
+def test_llamacpp_type_adapter_format_input_with_chat_handler():
+    adapter = LlamaCppTypeAdapter(has_chat_template=True)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == [{"role": "user", "content": "prompt"}]
+
+
+def test_llamacpp_type_adapter_format_input_with_metadata_template():
+    adapter = LlamaCppTypeAdapter(has_chat_template=True)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == [{"role": "user", "content": "prompt"}]
+
+
+def test_llamacpp_type_adapter_format_input_without_chat_capabilities():
+    adapter = LlamaCppTypeAdapter(has_chat_template=False)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == "prompt"
+
+
 def test_llamacpp_type_adapter_format_output_type(adapter, logits_processor):
     formatted = adapter.format_output_type(logits_processor)
     assert isinstance(formatted, LogitsProcessorList)
diff --git a/tests/models/test_mlxlm_type_adapter.py b/tests/models/test_mlxlm_type_adapter.py
index dd2feb16d..152eead01 100644
--- a/tests/models/test_mlxlm_type_adapter.py
+++ b/tests/models/test_mlxlm_type_adapter.py
@@ -1,5 +1,6 @@
 import pytest
 import io
+from unittest.mock import MagicMock
 
 from outlines_core import Index, Vocabulary
 from PIL import Image as PILImage
@@ -46,6 +47,34 @@ def image():
     return image
 
 
+def test_mlxlm_type_adapter_format_input_with_template():
+    tokenizer = MagicMock()
+    tokenizer.chat_template = "some_template"
+    tokenizer.apply_chat_template.return_value = "formatted_prompt"
+
+    adapter = MLXLMTypeAdapter(tokenizer=tokenizer, has_chat_template=True)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == "formatted_prompt"
+    tokenizer.apply_chat_template.assert_called_once_with(
+        [{"role": "user", "content": "prompt"}],
+        tokenize=False,
+        add_generation_prompt=True,
+    )
+
+
+def test_mlxlm_type_adapter_format_input_without_template():
+    tokenizer = MagicMock()
+    tokenizer.chat_template = None
+
+    adapter = MLXLMTypeAdapter(tokenizer=tokenizer, has_chat_template=False)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == "prompt"
+
+
 @pytest.mark.skipif(not HAS_MLX, reason="MLX tests require Apple Silicon")
 def test_mlxlm_type_adapter_format_input(adapter, image):
     # Anything else than a string/Chat (invalid)
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
index da7bbfcd2..571e3c2b6 100644
--- a/tests/models/test_transformers_type_adapter.py
+++ b/tests/models/test_transformers_type_adapter.py
@@ -1,5 +1,6 @@
 import io
 import pytest
+from unittest.mock import MagicMock
 
 import transformers
 from transformers import LogitsProcessorList
@@ -16,17 +17,27 @@
 
 @pytest.fixture
 def adapter():
-    tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
-    type_adapter = TransformersTypeAdapter(tokenizer=tokenizer)
-    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
-    type_adapter.tokenizer.chat_template = chat_template
+    tokenizer = MagicMock()
+    chat_template = "{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}{% if add_generation_prompt %}assistant:{% endif %}"
+    tokenizer.chat_template = chat_template
 
+    def apply_chat_template(messages, tokenize=False, add_generation_prompt=False):
+        res = ""
+        for msg in messages:
+            res += f"{msg['role']}: {msg['content']}"
+        if add_generation_prompt:
+            res += "assistant:"
+        return res
+
+    tokenizer.apply_chat_template.side_effect = apply_chat_template
+
+    type_adapter = TransformersTypeAdapter(tokenizer=tokenizer, has_chat_template=True)
     return type_adapter
 
 @pytest.fixture
 def logits_processor():
-    vocabulary = Vocabulary.from_pretrained("openai-community/gpt2")
-    index = Index(r"[0-9]{3}", vocabulary)
+    index = MagicMock()
+    index.tensor_library_name = "torch"
     return OutlinesCoreLogitsProcessor(index, "torch")
 
 @pytest.fixture
@@ -50,10 +61,17 @@ def test_transformers_type_adapter_format_input(adapter, image):
     with pytest.raises(TypeError, match="is not available."):
         adapter.format_input(["prompt", Image(image)])
 
-    # string
+    # string with chat template
+    # The fixture sets a chat template, so it should be formatted
+    assert adapter.format_input("Hello, world!") == "user: Hello, world!assistant:"
+
+    # string without chat template
+    adapter.has_chat_template = False
     assert adapter.format_input("Hello, world!") == "Hello, world!"
 
     # chat
+    # Restore chat template for chat test
+    adapter.has_chat_template = True
     assert isinstance(adapter.format_input(Chat(messages=[
         {"role": "system", "content": "You are a helpful assistant."},
         {"role": "user", "content": "Hello, world!"},
diff --git a/tests/models/test_vllm_offline_type_adapter.py b/tests/models/test_vllm_offline_type_adapter.py
index c431230e7..77bcb7353 100644
--- a/tests/models/test_vllm_offline_type_adapter.py
+++ b/tests/models/test_vllm_offline_type_adapter.py
@@ -1,5 +1,6 @@
 import io
 import json
+from unittest.mock import MagicMock
 
 import pytest
 from PIL import Image as PILImage
@@ -65,6 +66,22 @@ def test_vllm_offline_type_adapter_input_text(type_adapter):
     assert result == message
 
 
+def test_vllm_offline_type_adapter_input_text_with_template():
+    adapter = VLLMOfflineTypeAdapter(has_chat_template=True)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == [{"role": "user", "content": "prompt"}]
+
+
+def test_vllm_offline_type_adapter_input_text_without_template():
+    adapter = VLLMOfflineTypeAdapter(has_chat_template=False)
+    message = "prompt"
+    result = adapter.format_input(message)
+
+    assert result == "prompt"
+
+
 def test_vllm_offline_type_adapter_input_chat(type_adapter):
     model_input = Chat(messages=[
         {"role": "system", "content": "prompt"},

From 42c7842888a6780f4f66656820cc29da0bc7d832 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Wed, 3 Dec 2025 17:45:13 +0800
Subject: [PATCH 497/505] fix: update stream tests to handle potential empty
 chunks

---
 tests/models/test_llamacpp.py | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 776da800b..4d41c7203 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -171,8 +171,10 @@ class Foo(BaseModel):
 
     generator = model.stream("foo?", Foo)
 
-    x = next(generator)
-    assert x == "{"
+    # NOTE: The first few chunks may be empty (role info, control tokens, finish chunks)
+    # Relavant issue: https://github.com/abetlen/llama-cpp-python/issues/372
+    first_non_empty_token = next(x for x in generator if x)
+    assert first_non_empty_token == "{"
 
 
 def test_llamacpp_stream_cfg(model, ebnf_grammar):
@@ -204,8 +206,8 @@ class Foo(Enum):
 
     generator = model.stream("foo?", Foo)
 
-    x = next(generator)
-    assert x[0] in ("B", "F")
+    first_non_empty_token = next(x for x in generator if x)
+    assert first_non_empty_token[0] in ("B", "F")
 
 
 def test_llamacpp_stream_text_stop(model):

From cc4a06e8f815e3ca0f61e345a508b99e0e6d1cc5 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Wed, 3 Dec 2025 18:35:30 +0800
Subject: [PATCH 498/505] fix: add chat_mode parameter to LlamaCpp and
 from_llamacpp for better chat handling

---
 docs/features/models/llamacpp.md | 19 +++++++++++++++-
 outlines/models/llamacpp.py      | 39 +++++++++++++++++---------------
 tests/models/test_llamacpp.py    | 17 ++++++++++++++
 3 files changed, 56 insertions(+), 19 deletions(-)

diff --git a/docs/features/models/llamacpp.md b/docs/features/models/llamacpp.md
index 0f1717e24..867d325c5 100644
--- a/docs/features/models/llamacpp.md
+++ b/docs/features/models/llamacpp.md
@@ -15,7 +15,9 @@ Outlines provides an integration with [Llama.cpp](https://github.com/ggerganov/l
 
 ## Model Initialization
 
-To load the model, you can use the `from_llamacpp` function. The single argument of the function is a `Llama` model instance from the `llama_cpp` library. Consult the [Llama class API reference](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama) for detailed information on how to create a model instance and on the various available parameters.
+To load the model, you can use the `from_llamacpp` function. The first argument of the function is a `Llama` model instance from the `llama_cpp` library. Consult the [Llama class API reference](https://llama-cpp-python.readthedocs.io/en/latest/api-reference/#llama_cpp.Llama) for detailed information on how to create a model instance and on the various available parameters.
+
+You can also pass a `chat_mode` argument to `from_llamacpp`. If `True` (default), the model will regard all `str` inputs as user messages in a chat conversation. If `False`, the model will regard all `str` inputs as plain text prompts.
 
 For instance:
 
@@ -31,6 +33,21 @@ model = outlines.from_llamacpp(
 )
 ```
 
+You can also disable chat mode:
+
+```python
+import outlines
+from llama_cpp import Llama
+
+model = outlines.from_llamacpp(
+    Llama.from_pretrained(
+        repo_id="TheBloke/Mistral-7B-Instruct-v0.2-GGUF",
+        filename="mistral-7b-instruct-v0.2.Q5_K_M.gguf",
+    ),
+    chat_mode=False,
+)
+```
+
 ## Text Generation
 
 To generate text, you can simply call the model with a prompt.
diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index a560bbb19..d256d2788 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -238,17 +238,26 @@ class LlamaCpp(Model):
 
     tensor_library_name = "numpy"
 
-    def __init__(self, model: "Llama"):
+    def __init__(self, model: "Llama", chat_mode: bool = True):
         """
         Parameters
         ----------
         model
             A `llama_cpp.Llama` model instance.
+        chat_mode
+            Whether to enable chat mode. If `False`, the model will regard
+            all `str` inputs as plain text prompts. If `True`, the model will
+            regard all `str` inputs as user messages in a chat conversation.
 
         """
         self.model = model
         self.tokenizer = LlamaCppTokenizer(self.model)
-        self.type_adapter = LlamaCppTypeAdapter(has_chat_template=self._check_chat_template())
+
+        # Note: llama-cpp-python provides a default chat-template fallback even when
+        # the user hasn't explicitly configured one:
+        # https://github.com/abetlen/llama-cpp-python/blob/c37132b/llama_cpp/llama.py#L540-L545
+        # We keep the default as True because the upstream library generally favors chat-style usage.
+        self.type_adapter = LlamaCppTypeAdapter(has_chat_template=chat_mode)
 
     def generate(
         self,
@@ -277,14 +286,14 @@ def generate(
         """
         prompt = self.type_adapter.format_input(model_input)
 
-        if isinstance(prompt, str):
+        if isinstance(prompt, str):  # TODO: add tests
             completion = self.model(
                 prompt,
                 logits_processor=self.type_adapter.format_output_type(output_type),
                 **inference_kwargs,
             )
             result = completion["choices"][0]["text"]
-        elif isinstance(prompt, list): # pragma: no cover
+        elif isinstance(prompt, list):
             completion = self.model.create_chat_completion(
                 prompt,
                 logits_processor=self.type_adapter.format_output_type(output_type),
@@ -331,7 +340,7 @@ def generate_stream(
         """
         prompt = self.type_adapter.format_input(model_input)
 
-        if isinstance(prompt, str):
+        if isinstance(prompt, str):  # TODO: add tests
             generator = self.model(
                 prompt,
                 logits_processor=self.type_adapter.format_output_type(output_type),
@@ -341,7 +350,7 @@ def generate_stream(
             for chunk in generator:
                 yield chunk["choices"][0]["text"]
 
-        elif isinstance(prompt, list): # pragma: no cover
+        elif isinstance(prompt, list):
             generator = self.model.create_chat_completion(
                 prompt,
                 logits_processor=self.type_adapter.format_output_type(output_type),
@@ -351,18 +360,8 @@ def generate_stream(
             for chunk in generator:
                 yield chunk["choices"][0]["delta"].get("content", "")
 
-    def _check_chat_template(self) -> bool:
-        """Check if the model has a chat template defined."""
-        return (
-            self.model.chat_format is not None
-            or self.model.chat_handler is not None
-            or (
-                hasattr(self.model, "metadata")
-                and self.model.metadata.get("tokenizer.chat_template") is not None
-            )
-        )
 
-def from_llamacpp(model: "Llama"):
+def from_llamacpp(model: "Llama", chat_mode: bool = True) -> LlamaCpp:
     """Create an Outlines `LlamaCpp` model instance from a
     `llama_cpp.Llama` instance.
 
@@ -370,6 +369,10 @@ def from_llamacpp(model: "Llama"):
     ----------
     model
         A `llama_cpp.Llama` instance.
+    chat_mode
+        Whether to enable chat mode. If `False`, the model will regard
+        all `str` inputs as plain text prompts. If `True`, the model will
+        regard all `str` inputs as user messages in a chat conversation.
 
     Returns
     -------
@@ -377,4 +380,4 @@ def from_llamacpp(model: "Llama"):
         An Outlines `LlamaCpp` model instance.
 
     """
-    return LlamaCpp(model)
+    return LlamaCpp(model, chat_mode=chat_mode)
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index 4d41c7203..bead19367 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -41,6 +41,16 @@ def model(tmp_path_factory):
         )
     )
 
+@pytest.fixture(scope="session")
+def model_no_chat(tmp_path_factory):
+    return LlamaCpp(
+        Llama.from_pretrained(
+            repo_id="tensorblock/Llama3-1B-Base-GGUF",
+            filename="Llama3-1B-Base-Q2_K.gguf",
+        ),
+        chat_mode=False
+    )
+
 @pytest.fixture
 def lark_grammar():
     return """
@@ -223,3 +233,10 @@ def test_llamacpp_batch(model):
         model.batch(
             ["Respond with one word.", "Respond with one word."],
         )
+
+def test_llamacpp_no_chat(model_no_chat):
+    result = model_no_chat.generate("Respond with one word. Not more.", None)
+    assert isinstance(result, str)
+
+    generator = model_no_chat.stream("Respond with one word. Not more.", None)
+    assert isinstance(next(generator), str)

From 635741147885a0335ee3da818bc9845be382f2a7 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Wed, 3 Dec 2025 18:46:43 +0800
Subject: [PATCH 499/505] fix: remove TODO comments for adding tests in
 LlamaCpp class methods

---
 outlines/models/llamacpp.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index d256d2788..3ab348a08 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -286,7 +286,7 @@ def generate(
         """
         prompt = self.type_adapter.format_input(model_input)
 
-        if isinstance(prompt, str):  # TODO: add tests
+        if isinstance(prompt, str):
             completion = self.model(
                 prompt,
                 logits_processor=self.type_adapter.format_output_type(output_type),
@@ -340,7 +340,7 @@ def generate_stream(
         """
         prompt = self.type_adapter.format_input(model_input)
 
-        if isinstance(prompt, str):  # TODO: add tests
+        if isinstance(prompt, str):
             generator = self.model(
                 prompt,
                 logits_processor=self.type_adapter.format_output_type(output_type),

From 3a24576b4adc702d3ac811c30b54b79925dbe370 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Wed, 3 Dec 2025 19:37:00 +0800
Subject: [PATCH 500/505] fix: improve error handling for unexpected prompt
 types and update stream test assertions

---
 outlines/models/llamacpp.py   | 5 ++++-
 tests/models/test_llamacpp.py | 3 ++-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/outlines/models/llamacpp.py b/outlines/models/llamacpp.py
index 3ab348a08..e931d9c8a 100644
--- a/outlines/models/llamacpp.py
+++ b/outlines/models/llamacpp.py
@@ -300,6 +300,8 @@ def generate(
                 **inference_kwargs,
             )
             result = completion["choices"][0]["message"]["content"]
+        else:  # Never reached  # pragma: no cover
+            raise ValueError("Unexpected prompt type.")
 
         self.model.reset()
 
@@ -359,7 +361,8 @@ def generate_stream(
             )
             for chunk in generator:
                 yield chunk["choices"][0]["delta"].get("content", "")
-
+        else:  # Never reached  # pragma: no cover
+            raise ValueError("Unexpected prompt type.")
 
 def from_llamacpp(model: "Llama", chat_mode: bool = True) -> LlamaCpp:
     """Create an Outlines `LlamaCpp` model instance from a
diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index bead19367..e49f8e1ba 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -239,4 +239,5 @@ def test_llamacpp_no_chat(model_no_chat):
     assert isinstance(result, str)
 
     generator = model_no_chat.stream("Respond with one word. Not more.", None)
-    assert isinstance(next(generator), str)
+    for x in generator:
+        assert isinstance(x, str)

From d4a73ba757529fdc38e0f5a40d9de5eaf5507593 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 4 Dec 2025 14:39:38 +0800
Subject: [PATCH 501/505] test: simplify some tests

---
 tests/models/test_llamacpp_type_adapter.py    | 21 ++-------------
 .../models/test_transformers_type_adapter.py  | 26 ++++++-------------
 .../models/test_vllm_offline_type_adapter.py  |  1 -
 3 files changed, 10 insertions(+), 38 deletions(-)

diff --git a/tests/models/test_llamacpp_type_adapter.py b/tests/models/test_llamacpp_type_adapter.py
index 167d50f4e..86f9fdf9f 100644
--- a/tests/models/test_llamacpp_type_adapter.py
+++ b/tests/models/test_llamacpp_type_adapter.py
@@ -1,6 +1,5 @@
 import pytest
 import io
-from unittest.mock import MagicMock
 
 from llama_cpp import LogitsProcessorList
 from PIL import Image as PILImage
@@ -63,7 +62,7 @@ def test_llamacpp_type_adapter_format_input(adapter, image):
         ]))
 
 
-def test_llamacpp_type_adapter_format_input_with_chat_format():
+def test_llamacpp_type_adapter_format_input_with_chat_template():
     adapter = LlamaCppTypeAdapter(has_chat_template=True)
     message = "prompt"
     result = adapter.format_input(message)
@@ -71,23 +70,7 @@ def test_llamacpp_type_adapter_format_input_with_chat_format():
     assert result == [{"role": "user", "content": "prompt"}]
 
 
-def test_llamacpp_type_adapter_format_input_with_chat_handler():
-    adapter = LlamaCppTypeAdapter(has_chat_template=True)
-    message = "prompt"
-    result = adapter.format_input(message)
-
-    assert result == [{"role": "user", "content": "prompt"}]
-
-
-def test_llamacpp_type_adapter_format_input_with_metadata_template():
-    adapter = LlamaCppTypeAdapter(has_chat_template=True)
-    message = "prompt"
-    result = adapter.format_input(message)
-
-    assert result == [{"role": "user", "content": "prompt"}]
-
-
-def test_llamacpp_type_adapter_format_input_without_chat_capabilities():
+def test_llamacpp_type_adapter_format_input_without_chat_template():
     adapter = LlamaCppTypeAdapter(has_chat_template=False)
     message = "prompt"
     result = adapter.format_input(message)
diff --git a/tests/models/test_transformers_type_adapter.py b/tests/models/test_transformers_type_adapter.py
index 571e3c2b6..e5318f9f3 100644
--- a/tests/models/test_transformers_type_adapter.py
+++ b/tests/models/test_transformers_type_adapter.py
@@ -1,6 +1,5 @@
 import io
 import pytest
-from unittest.mock import MagicMock
 
 import transformers
 from transformers import LogitsProcessorList
@@ -17,27 +16,17 @@
 
 @pytest.fixture
 def adapter():
-    tokenizer = MagicMock()
-    chat_template = "{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}{% if add_generation_prompt %}assistant:{% endif %}"
-    tokenizer.chat_template = chat_template
+    tokenizer = transformers.AutoTokenizer.from_pretrained(MODEL_NAME)
+    type_adapter = TransformersTypeAdapter(tokenizer=tokenizer)
+    chat_template = '{% for message in messages %}{{ message.role }}: {{ message.content }}{% endfor %}'
+    type_adapter.tokenizer.chat_template = chat_template
 
-    def apply_chat_template(messages, tokenize=False, add_generation_prompt=False):
-        res = ""
-        for msg in messages:
-            res += f"{msg['role']}: {msg['content']}"
-        if add_generation_prompt:
-            res += "assistant:"
-        return res
-
-    tokenizer.apply_chat_template.side_effect = apply_chat_template
-
-    type_adapter = TransformersTypeAdapter(tokenizer=tokenizer, has_chat_template=True)
     return type_adapter
 
 @pytest.fixture
 def logits_processor():
-    index = MagicMock()
-    index.tensor_library_name = "torch"
+    vocabulary = Vocabulary.from_pretrained("openai-community/gpt2")
+    index = Index(r"[0-9]{3}", vocabulary)
     return OutlinesCoreLogitsProcessor(index, "torch")
 
 @pytest.fixture
@@ -63,7 +52,8 @@ def test_transformers_type_adapter_format_input(adapter, image):
 
     # string with chat template
     # The fixture sets a chat template, so it should be formatted
-    assert adapter.format_input("Hello, world!") == "user: Hello, world!assistant:"
+    adapter.has_chat_template = True
+    assert adapter.format_input("Hello, world!") == "user: Hello, world!"
 
     # string without chat template
     adapter.has_chat_template = False
diff --git a/tests/models/test_vllm_offline_type_adapter.py b/tests/models/test_vllm_offline_type_adapter.py
index 77bcb7353..bb9b9fc69 100644
--- a/tests/models/test_vllm_offline_type_adapter.py
+++ b/tests/models/test_vllm_offline_type_adapter.py
@@ -1,6 +1,5 @@
 import io
 import json
-from unittest.mock import MagicMock
 
 import pytest
 from PIL import Image as PILImage

From 272ee0aec23ce2877766f21088d66c0aaa27fab5 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 4 Dec 2025 15:31:50 +0800
Subject: [PATCH 502/505] fix: add tests for _check_hf_chat_template function
 in tokenizer module

---
 tests/models/test_tokenizer.py | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/tests/models/test_tokenizer.py b/tests/models/test_tokenizer.py
index 831f7fe3e..13ba45f3e 100644
--- a/tests/models/test_tokenizer.py
+++ b/tests/models/test_tokenizer.py
@@ -1,8 +1,14 @@
 import pytest
 
-from outlines.models.tokenizer import Tokenizer
+from outlines.models.tokenizer import Tokenizer, _check_hf_chat_template
 
 
 def test_tokenizer():
     with pytest.raises(TypeError, match="instantiate abstract"):
         Tokenizer()
+
+def test_check_hf_chat_template():
+    from transformers import AutoTokenizer
+
+    assert _check_hf_chat_template(AutoTokenizer.from_pretrained("openai-community/gpt2")) is False
+    assert _check_hf_chat_template(AutoTokenizer.from_pretrained("Qwen/Qwen3-0.6B")) is True

From 6557b3d6bfc12c4908042535189923d9ea71e9c0 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 4 Dec 2025 15:47:12 +0800
Subject: [PATCH 503/505] fix: remove unused AnyTokenizer import from
 vllm_offline.py

---
 outlines/models/vllm_offline.py | 1 -
 1 file changed, 1 deletion(-)

diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index c74a78570..1182944e8 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -13,7 +13,6 @@
     from vllm import LLM
     from vllm.sampling_params import SamplingParams
     from vllm.transformers_utils.tokenizer import (
-        AnyTokenizer,
         PreTrainedTokenizer,
         PreTrainedTokenizerFast,
         TokenizerBase

From 37ba815ee397de3da872d50898fc429972a2d3d8 Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 4 Dec 2025 16:05:29 +0800
Subject: [PATCH 504/505] fix: correct typo

---
 tests/models/test_llamacpp.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tests/models/test_llamacpp.py b/tests/models/test_llamacpp.py
index e49f8e1ba..18c57ce6c 100644
--- a/tests/models/test_llamacpp.py
+++ b/tests/models/test_llamacpp.py
@@ -182,7 +182,7 @@ class Foo(BaseModel):
     generator = model.stream("foo?", Foo)
 
     # NOTE: The first few chunks may be empty (role info, control tokens, finish chunks)
-    # Relavant issue: https://github.com/abetlen/llama-cpp-python/issues/372
+    # Relevant issue: https://github.com/abetlen/llama-cpp-python/issues/372
     first_non_empty_token = next(x for x in generator if x)
     assert first_non_empty_token == "{"
 

From b98ee3128e7f4a544faf21747fe2c476cf456c4a Mon Sep 17 00:00:00 2001
From: Shichao Song <60967965+Ki-Seki@users.noreply.github.com>
Date: Thu, 4 Dec 2025 16:10:43 +0800
Subject: [PATCH 505/505] fix: move tokenizer imports to the appropriate
 location in vllm_offline.py so that no runtime errors occur

---
 outlines/models/vllm_offline.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/outlines/models/vllm_offline.py b/outlines/models/vllm_offline.py
index 1182944e8..59d1eb477 100644
--- a/outlines/models/vllm_offline.py
+++ b/outlines/models/vllm_offline.py
@@ -12,11 +12,6 @@
 if TYPE_CHECKING:
     from vllm import LLM
     from vllm.sampling_params import SamplingParams
-    from vllm.transformers_utils.tokenizer import (
-        PreTrainedTokenizer,
-        PreTrainedTokenizerFast,
-        TokenizerBase
-    )
 
 __all__ = ["VLLMOffline", "from_vllm_offline"]
 
@@ -254,6 +249,11 @@ def generate_stream(self, model_input, output_type, **inference_kwargs):
 
     def _check_chat_template(self) -> bool:
         """Check if the tokenizer has a chat template."""
+        from vllm.transformers_utils.tokenizer import (
+            PreTrainedTokenizer,
+            PreTrainedTokenizerFast,
+            TokenizerBase
+        )
         from outlines.models.tokenizer import _check_hf_chat_template
 
         if isinstance(self.tokenizer, (PreTrainedTokenizer, PreTrainedTokenizerFast)):